{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.487747048960998,
  "eval_steps": 87,
  "global_step": 90000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000541941165512219,
      "grad_norm": 18.66890525817871,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 7.3063,
      "step": 10
    },
    {
      "epoch": 0.001083882331024438,
      "grad_norm": 14.986076354980469,
      "learning_rate": 6.666666666666667e-07,
      "loss": 6.602,
      "step": 20
    },
    {
      "epoch": 0.0016258234965366573,
      "grad_norm": 7.8340582847595215,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 5.4589,
      "step": 30
    },
    {
      "epoch": 0.002167764662048876,
      "grad_norm": 3.6684656143188477,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 4.4819,
      "step": 40
    },
    {
      "epoch": 0.0027097058275610954,
      "grad_norm": 1.7751961946487427,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 3.8516,
      "step": 50
    },
    {
      "epoch": 0.0032516469930733145,
      "grad_norm": 0.8667094707489014,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 3.471,
      "step": 60
    },
    {
      "epoch": 0.0037935881585855337,
      "grad_norm": 0.48009157180786133,
      "learning_rate": 2.3333333333333336e-06,
      "loss": 3.2603,
      "step": 70
    },
    {
      "epoch": 0.004335529324097752,
      "grad_norm": 0.31772586703300476,
      "learning_rate": 2.666666666666667e-06,
      "loss": 3.1565,
      "step": 80
    },
    {
      "epoch": 0.004714888139956306,
      "eval_loss": 3.0095248222351074,
      "eval_runtime": 22.3757,
      "eval_samples_per_second": 223.456,
      "eval_steps_per_second": 1.207,
      "step": 87
    },
    {
      "epoch": 0.0048774704896099716,
      "grad_norm": 0.22845230996608734,
      "learning_rate": 3e-06,
      "loss": 3.0993,
      "step": 90
    },
    {
      "epoch": 0.005419411655122191,
      "grad_norm": 0.1711948961019516,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 3.067,
      "step": 100
    },
    {
      "epoch": 0.00596135282063441,
      "grad_norm": 0.24801534414291382,
      "learning_rate": 3.666666666666667e-06,
      "loss": 3.0429,
      "step": 110
    },
    {
      "epoch": 0.006503293986146629,
      "grad_norm": 0.1762855052947998,
      "learning_rate": 4.000000000000001e-06,
      "loss": 3.0244,
      "step": 120
    },
    {
      "epoch": 0.007045235151658848,
      "grad_norm": 0.20275895297527313,
      "learning_rate": 4.333333333333334e-06,
      "loss": 3.0101,
      "step": 130
    },
    {
      "epoch": 0.007587176317171067,
      "grad_norm": 0.20807678997516632,
      "learning_rate": 4.666666666666667e-06,
      "loss": 2.9962,
      "step": 140
    },
    {
      "epoch": 0.008129117482683286,
      "grad_norm": 0.21902145445346832,
      "learning_rate": 5e-06,
      "loss": 2.9879,
      "step": 150
    },
    {
      "epoch": 0.008671058648195505,
      "grad_norm": 0.2640894055366516,
      "learning_rate": 5.333333333333334e-06,
      "loss": 2.9807,
      "step": 160
    },
    {
      "epoch": 0.009212999813707724,
      "grad_norm": 0.2247193157672882,
      "learning_rate": 5.666666666666667e-06,
      "loss": 2.9764,
      "step": 170
    },
    {
      "epoch": 0.009429776279912611,
      "eval_loss": 2.909708023071289,
      "eval_runtime": 21.9944,
      "eval_samples_per_second": 227.331,
      "eval_steps_per_second": 1.228,
      "step": 174
    },
    {
      "epoch": 0.009754940979219943,
      "grad_norm": 0.1678290069103241,
      "learning_rate": 6e-06,
      "loss": 2.9725,
      "step": 180
    },
    {
      "epoch": 0.010296882144732162,
      "grad_norm": 0.16639265418052673,
      "learning_rate": 6.333333333333334e-06,
      "loss": 2.9687,
      "step": 190
    },
    {
      "epoch": 0.010838823310244381,
      "grad_norm": 0.16535356640815735,
      "learning_rate": 6.666666666666667e-06,
      "loss": 2.968,
      "step": 200
    },
    {
      "epoch": 0.0113807644757566,
      "grad_norm": 0.12626643478870392,
      "learning_rate": 7.000000000000001e-06,
      "loss": 2.9656,
      "step": 210
    },
    {
      "epoch": 0.01192270564126882,
      "grad_norm": 0.16154739260673523,
      "learning_rate": 7.333333333333334e-06,
      "loss": 2.9658,
      "step": 220
    },
    {
      "epoch": 0.012464646806781039,
      "grad_norm": 0.10952357947826385,
      "learning_rate": 7.666666666666667e-06,
      "loss": 2.9626,
      "step": 230
    },
    {
      "epoch": 0.013006587972293258,
      "grad_norm": 0.13569523394107819,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.9611,
      "step": 240
    },
    {
      "epoch": 0.013548529137805477,
      "grad_norm": 0.18444740772247314,
      "learning_rate": 8.333333333333334e-06,
      "loss": 2.9599,
      "step": 250
    },
    {
      "epoch": 0.014090470303317696,
      "grad_norm": 0.1738782823085785,
      "learning_rate": 8.666666666666668e-06,
      "loss": 2.9588,
      "step": 260
    },
    {
      "epoch": 0.014144664419868919,
      "eval_loss": 2.9037859439849854,
      "eval_runtime": 21.9954,
      "eval_samples_per_second": 227.321,
      "eval_steps_per_second": 1.228,
      "step": 261
    },
    {
      "epoch": 0.014632411468829916,
      "grad_norm": 0.1825179159641266,
      "learning_rate": 9e-06,
      "loss": 2.9562,
      "step": 270
    },
    {
      "epoch": 0.015174352634342135,
      "grad_norm": 0.1394873708486557,
      "learning_rate": 9.333333333333334e-06,
      "loss": 2.9553,
      "step": 280
    },
    {
      "epoch": 0.015716293799854352,
      "grad_norm": 0.11560714244842529,
      "learning_rate": 9.666666666666667e-06,
      "loss": 2.9563,
      "step": 290
    },
    {
      "epoch": 0.01625823496536657,
      "grad_norm": 0.16794171929359436,
      "learning_rate": 1e-05,
      "loss": 2.9527,
      "step": 300
    },
    {
      "epoch": 0.01680017613087879,
      "grad_norm": 0.11284048855304718,
      "learning_rate": 1.0333333333333333e-05,
      "loss": 2.9529,
      "step": 310
    },
    {
      "epoch": 0.01734211729639101,
      "grad_norm": 0.12263070046901703,
      "learning_rate": 1.0666666666666667e-05,
      "loss": 2.9521,
      "step": 320
    },
    {
      "epoch": 0.01788405846190323,
      "grad_norm": 0.22584529221057892,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 2.9527,
      "step": 330
    },
    {
      "epoch": 0.018425999627415448,
      "grad_norm": 0.15804457664489746,
      "learning_rate": 1.1333333333333334e-05,
      "loss": 2.9509,
      "step": 340
    },
    {
      "epoch": 0.018859552559825223,
      "eval_loss": 2.8973894119262695,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.324,
      "eval_steps_per_second": 1.228,
      "step": 348
    },
    {
      "epoch": 0.018967940792927667,
      "grad_norm": 0.20428232848644257,
      "learning_rate": 1.1666666666666668e-05,
      "loss": 2.9489,
      "step": 350
    },
    {
      "epoch": 0.019509881958439886,
      "grad_norm": 0.18693846464157104,
      "learning_rate": 1.2e-05,
      "loss": 2.95,
      "step": 360
    },
    {
      "epoch": 0.020051823123952105,
      "grad_norm": 0.113795205950737,
      "learning_rate": 1.2333333333333334e-05,
      "loss": 2.9469,
      "step": 370
    },
    {
      "epoch": 0.020593764289464325,
      "grad_norm": 0.18272531032562256,
      "learning_rate": 1.2666666666666668e-05,
      "loss": 2.9474,
      "step": 380
    },
    {
      "epoch": 0.021135705454976544,
      "grad_norm": 0.14928747713565826,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 2.9478,
      "step": 390
    },
    {
      "epoch": 0.021677646620488763,
      "grad_norm": 0.28558146953582764,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 2.9453,
      "step": 400
    },
    {
      "epoch": 0.022219587786000982,
      "grad_norm": 0.2676473557949066,
      "learning_rate": 1.3666666666666666e-05,
      "loss": 2.9447,
      "step": 410
    },
    {
      "epoch": 0.0227615289515132,
      "grad_norm": 0.18305936455726624,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 2.9433,
      "step": 420
    },
    {
      "epoch": 0.02330347011702542,
      "grad_norm": 0.21853986382484436,
      "learning_rate": 1.4333333333333334e-05,
      "loss": 2.9438,
      "step": 430
    },
    {
      "epoch": 0.02357444069978153,
      "eval_loss": 2.8930819034576416,
      "eval_runtime": 21.9945,
      "eval_samples_per_second": 227.33,
      "eval_steps_per_second": 1.228,
      "step": 435
    },
    {
      "epoch": 0.02384541128253764,
      "grad_norm": 0.218394935131073,
      "learning_rate": 1.4666666666666668e-05,
      "loss": 2.942,
      "step": 440
    },
    {
      "epoch": 0.02438735244804986,
      "grad_norm": 0.15290504693984985,
      "learning_rate": 1.5e-05,
      "loss": 2.9412,
      "step": 450
    },
    {
      "epoch": 0.024929293613562078,
      "grad_norm": 0.2329382598400116,
      "learning_rate": 1.5333333333333334e-05,
      "loss": 2.9393,
      "step": 460
    },
    {
      "epoch": 0.025471234779074297,
      "grad_norm": 0.2563980519771576,
      "learning_rate": 1.5666666666666667e-05,
      "loss": 2.9404,
      "step": 470
    },
    {
      "epoch": 0.026013175944586516,
      "grad_norm": 0.18090900778770447,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.9384,
      "step": 480
    },
    {
      "epoch": 0.026555117110098735,
      "grad_norm": 0.10313646495342255,
      "learning_rate": 1.6333333333333335e-05,
      "loss": 2.9383,
      "step": 490
    },
    {
      "epoch": 0.027097058275610954,
      "grad_norm": 0.127348393201828,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 2.9374,
      "step": 500
    },
    {
      "epoch": 0.027638999441123174,
      "grad_norm": 0.39012327790260315,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 2.9369,
      "step": 510
    },
    {
      "epoch": 0.028180940606635393,
      "grad_norm": 0.26286575198173523,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 2.9365,
      "step": 520
    },
    {
      "epoch": 0.028289328839737837,
      "eval_loss": 2.885981798171997,
      "eval_runtime": 35.0266,
      "eval_samples_per_second": 142.749,
      "eval_steps_per_second": 0.771,
      "step": 522
    },
    {
      "epoch": 0.028722881772147612,
      "grad_norm": 0.34882956743240356,
      "learning_rate": 1.7666666666666668e-05,
      "loss": 2.9374,
      "step": 530
    },
    {
      "epoch": 0.02926482293765983,
      "grad_norm": 0.2220120131969452,
      "learning_rate": 1.8e-05,
      "loss": 2.934,
      "step": 540
    },
    {
      "epoch": 0.02980676410317205,
      "grad_norm": 0.11112108826637268,
      "learning_rate": 1.8333333333333333e-05,
      "loss": 2.9332,
      "step": 550
    },
    {
      "epoch": 0.03034870526868427,
      "grad_norm": 0.1723136603832245,
      "learning_rate": 1.866666666666667e-05,
      "loss": 2.9332,
      "step": 560
    },
    {
      "epoch": 0.03089064643419649,
      "grad_norm": 0.1271258443593979,
      "learning_rate": 1.9e-05,
      "loss": 2.9326,
      "step": 570
    },
    {
      "epoch": 0.031432587599708704,
      "grad_norm": 0.2465275079011917,
      "learning_rate": 1.9333333333333333e-05,
      "loss": 2.9304,
      "step": 580
    },
    {
      "epoch": 0.03197452876522092,
      "grad_norm": 0.24456460773944855,
      "learning_rate": 1.9666666666666666e-05,
      "loss": 2.9304,
      "step": 590
    },
    {
      "epoch": 0.03251646993073314,
      "grad_norm": 0.18503615260124207,
      "learning_rate": 2e-05,
      "loss": 2.9272,
      "step": 600
    },
    {
      "epoch": 0.03300421697969414,
      "eval_loss": 2.8818912506103516,
      "eval_runtime": 21.9944,
      "eval_samples_per_second": 227.331,
      "eval_steps_per_second": 1.228,
      "step": 609
    },
    {
      "epoch": 0.03305841109624536,
      "grad_norm": 0.18611077964305878,
      "learning_rate": 2.0333333333333334e-05,
      "loss": 2.9287,
      "step": 610
    },
    {
      "epoch": 0.03360035226175758,
      "grad_norm": 0.2303771823644638,
      "learning_rate": 2.0666666666666666e-05,
      "loss": 2.9272,
      "step": 620
    },
    {
      "epoch": 0.0341422934272698,
      "grad_norm": 0.12826542556285858,
      "learning_rate": 2.1e-05,
      "loss": 2.926,
      "step": 630
    },
    {
      "epoch": 0.03468423459278202,
      "grad_norm": 0.18119564652442932,
      "learning_rate": 2.1333333333333335e-05,
      "loss": 2.9259,
      "step": 640
    },
    {
      "epoch": 0.03522617575829424,
      "grad_norm": 0.17282642424106598,
      "learning_rate": 2.1666666666666667e-05,
      "loss": 2.9264,
      "step": 650
    },
    {
      "epoch": 0.03576811692380646,
      "grad_norm": 0.18431398272514343,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 2.9243,
      "step": 660
    },
    {
      "epoch": 0.03631005808931868,
      "grad_norm": 0.2258967012166977,
      "learning_rate": 2.2333333333333335e-05,
      "loss": 2.9244,
      "step": 670
    },
    {
      "epoch": 0.036851999254830896,
      "grad_norm": 0.14141803979873657,
      "learning_rate": 2.2666666666666668e-05,
      "loss": 2.9226,
      "step": 680
    },
    {
      "epoch": 0.037393940420343115,
      "grad_norm": 0.3774568736553192,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 2.9226,
      "step": 690
    },
    {
      "epoch": 0.037719105119650445,
      "eval_loss": 2.8792285919189453,
      "eval_runtime": 21.9939,
      "eval_samples_per_second": 227.336,
      "eval_steps_per_second": 1.228,
      "step": 696
    },
    {
      "epoch": 0.037935881585855334,
      "grad_norm": 0.2437097430229187,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 2.9214,
      "step": 700
    },
    {
      "epoch": 0.03847782275136755,
      "grad_norm": 0.18445032835006714,
      "learning_rate": 2.3666666666666668e-05,
      "loss": 2.9201,
      "step": 710
    },
    {
      "epoch": 0.03901976391687977,
      "grad_norm": 0.12405350804328918,
      "learning_rate": 2.4e-05,
      "loss": 2.919,
      "step": 720
    },
    {
      "epoch": 0.03956170508239199,
      "grad_norm": 0.14779530465602875,
      "learning_rate": 2.4333333333333336e-05,
      "loss": 2.919,
      "step": 730
    },
    {
      "epoch": 0.04010364624790421,
      "grad_norm": 0.16767142713069916,
      "learning_rate": 2.466666666666667e-05,
      "loss": 2.9153,
      "step": 740
    },
    {
      "epoch": 0.04064558741341643,
      "grad_norm": 0.22561559081077576,
      "learning_rate": 2.5e-05,
      "loss": 2.916,
      "step": 750
    },
    {
      "epoch": 0.04118752857892865,
      "grad_norm": 0.1435348242521286,
      "learning_rate": 2.5333333333333337e-05,
      "loss": 2.9134,
      "step": 760
    },
    {
      "epoch": 0.04172946974444087,
      "grad_norm": 0.25675374269485474,
      "learning_rate": 2.5666666666666666e-05,
      "loss": 2.9133,
      "step": 770
    },
    {
      "epoch": 0.04227141090995309,
      "grad_norm": 0.22575390338897705,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 2.9124,
      "step": 780
    },
    {
      "epoch": 0.042433993259606756,
      "eval_loss": 2.874786853790283,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.324,
      "eval_steps_per_second": 1.228,
      "step": 783
    },
    {
      "epoch": 0.04281335207546531,
      "grad_norm": 0.17684337496757507,
      "learning_rate": 2.633333333333333e-05,
      "loss": 2.9118,
      "step": 790
    },
    {
      "epoch": 0.043355293240977526,
      "grad_norm": 0.27737951278686523,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 2.9098,
      "step": 800
    },
    {
      "epoch": 0.043897234406489745,
      "grad_norm": 0.28385818004608154,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 2.9101,
      "step": 810
    },
    {
      "epoch": 0.044439175572001964,
      "grad_norm": 0.21399657428264618,
      "learning_rate": 2.733333333333333e-05,
      "loss": 2.909,
      "step": 820
    },
    {
      "epoch": 0.04498111673751418,
      "grad_norm": 0.20526191592216492,
      "learning_rate": 2.7666666666666667e-05,
      "loss": 2.906,
      "step": 830
    },
    {
      "epoch": 0.0455230579030264,
      "grad_norm": 0.3085150420665741,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 2.9058,
      "step": 840
    },
    {
      "epoch": 0.04606499906853862,
      "grad_norm": 0.18636156618595123,
      "learning_rate": 2.8333333333333335e-05,
      "loss": 2.9055,
      "step": 850
    },
    {
      "epoch": 0.04660694023405084,
      "grad_norm": 0.23989151418209076,
      "learning_rate": 2.8666666666666668e-05,
      "loss": 2.9046,
      "step": 860
    },
    {
      "epoch": 0.04714888139956306,
      "grad_norm": 0.23799440264701843,
      "learning_rate": 2.9e-05,
      "loss": 2.9056,
      "step": 870
    },
    {
      "epoch": 0.04714888139956306,
      "eval_loss": 2.871436595916748,
      "eval_runtime": 21.8935,
      "eval_samples_per_second": 228.379,
      "eval_steps_per_second": 1.233,
      "step": 870
    },
    {
      "epoch": 0.04769082256507528,
      "grad_norm": 0.1619342863559723,
      "learning_rate": 2.9333333333333336e-05,
      "loss": 2.9041,
      "step": 880
    },
    {
      "epoch": 0.0482327637305875,
      "grad_norm": 0.28320956230163574,
      "learning_rate": 2.9666666666666672e-05,
      "loss": 2.9025,
      "step": 890
    },
    {
      "epoch": 0.04877470489609972,
      "grad_norm": 0.18892525136470795,
      "learning_rate": 3e-05,
      "loss": 2.8997,
      "step": 900
    },
    {
      "epoch": 0.049316646061611936,
      "grad_norm": 0.29564598202705383,
      "learning_rate": 3.0333333333333337e-05,
      "loss": 2.9001,
      "step": 910
    },
    {
      "epoch": 0.049858587227124156,
      "grad_norm": 0.24769143760204315,
      "learning_rate": 3.066666666666667e-05,
      "loss": 2.8987,
      "step": 920
    },
    {
      "epoch": 0.050400528392636375,
      "grad_norm": 0.21904461085796356,
      "learning_rate": 3.1e-05,
      "loss": 2.8968,
      "step": 930
    },
    {
      "epoch": 0.050942469558148594,
      "grad_norm": 0.29892289638519287,
      "learning_rate": 3.1333333333333334e-05,
      "loss": 2.8972,
      "step": 940
    },
    {
      "epoch": 0.05148441072366081,
      "grad_norm": 0.20083844661712646,
      "learning_rate": 3.1666666666666666e-05,
      "loss": 2.8951,
      "step": 950
    },
    {
      "epoch": 0.051863769539519364,
      "eval_loss": 2.865743398666382,
      "eval_runtime": 21.7153,
      "eval_samples_per_second": 230.253,
      "eval_steps_per_second": 1.243,
      "step": 957
    },
    {
      "epoch": 0.05202635188917303,
      "grad_norm": 0.26588037610054016,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 2.8932,
      "step": 960
    },
    {
      "epoch": 0.05256829305468525,
      "grad_norm": 0.23635846376419067,
      "learning_rate": 3.233333333333333e-05,
      "loss": 2.8911,
      "step": 970
    },
    {
      "epoch": 0.05311023422019747,
      "grad_norm": 0.34888023138046265,
      "learning_rate": 3.266666666666667e-05,
      "loss": 2.891,
      "step": 980
    },
    {
      "epoch": 0.05365217538570969,
      "grad_norm": 0.34451714158058167,
      "learning_rate": 3.3e-05,
      "loss": 2.8907,
      "step": 990
    },
    {
      "epoch": 0.05419411655122191,
      "grad_norm": 0.2428581416606903,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.8909,
      "step": 1000
    },
    {
      "epoch": 0.05473605771673413,
      "grad_norm": 0.17025752365589142,
      "learning_rate": 3.366666666666667e-05,
      "loss": 2.8891,
      "step": 1010
    },
    {
      "epoch": 0.05527799888224635,
      "grad_norm": 0.23046477138996124,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 2.8887,
      "step": 1020
    },
    {
      "epoch": 0.055819940047758566,
      "grad_norm": 0.28877273201942444,
      "learning_rate": 3.433333333333333e-05,
      "loss": 2.8855,
      "step": 1030
    },
    {
      "epoch": 0.056361881213270786,
      "grad_norm": 0.16221983730793,
      "learning_rate": 3.466666666666667e-05,
      "loss": 2.8857,
      "step": 1040
    },
    {
      "epoch": 0.056578657679475675,
      "eval_loss": 2.8612656593322754,
      "eval_runtime": 22.0002,
      "eval_samples_per_second": 227.27,
      "eval_steps_per_second": 1.227,
      "step": 1044
    },
    {
      "epoch": 0.056903822378783005,
      "grad_norm": 0.10896836966276169,
      "learning_rate": 3.5e-05,
      "loss": 2.8861,
      "step": 1050
    },
    {
      "epoch": 0.057445763544295224,
      "grad_norm": 0.2890329360961914,
      "learning_rate": 3.5333333333333336e-05,
      "loss": 2.8826,
      "step": 1060
    },
    {
      "epoch": 0.05798770470980744,
      "grad_norm": 0.4240993857383728,
      "learning_rate": 3.566666666666667e-05,
      "loss": 2.885,
      "step": 1070
    },
    {
      "epoch": 0.05852964587531966,
      "grad_norm": 0.33681583404541016,
      "learning_rate": 3.6e-05,
      "loss": 2.8837,
      "step": 1080
    },
    {
      "epoch": 0.05907158704083188,
      "grad_norm": 0.23618997633457184,
      "learning_rate": 3.633333333333333e-05,
      "loss": 2.8846,
      "step": 1090
    },
    {
      "epoch": 0.0596135282063441,
      "grad_norm": 0.1710646152496338,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 2.881,
      "step": 1100
    },
    {
      "epoch": 0.06015546937185632,
      "grad_norm": 0.2047516405582428,
      "learning_rate": 3.7e-05,
      "loss": 2.8773,
      "step": 1110
    },
    {
      "epoch": 0.06069741053736854,
      "grad_norm": 0.20800012350082397,
      "learning_rate": 3.733333333333334e-05,
      "loss": 2.8787,
      "step": 1120
    },
    {
      "epoch": 0.06123935170288076,
      "grad_norm": 0.2583804428577423,
      "learning_rate": 3.766666666666667e-05,
      "loss": 2.876,
      "step": 1130
    },
    {
      "epoch": 0.06129354581943198,
      "eval_loss": 2.860222578048706,
      "eval_runtime": 26.0987,
      "eval_samples_per_second": 191.581,
      "eval_steps_per_second": 1.035,
      "step": 1131
    },
    {
      "epoch": 0.06178129286839298,
      "grad_norm": 0.1588805764913559,
      "learning_rate": 3.8e-05,
      "loss": 2.8741,
      "step": 1140
    },
    {
      "epoch": 0.062323234033905196,
      "grad_norm": 0.22128504514694214,
      "learning_rate": 3.8333333333333334e-05,
      "loss": 2.8781,
      "step": 1150
    },
    {
      "epoch": 0.06286517519941741,
      "grad_norm": 0.2544479966163635,
      "learning_rate": 3.866666666666667e-05,
      "loss": 2.8762,
      "step": 1160
    },
    {
      "epoch": 0.06340711636492963,
      "grad_norm": 0.3215446472167969,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 2.8758,
      "step": 1170
    },
    {
      "epoch": 0.06394905753044185,
      "grad_norm": 0.20564351975917816,
      "learning_rate": 3.933333333333333e-05,
      "loss": 2.8725,
      "step": 1180
    },
    {
      "epoch": 0.06449099869595407,
      "grad_norm": 0.24028970301151276,
      "learning_rate": 3.966666666666667e-05,
      "loss": 2.8728,
      "step": 1190
    },
    {
      "epoch": 0.06503293986146629,
      "grad_norm": 0.21962830424308777,
      "learning_rate": 4e-05,
      "loss": 2.8702,
      "step": 1200
    },
    {
      "epoch": 0.0655748810269785,
      "grad_norm": 0.281931072473526,
      "learning_rate": 4.0333333333333336e-05,
      "loss": 2.869,
      "step": 1210
    },
    {
      "epoch": 0.06600843395938828,
      "eval_loss": 2.8557112216949463,
      "eval_runtime": 21.9913,
      "eval_samples_per_second": 227.363,
      "eval_steps_per_second": 1.228,
      "step": 1218
    },
    {
      "epoch": 0.06611682219249072,
      "grad_norm": 0.16926707327365875,
      "learning_rate": 4.066666666666667e-05,
      "loss": 2.8687,
      "step": 1220
    },
    {
      "epoch": 0.06665876335800294,
      "grad_norm": 0.3461899757385254,
      "learning_rate": 4.1e-05,
      "loss": 2.8666,
      "step": 1230
    },
    {
      "epoch": 0.06720070452351516,
      "grad_norm": 0.324008047580719,
      "learning_rate": 4.133333333333333e-05,
      "loss": 2.8691,
      "step": 1240
    },
    {
      "epoch": 0.06774264568902738,
      "grad_norm": 0.2589833438396454,
      "learning_rate": 4.166666666666667e-05,
      "loss": 2.8676,
      "step": 1250
    },
    {
      "epoch": 0.0682845868545396,
      "grad_norm": 0.21630503237247467,
      "learning_rate": 4.2e-05,
      "loss": 2.8649,
      "step": 1260
    },
    {
      "epoch": 0.06882652802005182,
      "grad_norm": 0.19127613306045532,
      "learning_rate": 4.233333333333334e-05,
      "loss": 2.8643,
      "step": 1270
    },
    {
      "epoch": 0.06936846918556404,
      "grad_norm": 0.2493632286787033,
      "learning_rate": 4.266666666666667e-05,
      "loss": 2.8633,
      "step": 1280
    },
    {
      "epoch": 0.06991041035107626,
      "grad_norm": 0.26412254571914673,
      "learning_rate": 4.3e-05,
      "loss": 2.8644,
      "step": 1290
    },
    {
      "epoch": 0.07045235151658848,
      "grad_norm": 0.27580294013023376,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 2.8618,
      "step": 1300
    },
    {
      "epoch": 0.0707233220993446,
      "eval_loss": 2.8502135276794434,
      "eval_runtime": 21.9938,
      "eval_samples_per_second": 227.337,
      "eval_steps_per_second": 1.228,
      "step": 1305
    },
    {
      "epoch": 0.0709942926821007,
      "grad_norm": 0.37026703357696533,
      "learning_rate": 4.3666666666666666e-05,
      "loss": 2.8585,
      "step": 1310
    },
    {
      "epoch": 0.07153623384761292,
      "grad_norm": 0.3927730917930603,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 2.8594,
      "step": 1320
    },
    {
      "epoch": 0.07207817501312513,
      "grad_norm": 0.21110251545906067,
      "learning_rate": 4.433333333333334e-05,
      "loss": 2.858,
      "step": 1330
    },
    {
      "epoch": 0.07262011617863735,
      "grad_norm": 0.29518288373947144,
      "learning_rate": 4.466666666666667e-05,
      "loss": 2.8537,
      "step": 1340
    },
    {
      "epoch": 0.07316205734414957,
      "grad_norm": 0.27836179733276367,
      "learning_rate": 4.5e-05,
      "loss": 2.8552,
      "step": 1350
    },
    {
      "epoch": 0.07370399850966179,
      "grad_norm": 0.39191293716430664,
      "learning_rate": 4.5333333333333335e-05,
      "loss": 2.8549,
      "step": 1360
    },
    {
      "epoch": 0.07424593967517401,
      "grad_norm": 0.1891399621963501,
      "learning_rate": 4.566666666666667e-05,
      "loss": 2.8531,
      "step": 1370
    },
    {
      "epoch": 0.07478788084068623,
      "grad_norm": 0.3418962061405182,
      "learning_rate": 4.600000000000001e-05,
      "loss": 2.8513,
      "step": 1380
    },
    {
      "epoch": 0.07532982200619845,
      "grad_norm": 0.2758435308933258,
      "learning_rate": 4.633333333333333e-05,
      "loss": 2.8523,
      "step": 1390
    },
    {
      "epoch": 0.07543821023930089,
      "eval_loss": 2.8491737842559814,
      "eval_runtime": 21.9942,
      "eval_samples_per_second": 227.333,
      "eval_steps_per_second": 1.228,
      "step": 1392
    },
    {
      "epoch": 0.07587176317171067,
      "grad_norm": 0.5267038941383362,
      "learning_rate": 4.666666666666667e-05,
      "loss": 2.8527,
      "step": 1400
    },
    {
      "epoch": 0.07641370433722289,
      "grad_norm": 0.31036752462387085,
      "learning_rate": 4.7e-05,
      "loss": 2.8491,
      "step": 1410
    },
    {
      "epoch": 0.0769556455027351,
      "grad_norm": 0.2229401171207428,
      "learning_rate": 4.7333333333333336e-05,
      "loss": 2.849,
      "step": 1420
    },
    {
      "epoch": 0.07749758666824733,
      "grad_norm": 0.40294215083122253,
      "learning_rate": 4.766666666666667e-05,
      "loss": 2.847,
      "step": 1430
    },
    {
      "epoch": 0.07803952783375954,
      "grad_norm": 0.25234729051589966,
      "learning_rate": 4.8e-05,
      "loss": 2.8497,
      "step": 1440
    },
    {
      "epoch": 0.07858146899927176,
      "grad_norm": 0.2245229333639145,
      "learning_rate": 4.8333333333333334e-05,
      "loss": 2.8429,
      "step": 1450
    },
    {
      "epoch": 0.07912341016478398,
      "grad_norm": 0.37367942929267883,
      "learning_rate": 4.866666666666667e-05,
      "loss": 2.8461,
      "step": 1460
    },
    {
      "epoch": 0.0796653513302962,
      "grad_norm": 0.36259761452674866,
      "learning_rate": 4.9e-05,
      "loss": 2.8394,
      "step": 1470
    },
    {
      "epoch": 0.0801530983792572,
      "eval_loss": 2.847864866256714,
      "eval_runtime": 21.994,
      "eval_samples_per_second": 227.334,
      "eval_steps_per_second": 1.228,
      "step": 1479
    },
    {
      "epoch": 0.08020729249580842,
      "grad_norm": 0.3754774034023285,
      "learning_rate": 4.933333333333334e-05,
      "loss": 2.8415,
      "step": 1480
    },
    {
      "epoch": 0.08074923366132064,
      "grad_norm": 0.5675404071807861,
      "learning_rate": 4.966666666666667e-05,
      "loss": 2.8442,
      "step": 1490
    },
    {
      "epoch": 0.08129117482683286,
      "grad_norm": 0.23455142974853516,
      "learning_rate": 5e-05,
      "loss": 2.8377,
      "step": 1500
    },
    {
      "epoch": 0.08183311599234508,
      "grad_norm": 0.34516745805740356,
      "learning_rate": 5.0333333333333335e-05,
      "loss": 2.8378,
      "step": 1510
    },
    {
      "epoch": 0.0823750571578573,
      "grad_norm": 0.26785463094711304,
      "learning_rate": 5.0666666666666674e-05,
      "loss": 2.8371,
      "step": 1520
    },
    {
      "epoch": 0.08291699832336952,
      "grad_norm": 0.39150965213775635,
      "learning_rate": 5.1000000000000006e-05,
      "loss": 2.8366,
      "step": 1530
    },
    {
      "epoch": 0.08345893948888174,
      "grad_norm": 0.322078138589859,
      "learning_rate": 5.133333333333333e-05,
      "loss": 2.8332,
      "step": 1540
    },
    {
      "epoch": 0.08400088065439396,
      "grad_norm": 0.34766703844070435,
      "learning_rate": 5.166666666666667e-05,
      "loss": 2.8366,
      "step": 1550
    },
    {
      "epoch": 0.08454282181990617,
      "grad_norm": 0.32657214999198914,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 2.8305,
      "step": 1560
    },
    {
      "epoch": 0.08486798651921351,
      "eval_loss": 2.8430960178375244,
      "eval_runtime": 21.9941,
      "eval_samples_per_second": 227.334,
      "eval_steps_per_second": 1.228,
      "step": 1566
    },
    {
      "epoch": 0.0850847629854184,
      "grad_norm": 0.5440207719802856,
      "learning_rate": 5.2333333333333336e-05,
      "loss": 2.8311,
      "step": 1570
    },
    {
      "epoch": 0.08562670415093061,
      "grad_norm": 0.43800434470176697,
      "learning_rate": 5.266666666666666e-05,
      "loss": 2.834,
      "step": 1580
    },
    {
      "epoch": 0.08616864531644283,
      "grad_norm": 0.28220343589782715,
      "learning_rate": 5.300000000000001e-05,
      "loss": 2.8264,
      "step": 1590
    },
    {
      "epoch": 0.08671058648195505,
      "grad_norm": 0.5123216509819031,
      "learning_rate": 5.333333333333333e-05,
      "loss": 2.8261,
      "step": 1600
    },
    {
      "epoch": 0.08725252764746727,
      "grad_norm": 0.3581772744655609,
      "learning_rate": 5.3666666666666666e-05,
      "loss": 2.8278,
      "step": 1610
    },
    {
      "epoch": 0.08779446881297949,
      "grad_norm": 0.39921578764915466,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 2.8234,
      "step": 1620
    },
    {
      "epoch": 0.08833640997849171,
      "grad_norm": 0.36354538798332214,
      "learning_rate": 5.433333333333334e-05,
      "loss": 2.8229,
      "step": 1630
    },
    {
      "epoch": 0.08887835114400393,
      "grad_norm": 0.30660825967788696,
      "learning_rate": 5.466666666666666e-05,
      "loss": 2.8225,
      "step": 1640
    },
    {
      "epoch": 0.08942029230951615,
      "grad_norm": 0.30880966782569885,
      "learning_rate": 5.500000000000001e-05,
      "loss": 2.8206,
      "step": 1650
    },
    {
      "epoch": 0.08958287465916981,
      "eval_loss": 2.8423779010772705,
      "eval_runtime": 21.999,
      "eval_samples_per_second": 227.283,
      "eval_steps_per_second": 1.227,
      "step": 1653
    },
    {
      "epoch": 0.08996223347502837,
      "grad_norm": 0.2573600113391876,
      "learning_rate": 5.5333333333333334e-05,
      "loss": 2.8159,
      "step": 1660
    },
    {
      "epoch": 0.09050417464054059,
      "grad_norm": 0.30160731077194214,
      "learning_rate": 5.566666666666667e-05,
      "loss": 2.8184,
      "step": 1670
    },
    {
      "epoch": 0.0910461158060528,
      "grad_norm": 0.2629443109035492,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 2.8192,
      "step": 1680
    },
    {
      "epoch": 0.09158805697156502,
      "grad_norm": 0.23509126901626587,
      "learning_rate": 5.633333333333334e-05,
      "loss": 2.8148,
      "step": 1690
    },
    {
      "epoch": 0.09212999813707724,
      "grad_norm": 0.42312270402908325,
      "learning_rate": 5.666666666666667e-05,
      "loss": 2.813,
      "step": 1700
    },
    {
      "epoch": 0.09267193930258946,
      "grad_norm": 0.46546509861946106,
      "learning_rate": 5.6999999999999996e-05,
      "loss": 2.8086,
      "step": 1710
    },
    {
      "epoch": 0.09321388046810168,
      "grad_norm": 0.23622998595237732,
      "learning_rate": 5.7333333333333336e-05,
      "loss": 2.8111,
      "step": 1720
    },
    {
      "epoch": 0.0937558216336139,
      "grad_norm": 0.607647180557251,
      "learning_rate": 5.766666666666667e-05,
      "loss": 2.8102,
      "step": 1730
    },
    {
      "epoch": 0.09429776279912612,
      "grad_norm": 0.40518462657928467,
      "learning_rate": 5.8e-05,
      "loss": 2.8072,
      "step": 1740
    },
    {
      "epoch": 0.09429776279912612,
      "eval_loss": 2.834738254547119,
      "eval_runtime": 21.6631,
      "eval_samples_per_second": 230.807,
      "eval_steps_per_second": 1.246,
      "step": 1740
    },
    {
      "epoch": 0.09483970396463834,
      "grad_norm": 0.2947821021080017,
      "learning_rate": 5.833333333333334e-05,
      "loss": 2.807,
      "step": 1750
    },
    {
      "epoch": 0.09538164513015056,
      "grad_norm": 0.304720938205719,
      "learning_rate": 5.866666666666667e-05,
      "loss": 2.8086,
      "step": 1760
    },
    {
      "epoch": 0.09592358629566278,
      "grad_norm": 0.37076836824417114,
      "learning_rate": 5.9e-05,
      "loss": 2.8035,
      "step": 1770
    },
    {
      "epoch": 0.096465527461175,
      "grad_norm": 0.4996398389339447,
      "learning_rate": 5.9333333333333343e-05,
      "loss": 2.804,
      "step": 1780
    },
    {
      "epoch": 0.09700746862668722,
      "grad_norm": 0.37258633971214294,
      "learning_rate": 5.966666666666667e-05,
      "loss": 2.8074,
      "step": 1790
    },
    {
      "epoch": 0.09754940979219943,
      "grad_norm": 0.2905193567276001,
      "learning_rate": 6e-05,
      "loss": 2.8006,
      "step": 1800
    },
    {
      "epoch": 0.09809135095771165,
      "grad_norm": 0.27609148621559143,
      "learning_rate": 6.033333333333334e-05,
      "loss": 2.7965,
      "step": 1810
    },
    {
      "epoch": 0.09863329212322387,
      "grad_norm": 0.3159330189228058,
      "learning_rate": 6.066666666666667e-05,
      "loss": 2.7977,
      "step": 1820
    },
    {
      "epoch": 0.09901265093908243,
      "eval_loss": 2.8351926803588867,
      "eval_runtime": 21.9919,
      "eval_samples_per_second": 227.356,
      "eval_steps_per_second": 1.228,
      "step": 1827
    },
    {
      "epoch": 0.09917523328873609,
      "grad_norm": 0.3412172794342041,
      "learning_rate": 6.1e-05,
      "loss": 2.7945,
      "step": 1830
    },
    {
      "epoch": 0.09971717445424831,
      "grad_norm": 0.2713095247745514,
      "learning_rate": 6.133333333333334e-05,
      "loss": 2.7904,
      "step": 1840
    },
    {
      "epoch": 0.10025911561976053,
      "grad_norm": 0.3899465799331665,
      "learning_rate": 6.166666666666667e-05,
      "loss": 2.7907,
      "step": 1850
    },
    {
      "epoch": 0.10080105678527275,
      "grad_norm": 0.2539375424385071,
      "learning_rate": 6.2e-05,
      "loss": 2.7935,
      "step": 1860
    },
    {
      "epoch": 0.10134299795078497,
      "grad_norm": 0.6003013253211975,
      "learning_rate": 6.233333333333334e-05,
      "loss": 2.7854,
      "step": 1870
    },
    {
      "epoch": 0.10188493911629719,
      "grad_norm": 0.410022497177124,
      "learning_rate": 6.266666666666667e-05,
      "loss": 2.7879,
      "step": 1880
    },
    {
      "epoch": 0.10242688028180941,
      "grad_norm": 0.24878162145614624,
      "learning_rate": 6.3e-05,
      "loss": 2.7848,
      "step": 1890
    },
    {
      "epoch": 0.10296882144732163,
      "grad_norm": 0.5053821802139282,
      "learning_rate": 6.333333333333333e-05,
      "loss": 2.7876,
      "step": 1900
    },
    {
      "epoch": 0.10351076261283385,
      "grad_norm": 0.43480825424194336,
      "learning_rate": 6.366666666666668e-05,
      "loss": 2.7826,
      "step": 1910
    },
    {
      "epoch": 0.10372753907903873,
      "eval_loss": 2.837707996368408,
      "eval_runtime": 22.0381,
      "eval_samples_per_second": 226.88,
      "eval_steps_per_second": 1.225,
      "step": 1914
    },
    {
      "epoch": 0.10405270377834606,
      "grad_norm": 0.3400084674358368,
      "learning_rate": 6.400000000000001e-05,
      "loss": 2.7832,
      "step": 1920
    },
    {
      "epoch": 0.10459464494385828,
      "grad_norm": 0.2867070436477661,
      "learning_rate": 6.433333333333333e-05,
      "loss": 2.7821,
      "step": 1930
    },
    {
      "epoch": 0.1051365861093705,
      "grad_norm": 0.3295210003852844,
      "learning_rate": 6.466666666666666e-05,
      "loss": 2.776,
      "step": 1940
    },
    {
      "epoch": 0.10567852727488272,
      "grad_norm": 0.29053670167922974,
      "learning_rate": 6.500000000000001e-05,
      "loss": 2.7776,
      "step": 1950
    },
    {
      "epoch": 0.10622046844039494,
      "grad_norm": 0.4241558313369751,
      "learning_rate": 6.533333333333334e-05,
      "loss": 2.7788,
      "step": 1960
    },
    {
      "epoch": 0.10676240960590716,
      "grad_norm": 0.6300092339515686,
      "learning_rate": 6.566666666666666e-05,
      "loss": 2.7763,
      "step": 1970
    },
    {
      "epoch": 0.10730435077141938,
      "grad_norm": 0.29351305961608887,
      "learning_rate": 6.6e-05,
      "loss": 2.7713,
      "step": 1980
    },
    {
      "epoch": 0.1078462919369316,
      "grad_norm": 0.5574642419815063,
      "learning_rate": 6.633333333333334e-05,
      "loss": 2.7693,
      "step": 1990
    },
    {
      "epoch": 0.10838823310244382,
      "grad_norm": 0.34177178144454956,
      "learning_rate": 6.666666666666667e-05,
      "loss": 2.7661,
      "step": 2000
    },
    {
      "epoch": 0.10844242721899504,
      "eval_loss": 2.8290762901306152,
      "eval_runtime": 21.9941,
      "eval_samples_per_second": 227.334,
      "eval_steps_per_second": 1.228,
      "step": 2001
    },
    {
      "epoch": 0.10893017426795604,
      "grad_norm": 0.5755301117897034,
      "learning_rate": 6.7e-05,
      "loss": 2.7709,
      "step": 2010
    },
    {
      "epoch": 0.10947211543346826,
      "grad_norm": 0.4249889850616455,
      "learning_rate": 6.733333333333333e-05,
      "loss": 2.7636,
      "step": 2020
    },
    {
      "epoch": 0.11001405659898048,
      "grad_norm": 0.3331117033958435,
      "learning_rate": 6.766666666666667e-05,
      "loss": 2.768,
      "step": 2030
    },
    {
      "epoch": 0.1105559977644927,
      "grad_norm": 0.6131373643875122,
      "learning_rate": 6.800000000000001e-05,
      "loss": 2.7659,
      "step": 2040
    },
    {
      "epoch": 0.11109793893000491,
      "grad_norm": 0.36327242851257324,
      "learning_rate": 6.833333333333333e-05,
      "loss": 2.7612,
      "step": 2050
    },
    {
      "epoch": 0.11163988009551713,
      "grad_norm": 0.4649696946144104,
      "learning_rate": 6.866666666666666e-05,
      "loss": 2.7571,
      "step": 2060
    },
    {
      "epoch": 0.11218182126102935,
      "grad_norm": 0.4964756965637207,
      "learning_rate": 6.9e-05,
      "loss": 2.7529,
      "step": 2070
    },
    {
      "epoch": 0.11272376242654157,
      "grad_norm": 0.261391818523407,
      "learning_rate": 6.933333333333334e-05,
      "loss": 2.7493,
      "step": 2080
    },
    {
      "epoch": 0.11315731535895135,
      "eval_loss": 2.822695732116699,
      "eval_runtime": 21.997,
      "eval_samples_per_second": 227.304,
      "eval_steps_per_second": 1.227,
      "step": 2088
    },
    {
      "epoch": 0.11326570359205379,
      "grad_norm": 0.4503042995929718,
      "learning_rate": 6.966666666666668e-05,
      "loss": 2.7474,
      "step": 2090
    },
    {
      "epoch": 0.11380764475756601,
      "grad_norm": 0.5234399437904358,
      "learning_rate": 7e-05,
      "loss": 2.7477,
      "step": 2100
    },
    {
      "epoch": 0.11434958592307823,
      "grad_norm": 0.4587494432926178,
      "learning_rate": 7.033333333333334e-05,
      "loss": 2.7441,
      "step": 2110
    },
    {
      "epoch": 0.11489152708859045,
      "grad_norm": 0.3193627893924713,
      "learning_rate": 7.066666666666667e-05,
      "loss": 2.7529,
      "step": 2120
    },
    {
      "epoch": 0.11543346825410267,
      "grad_norm": 0.60213702917099,
      "learning_rate": 7.1e-05,
      "loss": 2.7452,
      "step": 2130
    },
    {
      "epoch": 0.11597540941961489,
      "grad_norm": 0.495370477437973,
      "learning_rate": 7.133333333333334e-05,
      "loss": 2.7431,
      "step": 2140
    },
    {
      "epoch": 0.1165173505851271,
      "grad_norm": 0.46439045667648315,
      "learning_rate": 7.166666666666667e-05,
      "loss": 2.7463,
      "step": 2150
    },
    {
      "epoch": 0.11705929175063932,
      "grad_norm": 0.4808029234409332,
      "learning_rate": 7.2e-05,
      "loss": 2.7399,
      "step": 2160
    },
    {
      "epoch": 0.11760123291615154,
      "grad_norm": 0.40961867570877075,
      "learning_rate": 7.233333333333335e-05,
      "loss": 2.7369,
      "step": 2170
    },
    {
      "epoch": 0.11787220349890765,
      "eval_loss": 2.823870897293091,
      "eval_runtime": 21.9942,
      "eval_samples_per_second": 227.333,
      "eval_steps_per_second": 1.228,
      "step": 2175
    },
    {
      "epoch": 0.11814317408166376,
      "grad_norm": 0.4309645891189575,
      "learning_rate": 7.266666666666667e-05,
      "loss": 2.7333,
      "step": 2180
    },
    {
      "epoch": 0.11868511524717598,
      "grad_norm": 0.46611249446868896,
      "learning_rate": 7.3e-05,
      "loss": 2.7364,
      "step": 2190
    },
    {
      "epoch": 0.1192270564126882,
      "grad_norm": 0.3425257205963135,
      "learning_rate": 7.333333333333333e-05,
      "loss": 2.7344,
      "step": 2200
    },
    {
      "epoch": 0.11976899757820042,
      "grad_norm": 0.6439690589904785,
      "learning_rate": 7.366666666666668e-05,
      "loss": 2.7389,
      "step": 2210
    },
    {
      "epoch": 0.12031093874371264,
      "grad_norm": 0.9094191193580627,
      "learning_rate": 7.4e-05,
      "loss": 2.727,
      "step": 2220
    },
    {
      "epoch": 0.12085287990922486,
      "grad_norm": 0.4872890114784241,
      "learning_rate": 7.433333333333333e-05,
      "loss": 2.7286,
      "step": 2230
    },
    {
      "epoch": 0.12139482107473708,
      "grad_norm": 0.4804534614086151,
      "learning_rate": 7.466666666666667e-05,
      "loss": 2.7233,
      "step": 2240
    },
    {
      "epoch": 0.1219367622402493,
      "grad_norm": 0.330377995967865,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.7211,
      "step": 2250
    },
    {
      "epoch": 0.12247870340576152,
      "grad_norm": 0.9292433857917786,
      "learning_rate": 7.533333333333334e-05,
      "loss": 2.7224,
      "step": 2260
    },
    {
      "epoch": 0.12258709163886396,
      "eval_loss": 2.818483352661133,
      "eval_runtime": 21.9931,
      "eval_samples_per_second": 227.344,
      "eval_steps_per_second": 1.228,
      "step": 2262
    },
    {
      "epoch": 0.12302064457127374,
      "grad_norm": 0.4465082287788391,
      "learning_rate": 7.566666666666667e-05,
      "loss": 2.7214,
      "step": 2270
    },
    {
      "epoch": 0.12356258573678595,
      "grad_norm": 0.300624817609787,
      "learning_rate": 7.6e-05,
      "loss": 2.722,
      "step": 2280
    },
    {
      "epoch": 0.12410452690229817,
      "grad_norm": 0.41431039571762085,
      "learning_rate": 7.633333333333334e-05,
      "loss": 2.7135,
      "step": 2290
    },
    {
      "epoch": 0.12464646806781039,
      "grad_norm": 0.3542834520339966,
      "learning_rate": 7.666666666666667e-05,
      "loss": 2.7118,
      "step": 2300
    },
    {
      "epoch": 0.1251884092333226,
      "grad_norm": 0.42152753472328186,
      "learning_rate": 7.7e-05,
      "loss": 2.7193,
      "step": 2310
    },
    {
      "epoch": 0.12573035039883482,
      "grad_norm": 0.4453124701976776,
      "learning_rate": 7.733333333333333e-05,
      "loss": 2.7154,
      "step": 2320
    },
    {
      "epoch": 0.12627229156434705,
      "grad_norm": 0.6320910453796387,
      "learning_rate": 7.766666666666667e-05,
      "loss": 2.7098,
      "step": 2330
    },
    {
      "epoch": 0.12681423272985926,
      "grad_norm": 0.3247159421443939,
      "learning_rate": 7.800000000000001e-05,
      "loss": 2.7067,
      "step": 2340
    },
    {
      "epoch": 0.12730197977882027,
      "eval_loss": 2.813124179840088,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.323,
      "eval_steps_per_second": 1.228,
      "step": 2349
    },
    {
      "epoch": 0.1273561738953715,
      "grad_norm": 0.34726181626319885,
      "learning_rate": 7.833333333333333e-05,
      "loss": 2.7054,
      "step": 2350
    },
    {
      "epoch": 0.1278981150608837,
      "grad_norm": 0.5048098564147949,
      "learning_rate": 7.866666666666666e-05,
      "loss": 2.7063,
      "step": 2360
    },
    {
      "epoch": 0.12844005622639593,
      "grad_norm": 0.4321523606777191,
      "learning_rate": 7.900000000000001e-05,
      "loss": 2.7011,
      "step": 2370
    },
    {
      "epoch": 0.12898199739190813,
      "grad_norm": 0.2796650826931,
      "learning_rate": 7.933333333333334e-05,
      "loss": 2.6997,
      "step": 2380
    },
    {
      "epoch": 0.12952393855742036,
      "grad_norm": 0.5107274651527405,
      "learning_rate": 7.966666666666666e-05,
      "loss": 2.6933,
      "step": 2390
    },
    {
      "epoch": 0.13006587972293257,
      "grad_norm": 0.6369110345840454,
      "learning_rate": 8e-05,
      "loss": 2.6989,
      "step": 2400
    },
    {
      "epoch": 0.1306078208884448,
      "grad_norm": 0.6331022381782532,
      "learning_rate": 8.033333333333334e-05,
      "loss": 2.6923,
      "step": 2410
    },
    {
      "epoch": 0.131149762053957,
      "grad_norm": 0.36579135060310364,
      "learning_rate": 8.066666666666667e-05,
      "loss": 2.6889,
      "step": 2420
    },
    {
      "epoch": 0.13169170321946924,
      "grad_norm": 0.9732735753059387,
      "learning_rate": 8.1e-05,
      "loss": 2.6948,
      "step": 2430
    },
    {
      "epoch": 0.13201686791877656,
      "eval_loss": 2.808922290802002,
      "eval_runtime": 21.9927,
      "eval_samples_per_second": 227.348,
      "eval_steps_per_second": 1.228,
      "step": 2436
    },
    {
      "epoch": 0.13223364438498145,
      "grad_norm": 0.5418347120285034,
      "learning_rate": 8.133333333333334e-05,
      "loss": 2.6871,
      "step": 2440
    },
    {
      "epoch": 0.13277558555049368,
      "grad_norm": 0.43290242552757263,
      "learning_rate": 8.166666666666667e-05,
      "loss": 2.6827,
      "step": 2450
    },
    {
      "epoch": 0.13331752671600589,
      "grad_norm": 0.6338348984718323,
      "learning_rate": 8.2e-05,
      "loss": 2.6821,
      "step": 2460
    },
    {
      "epoch": 0.13385946788151812,
      "grad_norm": 0.604308545589447,
      "learning_rate": 8.233333333333333e-05,
      "loss": 2.6774,
      "step": 2470
    },
    {
      "epoch": 0.13440140904703032,
      "grad_norm": 0.49863043427467346,
      "learning_rate": 8.266666666666667e-05,
      "loss": 2.6781,
      "step": 2480
    },
    {
      "epoch": 0.13494335021254256,
      "grad_norm": 0.9325143098831177,
      "learning_rate": 8.3e-05,
      "loss": 2.6793,
      "step": 2490
    },
    {
      "epoch": 0.13548529137805476,
      "grad_norm": 0.5320800542831421,
      "learning_rate": 8.333333333333334e-05,
      "loss": 2.6797,
      "step": 2500
    },
    {
      "epoch": 0.136027232543567,
      "grad_norm": 0.3204202353954315,
      "learning_rate": 8.366666666666668e-05,
      "loss": 2.6757,
      "step": 2510
    },
    {
      "epoch": 0.1365691737090792,
      "grad_norm": 0.2703516483306885,
      "learning_rate": 8.4e-05,
      "loss": 2.6667,
      "step": 2520
    },
    {
      "epoch": 0.13673175605873286,
      "eval_loss": 2.8014180660247803,
      "eval_runtime": 21.9979,
      "eval_samples_per_second": 227.295,
      "eval_steps_per_second": 1.227,
      "step": 2523
    },
    {
      "epoch": 0.13711111487459143,
      "grad_norm": 0.4166392385959625,
      "learning_rate": 8.433333333333334e-05,
      "loss": 2.672,
      "step": 2530
    },
    {
      "epoch": 0.13765305604010364,
      "grad_norm": 0.6374923586845398,
      "learning_rate": 8.466666666666667e-05,
      "loss": 2.666,
      "step": 2540
    },
    {
      "epoch": 0.13819499720561587,
      "grad_norm": 0.41108188033103943,
      "learning_rate": 8.5e-05,
      "loss": 2.6661,
      "step": 2550
    },
    {
      "epoch": 0.13873693837112808,
      "grad_norm": 0.6000506281852722,
      "learning_rate": 8.533333333333334e-05,
      "loss": 2.6677,
      "step": 2560
    },
    {
      "epoch": 0.1392788795366403,
      "grad_norm": 0.3874584436416626,
      "learning_rate": 8.566666666666667e-05,
      "loss": 2.6664,
      "step": 2570
    },
    {
      "epoch": 0.13982082070215252,
      "grad_norm": 1.0246257781982422,
      "learning_rate": 8.6e-05,
      "loss": 2.6662,
      "step": 2580
    },
    {
      "epoch": 0.14036276186766475,
      "grad_norm": 0.6086533069610596,
      "learning_rate": 8.633333333333334e-05,
      "loss": 2.6615,
      "step": 2590
    },
    {
      "epoch": 0.14090470303317695,
      "grad_norm": 0.2854156792163849,
      "learning_rate": 8.666666666666667e-05,
      "loss": 2.6512,
      "step": 2600
    },
    {
      "epoch": 0.1414466441986892,
      "grad_norm": 0.6454458236694336,
      "learning_rate": 8.7e-05,
      "loss": 2.6608,
      "step": 2610
    },
    {
      "epoch": 0.1414466441986892,
      "eval_loss": 2.8015899658203125,
      "eval_runtime": 21.9849,
      "eval_samples_per_second": 227.429,
      "eval_steps_per_second": 1.228,
      "step": 2610
    },
    {
      "epoch": 0.1419885853642014,
      "grad_norm": 0.553676426410675,
      "learning_rate": 8.733333333333333e-05,
      "loss": 2.6523,
      "step": 2620
    },
    {
      "epoch": 0.14253052652971362,
      "grad_norm": 0.6649917364120483,
      "learning_rate": 8.766666666666668e-05,
      "loss": 2.6496,
      "step": 2630
    },
    {
      "epoch": 0.14307246769522583,
      "grad_norm": 0.3380162715911865,
      "learning_rate": 8.800000000000001e-05,
      "loss": 2.6456,
      "step": 2640
    },
    {
      "epoch": 0.14361440886073806,
      "grad_norm": 0.4091057777404785,
      "learning_rate": 8.833333333333333e-05,
      "loss": 2.6397,
      "step": 2650
    },
    {
      "epoch": 0.14415635002625027,
      "grad_norm": 0.5197025537490845,
      "learning_rate": 8.866666666666668e-05,
      "loss": 2.6446,
      "step": 2660
    },
    {
      "epoch": 0.1446982911917625,
      "grad_norm": 0.3578343093395233,
      "learning_rate": 8.900000000000001e-05,
      "loss": 2.6438,
      "step": 2670
    },
    {
      "epoch": 0.1452402323572747,
      "grad_norm": 0.7891026735305786,
      "learning_rate": 8.933333333333334e-05,
      "loss": 2.6442,
      "step": 2680
    },
    {
      "epoch": 0.14578217352278694,
      "grad_norm": 0.49294915795326233,
      "learning_rate": 8.966666666666666e-05,
      "loss": 2.6435,
      "step": 2690
    },
    {
      "epoch": 0.14616153233864548,
      "eval_loss": 2.7972233295440674,
      "eval_runtime": 21.9971,
      "eval_samples_per_second": 227.303,
      "eval_steps_per_second": 1.227,
      "step": 2697
    },
    {
      "epoch": 0.14632411468829915,
      "grad_norm": 0.5119202136993408,
      "learning_rate": 9e-05,
      "loss": 2.6435,
      "step": 2700
    },
    {
      "epoch": 0.14686605585381138,
      "grad_norm": 0.5359740853309631,
      "learning_rate": 9.033333333333334e-05,
      "loss": 2.6324,
      "step": 2710
    },
    {
      "epoch": 0.14740799701932358,
      "grad_norm": 0.6155476570129395,
      "learning_rate": 9.066666666666667e-05,
      "loss": 2.6331,
      "step": 2720
    },
    {
      "epoch": 0.14794993818483582,
      "grad_norm": 0.9409157633781433,
      "learning_rate": 9.1e-05,
      "loss": 2.6363,
      "step": 2730
    },
    {
      "epoch": 0.14849187935034802,
      "grad_norm": 0.3532879650592804,
      "learning_rate": 9.133333333333334e-05,
      "loss": 2.6232,
      "step": 2740
    },
    {
      "epoch": 0.14903382051586025,
      "grad_norm": 0.38057300448417664,
      "learning_rate": 9.166666666666667e-05,
      "loss": 2.6234,
      "step": 2750
    },
    {
      "epoch": 0.14957576168137246,
      "grad_norm": 0.6639463901519775,
      "learning_rate": 9.200000000000001e-05,
      "loss": 2.6245,
      "step": 2760
    },
    {
      "epoch": 0.1501177028468847,
      "grad_norm": 0.6744067668914795,
      "learning_rate": 9.233333333333333e-05,
      "loss": 2.6265,
      "step": 2770
    },
    {
      "epoch": 0.1506596440123969,
      "grad_norm": 0.5296260714530945,
      "learning_rate": 9.266666666666666e-05,
      "loss": 2.6255,
      "step": 2780
    },
    {
      "epoch": 0.15087642047860178,
      "eval_loss": 2.7909934520721436,
      "eval_runtime": 21.9988,
      "eval_samples_per_second": 227.285,
      "eval_steps_per_second": 1.227,
      "step": 2784
    },
    {
      "epoch": 0.15120158517790913,
      "grad_norm": 1.1150288581848145,
      "learning_rate": 9.300000000000001e-05,
      "loss": 2.6247,
      "step": 2790
    },
    {
      "epoch": 0.15174352634342134,
      "grad_norm": 0.3209221661090851,
      "learning_rate": 9.333333333333334e-05,
      "loss": 2.6256,
      "step": 2800
    },
    {
      "epoch": 0.15228546750893357,
      "grad_norm": 0.7962296605110168,
      "learning_rate": 9.366666666666668e-05,
      "loss": 2.6113,
      "step": 2810
    },
    {
      "epoch": 0.15282740867444577,
      "grad_norm": 0.345284640789032,
      "learning_rate": 9.4e-05,
      "loss": 2.6234,
      "step": 2820
    },
    {
      "epoch": 0.153369349839958,
      "grad_norm": 0.3219437003135681,
      "learning_rate": 9.433333333333334e-05,
      "loss": 2.619,
      "step": 2830
    },
    {
      "epoch": 0.1539112910054702,
      "grad_norm": 0.3087829351425171,
      "learning_rate": 9.466666666666667e-05,
      "loss": 2.605,
      "step": 2840
    },
    {
      "epoch": 0.15445323217098245,
      "grad_norm": 0.3032431900501251,
      "learning_rate": 9.5e-05,
      "loss": 2.6083,
      "step": 2850
    },
    {
      "epoch": 0.15499517333649465,
      "grad_norm": 0.3768031597137451,
      "learning_rate": 9.533333333333334e-05,
      "loss": 2.6076,
      "step": 2860
    },
    {
      "epoch": 0.15553711450200688,
      "grad_norm": 0.37252676486968994,
      "learning_rate": 9.566666666666667e-05,
      "loss": 2.6045,
      "step": 2870
    },
    {
      "epoch": 0.1555913086185581,
      "eval_loss": 2.7903635501861572,
      "eval_runtime": 21.9938,
      "eval_samples_per_second": 227.337,
      "eval_steps_per_second": 1.228,
      "step": 2871
    },
    {
      "epoch": 0.1560790556675191,
      "grad_norm": 0.9097828269004822,
      "learning_rate": 9.6e-05,
      "loss": 2.6135,
      "step": 2880
    },
    {
      "epoch": 0.15662099683303132,
      "grad_norm": 0.5291810631752014,
      "learning_rate": 9.633333333333335e-05,
      "loss": 2.5981,
      "step": 2890
    },
    {
      "epoch": 0.15716293799854353,
      "grad_norm": 0.7748283743858337,
      "learning_rate": 9.666666666666667e-05,
      "loss": 2.5933,
      "step": 2900
    },
    {
      "epoch": 0.15770487916405576,
      "grad_norm": 0.4185350239276886,
      "learning_rate": 9.7e-05,
      "loss": 2.5929,
      "step": 2910
    },
    {
      "epoch": 0.15824682032956797,
      "grad_norm": 0.32910650968551636,
      "learning_rate": 9.733333333333335e-05,
      "loss": 2.5889,
      "step": 2920
    },
    {
      "epoch": 0.1587887614950802,
      "grad_norm": 0.32221078872680664,
      "learning_rate": 9.766666666666668e-05,
      "loss": 2.5938,
      "step": 2930
    },
    {
      "epoch": 0.1593307026605924,
      "grad_norm": 0.6913098096847534,
      "learning_rate": 9.8e-05,
      "loss": 2.5899,
      "step": 2940
    },
    {
      "epoch": 0.15987264382610464,
      "grad_norm": 1.0831592082977295,
      "learning_rate": 9.833333333333333e-05,
      "loss": 2.5864,
      "step": 2950
    },
    {
      "epoch": 0.1603061967585144,
      "eval_loss": 2.782402276992798,
      "eval_runtime": 21.9947,
      "eval_samples_per_second": 227.328,
      "eval_steps_per_second": 1.228,
      "step": 2958
    },
    {
      "epoch": 0.16041458499161684,
      "grad_norm": 0.6254389882087708,
      "learning_rate": 9.866666666666668e-05,
      "loss": 2.5797,
      "step": 2960
    },
    {
      "epoch": 0.16095652615712908,
      "grad_norm": 0.34385034441947937,
      "learning_rate": 9.900000000000001e-05,
      "loss": 2.5846,
      "step": 2970
    },
    {
      "epoch": 0.16149846732264128,
      "grad_norm": 0.7574624419212341,
      "learning_rate": 9.933333333333334e-05,
      "loss": 2.5805,
      "step": 2980
    },
    {
      "epoch": 0.16204040848815351,
      "grad_norm": 0.36409491300582886,
      "learning_rate": 9.966666666666667e-05,
      "loss": 2.5852,
      "step": 2990
    },
    {
      "epoch": 0.16258234965366572,
      "grad_norm": 0.7688263654708862,
      "learning_rate": 0.0001,
      "loss": 2.585,
      "step": 3000
    },
    {
      "epoch": 0.16312429081917795,
      "grad_norm": 0.9283438920974731,
      "learning_rate": 9.99999069936256e-05,
      "loss": 2.582,
      "step": 3010
    },
    {
      "epoch": 0.16366623198469016,
      "grad_norm": 0.5236392617225647,
      "learning_rate": 9.999962797488683e-05,
      "loss": 2.5818,
      "step": 3020
    },
    {
      "epoch": 0.1642081731502024,
      "grad_norm": 0.4604909121990204,
      "learning_rate": 9.999916294493705e-05,
      "loss": 2.574,
      "step": 3030
    },
    {
      "epoch": 0.1647501143157146,
      "grad_norm": 0.5703723430633545,
      "learning_rate": 9.999851190569852e-05,
      "loss": 2.5638,
      "step": 3040
    },
    {
      "epoch": 0.1650210848984707,
      "eval_loss": 2.777949571609497,
      "eval_runtime": 21.9996,
      "eval_samples_per_second": 227.277,
      "eval_steps_per_second": 1.227,
      "step": 3045
    },
    {
      "epoch": 0.16529205548122683,
      "grad_norm": 1.0797618627548218,
      "learning_rate": 9.99976748598624e-05,
      "loss": 2.5687,
      "step": 3050
    },
    {
      "epoch": 0.16583399664673903,
      "grad_norm": 0.7626991271972656,
      "learning_rate": 9.999665181088869e-05,
      "loss": 2.5726,
      "step": 3060
    },
    {
      "epoch": 0.16637593781225127,
      "grad_norm": 0.46589240431785583,
      "learning_rate": 9.999544276300629e-05,
      "loss": 2.5625,
      "step": 3070
    },
    {
      "epoch": 0.16691787897776347,
      "grad_norm": 0.4888695478439331,
      "learning_rate": 9.999404772121297e-05,
      "loss": 2.5547,
      "step": 3080
    },
    {
      "epoch": 0.1674598201432757,
      "grad_norm": 1.3562276363372803,
      "learning_rate": 9.999246669127524e-05,
      "loss": 2.5644,
      "step": 3090
    },
    {
      "epoch": 0.1680017613087879,
      "grad_norm": 0.3566310405731201,
      "learning_rate": 9.999069967972854e-05,
      "loss": 2.5584,
      "step": 3100
    },
    {
      "epoch": 0.16854370247430014,
      "grad_norm": 0.7351515293121338,
      "learning_rate": 9.998874669387696e-05,
      "loss": 2.5589,
      "step": 3110
    },
    {
      "epoch": 0.16908564363981235,
      "grad_norm": 0.5617663264274597,
      "learning_rate": 9.998660774179343e-05,
      "loss": 2.566,
      "step": 3120
    },
    {
      "epoch": 0.16962758480532458,
      "grad_norm": 0.7777019143104553,
      "learning_rate": 9.998428283231952e-05,
      "loss": 2.5553,
      "step": 3130
    },
    {
      "epoch": 0.16973597303842702,
      "eval_loss": 2.773970127105713,
      "eval_runtime": 21.9948,
      "eval_samples_per_second": 227.326,
      "eval_steps_per_second": 1.228,
      "step": 3132
    },
    {
      "epoch": 0.1701695259708368,
      "grad_norm": 0.8600534796714783,
      "learning_rate": 9.998177197506557e-05,
      "loss": 2.5529,
      "step": 3140
    },
    {
      "epoch": 0.17071146713634902,
      "grad_norm": 0.3683708906173706,
      "learning_rate": 9.997907518041047e-05,
      "loss": 2.5552,
      "step": 3150
    },
    {
      "epoch": 0.17125340830186123,
      "grad_norm": 0.5638754367828369,
      "learning_rate": 9.997619245950172e-05,
      "loss": 2.5465,
      "step": 3160
    },
    {
      "epoch": 0.17179534946737346,
      "grad_norm": 0.4635675251483917,
      "learning_rate": 9.997312382425543e-05,
      "loss": 2.5419,
      "step": 3170
    },
    {
      "epoch": 0.17233729063288566,
      "grad_norm": 0.48697274923324585,
      "learning_rate": 9.99698692873561e-05,
      "loss": 2.5409,
      "step": 3180
    },
    {
      "epoch": 0.1728792317983979,
      "grad_norm": 0.48942264914512634,
      "learning_rate": 9.99664288622568e-05,
      "loss": 2.5347,
      "step": 3190
    },
    {
      "epoch": 0.1734211729639101,
      "grad_norm": 0.3773297369480133,
      "learning_rate": 9.996280256317887e-05,
      "loss": 2.5432,
      "step": 3200
    },
    {
      "epoch": 0.17396311412942234,
      "grad_norm": 0.8151489496231079,
      "learning_rate": 9.995899040511207e-05,
      "loss": 2.5359,
      "step": 3210
    },
    {
      "epoch": 0.17445086117838332,
      "eval_loss": 2.7794876098632812,
      "eval_runtime": 21.6952,
      "eval_samples_per_second": 230.465,
      "eval_steps_per_second": 1.245,
      "step": 3219
    },
    {
      "epoch": 0.17450505529493454,
      "grad_norm": 0.6645236015319824,
      "learning_rate": 9.995499240381441e-05,
      "loss": 2.5412,
      "step": 3220
    },
    {
      "epoch": 0.17504699646044677,
      "grad_norm": 0.31391623616218567,
      "learning_rate": 9.995080857581208e-05,
      "loss": 2.5403,
      "step": 3230
    },
    {
      "epoch": 0.17558893762595898,
      "grad_norm": 0.6431202292442322,
      "learning_rate": 9.994643893839943e-05,
      "loss": 2.5346,
      "step": 3240
    },
    {
      "epoch": 0.1761308787914712,
      "grad_norm": 0.6171109676361084,
      "learning_rate": 9.994188350963887e-05,
      "loss": 2.5362,
      "step": 3250
    },
    {
      "epoch": 0.17667281995698342,
      "grad_norm": 0.3944205343723297,
      "learning_rate": 9.993714230836076e-05,
      "loss": 2.5302,
      "step": 3260
    },
    {
      "epoch": 0.17721476112249565,
      "grad_norm": 0.462694376707077,
      "learning_rate": 9.993221535416346e-05,
      "loss": 2.5301,
      "step": 3270
    },
    {
      "epoch": 0.17775670228800786,
      "grad_norm": 0.39359036087989807,
      "learning_rate": 9.992710266741307e-05,
      "loss": 2.5236,
      "step": 3280
    },
    {
      "epoch": 0.1782986434535201,
      "grad_norm": 0.6656551361083984,
      "learning_rate": 9.99218042692435e-05,
      "loss": 2.5282,
      "step": 3290
    },
    {
      "epoch": 0.1788405846190323,
      "grad_norm": 0.36592236161231995,
      "learning_rate": 9.991632018155627e-05,
      "loss": 2.5288,
      "step": 3300
    },
    {
      "epoch": 0.17916574931833962,
      "eval_loss": 2.763190746307373,
      "eval_runtime": 21.9963,
      "eval_samples_per_second": 227.311,
      "eval_steps_per_second": 1.227,
      "step": 3306
    },
    {
      "epoch": 0.17938252578454453,
      "grad_norm": 1.2581602334976196,
      "learning_rate": 9.991065042702054e-05,
      "loss": 2.5198,
      "step": 3310
    },
    {
      "epoch": 0.17992446695005673,
      "grad_norm": 0.9863691926002502,
      "learning_rate": 9.990479502907287e-05,
      "loss": 2.5248,
      "step": 3320
    },
    {
      "epoch": 0.18046640811556897,
      "grad_norm": 0.538944661617279,
      "learning_rate": 9.989875401191725e-05,
      "loss": 2.5226,
      "step": 3330
    },
    {
      "epoch": 0.18100834928108117,
      "grad_norm": 0.8944976925849915,
      "learning_rate": 9.989252740052489e-05,
      "loss": 2.512,
      "step": 3340
    },
    {
      "epoch": 0.1815502904465934,
      "grad_norm": 0.5679588913917542,
      "learning_rate": 9.988611522063423e-05,
      "loss": 2.5097,
      "step": 3350
    },
    {
      "epoch": 0.1820922316121056,
      "grad_norm": 0.40262141823768616,
      "learning_rate": 9.987951749875081e-05,
      "loss": 2.509,
      "step": 3360
    },
    {
      "epoch": 0.18263417277761784,
      "grad_norm": 0.3973683714866638,
      "learning_rate": 9.987273426214702e-05,
      "loss": 2.5156,
      "step": 3370
    },
    {
      "epoch": 0.18317611394313005,
      "grad_norm": 0.4501461982727051,
      "learning_rate": 9.98657655388622e-05,
      "loss": 2.5193,
      "step": 3380
    },
    {
      "epoch": 0.18371805510864228,
      "grad_norm": 0.3253689110279083,
      "learning_rate": 9.985861135770237e-05,
      "loss": 2.5033,
      "step": 3390
    },
    {
      "epoch": 0.18388063745829594,
      "eval_loss": 2.767366647720337,
      "eval_runtime": 22.0,
      "eval_samples_per_second": 227.273,
      "eval_steps_per_second": 1.227,
      "step": 3393
    },
    {
      "epoch": 0.18425999627415449,
      "grad_norm": 0.737877368927002,
      "learning_rate": 9.985127174824017e-05,
      "loss": 2.5085,
      "step": 3400
    },
    {
      "epoch": 0.18480193743966672,
      "grad_norm": 1.0722659826278687,
      "learning_rate": 9.984374674081472e-05,
      "loss": 2.4993,
      "step": 3410
    },
    {
      "epoch": 0.18534387860517892,
      "grad_norm": 0.7790729403495789,
      "learning_rate": 9.983603636653154e-05,
      "loss": 2.4982,
      "step": 3420
    },
    {
      "epoch": 0.18588581977069116,
      "grad_norm": 0.3918309211730957,
      "learning_rate": 9.982814065726233e-05,
      "loss": 2.4979,
      "step": 3430
    },
    {
      "epoch": 0.18642776093620336,
      "grad_norm": 0.31229421496391296,
      "learning_rate": 9.982005964564495e-05,
      "loss": 2.4927,
      "step": 3440
    },
    {
      "epoch": 0.1869697021017156,
      "grad_norm": 0.4041496813297272,
      "learning_rate": 9.981179336508322e-05,
      "loss": 2.4964,
      "step": 3450
    },
    {
      "epoch": 0.1875116432672278,
      "grad_norm": 0.3821089267730713,
      "learning_rate": 9.980334184974672e-05,
      "loss": 2.4913,
      "step": 3460
    },
    {
      "epoch": 0.18805358443274003,
      "grad_norm": 0.4665864408016205,
      "learning_rate": 9.979470513457084e-05,
      "loss": 2.487,
      "step": 3470
    },
    {
      "epoch": 0.18859552559825224,
      "grad_norm": 0.509825587272644,
      "learning_rate": 9.978588325525639e-05,
      "loss": 2.4932,
      "step": 3480
    },
    {
      "epoch": 0.18859552559825224,
      "eval_loss": 2.7670769691467285,
      "eval_runtime": 21.9697,
      "eval_samples_per_second": 227.587,
      "eval_steps_per_second": 1.229,
      "step": 3480
    },
    {
      "epoch": 0.18913746676376447,
      "grad_norm": 1.2350674867630005,
      "learning_rate": 9.977687624826966e-05,
      "loss": 2.4936,
      "step": 3490
    },
    {
      "epoch": 0.18967940792927668,
      "grad_norm": 0.7966161370277405,
      "learning_rate": 9.97676841508422e-05,
      "loss": 2.4861,
      "step": 3500
    },
    {
      "epoch": 0.1902213490947889,
      "grad_norm": 0.3995961844921112,
      "learning_rate": 9.975830700097056e-05,
      "loss": 2.488,
      "step": 3510
    },
    {
      "epoch": 0.19076329026030112,
      "grad_norm": 0.3042275309562683,
      "learning_rate": 9.974874483741632e-05,
      "loss": 2.4736,
      "step": 3520
    },
    {
      "epoch": 0.19130523142581332,
      "grad_norm": 0.31050169467926025,
      "learning_rate": 9.973899769970578e-05,
      "loss": 2.4825,
      "step": 3530
    },
    {
      "epoch": 0.19184717259132555,
      "grad_norm": 0.323839008808136,
      "learning_rate": 9.972906562812986e-05,
      "loss": 2.4788,
      "step": 3540
    },
    {
      "epoch": 0.19238911375683776,
      "grad_norm": 0.8257955312728882,
      "learning_rate": 9.971894866374397e-05,
      "loss": 2.4857,
      "step": 3550
    },
    {
      "epoch": 0.19293105492235,
      "grad_norm": 0.547508955001831,
      "learning_rate": 9.970864684836776e-05,
      "loss": 2.4832,
      "step": 3560
    },
    {
      "epoch": 0.19331041373820854,
      "eval_loss": 2.761841297149658,
      "eval_runtime": 21.9898,
      "eval_samples_per_second": 227.379,
      "eval_steps_per_second": 1.228,
      "step": 3567
    },
    {
      "epoch": 0.1934729960878622,
      "grad_norm": 0.7686718106269836,
      "learning_rate": 9.969816022458495e-05,
      "loss": 2.4779,
      "step": 3570
    },
    {
      "epoch": 0.19401493725337443,
      "grad_norm": 0.3649592995643616,
      "learning_rate": 9.968748883574324e-05,
      "loss": 2.4733,
      "step": 3580
    },
    {
      "epoch": 0.19455687841888664,
      "grad_norm": 0.7720659375190735,
      "learning_rate": 9.967663272595408e-05,
      "loss": 2.465,
      "step": 3590
    },
    {
      "epoch": 0.19509881958439887,
      "grad_norm": 0.47627437114715576,
      "learning_rate": 9.966559194009244e-05,
      "loss": 2.4662,
      "step": 3600
    },
    {
      "epoch": 0.19564076074991107,
      "grad_norm": 0.3539658486843109,
      "learning_rate": 9.965436652379671e-05,
      "loss": 2.4693,
      "step": 3610
    },
    {
      "epoch": 0.1961827019154233,
      "grad_norm": 1.2727680206298828,
      "learning_rate": 9.964295652346844e-05,
      "loss": 2.465,
      "step": 3620
    },
    {
      "epoch": 0.1967246430809355,
      "grad_norm": 0.34611645340919495,
      "learning_rate": 9.963136198627224e-05,
      "loss": 2.4634,
      "step": 3630
    },
    {
      "epoch": 0.19726658424644775,
      "grad_norm": 0.7994367480278015,
      "learning_rate": 9.961958296013543e-05,
      "loss": 2.4702,
      "step": 3640
    },
    {
      "epoch": 0.19780852541195995,
      "grad_norm": 1.1131157875061035,
      "learning_rate": 9.960761949374802e-05,
      "loss": 2.4639,
      "step": 3650
    },
    {
      "epoch": 0.19802530187816486,
      "eval_loss": 2.7547664642333984,
      "eval_runtime": 21.9926,
      "eval_samples_per_second": 227.35,
      "eval_steps_per_second": 1.228,
      "step": 3654
    },
    {
      "epoch": 0.19835046657747218,
      "grad_norm": 1.1642026901245117,
      "learning_rate": 9.959547163656238e-05,
      "loss": 2.4701,
      "step": 3660
    },
    {
      "epoch": 0.1988924077429844,
      "grad_norm": 0.38460174202919006,
      "learning_rate": 9.958313943879311e-05,
      "loss": 2.4701,
      "step": 3670
    },
    {
      "epoch": 0.19943434890849662,
      "grad_norm": 0.4307605028152466,
      "learning_rate": 9.957062295141675e-05,
      "loss": 2.46,
      "step": 3680
    },
    {
      "epoch": 0.19997629007400883,
      "grad_norm": 0.952154815196991,
      "learning_rate": 9.955792222617171e-05,
      "loss": 2.4643,
      "step": 3690
    },
    {
      "epoch": 0.20051823123952106,
      "grad_norm": 0.3698415756225586,
      "learning_rate": 9.95450373155579e-05,
      "loss": 2.4557,
      "step": 3700
    },
    {
      "epoch": 0.20106017240503327,
      "grad_norm": 0.3291034996509552,
      "learning_rate": 9.953196827283659e-05,
      "loss": 2.4598,
      "step": 3710
    },
    {
      "epoch": 0.2016021135705455,
      "grad_norm": 0.31008443236351013,
      "learning_rate": 9.95187151520302e-05,
      "loss": 2.4511,
      "step": 3720
    },
    {
      "epoch": 0.2021440547360577,
      "grad_norm": 0.5721209645271301,
      "learning_rate": 9.950527800792205e-05,
      "loss": 2.4536,
      "step": 3730
    },
    {
      "epoch": 0.20268599590156994,
      "grad_norm": 0.2794472575187683,
      "learning_rate": 9.949165689605615e-05,
      "loss": 2.4425,
      "step": 3740
    },
    {
      "epoch": 0.20274019001812116,
      "eval_loss": 2.7546393871307373,
      "eval_runtime": 22.0026,
      "eval_samples_per_second": 227.246,
      "eval_steps_per_second": 1.227,
      "step": 3741
    },
    {
      "epoch": 0.20322793706708214,
      "grad_norm": 0.844109058380127,
      "learning_rate": 9.947785187273695e-05,
      "loss": 2.4386,
      "step": 3750
    },
    {
      "epoch": 0.20376987823259438,
      "grad_norm": 0.3052925765514374,
      "learning_rate": 9.946386299502911e-05,
      "loss": 2.448,
      "step": 3760
    },
    {
      "epoch": 0.20431181939810658,
      "grad_norm": 0.5775868892669678,
      "learning_rate": 9.94496903207573e-05,
      "loss": 2.4449,
      "step": 3770
    },
    {
      "epoch": 0.20485376056361881,
      "grad_norm": 0.7839900255203247,
      "learning_rate": 9.943533390850595e-05,
      "loss": 2.4545,
      "step": 3780
    },
    {
      "epoch": 0.20539570172913102,
      "grad_norm": 0.4013843834400177,
      "learning_rate": 9.942079381761889e-05,
      "loss": 2.4563,
      "step": 3790
    },
    {
      "epoch": 0.20593764289464325,
      "grad_norm": 0.3987264633178711,
      "learning_rate": 9.940607010819937e-05,
      "loss": 2.4431,
      "step": 3800
    },
    {
      "epoch": 0.20647958406015546,
      "grad_norm": 0.3003256022930145,
      "learning_rate": 9.93911628411095e-05,
      "loss": 2.4424,
      "step": 3810
    },
    {
      "epoch": 0.2070215252256677,
      "grad_norm": 0.6472198367118835,
      "learning_rate": 9.937607207797024e-05,
      "loss": 2.4379,
      "step": 3820
    },
    {
      "epoch": 0.20745507815807745,
      "eval_loss": 2.7585620880126953,
      "eval_runtime": 21.9936,
      "eval_samples_per_second": 227.339,
      "eval_steps_per_second": 1.228,
      "step": 3828
    },
    {
      "epoch": 0.2075634663911799,
      "grad_norm": 0.4794847071170807,
      "learning_rate": 9.9360797881161e-05,
      "loss": 2.4339,
      "step": 3830
    },
    {
      "epoch": 0.20810540755669213,
      "grad_norm": 0.3864356577396393,
      "learning_rate": 9.934534031381946e-05,
      "loss": 2.4437,
      "step": 3840
    },
    {
      "epoch": 0.20864734872220433,
      "grad_norm": 0.4814181923866272,
      "learning_rate": 9.932969943984126e-05,
      "loss": 2.4322,
      "step": 3850
    },
    {
      "epoch": 0.20918928988771657,
      "grad_norm": 0.6743614077568054,
      "learning_rate": 9.931387532387981e-05,
      "loss": 2.4356,
      "step": 3860
    },
    {
      "epoch": 0.20973123105322877,
      "grad_norm": 0.6256967186927795,
      "learning_rate": 9.929786803134592e-05,
      "loss": 2.4571,
      "step": 3870
    },
    {
      "epoch": 0.210273172218741,
      "grad_norm": 0.45648303627967834,
      "learning_rate": 9.928167762840761e-05,
      "loss": 2.432,
      "step": 3880
    },
    {
      "epoch": 0.2108151133842532,
      "grad_norm": 0.4223450720310211,
      "learning_rate": 9.926530418198978e-05,
      "loss": 2.4279,
      "step": 3890
    },
    {
      "epoch": 0.21135705454976544,
      "grad_norm": 0.40104198455810547,
      "learning_rate": 9.924874775977402e-05,
      "loss": 2.4261,
      "step": 3900
    },
    {
      "epoch": 0.21189899571527765,
      "grad_norm": 0.7316650748252869,
      "learning_rate": 9.923200843019818e-05,
      "loss": 2.4293,
      "step": 3910
    },
    {
      "epoch": 0.21216996629803378,
      "eval_loss": 2.747528553009033,
      "eval_runtime": 22.0,
      "eval_samples_per_second": 227.273,
      "eval_steps_per_second": 1.227,
      "step": 3915
    },
    {
      "epoch": 0.21244093688078988,
      "grad_norm": 0.39364123344421387,
      "learning_rate": 9.921508626245628e-05,
      "loss": 2.4287,
      "step": 3920
    },
    {
      "epoch": 0.2129828780463021,
      "grad_norm": 0.5997620224952698,
      "learning_rate": 9.919798132649803e-05,
      "loss": 2.4304,
      "step": 3930
    },
    {
      "epoch": 0.21352481921181432,
      "grad_norm": 0.6328970193862915,
      "learning_rate": 9.91806936930287e-05,
      "loss": 2.4236,
      "step": 3940
    },
    {
      "epoch": 0.21406676037732653,
      "grad_norm": 0.8760083913803101,
      "learning_rate": 9.916322343350875e-05,
      "loss": 2.4235,
      "step": 3950
    },
    {
      "epoch": 0.21460870154283876,
      "grad_norm": 0.9582383036613464,
      "learning_rate": 9.914557062015352e-05,
      "loss": 2.4171,
      "step": 3960
    },
    {
      "epoch": 0.21515064270835096,
      "grad_norm": 0.7223474383354187,
      "learning_rate": 9.912773532593297e-05,
      "loss": 2.4194,
      "step": 3970
    },
    {
      "epoch": 0.2156925838738632,
      "grad_norm": 0.7395852208137512,
      "learning_rate": 9.910971762457138e-05,
      "loss": 2.412,
      "step": 3980
    },
    {
      "epoch": 0.2162345250393754,
      "grad_norm": 0.352475643157959,
      "learning_rate": 9.909151759054702e-05,
      "loss": 2.4086,
      "step": 3990
    },
    {
      "epoch": 0.21677646620488764,
      "grad_norm": 0.43535661697387695,
      "learning_rate": 9.907313529909185e-05,
      "loss": 2.4128,
      "step": 4000
    },
    {
      "epoch": 0.21688485443799008,
      "eval_loss": 2.742093563079834,
      "eval_runtime": 21.9979,
      "eval_samples_per_second": 227.294,
      "eval_steps_per_second": 1.227,
      "step": 4002
    },
    {
      "epoch": 0.21731840737039984,
      "grad_norm": 0.4158700406551361,
      "learning_rate": 9.905457082619124e-05,
      "loss": 2.4054,
      "step": 4010
    },
    {
      "epoch": 0.21786034853591207,
      "grad_norm": 0.6706676483154297,
      "learning_rate": 9.903582424858355e-05,
      "loss": 2.3992,
      "step": 4020
    },
    {
      "epoch": 0.21840228970142428,
      "grad_norm": 1.0080608129501343,
      "learning_rate": 9.901689564375998e-05,
      "loss": 2.4159,
      "step": 4030
    },
    {
      "epoch": 0.2189442308669365,
      "grad_norm": 0.8006541728973389,
      "learning_rate": 9.899778508996412e-05,
      "loss": 2.413,
      "step": 4040
    },
    {
      "epoch": 0.21948617203244872,
      "grad_norm": 0.7094182968139648,
      "learning_rate": 9.89784926661917e-05,
      "loss": 2.4095,
      "step": 4050
    },
    {
      "epoch": 0.22002811319796095,
      "grad_norm": 0.5873300433158875,
      "learning_rate": 9.895901845219013e-05,
      "loss": 2.4033,
      "step": 4060
    },
    {
      "epoch": 0.22057005436347316,
      "grad_norm": 0.65219646692276,
      "learning_rate": 9.893936252845842e-05,
      "loss": 2.3991,
      "step": 4070
    },
    {
      "epoch": 0.2211119955289854,
      "grad_norm": 0.5918698310852051,
      "learning_rate": 9.891952497624662e-05,
      "loss": 2.4142,
      "step": 4080
    },
    {
      "epoch": 0.22159974257794637,
      "eval_loss": 2.7375481128692627,
      "eval_runtime": 21.9899,
      "eval_samples_per_second": 227.377,
      "eval_steps_per_second": 1.228,
      "step": 4089
    },
    {
      "epoch": 0.2216539366944976,
      "grad_norm": 1.1681138277053833,
      "learning_rate": 9.889950587755549e-05,
      "loss": 2.4009,
      "step": 4090
    },
    {
      "epoch": 0.22219587786000983,
      "grad_norm": 0.3698543310165405,
      "learning_rate": 9.88793053151364e-05,
      "loss": 2.398,
      "step": 4100
    },
    {
      "epoch": 0.22273781902552203,
      "grad_norm": 0.3500048816204071,
      "learning_rate": 9.885892337249069e-05,
      "loss": 2.3997,
      "step": 4110
    },
    {
      "epoch": 0.22327976019103427,
      "grad_norm": 0.9475021958351135,
      "learning_rate": 9.88383601338695e-05,
      "loss": 2.4023,
      "step": 4120
    },
    {
      "epoch": 0.22382170135654647,
      "grad_norm": 0.37473389506340027,
      "learning_rate": 9.881761568427335e-05,
      "loss": 2.3969,
      "step": 4130
    },
    {
      "epoch": 0.2243636425220587,
      "grad_norm": 0.46967896819114685,
      "learning_rate": 9.879669010945189e-05,
      "loss": 2.3995,
      "step": 4140
    },
    {
      "epoch": 0.2249055836875709,
      "grad_norm": 0.40588507056236267,
      "learning_rate": 9.877558349590341e-05,
      "loss": 2.3983,
      "step": 4150
    },
    {
      "epoch": 0.22544752485308314,
      "grad_norm": 0.32905760407447815,
      "learning_rate": 9.875429593087454e-05,
      "loss": 2.3965,
      "step": 4160
    },
    {
      "epoch": 0.22598946601859535,
      "grad_norm": 0.7571531534194946,
      "learning_rate": 9.873282750235993e-05,
      "loss": 2.3858,
      "step": 4170
    },
    {
      "epoch": 0.2263146307179027,
      "eval_loss": 2.7395927906036377,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.323,
      "eval_steps_per_second": 1.228,
      "step": 4176
    },
    {
      "epoch": 0.22653140718410758,
      "grad_norm": 0.4100936949253082,
      "learning_rate": 9.871117829910181e-05,
      "loss": 2.3913,
      "step": 4180
    },
    {
      "epoch": 0.22707334834961979,
      "grad_norm": 1.1051249504089355,
      "learning_rate": 9.868934841058972e-05,
      "loss": 2.385,
      "step": 4190
    },
    {
      "epoch": 0.22761528951513202,
      "grad_norm": 0.4184013903141022,
      "learning_rate": 9.866733792706003e-05,
      "loss": 2.3901,
      "step": 4200
    },
    {
      "epoch": 0.22815723068064422,
      "grad_norm": 0.8634352087974548,
      "learning_rate": 9.864514693949563e-05,
      "loss": 2.382,
      "step": 4210
    },
    {
      "epoch": 0.22869917184615646,
      "grad_norm": 0.42947426438331604,
      "learning_rate": 9.86227755396256e-05,
      "loss": 2.3847,
      "step": 4220
    },
    {
      "epoch": 0.22924111301166866,
      "grad_norm": 0.7337902784347534,
      "learning_rate": 9.860022381992467e-05,
      "loss": 2.3877,
      "step": 4230
    },
    {
      "epoch": 0.2297830541771809,
      "grad_norm": 0.33461204171180725,
      "learning_rate": 9.857749187361308e-05,
      "loss": 2.3771,
      "step": 4240
    },
    {
      "epoch": 0.2303249953426931,
      "grad_norm": 0.4798398017883301,
      "learning_rate": 9.85545797946559e-05,
      "loss": 2.3745,
      "step": 4250
    },
    {
      "epoch": 0.23086693650820533,
      "grad_norm": 0.7971112728118896,
      "learning_rate": 9.853148767776293e-05,
      "loss": 2.3856,
      "step": 4260
    },
    {
      "epoch": 0.231029518857859,
      "eval_loss": 2.7393031120300293,
      "eval_runtime": 21.9959,
      "eval_samples_per_second": 227.316,
      "eval_steps_per_second": 1.228,
      "step": 4263
    },
    {
      "epoch": 0.23140887767371754,
      "grad_norm": 0.5765118598937988,
      "learning_rate": 9.85082156183881e-05,
      "loss": 2.3818,
      "step": 4270
    },
    {
      "epoch": 0.23195081883922977,
      "grad_norm": 0.9162944555282593,
      "learning_rate": 9.848476371272922e-05,
      "loss": 2.3756,
      "step": 4280
    },
    {
      "epoch": 0.23249276000474198,
      "grad_norm": 0.757483959197998,
      "learning_rate": 9.846113205772746e-05,
      "loss": 2.3754,
      "step": 4290
    },
    {
      "epoch": 0.2330347011702542,
      "grad_norm": 0.5934547185897827,
      "learning_rate": 9.8437320751067e-05,
      "loss": 2.3688,
      "step": 4300
    },
    {
      "epoch": 0.23357664233576642,
      "grad_norm": 0.540972888469696,
      "learning_rate": 9.841332989117469e-05,
      "loss": 2.3713,
      "step": 4310
    },
    {
      "epoch": 0.23411858350127865,
      "grad_norm": 0.41171032190322876,
      "learning_rate": 9.838915957721953e-05,
      "loss": 2.3675,
      "step": 4320
    },
    {
      "epoch": 0.23466052466679085,
      "grad_norm": 0.5488543510437012,
      "learning_rate": 9.83648099091123e-05,
      "loss": 2.369,
      "step": 4330
    },
    {
      "epoch": 0.2352024658323031,
      "grad_norm": 0.4907141923904419,
      "learning_rate": 9.834028098750525e-05,
      "loss": 2.3694,
      "step": 4340
    },
    {
      "epoch": 0.2357444069978153,
      "grad_norm": 0.3236653804779053,
      "learning_rate": 9.83155729137915e-05,
      "loss": 2.3596,
      "step": 4350
    },
    {
      "epoch": 0.2357444069978153,
      "eval_loss": 2.7318739891052246,
      "eval_runtime": 21.9575,
      "eval_samples_per_second": 227.713,
      "eval_steps_per_second": 1.23,
      "step": 4350
    },
    {
      "epoch": 0.23628634816332753,
      "grad_norm": 0.29620563983917236,
      "learning_rate": 9.82906857901048e-05,
      "loss": 2.3661,
      "step": 4360
    },
    {
      "epoch": 0.23682828932883973,
      "grad_norm": 0.9057065844535828,
      "learning_rate": 9.826561971931891e-05,
      "loss": 2.366,
      "step": 4370
    },
    {
      "epoch": 0.23737023049435196,
      "grad_norm": 0.9793679118156433,
      "learning_rate": 9.824037480504741e-05,
      "loss": 2.3657,
      "step": 4380
    },
    {
      "epoch": 0.23791217165986417,
      "grad_norm": 0.6782917976379395,
      "learning_rate": 9.821495115164309e-05,
      "loss": 2.3674,
      "step": 4390
    },
    {
      "epoch": 0.2384541128253764,
      "grad_norm": 0.3981853425502777,
      "learning_rate": 9.818934886419756e-05,
      "loss": 2.3678,
      "step": 4400
    },
    {
      "epoch": 0.2389960539908886,
      "grad_norm": 0.5394397974014282,
      "learning_rate": 9.816356804854089e-05,
      "loss": 2.3544,
      "step": 4410
    },
    {
      "epoch": 0.23953799515640084,
      "grad_norm": 0.45772433280944824,
      "learning_rate": 9.813760881124107e-05,
      "loss": 2.3546,
      "step": 4420
    },
    {
      "epoch": 0.24007993632191305,
      "grad_norm": 0.5437049269676208,
      "learning_rate": 9.811147125960364e-05,
      "loss": 2.3592,
      "step": 4430
    },
    {
      "epoch": 0.24045929513777162,
      "eval_loss": 2.728144407272339,
      "eval_runtime": 21.9944,
      "eval_samples_per_second": 227.331,
      "eval_steps_per_second": 1.228,
      "step": 4437
    },
    {
      "epoch": 0.24062187748742528,
      "grad_norm": 0.8073667883872986,
      "learning_rate": 9.808515550167124e-05,
      "loss": 2.3523,
      "step": 4440
    },
    {
      "epoch": 0.24116381865293748,
      "grad_norm": 0.40337878465652466,
      "learning_rate": 9.805866164622311e-05,
      "loss": 2.3574,
      "step": 4450
    },
    {
      "epoch": 0.24170575981844972,
      "grad_norm": 0.7153369188308716,
      "learning_rate": 9.803198980277467e-05,
      "loss": 2.3557,
      "step": 4460
    },
    {
      "epoch": 0.24224770098396192,
      "grad_norm": 0.37222856283187866,
      "learning_rate": 9.800514008157711e-05,
      "loss": 2.3553,
      "step": 4470
    },
    {
      "epoch": 0.24278964214947416,
      "grad_norm": 0.3131774067878723,
      "learning_rate": 9.79781125936169e-05,
      "loss": 2.3562,
      "step": 4480
    },
    {
      "epoch": 0.24333158331498636,
      "grad_norm": 0.5932138562202454,
      "learning_rate": 9.79509074506153e-05,
      "loss": 2.3422,
      "step": 4490
    },
    {
      "epoch": 0.2438735244804986,
      "grad_norm": 0.6757357716560364,
      "learning_rate": 9.792352476502796e-05,
      "loss": 2.349,
      "step": 4500
    },
    {
      "epoch": 0.2444154656460108,
      "grad_norm": 0.9189159870147705,
      "learning_rate": 9.789596465004437e-05,
      "loss": 2.341,
      "step": 4510
    },
    {
      "epoch": 0.24495740681152303,
      "grad_norm": 0.40294119715690613,
      "learning_rate": 9.786822721958751e-05,
      "loss": 2.3526,
      "step": 4520
    },
    {
      "epoch": 0.24517418327772791,
      "eval_loss": 2.7286596298217773,
      "eval_runtime": 21.9954,
      "eval_samples_per_second": 227.32,
      "eval_steps_per_second": 1.228,
      "step": 4524
    },
    {
      "epoch": 0.24549934797703524,
      "grad_norm": 0.4049138128757477,
      "learning_rate": 9.784031258831325e-05,
      "loss": 2.3427,
      "step": 4530
    },
    {
      "epoch": 0.24604128914254747,
      "grad_norm": 0.5206155776977539,
      "learning_rate": 9.781222087161003e-05,
      "loss": 2.347,
      "step": 4540
    },
    {
      "epoch": 0.24658323030805968,
      "grad_norm": 0.3420925736427307,
      "learning_rate": 9.77839521855982e-05,
      "loss": 2.3421,
      "step": 4550
    },
    {
      "epoch": 0.2471251714735719,
      "grad_norm": 1.2839010953903198,
      "learning_rate": 9.775550664712966e-05,
      "loss": 2.3492,
      "step": 4560
    },
    {
      "epoch": 0.2476671126390841,
      "grad_norm": 0.6513694524765015,
      "learning_rate": 9.772688437378738e-05,
      "loss": 2.3417,
      "step": 4570
    },
    {
      "epoch": 0.24820905380459635,
      "grad_norm": 0.7786572575569153,
      "learning_rate": 9.769808548388488e-05,
      "loss": 2.333,
      "step": 4580
    },
    {
      "epoch": 0.24875099497010855,
      "grad_norm": 0.8283511996269226,
      "learning_rate": 9.766911009646569e-05,
      "loss": 2.3358,
      "step": 4590
    },
    {
      "epoch": 0.24929293613562079,
      "grad_norm": 0.5858640074729919,
      "learning_rate": 9.763995833130299e-05,
      "loss": 2.3424,
      "step": 4600
    },
    {
      "epoch": 0.249834877301133,
      "grad_norm": 0.41306570172309875,
      "learning_rate": 9.761063030889898e-05,
      "loss": 2.3338,
      "step": 4610
    },
    {
      "epoch": 0.2498890714176842,
      "eval_loss": 2.723004102706909,
      "eval_runtime": 21.9968,
      "eval_samples_per_second": 227.306,
      "eval_steps_per_second": 1.227,
      "step": 4611
    },
    {
      "epoch": 0.2503768184666452,
      "grad_norm": 0.32257717847824097,
      "learning_rate": 9.758112615048448e-05,
      "loss": 2.3332,
      "step": 4620
    },
    {
      "epoch": 0.25091875963215743,
      "grad_norm": 0.31163015961647034,
      "learning_rate": 9.755144597801837e-05,
      "loss": 2.3306,
      "step": 4630
    },
    {
      "epoch": 0.25146070079766963,
      "grad_norm": 0.6164253950119019,
      "learning_rate": 9.752158991418708e-05,
      "loss": 2.3288,
      "step": 4640
    },
    {
      "epoch": 0.2520026419631819,
      "grad_norm": 0.6337904930114746,
      "learning_rate": 9.749155808240415e-05,
      "loss": 2.3271,
      "step": 4650
    },
    {
      "epoch": 0.2525445831286941,
      "grad_norm": 0.3386767506599426,
      "learning_rate": 9.746135060680966e-05,
      "loss": 2.3342,
      "step": 4660
    },
    {
      "epoch": 0.2530865242942063,
      "grad_norm": 0.9337306022644043,
      "learning_rate": 9.743096761226972e-05,
      "loss": 2.3299,
      "step": 4670
    },
    {
      "epoch": 0.2536284654597185,
      "grad_norm": 0.31421270966529846,
      "learning_rate": 9.7400409224376e-05,
      "loss": 2.3302,
      "step": 4680
    },
    {
      "epoch": 0.25417040662523077,
      "grad_norm": 0.33126044273376465,
      "learning_rate": 9.736967556944516e-05,
      "loss": 2.325,
      "step": 4690
    },
    {
      "epoch": 0.25460395955764054,
      "eval_loss": 2.722799301147461,
      "eval_runtime": 25.1071,
      "eval_samples_per_second": 199.147,
      "eval_steps_per_second": 1.075,
      "step": 4698
    },
    {
      "epoch": 0.254712347790743,
      "grad_norm": 0.3403181731700897,
      "learning_rate": 9.733876677451833e-05,
      "loss": 2.319,
      "step": 4700
    },
    {
      "epoch": 0.2552542889562552,
      "grad_norm": 0.4288688004016876,
      "learning_rate": 9.730768296736064e-05,
      "loss": 2.322,
      "step": 4710
    },
    {
      "epoch": 0.2557962301217674,
      "grad_norm": 1.1494555473327637,
      "learning_rate": 9.727642427646061e-05,
      "loss": 2.319,
      "step": 4720
    },
    {
      "epoch": 0.25633817128727965,
      "grad_norm": 1.3199281692504883,
      "learning_rate": 9.72449908310297e-05,
      "loss": 2.3299,
      "step": 4730
    },
    {
      "epoch": 0.25688011245279185,
      "grad_norm": 1.220360279083252,
      "learning_rate": 9.721338276100172e-05,
      "loss": 2.3441,
      "step": 4740
    },
    {
      "epoch": 0.25742205361830406,
      "grad_norm": 1.106679916381836,
      "learning_rate": 9.718160019703232e-05,
      "loss": 2.3379,
      "step": 4750
    },
    {
      "epoch": 0.25796399478381626,
      "grad_norm": 1.3295619487762451,
      "learning_rate": 9.714964327049842e-05,
      "loss": 2.3501,
      "step": 4760
    },
    {
      "epoch": 0.2585059359493285,
      "grad_norm": 0.6279563307762146,
      "learning_rate": 9.711751211349773e-05,
      "loss": 2.3166,
      "step": 4770
    },
    {
      "epoch": 0.25904787711484073,
      "grad_norm": 0.3395664095878601,
      "learning_rate": 9.70852068588481e-05,
      "loss": 2.3087,
      "step": 4780
    },
    {
      "epoch": 0.25931884769759683,
      "eval_loss": 2.721379518508911,
      "eval_runtime": 21.9931,
      "eval_samples_per_second": 227.344,
      "eval_steps_per_second": 1.228,
      "step": 4785
    },
    {
      "epoch": 0.25958981828035294,
      "grad_norm": 0.32632923126220703,
      "learning_rate": 9.705272764008709e-05,
      "loss": 2.3064,
      "step": 4790
    },
    {
      "epoch": 0.26013175944586514,
      "grad_norm": 0.3286370038986206,
      "learning_rate": 9.702007459147134e-05,
      "loss": 2.3013,
      "step": 4800
    },
    {
      "epoch": 0.2606737006113774,
      "grad_norm": 0.6520362496376038,
      "learning_rate": 9.698724784797604e-05,
      "loss": 2.3196,
      "step": 4810
    },
    {
      "epoch": 0.2612156417768896,
      "grad_norm": 0.6383447051048279,
      "learning_rate": 9.695424754529434e-05,
      "loss": 2.3109,
      "step": 4820
    },
    {
      "epoch": 0.2617575829424018,
      "grad_norm": 0.3725563585758209,
      "learning_rate": 9.692107381983684e-05,
      "loss": 2.3072,
      "step": 4830
    },
    {
      "epoch": 0.262299524107914,
      "grad_norm": 0.5179880857467651,
      "learning_rate": 9.688772680873103e-05,
      "loss": 2.3065,
      "step": 4840
    },
    {
      "epoch": 0.2628414652734263,
      "grad_norm": 0.5016129612922668,
      "learning_rate": 9.685420664982067e-05,
      "loss": 2.3088,
      "step": 4850
    },
    {
      "epoch": 0.2633834064389385,
      "grad_norm": 0.5394784212112427,
      "learning_rate": 9.682051348166523e-05,
      "loss": 2.2979,
      "step": 4860
    },
    {
      "epoch": 0.2639253476044507,
      "grad_norm": 0.5578838586807251,
      "learning_rate": 9.678664744353935e-05,
      "loss": 2.3034,
      "step": 4870
    },
    {
      "epoch": 0.26403373583755313,
      "eval_loss": 2.723337173461914,
      "eval_runtime": 21.9924,
      "eval_samples_per_second": 227.351,
      "eval_steps_per_second": 1.228,
      "step": 4872
    },
    {
      "epoch": 0.2644672887699629,
      "grad_norm": 0.34530109167099,
      "learning_rate": 9.675260867543224e-05,
      "loss": 2.3096,
      "step": 4880
    },
    {
      "epoch": 0.26500922993547515,
      "grad_norm": 0.7636291980743408,
      "learning_rate": 9.671839731804716e-05,
      "loss": 2.307,
      "step": 4890
    },
    {
      "epoch": 0.26555117110098736,
      "grad_norm": 0.3167097568511963,
      "learning_rate": 9.66840135128007e-05,
      "loss": 2.3068,
      "step": 4900
    },
    {
      "epoch": 0.26609311226649957,
      "grad_norm": 0.45626381039619446,
      "learning_rate": 9.664945740182235e-05,
      "loss": 2.2948,
      "step": 4910
    },
    {
      "epoch": 0.26663505343201177,
      "grad_norm": 0.695152759552002,
      "learning_rate": 9.661472912795383e-05,
      "loss": 2.2975,
      "step": 4920
    },
    {
      "epoch": 0.26717699459752403,
      "grad_norm": 0.30375921726226807,
      "learning_rate": 9.65798288347485e-05,
      "loss": 2.3041,
      "step": 4930
    },
    {
      "epoch": 0.26771893576303624,
      "grad_norm": 0.8904628753662109,
      "learning_rate": 9.654475666647078e-05,
      "loss": 2.2957,
      "step": 4940
    },
    {
      "epoch": 0.26826087692854844,
      "grad_norm": 1.199233889579773,
      "learning_rate": 9.650951276809561e-05,
      "loss": 2.2994,
      "step": 4950
    },
    {
      "epoch": 0.2687486239775094,
      "eval_loss": 2.7227141857147217,
      "eval_runtime": 21.9965,
      "eval_samples_per_second": 227.309,
      "eval_steps_per_second": 1.227,
      "step": 4959
    },
    {
      "epoch": 0.26880281809406065,
      "grad_norm": 2.123131513595581,
      "learning_rate": 9.647409728530772e-05,
      "loss": 2.3143,
      "step": 4960
    },
    {
      "epoch": 0.2693447592595729,
      "grad_norm": 0.7237464189529419,
      "learning_rate": 9.643851036450115e-05,
      "loss": 2.3183,
      "step": 4970
    },
    {
      "epoch": 0.2698867004250851,
      "grad_norm": 0.6150104403495789,
      "learning_rate": 9.640275215277858e-05,
      "loss": 2.3048,
      "step": 4980
    },
    {
      "epoch": 0.2704286415905973,
      "grad_norm": 0.3302861154079437,
      "learning_rate": 9.636682279795076e-05,
      "loss": 2.2903,
      "step": 4990
    },
    {
      "epoch": 0.2709705827561095,
      "grad_norm": 0.7235733270645142,
      "learning_rate": 9.633072244853587e-05,
      "loss": 2.2807,
      "step": 5000
    },
    {
      "epoch": 0.2715125239216218,
      "grad_norm": 0.36814624071121216,
      "learning_rate": 9.629445125375891e-05,
      "loss": 2.2886,
      "step": 5010
    },
    {
      "epoch": 0.272054465087134,
      "grad_norm": 0.5725374221801758,
      "learning_rate": 9.625800936355108e-05,
      "loss": 2.2996,
      "step": 5020
    },
    {
      "epoch": 0.2725964062526462,
      "grad_norm": 0.45517677068710327,
      "learning_rate": 9.62213969285492e-05,
      "loss": 2.2876,
      "step": 5030
    },
    {
      "epoch": 0.2731383474181584,
      "grad_norm": 0.5474647879600525,
      "learning_rate": 9.618461410009503e-05,
      "loss": 2.2838,
      "step": 5040
    },
    {
      "epoch": 0.2734635121174657,
      "eval_loss": 2.713071346282959,
      "eval_runtime": 21.9919,
      "eval_samples_per_second": 227.356,
      "eval_steps_per_second": 1.228,
      "step": 5046
    },
    {
      "epoch": 0.27368028858367066,
      "grad_norm": 0.3035077452659607,
      "learning_rate": 9.614766103023473e-05,
      "loss": 2.2759,
      "step": 5050
    },
    {
      "epoch": 0.27422222974918287,
      "grad_norm": 0.6675074100494385,
      "learning_rate": 9.611053787171804e-05,
      "loss": 2.2889,
      "step": 5060
    },
    {
      "epoch": 0.27476417091469507,
      "grad_norm": 0.5895007252693176,
      "learning_rate": 9.607324477799793e-05,
      "loss": 2.2786,
      "step": 5070
    },
    {
      "epoch": 0.2753061120802073,
      "grad_norm": 0.781353771686554,
      "learning_rate": 9.603578190322974e-05,
      "loss": 2.2852,
      "step": 5080
    },
    {
      "epoch": 0.27584805324571954,
      "grad_norm": 0.3827020823955536,
      "learning_rate": 9.599814940227062e-05,
      "loss": 2.2813,
      "step": 5090
    },
    {
      "epoch": 0.27638999441123174,
      "grad_norm": 0.6647722125053406,
      "learning_rate": 9.59603474306789e-05,
      "loss": 2.279,
      "step": 5100
    },
    {
      "epoch": 0.27693193557674395,
      "grad_norm": 0.3594939112663269,
      "learning_rate": 9.592237614471346e-05,
      "loss": 2.2697,
      "step": 5110
    },
    {
      "epoch": 0.27747387674225615,
      "grad_norm": 0.3628252446651459,
      "learning_rate": 9.588423570133301e-05,
      "loss": 2.2752,
      "step": 5120
    },
    {
      "epoch": 0.2780158179077684,
      "grad_norm": 0.322070449590683,
      "learning_rate": 9.584592625819555e-05,
      "loss": 2.2714,
      "step": 5130
    },
    {
      "epoch": 0.2781784002574221,
      "eval_loss": 2.711329698562622,
      "eval_runtime": 22.0018,
      "eval_samples_per_second": 227.255,
      "eval_steps_per_second": 1.227,
      "step": 5133
    },
    {
      "epoch": 0.2785577590732806,
      "grad_norm": 0.4778302311897278,
      "learning_rate": 9.580744797365761e-05,
      "loss": 2.2721,
      "step": 5140
    },
    {
      "epoch": 0.2790997002387928,
      "grad_norm": 0.44883444905281067,
      "learning_rate": 9.57688010067737e-05,
      "loss": 2.2807,
      "step": 5150
    },
    {
      "epoch": 0.27964164140430503,
      "grad_norm": 0.9123862981796265,
      "learning_rate": 9.572998551729552e-05,
      "loss": 2.2729,
      "step": 5160
    },
    {
      "epoch": 0.2801835825698173,
      "grad_norm": 2.519432544708252,
      "learning_rate": 9.569100166567143e-05,
      "loss": 2.2818,
      "step": 5170
    },
    {
      "epoch": 0.2807255237353295,
      "grad_norm": 1.1521965265274048,
      "learning_rate": 9.565184961304577e-05,
      "loss": 2.2977,
      "step": 5180
    },
    {
      "epoch": 0.2812674649008417,
      "grad_norm": 1.3088760375976562,
      "learning_rate": 9.561252952125808e-05,
      "loss": 2.2886,
      "step": 5190
    },
    {
      "epoch": 0.2818094060663539,
      "grad_norm": 0.45803970098495483,
      "learning_rate": 9.557304155284256e-05,
      "loss": 2.2879,
      "step": 5200
    },
    {
      "epoch": 0.28235134723186617,
      "grad_norm": 0.4502812623977661,
      "learning_rate": 9.553338587102732e-05,
      "loss": 2.2774,
      "step": 5210
    },
    {
      "epoch": 0.2828932883973784,
      "grad_norm": 0.3390588164329529,
      "learning_rate": 9.549356263973376e-05,
      "loss": 2.2749,
      "step": 5220
    },
    {
      "epoch": 0.2828932883973784,
      "eval_loss": 2.7108614444732666,
      "eval_runtime": 21.8696,
      "eval_samples_per_second": 228.628,
      "eval_steps_per_second": 1.235,
      "step": 5220
    },
    {
      "epoch": 0.2834352295628906,
      "grad_norm": 0.31600069999694824,
      "learning_rate": 9.545357202357584e-05,
      "loss": 2.2686,
      "step": 5230
    },
    {
      "epoch": 0.2839771707284028,
      "grad_norm": 0.6420486569404602,
      "learning_rate": 9.541341418785944e-05,
      "loss": 2.2669,
      "step": 5240
    },
    {
      "epoch": 0.28451911189391504,
      "grad_norm": 0.6971787810325623,
      "learning_rate": 9.537308929858167e-05,
      "loss": 2.2553,
      "step": 5250
    },
    {
      "epoch": 0.28506105305942725,
      "grad_norm": 0.35600554943084717,
      "learning_rate": 9.533259752243015e-05,
      "loss": 2.2607,
      "step": 5260
    },
    {
      "epoch": 0.28560299422493945,
      "grad_norm": 0.4857500195503235,
      "learning_rate": 9.529193902678236e-05,
      "loss": 2.2604,
      "step": 5270
    },
    {
      "epoch": 0.28614493539045166,
      "grad_norm": 0.41848480701446533,
      "learning_rate": 9.525111397970495e-05,
      "loss": 2.2627,
      "step": 5280
    },
    {
      "epoch": 0.28668687655596387,
      "grad_norm": 0.39947494864463806,
      "learning_rate": 9.521012254995298e-05,
      "loss": 2.259,
      "step": 5290
    },
    {
      "epoch": 0.2872288177214761,
      "grad_norm": 0.6807723045349121,
      "learning_rate": 9.516896490696936e-05,
      "loss": 2.2634,
      "step": 5300
    },
    {
      "epoch": 0.28760817653733467,
      "eval_loss": 2.706552267074585,
      "eval_runtime": 21.9925,
      "eval_samples_per_second": 227.35,
      "eval_steps_per_second": 1.228,
      "step": 5307
    },
    {
      "epoch": 0.28777075888698833,
      "grad_norm": 0.7820791602134705,
      "learning_rate": 9.512764122088394e-05,
      "loss": 2.2567,
      "step": 5310
    },
    {
      "epoch": 0.28831270005250054,
      "grad_norm": 0.6624732613563538,
      "learning_rate": 9.508615166251305e-05,
      "loss": 2.263,
      "step": 5320
    },
    {
      "epoch": 0.28885464121801274,
      "grad_norm": 0.3670085072517395,
      "learning_rate": 9.504449640335858e-05,
      "loss": 2.2528,
      "step": 5330
    },
    {
      "epoch": 0.289396582383525,
      "grad_norm": 0.36448076367378235,
      "learning_rate": 9.500267561560746e-05,
      "loss": 2.2564,
      "step": 5340
    },
    {
      "epoch": 0.2899385235490372,
      "grad_norm": 0.3871496021747589,
      "learning_rate": 9.496068947213073e-05,
      "loss": 2.2561,
      "step": 5350
    },
    {
      "epoch": 0.2904804647145494,
      "grad_norm": 0.5003486275672913,
      "learning_rate": 9.491853814648305e-05,
      "loss": 2.2539,
      "step": 5360
    },
    {
      "epoch": 0.2910224058800616,
      "grad_norm": 1.1620166301727295,
      "learning_rate": 9.487622181290183e-05,
      "loss": 2.2517,
      "step": 5370
    },
    {
      "epoch": 0.2915643470455739,
      "grad_norm": 0.725284218788147,
      "learning_rate": 9.483374064630656e-05,
      "loss": 2.2456,
      "step": 5380
    },
    {
      "epoch": 0.2921062882110861,
      "grad_norm": 0.541685938835144,
      "learning_rate": 9.479109482229812e-05,
      "loss": 2.2553,
      "step": 5390
    },
    {
      "epoch": 0.29232306467729097,
      "eval_loss": 2.7066264152526855,
      "eval_runtime": 21.9975,
      "eval_samples_per_second": 227.298,
      "eval_steps_per_second": 1.227,
      "step": 5394
    },
    {
      "epoch": 0.2926482293765983,
      "grad_norm": 0.40974506735801697,
      "learning_rate": 9.474828451715798e-05,
      "loss": 2.2516,
      "step": 5400
    },
    {
      "epoch": 0.2931901705421105,
      "grad_norm": 0.5276529788970947,
      "learning_rate": 9.470530990784752e-05,
      "loss": 2.2554,
      "step": 5410
    },
    {
      "epoch": 0.29373211170762276,
      "grad_norm": 0.9658313393592834,
      "learning_rate": 9.466217117200735e-05,
      "loss": 2.2501,
      "step": 5420
    },
    {
      "epoch": 0.29427405287313496,
      "grad_norm": 0.6076865196228027,
      "learning_rate": 9.461886848795642e-05,
      "loss": 2.2438,
      "step": 5430
    },
    {
      "epoch": 0.29481599403864717,
      "grad_norm": 0.5501825213432312,
      "learning_rate": 9.457540203469142e-05,
      "loss": 2.2419,
      "step": 5440
    },
    {
      "epoch": 0.29535793520415937,
      "grad_norm": 0.41925522685050964,
      "learning_rate": 9.453177199188603e-05,
      "loss": 2.2496,
      "step": 5450
    },
    {
      "epoch": 0.29589987636967163,
      "grad_norm": 0.4038546681404114,
      "learning_rate": 9.448797853989013e-05,
      "loss": 2.255,
      "step": 5460
    },
    {
      "epoch": 0.29644181753518384,
      "grad_norm": 0.3127026855945587,
      "learning_rate": 9.444402185972901e-05,
      "loss": 2.2396,
      "step": 5470
    },
    {
      "epoch": 0.29698375870069604,
      "grad_norm": 0.6538156867027283,
      "learning_rate": 9.439990213310277e-05,
      "loss": 2.2377,
      "step": 5480
    },
    {
      "epoch": 0.29703795281724726,
      "eval_loss": 2.7067112922668457,
      "eval_runtime": 21.9955,
      "eval_samples_per_second": 227.32,
      "eval_steps_per_second": 1.228,
      "step": 5481
    },
    {
      "epoch": 0.29752569986620825,
      "grad_norm": 0.6304920315742493,
      "learning_rate": 9.435561954238548e-05,
      "loss": 2.2417,
      "step": 5490
    },
    {
      "epoch": 0.2980676410317205,
      "grad_norm": 0.5496434569358826,
      "learning_rate": 9.431117427062434e-05,
      "loss": 2.2353,
      "step": 5500
    },
    {
      "epoch": 0.2986095821972327,
      "grad_norm": 0.7714524865150452,
      "learning_rate": 9.426656650153909e-05,
      "loss": 2.2292,
      "step": 5510
    },
    {
      "epoch": 0.2991515233627449,
      "grad_norm": 0.7220959663391113,
      "learning_rate": 9.422179641952113e-05,
      "loss": 2.2506,
      "step": 5520
    },
    {
      "epoch": 0.2996934645282571,
      "grad_norm": 0.5006898045539856,
      "learning_rate": 9.417686420963283e-05,
      "loss": 2.2222,
      "step": 5530
    },
    {
      "epoch": 0.3002354056937694,
      "grad_norm": 0.3837469220161438,
      "learning_rate": 9.413177005760672e-05,
      "loss": 2.2429,
      "step": 5540
    },
    {
      "epoch": 0.3007773468592816,
      "grad_norm": 0.34782886505126953,
      "learning_rate": 9.408651414984472e-05,
      "loss": 2.2423,
      "step": 5550
    },
    {
      "epoch": 0.3013192880247938,
      "grad_norm": 0.35922521352767944,
      "learning_rate": 9.404109667341746e-05,
      "loss": 2.2349,
      "step": 5560
    },
    {
      "epoch": 0.30175284095720356,
      "eval_loss": 2.698547124862671,
      "eval_runtime": 21.9987,
      "eval_samples_per_second": 227.287,
      "eval_steps_per_second": 1.227,
      "step": 5568
    },
    {
      "epoch": 0.301861229190306,
      "grad_norm": 0.4572945237159729,
      "learning_rate": 9.399551781606329e-05,
      "loss": 2.2244,
      "step": 5570
    },
    {
      "epoch": 0.30240317035581826,
      "grad_norm": 0.3920208513736725,
      "learning_rate": 9.394977776618779e-05,
      "loss": 2.2299,
      "step": 5580
    },
    {
      "epoch": 0.30294511152133047,
      "grad_norm": 0.9155771136283875,
      "learning_rate": 9.390387671286279e-05,
      "loss": 2.2424,
      "step": 5590
    },
    {
      "epoch": 0.3034870526868427,
      "grad_norm": 0.34311428666114807,
      "learning_rate": 9.38578148458256e-05,
      "loss": 2.2273,
      "step": 5600
    },
    {
      "epoch": 0.3040289938523549,
      "grad_norm": 0.39582574367523193,
      "learning_rate": 9.381159235547839e-05,
      "loss": 2.2342,
      "step": 5610
    },
    {
      "epoch": 0.30457093501786714,
      "grad_norm": 0.37435415387153625,
      "learning_rate": 9.376520943288716e-05,
      "loss": 2.2312,
      "step": 5620
    },
    {
      "epoch": 0.30511287618337934,
      "grad_norm": 0.5571390986442566,
      "learning_rate": 9.371866626978118e-05,
      "loss": 2.2193,
      "step": 5630
    },
    {
      "epoch": 0.30565481734889155,
      "grad_norm": 0.348417192697525,
      "learning_rate": 9.367196305855199e-05,
      "loss": 2.24,
      "step": 5640
    },
    {
      "epoch": 0.30619675851440376,
      "grad_norm": 0.9660398364067078,
      "learning_rate": 9.362509999225281e-05,
      "loss": 2.2291,
      "step": 5650
    },
    {
      "epoch": 0.3064677290971599,
      "eval_loss": 2.700347900390625,
      "eval_runtime": 21.9973,
      "eval_samples_per_second": 227.301,
      "eval_steps_per_second": 1.227,
      "step": 5655
    },
    {
      "epoch": 0.306738699679916,
      "grad_norm": 0.6863952875137329,
      "learning_rate": 9.357807726459754e-05,
      "loss": 2.2291,
      "step": 5660
    },
    {
      "epoch": 0.3072806408454282,
      "grad_norm": 0.4206200838088989,
      "learning_rate": 9.353089506996016e-05,
      "loss": 2.2224,
      "step": 5670
    },
    {
      "epoch": 0.3078225820109404,
      "grad_norm": 0.9796401262283325,
      "learning_rate": 9.348355360337374e-05,
      "loss": 2.2199,
      "step": 5680
    },
    {
      "epoch": 0.30836452317645263,
      "grad_norm": 0.8994840979576111,
      "learning_rate": 9.343605306052977e-05,
      "loss": 2.2317,
      "step": 5690
    },
    {
      "epoch": 0.3089064643419649,
      "grad_norm": 0.6637131571769714,
      "learning_rate": 9.338839363777728e-05,
      "loss": 2.2211,
      "step": 5700
    },
    {
      "epoch": 0.3094484055074771,
      "grad_norm": 0.6918546557426453,
      "learning_rate": 9.334057553212204e-05,
      "loss": 2.2204,
      "step": 5710
    },
    {
      "epoch": 0.3099903466729893,
      "grad_norm": 0.3313464820384979,
      "learning_rate": 9.32925989412258e-05,
      "loss": 2.2202,
      "step": 5720
    },
    {
      "epoch": 0.3105322878385015,
      "grad_norm": 0.4739852249622345,
      "learning_rate": 9.324446406340537e-05,
      "loss": 2.2164,
      "step": 5730
    },
    {
      "epoch": 0.31107422900401377,
      "grad_norm": 1.1123601198196411,
      "learning_rate": 9.319617109763188e-05,
      "loss": 2.2114,
      "step": 5740
    },
    {
      "epoch": 0.3111826172371162,
      "eval_loss": 2.697807550430298,
      "eval_runtime": 29.7475,
      "eval_samples_per_second": 168.081,
      "eval_steps_per_second": 0.908,
      "step": 5742
    },
    {
      "epoch": 0.311616170169526,
      "grad_norm": 0.33748936653137207,
      "learning_rate": 9.314772024352995e-05,
      "loss": 2.2157,
      "step": 5750
    },
    {
      "epoch": 0.3121581113350382,
      "grad_norm": 0.35050588846206665,
      "learning_rate": 9.309911170137682e-05,
      "loss": 2.2109,
      "step": 5760
    },
    {
      "epoch": 0.3127000525005504,
      "grad_norm": 0.5154600739479065,
      "learning_rate": 9.30503456721016e-05,
      "loss": 2.2026,
      "step": 5770
    },
    {
      "epoch": 0.31324199366606265,
      "grad_norm": 0.5823956727981567,
      "learning_rate": 9.300142235728432e-05,
      "loss": 2.2115,
      "step": 5780
    },
    {
      "epoch": 0.31378393483157485,
      "grad_norm": 0.7447882890701294,
      "learning_rate": 9.295234195915523e-05,
      "loss": 2.2098,
      "step": 5790
    },
    {
      "epoch": 0.31432587599708706,
      "grad_norm": 0.6827070116996765,
      "learning_rate": 9.290310468059389e-05,
      "loss": 2.2105,
      "step": 5800
    },
    {
      "epoch": 0.31486781716259926,
      "grad_norm": 0.42079001665115356,
      "learning_rate": 9.285371072512831e-05,
      "loss": 2.2102,
      "step": 5810
    },
    {
      "epoch": 0.3154097583281115,
      "grad_norm": 0.7562916874885559,
      "learning_rate": 9.280416029693419e-05,
      "loss": 2.2039,
      "step": 5820
    },
    {
      "epoch": 0.3158975053770725,
      "eval_loss": 2.691206455230713,
      "eval_runtime": 48.3032,
      "eval_samples_per_second": 103.513,
      "eval_steps_per_second": 0.559,
      "step": 5829
    },
    {
      "epoch": 0.3159516994936237,
      "grad_norm": 0.6581352949142456,
      "learning_rate": 9.275445360083398e-05,
      "loss": 2.2065,
      "step": 5830
    },
    {
      "epoch": 0.31649364065913593,
      "grad_norm": 0.8172852993011475,
      "learning_rate": 9.270459084229612e-05,
      "loss": 2.2078,
      "step": 5840
    },
    {
      "epoch": 0.31703558182464814,
      "grad_norm": 0.8475619554519653,
      "learning_rate": 9.265457222743414e-05,
      "loss": 2.2049,
      "step": 5850
    },
    {
      "epoch": 0.3175775229901604,
      "grad_norm": 0.37048637866973877,
      "learning_rate": 9.260439796300582e-05,
      "loss": 2.2131,
      "step": 5860
    },
    {
      "epoch": 0.3181194641556726,
      "grad_norm": 0.4438531994819641,
      "learning_rate": 9.255406825641233e-05,
      "loss": 2.1972,
      "step": 5870
    },
    {
      "epoch": 0.3186614053211848,
      "grad_norm": 0.7300620079040527,
      "learning_rate": 9.250358331569737e-05,
      "loss": 2.1957,
      "step": 5880
    },
    {
      "epoch": 0.319203346486697,
      "grad_norm": 0.42140352725982666,
      "learning_rate": 9.245294334954636e-05,
      "loss": 2.2043,
      "step": 5890
    },
    {
      "epoch": 0.3197452876522093,
      "grad_norm": 1.2300440073013306,
      "learning_rate": 9.24021485672855e-05,
      "loss": 2.2077,
      "step": 5900
    },
    {
      "epoch": 0.3202872288177215,
      "grad_norm": 0.9198090434074402,
      "learning_rate": 9.23511991788809e-05,
      "loss": 2.1974,
      "step": 5910
    },
    {
      "epoch": 0.3206123935170288,
      "eval_loss": 2.7107741832733154,
      "eval_runtime": 52.6522,
      "eval_samples_per_second": 94.963,
      "eval_steps_per_second": 0.513,
      "step": 5916
    },
    {
      "epoch": 0.3208291699832337,
      "grad_norm": 0.9036116600036621,
      "learning_rate": 9.230009539493787e-05,
      "loss": 2.2209,
      "step": 5920
    },
    {
      "epoch": 0.3213711111487459,
      "grad_norm": 0.5668349266052246,
      "learning_rate": 9.224883742669982e-05,
      "loss": 2.2166,
      "step": 5930
    },
    {
      "epoch": 0.32191305231425815,
      "grad_norm": 0.5860267281532288,
      "learning_rate": 9.219742548604756e-05,
      "loss": 2.1962,
      "step": 5940
    },
    {
      "epoch": 0.32245499347977036,
      "grad_norm": 0.3655729591846466,
      "learning_rate": 9.214585978549832e-05,
      "loss": 2.2041,
      "step": 5950
    },
    {
      "epoch": 0.32299693464528256,
      "grad_norm": 0.34334370493888855,
      "learning_rate": 9.209414053820495e-05,
      "loss": 2.198,
      "step": 5960
    },
    {
      "epoch": 0.32353887581079477,
      "grad_norm": 0.39599186182022095,
      "learning_rate": 9.2042267957955e-05,
      "loss": 2.1826,
      "step": 5970
    },
    {
      "epoch": 0.32408081697630703,
      "grad_norm": 0.46203556656837463,
      "learning_rate": 9.199024225916982e-05,
      "loss": 2.1947,
      "step": 5980
    },
    {
      "epoch": 0.32462275814181923,
      "grad_norm": 0.43823203444480896,
      "learning_rate": 9.193806365690371e-05,
      "loss": 2.1897,
      "step": 5990
    },
    {
      "epoch": 0.32516469930733144,
      "grad_norm": 0.3208545744419098,
      "learning_rate": 9.1885732366843e-05,
      "loss": 2.1952,
      "step": 6000
    },
    {
      "epoch": 0.00016258234965366573,
      "eval_loss": 2.6887643337249756,
      "eval_runtime": 22.0393,
      "eval_samples_per_second": 226.867,
      "eval_steps_per_second": 1.225,
      "step": 6003
    },
    {
      "epoch": 0.000541941165512219,
      "grad_norm": 0.34268251061439514,
      "learning_rate": 9.183324860530519e-05,
      "loss": 2.1881,
      "step": 6010
    },
    {
      "epoch": 0.001083882331024438,
      "grad_norm": 0.6927510499954224,
      "learning_rate": 9.178061258923802e-05,
      "loss": 2.1964,
      "step": 6020
    },
    {
      "epoch": 0.0016258234965366573,
      "grad_norm": 0.7226220965385437,
      "learning_rate": 9.172782453621862e-05,
      "loss": 2.1923,
      "step": 6030
    },
    {
      "epoch": 0.002167764662048876,
      "grad_norm": 0.6258693337440491,
      "learning_rate": 9.167488466445255e-05,
      "loss": 2.1977,
      "step": 6040
    },
    {
      "epoch": 0.0027097058275610954,
      "grad_norm": 0.9858429431915283,
      "learning_rate": 9.162179319277296e-05,
      "loss": 2.1936,
      "step": 6050
    },
    {
      "epoch": 0.0032516469930733145,
      "grad_norm": 0.6434946656227112,
      "learning_rate": 9.156855034063963e-05,
      "loss": 2.2008,
      "step": 6060
    },
    {
      "epoch": 0.0037935881585855337,
      "grad_norm": 0.49262088537216187,
      "learning_rate": 9.151515632813808e-05,
      "loss": 2.1885,
      "step": 6070
    },
    {
      "epoch": 0.004335529324097752,
      "grad_norm": 0.5512810945510864,
      "learning_rate": 9.14616113759787e-05,
      "loss": 2.1764,
      "step": 6080
    },
    {
      "epoch": 0.0048774704896099716,
      "grad_norm": 0.38034671545028687,
      "learning_rate": 9.140791570549583e-05,
      "loss": 2.1774,
      "step": 6090
    },
    {
      "epoch": 0.0048774704896099716,
      "eval_loss": 2.6898508071899414,
      "eval_runtime": 21.9661,
      "eval_samples_per_second": 227.624,
      "eval_steps_per_second": 1.229,
      "step": 6090
    },
    {
      "epoch": 0.005419411655122191,
      "grad_norm": 0.9372425079345703,
      "learning_rate": 9.135406953864675e-05,
      "loss": 2.1785,
      "step": 6100
    },
    {
      "epoch": 0.00596135282063441,
      "grad_norm": 0.36292022466659546,
      "learning_rate": 9.130007309801089e-05,
      "loss": 2.1849,
      "step": 6110
    },
    {
      "epoch": 0.006503293986146629,
      "grad_norm": 0.47751471400260925,
      "learning_rate": 9.12459266067888e-05,
      "loss": 2.1846,
      "step": 6120
    },
    {
      "epoch": 0.007045235151658848,
      "grad_norm": 0.5181723237037659,
      "learning_rate": 9.119163028880136e-05,
      "loss": 2.1772,
      "step": 6130
    },
    {
      "epoch": 0.007587176317171067,
      "grad_norm": 0.641890823841095,
      "learning_rate": 9.113718436848873e-05,
      "loss": 2.1719,
      "step": 6140
    },
    {
      "epoch": 0.008129117482683286,
      "grad_norm": 0.6258209347724915,
      "learning_rate": 9.108258907090944e-05,
      "loss": 2.1779,
      "step": 6150
    },
    {
      "epoch": 0.008671058648195505,
      "grad_norm": 0.41677314043045044,
      "learning_rate": 9.102784462173954e-05,
      "loss": 2.1814,
      "step": 6160
    },
    {
      "epoch": 0.009212999813707724,
      "grad_norm": 0.4094899594783783,
      "learning_rate": 9.097295124727161e-05,
      "loss": 2.169,
      "step": 6170
    },
    {
      "epoch": 0.009592358629566278,
      "eval_loss": 2.683100461959839,
      "eval_runtime": 22.9235,
      "eval_samples_per_second": 218.117,
      "eval_steps_per_second": 1.178,
      "step": 6177
    },
    {
      "epoch": 0.009754940979219943,
      "grad_norm": 0.6764087080955505,
      "learning_rate": 9.091790917441381e-05,
      "loss": 2.1747,
      "step": 6180
    },
    {
      "epoch": 0.010296882144732162,
      "grad_norm": 0.6722964644432068,
      "learning_rate": 9.086271863068893e-05,
      "loss": 2.1805,
      "step": 6190
    },
    {
      "epoch": 0.010838823310244381,
      "grad_norm": 0.5901824235916138,
      "learning_rate": 9.080737984423358e-05,
      "loss": 2.1904,
      "step": 6200
    },
    {
      "epoch": 0.0113807644757566,
      "grad_norm": 0.4733932614326477,
      "learning_rate": 9.075189304379703e-05,
      "loss": 2.1834,
      "step": 6210
    },
    {
      "epoch": 0.01192270564126882,
      "grad_norm": 0.6376433968544006,
      "learning_rate": 9.06962584587405e-05,
      "loss": 2.1765,
      "step": 6220
    },
    {
      "epoch": 0.012464646806781039,
      "grad_norm": 0.559138298034668,
      "learning_rate": 9.0640476319036e-05,
      "loss": 2.1715,
      "step": 6230
    },
    {
      "epoch": 0.013006587972293258,
      "grad_norm": 0.7046493887901306,
      "learning_rate": 9.05845468552655e-05,
      "loss": 2.1645,
      "step": 6240
    },
    {
      "epoch": 0.013548529137805477,
      "grad_norm": 1.1869386434555054,
      "learning_rate": 9.052847029861999e-05,
      "loss": 2.1667,
      "step": 6250
    },
    {
      "epoch": 0.014090470303317696,
      "grad_norm": 0.5609129071235657,
      "learning_rate": 9.047224688089845e-05,
      "loss": 2.1764,
      "step": 6260
    },
    {
      "epoch": 0.014307246769522584,
      "eval_loss": 2.6828575134277344,
      "eval_runtime": 21.9786,
      "eval_samples_per_second": 227.494,
      "eval_steps_per_second": 1.228,
      "step": 6264
    },
    {
      "epoch": 0.014632411468829916,
      "grad_norm": 0.3735623359680176,
      "learning_rate": 9.041587683450695e-05,
      "loss": 2.159,
      "step": 6270
    },
    {
      "epoch": 0.015174352634342135,
      "grad_norm": 0.5112239718437195,
      "learning_rate": 9.035936039245761e-05,
      "loss": 2.1625,
      "step": 6280
    },
    {
      "epoch": 0.015716293799854352,
      "grad_norm": 0.7340002059936523,
      "learning_rate": 9.030269778836777e-05,
      "loss": 2.1693,
      "step": 6290
    },
    {
      "epoch": 0.01625823496536657,
      "grad_norm": 0.5309348106384277,
      "learning_rate": 9.024588925645889e-05,
      "loss": 2.1689,
      "step": 6300
    },
    {
      "epoch": 0.01680017613087879,
      "grad_norm": 0.33672529458999634,
      "learning_rate": 9.018893503155569e-05,
      "loss": 2.1676,
      "step": 6310
    },
    {
      "epoch": 0.01734211729639101,
      "grad_norm": 0.43856242299079895,
      "learning_rate": 9.013183534908508e-05,
      "loss": 2.1618,
      "step": 6320
    },
    {
      "epoch": 0.01788405846190323,
      "grad_norm": 0.385219931602478,
      "learning_rate": 9.007459044507528e-05,
      "loss": 2.1699,
      "step": 6330
    },
    {
      "epoch": 0.018425999627415448,
      "grad_norm": 0.5022485256195068,
      "learning_rate": 9.001720055615476e-05,
      "loss": 2.1621,
      "step": 6340
    },
    {
      "epoch": 0.018967940792927667,
      "grad_norm": 0.38395243883132935,
      "learning_rate": 8.995966591955132e-05,
      "loss": 2.1586,
      "step": 6350
    },
    {
      "epoch": 0.01902213490947889,
      "eval_loss": 2.6814119815826416,
      "eval_runtime": 23.8855,
      "eval_samples_per_second": 209.332,
      "eval_steps_per_second": 1.13,
      "step": 6351
    },
    {
      "epoch": 0.019509881958439886,
      "grad_norm": 0.3883703649044037,
      "learning_rate": 8.990198677309109e-05,
      "loss": 2.1581,
      "step": 6360
    },
    {
      "epoch": 0.020051823123952105,
      "grad_norm": 0.3477310538291931,
      "learning_rate": 8.984416335519754e-05,
      "loss": 2.1587,
      "step": 6370
    },
    {
      "epoch": 0.020593764289464325,
      "grad_norm": 0.35369783639907837,
      "learning_rate": 8.978619590489055e-05,
      "loss": 2.164,
      "step": 6380
    },
    {
      "epoch": 0.021135705454976544,
      "grad_norm": 0.6048836708068848,
      "learning_rate": 8.972808466178529e-05,
      "loss": 2.1553,
      "step": 6390
    },
    {
      "epoch": 0.021677646620488763,
      "grad_norm": 0.5932040214538574,
      "learning_rate": 8.966982986609141e-05,
      "loss": 2.1536,
      "step": 6400
    },
    {
      "epoch": 0.022219587786000982,
      "grad_norm": 0.6168770790100098,
      "learning_rate": 8.961143175861187e-05,
      "loss": 2.1562,
      "step": 6410
    },
    {
      "epoch": 0.0227615289515132,
      "grad_norm": 0.41919735074043274,
      "learning_rate": 8.955289058074207e-05,
      "loss": 2.1598,
      "step": 6420
    },
    {
      "epoch": 0.02330347011702542,
      "grad_norm": 0.6106112003326416,
      "learning_rate": 8.94942065744688e-05,
      "loss": 2.1558,
      "step": 6430
    },
    {
      "epoch": 0.023737023049435195,
      "eval_loss": 2.683075189590454,
      "eval_runtime": 21.975,
      "eval_samples_per_second": 227.532,
      "eval_steps_per_second": 1.229,
      "step": 6438
    },
    {
      "epoch": 0.02384541128253764,
      "grad_norm": 0.7993687987327576,
      "learning_rate": 8.943537998236922e-05,
      "loss": 2.1442,
      "step": 6440
    },
    {
      "epoch": 0.02438735244804986,
      "grad_norm": 0.6228429079055786,
      "learning_rate": 8.937641104760994e-05,
      "loss": 2.1479,
      "step": 6450
    },
    {
      "epoch": 0.024929293613562078,
      "grad_norm": 0.47318562865257263,
      "learning_rate": 8.931730001394591e-05,
      "loss": 2.1512,
      "step": 6460
    },
    {
      "epoch": 0.025471234779074297,
      "grad_norm": 0.3833015263080597,
      "learning_rate": 8.92580471257195e-05,
      "loss": 2.1565,
      "step": 6470
    },
    {
      "epoch": 0.026013175944586516,
      "grad_norm": 0.357759565114975,
      "learning_rate": 8.919865262785941e-05,
      "loss": 2.15,
      "step": 6480
    },
    {
      "epoch": 0.026555117110098735,
      "grad_norm": 0.478135347366333,
      "learning_rate": 8.913911676587976e-05,
      "loss": 2.1399,
      "step": 6490
    },
    {
      "epoch": 0.027097058275610954,
      "grad_norm": 0.3849744498729706,
      "learning_rate": 8.907943978587896e-05,
      "loss": 2.1505,
      "step": 6500
    },
    {
      "epoch": 0.027638999441123174,
      "grad_norm": 0.7612195611000061,
      "learning_rate": 8.901962193453875e-05,
      "loss": 2.1517,
      "step": 6510
    },
    {
      "epoch": 0.028180940606635393,
      "grad_norm": 0.39389902353286743,
      "learning_rate": 8.895966345912322e-05,
      "loss": 2.1512,
      "step": 6520
    },
    {
      "epoch": 0.028451911189391502,
      "eval_loss": 2.684783935546875,
      "eval_runtime": 21.9732,
      "eval_samples_per_second": 227.55,
      "eval_steps_per_second": 1.229,
      "step": 6525
    },
    {
      "epoch": 0.028722881772147612,
      "grad_norm": 0.3195110559463501,
      "learning_rate": 8.889956460747773e-05,
      "loss": 2.1501,
      "step": 6530
    },
    {
      "epoch": 0.02926482293765983,
      "grad_norm": 0.34488433599472046,
      "learning_rate": 8.883932562802787e-05,
      "loss": 2.1392,
      "step": 6540
    },
    {
      "epoch": 0.02980676410317205,
      "grad_norm": 0.49418726563453674,
      "learning_rate": 8.877894676977848e-05,
      "loss": 2.148,
      "step": 6550
    },
    {
      "epoch": 0.03034870526868427,
      "grad_norm": 0.48197141289711,
      "learning_rate": 8.871842828231265e-05,
      "loss": 2.1441,
      "step": 6560
    },
    {
      "epoch": 0.03089064643419649,
      "grad_norm": 0.5270874500274658,
      "learning_rate": 8.865777041579057e-05,
      "loss": 2.1395,
      "step": 6570
    },
    {
      "epoch": 0.031432587599708704,
      "grad_norm": 0.42525890469551086,
      "learning_rate": 8.859697342094864e-05,
      "loss": 2.1438,
      "step": 6580
    },
    {
      "epoch": 0.03197452876522092,
      "grad_norm": 0.3241468071937561,
      "learning_rate": 8.85360375490983e-05,
      "loss": 2.1313,
      "step": 6590
    },
    {
      "epoch": 0.03251646993073314,
      "grad_norm": 0.4242691695690155,
      "learning_rate": 8.84749630521251e-05,
      "loss": 2.1338,
      "step": 6600
    },
    {
      "epoch": 0.03305841109624536,
      "grad_norm": 0.3246161937713623,
      "learning_rate": 8.84137501824876e-05,
      "loss": 2.1438,
      "step": 6610
    },
    {
      "epoch": 0.03316679932934781,
      "eval_loss": 2.6783671379089355,
      "eval_runtime": 21.9713,
      "eval_samples_per_second": 227.57,
      "eval_steps_per_second": 1.229,
      "step": 6612
    },
    {
      "epoch": 0.03360035226175758,
      "grad_norm": 0.37965840101242065,
      "learning_rate": 8.835239919321632e-05,
      "loss": 2.1324,
      "step": 6620
    },
    {
      "epoch": 0.0341422934272698,
      "grad_norm": 0.6797935962677002,
      "learning_rate": 8.829091033791274e-05,
      "loss": 2.1389,
      "step": 6630
    },
    {
      "epoch": 0.03468423459278202,
      "grad_norm": 0.7013537287712097,
      "learning_rate": 8.822928387074821e-05,
      "loss": 2.1369,
      "step": 6640
    },
    {
      "epoch": 0.03522617575829424,
      "grad_norm": 0.344716340303421,
      "learning_rate": 8.816752004646294e-05,
      "loss": 2.1294,
      "step": 6650
    },
    {
      "epoch": 0.03576811692380646,
      "grad_norm": 0.33730098605155945,
      "learning_rate": 8.810561912036489e-05,
      "loss": 2.148,
      "step": 6660
    },
    {
      "epoch": 0.03631005808931868,
      "grad_norm": 0.41755759716033936,
      "learning_rate": 8.804358134832874e-05,
      "loss": 2.133,
      "step": 6670
    },
    {
      "epoch": 0.036851999254830896,
      "grad_norm": 1.0376006364822388,
      "learning_rate": 8.798140698679489e-05,
      "loss": 2.1425,
      "step": 6680
    },
    {
      "epoch": 0.037393940420343115,
      "grad_norm": 0.673877477645874,
      "learning_rate": 8.791909629276827e-05,
      "loss": 2.1426,
      "step": 6690
    },
    {
      "epoch": 0.037881687469304114,
      "eval_loss": 2.6814136505126953,
      "eval_runtime": 21.9797,
      "eval_samples_per_second": 227.482,
      "eval_steps_per_second": 1.228,
      "step": 6699
    },
    {
      "epoch": 0.037935881585855334,
      "grad_norm": 0.5003832578659058,
      "learning_rate": 8.785664952381746e-05,
      "loss": 2.1438,
      "step": 6700
    },
    {
      "epoch": 0.03847782275136755,
      "grad_norm": 0.4365040957927704,
      "learning_rate": 8.77940669380734e-05,
      "loss": 2.1297,
      "step": 6710
    },
    {
      "epoch": 0.03901976391687977,
      "grad_norm": 0.3096308708190918,
      "learning_rate": 8.773134879422856e-05,
      "loss": 2.134,
      "step": 6720
    },
    {
      "epoch": 0.03956170508239199,
      "grad_norm": 0.9517742395401001,
      "learning_rate": 8.766849535153568e-05,
      "loss": 2.1274,
      "step": 6730
    },
    {
      "epoch": 0.04010364624790421,
      "grad_norm": 0.4467058777809143,
      "learning_rate": 8.760550686980681e-05,
      "loss": 2.1337,
      "step": 6740
    },
    {
      "epoch": 0.04064558741341643,
      "grad_norm": 0.619213879108429,
      "learning_rate": 8.754238360941218e-05,
      "loss": 2.1252,
      "step": 6750
    },
    {
      "epoch": 0.04118752857892865,
      "grad_norm": 0.5627840161323547,
      "learning_rate": 8.747912583127913e-05,
      "loss": 2.1229,
      "step": 6760
    },
    {
      "epoch": 0.04172946974444087,
      "grad_norm": 0.34763067960739136,
      "learning_rate": 8.741573379689109e-05,
      "loss": 2.1238,
      "step": 6770
    },
    {
      "epoch": 0.04227141090995309,
      "grad_norm": 0.45356321334838867,
      "learning_rate": 8.735220776828641e-05,
      "loss": 2.1258,
      "step": 6780
    },
    {
      "epoch": 0.04259657560926042,
      "eval_loss": 2.6720757484436035,
      "eval_runtime": 25.5872,
      "eval_samples_per_second": 195.41,
      "eval_steps_per_second": 1.055,
      "step": 6786
    },
    {
      "epoch": 0.04281335207546531,
      "grad_norm": 0.3579098880290985,
      "learning_rate": 8.728854800805733e-05,
      "loss": 2.122,
      "step": 6790
    },
    {
      "epoch": 0.043355293240977526,
      "grad_norm": 0.49912065267562866,
      "learning_rate": 8.722475477934894e-05,
      "loss": 2.1271,
      "step": 6800
    },
    {
      "epoch": 0.043897234406489745,
      "grad_norm": 0.5217333436012268,
      "learning_rate": 8.716082834585797e-05,
      "loss": 2.1158,
      "step": 6810
    },
    {
      "epoch": 0.044439175572001964,
      "grad_norm": 0.6708068251609802,
      "learning_rate": 8.709676897183176e-05,
      "loss": 2.1309,
      "step": 6820
    },
    {
      "epoch": 0.04498111673751418,
      "grad_norm": 0.4707426130771637,
      "learning_rate": 8.703257692206724e-05,
      "loss": 2.1155,
      "step": 6830
    },
    {
      "epoch": 0.0455230579030264,
      "grad_norm": 0.5237321257591248,
      "learning_rate": 8.696825246190972e-05,
      "loss": 2.1245,
      "step": 6840
    },
    {
      "epoch": 0.04606499906853862,
      "grad_norm": 0.6378241181373596,
      "learning_rate": 8.690379585725186e-05,
      "loss": 2.118,
      "step": 6850
    },
    {
      "epoch": 0.04660694023405084,
      "grad_norm": 0.33660611510276794,
      "learning_rate": 8.683920737453254e-05,
      "loss": 2.128,
      "step": 6860
    },
    {
      "epoch": 0.04714888139956306,
      "grad_norm": 0.2975222170352936,
      "learning_rate": 8.677448728073583e-05,
      "loss": 2.1156,
      "step": 6870
    },
    {
      "epoch": 0.04731146374921673,
      "eval_loss": 2.673781394958496,
      "eval_runtime": 22.7337,
      "eval_samples_per_second": 219.938,
      "eval_steps_per_second": 1.188,
      "step": 6873
    },
    {
      "epoch": 0.04769082256507528,
      "grad_norm": 0.34510669112205505,
      "learning_rate": 8.670963584338975e-05,
      "loss": 2.1145,
      "step": 6880
    },
    {
      "epoch": 0.0482327637305875,
      "grad_norm": 0.3134111166000366,
      "learning_rate": 8.664465333056526e-05,
      "loss": 2.1195,
      "step": 6890
    },
    {
      "epoch": 0.04877470489609972,
      "grad_norm": 0.37668564915657043,
      "learning_rate": 8.657954001087521e-05,
      "loss": 2.1195,
      "step": 6900
    },
    {
      "epoch": 0.049316646061611936,
      "grad_norm": 0.8455840945243835,
      "learning_rate": 8.651429615347309e-05,
      "loss": 2.1116,
      "step": 6910
    },
    {
      "epoch": 0.049858587227124156,
      "grad_norm": 1.0009831190109253,
      "learning_rate": 8.644892202805195e-05,
      "loss": 2.1119,
      "step": 6920
    },
    {
      "epoch": 0.050400528392636375,
      "grad_norm": 0.9167200326919556,
      "learning_rate": 8.638341790484341e-05,
      "loss": 2.1142,
      "step": 6930
    },
    {
      "epoch": 0.050942469558148594,
      "grad_norm": 0.6210402250289917,
      "learning_rate": 8.631778405461638e-05,
      "loss": 2.1147,
      "step": 6940
    },
    {
      "epoch": 0.05148441072366081,
      "grad_norm": 0.6486452221870422,
      "learning_rate": 8.625202074867607e-05,
      "loss": 2.107,
      "step": 6950
    },
    {
      "epoch": 0.05202635188917303,
      "grad_norm": 0.2966337502002716,
      "learning_rate": 8.618612825886272e-05,
      "loss": 2.0978,
      "step": 6960
    },
    {
      "epoch": 0.05202635188917303,
      "eval_loss": 2.667224168777466,
      "eval_runtime": 21.9683,
      "eval_samples_per_second": 227.601,
      "eval_steps_per_second": 1.229,
      "step": 6960
    },
    {
      "epoch": 0.05256829305468525,
      "grad_norm": 1.5529571771621704,
      "learning_rate": 8.612010685755066e-05,
      "loss": 2.1092,
      "step": 6970
    },
    {
      "epoch": 0.05311023422019747,
      "grad_norm": 0.6471717357635498,
      "learning_rate": 8.605395681764706e-05,
      "loss": 2.1193,
      "step": 6980
    },
    {
      "epoch": 0.05365217538570969,
      "grad_norm": 0.5466669797897339,
      "learning_rate": 8.598767841259078e-05,
      "loss": 2.1173,
      "step": 6990
    },
    {
      "epoch": 0.05419411655122191,
      "grad_norm": 0.3719353973865509,
      "learning_rate": 8.592127191635138e-05,
      "loss": 2.1186,
      "step": 7000
    },
    {
      "epoch": 0.05473605771673413,
      "grad_norm": 0.6376820206642151,
      "learning_rate": 8.585473760342786e-05,
      "loss": 2.1094,
      "step": 7010
    },
    {
      "epoch": 0.05527799888224635,
      "grad_norm": 0.36836645007133484,
      "learning_rate": 8.578807574884756e-05,
      "loss": 2.1028,
      "step": 7020
    },
    {
      "epoch": 0.055819940047758566,
      "grad_norm": 0.318914532661438,
      "learning_rate": 8.572128662816498e-05,
      "loss": 2.107,
      "step": 7030
    },
    {
      "epoch": 0.056361881213270786,
      "grad_norm": 0.5992175340652466,
      "learning_rate": 8.56543705174608e-05,
      "loss": 2.1014,
      "step": 7040
    },
    {
      "epoch": 0.056741240029129336,
      "eval_loss": 2.667701244354248,
      "eval_runtime": 21.977,
      "eval_samples_per_second": 227.51,
      "eval_steps_per_second": 1.229,
      "step": 7047
    },
    {
      "epoch": 0.056903822378783005,
      "grad_norm": 0.3516386151313782,
      "learning_rate": 8.558732769334055e-05,
      "loss": 2.1166,
      "step": 7050
    },
    {
      "epoch": 0.057445763544295224,
      "grad_norm": 1.2428702116012573,
      "learning_rate": 8.552015843293358e-05,
      "loss": 2.1083,
      "step": 7060
    },
    {
      "epoch": 0.05798770470980744,
      "grad_norm": 0.812322735786438,
      "learning_rate": 8.545286301389183e-05,
      "loss": 2.1043,
      "step": 7070
    },
    {
      "epoch": 0.05852964587531966,
      "grad_norm": 0.5382115840911865,
      "learning_rate": 8.538544171438879e-05,
      "loss": 2.0998,
      "step": 7080
    },
    {
      "epoch": 0.05907158704083188,
      "grad_norm": 0.7883318662643433,
      "learning_rate": 8.531789481311824e-05,
      "loss": 2.1118,
      "step": 7090
    },
    {
      "epoch": 0.0596135282063441,
      "grad_norm": 0.3522588312625885,
      "learning_rate": 8.525022258929319e-05,
      "loss": 2.1094,
      "step": 7100
    },
    {
      "epoch": 0.06015546937185632,
      "grad_norm": 0.5178406238555908,
      "learning_rate": 8.518242532264468e-05,
      "loss": 2.0969,
      "step": 7110
    },
    {
      "epoch": 0.06069741053736854,
      "grad_norm": 0.6928080916404724,
      "learning_rate": 8.511450329342061e-05,
      "loss": 2.0937,
      "step": 7120
    },
    {
      "epoch": 0.06123935170288076,
      "grad_norm": 0.33726730942726135,
      "learning_rate": 8.504645678238462e-05,
      "loss": 2.0893,
      "step": 7130
    },
    {
      "epoch": 0.06145612816908565,
      "eval_loss": 2.6699295043945312,
      "eval_runtime": 21.9715,
      "eval_samples_per_second": 227.567,
      "eval_steps_per_second": 1.229,
      "step": 7134
    },
    {
      "epoch": 0.06178129286839298,
      "grad_norm": 0.5048802495002747,
      "learning_rate": 8.497828607081488e-05,
      "loss": 2.1111,
      "step": 7140
    },
    {
      "epoch": 0.062323234033905196,
      "grad_norm": 0.3473956286907196,
      "learning_rate": 8.490999144050299e-05,
      "loss": 2.1016,
      "step": 7150
    },
    {
      "epoch": 0.06286517519941741,
      "grad_norm": 0.3585287928581238,
      "learning_rate": 8.484157317375276e-05,
      "loss": 2.091,
      "step": 7160
    },
    {
      "epoch": 0.06340711636492963,
      "grad_norm": 0.929985761642456,
      "learning_rate": 8.47730315533791e-05,
      "loss": 2.103,
      "step": 7170
    },
    {
      "epoch": 0.06394905753044185,
      "grad_norm": 0.5229865312576294,
      "learning_rate": 8.470436686270678e-05,
      "loss": 2.09,
      "step": 7180
    },
    {
      "epoch": 0.06449099869595407,
      "grad_norm": 0.6161647439002991,
      "learning_rate": 8.463557938556928e-05,
      "loss": 2.0994,
      "step": 7190
    },
    {
      "epoch": 0.06503293986146629,
      "grad_norm": 0.4682093858718872,
      "learning_rate": 8.456666940630772e-05,
      "loss": 2.094,
      "step": 7200
    },
    {
      "epoch": 0.0655748810269785,
      "grad_norm": 0.35867783427238464,
      "learning_rate": 8.449763720976947e-05,
      "loss": 2.0879,
      "step": 7210
    },
    {
      "epoch": 0.06611682219249072,
      "grad_norm": 0.44877171516418457,
      "learning_rate": 8.442848308130723e-05,
      "loss": 2.0922,
      "step": 7220
    },
    {
      "epoch": 0.06617101630904194,
      "eval_loss": 2.6712543964385986,
      "eval_runtime": 21.9738,
      "eval_samples_per_second": 227.544,
      "eval_steps_per_second": 1.229,
      "step": 7221
    },
    {
      "epoch": 0.06665876335800294,
      "grad_norm": 0.4711715877056122,
      "learning_rate": 8.435920730677762e-05,
      "loss": 2.0889,
      "step": 7230
    },
    {
      "epoch": 0.06720070452351516,
      "grad_norm": 0.5644111037254333,
      "learning_rate": 8.428981017254012e-05,
      "loss": 2.0885,
      "step": 7240
    },
    {
      "epoch": 0.06774264568902738,
      "grad_norm": 0.6288381814956665,
      "learning_rate": 8.42202919654559e-05,
      "loss": 2.0922,
      "step": 7250
    },
    {
      "epoch": 0.0682845868545396,
      "grad_norm": 0.39220157265663147,
      "learning_rate": 8.41506529728866e-05,
      "loss": 2.0756,
      "step": 7260
    },
    {
      "epoch": 0.06882652802005182,
      "grad_norm": 0.4057351052761078,
      "learning_rate": 8.408089348269307e-05,
      "loss": 2.0918,
      "step": 7270
    },
    {
      "epoch": 0.06936846918556404,
      "grad_norm": 0.4261212944984436,
      "learning_rate": 8.401101378323434e-05,
      "loss": 2.0806,
      "step": 7280
    },
    {
      "epoch": 0.06991041035107626,
      "grad_norm": 0.6397058367729187,
      "learning_rate": 8.394101416336627e-05,
      "loss": 2.0885,
      "step": 7290
    },
    {
      "epoch": 0.07045235151658848,
      "grad_norm": 0.7682284712791443,
      "learning_rate": 8.387089491244048e-05,
      "loss": 2.0817,
      "step": 7300
    },
    {
      "epoch": 0.07088590444899825,
      "eval_loss": 2.657864809036255,
      "eval_runtime": 21.9777,
      "eval_samples_per_second": 227.503,
      "eval_steps_per_second": 1.229,
      "step": 7308
    },
    {
      "epoch": 0.0709942926821007,
      "grad_norm": 0.4054642617702484,
      "learning_rate": 8.380065632030305e-05,
      "loss": 2.093,
      "step": 7310
    },
    {
      "epoch": 0.07153623384761292,
      "grad_norm": 0.42616334557533264,
      "learning_rate": 8.37302986772934e-05,
      "loss": 2.0864,
      "step": 7320
    },
    {
      "epoch": 0.07207817501312513,
      "grad_norm": 0.5650444030761719,
      "learning_rate": 8.365982227424306e-05,
      "loss": 2.0884,
      "step": 7330
    },
    {
      "epoch": 0.07262011617863735,
      "grad_norm": 0.9685772061347961,
      "learning_rate": 8.358922740247447e-05,
      "loss": 2.0863,
      "step": 7340
    },
    {
      "epoch": 0.07316205734414957,
      "grad_norm": 0.8171983957290649,
      "learning_rate": 8.351851435379974e-05,
      "loss": 2.0896,
      "step": 7350
    },
    {
      "epoch": 0.07370399850966179,
      "grad_norm": 0.6826300024986267,
      "learning_rate": 8.34476834205195e-05,
      "loss": 2.0914,
      "step": 7360
    },
    {
      "epoch": 0.07424593967517401,
      "grad_norm": 0.8995702266693115,
      "learning_rate": 8.337673489542172e-05,
      "loss": 2.0719,
      "step": 7370
    },
    {
      "epoch": 0.07478788084068623,
      "grad_norm": 0.3203958570957184,
      "learning_rate": 8.330566907178038e-05,
      "loss": 2.0861,
      "step": 7380
    },
    {
      "epoch": 0.07532982200619845,
      "grad_norm": 0.3850855827331543,
      "learning_rate": 8.323448624335435e-05,
      "loss": 2.0916,
      "step": 7390
    },
    {
      "epoch": 0.07560079258895457,
      "eval_loss": 2.671057939529419,
      "eval_runtime": 21.9786,
      "eval_samples_per_second": 227.494,
      "eval_steps_per_second": 1.228,
      "step": 7395
    },
    {
      "epoch": 0.07587176317171067,
      "grad_norm": 1.0707188844680786,
      "learning_rate": 8.316318670438614e-05,
      "loss": 2.0833,
      "step": 7400
    },
    {
      "epoch": 0.07641370433722289,
      "grad_norm": 0.36094823479652405,
      "learning_rate": 8.309177074960073e-05,
      "loss": 2.087,
      "step": 7410
    },
    {
      "epoch": 0.0769556455027351,
      "grad_norm": 0.34225502610206604,
      "learning_rate": 8.30202386742043e-05,
      "loss": 2.0898,
      "step": 7420
    },
    {
      "epoch": 0.07749758666824733,
      "grad_norm": 0.3515205681324005,
      "learning_rate": 8.294859077388301e-05,
      "loss": 2.0787,
      "step": 7430
    },
    {
      "epoch": 0.07803952783375954,
      "grad_norm": 0.4942980408668518,
      "learning_rate": 8.287682734480182e-05,
      "loss": 2.0727,
      "step": 7440
    },
    {
      "epoch": 0.07858146899927176,
      "grad_norm": 0.5919073224067688,
      "learning_rate": 8.280494868360325e-05,
      "loss": 2.085,
      "step": 7450
    },
    {
      "epoch": 0.07912341016478398,
      "grad_norm": 0.5980909466743469,
      "learning_rate": 8.27329550874061e-05,
      "loss": 2.0768,
      "step": 7460
    },
    {
      "epoch": 0.0796653513302962,
      "grad_norm": 0.4768329858779907,
      "learning_rate": 8.266084685380434e-05,
      "loss": 2.0824,
      "step": 7470
    },
    {
      "epoch": 0.08020729249580842,
      "grad_norm": 0.6061816215515137,
      "learning_rate": 8.258862428086572e-05,
      "loss": 2.0784,
      "step": 7480
    },
    {
      "epoch": 0.08031568072891086,
      "eval_loss": 2.6704938411712646,
      "eval_runtime": 21.9746,
      "eval_samples_per_second": 227.536,
      "eval_steps_per_second": 1.229,
      "step": 7482
    },
    {
      "epoch": 0.08074923366132064,
      "grad_norm": 0.4925580322742462,
      "learning_rate": 8.251628766713068e-05,
      "loss": 2.0779,
      "step": 7490
    },
    {
      "epoch": 0.08129117482683286,
      "grad_norm": 0.38380166888237,
      "learning_rate": 8.244383731161109e-05,
      "loss": 2.0689,
      "step": 7500
    },
    {
      "epoch": 0.08183311599234508,
      "grad_norm": 0.7923119068145752,
      "learning_rate": 8.237127351378889e-05,
      "loss": 2.0795,
      "step": 7510
    },
    {
      "epoch": 0.0823750571578573,
      "grad_norm": 0.6250379085540771,
      "learning_rate": 8.229859657361504e-05,
      "loss": 2.0803,
      "step": 7520
    },
    {
      "epoch": 0.08291699832336952,
      "grad_norm": 0.32212239503860474,
      "learning_rate": 8.222580679150813e-05,
      "loss": 2.0658,
      "step": 7530
    },
    {
      "epoch": 0.08345893948888174,
      "grad_norm": 0.35104089975357056,
      "learning_rate": 8.215290446835322e-05,
      "loss": 2.0689,
      "step": 7540
    },
    {
      "epoch": 0.08400088065439396,
      "grad_norm": 0.6346544623374939,
      "learning_rate": 8.207988990550055e-05,
      "loss": 2.0708,
      "step": 7550
    },
    {
      "epoch": 0.08454282181990617,
      "grad_norm": 0.3987348675727844,
      "learning_rate": 8.200676340476437e-05,
      "loss": 2.0642,
      "step": 7560
    },
    {
      "epoch": 0.08503056886886717,
      "eval_loss": 2.6780905723571777,
      "eval_runtime": 21.979,
      "eval_samples_per_second": 227.49,
      "eval_steps_per_second": 1.228,
      "step": 7569
    },
    {
      "epoch": 0.0850847629854184,
      "grad_norm": 0.9453224539756775,
      "learning_rate": 8.193352526842159e-05,
      "loss": 2.0647,
      "step": 7570
    },
    {
      "epoch": 0.08562670415093061,
      "grad_norm": 0.3680925667285919,
      "learning_rate": 8.186017579921055e-05,
      "loss": 2.0611,
      "step": 7580
    },
    {
      "epoch": 0.08616864531644283,
      "grad_norm": 0.8765574097633362,
      "learning_rate": 8.178671530032988e-05,
      "loss": 2.0722,
      "step": 7590
    },
    {
      "epoch": 0.08671058648195505,
      "grad_norm": 0.674678385257721,
      "learning_rate": 8.171314407543708e-05,
      "loss": 2.0738,
      "step": 7600
    },
    {
      "epoch": 0.08725252764746727,
      "grad_norm": 0.582229495048523,
      "learning_rate": 8.163946242864744e-05,
      "loss": 2.0725,
      "step": 7610
    },
    {
      "epoch": 0.08779446881297949,
      "grad_norm": 0.4998132288455963,
      "learning_rate": 8.15656706645326e-05,
      "loss": 2.0749,
      "step": 7620
    },
    {
      "epoch": 0.08833640997849171,
      "grad_norm": 0.38911086320877075,
      "learning_rate": 8.149176908811947e-05,
      "loss": 2.0665,
      "step": 7630
    },
    {
      "epoch": 0.08887835114400393,
      "grad_norm": 0.7126851081848145,
      "learning_rate": 8.141775800488877e-05,
      "loss": 2.0599,
      "step": 7640
    },
    {
      "epoch": 0.08942029230951615,
      "grad_norm": 0.6182229518890381,
      "learning_rate": 8.134363772077399e-05,
      "loss": 2.0657,
      "step": 7650
    },
    {
      "epoch": 0.08974545700882348,
      "eval_loss": 2.663647174835205,
      "eval_runtime": 21.9735,
      "eval_samples_per_second": 227.547,
      "eval_steps_per_second": 1.229,
      "step": 7656
    },
    {
      "epoch": 0.08996223347502837,
      "grad_norm": 0.5466020107269287,
      "learning_rate": 8.126940854215997e-05,
      "loss": 2.0764,
      "step": 7660
    },
    {
      "epoch": 0.09050417464054059,
      "grad_norm": 0.7600970268249512,
      "learning_rate": 8.119507077588165e-05,
      "loss": 2.0711,
      "step": 7670
    },
    {
      "epoch": 0.0910461158060528,
      "grad_norm": 0.3224121034145355,
      "learning_rate": 8.11206247292229e-05,
      "loss": 2.0595,
      "step": 7680
    },
    {
      "epoch": 0.09158805697156502,
      "grad_norm": 0.3413337171077728,
      "learning_rate": 8.10460707099151e-05,
      "loss": 2.0586,
      "step": 7690
    },
    {
      "epoch": 0.09212999813707724,
      "grad_norm": 0.337230384349823,
      "learning_rate": 8.097140902613601e-05,
      "loss": 2.0525,
      "step": 7700
    },
    {
      "epoch": 0.09267193930258946,
      "grad_norm": 1.0965609550476074,
      "learning_rate": 8.089663998650839e-05,
      "loss": 2.0618,
      "step": 7710
    },
    {
      "epoch": 0.09321388046810168,
      "grad_norm": 0.7565444707870483,
      "learning_rate": 8.082176390009878e-05,
      "loss": 2.0523,
      "step": 7720
    },
    {
      "epoch": 0.0937558216336139,
      "grad_norm": 0.49668484926223755,
      "learning_rate": 8.074678107641623e-05,
      "loss": 2.0595,
      "step": 7730
    },
    {
      "epoch": 0.09429776279912612,
      "grad_norm": 0.9314787983894348,
      "learning_rate": 8.067169182541099e-05,
      "loss": 2.0604,
      "step": 7740
    },
    {
      "epoch": 0.09446034514877978,
      "eval_loss": 2.659668445587158,
      "eval_runtime": 21.9747,
      "eval_samples_per_second": 227.535,
      "eval_steps_per_second": 1.229,
      "step": 7743
    },
    {
      "epoch": 0.09483970396463834,
      "grad_norm": 0.6480481624603271,
      "learning_rate": 8.059649645747325e-05,
      "loss": 2.0626,
      "step": 7750
    },
    {
      "epoch": 0.09538164513015056,
      "grad_norm": 0.9633429050445557,
      "learning_rate": 8.052119528343181e-05,
      "loss": 2.0628,
      "step": 7760
    },
    {
      "epoch": 0.09592358629566278,
      "grad_norm": 0.5564477443695068,
      "learning_rate": 8.044578861455286e-05,
      "loss": 2.0621,
      "step": 7770
    },
    {
      "epoch": 0.096465527461175,
      "grad_norm": 0.613134503364563,
      "learning_rate": 8.037027676253866e-05,
      "loss": 2.059,
      "step": 7780
    },
    {
      "epoch": 0.09700746862668722,
      "grad_norm": 0.7734301686286926,
      "learning_rate": 8.029466003952628e-05,
      "loss": 2.062,
      "step": 7790
    },
    {
      "epoch": 0.09754940979219943,
      "grad_norm": 0.31820887327194214,
      "learning_rate": 8.021893875808625e-05,
      "loss": 2.0541,
      "step": 7800
    },
    {
      "epoch": 0.09809135095771165,
      "grad_norm": 0.3358112573623657,
      "learning_rate": 8.014311323122131e-05,
      "loss": 2.0547,
      "step": 7810
    },
    {
      "epoch": 0.09863329212322387,
      "grad_norm": 0.5778986811637878,
      "learning_rate": 8.006718377236514e-05,
      "loss": 2.057,
      "step": 7820
    },
    {
      "epoch": 0.09917523328873609,
      "grad_norm": 0.318333238363266,
      "learning_rate": 7.999115069538099e-05,
      "loss": 2.0503,
      "step": 7830
    },
    {
      "epoch": 0.09917523328873609,
      "eval_loss": 2.6534852981567383,
      "eval_runtime": 21.973,
      "eval_samples_per_second": 227.552,
      "eval_steps_per_second": 1.229,
      "step": 7830
    },
    {
      "epoch": 0.09971717445424831,
      "grad_norm": 0.5033796429634094,
      "learning_rate": 7.991501431456047e-05,
      "loss": 2.0504,
      "step": 7840
    },
    {
      "epoch": 0.10025911561976053,
      "grad_norm": 0.39477086067199707,
      "learning_rate": 7.983877494462219e-05,
      "loss": 2.0601,
      "step": 7850
    },
    {
      "epoch": 0.10080105678527275,
      "grad_norm": 0.37355226278305054,
      "learning_rate": 7.976243290071045e-05,
      "loss": 2.0529,
      "step": 7860
    },
    {
      "epoch": 0.10134299795078497,
      "grad_norm": 0.6517744064331055,
      "learning_rate": 7.968598849839404e-05,
      "loss": 2.0555,
      "step": 7870
    },
    {
      "epoch": 0.10188493911629719,
      "grad_norm": 0.3515459895133972,
      "learning_rate": 7.960944205366478e-05,
      "loss": 2.0515,
      "step": 7880
    },
    {
      "epoch": 0.10242688028180941,
      "grad_norm": 0.590758740901947,
      "learning_rate": 7.953279388293634e-05,
      "loss": 2.0536,
      "step": 7890
    },
    {
      "epoch": 0.10296882144732163,
      "grad_norm": 0.3137708008289337,
      "learning_rate": 7.945604430304289e-05,
      "loss": 2.0495,
      "step": 7900
    },
    {
      "epoch": 0.10351076261283385,
      "grad_norm": 0.5091531872749329,
      "learning_rate": 7.93791936312377e-05,
      "loss": 2.0461,
      "step": 7910
    },
    {
      "epoch": 0.1038901214286924,
      "eval_loss": 2.6576335430145264,
      "eval_runtime": 21.9728,
      "eval_samples_per_second": 227.554,
      "eval_steps_per_second": 1.229,
      "step": 7917
    },
    {
      "epoch": 0.10405270377834606,
      "grad_norm": 0.33678263425827026,
      "learning_rate": 7.930224218519207e-05,
      "loss": 2.0508,
      "step": 7920
    },
    {
      "epoch": 0.10459464494385828,
      "grad_norm": 0.36633098125457764,
      "learning_rate": 7.922519028299376e-05,
      "loss": 2.0451,
      "step": 7930
    },
    {
      "epoch": 0.1051365861093705,
      "grad_norm": 0.531390905380249,
      "learning_rate": 7.914803824314579e-05,
      "loss": 2.0512,
      "step": 7940
    },
    {
      "epoch": 0.10567852727488272,
      "grad_norm": 0.6033332347869873,
      "learning_rate": 7.907078638456506e-05,
      "loss": 2.0421,
      "step": 7950
    },
    {
      "epoch": 0.10622046844039494,
      "grad_norm": 0.7366685271263123,
      "learning_rate": 7.899343502658123e-05,
      "loss": 2.0543,
      "step": 7960
    },
    {
      "epoch": 0.10676240960590716,
      "grad_norm": 0.5616899132728577,
      "learning_rate": 7.891598448893508e-05,
      "loss": 2.0458,
      "step": 7970
    },
    {
      "epoch": 0.10730435077141938,
      "grad_norm": 0.7124751806259155,
      "learning_rate": 7.88384350917775e-05,
      "loss": 2.0412,
      "step": 7980
    },
    {
      "epoch": 0.1078462919369316,
      "grad_norm": 0.5059435963630676,
      "learning_rate": 7.876078715566794e-05,
      "loss": 2.0452,
      "step": 7990
    },
    {
      "epoch": 0.10838823310244382,
      "grad_norm": 0.297770619392395,
      "learning_rate": 7.868304100157318e-05,
      "loss": 2.0514,
      "step": 8000
    },
    {
      "epoch": 0.1086050095686487,
      "eval_loss": 2.6553268432617188,
      "eval_runtime": 21.9655,
      "eval_samples_per_second": 227.63,
      "eval_steps_per_second": 1.229,
      "step": 8004
    },
    {
      "epoch": 0.10893017426795604,
      "grad_norm": 0.890073299407959,
      "learning_rate": 7.860519695086608e-05,
      "loss": 2.0421,
      "step": 8010
    },
    {
      "epoch": 0.10947211543346826,
      "grad_norm": 0.9192163348197937,
      "learning_rate": 7.852725532532405e-05,
      "loss": 2.0366,
      "step": 8020
    },
    {
      "epoch": 0.11001405659898048,
      "grad_norm": 0.3507506847381592,
      "learning_rate": 7.84492164471279e-05,
      "loss": 2.0436,
      "step": 8030
    },
    {
      "epoch": 0.1105559977644927,
      "grad_norm": 0.315594881772995,
      "learning_rate": 7.837108063886046e-05,
      "loss": 2.0421,
      "step": 8040
    },
    {
      "epoch": 0.11109793893000491,
      "grad_norm": 0.6328078508377075,
      "learning_rate": 7.829284822350516e-05,
      "loss": 2.0414,
      "step": 8050
    },
    {
      "epoch": 0.11163988009551713,
      "grad_norm": 0.6946396231651306,
      "learning_rate": 7.821451952444487e-05,
      "loss": 2.0334,
      "step": 8060
    },
    {
      "epoch": 0.11218182126102935,
      "grad_norm": 0.3898144066333771,
      "learning_rate": 7.813609486546036e-05,
      "loss": 2.0386,
      "step": 8070
    },
    {
      "epoch": 0.11272376242654157,
      "grad_norm": 0.8166786432266235,
      "learning_rate": 7.805757457072913e-05,
      "loss": 2.0515,
      "step": 8080
    },
    {
      "epoch": 0.11326570359205379,
      "grad_norm": 0.36218130588531494,
      "learning_rate": 7.797895896482396e-05,
      "loss": 2.0395,
      "step": 8090
    },
    {
      "epoch": 0.11331989770860501,
      "eval_loss": 2.6553170680999756,
      "eval_runtime": 21.9791,
      "eval_samples_per_second": 227.489,
      "eval_steps_per_second": 1.228,
      "step": 8091
    },
    {
      "epoch": 0.11380764475756601,
      "grad_norm": 0.34464749693870544,
      "learning_rate": 7.790024837271165e-05,
      "loss": 2.0413,
      "step": 8100
    },
    {
      "epoch": 0.11434958592307823,
      "grad_norm": 0.46647629141807556,
      "learning_rate": 7.782144311975158e-05,
      "loss": 2.047,
      "step": 8110
    },
    {
      "epoch": 0.11489152708859045,
      "grad_norm": 0.30626147985458374,
      "learning_rate": 7.77425435316945e-05,
      "loss": 2.0349,
      "step": 8120
    },
    {
      "epoch": 0.11543346825410267,
      "grad_norm": 0.3106520473957062,
      "learning_rate": 7.7663549934681e-05,
      "loss": 2.031,
      "step": 8130
    },
    {
      "epoch": 0.11597540941961489,
      "grad_norm": 0.6385337710380554,
      "learning_rate": 7.758446265524038e-05,
      "loss": 2.026,
      "step": 8140
    },
    {
      "epoch": 0.1165173505851271,
      "grad_norm": 0.7750070095062256,
      "learning_rate": 7.750528202028912e-05,
      "loss": 2.0243,
      "step": 8150
    },
    {
      "epoch": 0.11705929175063932,
      "grad_norm": 0.5177359580993652,
      "learning_rate": 7.74260083571296e-05,
      "loss": 2.0276,
      "step": 8160
    },
    {
      "epoch": 0.11760123291615154,
      "grad_norm": 0.9061049818992615,
      "learning_rate": 7.734664199344876e-05,
      "loss": 2.0382,
      "step": 8170
    },
    {
      "epoch": 0.11803478584856132,
      "eval_loss": 2.6514930725097656,
      "eval_runtime": 21.9691,
      "eval_samples_per_second": 227.592,
      "eval_steps_per_second": 1.229,
      "step": 8178
    },
    {
      "epoch": 0.11814317408166376,
      "grad_norm": 0.8364623188972473,
      "learning_rate": 7.726718325731671e-05,
      "loss": 2.0361,
      "step": 8180
    },
    {
      "epoch": 0.11868511524717598,
      "grad_norm": 0.8486116528511047,
      "learning_rate": 7.718763247718542e-05,
      "loss": 2.0466,
      "step": 8190
    },
    {
      "epoch": 0.1192270564126882,
      "grad_norm": 0.9383118152618408,
      "learning_rate": 7.710798998188731e-05,
      "loss": 2.0303,
      "step": 8200
    },
    {
      "epoch": 0.11976899757820042,
      "grad_norm": 0.6073142886161804,
      "learning_rate": 7.702825610063393e-05,
      "loss": 2.0357,
      "step": 8210
    },
    {
      "epoch": 0.12031093874371264,
      "grad_norm": 0.46166083216667175,
      "learning_rate": 7.694843116301458e-05,
      "loss": 2.0319,
      "step": 8220
    },
    {
      "epoch": 0.12085287990922486,
      "grad_norm": 0.5513178110122681,
      "learning_rate": 7.686851549899494e-05,
      "loss": 2.0273,
      "step": 8230
    },
    {
      "epoch": 0.12139482107473708,
      "grad_norm": 0.4025670289993286,
      "learning_rate": 7.678850943891573e-05,
      "loss": 2.0338,
      "step": 8240
    },
    {
      "epoch": 0.1219367622402493,
      "grad_norm": 0.7250736355781555,
      "learning_rate": 7.670841331349134e-05,
      "loss": 2.0223,
      "step": 8250
    },
    {
      "epoch": 0.12247870340576152,
      "grad_norm": 0.5293325185775757,
      "learning_rate": 7.662822745380845e-05,
      "loss": 2.0363,
      "step": 8260
    },
    {
      "epoch": 0.12274967398851762,
      "eval_loss": 2.6460092067718506,
      "eval_runtime": 21.9726,
      "eval_samples_per_second": 227.556,
      "eval_steps_per_second": 1.229,
      "step": 8265
    },
    {
      "epoch": 0.12302064457127374,
      "grad_norm": 0.4530751407146454,
      "learning_rate": 7.654795219132465e-05,
      "loss": 2.0366,
      "step": 8270
    },
    {
      "epoch": 0.12356258573678595,
      "grad_norm": 0.47878649830818176,
      "learning_rate": 7.64675878578671e-05,
      "loss": 2.0289,
      "step": 8280
    },
    {
      "epoch": 0.12410452690229817,
      "grad_norm": 0.5691394209861755,
      "learning_rate": 7.638713478563116e-05,
      "loss": 2.023,
      "step": 8290
    },
    {
      "epoch": 0.12464646806781039,
      "grad_norm": 0.5451269149780273,
      "learning_rate": 7.630659330717899e-05,
      "loss": 2.0293,
      "step": 8300
    },
    {
      "epoch": 0.1251884092333226,
      "grad_norm": 0.2928202450275421,
      "learning_rate": 7.622596375543815e-05,
      "loss": 2.0301,
      "step": 8310
    },
    {
      "epoch": 0.12573035039883482,
      "grad_norm": 0.4254739582538605,
      "learning_rate": 7.614524646370034e-05,
      "loss": 2.0301,
      "step": 8320
    },
    {
      "epoch": 0.12627229156434705,
      "grad_norm": 0.40174198150634766,
      "learning_rate": 7.606444176561989e-05,
      "loss": 2.0225,
      "step": 8330
    },
    {
      "epoch": 0.12681423272985926,
      "grad_norm": 0.32534322142601013,
      "learning_rate": 7.598354999521243e-05,
      "loss": 2.023,
      "step": 8340
    },
    {
      "epoch": 0.1273561738953715,
      "grad_norm": 0.6066051125526428,
      "learning_rate": 7.590257148685352e-05,
      "loss": 2.0251,
      "step": 8350
    },
    {
      "epoch": 0.12746456212847393,
      "eval_loss": 2.6490955352783203,
      "eval_runtime": 21.9681,
      "eval_samples_per_second": 227.602,
      "eval_steps_per_second": 1.229,
      "step": 8352
    },
    {
      "epoch": 0.1278981150608837,
      "grad_norm": 0.6870770454406738,
      "learning_rate": 7.582150657527732e-05,
      "loss": 2.021,
      "step": 8360
    },
    {
      "epoch": 0.12844005622639593,
      "grad_norm": 0.6110761761665344,
      "learning_rate": 7.574035559557506e-05,
      "loss": 2.0273,
      "step": 8370
    },
    {
      "epoch": 0.12898199739190813,
      "grad_norm": 0.32752725481987,
      "learning_rate": 7.565911888319375e-05,
      "loss": 2.0185,
      "step": 8380
    },
    {
      "epoch": 0.12952393855742036,
      "grad_norm": 0.5558075904846191,
      "learning_rate": 7.557779677393486e-05,
      "loss": 2.011,
      "step": 8390
    },
    {
      "epoch": 0.13006587972293257,
      "grad_norm": 0.46736517548561096,
      "learning_rate": 7.549638960395283e-05,
      "loss": 2.0255,
      "step": 8400
    },
    {
      "epoch": 0.1306078208884448,
      "grad_norm": 0.35495907068252563,
      "learning_rate": 7.541489770975365e-05,
      "loss": 2.0289,
      "step": 8410
    },
    {
      "epoch": 0.131149762053957,
      "grad_norm": 0.6544122099876404,
      "learning_rate": 7.533332142819358e-05,
      "loss": 2.0241,
      "step": 8420
    },
    {
      "epoch": 0.13169170321946924,
      "grad_norm": 0.4280942380428314,
      "learning_rate": 7.52516610964777e-05,
      "loss": 2.0245,
      "step": 8430
    },
    {
      "epoch": 0.13217945026843023,
      "eval_loss": 2.6533358097076416,
      "eval_runtime": 21.9742,
      "eval_samples_per_second": 227.54,
      "eval_steps_per_second": 1.229,
      "step": 8439
    },
    {
      "epoch": 0.13223364438498145,
      "grad_norm": 0.4303838610649109,
      "learning_rate": 7.516991705215853e-05,
      "loss": 2.0144,
      "step": 8440
    },
    {
      "epoch": 0.13277558555049368,
      "grad_norm": 0.7431203722953796,
      "learning_rate": 7.508808963313461e-05,
      "loss": 2.0236,
      "step": 8450
    },
    {
      "epoch": 0.13331752671600589,
      "grad_norm": 0.6149893403053284,
      "learning_rate": 7.500617917764908e-05,
      "loss": 2.0145,
      "step": 8460
    },
    {
      "epoch": 0.13385946788151812,
      "grad_norm": 0.3794841468334198,
      "learning_rate": 7.492418602428841e-05,
      "loss": 2.0113,
      "step": 8470
    },
    {
      "epoch": 0.13440140904703032,
      "grad_norm": 0.4046865403652191,
      "learning_rate": 7.484211051198085e-05,
      "loss": 2.0092,
      "step": 8480
    },
    {
      "epoch": 0.13494335021254256,
      "grad_norm": 0.4416908025741577,
      "learning_rate": 7.475995297999507e-05,
      "loss": 2.0194,
      "step": 8490
    },
    {
      "epoch": 0.13548529137805476,
      "grad_norm": 0.4587673246860504,
      "learning_rate": 7.467771376793887e-05,
      "loss": 2.0126,
      "step": 8500
    },
    {
      "epoch": 0.136027232543567,
      "grad_norm": 0.4561171233654022,
      "learning_rate": 7.459539321575758e-05,
      "loss": 2.0158,
      "step": 8510
    },
    {
      "epoch": 0.1365691737090792,
      "grad_norm": 0.4048430323600769,
      "learning_rate": 7.451299166373283e-05,
      "loss": 2.0094,
      "step": 8520
    },
    {
      "epoch": 0.13689433840838655,
      "eval_loss": 2.6495518684387207,
      "eval_runtime": 21.9712,
      "eval_samples_per_second": 227.571,
      "eval_steps_per_second": 1.229,
      "step": 8526
    },
    {
      "epoch": 0.13711111487459143,
      "grad_norm": 0.9034155011177063,
      "learning_rate": 7.443050945248102e-05,
      "loss": 2.0219,
      "step": 8530
    },
    {
      "epoch": 0.13765305604010364,
      "grad_norm": 0.7499250173568726,
      "learning_rate": 7.434794692295202e-05,
      "loss": 2.0136,
      "step": 8540
    },
    {
      "epoch": 0.13819499720561587,
      "grad_norm": 0.6840431094169617,
      "learning_rate": 7.426530441642766e-05,
      "loss": 2.0084,
      "step": 8550
    },
    {
      "epoch": 0.13873693837112808,
      "grad_norm": 0.44849923253059387,
      "learning_rate": 7.418258227452038e-05,
      "loss": 2.0216,
      "step": 8560
    },
    {
      "epoch": 0.1392788795366403,
      "grad_norm": 0.3296073079109192,
      "learning_rate": 7.40997808391718e-05,
      "loss": 2.0037,
      "step": 8570
    },
    {
      "epoch": 0.13982082070215252,
      "grad_norm": 0.34490448236465454,
      "learning_rate": 7.401690045265133e-05,
      "loss": 2.0104,
      "step": 8580
    },
    {
      "epoch": 0.14036276186766475,
      "grad_norm": 0.3491741418838501,
      "learning_rate": 7.39339414575547e-05,
      "loss": 2.0163,
      "step": 8590
    },
    {
      "epoch": 0.14090470303317695,
      "grad_norm": 0.37607693672180176,
      "learning_rate": 7.385090419680259e-05,
      "loss": 2.0061,
      "step": 8600
    },
    {
      "epoch": 0.1414466441986892,
      "grad_norm": 0.6395640969276428,
      "learning_rate": 7.37677890136392e-05,
      "loss": 2.0192,
      "step": 8610
    },
    {
      "epoch": 0.14160922654834285,
      "eval_loss": 2.6450469493865967,
      "eval_runtime": 21.9781,
      "eval_samples_per_second": 227.499,
      "eval_steps_per_second": 1.228,
      "step": 8613
    },
    {
      "epoch": 0.1419885853642014,
      "grad_norm": 0.44370248913764954,
      "learning_rate": 7.368459625163084e-05,
      "loss": 1.9993,
      "step": 8620
    },
    {
      "epoch": 0.14253052652971362,
      "grad_norm": 0.8615736961364746,
      "learning_rate": 7.360132625466452e-05,
      "loss": 1.9997,
      "step": 8630
    },
    {
      "epoch": 0.14307246769522583,
      "grad_norm": 0.45963114500045776,
      "learning_rate": 7.351797936694645e-05,
      "loss": 2.0108,
      "step": 8640
    },
    {
      "epoch": 0.14361440886073806,
      "grad_norm": 0.34752896428108215,
      "learning_rate": 7.34345559330007e-05,
      "loss": 2.0089,
      "step": 8650
    },
    {
      "epoch": 0.14415635002625027,
      "grad_norm": 0.36175432801246643,
      "learning_rate": 7.33510562976678e-05,
      "loss": 2.0071,
      "step": 8660
    },
    {
      "epoch": 0.1446982911917625,
      "grad_norm": 0.32326793670654297,
      "learning_rate": 7.326748080610324e-05,
      "loss": 1.9989,
      "step": 8670
    },
    {
      "epoch": 0.1452402323572747,
      "grad_norm": 1.1309995651245117,
      "learning_rate": 7.318382980377603e-05,
      "loss": 2.0054,
      "step": 8680
    },
    {
      "epoch": 0.14578217352278694,
      "grad_norm": 0.31617847084999084,
      "learning_rate": 7.310010363646736e-05,
      "loss": 2.0057,
      "step": 8690
    },
    {
      "epoch": 0.14632411468829915,
      "grad_norm": 0.34660130739212036,
      "learning_rate": 7.301630265026908e-05,
      "loss": 2.0027,
      "step": 8700
    },
    {
      "epoch": 0.14632411468829915,
      "eval_loss": 2.6555795669555664,
      "eval_runtime": 21.9664,
      "eval_samples_per_second": 227.62,
      "eval_steps_per_second": 1.229,
      "step": 8700
    },
    {
      "epoch": 0.14686605585381138,
      "grad_norm": 0.421053409576416,
      "learning_rate": 7.293242719158241e-05,
      "loss": 2.0116,
      "step": 8710
    },
    {
      "epoch": 0.14740799701932358,
      "grad_norm": 0.5713244080543518,
      "learning_rate": 7.284847760711628e-05,
      "loss": 2.0094,
      "step": 8720
    },
    {
      "epoch": 0.14794993818483582,
      "grad_norm": 0.5165607929229736,
      "learning_rate": 7.27644542438861e-05,
      "loss": 2.0045,
      "step": 8730
    },
    {
      "epoch": 0.14849187935034802,
      "grad_norm": 0.7992691993713379,
      "learning_rate": 7.268035744921225e-05,
      "loss": 1.9992,
      "step": 8740
    },
    {
      "epoch": 0.14903382051586025,
      "grad_norm": 0.42167598009109497,
      "learning_rate": 7.259618757071866e-05,
      "loss": 1.9992,
      "step": 8750
    },
    {
      "epoch": 0.14957576168137246,
      "grad_norm": 0.4957919418811798,
      "learning_rate": 7.251194495633132e-05,
      "loss": 2.0026,
      "step": 8760
    },
    {
      "epoch": 0.1501177028468847,
      "grad_norm": 0.512251079082489,
      "learning_rate": 7.24276299542769e-05,
      "loss": 1.9965,
      "step": 8770
    },
    {
      "epoch": 0.1506596440123969,
      "grad_norm": 0.30542322993278503,
      "learning_rate": 7.234324291308129e-05,
      "loss": 1.9981,
      "step": 8780
    },
    {
      "epoch": 0.15103900282825547,
      "eval_loss": 2.644069194793701,
      "eval_runtime": 21.9682,
      "eval_samples_per_second": 227.602,
      "eval_steps_per_second": 1.229,
      "step": 8787
    },
    {
      "epoch": 0.15120158517790913,
      "grad_norm": 0.7472847104072571,
      "learning_rate": 7.225878418156819e-05,
      "loss": 2.0041,
      "step": 8790
    },
    {
      "epoch": 0.15174352634342134,
      "grad_norm": 0.3889388144016266,
      "learning_rate": 7.217425410885759e-05,
      "loss": 1.9959,
      "step": 8800
    },
    {
      "epoch": 0.15228546750893357,
      "grad_norm": 0.40479448437690735,
      "learning_rate": 7.208965304436444e-05,
      "loss": 2.0035,
      "step": 8810
    },
    {
      "epoch": 0.15282740867444577,
      "grad_norm": 0.5486807227134705,
      "learning_rate": 7.200498133779706e-05,
      "loss": 2.0039,
      "step": 8820
    },
    {
      "epoch": 0.153369349839958,
      "grad_norm": 0.3424724042415619,
      "learning_rate": 7.192023933915586e-05,
      "loss": 2.0085,
      "step": 8830
    },
    {
      "epoch": 0.1539112910054702,
      "grad_norm": 0.3559320867061615,
      "learning_rate": 7.18354273987318e-05,
      "loss": 1.9998,
      "step": 8840
    },
    {
      "epoch": 0.15445323217098245,
      "grad_norm": 0.4217950105667114,
      "learning_rate": 7.175054586710486e-05,
      "loss": 2.0115,
      "step": 8850
    },
    {
      "epoch": 0.15499517333649465,
      "grad_norm": 0.8350193500518799,
      "learning_rate": 7.166559509514283e-05,
      "loss": 2.0065,
      "step": 8860
    },
    {
      "epoch": 0.15553711450200688,
      "grad_norm": 0.43232548236846924,
      "learning_rate": 7.158057543399957e-05,
      "loss": 2.0002,
      "step": 8870
    },
    {
      "epoch": 0.15575389096821177,
      "eval_loss": 2.642789602279663,
      "eval_runtime": 21.9675,
      "eval_samples_per_second": 227.609,
      "eval_steps_per_second": 1.229,
      "step": 8874
    },
    {
      "epoch": 0.1560790556675191,
      "grad_norm": 0.8233237862586975,
      "learning_rate": 7.149548723511377e-05,
      "loss": 2.004,
      "step": 8880
    },
    {
      "epoch": 0.15662099683303132,
      "grad_norm": 0.7288250923156738,
      "learning_rate": 7.141033085020747e-05,
      "loss": 1.997,
      "step": 8890
    },
    {
      "epoch": 0.15716293799854353,
      "grad_norm": 0.6390490531921387,
      "learning_rate": 7.132510663128448e-05,
      "loss": 2.0031,
      "step": 8900
    },
    {
      "epoch": 0.15770487916405576,
      "grad_norm": 0.41019633412361145,
      "learning_rate": 7.123981493062907e-05,
      "loss": 1.9946,
      "step": 8910
    },
    {
      "epoch": 0.15824682032956797,
      "grad_norm": 0.541032075881958,
      "learning_rate": 7.115445610080444e-05,
      "loss": 1.9958,
      "step": 8920
    },
    {
      "epoch": 0.1587887614950802,
      "grad_norm": 0.5177111029624939,
      "learning_rate": 7.106903049465123e-05,
      "loss": 1.9975,
      "step": 8930
    },
    {
      "epoch": 0.1593307026605924,
      "grad_norm": 0.3048165738582611,
      "learning_rate": 7.098353846528619e-05,
      "loss": 1.9908,
      "step": 8940
    },
    {
      "epoch": 0.15987264382610464,
      "grad_norm": 0.33911651372909546,
      "learning_rate": 7.089798036610058e-05,
      "loss": 2.002,
      "step": 8950
    },
    {
      "epoch": 0.16041458499161684,
      "grad_norm": 0.9156370162963867,
      "learning_rate": 7.081235655075878e-05,
      "loss": 1.9849,
      "step": 8960
    },
    {
      "epoch": 0.16046877910816806,
      "eval_loss": 2.637845754623413,
      "eval_runtime": 21.9743,
      "eval_samples_per_second": 227.538,
      "eval_steps_per_second": 1.229,
      "step": 8961
    },
    {
      "epoch": 0.16095652615712908,
      "grad_norm": 0.5576390027999878,
      "learning_rate": 7.072666737319683e-05,
      "loss": 1.9829,
      "step": 8970
    },
    {
      "epoch": 0.16149846732264128,
      "grad_norm": 0.36364230513572693,
      "learning_rate": 7.064091318762089e-05,
      "loss": 1.9983,
      "step": 8980
    },
    {
      "epoch": 0.16204040848815351,
      "grad_norm": 0.7000935077667236,
      "learning_rate": 7.055509434850597e-05,
      "loss": 1.9852,
      "step": 8990
    },
    {
      "epoch": 0.16258234965366572,
      "grad_norm": 0.6612693667411804,
      "learning_rate": 7.046921121059417e-05,
      "loss": 1.9829,
      "step": 9000
    },
    {
      "epoch": 0.16312429081917795,
      "grad_norm": 0.6157468557357788,
      "learning_rate": 7.038326412889353e-05,
      "loss": 1.9901,
      "step": 9010
    },
    {
      "epoch": 0.16366623198469016,
      "grad_norm": 0.33146920800209045,
      "learning_rate": 7.029725345867628e-05,
      "loss": 1.9893,
      "step": 9020
    },
    {
      "epoch": 0.1642081731502024,
      "grad_norm": 0.39117079973220825,
      "learning_rate": 7.02111795554776e-05,
      "loss": 1.9896,
      "step": 9030
    },
    {
      "epoch": 0.1647501143157146,
      "grad_norm": 0.4249345660209656,
      "learning_rate": 7.0125042775094e-05,
      "loss": 1.9823,
      "step": 9040
    },
    {
      "epoch": 0.1651836672481244,
      "eval_loss": 2.6504335403442383,
      "eval_runtime": 21.9717,
      "eval_samples_per_second": 227.566,
      "eval_steps_per_second": 1.229,
      "step": 9048
    },
    {
      "epoch": 0.16529205548122683,
      "grad_norm": 1.197481632232666,
      "learning_rate": 7.003884347358187e-05,
      "loss": 1.9937,
      "step": 9050
    },
    {
      "epoch": 0.16583399664673903,
      "grad_norm": 0.7394921183586121,
      "learning_rate": 6.995258200725613e-05,
      "loss": 1.9839,
      "step": 9060
    },
    {
      "epoch": 0.16637593781225127,
      "grad_norm": 0.6299876570701599,
      "learning_rate": 6.986625873268857e-05,
      "loss": 1.9854,
      "step": 9070
    },
    {
      "epoch": 0.16691787897776347,
      "grad_norm": 0.7766085863113403,
      "learning_rate": 6.977987400670654e-05,
      "loss": 1.9938,
      "step": 9080
    },
    {
      "epoch": 0.1674598201432757,
      "grad_norm": 0.40683385729789734,
      "learning_rate": 6.969342818639138e-05,
      "loss": 1.9844,
      "step": 9090
    },
    {
      "epoch": 0.1680017613087879,
      "grad_norm": 0.33413490653038025,
      "learning_rate": 6.960692162907695e-05,
      "loss": 1.9802,
      "step": 9100
    },
    {
      "epoch": 0.16854370247430014,
      "grad_norm": 0.3267943859100342,
      "learning_rate": 6.952035469234823e-05,
      "loss": 1.9908,
      "step": 9110
    },
    {
      "epoch": 0.16908564363981235,
      "grad_norm": 0.3775721788406372,
      "learning_rate": 6.943372773403972e-05,
      "loss": 1.9816,
      "step": 9120
    },
    {
      "epoch": 0.16962758480532458,
      "grad_norm": 0.42679643630981445,
      "learning_rate": 6.934704111223407e-05,
      "loss": 1.9814,
      "step": 9130
    },
    {
      "epoch": 0.16989855538808069,
      "eval_loss": 2.642576217651367,
      "eval_runtime": 21.9695,
      "eval_samples_per_second": 227.588,
      "eval_steps_per_second": 1.229,
      "step": 9135
    },
    {
      "epoch": 0.1701695259708368,
      "grad_norm": 0.542743444442749,
      "learning_rate": 6.926029518526054e-05,
      "loss": 1.995,
      "step": 9140
    },
    {
      "epoch": 0.17071146713634902,
      "grad_norm": 0.298638254404068,
      "learning_rate": 6.917349031169353e-05,
      "loss": 1.9888,
      "step": 9150
    },
    {
      "epoch": 0.17125340830186123,
      "grad_norm": 0.6379884481430054,
      "learning_rate": 6.908662685035111e-05,
      "loss": 1.9863,
      "step": 9160
    },
    {
      "epoch": 0.17179534946737346,
      "grad_norm": 0.3423231542110443,
      "learning_rate": 6.899970516029355e-05,
      "loss": 1.9858,
      "step": 9170
    },
    {
      "epoch": 0.17233729063288566,
      "grad_norm": 0.41656017303466797,
      "learning_rate": 6.891272560082177e-05,
      "loss": 1.9928,
      "step": 9180
    },
    {
      "epoch": 0.1728792317983979,
      "grad_norm": 0.41668909788131714,
      "learning_rate": 6.882568853147594e-05,
      "loss": 1.9793,
      "step": 9190
    },
    {
      "epoch": 0.1734211729639101,
      "grad_norm": 0.45455989241600037,
      "learning_rate": 6.873859431203393e-05,
      "loss": 1.9811,
      "step": 9200
    },
    {
      "epoch": 0.17396311412942234,
      "grad_norm": 0.421617716550827,
      "learning_rate": 6.865144330250984e-05,
      "loss": 1.9803,
      "step": 9210
    },
    {
      "epoch": 0.17450505529493454,
      "grad_norm": 0.33842185139656067,
      "learning_rate": 6.856423586315258e-05,
      "loss": 1.9708,
      "step": 9220
    },
    {
      "epoch": 0.17461344352803698,
      "eval_loss": 2.636101245880127,
      "eval_runtime": 21.9746,
      "eval_samples_per_second": 227.535,
      "eval_steps_per_second": 1.229,
      "step": 9222
    },
    {
      "epoch": 0.17504699646044677,
      "grad_norm": 0.4086952805519104,
      "learning_rate": 6.847697235444422e-05,
      "loss": 1.9812,
      "step": 9230
    },
    {
      "epoch": 0.17558893762595898,
      "grad_norm": 0.573617696762085,
      "learning_rate": 6.83896531370987e-05,
      "loss": 1.9738,
      "step": 9240
    },
    {
      "epoch": 0.1761308787914712,
      "grad_norm": 0.5594168901443481,
      "learning_rate": 6.830227857206014e-05,
      "loss": 1.9763,
      "step": 9250
    },
    {
      "epoch": 0.17667281995698342,
      "grad_norm": 0.820530116558075,
      "learning_rate": 6.821484902050152e-05,
      "loss": 1.9754,
      "step": 9260
    },
    {
      "epoch": 0.17721476112249565,
      "grad_norm": 0.33057236671447754,
      "learning_rate": 6.81273648438231e-05,
      "loss": 1.9775,
      "step": 9270
    },
    {
      "epoch": 0.17775670228800786,
      "grad_norm": 0.4354369640350342,
      "learning_rate": 6.803982640365092e-05,
      "loss": 1.9864,
      "step": 9280
    },
    {
      "epoch": 0.1782986434535201,
      "grad_norm": 0.3154381811618805,
      "learning_rate": 6.795223406183532e-05,
      "loss": 1.9862,
      "step": 9290
    },
    {
      "epoch": 0.1788405846190323,
      "grad_norm": 0.31054070591926575,
      "learning_rate": 6.78645881804495e-05,
      "loss": 1.9838,
      "step": 9300
    },
    {
      "epoch": 0.1793283316679933,
      "eval_loss": 2.6388654708862305,
      "eval_runtime": 21.9681,
      "eval_samples_per_second": 227.603,
      "eval_steps_per_second": 1.229,
      "step": 9309
    },
    {
      "epoch": 0.17938252578454453,
      "grad_norm": 0.3508462607860565,
      "learning_rate": 6.777688912178787e-05,
      "loss": 1.9744,
      "step": 9310
    },
    {
      "epoch": 0.17992446695005673,
      "grad_norm": 0.34929049015045166,
      "learning_rate": 6.768913724836477e-05,
      "loss": 1.9707,
      "step": 9320
    },
    {
      "epoch": 0.18046640811556897,
      "grad_norm": 0.32265743613243103,
      "learning_rate": 6.760133292291277e-05,
      "loss": 1.9676,
      "step": 9330
    },
    {
      "epoch": 0.18100834928108117,
      "grad_norm": 0.32434341311454773,
      "learning_rate": 6.751347650838134e-05,
      "loss": 1.96,
      "step": 9340
    },
    {
      "epoch": 0.1815502904465934,
      "grad_norm": 0.5159777998924255,
      "learning_rate": 6.742556836793518e-05,
      "loss": 1.9691,
      "step": 9350
    },
    {
      "epoch": 0.1820922316121056,
      "grad_norm": 0.30936720967292786,
      "learning_rate": 6.733760886495284e-05,
      "loss": 1.9696,
      "step": 9360
    },
    {
      "epoch": 0.18263417277761784,
      "grad_norm": 0.3295222222805023,
      "learning_rate": 6.724959836302518e-05,
      "loss": 1.9852,
      "step": 9370
    },
    {
      "epoch": 0.18317611394313005,
      "grad_norm": 0.631144106388092,
      "learning_rate": 6.716153722595392e-05,
      "loss": 1.9825,
      "step": 9380
    },
    {
      "epoch": 0.18371805510864228,
      "grad_norm": 0.4007103741168976,
      "learning_rate": 6.707342581775e-05,
      "loss": 1.9684,
      "step": 9390
    },
    {
      "epoch": 0.1840432198079496,
      "eval_loss": 2.635845422744751,
      "eval_runtime": 21.9618,
      "eval_samples_per_second": 227.668,
      "eval_steps_per_second": 1.229,
      "step": 9396
    },
    {
      "epoch": 0.18425999627415449,
      "grad_norm": 0.7023712396621704,
      "learning_rate": 6.69852645026322e-05,
      "loss": 1.973,
      "step": 9400
    },
    {
      "epoch": 0.18480193743966672,
      "grad_norm": 0.6182842254638672,
      "learning_rate": 6.689705364502562e-05,
      "loss": 1.9714,
      "step": 9410
    },
    {
      "epoch": 0.18534387860517892,
      "grad_norm": 0.5204935669898987,
      "learning_rate": 6.680879360956012e-05,
      "loss": 1.9647,
      "step": 9420
    },
    {
      "epoch": 0.18588581977069116,
      "grad_norm": 0.3946668803691864,
      "learning_rate": 6.672048476106886e-05,
      "loss": 1.9776,
      "step": 9430
    },
    {
      "epoch": 0.18642776093620336,
      "grad_norm": 0.44481825828552246,
      "learning_rate": 6.663212746458676e-05,
      "loss": 1.9718,
      "step": 9440
    },
    {
      "epoch": 0.1869697021017156,
      "grad_norm": 0.4806019365787506,
      "learning_rate": 6.6543722085349e-05,
      "loss": 1.9699,
      "step": 9450
    },
    {
      "epoch": 0.1875116432672278,
      "grad_norm": 0.3834541141986847,
      "learning_rate": 6.645526898878955e-05,
      "loss": 1.9617,
      "step": 9460
    },
    {
      "epoch": 0.18805358443274003,
      "grad_norm": 0.587719738483429,
      "learning_rate": 6.636676854053958e-05,
      "loss": 1.9663,
      "step": 9470
    },
    {
      "epoch": 0.18859552559825224,
      "grad_norm": 0.48372459411621094,
      "learning_rate": 6.627822110642603e-05,
      "loss": 1.9715,
      "step": 9480
    },
    {
      "epoch": 0.1887581079479059,
      "eval_loss": 2.629991292953491,
      "eval_runtime": 21.9683,
      "eval_samples_per_second": 227.601,
      "eval_steps_per_second": 1.229,
      "step": 9483
    },
    {
      "epoch": 0.18913746676376447,
      "grad_norm": 0.368070513010025,
      "learning_rate": 6.618962705247003e-05,
      "loss": 1.9678,
      "step": 9490
    },
    {
      "epoch": 0.18967940792927668,
      "grad_norm": 0.3025422692298889,
      "learning_rate": 6.610098674488546e-05,
      "loss": 1.9647,
      "step": 9500
    },
    {
      "epoch": 0.1902213490947889,
      "grad_norm": 0.4809657037258148,
      "learning_rate": 6.601230055007734e-05,
      "loss": 1.9742,
      "step": 9510
    },
    {
      "epoch": 0.19076329026030112,
      "grad_norm": 0.3071550726890564,
      "learning_rate": 6.592356883464043e-05,
      "loss": 1.9588,
      "step": 9520
    },
    {
      "epoch": 0.19130523142581332,
      "grad_norm": 0.3039422035217285,
      "learning_rate": 6.583479196535763e-05,
      "loss": 1.961,
      "step": 9530
    },
    {
      "epoch": 0.19184717259132555,
      "grad_norm": 1.0146269798278809,
      "learning_rate": 6.574597030919844e-05,
      "loss": 1.9722,
      "step": 9540
    },
    {
      "epoch": 0.19238911375683776,
      "grad_norm": 0.3183740973472595,
      "learning_rate": 6.565710423331757e-05,
      "loss": 1.9659,
      "step": 9550
    },
    {
      "epoch": 0.19293105492235,
      "grad_norm": 0.373046338558197,
      "learning_rate": 6.556819410505331e-05,
      "loss": 1.9615,
      "step": 9560
    },
    {
      "epoch": 0.1934729960878622,
      "grad_norm": 0.36389511823654175,
      "learning_rate": 6.547924029192603e-05,
      "loss": 1.9639,
      "step": 9570
    },
    {
      "epoch": 0.1934729960878622,
      "eval_loss": 2.6214120388031006,
      "eval_runtime": 21.9741,
      "eval_samples_per_second": 227.54,
      "eval_steps_per_second": 1.229,
      "step": 9570
    },
    {
      "epoch": 0.19401493725337443,
      "grad_norm": 0.3653751015663147,
      "learning_rate": 6.539024316163671e-05,
      "loss": 1.9665,
      "step": 9580
    },
    {
      "epoch": 0.19455687841888664,
      "grad_norm": 0.4899474084377289,
      "learning_rate": 6.530120308206536e-05,
      "loss": 1.968,
      "step": 9590
    },
    {
      "epoch": 0.19509881958439887,
      "grad_norm": 0.3284156322479248,
      "learning_rate": 6.521212042126951e-05,
      "loss": 1.9563,
      "step": 9600
    },
    {
      "epoch": 0.19564076074991107,
      "grad_norm": 0.6163231134414673,
      "learning_rate": 6.512299554748281e-05,
      "loss": 1.9764,
      "step": 9610
    },
    {
      "epoch": 0.1961827019154233,
      "grad_norm": 0.5178772807121277,
      "learning_rate": 6.503382882911322e-05,
      "loss": 1.97,
      "step": 9620
    },
    {
      "epoch": 0.1967246430809355,
      "grad_norm": 0.33963051438331604,
      "learning_rate": 6.494462063474181e-05,
      "loss": 1.9612,
      "step": 9630
    },
    {
      "epoch": 0.19726658424644775,
      "grad_norm": 0.2811043858528137,
      "learning_rate": 6.485537133312107e-05,
      "loss": 1.9682,
      "step": 9640
    },
    {
      "epoch": 0.19780852541195995,
      "grad_norm": 0.3501632511615753,
      "learning_rate": 6.476608129317342e-05,
      "loss": 1.9547,
      "step": 9650
    },
    {
      "epoch": 0.19818788422781852,
      "eval_loss": 2.6299521923065186,
      "eval_runtime": 21.9699,
      "eval_samples_per_second": 227.584,
      "eval_steps_per_second": 1.229,
      "step": 9657
    },
    {
      "epoch": 0.19835046657747218,
      "grad_norm": 0.3678341209888458,
      "learning_rate": 6.46767508839896e-05,
      "loss": 1.9613,
      "step": 9660
    },
    {
      "epoch": 0.1988924077429844,
      "grad_norm": 0.62614905834198,
      "learning_rate": 6.458738047482731e-05,
      "loss": 1.9626,
      "step": 9670
    },
    {
      "epoch": 0.19943434890849662,
      "grad_norm": 0.7425987124443054,
      "learning_rate": 6.449797043510954e-05,
      "loss": 1.9591,
      "step": 9680
    },
    {
      "epoch": 0.19997629007400883,
      "grad_norm": 0.3964003324508667,
      "learning_rate": 6.440852113442314e-05,
      "loss": 1.9638,
      "step": 9690
    },
    {
      "epoch": 0.20051823123952106,
      "grad_norm": 0.4931100308895111,
      "learning_rate": 6.431903294251721e-05,
      "loss": 1.9487,
      "step": 9700
    },
    {
      "epoch": 0.20106017240503327,
      "grad_norm": 0.31457212567329407,
      "learning_rate": 6.422950622930164e-05,
      "loss": 1.9618,
      "step": 9710
    },
    {
      "epoch": 0.2016021135705455,
      "grad_norm": 0.44526612758636475,
      "learning_rate": 6.413994136484553e-05,
      "loss": 1.9584,
      "step": 9720
    },
    {
      "epoch": 0.2021440547360577,
      "grad_norm": 0.3651030957698822,
      "learning_rate": 6.405033871937572e-05,
      "loss": 1.9614,
      "step": 9730
    },
    {
      "epoch": 0.20268599590156994,
      "grad_norm": 0.639445960521698,
      "learning_rate": 6.396069866327519e-05,
      "loss": 1.964,
      "step": 9740
    },
    {
      "epoch": 0.20290277236777482,
      "eval_loss": 2.6391215324401855,
      "eval_runtime": 21.9615,
      "eval_samples_per_second": 227.671,
      "eval_steps_per_second": 1.229,
      "step": 9744
    },
    {
      "epoch": 0.20322793706708214,
      "grad_norm": 0.45781099796295166,
      "learning_rate": 6.387102156708152e-05,
      "loss": 1.9654,
      "step": 9750
    },
    {
      "epoch": 0.20376987823259438,
      "grad_norm": 0.3018638789653778,
      "learning_rate": 6.37813078014855e-05,
      "loss": 1.9603,
      "step": 9760
    },
    {
      "epoch": 0.20431181939810658,
      "grad_norm": 0.2993408739566803,
      "learning_rate": 6.369155773732945e-05,
      "loss": 1.9519,
      "step": 9770
    },
    {
      "epoch": 0.20485376056361881,
      "grad_norm": 0.35188788175582886,
      "learning_rate": 6.360177174560567e-05,
      "loss": 1.945,
      "step": 9780
    },
    {
      "epoch": 0.20539570172913102,
      "grad_norm": 0.5056437849998474,
      "learning_rate": 6.351195019745508e-05,
      "loss": 1.9569,
      "step": 9790
    },
    {
      "epoch": 0.20593764289464325,
      "grad_norm": 0.4438895583152771,
      "learning_rate": 6.342209346416553e-05,
      "loss": 1.9503,
      "step": 9800
    },
    {
      "epoch": 0.20647958406015546,
      "grad_norm": 0.8153665661811829,
      "learning_rate": 6.333220191717026e-05,
      "loss": 1.9529,
      "step": 9810
    },
    {
      "epoch": 0.2070215252256677,
      "grad_norm": 0.47668522596359253,
      "learning_rate": 6.324227592804651e-05,
      "loss": 1.9515,
      "step": 9820
    },
    {
      "epoch": 0.2075634663911799,
      "grad_norm": 0.3650936782360077,
      "learning_rate": 6.315231586851382e-05,
      "loss": 1.9473,
      "step": 9830
    },
    {
      "epoch": 0.20761766050773112,
      "eval_loss": 2.6374318599700928,
      "eval_runtime": 21.9702,
      "eval_samples_per_second": 227.581,
      "eval_steps_per_second": 1.229,
      "step": 9831
    },
    {
      "epoch": 0.20810540755669213,
      "grad_norm": 0.4178442060947418,
      "learning_rate": 6.306232211043262e-05,
      "loss": 1.9593,
      "step": 9840
    },
    {
      "epoch": 0.20864734872220433,
      "grad_norm": 0.39025014638900757,
      "learning_rate": 6.297229502580257e-05,
      "loss": 1.957,
      "step": 9850
    },
    {
      "epoch": 0.20918928988771657,
      "grad_norm": 0.573856770992279,
      "learning_rate": 6.288223498676114e-05,
      "loss": 1.9455,
      "step": 9860
    },
    {
      "epoch": 0.20973123105322877,
      "grad_norm": 0.32634246349334717,
      "learning_rate": 6.279214236558201e-05,
      "loss": 1.9464,
      "step": 9870
    },
    {
      "epoch": 0.210273172218741,
      "grad_norm": 0.48771846294403076,
      "learning_rate": 6.270201753467351e-05,
      "loss": 1.9471,
      "step": 9880
    },
    {
      "epoch": 0.2108151133842532,
      "grad_norm": 0.3284076154232025,
      "learning_rate": 6.261186086657722e-05,
      "loss": 1.9512,
      "step": 9890
    },
    {
      "epoch": 0.21135705454976544,
      "grad_norm": 0.4014468789100647,
      "learning_rate": 6.252167273396614e-05,
      "loss": 1.9596,
      "step": 9900
    },
    {
      "epoch": 0.21189899571527765,
      "grad_norm": 0.2978026568889618,
      "learning_rate": 6.24314535096435e-05,
      "loss": 1.95,
      "step": 9910
    },
    {
      "epoch": 0.21233254864768744,
      "eval_loss": 2.6269800662994385,
      "eval_runtime": 21.9722,
      "eval_samples_per_second": 227.561,
      "eval_steps_per_second": 1.229,
      "step": 9918
    },
    {
      "epoch": 0.21244093688078988,
      "grad_norm": 0.36181050539016724,
      "learning_rate": 6.234120356654096e-05,
      "loss": 1.9543,
      "step": 9920
    },
    {
      "epoch": 0.2129828780463021,
      "grad_norm": 0.7839191555976868,
      "learning_rate": 6.225092327771723e-05,
      "loss": 1.9515,
      "step": 9930
    },
    {
      "epoch": 0.21352481921181432,
      "grad_norm": 1.2086142301559448,
      "learning_rate": 6.216061301635633e-05,
      "loss": 1.9502,
      "step": 9940
    },
    {
      "epoch": 0.21406676037732653,
      "grad_norm": 0.30762261152267456,
      "learning_rate": 6.207027315576635e-05,
      "loss": 1.9565,
      "step": 9950
    },
    {
      "epoch": 0.21460870154283876,
      "grad_norm": 0.31394830346107483,
      "learning_rate": 6.197990406937757e-05,
      "loss": 1.9466,
      "step": 9960
    },
    {
      "epoch": 0.21515064270835096,
      "grad_norm": 0.3644161820411682,
      "learning_rate": 6.188950613074122e-05,
      "loss": 1.954,
      "step": 9970
    },
    {
      "epoch": 0.2156925838738632,
      "grad_norm": 0.34890973567962646,
      "learning_rate": 6.179907971352766e-05,
      "loss": 1.936,
      "step": 9980
    },
    {
      "epoch": 0.2162345250393754,
      "grad_norm": 0.4451602101325989,
      "learning_rate": 6.170862519152505e-05,
      "loss": 1.9447,
      "step": 9990
    },
    {
      "epoch": 0.21677646620488764,
      "grad_norm": 0.33436113595962524,
      "learning_rate": 6.161814293863772e-05,
      "loss": 1.9497,
      "step": 10000
    },
    {
      "epoch": 0.21704743678764374,
      "eval_loss": 2.6255741119384766,
      "eval_runtime": 21.9675,
      "eval_samples_per_second": 227.609,
      "eval_steps_per_second": 1.229,
      "step": 10005
    },
    {
      "epoch": 0.21731840737039984,
      "grad_norm": 0.6699984669685364,
      "learning_rate": 6.152763332888464e-05,
      "loss": 1.9456,
      "step": 10010
    },
    {
      "epoch": 0.21786034853591207,
      "grad_norm": 0.3605382740497589,
      "learning_rate": 6.143709673639778e-05,
      "loss": 1.9341,
      "step": 10020
    },
    {
      "epoch": 0.21840228970142428,
      "grad_norm": 0.35184648633003235,
      "learning_rate": 6.134653353542074e-05,
      "loss": 1.9538,
      "step": 10030
    },
    {
      "epoch": 0.2189442308669365,
      "grad_norm": 0.3720100522041321,
      "learning_rate": 6.125594410030706e-05,
      "loss": 1.9441,
      "step": 10040
    },
    {
      "epoch": 0.21948617203244872,
      "grad_norm": 0.5039480924606323,
      "learning_rate": 6.116532880551876e-05,
      "loss": 1.9407,
      "step": 10050
    },
    {
      "epoch": 0.22002811319796095,
      "grad_norm": 0.3840113878250122,
      "learning_rate": 6.107468802562472e-05,
      "loss": 1.9444,
      "step": 10060
    },
    {
      "epoch": 0.22057005436347316,
      "grad_norm": 0.5373072028160095,
      "learning_rate": 6.098402213529916e-05,
      "loss": 1.948,
      "step": 10070
    },
    {
      "epoch": 0.2211119955289854,
      "grad_norm": 0.2886420786380768,
      "learning_rate": 6.089333150932014e-05,
      "loss": 1.9425,
      "step": 10080
    },
    {
      "epoch": 0.2216539366944976,
      "grad_norm": 0.2997380495071411,
      "learning_rate": 6.0802616522567914e-05,
      "loss": 1.9477,
      "step": 10090
    },
    {
      "epoch": 0.22176232492760004,
      "eval_loss": 2.6249098777770996,
      "eval_runtime": 21.9708,
      "eval_samples_per_second": 227.574,
      "eval_steps_per_second": 1.229,
      "step": 10092
    },
    {
      "epoch": 0.22219587786000983,
      "grad_norm": 0.3256387412548065,
      "learning_rate": 6.0711877550023474e-05,
      "loss": 1.9371,
      "step": 10100
    },
    {
      "epoch": 0.22273781902552203,
      "grad_norm": 0.37883371114730835,
      "learning_rate": 6.062111496676694e-05,
      "loss": 1.9464,
      "step": 10110
    },
    {
      "epoch": 0.22327976019103427,
      "grad_norm": 0.42293113470077515,
      "learning_rate": 6.053032914797605e-05,
      "loss": 1.9389,
      "step": 10120
    },
    {
      "epoch": 0.22382170135654647,
      "grad_norm": 0.9902483820915222,
      "learning_rate": 6.043952046892457e-05,
      "loss": 1.9459,
      "step": 10130
    },
    {
      "epoch": 0.2243636425220587,
      "grad_norm": 0.28801223635673523,
      "learning_rate": 6.034868930498076e-05,
      "loss": 1.9296,
      "step": 10140
    },
    {
      "epoch": 0.2249055836875709,
      "grad_norm": 0.4709838926792145,
      "learning_rate": 6.025783603160583e-05,
      "loss": 1.9471,
      "step": 10150
    },
    {
      "epoch": 0.22544752485308314,
      "grad_norm": 0.5633691549301147,
      "learning_rate": 6.016696102435241e-05,
      "loss": 1.9371,
      "step": 10160
    },
    {
      "epoch": 0.22598946601859535,
      "grad_norm": 0.6367583870887756,
      "learning_rate": 6.0076064658862884e-05,
      "loss": 1.9431,
      "step": 10170
    },
    {
      "epoch": 0.22647721306755636,
      "eval_loss": 2.6244375705718994,
      "eval_runtime": 21.9646,
      "eval_samples_per_second": 227.639,
      "eval_steps_per_second": 1.229,
      "step": 10179
    },
    {
      "epoch": 0.22653140718410758,
      "grad_norm": 0.6641575694084167,
      "learning_rate": 5.998514731086805e-05,
      "loss": 1.9408,
      "step": 10180
    },
    {
      "epoch": 0.22707334834961979,
      "grad_norm": 0.45785051584243774,
      "learning_rate": 5.9894209356185314e-05,
      "loss": 1.9407,
      "step": 10190
    },
    {
      "epoch": 0.22761528951513202,
      "grad_norm": 0.473785936832428,
      "learning_rate": 5.980325117071736e-05,
      "loss": 1.9439,
      "step": 10200
    },
    {
      "epoch": 0.22815723068064422,
      "grad_norm": 0.6695595979690552,
      "learning_rate": 5.971227313045043e-05,
      "loss": 1.945,
      "step": 10210
    },
    {
      "epoch": 0.22869917184615646,
      "grad_norm": 0.502714991569519,
      "learning_rate": 5.9621275611452874e-05,
      "loss": 1.9456,
      "step": 10220
    },
    {
      "epoch": 0.22924111301166866,
      "grad_norm": 0.6983989477157593,
      "learning_rate": 5.9530258989873555e-05,
      "loss": 1.9517,
      "step": 10230
    },
    {
      "epoch": 0.2297830541771809,
      "grad_norm": 0.3308335840702057,
      "learning_rate": 5.943922364194029e-05,
      "loss": 1.9387,
      "step": 10240
    },
    {
      "epoch": 0.2303249953426931,
      "grad_norm": 0.2856954038143158,
      "learning_rate": 5.93481699439583e-05,
      "loss": 1.9409,
      "step": 10250
    },
    {
      "epoch": 0.23086693650820533,
      "grad_norm": 0.3123374879360199,
      "learning_rate": 5.925709827230868e-05,
      "loss": 1.933,
      "step": 10260
    },
    {
      "epoch": 0.23119210120751266,
      "eval_loss": 2.6269350051879883,
      "eval_runtime": 21.9652,
      "eval_samples_per_second": 227.633,
      "eval_steps_per_second": 1.229,
      "step": 10266
    },
    {
      "epoch": 0.23140887767371754,
      "grad_norm": 0.3088303506374359,
      "learning_rate": 5.91660090034468e-05,
      "loss": 1.9344,
      "step": 10270
    },
    {
      "epoch": 0.23195081883922977,
      "grad_norm": 0.3397095501422882,
      "learning_rate": 5.907490251390079e-05,
      "loss": 1.9421,
      "step": 10280
    },
    {
      "epoch": 0.23249276000474198,
      "grad_norm": 0.40521690249443054,
      "learning_rate": 5.898377918026993e-05,
      "loss": 1.9369,
      "step": 10290
    },
    {
      "epoch": 0.2330347011702542,
      "grad_norm": 0.3461083769798279,
      "learning_rate": 5.889263937922315e-05,
      "loss": 1.937,
      "step": 10300
    },
    {
      "epoch": 0.23357664233576642,
      "grad_norm": 0.47172901034355164,
      "learning_rate": 5.8801483487497476e-05,
      "loss": 1.9437,
      "step": 10310
    },
    {
      "epoch": 0.23411858350127865,
      "grad_norm": 0.4379764795303345,
      "learning_rate": 5.87103118818964e-05,
      "loss": 1.92,
      "step": 10320
    },
    {
      "epoch": 0.23466052466679085,
      "grad_norm": 0.7908567190170288,
      "learning_rate": 5.861912493928837e-05,
      "loss": 1.9329,
      "step": 10330
    },
    {
      "epoch": 0.2352024658323031,
      "grad_norm": 0.5137506723403931,
      "learning_rate": 5.852792303660528e-05,
      "loss": 1.9414,
      "step": 10340
    },
    {
      "epoch": 0.2357444069978153,
      "grad_norm": 0.31868940591812134,
      "learning_rate": 5.8436706550840805e-05,
      "loss": 1.9387,
      "step": 10350
    },
    {
      "epoch": 0.23590698934746895,
      "eval_loss": 2.6262524127960205,
      "eval_runtime": 21.9674,
      "eval_samples_per_second": 227.61,
      "eval_steps_per_second": 1.229,
      "step": 10353
    },
    {
      "epoch": 0.23628634816332753,
      "grad_norm": 0.3951703608036041,
      "learning_rate": 5.834547585904898e-05,
      "loss": 1.9313,
      "step": 10360
    },
    {
      "epoch": 0.23682828932883973,
      "grad_norm": 0.3555956780910492,
      "learning_rate": 5.8254231338342446e-05,
      "loss": 1.9298,
      "step": 10370
    },
    {
      "epoch": 0.23737023049435196,
      "grad_norm": 0.3569830358028412,
      "learning_rate": 5.8162973365891106e-05,
      "loss": 1.9423,
      "step": 10380
    },
    {
      "epoch": 0.23791217165986417,
      "grad_norm": 0.34016862511634827,
      "learning_rate": 5.807170231892042e-05,
      "loss": 1.9239,
      "step": 10390
    },
    {
      "epoch": 0.2384541128253764,
      "grad_norm": 0.32383885979652405,
      "learning_rate": 5.7980418574709924e-05,
      "loss": 1.9378,
      "step": 10400
    },
    {
      "epoch": 0.2389960539908886,
      "grad_norm": 0.42014604806900024,
      "learning_rate": 5.788912251059162e-05,
      "loss": 1.9331,
      "step": 10410
    },
    {
      "epoch": 0.23953799515640084,
      "grad_norm": 0.3834390640258789,
      "learning_rate": 5.7797814503948414e-05,
      "loss": 1.9344,
      "step": 10420
    },
    {
      "epoch": 0.24007993632191305,
      "grad_norm": 0.33072102069854736,
      "learning_rate": 5.770649493221262e-05,
      "loss": 1.9281,
      "step": 10430
    },
    {
      "epoch": 0.24062187748742528,
      "grad_norm": 0.41425782442092896,
      "learning_rate": 5.7615164172864346e-05,
      "loss": 1.9247,
      "step": 10440
    },
    {
      "epoch": 0.24062187748742528,
      "eval_loss": 2.6209261417388916,
      "eval_runtime": 21.9653,
      "eval_samples_per_second": 227.632,
      "eval_steps_per_second": 1.229,
      "step": 10440
    },
    {
      "epoch": 0.24116381865293748,
      "grad_norm": 0.4126664400100708,
      "learning_rate": 5.7523822603429924e-05,
      "loss": 1.9254,
      "step": 10450
    },
    {
      "epoch": 0.24170575981844972,
      "grad_norm": 0.4077990651130676,
      "learning_rate": 5.7432470601480394e-05,
      "loss": 1.9298,
      "step": 10460
    },
    {
      "epoch": 0.24224770098396192,
      "grad_norm": 0.33790072798728943,
      "learning_rate": 5.7341108544629894e-05,
      "loss": 1.9214,
      "step": 10470
    },
    {
      "epoch": 0.24278964214947416,
      "grad_norm": 0.5326333045959473,
      "learning_rate": 5.724973681053417e-05,
      "loss": 1.9277,
      "step": 10480
    },
    {
      "epoch": 0.24333158331498636,
      "grad_norm": 0.8307605981826782,
      "learning_rate": 5.715835577688894e-05,
      "loss": 1.9399,
      "step": 10490
    },
    {
      "epoch": 0.2438735244804986,
      "grad_norm": 0.29457995295524597,
      "learning_rate": 5.706696582142834e-05,
      "loss": 1.9274,
      "step": 10500
    },
    {
      "epoch": 0.2444154656460108,
      "grad_norm": 0.47820374369621277,
      "learning_rate": 5.697556732192343e-05,
      "loss": 1.9332,
      "step": 10510
    },
    {
      "epoch": 0.24495740681152303,
      "grad_norm": 0.615271270275116,
      "learning_rate": 5.688416065618057e-05,
      "loss": 1.9358,
      "step": 10520
    },
    {
      "epoch": 0.24533676562738158,
      "eval_loss": 2.623511552810669,
      "eval_runtime": 21.9658,
      "eval_samples_per_second": 227.627,
      "eval_steps_per_second": 1.229,
      "step": 10527
    },
    {
      "epoch": 0.24549934797703524,
      "grad_norm": 0.3703593909740448,
      "learning_rate": 5.679274620203986e-05,
      "loss": 1.9345,
      "step": 10530
    },
    {
      "epoch": 0.24604128914254747,
      "grad_norm": 0.3069552779197693,
      "learning_rate": 5.670132433737363e-05,
      "loss": 1.9229,
      "step": 10540
    },
    {
      "epoch": 0.24658323030805968,
      "grad_norm": 0.30190664529800415,
      "learning_rate": 5.660989544008479e-05,
      "loss": 1.9288,
      "step": 10550
    },
    {
      "epoch": 0.2471251714735719,
      "grad_norm": 0.5308563113212585,
      "learning_rate": 5.651845988810538e-05,
      "loss": 1.9246,
      "step": 10560
    },
    {
      "epoch": 0.2476671126390841,
      "grad_norm": 0.33339574933052063,
      "learning_rate": 5.642701805939491e-05,
      "loss": 1.9239,
      "step": 10570
    },
    {
      "epoch": 0.24820905380459635,
      "grad_norm": 0.6166539788246155,
      "learning_rate": 5.633557033193884e-05,
      "loss": 1.9311,
      "step": 10580
    },
    {
      "epoch": 0.24875099497010855,
      "grad_norm": 0.3371501863002777,
      "learning_rate": 5.624411708374703e-05,
      "loss": 1.9148,
      "step": 10590
    },
    {
      "epoch": 0.24929293613562079,
      "grad_norm": 0.5741251111030579,
      "learning_rate": 5.615265869285212e-05,
      "loss": 1.936,
      "step": 10600
    },
    {
      "epoch": 0.249834877301133,
      "grad_norm": 0.8390110731124878,
      "learning_rate": 5.606119553730808e-05,
      "loss": 1.9235,
      "step": 10610
    },
    {
      "epoch": 0.2500516537673379,
      "eval_loss": 2.6314127445220947,
      "eval_runtime": 21.9728,
      "eval_samples_per_second": 227.554,
      "eval_steps_per_second": 1.229,
      "step": 10614
    },
    {
      "epoch": 0.2503768184666452,
      "grad_norm": 0.31572863459587097,
      "learning_rate": 5.596972799518849e-05,
      "loss": 1.9226,
      "step": 10620
    },
    {
      "epoch": 0.25091875963215743,
      "grad_norm": 0.3893384635448456,
      "learning_rate": 5.5878256444585134e-05,
      "loss": 1.9379,
      "step": 10630
    },
    {
      "epoch": 0.25146070079766963,
      "grad_norm": 0.4373924136161804,
      "learning_rate": 5.578678126360632e-05,
      "loss": 1.9296,
      "step": 10640
    },
    {
      "epoch": 0.2520026419631819,
      "grad_norm": 0.2836342751979828,
      "learning_rate": 5.5695302830375374e-05,
      "loss": 1.9231,
      "step": 10650
    },
    {
      "epoch": 0.2525445831286941,
      "grad_norm": 0.3092157244682312,
      "learning_rate": 5.5603821523029084e-05,
      "loss": 1.9191,
      "step": 10660
    },
    {
      "epoch": 0.2530865242942063,
      "grad_norm": 0.6483246088027954,
      "learning_rate": 5.551233771971611e-05,
      "loss": 1.9202,
      "step": 10670
    },
    {
      "epoch": 0.2536284654597185,
      "grad_norm": 0.34343624114990234,
      "learning_rate": 5.54208517985954e-05,
      "loss": 1.9297,
      "step": 10680
    },
    {
      "epoch": 0.25417040662523077,
      "grad_norm": 0.29993903636932373,
      "learning_rate": 5.532936413783469e-05,
      "loss": 1.9228,
      "step": 10690
    },
    {
      "epoch": 0.254712347790743,
      "grad_norm": 0.4231652319431305,
      "learning_rate": 5.5237875115608905e-05,
      "loss": 1.94,
      "step": 10700
    },
    {
      "epoch": 0.2547665419072942,
      "eval_loss": 2.6222281455993652,
      "eval_runtime": 21.972,
      "eval_samples_per_second": 227.563,
      "eval_steps_per_second": 1.229,
      "step": 10701
    },
    {
      "epoch": 0.2552542889562552,
      "grad_norm": 0.31495311856269836,
      "learning_rate": 5.51463851100986e-05,
      "loss": 1.9252,
      "step": 10710
    },
    {
      "epoch": 0.2557962301217674,
      "grad_norm": 0.5191118717193604,
      "learning_rate": 5.5054894499488353e-05,
      "loss": 1.9118,
      "step": 10720
    },
    {
      "epoch": 0.25633817128727965,
      "grad_norm": 0.6403262615203857,
      "learning_rate": 5.496340366196527e-05,
      "loss": 1.9308,
      "step": 10730
    },
    {
      "epoch": 0.25688011245279185,
      "grad_norm": 0.29355281591415405,
      "learning_rate": 5.4871912975717444e-05,
      "loss": 1.925,
      "step": 10740
    },
    {
      "epoch": 0.25742205361830406,
      "grad_norm": 0.4204306900501251,
      "learning_rate": 5.478042281893225e-05,
      "loss": 1.92,
      "step": 10750
    },
    {
      "epoch": 0.25796399478381626,
      "grad_norm": 0.5830504894256592,
      "learning_rate": 5.468893356979498e-05,
      "loss": 1.9133,
      "step": 10760
    },
    {
      "epoch": 0.2585059359493285,
      "grad_norm": 0.2985590994358063,
      "learning_rate": 5.459744560648707e-05,
      "loss": 1.9221,
      "step": 10770
    },
    {
      "epoch": 0.25904787711484073,
      "grad_norm": 0.29512789845466614,
      "learning_rate": 5.4505959307184675e-05,
      "loss": 1.9203,
      "step": 10780
    },
    {
      "epoch": 0.2594814300472505,
      "eval_loss": 2.625079870223999,
      "eval_runtime": 21.9676,
      "eval_samples_per_second": 227.608,
      "eval_steps_per_second": 1.229,
      "step": 10788
    },
    {
      "epoch": 0.25958981828035294,
      "grad_norm": 0.3882138133049011,
      "learning_rate": 5.441447505005714e-05,
      "loss": 1.9178,
      "step": 10790
    },
    {
      "epoch": 0.26013175944586514,
      "grad_norm": 0.5985729098320007,
      "learning_rate": 5.432299321326526e-05,
      "loss": 1.9214,
      "step": 10800
    },
    {
      "epoch": 0.2606737006113774,
      "grad_norm": 0.5359235405921936,
      "learning_rate": 5.423151417495991e-05,
      "loss": 1.9182,
      "step": 10810
    },
    {
      "epoch": 0.2612156417768896,
      "grad_norm": 0.2866165041923523,
      "learning_rate": 5.4140038313280364e-05,
      "loss": 1.9205,
      "step": 10820
    },
    {
      "epoch": 0.2617575829424018,
      "grad_norm": 0.7025465965270996,
      "learning_rate": 5.404856600635273e-05,
      "loss": 1.9221,
      "step": 10830
    },
    {
      "epoch": 0.262299524107914,
      "grad_norm": 0.48262158036231995,
      "learning_rate": 5.39570976322885e-05,
      "loss": 1.9164,
      "step": 10840
    },
    {
      "epoch": 0.2628414652734263,
      "grad_norm": 0.331333190202713,
      "learning_rate": 5.386563356918286e-05,
      "loss": 1.9203,
      "step": 10850
    },
    {
      "epoch": 0.2633834064389385,
      "grad_norm": 0.559256374835968,
      "learning_rate": 5.3774174195113145e-05,
      "loss": 1.919,
      "step": 10860
    },
    {
      "epoch": 0.2639253476044507,
      "grad_norm": 0.4558927118778229,
      "learning_rate": 5.368271988813741e-05,
      "loss": 1.9186,
      "step": 10870
    },
    {
      "epoch": 0.2641963181872068,
      "eval_loss": 2.6204073429107666,
      "eval_runtime": 21.9646,
      "eval_samples_per_second": 227.639,
      "eval_steps_per_second": 1.229,
      "step": 10875
    },
    {
      "epoch": 0.2644672887699629,
      "grad_norm": 0.32578545808792114,
      "learning_rate": 5.3591271026292645e-05,
      "loss": 1.9178,
      "step": 10880
    },
    {
      "epoch": 0.26500922993547515,
      "grad_norm": 0.5844838619232178,
      "learning_rate": 5.349982798759341e-05,
      "loss": 1.9306,
      "step": 10890
    },
    {
      "epoch": 0.26555117110098736,
      "grad_norm": 0.37786585092544556,
      "learning_rate": 5.340839115003019e-05,
      "loss": 1.9236,
      "step": 10900
    },
    {
      "epoch": 0.26609311226649957,
      "grad_norm": 0.33021968603134155,
      "learning_rate": 5.331696089156776e-05,
      "loss": 1.9097,
      "step": 10910
    },
    {
      "epoch": 0.26663505343201177,
      "grad_norm": 0.3873339891433716,
      "learning_rate": 5.322553759014383e-05,
      "loss": 1.9097,
      "step": 10920
    },
    {
      "epoch": 0.26717699459752403,
      "grad_norm": 0.32818225026130676,
      "learning_rate": 5.313412162366723e-05,
      "loss": 1.9136,
      "step": 10930
    },
    {
      "epoch": 0.26771893576303624,
      "grad_norm": 0.6563804149627686,
      "learning_rate": 5.304271337001652e-05,
      "loss": 1.9173,
      "step": 10940
    },
    {
      "epoch": 0.26826087692854844,
      "grad_norm": 0.3008669912815094,
      "learning_rate": 5.295131320703841e-05,
      "loss": 1.9104,
      "step": 10950
    },
    {
      "epoch": 0.26880281809406065,
      "grad_norm": 0.3501867353916168,
      "learning_rate": 5.2859921512546104e-05,
      "loss": 1.9232,
      "step": 10960
    },
    {
      "epoch": 0.2689112063271631,
      "eval_loss": 2.611741542816162,
      "eval_runtime": 21.9709,
      "eval_samples_per_second": 227.574,
      "eval_steps_per_second": 1.229,
      "step": 10962
    },
    {
      "epoch": 0.2693447592595729,
      "grad_norm": 0.2955268621444702,
      "learning_rate": 5.276853866431787e-05,
      "loss": 1.906,
      "step": 10970
    },
    {
      "epoch": 0.2698867004250851,
      "grad_norm": 0.44501546025276184,
      "learning_rate": 5.267716504009533e-05,
      "loss": 1.9085,
      "step": 10980
    },
    {
      "epoch": 0.2704286415905973,
      "grad_norm": 0.3195658326148987,
      "learning_rate": 5.258580101758203e-05,
      "loss": 1.9186,
      "step": 10990
    },
    {
      "epoch": 0.2709705827561095,
      "grad_norm": 0.5696297883987427,
      "learning_rate": 5.2494446974441837e-05,
      "loss": 1.9061,
      "step": 11000
    },
    {
      "epoch": 0.2715125239216218,
      "grad_norm": 0.29219719767570496,
      "learning_rate": 5.2403103288297314e-05,
      "loss": 1.9114,
      "step": 11010
    },
    {
      "epoch": 0.272054465087134,
      "grad_norm": 0.3206324279308319,
      "learning_rate": 5.231177033672824e-05,
      "loss": 1.9041,
      "step": 11020
    },
    {
      "epoch": 0.2725964062526462,
      "grad_norm": 0.33854442834854126,
      "learning_rate": 5.222044849727005e-05,
      "loss": 1.9057,
      "step": 11030
    },
    {
      "epoch": 0.2731383474181584,
      "grad_norm": 0.2936360538005829,
      "learning_rate": 5.212913814741219e-05,
      "loss": 1.918,
      "step": 11040
    },
    {
      "epoch": 0.27362609446711944,
      "eval_loss": 2.61767840385437,
      "eval_runtime": 21.6394,
      "eval_samples_per_second": 231.06,
      "eval_steps_per_second": 1.248,
      "step": 11049
    },
    {
      "epoch": 0.27368028858367066,
      "grad_norm": 0.34562787413597107,
      "learning_rate": 5.203783966459665e-05,
      "loss": 1.9068,
      "step": 11050
    },
    {
      "epoch": 0.27422222974918287,
      "grad_norm": 0.5706035494804382,
      "learning_rate": 5.1946553426216394e-05,
      "loss": 1.9123,
      "step": 11060
    },
    {
      "epoch": 0.27476417091469507,
      "grad_norm": 0.5709580183029175,
      "learning_rate": 5.1855279809613675e-05,
      "loss": 1.9214,
      "step": 11070
    },
    {
      "epoch": 0.2753061120802073,
      "grad_norm": 0.3960080146789551,
      "learning_rate": 5.1764019192078686e-05,
      "loss": 1.913,
      "step": 11080
    },
    {
      "epoch": 0.27584805324571954,
      "grad_norm": 0.4222663342952728,
      "learning_rate": 5.16727719508478e-05,
      "loss": 1.9182,
      "step": 11090
    },
    {
      "epoch": 0.27638999441123174,
      "grad_norm": 0.6161875128746033,
      "learning_rate": 5.158153846310214e-05,
      "loss": 1.9089,
      "step": 11100
    },
    {
      "epoch": 0.27693193557674395,
      "grad_norm": 0.5677464008331299,
      "learning_rate": 5.149031910596599e-05,
      "loss": 1.8998,
      "step": 11110
    },
    {
      "epoch": 0.27747387674225615,
      "grad_norm": 0.3052445352077484,
      "learning_rate": 5.139911425650518e-05,
      "loss": 1.9103,
      "step": 11120
    },
    {
      "epoch": 0.2780158179077684,
      "grad_norm": 0.2828851044178009,
      "learning_rate": 5.1307924291725583e-05,
      "loss": 1.907,
      "step": 11130
    },
    {
      "epoch": 0.27834098260707574,
      "eval_loss": 2.613119125366211,
      "eval_runtime": 21.9757,
      "eval_samples_per_second": 227.524,
      "eval_steps_per_second": 1.229,
      "step": 11136
    },
    {
      "epoch": 0.2785577590732806,
      "grad_norm": 0.33760419487953186,
      "learning_rate": 5.121674958857159e-05,
      "loss": 1.9194,
      "step": 11140
    },
    {
      "epoch": 0.2790997002387928,
      "grad_norm": 0.3404898941516876,
      "learning_rate": 5.112559052392444e-05,
      "loss": 1.9043,
      "step": 11150
    },
    {
      "epoch": 0.27964164140430503,
      "grad_norm": 0.3572676181793213,
      "learning_rate": 5.103444747460079e-05,
      "loss": 1.9091,
      "step": 11160
    },
    {
      "epoch": 0.2801835825698173,
      "grad_norm": 0.3610747456550598,
      "learning_rate": 5.0943320817351034e-05,
      "loss": 1.9049,
      "step": 11170
    },
    {
      "epoch": 0.2807255237353295,
      "grad_norm": 0.46232640743255615,
      "learning_rate": 5.085221092885785e-05,
      "loss": 1.9233,
      "step": 11180
    },
    {
      "epoch": 0.2812674649008417,
      "grad_norm": 0.27908554673194885,
      "learning_rate": 5.076111818573459e-05,
      "loss": 1.9064,
      "step": 11190
    },
    {
      "epoch": 0.2818094060663539,
      "grad_norm": 0.6692981719970703,
      "learning_rate": 5.0670042964523745e-05,
      "loss": 1.9067,
      "step": 11200
    },
    {
      "epoch": 0.28235134723186617,
      "grad_norm": 0.3264489769935608,
      "learning_rate": 5.057898564169534e-05,
      "loss": 1.8999,
      "step": 11210
    },
    {
      "epoch": 0.2828932883973784,
      "grad_norm": 0.5410310626029968,
      "learning_rate": 5.048794659364546e-05,
      "loss": 1.9098,
      "step": 11220
    },
    {
      "epoch": 0.28305587074703203,
      "eval_loss": 2.6135261058807373,
      "eval_runtime": 21.9689,
      "eval_samples_per_second": 227.594,
      "eval_steps_per_second": 1.229,
      "step": 11223
    },
    {
      "epoch": 0.2834352295628906,
      "grad_norm": 0.6051361560821533,
      "learning_rate": 5.0396926196694626e-05,
      "loss": 1.9124,
      "step": 11230
    },
    {
      "epoch": 0.2839771707284028,
      "grad_norm": 0.5583413243293762,
      "learning_rate": 5.030592482708626e-05,
      "loss": 1.9025,
      "step": 11240
    },
    {
      "epoch": 0.28451911189391504,
      "grad_norm": 0.6230108737945557,
      "learning_rate": 5.021494286098514e-05,
      "loss": 1.9113,
      "step": 11250
    },
    {
      "epoch": 0.28506105305942725,
      "grad_norm": 0.538650631904602,
      "learning_rate": 5.0123980674475824e-05,
      "loss": 1.9034,
      "step": 11260
    },
    {
      "epoch": 0.28560299422493945,
      "grad_norm": 0.6079558730125427,
      "learning_rate": 5.003303864356115e-05,
      "loss": 1.9154,
      "step": 11270
    },
    {
      "epoch": 0.28614493539045166,
      "grad_norm": 0.5191447138786316,
      "learning_rate": 4.994211714416058e-05,
      "loss": 1.8997,
      "step": 11280
    },
    {
      "epoch": 0.28668687655596387,
      "grad_norm": 0.2797117233276367,
      "learning_rate": 4.9851216552108745e-05,
      "loss": 1.8978,
      "step": 11290
    },
    {
      "epoch": 0.2872288177214761,
      "grad_norm": 0.29481279850006104,
      "learning_rate": 4.976033724315385e-05,
      "loss": 1.9152,
      "step": 11300
    },
    {
      "epoch": 0.28777075888698833,
      "grad_norm": 0.390680193901062,
      "learning_rate": 4.966947959295612e-05,
      "loss": 1.9112,
      "step": 11310
    },
    {
      "epoch": 0.28777075888698833,
      "eval_loss": 2.617300033569336,
      "eval_runtime": 21.9665,
      "eval_samples_per_second": 227.62,
      "eval_steps_per_second": 1.229,
      "step": 11310
    },
    {
      "epoch": 0.28831270005250054,
      "grad_norm": 0.6555575132369995,
      "learning_rate": 4.957864397708625e-05,
      "loss": 1.8965,
      "step": 11320
    },
    {
      "epoch": 0.28885464121801274,
      "grad_norm": 0.3799804449081421,
      "learning_rate": 4.948783077102385e-05,
      "loss": 1.9039,
      "step": 11330
    },
    {
      "epoch": 0.289396582383525,
      "grad_norm": 0.2934130132198334,
      "learning_rate": 4.9397040350155876e-05,
      "loss": 1.9078,
      "step": 11340
    },
    {
      "epoch": 0.2899385235490372,
      "grad_norm": 0.2916678190231323,
      "learning_rate": 4.930627308977517e-05,
      "loss": 1.892,
      "step": 11350
    },
    {
      "epoch": 0.2904804647145494,
      "grad_norm": 0.30668988823890686,
      "learning_rate": 4.921552936507876e-05,
      "loss": 1.9098,
      "step": 11360
    },
    {
      "epoch": 0.2910224058800616,
      "grad_norm": 0.4994325637817383,
      "learning_rate": 4.912480955116642e-05,
      "loss": 1.9045,
      "step": 11370
    },
    {
      "epoch": 0.2915643470455739,
      "grad_norm": 1.1822820901870728,
      "learning_rate": 4.9034114023039076e-05,
      "loss": 1.8957,
      "step": 11380
    },
    {
      "epoch": 0.2921062882110861,
      "grad_norm": 0.3069334626197815,
      "learning_rate": 4.894344315559729e-05,
      "loss": 1.9096,
      "step": 11390
    },
    {
      "epoch": 0.29248564702694463,
      "eval_loss": 2.6101443767547607,
      "eval_runtime": 21.9727,
      "eval_samples_per_second": 227.556,
      "eval_steps_per_second": 1.229,
      "step": 11397
    },
    {
      "epoch": 0.2926482293765983,
      "grad_norm": 0.3290071487426758,
      "learning_rate": 4.885279732363967e-05,
      "loss": 1.9007,
      "step": 11400
    },
    {
      "epoch": 0.2931901705421105,
      "grad_norm": 0.27592140436172485,
      "learning_rate": 4.876217690186131e-05,
      "loss": 1.8886,
      "step": 11410
    },
    {
      "epoch": 0.29373211170762276,
      "grad_norm": 0.5586705207824707,
      "learning_rate": 4.867158226485231e-05,
      "loss": 1.8874,
      "step": 11420
    },
    {
      "epoch": 0.29427405287313496,
      "grad_norm": 0.5295722484588623,
      "learning_rate": 4.858101378709616e-05,
      "loss": 1.8926,
      "step": 11430
    },
    {
      "epoch": 0.29481599403864717,
      "grad_norm": 0.3275451064109802,
      "learning_rate": 4.8490471842968267e-05,
      "loss": 1.908,
      "step": 11440
    },
    {
      "epoch": 0.29535793520415937,
      "grad_norm": 0.2957296371459961,
      "learning_rate": 4.8399956806734234e-05,
      "loss": 1.899,
      "step": 11450
    },
    {
      "epoch": 0.29589987636967163,
      "grad_norm": 0.30543988943099976,
      "learning_rate": 4.830946905254861e-05,
      "loss": 1.9042,
      "step": 11460
    },
    {
      "epoch": 0.29644181753518384,
      "grad_norm": 0.5068356394767761,
      "learning_rate": 4.821900895445302e-05,
      "loss": 1.8957,
      "step": 11470
    },
    {
      "epoch": 0.29698375870069604,
      "grad_norm": 0.3116234838962555,
      "learning_rate": 4.812857688637486e-05,
      "loss": 1.9,
      "step": 11480
    },
    {
      "epoch": 0.2972005351669009,
      "eval_loss": 2.6111230850219727,
      "eval_runtime": 21.9729,
      "eval_samples_per_second": 227.553,
      "eval_steps_per_second": 1.229,
      "step": 11484
    },
    {
      "epoch": 0.29752569986620825,
      "grad_norm": 0.31971922516822815,
      "learning_rate": 4.8038173222125645e-05,
      "loss": 1.9074,
      "step": 11490
    },
    {
      "epoch": 0.2980676410317205,
      "grad_norm": 0.3503149747848511,
      "learning_rate": 4.7947798335399416e-05,
      "loss": 1.899,
      "step": 11500
    },
    {
      "epoch": 0.2986095821972327,
      "grad_norm": 0.30250006914138794,
      "learning_rate": 4.7857452599771354e-05,
      "loss": 1.9063,
      "step": 11510
    },
    {
      "epoch": 0.2991515233627449,
      "grad_norm": 0.7795620560646057,
      "learning_rate": 4.77671363886961e-05,
      "loss": 1.9005,
      "step": 11520
    },
    {
      "epoch": 0.2996934645282571,
      "grad_norm": 0.5038977265357971,
      "learning_rate": 4.7676850075506185e-05,
      "loss": 1.8991,
      "step": 11530
    },
    {
      "epoch": 0.3002354056937694,
      "grad_norm": 0.4416648745536804,
      "learning_rate": 4.758659403341069e-05,
      "loss": 1.893,
      "step": 11540
    },
    {
      "epoch": 0.3007773468592816,
      "grad_norm": 0.3159092366695404,
      "learning_rate": 4.749636863549346e-05,
      "loss": 1.9063,
      "step": 11550
    },
    {
      "epoch": 0.3013192880247938,
      "grad_norm": 0.33458906412124634,
      "learning_rate": 4.740617425471168e-05,
      "loss": 1.9001,
      "step": 11560
    },
    {
      "epoch": 0.301861229190306,
      "grad_norm": 0.5140174627304077,
      "learning_rate": 4.731601126389438e-05,
      "loss": 1.899,
      "step": 11570
    },
    {
      "epoch": 0.3019154233068572,
      "eval_loss": 2.6166553497314453,
      "eval_runtime": 21.9725,
      "eval_samples_per_second": 227.558,
      "eval_steps_per_second": 1.229,
      "step": 11571
    },
    {
      "epoch": 0.30240317035581826,
      "grad_norm": 0.570547878742218,
      "learning_rate": 4.722588003574077e-05,
      "loss": 1.895,
      "step": 11580
    },
    {
      "epoch": 0.30294511152133047,
      "grad_norm": 0.44531741738319397,
      "learning_rate": 4.7135780942818817e-05,
      "loss": 1.9075,
      "step": 11590
    },
    {
      "epoch": 0.3034870526868427,
      "grad_norm": 0.47997337579727173,
      "learning_rate": 4.704571435756363e-05,
      "loss": 1.8995,
      "step": 11600
    },
    {
      "epoch": 0.3040289938523549,
      "grad_norm": 0.47242090106010437,
      "learning_rate": 4.6955680652275916e-05,
      "loss": 1.8903,
      "step": 11610
    },
    {
      "epoch": 0.30457093501786714,
      "grad_norm": 0.28598716855049133,
      "learning_rate": 4.6865680199120545e-05,
      "loss": 1.8922,
      "step": 11620
    },
    {
      "epoch": 0.30511287618337934,
      "grad_norm": 0.44576406478881836,
      "learning_rate": 4.677571337012484e-05,
      "loss": 1.8952,
      "step": 11630
    },
    {
      "epoch": 0.30565481734889155,
      "grad_norm": 0.3293391764163971,
      "learning_rate": 4.668578053717721e-05,
      "loss": 1.8935,
      "step": 11640
    },
    {
      "epoch": 0.30619675851440376,
      "grad_norm": 0.31350746750831604,
      "learning_rate": 4.65958820720255e-05,
      "loss": 1.8906,
      "step": 11650
    },
    {
      "epoch": 0.3066303114468136,
      "eval_loss": 2.620225191116333,
      "eval_runtime": 21.9691,
      "eval_samples_per_second": 227.592,
      "eval_steps_per_second": 1.229,
      "step": 11658
    },
    {
      "epoch": 0.306738699679916,
      "grad_norm": 0.3374342918395996,
      "learning_rate": 4.650601834627549e-05,
      "loss": 1.8892,
      "step": 11660
    },
    {
      "epoch": 0.3072806408454282,
      "grad_norm": 0.365682452917099,
      "learning_rate": 4.641618973138942e-05,
      "loss": 1.892,
      "step": 11670
    },
    {
      "epoch": 0.3078225820109404,
      "grad_norm": 0.3332476019859314,
      "learning_rate": 4.6326396598684296e-05,
      "loss": 1.8927,
      "step": 11680
    },
    {
      "epoch": 0.30836452317645263,
      "grad_norm": 1.3207186460494995,
      "learning_rate": 4.6236639319330524e-05,
      "loss": 1.8873,
      "step": 11690
    },
    {
      "epoch": 0.3089064643419649,
      "grad_norm": 0.3541896641254425,
      "learning_rate": 4.614691826435028e-05,
      "loss": 1.8908,
      "step": 11700
    },
    {
      "epoch": 0.3094484055074771,
      "grad_norm": 0.486902117729187,
      "learning_rate": 4.605723380461603e-05,
      "loss": 1.8982,
      "step": 11710
    },
    {
      "epoch": 0.3099903466729893,
      "grad_norm": 0.33585986495018005,
      "learning_rate": 4.596758631084892e-05,
      "loss": 1.901,
      "step": 11720
    },
    {
      "epoch": 0.3105322878385015,
      "grad_norm": 0.37316057085990906,
      "learning_rate": 4.587797615361735e-05,
      "loss": 1.8892,
      "step": 11730
    },
    {
      "epoch": 0.31107422900401377,
      "grad_norm": 0.4041992723941803,
      "learning_rate": 4.578840370333534e-05,
      "loss": 1.8928,
      "step": 11740
    },
    {
      "epoch": 0.31134519958676987,
      "eval_loss": 2.6146914958953857,
      "eval_runtime": 21.9662,
      "eval_samples_per_second": 227.622,
      "eval_steps_per_second": 1.229,
      "step": 11745
    },
    {
      "epoch": 0.311616170169526,
      "grad_norm": 0.2976398169994354,
      "learning_rate": 4.569886933026107e-05,
      "loss": 1.8848,
      "step": 11750
    },
    {
      "epoch": 0.3121581113350382,
      "grad_norm": 0.3247699737548828,
      "learning_rate": 4.5609373404495316e-05,
      "loss": 1.8998,
      "step": 11760
    },
    {
      "epoch": 0.3127000525005504,
      "grad_norm": 0.29927968978881836,
      "learning_rate": 4.55199162959799e-05,
      "loss": 1.8883,
      "step": 11770
    },
    {
      "epoch": 0.31324199366606265,
      "grad_norm": 0.301215797662735,
      "learning_rate": 4.543049837449626e-05,
      "loss": 1.8916,
      "step": 11780
    },
    {
      "epoch": 0.31378393483157485,
      "grad_norm": 0.3488280773162842,
      "learning_rate": 4.534112000966377e-05,
      "loss": 1.9006,
      "step": 11790
    },
    {
      "epoch": 0.31432587599708706,
      "grad_norm": 0.36820992827415466,
      "learning_rate": 4.5251781570938324e-05,
      "loss": 1.8888,
      "step": 11800
    },
    {
      "epoch": 0.31486781716259926,
      "grad_norm": 0.33846035599708557,
      "learning_rate": 4.51624834276108e-05,
      "loss": 1.8845,
      "step": 11810
    },
    {
      "epoch": 0.3154097583281115,
      "grad_norm": 0.3961305320262909,
      "learning_rate": 4.5073225948805476e-05,
      "loss": 1.8928,
      "step": 11820
    },
    {
      "epoch": 0.3159516994936237,
      "grad_norm": 0.29425087571144104,
      "learning_rate": 4.498400950347855e-05,
      "loss": 1.8951,
      "step": 11830
    },
    {
      "epoch": 0.31606008772672617,
      "eval_loss": 2.6105449199676514,
      "eval_runtime": 22.294,
      "eval_samples_per_second": 224.276,
      "eval_steps_per_second": 1.211,
      "step": 11832
    },
    {
      "epoch": 0.31649364065913593,
      "grad_norm": 0.32833293080329895,
      "learning_rate": 4.4894834460416626e-05,
      "loss": 1.8865,
      "step": 11840
    },
    {
      "epoch": 0.31703558182464814,
      "grad_norm": 0.274554580450058,
      "learning_rate": 4.480570118823511e-05,
      "loss": 1.8974,
      "step": 11850
    },
    {
      "epoch": 0.3175775229901604,
      "grad_norm": 0.3351888656616211,
      "learning_rate": 4.471661005537682e-05,
      "loss": 1.8834,
      "step": 11860
    },
    {
      "epoch": 0.3181194641556726,
      "grad_norm": 0.6972358226776123,
      "learning_rate": 4.462756143011031e-05,
      "loss": 1.8822,
      "step": 11870
    },
    {
      "epoch": 0.3186614053211848,
      "grad_norm": 0.36904025077819824,
      "learning_rate": 4.453855568052847e-05,
      "loss": 1.8972,
      "step": 11880
    },
    {
      "epoch": 0.319203346486697,
      "grad_norm": 0.27459755539894104,
      "learning_rate": 4.444959317454696e-05,
      "loss": 1.8854,
      "step": 11890
    },
    {
      "epoch": 0.3197452876522093,
      "grad_norm": 0.3487244248390198,
      "learning_rate": 4.436067427990266e-05,
      "loss": 1.8857,
      "step": 11900
    },
    {
      "epoch": 0.3202872288177215,
      "grad_norm": 0.3203682005405426,
      "learning_rate": 4.42717993641522e-05,
      "loss": 1.8869,
      "step": 11910
    },
    {
      "epoch": 0.32077497586668247,
      "eval_loss": 2.613398313522339,
      "eval_runtime": 21.9703,
      "eval_samples_per_second": 227.58,
      "eval_steps_per_second": 1.229,
      "step": 11919
    },
    {
      "epoch": 0.3208291699832337,
      "grad_norm": 0.30035504698753357,
      "learning_rate": 4.418296879467041e-05,
      "loss": 1.8942,
      "step": 11920
    },
    {
      "epoch": 0.3213711111487459,
      "grad_norm": 0.38128897547721863,
      "learning_rate": 4.409418293864881e-05,
      "loss": 1.8833,
      "step": 11930
    },
    {
      "epoch": 0.32191305231425815,
      "grad_norm": 0.31847649812698364,
      "learning_rate": 4.400544216309409e-05,
      "loss": 1.8946,
      "step": 11940
    },
    {
      "epoch": 0.32245499347977036,
      "grad_norm": 0.6311419606208801,
      "learning_rate": 4.3916746834826604e-05,
      "loss": 1.887,
      "step": 11950
    },
    {
      "epoch": 0.32299693464528256,
      "grad_norm": 0.5365369319915771,
      "learning_rate": 4.3828097320478825e-05,
      "loss": 1.881,
      "step": 11960
    },
    {
      "epoch": 0.32353887581079477,
      "grad_norm": 0.3272137939929962,
      "learning_rate": 4.3739493986493864e-05,
      "loss": 1.8887,
      "step": 11970
    },
    {
      "epoch": 0.32408081697630703,
      "grad_norm": 0.3870546519756317,
      "learning_rate": 4.3650937199123934e-05,
      "loss": 1.8752,
      "step": 11980
    },
    {
      "epoch": 0.32462275814181923,
      "grad_norm": 0.47655603289604187,
      "learning_rate": 4.356242732442887e-05,
      "loss": 1.8959,
      "step": 11990
    },
    {
      "epoch": 0.32516469930733144,
      "grad_norm": 0.5217764377593994,
      "learning_rate": 4.3473964728274516e-05,
      "loss": 1.8831,
      "step": 12000
    },
    {
      "epoch": 0.32548986400663876,
      "eval_loss": 2.6064906120300293,
      "eval_runtime": 21.97,
      "eval_samples_per_second": 227.583,
      "eval_steps_per_second": 1.229,
      "step": 12006
    },
    {
      "epoch": 0.32570664047284364,
      "grad_norm": 0.4419178068637848,
      "learning_rate": 4.338554977633138e-05,
      "loss": 1.8843,
      "step": 12010
    },
    {
      "epoch": 0.3262485816383559,
      "grad_norm": 0.38523489236831665,
      "learning_rate": 4.3297182834072944e-05,
      "loss": 1.8829,
      "step": 12020
    },
    {
      "epoch": 0.3267905228038681,
      "grad_norm": 0.537399411201477,
      "learning_rate": 4.3208864266774294e-05,
      "loss": 1.8814,
      "step": 12030
    },
    {
      "epoch": 0.3273324639693803,
      "grad_norm": 0.3097943365573883,
      "learning_rate": 4.312059443951051e-05,
      "loss": 1.8777,
      "step": 12040
    },
    {
      "epoch": 0.3278744051348925,
      "grad_norm": 0.3149316608905792,
      "learning_rate": 4.303237371715524e-05,
      "loss": 1.8874,
      "step": 12050
    },
    {
      "epoch": 0.3284163463004048,
      "grad_norm": 0.35441088676452637,
      "learning_rate": 4.2944202464379125e-05,
      "loss": 1.8776,
      "step": 12060
    },
    {
      "epoch": 0.328958287465917,
      "grad_norm": 0.34289848804473877,
      "learning_rate": 4.2856081045648285e-05,
      "loss": 1.8946,
      "step": 12070
    },
    {
      "epoch": 0.3295002286314292,
      "grad_norm": 0.3778747320175171,
      "learning_rate": 4.276800982522293e-05,
      "loss": 1.8779,
      "step": 12080
    },
    {
      "epoch": 0.3300421697969414,
      "grad_norm": 0.4332473874092102,
      "learning_rate": 4.26799891671557e-05,
      "loss": 1.8827,
      "step": 12090
    },
    {
      "epoch": 0.33020475214659506,
      "eval_loss": 2.6110384464263916,
      "eval_runtime": 21.9727,
      "eval_samples_per_second": 227.555,
      "eval_steps_per_second": 1.229,
      "step": 12093
    },
    {
      "epoch": 0.33058411096245366,
      "grad_norm": 0.334953248500824,
      "learning_rate": 4.2592019435290266e-05,
      "loss": 1.8724,
      "step": 12100
    },
    {
      "epoch": 0.33112605212796586,
      "grad_norm": 0.36350664496421814,
      "learning_rate": 4.2504100993259774e-05,
      "loss": 1.8744,
      "step": 12110
    },
    {
      "epoch": 0.33166799329347807,
      "grad_norm": 0.44730937480926514,
      "learning_rate": 4.241623420448533e-05,
      "loss": 1.8722,
      "step": 12120
    },
    {
      "epoch": 0.3322099344589903,
      "grad_norm": 0.6003024578094482,
      "learning_rate": 4.2328419432174605e-05,
      "loss": 1.8838,
      "step": 12130
    },
    {
      "epoch": 0.33275187562450254,
      "grad_norm": 0.27411770820617676,
      "learning_rate": 4.224065703932016e-05,
      "loss": 1.8716,
      "step": 12140
    },
    {
      "epoch": 0.33329381679001474,
      "grad_norm": 0.3935556411743164,
      "learning_rate": 4.215294738869808e-05,
      "loss": 1.8874,
      "step": 12150
    },
    {
      "epoch": 0.33383575795552695,
      "grad_norm": 0.32209470868110657,
      "learning_rate": 4.206529084286649e-05,
      "loss": 1.877,
      "step": 12160
    },
    {
      "epoch": 0.33437769912103915,
      "grad_norm": 0.5299956202507019,
      "learning_rate": 4.197768776416387e-05,
      "loss": 1.8806,
      "step": 12170
    },
    {
      "epoch": 0.3349196402865514,
      "grad_norm": 0.5041821002960205,
      "learning_rate": 4.1890138514707835e-05,
      "loss": 1.8764,
      "step": 12180
    },
    {
      "epoch": 0.3349196402865514,
      "eval_loss": 2.6070642471313477,
      "eval_runtime": 21.9587,
      "eval_samples_per_second": 227.7,
      "eval_steps_per_second": 1.23,
      "step": 12180
    },
    {
      "epoch": 0.3354615814520636,
      "grad_norm": 0.4416036307811737,
      "learning_rate": 4.180264345639339e-05,
      "loss": 1.8713,
      "step": 12190
    },
    {
      "epoch": 0.3360035226175758,
      "grad_norm": 0.2827119827270508,
      "learning_rate": 4.171520295089153e-05,
      "loss": 1.8777,
      "step": 12200
    },
    {
      "epoch": 0.33654546378308803,
      "grad_norm": 0.374304860830307,
      "learning_rate": 4.1627817359647846e-05,
      "loss": 1.8803,
      "step": 12210
    },
    {
      "epoch": 0.3370874049486003,
      "grad_norm": 0.27908772230148315,
      "learning_rate": 4.1540487043880824e-05,
      "loss": 1.8862,
      "step": 12220
    },
    {
      "epoch": 0.3376293461141125,
      "grad_norm": 0.41203275322914124,
      "learning_rate": 4.145321236458053e-05,
      "loss": 1.8731,
      "step": 12230
    },
    {
      "epoch": 0.3381712872796247,
      "grad_norm": 0.332907497882843,
      "learning_rate": 4.136599368250704e-05,
      "loss": 1.8844,
      "step": 12240
    },
    {
      "epoch": 0.3387132284451369,
      "grad_norm": 0.302048921585083,
      "learning_rate": 4.1278831358188915e-05,
      "loss": 1.8878,
      "step": 12250
    },
    {
      "epoch": 0.33925516961064917,
      "grad_norm": 0.40147966146469116,
      "learning_rate": 4.119172575192185e-05,
      "loss": 1.876,
      "step": 12260
    },
    {
      "epoch": 0.3396345284265077,
      "eval_loss": 2.6118130683898926,
      "eval_runtime": 21.9708,
      "eval_samples_per_second": 227.575,
      "eval_steps_per_second": 1.229,
      "step": 12267
    },
    {
      "epoch": 0.33979711077616137,
      "grad_norm": 0.2979309856891632,
      "learning_rate": 4.110467722376697e-05,
      "loss": 1.8913,
      "step": 12270
    },
    {
      "epoch": 0.3403390519416736,
      "grad_norm": 0.3392654061317444,
      "learning_rate": 4.1017686133549524e-05,
      "loss": 1.8759,
      "step": 12280
    },
    {
      "epoch": 0.3408809931071858,
      "grad_norm": 0.44583526253700256,
      "learning_rate": 4.093075284085738e-05,
      "loss": 1.8815,
      "step": 12290
    },
    {
      "epoch": 0.34142293427269804,
      "grad_norm": 0.3814590573310852,
      "learning_rate": 4.084387770503939e-05,
      "loss": 1.8834,
      "step": 12300
    },
    {
      "epoch": 0.34196487543821025,
      "grad_norm": 0.2939944565296173,
      "learning_rate": 4.0757061085204084e-05,
      "loss": 1.8648,
      "step": 12310
    },
    {
      "epoch": 0.34250681660372245,
      "grad_norm": 0.41222232580184937,
      "learning_rate": 4.0670303340218085e-05,
      "loss": 1.8683,
      "step": 12320
    },
    {
      "epoch": 0.34304875776923466,
      "grad_norm": 0.307704359292984,
      "learning_rate": 4.058360482870464e-05,
      "loss": 1.8837,
      "step": 12330
    },
    {
      "epoch": 0.3435906989347469,
      "grad_norm": 0.41536006331443787,
      "learning_rate": 4.049696590904218e-05,
      "loss": 1.8883,
      "step": 12340
    },
    {
      "epoch": 0.3441326401002591,
      "grad_norm": 0.29493269324302673,
      "learning_rate": 4.0410386939362774e-05,
      "loss": 1.8749,
      "step": 12350
    },
    {
      "epoch": 0.344349416566464,
      "eval_loss": 2.604987144470215,
      "eval_runtime": 21.9698,
      "eval_samples_per_second": 227.585,
      "eval_steps_per_second": 1.229,
      "step": 12354
    },
    {
      "epoch": 0.34467458126577133,
      "grad_norm": 0.3915794789791107,
      "learning_rate": 4.032386827755069e-05,
      "loss": 1.8712,
      "step": 12360
    },
    {
      "epoch": 0.34521652243128353,
      "grad_norm": 0.3644906282424927,
      "learning_rate": 4.0237410281240915e-05,
      "loss": 1.8795,
      "step": 12370
    },
    {
      "epoch": 0.3457584635967958,
      "grad_norm": 0.29800209403038025,
      "learning_rate": 4.015101330781764e-05,
      "loss": 1.8704,
      "step": 12380
    },
    {
      "epoch": 0.346300404762308,
      "grad_norm": 0.3081452250480652,
      "learning_rate": 4.0064677714412856e-05,
      "loss": 1.8776,
      "step": 12390
    },
    {
      "epoch": 0.3468423459278202,
      "grad_norm": 0.3264562487602234,
      "learning_rate": 3.997840385790481e-05,
      "loss": 1.8819,
      "step": 12400
    },
    {
      "epoch": 0.3473842870933324,
      "grad_norm": 0.30599161982536316,
      "learning_rate": 3.989219209491652e-05,
      "loss": 1.8903,
      "step": 12410
    },
    {
      "epoch": 0.34792622825884467,
      "grad_norm": 0.31660717725753784,
      "learning_rate": 3.98060427818144e-05,
      "loss": 1.8748,
      "step": 12420
    },
    {
      "epoch": 0.3484681694243569,
      "grad_norm": 0.41350027918815613,
      "learning_rate": 3.971995627470668e-05,
      "loss": 1.8775,
      "step": 12430
    },
    {
      "epoch": 0.3490101105898691,
      "grad_norm": 0.43306195735931396,
      "learning_rate": 3.963393292944195e-05,
      "loss": 1.8683,
      "step": 12440
    },
    {
      "epoch": 0.3490643047064203,
      "eval_loss": 2.6103334426879883,
      "eval_runtime": 21.967,
      "eval_samples_per_second": 227.614,
      "eval_steps_per_second": 1.229,
      "step": 12441
    },
    {
      "epoch": 0.3495520517553813,
      "grad_norm": 0.3944796323776245,
      "learning_rate": 3.954797310160777e-05,
      "loss": 1.8695,
      "step": 12450
    },
    {
      "epoch": 0.35009399292089355,
      "grad_norm": 0.3327912986278534,
      "learning_rate": 3.946207714652911e-05,
      "loss": 1.8703,
      "step": 12460
    },
    {
      "epoch": 0.35063593408640575,
      "grad_norm": 0.38429713249206543,
      "learning_rate": 3.937624541926689e-05,
      "loss": 1.8715,
      "step": 12470
    },
    {
      "epoch": 0.35117787525191796,
      "grad_norm": 0.6065689325332642,
      "learning_rate": 3.9290478274616605e-05,
      "loss": 1.8717,
      "step": 12480
    },
    {
      "epoch": 0.35171981641743016,
      "grad_norm": 0.31308409571647644,
      "learning_rate": 3.920477606710673e-05,
      "loss": 1.8792,
      "step": 12490
    },
    {
      "epoch": 0.3522617575829424,
      "grad_norm": 0.5742292404174805,
      "learning_rate": 3.911913915099734e-05,
      "loss": 1.8788,
      "step": 12500
    },
    {
      "epoch": 0.35280369874845463,
      "grad_norm": 0.4109736979007721,
      "learning_rate": 3.903356788027863e-05,
      "loss": 1.8666,
      "step": 12510
    },
    {
      "epoch": 0.35334563991396684,
      "grad_norm": 0.6786653399467468,
      "learning_rate": 3.894806260866941e-05,
      "loss": 1.8743,
      "step": 12520
    },
    {
      "epoch": 0.3537791928463766,
      "eval_loss": 2.6084859371185303,
      "eval_runtime": 21.9673,
      "eval_samples_per_second": 227.611,
      "eval_steps_per_second": 1.229,
      "step": 12528
    },
    {
      "epoch": 0.35388758107947904,
      "grad_norm": 0.2924114465713501,
      "learning_rate": 3.886262368961571e-05,
      "loss": 1.8715,
      "step": 12530
    },
    {
      "epoch": 0.3544295222449913,
      "grad_norm": 0.36380696296691895,
      "learning_rate": 3.877725147628925e-05,
      "loss": 1.8659,
      "step": 12540
    },
    {
      "epoch": 0.3549714634105035,
      "grad_norm": 0.3204795718193054,
      "learning_rate": 3.869194632158603e-05,
      "loss": 1.8751,
      "step": 12550
    },
    {
      "epoch": 0.3555134045760157,
      "grad_norm": 0.5478801131248474,
      "learning_rate": 3.8606708578124875e-05,
      "loss": 1.8603,
      "step": 12560
    },
    {
      "epoch": 0.3560553457415279,
      "grad_norm": 0.5209506154060364,
      "learning_rate": 3.852153859824593e-05,
      "loss": 1.868,
      "step": 12570
    },
    {
      "epoch": 0.3565972869070402,
      "grad_norm": 0.43026360869407654,
      "learning_rate": 3.8436436734009243e-05,
      "loss": 1.8737,
      "step": 12580
    },
    {
      "epoch": 0.3571392280725524,
      "grad_norm": 0.3033788800239563,
      "learning_rate": 3.83514033371933e-05,
      "loss": 1.874,
      "step": 12590
    },
    {
      "epoch": 0.3576811692380646,
      "grad_norm": 0.460555762052536,
      "learning_rate": 3.8266438759293555e-05,
      "loss": 1.8764,
      "step": 12600
    },
    {
      "epoch": 0.3582231104035768,
      "grad_norm": 0.335553377866745,
      "learning_rate": 3.818154335152101e-05,
      "loss": 1.8718,
      "step": 12610
    },
    {
      "epoch": 0.3584940809863329,
      "eval_loss": 2.607743740081787,
      "eval_runtime": 21.9725,
      "eval_samples_per_second": 227.557,
      "eval_steps_per_second": 1.229,
      "step": 12615
    },
    {
      "epoch": 0.35876505156908906,
      "grad_norm": 0.2863267958164215,
      "learning_rate": 3.8096717464800735e-05,
      "loss": 1.8626,
      "step": 12620
    },
    {
      "epoch": 0.35930699273460126,
      "grad_norm": 0.386030375957489,
      "learning_rate": 3.8011961449770403e-05,
      "loss": 1.8737,
      "step": 12630
    },
    {
      "epoch": 0.35984893390011347,
      "grad_norm": 0.5605409145355225,
      "learning_rate": 3.7927275656778936e-05,
      "loss": 1.8686,
      "step": 12640
    },
    {
      "epoch": 0.36039087506562567,
      "grad_norm": 0.32051485776901245,
      "learning_rate": 3.7842660435884916e-05,
      "loss": 1.8766,
      "step": 12650
    },
    {
      "epoch": 0.36093281623113793,
      "grad_norm": 0.31499314308166504,
      "learning_rate": 3.775811613685518e-05,
      "loss": 1.867,
      "step": 12660
    },
    {
      "epoch": 0.36147475739665014,
      "grad_norm": 0.280678927898407,
      "learning_rate": 3.767364310916353e-05,
      "loss": 1.8651,
      "step": 12670
    },
    {
      "epoch": 0.36201669856216234,
      "grad_norm": 0.26214084029197693,
      "learning_rate": 3.7589241701989005e-05,
      "loss": 1.8702,
      "step": 12680
    },
    {
      "epoch": 0.36255863972767455,
      "grad_norm": 0.31816792488098145,
      "learning_rate": 3.750491226421473e-05,
      "loss": 1.8672,
      "step": 12690
    },
    {
      "epoch": 0.3631005808931868,
      "grad_norm": 0.3030293881893158,
      "learning_rate": 3.7420655144426256e-05,
      "loss": 1.8554,
      "step": 12700
    },
    {
      "epoch": 0.36320896912628925,
      "eval_loss": 2.6091229915618896,
      "eval_runtime": 21.9746,
      "eval_samples_per_second": 227.536,
      "eval_steps_per_second": 1.229,
      "step": 12702
    },
    {
      "epoch": 0.363642522058699,
      "grad_norm": 0.4357650578022003,
      "learning_rate": 3.733647069091016e-05,
      "loss": 1.8616,
      "step": 12710
    },
    {
      "epoch": 0.3641844632242112,
      "grad_norm": 0.3486103415489197,
      "learning_rate": 3.725235925165278e-05,
      "loss": 1.8744,
      "step": 12720
    },
    {
      "epoch": 0.3647264043897234,
      "grad_norm": 0.35597971081733704,
      "learning_rate": 3.716832117433853e-05,
      "loss": 1.8693,
      "step": 12730
    },
    {
      "epoch": 0.3652683455552357,
      "grad_norm": 0.6880526542663574,
      "learning_rate": 3.7084356806348566e-05,
      "loss": 1.8561,
      "step": 12740
    },
    {
      "epoch": 0.3658102867207479,
      "grad_norm": 0.36243849992752075,
      "learning_rate": 3.7000466494759445e-05,
      "loss": 1.869,
      "step": 12750
    },
    {
      "epoch": 0.3663522278862601,
      "grad_norm": 0.35954001545906067,
      "learning_rate": 3.691665058634153e-05,
      "loss": 1.8669,
      "step": 12760
    },
    {
      "epoch": 0.3668941690517723,
      "grad_norm": 0.2994548976421356,
      "learning_rate": 3.683290942755767e-05,
      "loss": 1.8587,
      "step": 12770
    },
    {
      "epoch": 0.36743611021728456,
      "grad_norm": 0.5059475302696228,
      "learning_rate": 3.674924336456173e-05,
      "loss": 1.8584,
      "step": 12780
    },
    {
      "epoch": 0.36792385726624555,
      "eval_loss": 2.6061477661132812,
      "eval_runtime": 21.9728,
      "eval_samples_per_second": 227.554,
      "eval_steps_per_second": 1.229,
      "step": 12789
    },
    {
      "epoch": 0.36797805138279677,
      "grad_norm": 0.4944458305835724,
      "learning_rate": 3.6665652743197075e-05,
      "loss": 1.8711,
      "step": 12790
    },
    {
      "epoch": 0.36851999254830897,
      "grad_norm": 0.3134331703186035,
      "learning_rate": 3.658213790899537e-05,
      "loss": 1.8633,
      "step": 12800
    },
    {
      "epoch": 0.3690619337138212,
      "grad_norm": 0.35285812616348267,
      "learning_rate": 3.649869920717487e-05,
      "loss": 1.8843,
      "step": 12810
    },
    {
      "epoch": 0.36960387487933344,
      "grad_norm": 0.3203577399253845,
      "learning_rate": 3.641533698263917e-05,
      "loss": 1.859,
      "step": 12820
    },
    {
      "epoch": 0.37014581604484564,
      "grad_norm": 0.35432031750679016,
      "learning_rate": 3.6332051579975815e-05,
      "loss": 1.8704,
      "step": 12830
    },
    {
      "epoch": 0.37068775721035785,
      "grad_norm": 0.3153700828552246,
      "learning_rate": 3.624884334345465e-05,
      "loss": 1.8622,
      "step": 12840
    },
    {
      "epoch": 0.37122969837587005,
      "grad_norm": 0.7621405124664307,
      "learning_rate": 3.616571261702669e-05,
      "loss": 1.8665,
      "step": 12850
    },
    {
      "epoch": 0.3717716395413823,
      "grad_norm": 0.5706172585487366,
      "learning_rate": 3.6082659744322464e-05,
      "loss": 1.8578,
      "step": 12860
    },
    {
      "epoch": 0.3723135807068945,
      "grad_norm": 0.6530439853668213,
      "learning_rate": 3.59996850686507e-05,
      "loss": 1.8579,
      "step": 12870
    },
    {
      "epoch": 0.37263874540620184,
      "eval_loss": 2.6071505546569824,
      "eval_runtime": 21.9724,
      "eval_samples_per_second": 227.558,
      "eval_steps_per_second": 1.229,
      "step": 12876
    },
    {
      "epoch": 0.3728555218724067,
      "grad_norm": 0.49953022599220276,
      "learning_rate": 3.591678893299693e-05,
      "loss": 1.8671,
      "step": 12880
    },
    {
      "epoch": 0.37339746303791893,
      "grad_norm": 0.37402573227882385,
      "learning_rate": 3.583397168002196e-05,
      "loss": 1.8639,
      "step": 12890
    },
    {
      "epoch": 0.3739394042034312,
      "grad_norm": 0.281125545501709,
      "learning_rate": 3.575123365206057e-05,
      "loss": 1.8736,
      "step": 12900
    },
    {
      "epoch": 0.3744813453689434,
      "grad_norm": 0.29488998651504517,
      "learning_rate": 3.566857519112008e-05,
      "loss": 1.8704,
      "step": 12910
    },
    {
      "epoch": 0.3750232865344556,
      "grad_norm": 0.29372894763946533,
      "learning_rate": 3.558599663887886e-05,
      "loss": 1.868,
      "step": 12920
    },
    {
      "epoch": 0.3755652276999678,
      "grad_norm": 0.27853238582611084,
      "learning_rate": 3.550349833668499e-05,
      "loss": 1.8644,
      "step": 12930
    },
    {
      "epoch": 0.37610716886548007,
      "grad_norm": 0.2777658998966217,
      "learning_rate": 3.542108062555483e-05,
      "loss": 1.8621,
      "step": 12940
    },
    {
      "epoch": 0.3766491100309923,
      "grad_norm": 0.3306451737880707,
      "learning_rate": 3.5338743846171574e-05,
      "loss": 1.8612,
      "step": 12950
    },
    {
      "epoch": 0.3771910511965045,
      "grad_norm": 0.4671363830566406,
      "learning_rate": 3.525648833888393e-05,
      "loss": 1.857,
      "step": 12960
    },
    {
      "epoch": 0.37735363354615814,
      "eval_loss": 2.6051692962646484,
      "eval_runtime": 21.9698,
      "eval_samples_per_second": 227.585,
      "eval_steps_per_second": 1.229,
      "step": 12963
    },
    {
      "epoch": 0.3777329923620167,
      "grad_norm": 0.3648098111152649,
      "learning_rate": 3.5174314443704634e-05,
      "loss": 1.8643,
      "step": 12970
    },
    {
      "epoch": 0.37827493352752894,
      "grad_norm": 0.4720825254917145,
      "learning_rate": 3.5092222500309066e-05,
      "loss": 1.8606,
      "step": 12980
    },
    {
      "epoch": 0.37881687469304115,
      "grad_norm": 0.3330930173397064,
      "learning_rate": 3.501021284803384e-05,
      "loss": 1.8627,
      "step": 12990
    },
    {
      "epoch": 0.37935881585855336,
      "grad_norm": 0.45893749594688416,
      "learning_rate": 3.492828582587541e-05,
      "loss": 1.8577,
      "step": 13000
    },
    {
      "epoch": 0.37990075702406556,
      "grad_norm": 0.2954985201358795,
      "learning_rate": 3.4846441772488706e-05,
      "loss": 1.8512,
      "step": 13010
    },
    {
      "epoch": 0.3804426981895778,
      "grad_norm": 0.2998946011066437,
      "learning_rate": 3.476468102618564e-05,
      "loss": 1.8632,
      "step": 13020
    },
    {
      "epoch": 0.38098463935509,
      "grad_norm": 0.3708723187446594,
      "learning_rate": 3.4683003924933823e-05,
      "loss": 1.8694,
      "step": 13030
    },
    {
      "epoch": 0.38152658052060223,
      "grad_norm": 0.48674148321151733,
      "learning_rate": 3.4601410806355055e-05,
      "loss": 1.864,
      "step": 13040
    },
    {
      "epoch": 0.38206852168611444,
      "grad_norm": 0.35994434356689453,
      "learning_rate": 3.4519902007724026e-05,
      "loss": 1.8574,
      "step": 13050
    },
    {
      "epoch": 0.38206852168611444,
      "eval_loss": 2.6076242923736572,
      "eval_runtime": 21.9664,
      "eval_samples_per_second": 227.621,
      "eval_steps_per_second": 1.229,
      "step": 13050
    },
    {
      "epoch": 0.38261046285162664,
      "grad_norm": 0.30608904361724854,
      "learning_rate": 3.443847786596682e-05,
      "loss": 1.8665,
      "step": 13060
    },
    {
      "epoch": 0.3831524040171389,
      "grad_norm": 0.2905879616737366,
      "learning_rate": 3.435713871765969e-05,
      "loss": 1.8563,
      "step": 13070
    },
    {
      "epoch": 0.3836943451826511,
      "grad_norm": 0.34675753116607666,
      "learning_rate": 3.427588489902748e-05,
      "loss": 1.8582,
      "step": 13080
    },
    {
      "epoch": 0.3842362863481633,
      "grad_norm": 0.2855149209499359,
      "learning_rate": 3.419471674594226e-05,
      "loss": 1.8635,
      "step": 13090
    },
    {
      "epoch": 0.3847782275136755,
      "grad_norm": 0.3102046847343445,
      "learning_rate": 3.4113634593922126e-05,
      "loss": 1.8736,
      "step": 13100
    },
    {
      "epoch": 0.3853201686791878,
      "grad_norm": 0.34414398670196533,
      "learning_rate": 3.4032638778129576e-05,
      "loss": 1.8504,
      "step": 13110
    },
    {
      "epoch": 0.3858621098447,
      "grad_norm": 0.34666258096694946,
      "learning_rate": 3.395172963337029e-05,
      "loss": 1.8708,
      "step": 13120
    },
    {
      "epoch": 0.3864040510102122,
      "grad_norm": 0.2988355755805969,
      "learning_rate": 3.387090749409167e-05,
      "loss": 1.8635,
      "step": 13130
    },
    {
      "epoch": 0.38678340982607073,
      "eval_loss": 2.610823631286621,
      "eval_runtime": 21.9709,
      "eval_samples_per_second": 227.574,
      "eval_steps_per_second": 1.229,
      "step": 13137
    },
    {
      "epoch": 0.3869459921757244,
      "grad_norm": 0.2701932489871979,
      "learning_rate": 3.3790172694381385e-05,
      "loss": 1.8619,
      "step": 13140
    },
    {
      "epoch": 0.38748793334123666,
      "grad_norm": 0.31789588928222656,
      "learning_rate": 3.370952556796621e-05,
      "loss": 1.8527,
      "step": 13150
    },
    {
      "epoch": 0.38802987450674886,
      "grad_norm": 0.3052840232849121,
      "learning_rate": 3.362896644821042e-05,
      "loss": 1.8559,
      "step": 13160
    },
    {
      "epoch": 0.38857181567226107,
      "grad_norm": 0.2687411606311798,
      "learning_rate": 3.3548495668114536e-05,
      "loss": 1.8587,
      "step": 13170
    },
    {
      "epoch": 0.3891137568377733,
      "grad_norm": 0.5188722014427185,
      "learning_rate": 3.346811356031394e-05,
      "loss": 1.8567,
      "step": 13180
    },
    {
      "epoch": 0.38965569800328553,
      "grad_norm": 0.33732667565345764,
      "learning_rate": 3.3387820457077403e-05,
      "loss": 1.8539,
      "step": 13190
    },
    {
      "epoch": 0.39019763916879774,
      "grad_norm": 0.31062641739845276,
      "learning_rate": 3.3307616690305875e-05,
      "loss": 1.8569,
      "step": 13200
    },
    {
      "epoch": 0.39073958033430994,
      "grad_norm": 0.30725353956222534,
      "learning_rate": 3.322750259153096e-05,
      "loss": 1.8489,
      "step": 13210
    },
    {
      "epoch": 0.39128152149982215,
      "grad_norm": 0.3953590989112854,
      "learning_rate": 3.314747849191362e-05,
      "loss": 1.8627,
      "step": 13220
    },
    {
      "epoch": 0.3914982979660271,
      "eval_loss": 2.6103906631469727,
      "eval_runtime": 21.9715,
      "eval_samples_per_second": 227.567,
      "eval_steps_per_second": 1.229,
      "step": 13224
    },
    {
      "epoch": 0.3918234626653344,
      "grad_norm": 0.284915953874588,
      "learning_rate": 3.306754472224285e-05,
      "loss": 1.8606,
      "step": 13230
    },
    {
      "epoch": 0.3923654038308466,
      "grad_norm": 0.5332658886909485,
      "learning_rate": 3.2987701612934174e-05,
      "loss": 1.8526,
      "step": 13240
    },
    {
      "epoch": 0.3929073449963588,
      "grad_norm": 0.330483078956604,
      "learning_rate": 3.290794949402837e-05,
      "loss": 1.8603,
      "step": 13250
    },
    {
      "epoch": 0.393449286161871,
      "grad_norm": 0.4462834298610687,
      "learning_rate": 3.282828869519019e-05,
      "loss": 1.8646,
      "step": 13260
    },
    {
      "epoch": 0.3939912273273833,
      "grad_norm": 0.3231438100337982,
      "learning_rate": 3.2748719545706776e-05,
      "loss": 1.8528,
      "step": 13270
    },
    {
      "epoch": 0.3945331684928955,
      "grad_norm": 0.2627573311328888,
      "learning_rate": 3.266924237448655e-05,
      "loss": 1.857,
      "step": 13280
    },
    {
      "epoch": 0.3950751096584077,
      "grad_norm": 0.2972419261932373,
      "learning_rate": 3.2589857510057634e-05,
      "loss": 1.8477,
      "step": 13290
    },
    {
      "epoch": 0.3956170508239199,
      "grad_norm": 0.3084324598312378,
      "learning_rate": 3.251056528056658e-05,
      "loss": 1.8504,
      "step": 13300
    },
    {
      "epoch": 0.39615899198943216,
      "grad_norm": 0.3726852536201477,
      "learning_rate": 3.2431366013777156e-05,
      "loss": 1.8487,
      "step": 13310
    },
    {
      "epoch": 0.3962131861059834,
      "eval_loss": 2.607210397720337,
      "eval_runtime": 21.9736,
      "eval_samples_per_second": 227.546,
      "eval_steps_per_second": 1.229,
      "step": 13311
    },
    {
      "epoch": 0.39670093315494437,
      "grad_norm": 0.3420845568180084,
      "learning_rate": 3.235226003706872e-05,
      "loss": 1.8553,
      "step": 13320
    },
    {
      "epoch": 0.3972428743204566,
      "grad_norm": 0.3318046033382416,
      "learning_rate": 3.227324767743507e-05,
      "loss": 1.8592,
      "step": 13330
    },
    {
      "epoch": 0.3977848154859688,
      "grad_norm": 0.2706853449344635,
      "learning_rate": 3.2194329261482985e-05,
      "loss": 1.8574,
      "step": 13340
    },
    {
      "epoch": 0.39832675665148104,
      "grad_norm": 0.27932029962539673,
      "learning_rate": 3.211550511543095e-05,
      "loss": 1.8565,
      "step": 13350
    },
    {
      "epoch": 0.39886869781699325,
      "grad_norm": 0.29047316312789917,
      "learning_rate": 3.203677556510779e-05,
      "loss": 1.859,
      "step": 13360
    },
    {
      "epoch": 0.39941063898250545,
      "grad_norm": 0.3240620791912079,
      "learning_rate": 3.195814093595127e-05,
      "loss": 1.8506,
      "step": 13370
    },
    {
      "epoch": 0.39995258014801766,
      "grad_norm": 0.2864803969860077,
      "learning_rate": 3.18796015530068e-05,
      "loss": 1.8585,
      "step": 13380
    },
    {
      "epoch": 0.4004945213135299,
      "grad_norm": 0.28760528564453125,
      "learning_rate": 3.180115774092609e-05,
      "loss": 1.857,
      "step": 13390
    },
    {
      "epoch": 0.4009280742459397,
      "eval_loss": 2.60837984085083,
      "eval_runtime": 21.9751,
      "eval_samples_per_second": 227.53,
      "eval_steps_per_second": 1.229,
      "step": 13398
    },
    {
      "epoch": 0.4010364624790421,
      "grad_norm": 0.47491455078125,
      "learning_rate": 3.172280982396577e-05,
      "loss": 1.8524,
      "step": 13400
    },
    {
      "epoch": 0.4015784036445543,
      "grad_norm": 0.7039004564285278,
      "learning_rate": 3.164455812598609e-05,
      "loss": 1.8448,
      "step": 13410
    },
    {
      "epoch": 0.40212034481006653,
      "grad_norm": 0.5090378522872925,
      "learning_rate": 3.15664029704496e-05,
      "loss": 1.8595,
      "step": 13420
    },
    {
      "epoch": 0.4026622859755788,
      "grad_norm": 0.3467673361301422,
      "learning_rate": 3.148834468041973e-05,
      "loss": 1.8602,
      "step": 13430
    },
    {
      "epoch": 0.403204227141091,
      "grad_norm": 0.3230932056903839,
      "learning_rate": 3.141038357855953e-05,
      "loss": 1.8451,
      "step": 13440
    },
    {
      "epoch": 0.4037461683066032,
      "grad_norm": 0.3669589161872864,
      "learning_rate": 3.133251998713032e-05,
      "loss": 1.8655,
      "step": 13450
    },
    {
      "epoch": 0.4042881094721154,
      "grad_norm": 0.29886960983276367,
      "learning_rate": 3.1254754227990294e-05,
      "loss": 1.8554,
      "step": 13460
    },
    {
      "epoch": 0.40483005063762767,
      "grad_norm": 0.3277972638607025,
      "learning_rate": 3.1177086622593345e-05,
      "loss": 1.8566,
      "step": 13470
    },
    {
      "epoch": 0.4053719918031399,
      "grad_norm": 0.2972913980484009,
      "learning_rate": 3.109951749198755e-05,
      "loss": 1.8554,
      "step": 13480
    },
    {
      "epoch": 0.405642962385896,
      "eval_loss": 2.604696273803711,
      "eval_runtime": 22.1189,
      "eval_samples_per_second": 226.051,
      "eval_steps_per_second": 1.221,
      "step": 13485
    },
    {
      "epoch": 0.4059139329686521,
      "grad_norm": 0.41642656922340393,
      "learning_rate": 3.102204715681397e-05,
      "loss": 1.8526,
      "step": 13490
    },
    {
      "epoch": 0.4064558741341643,
      "grad_norm": 0.3331157863140106,
      "learning_rate": 3.0944675937305254e-05,
      "loss": 1.8621,
      "step": 13500
    },
    {
      "epoch": 0.40699781529967655,
      "grad_norm": 0.2806148827075958,
      "learning_rate": 3.086740415328436e-05,
      "loss": 1.8597,
      "step": 13510
    },
    {
      "epoch": 0.40753975646518875,
      "grad_norm": 0.328106552362442,
      "learning_rate": 3.0790232124163256e-05,
      "loss": 1.8543,
      "step": 13520
    },
    {
      "epoch": 0.40808169763070096,
      "grad_norm": 0.27871695160865784,
      "learning_rate": 3.0713160168941494e-05,
      "loss": 1.8546,
      "step": 13530
    },
    {
      "epoch": 0.40862363879621316,
      "grad_norm": 0.3508320748806,
      "learning_rate": 3.0636188606205e-05,
      "loss": 1.8454,
      "step": 13540
    },
    {
      "epoch": 0.4091655799617254,
      "grad_norm": 0.29558315873146057,
      "learning_rate": 3.0559317754124706e-05,
      "loss": 1.8595,
      "step": 13550
    },
    {
      "epoch": 0.40970752112723763,
      "grad_norm": 0.31710633635520935,
      "learning_rate": 3.048254793045524e-05,
      "loss": 1.8416,
      "step": 13560
    },
    {
      "epoch": 0.41024946229274983,
      "grad_norm": 0.31882941722869873,
      "learning_rate": 3.040587945253362e-05,
      "loss": 1.8512,
      "step": 13570
    },
    {
      "epoch": 0.4103578505258523,
      "eval_loss": 2.610496759414673,
      "eval_runtime": 21.9645,
      "eval_samples_per_second": 227.64,
      "eval_steps_per_second": 1.229,
      "step": 13572
    },
    {
      "epoch": 0.41079140345826204,
      "grad_norm": 0.28335651755332947,
      "learning_rate": 3.032931263727796e-05,
      "loss": 1.852,
      "step": 13580
    },
    {
      "epoch": 0.4113333446237743,
      "grad_norm": 0.40349459648132324,
      "learning_rate": 3.0252847801186135e-05,
      "loss": 1.8441,
      "step": 13590
    },
    {
      "epoch": 0.4118752857892865,
      "grad_norm": 0.2962387204170227,
      "learning_rate": 3.0176485260334398e-05,
      "loss": 1.8472,
      "step": 13600
    },
    {
      "epoch": 0.4124172269547987,
      "grad_norm": 0.27455824613571167,
      "learning_rate": 3.0100225330376282e-05,
      "loss": 1.848,
      "step": 13610
    },
    {
      "epoch": 0.4129591681203109,
      "grad_norm": 0.524045467376709,
      "learning_rate": 3.0024068326541056e-05,
      "loss": 1.8442,
      "step": 13620
    },
    {
      "epoch": 0.4135011092858232,
      "grad_norm": 0.28335145115852356,
      "learning_rate": 2.994801456363263e-05,
      "loss": 1.8395,
      "step": 13630
    },
    {
      "epoch": 0.4140430504513354,
      "grad_norm": 0.2653671205043793,
      "learning_rate": 2.987206435602809e-05,
      "loss": 1.8421,
      "step": 13640
    },
    {
      "epoch": 0.4145849916168476,
      "grad_norm": 0.37382206320762634,
      "learning_rate": 2.979621801767643e-05,
      "loss": 1.847,
      "step": 13650
    },
    {
      "epoch": 0.41507273866580857,
      "eval_loss": 2.604285717010498,
      "eval_runtime": 21.965,
      "eval_samples_per_second": 227.635,
      "eval_steps_per_second": 1.229,
      "step": 13659
    },
    {
      "epoch": 0.4151269327823598,
      "grad_norm": 0.2956138849258423,
      "learning_rate": 2.972047586209739e-05,
      "loss": 1.8429,
      "step": 13660
    },
    {
      "epoch": 0.41566887394787205,
      "grad_norm": 0.3020017147064209,
      "learning_rate": 2.9644838202379988e-05,
      "loss": 1.8395,
      "step": 13670
    },
    {
      "epoch": 0.41621081511338426,
      "grad_norm": 0.36557555198669434,
      "learning_rate": 2.956930535118129e-05,
      "loss": 1.8314,
      "step": 13680
    },
    {
      "epoch": 0.41675275627889646,
      "grad_norm": 0.2863655686378479,
      "learning_rate": 2.9493877620725208e-05,
      "loss": 1.8437,
      "step": 13690
    },
    {
      "epoch": 0.41729469744440867,
      "grad_norm": 0.30269670486450195,
      "learning_rate": 2.9418555322800983e-05,
      "loss": 1.8538,
      "step": 13700
    },
    {
      "epoch": 0.41783663860992093,
      "grad_norm": 0.3189536929130554,
      "learning_rate": 2.9343338768762175e-05,
      "loss": 1.832,
      "step": 13710
    },
    {
      "epoch": 0.41837857977543313,
      "grad_norm": 0.3305031657218933,
      "learning_rate": 2.9268228269525178e-05,
      "loss": 1.8469,
      "step": 13720
    },
    {
      "epoch": 0.41892052094094534,
      "grad_norm": 0.30250081419944763,
      "learning_rate": 2.9193224135567965e-05,
      "loss": 1.8563,
      "step": 13730
    },
    {
      "epoch": 0.41946246210645755,
      "grad_norm": 0.2988894581794739,
      "learning_rate": 2.9118326676928938e-05,
      "loss": 1.8563,
      "step": 13740
    },
    {
      "epoch": 0.4197876268057649,
      "eval_loss": 2.606349229812622,
      "eval_runtime": 21.9714,
      "eval_samples_per_second": 227.568,
      "eval_steps_per_second": 1.229,
      "step": 13746
    },
    {
      "epoch": 0.4200044032719698,
      "grad_norm": 0.32497337460517883,
      "learning_rate": 2.904353620320542e-05,
      "loss": 1.8507,
      "step": 13750
    },
    {
      "epoch": 0.420546344437482,
      "grad_norm": 0.4472458064556122,
      "learning_rate": 2.8968853023552555e-05,
      "loss": 1.8343,
      "step": 13760
    },
    {
      "epoch": 0.4210882856029942,
      "grad_norm": 0.4618549942970276,
      "learning_rate": 2.8894277446682028e-05,
      "loss": 1.8497,
      "step": 13770
    },
    {
      "epoch": 0.4216302267685064,
      "grad_norm": 0.3211345970630646,
      "learning_rate": 2.8819809780860625e-05,
      "loss": 1.8459,
      "step": 13780
    },
    {
      "epoch": 0.4221721679340187,
      "grad_norm": 0.2894679307937622,
      "learning_rate": 2.87454503339092e-05,
      "loss": 1.8519,
      "step": 13790
    },
    {
      "epoch": 0.4227141090995309,
      "grad_norm": 0.2763760983943939,
      "learning_rate": 2.867119941320114e-05,
      "loss": 1.8481,
      "step": 13800
    },
    {
      "epoch": 0.4232560502650431,
      "grad_norm": 0.318680077791214,
      "learning_rate": 2.859705732566129e-05,
      "loss": 1.8437,
      "step": 13810
    },
    {
      "epoch": 0.4237979914305553,
      "grad_norm": 0.39189639687538147,
      "learning_rate": 2.852302437776465e-05,
      "loss": 1.8552,
      "step": 13820
    },
    {
      "epoch": 0.42433993259606756,
      "grad_norm": 0.45934727787971497,
      "learning_rate": 2.844910087553503e-05,
      "loss": 1.8426,
      "step": 13830
    },
    {
      "epoch": 0.4245025149457212,
      "eval_loss": 2.608975648880005,
      "eval_runtime": 21.9733,
      "eval_samples_per_second": 227.548,
      "eval_steps_per_second": 1.229,
      "step": 13833
    },
    {
      "epoch": 0.42488187376157976,
      "grad_norm": 0.2927630543708801,
      "learning_rate": 2.8375287124543835e-05,
      "loss": 1.855,
      "step": 13840
    },
    {
      "epoch": 0.42542381492709197,
      "grad_norm": 0.27992621064186096,
      "learning_rate": 2.830158342990884e-05,
      "loss": 1.845,
      "step": 13850
    },
    {
      "epoch": 0.4259657560926042,
      "grad_norm": 0.33510464429855347,
      "learning_rate": 2.8227990096292827e-05,
      "loss": 1.8371,
      "step": 13860
    },
    {
      "epoch": 0.42650769725811644,
      "grad_norm": 0.4821888208389282,
      "learning_rate": 2.8154507427902467e-05,
      "loss": 1.8491,
      "step": 13870
    },
    {
      "epoch": 0.42704963842362864,
      "grad_norm": 0.29919731616973877,
      "learning_rate": 2.808113572848692e-05,
      "loss": 1.8429,
      "step": 13880
    },
    {
      "epoch": 0.42759157958914085,
      "grad_norm": 0.31553396582603455,
      "learning_rate": 2.8007875301336662e-05,
      "loss": 1.8436,
      "step": 13890
    },
    {
      "epoch": 0.42813352075465305,
      "grad_norm": 0.30432793498039246,
      "learning_rate": 2.7934726449282213e-05,
      "loss": 1.8395,
      "step": 13900
    },
    {
      "epoch": 0.4286754619201653,
      "grad_norm": 0.3555530905723572,
      "learning_rate": 2.7861689474692898e-05,
      "loss": 1.842,
      "step": 13910
    },
    {
      "epoch": 0.4292174030856775,
      "grad_norm": 0.32901936769485474,
      "learning_rate": 2.7788764679475538e-05,
      "loss": 1.8336,
      "step": 13920
    },
    {
      "epoch": 0.4292174030856775,
      "eval_loss": 2.611562728881836,
      "eval_runtime": 21.9692,
      "eval_samples_per_second": 227.591,
      "eval_steps_per_second": 1.229,
      "step": 13920
    },
    {
      "epoch": 0.4297593442511897,
      "grad_norm": 0.5692172646522522,
      "learning_rate": 2.7715952365073324e-05,
      "loss": 1.8485,
      "step": 13930
    },
    {
      "epoch": 0.43030128541670193,
      "grad_norm": 0.28073015809059143,
      "learning_rate": 2.7643252832464423e-05,
      "loss": 1.8312,
      "step": 13940
    },
    {
      "epoch": 0.4308432265822142,
      "grad_norm": 0.4343844950199127,
      "learning_rate": 2.7570666382160843e-05,
      "loss": 1.8512,
      "step": 13950
    },
    {
      "epoch": 0.4313851677477264,
      "grad_norm": 0.26866263151168823,
      "learning_rate": 2.7498193314207137e-05,
      "loss": 1.8367,
      "step": 13960
    },
    {
      "epoch": 0.4319271089132386,
      "grad_norm": 0.5220692753791809,
      "learning_rate": 2.742583392817918e-05,
      "loss": 1.8317,
      "step": 13970
    },
    {
      "epoch": 0.4324690500787508,
      "grad_norm": 0.3568936288356781,
      "learning_rate": 2.7353588523182943e-05,
      "loss": 1.8356,
      "step": 13980
    },
    {
      "epoch": 0.43301099124426307,
      "grad_norm": 0.28348490595817566,
      "learning_rate": 2.7281457397853237e-05,
      "loss": 1.8416,
      "step": 13990
    },
    {
      "epoch": 0.43355293240977527,
      "grad_norm": 0.32432541251182556,
      "learning_rate": 2.720944085035248e-05,
      "loss": 1.8346,
      "step": 14000
    },
    {
      "epoch": 0.4339322912256338,
      "eval_loss": 2.6060945987701416,
      "eval_runtime": 21.9781,
      "eval_samples_per_second": 227.499,
      "eval_steps_per_second": 1.228,
      "step": 14007
    },
    {
      "epoch": 0.4340948735752875,
      "grad_norm": 0.30824974179267883,
      "learning_rate": 2.7137539178369464e-05,
      "loss": 1.8423,
      "step": 14010
    },
    {
      "epoch": 0.4346368147407997,
      "grad_norm": 0.387539803981781,
      "learning_rate": 2.7065752679118128e-05,
      "loss": 1.8532,
      "step": 14020
    },
    {
      "epoch": 0.43517875590631194,
      "grad_norm": 0.3984427750110626,
      "learning_rate": 2.6994081649336366e-05,
      "loss": 1.8505,
      "step": 14030
    },
    {
      "epoch": 0.43572069707182415,
      "grad_norm": 0.39461550116539,
      "learning_rate": 2.6922526385284737e-05,
      "loss": 1.8425,
      "step": 14040
    },
    {
      "epoch": 0.43626263823733635,
      "grad_norm": 0.26579055190086365,
      "learning_rate": 2.685108718274525e-05,
      "loss": 1.8379,
      "step": 14050
    },
    {
      "epoch": 0.43680457940284856,
      "grad_norm": 0.3582479655742645,
      "learning_rate": 2.6779764337020195e-05,
      "loss": 1.8338,
      "step": 14060
    },
    {
      "epoch": 0.4373465205683608,
      "grad_norm": 0.4719529151916504,
      "learning_rate": 2.6708558142930862e-05,
      "loss": 1.8382,
      "step": 14070
    },
    {
      "epoch": 0.437888461733873,
      "grad_norm": 0.3371265232563019,
      "learning_rate": 2.6637468894816366e-05,
      "loss": 1.844,
      "step": 14080
    },
    {
      "epoch": 0.43843040289938523,
      "grad_norm": 0.27364712953567505,
      "learning_rate": 2.656649688653242e-05,
      "loss": 1.8391,
      "step": 14090
    },
    {
      "epoch": 0.4386471793655901,
      "eval_loss": 2.606126308441162,
      "eval_runtime": 21.9732,
      "eval_samples_per_second": 227.549,
      "eval_steps_per_second": 1.229,
      "step": 14094
    },
    {
      "epoch": 0.43897234406489744,
      "grad_norm": 0.3573056757450104,
      "learning_rate": 2.6495642411450082e-05,
      "loss": 1.8407,
      "step": 14100
    },
    {
      "epoch": 0.4395142852304097,
      "grad_norm": 0.26993247866630554,
      "learning_rate": 2.64249057624546e-05,
      "loss": 1.8321,
      "step": 14110
    },
    {
      "epoch": 0.4400562263959219,
      "grad_norm": 0.48967689275741577,
      "learning_rate": 2.6354287231944154e-05,
      "loss": 1.8435,
      "step": 14120
    },
    {
      "epoch": 0.4405981675614341,
      "grad_norm": 0.3284403383731842,
      "learning_rate": 2.6283787111828666e-05,
      "loss": 1.8477,
      "step": 14130
    },
    {
      "epoch": 0.4411401087269463,
      "grad_norm": 0.33071354031562805,
      "learning_rate": 2.6213405693528638e-05,
      "loss": 1.8433,
      "step": 14140
    },
    {
      "epoch": 0.4416820498924586,
      "grad_norm": 0.2832528352737427,
      "learning_rate": 2.6143143267973846e-05,
      "loss": 1.846,
      "step": 14150
    },
    {
      "epoch": 0.4422239910579708,
      "grad_norm": 0.3607860207557678,
      "learning_rate": 2.6073000125602236e-05,
      "loss": 1.8488,
      "step": 14160
    },
    {
      "epoch": 0.442765932223483,
      "grad_norm": 0.26745522022247314,
      "learning_rate": 2.600297655635866e-05,
      "loss": 1.8377,
      "step": 14170
    },
    {
      "epoch": 0.4433078733889952,
      "grad_norm": 0.28432849049568176,
      "learning_rate": 2.5933072849693706e-05,
      "loss": 1.8432,
      "step": 14180
    },
    {
      "epoch": 0.4433620675055464,
      "eval_loss": 2.6039960384368896,
      "eval_runtime": 21.9648,
      "eval_samples_per_second": 227.637,
      "eval_steps_per_second": 1.229,
      "step": 14181
    },
    {
      "epoch": 0.44384981455450745,
      "grad_norm": 0.274472177028656,
      "learning_rate": 2.5863289294562497e-05,
      "loss": 1.8296,
      "step": 14190
    },
    {
      "epoch": 0.44439175572001965,
      "grad_norm": 0.30612727999687195,
      "learning_rate": 2.5793626179423514e-05,
      "loss": 1.8461,
      "step": 14200
    },
    {
      "epoch": 0.44493369688553186,
      "grad_norm": 0.2670454978942871,
      "learning_rate": 2.5724083792237363e-05,
      "loss": 1.8464,
      "step": 14210
    },
    {
      "epoch": 0.44547563805104406,
      "grad_norm": 0.3071325719356537,
      "learning_rate": 2.5654662420465613e-05,
      "loss": 1.8413,
      "step": 14220
    },
    {
      "epoch": 0.4460175792165563,
      "grad_norm": 0.35536590218544006,
      "learning_rate": 2.5585362351069586e-05,
      "loss": 1.8479,
      "step": 14230
    },
    {
      "epoch": 0.44655952038206853,
      "grad_norm": 0.3007301390171051,
      "learning_rate": 2.5516183870509212e-05,
      "loss": 1.8459,
      "step": 14240
    },
    {
      "epoch": 0.44710146154758074,
      "grad_norm": 0.3410910665988922,
      "learning_rate": 2.544712726474182e-05,
      "loss": 1.8482,
      "step": 14250
    },
    {
      "epoch": 0.44764340271309294,
      "grad_norm": 0.3722754418849945,
      "learning_rate": 2.5378192819220954e-05,
      "loss": 1.8384,
      "step": 14260
    },
    {
      "epoch": 0.44807695564550276,
      "eval_loss": 2.605301856994629,
      "eval_runtime": 22.0831,
      "eval_samples_per_second": 226.417,
      "eval_steps_per_second": 1.223,
      "step": 14268
    },
    {
      "epoch": 0.4481853438786052,
      "grad_norm": 0.27585405111312866,
      "learning_rate": 2.5309380818895133e-05,
      "loss": 1.8416,
      "step": 14270
    },
    {
      "epoch": 0.4487272850441174,
      "grad_norm": 0.2805611193180084,
      "learning_rate": 2.524069154820684e-05,
      "loss": 1.8404,
      "step": 14280
    },
    {
      "epoch": 0.4492692262096296,
      "grad_norm": 0.4039418697357178,
      "learning_rate": 2.5172125291091147e-05,
      "loss": 1.8429,
      "step": 14290
    },
    {
      "epoch": 0.4498111673751418,
      "grad_norm": 0.27978357672691345,
      "learning_rate": 2.510368233097472e-05,
      "loss": 1.8414,
      "step": 14300
    },
    {
      "epoch": 0.4503531085406541,
      "grad_norm": 0.2985621690750122,
      "learning_rate": 2.5035362950774504e-05,
      "loss": 1.8439,
      "step": 14310
    },
    {
      "epoch": 0.4508950497061663,
      "grad_norm": 0.28868043422698975,
      "learning_rate": 2.496716743289659e-05,
      "loss": 1.8448,
      "step": 14320
    },
    {
      "epoch": 0.4514369908716785,
      "grad_norm": 0.3035377860069275,
      "learning_rate": 2.4899096059235144e-05,
      "loss": 1.8442,
      "step": 14330
    },
    {
      "epoch": 0.4519789320371907,
      "grad_norm": 0.29208043217658997,
      "learning_rate": 2.4831149111171117e-05,
      "loss": 1.8354,
      "step": 14340
    },
    {
      "epoch": 0.45252087320270296,
      "grad_norm": 0.36014124751091003,
      "learning_rate": 2.476332686957113e-05,
      "loss": 1.8325,
      "step": 14350
    },
    {
      "epoch": 0.45279184378545906,
      "eval_loss": 2.5989575386047363,
      "eval_runtime": 21.9695,
      "eval_samples_per_second": 227.588,
      "eval_steps_per_second": 1.229,
      "step": 14355
    },
    {
      "epoch": 0.45306281436821516,
      "grad_norm": 0.4234754741191864,
      "learning_rate": 2.4695629614786373e-05,
      "loss": 1.8427,
      "step": 14360
    },
    {
      "epoch": 0.45360475553372737,
      "grad_norm": 0.2819061577320099,
      "learning_rate": 2.462805762665128e-05,
      "loss": 1.8368,
      "step": 14370
    },
    {
      "epoch": 0.45414669669923957,
      "grad_norm": 0.4023802876472473,
      "learning_rate": 2.4560611184482604e-05,
      "loss": 1.8363,
      "step": 14380
    },
    {
      "epoch": 0.45468863786475183,
      "grad_norm": 0.4065036475658417,
      "learning_rate": 2.4493290567078052e-05,
      "loss": 1.8344,
      "step": 14390
    },
    {
      "epoch": 0.45523057903026404,
      "grad_norm": 0.3405419886112213,
      "learning_rate": 2.442609605271524e-05,
      "loss": 1.8358,
      "step": 14400
    },
    {
      "epoch": 0.45577252019577624,
      "grad_norm": 0.3849346339702606,
      "learning_rate": 2.4359027919150578e-05,
      "loss": 1.8321,
      "step": 14410
    },
    {
      "epoch": 0.45631446136128845,
      "grad_norm": 0.27683576941490173,
      "learning_rate": 2.4292086443617964e-05,
      "loss": 1.8423,
      "step": 14420
    },
    {
      "epoch": 0.4568564025268007,
      "grad_norm": 0.30656400322914124,
      "learning_rate": 2.4225271902827808e-05,
      "loss": 1.8341,
      "step": 14430
    },
    {
      "epoch": 0.4573983436923129,
      "grad_norm": 0.46740400791168213,
      "learning_rate": 2.4158584572965827e-05,
      "loss": 1.8442,
      "step": 14440
    },
    {
      "epoch": 0.45750673192541536,
      "eval_loss": 2.6014182567596436,
      "eval_runtime": 21.98,
      "eval_samples_per_second": 227.479,
      "eval_steps_per_second": 1.228,
      "step": 14442
    },
    {
      "epoch": 0.4579402848578251,
      "grad_norm": 0.4761998951435089,
      "learning_rate": 2.4092024729691855e-05,
      "loss": 1.8453,
      "step": 14450
    },
    {
      "epoch": 0.4584822260233373,
      "grad_norm": 0.3128609359264374,
      "learning_rate": 2.4025592648138807e-05,
      "loss": 1.832,
      "step": 14460
    },
    {
      "epoch": 0.4590241671888496,
      "grad_norm": 0.31973832845687866,
      "learning_rate": 2.3959288602911398e-05,
      "loss": 1.8325,
      "step": 14470
    },
    {
      "epoch": 0.4595661083543618,
      "grad_norm": 0.2859000265598297,
      "learning_rate": 2.3893112868085134e-05,
      "loss": 1.8248,
      "step": 14480
    },
    {
      "epoch": 0.460108049519874,
      "grad_norm": 0.2959967255592346,
      "learning_rate": 2.382706571720516e-05,
      "loss": 1.8338,
      "step": 14490
    },
    {
      "epoch": 0.4606499906853862,
      "grad_norm": 0.26421812176704407,
      "learning_rate": 2.376114742328507e-05,
      "loss": 1.8357,
      "step": 14500
    },
    {
      "epoch": 0.46119193185089846,
      "grad_norm": 0.29020607471466064,
      "learning_rate": 2.3695358258805813e-05,
      "loss": 1.8357,
      "step": 14510
    },
    {
      "epoch": 0.46173387301641067,
      "grad_norm": 0.5411327481269836,
      "learning_rate": 2.3629698495714577e-05,
      "loss": 1.8333,
      "step": 14520
    },
    {
      "epoch": 0.46222162006537165,
      "eval_loss": 2.5997419357299805,
      "eval_runtime": 21.9696,
      "eval_samples_per_second": 227.587,
      "eval_steps_per_second": 1.229,
      "step": 14529
    },
    {
      "epoch": 0.4622758141819229,
      "grad_norm": 0.4464700222015381,
      "learning_rate": 2.356416840542364e-05,
      "loss": 1.8359,
      "step": 14530
    },
    {
      "epoch": 0.4628177553474351,
      "grad_norm": 0.37165167927742004,
      "learning_rate": 2.3498768258809296e-05,
      "loss": 1.8324,
      "step": 14540
    },
    {
      "epoch": 0.46335969651294734,
      "grad_norm": 0.3330148458480835,
      "learning_rate": 2.343349832621067e-05,
      "loss": 1.8297,
      "step": 14550
    },
    {
      "epoch": 0.46390163767845954,
      "grad_norm": 0.5441980361938477,
      "learning_rate": 2.3368358877428643e-05,
      "loss": 1.8247,
      "step": 14560
    },
    {
      "epoch": 0.46444357884397175,
      "grad_norm": 0.2869478464126587,
      "learning_rate": 2.3303350181724716e-05,
      "loss": 1.8367,
      "step": 14570
    },
    {
      "epoch": 0.46498552000948395,
      "grad_norm": 0.30734243988990784,
      "learning_rate": 2.3238472507819923e-05,
      "loss": 1.8368,
      "step": 14580
    },
    {
      "epoch": 0.4655274611749962,
      "grad_norm": 0.27206000685691833,
      "learning_rate": 2.3173726123893675e-05,
      "loss": 1.8442,
      "step": 14590
    },
    {
      "epoch": 0.4660694023405084,
      "grad_norm": 0.34638717770576477,
      "learning_rate": 2.3109111297582744e-05,
      "loss": 1.8407,
      "step": 14600
    },
    {
      "epoch": 0.4666113435060206,
      "grad_norm": 0.29559314250946045,
      "learning_rate": 2.3044628295980027e-05,
      "loss": 1.8357,
      "step": 14610
    },
    {
      "epoch": 0.46693650820532795,
      "eval_loss": 2.6009130477905273,
      "eval_runtime": 21.9702,
      "eval_samples_per_second": 227.581,
      "eval_steps_per_second": 1.229,
      "step": 14616
    },
    {
      "epoch": 0.46715328467153283,
      "grad_norm": 0.31058135628700256,
      "learning_rate": 2.2980277385633533e-05,
      "loss": 1.8294,
      "step": 14620
    },
    {
      "epoch": 0.4676952258370451,
      "grad_norm": 0.27876853942871094,
      "learning_rate": 2.291605883254525e-05,
      "loss": 1.8333,
      "step": 14630
    },
    {
      "epoch": 0.4682371670025573,
      "grad_norm": 0.3003843128681183,
      "learning_rate": 2.2851972902170053e-05,
      "loss": 1.8333,
      "step": 14640
    },
    {
      "epoch": 0.4687791081680695,
      "grad_norm": 0.4129401743412018,
      "learning_rate": 2.2788019859414646e-05,
      "loss": 1.8185,
      "step": 14650
    },
    {
      "epoch": 0.4693210493335817,
      "grad_norm": 0.2741876542568207,
      "learning_rate": 2.2724199968636357e-05,
      "loss": 1.8215,
      "step": 14660
    },
    {
      "epoch": 0.46986299049909397,
      "grad_norm": 0.34926167130470276,
      "learning_rate": 2.266051349364216e-05,
      "loss": 1.8269,
      "step": 14670
    },
    {
      "epoch": 0.4704049316646062,
      "grad_norm": 0.3961677849292755,
      "learning_rate": 2.2596960697687518e-05,
      "loss": 1.8275,
      "step": 14680
    },
    {
      "epoch": 0.4709468728301184,
      "grad_norm": 0.3206305503845215,
      "learning_rate": 2.2533541843475344e-05,
      "loss": 1.836,
      "step": 14690
    },
    {
      "epoch": 0.4714888139956306,
      "grad_norm": 0.349286824464798,
      "learning_rate": 2.247025719315484e-05,
      "loss": 1.8297,
      "step": 14700
    },
    {
      "epoch": 0.47165139634528425,
      "eval_loss": 2.6037356853485107,
      "eval_runtime": 22.0801,
      "eval_samples_per_second": 226.448,
      "eval_steps_per_second": 1.223,
      "step": 14703
    },
    {
      "epoch": 0.47203075516114285,
      "grad_norm": 0.3154524862766266,
      "learning_rate": 2.240710700832052e-05,
      "loss": 1.8323,
      "step": 14710
    },
    {
      "epoch": 0.47257269632665505,
      "grad_norm": 0.30774760246276855,
      "learning_rate": 2.2344091550011033e-05,
      "loss": 1.8456,
      "step": 14720
    },
    {
      "epoch": 0.47311463749216726,
      "grad_norm": 0.3301653265953064,
      "learning_rate": 2.228121107870812e-05,
      "loss": 1.8384,
      "step": 14730
    },
    {
      "epoch": 0.47365657865767946,
      "grad_norm": 0.30092665553092957,
      "learning_rate": 2.2218465854335535e-05,
      "loss": 1.8373,
      "step": 14740
    },
    {
      "epoch": 0.4741985198231917,
      "grad_norm": 0.28236475586891174,
      "learning_rate": 2.215585613625798e-05,
      "loss": 1.8184,
      "step": 14750
    },
    {
      "epoch": 0.4747404609887039,
      "grad_norm": 0.28573015332221985,
      "learning_rate": 2.209338218328006e-05,
      "loss": 1.841,
      "step": 14760
    },
    {
      "epoch": 0.47528240215421613,
      "grad_norm": 0.3064768314361572,
      "learning_rate": 2.2031044253645117e-05,
      "loss": 1.8275,
      "step": 14770
    },
    {
      "epoch": 0.47582434331972834,
      "grad_norm": 0.2896689772605896,
      "learning_rate": 2.1968842605034262e-05,
      "loss": 1.8326,
      "step": 14780
    },
    {
      "epoch": 0.4763662844852406,
      "grad_norm": 0.28397008776664734,
      "learning_rate": 2.190677749456526e-05,
      "loss": 1.8242,
      "step": 14790
    },
    {
      "epoch": 0.4763662844852406,
      "eval_loss": 2.599595785140991,
      "eval_runtime": 22.3123,
      "eval_samples_per_second": 224.092,
      "eval_steps_per_second": 1.21,
      "step": 14790
    },
    {
      "epoch": 0.4769082256507528,
      "grad_norm": 0.7356740236282349,
      "learning_rate": 2.1844849178791486e-05,
      "loss": 1.8301,
      "step": 14800
    },
    {
      "epoch": 0.477450166816265,
      "grad_norm": 0.40485429763793945,
      "learning_rate": 2.1783057913700865e-05,
      "loss": 1.825,
      "step": 14810
    },
    {
      "epoch": 0.4779921079817772,
      "grad_norm": 0.2894582748413086,
      "learning_rate": 2.17214039547148e-05,
      "loss": 1.8295,
      "step": 14820
    },
    {
      "epoch": 0.4785340491472894,
      "grad_norm": 0.28514474630355835,
      "learning_rate": 2.1659887556687102e-05,
      "loss": 1.8327,
      "step": 14830
    },
    {
      "epoch": 0.4790759903128017,
      "grad_norm": 0.3327902555465698,
      "learning_rate": 2.1598508973903004e-05,
      "loss": 1.8393,
      "step": 14840
    },
    {
      "epoch": 0.4796179314783139,
      "grad_norm": 0.28141072392463684,
      "learning_rate": 2.1537268460078018e-05,
      "loss": 1.8279,
      "step": 14850
    },
    {
      "epoch": 0.4801598726438261,
      "grad_norm": 0.5223854184150696,
      "learning_rate": 2.147616626835694e-05,
      "loss": 1.8246,
      "step": 14860
    },
    {
      "epoch": 0.4807018138093383,
      "grad_norm": 0.28995490074157715,
      "learning_rate": 2.141520265131284e-05,
      "loss": 1.8196,
      "step": 14870
    },
    {
      "epoch": 0.4810811726251969,
      "eval_loss": 2.6000826358795166,
      "eval_runtime": 21.9715,
      "eval_samples_per_second": 227.568,
      "eval_steps_per_second": 1.229,
      "step": 14877
    },
    {
      "epoch": 0.48124375497485056,
      "grad_norm": 0.3166589140892029,
      "learning_rate": 2.1354377860945925e-05,
      "loss": 1.829,
      "step": 14880
    },
    {
      "epoch": 0.48178569614036276,
      "grad_norm": 0.304440975189209,
      "learning_rate": 2.1293692148682553e-05,
      "loss": 1.8353,
      "step": 14890
    },
    {
      "epoch": 0.48232763730587497,
      "grad_norm": 0.3413919508457184,
      "learning_rate": 2.1233145765374202e-05,
      "loss": 1.8222,
      "step": 14900
    },
    {
      "epoch": 0.4828695784713872,
      "grad_norm": 0.31630587577819824,
      "learning_rate": 2.1172738961296396e-05,
      "loss": 1.8281,
      "step": 14910
    },
    {
      "epoch": 0.48341151963689943,
      "grad_norm": 0.3274076581001282,
      "learning_rate": 2.1112471986147723e-05,
      "loss": 1.8278,
      "step": 14920
    },
    {
      "epoch": 0.48395346080241164,
      "grad_norm": 0.3373514711856842,
      "learning_rate": 2.1052345089048765e-05,
      "loss": 1.818,
      "step": 14930
    },
    {
      "epoch": 0.48449540196792384,
      "grad_norm": 0.43474531173706055,
      "learning_rate": 2.0992358518541025e-05,
      "loss": 1.8377,
      "step": 14940
    },
    {
      "epoch": 0.48503734313343605,
      "grad_norm": 0.28481611609458923,
      "learning_rate": 2.093251252258602e-05,
      "loss": 1.8339,
      "step": 14950
    },
    {
      "epoch": 0.4855792842989483,
      "grad_norm": 0.38067948818206787,
      "learning_rate": 2.0872807348564134e-05,
      "loss": 1.8251,
      "step": 14960
    },
    {
      "epoch": 0.4857960607651532,
      "eval_loss": 2.606252431869507,
      "eval_runtime": 21.966,
      "eval_samples_per_second": 227.624,
      "eval_steps_per_second": 1.229,
      "step": 14964
    },
    {
      "epoch": 0.4861212254644605,
      "grad_norm": 0.30156803131103516,
      "learning_rate": 2.0813243243273694e-05,
      "loss": 1.832,
      "step": 14970
    },
    {
      "epoch": 0.4866631666299727,
      "grad_norm": 0.2849683463573456,
      "learning_rate": 2.075382045292987e-05,
      "loss": 1.8327,
      "step": 14980
    },
    {
      "epoch": 0.4872051077954849,
      "grad_norm": 0.2862844169139862,
      "learning_rate": 2.0694539223163674e-05,
      "loss": 1.8226,
      "step": 14990
    },
    {
      "epoch": 0.4877470489609972,
      "grad_norm": 0.3448987603187561,
      "learning_rate": 2.0635399799021005e-05,
      "loss": 1.827,
      "step": 15000
    },
    {
      "epoch": 0.000541941165512219,
      "grad_norm": 0.3310023248195648,
      "learning_rate": 2.0576402424961567e-05,
      "loss": 1.821,
      "step": 15010
    },
    {
      "epoch": 0.001083882331024438,
      "grad_norm": 0.2753497064113617,
      "learning_rate": 2.0517547344857874e-05,
      "loss": 1.8207,
      "step": 15020
    },
    {
      "epoch": 0.0016258234965366573,
      "grad_norm": 0.2612669765949249,
      "learning_rate": 2.04588348019943e-05,
      "loss": 1.8235,
      "step": 15030
    },
    {
      "epoch": 0.002167764662048876,
      "grad_norm": 0.2987896502017975,
      "learning_rate": 2.0400265039065938e-05,
      "loss": 1.8197,
      "step": 15040
    },
    {
      "epoch": 0.0027097058275610954,
      "grad_norm": 0.27388012409210205,
      "learning_rate": 2.0341838298177776e-05,
      "loss": 1.8277,
      "step": 15050
    },
    {
      "epoch": 0.002763899944112317,
      "eval_loss": 2.6046245098114014,
      "eval_runtime": 23.5773,
      "eval_samples_per_second": 212.068,
      "eval_steps_per_second": 1.145,
      "step": 15051
    },
    {
      "epoch": 0.0032516469930733145,
      "grad_norm": 0.3995117247104645,
      "learning_rate": 2.0283554820843547e-05,
      "loss": 1.8286,
      "step": 15060
    },
    {
      "epoch": 0.0037935881585855337,
      "grad_norm": 0.2957726716995239,
      "learning_rate": 2.022541484798479e-05,
      "loss": 1.8287,
      "step": 15070
    },
    {
      "epoch": 0.004335529324097752,
      "grad_norm": 0.2851552963256836,
      "learning_rate": 2.0167418619929908e-05,
      "loss": 1.8285,
      "step": 15080
    },
    {
      "epoch": 0.0048774704896099716,
      "grad_norm": 0.3845987021923065,
      "learning_rate": 2.010956637641303e-05,
      "loss": 1.8201,
      "step": 15090
    },
    {
      "epoch": 0.005419411655122191,
      "grad_norm": 0.2890310287475586,
      "learning_rate": 2.0051858356573155e-05,
      "loss": 1.8243,
      "step": 15100
    },
    {
      "epoch": 0.00596135282063441,
      "grad_norm": 0.27605000138282776,
      "learning_rate": 1.9994294798953134e-05,
      "loss": 1.8272,
      "step": 15110
    },
    {
      "epoch": 0.006503293986146629,
      "grad_norm": 0.39410722255706787,
      "learning_rate": 1.9936875941498646e-05,
      "loss": 1.8233,
      "step": 15120
    },
    {
      "epoch": 0.007045235151658848,
      "grad_norm": 0.37865981459617615,
      "learning_rate": 1.9879602021557226e-05,
      "loss": 1.8267,
      "step": 15130
    },
    {
      "epoch": 0.007478788084068624,
      "eval_loss": 2.604631185531616,
      "eval_runtime": 23.8764,
      "eval_samples_per_second": 209.412,
      "eval_steps_per_second": 1.131,
      "step": 15138
    },
    {
      "epoch": 0.007587176317171067,
      "grad_norm": 0.28183528780937195,
      "learning_rate": 1.98224732758773e-05,
      "loss": 1.8224,
      "step": 15140
    },
    {
      "epoch": 0.008129117482683286,
      "grad_norm": 0.2929738759994507,
      "learning_rate": 1.97654899406072e-05,
      "loss": 1.8245,
      "step": 15150
    },
    {
      "epoch": 0.008671058648195505,
      "grad_norm": 0.28582409024238586,
      "learning_rate": 1.9708652251294206e-05,
      "loss": 1.8136,
      "step": 15160
    },
    {
      "epoch": 0.009212999813707724,
      "grad_norm": 0.2901137173175812,
      "learning_rate": 1.9651960442883528e-05,
      "loss": 1.8254,
      "step": 15170
    },
    {
      "epoch": 0.009754940979219943,
      "grad_norm": 0.26637405157089233,
      "learning_rate": 1.9595414749717363e-05,
      "loss": 1.8244,
      "step": 15180
    },
    {
      "epoch": 0.010296882144732162,
      "grad_norm": 0.2925315201282501,
      "learning_rate": 1.9539015405533935e-05,
      "loss": 1.8162,
      "step": 15190
    },
    {
      "epoch": 0.010838823310244381,
      "grad_norm": 0.3221004009246826,
      "learning_rate": 1.9482762643466504e-05,
      "loss": 1.8219,
      "step": 15200
    },
    {
      "epoch": 0.0113807644757566,
      "grad_norm": 0.3065180480480194,
      "learning_rate": 1.9426656696042424e-05,
      "loss": 1.8196,
      "step": 15210
    },
    {
      "epoch": 0.01192270564126882,
      "grad_norm": 0.3092059791088104,
      "learning_rate": 1.9370697795182187e-05,
      "loss": 1.8217,
      "step": 15220
    },
    {
      "epoch": 0.01219367622402493,
      "eval_loss": 2.6005666255950928,
      "eval_runtime": 22.3369,
      "eval_samples_per_second": 223.845,
      "eval_steps_per_second": 1.209,
      "step": 15225
    },
    {
      "epoch": 0.012464646806781039,
      "grad_norm": 0.3684033155441284,
      "learning_rate": 1.9314886172198426e-05,
      "loss": 1.8316,
      "step": 15230
    },
    {
      "epoch": 0.013006587972293258,
      "grad_norm": 0.3613276779651642,
      "learning_rate": 1.9259222057794996e-05,
      "loss": 1.8332,
      "step": 15240
    },
    {
      "epoch": 0.013548529137805477,
      "grad_norm": 0.26722803711891174,
      "learning_rate": 1.9203705682066007e-05,
      "loss": 1.8249,
      "step": 15250
    },
    {
      "epoch": 0.014090470303317696,
      "grad_norm": 0.30705708265304565,
      "learning_rate": 1.9148337274494862e-05,
      "loss": 1.8228,
      "step": 15260
    },
    {
      "epoch": 0.014632411468829916,
      "grad_norm": 0.32335802912712097,
      "learning_rate": 1.9093117063953354e-05,
      "loss": 1.8176,
      "step": 15270
    },
    {
      "epoch": 0.015174352634342135,
      "grad_norm": 0.33310064673423767,
      "learning_rate": 1.9038045278700653e-05,
      "loss": 1.824,
      "step": 15280
    },
    {
      "epoch": 0.015716293799854352,
      "grad_norm": 0.27640849351882935,
      "learning_rate": 1.8983122146382406e-05,
      "loss": 1.8284,
      "step": 15290
    },
    {
      "epoch": 0.01625823496536657,
      "grad_norm": 0.27379941940307617,
      "learning_rate": 1.892834789402979e-05,
      "loss": 1.8175,
      "step": 15300
    },
    {
      "epoch": 0.01680017613087879,
      "grad_norm": 0.32370367646217346,
      "learning_rate": 1.8873722748058552e-05,
      "loss": 1.8272,
      "step": 15310
    },
    {
      "epoch": 0.016908564363981235,
      "eval_loss": 2.6025352478027344,
      "eval_runtime": 25.0337,
      "eval_samples_per_second": 199.731,
      "eval_steps_per_second": 1.079,
      "step": 15312
    },
    {
      "epoch": 0.01734211729639101,
      "grad_norm": 0.3098716139793396,
      "learning_rate": 1.8819246934268125e-05,
      "loss": 1.8247,
      "step": 15320
    },
    {
      "epoch": 0.01788405846190323,
      "grad_norm": 0.2977130115032196,
      "learning_rate": 1.8764920677840632e-05,
      "loss": 1.8229,
      "step": 15330
    },
    {
      "epoch": 0.018425999627415448,
      "grad_norm": 0.28470513224601746,
      "learning_rate": 1.871074420333999e-05,
      "loss": 1.817,
      "step": 15340
    },
    {
      "epoch": 0.018967940792927667,
      "grad_norm": 0.2921582758426666,
      "learning_rate": 1.8656717734710975e-05,
      "loss": 1.8392,
      "step": 15350
    },
    {
      "epoch": 0.019509881958439886,
      "grad_norm": 0.30813974142074585,
      "learning_rate": 1.8602841495278294e-05,
      "loss": 1.8216,
      "step": 15360
    },
    {
      "epoch": 0.020051823123952105,
      "grad_norm": 0.27432745695114136,
      "learning_rate": 1.8549115707745656e-05,
      "loss": 1.8257,
      "step": 15370
    },
    {
      "epoch": 0.020593764289464325,
      "grad_norm": 0.2718896269798279,
      "learning_rate": 1.84955405941949e-05,
      "loss": 1.8297,
      "step": 15380
    },
    {
      "epoch": 0.021135705454976544,
      "grad_norm": 0.2647911310195923,
      "learning_rate": 1.8442116376084985e-05,
      "loss": 1.8202,
      "step": 15390
    },
    {
      "epoch": 0.021623452503937542,
      "eval_loss": 2.602659225463867,
      "eval_runtime": 26.4511,
      "eval_samples_per_second": 189.028,
      "eval_steps_per_second": 1.021,
      "step": 15399
    },
    {
      "epoch": 0.021677646620488763,
      "grad_norm": 0.3035239279270172,
      "learning_rate": 1.8388843274251156e-05,
      "loss": 1.825,
      "step": 15400
    },
    {
      "epoch": 0.022219587786000982,
      "grad_norm": 0.31842127442359924,
      "learning_rate": 1.8335721508903987e-05,
      "loss": 1.8171,
      "step": 15410
    },
    {
      "epoch": 0.0227615289515132,
      "grad_norm": 0.3104911744594574,
      "learning_rate": 1.8282751299628486e-05,
      "loss": 1.8218,
      "step": 15420
    },
    {
      "epoch": 0.02330347011702542,
      "grad_norm": 0.3664672076702118,
      "learning_rate": 1.822993286538321e-05,
      "loss": 1.8127,
      "step": 15430
    },
    {
      "epoch": 0.02384541128253764,
      "grad_norm": 0.2692658007144928,
      "learning_rate": 1.8177266424499313e-05,
      "loss": 1.8264,
      "step": 15440
    },
    {
      "epoch": 0.02438735244804986,
      "grad_norm": 0.3049471378326416,
      "learning_rate": 1.812475219467966e-05,
      "loss": 1.8147,
      "step": 15450
    },
    {
      "epoch": 0.024929293613562078,
      "grad_norm": 0.32675981521606445,
      "learning_rate": 1.8072390392997954e-05,
      "loss": 1.8286,
      "step": 15460
    },
    {
      "epoch": 0.025471234779074297,
      "grad_norm": 0.2658865451812744,
      "learning_rate": 1.8020181235897797e-05,
      "loss": 1.8234,
      "step": 15470
    },
    {
      "epoch": 0.026013175944586516,
      "grad_norm": 0.28163376450538635,
      "learning_rate": 1.796812493919185e-05,
      "loss": 1.8217,
      "step": 15480
    },
    {
      "epoch": 0.026338340643893846,
      "eval_loss": 2.600201368331909,
      "eval_runtime": 22.1652,
      "eval_samples_per_second": 225.579,
      "eval_steps_per_second": 1.218,
      "step": 15486
    },
    {
      "epoch": 0.026555117110098735,
      "grad_norm": 0.32032302021980286,
      "learning_rate": 1.791622171806088e-05,
      "loss": 1.8177,
      "step": 15490
    },
    {
      "epoch": 0.027097058275610954,
      "grad_norm": 0.28977230191230774,
      "learning_rate": 1.78644717870529e-05,
      "loss": 1.8137,
      "step": 15500
    },
    {
      "epoch": 0.027638999441123174,
      "grad_norm": 0.332872599363327,
      "learning_rate": 1.781287536008229e-05,
      "loss": 1.8214,
      "step": 15510
    },
    {
      "epoch": 0.028180940606635393,
      "grad_norm": 0.27174341678619385,
      "learning_rate": 1.77614326504289e-05,
      "loss": 1.8277,
      "step": 15520
    },
    {
      "epoch": 0.028722881772147612,
      "grad_norm": 0.27996736764907837,
      "learning_rate": 1.771014387073715e-05,
      "loss": 1.8163,
      "step": 15530
    },
    {
      "epoch": 0.02926482293765983,
      "grad_norm": 0.26476001739501953,
      "learning_rate": 1.765900923301523e-05,
      "loss": 1.8268,
      "step": 15540
    },
    {
      "epoch": 0.02980676410317205,
      "grad_norm": 0.3071003556251526,
      "learning_rate": 1.760802894863412e-05,
      "loss": 1.8231,
      "step": 15550
    },
    {
      "epoch": 0.03034870526868427,
      "grad_norm": 0.31128978729248047,
      "learning_rate": 1.7557203228326737e-05,
      "loss": 1.8213,
      "step": 15560
    },
    {
      "epoch": 0.03089064643419649,
      "grad_norm": 0.27923986315727234,
      "learning_rate": 1.7506532282187166e-05,
      "loss": 1.8192,
      "step": 15570
    },
    {
      "epoch": 0.031053228783850154,
      "eval_loss": 2.6021640300750732,
      "eval_runtime": 22.1843,
      "eval_samples_per_second": 225.384,
      "eval_steps_per_second": 1.217,
      "step": 15573
    },
    {
      "epoch": 0.031432587599708704,
      "grad_norm": 0.324139267206192,
      "learning_rate": 1.7456016319669642e-05,
      "loss": 1.827,
      "step": 15580
    },
    {
      "epoch": 0.03197452876522092,
      "grad_norm": 0.27704742550849915,
      "learning_rate": 1.7405655549587816e-05,
      "loss": 1.8165,
      "step": 15590
    },
    {
      "epoch": 0.03251646993073314,
      "grad_norm": 0.3016372323036194,
      "learning_rate": 1.7355450180113808e-05,
      "loss": 1.8272,
      "step": 15600
    },
    {
      "epoch": 0.03305841109624536,
      "grad_norm": 0.2791915833950043,
      "learning_rate": 1.730540041877733e-05,
      "loss": 1.8166,
      "step": 15610
    },
    {
      "epoch": 0.03360035226175758,
      "grad_norm": 0.40646281838417053,
      "learning_rate": 1.7255506472464936e-05,
      "loss": 1.828,
      "step": 15620
    },
    {
      "epoch": 0.0341422934272698,
      "grad_norm": 0.32504037022590637,
      "learning_rate": 1.7205768547419077e-05,
      "loss": 1.8265,
      "step": 15630
    },
    {
      "epoch": 0.03468423459278202,
      "grad_norm": 0.2760820984840393,
      "learning_rate": 1.7156186849237244e-05,
      "loss": 1.8341,
      "step": 15640
    },
    {
      "epoch": 0.03522617575829424,
      "grad_norm": 0.2752606272697449,
      "learning_rate": 1.7106761582871205e-05,
      "loss": 1.8145,
      "step": 15650
    },
    {
      "epoch": 0.03576811692380646,
      "grad_norm": 0.3181304633617401,
      "learning_rate": 1.7057492952626025e-05,
      "loss": 1.8322,
      "step": 15660
    },
    {
      "epoch": 0.03576811692380646,
      "eval_loss": 2.6028239727020264,
      "eval_runtime": 21.9861,
      "eval_samples_per_second": 227.417,
      "eval_steps_per_second": 1.228,
      "step": 15660
    },
    {
      "epoch": 0.03631005808931868,
      "grad_norm": 0.2869409918785095,
      "learning_rate": 1.7008381162159358e-05,
      "loss": 1.8151,
      "step": 15670
    },
    {
      "epoch": 0.036851999254830896,
      "grad_norm": 0.30195239186286926,
      "learning_rate": 1.6959426414480516e-05,
      "loss": 1.8268,
      "step": 15680
    },
    {
      "epoch": 0.037393940420343115,
      "grad_norm": 0.2631888687610626,
      "learning_rate": 1.6910628911949644e-05,
      "loss": 1.8166,
      "step": 15690
    },
    {
      "epoch": 0.037935881585855334,
      "grad_norm": 0.29582157731056213,
      "learning_rate": 1.6861988856276946e-05,
      "loss": 1.8089,
      "step": 15700
    },
    {
      "epoch": 0.03847782275136755,
      "grad_norm": 0.27693384885787964,
      "learning_rate": 1.6813506448521727e-05,
      "loss": 1.812,
      "step": 15710
    },
    {
      "epoch": 0.03901976391687977,
      "grad_norm": 0.282247930765152,
      "learning_rate": 1.6765181889091675e-05,
      "loss": 1.819,
      "step": 15720
    },
    {
      "epoch": 0.03956170508239199,
      "grad_norm": 0.3580980896949768,
      "learning_rate": 1.671701537774202e-05,
      "loss": 1.8174,
      "step": 15730
    },
    {
      "epoch": 0.04010364624790421,
      "grad_norm": 0.2820577621459961,
      "learning_rate": 1.666900711357463e-05,
      "loss": 1.8327,
      "step": 15740
    },
    {
      "epoch": 0.04048300506376277,
      "eval_loss": 2.6001157760620117,
      "eval_runtime": 25.7544,
      "eval_samples_per_second": 194.141,
      "eval_steps_per_second": 1.048,
      "step": 15747
    },
    {
      "epoch": 0.04064558741341643,
      "grad_norm": 0.3680673837661743,
      "learning_rate": 1.6621157295037298e-05,
      "loss": 1.8087,
      "step": 15750
    },
    {
      "epoch": 0.04118752857892865,
      "grad_norm": 0.3296014964580536,
      "learning_rate": 1.65734661199228e-05,
      "loss": 1.8183,
      "step": 15760
    },
    {
      "epoch": 0.04172946974444087,
      "grad_norm": 0.29352930188179016,
      "learning_rate": 1.652593378536816e-05,
      "loss": 1.8075,
      "step": 15770
    },
    {
      "epoch": 0.04227141090995309,
      "grad_norm": 0.36422020196914673,
      "learning_rate": 1.6478560487853866e-05,
      "loss": 1.8151,
      "step": 15780
    },
    {
      "epoch": 0.04281335207546531,
      "grad_norm": 0.30587121844291687,
      "learning_rate": 1.6431346423202945e-05,
      "loss": 1.8174,
      "step": 15790
    },
    {
      "epoch": 0.043355293240977526,
      "grad_norm": 0.29669710993766785,
      "learning_rate": 1.6384291786580247e-05,
      "loss": 1.8186,
      "step": 15800
    },
    {
      "epoch": 0.043897234406489745,
      "grad_norm": 0.3331429362297058,
      "learning_rate": 1.633739677249159e-05,
      "loss": 1.8162,
      "step": 15810
    },
    {
      "epoch": 0.044439175572001964,
      "grad_norm": 0.3026489019393921,
      "learning_rate": 1.6290661574782995e-05,
      "loss": 1.8103,
      "step": 15820
    },
    {
      "epoch": 0.04498111673751418,
      "grad_norm": 0.29594072699546814,
      "learning_rate": 1.624408638663985e-05,
      "loss": 1.8055,
      "step": 15830
    },
    {
      "epoch": 0.04519789320371907,
      "eval_loss": 2.599421977996826,
      "eval_runtime": 23.9096,
      "eval_samples_per_second": 209.121,
      "eval_steps_per_second": 1.129,
      "step": 15834
    },
    {
      "epoch": 0.0455230579030264,
      "grad_norm": 0.286182165145874,
      "learning_rate": 1.619767140058614e-05,
      "loss": 1.8213,
      "step": 15840
    },
    {
      "epoch": 0.04606499906853862,
      "grad_norm": 0.3165307343006134,
      "learning_rate": 1.6151416808483603e-05,
      "loss": 1.816,
      "step": 15850
    },
    {
      "epoch": 0.04660694023405084,
      "grad_norm": 0.3465806245803833,
      "learning_rate": 1.6105322801531005e-05,
      "loss": 1.8199,
      "step": 15860
    },
    {
      "epoch": 0.04714888139956306,
      "grad_norm": 0.31671446561813354,
      "learning_rate": 1.605938957026329e-05,
      "loss": 1.8164,
      "step": 15870
    },
    {
      "epoch": 0.04769082256507528,
      "grad_norm": 0.2874152362346649,
      "learning_rate": 1.6013617304550827e-05,
      "loss": 1.8194,
      "step": 15880
    },
    {
      "epoch": 0.0482327637305875,
      "grad_norm": 0.28032568097114563,
      "learning_rate": 1.5968006193598626e-05,
      "loss": 1.8211,
      "step": 15890
    },
    {
      "epoch": 0.04877470489609972,
      "grad_norm": 0.3768242597579956,
      "learning_rate": 1.5922556425945532e-05,
      "loss": 1.8175,
      "step": 15900
    },
    {
      "epoch": 0.049316646061611936,
      "grad_norm": 0.32226991653442383,
      "learning_rate": 1.5877268189463456e-05,
      "loss": 1.8246,
      "step": 15910
    },
    {
      "epoch": 0.049858587227124156,
      "grad_norm": 0.29384303092956543,
      "learning_rate": 1.583214167135661e-05,
      "loss": 1.8125,
      "step": 15920
    },
    {
      "epoch": 0.049912781343675376,
      "eval_loss": 2.5980074405670166,
      "eval_runtime": 24.6083,
      "eval_samples_per_second": 203.184,
      "eval_steps_per_second": 1.097,
      "step": 15921
    },
    {
      "epoch": 0.050400528392636375,
      "grad_norm": 0.30090948939323425,
      "learning_rate": 1.57871770581607e-05,
      "loss": 1.8066,
      "step": 15930
    },
    {
      "epoch": 0.050942469558148594,
      "grad_norm": 0.28371649980545044,
      "learning_rate": 1.5742374535742233e-05,
      "loss": 1.8208,
      "step": 15940
    },
    {
      "epoch": 0.05148441072366081,
      "grad_norm": 0.2772349417209625,
      "learning_rate": 1.569773428929765e-05,
      "loss": 1.8237,
      "step": 15950
    },
    {
      "epoch": 0.05202635188917303,
      "grad_norm": 0.28294837474823,
      "learning_rate": 1.5653256503352603e-05,
      "loss": 1.8141,
      "step": 15960
    },
    {
      "epoch": 0.05256829305468525,
      "grad_norm": 0.264106810092926,
      "learning_rate": 1.5608941361761224e-05,
      "loss": 1.8101,
      "step": 15970
    },
    {
      "epoch": 0.05311023422019747,
      "grad_norm": 0.2991582155227661,
      "learning_rate": 1.5564789047705296e-05,
      "loss": 1.8162,
      "step": 15980
    },
    {
      "epoch": 0.05365217538570969,
      "grad_norm": 0.2891395688056946,
      "learning_rate": 1.5520799743693585e-05,
      "loss": 1.8405,
      "step": 15990
    },
    {
      "epoch": 0.05419411655122191,
      "grad_norm": 0.29393982887268066,
      "learning_rate": 1.5476973631561003e-05,
      "loss": 1.8179,
      "step": 16000
    },
    {
      "epoch": 0.05462766948363168,
      "eval_loss": 2.5983400344848633,
      "eval_runtime": 24.2547,
      "eval_samples_per_second": 206.145,
      "eval_steps_per_second": 1.113,
      "step": 16008
    },
    {
      "epoch": 0.05473605771673413,
      "grad_norm": 0.2763077914714813,
      "learning_rate": 1.5433310892467897e-05,
      "loss": 1.8226,
      "step": 16010
    },
    {
      "epoch": 0.05527799888224635,
      "grad_norm": 0.27592843770980835,
      "learning_rate": 1.5389811706899284e-05,
      "loss": 1.8173,
      "step": 16020
    },
    {
      "epoch": 0.055819940047758566,
      "grad_norm": 0.31375107169151306,
      "learning_rate": 1.5346476254664132e-05,
      "loss": 1.8118,
      "step": 16030
    },
    {
      "epoch": 0.056361881213270786,
      "grad_norm": 0.3110925853252411,
      "learning_rate": 1.5303304714894568e-05,
      "loss": 1.8171,
      "step": 16040
    },
    {
      "epoch": 0.056903822378783005,
      "grad_norm": 0.2970430254936218,
      "learning_rate": 1.526029726604521e-05,
      "loss": 1.8095,
      "step": 16050
    },
    {
      "epoch": 0.057445763544295224,
      "grad_norm": 0.27963101863861084,
      "learning_rate": 1.5217454085892358e-05,
      "loss": 1.8202,
      "step": 16060
    },
    {
      "epoch": 0.05798770470980744,
      "grad_norm": 0.31473496556282043,
      "learning_rate": 1.5174775351533289e-05,
      "loss": 1.831,
      "step": 16070
    },
    {
      "epoch": 0.05852964587531966,
      "grad_norm": 0.2634471654891968,
      "learning_rate": 1.5132261239385533e-05,
      "loss": 1.8102,
      "step": 16080
    },
    {
      "epoch": 0.05907158704083188,
      "grad_norm": 0.36205658316612244,
      "learning_rate": 1.5089911925186124e-05,
      "loss": 1.814,
      "step": 16090
    },
    {
      "epoch": 0.05934255762358799,
      "eval_loss": 2.5985541343688965,
      "eval_runtime": 28.1025,
      "eval_samples_per_second": 177.92,
      "eval_steps_per_second": 0.961,
      "step": 16095
    },
    {
      "epoch": 0.0596135282063441,
      "grad_norm": 0.5245100855827332,
      "learning_rate": 1.5047727583990898e-05,
      "loss": 1.8204,
      "step": 16100
    },
    {
      "epoch": 0.06015546937185632,
      "grad_norm": 0.316148966550827,
      "learning_rate": 1.5005708390173756e-05,
      "loss": 1.8179,
      "step": 16110
    },
    {
      "epoch": 0.06069741053736854,
      "grad_norm": 0.29313069581985474,
      "learning_rate": 1.4963854517425934e-05,
      "loss": 1.8179,
      "step": 16120
    },
    {
      "epoch": 0.06123935170288076,
      "grad_norm": 0.26670292019844055,
      "learning_rate": 1.4922166138755289e-05,
      "loss": 1.8124,
      "step": 16130
    },
    {
      "epoch": 0.06178129286839298,
      "grad_norm": 0.3401847183704376,
      "learning_rate": 1.4880643426485602e-05,
      "loss": 1.8163,
      "step": 16140
    },
    {
      "epoch": 0.062323234033905196,
      "grad_norm": 0.31569597125053406,
      "learning_rate": 1.4839286552255843e-05,
      "loss": 1.8118,
      "step": 16150
    },
    {
      "epoch": 0.06286517519941741,
      "grad_norm": 0.2905227839946747,
      "learning_rate": 1.4798095687019482e-05,
      "loss": 1.8281,
      "step": 16160
    },
    {
      "epoch": 0.06340711636492963,
      "grad_norm": 0.27679443359375,
      "learning_rate": 1.4757071001043765e-05,
      "loss": 1.8123,
      "step": 16170
    },
    {
      "epoch": 0.06394905753044185,
      "grad_norm": 0.2783887982368469,
      "learning_rate": 1.4716212663909016e-05,
      "loss": 1.806,
      "step": 16180
    },
    {
      "epoch": 0.0640574457635443,
      "eval_loss": 2.598423957824707,
      "eval_runtime": 22.218,
      "eval_samples_per_second": 225.043,
      "eval_steps_per_second": 1.215,
      "step": 16182
    },
    {
      "epoch": 0.06449099869595407,
      "grad_norm": 0.2765551507472992,
      "learning_rate": 1.4675520844507932e-05,
      "loss": 1.8033,
      "step": 16190
    },
    {
      "epoch": 0.06503293986146629,
      "grad_norm": 0.2806571424007416,
      "learning_rate": 1.4634995711044892e-05,
      "loss": 1.8129,
      "step": 16200
    },
    {
      "epoch": 0.0655748810269785,
      "grad_norm": 0.3304462134838104,
      "learning_rate": 1.4594637431035272e-05,
      "loss": 1.8113,
      "step": 16210
    },
    {
      "epoch": 0.06611682219249072,
      "grad_norm": 0.29869329929351807,
      "learning_rate": 1.4554446171304736e-05,
      "loss": 1.8068,
      "step": 16220
    },
    {
      "epoch": 0.06665876335800294,
      "grad_norm": 0.2936442196369171,
      "learning_rate": 1.451442209798852e-05,
      "loss": 1.8089,
      "step": 16230
    },
    {
      "epoch": 0.06720070452351516,
      "grad_norm": 0.2957615256309509,
      "learning_rate": 1.4474565376530819e-05,
      "loss": 1.7961,
      "step": 16240
    },
    {
      "epoch": 0.06774264568902738,
      "grad_norm": 0.2832295298576355,
      "learning_rate": 1.4434876171684023e-05,
      "loss": 1.814,
      "step": 16250
    },
    {
      "epoch": 0.0682845868545396,
      "grad_norm": 0.2817908525466919,
      "learning_rate": 1.4395354647508113e-05,
      "loss": 1.8179,
      "step": 16260
    },
    {
      "epoch": 0.0687723339035006,
      "eval_loss": 2.5987160205841064,
      "eval_runtime": 31.5857,
      "eval_samples_per_second": 158.299,
      "eval_steps_per_second": 0.855,
      "step": 16269
    },
    {
      "epoch": 0.06882652802005182,
      "grad_norm": 0.27101555466651917,
      "learning_rate": 1.4356000967369915e-05,
      "loss": 1.8071,
      "step": 16270
    },
    {
      "epoch": 0.06936846918556404,
      "grad_norm": 0.29149556159973145,
      "learning_rate": 1.4316815293942438e-05,
      "loss": 1.8095,
      "step": 16280
    },
    {
      "epoch": 0.06991041035107626,
      "grad_norm": 0.3395109176635742,
      "learning_rate": 1.4277797789204242e-05,
      "loss": 1.8077,
      "step": 16290
    },
    {
      "epoch": 0.07045235151658848,
      "grad_norm": 0.2934078276157379,
      "learning_rate": 1.4238948614438738e-05,
      "loss": 1.8234,
      "step": 16300
    },
    {
      "epoch": 0.0709942926821007,
      "grad_norm": 0.40953195095062256,
      "learning_rate": 1.4200267930233512e-05,
      "loss": 1.8193,
      "step": 16310
    },
    {
      "epoch": 0.07153623384761292,
      "grad_norm": 0.30236881971359253,
      "learning_rate": 1.4161755896479711e-05,
      "loss": 1.804,
      "step": 16320
    },
    {
      "epoch": 0.07207817501312513,
      "grad_norm": 0.3711455762386322,
      "learning_rate": 1.4123412672371278e-05,
      "loss": 1.801,
      "step": 16330
    },
    {
      "epoch": 0.07262011617863735,
      "grad_norm": 0.33262383937835693,
      "learning_rate": 1.4085238416404434e-05,
      "loss": 1.8189,
      "step": 16340
    },
    {
      "epoch": 0.07316205734414957,
      "grad_norm": 0.2707761228084564,
      "learning_rate": 1.4047233286376901e-05,
      "loss": 1.8234,
      "step": 16350
    },
    {
      "epoch": 0.07348722204345691,
      "eval_loss": 2.5961382389068604,
      "eval_runtime": 23.0319,
      "eval_samples_per_second": 217.09,
      "eval_steps_per_second": 1.172,
      "step": 16356
    },
    {
      "epoch": 0.07370399850966179,
      "grad_norm": 0.277217835187912,
      "learning_rate": 1.4009397439387317e-05,
      "loss": 1.8101,
      "step": 16360
    },
    {
      "epoch": 0.07424593967517401,
      "grad_norm": 0.324625700712204,
      "learning_rate": 1.3971731031834595e-05,
      "loss": 1.8177,
      "step": 16370
    },
    {
      "epoch": 0.07478788084068623,
      "grad_norm": 0.27004820108413696,
      "learning_rate": 1.3934234219417198e-05,
      "loss": 1.8239,
      "step": 16380
    },
    {
      "epoch": 0.07532982200619845,
      "grad_norm": 0.31389129161834717,
      "learning_rate": 1.3896907157132564e-05,
      "loss": 1.8044,
      "step": 16390
    },
    {
      "epoch": 0.07587176317171067,
      "grad_norm": 0.334043025970459,
      "learning_rate": 1.3859749999276485e-05,
      "loss": 1.8114,
      "step": 16400
    },
    {
      "epoch": 0.07641370433722289,
      "grad_norm": 0.28280240297317505,
      "learning_rate": 1.382276289944239e-05,
      "loss": 1.81,
      "step": 16410
    },
    {
      "epoch": 0.0769556455027351,
      "grad_norm": 0.2850959002971649,
      "learning_rate": 1.3785946010520792e-05,
      "loss": 1.8064,
      "step": 16420
    },
    {
      "epoch": 0.07749758666824733,
      "grad_norm": 0.33790093660354614,
      "learning_rate": 1.3749299484698571e-05,
      "loss": 1.8038,
      "step": 16430
    },
    {
      "epoch": 0.07803952783375954,
      "grad_norm": 0.3782959282398224,
      "learning_rate": 1.3712823473458422e-05,
      "loss": 1.814,
      "step": 16440
    },
    {
      "epoch": 0.0782021101834132,
      "eval_loss": 2.5989608764648438,
      "eval_runtime": 23.6056,
      "eval_samples_per_second": 211.814,
      "eval_steps_per_second": 1.144,
      "step": 16443
    },
    {
      "epoch": 0.07858146899927176,
      "grad_norm": 0.34481021761894226,
      "learning_rate": 1.36765181275782e-05,
      "loss": 1.8129,
      "step": 16450
    },
    {
      "epoch": 0.07912341016478398,
      "grad_norm": 0.2746882438659668,
      "learning_rate": 1.3640383597130288e-05,
      "loss": 1.8063,
      "step": 16460
    },
    {
      "epoch": 0.0796653513302962,
      "grad_norm": 0.38598158955574036,
      "learning_rate": 1.3604420031480968e-05,
      "loss": 1.8072,
      "step": 16470
    },
    {
      "epoch": 0.08020729249580842,
      "grad_norm": 0.267660915851593,
      "learning_rate": 1.356862757928985e-05,
      "loss": 1.7979,
      "step": 16480
    },
    {
      "epoch": 0.08074923366132064,
      "grad_norm": 0.33238664269447327,
      "learning_rate": 1.3533006388509187e-05,
      "loss": 1.8071,
      "step": 16490
    },
    {
      "epoch": 0.08129117482683286,
      "grad_norm": 0.28862419724464417,
      "learning_rate": 1.3497556606383354e-05,
      "loss": 1.8117,
      "step": 16500
    },
    {
      "epoch": 0.08183311599234508,
      "grad_norm": 0.36335626244544983,
      "learning_rate": 1.3462278379448148e-05,
      "loss": 1.8215,
      "step": 16510
    },
    {
      "epoch": 0.0823750571578573,
      "grad_norm": 0.31064873933792114,
      "learning_rate": 1.3427171853530232e-05,
      "loss": 1.803,
      "step": 16520
    },
    {
      "epoch": 0.08291699832336952,
      "grad_norm": 0.3698948919773102,
      "learning_rate": 1.3392237173746527e-05,
      "loss": 1.8072,
      "step": 16530
    },
    {
      "epoch": 0.08291699832336952,
      "eval_loss": 2.5981791019439697,
      "eval_runtime": 21.9635,
      "eval_samples_per_second": 227.65,
      "eval_steps_per_second": 1.229,
      "step": 16530
    },
    {
      "epoch": 0.08345893948888174,
      "grad_norm": 0.272014319896698,
      "learning_rate": 1.3357474484503613e-05,
      "loss": 1.8132,
      "step": 16540
    },
    {
      "epoch": 0.08400088065439396,
      "grad_norm": 0.2674700617790222,
      "learning_rate": 1.3322883929497105e-05,
      "loss": 1.8112,
      "step": 16550
    },
    {
      "epoch": 0.08454282181990617,
      "grad_norm": 0.3143172264099121,
      "learning_rate": 1.3288465651711113e-05,
      "loss": 1.8035,
      "step": 16560
    },
    {
      "epoch": 0.0850847629854184,
      "grad_norm": 0.2932548224925995,
      "learning_rate": 1.3254219793417596e-05,
      "loss": 1.8099,
      "step": 16570
    },
    {
      "epoch": 0.08562670415093061,
      "grad_norm": 0.28410300612449646,
      "learning_rate": 1.3220146496175808e-05,
      "loss": 1.8086,
      "step": 16580
    },
    {
      "epoch": 0.08616864531644283,
      "grad_norm": 0.2889862358570099,
      "learning_rate": 1.3186245900831692e-05,
      "loss": 1.8116,
      "step": 16590
    },
    {
      "epoch": 0.08671058648195505,
      "grad_norm": 0.26894381642341614,
      "learning_rate": 1.3152518147517309e-05,
      "loss": 1.8031,
      "step": 16600
    },
    {
      "epoch": 0.08725252764746727,
      "grad_norm": 0.265250027179718,
      "learning_rate": 1.3118963375650279e-05,
      "loss": 1.8048,
      "step": 16610
    },
    {
      "epoch": 0.08763188646332583,
      "eval_loss": 2.597559690475464,
      "eval_runtime": 27.4632,
      "eval_samples_per_second": 182.062,
      "eval_steps_per_second": 0.983,
      "step": 16617
    },
    {
      "epoch": 0.08779446881297949,
      "grad_norm": 0.27796122431755066,
      "learning_rate": 1.3085581723933146e-05,
      "loss": 1.8049,
      "step": 16620
    },
    {
      "epoch": 0.08833640997849171,
      "grad_norm": 0.3169678747653961,
      "learning_rate": 1.3052373330352873e-05,
      "loss": 1.8094,
      "step": 16630
    },
    {
      "epoch": 0.08887835114400393,
      "grad_norm": 0.3725588321685791,
      "learning_rate": 1.3019338332180223e-05,
      "loss": 1.8171,
      "step": 16640
    },
    {
      "epoch": 0.08942029230951615,
      "grad_norm": 0.3112504780292511,
      "learning_rate": 1.2986476865969215e-05,
      "loss": 1.8162,
      "step": 16650
    },
    {
      "epoch": 0.08996223347502837,
      "grad_norm": 0.33968761563301086,
      "learning_rate": 1.2953789067556545e-05,
      "loss": 1.8147,
      "step": 16660
    },
    {
      "epoch": 0.09050417464054059,
      "grad_norm": 0.3981771469116211,
      "learning_rate": 1.2921275072061061e-05,
      "loss": 1.8085,
      "step": 16670
    },
    {
      "epoch": 0.0910461158060528,
      "grad_norm": 0.35043928027153015,
      "learning_rate": 1.2888935013883141e-05,
      "loss": 1.8226,
      "step": 16680
    },
    {
      "epoch": 0.09158805697156502,
      "grad_norm": 0.26825150847435,
      "learning_rate": 1.2856769026704188e-05,
      "loss": 1.8066,
      "step": 16690
    },
    {
      "epoch": 0.09212999813707724,
      "grad_norm": 0.27858951687812805,
      "learning_rate": 1.2824777243486063e-05,
      "loss": 1.8188,
      "step": 16700
    },
    {
      "epoch": 0.09234677460328213,
      "eval_loss": 2.594632863998413,
      "eval_runtime": 22.2955,
      "eval_samples_per_second": 224.26,
      "eval_steps_per_second": 1.211,
      "step": 16704
    },
    {
      "epoch": 0.09267193930258946,
      "grad_norm": 0.27273324131965637,
      "learning_rate": 1.2792959796470527e-05,
      "loss": 1.804,
      "step": 16710
    },
    {
      "epoch": 0.09321388046810168,
      "grad_norm": 0.3499986529350281,
      "learning_rate": 1.2761316817178725e-05,
      "loss": 1.8051,
      "step": 16720
    },
    {
      "epoch": 0.0937558216336139,
      "grad_norm": 0.26582643389701843,
      "learning_rate": 1.2729848436410596e-05,
      "loss": 1.8208,
      "step": 16730
    },
    {
      "epoch": 0.09429776279912612,
      "grad_norm": 0.3342505693435669,
      "learning_rate": 1.2698554784244355e-05,
      "loss": 1.8129,
      "step": 16740
    },
    {
      "epoch": 0.09483970396463834,
      "grad_norm": 0.28335967659950256,
      "learning_rate": 1.2667435990035972e-05,
      "loss": 1.8078,
      "step": 16750
    },
    {
      "epoch": 0.09538164513015056,
      "grad_norm": 0.2727366089820862,
      "learning_rate": 1.2636492182418594e-05,
      "loss": 1.8062,
      "step": 16760
    },
    {
      "epoch": 0.09592358629566278,
      "grad_norm": 0.31147339940071106,
      "learning_rate": 1.2605723489302087e-05,
      "loss": 1.8107,
      "step": 16770
    },
    {
      "epoch": 0.096465527461175,
      "grad_norm": 0.2873106002807617,
      "learning_rate": 1.2575130037872418e-05,
      "loss": 1.8043,
      "step": 16780
    },
    {
      "epoch": 0.09700746862668722,
      "grad_norm": 0.32366591691970825,
      "learning_rate": 1.2544711954591192e-05,
      "loss": 1.8257,
      "step": 16790
    },
    {
      "epoch": 0.09706166274323844,
      "eval_loss": 2.5952253341674805,
      "eval_runtime": 27.6353,
      "eval_samples_per_second": 180.928,
      "eval_steps_per_second": 0.977,
      "step": 16791
    },
    {
      "epoch": 0.09754940979219943,
      "grad_norm": 0.2702518105506897,
      "learning_rate": 1.2514469365195094e-05,
      "loss": 1.8183,
      "step": 16800
    },
    {
      "epoch": 0.09809135095771165,
      "grad_norm": 0.28163909912109375,
      "learning_rate": 1.2484402394695408e-05,
      "loss": 1.8045,
      "step": 16810
    },
    {
      "epoch": 0.09863329212322387,
      "grad_norm": 0.2960811257362366,
      "learning_rate": 1.2454511167377447e-05,
      "loss": 1.8087,
      "step": 16820
    },
    {
      "epoch": 0.09917523328873609,
      "grad_norm": 0.28118273615837097,
      "learning_rate": 1.2424795806800102e-05,
      "loss": 1.812,
      "step": 16830
    },
    {
      "epoch": 0.09971717445424831,
      "grad_norm": 0.30942973494529724,
      "learning_rate": 1.2395256435795278e-05,
      "loss": 1.8082,
      "step": 16840
    },
    {
      "epoch": 0.10025911561976053,
      "grad_norm": 0.33988916873931885,
      "learning_rate": 1.2365893176467404e-05,
      "loss": 1.812,
      "step": 16850
    },
    {
      "epoch": 0.10080105678527275,
      "grad_norm": 0.3102359473705292,
      "learning_rate": 1.2336706150192946e-05,
      "loss": 1.8144,
      "step": 16860
    },
    {
      "epoch": 0.10134299795078497,
      "grad_norm": 0.3185904920101166,
      "learning_rate": 1.2307695477619872e-05,
      "loss": 1.805,
      "step": 16870
    },
    {
      "epoch": 0.10177655088319475,
      "eval_loss": 2.5944910049438477,
      "eval_runtime": 22.1385,
      "eval_samples_per_second": 225.85,
      "eval_steps_per_second": 1.22,
      "step": 16878
    },
    {
      "epoch": 0.10188493911629719,
      "grad_norm": 0.28458863496780396,
      "learning_rate": 1.2278861278667187e-05,
      "loss": 1.795,
      "step": 16880
    },
    {
      "epoch": 0.10242688028180941,
      "grad_norm": 0.31006863713264465,
      "learning_rate": 1.2250203672524424e-05,
      "loss": 1.7943,
      "step": 16890
    },
    {
      "epoch": 0.10296882144732163,
      "grad_norm": 0.28029975295066833,
      "learning_rate": 1.2221722777651119e-05,
      "loss": 1.8163,
      "step": 16900
    },
    {
      "epoch": 0.10351076261283385,
      "grad_norm": 0.27603140473365784,
      "learning_rate": 1.219341871177639e-05,
      "loss": 1.814,
      "step": 16910
    },
    {
      "epoch": 0.10405270377834606,
      "grad_norm": 0.2757927179336548,
      "learning_rate": 1.2165291591898383e-05,
      "loss": 1.8138,
      "step": 16920
    },
    {
      "epoch": 0.10459464494385828,
      "grad_norm": 0.2880472242832184,
      "learning_rate": 1.2137341534283842e-05,
      "loss": 1.8064,
      "step": 16930
    },
    {
      "epoch": 0.1051365861093705,
      "grad_norm": 0.2712019979953766,
      "learning_rate": 1.2109568654467594e-05,
      "loss": 1.8005,
      "step": 16940
    },
    {
      "epoch": 0.10567852727488272,
      "grad_norm": 0.2719828188419342,
      "learning_rate": 1.2081973067252051e-05,
      "loss": 1.8146,
      "step": 16950
    },
    {
      "epoch": 0.10622046844039494,
      "grad_norm": 0.3635210394859314,
      "learning_rate": 1.205455488670681e-05,
      "loss": 1.8189,
      "step": 16960
    },
    {
      "epoch": 0.10649143902315104,
      "eval_loss": 2.596562385559082,
      "eval_runtime": 28.8968,
      "eval_samples_per_second": 173.029,
      "eval_steps_per_second": 0.934,
      "step": 16965
    },
    {
      "epoch": 0.10676240960590716,
      "grad_norm": 0.2864342927932739,
      "learning_rate": 1.2027314226168121e-05,
      "loss": 1.8097,
      "step": 16970
    },
    {
      "epoch": 0.10730435077141938,
      "grad_norm": 0.3084823489189148,
      "learning_rate": 1.2000251198238424e-05,
      "loss": 1.8093,
      "step": 16980
    },
    {
      "epoch": 0.1078462919369316,
      "grad_norm": 0.3098483383655548,
      "learning_rate": 1.1973365914785909e-05,
      "loss": 1.8137,
      "step": 16990
    },
    {
      "epoch": 0.10838823310244382,
      "grad_norm": 0.30631518363952637,
      "learning_rate": 1.1946658486944022e-05,
      "loss": 1.8042,
      "step": 17000
    },
    {
      "epoch": 0.10893017426795604,
      "grad_norm": 0.33063915371894836,
      "learning_rate": 1.1920129025111034e-05,
      "loss": 1.8086,
      "step": 17010
    },
    {
      "epoch": 0.10947211543346826,
      "grad_norm": 0.2645542025566101,
      "learning_rate": 1.1893777638949576e-05,
      "loss": 1.8219,
      "step": 17020
    },
    {
      "epoch": 0.11001405659898048,
      "grad_norm": 0.27368003129959106,
      "learning_rate": 1.1867604437386164e-05,
      "loss": 1.8101,
      "step": 17030
    },
    {
      "epoch": 0.1105559977644927,
      "grad_norm": 0.2701696455478668,
      "learning_rate": 1.1841609528610796e-05,
      "loss": 1.8174,
      "step": 17040
    },
    {
      "epoch": 0.11109793893000491,
      "grad_norm": 0.28612881898880005,
      "learning_rate": 1.1815793020076448e-05,
      "loss": 1.8077,
      "step": 17050
    },
    {
      "epoch": 0.11120632716310735,
      "eval_loss": 2.595874071121216,
      "eval_runtime": 29.3622,
      "eval_samples_per_second": 170.287,
      "eval_steps_per_second": 0.92,
      "step": 17052
    },
    {
      "epoch": 0.11163988009551713,
      "grad_norm": 0.33459609746932983,
      "learning_rate": 1.1790155018498666e-05,
      "loss": 1.7929,
      "step": 17060
    },
    {
      "epoch": 0.11218182126102935,
      "grad_norm": 0.27914193272590637,
      "learning_rate": 1.176469562985513e-05,
      "loss": 1.8074,
      "step": 17070
    },
    {
      "epoch": 0.11272376242654157,
      "grad_norm": 0.26985886693000793,
      "learning_rate": 1.1739414959385191e-05,
      "loss": 1.8024,
      "step": 17080
    },
    {
      "epoch": 0.11326570359205379,
      "grad_norm": 0.4287962019443512,
      "learning_rate": 1.1714313111589445e-05,
      "loss": 1.7996,
      "step": 17090
    },
    {
      "epoch": 0.11380764475756601,
      "grad_norm": 0.29455164074897766,
      "learning_rate": 1.1689390190229312e-05,
      "loss": 1.8117,
      "step": 17100
    },
    {
      "epoch": 0.11434958592307823,
      "grad_norm": 0.26276010274887085,
      "learning_rate": 1.1664646298326582e-05,
      "loss": 1.8055,
      "step": 17110
    },
    {
      "epoch": 0.11489152708859045,
      "grad_norm": 0.2993430495262146,
      "learning_rate": 1.1640081538163036e-05,
      "loss": 1.8146,
      "step": 17120
    },
    {
      "epoch": 0.11543346825410267,
      "grad_norm": 0.3164224922657013,
      "learning_rate": 1.1615696011279965e-05,
      "loss": 1.8064,
      "step": 17130
    },
    {
      "epoch": 0.11592121530306367,
      "eval_loss": 2.596324920654297,
      "eval_runtime": 25.6516,
      "eval_samples_per_second": 194.919,
      "eval_steps_per_second": 1.053,
      "step": 17139
    },
    {
      "epoch": 0.11597540941961489,
      "grad_norm": 0.3307249844074249,
      "learning_rate": 1.1591489818477795e-05,
      "loss": 1.8039,
      "step": 17140
    },
    {
      "epoch": 0.1165173505851271,
      "grad_norm": 0.29287073016166687,
      "learning_rate": 1.1567463059815642e-05,
      "loss": 1.8227,
      "step": 17150
    },
    {
      "epoch": 0.11705929175063932,
      "grad_norm": 0.282429039478302,
      "learning_rate": 1.1543615834610914e-05,
      "loss": 1.8041,
      "step": 17160
    },
    {
      "epoch": 0.11760123291615154,
      "grad_norm": 0.2842390835285187,
      "learning_rate": 1.1519948241438899e-05,
      "loss": 1.8088,
      "step": 17170
    },
    {
      "epoch": 0.11814317408166376,
      "grad_norm": 0.2744342088699341,
      "learning_rate": 1.149646037813236e-05,
      "loss": 1.8148,
      "step": 17180
    },
    {
      "epoch": 0.11868511524717598,
      "grad_norm": 0.28901031613349915,
      "learning_rate": 1.1473152341781114e-05,
      "loss": 1.8048,
      "step": 17190
    },
    {
      "epoch": 0.1192270564126882,
      "grad_norm": 0.2894619107246399,
      "learning_rate": 1.1450024228731648e-05,
      "loss": 1.8114,
      "step": 17200
    },
    {
      "epoch": 0.11976899757820042,
      "grad_norm": 0.2956114709377289,
      "learning_rate": 1.1427076134586722e-05,
      "loss": 1.8023,
      "step": 17210
    },
    {
      "epoch": 0.12031093874371264,
      "grad_norm": 0.28053778409957886,
      "learning_rate": 1.1404308154204944e-05,
      "loss": 1.8051,
      "step": 17220
    },
    {
      "epoch": 0.12063610344301996,
      "eval_loss": 2.5927584171295166,
      "eval_runtime": 23.5866,
      "eval_samples_per_second": 211.984,
      "eval_steps_per_second": 1.145,
      "step": 17226
    },
    {
      "epoch": 0.12085287990922486,
      "grad_norm": 0.3586195707321167,
      "learning_rate": 1.1381720381700439e-05,
      "loss": 1.8011,
      "step": 17230
    },
    {
      "epoch": 0.12139482107473708,
      "grad_norm": 0.2799775004386902,
      "learning_rate": 1.1359312910442378e-05,
      "loss": 1.8082,
      "step": 17240
    },
    {
      "epoch": 0.1219367622402493,
      "grad_norm": 0.2748625874519348,
      "learning_rate": 1.1337085833054672e-05,
      "loss": 1.7877,
      "step": 17250
    },
    {
      "epoch": 0.12247870340576152,
      "grad_norm": 0.27648496627807617,
      "learning_rate": 1.131503924141553e-05,
      "loss": 1.7975,
      "step": 17260
    },
    {
      "epoch": 0.12302064457127374,
      "grad_norm": 0.27565595507621765,
      "learning_rate": 1.1293173226657108e-05,
      "loss": 1.8062,
      "step": 17270
    },
    {
      "epoch": 0.12356258573678595,
      "grad_norm": 0.2683922052383423,
      "learning_rate": 1.1271487879165137e-05,
      "loss": 1.804,
      "step": 17280
    },
    {
      "epoch": 0.12410452690229817,
      "grad_norm": 0.27080050110816956,
      "learning_rate": 1.1249983288578525e-05,
      "loss": 1.8092,
      "step": 17290
    },
    {
      "epoch": 0.12464646806781039,
      "grad_norm": 0.3873322308063507,
      "learning_rate": 1.1228659543789005e-05,
      "loss": 1.8062,
      "step": 17300
    },
    {
      "epoch": 0.1251884092333226,
      "grad_norm": 0.3005862534046173,
      "learning_rate": 1.1207516732940766e-05,
      "loss": 1.8053,
      "step": 17310
    },
    {
      "epoch": 0.12535099158297627,
      "eval_loss": 2.597273826599121,
      "eval_runtime": 26.7305,
      "eval_samples_per_second": 187.052,
      "eval_steps_per_second": 1.01,
      "step": 17313
    },
    {
      "epoch": 0.12573035039883482,
      "grad_norm": 0.27348408102989197,
      "learning_rate": 1.118655494343008e-05,
      "loss": 1.8239,
      "step": 17320
    },
    {
      "epoch": 0.12627229156434705,
      "grad_norm": 0.3401775360107422,
      "learning_rate": 1.1165774261904948e-05,
      "loss": 1.7989,
      "step": 17330
    },
    {
      "epoch": 0.12681423272985926,
      "grad_norm": 0.2691876292228699,
      "learning_rate": 1.114517477426475e-05,
      "loss": 1.8121,
      "step": 17340
    },
    {
      "epoch": 0.1273561738953715,
      "grad_norm": 0.2959340214729309,
      "learning_rate": 1.1124756565659861e-05,
      "loss": 1.7961,
      "step": 17350
    },
    {
      "epoch": 0.1278981150608837,
      "grad_norm": 0.28132373094558716,
      "learning_rate": 1.1104519720491333e-05,
      "loss": 1.8025,
      "step": 17360
    },
    {
      "epoch": 0.12844005622639593,
      "grad_norm": 0.322965145111084,
      "learning_rate": 1.108446432241052e-05,
      "loss": 1.811,
      "step": 17370
    },
    {
      "epoch": 0.12898199739190813,
      "grad_norm": 0.28159090876579285,
      "learning_rate": 1.106459045431875e-05,
      "loss": 1.806,
      "step": 17380
    },
    {
      "epoch": 0.12952393855742036,
      "grad_norm": 0.25992992520332336,
      "learning_rate": 1.1044898198366984e-05,
      "loss": 1.8067,
      "step": 17390
    },
    {
      "epoch": 0.13006587972293257,
      "grad_norm": 0.3366539180278778,
      "learning_rate": 1.1025387635955451e-05,
      "loss": 1.8015,
      "step": 17400
    },
    {
      "epoch": 0.13006587972293257,
      "eval_loss": 2.593247890472412,
      "eval_runtime": 21.9708,
      "eval_samples_per_second": 227.575,
      "eval_steps_per_second": 1.229,
      "step": 17400
    },
    {
      "epoch": 0.1306078208884448,
      "grad_norm": 0.2966313660144806,
      "learning_rate": 1.1006058847733338e-05,
      "loss": 1.8048,
      "step": 17410
    },
    {
      "epoch": 0.131149762053957,
      "grad_norm": 0.27395039796829224,
      "learning_rate": 1.098691191359845e-05,
      "loss": 1.8117,
      "step": 17420
    },
    {
      "epoch": 0.13169170321946924,
      "grad_norm": 0.331516295671463,
      "learning_rate": 1.096794691269686e-05,
      "loss": 1.805,
      "step": 17430
    },
    {
      "epoch": 0.13223364438498145,
      "grad_norm": 0.30107611417770386,
      "learning_rate": 1.0949163923422624e-05,
      "loss": 1.8032,
      "step": 17440
    },
    {
      "epoch": 0.13277558555049368,
      "grad_norm": 0.2857365310192108,
      "learning_rate": 1.0930563023417417e-05,
      "loss": 1.8035,
      "step": 17450
    },
    {
      "epoch": 0.13331752671600589,
      "grad_norm": 0.2834301292896271,
      "learning_rate": 1.0912144289570228e-05,
      "loss": 1.7969,
      "step": 17460
    },
    {
      "epoch": 0.13385946788151812,
      "grad_norm": 0.275591641664505,
      "learning_rate": 1.089390779801704e-05,
      "loss": 1.7935,
      "step": 17470
    },
    {
      "epoch": 0.13440140904703032,
      "grad_norm": 0.27701419591903687,
      "learning_rate": 1.0875853624140517e-05,
      "loss": 1.8203,
      "step": 17480
    },
    {
      "epoch": 0.1347807678628889,
      "eval_loss": 2.59383487701416,
      "eval_runtime": 26.2522,
      "eval_samples_per_second": 190.46,
      "eval_steps_per_second": 1.028,
      "step": 17487
    },
    {
      "epoch": 0.13494335021254256,
      "grad_norm": 0.3042493164539337,
      "learning_rate": 1.0857981842569686e-05,
      "loss": 1.7945,
      "step": 17490
    },
    {
      "epoch": 0.13548529137805476,
      "grad_norm": 0.3667808473110199,
      "learning_rate": 1.0840292527179657e-05,
      "loss": 1.8075,
      "step": 17500
    },
    {
      "epoch": 0.136027232543567,
      "grad_norm": 0.28393447399139404,
      "learning_rate": 1.0822785751091268e-05,
      "loss": 1.7968,
      "step": 17510
    },
    {
      "epoch": 0.1365691737090792,
      "grad_norm": 0.2786491811275482,
      "learning_rate": 1.0805461586670825e-05,
      "loss": 1.8161,
      "step": 17520
    },
    {
      "epoch": 0.13711111487459143,
      "grad_norm": 0.27336838841438293,
      "learning_rate": 1.0788320105529776e-05,
      "loss": 1.7943,
      "step": 17530
    },
    {
      "epoch": 0.13765305604010364,
      "grad_norm": 0.2629176676273346,
      "learning_rate": 1.0771361378524431e-05,
      "loss": 1.8007,
      "step": 17540
    },
    {
      "epoch": 0.13819499720561587,
      "grad_norm": 0.26536351442337036,
      "learning_rate": 1.0754585475755682e-05,
      "loss": 1.7939,
      "step": 17550
    },
    {
      "epoch": 0.13873693837112808,
      "grad_norm": 0.30199551582336426,
      "learning_rate": 1.0737992466568667e-05,
      "loss": 1.7971,
      "step": 17560
    },
    {
      "epoch": 0.1392788795366403,
      "grad_norm": 0.2917165160179138,
      "learning_rate": 1.072158241955252e-05,
      "loss": 1.8053,
      "step": 17570
    },
    {
      "epoch": 0.1394956560028452,
      "eval_loss": 2.5910274982452393,
      "eval_runtime": 27.7651,
      "eval_samples_per_second": 180.082,
      "eval_steps_per_second": 0.972,
      "step": 17574
    },
    {
      "epoch": 0.13982082070215252,
      "grad_norm": 0.3207840025424957,
      "learning_rate": 1.0705355402540095e-05,
      "loss": 1.8075,
      "step": 17580
    },
    {
      "epoch": 0.14036276186766475,
      "grad_norm": 0.2685806155204773,
      "learning_rate": 1.0689311482607655e-05,
      "loss": 1.8048,
      "step": 17590
    },
    {
      "epoch": 0.14090470303317695,
      "grad_norm": 0.2922206521034241,
      "learning_rate": 1.0673450726074611e-05,
      "loss": 1.7965,
      "step": 17600
    },
    {
      "epoch": 0.1414466441986892,
      "grad_norm": 0.3959106504917145,
      "learning_rate": 1.0657773198503267e-05,
      "loss": 1.8087,
      "step": 17610
    },
    {
      "epoch": 0.1419885853642014,
      "grad_norm": 0.33581146597862244,
      "learning_rate": 1.0642278964698486e-05,
      "loss": 1.8066,
      "step": 17620
    },
    {
      "epoch": 0.14253052652971362,
      "grad_norm": 0.2724241614341736,
      "learning_rate": 1.0626968088707508e-05,
      "loss": 1.8005,
      "step": 17630
    },
    {
      "epoch": 0.14307246769522583,
      "grad_norm": 0.2701176404953003,
      "learning_rate": 1.0611840633819626e-05,
      "loss": 1.8039,
      "step": 17640
    },
    {
      "epoch": 0.14361440886073806,
      "grad_norm": 0.2956571877002716,
      "learning_rate": 1.0596896662565925e-05,
      "loss": 1.7962,
      "step": 17650
    },
    {
      "epoch": 0.14415635002625027,
      "grad_norm": 0.29787176847457886,
      "learning_rate": 1.0582136236719073e-05,
      "loss": 1.8094,
      "step": 17660
    },
    {
      "epoch": 0.1442105441428015,
      "eval_loss": 2.5948381423950195,
      "eval_runtime": 22.6082,
      "eval_samples_per_second": 221.158,
      "eval_steps_per_second": 1.194,
      "step": 17661
    },
    {
      "epoch": 0.1446982911917625,
      "grad_norm": 0.2762181758880615,
      "learning_rate": 1.0567559417292992e-05,
      "loss": 1.8069,
      "step": 17670
    },
    {
      "epoch": 0.1452402323572747,
      "grad_norm": 0.2777063548564911,
      "learning_rate": 1.0553166264542675e-05,
      "loss": 1.8085,
      "step": 17680
    },
    {
      "epoch": 0.14578217352278694,
      "grad_norm": 0.2686801552772522,
      "learning_rate": 1.05389568379639e-05,
      "loss": 1.7995,
      "step": 17690
    },
    {
      "epoch": 0.14632411468829915,
      "grad_norm": 0.3426153361797333,
      "learning_rate": 1.0524931196292987e-05,
      "loss": 1.8079,
      "step": 17700
    },
    {
      "epoch": 0.14686605585381138,
      "grad_norm": 0.30656588077545166,
      "learning_rate": 1.0511089397506558e-05,
      "loss": 1.8001,
      "step": 17710
    },
    {
      "epoch": 0.14740799701932358,
      "grad_norm": 0.26832064986228943,
      "learning_rate": 1.0497431498821308e-05,
      "loss": 1.8022,
      "step": 17720
    },
    {
      "epoch": 0.14794993818483582,
      "grad_norm": 0.31563130021095276,
      "learning_rate": 1.0483957556693748e-05,
      "loss": 1.7997,
      "step": 17730
    },
    {
      "epoch": 0.14849187935034802,
      "grad_norm": 0.2889915406703949,
      "learning_rate": 1.0470667626820001e-05,
      "loss": 1.7999,
      "step": 17740
    },
    {
      "epoch": 0.1489254322827578,
      "eval_loss": 2.596463918685913,
      "eval_runtime": 22.7533,
      "eval_samples_per_second": 219.749,
      "eval_steps_per_second": 1.187,
      "step": 17748
    },
    {
      "epoch": 0.14903382051586025,
      "grad_norm": 0.2632191777229309,
      "learning_rate": 1.0457561764135531e-05,
      "loss": 1.7958,
      "step": 17750
    },
    {
      "epoch": 0.14957576168137246,
      "grad_norm": 0.3954158425331116,
      "learning_rate": 1.0444640022814952e-05,
      "loss": 1.8029,
      "step": 17760
    },
    {
      "epoch": 0.1501177028468847,
      "grad_norm": 0.32601505517959595,
      "learning_rate": 1.0431902456271798e-05,
      "loss": 1.8048,
      "step": 17770
    },
    {
      "epoch": 0.1506596440123969,
      "grad_norm": 0.3632460832595825,
      "learning_rate": 1.0419349117158271e-05,
      "loss": 1.7984,
      "step": 17780
    },
    {
      "epoch": 0.15120158517790913,
      "grad_norm": 0.2885991036891937,
      "learning_rate": 1.040698005736507e-05,
      "loss": 1.7987,
      "step": 17790
    },
    {
      "epoch": 0.15174352634342134,
      "grad_norm": 0.3226932883262634,
      "learning_rate": 1.0394795328021151e-05,
      "loss": 1.8114,
      "step": 17800
    },
    {
      "epoch": 0.15228546750893357,
      "grad_norm": 0.2789970934391022,
      "learning_rate": 1.038279497949351e-05,
      "loss": 1.7995,
      "step": 17810
    },
    {
      "epoch": 0.15282740867444577,
      "grad_norm": 0.2944728136062622,
      "learning_rate": 1.0370979061386985e-05,
      "loss": 1.8049,
      "step": 17820
    },
    {
      "epoch": 0.153369349839958,
      "grad_norm": 0.28693804144859314,
      "learning_rate": 1.0359347622544055e-05,
      "loss": 1.79,
      "step": 17830
    },
    {
      "epoch": 0.1536403204227141,
      "eval_loss": 2.597463369369507,
      "eval_runtime": 23.0476,
      "eval_samples_per_second": 216.942,
      "eval_steps_per_second": 1.171,
      "step": 17835
    },
    {
      "epoch": 0.1539112910054702,
      "grad_norm": 0.2763684391975403,
      "learning_rate": 1.0347900711044624e-05,
      "loss": 1.81,
      "step": 17840
    },
    {
      "epoch": 0.15445323217098245,
      "grad_norm": 0.2787637412548065,
      "learning_rate": 1.0336638374205845e-05,
      "loss": 1.8125,
      "step": 17850
    },
    {
      "epoch": 0.15499517333649465,
      "grad_norm": 0.2788551449775696,
      "learning_rate": 1.0325560658581893e-05,
      "loss": 1.816,
      "step": 17860
    },
    {
      "epoch": 0.15553711450200688,
      "grad_norm": 0.2845081686973572,
      "learning_rate": 1.0314667609963802e-05,
      "loss": 1.7967,
      "step": 17870
    },
    {
      "epoch": 0.1560790556675191,
      "grad_norm": 0.37821102142333984,
      "learning_rate": 1.0303959273379258e-05,
      "loss": 1.7942,
      "step": 17880
    },
    {
      "epoch": 0.15662099683303132,
      "grad_norm": 0.2836483418941498,
      "learning_rate": 1.0293435693092411e-05,
      "loss": 1.814,
      "step": 17890
    },
    {
      "epoch": 0.15716293799854353,
      "grad_norm": 0.2846856415271759,
      "learning_rate": 1.0283096912603723e-05,
      "loss": 1.8124,
      "step": 17900
    },
    {
      "epoch": 0.15770487916405576,
      "grad_norm": 0.2603934705257416,
      "learning_rate": 1.027294297464974e-05,
      "loss": 1.8014,
      "step": 17910
    },
    {
      "epoch": 0.15824682032956797,
      "grad_norm": 0.34290361404418945,
      "learning_rate": 1.0262973921202948e-05,
      "loss": 1.7978,
      "step": 17920
    },
    {
      "epoch": 0.1583552085626704,
      "eval_loss": 2.5988216400146484,
      "eval_runtime": 25.8064,
      "eval_samples_per_second": 193.75,
      "eval_steps_per_second": 1.046,
      "step": 17922
    },
    {
      "epoch": 0.1587887614950802,
      "grad_norm": 0.28911086916923523,
      "learning_rate": 1.0253189793471598e-05,
      "loss": 1.8057,
      "step": 17930
    },
    {
      "epoch": 0.1593307026605924,
      "grad_norm": 0.297972708940506,
      "learning_rate": 1.0243590631899517e-05,
      "loss": 1.8037,
      "step": 17940
    },
    {
      "epoch": 0.15987264382610464,
      "grad_norm": 0.27581244707107544,
      "learning_rate": 1.023417647616596e-05,
      "loss": 1.8009,
      "step": 17950
    },
    {
      "epoch": 0.16041458499161684,
      "grad_norm": 0.2763090133666992,
      "learning_rate": 1.0224947365185444e-05,
      "loss": 1.8058,
      "step": 17960
    },
    {
      "epoch": 0.16095652615712908,
      "grad_norm": 0.285412073135376,
      "learning_rate": 1.0215903337107571e-05,
      "loss": 1.8179,
      "step": 17970
    },
    {
      "epoch": 0.16149846732264128,
      "grad_norm": 0.31604015827178955,
      "learning_rate": 1.0207044429316883e-05,
      "loss": 1.8074,
      "step": 17980
    },
    {
      "epoch": 0.16204040848815351,
      "grad_norm": 0.2986457645893097,
      "learning_rate": 1.0198370678432713e-05,
      "loss": 1.7948,
      "step": 17990
    },
    {
      "epoch": 0.16258234965366572,
      "grad_norm": 0.28448864817619324,
      "learning_rate": 1.0189882120309019e-05,
      "loss": 1.794,
      "step": 18000
    },
    {
      "epoch": 0.16307009670262673,
      "eval_loss": 2.5937376022338867,
      "eval_runtime": 24.5121,
      "eval_samples_per_second": 203.981,
      "eval_steps_per_second": 1.101,
      "step": 18009
    },
    {
      "epoch": 0.16312429081917795,
      "grad_norm": 0.3744131922721863,
      "learning_rate": 1.0181578790034248e-05,
      "loss": 1.7963,
      "step": 18010
    },
    {
      "epoch": 0.16366623198469016,
      "grad_norm": 0.35869526863098145,
      "learning_rate": 1.0173460721931186e-05,
      "loss": 1.7989,
      "step": 18020
    },
    {
      "epoch": 0.1642081731502024,
      "grad_norm": 0.35815170407295227,
      "learning_rate": 1.0165527949556814e-05,
      "loss": 1.8044,
      "step": 18030
    },
    {
      "epoch": 0.1647501143157146,
      "grad_norm": 0.3258884847164154,
      "learning_rate": 1.015778050570217e-05,
      "loss": 1.7979,
      "step": 18040
    },
    {
      "epoch": 0.16529205548122683,
      "grad_norm": 0.2814440429210663,
      "learning_rate": 1.0150218422392213e-05,
      "loss": 1.7945,
      "step": 18050
    },
    {
      "epoch": 0.16583399664673903,
      "grad_norm": 0.2600628733634949,
      "learning_rate": 1.0142841730885705e-05,
      "loss": 1.8,
      "step": 18060
    },
    {
      "epoch": 0.16637593781225127,
      "grad_norm": 0.2741418182849884,
      "learning_rate": 1.0135650461675054e-05,
      "loss": 1.7987,
      "step": 18070
    },
    {
      "epoch": 0.16691787897776347,
      "grad_norm": 0.286240816116333,
      "learning_rate": 1.0128644644486213e-05,
      "loss": 1.8017,
      "step": 18080
    },
    {
      "epoch": 0.1674598201432757,
      "grad_norm": 0.2839854061603546,
      "learning_rate": 1.0121824308278545e-05,
      "loss": 1.793,
      "step": 18090
    },
    {
      "epoch": 0.16778498484258303,
      "eval_loss": 2.5955307483673096,
      "eval_runtime": 23.2564,
      "eval_samples_per_second": 214.994,
      "eval_steps_per_second": 1.161,
      "step": 18096
    },
    {
      "epoch": 0.1680017613087879,
      "grad_norm": 0.2708841860294342,
      "learning_rate": 1.0115189481244706e-05,
      "loss": 1.7934,
      "step": 18100
    },
    {
      "epoch": 0.16854370247430014,
      "grad_norm": 0.32778841257095337,
      "learning_rate": 1.0108740190810519e-05,
      "loss": 1.8111,
      "step": 18110
    },
    {
      "epoch": 0.16908564363981235,
      "grad_norm": 0.28490176796913147,
      "learning_rate": 1.0102476463634891e-05,
      "loss": 1.8004,
      "step": 18120
    },
    {
      "epoch": 0.16962758480532458,
      "grad_norm": 0.28157317638397217,
      "learning_rate": 1.0096398325609658e-05,
      "loss": 1.8033,
      "step": 18130
    },
    {
      "epoch": 0.1701695259708368,
      "grad_norm": 0.29462599754333496,
      "learning_rate": 1.0090505801859521e-05,
      "loss": 1.8047,
      "step": 18140
    },
    {
      "epoch": 0.17071146713634902,
      "grad_norm": 0.3192684054374695,
      "learning_rate": 1.0084798916741909e-05,
      "loss": 1.8049,
      "step": 18150
    },
    {
      "epoch": 0.17125340830186123,
      "grad_norm": 0.2903258204460144,
      "learning_rate": 1.0079277693846895e-05,
      "loss": 1.7911,
      "step": 18160
    },
    {
      "epoch": 0.17179534946737346,
      "grad_norm": 0.26754647493362427,
      "learning_rate": 1.0073942155997108e-05,
      "loss": 1.7938,
      "step": 18170
    },
    {
      "epoch": 0.17233729063288566,
      "grad_norm": 0.26055288314819336,
      "learning_rate": 1.0068792325247608e-05,
      "loss": 1.8061,
      "step": 18180
    },
    {
      "epoch": 0.17249987298253933,
      "eval_loss": 2.5973074436187744,
      "eval_runtime": 25.1042,
      "eval_samples_per_second": 199.169,
      "eval_steps_per_second": 1.076,
      "step": 18183
    },
    {
      "epoch": 0.1728792317983979,
      "grad_norm": 0.3081805109977722,
      "learning_rate": 1.0063828222885822e-05,
      "loss": 1.81,
      "step": 18190
    },
    {
      "epoch": 0.1734211729639101,
      "grad_norm": 0.30028587579727173,
      "learning_rate": 1.0059049869431447e-05,
      "loss": 1.7991,
      "step": 18200
    },
    {
      "epoch": 0.17396311412942234,
      "grad_norm": 0.29933491349220276,
      "learning_rate": 1.0054457284636363e-05,
      "loss": 1.7965,
      "step": 18210
    },
    {
      "epoch": 0.17450505529493454,
      "grad_norm": 0.2657584547996521,
      "learning_rate": 1.0050050487484561e-05,
      "loss": 1.7991,
      "step": 18220
    },
    {
      "epoch": 0.17504699646044677,
      "grad_norm": 0.28480470180511475,
      "learning_rate": 1.004582949619204e-05,
      "loss": 1.7998,
      "step": 18230
    },
    {
      "epoch": 0.17558893762595898,
      "grad_norm": 0.3122563660144806,
      "learning_rate": 1.004179432820677e-05,
      "loss": 1.8084,
      "step": 18240
    },
    {
      "epoch": 0.1761308787914712,
      "grad_norm": 0.2768636643886566,
      "learning_rate": 1.0037945000208584e-05,
      "loss": 1.8065,
      "step": 18250
    },
    {
      "epoch": 0.17667281995698342,
      "grad_norm": 0.2690199017524719,
      "learning_rate": 1.0034281528109125e-05,
      "loss": 1.8072,
      "step": 18260
    },
    {
      "epoch": 0.17721476112249565,
      "grad_norm": 0.27862584590911865,
      "learning_rate": 1.0030803927051793e-05,
      "loss": 1.7843,
      "step": 18270
    },
    {
      "epoch": 0.17721476112249565,
      "eval_loss": 2.597975730895996,
      "eval_runtime": 21.9736,
      "eval_samples_per_second": 227.545,
      "eval_steps_per_second": 1.229,
      "step": 18270
    },
    {
      "epoch": 0.17775670228800786,
      "grad_norm": 0.3340103328227997,
      "learning_rate": 1.0027512211411644e-05,
      "loss": 1.7988,
      "step": 18280
    },
    {
      "epoch": 0.1782986434535201,
      "grad_norm": 0.33370035886764526,
      "learning_rate": 1.0024406394795373e-05,
      "loss": 1.8053,
      "step": 18290
    },
    {
      "epoch": 0.1788405846190323,
      "grad_norm": 0.2686094343662262,
      "learning_rate": 1.0021486490041246e-05,
      "loss": 1.7932,
      "step": 18300
    },
    {
      "epoch": 0.17938252578454453,
      "grad_norm": 0.27461278438568115,
      "learning_rate": 1.0018752509219017e-05,
      "loss": 1.8024,
      "step": 18310
    },
    {
      "epoch": 0.17992446695005673,
      "grad_norm": 0.26994603872299194,
      "learning_rate": 1.0016204463629918e-05,
      "loss": 1.794,
      "step": 18320
    },
    {
      "epoch": 0.18046640811556897,
      "grad_norm": 0.28709709644317627,
      "learning_rate": 1.0013842363806594e-05,
      "loss": 1.7986,
      "step": 18330
    },
    {
      "epoch": 0.18100834928108117,
      "grad_norm": 0.3366316556930542,
      "learning_rate": 1.0011666219513065e-05,
      "loss": 1.794,
      "step": 18340
    },
    {
      "epoch": 0.1815502904465934,
      "grad_norm": 0.3002452552318573,
      "learning_rate": 1.0009676039744664e-05,
      "loss": 1.8091,
      "step": 18350
    },
    {
      "epoch": 0.18192964926245195,
      "eval_loss": 2.598811149597168,
      "eval_runtime": 26.2348,
      "eval_samples_per_second": 190.586,
      "eval_steps_per_second": 1.029,
      "step": 18357
    },
    {
      "epoch": 0.1820922316121056,
      "grad_norm": 0.276909738779068,
      "learning_rate": 1.0007871832728043e-05,
      "loss": 1.7919,
      "step": 18360
    },
    {
      "epoch": 0.18263417277761784,
      "grad_norm": 0.2699563503265381,
      "learning_rate": 1.0006253605921098e-05,
      "loss": 1.7975,
      "step": 18370
    },
    {
      "epoch": 0.18317611394313005,
      "grad_norm": 0.30202600359916687,
      "learning_rate": 1.0004821366012962e-05,
      "loss": 1.7966,
      "step": 18380
    },
    {
      "epoch": 0.18371805510864228,
      "grad_norm": 0.2897915244102478,
      "learning_rate": 1.0003575118923961e-05,
      "loss": 1.8013,
      "step": 18390
    },
    {
      "epoch": 0.18425999627415449,
      "grad_norm": 0.2939542233943939,
      "learning_rate": 1.0002514869805599e-05,
      "loss": 1.8084,
      "step": 18400
    },
    {
      "epoch": 0.18480193743966672,
      "grad_norm": 0.2901172637939453,
      "learning_rate": 1.0001640623040548e-05,
      "loss": 1.8072,
      "step": 18410
    },
    {
      "epoch": 0.18534387860517892,
      "grad_norm": 0.4139680564403534,
      "learning_rate": 1.0000952382242605e-05,
      "loss": 1.798,
      "step": 18420
    },
    {
      "epoch": 0.18588581977069116,
      "grad_norm": 0.2750343978404999,
      "learning_rate": 1.000045015025669e-05,
      "loss": 1.7949,
      "step": 18430
    },
    {
      "epoch": 0.18642776093620336,
      "grad_norm": 0.28022781014442444,
      "learning_rate": 1.0000133929158849e-05,
      "loss": 1.8095,
      "step": 18440
    },
    {
      "epoch": 0.18664453740240824,
      "eval_loss": 2.5955660343170166,
      "eval_runtime": 23.3631,
      "eval_samples_per_second": 214.012,
      "eval_steps_per_second": 1.156,
      "step": 18444
    },
    {
      "epoch": 0.1869697021017156,
      "grad_norm": 0.26084405183792114,
      "learning_rate": 1.000000372025621e-05,
      "loss": 1.815,
      "step": 18450
    },
    {
      "epoch": 1.0004335529324098,
      "grad_norm": 7.159980297088623,
      "learning_rate": 9.350101516641391e-05,
      "loss": 1.9598,
      "step": 18460
    },
    {
      "epoch": 1.000975494097922,
      "grad_norm": 6.029938697814941,
      "learning_rate": 9.349281376788221e-05,
      "loss": 2.0867,
      "step": 18470
    },
    {
      "epoch": 1.0015174352634342,
      "grad_norm": 4.406125068664551,
      "learning_rate": 9.348460760102786e-05,
      "loss": 2.0069,
      "step": 18480
    },
    {
      "epoch": 1.0020593764289465,
      "grad_norm": 4.879397392272949,
      "learning_rate": 9.347639666686739e-05,
      "loss": 2.0173,
      "step": 18490
    },
    {
      "epoch": 1.0026013175944586,
      "grad_norm": 3.420335292816162,
      "learning_rate": 9.346818096641794e-05,
      "loss": 1.9867,
      "step": 18500
    },
    {
      "epoch": 1.003143258759971,
      "grad_norm": 1.2602205276489258,
      "learning_rate": 9.345996050069725e-05,
      "loss": 1.9279,
      "step": 18510
    },
    {
      "epoch": 1.003685199925483,
      "grad_norm": 3.904205560684204,
      "learning_rate": 9.345173527072363e-05,
      "loss": 1.9165,
      "step": 18520
    },
    {
      "epoch": 1.0042271410909953,
      "grad_norm": 1.9543260335922241,
      "learning_rate": 9.344350527751599e-05,
      "loss": 1.9421,
      "step": 18530
    },
    {
      "epoch": 1.0042813352075466,
      "eval_loss": 2.611534595489502,
      "eval_runtime": 59.2714,
      "eval_samples_per_second": 84.358,
      "eval_steps_per_second": 0.456,
      "step": 18531
    },
    {
      "epoch": 1.0047690822565076,
      "grad_norm": 1.9394625425338745,
      "learning_rate": 9.343527052209384e-05,
      "loss": 1.8943,
      "step": 18540
    },
    {
      "epoch": 1.0053110234220197,
      "grad_norm": 1.3751671314239502,
      "learning_rate": 9.342703100547722e-05,
      "loss": 1.8667,
      "step": 18550
    },
    {
      "epoch": 1.005852964587532,
      "grad_norm": 1.4693511724472046,
      "learning_rate": 9.341878672868683e-05,
      "loss": 1.8609,
      "step": 18560
    },
    {
      "epoch": 1.0063949057530441,
      "grad_norm": 1.3482840061187744,
      "learning_rate": 9.341053769274396e-05,
      "loss": 1.8552,
      "step": 18570
    },
    {
      "epoch": 1.0069368469185564,
      "grad_norm": 1.061438798904419,
      "learning_rate": 9.340228389867041e-05,
      "loss": 1.8599,
      "step": 18580
    },
    {
      "epoch": 1.0074787880840685,
      "grad_norm": 0.778508722782135,
      "learning_rate": 9.339402534748868e-05,
      "loss": 1.8528,
      "step": 18590
    },
    {
      "epoch": 1.0080207292495809,
      "grad_norm": 0.43868595361709595,
      "learning_rate": 9.338576204022178e-05,
      "loss": 1.8457,
      "step": 18600
    },
    {
      "epoch": 1.008562670415093,
      "grad_norm": 0.531323254108429,
      "learning_rate": 9.337749397789332e-05,
      "loss": 1.8489,
      "step": 18610
    },
    {
      "epoch": 1.008996223347503,
      "eval_loss": 2.614398717880249,
      "eval_runtime": 21.9914,
      "eval_samples_per_second": 227.361,
      "eval_steps_per_second": 1.228,
      "step": 18618
    },
    {
      "epoch": 1.0091046115806053,
      "grad_norm": 0.6396978497505188,
      "learning_rate": 9.336922116152755e-05,
      "loss": 1.844,
      "step": 18620
    },
    {
      "epoch": 1.0096465527461176,
      "grad_norm": 0.44661280512809753,
      "learning_rate": 9.336094359214923e-05,
      "loss": 1.8426,
      "step": 18630
    },
    {
      "epoch": 1.0101884939116297,
      "grad_norm": 0.55703204870224,
      "learning_rate": 9.335266127078376e-05,
      "loss": 1.8513,
      "step": 18640
    },
    {
      "epoch": 1.010730435077142,
      "grad_norm": 0.8314570784568787,
      "learning_rate": 9.334437419845713e-05,
      "loss": 1.8503,
      "step": 18650
    },
    {
      "epoch": 1.011272376242654,
      "grad_norm": 0.5193946957588196,
      "learning_rate": 9.333608237619589e-05,
      "loss": 1.8619,
      "step": 18660
    },
    {
      "epoch": 1.0118143174081664,
      "grad_norm": 0.3355676829814911,
      "learning_rate": 9.33277858050272e-05,
      "loss": 1.8414,
      "step": 18670
    },
    {
      "epoch": 1.0123562585736785,
      "grad_norm": 0.4459117650985718,
      "learning_rate": 9.331948448597882e-05,
      "loss": 1.8392,
      "step": 18680
    },
    {
      "epoch": 1.0128981997391908,
      "grad_norm": 0.7514176368713379,
      "learning_rate": 9.331117842007906e-05,
      "loss": 1.842,
      "step": 18690
    },
    {
      "epoch": 1.0134401409047031,
      "grad_norm": 0.4279676377773285,
      "learning_rate": 9.330286760835686e-05,
      "loss": 1.8378,
      "step": 18700
    },
    {
      "epoch": 1.013711111487459,
      "eval_loss": 2.6156516075134277,
      "eval_runtime": 21.9911,
      "eval_samples_per_second": 227.364,
      "eval_steps_per_second": 1.228,
      "step": 18705
    },
    {
      "epoch": 1.0139820820702152,
      "grad_norm": 0.43919897079467773,
      "learning_rate": 9.32945520518417e-05,
      "loss": 1.8378,
      "step": 18710
    },
    {
      "epoch": 1.0145240232357275,
      "grad_norm": 0.4081036150455475,
      "learning_rate": 9.32862317515637e-05,
      "loss": 1.8381,
      "step": 18720
    },
    {
      "epoch": 1.0150659644012396,
      "grad_norm": 1.0370370149612427,
      "learning_rate": 9.327790670855352e-05,
      "loss": 1.8482,
      "step": 18730
    },
    {
      "epoch": 1.015607905566752,
      "grad_norm": 0.717692494392395,
      "learning_rate": 9.326957692384247e-05,
      "loss": 1.8335,
      "step": 18740
    },
    {
      "epoch": 1.016149846732264,
      "grad_norm": 0.5903939604759216,
      "learning_rate": 9.326124239846237e-05,
      "loss": 1.8393,
      "step": 18750
    },
    {
      "epoch": 1.0166917878977764,
      "grad_norm": 0.5651223063468933,
      "learning_rate": 9.325290313344565e-05,
      "loss": 1.8397,
      "step": 18760
    },
    {
      "epoch": 1.0172337290632885,
      "grad_norm": 0.7575017809867859,
      "learning_rate": 9.32445591298254e-05,
      "loss": 1.8482,
      "step": 18770
    },
    {
      "epoch": 1.0177756702288008,
      "grad_norm": 1.311630129814148,
      "learning_rate": 9.32362103886352e-05,
      "loss": 1.844,
      "step": 18780
    },
    {
      "epoch": 1.018317611394313,
      "grad_norm": 1.2291172742843628,
      "learning_rate": 9.322785691090925e-05,
      "loss": 1.8475,
      "step": 18790
    },
    {
      "epoch": 1.0184259996274154,
      "eval_loss": 2.618657350540161,
      "eval_runtime": 21.9861,
      "eval_samples_per_second": 227.417,
      "eval_steps_per_second": 1.228,
      "step": 18792
    },
    {
      "epoch": 1.0188595525598252,
      "grad_norm": 0.6886726021766663,
      "learning_rate": 9.321949869768236e-05,
      "loss": 1.8358,
      "step": 18800
    },
    {
      "epoch": 1.0194014937253375,
      "grad_norm": 0.8392327427864075,
      "learning_rate": 9.32111357499899e-05,
      "loss": 1.8462,
      "step": 18810
    },
    {
      "epoch": 1.0199434348908496,
      "grad_norm": 0.35495704412460327,
      "learning_rate": 9.320276806886784e-05,
      "loss": 1.834,
      "step": 18820
    },
    {
      "epoch": 1.020485376056362,
      "grad_norm": 0.45716392993927,
      "learning_rate": 9.319439565535274e-05,
      "loss": 1.8272,
      "step": 18830
    },
    {
      "epoch": 1.021027317221874,
      "grad_norm": 0.31467896699905396,
      "learning_rate": 9.318601851048172e-05,
      "loss": 1.8398,
      "step": 18840
    },
    {
      "epoch": 1.0215692583873863,
      "grad_norm": 0.7633432149887085,
      "learning_rate": 9.31776366352925e-05,
      "loss": 1.839,
      "step": 18850
    },
    {
      "epoch": 1.0221111995528986,
      "grad_norm": 0.40369927883148193,
      "learning_rate": 9.316925003082344e-05,
      "loss": 1.8429,
      "step": 18860
    },
    {
      "epoch": 1.0226531407184107,
      "grad_norm": 0.7707934975624084,
      "learning_rate": 9.316085869811338e-05,
      "loss": 1.8314,
      "step": 18870
    },
    {
      "epoch": 1.0231408877673718,
      "eval_loss": 2.6104660034179688,
      "eval_runtime": 21.9934,
      "eval_samples_per_second": 227.341,
      "eval_steps_per_second": 1.228,
      "step": 18879
    },
    {
      "epoch": 1.023195081883923,
      "grad_norm": 0.791067361831665,
      "learning_rate": 9.315246263820181e-05,
      "loss": 1.8436,
      "step": 18880
    },
    {
      "epoch": 1.0237370230494351,
      "grad_norm": 0.3280937671661377,
      "learning_rate": 9.314406185212883e-05,
      "loss": 1.8302,
      "step": 18890
    },
    {
      "epoch": 1.0242789642149475,
      "grad_norm": 0.4388424754142761,
      "learning_rate": 9.313565634093507e-05,
      "loss": 1.8393,
      "step": 18900
    },
    {
      "epoch": 1.0248209053804596,
      "grad_norm": 0.4343729019165039,
      "learning_rate": 9.312724610566177e-05,
      "loss": 1.8377,
      "step": 18910
    },
    {
      "epoch": 1.0253628465459719,
      "grad_norm": 0.7564290761947632,
      "learning_rate": 9.311883114735077e-05,
      "loss": 1.8383,
      "step": 18920
    },
    {
      "epoch": 1.025904787711484,
      "grad_norm": 0.3451434075832367,
      "learning_rate": 9.311041146704443e-05,
      "loss": 1.8312,
      "step": 18930
    },
    {
      "epoch": 1.0264467288769963,
      "grad_norm": 0.44154074788093567,
      "learning_rate": 9.31019870657858e-05,
      "loss": 1.832,
      "step": 18940
    },
    {
      "epoch": 1.0269886700425086,
      "grad_norm": 0.5464621782302856,
      "learning_rate": 9.309355794461843e-05,
      "loss": 1.832,
      "step": 18950
    },
    {
      "epoch": 1.0275306112080207,
      "grad_norm": 0.737061619758606,
      "learning_rate": 9.308512410458648e-05,
      "loss": 1.8379,
      "step": 18960
    },
    {
      "epoch": 1.0278557759073281,
      "eval_loss": 2.598432779312134,
      "eval_runtime": 21.9905,
      "eval_samples_per_second": 227.371,
      "eval_steps_per_second": 1.228,
      "step": 18966
    },
    {
      "epoch": 1.028072552373533,
      "grad_norm": 0.5460038185119629,
      "learning_rate": 9.307668554673471e-05,
      "loss": 1.8363,
      "step": 18970
    },
    {
      "epoch": 1.028614493539045,
      "grad_norm": 0.365409791469574,
      "learning_rate": 9.306824227210845e-05,
      "loss": 1.8447,
      "step": 18980
    },
    {
      "epoch": 1.0291564347045574,
      "grad_norm": 0.37528592348098755,
      "learning_rate": 9.305979428175361e-05,
      "loss": 1.8298,
      "step": 18990
    },
    {
      "epoch": 1.0296983758700695,
      "grad_norm": 0.37354302406311035,
      "learning_rate": 9.30513415767167e-05,
      "loss": 1.8439,
      "step": 19000
    },
    {
      "epoch": 1.0302403170355818,
      "grad_norm": 0.4735754430294037,
      "learning_rate": 9.304288415804479e-05,
      "loss": 1.8267,
      "step": 19010
    },
    {
      "epoch": 1.0307822582010941,
      "grad_norm": 0.30185240507125854,
      "learning_rate": 9.303442202678555e-05,
      "loss": 1.8362,
      "step": 19020
    },
    {
      "epoch": 1.0313241993666062,
      "grad_norm": 0.3827216625213623,
      "learning_rate": 9.302595518398723e-05,
      "loss": 1.8311,
      "step": 19030
    },
    {
      "epoch": 1.0318661405321186,
      "grad_norm": 0.3216041624546051,
      "learning_rate": 9.301748363069872e-05,
      "loss": 1.8422,
      "step": 19040
    },
    {
      "epoch": 1.0324080816976307,
      "grad_norm": 0.34851735830307007,
      "learning_rate": 9.300900736796934e-05,
      "loss": 1.8369,
      "step": 19050
    },
    {
      "epoch": 1.0325706640472845,
      "eval_loss": 2.6064109802246094,
      "eval_runtime": 21.9894,
      "eval_samples_per_second": 227.383,
      "eval_steps_per_second": 1.228,
      "step": 19053
    },
    {
      "epoch": 1.032950022863143,
      "grad_norm": 0.6215195059776306,
      "learning_rate": 9.300052639684918e-05,
      "loss": 1.834,
      "step": 19060
    },
    {
      "epoch": 1.033491964028655,
      "grad_norm": 0.6960977911949158,
      "learning_rate": 9.299204071838878e-05,
      "loss": 1.8224,
      "step": 19070
    },
    {
      "epoch": 1.0340339051941674,
      "grad_norm": 0.680069088935852,
      "learning_rate": 9.298355033363931e-05,
      "loss": 1.8277,
      "step": 19080
    },
    {
      "epoch": 1.0345758463596795,
      "grad_norm": 0.8072498440742493,
      "learning_rate": 9.297505524365254e-05,
      "loss": 1.8475,
      "step": 19090
    },
    {
      "epoch": 1.0351177875251918,
      "grad_norm": 0.547581672668457,
      "learning_rate": 9.296655544948081e-05,
      "loss": 1.8269,
      "step": 19100
    },
    {
      "epoch": 1.035659728690704,
      "grad_norm": 0.33486223220825195,
      "learning_rate": 9.295805095217701e-05,
      "loss": 1.8352,
      "step": 19110
    },
    {
      "epoch": 1.0362016698562162,
      "grad_norm": 0.5130099654197693,
      "learning_rate": 9.294954175279466e-05,
      "loss": 1.8332,
      "step": 19120
    },
    {
      "epoch": 1.0367436110217285,
      "grad_norm": 0.5673821568489075,
      "learning_rate": 9.294102785238785e-05,
      "loss": 1.8172,
      "step": 19130
    },
    {
      "epoch": 1.0372855521872406,
      "grad_norm": 0.5935150980949402,
      "learning_rate": 9.293250925201122e-05,
      "loss": 1.83,
      "step": 19140
    },
    {
      "epoch": 1.0372855521872406,
      "eval_loss": 2.619783878326416,
      "eval_runtime": 21.976,
      "eval_samples_per_second": 227.521,
      "eval_steps_per_second": 1.229,
      "step": 19140
    },
    {
      "epoch": 1.037827493352753,
      "grad_norm": 0.8336159586906433,
      "learning_rate": 9.292398595272003e-05,
      "loss": 1.8289,
      "step": 19150
    },
    {
      "epoch": 1.038369434518265,
      "grad_norm": 0.7801002264022827,
      "learning_rate": 9.291545795557011e-05,
      "loss": 1.8325,
      "step": 19160
    },
    {
      "epoch": 1.0389113756837773,
      "grad_norm": 0.3641831576824188,
      "learning_rate": 9.290692526161787e-05,
      "loss": 1.8289,
      "step": 19170
    },
    {
      "epoch": 1.0394533168492897,
      "grad_norm": 0.34061098098754883,
      "learning_rate": 9.289838787192032e-05,
      "loss": 1.8234,
      "step": 19180
    },
    {
      "epoch": 1.0399952580148017,
      "grad_norm": 0.31556975841522217,
      "learning_rate": 9.288984578753502e-05,
      "loss": 1.8252,
      "step": 19190
    },
    {
      "epoch": 1.040537199180314,
      "grad_norm": 0.3090665638446808,
      "learning_rate": 9.28812990095201e-05,
      "loss": 1.8194,
      "step": 19200
    },
    {
      "epoch": 1.0410791403458262,
      "grad_norm": 0.29080694913864136,
      "learning_rate": 9.287274753893435e-05,
      "loss": 1.8342,
      "step": 19210
    },
    {
      "epoch": 1.0416210815113385,
      "grad_norm": 0.5005635619163513,
      "learning_rate": 9.286419137683704e-05,
      "loss": 1.8401,
      "step": 19220
    },
    {
      "epoch": 1.042000440327197,
      "eval_loss": 2.6065304279327393,
      "eval_runtime": 21.9863,
      "eval_samples_per_second": 227.414,
      "eval_steps_per_second": 1.228,
      "step": 19227
    },
    {
      "epoch": 1.0421630226768506,
      "grad_norm": 0.6064873933792114,
      "learning_rate": 9.28556305242881e-05,
      "loss": 1.822,
      "step": 19230
    },
    {
      "epoch": 1.0427049638423629,
      "grad_norm": 0.30239003896713257,
      "learning_rate": 9.284706498234798e-05,
      "loss": 1.8317,
      "step": 19240
    },
    {
      "epoch": 1.043246905007875,
      "grad_norm": 0.45904672145843506,
      "learning_rate": 9.28384947520778e-05,
      "loss": 1.8207,
      "step": 19250
    },
    {
      "epoch": 1.0437888461733873,
      "grad_norm": 0.4452885389328003,
      "learning_rate": 9.282991983453917e-05,
      "loss": 1.832,
      "step": 19260
    },
    {
      "epoch": 1.0443307873388996,
      "grad_norm": 0.5682956576347351,
      "learning_rate": 9.28213402307943e-05,
      "loss": 1.8331,
      "step": 19270
    },
    {
      "epoch": 1.0448727285044117,
      "grad_norm": 0.886755108833313,
      "learning_rate": 9.281275594190599e-05,
      "loss": 1.8377,
      "step": 19280
    },
    {
      "epoch": 1.045414669669924,
      "grad_norm": 0.9180141687393188,
      "learning_rate": 9.280416696893766e-05,
      "loss": 1.8277,
      "step": 19290
    },
    {
      "epoch": 1.0459566108354361,
      "grad_norm": 0.3015201687812805,
      "learning_rate": 9.279557331295325e-05,
      "loss": 1.8243,
      "step": 19300
    },
    {
      "epoch": 1.0464985520009484,
      "grad_norm": 0.36977553367614746,
      "learning_rate": 9.278697497501731e-05,
      "loss": 1.8161,
      "step": 19310
    },
    {
      "epoch": 1.0467153284671533,
      "eval_loss": 2.6029727458953857,
      "eval_runtime": 21.9909,
      "eval_samples_per_second": 227.367,
      "eval_steps_per_second": 1.228,
      "step": 19314
    },
    {
      "epoch": 1.0470404931664605,
      "grad_norm": 0.7295284867286682,
      "learning_rate": 9.277837195619496e-05,
      "loss": 1.8169,
      "step": 19320
    },
    {
      "epoch": 1.0475824343319728,
      "grad_norm": 0.3548785448074341,
      "learning_rate": 9.276976425755192e-05,
      "loss": 1.8258,
      "step": 19330
    },
    {
      "epoch": 1.0481243754974852,
      "grad_norm": 0.581323504447937,
      "learning_rate": 9.276115188015445e-05,
      "loss": 1.8386,
      "step": 19340
    },
    {
      "epoch": 1.0486663166629973,
      "grad_norm": 0.33570852875709534,
      "learning_rate": 9.275253482506945e-05,
      "loss": 1.8224,
      "step": 19350
    },
    {
      "epoch": 1.0492082578285096,
      "grad_norm": 0.3555627167224884,
      "learning_rate": 9.274391309336432e-05,
      "loss": 1.8235,
      "step": 19360
    },
    {
      "epoch": 1.0497501989940217,
      "grad_norm": 0.31758037209510803,
      "learning_rate": 9.27352866861071e-05,
      "loss": 1.8283,
      "step": 19370
    },
    {
      "epoch": 1.050292140159534,
      "grad_norm": 0.3359634280204773,
      "learning_rate": 9.272665560436642e-05,
      "loss": 1.8207,
      "step": 19380
    },
    {
      "epoch": 1.050834081325046,
      "grad_norm": 0.2850840389728546,
      "learning_rate": 9.271801984921142e-05,
      "loss": 1.8291,
      "step": 19390
    },
    {
      "epoch": 1.0513760224905584,
      "grad_norm": 0.29905441403388977,
      "learning_rate": 9.270937942171189e-05,
      "loss": 1.8188,
      "step": 19400
    },
    {
      "epoch": 1.0514302166071097,
      "eval_loss": 2.616773843765259,
      "eval_runtime": 22.0168,
      "eval_samples_per_second": 227.1,
      "eval_steps_per_second": 1.226,
      "step": 19401
    },
    {
      "epoch": 1.0519179636560705,
      "grad_norm": 0.3919777274131775,
      "learning_rate": 9.270073432293814e-05,
      "loss": 1.8129,
      "step": 19410
    },
    {
      "epoch": 1.0524599048215828,
      "grad_norm": 0.37216314673423767,
      "learning_rate": 9.26920845539611e-05,
      "loss": 1.8217,
      "step": 19420
    },
    {
      "epoch": 1.0530018459870951,
      "grad_norm": 0.3143322765827179,
      "learning_rate": 9.268343011585227e-05,
      "loss": 1.8239,
      "step": 19430
    },
    {
      "epoch": 1.0535437871526072,
      "grad_norm": 0.6800436973571777,
      "learning_rate": 9.267477100968372e-05,
      "loss": 1.8202,
      "step": 19440
    },
    {
      "epoch": 1.0540857283181195,
      "grad_norm": 0.35689041018486023,
      "learning_rate": 9.266610723652813e-05,
      "loss": 1.82,
      "step": 19450
    },
    {
      "epoch": 1.0546276694836316,
      "grad_norm": 0.29921287298202515,
      "learning_rate": 9.265743879745867e-05,
      "loss": 1.8124,
      "step": 19460
    },
    {
      "epoch": 1.055169610649144,
      "grad_norm": 0.47158485651016235,
      "learning_rate": 9.264876569354921e-05,
      "loss": 1.8158,
      "step": 19470
    },
    {
      "epoch": 1.055711551814656,
      "grad_norm": 0.32789716124534607,
      "learning_rate": 9.26400879258741e-05,
      "loss": 1.8313,
      "step": 19480
    },
    {
      "epoch": 1.0561451047470658,
      "eval_loss": 2.605008840560913,
      "eval_runtime": 21.9889,
      "eval_samples_per_second": 227.388,
      "eval_steps_per_second": 1.228,
      "step": 19488
    },
    {
      "epoch": 1.0562534929801684,
      "grad_norm": 0.3259328007698059,
      "learning_rate": 9.263140549550832e-05,
      "loss": 1.8178,
      "step": 19490
    },
    {
      "epoch": 1.0567954341456804,
      "grad_norm": 0.27712419629096985,
      "learning_rate": 9.26227184035274e-05,
      "loss": 1.8237,
      "step": 19500
    },
    {
      "epoch": 1.0573373753111928,
      "grad_norm": 0.7220262289047241,
      "learning_rate": 9.261402665100746e-05,
      "loss": 1.8208,
      "step": 19510
    },
    {
      "epoch": 1.057879316476705,
      "grad_norm": 0.4974667429924011,
      "learning_rate": 9.260533023902522e-05,
      "loss": 1.8214,
      "step": 19520
    },
    {
      "epoch": 1.0584212576422172,
      "grad_norm": 0.4250205457210541,
      "learning_rate": 9.259662916865792e-05,
      "loss": 1.8159,
      "step": 19530
    },
    {
      "epoch": 1.0589631988077295,
      "grad_norm": 0.4620699882507324,
      "learning_rate": 9.258792344098344e-05,
      "loss": 1.806,
      "step": 19540
    },
    {
      "epoch": 1.0595051399732416,
      "grad_norm": 0.36918604373931885,
      "learning_rate": 9.257921305708018e-05,
      "loss": 1.8078,
      "step": 19550
    },
    {
      "epoch": 1.060047081138754,
      "grad_norm": 0.3109738528728485,
      "learning_rate": 9.257049801802716e-05,
      "loss": 1.8172,
      "step": 19560
    },
    {
      "epoch": 1.060589022304266,
      "grad_norm": 0.5203297138214111,
      "learning_rate": 9.256177832490398e-05,
      "loss": 1.816,
      "step": 19570
    },
    {
      "epoch": 1.0608599928870222,
      "eval_loss": 2.603193759918213,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.323,
      "eval_steps_per_second": 1.228,
      "step": 19575
    },
    {
      "epoch": 1.0611309634697783,
      "grad_norm": 0.29931849241256714,
      "learning_rate": 9.255305397879076e-05,
      "loss": 1.8155,
      "step": 19580
    },
    {
      "epoch": 1.0616729046352906,
      "grad_norm": 0.3699570298194885,
      "learning_rate": 9.254432498076826e-05,
      "loss": 1.8154,
      "step": 19590
    },
    {
      "epoch": 1.0622148458008027,
      "grad_norm": 0.2735179364681244,
      "learning_rate": 9.253559133191779e-05,
      "loss": 1.8209,
      "step": 19600
    },
    {
      "epoch": 1.062756786966315,
      "grad_norm": 0.31726008653640747,
      "learning_rate": 9.252685303332123e-05,
      "loss": 1.8278,
      "step": 19610
    },
    {
      "epoch": 1.0632987281318271,
      "grad_norm": 0.6221383213996887,
      "learning_rate": 9.251811008606102e-05,
      "loss": 1.825,
      "step": 19620
    },
    {
      "epoch": 1.0638406692973394,
      "grad_norm": 0.8196062445640564,
      "learning_rate": 9.250936249122023e-05,
      "loss": 1.8196,
      "step": 19630
    },
    {
      "epoch": 1.0643826104628515,
      "grad_norm": 0.2812905013561249,
      "learning_rate": 9.250061024988246e-05,
      "loss": 1.8123,
      "step": 19640
    },
    {
      "epoch": 1.0649245516283639,
      "grad_norm": 0.4467778205871582,
      "learning_rate": 9.249185336313191e-05,
      "loss": 1.8098,
      "step": 19650
    },
    {
      "epoch": 1.0654664927938762,
      "grad_norm": 0.36650586128234863,
      "learning_rate": 9.248309183205334e-05,
      "loss": 1.8245,
      "step": 19660
    },
    {
      "epoch": 1.0655748810269785,
      "eval_loss": 2.5961737632751465,
      "eval_runtime": 21.9861,
      "eval_samples_per_second": 227.416,
      "eval_steps_per_second": 1.228,
      "step": 19662
    },
    {
      "epoch": 1.0660084339593883,
      "grad_norm": 0.351523220539093,
      "learning_rate": 9.247432565773209e-05,
      "loss": 1.8098,
      "step": 19670
    },
    {
      "epoch": 1.0665503751249006,
      "grad_norm": 0.4025458097457886,
      "learning_rate": 9.246555484125407e-05,
      "loss": 1.8208,
      "step": 19680
    },
    {
      "epoch": 1.0670923162904127,
      "grad_norm": 0.5277726650238037,
      "learning_rate": 9.245677938370578e-05,
      "loss": 1.8235,
      "step": 19690
    },
    {
      "epoch": 1.067634257455925,
      "grad_norm": 0.3863029181957245,
      "learning_rate": 9.244799928617427e-05,
      "loss": 1.8074,
      "step": 19700
    },
    {
      "epoch": 1.068176198621437,
      "grad_norm": 0.4278218746185303,
      "learning_rate": 9.24392145497472e-05,
      "loss": 1.8179,
      "step": 19710
    },
    {
      "epoch": 1.0687181397869494,
      "grad_norm": 0.30992093682289124,
      "learning_rate": 9.243042517551277e-05,
      "loss": 1.8083,
      "step": 19720
    },
    {
      "epoch": 1.0692600809524615,
      "grad_norm": 0.92547607421875,
      "learning_rate": 9.242163116455979e-05,
      "loss": 1.8194,
      "step": 19730
    },
    {
      "epoch": 1.0698020221179738,
      "grad_norm": 0.297558069229126,
      "learning_rate": 9.241283251797759e-05,
      "loss": 1.8245,
      "step": 19740
    },
    {
      "epoch": 1.0702897691669349,
      "eval_loss": 2.59582781791687,
      "eval_runtime": 21.9957,
      "eval_samples_per_second": 227.317,
      "eval_steps_per_second": 1.228,
      "step": 19749
    },
    {
      "epoch": 1.0703439632834861,
      "grad_norm": 0.34201860427856445,
      "learning_rate": 9.240402923685613e-05,
      "loss": 1.8202,
      "step": 19750
    },
    {
      "epoch": 1.0708859044489982,
      "grad_norm": 0.5197858810424805,
      "learning_rate": 9.239522132228594e-05,
      "loss": 1.8104,
      "step": 19760
    },
    {
      "epoch": 1.0714278456145105,
      "grad_norm": 0.5215007662773132,
      "learning_rate": 9.238640877535809e-05,
      "loss": 1.8128,
      "step": 19770
    },
    {
      "epoch": 1.0719697867800226,
      "grad_norm": 0.3624344766139984,
      "learning_rate": 9.237759159716421e-05,
      "loss": 1.8203,
      "step": 19780
    },
    {
      "epoch": 1.072511727945535,
      "grad_norm": 0.3879830241203308,
      "learning_rate": 9.236876978879657e-05,
      "loss": 1.8119,
      "step": 19790
    },
    {
      "epoch": 1.073053669111047,
      "grad_norm": 0.3035085201263428,
      "learning_rate": 9.235994335134798e-05,
      "loss": 1.8118,
      "step": 19800
    },
    {
      "epoch": 1.0735956102765594,
      "grad_norm": 1.0123307704925537,
      "learning_rate": 9.23511122859118e-05,
      "loss": 1.8119,
      "step": 19810
    },
    {
      "epoch": 1.0741375514420715,
      "grad_norm": 0.46915149688720703,
      "learning_rate": 9.234227659358197e-05,
      "loss": 1.8163,
      "step": 19820
    },
    {
      "epoch": 1.0746794926075838,
      "grad_norm": 0.48051413893699646,
      "learning_rate": 9.233343627545307e-05,
      "loss": 1.8178,
      "step": 19830
    },
    {
      "epoch": 1.0750046573068912,
      "eval_loss": 2.6102840900421143,
      "eval_runtime": 21.9902,
      "eval_samples_per_second": 227.374,
      "eval_steps_per_second": 1.228,
      "step": 19836
    },
    {
      "epoch": 1.075221433773096,
      "grad_norm": 0.7908685207366943,
      "learning_rate": 9.232459133262016e-05,
      "loss": 1.8174,
      "step": 19840
    },
    {
      "epoch": 1.0757633749386082,
      "grad_norm": 0.4490140974521637,
      "learning_rate": 9.231574176617893e-05,
      "loss": 1.821,
      "step": 19850
    },
    {
      "epoch": 1.0763053161041205,
      "grad_norm": 0.6310442686080933,
      "learning_rate": 9.230688757722562e-05,
      "loss": 1.8002,
      "step": 19860
    },
    {
      "epoch": 1.0768472572696326,
      "grad_norm": 0.47709372639656067,
      "learning_rate": 9.229802876685702e-05,
      "loss": 1.817,
      "step": 19870
    },
    {
      "epoch": 1.077389198435145,
      "grad_norm": 0.3235110640525818,
      "learning_rate": 9.228916533617057e-05,
      "loss": 1.8022,
      "step": 19880
    },
    {
      "epoch": 1.077931139600657,
      "grad_norm": 0.2803106904029846,
      "learning_rate": 9.228029728626421e-05,
      "loss": 1.8075,
      "step": 19890
    },
    {
      "epoch": 1.0784730807661693,
      "grad_norm": 0.290698766708374,
      "learning_rate": 9.227142461823648e-05,
      "loss": 1.8064,
      "step": 19900
    },
    {
      "epoch": 1.0790150219316816,
      "grad_norm": 0.7563320398330688,
      "learning_rate": 9.226254733318648e-05,
      "loss": 1.8079,
      "step": 19910
    },
    {
      "epoch": 1.0795569630971937,
      "grad_norm": 0.4314441382884979,
      "learning_rate": 9.22536654322139e-05,
      "loss": 1.7955,
      "step": 19920
    },
    {
      "epoch": 1.0797195454468473,
      "eval_loss": 2.596864938735962,
      "eval_runtime": 21.9923,
      "eval_samples_per_second": 227.352,
      "eval_steps_per_second": 1.228,
      "step": 19923
    },
    {
      "epoch": 1.080098904262706,
      "grad_norm": 0.30512261390686035,
      "learning_rate": 9.224477891641897e-05,
      "loss": 1.8072,
      "step": 19930
    },
    {
      "epoch": 1.0806408454282181,
      "grad_norm": 0.5094988346099854,
      "learning_rate": 9.223588778690255e-05,
      "loss": 1.8176,
      "step": 19940
    },
    {
      "epoch": 1.0811827865937305,
      "grad_norm": 0.47923043370246887,
      "learning_rate": 9.222699204476599e-05,
      "loss": 1.8203,
      "step": 19950
    },
    {
      "epoch": 1.0817247277592426,
      "grad_norm": 0.3491983711719513,
      "learning_rate": 9.221809169111129e-05,
      "loss": 1.8097,
      "step": 19960
    },
    {
      "epoch": 1.0822666689247549,
      "grad_norm": 0.5780982971191406,
      "learning_rate": 9.220918672704099e-05,
      "loss": 1.8065,
      "step": 19970
    },
    {
      "epoch": 1.0828086100902672,
      "grad_norm": 0.37517765164375305,
      "learning_rate": 9.220027715365817e-05,
      "loss": 1.8102,
      "step": 19980
    },
    {
      "epoch": 1.0833505512557793,
      "grad_norm": 0.582490086555481,
      "learning_rate": 9.219136297206652e-05,
      "loss": 1.8215,
      "step": 19990
    },
    {
      "epoch": 1.0838924924212916,
      "grad_norm": 0.5589390397071838,
      "learning_rate": 9.218244418337028e-05,
      "loss": 1.8206,
      "step": 20000
    },
    {
      "epoch": 1.0844344335868037,
      "grad_norm": 0.5232866406440735,
      "learning_rate": 9.217352078867431e-05,
      "loss": 1.8069,
      "step": 20010
    },
    {
      "epoch": 1.0844344335868037,
      "eval_loss": 2.6093404293060303,
      "eval_runtime": 21.9876,
      "eval_samples_per_second": 227.401,
      "eval_steps_per_second": 1.228,
      "step": 20010
    },
    {
      "epoch": 1.084976374752316,
      "grad_norm": 0.3265039026737213,
      "learning_rate": 9.216459278908398e-05,
      "loss": 1.7983,
      "step": 20020
    },
    {
      "epoch": 1.085518315917828,
      "grad_norm": 0.3922523558139801,
      "learning_rate": 9.215566018570523e-05,
      "loss": 1.7976,
      "step": 20030
    },
    {
      "epoch": 1.0860602570833404,
      "grad_norm": 0.35444027185440063,
      "learning_rate": 9.214672297964461e-05,
      "loss": 1.7977,
      "step": 20040
    },
    {
      "epoch": 1.0866021982488525,
      "grad_norm": 0.5408755540847778,
      "learning_rate": 9.213778117200926e-05,
      "loss": 1.8022,
      "step": 20050
    },
    {
      "epoch": 1.0871441394143648,
      "grad_norm": 0.5063872337341309,
      "learning_rate": 9.212883476390677e-05,
      "loss": 1.8088,
      "step": 20060
    },
    {
      "epoch": 1.0876860805798771,
      "grad_norm": 0.39630988240242004,
      "learning_rate": 9.211988375644543e-05,
      "loss": 1.7955,
      "step": 20070
    },
    {
      "epoch": 1.0882280217453892,
      "grad_norm": 0.3347756564617157,
      "learning_rate": 9.211092815073407e-05,
      "loss": 1.7925,
      "step": 20080
    },
    {
      "epoch": 1.0887699629109016,
      "grad_norm": 0.3801024854183197,
      "learning_rate": 9.210196794788203e-05,
      "loss": 1.8096,
      "step": 20090
    },
    {
      "epoch": 1.08914932172676,
      "eval_loss": 2.6098368167877197,
      "eval_runtime": 21.9921,
      "eval_samples_per_second": 227.354,
      "eval_steps_per_second": 1.228,
      "step": 20097
    },
    {
      "epoch": 1.0893119040764137,
      "grad_norm": 0.4123273491859436,
      "learning_rate": 9.20930031489993e-05,
      "loss": 1.7988,
      "step": 20100
    },
    {
      "epoch": 1.089853845241926,
      "grad_norm": 0.6030639410018921,
      "learning_rate": 9.208403375519637e-05,
      "loss": 1.7927,
      "step": 20110
    },
    {
      "epoch": 1.090395786407438,
      "grad_norm": 0.3275871276855469,
      "learning_rate": 9.207505976758434e-05,
      "loss": 1.7937,
      "step": 20120
    },
    {
      "epoch": 1.0909377275729504,
      "grad_norm": 0.4877399206161499,
      "learning_rate": 9.206608118727488e-05,
      "loss": 1.811,
      "step": 20130
    },
    {
      "epoch": 1.0914796687384625,
      "grad_norm": 0.2985506057739258,
      "learning_rate": 9.20570980153802e-05,
      "loss": 1.812,
      "step": 20140
    },
    {
      "epoch": 1.0920216099039748,
      "grad_norm": 0.29198411107063293,
      "learning_rate": 9.204811025301311e-05,
      "loss": 1.803,
      "step": 20150
    },
    {
      "epoch": 1.092563551069487,
      "grad_norm": 0.3116692006587982,
      "learning_rate": 9.203911790128696e-05,
      "loss": 1.8113,
      "step": 20160
    },
    {
      "epoch": 1.0931054922349992,
      "grad_norm": 0.3624420464038849,
      "learning_rate": 9.20301209613157e-05,
      "loss": 1.794,
      "step": 20170
    },
    {
      "epoch": 1.0936474334005115,
      "grad_norm": 0.5469610095024109,
      "learning_rate": 9.202111943421381e-05,
      "loss": 1.8038,
      "step": 20180
    },
    {
      "epoch": 1.0938642098667164,
      "eval_loss": 2.607426404953003,
      "eval_runtime": 21.9928,
      "eval_samples_per_second": 227.347,
      "eval_steps_per_second": 1.228,
      "step": 20184
    },
    {
      "epoch": 1.0941893745660236,
      "grad_norm": 0.5207812786102295,
      "learning_rate": 9.201211332109639e-05,
      "loss": 1.7874,
      "step": 20190
    },
    {
      "epoch": 1.094731315731536,
      "grad_norm": 0.3748219311237335,
      "learning_rate": 9.200310262307905e-05,
      "loss": 1.7988,
      "step": 20200
    },
    {
      "epoch": 1.095273256897048,
      "grad_norm": 0.34070441126823425,
      "learning_rate": 9.199408734127801e-05,
      "loss": 1.8057,
      "step": 20210
    },
    {
      "epoch": 1.0958151980625603,
      "grad_norm": 0.32609522342681885,
      "learning_rate": 9.198506747681005e-05,
      "loss": 1.8109,
      "step": 20220
    },
    {
      "epoch": 1.0963571392280727,
      "grad_norm": 0.4092683792114258,
      "learning_rate": 9.19760430307925e-05,
      "loss": 1.7954,
      "step": 20230
    },
    {
      "epoch": 1.0968990803935847,
      "grad_norm": 0.5064677000045776,
      "learning_rate": 9.196701400434327e-05,
      "loss": 1.8074,
      "step": 20240
    },
    {
      "epoch": 1.097441021559097,
      "grad_norm": 0.3224587142467499,
      "learning_rate": 9.195798039858084e-05,
      "loss": 1.7973,
      "step": 20250
    },
    {
      "epoch": 1.0979829627246092,
      "grad_norm": 0.5034688711166382,
      "learning_rate": 9.194894221462427e-05,
      "loss": 1.7994,
      "step": 20260
    },
    {
      "epoch": 1.0985249038901215,
      "grad_norm": 0.3772341310977936,
      "learning_rate": 9.193989945359314e-05,
      "loss": 1.7817,
      "step": 20270
    },
    {
      "epoch": 1.0985790980066727,
      "eval_loss": 2.6105637550354004,
      "eval_runtime": 21.9937,
      "eval_samples_per_second": 227.338,
      "eval_steps_per_second": 1.228,
      "step": 20271
    },
    {
      "epoch": 1.0990668450556336,
      "grad_norm": 0.8809638619422913,
      "learning_rate": 9.193085211660764e-05,
      "loss": 1.8056,
      "step": 20280
    },
    {
      "epoch": 1.0996087862211459,
      "grad_norm": 0.6711058020591736,
      "learning_rate": 9.192180020478852e-05,
      "loss": 1.7951,
      "step": 20290
    },
    {
      "epoch": 1.1001507273866582,
      "grad_norm": 0.8903740048408508,
      "learning_rate": 9.19127437192571e-05,
      "loss": 1.8076,
      "step": 20300
    },
    {
      "epoch": 1.1006926685521703,
      "grad_norm": 0.8000921010971069,
      "learning_rate": 9.190368266113524e-05,
      "loss": 1.811,
      "step": 20310
    },
    {
      "epoch": 1.1012346097176826,
      "grad_norm": 0.4838610589504242,
      "learning_rate": 9.189461703154538e-05,
      "loss": 1.809,
      "step": 20320
    },
    {
      "epoch": 1.1017765508831947,
      "grad_norm": 0.5514103174209595,
      "learning_rate": 9.188554683161056e-05,
      "loss": 1.7987,
      "step": 20330
    },
    {
      "epoch": 1.102318492048707,
      "grad_norm": 0.2846459448337555,
      "learning_rate": 9.187647206245434e-05,
      "loss": 1.8004,
      "step": 20340
    },
    {
      "epoch": 1.1028604332142191,
      "grad_norm": 0.45695552229881287,
      "learning_rate": 9.186739272520085e-05,
      "loss": 1.7999,
      "step": 20350
    },
    {
      "epoch": 1.1032939861466289,
      "eval_loss": 2.5855958461761475,
      "eval_runtime": 21.9945,
      "eval_samples_per_second": 227.33,
      "eval_steps_per_second": 1.228,
      "step": 20358
    },
    {
      "epoch": 1.1034023743797314,
      "grad_norm": 0.3189384937286377,
      "learning_rate": 9.185830882097482e-05,
      "loss": 1.8023,
      "step": 20360
    },
    {
      "epoch": 1.1039443155452435,
      "grad_norm": 0.5680913925170898,
      "learning_rate": 9.184922035090151e-05,
      "loss": 1.8008,
      "step": 20370
    },
    {
      "epoch": 1.1044862567107558,
      "grad_norm": 0.31937944889068604,
      "learning_rate": 9.184012731610676e-05,
      "loss": 1.7983,
      "step": 20380
    },
    {
      "epoch": 1.1050281978762682,
      "grad_norm": 0.9807746410369873,
      "learning_rate": 9.1831029717717e-05,
      "loss": 1.7981,
      "step": 20390
    },
    {
      "epoch": 1.1055701390417803,
      "grad_norm": 1.0870219469070435,
      "learning_rate": 9.182192755685917e-05,
      "loss": 1.7879,
      "step": 20400
    },
    {
      "epoch": 1.1061120802072926,
      "grad_norm": 0.45273005962371826,
      "learning_rate": 9.181282083466082e-05,
      "loss": 1.7992,
      "step": 20410
    },
    {
      "epoch": 1.1066540213728047,
      "grad_norm": 0.7691323757171631,
      "learning_rate": 9.180370955225006e-05,
      "loss": 1.7918,
      "step": 20420
    },
    {
      "epoch": 1.107195962538317,
      "grad_norm": 0.5909658670425415,
      "learning_rate": 9.179459371075554e-05,
      "loss": 1.802,
      "step": 20430
    },
    {
      "epoch": 1.107737903703829,
      "grad_norm": 0.3207308053970337,
      "learning_rate": 9.178547331130649e-05,
      "loss": 1.7934,
      "step": 20440
    },
    {
      "epoch": 1.1080088742865852,
      "eval_loss": 2.5942413806915283,
      "eval_runtime": 21.9938,
      "eval_samples_per_second": 227.337,
      "eval_steps_per_second": 1.228,
      "step": 20445
    },
    {
      "epoch": 1.1082798448693414,
      "grad_norm": 0.3061155378818512,
      "learning_rate": 9.177634835503272e-05,
      "loss": 1.7894,
      "step": 20450
    },
    {
      "epoch": 1.1088217860348535,
      "grad_norm": 0.35638898611068726,
      "learning_rate": 9.176721884306459e-05,
      "loss": 1.7893,
      "step": 20460
    },
    {
      "epoch": 1.1093637272003658,
      "grad_norm": 0.351608008146286,
      "learning_rate": 9.1758084776533e-05,
      "loss": 1.7951,
      "step": 20470
    },
    {
      "epoch": 1.1099056683658781,
      "grad_norm": 0.3291257321834564,
      "learning_rate": 9.174894615656948e-05,
      "loss": 1.7911,
      "step": 20480
    },
    {
      "epoch": 1.1104476095313902,
      "grad_norm": 0.2714458703994751,
      "learning_rate": 9.173980298430604e-05,
      "loss": 1.795,
      "step": 20490
    },
    {
      "epoch": 1.1109895506969025,
      "grad_norm": 0.3953942358493805,
      "learning_rate": 9.173065526087531e-05,
      "loss": 1.7912,
      "step": 20500
    },
    {
      "epoch": 1.1115314918624146,
      "grad_norm": 0.3956896662712097,
      "learning_rate": 9.17215029874105e-05,
      "loss": 1.7911,
      "step": 20510
    },
    {
      "epoch": 1.112073433027927,
      "grad_norm": 0.3793238401412964,
      "learning_rate": 9.17123461650453e-05,
      "loss": 1.7972,
      "step": 20520
    },
    {
      "epoch": 1.112615374193439,
      "grad_norm": 0.4785431921482086,
      "learning_rate": 9.170318479491406e-05,
      "loss": 1.7961,
      "step": 20530
    },
    {
      "epoch": 1.1127237624265416,
      "eval_loss": 2.584057331085205,
      "eval_runtime": 21.9901,
      "eval_samples_per_second": 227.375,
      "eval_steps_per_second": 1.228,
      "step": 20532
    },
    {
      "epoch": 1.1131573153589513,
      "grad_norm": 0.4006010890007019,
      "learning_rate": 9.169401887815164e-05,
      "loss": 1.7861,
      "step": 20540
    },
    {
      "epoch": 1.1136992565244637,
      "grad_norm": 0.2974483072757721,
      "learning_rate": 9.168484841589346e-05,
      "loss": 1.7885,
      "step": 20550
    },
    {
      "epoch": 1.1142411976899758,
      "grad_norm": 0.7173396348953247,
      "learning_rate": 9.167567340927552e-05,
      "loss": 1.8022,
      "step": 20560
    },
    {
      "epoch": 1.114783138855488,
      "grad_norm": 0.46824324131011963,
      "learning_rate": 9.166649385943441e-05,
      "loss": 1.7895,
      "step": 20570
    },
    {
      "epoch": 1.1153250800210002,
      "grad_norm": 0.6654873490333557,
      "learning_rate": 9.165730976750722e-05,
      "loss": 1.7898,
      "step": 20580
    },
    {
      "epoch": 1.1158670211865125,
      "grad_norm": 0.27924737334251404,
      "learning_rate": 9.164812113463165e-05,
      "loss": 1.783,
      "step": 20590
    },
    {
      "epoch": 1.1164089623520246,
      "grad_norm": 0.5301341414451599,
      "learning_rate": 9.163892796194594e-05,
      "loss": 1.7854,
      "step": 20600
    },
    {
      "epoch": 1.116950903517537,
      "grad_norm": 0.357898086309433,
      "learning_rate": 9.162973025058891e-05,
      "loss": 1.7834,
      "step": 20610
    },
    {
      "epoch": 1.117438650566498,
      "eval_loss": 2.5919032096862793,
      "eval_runtime": 21.993,
      "eval_samples_per_second": 227.345,
      "eval_steps_per_second": 1.228,
      "step": 20619
    },
    {
      "epoch": 1.1174928446830492,
      "grad_norm": 0.3866696059703827,
      "learning_rate": 9.162052800169992e-05,
      "loss": 1.7898,
      "step": 20620
    },
    {
      "epoch": 1.1180347858485613,
      "grad_norm": 0.30722272396087646,
      "learning_rate": 9.161132121641892e-05,
      "loss": 1.7837,
      "step": 20630
    },
    {
      "epoch": 1.1185767270140736,
      "grad_norm": 0.36760374903678894,
      "learning_rate": 9.160210989588639e-05,
      "loss": 1.7846,
      "step": 20640
    },
    {
      "epoch": 1.1191186681795857,
      "grad_norm": 0.48524531722068787,
      "learning_rate": 9.15928940412434e-05,
      "loss": 1.7822,
      "step": 20650
    },
    {
      "epoch": 1.119660609345098,
      "grad_norm": 0.41821709275245667,
      "learning_rate": 9.158367365363157e-05,
      "loss": 1.7902,
      "step": 20660
    },
    {
      "epoch": 1.1202025505106101,
      "grad_norm": 0.5472697615623474,
      "learning_rate": 9.157444873419307e-05,
      "loss": 1.7975,
      "step": 20670
    },
    {
      "epoch": 1.1207444916761224,
      "grad_norm": 0.3065122663974762,
      "learning_rate": 9.156521928407066e-05,
      "loss": 1.7851,
      "step": 20680
    },
    {
      "epoch": 1.1212864328416345,
      "grad_norm": 0.5732535719871521,
      "learning_rate": 9.155598530440763e-05,
      "loss": 1.7849,
      "step": 20690
    },
    {
      "epoch": 1.1218283740071469,
      "grad_norm": 0.3151445686817169,
      "learning_rate": 9.154674679634786e-05,
      "loss": 1.7931,
      "step": 20700
    },
    {
      "epoch": 1.122153538706454,
      "eval_loss": 2.5944461822509766,
      "eval_runtime": 21.9928,
      "eval_samples_per_second": 227.348,
      "eval_steps_per_second": 1.228,
      "step": 20706
    },
    {
      "epoch": 1.1223703151726592,
      "grad_norm": 0.3001176714897156,
      "learning_rate": 9.153750376103575e-05,
      "loss": 1.7924,
      "step": 20710
    },
    {
      "epoch": 1.1229122563381713,
      "grad_norm": 0.2985573709011078,
      "learning_rate": 9.152825619961633e-05,
      "loss": 1.7838,
      "step": 20720
    },
    {
      "epoch": 1.1234541975036836,
      "grad_norm": 0.3202672600746155,
      "learning_rate": 9.151900411323509e-05,
      "loss": 1.7899,
      "step": 20730
    },
    {
      "epoch": 1.1239961386691957,
      "grad_norm": 0.3624779284000397,
      "learning_rate": 9.150974750303818e-05,
      "loss": 1.779,
      "step": 20740
    },
    {
      "epoch": 1.124538079834708,
      "grad_norm": 0.47234293818473816,
      "learning_rate": 9.150048637017226e-05,
      "loss": 1.7977,
      "step": 20750
    },
    {
      "epoch": 1.12508002100022,
      "grad_norm": 0.4014720618724823,
      "learning_rate": 9.149122071578457e-05,
      "loss": 1.7887,
      "step": 20760
    },
    {
      "epoch": 1.1256219621657324,
      "grad_norm": 0.8474875688552856,
      "learning_rate": 9.148195054102289e-05,
      "loss": 1.7889,
      "step": 20770
    },
    {
      "epoch": 1.1261639033312445,
      "grad_norm": 0.3223975598812103,
      "learning_rate": 9.147267584703554e-05,
      "loss": 1.7931,
      "step": 20780
    },
    {
      "epoch": 1.1267058444967568,
      "grad_norm": 0.5686293840408325,
      "learning_rate": 9.146339663497148e-05,
      "loss": 1.7829,
      "step": 20790
    },
    {
      "epoch": 1.1268684268464104,
      "eval_loss": 2.596271276473999,
      "eval_runtime": 21.9897,
      "eval_samples_per_second": 227.38,
      "eval_steps_per_second": 1.228,
      "step": 20793
    },
    {
      "epoch": 1.1272477856622691,
      "grad_norm": 0.4797728359699249,
      "learning_rate": 9.145411290598013e-05,
      "loss": 1.7919,
      "step": 20800
    },
    {
      "epoch": 1.1277897268277812,
      "grad_norm": 0.33035770058631897,
      "learning_rate": 9.144482466121157e-05,
      "loss": 1.7981,
      "step": 20810
    },
    {
      "epoch": 1.1283316679932935,
      "grad_norm": 0.6441662311553955,
      "learning_rate": 9.143553190181633e-05,
      "loss": 1.7875,
      "step": 20820
    },
    {
      "epoch": 1.1288736091588056,
      "grad_norm": 0.5361613035202026,
      "learning_rate": 9.14262346289456e-05,
      "loss": 1.7806,
      "step": 20830
    },
    {
      "epoch": 1.129415550324318,
      "grad_norm": 0.32189682126045227,
      "learning_rate": 9.141693284375106e-05,
      "loss": 1.7967,
      "step": 20840
    },
    {
      "epoch": 1.12995749148983,
      "grad_norm": 0.28092676401138306,
      "learning_rate": 9.140762654738499e-05,
      "loss": 1.7902,
      "step": 20850
    },
    {
      "epoch": 1.1304994326553424,
      "grad_norm": 0.29651352763175964,
      "learning_rate": 9.139831574100022e-05,
      "loss": 1.7905,
      "step": 20860
    },
    {
      "epoch": 1.1310413738208545,
      "grad_norm": 0.3914618492126465,
      "learning_rate": 9.138900042575012e-05,
      "loss": 1.7863,
      "step": 20870
    },
    {
      "epoch": 1.1315833149863668,
      "grad_norm": 0.33878007531166077,
      "learning_rate": 9.137968060278863e-05,
      "loss": 1.7872,
      "step": 20880
    },
    {
      "epoch": 1.1315833149863668,
      "eval_loss": 2.5968291759490967,
      "eval_runtime": 21.9881,
      "eval_samples_per_second": 227.396,
      "eval_steps_per_second": 1.228,
      "step": 20880
    },
    {
      "epoch": 1.132125256151879,
      "grad_norm": 0.584201455116272,
      "learning_rate": 9.137035627327026e-05,
      "loss": 1.7768,
      "step": 20890
    },
    {
      "epoch": 1.1326671973173912,
      "grad_norm": 0.3373468816280365,
      "learning_rate": 9.136102743835004e-05,
      "loss": 1.7857,
      "step": 20900
    },
    {
      "epoch": 1.1332091384829035,
      "grad_norm": 0.2604641020298004,
      "learning_rate": 9.13516940991836e-05,
      "loss": 1.7847,
      "step": 20910
    },
    {
      "epoch": 1.1337510796484156,
      "grad_norm": 0.28360849618911743,
      "learning_rate": 9.134235625692714e-05,
      "loss": 1.7903,
      "step": 20920
    },
    {
      "epoch": 1.134293020813928,
      "grad_norm": 0.4277685284614563,
      "learning_rate": 9.133301391273736e-05,
      "loss": 1.7813,
      "step": 20930
    },
    {
      "epoch": 1.1348349619794402,
      "grad_norm": 0.6790282726287842,
      "learning_rate": 9.132366706777155e-05,
      "loss": 1.792,
      "step": 20940
    },
    {
      "epoch": 1.1353769031449523,
      "grad_norm": 0.4232507050037384,
      "learning_rate": 9.13143157231876e-05,
      "loss": 1.7832,
      "step": 20950
    },
    {
      "epoch": 1.1359188443104646,
      "grad_norm": 0.5046867728233337,
      "learning_rate": 9.130495988014384e-05,
      "loss": 1.7758,
      "step": 20960
    },
    {
      "epoch": 1.1362982031263231,
      "eval_loss": 2.590829372406006,
      "eval_runtime": 21.9886,
      "eval_samples_per_second": 227.391,
      "eval_steps_per_second": 1.228,
      "step": 20967
    },
    {
      "epoch": 1.1364607854759767,
      "grad_norm": 0.3382951021194458,
      "learning_rate": 9.129559953979928e-05,
      "loss": 1.7858,
      "step": 20970
    },
    {
      "epoch": 1.137002726641489,
      "grad_norm": 0.33747395873069763,
      "learning_rate": 9.128623470331343e-05,
      "loss": 1.7828,
      "step": 20980
    },
    {
      "epoch": 1.1375446678070011,
      "grad_norm": 0.2978461682796478,
      "learning_rate": 9.127686537184636e-05,
      "loss": 1.7839,
      "step": 20990
    },
    {
      "epoch": 1.1380866089725135,
      "grad_norm": 0.5556627511978149,
      "learning_rate": 9.126749154655872e-05,
      "loss": 1.7826,
      "step": 21000
    },
    {
      "epoch": 1.1386285501380256,
      "grad_norm": 0.3020962178707123,
      "learning_rate": 9.125811322861168e-05,
      "loss": 1.7779,
      "step": 21010
    },
    {
      "epoch": 1.1391704913035379,
      "grad_norm": 0.5155685544013977,
      "learning_rate": 9.1248730419167e-05,
      "loss": 1.7912,
      "step": 21020
    },
    {
      "epoch": 1.1397124324690502,
      "grad_norm": 1.0742664337158203,
      "learning_rate": 9.123934311938696e-05,
      "loss": 1.7943,
      "step": 21030
    },
    {
      "epoch": 1.1402543736345623,
      "grad_norm": 0.6324855089187622,
      "learning_rate": 9.122995133043442e-05,
      "loss": 1.7778,
      "step": 21040
    },
    {
      "epoch": 1.1407963148000746,
      "grad_norm": 0.33419859409332275,
      "learning_rate": 9.122055505347283e-05,
      "loss": 1.7839,
      "step": 21050
    },
    {
      "epoch": 1.1410130912662795,
      "eval_loss": 2.5992467403411865,
      "eval_runtime": 21.9929,
      "eval_samples_per_second": 227.346,
      "eval_steps_per_second": 1.228,
      "step": 21054
    },
    {
      "epoch": 1.1413382559655867,
      "grad_norm": 0.30402234196662903,
      "learning_rate": 9.12111542896661e-05,
      "loss": 1.7818,
      "step": 21060
    },
    {
      "epoch": 1.141880197131099,
      "grad_norm": 0.28443703055381775,
      "learning_rate": 9.120174904017882e-05,
      "loss": 1.7698,
      "step": 21070
    },
    {
      "epoch": 1.142422138296611,
      "grad_norm": 0.29675841331481934,
      "learning_rate": 9.119233930617603e-05,
      "loss": 1.7764,
      "step": 21080
    },
    {
      "epoch": 1.1429640794621234,
      "grad_norm": 0.6041000485420227,
      "learning_rate": 9.118292508882338e-05,
      "loss": 1.7697,
      "step": 21090
    },
    {
      "epoch": 1.1435060206276355,
      "grad_norm": 0.3296681046485901,
      "learning_rate": 9.117350638928706e-05,
      "loss": 1.7884,
      "step": 21100
    },
    {
      "epoch": 1.1440479617931478,
      "grad_norm": 0.5075874924659729,
      "learning_rate": 9.11640832087338e-05,
      "loss": 1.7874,
      "step": 21110
    },
    {
      "epoch": 1.1445899029586601,
      "grad_norm": 0.5272268056869507,
      "learning_rate": 9.115465554833095e-05,
      "loss": 1.7755,
      "step": 21120
    },
    {
      "epoch": 1.1451318441241722,
      "grad_norm": 0.6501821875572205,
      "learning_rate": 9.114522340924631e-05,
      "loss": 1.7839,
      "step": 21130
    },
    {
      "epoch": 1.1456737852896846,
      "grad_norm": 0.2923777103424072,
      "learning_rate": 9.113578679264835e-05,
      "loss": 1.7769,
      "step": 21140
    },
    {
      "epoch": 1.1457279794062356,
      "eval_loss": 2.5842106342315674,
      "eval_runtime": 22.0002,
      "eval_samples_per_second": 227.27,
      "eval_steps_per_second": 1.227,
      "step": 21141
    },
    {
      "epoch": 1.1462157264551966,
      "grad_norm": 0.3021109402179718,
      "learning_rate": 9.1126345699706e-05,
      "loss": 1.7834,
      "step": 21150
    },
    {
      "epoch": 1.146757667620709,
      "grad_norm": 0.30015039443969727,
      "learning_rate": 9.111690013158877e-05,
      "loss": 1.7723,
      "step": 21160
    },
    {
      "epoch": 1.147299608786221,
      "grad_norm": 0.7381042838096619,
      "learning_rate": 9.110745008946678e-05,
      "loss": 1.7824,
      "step": 21170
    },
    {
      "epoch": 1.1478415499517334,
      "grad_norm": 0.39448803663253784,
      "learning_rate": 9.109799557451062e-05,
      "loss": 1.7873,
      "step": 21180
    },
    {
      "epoch": 1.1483834911172455,
      "grad_norm": 0.2745918035507202,
      "learning_rate": 9.108853658789149e-05,
      "loss": 1.7832,
      "step": 21190
    },
    {
      "epoch": 1.1489254322827578,
      "grad_norm": 0.3459675908088684,
      "learning_rate": 9.107907313078115e-05,
      "loss": 1.7695,
      "step": 21200
    },
    {
      "epoch": 1.14946737344827,
      "grad_norm": 0.30421921610832214,
      "learning_rate": 9.106960520435183e-05,
      "loss": 1.7735,
      "step": 21210
    },
    {
      "epoch": 1.1500093146137822,
      "grad_norm": 0.2987232804298401,
      "learning_rate": 9.106013280977645e-05,
      "loss": 1.7777,
      "step": 21220
    },
    {
      "epoch": 1.150442867546192,
      "eval_loss": 2.5747172832489014,
      "eval_runtime": 21.9936,
      "eval_samples_per_second": 227.339,
      "eval_steps_per_second": 1.228,
      "step": 21228
    },
    {
      "epoch": 1.1505512557792945,
      "grad_norm": 0.29645800590515137,
      "learning_rate": 9.105065594822838e-05,
      "loss": 1.7794,
      "step": 21230
    },
    {
      "epoch": 1.1510931969448066,
      "grad_norm": 0.3672481179237366,
      "learning_rate": 9.104117462088154e-05,
      "loss": 1.7751,
      "step": 21240
    },
    {
      "epoch": 1.151635138110319,
      "grad_norm": 0.27135440707206726,
      "learning_rate": 9.103168882891047e-05,
      "loss": 1.7841,
      "step": 21250
    },
    {
      "epoch": 1.1521770792758312,
      "grad_norm": 0.5033676624298096,
      "learning_rate": 9.102219857349024e-05,
      "loss": 1.776,
      "step": 21260
    },
    {
      "epoch": 1.1527190204413433,
      "grad_norm": 0.4527705907821655,
      "learning_rate": 9.101270385579643e-05,
      "loss": 1.776,
      "step": 21270
    },
    {
      "epoch": 1.1532609616068557,
      "grad_norm": 0.4339083433151245,
      "learning_rate": 9.100320467700521e-05,
      "loss": 1.7749,
      "step": 21280
    },
    {
      "epoch": 1.1538029027723677,
      "grad_norm": 0.28354117274284363,
      "learning_rate": 9.099370103829332e-05,
      "loss": 1.7802,
      "step": 21290
    },
    {
      "epoch": 1.15434484393788,
      "grad_norm": 0.2814236283302307,
      "learning_rate": 9.098419294083801e-05,
      "loss": 1.7651,
      "step": 21300
    },
    {
      "epoch": 1.1548867851033922,
      "grad_norm": 0.32581958174705505,
      "learning_rate": 9.09746803858171e-05,
      "loss": 1.7867,
      "step": 21310
    },
    {
      "epoch": 1.1551577556861483,
      "eval_loss": 2.5878617763519287,
      "eval_runtime": 21.9939,
      "eval_samples_per_second": 227.336,
      "eval_steps_per_second": 1.228,
      "step": 21315
    },
    {
      "epoch": 1.1554287262689045,
      "grad_norm": 0.742839515209198,
      "learning_rate": 9.096516337440898e-05,
      "loss": 1.7793,
      "step": 21320
    },
    {
      "epoch": 1.1559706674344166,
      "grad_norm": 0.27201151847839355,
      "learning_rate": 9.095564190779257e-05,
      "loss": 1.7819,
      "step": 21330
    },
    {
      "epoch": 1.1565126085999289,
      "grad_norm": 0.2983904778957367,
      "learning_rate": 9.094611598714733e-05,
      "loss": 1.7799,
      "step": 21340
    },
    {
      "epoch": 1.1570545497654412,
      "grad_norm": 0.2776441276073456,
      "learning_rate": 9.09365856136533e-05,
      "loss": 1.7716,
      "step": 21350
    },
    {
      "epoch": 1.1575964909309533,
      "grad_norm": 0.27620989084243774,
      "learning_rate": 9.092705078849108e-05,
      "loss": 1.7662,
      "step": 21360
    },
    {
      "epoch": 1.1581384320964656,
      "grad_norm": 0.4133933484554291,
      "learning_rate": 9.091751151284178e-05,
      "loss": 1.7788,
      "step": 21370
    },
    {
      "epoch": 1.1586803732619777,
      "grad_norm": 0.587293267250061,
      "learning_rate": 9.090796778788709e-05,
      "loss": 1.7711,
      "step": 21380
    },
    {
      "epoch": 1.15922231442749,
      "grad_norm": 0.3694206178188324,
      "learning_rate": 9.089841961480927e-05,
      "loss": 1.7688,
      "step": 21390
    },
    {
      "epoch": 1.1597642555930021,
      "grad_norm": 0.5002974271774292,
      "learning_rate": 9.088886699479105e-05,
      "loss": 1.7743,
      "step": 21400
    },
    {
      "epoch": 1.1598726438261047,
      "eval_loss": 2.5802183151245117,
      "eval_runtime": 21.9908,
      "eval_samples_per_second": 227.368,
      "eval_steps_per_second": 1.228,
      "step": 21402
    },
    {
      "epoch": 1.1603061967585144,
      "grad_norm": 0.4875272810459137,
      "learning_rate": 9.087930992901581e-05,
      "loss": 1.7803,
      "step": 21410
    },
    {
      "epoch": 1.1608481379240265,
      "grad_norm": 0.33327093720436096,
      "learning_rate": 9.086974841866743e-05,
      "loss": 1.7743,
      "step": 21420
    },
    {
      "epoch": 1.1613900790895388,
      "grad_norm": 0.6208361983299255,
      "learning_rate": 9.086018246493037e-05,
      "loss": 1.7694,
      "step": 21430
    },
    {
      "epoch": 1.1619320202550512,
      "grad_norm": 0.4120174050331116,
      "learning_rate": 9.085061206898957e-05,
      "loss": 1.7802,
      "step": 21440
    },
    {
      "epoch": 1.1624739614205633,
      "grad_norm": 0.3111993074417114,
      "learning_rate": 9.08410372320306e-05,
      "loss": 1.7863,
      "step": 21450
    },
    {
      "epoch": 1.1630159025860756,
      "grad_norm": 0.2776833772659302,
      "learning_rate": 9.083145795523955e-05,
      "loss": 1.7794,
      "step": 21460
    },
    {
      "epoch": 1.1635578437515877,
      "grad_norm": 0.7861410975456238,
      "learning_rate": 9.082187423980304e-05,
      "loss": 1.7772,
      "step": 21470
    },
    {
      "epoch": 1.1640997849171,
      "grad_norm": 0.3405260741710663,
      "learning_rate": 9.081228608690828e-05,
      "loss": 1.7832,
      "step": 21480
    },
    {
      "epoch": 1.164587531966061,
      "eval_loss": 2.5907578468322754,
      "eval_runtime": 21.9898,
      "eval_samples_per_second": 227.378,
      "eval_steps_per_second": 1.228,
      "step": 21489
    },
    {
      "epoch": 1.164641726082612,
      "grad_norm": 0.28428077697753906,
      "learning_rate": 9.080269349774301e-05,
      "loss": 1.7735,
      "step": 21490
    },
    {
      "epoch": 1.1651836672481244,
      "grad_norm": 0.28834155201911926,
      "learning_rate": 9.079309647349549e-05,
      "loss": 1.7815,
      "step": 21500
    },
    {
      "epoch": 1.1657256084136365,
      "grad_norm": 0.35341453552246094,
      "learning_rate": 9.078349501535461e-05,
      "loss": 1.7785,
      "step": 21510
    },
    {
      "epoch": 1.1662675495791488,
      "grad_norm": 0.5031281113624573,
      "learning_rate": 9.077388912450969e-05,
      "loss": 1.7797,
      "step": 21520
    },
    {
      "epoch": 1.1668094907446611,
      "grad_norm": 0.439062237739563,
      "learning_rate": 9.076427880215072e-05,
      "loss": 1.7657,
      "step": 21530
    },
    {
      "epoch": 1.1673514319101732,
      "grad_norm": 0.356479287147522,
      "learning_rate": 9.075466404946814e-05,
      "loss": 1.7763,
      "step": 21540
    },
    {
      "epoch": 1.1678933730756855,
      "grad_norm": 0.3379020094871521,
      "learning_rate": 9.074504486765304e-05,
      "loss": 1.7652,
      "step": 21550
    },
    {
      "epoch": 1.1684353142411976,
      "grad_norm": 0.3034111559391022,
      "learning_rate": 9.073542125789695e-05,
      "loss": 1.7658,
      "step": 21560
    },
    {
      "epoch": 1.16897725540671,
      "grad_norm": 0.4461290240287781,
      "learning_rate": 9.072579322139202e-05,
      "loss": 1.7678,
      "step": 21570
    },
    {
      "epoch": 1.1693024201060171,
      "eval_loss": 2.5929572582244873,
      "eval_runtime": 21.9943,
      "eval_samples_per_second": 227.332,
      "eval_steps_per_second": 1.228,
      "step": 21576
    },
    {
      "epoch": 1.1695191965722223,
      "grad_norm": 0.25855252146720886,
      "learning_rate": 9.071616075933095e-05,
      "loss": 1.7704,
      "step": 21580
    },
    {
      "epoch": 1.1700611377377343,
      "grad_norm": 0.27765700221061707,
      "learning_rate": 9.070652387290695e-05,
      "loss": 1.7575,
      "step": 21590
    },
    {
      "epoch": 1.1706030789032467,
      "grad_norm": 0.26034823060035706,
      "learning_rate": 9.069688256331377e-05,
      "loss": 1.7625,
      "step": 21600
    },
    {
      "epoch": 1.1711450200687588,
      "grad_norm": 1.0148969888687134,
      "learning_rate": 9.068723683174578e-05,
      "loss": 1.7771,
      "step": 21610
    },
    {
      "epoch": 1.171686961234271,
      "grad_norm": 0.3433249592781067,
      "learning_rate": 9.067758667939782e-05,
      "loss": 1.769,
      "step": 21620
    },
    {
      "epoch": 1.1722289023997832,
      "grad_norm": 0.3450257480144501,
      "learning_rate": 9.066793210746533e-05,
      "loss": 1.7761,
      "step": 21630
    },
    {
      "epoch": 1.1727708435652955,
      "grad_norm": 0.29265645146369934,
      "learning_rate": 9.065827311714426e-05,
      "loss": 1.7633,
      "step": 21640
    },
    {
      "epoch": 1.1733127847308076,
      "grad_norm": 0.29442882537841797,
      "learning_rate": 9.064860970963112e-05,
      "loss": 1.7706,
      "step": 21650
    },
    {
      "epoch": 1.17385472589632,
      "grad_norm": 0.4089371860027313,
      "learning_rate": 9.063894188612298e-05,
      "loss": 1.7673,
      "step": 21660
    },
    {
      "epoch": 1.1740173082459735,
      "eval_loss": 2.585028648376465,
      "eval_runtime": 21.9891,
      "eval_samples_per_second": 227.386,
      "eval_steps_per_second": 1.228,
      "step": 21663
    },
    {
      "epoch": 1.1743966670618322,
      "grad_norm": 0.5681344866752625,
      "learning_rate": 9.062926964781746e-05,
      "loss": 1.7797,
      "step": 21670
    },
    {
      "epoch": 1.1749386082273443,
      "grad_norm": 0.5952972769737244,
      "learning_rate": 9.061959299591269e-05,
      "loss": 1.7781,
      "step": 21680
    },
    {
      "epoch": 1.1754805493928566,
      "grad_norm": 0.30679190158843994,
      "learning_rate": 9.060991193160739e-05,
      "loss": 1.7685,
      "step": 21690
    },
    {
      "epoch": 1.1760224905583687,
      "grad_norm": 0.37840238213539124,
      "learning_rate": 9.06002264561008e-05,
      "loss": 1.7613,
      "step": 21700
    },
    {
      "epoch": 1.176564431723881,
      "grad_norm": 0.6132546663284302,
      "learning_rate": 9.059053657059272e-05,
      "loss": 1.7842,
      "step": 21710
    },
    {
      "epoch": 1.1771063728893931,
      "grad_norm": 0.37106719613075256,
      "learning_rate": 9.058084227628351e-05,
      "loss": 1.763,
      "step": 21720
    },
    {
      "epoch": 1.1776483140549054,
      "grad_norm": 0.2650188207626343,
      "learning_rate": 9.057114357437401e-05,
      "loss": 1.7582,
      "step": 21730
    },
    {
      "epoch": 1.1781902552204175,
      "grad_norm": 0.4426226019859314,
      "learning_rate": 9.056144046606568e-05,
      "loss": 1.7725,
      "step": 21740
    },
    {
      "epoch": 1.1787321963859299,
      "grad_norm": 0.39338555932044983,
      "learning_rate": 9.05517329525605e-05,
      "loss": 1.7712,
      "step": 21750
    },
    {
      "epoch": 1.1787321963859299,
      "eval_loss": 2.59128999710083,
      "eval_runtime": 21.9834,
      "eval_samples_per_second": 227.444,
      "eval_steps_per_second": 1.228,
      "step": 21750
    },
    {
      "epoch": 1.1792741375514422,
      "grad_norm": 0.3727186918258667,
      "learning_rate": 9.054202103506098e-05,
      "loss": 1.7642,
      "step": 21760
    },
    {
      "epoch": 1.1798160787169543,
      "grad_norm": 0.4082069993019104,
      "learning_rate": 9.053230471477023e-05,
      "loss": 1.7736,
      "step": 21770
    },
    {
      "epoch": 1.1803580198824666,
      "grad_norm": 0.29831764101982117,
      "learning_rate": 9.052258399289182e-05,
      "loss": 1.7592,
      "step": 21780
    },
    {
      "epoch": 1.1808999610479787,
      "grad_norm": 0.4205126166343689,
      "learning_rate": 9.051285887062993e-05,
      "loss": 1.7677,
      "step": 21790
    },
    {
      "epoch": 1.181441902213491,
      "grad_norm": 0.38999685645103455,
      "learning_rate": 9.050312934918926e-05,
      "loss": 1.7681,
      "step": 21800
    },
    {
      "epoch": 1.181983843379003,
      "grad_norm": 0.34131819009780884,
      "learning_rate": 9.049339542977507e-05,
      "loss": 1.7568,
      "step": 21810
    },
    {
      "epoch": 1.1825257845445154,
      "grad_norm": 0.5073003768920898,
      "learning_rate": 9.048365711359317e-05,
      "loss": 1.7684,
      "step": 21820
    },
    {
      "epoch": 1.1830677257100275,
      "grad_norm": 0.47667044401168823,
      "learning_rate": 9.047391440184985e-05,
      "loss": 1.7634,
      "step": 21830
    },
    {
      "epoch": 1.1834470845258862,
      "eval_loss": 2.5854649543762207,
      "eval_runtime": 21.9926,
      "eval_samples_per_second": 227.349,
      "eval_steps_per_second": 1.228,
      "step": 21837
    },
    {
      "epoch": 1.1836096668755398,
      "grad_norm": 0.8095766305923462,
      "learning_rate": 9.046416729575205e-05,
      "loss": 1.7581,
      "step": 21840
    },
    {
      "epoch": 1.1841516080410521,
      "grad_norm": 0.3507765531539917,
      "learning_rate": 9.045441579650717e-05,
      "loss": 1.7646,
      "step": 21850
    },
    {
      "epoch": 1.1846935492065642,
      "grad_norm": 0.29202473163604736,
      "learning_rate": 9.044465990532318e-05,
      "loss": 1.7758,
      "step": 21860
    },
    {
      "epoch": 1.1852354903720765,
      "grad_norm": 0.265829861164093,
      "learning_rate": 9.043489962340861e-05,
      "loss": 1.7639,
      "step": 21870
    },
    {
      "epoch": 1.1857774315375886,
      "grad_norm": 0.2809160649776459,
      "learning_rate": 9.042513495197252e-05,
      "loss": 1.7516,
      "step": 21880
    },
    {
      "epoch": 1.186319372703101,
      "grad_norm": 0.3107336163520813,
      "learning_rate": 9.04153658922245e-05,
      "loss": 1.7647,
      "step": 21890
    },
    {
      "epoch": 1.186861313868613,
      "grad_norm": 0.5341174006462097,
      "learning_rate": 9.040559244537473e-05,
      "loss": 1.7625,
      "step": 21900
    },
    {
      "epoch": 1.1874032550341254,
      "grad_norm": 0.25601211190223694,
      "learning_rate": 9.039581461263388e-05,
      "loss": 1.7634,
      "step": 21910
    },
    {
      "epoch": 1.1879451961996375,
      "grad_norm": 0.6476098299026489,
      "learning_rate": 9.038603239521318e-05,
      "loss": 1.7637,
      "step": 21920
    },
    {
      "epoch": 1.1881619726658426,
      "eval_loss": 2.5872724056243896,
      "eval_runtime": 21.9906,
      "eval_samples_per_second": 227.37,
      "eval_steps_per_second": 1.228,
      "step": 21924
    },
    {
      "epoch": 1.1884871373651498,
      "grad_norm": 0.30727824568748474,
      "learning_rate": 9.037624579432442e-05,
      "loss": 1.7608,
      "step": 21930
    },
    {
      "epoch": 1.189029078530662,
      "grad_norm": 0.256548672914505,
      "learning_rate": 9.036645481117992e-05,
      "loss": 1.7675,
      "step": 21940
    },
    {
      "epoch": 1.1895710196961742,
      "grad_norm": 0.3580315113067627,
      "learning_rate": 9.035665944699254e-05,
      "loss": 1.7716,
      "step": 21950
    },
    {
      "epoch": 1.1901129608616865,
      "grad_norm": 0.2569289207458496,
      "learning_rate": 9.034685970297571e-05,
      "loss": 1.7627,
      "step": 21960
    },
    {
      "epoch": 1.1906549020271986,
      "grad_norm": 0.3389548361301422,
      "learning_rate": 9.033705558034335e-05,
      "loss": 1.7617,
      "step": 21970
    },
    {
      "epoch": 1.191196843192711,
      "grad_norm": 0.35166749358177185,
      "learning_rate": 9.032724708030995e-05,
      "loss": 1.7634,
      "step": 21980
    },
    {
      "epoch": 1.1917387843582232,
      "grad_norm": 0.47258156538009644,
      "learning_rate": 9.031743420409058e-05,
      "loss": 1.7606,
      "step": 21990
    },
    {
      "epoch": 1.1922807255237353,
      "grad_norm": 0.3378153443336487,
      "learning_rate": 9.030761695290077e-05,
      "loss": 1.7602,
      "step": 22000
    },
    {
      "epoch": 1.1928226666892476,
      "grad_norm": 0.5784263610839844,
      "learning_rate": 9.029779532795668e-05,
      "loss": 1.7585,
      "step": 22010
    },
    {
      "epoch": 1.1928768608057987,
      "eval_loss": 2.6090340614318848,
      "eval_runtime": 21.9879,
      "eval_samples_per_second": 227.398,
      "eval_steps_per_second": 1.228,
      "step": 22011
    },
    {
      "epoch": 1.1933646078547597,
      "grad_norm": 0.5027700066566467,
      "learning_rate": 9.028796933047495e-05,
      "loss": 1.7604,
      "step": 22020
    },
    {
      "epoch": 1.193906549020272,
      "grad_norm": 0.3202950060367584,
      "learning_rate": 9.027813896167278e-05,
      "loss": 1.7642,
      "step": 22030
    },
    {
      "epoch": 1.1944484901857841,
      "grad_norm": 0.3492630422115326,
      "learning_rate": 9.026830422276792e-05,
      "loss": 1.7612,
      "step": 22040
    },
    {
      "epoch": 1.1949904313512965,
      "grad_norm": 0.6746916770935059,
      "learning_rate": 9.025846511497864e-05,
      "loss": 1.7562,
      "step": 22050
    },
    {
      "epoch": 1.1955323725168086,
      "grad_norm": 0.42550233006477356,
      "learning_rate": 9.02486216395238e-05,
      "loss": 1.7618,
      "step": 22060
    },
    {
      "epoch": 1.1960743136823209,
      "grad_norm": 0.4293383061885834,
      "learning_rate": 9.023877379762274e-05,
      "loss": 1.763,
      "step": 22070
    },
    {
      "epoch": 1.1966162548478332,
      "grad_norm": 0.3266971707344055,
      "learning_rate": 9.022892159049537e-05,
      "loss": 1.7669,
      "step": 22080
    },
    {
      "epoch": 1.1971581960133453,
      "grad_norm": 0.99581378698349,
      "learning_rate": 9.021906501936213e-05,
      "loss": 1.7648,
      "step": 22090
    },
    {
      "epoch": 1.197591748945755,
      "eval_loss": 2.58486270904541,
      "eval_runtime": 21.9939,
      "eval_samples_per_second": 227.336,
      "eval_steps_per_second": 1.228,
      "step": 22098
    },
    {
      "epoch": 1.1977001371788576,
      "grad_norm": 0.5003217458724976,
      "learning_rate": 9.020920408544404e-05,
      "loss": 1.7659,
      "step": 22100
    },
    {
      "epoch": 1.1982420783443697,
      "grad_norm": 0.5149915218353271,
      "learning_rate": 9.019933878996259e-05,
      "loss": 1.7496,
      "step": 22110
    },
    {
      "epoch": 1.198784019509882,
      "grad_norm": 0.3475677967071533,
      "learning_rate": 9.018946913413989e-05,
      "loss": 1.7642,
      "step": 22120
    },
    {
      "epoch": 1.199325960675394,
      "grad_norm": 0.3139243423938751,
      "learning_rate": 9.017959511919853e-05,
      "loss": 1.7556,
      "step": 22130
    },
    {
      "epoch": 1.1998679018409064,
      "grad_norm": 0.32470259070396423,
      "learning_rate": 9.016971674636165e-05,
      "loss": 1.7616,
      "step": 22140
    },
    {
      "epoch": 1.2004098430064185,
      "grad_norm": 0.6316733360290527,
      "learning_rate": 9.015983401685296e-05,
      "loss": 1.7633,
      "step": 22150
    },
    {
      "epoch": 1.2009517841719308,
      "grad_norm": 0.6149379014968872,
      "learning_rate": 9.014994693189667e-05,
      "loss": 1.7599,
      "step": 22160
    },
    {
      "epoch": 1.2014937253374431,
      "grad_norm": 0.4740849435329437,
      "learning_rate": 9.014005549271757e-05,
      "loss": 1.7587,
      "step": 22170
    },
    {
      "epoch": 1.2020356665029552,
      "grad_norm": 0.4013853371143341,
      "learning_rate": 9.013015970054096e-05,
      "loss": 1.7523,
      "step": 22180
    },
    {
      "epoch": 1.2023066370857114,
      "eval_loss": 2.5968024730682373,
      "eval_runtime": 21.9857,
      "eval_samples_per_second": 227.42,
      "eval_steps_per_second": 1.228,
      "step": 22185
    },
    {
      "epoch": 1.2025776076684676,
      "grad_norm": 0.5497053861618042,
      "learning_rate": 9.012025955659269e-05,
      "loss": 1.7546,
      "step": 22190
    },
    {
      "epoch": 1.2031195488339796,
      "grad_norm": 0.3100571632385254,
      "learning_rate": 9.011035506209912e-05,
      "loss": 1.7637,
      "step": 22200
    },
    {
      "epoch": 1.203661489999492,
      "grad_norm": 0.298776239156723,
      "learning_rate": 9.010044621828722e-05,
      "loss": 1.7663,
      "step": 22210
    },
    {
      "epoch": 1.204203431165004,
      "grad_norm": 0.36760690808296204,
      "learning_rate": 9.009053302638444e-05,
      "loss": 1.7575,
      "step": 22220
    },
    {
      "epoch": 1.2047453723305164,
      "grad_norm": 0.2956767678260803,
      "learning_rate": 9.008061548761876e-05,
      "loss": 1.7624,
      "step": 22230
    },
    {
      "epoch": 1.2052873134960285,
      "grad_norm": 0.5647885203361511,
      "learning_rate": 9.007069360321873e-05,
      "loss": 1.7409,
      "step": 22240
    },
    {
      "epoch": 1.2058292546615408,
      "grad_norm": 0.26180022954940796,
      "learning_rate": 9.006076737441347e-05,
      "loss": 1.7539,
      "step": 22250
    },
    {
      "epoch": 1.206371195827053,
      "grad_norm": 0.29279986023902893,
      "learning_rate": 9.005083680243254e-05,
      "loss": 1.7551,
      "step": 22260
    },
    {
      "epoch": 1.2069131369925652,
      "grad_norm": 0.3438403904438019,
      "learning_rate": 9.004090188850612e-05,
      "loss": 1.7585,
      "step": 22270
    },
    {
      "epoch": 1.2070215252256677,
      "eval_loss": 2.595259666442871,
      "eval_runtime": 21.992,
      "eval_samples_per_second": 227.356,
      "eval_steps_per_second": 1.228,
      "step": 22272
    },
    {
      "epoch": 1.2074550781580775,
      "grad_norm": 0.46787571907043457,
      "learning_rate": 9.003096263386492e-05,
      "loss": 1.7648,
      "step": 22280
    },
    {
      "epoch": 1.2079970193235896,
      "grad_norm": 0.5319798588752747,
      "learning_rate": 9.002101903974016e-05,
      "loss": 1.7574,
      "step": 22290
    },
    {
      "epoch": 1.208538960489102,
      "grad_norm": 0.45188024640083313,
      "learning_rate": 9.00110711073636e-05,
      "loss": 1.7565,
      "step": 22300
    },
    {
      "epoch": 1.2090809016546142,
      "grad_norm": 0.4394822120666504,
      "learning_rate": 9.000111883796756e-05,
      "loss": 1.7594,
      "step": 22310
    },
    {
      "epoch": 1.2096228428201263,
      "grad_norm": 0.290712833404541,
      "learning_rate": 8.999116223278486e-05,
      "loss": 1.7592,
      "step": 22320
    },
    {
      "epoch": 1.2101647839856386,
      "grad_norm": 0.6723989248275757,
      "learning_rate": 8.998120129304892e-05,
      "loss": 1.752,
      "step": 22330
    },
    {
      "epoch": 1.2107067251511507,
      "grad_norm": 0.3058725595474243,
      "learning_rate": 8.997123601999364e-05,
      "loss": 1.7487,
      "step": 22340
    },
    {
      "epoch": 1.211248666316663,
      "grad_norm": 0.25992143154144287,
      "learning_rate": 8.996126641485345e-05,
      "loss": 1.7464,
      "step": 22350
    },
    {
      "epoch": 1.211736413365624,
      "eval_loss": 2.6008663177490234,
      "eval_runtime": 21.9912,
      "eval_samples_per_second": 227.363,
      "eval_steps_per_second": 1.228,
      "step": 22359
    },
    {
      "epoch": 1.2117906074821752,
      "grad_norm": 0.30580416321754456,
      "learning_rate": 8.995129247886339e-05,
      "loss": 1.7577,
      "step": 22360
    },
    {
      "epoch": 1.2123325486476875,
      "grad_norm": 0.2877110540866852,
      "learning_rate": 8.994131421325893e-05,
      "loss": 1.7556,
      "step": 22370
    },
    {
      "epoch": 1.2128744898131996,
      "grad_norm": 0.390491247177124,
      "learning_rate": 8.993133161927618e-05,
      "loss": 1.7553,
      "step": 22380
    },
    {
      "epoch": 1.2134164309787119,
      "grad_norm": 0.3429940342903137,
      "learning_rate": 8.992134469815173e-05,
      "loss": 1.7507,
      "step": 22390
    },
    {
      "epoch": 1.2139583721442242,
      "grad_norm": 0.29374656081199646,
      "learning_rate": 8.99113534511227e-05,
      "loss": 1.7548,
      "step": 22400
    },
    {
      "epoch": 1.2145003133097363,
      "grad_norm": 0.37793880701065063,
      "learning_rate": 8.99013578794268e-05,
      "loss": 1.7504,
      "step": 22410
    },
    {
      "epoch": 1.2150422544752486,
      "grad_norm": 0.312223345041275,
      "learning_rate": 8.989135798430218e-05,
      "loss": 1.7373,
      "step": 22420
    },
    {
      "epoch": 1.2155841956407607,
      "grad_norm": 0.355259507894516,
      "learning_rate": 8.988135376698764e-05,
      "loss": 1.7562,
      "step": 22430
    },
    {
      "epoch": 1.216126136806273,
      "grad_norm": 0.5126851797103882,
      "learning_rate": 8.987134522872242e-05,
      "loss": 1.7551,
      "step": 22440
    },
    {
      "epoch": 1.2164513015055802,
      "eval_loss": 2.5885868072509766,
      "eval_runtime": 21.9949,
      "eval_samples_per_second": 227.325,
      "eval_steps_per_second": 1.228,
      "step": 22446
    },
    {
      "epoch": 1.2166680779717851,
      "grad_norm": 0.3533921241760254,
      "learning_rate": 8.986133237074636e-05,
      "loss": 1.7599,
      "step": 22450
    },
    {
      "epoch": 1.2172100191372974,
      "grad_norm": 0.3277350962162018,
      "learning_rate": 8.98513151942998e-05,
      "loss": 1.7425,
      "step": 22460
    },
    {
      "epoch": 1.2177519603028095,
      "grad_norm": 0.45636269450187683,
      "learning_rate": 8.984129370062362e-05,
      "loss": 1.7538,
      "step": 22470
    },
    {
      "epoch": 1.2182939014683218,
      "grad_norm": 0.7295756340026855,
      "learning_rate": 8.983126789095925e-05,
      "loss": 1.7601,
      "step": 22480
    },
    {
      "epoch": 1.2188358426338342,
      "grad_norm": 0.5607370734214783,
      "learning_rate": 8.982123776654862e-05,
      "loss": 1.7585,
      "step": 22490
    },
    {
      "epoch": 1.2193777837993462,
      "grad_norm": 0.29513418674468994,
      "learning_rate": 8.981120332863423e-05,
      "loss": 1.7626,
      "step": 22500
    },
    {
      "epoch": 1.2199197249648586,
      "grad_norm": 0.2580159306526184,
      "learning_rate": 8.980116457845911e-05,
      "loss": 1.756,
      "step": 22510
    },
    {
      "epoch": 1.2204616661303707,
      "grad_norm": 0.31364554166793823,
      "learning_rate": 8.979112151726684e-05,
      "loss": 1.7503,
      "step": 22520
    },
    {
      "epoch": 1.221003607295883,
      "grad_norm": 0.8860167860984802,
      "learning_rate": 8.978107414630146e-05,
      "loss": 1.7357,
      "step": 22530
    },
    {
      "epoch": 1.2211661896455366,
      "eval_loss": 2.586221933364868,
      "eval_runtime": 21.9909,
      "eval_samples_per_second": 227.367,
      "eval_steps_per_second": 1.228,
      "step": 22533
    },
    {
      "epoch": 1.221545548461395,
      "grad_norm": 0.550439178943634,
      "learning_rate": 8.977102246680762e-05,
      "loss": 1.7482,
      "step": 22540
    },
    {
      "epoch": 1.2220874896269074,
      "grad_norm": 0.8875564932823181,
      "learning_rate": 8.976096648003048e-05,
      "loss": 1.7466,
      "step": 22550
    },
    {
      "epoch": 1.2226294307924195,
      "grad_norm": 0.5859338045120239,
      "learning_rate": 8.975090618721573e-05,
      "loss": 1.7587,
      "step": 22560
    },
    {
      "epoch": 1.2231713719579318,
      "grad_norm": 0.27543067932128906,
      "learning_rate": 8.97408415896096e-05,
      "loss": 1.7534,
      "step": 22570
    },
    {
      "epoch": 1.2237133131234441,
      "grad_norm": 0.3466237485408783,
      "learning_rate": 8.973077268845884e-05,
      "loss": 1.7601,
      "step": 22580
    },
    {
      "epoch": 1.2242552542889562,
      "grad_norm": 0.2768082916736603,
      "learning_rate": 8.972069948501074e-05,
      "loss": 1.7504,
      "step": 22590
    },
    {
      "epoch": 1.2247971954544685,
      "grad_norm": 0.45373255014419556,
      "learning_rate": 8.971062198051315e-05,
      "loss": 1.7479,
      "step": 22600
    },
    {
      "epoch": 1.2253391366199806,
      "grad_norm": 0.6204401850700378,
      "learning_rate": 8.970054017621437e-05,
      "loss": 1.7478,
      "step": 22610
    },
    {
      "epoch": 1.225881077785493,
      "grad_norm": 0.3263530135154724,
      "learning_rate": 8.969045407336336e-05,
      "loss": 1.7486,
      "step": 22620
    },
    {
      "epoch": 1.225881077785493,
      "eval_loss": 2.593569040298462,
      "eval_runtime": 21.9692,
      "eval_samples_per_second": 227.591,
      "eval_steps_per_second": 1.229,
      "step": 22620
    },
    {
      "epoch": 1.2264230189510053,
      "grad_norm": 0.48239004611968994,
      "learning_rate": 8.968036367320952e-05,
      "loss": 1.7489,
      "step": 22630
    },
    {
      "epoch": 1.2269649601165173,
      "grad_norm": 0.3203703761100769,
      "learning_rate": 8.967026897700277e-05,
      "loss": 1.7437,
      "step": 22640
    },
    {
      "epoch": 1.2275069012820297,
      "grad_norm": 0.6340882182121277,
      "learning_rate": 8.966016998599362e-05,
      "loss": 1.7371,
      "step": 22650
    },
    {
      "epoch": 1.2280488424475418,
      "grad_norm": 0.30104759335517883,
      "learning_rate": 8.96500667014331e-05,
      "loss": 1.7637,
      "step": 22660
    },
    {
      "epoch": 1.228590783613054,
      "grad_norm": 0.2778589129447937,
      "learning_rate": 8.963995912457275e-05,
      "loss": 1.7441,
      "step": 22670
    },
    {
      "epoch": 1.2291327247785662,
      "grad_norm": 0.3374054431915283,
      "learning_rate": 8.962984725666465e-05,
      "loss": 1.7528,
      "step": 22680
    },
    {
      "epoch": 1.2296746659440785,
      "grad_norm": 0.25601983070373535,
      "learning_rate": 8.961973109896144e-05,
      "loss": 1.7444,
      "step": 22690
    },
    {
      "epoch": 1.2302166071095906,
      "grad_norm": 0.2800886929035187,
      "learning_rate": 8.960961065271622e-05,
      "loss": 1.7444,
      "step": 22700
    },
    {
      "epoch": 1.2305959659254493,
      "eval_loss": 2.58610463142395,
      "eval_runtime": 21.9907,
      "eval_samples_per_second": 227.368,
      "eval_steps_per_second": 1.228,
      "step": 22707
    },
    {
      "epoch": 1.230758548275103,
      "grad_norm": 0.4286305904388428,
      "learning_rate": 8.95994859191827e-05,
      "loss": 1.7398,
      "step": 22710
    },
    {
      "epoch": 1.2313004894406152,
      "grad_norm": 0.5830126404762268,
      "learning_rate": 8.95893568996151e-05,
      "loss": 1.7458,
      "step": 22720
    },
    {
      "epoch": 1.2318424306061273,
      "grad_norm": 0.30274108052253723,
      "learning_rate": 8.957922359526812e-05,
      "loss": 1.7443,
      "step": 22730
    },
    {
      "epoch": 1.2323843717716396,
      "grad_norm": 0.5309381484985352,
      "learning_rate": 8.956908600739707e-05,
      "loss": 1.7477,
      "step": 22740
    },
    {
      "epoch": 1.2329263129371517,
      "grad_norm": 0.2998747229576111,
      "learning_rate": 8.95589441372577e-05,
      "loss": 1.7574,
      "step": 22750
    },
    {
      "epoch": 1.233468254102664,
      "grad_norm": 0.3685966432094574,
      "learning_rate": 8.954879798610637e-05,
      "loss": 1.74,
      "step": 22760
    },
    {
      "epoch": 1.2340101952681761,
      "grad_norm": 0.6951805949211121,
      "learning_rate": 8.953864755519995e-05,
      "loss": 1.7444,
      "step": 22770
    },
    {
      "epoch": 1.2345521364336884,
      "grad_norm": 0.3382275700569153,
      "learning_rate": 8.952849284579585e-05,
      "loss": 1.7352,
      "step": 22780
    },
    {
      "epoch": 1.2350940775992005,
      "grad_norm": 0.26488932967185974,
      "learning_rate": 8.951833385915193e-05,
      "loss": 1.7461,
      "step": 22790
    },
    {
      "epoch": 1.2353108540654056,
      "eval_loss": 2.575146436691284,
      "eval_runtime": 21.9928,
      "eval_samples_per_second": 227.347,
      "eval_steps_per_second": 1.228,
      "step": 22794
    },
    {
      "epoch": 1.2356360187647129,
      "grad_norm": 0.46747535467147827,
      "learning_rate": 8.950817059652669e-05,
      "loss": 1.7437,
      "step": 22800
    },
    {
      "epoch": 1.2361779599302252,
      "grad_norm": 0.29368457198143005,
      "learning_rate": 8.949800305917909e-05,
      "loss": 1.7458,
      "step": 22810
    },
    {
      "epoch": 1.2367199010957373,
      "grad_norm": 0.3844955265522003,
      "learning_rate": 8.948783124836866e-05,
      "loss": 1.7403,
      "step": 22820
    },
    {
      "epoch": 1.2372618422612496,
      "grad_norm": 0.6568012237548828,
      "learning_rate": 8.94776551653554e-05,
      "loss": 1.7358,
      "step": 22830
    },
    {
      "epoch": 1.2378037834267617,
      "grad_norm": 0.34444165229797363,
      "learning_rate": 8.946747481139992e-05,
      "loss": 1.7468,
      "step": 22840
    },
    {
      "epoch": 1.238345724592274,
      "grad_norm": 0.44042837619781494,
      "learning_rate": 8.945729018776331e-05,
      "loss": 1.7455,
      "step": 22850
    },
    {
      "epoch": 1.238887665757786,
      "grad_norm": 0.32380804419517517,
      "learning_rate": 8.944710129570719e-05,
      "loss": 1.7432,
      "step": 22860
    },
    {
      "epoch": 1.2394296069232984,
      "grad_norm": 0.608935534954071,
      "learning_rate": 8.943690813649369e-05,
      "loss": 1.7578,
      "step": 22870
    },
    {
      "epoch": 1.2399715480888105,
      "grad_norm": 0.5230217576026917,
      "learning_rate": 8.942671071138554e-05,
      "loss": 1.7458,
      "step": 22880
    },
    {
      "epoch": 1.2400257422053618,
      "eval_loss": 2.581331253051758,
      "eval_runtime": 21.9957,
      "eval_samples_per_second": 227.317,
      "eval_steps_per_second": 1.228,
      "step": 22881
    },
    {
      "epoch": 1.2405134892543228,
      "grad_norm": 0.5061838030815125,
      "learning_rate": 8.941650902164595e-05,
      "loss": 1.7455,
      "step": 22890
    },
    {
      "epoch": 1.2410554304198351,
      "grad_norm": 0.405821293592453,
      "learning_rate": 8.940630306853861e-05,
      "loss": 1.7405,
      "step": 22900
    },
    {
      "epoch": 1.2415973715853472,
      "grad_norm": 0.45560845732688904,
      "learning_rate": 8.939609285332785e-05,
      "loss": 1.7459,
      "step": 22910
    },
    {
      "epoch": 1.2421393127508595,
      "grad_norm": 0.5665388703346252,
      "learning_rate": 8.938587837727842e-05,
      "loss": 1.7552,
      "step": 22920
    },
    {
      "epoch": 1.2426812539163716,
      "grad_norm": 0.37847816944122314,
      "learning_rate": 8.937565964165569e-05,
      "loss": 1.7422,
      "step": 22930
    },
    {
      "epoch": 1.243223195081884,
      "grad_norm": 0.4069490432739258,
      "learning_rate": 8.936543664772546e-05,
      "loss": 1.7435,
      "step": 22940
    },
    {
      "epoch": 1.2437651362473963,
      "grad_norm": 0.3240896761417389,
      "learning_rate": 8.935520939675414e-05,
      "loss": 1.7512,
      "step": 22950
    },
    {
      "epoch": 1.2443070774129084,
      "grad_norm": 0.4123472273349762,
      "learning_rate": 8.934497789000865e-05,
      "loss": 1.7425,
      "step": 22960
    },
    {
      "epoch": 1.2447406303453181,
      "eval_loss": 2.5666730403900146,
      "eval_runtime": 21.9906,
      "eval_samples_per_second": 227.37,
      "eval_steps_per_second": 1.228,
      "step": 22968
    },
    {
      "epoch": 1.2448490185784207,
      "grad_norm": 0.2759721577167511,
      "learning_rate": 8.933474212875642e-05,
      "loss": 1.736,
      "step": 22970
    },
    {
      "epoch": 1.2453909597439328,
      "grad_norm": 0.438054621219635,
      "learning_rate": 8.932450211426537e-05,
      "loss": 1.7408,
      "step": 22980
    },
    {
      "epoch": 1.245932900909445,
      "grad_norm": 0.3276619613170624,
      "learning_rate": 8.931425784780405e-05,
      "loss": 1.748,
      "step": 22990
    },
    {
      "epoch": 1.2464748420749572,
      "grad_norm": 0.27382031083106995,
      "learning_rate": 8.930400933064144e-05,
      "loss": 1.7394,
      "step": 23000
    },
    {
      "epoch": 1.2470167832404695,
      "grad_norm": 0.24581308662891388,
      "learning_rate": 8.929375656404707e-05,
      "loss": 1.7361,
      "step": 23010
    },
    {
      "epoch": 1.2475587244059816,
      "grad_norm": 0.2953229546546936,
      "learning_rate": 8.928349954929103e-05,
      "loss": 1.7587,
      "step": 23020
    },
    {
      "epoch": 1.248100665571494,
      "grad_norm": 0.48332518339157104,
      "learning_rate": 8.927323828764393e-05,
      "loss": 1.7334,
      "step": 23030
    },
    {
      "epoch": 1.2486426067370062,
      "grad_norm": 0.8373629450798035,
      "learning_rate": 8.926297278037685e-05,
      "loss": 1.7477,
      "step": 23040
    },
    {
      "epoch": 1.2491845479025183,
      "grad_norm": 0.2879384458065033,
      "learning_rate": 8.925270302876146e-05,
      "loss": 1.7373,
      "step": 23050
    },
    {
      "epoch": 1.2494555184852745,
      "eval_loss": 2.5645227432250977,
      "eval_runtime": 21.9888,
      "eval_samples_per_second": 227.388,
      "eval_steps_per_second": 1.228,
      "step": 23055
    },
    {
      "epoch": 1.2497264890680306,
      "grad_norm": 0.2931196689605713,
      "learning_rate": 8.924242903406993e-05,
      "loss": 1.747,
      "step": 23060
    },
    {
      "epoch": 1.2502684302335427,
      "grad_norm": 0.5951820015907288,
      "learning_rate": 8.923215079757496e-05,
      "loss": 1.7395,
      "step": 23070
    },
    {
      "epoch": 1.250810371399055,
      "grad_norm": 0.30423420667648315,
      "learning_rate": 8.922186832054977e-05,
      "loss": 1.7422,
      "step": 23080
    },
    {
      "epoch": 1.2513523125645671,
      "grad_norm": 0.3447738289833069,
      "learning_rate": 8.92115816042681e-05,
      "loss": 1.7358,
      "step": 23090
    },
    {
      "epoch": 1.2518942537300795,
      "grad_norm": 0.3407246470451355,
      "learning_rate": 8.920129065000424e-05,
      "loss": 1.7486,
      "step": 23100
    },
    {
      "epoch": 1.2524361948955915,
      "grad_norm": 0.33367615938186646,
      "learning_rate": 8.919099545903299e-05,
      "loss": 1.7355,
      "step": 23110
    },
    {
      "epoch": 1.2529781360611039,
      "grad_norm": 0.32768481969833374,
      "learning_rate": 8.918069603262965e-05,
      "loss": 1.7406,
      "step": 23120
    },
    {
      "epoch": 1.2535200772266162,
      "grad_norm": 0.31800198554992676,
      "learning_rate": 8.91703923720701e-05,
      "loss": 1.7452,
      "step": 23130
    },
    {
      "epoch": 1.2540620183921283,
      "grad_norm": 0.5581735372543335,
      "learning_rate": 8.916008447863068e-05,
      "loss": 1.743,
      "step": 23140
    },
    {
      "epoch": 1.2541704066252308,
      "eval_loss": 2.5848233699798584,
      "eval_runtime": 21.9922,
      "eval_samples_per_second": 227.353,
      "eval_steps_per_second": 1.228,
      "step": 23142
    },
    {
      "epoch": 1.2546039595576406,
      "grad_norm": 0.5674453377723694,
      "learning_rate": 8.914977235358831e-05,
      "loss": 1.7393,
      "step": 23150
    },
    {
      "epoch": 1.2551459007231527,
      "grad_norm": 0.3421599864959717,
      "learning_rate": 8.913945599822043e-05,
      "loss": 1.7402,
      "step": 23160
    },
    {
      "epoch": 1.255687841888665,
      "grad_norm": 0.31315505504608154,
      "learning_rate": 8.912913541380492e-05,
      "loss": 1.7363,
      "step": 23170
    },
    {
      "epoch": 1.2562297830541773,
      "grad_norm": 0.30993425846099854,
      "learning_rate": 8.91188106016203e-05,
      "loss": 1.7405,
      "step": 23180
    },
    {
      "epoch": 1.2567717242196894,
      "grad_norm": 0.2763616144657135,
      "learning_rate": 8.910848156294555e-05,
      "loss": 1.7464,
      "step": 23190
    },
    {
      "epoch": 1.2573136653852015,
      "grad_norm": 0.3239862322807312,
      "learning_rate": 8.90981482990602e-05,
      "loss": 1.7422,
      "step": 23200
    },
    {
      "epoch": 1.2578556065507138,
      "grad_norm": 0.3442568778991699,
      "learning_rate": 8.908781081124427e-05,
      "loss": 1.7433,
      "step": 23210
    },
    {
      "epoch": 1.2583975477162261,
      "grad_norm": 0.2833685278892517,
      "learning_rate": 8.907746910077834e-05,
      "loss": 1.7337,
      "step": 23220
    },
    {
      "epoch": 1.2588852947651872,
      "eval_loss": 2.577096939086914,
      "eval_runtime": 21.9824,
      "eval_samples_per_second": 227.454,
      "eval_steps_per_second": 1.228,
      "step": 23229
    },
    {
      "epoch": 1.2589394888817382,
      "grad_norm": 0.2630921006202698,
      "learning_rate": 8.906712316894346e-05,
      "loss": 1.7534,
      "step": 23230
    },
    {
      "epoch": 1.2594814300472505,
      "grad_norm": 0.5170078277587891,
      "learning_rate": 8.90567730170213e-05,
      "loss": 1.7483,
      "step": 23240
    },
    {
      "epoch": 1.2600233712127626,
      "grad_norm": 0.3840513527393341,
      "learning_rate": 8.904641864629394e-05,
      "loss": 1.7339,
      "step": 23250
    },
    {
      "epoch": 1.260565312378275,
      "grad_norm": 0.29101109504699707,
      "learning_rate": 8.903606005804406e-05,
      "loss": 1.7207,
      "step": 23260
    },
    {
      "epoch": 1.2611072535437873,
      "grad_norm": 0.34224191308021545,
      "learning_rate": 8.902569725355482e-05,
      "loss": 1.7333,
      "step": 23270
    },
    {
      "epoch": 1.2616491947092994,
      "grad_norm": 0.8239234089851379,
      "learning_rate": 8.901533023410994e-05,
      "loss": 1.7417,
      "step": 23280
    },
    {
      "epoch": 1.2621911358748115,
      "grad_norm": 0.7893701791763306,
      "learning_rate": 8.900495900099362e-05,
      "loss": 1.7398,
      "step": 23290
    },
    {
      "epoch": 1.2627330770403238,
      "grad_norm": 0.31907394528388977,
      "learning_rate": 8.899458355549061e-05,
      "loss": 1.739,
      "step": 23300
    },
    {
      "epoch": 1.263275018205836,
      "grad_norm": 0.26405349373817444,
      "learning_rate": 8.898420389888619e-05,
      "loss": 1.747,
      "step": 23310
    },
    {
      "epoch": 1.2636001829051433,
      "eval_loss": 2.575762987136841,
      "eval_runtime": 21.9898,
      "eval_samples_per_second": 227.378,
      "eval_steps_per_second": 1.228,
      "step": 23316
    },
    {
      "epoch": 1.2638169593713482,
      "grad_norm": 0.2704724669456482,
      "learning_rate": 8.897382003246614e-05,
      "loss": 1.7478,
      "step": 23320
    },
    {
      "epoch": 1.2643589005368605,
      "grad_norm": 0.28200745582580566,
      "learning_rate": 8.896343195751677e-05,
      "loss": 1.7414,
      "step": 23330
    },
    {
      "epoch": 1.2649008417023726,
      "grad_norm": 0.367416113615036,
      "learning_rate": 8.895303967532489e-05,
      "loss": 1.7448,
      "step": 23340
    },
    {
      "epoch": 1.265442782867885,
      "grad_norm": 0.41314128041267395,
      "learning_rate": 8.894264318717786e-05,
      "loss": 1.7301,
      "step": 23350
    },
    {
      "epoch": 1.2659847240333972,
      "grad_norm": 0.346835196018219,
      "learning_rate": 8.893224249436357e-05,
      "loss": 1.7309,
      "step": 23360
    },
    {
      "epoch": 1.2665266651989093,
      "grad_norm": 0.37419700622558594,
      "learning_rate": 8.892183759817039e-05,
      "loss": 1.7363,
      "step": 23370
    },
    {
      "epoch": 1.2670686063644216,
      "grad_norm": 0.5215455293655396,
      "learning_rate": 8.891142849988725e-05,
      "loss": 1.7392,
      "step": 23380
    },
    {
      "epoch": 1.2676105475299337,
      "grad_norm": 0.32529744505882263,
      "learning_rate": 8.890101520080357e-05,
      "loss": 1.7349,
      "step": 23390
    },
    {
      "epoch": 1.268152488695446,
      "grad_norm": 0.5487125515937805,
      "learning_rate": 8.889059770220931e-05,
      "loss": 1.734,
      "step": 23400
    },
    {
      "epoch": 1.2683150710450997,
      "eval_loss": 2.5602195262908936,
      "eval_runtime": 21.9906,
      "eval_samples_per_second": 227.37,
      "eval_steps_per_second": 1.228,
      "step": 23403
    },
    {
      "epoch": 1.2686944298609582,
      "grad_norm": 0.465263694524765,
      "learning_rate": 8.888017600539493e-05,
      "loss": 1.7317,
      "step": 23410
    },
    {
      "epoch": 1.2692363710264705,
      "grad_norm": 0.3183075487613678,
      "learning_rate": 8.886975011165146e-05,
      "loss": 1.7385,
      "step": 23420
    },
    {
      "epoch": 1.2697783121919826,
      "grad_norm": 0.5280638933181763,
      "learning_rate": 8.885932002227039e-05,
      "loss": 1.732,
      "step": 23430
    },
    {
      "epoch": 1.2703202533574949,
      "grad_norm": 0.3372764587402344,
      "learning_rate": 8.884888573854375e-05,
      "loss": 1.7314,
      "step": 23440
    },
    {
      "epoch": 1.2708621945230072,
      "grad_norm": 0.374603807926178,
      "learning_rate": 8.883844726176412e-05,
      "loss": 1.731,
      "step": 23450
    },
    {
      "epoch": 1.2714041356885193,
      "grad_norm": 0.3078954219818115,
      "learning_rate": 8.882800459322453e-05,
      "loss": 1.7377,
      "step": 23460
    },
    {
      "epoch": 1.2719460768540316,
      "grad_norm": 0.2972983419895172,
      "learning_rate": 8.881755773421863e-05,
      "loss": 1.734,
      "step": 23470
    },
    {
      "epoch": 1.2724880180195437,
      "grad_norm": 0.30043935775756836,
      "learning_rate": 8.880710668604047e-05,
      "loss": 1.7393,
      "step": 23480
    },
    {
      "epoch": 1.273029959185056,
      "grad_norm": 0.4024638533592224,
      "learning_rate": 8.879665144998473e-05,
      "loss": 1.738,
      "step": 23490
    },
    {
      "epoch": 1.273029959185056,
      "eval_loss": 2.574789047241211,
      "eval_runtime": 21.9656,
      "eval_samples_per_second": 227.628,
      "eval_steps_per_second": 1.229,
      "step": 23490
    },
    {
      "epoch": 1.273571900350568,
      "grad_norm": 0.33888739347457886,
      "learning_rate": 8.878619202734653e-05,
      "loss": 1.7342,
      "step": 23500
    },
    {
      "epoch": 1.2741138415160804,
      "grad_norm": 0.40043386816978455,
      "learning_rate": 8.877572841942153e-05,
      "loss": 1.7383,
      "step": 23510
    },
    {
      "epoch": 1.2746557826815925,
      "grad_norm": 0.46869125962257385,
      "learning_rate": 8.876526062750597e-05,
      "loss": 1.7412,
      "step": 23520
    },
    {
      "epoch": 1.2751977238471048,
      "grad_norm": 0.41995322704315186,
      "learning_rate": 8.875478865289649e-05,
      "loss": 1.7332,
      "step": 23530
    },
    {
      "epoch": 1.2757396650126172,
      "grad_norm": 0.28642240166664124,
      "learning_rate": 8.874431249689033e-05,
      "loss": 1.7313,
      "step": 23540
    },
    {
      "epoch": 1.2762816061781292,
      "grad_norm": 0.5792128443717957,
      "learning_rate": 8.873383216078527e-05,
      "loss": 1.7324,
      "step": 23550
    },
    {
      "epoch": 1.2768235473436416,
      "grad_norm": 0.6733444929122925,
      "learning_rate": 8.872334764587952e-05,
      "loss": 1.7357,
      "step": 23560
    },
    {
      "epoch": 1.2773654885091537,
      "grad_norm": 0.27996379137039185,
      "learning_rate": 8.87128589534719e-05,
      "loss": 1.723,
      "step": 23570
    },
    {
      "epoch": 1.2777448473250121,
      "eval_loss": 2.5722544193267822,
      "eval_runtime": 21.9913,
      "eval_samples_per_second": 227.362,
      "eval_steps_per_second": 1.228,
      "step": 23577
    },
    {
      "epoch": 1.277907429674666,
      "grad_norm": 0.29412564635276794,
      "learning_rate": 8.870236608486165e-05,
      "loss": 1.7301,
      "step": 23580
    },
    {
      "epoch": 1.2784493708401783,
      "grad_norm": 0.2967008352279663,
      "learning_rate": 8.869186904134862e-05,
      "loss": 1.7411,
      "step": 23590
    },
    {
      "epoch": 1.2789913120056904,
      "grad_norm": 1.056755542755127,
      "learning_rate": 8.868136782423314e-05,
      "loss": 1.735,
      "step": 23600
    },
    {
      "epoch": 1.2795332531712025,
      "grad_norm": 0.6758991479873657,
      "learning_rate": 8.867086243481603e-05,
      "loss": 1.7431,
      "step": 23610
    },
    {
      "epoch": 1.2800751943367148,
      "grad_norm": 0.33271247148513794,
      "learning_rate": 8.866035287439868e-05,
      "loss": 1.7338,
      "step": 23620
    },
    {
      "epoch": 1.280617135502227,
      "grad_norm": 0.40084099769592285,
      "learning_rate": 8.864983914428293e-05,
      "loss": 1.7385,
      "step": 23630
    },
    {
      "epoch": 1.2811590766677392,
      "grad_norm": 0.3270736038684845,
      "learning_rate": 8.863932124577123e-05,
      "loss": 1.716,
      "step": 23640
    },
    {
      "epoch": 1.2817010178332515,
      "grad_norm": 0.3113623559474945,
      "learning_rate": 8.862879918016643e-05,
      "loss": 1.7371,
      "step": 23650
    },
    {
      "epoch": 1.2822429589987636,
      "grad_norm": 0.6056118011474609,
      "learning_rate": 8.861827294877201e-05,
      "loss": 1.7375,
      "step": 23660
    },
    {
      "epoch": 1.2824597354649687,
      "eval_loss": 2.576855182647705,
      "eval_runtime": 21.9914,
      "eval_samples_per_second": 227.361,
      "eval_steps_per_second": 1.228,
      "step": 23664
    },
    {
      "epoch": 1.282784900164276,
      "grad_norm": 0.3462965488433838,
      "learning_rate": 8.86077425528919e-05,
      "loss": 1.7308,
      "step": 23670
    },
    {
      "epoch": 1.2833268413297882,
      "grad_norm": 0.2960757315158844,
      "learning_rate": 8.859720799383054e-05,
      "loss": 1.7263,
      "step": 23680
    },
    {
      "epoch": 1.2838687824953003,
      "grad_norm": 0.29280513525009155,
      "learning_rate": 8.858666927289292e-05,
      "loss": 1.7369,
      "step": 23690
    },
    {
      "epoch": 1.2844107236608124,
      "grad_norm": 0.602401614189148,
      "learning_rate": 8.857612639138451e-05,
      "loss": 1.7398,
      "step": 23700
    },
    {
      "epoch": 1.2849526648263248,
      "grad_norm": 0.4261922240257263,
      "learning_rate": 8.856557935061137e-05,
      "loss": 1.7272,
      "step": 23710
    },
    {
      "epoch": 1.285494605991837,
      "grad_norm": 0.32072654366493225,
      "learning_rate": 8.855502815187996e-05,
      "loss": 1.7529,
      "step": 23720
    },
    {
      "epoch": 1.2860365471573492,
      "grad_norm": 0.45794838666915894,
      "learning_rate": 8.854447279649737e-05,
      "loss": 1.7268,
      "step": 23730
    },
    {
      "epoch": 1.2865784883228615,
      "grad_norm": 0.8380472660064697,
      "learning_rate": 8.85339132857711e-05,
      "loss": 1.732,
      "step": 23740
    },
    {
      "epoch": 1.2871204294883736,
      "grad_norm": 0.25734448432922363,
      "learning_rate": 8.852334962100926e-05,
      "loss": 1.7341,
      "step": 23750
    },
    {
      "epoch": 1.2871746236049249,
      "eval_loss": 2.560034990310669,
      "eval_runtime": 21.9846,
      "eval_samples_per_second": 227.432,
      "eval_steps_per_second": 1.228,
      "step": 23751
    },
    {
      "epoch": 1.287662370653886,
      "grad_norm": 0.32123035192489624,
      "learning_rate": 8.851278180352041e-05,
      "loss": 1.7288,
      "step": 23760
    },
    {
      "epoch": 1.2882043118193982,
      "grad_norm": 0.41775065660476685,
      "learning_rate": 8.850220983461365e-05,
      "loss": 1.7305,
      "step": 23770
    },
    {
      "epoch": 1.2887462529849103,
      "grad_norm": 0.3736642599105835,
      "learning_rate": 8.849163371559858e-05,
      "loss": 1.7276,
      "step": 23780
    },
    {
      "epoch": 1.2892881941504226,
      "grad_norm": 0.3616468906402588,
      "learning_rate": 8.848105344778532e-05,
      "loss": 1.731,
      "step": 23790
    },
    {
      "epoch": 1.2898301353159347,
      "grad_norm": 0.39535510540008545,
      "learning_rate": 8.847046903248453e-05,
      "loss": 1.7199,
      "step": 23800
    },
    {
      "epoch": 1.290372076481447,
      "grad_norm": 0.27410778403282166,
      "learning_rate": 8.845988047100736e-05,
      "loss": 1.729,
      "step": 23810
    },
    {
      "epoch": 1.2909140176469591,
      "grad_norm": 0.2933545708656311,
      "learning_rate": 8.844928776466547e-05,
      "loss": 1.7338,
      "step": 23820
    },
    {
      "epoch": 1.2914559588124714,
      "grad_norm": 0.44660070538520813,
      "learning_rate": 8.843869091477102e-05,
      "loss": 1.7316,
      "step": 23830
    },
    {
      "epoch": 1.2918895117448812,
      "eval_loss": 2.552804946899414,
      "eval_runtime": 21.9898,
      "eval_samples_per_second": 227.378,
      "eval_steps_per_second": 1.228,
      "step": 23838
    },
    {
      "epoch": 1.2919978999779835,
      "grad_norm": 0.3844490945339203,
      "learning_rate": 8.842808992263672e-05,
      "loss": 1.7361,
      "step": 23840
    },
    {
      "epoch": 1.2925398411434958,
      "grad_norm": 0.5550277829170227,
      "learning_rate": 8.841748478957577e-05,
      "loss": 1.7385,
      "step": 23850
    },
    {
      "epoch": 1.2930817823090082,
      "grad_norm": 0.32363656163215637,
      "learning_rate": 8.840687551690189e-05,
      "loss": 1.7243,
      "step": 23860
    },
    {
      "epoch": 1.2936237234745203,
      "grad_norm": 0.4406464993953705,
      "learning_rate": 8.839626210592931e-05,
      "loss": 1.7437,
      "step": 23870
    },
    {
      "epoch": 1.2941656646400326,
      "grad_norm": 0.3115149140357971,
      "learning_rate": 8.838564455797275e-05,
      "loss": 1.7447,
      "step": 23880
    },
    {
      "epoch": 1.2947076058055447,
      "grad_norm": 0.27961403131484985,
      "learning_rate": 8.837502287434752e-05,
      "loss": 1.7287,
      "step": 23890
    },
    {
      "epoch": 1.295249546971057,
      "grad_norm": 0.49664953351020813,
      "learning_rate": 8.836439705636935e-05,
      "loss": 1.7304,
      "step": 23900
    },
    {
      "epoch": 1.2957914881365693,
      "grad_norm": 0.6754089593887329,
      "learning_rate": 8.835376710535451e-05,
      "loss": 1.7334,
      "step": 23910
    },
    {
      "epoch": 1.2963334293020814,
      "grad_norm": 0.2822556793689728,
      "learning_rate": 8.834313302261982e-05,
      "loss": 1.7226,
      "step": 23920
    },
    {
      "epoch": 1.2966043998848376,
      "eval_loss": 2.5520572662353516,
      "eval_runtime": 21.9902,
      "eval_samples_per_second": 227.373,
      "eval_steps_per_second": 1.228,
      "step": 23925
    },
    {
      "epoch": 1.2968753704675935,
      "grad_norm": 0.41642260551452637,
      "learning_rate": 8.833249480948257e-05,
      "loss": 1.7223,
      "step": 23930
    },
    {
      "epoch": 1.2974173116331058,
      "grad_norm": 0.6077197790145874,
      "learning_rate": 8.832185246726057e-05,
      "loss": 1.7211,
      "step": 23940
    },
    {
      "epoch": 1.2979592527986181,
      "grad_norm": 0.7251920700073242,
      "learning_rate": 8.831120599727217e-05,
      "loss": 1.7233,
      "step": 23950
    },
    {
      "epoch": 1.2985011939641302,
      "grad_norm": 0.7429440021514893,
      "learning_rate": 8.83005554008362e-05,
      "loss": 1.727,
      "step": 23960
    },
    {
      "epoch": 1.2990431351296425,
      "grad_norm": 0.33626994490623474,
      "learning_rate": 8.828990067927199e-05,
      "loss": 1.7237,
      "step": 23970
    },
    {
      "epoch": 1.2995850762951546,
      "grad_norm": 0.4045674502849579,
      "learning_rate": 8.827924183389941e-05,
      "loss": 1.7197,
      "step": 23980
    },
    {
      "epoch": 1.300127017460667,
      "grad_norm": 0.43031617999076843,
      "learning_rate": 8.826857886603885e-05,
      "loss": 1.7181,
      "step": 23990
    },
    {
      "epoch": 1.3006689586261793,
      "grad_norm": 0.4407658874988556,
      "learning_rate": 8.825791177701116e-05,
      "loss": 1.7297,
      "step": 24000
    },
    {
      "epoch": 1.3012108997916914,
      "grad_norm": 0.4655308425426483,
      "learning_rate": 8.824724056813775e-05,
      "loss": 1.7271,
      "step": 24010
    },
    {
      "epoch": 1.3013192880247937,
      "eval_loss": 2.550110340118408,
      "eval_runtime": 21.9873,
      "eval_samples_per_second": 227.404,
      "eval_steps_per_second": 1.228,
      "step": 24012
    },
    {
      "epoch": 1.3017528409572034,
      "grad_norm": 0.47410663962364197,
      "learning_rate": 8.823656524074054e-05,
      "loss": 1.7271,
      "step": 24020
    },
    {
      "epoch": 1.3022947821227158,
      "grad_norm": 0.397776335477829,
      "learning_rate": 8.822588579614192e-05,
      "loss": 1.7268,
      "step": 24030
    },
    {
      "epoch": 1.302836723288228,
      "grad_norm": 0.4908923804759979,
      "learning_rate": 8.821520223566483e-05,
      "loss": 1.7312,
      "step": 24040
    },
    {
      "epoch": 1.3033786644537402,
      "grad_norm": 0.475315660238266,
      "learning_rate": 8.820451456063268e-05,
      "loss": 1.7358,
      "step": 24050
    },
    {
      "epoch": 1.3039206056192525,
      "grad_norm": 0.3519163131713867,
      "learning_rate": 8.819382277236943e-05,
      "loss": 1.7313,
      "step": 24060
    },
    {
      "epoch": 1.3044625467847646,
      "grad_norm": 0.6225571632385254,
      "learning_rate": 8.818312687219953e-05,
      "loss": 1.7269,
      "step": 24070
    },
    {
      "epoch": 1.305004487950277,
      "grad_norm": 0.3213563859462738,
      "learning_rate": 8.817242686144793e-05,
      "loss": 1.7273,
      "step": 24080
    },
    {
      "epoch": 1.3055464291157892,
      "grad_norm": 0.3261583149433136,
      "learning_rate": 8.816172274144013e-05,
      "loss": 1.7208,
      "step": 24090
    },
    {
      "epoch": 1.30603417616475,
      "eval_loss": 2.560994863510132,
      "eval_runtime": 21.9922,
      "eval_samples_per_second": 227.353,
      "eval_steps_per_second": 1.228,
      "step": 24099
    },
    {
      "epoch": 1.3060883702813013,
      "grad_norm": 0.35049372911453247,
      "learning_rate": 8.815101451350207e-05,
      "loss": 1.7158,
      "step": 24100
    },
    {
      "epoch": 1.3066303114468136,
      "grad_norm": 0.253828227519989,
      "learning_rate": 8.814030217896026e-05,
      "loss": 1.7282,
      "step": 24110
    },
    {
      "epoch": 1.3071722526123257,
      "grad_norm": 0.32618460059165955,
      "learning_rate": 8.81295857391417e-05,
      "loss": 1.7182,
      "step": 24120
    },
    {
      "epoch": 1.307714193777838,
      "grad_norm": 0.2585597634315491,
      "learning_rate": 8.811886519537391e-05,
      "loss": 1.7322,
      "step": 24130
    },
    {
      "epoch": 1.3082561349433501,
      "grad_norm": 0.29455363750457764,
      "learning_rate": 8.810814054898488e-05,
      "loss": 1.71,
      "step": 24140
    },
    {
      "epoch": 1.3087980761088625,
      "grad_norm": 0.32605740427970886,
      "learning_rate": 8.809741180130313e-05,
      "loss": 1.7158,
      "step": 24150
    },
    {
      "epoch": 1.3093400172743745,
      "grad_norm": 0.422584593296051,
      "learning_rate": 8.808667895365771e-05,
      "loss": 1.717,
      "step": 24160
    },
    {
      "epoch": 1.3098819584398869,
      "grad_norm": 0.3378809988498688,
      "learning_rate": 8.807594200737815e-05,
      "loss": 1.7302,
      "step": 24170
    },
    {
      "epoch": 1.3104238996053992,
      "grad_norm": 0.4901425242424011,
      "learning_rate": 8.806520096379448e-05,
      "loss": 1.7179,
      "step": 24180
    },
    {
      "epoch": 1.3107490643047064,
      "eval_loss": 2.562967300415039,
      "eval_runtime": 21.9941,
      "eval_samples_per_second": 227.334,
      "eval_steps_per_second": 1.228,
      "step": 24186
    },
    {
      "epoch": 1.3109658407709113,
      "grad_norm": 0.39655086398124695,
      "learning_rate": 8.805445582423728e-05,
      "loss": 1.7215,
      "step": 24190
    },
    {
      "epoch": 1.3115077819364236,
      "grad_norm": 0.3672724962234497,
      "learning_rate": 8.804370659003762e-05,
      "loss": 1.7214,
      "step": 24200
    },
    {
      "epoch": 1.3120497231019357,
      "grad_norm": 0.4074071943759918,
      "learning_rate": 8.803295326252701e-05,
      "loss": 1.7172,
      "step": 24210
    },
    {
      "epoch": 1.312591664267448,
      "grad_norm": 0.4320417046546936,
      "learning_rate": 8.802219584303758e-05,
      "loss": 1.7183,
      "step": 24220
    },
    {
      "epoch": 1.3131336054329603,
      "grad_norm": 0.5554030537605286,
      "learning_rate": 8.80114343329019e-05,
      "loss": 1.7136,
      "step": 24230
    },
    {
      "epoch": 1.3136755465984724,
      "grad_norm": 0.3394809663295746,
      "learning_rate": 8.800066873345306e-05,
      "loss": 1.7206,
      "step": 24240
    },
    {
      "epoch": 1.3142174877639845,
      "grad_norm": 0.2953517436981201,
      "learning_rate": 8.798989904602465e-05,
      "loss": 1.7286,
      "step": 24250
    },
    {
      "epoch": 1.3147594289294968,
      "grad_norm": 0.3113909959793091,
      "learning_rate": 8.797912527195078e-05,
      "loss": 1.7155,
      "step": 24260
    },
    {
      "epoch": 1.3153013700950091,
      "grad_norm": 0.26612791419029236,
      "learning_rate": 8.796834741256605e-05,
      "loss": 1.7277,
      "step": 24270
    },
    {
      "epoch": 1.3154639524446627,
      "eval_loss": 2.5562336444854736,
      "eval_runtime": 21.9909,
      "eval_samples_per_second": 227.367,
      "eval_steps_per_second": 1.228,
      "step": 24273
    },
    {
      "epoch": 1.3158433112605212,
      "grad_norm": 0.3873152434825897,
      "learning_rate": 8.795756546920556e-05,
      "loss": 1.7147,
      "step": 24280
    },
    {
      "epoch": 1.3163852524260335,
      "grad_norm": 0.30190661549568176,
      "learning_rate": 8.794677944320497e-05,
      "loss": 1.7135,
      "step": 24290
    },
    {
      "epoch": 1.3169271935915456,
      "grad_norm": 0.6682907342910767,
      "learning_rate": 8.793598933590036e-05,
      "loss": 1.729,
      "step": 24300
    },
    {
      "epoch": 1.317469134757058,
      "grad_norm": 0.280906617641449,
      "learning_rate": 8.79251951486284e-05,
      "loss": 1.7061,
      "step": 24310
    },
    {
      "epoch": 1.3180110759225703,
      "grad_norm": 0.5581380724906921,
      "learning_rate": 8.79143968827262e-05,
      "loss": 1.7343,
      "step": 24320
    },
    {
      "epoch": 1.3185530170880824,
      "grad_norm": 0.4470410645008087,
      "learning_rate": 8.790359453953145e-05,
      "loss": 1.7278,
      "step": 24330
    },
    {
      "epoch": 1.3190949582535945,
      "grad_norm": 0.4987182021141052,
      "learning_rate": 8.789278812038222e-05,
      "loss": 1.7266,
      "step": 24340
    },
    {
      "epoch": 1.3196368994191068,
      "grad_norm": 1.2388598918914795,
      "learning_rate": 8.788197762661723e-05,
      "loss": 1.7193,
      "step": 24350
    },
    {
      "epoch": 1.320178840584619,
      "grad_norm": 0.8786284327507019,
      "learning_rate": 8.78711630595756e-05,
      "loss": 1.7217,
      "step": 24360
    },
    {
      "epoch": 1.320178840584619,
      "eval_loss": 2.568453073501587,
      "eval_runtime": 21.9891,
      "eval_samples_per_second": 227.386,
      "eval_steps_per_second": 1.228,
      "step": 24360
    },
    {
      "epoch": 1.3207207817501312,
      "grad_norm": 0.41298526525497437,
      "learning_rate": 8.7860344420597e-05,
      "loss": 1.7191,
      "step": 24370
    },
    {
      "epoch": 1.3212627229156435,
      "grad_norm": 0.27922841906547546,
      "learning_rate": 8.78495217110216e-05,
      "loss": 1.7214,
      "step": 24380
    },
    {
      "epoch": 1.3218046640811556,
      "grad_norm": 0.3357425034046173,
      "learning_rate": 8.783869493219008e-05,
      "loss": 1.7165,
      "step": 24390
    },
    {
      "epoch": 1.322346605246668,
      "grad_norm": 0.6415514945983887,
      "learning_rate": 8.782786408544358e-05,
      "loss": 1.7259,
      "step": 24400
    },
    {
      "epoch": 1.3228885464121802,
      "grad_norm": 0.5723456144332886,
      "learning_rate": 8.78170291721238e-05,
      "loss": 1.7221,
      "step": 24410
    },
    {
      "epoch": 1.3234304875776923,
      "grad_norm": 0.27751269936561584,
      "learning_rate": 8.780619019357295e-05,
      "loss": 1.7241,
      "step": 24420
    },
    {
      "epoch": 1.3239724287432046,
      "grad_norm": 0.5434432029724121,
      "learning_rate": 8.779534715113368e-05,
      "loss": 1.7284,
      "step": 24430
    },
    {
      "epoch": 1.3245143699087167,
      "grad_norm": 0.3848007917404175,
      "learning_rate": 8.778450004614918e-05,
      "loss": 1.7164,
      "step": 24440
    },
    {
      "epoch": 1.3248937287245752,
      "eval_loss": 2.563822031021118,
      "eval_runtime": 21.9907,
      "eval_samples_per_second": 227.369,
      "eval_steps_per_second": 1.228,
      "step": 24447
    },
    {
      "epoch": 1.325056311074229,
      "grad_norm": 0.3684322237968445,
      "learning_rate": 8.777364887996315e-05,
      "loss": 1.7197,
      "step": 24450
    },
    {
      "epoch": 1.3255982522397411,
      "grad_norm": 0.3419678807258606,
      "learning_rate": 8.77627936539198e-05,
      "loss": 1.7221,
      "step": 24460
    },
    {
      "epoch": 1.3261401934052535,
      "grad_norm": 0.2709847688674927,
      "learning_rate": 8.77519343693638e-05,
      "loss": 1.7159,
      "step": 24470
    },
    {
      "epoch": 1.3266821345707656,
      "grad_norm": 0.29516398906707764,
      "learning_rate": 8.774107102764038e-05,
      "loss": 1.72,
      "step": 24480
    },
    {
      "epoch": 1.3272240757362779,
      "grad_norm": 0.29573437571525574,
      "learning_rate": 8.773020363009521e-05,
      "loss": 1.7187,
      "step": 24490
    },
    {
      "epoch": 1.3277660169017902,
      "grad_norm": 0.44336891174316406,
      "learning_rate": 8.771933217807453e-05,
      "loss": 1.7106,
      "step": 24500
    },
    {
      "epoch": 1.3283079580673023,
      "grad_norm": 0.3083667755126953,
      "learning_rate": 8.770845667292503e-05,
      "loss": 1.7229,
      "step": 24510
    },
    {
      "epoch": 1.3288498992328146,
      "grad_norm": 0.4367120862007141,
      "learning_rate": 8.769757711599391e-05,
      "loss": 1.7231,
      "step": 24520
    },
    {
      "epoch": 1.3293918403983267,
      "grad_norm": 0.40371039509773254,
      "learning_rate": 8.768669350862892e-05,
      "loss": 1.7188,
      "step": 24530
    },
    {
      "epoch": 1.3296086168645316,
      "eval_loss": 2.572169542312622,
      "eval_runtime": 21.9901,
      "eval_samples_per_second": 227.375,
      "eval_steps_per_second": 1.228,
      "step": 24534
    },
    {
      "epoch": 1.329933781563839,
      "grad_norm": 0.7102428674697876,
      "learning_rate": 8.767580585217823e-05,
      "loss": 1.7157,
      "step": 24540
    },
    {
      "epoch": 1.3304757227293513,
      "grad_norm": 0.35714074969291687,
      "learning_rate": 8.766491414799057e-05,
      "loss": 1.7066,
      "step": 24550
    },
    {
      "epoch": 1.3310176638948634,
      "grad_norm": 0.29054415225982666,
      "learning_rate": 8.765401839741517e-05,
      "loss": 1.7183,
      "step": 24560
    },
    {
      "epoch": 1.3315596050603755,
      "grad_norm": 0.26257362961769104,
      "learning_rate": 8.764311860180175e-05,
      "loss": 1.7163,
      "step": 24570
    },
    {
      "epoch": 1.3321015462258878,
      "grad_norm": 0.3010044991970062,
      "learning_rate": 8.763221476250051e-05,
      "loss": 1.7215,
      "step": 24580
    },
    {
      "epoch": 1.3326434873914001,
      "grad_norm": 0.33944013714790344,
      "learning_rate": 8.762130688086219e-05,
      "loss": 1.72,
      "step": 24590
    },
    {
      "epoch": 1.3331854285569122,
      "grad_norm": 0.5436154007911682,
      "learning_rate": 8.761039495823799e-05,
      "loss": 1.7121,
      "step": 24600
    },
    {
      "epoch": 1.3337273697224246,
      "grad_norm": 0.4238379895687103,
      "learning_rate": 8.759947899597964e-05,
      "loss": 1.7195,
      "step": 24610
    },
    {
      "epoch": 1.3342693108879367,
      "grad_norm": 0.2585737705230713,
      "learning_rate": 8.758855899543939e-05,
      "loss": 1.7015,
      "step": 24620
    },
    {
      "epoch": 1.334323505004488,
      "eval_loss": 2.5573627948760986,
      "eval_runtime": 21.9891,
      "eval_samples_per_second": 227.386,
      "eval_steps_per_second": 1.228,
      "step": 24621
    },
    {
      "epoch": 1.334811252053449,
      "grad_norm": 0.35536953806877136,
      "learning_rate": 8.75776349579699e-05,
      "loss": 1.7164,
      "step": 24630
    },
    {
      "epoch": 1.3353531932189613,
      "grad_norm": 0.4927321672439575,
      "learning_rate": 8.756670688492445e-05,
      "loss": 1.7181,
      "step": 24640
    },
    {
      "epoch": 1.3358951343844734,
      "grad_norm": 0.3061029613018036,
      "learning_rate": 8.755577477765674e-05,
      "loss": 1.7136,
      "step": 24650
    },
    {
      "epoch": 1.3364370755499855,
      "grad_norm": 0.27562111616134644,
      "learning_rate": 8.7544838637521e-05,
      "loss": 1.7222,
      "step": 24660
    },
    {
      "epoch": 1.3369790167154978,
      "grad_norm": 0.46605604887008667,
      "learning_rate": 8.753389846587194e-05,
      "loss": 1.7085,
      "step": 24670
    },
    {
      "epoch": 1.33752095788101,
      "grad_norm": 0.47294819355010986,
      "learning_rate": 8.752295426406479e-05,
      "loss": 1.711,
      "step": 24680
    },
    {
      "epoch": 1.3380628990465222,
      "grad_norm": 0.42514827847480774,
      "learning_rate": 8.751200603345524e-05,
      "loss": 1.7081,
      "step": 24690
    },
    {
      "epoch": 1.3386048402120345,
      "grad_norm": 0.46494096517562866,
      "learning_rate": 8.750105377539957e-05,
      "loss": 1.6989,
      "step": 24700
    },
    {
      "epoch": 1.3390383931444443,
      "eval_loss": 2.5702195167541504,
      "eval_runtime": 21.9879,
      "eval_samples_per_second": 227.398,
      "eval_steps_per_second": 1.228,
      "step": 24708
    },
    {
      "epoch": 1.3391467813775466,
      "grad_norm": 0.36518365144729614,
      "learning_rate": 8.749009749125445e-05,
      "loss": 1.7115,
      "step": 24710
    },
    {
      "epoch": 1.339688722543059,
      "grad_norm": 0.32633692026138306,
      "learning_rate": 8.747913718237712e-05,
      "loss": 1.7166,
      "step": 24720
    },
    {
      "epoch": 1.3402306637085712,
      "grad_norm": 0.370088130235672,
      "learning_rate": 8.746817285012527e-05,
      "loss": 1.714,
      "step": 24730
    },
    {
      "epoch": 1.3407726048740833,
      "grad_norm": 0.25269201397895813,
      "learning_rate": 8.745720449585714e-05,
      "loss": 1.7212,
      "step": 24740
    },
    {
      "epoch": 1.3413145460395957,
      "grad_norm": 0.3450757563114166,
      "learning_rate": 8.744623212093142e-05,
      "loss": 1.7241,
      "step": 24750
    },
    {
      "epoch": 1.3418564872051078,
      "grad_norm": 0.26472437381744385,
      "learning_rate": 8.743525572670734e-05,
      "loss": 1.7087,
      "step": 24760
    },
    {
      "epoch": 1.34239842837062,
      "grad_norm": 0.5389896035194397,
      "learning_rate": 8.74242753145446e-05,
      "loss": 1.7157,
      "step": 24770
    },
    {
      "epoch": 1.3429403695361322,
      "grad_norm": 0.6463091969490051,
      "learning_rate": 8.74132908858034e-05,
      "loss": 1.7056,
      "step": 24780
    },
    {
      "epoch": 1.3434823107016445,
      "grad_norm": 0.40934550762176514,
      "learning_rate": 8.740230244184448e-05,
      "loss": 1.7137,
      "step": 24790
    },
    {
      "epoch": 1.3437532812844006,
      "eval_loss": 2.5633254051208496,
      "eval_runtime": 21.9886,
      "eval_samples_per_second": 227.391,
      "eval_steps_per_second": 1.228,
      "step": 24795
    },
    {
      "epoch": 1.3440242518671566,
      "grad_norm": 0.4404658377170563,
      "learning_rate": 8.739130998402898e-05,
      "loss": 1.7126,
      "step": 24800
    },
    {
      "epoch": 1.3445661930326689,
      "grad_norm": 0.37851405143737793,
      "learning_rate": 8.738031351371863e-05,
      "loss": 1.7127,
      "step": 24810
    },
    {
      "epoch": 1.3451081341981812,
      "grad_norm": 0.30251404643058777,
      "learning_rate": 8.736931303227563e-05,
      "loss": 1.7293,
      "step": 24820
    },
    {
      "epoch": 1.3456500753636933,
      "grad_norm": 0.5295489430427551,
      "learning_rate": 8.735830854106267e-05,
      "loss": 1.7092,
      "step": 24830
    },
    {
      "epoch": 1.3461920165292056,
      "grad_norm": 0.54742431640625,
      "learning_rate": 8.734730004144292e-05,
      "loss": 1.7201,
      "step": 24840
    },
    {
      "epoch": 1.3467339576947177,
      "grad_norm": 0.6740279793739319,
      "learning_rate": 8.733628753478009e-05,
      "loss": 1.7187,
      "step": 24850
    },
    {
      "epoch": 1.34727589886023,
      "grad_norm": 0.4922647476196289,
      "learning_rate": 8.732527102243835e-05,
      "loss": 1.7185,
      "step": 24860
    },
    {
      "epoch": 1.3478178400257423,
      "grad_norm": 0.34301790595054626,
      "learning_rate": 8.731425050578238e-05,
      "loss": 1.709,
      "step": 24870
    },
    {
      "epoch": 1.3483597811912544,
      "grad_norm": 0.6136478185653687,
      "learning_rate": 8.730322598617734e-05,
      "loss": 1.7129,
      "step": 24880
    },
    {
      "epoch": 1.3484681694243568,
      "eval_loss": 2.5651490688323975,
      "eval_runtime": 21.9924,
      "eval_samples_per_second": 227.351,
      "eval_steps_per_second": 1.228,
      "step": 24882
    },
    {
      "epoch": 1.3489017223567665,
      "grad_norm": 0.44811177253723145,
      "learning_rate": 8.729219746498892e-05,
      "loss": 1.7175,
      "step": 24890
    },
    {
      "epoch": 1.3494436635222788,
      "grad_norm": 0.35493898391723633,
      "learning_rate": 8.728116494358325e-05,
      "loss": 1.7221,
      "step": 24900
    },
    {
      "epoch": 1.3499856046877912,
      "grad_norm": 0.3304137885570526,
      "learning_rate": 8.727012842332706e-05,
      "loss": 1.7252,
      "step": 24910
    },
    {
      "epoch": 1.3505275458533033,
      "grad_norm": 0.2520520091056824,
      "learning_rate": 8.725908790558746e-05,
      "loss": 1.7111,
      "step": 24920
    },
    {
      "epoch": 1.3510694870188156,
      "grad_norm": 0.48782554268836975,
      "learning_rate": 8.724804339173208e-05,
      "loss": 1.7112,
      "step": 24930
    },
    {
      "epoch": 1.3516114281843277,
      "grad_norm": 0.3623749017715454,
      "learning_rate": 8.72369948831291e-05,
      "loss": 1.7057,
      "step": 24940
    },
    {
      "epoch": 1.35215336934984,
      "grad_norm": 0.31124410033226013,
      "learning_rate": 8.722594238114716e-05,
      "loss": 1.702,
      "step": 24950
    },
    {
      "epoch": 1.3526953105153523,
      "grad_norm": 0.3612881600856781,
      "learning_rate": 8.721488588715539e-05,
      "loss": 1.7136,
      "step": 24960
    },
    {
      "epoch": 1.3531830575643131,
      "eval_loss": 2.5710861682891846,
      "eval_runtime": 21.9923,
      "eval_samples_per_second": 227.352,
      "eval_steps_per_second": 1.228,
      "step": 24969
    },
    {
      "epoch": 1.3532372516808644,
      "grad_norm": 0.3691853880882263,
      "learning_rate": 8.720382540252341e-05,
      "loss": 1.7128,
      "step": 24970
    },
    {
      "epoch": 1.3537791928463765,
      "grad_norm": 0.33288514614105225,
      "learning_rate": 8.719276092862137e-05,
      "loss": 1.6998,
      "step": 24980
    },
    {
      "epoch": 1.3543211340118888,
      "grad_norm": 0.28774330019950867,
      "learning_rate": 8.718169246681986e-05,
      "loss": 1.7154,
      "step": 24990
    },
    {
      "epoch": 1.3548630751774011,
      "grad_norm": 0.44046953320503235,
      "learning_rate": 8.717062001849001e-05,
      "loss": 1.6982,
      "step": 25000
    },
    {
      "epoch": 1.3554050163429132,
      "grad_norm": 0.40893176198005676,
      "learning_rate": 8.715954358500342e-05,
      "loss": 1.7179,
      "step": 25010
    },
    {
      "epoch": 1.3559469575084255,
      "grad_norm": 0.46221449971199036,
      "learning_rate": 8.714846316773218e-05,
      "loss": 1.7108,
      "step": 25020
    },
    {
      "epoch": 1.3564888986739376,
      "grad_norm": 0.4240012466907501,
      "learning_rate": 8.713737876804891e-05,
      "loss": 1.7114,
      "step": 25030
    },
    {
      "epoch": 1.35703083983945,
      "grad_norm": 0.39423850178718567,
      "learning_rate": 8.71262903873267e-05,
      "loss": 1.7035,
      "step": 25040
    },
    {
      "epoch": 1.3575727810049623,
      "grad_norm": 0.26944607496261597,
      "learning_rate": 8.711519802693911e-05,
      "loss": 1.7109,
      "step": 25050
    },
    {
      "epoch": 1.3578979457042695,
      "eval_loss": 2.564357280731201,
      "eval_runtime": 21.9966,
      "eval_samples_per_second": 227.308,
      "eval_steps_per_second": 1.227,
      "step": 25056
    },
    {
      "epoch": 1.3581147221704744,
      "grad_norm": 0.353392630815506,
      "learning_rate": 8.710410168826021e-05,
      "loss": 1.7087,
      "step": 25060
    },
    {
      "epoch": 1.3586566633359867,
      "grad_norm": 0.41630539298057556,
      "learning_rate": 8.709300137266456e-05,
      "loss": 1.7018,
      "step": 25070
    },
    {
      "epoch": 1.3591986045014988,
      "grad_norm": 0.3117140233516693,
      "learning_rate": 8.708189708152727e-05,
      "loss": 1.7139,
      "step": 25080
    },
    {
      "epoch": 1.359740545667011,
      "grad_norm": 0.522264301776886,
      "learning_rate": 8.707078881622381e-05,
      "loss": 1.7087,
      "step": 25090
    },
    {
      "epoch": 1.3602824868325232,
      "grad_norm": 0.2718927562236786,
      "learning_rate": 8.70596765781303e-05,
      "loss": 1.702,
      "step": 25100
    },
    {
      "epoch": 1.3608244279980355,
      "grad_norm": 0.5048019289970398,
      "learning_rate": 8.704856036862325e-05,
      "loss": 1.7174,
      "step": 25110
    },
    {
      "epoch": 1.3613663691635476,
      "grad_norm": 0.29533931612968445,
      "learning_rate": 8.703744018907966e-05,
      "loss": 1.7028,
      "step": 25120
    },
    {
      "epoch": 1.36190831032906,
      "grad_norm": 0.6198928356170654,
      "learning_rate": 8.702631604087709e-05,
      "loss": 1.7008,
      "step": 25130
    },
    {
      "epoch": 1.3624502514945722,
      "grad_norm": 0.3209061026573181,
      "learning_rate": 8.701518792539351e-05,
      "loss": 1.7072,
      "step": 25140
    },
    {
      "epoch": 1.3626128338442258,
      "eval_loss": 2.5585379600524902,
      "eval_runtime": 21.9846,
      "eval_samples_per_second": 227.432,
      "eval_steps_per_second": 1.228,
      "step": 25143
    },
    {
      "epoch": 1.3629921926600843,
      "grad_norm": 0.2755369246006012,
      "learning_rate": 8.700405584400747e-05,
      "loss": 1.7093,
      "step": 25150
    },
    {
      "epoch": 1.3635341338255966,
      "grad_norm": 0.24208010733127594,
      "learning_rate": 8.699291979809794e-05,
      "loss": 1.7136,
      "step": 25160
    },
    {
      "epoch": 1.3640760749911087,
      "grad_norm": 0.44155067205429077,
      "learning_rate": 8.69817797890444e-05,
      "loss": 1.6991,
      "step": 25170
    },
    {
      "epoch": 1.364618016156621,
      "grad_norm": 0.24571721255779266,
      "learning_rate": 8.697063581822682e-05,
      "loss": 1.6956,
      "step": 25180
    },
    {
      "epoch": 1.3651599573221331,
      "grad_norm": 0.2612999975681305,
      "learning_rate": 8.695948788702568e-05,
      "loss": 1.7069,
      "step": 25190
    },
    {
      "epoch": 1.3657018984876454,
      "grad_norm": 0.47256600856781006,
      "learning_rate": 8.694833599682194e-05,
      "loss": 1.6981,
      "step": 25200
    },
    {
      "epoch": 1.3662438396531575,
      "grad_norm": 0.3326663374900818,
      "learning_rate": 8.693718014899705e-05,
      "loss": 1.7145,
      "step": 25210
    },
    {
      "epoch": 1.3667857808186699,
      "grad_norm": 0.6608891487121582,
      "learning_rate": 8.692602034493292e-05,
      "loss": 1.6981,
      "step": 25220
    },
    {
      "epoch": 1.3673277219841822,
      "grad_norm": 0.3640691041946411,
      "learning_rate": 8.691485658601203e-05,
      "loss": 1.7105,
      "step": 25230
    },
    {
      "epoch": 1.3673277219841822,
      "eval_loss": 2.5621073246002197,
      "eval_runtime": 21.9283,
      "eval_samples_per_second": 228.016,
      "eval_steps_per_second": 1.231,
      "step": 25230
    },
    {
      "epoch": 1.3678696631496943,
      "grad_norm": 0.26905468106269836,
      "learning_rate": 8.690368887361724e-05,
      "loss": 1.7063,
      "step": 25240
    },
    {
      "epoch": 1.3684116043152066,
      "grad_norm": 0.33112001419067383,
      "learning_rate": 8.689251720913199e-05,
      "loss": 1.7043,
      "step": 25250
    },
    {
      "epoch": 1.3689535454807187,
      "grad_norm": 0.6896092891693115,
      "learning_rate": 8.688134159394017e-05,
      "loss": 1.7025,
      "step": 25260
    },
    {
      "epoch": 1.369495486646231,
      "grad_norm": 0.26553961634635925,
      "learning_rate": 8.687016202942617e-05,
      "loss": 1.7046,
      "step": 25270
    },
    {
      "epoch": 1.3700374278117433,
      "grad_norm": 0.6590327024459839,
      "learning_rate": 8.685897851697486e-05,
      "loss": 1.6894,
      "step": 25280
    },
    {
      "epoch": 1.3705793689772554,
      "grad_norm": 0.264779657125473,
      "learning_rate": 8.684779105797164e-05,
      "loss": 1.7135,
      "step": 25290
    },
    {
      "epoch": 1.3711213101427675,
      "grad_norm": 0.29008379578590393,
      "learning_rate": 8.68365996538023e-05,
      "loss": 1.7019,
      "step": 25300
    },
    {
      "epoch": 1.3716632513082798,
      "grad_norm": 0.28447601199150085,
      "learning_rate": 8.682540430585322e-05,
      "loss": 1.7155,
      "step": 25310
    },
    {
      "epoch": 1.3720426101241383,
      "eval_loss": 2.554466962814331,
      "eval_runtime": 21.9935,
      "eval_samples_per_second": 227.34,
      "eval_steps_per_second": 1.228,
      "step": 25317
    },
    {
      "epoch": 1.3722051924737921,
      "grad_norm": 0.4446844160556793,
      "learning_rate": 8.681420501551123e-05,
      "loss": 1.6982,
      "step": 25320
    },
    {
      "epoch": 1.3727471336393042,
      "grad_norm": 0.3506196439266205,
      "learning_rate": 8.680300178416364e-05,
      "loss": 1.6931,
      "step": 25330
    },
    {
      "epoch": 1.3732890748048165,
      "grad_norm": 0.28431805968284607,
      "learning_rate": 8.679179461319828e-05,
      "loss": 1.7094,
      "step": 25340
    },
    {
      "epoch": 1.3738310159703286,
      "grad_norm": 0.3544039726257324,
      "learning_rate": 8.678058350400342e-05,
      "loss": 1.6993,
      "step": 25350
    },
    {
      "epoch": 1.374372957135841,
      "grad_norm": 0.4839727580547333,
      "learning_rate": 8.676936845796785e-05,
      "loss": 1.712,
      "step": 25360
    },
    {
      "epoch": 1.3749148983013533,
      "grad_norm": 0.34992125630378723,
      "learning_rate": 8.675814947648085e-05,
      "loss": 1.7031,
      "step": 25370
    },
    {
      "epoch": 1.3754568394668654,
      "grad_norm": 0.37806737422943115,
      "learning_rate": 8.674692656093216e-05,
      "loss": 1.7053,
      "step": 25380
    },
    {
      "epoch": 1.3759987806323775,
      "grad_norm": 0.24221035838127136,
      "learning_rate": 8.673569971271206e-05,
      "loss": 1.702,
      "step": 25390
    },
    {
      "epoch": 1.3765407217978898,
      "grad_norm": 0.5913215279579163,
      "learning_rate": 8.672446893321125e-05,
      "loss": 1.6931,
      "step": 25400
    },
    {
      "epoch": 1.3767574982640947,
      "eval_loss": 2.552272081375122,
      "eval_runtime": 21.9916,
      "eval_samples_per_second": 227.36,
      "eval_steps_per_second": 1.228,
      "step": 25404
    },
    {
      "epoch": 1.377082662963402,
      "grad_norm": 0.29546642303466797,
      "learning_rate": 8.671323422382097e-05,
      "loss": 1.7001,
      "step": 25410
    },
    {
      "epoch": 1.3776246041289142,
      "grad_norm": 0.48717227578163147,
      "learning_rate": 8.67019955859329e-05,
      "loss": 1.7133,
      "step": 25420
    },
    {
      "epoch": 1.3781665452944265,
      "grad_norm": 0.3169041872024536,
      "learning_rate": 8.669075302093928e-05,
      "loss": 1.6972,
      "step": 25430
    },
    {
      "epoch": 1.3787084864599386,
      "grad_norm": 0.36612847447395325,
      "learning_rate": 8.667950653023274e-05,
      "loss": 1.6897,
      "step": 25440
    },
    {
      "epoch": 1.379250427625451,
      "grad_norm": 0.2583743929862976,
      "learning_rate": 8.666825611520648e-05,
      "loss": 1.7026,
      "step": 25450
    },
    {
      "epoch": 1.3797923687909632,
      "grad_norm": 0.34147560596466064,
      "learning_rate": 8.665700177725415e-05,
      "loss": 1.6969,
      "step": 25460
    },
    {
      "epoch": 1.3803343099564753,
      "grad_norm": 0.43001192808151245,
      "learning_rate": 8.664574351776988e-05,
      "loss": 1.708,
      "step": 25470
    },
    {
      "epoch": 1.3808762511219876,
      "grad_norm": 0.41845470666885376,
      "learning_rate": 8.663448133814826e-05,
      "loss": 1.7032,
      "step": 25480
    },
    {
      "epoch": 1.3814181922874997,
      "grad_norm": 0.3702855706214905,
      "learning_rate": 8.662321523978448e-05,
      "loss": 1.6965,
      "step": 25490
    },
    {
      "epoch": 1.381472386404051,
      "eval_loss": 2.5554749965667725,
      "eval_runtime": 21.99,
      "eval_samples_per_second": 227.376,
      "eval_steps_per_second": 1.228,
      "step": 25491
    },
    {
      "epoch": 1.381960133453012,
      "grad_norm": 0.4040187895298004,
      "learning_rate": 8.661194522407408e-05,
      "loss": 1.7025,
      "step": 25500
    },
    {
      "epoch": 1.3825020746185241,
      "grad_norm": 0.5149408578872681,
      "learning_rate": 8.660067129241315e-05,
      "loss": 1.698,
      "step": 25510
    },
    {
      "epoch": 1.3830440157840365,
      "grad_norm": 0.328133761882782,
      "learning_rate": 8.658939344619824e-05,
      "loss": 1.7068,
      "step": 25520
    },
    {
      "epoch": 1.3835859569495486,
      "grad_norm": 0.34496381878852844,
      "learning_rate": 8.657811168682644e-05,
      "loss": 1.7131,
      "step": 25530
    },
    {
      "epoch": 1.3841278981150609,
      "grad_norm": 0.29763221740722656,
      "learning_rate": 8.656682601569524e-05,
      "loss": 1.6992,
      "step": 25540
    },
    {
      "epoch": 1.3846698392805732,
      "grad_norm": 0.3687681257724762,
      "learning_rate": 8.655553643420268e-05,
      "loss": 1.7021,
      "step": 25550
    },
    {
      "epoch": 1.3852117804460853,
      "grad_norm": 0.36310508847236633,
      "learning_rate": 8.654424294374729e-05,
      "loss": 1.6958,
      "step": 25560
    },
    {
      "epoch": 1.3857537216115976,
      "grad_norm": 0.7248652577400208,
      "learning_rate": 8.653294554572802e-05,
      "loss": 1.7009,
      "step": 25570
    },
    {
      "epoch": 1.3861872745440074,
      "eval_loss": 2.5556063652038574,
      "eval_runtime": 21.9953,
      "eval_samples_per_second": 227.321,
      "eval_steps_per_second": 1.228,
      "step": 25578
    },
    {
      "epoch": 1.3862956627771097,
      "grad_norm": 0.2832872271537781,
      "learning_rate": 8.652164424154438e-05,
      "loss": 1.6906,
      "step": 25580
    },
    {
      "epoch": 1.386837603942622,
      "grad_norm": 0.6250040531158447,
      "learning_rate": 8.651033903259629e-05,
      "loss": 1.6885,
      "step": 25590
    },
    {
      "epoch": 1.3873795451081343,
      "grad_norm": 0.36408352851867676,
      "learning_rate": 8.649902992028419e-05,
      "loss": 1.697,
      "step": 25600
    },
    {
      "epoch": 1.3879214862736464,
      "grad_norm": 0.344611257314682,
      "learning_rate": 8.648771690600905e-05,
      "loss": 1.7002,
      "step": 25610
    },
    {
      "epoch": 1.3884634274391585,
      "grad_norm": 0.5186106562614441,
      "learning_rate": 8.647639999117221e-05,
      "loss": 1.6938,
      "step": 25620
    },
    {
      "epoch": 1.3890053686046708,
      "grad_norm": 0.3214333951473236,
      "learning_rate": 8.646507917717562e-05,
      "loss": 1.6959,
      "step": 25630
    },
    {
      "epoch": 1.3895473097701831,
      "grad_norm": 0.31978529691696167,
      "learning_rate": 8.645375446542162e-05,
      "loss": 1.6994,
      "step": 25640
    },
    {
      "epoch": 1.3900892509356952,
      "grad_norm": 0.3708828389644623,
      "learning_rate": 8.644242585731309e-05,
      "loss": 1.7054,
      "step": 25650
    },
    {
      "epoch": 1.3906311921012076,
      "grad_norm": 0.2623041272163391,
      "learning_rate": 8.643109335425335e-05,
      "loss": 1.6969,
      "step": 25660
    },
    {
      "epoch": 1.3909021626839637,
      "eval_loss": 2.5598154067993164,
      "eval_runtime": 21.9908,
      "eval_samples_per_second": 227.368,
      "eval_steps_per_second": 1.228,
      "step": 25665
    },
    {
      "epoch": 1.3911731332667197,
      "grad_norm": 0.23545025289058685,
      "learning_rate": 8.641975695764621e-05,
      "loss": 1.6934,
      "step": 25670
    },
    {
      "epoch": 1.391715074432232,
      "grad_norm": 0.31226447224617004,
      "learning_rate": 8.640841666889599e-05,
      "loss": 1.705,
      "step": 25680
    },
    {
      "epoch": 1.3922570155977443,
      "grad_norm": 0.25354868173599243,
      "learning_rate": 8.639707248940748e-05,
      "loss": 1.6927,
      "step": 25690
    },
    {
      "epoch": 1.3927989567632564,
      "grad_norm": 0.2758951783180237,
      "learning_rate": 8.638572442058596e-05,
      "loss": 1.6954,
      "step": 25700
    },
    {
      "epoch": 1.3933408979287685,
      "grad_norm": 0.2809247672557831,
      "learning_rate": 8.637437246383714e-05,
      "loss": 1.691,
      "step": 25710
    },
    {
      "epoch": 1.3938828390942808,
      "grad_norm": 0.27451246976852417,
      "learning_rate": 8.636301662056727e-05,
      "loss": 1.6959,
      "step": 25720
    },
    {
      "epoch": 1.394424780259793,
      "grad_norm": 0.5022933483123779,
      "learning_rate": 8.635165689218309e-05,
      "loss": 1.6908,
      "step": 25730
    },
    {
      "epoch": 1.3949667214253052,
      "grad_norm": 0.23696884512901306,
      "learning_rate": 8.634029328009174e-05,
      "loss": 1.6908,
      "step": 25740
    },
    {
      "epoch": 1.3955086625908175,
      "grad_norm": 0.2913939356803894,
      "learning_rate": 8.632892578570094e-05,
      "loss": 1.6963,
      "step": 25750
    },
    {
      "epoch": 1.3956170508239198,
      "eval_loss": 2.5667457580566406,
      "eval_runtime": 21.9931,
      "eval_samples_per_second": 227.344,
      "eval_steps_per_second": 1.228,
      "step": 25752
    },
    {
      "epoch": 1.3960506037563296,
      "grad_norm": 0.28175458312034607,
      "learning_rate": 8.631755441041884e-05,
      "loss": 1.7013,
      "step": 25760
    },
    {
      "epoch": 1.396592544921842,
      "grad_norm": 0.4736407399177551,
      "learning_rate": 8.630617915565405e-05,
      "loss": 1.7113,
      "step": 25770
    },
    {
      "epoch": 1.3971344860873542,
      "grad_norm": 0.33866068720817566,
      "learning_rate": 8.629480002281572e-05,
      "loss": 1.6952,
      "step": 25780
    },
    {
      "epoch": 1.3976764272528663,
      "grad_norm": 0.2977088689804077,
      "learning_rate": 8.628341701331341e-05,
      "loss": 1.7036,
      "step": 25790
    },
    {
      "epoch": 1.3982183684183787,
      "grad_norm": 0.26200008392333984,
      "learning_rate": 8.627203012855723e-05,
      "loss": 1.708,
      "step": 25800
    },
    {
      "epoch": 1.3987603095838907,
      "grad_norm": 0.6198749542236328,
      "learning_rate": 8.626063936995774e-05,
      "loss": 1.6924,
      "step": 25810
    },
    {
      "epoch": 1.399302250749403,
      "grad_norm": 0.6624993681907654,
      "learning_rate": 8.624924473892595e-05,
      "loss": 1.6978,
      "step": 25820
    },
    {
      "epoch": 1.3998441919149152,
      "grad_norm": 0.32168132066726685,
      "learning_rate": 8.62378462368734e-05,
      "loss": 1.7015,
      "step": 25830
    },
    {
      "epoch": 1.4003319389638762,
      "eval_loss": 2.560290575027466,
      "eval_runtime": 21.9932,
      "eval_samples_per_second": 227.343,
      "eval_steps_per_second": 1.228,
      "step": 25839
    },
    {
      "epoch": 1.4003861330804275,
      "grad_norm": 0.39742761850357056,
      "learning_rate": 8.622644386521207e-05,
      "loss": 1.6899,
      "step": 25840
    },
    {
      "epoch": 1.4009280742459396,
      "grad_norm": 0.4070427417755127,
      "learning_rate": 8.621503762535445e-05,
      "loss": 1.6905,
      "step": 25850
    },
    {
      "epoch": 1.4014700154114519,
      "grad_norm": 0.33487892150878906,
      "learning_rate": 8.620362751871349e-05,
      "loss": 1.6978,
      "step": 25860
    },
    {
      "epoch": 1.4020119565769642,
      "grad_norm": 0.2614098787307739,
      "learning_rate": 8.619221354670264e-05,
      "loss": 1.682,
      "step": 25870
    },
    {
      "epoch": 1.4025538977424763,
      "grad_norm": 0.3438650369644165,
      "learning_rate": 8.618079571073578e-05,
      "loss": 1.6924,
      "step": 25880
    },
    {
      "epoch": 1.4030958389079886,
      "grad_norm": 0.3044370710849762,
      "learning_rate": 8.616937401222734e-05,
      "loss": 1.6948,
      "step": 25890
    },
    {
      "epoch": 1.4036377800735007,
      "grad_norm": 0.3689815402030945,
      "learning_rate": 8.615794845259215e-05,
      "loss": 1.7012,
      "step": 25900
    },
    {
      "epoch": 1.404179721239013,
      "grad_norm": 0.2958444356918335,
      "learning_rate": 8.61465190332456e-05,
      "loss": 1.6935,
      "step": 25910
    },
    {
      "epoch": 1.4047216624045253,
      "grad_norm": 0.4378800392150879,
      "learning_rate": 8.613508575560348e-05,
      "loss": 1.6843,
      "step": 25920
    },
    {
      "epoch": 1.4050468271038326,
      "eval_loss": 2.5524935722351074,
      "eval_runtime": 21.9905,
      "eval_samples_per_second": 227.371,
      "eval_steps_per_second": 1.228,
      "step": 25926
    },
    {
      "epoch": 1.4052636035700374,
      "grad_norm": 0.2674407362937927,
      "learning_rate": 8.612364862108211e-05,
      "loss": 1.6896,
      "step": 25930
    },
    {
      "epoch": 1.4058055447355495,
      "grad_norm": 0.37812310457229614,
      "learning_rate": 8.611220763109829e-05,
      "loss": 1.6967,
      "step": 25940
    },
    {
      "epoch": 1.4063474859010618,
      "grad_norm": 0.25375431776046753,
      "learning_rate": 8.610076278706926e-05,
      "loss": 1.6858,
      "step": 25950
    },
    {
      "epoch": 1.4068894270665742,
      "grad_norm": 0.30164968967437744,
      "learning_rate": 8.608931409041273e-05,
      "loss": 1.695,
      "step": 25960
    },
    {
      "epoch": 1.4074313682320863,
      "grad_norm": 0.3200394809246063,
      "learning_rate": 8.607786154254699e-05,
      "loss": 1.6881,
      "step": 25970
    },
    {
      "epoch": 1.4079733093975986,
      "grad_norm": 0.3104444742202759,
      "learning_rate": 8.606640514489067e-05,
      "loss": 1.6972,
      "step": 25980
    },
    {
      "epoch": 1.4085152505631107,
      "grad_norm": 0.3415814936161041,
      "learning_rate": 8.605494489886297e-05,
      "loss": 1.6989,
      "step": 25990
    },
    {
      "epoch": 1.409057191728623,
      "grad_norm": 0.3201243281364441,
      "learning_rate": 8.604348080588351e-05,
      "loss": 1.6955,
      "step": 26000
    },
    {
      "epoch": 1.4095991328941353,
      "grad_norm": 0.2451564520597458,
      "learning_rate": 8.603201286737243e-05,
      "loss": 1.6988,
      "step": 26010
    },
    {
      "epoch": 1.409761715243789,
      "eval_loss": 2.557467222213745,
      "eval_runtime": 21.9947,
      "eval_samples_per_second": 227.328,
      "eval_steps_per_second": 1.228,
      "step": 26013
    },
    {
      "epoch": 1.4101410740596474,
      "grad_norm": 0.31296998262405396,
      "learning_rate": 8.602054108475032e-05,
      "loss": 1.6994,
      "step": 26020
    },
    {
      "epoch": 1.4106830152251595,
      "grad_norm": 0.6278992891311646,
      "learning_rate": 8.600906545943827e-05,
      "loss": 1.7028,
      "step": 26030
    },
    {
      "epoch": 1.4112249563906718,
      "grad_norm": 0.3286595642566681,
      "learning_rate": 8.59975859928578e-05,
      "loss": 1.7018,
      "step": 26040
    },
    {
      "epoch": 1.4117668975561841,
      "grad_norm": 0.26818498969078064,
      "learning_rate": 8.598610268643097e-05,
      "loss": 1.686,
      "step": 26050
    },
    {
      "epoch": 1.4123088387216962,
      "grad_norm": 0.3351867198944092,
      "learning_rate": 8.597461554158025e-05,
      "loss": 1.6841,
      "step": 26060
    },
    {
      "epoch": 1.4128507798872085,
      "grad_norm": 0.43873295187950134,
      "learning_rate": 8.596312455972866e-05,
      "loss": 1.7027,
      "step": 26070
    },
    {
      "epoch": 1.4133927210527206,
      "grad_norm": 0.3023853600025177,
      "learning_rate": 8.595162974229963e-05,
      "loss": 1.6949,
      "step": 26080
    },
    {
      "epoch": 1.413934662218233,
      "grad_norm": 0.30039265751838684,
      "learning_rate": 8.594013109071705e-05,
      "loss": 1.6887,
      "step": 26090
    },
    {
      "epoch": 1.4144766033837453,
      "grad_norm": 0.32320213317871094,
      "learning_rate": 8.59286286064054e-05,
      "loss": 1.6913,
      "step": 26100
    },
    {
      "epoch": 1.4144766033837453,
      "eval_loss": 2.557053327560425,
      "eval_runtime": 21.9348,
      "eval_samples_per_second": 227.948,
      "eval_steps_per_second": 1.231,
      "step": 26100
    },
    {
      "epoch": 1.4150185445492574,
      "grad_norm": 0.2611085772514343,
      "learning_rate": 8.591712229078949e-05,
      "loss": 1.6902,
      "step": 26110
    },
    {
      "epoch": 1.4155604857147697,
      "grad_norm": 0.6301568150520325,
      "learning_rate": 8.59056121452947e-05,
      "loss": 1.6871,
      "step": 26120
    },
    {
      "epoch": 1.4161024268802818,
      "grad_norm": 0.29995056986808777,
      "learning_rate": 8.589409817134685e-05,
      "loss": 1.6999,
      "step": 26130
    },
    {
      "epoch": 1.416644368045794,
      "grad_norm": 0.30796995759010315,
      "learning_rate": 8.588258037037227e-05,
      "loss": 1.6911,
      "step": 26140
    },
    {
      "epoch": 1.4171863092113062,
      "grad_norm": 0.28592872619628906,
      "learning_rate": 8.58710587437977e-05,
      "loss": 1.6882,
      "step": 26150
    },
    {
      "epoch": 1.4177282503768185,
      "grad_norm": 0.264588862657547,
      "learning_rate": 8.58595332930504e-05,
      "loss": 1.6967,
      "step": 26160
    },
    {
      "epoch": 1.4182701915423306,
      "grad_norm": 0.3666386306285858,
      "learning_rate": 8.58480040195581e-05,
      "loss": 1.682,
      "step": 26170
    },
    {
      "epoch": 1.418812132707843,
      "grad_norm": 0.38330572843551636,
      "learning_rate": 8.5836470924749e-05,
      "loss": 1.6829,
      "step": 26180
    },
    {
      "epoch": 1.4191914915237014,
      "eval_loss": 2.5586886405944824,
      "eval_runtime": 21.9928,
      "eval_samples_per_second": 227.348,
      "eval_steps_per_second": 1.228,
      "step": 26187
    },
    {
      "epoch": 1.4193540738733552,
      "grad_norm": 0.32240378856658936,
      "learning_rate": 8.582493401005175e-05,
      "loss": 1.6968,
      "step": 26190
    },
    {
      "epoch": 1.4198960150388673,
      "grad_norm": 0.3624284267425537,
      "learning_rate": 8.581339327689549e-05,
      "loss": 1.7,
      "step": 26200
    },
    {
      "epoch": 1.4204379562043796,
      "grad_norm": 0.3740536868572235,
      "learning_rate": 8.580184872670987e-05,
      "loss": 1.6942,
      "step": 26210
    },
    {
      "epoch": 1.4209798973698917,
      "grad_norm": 0.5000831484794617,
      "learning_rate": 8.579030036092495e-05,
      "loss": 1.6869,
      "step": 26220
    },
    {
      "epoch": 1.421521838535404,
      "grad_norm": 0.4658829867839813,
      "learning_rate": 8.57787481809713e-05,
      "loss": 1.6873,
      "step": 26230
    },
    {
      "epoch": 1.4220637797009164,
      "grad_norm": 0.3881395161151886,
      "learning_rate": 8.576719218827996e-05,
      "loss": 1.6912,
      "step": 26240
    },
    {
      "epoch": 1.4226057208664284,
      "grad_norm": 0.6568459868431091,
      "learning_rate": 8.575563238428241e-05,
      "loss": 1.6982,
      "step": 26250
    },
    {
      "epoch": 1.4231476620319405,
      "grad_norm": 0.2987072765827179,
      "learning_rate": 8.574406877041067e-05,
      "loss": 1.6951,
      "step": 26260
    },
    {
      "epoch": 1.4236896031974529,
      "grad_norm": 0.3211478590965271,
      "learning_rate": 8.573250134809715e-05,
      "loss": 1.7003,
      "step": 26270
    },
    {
      "epoch": 1.4239063796636577,
      "eval_loss": 2.5507450103759766,
      "eval_runtime": 21.9925,
      "eval_samples_per_second": 227.35,
      "eval_steps_per_second": 1.228,
      "step": 26274
    },
    {
      "epoch": 1.4242315443629652,
      "grad_norm": 0.327347993850708,
      "learning_rate": 8.57209301187748e-05,
      "loss": 1.6815,
      "step": 26280
    },
    {
      "epoch": 1.4247734855284773,
      "grad_norm": 0.3275546133518219,
      "learning_rate": 8.5709355083877e-05,
      "loss": 1.6945,
      "step": 26290
    },
    {
      "epoch": 1.4253154266939896,
      "grad_norm": 0.36542415618896484,
      "learning_rate": 8.569777624483763e-05,
      "loss": 1.695,
      "step": 26300
    },
    {
      "epoch": 1.4258573678595017,
      "grad_norm": 0.52955561876297,
      "learning_rate": 8.568619360309102e-05,
      "loss": 1.6802,
      "step": 26310
    },
    {
      "epoch": 1.426399309025014,
      "grad_norm": 0.9618374109268188,
      "learning_rate": 8.567460716007195e-05,
      "loss": 1.6978,
      "step": 26320
    },
    {
      "epoch": 1.4269412501905263,
      "grad_norm": 0.7096937298774719,
      "learning_rate": 8.566301691721574e-05,
      "loss": 1.6956,
      "step": 26330
    },
    {
      "epoch": 1.4274831913560384,
      "grad_norm": 0.2932894825935364,
      "learning_rate": 8.56514228759581e-05,
      "loss": 1.6919,
      "step": 26340
    },
    {
      "epoch": 1.4280251325215505,
      "grad_norm": 0.38948243856430054,
      "learning_rate": 8.563982503773527e-05,
      "loss": 1.6941,
      "step": 26350
    },
    {
      "epoch": 1.4285670736870628,
      "grad_norm": 0.2520905137062073,
      "learning_rate": 8.562822340398395e-05,
      "loss": 1.6801,
      "step": 26360
    },
    {
      "epoch": 1.428621267803614,
      "eval_loss": 2.5581541061401367,
      "eval_runtime": 21.991,
      "eval_samples_per_second": 227.366,
      "eval_steps_per_second": 1.228,
      "step": 26361
    },
    {
      "epoch": 1.4291090148525751,
      "grad_norm": 0.26686540246009827,
      "learning_rate": 8.56166179761413e-05,
      "loss": 1.6883,
      "step": 26370
    },
    {
      "epoch": 1.4296509560180872,
      "grad_norm": 0.45527172088623047,
      "learning_rate": 8.560500875564491e-05,
      "loss": 1.6885,
      "step": 26380
    },
    {
      "epoch": 1.4301928971835995,
      "grad_norm": 0.33151310682296753,
      "learning_rate": 8.559339574393294e-05,
      "loss": 1.6901,
      "step": 26390
    },
    {
      "epoch": 1.4307348383491116,
      "grad_norm": 0.4582711458206177,
      "learning_rate": 8.558177894244392e-05,
      "loss": 1.6876,
      "step": 26400
    },
    {
      "epoch": 1.431276779514624,
      "grad_norm": 0.23842747509479523,
      "learning_rate": 8.557015835261688e-05,
      "loss": 1.7054,
      "step": 26410
    },
    {
      "epoch": 1.4318187206801363,
      "grad_norm": 0.4057098627090454,
      "learning_rate": 8.555853397589136e-05,
      "loss": 1.7013,
      "step": 26420
    },
    {
      "epoch": 1.4323606618456484,
      "grad_norm": 0.4992932081222534,
      "learning_rate": 8.554690581370732e-05,
      "loss": 1.6931,
      "step": 26430
    },
    {
      "epoch": 1.4329026030111607,
      "grad_norm": 0.4780980944633484,
      "learning_rate": 8.553527386750521e-05,
      "loss": 1.6896,
      "step": 26440
    },
    {
      "epoch": 1.4333361559435704,
      "eval_loss": 2.556183099746704,
      "eval_runtime": 21.9936,
      "eval_samples_per_second": 227.339,
      "eval_steps_per_second": 1.228,
      "step": 26448
    },
    {
      "epoch": 1.4334445441766728,
      "grad_norm": 0.5446142554283142,
      "learning_rate": 8.552363813872594e-05,
      "loss": 1.6788,
      "step": 26450
    },
    {
      "epoch": 1.433986485342185,
      "grad_norm": 0.48127421736717224,
      "learning_rate": 8.551199862881091e-05,
      "loss": 1.6734,
      "step": 26460
    },
    {
      "epoch": 1.4345284265076972,
      "grad_norm": 0.30122366547584534,
      "learning_rate": 8.550035533920194e-05,
      "loss": 1.6897,
      "step": 26470
    },
    {
      "epoch": 1.4350703676732095,
      "grad_norm": 0.36052316427230835,
      "learning_rate": 8.548870827134139e-05,
      "loss": 1.6886,
      "step": 26480
    },
    {
      "epoch": 1.4356123088387216,
      "grad_norm": 0.47297677397727966,
      "learning_rate": 8.547705742667203e-05,
      "loss": 1.6943,
      "step": 26490
    },
    {
      "epoch": 1.436154250004234,
      "grad_norm": 0.26182791590690613,
      "learning_rate": 8.54654028066371e-05,
      "loss": 1.6938,
      "step": 26500
    },
    {
      "epoch": 1.4366961911697462,
      "grad_norm": 0.3834492564201355,
      "learning_rate": 8.545374441268033e-05,
      "loss": 1.6822,
      "step": 26510
    },
    {
      "epoch": 1.4372381323352583,
      "grad_norm": 0.29216572642326355,
      "learning_rate": 8.544208224624592e-05,
      "loss": 1.6838,
      "step": 26520
    },
    {
      "epoch": 1.4377800735007706,
      "grad_norm": 0.2463953047990799,
      "learning_rate": 8.543041630877856e-05,
      "loss": 1.6817,
      "step": 26530
    },
    {
      "epoch": 1.4380510440835268,
      "eval_loss": 2.5639803409576416,
      "eval_runtime": 21.9922,
      "eval_samples_per_second": 227.353,
      "eval_steps_per_second": 1.228,
      "step": 26535
    },
    {
      "epoch": 1.4383220146662827,
      "grad_norm": 0.6637621521949768,
      "learning_rate": 8.54187466017233e-05,
      "loss": 1.6862,
      "step": 26540
    },
    {
      "epoch": 1.438863955831795,
      "grad_norm": 0.33043143153190613,
      "learning_rate": 8.540707312652581e-05,
      "loss": 1.704,
      "step": 26550
    },
    {
      "epoch": 1.4394058969973074,
      "grad_norm": 0.36580994725227356,
      "learning_rate": 8.53953958846321e-05,
      "loss": 1.6747,
      "step": 26560
    },
    {
      "epoch": 1.4399478381628195,
      "grad_norm": 0.5655596256256104,
      "learning_rate": 8.538371487748871e-05,
      "loss": 1.688,
      "step": 26570
    },
    {
      "epoch": 1.4404897793283316,
      "grad_norm": 0.41566041111946106,
      "learning_rate": 8.537203010654266e-05,
      "loss": 1.6908,
      "step": 26580
    },
    {
      "epoch": 1.4410317204938439,
      "grad_norm": 0.3370110094547272,
      "learning_rate": 8.536034157324135e-05,
      "loss": 1.6906,
      "step": 26590
    },
    {
      "epoch": 1.4415736616593562,
      "grad_norm": 0.4993593692779541,
      "learning_rate": 8.534864927903274e-05,
      "loss": 1.6829,
      "step": 26600
    },
    {
      "epoch": 1.4421156028248683,
      "grad_norm": 0.28498584032058716,
      "learning_rate": 8.533695322536523e-05,
      "loss": 1.6933,
      "step": 26610
    },
    {
      "epoch": 1.4426575439903806,
      "grad_norm": 0.7398360371589661,
      "learning_rate": 8.532525341368767e-05,
      "loss": 1.686,
      "step": 26620
    },
    {
      "epoch": 1.442765932223483,
      "eval_loss": 2.5482335090637207,
      "eval_runtime": 21.988,
      "eval_samples_per_second": 227.397,
      "eval_steps_per_second": 1.228,
      "step": 26622
    },
    {
      "epoch": 1.4431994851558927,
      "grad_norm": 0.40166711807250977,
      "learning_rate": 8.531354984544936e-05,
      "loss": 1.6862,
      "step": 26630
    },
    {
      "epoch": 1.443741426321405,
      "grad_norm": 0.43468061089515686,
      "learning_rate": 8.530184252210012e-05,
      "loss": 1.6952,
      "step": 26640
    },
    {
      "epoch": 1.4442833674869173,
      "grad_norm": 0.39255911111831665,
      "learning_rate": 8.529013144509017e-05,
      "loss": 1.6928,
      "step": 26650
    },
    {
      "epoch": 1.4448253086524294,
      "grad_norm": 0.3616562783718109,
      "learning_rate": 8.527841661587024e-05,
      "loss": 1.6716,
      "step": 26660
    },
    {
      "epoch": 1.4453672498179415,
      "grad_norm": 0.5667160153388977,
      "learning_rate": 8.526669803589153e-05,
      "loss": 1.6842,
      "step": 26670
    },
    {
      "epoch": 1.4459091909834538,
      "grad_norm": 0.6671094298362732,
      "learning_rate": 8.525497570660567e-05,
      "loss": 1.6756,
      "step": 26680
    },
    {
      "epoch": 1.4464511321489661,
      "grad_norm": 0.4949670732021332,
      "learning_rate": 8.524324962946477e-05,
      "loss": 1.6857,
      "step": 26690
    },
    {
      "epoch": 1.4469930733144782,
      "grad_norm": 0.2640099823474884,
      "learning_rate": 8.52315198059214e-05,
      "loss": 1.6939,
      "step": 26700
    },
    {
      "epoch": 1.4474808203634393,
      "eval_loss": 2.541543960571289,
      "eval_runtime": 21.9948,
      "eval_samples_per_second": 227.326,
      "eval_steps_per_second": 1.228,
      "step": 26709
    },
    {
      "epoch": 1.4475350144799906,
      "grad_norm": 0.24292974174022675,
      "learning_rate": 8.521978623742861e-05,
      "loss": 1.6733,
      "step": 26710
    },
    {
      "epoch": 1.4480769556455027,
      "grad_norm": 0.40531477332115173,
      "learning_rate": 8.520804892543993e-05,
      "loss": 1.6769,
      "step": 26720
    },
    {
      "epoch": 1.448618896811015,
      "grad_norm": 0.37671342492103577,
      "learning_rate": 8.519630787140926e-05,
      "loss": 1.6894,
      "step": 26730
    },
    {
      "epoch": 1.4491608379765273,
      "grad_norm": 0.3520098328590393,
      "learning_rate": 8.51845630767911e-05,
      "loss": 1.6852,
      "step": 26740
    },
    {
      "epoch": 1.4497027791420394,
      "grad_norm": 0.3429430425167084,
      "learning_rate": 8.517281454304031e-05,
      "loss": 1.6855,
      "step": 26750
    },
    {
      "epoch": 1.4502447203075517,
      "grad_norm": 0.22939632833003998,
      "learning_rate": 8.516106227161223e-05,
      "loss": 1.6898,
      "step": 26760
    },
    {
      "epoch": 1.4507866614730638,
      "grad_norm": 0.2420610934495926,
      "learning_rate": 8.514930626396273e-05,
      "loss": 1.6847,
      "step": 26770
    },
    {
      "epoch": 1.451328602638576,
      "grad_norm": 0.3113659620285034,
      "learning_rate": 8.513754652154805e-05,
      "loss": 1.6867,
      "step": 26780
    },
    {
      "epoch": 1.4518705438040882,
      "grad_norm": 0.34163153171539307,
      "learning_rate": 8.512578304582495e-05,
      "loss": 1.695,
      "step": 26790
    },
    {
      "epoch": 1.4521957085033956,
      "eval_loss": 2.5400216579437256,
      "eval_runtime": 21.9954,
      "eval_samples_per_second": 227.32,
      "eval_steps_per_second": 1.228,
      "step": 26796
    },
    {
      "epoch": 1.4524124849696005,
      "grad_norm": 0.26067569851875305,
      "learning_rate": 8.511401583825066e-05,
      "loss": 1.6843,
      "step": 26800
    },
    {
      "epoch": 1.4529544261351126,
      "grad_norm": 0.34139981865882874,
      "learning_rate": 8.51022449002828e-05,
      "loss": 1.68,
      "step": 26810
    },
    {
      "epoch": 1.453496367300625,
      "grad_norm": 0.40918228030204773,
      "learning_rate": 8.509047023337954e-05,
      "loss": 1.6758,
      "step": 26820
    },
    {
      "epoch": 1.4540383084661372,
      "grad_norm": 0.33097583055496216,
      "learning_rate": 8.507869183899947e-05,
      "loss": 1.6781,
      "step": 26830
    },
    {
      "epoch": 1.4545802496316493,
      "grad_norm": 0.5380516648292542,
      "learning_rate": 8.506690971860167e-05,
      "loss": 1.6753,
      "step": 26840
    },
    {
      "epoch": 1.4551221907971617,
      "grad_norm": 0.3190267086029053,
      "learning_rate": 8.50551238736456e-05,
      "loss": 1.6841,
      "step": 26850
    },
    {
      "epoch": 1.4556641319626737,
      "grad_norm": 0.40389493107795715,
      "learning_rate": 8.504333430559127e-05,
      "loss": 1.687,
      "step": 26860
    },
    {
      "epoch": 1.456206073128186,
      "grad_norm": 0.3580215573310852,
      "learning_rate": 8.503154101589915e-05,
      "loss": 1.6818,
      "step": 26870
    },
    {
      "epoch": 1.4567480142936984,
      "grad_norm": 0.2765572667121887,
      "learning_rate": 8.501974400603009e-05,
      "loss": 1.6859,
      "step": 26880
    },
    {
      "epoch": 1.456910596643352,
      "eval_loss": 2.5574100017547607,
      "eval_runtime": 21.9907,
      "eval_samples_per_second": 227.368,
      "eval_steps_per_second": 1.228,
      "step": 26883
    },
    {
      "epoch": 1.4572899554592105,
      "grad_norm": 0.2352159172296524,
      "learning_rate": 8.500794327744547e-05,
      "loss": 1.679,
      "step": 26890
    },
    {
      "epoch": 1.4578318966247226,
      "grad_norm": 0.241908460855484,
      "learning_rate": 8.499613883160712e-05,
      "loss": 1.6749,
      "step": 26900
    },
    {
      "epoch": 1.4583738377902349,
      "grad_norm": 0.46047091484069824,
      "learning_rate": 8.498433066997733e-05,
      "loss": 1.6867,
      "step": 26910
    },
    {
      "epoch": 1.4589157789557472,
      "grad_norm": 0.4493439793586731,
      "learning_rate": 8.497251879401884e-05,
      "loss": 1.6741,
      "step": 26920
    },
    {
      "epoch": 1.4594577201212593,
      "grad_norm": 0.26538506150245667,
      "learning_rate": 8.496070320519484e-05,
      "loss": 1.6759,
      "step": 26930
    },
    {
      "epoch": 1.4599996612867716,
      "grad_norm": 0.3610468804836273,
      "learning_rate": 8.494888390496901e-05,
      "loss": 1.6859,
      "step": 26940
    },
    {
      "epoch": 1.4605416024522837,
      "grad_norm": 0.3829333782196045,
      "learning_rate": 8.493706089480546e-05,
      "loss": 1.6795,
      "step": 26950
    },
    {
      "epoch": 1.461083543617796,
      "grad_norm": 0.26342180371284485,
      "learning_rate": 8.492523417616878e-05,
      "loss": 1.6823,
      "step": 26960
    },
    {
      "epoch": 1.4616254847833083,
      "grad_norm": 0.3344675302505493,
      "learning_rate": 8.491340375052403e-05,
      "loss": 1.6862,
      "step": 26970
    },
    {
      "epoch": 1.4616254847833083,
      "eval_loss": 2.554945230484009,
      "eval_runtime": 21.9254,
      "eval_samples_per_second": 228.046,
      "eval_steps_per_second": 1.231,
      "step": 26970
    },
    {
      "epoch": 1.4621674259488204,
      "grad_norm": 0.28914472460746765,
      "learning_rate": 8.490156961933666e-05,
      "loss": 1.6751,
      "step": 26980
    },
    {
      "epoch": 1.4627093671143325,
      "grad_norm": 0.46620985865592957,
      "learning_rate": 8.488973178407269e-05,
      "loss": 1.689,
      "step": 26990
    },
    {
      "epoch": 1.4632513082798448,
      "grad_norm": 0.2889445722103119,
      "learning_rate": 8.487789024619852e-05,
      "loss": 1.6843,
      "step": 27000
    },
    {
      "epoch": 1.0005419411655123,
      "grad_norm": 0.23382456600666046,
      "learning_rate": 8.486604500718102e-05,
      "loss": 1.6714,
      "step": 27010
    },
    {
      "epoch": 1.0010838823310244,
      "grad_norm": 0.554624617099762,
      "learning_rate": 8.485419606848751e-05,
      "loss": 1.675,
      "step": 27020
    },
    {
      "epoch": 1.0016258234965367,
      "grad_norm": 0.48615115880966187,
      "learning_rate": 8.484234343158583e-05,
      "loss": 1.6822,
      "step": 27030
    },
    {
      "epoch": 1.0021677646620488,
      "grad_norm": 0.348832368850708,
      "learning_rate": 8.48304870979442e-05,
      "loss": 1.6722,
      "step": 27040
    },
    {
      "epoch": 1.0027097058275611,
      "grad_norm": 0.5944726467132568,
      "learning_rate": 8.481862706903132e-05,
      "loss": 1.6743,
      "step": 27050
    },
    {
      "epoch": 1.0030890646434196,
      "eval_loss": 2.5551207065582275,
      "eval_runtime": 22.1704,
      "eval_samples_per_second": 225.526,
      "eval_steps_per_second": 1.218,
      "step": 27057
    },
    {
      "epoch": 1.0032516469930732,
      "grad_norm": 0.34727635979652405,
      "learning_rate": 8.48067633463164e-05,
      "loss": 1.6697,
      "step": 27060
    },
    {
      "epoch": 1.0037935881585855,
      "grad_norm": 0.37410902976989746,
      "learning_rate": 8.479489593126904e-05,
      "loss": 1.666,
      "step": 27070
    },
    {
      "epoch": 1.0043355293240976,
      "grad_norm": 0.3332069516181946,
      "learning_rate": 8.478302482535933e-05,
      "loss": 1.6891,
      "step": 27080
    },
    {
      "epoch": 1.00487747048961,
      "grad_norm": 0.2845593988895416,
      "learning_rate": 8.477115003005781e-05,
      "loss": 1.6745,
      "step": 27090
    },
    {
      "epoch": 1.0054194116551223,
      "grad_norm": 0.367365300655365,
      "learning_rate": 8.475927154683547e-05,
      "loss": 1.6853,
      "step": 27100
    },
    {
      "epoch": 1.0059613528206344,
      "grad_norm": 0.2617562711238861,
      "learning_rate": 8.474738937716378e-05,
      "loss": 1.6763,
      "step": 27110
    },
    {
      "epoch": 1.0065032939861467,
      "grad_norm": 0.4524533748626709,
      "learning_rate": 8.473550352251465e-05,
      "loss": 1.6789,
      "step": 27120
    },
    {
      "epoch": 1.0070452351516588,
      "grad_norm": 0.39219167828559875,
      "learning_rate": 8.472361398436043e-05,
      "loss": 1.6915,
      "step": 27130
    },
    {
      "epoch": 1.007587176317171,
      "grad_norm": 0.47083818912506104,
      "learning_rate": 8.471172076417397e-05,
      "loss": 1.6807,
      "step": 27140
    },
    {
      "epoch": 1.007803952783376,
      "eval_loss": 2.5521044731140137,
      "eval_runtime": 21.9818,
      "eval_samples_per_second": 227.461,
      "eval_steps_per_second": 1.228,
      "step": 27144
    },
    {
      "epoch": 1.0081291174826832,
      "grad_norm": 0.33303186297416687,
      "learning_rate": 8.469982386342851e-05,
      "loss": 1.6765,
      "step": 27150
    },
    {
      "epoch": 1.0086710586481955,
      "grad_norm": 0.28193533420562744,
      "learning_rate": 8.468792328359783e-05,
      "loss": 1.674,
      "step": 27160
    },
    {
      "epoch": 1.0092129998137078,
      "grad_norm": 0.35703226923942566,
      "learning_rate": 8.467601902615613e-05,
      "loss": 1.6827,
      "step": 27170
    },
    {
      "epoch": 1.00975494097922,
      "grad_norm": 0.36111465096473694,
      "learning_rate": 8.4664111092578e-05,
      "loss": 1.6728,
      "step": 27180
    },
    {
      "epoch": 1.0102968821447322,
      "grad_norm": 0.28821465373039246,
      "learning_rate": 8.465219948433858e-05,
      "loss": 1.6828,
      "step": 27190
    },
    {
      "epoch": 1.0108388233102443,
      "grad_norm": 0.3027511537075043,
      "learning_rate": 8.464028420291344e-05,
      "loss": 1.6751,
      "step": 27200
    },
    {
      "epoch": 1.0113807644757566,
      "grad_norm": 0.28520819544792175,
      "learning_rate": 8.462836524977856e-05,
      "loss": 1.6664,
      "step": 27210
    },
    {
      "epoch": 1.0119227056412687,
      "grad_norm": 0.4790101647377014,
      "learning_rate": 8.461644262641044e-05,
      "loss": 1.6729,
      "step": 27220
    },
    {
      "epoch": 1.012464646806781,
      "grad_norm": 0.43564823269844055,
      "learning_rate": 8.460451633428599e-05,
      "loss": 1.6749,
      "step": 27230
    },
    {
      "epoch": 1.0125188409233323,
      "eval_loss": 2.557983636856079,
      "eval_runtime": 21.9815,
      "eval_samples_per_second": 227.464,
      "eval_steps_per_second": 1.228,
      "step": 27231
    },
    {
      "epoch": 1.0130065879722931,
      "grad_norm": 0.24843677878379822,
      "learning_rate": 8.459258637488259e-05,
      "loss": 1.6682,
      "step": 27240
    },
    {
      "epoch": 1.0135485291378055,
      "grad_norm": 0.2693169414997101,
      "learning_rate": 8.458065274967806e-05,
      "loss": 1.6843,
      "step": 27250
    },
    {
      "epoch": 1.0140904703033178,
      "grad_norm": 0.2424820065498352,
      "learning_rate": 8.45687154601507e-05,
      "loss": 1.6775,
      "step": 27260
    },
    {
      "epoch": 1.0146324114688299,
      "grad_norm": 0.37894406914711,
      "learning_rate": 8.455677450777922e-05,
      "loss": 1.6716,
      "step": 27270
    },
    {
      "epoch": 1.0151743526343422,
      "grad_norm": 0.771432638168335,
      "learning_rate": 8.454482989404286e-05,
      "loss": 1.6872,
      "step": 27280
    },
    {
      "epoch": 1.0157162937998543,
      "grad_norm": 0.28843340277671814,
      "learning_rate": 8.453288162042124e-05,
      "loss": 1.681,
      "step": 27290
    },
    {
      "epoch": 1.0162582349653666,
      "grad_norm": 0.31491175293922424,
      "learning_rate": 8.452092968839446e-05,
      "loss": 1.6717,
      "step": 27300
    },
    {
      "epoch": 1.0168001761308787,
      "grad_norm": 0.38847339153289795,
      "learning_rate": 8.450897409944309e-05,
      "loss": 1.6786,
      "step": 27310
    },
    {
      "epoch": 1.0172337290632885,
      "eval_loss": 2.552642345428467,
      "eval_runtime": 21.9929,
      "eval_samples_per_second": 227.346,
      "eval_steps_per_second": 1.228,
      "step": 27318
    },
    {
      "epoch": 1.017342117296391,
      "grad_norm": 0.6366086602210999,
      "learning_rate": 8.44970148550481e-05,
      "loss": 1.6791,
      "step": 27320
    },
    {
      "epoch": 1.0178840584619033,
      "grad_norm": 0.5009103417396545,
      "learning_rate": 8.448505195669099e-05,
      "loss": 1.6762,
      "step": 27330
    },
    {
      "epoch": 1.0184259996274154,
      "grad_norm": 0.31960320472717285,
      "learning_rate": 8.447308540585364e-05,
      "loss": 1.676,
      "step": 27340
    },
    {
      "epoch": 1.0189679407929277,
      "grad_norm": 0.5080009698867798,
      "learning_rate": 8.446111520401845e-05,
      "loss": 1.6768,
      "step": 27350
    },
    {
      "epoch": 1.0195098819584398,
      "grad_norm": 0.44143038988113403,
      "learning_rate": 8.44491413526682e-05,
      "loss": 1.6816,
      "step": 27360
    },
    {
      "epoch": 1.0200518231239522,
      "grad_norm": 0.3140455186367035,
      "learning_rate": 8.443716385328618e-05,
      "loss": 1.6696,
      "step": 27370
    },
    {
      "epoch": 1.0205937642894642,
      "grad_norm": 0.5069706439971924,
      "learning_rate": 8.442518270735611e-05,
      "loss": 1.6752,
      "step": 27380
    },
    {
      "epoch": 1.0211357054549766,
      "grad_norm": 1.0147614479064941,
      "learning_rate": 8.441319791636215e-05,
      "loss": 1.6684,
      "step": 27390
    },
    {
      "epoch": 1.0216776466204887,
      "grad_norm": 0.2649192214012146,
      "learning_rate": 8.440120948178895e-05,
      "loss": 1.6823,
      "step": 27400
    },
    {
      "epoch": 1.0219486172032448,
      "eval_loss": 2.5535383224487305,
      "eval_runtime": 21.989,
      "eval_samples_per_second": 227.386,
      "eval_steps_per_second": 1.228,
      "step": 27405
    },
    {
      "epoch": 1.022219587786001,
      "grad_norm": 0.2985130250453949,
      "learning_rate": 8.438921740512153e-05,
      "loss": 1.6679,
      "step": 27410
    },
    {
      "epoch": 1.0227615289515133,
      "grad_norm": 0.3355660140514374,
      "learning_rate": 8.437722168784548e-05,
      "loss": 1.6648,
      "step": 27420
    },
    {
      "epoch": 1.0233034701170254,
      "grad_norm": 0.30278173089027405,
      "learning_rate": 8.436522233144675e-05,
      "loss": 1.6676,
      "step": 27430
    },
    {
      "epoch": 1.0238454112825377,
      "grad_norm": 0.24776387214660645,
      "learning_rate": 8.435321933741177e-05,
      "loss": 1.6718,
      "step": 27440
    },
    {
      "epoch": 1.0243873524480498,
      "grad_norm": 0.25117209553718567,
      "learning_rate": 8.434121270722741e-05,
      "loss": 1.6791,
      "step": 27450
    },
    {
      "epoch": 1.024929293613562,
      "grad_norm": 0.23679675161838531,
      "learning_rate": 8.4329202442381e-05,
      "loss": 1.6793,
      "step": 27460
    },
    {
      "epoch": 1.0254712347790742,
      "grad_norm": 0.4516010582447052,
      "learning_rate": 8.431718854436034e-05,
      "loss": 1.6673,
      "step": 27470
    },
    {
      "epoch": 1.0260131759445865,
      "grad_norm": 0.2465139925479889,
      "learning_rate": 8.430517101465364e-05,
      "loss": 1.6743,
      "step": 27480
    },
    {
      "epoch": 1.0265551171100988,
      "grad_norm": 0.3209734857082367,
      "learning_rate": 8.429314985474958e-05,
      "loss": 1.67,
      "step": 27490
    },
    {
      "epoch": 1.0266635053432012,
      "eval_loss": 2.5473275184631348,
      "eval_runtime": 21.9881,
      "eval_samples_per_second": 227.396,
      "eval_steps_per_second": 1.228,
      "step": 27492
    },
    {
      "epoch": 1.027097058275611,
      "grad_norm": 0.2319590449333191,
      "learning_rate": 8.42811250661373e-05,
      "loss": 1.6803,
      "step": 27500
    },
    {
      "epoch": 1.0276389994411232,
      "grad_norm": 0.3213670253753662,
      "learning_rate": 8.426909665030637e-05,
      "loss": 1.6656,
      "step": 27510
    },
    {
      "epoch": 1.0281809406066353,
      "grad_norm": 0.29917410016059875,
      "learning_rate": 8.425706460874682e-05,
      "loss": 1.6738,
      "step": 27520
    },
    {
      "epoch": 1.0287228817721477,
      "grad_norm": 0.24403968453407288,
      "learning_rate": 8.424502894294913e-05,
      "loss": 1.6695,
      "step": 27530
    },
    {
      "epoch": 1.0292648229376598,
      "grad_norm": 0.3891294002532959,
      "learning_rate": 8.423298965440423e-05,
      "loss": 1.6619,
      "step": 27540
    },
    {
      "epoch": 1.029806764103172,
      "grad_norm": 0.4318540692329407,
      "learning_rate": 8.422094674460348e-05,
      "loss": 1.6751,
      "step": 27550
    },
    {
      "epoch": 1.0303487052686842,
      "grad_norm": 0.3217497766017914,
      "learning_rate": 8.420890021503872e-05,
      "loss": 1.6667,
      "step": 27560
    },
    {
      "epoch": 1.0308906464341965,
      "grad_norm": 0.31174400448799133,
      "learning_rate": 8.419685006720221e-05,
      "loss": 1.6779,
      "step": 27570
    },
    {
      "epoch": 1.0313783934831575,
      "eval_loss": 2.553981304168701,
      "eval_runtime": 22.0351,
      "eval_samples_per_second": 226.91,
      "eval_steps_per_second": 1.225,
      "step": 27579
    },
    {
      "epoch": 1.0314325875997088,
      "grad_norm": 0.25709420442581177,
      "learning_rate": 8.418479630258666e-05,
      "loss": 1.681,
      "step": 27580
    },
    {
      "epoch": 1.0319745287652209,
      "grad_norm": 0.2779182195663452,
      "learning_rate": 8.417273892268527e-05,
      "loss": 1.6807,
      "step": 27590
    },
    {
      "epoch": 1.0325164699307332,
      "grad_norm": 0.41596314311027527,
      "learning_rate": 8.416067792899162e-05,
      "loss": 1.6761,
      "step": 27600
    },
    {
      "epoch": 1.0330584110962453,
      "grad_norm": 0.2902078628540039,
      "learning_rate": 8.41486133229998e-05,
      "loss": 1.6773,
      "step": 27610
    },
    {
      "epoch": 1.0336003522617576,
      "grad_norm": 0.2825920879840851,
      "learning_rate": 8.413654510620432e-05,
      "loss": 1.6655,
      "step": 27620
    },
    {
      "epoch": 1.0341422934272697,
      "grad_norm": 0.3873823285102844,
      "learning_rate": 8.412447328010012e-05,
      "loss": 1.6638,
      "step": 27630
    },
    {
      "epoch": 1.034684234592782,
      "grad_norm": 1.103782057762146,
      "learning_rate": 8.411239784618263e-05,
      "loss": 1.6825,
      "step": 27640
    },
    {
      "epoch": 1.0352261757582943,
      "grad_norm": 0.632357120513916,
      "learning_rate": 8.410031880594768e-05,
      "loss": 1.6624,
      "step": 27650
    },
    {
      "epoch": 1.0357681169238064,
      "grad_norm": 0.3010210394859314,
      "learning_rate": 8.408823616089157e-05,
      "loss": 1.6669,
      "step": 27660
    },
    {
      "epoch": 1.0360932816231139,
      "eval_loss": 2.547511100769043,
      "eval_runtime": 21.993,
      "eval_samples_per_second": 227.345,
      "eval_steps_per_second": 1.228,
      "step": 27666
    },
    {
      "epoch": 1.0363100580893188,
      "grad_norm": 0.5328647494316101,
      "learning_rate": 8.407614991251106e-05,
      "loss": 1.6691,
      "step": 27670
    },
    {
      "epoch": 1.0368519992548308,
      "grad_norm": 0.328873872756958,
      "learning_rate": 8.406406006230333e-05,
      "loss": 1.6752,
      "step": 27680
    },
    {
      "epoch": 1.0373939404203432,
      "grad_norm": 0.5136923789978027,
      "learning_rate": 8.405196661176604e-05,
      "loss": 1.6724,
      "step": 27690
    },
    {
      "epoch": 1.0379358815858553,
      "grad_norm": 0.2645358443260193,
      "learning_rate": 8.403986956239724e-05,
      "loss": 1.67,
      "step": 27700
    },
    {
      "epoch": 1.0384778227513676,
      "grad_norm": 0.31197217106819153,
      "learning_rate": 8.402776891569547e-05,
      "loss": 1.6693,
      "step": 27710
    },
    {
      "epoch": 1.0390197639168797,
      "grad_norm": 0.3356460928916931,
      "learning_rate": 8.401566467315973e-05,
      "loss": 1.668,
      "step": 27720
    },
    {
      "epoch": 1.039561705082392,
      "grad_norm": 0.3273829221725464,
      "learning_rate": 8.400355683628941e-05,
      "loss": 1.6722,
      "step": 27730
    },
    {
      "epoch": 1.0401036462479043,
      "grad_norm": 0.26531314849853516,
      "learning_rate": 8.399144540658441e-05,
      "loss": 1.6708,
      "step": 27740
    },
    {
      "epoch": 1.0406455874134164,
      "grad_norm": 0.2471926361322403,
      "learning_rate": 8.397933038554502e-05,
      "loss": 1.6663,
      "step": 27750
    },
    {
      "epoch": 1.04080816976307,
      "eval_loss": 2.5492753982543945,
      "eval_runtime": 21.9881,
      "eval_samples_per_second": 227.396,
      "eval_steps_per_second": 1.228,
      "step": 27753
    },
    {
      "epoch": 1.0411875285789287,
      "grad_norm": 0.35514238476753235,
      "learning_rate": 8.396721177467197e-05,
      "loss": 1.6734,
      "step": 27760
    },
    {
      "epoch": 1.0417294697444408,
      "grad_norm": 0.33591434359550476,
      "learning_rate": 8.39550895754665e-05,
      "loss": 1.6738,
      "step": 27770
    },
    {
      "epoch": 1.0422714109099531,
      "grad_norm": 0.2625053822994232,
      "learning_rate": 8.394296378943027e-05,
      "loss": 1.6687,
      "step": 27780
    },
    {
      "epoch": 1.0428133520754652,
      "grad_norm": 0.2579818665981293,
      "learning_rate": 8.39308344180653e-05,
      "loss": 1.6667,
      "step": 27790
    },
    {
      "epoch": 1.0433552932409775,
      "grad_norm": 0.25735151767730713,
      "learning_rate": 8.39187014628742e-05,
      "loss": 1.6705,
      "step": 27800
    },
    {
      "epoch": 1.0438972344064898,
      "grad_norm": 0.259867399930954,
      "learning_rate": 8.390656492535992e-05,
      "loss": 1.6795,
      "step": 27810
    },
    {
      "epoch": 1.044439175572002,
      "grad_norm": 0.26979291439056396,
      "learning_rate": 8.389442480702585e-05,
      "loss": 1.6748,
      "step": 27820
    },
    {
      "epoch": 1.0449811167375143,
      "grad_norm": 0.3980228006839752,
      "learning_rate": 8.38822811093759e-05,
      "loss": 1.6651,
      "step": 27830
    },
    {
      "epoch": 1.0455230579030264,
      "grad_norm": 0.39179670810699463,
      "learning_rate": 8.387013383391435e-05,
      "loss": 1.6701,
      "step": 27840
    },
    {
      "epoch": 1.0455230579030264,
      "eval_loss": 2.563586950302124,
      "eval_runtime": 21.9807,
      "eval_samples_per_second": 227.472,
      "eval_steps_per_second": 1.228,
      "step": 27840
    },
    {
      "epoch": 1.0460649990685387,
      "grad_norm": 0.5658113956451416,
      "learning_rate": 8.385798298214596e-05,
      "loss": 1.672,
      "step": 27850
    },
    {
      "epoch": 1.0466069402340508,
      "grad_norm": 0.5975947380065918,
      "learning_rate": 8.384582855557594e-05,
      "loss": 1.6522,
      "step": 27860
    },
    {
      "epoch": 1.047148881399563,
      "grad_norm": 0.28900066018104553,
      "learning_rate": 8.383367055570991e-05,
      "loss": 1.6596,
      "step": 27870
    },
    {
      "epoch": 1.0476908225650752,
      "grad_norm": 0.38573452830314636,
      "learning_rate": 8.382150898405396e-05,
      "loss": 1.6625,
      "step": 27880
    },
    {
      "epoch": 1.0482327637305875,
      "grad_norm": 0.2802741229534149,
      "learning_rate": 8.380934384211458e-05,
      "loss": 1.6651,
      "step": 27890
    },
    {
      "epoch": 1.0487747048960998,
      "grad_norm": 0.31623005867004395,
      "learning_rate": 8.37971751313988e-05,
      "loss": 1.6644,
      "step": 27900
    },
    {
      "epoch": 1.049316646061612,
      "grad_norm": 0.2634221911430359,
      "learning_rate": 8.378500285341397e-05,
      "loss": 1.6651,
      "step": 27910
    },
    {
      "epoch": 1.0498585872271242,
      "grad_norm": 0.2507796287536621,
      "learning_rate": 8.377282700966795e-05,
      "loss": 1.6699,
      "step": 27920
    },
    {
      "epoch": 1.0502379460429827,
      "eval_loss": 2.5567116737365723,
      "eval_runtime": 21.9902,
      "eval_samples_per_second": 227.374,
      "eval_steps_per_second": 1.228,
      "step": 27927
    },
    {
      "epoch": 1.0504005283926363,
      "grad_norm": 0.38170045614242554,
      "learning_rate": 8.376064760166907e-05,
      "loss": 1.674,
      "step": 27930
    },
    {
      "epoch": 1.0509424695581486,
      "grad_norm": 0.31428414583206177,
      "learning_rate": 8.3748464630926e-05,
      "loss": 1.6715,
      "step": 27940
    },
    {
      "epoch": 1.0514844107236607,
      "grad_norm": 0.30629944801330566,
      "learning_rate": 8.373627809894796e-05,
      "loss": 1.6685,
      "step": 27950
    },
    {
      "epoch": 1.052026351889173,
      "grad_norm": 0.4108351171016693,
      "learning_rate": 8.372408800724454e-05,
      "loss": 1.6599,
      "step": 27960
    },
    {
      "epoch": 1.0525682930546854,
      "grad_norm": 0.4083997309207916,
      "learning_rate": 8.371189435732581e-05,
      "loss": 1.6629,
      "step": 27970
    },
    {
      "epoch": 1.0531102342201974,
      "grad_norm": 0.5288533568382263,
      "learning_rate": 8.369969715070226e-05,
      "loss": 1.6598,
      "step": 27980
    },
    {
      "epoch": 1.0536521753857098,
      "grad_norm": 0.35696470737457275,
      "learning_rate": 8.368749638888484e-05,
      "loss": 1.6646,
      "step": 27990
    },
    {
      "epoch": 1.0541941165512219,
      "grad_norm": 0.3221840262413025,
      "learning_rate": 8.36752920733849e-05,
      "loss": 1.6577,
      "step": 28000
    },
    {
      "epoch": 1.0547360577167342,
      "grad_norm": 0.35849282145500183,
      "learning_rate": 8.366308420571429e-05,
      "loss": 1.6604,
      "step": 28010
    },
    {
      "epoch": 1.054952834182939,
      "eval_loss": 2.547173500061035,
      "eval_runtime": 21.9831,
      "eval_samples_per_second": 227.447,
      "eval_steps_per_second": 1.228,
      "step": 28014
    },
    {
      "epoch": 1.0552779988822463,
      "grad_norm": 0.2584594488143921,
      "learning_rate": 8.365087278738524e-05,
      "loss": 1.6827,
      "step": 28020
    },
    {
      "epoch": 1.0558199400477586,
      "grad_norm": 0.2509235441684723,
      "learning_rate": 8.363865781991046e-05,
      "loss": 1.6627,
      "step": 28030
    },
    {
      "epoch": 1.0563618812132707,
      "grad_norm": 0.36739498376846313,
      "learning_rate": 8.36264393048031e-05,
      "loss": 1.666,
      "step": 28040
    },
    {
      "epoch": 1.056903822378783,
      "grad_norm": 0.41700538992881775,
      "learning_rate": 8.36142172435767e-05,
      "loss": 1.6799,
      "step": 28050
    },
    {
      "epoch": 1.0574457635442953,
      "grad_norm": 0.31023573875427246,
      "learning_rate": 8.360199163774531e-05,
      "loss": 1.6701,
      "step": 28060
    },
    {
      "epoch": 1.0579877047098074,
      "grad_norm": 0.23943783342838287,
      "learning_rate": 8.358976248882337e-05,
      "loss": 1.656,
      "step": 28070
    },
    {
      "epoch": 1.0585296458753197,
      "grad_norm": 0.8542354702949524,
      "learning_rate": 8.357752979832578e-05,
      "loss": 1.6668,
      "step": 28080
    },
    {
      "epoch": 1.0590715870408318,
      "grad_norm": 0.4590936601161957,
      "learning_rate": 8.356529356776786e-05,
      "loss": 1.6735,
      "step": 28090
    },
    {
      "epoch": 1.0596135282063441,
      "grad_norm": 0.61350017786026,
      "learning_rate": 8.355305379866541e-05,
      "loss": 1.667,
      "step": 28100
    },
    {
      "epoch": 1.0596677223228954,
      "eval_loss": 2.5593972206115723,
      "eval_runtime": 21.9806,
      "eval_samples_per_second": 227.474,
      "eval_steps_per_second": 1.228,
      "step": 28101
    },
    {
      "epoch": 1.0601554693718562,
      "grad_norm": 0.7067118883132935,
      "learning_rate": 8.354081049253462e-05,
      "loss": 1.6765,
      "step": 28110
    },
    {
      "epoch": 1.0606974105373685,
      "grad_norm": 0.27008959650993347,
      "learning_rate": 8.35285636508921e-05,
      "loss": 1.6628,
      "step": 28120
    },
    {
      "epoch": 1.0612393517028809,
      "grad_norm": 0.2994672358036041,
      "learning_rate": 8.351631327525501e-05,
      "loss": 1.6551,
      "step": 28130
    },
    {
      "epoch": 1.061781292868393,
      "grad_norm": 0.3203209340572357,
      "learning_rate": 8.350405936714083e-05,
      "loss": 1.6561,
      "step": 28140
    },
    {
      "epoch": 1.0623232340339053,
      "grad_norm": 0.3424260914325714,
      "learning_rate": 8.349180192806753e-05,
      "loss": 1.67,
      "step": 28150
    },
    {
      "epoch": 1.0628651751994174,
      "grad_norm": 0.2762835621833801,
      "learning_rate": 8.34795409595535e-05,
      "loss": 1.6691,
      "step": 28160
    },
    {
      "epoch": 1.0634071163649297,
      "grad_norm": 0.3313772976398468,
      "learning_rate": 8.346727646311758e-05,
      "loss": 1.6552,
      "step": 28170
    },
    {
      "epoch": 1.0639490575304418,
      "grad_norm": 0.2399122714996338,
      "learning_rate": 8.345500844027905e-05,
      "loss": 1.6666,
      "step": 28180
    },
    {
      "epoch": 1.0643826104628515,
      "eval_loss": 2.549626350402832,
      "eval_runtime": 21.9895,
      "eval_samples_per_second": 227.381,
      "eval_steps_per_second": 1.228,
      "step": 28188
    },
    {
      "epoch": 1.064490998695954,
      "grad_norm": 0.4699346721172333,
      "learning_rate": 8.344273689255761e-05,
      "loss": 1.668,
      "step": 28190
    },
    {
      "epoch": 1.0650329398614662,
      "grad_norm": 0.25492924451828003,
      "learning_rate": 8.343046182147342e-05,
      "loss": 1.6632,
      "step": 28200
    },
    {
      "epoch": 1.0655748810269785,
      "grad_norm": 0.4417177438735962,
      "learning_rate": 8.341818322854705e-05,
      "loss": 1.6678,
      "step": 28210
    },
    {
      "epoch": 1.0661168221924908,
      "grad_norm": 0.28891265392303467,
      "learning_rate": 8.340590111529952e-05,
      "loss": 1.6606,
      "step": 28220
    },
    {
      "epoch": 1.066658763358003,
      "grad_norm": 0.29153016209602356,
      "learning_rate": 8.339361548325226e-05,
      "loss": 1.6658,
      "step": 28230
    },
    {
      "epoch": 1.0672007045235152,
      "grad_norm": 0.2998339533805847,
      "learning_rate": 8.338132633392723e-05,
      "loss": 1.6612,
      "step": 28240
    },
    {
      "epoch": 1.0677426456890273,
      "grad_norm": 0.7401821613311768,
      "learning_rate": 8.336903366884668e-05,
      "loss": 1.6711,
      "step": 28250
    },
    {
      "epoch": 1.0682845868545396,
      "grad_norm": 0.3623445928096771,
      "learning_rate": 8.335673748953342e-05,
      "loss": 1.6542,
      "step": 28260
    },
    {
      "epoch": 1.0688265280200517,
      "grad_norm": 0.48223739862442017,
      "learning_rate": 8.334443779751063e-05,
      "loss": 1.6577,
      "step": 28270
    },
    {
      "epoch": 1.069097498602808,
      "eval_loss": 2.5425007343292236,
      "eval_runtime": 21.9868,
      "eval_samples_per_second": 227.409,
      "eval_steps_per_second": 1.228,
      "step": 28275
    },
    {
      "epoch": 1.069368469185564,
      "grad_norm": 0.4789433479309082,
      "learning_rate": 8.333213459430195e-05,
      "loss": 1.6562,
      "step": 28280
    },
    {
      "epoch": 1.0699104103510764,
      "grad_norm": 0.5614416003227234,
      "learning_rate": 8.331982788143142e-05,
      "loss": 1.6689,
      "step": 28290
    },
    {
      "epoch": 1.0704523515165885,
      "grad_norm": 0.4019344449043274,
      "learning_rate": 8.330751766042358e-05,
      "loss": 1.6564,
      "step": 28300
    },
    {
      "epoch": 1.0709942926821008,
      "grad_norm": 0.3532206118106842,
      "learning_rate": 8.329520393280335e-05,
      "loss": 1.6645,
      "step": 28310
    },
    {
      "epoch": 1.0715362338476129,
      "grad_norm": 0.3050895631313324,
      "learning_rate": 8.32828867000961e-05,
      "loss": 1.6581,
      "step": 28320
    },
    {
      "epoch": 1.0720781750131252,
      "grad_norm": 0.4680069386959076,
      "learning_rate": 8.327056596382766e-05,
      "loss": 1.6648,
      "step": 28330
    },
    {
      "epoch": 1.0726201161786373,
      "grad_norm": 0.5191404819488525,
      "learning_rate": 8.325824172552422e-05,
      "loss": 1.6579,
      "step": 28340
    },
    {
      "epoch": 1.0731620573441496,
      "grad_norm": 0.30772343277931213,
      "learning_rate": 8.324591398671249e-05,
      "loss": 1.6623,
      "step": 28350
    },
    {
      "epoch": 1.0737039985096617,
      "grad_norm": 0.351131796836853,
      "learning_rate": 8.323358274891956e-05,
      "loss": 1.6632,
      "step": 28360
    },
    {
      "epoch": 1.0738123867427642,
      "eval_loss": 2.5413732528686523,
      "eval_runtime": 21.9839,
      "eval_samples_per_second": 227.439,
      "eval_steps_per_second": 1.228,
      "step": 28362
    },
    {
      "epoch": 1.074245939675174,
      "grad_norm": 0.42983412742614746,
      "learning_rate": 8.322124801367298e-05,
      "loss": 1.652,
      "step": 28370
    },
    {
      "epoch": 1.0747878808406863,
      "grad_norm": 0.2206849902868271,
      "learning_rate": 8.320890978250073e-05,
      "loss": 1.6608,
      "step": 28380
    },
    {
      "epoch": 1.0753298220061984,
      "grad_norm": 0.2790736258029938,
      "learning_rate": 8.31965680569312e-05,
      "loss": 1.6622,
      "step": 28390
    },
    {
      "epoch": 1.0758717631717107,
      "grad_norm": 0.44509637355804443,
      "learning_rate": 8.318422283849323e-05,
      "loss": 1.6611,
      "step": 28400
    },
    {
      "epoch": 1.0764137043372228,
      "grad_norm": 0.24993574619293213,
      "learning_rate": 8.317187412871611e-05,
      "loss": 1.6573,
      "step": 28410
    },
    {
      "epoch": 1.0769556455027351,
      "grad_norm": 0.325591504573822,
      "learning_rate": 8.315952192912952e-05,
      "loss": 1.6501,
      "step": 28420
    },
    {
      "epoch": 1.0774975866682472,
      "grad_norm": 0.2710714638233185,
      "learning_rate": 8.314716624126362e-05,
      "loss": 1.6609,
      "step": 28430
    },
    {
      "epoch": 1.0780395278337596,
      "grad_norm": 0.5176997780799866,
      "learning_rate": 8.313480706664898e-05,
      "loss": 1.6707,
      "step": 28440
    },
    {
      "epoch": 1.0785272748827206,
      "eval_loss": 2.5304479598999023,
      "eval_runtime": 21.9868,
      "eval_samples_per_second": 227.409,
      "eval_steps_per_second": 1.228,
      "step": 28449
    },
    {
      "epoch": 1.0785814689992717,
      "grad_norm": 0.5030739307403564,
      "learning_rate": 8.312244440681657e-05,
      "loss": 1.6555,
      "step": 28450
    },
    {
      "epoch": 1.079123410164784,
      "grad_norm": 0.44651323556900024,
      "learning_rate": 8.311007826329786e-05,
      "loss": 1.6696,
      "step": 28460
    },
    {
      "epoch": 1.0796653513302963,
      "grad_norm": 0.6648367047309875,
      "learning_rate": 8.309770863762468e-05,
      "loss": 1.6589,
      "step": 28470
    },
    {
      "epoch": 1.0802072924958084,
      "grad_norm": 0.24871893227100372,
      "learning_rate": 8.308533553132935e-05,
      "loss": 1.6527,
      "step": 28480
    },
    {
      "epoch": 1.0807492336613207,
      "grad_norm": 0.4465760588645935,
      "learning_rate": 8.307295894594456e-05,
      "loss": 1.6585,
      "step": 28490
    },
    {
      "epoch": 1.0812911748268328,
      "grad_norm": 0.313731849193573,
      "learning_rate": 8.306057888300352e-05,
      "loss": 1.6579,
      "step": 28500
    },
    {
      "epoch": 1.081833115992345,
      "grad_norm": 0.2809997797012329,
      "learning_rate": 8.304819534403981e-05,
      "loss": 1.6533,
      "step": 28510
    },
    {
      "epoch": 1.0823750571578572,
      "grad_norm": 0.24525237083435059,
      "learning_rate": 8.303580833058742e-05,
      "loss": 1.6607,
      "step": 28520
    },
    {
      "epoch": 1.0829169983233695,
      "grad_norm": 0.2652153968811035,
      "learning_rate": 8.302341784418081e-05,
      "loss": 1.6687,
      "step": 28530
    },
    {
      "epoch": 1.083242163022677,
      "eval_loss": 2.5320417881011963,
      "eval_runtime": 21.9915,
      "eval_samples_per_second": 227.361,
      "eval_steps_per_second": 1.228,
      "step": 28536
    },
    {
      "epoch": 1.0834589394888818,
      "grad_norm": 0.24303193390369415,
      "learning_rate": 8.301102388635487e-05,
      "loss": 1.6608,
      "step": 28540
    },
    {
      "epoch": 1.084000880654394,
      "grad_norm": 0.26834824681282043,
      "learning_rate": 8.299862645864491e-05,
      "loss": 1.6616,
      "step": 28550
    },
    {
      "epoch": 1.0845428218199062,
      "grad_norm": 0.2935827374458313,
      "learning_rate": 8.298622556258664e-05,
      "loss": 1.6588,
      "step": 28560
    },
    {
      "epoch": 1.0850847629854183,
      "grad_norm": 0.27826014161109924,
      "learning_rate": 8.297382119971627e-05,
      "loss": 1.6731,
      "step": 28570
    },
    {
      "epoch": 1.0856267041509307,
      "grad_norm": 0.35834673047065735,
      "learning_rate": 8.296141337157038e-05,
      "loss": 1.657,
      "step": 28580
    },
    {
      "epoch": 1.0861686453164427,
      "grad_norm": 0.2979893386363983,
      "learning_rate": 8.294900207968597e-05,
      "loss": 1.6546,
      "step": 28590
    },
    {
      "epoch": 1.086710586481955,
      "grad_norm": 0.2854683995246887,
      "learning_rate": 8.293658732560056e-05,
      "loss": 1.6656,
      "step": 28600
    },
    {
      "epoch": 1.0872525276474674,
      "grad_norm": 0.37426939606666565,
      "learning_rate": 8.292416911085198e-05,
      "loss": 1.6513,
      "step": 28610
    },
    {
      "epoch": 1.0877944688129795,
      "grad_norm": 0.34777402877807617,
      "learning_rate": 8.291174743697857e-05,
      "loss": 1.6653,
      "step": 28620
    },
    {
      "epoch": 1.087957051162633,
      "eval_loss": 2.53008770942688,
      "eval_runtime": 21.9855,
      "eval_samples_per_second": 227.422,
      "eval_steps_per_second": 1.228,
      "step": 28623
    },
    {
      "epoch": 1.0883364099784918,
      "grad_norm": 0.28725841641426086,
      "learning_rate": 8.289932230551907e-05,
      "loss": 1.651,
      "step": 28630
    },
    {
      "epoch": 1.0888783511440039,
      "grad_norm": 0.28935521841049194,
      "learning_rate": 8.288689371801265e-05,
      "loss": 1.6616,
      "step": 28640
    },
    {
      "epoch": 1.0894202923095162,
      "grad_norm": 0.30465996265411377,
      "learning_rate": 8.287446167599891e-05,
      "loss": 1.6554,
      "step": 28650
    },
    {
      "epoch": 1.0899622334750283,
      "grad_norm": 0.5029714703559875,
      "learning_rate": 8.286202618101788e-05,
      "loss": 1.668,
      "step": 28660
    },
    {
      "epoch": 1.0905041746405406,
      "grad_norm": 0.2679192125797272,
      "learning_rate": 8.284958723461001e-05,
      "loss": 1.661,
      "step": 28670
    },
    {
      "epoch": 1.0910461158060527,
      "grad_norm": 0.3121162950992584,
      "learning_rate": 8.283714483831617e-05,
      "loss": 1.6601,
      "step": 28680
    },
    {
      "epoch": 1.091588056971565,
      "grad_norm": 0.2876189947128296,
      "learning_rate": 8.28246989936777e-05,
      "loss": 1.65,
      "step": 28690
    },
    {
      "epoch": 1.0921299981370773,
      "grad_norm": 0.30062171816825867,
      "learning_rate": 8.281224970223632e-05,
      "loss": 1.6531,
      "step": 28700
    },
    {
      "epoch": 1.0926719393025894,
      "grad_norm": 0.44612184166908264,
      "learning_rate": 8.27997969655342e-05,
      "loss": 1.6509,
      "step": 28710
    },
    {
      "epoch": 1.0926719393025894,
      "eval_loss": 2.5466320514678955,
      "eval_runtime": 21.9837,
      "eval_samples_per_second": 227.441,
      "eval_steps_per_second": 1.228,
      "step": 28710
    },
    {
      "epoch": 1.0932138804681018,
      "grad_norm": 0.2449900060892105,
      "learning_rate": 8.27873407851139e-05,
      "loss": 1.6529,
      "step": 28720
    },
    {
      "epoch": 1.0937558216336138,
      "grad_norm": 0.4357602894306183,
      "learning_rate": 8.27748811625185e-05,
      "loss": 1.6595,
      "step": 28730
    },
    {
      "epoch": 1.0942977627991262,
      "grad_norm": 0.34103602170944214,
      "learning_rate": 8.276241809929137e-05,
      "loss": 1.6567,
      "step": 28740
    },
    {
      "epoch": 1.0948397039646383,
      "grad_norm": 0.3430241346359253,
      "learning_rate": 8.274995159697646e-05,
      "loss": 1.6582,
      "step": 28750
    },
    {
      "epoch": 1.0953816451301506,
      "grad_norm": 0.40874654054641724,
      "learning_rate": 8.273748165711799e-05,
      "loss": 1.6566,
      "step": 28760
    },
    {
      "epoch": 1.0959235862956627,
      "grad_norm": 0.26068803668022156,
      "learning_rate": 8.272500828126075e-05,
      "loss": 1.6484,
      "step": 28770
    },
    {
      "epoch": 1.096465527461175,
      "grad_norm": 0.3570903241634369,
      "learning_rate": 8.271253147094981e-05,
      "loss": 1.6562,
      "step": 28780
    },
    {
      "epoch": 1.0970074686266873,
      "grad_norm": 0.2552138864994049,
      "learning_rate": 8.270005122773083e-05,
      "loss": 1.657,
      "step": 28790
    },
    {
      "epoch": 1.0973868274425458,
      "eval_loss": 2.531879186630249,
      "eval_runtime": 21.9807,
      "eval_samples_per_second": 227.473,
      "eval_steps_per_second": 1.228,
      "step": 28797
    },
    {
      "epoch": 1.0975494097921994,
      "grad_norm": 0.374850869178772,
      "learning_rate": 8.268756755314973e-05,
      "loss": 1.6501,
      "step": 28800
    },
    {
      "epoch": 1.0980913509577117,
      "grad_norm": 0.2875063419342041,
      "learning_rate": 8.2675080448753e-05,
      "loss": 1.6515,
      "step": 28810
    },
    {
      "epoch": 1.0986332921232238,
      "grad_norm": 0.2932276725769043,
      "learning_rate": 8.266258991608743e-05,
      "loss": 1.6496,
      "step": 28820
    },
    {
      "epoch": 1.0991752332887361,
      "grad_norm": 0.28365346789360046,
      "learning_rate": 8.265009595670034e-05,
      "loss": 1.6466,
      "step": 28830
    },
    {
      "epoch": 1.0997171744542482,
      "grad_norm": 0.32967546582221985,
      "learning_rate": 8.263759857213939e-05,
      "loss": 1.6494,
      "step": 28840
    },
    {
      "epoch": 1.1002591156197605,
      "grad_norm": 0.5435857772827148,
      "learning_rate": 8.262509776395274e-05,
      "loss": 1.661,
      "step": 28850
    },
    {
      "epoch": 1.1008010567852728,
      "grad_norm": 0.26168474555015564,
      "learning_rate": 8.26125935336889e-05,
      "loss": 1.6467,
      "step": 28860
    },
    {
      "epoch": 1.101342997950785,
      "grad_norm": 0.5927068591117859,
      "learning_rate": 8.260008588289687e-05,
      "loss": 1.6555,
      "step": 28870
    },
    {
      "epoch": 1.1018849391162973,
      "grad_norm": 0.28580188751220703,
      "learning_rate": 8.258757481312601e-05,
      "loss": 1.6536,
      "step": 28880
    },
    {
      "epoch": 1.1021017155825021,
      "eval_loss": 2.5326905250549316,
      "eval_runtime": 21.9872,
      "eval_samples_per_second": 227.405,
      "eval_steps_per_second": 1.228,
      "step": 28884
    },
    {
      "epoch": 1.1024268802818094,
      "grad_norm": 0.23324058949947357,
      "learning_rate": 8.257506032592617e-05,
      "loss": 1.6618,
      "step": 28890
    },
    {
      "epoch": 1.1029688214473217,
      "grad_norm": 0.37625429034233093,
      "learning_rate": 8.256254242284755e-05,
      "loss": 1.6533,
      "step": 28900
    },
    {
      "epoch": 1.1035107626128338,
      "grad_norm": 0.2690766453742981,
      "learning_rate": 8.255002110544087e-05,
      "loss": 1.6568,
      "step": 28910
    },
    {
      "epoch": 1.104052703778346,
      "grad_norm": 0.3211461007595062,
      "learning_rate": 8.253749637525717e-05,
      "loss": 1.6532,
      "step": 28920
    },
    {
      "epoch": 1.1045946449438584,
      "grad_norm": 0.30158957839012146,
      "learning_rate": 8.252496823384798e-05,
      "loss": 1.6473,
      "step": 28930
    },
    {
      "epoch": 1.1051365861093705,
      "grad_norm": 0.26189902424812317,
      "learning_rate": 8.251243668276524e-05,
      "loss": 1.6586,
      "step": 28940
    },
    {
      "epoch": 1.1056785272748828,
      "grad_norm": 0.3414219915866852,
      "learning_rate": 8.249990172356128e-05,
      "loss": 1.6553,
      "step": 28950
    },
    {
      "epoch": 1.106220468440395,
      "grad_norm": 0.2272077202796936,
      "learning_rate": 8.248736335778888e-05,
      "loss": 1.662,
      "step": 28960
    },
    {
      "epoch": 1.1067624096059072,
      "grad_norm": 0.3499011695384979,
      "learning_rate": 8.247482158700126e-05,
      "loss": 1.6621,
      "step": 28970
    },
    {
      "epoch": 1.1068166037224585,
      "eval_loss": 2.527860403060913,
      "eval_runtime": 21.9887,
      "eval_samples_per_second": 227.389,
      "eval_steps_per_second": 1.228,
      "step": 28971
    },
    {
      "epoch": 1.1073043507714193,
      "grad_norm": 0.44124558568000793,
      "learning_rate": 8.246227641275199e-05,
      "loss": 1.6541,
      "step": 28980
    },
    {
      "epoch": 1.1078462919369316,
      "grad_norm": 0.807167649269104,
      "learning_rate": 8.244972783659518e-05,
      "loss": 1.6499,
      "step": 28990
    },
    {
      "epoch": 1.1083882331024437,
      "grad_norm": 0.5441895723342896,
      "learning_rate": 8.243717586008525e-05,
      "loss": 1.6535,
      "step": 29000
    },
    {
      "epoch": 1.108930174267956,
      "grad_norm": 0.4006600081920624,
      "learning_rate": 8.24246204847771e-05,
      "loss": 1.6554,
      "step": 29010
    },
    {
      "epoch": 1.1094721154334684,
      "grad_norm": 0.26644378900527954,
      "learning_rate": 8.241206171222602e-05,
      "loss": 1.6445,
      "step": 29020
    },
    {
      "epoch": 1.1100140565989804,
      "grad_norm": 0.3243505358695984,
      "learning_rate": 8.239949954398777e-05,
      "loss": 1.6578,
      "step": 29030
    },
    {
      "epoch": 1.1105559977644928,
      "grad_norm": 0.39492732286453247,
      "learning_rate": 8.238693398161844e-05,
      "loss": 1.6574,
      "step": 29040
    },
    {
      "epoch": 1.1110979389300049,
      "grad_norm": 0.36204707622528076,
      "learning_rate": 8.237436502667467e-05,
      "loss": 1.647,
      "step": 29050
    },
    {
      "epoch": 1.1115314918624146,
      "eval_loss": 2.537062644958496,
      "eval_runtime": 21.9899,
      "eval_samples_per_second": 227.377,
      "eval_steps_per_second": 1.228,
      "step": 29058
    },
    {
      "epoch": 1.1116398800955172,
      "grad_norm": 0.4409829080104828,
      "learning_rate": 8.236179268071337e-05,
      "loss": 1.6599,
      "step": 29060
    },
    {
      "epoch": 1.1121818212610293,
      "grad_norm": 0.4123667776584625,
      "learning_rate": 8.2349216945292e-05,
      "loss": 1.6573,
      "step": 29070
    },
    {
      "epoch": 1.1127237624265416,
      "grad_norm": 0.29760655760765076,
      "learning_rate": 8.233663782196837e-05,
      "loss": 1.6542,
      "step": 29080
    },
    {
      "epoch": 1.1132657035920537,
      "grad_norm": 0.34215226769447327,
      "learning_rate": 8.232405531230074e-05,
      "loss": 1.6521,
      "step": 29090
    },
    {
      "epoch": 1.113807644757566,
      "grad_norm": 0.6061283349990845,
      "learning_rate": 8.231146941784776e-05,
      "loss": 1.6578,
      "step": 29100
    },
    {
      "epoch": 1.1143495859230783,
      "grad_norm": 0.5324410200119019,
      "learning_rate": 8.229888014016853e-05,
      "loss": 1.6522,
      "step": 29110
    },
    {
      "epoch": 1.1148915270885904,
      "grad_norm": 0.3367980420589447,
      "learning_rate": 8.228628748082256e-05,
      "loss": 1.6521,
      "step": 29120
    },
    {
      "epoch": 1.1154334682541027,
      "grad_norm": 0.34855327010154724,
      "learning_rate": 8.227369144136974e-05,
      "loss": 1.6478,
      "step": 29130
    },
    {
      "epoch": 1.1159754094196148,
      "grad_norm": 0.5364739298820496,
      "learning_rate": 8.226109202337043e-05,
      "loss": 1.6525,
      "step": 29140
    },
    {
      "epoch": 1.116246380002371,
      "eval_loss": 2.545180320739746,
      "eval_runtime": 21.9866,
      "eval_samples_per_second": 227.411,
      "eval_steps_per_second": 1.228,
      "step": 29145
    },
    {
      "epoch": 1.1165173505851271,
      "grad_norm": 0.36383602023124695,
      "learning_rate": 8.224848922838541e-05,
      "loss": 1.66,
      "step": 29150
    },
    {
      "epoch": 1.1170592917506392,
      "grad_norm": 0.5313844680786133,
      "learning_rate": 8.223588305797585e-05,
      "loss": 1.6447,
      "step": 29160
    },
    {
      "epoch": 1.1176012329161515,
      "grad_norm": 0.2722112536430359,
      "learning_rate": 8.222327351370332e-05,
      "loss": 1.6562,
      "step": 29170
    },
    {
      "epoch": 1.1181431740816639,
      "grad_norm": 0.39539259672164917,
      "learning_rate": 8.221066059712988e-05,
      "loss": 1.6494,
      "step": 29180
    },
    {
      "epoch": 1.118685115247176,
      "grad_norm": 0.4637277126312256,
      "learning_rate": 8.219804430981794e-05,
      "loss": 1.6445,
      "step": 29190
    },
    {
      "epoch": 1.1192270564126883,
      "grad_norm": 0.3050798177719116,
      "learning_rate": 8.218542465333035e-05,
      "loss": 1.6477,
      "step": 29200
    },
    {
      "epoch": 1.1197689975782004,
      "grad_norm": 0.3634278476238251,
      "learning_rate": 8.217280162923036e-05,
      "loss": 1.662,
      "step": 29210
    },
    {
      "epoch": 1.1203109387437127,
      "grad_norm": 0.2769427001476288,
      "learning_rate": 8.216017523908172e-05,
      "loss": 1.655,
      "step": 29220
    },
    {
      "epoch": 1.1208528799092248,
      "grad_norm": 0.2443336397409439,
      "learning_rate": 8.214754548444846e-05,
      "loss": 1.6374,
      "step": 29230
    },
    {
      "epoch": 1.1209612681423273,
      "eval_loss": 2.541795015335083,
      "eval_runtime": 21.9864,
      "eval_samples_per_second": 227.413,
      "eval_steps_per_second": 1.228,
      "step": 29232
    },
    {
      "epoch": 1.121394821074737,
      "grad_norm": 0.2958986163139343,
      "learning_rate": 8.213491236689514e-05,
      "loss": 1.6575,
      "step": 29240
    },
    {
      "epoch": 1.1219367622402494,
      "grad_norm": 0.2536433935165405,
      "learning_rate": 8.212227588798667e-05,
      "loss": 1.6449,
      "step": 29250
    },
    {
      "epoch": 1.1224787034057615,
      "grad_norm": 0.2808546721935272,
      "learning_rate": 8.210963604928842e-05,
      "loss": 1.664,
      "step": 29260
    },
    {
      "epoch": 1.1230206445712738,
      "grad_norm": 0.24417255818843842,
      "learning_rate": 8.209699285236618e-05,
      "loss": 1.6619,
      "step": 29270
    },
    {
      "epoch": 1.123562585736786,
      "grad_norm": 0.22852517664432526,
      "learning_rate": 8.208434629878607e-05,
      "loss": 1.6587,
      "step": 29280
    },
    {
      "epoch": 1.1241045269022982,
      "grad_norm": 0.3301936388015747,
      "learning_rate": 8.207169639011474e-05,
      "loss": 1.6574,
      "step": 29290
    },
    {
      "epoch": 1.1246464680678103,
      "grad_norm": 0.3570113480091095,
      "learning_rate": 8.205904312791921e-05,
      "loss": 1.648,
      "step": 29300
    },
    {
      "epoch": 1.1251884092333226,
      "grad_norm": 0.4051576256752014,
      "learning_rate": 8.20463865137669e-05,
      "loss": 1.6463,
      "step": 29310
    },
    {
      "epoch": 1.1256761562822837,
      "eval_loss": 2.5262796878814697,
      "eval_runtime": 21.9856,
      "eval_samples_per_second": 227.421,
      "eval_steps_per_second": 1.228,
      "step": 29319
    },
    {
      "epoch": 1.1257303503988347,
      "grad_norm": 0.3884764611721039,
      "learning_rate": 8.203372654922563e-05,
      "loss": 1.6512,
      "step": 29320
    },
    {
      "epoch": 1.126272291564347,
      "grad_norm": 0.7175701260566711,
      "learning_rate": 8.20210632358637e-05,
      "loss": 1.6392,
      "step": 29330
    },
    {
      "epoch": 1.1268142327298594,
      "grad_norm": 0.44228026270866394,
      "learning_rate": 8.200839657524976e-05,
      "loss": 1.6573,
      "step": 29340
    },
    {
      "epoch": 1.1273561738953715,
      "grad_norm": 0.25877845287323,
      "learning_rate": 8.199572656895291e-05,
      "loss": 1.6494,
      "step": 29350
    },
    {
      "epoch": 1.1278981150608838,
      "grad_norm": 0.42708685994148254,
      "learning_rate": 8.198305321854267e-05,
      "loss": 1.6489,
      "step": 29360
    },
    {
      "epoch": 1.1284400562263959,
      "grad_norm": 0.6025623083114624,
      "learning_rate": 8.197037652558895e-05,
      "loss": 1.6507,
      "step": 29370
    },
    {
      "epoch": 1.1289819973919082,
      "grad_norm": 0.4976705312728882,
      "learning_rate": 8.195769649166205e-05,
      "loss": 1.6528,
      "step": 29380
    },
    {
      "epoch": 1.1295239385574203,
      "grad_norm": 0.27502158284187317,
      "learning_rate": 8.194501311833277e-05,
      "loss": 1.659,
      "step": 29390
    },
    {
      "epoch": 1.1300658797229326,
      "grad_norm": 0.26740071177482605,
      "learning_rate": 8.193232640717223e-05,
      "loss": 1.6435,
      "step": 29400
    },
    {
      "epoch": 1.13039104442224,
      "eval_loss": 2.532335042953491,
      "eval_runtime": 21.9808,
      "eval_samples_per_second": 227.471,
      "eval_steps_per_second": 1.228,
      "step": 29406
    },
    {
      "epoch": 1.1306078208884447,
      "grad_norm": 0.21525554358959198,
      "learning_rate": 8.191963635975204e-05,
      "loss": 1.651,
      "step": 29410
    },
    {
      "epoch": 1.131149762053957,
      "grad_norm": 0.3878322243690491,
      "learning_rate": 8.190694297764417e-05,
      "loss": 1.6369,
      "step": 29420
    },
    {
      "epoch": 1.1316917032194693,
      "grad_norm": 0.2476329803466797,
      "learning_rate": 8.189424626242102e-05,
      "loss": 1.6483,
      "step": 29430
    },
    {
      "epoch": 1.1322336443849814,
      "grad_norm": 0.40698474645614624,
      "learning_rate": 8.18815462156554e-05,
      "loss": 1.6495,
      "step": 29440
    },
    {
      "epoch": 1.1327755855504937,
      "grad_norm": 0.2684831917285919,
      "learning_rate": 8.186884283892056e-05,
      "loss": 1.6424,
      "step": 29450
    },
    {
      "epoch": 1.1333175267160058,
      "grad_norm": 0.46887609362602234,
      "learning_rate": 8.185613613379011e-05,
      "loss": 1.6439,
      "step": 29460
    },
    {
      "epoch": 1.1338594678815181,
      "grad_norm": 0.34955888986587524,
      "learning_rate": 8.184342610183812e-05,
      "loss": 1.6492,
      "step": 29470
    },
    {
      "epoch": 1.1344014090470302,
      "grad_norm": 0.3082946240901947,
      "learning_rate": 8.183071274463903e-05,
      "loss": 1.6533,
      "step": 29480
    },
    {
      "epoch": 1.1349433502125426,
      "grad_norm": 0.35093745589256287,
      "learning_rate": 8.181799606376773e-05,
      "loss": 1.6564,
      "step": 29490
    },
    {
      "epoch": 1.1351059325621962,
      "eval_loss": 2.538316249847412,
      "eval_runtime": 21.9822,
      "eval_samples_per_second": 227.456,
      "eval_steps_per_second": 1.228,
      "step": 29493
    },
    {
      "epoch": 1.1354852913780547,
      "grad_norm": 0.26975366473197937,
      "learning_rate": 8.180527606079953e-05,
      "loss": 1.6604,
      "step": 29500
    },
    {
      "epoch": 1.136027232543567,
      "grad_norm": 0.2648986577987671,
      "learning_rate": 8.17925527373101e-05,
      "loss": 1.6453,
      "step": 29510
    },
    {
      "epoch": 1.1365691737090793,
      "grad_norm": 0.2807252109050751,
      "learning_rate": 8.177982609487556e-05,
      "loss": 1.634,
      "step": 29520
    },
    {
      "epoch": 1.1371111148745914,
      "grad_norm": 0.36864936351776123,
      "learning_rate": 8.176709613507243e-05,
      "loss": 1.6542,
      "step": 29530
    },
    {
      "epoch": 1.1376530560401037,
      "grad_norm": 0.3327690362930298,
      "learning_rate": 8.175436285947764e-05,
      "loss": 1.6508,
      "step": 29540
    },
    {
      "epoch": 1.1381949972056158,
      "grad_norm": 0.44101572036743164,
      "learning_rate": 8.174162626966853e-05,
      "loss": 1.6418,
      "step": 29550
    },
    {
      "epoch": 1.138736938371128,
      "grad_norm": 0.24692606925964355,
      "learning_rate": 8.172888636722288e-05,
      "loss": 1.6477,
      "step": 29560
    },
    {
      "epoch": 1.1392788795366404,
      "grad_norm": 0.41825541853904724,
      "learning_rate": 8.171614315371881e-05,
      "loss": 1.6572,
      "step": 29570
    },
    {
      "epoch": 1.1398208207021525,
      "grad_norm": 0.3095550835132599,
      "learning_rate": 8.170339663073492e-05,
      "loss": 1.6457,
      "step": 29580
    },
    {
      "epoch": 1.1398208207021525,
      "eval_loss": 2.5319855213165283,
      "eval_runtime": 21.9838,
      "eval_samples_per_second": 227.44,
      "eval_steps_per_second": 1.228,
      "step": 29580
    },
    {
      "epoch": 1.1403627618676648,
      "grad_norm": 0.5016186833381653,
      "learning_rate": 8.16906467998502e-05,
      "loss": 1.6549,
      "step": 29590
    },
    {
      "epoch": 1.140904703033177,
      "grad_norm": 0.4376733899116516,
      "learning_rate": 8.1677893662644e-05,
      "loss": 1.6502,
      "step": 29600
    },
    {
      "epoch": 1.1414466441986892,
      "grad_norm": 0.285548597574234,
      "learning_rate": 8.16651372206962e-05,
      "loss": 1.6387,
      "step": 29610
    },
    {
      "epoch": 1.1419885853642013,
      "grad_norm": 0.22487492859363556,
      "learning_rate": 8.165237747558694e-05,
      "loss": 1.6547,
      "step": 29620
    },
    {
      "epoch": 1.1425305265297137,
      "grad_norm": 0.37438520789146423,
      "learning_rate": 8.16396144288969e-05,
      "loss": 1.6615,
      "step": 29630
    },
    {
      "epoch": 1.1430724676952257,
      "grad_norm": 0.2509768009185791,
      "learning_rate": 8.162684808220708e-05,
      "loss": 1.6561,
      "step": 29640
    },
    {
      "epoch": 1.143614408860738,
      "grad_norm": 0.32760903239250183,
      "learning_rate": 8.161407843709889e-05,
      "loss": 1.6438,
      "step": 29650
    },
    {
      "epoch": 1.1441563500262504,
      "grad_norm": 0.29007697105407715,
      "learning_rate": 8.160130549515423e-05,
      "loss": 1.6537,
      "step": 29660
    },
    {
      "epoch": 1.1445357088421089,
      "eval_loss": 2.5295541286468506,
      "eval_runtime": 21.9895,
      "eval_samples_per_second": 227.382,
      "eval_steps_per_second": 1.228,
      "step": 29667
    },
    {
      "epoch": 1.1446982911917625,
      "grad_norm": 0.3338105380535126,
      "learning_rate": 8.158852925795534e-05,
      "loss": 1.6412,
      "step": 29670
    },
    {
      "epoch": 1.1452402323572748,
      "grad_norm": 0.2998930811882019,
      "learning_rate": 8.157574972708488e-05,
      "loss": 1.6408,
      "step": 29680
    },
    {
      "epoch": 1.1457821735227869,
      "grad_norm": 0.2588035464286804,
      "learning_rate": 8.156296690412593e-05,
      "loss": 1.6518,
      "step": 29690
    },
    {
      "epoch": 1.1463241146882992,
      "grad_norm": 0.24682050943374634,
      "learning_rate": 8.155018079066193e-05,
      "loss": 1.6513,
      "step": 29700
    },
    {
      "epoch": 1.1468660558538113,
      "grad_norm": 0.34920454025268555,
      "learning_rate": 8.153739138827684e-05,
      "loss": 1.6486,
      "step": 29710
    },
    {
      "epoch": 1.1474079970193236,
      "grad_norm": 0.24906253814697266,
      "learning_rate": 8.15245986985549e-05,
      "loss": 1.6509,
      "step": 29720
    },
    {
      "epoch": 1.1479499381848357,
      "grad_norm": 0.3515869677066803,
      "learning_rate": 8.151180272308085e-05,
      "loss": 1.6521,
      "step": 29730
    },
    {
      "epoch": 1.148491879350348,
      "grad_norm": 0.2527877688407898,
      "learning_rate": 8.149900346343975e-05,
      "loss": 1.6504,
      "step": 29740
    },
    {
      "epoch": 1.1490338205158603,
      "grad_norm": 0.29912281036376953,
      "learning_rate": 8.148620092121718e-05,
      "loss": 1.6499,
      "step": 29750
    },
    {
      "epoch": 1.1492505969820652,
      "eval_loss": 2.5310330390930176,
      "eval_runtime": 21.9865,
      "eval_samples_per_second": 227.412,
      "eval_steps_per_second": 1.228,
      "step": 29754
    },
    {
      "epoch": 1.1495757616813724,
      "grad_norm": 0.34561511874198914,
      "learning_rate": 8.147339509799902e-05,
      "loss": 1.6445,
      "step": 29760
    },
    {
      "epoch": 1.1501177028468847,
      "grad_norm": 0.2445221096277237,
      "learning_rate": 8.146058599537162e-05,
      "loss": 1.6496,
      "step": 29770
    },
    {
      "epoch": 1.1506596440123968,
      "grad_norm": 0.2948252856731415,
      "learning_rate": 8.144777361492168e-05,
      "loss": 1.6419,
      "step": 29780
    },
    {
      "epoch": 1.1512015851779092,
      "grad_norm": 0.297454833984375,
      "learning_rate": 8.14349579582364e-05,
      "loss": 1.6407,
      "step": 29790
    },
    {
      "epoch": 1.1517435263434213,
      "grad_norm": 0.2834935188293457,
      "learning_rate": 8.142213902690329e-05,
      "loss": 1.6598,
      "step": 29800
    },
    {
      "epoch": 1.1522854675089336,
      "grad_norm": 0.4210936427116394,
      "learning_rate": 8.140931682251029e-05,
      "loss": 1.6401,
      "step": 29810
    },
    {
      "epoch": 1.1528274086744457,
      "grad_norm": 0.33858510851860046,
      "learning_rate": 8.13964913466458e-05,
      "loss": 1.646,
      "step": 29820
    },
    {
      "epoch": 1.153369349839958,
      "grad_norm": 0.3140000104904175,
      "learning_rate": 8.138366260089855e-05,
      "loss": 1.6428,
      "step": 29830
    },
    {
      "epoch": 1.1539112910054703,
      "grad_norm": 0.24602891504764557,
      "learning_rate": 8.137083058685774e-05,
      "loss": 1.6408,
      "step": 29840
    },
    {
      "epoch": 1.1539654851220214,
      "eval_loss": 2.540917158126831,
      "eval_runtime": 21.9857,
      "eval_samples_per_second": 227.421,
      "eval_steps_per_second": 1.228,
      "step": 29841
    },
    {
      "epoch": 1.1544532321709824,
      "grad_norm": 0.23379047214984894,
      "learning_rate": 8.135799530611292e-05,
      "loss": 1.6488,
      "step": 29850
    },
    {
      "epoch": 1.1549951733364947,
      "grad_norm": 0.3018862307071686,
      "learning_rate": 8.134515676025407e-05,
      "loss": 1.6542,
      "step": 29860
    },
    {
      "epoch": 1.1555371145020068,
      "grad_norm": 0.24227862060070038,
      "learning_rate": 8.133231495087159e-05,
      "loss": 1.6456,
      "step": 29870
    },
    {
      "epoch": 1.1560790556675191,
      "grad_norm": 0.39517369866371155,
      "learning_rate": 8.131946987955627e-05,
      "loss": 1.6424,
      "step": 29880
    },
    {
      "epoch": 1.1566209968330314,
      "grad_norm": 0.3524185121059418,
      "learning_rate": 8.130662154789926e-05,
      "loss": 1.6485,
      "step": 29890
    },
    {
      "epoch": 1.1571629379985435,
      "grad_norm": 0.43569809198379517,
      "learning_rate": 8.12937699574922e-05,
      "loss": 1.6421,
      "step": 29900
    },
    {
      "epoch": 1.1577048791640558,
      "grad_norm": 0.3304058909416199,
      "learning_rate": 8.128091510992705e-05,
      "loss": 1.652,
      "step": 29910
    },
    {
      "epoch": 1.158246820329568,
      "grad_norm": 0.4421805143356323,
      "learning_rate": 8.126805700679628e-05,
      "loss": 1.6417,
      "step": 29920
    },
    {
      "epoch": 1.1586803732619777,
      "eval_loss": 2.530564785003662,
      "eval_runtime": 21.9874,
      "eval_samples_per_second": 227.403,
      "eval_steps_per_second": 1.228,
      "step": 29928
    },
    {
      "epoch": 1.1587887614950803,
      "grad_norm": 0.3330616056919098,
      "learning_rate": 8.125519564969263e-05,
      "loss": 1.6476,
      "step": 29930
    },
    {
      "epoch": 1.1593307026605923,
      "grad_norm": 0.36129215359687805,
      "learning_rate": 8.124233104020932e-05,
      "loss": 1.6414,
      "step": 29940
    },
    {
      "epoch": 1.1598726438261047,
      "grad_norm": 0.24318821728229523,
      "learning_rate": 8.122946317993999e-05,
      "loss": 1.6465,
      "step": 29950
    },
    {
      "epoch": 1.1604145849916168,
      "grad_norm": 0.2985212206840515,
      "learning_rate": 8.121659207047864e-05,
      "loss": 1.6386,
      "step": 29960
    },
    {
      "epoch": 1.160956526157129,
      "grad_norm": 0.33473873138427734,
      "learning_rate": 8.120371771341968e-05,
      "loss": 1.6402,
      "step": 29970
    },
    {
      "epoch": 1.1614984673226414,
      "grad_norm": 0.2525531053543091,
      "learning_rate": 8.119084011035794e-05,
      "loss": 1.6486,
      "step": 29980
    },
    {
      "epoch": 1.1620404084881535,
      "grad_norm": 0.36281898617744446,
      "learning_rate": 8.117795926288862e-05,
      "loss": 1.6468,
      "step": 29990
    },
    {
      "epoch": 1.1625823496536658,
      "grad_norm": 0.355827659368515,
      "learning_rate": 8.116507517260737e-05,
      "loss": 1.6445,
      "step": 30000
    },
    {
      "epoch": 1.163124290819178,
      "grad_norm": 0.38084909319877625,
      "learning_rate": 8.11521878411102e-05,
      "loss": 1.6413,
      "step": 30010
    },
    {
      "epoch": 1.163395261401934,
      "eval_loss": 2.527758836746216,
      "eval_runtime": 21.9849,
      "eval_samples_per_second": 227.428,
      "eval_steps_per_second": 1.228,
      "step": 30015
    },
    {
      "epoch": 1.1636662319846902,
      "grad_norm": 0.2793338894844055,
      "learning_rate": 8.113929726999354e-05,
      "loss": 1.6487,
      "step": 30020
    },
    {
      "epoch": 1.1642081731502023,
      "grad_norm": 0.31127527356147766,
      "learning_rate": 8.112640346085424e-05,
      "loss": 1.6481,
      "step": 30030
    },
    {
      "epoch": 1.1647501143157146,
      "grad_norm": 0.26151424646377563,
      "learning_rate": 8.11135064152895e-05,
      "loss": 1.6467,
      "step": 30040
    },
    {
      "epoch": 1.1652920554812267,
      "grad_norm": 0.3043818771839142,
      "learning_rate": 8.110060613489693e-05,
      "loss": 1.6401,
      "step": 30050
    },
    {
      "epoch": 1.165833996646739,
      "grad_norm": 0.26495102047920227,
      "learning_rate": 8.108770262127463e-05,
      "loss": 1.6412,
      "step": 30060
    },
    {
      "epoch": 1.1663759378122514,
      "grad_norm": 0.31619176268577576,
      "learning_rate": 8.107479587602097e-05,
      "loss": 1.6383,
      "step": 30070
    },
    {
      "epoch": 1.1669178789777634,
      "grad_norm": 0.22748631238937378,
      "learning_rate": 8.106188590073481e-05,
      "loss": 1.637,
      "step": 30080
    },
    {
      "epoch": 1.1674598201432758,
      "grad_norm": 0.4346810579299927,
      "learning_rate": 8.104897269701538e-05,
      "loss": 1.647,
      "step": 30090
    },
    {
      "epoch": 1.1680017613087879,
      "grad_norm": 0.2576606571674347,
      "learning_rate": 8.103605626646229e-05,
      "loss": 1.6389,
      "step": 30100
    },
    {
      "epoch": 1.1681101495418904,
      "eval_loss": 2.51259708404541,
      "eval_runtime": 21.9879,
      "eval_samples_per_second": 227.398,
      "eval_steps_per_second": 1.228,
      "step": 30102
    },
    {
      "epoch": 1.1685437024743002,
      "grad_norm": 0.31433582305908203,
      "learning_rate": 8.10231366106756e-05,
      "loss": 1.6312,
      "step": 30110
    },
    {
      "epoch": 1.1690856436398123,
      "grad_norm": 0.5255588889122009,
      "learning_rate": 8.101021373125573e-05,
      "loss": 1.634,
      "step": 30120
    },
    {
      "epoch": 1.1696275848053246,
      "grad_norm": 0.23931747674942017,
      "learning_rate": 8.099728762980349e-05,
      "loss": 1.6314,
      "step": 30130
    },
    {
      "epoch": 1.1701695259708367,
      "grad_norm": 0.2556639015674591,
      "learning_rate": 8.098435830792013e-05,
      "loss": 1.6428,
      "step": 30140
    },
    {
      "epoch": 1.170711467136349,
      "grad_norm": 0.4357036352157593,
      "learning_rate": 8.097142576720728e-05,
      "loss": 1.6342,
      "step": 30150
    },
    {
      "epoch": 1.1712534083018613,
      "grad_norm": 0.37833094596862793,
      "learning_rate": 8.095849000926696e-05,
      "loss": 1.6384,
      "step": 30160
    },
    {
      "epoch": 1.1717953494673734,
      "grad_norm": 0.2795352041721344,
      "learning_rate": 8.094555103570161e-05,
      "loss": 1.6411,
      "step": 30170
    },
    {
      "epoch": 1.1723372906328857,
      "grad_norm": 0.3068492114543915,
      "learning_rate": 8.093260884811403e-05,
      "loss": 1.6403,
      "step": 30180
    },
    {
      "epoch": 1.1728250376818468,
      "eval_loss": 2.527621269226074,
      "eval_runtime": 21.9849,
      "eval_samples_per_second": 227.429,
      "eval_steps_per_second": 1.228,
      "step": 30189
    },
    {
      "epoch": 1.1728792317983978,
      "grad_norm": 0.6460037231445312,
      "learning_rate": 8.091966344810746e-05,
      "loss": 1.6299,
      "step": 30190
    },
    {
      "epoch": 1.1734211729639101,
      "grad_norm": 0.5416850447654724,
      "learning_rate": 8.090671483728553e-05,
      "loss": 1.6407,
      "step": 30200
    },
    {
      "epoch": 1.1739631141294224,
      "grad_norm": 0.23775923252105713,
      "learning_rate": 8.08937630172522e-05,
      "loss": 1.6432,
      "step": 30210
    },
    {
      "epoch": 1.1745050552949345,
      "grad_norm": 0.26255470514297485,
      "learning_rate": 8.088080798961196e-05,
      "loss": 1.6431,
      "step": 30220
    },
    {
      "epoch": 1.1750469964604469,
      "grad_norm": 0.2868313491344452,
      "learning_rate": 8.086784975596959e-05,
      "loss": 1.638,
      "step": 30230
    },
    {
      "epoch": 1.175588937625959,
      "grad_norm": 0.3102096617221832,
      "learning_rate": 8.085488831793029e-05,
      "loss": 1.6405,
      "step": 30240
    },
    {
      "epoch": 1.1761308787914713,
      "grad_norm": 0.3971962630748749,
      "learning_rate": 8.084192367709967e-05,
      "loss": 1.6356,
      "step": 30250
    },
    {
      "epoch": 1.1766728199569834,
      "grad_norm": 0.39245739579200745,
      "learning_rate": 8.082895583508374e-05,
      "loss": 1.6448,
      "step": 30260
    },
    {
      "epoch": 1.1772147611224957,
      "grad_norm": 0.31180277466773987,
      "learning_rate": 8.081598479348892e-05,
      "loss": 1.6339,
      "step": 30270
    },
    {
      "epoch": 1.177539925821803,
      "eval_loss": 2.5259082317352295,
      "eval_runtime": 21.9866,
      "eval_samples_per_second": 227.411,
      "eval_steps_per_second": 1.228,
      "step": 30276
    },
    {
      "epoch": 1.1777567022880078,
      "grad_norm": 0.3723970353603363,
      "learning_rate": 8.080301055392196e-05,
      "loss": 1.6444,
      "step": 30280
    },
    {
      "epoch": 1.17829864345352,
      "grad_norm": 0.24248439073562622,
      "learning_rate": 8.079003311799008e-05,
      "loss": 1.6399,
      "step": 30290
    },
    {
      "epoch": 1.1788405846190324,
      "grad_norm": 0.545952320098877,
      "learning_rate": 8.077705248730089e-05,
      "loss": 1.6446,
      "step": 30300
    },
    {
      "epoch": 1.1793825257845445,
      "grad_norm": 0.27861401438713074,
      "learning_rate": 8.076406866346233e-05,
      "loss": 1.6314,
      "step": 30310
    },
    {
      "epoch": 1.1799244669500568,
      "grad_norm": 0.3368987441062927,
      "learning_rate": 8.075108164808281e-05,
      "loss": 1.6445,
      "step": 30320
    },
    {
      "epoch": 1.180466408115569,
      "grad_norm": 0.5238324999809265,
      "learning_rate": 8.073809144277109e-05,
      "loss": 1.6504,
      "step": 30330
    },
    {
      "epoch": 1.1810083492810812,
      "grad_norm": 0.4157116413116455,
      "learning_rate": 8.072509804913634e-05,
      "loss": 1.6359,
      "step": 30340
    },
    {
      "epoch": 1.1815502904465933,
      "grad_norm": 0.4345547556877136,
      "learning_rate": 8.071210146878813e-05,
      "loss": 1.6455,
      "step": 30350
    },
    {
      "epoch": 1.1820922316121056,
      "grad_norm": 0.3230676054954529,
      "learning_rate": 8.069910170333643e-05,
      "loss": 1.6376,
      "step": 30360
    },
    {
      "epoch": 1.1822548139617592,
      "eval_loss": 2.5333669185638428,
      "eval_runtime": 21.9869,
      "eval_samples_per_second": 227.408,
      "eval_steps_per_second": 1.228,
      "step": 30363
    },
    {
      "epoch": 1.1826341727776177,
      "grad_norm": 0.3938463032245636,
      "learning_rate": 8.068609875439159e-05,
      "loss": 1.6364,
      "step": 30370
    },
    {
      "epoch": 1.18317611394313,
      "grad_norm": 0.29343992471694946,
      "learning_rate": 8.067309262356435e-05,
      "loss": 1.6359,
      "step": 30380
    },
    {
      "epoch": 1.1837180551086424,
      "grad_norm": 0.25108593702316284,
      "learning_rate": 8.066008331246586e-05,
      "loss": 1.6415,
      "step": 30390
    },
    {
      "epoch": 1.1842599962741545,
      "grad_norm": 0.6940186619758606,
      "learning_rate": 8.064707082270765e-05,
      "loss": 1.6353,
      "step": 30400
    },
    {
      "epoch": 1.1848019374396668,
      "grad_norm": 0.33529943227767944,
      "learning_rate": 8.063405515590166e-05,
      "loss": 1.6275,
      "step": 30410
    },
    {
      "epoch": 1.1853438786051789,
      "grad_norm": 0.3404565453529358,
      "learning_rate": 8.06210363136602e-05,
      "loss": 1.638,
      "step": 30420
    },
    {
      "epoch": 1.1858858197706912,
      "grad_norm": 0.4092163145542145,
      "learning_rate": 8.0608014297596e-05,
      "loss": 1.6319,
      "step": 30430
    },
    {
      "epoch": 1.1864277609362033,
      "grad_norm": 0.5055906772613525,
      "learning_rate": 8.059498910932216e-05,
      "loss": 1.6342,
      "step": 30440
    },
    {
      "epoch": 1.1869697021017156,
      "grad_norm": 0.38912099599838257,
      "learning_rate": 8.05819607504522e-05,
      "loss": 1.6364,
      "step": 30450
    },
    {
      "epoch": 1.1869697021017156,
      "eval_loss": 2.5219056606292725,
      "eval_runtime": 21.9805,
      "eval_samples_per_second": 227.474,
      "eval_steps_per_second": 1.228,
      "step": 30450
    },
    {
      "epoch": 1.1875116432672277,
      "grad_norm": 0.47896572947502136,
      "learning_rate": 8.056892922260001e-05,
      "loss": 1.6426,
      "step": 30460
    },
    {
      "epoch": 1.18805358443274,
      "grad_norm": 0.2856700122356415,
      "learning_rate": 8.055589452737988e-05,
      "loss": 1.6501,
      "step": 30470
    },
    {
      "epoch": 1.1885955255982523,
      "grad_norm": 0.2293747514486313,
      "learning_rate": 8.054285666640649e-05,
      "loss": 1.6401,
      "step": 30480
    },
    {
      "epoch": 1.1891374667637644,
      "grad_norm": 0.46321722865104675,
      "learning_rate": 8.052981564129489e-05,
      "loss": 1.6478,
      "step": 30490
    },
    {
      "epoch": 1.1896794079292767,
      "grad_norm": 0.39135417342185974,
      "learning_rate": 8.051677145366058e-05,
      "loss": 1.6481,
      "step": 30500
    },
    {
      "epoch": 1.1902213490947888,
      "grad_norm": 0.4103468954563141,
      "learning_rate": 8.050372410511941e-05,
      "loss": 1.6359,
      "step": 30510
    },
    {
      "epoch": 1.1907632902603011,
      "grad_norm": 0.2855057418346405,
      "learning_rate": 8.049067359728763e-05,
      "loss": 1.6489,
      "step": 30520
    },
    {
      "epoch": 1.1913052314258132,
      "grad_norm": 0.3194611966609955,
      "learning_rate": 8.047761993178186e-05,
      "loss": 1.6374,
      "step": 30530
    },
    {
      "epoch": 1.191684590241672,
      "eval_loss": 2.5290892124176025,
      "eval_runtime": 21.9827,
      "eval_samples_per_second": 227.452,
      "eval_steps_per_second": 1.228,
      "step": 30537
    },
    {
      "epoch": 1.1918471725913256,
      "grad_norm": 0.47105491161346436,
      "learning_rate": 8.046456311021916e-05,
      "loss": 1.6399,
      "step": 30540
    },
    {
      "epoch": 1.1923891137568376,
      "grad_norm": 0.3898428976535797,
      "learning_rate": 8.045150313421693e-05,
      "loss": 1.6325,
      "step": 30550
    },
    {
      "epoch": 1.19293105492235,
      "grad_norm": 0.23930542171001434,
      "learning_rate": 8.0438440005393e-05,
      "loss": 1.6297,
      "step": 30560
    },
    {
      "epoch": 1.1934729960878623,
      "grad_norm": 0.31366342306137085,
      "learning_rate": 8.042537372536556e-05,
      "loss": 1.634,
      "step": 30570
    },
    {
      "epoch": 1.1940149372533744,
      "grad_norm": 0.22141672670841217,
      "learning_rate": 8.041230429575319e-05,
      "loss": 1.6477,
      "step": 30580
    },
    {
      "epoch": 1.1945568784188867,
      "grad_norm": 0.5529452562332153,
      "learning_rate": 8.039923171817492e-05,
      "loss": 1.6397,
      "step": 30590
    },
    {
      "epoch": 1.1950988195843988,
      "grad_norm": 0.4657527208328247,
      "learning_rate": 8.038615599425008e-05,
      "loss": 1.6408,
      "step": 30600
    },
    {
      "epoch": 1.195640760749911,
      "grad_norm": 0.4929318130016327,
      "learning_rate": 8.037307712559847e-05,
      "loss": 1.6377,
      "step": 30610
    },
    {
      "epoch": 1.1961827019154234,
      "grad_norm": 0.38976770639419556,
      "learning_rate": 8.03599951138402e-05,
      "loss": 1.6393,
      "step": 30620
    },
    {
      "epoch": 1.196399478381628,
      "eval_loss": 2.534869432449341,
      "eval_runtime": 21.9803,
      "eval_samples_per_second": 227.476,
      "eval_steps_per_second": 1.228,
      "step": 30624
    },
    {
      "epoch": 1.1967246430809355,
      "grad_norm": 0.2525932788848877,
      "learning_rate": 8.034690996059584e-05,
      "loss": 1.6411,
      "step": 30630
    },
    {
      "epoch": 1.1972665842464478,
      "grad_norm": 0.2530060112476349,
      "learning_rate": 8.033382166748633e-05,
      "loss": 1.6505,
      "step": 30640
    },
    {
      "epoch": 1.19780852541196,
      "grad_norm": 0.46789705753326416,
      "learning_rate": 8.032073023613299e-05,
      "loss": 1.6235,
      "step": 30650
    },
    {
      "epoch": 1.1983504665774722,
      "grad_norm": 0.547118604183197,
      "learning_rate": 8.03076356681575e-05,
      "loss": 1.6362,
      "step": 30660
    },
    {
      "epoch": 1.1988924077429843,
      "grad_norm": 0.33731144666671753,
      "learning_rate": 8.029453796518198e-05,
      "loss": 1.6387,
      "step": 30670
    },
    {
      "epoch": 1.1994343489084967,
      "grad_norm": 0.22524616122245789,
      "learning_rate": 8.028143712882893e-05,
      "loss": 1.6338,
      "step": 30680
    },
    {
      "epoch": 1.1999762900740087,
      "grad_norm": 0.46151962876319885,
      "learning_rate": 8.02683331607212e-05,
      "loss": 1.6475,
      "step": 30690
    },
    {
      "epoch": 1.200518231239521,
      "grad_norm": 0.313071072101593,
      "learning_rate": 8.025522606248206e-05,
      "loss": 1.6416,
      "step": 30700
    },
    {
      "epoch": 1.2010601724050334,
      "grad_norm": 0.480162113904953,
      "learning_rate": 8.024211583573516e-05,
      "loss": 1.6333,
      "step": 30710
    },
    {
      "epoch": 1.2011143665215844,
      "eval_loss": 2.5296213626861572,
      "eval_runtime": 21.9849,
      "eval_samples_per_second": 227.429,
      "eval_steps_per_second": 1.228,
      "step": 30711
    },
    {
      "epoch": 1.2016021135705455,
      "grad_norm": 0.3885667622089386,
      "learning_rate": 8.022900248210455e-05,
      "loss": 1.6343,
      "step": 30720
    },
    {
      "epoch": 1.2021440547360578,
      "grad_norm": 0.34139159321784973,
      "learning_rate": 8.021588600321465e-05,
      "loss": 1.6332,
      "step": 30730
    },
    {
      "epoch": 1.2026859959015699,
      "grad_norm": 0.2892656922340393,
      "learning_rate": 8.020276640069025e-05,
      "loss": 1.6362,
      "step": 30740
    },
    {
      "epoch": 1.2032279370670822,
      "grad_norm": 0.4062507748603821,
      "learning_rate": 8.018964367615659e-05,
      "loss": 1.6388,
      "step": 30750
    },
    {
      "epoch": 1.2037698782325943,
      "grad_norm": 0.3714730739593506,
      "learning_rate": 8.017651783123922e-05,
      "loss": 1.63,
      "step": 30760
    },
    {
      "epoch": 1.2043118193981066,
      "grad_norm": 0.5153228044509888,
      "learning_rate": 8.016338886756412e-05,
      "loss": 1.6387,
      "step": 30770
    },
    {
      "epoch": 1.2048537605636187,
      "grad_norm": 0.27551186084747314,
      "learning_rate": 8.015025678675767e-05,
      "loss": 1.634,
      "step": 30780
    },
    {
      "epoch": 1.205395701729131,
      "grad_norm": 0.8653702139854431,
      "learning_rate": 8.013712159044658e-05,
      "loss": 1.6362,
      "step": 30790
    },
    {
      "epoch": 1.2058292546615408,
      "eval_loss": 2.5252997875213623,
      "eval_runtime": 21.986,
      "eval_samples_per_second": 227.418,
      "eval_steps_per_second": 1.228,
      "step": 30798
    },
    {
      "epoch": 1.2059376428946433,
      "grad_norm": 0.8155677914619446,
      "learning_rate": 8.012398328025804e-05,
      "loss": 1.6397,
      "step": 30800
    },
    {
      "epoch": 1.2064795840601554,
      "grad_norm": 0.49000170826911926,
      "learning_rate": 8.01108418578195e-05,
      "loss": 1.6407,
      "step": 30810
    },
    {
      "epoch": 1.2070215252256677,
      "grad_norm": 0.37893688678741455,
      "learning_rate": 8.009769732475889e-05,
      "loss": 1.6368,
      "step": 30820
    },
    {
      "epoch": 1.2075634663911798,
      "grad_norm": 0.3986130356788635,
      "learning_rate": 8.008454968270452e-05,
      "loss": 1.634,
      "step": 30830
    },
    {
      "epoch": 1.2081054075566922,
      "grad_norm": 0.33178648352622986,
      "learning_rate": 8.0071398933285e-05,
      "loss": 1.6315,
      "step": 30840
    },
    {
      "epoch": 1.2086473487222043,
      "grad_norm": 0.3417208790779114,
      "learning_rate": 8.005824507812947e-05,
      "loss": 1.6246,
      "step": 30850
    },
    {
      "epoch": 1.2091892898877166,
      "grad_norm": 0.4329122006893158,
      "learning_rate": 8.004508811886732e-05,
      "loss": 1.6451,
      "step": 30860
    },
    {
      "epoch": 1.2097312310532287,
      "grad_norm": 0.4540995657444,
      "learning_rate": 8.003192805712839e-05,
      "loss": 1.6188,
      "step": 30870
    },
    {
      "epoch": 1.210273172218741,
      "grad_norm": 0.41036659479141235,
      "learning_rate": 8.001876489454289e-05,
      "loss": 1.6392,
      "step": 30880
    },
    {
      "epoch": 1.2105441428014971,
      "eval_loss": 2.5213656425476074,
      "eval_runtime": 21.9838,
      "eval_samples_per_second": 227.441,
      "eval_steps_per_second": 1.228,
      "step": 30885
    },
    {
      "epoch": 1.2108151133842533,
      "grad_norm": 0.2853230834007263,
      "learning_rate": 8.00055986327414e-05,
      "loss": 1.6335,
      "step": 30890
    },
    {
      "epoch": 1.2113570545497654,
      "grad_norm": 0.23493582010269165,
      "learning_rate": 7.999242927335493e-05,
      "loss": 1.6369,
      "step": 30900
    },
    {
      "epoch": 1.2118989957152777,
      "grad_norm": 0.4317961037158966,
      "learning_rate": 7.997925681801484e-05,
      "loss": 1.6346,
      "step": 30910
    },
    {
      "epoch": 1.2124409368807898,
      "grad_norm": 0.29279986023902893,
      "learning_rate": 7.996608126835285e-05,
      "loss": 1.6326,
      "step": 30920
    },
    {
      "epoch": 1.2129828780463021,
      "grad_norm": 0.3215675950050354,
      "learning_rate": 7.995290262600109e-05,
      "loss": 1.6445,
      "step": 30930
    },
    {
      "epoch": 1.2135248192118144,
      "grad_norm": 0.6567270755767822,
      "learning_rate": 7.99397208925921e-05,
      "loss": 1.6359,
      "step": 30940
    },
    {
      "epoch": 1.2140667603773265,
      "grad_norm": 0.5213804841041565,
      "learning_rate": 7.992653606975877e-05,
      "loss": 1.6262,
      "step": 30950
    },
    {
      "epoch": 1.2146087015428388,
      "grad_norm": 0.3358005881309509,
      "learning_rate": 7.991334815913437e-05,
      "loss": 1.6371,
      "step": 30960
    },
    {
      "epoch": 1.215150642708351,
      "grad_norm": 0.26169344782829285,
      "learning_rate": 7.990015716235255e-05,
      "loss": 1.6329,
      "step": 30970
    },
    {
      "epoch": 1.2152590309414535,
      "eval_loss": 2.521756172180176,
      "eval_runtime": 21.9857,
      "eval_samples_per_second": 227.421,
      "eval_steps_per_second": 1.228,
      "step": 30972
    },
    {
      "epoch": 1.2156925838738633,
      "grad_norm": 0.3503156304359436,
      "learning_rate": 7.988696308104738e-05,
      "loss": 1.6396,
      "step": 30980
    },
    {
      "epoch": 1.2162345250393753,
      "grad_norm": 0.4474554657936096,
      "learning_rate": 7.987376591685325e-05,
      "loss": 1.6303,
      "step": 30990
    },
    {
      "epoch": 1.2167764662048877,
      "grad_norm": 0.25015437602996826,
      "learning_rate": 7.986056567140502e-05,
      "loss": 1.6229,
      "step": 31000
    },
    {
      "epoch": 1.2173184073703998,
      "grad_norm": 0.40038561820983887,
      "learning_rate": 7.984736234633784e-05,
      "loss": 1.6299,
      "step": 31010
    },
    {
      "epoch": 1.217860348535912,
      "grad_norm": 0.3585070073604584,
      "learning_rate": 7.983415594328729e-05,
      "loss": 1.644,
      "step": 31020
    },
    {
      "epoch": 1.2184022897014244,
      "grad_norm": 0.2341560274362564,
      "learning_rate": 7.98209464638893e-05,
      "loss": 1.6389,
      "step": 31030
    },
    {
      "epoch": 1.2189442308669365,
      "grad_norm": 0.32264870405197144,
      "learning_rate": 7.980773390978024e-05,
      "loss": 1.635,
      "step": 31040
    },
    {
      "epoch": 1.2194861720324488,
      "grad_norm": 0.46334415674209595,
      "learning_rate": 7.979451828259681e-05,
      "loss": 1.6315,
      "step": 31050
    },
    {
      "epoch": 1.2199739190814096,
      "eval_loss": 2.524176597595215,
      "eval_runtime": 21.9898,
      "eval_samples_per_second": 227.379,
      "eval_steps_per_second": 1.228,
      "step": 31059
    },
    {
      "epoch": 1.220028113197961,
      "grad_norm": 0.2632630467414856,
      "learning_rate": 7.978129958397612e-05,
      "loss": 1.6317,
      "step": 31060
    },
    {
      "epoch": 1.2205700543634732,
      "grad_norm": 0.5674657225608826,
      "learning_rate": 7.97680778155556e-05,
      "loss": 1.6294,
      "step": 31070
    },
    {
      "epoch": 1.2211119955289853,
      "grad_norm": 0.4556896686553955,
      "learning_rate": 7.975485297897312e-05,
      "loss": 1.633,
      "step": 31080
    },
    {
      "epoch": 1.2216539366944976,
      "grad_norm": 0.362857848405838,
      "learning_rate": 7.974162507586696e-05,
      "loss": 1.632,
      "step": 31090
    },
    {
      "epoch": 1.2221958778600097,
      "grad_norm": 0.3741185963153839,
      "learning_rate": 7.972839410787568e-05,
      "loss": 1.6396,
      "step": 31100
    },
    {
      "epoch": 1.222737819025522,
      "grad_norm": 0.5633114576339722,
      "learning_rate": 7.971516007663831e-05,
      "loss": 1.6354,
      "step": 31110
    },
    {
      "epoch": 1.2232797601910343,
      "grad_norm": 0.3228178322315216,
      "learning_rate": 7.970192298379421e-05,
      "loss": 1.6383,
      "step": 31120
    },
    {
      "epoch": 1.2238217013565464,
      "grad_norm": 0.39119747281074524,
      "learning_rate": 7.968868283098314e-05,
      "loss": 1.6271,
      "step": 31130
    },
    {
      "epoch": 1.2243636425220588,
      "grad_norm": 0.28056901693344116,
      "learning_rate": 7.967543961984522e-05,
      "loss": 1.6356,
      "step": 31140
    },
    {
      "epoch": 1.224688807221366,
      "eval_loss": 2.5236127376556396,
      "eval_runtime": 21.9859,
      "eval_samples_per_second": 227.418,
      "eval_steps_per_second": 1.228,
      "step": 31146
    },
    {
      "epoch": 1.2249055836875709,
      "grad_norm": 0.4328171908855438,
      "learning_rate": 7.966219335202097e-05,
      "loss": 1.6377,
      "step": 31150
    },
    {
      "epoch": 1.2254475248530832,
      "grad_norm": 0.3190264105796814,
      "learning_rate": 7.964894402915127e-05,
      "loss": 1.6327,
      "step": 31160
    },
    {
      "epoch": 1.2259894660185953,
      "grad_norm": 0.38925376534461975,
      "learning_rate": 7.963569165287743e-05,
      "loss": 1.6092,
      "step": 31170
    },
    {
      "epoch": 1.2265314071841076,
      "grad_norm": 0.2989320755004883,
      "learning_rate": 7.962243622484104e-05,
      "loss": 1.6305,
      "step": 31180
    },
    {
      "epoch": 1.2270733483496197,
      "grad_norm": 0.21814024448394775,
      "learning_rate": 7.960917774668415e-05,
      "loss": 1.6351,
      "step": 31190
    },
    {
      "epoch": 1.227615289515132,
      "grad_norm": 0.23537899553775787,
      "learning_rate": 7.95959162200492e-05,
      "loss": 1.6261,
      "step": 31200
    },
    {
      "epoch": 1.2281572306806443,
      "grad_norm": 0.3734976649284363,
      "learning_rate": 7.958265164657889e-05,
      "loss": 1.6297,
      "step": 31210
    },
    {
      "epoch": 1.2286991718461564,
      "grad_norm": 0.3379276990890503,
      "learning_rate": 7.956938402791644e-05,
      "loss": 1.622,
      "step": 31220
    },
    {
      "epoch": 1.2292411130116687,
      "grad_norm": 0.3800952434539795,
      "learning_rate": 7.955611336570537e-05,
      "loss": 1.6307,
      "step": 31230
    },
    {
      "epoch": 1.2294036953613223,
      "eval_loss": 2.5284194946289062,
      "eval_runtime": 21.9883,
      "eval_samples_per_second": 227.393,
      "eval_steps_per_second": 1.228,
      "step": 31233
    },
    {
      "epoch": 1.2297830541771808,
      "grad_norm": 0.36053764820098877,
      "learning_rate": 7.954283966158957e-05,
      "loss": 1.6258,
      "step": 31240
    },
    {
      "epoch": 1.2303249953426931,
      "grad_norm": 0.24756401777267456,
      "learning_rate": 7.952956291721335e-05,
      "loss": 1.6383,
      "step": 31250
    },
    {
      "epoch": 1.2308669365082054,
      "grad_norm": 0.24404959380626678,
      "learning_rate": 7.951628313422139e-05,
      "loss": 1.6229,
      "step": 31260
    },
    {
      "epoch": 1.2314088776737175,
      "grad_norm": 0.2366773635149002,
      "learning_rate": 7.95030003142587e-05,
      "loss": 1.6288,
      "step": 31270
    },
    {
      "epoch": 1.2319508188392299,
      "grad_norm": 0.6960218548774719,
      "learning_rate": 7.948971445897072e-05,
      "loss": 1.6308,
      "step": 31280
    },
    {
      "epoch": 1.232492760004742,
      "grad_norm": 0.24987642467021942,
      "learning_rate": 7.947642557000324e-05,
      "loss": 1.6346,
      "step": 31290
    },
    {
      "epoch": 1.2330347011702543,
      "grad_norm": 0.3360191583633423,
      "learning_rate": 7.946313364900242e-05,
      "loss": 1.6323,
      "step": 31300
    },
    {
      "epoch": 1.2335766423357664,
      "grad_norm": 0.3646528124809265,
      "learning_rate": 7.944983869761481e-05,
      "loss": 1.6235,
      "step": 31310
    },
    {
      "epoch": 1.2341185835012787,
      "grad_norm": 0.29639899730682373,
      "learning_rate": 7.943654071748734e-05,
      "loss": 1.6291,
      "step": 31320
    },
    {
      "epoch": 1.2341185835012787,
      "eval_loss": 2.529906749725342,
      "eval_runtime": 21.9492,
      "eval_samples_per_second": 227.799,
      "eval_steps_per_second": 1.23,
      "step": 31320
    },
    {
      "epoch": 1.2346605246667908,
      "grad_norm": 0.2218153476715088,
      "learning_rate": 7.942323971026729e-05,
      "loss": 1.6226,
      "step": 31330
    },
    {
      "epoch": 1.235202465832303,
      "grad_norm": 0.25773563981056213,
      "learning_rate": 7.940993567760235e-05,
      "loss": 1.6332,
      "step": 31340
    },
    {
      "epoch": 1.2357444069978154,
      "grad_norm": 0.26830193400382996,
      "learning_rate": 7.939662862114053e-05,
      "loss": 1.6398,
      "step": 31350
    },
    {
      "epoch": 1.2362863481633275,
      "grad_norm": 0.3047340214252472,
      "learning_rate": 7.938331854253031e-05,
      "loss": 1.6377,
      "step": 31360
    },
    {
      "epoch": 1.2368282893288398,
      "grad_norm": 0.39592602849006653,
      "learning_rate": 7.937000544342042e-05,
      "loss": 1.6235,
      "step": 31370
    },
    {
      "epoch": 1.237370230494352,
      "grad_norm": 0.39517736434936523,
      "learning_rate": 7.935668932546009e-05,
      "loss": 1.639,
      "step": 31380
    },
    {
      "epoch": 1.2379121716598642,
      "grad_norm": 0.3887443542480469,
      "learning_rate": 7.934337019029881e-05,
      "loss": 1.6261,
      "step": 31390
    },
    {
      "epoch": 1.2384541128253763,
      "grad_norm": 0.5671885013580322,
      "learning_rate": 7.933004803958654e-05,
      "loss": 1.6313,
      "step": 31400
    },
    {
      "epoch": 1.238833471641235,
      "eval_loss": 2.5245509147644043,
      "eval_runtime": 21.9815,
      "eval_samples_per_second": 227.464,
      "eval_steps_per_second": 1.228,
      "step": 31407
    },
    {
      "epoch": 1.2389960539908886,
      "grad_norm": 0.3325177729129791,
      "learning_rate": 7.931672287497353e-05,
      "loss": 1.6346,
      "step": 31410
    },
    {
      "epoch": 1.2395379951564007,
      "grad_norm": 0.36562344431877136,
      "learning_rate": 7.930339469811045e-05,
      "loss": 1.6241,
      "step": 31420
    },
    {
      "epoch": 1.240079936321913,
      "grad_norm": 0.2752797603607178,
      "learning_rate": 7.929006351064838e-05,
      "loss": 1.6228,
      "step": 31430
    },
    {
      "epoch": 1.2406218774874254,
      "grad_norm": 0.27838945388793945,
      "learning_rate": 7.927672931423869e-05,
      "loss": 1.6249,
      "step": 31440
    },
    {
      "epoch": 1.2411638186529375,
      "grad_norm": 0.40533244609832764,
      "learning_rate": 7.926339211053316e-05,
      "loss": 1.6243,
      "step": 31450
    },
    {
      "epoch": 1.2417057598184498,
      "grad_norm": 0.47712355852127075,
      "learning_rate": 7.925005190118397e-05,
      "loss": 1.6253,
      "step": 31460
    },
    {
      "epoch": 1.2422477009839619,
      "grad_norm": 0.3724726140499115,
      "learning_rate": 7.923670868784364e-05,
      "loss": 1.6356,
      "step": 31470
    },
    {
      "epoch": 1.2427896421494742,
      "grad_norm": 0.26425930857658386,
      "learning_rate": 7.922336247216505e-05,
      "loss": 1.6291,
      "step": 31480
    },
    {
      "epoch": 1.2433315833149863,
      "grad_norm": 0.2903880476951599,
      "learning_rate": 7.92100132558015e-05,
      "loss": 1.6299,
      "step": 31490
    },
    {
      "epoch": 1.2435483597811912,
      "eval_loss": 2.520765781402588,
      "eval_runtime": 21.9829,
      "eval_samples_per_second": 227.449,
      "eval_steps_per_second": 1.228,
      "step": 31494
    },
    {
      "epoch": 1.2438735244804986,
      "grad_norm": 0.3055003881454468,
      "learning_rate": 7.91966610404066e-05,
      "loss": 1.6376,
      "step": 31500
    },
    {
      "epoch": 1.2444154656460107,
      "grad_norm": 0.37320762872695923,
      "learning_rate": 7.918330582763438e-05,
      "loss": 1.6298,
      "step": 31510
    },
    {
      "epoch": 1.244957406811523,
      "grad_norm": 0.2776867747306824,
      "learning_rate": 7.916994761913923e-05,
      "loss": 1.6266,
      "step": 31520
    },
    {
      "epoch": 1.2454993479770353,
      "grad_norm": 0.2319355010986328,
      "learning_rate": 7.91565864165759e-05,
      "loss": 1.6252,
      "step": 31530
    },
    {
      "epoch": 1.2460412891425474,
      "grad_norm": 0.5147307515144348,
      "learning_rate": 7.914322222159953e-05,
      "loss": 1.6261,
      "step": 31540
    },
    {
      "epoch": 1.2465832303080597,
      "grad_norm": 0.31287848949432373,
      "learning_rate": 7.912985503586562e-05,
      "loss": 1.6304,
      "step": 31550
    },
    {
      "epoch": 1.2471251714735718,
      "grad_norm": 0.2941262423992157,
      "learning_rate": 7.911648486103002e-05,
      "loss": 1.6325,
      "step": 31560
    },
    {
      "epoch": 1.2476671126390841,
      "grad_norm": 0.2733207941055298,
      "learning_rate": 7.910311169874898e-05,
      "loss": 1.6273,
      "step": 31570
    },
    {
      "epoch": 1.2482090538045965,
      "grad_norm": 0.40679219365119934,
      "learning_rate": 7.908973555067911e-05,
      "loss": 1.6264,
      "step": 31580
    },
    {
      "epoch": 1.2482632479211475,
      "eval_loss": 2.5255589485168457,
      "eval_runtime": 21.9824,
      "eval_samples_per_second": 227.454,
      "eval_steps_per_second": 1.228,
      "step": 31581
    },
    {
      "epoch": 1.2487509949701086,
      "grad_norm": 0.27531698346138,
      "learning_rate": 7.907635641847739e-05,
      "loss": 1.6209,
      "step": 31590
    },
    {
      "epoch": 1.2492929361356209,
      "grad_norm": 0.26964595913887024,
      "learning_rate": 7.906297430380114e-05,
      "loss": 1.6253,
      "step": 31600
    },
    {
      "epoch": 1.249834877301133,
      "grad_norm": 0.25738173723220825,
      "learning_rate": 7.904958920830813e-05,
      "loss": 1.6205,
      "step": 31610
    },
    {
      "epoch": 1.2503768184666453,
      "grad_norm": 0.2674032151699066,
      "learning_rate": 7.903620113365644e-05,
      "loss": 1.6191,
      "step": 31620
    },
    {
      "epoch": 1.2509187596321574,
      "grad_norm": 0.42425525188446045,
      "learning_rate": 7.902281008150447e-05,
      "loss": 1.6345,
      "step": 31630
    },
    {
      "epoch": 1.2514607007976697,
      "grad_norm": 0.4274544417858124,
      "learning_rate": 7.90094160535111e-05,
      "loss": 1.6372,
      "step": 31640
    },
    {
      "epoch": 1.2520026419631818,
      "grad_norm": 0.36170271039009094,
      "learning_rate": 7.89960190513355e-05,
      "loss": 1.6288,
      "step": 31650
    },
    {
      "epoch": 1.252544583128694,
      "grad_norm": 0.35101965069770813,
      "learning_rate": 7.898261907663725e-05,
      "loss": 1.6218,
      "step": 31660
    },
    {
      "epoch": 1.2529781360611039,
      "eval_loss": 2.5185720920562744,
      "eval_runtime": 21.9807,
      "eval_samples_per_second": 227.472,
      "eval_steps_per_second": 1.228,
      "step": 31668
    },
    {
      "epoch": 1.2530865242942064,
      "grad_norm": 0.2534119188785553,
      "learning_rate": 7.896921613107626e-05,
      "loss": 1.6279,
      "step": 31670
    },
    {
      "epoch": 1.2536284654597185,
      "grad_norm": 0.4205189049243927,
      "learning_rate": 7.895581021631286e-05,
      "loss": 1.623,
      "step": 31680
    },
    {
      "epoch": 1.2541704066252308,
      "grad_norm": 0.29453709721565247,
      "learning_rate": 7.894240133400767e-05,
      "loss": 1.6259,
      "step": 31690
    },
    {
      "epoch": 1.254712347790743,
      "grad_norm": 0.3749118149280548,
      "learning_rate": 7.892898948582177e-05,
      "loss": 1.6186,
      "step": 31700
    },
    {
      "epoch": 1.2552542889562552,
      "grad_norm": 0.3719863295555115,
      "learning_rate": 7.891557467341653e-05,
      "loss": 1.6233,
      "step": 31710
    },
    {
      "epoch": 1.2557962301217673,
      "grad_norm": 0.2719629406929016,
      "learning_rate": 7.890215689845374e-05,
      "loss": 1.642,
      "step": 31720
    },
    {
      "epoch": 1.2563381712872796,
      "grad_norm": 0.29956310987472534,
      "learning_rate": 7.888873616259552e-05,
      "loss": 1.6394,
      "step": 31730
    },
    {
      "epoch": 1.2568801124527917,
      "grad_norm": 0.6171798706054688,
      "learning_rate": 7.887531246750438e-05,
      "loss": 1.6224,
      "step": 31740
    },
    {
      "epoch": 1.257422053618304,
      "grad_norm": 0.42021244764328003,
      "learning_rate": 7.886188581484318e-05,
      "loss": 1.6289,
      "step": 31750
    },
    {
      "epoch": 1.2576930242010602,
      "eval_loss": 2.518956422805786,
      "eval_runtime": 21.9853,
      "eval_samples_per_second": 227.425,
      "eval_steps_per_second": 1.228,
      "step": 31755
    },
    {
      "epoch": 1.2579639947838164,
      "grad_norm": 0.3666936159133911,
      "learning_rate": 7.884845620627518e-05,
      "loss": 1.6277,
      "step": 31760
    },
    {
      "epoch": 1.2585059359493285,
      "grad_norm": 0.3201642334461212,
      "learning_rate": 7.883502364346396e-05,
      "loss": 1.6311,
      "step": 31770
    },
    {
      "epoch": 1.2590478771148408,
      "grad_norm": 0.2944391369819641,
      "learning_rate": 7.882158812807349e-05,
      "loss": 1.6259,
      "step": 31780
    },
    {
      "epoch": 1.2595898182803529,
      "grad_norm": 0.2937580943107605,
      "learning_rate": 7.88081496617681e-05,
      "loss": 1.626,
      "step": 31790
    },
    {
      "epoch": 1.2601317594458652,
      "grad_norm": 0.3028720021247864,
      "learning_rate": 7.87947082462125e-05,
      "loss": 1.6325,
      "step": 31800
    },
    {
      "epoch": 1.2606737006113775,
      "grad_norm": 0.2833790183067322,
      "learning_rate": 7.878126388307173e-05,
      "loss": 1.6247,
      "step": 31810
    },
    {
      "epoch": 1.2612156417768896,
      "grad_norm": 0.264164000749588,
      "learning_rate": 7.876781657401125e-05,
      "loss": 1.627,
      "step": 31820
    },
    {
      "epoch": 1.2617575829424017,
      "grad_norm": 0.3081631064414978,
      "learning_rate": 7.875436632069687e-05,
      "loss": 1.6225,
      "step": 31830
    },
    {
      "epoch": 1.262299524107914,
      "grad_norm": 0.4150921404361725,
      "learning_rate": 7.874091312479468e-05,
      "loss": 1.6336,
      "step": 31840
    },
    {
      "epoch": 1.2624079123410166,
      "eval_loss": 2.5215582847595215,
      "eval_runtime": 21.9862,
      "eval_samples_per_second": 227.416,
      "eval_steps_per_second": 1.228,
      "step": 31842
    },
    {
      "epoch": 1.2628414652734263,
      "grad_norm": 0.689495861530304,
      "learning_rate": 7.872745698797128e-05,
      "loss": 1.6334,
      "step": 31850
    },
    {
      "epoch": 1.2633834064389384,
      "grad_norm": 0.501124918460846,
      "learning_rate": 7.87139979118935e-05,
      "loss": 1.6275,
      "step": 31860
    },
    {
      "epoch": 1.2639253476044507,
      "grad_norm": 0.2936415672302246,
      "learning_rate": 7.870053589822863e-05,
      "loss": 1.6368,
      "step": 31870
    },
    {
      "epoch": 1.2644672887699628,
      "grad_norm": 0.3334285616874695,
      "learning_rate": 7.868707094864427e-05,
      "loss": 1.6255,
      "step": 31880
    },
    {
      "epoch": 1.2650092299354752,
      "grad_norm": 0.2995965778827667,
      "learning_rate": 7.867360306480839e-05,
      "loss": 1.6214,
      "step": 31890
    },
    {
      "epoch": 1.2655511711009875,
      "grad_norm": 0.3712153732776642,
      "learning_rate": 7.866013224838933e-05,
      "loss": 1.634,
      "step": 31900
    },
    {
      "epoch": 1.2660931122664996,
      "grad_norm": 0.30742502212524414,
      "learning_rate": 7.864665850105583e-05,
      "loss": 1.6193,
      "step": 31910
    },
    {
      "epoch": 1.2666350534320117,
      "grad_norm": 0.27019500732421875,
      "learning_rate": 7.863318182447693e-05,
      "loss": 1.627,
      "step": 31920
    },
    {
      "epoch": 1.2671228004809727,
      "eval_loss": 2.5232045650482178,
      "eval_runtime": 21.9899,
      "eval_samples_per_second": 227.377,
      "eval_steps_per_second": 1.228,
      "step": 31929
    },
    {
      "epoch": 1.267176994597524,
      "grad_norm": 0.3056955337524414,
      "learning_rate": 7.861970222032207e-05,
      "loss": 1.6083,
      "step": 31930
    },
    {
      "epoch": 1.2677189357630363,
      "grad_norm": 0.28661999106407166,
      "learning_rate": 7.860621969026106e-05,
      "loss": 1.6225,
      "step": 31940
    },
    {
      "epoch": 1.2682608769285484,
      "grad_norm": 0.26536825299263,
      "learning_rate": 7.859273423596403e-05,
      "loss": 1.6166,
      "step": 31950
    },
    {
      "epoch": 1.2688028180940607,
      "grad_norm": 0.5294485092163086,
      "learning_rate": 7.85792458591015e-05,
      "loss": 1.6225,
      "step": 31960
    },
    {
      "epoch": 1.2693447592595728,
      "grad_norm": 0.3292939364910126,
      "learning_rate": 7.85657545613444e-05,
      "loss": 1.6334,
      "step": 31970
    },
    {
      "epoch": 1.2698867004250851,
      "grad_norm": 0.23607225716114044,
      "learning_rate": 7.85522603443639e-05,
      "loss": 1.6311,
      "step": 31980
    },
    {
      "epoch": 1.2704286415905974,
      "grad_norm": 0.28158169984817505,
      "learning_rate": 7.853876320983165e-05,
      "loss": 1.622,
      "step": 31990
    },
    {
      "epoch": 1.2709705827561095,
      "grad_norm": 0.2377581000328064,
      "learning_rate": 7.852526315941961e-05,
      "loss": 1.623,
      "step": 32000
    },
    {
      "epoch": 1.2715125239216218,
      "grad_norm": 0.514994740486145,
      "learning_rate": 7.851176019480012e-05,
      "loss": 1.6112,
      "step": 32010
    },
    {
      "epoch": 1.271837688620929,
      "eval_loss": 2.515641689300537,
      "eval_runtime": 21.988,
      "eval_samples_per_second": 227.397,
      "eval_steps_per_second": 1.228,
      "step": 32016
    },
    {
      "epoch": 1.272054465087134,
      "grad_norm": 0.3797670900821686,
      "learning_rate": 7.849825431764585e-05,
      "loss": 1.6232,
      "step": 32020
    },
    {
      "epoch": 1.2725964062526463,
      "grad_norm": 0.44352027773857117,
      "learning_rate": 7.848474552962984e-05,
      "loss": 1.6295,
      "step": 32030
    },
    {
      "epoch": 1.2731383474181583,
      "grad_norm": 0.2417132705450058,
      "learning_rate": 7.847123383242552e-05,
      "loss": 1.6249,
      "step": 32040
    },
    {
      "epoch": 1.2736802885836707,
      "grad_norm": 0.4180681109428406,
      "learning_rate": 7.845771922770667e-05,
      "loss": 1.6065,
      "step": 32050
    },
    {
      "epoch": 1.2742222297491828,
      "grad_norm": 0.29426735639572144,
      "learning_rate": 7.84442017171474e-05,
      "loss": 1.6243,
      "step": 32060
    },
    {
      "epoch": 1.274764170914695,
      "grad_norm": 0.37015318870544434,
      "learning_rate": 7.84306813024222e-05,
      "loss": 1.6223,
      "step": 32070
    },
    {
      "epoch": 1.2753061120802074,
      "grad_norm": 0.2594332993030548,
      "learning_rate": 7.841715798520592e-05,
      "loss": 1.6248,
      "step": 32080
    },
    {
      "epoch": 1.2758480532457195,
      "grad_norm": 0.23385967314243317,
      "learning_rate": 7.840363176717377e-05,
      "loss": 1.6098,
      "step": 32090
    },
    {
      "epoch": 1.2763899944112318,
      "grad_norm": 0.38536781072616577,
      "learning_rate": 7.839010265000136e-05,
      "loss": 1.6224,
      "step": 32100
    },
    {
      "epoch": 1.2765525767608854,
      "eval_loss": 2.518597364425659,
      "eval_runtime": 21.9837,
      "eval_samples_per_second": 227.441,
      "eval_steps_per_second": 1.228,
      "step": 32103
    },
    {
      "epoch": 1.276931935576744,
      "grad_norm": 0.2687205374240875,
      "learning_rate": 7.837657063536456e-05,
      "loss": 1.6277,
      "step": 32110
    },
    {
      "epoch": 1.2774738767422562,
      "grad_norm": 0.29153600335121155,
      "learning_rate": 7.83630357249397e-05,
      "loss": 1.6221,
      "step": 32120
    },
    {
      "epoch": 1.2780158179077685,
      "grad_norm": 0.42591148614883423,
      "learning_rate": 7.834949792040337e-05,
      "loss": 1.6185,
      "step": 32130
    },
    {
      "epoch": 1.2785577590732806,
      "grad_norm": 0.5398927927017212,
      "learning_rate": 7.833595722343263e-05,
      "loss": 1.6177,
      "step": 32140
    },
    {
      "epoch": 1.2790997002387927,
      "grad_norm": 0.3706096410751343,
      "learning_rate": 7.832241363570482e-05,
      "loss": 1.6266,
      "step": 32150
    },
    {
      "epoch": 1.279641641404305,
      "grad_norm": 0.34234559535980225,
      "learning_rate": 7.830886715889766e-05,
      "loss": 1.6242,
      "step": 32160
    },
    {
      "epoch": 1.2801835825698173,
      "grad_norm": 0.288562148809433,
      "learning_rate": 7.829531779468925e-05,
      "loss": 1.6247,
      "step": 32170
    },
    {
      "epoch": 1.2807255237353294,
      "grad_norm": 0.5007647275924683,
      "learning_rate": 7.8281765544758e-05,
      "loss": 1.6259,
      "step": 32180
    },
    {
      "epoch": 1.2812674649008418,
      "grad_norm": 0.2426513284444809,
      "learning_rate": 7.826821041078271e-05,
      "loss": 1.6284,
      "step": 32190
    },
    {
      "epoch": 1.2812674649008418,
      "eval_loss": 2.516087293624878,
      "eval_runtime": 21.9703,
      "eval_samples_per_second": 227.58,
      "eval_steps_per_second": 1.229,
      "step": 32190
    },
    {
      "epoch": 1.2818094060663539,
      "grad_norm": 0.2416008710861206,
      "learning_rate": 7.825465239444255e-05,
      "loss": 1.6283,
      "step": 32200
    },
    {
      "epoch": 1.2823513472318662,
      "grad_norm": 0.3502400517463684,
      "learning_rate": 7.824109149741701e-05,
      "loss": 1.6253,
      "step": 32210
    },
    {
      "epoch": 1.2828932883973785,
      "grad_norm": 0.2979862689971924,
      "learning_rate": 7.822752772138594e-05,
      "loss": 1.6324,
      "step": 32220
    },
    {
      "epoch": 1.2834352295628906,
      "grad_norm": 0.355049192905426,
      "learning_rate": 7.821396106802958e-05,
      "loss": 1.6229,
      "step": 32230
    },
    {
      "epoch": 1.2839771707284027,
      "grad_norm": 0.29975321888923645,
      "learning_rate": 7.820039153902852e-05,
      "loss": 1.6195,
      "step": 32240
    },
    {
      "epoch": 1.284519111893915,
      "grad_norm": 0.3482322692871094,
      "learning_rate": 7.81868191360637e-05,
      "loss": 1.6228,
      "step": 32250
    },
    {
      "epoch": 1.2850610530594273,
      "grad_norm": 0.20890609920024872,
      "learning_rate": 7.817324386081637e-05,
      "loss": 1.6212,
      "step": 32260
    },
    {
      "epoch": 1.2856029942249394,
      "grad_norm": 0.35245707631111145,
      "learning_rate": 7.81596657149682e-05,
      "loss": 1.613,
      "step": 32270
    },
    {
      "epoch": 1.285982353040798,
      "eval_loss": 2.5217950344085693,
      "eval_runtime": 21.9904,
      "eval_samples_per_second": 227.372,
      "eval_steps_per_second": 1.228,
      "step": 32277
    },
    {
      "epoch": 1.2861449353904517,
      "grad_norm": 0.3934645354747772,
      "learning_rate": 7.814608470020118e-05,
      "loss": 1.6187,
      "step": 32280
    },
    {
      "epoch": 1.2866868765559638,
      "grad_norm": 0.2991849482059479,
      "learning_rate": 7.81325008181977e-05,
      "loss": 1.6208,
      "step": 32290
    },
    {
      "epoch": 1.2872288177214761,
      "grad_norm": 0.31971707940101624,
      "learning_rate": 7.811891407064044e-05,
      "loss": 1.6314,
      "step": 32300
    },
    {
      "epoch": 1.2877707588869884,
      "grad_norm": 0.32109859585762024,
      "learning_rate": 7.810532445921248e-05,
      "loss": 1.6099,
      "step": 32310
    },
    {
      "epoch": 1.2883127000525005,
      "grad_norm": 0.4225994348526001,
      "learning_rate": 7.809173198559724e-05,
      "loss": 1.6177,
      "step": 32320
    },
    {
      "epoch": 1.2888546412180126,
      "grad_norm": 0.2910853326320648,
      "learning_rate": 7.807813665147847e-05,
      "loss": 1.6291,
      "step": 32330
    },
    {
      "epoch": 1.289396582383525,
      "grad_norm": 0.3097500205039978,
      "learning_rate": 7.806453845854036e-05,
      "loss": 1.6157,
      "step": 32340
    },
    {
      "epoch": 1.2899385235490373,
      "grad_norm": 0.2302774041891098,
      "learning_rate": 7.805093740846736e-05,
      "loss": 1.635,
      "step": 32350
    },
    {
      "epoch": 1.2904804647145494,
      "grad_norm": 0.49598991870880127,
      "learning_rate": 7.80373335029443e-05,
      "loss": 1.6259,
      "step": 32360
    },
    {
      "epoch": 1.2906972411807542,
      "eval_loss": 2.5177066326141357,
      "eval_runtime": 21.9888,
      "eval_samples_per_second": 227.389,
      "eval_steps_per_second": 1.228,
      "step": 32364
    },
    {
      "epoch": 1.2910224058800617,
      "grad_norm": 0.32611334323883057,
      "learning_rate": 7.802372674365639e-05,
      "loss": 1.6188,
      "step": 32370
    },
    {
      "epoch": 1.2915643470455738,
      "grad_norm": 0.31275832653045654,
      "learning_rate": 7.801011713228915e-05,
      "loss": 1.6243,
      "step": 32380
    },
    {
      "epoch": 1.292106288211086,
      "grad_norm": 0.31753793358802795,
      "learning_rate": 7.799650467052853e-05,
      "loss": 1.6231,
      "step": 32390
    },
    {
      "epoch": 1.2926482293765984,
      "grad_norm": 0.25154951214790344,
      "learning_rate": 7.798288936006073e-05,
      "loss": 1.6303,
      "step": 32400
    },
    {
      "epoch": 1.2931901705421105,
      "grad_norm": 0.264728844165802,
      "learning_rate": 7.796927120257237e-05,
      "loss": 1.6155,
      "step": 32410
    },
    {
      "epoch": 1.2937321117076228,
      "grad_norm": 0.27522847056388855,
      "learning_rate": 7.795565019975045e-05,
      "loss": 1.6155,
      "step": 32420
    },
    {
      "epoch": 1.294274052873135,
      "grad_norm": 0.3371419906616211,
      "learning_rate": 7.794202635328222e-05,
      "loss": 1.614,
      "step": 32430
    },
    {
      "epoch": 1.2948159940386472,
      "grad_norm": 0.25964683294296265,
      "learning_rate": 7.792839966485537e-05,
      "loss": 1.6086,
      "step": 32440
    },
    {
      "epoch": 1.2953579352041593,
      "grad_norm": 0.22096186876296997,
      "learning_rate": 7.791477013615794e-05,
      "loss": 1.608,
      "step": 32450
    },
    {
      "epoch": 1.2954121293207106,
      "eval_loss": 2.525634288787842,
      "eval_runtime": 21.9845,
      "eval_samples_per_second": 227.433,
      "eval_steps_per_second": 1.228,
      "step": 32451
    },
    {
      "epoch": 1.2958998763696716,
      "grad_norm": 0.4005647301673889,
      "learning_rate": 7.790113776887825e-05,
      "loss": 1.6187,
      "step": 32460
    },
    {
      "epoch": 1.2964418175351837,
      "grad_norm": 0.24895216524600983,
      "learning_rate": 7.788750256470506e-05,
      "loss": 1.6125,
      "step": 32470
    },
    {
      "epoch": 1.296983758700696,
      "grad_norm": 0.2178104668855667,
      "learning_rate": 7.787386452532739e-05,
      "loss": 1.6205,
      "step": 32480
    },
    {
      "epoch": 1.2975256998662084,
      "grad_norm": 0.30151644349098206,
      "learning_rate": 7.78602236524347e-05,
      "loss": 1.6093,
      "step": 32490
    },
    {
      "epoch": 1.2980676410317205,
      "grad_norm": 0.3658519983291626,
      "learning_rate": 7.784657994771676e-05,
      "loss": 1.6158,
      "step": 32500
    },
    {
      "epoch": 1.2986095821972328,
      "grad_norm": 0.40358707308769226,
      "learning_rate": 7.783293341286368e-05,
      "loss": 1.6271,
      "step": 32510
    },
    {
      "epoch": 1.2991515233627449,
      "grad_norm": 0.4737723469734192,
      "learning_rate": 7.781928404956594e-05,
      "loss": 1.6102,
      "step": 32520
    },
    {
      "epoch": 1.2996934645282572,
      "grad_norm": 0.3582918345928192,
      "learning_rate": 7.780563185951437e-05,
      "loss": 1.6199,
      "step": 32530
    },
    {
      "epoch": 1.300127017460667,
      "eval_loss": 2.526118040084839,
      "eval_runtime": 21.9907,
      "eval_samples_per_second": 227.369,
      "eval_steps_per_second": 1.228,
      "step": 32538
    },
    {
      "epoch": 1.3002354056937695,
      "grad_norm": 0.282720148563385,
      "learning_rate": 7.779197684440014e-05,
      "loss": 1.6162,
      "step": 32540
    },
    {
      "epoch": 1.3007773468592816,
      "grad_norm": 0.3246510624885559,
      "learning_rate": 7.777831900591477e-05,
      "loss": 1.6151,
      "step": 32550
    },
    {
      "epoch": 1.3013192880247937,
      "grad_norm": 0.311619371175766,
      "learning_rate": 7.776465834575013e-05,
      "loss": 1.6199,
      "step": 32560
    },
    {
      "epoch": 1.301861229190306,
      "grad_norm": 0.4281388223171234,
      "learning_rate": 7.775099486559845e-05,
      "loss": 1.6132,
      "step": 32570
    },
    {
      "epoch": 1.3024031703558183,
      "grad_norm": 0.39166897535324097,
      "learning_rate": 7.773732856715229e-05,
      "loss": 1.6158,
      "step": 32580
    },
    {
      "epoch": 1.3029451115213304,
      "grad_norm": 0.26505813002586365,
      "learning_rate": 7.772365945210459e-05,
      "loss": 1.6237,
      "step": 32590
    },
    {
      "epoch": 1.3034870526868427,
      "grad_norm": 0.4248243570327759,
      "learning_rate": 7.770998752214863e-05,
      "loss": 1.6218,
      "step": 32600
    },
    {
      "epoch": 1.3040289938523548,
      "grad_norm": 0.21477878093719482,
      "learning_rate": 7.769631277897801e-05,
      "loss": 1.6209,
      "step": 32610
    },
    {
      "epoch": 1.3045709350178671,
      "grad_norm": 0.29866376519203186,
      "learning_rate": 7.768263522428667e-05,
      "loss": 1.6122,
      "step": 32620
    },
    {
      "epoch": 1.3048419056006233,
      "eval_loss": 2.5290870666503906,
      "eval_runtime": 21.9868,
      "eval_samples_per_second": 227.409,
      "eval_steps_per_second": 1.228,
      "step": 32625
    },
    {
      "epoch": 1.3051128761833795,
      "grad_norm": 0.2718803882598877,
      "learning_rate": 7.766895485976899e-05,
      "loss": 1.6174,
      "step": 32630
    },
    {
      "epoch": 1.3056548173488915,
      "grad_norm": 0.22970491647720337,
      "learning_rate": 7.765527168711958e-05,
      "loss": 1.6211,
      "step": 32640
    },
    {
      "epoch": 1.3061967585144036,
      "grad_norm": 0.36656203866004944,
      "learning_rate": 7.764158570803348e-05,
      "loss": 1.6204,
      "step": 32650
    },
    {
      "epoch": 1.306738699679916,
      "grad_norm": 0.31001171469688416,
      "learning_rate": 7.762789692420604e-05,
      "loss": 1.6245,
      "step": 32660
    },
    {
      "epoch": 1.3072806408454283,
      "grad_norm": 0.3693181872367859,
      "learning_rate": 7.761420533733297e-05,
      "loss": 1.616,
      "step": 32670
    },
    {
      "epoch": 1.3078225820109404,
      "grad_norm": 0.3442032039165497,
      "learning_rate": 7.760051094911032e-05,
      "loss": 1.6168,
      "step": 32680
    },
    {
      "epoch": 1.3083645231764527,
      "grad_norm": 0.26000893115997314,
      "learning_rate": 7.75868137612345e-05,
      "loss": 1.6162,
      "step": 32690
    },
    {
      "epoch": 1.3089064643419648,
      "grad_norm": 0.27503514289855957,
      "learning_rate": 7.757311377540226e-05,
      "loss": 1.6104,
      "step": 32700
    },
    {
      "epoch": 1.309448405507477,
      "grad_norm": 0.395526647567749,
      "learning_rate": 7.75594109933107e-05,
      "loss": 1.6291,
      "step": 32710
    },
    {
      "epoch": 1.3095567937405796,
      "eval_loss": 2.5185840129852295,
      "eval_runtime": 21.986,
      "eval_samples_per_second": 227.418,
      "eval_steps_per_second": 1.228,
      "step": 32712
    },
    {
      "epoch": 1.3099903466729894,
      "grad_norm": 0.31528306007385254,
      "learning_rate": 7.75457054166572e-05,
      "loss": 1.6118,
      "step": 32720
    },
    {
      "epoch": 1.3105322878385015,
      "grad_norm": 0.3324233889579773,
      "learning_rate": 7.753199704713963e-05,
      "loss": 1.6176,
      "step": 32730
    },
    {
      "epoch": 1.3110742290040138,
      "grad_norm": 0.2883225679397583,
      "learning_rate": 7.75182858864561e-05,
      "loss": 1.6269,
      "step": 32740
    },
    {
      "epoch": 1.311616170169526,
      "grad_norm": 0.4456532597541809,
      "learning_rate": 7.750457193630507e-05,
      "loss": 1.6139,
      "step": 32750
    },
    {
      "epoch": 1.3121581113350382,
      "grad_norm": 0.26752784848213196,
      "learning_rate": 7.749085519838537e-05,
      "loss": 1.6183,
      "step": 32760
    },
    {
      "epoch": 1.3127000525005503,
      "grad_norm": 0.46080008149147034,
      "learning_rate": 7.747713567439617e-05,
      "loss": 1.6097,
      "step": 32770
    },
    {
      "epoch": 1.3132419936660626,
      "grad_norm": 0.2891246974468231,
      "learning_rate": 7.746341336603698e-05,
      "loss": 1.6118,
      "step": 32780
    },
    {
      "epoch": 1.3137839348315747,
      "grad_norm": 0.4009568989276886,
      "learning_rate": 7.744968827500769e-05,
      "loss": 1.63,
      "step": 32790
    },
    {
      "epoch": 1.3142716818805358,
      "eval_loss": 2.504746913909912,
      "eval_runtime": 21.9893,
      "eval_samples_per_second": 227.384,
      "eval_steps_per_second": 1.228,
      "step": 32799
    },
    {
      "epoch": 1.314325875997087,
      "grad_norm": 0.31230422854423523,
      "learning_rate": 7.743596040300848e-05,
      "loss": 1.6139,
      "step": 32800
    },
    {
      "epoch": 1.3148678171625994,
      "grad_norm": 0.4773191809654236,
      "learning_rate": 7.742222975173991e-05,
      "loss": 1.619,
      "step": 32810
    },
    {
      "epoch": 1.3154097583281115,
      "grad_norm": 0.36002716422080994,
      "learning_rate": 7.740849632290284e-05,
      "loss": 1.6178,
      "step": 32820
    },
    {
      "epoch": 1.3159516994936238,
      "grad_norm": 0.22423742711544037,
      "learning_rate": 7.739476011819854e-05,
      "loss": 1.6269,
      "step": 32830
    },
    {
      "epoch": 1.3164936406591359,
      "grad_norm": 0.43875738978385925,
      "learning_rate": 7.73810211393286e-05,
      "loss": 1.6173,
      "step": 32840
    },
    {
      "epoch": 1.3170355818246482,
      "grad_norm": 0.24483536183834076,
      "learning_rate": 7.736727938799492e-05,
      "loss": 1.6307,
      "step": 32850
    },
    {
      "epoch": 1.3175775229901605,
      "grad_norm": 0.24819006025791168,
      "learning_rate": 7.73535348658998e-05,
      "loss": 1.61,
      "step": 32860
    },
    {
      "epoch": 1.3181194641556726,
      "grad_norm": 0.2164914309978485,
      "learning_rate": 7.733978757474579e-05,
      "loss": 1.6216,
      "step": 32870
    },
    {
      "epoch": 1.3186614053211847,
      "grad_norm": 0.2493513524532318,
      "learning_rate": 7.732603751623591e-05,
      "loss": 1.614,
      "step": 32880
    },
    {
      "epoch": 1.3189865700204921,
      "eval_loss": 2.5233335494995117,
      "eval_runtime": 21.9884,
      "eval_samples_per_second": 227.392,
      "eval_steps_per_second": 1.228,
      "step": 32886
    },
    {
      "epoch": 1.319203346486697,
      "grad_norm": 0.2988925278186798,
      "learning_rate": 7.731228469207342e-05,
      "loss": 1.6109,
      "step": 32890
    },
    {
      "epoch": 1.3197452876522093,
      "grad_norm": 0.3509455621242523,
      "learning_rate": 7.7298529103962e-05,
      "loss": 1.6183,
      "step": 32900
    },
    {
      "epoch": 1.3202872288177214,
      "grad_norm": 0.33325713872909546,
      "learning_rate": 7.728477075360558e-05,
      "loss": 1.6207,
      "step": 32910
    },
    {
      "epoch": 1.3208291699832337,
      "grad_norm": 0.22428090870380402,
      "learning_rate": 7.72710096427085e-05,
      "loss": 1.6048,
      "step": 32920
    },
    {
      "epoch": 1.3213711111487458,
      "grad_norm": 0.4815099835395813,
      "learning_rate": 7.725724577297547e-05,
      "loss": 1.6148,
      "step": 32930
    },
    {
      "epoch": 1.3219130523142582,
      "grad_norm": 0.2657027542591095,
      "learning_rate": 7.724347914611142e-05,
      "loss": 1.6154,
      "step": 32940
    },
    {
      "epoch": 1.3224549934797705,
      "grad_norm": 0.4155304729938507,
      "learning_rate": 7.722970976382179e-05,
      "loss": 1.622,
      "step": 32950
    },
    {
      "epoch": 1.3229969346452826,
      "grad_norm": 0.49975696206092834,
      "learning_rate": 7.721593762781221e-05,
      "loss": 1.6162,
      "step": 32960
    },
    {
      "epoch": 1.3235388758107947,
      "grad_norm": 0.376020222902298,
      "learning_rate": 7.720216273978872e-05,
      "loss": 1.6212,
      "step": 32970
    },
    {
      "epoch": 1.3237014581604485,
      "eval_loss": 2.526895523071289,
      "eval_runtime": 21.9876,
      "eval_samples_per_second": 227.401,
      "eval_steps_per_second": 1.228,
      "step": 32973
    },
    {
      "epoch": 1.324080816976307,
      "grad_norm": 0.22721335291862488,
      "learning_rate": 7.71883851014577e-05,
      "loss": 1.6171,
      "step": 32980
    },
    {
      "epoch": 1.3246227581418193,
      "grad_norm": 0.2434106022119522,
      "learning_rate": 7.717460471452588e-05,
      "loss": 1.6035,
      "step": 32990
    },
    {
      "epoch": 1.3251646993073314,
      "grad_norm": 0.4650270938873291,
      "learning_rate": 7.716082158070031e-05,
      "loss": 1.6103,
      "step": 33000
    },
    {
      "epoch": 1.3257066404728437,
      "grad_norm": 0.2861434817314148,
      "learning_rate": 7.714703570168835e-05,
      "loss": 1.6158,
      "step": 33010
    },
    {
      "epoch": 1.3262485816383558,
      "grad_norm": 0.7163834571838379,
      "learning_rate": 7.713324707919777e-05,
      "loss": 1.6211,
      "step": 33020
    },
    {
      "epoch": 1.326790522803868,
      "grad_norm": 0.685391902923584,
      "learning_rate": 7.711945571493663e-05,
      "loss": 1.624,
      "step": 33030
    },
    {
      "epoch": 1.3273324639693804,
      "grad_norm": 0.33094045519828796,
      "learning_rate": 7.710566161061337e-05,
      "loss": 1.6251,
      "step": 33040
    },
    {
      "epoch": 1.3278744051348925,
      "grad_norm": 0.2844617962837219,
      "learning_rate": 7.70918647679367e-05,
      "loss": 1.6184,
      "step": 33050
    },
    {
      "epoch": 1.3284163463004048,
      "grad_norm": 0.3345138430595398,
      "learning_rate": 7.707806518861575e-05,
      "loss": 1.6111,
      "step": 33060
    },
    {
      "epoch": 1.3284163463004048,
      "eval_loss": 2.527109384536743,
      "eval_runtime": 21.9601,
      "eval_samples_per_second": 227.686,
      "eval_steps_per_second": 1.23,
      "step": 33060
    },
    {
      "epoch": 1.328958287465917,
      "grad_norm": 0.48074135184288025,
      "learning_rate": 7.706426287435991e-05,
      "loss": 1.611,
      "step": 33070
    },
    {
      "epoch": 1.3295002286314292,
      "grad_norm": 0.31359565258026123,
      "learning_rate": 7.7050457826879e-05,
      "loss": 1.6273,
      "step": 33080
    },
    {
      "epoch": 1.3300421697969413,
      "grad_norm": 0.23901283740997314,
      "learning_rate": 7.703665004788312e-05,
      "loss": 1.6191,
      "step": 33090
    },
    {
      "epoch": 1.3305841109624537,
      "grad_norm": 0.5294537544250488,
      "learning_rate": 7.70228395390827e-05,
      "loss": 1.6146,
      "step": 33100
    },
    {
      "epoch": 1.3311260521279658,
      "grad_norm": 0.2701181173324585,
      "learning_rate": 7.700902630218852e-05,
      "loss": 1.6156,
      "step": 33110
    },
    {
      "epoch": 1.331667993293478,
      "grad_norm": 0.25721871852874756,
      "learning_rate": 7.699521033891171e-05,
      "loss": 1.6191,
      "step": 33120
    },
    {
      "epoch": 1.3322099344589904,
      "grad_norm": 0.35411280393600464,
      "learning_rate": 7.698139165096375e-05,
      "loss": 1.6178,
      "step": 33130
    },
    {
      "epoch": 1.3327518756245025,
      "grad_norm": 0.4553796052932739,
      "learning_rate": 7.696757024005642e-05,
      "loss": 1.6116,
      "step": 33140
    },
    {
      "epoch": 1.3331312344403612,
      "eval_loss": 2.523940086364746,
      "eval_runtime": 21.9819,
      "eval_samples_per_second": 227.46,
      "eval_steps_per_second": 1.228,
      "step": 33147
    },
    {
      "epoch": 1.3332938167900148,
      "grad_norm": 0.22905197739601135,
      "learning_rate": 7.695374610790187e-05,
      "loss": 1.6081,
      "step": 33150
    },
    {
      "epoch": 1.333835757955527,
      "grad_norm": 0.3004699945449829,
      "learning_rate": 7.693991925621256e-05,
      "loss": 1.6163,
      "step": 33160
    },
    {
      "epoch": 1.3343776991210392,
      "grad_norm": 0.3088211119174957,
      "learning_rate": 7.69260896867013e-05,
      "loss": 1.6104,
      "step": 33170
    },
    {
      "epoch": 1.3349196402865515,
      "grad_norm": 0.2766474187374115,
      "learning_rate": 7.691225740108126e-05,
      "loss": 1.6134,
      "step": 33180
    },
    {
      "epoch": 1.3354615814520636,
      "grad_norm": 0.4962151050567627,
      "learning_rate": 7.68984224010659e-05,
      "loss": 1.6066,
      "step": 33190
    },
    {
      "epoch": 1.3360035226175757,
      "grad_norm": 0.3378833830356598,
      "learning_rate": 7.688458468836903e-05,
      "loss": 1.6132,
      "step": 33200
    },
    {
      "epoch": 1.336545463783088,
      "grad_norm": 0.34599658846855164,
      "learning_rate": 7.687074426470484e-05,
      "loss": 1.6161,
      "step": 33210
    },
    {
      "epoch": 1.3370874049486003,
      "grad_norm": 0.3025360703468323,
      "learning_rate": 7.68569011317878e-05,
      "loss": 1.6168,
      "step": 33220
    },
    {
      "epoch": 1.3376293461141124,
      "grad_norm": 0.32311034202575684,
      "learning_rate": 7.684305529133273e-05,
      "loss": 1.6177,
      "step": 33230
    },
    {
      "epoch": 1.3378461225803173,
      "eval_loss": 2.5135555267333984,
      "eval_runtime": 21.987,
      "eval_samples_per_second": 227.407,
      "eval_steps_per_second": 1.228,
      "step": 33234
    },
    {
      "epoch": 1.3381712872796248,
      "grad_norm": 0.25993430614471436,
      "learning_rate": 7.682920674505481e-05,
      "loss": 1.612,
      "step": 33240
    },
    {
      "epoch": 1.3387132284451368,
      "grad_norm": 0.35539132356643677,
      "learning_rate": 7.681535549466954e-05,
      "loss": 1.6175,
      "step": 33250
    },
    {
      "epoch": 1.3392551696106492,
      "grad_norm": 0.24199128150939941,
      "learning_rate": 7.680150154189275e-05,
      "loss": 1.6121,
      "step": 33260
    },
    {
      "epoch": 1.3397971107761615,
      "grad_norm": 0.4272856116294861,
      "learning_rate": 7.678764488844059e-05,
      "loss": 1.6128,
      "step": 33270
    },
    {
      "epoch": 1.3403390519416736,
      "grad_norm": 0.23739628493785858,
      "learning_rate": 7.677378553602958e-05,
      "loss": 1.6096,
      "step": 33280
    },
    {
      "epoch": 1.3408809931071857,
      "grad_norm": 0.2996535897254944,
      "learning_rate": 7.675992348637654e-05,
      "loss": 1.6071,
      "step": 33290
    },
    {
      "epoch": 1.341422934272698,
      "grad_norm": 0.38013362884521484,
      "learning_rate": 7.674605874119865e-05,
      "loss": 1.6145,
      "step": 33300
    },
    {
      "epoch": 1.3419648754382103,
      "grad_norm": 0.5391820073127747,
      "learning_rate": 7.673219130221342e-05,
      "loss": 1.6162,
      "step": 33310
    },
    {
      "epoch": 1.3425068166037224,
      "grad_norm": 0.5446195006370544,
      "learning_rate": 7.671832117113868e-05,
      "loss": 1.6039,
      "step": 33320
    },
    {
      "epoch": 1.3425610107202737,
      "eval_loss": 2.505586624145508,
      "eval_runtime": 21.9853,
      "eval_samples_per_second": 227.425,
      "eval_steps_per_second": 1.228,
      "step": 33321
    },
    {
      "epoch": 1.3430487577692347,
      "grad_norm": 0.3738638460636139,
      "learning_rate": 7.670444834969262e-05,
      "loss": 1.6136,
      "step": 33330
    },
    {
      "epoch": 1.3435906989347468,
      "grad_norm": 0.2111603319644928,
      "learning_rate": 7.669057283959371e-05,
      "loss": 1.6126,
      "step": 33340
    },
    {
      "epoch": 1.3441326401002591,
      "grad_norm": 0.27275893092155457,
      "learning_rate": 7.667669464256081e-05,
      "loss": 1.6204,
      "step": 33350
    },
    {
      "epoch": 1.3446745812657714,
      "grad_norm": 0.3274075984954834,
      "learning_rate": 7.66628137603131e-05,
      "loss": 1.6052,
      "step": 33360
    },
    {
      "epoch": 1.3452165224312835,
      "grad_norm": 0.22435316443443298,
      "learning_rate": 7.664893019457007e-05,
      "loss": 1.6097,
      "step": 33370
    },
    {
      "epoch": 1.3457584635967959,
      "grad_norm": 0.41898712515830994,
      "learning_rate": 7.663504394705155e-05,
      "loss": 1.6118,
      "step": 33380
    },
    {
      "epoch": 1.346300404762308,
      "grad_norm": 0.2666551172733307,
      "learning_rate": 7.662115501947772e-05,
      "loss": 1.5982,
      "step": 33390
    },
    {
      "epoch": 1.3468423459278203,
      "grad_norm": 0.4487745463848114,
      "learning_rate": 7.660726341356908e-05,
      "loss": 1.6143,
      "step": 33400
    },
    {
      "epoch": 1.34727589886023,
      "eval_loss": 2.506096363067627,
      "eval_runtime": 21.9887,
      "eval_samples_per_second": 227.39,
      "eval_steps_per_second": 1.228,
      "step": 33408
    },
    {
      "epoch": 1.3473842870933324,
      "grad_norm": 0.21905028820037842,
      "learning_rate": 7.659336913104645e-05,
      "loss": 1.6089,
      "step": 33410
    },
    {
      "epoch": 1.3479262282588447,
      "grad_norm": 0.21832507848739624,
      "learning_rate": 7.657947217363099e-05,
      "loss": 1.6071,
      "step": 33420
    },
    {
      "epoch": 1.3484681694243568,
      "grad_norm": 0.26740700006484985,
      "learning_rate": 7.656557254304423e-05,
      "loss": 1.6158,
      "step": 33430
    },
    {
      "epoch": 1.349010110589869,
      "grad_norm": 0.5225000977516174,
      "learning_rate": 7.655167024100798e-05,
      "loss": 1.6123,
      "step": 33440
    },
    {
      "epoch": 1.3495520517553814,
      "grad_norm": 0.3178088068962097,
      "learning_rate": 7.653776526924435e-05,
      "loss": 1.6182,
      "step": 33450
    },
    {
      "epoch": 1.3500939929208935,
      "grad_norm": 0.21684478223323822,
      "learning_rate": 7.65238576294759e-05,
      "loss": 1.6098,
      "step": 33460
    },
    {
      "epoch": 1.3506359340864058,
      "grad_norm": 0.5009759664535522,
      "learning_rate": 7.650994732342539e-05,
      "loss": 1.6073,
      "step": 33470
    },
    {
      "epoch": 1.351177875251918,
      "grad_norm": 0.5376147031784058,
      "learning_rate": 7.649603435281601e-05,
      "loss": 1.609,
      "step": 33480
    },
    {
      "epoch": 1.3517198164174302,
      "grad_norm": 0.49498310685157776,
      "learning_rate": 7.648211871937121e-05,
      "loss": 1.6146,
      "step": 33490
    },
    {
      "epoch": 1.3519907870001862,
      "eval_loss": 2.51250958442688,
      "eval_runtime": 21.983,
      "eval_samples_per_second": 227.448,
      "eval_steps_per_second": 1.228,
      "step": 33495
    },
    {
      "epoch": 1.3522617575829425,
      "grad_norm": 0.4050725996494293,
      "learning_rate": 7.64682004248148e-05,
      "loss": 1.5954,
      "step": 33500
    },
    {
      "epoch": 1.3528036987484546,
      "grad_norm": 0.2994046211242676,
      "learning_rate": 7.645427947087096e-05,
      "loss": 1.6119,
      "step": 33510
    },
    {
      "epoch": 1.3533456399139667,
      "grad_norm": 0.2522560656070709,
      "learning_rate": 7.64403558592641e-05,
      "loss": 1.613,
      "step": 33520
    },
    {
      "epoch": 1.353887581079479,
      "grad_norm": 0.20905892550945282,
      "learning_rate": 7.642642959171905e-05,
      "loss": 1.5925,
      "step": 33530
    },
    {
      "epoch": 1.3544295222449914,
      "grad_norm": 0.259552925825119,
      "learning_rate": 7.641250066996092e-05,
      "loss": 1.6032,
      "step": 33540
    },
    {
      "epoch": 1.3549714634105035,
      "grad_norm": 0.2907361686229706,
      "learning_rate": 7.639856909571517e-05,
      "loss": 1.6111,
      "step": 33550
    },
    {
      "epoch": 1.3555134045760158,
      "grad_norm": 0.2905162274837494,
      "learning_rate": 7.638463487070759e-05,
      "loss": 1.6076,
      "step": 33560
    },
    {
      "epoch": 1.3560553457415279,
      "grad_norm": 0.3089943826198578,
      "learning_rate": 7.637069799666428e-05,
      "loss": 1.6011,
      "step": 33570
    },
    {
      "epoch": 1.3565972869070402,
      "grad_norm": 0.34541210532188416,
      "learning_rate": 7.635675847531169e-05,
      "loss": 1.6062,
      "step": 33580
    },
    {
      "epoch": 1.3567056751401427,
      "eval_loss": 2.5141220092773438,
      "eval_runtime": 21.9917,
      "eval_samples_per_second": 227.359,
      "eval_steps_per_second": 1.228,
      "step": 33582
    },
    {
      "epoch": 1.3571392280725525,
      "grad_norm": 0.3026711642742157,
      "learning_rate": 7.634281630837656e-05,
      "loss": 1.6096,
      "step": 33590
    },
    {
      "epoch": 1.3576811692380646,
      "grad_norm": 0.37327638268470764,
      "learning_rate": 7.632887149758604e-05,
      "loss": 1.6083,
      "step": 33600
    },
    {
      "epoch": 1.3582231104035767,
      "grad_norm": 0.24328859150409698,
      "learning_rate": 7.63149240446675e-05,
      "loss": 1.6124,
      "step": 33610
    },
    {
      "epoch": 1.358765051569089,
      "grad_norm": 0.26186954975128174,
      "learning_rate": 7.630097395134873e-05,
      "loss": 1.5939,
      "step": 33620
    },
    {
      "epoch": 1.3593069927346013,
      "grad_norm": 0.31656497716903687,
      "learning_rate": 7.628702121935776e-05,
      "loss": 1.6096,
      "step": 33630
    },
    {
      "epoch": 1.3598489339001134,
      "grad_norm": 0.3277546763420105,
      "learning_rate": 7.627306585042302e-05,
      "loss": 1.6049,
      "step": 33640
    },
    {
      "epoch": 1.3603908750656257,
      "grad_norm": 0.25422918796539307,
      "learning_rate": 7.625910784627326e-05,
      "loss": 1.6149,
      "step": 33650
    },
    {
      "epoch": 1.3609328162311378,
      "grad_norm": 0.22749896347522736,
      "learning_rate": 7.62451472086375e-05,
      "loss": 1.6047,
      "step": 33660
    },
    {
      "epoch": 1.3614205632800989,
      "eval_loss": 2.5274205207824707,
      "eval_runtime": 21.9881,
      "eval_samples_per_second": 227.396,
      "eval_steps_per_second": 1.228,
      "step": 33669
    },
    {
      "epoch": 1.3614747573966501,
      "grad_norm": 0.3570140302181244,
      "learning_rate": 7.623118393924515e-05,
      "loss": 1.6078,
      "step": 33670
    },
    {
      "epoch": 1.3620166985621625,
      "grad_norm": 0.583695113658905,
      "learning_rate": 7.62172180398259e-05,
      "loss": 1.6051,
      "step": 33680
    },
    {
      "epoch": 1.3625586397276745,
      "grad_norm": 0.2449539452791214,
      "learning_rate": 7.620324951210981e-05,
      "loss": 1.6079,
      "step": 33690
    },
    {
      "epoch": 1.3631005808931869,
      "grad_norm": 0.33772215247154236,
      "learning_rate": 7.618927835782724e-05,
      "loss": 1.6193,
      "step": 33700
    },
    {
      "epoch": 1.363642522058699,
      "grad_norm": 0.28392404317855835,
      "learning_rate": 7.617530457870883e-05,
      "loss": 1.6043,
      "step": 33710
    },
    {
      "epoch": 1.3641844632242113,
      "grad_norm": 0.34906795620918274,
      "learning_rate": 7.616132817648565e-05,
      "loss": 1.6015,
      "step": 33720
    },
    {
      "epoch": 1.3647264043897234,
      "grad_norm": 0.23826880753040314,
      "learning_rate": 7.614734915288899e-05,
      "loss": 1.6132,
      "step": 33730
    },
    {
      "epoch": 1.3652683455552357,
      "grad_norm": 0.4068436026573181,
      "learning_rate": 7.613336750965053e-05,
      "loss": 1.6078,
      "step": 33740
    },
    {
      "epoch": 1.3658102867207478,
      "grad_norm": 0.45196202397346497,
      "learning_rate": 7.611938324850226e-05,
      "loss": 1.5997,
      "step": 33750
    },
    {
      "epoch": 1.3661354514200552,
      "eval_loss": 2.5162079334259033,
      "eval_runtime": 21.9869,
      "eval_samples_per_second": 227.408,
      "eval_steps_per_second": 1.228,
      "step": 33756
    },
    {
      "epoch": 1.36635222788626,
      "grad_norm": 0.2863083779811859,
      "learning_rate": 7.610539637117647e-05,
      "loss": 1.6013,
      "step": 33760
    },
    {
      "epoch": 1.3668941690517724,
      "grad_norm": 0.3467727303504944,
      "learning_rate": 7.609140687940585e-05,
      "loss": 1.6113,
      "step": 33770
    },
    {
      "epoch": 1.3674361102172845,
      "grad_norm": 0.8201264142990112,
      "learning_rate": 7.607741477492327e-05,
      "loss": 1.6092,
      "step": 33780
    },
    {
      "epoch": 1.3679780513827968,
      "grad_norm": 0.3005310893058777,
      "learning_rate": 7.606342005946207e-05,
      "loss": 1.6119,
      "step": 33790
    },
    {
      "epoch": 1.368519992548309,
      "grad_norm": 0.31185653805732727,
      "learning_rate": 7.604942273475585e-05,
      "loss": 1.6117,
      "step": 33800
    },
    {
      "epoch": 1.3690619337138212,
      "grad_norm": 0.4523196816444397,
      "learning_rate": 7.603542280253853e-05,
      "loss": 1.6103,
      "step": 33810
    },
    {
      "epoch": 1.3696038748793335,
      "grad_norm": 0.2638733386993408,
      "learning_rate": 7.602142026454435e-05,
      "loss": 1.6264,
      "step": 33820
    },
    {
      "epoch": 1.3701458160448456,
      "grad_norm": 0.25433018803596497,
      "learning_rate": 7.60074151225079e-05,
      "loss": 1.6129,
      "step": 33830
    },
    {
      "epoch": 1.3706877572103577,
      "grad_norm": 0.22881822288036346,
      "learning_rate": 7.599340737816406e-05,
      "loss": 1.5969,
      "step": 33840
    },
    {
      "epoch": 1.3708503395600116,
      "eval_loss": 2.5215718746185303,
      "eval_runtime": 21.9866,
      "eval_samples_per_second": 227.411,
      "eval_steps_per_second": 1.228,
      "step": 33843
    },
    {
      "epoch": 1.37122969837587,
      "grad_norm": 0.47086840867996216,
      "learning_rate": 7.597939703324807e-05,
      "loss": 1.6004,
      "step": 33850
    },
    {
      "epoch": 1.3717716395413824,
      "grad_norm": 0.25351575016975403,
      "learning_rate": 7.596538408949546e-05,
      "loss": 1.6025,
      "step": 33860
    },
    {
      "epoch": 1.3723135807068945,
      "grad_norm": 0.47730979323387146,
      "learning_rate": 7.595136854864208e-05,
      "loss": 1.6075,
      "step": 33870
    },
    {
      "epoch": 1.3728555218724068,
      "grad_norm": 0.22560712695121765,
      "learning_rate": 7.593735041242414e-05,
      "loss": 1.6045,
      "step": 33880
    },
    {
      "epoch": 1.3733974630379189,
      "grad_norm": 0.3164345920085907,
      "learning_rate": 7.592332968257812e-05,
      "loss": 1.612,
      "step": 33890
    },
    {
      "epoch": 1.3739394042034312,
      "grad_norm": 0.5792235136032104,
      "learning_rate": 7.590930636084087e-05,
      "loss": 1.6049,
      "step": 33900
    },
    {
      "epoch": 1.3744813453689435,
      "grad_norm": 0.2980400025844574,
      "learning_rate": 7.589528044894951e-05,
      "loss": 1.6049,
      "step": 33910
    },
    {
      "epoch": 1.3750232865344556,
      "grad_norm": 0.2572561800479889,
      "learning_rate": 7.588125194864154e-05,
      "loss": 1.5972,
      "step": 33920
    },
    {
      "epoch": 1.3755652276999677,
      "grad_norm": 0.3515624701976776,
      "learning_rate": 7.586722086165471e-05,
      "loss": 1.6086,
      "step": 33930
    },
    {
      "epoch": 1.3755652276999677,
      "eval_loss": 2.5238847732543945,
      "eval_runtime": 21.9722,
      "eval_samples_per_second": 227.56,
      "eval_steps_per_second": 1.229,
      "step": 33930
    },
    {
      "epoch": 1.37610716886548,
      "grad_norm": 0.22956761717796326,
      "learning_rate": 7.585318718972719e-05,
      "loss": 1.5948,
      "step": 33940
    },
    {
      "epoch": 1.3766491100309923,
      "grad_norm": 0.2527923583984375,
      "learning_rate": 7.583915093459736e-05,
      "loss": 1.5908,
      "step": 33950
    },
    {
      "epoch": 1.3771910511965044,
      "grad_norm": 0.3912854790687561,
      "learning_rate": 7.5825112098004e-05,
      "loss": 1.5966,
      "step": 33960
    },
    {
      "epoch": 1.3777329923620167,
      "grad_norm": 0.2859535813331604,
      "learning_rate": 7.581107068168615e-05,
      "loss": 1.6188,
      "step": 33970
    },
    {
      "epoch": 1.3782749335275288,
      "grad_norm": 0.29767870903015137,
      "learning_rate": 7.579702668738323e-05,
      "loss": 1.604,
      "step": 33980
    },
    {
      "epoch": 1.3788168746930412,
      "grad_norm": 0.3692917227745056,
      "learning_rate": 7.578298011683493e-05,
      "loss": 1.6166,
      "step": 33990
    },
    {
      "epoch": 1.3793588158585535,
      "grad_norm": 0.40221714973449707,
      "learning_rate": 7.576893097178128e-05,
      "loss": 1.6078,
      "step": 34000
    },
    {
      "epoch": 1.3799007570240656,
      "grad_norm": 0.4148884415626526,
      "learning_rate": 7.575487925396264e-05,
      "loss": 1.6028,
      "step": 34010
    },
    {
      "epoch": 1.380280115839924,
      "eval_loss": 2.53521466255188,
      "eval_runtime": 21.9888,
      "eval_samples_per_second": 227.389,
      "eval_steps_per_second": 1.228,
      "step": 34017
    },
    {
      "epoch": 1.3804426981895779,
      "grad_norm": 0.5252642035484314,
      "learning_rate": 7.574082496511966e-05,
      "loss": 1.6146,
      "step": 34020
    },
    {
      "epoch": 1.38098463935509,
      "grad_norm": 0.4250527322292328,
      "learning_rate": 7.572676810699333e-05,
      "loss": 1.6074,
      "step": 34030
    },
    {
      "epoch": 1.3815265805206023,
      "grad_norm": 0.31851446628570557,
      "learning_rate": 7.571270868132496e-05,
      "loss": 1.5978,
      "step": 34040
    },
    {
      "epoch": 1.3820685216861144,
      "grad_norm": 0.22338920831680298,
      "learning_rate": 7.569864668985617e-05,
      "loss": 1.5961,
      "step": 34050
    },
    {
      "epoch": 1.3826104628516267,
      "grad_norm": 0.2865997552871704,
      "learning_rate": 7.568458213432888e-05,
      "loss": 1.6073,
      "step": 34060
    },
    {
      "epoch": 1.3831524040171388,
      "grad_norm": 0.3791174590587616,
      "learning_rate": 7.567051501648536e-05,
      "loss": 1.6018,
      "step": 34070
    },
    {
      "epoch": 1.383694345182651,
      "grad_norm": 0.3320346474647522,
      "learning_rate": 7.565644533806818e-05,
      "loss": 1.5995,
      "step": 34080
    },
    {
      "epoch": 1.3842362863481634,
      "grad_norm": 0.2221497744321823,
      "learning_rate": 7.564237310082024e-05,
      "loss": 1.5972,
      "step": 34090
    },
    {
      "epoch": 1.3847782275136755,
      "grad_norm": 0.27021268010139465,
      "learning_rate": 7.562829830648474e-05,
      "loss": 1.6121,
      "step": 34100
    },
    {
      "epoch": 1.3849950039798804,
      "eval_loss": 2.520078420639038,
      "eval_runtime": 21.9871,
      "eval_samples_per_second": 227.407,
      "eval_steps_per_second": 1.228,
      "step": 34104
    },
    {
      "epoch": 1.3853201686791878,
      "grad_norm": 0.35636231303215027,
      "learning_rate": 7.56142209568052e-05,
      "loss": 1.6015,
      "step": 34110
    },
    {
      "epoch": 1.3858621098447,
      "grad_norm": 0.45814478397369385,
      "learning_rate": 7.560014105352546e-05,
      "loss": 1.6131,
      "step": 34120
    },
    {
      "epoch": 1.3864040510102122,
      "grad_norm": 0.6515801548957825,
      "learning_rate": 7.55860585983897e-05,
      "loss": 1.6189,
      "step": 34130
    },
    {
      "epoch": 1.3869459921757243,
      "grad_norm": 0.32424890995025635,
      "learning_rate": 7.557197359314237e-05,
      "loss": 1.6162,
      "step": 34140
    },
    {
      "epoch": 1.3874879333412367,
      "grad_norm": 0.24668356776237488,
      "learning_rate": 7.555788603952825e-05,
      "loss": 1.607,
      "step": 34150
    },
    {
      "epoch": 1.3880298745067488,
      "grad_norm": 0.2475864738225937,
      "learning_rate": 7.554379593929248e-05,
      "loss": 1.6014,
      "step": 34160
    },
    {
      "epoch": 1.388571815672261,
      "grad_norm": 0.2331823706626892,
      "learning_rate": 7.552970329418045e-05,
      "loss": 1.6068,
      "step": 34170
    },
    {
      "epoch": 1.3891137568377734,
      "grad_norm": 0.2390824258327484,
      "learning_rate": 7.551560810593792e-05,
      "loss": 1.5906,
      "step": 34180
    },
    {
      "epoch": 1.3896556980032855,
      "grad_norm": 0.26939457654953003,
      "learning_rate": 7.550151037631092e-05,
      "loss": 1.6275,
      "step": 34190
    },
    {
      "epoch": 1.3897098921198368,
      "eval_loss": 2.518341064453125,
      "eval_runtime": 21.9824,
      "eval_samples_per_second": 227.455,
      "eval_steps_per_second": 1.228,
      "step": 34191
    },
    {
      "epoch": 1.3901976391687978,
      "grad_norm": 0.262103796005249,
      "learning_rate": 7.548741010704583e-05,
      "loss": 1.5979,
      "step": 34200
    },
    {
      "epoch": 1.3907395803343099,
      "grad_norm": 0.34173309803009033,
      "learning_rate": 7.547330729988931e-05,
      "loss": 1.5959,
      "step": 34210
    },
    {
      "epoch": 1.3912815214998222,
      "grad_norm": 0.22995412349700928,
      "learning_rate": 7.545920195658837e-05,
      "loss": 1.605,
      "step": 34220
    },
    {
      "epoch": 1.3918234626653345,
      "grad_norm": 0.2825547158718109,
      "learning_rate": 7.544509407889033e-05,
      "loss": 1.5997,
      "step": 34230
    },
    {
      "epoch": 1.3923654038308466,
      "grad_norm": 0.3113856613636017,
      "learning_rate": 7.54309836685428e-05,
      "loss": 1.6061,
      "step": 34240
    },
    {
      "epoch": 1.3929073449963587,
      "grad_norm": 0.32774263620376587,
      "learning_rate": 7.54168707272937e-05,
      "loss": 1.5979,
      "step": 34250
    },
    {
      "epoch": 1.393449286161871,
      "grad_norm": 0.3357396721839905,
      "learning_rate": 7.540275525689131e-05,
      "loss": 1.6033,
      "step": 34260
    },
    {
      "epoch": 1.3939912273273833,
      "grad_norm": 0.35256245732307434,
      "learning_rate": 7.538863725908416e-05,
      "loss": 1.6,
      "step": 34270
    },
    {
      "epoch": 1.394424780259793,
      "eval_loss": 2.519773244857788,
      "eval_runtime": 21.9811,
      "eval_samples_per_second": 227.468,
      "eval_steps_per_second": 1.228,
      "step": 34278
    },
    {
      "epoch": 1.3945331684928954,
      "grad_norm": 0.31579533219337463,
      "learning_rate": 7.537451673562116e-05,
      "loss": 1.5972,
      "step": 34280
    },
    {
      "epoch": 1.3950751096584078,
      "grad_norm": 0.3092842996120453,
      "learning_rate": 7.536039368825147e-05,
      "loss": 1.5975,
      "step": 34290
    },
    {
      "epoch": 1.3956170508239198,
      "grad_norm": 0.4527590572834015,
      "learning_rate": 7.534626811872463e-05,
      "loss": 1.6089,
      "step": 34300
    },
    {
      "epoch": 1.3961589919894322,
      "grad_norm": 0.30094006657600403,
      "learning_rate": 7.53321400287904e-05,
      "loss": 1.6083,
      "step": 34310
    },
    {
      "epoch": 1.3967009331549445,
      "grad_norm": 0.3692034184932709,
      "learning_rate": 7.531800942019895e-05,
      "loss": 1.6021,
      "step": 34320
    },
    {
      "epoch": 1.3972428743204566,
      "grad_norm": 0.37879860401153564,
      "learning_rate": 7.530387629470072e-05,
      "loss": 1.6015,
      "step": 34330
    },
    {
      "epoch": 1.3977848154859687,
      "grad_norm": 0.5758041739463806,
      "learning_rate": 7.528974065404644e-05,
      "loss": 1.598,
      "step": 34340
    },
    {
      "epoch": 1.398326756651481,
      "grad_norm": 0.29847633838653564,
      "learning_rate": 7.527560249998716e-05,
      "loss": 1.6055,
      "step": 34350
    },
    {
      "epoch": 1.3988686978169933,
      "grad_norm": 0.22990530729293823,
      "learning_rate": 7.526146183427428e-05,
      "loss": 1.5928,
      "step": 34360
    },
    {
      "epoch": 1.3991396683997492,
      "eval_loss": 2.5169806480407715,
      "eval_runtime": 21.9875,
      "eval_samples_per_second": 227.402,
      "eval_steps_per_second": 1.228,
      "step": 34365
    },
    {
      "epoch": 1.3994106389825054,
      "grad_norm": 0.20360521972179413,
      "learning_rate": 7.524731865865947e-05,
      "loss": 1.5949,
      "step": 34370
    },
    {
      "epoch": 1.3999525801480177,
      "grad_norm": 0.24859194457530975,
      "learning_rate": 7.523317297489473e-05,
      "loss": 1.6024,
      "step": 34380
    },
    {
      "epoch": 1.4004945213135298,
      "grad_norm": 0.33308878540992737,
      "learning_rate": 7.521902478473238e-05,
      "loss": 1.598,
      "step": 34390
    },
    {
      "epoch": 1.4010364624790421,
      "grad_norm": 0.27835598587989807,
      "learning_rate": 7.520487408992501e-05,
      "loss": 1.5965,
      "step": 34400
    },
    {
      "epoch": 1.4015784036445544,
      "grad_norm": 0.36223334074020386,
      "learning_rate": 7.519072089222557e-05,
      "loss": 1.6055,
      "step": 34410
    },
    {
      "epoch": 1.4021203448100665,
      "grad_norm": 0.3996673822402954,
      "learning_rate": 7.517656519338728e-05,
      "loss": 1.6096,
      "step": 34420
    },
    {
      "epoch": 1.4026622859755788,
      "grad_norm": 0.25303345918655396,
      "learning_rate": 7.516240699516367e-05,
      "loss": 1.614,
      "step": 34430
    },
    {
      "epoch": 1.403204227141091,
      "grad_norm": 0.3061133325099945,
      "learning_rate": 7.514824629930865e-05,
      "loss": 1.6032,
      "step": 34440
    },
    {
      "epoch": 1.4037461683066033,
      "grad_norm": 0.3153916895389557,
      "learning_rate": 7.513408310757632e-05,
      "loss": 1.5952,
      "step": 34450
    },
    {
      "epoch": 1.4038545565397056,
      "eval_loss": 2.5186572074890137,
      "eval_runtime": 21.9793,
      "eval_samples_per_second": 227.487,
      "eval_steps_per_second": 1.228,
      "step": 34452
    },
    {
      "epoch": 1.4042881094721154,
      "grad_norm": 0.24946312606334686,
      "learning_rate": 7.51199174217212e-05,
      "loss": 1.6027,
      "step": 34460
    },
    {
      "epoch": 1.4048300506376277,
      "grad_norm": 0.23080390691757202,
      "learning_rate": 7.510574924349807e-05,
      "loss": 1.604,
      "step": 34470
    },
    {
      "epoch": 1.4053719918031398,
      "grad_norm": 0.36294880509376526,
      "learning_rate": 7.509157857466202e-05,
      "loss": 1.608,
      "step": 34480
    },
    {
      "epoch": 1.405913932968652,
      "grad_norm": 0.3913600742816925,
      "learning_rate": 7.507740541696844e-05,
      "loss": 1.6031,
      "step": 34490
    },
    {
      "epoch": 1.4064558741341644,
      "grad_norm": 0.36812445521354675,
      "learning_rate": 7.506322977217305e-05,
      "loss": 1.596,
      "step": 34500
    },
    {
      "epoch": 1.4069978152996765,
      "grad_norm": 0.5203189253807068,
      "learning_rate": 7.504905164203184e-05,
      "loss": 1.6107,
      "step": 34510
    },
    {
      "epoch": 1.4075397564651888,
      "grad_norm": 0.2861349880695343,
      "learning_rate": 7.503487102830116e-05,
      "loss": 1.6116,
      "step": 34520
    },
    {
      "epoch": 1.408081697630701,
      "grad_norm": 0.3074442446231842,
      "learning_rate": 7.502068793273765e-05,
      "loss": 1.602,
      "step": 34530
    },
    {
      "epoch": 1.408569444679662,
      "eval_loss": 2.5188817977905273,
      "eval_runtime": 21.988,
      "eval_samples_per_second": 227.396,
      "eval_steps_per_second": 1.228,
      "step": 34539
    },
    {
      "epoch": 1.4086236387962132,
      "grad_norm": 0.3228742778301239,
      "learning_rate": 7.500650235709826e-05,
      "loss": 1.5973,
      "step": 34540
    },
    {
      "epoch": 1.4091655799617255,
      "grad_norm": 0.3475671708583832,
      "learning_rate": 7.49923143031402e-05,
      "loss": 1.5987,
      "step": 34550
    },
    {
      "epoch": 1.4097075211272376,
      "grad_norm": 0.3168829083442688,
      "learning_rate": 7.497812377262102e-05,
      "loss": 1.6001,
      "step": 34560
    },
    {
      "epoch": 1.4102494622927497,
      "grad_norm": 0.2580108046531677,
      "learning_rate": 7.496393076729862e-05,
      "loss": 1.6122,
      "step": 34570
    },
    {
      "epoch": 1.410791403458262,
      "grad_norm": 0.28347161412239075,
      "learning_rate": 7.494973528893117e-05,
      "loss": 1.609,
      "step": 34580
    },
    {
      "epoch": 1.4113333446237744,
      "grad_norm": 0.25245901942253113,
      "learning_rate": 7.49355373392771e-05,
      "loss": 1.6036,
      "step": 34590
    },
    {
      "epoch": 1.4118752857892864,
      "grad_norm": 0.29171431064605713,
      "learning_rate": 7.492133692009524e-05,
      "loss": 1.5972,
      "step": 34600
    },
    {
      "epoch": 1.4124172269547988,
      "grad_norm": 0.3764426112174988,
      "learning_rate": 7.490713403314462e-05,
      "loss": 1.6124,
      "step": 34610
    },
    {
      "epoch": 1.4129591681203109,
      "grad_norm": 0.2615674138069153,
      "learning_rate": 7.489292868018469e-05,
      "loss": 1.6089,
      "step": 34620
    },
    {
      "epoch": 1.4132843328196183,
      "eval_loss": 2.516463279724121,
      "eval_runtime": 21.9878,
      "eval_samples_per_second": 227.399,
      "eval_steps_per_second": 1.228,
      "step": 34626
    },
    {
      "epoch": 1.4135011092858232,
      "grad_norm": 0.2315002828836441,
      "learning_rate": 7.487872086297513e-05,
      "loss": 1.6043,
      "step": 34630
    },
    {
      "epoch": 1.4140430504513355,
      "grad_norm": 0.25207579135894775,
      "learning_rate": 7.48645105832759e-05,
      "loss": 1.5948,
      "step": 34640
    },
    {
      "epoch": 1.4145849916168476,
      "grad_norm": 0.29729408025741577,
      "learning_rate": 7.485029784284737e-05,
      "loss": 1.6036,
      "step": 34650
    },
    {
      "epoch": 1.4151269327823597,
      "grad_norm": 0.2224244922399521,
      "learning_rate": 7.483608264345011e-05,
      "loss": 1.592,
      "step": 34660
    },
    {
      "epoch": 1.415668873947872,
      "grad_norm": 0.3714025616645813,
      "learning_rate": 7.482186498684504e-05,
      "loss": 1.5983,
      "step": 34670
    },
    {
      "epoch": 1.4162108151133843,
      "grad_norm": 0.29220449924468994,
      "learning_rate": 7.480764487479342e-05,
      "loss": 1.5968,
      "step": 34680
    },
    {
      "epoch": 1.4167527562788964,
      "grad_norm": 0.2586628198623657,
      "learning_rate": 7.479342230905674e-05,
      "loss": 1.5986,
      "step": 34690
    },
    {
      "epoch": 1.4172946974444087,
      "grad_norm": 0.2448311746120453,
      "learning_rate": 7.477919729139684e-05,
      "loss": 1.5924,
      "step": 34700
    },
    {
      "epoch": 1.4178366386099208,
      "grad_norm": 0.43788665533065796,
      "learning_rate": 7.476496982357585e-05,
      "loss": 1.6003,
      "step": 34710
    },
    {
      "epoch": 1.4179992209595746,
      "eval_loss": 2.5183908939361572,
      "eval_runtime": 21.9844,
      "eval_samples_per_second": 227.434,
      "eval_steps_per_second": 1.228,
      "step": 34713
    },
    {
      "epoch": 1.4183785797754331,
      "grad_norm": 0.3235478103160858,
      "learning_rate": 7.47507399073562e-05,
      "loss": 1.6102,
      "step": 34720
    },
    {
      "epoch": 1.4189205209409455,
      "grad_norm": 0.2464754730463028,
      "learning_rate": 7.473650754450066e-05,
      "loss": 1.5979,
      "step": 34730
    },
    {
      "epoch": 1.4194624621064575,
      "grad_norm": 0.4333379566669464,
      "learning_rate": 7.472227273677225e-05,
      "loss": 1.6088,
      "step": 34740
    },
    {
      "epoch": 1.4200044032719699,
      "grad_norm": 0.2818497121334076,
      "learning_rate": 7.47080354859343e-05,
      "loss": 1.6013,
      "step": 34750
    },
    {
      "epoch": 1.420546344437482,
      "grad_norm": 0.2427944540977478,
      "learning_rate": 7.469379579375049e-05,
      "loss": 1.5914,
      "step": 34760
    },
    {
      "epoch": 1.4210882856029943,
      "grad_norm": 0.29250580072402954,
      "learning_rate": 7.467955366198474e-05,
      "loss": 1.6074,
      "step": 34770
    },
    {
      "epoch": 1.4216302267685064,
      "grad_norm": 0.26771166920661926,
      "learning_rate": 7.466530909240135e-05,
      "loss": 1.6108,
      "step": 34780
    },
    {
      "epoch": 1.4221721679340187,
      "grad_norm": 0.29733744263648987,
      "learning_rate": 7.465106208676485e-05,
      "loss": 1.5965,
      "step": 34790
    },
    {
      "epoch": 1.4227141090995308,
      "grad_norm": 0.2804502546787262,
      "learning_rate": 7.463681264684007e-05,
      "loss": 1.5984,
      "step": 34800
    },
    {
      "epoch": 1.4227141090995308,
      "eval_loss": 2.5124473571777344,
      "eval_runtime": 21.9639,
      "eval_samples_per_second": 227.646,
      "eval_steps_per_second": 1.229,
      "step": 34800
    },
    {
      "epoch": 1.423256050265043,
      "grad_norm": 0.46871018409729004,
      "learning_rate": 7.46225607743922e-05,
      "loss": 1.5917,
      "step": 34810
    },
    {
      "epoch": 1.4237979914305554,
      "grad_norm": 0.26123303174972534,
      "learning_rate": 7.46083064711867e-05,
      "loss": 1.5963,
      "step": 34820
    },
    {
      "epoch": 1.4243399325960675,
      "grad_norm": 0.22404921054840088,
      "learning_rate": 7.459404973898932e-05,
      "loss": 1.6005,
      "step": 34830
    },
    {
      "epoch": 1.4248818737615798,
      "grad_norm": 0.26554566621780396,
      "learning_rate": 7.457979057956615e-05,
      "loss": 1.6021,
      "step": 34840
    },
    {
      "epoch": 1.425423814927092,
      "grad_norm": 0.2421078085899353,
      "learning_rate": 7.45655289946835e-05,
      "loss": 1.6013,
      "step": 34850
    },
    {
      "epoch": 1.4259657560926042,
      "grad_norm": 0.22835715115070343,
      "learning_rate": 7.455126498610807e-05,
      "loss": 1.5989,
      "step": 34860
    },
    {
      "epoch": 1.4265076972581165,
      "grad_norm": 0.37547022104263306,
      "learning_rate": 7.453699855560683e-05,
      "loss": 1.6129,
      "step": 34870
    },
    {
      "epoch": 1.4270496384236286,
      "grad_norm": 0.4508250057697296,
      "learning_rate": 7.452272970494702e-05,
      "loss": 1.5987,
      "step": 34880
    },
    {
      "epoch": 1.4274289972394871,
      "eval_loss": 2.5165185928344727,
      "eval_runtime": 22.3018,
      "eval_samples_per_second": 224.198,
      "eval_steps_per_second": 1.211,
      "step": 34887
    },
    {
      "epoch": 1.4275915795891407,
      "grad_norm": 0.28928378224372864,
      "learning_rate": 7.450845843589622e-05,
      "loss": 1.5965,
      "step": 34890
    },
    {
      "epoch": 1.428133520754653,
      "grad_norm": 0.24763907492160797,
      "learning_rate": 7.449418475022228e-05,
      "loss": 1.5987,
      "step": 34900
    },
    {
      "epoch": 1.4286754619201654,
      "grad_norm": 0.28781482577323914,
      "learning_rate": 7.447990864969336e-05,
      "loss": 1.59,
      "step": 34910
    },
    {
      "epoch": 1.4292174030856775,
      "grad_norm": 0.4051133990287781,
      "learning_rate": 7.446563013607795e-05,
      "loss": 1.5962,
      "step": 34920
    },
    {
      "epoch": 1.4297593442511898,
      "grad_norm": 0.472084641456604,
      "learning_rate": 7.445134921114477e-05,
      "loss": 1.5906,
      "step": 34930
    },
    {
      "epoch": 1.4303012854167019,
      "grad_norm": 0.495980441570282,
      "learning_rate": 7.443706587666292e-05,
      "loss": 1.6111,
      "step": 34940
    },
    {
      "epoch": 1.4308432265822142,
      "grad_norm": 0.41576093435287476,
      "learning_rate": 7.442278013440173e-05,
      "loss": 1.5951,
      "step": 34950
    },
    {
      "epoch": 1.4313851677477265,
      "grad_norm": 0.48290589451789856,
      "learning_rate": 7.440849198613084e-05,
      "loss": 1.5979,
      "step": 34960
    },
    {
      "epoch": 1.4319271089132386,
      "grad_norm": 0.3737516701221466,
      "learning_rate": 7.439420143362025e-05,
      "loss": 1.6024,
      "step": 34970
    },
    {
      "epoch": 1.4321438853794435,
      "eval_loss": 2.513838768005371,
      "eval_runtime": 21.9922,
      "eval_samples_per_second": 227.353,
      "eval_steps_per_second": 1.228,
      "step": 34974
    },
    {
      "epoch": 1.4324690500787507,
      "grad_norm": 0.22067669034004211,
      "learning_rate": 7.437990847864018e-05,
      "loss": 1.6073,
      "step": 34980
    },
    {
      "epoch": 1.433010991244263,
      "grad_norm": 0.35069945454597473,
      "learning_rate": 7.436561312296118e-05,
      "loss": 1.5947,
      "step": 34990
    },
    {
      "epoch": 1.4335529324097753,
      "grad_norm": 0.3185942471027374,
      "learning_rate": 7.435131536835412e-05,
      "loss": 1.5965,
      "step": 35000
    },
    {
      "epoch": 1.4340948735752874,
      "grad_norm": 0.3867363929748535,
      "learning_rate": 7.433701521659012e-05,
      "loss": 1.6054,
      "step": 35010
    },
    {
      "epoch": 1.4346368147407997,
      "grad_norm": 0.2255895733833313,
      "learning_rate": 7.432271266944063e-05,
      "loss": 1.5981,
      "step": 35020
    },
    {
      "epoch": 1.4351787559063118,
      "grad_norm": 0.5799124836921692,
      "learning_rate": 7.43084077286774e-05,
      "loss": 1.5975,
      "step": 35030
    },
    {
      "epoch": 1.4357206970718241,
      "grad_norm": 0.32704293727874756,
      "learning_rate": 7.429410039607241e-05,
      "loss": 1.598,
      "step": 35040
    },
    {
      "epoch": 1.4362626382373365,
      "grad_norm": 0.2843274772167206,
      "learning_rate": 7.427979067339805e-05,
      "loss": 1.6012,
      "step": 35050
    },
    {
      "epoch": 1.4368045794028486,
      "grad_norm": 0.21767373383045197,
      "learning_rate": 7.426547856242692e-05,
      "loss": 1.5984,
      "step": 35060
    },
    {
      "epoch": 1.4368587735193998,
      "eval_loss": 2.5202910900115967,
      "eval_runtime": 21.9854,
      "eval_samples_per_second": 227.423,
      "eval_steps_per_second": 1.228,
      "step": 35061
    },
    {
      "epoch": 1.4373465205683609,
      "grad_norm": 0.3761364221572876,
      "learning_rate": 7.425116406493196e-05,
      "loss": 1.6041,
      "step": 35070
    },
    {
      "epoch": 1.437888461733873,
      "grad_norm": 0.3286113142967224,
      "learning_rate": 7.423684718268637e-05,
      "loss": 1.5982,
      "step": 35080
    },
    {
      "epoch": 1.4384304028993853,
      "grad_norm": 0.3064782917499542,
      "learning_rate": 7.422252791746369e-05,
      "loss": 1.5936,
      "step": 35090
    },
    {
      "epoch": 1.4389723440648974,
      "grad_norm": 0.2235429883003235,
      "learning_rate": 7.420820627103768e-05,
      "loss": 1.5973,
      "step": 35100
    },
    {
      "epoch": 1.4395142852304097,
      "grad_norm": 0.33427155017852783,
      "learning_rate": 7.419388224518247e-05,
      "loss": 1.5904,
      "step": 35110
    },
    {
      "epoch": 1.4400562263959218,
      "grad_norm": 0.26824381947517395,
      "learning_rate": 7.417955584167246e-05,
      "loss": 1.6038,
      "step": 35120
    },
    {
      "epoch": 1.440598167561434,
      "grad_norm": 0.3559771776199341,
      "learning_rate": 7.416522706228235e-05,
      "loss": 1.5985,
      "step": 35130
    },
    {
      "epoch": 1.4411401087269464,
      "grad_norm": 0.43836888670921326,
      "learning_rate": 7.415089590878713e-05,
      "loss": 1.6008,
      "step": 35140
    },
    {
      "epoch": 1.4415736616593562,
      "eval_loss": 2.522282838821411,
      "eval_runtime": 21.9813,
      "eval_samples_per_second": 227.466,
      "eval_steps_per_second": 1.228,
      "step": 35148
    },
    {
      "epoch": 1.4416820498924585,
      "grad_norm": 0.2667721211910248,
      "learning_rate": 7.413656238296205e-05,
      "loss": 1.5959,
      "step": 35150
    },
    {
      "epoch": 1.4422239910579708,
      "grad_norm": 0.34478846192359924,
      "learning_rate": 7.412222648658271e-05,
      "loss": 1.5947,
      "step": 35160
    },
    {
      "epoch": 1.442765932223483,
      "grad_norm": 0.2894307076931,
      "learning_rate": 7.410788822142497e-05,
      "loss": 1.5881,
      "step": 35170
    },
    {
      "epoch": 1.4433078733889952,
      "grad_norm": 0.44409897923469543,
      "learning_rate": 7.409354758926501e-05,
      "loss": 1.5902,
      "step": 35180
    },
    {
      "epoch": 1.4438498145545076,
      "grad_norm": 0.3519600033760071,
      "learning_rate": 7.407920459187925e-05,
      "loss": 1.6024,
      "step": 35190
    },
    {
      "epoch": 1.4443917557200197,
      "grad_norm": 0.2908567786216736,
      "learning_rate": 7.406485923104447e-05,
      "loss": 1.5973,
      "step": 35200
    },
    {
      "epoch": 1.4449336968855317,
      "grad_norm": 0.33565568923950195,
      "learning_rate": 7.405051150853771e-05,
      "loss": 1.5992,
      "step": 35210
    },
    {
      "epoch": 1.445475638051044,
      "grad_norm": 0.3194481134414673,
      "learning_rate": 7.403616142613627e-05,
      "loss": 1.5931,
      "step": 35220
    },
    {
      "epoch": 1.4460175792165564,
      "grad_norm": 0.25139880180358887,
      "learning_rate": 7.402180898561781e-05,
      "loss": 1.599,
      "step": 35230
    },
    {
      "epoch": 1.4462885497993123,
      "eval_loss": 2.5243048667907715,
      "eval_runtime": 21.9879,
      "eval_samples_per_second": 227.398,
      "eval_steps_per_second": 1.228,
      "step": 35235
    },
    {
      "epoch": 1.4465595203820685,
      "grad_norm": 0.4910086989402771,
      "learning_rate": 7.400745418876024e-05,
      "loss": 1.5972,
      "step": 35240
    },
    {
      "epoch": 1.4471014615475808,
      "grad_norm": 0.2239486426115036,
      "learning_rate": 7.399309703734178e-05,
      "loss": 1.5989,
      "step": 35250
    },
    {
      "epoch": 1.4476434027130929,
      "grad_norm": 0.3110259175300598,
      "learning_rate": 7.39787375331409e-05,
      "loss": 1.6041,
      "step": 35260
    },
    {
      "epoch": 1.4481853438786052,
      "grad_norm": 0.394132137298584,
      "learning_rate": 7.396437567793642e-05,
      "loss": 1.5946,
      "step": 35270
    },
    {
      "epoch": 1.4487272850441175,
      "grad_norm": 0.3766900599002838,
      "learning_rate": 7.39500114735074e-05,
      "loss": 1.5982,
      "step": 35280
    },
    {
      "epoch": 1.4492692262096296,
      "grad_norm": 0.37917500734329224,
      "learning_rate": 7.393564492163326e-05,
      "loss": 1.599,
      "step": 35290
    },
    {
      "epoch": 1.4498111673751417,
      "grad_norm": 0.30892854928970337,
      "learning_rate": 7.392127602409361e-05,
      "loss": 1.5915,
      "step": 35300
    },
    {
      "epoch": 1.450353108540654,
      "grad_norm": 0.23008306324481964,
      "learning_rate": 7.390690478266844e-05,
      "loss": 1.6001,
      "step": 35310
    },
    {
      "epoch": 1.4508950497061663,
      "grad_norm": 0.23139609396457672,
      "learning_rate": 7.389253119913801e-05,
      "loss": 1.5941,
      "step": 35320
    },
    {
      "epoch": 1.4510034379392687,
      "eval_loss": 2.5347366333007812,
      "eval_runtime": 22.3055,
      "eval_samples_per_second": 224.16,
      "eval_steps_per_second": 1.21,
      "step": 35322
    },
    {
      "epoch": 1.4514369908716784,
      "grad_norm": 0.3306918740272522,
      "learning_rate": 7.387815527528283e-05,
      "loss": 1.6064,
      "step": 35330
    },
    {
      "epoch": 1.4519789320371908,
      "grad_norm": 0.3398854732513428,
      "learning_rate": 7.386377701288372e-05,
      "loss": 1.5957,
      "step": 35340
    },
    {
      "epoch": 1.4525208732027028,
      "grad_norm": 0.4003507196903229,
      "learning_rate": 7.384939641372184e-05,
      "loss": 1.5979,
      "step": 35350
    },
    {
      "epoch": 1.4530628143682152,
      "grad_norm": 0.23540987074375153,
      "learning_rate": 7.383501347957854e-05,
      "loss": 1.5936,
      "step": 35360
    },
    {
      "epoch": 1.4536047555337275,
      "grad_norm": 0.23208670318126678,
      "learning_rate": 7.382062821223555e-05,
      "loss": 1.5963,
      "step": 35370
    },
    {
      "epoch": 1.4541466966992396,
      "grad_norm": 0.2580191195011139,
      "learning_rate": 7.380624061347485e-05,
      "loss": 1.5969,
      "step": 35380
    },
    {
      "epoch": 1.4546886378647519,
      "grad_norm": 0.4060412645339966,
      "learning_rate": 7.37918506850787e-05,
      "loss": 1.599,
      "step": 35390
    },
    {
      "epoch": 1.455230579030264,
      "grad_norm": 0.2602030634880066,
      "learning_rate": 7.377745842882971e-05,
      "loss": 1.5982,
      "step": 35400
    },
    {
      "epoch": 1.455718326079225,
      "eval_loss": 2.522488832473755,
      "eval_runtime": 21.9903,
      "eval_samples_per_second": 227.373,
      "eval_steps_per_second": 1.228,
      "step": 35409
    },
    {
      "epoch": 1.4557725201957763,
      "grad_norm": 0.31275516748428345,
      "learning_rate": 7.376306384651066e-05,
      "loss": 1.5937,
      "step": 35410
    },
    {
      "epoch": 1.4563144613612884,
      "grad_norm": 0.34258827567100525,
      "learning_rate": 7.374866693990474e-05,
      "loss": 1.5911,
      "step": 35420
    },
    {
      "epoch": 1.4568564025268007,
      "grad_norm": 0.2919900417327881,
      "learning_rate": 7.373426771079534e-05,
      "loss": 1.5827,
      "step": 35430
    },
    {
      "epoch": 1.4573983436923128,
      "grad_norm": 0.4928779900074005,
      "learning_rate": 7.37198661609662e-05,
      "loss": 1.5888,
      "step": 35440
    },
    {
      "epoch": 1.4579402848578251,
      "grad_norm": 0.25346246361732483,
      "learning_rate": 7.370546229220133e-05,
      "loss": 1.5951,
      "step": 35450
    },
    {
      "epoch": 1.4584822260233374,
      "grad_norm": 0.21547643840312958,
      "learning_rate": 7.369105610628498e-05,
      "loss": 1.5987,
      "step": 35460
    },
    {
      "epoch": 1.4590241671888495,
      "grad_norm": 0.26982036232948303,
      "learning_rate": 7.367664760500177e-05,
      "loss": 1.5837,
      "step": 35470
    },
    {
      "epoch": 1.4595661083543618,
      "grad_norm": 0.34236806631088257,
      "learning_rate": 7.366223679013652e-05,
      "loss": 1.595,
      "step": 35480
    },
    {
      "epoch": 1.460108049519874,
      "grad_norm": 0.41274112462997437,
      "learning_rate": 7.364782366347439e-05,
      "loss": 1.5924,
      "step": 35490
    },
    {
      "epoch": 1.4604332142191814,
      "eval_loss": 2.516339063644409,
      "eval_runtime": 21.9874,
      "eval_samples_per_second": 227.403,
      "eval_steps_per_second": 1.228,
      "step": 35496
    },
    {
      "epoch": 1.4606499906853863,
      "grad_norm": 0.6666343212127686,
      "learning_rate": 7.363340822680085e-05,
      "loss": 1.6145,
      "step": 35500
    },
    {
      "epoch": 1.4611919318508986,
      "grad_norm": 0.57722008228302,
      "learning_rate": 7.36189904819016e-05,
      "loss": 1.5893,
      "step": 35510
    },
    {
      "epoch": 1.4617338730164107,
      "grad_norm": 0.3001037836074829,
      "learning_rate": 7.360457043056263e-05,
      "loss": 1.5977,
      "step": 35520
    },
    {
      "epoch": 1.4622758141819228,
      "grad_norm": 0.3289230763912201,
      "learning_rate": 7.359014807457026e-05,
      "loss": 1.5955,
      "step": 35530
    },
    {
      "epoch": 1.462817755347435,
      "grad_norm": 0.26572784781455994,
      "learning_rate": 7.357572341571103e-05,
      "loss": 1.5957,
      "step": 35540
    },
    {
      "epoch": 1.4633596965129474,
      "grad_norm": 0.3806922435760498,
      "learning_rate": 7.356129645577186e-05,
      "loss": 1.5925,
      "step": 35550
    },
    {
      "epoch": 1.4639016376784595,
      "grad_norm": 0.31142738461494446,
      "learning_rate": 7.354686719653986e-05,
      "loss": 1.5968,
      "step": 35560
    },
    {
      "epoch": 1.4644435788439718,
      "grad_norm": 0.26502105593681335,
      "learning_rate": 7.353243563980247e-05,
      "loss": 1.6025,
      "step": 35570
    },
    {
      "epoch": 1.464985520009484,
      "grad_norm": 0.3371651768684387,
      "learning_rate": 7.351800178734741e-05,
      "loss": 1.5903,
      "step": 35580
    },
    {
      "epoch": 1.4651481023591377,
      "eval_loss": 2.5116758346557617,
      "eval_runtime": 21.9879,
      "eval_samples_per_second": 227.398,
      "eval_steps_per_second": 1.228,
      "step": 35583
    },
    {
      "epoch": 1.4655274611749962,
      "grad_norm": 0.3568730056285858,
      "learning_rate": 7.350356564096269e-05,
      "loss": 1.5913,
      "step": 35590
    },
    {
      "epoch": 1.4660694023405085,
      "grad_norm": 0.29186302423477173,
      "learning_rate": 7.348912720243661e-05,
      "loss": 1.5942,
      "step": 35600
    },
    {
      "epoch": 1.4666113435060206,
      "grad_norm": 0.2645188868045807,
      "learning_rate": 7.347468647355771e-05,
      "loss": 1.5962,
      "step": 35610
    },
    {
      "epoch": 1.4671532846715327,
      "grad_norm": 0.24721835553646088,
      "learning_rate": 7.346024345611485e-05,
      "loss": 1.5911,
      "step": 35620
    },
    {
      "epoch": 1.467695225837045,
      "grad_norm": 0.3433522582054138,
      "learning_rate": 7.344579815189718e-05,
      "loss": 1.5934,
      "step": 35630
    },
    {
      "epoch": 1.4682371670025574,
      "grad_norm": 0.28682151436805725,
      "learning_rate": 7.343135056269412e-05,
      "loss": 1.5839,
      "step": 35640
    },
    {
      "epoch": 1.4687791081680694,
      "grad_norm": 0.28299185633659363,
      "learning_rate": 7.341690069029538e-05,
      "loss": 1.5926,
      "step": 35650
    },
    {
      "epoch": 1.4693210493335818,
      "grad_norm": 0.3892780840396881,
      "learning_rate": 7.340244853649095e-05,
      "loss": 1.589,
      "step": 35660
    },
    {
      "epoch": 1.4698629904990939,
      "grad_norm": 0.296934574842453,
      "learning_rate": 7.338799410307107e-05,
      "loss": 1.6075,
      "step": 35670
    },
    {
      "epoch": 1.4698629904990939,
      "eval_loss": 2.520735740661621,
      "eval_runtime": 21.9779,
      "eval_samples_per_second": 227.501,
      "eval_steps_per_second": 1.229,
      "step": 35670
    },
    {
      "epoch": 1.4704049316646062,
      "grad_norm": 0.2969922423362732,
      "learning_rate": 7.337353739182631e-05,
      "loss": 1.5992,
      "step": 35680
    },
    {
      "epoch": 1.4709468728301185,
      "grad_norm": 0.24185840785503387,
      "learning_rate": 7.335907840454753e-05,
      "loss": 1.5935,
      "step": 35690
    },
    {
      "epoch": 1.4714888139956306,
      "grad_norm": 0.272232323884964,
      "learning_rate": 7.334461714302582e-05,
      "loss": 1.6053,
      "step": 35700
    },
    {
      "epoch": 1.472030755161143,
      "grad_norm": 0.2702069878578186,
      "learning_rate": 7.333015360905257e-05,
      "loss": 1.5961,
      "step": 35710
    },
    {
      "epoch": 1.472572696326655,
      "grad_norm": 0.28312546014785767,
      "learning_rate": 7.331568780441948e-05,
      "loss": 1.5832,
      "step": 35720
    },
    {
      "epoch": 1.4731146374921673,
      "grad_norm": 0.2772080898284912,
      "learning_rate": 7.330121973091849e-05,
      "loss": 1.6027,
      "step": 35730
    },
    {
      "epoch": 1.4736565786576794,
      "grad_norm": 0.2347124218940735,
      "learning_rate": 7.328674939034188e-05,
      "loss": 1.5872,
      "step": 35740
    },
    {
      "epoch": 1.4741985198231917,
      "grad_norm": 0.30288901925086975,
      "learning_rate": 7.327227678448215e-05,
      "loss": 1.5851,
      "step": 35750
    },
    {
      "epoch": 1.4745778786390502,
      "eval_loss": 2.5225229263305664,
      "eval_runtime": 21.6393,
      "eval_samples_per_second": 231.061,
      "eval_steps_per_second": 1.248,
      "step": 35757
    },
    {
      "epoch": 1.4747404609887038,
      "grad_norm": 0.3605928122997284,
      "learning_rate": 7.325780191513208e-05,
      "loss": 1.5919,
      "step": 35760
    },
    {
      "epoch": 1.4752824021542161,
      "grad_norm": 0.30377334356307983,
      "learning_rate": 7.324332478408478e-05,
      "loss": 1.5886,
      "step": 35770
    },
    {
      "epoch": 1.4758243433197284,
      "grad_norm": 0.32762688398361206,
      "learning_rate": 7.322884539313361e-05,
      "loss": 1.5903,
      "step": 35780
    },
    {
      "epoch": 1.4763662844852405,
      "grad_norm": 0.29475733637809753,
      "learning_rate": 7.321436374407222e-05,
      "loss": 1.59,
      "step": 35790
    },
    {
      "epoch": 1.4769082256507529,
      "grad_norm": 0.2471809685230255,
      "learning_rate": 7.319987983869454e-05,
      "loss": 1.5876,
      "step": 35800
    },
    {
      "epoch": 1.477450166816265,
      "grad_norm": 0.36135897040367126,
      "learning_rate": 7.318539367879475e-05,
      "loss": 1.5946,
      "step": 35810
    },
    {
      "epoch": 1.4779921079817773,
      "grad_norm": 0.42719608545303345,
      "learning_rate": 7.317090526616733e-05,
      "loss": 1.593,
      "step": 35820
    },
    {
      "epoch": 1.4785340491472894,
      "grad_norm": 0.22060419619083405,
      "learning_rate": 7.315641460260708e-05,
      "loss": 1.5905,
      "step": 35830
    },
    {
      "epoch": 1.4790759903128017,
      "grad_norm": 0.27766963839530945,
      "learning_rate": 7.3141921689909e-05,
      "loss": 1.5923,
      "step": 35840
    },
    {
      "epoch": 1.4792927667790066,
      "eval_loss": 2.5205538272857666,
      "eval_runtime": 21.9888,
      "eval_samples_per_second": 227.388,
      "eval_steps_per_second": 1.228,
      "step": 35844
    },
    {
      "epoch": 1.4796179314783138,
      "grad_norm": 0.21527595818042755,
      "learning_rate": 7.312742652986844e-05,
      "loss": 1.5926,
      "step": 35850
    },
    {
      "epoch": 1.480159872643826,
      "grad_norm": 0.2089533805847168,
      "learning_rate": 7.311292912428097e-05,
      "loss": 1.582,
      "step": 35860
    },
    {
      "epoch": 1.4807018138093384,
      "grad_norm": 0.2418367564678192,
      "learning_rate": 7.309842947494248e-05,
      "loss": 1.5862,
      "step": 35870
    },
    {
      "epoch": 1.4812437549748505,
      "grad_norm": 0.2674960494041443,
      "learning_rate": 7.308392758364912e-05,
      "loss": 1.5865,
      "step": 35880
    },
    {
      "epoch": 1.4817856961403628,
      "grad_norm": 0.24217753112316132,
      "learning_rate": 7.306942345219733e-05,
      "loss": 1.5894,
      "step": 35890
    },
    {
      "epoch": 1.482327637305875,
      "grad_norm": 0.2955038249492645,
      "learning_rate": 7.305491708238381e-05,
      "loss": 1.5957,
      "step": 35900
    },
    {
      "epoch": 1.4828695784713872,
      "grad_norm": 0.4298487603664398,
      "learning_rate": 7.304040847600555e-05,
      "loss": 1.5872,
      "step": 35910
    },
    {
      "epoch": 1.4834115196368995,
      "grad_norm": 0.3208651542663574,
      "learning_rate": 7.30258976348598e-05,
      "loss": 1.5896,
      "step": 35920
    },
    {
      "epoch": 1.4839534608024116,
      "grad_norm": 0.3353649079799652,
      "learning_rate": 7.301138456074414e-05,
      "loss": 1.5863,
      "step": 35930
    },
    {
      "epoch": 1.484007654918963,
      "eval_loss": 2.5121777057647705,
      "eval_runtime": 21.989,
      "eval_samples_per_second": 227.386,
      "eval_steps_per_second": 1.228,
      "step": 35931
    },
    {
      "epoch": 1.4844954019679237,
      "grad_norm": 0.2195397913455963,
      "learning_rate": 7.299686925545633e-05,
      "loss": 1.5874,
      "step": 35940
    },
    {
      "epoch": 1.485037343133436,
      "grad_norm": 0.27811160683631897,
      "learning_rate": 7.298235172079451e-05,
      "loss": 1.5921,
      "step": 35950
    },
    {
      "epoch": 1.4855792842989484,
      "grad_norm": 0.3784521818161011,
      "learning_rate": 7.296783195855703e-05,
      "loss": 1.5812,
      "step": 35960
    },
    {
      "epoch": 1.4861212254644605,
      "grad_norm": 0.20319710671901703,
      "learning_rate": 7.295330997054251e-05,
      "loss": 1.5943,
      "step": 35970
    },
    {
      "epoch": 1.4866631666299728,
      "grad_norm": 0.45995742082595825,
      "learning_rate": 7.293878575854993e-05,
      "loss": 1.5899,
      "step": 35980
    },
    {
      "epoch": 1.4872051077954849,
      "grad_norm": 0.4735308289527893,
      "learning_rate": 7.292425932437843e-05,
      "loss": 1.5944,
      "step": 35990
    },
    {
      "epoch": 1.4877470489609972,
      "grad_norm": 0.2938894033432007,
      "learning_rate": 7.290973066982752e-05,
      "loss": 1.5885,
      "step": 36000
    },
    {
      "epoch": 1.4882889901265095,
      "grad_norm": 0.4378818869590759,
      "learning_rate": 7.289519979669693e-05,
      "loss": 1.5856,
      "step": 36010
    },
    {
      "epoch": 1.4887225430589193,
      "eval_loss": 2.5050690174102783,
      "eval_runtime": 21.9834,
      "eval_samples_per_second": 227.444,
      "eval_steps_per_second": 1.228,
      "step": 36018
    },
    {
      "epoch": 1.4888309312920216,
      "grad_norm": 0.5193374752998352,
      "learning_rate": 7.288066670678665e-05,
      "loss": 1.6,
      "step": 36020
    },
    {
      "epoch": 1.4893728724575337,
      "grad_norm": 0.3998728394508362,
      "learning_rate": 7.286613140189704e-05,
      "loss": 1.5935,
      "step": 36030
    },
    {
      "epoch": 1.489914813623046,
      "grad_norm": 0.2564713954925537,
      "learning_rate": 7.285159388382864e-05,
      "loss": 1.5882,
      "step": 36040
    },
    {
      "epoch": 1.4904567547885583,
      "grad_norm": 0.2355422079563141,
      "learning_rate": 7.283705415438228e-05,
      "loss": 1.5895,
      "step": 36050
    },
    {
      "epoch": 1.4909986959540704,
      "grad_norm": 0.226087749004364,
      "learning_rate": 7.282251221535908e-05,
      "loss": 1.5908,
      "step": 36060
    },
    {
      "epoch": 1.4915406371195827,
      "grad_norm": 0.3110443353652954,
      "learning_rate": 7.280796806856048e-05,
      "loss": 1.5861,
      "step": 36070
    },
    {
      "epoch": 1.4920825782850948,
      "grad_norm": 0.2450808882713318,
      "learning_rate": 7.27934217157881e-05,
      "loss": 1.5803,
      "step": 36080
    },
    {
      "epoch": 1.4926245194506071,
      "grad_norm": 0.27178075909614563,
      "learning_rate": 7.277887315884388e-05,
      "loss": 1.5887,
      "step": 36090
    },
    {
      "epoch": 1.4931664606161195,
      "grad_norm": 0.35005462169647217,
      "learning_rate": 7.276432239953004e-05,
      "loss": 1.5939,
      "step": 36100
    },
    {
      "epoch": 1.4934374311988754,
      "eval_loss": 2.5150177478790283,
      "eval_runtime": 21.9887,
      "eval_samples_per_second": 227.39,
      "eval_steps_per_second": 1.228,
      "step": 36105
    },
    {
      "epoch": 1.4937084017816316,
      "grad_norm": 0.27534955739974976,
      "learning_rate": 7.274976943964906e-05,
      "loss": 1.5847,
      "step": 36110
    },
    {
      "epoch": 1.4942503429471439,
      "grad_norm": 0.29644301533699036,
      "learning_rate": 7.273521428100372e-05,
      "loss": 1.5859,
      "step": 36120
    },
    {
      "epoch": 1.494792284112656,
      "grad_norm": 0.37286531925201416,
      "learning_rate": 7.272065692539701e-05,
      "loss": 1.6031,
      "step": 36130
    },
    {
      "epoch": 1.4953342252781683,
      "grad_norm": 0.3895072340965271,
      "learning_rate": 7.270609737463229e-05,
      "loss": 1.5959,
      "step": 36140
    },
    {
      "epoch": 1.4958761664436804,
      "grad_norm": 0.3287084996700287,
      "learning_rate": 7.269153563051306e-05,
      "loss": 1.5838,
      "step": 36150
    },
    {
      "epoch": 1.4964181076091927,
      "grad_norm": 0.4030684530735016,
      "learning_rate": 7.267697169484323e-05,
      "loss": 1.58,
      "step": 36160
    },
    {
      "epoch": 1.4969600487747048,
      "grad_norm": 0.32397374510765076,
      "learning_rate": 7.26624055694269e-05,
      "loss": 1.5884,
      "step": 36170
    },
    {
      "epoch": 1.497501989940217,
      "grad_norm": 0.268887996673584,
      "learning_rate": 7.264783725606843e-05,
      "loss": 1.5928,
      "step": 36180
    },
    {
      "epoch": 1.4980439311057294,
      "grad_norm": 0.32079851627349854,
      "learning_rate": 7.263326675657251e-05,
      "loss": 1.5925,
      "step": 36190
    },
    {
      "epoch": 1.4981523193388318,
      "eval_loss": 2.512747049331665,
      "eval_runtime": 21.9853,
      "eval_samples_per_second": 227.425,
      "eval_steps_per_second": 1.228,
      "step": 36192
    },
    {
      "epoch": 1.4985858722712415,
      "grad_norm": 0.27530887722969055,
      "learning_rate": 7.261869407274406e-05,
      "loss": 1.5871,
      "step": 36200
    },
    {
      "epoch": 1.4991278134367538,
      "grad_norm": 0.39330941438674927,
      "learning_rate": 7.260411920638828e-05,
      "loss": 1.5856,
      "step": 36210
    },
    {
      "epoch": 1.499669754602266,
      "grad_norm": 0.23756785690784454,
      "learning_rate": 7.258954215931064e-05,
      "loss": 1.5878,
      "step": 36220
    },
    {
      "epoch": 1.5002116957677782,
      "grad_norm": 0.3410539925098419,
      "learning_rate": 7.257496293331688e-05,
      "loss": 1.6006,
      "step": 36230
    },
    {
      "epoch": 1.5007536369332906,
      "grad_norm": 0.3402417302131653,
      "learning_rate": 7.256038153021303e-05,
      "loss": 1.5809,
      "step": 36240
    },
    {
      "epoch": 1.5012955780988027,
      "grad_norm": 0.25963348150253296,
      "learning_rate": 7.254579795180534e-05,
      "loss": 1.58,
      "step": 36250
    },
    {
      "epoch": 1.5018375192643147,
      "grad_norm": 0.37318065762519836,
      "learning_rate": 7.253121219990038e-05,
      "loss": 1.5912,
      "step": 36260
    },
    {
      "epoch": 1.502379460429827,
      "grad_norm": 0.45932379364967346,
      "learning_rate": 7.251662427630496e-05,
      "loss": 1.5933,
      "step": 36270
    },
    {
      "epoch": 1.502867207478788,
      "eval_loss": 2.5038368701934814,
      "eval_runtime": 21.9923,
      "eval_samples_per_second": 227.352,
      "eval_steps_per_second": 1.228,
      "step": 36279
    },
    {
      "epoch": 1.5029214015953394,
      "grad_norm": 0.3073781430721283,
      "learning_rate": 7.25020341828262e-05,
      "loss": 1.5869,
      "step": 36280
    },
    {
      "epoch": 1.5034633427608515,
      "grad_norm": 0.29523077607154846,
      "learning_rate": 7.24874419212714e-05,
      "loss": 1.5864,
      "step": 36290
    },
    {
      "epoch": 1.5040052839263638,
      "grad_norm": 0.3561224341392517,
      "learning_rate": 7.247284749344824e-05,
      "loss": 1.5904,
      "step": 36300
    },
    {
      "epoch": 1.5045472250918759,
      "grad_norm": 0.26353198289871216,
      "learning_rate": 7.245825090116457e-05,
      "loss": 1.5867,
      "step": 36310
    },
    {
      "epoch": 1.5050891662573882,
      "grad_norm": 0.2485145479440689,
      "learning_rate": 7.244365214622859e-05,
      "loss": 1.5951,
      "step": 36320
    },
    {
      "epoch": 1.5056311074229005,
      "grad_norm": 0.21996235847473145,
      "learning_rate": 7.242905123044872e-05,
      "loss": 1.5847,
      "step": 36330
    },
    {
      "epoch": 1.5061730485884126,
      "grad_norm": 0.32352975010871887,
      "learning_rate": 7.241444815563364e-05,
      "loss": 1.5844,
      "step": 36340
    },
    {
      "epoch": 1.5067149897539247,
      "grad_norm": 0.3241994380950928,
      "learning_rate": 7.239984292359232e-05,
      "loss": 1.5895,
      "step": 36350
    },
    {
      "epoch": 1.507256930919437,
      "grad_norm": 0.7218349575996399,
      "learning_rate": 7.238523553613402e-05,
      "loss": 1.5877,
      "step": 36360
    },
    {
      "epoch": 1.5075820956187442,
      "eval_loss": 2.5044937133789062,
      "eval_runtime": 21.9867,
      "eval_samples_per_second": 227.411,
      "eval_steps_per_second": 1.228,
      "step": 36366
    },
    {
      "epoch": 1.5077988720849493,
      "grad_norm": 0.5906317830085754,
      "learning_rate": 7.237062599506821e-05,
      "loss": 1.589,
      "step": 36370
    },
    {
      "epoch": 1.5083408132504617,
      "grad_norm": 0.43497881293296814,
      "learning_rate": 7.23560143022047e-05,
      "loss": 1.5833,
      "step": 36380
    },
    {
      "epoch": 1.5088827544159737,
      "grad_norm": 0.3339081406593323,
      "learning_rate": 7.234140045935345e-05,
      "loss": 1.6007,
      "step": 36390
    },
    {
      "epoch": 1.5094246955814858,
      "grad_norm": 0.2763802111148834,
      "learning_rate": 7.23267844683248e-05,
      "loss": 1.5971,
      "step": 36400
    },
    {
      "epoch": 1.5099666367469982,
      "grad_norm": 0.3865683376789093,
      "learning_rate": 7.231216633092934e-05,
      "loss": 1.5844,
      "step": 36410
    },
    {
      "epoch": 1.5105085779125105,
      "grad_norm": 0.32255837321281433,
      "learning_rate": 7.229754604897786e-05,
      "loss": 1.5878,
      "step": 36420
    },
    {
      "epoch": 1.5110505190780226,
      "grad_norm": 0.29356804490089417,
      "learning_rate": 7.228292362428148e-05,
      "loss": 1.5826,
      "step": 36430
    },
    {
      "epoch": 1.5115924602435347,
      "grad_norm": 0.2429652065038681,
      "learning_rate": 7.226829905865156e-05,
      "loss": 1.5873,
      "step": 36440
    },
    {
      "epoch": 1.512134401409047,
      "grad_norm": 0.39031389355659485,
      "learning_rate": 7.225367235389972e-05,
      "loss": 1.5894,
      "step": 36450
    },
    {
      "epoch": 1.5122969837587008,
      "eval_loss": 2.497527837753296,
      "eval_runtime": 21.9849,
      "eval_samples_per_second": 227.429,
      "eval_steps_per_second": 1.228,
      "step": 36453
    },
    {
      "epoch": 1.5126763425745593,
      "grad_norm": 0.27026987075805664,
      "learning_rate": 7.223904351183786e-05,
      "loss": 1.5877,
      "step": 36460
    },
    {
      "epoch": 1.5132182837400716,
      "grad_norm": 0.26341304183006287,
      "learning_rate": 7.222441253427813e-05,
      "loss": 1.5977,
      "step": 36470
    },
    {
      "epoch": 1.5137602249055837,
      "grad_norm": 0.2732945382595062,
      "learning_rate": 7.220977942303295e-05,
      "loss": 1.5945,
      "step": 36480
    },
    {
      "epoch": 1.5143021660710958,
      "grad_norm": 0.29458683729171753,
      "learning_rate": 7.219514417991505e-05,
      "loss": 1.5861,
      "step": 36490
    },
    {
      "epoch": 1.5148441072366081,
      "grad_norm": 0.5213008522987366,
      "learning_rate": 7.218050680673729e-05,
      "loss": 1.5707,
      "step": 36500
    },
    {
      "epoch": 1.5153860484021204,
      "grad_norm": 0.4199194014072418,
      "learning_rate": 7.216586730531296e-05,
      "loss": 1.5796,
      "step": 36510
    },
    {
      "epoch": 1.5159279895676325,
      "grad_norm": 0.23940829932689667,
      "learning_rate": 7.215122567745552e-05,
      "loss": 1.5849,
      "step": 36520
    },
    {
      "epoch": 1.5164699307331446,
      "grad_norm": 0.2722623944282532,
      "learning_rate": 7.213658192497869e-05,
      "loss": 1.5892,
      "step": 36530
    },
    {
      "epoch": 1.517011871898657,
      "grad_norm": 0.24977056682109833,
      "learning_rate": 7.212193604969652e-05,
      "loss": 1.5806,
      "step": 36540
    },
    {
      "epoch": 1.517011871898657,
      "eval_loss": 2.511324644088745,
      "eval_runtime": 21.9284,
      "eval_samples_per_second": 228.015,
      "eval_steps_per_second": 1.231,
      "step": 36540
    },
    {
      "epoch": 1.5175538130641693,
      "grad_norm": 0.4285825788974762,
      "learning_rate": 7.21072880534232e-05,
      "loss": 1.5742,
      "step": 36550
    },
    {
      "epoch": 1.5180957542296816,
      "grad_norm": 0.72999107837677,
      "learning_rate": 7.209263793797335e-05,
      "loss": 1.5943,
      "step": 36560
    },
    {
      "epoch": 1.5186376953951937,
      "grad_norm": 0.4707547426223755,
      "learning_rate": 7.207798570516172e-05,
      "loss": 1.5901,
      "step": 36570
    },
    {
      "epoch": 1.5191796365607058,
      "grad_norm": 0.25756052136421204,
      "learning_rate": 7.206333135680336e-05,
      "loss": 1.5844,
      "step": 36580
    },
    {
      "epoch": 1.519721577726218,
      "grad_norm": 0.21552801132202148,
      "learning_rate": 7.204867489471359e-05,
      "loss": 1.5989,
      "step": 36590
    },
    {
      "epoch": 1.5202635188917304,
      "grad_norm": 0.3025147616863251,
      "learning_rate": 7.203401632070798e-05,
      "loss": 1.5821,
      "step": 36600
    },
    {
      "epoch": 1.5208054600572425,
      "grad_norm": 0.3975871801376343,
      "learning_rate": 7.201935563660239e-05,
      "loss": 1.5842,
      "step": 36610
    },
    {
      "epoch": 1.5213474012227548,
      "grad_norm": 0.2734328806400299,
      "learning_rate": 7.200469284421292e-05,
      "loss": 1.5842,
      "step": 36620
    },
    {
      "epoch": 1.5217267600386133,
      "eval_loss": 2.4983890056610107,
      "eval_runtime": 21.9821,
      "eval_samples_per_second": 227.458,
      "eval_steps_per_second": 1.228,
      "step": 36627
    },
    {
      "epoch": 1.521889342388267,
      "grad_norm": 0.23720048367977142,
      "learning_rate": 7.199002794535593e-05,
      "loss": 1.5873,
      "step": 36630
    },
    {
      "epoch": 1.5224312835537792,
      "grad_norm": 0.26733702421188354,
      "learning_rate": 7.197536094184803e-05,
      "loss": 1.5868,
      "step": 36640
    },
    {
      "epoch": 1.5229732247192915,
      "grad_norm": 0.3890104293823242,
      "learning_rate": 7.196069183550612e-05,
      "loss": 1.5894,
      "step": 36650
    },
    {
      "epoch": 1.5235151658848036,
      "grad_norm": 0.23871298134326935,
      "learning_rate": 7.194602062814733e-05,
      "loss": 1.5899,
      "step": 36660
    },
    {
      "epoch": 1.5240571070503157,
      "grad_norm": 0.20540666580200195,
      "learning_rate": 7.19313473215891e-05,
      "loss": 1.585,
      "step": 36670
    },
    {
      "epoch": 1.524599048215828,
      "grad_norm": 0.2062886506319046,
      "learning_rate": 7.191667191764906e-05,
      "loss": 1.5882,
      "step": 36680
    },
    {
      "epoch": 1.5251409893813404,
      "grad_norm": 0.2321225106716156,
      "learning_rate": 7.190199441814516e-05,
      "loss": 1.5901,
      "step": 36690
    },
    {
      "epoch": 1.5256829305468527,
      "grad_norm": 0.4270740747451782,
      "learning_rate": 7.188731482489556e-05,
      "loss": 1.5668,
      "step": 36700
    },
    {
      "epoch": 1.5262248717123648,
      "grad_norm": 0.34541916847229004,
      "learning_rate": 7.187263313971872e-05,
      "loss": 1.5925,
      "step": 36710
    },
    {
      "epoch": 1.5264416481785696,
      "eval_loss": 2.493448495864868,
      "eval_runtime": 21.9846,
      "eval_samples_per_second": 227.432,
      "eval_steps_per_second": 1.228,
      "step": 36714
    },
    {
      "epoch": 1.5267668128778769,
      "grad_norm": 0.44076773524284363,
      "learning_rate": 7.185794936443334e-05,
      "loss": 1.5788,
      "step": 36720
    },
    {
      "epoch": 1.5273087540433892,
      "grad_norm": 0.2881830334663391,
      "learning_rate": 7.184326350085839e-05,
      "loss": 1.5871,
      "step": 36730
    },
    {
      "epoch": 1.5278506952089015,
      "grad_norm": 0.4518699645996094,
      "learning_rate": 7.182857555081307e-05,
      "loss": 1.5941,
      "step": 36740
    },
    {
      "epoch": 1.5283926363744136,
      "grad_norm": 0.48421424627304077,
      "learning_rate": 7.181388551611688e-05,
      "loss": 1.5865,
      "step": 36750
    },
    {
      "epoch": 1.5289345775399257,
      "grad_norm": 0.2462109625339508,
      "learning_rate": 7.179919339858955e-05,
      "loss": 1.5774,
      "step": 36760
    },
    {
      "epoch": 1.529476518705438,
      "grad_norm": 0.25756916403770447,
      "learning_rate": 7.178449920005108e-05,
      "loss": 1.5809,
      "step": 36770
    },
    {
      "epoch": 1.5300184598709503,
      "grad_norm": 0.28613096475601196,
      "learning_rate": 7.176980292232173e-05,
      "loss": 1.5815,
      "step": 36780
    },
    {
      "epoch": 1.5305604010364626,
      "grad_norm": 0.3218177258968353,
      "learning_rate": 7.175510456722198e-05,
      "loss": 1.5914,
      "step": 36790
    },
    {
      "epoch": 1.5311023422019747,
      "grad_norm": 0.42091110348701477,
      "learning_rate": 7.174040413657262e-05,
      "loss": 1.5778,
      "step": 36800
    },
    {
      "epoch": 1.5311565363185258,
      "eval_loss": 2.505159854888916,
      "eval_runtime": 21.9856,
      "eval_samples_per_second": 227.422,
      "eval_steps_per_second": 1.228,
      "step": 36801
    },
    {
      "epoch": 1.5316442833674868,
      "grad_norm": 0.23221740126609802,
      "learning_rate": 7.17257016321947e-05,
      "loss": 1.5965,
      "step": 36810
    },
    {
      "epoch": 1.5321862245329991,
      "grad_norm": 0.28910213708877563,
      "learning_rate": 7.171099705590946e-05,
      "loss": 1.5775,
      "step": 36820
    },
    {
      "epoch": 1.5327281656985114,
      "grad_norm": 0.25105226039886475,
      "learning_rate": 7.169629040953846e-05,
      "loss": 1.5871,
      "step": 36830
    },
    {
      "epoch": 1.5332701068640235,
      "grad_norm": 0.2661377489566803,
      "learning_rate": 7.168158169490347e-05,
      "loss": 1.5805,
      "step": 36840
    },
    {
      "epoch": 1.5338120480295356,
      "grad_norm": 0.28218016028404236,
      "learning_rate": 7.166687091382659e-05,
      "loss": 1.5916,
      "step": 36850
    },
    {
      "epoch": 1.534353989195048,
      "grad_norm": 0.3052498400211334,
      "learning_rate": 7.165215806813009e-05,
      "loss": 1.5885,
      "step": 36860
    },
    {
      "epoch": 1.5348959303605603,
      "grad_norm": 0.2858087718486786,
      "learning_rate": 7.163744315963656e-05,
      "loss": 1.5752,
      "step": 36870
    },
    {
      "epoch": 1.5354378715260726,
      "grad_norm": 0.2672872245311737,
      "learning_rate": 7.162272619016879e-05,
      "loss": 1.5828,
      "step": 36880
    },
    {
      "epoch": 1.5358714244584823,
      "eval_loss": 2.495072364807129,
      "eval_runtime": 21.9812,
      "eval_samples_per_second": 227.467,
      "eval_steps_per_second": 1.228,
      "step": 36888
    },
    {
      "epoch": 1.5359798126915847,
      "grad_norm": 0.3383979797363281,
      "learning_rate": 7.160800716154987e-05,
      "loss": 1.5818,
      "step": 36890
    },
    {
      "epoch": 1.5365217538570968,
      "grad_norm": 0.3658377528190613,
      "learning_rate": 7.159328607560312e-05,
      "loss": 1.5777,
      "step": 36900
    },
    {
      "epoch": 1.537063695022609,
      "grad_norm": 0.2234833836555481,
      "learning_rate": 7.157856293415216e-05,
      "loss": 1.5839,
      "step": 36910
    },
    {
      "epoch": 1.5376056361881214,
      "grad_norm": 0.2896660566329956,
      "learning_rate": 7.156383773902076e-05,
      "loss": 1.5878,
      "step": 36920
    },
    {
      "epoch": 1.5381475773536335,
      "grad_norm": 0.37888798117637634,
      "learning_rate": 7.15491104920331e-05,
      "loss": 1.5934,
      "step": 36930
    },
    {
      "epoch": 1.5386895185191458,
      "grad_norm": 0.49281513690948486,
      "learning_rate": 7.153438119501346e-05,
      "loss": 1.5918,
      "step": 36940
    },
    {
      "epoch": 1.539231459684658,
      "grad_norm": 0.5974851846694946,
      "learning_rate": 7.151964984978645e-05,
      "loss": 1.59,
      "step": 36950
    },
    {
      "epoch": 1.5397734008501702,
      "grad_norm": 0.28145140409469604,
      "learning_rate": 7.150491645817695e-05,
      "loss": 1.5886,
      "step": 36960
    },
    {
      "epoch": 1.5403153420156825,
      "grad_norm": 0.2434787005186081,
      "learning_rate": 7.149018102201007e-05,
      "loss": 1.5792,
      "step": 36970
    },
    {
      "epoch": 1.5405863125984385,
      "eval_loss": 2.5023746490478516,
      "eval_runtime": 21.9846,
      "eval_samples_per_second": 227.432,
      "eval_steps_per_second": 1.228,
      "step": 36975
    },
    {
      "epoch": 1.5408572831811946,
      "grad_norm": 0.2465183436870575,
      "learning_rate": 7.147544354311114e-05,
      "loss": 1.5838,
      "step": 36980
    },
    {
      "epoch": 1.5413992243467067,
      "grad_norm": 0.23497800529003143,
      "learning_rate": 7.146070402330578e-05,
      "loss": 1.5924,
      "step": 36990
    },
    {
      "epoch": 1.541941165512219,
      "grad_norm": 0.3130892217159271,
      "learning_rate": 7.14459624644199e-05,
      "loss": 1.5842,
      "step": 37000
    },
    {
      "epoch": 1.5424831066777314,
      "grad_norm": 0.26590847969055176,
      "learning_rate": 7.143121886827955e-05,
      "loss": 1.5869,
      "step": 37010
    },
    {
      "epoch": 1.5430250478432437,
      "grad_norm": 0.22026236355304718,
      "learning_rate": 7.141647323671119e-05,
      "loss": 1.5772,
      "step": 37020
    },
    {
      "epoch": 1.5435669890087558,
      "grad_norm": 0.3443725109100342,
      "learning_rate": 7.140172557154138e-05,
      "loss": 1.5823,
      "step": 37030
    },
    {
      "epoch": 1.5441089301742679,
      "grad_norm": 0.3271442949771881,
      "learning_rate": 7.1386975874597e-05,
      "loss": 1.5902,
      "step": 37040
    },
    {
      "epoch": 1.5446508713397802,
      "grad_norm": 0.3126424252986908,
      "learning_rate": 7.137222414770522e-05,
      "loss": 1.5913,
      "step": 37050
    },
    {
      "epoch": 1.5451928125052925,
      "grad_norm": 0.3820621073246002,
      "learning_rate": 7.135747039269339e-05,
      "loss": 1.5824,
      "step": 37060
    },
    {
      "epoch": 1.5453012007383948,
      "eval_loss": 2.5038020610809326,
      "eval_runtime": 21.984,
      "eval_samples_per_second": 227.438,
      "eval_steps_per_second": 1.228,
      "step": 37062
    },
    {
      "epoch": 1.5457347536708046,
      "grad_norm": 0.29371535778045654,
      "learning_rate": 7.134271461138913e-05,
      "loss": 1.5818,
      "step": 37070
    },
    {
      "epoch": 1.5462766948363167,
      "grad_norm": 0.3977007567882538,
      "learning_rate": 7.132795680562035e-05,
      "loss": 1.5777,
      "step": 37080
    },
    {
      "epoch": 1.546818636001829,
      "grad_norm": 0.21876034140586853,
      "learning_rate": 7.131319697721515e-05,
      "loss": 1.5834,
      "step": 37090
    },
    {
      "epoch": 1.5473605771673413,
      "grad_norm": 0.28337687253952026,
      "learning_rate": 7.129843512800196e-05,
      "loss": 1.5842,
      "step": 37100
    },
    {
      "epoch": 1.5479025183328536,
      "grad_norm": 0.2559991180896759,
      "learning_rate": 7.128367125980938e-05,
      "loss": 1.5773,
      "step": 37110
    },
    {
      "epoch": 1.5484444594983657,
      "grad_norm": 0.442965567111969,
      "learning_rate": 7.126890537446628e-05,
      "loss": 1.5765,
      "step": 37120
    },
    {
      "epoch": 1.5489864006638778,
      "grad_norm": 0.4674750566482544,
      "learning_rate": 7.125413747380183e-05,
      "loss": 1.5938,
      "step": 37130
    },
    {
      "epoch": 1.5495283418293901,
      "grad_norm": 0.30193501710891724,
      "learning_rate": 7.12393675596454e-05,
      "loss": 1.585,
      "step": 37140
    },
    {
      "epoch": 1.5500160888783512,
      "eval_loss": 2.501018524169922,
      "eval_runtime": 21.9863,
      "eval_samples_per_second": 227.415,
      "eval_steps_per_second": 1.228,
      "step": 37149
    },
    {
      "epoch": 1.5500702829949025,
      "grad_norm": 0.4235888719558716,
      "learning_rate": 7.122459563382663e-05,
      "loss": 1.5755,
      "step": 37150
    },
    {
      "epoch": 1.5506122241604146,
      "grad_norm": 0.2661120593547821,
      "learning_rate": 7.120982169817538e-05,
      "loss": 1.5908,
      "step": 37160
    },
    {
      "epoch": 1.5511541653259266,
      "grad_norm": 0.2776598334312439,
      "learning_rate": 7.119504575452179e-05,
      "loss": 1.5824,
      "step": 37170
    },
    {
      "epoch": 1.551696106491439,
      "grad_norm": 0.44518810510635376,
      "learning_rate": 7.118026780469626e-05,
      "loss": 1.5958,
      "step": 37180
    },
    {
      "epoch": 1.5522380476569513,
      "grad_norm": 0.19919413328170776,
      "learning_rate": 7.116548785052938e-05,
      "loss": 1.5844,
      "step": 37190
    },
    {
      "epoch": 1.5527799888224636,
      "grad_norm": 0.27062350511550903,
      "learning_rate": 7.115070589385206e-05,
      "loss": 1.5944,
      "step": 37200
    },
    {
      "epoch": 1.5533219299879757,
      "grad_norm": 0.25841081142425537,
      "learning_rate": 7.113592193649542e-05,
      "loss": 1.5645,
      "step": 37210
    },
    {
      "epoch": 1.5538638711534878,
      "grad_norm": 0.22730769217014313,
      "learning_rate": 7.112113598029083e-05,
      "loss": 1.5833,
      "step": 37220
    },
    {
      "epoch": 1.554405812319,
      "grad_norm": 0.22557532787322998,
      "learning_rate": 7.110634802706988e-05,
      "loss": 1.5742,
      "step": 37230
    },
    {
      "epoch": 1.5547309770183073,
      "eval_loss": 2.501530885696411,
      "eval_runtime": 21.9894,
      "eval_samples_per_second": 227.383,
      "eval_steps_per_second": 1.228,
      "step": 37236
    },
    {
      "epoch": 1.5549477534845124,
      "grad_norm": 0.24140417575836182,
      "learning_rate": 7.109155807866449e-05,
      "loss": 1.5807,
      "step": 37240
    },
    {
      "epoch": 1.5554896946500245,
      "grad_norm": 0.33359408378601074,
      "learning_rate": 7.107676613690673e-05,
      "loss": 1.5806,
      "step": 37250
    },
    {
      "epoch": 1.5560316358155368,
      "grad_norm": 0.3180777132511139,
      "learning_rate": 7.1061972203629e-05,
      "loss": 1.577,
      "step": 37260
    },
    {
      "epoch": 1.556573576981049,
      "grad_norm": 0.29410409927368164,
      "learning_rate": 7.104717628066387e-05,
      "loss": 1.5777,
      "step": 37270
    },
    {
      "epoch": 1.5571155181465612,
      "grad_norm": 0.4459421932697296,
      "learning_rate": 7.103237836984421e-05,
      "loss": 1.5918,
      "step": 37280
    },
    {
      "epoch": 1.5576574593120736,
      "grad_norm": 0.24809594452381134,
      "learning_rate": 7.101757847300312e-05,
      "loss": 1.583,
      "step": 37290
    },
    {
      "epoch": 1.5581994004775856,
      "grad_norm": 0.2839214503765106,
      "learning_rate": 7.100277659197396e-05,
      "loss": 1.5854,
      "step": 37300
    },
    {
      "epoch": 1.5587413416430977,
      "grad_norm": 0.32646796107292175,
      "learning_rate": 7.098797272859032e-05,
      "loss": 1.5824,
      "step": 37310
    },
    {
      "epoch": 1.55928328280861,
      "grad_norm": 0.3208850920200348,
      "learning_rate": 7.097316688468602e-05,
      "loss": 1.5793,
      "step": 37320
    },
    {
      "epoch": 1.5594458651582639,
      "eval_loss": 2.494997024536133,
      "eval_runtime": 21.9879,
      "eval_samples_per_second": 227.397,
      "eval_steps_per_second": 1.228,
      "step": 37323
    },
    {
      "epoch": 1.5598252239741224,
      "grad_norm": 0.2501738965511322,
      "learning_rate": 7.095835906209515e-05,
      "loss": 1.59,
      "step": 37330
    },
    {
      "epoch": 1.5603671651396347,
      "grad_norm": 0.3885467052459717,
      "learning_rate": 7.094354926265206e-05,
      "loss": 1.5841,
      "step": 37340
    },
    {
      "epoch": 1.5609091063051468,
      "grad_norm": 0.48911768198013306,
      "learning_rate": 7.09287374881913e-05,
      "loss": 1.5772,
      "step": 37350
    },
    {
      "epoch": 1.5614510474706589,
      "grad_norm": 0.3899608552455902,
      "learning_rate": 7.09139237405477e-05,
      "loss": 1.5811,
      "step": 37360
    },
    {
      "epoch": 1.5619929886361712,
      "grad_norm": 0.44720959663391113,
      "learning_rate": 7.089910802155631e-05,
      "loss": 1.5769,
      "step": 37370
    },
    {
      "epoch": 1.5625349298016835,
      "grad_norm": 0.3570062816143036,
      "learning_rate": 7.088429033305245e-05,
      "loss": 1.5878,
      "step": 37380
    },
    {
      "epoch": 1.5630768709671956,
      "grad_norm": 0.3376300036907196,
      "learning_rate": 7.086947067687167e-05,
      "loss": 1.5816,
      "step": 37390
    },
    {
      "epoch": 1.5636188121327077,
      "grad_norm": 0.24573005735874176,
      "learning_rate": 7.085464905484974e-05,
      "loss": 1.58,
      "step": 37400
    },
    {
      "epoch": 1.56416075329822,
      "grad_norm": 0.3362915813922882,
      "learning_rate": 7.083982546882275e-05,
      "loss": 1.5777,
      "step": 37410
    },
    {
      "epoch": 1.56416075329822,
      "eval_loss": 2.491037368774414,
      "eval_runtime": 21.982,
      "eval_samples_per_second": 227.459,
      "eval_steps_per_second": 1.228,
      "step": 37410
    },
    {
      "epoch": 1.5647026944637323,
      "grad_norm": 0.37921565771102905,
      "learning_rate": 7.082499992062693e-05,
      "loss": 1.5786,
      "step": 37420
    },
    {
      "epoch": 1.5652446356292447,
      "grad_norm": 0.3702291250228882,
      "learning_rate": 7.081017241209883e-05,
      "loss": 1.5851,
      "step": 37430
    },
    {
      "epoch": 1.5657865767947567,
      "grad_norm": 0.23378178477287292,
      "learning_rate": 7.07953429450752e-05,
      "loss": 1.5818,
      "step": 37440
    },
    {
      "epoch": 1.5663285179602688,
      "grad_norm": 0.3024660646915436,
      "learning_rate": 7.078051152139308e-05,
      "loss": 1.5752,
      "step": 37450
    },
    {
      "epoch": 1.5668704591257812,
      "grad_norm": 0.25815534591674805,
      "learning_rate": 7.076567814288971e-05,
      "loss": 1.5895,
      "step": 37460
    },
    {
      "epoch": 1.5674124002912935,
      "grad_norm": 0.35706350207328796,
      "learning_rate": 7.075084281140256e-05,
      "loss": 1.567,
      "step": 37470
    },
    {
      "epoch": 1.5679543414568056,
      "grad_norm": 0.34599047899246216,
      "learning_rate": 7.07360055287694e-05,
      "loss": 1.5831,
      "step": 37480
    },
    {
      "epoch": 1.5684962826223177,
      "grad_norm": 0.22840653359889984,
      "learning_rate": 7.072116629682819e-05,
      "loss": 1.5795,
      "step": 37490
    },
    {
      "epoch": 1.5688756414381764,
      "eval_loss": 2.498384714126587,
      "eval_runtime": 21.9865,
      "eval_samples_per_second": 227.412,
      "eval_steps_per_second": 1.228,
      "step": 37497
    },
    {
      "epoch": 1.56903822378783,
      "grad_norm": 0.24497516453266144,
      "learning_rate": 7.070632511741717e-05,
      "loss": 1.5712,
      "step": 37500
    },
    {
      "epoch": 1.5695801649533423,
      "grad_norm": 0.3624199628829956,
      "learning_rate": 7.069148199237476e-05,
      "loss": 1.5803,
      "step": 37510
    },
    {
      "epoch": 2.000325164699307,
      "grad_norm": 0.2853184640407562,
      "learning_rate": 2.0000000000000002e-07,
      "loss": 1.5294,
      "step": 37520
    },
    {
      "epoch": 2.0008671058648195,
      "grad_norm": 0.2135554403066635,
      "learning_rate": 5.333333333333333e-07,
      "loss": 1.5365,
      "step": 37530
    },
    {
      "epoch": 2.001409047030332,
      "grad_norm": 0.20927861332893372,
      "learning_rate": 8.666666666666667e-07,
      "loss": 1.5365,
      "step": 37540
    },
    {
      "epoch": 2.001950988195844,
      "grad_norm": 0.17029166221618652,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 1.5374,
      "step": 37550
    },
    {
      "epoch": 2.002492929361356,
      "grad_norm": 0.16770167648792267,
      "learning_rate": 1.5333333333333334e-06,
      "loss": 1.5412,
      "step": 37560
    },
    {
      "epoch": 2.0030348705268683,
      "grad_norm": 0.17810305953025818,
      "learning_rate": 1.8666666666666669e-06,
      "loss": 1.5236,
      "step": 37570
    },
    {
      "epoch": 2.0035768116923807,
      "grad_norm": 0.18917031586170197,
      "learning_rate": 2.2e-06,
      "loss": 1.5386,
      "step": 37580
    },
    {
      "epoch": 2.0037935881585853,
      "eval_loss": 2.4802141189575195,
      "eval_runtime": 22.3304,
      "eval_samples_per_second": 223.91,
      "eval_steps_per_second": 1.209,
      "step": 37584
    },
    {
      "epoch": 2.004118752857893,
      "grad_norm": 0.1666443794965744,
      "learning_rate": 2.5333333333333334e-06,
      "loss": 1.5323,
      "step": 37590
    },
    {
      "epoch": 2.0046606940234053,
      "grad_norm": 0.16958613693714142,
      "learning_rate": 2.8666666666666666e-06,
      "loss": 1.5331,
      "step": 37600
    },
    {
      "epoch": 2.005202635188917,
      "grad_norm": 0.16641965508460999,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 1.5381,
      "step": 37610
    },
    {
      "epoch": 2.0057445763544295,
      "grad_norm": 0.18020287156105042,
      "learning_rate": 3.5333333333333335e-06,
      "loss": 1.5293,
      "step": 37620
    },
    {
      "epoch": 2.006286517519942,
      "grad_norm": 0.16953766345977783,
      "learning_rate": 3.866666666666667e-06,
      "loss": 1.5372,
      "step": 37630
    },
    {
      "epoch": 2.006828458685454,
      "grad_norm": 0.16563622653484344,
      "learning_rate": 4.2000000000000004e-06,
      "loss": 1.53,
      "step": 37640
    },
    {
      "epoch": 2.007370399850966,
      "grad_norm": 0.1674739271402359,
      "learning_rate": 4.533333333333334e-06,
      "loss": 1.5224,
      "step": 37650
    },
    {
      "epoch": 2.0079123410164783,
      "grad_norm": 0.17399495840072632,
      "learning_rate": 4.866666666666667e-06,
      "loss": 1.5198,
      "step": 37660
    },
    {
      "epoch": 2.0084542821819906,
      "grad_norm": 0.166429802775383,
      "learning_rate": 5.2e-06,
      "loss": 1.5246,
      "step": 37670
    },
    {
      "epoch": 2.008508476298542,
      "eval_loss": 2.476987600326538,
      "eval_runtime": 27.0724,
      "eval_samples_per_second": 184.69,
      "eval_steps_per_second": 0.997,
      "step": 37671
    },
    {
      "epoch": 2.008996223347503,
      "grad_norm": 0.17933695018291473,
      "learning_rate": 5.5333333333333334e-06,
      "loss": 1.5297,
      "step": 37680
    },
    {
      "epoch": 2.0095381645130153,
      "grad_norm": 0.16821593046188354,
      "learning_rate": 5.866666666666667e-06,
      "loss": 1.5318,
      "step": 37690
    },
    {
      "epoch": 2.010080105678527,
      "grad_norm": 0.17084164917469025,
      "learning_rate": 6.2e-06,
      "loss": 1.5276,
      "step": 37700
    },
    {
      "epoch": 2.0106220468440394,
      "grad_norm": 0.17240796983242035,
      "learning_rate": 6.533333333333333e-06,
      "loss": 1.5238,
      "step": 37710
    },
    {
      "epoch": 2.0111639880095518,
      "grad_norm": 0.1693965643644333,
      "learning_rate": 6.866666666666667e-06,
      "loss": 1.5267,
      "step": 37720
    },
    {
      "epoch": 2.011705929175064,
      "grad_norm": 0.17528323829174042,
      "learning_rate": 7.2e-06,
      "loss": 1.5202,
      "step": 37730
    },
    {
      "epoch": 2.012247870340576,
      "grad_norm": 0.16867630183696747,
      "learning_rate": 7.533333333333334e-06,
      "loss": 1.5173,
      "step": 37740
    },
    {
      "epoch": 2.0127898115060883,
      "grad_norm": 0.17133750021457672,
      "learning_rate": 7.866666666666667e-06,
      "loss": 1.5128,
      "step": 37750
    },
    {
      "epoch": 2.013223364438498,
      "eval_loss": 2.475126028060913,
      "eval_runtime": 21.9812,
      "eval_samples_per_second": 227.467,
      "eval_steps_per_second": 1.228,
      "step": 37758
    },
    {
      "epoch": 2.0133317526716006,
      "grad_norm": 0.17712301015853882,
      "learning_rate": 8.200000000000001e-06,
      "loss": 1.5215,
      "step": 37760
    },
    {
      "epoch": 2.013873693837113,
      "grad_norm": 0.19257335364818573,
      "learning_rate": 8.533333333333334e-06,
      "loss": 1.5109,
      "step": 37770
    },
    {
      "epoch": 2.014415635002625,
      "grad_norm": 0.2007877379655838,
      "learning_rate": 8.866666666666668e-06,
      "loss": 1.5275,
      "step": 37780
    },
    {
      "epoch": 2.014957576168137,
      "grad_norm": 0.16571199893951416,
      "learning_rate": 9.2e-06,
      "loss": 1.5206,
      "step": 37790
    },
    {
      "epoch": 2.0154995173336494,
      "grad_norm": 0.17200104892253876,
      "learning_rate": 9.533333333333334e-06,
      "loss": 1.5205,
      "step": 37800
    },
    {
      "epoch": 2.0160414584991617,
      "grad_norm": 0.1716819703578949,
      "learning_rate": 9.866666666666667e-06,
      "loss": 1.5087,
      "step": 37810
    },
    {
      "epoch": 2.016583399664674,
      "grad_norm": 0.21321678161621094,
      "learning_rate": 1.02e-05,
      "loss": 1.5171,
      "step": 37820
    },
    {
      "epoch": 2.017125340830186,
      "grad_norm": 0.16358011960983276,
      "learning_rate": 1.0533333333333335e-05,
      "loss": 1.5169,
      "step": 37830
    },
    {
      "epoch": 2.0176672819956982,
      "grad_norm": 0.16521596908569336,
      "learning_rate": 1.0866666666666667e-05,
      "loss": 1.5277,
      "step": 37840
    },
    {
      "epoch": 2.0179382525784546,
      "eval_loss": 2.476113796234131,
      "eval_runtime": 21.9762,
      "eval_samples_per_second": 227.519,
      "eval_steps_per_second": 1.229,
      "step": 37845
    },
    {
      "epoch": 2.0182092231612105,
      "grad_norm": 0.17164713144302368,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 1.5241,
      "step": 37850
    },
    {
      "epoch": 2.018751164326723,
      "grad_norm": 0.21223784983158112,
      "learning_rate": 1.1533333333333334e-05,
      "loss": 1.5176,
      "step": 37860
    },
    {
      "epoch": 2.019293105492235,
      "grad_norm": 0.17650015652179718,
      "learning_rate": 1.1866666666666668e-05,
      "loss": 1.5128,
      "step": 37870
    },
    {
      "epoch": 2.019835046657747,
      "grad_norm": 0.23301972448825836,
      "learning_rate": 1.22e-05,
      "loss": 1.5302,
      "step": 37880
    },
    {
      "epoch": 2.0203769878232594,
      "grad_norm": 0.16798456013202667,
      "learning_rate": 1.2533333333333332e-05,
      "loss": 1.5146,
      "step": 37890
    },
    {
      "epoch": 2.0209189289887717,
      "grad_norm": 0.1657019704580307,
      "learning_rate": 1.2866666666666668e-05,
      "loss": 1.523,
      "step": 37900
    },
    {
      "epoch": 2.021460870154284,
      "grad_norm": 0.1911483108997345,
      "learning_rate": 1.32e-05,
      "loss": 1.5141,
      "step": 37910
    },
    {
      "epoch": 2.0220028113197963,
      "grad_norm": 0.22926867008209229,
      "learning_rate": 1.3533333333333335e-05,
      "loss": 1.5156,
      "step": 37920
    },
    {
      "epoch": 2.022544752485308,
      "grad_norm": 0.25321751832962036,
      "learning_rate": 1.3866666666666667e-05,
      "loss": 1.5193,
      "step": 37930
    },
    {
      "epoch": 2.0226531407184107,
      "eval_loss": 2.470311403274536,
      "eval_runtime": 21.9756,
      "eval_samples_per_second": 227.525,
      "eval_steps_per_second": 1.229,
      "step": 37932
    },
    {
      "epoch": 2.0230866936508205,
      "grad_norm": 0.20465686917304993,
      "learning_rate": 1.42e-05,
      "loss": 1.514,
      "step": 37940
    },
    {
      "epoch": 2.023628634816333,
      "grad_norm": 0.20215226709842682,
      "learning_rate": 1.4533333333333335e-05,
      "loss": 1.5079,
      "step": 37950
    },
    {
      "epoch": 2.024170575981845,
      "grad_norm": 0.18972061574459076,
      "learning_rate": 1.4866666666666668e-05,
      "loss": 1.5137,
      "step": 37960
    },
    {
      "epoch": 2.024712517147357,
      "grad_norm": 0.18428048491477966,
      "learning_rate": 1.52e-05,
      "loss": 1.5206,
      "step": 37970
    },
    {
      "epoch": 2.0252544583128693,
      "grad_norm": 0.21038727462291718,
      "learning_rate": 1.5533333333333333e-05,
      "loss": 1.515,
      "step": 37980
    },
    {
      "epoch": 2.0257963994783816,
      "grad_norm": 0.20297273993492126,
      "learning_rate": 1.586666666666667e-05,
      "loss": 1.5159,
      "step": 37990
    },
    {
      "epoch": 2.026338340643894,
      "grad_norm": 0.16452449560165405,
      "learning_rate": 1.62e-05,
      "loss": 1.5092,
      "step": 38000
    },
    {
      "epoch": 2.0268802818094063,
      "grad_norm": 0.25832870602607727,
      "learning_rate": 1.6533333333333333e-05,
      "loss": 1.5097,
      "step": 38010
    },
    {
      "epoch": 2.027368028858367,
      "eval_loss": 2.469764232635498,
      "eval_runtime": 21.9777,
      "eval_samples_per_second": 227.503,
      "eval_steps_per_second": 1.229,
      "step": 38019
    },
    {
      "epoch": 2.027422222974918,
      "grad_norm": 0.17639535665512085,
      "learning_rate": 1.6866666666666666e-05,
      "loss": 1.52,
      "step": 38020
    },
    {
      "epoch": 2.0279641641404305,
      "grad_norm": 0.2649475336074829,
      "learning_rate": 1.7199999999999998e-05,
      "loss": 1.5037,
      "step": 38030
    },
    {
      "epoch": 2.0285061053059428,
      "grad_norm": 0.19192355871200562,
      "learning_rate": 1.7533333333333334e-05,
      "loss": 1.501,
      "step": 38040
    },
    {
      "epoch": 2.029048046471455,
      "grad_norm": 0.19619537889957428,
      "learning_rate": 1.7866666666666666e-05,
      "loss": 1.5126,
      "step": 38050
    },
    {
      "epoch": 2.029589987636967,
      "grad_norm": 0.1901368945837021,
      "learning_rate": 1.8200000000000002e-05,
      "loss": 1.525,
      "step": 38060
    },
    {
      "epoch": 2.0301319288024793,
      "grad_norm": 0.1913931667804718,
      "learning_rate": 1.8533333333333334e-05,
      "loss": 1.5136,
      "step": 38070
    },
    {
      "epoch": 2.0306738699679916,
      "grad_norm": 0.1890285611152649,
      "learning_rate": 1.886666666666667e-05,
      "loss": 1.5198,
      "step": 38080
    },
    {
      "epoch": 2.031215811133504,
      "grad_norm": 0.19151568412780762,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 1.5177,
      "step": 38090
    },
    {
      "epoch": 2.0317577522990162,
      "grad_norm": 0.16688157618045807,
      "learning_rate": 1.9533333333333335e-05,
      "loss": 1.5084,
      "step": 38100
    },
    {
      "epoch": 2.0320829169983234,
      "eval_loss": 2.4747180938720703,
      "eval_runtime": 21.9798,
      "eval_samples_per_second": 227.481,
      "eval_steps_per_second": 1.228,
      "step": 38106
    },
    {
      "epoch": 2.032299693464528,
      "grad_norm": 0.21504206955432892,
      "learning_rate": 1.9866666666666667e-05,
      "loss": 1.5145,
      "step": 38110
    },
    {
      "epoch": 2.0328416346300404,
      "grad_norm": 0.19307386875152588,
      "learning_rate": 2.0200000000000003e-05,
      "loss": 1.514,
      "step": 38120
    },
    {
      "epoch": 2.0333835757955527,
      "grad_norm": 0.17073208093643188,
      "learning_rate": 2.0533333333333336e-05,
      "loss": 1.5158,
      "step": 38130
    },
    {
      "epoch": 2.033925516961065,
      "grad_norm": 0.22306814789772034,
      "learning_rate": 2.0866666666666668e-05,
      "loss": 1.5055,
      "step": 38140
    },
    {
      "epoch": 2.034467458126577,
      "grad_norm": 0.170424222946167,
      "learning_rate": 2.12e-05,
      "loss": 1.5103,
      "step": 38150
    },
    {
      "epoch": 2.0350093992920892,
      "grad_norm": 0.18648818135261536,
      "learning_rate": 2.1533333333333333e-05,
      "loss": 1.5109,
      "step": 38160
    },
    {
      "epoch": 2.0355513404576016,
      "grad_norm": 0.18736661970615387,
      "learning_rate": 2.186666666666667e-05,
      "loss": 1.507,
      "step": 38170
    },
    {
      "epoch": 2.036093281623114,
      "grad_norm": 0.20201878249645233,
      "learning_rate": 2.22e-05,
      "loss": 1.5157,
      "step": 38180
    },
    {
      "epoch": 2.036635222788626,
      "grad_norm": 0.19277946650981903,
      "learning_rate": 2.2533333333333333e-05,
      "loss": 1.5152,
      "step": 38190
    },
    {
      "epoch": 2.0367978051382796,
      "eval_loss": 2.474329710006714,
      "eval_runtime": 21.9764,
      "eval_samples_per_second": 227.517,
      "eval_steps_per_second": 1.229,
      "step": 38193
    },
    {
      "epoch": 2.037177163954138,
      "grad_norm": 0.1907971054315567,
      "learning_rate": 2.2866666666666666e-05,
      "loss": 1.5019,
      "step": 38200
    },
    {
      "epoch": 2.0377191051196504,
      "grad_norm": 0.2295379936695099,
      "learning_rate": 2.32e-05,
      "loss": 1.5046,
      "step": 38210
    },
    {
      "epoch": 2.0382610462851627,
      "grad_norm": 0.18477369844913483,
      "learning_rate": 2.3533333333333334e-05,
      "loss": 1.5057,
      "step": 38220
    },
    {
      "epoch": 2.038802987450675,
      "grad_norm": 0.19044053554534912,
      "learning_rate": 2.3866666666666666e-05,
      "loss": 1.5217,
      "step": 38230
    },
    {
      "epoch": 2.0393449286161873,
      "grad_norm": 0.18540959060192108,
      "learning_rate": 2.4200000000000002e-05,
      "loss": 1.52,
      "step": 38240
    },
    {
      "epoch": 2.039886869781699,
      "grad_norm": 0.23876672983169556,
      "learning_rate": 2.4533333333333334e-05,
      "loss": 1.5129,
      "step": 38250
    },
    {
      "epoch": 2.0404288109472115,
      "grad_norm": 0.19681920111179352,
      "learning_rate": 2.486666666666667e-05,
      "loss": 1.502,
      "step": 38260
    },
    {
      "epoch": 2.040970752112724,
      "grad_norm": 0.17679736018180847,
      "learning_rate": 2.5200000000000003e-05,
      "loss": 1.5122,
      "step": 38270
    },
    {
      "epoch": 2.041512693278236,
      "grad_norm": 0.18017658591270447,
      "learning_rate": 2.553333333333334e-05,
      "loss": 1.5145,
      "step": 38280
    },
    {
      "epoch": 2.041512693278236,
      "eval_loss": 2.4778501987457275,
      "eval_runtime": 22.0279,
      "eval_samples_per_second": 226.985,
      "eval_steps_per_second": 1.226,
      "step": 38280
    },
    {
      "epoch": 2.042054634443748,
      "grad_norm": 0.17411205172538757,
      "learning_rate": 2.5866666666666667e-05,
      "loss": 1.5064,
      "step": 38290
    },
    {
      "epoch": 2.0425965756092603,
      "grad_norm": 0.18675434589385986,
      "learning_rate": 2.6200000000000003e-05,
      "loss": 1.5071,
      "step": 38300
    },
    {
      "epoch": 2.0431385167747727,
      "grad_norm": 0.17742466926574707,
      "learning_rate": 2.6533333333333332e-05,
      "loss": 1.5147,
      "step": 38310
    },
    {
      "epoch": 2.043680457940285,
      "grad_norm": 0.2024219036102295,
      "learning_rate": 2.6866666666666668e-05,
      "loss": 1.5201,
      "step": 38320
    },
    {
      "epoch": 2.0442223991057973,
      "grad_norm": 0.19301582872867584,
      "learning_rate": 2.7200000000000004e-05,
      "loss": 1.5131,
      "step": 38330
    },
    {
      "epoch": 2.044764340271309,
      "grad_norm": 0.186081200838089,
      "learning_rate": 2.7533333333333333e-05,
      "loss": 1.5113,
      "step": 38340
    },
    {
      "epoch": 2.0453062814368215,
      "grad_norm": 0.2549179792404175,
      "learning_rate": 2.786666666666667e-05,
      "loss": 1.5133,
      "step": 38350
    },
    {
      "epoch": 2.045848222602334,
      "grad_norm": 0.23150603473186493,
      "learning_rate": 2.8199999999999998e-05,
      "loss": 1.5087,
      "step": 38360
    },
    {
      "epoch": 2.0462275814181923,
      "eval_loss": 2.479970932006836,
      "eval_runtime": 21.9756,
      "eval_samples_per_second": 227.525,
      "eval_steps_per_second": 1.229,
      "step": 38367
    },
    {
      "epoch": 2.046390163767846,
      "grad_norm": 0.199720561504364,
      "learning_rate": 2.8533333333333333e-05,
      "loss": 1.5038,
      "step": 38370
    },
    {
      "epoch": 2.046932104933358,
      "grad_norm": 0.16712939739227295,
      "learning_rate": 2.886666666666667e-05,
      "loss": 1.5047,
      "step": 38380
    },
    {
      "epoch": 2.0474740460988703,
      "grad_norm": 0.17325983941555023,
      "learning_rate": 2.9199999999999998e-05,
      "loss": 1.517,
      "step": 38390
    },
    {
      "epoch": 2.0480159872643826,
      "grad_norm": 0.19841617345809937,
      "learning_rate": 2.9533333333333334e-05,
      "loss": 1.5113,
      "step": 38400
    },
    {
      "epoch": 2.048557928429895,
      "grad_norm": 0.21291446685791016,
      "learning_rate": 2.986666666666667e-05,
      "loss": 1.5056,
      "step": 38410
    },
    {
      "epoch": 2.0490998695954072,
      "grad_norm": 0.33324041962623596,
      "learning_rate": 3.02e-05,
      "loss": 1.4991,
      "step": 38420
    },
    {
      "epoch": 2.049641810760919,
      "grad_norm": 0.19952350854873657,
      "learning_rate": 3.0533333333333335e-05,
      "loss": 1.5244,
      "step": 38430
    },
    {
      "epoch": 2.0501837519264314,
      "grad_norm": 0.19988487660884857,
      "learning_rate": 3.086666666666667e-05,
      "loss": 1.509,
      "step": 38440
    },
    {
      "epoch": 2.0507256930919437,
      "grad_norm": 0.3386477530002594,
      "learning_rate": 3.12e-05,
      "loss": 1.5031,
      "step": 38450
    },
    {
      "epoch": 2.0509424695581484,
      "eval_loss": 2.480910301208496,
      "eval_runtime": 21.9771,
      "eval_samples_per_second": 227.51,
      "eval_steps_per_second": 1.229,
      "step": 38454
    },
    {
      "epoch": 2.051267634257456,
      "grad_norm": 0.23942768573760986,
      "learning_rate": 3.153333333333334e-05,
      "loss": 1.5143,
      "step": 38460
    },
    {
      "epoch": 2.051809575422968,
      "grad_norm": 0.2345375418663025,
      "learning_rate": 3.1866666666666664e-05,
      "loss": 1.5095,
      "step": 38470
    },
    {
      "epoch": 2.0523515165884803,
      "grad_norm": 0.23772574961185455,
      "learning_rate": 3.2200000000000003e-05,
      "loss": 1.5036,
      "step": 38480
    },
    {
      "epoch": 2.0528934577539926,
      "grad_norm": 0.21411636471748352,
      "learning_rate": 3.253333333333333e-05,
      "loss": 1.5111,
      "step": 38490
    },
    {
      "epoch": 2.053435398919505,
      "grad_norm": 0.17845569550991058,
      "learning_rate": 3.286666666666667e-05,
      "loss": 1.5151,
      "step": 38500
    },
    {
      "epoch": 2.053977340085017,
      "grad_norm": 0.23491686582565308,
      "learning_rate": 3.32e-05,
      "loss": 1.5025,
      "step": 38510
    },
    {
      "epoch": 2.054519281250529,
      "grad_norm": 0.2281605452299118,
      "learning_rate": 3.353333333333333e-05,
      "loss": 1.5121,
      "step": 38520
    },
    {
      "epoch": 2.0550612224160414,
      "grad_norm": 0.19572101533412933,
      "learning_rate": 3.3866666666666665e-05,
      "loss": 1.5169,
      "step": 38530
    },
    {
      "epoch": 2.0556031635815537,
      "grad_norm": 0.2043548971414566,
      "learning_rate": 3.4200000000000005e-05,
      "loss": 1.507,
      "step": 38540
    },
    {
      "epoch": 2.055657357698105,
      "eval_loss": 2.4816620349884033,
      "eval_runtime": 22.0156,
      "eval_samples_per_second": 227.111,
      "eval_steps_per_second": 1.226,
      "step": 38541
    },
    {
      "epoch": 2.056145104747066,
      "grad_norm": 0.3473966717720032,
      "learning_rate": 3.453333333333334e-05,
      "loss": 1.503,
      "step": 38550
    },
    {
      "epoch": 2.056687045912578,
      "grad_norm": 0.17651808261871338,
      "learning_rate": 3.486666666666667e-05,
      "loss": 1.5191,
      "step": 38560
    },
    {
      "epoch": 2.05722898707809,
      "grad_norm": 0.31023022532463074,
      "learning_rate": 3.52e-05,
      "loss": 1.5095,
      "step": 38570
    },
    {
      "epoch": 2.0577709282436025,
      "grad_norm": 0.22835835814476013,
      "learning_rate": 3.5533333333333334e-05,
      "loss": 1.496,
      "step": 38580
    },
    {
      "epoch": 2.058312869409115,
      "grad_norm": 0.28630900382995605,
      "learning_rate": 3.586666666666667e-05,
      "loss": 1.5129,
      "step": 38590
    },
    {
      "epoch": 2.058854810574627,
      "grad_norm": 0.2917551100254059,
      "learning_rate": 3.62e-05,
      "loss": 1.5131,
      "step": 38600
    },
    {
      "epoch": 2.059396751740139,
      "grad_norm": 0.23212890326976776,
      "learning_rate": 3.653333333333334e-05,
      "loss": 1.5075,
      "step": 38610
    },
    {
      "epoch": 2.0599386929056513,
      "grad_norm": 0.3044704496860504,
      "learning_rate": 3.6866666666666664e-05,
      "loss": 1.5153,
      "step": 38620
    },
    {
      "epoch": 2.060372245838061,
      "eval_loss": 2.466387987136841,
      "eval_runtime": 21.9823,
      "eval_samples_per_second": 227.456,
      "eval_steps_per_second": 1.228,
      "step": 38628
    },
    {
      "epoch": 2.0604806340711637,
      "grad_norm": 0.1840297132730484,
      "learning_rate": 3.72e-05,
      "loss": 1.5098,
      "step": 38630
    },
    {
      "epoch": 2.061022575236676,
      "grad_norm": 0.20662543177604675,
      "learning_rate": 3.7533333333333335e-05,
      "loss": 1.5006,
      "step": 38640
    },
    {
      "epoch": 2.0615645164021883,
      "grad_norm": 0.227204829454422,
      "learning_rate": 3.786666666666667e-05,
      "loss": 1.5052,
      "step": 38650
    },
    {
      "epoch": 2.0621064575677,
      "grad_norm": 0.377437561750412,
      "learning_rate": 3.82e-05,
      "loss": 1.5002,
      "step": 38660
    },
    {
      "epoch": 2.0626483987332125,
      "grad_norm": 0.3552536964416504,
      "learning_rate": 3.853333333333334e-05,
      "loss": 1.5104,
      "step": 38670
    },
    {
      "epoch": 2.063190339898725,
      "grad_norm": 0.31461700797080994,
      "learning_rate": 3.8866666666666665e-05,
      "loss": 1.5048,
      "step": 38680
    },
    {
      "epoch": 2.063732281064237,
      "grad_norm": 0.20412826538085938,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 1.5058,
      "step": 38690
    },
    {
      "epoch": 2.064274222229749,
      "grad_norm": 0.22955581545829773,
      "learning_rate": 3.9533333333333337e-05,
      "loss": 1.5037,
      "step": 38700
    },
    {
      "epoch": 2.0648161633952613,
      "grad_norm": 0.2234533727169037,
      "learning_rate": 3.986666666666667e-05,
      "loss": 1.5108,
      "step": 38710
    },
    {
      "epoch": 2.0650871339780177,
      "eval_loss": 2.4609971046447754,
      "eval_runtime": 21.9738,
      "eval_samples_per_second": 227.543,
      "eval_steps_per_second": 1.229,
      "step": 38715
    },
    {
      "epoch": 2.0653581045607736,
      "grad_norm": 0.18891902267932892,
      "learning_rate": 4.02e-05,
      "loss": 1.504,
      "step": 38720
    },
    {
      "epoch": 2.065900045726286,
      "grad_norm": 0.21943628787994385,
      "learning_rate": 4.0533333333333334e-05,
      "loss": 1.5037,
      "step": 38730
    },
    {
      "epoch": 2.0664419868917983,
      "grad_norm": 0.24250586330890656,
      "learning_rate": 4.086666666666667e-05,
      "loss": 1.5022,
      "step": 38740
    },
    {
      "epoch": 2.06698392805731,
      "grad_norm": 0.2112322300672531,
      "learning_rate": 4.12e-05,
      "loss": 1.5052,
      "step": 38750
    },
    {
      "epoch": 2.0675258692228224,
      "grad_norm": 0.24614650011062622,
      "learning_rate": 4.153333333333334e-05,
      "loss": 1.5012,
      "step": 38760
    },
    {
      "epoch": 2.0680678103883348,
      "grad_norm": 0.25870656967163086,
      "learning_rate": 4.186666666666667e-05,
      "loss": 1.5078,
      "step": 38770
    },
    {
      "epoch": 2.068609751553847,
      "grad_norm": 0.18944492936134338,
      "learning_rate": 4.22e-05,
      "loss": 1.5,
      "step": 38780
    },
    {
      "epoch": 2.069151692719359,
      "grad_norm": 0.2570621073246002,
      "learning_rate": 4.2533333333333335e-05,
      "loss": 1.5076,
      "step": 38790
    },
    {
      "epoch": 2.0696936338848713,
      "grad_norm": 0.19097241759300232,
      "learning_rate": 4.286666666666667e-05,
      "loss": 1.5103,
      "step": 38800
    },
    {
      "epoch": 2.069802022117974,
      "eval_loss": 2.472243309020996,
      "eval_runtime": 21.9768,
      "eval_samples_per_second": 227.513,
      "eval_steps_per_second": 1.229,
      "step": 38802
    },
    {
      "epoch": 2.0702355750503836,
      "grad_norm": 0.2385208159685135,
      "learning_rate": 4.32e-05,
      "loss": 1.5022,
      "step": 38810
    },
    {
      "epoch": 2.070777516215896,
      "grad_norm": 0.3327140212059021,
      "learning_rate": 4.353333333333334e-05,
      "loss": 1.5071,
      "step": 38820
    },
    {
      "epoch": 2.071319457381408,
      "grad_norm": 0.36782822012901306,
      "learning_rate": 4.3866666666666665e-05,
      "loss": 1.5097,
      "step": 38830
    },
    {
      "epoch": 2.07186139854692,
      "grad_norm": 0.35685569047927856,
      "learning_rate": 4.4200000000000004e-05,
      "loss": 1.5072,
      "step": 38840
    },
    {
      "epoch": 2.0724033397124324,
      "grad_norm": 0.20131441950798035,
      "learning_rate": 4.4533333333333336e-05,
      "loss": 1.4922,
      "step": 38850
    },
    {
      "epoch": 2.0729452808779447,
      "grad_norm": 0.31305959820747375,
      "learning_rate": 4.486666666666667e-05,
      "loss": 1.509,
      "step": 38860
    },
    {
      "epoch": 2.073487222043457,
      "grad_norm": 0.2946673333644867,
      "learning_rate": 4.52e-05,
      "loss": 1.5089,
      "step": 38870
    },
    {
      "epoch": 2.0740291632089694,
      "grad_norm": 0.29778286814689636,
      "learning_rate": 4.553333333333333e-05,
      "loss": 1.4924,
      "step": 38880
    },
    {
      "epoch": 2.07451691025793,
      "eval_loss": 2.4727067947387695,
      "eval_runtime": 21.9813,
      "eval_samples_per_second": 227.466,
      "eval_steps_per_second": 1.228,
      "step": 38889
    },
    {
      "epoch": 2.0745711043744812,
      "grad_norm": 0.2734017074108124,
      "learning_rate": 4.5866666666666666e-05,
      "loss": 1.5117,
      "step": 38890
    },
    {
      "epoch": 2.0751130455399935,
      "grad_norm": 0.1901269257068634,
      "learning_rate": 4.6200000000000005e-05,
      "loss": 1.503,
      "step": 38900
    },
    {
      "epoch": 2.075654986705506,
      "grad_norm": 0.2383534014225006,
      "learning_rate": 4.653333333333334e-05,
      "loss": 1.4965,
      "step": 38910
    },
    {
      "epoch": 2.076196927871018,
      "grad_norm": 0.27332353591918945,
      "learning_rate": 4.686666666666667e-05,
      "loss": 1.5121,
      "step": 38920
    },
    {
      "epoch": 2.07673886903653,
      "grad_norm": 0.25885239243507385,
      "learning_rate": 4.72e-05,
      "loss": 1.5025,
      "step": 38930
    },
    {
      "epoch": 2.0772808102020424,
      "grad_norm": 0.19907046854496002,
      "learning_rate": 4.7533333333333334e-05,
      "loss": 1.5126,
      "step": 38940
    },
    {
      "epoch": 2.0778227513675547,
      "grad_norm": 0.19673128426074982,
      "learning_rate": 4.7866666666666674e-05,
      "loss": 1.5149,
      "step": 38950
    },
    {
      "epoch": 2.078364692533067,
      "grad_norm": 0.3096723258495331,
      "learning_rate": 4.82e-05,
      "loss": 1.509,
      "step": 38960
    },
    {
      "epoch": 2.0789066336985793,
      "grad_norm": 0.22866037487983704,
      "learning_rate": 4.853333333333334e-05,
      "loss": 1.5085,
      "step": 38970
    },
    {
      "epoch": 2.0792317983978865,
      "eval_loss": 2.4707653522491455,
      "eval_runtime": 21.9775,
      "eval_samples_per_second": 227.506,
      "eval_steps_per_second": 1.229,
      "step": 38976
    },
    {
      "epoch": 2.079448574864091,
      "grad_norm": 0.19826719164848328,
      "learning_rate": 4.886666666666667e-05,
      "loss": 1.5061,
      "step": 38980
    },
    {
      "epoch": 2.0799905160296035,
      "grad_norm": 0.3995167016983032,
      "learning_rate": 4.92e-05,
      "loss": 1.4995,
      "step": 38990
    },
    {
      "epoch": 2.080532457195116,
      "grad_norm": 0.3492770791053772,
      "learning_rate": 4.9533333333333336e-05,
      "loss": 1.5092,
      "step": 39000
    },
    {
      "epoch": 2.081074398360628,
      "grad_norm": 0.248097226023674,
      "learning_rate": 4.986666666666667e-05,
      "loss": 1.5146,
      "step": 39010
    },
    {
      "epoch": 2.08161633952614,
      "grad_norm": 0.3634330630302429,
      "learning_rate": 5.02e-05,
      "loss": 1.5122,
      "step": 39020
    },
    {
      "epoch": 2.0821582806916523,
      "grad_norm": 0.30074241757392883,
      "learning_rate": 5.053333333333333e-05,
      "loss": 1.5094,
      "step": 39030
    },
    {
      "epoch": 2.0827002218571646,
      "grad_norm": 0.19916889071464539,
      "learning_rate": 5.086666666666667e-05,
      "loss": 1.5129,
      "step": 39040
    },
    {
      "epoch": 2.083242163022677,
      "grad_norm": 0.19030417501926422,
      "learning_rate": 5.1200000000000004e-05,
      "loss": 1.5048,
      "step": 39050
    },
    {
      "epoch": 2.0837841041881893,
      "grad_norm": 0.2428630143404007,
      "learning_rate": 5.153333333333333e-05,
      "loss": 1.5089,
      "step": 39060
    },
    {
      "epoch": 2.0839466865378427,
      "eval_loss": 2.4720537662506104,
      "eval_runtime": 21.9794,
      "eval_samples_per_second": 227.486,
      "eval_steps_per_second": 1.228,
      "step": 39063
    },
    {
      "epoch": 2.084326045353701,
      "grad_norm": 0.2563911974430084,
      "learning_rate": 5.1866666666666676e-05,
      "loss": 1.5078,
      "step": 39070
    },
    {
      "epoch": 2.0848679865192135,
      "grad_norm": 0.20653724670410156,
      "learning_rate": 5.22e-05,
      "loss": 1.4982,
      "step": 39080
    },
    {
      "epoch": 2.0854099276847258,
      "grad_norm": 0.29985037446022034,
      "learning_rate": 5.2533333333333334e-05,
      "loss": 1.5188,
      "step": 39090
    },
    {
      "epoch": 2.085951868850238,
      "grad_norm": 0.2325274795293808,
      "learning_rate": 5.2866666666666666e-05,
      "loss": 1.4888,
      "step": 39100
    },
    {
      "epoch": 2.08649381001575,
      "grad_norm": 0.35616472363471985,
      "learning_rate": 5.3200000000000006e-05,
      "loss": 1.503,
      "step": 39110
    },
    {
      "epoch": 2.0870357511812623,
      "grad_norm": 0.23784513771533966,
      "learning_rate": 5.353333333333334e-05,
      "loss": 1.4995,
      "step": 39120
    },
    {
      "epoch": 2.0875776923467746,
      "grad_norm": 0.2512984275817871,
      "learning_rate": 5.3866666666666664e-05,
      "loss": 1.4988,
      "step": 39130
    },
    {
      "epoch": 2.088119633512287,
      "grad_norm": 0.31452298164367676,
      "learning_rate": 5.420000000000001e-05,
      "loss": 1.5064,
      "step": 39140
    },
    {
      "epoch": 2.0886615746777992,
      "grad_norm": 0.2067795991897583,
      "learning_rate": 5.4533333333333335e-05,
      "loss": 1.503,
      "step": 39150
    },
    {
      "epoch": 2.0886615746777992,
      "eval_loss": 2.478317975997925,
      "eval_runtime": 21.954,
      "eval_samples_per_second": 227.749,
      "eval_steps_per_second": 1.23,
      "step": 39150
    },
    {
      "epoch": 2.089203515843311,
      "grad_norm": 0.2181154042482376,
      "learning_rate": 5.486666666666667e-05,
      "loss": 1.5082,
      "step": 39160
    },
    {
      "epoch": 2.0897454570088234,
      "grad_norm": 0.25218093395233154,
      "learning_rate": 5.520000000000001e-05,
      "loss": 1.4949,
      "step": 39170
    },
    {
      "epoch": 2.0902873981743357,
      "grad_norm": 0.26879268884658813,
      "learning_rate": 5.553333333333334e-05,
      "loss": 1.5169,
      "step": 39180
    },
    {
      "epoch": 2.090829339339848,
      "grad_norm": 0.42360979318618774,
      "learning_rate": 5.5866666666666665e-05,
      "loss": 1.5001,
      "step": 39190
    },
    {
      "epoch": 2.09137128050536,
      "grad_norm": 0.3156111538410187,
      "learning_rate": 5.620000000000001e-05,
      "loss": 1.4959,
      "step": 39200
    },
    {
      "epoch": 2.0919132216708722,
      "grad_norm": 0.3810129761695862,
      "learning_rate": 5.6533333333333336e-05,
      "loss": 1.5044,
      "step": 39210
    },
    {
      "epoch": 2.0924551628363846,
      "grad_norm": 0.2980895936489105,
      "learning_rate": 5.686666666666667e-05,
      "loss": 1.509,
      "step": 39220
    },
    {
      "epoch": 2.092997104001897,
      "grad_norm": 0.2537708878517151,
      "learning_rate": 5.72e-05,
      "loss": 1.502,
      "step": 39230
    },
    {
      "epoch": 2.0933764628177554,
      "eval_loss": 2.486907482147217,
      "eval_runtime": 21.9802,
      "eval_samples_per_second": 227.478,
      "eval_steps_per_second": 1.228,
      "step": 39237
    },
    {
      "epoch": 2.093539045167409,
      "grad_norm": 0.2361438274383545,
      "learning_rate": 5.753333333333334e-05,
      "loss": 1.5046,
      "step": 39240
    },
    {
      "epoch": 2.094080986332921,
      "grad_norm": 0.3751475214958191,
      "learning_rate": 5.7866666666666666e-05,
      "loss": 1.5045,
      "step": 39250
    },
    {
      "epoch": 2.0946229274984334,
      "grad_norm": 0.28046801686286926,
      "learning_rate": 5.82e-05,
      "loss": 1.5143,
      "step": 39260
    },
    {
      "epoch": 2.0951648686639457,
      "grad_norm": 0.3462556302547455,
      "learning_rate": 5.853333333333334e-05,
      "loss": 1.5155,
      "step": 39270
    },
    {
      "epoch": 2.095706809829458,
      "grad_norm": 0.4051414728164673,
      "learning_rate": 5.886666666666667e-05,
      "loss": 1.5008,
      "step": 39280
    },
    {
      "epoch": 2.0962487509949703,
      "grad_norm": 0.348838746547699,
      "learning_rate": 5.92e-05,
      "loss": 1.5027,
      "step": 39290
    },
    {
      "epoch": 2.096790692160482,
      "grad_norm": 0.26631638407707214,
      "learning_rate": 5.953333333333334e-05,
      "loss": 1.5169,
      "step": 39300
    },
    {
      "epoch": 2.0973326333259945,
      "grad_norm": 0.20438840985298157,
      "learning_rate": 5.9866666666666674e-05,
      "loss": 1.5017,
      "step": 39310
    },
    {
      "epoch": 2.097874574491507,
      "grad_norm": 0.23914705216884613,
      "learning_rate": 6.02e-05,
      "loss": 1.5111,
      "step": 39320
    },
    {
      "epoch": 2.0980913509577115,
      "eval_loss": 2.485321521759033,
      "eval_runtime": 21.983,
      "eval_samples_per_second": 227.448,
      "eval_steps_per_second": 1.228,
      "step": 39324
    },
    {
      "epoch": 2.098416515657019,
      "grad_norm": 0.4190734028816223,
      "learning_rate": 6.053333333333333e-05,
      "loss": 1.5052,
      "step": 39330
    },
    {
      "epoch": 2.098958456822531,
      "grad_norm": 0.2981734871864319,
      "learning_rate": 6.086666666666667e-05,
      "loss": 1.5005,
      "step": 39340
    },
    {
      "epoch": 2.0995003979880433,
      "grad_norm": 0.28054559230804443,
      "learning_rate": 6.12e-05,
      "loss": 1.4988,
      "step": 39350
    },
    {
      "epoch": 2.1000423391535556,
      "grad_norm": 0.4990901052951813,
      "learning_rate": 6.153333333333333e-05,
      "loss": 1.4869,
      "step": 39360
    },
    {
      "epoch": 2.100584280319068,
      "grad_norm": 0.30956992506980896,
      "learning_rate": 6.186666666666668e-05,
      "loss": 1.4983,
      "step": 39370
    },
    {
      "epoch": 2.1011262214845803,
      "grad_norm": 0.25511854887008667,
      "learning_rate": 6.220000000000001e-05,
      "loss": 1.5014,
      "step": 39380
    },
    {
      "epoch": 2.101668162650092,
      "grad_norm": 0.5350297689437866,
      "learning_rate": 6.253333333333333e-05,
      "loss": 1.4979,
      "step": 39390
    },
    {
      "epoch": 2.1022101038156045,
      "grad_norm": 0.20907965302467346,
      "learning_rate": 6.286666666666667e-05,
      "loss": 1.5125,
      "step": 39400
    },
    {
      "epoch": 2.102752044981117,
      "grad_norm": 0.2567403018474579,
      "learning_rate": 6.32e-05,
      "loss": 1.4964,
      "step": 39410
    },
    {
      "epoch": 2.102806239097668,
      "eval_loss": 2.490506887435913,
      "eval_runtime": 21.9797,
      "eval_samples_per_second": 227.483,
      "eval_steps_per_second": 1.228,
      "step": 39411
    },
    {
      "epoch": 2.103293986146629,
      "grad_norm": 0.25373193621635437,
      "learning_rate": 6.353333333333334e-05,
      "loss": 1.5058,
      "step": 39420
    },
    {
      "epoch": 2.103835927312141,
      "grad_norm": 0.20478834211826324,
      "learning_rate": 6.386666666666667e-05,
      "loss": 1.4987,
      "step": 39430
    },
    {
      "epoch": 2.1043778684776533,
      "grad_norm": 0.2046637237071991,
      "learning_rate": 6.42e-05,
      "loss": 1.51,
      "step": 39440
    },
    {
      "epoch": 2.1049198096431656,
      "grad_norm": 0.2921828329563141,
      "learning_rate": 6.453333333333333e-05,
      "loss": 1.505,
      "step": 39450
    },
    {
      "epoch": 2.105461750808678,
      "grad_norm": 0.22420616447925568,
      "learning_rate": 6.486666666666667e-05,
      "loss": 1.5004,
      "step": 39460
    },
    {
      "epoch": 2.1060036919741902,
      "grad_norm": 0.39737066626548767,
      "learning_rate": 6.52e-05,
      "loss": 1.5068,
      "step": 39470
    },
    {
      "epoch": 2.106545633139702,
      "grad_norm": 0.21135340631008148,
      "learning_rate": 6.553333333333333e-05,
      "loss": 1.4962,
      "step": 39480
    },
    {
      "epoch": 2.1070875743052144,
      "grad_norm": 0.2920219600200653,
      "learning_rate": 6.586666666666666e-05,
      "loss": 1.5102,
      "step": 39490
    },
    {
      "epoch": 2.107521127237624,
      "eval_loss": 2.477649450302124,
      "eval_runtime": 22.0366,
      "eval_samples_per_second": 226.895,
      "eval_steps_per_second": 1.225,
      "step": 39498
    },
    {
      "epoch": 2.1076295154707267,
      "grad_norm": 0.34153836965560913,
      "learning_rate": 6.620000000000001e-05,
      "loss": 1.5078,
      "step": 39500
    },
    {
      "epoch": 2.108171456636239,
      "grad_norm": 0.27279841899871826,
      "learning_rate": 6.653333333333334e-05,
      "loss": 1.505,
      "step": 39510
    },
    {
      "epoch": 2.108713397801751,
      "grad_norm": 0.2870520055294037,
      "learning_rate": 6.686666666666666e-05,
      "loss": 1.5043,
      "step": 39520
    },
    {
      "epoch": 2.1092553389672632,
      "grad_norm": 0.34095126390457153,
      "learning_rate": 6.720000000000001e-05,
      "loss": 1.5083,
      "step": 39530
    },
    {
      "epoch": 2.1097972801327756,
      "grad_norm": 0.2083679437637329,
      "learning_rate": 6.753333333333334e-05,
      "loss": 1.4983,
      "step": 39540
    },
    {
      "epoch": 2.110339221298288,
      "grad_norm": 0.281099796295166,
      "learning_rate": 6.786666666666667e-05,
      "loss": 1.5089,
      "step": 39550
    },
    {
      "epoch": 2.1108811624638,
      "grad_norm": 0.29480496048927307,
      "learning_rate": 6.82e-05,
      "loss": 1.5133,
      "step": 39560
    },
    {
      "epoch": 2.111423103629312,
      "grad_norm": 0.38620585203170776,
      "learning_rate": 6.853333333333334e-05,
      "loss": 1.5034,
      "step": 39570
    },
    {
      "epoch": 2.1119650447948244,
      "grad_norm": 0.41479256749153137,
      "learning_rate": 6.886666666666667e-05,
      "loss": 1.5107,
      "step": 39580
    },
    {
      "epoch": 2.1122360153775808,
      "eval_loss": 2.4839091300964355,
      "eval_runtime": 22.0638,
      "eval_samples_per_second": 226.615,
      "eval_steps_per_second": 1.224,
      "step": 39585
    },
    {
      "epoch": 2.1125069859603367,
      "grad_norm": 0.3035687804222107,
      "learning_rate": 6.92e-05,
      "loss": 1.4977,
      "step": 39590
    },
    {
      "epoch": 2.113048927125849,
      "grad_norm": 0.34999310970306396,
      "learning_rate": 6.953333333333333e-05,
      "loss": 1.5053,
      "step": 39600
    },
    {
      "epoch": 2.113590868291361,
      "grad_norm": 0.27466028928756714,
      "learning_rate": 6.986666666666667e-05,
      "loss": 1.5103,
      "step": 39610
    },
    {
      "epoch": 2.114132809456873,
      "grad_norm": 0.4131866693496704,
      "learning_rate": 7.02e-05,
      "loss": 1.4978,
      "step": 39620
    },
    {
      "epoch": 2.1146747506223855,
      "grad_norm": 0.3529013991355896,
      "learning_rate": 7.053333333333334e-05,
      "loss": 1.5053,
      "step": 39630
    },
    {
      "epoch": 2.115216691787898,
      "grad_norm": 0.30995792150497437,
      "learning_rate": 7.086666666666666e-05,
      "loss": 1.513,
      "step": 39640
    },
    {
      "epoch": 2.11575863295341,
      "grad_norm": 0.3839752972126007,
      "learning_rate": 7.12e-05,
      "loss": 1.505,
      "step": 39650
    },
    {
      "epoch": 2.116300574118922,
      "grad_norm": 0.2367662787437439,
      "learning_rate": 7.153333333333334e-05,
      "loss": 1.5036,
      "step": 39660
    },
    {
      "epoch": 2.1168425152844343,
      "grad_norm": 0.22004197537899017,
      "learning_rate": 7.186666666666667e-05,
      "loss": 1.5015,
      "step": 39670
    },
    {
      "epoch": 2.116950903517537,
      "eval_loss": 2.4732017517089844,
      "eval_runtime": 21.9785,
      "eval_samples_per_second": 227.495,
      "eval_steps_per_second": 1.228,
      "step": 39672
    },
    {
      "epoch": 2.1173844564499467,
      "grad_norm": 0.2898196280002594,
      "learning_rate": 7.22e-05,
      "loss": 1.5044,
      "step": 39680
    },
    {
      "epoch": 2.117926397615459,
      "grad_norm": 0.30908989906311035,
      "learning_rate": 7.253333333333334e-05,
      "loss": 1.5071,
      "step": 39690
    },
    {
      "epoch": 2.1184683387809713,
      "grad_norm": 0.6186519265174866,
      "learning_rate": 7.286666666666667e-05,
      "loss": 1.522,
      "step": 39700
    },
    {
      "epoch": 2.119010279946483,
      "grad_norm": 0.47445744276046753,
      "learning_rate": 7.32e-05,
      "loss": 1.4992,
      "step": 39710
    },
    {
      "epoch": 2.1195522211119955,
      "grad_norm": 0.3463587760925293,
      "learning_rate": 7.353333333333334e-05,
      "loss": 1.512,
      "step": 39720
    },
    {
      "epoch": 2.120094162277508,
      "grad_norm": 0.3156580626964569,
      "learning_rate": 7.386666666666667e-05,
      "loss": 1.5026,
      "step": 39730
    },
    {
      "epoch": 2.12063610344302,
      "grad_norm": 0.3200650215148926,
      "learning_rate": 7.42e-05,
      "loss": 1.504,
      "step": 39740
    },
    {
      "epoch": 2.121178044608532,
      "grad_norm": 0.3450500965118408,
      "learning_rate": 7.453333333333333e-05,
      "loss": 1.4982,
      "step": 39750
    },
    {
      "epoch": 2.121665791657493,
      "eval_loss": 2.47827410697937,
      "eval_runtime": 22.0331,
      "eval_samples_per_second": 226.931,
      "eval_steps_per_second": 1.225,
      "step": 39759
    },
    {
      "epoch": 2.1217199857740443,
      "grad_norm": 0.221024289727211,
      "learning_rate": 7.486666666666668e-05,
      "loss": 1.5028,
      "step": 39760
    },
    {
      "epoch": 2.1222619269395566,
      "grad_norm": 0.2555961012840271,
      "learning_rate": 7.52e-05,
      "loss": 1.5123,
      "step": 39770
    },
    {
      "epoch": 2.122803868105069,
      "grad_norm": 0.34122392535209656,
      "learning_rate": 7.553333333333333e-05,
      "loss": 1.5043,
      "step": 39780
    },
    {
      "epoch": 2.1233458092705813,
      "grad_norm": 0.29069313406944275,
      "learning_rate": 7.586666666666668e-05,
      "loss": 1.4978,
      "step": 39790
    },
    {
      "epoch": 2.123887750436093,
      "grad_norm": 0.3182983696460724,
      "learning_rate": 7.620000000000001e-05,
      "loss": 1.5068,
      "step": 39800
    },
    {
      "epoch": 2.1244296916016054,
      "grad_norm": 0.5020242929458618,
      "learning_rate": 7.653333333333333e-05,
      "loss": 1.497,
      "step": 39810
    },
    {
      "epoch": 2.1249716327671178,
      "grad_norm": 0.27356261014938354,
      "learning_rate": 7.686666666666667e-05,
      "loss": 1.5029,
      "step": 39820
    },
    {
      "epoch": 2.12551357393263,
      "grad_norm": 0.315500408411026,
      "learning_rate": 7.72e-05,
      "loss": 1.5193,
      "step": 39830
    },
    {
      "epoch": 2.126055515098142,
      "grad_norm": 0.36083605885505676,
      "learning_rate": 7.753333333333334e-05,
      "loss": 1.5013,
      "step": 39840
    },
    {
      "epoch": 2.1263806797974496,
      "eval_loss": 2.4653656482696533,
      "eval_runtime": 21.9768,
      "eval_samples_per_second": 227.513,
      "eval_steps_per_second": 1.229,
      "step": 39846
    },
    {
      "epoch": 2.1265974562636543,
      "grad_norm": 0.3310469090938568,
      "learning_rate": 7.786666666666667e-05,
      "loss": 1.5136,
      "step": 39850
    },
    {
      "epoch": 2.1271393974291666,
      "grad_norm": 0.3014518618583679,
      "learning_rate": 7.82e-05,
      "loss": 1.5079,
      "step": 39860
    },
    {
      "epoch": 2.127681338594679,
      "grad_norm": 0.434340238571167,
      "learning_rate": 7.853333333333334e-05,
      "loss": 1.5123,
      "step": 39870
    },
    {
      "epoch": 2.128223279760191,
      "grad_norm": 0.34903663396835327,
      "learning_rate": 7.886666666666667e-05,
      "loss": 1.5069,
      "step": 39880
    },
    {
      "epoch": 2.128765220925703,
      "grad_norm": 0.2542589604854584,
      "learning_rate": 7.920000000000001e-05,
      "loss": 1.5123,
      "step": 39890
    },
    {
      "epoch": 2.1293071620912154,
      "grad_norm": 0.26018011569976807,
      "learning_rate": 7.953333333333333e-05,
      "loss": 1.5098,
      "step": 39900
    },
    {
      "epoch": 2.1298491032567277,
      "grad_norm": 0.19746781885623932,
      "learning_rate": 7.986666666666667e-05,
      "loss": 1.5074,
      "step": 39910
    },
    {
      "epoch": 2.13039104442224,
      "grad_norm": 0.2585922181606293,
      "learning_rate": 8.020000000000001e-05,
      "loss": 1.5005,
      "step": 39920
    },
    {
      "epoch": 2.1309329855877523,
      "grad_norm": 0.3514852523803711,
      "learning_rate": 8.053333333333334e-05,
      "loss": 1.5007,
      "step": 39930
    },
    {
      "epoch": 2.1310955679374057,
      "eval_loss": 2.4711849689483643,
      "eval_runtime": 21.9689,
      "eval_samples_per_second": 227.594,
      "eval_steps_per_second": 1.229,
      "step": 39933
    },
    {
      "epoch": 2.131474926753264,
      "grad_norm": 0.4120597839355469,
      "learning_rate": 8.086666666666666e-05,
      "loss": 1.5033,
      "step": 39940
    },
    {
      "epoch": 2.1320168679187765,
      "grad_norm": 0.27820542454719543,
      "learning_rate": 8.120000000000001e-05,
      "loss": 1.5124,
      "step": 39950
    },
    {
      "epoch": 2.132558809084289,
      "grad_norm": 0.33248940110206604,
      "learning_rate": 8.153333333333334e-05,
      "loss": 1.5004,
      "step": 39960
    },
    {
      "epoch": 2.133100750249801,
      "grad_norm": 0.4663317799568176,
      "learning_rate": 8.186666666666667e-05,
      "loss": 1.5161,
      "step": 39970
    },
    {
      "epoch": 2.133642691415313,
      "grad_norm": 0.463692843914032,
      "learning_rate": 8.22e-05,
      "loss": 1.5041,
      "step": 39980
    },
    {
      "epoch": 2.1341846325808254,
      "grad_norm": 0.3125864267349243,
      "learning_rate": 8.253333333333334e-05,
      "loss": 1.496,
      "step": 39990
    },
    {
      "epoch": 2.1347265737463377,
      "grad_norm": 0.4677470922470093,
      "learning_rate": 8.286666666666667e-05,
      "loss": 1.5,
      "step": 40000
    },
    {
      "epoch": 2.13526851491185,
      "grad_norm": 0.24888941645622253,
      "learning_rate": 8.32e-05,
      "loss": 1.5033,
      "step": 40010
    },
    {
      "epoch": 2.1358104560773623,
      "grad_norm": 0.26266810297966003,
      "learning_rate": 8.353333333333334e-05,
      "loss": 1.5042,
      "step": 40020
    },
    {
      "epoch": 2.1358104560773623,
      "eval_loss": 2.4674057960510254,
      "eval_runtime": 21.5786,
      "eval_samples_per_second": 231.711,
      "eval_steps_per_second": 1.251,
      "step": 40020
    },
    {
      "epoch": 2.136352397242874,
      "grad_norm": 0.36174437403678894,
      "learning_rate": 8.386666666666667e-05,
      "loss": 1.5109,
      "step": 40030
    },
    {
      "epoch": 2.1368943384083865,
      "grad_norm": 0.2121952325105667,
      "learning_rate": 8.42e-05,
      "loss": 1.4932,
      "step": 40040
    },
    {
      "epoch": 2.137436279573899,
      "grad_norm": 0.3189486861228943,
      "learning_rate": 8.453333333333335e-05,
      "loss": 1.51,
      "step": 40050
    },
    {
      "epoch": 2.137978220739411,
      "grad_norm": 0.36640292406082153,
      "learning_rate": 8.486666666666668e-05,
      "loss": 1.5087,
      "step": 40060
    },
    {
      "epoch": 2.138520161904923,
      "grad_norm": 0.22334055602550507,
      "learning_rate": 8.52e-05,
      "loss": 1.5019,
      "step": 40070
    },
    {
      "epoch": 2.1390621030704353,
      "grad_norm": 0.4532851576805115,
      "learning_rate": 8.553333333333333e-05,
      "loss": 1.5119,
      "step": 40080
    },
    {
      "epoch": 2.1396040442359476,
      "grad_norm": 0.30973002314567566,
      "learning_rate": 8.586666666666668e-05,
      "loss": 1.5028,
      "step": 40090
    },
    {
      "epoch": 2.14014598540146,
      "grad_norm": 0.27160486578941345,
      "learning_rate": 8.620000000000001e-05,
      "loss": 1.5112,
      "step": 40100
    },
    {
      "epoch": 2.1405253442173184,
      "eval_loss": 2.4754085540771484,
      "eval_runtime": 21.9771,
      "eval_samples_per_second": 227.51,
      "eval_steps_per_second": 1.229,
      "step": 40107
    },
    {
      "epoch": 2.1406879265669723,
      "grad_norm": 0.3385959267616272,
      "learning_rate": 8.653333333333333e-05,
      "loss": 1.5036,
      "step": 40110
    },
    {
      "epoch": 2.141229867732484,
      "grad_norm": 0.5644890069961548,
      "learning_rate": 8.686666666666667e-05,
      "loss": 1.5046,
      "step": 40120
    },
    {
      "epoch": 2.1417718088979965,
      "grad_norm": 0.4575164020061493,
      "learning_rate": 8.72e-05,
      "loss": 1.5059,
      "step": 40130
    },
    {
      "epoch": 2.1423137500635088,
      "grad_norm": 0.656599223613739,
      "learning_rate": 8.753333333333334e-05,
      "loss": 1.5027,
      "step": 40140
    },
    {
      "epoch": 2.142855691229021,
      "grad_norm": 0.4436124563217163,
      "learning_rate": 8.786666666666667e-05,
      "loss": 1.5026,
      "step": 40150
    },
    {
      "epoch": 2.1433976323945334,
      "grad_norm": 0.35058170557022095,
      "learning_rate": 8.82e-05,
      "loss": 1.5074,
      "step": 40160
    },
    {
      "epoch": 2.1439395735600453,
      "grad_norm": 0.2605305016040802,
      "learning_rate": 8.853333333333333e-05,
      "loss": 1.5077,
      "step": 40170
    },
    {
      "epoch": 2.1444815147255576,
      "grad_norm": 0.3072412312030792,
      "learning_rate": 8.886666666666668e-05,
      "loss": 1.5117,
      "step": 40180
    },
    {
      "epoch": 2.14502345589107,
      "grad_norm": 0.20943394303321838,
      "learning_rate": 8.92e-05,
      "loss": 1.5003,
      "step": 40190
    },
    {
      "epoch": 2.1452402323572746,
      "eval_loss": 2.478358507156372,
      "eval_runtime": 24.6282,
      "eval_samples_per_second": 203.019,
      "eval_steps_per_second": 1.096,
      "step": 40194
    },
    {
      "epoch": 2.1455653970565822,
      "grad_norm": 0.4068165719509125,
      "learning_rate": 8.953333333333333e-05,
      "loss": 1.5108,
      "step": 40200
    },
    {
      "epoch": 2.146107338222094,
      "grad_norm": 0.35715365409851074,
      "learning_rate": 8.986666666666666e-05,
      "loss": 1.5062,
      "step": 40210
    },
    {
      "epoch": 2.1466492793876064,
      "grad_norm": 0.4496524930000305,
      "learning_rate": 9.020000000000001e-05,
      "loss": 1.5063,
      "step": 40220
    },
    {
      "epoch": 2.1471912205531187,
      "grad_norm": 0.606382429599762,
      "learning_rate": 9.053333333333334e-05,
      "loss": 1.5041,
      "step": 40230
    },
    {
      "epoch": 2.147733161718631,
      "grad_norm": 0.3925984799861908,
      "learning_rate": 9.086666666666666e-05,
      "loss": 1.5084,
      "step": 40240
    },
    {
      "epoch": 2.148275102884143,
      "grad_norm": 0.29136770963668823,
      "learning_rate": 9.120000000000001e-05,
      "loss": 1.5013,
      "step": 40250
    },
    {
      "epoch": 2.1488170440496552,
      "grad_norm": 0.35544130206108093,
      "learning_rate": 9.153333333333334e-05,
      "loss": 1.5072,
      "step": 40260
    },
    {
      "epoch": 2.1493589852151676,
      "grad_norm": 0.3096676766872406,
      "learning_rate": 9.186666666666667e-05,
      "loss": 1.5037,
      "step": 40270
    },
    {
      "epoch": 2.14990092638068,
      "grad_norm": 0.2778821587562561,
      "learning_rate": 9.22e-05,
      "loss": 1.5112,
      "step": 40280
    },
    {
      "epoch": 2.149955120497231,
      "eval_loss": 2.4636154174804688,
      "eval_runtime": 24.7898,
      "eval_samples_per_second": 201.695,
      "eval_steps_per_second": 1.089,
      "step": 40281
    },
    {
      "epoch": 2.150442867546192,
      "grad_norm": 0.7011622786521912,
      "learning_rate": 9.253333333333334e-05,
      "loss": 1.5166,
      "step": 40290
    },
    {
      "epoch": 2.150984808711704,
      "grad_norm": 0.5509504079818726,
      "learning_rate": 9.286666666666667e-05,
      "loss": 1.5078,
      "step": 40300
    },
    {
      "epoch": 2.1515267498772164,
      "grad_norm": 0.28685396909713745,
      "learning_rate": 9.320000000000002e-05,
      "loss": 1.5159,
      "step": 40310
    },
    {
      "epoch": 2.1520686910427287,
      "grad_norm": 0.3149552345275879,
      "learning_rate": 9.353333333333333e-05,
      "loss": 1.5044,
      "step": 40320
    },
    {
      "epoch": 2.152610632208241,
      "grad_norm": 0.458866685628891,
      "learning_rate": 9.386666666666667e-05,
      "loss": 1.4973,
      "step": 40330
    },
    {
      "epoch": 2.1531525733737533,
      "grad_norm": 0.21605221927165985,
      "learning_rate": 9.42e-05,
      "loss": 1.5009,
      "step": 40340
    },
    {
      "epoch": 2.153694514539265,
      "grad_norm": 0.6214112043380737,
      "learning_rate": 9.453333333333335e-05,
      "loss": 1.5062,
      "step": 40350
    },
    {
      "epoch": 2.1542364557047775,
      "grad_norm": 0.5758342146873474,
      "learning_rate": 9.486666666666666e-05,
      "loss": 1.5175,
      "step": 40360
    },
    {
      "epoch": 2.1546700086371873,
      "eval_loss": 2.45977520942688,
      "eval_runtime": 21.979,
      "eval_samples_per_second": 227.489,
      "eval_steps_per_second": 1.228,
      "step": 40368
    },
    {
      "epoch": 2.15477839687029,
      "grad_norm": 0.3867745101451874,
      "learning_rate": 9.52e-05,
      "loss": 1.513,
      "step": 40370
    },
    {
      "epoch": 2.155320338035802,
      "grad_norm": 0.25635677576065063,
      "learning_rate": 9.553333333333334e-05,
      "loss": 1.5077,
      "step": 40380
    },
    {
      "epoch": 2.155862279201314,
      "grad_norm": 0.5329334735870361,
      "learning_rate": 9.586666666666667e-05,
      "loss": 1.5127,
      "step": 40390
    },
    {
      "epoch": 2.1564042203668263,
      "grad_norm": 0.4407355785369873,
      "learning_rate": 9.620000000000001e-05,
      "loss": 1.5063,
      "step": 40400
    },
    {
      "epoch": 2.1569461615323386,
      "grad_norm": 0.3090643584728241,
      "learning_rate": 9.653333333333334e-05,
      "loss": 1.5083,
      "step": 40410
    },
    {
      "epoch": 2.157488102697851,
      "grad_norm": 0.487301766872406,
      "learning_rate": 9.686666666666667e-05,
      "loss": 1.5051,
      "step": 40420
    },
    {
      "epoch": 2.1580300438633633,
      "grad_norm": 0.5774275064468384,
      "learning_rate": 9.72e-05,
      "loss": 1.5058,
      "step": 40430
    },
    {
      "epoch": 2.158571985028875,
      "grad_norm": 0.30334457755088806,
      "learning_rate": 9.753333333333334e-05,
      "loss": 1.514,
      "step": 40440
    },
    {
      "epoch": 2.1591139261943875,
      "grad_norm": 0.5308240056037903,
      "learning_rate": 9.786666666666667e-05,
      "loss": 1.5075,
      "step": 40450
    },
    {
      "epoch": 2.1593848967771434,
      "eval_loss": 2.4678995609283447,
      "eval_runtime": 22.396,
      "eval_samples_per_second": 223.254,
      "eval_steps_per_second": 1.206,
      "step": 40455
    },
    {
      "epoch": 2.1596558673599,
      "grad_norm": 0.5267515182495117,
      "learning_rate": 9.82e-05,
      "loss": 1.5033,
      "step": 40460
    },
    {
      "epoch": 2.160197808525412,
      "grad_norm": 0.4302433133125305,
      "learning_rate": 9.853333333333333e-05,
      "loss": 1.5039,
      "step": 40470
    },
    {
      "epoch": 2.160739749690924,
      "grad_norm": 0.2622862160205841,
      "learning_rate": 9.886666666666668e-05,
      "loss": 1.5074,
      "step": 40480
    },
    {
      "epoch": 2.1612816908564363,
      "grad_norm": 0.34495052695274353,
      "learning_rate": 9.92e-05,
      "loss": 1.5148,
      "step": 40490
    },
    {
      "epoch": 2.1618236320219486,
      "grad_norm": 0.28749775886535645,
      "learning_rate": 9.953333333333333e-05,
      "loss": 1.5168,
      "step": 40500
    },
    {
      "epoch": 2.162365573187461,
      "grad_norm": 0.34906288981437683,
      "learning_rate": 9.986666666666668e-05,
      "loss": 1.5094,
      "step": 40510
    },
    {
      "epoch": 2.1629075143529732,
      "grad_norm": 0.6776122450828552,
      "learning_rate": 9.999999899660717e-05,
      "loss": 1.5113,
      "step": 40520
    },
    {
      "epoch": 2.163449455518485,
      "grad_norm": 0.3558565080165863,
      "learning_rate": 9.99999928647622e-05,
      "loss": 1.5072,
      "step": 40530
    },
    {
      "epoch": 2.1639913966839974,
      "grad_norm": 0.2457047551870346,
      "learning_rate": 9.999998115851347e-05,
      "loss": 1.5147,
      "step": 40540
    },
    {
      "epoch": 2.1640997849171,
      "eval_loss": 2.471486806869507,
      "eval_runtime": 21.974,
      "eval_samples_per_second": 227.542,
      "eval_steps_per_second": 1.229,
      "step": 40542
    },
    {
      "epoch": 2.1645333378495097,
      "grad_norm": 0.520976722240448,
      "learning_rate": 9.999996387786247e-05,
      "loss": 1.5066,
      "step": 40550
    },
    {
      "epoch": 2.165075279015022,
      "grad_norm": 0.3190830945968628,
      "learning_rate": 9.99999410228113e-05,
      "loss": 1.5005,
      "step": 40560
    },
    {
      "epoch": 2.1656172201805344,
      "grad_norm": 0.31958481669425964,
      "learning_rate": 9.99999125933628e-05,
      "loss": 1.5157,
      "step": 40570
    },
    {
      "epoch": 2.1661591613460462,
      "grad_norm": 0.5813402533531189,
      "learning_rate": 9.999987858952052e-05,
      "loss": 1.5099,
      "step": 40580
    },
    {
      "epoch": 2.1667011025115586,
      "grad_norm": 0.34184807538986206,
      "learning_rate": 9.999983901128863e-05,
      "loss": 1.5005,
      "step": 40590
    },
    {
      "epoch": 2.167243043677071,
      "grad_norm": 0.3223097622394562,
      "learning_rate": 9.999979385867205e-05,
      "loss": 1.5134,
      "step": 40600
    },
    {
      "epoch": 2.167784984842583,
      "grad_norm": 0.20804668962955475,
      "learning_rate": 9.99997431316764e-05,
      "loss": 1.4992,
      "step": 40610
    },
    {
      "epoch": 2.168326926008095,
      "grad_norm": 0.25401726365089417,
      "learning_rate": 9.999968683030792e-05,
      "loss": 1.5153,
      "step": 40620
    },
    {
      "epoch": 2.168814673057056,
      "eval_loss": 2.468203067779541,
      "eval_runtime": 21.9753,
      "eval_samples_per_second": 227.529,
      "eval_steps_per_second": 1.229,
      "step": 40629
    },
    {
      "epoch": 2.1688688671736074,
      "grad_norm": 0.2348717451095581,
      "learning_rate": 9.999962495457362e-05,
      "loss": 1.5125,
      "step": 40630
    },
    {
      "epoch": 2.1694108083391197,
      "grad_norm": 0.43220508098602295,
      "learning_rate": 9.999955750448114e-05,
      "loss": 1.5078,
      "step": 40640
    },
    {
      "epoch": 2.169952749504632,
      "grad_norm": 0.28150442242622375,
      "learning_rate": 9.999948448003884e-05,
      "loss": 1.5077,
      "step": 40650
    },
    {
      "epoch": 2.170494690670144,
      "grad_norm": 0.24436496198177338,
      "learning_rate": 9.999940588125579e-05,
      "loss": 1.5158,
      "step": 40660
    },
    {
      "epoch": 2.171036631835656,
      "grad_norm": 0.3555525243282318,
      "learning_rate": 9.999932170814168e-05,
      "loss": 1.5009,
      "step": 40670
    },
    {
      "epoch": 2.1715785730011685,
      "grad_norm": 0.6536611318588257,
      "learning_rate": 9.999923196070698e-05,
      "loss": 1.5179,
      "step": 40680
    },
    {
      "epoch": 2.172120514166681,
      "grad_norm": 0.555479884147644,
      "learning_rate": 9.99991366389628e-05,
      "loss": 1.5034,
      "step": 40690
    },
    {
      "epoch": 2.172662455332193,
      "grad_norm": 0.27514633536338806,
      "learning_rate": 9.999903574292093e-05,
      "loss": 1.4983,
      "step": 40700
    },
    {
      "epoch": 2.173204396497705,
      "grad_norm": 0.2602129876613617,
      "learning_rate": 9.999892927259388e-05,
      "loss": 1.5167,
      "step": 40710
    },
    {
      "epoch": 2.1735295611970127,
      "eval_loss": 2.479363441467285,
      "eval_runtime": 21.9755,
      "eval_samples_per_second": 227.526,
      "eval_steps_per_second": 1.229,
      "step": 40716
    },
    {
      "epoch": 2.1737463376632173,
      "grad_norm": 0.3722988963127136,
      "learning_rate": 9.999881722799482e-05,
      "loss": 1.5071,
      "step": 40720
    },
    {
      "epoch": 2.1742882788287297,
      "grad_norm": 0.22509793937206268,
      "learning_rate": 9.999869960913767e-05,
      "loss": 1.5103,
      "step": 40730
    },
    {
      "epoch": 2.174830219994242,
      "grad_norm": 0.5048813819885254,
      "learning_rate": 9.999857641603697e-05,
      "loss": 1.5164,
      "step": 40740
    },
    {
      "epoch": 2.1753721611597543,
      "grad_norm": 0.3289371430873871,
      "learning_rate": 9.999844764870799e-05,
      "loss": 1.5028,
      "step": 40750
    },
    {
      "epoch": 2.175914102325266,
      "grad_norm": 0.5190215706825256,
      "learning_rate": 9.999831330716668e-05,
      "loss": 1.5106,
      "step": 40760
    },
    {
      "epoch": 2.1764560434907785,
      "grad_norm": 0.2816135883331299,
      "learning_rate": 9.999817339142969e-05,
      "loss": 1.5065,
      "step": 40770
    },
    {
      "epoch": 2.176997984656291,
      "grad_norm": 0.7596124410629272,
      "learning_rate": 9.999802790151434e-05,
      "loss": 1.5103,
      "step": 40780
    },
    {
      "epoch": 2.177539925821803,
      "grad_norm": 0.34131622314453125,
      "learning_rate": 9.999787683743863e-05,
      "loss": 1.5063,
      "step": 40790
    },
    {
      "epoch": 2.178081866987315,
      "grad_norm": 0.3968164026737213,
      "learning_rate": 9.999772019922132e-05,
      "loss": 1.5084,
      "step": 40800
    },
    {
      "epoch": 2.178244449336969,
      "eval_loss": 2.4752957820892334,
      "eval_runtime": 21.9803,
      "eval_samples_per_second": 227.477,
      "eval_steps_per_second": 1.228,
      "step": 40803
    },
    {
      "epoch": 2.1786238081528273,
      "grad_norm": 0.28055036067962646,
      "learning_rate": 9.999755798688178e-05,
      "loss": 1.5242,
      "step": 40810
    },
    {
      "epoch": 2.1791657493183396,
      "grad_norm": 0.35142403841018677,
      "learning_rate": 9.999739020044013e-05,
      "loss": 1.5095,
      "step": 40820
    },
    {
      "epoch": 2.179707690483852,
      "grad_norm": 0.2656286656856537,
      "learning_rate": 9.999721683991714e-05,
      "loss": 1.5034,
      "step": 40830
    },
    {
      "epoch": 2.1802496316493643,
      "grad_norm": 0.2500777840614319,
      "learning_rate": 9.999703790533428e-05,
      "loss": 1.5072,
      "step": 40840
    },
    {
      "epoch": 2.180791572814876,
      "grad_norm": 0.23317931592464447,
      "learning_rate": 9.999685339671372e-05,
      "loss": 1.5045,
      "step": 40850
    },
    {
      "epoch": 2.1813335139803884,
      "grad_norm": 0.6272380352020264,
      "learning_rate": 9.999666331407832e-05,
      "loss": 1.506,
      "step": 40860
    },
    {
      "epoch": 2.1818754551459008,
      "grad_norm": 0.28550583124160767,
      "learning_rate": 9.999646765745162e-05,
      "loss": 1.5038,
      "step": 40870
    },
    {
      "epoch": 2.182417396311413,
      "grad_norm": 0.3291584253311157,
      "learning_rate": 9.999626642685788e-05,
      "loss": 1.5097,
      "step": 40880
    },
    {
      "epoch": 2.182959337476925,
      "grad_norm": 0.6882104873657227,
      "learning_rate": 9.999605962232201e-05,
      "loss": 1.4979,
      "step": 40890
    },
    {
      "epoch": 2.182959337476925,
      "eval_loss": 2.482081174850464,
      "eval_runtime": 21.9465,
      "eval_samples_per_second": 227.827,
      "eval_steps_per_second": 1.23,
      "step": 40890
    },
    {
      "epoch": 2.1835012786424373,
      "grad_norm": 0.4152418375015259,
      "learning_rate": 9.999584724386959e-05,
      "loss": 1.4999,
      "step": 40900
    },
    {
      "epoch": 2.1840432198079496,
      "grad_norm": 0.40763190388679504,
      "learning_rate": 9.9995629291527e-05,
      "loss": 1.5092,
      "step": 40910
    },
    {
      "epoch": 2.184585160973462,
      "grad_norm": 0.29160091280937195,
      "learning_rate": 9.99954057653212e-05,
      "loss": 1.5012,
      "step": 40920
    },
    {
      "epoch": 2.185127102138974,
      "grad_norm": 0.40323200821876526,
      "learning_rate": 9.999517666527988e-05,
      "loss": 1.5059,
      "step": 40930
    },
    {
      "epoch": 2.185669043304486,
      "grad_norm": 0.30431511998176575,
      "learning_rate": 9.999494199143142e-05,
      "loss": 1.5109,
      "step": 40940
    },
    {
      "epoch": 2.1862109844699984,
      "grad_norm": 0.2627016007900238,
      "learning_rate": 9.999470174380489e-05,
      "loss": 1.5064,
      "step": 40950
    },
    {
      "epoch": 2.1867529256355107,
      "grad_norm": 0.47459253668785095,
      "learning_rate": 9.999445592243008e-05,
      "loss": 1.51,
      "step": 40960
    },
    {
      "epoch": 2.187294866801023,
      "grad_norm": 0.352264404296875,
      "learning_rate": 9.999420452733739e-05,
      "loss": 1.5102,
      "step": 40970
    },
    {
      "epoch": 2.1876742256168815,
      "eval_loss": 2.473921298980713,
      "eval_runtime": 21.9773,
      "eval_samples_per_second": 227.507,
      "eval_steps_per_second": 1.229,
      "step": 40977
    },
    {
      "epoch": 2.1878368079665353,
      "grad_norm": 0.35294270515441895,
      "learning_rate": 9.9993947558558e-05,
      "loss": 1.5141,
      "step": 40980
    },
    {
      "epoch": 2.188378749132047,
      "grad_norm": 0.4508706033229828,
      "learning_rate": 9.999368501612373e-05,
      "loss": 1.503,
      "step": 40990
    },
    {
      "epoch": 2.1889206902975595,
      "grad_norm": 0.5512154698371887,
      "learning_rate": 9.999341690006711e-05,
      "loss": 1.5284,
      "step": 41000
    },
    {
      "epoch": 2.189462631463072,
      "grad_norm": 0.4073977768421173,
      "learning_rate": 9.999314321042134e-05,
      "loss": 1.4996,
      "step": 41010
    },
    {
      "epoch": 2.190004572628584,
      "grad_norm": 0.458518385887146,
      "learning_rate": 9.999286394722031e-05,
      "loss": 1.5099,
      "step": 41020
    },
    {
      "epoch": 2.190546513794096,
      "grad_norm": 0.25982314348220825,
      "learning_rate": 9.999257911049866e-05,
      "loss": 1.5042,
      "step": 41030
    },
    {
      "epoch": 2.1910884549596084,
      "grad_norm": 0.2903514802455902,
      "learning_rate": 9.999228870029165e-05,
      "loss": 1.5033,
      "step": 41040
    },
    {
      "epoch": 2.1916303961251207,
      "grad_norm": 0.21735897660255432,
      "learning_rate": 9.999199271663524e-05,
      "loss": 1.5171,
      "step": 41050
    },
    {
      "epoch": 2.192172337290633,
      "grad_norm": 0.39294907450675964,
      "learning_rate": 9.999169115956612e-05,
      "loss": 1.5007,
      "step": 41060
    },
    {
      "epoch": 2.1923891137568376,
      "eval_loss": 2.489570140838623,
      "eval_runtime": 21.9756,
      "eval_samples_per_second": 227.525,
      "eval_steps_per_second": 1.229,
      "step": 41064
    },
    {
      "epoch": 2.1927142784561453,
      "grad_norm": 0.5847845673561096,
      "learning_rate": 9.999138402912161e-05,
      "loss": 1.5158,
      "step": 41070
    },
    {
      "epoch": 2.193256219621657,
      "grad_norm": 0.33957940340042114,
      "learning_rate": 9.99910713253398e-05,
      "loss": 1.5152,
      "step": 41080
    },
    {
      "epoch": 2.1937981607871695,
      "grad_norm": 0.3208034038543701,
      "learning_rate": 9.999075304825941e-05,
      "loss": 1.509,
      "step": 41090
    },
    {
      "epoch": 2.194340101952682,
      "grad_norm": 0.3303355872631073,
      "learning_rate": 9.999042919791985e-05,
      "loss": 1.5075,
      "step": 41100
    },
    {
      "epoch": 2.194882043118194,
      "grad_norm": 0.278929203748703,
      "learning_rate": 9.999009977436125e-05,
      "loss": 1.5121,
      "step": 41110
    },
    {
      "epoch": 2.195423984283706,
      "grad_norm": 0.21701566874980927,
      "learning_rate": 9.998976477762442e-05,
      "loss": 1.5121,
      "step": 41120
    },
    {
      "epoch": 2.1959659254492183,
      "grad_norm": 0.35072654485702515,
      "learning_rate": 9.998942420775086e-05,
      "loss": 1.5021,
      "step": 41130
    },
    {
      "epoch": 2.1965078666147306,
      "grad_norm": 0.36139407753944397,
      "learning_rate": 9.998907806478275e-05,
      "loss": 1.5029,
      "step": 41140
    },
    {
      "epoch": 2.197049807780243,
      "grad_norm": 0.31903785467147827,
      "learning_rate": 9.998872634876297e-05,
      "loss": 1.506,
      "step": 41150
    },
    {
      "epoch": 2.1971040018967942,
      "eval_loss": 2.4797050952911377,
      "eval_runtime": 21.9824,
      "eval_samples_per_second": 227.454,
      "eval_steps_per_second": 1.228,
      "step": 41151
    },
    {
      "epoch": 2.1975917489457553,
      "grad_norm": 0.6769224405288696,
      "learning_rate": 9.99883690597351e-05,
      "loss": 1.5126,
      "step": 41160
    },
    {
      "epoch": 2.198133690111267,
      "grad_norm": 0.44375690817832947,
      "learning_rate": 9.998800619774339e-05,
      "loss": 1.5127,
      "step": 41170
    },
    {
      "epoch": 2.1986756312767795,
      "grad_norm": 0.22975300252437592,
      "learning_rate": 9.998763776283277e-05,
      "loss": 1.5153,
      "step": 41180
    },
    {
      "epoch": 2.1992175724422918,
      "grad_norm": 0.250010222196579,
      "learning_rate": 9.998726375504892e-05,
      "loss": 1.5109,
      "step": 41190
    },
    {
      "epoch": 2.199759513607804,
      "grad_norm": 0.2538135349750519,
      "learning_rate": 9.998688417443815e-05,
      "loss": 1.5127,
      "step": 41200
    },
    {
      "epoch": 2.2003014547733164,
      "grad_norm": 0.25529077649116516,
      "learning_rate": 9.998649902104748e-05,
      "loss": 1.5056,
      "step": 41210
    },
    {
      "epoch": 2.2008433959388283,
      "grad_norm": 0.3090484142303467,
      "learning_rate": 9.998610829492462e-05,
      "loss": 1.5212,
      "step": 41220
    },
    {
      "epoch": 2.2013853371043406,
      "grad_norm": 0.2540747821331024,
      "learning_rate": 9.998571199611799e-05,
      "loss": 1.5252,
      "step": 41230
    },
    {
      "epoch": 2.2018188900367504,
      "eval_loss": 2.49056077003479,
      "eval_runtime": 21.9807,
      "eval_samples_per_second": 227.473,
      "eval_steps_per_second": 1.228,
      "step": 41238
    },
    {
      "epoch": 2.201927278269853,
      "grad_norm": 0.35519057512283325,
      "learning_rate": 9.998531012467664e-05,
      "loss": 1.5082,
      "step": 41240
    },
    {
      "epoch": 2.202469219435365,
      "grad_norm": 0.24406082928180695,
      "learning_rate": 9.998490268065038e-05,
      "loss": 1.5048,
      "step": 41250
    },
    {
      "epoch": 2.203011160600877,
      "grad_norm": 0.3896248936653137,
      "learning_rate": 9.998448966408971e-05,
      "loss": 1.4991,
      "step": 41260
    },
    {
      "epoch": 2.2035531017663894,
      "grad_norm": 0.278812974691391,
      "learning_rate": 9.998407107504575e-05,
      "loss": 1.5014,
      "step": 41270
    },
    {
      "epoch": 2.2040950429319017,
      "grad_norm": 0.5823280811309814,
      "learning_rate": 9.998364691357036e-05,
      "loss": 1.5131,
      "step": 41280
    },
    {
      "epoch": 2.204636984097414,
      "grad_norm": 0.3981192111968994,
      "learning_rate": 9.998321717971609e-05,
      "loss": 1.5001,
      "step": 41290
    },
    {
      "epoch": 2.205178925262926,
      "grad_norm": 0.6286061406135559,
      "learning_rate": 9.998278187353616e-05,
      "loss": 1.5109,
      "step": 41300
    },
    {
      "epoch": 2.2057208664284382,
      "grad_norm": 0.2553674280643463,
      "learning_rate": 9.998234099508454e-05,
      "loss": 1.5113,
      "step": 41310
    },
    {
      "epoch": 2.2062628075939505,
      "grad_norm": 0.48403024673461914,
      "learning_rate": 9.998189454441579e-05,
      "loss": 1.5138,
      "step": 41320
    },
    {
      "epoch": 2.2065337781767065,
      "eval_loss": 2.478606700897217,
      "eval_runtime": 21.9762,
      "eval_samples_per_second": 227.519,
      "eval_steps_per_second": 1.229,
      "step": 41325
    },
    {
      "epoch": 2.206804748759463,
      "grad_norm": 0.3329886794090271,
      "learning_rate": 9.998144252158523e-05,
      "loss": 1.5122,
      "step": 41330
    },
    {
      "epoch": 2.207346689924975,
      "grad_norm": 0.2454531043767929,
      "learning_rate": 9.998098492664888e-05,
      "loss": 1.5094,
      "step": 41340
    },
    {
      "epoch": 2.207888631090487,
      "grad_norm": 0.3834342360496521,
      "learning_rate": 9.998052175966339e-05,
      "loss": 1.5042,
      "step": 41350
    },
    {
      "epoch": 2.2084305722559994,
      "grad_norm": 0.27599623799324036,
      "learning_rate": 9.998005302068616e-05,
      "loss": 1.5139,
      "step": 41360
    },
    {
      "epoch": 2.2089725134215117,
      "grad_norm": 0.2955540716648102,
      "learning_rate": 9.997957870977525e-05,
      "loss": 1.5212,
      "step": 41370
    },
    {
      "epoch": 2.209514454587024,
      "grad_norm": 0.25166115164756775,
      "learning_rate": 9.99790988269894e-05,
      "loss": 1.5021,
      "step": 41380
    },
    {
      "epoch": 2.2100563957525363,
      "grad_norm": 0.3993496298789978,
      "learning_rate": 9.997861337238807e-05,
      "loss": 1.4986,
      "step": 41390
    },
    {
      "epoch": 2.210598336918048,
      "grad_norm": 0.3200553059577942,
      "learning_rate": 9.99781223460314e-05,
      "loss": 1.5055,
      "step": 41400
    },
    {
      "epoch": 2.2111402780835605,
      "grad_norm": 0.3932972550392151,
      "learning_rate": 9.99776257479802e-05,
      "loss": 1.5082,
      "step": 41410
    },
    {
      "epoch": 2.211248666316663,
      "eval_loss": 2.4762344360351562,
      "eval_runtime": 21.9757,
      "eval_samples_per_second": 227.524,
      "eval_steps_per_second": 1.229,
      "step": 41412
    },
    {
      "epoch": 2.211682219249073,
      "grad_norm": 0.22166556119918823,
      "learning_rate": 9.997712357829599e-05,
      "loss": 1.5199,
      "step": 41420
    },
    {
      "epoch": 2.212224160414585,
      "grad_norm": 0.4338929057121277,
      "learning_rate": 9.997661583704098e-05,
      "loss": 1.51,
      "step": 41430
    },
    {
      "epoch": 2.212766101580097,
      "grad_norm": 0.2761380672454834,
      "learning_rate": 9.99761025242781e-05,
      "loss": 1.524,
      "step": 41440
    },
    {
      "epoch": 2.2133080427456093,
      "grad_norm": 0.2140953242778778,
      "learning_rate": 9.997558364007087e-05,
      "loss": 1.5159,
      "step": 41450
    },
    {
      "epoch": 2.2138499839111216,
      "grad_norm": 0.2510915696620941,
      "learning_rate": 9.997505918448364e-05,
      "loss": 1.5055,
      "step": 41460
    },
    {
      "epoch": 2.214391925076634,
      "grad_norm": 0.38897550106048584,
      "learning_rate": 9.997452915758131e-05,
      "loss": 1.4971,
      "step": 41470
    },
    {
      "epoch": 2.2149338662421463,
      "grad_norm": 0.29425960779190063,
      "learning_rate": 9.997399355942958e-05,
      "loss": 1.5033,
      "step": 41480
    },
    {
      "epoch": 2.215475807407658,
      "grad_norm": 0.43580734729766846,
      "learning_rate": 9.997345239009477e-05,
      "loss": 1.5049,
      "step": 41490
    },
    {
      "epoch": 2.215963554456619,
      "eval_loss": 2.4717535972595215,
      "eval_runtime": 22.0352,
      "eval_samples_per_second": 226.909,
      "eval_steps_per_second": 1.225,
      "step": 41499
    },
    {
      "epoch": 2.2160177485731705,
      "grad_norm": 0.25963518023490906,
      "learning_rate": 9.997290564964395e-05,
      "loss": 1.4924,
      "step": 41500
    },
    {
      "epoch": 2.216559689738683,
      "grad_norm": 0.23817171156406403,
      "learning_rate": 9.997235333814482e-05,
      "loss": 1.497,
      "step": 41510
    },
    {
      "epoch": 2.217101630904195,
      "grad_norm": 0.19800572097301483,
      "learning_rate": 9.997179545566582e-05,
      "loss": 1.5142,
      "step": 41520
    },
    {
      "epoch": 2.217643572069707,
      "grad_norm": 0.21633438766002655,
      "learning_rate": 9.997123200227606e-05,
      "loss": 1.5033,
      "step": 41530
    },
    {
      "epoch": 2.2181855132352193,
      "grad_norm": 0.24905604124069214,
      "learning_rate": 9.99706629780453e-05,
      "loss": 1.4983,
      "step": 41540
    },
    {
      "epoch": 2.2187274544007316,
      "grad_norm": 0.47959622740745544,
      "learning_rate": 9.997008838304404e-05,
      "loss": 1.5122,
      "step": 41550
    },
    {
      "epoch": 2.219269395566244,
      "grad_norm": 0.5474976301193237,
      "learning_rate": 9.996950821734351e-05,
      "loss": 1.5153,
      "step": 41560
    },
    {
      "epoch": 2.2198113367317562,
      "grad_norm": 0.38861730694770813,
      "learning_rate": 9.996892248101553e-05,
      "loss": 1.5109,
      "step": 41570
    },
    {
      "epoch": 2.220353277897268,
      "grad_norm": 0.48361384868621826,
      "learning_rate": 9.996833117413266e-05,
      "loss": 1.503,
      "step": 41580
    },
    {
      "epoch": 2.2206784425965758,
      "eval_loss": 2.471097946166992,
      "eval_runtime": 21.9727,
      "eval_samples_per_second": 227.555,
      "eval_steps_per_second": 1.229,
      "step": 41586
    },
    {
      "epoch": 2.2208952190627804,
      "grad_norm": 0.24642570316791534,
      "learning_rate": 9.996773429676815e-05,
      "loss": 1.5177,
      "step": 41590
    },
    {
      "epoch": 2.2214371602282927,
      "grad_norm": 0.2803477942943573,
      "learning_rate": 9.996713184899595e-05,
      "loss": 1.514,
      "step": 41600
    },
    {
      "epoch": 2.221979101393805,
      "grad_norm": 0.555143415927887,
      "learning_rate": 9.996652383089069e-05,
      "loss": 1.5011,
      "step": 41610
    },
    {
      "epoch": 2.2225210425593174,
      "grad_norm": 0.2646414041519165,
      "learning_rate": 9.996591024252768e-05,
      "loss": 1.5026,
      "step": 41620
    },
    {
      "epoch": 2.2230629837248292,
      "grad_norm": 0.5564790964126587,
      "learning_rate": 9.996529108398294e-05,
      "loss": 1.5104,
      "step": 41630
    },
    {
      "epoch": 2.2236049248903416,
      "grad_norm": 0.5015528798103333,
      "learning_rate": 9.996466635533316e-05,
      "loss": 1.5151,
      "step": 41640
    },
    {
      "epoch": 2.224146866055854,
      "grad_norm": 0.3904198408126831,
      "learning_rate": 9.996403605665572e-05,
      "loss": 1.5045,
      "step": 41650
    },
    {
      "epoch": 2.224688807221366,
      "grad_norm": 0.27617716789245605,
      "learning_rate": 9.996340018802872e-05,
      "loss": 1.5048,
      "step": 41660
    },
    {
      "epoch": 2.225230748386878,
      "grad_norm": 0.2563170790672302,
      "learning_rate": 9.99627587495309e-05,
      "loss": 1.507,
      "step": 41670
    },
    {
      "epoch": 2.225393330736532,
      "eval_loss": 2.475794553756714,
      "eval_runtime": 21.9846,
      "eval_samples_per_second": 227.432,
      "eval_steps_per_second": 1.228,
      "step": 41673
    },
    {
      "epoch": 2.2257726895523904,
      "grad_norm": 0.20860429108142853,
      "learning_rate": 9.996211174124174e-05,
      "loss": 1.5048,
      "step": 41680
    },
    {
      "epoch": 2.2263146307179027,
      "grad_norm": 0.2962222397327423,
      "learning_rate": 9.996145916324138e-05,
      "loss": 1.4984,
      "step": 41690
    },
    {
      "epoch": 2.226856571883415,
      "grad_norm": 0.285791277885437,
      "learning_rate": 9.996080101561066e-05,
      "loss": 1.5111,
      "step": 41700
    },
    {
      "epoch": 2.2273985130489273,
      "grad_norm": 0.2907065451145172,
      "learning_rate": 9.996013729843113e-05,
      "loss": 1.5012,
      "step": 41710
    },
    {
      "epoch": 2.227940454214439,
      "grad_norm": 0.3030034303665161,
      "learning_rate": 9.995946801178498e-05,
      "loss": 1.5042,
      "step": 41720
    },
    {
      "epoch": 2.2284823953799515,
      "grad_norm": 0.3327134847640991,
      "learning_rate": 9.995879315575511e-05,
      "loss": 1.5058,
      "step": 41730
    },
    {
      "epoch": 2.229024336545464,
      "grad_norm": 0.2516813576221466,
      "learning_rate": 9.995811273042515e-05,
      "loss": 1.5054,
      "step": 41740
    },
    {
      "epoch": 2.229566277710976,
      "grad_norm": 0.43475955724716187,
      "learning_rate": 9.995742673587937e-05,
      "loss": 1.5044,
      "step": 41750
    },
    {
      "epoch": 2.230108218876488,
      "grad_norm": 0.4820566773414612,
      "learning_rate": 9.995673517220277e-05,
      "loss": 1.4987,
      "step": 41760
    },
    {
      "epoch": 2.230108218876488,
      "eval_loss": 2.476296901702881,
      "eval_runtime": 21.9717,
      "eval_samples_per_second": 227.565,
      "eval_steps_per_second": 1.229,
      "step": 41760
    },
    {
      "epoch": 2.2306501600420003,
      "grad_norm": 0.503626823425293,
      "learning_rate": 9.995603803948098e-05,
      "loss": 1.5076,
      "step": 41770
    },
    {
      "epoch": 2.2311921012075127,
      "grad_norm": 0.3283330500125885,
      "learning_rate": 9.995533533780038e-05,
      "loss": 1.5055,
      "step": 41780
    },
    {
      "epoch": 2.231734042373025,
      "grad_norm": 0.5820431113243103,
      "learning_rate": 9.9954627067248e-05,
      "loss": 1.5118,
      "step": 41790
    },
    {
      "epoch": 2.2322759835385373,
      "grad_norm": 0.29778963327407837,
      "learning_rate": 9.995391322791162e-05,
      "loss": 1.5087,
      "step": 41800
    },
    {
      "epoch": 2.232817924704049,
      "grad_norm": 0.4030333459377289,
      "learning_rate": 9.995319381987962e-05,
      "loss": 1.4998,
      "step": 41810
    },
    {
      "epoch": 2.2333598658695615,
      "grad_norm": 0.22505435347557068,
      "learning_rate": 9.995246884324115e-05,
      "loss": 1.5041,
      "step": 41820
    },
    {
      "epoch": 2.233901807035074,
      "grad_norm": 0.4648926258087158,
      "learning_rate": 9.995173829808599e-05,
      "loss": 1.4911,
      "step": 41830
    },
    {
      "epoch": 2.234443748200586,
      "grad_norm": 0.2858235239982605,
      "learning_rate": 9.995100218450467e-05,
      "loss": 1.5019,
      "step": 41840
    },
    {
      "epoch": 2.2348231070164446,
      "eval_loss": 2.4754035472869873,
      "eval_runtime": 21.9771,
      "eval_samples_per_second": 227.51,
      "eval_steps_per_second": 1.229,
      "step": 41847
    },
    {
      "epoch": 2.2349856893660984,
      "grad_norm": 0.48889511823654175,
      "learning_rate": 9.995026050258835e-05,
      "loss": 1.5035,
      "step": 41850
    },
    {
      "epoch": 2.2355276305316103,
      "grad_norm": 0.32360124588012695,
      "learning_rate": 9.994951325242891e-05,
      "loss": 1.4995,
      "step": 41860
    },
    {
      "epoch": 2.2360695716971226,
      "grad_norm": 0.3390207588672638,
      "learning_rate": 9.994876043411891e-05,
      "loss": 1.5077,
      "step": 41870
    },
    {
      "epoch": 2.236611512862635,
      "grad_norm": 0.3765084743499756,
      "learning_rate": 9.994800204775163e-05,
      "loss": 1.5077,
      "step": 41880
    },
    {
      "epoch": 2.2371534540281472,
      "grad_norm": 0.26570597290992737,
      "learning_rate": 9.994723809342099e-05,
      "loss": 1.5082,
      "step": 41890
    },
    {
      "epoch": 2.237695395193659,
      "grad_norm": 0.2624066472053528,
      "learning_rate": 9.994646857122165e-05,
      "loss": 1.5153,
      "step": 41900
    },
    {
      "epoch": 2.2382373363591714,
      "grad_norm": 0.5075863003730774,
      "learning_rate": 9.994569348124891e-05,
      "loss": 1.5177,
      "step": 41910
    },
    {
      "epoch": 2.2387792775246838,
      "grad_norm": 0.2840348184108734,
      "learning_rate": 9.99449128235988e-05,
      "loss": 1.507,
      "step": 41920
    },
    {
      "epoch": 2.239321218690196,
      "grad_norm": 0.23727837204933167,
      "learning_rate": 9.994412659836802e-05,
      "loss": 1.5062,
      "step": 41930
    },
    {
      "epoch": 2.2395379951564007,
      "eval_loss": 2.478437662124634,
      "eval_runtime": 21.9787,
      "eval_samples_per_second": 227.493,
      "eval_steps_per_second": 1.228,
      "step": 41934
    },
    {
      "epoch": 2.239863159855708,
      "grad_norm": 0.5313767790794373,
      "learning_rate": 9.994333480565397e-05,
      "loss": 1.5117,
      "step": 41940
    },
    {
      "epoch": 2.2404051010212203,
      "grad_norm": 0.28299862146377563,
      "learning_rate": 9.994253744555473e-05,
      "loss": 1.5064,
      "step": 41950
    },
    {
      "epoch": 2.2409470421867326,
      "grad_norm": 0.2567594349384308,
      "learning_rate": 9.994173451816906e-05,
      "loss": 1.5007,
      "step": 41960
    },
    {
      "epoch": 2.241488983352245,
      "grad_norm": 0.24751020967960358,
      "learning_rate": 9.994092602359645e-05,
      "loss": 1.5152,
      "step": 41970
    },
    {
      "epoch": 2.242030924517757,
      "grad_norm": 0.5088745951652527,
      "learning_rate": 9.994011196193702e-05,
      "loss": 1.5042,
      "step": 41980
    },
    {
      "epoch": 2.242572865683269,
      "grad_norm": 0.2354600876569748,
      "learning_rate": 9.993929233329164e-05,
      "loss": 1.5111,
      "step": 41990
    },
    {
      "epoch": 2.2431148068487814,
      "grad_norm": 0.22222284972667694,
      "learning_rate": 9.993846713776183e-05,
      "loss": 1.5087,
      "step": 42000
    },
    {
      "epoch": 2.2436567480142937,
      "grad_norm": 0.20365022122859955,
      "learning_rate": 9.993763637544983e-05,
      "loss": 1.5099,
      "step": 42010
    },
    {
      "epoch": 2.244198689179806,
      "grad_norm": 0.2721654176712036,
      "learning_rate": 9.993680004645851e-05,
      "loss": 1.5031,
      "step": 42020
    },
    {
      "epoch": 2.2442528832963573,
      "eval_loss": 2.4772350788116455,
      "eval_runtime": 21.9795,
      "eval_samples_per_second": 227.485,
      "eval_steps_per_second": 1.228,
      "step": 42021
    },
    {
      "epoch": 2.2447406303453183,
      "grad_norm": 0.3124418556690216,
      "learning_rate": 9.993595815089151e-05,
      "loss": 1.503,
      "step": 42030
    },
    {
      "epoch": 2.24528257151083,
      "grad_norm": 0.304231196641922,
      "learning_rate": 9.993511068885311e-05,
      "loss": 1.4958,
      "step": 42040
    },
    {
      "epoch": 2.2458245126763425,
      "grad_norm": 0.23490315675735474,
      "learning_rate": 9.99342576604483e-05,
      "loss": 1.5052,
      "step": 42050
    },
    {
      "epoch": 2.246366453841855,
      "grad_norm": 0.5602021217346191,
      "learning_rate": 9.99333990657827e-05,
      "loss": 1.5007,
      "step": 42060
    },
    {
      "epoch": 2.246908395007367,
      "grad_norm": 0.25727346539497375,
      "learning_rate": 9.993253490496272e-05,
      "loss": 1.5107,
      "step": 42070
    },
    {
      "epoch": 2.247450336172879,
      "grad_norm": 0.2954048216342926,
      "learning_rate": 9.99316651780954e-05,
      "loss": 1.5127,
      "step": 42080
    },
    {
      "epoch": 2.2479922773383914,
      "grad_norm": 0.23384593427181244,
      "learning_rate": 9.993078988528848e-05,
      "loss": 1.5024,
      "step": 42090
    },
    {
      "epoch": 2.2485342185039037,
      "grad_norm": 0.2479962408542633,
      "learning_rate": 9.992990902665037e-05,
      "loss": 1.5007,
      "step": 42100
    },
    {
      "epoch": 2.2489677714363134,
      "eval_loss": 2.471235990524292,
      "eval_runtime": 21.9804,
      "eval_samples_per_second": 227.476,
      "eval_steps_per_second": 1.228,
      "step": 42108
    },
    {
      "epoch": 2.249076159669416,
      "grad_norm": 0.26295939087867737,
      "learning_rate": 9.99290226022902e-05,
      "loss": 1.5132,
      "step": 42110
    },
    {
      "epoch": 2.2496181008349283,
      "grad_norm": 0.6523755192756653,
      "learning_rate": 9.992813061231775e-05,
      "loss": 1.5043,
      "step": 42120
    },
    {
      "epoch": 2.25016004200044,
      "grad_norm": 0.3719354569911957,
      "learning_rate": 9.992723305684357e-05,
      "loss": 1.5014,
      "step": 42130
    },
    {
      "epoch": 2.2507019831659525,
      "grad_norm": 0.2825568616390228,
      "learning_rate": 9.99263299359788e-05,
      "loss": 1.5012,
      "step": 42140
    },
    {
      "epoch": 2.251243924331465,
      "grad_norm": 0.4288165867328644,
      "learning_rate": 9.992542124983531e-05,
      "loss": 1.5053,
      "step": 42150
    },
    {
      "epoch": 2.251785865496977,
      "grad_norm": 0.22682829201221466,
      "learning_rate": 9.992450699852571e-05,
      "loss": 1.501,
      "step": 42160
    },
    {
      "epoch": 2.252327806662489,
      "grad_norm": 0.409411758184433,
      "learning_rate": 9.992358718216321e-05,
      "loss": 1.5048,
      "step": 42170
    },
    {
      "epoch": 2.2528697478280013,
      "grad_norm": 0.6447529792785645,
      "learning_rate": 9.992266180086178e-05,
      "loss": 1.5141,
      "step": 42180
    },
    {
      "epoch": 2.2534116889935136,
      "grad_norm": 0.35520994663238525,
      "learning_rate": 9.992173085473603e-05,
      "loss": 1.5043,
      "step": 42190
    },
    {
      "epoch": 2.2536826595762696,
      "eval_loss": 2.463776111602783,
      "eval_runtime": 21.9731,
      "eval_samples_per_second": 227.551,
      "eval_steps_per_second": 1.229,
      "step": 42195
    },
    {
      "epoch": 2.253953630159026,
      "grad_norm": 0.24560990929603577,
      "learning_rate": 9.992079434390129e-05,
      "loss": 1.5081,
      "step": 42200
    },
    {
      "epoch": 2.2544955713245383,
      "grad_norm": 0.41237837076187134,
      "learning_rate": 9.991985226847358e-05,
      "loss": 1.5058,
      "step": 42210
    },
    {
      "epoch": 2.25503751249005,
      "grad_norm": 0.2179434448480606,
      "learning_rate": 9.99189046285696e-05,
      "loss": 1.5061,
      "step": 42220
    },
    {
      "epoch": 2.2555794536555624,
      "grad_norm": 0.2224990427494049,
      "learning_rate": 9.991795142430672e-05,
      "loss": 1.4924,
      "step": 42230
    },
    {
      "epoch": 2.2561213948210748,
      "grad_norm": 0.3237124979496002,
      "learning_rate": 9.991699265580304e-05,
      "loss": 1.5058,
      "step": 42240
    },
    {
      "epoch": 2.256663335986587,
      "grad_norm": 0.2218666821718216,
      "learning_rate": 9.991602832317731e-05,
      "loss": 1.4995,
      "step": 42250
    },
    {
      "epoch": 2.2572052771520994,
      "grad_norm": 0.38904523849487305,
      "learning_rate": 9.991505842654901e-05,
      "loss": 1.503,
      "step": 42260
    },
    {
      "epoch": 2.2577472183176113,
      "grad_norm": 0.341505765914917,
      "learning_rate": 9.991408296603827e-05,
      "loss": 1.5107,
      "step": 42270
    },
    {
      "epoch": 2.2582891594831236,
      "grad_norm": 0.242193341255188,
      "learning_rate": 9.991310194176593e-05,
      "loss": 1.495,
      "step": 42280
    },
    {
      "epoch": 2.258397547716226,
      "eval_loss": 2.4719505310058594,
      "eval_runtime": 21.9979,
      "eval_samples_per_second": 227.294,
      "eval_steps_per_second": 1.227,
      "step": 42282
    },
    {
      "epoch": 2.258831100648636,
      "grad_norm": 0.26852932572364807,
      "learning_rate": 9.991211535385352e-05,
      "loss": 1.504,
      "step": 42290
    },
    {
      "epoch": 2.259373041814148,
      "grad_norm": 0.32731330394744873,
      "learning_rate": 9.991112320242322e-05,
      "loss": 1.4927,
      "step": 42300
    },
    {
      "epoch": 2.25991498297966,
      "grad_norm": 0.30170223116874695,
      "learning_rate": 9.991012548759801e-05,
      "loss": 1.5,
      "step": 42310
    },
    {
      "epoch": 2.2604569241451724,
      "grad_norm": 0.2903915345668793,
      "learning_rate": 9.990912220950141e-05,
      "loss": 1.4963,
      "step": 42320
    },
    {
      "epoch": 2.2609988653106847,
      "grad_norm": 0.2501979470252991,
      "learning_rate": 9.990811336825771e-05,
      "loss": 1.5035,
      "step": 42330
    },
    {
      "epoch": 2.261540806476197,
      "grad_norm": 0.3588942289352417,
      "learning_rate": 9.990709896399191e-05,
      "loss": 1.506,
      "step": 42340
    },
    {
      "epoch": 2.262082747641709,
      "grad_norm": 0.2870718240737915,
      "learning_rate": 9.990607899682967e-05,
      "loss": 1.4929,
      "step": 42350
    },
    {
      "epoch": 2.2626246888072212,
      "grad_norm": 0.2925523817539215,
      "learning_rate": 9.990505346689732e-05,
      "loss": 1.5084,
      "step": 42360
    },
    {
      "epoch": 2.2631124358561823,
      "eval_loss": 2.4687459468841553,
      "eval_runtime": 22.2759,
      "eval_samples_per_second": 224.458,
      "eval_steps_per_second": 1.212,
      "step": 42369
    },
    {
      "epoch": 2.2631666299727335,
      "grad_norm": 0.28121358156204224,
      "learning_rate": 9.99040223743219e-05,
      "loss": 1.5021,
      "step": 42370
    },
    {
      "epoch": 2.263708571138246,
      "grad_norm": 0.40164533257484436,
      "learning_rate": 9.990298571923114e-05,
      "loss": 1.4933,
      "step": 42380
    },
    {
      "epoch": 2.264250512303758,
      "grad_norm": 0.1876879632472992,
      "learning_rate": 9.990194350175346e-05,
      "loss": 1.5118,
      "step": 42390
    },
    {
      "epoch": 2.26479245346927,
      "grad_norm": 0.3837462067604065,
      "learning_rate": 9.990089572201796e-05,
      "loss": 1.4987,
      "step": 42400
    },
    {
      "epoch": 2.2653343946347824,
      "grad_norm": 0.3707353472709656,
      "learning_rate": 9.989984238015445e-05,
      "loss": 1.5162,
      "step": 42410
    },
    {
      "epoch": 2.2658763358002947,
      "grad_norm": 0.5873448252677917,
      "learning_rate": 9.98987834762934e-05,
      "loss": 1.4921,
      "step": 42420
    },
    {
      "epoch": 2.266418276965807,
      "grad_norm": 0.43937060236930847,
      "learning_rate": 9.989771901056598e-05,
      "loss": 1.5083,
      "step": 42430
    },
    {
      "epoch": 2.2669602181313193,
      "grad_norm": 0.3484452962875366,
      "learning_rate": 9.989664898310405e-05,
      "loss": 1.5054,
      "step": 42440
    },
    {
      "epoch": 2.267502159296831,
      "grad_norm": 0.3318488597869873,
      "learning_rate": 9.98955733940402e-05,
      "loss": 1.4976,
      "step": 42450
    },
    {
      "epoch": 2.267827323996139,
      "eval_loss": 2.4685611724853516,
      "eval_runtime": 22.0052,
      "eval_samples_per_second": 227.219,
      "eval_steps_per_second": 1.227,
      "step": 42456
    },
    {
      "epoch": 2.2680441004623435,
      "grad_norm": 0.3194591999053955,
      "learning_rate": 9.989449224350758e-05,
      "loss": 1.504,
      "step": 42460
    },
    {
      "epoch": 2.268586041627856,
      "grad_norm": 0.2176416665315628,
      "learning_rate": 9.989340553164021e-05,
      "loss": 1.5006,
      "step": 42470
    },
    {
      "epoch": 2.269127982793368,
      "grad_norm": 0.2666696608066559,
      "learning_rate": 9.989231325857266e-05,
      "loss": 1.5008,
      "step": 42480
    },
    {
      "epoch": 2.2696699239588805,
      "grad_norm": 0.5212827324867249,
      "learning_rate": 9.989121542444023e-05,
      "loss": 1.5013,
      "step": 42490
    },
    {
      "epoch": 2.2702118651243923,
      "grad_norm": 0.33473333716392517,
      "learning_rate": 9.989011202937896e-05,
      "loss": 1.5156,
      "step": 42500
    },
    {
      "epoch": 2.2707538062899046,
      "grad_norm": 0.46315068006515503,
      "learning_rate": 9.988900307352549e-05,
      "loss": 1.4865,
      "step": 42510
    },
    {
      "epoch": 2.271295747455417,
      "grad_norm": 0.3576599657535553,
      "learning_rate": 9.98878885570172e-05,
      "loss": 1.5077,
      "step": 42520
    },
    {
      "epoch": 2.2718376886209293,
      "grad_norm": 0.22602027654647827,
      "learning_rate": 9.988676847999218e-05,
      "loss": 1.5185,
      "step": 42530
    },
    {
      "epoch": 2.272379629786441,
      "grad_norm": 0.2150840014219284,
      "learning_rate": 9.988564284258916e-05,
      "loss": 1.5062,
      "step": 42540
    },
    {
      "epoch": 2.272542212136095,
      "eval_loss": 2.4602785110473633,
      "eval_runtime": 21.997,
      "eval_samples_per_second": 227.304,
      "eval_steps_per_second": 1.227,
      "step": 42543
    },
    {
      "epoch": 2.2729215709519535,
      "grad_norm": 0.3397773206233978,
      "learning_rate": 9.988451164494757e-05,
      "loss": 1.5026,
      "step": 42550
    },
    {
      "epoch": 2.273463512117466,
      "grad_norm": 0.213679701089859,
      "learning_rate": 9.988337488720753e-05,
      "loss": 1.5014,
      "step": 42560
    },
    {
      "epoch": 2.274005453282978,
      "grad_norm": 0.21069695055484772,
      "learning_rate": 9.98822325695099e-05,
      "loss": 1.5098,
      "step": 42570
    },
    {
      "epoch": 2.27454739444849,
      "grad_norm": 0.38944995403289795,
      "learning_rate": 9.988108469199613e-05,
      "loss": 1.496,
      "step": 42580
    },
    {
      "epoch": 2.2750893356140023,
      "grad_norm": 0.36275213956832886,
      "learning_rate": 9.987993125480848e-05,
      "loss": 1.5009,
      "step": 42590
    },
    {
      "epoch": 2.2756312767795146,
      "grad_norm": 0.27559274435043335,
      "learning_rate": 9.987877225808976e-05,
      "loss": 1.4994,
      "step": 42600
    },
    {
      "epoch": 2.276173217945027,
      "grad_norm": 0.365401029586792,
      "learning_rate": 9.987760770198359e-05,
      "loss": 1.5061,
      "step": 42610
    },
    {
      "epoch": 2.2767151591105392,
      "grad_norm": 0.29308363795280457,
      "learning_rate": 9.987643758663422e-05,
      "loss": 1.4963,
      "step": 42620
    },
    {
      "epoch": 2.277257100276051,
      "grad_norm": 0.43470004200935364,
      "learning_rate": 9.98752619121866e-05,
      "loss": 1.496,
      "step": 42630
    },
    {
      "epoch": 2.277257100276051,
      "eval_loss": 2.4641952514648438,
      "eval_runtime": 22.0028,
      "eval_samples_per_second": 227.243,
      "eval_steps_per_second": 1.227,
      "step": 42630
    },
    {
      "epoch": 2.2777990414415634,
      "grad_norm": 0.21174076199531555,
      "learning_rate": 9.987408067878636e-05,
      "loss": 1.5095,
      "step": 42640
    },
    {
      "epoch": 2.2783409826070757,
      "grad_norm": 0.4398151934146881,
      "learning_rate": 9.987289388657982e-05,
      "loss": 1.4995,
      "step": 42650
    },
    {
      "epoch": 2.278882923772588,
      "grad_norm": 0.41599544882774353,
      "learning_rate": 9.987170153571402e-05,
      "loss": 1.4949,
      "step": 42660
    },
    {
      "epoch": 2.2794248649381004,
      "grad_norm": 0.34385424852371216,
      "learning_rate": 9.987050362633663e-05,
      "loss": 1.5031,
      "step": 42670
    },
    {
      "epoch": 2.2799668061036122,
      "grad_norm": 0.242776557803154,
      "learning_rate": 9.986930015859607e-05,
      "loss": 1.4902,
      "step": 42680
    },
    {
      "epoch": 2.2805087472691246,
      "grad_norm": 0.4597446322441101,
      "learning_rate": 9.986809113264142e-05,
      "loss": 1.5039,
      "step": 42690
    },
    {
      "epoch": 2.281050688434637,
      "grad_norm": 0.25635120272636414,
      "learning_rate": 9.986687654862242e-05,
      "loss": 1.5071,
      "step": 42700
    },
    {
      "epoch": 2.281592629600149,
      "grad_norm": 0.23346717655658722,
      "learning_rate": 9.986565640668955e-05,
      "loss": 1.5025,
      "step": 42710
    },
    {
      "epoch": 2.2819719884160077,
      "eval_loss": 2.4699313640594482,
      "eval_runtime": 22.0353,
      "eval_samples_per_second": 226.908,
      "eval_steps_per_second": 1.225,
      "step": 42717
    },
    {
      "epoch": 2.282134570765661,
      "grad_norm": 0.4288314878940582,
      "learning_rate": 9.986443070699393e-05,
      "loss": 1.5069,
      "step": 42720
    },
    {
      "epoch": 2.2826765119311734,
      "grad_norm": 0.3048160672187805,
      "learning_rate": 9.986319944968745e-05,
      "loss": 1.5084,
      "step": 42730
    },
    {
      "epoch": 2.2832184530966857,
      "grad_norm": 0.2685145139694214,
      "learning_rate": 9.986196263492259e-05,
      "loss": 1.5078,
      "step": 42740
    },
    {
      "epoch": 2.283760394262198,
      "grad_norm": 0.36465343832969666,
      "learning_rate": 9.986072026285257e-05,
      "loss": 1.4972,
      "step": 42750
    },
    {
      "epoch": 2.28430233542771,
      "grad_norm": 0.27710703015327454,
      "learning_rate": 9.985947233363128e-05,
      "loss": 1.4929,
      "step": 42760
    },
    {
      "epoch": 2.284844276593222,
      "grad_norm": 0.5926934480667114,
      "learning_rate": 9.985821884741333e-05,
      "loss": 1.5015,
      "step": 42770
    },
    {
      "epoch": 2.2853862177587345,
      "grad_norm": 0.34573689103126526,
      "learning_rate": 9.985695980435398e-05,
      "loss": 1.5047,
      "step": 42780
    },
    {
      "epoch": 2.285928158924247,
      "grad_norm": 0.33093443512916565,
      "learning_rate": 9.985569520460918e-05,
      "loss": 1.4997,
      "step": 42790
    },
    {
      "epoch": 2.286470100089759,
      "grad_norm": 0.5260155200958252,
      "learning_rate": 9.985442504833562e-05,
      "loss": 1.5034,
      "step": 42800
    },
    {
      "epoch": 2.286686876555964,
      "eval_loss": 2.4743220806121826,
      "eval_runtime": 22.0148,
      "eval_samples_per_second": 227.12,
      "eval_steps_per_second": 1.226,
      "step": 42804
    },
    {
      "epoch": 2.287012041255271,
      "grad_norm": 0.22900910675525665,
      "learning_rate": 9.985314933569063e-05,
      "loss": 1.5057,
      "step": 42810
    },
    {
      "epoch": 2.2875539824207833,
      "grad_norm": 0.47526514530181885,
      "learning_rate": 9.985186806683222e-05,
      "loss": 1.4987,
      "step": 42820
    },
    {
      "epoch": 2.2880959235862957,
      "grad_norm": 0.2705422043800354,
      "learning_rate": 9.985058124191914e-05,
      "loss": 1.5031,
      "step": 42830
    },
    {
      "epoch": 2.288637864751808,
      "grad_norm": 0.21489684283733368,
      "learning_rate": 9.984928886111076e-05,
      "loss": 1.5001,
      "step": 42840
    },
    {
      "epoch": 2.2891798059173203,
      "grad_norm": 0.44066211581230164,
      "learning_rate": 9.98479909245672e-05,
      "loss": 1.5006,
      "step": 42850
    },
    {
      "epoch": 2.289721747082832,
      "grad_norm": 0.4437292516231537,
      "learning_rate": 9.984668743244921e-05,
      "loss": 1.5121,
      "step": 42860
    },
    {
      "epoch": 2.2902636882483445,
      "grad_norm": 0.41992589831352234,
      "learning_rate": 9.984537838491833e-05,
      "loss": 1.4998,
      "step": 42870
    },
    {
      "epoch": 2.290805629413857,
      "grad_norm": 0.26407718658447266,
      "learning_rate": 9.984406378213664e-05,
      "loss": 1.5009,
      "step": 42880
    },
    {
      "epoch": 2.291347570579369,
      "grad_norm": 0.35871899127960205,
      "learning_rate": 9.984274362426703e-05,
      "loss": 1.5003,
      "step": 42890
    },
    {
      "epoch": 2.2914017646959204,
      "eval_loss": 2.4577038288116455,
      "eval_runtime": 22.0057,
      "eval_samples_per_second": 227.214,
      "eval_steps_per_second": 1.227,
      "step": 42891
    },
    {
      "epoch": 2.2918895117448814,
      "grad_norm": 0.38924890756607056,
      "learning_rate": 9.984141791147303e-05,
      "loss": 1.5098,
      "step": 42900
    },
    {
      "epoch": 2.2924314529103933,
      "grad_norm": 0.46913039684295654,
      "learning_rate": 9.984008664391888e-05,
      "loss": 1.5006,
      "step": 42910
    },
    {
      "epoch": 2.2929733940759056,
      "grad_norm": 0.6839142441749573,
      "learning_rate": 9.983874982176944e-05,
      "loss": 1.5011,
      "step": 42920
    },
    {
      "epoch": 2.293515335241418,
      "grad_norm": 0.37098947167396545,
      "learning_rate": 9.983740744519037e-05,
      "loss": 1.5078,
      "step": 42930
    },
    {
      "epoch": 2.2940572764069302,
      "grad_norm": 0.4213247001171112,
      "learning_rate": 9.983605951434792e-05,
      "loss": 1.4979,
      "step": 42940
    },
    {
      "epoch": 2.294599217572442,
      "grad_norm": 0.37451714277267456,
      "learning_rate": 9.983470602940907e-05,
      "loss": 1.5055,
      "step": 42950
    },
    {
      "epoch": 2.2951411587379544,
      "grad_norm": 0.3335314989089966,
      "learning_rate": 9.983334699054149e-05,
      "loss": 1.4985,
      "step": 42960
    },
    {
      "epoch": 2.2956830999034668,
      "grad_norm": 0.5755605697631836,
      "learning_rate": 9.983198239791353e-05,
      "loss": 1.5072,
      "step": 42970
    },
    {
      "epoch": 2.2961166528358765,
      "eval_loss": 2.46994686126709,
      "eval_runtime": 22.0001,
      "eval_samples_per_second": 227.272,
      "eval_steps_per_second": 1.227,
      "step": 42978
    },
    {
      "epoch": 2.296225041068979,
      "grad_norm": 0.4055348038673401,
      "learning_rate": 9.983061225169423e-05,
      "loss": 1.514,
      "step": 42980
    },
    {
      "epoch": 2.296766982234491,
      "grad_norm": 0.41009873151779175,
      "learning_rate": 9.982923655205335e-05,
      "loss": 1.4959,
      "step": 42990
    },
    {
      "epoch": 2.2973089234000033,
      "grad_norm": 0.30407077074050903,
      "learning_rate": 9.982785529916124e-05,
      "loss": 1.4958,
      "step": 43000
    },
    {
      "epoch": 2.2978508645655156,
      "grad_norm": 0.3343801498413086,
      "learning_rate": 9.982646849318906e-05,
      "loss": 1.5003,
      "step": 43010
    },
    {
      "epoch": 2.298392805731028,
      "grad_norm": 0.4572197496891022,
      "learning_rate": 9.982507613430856e-05,
      "loss": 1.4998,
      "step": 43020
    },
    {
      "epoch": 2.29893474689654,
      "grad_norm": 0.3537229895591736,
      "learning_rate": 9.982367822269225e-05,
      "loss": 1.5059,
      "step": 43030
    },
    {
      "epoch": 2.299476688062052,
      "grad_norm": 0.4736888110637665,
      "learning_rate": 9.982227475851328e-05,
      "loss": 1.4988,
      "step": 43040
    },
    {
      "epoch": 2.3000186292275644,
      "grad_norm": 0.32492560148239136,
      "learning_rate": 9.982086574194552e-05,
      "loss": 1.5023,
      "step": 43050
    },
    {
      "epoch": 2.3005605703930767,
      "grad_norm": 0.19111861288547516,
      "learning_rate": 9.98194511731635e-05,
      "loss": 1.5075,
      "step": 43060
    },
    {
      "epoch": 2.3008315409758326,
      "eval_loss": 2.4626097679138184,
      "eval_runtime": 22.0143,
      "eval_samples_per_second": 227.125,
      "eval_steps_per_second": 1.226,
      "step": 43065
    },
    {
      "epoch": 2.301102511558589,
      "grad_norm": 0.416725754737854,
      "learning_rate": 9.981803105234246e-05,
      "loss": 1.5009,
      "step": 43070
    },
    {
      "epoch": 2.3016444527241013,
      "grad_norm": 0.5482866168022156,
      "learning_rate": 9.981660537965833e-05,
      "loss": 1.5044,
      "step": 43080
    },
    {
      "epoch": 2.302186393889613,
      "grad_norm": 0.22884666919708252,
      "learning_rate": 9.981517415528766e-05,
      "loss": 1.4965,
      "step": 43090
    },
    {
      "epoch": 2.3027283350551255,
      "grad_norm": 0.3131319284439087,
      "learning_rate": 9.98137373794078e-05,
      "loss": 1.5089,
      "step": 43100
    },
    {
      "epoch": 2.303270276220638,
      "grad_norm": 0.22370652854442596,
      "learning_rate": 9.981229505219673e-05,
      "loss": 1.5051,
      "step": 43110
    },
    {
      "epoch": 2.30381221738615,
      "grad_norm": 0.2758335471153259,
      "learning_rate": 9.98108471738331e-05,
      "loss": 1.5001,
      "step": 43120
    },
    {
      "epoch": 2.3043541585516625,
      "grad_norm": 0.309836208820343,
      "learning_rate": 9.980939374449627e-05,
      "loss": 1.5022,
      "step": 43130
    },
    {
      "epoch": 2.3048960997171744,
      "grad_norm": 0.2783421576023102,
      "learning_rate": 9.980793476436628e-05,
      "loss": 1.4996,
      "step": 43140
    },
    {
      "epoch": 2.3054380408826867,
      "grad_norm": 0.26890337467193604,
      "learning_rate": 9.980647023362388e-05,
      "loss": 1.5025,
      "step": 43150
    },
    {
      "epoch": 2.305546429115789,
      "eval_loss": 2.4663755893707275,
      "eval_runtime": 21.9786,
      "eval_samples_per_second": 227.494,
      "eval_steps_per_second": 1.228,
      "step": 43152
    },
    {
      "epoch": 2.305979982048199,
      "grad_norm": 0.26960644125938416,
      "learning_rate": 9.980500015245045e-05,
      "loss": 1.5067,
      "step": 43160
    },
    {
      "epoch": 2.3065219232137113,
      "grad_norm": 0.32048168778419495,
      "learning_rate": 9.980352452102815e-05,
      "loss": 1.4944,
      "step": 43170
    },
    {
      "epoch": 2.307063864379223,
      "grad_norm": 0.36742621660232544,
      "learning_rate": 9.980204333953974e-05,
      "loss": 1.4997,
      "step": 43180
    },
    {
      "epoch": 2.3076058055447355,
      "grad_norm": 0.38924074172973633,
      "learning_rate": 9.980055660816872e-05,
      "loss": 1.5018,
      "step": 43190
    },
    {
      "epoch": 2.308147746710248,
      "grad_norm": 0.7526839971542358,
      "learning_rate": 9.979906432709925e-05,
      "loss": 1.5036,
      "step": 43200
    },
    {
      "epoch": 2.30868968787576,
      "grad_norm": 0.49918413162231445,
      "learning_rate": 9.97975664965162e-05,
      "loss": 1.5053,
      "step": 43210
    },
    {
      "epoch": 2.309231629041272,
      "grad_norm": 0.21481403708457947,
      "learning_rate": 9.979606311660506e-05,
      "loss": 1.5091,
      "step": 43220
    },
    {
      "epoch": 2.3097735702067843,
      "grad_norm": 0.26321297883987427,
      "learning_rate": 9.979455418755215e-05,
      "loss": 1.4983,
      "step": 43230
    },
    {
      "epoch": 2.3102613172557454,
      "eval_loss": 2.472028970718384,
      "eval_runtime": 21.9804,
      "eval_samples_per_second": 227.476,
      "eval_steps_per_second": 1.228,
      "step": 43239
    },
    {
      "epoch": 2.3103155113722966,
      "grad_norm": 0.2163982391357422,
      "learning_rate": 9.979303970954434e-05,
      "loss": 1.5075,
      "step": 43240
    },
    {
      "epoch": 2.310857452537809,
      "grad_norm": 0.2770186960697174,
      "learning_rate": 9.979151968276922e-05,
      "loss": 1.5033,
      "step": 43250
    },
    {
      "epoch": 2.3113993937033213,
      "grad_norm": 0.3074730932712555,
      "learning_rate": 9.978999410741514e-05,
      "loss": 1.4884,
      "step": 43260
    },
    {
      "epoch": 2.311941334868833,
      "grad_norm": 0.30169418454170227,
      "learning_rate": 9.978846298367103e-05,
      "loss": 1.4956,
      "step": 43270
    },
    {
      "epoch": 2.3124832760343454,
      "grad_norm": 0.23095916211605072,
      "learning_rate": 9.978692631172657e-05,
      "loss": 1.4884,
      "step": 43280
    },
    {
      "epoch": 2.3130252171998578,
      "grad_norm": 0.28143152594566345,
      "learning_rate": 9.978538409177212e-05,
      "loss": 1.494,
      "step": 43290
    },
    {
      "epoch": 2.31356715836537,
      "grad_norm": 0.3327721953392029,
      "learning_rate": 9.978383632399876e-05,
      "loss": 1.5017,
      "step": 43300
    },
    {
      "epoch": 2.3141090995308824,
      "grad_norm": 0.2959439754486084,
      "learning_rate": 9.978228300859817e-05,
      "loss": 1.4953,
      "step": 43310
    },
    {
      "epoch": 2.3146510406963943,
      "grad_norm": 0.3569033741950989,
      "learning_rate": 9.978072414576277e-05,
      "loss": 1.509,
      "step": 43320
    },
    {
      "epoch": 2.314976205395702,
      "eval_loss": 2.471214771270752,
      "eval_runtime": 21.9774,
      "eval_samples_per_second": 227.506,
      "eval_steps_per_second": 1.229,
      "step": 43326
    },
    {
      "epoch": 2.3151929818619066,
      "grad_norm": 0.37210530042648315,
      "learning_rate": 9.97791597356857e-05,
      "loss": 1.4907,
      "step": 43330
    },
    {
      "epoch": 2.315734923027419,
      "grad_norm": 0.22837452590465546,
      "learning_rate": 9.977758977856074e-05,
      "loss": 1.494,
      "step": 43340
    },
    {
      "epoch": 2.316276864192931,
      "grad_norm": 0.351477712392807,
      "learning_rate": 9.977601427458235e-05,
      "loss": 1.4953,
      "step": 43350
    },
    {
      "epoch": 2.316818805358443,
      "grad_norm": 0.4676024913787842,
      "learning_rate": 9.97744332239457e-05,
      "loss": 1.5027,
      "step": 43360
    },
    {
      "epoch": 2.3173607465239554,
      "grad_norm": 0.3546915054321289,
      "learning_rate": 9.977284662684668e-05,
      "loss": 1.509,
      "step": 43370
    },
    {
      "epoch": 2.3179026876894677,
      "grad_norm": 0.45756760239601135,
      "learning_rate": 9.977125448348178e-05,
      "loss": 1.4956,
      "step": 43380
    },
    {
      "epoch": 2.31844462885498,
      "grad_norm": 0.44374728202819824,
      "learning_rate": 9.976965679404827e-05,
      "loss": 1.4851,
      "step": 43390
    },
    {
      "epoch": 2.318986570020492,
      "grad_norm": 0.2877369225025177,
      "learning_rate": 9.976805355874404e-05,
      "loss": 1.4979,
      "step": 43400
    },
    {
      "epoch": 2.3195285111860042,
      "grad_norm": 0.4321652352809906,
      "learning_rate": 9.976644477776768e-05,
      "loss": 1.5,
      "step": 43410
    },
    {
      "epoch": 2.319691093535658,
      "eval_loss": 2.4682838916778564,
      "eval_runtime": 21.9705,
      "eval_samples_per_second": 227.578,
      "eval_steps_per_second": 1.229,
      "step": 43413
    },
    {
      "epoch": 2.3200704523515165,
      "grad_norm": 0.3509621024131775,
      "learning_rate": 9.976483045131853e-05,
      "loss": 1.4899,
      "step": 43420
    },
    {
      "epoch": 2.320612393517029,
      "grad_norm": 0.2721761167049408,
      "learning_rate": 9.976321057959651e-05,
      "loss": 1.4967,
      "step": 43430
    },
    {
      "epoch": 2.321154334682541,
      "grad_norm": 0.27356627583503723,
      "learning_rate": 9.976158516280231e-05,
      "loss": 1.5021,
      "step": 43440
    },
    {
      "epoch": 2.321696275848053,
      "grad_norm": 0.30389806628227234,
      "learning_rate": 9.975995420113729e-05,
      "loss": 1.4943,
      "step": 43450
    },
    {
      "epoch": 2.3222382170135654,
      "grad_norm": 0.35005050897598267,
      "learning_rate": 9.975831769480345e-05,
      "loss": 1.5019,
      "step": 43460
    },
    {
      "epoch": 2.3227801581790777,
      "grad_norm": 0.19983713328838348,
      "learning_rate": 9.975667564400355e-05,
      "loss": 1.5049,
      "step": 43470
    },
    {
      "epoch": 2.32332209934459,
      "grad_norm": 0.27769649028778076,
      "learning_rate": 9.975502804894097e-05,
      "loss": 1.4946,
      "step": 43480
    },
    {
      "epoch": 2.3238640405101023,
      "grad_norm": 0.20963497459888458,
      "learning_rate": 9.975337490981984e-05,
      "loss": 1.4937,
      "step": 43490
    },
    {
      "epoch": 2.324405981675614,
      "grad_norm": 0.21975292265415192,
      "learning_rate": 9.975171622684492e-05,
      "loss": 1.4993,
      "step": 43500
    },
    {
      "epoch": 2.324405981675614,
      "eval_loss": 2.477883815765381,
      "eval_runtime": 21.9663,
      "eval_samples_per_second": 227.622,
      "eval_steps_per_second": 1.229,
      "step": 43500
    },
    {
      "epoch": 2.3249479228411265,
      "grad_norm": 0.3812916874885559,
      "learning_rate": 9.97500520002217e-05,
      "loss": 1.5059,
      "step": 43510
    },
    {
      "epoch": 2.325489864006639,
      "grad_norm": 0.29676496982574463,
      "learning_rate": 9.974838223015631e-05,
      "loss": 1.4986,
      "step": 43520
    },
    {
      "epoch": 2.326031805172151,
      "grad_norm": 0.3537111282348633,
      "learning_rate": 9.97467069168556e-05,
      "loss": 1.4903,
      "step": 43530
    },
    {
      "epoch": 2.3265737463376635,
      "grad_norm": 0.2975006699562073,
      "learning_rate": 9.974502606052711e-05,
      "loss": 1.4985,
      "step": 43540
    },
    {
      "epoch": 2.3271156875031753,
      "grad_norm": 0.42832067608833313,
      "learning_rate": 9.974333966137907e-05,
      "loss": 1.4866,
      "step": 43550
    },
    {
      "epoch": 2.3276576286686876,
      "grad_norm": 0.29303500056266785,
      "learning_rate": 9.974164771962035e-05,
      "loss": 1.496,
      "step": 43560
    },
    {
      "epoch": 2.3281995698342,
      "grad_norm": 0.330950528383255,
      "learning_rate": 9.973995023546055e-05,
      "loss": 1.4929,
      "step": 43570
    },
    {
      "epoch": 2.3287415109997123,
      "grad_norm": 0.20558315515518188,
      "learning_rate": 9.973824720911e-05,
      "loss": 1.5115,
      "step": 43580
    },
    {
      "epoch": 2.3291208698155708,
      "eval_loss": 2.4667203426361084,
      "eval_runtime": 21.9785,
      "eval_samples_per_second": 227.495,
      "eval_steps_per_second": 1.228,
      "step": 43587
    },
    {
      "epoch": 2.329283452165224,
      "grad_norm": 0.36295872926712036,
      "learning_rate": 9.973653864077958e-05,
      "loss": 1.4857,
      "step": 43590
    },
    {
      "epoch": 2.3298253933307365,
      "grad_norm": 0.6293714046478271,
      "learning_rate": 9.973482453068099e-05,
      "loss": 1.497,
      "step": 43600
    },
    {
      "epoch": 2.3303673344962488,
      "grad_norm": 0.4836002290248871,
      "learning_rate": 9.973310487902657e-05,
      "loss": 1.5107,
      "step": 43610
    },
    {
      "epoch": 2.330909275661761,
      "grad_norm": 0.3348879814147949,
      "learning_rate": 9.973137968602932e-05,
      "loss": 1.4975,
      "step": 43620
    },
    {
      "epoch": 2.331451216827273,
      "grad_norm": 0.24856191873550415,
      "learning_rate": 9.972964895190295e-05,
      "loss": 1.4994,
      "step": 43630
    },
    {
      "epoch": 2.3319931579927853,
      "grad_norm": 0.2245061695575714,
      "learning_rate": 9.972791267686188e-05,
      "loss": 1.494,
      "step": 43640
    },
    {
      "epoch": 2.3325350991582976,
      "grad_norm": 0.2887219786643982,
      "learning_rate": 9.972617086112116e-05,
      "loss": 1.5022,
      "step": 43650
    },
    {
      "epoch": 2.33307704032381,
      "grad_norm": 0.5362899303436279,
      "learning_rate": 9.97244235048966e-05,
      "loss": 1.5091,
      "step": 43660
    },
    {
      "epoch": 2.3336189814893222,
      "grad_norm": 0.4345687925815582,
      "learning_rate": 9.972267060840461e-05,
      "loss": 1.5024,
      "step": 43670
    },
    {
      "epoch": 2.333835757955527,
      "eval_loss": 2.4626216888427734,
      "eval_runtime": 21.9768,
      "eval_samples_per_second": 227.513,
      "eval_steps_per_second": 1.229,
      "step": 43674
    },
    {
      "epoch": 2.334160922654834,
      "grad_norm": 0.42070385813713074,
      "learning_rate": 9.972091217186236e-05,
      "loss": 1.5046,
      "step": 43680
    },
    {
      "epoch": 2.3347028638203464,
      "grad_norm": 0.5176911950111389,
      "learning_rate": 9.971914819548766e-05,
      "loss": 1.5029,
      "step": 43690
    },
    {
      "epoch": 2.3352448049858587,
      "grad_norm": 0.3545333743095398,
      "learning_rate": 9.971737867949903e-05,
      "loss": 1.4942,
      "step": 43700
    },
    {
      "epoch": 2.335786746151371,
      "grad_norm": 0.2512124180793762,
      "learning_rate": 9.971560362411569e-05,
      "loss": 1.5021,
      "step": 43710
    },
    {
      "epoch": 2.3363286873168834,
      "grad_norm": 0.32333534955978394,
      "learning_rate": 9.971382302955748e-05,
      "loss": 1.4935,
      "step": 43720
    },
    {
      "epoch": 2.3368706284823952,
      "grad_norm": 0.431264728307724,
      "learning_rate": 9.971203689604504e-05,
      "loss": 1.5011,
      "step": 43730
    },
    {
      "epoch": 2.3374125696479076,
      "grad_norm": 0.26169353723526,
      "learning_rate": 9.971024522379957e-05,
      "loss": 1.5027,
      "step": 43740
    },
    {
      "epoch": 2.33795451081342,
      "grad_norm": 0.44165289402008057,
      "learning_rate": 9.970844801304303e-05,
      "loss": 1.4827,
      "step": 43750
    },
    {
      "epoch": 2.338496451978932,
      "grad_norm": 0.3183143734931946,
      "learning_rate": 9.970664526399806e-05,
      "loss": 1.4988,
      "step": 43760
    },
    {
      "epoch": 2.3385506460954835,
      "eval_loss": 2.4591445922851562,
      "eval_runtime": 21.9816,
      "eval_samples_per_second": 227.463,
      "eval_steps_per_second": 1.228,
      "step": 43761
    },
    {
      "epoch": 2.3390383931444445,
      "grad_norm": 0.22885079681873322,
      "learning_rate": 9.970483697688798e-05,
      "loss": 1.4994,
      "step": 43770
    },
    {
      "epoch": 2.3395803343099564,
      "grad_norm": 0.20680442452430725,
      "learning_rate": 9.970302315193677e-05,
      "loss": 1.4886,
      "step": 43780
    },
    {
      "epoch": 2.3401222754754687,
      "grad_norm": 0.28351494669914246,
      "learning_rate": 9.970120378936914e-05,
      "loss": 1.4918,
      "step": 43790
    },
    {
      "epoch": 2.340664216640981,
      "grad_norm": 0.3337201774120331,
      "learning_rate": 9.969937888941046e-05,
      "loss": 1.4897,
      "step": 43800
    },
    {
      "epoch": 2.3412061578064933,
      "grad_norm": 0.37120136618614197,
      "learning_rate": 9.969754845228676e-05,
      "loss": 1.4988,
      "step": 43810
    },
    {
      "epoch": 2.341748098972005,
      "grad_norm": 0.2082194685935974,
      "learning_rate": 9.969571247822486e-05,
      "loss": 1.5011,
      "step": 43820
    },
    {
      "epoch": 2.3422900401375175,
      "grad_norm": 0.3116413652896881,
      "learning_rate": 9.969387096745211e-05,
      "loss": 1.4943,
      "step": 43830
    },
    {
      "epoch": 2.34283198130303,
      "grad_norm": 0.18814930319786072,
      "learning_rate": 9.969202392019668e-05,
      "loss": 1.5032,
      "step": 43840
    },
    {
      "epoch": 2.3432655342354396,
      "eval_loss": 2.4612669944763184,
      "eval_runtime": 21.9782,
      "eval_samples_per_second": 227.498,
      "eval_steps_per_second": 1.228,
      "step": 43848
    },
    {
      "epoch": 2.343373922468542,
      "grad_norm": 0.19654281437397003,
      "learning_rate": 9.969017133668738e-05,
      "loss": 1.4942,
      "step": 43850
    },
    {
      "epoch": 2.343915863634054,
      "grad_norm": 0.49275675415992737,
      "learning_rate": 9.968831321715365e-05,
      "loss": 1.4915,
      "step": 43860
    },
    {
      "epoch": 2.3444578047995663,
      "grad_norm": 0.26238903403282166,
      "learning_rate": 9.968644956182572e-05,
      "loss": 1.5072,
      "step": 43870
    },
    {
      "epoch": 2.3449997459650787,
      "grad_norm": 0.43797948956489563,
      "learning_rate": 9.968458037093442e-05,
      "loss": 1.4991,
      "step": 43880
    },
    {
      "epoch": 2.345541687130591,
      "grad_norm": 0.34569740295410156,
      "learning_rate": 9.968270564471131e-05,
      "loss": 1.5006,
      "step": 43890
    },
    {
      "epoch": 2.3460836282961033,
      "grad_norm": 0.3247714340686798,
      "learning_rate": 9.96808253833886e-05,
      "loss": 1.5022,
      "step": 43900
    },
    {
      "epoch": 2.346625569461615,
      "grad_norm": 0.3044470250606537,
      "learning_rate": 9.967893958719924e-05,
      "loss": 1.4978,
      "step": 43910
    },
    {
      "epoch": 2.3471675106271275,
      "grad_norm": 0.26124030351638794,
      "learning_rate": 9.967704825637681e-05,
      "loss": 1.5,
      "step": 43920
    },
    {
      "epoch": 2.34770945179264,
      "grad_norm": 0.2891956567764282,
      "learning_rate": 9.967515139115562e-05,
      "loss": 1.5099,
      "step": 43930
    },
    {
      "epoch": 2.3479804223753957,
      "eval_loss": 2.4739420413970947,
      "eval_runtime": 21.9776,
      "eval_samples_per_second": 227.505,
      "eval_steps_per_second": 1.229,
      "step": 43935
    },
    {
      "epoch": 2.348251392958152,
      "grad_norm": 0.3045171797275543,
      "learning_rate": 9.967324899177062e-05,
      "loss": 1.4922,
      "step": 43940
    },
    {
      "epoch": 2.3487933341236644,
      "grad_norm": 0.26287245750427246,
      "learning_rate": 9.96713410584575e-05,
      "loss": 1.5083,
      "step": 43950
    },
    {
      "epoch": 2.3493352752891763,
      "grad_norm": 0.37551262974739075,
      "learning_rate": 9.96694275914526e-05,
      "loss": 1.4992,
      "step": 43960
    },
    {
      "epoch": 2.3498772164546886,
      "grad_norm": 0.4330720901489258,
      "learning_rate": 9.966750859099294e-05,
      "loss": 1.4883,
      "step": 43970
    },
    {
      "epoch": 2.350419157620201,
      "grad_norm": 0.21787714958190918,
      "learning_rate": 9.966558405731624e-05,
      "loss": 1.4937,
      "step": 43980
    },
    {
      "epoch": 2.3509610987857132,
      "grad_norm": 0.28836148977279663,
      "learning_rate": 9.96636539906609e-05,
      "loss": 1.4988,
      "step": 43990
    },
    {
      "epoch": 2.351503039951225,
      "grad_norm": 0.29318341612815857,
      "learning_rate": 9.966171839126601e-05,
      "loss": 1.5102,
      "step": 44000
    },
    {
      "epoch": 2.3520449811167374,
      "grad_norm": 0.5823042988777161,
      "learning_rate": 9.965977725937138e-05,
      "loss": 1.4957,
      "step": 44010
    },
    {
      "epoch": 2.3525869222822497,
      "grad_norm": 0.5775142908096313,
      "learning_rate": 9.96578305952174e-05,
      "loss": 1.4923,
      "step": 44020
    },
    {
      "epoch": 2.3526953105153523,
      "eval_loss": 2.4744277000427246,
      "eval_runtime": 21.978,
      "eval_samples_per_second": 227.5,
      "eval_steps_per_second": 1.228,
      "step": 44022
    },
    {
      "epoch": 2.353128863447762,
      "grad_norm": 0.6233053803443909,
      "learning_rate": 9.965587839904527e-05,
      "loss": 1.507,
      "step": 44030
    },
    {
      "epoch": 2.353670804613274,
      "grad_norm": 0.5592018365859985,
      "learning_rate": 9.965392067109679e-05,
      "loss": 1.4945,
      "step": 44040
    },
    {
      "epoch": 2.3542127457787863,
      "grad_norm": 0.3838133215904236,
      "learning_rate": 9.965195741161449e-05,
      "loss": 1.4825,
      "step": 44050
    },
    {
      "epoch": 2.3547546869442986,
      "grad_norm": 0.2884215712547302,
      "learning_rate": 9.964998862084157e-05,
      "loss": 1.4858,
      "step": 44060
    },
    {
      "epoch": 2.355296628109811,
      "grad_norm": 0.22418469190597534,
      "learning_rate": 9.96480142990219e-05,
      "loss": 1.493,
      "step": 44070
    },
    {
      "epoch": 2.355838569275323,
      "grad_norm": 0.34609487652778625,
      "learning_rate": 9.964603444640007e-05,
      "loss": 1.5015,
      "step": 44080
    },
    {
      "epoch": 2.356380510440835,
      "grad_norm": 0.2725040912628174,
      "learning_rate": 9.964404906322132e-05,
      "loss": 1.4937,
      "step": 44090
    },
    {
      "epoch": 2.3569224516063474,
      "grad_norm": 0.6061669588088989,
      "learning_rate": 9.96420581497316e-05,
      "loss": 1.5041,
      "step": 44100
    },
    {
      "epoch": 2.3574101986553084,
      "eval_loss": 2.4810047149658203,
      "eval_runtime": 21.9811,
      "eval_samples_per_second": 227.468,
      "eval_steps_per_second": 1.228,
      "step": 44109
    },
    {
      "epoch": 2.3574643927718597,
      "grad_norm": 0.509492814540863,
      "learning_rate": 9.964006170617754e-05,
      "loss": 1.5008,
      "step": 44110
    },
    {
      "epoch": 2.358006333937372,
      "grad_norm": 0.4258333146572113,
      "learning_rate": 9.963805973280643e-05,
      "loss": 1.4977,
      "step": 44120
    },
    {
      "epoch": 2.3585482751028843,
      "grad_norm": 0.27747806906700134,
      "learning_rate": 9.96360522298663e-05,
      "loss": 1.4924,
      "step": 44130
    },
    {
      "epoch": 2.359090216268396,
      "grad_norm": 0.339024156332016,
      "learning_rate": 9.963403919760579e-05,
      "loss": 1.494,
      "step": 44140
    },
    {
      "epoch": 2.3596321574339085,
      "grad_norm": 0.3902308940887451,
      "learning_rate": 9.963202063627429e-05,
      "loss": 1.5009,
      "step": 44150
    },
    {
      "epoch": 2.360174098599421,
      "grad_norm": 0.20376642048358917,
      "learning_rate": 9.962999654612185e-05,
      "loss": 1.5112,
      "step": 44160
    },
    {
      "epoch": 2.360716039764933,
      "grad_norm": 0.28291308879852295,
      "learning_rate": 9.96279669273992e-05,
      "loss": 1.5011,
      "step": 44170
    },
    {
      "epoch": 2.3612579809304455,
      "grad_norm": 0.3442187011241913,
      "learning_rate": 9.962593178035776e-05,
      "loss": 1.5062,
      "step": 44180
    },
    {
      "epoch": 2.3617999220959573,
      "grad_norm": 0.23570683598518372,
      "learning_rate": 9.962389110524964e-05,
      "loss": 1.4994,
      "step": 44190
    },
    {
      "epoch": 2.362125086795265,
      "eval_loss": 2.45691180229187,
      "eval_runtime": 21.9778,
      "eval_samples_per_second": 227.503,
      "eval_steps_per_second": 1.229,
      "step": 44196
    },
    {
      "epoch": 2.3623418632614697,
      "grad_norm": 0.38584163784980774,
      "learning_rate": 9.962184490232763e-05,
      "loss": 1.5001,
      "step": 44200
    },
    {
      "epoch": 2.362883804426982,
      "grad_norm": 0.2459559589624405,
      "learning_rate": 9.96197931718452e-05,
      "loss": 1.4962,
      "step": 44210
    },
    {
      "epoch": 2.3634257455924943,
      "grad_norm": 0.217377707362175,
      "learning_rate": 9.96177359140565e-05,
      "loss": 1.4944,
      "step": 44220
    },
    {
      "epoch": 2.363967686758006,
      "grad_norm": 0.5001499652862549,
      "learning_rate": 9.96156731292164e-05,
      "loss": 1.4937,
      "step": 44230
    },
    {
      "epoch": 2.3645096279235185,
      "grad_norm": 0.33417242765426636,
      "learning_rate": 9.961360481758043e-05,
      "loss": 1.4935,
      "step": 44240
    },
    {
      "epoch": 2.365051569089031,
      "grad_norm": 0.4141455590724945,
      "learning_rate": 9.961153097940475e-05,
      "loss": 1.5026,
      "step": 44250
    },
    {
      "epoch": 2.365593510254543,
      "grad_norm": 0.6103190779685974,
      "learning_rate": 9.960945161494633e-05,
      "loss": 1.4783,
      "step": 44260
    },
    {
      "epoch": 2.366135451420055,
      "grad_norm": 0.4958433508872986,
      "learning_rate": 9.96073667244627e-05,
      "loss": 1.4915,
      "step": 44270
    },
    {
      "epoch": 2.3666773925855673,
      "grad_norm": 0.30558812618255615,
      "learning_rate": 9.960527630821217e-05,
      "loss": 1.4999,
      "step": 44280
    },
    {
      "epoch": 2.366839974935221,
      "eval_loss": 2.4537668228149414,
      "eval_runtime": 21.9779,
      "eval_samples_per_second": 227.501,
      "eval_steps_per_second": 1.229,
      "step": 44283
    },
    {
      "epoch": 2.3672193337510796,
      "grad_norm": 0.36779430508613586,
      "learning_rate": 9.960318036645363e-05,
      "loss": 1.4988,
      "step": 44290
    },
    {
      "epoch": 2.367761274916592,
      "grad_norm": 0.2908129096031189,
      "learning_rate": 9.960107889944679e-05,
      "loss": 1.5003,
      "step": 44300
    },
    {
      "epoch": 2.3683032160821043,
      "grad_norm": 0.5681326389312744,
      "learning_rate": 9.959897190745192e-05,
      "loss": 1.4934,
      "step": 44310
    },
    {
      "epoch": 2.368845157247616,
      "grad_norm": 0.40029388666152954,
      "learning_rate": 9.959685939073002e-05,
      "loss": 1.5063,
      "step": 44320
    },
    {
      "epoch": 2.3693870984131284,
      "grad_norm": 0.42044511437416077,
      "learning_rate": 9.959474134954283e-05,
      "loss": 1.4933,
      "step": 44330
    },
    {
      "epoch": 2.3699290395786408,
      "grad_norm": 0.590247392654419,
      "learning_rate": 9.959261778415267e-05,
      "loss": 1.4867,
      "step": 44340
    },
    {
      "epoch": 2.370470980744153,
      "grad_norm": 0.2779798209667206,
      "learning_rate": 9.959048869482262e-05,
      "loss": 1.5081,
      "step": 44350
    },
    {
      "epoch": 2.3710129219096654,
      "grad_norm": 0.48375606536865234,
      "learning_rate": 9.958835408181643e-05,
      "loss": 1.4951,
      "step": 44360
    },
    {
      "epoch": 2.3715548630751773,
      "grad_norm": 0.251369446516037,
      "learning_rate": 9.958621394539854e-05,
      "loss": 1.4872,
      "step": 44370
    },
    {
      "epoch": 2.3715548630751773,
      "eval_loss": 2.4517953395843506,
      "eval_runtime": 21.9682,
      "eval_samples_per_second": 227.602,
      "eval_steps_per_second": 1.229,
      "step": 44370
    },
    {
      "epoch": 2.3720968042406896,
      "grad_norm": 0.24449922144412994,
      "learning_rate": 9.9584068285834e-05,
      "loss": 1.4872,
      "step": 44380
    },
    {
      "epoch": 2.372638745406202,
      "grad_norm": 0.33364132046699524,
      "learning_rate": 9.95819171033887e-05,
      "loss": 1.4842,
      "step": 44390
    },
    {
      "epoch": 2.373180686571714,
      "grad_norm": 0.30811458826065063,
      "learning_rate": 9.957976039832901e-05,
      "loss": 1.4935,
      "step": 44400
    },
    {
      "epoch": 2.373722627737226,
      "grad_norm": 0.2445499747991562,
      "learning_rate": 9.957759817092218e-05,
      "loss": 1.5017,
      "step": 44410
    },
    {
      "epoch": 2.3742645689027384,
      "grad_norm": 0.2053236961364746,
      "learning_rate": 9.957543042143601e-05,
      "loss": 1.5071,
      "step": 44420
    },
    {
      "epoch": 2.3748065100682507,
      "grad_norm": 0.4515707492828369,
      "learning_rate": 9.957325715013905e-05,
      "loss": 1.4962,
      "step": 44430
    },
    {
      "epoch": 2.375348451233763,
      "grad_norm": 0.26038962602615356,
      "learning_rate": 9.957107835730052e-05,
      "loss": 1.4925,
      "step": 44440
    },
    {
      "epoch": 2.375890392399275,
      "grad_norm": 0.3523120880126953,
      "learning_rate": 9.95688940431903e-05,
      "loss": 1.4928,
      "step": 44450
    },
    {
      "epoch": 2.376269751215134,
      "eval_loss": 2.4564239978790283,
      "eval_runtime": 21.9781,
      "eval_samples_per_second": 227.499,
      "eval_steps_per_second": 1.228,
      "step": 44457
    },
    {
      "epoch": 2.3764323335647872,
      "grad_norm": 0.24596655368804932,
      "learning_rate": 9.956670420807899e-05,
      "loss": 1.5008,
      "step": 44460
    },
    {
      "epoch": 2.3769742747302995,
      "grad_norm": 0.47213152050971985,
      "learning_rate": 9.956450885223785e-05,
      "loss": 1.5027,
      "step": 44470
    },
    {
      "epoch": 2.377516215895812,
      "grad_norm": 0.3502696752548218,
      "learning_rate": 9.956230797593884e-05,
      "loss": 1.4894,
      "step": 44480
    },
    {
      "epoch": 2.378058157061324,
      "grad_norm": 0.31429505348205566,
      "learning_rate": 9.95601015794546e-05,
      "loss": 1.4878,
      "step": 44490
    },
    {
      "epoch": 2.378600098226836,
      "grad_norm": 0.22514714300632477,
      "learning_rate": 9.95578896630584e-05,
      "loss": 1.4924,
      "step": 44500
    },
    {
      "epoch": 2.3791420393923484,
      "grad_norm": 0.27507612109184265,
      "learning_rate": 9.95556722270243e-05,
      "loss": 1.5104,
      "step": 44510
    },
    {
      "epoch": 2.3796839805578607,
      "grad_norm": 0.3192688226699829,
      "learning_rate": 9.955344927162698e-05,
      "loss": 1.4992,
      "step": 44520
    },
    {
      "epoch": 2.380225921723373,
      "grad_norm": 0.20219798386096954,
      "learning_rate": 9.955122079714177e-05,
      "loss": 1.51,
      "step": 44530
    },
    {
      "epoch": 2.3807678628888853,
      "grad_norm": 0.21116623282432556,
      "learning_rate": 9.954898680384476e-05,
      "loss": 1.4943,
      "step": 44540
    },
    {
      "epoch": 2.38098463935509,
      "eval_loss": 2.4590439796447754,
      "eval_runtime": 22.3001,
      "eval_samples_per_second": 224.215,
      "eval_steps_per_second": 1.211,
      "step": 44544
    },
    {
      "epoch": 2.381309804054397,
      "grad_norm": 0.3956089913845062,
      "learning_rate": 9.954674729201268e-05,
      "loss": 1.4822,
      "step": 44550
    },
    {
      "epoch": 2.3818517452199095,
      "grad_norm": 0.5214465856552124,
      "learning_rate": 9.954450226192295e-05,
      "loss": 1.4843,
      "step": 44560
    },
    {
      "epoch": 2.382393686385422,
      "grad_norm": 0.4375009536743164,
      "learning_rate": 9.954225171385366e-05,
      "loss": 1.509,
      "step": 44570
    },
    {
      "epoch": 2.382935627550934,
      "grad_norm": 0.22339291870594025,
      "learning_rate": 9.953999564808362e-05,
      "loss": 1.4914,
      "step": 44580
    },
    {
      "epoch": 2.3834775687164464,
      "grad_norm": 0.2312263697385788,
      "learning_rate": 9.953773406489229e-05,
      "loss": 1.5064,
      "step": 44590
    },
    {
      "epoch": 2.3840195098819583,
      "grad_norm": 0.21803376078605652,
      "learning_rate": 9.953546696455984e-05,
      "loss": 1.5093,
      "step": 44600
    },
    {
      "epoch": 2.3845614510474706,
      "grad_norm": 0.33443108201026917,
      "learning_rate": 9.953319434736708e-05,
      "loss": 1.4916,
      "step": 44610
    },
    {
      "epoch": 2.385103392212983,
      "grad_norm": 0.30027446150779724,
      "learning_rate": 9.953091621359556e-05,
      "loss": 1.4997,
      "step": 44620
    },
    {
      "epoch": 2.3856453333784953,
      "grad_norm": 0.32896170020103455,
      "learning_rate": 9.952863256352746e-05,
      "loss": 1.4882,
      "step": 44630
    },
    {
      "epoch": 2.3856995274950465,
      "eval_loss": 2.4726407527923584,
      "eval_runtime": 21.9784,
      "eval_samples_per_second": 227.496,
      "eval_steps_per_second": 1.228,
      "step": 44631
    },
    {
      "epoch": 2.386187274544007,
      "grad_norm": 0.41596585512161255,
      "learning_rate": 9.95263433974457e-05,
      "loss": 1.4935,
      "step": 44640
    },
    {
      "epoch": 2.3867292157095195,
      "grad_norm": 0.2066296935081482,
      "learning_rate": 9.952404871563383e-05,
      "loss": 1.4963,
      "step": 44650
    },
    {
      "epoch": 2.3872711568750318,
      "grad_norm": 0.5098561644554138,
      "learning_rate": 9.952174851837609e-05,
      "loss": 1.4899,
      "step": 44660
    },
    {
      "epoch": 2.387813098040544,
      "grad_norm": 0.3089795410633087,
      "learning_rate": 9.951944280595745e-05,
      "loss": 1.4808,
      "step": 44670
    },
    {
      "epoch": 2.388355039206056,
      "grad_norm": 0.3919314742088318,
      "learning_rate": 9.951713157866352e-05,
      "loss": 1.4988,
      "step": 44680
    },
    {
      "epoch": 2.3888969803715683,
      "grad_norm": 0.6062659025192261,
      "learning_rate": 9.95148148367806e-05,
      "loss": 1.494,
      "step": 44690
    },
    {
      "epoch": 2.3894389215370806,
      "grad_norm": 0.40539994835853577,
      "learning_rate": 9.951249258059569e-05,
      "loss": 1.4863,
      "step": 44700
    },
    {
      "epoch": 2.389980862702593,
      "grad_norm": 0.3886335492134094,
      "learning_rate": 9.951016481039646e-05,
      "loss": 1.4878,
      "step": 44710
    },
    {
      "epoch": 2.3904144156350027,
      "eval_loss": 2.4730005264282227,
      "eval_runtime": 21.9758,
      "eval_samples_per_second": 227.523,
      "eval_steps_per_second": 1.229,
      "step": 44718
    },
    {
      "epoch": 2.3905228038681052,
      "grad_norm": 0.39864251017570496,
      "learning_rate": 9.950783152647124e-05,
      "loss": 1.4932,
      "step": 44720
    },
    {
      "epoch": 2.391064745033617,
      "grad_norm": 0.21343019604682922,
      "learning_rate": 9.950549272910909e-05,
      "loss": 1.4889,
      "step": 44730
    },
    {
      "epoch": 2.3916066861991294,
      "grad_norm": 0.33363330364227295,
      "learning_rate": 9.950314841859973e-05,
      "loss": 1.484,
      "step": 44740
    },
    {
      "epoch": 2.3921486273646417,
      "grad_norm": 0.41146817803382874,
      "learning_rate": 9.950079859523354e-05,
      "loss": 1.497,
      "step": 44750
    },
    {
      "epoch": 2.392690568530154,
      "grad_norm": 0.27292779088020325,
      "learning_rate": 9.949844325930165e-05,
      "loss": 1.4942,
      "step": 44760
    },
    {
      "epoch": 2.3932325096956664,
      "grad_norm": 0.2181469202041626,
      "learning_rate": 9.94960824110958e-05,
      "loss": 1.4911,
      "step": 44770
    },
    {
      "epoch": 2.3937744508611782,
      "grad_norm": 0.21608904004096985,
      "learning_rate": 9.94937160509084e-05,
      "loss": 1.4981,
      "step": 44780
    },
    {
      "epoch": 2.3943163920266906,
      "grad_norm": 0.2778734266757965,
      "learning_rate": 9.949134417903267e-05,
      "loss": 1.4898,
      "step": 44790
    },
    {
      "epoch": 2.394858333192203,
      "grad_norm": 0.3089962303638458,
      "learning_rate": 9.948896679576238e-05,
      "loss": 1.4905,
      "step": 44800
    },
    {
      "epoch": 2.395129303774959,
      "eval_loss": 2.4715840816497803,
      "eval_runtime": 21.9797,
      "eval_samples_per_second": 227.482,
      "eval_steps_per_second": 1.228,
      "step": 44805
    },
    {
      "epoch": 2.395400274357715,
      "grad_norm": 0.3231702148914337,
      "learning_rate": 9.948658390139203e-05,
      "loss": 1.5066,
      "step": 44810
    },
    {
      "epoch": 2.3959422155232275,
      "grad_norm": 0.40103909373283386,
      "learning_rate": 9.94841954962168e-05,
      "loss": 1.4987,
      "step": 44820
    },
    {
      "epoch": 2.3964841566887394,
      "grad_norm": 0.40768495202064514,
      "learning_rate": 9.948180158053257e-05,
      "loss": 1.4982,
      "step": 44830
    },
    {
      "epoch": 2.3970260978542517,
      "grad_norm": 0.23787713050842285,
      "learning_rate": 9.947940215463589e-05,
      "loss": 1.4838,
      "step": 44840
    },
    {
      "epoch": 2.397568039019764,
      "grad_norm": 0.3806591331958771,
      "learning_rate": 9.947699721882396e-05,
      "loss": 1.4936,
      "step": 44850
    },
    {
      "epoch": 2.3981099801852763,
      "grad_norm": 0.26466140151023865,
      "learning_rate": 9.947458677339473e-05,
      "loss": 1.4965,
      "step": 44860
    },
    {
      "epoch": 2.398651921350788,
      "grad_norm": 0.2688174247741699,
      "learning_rate": 9.947217081864678e-05,
      "loss": 1.494,
      "step": 44870
    },
    {
      "epoch": 2.3991938625163005,
      "grad_norm": 0.5081690549850464,
      "learning_rate": 9.94697493548794e-05,
      "loss": 1.4951,
      "step": 44880
    },
    {
      "epoch": 2.399735803681813,
      "grad_norm": 0.23584547638893127,
      "learning_rate": 9.946732238239251e-05,
      "loss": 1.4915,
      "step": 44890
    },
    {
      "epoch": 2.3998441919149154,
      "eval_loss": 2.4753541946411133,
      "eval_runtime": 21.9826,
      "eval_samples_per_second": 227.453,
      "eval_steps_per_second": 1.228,
      "step": 44892
    },
    {
      "epoch": 2.400277744847325,
      "grad_norm": 0.41271650791168213,
      "learning_rate": 9.946488990148679e-05,
      "loss": 1.4985,
      "step": 44900
    },
    {
      "epoch": 2.400819686012837,
      "grad_norm": 0.24468545615673065,
      "learning_rate": 9.946245191246358e-05,
      "loss": 1.4974,
      "step": 44910
    },
    {
      "epoch": 2.4013616271783493,
      "grad_norm": 0.3811487853527069,
      "learning_rate": 9.946000841562482e-05,
      "loss": 1.4896,
      "step": 44920
    },
    {
      "epoch": 2.4019035683438617,
      "grad_norm": 0.4173762798309326,
      "learning_rate": 9.945755941127327e-05,
      "loss": 1.4897,
      "step": 44930
    },
    {
      "epoch": 2.402445509509374,
      "grad_norm": 0.5052602291107178,
      "learning_rate": 9.945510489971228e-05,
      "loss": 1.4964,
      "step": 44940
    },
    {
      "epoch": 2.4029874506748863,
      "grad_norm": 0.3579169809818268,
      "learning_rate": 9.945264488124589e-05,
      "loss": 1.4941,
      "step": 44950
    },
    {
      "epoch": 2.403529391840398,
      "grad_norm": 0.2974871098995209,
      "learning_rate": 9.945017935617885e-05,
      "loss": 1.5033,
      "step": 44960
    },
    {
      "epoch": 2.4040713330059105,
      "grad_norm": 0.34116071462631226,
      "learning_rate": 9.944770832481656e-05,
      "loss": 1.4966,
      "step": 44970
    },
    {
      "epoch": 2.4045590800548715,
      "eval_loss": 2.4695467948913574,
      "eval_runtime": 21.9753,
      "eval_samples_per_second": 227.528,
      "eval_steps_per_second": 1.229,
      "step": 44979
    },
    {
      "epoch": 2.404613274171423,
      "grad_norm": 0.2582961916923523,
      "learning_rate": 9.944523178746516e-05,
      "loss": 1.499,
      "step": 44980
    },
    {
      "epoch": 2.405155215336935,
      "grad_norm": 0.32894718647003174,
      "learning_rate": 9.94427497444314e-05,
      "loss": 1.4877,
      "step": 44990
    },
    {
      "epoch": 2.4056971565024474,
      "grad_norm": 0.38005682826042175,
      "learning_rate": 9.944026219602274e-05,
      "loss": 1.4999,
      "step": 45000
    },
    {
      "epoch": 2.0005419411655123,
      "grad_norm": 0.2491222321987152,
      "learning_rate": 9.943776914254736e-05,
      "loss": 1.5076,
      "step": 45010
    },
    {
      "epoch": 2.0010838823310246,
      "grad_norm": 0.3733494281768799,
      "learning_rate": 9.943527058431406e-05,
      "loss": 1.517,
      "step": 45020
    },
    {
      "epoch": 2.0016258234965365,
      "grad_norm": 0.33631569147109985,
      "learning_rate": 9.943276652163235e-05,
      "loss": 1.5062,
      "step": 45030
    },
    {
      "epoch": 2.002167764662049,
      "grad_norm": 0.21265359222888947,
      "learning_rate": 9.943025695481244e-05,
      "loss": 1.5177,
      "step": 45040
    },
    {
      "epoch": 2.002709705827561,
      "grad_norm": 0.18703655898571014,
      "learning_rate": 9.94277418841652e-05,
      "loss": 1.5098,
      "step": 45050
    },
    {
      "epoch": 2.0032516469930735,
      "grad_norm": 0.44634345173835754,
      "learning_rate": 9.942522131000216e-05,
      "loss": 1.5068,
      "step": 45060
    },
    {
      "epoch": 2.0035768116923807,
      "eval_loss": 2.45697283744812,
      "eval_runtime": 28.4848,
      "eval_samples_per_second": 175.532,
      "eval_steps_per_second": 0.948,
      "step": 45066
    },
    {
      "epoch": 2.0037935881585853,
      "grad_norm": 0.22693060338497162,
      "learning_rate": 9.942269523263559e-05,
      "loss": 1.5085,
      "step": 45070
    },
    {
      "epoch": 2.0043355293240976,
      "grad_norm": 0.22670039534568787,
      "learning_rate": 9.942016365237841e-05,
      "loss": 1.5155,
      "step": 45080
    },
    {
      "epoch": 2.00487747048961,
      "grad_norm": 0.2845227122306824,
      "learning_rate": 9.94176265695442e-05,
      "loss": 1.5119,
      "step": 45090
    },
    {
      "epoch": 2.0054194116551223,
      "grad_norm": 0.2120792120695114,
      "learning_rate": 9.941508398444725e-05,
      "loss": 1.5108,
      "step": 45100
    },
    {
      "epoch": 2.0059613528206346,
      "grad_norm": 0.2501295804977417,
      "learning_rate": 9.941253589740255e-05,
      "loss": 1.5151,
      "step": 45110
    },
    {
      "epoch": 2.0065032939861465,
      "grad_norm": 0.21968966722488403,
      "learning_rate": 9.940998230872569e-05,
      "loss": 1.5051,
      "step": 45120
    },
    {
      "epoch": 2.007045235151659,
      "grad_norm": 0.2440696805715561,
      "learning_rate": 9.940742321873304e-05,
      "loss": 1.5188,
      "step": 45130
    },
    {
      "epoch": 2.007587176317171,
      "grad_norm": 0.26428186893463135,
      "learning_rate": 9.940485862774162e-05,
      "loss": 1.5097,
      "step": 45140
    },
    {
      "epoch": 2.0081291174826834,
      "grad_norm": 0.2735063135623932,
      "learning_rate": 9.940228853606908e-05,
      "loss": 1.5153,
      "step": 45150
    },
    {
      "epoch": 2.008291699832337,
      "eval_loss": 2.474893093109131,
      "eval_runtime": 22.0407,
      "eval_samples_per_second": 226.853,
      "eval_steps_per_second": 1.225,
      "step": 45153
    },
    {
      "epoch": 2.0086710586481953,
      "grad_norm": 0.21670082211494446,
      "learning_rate": 9.939971294403382e-05,
      "loss": 1.5133,
      "step": 45160
    },
    {
      "epoch": 2.0092129998137076,
      "grad_norm": 0.219325989484787,
      "learning_rate": 9.939713185195486e-05,
      "loss": 1.5115,
      "step": 45170
    },
    {
      "epoch": 2.00975494097922,
      "grad_norm": 0.21777480840682983,
      "learning_rate": 9.939454526015199e-05,
      "loss": 1.5066,
      "step": 45180
    },
    {
      "epoch": 2.0102968821447322,
      "grad_norm": 0.24728938937187195,
      "learning_rate": 9.939195316894558e-05,
      "loss": 1.5173,
      "step": 45190
    },
    {
      "epoch": 2.0108388233102445,
      "grad_norm": 0.3391672670841217,
      "learning_rate": 9.938935557865676e-05,
      "loss": 1.4985,
      "step": 45200
    },
    {
      "epoch": 2.0113807644757564,
      "grad_norm": 0.22963690757751465,
      "learning_rate": 9.938675248960726e-05,
      "loss": 1.5031,
      "step": 45210
    },
    {
      "epoch": 2.0119227056412687,
      "grad_norm": 0.2681106626987457,
      "learning_rate": 9.93841439021196e-05,
      "loss": 1.5047,
      "step": 45220
    },
    {
      "epoch": 2.012464646806781,
      "grad_norm": 0.37070515751838684,
      "learning_rate": 9.938152981651687e-05,
      "loss": 1.5039,
      "step": 45230
    },
    {
      "epoch": 2.0130065879722934,
      "grad_norm": 0.342690110206604,
      "learning_rate": 9.937891023312292e-05,
      "loss": 1.5114,
      "step": 45240
    },
    {
      "epoch": 2.0130065879722934,
      "eval_loss": 2.4622411727905273,
      "eval_runtime": 21.7527,
      "eval_samples_per_second": 229.857,
      "eval_steps_per_second": 1.241,
      "step": 45240
    },
    {
      "epoch": 2.0135485291378057,
      "grad_norm": 0.4903299808502197,
      "learning_rate": 9.937628515226225e-05,
      "loss": 1.5174,
      "step": 45250
    },
    {
      "epoch": 2.0140904703033176,
      "grad_norm": 0.2701393961906433,
      "learning_rate": 9.937365457426003e-05,
      "loss": 1.5235,
      "step": 45260
    },
    {
      "epoch": 2.01463241146883,
      "grad_norm": 0.21151518821716309,
      "learning_rate": 9.937101849944213e-05,
      "loss": 1.4979,
      "step": 45270
    },
    {
      "epoch": 2.015174352634342,
      "grad_norm": 0.1973683089017868,
      "learning_rate": 9.936837692813511e-05,
      "loss": 1.5211,
      "step": 45280
    },
    {
      "epoch": 2.0157162937998545,
      "grad_norm": 0.2013273984193802,
      "learning_rate": 9.936572986066616e-05,
      "loss": 1.5038,
      "step": 45290
    },
    {
      "epoch": 2.0162582349653664,
      "grad_norm": 0.28360608220100403,
      "learning_rate": 9.936307729736324e-05,
      "loss": 1.5059,
      "step": 45300
    },
    {
      "epoch": 2.0168001761308787,
      "grad_norm": 0.2819204032421112,
      "learning_rate": 9.93604192385549e-05,
      "loss": 1.5104,
      "step": 45310
    },
    {
      "epoch": 2.017342117296391,
      "grad_norm": 0.20827263593673706,
      "learning_rate": 9.935775568457041e-05,
      "loss": 1.4944,
      "step": 45320
    },
    {
      "epoch": 2.0177214761122495,
      "eval_loss": 2.4640448093414307,
      "eval_runtime": 22.083,
      "eval_samples_per_second": 226.419,
      "eval_steps_per_second": 1.223,
      "step": 45327
    },
    {
      "epoch": 2.0178840584619033,
      "grad_norm": 0.21158994734287262,
      "learning_rate": 9.935508663573973e-05,
      "loss": 1.5125,
      "step": 45330
    },
    {
      "epoch": 2.0184259996274156,
      "grad_norm": 0.3482116162776947,
      "learning_rate": 9.935241209239349e-05,
      "loss": 1.5073,
      "step": 45340
    },
    {
      "epoch": 2.0189679407929275,
      "grad_norm": 0.2513381540775299,
      "learning_rate": 9.9349732054863e-05,
      "loss": 1.4914,
      "step": 45350
    },
    {
      "epoch": 2.01950988195844,
      "grad_norm": 0.21970421075820923,
      "learning_rate": 9.934704652348024e-05,
      "loss": 1.511,
      "step": 45360
    },
    {
      "epoch": 2.020051823123952,
      "grad_norm": 0.20519542694091797,
      "learning_rate": 9.934435549857793e-05,
      "loss": 1.4999,
      "step": 45370
    },
    {
      "epoch": 2.0205937642894645,
      "grad_norm": 0.2705243229866028,
      "learning_rate": 9.934165898048934e-05,
      "loss": 1.5099,
      "step": 45380
    },
    {
      "epoch": 2.0211357054549763,
      "grad_norm": 0.3955739140510559,
      "learning_rate": 9.933895696954857e-05,
      "loss": 1.5111,
      "step": 45390
    },
    {
      "epoch": 2.0216776466204887,
      "grad_norm": 0.39523282647132874,
      "learning_rate": 9.933624946609031e-05,
      "loss": 1.5154,
      "step": 45400
    },
    {
      "epoch": 2.022219587786001,
      "grad_norm": 0.22369325160980225,
      "learning_rate": 9.933353647044995e-05,
      "loss": 1.5127,
      "step": 45410
    },
    {
      "epoch": 2.022436364252206,
      "eval_loss": 2.4598445892333984,
      "eval_runtime": 22.0511,
      "eval_samples_per_second": 226.746,
      "eval_steps_per_second": 1.224,
      "step": 45414
    },
    {
      "epoch": 2.0227615289515133,
      "grad_norm": 0.29554933309555054,
      "learning_rate": 9.933081798296358e-05,
      "loss": 1.507,
      "step": 45420
    },
    {
      "epoch": 2.0233034701170256,
      "grad_norm": 0.2695947289466858,
      "learning_rate": 9.932809400396793e-05,
      "loss": 1.505,
      "step": 45430
    },
    {
      "epoch": 2.0238454112825375,
      "grad_norm": 0.25676560401916504,
      "learning_rate": 9.932536453380045e-05,
      "loss": 1.5104,
      "step": 45440
    },
    {
      "epoch": 2.02438735244805,
      "grad_norm": 0.21625731885433197,
      "learning_rate": 9.932262957279926e-05,
      "loss": 1.5112,
      "step": 45450
    },
    {
      "epoch": 2.024929293613562,
      "grad_norm": 0.22126367688179016,
      "learning_rate": 9.931988912130313e-05,
      "loss": 1.4907,
      "step": 45460
    },
    {
      "epoch": 2.0254712347790744,
      "grad_norm": 0.2600845396518707,
      "learning_rate": 9.931714317965158e-05,
      "loss": 1.5057,
      "step": 45470
    },
    {
      "epoch": 2.0260131759445863,
      "grad_norm": 0.2164200097322464,
      "learning_rate": 9.931439174818472e-05,
      "loss": 1.5125,
      "step": 45480
    },
    {
      "epoch": 2.0265551171100986,
      "grad_norm": 0.3730428218841553,
      "learning_rate": 9.931163482724341e-05,
      "loss": 1.5166,
      "step": 45490
    },
    {
      "epoch": 2.027097058275611,
      "grad_norm": 0.2146274298429489,
      "learning_rate": 9.930887241716915e-05,
      "loss": 1.5042,
      "step": 45500
    },
    {
      "epoch": 2.027151252392162,
      "eval_loss": 2.4563684463500977,
      "eval_runtime": 22.0541,
      "eval_samples_per_second": 226.715,
      "eval_steps_per_second": 1.224,
      "step": 45501
    },
    {
      "epoch": 2.0276389994411232,
      "grad_norm": 0.2864300310611725,
      "learning_rate": 9.930610451830417e-05,
      "loss": 1.5001,
      "step": 45510
    },
    {
      "epoch": 2.0281809406066356,
      "grad_norm": 0.37505003809928894,
      "learning_rate": 9.93033311309913e-05,
      "loss": 1.5074,
      "step": 45520
    },
    {
      "epoch": 2.0287228817721474,
      "grad_norm": 0.3186861276626587,
      "learning_rate": 9.93005522555741e-05,
      "loss": 1.5096,
      "step": 45530
    },
    {
      "epoch": 2.0292648229376598,
      "grad_norm": 0.31974560022354126,
      "learning_rate": 9.929776789239685e-05,
      "loss": 1.5028,
      "step": 45540
    },
    {
      "epoch": 2.029806764103172,
      "grad_norm": 0.22991715371608734,
      "learning_rate": 9.929497804180442e-05,
      "loss": 1.4912,
      "step": 45550
    },
    {
      "epoch": 2.0303487052686844,
      "grad_norm": 0.38136276602745056,
      "learning_rate": 9.929218270414243e-05,
      "loss": 1.5082,
      "step": 45560
    },
    {
      "epoch": 2.0308906464341967,
      "grad_norm": 0.2910080850124359,
      "learning_rate": 9.928938187975714e-05,
      "loss": 1.5081,
      "step": 45570
    },
    {
      "epoch": 2.0314325875997086,
      "grad_norm": 0.20356157422065735,
      "learning_rate": 9.928657556899551e-05,
      "loss": 1.4984,
      "step": 45580
    },
    {
      "epoch": 2.0318661405321183,
      "eval_loss": 2.454059600830078,
      "eval_runtime": 22.0277,
      "eval_samples_per_second": 226.987,
      "eval_steps_per_second": 1.226,
      "step": 45588
    },
    {
      "epoch": 2.031974528765221,
      "grad_norm": 0.32182246446609497,
      "learning_rate": 9.928376377220517e-05,
      "loss": 1.507,
      "step": 45590
    },
    {
      "epoch": 2.032516469930733,
      "grad_norm": 0.19996988773345947,
      "learning_rate": 9.928094648973443e-05,
      "loss": 1.5189,
      "step": 45600
    },
    {
      "epoch": 2.0330584110962455,
      "grad_norm": 0.2483881115913391,
      "learning_rate": 9.92781237219323e-05,
      "loss": 1.5088,
      "step": 45610
    },
    {
      "epoch": 2.0336003522617574,
      "grad_norm": 0.3291274309158325,
      "learning_rate": 9.927529546914842e-05,
      "loss": 1.5016,
      "step": 45620
    },
    {
      "epoch": 2.0341422934272697,
      "grad_norm": 0.21742436289787292,
      "learning_rate": 9.927246173173318e-05,
      "loss": 1.507,
      "step": 45630
    },
    {
      "epoch": 2.034684234592782,
      "grad_norm": 0.34620094299316406,
      "learning_rate": 9.926962251003758e-05,
      "loss": 1.5015,
      "step": 45640
    },
    {
      "epoch": 2.0352261757582943,
      "grad_norm": 0.2289854735136032,
      "learning_rate": 9.926677780441335e-05,
      "loss": 1.5091,
      "step": 45650
    },
    {
      "epoch": 2.0357681169238067,
      "grad_norm": 0.32256433367729187,
      "learning_rate": 9.926392761521286e-05,
      "loss": 1.5123,
      "step": 45660
    },
    {
      "epoch": 2.0363100580893185,
      "grad_norm": 0.22222593426704407,
      "learning_rate": 9.926107194278921e-05,
      "loss": 1.5074,
      "step": 45670
    },
    {
      "epoch": 2.036581028672075,
      "eval_loss": 2.465963125228882,
      "eval_runtime": 21.9821,
      "eval_samples_per_second": 227.458,
      "eval_steps_per_second": 1.228,
      "step": 45675
    },
    {
      "epoch": 2.036851999254831,
      "grad_norm": 0.23388569056987762,
      "learning_rate": 9.925821078749612e-05,
      "loss": 1.5062,
      "step": 45680
    },
    {
      "epoch": 2.037393940420343,
      "grad_norm": 0.22175799310207367,
      "learning_rate": 9.925534414968802e-05,
      "loss": 1.4969,
      "step": 45690
    },
    {
      "epoch": 2.0379358815858555,
      "grad_norm": 0.5243973135948181,
      "learning_rate": 9.925247202972004e-05,
      "loss": 1.5039,
      "step": 45700
    },
    {
      "epoch": 2.0384778227513674,
      "grad_norm": 0.3757253885269165,
      "learning_rate": 9.924959442794794e-05,
      "loss": 1.5076,
      "step": 45710
    },
    {
      "epoch": 2.0390197639168797,
      "grad_norm": 0.4267645478248596,
      "learning_rate": 9.92467113447282e-05,
      "loss": 1.5162,
      "step": 45720
    },
    {
      "epoch": 2.039561705082392,
      "grad_norm": 0.19706867635250092,
      "learning_rate": 9.924382278041796e-05,
      "loss": 1.5014,
      "step": 45730
    },
    {
      "epoch": 2.0401036462479043,
      "grad_norm": 0.3324958086013794,
      "learning_rate": 9.924092873537506e-05,
      "loss": 1.5019,
      "step": 45740
    },
    {
      "epoch": 2.0406455874134166,
      "grad_norm": 0.2539781630039215,
      "learning_rate": 9.923802920995794e-05,
      "loss": 1.5161,
      "step": 45750
    },
    {
      "epoch": 2.0411875285789285,
      "grad_norm": 0.4110063314437866,
      "learning_rate": 9.923512420452585e-05,
      "loss": 1.4998,
      "step": 45760
    },
    {
      "epoch": 2.041295916812031,
      "eval_loss": 2.4641757011413574,
      "eval_runtime": 21.9849,
      "eval_samples_per_second": 227.429,
      "eval_steps_per_second": 1.228,
      "step": 45762
    },
    {
      "epoch": 2.041729469744441,
      "grad_norm": 0.22075022757053375,
      "learning_rate": 9.923221371943862e-05,
      "loss": 1.4926,
      "step": 45770
    },
    {
      "epoch": 2.042271410909953,
      "grad_norm": 0.3489071726799011,
      "learning_rate": 9.92292977550568e-05,
      "loss": 1.5091,
      "step": 45780
    },
    {
      "epoch": 2.0428133520754654,
      "grad_norm": 0.364443302154541,
      "learning_rate": 9.922637631174158e-05,
      "loss": 1.5006,
      "step": 45790
    },
    {
      "epoch": 2.0433552932409773,
      "grad_norm": 0.26170814037323,
      "learning_rate": 9.922344938985489e-05,
      "loss": 1.5044,
      "step": 45800
    },
    {
      "epoch": 2.0438972344064896,
      "grad_norm": 0.2701292634010315,
      "learning_rate": 9.922051698975927e-05,
      "loss": 1.5028,
      "step": 45810
    },
    {
      "epoch": 2.044439175572002,
      "grad_norm": 0.5133768320083618,
      "learning_rate": 9.921757911181801e-05,
      "loss": 1.514,
      "step": 45820
    },
    {
      "epoch": 2.0449811167375143,
      "grad_norm": 0.20944929122924805,
      "learning_rate": 9.921463575639503e-05,
      "loss": 1.5047,
      "step": 45830
    },
    {
      "epoch": 2.0455230579030266,
      "grad_norm": 0.2594199776649475,
      "learning_rate": 9.921168692385491e-05,
      "loss": 1.5131,
      "step": 45840
    },
    {
      "epoch": 2.0460108049519876,
      "eval_loss": 2.4567441940307617,
      "eval_runtime": 21.9911,
      "eval_samples_per_second": 227.365,
      "eval_steps_per_second": 1.228,
      "step": 45849
    },
    {
      "epoch": 2.0460649990685384,
      "grad_norm": 0.4213975667953491,
      "learning_rate": 9.920873261456297e-05,
      "loss": 1.4995,
      "step": 45850
    },
    {
      "epoch": 2.0466069402340508,
      "grad_norm": 0.33704066276550293,
      "learning_rate": 9.920577282888515e-05,
      "loss": 1.5053,
      "step": 45860
    },
    {
      "epoch": 2.047148881399563,
      "grad_norm": 0.3914563059806824,
      "learning_rate": 9.920280756718815e-05,
      "loss": 1.508,
      "step": 45870
    },
    {
      "epoch": 2.0476908225650754,
      "grad_norm": 0.22689972817897797,
      "learning_rate": 9.919983682983924e-05,
      "loss": 1.5135,
      "step": 45880
    },
    {
      "epoch": 2.0482327637305877,
      "grad_norm": 0.24155005812644958,
      "learning_rate": 9.919686061720645e-05,
      "loss": 1.5056,
      "step": 45890
    },
    {
      "epoch": 2.0487747048960996,
      "grad_norm": 0.22122272849082947,
      "learning_rate": 9.919387892965845e-05,
      "loss": 1.5075,
      "step": 45900
    },
    {
      "epoch": 2.049316646061612,
      "grad_norm": 0.18150360882282257,
      "learning_rate": 9.919089176756458e-05,
      "loss": 1.5084,
      "step": 45910
    },
    {
      "epoch": 2.049858587227124,
      "grad_norm": 0.31947770714759827,
      "learning_rate": 9.918789913129491e-05,
      "loss": 1.5126,
      "step": 45920
    },
    {
      "epoch": 2.0504005283926365,
      "grad_norm": 0.21400408446788788,
      "learning_rate": 9.918490102122014e-05,
      "loss": 1.5074,
      "step": 45930
    },
    {
      "epoch": 2.0507256930919437,
      "eval_loss": 2.472001552581787,
      "eval_runtime": 21.9893,
      "eval_samples_per_second": 227.383,
      "eval_steps_per_second": 1.228,
      "step": 45936
    },
    {
      "epoch": 2.0509424695581484,
      "grad_norm": 0.4493444263935089,
      "learning_rate": 9.918189743771165e-05,
      "loss": 1.5229,
      "step": 45940
    },
    {
      "epoch": 2.0514844107236607,
      "grad_norm": 0.28981345891952515,
      "learning_rate": 9.917888838114155e-05,
      "loss": 1.5128,
      "step": 45950
    },
    {
      "epoch": 2.052026351889173,
      "grad_norm": 0.3488079011440277,
      "learning_rate": 9.917587385188255e-05,
      "loss": 1.4949,
      "step": 45960
    },
    {
      "epoch": 2.0525682930546854,
      "grad_norm": 0.25213801860809326,
      "learning_rate": 9.917285385030808e-05,
      "loss": 1.5046,
      "step": 45970
    },
    {
      "epoch": 2.0531102342201977,
      "grad_norm": 0.1905914545059204,
      "learning_rate": 9.916982837679226e-05,
      "loss": 1.5046,
      "step": 45980
    },
    {
      "epoch": 2.0536521753857095,
      "grad_norm": 0.2938050329685211,
      "learning_rate": 9.916679743170986e-05,
      "loss": 1.5087,
      "step": 45990
    },
    {
      "epoch": 2.054194116551222,
      "grad_norm": 0.3296414613723755,
      "learning_rate": 9.916376101543636e-05,
      "loss": 1.5114,
      "step": 46000
    },
    {
      "epoch": 2.054736057716734,
      "grad_norm": 0.32699063420295715,
      "learning_rate": 9.916071912834789e-05,
      "loss": 1.5049,
      "step": 46010
    },
    {
      "epoch": 2.0552779988822465,
      "grad_norm": 0.28838619589805603,
      "learning_rate": 9.915767177082125e-05,
      "loss": 1.5045,
      "step": 46020
    },
    {
      "epoch": 2.0554405812319,
      "eval_loss": 2.4637975692749023,
      "eval_runtime": 21.9861,
      "eval_samples_per_second": 227.417,
      "eval_steps_per_second": 1.228,
      "step": 46023
    },
    {
      "epoch": 2.0558199400477584,
      "grad_norm": 0.23613585531711578,
      "learning_rate": 9.915461894323395e-05,
      "loss": 1.5072,
      "step": 46030
    },
    {
      "epoch": 2.0563618812132707,
      "grad_norm": 0.33887359499931335,
      "learning_rate": 9.915156064596414e-05,
      "loss": 1.497,
      "step": 46040
    },
    {
      "epoch": 2.056903822378783,
      "grad_norm": 0.24096645414829254,
      "learning_rate": 9.914849687939071e-05,
      "loss": 1.4923,
      "step": 46050
    },
    {
      "epoch": 2.0574457635442953,
      "grad_norm": 0.29144373536109924,
      "learning_rate": 9.914542764389314e-05,
      "loss": 1.4964,
      "step": 46060
    },
    {
      "epoch": 2.0579877047098076,
      "grad_norm": 0.34292078018188477,
      "learning_rate": 9.914235293985167e-05,
      "loss": 1.514,
      "step": 46070
    },
    {
      "epoch": 2.0585296458753195,
      "grad_norm": 0.286112904548645,
      "learning_rate": 9.913927276764715e-05,
      "loss": 1.5068,
      "step": 46080
    },
    {
      "epoch": 2.059071587040832,
      "grad_norm": 0.2864058017730713,
      "learning_rate": 9.913618712766117e-05,
      "loss": 1.5067,
      "step": 46090
    },
    {
      "epoch": 2.059613528206344,
      "grad_norm": 0.2555348575115204,
      "learning_rate": 9.913309602027593e-05,
      "loss": 1.5068,
      "step": 46100
    },
    {
      "epoch": 2.0601554693718565,
      "grad_norm": 0.21725402772426605,
      "learning_rate": 9.912999944587437e-05,
      "loss": 1.5055,
      "step": 46110
    },
    {
      "epoch": 2.0601554693718565,
      "eval_loss": 2.46505069732666,
      "eval_runtime": 22.3497,
      "eval_samples_per_second": 223.717,
      "eval_steps_per_second": 1.208,
      "step": 46110
    },
    {
      "epoch": 2.0606974105373683,
      "grad_norm": 0.3962418735027313,
      "learning_rate": 9.912689740484007e-05,
      "loss": 1.5052,
      "step": 46120
    },
    {
      "epoch": 2.0612393517028806,
      "grad_norm": 0.22441263496875763,
      "learning_rate": 9.91237898975573e-05,
      "loss": 1.5036,
      "step": 46130
    },
    {
      "epoch": 2.061781292868393,
      "grad_norm": 0.3478442430496216,
      "learning_rate": 9.9120676924411e-05,
      "loss": 1.5032,
      "step": 46140
    },
    {
      "epoch": 2.0623232340339053,
      "grad_norm": 0.41713225841522217,
      "learning_rate": 9.911755848578681e-05,
      "loss": 1.5111,
      "step": 46150
    },
    {
      "epoch": 2.0628651751994176,
      "grad_norm": 0.20105630159378052,
      "learning_rate": 9.9114434582071e-05,
      "loss": 1.5153,
      "step": 46160
    },
    {
      "epoch": 2.0634071163649295,
      "grad_norm": 0.48680949211120605,
      "learning_rate": 9.911130521365057e-05,
      "loss": 1.5127,
      "step": 46170
    },
    {
      "epoch": 2.0639490575304418,
      "grad_norm": 0.2089950144290924,
      "learning_rate": 9.910817038091315e-05,
      "loss": 1.4976,
      "step": 46180
    },
    {
      "epoch": 2.064490998695954,
      "grad_norm": 0.35045289993286133,
      "learning_rate": 9.910503008424709e-05,
      "loss": 1.4987,
      "step": 46190
    },
    {
      "epoch": 2.0648703575118126,
      "eval_loss": 2.4648759365081787,
      "eval_runtime": 22.0369,
      "eval_samples_per_second": 226.892,
      "eval_steps_per_second": 1.225,
      "step": 46197
    },
    {
      "epoch": 2.0650329398614664,
      "grad_norm": 0.30531641840934753,
      "learning_rate": 9.910188432404139e-05,
      "loss": 1.5034,
      "step": 46200
    },
    {
      "epoch": 2.0655748810269783,
      "grad_norm": 0.2192566692829132,
      "learning_rate": 9.909873310068571e-05,
      "loss": 1.5177,
      "step": 46210
    },
    {
      "epoch": 2.0661168221924906,
      "grad_norm": 0.39566564559936523,
      "learning_rate": 9.909557641457047e-05,
      "loss": 1.5048,
      "step": 46220
    },
    {
      "epoch": 2.066658763358003,
      "grad_norm": 0.4346673786640167,
      "learning_rate": 9.909241426608664e-05,
      "loss": 1.4999,
      "step": 46230
    },
    {
      "epoch": 2.0672007045235152,
      "grad_norm": 0.2114131599664688,
      "learning_rate": 9.908924665562595e-05,
      "loss": 1.5104,
      "step": 46240
    },
    {
      "epoch": 2.0677426456890275,
      "grad_norm": 0.21276456117630005,
      "learning_rate": 9.908607358358082e-05,
      "loss": 1.5089,
      "step": 46250
    },
    {
      "epoch": 2.0682845868545394,
      "grad_norm": 0.20837821066379547,
      "learning_rate": 9.90828950503443e-05,
      "loss": 1.508,
      "step": 46260
    },
    {
      "epoch": 2.0688265280200517,
      "grad_norm": 0.22510722279548645,
      "learning_rate": 9.907971105631014e-05,
      "loss": 1.4973,
      "step": 46270
    },
    {
      "epoch": 2.069368469185564,
      "grad_norm": 0.2420571893453598,
      "learning_rate": 9.907652160187272e-05,
      "loss": 1.5006,
      "step": 46280
    },
    {
      "epoch": 2.069585245651769,
      "eval_loss": 2.455404043197632,
      "eval_runtime": 22.0244,
      "eval_samples_per_second": 227.021,
      "eval_steps_per_second": 1.226,
      "step": 46284
    },
    {
      "epoch": 2.0699104103510764,
      "grad_norm": 0.21104340255260468,
      "learning_rate": 9.907332668742718e-05,
      "loss": 1.5047,
      "step": 46290
    },
    {
      "epoch": 2.0704523515165887,
      "grad_norm": 0.3561536967754364,
      "learning_rate": 9.907012631336927e-05,
      "loss": 1.504,
      "step": 46300
    },
    {
      "epoch": 2.0709942926821006,
      "grad_norm": 0.19159241020679474,
      "learning_rate": 9.906692048009546e-05,
      "loss": 1.5021,
      "step": 46310
    },
    {
      "epoch": 2.071536233847613,
      "grad_norm": 0.4784151315689087,
      "learning_rate": 9.906370918800283e-05,
      "loss": 1.4997,
      "step": 46320
    },
    {
      "epoch": 2.072078175013125,
      "grad_norm": 0.2709428369998932,
      "learning_rate": 9.906049243748925e-05,
      "loss": 1.5038,
      "step": 46330
    },
    {
      "epoch": 2.0726201161786375,
      "grad_norm": 0.3371290862560272,
      "learning_rate": 9.905727022895313e-05,
      "loss": 1.5185,
      "step": 46340
    },
    {
      "epoch": 2.0731620573441494,
      "grad_norm": 0.3282695710659027,
      "learning_rate": 9.905404256279367e-05,
      "loss": 1.5043,
      "step": 46350
    },
    {
      "epoch": 2.0737039985096617,
      "grad_norm": 0.22218751907348633,
      "learning_rate": 9.905080943941068e-05,
      "loss": 1.5047,
      "step": 46360
    },
    {
      "epoch": 2.074245939675174,
      "grad_norm": 0.4120001494884491,
      "learning_rate": 9.904757085920466e-05,
      "loss": 1.4932,
      "step": 46370
    },
    {
      "epoch": 2.0743001337917253,
      "eval_loss": 2.4604697227478027,
      "eval_runtime": 22.0263,
      "eval_samples_per_second": 227.001,
      "eval_steps_per_second": 1.226,
      "step": 46371
    },
    {
      "epoch": 2.0747878808406863,
      "grad_norm": 0.2726476788520813,
      "learning_rate": 9.90443268225768e-05,
      "loss": 1.4971,
      "step": 46380
    },
    {
      "epoch": 2.0753298220061986,
      "grad_norm": 0.25668570399284363,
      "learning_rate": 9.904107732992896e-05,
      "loss": 1.5062,
      "step": 46390
    },
    {
      "epoch": 2.0758717631717105,
      "grad_norm": 0.3078915476799011,
      "learning_rate": 9.903782238166364e-05,
      "loss": 1.504,
      "step": 46400
    },
    {
      "epoch": 2.076413704337223,
      "grad_norm": 0.23746353387832642,
      "learning_rate": 9.903456197818411e-05,
      "loss": 1.5084,
      "step": 46410
    },
    {
      "epoch": 2.076955645502735,
      "grad_norm": 0.1999620497226715,
      "learning_rate": 9.90312961198942e-05,
      "loss": 1.4992,
      "step": 46420
    },
    {
      "epoch": 2.0774975866682475,
      "grad_norm": 0.3216739594936371,
      "learning_rate": 9.90280248071985e-05,
      "loss": 1.5011,
      "step": 46430
    },
    {
      "epoch": 2.0780395278337593,
      "grad_norm": 0.21200136840343475,
      "learning_rate": 9.902474804050224e-05,
      "loss": 1.4959,
      "step": 46440
    },
    {
      "epoch": 2.0785814689992717,
      "grad_norm": 0.21161618828773499,
      "learning_rate": 9.902146582021133e-05,
      "loss": 1.5019,
      "step": 46450
    },
    {
      "epoch": 2.0790150219316814,
      "eval_loss": 2.4619641304016113,
      "eval_runtime": 21.9868,
      "eval_samples_per_second": 227.409,
      "eval_steps_per_second": 1.228,
      "step": 46458
    },
    {
      "epoch": 2.079123410164784,
      "grad_norm": 0.28815415501594543,
      "learning_rate": 9.901817814673236e-05,
      "loss": 1.4998,
      "step": 46460
    },
    {
      "epoch": 2.0796653513302963,
      "grad_norm": 0.21010233461856842,
      "learning_rate": 9.901488502047257e-05,
      "loss": 1.5081,
      "step": 46470
    },
    {
      "epoch": 2.0802072924958086,
      "grad_norm": 0.21125809848308563,
      "learning_rate": 9.901158644183993e-05,
      "loss": 1.5111,
      "step": 46480
    },
    {
      "epoch": 2.0807492336613205,
      "grad_norm": 0.2528863251209259,
      "learning_rate": 9.900828241124303e-05,
      "loss": 1.4949,
      "step": 46490
    },
    {
      "epoch": 2.081291174826833,
      "grad_norm": 0.5834783911705017,
      "learning_rate": 9.900497292909119e-05,
      "loss": 1.5182,
      "step": 46500
    },
    {
      "epoch": 2.081833115992345,
      "grad_norm": 0.47388049960136414,
      "learning_rate": 9.900165799579434e-05,
      "loss": 1.4988,
      "step": 46510
    },
    {
      "epoch": 2.0823750571578574,
      "grad_norm": 0.2034447342157364,
      "learning_rate": 9.899833761176312e-05,
      "loss": 1.4929,
      "step": 46520
    },
    {
      "epoch": 2.0829169983233697,
      "grad_norm": 0.2005794793367386,
      "learning_rate": 9.899501177740889e-05,
      "loss": 1.4957,
      "step": 46530
    },
    {
      "epoch": 2.0834589394888816,
      "grad_norm": 0.3143729269504547,
      "learning_rate": 9.899168049314358e-05,
      "loss": 1.4937,
      "step": 46540
    },
    {
      "epoch": 2.083729910071638,
      "eval_loss": 2.4592928886413574,
      "eval_runtime": 21.99,
      "eval_samples_per_second": 227.376,
      "eval_steps_per_second": 1.228,
      "step": 46545
    },
    {
      "epoch": 2.084000880654394,
      "grad_norm": 0.3189544379711151,
      "learning_rate": 9.898834375937991e-05,
      "loss": 1.5108,
      "step": 46550
    },
    {
      "epoch": 2.0845428218199062,
      "grad_norm": 0.4922240674495697,
      "learning_rate": 9.898500157653118e-05,
      "loss": 1.51,
      "step": 46560
    },
    {
      "epoch": 2.0850847629854186,
      "grad_norm": 0.18532593548297882,
      "learning_rate": 9.898165394501142e-05,
      "loss": 1.5036,
      "step": 46570
    },
    {
      "epoch": 2.0856267041509304,
      "grad_norm": 0.26253679394721985,
      "learning_rate": 9.897830086523531e-05,
      "loss": 1.5052,
      "step": 46580
    },
    {
      "epoch": 2.0861686453164427,
      "grad_norm": 0.36160239577293396,
      "learning_rate": 9.897494233761823e-05,
      "loss": 1.5125,
      "step": 46590
    },
    {
      "epoch": 2.086710586481955,
      "grad_norm": 0.25832414627075195,
      "learning_rate": 9.897157836257621e-05,
      "loss": 1.5106,
      "step": 46600
    },
    {
      "epoch": 2.0872525276474674,
      "grad_norm": 0.32197335362434387,
      "learning_rate": 9.896820894052598e-05,
      "loss": 1.5029,
      "step": 46610
    },
    {
      "epoch": 2.0877944688129797,
      "grad_norm": 0.18866315484046936,
      "learning_rate": 9.896483407188492e-05,
      "loss": 1.502,
      "step": 46620
    },
    {
      "epoch": 2.0883364099784916,
      "grad_norm": 0.25076818466186523,
      "learning_rate": 9.896145375707106e-05,
      "loss": 1.5093,
      "step": 46630
    },
    {
      "epoch": 2.088444798211594,
      "eval_loss": 2.461345911026001,
      "eval_runtime": 26.7525,
      "eval_samples_per_second": 186.898,
      "eval_steps_per_second": 1.009,
      "step": 46632
    },
    {
      "epoch": 2.088878351144004,
      "grad_norm": 0.20240730047225952,
      "learning_rate": 9.89580679965032e-05,
      "loss": 1.508,
      "step": 46640
    },
    {
      "epoch": 2.089420292309516,
      "grad_norm": 0.2117396742105484,
      "learning_rate": 9.895467679060071e-05,
      "loss": 1.4989,
      "step": 46650
    },
    {
      "epoch": 2.0899622334750285,
      "grad_norm": 0.3446267545223236,
      "learning_rate": 9.89512801397837e-05,
      "loss": 1.5041,
      "step": 46660
    },
    {
      "epoch": 2.0905041746405404,
      "grad_norm": 0.22244545817375183,
      "learning_rate": 9.89478780444729e-05,
      "loss": 1.5003,
      "step": 46670
    },
    {
      "epoch": 2.0910461158060527,
      "grad_norm": 0.18350328505039215,
      "learning_rate": 9.894447050508981e-05,
      "loss": 1.5068,
      "step": 46680
    },
    {
      "epoch": 2.091588056971565,
      "grad_norm": 0.2232082635164261,
      "learning_rate": 9.894105752205648e-05,
      "loss": 1.5096,
      "step": 46690
    },
    {
      "epoch": 2.0921299981370773,
      "grad_norm": 0.2229575365781784,
      "learning_rate": 9.893763909579571e-05,
      "loss": 1.4963,
      "step": 46700
    },
    {
      "epoch": 2.0926719393025897,
      "grad_norm": 0.22202491760253906,
      "learning_rate": 9.893421522673098e-05,
      "loss": 1.4979,
      "step": 46710
    },
    {
      "epoch": 2.0931596863515503,
      "eval_loss": 2.4592254161834717,
      "eval_runtime": 22.1699,
      "eval_samples_per_second": 225.531,
      "eval_steps_per_second": 1.218,
      "step": 46719
    },
    {
      "epoch": 2.0932138804681015,
      "grad_norm": 0.28266942501068115,
      "learning_rate": 9.89307859152864e-05,
      "loss": 1.5001,
      "step": 46720
    },
    {
      "epoch": 2.093755821633614,
      "grad_norm": 0.24115784466266632,
      "learning_rate": 9.89273511618868e-05,
      "loss": 1.501,
      "step": 46730
    },
    {
      "epoch": 2.094297762799126,
      "grad_norm": 0.3385097086429596,
      "learning_rate": 9.892391096695766e-05,
      "loss": 1.5128,
      "step": 46740
    },
    {
      "epoch": 2.0948397039646385,
      "grad_norm": 0.2886834144592285,
      "learning_rate": 9.89204653309251e-05,
      "loss": 1.4963,
      "step": 46750
    },
    {
      "epoch": 2.0953816451301503,
      "grad_norm": 0.40250498056411743,
      "learning_rate": 9.891701425421599e-05,
      "loss": 1.5119,
      "step": 46760
    },
    {
      "epoch": 2.0959235862956627,
      "grad_norm": 0.20079436898231506,
      "learning_rate": 9.891355773725783e-05,
      "loss": 1.5029,
      "step": 46770
    },
    {
      "epoch": 2.096465527461175,
      "grad_norm": 0.26996469497680664,
      "learning_rate": 9.891009578047879e-05,
      "loss": 1.4942,
      "step": 46780
    },
    {
      "epoch": 2.0970074686266873,
      "grad_norm": 0.31331729888916016,
      "learning_rate": 9.890662838430771e-05,
      "loss": 1.5042,
      "step": 46790
    },
    {
      "epoch": 2.0975494097921996,
      "grad_norm": 0.27219855785369873,
      "learning_rate": 9.890315554917415e-05,
      "loss": 1.5004,
      "step": 46800
    },
    {
      "epoch": 2.097874574491507,
      "eval_loss": 2.4539170265197754,
      "eval_runtime": 22.0178,
      "eval_samples_per_second": 227.089,
      "eval_steps_per_second": 1.226,
      "step": 46806
    },
    {
      "epoch": 2.0980913509577115,
      "grad_norm": 0.2593035399913788,
      "learning_rate": 9.889967727550829e-05,
      "loss": 1.495,
      "step": 46810
    },
    {
      "epoch": 2.098633292123224,
      "grad_norm": 0.219623863697052,
      "learning_rate": 9.889619356374097e-05,
      "loss": 1.5117,
      "step": 46820
    },
    {
      "epoch": 2.099175233288736,
      "grad_norm": 0.3427010774612427,
      "learning_rate": 9.88927044143038e-05,
      "loss": 1.5115,
      "step": 46830
    },
    {
      "epoch": 2.0997171744542484,
      "grad_norm": 0.2844060957431793,
      "learning_rate": 9.888920982762895e-05,
      "loss": 1.5038,
      "step": 46840
    },
    {
      "epoch": 2.1002591156197603,
      "grad_norm": 0.6262246966362,
      "learning_rate": 9.888570980414935e-05,
      "loss": 1.4964,
      "step": 46850
    },
    {
      "epoch": 2.1008010567852726,
      "grad_norm": 0.3818177878856659,
      "learning_rate": 9.888220434429856e-05,
      "loss": 1.499,
      "step": 46860
    },
    {
      "epoch": 2.101342997950785,
      "grad_norm": 0.19796334207057953,
      "learning_rate": 9.887869344851081e-05,
      "loss": 1.5019,
      "step": 46870
    },
    {
      "epoch": 2.1018849391162973,
      "grad_norm": 0.40566131472587585,
      "learning_rate": 9.8875177117221e-05,
      "loss": 1.4951,
      "step": 46880
    },
    {
      "epoch": 2.1024268802818096,
      "grad_norm": 0.2360570877790451,
      "learning_rate": 9.887165535086473e-05,
      "loss": 1.501,
      "step": 46890
    },
    {
      "epoch": 2.102589462631463,
      "eval_loss": 2.4633328914642334,
      "eval_runtime": 22.3971,
      "eval_samples_per_second": 223.243,
      "eval_steps_per_second": 1.206,
      "step": 46893
    },
    {
      "epoch": 2.1029688214473214,
      "grad_norm": 0.41195404529571533,
      "learning_rate": 9.88681281498783e-05,
      "loss": 1.5036,
      "step": 46900
    },
    {
      "epoch": 2.1035107626128338,
      "grad_norm": 0.30758100748062134,
      "learning_rate": 9.886459551469858e-05,
      "loss": 1.5054,
      "step": 46910
    },
    {
      "epoch": 2.104052703778346,
      "grad_norm": 0.40162351727485657,
      "learning_rate": 9.886105744576322e-05,
      "loss": 1.4949,
      "step": 46920
    },
    {
      "epoch": 2.1045946449438584,
      "grad_norm": 0.32191726565361023,
      "learning_rate": 9.88575139435105e-05,
      "loss": 1.5065,
      "step": 46930
    },
    {
      "epoch": 2.1051365861093707,
      "grad_norm": 0.5494133830070496,
      "learning_rate": 9.885396500837934e-05,
      "loss": 1.5051,
      "step": 46940
    },
    {
      "epoch": 2.1056785272748826,
      "grad_norm": 0.23451875150203705,
      "learning_rate": 9.88504106408094e-05,
      "loss": 1.4936,
      "step": 46950
    },
    {
      "epoch": 2.106220468440395,
      "grad_norm": 0.2888700067996979,
      "learning_rate": 9.884685084124098e-05,
      "loss": 1.4968,
      "step": 46960
    },
    {
      "epoch": 2.106762409605907,
      "grad_norm": 0.24672222137451172,
      "learning_rate": 9.884328561011503e-05,
      "loss": 1.4998,
      "step": 46970
    },
    {
      "epoch": 2.1073043507714195,
      "grad_norm": 0.2256728857755661,
      "learning_rate": 9.883971494787322e-05,
      "loss": 1.5046,
      "step": 46980
    },
    {
      "epoch": 2.1073043507714195,
      "eval_loss": 2.471791982650757,
      "eval_runtime": 21.9693,
      "eval_samples_per_second": 227.591,
      "eval_steps_per_second": 1.229,
      "step": 46980
    },
    {
      "epoch": 2.1078462919369314,
      "grad_norm": 0.21322891116142273,
      "learning_rate": 9.883613885495785e-05,
      "loss": 1.5042,
      "step": 46990
    },
    {
      "epoch": 2.1083882331024437,
      "grad_norm": 0.35615795850753784,
      "learning_rate": 9.88325573318119e-05,
      "loss": 1.4933,
      "step": 47000
    },
    {
      "epoch": 2.108930174267956,
      "grad_norm": 0.3552355170249939,
      "learning_rate": 9.882897037887907e-05,
      "loss": 1.5039,
      "step": 47010
    },
    {
      "epoch": 2.1094721154334684,
      "grad_norm": 0.6598255038261414,
      "learning_rate": 9.882537799660368e-05,
      "loss": 1.4987,
      "step": 47020
    },
    {
      "epoch": 2.1100140565989807,
      "grad_norm": 0.22862312197685242,
      "learning_rate": 9.882178018543071e-05,
      "loss": 1.5196,
      "step": 47030
    },
    {
      "epoch": 2.1105559977644925,
      "grad_norm": 0.3337612450122833,
      "learning_rate": 9.881817694580588e-05,
      "loss": 1.5018,
      "step": 47040
    },
    {
      "epoch": 2.111097938930005,
      "grad_norm": 0.24085493385791779,
      "learning_rate": 9.881456827817553e-05,
      "loss": 1.5019,
      "step": 47050
    },
    {
      "epoch": 2.111639880095517,
      "grad_norm": 0.20780259370803833,
      "learning_rate": 9.881095418298668e-05,
      "loss": 1.5074,
      "step": 47060
    },
    {
      "epoch": 2.1120192389113757,
      "eval_loss": 2.4535341262817383,
      "eval_runtime": 21.9889,
      "eval_samples_per_second": 227.388,
      "eval_steps_per_second": 1.228,
      "step": 47067
    },
    {
      "epoch": 2.1121818212610295,
      "grad_norm": 0.41703495383262634,
      "learning_rate": 9.880733466068704e-05,
      "loss": 1.5105,
      "step": 47070
    },
    {
      "epoch": 2.1127237624265414,
      "grad_norm": 0.3521629571914673,
      "learning_rate": 9.880370971172497e-05,
      "loss": 1.5062,
      "step": 47080
    },
    {
      "epoch": 2.1132657035920537,
      "grad_norm": 0.3287181854248047,
      "learning_rate": 9.880007933654953e-05,
      "loss": 1.4956,
      "step": 47090
    },
    {
      "epoch": 2.113807644757566,
      "grad_norm": 0.3003343939781189,
      "learning_rate": 9.87964435356104e-05,
      "loss": 1.5089,
      "step": 47100
    },
    {
      "epoch": 2.1143495859230783,
      "grad_norm": 0.2599339187145233,
      "learning_rate": 9.879280230935801e-05,
      "loss": 1.4982,
      "step": 47110
    },
    {
      "epoch": 2.1148915270885906,
      "grad_norm": 0.22060526907444,
      "learning_rate": 9.878915565824341e-05,
      "loss": 1.4989,
      "step": 47120
    },
    {
      "epoch": 2.1154334682541025,
      "grad_norm": 0.28959226608276367,
      "learning_rate": 9.878550358271829e-05,
      "loss": 1.5042,
      "step": 47130
    },
    {
      "epoch": 2.115975409419615,
      "grad_norm": 0.31336092948913574,
      "learning_rate": 9.878184608323509e-05,
      "loss": 1.4997,
      "step": 47140
    },
    {
      "epoch": 2.116517350585127,
      "grad_norm": 0.23946624994277954,
      "learning_rate": 9.877818316024689e-05,
      "loss": 1.5056,
      "step": 47150
    },
    {
      "epoch": 2.116734127051332,
      "eval_loss": 2.464033365249634,
      "eval_runtime": 21.9877,
      "eval_samples_per_second": 227.4,
      "eval_steps_per_second": 1.228,
      "step": 47154
    },
    {
      "epoch": 2.1170592917506394,
      "grad_norm": 0.23651756346225739,
      "learning_rate": 9.877451481420742e-05,
      "loss": 1.502,
      "step": 47160
    },
    {
      "epoch": 2.1176012329161518,
      "grad_norm": 0.2244928926229477,
      "learning_rate": 9.877084104557111e-05,
      "loss": 1.5014,
      "step": 47170
    },
    {
      "epoch": 2.1181431740816636,
      "grad_norm": 0.24885980784893036,
      "learning_rate": 9.876716185479303e-05,
      "loss": 1.5075,
      "step": 47180
    },
    {
      "epoch": 2.118685115247176,
      "grad_norm": 0.22047527134418488,
      "learning_rate": 9.876347724232898e-05,
      "loss": 1.5042,
      "step": 47190
    },
    {
      "epoch": 2.1192270564126883,
      "grad_norm": 0.2784343659877777,
      "learning_rate": 9.875978720863536e-05,
      "loss": 1.5095,
      "step": 47200
    },
    {
      "epoch": 2.1197689975782006,
      "grad_norm": 0.41502806544303894,
      "learning_rate": 9.875609175416929e-05,
      "loss": 1.498,
      "step": 47210
    },
    {
      "epoch": 2.1203109387437125,
      "grad_norm": 0.23642586171627045,
      "learning_rate": 9.875239087938853e-05,
      "loss": 1.4923,
      "step": 47220
    },
    {
      "epoch": 2.1208528799092248,
      "grad_norm": 0.3005571663379669,
      "learning_rate": 9.874868458475155e-05,
      "loss": 1.5045,
      "step": 47230
    },
    {
      "epoch": 2.121394821074737,
      "grad_norm": 0.2985764443874359,
      "learning_rate": 9.874497287071747e-05,
      "loss": 1.5013,
      "step": 47240
    },
    {
      "epoch": 2.1214490151912884,
      "eval_loss": 2.4620718955993652,
      "eval_runtime": 22.3871,
      "eval_samples_per_second": 223.343,
      "eval_steps_per_second": 1.206,
      "step": 47241
    },
    {
      "epoch": 2.1219367622402494,
      "grad_norm": 0.3452117443084717,
      "learning_rate": 9.874125573774608e-05,
      "loss": 1.5131,
      "step": 47250
    },
    {
      "epoch": 2.1224787034057617,
      "grad_norm": 0.33107510209083557,
      "learning_rate": 9.873753318629781e-05,
      "loss": 1.4932,
      "step": 47260
    },
    {
      "epoch": 2.1230206445712736,
      "grad_norm": 0.20669598877429962,
      "learning_rate": 9.873380521683383e-05,
      "loss": 1.5066,
      "step": 47270
    },
    {
      "epoch": 2.123562585736786,
      "grad_norm": 0.2573641538619995,
      "learning_rate": 9.873007182981592e-05,
      "loss": 1.5003,
      "step": 47280
    },
    {
      "epoch": 2.1241045269022982,
      "grad_norm": 0.2514735460281372,
      "learning_rate": 9.87263330257066e-05,
      "loss": 1.4996,
      "step": 47290
    },
    {
      "epoch": 2.1246464680678105,
      "grad_norm": 0.7629513740539551,
      "learning_rate": 9.872258880496896e-05,
      "loss": 1.5085,
      "step": 47300
    },
    {
      "epoch": 2.1251884092333224,
      "grad_norm": 0.2581771910190582,
      "learning_rate": 9.871883916806686e-05,
      "loss": 1.5017,
      "step": 47310
    },
    {
      "epoch": 2.1257303503988347,
      "grad_norm": 0.22503337264060974,
      "learning_rate": 9.871508411546475e-05,
      "loss": 1.4997,
      "step": 47320
    },
    {
      "epoch": 2.1261639033312445,
      "eval_loss": 2.4661858081817627,
      "eval_runtime": 25.2283,
      "eval_samples_per_second": 198.19,
      "eval_steps_per_second": 1.07,
      "step": 47328
    },
    {
      "epoch": 2.126272291564347,
      "grad_norm": 0.3019942045211792,
      "learning_rate": 9.871132364762782e-05,
      "loss": 1.5063,
      "step": 47330
    },
    {
      "epoch": 2.1268142327298594,
      "grad_norm": 0.3621060848236084,
      "learning_rate": 9.87075577650219e-05,
      "loss": 1.4999,
      "step": 47340
    },
    {
      "epoch": 2.1273561738953717,
      "grad_norm": 0.18868303298950195,
      "learning_rate": 9.870378646811347e-05,
      "loss": 1.5056,
      "step": 47350
    },
    {
      "epoch": 2.1278981150608836,
      "grad_norm": 0.20302945375442505,
      "learning_rate": 9.870000975736974e-05,
      "loss": 1.5063,
      "step": 47360
    },
    {
      "epoch": 2.128440056226396,
      "grad_norm": 0.41298550367355347,
      "learning_rate": 9.86962276332585e-05,
      "loss": 1.5067,
      "step": 47370
    },
    {
      "epoch": 2.128981997391908,
      "grad_norm": 0.25545287132263184,
      "learning_rate": 9.869244009624831e-05,
      "loss": 1.4994,
      "step": 47380
    },
    {
      "epoch": 2.1295239385574205,
      "grad_norm": 0.33961889147758484,
      "learning_rate": 9.868864714680832e-05,
      "loss": 1.5066,
      "step": 47390
    },
    {
      "epoch": 2.1300658797229324,
      "grad_norm": 0.29086834192276,
      "learning_rate": 9.868484878540842e-05,
      "loss": 1.505,
      "step": 47400
    },
    {
      "epoch": 2.1306078208884447,
      "grad_norm": 0.2222570925951004,
      "learning_rate": 9.868104501251909e-05,
      "loss": 1.4965,
      "step": 47410
    },
    {
      "epoch": 2.130878791471201,
      "eval_loss": 2.467705249786377,
      "eval_runtime": 22.0169,
      "eval_samples_per_second": 227.099,
      "eval_steps_per_second": 1.226,
      "step": 47415
    },
    {
      "epoch": 2.131149762053957,
      "grad_norm": 0.25193503499031067,
      "learning_rate": 9.867723582861155e-05,
      "loss": 1.5039,
      "step": 47420
    },
    {
      "epoch": 2.1316917032194693,
      "grad_norm": 0.2993335723876953,
      "learning_rate": 9.867342123415768e-05,
      "loss": 1.5052,
      "step": 47430
    },
    {
      "epoch": 2.1322336443849816,
      "grad_norm": 0.2402217835187912,
      "learning_rate": 9.866960122962998e-05,
      "loss": 1.4947,
      "step": 47440
    },
    {
      "epoch": 2.1327755855504935,
      "grad_norm": 0.47997650504112244,
      "learning_rate": 9.866577581550169e-05,
      "loss": 1.5065,
      "step": 47450
    },
    {
      "epoch": 2.133317526716006,
      "grad_norm": 0.3690037429332733,
      "learning_rate": 9.866194499224665e-05,
      "loss": 1.5024,
      "step": 47460
    },
    {
      "epoch": 2.133859467881518,
      "grad_norm": 0.2865743935108185,
      "learning_rate": 9.865810876033946e-05,
      "loss": 1.4975,
      "step": 47470
    },
    {
      "epoch": 2.1344014090470305,
      "grad_norm": 0.2986805737018585,
      "learning_rate": 9.865426712025527e-05,
      "loss": 1.4964,
      "step": 47480
    },
    {
      "epoch": 2.1349433502125423,
      "grad_norm": 0.44685614109039307,
      "learning_rate": 9.865042007247001e-05,
      "loss": 1.4987,
      "step": 47490
    },
    {
      "epoch": 2.1354852913780547,
      "grad_norm": 0.26483583450317383,
      "learning_rate": 9.864656761746021e-05,
      "loss": 1.4933,
      "step": 47500
    },
    {
      "epoch": 2.135593679611157,
      "eval_loss": 2.4674875736236572,
      "eval_runtime": 22.0484,
      "eval_samples_per_second": 226.774,
      "eval_steps_per_second": 1.225,
      "step": 47502
    },
    {
      "epoch": 2.136027232543567,
      "grad_norm": 0.2297901213169098,
      "learning_rate": 9.864270975570313e-05,
      "loss": 1.5029,
      "step": 47510
    },
    {
      "epoch": 2.1365691737090793,
      "grad_norm": 0.251094251871109,
      "learning_rate": 9.863884648767662e-05,
      "loss": 1.5057,
      "step": 47520
    },
    {
      "epoch": 2.1371111148745916,
      "grad_norm": 0.36560431122779846,
      "learning_rate": 9.863497781385928e-05,
      "loss": 1.4962,
      "step": 47530
    },
    {
      "epoch": 2.1376530560401035,
      "grad_norm": 0.3538174033164978,
      "learning_rate": 9.863110373473033e-05,
      "loss": 1.4831,
      "step": 47540
    },
    {
      "epoch": 2.138194997205616,
      "grad_norm": 0.20020203292369843,
      "learning_rate": 9.862722425076968e-05,
      "loss": 1.5092,
      "step": 47550
    },
    {
      "epoch": 2.138736938371128,
      "grad_norm": 0.3080503046512604,
      "learning_rate": 9.862333936245789e-05,
      "loss": 1.5111,
      "step": 47560
    },
    {
      "epoch": 2.1392788795366404,
      "grad_norm": 0.3943610191345215,
      "learning_rate": 9.861944907027624e-05,
      "loss": 1.5024,
      "step": 47570
    },
    {
      "epoch": 2.1398208207021527,
      "grad_norm": 0.25787243247032166,
      "learning_rate": 9.861555337470658e-05,
      "loss": 1.4955,
      "step": 47580
    },
    {
      "epoch": 2.1403085677511133,
      "eval_loss": 2.4601619243621826,
      "eval_runtime": 22.0199,
      "eval_samples_per_second": 227.067,
      "eval_steps_per_second": 1.226,
      "step": 47589
    },
    {
      "epoch": 2.1403627618676646,
      "grad_norm": 0.2594373822212219,
      "learning_rate": 9.861165227623154e-05,
      "loss": 1.5006,
      "step": 47590
    },
    {
      "epoch": 2.140904703033177,
      "grad_norm": 0.1998678594827652,
      "learning_rate": 9.860774577533438e-05,
      "loss": 1.493,
      "step": 47600
    },
    {
      "epoch": 2.1414466441986892,
      "grad_norm": 0.30177435278892517,
      "learning_rate": 9.860383387249897e-05,
      "loss": 1.5081,
      "step": 47610
    },
    {
      "epoch": 2.1419885853642016,
      "grad_norm": 0.3916855752468109,
      "learning_rate": 9.859991656820994e-05,
      "loss": 1.4935,
      "step": 47620
    },
    {
      "epoch": 2.1425305265297134,
      "grad_norm": 0.3234284520149231,
      "learning_rate": 9.859599386295255e-05,
      "loss": 1.5059,
      "step": 47630
    },
    {
      "epoch": 2.1430724676952257,
      "grad_norm": 0.4442438781261444,
      "learning_rate": 9.85920657572127e-05,
      "loss": 1.5093,
      "step": 47640
    },
    {
      "epoch": 2.143614408860738,
      "grad_norm": 0.194505512714386,
      "learning_rate": 9.858813225147702e-05,
      "loss": 1.4962,
      "step": 47650
    },
    {
      "epoch": 2.1441563500262504,
      "grad_norm": 0.31255653500556946,
      "learning_rate": 9.858419334623273e-05,
      "loss": 1.5066,
      "step": 47660
    },
    {
      "epoch": 2.1446982911917627,
      "grad_norm": 0.2588372230529785,
      "learning_rate": 9.858024904196782e-05,
      "loss": 1.502,
      "step": 47670
    },
    {
      "epoch": 2.14502345589107,
      "eval_loss": 2.4643874168395996,
      "eval_runtime": 21.985,
      "eval_samples_per_second": 227.428,
      "eval_steps_per_second": 1.228,
      "step": 47676
    },
    {
      "epoch": 2.1452402323572746,
      "grad_norm": 0.27704402804374695,
      "learning_rate": 9.857629933917084e-05,
      "loss": 1.5025,
      "step": 47680
    },
    {
      "epoch": 2.145782173522787,
      "grad_norm": 0.34092339873313904,
      "learning_rate": 9.857234423833111e-05,
      "loss": 1.4959,
      "step": 47690
    },
    {
      "epoch": 2.146324114688299,
      "grad_norm": 0.37414854764938354,
      "learning_rate": 9.856838373993852e-05,
      "loss": 1.494,
      "step": 47700
    },
    {
      "epoch": 2.1468660558538115,
      "grad_norm": 0.19338767230510712,
      "learning_rate": 9.856441784448373e-05,
      "loss": 1.5082,
      "step": 47710
    },
    {
      "epoch": 2.1474079970193234,
      "grad_norm": 0.23796340823173523,
      "learning_rate": 9.8560446552458e-05,
      "loss": 1.4824,
      "step": 47720
    },
    {
      "epoch": 2.1479499381848357,
      "grad_norm": 0.3311995565891266,
      "learning_rate": 9.855646986435325e-05,
      "loss": 1.5069,
      "step": 47730
    },
    {
      "epoch": 2.148491879350348,
      "grad_norm": 0.2240724116563797,
      "learning_rate": 9.855248778066212e-05,
      "loss": 1.4987,
      "step": 47740
    },
    {
      "epoch": 2.1490338205158603,
      "grad_norm": 0.22334522008895874,
      "learning_rate": 9.854850030187791e-05,
      "loss": 1.4981,
      "step": 47750
    },
    {
      "epoch": 2.1495757616813727,
      "grad_norm": 0.21262875199317932,
      "learning_rate": 9.854450742849452e-05,
      "loss": 1.5029,
      "step": 47760
    },
    {
      "epoch": 2.149738344031026,
      "eval_loss": 2.4613311290740967,
      "eval_runtime": 21.9832,
      "eval_samples_per_second": 227.447,
      "eval_steps_per_second": 1.228,
      "step": 47763
    },
    {
      "epoch": 2.1501177028468845,
      "grad_norm": 0.2315114289522171,
      "learning_rate": 9.854050916100664e-05,
      "loss": 1.5089,
      "step": 47770
    },
    {
      "epoch": 2.150659644012397,
      "grad_norm": 0.34692510962486267,
      "learning_rate": 9.853650549990949e-05,
      "loss": 1.488,
      "step": 47780
    },
    {
      "epoch": 2.151201585177909,
      "grad_norm": 0.2978134751319885,
      "learning_rate": 9.853249644569907e-05,
      "loss": 1.5069,
      "step": 47790
    },
    {
      "epoch": 2.1517435263434215,
      "grad_norm": 0.2450045347213745,
      "learning_rate": 9.852848199887198e-05,
      "loss": 1.4962,
      "step": 47800
    },
    {
      "epoch": 2.152285467508934,
      "grad_norm": 0.27408814430236816,
      "learning_rate": 9.852446215992552e-05,
      "loss": 1.5035,
      "step": 47810
    },
    {
      "epoch": 2.1528274086744457,
      "grad_norm": 0.288369745016098,
      "learning_rate": 9.852043692935766e-05,
      "loss": 1.5054,
      "step": 47820
    },
    {
      "epoch": 2.153369349839958,
      "grad_norm": 0.33964666724205017,
      "learning_rate": 9.851640630766703e-05,
      "loss": 1.5055,
      "step": 47830
    },
    {
      "epoch": 2.1539112910054703,
      "grad_norm": 0.3584079444408417,
      "learning_rate": 9.851237029535289e-05,
      "loss": 1.4909,
      "step": 47840
    },
    {
      "epoch": 2.1544532321709826,
      "grad_norm": 0.3552767336368561,
      "learning_rate": 9.850832889291525e-05,
      "loss": 1.4932,
      "step": 47850
    },
    {
      "epoch": 2.1544532321709826,
      "eval_loss": 2.4597816467285156,
      "eval_runtime": 21.9841,
      "eval_samples_per_second": 227.437,
      "eval_steps_per_second": 1.228,
      "step": 47850
    },
    {
      "epoch": 2.1549951733364945,
      "grad_norm": 0.24502573907375336,
      "learning_rate": 9.850428210085471e-05,
      "loss": 1.4988,
      "step": 47860
    },
    {
      "epoch": 2.155537114502007,
      "grad_norm": 0.26566943526268005,
      "learning_rate": 9.850022991967258e-05,
      "loss": 1.5044,
      "step": 47870
    },
    {
      "epoch": 2.156079055667519,
      "grad_norm": 0.22309494018554688,
      "learning_rate": 9.849617234987083e-05,
      "loss": 1.5052,
      "step": 47880
    },
    {
      "epoch": 2.1566209968330314,
      "grad_norm": 0.3310026526451111,
      "learning_rate": 9.849210939195209e-05,
      "loss": 1.5026,
      "step": 47890
    },
    {
      "epoch": 2.1571629379985433,
      "grad_norm": 0.24902848899364471,
      "learning_rate": 9.848804104641966e-05,
      "loss": 1.5058,
      "step": 47900
    },
    {
      "epoch": 2.1577048791640556,
      "grad_norm": 0.21098916232585907,
      "learning_rate": 9.848396731377751e-05,
      "loss": 1.5041,
      "step": 47910
    },
    {
      "epoch": 2.158246820329568,
      "grad_norm": 0.23908697068691254,
      "learning_rate": 9.84798881945303e-05,
      "loss": 1.5018,
      "step": 47920
    },
    {
      "epoch": 2.1587887614950803,
      "grad_norm": 0.41047078371047974,
      "learning_rate": 9.847580368918329e-05,
      "loss": 1.5021,
      "step": 47930
    },
    {
      "epoch": 2.1591681203109387,
      "eval_loss": 2.4657371044158936,
      "eval_runtime": 21.9884,
      "eval_samples_per_second": 227.392,
      "eval_steps_per_second": 1.228,
      "step": 47937
    },
    {
      "epoch": 2.1593307026605926,
      "grad_norm": 0.4476509988307953,
      "learning_rate": 9.847171379824248e-05,
      "loss": 1.5102,
      "step": 47940
    },
    {
      "epoch": 2.1598726438261044,
      "grad_norm": 0.43281930685043335,
      "learning_rate": 9.846761852221449e-05,
      "loss": 1.4999,
      "step": 47950
    },
    {
      "epoch": 2.1604145849916168,
      "grad_norm": 0.20760992169380188,
      "learning_rate": 9.846351786160665e-05,
      "loss": 1.5029,
      "step": 47960
    },
    {
      "epoch": 2.160956526157129,
      "grad_norm": 0.24444232881069183,
      "learning_rate": 9.84594118169269e-05,
      "loss": 1.5091,
      "step": 47970
    },
    {
      "epoch": 2.1614984673226414,
      "grad_norm": 0.25241169333457947,
      "learning_rate": 9.84553003886839e-05,
      "loss": 1.5019,
      "step": 47980
    },
    {
      "epoch": 2.1620404084881537,
      "grad_norm": 0.26885291934013367,
      "learning_rate": 9.845118357738696e-05,
      "loss": 1.5068,
      "step": 47990
    },
    {
      "epoch": 2.1625823496536656,
      "grad_norm": 0.26971253752708435,
      "learning_rate": 9.844706138354603e-05,
      "loss": 1.5013,
      "step": 48000
    },
    {
      "epoch": 2.163124290819178,
      "grad_norm": 0.20452235639095306,
      "learning_rate": 9.844293380767178e-05,
      "loss": 1.4947,
      "step": 48010
    },
    {
      "epoch": 2.16366623198469,
      "grad_norm": 0.19715271890163422,
      "learning_rate": 9.843880085027551e-05,
      "loss": 1.496,
      "step": 48020
    },
    {
      "epoch": 2.163883008450895,
      "eval_loss": 2.4602131843566895,
      "eval_runtime": 21.9834,
      "eval_samples_per_second": 227.444,
      "eval_steps_per_second": 1.228,
      "step": 48024
    },
    {
      "epoch": 2.1642081731502025,
      "grad_norm": 0.30456554889678955,
      "learning_rate": 9.843466251186916e-05,
      "loss": 1.5103,
      "step": 48030
    },
    {
      "epoch": 2.1647501143157144,
      "grad_norm": 0.24360065162181854,
      "learning_rate": 9.843051879296539e-05,
      "loss": 1.5072,
      "step": 48040
    },
    {
      "epoch": 2.1652920554812267,
      "grad_norm": 0.2575078308582306,
      "learning_rate": 9.842636969407753e-05,
      "loss": 1.5027,
      "step": 48050
    },
    {
      "epoch": 2.165833996646739,
      "grad_norm": 0.1880762130022049,
      "learning_rate": 9.842221521571951e-05,
      "loss": 1.4936,
      "step": 48060
    },
    {
      "epoch": 2.1663759378122514,
      "grad_norm": 0.24595960974693298,
      "learning_rate": 9.841805535840602e-05,
      "loss": 1.501,
      "step": 48070
    },
    {
      "epoch": 2.1669178789777637,
      "grad_norm": 0.21159407496452332,
      "learning_rate": 9.841389012265231e-05,
      "loss": 1.4968,
      "step": 48080
    },
    {
      "epoch": 2.1674598201432755,
      "grad_norm": 0.4702402353286743,
      "learning_rate": 9.840971950897437e-05,
      "loss": 1.4929,
      "step": 48090
    },
    {
      "epoch": 2.168001761308788,
      "grad_norm": 0.23953235149383545,
      "learning_rate": 9.840554351788885e-05,
      "loss": 1.4948,
      "step": 48100
    },
    {
      "epoch": 2.1685437024743,
      "grad_norm": 0.21100491285324097,
      "learning_rate": 9.840136214991305e-05,
      "loss": 1.4927,
      "step": 48110
    },
    {
      "epoch": 2.1685978965908514,
      "eval_loss": 2.4579579830169678,
      "eval_runtime": 21.9856,
      "eval_samples_per_second": 227.422,
      "eval_steps_per_second": 1.228,
      "step": 48111
    },
    {
      "epoch": 2.1690856436398125,
      "grad_norm": 0.2241096794605255,
      "learning_rate": 9.839717540556495e-05,
      "loss": 1.4958,
      "step": 48120
    },
    {
      "epoch": 2.1696275848053244,
      "grad_norm": 0.19508574903011322,
      "learning_rate": 9.839298328536316e-05,
      "loss": 1.4886,
      "step": 48130
    },
    {
      "epoch": 2.1701695259708367,
      "grad_norm": 0.3244903087615967,
      "learning_rate": 9.838878578982699e-05,
      "loss": 1.4895,
      "step": 48140
    },
    {
      "epoch": 2.170711467136349,
      "grad_norm": 0.3189353346824646,
      "learning_rate": 9.838458291947641e-05,
      "loss": 1.4833,
      "step": 48150
    },
    {
      "epoch": 2.1712534083018613,
      "grad_norm": 0.20821264386177063,
      "learning_rate": 9.838037467483207e-05,
      "loss": 1.498,
      "step": 48160
    },
    {
      "epoch": 2.1717953494673736,
      "grad_norm": 0.21814101934432983,
      "learning_rate": 9.837616105641523e-05,
      "loss": 1.4939,
      "step": 48170
    },
    {
      "epoch": 2.1723372906328855,
      "grad_norm": 0.2125815451145172,
      "learning_rate": 9.837194206474789e-05,
      "loss": 1.4896,
      "step": 48180
    },
    {
      "epoch": 2.172879231798398,
      "grad_norm": 0.20504556596279144,
      "learning_rate": 9.83677177003527e-05,
      "loss": 1.498,
      "step": 48190
    },
    {
      "epoch": 2.1733127847308076,
      "eval_loss": 2.4657247066497803,
      "eval_runtime": 22.1285,
      "eval_samples_per_second": 225.953,
      "eval_steps_per_second": 1.22,
      "step": 48198
    },
    {
      "epoch": 2.17342117296391,
      "grad_norm": 0.21012401580810547,
      "learning_rate": 9.836348796375288e-05,
      "loss": 1.4979,
      "step": 48200
    },
    {
      "epoch": 2.1739631141294224,
      "grad_norm": 0.42915037274360657,
      "learning_rate": 9.835925285547245e-05,
      "loss": 1.4883,
      "step": 48210
    },
    {
      "epoch": 2.1745050552949348,
      "grad_norm": 0.350690633058548,
      "learning_rate": 9.835501237603603e-05,
      "loss": 1.5056,
      "step": 48220
    },
    {
      "epoch": 2.1750469964604466,
      "grad_norm": 0.23836711049079895,
      "learning_rate": 9.835076652596891e-05,
      "loss": 1.5043,
      "step": 48230
    },
    {
      "epoch": 2.175588937625959,
      "grad_norm": 0.4392627477645874,
      "learning_rate": 9.834651530579703e-05,
      "loss": 1.489,
      "step": 48240
    },
    {
      "epoch": 2.1761308787914713,
      "grad_norm": 0.1950349509716034,
      "learning_rate": 9.834225871604701e-05,
      "loss": 1.4914,
      "step": 48250
    },
    {
      "epoch": 2.1766728199569836,
      "grad_norm": 0.36042729020118713,
      "learning_rate": 9.833799675724619e-05,
      "loss": 1.5052,
      "step": 48260
    },
    {
      "epoch": 2.1772147611224955,
      "grad_norm": 0.3218774199485779,
      "learning_rate": 9.833372942992248e-05,
      "loss": 1.4963,
      "step": 48270
    },
    {
      "epoch": 2.1777567022880078,
      "grad_norm": 0.17845819890499115,
      "learning_rate": 9.832945673460448e-05,
      "loss": 1.494,
      "step": 48280
    },
    {
      "epoch": 2.178027672870764,
      "eval_loss": 2.4591383934020996,
      "eval_runtime": 22.2137,
      "eval_samples_per_second": 225.086,
      "eval_steps_per_second": 1.215,
      "step": 48285
    },
    {
      "epoch": 2.17829864345352,
      "grad_norm": 0.36038216948509216,
      "learning_rate": 9.832517867182151e-05,
      "loss": 1.4997,
      "step": 48290
    },
    {
      "epoch": 2.1788405846190324,
      "grad_norm": 0.2766604721546173,
      "learning_rate": 9.832089524210352e-05,
      "loss": 1.4955,
      "step": 48300
    },
    {
      "epoch": 2.1793825257845447,
      "grad_norm": 0.2935325801372528,
      "learning_rate": 9.831660644598109e-05,
      "loss": 1.5001,
      "step": 48310
    },
    {
      "epoch": 2.1799244669500566,
      "grad_norm": 0.3200201094150543,
      "learning_rate": 9.831231228398553e-05,
      "loss": 1.4945,
      "step": 48320
    },
    {
      "epoch": 2.180466408115569,
      "grad_norm": 0.27418839931488037,
      "learning_rate": 9.830801275664876e-05,
      "loss": 1.5014,
      "step": 48330
    },
    {
      "epoch": 2.1810083492810812,
      "grad_norm": 0.26256710290908813,
      "learning_rate": 9.830370786450339e-05,
      "loss": 1.5014,
      "step": 48340
    },
    {
      "epoch": 2.1815502904465935,
      "grad_norm": 0.4114410877227783,
      "learning_rate": 9.82993976080827e-05,
      "loss": 1.51,
      "step": 48350
    },
    {
      "epoch": 2.1820922316121054,
      "grad_norm": 0.3471300005912781,
      "learning_rate": 9.829508198792062e-05,
      "loss": 1.5092,
      "step": 48360
    },
    {
      "epoch": 2.1826341727776177,
      "grad_norm": 0.356444776058197,
      "learning_rate": 9.829076100455176e-05,
      "loss": 1.4959,
      "step": 48370
    },
    {
      "epoch": 2.1827425610107203,
      "eval_loss": 2.4686012268066406,
      "eval_runtime": 22.0175,
      "eval_samples_per_second": 227.092,
      "eval_steps_per_second": 1.226,
      "step": 48372
    },
    {
      "epoch": 2.18317611394313,
      "grad_norm": 0.2611772418022156,
      "learning_rate": 9.828643465851137e-05,
      "loss": 1.4966,
      "step": 48380
    },
    {
      "epoch": 2.1837180551086424,
      "grad_norm": 0.43987488746643066,
      "learning_rate": 9.828210295033537e-05,
      "loss": 1.4965,
      "step": 48390
    },
    {
      "epoch": 2.1842599962741547,
      "grad_norm": 0.24238485097885132,
      "learning_rate": 9.82777658805604e-05,
      "loss": 1.4908,
      "step": 48400
    },
    {
      "epoch": 2.1848019374396666,
      "grad_norm": 0.2008477747440338,
      "learning_rate": 9.827342344972366e-05,
      "loss": 1.4946,
      "step": 48410
    },
    {
      "epoch": 2.185343878605179,
      "grad_norm": 0.21971365809440613,
      "learning_rate": 9.826907565836311e-05,
      "loss": 1.4978,
      "step": 48420
    },
    {
      "epoch": 2.185885819770691,
      "grad_norm": 0.35851284861564636,
      "learning_rate": 9.826472250701732e-05,
      "loss": 1.4988,
      "step": 48430
    },
    {
      "epoch": 2.1864277609362035,
      "grad_norm": 0.24326647818088531,
      "learning_rate": 9.826036399622553e-05,
      "loss": 1.4949,
      "step": 48440
    },
    {
      "epoch": 2.186969702101716,
      "grad_norm": 0.23622386157512665,
      "learning_rate": 9.82560001265277e-05,
      "loss": 1.4959,
      "step": 48450
    },
    {
      "epoch": 2.1874574491506764,
      "eval_loss": 2.4709348678588867,
      "eval_runtime": 22.0771,
      "eval_samples_per_second": 226.479,
      "eval_steps_per_second": 1.223,
      "step": 48459
    },
    {
      "epoch": 2.1875116432672277,
      "grad_norm": 0.22727850079536438,
      "learning_rate": 9.825163089846434e-05,
      "loss": 1.4841,
      "step": 48460
    },
    {
      "epoch": 2.18805358443274,
      "grad_norm": 0.291652649641037,
      "learning_rate": 9.824725631257674e-05,
      "loss": 1.4959,
      "step": 48470
    },
    {
      "epoch": 2.1885955255982523,
      "grad_norm": 0.27845892310142517,
      "learning_rate": 9.824287636940678e-05,
      "loss": 1.4919,
      "step": 48480
    },
    {
      "epoch": 2.1891374667637646,
      "grad_norm": 0.3905391991138458,
      "learning_rate": 9.823849106949704e-05,
      "loss": 1.4948,
      "step": 48490
    },
    {
      "epoch": 2.1896794079292765,
      "grad_norm": 0.2136894315481186,
      "learning_rate": 9.823410041339075e-05,
      "loss": 1.4967,
      "step": 48500
    },
    {
      "epoch": 2.190221349094789,
      "grad_norm": 0.3162279725074768,
      "learning_rate": 9.82297044016318e-05,
      "loss": 1.4927,
      "step": 48510
    },
    {
      "epoch": 2.190763290260301,
      "grad_norm": 0.24722278118133545,
      "learning_rate": 9.822530303476476e-05,
      "loss": 1.4965,
      "step": 48520
    },
    {
      "epoch": 2.1913052314258135,
      "grad_norm": 0.2130030393600464,
      "learning_rate": 9.822089631333484e-05,
      "loss": 1.5057,
      "step": 48530
    },
    {
      "epoch": 2.1918471725913253,
      "grad_norm": 0.2408854067325592,
      "learning_rate": 9.821648423788794e-05,
      "loss": 1.5172,
      "step": 48540
    },
    {
      "epoch": 2.192172337290633,
      "eval_loss": 2.4656124114990234,
      "eval_runtime": 22.0104,
      "eval_samples_per_second": 227.165,
      "eval_steps_per_second": 1.227,
      "step": 48546
    },
    {
      "epoch": 2.1923891137568376,
      "grad_norm": 0.21235163509845734,
      "learning_rate": 9.821206680897059e-05,
      "loss": 1.4993,
      "step": 48550
    },
    {
      "epoch": 2.19293105492235,
      "grad_norm": 0.365730345249176,
      "learning_rate": 9.820764402713002e-05,
      "loss": 1.4979,
      "step": 48560
    },
    {
      "epoch": 2.1934729960878623,
      "grad_norm": 0.26309072971343994,
      "learning_rate": 9.82032158929141e-05,
      "loss": 1.4962,
      "step": 48570
    },
    {
      "epoch": 2.1940149372533746,
      "grad_norm": 0.3084958493709564,
      "learning_rate": 9.819878240687136e-05,
      "loss": 1.4912,
      "step": 48580
    },
    {
      "epoch": 2.1945568784188865,
      "grad_norm": 0.2697133719921112,
      "learning_rate": 9.8194343569551e-05,
      "loss": 1.4936,
      "step": 48590
    },
    {
      "epoch": 2.195098819584399,
      "grad_norm": 0.31936559081077576,
      "learning_rate": 9.81898993815029e-05,
      "loss": 1.4938,
      "step": 48600
    },
    {
      "epoch": 2.195640760749911,
      "grad_norm": 0.2950705289840698,
      "learning_rate": 9.818544984327756e-05,
      "loss": 1.501,
      "step": 48610
    },
    {
      "epoch": 2.1961827019154234,
      "grad_norm": 0.49501827359199524,
      "learning_rate": 9.818099495542619e-05,
      "loss": 1.5048,
      "step": 48620
    },
    {
      "epoch": 2.1967246430809357,
      "grad_norm": 0.22042915225028992,
      "learning_rate": 9.817653471850066e-05,
      "loss": 1.4881,
      "step": 48630
    },
    {
      "epoch": 2.196887225430589,
      "eval_loss": 2.4624252319335938,
      "eval_runtime": 22.0712,
      "eval_samples_per_second": 226.54,
      "eval_steps_per_second": 1.223,
      "step": 48633
    },
    {
      "epoch": 2.1972665842464476,
      "grad_norm": 0.26132088899612427,
      "learning_rate": 9.817206913305344e-05,
      "loss": 1.497,
      "step": 48640
    },
    {
      "epoch": 2.19780852541196,
      "grad_norm": 0.19975581765174866,
      "learning_rate": 9.816759819963773e-05,
      "loss": 1.483,
      "step": 48650
    },
    {
      "epoch": 2.1983504665774722,
      "grad_norm": 0.2904106080532074,
      "learning_rate": 9.816312191880738e-05,
      "loss": 1.4892,
      "step": 48660
    },
    {
      "epoch": 2.1988924077429846,
      "grad_norm": 0.2334350049495697,
      "learning_rate": 9.815864029111688e-05,
      "loss": 1.4904,
      "step": 48670
    },
    {
      "epoch": 2.1994343489084964,
      "grad_norm": 0.24541063606739044,
      "learning_rate": 9.81541533171214e-05,
      "loss": 1.4901,
      "step": 48680
    },
    {
      "epoch": 2.1999762900740087,
      "grad_norm": 0.1976650059223175,
      "learning_rate": 9.814966099737676e-05,
      "loss": 1.479,
      "step": 48690
    },
    {
      "epoch": 2.200518231239521,
      "grad_norm": 0.22315511107444763,
      "learning_rate": 9.814516333243945e-05,
      "loss": 1.5053,
      "step": 48700
    },
    {
      "epoch": 2.2010601724050334,
      "grad_norm": 0.3041779100894928,
      "learning_rate": 9.814066032286664e-05,
      "loss": 1.4908,
      "step": 48710
    },
    {
      "epoch": 2.2016021135705457,
      "grad_norm": 0.3959655463695526,
      "learning_rate": 9.813615196921611e-05,
      "loss": 1.497,
      "step": 48720
    },
    {
      "epoch": 2.2016021135705457,
      "eval_loss": 2.467682361602783,
      "eval_runtime": 21.8632,
      "eval_samples_per_second": 228.695,
      "eval_steps_per_second": 1.235,
      "step": 48720
    },
    {
      "epoch": 2.2021440547360576,
      "grad_norm": 0.2074437290430069,
      "learning_rate": 9.813163827204634e-05,
      "loss": 1.4933,
      "step": 48730
    },
    {
      "epoch": 2.20268599590157,
      "grad_norm": 0.24793848395347595,
      "learning_rate": 9.812711923191651e-05,
      "loss": 1.5001,
      "step": 48740
    },
    {
      "epoch": 2.203227937067082,
      "grad_norm": 0.2648943364620209,
      "learning_rate": 9.812259484938638e-05,
      "loss": 1.4939,
      "step": 48750
    },
    {
      "epoch": 2.2037698782325945,
      "grad_norm": 0.4681974947452545,
      "learning_rate": 9.811806512501641e-05,
      "loss": 1.4988,
      "step": 48760
    },
    {
      "epoch": 2.2043118193981064,
      "grad_norm": 0.30370041728019714,
      "learning_rate": 9.811353005936774e-05,
      "loss": 1.4922,
      "step": 48770
    },
    {
      "epoch": 2.2048537605636187,
      "grad_norm": 0.26227137446403503,
      "learning_rate": 9.810898965300213e-05,
      "loss": 1.4986,
      "step": 48780
    },
    {
      "epoch": 2.205395701729131,
      "grad_norm": 0.3216424286365509,
      "learning_rate": 9.810444390648207e-05,
      "loss": 1.493,
      "step": 48790
    },
    {
      "epoch": 2.2059376428946433,
      "grad_norm": 0.19285954535007477,
      "learning_rate": 9.809989282037062e-05,
      "loss": 1.4856,
      "step": 48800
    },
    {
      "epoch": 2.206317001710502,
      "eval_loss": 2.4649438858032227,
      "eval_runtime": 22.0562,
      "eval_samples_per_second": 226.694,
      "eval_steps_per_second": 1.224,
      "step": 48807
    },
    {
      "epoch": 2.2064795840601557,
      "grad_norm": 0.5123102068901062,
      "learning_rate": 9.809533639523156e-05,
      "loss": 1.494,
      "step": 48810
    },
    {
      "epoch": 2.2070215252256675,
      "grad_norm": 0.23556865751743317,
      "learning_rate": 9.809077463162935e-05,
      "loss": 1.4987,
      "step": 48820
    },
    {
      "epoch": 2.20756346639118,
      "grad_norm": 0.25429365038871765,
      "learning_rate": 9.808620753012906e-05,
      "loss": 1.492,
      "step": 48830
    },
    {
      "epoch": 2.208105407556692,
      "grad_norm": 0.2279432713985443,
      "learning_rate": 9.808163509129643e-05,
      "loss": 1.4876,
      "step": 48840
    },
    {
      "epoch": 2.2086473487222045,
      "grad_norm": 0.23476704955101013,
      "learning_rate": 9.807705731569789e-05,
      "loss": 1.4923,
      "step": 48850
    },
    {
      "epoch": 2.209189289887717,
      "grad_norm": 0.19318480789661407,
      "learning_rate": 9.80724742039005e-05,
      "loss": 1.4884,
      "step": 48860
    },
    {
      "epoch": 2.2097312310532287,
      "grad_norm": 0.2585408091545105,
      "learning_rate": 9.806788575647204e-05,
      "loss": 1.5005,
      "step": 48870
    },
    {
      "epoch": 2.210273172218741,
      "grad_norm": 0.23720701038837433,
      "learning_rate": 9.806329197398085e-05,
      "loss": 1.4892,
      "step": 48880
    },
    {
      "epoch": 2.2108151133842533,
      "grad_norm": 0.3471413552761078,
      "learning_rate": 9.805869285699602e-05,
      "loss": 1.4984,
      "step": 48890
    },
    {
      "epoch": 2.211031889850458,
      "eval_loss": 2.4734652042388916,
      "eval_runtime": 22.0944,
      "eval_samples_per_second": 226.302,
      "eval_steps_per_second": 1.222,
      "step": 48894
    },
    {
      "epoch": 2.2113570545497656,
      "grad_norm": 0.4286750555038452,
      "learning_rate": 9.805408840608725e-05,
      "loss": 1.4975,
      "step": 48900
    },
    {
      "epoch": 2.2118989957152775,
      "grad_norm": 0.4188242256641388,
      "learning_rate": 9.804947862182494e-05,
      "loss": 1.4967,
      "step": 48910
    },
    {
      "epoch": 2.21244093688079,
      "grad_norm": 0.3151668310165405,
      "learning_rate": 9.804486350478013e-05,
      "loss": 1.4959,
      "step": 48920
    },
    {
      "epoch": 2.212982878046302,
      "grad_norm": 0.24722276628017426,
      "learning_rate": 9.804024305552451e-05,
      "loss": 1.486,
      "step": 48930
    },
    {
      "epoch": 2.2135248192118144,
      "grad_norm": 0.33431771397590637,
      "learning_rate": 9.803561727463042e-05,
      "loss": 1.4907,
      "step": 48940
    },
    {
      "epoch": 2.2140667603773263,
      "grad_norm": 0.20495231449604034,
      "learning_rate": 9.803098616267093e-05,
      "loss": 1.5058,
      "step": 48950
    },
    {
      "epoch": 2.2146087015428386,
      "grad_norm": 0.2133825570344925,
      "learning_rate": 9.802634972021967e-05,
      "loss": 1.4913,
      "step": 48960
    },
    {
      "epoch": 2.215150642708351,
      "grad_norm": 0.32575109601020813,
      "learning_rate": 9.802170794785103e-05,
      "loss": 1.4912,
      "step": 48970
    },
    {
      "epoch": 2.2156925838738633,
      "grad_norm": 0.24184833467006683,
      "learning_rate": 9.801706084613997e-05,
      "loss": 1.4915,
      "step": 48980
    },
    {
      "epoch": 2.2157467779904145,
      "eval_loss": 2.473459243774414,
      "eval_runtime": 22.0863,
      "eval_samples_per_second": 226.385,
      "eval_steps_per_second": 1.222,
      "step": 48981
    },
    {
      "epoch": 2.2162345250393756,
      "grad_norm": 0.281916081905365,
      "learning_rate": 9.801240841566221e-05,
      "loss": 1.4939,
      "step": 48990
    },
    {
      "epoch": 2.2167764662048874,
      "grad_norm": 0.24919818341732025,
      "learning_rate": 9.8007750656994e-05,
      "loss": 1.5073,
      "step": 49000
    },
    {
      "epoch": 2.2173184073703998,
      "grad_norm": 0.3815664052963257,
      "learning_rate": 9.800308757071238e-05,
      "loss": 1.4974,
      "step": 49010
    },
    {
      "epoch": 2.217860348535912,
      "grad_norm": 0.21346397697925568,
      "learning_rate": 9.799841915739496e-05,
      "loss": 1.49,
      "step": 49020
    },
    {
      "epoch": 2.2184022897014244,
      "grad_norm": 0.2835428714752197,
      "learning_rate": 9.799374541762005e-05,
      "loss": 1.4981,
      "step": 49030
    },
    {
      "epoch": 2.2189442308669367,
      "grad_norm": 0.20683638751506805,
      "learning_rate": 9.798906635196665e-05,
      "loss": 1.5014,
      "step": 49040
    },
    {
      "epoch": 2.2194861720324486,
      "grad_norm": 0.28130945563316345,
      "learning_rate": 9.798438196101432e-05,
      "loss": 1.4794,
      "step": 49050
    },
    {
      "epoch": 2.220028113197961,
      "grad_norm": 0.6789763569831848,
      "learning_rate": 9.797969224534338e-05,
      "loss": 1.504,
      "step": 49060
    },
    {
      "epoch": 2.2204616661303707,
      "eval_loss": 2.4563019275665283,
      "eval_runtime": 22.0507,
      "eval_samples_per_second": 226.75,
      "eval_steps_per_second": 1.224,
      "step": 49068
    },
    {
      "epoch": 2.220570054363473,
      "grad_norm": 0.5758925676345825,
      "learning_rate": 9.797499720553476e-05,
      "loss": 1.4929,
      "step": 49070
    },
    {
      "epoch": 2.2211119955289855,
      "grad_norm": 0.26593759655952454,
      "learning_rate": 9.797029684217008e-05,
      "loss": 1.4907,
      "step": 49080
    },
    {
      "epoch": 2.2216539366944974,
      "grad_norm": 0.2138567566871643,
      "learning_rate": 9.796559115583158e-05,
      "loss": 1.4922,
      "step": 49090
    },
    {
      "epoch": 2.2221958778600097,
      "grad_norm": 0.31835222244262695,
      "learning_rate": 9.796088014710218e-05,
      "loss": 1.4927,
      "step": 49100
    },
    {
      "epoch": 2.222737819025522,
      "grad_norm": 0.19409584999084473,
      "learning_rate": 9.795616381656546e-05,
      "loss": 1.4954,
      "step": 49110
    },
    {
      "epoch": 2.2232797601910343,
      "grad_norm": 0.273907870054245,
      "learning_rate": 9.795144216480566e-05,
      "loss": 1.4999,
      "step": 49120
    },
    {
      "epoch": 2.2238217013565467,
      "grad_norm": 0.19406692683696747,
      "learning_rate": 9.794671519240768e-05,
      "loss": 1.4948,
      "step": 49130
    },
    {
      "epoch": 2.2243636425220585,
      "grad_norm": 0.262638121843338,
      "learning_rate": 9.79419828999571e-05,
      "loss": 1.4956,
      "step": 49140
    },
    {
      "epoch": 2.224905583687571,
      "grad_norm": 0.26331815123558044,
      "learning_rate": 9.79372452880401e-05,
      "loss": 1.4954,
      "step": 49150
    },
    {
      "epoch": 2.2251765542703272,
      "eval_loss": 2.458817958831787,
      "eval_runtime": 22.1352,
      "eval_samples_per_second": 225.885,
      "eval_steps_per_second": 1.22,
      "step": 49155
    },
    {
      "epoch": 2.225447524853083,
      "grad_norm": 0.35624048113822937,
      "learning_rate": 9.793250235724358e-05,
      "loss": 1.4944,
      "step": 49160
    },
    {
      "epoch": 2.2259894660185955,
      "grad_norm": 0.24932223558425903,
      "learning_rate": 9.792775410815506e-05,
      "loss": 1.4867,
      "step": 49170
    },
    {
      "epoch": 2.2265314071841074,
      "grad_norm": 0.2496650665998459,
      "learning_rate": 9.792300054136272e-05,
      "loss": 1.4798,
      "step": 49180
    },
    {
      "epoch": 2.2270733483496197,
      "grad_norm": 0.29679879546165466,
      "learning_rate": 9.791824165745543e-05,
      "loss": 1.493,
      "step": 49190
    },
    {
      "epoch": 2.227615289515132,
      "grad_norm": 0.3611978590488434,
      "learning_rate": 9.79134774570227e-05,
      "loss": 1.4927,
      "step": 49200
    },
    {
      "epoch": 2.2281572306806443,
      "grad_norm": 0.289468377828598,
      "learning_rate": 9.790870794065469e-05,
      "loss": 1.5052,
      "step": 49210
    },
    {
      "epoch": 2.2286991718461566,
      "grad_norm": 0.31869956851005554,
      "learning_rate": 9.790393310894225e-05,
      "loss": 1.5031,
      "step": 49220
    },
    {
      "epoch": 2.2292411130116685,
      "grad_norm": 0.1822601705789566,
      "learning_rate": 9.789915296247682e-05,
      "loss": 1.4961,
      "step": 49230
    },
    {
      "epoch": 2.229783054177181,
      "grad_norm": 0.19582431018352509,
      "learning_rate": 9.789436750185059e-05,
      "loss": 1.4834,
      "step": 49240
    },
    {
      "epoch": 2.2298914424102834,
      "eval_loss": 2.4611740112304688,
      "eval_runtime": 23.4689,
      "eval_samples_per_second": 213.048,
      "eval_steps_per_second": 1.15,
      "step": 49242
    },
    {
      "epoch": 2.230324995342693,
      "grad_norm": 0.3001343607902527,
      "learning_rate": 9.788957672765634e-05,
      "loss": 1.4975,
      "step": 49250
    },
    {
      "epoch": 2.2308669365082054,
      "grad_norm": 0.32307741045951843,
      "learning_rate": 9.788478064048753e-05,
      "loss": 1.4875,
      "step": 49260
    },
    {
      "epoch": 2.2314088776737178,
      "grad_norm": 0.21440020203590393,
      "learning_rate": 9.787997924093829e-05,
      "loss": 1.5025,
      "step": 49270
    },
    {
      "epoch": 2.2319508188392296,
      "grad_norm": 0.21042481064796448,
      "learning_rate": 9.787517252960335e-05,
      "loss": 1.5006,
      "step": 49280
    },
    {
      "epoch": 2.232492760004742,
      "grad_norm": 0.22425493597984314,
      "learning_rate": 9.787036050707822e-05,
      "loss": 1.4948,
      "step": 49290
    },
    {
      "epoch": 2.2330347011702543,
      "grad_norm": 0.24447576701641083,
      "learning_rate": 9.786554317395894e-05,
      "loss": 1.4939,
      "step": 49300
    },
    {
      "epoch": 2.2335766423357666,
      "grad_norm": 0.3083951771259308,
      "learning_rate": 9.786072053084228e-05,
      "loss": 1.4949,
      "step": 49310
    },
    {
      "epoch": 2.2341185835012785,
      "grad_norm": 0.24689245223999023,
      "learning_rate": 9.785589257832565e-05,
      "loss": 1.489,
      "step": 49320
    },
    {
      "epoch": 2.2346063305502395,
      "eval_loss": 2.45898175239563,
      "eval_runtime": 22.1974,
      "eval_samples_per_second": 225.252,
      "eval_steps_per_second": 1.216,
      "step": 49329
    },
    {
      "epoch": 2.2346605246667908,
      "grad_norm": 0.3437364995479584,
      "learning_rate": 9.78510593170071e-05,
      "loss": 1.4934,
      "step": 49330
    },
    {
      "epoch": 2.235202465832303,
      "grad_norm": 0.34142521023750305,
      "learning_rate": 9.784622074748537e-05,
      "loss": 1.4972,
      "step": 49340
    },
    {
      "epoch": 2.2357444069978154,
      "grad_norm": 0.3373830020427704,
      "learning_rate": 9.784137687035983e-05,
      "loss": 1.4975,
      "step": 49350
    },
    {
      "epoch": 2.2362863481633277,
      "grad_norm": 0.2561146318912506,
      "learning_rate": 9.78365276862305e-05,
      "loss": 1.4934,
      "step": 49360
    },
    {
      "epoch": 2.2368282893288396,
      "grad_norm": 0.19827355444431305,
      "learning_rate": 9.783167319569813e-05,
      "loss": 1.4969,
      "step": 49370
    },
    {
      "epoch": 2.237370230494352,
      "grad_norm": 0.2824994921684265,
      "learning_rate": 9.782681339936404e-05,
      "loss": 1.4843,
      "step": 49380
    },
    {
      "epoch": 2.2379121716598642,
      "grad_norm": 0.21020928025245667,
      "learning_rate": 9.782194829783022e-05,
      "loss": 1.4916,
      "step": 49390
    },
    {
      "epoch": 2.2384541128253765,
      "grad_norm": 0.30868810415267944,
      "learning_rate": 9.781707789169937e-05,
      "loss": 1.4969,
      "step": 49400
    },
    {
      "epoch": 2.2389960539908884,
      "grad_norm": 0.41937023401260376,
      "learning_rate": 9.781220218157479e-05,
      "loss": 1.4893,
      "step": 49410
    },
    {
      "epoch": 2.239321218690196,
      "eval_loss": 2.4543631076812744,
      "eval_runtime": 22.0555,
      "eval_samples_per_second": 226.7,
      "eval_steps_per_second": 1.224,
      "step": 49416
    },
    {
      "epoch": 2.2395379951564007,
      "grad_norm": 0.6382551789283752,
      "learning_rate": 9.780732116806052e-05,
      "loss": 1.5002,
      "step": 49420
    },
    {
      "epoch": 2.240079936321913,
      "grad_norm": 0.2832179069519043,
      "learning_rate": 9.780243485176111e-05,
      "loss": 1.4967,
      "step": 49430
    },
    {
      "epoch": 2.2406218774874254,
      "grad_norm": 0.19235193729400635,
      "learning_rate": 9.779754323328192e-05,
      "loss": 1.4979,
      "step": 49440
    },
    {
      "epoch": 2.2411638186529377,
      "grad_norm": 0.26842355728149414,
      "learning_rate": 9.779264631322888e-05,
      "loss": 1.4917,
      "step": 49450
    },
    {
      "epoch": 2.2417057598184496,
      "grad_norm": 0.28281205892562866,
      "learning_rate": 9.778774409220859e-05,
      "loss": 1.4907,
      "step": 49460
    },
    {
      "epoch": 2.242247700983962,
      "grad_norm": 0.3516809344291687,
      "learning_rate": 9.778283657082832e-05,
      "loss": 1.4887,
      "step": 49470
    },
    {
      "epoch": 2.242789642149474,
      "grad_norm": 0.23774586617946625,
      "learning_rate": 9.777792374969603e-05,
      "loss": 1.4881,
      "step": 49480
    },
    {
      "epoch": 2.2433315833149865,
      "grad_norm": 0.32771846652030945,
      "learning_rate": 9.777300562942027e-05,
      "loss": 1.4883,
      "step": 49490
    },
    {
      "epoch": 2.243873524480499,
      "grad_norm": 0.25199854373931885,
      "learning_rate": 9.776808221061024e-05,
      "loss": 1.4944,
      "step": 49500
    },
    {
      "epoch": 2.244036106830152,
      "eval_loss": 2.4585046768188477,
      "eval_runtime": 22.068,
      "eval_samples_per_second": 226.572,
      "eval_steps_per_second": 1.223,
      "step": 49503
    },
    {
      "epoch": 2.2444154656460107,
      "grad_norm": 0.26867803931236267,
      "learning_rate": 9.776315349387589e-05,
      "loss": 1.4877,
      "step": 49510
    },
    {
      "epoch": 2.244957406811523,
      "grad_norm": 0.2837635576725006,
      "learning_rate": 9.775821947982775e-05,
      "loss": 1.4914,
      "step": 49520
    },
    {
      "epoch": 2.2454993479770353,
      "grad_norm": 0.24937689304351807,
      "learning_rate": 9.775328016907701e-05,
      "loss": 1.4903,
      "step": 49530
    },
    {
      "epoch": 2.2460412891425476,
      "grad_norm": 0.20561334490776062,
      "learning_rate": 9.774833556223554e-05,
      "loss": 1.4971,
      "step": 49540
    },
    {
      "epoch": 2.2465832303080595,
      "grad_norm": 0.2335055023431778,
      "learning_rate": 9.774338565991586e-05,
      "loss": 1.4979,
      "step": 49550
    },
    {
      "epoch": 2.247125171473572,
      "grad_norm": 0.3294164836406708,
      "learning_rate": 9.773843046273111e-05,
      "loss": 1.4996,
      "step": 49560
    },
    {
      "epoch": 2.247667112639084,
      "grad_norm": 0.3552034795284271,
      "learning_rate": 9.773346997129518e-05,
      "loss": 1.5011,
      "step": 49570
    },
    {
      "epoch": 2.2482090538045965,
      "grad_norm": 0.36698946356773376,
      "learning_rate": 9.77285041862225e-05,
      "loss": 1.5012,
      "step": 49580
    },
    {
      "epoch": 2.2487509949701083,
      "grad_norm": 0.2022208720445633,
      "learning_rate": 9.772353310812824e-05,
      "loss": 1.5007,
      "step": 49590
    },
    {
      "epoch": 2.2487509949701083,
      "eval_loss": 2.4587907791137695,
      "eval_runtime": 22.0232,
      "eval_samples_per_second": 227.033,
      "eval_steps_per_second": 1.226,
      "step": 49590
    },
    {
      "epoch": 2.2492929361356206,
      "grad_norm": 0.3783475160598755,
      "learning_rate": 9.771855673762818e-05,
      "loss": 1.4963,
      "step": 49600
    },
    {
      "epoch": 2.249834877301133,
      "grad_norm": 0.4129343628883362,
      "learning_rate": 9.771357507533878e-05,
      "loss": 1.4945,
      "step": 49610
    },
    {
      "epoch": 2.2503768184666453,
      "grad_norm": 0.2977985739707947,
      "learning_rate": 9.770858812187715e-05,
      "loss": 1.4925,
      "step": 49620
    },
    {
      "epoch": 2.2509187596321576,
      "grad_norm": 0.22732873260974884,
      "learning_rate": 9.770359587786105e-05,
      "loss": 1.4973,
      "step": 49630
    },
    {
      "epoch": 2.2514607007976695,
      "grad_norm": 0.27571314573287964,
      "learning_rate": 9.769859834390887e-05,
      "loss": 1.4836,
      "step": 49640
    },
    {
      "epoch": 2.252002641963182,
      "grad_norm": 0.19114813208580017,
      "learning_rate": 9.769359552063972e-05,
      "loss": 1.4889,
      "step": 49650
    },
    {
      "epoch": 2.252544583128694,
      "grad_norm": 0.193264439702034,
      "learning_rate": 9.76885874086733e-05,
      "loss": 1.5037,
      "step": 49660
    },
    {
      "epoch": 2.2530865242942064,
      "grad_norm": 0.2822969853878021,
      "learning_rate": 9.768357400863004e-05,
      "loss": 1.4948,
      "step": 49670
    },
    {
      "epoch": 2.253465883110065,
      "eval_loss": 2.4479169845581055,
      "eval_runtime": 21.9892,
      "eval_samples_per_second": 227.384,
      "eval_steps_per_second": 1.228,
      "step": 49677
    },
    {
      "epoch": 2.2536284654597187,
      "grad_norm": 0.2823607325553894,
      "learning_rate": 9.767855532113091e-05,
      "loss": 1.4953,
      "step": 49680
    },
    {
      "epoch": 2.2541704066252306,
      "grad_norm": 0.29018524289131165,
      "learning_rate": 9.767353134679765e-05,
      "loss": 1.4909,
      "step": 49690
    },
    {
      "epoch": 2.254712347790743,
      "grad_norm": 0.2224309891462326,
      "learning_rate": 9.76685020862526e-05,
      "loss": 1.4952,
      "step": 49700
    },
    {
      "epoch": 2.2552542889562552,
      "grad_norm": 0.25766855478286743,
      "learning_rate": 9.766346754011875e-05,
      "loss": 1.4953,
      "step": 49710
    },
    {
      "epoch": 2.2557962301217676,
      "grad_norm": 0.2575187385082245,
      "learning_rate": 9.765842770901979e-05,
      "loss": 1.4876,
      "step": 49720
    },
    {
      "epoch": 2.25633817128728,
      "grad_norm": 0.3350069522857666,
      "learning_rate": 9.765338259358e-05,
      "loss": 1.4925,
      "step": 49730
    },
    {
      "epoch": 2.2568801124527917,
      "grad_norm": 0.21751070022583008,
      "learning_rate": 9.764833219442435e-05,
      "loss": 1.4812,
      "step": 49740
    },
    {
      "epoch": 2.257422053618304,
      "grad_norm": 0.19443929195404053,
      "learning_rate": 9.764327651217847e-05,
      "loss": 1.4924,
      "step": 49750
    },
    {
      "epoch": 2.2579639947838164,
      "grad_norm": 0.21549266576766968,
      "learning_rate": 9.763821554746864e-05,
      "loss": 1.484,
      "step": 49760
    },
    {
      "epoch": 2.258180771250021,
      "eval_loss": 2.451836585998535,
      "eval_runtime": 21.9854,
      "eval_samples_per_second": 227.423,
      "eval_steps_per_second": 1.228,
      "step": 49764
    },
    {
      "epoch": 2.2585059359493287,
      "grad_norm": 0.30445384979248047,
      "learning_rate": 9.763314930092178e-05,
      "loss": 1.4924,
      "step": 49770
    },
    {
      "epoch": 2.2590478771148406,
      "grad_norm": 0.3584153950214386,
      "learning_rate": 9.762807777316548e-05,
      "loss": 1.4855,
      "step": 49780
    },
    {
      "epoch": 2.259589818280353,
      "grad_norm": 0.2762848138809204,
      "learning_rate": 9.762300096482799e-05,
      "loss": 1.5014,
      "step": 49790
    },
    {
      "epoch": 2.260131759445865,
      "grad_norm": 0.22628115117549896,
      "learning_rate": 9.761791887653817e-05,
      "loss": 1.4932,
      "step": 49800
    },
    {
      "epoch": 2.2606737006113775,
      "grad_norm": 0.3497247099876404,
      "learning_rate": 9.761283150892561e-05,
      "loss": 1.5017,
      "step": 49810
    },
    {
      "epoch": 2.2612156417768894,
      "grad_norm": 0.40231335163116455,
      "learning_rate": 9.760773886262046e-05,
      "loss": 1.4924,
      "step": 49820
    },
    {
      "epoch": 2.2617575829424017,
      "grad_norm": 0.2883796989917755,
      "learning_rate": 9.760264093825361e-05,
      "loss": 1.4829,
      "step": 49830
    },
    {
      "epoch": 2.262299524107914,
      "grad_norm": 0.4345490038394928,
      "learning_rate": 9.759753773645658e-05,
      "loss": 1.4926,
      "step": 49840
    },
    {
      "epoch": 2.2628414652734263,
      "grad_norm": 0.2081182599067688,
      "learning_rate": 9.759242925786151e-05,
      "loss": 1.5033,
      "step": 49850
    },
    {
      "epoch": 2.2628956593899776,
      "eval_loss": 2.4485983848571777,
      "eval_runtime": 22.0488,
      "eval_samples_per_second": 226.769,
      "eval_steps_per_second": 1.225,
      "step": 49851
    },
    {
      "epoch": 2.2633834064389386,
      "grad_norm": 0.29944297671318054,
      "learning_rate": 9.758731550310122e-05,
      "loss": 1.4909,
      "step": 49860
    },
    {
      "epoch": 2.2639253476044505,
      "grad_norm": 0.37277737259864807,
      "learning_rate": 9.758219647280919e-05,
      "loss": 1.4931,
      "step": 49870
    },
    {
      "epoch": 2.264467288769963,
      "grad_norm": 0.2762742340564728,
      "learning_rate": 9.757707216761953e-05,
      "loss": 1.4982,
      "step": 49880
    },
    {
      "epoch": 2.265009229935475,
      "grad_norm": 0.22631752490997314,
      "learning_rate": 9.7571942588167e-05,
      "loss": 1.4871,
      "step": 49890
    },
    {
      "epoch": 2.2655511711009875,
      "grad_norm": 0.22785073518753052,
      "learning_rate": 9.756680773508708e-05,
      "loss": 1.4866,
      "step": 49900
    },
    {
      "epoch": 2.2660931122665,
      "grad_norm": 0.3055456876754761,
      "learning_rate": 9.756166760901581e-05,
      "loss": 1.4829,
      "step": 49910
    },
    {
      "epoch": 2.2666350534320117,
      "grad_norm": 0.25249767303466797,
      "learning_rate": 9.755652221058993e-05,
      "loss": 1.501,
      "step": 49920
    },
    {
      "epoch": 2.267176994597524,
      "grad_norm": 0.25835105776786804,
      "learning_rate": 9.755137154044686e-05,
      "loss": 1.4893,
      "step": 49930
    },
    {
      "epoch": 2.2676105475299337,
      "eval_loss": 2.449587345123291,
      "eval_runtime": 22.013,
      "eval_samples_per_second": 227.138,
      "eval_steps_per_second": 1.227,
      "step": 49938
    },
    {
      "epoch": 2.2677189357630363,
      "grad_norm": 0.3590300679206848,
      "learning_rate": 9.754621559922461e-05,
      "loss": 1.4947,
      "step": 49940
    },
    {
      "epoch": 2.2682608769285486,
      "grad_norm": 0.24853801727294922,
      "learning_rate": 9.75410543875619e-05,
      "loss": 1.4931,
      "step": 49950
    },
    {
      "epoch": 2.2688028180940605,
      "grad_norm": 0.19558899104595184,
      "learning_rate": 9.753588790609807e-05,
      "loss": 1.4906,
      "step": 49960
    },
    {
      "epoch": 2.269344759259573,
      "grad_norm": 0.2478579878807068,
      "learning_rate": 9.75307161554731e-05,
      "loss": 1.4903,
      "step": 49970
    },
    {
      "epoch": 2.269886700425085,
      "grad_norm": 0.32022833824157715,
      "learning_rate": 9.752553913632768e-05,
      "loss": 1.4904,
      "step": 49980
    },
    {
      "epoch": 2.2704286415905974,
      "grad_norm": 0.22632542252540588,
      "learning_rate": 9.752035684930309e-05,
      "loss": 1.4888,
      "step": 49990
    },
    {
      "epoch": 2.2709705827561093,
      "grad_norm": 0.29843834042549133,
      "learning_rate": 9.75151692950413e-05,
      "loss": 1.4982,
      "step": 50000
    },
    {
      "epoch": 2.2715125239216216,
      "grad_norm": 0.24948932230472565,
      "learning_rate": 9.750997647418492e-05,
      "loss": 1.4907,
      "step": 50010
    },
    {
      "epoch": 2.272054465087134,
      "grad_norm": 0.26221540570259094,
      "learning_rate": 9.750477838737721e-05,
      "loss": 1.4931,
      "step": 50020
    },
    {
      "epoch": 2.27232543566989,
      "eval_loss": 2.4510858058929443,
      "eval_runtime": 22.1311,
      "eval_samples_per_second": 225.927,
      "eval_steps_per_second": 1.22,
      "step": 50025
    },
    {
      "epoch": 2.2725964062526463,
      "grad_norm": 0.4738941192626953,
      "learning_rate": 9.74995750352621e-05,
      "loss": 1.5008,
      "step": 50030
    },
    {
      "epoch": 2.2731383474181586,
      "grad_norm": 0.29958027601242065,
      "learning_rate": 9.749436641848415e-05,
      "loss": 1.4866,
      "step": 50040
    },
    {
      "epoch": 2.2736802885836704,
      "grad_norm": 0.21620671451091766,
      "learning_rate": 9.748915253768856e-05,
      "loss": 1.4908,
      "step": 50050
    },
    {
      "epoch": 2.2742222297491828,
      "grad_norm": 0.21653667092323303,
      "learning_rate": 9.748393339352125e-05,
      "loss": 1.4871,
      "step": 50060
    },
    {
      "epoch": 2.274764170914695,
      "grad_norm": 0.2310364544391632,
      "learning_rate": 9.747870898662871e-05,
      "loss": 1.4929,
      "step": 50070
    },
    {
      "epoch": 2.2753061120802074,
      "grad_norm": 0.23432140052318573,
      "learning_rate": 9.747347931765812e-05,
      "loss": 1.4955,
      "step": 50080
    },
    {
      "epoch": 2.2758480532457197,
      "grad_norm": 0.22089755535125732,
      "learning_rate": 9.746824438725731e-05,
      "loss": 1.489,
      "step": 50090
    },
    {
      "epoch": 2.2763899944112316,
      "grad_norm": 0.3632870018482208,
      "learning_rate": 9.746300419607479e-05,
      "loss": 1.4839,
      "step": 50100
    },
    {
      "epoch": 2.276931935576744,
      "grad_norm": 0.18700088560581207,
      "learning_rate": 9.745775874475963e-05,
      "loss": 1.4952,
      "step": 50110
    },
    {
      "epoch": 2.2770403238098464,
      "eval_loss": 2.4573004245758057,
      "eval_runtime": 22.0595,
      "eval_samples_per_second": 226.66,
      "eval_steps_per_second": 1.224,
      "step": 50112
    },
    {
      "epoch": 2.277473876742256,
      "grad_norm": 0.23492306470870972,
      "learning_rate": 9.745250803396166e-05,
      "loss": 1.4832,
      "step": 50120
    },
    {
      "epoch": 2.2780158179077685,
      "grad_norm": 0.35017403960227966,
      "learning_rate": 9.744725206433131e-05,
      "loss": 1.4903,
      "step": 50130
    },
    {
      "epoch": 2.278557759073281,
      "grad_norm": 0.2380804866552353,
      "learning_rate": 9.744199083651968e-05,
      "loss": 1.492,
      "step": 50140
    },
    {
      "epoch": 2.2790997002387927,
      "grad_norm": 0.2782959043979645,
      "learning_rate": 9.743672435117846e-05,
      "loss": 1.4848,
      "step": 50150
    },
    {
      "epoch": 2.279641641404305,
      "grad_norm": 0.19856207072734833,
      "learning_rate": 9.743145260896009e-05,
      "loss": 1.4947,
      "step": 50160
    },
    {
      "epoch": 2.2801835825698173,
      "grad_norm": 0.3704603612422943,
      "learning_rate": 9.742617561051758e-05,
      "loss": 1.4797,
      "step": 50170
    },
    {
      "epoch": 2.2807255237353297,
      "grad_norm": 0.38659003376960754,
      "learning_rate": 9.742089335650462e-05,
      "loss": 1.4944,
      "step": 50180
    },
    {
      "epoch": 2.2812674649008415,
      "grad_norm": 0.2731989324092865,
      "learning_rate": 9.741560584757559e-05,
      "loss": 1.4955,
      "step": 50190
    },
    {
      "epoch": 2.2817552119498026,
      "eval_loss": 2.4447662830352783,
      "eval_runtime": 22.0833,
      "eval_samples_per_second": 226.415,
      "eval_steps_per_second": 1.223,
      "step": 50199
    },
    {
      "epoch": 2.281809406066354,
      "grad_norm": 0.312407523393631,
      "learning_rate": 9.741031308438543e-05,
      "loss": 1.4969,
      "step": 50200
    },
    {
      "epoch": 2.282351347231866,
      "grad_norm": 0.2855091691017151,
      "learning_rate": 9.740501506758983e-05,
      "loss": 1.4872,
      "step": 50210
    },
    {
      "epoch": 2.2828932883973785,
      "grad_norm": 0.33906736969947815,
      "learning_rate": 9.739971179784508e-05,
      "loss": 1.4888,
      "step": 50220
    },
    {
      "epoch": 2.2834352295628904,
      "grad_norm": 0.21002908051013947,
      "learning_rate": 9.739440327580809e-05,
      "loss": 1.49,
      "step": 50230
    },
    {
      "epoch": 2.2839771707284027,
      "grad_norm": 0.24895620346069336,
      "learning_rate": 9.738908950213648e-05,
      "loss": 1.4906,
      "step": 50240
    },
    {
      "epoch": 2.284519111893915,
      "grad_norm": 0.2253609001636505,
      "learning_rate": 9.738377047748852e-05,
      "loss": 1.5001,
      "step": 50250
    },
    {
      "epoch": 2.2850610530594273,
      "grad_norm": 0.30024558305740356,
      "learning_rate": 9.737844620252307e-05,
      "loss": 1.4887,
      "step": 50260
    },
    {
      "epoch": 2.2856029942249396,
      "grad_norm": 0.3111526370048523,
      "learning_rate": 9.737311667789967e-05,
      "loss": 1.4861,
      "step": 50270
    },
    {
      "epoch": 2.2861449353904515,
      "grad_norm": 0.22137415409088135,
      "learning_rate": 9.736778190427859e-05,
      "loss": 1.4972,
      "step": 50280
    },
    {
      "epoch": 2.286470100089759,
      "eval_loss": 2.4479713439941406,
      "eval_runtime": 21.9852,
      "eval_samples_per_second": 227.425,
      "eval_steps_per_second": 1.228,
      "step": 50286
    },
    {
      "epoch": 2.286686876555964,
      "grad_norm": 0.18520957231521606,
      "learning_rate": 9.73624418823206e-05,
      "loss": 1.4849,
      "step": 50290
    },
    {
      "epoch": 2.287228817721476,
      "grad_norm": 0.30916085839271545,
      "learning_rate": 9.735709661268723e-05,
      "loss": 1.4852,
      "step": 50300
    },
    {
      "epoch": 2.2877707588869884,
      "grad_norm": 0.5676615238189697,
      "learning_rate": 9.735174609604063e-05,
      "loss": 1.4893,
      "step": 50310
    },
    {
      "epoch": 2.2883127000525008,
      "grad_norm": 0.40237462520599365,
      "learning_rate": 9.734639033304361e-05,
      "loss": 1.4901,
      "step": 50320
    },
    {
      "epoch": 2.2888546412180126,
      "grad_norm": 0.24099163711071014,
      "learning_rate": 9.734102932435959e-05,
      "loss": 1.4842,
      "step": 50330
    },
    {
      "epoch": 2.289396582383525,
      "grad_norm": 0.2044031023979187,
      "learning_rate": 9.73356630706527e-05,
      "loss": 1.4834,
      "step": 50340
    },
    {
      "epoch": 2.2899385235490373,
      "grad_norm": 0.22731153666973114,
      "learning_rate": 9.733029157258765e-05,
      "loss": 1.471,
      "step": 50350
    },
    {
      "epoch": 2.2904804647145496,
      "grad_norm": 0.19987347722053528,
      "learning_rate": 9.732491483082987e-05,
      "loss": 1.4786,
      "step": 50360
    },
    {
      "epoch": 2.2910224058800615,
      "grad_norm": 0.23623400926589966,
      "learning_rate": 9.73195328460454e-05,
      "loss": 1.4864,
      "step": 50370
    },
    {
      "epoch": 2.2911849882297153,
      "eval_loss": 2.4462499618530273,
      "eval_runtime": 22.0809,
      "eval_samples_per_second": 226.44,
      "eval_steps_per_second": 1.223,
      "step": 50373
    },
    {
      "epoch": 2.2915643470455738,
      "grad_norm": 0.2596568167209625,
      "learning_rate": 9.731414561890093e-05,
      "loss": 1.4828,
      "step": 50380
    },
    {
      "epoch": 2.292106288211086,
      "grad_norm": 0.22793002426624298,
      "learning_rate": 9.730875315006381e-05,
      "loss": 1.4874,
      "step": 50390
    },
    {
      "epoch": 2.2926482293765984,
      "grad_norm": 0.34478330612182617,
      "learning_rate": 9.730335544020204e-05,
      "loss": 1.494,
      "step": 50400
    },
    {
      "epoch": 2.2931901705421103,
      "grad_norm": 0.3564873933792114,
      "learning_rate": 9.729795248998425e-05,
      "loss": 1.4937,
      "step": 50410
    },
    {
      "epoch": 2.2937321117076226,
      "grad_norm": 0.21651984751224518,
      "learning_rate": 9.729254430007976e-05,
      "loss": 1.4791,
      "step": 50420
    },
    {
      "epoch": 2.294274052873135,
      "grad_norm": 0.25641965866088867,
      "learning_rate": 9.728713087115848e-05,
      "loss": 1.4991,
      "step": 50430
    },
    {
      "epoch": 2.2948159940386472,
      "grad_norm": 0.3314838707447052,
      "learning_rate": 9.728171220389104e-05,
      "loss": 1.5014,
      "step": 50440
    },
    {
      "epoch": 2.2953579352041595,
      "grad_norm": 0.18577730655670166,
      "learning_rate": 9.727628829894866e-05,
      "loss": 1.4913,
      "step": 50450
    },
    {
      "epoch": 2.2958998763696714,
      "grad_norm": 0.3247391879558563,
      "learning_rate": 9.727085915700321e-05,
      "loss": 1.4837,
      "step": 50460
    },
    {
      "epoch": 2.2958998763696714,
      "eval_loss": 2.445127010345459,
      "eval_runtime": 22.0909,
      "eval_samples_per_second": 226.337,
      "eval_steps_per_second": 1.222,
      "step": 50460
    },
    {
      "epoch": 2.2964418175351837,
      "grad_norm": 0.2847256660461426,
      "learning_rate": 9.726542477872726e-05,
      "loss": 1.4776,
      "step": 50470
    },
    {
      "epoch": 2.296983758700696,
      "grad_norm": 0.23234505951404572,
      "learning_rate": 9.725998516479399e-05,
      "loss": 1.4889,
      "step": 50480
    },
    {
      "epoch": 2.2975256998662084,
      "grad_norm": 0.20872457325458527,
      "learning_rate": 9.725454031587725e-05,
      "loss": 1.4975,
      "step": 50490
    },
    {
      "epoch": 2.2980676410317207,
      "grad_norm": 0.24152375757694244,
      "learning_rate": 9.724909023265147e-05,
      "loss": 1.4884,
      "step": 50500
    },
    {
      "epoch": 2.2986095821972325,
      "grad_norm": 0.45855534076690674,
      "learning_rate": 9.724363491579185e-05,
      "loss": 1.4975,
      "step": 50510
    },
    {
      "epoch": 2.299151523362745,
      "grad_norm": 0.37530308961868286,
      "learning_rate": 9.723817436597413e-05,
      "loss": 1.4934,
      "step": 50520
    },
    {
      "epoch": 2.299693464528257,
      "grad_norm": 0.3856462240219116,
      "learning_rate": 9.723270858387474e-05,
      "loss": 1.4977,
      "step": 50530
    },
    {
      "epoch": 2.3002354056937695,
      "grad_norm": 0.21550019085407257,
      "learning_rate": 9.722723757017078e-05,
      "loss": 1.5,
      "step": 50540
    },
    {
      "epoch": 2.300614764509628,
      "eval_loss": 2.45145320892334,
      "eval_runtime": 22.1166,
      "eval_samples_per_second": 226.074,
      "eval_steps_per_second": 1.221,
      "step": 50547
    },
    {
      "epoch": 2.300777346859282,
      "grad_norm": 0.2878018617630005,
      "learning_rate": 9.722176132553995e-05,
      "loss": 1.498,
      "step": 50550
    },
    {
      "epoch": 2.3013192880247937,
      "grad_norm": 0.3644629418849945,
      "learning_rate": 9.721627985066064e-05,
      "loss": 1.4972,
      "step": 50560
    },
    {
      "epoch": 2.301861229190306,
      "grad_norm": 0.22482427954673767,
      "learning_rate": 9.721079314621186e-05,
      "loss": 1.4867,
      "step": 50570
    },
    {
      "epoch": 2.3024031703558183,
      "grad_norm": 0.3207145631313324,
      "learning_rate": 9.72053012128733e-05,
      "loss": 1.481,
      "step": 50580
    },
    {
      "epoch": 2.3029451115213306,
      "grad_norm": 0.26092955470085144,
      "learning_rate": 9.719980405132527e-05,
      "loss": 1.4841,
      "step": 50590
    },
    {
      "epoch": 2.3034870526868425,
      "grad_norm": 0.30404138565063477,
      "learning_rate": 9.71943016622487e-05,
      "loss": 1.4924,
      "step": 50600
    },
    {
      "epoch": 2.304028993852355,
      "grad_norm": 0.23371928930282593,
      "learning_rate": 9.718879404632525e-05,
      "loss": 1.4786,
      "step": 50610
    },
    {
      "epoch": 2.304570935017867,
      "grad_norm": 0.2730085551738739,
      "learning_rate": 9.718328120423715e-05,
      "loss": 1.4862,
      "step": 50620
    },
    {
      "epoch": 2.3051128761833795,
      "grad_norm": 0.19008232653141022,
      "learning_rate": 9.717776313666731e-05,
      "loss": 1.4935,
      "step": 50630
    },
    {
      "epoch": 2.305329652649584,
      "eval_loss": 2.4463045597076416,
      "eval_runtime": 22.0762,
      "eval_samples_per_second": 226.488,
      "eval_steps_per_second": 1.223,
      "step": 50634
    },
    {
      "epoch": 2.3056548173488913,
      "grad_norm": 0.30709657073020935,
      "learning_rate": 9.717223984429931e-05,
      "loss": 1.4934,
      "step": 50640
    },
    {
      "epoch": 2.3061967585144036,
      "grad_norm": 0.3597027361392975,
      "learning_rate": 9.716671132781731e-05,
      "loss": 1.488,
      "step": 50650
    },
    {
      "epoch": 2.306738699679916,
      "grad_norm": 0.2394603192806244,
      "learning_rate": 9.71611775879062e-05,
      "loss": 1.4865,
      "step": 50660
    },
    {
      "epoch": 2.3072806408454283,
      "grad_norm": 0.2578514814376831,
      "learning_rate": 9.715563862525145e-05,
      "loss": 1.4828,
      "step": 50670
    },
    {
      "epoch": 2.3078225820109406,
      "grad_norm": 0.2434462159872055,
      "learning_rate": 9.715009444053921e-05,
      "loss": 1.4906,
      "step": 50680
    },
    {
      "epoch": 2.3083645231764525,
      "grad_norm": 0.3057672381401062,
      "learning_rate": 9.714454503445626e-05,
      "loss": 1.4889,
      "step": 50690
    },
    {
      "epoch": 2.308906464341965,
      "grad_norm": 0.2835559546947479,
      "learning_rate": 9.713899040769004e-05,
      "loss": 1.4879,
      "step": 50700
    },
    {
      "epoch": 2.309448405507477,
      "grad_norm": 0.23302173614501953,
      "learning_rate": 9.713343056092866e-05,
      "loss": 1.4891,
      "step": 50710
    },
    {
      "epoch": 2.3099903466729894,
      "grad_norm": 0.2765030562877655,
      "learning_rate": 9.71278654948608e-05,
      "loss": 1.4879,
      "step": 50720
    },
    {
      "epoch": 2.3100445407895407,
      "eval_loss": 2.4514873027801514,
      "eval_runtime": 22.1411,
      "eval_samples_per_second": 225.825,
      "eval_steps_per_second": 1.219,
      "step": 50721
    },
    {
      "epoch": 2.3105322878385017,
      "grad_norm": 0.3144146502017975,
      "learning_rate": 9.712229521017588e-05,
      "loss": 1.4854,
      "step": 50730
    },
    {
      "epoch": 2.3110742290040136,
      "grad_norm": 0.2929839491844177,
      "learning_rate": 9.71167197075639e-05,
      "loss": 1.4764,
      "step": 50740
    },
    {
      "epoch": 2.311616170169526,
      "grad_norm": 0.38944029808044434,
      "learning_rate": 9.711113898771554e-05,
      "loss": 1.4945,
      "step": 50750
    },
    {
      "epoch": 2.3121581113350382,
      "grad_norm": 0.28745847940444946,
      "learning_rate": 9.71055530513221e-05,
      "loss": 1.4949,
      "step": 50760
    },
    {
      "epoch": 2.3127000525005506,
      "grad_norm": 0.26115939021110535,
      "learning_rate": 9.709996189907557e-05,
      "loss": 1.4907,
      "step": 50770
    },
    {
      "epoch": 2.313241993666063,
      "grad_norm": 0.24977637827396393,
      "learning_rate": 9.709436553166853e-05,
      "loss": 1.4829,
      "step": 50780
    },
    {
      "epoch": 2.3137839348315747,
      "grad_norm": 0.35427191853523254,
      "learning_rate": 9.708876394979424e-05,
      "loss": 1.4819,
      "step": 50790
    },
    {
      "epoch": 2.314325875997087,
      "grad_norm": 0.24472922086715698,
      "learning_rate": 9.708315715414661e-05,
      "loss": 1.4968,
      "step": 50800
    },
    {
      "epoch": 2.314759428929497,
      "eval_loss": 2.440610647201538,
      "eval_runtime": 21.9959,
      "eval_samples_per_second": 227.315,
      "eval_steps_per_second": 1.227,
      "step": 50808
    },
    {
      "epoch": 2.3148678171625994,
      "grad_norm": 0.338789165019989,
      "learning_rate": 9.707754514542017e-05,
      "loss": 1.4834,
      "step": 50810
    },
    {
      "epoch": 2.3154097583281117,
      "grad_norm": 0.30206480622291565,
      "learning_rate": 9.707192792431014e-05,
      "loss": 1.4907,
      "step": 50820
    },
    {
      "epoch": 2.3159516994936236,
      "grad_norm": 0.23182201385498047,
      "learning_rate": 9.70663054915123e-05,
      "loss": 1.4917,
      "step": 50830
    },
    {
      "epoch": 2.316493640659136,
      "grad_norm": 0.3595997989177704,
      "learning_rate": 9.70606778477232e-05,
      "loss": 1.4962,
      "step": 50840
    },
    {
      "epoch": 2.317035581824648,
      "grad_norm": 0.21878936886787415,
      "learning_rate": 9.705504499363993e-05,
      "loss": 1.4847,
      "step": 50850
    },
    {
      "epoch": 2.3175775229901605,
      "grad_norm": 0.19247514009475708,
      "learning_rate": 9.704940692996027e-05,
      "loss": 1.4883,
      "step": 50860
    },
    {
      "epoch": 2.3181194641556724,
      "grad_norm": 0.20777463912963867,
      "learning_rate": 9.704376365738262e-05,
      "loss": 1.5016,
      "step": 50870
    },
    {
      "epoch": 2.3186614053211847,
      "grad_norm": 0.2756713628768921,
      "learning_rate": 9.703811517660609e-05,
      "loss": 1.494,
      "step": 50880
    },
    {
      "epoch": 2.319203346486697,
      "grad_norm": 0.3496546745300293,
      "learning_rate": 9.703246148833037e-05,
      "loss": 1.495,
      "step": 50890
    },
    {
      "epoch": 2.319474317069453,
      "eval_loss": 2.451153039932251,
      "eval_runtime": 22.0437,
      "eval_samples_per_second": 226.823,
      "eval_steps_per_second": 1.225,
      "step": 50895
    },
    {
      "epoch": 2.3197452876522093,
      "grad_norm": 0.28930479288101196,
      "learning_rate": 9.702680259325579e-05,
      "loss": 1.4995,
      "step": 50900
    },
    {
      "epoch": 2.3202872288177216,
      "grad_norm": 0.1946384757757187,
      "learning_rate": 9.702113849208337e-05,
      "loss": 1.4856,
      "step": 50910
    },
    {
      "epoch": 2.3208291699832335,
      "grad_norm": 0.23787322640419006,
      "learning_rate": 9.701546918551475e-05,
      "loss": 1.4841,
      "step": 50920
    },
    {
      "epoch": 2.321371111148746,
      "grad_norm": 0.2119310200214386,
      "learning_rate": 9.700979467425222e-05,
      "loss": 1.4882,
      "step": 50930
    },
    {
      "epoch": 2.321913052314258,
      "grad_norm": 0.23019666969776154,
      "learning_rate": 9.700411495899872e-05,
      "loss": 1.4834,
      "step": 50940
    },
    {
      "epoch": 2.3224549934797705,
      "grad_norm": 0.22270162403583527,
      "learning_rate": 9.699843004045782e-05,
      "loss": 1.4819,
      "step": 50950
    },
    {
      "epoch": 2.322996934645283,
      "grad_norm": 0.3237341642379761,
      "learning_rate": 9.699273991933373e-05,
      "loss": 1.4903,
      "step": 50960
    },
    {
      "epoch": 2.3235388758107947,
      "grad_norm": 0.26717808842658997,
      "learning_rate": 9.698704459633136e-05,
      "loss": 1.4825,
      "step": 50970
    },
    {
      "epoch": 2.324080816976307,
      "grad_norm": 0.27529168128967285,
      "learning_rate": 9.698134407215618e-05,
      "loss": 1.4881,
      "step": 50980
    },
    {
      "epoch": 2.3241892052094095,
      "eval_loss": 2.4506008625030518,
      "eval_runtime": 22.0206,
      "eval_samples_per_second": 227.06,
      "eval_steps_per_second": 1.226,
      "step": 50982
    },
    {
      "epoch": 2.3246227581418193,
      "grad_norm": 0.17906807363033295,
      "learning_rate": 9.697563834751436e-05,
      "loss": 1.4853,
      "step": 50990
    },
    {
      "epoch": 2.3251646993073316,
      "grad_norm": 0.2530919313430786,
      "learning_rate": 9.69699274231127e-05,
      "loss": 1.4873,
      "step": 51000
    },
    {
      "epoch": 2.3257066404728435,
      "grad_norm": 0.20605182647705078,
      "learning_rate": 9.696421129965865e-05,
      "loss": 1.4754,
      "step": 51010
    },
    {
      "epoch": 2.326248581638356,
      "grad_norm": 0.3275115489959717,
      "learning_rate": 9.69584899778603e-05,
      "loss": 1.4704,
      "step": 51020
    },
    {
      "epoch": 2.326790522803868,
      "grad_norm": 0.4294751286506653,
      "learning_rate": 9.695276345842638e-05,
      "loss": 1.4987,
      "step": 51030
    },
    {
      "epoch": 2.3273324639693804,
      "grad_norm": 0.22030548751354218,
      "learning_rate": 9.694703174206624e-05,
      "loss": 1.4899,
      "step": 51040
    },
    {
      "epoch": 2.3278744051348923,
      "grad_norm": 0.263075053691864,
      "learning_rate": 9.694129482948994e-05,
      "loss": 1.4819,
      "step": 51050
    },
    {
      "epoch": 2.3284163463004046,
      "grad_norm": 0.26459792256355286,
      "learning_rate": 9.693555272140813e-05,
      "loss": 1.499,
      "step": 51060
    },
    {
      "epoch": 2.3289040933493657,
      "eval_loss": 2.4480111598968506,
      "eval_runtime": 22.0391,
      "eval_samples_per_second": 226.869,
      "eval_steps_per_second": 1.225,
      "step": 51069
    },
    {
      "epoch": 2.328958287465917,
      "grad_norm": 0.19399993121623993,
      "learning_rate": 9.692980541853211e-05,
      "loss": 1.4926,
      "step": 51070
    },
    {
      "epoch": 2.3295002286314292,
      "grad_norm": 0.2647286057472229,
      "learning_rate": 9.692405292157384e-05,
      "loss": 1.4846,
      "step": 51080
    },
    {
      "epoch": 2.3300421697969416,
      "grad_norm": 0.2626708149909973,
      "learning_rate": 9.69182952312459e-05,
      "loss": 1.4875,
      "step": 51090
    },
    {
      "epoch": 2.3305841109624534,
      "grad_norm": 0.3158663213253021,
      "learning_rate": 9.691253234826155e-05,
      "loss": 1.4902,
      "step": 51100
    },
    {
      "epoch": 2.3311260521279658,
      "grad_norm": 0.18909795582294464,
      "learning_rate": 9.690676427333467e-05,
      "loss": 1.4898,
      "step": 51110
    },
    {
      "epoch": 2.331667993293478,
      "grad_norm": 0.3357420265674591,
      "learning_rate": 9.690099100717974e-05,
      "loss": 1.4999,
      "step": 51120
    },
    {
      "epoch": 2.3322099344589904,
      "grad_norm": 0.36199477314949036,
      "learning_rate": 9.689521255051198e-05,
      "loss": 1.4957,
      "step": 51130
    },
    {
      "epoch": 2.3327518756245027,
      "grad_norm": 0.36792802810668945,
      "learning_rate": 9.688942890404718e-05,
      "loss": 1.4897,
      "step": 51140
    },
    {
      "epoch": 2.3332938167900146,
      "grad_norm": 0.22224745154380798,
      "learning_rate": 9.68836400685018e-05,
      "loss": 1.4866,
      "step": 51150
    },
    {
      "epoch": 2.3336189814893222,
      "eval_loss": 2.449270248413086,
      "eval_runtime": 22.033,
      "eval_samples_per_second": 226.933,
      "eval_steps_per_second": 1.225,
      "step": 51156
    },
    {
      "epoch": 2.333835757955527,
      "grad_norm": 0.20259080827236176,
      "learning_rate": 9.687784604459292e-05,
      "loss": 1.4879,
      "step": 51160
    },
    {
      "epoch": 2.334377699121039,
      "grad_norm": 0.41556867957115173,
      "learning_rate": 9.687204683303829e-05,
      "loss": 1.4895,
      "step": 51170
    },
    {
      "epoch": 2.3349196402865515,
      "grad_norm": 0.24334754049777985,
      "learning_rate": 9.686624243455627e-05,
      "loss": 1.4751,
      "step": 51180
    },
    {
      "epoch": 2.335461581452064,
      "grad_norm": 0.20898622274398804,
      "learning_rate": 9.686043284986593e-05,
      "loss": 1.4863,
      "step": 51190
    },
    {
      "epoch": 2.3360035226175757,
      "grad_norm": 0.22658079862594604,
      "learning_rate": 9.685461807968688e-05,
      "loss": 1.4837,
      "step": 51200
    },
    {
      "epoch": 2.336545463783088,
      "grad_norm": 0.29090017080307007,
      "learning_rate": 9.684879812473948e-05,
      "loss": 1.4827,
      "step": 51210
    },
    {
      "epoch": 2.3370874049486003,
      "grad_norm": 0.27824729681015015,
      "learning_rate": 9.684297298574465e-05,
      "loss": 1.4882,
      "step": 51220
    },
    {
      "epoch": 2.3376293461141127,
      "grad_norm": 0.2605385482311249,
      "learning_rate": 9.683714266342398e-05,
      "loss": 1.4894,
      "step": 51230
    },
    {
      "epoch": 2.3381712872796245,
      "grad_norm": 0.34890076518058777,
      "learning_rate": 9.683130715849973e-05,
      "loss": 1.4949,
      "step": 51240
    },
    {
      "epoch": 2.3383338696292784,
      "eval_loss": 2.450643539428711,
      "eval_runtime": 22.0782,
      "eval_samples_per_second": 226.468,
      "eval_steps_per_second": 1.223,
      "step": 51243
    },
    {
      "epoch": 2.338713228445137,
      "grad_norm": 0.2740577459335327,
      "learning_rate": 9.682546647169478e-05,
      "loss": 1.4953,
      "step": 51250
    },
    {
      "epoch": 2.339255169610649,
      "grad_norm": 0.29385995864868164,
      "learning_rate": 9.68196206037326e-05,
      "loss": 1.4833,
      "step": 51260
    },
    {
      "epoch": 2.3397971107761615,
      "grad_norm": 0.2904885709285736,
      "learning_rate": 9.681376955533739e-05,
      "loss": 1.4861,
      "step": 51270
    },
    {
      "epoch": 2.3403390519416734,
      "grad_norm": 0.19600605964660645,
      "learning_rate": 9.680791332723396e-05,
      "loss": 1.4891,
      "step": 51280
    },
    {
      "epoch": 2.3408809931071857,
      "grad_norm": 0.35659798979759216,
      "learning_rate": 9.680205192014772e-05,
      "loss": 1.481,
      "step": 51290
    },
    {
      "epoch": 2.341422934272698,
      "grad_norm": 0.20791636407375336,
      "learning_rate": 9.679618533480478e-05,
      "loss": 1.4846,
      "step": 51300
    },
    {
      "epoch": 2.3419648754382103,
      "grad_norm": 0.36341404914855957,
      "learning_rate": 9.679031357193186e-05,
      "loss": 1.487,
      "step": 51310
    },
    {
      "epoch": 2.3425068166037226,
      "grad_norm": 0.2511788308620453,
      "learning_rate": 9.678443663225633e-05,
      "loss": 1.4911,
      "step": 51320
    },
    {
      "epoch": 2.3430487577692345,
      "grad_norm": 0.22106388211250305,
      "learning_rate": 9.67785545165062e-05,
      "loss": 1.4855,
      "step": 51330
    },
    {
      "epoch": 2.3430487577692345,
      "eval_loss": 2.449833631515503,
      "eval_runtime": 22.0541,
      "eval_samples_per_second": 226.716,
      "eval_steps_per_second": 1.224,
      "step": 51330
    },
    {
      "epoch": 2.343590698934747,
      "grad_norm": 0.2117205709218979,
      "learning_rate": 9.677266722541012e-05,
      "loss": 1.4836,
      "step": 51340
    },
    {
      "epoch": 2.344132640100259,
      "grad_norm": 0.261532187461853,
      "learning_rate": 9.67667747596974e-05,
      "loss": 1.4843,
      "step": 51350
    },
    {
      "epoch": 2.3446745812657714,
      "grad_norm": 0.25580722093582153,
      "learning_rate": 9.676087712009794e-05,
      "loss": 1.4842,
      "step": 51360
    },
    {
      "epoch": 2.3452165224312838,
      "grad_norm": 0.21774953603744507,
      "learning_rate": 9.675497430734233e-05,
      "loss": 1.4816,
      "step": 51370
    },
    {
      "epoch": 2.3457584635967956,
      "grad_norm": 0.4116745591163635,
      "learning_rate": 9.67490663221618e-05,
      "loss": 1.4927,
      "step": 51380
    },
    {
      "epoch": 2.346300404762308,
      "grad_norm": 0.35664743185043335,
      "learning_rate": 9.674315316528816e-05,
      "loss": 1.4871,
      "step": 51390
    },
    {
      "epoch": 2.3468423459278203,
      "grad_norm": 0.25654247403144836,
      "learning_rate": 9.673723483745395e-05,
      "loss": 1.4728,
      "step": 51400
    },
    {
      "epoch": 2.3473842870933326,
      "grad_norm": 0.32242023944854736,
      "learning_rate": 9.673131133939229e-05,
      "loss": 1.4841,
      "step": 51410
    },
    {
      "epoch": 2.347763645909191,
      "eval_loss": 2.4569506645202637,
      "eval_runtime": 21.6699,
      "eval_samples_per_second": 230.735,
      "eval_steps_per_second": 1.246,
      "step": 51417
    },
    {
      "epoch": 2.347926228258845,
      "grad_norm": 0.19315354526042938,
      "learning_rate": 9.672538267183699e-05,
      "loss": 1.4803,
      "step": 51420
    },
    {
      "epoch": 2.3484681694243568,
      "grad_norm": 0.2572130262851715,
      "learning_rate": 9.67194488355224e-05,
      "loss": 1.4969,
      "step": 51430
    },
    {
      "epoch": 2.349010110589869,
      "grad_norm": 0.5099160671234131,
      "learning_rate": 9.671350983118365e-05,
      "loss": 1.4876,
      "step": 51440
    },
    {
      "epoch": 2.3495520517553814,
      "grad_norm": 0.2558972239494324,
      "learning_rate": 9.670756565955637e-05,
      "loss": 1.4961,
      "step": 51450
    },
    {
      "epoch": 2.3500939929208937,
      "grad_norm": 0.5650994181632996,
      "learning_rate": 9.670161632137696e-05,
      "loss": 1.4858,
      "step": 51460
    },
    {
      "epoch": 2.3506359340864056,
      "grad_norm": 0.45410850644111633,
      "learning_rate": 9.669566181738236e-05,
      "loss": 1.4921,
      "step": 51470
    },
    {
      "epoch": 2.351177875251918,
      "grad_norm": 0.37703588604927063,
      "learning_rate": 9.668970214831019e-05,
      "loss": 1.4899,
      "step": 51480
    },
    {
      "epoch": 2.35171981641743,
      "grad_norm": 0.20002073049545288,
      "learning_rate": 9.668373731489872e-05,
      "loss": 1.4873,
      "step": 51490
    },
    {
      "epoch": 2.3522617575829425,
      "grad_norm": 0.2333013415336609,
      "learning_rate": 9.667776731788685e-05,
      "loss": 1.4952,
      "step": 51500
    },
    {
      "epoch": 2.352478534049147,
      "eval_loss": 2.4526426792144775,
      "eval_runtime": 22.019,
      "eval_samples_per_second": 227.077,
      "eval_steps_per_second": 1.226,
      "step": 51504
    },
    {
      "epoch": 2.3528036987484544,
      "grad_norm": 0.28557243943214417,
      "learning_rate": 9.667179215801411e-05,
      "loss": 1.4851,
      "step": 51510
    },
    {
      "epoch": 2.3533456399139667,
      "grad_norm": 0.2601202130317688,
      "learning_rate": 9.666581183602069e-05,
      "loss": 1.4857,
      "step": 51520
    },
    {
      "epoch": 2.353887581079479,
      "grad_norm": 0.37464073300361633,
      "learning_rate": 9.665982635264736e-05,
      "loss": 1.491,
      "step": 51530
    },
    {
      "epoch": 2.3544295222449914,
      "grad_norm": 0.3240549862384796,
      "learning_rate": 9.665383570863565e-05,
      "loss": 1.478,
      "step": 51540
    },
    {
      "epoch": 2.3549714634105037,
      "grad_norm": 0.18953123688697815,
      "learning_rate": 9.66478399047276e-05,
      "loss": 1.4791,
      "step": 51550
    },
    {
      "epoch": 2.3555134045760155,
      "grad_norm": 0.24826766550540924,
      "learning_rate": 9.664183894166595e-05,
      "loss": 1.4875,
      "step": 51560
    },
    {
      "epoch": 2.356055345741528,
      "grad_norm": 0.2706245481967926,
      "learning_rate": 9.66358328201941e-05,
      "loss": 1.4802,
      "step": 51570
    },
    {
      "epoch": 2.35659728690704,
      "grad_norm": 0.3040546178817749,
      "learning_rate": 9.662982154105604e-05,
      "loss": 1.4826,
      "step": 51580
    },
    {
      "epoch": 2.3571392280725525,
      "grad_norm": 0.28579655289649963,
      "learning_rate": 9.66238051049964e-05,
      "loss": 1.4826,
      "step": 51590
    },
    {
      "epoch": 2.3571934221891038,
      "eval_loss": 2.4456629753112793,
      "eval_runtime": 22.0034,
      "eval_samples_per_second": 227.238,
      "eval_steps_per_second": 1.227,
      "step": 51591
    },
    {
      "epoch": 2.357681169238065,
      "grad_norm": 0.2639320194721222,
      "learning_rate": 9.661778351276052e-05,
      "loss": 1.4805,
      "step": 51600
    },
    {
      "epoch": 2.3582231104035767,
      "grad_norm": 0.3113354742527008,
      "learning_rate": 9.66117567650943e-05,
      "loss": 1.4745,
      "step": 51610
    },
    {
      "epoch": 2.358765051569089,
      "grad_norm": 0.2081727683544159,
      "learning_rate": 9.660572486274432e-05,
      "loss": 1.4801,
      "step": 51620
    },
    {
      "epoch": 2.3593069927346013,
      "grad_norm": 0.2256453037261963,
      "learning_rate": 9.659968780645777e-05,
      "loss": 1.4801,
      "step": 51630
    },
    {
      "epoch": 2.3598489339001136,
      "grad_norm": 0.23762857913970947,
      "learning_rate": 9.659364559698249e-05,
      "loss": 1.4833,
      "step": 51640
    },
    {
      "epoch": 2.3603908750656255,
      "grad_norm": 0.28543922305107117,
      "learning_rate": 9.6587598235067e-05,
      "loss": 1.4877,
      "step": 51650
    },
    {
      "epoch": 2.360932816231138,
      "grad_norm": 0.32550597190856934,
      "learning_rate": 9.658154572146039e-05,
      "loss": 1.4831,
      "step": 51660
    },
    {
      "epoch": 2.36147475739665,
      "grad_norm": 0.3355502188205719,
      "learning_rate": 9.65754880569124e-05,
      "loss": 1.4936,
      "step": 51670
    },
    {
      "epoch": 2.36190831032906,
      "eval_loss": 2.46115779876709,
      "eval_runtime": 22.0455,
      "eval_samples_per_second": 226.804,
      "eval_steps_per_second": 1.225,
      "step": 51678
    },
    {
      "epoch": 2.3620166985621625,
      "grad_norm": 0.3635583519935608,
      "learning_rate": 9.65694252421735e-05,
      "loss": 1.4941,
      "step": 51680
    },
    {
      "epoch": 2.3625586397276743,
      "grad_norm": 0.2427382469177246,
      "learning_rate": 9.656335727799464e-05,
      "loss": 1.4888,
      "step": 51690
    },
    {
      "epoch": 2.3631005808931866,
      "grad_norm": 0.178348109126091,
      "learning_rate": 9.655728416512754e-05,
      "loss": 1.4835,
      "step": 51700
    },
    {
      "epoch": 2.363642522058699,
      "grad_norm": 0.22014687955379486,
      "learning_rate": 9.65512059043245e-05,
      "loss": 1.485,
      "step": 51710
    },
    {
      "epoch": 2.3641844632242113,
      "grad_norm": 0.21226602792739868,
      "learning_rate": 9.654512249633848e-05,
      "loss": 1.4826,
      "step": 51720
    },
    {
      "epoch": 2.3647264043897236,
      "grad_norm": 0.28715527057647705,
      "learning_rate": 9.653903394192304e-05,
      "loss": 1.4772,
      "step": 51730
    },
    {
      "epoch": 2.3652683455552355,
      "grad_norm": 0.24136418104171753,
      "learning_rate": 9.653294024183243e-05,
      "loss": 1.4856,
      "step": 51740
    },
    {
      "epoch": 2.365810286720748,
      "grad_norm": 0.20212812721729279,
      "learning_rate": 9.65268413968215e-05,
      "loss": 1.4721,
      "step": 51750
    },
    {
      "epoch": 2.36635222788626,
      "grad_norm": 0.2756584882736206,
      "learning_rate": 9.652073740764576e-05,
      "loss": 1.4901,
      "step": 51760
    },
    {
      "epoch": 2.366623198469016,
      "eval_loss": 2.4501538276672363,
      "eval_runtime": 22.0902,
      "eval_samples_per_second": 226.345,
      "eval_steps_per_second": 1.222,
      "step": 51765
    },
    {
      "epoch": 2.3668941690517724,
      "grad_norm": 0.27957209944725037,
      "learning_rate": 9.651462827506131e-05,
      "loss": 1.484,
      "step": 51770
    },
    {
      "epoch": 2.3674361102172847,
      "grad_norm": 0.3119787871837616,
      "learning_rate": 9.650851399982495e-05,
      "loss": 1.4874,
      "step": 51780
    },
    {
      "epoch": 2.3679780513827966,
      "grad_norm": 0.37608903646469116,
      "learning_rate": 9.65023945826941e-05,
      "loss": 1.4897,
      "step": 51790
    },
    {
      "epoch": 2.368519992548309,
      "grad_norm": 0.1750231236219406,
      "learning_rate": 9.64962700244268e-05,
      "loss": 1.4893,
      "step": 51800
    },
    {
      "epoch": 2.3690619337138212,
      "grad_norm": 0.20804855227470398,
      "learning_rate": 9.649014032578174e-05,
      "loss": 1.4854,
      "step": 51810
    },
    {
      "epoch": 2.3696038748793335,
      "grad_norm": 0.19923941791057587,
      "learning_rate": 9.648400548751819e-05,
      "loss": 1.4891,
      "step": 51820
    },
    {
      "epoch": 2.370145816044846,
      "grad_norm": 0.3422909379005432,
      "learning_rate": 9.647786551039617e-05,
      "loss": 1.4859,
      "step": 51830
    },
    {
      "epoch": 2.3706877572103577,
      "grad_norm": 0.2580278813838959,
      "learning_rate": 9.647172039517623e-05,
      "loss": 1.4886,
      "step": 51840
    },
    {
      "epoch": 2.37122969837587,
      "grad_norm": 0.21753491461277008,
      "learning_rate": 9.646557014261966e-05,
      "loss": 1.4819,
      "step": 51850
    },
    {
      "epoch": 2.3713380866089726,
      "eval_loss": 2.452852964401245,
      "eval_runtime": 22.005,
      "eval_samples_per_second": 227.221,
      "eval_steps_per_second": 1.227,
      "step": 51852
    },
    {
      "epoch": 2.3717716395413824,
      "grad_norm": 0.3423318862915039,
      "learning_rate": 9.645941475348825e-05,
      "loss": 1.4947,
      "step": 51860
    },
    {
      "epoch": 2.3723135807068947,
      "grad_norm": 0.28139030933380127,
      "learning_rate": 9.645325422854454e-05,
      "loss": 1.4803,
      "step": 51870
    },
    {
      "epoch": 2.3728555218724066,
      "grad_norm": 0.21756911277770996,
      "learning_rate": 9.644708856855168e-05,
      "loss": 1.4883,
      "step": 51880
    },
    {
      "epoch": 2.373397463037919,
      "grad_norm": 0.22105887532234192,
      "learning_rate": 9.644091777427344e-05,
      "loss": 1.4905,
      "step": 51890
    },
    {
      "epoch": 2.373939404203431,
      "grad_norm": 0.2563340365886688,
      "learning_rate": 9.643474184647422e-05,
      "loss": 1.4769,
      "step": 51900
    },
    {
      "epoch": 2.3744813453689435,
      "grad_norm": 0.19484004378318787,
      "learning_rate": 9.642856078591906e-05,
      "loss": 1.4858,
      "step": 51910
    },
    {
      "epoch": 2.3750232865344554,
      "grad_norm": 0.4082275629043579,
      "learning_rate": 9.642237459337366e-05,
      "loss": 1.4934,
      "step": 51920
    },
    {
      "epoch": 2.3755652276999677,
      "grad_norm": 0.26656705141067505,
      "learning_rate": 9.641618326960435e-05,
      "loss": 1.4833,
      "step": 51930
    },
    {
      "epoch": 2.3760529747489287,
      "eval_loss": 2.4559152126312256,
      "eval_runtime": 22.0417,
      "eval_samples_per_second": 226.843,
      "eval_steps_per_second": 1.225,
      "step": 51939
    },
    {
      "epoch": 2.37610716886548,
      "grad_norm": 0.38334181904792786,
      "learning_rate": 9.640998681537805e-05,
      "loss": 1.4832,
      "step": 51940
    },
    {
      "epoch": 2.3766491100309923,
      "grad_norm": 0.34898388385772705,
      "learning_rate": 9.640378523146238e-05,
      "loss": 1.489,
      "step": 51950
    },
    {
      "epoch": 2.3771910511965046,
      "grad_norm": 0.23095417022705078,
      "learning_rate": 9.639757851862553e-05,
      "loss": 1.4816,
      "step": 51960
    },
    {
      "epoch": 2.3777329923620165,
      "grad_norm": 0.19512107968330383,
      "learning_rate": 9.63913666776364e-05,
      "loss": 1.4864,
      "step": 51970
    },
    {
      "epoch": 2.378274933527529,
      "grad_norm": 0.4121658205986023,
      "learning_rate": 9.638514970926447e-05,
      "loss": 1.4825,
      "step": 51980
    },
    {
      "epoch": 2.378816874693041,
      "grad_norm": 0.3080250024795532,
      "learning_rate": 9.637892761427987e-05,
      "loss": 1.4921,
      "step": 51990
    },
    {
      "epoch": 2.3793588158585535,
      "grad_norm": 0.25781115889549255,
      "learning_rate": 9.637270039345335e-05,
      "loss": 1.4752,
      "step": 52000
    },
    {
      "epoch": 2.379900757024066,
      "grad_norm": 0.2879182994365692,
      "learning_rate": 9.636646804755635e-05,
      "loss": 1.4941,
      "step": 52010
    },
    {
      "epoch": 2.3804426981895777,
      "grad_norm": 0.2381771355867386,
      "learning_rate": 9.636023057736088e-05,
      "loss": 1.4927,
      "step": 52020
    },
    {
      "epoch": 2.3807678628888853,
      "eval_loss": 2.451416492462158,
      "eval_runtime": 22.0461,
      "eval_samples_per_second": 226.797,
      "eval_steps_per_second": 1.225,
      "step": 52026
    },
    {
      "epoch": 2.38098463935509,
      "grad_norm": 0.33801183104515076,
      "learning_rate": 9.63539879836396e-05,
      "loss": 1.4843,
      "step": 52030
    },
    {
      "epoch": 2.3815265805206023,
      "grad_norm": 0.19671766459941864,
      "learning_rate": 9.634774026716585e-05,
      "loss": 1.4758,
      "step": 52040
    },
    {
      "epoch": 2.3820685216861146,
      "grad_norm": 0.17954659461975098,
      "learning_rate": 9.634148742871353e-05,
      "loss": 1.4805,
      "step": 52050
    },
    {
      "epoch": 2.3826104628516265,
      "grad_norm": 0.31026336550712585,
      "learning_rate": 9.633522946905725e-05,
      "loss": 1.4882,
      "step": 52060
    },
    {
      "epoch": 2.383152404017139,
      "grad_norm": 0.19763007760047913,
      "learning_rate": 9.632896638897219e-05,
      "loss": 1.4804,
      "step": 52070
    },
    {
      "epoch": 2.383694345182651,
      "grad_norm": 0.37613585591316223,
      "learning_rate": 9.63226981892342e-05,
      "loss": 1.4732,
      "step": 52080
    },
    {
      "epoch": 2.3842362863481634,
      "grad_norm": 0.3250291347503662,
      "learning_rate": 9.631642487061978e-05,
      "loss": 1.4788,
      "step": 52090
    },
    {
      "epoch": 2.3847782275136753,
      "grad_norm": 0.4253854751586914,
      "learning_rate": 9.631014643390602e-05,
      "loss": 1.4871,
      "step": 52100
    },
    {
      "epoch": 2.3853201686791876,
      "grad_norm": 0.1921018362045288,
      "learning_rate": 9.630386287987067e-05,
      "loss": 1.4825,
      "step": 52110
    },
    {
      "epoch": 2.3854827510288414,
      "eval_loss": 2.4489946365356445,
      "eval_runtime": 21.9884,
      "eval_samples_per_second": 227.393,
      "eval_steps_per_second": 1.228,
      "step": 52113
    },
    {
      "epoch": 2.3858621098447,
      "grad_norm": 0.2642991840839386,
      "learning_rate": 9.629757420929212e-05,
      "loss": 1.4773,
      "step": 52120
    },
    {
      "epoch": 2.3864040510102122,
      "grad_norm": 0.20693929493427277,
      "learning_rate": 9.629128042294936e-05,
      "loss": 1.4714,
      "step": 52130
    },
    {
      "epoch": 2.3869459921757246,
      "grad_norm": 0.2861025631427765,
      "learning_rate": 9.628498152162205e-05,
      "loss": 1.4816,
      "step": 52140
    },
    {
      "epoch": 2.3874879333412364,
      "grad_norm": 0.1987767517566681,
      "learning_rate": 9.627867750609047e-05,
      "loss": 1.4794,
      "step": 52150
    },
    {
      "epoch": 2.3880298745067488,
      "grad_norm": 0.19934241473674774,
      "learning_rate": 9.627236837713553e-05,
      "loss": 1.4802,
      "step": 52160
    },
    {
      "epoch": 2.388571815672261,
      "grad_norm": 0.2977670133113861,
      "learning_rate": 9.626605413553881e-05,
      "loss": 1.4917,
      "step": 52170
    },
    {
      "epoch": 2.3891137568377734,
      "grad_norm": 0.22710345685482025,
      "learning_rate": 9.625973478208243e-05,
      "loss": 1.4835,
      "step": 52180
    },
    {
      "epoch": 2.3896556980032857,
      "grad_norm": 0.2273036539554596,
      "learning_rate": 9.625341031754925e-05,
      "loss": 1.4893,
      "step": 52190
    },
    {
      "epoch": 2.3901976391687976,
      "grad_norm": 0.2654477655887604,
      "learning_rate": 9.624708074272272e-05,
      "loss": 1.4888,
      "step": 52200
    },
    {
      "epoch": 2.3901976391687976,
      "eval_loss": 2.4439258575439453,
      "eval_runtime": 21.9105,
      "eval_samples_per_second": 228.201,
      "eval_steps_per_second": 1.232,
      "step": 52200
    },
    {
      "epoch": 2.39073958033431,
      "grad_norm": 0.2779911160469055,
      "learning_rate": 9.624074605838688e-05,
      "loss": 1.4723,
      "step": 52210
    },
    {
      "epoch": 2.391281521499822,
      "grad_norm": 0.20526061952114105,
      "learning_rate": 9.623440626532649e-05,
      "loss": 1.4886,
      "step": 52220
    },
    {
      "epoch": 2.3918234626653345,
      "grad_norm": 0.23855385184288025,
      "learning_rate": 9.622806136432684e-05,
      "loss": 1.4846,
      "step": 52230
    },
    {
      "epoch": 2.392365403830847,
      "grad_norm": 0.1878088265657425,
      "learning_rate": 9.622171135617397e-05,
      "loss": 1.4781,
      "step": 52240
    },
    {
      "epoch": 2.3929073449963587,
      "grad_norm": 0.28239238262176514,
      "learning_rate": 9.621535624165446e-05,
      "loss": 1.4894,
      "step": 52250
    },
    {
      "epoch": 2.393449286161871,
      "grad_norm": 0.24423588812351227,
      "learning_rate": 9.620899602155557e-05,
      "loss": 1.4842,
      "step": 52260
    },
    {
      "epoch": 2.3939912273273833,
      "grad_norm": 0.20243407785892487,
      "learning_rate": 9.620263069666514e-05,
      "loss": 1.4767,
      "step": 52270
    },
    {
      "epoch": 2.3945331684928957,
      "grad_norm": 0.3433877229690552,
      "learning_rate": 9.619626026777172e-05,
      "loss": 1.4847,
      "step": 52280
    },
    {
      "epoch": 2.394912527308754,
      "eval_loss": 2.4462528228759766,
      "eval_runtime": 22.0531,
      "eval_samples_per_second": 226.726,
      "eval_steps_per_second": 1.224,
      "step": 52287
    },
    {
      "epoch": 2.3950751096584075,
      "grad_norm": 0.2804941236972809,
      "learning_rate": 9.618988473566442e-05,
      "loss": 1.4887,
      "step": 52290
    },
    {
      "epoch": 2.39561705082392,
      "grad_norm": 0.21407003700733185,
      "learning_rate": 9.618350410113304e-05,
      "loss": 1.4823,
      "step": 52300
    },
    {
      "epoch": 2.396158991989432,
      "grad_norm": 0.2938377261161804,
      "learning_rate": 9.617711836496797e-05,
      "loss": 1.489,
      "step": 52310
    },
    {
      "epoch": 2.3967009331549445,
      "grad_norm": 0.19561982154846191,
      "learning_rate": 9.617072752796025e-05,
      "loss": 1.4788,
      "step": 52320
    },
    {
      "epoch": 2.3972428743204564,
      "grad_norm": 0.2788771688938141,
      "learning_rate": 9.616433159090154e-05,
      "loss": 1.4788,
      "step": 52330
    },
    {
      "epoch": 2.3977848154859687,
      "grad_norm": 0.24922244250774384,
      "learning_rate": 9.615793055458415e-05,
      "loss": 1.4827,
      "step": 52340
    },
    {
      "epoch": 2.398326756651481,
      "grad_norm": 0.18937751650810242,
      "learning_rate": 9.615152441980104e-05,
      "loss": 1.4771,
      "step": 52350
    },
    {
      "epoch": 2.3988686978169933,
      "grad_norm": 0.19057206809520721,
      "learning_rate": 9.614511318734572e-05,
      "loss": 1.4782,
      "step": 52360
    },
    {
      "epoch": 2.3994106389825056,
      "grad_norm": 0.40637481212615967,
      "learning_rate": 9.613869685801242e-05,
      "loss": 1.4894,
      "step": 52370
    },
    {
      "epoch": 2.3996274154487103,
      "eval_loss": 2.4485349655151367,
      "eval_runtime": 25.128,
      "eval_samples_per_second": 198.981,
      "eval_steps_per_second": 1.074,
      "step": 52374
    },
    {
      "epoch": 2.3999525801480175,
      "grad_norm": 0.18058116734027863,
      "learning_rate": 9.613227543259595e-05,
      "loss": 1.4847,
      "step": 52380
    },
    {
      "epoch": 2.40049452131353,
      "grad_norm": 0.2915930151939392,
      "learning_rate": 9.61258489118918e-05,
      "loss": 1.4771,
      "step": 52390
    },
    {
      "epoch": 2.401036462479042,
      "grad_norm": 0.20396968722343445,
      "learning_rate": 9.611941729669602e-05,
      "loss": 1.4799,
      "step": 52400
    },
    {
      "epoch": 2.4015784036445544,
      "grad_norm": 0.18857981264591217,
      "learning_rate": 9.611298058780536e-05,
      "loss": 1.4857,
      "step": 52410
    },
    {
      "epoch": 2.4021203448100668,
      "grad_norm": 0.3981879651546478,
      "learning_rate": 9.610653878601715e-05,
      "loss": 1.4748,
      "step": 52420
    },
    {
      "epoch": 2.4026622859755786,
      "grad_norm": 0.17841531336307526,
      "learning_rate": 9.610009189212938e-05,
      "loss": 1.4865,
      "step": 52430
    },
    {
      "epoch": 2.403204227141091,
      "grad_norm": 0.37024664878845215,
      "learning_rate": 9.609363990694067e-05,
      "loss": 1.4806,
      "step": 52440
    },
    {
      "epoch": 2.4037461683066033,
      "grad_norm": 0.19459642469882965,
      "learning_rate": 9.608718283125026e-05,
      "loss": 1.4754,
      "step": 52450
    },
    {
      "epoch": 2.4042881094721156,
      "grad_norm": 0.2543799579143524,
      "learning_rate": 9.608072066585803e-05,
      "loss": 1.4872,
      "step": 52460
    },
    {
      "epoch": 2.404342303588667,
      "eval_loss": 2.44846773147583,
      "eval_runtime": 21.9857,
      "eval_samples_per_second": 227.42,
      "eval_steps_per_second": 1.228,
      "step": 52461
    },
    {
      "epoch": 2.404830050637628,
      "grad_norm": 0.31468695402145386,
      "learning_rate": 9.607425341156447e-05,
      "loss": 1.4677,
      "step": 52470
    },
    {
      "epoch": 2.4053719918031398,
      "grad_norm": 0.31996336579322815,
      "learning_rate": 9.606778106917071e-05,
      "loss": 1.4961,
      "step": 52480
    },
    {
      "epoch": 2.405913932968652,
      "grad_norm": 0.22491812705993652,
      "learning_rate": 9.606130363947856e-05,
      "loss": 1.4923,
      "step": 52490
    },
    {
      "epoch": 2.4064558741341644,
      "grad_norm": 0.18299773335456848,
      "learning_rate": 9.605482112329037e-05,
      "loss": 1.4828,
      "step": 52500
    },
    {
      "epoch": 2.4069978152996767,
      "grad_norm": 0.17865203320980072,
      "learning_rate": 9.604833352140918e-05,
      "loss": 1.4884,
      "step": 52510
    },
    {
      "epoch": 2.4075397564651886,
      "grad_norm": 0.31600865721702576,
      "learning_rate": 9.604184083463863e-05,
      "loss": 1.4814,
      "step": 52520
    },
    {
      "epoch": 2.408081697630701,
      "grad_norm": 0.22451327741146088,
      "learning_rate": 9.603534306378305e-05,
      "loss": 1.4824,
      "step": 52530
    },
    {
      "epoch": 2.408623638796213,
      "grad_norm": 0.1985584944486618,
      "learning_rate": 9.602884020964734e-05,
      "loss": 1.4852,
      "step": 52540
    },
    {
      "epoch": 2.409057191728623,
      "eval_loss": 2.45249605178833,
      "eval_runtime": 22.1056,
      "eval_samples_per_second": 226.187,
      "eval_steps_per_second": 1.221,
      "step": 52548
    },
    {
      "epoch": 2.4091655799617255,
      "grad_norm": 0.2913689911365509,
      "learning_rate": 9.6022332273037e-05,
      "loss": 1.4858,
      "step": 52550
    },
    {
      "epoch": 2.4097075211272374,
      "grad_norm": 0.28544992208480835,
      "learning_rate": 9.601581925475825e-05,
      "loss": 1.4845,
      "step": 52560
    },
    {
      "epoch": 2.4102494622927497,
      "grad_norm": 0.34636637568473816,
      "learning_rate": 9.600930115561791e-05,
      "loss": 1.4848,
      "step": 52570
    },
    {
      "epoch": 2.410791403458262,
      "grad_norm": 0.22824904322624207,
      "learning_rate": 9.600277797642335e-05,
      "loss": 1.4804,
      "step": 52580
    },
    {
      "epoch": 2.4113333446237744,
      "grad_norm": 0.2189302295446396,
      "learning_rate": 9.59962497179827e-05,
      "loss": 1.4794,
      "step": 52590
    },
    {
      "epoch": 2.4118752857892867,
      "grad_norm": 0.1900554746389389,
      "learning_rate": 9.59897163811046e-05,
      "loss": 1.4769,
      "step": 52600
    },
    {
      "epoch": 2.4124172269547985,
      "grad_norm": 0.2690756618976593,
      "learning_rate": 9.59831779665984e-05,
      "loss": 1.4868,
      "step": 52610
    },
    {
      "epoch": 2.412959168120311,
      "grad_norm": 0.22319385409355164,
      "learning_rate": 9.597663447527407e-05,
      "loss": 1.4726,
      "step": 52620
    },
    {
      "epoch": 2.413501109285823,
      "grad_norm": 0.22841401398181915,
      "learning_rate": 9.597008590794211e-05,
      "loss": 1.4806,
      "step": 52630
    },
    {
      "epoch": 2.413772079868579,
      "eval_loss": 2.4524683952331543,
      "eval_runtime": 22.0707,
      "eval_samples_per_second": 226.544,
      "eval_steps_per_second": 1.223,
      "step": 52635
    },
    {
      "epoch": 2.4140430504513355,
      "grad_norm": 0.20807218551635742,
      "learning_rate": 9.596353226541382e-05,
      "loss": 1.4779,
      "step": 52640
    },
    {
      "epoch": 2.414584991616848,
      "grad_norm": 0.4142747223377228,
      "learning_rate": 9.595697354850101e-05,
      "loss": 1.4986,
      "step": 52650
    },
    {
      "epoch": 2.4151269327823597,
      "grad_norm": 0.21264301240444183,
      "learning_rate": 9.59504097580161e-05,
      "loss": 1.4877,
      "step": 52660
    },
    {
      "epoch": 2.415668873947872,
      "grad_norm": 0.5480301976203918,
      "learning_rate": 9.594384089477224e-05,
      "loss": 1.4897,
      "step": 52670
    },
    {
      "epoch": 2.4162108151133843,
      "grad_norm": 0.42489802837371826,
      "learning_rate": 9.593726695958313e-05,
      "loss": 1.4774,
      "step": 52680
    },
    {
      "epoch": 2.4167527562788966,
      "grad_norm": 0.17096111178398132,
      "learning_rate": 9.593068795326312e-05,
      "loss": 1.4863,
      "step": 52690
    },
    {
      "epoch": 2.4172946974444085,
      "grad_norm": 0.19619131088256836,
      "learning_rate": 9.592410387662717e-05,
      "loss": 1.4682,
      "step": 52700
    },
    {
      "epoch": 2.417836638609921,
      "grad_norm": 0.18955209851264954,
      "learning_rate": 9.591751473049095e-05,
      "loss": 1.4765,
      "step": 52710
    },
    {
      "epoch": 2.418378579775433,
      "grad_norm": 0.1831616461277008,
      "learning_rate": 9.591092051567063e-05,
      "loss": 1.4783,
      "step": 52720
    },
    {
      "epoch": 2.4184869680085357,
      "eval_loss": 2.446958303451538,
      "eval_runtime": 22.1086,
      "eval_samples_per_second": 226.156,
      "eval_steps_per_second": 1.221,
      "step": 52722
    },
    {
      "epoch": 2.4189205209409455,
      "grad_norm": 0.2192278653383255,
      "learning_rate": 9.590432123298307e-05,
      "loss": 1.4874,
      "step": 52730
    },
    {
      "epoch": 2.4194624621064573,
      "grad_norm": 0.19387726485729218,
      "learning_rate": 9.589771688324582e-05,
      "loss": 1.4803,
      "step": 52740
    },
    {
      "epoch": 2.4200044032719696,
      "grad_norm": 0.280454158782959,
      "learning_rate": 9.589110746727692e-05,
      "loss": 1.4762,
      "step": 52750
    },
    {
      "epoch": 2.420546344437482,
      "grad_norm": 0.18469981849193573,
      "learning_rate": 9.588449298589518e-05,
      "loss": 1.478,
      "step": 52760
    },
    {
      "epoch": 2.4210882856029943,
      "grad_norm": 0.21699224412441254,
      "learning_rate": 9.587787343991996e-05,
      "loss": 1.473,
      "step": 52770
    },
    {
      "epoch": 2.4216302267685066,
      "grad_norm": 0.373770147562027,
      "learning_rate": 9.587124883017126e-05,
      "loss": 1.4781,
      "step": 52780
    },
    {
      "epoch": 2.4221721679340185,
      "grad_norm": 0.305627703666687,
      "learning_rate": 9.586461915746968e-05,
      "loss": 1.4829,
      "step": 52790
    },
    {
      "epoch": 2.4227141090995308,
      "grad_norm": 0.23528113961219788,
      "learning_rate": 9.585798442263651e-05,
      "loss": 1.4835,
      "step": 52800
    },
    {
      "epoch": 2.423201856148492,
      "eval_loss": 2.444547653198242,
      "eval_runtime": 22.0829,
      "eval_samples_per_second": 226.42,
      "eval_steps_per_second": 1.223,
      "step": 52809
    },
    {
      "epoch": 2.423256050265043,
      "grad_norm": 0.33448222279548645,
      "learning_rate": 9.58513446264936e-05,
      "loss": 1.4654,
      "step": 52810
    },
    {
      "epoch": 2.4237979914305554,
      "grad_norm": 0.2976095974445343,
      "learning_rate": 9.584469976986349e-05,
      "loss": 1.4784,
      "step": 52820
    },
    {
      "epoch": 2.4243399325960677,
      "grad_norm": 0.19375227391719818,
      "learning_rate": 9.58380498535693e-05,
      "loss": 1.4728,
      "step": 52830
    },
    {
      "epoch": 2.4248818737615796,
      "grad_norm": 0.30406102538108826,
      "learning_rate": 9.583139487843479e-05,
      "loss": 1.4812,
      "step": 52840
    },
    {
      "epoch": 2.425423814927092,
      "grad_norm": 0.3576555550098419,
      "learning_rate": 9.582473484528436e-05,
      "loss": 1.4949,
      "step": 52850
    },
    {
      "epoch": 2.4259657560926042,
      "grad_norm": 0.21541939675807953,
      "learning_rate": 9.581806975494303e-05,
      "loss": 1.4831,
      "step": 52860
    },
    {
      "epoch": 2.4265076972581165,
      "grad_norm": 0.20319156348705292,
      "learning_rate": 9.581139960823642e-05,
      "loss": 1.471,
      "step": 52870
    },
    {
      "epoch": 2.427049638423629,
      "grad_norm": 0.3167076110839844,
      "learning_rate": 9.58047244059908e-05,
      "loss": 1.4771,
      "step": 52880
    },
    {
      "epoch": 2.4275915795891407,
      "grad_norm": 0.41747766733169556,
      "learning_rate": 9.579804414903311e-05,
      "loss": 1.4705,
      "step": 52890
    },
    {
      "epoch": 2.4279167442884484,
      "eval_loss": 2.454636573791504,
      "eval_runtime": 21.9836,
      "eval_samples_per_second": 227.442,
      "eval_steps_per_second": 1.228,
      "step": 52896
    },
    {
      "epoch": 2.428133520754653,
      "grad_norm": 0.37934955954551697,
      "learning_rate": 9.579135883819082e-05,
      "loss": 1.4835,
      "step": 52900
    },
    {
      "epoch": 2.4286754619201654,
      "grad_norm": 0.1990528702735901,
      "learning_rate": 9.578466847429208e-05,
      "loss": 1.4811,
      "step": 52910
    },
    {
      "epoch": 2.4292174030856777,
      "grad_norm": 0.1967354118824005,
      "learning_rate": 9.57779730581657e-05,
      "loss": 1.4862,
      "step": 52920
    },
    {
      "epoch": 2.4297593442511896,
      "grad_norm": 0.22842636704444885,
      "learning_rate": 9.577127259064106e-05,
      "loss": 1.4771,
      "step": 52930
    },
    {
      "epoch": 2.430301285416702,
      "grad_norm": 0.34596872329711914,
      "learning_rate": 9.576456707254817e-05,
      "loss": 1.4861,
      "step": 52940
    },
    {
      "epoch": 2.430843226582214,
      "grad_norm": 0.2179035097360611,
      "learning_rate": 9.57578565047177e-05,
      "loss": 1.4817,
      "step": 52950
    },
    {
      "epoch": 2.4313851677477265,
      "grad_norm": 0.3425596356391907,
      "learning_rate": 9.57511408879809e-05,
      "loss": 1.4933,
      "step": 52960
    },
    {
      "epoch": 2.4319271089132384,
      "grad_norm": 0.33133429288864136,
      "learning_rate": 9.574442022316972e-05,
      "loss": 1.4704,
      "step": 52970
    },
    {
      "epoch": 2.4324690500787507,
      "grad_norm": 0.2703513205051422,
      "learning_rate": 9.573769451111665e-05,
      "loss": 1.4756,
      "step": 52980
    },
    {
      "epoch": 2.4326316324284045,
      "eval_loss": 2.4538090229034424,
      "eval_runtime": 22.1163,
      "eval_samples_per_second": 226.078,
      "eval_steps_per_second": 1.221,
      "step": 52983
    },
    {
      "epoch": 2.433010991244263,
      "grad_norm": 0.2822078764438629,
      "learning_rate": 9.573096375265484e-05,
      "loss": 1.4883,
      "step": 52990
    },
    {
      "epoch": 2.4335529324097753,
      "grad_norm": 0.2804018259048462,
      "learning_rate": 9.572422794861808e-05,
      "loss": 1.4807,
      "step": 53000
    },
    {
      "epoch": 2.4340948735752876,
      "grad_norm": 0.41560617089271545,
      "learning_rate": 9.571748709984076e-05,
      "loss": 1.4775,
      "step": 53010
    },
    {
      "epoch": 2.4346368147407995,
      "grad_norm": 0.43294498324394226,
      "learning_rate": 9.571074120715794e-05,
      "loss": 1.4794,
      "step": 53020
    },
    {
      "epoch": 2.435178755906312,
      "grad_norm": 0.22881726920604706,
      "learning_rate": 9.570399027140523e-05,
      "loss": 1.4722,
      "step": 53030
    },
    {
      "epoch": 2.435720697071824,
      "grad_norm": 0.25796639919281006,
      "learning_rate": 9.569723429341892e-05,
      "loss": 1.4782,
      "step": 53040
    },
    {
      "epoch": 2.4362626382373365,
      "grad_norm": 0.2922258973121643,
      "learning_rate": 9.569047327403593e-05,
      "loss": 1.4862,
      "step": 53050
    },
    {
      "epoch": 2.436804579402849,
      "grad_norm": 0.2601679265499115,
      "learning_rate": 9.568370721409376e-05,
      "loss": 1.4967,
      "step": 53060
    },
    {
      "epoch": 2.4373465205683607,
      "grad_norm": 0.2096012830734253,
      "learning_rate": 9.567693611443057e-05,
      "loss": 1.4771,
      "step": 53070
    },
    {
      "epoch": 2.4373465205683607,
      "eval_loss": 2.448545217514038,
      "eval_runtime": 22.001,
      "eval_samples_per_second": 227.262,
      "eval_steps_per_second": 1.227,
      "step": 53070
    },
    {
      "epoch": 2.437888461733873,
      "grad_norm": 0.18229645490646362,
      "learning_rate": 9.567015997588516e-05,
      "loss": 1.4764,
      "step": 53080
    },
    {
      "epoch": 2.4384304028993853,
      "grad_norm": 0.20466655492782593,
      "learning_rate": 9.566337879929687e-05,
      "loss": 1.478,
      "step": 53090
    },
    {
      "epoch": 2.4389723440648976,
      "grad_norm": 0.27803587913513184,
      "learning_rate": 9.565659258550576e-05,
      "loss": 1.4613,
      "step": 53100
    },
    {
      "epoch": 2.43951428523041,
      "grad_norm": 0.2388664335012436,
      "learning_rate": 9.564980133535249e-05,
      "loss": 1.4798,
      "step": 53110
    },
    {
      "epoch": 2.440056226395922,
      "grad_norm": 0.20684485137462616,
      "learning_rate": 9.564300504967832e-05,
      "loss": 1.4779,
      "step": 53120
    },
    {
      "epoch": 2.440598167561434,
      "grad_norm": 0.22506524622440338,
      "learning_rate": 9.563620372932513e-05,
      "loss": 1.4851,
      "step": 53130
    },
    {
      "epoch": 2.4411401087269464,
      "grad_norm": 0.2067488580942154,
      "learning_rate": 9.562939737513544e-05,
      "loss": 1.476,
      "step": 53140
    },
    {
      "epoch": 2.4416820498924587,
      "grad_norm": 0.27859237790107727,
      "learning_rate": 9.56225859879524e-05,
      "loss": 1.4776,
      "step": 53150
    },
    {
      "epoch": 2.4420614087083172,
      "eval_loss": 2.4477126598358154,
      "eval_runtime": 22.0385,
      "eval_samples_per_second": 226.876,
      "eval_steps_per_second": 1.225,
      "step": 53157
    },
    {
      "epoch": 2.4422239910579706,
      "grad_norm": 0.31960025429725647,
      "learning_rate": 9.561576956861978e-05,
      "loss": 1.4773,
      "step": 53160
    },
    {
      "epoch": 2.442765932223483,
      "grad_norm": 0.23332948982715607,
      "learning_rate": 9.560894811798198e-05,
      "loss": 1.4791,
      "step": 53170
    },
    {
      "epoch": 2.4433078733889952,
      "grad_norm": 0.20001184940338135,
      "learning_rate": 9.560212163688395e-05,
      "loss": 1.4825,
      "step": 53180
    },
    {
      "epoch": 2.4438498145545076,
      "grad_norm": 0.2940216064453125,
      "learning_rate": 9.559529012617141e-05,
      "loss": 1.4872,
      "step": 53190
    },
    {
      "epoch": 2.4443917557200194,
      "grad_norm": 0.25465646386146545,
      "learning_rate": 9.558845358669055e-05,
      "loss": 1.4896,
      "step": 53200
    },
    {
      "epoch": 2.4449336968855317,
      "grad_norm": 0.2544267177581787,
      "learning_rate": 9.55816120192883e-05,
      "loss": 1.4844,
      "step": 53210
    },
    {
      "epoch": 2.445475638051044,
      "grad_norm": 0.30805036425590515,
      "learning_rate": 9.557476542481212e-05,
      "loss": 1.4777,
      "step": 53220
    },
    {
      "epoch": 2.4460175792165564,
      "grad_norm": 0.23450466990470886,
      "learning_rate": 9.556791380411017e-05,
      "loss": 1.486,
      "step": 53230
    },
    {
      "epoch": 2.4465595203820687,
      "grad_norm": 0.2862785756587982,
      "learning_rate": 9.556105715803116e-05,
      "loss": 1.4807,
      "step": 53240
    },
    {
      "epoch": 2.4467762968482734,
      "eval_loss": 2.448235273361206,
      "eval_runtime": 22.0612,
      "eval_samples_per_second": 226.642,
      "eval_steps_per_second": 1.224,
      "step": 53244
    },
    {
      "epoch": 2.4471014615475806,
      "grad_norm": 0.23192840814590454,
      "learning_rate": 9.555419548742452e-05,
      "loss": 1.4753,
      "step": 53250
    },
    {
      "epoch": 2.447643402713093,
      "grad_norm": 0.3470098078250885,
      "learning_rate": 9.554732879314019e-05,
      "loss": 1.4847,
      "step": 53260
    },
    {
      "epoch": 2.448185343878605,
      "grad_norm": 0.3917534351348877,
      "learning_rate": 9.554045707602882e-05,
      "loss": 1.481,
      "step": 53270
    },
    {
      "epoch": 2.4487272850441175,
      "grad_norm": 0.36082231998443604,
      "learning_rate": 9.553358033694164e-05,
      "loss": 1.4753,
      "step": 53280
    },
    {
      "epoch": 2.44926922620963,
      "grad_norm": 0.24314887821674347,
      "learning_rate": 9.552669857673049e-05,
      "loss": 1.488,
      "step": 53290
    },
    {
      "epoch": 2.4498111673751417,
      "grad_norm": 0.27214857935905457,
      "learning_rate": 9.551981179624789e-05,
      "loss": 1.4761,
      "step": 53300
    },
    {
      "epoch": 2.450353108540654,
      "grad_norm": 0.22485631704330444,
      "learning_rate": 9.55129199963469e-05,
      "loss": 1.4849,
      "step": 53310
    },
    {
      "epoch": 2.4508950497061663,
      "grad_norm": 0.2175171673297882,
      "learning_rate": 9.55060231778813e-05,
      "loss": 1.4772,
      "step": 53320
    },
    {
      "epoch": 2.4514369908716787,
      "grad_norm": 0.5146223306655884,
      "learning_rate": 9.549912134170539e-05,
      "loss": 1.4912,
      "step": 53330
    },
    {
      "epoch": 2.45149118498823,
      "eval_loss": 2.4793365001678467,
      "eval_runtime": 22.2104,
      "eval_samples_per_second": 225.12,
      "eval_steps_per_second": 1.216,
      "step": 53331
    },
    {
      "epoch": 2.4519789320371905,
      "grad_norm": 0.2823374271392822,
      "learning_rate": 9.549221448867415e-05,
      "loss": 1.4967,
      "step": 53340
    },
    {
      "epoch": 2.452520873202703,
      "grad_norm": 0.26054081320762634,
      "learning_rate": 9.54853026196432e-05,
      "loss": 1.4803,
      "step": 53350
    },
    {
      "epoch": 2.453062814368215,
      "grad_norm": 0.27415192127227783,
      "learning_rate": 9.547838573546872e-05,
      "loss": 1.487,
      "step": 53360
    },
    {
      "epoch": 2.4536047555337275,
      "grad_norm": 0.19018392264842987,
      "learning_rate": 9.547146383700756e-05,
      "loss": 1.4945,
      "step": 53370
    },
    {
      "epoch": 2.4541466966992393,
      "grad_norm": 0.23459668457508087,
      "learning_rate": 9.546453692511715e-05,
      "loss": 1.4807,
      "step": 53380
    },
    {
      "epoch": 2.4546886378647517,
      "grad_norm": 0.30195194482803345,
      "learning_rate": 9.545760500065562e-05,
      "loss": 1.4858,
      "step": 53390
    },
    {
      "epoch": 2.455230579030264,
      "grad_norm": 0.23426523804664612,
      "learning_rate": 9.54506680644816e-05,
      "loss": 1.4805,
      "step": 53400
    },
    {
      "epoch": 2.4557725201957763,
      "grad_norm": 0.23431915044784546,
      "learning_rate": 9.544372611745444e-05,
      "loss": 1.4824,
      "step": 53410
    },
    {
      "epoch": 2.456206073128186,
      "eval_loss": 2.459233283996582,
      "eval_runtime": 22.0413,
      "eval_samples_per_second": 226.847,
      "eval_steps_per_second": 1.225,
      "step": 53418
    },
    {
      "epoch": 2.4563144613612886,
      "grad_norm": 0.19480261206626892,
      "learning_rate": 9.54367791604341e-05,
      "loss": 1.4798,
      "step": 53420
    },
    {
      "epoch": 2.4568564025268005,
      "grad_norm": 0.23288105428218842,
      "learning_rate": 9.54298271942811e-05,
      "loss": 1.4837,
      "step": 53430
    },
    {
      "epoch": 2.457398343692313,
      "grad_norm": 0.21174949407577515,
      "learning_rate": 9.542287021985665e-05,
      "loss": 1.4851,
      "step": 53440
    },
    {
      "epoch": 2.457940284857825,
      "grad_norm": 0.4950149953365326,
      "learning_rate": 9.541590823802252e-05,
      "loss": 1.4755,
      "step": 53450
    },
    {
      "epoch": 2.4584822260233374,
      "grad_norm": 0.2723096013069153,
      "learning_rate": 9.540894124964115e-05,
      "loss": 1.4802,
      "step": 53460
    },
    {
      "epoch": 2.4590241671888498,
      "grad_norm": 0.17027664184570312,
      "learning_rate": 9.54019692555756e-05,
      "loss": 1.4746,
      "step": 53470
    },
    {
      "epoch": 2.4595661083543616,
      "grad_norm": 0.30771180987358093,
      "learning_rate": 9.539499225668948e-05,
      "loss": 1.4835,
      "step": 53480
    },
    {
      "epoch": 2.460108049519874,
      "grad_norm": 0.21357007324695587,
      "learning_rate": 9.538801025384709e-05,
      "loss": 1.4775,
      "step": 53490
    },
    {
      "epoch": 2.4606499906853863,
      "grad_norm": 0.28452086448669434,
      "learning_rate": 9.538102324791336e-05,
      "loss": 1.4832,
      "step": 53500
    },
    {
      "epoch": 2.460920961268142,
      "eval_loss": 2.4530394077301025,
      "eval_runtime": 22.1982,
      "eval_samples_per_second": 225.244,
      "eval_steps_per_second": 1.216,
      "step": 53505
    },
    {
      "epoch": 2.4611919318508986,
      "grad_norm": 0.20882080495357513,
      "learning_rate": 9.537403123975378e-05,
      "loss": 1.4729,
      "step": 53510
    },
    {
      "epoch": 2.461733873016411,
      "grad_norm": 0.2761918306350708,
      "learning_rate": 9.536703423023449e-05,
      "loss": 1.4739,
      "step": 53520
    },
    {
      "epoch": 2.4622758141819228,
      "grad_norm": 0.1824912279844284,
      "learning_rate": 9.536003222022225e-05,
      "loss": 1.48,
      "step": 53530
    },
    {
      "epoch": 2.462817755347435,
      "grad_norm": 0.22570142149925232,
      "learning_rate": 9.535302521058445e-05,
      "loss": 1.4832,
      "step": 53540
    },
    {
      "epoch": 2.4633596965129474,
      "grad_norm": 0.3082813620567322,
      "learning_rate": 9.534601320218909e-05,
      "loss": 1.4899,
      "step": 53550
    },
    {
      "epoch": 2.4639016376784597,
      "grad_norm": 0.17578811943531036,
      "learning_rate": 9.533899619590477e-05,
      "loss": 1.4744,
      "step": 53560
    },
    {
      "epoch": 2.4644435788439716,
      "grad_norm": 0.20538325607776642,
      "learning_rate": 9.533197419260073e-05,
      "loss": 1.4834,
      "step": 53570
    },
    {
      "epoch": 2.464985520009484,
      "grad_norm": 0.2888244688510895,
      "learning_rate": 9.532494719314686e-05,
      "loss": 1.4833,
      "step": 53580
    },
    {
      "epoch": 2.465527461174996,
      "grad_norm": 0.26800277829170227,
      "learning_rate": 9.531791519841359e-05,
      "loss": 1.4856,
      "step": 53590
    },
    {
      "epoch": 2.4656358494080988,
      "eval_loss": 2.4533004760742188,
      "eval_runtime": 22.8599,
      "eval_samples_per_second": 218.724,
      "eval_steps_per_second": 1.181,
      "step": 53592
    },
    {
      "epoch": 2.4660694023405085,
      "grad_norm": 0.24594108760356903,
      "learning_rate": 9.531087820927201e-05,
      "loss": 1.4797,
      "step": 53600
    },
    {
      "epoch": 2.4666113435060204,
      "grad_norm": 0.18168674409389496,
      "learning_rate": 9.530383622659386e-05,
      "loss": 1.4837,
      "step": 53610
    },
    {
      "epoch": 2.4671532846715327,
      "grad_norm": 0.20581193268299103,
      "learning_rate": 9.529678925125148e-05,
      "loss": 1.4761,
      "step": 53620
    },
    {
      "epoch": 2.467695225837045,
      "grad_norm": 0.33261099457740784,
      "learning_rate": 9.528973728411778e-05,
      "loss": 1.4789,
      "step": 53630
    },
    {
      "epoch": 2.4682371670025574,
      "grad_norm": 0.3050486147403717,
      "learning_rate": 9.528268032606636e-05,
      "loss": 1.4759,
      "step": 53640
    },
    {
      "epoch": 2.4687791081680697,
      "grad_norm": 0.24290646612644196,
      "learning_rate": 9.527561837797136e-05,
      "loss": 1.4944,
      "step": 53650
    },
    {
      "epoch": 2.4693210493335815,
      "grad_norm": 0.26528191566467285,
      "learning_rate": 9.526855144070763e-05,
      "loss": 1.4721,
      "step": 53660
    },
    {
      "epoch": 2.469862990499094,
      "grad_norm": 0.1806386113166809,
      "learning_rate": 9.52614795151506e-05,
      "loss": 1.4691,
      "step": 53670
    },
    {
      "epoch": 2.470350737548055,
      "eval_loss": 2.45682430267334,
      "eval_runtime": 21.9989,
      "eval_samples_per_second": 227.284,
      "eval_steps_per_second": 1.227,
      "step": 53679
    },
    {
      "epoch": 2.470404931664606,
      "grad_norm": 0.27701669931411743,
      "learning_rate": 9.525440260217627e-05,
      "loss": 1.4868,
      "step": 53680
    },
    {
      "epoch": 2.4709468728301185,
      "grad_norm": 0.23009248077869415,
      "learning_rate": 9.52473207026613e-05,
      "loss": 1.4797,
      "step": 53690
    },
    {
      "epoch": 2.471488813995631,
      "grad_norm": 0.19565734267234802,
      "learning_rate": 9.524023381748298e-05,
      "loss": 1.4742,
      "step": 53700
    },
    {
      "epoch": 2.4720307551611427,
      "grad_norm": 0.3481823801994324,
      "learning_rate": 9.523314194751921e-05,
      "loss": 1.4823,
      "step": 53710
    },
    {
      "epoch": 2.472572696326655,
      "grad_norm": 0.1881093680858612,
      "learning_rate": 9.522604509364849e-05,
      "loss": 1.4707,
      "step": 53720
    },
    {
      "epoch": 2.4731146374921673,
      "grad_norm": 0.4059028625488281,
      "learning_rate": 9.521894325674994e-05,
      "loss": 1.4749,
      "step": 53730
    },
    {
      "epoch": 2.4736565786576796,
      "grad_norm": 0.3592798709869385,
      "learning_rate": 9.521183643770333e-05,
      "loss": 1.477,
      "step": 53740
    },
    {
      "epoch": 2.474198519823192,
      "grad_norm": 0.2864842712879181,
      "learning_rate": 9.520472463738899e-05,
      "loss": 1.4784,
      "step": 53750
    },
    {
      "epoch": 2.474740460988704,
      "grad_norm": 0.20083869993686676,
      "learning_rate": 9.519760785668791e-05,
      "loss": 1.4775,
      "step": 53760
    },
    {
      "epoch": 2.4750656256880115,
      "eval_loss": 2.4518940448760986,
      "eval_runtime": 33.4746,
      "eval_samples_per_second": 149.367,
      "eval_steps_per_second": 0.807,
      "step": 53766
    },
    {
      "epoch": 2.475282402154216,
      "grad_norm": 0.25543951988220215,
      "learning_rate": 9.519048609648169e-05,
      "loss": 1.4624,
      "step": 53770
    },
    {
      "epoch": 2.4758243433197284,
      "grad_norm": 0.2484443336725235,
      "learning_rate": 9.518335935765256e-05,
      "loss": 1.483,
      "step": 53780
    },
    {
      "epoch": 2.4763662844852408,
      "grad_norm": 0.3281131088733673,
      "learning_rate": 9.517622764108331e-05,
      "loss": 1.4762,
      "step": 53790
    },
    {
      "epoch": 2.4769082256507526,
      "grad_norm": 0.23684971034526825,
      "learning_rate": 9.516909094765741e-05,
      "loss": 1.4772,
      "step": 53800
    },
    {
      "epoch": 2.477450166816265,
      "grad_norm": 0.2768543064594269,
      "learning_rate": 9.516194927825894e-05,
      "loss": 1.4877,
      "step": 53810
    },
    {
      "epoch": 2.4779921079817773,
      "grad_norm": 0.20820960402488708,
      "learning_rate": 9.515480263377253e-05,
      "loss": 1.4711,
      "step": 53820
    },
    {
      "epoch": 2.4785340491472896,
      "grad_norm": 0.18678176403045654,
      "learning_rate": 9.514765101508353e-05,
      "loss": 1.4923,
      "step": 53830
    },
    {
      "epoch": 2.4790759903128015,
      "grad_norm": 0.22122722864151,
      "learning_rate": 9.514049442307782e-05,
      "loss": 1.4837,
      "step": 53840
    },
    {
      "epoch": 2.4796179314783138,
      "grad_norm": 0.3165682256221771,
      "learning_rate": 9.513333285864192e-05,
      "loss": 1.4856,
      "step": 53850
    },
    {
      "epoch": 2.4797805138279676,
      "eval_loss": 2.4556586742401123,
      "eval_runtime": 33.9488,
      "eval_samples_per_second": 147.281,
      "eval_steps_per_second": 0.795,
      "step": 53853
    },
    {
      "epoch": 2.480159872643826,
      "grad_norm": 0.2486167550086975,
      "learning_rate": 9.5126166322663e-05,
      "loss": 1.4829,
      "step": 53860
    },
    {
      "epoch": 2.4807018138093384,
      "grad_norm": 0.21120484173297882,
      "learning_rate": 9.51189948160288e-05,
      "loss": 1.479,
      "step": 53870
    },
    {
      "epoch": 2.4812437549748507,
      "grad_norm": 0.18009105324745178,
      "learning_rate": 9.511181833962772e-05,
      "loss": 1.4698,
      "step": 53880
    },
    {
      "epoch": 2.4817856961403626,
      "grad_norm": 0.19406986236572266,
      "learning_rate": 9.51046368943487e-05,
      "loss": 1.484,
      "step": 53890
    },
    {
      "epoch": 2.482327637305875,
      "grad_norm": 0.22672545909881592,
      "learning_rate": 9.50974504810814e-05,
      "loss": 1.4773,
      "step": 53900
    },
    {
      "epoch": 2.4828695784713872,
      "grad_norm": 0.21448560059070587,
      "learning_rate": 9.509025910071602e-05,
      "loss": 1.4781,
      "step": 53910
    },
    {
      "epoch": 2.4834115196368995,
      "grad_norm": 0.3304313123226166,
      "learning_rate": 9.508306275414339e-05,
      "loss": 1.472,
      "step": 53920
    },
    {
      "epoch": 2.483953460802412,
      "grad_norm": 0.29814818501472473,
      "learning_rate": 9.507586144225497e-05,
      "loss": 1.4737,
      "step": 53930
    },
    {
      "epoch": 2.4844954019679237,
      "grad_norm": 0.20384559035301208,
      "learning_rate": 9.506865516594282e-05,
      "loss": 1.4748,
      "step": 53940
    },
    {
      "epoch": 2.4844954019679237,
      "eval_loss": 2.4536936283111572,
      "eval_runtime": 21.8133,
      "eval_samples_per_second": 229.218,
      "eval_steps_per_second": 1.238,
      "step": 53940
    },
    {
      "epoch": 2.485037343133436,
      "grad_norm": 0.1892993301153183,
      "learning_rate": 9.506144392609965e-05,
      "loss": 1.4846,
      "step": 53950
    },
    {
      "epoch": 2.4855792842989484,
      "grad_norm": 0.37276574969291687,
      "learning_rate": 9.505422772361872e-05,
      "loss": 1.4809,
      "step": 53960
    },
    {
      "epoch": 2.4861212254644607,
      "grad_norm": 0.5390621423721313,
      "learning_rate": 9.504700655939396e-05,
      "loss": 1.4752,
      "step": 53970
    },
    {
      "epoch": 2.4866631666299726,
      "grad_norm": 0.26775336265563965,
      "learning_rate": 9.50397804343199e-05,
      "loss": 1.4858,
      "step": 53980
    },
    {
      "epoch": 2.487205107795485,
      "grad_norm": 0.18470676243305206,
      "learning_rate": 9.503254934929165e-05,
      "loss": 1.4741,
      "step": 53990
    },
    {
      "epoch": 2.487747048960997,
      "grad_norm": 0.24282990396022797,
      "learning_rate": 9.502531330520501e-05,
      "loss": 1.4781,
      "step": 54000
    },
    {
      "epoch": 2.4882889901265095,
      "grad_norm": 0.19152526557445526,
      "learning_rate": 9.501807230295634e-05,
      "loss": 1.4714,
      "step": 54010
    },
    {
      "epoch": 2.4888309312920214,
      "grad_norm": 0.24509835243225098,
      "learning_rate": 9.50108263434426e-05,
      "loss": 1.4673,
      "step": 54020
    },
    {
      "epoch": 2.4892102901078803,
      "eval_loss": 2.4450418949127197,
      "eval_runtime": 21.9853,
      "eval_samples_per_second": 227.425,
      "eval_steps_per_second": 1.228,
      "step": 54027
    },
    {
      "epoch": 2.4893728724575337,
      "grad_norm": 0.20996253192424774,
      "learning_rate": 9.500357542756139e-05,
      "loss": 1.4696,
      "step": 54030
    },
    {
      "epoch": 2.489914813623046,
      "grad_norm": 0.3065512776374817,
      "learning_rate": 9.499631955621097e-05,
      "loss": 1.4837,
      "step": 54040
    },
    {
      "epoch": 2.4904567547885583,
      "grad_norm": 0.2927444577217102,
      "learning_rate": 9.498905873029012e-05,
      "loss": 1.4833,
      "step": 54050
    },
    {
      "epoch": 2.4909986959540706,
      "grad_norm": 0.3556462228298187,
      "learning_rate": 9.498179295069827e-05,
      "loss": 1.4744,
      "step": 54060
    },
    {
      "epoch": 2.4915406371195825,
      "grad_norm": 0.3211055099964142,
      "learning_rate": 9.49745222183355e-05,
      "loss": 1.4723,
      "step": 54070
    },
    {
      "epoch": 2.492082578285095,
      "grad_norm": 0.510176956653595,
      "learning_rate": 9.496724653410249e-05,
      "loss": 1.4706,
      "step": 54080
    },
    {
      "epoch": 2.492624519450607,
      "grad_norm": 0.20977604389190674,
      "learning_rate": 9.495996589890048e-05,
      "loss": 1.4692,
      "step": 54090
    },
    {
      "epoch": 2.4931664606161195,
      "grad_norm": 0.19488461315631866,
      "learning_rate": 9.495268031363138e-05,
      "loss": 1.4821,
      "step": 54100
    },
    {
      "epoch": 2.493708401781632,
      "grad_norm": 0.21856850385665894,
      "learning_rate": 9.494538977919771e-05,
      "loss": 1.4704,
      "step": 54110
    },
    {
      "epoch": 2.4939251782478364,
      "eval_loss": 2.4471306800842285,
      "eval_runtime": 21.9864,
      "eval_samples_per_second": 227.414,
      "eval_steps_per_second": 1.228,
      "step": 54114
    },
    {
      "epoch": 2.4942503429471437,
      "grad_norm": 0.21702884137630463,
      "learning_rate": 9.493809429650256e-05,
      "loss": 1.4802,
      "step": 54120
    },
    {
      "epoch": 2.494792284112656,
      "grad_norm": 0.17151543498039246,
      "learning_rate": 9.49307938664497e-05,
      "loss": 1.4798,
      "step": 54130
    },
    {
      "epoch": 2.4953342252781683,
      "grad_norm": 0.20608776807785034,
      "learning_rate": 9.492348848994345e-05,
      "loss": 1.4736,
      "step": 54140
    },
    {
      "epoch": 2.4958761664436806,
      "grad_norm": 0.29367008805274963,
      "learning_rate": 9.491617816788879e-05,
      "loss": 1.4799,
      "step": 54150
    },
    {
      "epoch": 2.496418107609193,
      "grad_norm": 0.38167864084243774,
      "learning_rate": 9.490886290119128e-05,
      "loss": 1.4636,
      "step": 54160
    },
    {
      "epoch": 2.496960048774705,
      "grad_norm": 0.18490886688232422,
      "learning_rate": 9.490154269075708e-05,
      "loss": 1.4694,
      "step": 54170
    },
    {
      "epoch": 2.497501989940217,
      "grad_norm": 0.44868624210357666,
      "learning_rate": 9.489421753749303e-05,
      "loss": 1.4783,
      "step": 54180
    },
    {
      "epoch": 2.4980439311057294,
      "grad_norm": 0.34574294090270996,
      "learning_rate": 9.488688744230649e-05,
      "loss": 1.4775,
      "step": 54190
    },
    {
      "epoch": 2.4985858722712417,
      "grad_norm": 0.34734290838241577,
      "learning_rate": 9.487955240610553e-05,
      "loss": 1.4725,
      "step": 54200
    },
    {
      "epoch": 2.498640066387793,
      "eval_loss": 2.442676305770874,
      "eval_runtime": 21.9936,
      "eval_samples_per_second": 227.339,
      "eval_steps_per_second": 1.228,
      "step": 54201
    },
    {
      "epoch": 2.4991278134367536,
      "grad_norm": 0.18172703683376312,
      "learning_rate": 9.487221242979874e-05,
      "loss": 1.4862,
      "step": 54210
    },
    {
      "epoch": 2.499669754602266,
      "grad_norm": 0.4136309027671814,
      "learning_rate": 9.486486751429539e-05,
      "loss": 1.4723,
      "step": 54220
    },
    {
      "epoch": 2.5002116957677782,
      "grad_norm": 0.2756052315235138,
      "learning_rate": 9.485751766050533e-05,
      "loss": 1.4846,
      "step": 54230
    },
    {
      "epoch": 2.5007536369332906,
      "grad_norm": 0.34378984570503235,
      "learning_rate": 9.485016286933902e-05,
      "loss": 1.4631,
      "step": 54240
    },
    {
      "epoch": 2.5012955780988024,
      "grad_norm": 0.2502152919769287,
      "learning_rate": 9.484280314170754e-05,
      "loss": 1.4684,
      "step": 54250
    },
    {
      "epoch": 2.5018375192643147,
      "grad_norm": 0.20602858066558838,
      "learning_rate": 9.483543847852258e-05,
      "loss": 1.461,
      "step": 54260
    },
    {
      "epoch": 2.502379460429827,
      "grad_norm": 0.19559136033058167,
      "learning_rate": 9.482806888069648e-05,
      "loss": 1.4821,
      "step": 54270
    },
    {
      "epoch": 2.5029214015953394,
      "grad_norm": 0.19327175617218018,
      "learning_rate": 9.48206943491421e-05,
      "loss": 1.4777,
      "step": 54280
    },
    {
      "epoch": 2.503354954527749,
      "eval_loss": 2.4406850337982178,
      "eval_runtime": 21.9849,
      "eval_samples_per_second": 227.429,
      "eval_steps_per_second": 1.228,
      "step": 54288
    },
    {
      "epoch": 2.5034633427608517,
      "grad_norm": 0.19267085194587708,
      "learning_rate": 9.4813314884773e-05,
      "loss": 1.48,
      "step": 54290
    },
    {
      "epoch": 2.5040052839263636,
      "grad_norm": 0.21109183132648468,
      "learning_rate": 9.48059304885033e-05,
      "loss": 1.4782,
      "step": 54300
    },
    {
      "epoch": 2.504547225091876,
      "grad_norm": 0.2512279152870178,
      "learning_rate": 9.479854116124775e-05,
      "loss": 1.4898,
      "step": 54310
    },
    {
      "epoch": 2.505089166257388,
      "grad_norm": 0.24899804592132568,
      "learning_rate": 9.479114690392172e-05,
      "loss": 1.4642,
      "step": 54320
    },
    {
      "epoch": 2.5056311074229005,
      "grad_norm": 0.20213483273983002,
      "learning_rate": 9.478374771744115e-05,
      "loss": 1.4686,
      "step": 54330
    },
    {
      "epoch": 2.506173048588413,
      "grad_norm": 0.1799444556236267,
      "learning_rate": 9.477634360272265e-05,
      "loss": 1.4745,
      "step": 54340
    },
    {
      "epoch": 2.5067149897539247,
      "grad_norm": 0.347083181142807,
      "learning_rate": 9.476893456068338e-05,
      "loss": 1.48,
      "step": 54350
    },
    {
      "epoch": 2.507256930919437,
      "grad_norm": 0.2862820625305176,
      "learning_rate": 9.476152059224117e-05,
      "loss": 1.4804,
      "step": 54360
    },
    {
      "epoch": 2.5077988720849493,
      "grad_norm": 0.285040020942688,
      "learning_rate": 9.475410169831442e-05,
      "loss": 1.4798,
      "step": 54370
    },
    {
      "epoch": 2.5080698426677053,
      "eval_loss": 2.4463353157043457,
      "eval_runtime": 22.1709,
      "eval_samples_per_second": 225.521,
      "eval_steps_per_second": 1.218,
      "step": 54375
    },
    {
      "epoch": 2.5083408132504617,
      "grad_norm": 0.2216855138540268,
      "learning_rate": 9.474667787982213e-05,
      "loss": 1.4854,
      "step": 54380
    },
    {
      "epoch": 2.508882754415974,
      "grad_norm": 0.307620108127594,
      "learning_rate": 9.473924913768396e-05,
      "loss": 1.4787,
      "step": 54390
    },
    {
      "epoch": 2.509424695581486,
      "grad_norm": 0.30470430850982666,
      "learning_rate": 9.473181547282013e-05,
      "loss": 1.4721,
      "step": 54400
    },
    {
      "epoch": 2.509966636746998,
      "grad_norm": 0.35153964161872864,
      "learning_rate": 9.47243768861515e-05,
      "loss": 1.4701,
      "step": 54410
    },
    {
      "epoch": 2.5105085779125105,
      "grad_norm": 0.24849040806293488,
      "learning_rate": 9.471693337859953e-05,
      "loss": 1.4609,
      "step": 54420
    },
    {
      "epoch": 2.5110505190780223,
      "grad_norm": 0.3956458270549774,
      "learning_rate": 9.470948495108628e-05,
      "loss": 1.4819,
      "step": 54430
    },
    {
      "epoch": 2.5115924602435347,
      "grad_norm": 0.21388791501522064,
      "learning_rate": 9.470203160453445e-05,
      "loss": 1.4825,
      "step": 54440
    },
    {
      "epoch": 2.512134401409047,
      "grad_norm": 0.35142725706100464,
      "learning_rate": 9.46945733398673e-05,
      "loss": 1.4787,
      "step": 54450
    },
    {
      "epoch": 2.5126763425745593,
      "grad_norm": 0.22184672951698303,
      "learning_rate": 9.468711015800874e-05,
      "loss": 1.4662,
      "step": 54460
    },
    {
      "epoch": 2.512784730807662,
      "eval_loss": 2.446195363998413,
      "eval_runtime": 22.0811,
      "eval_samples_per_second": 226.438,
      "eval_steps_per_second": 1.223,
      "step": 54462
    },
    {
      "epoch": 2.5132182837400716,
      "grad_norm": 0.27088162302970886,
      "learning_rate": 9.46796420598833e-05,
      "loss": 1.473,
      "step": 54470
    },
    {
      "epoch": 2.5137602249055835,
      "grad_norm": 0.2419961541891098,
      "learning_rate": 9.467216904641606e-05,
      "loss": 1.4824,
      "step": 54480
    },
    {
      "epoch": 2.514302166071096,
      "grad_norm": 0.2746829688549042,
      "learning_rate": 9.466469111853275e-05,
      "loss": 1.4729,
      "step": 54490
    },
    {
      "epoch": 2.514844107236608,
      "grad_norm": 0.2866189479827881,
      "learning_rate": 9.465720827715972e-05,
      "loss": 1.476,
      "step": 54500
    },
    {
      "epoch": 2.5153860484021204,
      "grad_norm": 0.19014035165309906,
      "learning_rate": 9.464972052322388e-05,
      "loss": 1.4803,
      "step": 54510
    },
    {
      "epoch": 2.5159279895676327,
      "grad_norm": 0.2546852231025696,
      "learning_rate": 9.464222785765284e-05,
      "loss": 1.4749,
      "step": 54520
    },
    {
      "epoch": 2.5164699307331446,
      "grad_norm": 0.3189975917339325,
      "learning_rate": 9.46347302813747e-05,
      "loss": 1.4838,
      "step": 54530
    },
    {
      "epoch": 2.517011871898657,
      "grad_norm": 0.25641825795173645,
      "learning_rate": 9.462722779531825e-05,
      "loss": 1.4775,
      "step": 54540
    },
    {
      "epoch": 2.517499618947618,
      "eval_loss": 2.4492218494415283,
      "eval_runtime": 22.0696,
      "eval_samples_per_second": 226.556,
      "eval_steps_per_second": 1.223,
      "step": 54549
    },
    {
      "epoch": 2.5175538130641693,
      "grad_norm": 0.36368492245674133,
      "learning_rate": 9.461972040041286e-05,
      "loss": 1.4834,
      "step": 54550
    },
    {
      "epoch": 2.5180957542296816,
      "grad_norm": 0.41073617339134216,
      "learning_rate": 9.461220809758854e-05,
      "loss": 1.475,
      "step": 54560
    },
    {
      "epoch": 2.518637695395194,
      "grad_norm": 0.42227840423583984,
      "learning_rate": 9.460469088777585e-05,
      "loss": 1.4782,
      "step": 54570
    },
    {
      "epoch": 2.5191796365607058,
      "grad_norm": 0.24830390512943268,
      "learning_rate": 9.459716877190599e-05,
      "loss": 1.4789,
      "step": 54580
    },
    {
      "epoch": 2.519721577726218,
      "grad_norm": 0.2122548371553421,
      "learning_rate": 9.458964175091078e-05,
      "loss": 1.4714,
      "step": 54590
    },
    {
      "epoch": 2.5202635188917304,
      "grad_norm": 0.1787809282541275,
      "learning_rate": 9.458210982572264e-05,
      "loss": 1.4732,
      "step": 54600
    },
    {
      "epoch": 2.5208054600572423,
      "grad_norm": 0.19361905753612518,
      "learning_rate": 9.457457299727458e-05,
      "loss": 1.4837,
      "step": 54610
    },
    {
      "epoch": 2.521347401222755,
      "grad_norm": 0.459349662065506,
      "learning_rate": 9.456703126650023e-05,
      "loss": 1.4909,
      "step": 54620
    },
    {
      "epoch": 2.521889342388267,
      "grad_norm": 0.31457701325416565,
      "learning_rate": 9.455948463433384e-05,
      "loss": 1.4623,
      "step": 54630
    },
    {
      "epoch": 2.5222145070875746,
      "eval_loss": 2.4459049701690674,
      "eval_runtime": 22.0427,
      "eval_samples_per_second": 226.832,
      "eval_steps_per_second": 1.225,
      "step": 54636
    },
    {
      "epoch": 2.522431283553779,
      "grad_norm": 0.2660439610481262,
      "learning_rate": 9.455193310171022e-05,
      "loss": 1.4708,
      "step": 54640
    },
    {
      "epoch": 2.5229732247192915,
      "grad_norm": 0.22614122927188873,
      "learning_rate": 9.454437666956486e-05,
      "loss": 1.4763,
      "step": 54650
    },
    {
      "epoch": 2.5235151658848034,
      "grad_norm": 0.23727241158485413,
      "learning_rate": 9.45368153388338e-05,
      "loss": 1.4875,
      "step": 54660
    },
    {
      "epoch": 2.5240571070503157,
      "grad_norm": 0.20123521983623505,
      "learning_rate": 9.452924911045372e-05,
      "loss": 1.4907,
      "step": 54670
    },
    {
      "epoch": 2.524599048215828,
      "grad_norm": 0.28257882595062256,
      "learning_rate": 9.452167798536186e-05,
      "loss": 1.4782,
      "step": 54680
    },
    {
      "epoch": 2.5251409893813404,
      "grad_norm": 0.16655373573303223,
      "learning_rate": 9.451410196449613e-05,
      "loss": 1.4794,
      "step": 54690
    },
    {
      "epoch": 2.5256829305468527,
      "grad_norm": 0.28501424193382263,
      "learning_rate": 9.450652104879499e-05,
      "loss": 1.4713,
      "step": 54700
    },
    {
      "epoch": 2.5262248717123645,
      "grad_norm": 0.24294120073318481,
      "learning_rate": 9.449893523919754e-05,
      "loss": 1.4853,
      "step": 54710
    },
    {
      "epoch": 2.526766812877877,
      "grad_norm": 0.5156522393226624,
      "learning_rate": 9.44913445366435e-05,
      "loss": 1.4686,
      "step": 54720
    },
    {
      "epoch": 2.5269293952275307,
      "eval_loss": 2.447787046432495,
      "eval_runtime": 22.1759,
      "eval_samples_per_second": 225.47,
      "eval_steps_per_second": 1.218,
      "step": 54723
    },
    {
      "epoch": 2.527308754043389,
      "grad_norm": 0.21539968252182007,
      "learning_rate": 9.448374894207314e-05,
      "loss": 1.4801,
      "step": 54730
    },
    {
      "epoch": 2.5278506952089015,
      "grad_norm": 0.268772155046463,
      "learning_rate": 9.447614845642738e-05,
      "loss": 1.4644,
      "step": 54740
    },
    {
      "epoch": 2.528392636374414,
      "grad_norm": 0.1783050000667572,
      "learning_rate": 9.446854308064774e-05,
      "loss": 1.4827,
      "step": 54750
    },
    {
      "epoch": 2.5289345775399257,
      "grad_norm": 0.17679934203624725,
      "learning_rate": 9.446093281567635e-05,
      "loss": 1.4651,
      "step": 54760
    },
    {
      "epoch": 2.529476518705438,
      "grad_norm": 0.2454293668270111,
      "learning_rate": 9.445331766245592e-05,
      "loss": 1.4663,
      "step": 54770
    },
    {
      "epoch": 2.5300184598709503,
      "grad_norm": 0.1850767731666565,
      "learning_rate": 9.444569762192977e-05,
      "loss": 1.4661,
      "step": 54780
    },
    {
      "epoch": 2.5305604010364626,
      "grad_norm": 0.2817839980125427,
      "learning_rate": 9.443807269504187e-05,
      "loss": 1.4777,
      "step": 54790
    },
    {
      "epoch": 2.531102342201975,
      "grad_norm": 0.2719174921512604,
      "learning_rate": 9.443044288273675e-05,
      "loss": 1.4745,
      "step": 54800
    },
    {
      "epoch": 2.531644283367487,
      "grad_norm": 0.7066235542297363,
      "learning_rate": 9.442280818595955e-05,
      "loss": 1.466,
      "step": 54810
    },
    {
      "epoch": 2.531644283367487,
      "eval_loss": 2.447768211364746,
      "eval_runtime": 21.9147,
      "eval_samples_per_second": 228.157,
      "eval_steps_per_second": 1.232,
      "step": 54810
    },
    {
      "epoch": 2.532186224532999,
      "grad_norm": 0.3116154968738556,
      "learning_rate": 9.441516860565602e-05,
      "loss": 1.4752,
      "step": 54820
    },
    {
      "epoch": 2.5327281656985114,
      "grad_norm": 0.29827791452407837,
      "learning_rate": 9.440752414277254e-05,
      "loss": 1.4729,
      "step": 54830
    },
    {
      "epoch": 2.5332701068640233,
      "grad_norm": 0.2395966500043869,
      "learning_rate": 9.439987479825607e-05,
      "loss": 1.4742,
      "step": 54840
    },
    {
      "epoch": 2.5338120480295356,
      "grad_norm": 0.21512676775455475,
      "learning_rate": 9.439222057305414e-05,
      "loss": 1.4761,
      "step": 54850
    },
    {
      "epoch": 2.534353989195048,
      "grad_norm": 0.19757066667079926,
      "learning_rate": 9.438456146811496e-05,
      "loss": 1.4736,
      "step": 54860
    },
    {
      "epoch": 2.5348959303605603,
      "grad_norm": 0.2103499174118042,
      "learning_rate": 9.43768974843873e-05,
      "loss": 1.4746,
      "step": 54870
    },
    {
      "epoch": 2.5354378715260726,
      "grad_norm": 0.19325312972068787,
      "learning_rate": 9.436922862282052e-05,
      "loss": 1.4561,
      "step": 54880
    },
    {
      "epoch": 2.5359798126915845,
      "grad_norm": 0.2294219732284546,
      "learning_rate": 9.436155488436464e-05,
      "loss": 1.4783,
      "step": 54890
    },
    {
      "epoch": 2.5363591715074434,
      "eval_loss": 2.4521801471710205,
      "eval_runtime": 22.0914,
      "eval_samples_per_second": 226.332,
      "eval_steps_per_second": 1.222,
      "step": 54897
    },
    {
      "epoch": 2.5365217538570968,
      "grad_norm": 0.40045541524887085,
      "learning_rate": 9.435387626997024e-05,
      "loss": 1.4763,
      "step": 54900
    },
    {
      "epoch": 2.537063695022609,
      "grad_norm": 0.2468022257089615,
      "learning_rate": 9.434619278058848e-05,
      "loss": 1.4752,
      "step": 54910
    },
    {
      "epoch": 2.5376056361881214,
      "grad_norm": 0.2977105975151062,
      "learning_rate": 9.433850441717122e-05,
      "loss": 1.4786,
      "step": 54920
    },
    {
      "epoch": 2.5381475773536337,
      "grad_norm": 0.20533820986747742,
      "learning_rate": 9.433081118067078e-05,
      "loss": 1.479,
      "step": 54930
    },
    {
      "epoch": 2.5386895185191456,
      "grad_norm": 0.2988840937614441,
      "learning_rate": 9.432311307204024e-05,
      "loss": 1.4668,
      "step": 54940
    },
    {
      "epoch": 2.539231459684658,
      "grad_norm": 0.24603962898254395,
      "learning_rate": 9.431541009223316e-05,
      "loss": 1.4821,
      "step": 54950
    },
    {
      "epoch": 2.5397734008501702,
      "grad_norm": 0.1960807591676712,
      "learning_rate": 9.430770224220376e-05,
      "loss": 1.4664,
      "step": 54960
    },
    {
      "epoch": 2.5403153420156825,
      "grad_norm": 0.20046474039554596,
      "learning_rate": 9.429998952290688e-05,
      "loss": 1.4732,
      "step": 54970
    },
    {
      "epoch": 2.540857283181195,
      "grad_norm": 0.1977715790271759,
      "learning_rate": 9.429227193529791e-05,
      "loss": 1.4807,
      "step": 54980
    },
    {
      "epoch": 2.5410740596473995,
      "eval_loss": 2.445849657058716,
      "eval_runtime": 21.9883,
      "eval_samples_per_second": 227.394,
      "eval_steps_per_second": 1.228,
      "step": 54984
    },
    {
      "epoch": 2.5413992243467067,
      "grad_norm": 0.20431387424468994,
      "learning_rate": 9.42845494803329e-05,
      "loss": 1.4758,
      "step": 54990
    },
    {
      "epoch": 2.541941165512219,
      "grad_norm": 0.3124183416366577,
      "learning_rate": 9.427682215896846e-05,
      "loss": 1.4776,
      "step": 55000
    },
    {
      "epoch": 2.5424831066777314,
      "grad_norm": 0.17832711338996887,
      "learning_rate": 9.426908997216179e-05,
      "loss": 1.4732,
      "step": 55010
    },
    {
      "epoch": 2.5430250478432437,
      "grad_norm": 0.2197619527578354,
      "learning_rate": 9.426135292087076e-05,
      "loss": 1.4652,
      "step": 55020
    },
    {
      "epoch": 2.543566989008756,
      "grad_norm": 0.3124410808086395,
      "learning_rate": 9.425361100605378e-05,
      "loss": 1.4776,
      "step": 55030
    },
    {
      "epoch": 2.544108930174268,
      "grad_norm": 0.28143149614334106,
      "learning_rate": 9.424586422866989e-05,
      "loss": 1.4755,
      "step": 55040
    },
    {
      "epoch": 2.54465087133978,
      "grad_norm": 0.3505375385284424,
      "learning_rate": 9.423811258967872e-05,
      "loss": 1.4826,
      "step": 55050
    },
    {
      "epoch": 2.5451928125052925,
      "grad_norm": 0.30467939376831055,
      "learning_rate": 9.423035609004054e-05,
      "loss": 1.4683,
      "step": 55060
    },
    {
      "epoch": 2.5457347536708044,
      "grad_norm": 0.2334168255329132,
      "learning_rate": 9.422259473071615e-05,
      "loss": 1.4676,
      "step": 55070
    },
    {
      "epoch": 2.545788947787356,
      "eval_loss": 2.4590961933135986,
      "eval_runtime": 22.0238,
      "eval_samples_per_second": 227.027,
      "eval_steps_per_second": 1.226,
      "step": 55071
    },
    {
      "epoch": 2.5462766948363167,
      "grad_norm": 0.24799096584320068,
      "learning_rate": 9.421482851266702e-05,
      "loss": 1.4726,
      "step": 55080
    },
    {
      "epoch": 2.546818636001829,
      "grad_norm": 0.17815372347831726,
      "learning_rate": 9.420705743685519e-05,
      "loss": 1.4807,
      "step": 55090
    },
    {
      "epoch": 2.5473605771673413,
      "grad_norm": 0.1867280900478363,
      "learning_rate": 9.419928150424328e-05,
      "loss": 1.479,
      "step": 55100
    },
    {
      "epoch": 2.5479025183328536,
      "grad_norm": 0.18417419493198395,
      "learning_rate": 9.419150071579459e-05,
      "loss": 1.4774,
      "step": 55110
    },
    {
      "epoch": 2.5484444594983655,
      "grad_norm": 0.24251243472099304,
      "learning_rate": 9.418371507247294e-05,
      "loss": 1.4771,
      "step": 55120
    },
    {
      "epoch": 2.548986400663878,
      "grad_norm": 0.22498652338981628,
      "learning_rate": 9.417592457524278e-05,
      "loss": 1.4618,
      "step": 55130
    },
    {
      "epoch": 2.54952834182939,
      "grad_norm": 0.3402760624885559,
      "learning_rate": 9.416812922506917e-05,
      "loss": 1.4772,
      "step": 55140
    },
    {
      "epoch": 2.5500702829949025,
      "grad_norm": 0.22304527461528778,
      "learning_rate": 9.416032902291778e-05,
      "loss": 1.4924,
      "step": 55150
    },
    {
      "epoch": 2.5505038359273122,
      "eval_loss": 2.4431533813476562,
      "eval_runtime": 22.327,
      "eval_samples_per_second": 223.944,
      "eval_steps_per_second": 1.209,
      "step": 55158
    },
    {
      "epoch": 2.5506122241604148,
      "grad_norm": 0.17721644043922424,
      "learning_rate": 9.415252396975482e-05,
      "loss": 1.4695,
      "step": 55160
    },
    {
      "epoch": 2.5511541653259266,
      "grad_norm": 0.2341170608997345,
      "learning_rate": 9.414471406654718e-05,
      "loss": 1.4676,
      "step": 55170
    },
    {
      "epoch": 2.551696106491439,
      "grad_norm": 0.17687572538852692,
      "learning_rate": 9.413689931426232e-05,
      "loss": 1.4775,
      "step": 55180
    },
    {
      "epoch": 2.5522380476569513,
      "grad_norm": 0.3460327982902527,
      "learning_rate": 9.412907971386828e-05,
      "loss": 1.4701,
      "step": 55190
    },
    {
      "epoch": 2.5527799888224636,
      "grad_norm": 0.3097844123840332,
      "learning_rate": 9.412125526633373e-05,
      "loss": 1.4664,
      "step": 55200
    },
    {
      "epoch": 2.553321929987976,
      "grad_norm": 0.2532496452331543,
      "learning_rate": 9.411342597262794e-05,
      "loss": 1.4723,
      "step": 55210
    },
    {
      "epoch": 2.553863871153488,
      "grad_norm": 0.18894581496715546,
      "learning_rate": 9.410559183372072e-05,
      "loss": 1.4768,
      "step": 55220
    },
    {
      "epoch": 2.554405812319,
      "grad_norm": 0.3138901889324188,
      "learning_rate": 9.409775285058259e-05,
      "loss": 1.4703,
      "step": 55230
    },
    {
      "epoch": 2.5549477534845124,
      "grad_norm": 0.5384250283241272,
      "learning_rate": 9.408990902418458e-05,
      "loss": 1.4734,
      "step": 55240
    },
    {
      "epoch": 2.5552187240672684,
      "eval_loss": 2.4425833225250244,
      "eval_runtime": 22.3536,
      "eval_samples_per_second": 223.678,
      "eval_steps_per_second": 1.208,
      "step": 55245
    },
    {
      "epoch": 2.0001083882331026,
      "grad_norm": 0.246513232588768,
      "learning_rate": 9.408206035549835e-05,
      "loss": 1.4546,
      "step": 55250
    },
    {
      "epoch": 2.000650329398615,
      "grad_norm": 0.18351303040981293,
      "learning_rate": 9.407420684549616e-05,
      "loss": 1.4714,
      "step": 55260
    },
    {
      "epoch": 2.0011922705641267,
      "grad_norm": 0.38743826746940613,
      "learning_rate": 9.406634849515087e-05,
      "loss": 1.4758,
      "step": 55270
    },
    {
      "epoch": 2.001734211729639,
      "grad_norm": 0.2079382985830307,
      "learning_rate": 9.405848530543593e-05,
      "loss": 1.4662,
      "step": 55280
    },
    {
      "epoch": 2.0022761528951514,
      "grad_norm": 0.19639058411121368,
      "learning_rate": 9.40506172773254e-05,
      "loss": 1.4805,
      "step": 55290
    },
    {
      "epoch": 2.0028180940606637,
      "grad_norm": 0.19277872145175934,
      "learning_rate": 9.404274441179397e-05,
      "loss": 1.4692,
      "step": 55300
    },
    {
      "epoch": 2.0033600352261756,
      "grad_norm": 0.3094417452812195,
      "learning_rate": 9.403486670981685e-05,
      "loss": 1.4718,
      "step": 55310
    },
    {
      "epoch": 2.003901976391688,
      "grad_norm": 0.2434214949607849,
      "learning_rate": 9.402698417236991e-05,
      "loss": 1.4711,
      "step": 55320
    },
    {
      "epoch": 2.0044439175572,
      "grad_norm": 0.37863555550575256,
      "learning_rate": 9.401909680042962e-05,
      "loss": 1.4704,
      "step": 55330
    },
    {
      "epoch": 2.0045523057903027,
      "eval_loss": 2.450153112411499,
      "eval_runtime": 62.0654,
      "eval_samples_per_second": 80.56,
      "eval_steps_per_second": 0.435,
      "step": 55332
    },
    {
      "epoch": 2.0049858587227125,
      "grad_norm": 0.7027290463447571,
      "learning_rate": 9.401120459497302e-05,
      "loss": 1.4786,
      "step": 55340
    },
    {
      "epoch": 2.005527799888225,
      "grad_norm": 0.42648592591285706,
      "learning_rate": 9.400330755697774e-05,
      "loss": 1.4755,
      "step": 55350
    },
    {
      "epoch": 2.0060697410537367,
      "grad_norm": 0.22446554899215698,
      "learning_rate": 9.399540568742209e-05,
      "loss": 1.4791,
      "step": 55360
    },
    {
      "epoch": 2.006611682219249,
      "grad_norm": 0.2071634829044342,
      "learning_rate": 9.398749898728487e-05,
      "loss": 1.4712,
      "step": 55370
    },
    {
      "epoch": 2.0071536233847613,
      "grad_norm": 0.22032909095287323,
      "learning_rate": 9.397958745754554e-05,
      "loss": 1.4703,
      "step": 55380
    },
    {
      "epoch": 2.0076955645502736,
      "grad_norm": 0.24659113585948944,
      "learning_rate": 9.397167109918416e-05,
      "loss": 1.4654,
      "step": 55390
    },
    {
      "epoch": 2.0082375057157855,
      "grad_norm": 0.23570233583450317,
      "learning_rate": 9.396374991318135e-05,
      "loss": 1.4684,
      "step": 55400
    },
    {
      "epoch": 2.008779446881298,
      "grad_norm": 0.3380047380924225,
      "learning_rate": 9.395582390051838e-05,
      "loss": 1.4691,
      "step": 55410
    },
    {
      "epoch": 2.009267193930259,
      "eval_loss": 2.4444944858551025,
      "eval_runtime": 24.1261,
      "eval_samples_per_second": 207.245,
      "eval_steps_per_second": 1.119,
      "step": 55419
    },
    {
      "epoch": 2.00932138804681,
      "grad_norm": 0.2984738051891327,
      "learning_rate": 9.394789306217707e-05,
      "loss": 1.4741,
      "step": 55420
    },
    {
      "epoch": 2.0098633292123225,
      "grad_norm": 0.2363492250442505,
      "learning_rate": 9.393995739913985e-05,
      "loss": 1.474,
      "step": 55430
    },
    {
      "epoch": 2.010405270377835,
      "grad_norm": 0.18736572563648224,
      "learning_rate": 9.393201691238976e-05,
      "loss": 1.4761,
      "step": 55440
    },
    {
      "epoch": 2.0109472115433467,
      "grad_norm": 0.27000486850738525,
      "learning_rate": 9.392407160291045e-05,
      "loss": 1.4624,
      "step": 55450
    },
    {
      "epoch": 2.011489152708859,
      "grad_norm": 0.26010996103286743,
      "learning_rate": 9.391612147168617e-05,
      "loss": 1.4704,
      "step": 55460
    },
    {
      "epoch": 2.0120310938743713,
      "grad_norm": 0.2281525433063507,
      "learning_rate": 9.390816651970168e-05,
      "loss": 1.465,
      "step": 55470
    },
    {
      "epoch": 2.0125730350398836,
      "grad_norm": 0.20982374250888824,
      "learning_rate": 9.390020674794245e-05,
      "loss": 1.4664,
      "step": 55480
    },
    {
      "epoch": 2.013114976205396,
      "grad_norm": 0.20972926914691925,
      "learning_rate": 9.389224215739452e-05,
      "loss": 1.4597,
      "step": 55490
    },
    {
      "epoch": 2.013656917370908,
      "grad_norm": 0.21081332862377167,
      "learning_rate": 9.388427274904447e-05,
      "loss": 1.4585,
      "step": 55500
    },
    {
      "epoch": 2.0139820820702155,
      "eval_loss": 2.451312780380249,
      "eval_runtime": 21.9802,
      "eval_samples_per_second": 227.477,
      "eval_steps_per_second": 1.228,
      "step": 55506
    },
    {
      "epoch": 2.01419885853642,
      "grad_norm": 0.22025637328624725,
      "learning_rate": 9.387629852387952e-05,
      "loss": 1.4756,
      "step": 55510
    },
    {
      "epoch": 2.0147407997019324,
      "grad_norm": 0.18817798793315887,
      "learning_rate": 9.38683194828875e-05,
      "loss": 1.4721,
      "step": 55520
    },
    {
      "epoch": 2.0152827408674447,
      "grad_norm": 0.30482780933380127,
      "learning_rate": 9.386033562705681e-05,
      "loss": 1.4679,
      "step": 55530
    },
    {
      "epoch": 2.0158246820329566,
      "grad_norm": 0.20792457461357117,
      "learning_rate": 9.385234695737647e-05,
      "loss": 1.4554,
      "step": 55540
    },
    {
      "epoch": 2.016366623198469,
      "grad_norm": 0.39416491985321045,
      "learning_rate": 9.384435347483606e-05,
      "loss": 1.4665,
      "step": 55550
    },
    {
      "epoch": 2.0169085643639812,
      "grad_norm": 0.29534631967544556,
      "learning_rate": 9.383635518042579e-05,
      "loss": 1.4604,
      "step": 55560
    },
    {
      "epoch": 2.0174505055294936,
      "grad_norm": 0.20709377527236938,
      "learning_rate": 9.382835207513644e-05,
      "loss": 1.4783,
      "step": 55570
    },
    {
      "epoch": 2.017992446695006,
      "grad_norm": 0.1838233321905136,
      "learning_rate": 9.382034415995942e-05,
      "loss": 1.4729,
      "step": 55580
    },
    {
      "epoch": 2.0185343878605178,
      "grad_norm": 0.20121820271015167,
      "learning_rate": 9.381233143588671e-05,
      "loss": 1.469,
      "step": 55590
    },
    {
      "epoch": 2.0186969702101716,
      "eval_loss": 2.4466960430145264,
      "eval_runtime": 22.0147,
      "eval_samples_per_second": 227.121,
      "eval_steps_per_second": 1.226,
      "step": 55593
    },
    {
      "epoch": 2.01907632902603,
      "grad_norm": 0.24472562968730927,
      "learning_rate": 9.380431390391089e-05,
      "loss": 1.4665,
      "step": 55600
    },
    {
      "epoch": 2.0196182701915424,
      "grad_norm": 0.2527831196784973,
      "learning_rate": 9.379629156502513e-05,
      "loss": 1.4773,
      "step": 55610
    },
    {
      "epoch": 2.0201602113570547,
      "grad_norm": 0.23660998046398163,
      "learning_rate": 9.37882644202232e-05,
      "loss": 1.4686,
      "step": 55620
    },
    {
      "epoch": 2.0207021525225666,
      "grad_norm": 0.2622048258781433,
      "learning_rate": 9.378023247049949e-05,
      "loss": 1.4686,
      "step": 55630
    },
    {
      "epoch": 2.021244093688079,
      "grad_norm": 0.1906578242778778,
      "learning_rate": 9.377219571684895e-05,
      "loss": 1.4706,
      "step": 55640
    },
    {
      "epoch": 2.021786034853591,
      "grad_norm": 0.2571401000022888,
      "learning_rate": 9.376415416026712e-05,
      "loss": 1.4626,
      "step": 55650
    },
    {
      "epoch": 2.0223279760191035,
      "grad_norm": 0.37901771068573,
      "learning_rate": 9.375610780175017e-05,
      "loss": 1.4766,
      "step": 55660
    },
    {
      "epoch": 2.022869917184616,
      "grad_norm": 0.23881062865257263,
      "learning_rate": 9.374805664229484e-05,
      "loss": 1.4653,
      "step": 55670
    },
    {
      "epoch": 2.0234118583501277,
      "grad_norm": 0.5407462120056152,
      "learning_rate": 9.37400006828985e-05,
      "loss": 1.4641,
      "step": 55680
    },
    {
      "epoch": 2.0234118583501277,
      "eval_loss": 2.4462928771972656,
      "eval_runtime": 21.8259,
      "eval_samples_per_second": 229.086,
      "eval_steps_per_second": 1.237,
      "step": 55680
    },
    {
      "epoch": 2.02395379951564,
      "grad_norm": 0.24240484833717346,
      "learning_rate": 9.373193992455907e-05,
      "loss": 1.4575,
      "step": 55690
    },
    {
      "epoch": 2.0244957406811523,
      "grad_norm": 0.3244767487049103,
      "learning_rate": 9.372387436827507e-05,
      "loss": 1.4706,
      "step": 55700
    },
    {
      "epoch": 2.0250376818466647,
      "grad_norm": 0.22140353918075562,
      "learning_rate": 9.371580401504564e-05,
      "loss": 1.4656,
      "step": 55710
    },
    {
      "epoch": 2.0255796230121765,
      "grad_norm": 0.21475926041603088,
      "learning_rate": 9.370772886587049e-05,
      "loss": 1.4706,
      "step": 55720
    },
    {
      "epoch": 2.026121564177689,
      "grad_norm": 0.2393941879272461,
      "learning_rate": 9.369964892174995e-05,
      "loss": 1.4622,
      "step": 55730
    },
    {
      "epoch": 2.026663505343201,
      "grad_norm": 0.23679640889167786,
      "learning_rate": 9.369156418368491e-05,
      "loss": 1.4604,
      "step": 55740
    },
    {
      "epoch": 2.0272054465087135,
      "grad_norm": 0.2113851010799408,
      "learning_rate": 9.368347465267688e-05,
      "loss": 1.4713,
      "step": 55750
    },
    {
      "epoch": 2.027747387674226,
      "grad_norm": 0.25252246856689453,
      "learning_rate": 9.367538032972797e-05,
      "loss": 1.4595,
      "step": 55760
    },
    {
      "epoch": 2.0281267464900843,
      "eval_loss": 2.453479051589966,
      "eval_runtime": 21.9786,
      "eval_samples_per_second": 227.494,
      "eval_steps_per_second": 1.228,
      "step": 55767
    },
    {
      "epoch": 2.0282893288397377,
      "grad_norm": 0.23864135146141052,
      "learning_rate": 9.366728121584084e-05,
      "loss": 1.4588,
      "step": 55770
    },
    {
      "epoch": 2.02883127000525,
      "grad_norm": 0.24793609976768494,
      "learning_rate": 9.365917731201879e-05,
      "loss": 1.4593,
      "step": 55780
    },
    {
      "epoch": 2.0293732111707623,
      "grad_norm": 0.21740341186523438,
      "learning_rate": 9.365106861926571e-05,
      "loss": 1.4745,
      "step": 55790
    },
    {
      "epoch": 2.0299151523362746,
      "grad_norm": 0.19946753978729248,
      "learning_rate": 9.364295513858604e-05,
      "loss": 1.4723,
      "step": 55800
    },
    {
      "epoch": 2.030457093501787,
      "grad_norm": 0.2608671486377716,
      "learning_rate": 9.363483687098487e-05,
      "loss": 1.4651,
      "step": 55810
    },
    {
      "epoch": 2.030999034667299,
      "grad_norm": 0.2773636281490326,
      "learning_rate": 9.362671381746784e-05,
      "loss": 1.4797,
      "step": 55820
    },
    {
      "epoch": 2.031540975832811,
      "grad_norm": 0.3309532105922699,
      "learning_rate": 9.361858597904119e-05,
      "loss": 1.4591,
      "step": 55830
    },
    {
      "epoch": 2.0320829169983234,
      "grad_norm": 0.343264102935791,
      "learning_rate": 9.36104533567118e-05,
      "loss": 1.4657,
      "step": 55840
    },
    {
      "epoch": 2.0326248581638358,
      "grad_norm": 0.21975140273571014,
      "learning_rate": 9.360231595148708e-05,
      "loss": 1.4684,
      "step": 55850
    },
    {
      "epoch": 2.0328416346300404,
      "eval_loss": 2.453883409500122,
      "eval_runtime": 21.9811,
      "eval_samples_per_second": 227.468,
      "eval_steps_per_second": 1.228,
      "step": 55854
    },
    {
      "epoch": 2.0331667993293476,
      "grad_norm": 0.2824975848197937,
      "learning_rate": 9.359417376437503e-05,
      "loss": 1.4678,
      "step": 55860
    },
    {
      "epoch": 2.03370874049486,
      "grad_norm": 0.41226598620414734,
      "learning_rate": 9.35860267963843e-05,
      "loss": 1.4693,
      "step": 55870
    },
    {
      "epoch": 2.0342506816603723,
      "grad_norm": 0.37096548080444336,
      "learning_rate": 9.35778750485241e-05,
      "loss": 1.461,
      "step": 55880
    },
    {
      "epoch": 2.0347926228258846,
      "grad_norm": 0.3635852038860321,
      "learning_rate": 9.356971852180421e-05,
      "loss": 1.4598,
      "step": 55890
    },
    {
      "epoch": 2.035334563991397,
      "grad_norm": 0.32016879320144653,
      "learning_rate": 9.356155721723506e-05,
      "loss": 1.4612,
      "step": 55900
    },
    {
      "epoch": 2.0358765051569088,
      "grad_norm": 0.23647364974021912,
      "learning_rate": 9.355339113582761e-05,
      "loss": 1.4744,
      "step": 55910
    },
    {
      "epoch": 2.036418446322421,
      "grad_norm": 0.40886321663856506,
      "learning_rate": 9.354522027859347e-05,
      "loss": 1.4683,
      "step": 55920
    },
    {
      "epoch": 2.0369603874879334,
      "grad_norm": 0.3399202525615692,
      "learning_rate": 9.353704464654477e-05,
      "loss": 1.4559,
      "step": 55930
    },
    {
      "epoch": 2.0375023286534457,
      "grad_norm": 0.2759416401386261,
      "learning_rate": 9.35288642406943e-05,
      "loss": 1.4592,
      "step": 55940
    },
    {
      "epoch": 2.037556522769997,
      "eval_loss": 2.454275608062744,
      "eval_runtime": 21.9844,
      "eval_samples_per_second": 227.434,
      "eval_steps_per_second": 1.228,
      "step": 55941
    },
    {
      "epoch": 2.0380442698189576,
      "grad_norm": 0.193961039185524,
      "learning_rate": 9.352067906205538e-05,
      "loss": 1.4633,
      "step": 55950
    },
    {
      "epoch": 2.03858621098447,
      "grad_norm": 0.23220661282539368,
      "learning_rate": 9.3512489111642e-05,
      "loss": 1.4639,
      "step": 55960
    },
    {
      "epoch": 2.039128152149982,
      "grad_norm": 0.24919536709785461,
      "learning_rate": 9.350429439046867e-05,
      "loss": 1.4727,
      "step": 55970
    },
    {
      "epoch": 2.0396700933154945,
      "grad_norm": 0.3676759898662567,
      "learning_rate": 9.349609489955053e-05,
      "loss": 1.477,
      "step": 55980
    },
    {
      "epoch": 2.040212034481007,
      "grad_norm": 0.5084336996078491,
      "learning_rate": 9.348789063990328e-05,
      "loss": 1.4609,
      "step": 55990
    },
    {
      "epoch": 2.0407539756465187,
      "grad_norm": 0.33119305968284607,
      "learning_rate": 9.347968161254321e-05,
      "loss": 1.4632,
      "step": 56000
    },
    {
      "epoch": 2.041295916812031,
      "grad_norm": 0.23832644522190094,
      "learning_rate": 9.347146781848726e-05,
      "loss": 1.4682,
      "step": 56010
    },
    {
      "epoch": 2.0418378579775434,
      "grad_norm": 0.2528223395347595,
      "learning_rate": 9.346324925875293e-05,
      "loss": 1.4668,
      "step": 56020
    },
    {
      "epoch": 2.042271410909953,
      "eval_loss": 2.461331844329834,
      "eval_runtime": 21.9887,
      "eval_samples_per_second": 227.39,
      "eval_steps_per_second": 1.228,
      "step": 56028
    },
    {
      "epoch": 2.0423797991430557,
      "grad_norm": 0.30627909302711487,
      "learning_rate": 9.345502593435824e-05,
      "loss": 1.4561,
      "step": 56030
    },
    {
      "epoch": 2.0429217403085675,
      "grad_norm": 0.275127112865448,
      "learning_rate": 9.34467978463219e-05,
      "loss": 1.4664,
      "step": 56040
    },
    {
      "epoch": 2.04346368147408,
      "grad_norm": 0.25963136553764343,
      "learning_rate": 9.343856499566315e-05,
      "loss": 1.471,
      "step": 56050
    },
    {
      "epoch": 2.044005622639592,
      "grad_norm": 0.18673470616340637,
      "learning_rate": 9.343032738340187e-05,
      "loss": 1.4709,
      "step": 56060
    },
    {
      "epoch": 2.0445475638051045,
      "grad_norm": 0.23496182262897491,
      "learning_rate": 9.342208501055849e-05,
      "loss": 1.4652,
      "step": 56070
    },
    {
      "epoch": 2.045089504970617,
      "grad_norm": 0.23610053956508636,
      "learning_rate": 9.341383787815402e-05,
      "loss": 1.4712,
      "step": 56080
    },
    {
      "epoch": 2.0456314461361287,
      "grad_norm": 0.2762158215045929,
      "learning_rate": 9.340558598721008e-05,
      "loss": 1.4671,
      "step": 56090
    },
    {
      "epoch": 2.046173387301641,
      "grad_norm": 0.23105528950691223,
      "learning_rate": 9.339732933874891e-05,
      "loss": 1.4625,
      "step": 56100
    },
    {
      "epoch": 2.0467153284671533,
      "grad_norm": 0.4698254466056824,
      "learning_rate": 9.338906793379327e-05,
      "loss": 1.4586,
      "step": 56110
    },
    {
      "epoch": 2.0469862990499093,
      "eval_loss": 2.449103832244873,
      "eval_runtime": 22.0267,
      "eval_samples_per_second": 226.997,
      "eval_steps_per_second": 1.226,
      "step": 56115
    },
    {
      "epoch": 2.0472572696326656,
      "grad_norm": 0.311443954706192,
      "learning_rate": 9.338080177336656e-05,
      "loss": 1.4692,
      "step": 56120
    },
    {
      "epoch": 2.0477992107981775,
      "grad_norm": 0.3137112557888031,
      "learning_rate": 9.337253085849276e-05,
      "loss": 1.4666,
      "step": 56130
    },
    {
      "epoch": 2.04834115196369,
      "grad_norm": 0.27474987506866455,
      "learning_rate": 9.336425519019644e-05,
      "loss": 1.4654,
      "step": 56140
    },
    {
      "epoch": 2.048883093129202,
      "grad_norm": 0.21653681993484497,
      "learning_rate": 9.335597476950275e-05,
      "loss": 1.4556,
      "step": 56150
    },
    {
      "epoch": 2.0494250342947145,
      "grad_norm": 0.21602825820446014,
      "learning_rate": 9.334768959743742e-05,
      "loss": 1.4701,
      "step": 56160
    },
    {
      "epoch": 2.0499669754602268,
      "grad_norm": 0.2874417006969452,
      "learning_rate": 9.333939967502681e-05,
      "loss": 1.4702,
      "step": 56170
    },
    {
      "epoch": 2.0505089166257386,
      "grad_norm": 0.3574830889701843,
      "learning_rate": 9.333110500329781e-05,
      "loss": 1.4587,
      "step": 56180
    },
    {
      "epoch": 2.051050857791251,
      "grad_norm": 0.33918556571006775,
      "learning_rate": 9.332280558327795e-05,
      "loss": 1.4667,
      "step": 56190
    },
    {
      "epoch": 2.0515927989567633,
      "grad_norm": 0.25858092308044434,
      "learning_rate": 9.33145014159953e-05,
      "loss": 1.4654,
      "step": 56200
    },
    {
      "epoch": 2.051701187189866,
      "eval_loss": 2.4660325050354004,
      "eval_runtime": 21.9813,
      "eval_samples_per_second": 227.466,
      "eval_steps_per_second": 1.228,
      "step": 56202
    },
    {
      "epoch": 2.0521347401222756,
      "grad_norm": 0.24299031496047974,
      "learning_rate": 9.330619250247855e-05,
      "loss": 1.4677,
      "step": 56210
    },
    {
      "epoch": 2.052676681287788,
      "grad_norm": 0.2615341246128082,
      "learning_rate": 9.3297878843757e-05,
      "loss": 1.4633,
      "step": 56220
    },
    {
      "epoch": 2.0532186224533,
      "grad_norm": 0.34123480319976807,
      "learning_rate": 9.328956044086049e-05,
      "loss": 1.4689,
      "step": 56230
    },
    {
      "epoch": 2.053760563618812,
      "grad_norm": 0.3530578315258026,
      "learning_rate": 9.328123729481947e-05,
      "loss": 1.4615,
      "step": 56240
    },
    {
      "epoch": 2.0543025047843244,
      "grad_norm": 0.18291862308979034,
      "learning_rate": 9.327290940666497e-05,
      "loss": 1.4621,
      "step": 56250
    },
    {
      "epoch": 2.0548444459498367,
      "grad_norm": 0.19218549132347107,
      "learning_rate": 9.326457677742861e-05,
      "loss": 1.4736,
      "step": 56260
    },
    {
      "epoch": 2.0553863871153486,
      "grad_norm": 0.26161691546440125,
      "learning_rate": 9.325623940814263e-05,
      "loss": 1.4674,
      "step": 56270
    },
    {
      "epoch": 2.055928328280861,
      "grad_norm": 0.2795359194278717,
      "learning_rate": 9.324789729983979e-05,
      "loss": 1.4581,
      "step": 56280
    },
    {
      "epoch": 2.056416075329822,
      "eval_loss": 2.465681314468384,
      "eval_runtime": 21.9832,
      "eval_samples_per_second": 227.447,
      "eval_steps_per_second": 1.228,
      "step": 56289
    },
    {
      "epoch": 2.0564702694463732,
      "grad_norm": 0.19691872596740723,
      "learning_rate": 9.323955045355349e-05,
      "loss": 1.4692,
      "step": 56290
    },
    {
      "epoch": 2.0570122106118855,
      "grad_norm": 0.33033034205436707,
      "learning_rate": 9.323119887031769e-05,
      "loss": 1.4676,
      "step": 56300
    },
    {
      "epoch": 2.057554151777398,
      "grad_norm": 0.2493034452199936,
      "learning_rate": 9.322284255116696e-05,
      "loss": 1.4552,
      "step": 56310
    },
    {
      "epoch": 2.0580960929429097,
      "grad_norm": 0.3580038249492645,
      "learning_rate": 9.321448149713645e-05,
      "loss": 1.4593,
      "step": 56320
    },
    {
      "epoch": 2.058638034108422,
      "grad_norm": 0.2858411967754364,
      "learning_rate": 9.320611570926189e-05,
      "loss": 1.4693,
      "step": 56330
    },
    {
      "epoch": 2.0591799752739344,
      "grad_norm": 0.35059741139411926,
      "learning_rate": 9.319774518857958e-05,
      "loss": 1.468,
      "step": 56340
    },
    {
      "epoch": 2.0597219164394467,
      "grad_norm": 0.30402854084968567,
      "learning_rate": 9.318936993612643e-05,
      "loss": 1.4626,
      "step": 56350
    },
    {
      "epoch": 2.0602638576049586,
      "grad_norm": 0.33084359765052795,
      "learning_rate": 9.318098995293993e-05,
      "loss": 1.4748,
      "step": 56360
    },
    {
      "epoch": 2.060805798770471,
      "grad_norm": 0.21346881985664368,
      "learning_rate": 9.317260524005817e-05,
      "loss": 1.4647,
      "step": 56370
    },
    {
      "epoch": 2.061130963469778,
      "eval_loss": 2.4557716846466064,
      "eval_runtime": 21.9864,
      "eval_samples_per_second": 227.413,
      "eval_steps_per_second": 1.228,
      "step": 56376
    },
    {
      "epoch": 2.061347739935983,
      "grad_norm": 0.32445028424263,
      "learning_rate": 9.31642157985198e-05,
      "loss": 1.4523,
      "step": 56380
    },
    {
      "epoch": 2.0618896811014955,
      "grad_norm": 0.2711530029773712,
      "learning_rate": 9.315582162936407e-05,
      "loss": 1.4613,
      "step": 56390
    },
    {
      "epoch": 2.062431622267008,
      "grad_norm": 0.23968590795993805,
      "learning_rate": 9.314742273363082e-05,
      "loss": 1.4578,
      "step": 56400
    },
    {
      "epoch": 2.0629735634325197,
      "grad_norm": 0.29744964838027954,
      "learning_rate": 9.313901911236046e-05,
      "loss": 1.4645,
      "step": 56410
    },
    {
      "epoch": 2.063515504598032,
      "grad_norm": 0.23724764585494995,
      "learning_rate": 9.313061076659398e-05,
      "loss": 1.4636,
      "step": 56420
    },
    {
      "epoch": 2.0640574457635443,
      "grad_norm": 0.18018679320812225,
      "learning_rate": 9.312219769737299e-05,
      "loss": 1.4586,
      "step": 56430
    },
    {
      "epoch": 2.0645993869290566,
      "grad_norm": 0.25777846574783325,
      "learning_rate": 9.311377990573967e-05,
      "loss": 1.464,
      "step": 56440
    },
    {
      "epoch": 2.065141328094569,
      "grad_norm": 0.2517983317375183,
      "learning_rate": 9.310535739273675e-05,
      "loss": 1.4611,
      "step": 56450
    },
    {
      "epoch": 2.065683269260081,
      "grad_norm": 0.32330676913261414,
      "learning_rate": 9.30969301594076e-05,
      "loss": 1.4483,
      "step": 56460
    },
    {
      "epoch": 2.0658458516097347,
      "eval_loss": 2.4468648433685303,
      "eval_runtime": 23.3942,
      "eval_samples_per_second": 213.728,
      "eval_steps_per_second": 1.154,
      "step": 56463
    },
    {
      "epoch": 2.066225210425593,
      "grad_norm": 0.3357566297054291,
      "learning_rate": 9.308849820679614e-05,
      "loss": 1.4712,
      "step": 56470
    },
    {
      "epoch": 2.0667671515911055,
      "grad_norm": 0.18558350205421448,
      "learning_rate": 9.308006153594692e-05,
      "loss": 1.4578,
      "step": 56480
    },
    {
      "epoch": 2.067309092756618,
      "grad_norm": 0.324517160654068,
      "learning_rate": 9.307162014790496e-05,
      "loss": 1.457,
      "step": 56490
    },
    {
      "epoch": 2.0678510339221297,
      "grad_norm": 0.2188093215227127,
      "learning_rate": 9.3063174043716e-05,
      "loss": 1.4654,
      "step": 56500
    },
    {
      "epoch": 2.068392975087642,
      "grad_norm": 0.19635216891765594,
      "learning_rate": 9.30547232244263e-05,
      "loss": 1.4495,
      "step": 56510
    },
    {
      "epoch": 2.0689349162531543,
      "grad_norm": 0.18253515660762787,
      "learning_rate": 9.304626769108271e-05,
      "loss": 1.4619,
      "step": 56520
    },
    {
      "epoch": 2.0694768574186666,
      "grad_norm": 0.19728650152683258,
      "learning_rate": 9.303780744473265e-05,
      "loss": 1.4629,
      "step": 56530
    },
    {
      "epoch": 2.0700187985841785,
      "grad_norm": 0.27253544330596924,
      "learning_rate": 9.302934248642414e-05,
      "loss": 1.4604,
      "step": 56540
    },
    {
      "epoch": 2.070560739749691,
      "grad_norm": 0.22469353675842285,
      "learning_rate": 9.30208728172058e-05,
      "loss": 1.4624,
      "step": 56550
    },
    {
      "epoch": 2.070560739749691,
      "eval_loss": 2.449016809463501,
      "eval_runtime": 21.9605,
      "eval_samples_per_second": 227.682,
      "eval_steps_per_second": 1.229,
      "step": 56550
    },
    {
      "epoch": 2.071102680915203,
      "grad_norm": 0.4223212003707886,
      "learning_rate": 9.301239843812681e-05,
      "loss": 1.4569,
      "step": 56560
    },
    {
      "epoch": 2.0716446220807154,
      "grad_norm": 0.1832091063261032,
      "learning_rate": 9.30039193502369e-05,
      "loss": 1.4669,
      "step": 56570
    },
    {
      "epoch": 2.0721865632462277,
      "grad_norm": 0.19555889070034027,
      "learning_rate": 9.29954355545865e-05,
      "loss": 1.4459,
      "step": 56580
    },
    {
      "epoch": 2.0727285044117396,
      "grad_norm": 0.25337639451026917,
      "learning_rate": 9.29869470522265e-05,
      "loss": 1.4591,
      "step": 56590
    },
    {
      "epoch": 2.073270445577252,
      "grad_norm": 0.2375129759311676,
      "learning_rate": 9.29784538442084e-05,
      "loss": 1.4576,
      "step": 56600
    },
    {
      "epoch": 2.0738123867427642,
      "grad_norm": 0.3068578243255615,
      "learning_rate": 9.296995593158433e-05,
      "loss": 1.4535,
      "step": 56610
    },
    {
      "epoch": 2.0743543279082766,
      "grad_norm": 0.37019553780555725,
      "learning_rate": 9.296145331540696e-05,
      "loss": 1.4583,
      "step": 56620
    },
    {
      "epoch": 2.074896269073789,
      "grad_norm": 0.26135745644569397,
      "learning_rate": 9.295294599672959e-05,
      "loss": 1.463,
      "step": 56630
    },
    {
      "epoch": 2.0752756278896474,
      "eval_loss": 2.45114803314209,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.323,
      "eval_steps_per_second": 1.228,
      "step": 56637
    },
    {
      "epoch": 2.0754382102393008,
      "grad_norm": 0.23497585952281952,
      "learning_rate": 9.294443397660603e-05,
      "loss": 1.4568,
      "step": 56640
    },
    {
      "epoch": 2.075980151404813,
      "grad_norm": 0.20370684564113617,
      "learning_rate": 9.293591725609072e-05,
      "loss": 1.4567,
      "step": 56650
    },
    {
      "epoch": 2.0765220925703254,
      "grad_norm": 0.2905084490776062,
      "learning_rate": 9.292739583623869e-05,
      "loss": 1.4572,
      "step": 56660
    },
    {
      "epoch": 2.0770640337358377,
      "grad_norm": 0.17985448241233826,
      "learning_rate": 9.291886971810554e-05,
      "loss": 1.465,
      "step": 56670
    },
    {
      "epoch": 2.0776059749013496,
      "grad_norm": 0.2356506884098053,
      "learning_rate": 9.291033890274743e-05,
      "loss": 1.4714,
      "step": 56680
    },
    {
      "epoch": 2.078147916066862,
      "grad_norm": 0.22777162492275238,
      "learning_rate": 9.290180339122111e-05,
      "loss": 1.458,
      "step": 56690
    },
    {
      "epoch": 2.078689857232374,
      "grad_norm": 0.17766469717025757,
      "learning_rate": 9.289326318458396e-05,
      "loss": 1.466,
      "step": 56700
    },
    {
      "epoch": 2.0792317983978865,
      "grad_norm": 0.18408410251140594,
      "learning_rate": 9.288471828389387e-05,
      "loss": 1.4562,
      "step": 56710
    },
    {
      "epoch": 2.079773739563399,
      "grad_norm": 0.2577512264251709,
      "learning_rate": 9.287616869020935e-05,
      "loss": 1.4582,
      "step": 56720
    },
    {
      "epoch": 2.0799905160296035,
      "eval_loss": 2.459651231765747,
      "eval_runtime": 21.9706,
      "eval_samples_per_second": 227.577,
      "eval_steps_per_second": 1.229,
      "step": 56724
    },
    {
      "epoch": 2.0803156807289107,
      "grad_norm": 0.19766086339950562,
      "learning_rate": 9.286761440458952e-05,
      "loss": 1.4558,
      "step": 56730
    },
    {
      "epoch": 2.080857621894423,
      "grad_norm": 0.33140110969543457,
      "learning_rate": 9.285905542809401e-05,
      "loss": 1.4656,
      "step": 56740
    },
    {
      "epoch": 2.0813995630599353,
      "grad_norm": 0.26236221194267273,
      "learning_rate": 9.285049176178308e-05,
      "loss": 1.4667,
      "step": 56750
    },
    {
      "epoch": 2.0819415042254477,
      "grad_norm": 0.39471328258514404,
      "learning_rate": 9.284192340671756e-05,
      "loss": 1.4631,
      "step": 56760
    },
    {
      "epoch": 2.0824834453909595,
      "grad_norm": 0.19031473994255066,
      "learning_rate": 9.283335036395886e-05,
      "loss": 1.4657,
      "step": 56770
    },
    {
      "epoch": 2.083025386556472,
      "grad_norm": 0.27123594284057617,
      "learning_rate": 9.282477263456899e-05,
      "loss": 1.4586,
      "step": 56780
    },
    {
      "epoch": 2.083567327721984,
      "grad_norm": 0.4629122018814087,
      "learning_rate": 9.281619021961049e-05,
      "loss": 1.4637,
      "step": 56790
    },
    {
      "epoch": 2.0841092688874965,
      "grad_norm": 0.2233474850654602,
      "learning_rate": 9.280760312014651e-05,
      "loss": 1.4612,
      "step": 56800
    },
    {
      "epoch": 2.084651210053009,
      "grad_norm": 0.25030142068862915,
      "learning_rate": 9.279901133724082e-05,
      "loss": 1.446,
      "step": 56810
    },
    {
      "epoch": 2.0847054041695596,
      "eval_loss": 2.456845998764038,
      "eval_runtime": 21.9864,
      "eval_samples_per_second": 227.413,
      "eval_steps_per_second": 1.228,
      "step": 56811
    },
    {
      "epoch": 2.0851931512185207,
      "grad_norm": 0.2584102153778076,
      "learning_rate": 9.279041487195772e-05,
      "loss": 1.468,
      "step": 56820
    },
    {
      "epoch": 2.085735092384033,
      "grad_norm": 0.2037636637687683,
      "learning_rate": 9.27818137253621e-05,
      "loss": 1.457,
      "step": 56830
    },
    {
      "epoch": 2.0862770335495453,
      "grad_norm": 0.2464050054550171,
      "learning_rate": 9.27732078985194e-05,
      "loss": 1.4455,
      "step": 56840
    },
    {
      "epoch": 2.0868189747150576,
      "grad_norm": 0.24278301000595093,
      "learning_rate": 9.27645973924957e-05,
      "loss": 1.4556,
      "step": 56850
    },
    {
      "epoch": 2.08736091588057,
      "grad_norm": 0.2788692116737366,
      "learning_rate": 9.275598220835765e-05,
      "loss": 1.4473,
      "step": 56860
    },
    {
      "epoch": 2.087902857046082,
      "grad_norm": 0.21799428761005402,
      "learning_rate": 9.274736234717243e-05,
      "loss": 1.4548,
      "step": 56870
    },
    {
      "epoch": 2.088444798211594,
      "grad_norm": 0.2989203929901123,
      "learning_rate": 9.273873781000786e-05,
      "loss": 1.4623,
      "step": 56880
    },
    {
      "epoch": 2.0889867393771064,
      "grad_norm": 0.21010935306549072,
      "learning_rate": 9.273010859793231e-05,
      "loss": 1.456,
      "step": 56890
    },
    {
      "epoch": 2.089420292309516,
      "eval_loss": 2.455754041671753,
      "eval_runtime": 21.9865,
      "eval_samples_per_second": 227.412,
      "eval_steps_per_second": 1.228,
      "step": 56898
    },
    {
      "epoch": 2.0895286805426188,
      "grad_norm": 0.18887412548065186,
      "learning_rate": 9.272147471201467e-05,
      "loss": 1.456,
      "step": 56900
    },
    {
      "epoch": 2.0900706217081306,
      "grad_norm": 0.26745525002479553,
      "learning_rate": 9.271283615332455e-05,
      "loss": 1.4589,
      "step": 56910
    },
    {
      "epoch": 2.090612562873643,
      "grad_norm": 0.18780305981636047,
      "learning_rate": 9.270419292293201e-05,
      "loss": 1.4585,
      "step": 56920
    },
    {
      "epoch": 2.0911545040391553,
      "grad_norm": 0.21684403717517853,
      "learning_rate": 9.269554502190775e-05,
      "loss": 1.4464,
      "step": 56930
    },
    {
      "epoch": 2.0916964452046676,
      "grad_norm": 0.4031422436237335,
      "learning_rate": 9.268689245132302e-05,
      "loss": 1.4465,
      "step": 56940
    },
    {
      "epoch": 2.09223838637018,
      "grad_norm": 0.31991732120513916,
      "learning_rate": 9.267823521224967e-05,
      "loss": 1.4656,
      "step": 56950
    },
    {
      "epoch": 2.0927803275356918,
      "grad_norm": 0.29410219192504883,
      "learning_rate": 9.266957330576015e-05,
      "loss": 1.452,
      "step": 56960
    },
    {
      "epoch": 2.093322268701204,
      "grad_norm": 0.3743274211883545,
      "learning_rate": 9.26609067329274e-05,
      "loss": 1.4552,
      "step": 56970
    },
    {
      "epoch": 2.0938642098667164,
      "grad_norm": 0.5883728265762329,
      "learning_rate": 9.265223549482505e-05,
      "loss": 1.4578,
      "step": 56980
    },
    {
      "epoch": 2.0941351804494723,
      "eval_loss": 2.460325002670288,
      "eval_runtime": 22.2713,
      "eval_samples_per_second": 224.505,
      "eval_steps_per_second": 1.212,
      "step": 56985
    },
    {
      "epoch": 2.0944061510322287,
      "grad_norm": 0.3245681822299957,
      "learning_rate": 9.264355959252722e-05,
      "loss": 1.4606,
      "step": 56990
    },
    {
      "epoch": 2.0949480921977406,
      "grad_norm": 0.255845844745636,
      "learning_rate": 9.263487902710869e-05,
      "loss": 1.4688,
      "step": 57000
    },
    {
      "epoch": 2.095490033363253,
      "grad_norm": 0.17143049836158752,
      "learning_rate": 9.262619379964474e-05,
      "loss": 1.4586,
      "step": 57010
    },
    {
      "epoch": 2.096031974528765,
      "grad_norm": 0.2582264244556427,
      "learning_rate": 9.261750391121122e-05,
      "loss": 1.4485,
      "step": 57020
    },
    {
      "epoch": 2.0965739156942775,
      "grad_norm": 0.3058115541934967,
      "learning_rate": 9.260880936288466e-05,
      "loss": 1.4608,
      "step": 57030
    },
    {
      "epoch": 2.09711585685979,
      "grad_norm": 0.3971554934978485,
      "learning_rate": 9.260011015574207e-05,
      "loss": 1.4587,
      "step": 57040
    },
    {
      "epoch": 2.0976577980253017,
      "grad_norm": 0.21815316379070282,
      "learning_rate": 9.25914062908611e-05,
      "loss": 1.4634,
      "step": 57050
    },
    {
      "epoch": 2.098199739190814,
      "grad_norm": 0.18554462492465973,
      "learning_rate": 9.258269776931989e-05,
      "loss": 1.4547,
      "step": 57060
    },
    {
      "epoch": 2.0987416803563264,
      "grad_norm": 0.2115841954946518,
      "learning_rate": 9.257398459219727e-05,
      "loss": 1.4534,
      "step": 57070
    },
    {
      "epoch": 2.098850068589429,
      "eval_loss": 2.4590704441070557,
      "eval_runtime": 21.9818,
      "eval_samples_per_second": 227.461,
      "eval_steps_per_second": 1.228,
      "step": 57072
    },
    {
      "epoch": 2.0992836215218387,
      "grad_norm": 0.3519175350666046,
      "learning_rate": 9.256526676057257e-05,
      "loss": 1.4481,
      "step": 57080
    },
    {
      "epoch": 2.0998255626873505,
      "grad_norm": 0.3369831144809723,
      "learning_rate": 9.25565442755257e-05,
      "loss": 1.4422,
      "step": 57090
    },
    {
      "epoch": 2.100367503852863,
      "grad_norm": 0.3334982395172119,
      "learning_rate": 9.254781713813719e-05,
      "loss": 1.444,
      "step": 57100
    },
    {
      "epoch": 2.100909445018375,
      "grad_norm": 0.3039768636226654,
      "learning_rate": 9.253908534948811e-05,
      "loss": 1.4585,
      "step": 57110
    },
    {
      "epoch": 2.1014513861838875,
      "grad_norm": 0.3292928636074066,
      "learning_rate": 9.253034891066011e-05,
      "loss": 1.4428,
      "step": 57120
    },
    {
      "epoch": 2.1019933273494,
      "grad_norm": 0.2547774910926819,
      "learning_rate": 9.252160782273544e-05,
      "loss": 1.4511,
      "step": 57130
    },
    {
      "epoch": 2.1025352685149117,
      "grad_norm": 0.19933480024337769,
      "learning_rate": 9.251286208679688e-05,
      "loss": 1.456,
      "step": 57140
    },
    {
      "epoch": 2.103077209680424,
      "grad_norm": 0.20023803412914276,
      "learning_rate": 9.250411170392783e-05,
      "loss": 1.4615,
      "step": 57150
    },
    {
      "epoch": 2.103564956729385,
      "eval_loss": 2.44368839263916,
      "eval_runtime": 21.9837,
      "eval_samples_per_second": 227.441,
      "eval_steps_per_second": 1.228,
      "step": 57159
    },
    {
      "epoch": 2.1036191508459363,
      "grad_norm": 0.3000018894672394,
      "learning_rate": 9.249535667521225e-05,
      "loss": 1.4423,
      "step": 57160
    },
    {
      "epoch": 2.1041610920114486,
      "grad_norm": 0.18277384340763092,
      "learning_rate": 9.248659700173469e-05,
      "loss": 1.4596,
      "step": 57170
    },
    {
      "epoch": 2.1047030331769605,
      "grad_norm": 0.33772820234298706,
      "learning_rate": 9.247783268458023e-05,
      "loss": 1.4496,
      "step": 57180
    },
    {
      "epoch": 2.105244974342473,
      "grad_norm": 0.2701638340950012,
      "learning_rate": 9.246906372483456e-05,
      "loss": 1.4604,
      "step": 57190
    },
    {
      "epoch": 2.105786915507985,
      "grad_norm": 0.33614516258239746,
      "learning_rate": 9.246029012358398e-05,
      "loss": 1.4474,
      "step": 57200
    },
    {
      "epoch": 2.1063288566734975,
      "grad_norm": 0.18986083567142487,
      "learning_rate": 9.245151188191526e-05,
      "loss": 1.448,
      "step": 57210
    },
    {
      "epoch": 2.1068707978390098,
      "grad_norm": 0.19249024987220764,
      "learning_rate": 9.244272900091586e-05,
      "loss": 1.4549,
      "step": 57220
    },
    {
      "epoch": 2.1074127390045216,
      "grad_norm": 0.21236182749271393,
      "learning_rate": 9.243394148167376e-05,
      "loss": 1.4536,
      "step": 57230
    },
    {
      "epoch": 2.107954680170034,
      "grad_norm": 0.37698227167129517,
      "learning_rate": 9.242514932527751e-05,
      "loss": 1.4607,
      "step": 57240
    },
    {
      "epoch": 2.108279844869341,
      "eval_loss": 2.428705930709839,
      "eval_runtime": 21.9743,
      "eval_samples_per_second": 227.539,
      "eval_steps_per_second": 1.229,
      "step": 57246
    },
    {
      "epoch": 2.1084966213355463,
      "grad_norm": 0.26461127400398254,
      "learning_rate": 9.241635253281624e-05,
      "loss": 1.4558,
      "step": 57250
    },
    {
      "epoch": 2.1090385625010586,
      "grad_norm": 0.20196294784545898,
      "learning_rate": 9.240755110537966e-05,
      "loss": 1.4561,
      "step": 57260
    },
    {
      "epoch": 2.109580503666571,
      "grad_norm": 0.40879595279693604,
      "learning_rate": 9.239874504405806e-05,
      "loss": 1.4488,
      "step": 57270
    },
    {
      "epoch": 2.1101224448320828,
      "grad_norm": 0.2242775410413742,
      "learning_rate": 9.238993434994229e-05,
      "loss": 1.4513,
      "step": 57280
    },
    {
      "epoch": 2.110664385997595,
      "grad_norm": 0.25892990827560425,
      "learning_rate": 9.238111902412379e-05,
      "loss": 1.4625,
      "step": 57290
    },
    {
      "epoch": 2.1112063271631074,
      "grad_norm": 0.1938840001821518,
      "learning_rate": 9.237229906769456e-05,
      "loss": 1.4614,
      "step": 57300
    },
    {
      "epoch": 2.1117482683286197,
      "grad_norm": 0.2417358160018921,
      "learning_rate": 9.236347448174717e-05,
      "loss": 1.4616,
      "step": 57310
    },
    {
      "epoch": 2.1122902094941316,
      "grad_norm": 0.26493963599205017,
      "learning_rate": 9.235464526737477e-05,
      "loss": 1.4459,
      "step": 57320
    },
    {
      "epoch": 2.112832150659644,
      "grad_norm": 0.19729004800319672,
      "learning_rate": 9.234581142567112e-05,
      "loss": 1.4565,
      "step": 57330
    },
    {
      "epoch": 2.1129947330092977,
      "eval_loss": 2.458986520767212,
      "eval_runtime": 21.9806,
      "eval_samples_per_second": 227.473,
      "eval_steps_per_second": 1.228,
      "step": 57333
    },
    {
      "epoch": 2.1133740918251562,
      "grad_norm": 0.2965938448905945,
      "learning_rate": 9.233697295773046e-05,
      "loss": 1.4564,
      "step": 57340
    },
    {
      "epoch": 2.1139160329906685,
      "grad_norm": 0.281646728515625,
      "learning_rate": 9.23281298646477e-05,
      "loss": 1.4464,
      "step": 57350
    },
    {
      "epoch": 2.114457974156181,
      "grad_norm": 0.2671820819377899,
      "learning_rate": 9.231928214751828e-05,
      "loss": 1.4495,
      "step": 57360
    },
    {
      "epoch": 2.1149999153216927,
      "grad_norm": 0.20213744044303894,
      "learning_rate": 9.23104298074382e-05,
      "loss": 1.4672,
      "step": 57370
    },
    {
      "epoch": 2.115541856487205,
      "grad_norm": 0.3106236755847931,
      "learning_rate": 9.230157284550407e-05,
      "loss": 1.4501,
      "step": 57380
    },
    {
      "epoch": 2.1160837976527174,
      "grad_norm": 0.26210352778434753,
      "learning_rate": 9.229271126281304e-05,
      "loss": 1.454,
      "step": 57390
    },
    {
      "epoch": 2.1166257388182297,
      "grad_norm": 0.2264351099729538,
      "learning_rate": 9.228384506046286e-05,
      "loss": 1.4441,
      "step": 57400
    },
    {
      "epoch": 2.1171676799837416,
      "grad_norm": 0.19489522278308868,
      "learning_rate": 9.22749742395518e-05,
      "loss": 1.4577,
      "step": 57410
    },
    {
      "epoch": 2.117709621149254,
      "grad_norm": 0.20375080406665802,
      "learning_rate": 9.226609880117877e-05,
      "loss": 1.4462,
      "step": 57420
    },
    {
      "epoch": 2.117709621149254,
      "eval_loss": 2.4545302391052246,
      "eval_runtime": 21.9693,
      "eval_samples_per_second": 227.591,
      "eval_steps_per_second": 1.229,
      "step": 57420
    },
    {
      "epoch": 2.118251562314766,
      "grad_norm": 0.29413971304893494,
      "learning_rate": 9.225721874644324e-05,
      "loss": 1.465,
      "step": 57430
    },
    {
      "epoch": 2.1187935034802785,
      "grad_norm": 0.2669055461883545,
      "learning_rate": 9.224833407644519e-05,
      "loss": 1.454,
      "step": 57440
    },
    {
      "epoch": 2.119335444645791,
      "grad_norm": 0.2955906391143799,
      "learning_rate": 9.223944479228522e-05,
      "loss": 1.4574,
      "step": 57450
    },
    {
      "epoch": 2.1198773858113027,
      "grad_norm": 0.3815632462501526,
      "learning_rate": 9.223055089506451e-05,
      "loss": 1.4574,
      "step": 57460
    },
    {
      "epoch": 2.120419326976815,
      "grad_norm": 0.3034198582172394,
      "learning_rate": 9.22216523858848e-05,
      "loss": 1.4515,
      "step": 57470
    },
    {
      "epoch": 2.1209612681423273,
      "grad_norm": 0.22253744304180145,
      "learning_rate": 9.22127492658484e-05,
      "loss": 1.4485,
      "step": 57480
    },
    {
      "epoch": 2.1215032093078396,
      "grad_norm": 0.39392632246017456,
      "learning_rate": 9.220384153605816e-05,
      "loss": 1.4488,
      "step": 57490
    },
    {
      "epoch": 2.122045150473352,
      "grad_norm": 0.19380737841129303,
      "learning_rate": 9.219492919761758e-05,
      "loss": 1.448,
      "step": 57500
    },
    {
      "epoch": 2.1224245092892104,
      "eval_loss": 2.4559643268585205,
      "eval_runtime": 21.9784,
      "eval_samples_per_second": 227.496,
      "eval_steps_per_second": 1.228,
      "step": 57507
    },
    {
      "epoch": 2.122587091638864,
      "grad_norm": 0.1623711884021759,
      "learning_rate": 9.218601225163063e-05,
      "loss": 1.4629,
      "step": 57510
    },
    {
      "epoch": 2.123129032804376,
      "grad_norm": 0.27015969157218933,
      "learning_rate": 9.217709069920196e-05,
      "loss": 1.4466,
      "step": 57520
    },
    {
      "epoch": 2.1236709739698885,
      "grad_norm": 0.20501737296581268,
      "learning_rate": 9.216816454143667e-05,
      "loss": 1.4536,
      "step": 57530
    },
    {
      "epoch": 2.124212915135401,
      "grad_norm": 0.22715558111667633,
      "learning_rate": 9.215923377944055e-05,
      "loss": 1.4488,
      "step": 57540
    },
    {
      "epoch": 2.1247548563009127,
      "grad_norm": 0.27195459604263306,
      "learning_rate": 9.215029841431984e-05,
      "loss": 1.4442,
      "step": 57550
    },
    {
      "epoch": 2.125296797466425,
      "grad_norm": 0.2515755295753479,
      "learning_rate": 9.21413584471815e-05,
      "loss": 1.4584,
      "step": 57560
    },
    {
      "epoch": 2.1258387386319373,
      "grad_norm": 0.2585810720920563,
      "learning_rate": 9.213241387913289e-05,
      "loss": 1.4515,
      "step": 57570
    },
    {
      "epoch": 2.1263806797974496,
      "grad_norm": 0.3669833540916443,
      "learning_rate": 9.212346471128207e-05,
      "loss": 1.4564,
      "step": 57580
    },
    {
      "epoch": 2.1269226209629615,
      "grad_norm": 0.2513576149940491,
      "learning_rate": 9.211451094473764e-05,
      "loss": 1.4557,
      "step": 57590
    },
    {
      "epoch": 2.1271393974291666,
      "eval_loss": 2.456778049468994,
      "eval_runtime": 22.6763,
      "eval_samples_per_second": 220.495,
      "eval_steps_per_second": 1.191,
      "step": 57594
    },
    {
      "epoch": 2.127464562128474,
      "grad_norm": 0.2454679310321808,
      "learning_rate": 9.210555258060871e-05,
      "loss": 1.4549,
      "step": 57600
    },
    {
      "epoch": 2.128006503293986,
      "grad_norm": 0.3271390199661255,
      "learning_rate": 9.209658962000502e-05,
      "loss": 1.4548,
      "step": 57610
    },
    {
      "epoch": 2.1285484444594984,
      "grad_norm": 0.2133026272058487,
      "learning_rate": 9.208762206403687e-05,
      "loss": 1.4508,
      "step": 57620
    },
    {
      "epoch": 2.1290903856250107,
      "grad_norm": 0.20085935294628143,
      "learning_rate": 9.20786499138151e-05,
      "loss": 1.4557,
      "step": 57630
    },
    {
      "epoch": 2.1296323267905226,
      "grad_norm": 0.3363330662250519,
      "learning_rate": 9.206967317045119e-05,
      "loss": 1.4512,
      "step": 57640
    },
    {
      "epoch": 2.130174267956035,
      "grad_norm": 0.2005142867565155,
      "learning_rate": 9.20606918350571e-05,
      "loss": 1.4578,
      "step": 57650
    },
    {
      "epoch": 2.1307162091215472,
      "grad_norm": 0.32124650478363037,
      "learning_rate": 9.205170590874539e-05,
      "loss": 1.4407,
      "step": 57660
    },
    {
      "epoch": 2.1312581502870596,
      "grad_norm": 0.406781405210495,
      "learning_rate": 9.204271539262922e-05,
      "loss": 1.4442,
      "step": 57670
    },
    {
      "epoch": 2.131800091452572,
      "grad_norm": 0.19096383452415466,
      "learning_rate": 9.20337202878223e-05,
      "loss": 1.4632,
      "step": 57680
    },
    {
      "epoch": 2.1318542855691227,
      "eval_loss": 2.4498584270477295,
      "eval_runtime": 21.9755,
      "eval_samples_per_second": 227.526,
      "eval_steps_per_second": 1.229,
      "step": 57681
    },
    {
      "epoch": 2.1323420326180837,
      "grad_norm": 0.21396368741989136,
      "learning_rate": 9.202472059543888e-05,
      "loss": 1.4473,
      "step": 57690
    },
    {
      "epoch": 2.132883973783596,
      "grad_norm": 0.33212918043136597,
      "learning_rate": 9.201571631659384e-05,
      "loss": 1.4547,
      "step": 57700
    },
    {
      "epoch": 2.1334259149491084,
      "grad_norm": 0.3928041458129883,
      "learning_rate": 9.200670745240255e-05,
      "loss": 1.4552,
      "step": 57710
    },
    {
      "epoch": 2.1339678561146207,
      "grad_norm": 0.2717365622520447,
      "learning_rate": 9.199769400398101e-05,
      "loss": 1.4431,
      "step": 57720
    },
    {
      "epoch": 2.1345097972801326,
      "grad_norm": 0.19520139694213867,
      "learning_rate": 9.198867597244577e-05,
      "loss": 1.4426,
      "step": 57730
    },
    {
      "epoch": 2.135051738445645,
      "grad_norm": 0.4766680896282196,
      "learning_rate": 9.197965335891394e-05,
      "loss": 1.4433,
      "step": 57740
    },
    {
      "epoch": 2.135593679611157,
      "grad_norm": 0.5358844995498657,
      "learning_rate": 9.19706261645032e-05,
      "loss": 1.4486,
      "step": 57750
    },
    {
      "epoch": 2.1361356207766695,
      "grad_norm": 0.40660426020622253,
      "learning_rate": 9.196159439033179e-05,
      "loss": 1.4481,
      "step": 57760
    },
    {
      "epoch": 2.1365691737090793,
      "eval_loss": 2.4472365379333496,
      "eval_runtime": 21.9775,
      "eval_samples_per_second": 227.505,
      "eval_steps_per_second": 1.229,
      "step": 57768
    },
    {
      "epoch": 2.136677561942182,
      "grad_norm": 0.21572063863277435,
      "learning_rate": 9.195255803751855e-05,
      "loss": 1.4457,
      "step": 57770
    },
    {
      "epoch": 2.1372195031076937,
      "grad_norm": 0.29680702090263367,
      "learning_rate": 9.194351710718285e-05,
      "loss": 1.4545,
      "step": 57780
    },
    {
      "epoch": 2.137761444273206,
      "grad_norm": 0.27061325311660767,
      "learning_rate": 9.193447160044465e-05,
      "loss": 1.4451,
      "step": 57790
    },
    {
      "epoch": 2.1383033854387183,
      "grad_norm": 0.2027037888765335,
      "learning_rate": 9.192542151842447e-05,
      "loss": 1.4497,
      "step": 57800
    },
    {
      "epoch": 2.1388453266042307,
      "grad_norm": 0.20895957946777344,
      "learning_rate": 9.191636686224336e-05,
      "loss": 1.4519,
      "step": 57810
    },
    {
      "epoch": 2.1393872677697425,
      "grad_norm": 0.18710966408252716,
      "learning_rate": 9.1907307633023e-05,
      "loss": 1.4492,
      "step": 57820
    },
    {
      "epoch": 2.139929208935255,
      "grad_norm": 0.270098477602005,
      "learning_rate": 9.189824383188562e-05,
      "loss": 1.4581,
      "step": 57830
    },
    {
      "epoch": 2.140471150100767,
      "grad_norm": 0.21558840572834015,
      "learning_rate": 9.1889175459954e-05,
      "loss": 1.4398,
      "step": 57840
    },
    {
      "epoch": 2.1410130912662795,
      "grad_norm": 0.2560511827468872,
      "learning_rate": 9.188010251835147e-05,
      "loss": 1.4501,
      "step": 57850
    },
    {
      "epoch": 2.1412840618490354,
      "eval_loss": 2.448354482650757,
      "eval_runtime": 21.9774,
      "eval_samples_per_second": 227.506,
      "eval_steps_per_second": 1.229,
      "step": 57855
    },
    {
      "epoch": 2.141555032431792,
      "grad_norm": 0.3320198357105255,
      "learning_rate": 9.187102500820195e-05,
      "loss": 1.4466,
      "step": 57860
    },
    {
      "epoch": 2.1420969735973037,
      "grad_norm": 0.37881776690483093,
      "learning_rate": 9.186194293062993e-05,
      "loss": 1.446,
      "step": 57870
    },
    {
      "epoch": 2.142638914762816,
      "grad_norm": 0.38807186484336853,
      "learning_rate": 9.185285628676045e-05,
      "loss": 1.4483,
      "step": 57880
    },
    {
      "epoch": 2.1431808559283283,
      "grad_norm": 0.27356836199760437,
      "learning_rate": 9.184376507771916e-05,
      "loss": 1.4483,
      "step": 57890
    },
    {
      "epoch": 2.1437227970938406,
      "grad_norm": 0.31562650203704834,
      "learning_rate": 9.18346693046322e-05,
      "loss": 1.4557,
      "step": 57900
    },
    {
      "epoch": 2.144264738259353,
      "grad_norm": 0.3229880630970001,
      "learning_rate": 9.182556896862632e-05,
      "loss": 1.4511,
      "step": 57910
    },
    {
      "epoch": 2.144806679424865,
      "grad_norm": 0.36681440472602844,
      "learning_rate": 9.181646407082885e-05,
      "loss": 1.4448,
      "step": 57920
    },
    {
      "epoch": 2.145348620590377,
      "grad_norm": 0.4348771274089813,
      "learning_rate": 9.180735461236764e-05,
      "loss": 1.4456,
      "step": 57930
    },
    {
      "epoch": 2.1458905617558894,
      "grad_norm": 0.35928380489349365,
      "learning_rate": 9.179824059437113e-05,
      "loss": 1.448,
      "step": 57940
    },
    {
      "epoch": 2.145998949988992,
      "eval_loss": 2.451070547103882,
      "eval_runtime": 21.9847,
      "eval_samples_per_second": 227.431,
      "eval_steps_per_second": 1.228,
      "step": 57942
    },
    {
      "epoch": 2.1464325029214018,
      "grad_norm": 0.3611794710159302,
      "learning_rate": 9.178912201796836e-05,
      "loss": 1.4492,
      "step": 57950
    },
    {
      "epoch": 2.1469744440869136,
      "grad_norm": 0.21141478419303894,
      "learning_rate": 9.177999888428886e-05,
      "loss": 1.4474,
      "step": 57960
    },
    {
      "epoch": 2.147516385252426,
      "grad_norm": 0.2665724456310272,
      "learning_rate": 9.17708711944628e-05,
      "loss": 1.4562,
      "step": 57970
    },
    {
      "epoch": 2.1480583264179383,
      "grad_norm": 0.3022826015949249,
      "learning_rate": 9.176173894962086e-05,
      "loss": 1.4337,
      "step": 57980
    },
    {
      "epoch": 2.1486002675834506,
      "grad_norm": 0.27852869033813477,
      "learning_rate": 9.175260215089429e-05,
      "loss": 1.4481,
      "step": 57990
    },
    {
      "epoch": 2.149142208748963,
      "grad_norm": 0.22829607129096985,
      "learning_rate": 9.174346079941492e-05,
      "loss": 1.4468,
      "step": 58000
    },
    {
      "epoch": 2.1496841499144748,
      "grad_norm": 0.24403288960456848,
      "learning_rate": 9.173431489631517e-05,
      "loss": 1.4571,
      "step": 58010
    },
    {
      "epoch": 2.150226091079987,
      "grad_norm": 0.23664097487926483,
      "learning_rate": 9.172516444272796e-05,
      "loss": 1.4461,
      "step": 58020
    },
    {
      "epoch": 2.150713838128948,
      "eval_loss": 2.4530248641967773,
      "eval_runtime": 21.9794,
      "eval_samples_per_second": 227.486,
      "eval_steps_per_second": 1.228,
      "step": 58029
    },
    {
      "epoch": 2.1507680322454994,
      "grad_norm": 0.19086076319217682,
      "learning_rate": 9.171600943978683e-05,
      "loss": 1.4514,
      "step": 58030
    },
    {
      "epoch": 2.1513099734110117,
      "grad_norm": 0.22935578227043152,
      "learning_rate": 9.170684988862586e-05,
      "loss": 1.456,
      "step": 58040
    },
    {
      "epoch": 2.1518519145765236,
      "grad_norm": 0.21930955350399017,
      "learning_rate": 9.16976857903797e-05,
      "loss": 1.4513,
      "step": 58050
    },
    {
      "epoch": 2.152393855742036,
      "grad_norm": 0.2644178867340088,
      "learning_rate": 9.168851714618352e-05,
      "loss": 1.4377,
      "step": 58060
    },
    {
      "epoch": 2.152935796907548,
      "grad_norm": 0.3892003297805786,
      "learning_rate": 9.167934395717314e-05,
      "loss": 1.4409,
      "step": 58070
    },
    {
      "epoch": 2.1534777380730605,
      "grad_norm": 0.40704429149627686,
      "learning_rate": 9.167016622448488e-05,
      "loss": 1.4479,
      "step": 58080
    },
    {
      "epoch": 2.154019679238573,
      "grad_norm": 0.33170846104621887,
      "learning_rate": 9.166098394925562e-05,
      "loss": 1.4459,
      "step": 58090
    },
    {
      "epoch": 2.1545616204040847,
      "grad_norm": 0.18887291848659515,
      "learning_rate": 9.165179713262286e-05,
      "loss": 1.4583,
      "step": 58100
    },
    {
      "epoch": 2.155103561569597,
      "grad_norm": 0.2687866687774658,
      "learning_rate": 9.164260577572456e-05,
      "loss": 1.4504,
      "step": 58110
    },
    {
      "epoch": 2.1554287262689042,
      "eval_loss": 2.450887680053711,
      "eval_runtime": 21.9808,
      "eval_samples_per_second": 227.471,
      "eval_steps_per_second": 1.228,
      "step": 58116
    },
    {
      "epoch": 2.1556455027351094,
      "grad_norm": 0.21473592519760132,
      "learning_rate": 9.163340987969938e-05,
      "loss": 1.4491,
      "step": 58120
    },
    {
      "epoch": 2.1561874439006217,
      "grad_norm": 0.1952633261680603,
      "learning_rate": 9.162420944568641e-05,
      "loss": 1.4505,
      "step": 58130
    },
    {
      "epoch": 2.156729385066134,
      "grad_norm": 0.21670879423618317,
      "learning_rate": 9.161500447482539e-05,
      "loss": 1.4447,
      "step": 58140
    },
    {
      "epoch": 2.157271326231646,
      "grad_norm": 0.27638787031173706,
      "learning_rate": 9.160579496825656e-05,
      "loss": 1.4444,
      "step": 58150
    },
    {
      "epoch": 2.157813267397158,
      "grad_norm": 0.2775803804397583,
      "learning_rate": 9.15965809271208e-05,
      "loss": 1.4442,
      "step": 58160
    },
    {
      "epoch": 2.1583552085626705,
      "grad_norm": 0.2165958732366562,
      "learning_rate": 9.158736235255949e-05,
      "loss": 1.4533,
      "step": 58170
    },
    {
      "epoch": 2.158897149728183,
      "grad_norm": 0.3519152104854584,
      "learning_rate": 9.157813924571455e-05,
      "loss": 1.4455,
      "step": 58180
    },
    {
      "epoch": 2.1594390908936947,
      "grad_norm": 0.23166511952877045,
      "learning_rate": 9.156891160772854e-05,
      "loss": 1.4437,
      "step": 58190
    },
    {
      "epoch": 2.159981032059207,
      "grad_norm": 0.32799333333969116,
      "learning_rate": 9.155967943974453e-05,
      "loss": 1.4477,
      "step": 58200
    },
    {
      "epoch": 2.160143614408861,
      "eval_loss": 2.4706532955169678,
      "eval_runtime": 21.9762,
      "eval_samples_per_second": 227.518,
      "eval_steps_per_second": 1.229,
      "step": 58203
    },
    {
      "epoch": 2.1605229732247193,
      "grad_norm": 0.26824527978897095,
      "learning_rate": 9.155044274290614e-05,
      "loss": 1.4423,
      "step": 58210
    },
    {
      "epoch": 2.1610649143902316,
      "grad_norm": 0.3500238358974457,
      "learning_rate": 9.15412015183576e-05,
      "loss": 1.4509,
      "step": 58220
    },
    {
      "epoch": 2.1616068555557435,
      "grad_norm": 0.38133561611175537,
      "learning_rate": 9.153195576724367e-05,
      "loss": 1.4519,
      "step": 58230
    },
    {
      "epoch": 2.162148796721256,
      "grad_norm": 0.27178552746772766,
      "learning_rate": 9.152270549070964e-05,
      "loss": 1.4541,
      "step": 58240
    },
    {
      "epoch": 2.162690737886768,
      "grad_norm": 0.24917040765285492,
      "learning_rate": 9.151345068990146e-05,
      "loss": 1.449,
      "step": 58250
    },
    {
      "epoch": 2.1632326790522804,
      "grad_norm": 0.39437514543533325,
      "learning_rate": 9.150419136596551e-05,
      "loss": 1.4438,
      "step": 58260
    },
    {
      "epoch": 2.1637746202177928,
      "grad_norm": 0.346034437417984,
      "learning_rate": 9.149492752004882e-05,
      "loss": 1.4544,
      "step": 58270
    },
    {
      "epoch": 2.1643165613833046,
      "grad_norm": 0.3629694879055023,
      "learning_rate": 9.148565915329896e-05,
      "loss": 1.4454,
      "step": 58280
    },
    {
      "epoch": 2.164858502548817,
      "grad_norm": 0.20559094846248627,
      "learning_rate": 9.147638626686404e-05,
      "loss": 1.4417,
      "step": 58290
    },
    {
      "epoch": 2.164858502548817,
      "eval_loss": 2.461625099182129,
      "eval_runtime": 21.9403,
      "eval_samples_per_second": 227.891,
      "eval_steps_per_second": 1.231,
      "step": 58290
    },
    {
      "epoch": 2.1654004437143293,
      "grad_norm": 0.1901310533285141,
      "learning_rate": 9.146710886189276e-05,
      "loss": 1.4454,
      "step": 58300
    },
    {
      "epoch": 2.1659423848798416,
      "grad_norm": 0.2069934457540512,
      "learning_rate": 9.145782693953435e-05,
      "loss": 1.4521,
      "step": 58310
    },
    {
      "epoch": 2.166484326045354,
      "grad_norm": 0.34269529581069946,
      "learning_rate": 9.144854050093863e-05,
      "loss": 1.4463,
      "step": 58320
    },
    {
      "epoch": 2.1670262672108658,
      "grad_norm": 0.2547290325164795,
      "learning_rate": 9.143924954725595e-05,
      "loss": 1.4447,
      "step": 58330
    },
    {
      "epoch": 2.167568208376378,
      "grad_norm": 0.2823292016983032,
      "learning_rate": 9.142995407963724e-05,
      "loss": 1.4402,
      "step": 58340
    },
    {
      "epoch": 2.1681101495418904,
      "grad_norm": 0.3173072934150696,
      "learning_rate": 9.1420654099234e-05,
      "loss": 1.4404,
      "step": 58350
    },
    {
      "epoch": 2.1686520907074027,
      "grad_norm": 0.23170779645442963,
      "learning_rate": 9.141134960719824e-05,
      "loss": 1.4542,
      "step": 58360
    },
    {
      "epoch": 2.1691940318729146,
      "grad_norm": 0.33306044340133667,
      "learning_rate": 9.140204060468257e-05,
      "loss": 1.4494,
      "step": 58370
    },
    {
      "epoch": 2.1695733906887735,
      "eval_loss": 2.4640088081359863,
      "eval_runtime": 21.9887,
      "eval_samples_per_second": 227.39,
      "eval_steps_per_second": 1.228,
      "step": 58377
    },
    {
      "epoch": 2.169735973038427,
      "grad_norm": 0.23135629296302795,
      "learning_rate": 9.139272709284015e-05,
      "loss": 1.4456,
      "step": 58380
    },
    {
      "epoch": 2.1702779142039392,
      "grad_norm": 0.22279199957847595,
      "learning_rate": 9.138340907282472e-05,
      "loss": 1.4514,
      "step": 58390
    },
    {
      "epoch": 2.1708198553694515,
      "grad_norm": 0.27086111903190613,
      "learning_rate": 9.137408654579051e-05,
      "loss": 1.4481,
      "step": 58400
    },
    {
      "epoch": 2.171361796534964,
      "grad_norm": 0.21697980165481567,
      "learning_rate": 9.136475951289239e-05,
      "loss": 1.4424,
      "step": 58410
    },
    {
      "epoch": 2.1719037377004757,
      "grad_norm": 0.32999521493911743,
      "learning_rate": 9.135542797528574e-05,
      "loss": 1.4552,
      "step": 58420
    },
    {
      "epoch": 2.172445678865988,
      "grad_norm": 0.3084947466850281,
      "learning_rate": 9.134609193412652e-05,
      "loss": 1.433,
      "step": 58430
    },
    {
      "epoch": 2.1729876200315004,
      "grad_norm": 0.21755863726139069,
      "learning_rate": 9.13367513905712e-05,
      "loss": 1.4449,
      "step": 58440
    },
    {
      "epoch": 2.1735295611970127,
      "grad_norm": 0.1901983916759491,
      "learning_rate": 9.13274063457769e-05,
      "loss": 1.4482,
      "step": 58450
    },
    {
      "epoch": 2.1740715023625246,
      "grad_norm": 0.18798936903476715,
      "learning_rate": 9.131805680090122e-05,
      "loss": 1.4472,
      "step": 58460
    },
    {
      "epoch": 2.1742882788287297,
      "eval_loss": 2.459465503692627,
      "eval_runtime": 21.9819,
      "eval_samples_per_second": 227.459,
      "eval_steps_per_second": 1.228,
      "step": 58464
    },
    {
      "epoch": 2.174613443528037,
      "grad_norm": 0.2203335165977478,
      "learning_rate": 9.130870275710235e-05,
      "loss": 1.4538,
      "step": 58470
    },
    {
      "epoch": 2.175155384693549,
      "grad_norm": 0.22482015192508698,
      "learning_rate": 9.1299344215539e-05,
      "loss": 1.4545,
      "step": 58480
    },
    {
      "epoch": 2.1756973258590615,
      "grad_norm": 0.3685661852359772,
      "learning_rate": 9.12899811773705e-05,
      "loss": 1.441,
      "step": 58490
    },
    {
      "epoch": 2.176239267024574,
      "grad_norm": 0.27606692910194397,
      "learning_rate": 9.128061364375668e-05,
      "loss": 1.4412,
      "step": 58500
    },
    {
      "epoch": 2.1767812081900857,
      "grad_norm": 0.2645034193992615,
      "learning_rate": 9.127124161585795e-05,
      "loss": 1.4463,
      "step": 58510
    },
    {
      "epoch": 2.177323149355598,
      "grad_norm": 0.4667447805404663,
      "learning_rate": 9.126186509483529e-05,
      "loss": 1.4512,
      "step": 58520
    },
    {
      "epoch": 2.1778650905211103,
      "grad_norm": 0.5272287726402283,
      "learning_rate": 9.125248408185021e-05,
      "loss": 1.4451,
      "step": 58530
    },
    {
      "epoch": 2.1784070316866226,
      "grad_norm": 0.2895019054412842,
      "learning_rate": 9.12430985780648e-05,
      "loss": 1.4515,
      "step": 58540
    },
    {
      "epoch": 2.178948972852135,
      "grad_norm": 0.25505638122558594,
      "learning_rate": 9.123370858464169e-05,
      "loss": 1.4515,
      "step": 58550
    },
    {
      "epoch": 2.179003166968686,
      "eval_loss": 2.462846279144287,
      "eval_runtime": 21.9798,
      "eval_samples_per_second": 227.481,
      "eval_steps_per_second": 1.228,
      "step": 58551
    },
    {
      "epoch": 2.179490914017647,
      "grad_norm": 0.24811875820159912,
      "learning_rate": 9.122431410274406e-05,
      "loss": 1.4464,
      "step": 58560
    },
    {
      "epoch": 2.180032855183159,
      "grad_norm": 0.22253675758838654,
      "learning_rate": 9.12149151335357e-05,
      "loss": 1.4418,
      "step": 58570
    },
    {
      "epoch": 2.1805747963486715,
      "grad_norm": 0.3881581127643585,
      "learning_rate": 9.120551167818084e-05,
      "loss": 1.4429,
      "step": 58580
    },
    {
      "epoch": 2.181116737514184,
      "grad_norm": 0.19040873646736145,
      "learning_rate": 9.119610373784442e-05,
      "loss": 1.4412,
      "step": 58590
    },
    {
      "epoch": 2.1816586786796957,
      "grad_norm": 0.18105448782444,
      "learning_rate": 9.118669131369179e-05,
      "loss": 1.4504,
      "step": 58600
    },
    {
      "epoch": 2.182200619845208,
      "grad_norm": 0.23432651162147522,
      "learning_rate": 9.117727440688896e-05,
      "loss": 1.4373,
      "step": 58610
    },
    {
      "epoch": 2.1827425610107203,
      "grad_norm": 0.22334584593772888,
      "learning_rate": 9.116785301860244e-05,
      "loss": 1.4414,
      "step": 58620
    },
    {
      "epoch": 2.1832845021762326,
      "grad_norm": 0.3361223638057709,
      "learning_rate": 9.115842714999931e-05,
      "loss": 1.4394,
      "step": 58630
    },
    {
      "epoch": 2.1837180551086424,
      "eval_loss": 2.464482545852661,
      "eval_runtime": 21.9821,
      "eval_samples_per_second": 227.458,
      "eval_steps_per_second": 1.228,
      "step": 58638
    },
    {
      "epoch": 2.183826443341745,
      "grad_norm": 0.5349717140197754,
      "learning_rate": 9.11489968022472e-05,
      "loss": 1.4439,
      "step": 58640
    },
    {
      "epoch": 2.184368384507257,
      "grad_norm": 0.5961711406707764,
      "learning_rate": 9.113956197651434e-05,
      "loss": 1.4424,
      "step": 58650
    },
    {
      "epoch": 2.184910325672769,
      "grad_norm": 0.3221561312675476,
      "learning_rate": 9.113012267396943e-05,
      "loss": 1.4393,
      "step": 58660
    },
    {
      "epoch": 2.1854522668382814,
      "grad_norm": 0.36589309573173523,
      "learning_rate": 9.112067889578178e-05,
      "loss": 1.4436,
      "step": 58670
    },
    {
      "epoch": 2.1859942080037937,
      "grad_norm": 0.21223586797714233,
      "learning_rate": 9.111123064312125e-05,
      "loss": 1.4466,
      "step": 58680
    },
    {
      "epoch": 2.1865361491693056,
      "grad_norm": 0.21100690960884094,
      "learning_rate": 9.110177791715824e-05,
      "loss": 1.4415,
      "step": 58690
    },
    {
      "epoch": 2.187078090334818,
      "grad_norm": 0.3193996548652649,
      "learning_rate": 9.109232071906373e-05,
      "loss": 1.4573,
      "step": 58700
    },
    {
      "epoch": 2.1876200315003302,
      "grad_norm": 0.18448546528816223,
      "learning_rate": 9.108285905000922e-05,
      "loss": 1.4535,
      "step": 58710
    },
    {
      "epoch": 2.1881619726658426,
      "grad_norm": 0.3342556953430176,
      "learning_rate": 9.107339291116679e-05,
      "loss": 1.4356,
      "step": 58720
    },
    {
      "epoch": 2.1884329432485985,
      "eval_loss": 2.4611220359802246,
      "eval_runtime": 21.982,
      "eval_samples_per_second": 227.459,
      "eval_steps_per_second": 1.228,
      "step": 58725
    },
    {
      "epoch": 2.188703913831355,
      "grad_norm": 0.43229833245277405,
      "learning_rate": 9.106392230370906e-05,
      "loss": 1.4644,
      "step": 58730
    },
    {
      "epoch": 2.1892458549968667,
      "grad_norm": 0.36450695991516113,
      "learning_rate": 9.10544472288092e-05,
      "loss": 1.4381,
      "step": 58740
    },
    {
      "epoch": 2.189787796162379,
      "grad_norm": 0.18473106622695923,
      "learning_rate": 9.104496768764096e-05,
      "loss": 1.4404,
      "step": 58750
    },
    {
      "epoch": 2.1903297373278914,
      "grad_norm": 0.3914441168308258,
      "learning_rate": 9.103548368137863e-05,
      "loss": 1.4448,
      "step": 58760
    },
    {
      "epoch": 2.1908716784934037,
      "grad_norm": 0.38350018858909607,
      "learning_rate": 9.102599521119701e-05,
      "loss": 1.4437,
      "step": 58770
    },
    {
      "epoch": 2.191413619658916,
      "grad_norm": 0.1921490877866745,
      "learning_rate": 9.101650227827152e-05,
      "loss": 1.4433,
      "step": 58780
    },
    {
      "epoch": 2.191955560824428,
      "grad_norm": 0.25018230080604553,
      "learning_rate": 9.100700488377809e-05,
      "loss": 1.4451,
      "step": 58790
    },
    {
      "epoch": 2.19249750198994,
      "grad_norm": 0.5067526698112488,
      "learning_rate": 9.099750302889323e-05,
      "loss": 1.4481,
      "step": 58800
    },
    {
      "epoch": 2.1930394431554525,
      "grad_norm": 0.27210572361946106,
      "learning_rate": 9.098799671479397e-05,
      "loss": 1.4553,
      "step": 58810
    },
    {
      "epoch": 2.193147831388555,
      "eval_loss": 2.46246600151062,
      "eval_runtime": 21.9768,
      "eval_samples_per_second": 227.512,
      "eval_steps_per_second": 1.229,
      "step": 58812
    },
    {
      "epoch": 2.193581384320965,
      "grad_norm": 0.1989605873823166,
      "learning_rate": 9.097848594265793e-05,
      "loss": 1.4463,
      "step": 58820
    },
    {
      "epoch": 2.1941233254864767,
      "grad_norm": 0.18576236069202423,
      "learning_rate": 9.096897071366326e-05,
      "loss": 1.4411,
      "step": 58830
    },
    {
      "epoch": 2.194665266651989,
      "grad_norm": 0.2913901209831238,
      "learning_rate": 9.095945102898865e-05,
      "loss": 1.4448,
      "step": 58840
    },
    {
      "epoch": 2.1952072078175013,
      "grad_norm": 0.5270187854766846,
      "learning_rate": 9.094992688981337e-05,
      "loss": 1.454,
      "step": 58850
    },
    {
      "epoch": 2.1957491489830137,
      "grad_norm": 0.2984652519226074,
      "learning_rate": 9.094039829731721e-05,
      "loss": 1.4452,
      "step": 58860
    },
    {
      "epoch": 2.1962910901485255,
      "grad_norm": 0.3765738308429718,
      "learning_rate": 9.093086525268059e-05,
      "loss": 1.4389,
      "step": 58870
    },
    {
      "epoch": 2.196833031314038,
      "grad_norm": 0.31733348965644836,
      "learning_rate": 9.092132775708432e-05,
      "loss": 1.4405,
      "step": 58880
    },
    {
      "epoch": 2.19737497247955,
      "grad_norm": 0.19566883146762848,
      "learning_rate": 9.091178581170996e-05,
      "loss": 1.451,
      "step": 58890
    },
    {
      "epoch": 2.197862719528511,
      "eval_loss": 2.460601568222046,
      "eval_runtime": 21.9826,
      "eval_samples_per_second": 227.453,
      "eval_steps_per_second": 1.228,
      "step": 58899
    },
    {
      "epoch": 2.1979169136450625,
      "grad_norm": 0.2175428867340088,
      "learning_rate": 9.090223941773949e-05,
      "loss": 1.4446,
      "step": 58900
    },
    {
      "epoch": 2.198458854810575,
      "grad_norm": 0.17789192497730255,
      "learning_rate": 9.089268857635546e-05,
      "loss": 1.4443,
      "step": 58910
    },
    {
      "epoch": 2.1990007959760867,
      "grad_norm": 0.17181305587291718,
      "learning_rate": 9.0883133288741e-05,
      "loss": 1.4535,
      "step": 58920
    },
    {
      "epoch": 2.199542737141599,
      "grad_norm": 0.3057771325111389,
      "learning_rate": 9.087357355607977e-05,
      "loss": 1.4402,
      "step": 58930
    },
    {
      "epoch": 2.2000846783071113,
      "grad_norm": 0.3295128643512726,
      "learning_rate": 9.086400937955601e-05,
      "loss": 1.4578,
      "step": 58940
    },
    {
      "epoch": 2.2006266194726236,
      "grad_norm": 0.3450293242931366,
      "learning_rate": 9.085444076035448e-05,
      "loss": 1.451,
      "step": 58950
    },
    {
      "epoch": 2.201168560638136,
      "grad_norm": 0.31713640689849854,
      "learning_rate": 9.084486769966047e-05,
      "loss": 1.4599,
      "step": 58960
    },
    {
      "epoch": 2.201710501803648,
      "grad_norm": 0.23452381789684296,
      "learning_rate": 9.083529019865988e-05,
      "loss": 1.4549,
      "step": 58970
    },
    {
      "epoch": 2.20225244296916,
      "grad_norm": 0.19139593839645386,
      "learning_rate": 9.082570825853912e-05,
      "loss": 1.4405,
      "step": 58980
    },
    {
      "epoch": 2.2025776076684673,
      "eval_loss": 2.4606852531433105,
      "eval_runtime": 21.9818,
      "eval_samples_per_second": 227.461,
      "eval_steps_per_second": 1.228,
      "step": 58986
    },
    {
      "epoch": 2.2027943841346724,
      "grad_norm": 0.20362070202827454,
      "learning_rate": 9.081612188048518e-05,
      "loss": 1.4418,
      "step": 58990
    },
    {
      "epoch": 2.2033363253001848,
      "grad_norm": 0.24732381105422974,
      "learning_rate": 9.080653106568555e-05,
      "loss": 1.4289,
      "step": 59000
    },
    {
      "epoch": 2.2038782664656966,
      "grad_norm": 0.2842128872871399,
      "learning_rate": 9.07969358153283e-05,
      "loss": 1.4486,
      "step": 59010
    },
    {
      "epoch": 2.204420207631209,
      "grad_norm": 0.4157765805721283,
      "learning_rate": 9.078733613060205e-05,
      "loss": 1.4461,
      "step": 59020
    },
    {
      "epoch": 2.2049621487967213,
      "grad_norm": 0.5429375171661377,
      "learning_rate": 9.0777732012696e-05,
      "loss": 1.4433,
      "step": 59030
    },
    {
      "epoch": 2.2055040899622336,
      "grad_norm": 0.29770663380622864,
      "learning_rate": 9.076812346279982e-05,
      "loss": 1.4464,
      "step": 59040
    },
    {
      "epoch": 2.206046031127746,
      "grad_norm": 0.22405195236206055,
      "learning_rate": 9.075851048210379e-05,
      "loss": 1.4495,
      "step": 59050
    },
    {
      "epoch": 2.2065879722932578,
      "grad_norm": 0.43941786885261536,
      "learning_rate": 9.074889307179873e-05,
      "loss": 1.4481,
      "step": 59060
    },
    {
      "epoch": 2.20712991345877,
      "grad_norm": 0.20217150449752808,
      "learning_rate": 9.073927123307602e-05,
      "loss": 1.4482,
      "step": 59070
    },
    {
      "epoch": 2.207292495808424,
      "eval_loss": 2.457340717315674,
      "eval_runtime": 21.9825,
      "eval_samples_per_second": 227.453,
      "eval_steps_per_second": 1.228,
      "step": 59073
    },
    {
      "epoch": 2.2076718546242824,
      "grad_norm": 0.34927546977996826,
      "learning_rate": 9.072964496712755e-05,
      "loss": 1.4482,
      "step": 59080
    },
    {
      "epoch": 2.2082137957897947,
      "grad_norm": 0.2522357702255249,
      "learning_rate": 9.072001427514578e-05,
      "loss": 1.4505,
      "step": 59090
    },
    {
      "epoch": 2.2087557369553066,
      "grad_norm": 0.2815626263618469,
      "learning_rate": 9.07103791583237e-05,
      "loss": 1.4479,
      "step": 59100
    },
    {
      "epoch": 2.209297678120819,
      "grad_norm": 0.22079609334468842,
      "learning_rate": 9.070073961785491e-05,
      "loss": 1.4494,
      "step": 59110
    },
    {
      "epoch": 2.209839619286331,
      "grad_norm": 0.2886086702346802,
      "learning_rate": 9.06910956549335e-05,
      "loss": 1.4356,
      "step": 59120
    },
    {
      "epoch": 2.2103815604518435,
      "grad_norm": 0.39057257771492004,
      "learning_rate": 9.068144727075409e-05,
      "loss": 1.4445,
      "step": 59130
    },
    {
      "epoch": 2.210923501617356,
      "grad_norm": 0.41756001114845276,
      "learning_rate": 9.067179446651195e-05,
      "loss": 1.4419,
      "step": 59140
    },
    {
      "epoch": 2.2114654427828677,
      "grad_norm": 0.30022770166397095,
      "learning_rate": 9.066213724340274e-05,
      "loss": 1.4551,
      "step": 59150
    },
    {
      "epoch": 2.21200738394838,
      "grad_norm": 0.20884348452091217,
      "learning_rate": 9.065247560262282e-05,
      "loss": 1.456,
      "step": 59160
    },
    {
      "epoch": 2.21200738394838,
      "eval_loss": 2.453835964202881,
      "eval_runtime": 21.9755,
      "eval_samples_per_second": 227.526,
      "eval_steps_per_second": 1.229,
      "step": 59160
    },
    {
      "epoch": 2.2125493251138924,
      "grad_norm": 0.3450928330421448,
      "learning_rate": 9.0642809545369e-05,
      "loss": 1.4548,
      "step": 59170
    },
    {
      "epoch": 2.2130912662794047,
      "grad_norm": 0.3694857358932495,
      "learning_rate": 9.063313907283868e-05,
      "loss": 1.4562,
      "step": 59180
    },
    {
      "epoch": 2.213633207444917,
      "grad_norm": 0.28221815824508667,
      "learning_rate": 9.06234641862298e-05,
      "loss": 1.4502,
      "step": 59190
    },
    {
      "epoch": 2.214175148610429,
      "grad_norm": 0.207607701420784,
      "learning_rate": 9.061378488674084e-05,
      "loss": 1.4369,
      "step": 59200
    },
    {
      "epoch": 2.214717089775941,
      "grad_norm": 0.3098548948764801,
      "learning_rate": 9.060410117557083e-05,
      "loss": 1.4371,
      "step": 59210
    },
    {
      "epoch": 2.2152590309414535,
      "grad_norm": 0.3138673007488251,
      "learning_rate": 9.059441305391932e-05,
      "loss": 1.443,
      "step": 59220
    },
    {
      "epoch": 2.215800972106966,
      "grad_norm": 0.3069899380207062,
      "learning_rate": 9.058472052298649e-05,
      "loss": 1.4424,
      "step": 59230
    },
    {
      "epoch": 2.2163429132724777,
      "grad_norm": 0.4762488305568695,
      "learning_rate": 9.057502358397296e-05,
      "loss": 1.4312,
      "step": 59240
    },
    {
      "epoch": 2.2167222720883366,
      "eval_loss": 2.4593636989593506,
      "eval_runtime": 21.9809,
      "eval_samples_per_second": 227.471,
      "eval_steps_per_second": 1.228,
      "step": 59247
    },
    {
      "epoch": 2.21688485443799,
      "grad_norm": 0.2531556487083435,
      "learning_rate": 9.056532223807995e-05,
      "loss": 1.4506,
      "step": 59250
    },
    {
      "epoch": 2.2174267956035023,
      "grad_norm": 0.42446276545524597,
      "learning_rate": 9.055561648650924e-05,
      "loss": 1.4452,
      "step": 59260
    },
    {
      "epoch": 2.2179687367690146,
      "grad_norm": 0.2571756839752197,
      "learning_rate": 9.054590633046313e-05,
      "loss": 1.4384,
      "step": 59270
    },
    {
      "epoch": 2.2185106779345265,
      "grad_norm": 0.29887983202934265,
      "learning_rate": 9.053619177114446e-05,
      "loss": 1.4408,
      "step": 59280
    },
    {
      "epoch": 2.219052619100039,
      "grad_norm": 0.22026404738426208,
      "learning_rate": 9.052647280975664e-05,
      "loss": 1.4522,
      "step": 59290
    },
    {
      "epoch": 2.219594560265551,
      "grad_norm": 0.20031945407390594,
      "learning_rate": 9.051674944750362e-05,
      "loss": 1.4536,
      "step": 59300
    },
    {
      "epoch": 2.2201365014310634,
      "grad_norm": 0.2620982527732849,
      "learning_rate": 9.050702168558987e-05,
      "loss": 1.4432,
      "step": 59310
    },
    {
      "epoch": 2.2206784425965758,
      "grad_norm": 0.35358208417892456,
      "learning_rate": 9.049728952522042e-05,
      "loss": 1.444,
      "step": 59320
    },
    {
      "epoch": 2.2212203837620876,
      "grad_norm": 0.36926135420799255,
      "learning_rate": 9.048755296760087e-05,
      "loss": 1.4593,
      "step": 59330
    },
    {
      "epoch": 2.2214371602282927,
      "eval_loss": 2.448256015777588,
      "eval_runtime": 21.981,
      "eval_samples_per_second": 227.47,
      "eval_steps_per_second": 1.228,
      "step": 59334
    },
    {
      "epoch": 2.2217623249276,
      "grad_norm": 0.17719149589538574,
      "learning_rate": 9.047781201393731e-05,
      "loss": 1.4395,
      "step": 59340
    },
    {
      "epoch": 2.2223042660931123,
      "grad_norm": 0.23782901465892792,
      "learning_rate": 9.046806666543645e-05,
      "loss": 1.4385,
      "step": 59350
    },
    {
      "epoch": 2.2228462072586246,
      "grad_norm": 0.36395078897476196,
      "learning_rate": 9.045831692330546e-05,
      "loss": 1.4533,
      "step": 59360
    },
    {
      "epoch": 2.223388148424137,
      "grad_norm": 0.4966432452201843,
      "learning_rate": 9.044856278875212e-05,
      "loss": 1.4494,
      "step": 59370
    },
    {
      "epoch": 2.2239300895896488,
      "grad_norm": 0.4597684442996979,
      "learning_rate": 9.043880426298475e-05,
      "loss": 1.4468,
      "step": 59380
    },
    {
      "epoch": 2.224472030755161,
      "grad_norm": 0.2446877360343933,
      "learning_rate": 9.042904134721214e-05,
      "loss": 1.4415,
      "step": 59390
    },
    {
      "epoch": 2.2250139719206734,
      "grad_norm": 0.2874411940574646,
      "learning_rate": 9.041927404264371e-05,
      "loss": 1.4472,
      "step": 59400
    },
    {
      "epoch": 2.2255559130861857,
      "grad_norm": 0.5566506385803223,
      "learning_rate": 9.040950235048938e-05,
      "loss": 1.446,
      "step": 59410
    },
    {
      "epoch": 2.2260978542516976,
      "grad_norm": 0.19569170475006104,
      "learning_rate": 9.039972627195965e-05,
      "loss": 1.4365,
      "step": 59420
    },
    {
      "epoch": 2.226152048368249,
      "eval_loss": 2.458207607269287,
      "eval_runtime": 21.9788,
      "eval_samples_per_second": 227.492,
      "eval_steps_per_second": 1.228,
      "step": 59421
    },
    {
      "epoch": 2.22663979541721,
      "grad_norm": 0.41606298089027405,
      "learning_rate": 9.038994580826549e-05,
      "loss": 1.4496,
      "step": 59430
    },
    {
      "epoch": 2.2271817365827222,
      "grad_norm": 0.21181747317314148,
      "learning_rate": 9.03801609606185e-05,
      "loss": 1.4425,
      "step": 59440
    },
    {
      "epoch": 2.2277236777482345,
      "grad_norm": 0.3672482371330261,
      "learning_rate": 9.03703717302308e-05,
      "loss": 1.4444,
      "step": 59450
    },
    {
      "epoch": 2.228265618913747,
      "grad_norm": 0.2655334174633026,
      "learning_rate": 9.0360578118315e-05,
      "loss": 1.4421,
      "step": 59460
    },
    {
      "epoch": 2.2288075600792587,
      "grad_norm": 0.20638707280158997,
      "learning_rate": 9.035078012608431e-05,
      "loss": 1.4422,
      "step": 59470
    },
    {
      "epoch": 2.229349501244771,
      "grad_norm": 0.5288010239601135,
      "learning_rate": 9.034097775475244e-05,
      "loss": 1.4441,
      "step": 59480
    },
    {
      "epoch": 2.2298914424102834,
      "grad_norm": 0.37558361887931824,
      "learning_rate": 9.03311710055337e-05,
      "loss": 1.4386,
      "step": 59490
    },
    {
      "epoch": 2.2304333835757957,
      "grad_norm": 0.30273330211639404,
      "learning_rate": 9.032135987964287e-05,
      "loss": 1.4426,
      "step": 59500
    },
    {
      "epoch": 2.2308669365082054,
      "eval_loss": 2.4551827907562256,
      "eval_runtime": 21.9794,
      "eval_samples_per_second": 227.485,
      "eval_steps_per_second": 1.228,
      "step": 59508
    },
    {
      "epoch": 2.2309753247413076,
      "grad_norm": 0.34168776869773865,
      "learning_rate": 9.031154437829533e-05,
      "loss": 1.4485,
      "step": 59510
    },
    {
      "epoch": 2.23151726590682,
      "grad_norm": 0.2830340266227722,
      "learning_rate": 9.030172450270699e-05,
      "loss": 1.446,
      "step": 59520
    },
    {
      "epoch": 2.232059207072332,
      "grad_norm": 0.18490228056907654,
      "learning_rate": 9.029190025409426e-05,
      "loss": 1.4459,
      "step": 59530
    },
    {
      "epoch": 2.2326011482378445,
      "grad_norm": 0.32973939180374146,
      "learning_rate": 9.028207163367417e-05,
      "loss": 1.4491,
      "step": 59540
    },
    {
      "epoch": 2.233143089403357,
      "grad_norm": 0.2619604766368866,
      "learning_rate": 9.027223864266423e-05,
      "loss": 1.4446,
      "step": 59550
    },
    {
      "epoch": 2.2336850305688687,
      "grad_norm": 0.3706190884113312,
      "learning_rate": 9.02624012822825e-05,
      "loss": 1.4369,
      "step": 59560
    },
    {
      "epoch": 2.234226971734381,
      "grad_norm": 0.1908445507287979,
      "learning_rate": 9.025255955374758e-05,
      "loss": 1.4321,
      "step": 59570
    },
    {
      "epoch": 2.2347689128998933,
      "grad_norm": 0.25038012862205505,
      "learning_rate": 9.024271345827864e-05,
      "loss": 1.4451,
      "step": 59580
    },
    {
      "epoch": 2.2353108540654056,
      "grad_norm": 0.22101671993732452,
      "learning_rate": 9.023286299709538e-05,
      "loss": 1.4376,
      "step": 59590
    },
    {
      "epoch": 2.2355818246481616,
      "eval_loss": 2.449619770050049,
      "eval_runtime": 21.9821,
      "eval_samples_per_second": 227.458,
      "eval_steps_per_second": 1.228,
      "step": 59595
    },
    {
      "epoch": 2.235852795230918,
      "grad_norm": 0.2644299268722534,
      "learning_rate": 9.022300817141799e-05,
      "loss": 1.4417,
      "step": 59600
    },
    {
      "epoch": 2.23639473639643,
      "grad_norm": 0.18024158477783203,
      "learning_rate": 9.02131489824673e-05,
      "loss": 1.4442,
      "step": 59610
    },
    {
      "epoch": 2.236936677561942,
      "grad_norm": 0.2031889110803604,
      "learning_rate": 9.020328543146457e-05,
      "loss": 1.4512,
      "step": 59620
    },
    {
      "epoch": 2.2374786187274545,
      "grad_norm": 0.2275354415178299,
      "learning_rate": 9.01934175196317e-05,
      "loss": 1.4478,
      "step": 59630
    },
    {
      "epoch": 2.2380205598929668,
      "grad_norm": 0.1913514882326126,
      "learning_rate": 9.018354524819104e-05,
      "loss": 1.4532,
      "step": 59640
    },
    {
      "epoch": 2.2385625010584786,
      "grad_norm": 0.17917923629283905,
      "learning_rate": 9.017366861836555e-05,
      "loss": 1.4534,
      "step": 59650
    },
    {
      "epoch": 2.239104442223991,
      "grad_norm": 0.1910235583782196,
      "learning_rate": 9.016378763137872e-05,
      "loss": 1.4452,
      "step": 59660
    },
    {
      "epoch": 2.2396463833895033,
      "grad_norm": 0.2699836194515228,
      "learning_rate": 9.015390228845454e-05,
      "loss": 1.4452,
      "step": 59670
    },
    {
      "epoch": 2.2401883245550156,
      "grad_norm": 0.5121615529060364,
      "learning_rate": 9.014401259081754e-05,
      "loss": 1.4472,
      "step": 59680
    },
    {
      "epoch": 2.240296712788118,
      "eval_loss": 2.445277452468872,
      "eval_runtime": 21.9791,
      "eval_samples_per_second": 227.489,
      "eval_steps_per_second": 1.228,
      "step": 59682
    },
    {
      "epoch": 2.240730265720528,
      "grad_norm": 0.3202587962150574,
      "learning_rate": 9.013411853969286e-05,
      "loss": 1.447,
      "step": 59690
    },
    {
      "epoch": 2.24127220688604,
      "grad_norm": 0.29666054248809814,
      "learning_rate": 9.012422013630611e-05,
      "loss": 1.4478,
      "step": 59700
    },
    {
      "epoch": 2.241814148051552,
      "grad_norm": 0.2042292356491089,
      "learning_rate": 9.011431738188348e-05,
      "loss": 1.4473,
      "step": 59710
    },
    {
      "epoch": 2.2423560892170644,
      "grad_norm": 0.1927926242351532,
      "learning_rate": 9.010441027765165e-05,
      "loss": 1.4467,
      "step": 59720
    },
    {
      "epoch": 2.2428980303825767,
      "grad_norm": 0.3930176794528961,
      "learning_rate": 9.009449882483788e-05,
      "loss": 1.4493,
      "step": 59730
    },
    {
      "epoch": 2.2434399715480886,
      "grad_norm": 0.31526270508766174,
      "learning_rate": 9.008458302466995e-05,
      "loss": 1.4487,
      "step": 59740
    },
    {
      "epoch": 2.243981912713601,
      "grad_norm": 0.2965898811817169,
      "learning_rate": 9.007466287837622e-05,
      "loss": 1.4528,
      "step": 59750
    },
    {
      "epoch": 2.2445238538791132,
      "grad_norm": 0.3894706964492798,
      "learning_rate": 9.006473838718551e-05,
      "loss": 1.443,
      "step": 59760
    },
    {
      "epoch": 2.2450116009280743,
      "eval_loss": 2.4487032890319824,
      "eval_runtime": 21.9873,
      "eval_samples_per_second": 227.404,
      "eval_steps_per_second": 1.228,
      "step": 59769
    },
    {
      "epoch": 2.2450657950446256,
      "grad_norm": 0.2878785729408264,
      "learning_rate": 9.005480955232726e-05,
      "loss": 1.4379,
      "step": 59770
    },
    {
      "epoch": 2.245607736210138,
      "grad_norm": 0.2061561793088913,
      "learning_rate": 9.004487637503139e-05,
      "loss": 1.4331,
      "step": 59780
    },
    {
      "epoch": 2.2461496773756497,
      "grad_norm": 0.24609437584877014,
      "learning_rate": 9.00349388565284e-05,
      "loss": 1.4486,
      "step": 59790
    },
    {
      "epoch": 2.246691618541162,
      "grad_norm": 0.24656355381011963,
      "learning_rate": 9.002499699804927e-05,
      "loss": 1.442,
      "step": 59800
    },
    {
      "epoch": 2.2472335597066744,
      "grad_norm": 0.23344773054122925,
      "learning_rate": 9.001505080082558e-05,
      "loss": 1.452,
      "step": 59810
    },
    {
      "epoch": 2.2477755008721867,
      "grad_norm": 0.2287297546863556,
      "learning_rate": 9.000510026608942e-05,
      "loss": 1.4439,
      "step": 59820
    },
    {
      "epoch": 2.248317442037699,
      "grad_norm": 0.2048652619123459,
      "learning_rate": 8.999514539507342e-05,
      "loss": 1.4392,
      "step": 59830
    },
    {
      "epoch": 2.248859383203211,
      "grad_norm": 0.2534552812576294,
      "learning_rate": 8.998518618901073e-05,
      "loss": 1.4449,
      "step": 59840
    },
    {
      "epoch": 2.249401324368723,
      "grad_norm": 0.22183406352996826,
      "learning_rate": 8.997522264913508e-05,
      "loss": 1.4556,
      "step": 59850
    },
    {
      "epoch": 2.2497264890680304,
      "eval_loss": 2.4434633255004883,
      "eval_runtime": 21.9826,
      "eval_samples_per_second": 227.452,
      "eval_steps_per_second": 1.228,
      "step": 59856
    },
    {
      "epoch": 2.2499432655342355,
      "grad_norm": 0.26273661851882935,
      "learning_rate": 8.996525477668068e-05,
      "loss": 1.4403,
      "step": 59860
    },
    {
      "epoch": 2.250485206699748,
      "grad_norm": 0.2545994222164154,
      "learning_rate": 8.995528257288233e-05,
      "loss": 1.4377,
      "step": 59870
    },
    {
      "epoch": 2.2510271478652597,
      "grad_norm": 0.2006346583366394,
      "learning_rate": 8.994530603897534e-05,
      "loss": 1.4457,
      "step": 59880
    },
    {
      "epoch": 2.251569089030772,
      "grad_norm": 0.17895658314228058,
      "learning_rate": 8.993532517619554e-05,
      "loss": 1.4424,
      "step": 59890
    },
    {
      "epoch": 2.2521110301962843,
      "grad_norm": 0.2239982634782791,
      "learning_rate": 8.992533998577936e-05,
      "loss": 1.4329,
      "step": 59900
    },
    {
      "epoch": 2.2526529713617967,
      "grad_norm": 0.2703670263290405,
      "learning_rate": 8.991535046896367e-05,
      "loss": 1.456,
      "step": 59910
    },
    {
      "epoch": 2.2531949125273085,
      "grad_norm": 0.23329605162143707,
      "learning_rate": 8.990535662698596e-05,
      "loss": 1.4411,
      "step": 59920
    },
    {
      "epoch": 2.253736853692821,
      "grad_norm": 0.25912684202194214,
      "learning_rate": 8.989535846108421e-05,
      "loss": 1.446,
      "step": 59930
    },
    {
      "epoch": 2.254278794858333,
      "grad_norm": 0.22521668672561646,
      "learning_rate": 8.988535597249696e-05,
      "loss": 1.4516,
      "step": 59940
    },
    {
      "epoch": 2.254441377207987,
      "eval_loss": 2.432748556137085,
      "eval_runtime": 21.9825,
      "eval_samples_per_second": 227.454,
      "eval_steps_per_second": 1.228,
      "step": 59943
    },
    {
      "epoch": 2.2548207360238455,
      "grad_norm": 0.20753563940525055,
      "learning_rate": 8.987534916246327e-05,
      "loss": 1.4364,
      "step": 59950
    },
    {
      "epoch": 2.255362677189358,
      "grad_norm": 0.18667367100715637,
      "learning_rate": 8.986533803222272e-05,
      "loss": 1.4418,
      "step": 59960
    },
    {
      "epoch": 2.2559046183548697,
      "grad_norm": 0.2076788991689682,
      "learning_rate": 8.985532258301549e-05,
      "loss": 1.4435,
      "step": 59970
    },
    {
      "epoch": 2.256446559520382,
      "grad_norm": 0.2706129550933838,
      "learning_rate": 8.98453028160822e-05,
      "loss": 1.4415,
      "step": 59980
    },
    {
      "epoch": 2.2569885006858943,
      "grad_norm": 0.21341146528720856,
      "learning_rate": 8.98352787326641e-05,
      "loss": 1.4393,
      "step": 59990
    },
    {
      "epoch": 2.2575304418514066,
      "grad_norm": 0.21545018255710602,
      "learning_rate": 8.982525033400289e-05,
      "loss": 1.4484,
      "step": 60000
    },
    {
      "epoch": 2.258072383016919,
      "grad_norm": 0.23385457694530487,
      "learning_rate": 8.98152176213409e-05,
      "loss": 1.4451,
      "step": 60010
    },
    {
      "epoch": 2.258614324182431,
      "grad_norm": 0.20415639877319336,
      "learning_rate": 8.980518059592088e-05,
      "loss": 1.4394,
      "step": 60020
    },
    {
      "epoch": 2.259156265347943,
      "grad_norm": 0.25711020827293396,
      "learning_rate": 8.97951392589862e-05,
      "loss": 1.4322,
      "step": 60030
    },
    {
      "epoch": 2.259156265347943,
      "eval_loss": 2.4509191513061523,
      "eval_runtime": 21.9779,
      "eval_samples_per_second": 227.502,
      "eval_steps_per_second": 1.229,
      "step": 60030
    },
    {
      "epoch": 2.2596982065134554,
      "grad_norm": 0.3049508333206177,
      "learning_rate": 8.978509361178073e-05,
      "loss": 1.4373,
      "step": 60040
    },
    {
      "epoch": 2.2602401476789677,
      "grad_norm": 0.2312183976173401,
      "learning_rate": 8.97750436555489e-05,
      "loss": 1.4361,
      "step": 60050
    },
    {
      "epoch": 2.26078208884448,
      "grad_norm": 0.2977278232574463,
      "learning_rate": 8.976498939153562e-05,
      "loss": 1.4418,
      "step": 60060
    },
    {
      "epoch": 2.261324030009992,
      "grad_norm": 0.18335352838039398,
      "learning_rate": 8.975493082098639e-05,
      "loss": 1.4427,
      "step": 60070
    },
    {
      "epoch": 2.2618659711755043,
      "grad_norm": 0.22939863801002502,
      "learning_rate": 8.974486794514723e-05,
      "loss": 1.4411,
      "step": 60080
    },
    {
      "epoch": 2.2624079123410166,
      "grad_norm": 0.17828980088233948,
      "learning_rate": 8.973480076526469e-05,
      "loss": 1.4399,
      "step": 60090
    },
    {
      "epoch": 2.262949853506529,
      "grad_norm": 0.2951532304286957,
      "learning_rate": 8.97247292825858e-05,
      "loss": 1.4469,
      "step": 60100
    },
    {
      "epoch": 2.2634917946720408,
      "grad_norm": 0.2627731263637543,
      "learning_rate": 8.971465349835824e-05,
      "loss": 1.4371,
      "step": 60110
    },
    {
      "epoch": 2.2638711534878997,
      "eval_loss": 2.4536399841308594,
      "eval_runtime": 21.9812,
      "eval_samples_per_second": 227.467,
      "eval_steps_per_second": 1.228,
      "step": 60117
    },
    {
      "epoch": 2.264033735837553,
      "grad_norm": 0.1896529346704483,
      "learning_rate": 8.970457341383011e-05,
      "loss": 1.4413,
      "step": 60120
    },
    {
      "epoch": 2.2645756770030654,
      "grad_norm": 0.1765110045671463,
      "learning_rate": 8.969448903025008e-05,
      "loss": 1.445,
      "step": 60130
    },
    {
      "epoch": 2.2651176181685777,
      "grad_norm": 0.19842223823070526,
      "learning_rate": 8.96844003488674e-05,
      "loss": 1.4459,
      "step": 60140
    },
    {
      "epoch": 2.2656595593340896,
      "grad_norm": 0.23995903134346008,
      "learning_rate": 8.967430737093179e-05,
      "loss": 1.4476,
      "step": 60150
    },
    {
      "epoch": 2.266201500499602,
      "grad_norm": 0.3694940209388733,
      "learning_rate": 8.966421009769352e-05,
      "loss": 1.4348,
      "step": 60160
    },
    {
      "epoch": 2.266743441665114,
      "grad_norm": 0.27862706780433655,
      "learning_rate": 8.965410853040338e-05,
      "loss": 1.4425,
      "step": 60170
    },
    {
      "epoch": 2.2672853828306265,
      "grad_norm": 0.2391398549079895,
      "learning_rate": 8.964400267031274e-05,
      "loss": 1.4437,
      "step": 60180
    },
    {
      "epoch": 2.267827323996139,
      "grad_norm": 0.17560423910617828,
      "learning_rate": 8.963389251867346e-05,
      "loss": 1.4396,
      "step": 60190
    },
    {
      "epoch": 2.2683692651616507,
      "grad_norm": 0.2710578143596649,
      "learning_rate": 8.962377807673795e-05,
      "loss": 1.4407,
      "step": 60200
    },
    {
      "epoch": 2.268586041627856,
      "eval_loss": 2.433633327484131,
      "eval_runtime": 21.9783,
      "eval_samples_per_second": 227.497,
      "eval_steps_per_second": 1.228,
      "step": 60204
    },
    {
      "epoch": 2.268911206327163,
      "grad_norm": 0.2406301647424698,
      "learning_rate": 8.961365934575913e-05,
      "loss": 1.4453,
      "step": 60210
    },
    {
      "epoch": 2.2694531474926753,
      "grad_norm": 0.48485299944877625,
      "learning_rate": 8.960353632699046e-05,
      "loss": 1.4345,
      "step": 60220
    },
    {
      "epoch": 2.2699950886581877,
      "grad_norm": 0.270643025636673,
      "learning_rate": 8.959340902168594e-05,
      "loss": 1.4527,
      "step": 60230
    },
    {
      "epoch": 2.2705370298237,
      "grad_norm": 0.22174952924251556,
      "learning_rate": 8.95832774311001e-05,
      "loss": 1.4415,
      "step": 60240
    },
    {
      "epoch": 2.271078970989212,
      "grad_norm": 0.4398956298828125,
      "learning_rate": 8.957314155648801e-05,
      "loss": 1.4397,
      "step": 60250
    },
    {
      "epoch": 2.271620912154724,
      "grad_norm": 0.2305718958377838,
      "learning_rate": 8.956300139910525e-05,
      "loss": 1.4484,
      "step": 60260
    },
    {
      "epoch": 2.2721628533202365,
      "grad_norm": 0.2533299922943115,
      "learning_rate": 8.955285696020793e-05,
      "loss": 1.4524,
      "step": 60270
    },
    {
      "epoch": 2.272704794485749,
      "grad_norm": 0.20475125312805176,
      "learning_rate": 8.954270824105268e-05,
      "loss": 1.4427,
      "step": 60280
    },
    {
      "epoch": 2.2732467356512607,
      "grad_norm": 0.20677226781845093,
      "learning_rate": 8.953255524289671e-05,
      "loss": 1.4397,
      "step": 60290
    },
    {
      "epoch": 2.273300929767812,
      "eval_loss": 2.436509132385254,
      "eval_runtime": 22.2954,
      "eval_samples_per_second": 224.261,
      "eval_steps_per_second": 1.211,
      "step": 60291
    },
    {
      "epoch": 2.273788676816773,
      "grad_norm": 0.21641433238983154,
      "learning_rate": 8.952239796699774e-05,
      "loss": 1.4478,
      "step": 60300
    },
    {
      "epoch": 2.2743306179822853,
      "grad_norm": 0.18868723511695862,
      "learning_rate": 8.951223641461398e-05,
      "loss": 1.4441,
      "step": 60310
    },
    {
      "epoch": 2.2748725591477976,
      "grad_norm": 0.22199149429798126,
      "learning_rate": 8.950207058700422e-05,
      "loss": 1.4376,
      "step": 60320
    },
    {
      "epoch": 2.2754145003133095,
      "grad_norm": 0.21569089591503143,
      "learning_rate": 8.949190048542773e-05,
      "loss": 1.4365,
      "step": 60330
    },
    {
      "epoch": 2.275956441478822,
      "grad_norm": 0.2327711433172226,
      "learning_rate": 8.948172611114438e-05,
      "loss": 1.4375,
      "step": 60340
    },
    {
      "epoch": 2.276498382644334,
      "grad_norm": 0.25674811005592346,
      "learning_rate": 8.947154746541448e-05,
      "loss": 1.442,
      "step": 60350
    },
    {
      "epoch": 2.2770403238098464,
      "grad_norm": 0.26655909419059753,
      "learning_rate": 8.946136454949895e-05,
      "loss": 1.4425,
      "step": 60360
    },
    {
      "epoch": 2.2775822649753588,
      "grad_norm": 0.2611353099346161,
      "learning_rate": 8.94511773646592e-05,
      "loss": 1.4407,
      "step": 60370
    },
    {
      "epoch": 2.2780158179077685,
      "eval_loss": 2.4322497844696045,
      "eval_runtime": 21.9808,
      "eval_samples_per_second": 227.471,
      "eval_steps_per_second": 1.228,
      "step": 60378
    },
    {
      "epoch": 2.2781242061408706,
      "grad_norm": 0.22814051806926727,
      "learning_rate": 8.944098591215717e-05,
      "loss": 1.4372,
      "step": 60380
    },
    {
      "epoch": 2.278666147306383,
      "grad_norm": 0.3651120066642761,
      "learning_rate": 8.943079019325533e-05,
      "loss": 1.4415,
      "step": 60390
    },
    {
      "epoch": 2.2792080884718953,
      "grad_norm": 0.3389890193939209,
      "learning_rate": 8.94205902092167e-05,
      "loss": 1.4422,
      "step": 60400
    },
    {
      "epoch": 2.2797500296374076,
      "grad_norm": 0.32138633728027344,
      "learning_rate": 8.941038596130477e-05,
      "loss": 1.428,
      "step": 60410
    },
    {
      "epoch": 2.28029197080292,
      "grad_norm": 0.2286197692155838,
      "learning_rate": 8.940017745078363e-05,
      "loss": 1.4454,
      "step": 60420
    },
    {
      "epoch": 2.2808339119684318,
      "grad_norm": 0.26349860429763794,
      "learning_rate": 8.938996467891786e-05,
      "loss": 1.4401,
      "step": 60430
    },
    {
      "epoch": 2.281375853133944,
      "grad_norm": 0.1832720935344696,
      "learning_rate": 8.937974764697258e-05,
      "loss": 1.451,
      "step": 60440
    },
    {
      "epoch": 2.2819177942994564,
      "grad_norm": 0.18660111725330353,
      "learning_rate": 8.936952635621341e-05,
      "loss": 1.4453,
      "step": 60450
    },
    {
      "epoch": 2.2824597354649687,
      "grad_norm": 0.22778889536857605,
      "learning_rate": 8.935930080790655e-05,
      "loss": 1.4451,
      "step": 60460
    },
    {
      "epoch": 2.2827307060477247,
      "eval_loss": 2.4412660598754883,
      "eval_runtime": 21.9845,
      "eval_samples_per_second": 227.433,
      "eval_steps_per_second": 1.228,
      "step": 60465
    },
    {
      "epoch": 2.283001676630481,
      "grad_norm": 0.4049639105796814,
      "learning_rate": 8.934907100331865e-05,
      "loss": 1.4458,
      "step": 60470
    },
    {
      "epoch": 2.283543617795993,
      "grad_norm": 0.20592905580997467,
      "learning_rate": 8.933883694371699e-05,
      "loss": 1.441,
      "step": 60480
    },
    {
      "epoch": 2.2840855589615052,
      "grad_norm": 0.5759056806564331,
      "learning_rate": 8.932859863036927e-05,
      "loss": 1.4342,
      "step": 60490
    },
    {
      "epoch": 2.2846275001270175,
      "grad_norm": 0.3856758177280426,
      "learning_rate": 8.931835606454382e-05,
      "loss": 1.4307,
      "step": 60500
    },
    {
      "epoch": 2.28516944129253,
      "grad_norm": 0.24121706187725067,
      "learning_rate": 8.930810924750939e-05,
      "loss": 1.4466,
      "step": 60510
    },
    {
      "epoch": 2.2857113824580417,
      "grad_norm": 0.3475324511528015,
      "learning_rate": 8.929785818053534e-05,
      "loss": 1.4371,
      "step": 60520
    },
    {
      "epoch": 2.286253323623554,
      "grad_norm": 0.2951948642730713,
      "learning_rate": 8.928760286489155e-05,
      "loss": 1.4513,
      "step": 60530
    },
    {
      "epoch": 2.2867952647890664,
      "grad_norm": 0.27565380930900574,
      "learning_rate": 8.927734330184835e-05,
      "loss": 1.4397,
      "step": 60540
    },
    {
      "epoch": 2.2873372059545787,
      "grad_norm": 0.22539940476417542,
      "learning_rate": 8.92670794926767e-05,
      "loss": 1.4414,
      "step": 60550
    },
    {
      "epoch": 2.2874455941876812,
      "eval_loss": 2.4488258361816406,
      "eval_runtime": 21.9792,
      "eval_samples_per_second": 227.487,
      "eval_steps_per_second": 1.228,
      "step": 60552
    },
    {
      "epoch": 2.2878791471200906,
      "grad_norm": 0.2256643921136856,
      "learning_rate": 8.9256811438648e-05,
      "loss": 1.4383,
      "step": 60560
    },
    {
      "epoch": 2.288421088285603,
      "grad_norm": 0.23651210963726044,
      "learning_rate": 8.924653914103423e-05,
      "loss": 1.44,
      "step": 60570
    },
    {
      "epoch": 2.288963029451115,
      "grad_norm": 0.2763255834579468,
      "learning_rate": 8.923626260110789e-05,
      "loss": 1.4377,
      "step": 60580
    },
    {
      "epoch": 2.2895049706166275,
      "grad_norm": 0.34174492955207825,
      "learning_rate": 8.922598182014197e-05,
      "loss": 1.4544,
      "step": 60590
    },
    {
      "epoch": 2.29004691178214,
      "grad_norm": 0.26863643527030945,
      "learning_rate": 8.921569679941e-05,
      "loss": 1.4435,
      "step": 60600
    },
    {
      "epoch": 2.2905888529476517,
      "grad_norm": 0.24815820157527924,
      "learning_rate": 8.920540754018608e-05,
      "loss": 1.4385,
      "step": 60610
    },
    {
      "epoch": 2.291130794113164,
      "grad_norm": 0.3145918548107147,
      "learning_rate": 8.919511404374476e-05,
      "loss": 1.446,
      "step": 60620
    },
    {
      "epoch": 2.2916727352786763,
      "grad_norm": 0.2556162178516388,
      "learning_rate": 8.918481631136119e-05,
      "loss": 1.4419,
      "step": 60630
    },
    {
      "epoch": 2.2921604823276374,
      "eval_loss": 2.4316673278808594,
      "eval_runtime": 21.9778,
      "eval_samples_per_second": 227.503,
      "eval_steps_per_second": 1.229,
      "step": 60639
    },
    {
      "epoch": 2.2922146764441886,
      "grad_norm": 0.32888561487197876,
      "learning_rate": 8.9174514344311e-05,
      "loss": 1.4446,
      "step": 60640
    },
    {
      "epoch": 2.292756617609701,
      "grad_norm": 0.289276123046875,
      "learning_rate": 8.916420814387033e-05,
      "loss": 1.4418,
      "step": 60650
    },
    {
      "epoch": 2.293298558775213,
      "grad_norm": 0.4594111740589142,
      "learning_rate": 8.915389771131587e-05,
      "loss": 1.4414,
      "step": 60660
    },
    {
      "epoch": 2.293840499940725,
      "grad_norm": 0.2204999476671219,
      "learning_rate": 8.914358304792486e-05,
      "loss": 1.4378,
      "step": 60670
    },
    {
      "epoch": 2.2943824411062375,
      "grad_norm": 0.22972345352172852,
      "learning_rate": 8.913326415497501e-05,
      "loss": 1.4386,
      "step": 60680
    },
    {
      "epoch": 2.2949243822717498,
      "grad_norm": 0.2072676122188568,
      "learning_rate": 8.912294103374459e-05,
      "loss": 1.4439,
      "step": 60690
    },
    {
      "epoch": 2.295466323437262,
      "grad_norm": 0.18969523906707764,
      "learning_rate": 8.911261368551238e-05,
      "loss": 1.4416,
      "step": 60700
    },
    {
      "epoch": 2.296008264602774,
      "grad_norm": 0.24862347543239594,
      "learning_rate": 8.910228211155771e-05,
      "loss": 1.4567,
      "step": 60710
    },
    {
      "epoch": 2.2965502057682863,
      "grad_norm": 0.3798186779022217,
      "learning_rate": 8.909194631316035e-05,
      "loss": 1.4435,
      "step": 60720
    },
    {
      "epoch": 2.2968753704675935,
      "eval_loss": 2.450904369354248,
      "eval_runtime": 21.9806,
      "eval_samples_per_second": 227.474,
      "eval_steps_per_second": 1.228,
      "step": 60726
    },
    {
      "epoch": 2.2970921469337986,
      "grad_norm": 0.32553133368492126,
      "learning_rate": 8.908160629160073e-05,
      "loss": 1.4286,
      "step": 60730
    },
    {
      "epoch": 2.297634088099311,
      "grad_norm": 0.3055340647697449,
      "learning_rate": 8.907126204815966e-05,
      "loss": 1.4405,
      "step": 60740
    },
    {
      "epoch": 2.298176029264823,
      "grad_norm": 0.28653863072395325,
      "learning_rate": 8.906091358411858e-05,
      "loss": 1.4405,
      "step": 60750
    },
    {
      "epoch": 2.298717970430335,
      "grad_norm": 0.2887609004974365,
      "learning_rate": 8.905056090075942e-05,
      "loss": 1.4377,
      "step": 60760
    },
    {
      "epoch": 2.2992599115958474,
      "grad_norm": 0.23055769503116608,
      "learning_rate": 8.90402039993646e-05,
      "loss": 1.4483,
      "step": 60770
    },
    {
      "epoch": 2.2998018527613597,
      "grad_norm": 0.20625241100788116,
      "learning_rate": 8.90298428812171e-05,
      "loss": 1.4344,
      "step": 60780
    },
    {
      "epoch": 2.3003437939268716,
      "grad_norm": 0.1865055412054062,
      "learning_rate": 8.901947754760039e-05,
      "loss": 1.4477,
      "step": 60790
    },
    {
      "epoch": 2.300885735092384,
      "grad_norm": 0.36700624227523804,
      "learning_rate": 8.90091079997985e-05,
      "loss": 1.4487,
      "step": 60800
    },
    {
      "epoch": 2.3014276762578962,
      "grad_norm": 0.2842310667037964,
      "learning_rate": 8.899873423909599e-05,
      "loss": 1.4478,
      "step": 60810
    },
    {
      "epoch": 2.30159025860755,
      "eval_loss": 2.4509172439575195,
      "eval_runtime": 35.9865,
      "eval_samples_per_second": 138.941,
      "eval_steps_per_second": 0.75,
      "step": 60813
    },
    {
      "epoch": 2.3019696174234086,
      "grad_norm": 0.24928165972232819,
      "learning_rate": 8.898835626677786e-05,
      "loss": 1.4322,
      "step": 60820
    },
    {
      "epoch": 2.302511558588921,
      "grad_norm": 0.22808949649333954,
      "learning_rate": 8.897797408412973e-05,
      "loss": 1.4377,
      "step": 60830
    },
    {
      "epoch": 2.3030534997544327,
      "grad_norm": 0.22761550545692444,
      "learning_rate": 8.896758769243769e-05,
      "loss": 1.4548,
      "step": 60840
    },
    {
      "epoch": 2.303595440919945,
      "grad_norm": 0.24886593222618103,
      "learning_rate": 8.895719709298837e-05,
      "loss": 1.4398,
      "step": 60850
    },
    {
      "epoch": 2.3041373820854574,
      "grad_norm": 0.1771778017282486,
      "learning_rate": 8.894680228706888e-05,
      "loss": 1.4439,
      "step": 60860
    },
    {
      "epoch": 2.3046793232509697,
      "grad_norm": 0.28259819746017456,
      "learning_rate": 8.893640327596692e-05,
      "loss": 1.4408,
      "step": 60870
    },
    {
      "epoch": 2.305221264416482,
      "grad_norm": 0.23001182079315186,
      "learning_rate": 8.892600006097065e-05,
      "loss": 1.4418,
      "step": 60880
    },
    {
      "epoch": 2.305763205581994,
      "grad_norm": 0.21615125238895416,
      "learning_rate": 8.891559264336879e-05,
      "loss": 1.4419,
      "step": 60890
    },
    {
      "epoch": 2.306305146747506,
      "grad_norm": 0.2845569849014282,
      "learning_rate": 8.890518102445055e-05,
      "loss": 1.4417,
      "step": 60900
    },
    {
      "epoch": 2.306305146747506,
      "eval_loss": 2.452852487564087,
      "eval_runtime": 21.8629,
      "eval_samples_per_second": 228.698,
      "eval_steps_per_second": 1.235,
      "step": 60900
    },
    {
      "epoch": 2.3068470879130185,
      "grad_norm": 0.21112145483493805,
      "learning_rate": 8.88947652055057e-05,
      "loss": 1.4355,
      "step": 60910
    },
    {
      "epoch": 2.307389029078531,
      "grad_norm": 0.2809900641441345,
      "learning_rate": 8.888434518782449e-05,
      "loss": 1.441,
      "step": 60920
    },
    {
      "epoch": 2.3079309702440427,
      "grad_norm": 0.19691738486289978,
      "learning_rate": 8.88739209726977e-05,
      "loss": 1.4439,
      "step": 60930
    },
    {
      "epoch": 2.308472911409555,
      "grad_norm": 0.5515458583831787,
      "learning_rate": 8.886349256141665e-05,
      "loss": 1.4473,
      "step": 60940
    },
    {
      "epoch": 2.3090148525750673,
      "grad_norm": 0.34075281023979187,
      "learning_rate": 8.885305995527317e-05,
      "loss": 1.4532,
      "step": 60950
    },
    {
      "epoch": 2.3095567937405796,
      "grad_norm": 0.2703841030597687,
      "learning_rate": 8.884262315555958e-05,
      "loss": 1.4428,
      "step": 60960
    },
    {
      "epoch": 2.3100987349060915,
      "grad_norm": 0.24917836487293243,
      "learning_rate": 8.88321821635688e-05,
      "loss": 1.4448,
      "step": 60970
    },
    {
      "epoch": 2.310640676071604,
      "grad_norm": 0.22919951379299164,
      "learning_rate": 8.882173698059414e-05,
      "loss": 1.4484,
      "step": 60980
    },
    {
      "epoch": 2.3110200348874628,
      "eval_loss": 2.4533286094665527,
      "eval_runtime": 21.9805,
      "eval_samples_per_second": 227.474,
      "eval_steps_per_second": 1.228,
      "step": 60987
    },
    {
      "epoch": 2.311182617237116,
      "grad_norm": 0.2503993511199951,
      "learning_rate": 8.881128760792957e-05,
      "loss": 1.4377,
      "step": 60990
    },
    {
      "epoch": 2.3117245584026285,
      "grad_norm": 0.2338937520980835,
      "learning_rate": 8.880083404686947e-05,
      "loss": 1.4339,
      "step": 61000
    },
    {
      "epoch": 2.312266499568141,
      "grad_norm": 0.2059098482131958,
      "learning_rate": 8.879037629870878e-05,
      "loss": 1.4293,
      "step": 61010
    },
    {
      "epoch": 2.3128084407336527,
      "grad_norm": 0.20788943767547607,
      "learning_rate": 8.8779914364743e-05,
      "loss": 1.429,
      "step": 61020
    },
    {
      "epoch": 2.313350381899165,
      "grad_norm": 0.2890428900718689,
      "learning_rate": 8.87694482462681e-05,
      "loss": 1.4375,
      "step": 61030
    },
    {
      "epoch": 2.3138923230646773,
      "grad_norm": 0.235799640417099,
      "learning_rate": 8.875897794458053e-05,
      "loss": 1.4417,
      "step": 61040
    },
    {
      "epoch": 2.3144342642301896,
      "grad_norm": 0.39794206619262695,
      "learning_rate": 8.874850346097736e-05,
      "loss": 1.4408,
      "step": 61050
    },
    {
      "epoch": 2.314976205395702,
      "grad_norm": 0.46668577194213867,
      "learning_rate": 8.87380247967561e-05,
      "loss": 1.4364,
      "step": 61060
    },
    {
      "epoch": 2.315518146561214,
      "grad_norm": 0.21078045666217804,
      "learning_rate": 8.872754195321479e-05,
      "loss": 1.4361,
      "step": 61070
    },
    {
      "epoch": 2.315734923027419,
      "eval_loss": 2.4492809772491455,
      "eval_runtime": 21.9892,
      "eval_samples_per_second": 227.384,
      "eval_steps_per_second": 1.228,
      "step": 61074
    },
    {
      "epoch": 2.316060087726726,
      "grad_norm": 0.27754735946655273,
      "learning_rate": 8.871705493165203e-05,
      "loss": 1.4314,
      "step": 61080
    },
    {
      "epoch": 2.3166020288922384,
      "grad_norm": 0.22259338200092316,
      "learning_rate": 8.870656373336689e-05,
      "loss": 1.4417,
      "step": 61090
    },
    {
      "epoch": 2.3171439700577507,
      "grad_norm": 0.21421173214912415,
      "learning_rate": 8.869606835965896e-05,
      "loss": 1.4432,
      "step": 61100
    },
    {
      "epoch": 2.317685911223263,
      "grad_norm": 0.2673332691192627,
      "learning_rate": 8.868556881182838e-05,
      "loss": 1.4422,
      "step": 61110
    },
    {
      "epoch": 2.318227852388775,
      "grad_norm": 0.18308067321777344,
      "learning_rate": 8.867506509117578e-05,
      "loss": 1.426,
      "step": 61120
    },
    {
      "epoch": 2.3187697935542873,
      "grad_norm": 0.3170804977416992,
      "learning_rate": 8.866455719900234e-05,
      "loss": 1.4377,
      "step": 61130
    },
    {
      "epoch": 2.3193117347197996,
      "grad_norm": 0.28330197930336,
      "learning_rate": 8.865404513660968e-05,
      "loss": 1.437,
      "step": 61140
    },
    {
      "epoch": 2.319853675885312,
      "grad_norm": 0.17759016156196594,
      "learning_rate": 8.864352890530005e-05,
      "loss": 1.439,
      "step": 61150
    },
    {
      "epoch": 2.3203956170508238,
      "grad_norm": 0.33059704303741455,
      "learning_rate": 8.863300850637611e-05,
      "loss": 1.4346,
      "step": 61160
    },
    {
      "epoch": 2.320449811167375,
      "eval_loss": 2.450498580932617,
      "eval_runtime": 21.984,
      "eval_samples_per_second": 227.438,
      "eval_steps_per_second": 1.228,
      "step": 61161
    },
    {
      "epoch": 2.320937558216336,
      "grad_norm": 0.26688456535339355,
      "learning_rate": 8.862248394114111e-05,
      "loss": 1.4378,
      "step": 61170
    },
    {
      "epoch": 2.3214794993818484,
      "grad_norm": 0.25356730818748474,
      "learning_rate": 8.861195521089875e-05,
      "loss": 1.439,
      "step": 61180
    },
    {
      "epoch": 2.3220214405473607,
      "grad_norm": 0.33288466930389404,
      "learning_rate": 8.860142231695331e-05,
      "loss": 1.4367,
      "step": 61190
    },
    {
      "epoch": 2.3225633817128726,
      "grad_norm": 0.22826845943927765,
      "learning_rate": 8.859088526060957e-05,
      "loss": 1.4443,
      "step": 61200
    },
    {
      "epoch": 2.323105322878385,
      "grad_norm": 0.214629128575325,
      "learning_rate": 8.85803440431728e-05,
      "loss": 1.4391,
      "step": 61210
    },
    {
      "epoch": 2.323647264043897,
      "grad_norm": 0.21806223690509796,
      "learning_rate": 8.856979866594879e-05,
      "loss": 1.4335,
      "step": 61220
    },
    {
      "epoch": 2.3241892052094095,
      "grad_norm": 0.20262044668197632,
      "learning_rate": 8.855924913024388e-05,
      "loss": 1.4399,
      "step": 61230
    },
    {
      "epoch": 2.324731146374922,
      "grad_norm": 0.267966628074646,
      "learning_rate": 8.854869543736487e-05,
      "loss": 1.4412,
      "step": 61240
    },
    {
      "epoch": 2.3251646993073316,
      "eval_loss": 2.4549691677093506,
      "eval_runtime": 21.9814,
      "eval_samples_per_second": 227.465,
      "eval_steps_per_second": 1.228,
      "step": 61248
    },
    {
      "epoch": 2.3252730875404337,
      "grad_norm": 0.23619084060192108,
      "learning_rate": 8.853813758861915e-05,
      "loss": 1.442,
      "step": 61250
    },
    {
      "epoch": 2.325815028705946,
      "grad_norm": 0.4099680185317993,
      "learning_rate": 8.852757558531453e-05,
      "loss": 1.4374,
      "step": 61260
    },
    {
      "epoch": 2.3263569698714583,
      "grad_norm": 0.20198297500610352,
      "learning_rate": 8.851700942875943e-05,
      "loss": 1.4314,
      "step": 61270
    },
    {
      "epoch": 2.3268989110369707,
      "grad_norm": 0.22473087906837463,
      "learning_rate": 8.850643912026269e-05,
      "loss": 1.4323,
      "step": 61280
    },
    {
      "epoch": 2.327440852202483,
      "grad_norm": 0.19374053180217743,
      "learning_rate": 8.849586466113376e-05,
      "loss": 1.4348,
      "step": 61290
    },
    {
      "epoch": 2.327982793367995,
      "grad_norm": 0.20663167536258698,
      "learning_rate": 8.848528605268255e-05,
      "loss": 1.4335,
      "step": 61300
    },
    {
      "epoch": 2.328524734533507,
      "grad_norm": 0.2488778531551361,
      "learning_rate": 8.847470329621945e-05,
      "loss": 1.4479,
      "step": 61310
    },
    {
      "epoch": 2.3290666756990195,
      "grad_norm": 0.181378573179245,
      "learning_rate": 8.846411639305546e-05,
      "loss": 1.4368,
      "step": 61320
    },
    {
      "epoch": 2.329608616864532,
      "grad_norm": 0.2096365988254547,
      "learning_rate": 8.845352534450202e-05,
      "loss": 1.4265,
      "step": 61330
    },
    {
      "epoch": 2.3298795874472877,
      "eval_loss": 2.452427625656128,
      "eval_runtime": 21.9811,
      "eval_samples_per_second": 227.468,
      "eval_steps_per_second": 1.228,
      "step": 61335
    },
    {
      "epoch": 2.3301505580300437,
      "grad_norm": 0.3072015345096588,
      "learning_rate": 8.84429301518711e-05,
      "loss": 1.4511,
      "step": 61340
    },
    {
      "epoch": 2.330692499195556,
      "grad_norm": 0.21584992110729218,
      "learning_rate": 8.843233081647519e-05,
      "loss": 1.4421,
      "step": 61350
    },
    {
      "epoch": 2.3312344403610683,
      "grad_norm": 0.5617396831512451,
      "learning_rate": 8.842172733962727e-05,
      "loss": 1.4423,
      "step": 61360
    },
    {
      "epoch": 2.3317763815265806,
      "grad_norm": 0.194200336933136,
      "learning_rate": 8.841111972264088e-05,
      "loss": 1.4333,
      "step": 61370
    },
    {
      "epoch": 2.3323183226920925,
      "grad_norm": 0.457123339176178,
      "learning_rate": 8.840050796683005e-05,
      "loss": 1.441,
      "step": 61380
    },
    {
      "epoch": 2.332860263857605,
      "grad_norm": 0.24595646560192108,
      "learning_rate": 8.838989207350928e-05,
      "loss": 1.4481,
      "step": 61390
    },
    {
      "epoch": 2.333402205023117,
      "grad_norm": 0.22267189621925354,
      "learning_rate": 8.837927204399367e-05,
      "loss": 1.4456,
      "step": 61400
    },
    {
      "epoch": 2.3339441461886294,
      "grad_norm": 0.18372011184692383,
      "learning_rate": 8.836864787959873e-05,
      "loss": 1.4442,
      "step": 61410
    },
    {
      "epoch": 2.3344860873541418,
      "grad_norm": 0.18260882794857025,
      "learning_rate": 8.835801958164059e-05,
      "loss": 1.4382,
      "step": 61420
    },
    {
      "epoch": 2.334594475587244,
      "eval_loss": 2.4512012004852295,
      "eval_runtime": 21.9808,
      "eval_samples_per_second": 227.471,
      "eval_steps_per_second": 1.228,
      "step": 61422
    },
    {
      "epoch": 2.3350280285196536,
      "grad_norm": 0.238543301820755,
      "learning_rate": 8.834738715143577e-05,
      "loss": 1.4325,
      "step": 61430
    },
    {
      "epoch": 2.335569969685166,
      "grad_norm": 0.47301405668258667,
      "learning_rate": 8.833675059030144e-05,
      "loss": 1.4422,
      "step": 61440
    },
    {
      "epoch": 2.3361119108506783,
      "grad_norm": 0.23695412278175354,
      "learning_rate": 8.832610989955518e-05,
      "loss": 1.4361,
      "step": 61450
    },
    {
      "epoch": 2.3366538520161906,
      "grad_norm": 0.23388248682022095,
      "learning_rate": 8.831546508051511e-05,
      "loss": 1.441,
      "step": 61460
    },
    {
      "epoch": 2.337195793181703,
      "grad_norm": 0.2939755320549011,
      "learning_rate": 8.830481613449985e-05,
      "loss": 1.4408,
      "step": 61470
    },
    {
      "epoch": 2.3377377343472148,
      "grad_norm": 0.3122168779373169,
      "learning_rate": 8.82941630628286e-05,
      "loss": 1.4366,
      "step": 61480
    },
    {
      "epoch": 2.338279675512727,
      "grad_norm": 0.4362255036830902,
      "learning_rate": 8.828350586682096e-05,
      "loss": 1.4309,
      "step": 61490
    },
    {
      "epoch": 2.3388216166782394,
      "grad_norm": 0.3685493469238281,
      "learning_rate": 8.82728445477971e-05,
      "loss": 1.4369,
      "step": 61500
    },
    {
      "epoch": 2.3393093637272004,
      "eval_loss": 2.454096555709839,
      "eval_runtime": 21.9765,
      "eval_samples_per_second": 227.516,
      "eval_steps_per_second": 1.229,
      "step": 61509
    },
    {
      "epoch": 2.3393635578437517,
      "grad_norm": 0.3012941777706146,
      "learning_rate": 8.826217910707774e-05,
      "loss": 1.4327,
      "step": 61510
    },
    {
      "epoch": 2.339905499009264,
      "grad_norm": 0.4149247407913208,
      "learning_rate": 8.825150954598403e-05,
      "loss": 1.4361,
      "step": 61520
    },
    {
      "epoch": 2.340447440174776,
      "grad_norm": 0.17556257545948029,
      "learning_rate": 8.824083586583767e-05,
      "loss": 1.4273,
      "step": 61530
    },
    {
      "epoch": 2.3409893813402882,
      "grad_norm": 0.42168980836868286,
      "learning_rate": 8.823015806796089e-05,
      "loss": 1.4306,
      "step": 61540
    },
    {
      "epoch": 2.3415313225058005,
      "grad_norm": 0.29091402888298035,
      "learning_rate": 8.82194761536764e-05,
      "loss": 1.4425,
      "step": 61550
    },
    {
      "epoch": 2.342073263671313,
      "grad_norm": 0.33587974309921265,
      "learning_rate": 8.820879012430742e-05,
      "loss": 1.4383,
      "step": 61560
    },
    {
      "epoch": 2.3426152048368247,
      "grad_norm": 0.2704252600669861,
      "learning_rate": 8.81980999811777e-05,
      "loss": 1.4397,
      "step": 61570
    },
    {
      "epoch": 2.343157146002337,
      "grad_norm": 0.21759971976280212,
      "learning_rate": 8.81874057256115e-05,
      "loss": 1.4436,
      "step": 61580
    },
    {
      "epoch": 2.3436990871678494,
      "grad_norm": 0.28754401206970215,
      "learning_rate": 8.817670735893356e-05,
      "loss": 1.4314,
      "step": 61590
    },
    {
      "epoch": 2.3440242518671566,
      "eval_loss": 2.4510679244995117,
      "eval_runtime": 21.9791,
      "eval_samples_per_second": 227.489,
      "eval_steps_per_second": 1.228,
      "step": 61596
    },
    {
      "epoch": 2.3442410283333617,
      "grad_norm": 0.39755532145500183,
      "learning_rate": 8.816600488246914e-05,
      "loss": 1.441,
      "step": 61600
    },
    {
      "epoch": 2.3447829694988735,
      "grad_norm": 0.27198687195777893,
      "learning_rate": 8.815529829754403e-05,
      "loss": 1.4492,
      "step": 61610
    },
    {
      "epoch": 2.345324910664386,
      "grad_norm": 0.2784078121185303,
      "learning_rate": 8.814458760548452e-05,
      "loss": 1.4396,
      "step": 61620
    },
    {
      "epoch": 2.345866851829898,
      "grad_norm": 0.205166295170784,
      "learning_rate": 8.81338728076174e-05,
      "loss": 1.4402,
      "step": 61630
    },
    {
      "epoch": 2.3464087929954105,
      "grad_norm": 0.23504230380058289,
      "learning_rate": 8.812315390526997e-05,
      "loss": 1.4417,
      "step": 61640
    },
    {
      "epoch": 2.346950734160923,
      "grad_norm": 0.21687854826450348,
      "learning_rate": 8.811243089977002e-05,
      "loss": 1.4379,
      "step": 61650
    },
    {
      "epoch": 2.3474926753264347,
      "grad_norm": 0.27661964297294617,
      "learning_rate": 8.810170379244591e-05,
      "loss": 1.446,
      "step": 61660
    },
    {
      "epoch": 2.348034616491947,
      "grad_norm": 0.23024803400039673,
      "learning_rate": 8.809097258462645e-05,
      "loss": 1.4403,
      "step": 61670
    },
    {
      "epoch": 2.3485765576574593,
      "grad_norm": 0.4340742230415344,
      "learning_rate": 8.808023727764095e-05,
      "loss": 1.439,
      "step": 61680
    },
    {
      "epoch": 2.348739140007113,
      "eval_loss": 2.4533681869506836,
      "eval_runtime": 21.9799,
      "eval_samples_per_second": 227.481,
      "eval_steps_per_second": 1.228,
      "step": 61683
    },
    {
      "epoch": 2.3491184988229716,
      "grad_norm": 0.2405279576778412,
      "learning_rate": 8.806949787281929e-05,
      "loss": 1.444,
      "step": 61690
    },
    {
      "epoch": 2.349660439988484,
      "grad_norm": 0.3794168531894684,
      "learning_rate": 8.805875437149182e-05,
      "loss": 1.4397,
      "step": 61700
    },
    {
      "epoch": 2.350202381153996,
      "grad_norm": 0.272239625453949,
      "learning_rate": 8.804800677498935e-05,
      "loss": 1.4343,
      "step": 61710
    },
    {
      "epoch": 2.350744322319508,
      "grad_norm": 0.20025727152824402,
      "learning_rate": 8.803725508464332e-05,
      "loss": 1.4387,
      "step": 61720
    },
    {
      "epoch": 2.3512862634850205,
      "grad_norm": 0.24284347891807556,
      "learning_rate": 8.802649930178553e-05,
      "loss": 1.4398,
      "step": 61730
    },
    {
      "epoch": 2.3518282046505328,
      "grad_norm": 0.20169271528720856,
      "learning_rate": 8.801573942774842e-05,
      "loss": 1.4471,
      "step": 61740
    },
    {
      "epoch": 2.352370145816045,
      "grad_norm": 0.2248934954404831,
      "learning_rate": 8.800497546386484e-05,
      "loss": 1.4281,
      "step": 61750
    },
    {
      "epoch": 2.352912086981557,
      "grad_norm": 0.2882291376590729,
      "learning_rate": 8.79942074114682e-05,
      "loss": 1.4445,
      "step": 61760
    },
    {
      "epoch": 2.3534540281470693,
      "grad_norm": 0.544728696346283,
      "learning_rate": 8.798343527189238e-05,
      "loss": 1.4378,
      "step": 61770
    },
    {
      "epoch": 2.3534540281470693,
      "eval_loss": 2.448328971862793,
      "eval_runtime": 21.7136,
      "eval_samples_per_second": 230.271,
      "eval_steps_per_second": 1.243,
      "step": 61770
    },
    {
      "epoch": 2.3539959693125816,
      "grad_norm": 0.3125400245189667,
      "learning_rate": 8.797265904647179e-05,
      "loss": 1.4284,
      "step": 61780
    },
    {
      "epoch": 2.354537910478094,
      "grad_norm": 0.3059818744659424,
      "learning_rate": 8.796187873654138e-05,
      "loss": 1.4296,
      "step": 61790
    },
    {
      "epoch": 2.355079851643606,
      "grad_norm": 0.2530340254306793,
      "learning_rate": 8.795109434343652e-05,
      "loss": 1.4302,
      "step": 61800
    },
    {
      "epoch": 2.355621792809118,
      "grad_norm": 0.2673911452293396,
      "learning_rate": 8.794030586849315e-05,
      "loss": 1.4459,
      "step": 61810
    },
    {
      "epoch": 2.3561637339746304,
      "grad_norm": 0.3139052391052246,
      "learning_rate": 8.792951331304771e-05,
      "loss": 1.4379,
      "step": 61820
    },
    {
      "epoch": 2.3567056751401427,
      "grad_norm": 0.19884108006954193,
      "learning_rate": 8.791871667843713e-05,
      "loss": 1.4348,
      "step": 61830
    },
    {
      "epoch": 2.3572476163056546,
      "grad_norm": 0.2218913733959198,
      "learning_rate": 8.790791596599882e-05,
      "loss": 1.4465,
      "step": 61840
    },
    {
      "epoch": 2.357789557471167,
      "grad_norm": 0.18263356387615204,
      "learning_rate": 8.789711117707078e-05,
      "loss": 1.4378,
      "step": 61850
    },
    {
      "epoch": 2.3581689162870254,
      "eval_loss": 2.4493706226348877,
      "eval_runtime": 21.983,
      "eval_samples_per_second": 227.448,
      "eval_steps_per_second": 1.228,
      "step": 61857
    },
    {
      "epoch": 2.3583314986366792,
      "grad_norm": 0.2500559091567993,
      "learning_rate": 8.788630231299142e-05,
      "loss": 1.4311,
      "step": 61860
    },
    {
      "epoch": 2.3588734398021916,
      "grad_norm": 0.27540549635887146,
      "learning_rate": 8.787548937509971e-05,
      "loss": 1.431,
      "step": 61870
    },
    {
      "epoch": 2.359415380967704,
      "grad_norm": 0.23163899779319763,
      "learning_rate": 8.786467236473511e-05,
      "loss": 1.4425,
      "step": 61880
    },
    {
      "epoch": 2.3599573221332157,
      "grad_norm": 0.19727367162704468,
      "learning_rate": 8.785385128323759e-05,
      "loss": 1.45,
      "step": 61890
    },
    {
      "epoch": 2.360499263298728,
      "grad_norm": 0.2537407875061035,
      "learning_rate": 8.784302613194758e-05,
      "loss": 1.4446,
      "step": 61900
    },
    {
      "epoch": 2.3610412044642404,
      "grad_norm": 0.208670973777771,
      "learning_rate": 8.783219691220611e-05,
      "loss": 1.4393,
      "step": 61910
    },
    {
      "epoch": 2.3615831456297527,
      "grad_norm": 0.18386749923229218,
      "learning_rate": 8.782136362535462e-05,
      "loss": 1.4468,
      "step": 61920
    },
    {
      "epoch": 2.362125086795265,
      "grad_norm": 0.19878651201725006,
      "learning_rate": 8.781052627273512e-05,
      "loss": 1.444,
      "step": 61930
    },
    {
      "epoch": 2.362667027960777,
      "grad_norm": 0.27927711606025696,
      "learning_rate": 8.779968485569004e-05,
      "loss": 1.4331,
      "step": 61940
    },
    {
      "epoch": 2.362883804426982,
      "eval_loss": 2.4520583152770996,
      "eval_runtime": 21.9818,
      "eval_samples_per_second": 227.461,
      "eval_steps_per_second": 1.228,
      "step": 61944
    },
    {
      "epoch": 2.363208969126289,
      "grad_norm": 0.5157132744789124,
      "learning_rate": 8.778883937556243e-05,
      "loss": 1.4434,
      "step": 61950
    },
    {
      "epoch": 2.3637509102918015,
      "grad_norm": 0.6511501669883728,
      "learning_rate": 8.777798983369573e-05,
      "loss": 1.4315,
      "step": 61960
    },
    {
      "epoch": 2.364292851457314,
      "grad_norm": 0.21835680305957794,
      "learning_rate": 8.776713623143397e-05,
      "loss": 1.4358,
      "step": 61970
    },
    {
      "epoch": 2.3648347926228257,
      "grad_norm": 0.1703735738992691,
      "learning_rate": 8.775627857012165e-05,
      "loss": 1.4404,
      "step": 61980
    },
    {
      "epoch": 2.365376733788338,
      "grad_norm": 0.17746131122112274,
      "learning_rate": 8.774541685110373e-05,
      "loss": 1.4322,
      "step": 61990
    },
    {
      "epoch": 2.3659186749538503,
      "grad_norm": 0.34184667468070984,
      "learning_rate": 8.773455107572574e-05,
      "loss": 1.4246,
      "step": 62000
    },
    {
      "epoch": 2.3664606161193626,
      "grad_norm": 0.2798058092594147,
      "learning_rate": 8.772368124533369e-05,
      "loss": 1.4387,
      "step": 62010
    },
    {
      "epoch": 2.3670025572848745,
      "grad_norm": 0.40121662616729736,
      "learning_rate": 8.771280736127407e-05,
      "loss": 1.4394,
      "step": 62020
    },
    {
      "epoch": 2.367544498450387,
      "grad_norm": 0.26767289638519287,
      "learning_rate": 8.770192942489392e-05,
      "loss": 1.4371,
      "step": 62030
    },
    {
      "epoch": 2.367598692566938,
      "eval_loss": 2.441128969192505,
      "eval_runtime": 21.982,
      "eval_samples_per_second": 227.459,
      "eval_steps_per_second": 1.228,
      "step": 62031
    },
    {
      "epoch": 2.368086439615899,
      "grad_norm": 0.32896319031715393,
      "learning_rate": 8.76910474375407e-05,
      "loss": 1.4328,
      "step": 62040
    },
    {
      "epoch": 2.3686283807814115,
      "grad_norm": 0.4783940017223358,
      "learning_rate": 8.768016140056247e-05,
      "loss": 1.4453,
      "step": 62050
    },
    {
      "epoch": 2.369170321946924,
      "grad_norm": 0.29626142978668213,
      "learning_rate": 8.766927131530774e-05,
      "loss": 1.4437,
      "step": 62060
    },
    {
      "epoch": 2.3697122631124357,
      "grad_norm": 0.31796202063560486,
      "learning_rate": 8.765837718312549e-05,
      "loss": 1.4328,
      "step": 62070
    },
    {
      "epoch": 2.370254204277948,
      "grad_norm": 0.6041905283927917,
      "learning_rate": 8.764747900536529e-05,
      "loss": 1.4374,
      "step": 62080
    },
    {
      "epoch": 2.3707961454434603,
      "grad_norm": 0.2997055649757385,
      "learning_rate": 8.76365767833771e-05,
      "loss": 1.4417,
      "step": 62090
    },
    {
      "epoch": 2.3713380866089726,
      "grad_norm": 0.23351405560970306,
      "learning_rate": 8.76256705185115e-05,
      "loss": 1.4327,
      "step": 62100
    },
    {
      "epoch": 2.371880027774485,
      "grad_norm": 0.18303707242012024,
      "learning_rate": 8.761476021211947e-05,
      "loss": 1.4324,
      "step": 62110
    },
    {
      "epoch": 2.3723135807068947,
      "eval_loss": 2.431164503097534,
      "eval_runtime": 21.98,
      "eval_samples_per_second": 227.48,
      "eval_steps_per_second": 1.228,
      "step": 62118
    },
    {
      "epoch": 2.372421968939997,
      "grad_norm": 0.17883995175361633,
      "learning_rate": 8.760384586555255e-05,
      "loss": 1.4199,
      "step": 62120
    },
    {
      "epoch": 2.372963910105509,
      "grad_norm": 0.1785135418176651,
      "learning_rate": 8.759292748016275e-05,
      "loss": 1.4307,
      "step": 62130
    },
    {
      "epoch": 2.3735058512710214,
      "grad_norm": 0.3247034549713135,
      "learning_rate": 8.75820050573026e-05,
      "loss": 1.4367,
      "step": 62140
    },
    {
      "epoch": 2.3740477924365337,
      "grad_norm": 0.2257276326417923,
      "learning_rate": 8.757107859832512e-05,
      "loss": 1.4524,
      "step": 62150
    },
    {
      "epoch": 2.374589733602046,
      "grad_norm": 0.34088021516799927,
      "learning_rate": 8.756014810458384e-05,
      "loss": 1.4359,
      "step": 62160
    },
    {
      "epoch": 2.375131674767558,
      "grad_norm": 0.2236095666885376,
      "learning_rate": 8.754921357743277e-05,
      "loss": 1.439,
      "step": 62170
    },
    {
      "epoch": 2.3756736159330702,
      "grad_norm": 0.2399052530527115,
      "learning_rate": 8.753827501822643e-05,
      "loss": 1.4326,
      "step": 62180
    },
    {
      "epoch": 2.3762155570985826,
      "grad_norm": 0.18329595029354095,
      "learning_rate": 8.752733242831985e-05,
      "loss": 1.4373,
      "step": 62190
    },
    {
      "epoch": 2.376757498264095,
      "grad_norm": 0.2353314459323883,
      "learning_rate": 8.751638580906856e-05,
      "loss": 1.441,
      "step": 62200
    },
    {
      "epoch": 2.377028468846851,
      "eval_loss": 2.4490318298339844,
      "eval_runtime": 21.9823,
      "eval_samples_per_second": 227.456,
      "eval_steps_per_second": 1.228,
      "step": 62205
    },
    {
      "epoch": 2.3772994394296068,
      "grad_norm": 0.30364421010017395,
      "learning_rate": 8.750543516182855e-05,
      "loss": 1.4362,
      "step": 62210
    },
    {
      "epoch": 2.377841380595119,
      "grad_norm": 0.1798836588859558,
      "learning_rate": 8.749448048795637e-05,
      "loss": 1.4259,
      "step": 62220
    },
    {
      "epoch": 2.3783833217606314,
      "grad_norm": 0.24086004495620728,
      "learning_rate": 8.748352178880902e-05,
      "loss": 1.4362,
      "step": 62230
    },
    {
      "epoch": 2.3789252629261437,
      "grad_norm": 0.23006141185760498,
      "learning_rate": 8.747255906574402e-05,
      "loss": 1.4397,
      "step": 62240
    },
    {
      "epoch": 2.3794672040916556,
      "grad_norm": 0.23331311345100403,
      "learning_rate": 8.74615923201194e-05,
      "loss": 1.4479,
      "step": 62250
    },
    {
      "epoch": 2.380009145257168,
      "grad_norm": 0.2259208709001541,
      "learning_rate": 8.745062155329363e-05,
      "loss": 1.4515,
      "step": 62260
    },
    {
      "epoch": 2.38055108642268,
      "grad_norm": 0.3277144730091095,
      "learning_rate": 8.743964676662576e-05,
      "loss": 1.4405,
      "step": 62270
    },
    {
      "epoch": 2.3810930275881925,
      "grad_norm": 0.25958386063575745,
      "learning_rate": 8.742866796147528e-05,
      "loss": 1.4254,
      "step": 62280
    },
    {
      "epoch": 2.381634968753705,
      "grad_norm": 0.2229194939136505,
      "learning_rate": 8.741768513920222e-05,
      "loss": 1.426,
      "step": 62290
    },
    {
      "epoch": 2.381743356986807,
      "eval_loss": 2.435068368911743,
      "eval_runtime": 21.9789,
      "eval_samples_per_second": 227.491,
      "eval_steps_per_second": 1.228,
      "step": 62292
    },
    {
      "epoch": 2.3821769099192167,
      "grad_norm": 0.4171644449234009,
      "learning_rate": 8.740669830116706e-05,
      "loss": 1.4468,
      "step": 62300
    },
    {
      "epoch": 2.382718851084729,
      "grad_norm": 0.17529958486557007,
      "learning_rate": 8.739570744873081e-05,
      "loss": 1.4307,
      "step": 62310
    },
    {
      "epoch": 2.3832607922502413,
      "grad_norm": 0.25872156023979187,
      "learning_rate": 8.738471258325499e-05,
      "loss": 1.4456,
      "step": 62320
    },
    {
      "epoch": 2.3838027334157537,
      "grad_norm": 0.18120713531970978,
      "learning_rate": 8.737371370610156e-05,
      "loss": 1.4495,
      "step": 62330
    },
    {
      "epoch": 2.384344674581266,
      "grad_norm": 0.3255040645599365,
      "learning_rate": 8.736271081863302e-05,
      "loss": 1.4378,
      "step": 62340
    },
    {
      "epoch": 2.384886615746778,
      "grad_norm": 0.22683602571487427,
      "learning_rate": 8.735170392221237e-05,
      "loss": 1.4417,
      "step": 62350
    },
    {
      "epoch": 2.38542855691229,
      "grad_norm": 0.5052164196968079,
      "learning_rate": 8.73406930182031e-05,
      "loss": 1.4369,
      "step": 62360
    },
    {
      "epoch": 2.3859704980778025,
      "grad_norm": 0.27900445461273193,
      "learning_rate": 8.732967810796918e-05,
      "loss": 1.4283,
      "step": 62370
    },
    {
      "epoch": 2.3864582451267635,
      "eval_loss": 2.445622682571411,
      "eval_runtime": 21.9893,
      "eval_samples_per_second": 227.383,
      "eval_steps_per_second": 1.228,
      "step": 62379
    },
    {
      "epoch": 2.386512439243315,
      "grad_norm": 0.30575114488601685,
      "learning_rate": 8.73186591928751e-05,
      "loss": 1.438,
      "step": 62380
    },
    {
      "epoch": 2.387054380408827,
      "grad_norm": 0.2561473846435547,
      "learning_rate": 8.730763627428585e-05,
      "loss": 1.4324,
      "step": 62390
    },
    {
      "epoch": 2.387596321574339,
      "grad_norm": 0.2688450515270233,
      "learning_rate": 8.729660935356686e-05,
      "loss": 1.4184,
      "step": 62400
    },
    {
      "epoch": 2.3881382627398513,
      "grad_norm": 0.3544863164424896,
      "learning_rate": 8.728557843208413e-05,
      "loss": 1.4436,
      "step": 62410
    },
    {
      "epoch": 2.3886802039053636,
      "grad_norm": 0.30758488178253174,
      "learning_rate": 8.727454351120413e-05,
      "loss": 1.427,
      "step": 62420
    },
    {
      "epoch": 2.389222145070876,
      "grad_norm": 0.41708388924598694,
      "learning_rate": 8.726350459229379e-05,
      "loss": 1.4349,
      "step": 62430
    },
    {
      "epoch": 2.389764086236388,
      "grad_norm": 0.5974102020263672,
      "learning_rate": 8.725246167672057e-05,
      "loss": 1.4272,
      "step": 62440
    },
    {
      "epoch": 2.3903060274019,
      "grad_norm": 0.19719624519348145,
      "learning_rate": 8.724141476585244e-05,
      "loss": 1.4359,
      "step": 62450
    },
    {
      "epoch": 2.3908479685674124,
      "grad_norm": 0.3106327950954437,
      "learning_rate": 8.723036386105781e-05,
      "loss": 1.4333,
      "step": 62460
    },
    {
      "epoch": 2.3911731332667197,
      "eval_loss": 2.4473304748535156,
      "eval_runtime": 21.9738,
      "eval_samples_per_second": 227.544,
      "eval_steps_per_second": 1.229,
      "step": 62466
    },
    {
      "epoch": 2.3913899097329248,
      "grad_norm": 0.2692268192768097,
      "learning_rate": 8.721930896370563e-05,
      "loss": 1.421,
      "step": 62470
    },
    {
      "epoch": 2.3919318508984366,
      "grad_norm": 0.2162933647632599,
      "learning_rate": 8.720825007516535e-05,
      "loss": 1.4401,
      "step": 62480
    },
    {
      "epoch": 2.392473792063949,
      "grad_norm": 0.3203500509262085,
      "learning_rate": 8.719718719680687e-05,
      "loss": 1.4334,
      "step": 62490
    },
    {
      "epoch": 2.3930157332294613,
      "grad_norm": 0.3285951018333435,
      "learning_rate": 8.718612033000066e-05,
      "loss": 1.4321,
      "step": 62500
    },
    {
      "epoch": 2.3935576743949736,
      "grad_norm": 0.4296458065509796,
      "learning_rate": 8.717504947611757e-05,
      "loss": 1.4345,
      "step": 62510
    },
    {
      "epoch": 2.394099615560486,
      "grad_norm": 0.1817467212677002,
      "learning_rate": 8.716397463652907e-05,
      "loss": 1.4386,
      "step": 62520
    },
    {
      "epoch": 2.3946415567259978,
      "grad_norm": 0.3981388807296753,
      "learning_rate": 8.7152895812607e-05,
      "loss": 1.4317,
      "step": 62530
    },
    {
      "epoch": 2.39518349789151,
      "grad_norm": 0.2751355767250061,
      "learning_rate": 8.71418130057238e-05,
      "loss": 1.4464,
      "step": 62540
    },
    {
      "epoch": 2.3957254390570224,
      "grad_norm": 0.2315666377544403,
      "learning_rate": 8.713072621725235e-05,
      "loss": 1.4424,
      "step": 62550
    },
    {
      "epoch": 2.3958880214066762,
      "eval_loss": 2.4334030151367188,
      "eval_runtime": 22.0386,
      "eval_samples_per_second": 226.874,
      "eval_steps_per_second": 1.225,
      "step": 62553
    },
    {
      "epoch": 2.3962673802225347,
      "grad_norm": 0.2237350344657898,
      "learning_rate": 8.711963544856606e-05,
      "loss": 1.4372,
      "step": 62560
    },
    {
      "epoch": 2.396809321388047,
      "grad_norm": 0.17813153564929962,
      "learning_rate": 8.710854070103876e-05,
      "loss": 1.4269,
      "step": 62570
    },
    {
      "epoch": 2.397351262553559,
      "grad_norm": 0.2021496742963791,
      "learning_rate": 8.709744197604483e-05,
      "loss": 1.4307,
      "step": 62580
    },
    {
      "epoch": 2.397893203719071,
      "grad_norm": 0.1962280124425888,
      "learning_rate": 8.708633927495916e-05,
      "loss": 1.4401,
      "step": 62590
    },
    {
      "epoch": 2.3984351448845835,
      "grad_norm": 0.29893505573272705,
      "learning_rate": 8.707523259915707e-05,
      "loss": 1.4374,
      "step": 62600
    },
    {
      "epoch": 2.398977086050096,
      "grad_norm": 0.24217230081558228,
      "learning_rate": 8.706412195001444e-05,
      "loss": 1.4346,
      "step": 62610
    },
    {
      "epoch": 2.3995190272156077,
      "grad_norm": 0.24626600742340088,
      "learning_rate": 8.705300732890756e-05,
      "loss": 1.4344,
      "step": 62620
    },
    {
      "epoch": 2.40006096838112,
      "grad_norm": 0.17191511392593384,
      "learning_rate": 8.704188873721332e-05,
      "loss": 1.443,
      "step": 62630
    },
    {
      "epoch": 2.4006029095466324,
      "grad_norm": 0.20966501533985138,
      "learning_rate": 8.703076617630901e-05,
      "loss": 1.4345,
      "step": 62640
    },
    {
      "epoch": 2.4006029095466324,
      "eval_loss": 2.4349613189697266,
      "eval_runtime": 21.971,
      "eval_samples_per_second": 227.572,
      "eval_steps_per_second": 1.229,
      "step": 62640
    },
    {
      "epoch": 2.4011448507121447,
      "grad_norm": 0.190885990858078,
      "learning_rate": 8.701963964757245e-05,
      "loss": 1.4334,
      "step": 62650
    },
    {
      "epoch": 2.4016867918776565,
      "grad_norm": 0.26069727540016174,
      "learning_rate": 8.700850915238195e-05,
      "loss": 1.4388,
      "step": 62660
    },
    {
      "epoch": 2.402228733043169,
      "grad_norm": 0.33938780426979065,
      "learning_rate": 8.699737469211629e-05,
      "loss": 1.4331,
      "step": 62670
    },
    {
      "epoch": 2.402770674208681,
      "grad_norm": 0.23580268025398254,
      "learning_rate": 8.698623626815478e-05,
      "loss": 1.4318,
      "step": 62680
    },
    {
      "epoch": 2.4033126153741935,
      "grad_norm": 0.23255738615989685,
      "learning_rate": 8.697509388187721e-05,
      "loss": 1.4455,
      "step": 62690
    },
    {
      "epoch": 2.403854556539706,
      "grad_norm": 0.3351677656173706,
      "learning_rate": 8.696394753466381e-05,
      "loss": 1.4373,
      "step": 62700
    },
    {
      "epoch": 2.4043964977052177,
      "grad_norm": 0.18452972173690796,
      "learning_rate": 8.695279722789536e-05,
      "loss": 1.4417,
      "step": 62710
    },
    {
      "epoch": 2.40493843887073,
      "grad_norm": 0.21630007028579712,
      "learning_rate": 8.694164296295311e-05,
      "loss": 1.434,
      "step": 62720
    },
    {
      "epoch": 2.4053177976865885,
      "eval_loss": 2.435746908187866,
      "eval_runtime": 22.0344,
      "eval_samples_per_second": 226.918,
      "eval_steps_per_second": 1.225,
      "step": 62727
    },
    {
      "epoch": 2.4054803800362423,
      "grad_norm": 0.3593086004257202,
      "learning_rate": 8.693048474121883e-05,
      "loss": 1.4383,
      "step": 62730
    },
    {
      "epoch": 2.4060223212017546,
      "grad_norm": 0.4216225743293762,
      "learning_rate": 8.69193225640747e-05,
      "loss": 1.4458,
      "step": 62740
    },
    {
      "epoch": 2.406564262367267,
      "grad_norm": 0.2987191379070282,
      "learning_rate": 8.690815643290348e-05,
      "loss": 1.4488,
      "step": 62750
    },
    {
      "epoch": 2.407106203532779,
      "grad_norm": 0.2709275186061859,
      "learning_rate": 8.68969863490884e-05,
      "loss": 1.4483,
      "step": 62760
    },
    {
      "epoch": 2.407648144698291,
      "grad_norm": 0.17814864218235016,
      "learning_rate": 8.68858123140131e-05,
      "loss": 1.4451,
      "step": 62770
    },
    {
      "epoch": 2.4081900858638035,
      "grad_norm": 0.24348965287208557,
      "learning_rate": 8.687463432906182e-05,
      "loss": 1.4463,
      "step": 62780
    },
    {
      "epoch": 2.4087320270293158,
      "grad_norm": 0.21512000262737274,
      "learning_rate": 8.686345239561921e-05,
      "loss": 1.4508,
      "step": 62790
    },
    {
      "epoch": 2.409273968194828,
      "grad_norm": 0.3042374551296234,
      "learning_rate": 8.685226651507047e-05,
      "loss": 1.4606,
      "step": 62800
    },
    {
      "epoch": 2.40981590936034,
      "grad_norm": 0.22630225121974945,
      "learning_rate": 8.684107668880124e-05,
      "loss": 1.4499,
      "step": 62810
    },
    {
      "epoch": 2.410032685826545,
      "eval_loss": 2.45141339302063,
      "eval_runtime": 21.9807,
      "eval_samples_per_second": 227.472,
      "eval_steps_per_second": 1.228,
      "step": 62814
    },
    {
      "epoch": 2.4103578505258523,
      "grad_norm": 0.22491927444934845,
      "learning_rate": 8.682988291819766e-05,
      "loss": 1.4527,
      "step": 62820
    },
    {
      "epoch": 2.4108997916913646,
      "grad_norm": 0.18473847210407257,
      "learning_rate": 8.68186852046464e-05,
      "loss": 1.4372,
      "step": 62830
    },
    {
      "epoch": 2.411441732856877,
      "grad_norm": 0.18476863205432892,
      "learning_rate": 8.680748354953454e-05,
      "loss": 1.4481,
      "step": 62840
    },
    {
      "epoch": 2.411983674022389,
      "grad_norm": 0.20057182013988495,
      "learning_rate": 8.679627795424973e-05,
      "loss": 1.442,
      "step": 62850
    },
    {
      "epoch": 2.412525615187901,
      "grad_norm": 0.2545701563358307,
      "learning_rate": 8.678506842018002e-05,
      "loss": 1.4425,
      "step": 62860
    },
    {
      "epoch": 2.4130675563534134,
      "grad_norm": 0.18435277044773102,
      "learning_rate": 8.677385494871406e-05,
      "loss": 1.4428,
      "step": 62870
    },
    {
      "epoch": 2.4136094975189257,
      "grad_norm": 0.3235138952732086,
      "learning_rate": 8.676263754124089e-05,
      "loss": 1.4548,
      "step": 62880
    },
    {
      "epoch": 2.4141514386844376,
      "grad_norm": 0.36199185252189636,
      "learning_rate": 8.675141619915008e-05,
      "loss": 1.4421,
      "step": 62890
    },
    {
      "epoch": 2.41469337984995,
      "grad_norm": 0.19079086184501648,
      "learning_rate": 8.674019092383168e-05,
      "loss": 1.4568,
      "step": 62900
    },
    {
      "epoch": 2.414747573966501,
      "eval_loss": 2.44244122505188,
      "eval_runtime": 21.9822,
      "eval_samples_per_second": 227.456,
      "eval_steps_per_second": 1.228,
      "step": 62901
    },
    {
      "epoch": 2.4152353210154622,
      "grad_norm": 0.20537631213665009,
      "learning_rate": 8.672896171667623e-05,
      "loss": 1.454,
      "step": 62910
    },
    {
      "epoch": 2.4157772621809745,
      "grad_norm": 0.1923794001340866,
      "learning_rate": 8.671772857907476e-05,
      "loss": 1.444,
      "step": 62920
    },
    {
      "epoch": 2.416319203346487,
      "grad_norm": 0.3296695649623871,
      "learning_rate": 8.670649151241876e-05,
      "loss": 1.4404,
      "step": 62930
    },
    {
      "epoch": 2.4168611445119987,
      "grad_norm": 0.4826582670211792,
      "learning_rate": 8.669525051810028e-05,
      "loss": 1.4398,
      "step": 62940
    },
    {
      "epoch": 2.417403085677511,
      "grad_norm": 0.23408730328083038,
      "learning_rate": 8.668400559751175e-05,
      "loss": 1.4405,
      "step": 62950
    },
    {
      "epoch": 2.4179450268430234,
      "grad_norm": 0.461852103471756,
      "learning_rate": 8.667275675204617e-05,
      "loss": 1.4522,
      "step": 62960
    },
    {
      "epoch": 2.4184869680085357,
      "grad_norm": 0.22658464312553406,
      "learning_rate": 8.6661503983097e-05,
      "loss": 1.4552,
      "step": 62970
    },
    {
      "epoch": 2.419028909174048,
      "grad_norm": 0.1908838152885437,
      "learning_rate": 8.665024729205816e-05,
      "loss": 1.4352,
      "step": 62980
    },
    {
      "epoch": 2.4194624621064573,
      "eval_loss": 2.4345483779907227,
      "eval_runtime": 21.9764,
      "eval_samples_per_second": 227.517,
      "eval_steps_per_second": 1.229,
      "step": 62988
    },
    {
      "epoch": 2.41957085033956,
      "grad_norm": 0.24472688138484955,
      "learning_rate": 8.663898668032412e-05,
      "loss": 1.4473,
      "step": 62990
    },
    {
      "epoch": 2.420112791505072,
      "grad_norm": 0.2447074055671692,
      "learning_rate": 8.662772214928976e-05,
      "loss": 1.4536,
      "step": 63000
    },
    {
      "epoch": 3.0005419411655123,
      "grad_norm": 0.16394081711769104,
      "learning_rate": 8.661645370035048e-05,
      "loss": 1.4525,
      "step": 63010
    },
    {
      "epoch": 3.0010838823310246,
      "grad_norm": 0.2631014585494995,
      "learning_rate": 8.660518133490221e-05,
      "loss": 1.4492,
      "step": 63020
    },
    {
      "epoch": 3.0016258234965365,
      "grad_norm": 0.1863064467906952,
      "learning_rate": 8.659390505434127e-05,
      "loss": 1.4496,
      "step": 63030
    },
    {
      "epoch": 3.002167764662049,
      "grad_norm": 0.1799326092004776,
      "learning_rate": 8.658262486006455e-05,
      "loss": 1.4565,
      "step": 63040
    },
    {
      "epoch": 3.002709705827561,
      "grad_norm": 0.18670876324176788,
      "learning_rate": 8.657134075346938e-05,
      "loss": 1.4478,
      "step": 63050
    },
    {
      "epoch": 3.0032516469930735,
      "grad_norm": 0.23742614686489105,
      "learning_rate": 8.65600527359536e-05,
      "loss": 1.4542,
      "step": 63060
    },
    {
      "epoch": 3.0037935881585853,
      "grad_norm": 0.21548855304718018,
      "learning_rate": 8.654876080891547e-05,
      "loss": 1.4522,
      "step": 63070
    },
    {
      "epoch": 3.0040645587413417,
      "eval_loss": 2.447054624557495,
      "eval_runtime": 22.3909,
      "eval_samples_per_second": 223.305,
      "eval_steps_per_second": 1.206,
      "step": 63075
    },
    {
      "epoch": 3.0043355293240976,
      "grad_norm": 0.20155511796474457,
      "learning_rate": 8.653746497375385e-05,
      "loss": 1.4611,
      "step": 63080
    },
    {
      "epoch": 3.00487747048961,
      "grad_norm": 0.21091540157794952,
      "learning_rate": 8.652616523186797e-05,
      "loss": 1.444,
      "step": 63090
    },
    {
      "epoch": 3.0054194116551223,
      "grad_norm": 0.20185105502605438,
      "learning_rate": 8.651486158465764e-05,
      "loss": 1.453,
      "step": 63100
    },
    {
      "epoch": 3.0059613528206346,
      "grad_norm": 0.21705321967601776,
      "learning_rate": 8.650355403352307e-05,
      "loss": 1.4463,
      "step": 63110
    },
    {
      "epoch": 3.0065032939861465,
      "grad_norm": 0.22515122592449188,
      "learning_rate": 8.649224257986499e-05,
      "loss": 1.4393,
      "step": 63120
    },
    {
      "epoch": 3.007045235151659,
      "grad_norm": 0.2835371494293213,
      "learning_rate": 8.648092722508463e-05,
      "loss": 1.449,
      "step": 63130
    },
    {
      "epoch": 3.007587176317171,
      "grad_norm": 0.36949828267097473,
      "learning_rate": 8.64696079705837e-05,
      "loss": 1.4525,
      "step": 63140
    },
    {
      "epoch": 3.0081291174826834,
      "grad_norm": 0.23466630280017853,
      "learning_rate": 8.645828481776434e-05,
      "loss": 1.4465,
      "step": 63150
    },
    {
      "epoch": 3.0086710586481953,
      "grad_norm": 0.2540377676486969,
      "learning_rate": 8.644695776802925e-05,
      "loss": 1.454,
      "step": 63160
    },
    {
      "epoch": 3.008779446881298,
      "eval_loss": 2.4413795471191406,
      "eval_runtime": 21.9823,
      "eval_samples_per_second": 227.456,
      "eval_steps_per_second": 1.228,
      "step": 63162
    },
    {
      "epoch": 3.0092129998137076,
      "grad_norm": 0.3684237599372864,
      "learning_rate": 8.643562682278154e-05,
      "loss": 1.4465,
      "step": 63170
    },
    {
      "epoch": 3.00975494097922,
      "grad_norm": 0.1901058703660965,
      "learning_rate": 8.642429198342488e-05,
      "loss": 1.4547,
      "step": 63180
    },
    {
      "epoch": 3.0102968821447322,
      "grad_norm": 0.18713931739330292,
      "learning_rate": 8.641295325136336e-05,
      "loss": 1.441,
      "step": 63190
    },
    {
      "epoch": 3.0108388233102445,
      "grad_norm": 0.18807746469974518,
      "learning_rate": 8.640161062800155e-05,
      "loss": 1.4493,
      "step": 63200
    },
    {
      "epoch": 3.0113807644757564,
      "grad_norm": 0.19221310317516327,
      "learning_rate": 8.639026411474457e-05,
      "loss": 1.44,
      "step": 63210
    },
    {
      "epoch": 3.0119227056412687,
      "grad_norm": 0.2589902877807617,
      "learning_rate": 8.637891371299796e-05,
      "loss": 1.4451,
      "step": 63220
    },
    {
      "epoch": 3.012464646806781,
      "grad_norm": 0.22735221683979034,
      "learning_rate": 8.636755942416774e-05,
      "loss": 1.4497,
      "step": 63230
    },
    {
      "epoch": 3.0130065879722934,
      "grad_norm": 0.17304742336273193,
      "learning_rate": 8.635620124966043e-05,
      "loss": 1.4586,
      "step": 63240
    },
    {
      "epoch": 3.0134943350212544,
      "eval_loss": 2.4555513858795166,
      "eval_runtime": 21.9864,
      "eval_samples_per_second": 227.414,
      "eval_steps_per_second": 1.228,
      "step": 63249
    },
    {
      "epoch": 3.0135485291378057,
      "grad_norm": 0.19638432562351227,
      "learning_rate": 8.634483919088306e-05,
      "loss": 1.4583,
      "step": 63250
    },
    {
      "epoch": 3.0140904703033176,
      "grad_norm": 0.3353283703327179,
      "learning_rate": 8.633347324924309e-05,
      "loss": 1.4459,
      "step": 63260
    },
    {
      "epoch": 3.01463241146883,
      "grad_norm": 0.23562084138393402,
      "learning_rate": 8.63221034261485e-05,
      "loss": 1.449,
      "step": 63270
    },
    {
      "epoch": 3.015174352634342,
      "grad_norm": 0.23831287026405334,
      "learning_rate": 8.63107297230077e-05,
      "loss": 1.4461,
      "step": 63280
    },
    {
      "epoch": 3.0157162937998545,
      "grad_norm": 0.2882630228996277,
      "learning_rate": 8.629935214122968e-05,
      "loss": 1.453,
      "step": 63290
    },
    {
      "epoch": 3.0162582349653664,
      "grad_norm": 0.26988014578819275,
      "learning_rate": 8.628797068222378e-05,
      "loss": 1.451,
      "step": 63300
    },
    {
      "epoch": 3.0168001761308787,
      "grad_norm": 0.2622480094432831,
      "learning_rate": 8.627658534739992e-05,
      "loss": 1.4587,
      "step": 63310
    },
    {
      "epoch": 3.017342117296391,
      "grad_norm": 0.2087298184633255,
      "learning_rate": 8.626519613816844e-05,
      "loss": 1.4502,
      "step": 63320
    },
    {
      "epoch": 3.0178840584619033,
      "grad_norm": 0.20758210122585297,
      "learning_rate": 8.62538030559402e-05,
      "loss": 1.4498,
      "step": 63330
    },
    {
      "epoch": 3.0182092231612105,
      "eval_loss": 2.4477882385253906,
      "eval_runtime": 21.9833,
      "eval_samples_per_second": 227.445,
      "eval_steps_per_second": 1.228,
      "step": 63336
    },
    {
      "epoch": 3.0184259996274156,
      "grad_norm": 0.1754673719406128,
      "learning_rate": 8.624240610212656e-05,
      "loss": 1.4539,
      "step": 63340
    },
    {
      "epoch": 3.0189679407929275,
      "grad_norm": 0.15522533655166626,
      "learning_rate": 8.623100527813928e-05,
      "loss": 1.4482,
      "step": 63350
    },
    {
      "epoch": 3.01950988195844,
      "grad_norm": 0.22570842504501343,
      "learning_rate": 8.621960058539062e-05,
      "loss": 1.4564,
      "step": 63360
    },
    {
      "epoch": 3.020051823123952,
      "grad_norm": 0.20296502113342285,
      "learning_rate": 8.620819202529342e-05,
      "loss": 1.4586,
      "step": 63370
    },
    {
      "epoch": 3.0205937642894645,
      "grad_norm": 0.2882785201072693,
      "learning_rate": 8.619677959926089e-05,
      "loss": 1.4478,
      "step": 63380
    },
    {
      "epoch": 3.0211357054549763,
      "grad_norm": 0.3313170075416565,
      "learning_rate": 8.618536330870673e-05,
      "loss": 1.4577,
      "step": 63390
    },
    {
      "epoch": 3.0216776466204887,
      "grad_norm": 0.19241797924041748,
      "learning_rate": 8.617394315504516e-05,
      "loss": 1.4574,
      "step": 63400
    },
    {
      "epoch": 3.022219587786001,
      "grad_norm": 0.32396382093429565,
      "learning_rate": 8.616251913969085e-05,
      "loss": 1.4519,
      "step": 63410
    },
    {
      "epoch": 3.0227615289515133,
      "grad_norm": 0.1714586317539215,
      "learning_rate": 8.615109126405897e-05,
      "loss": 1.4567,
      "step": 63420
    },
    {
      "epoch": 3.0229241113011667,
      "eval_loss": 2.4477341175079346,
      "eval_runtime": 21.9816,
      "eval_samples_per_second": 227.463,
      "eval_steps_per_second": 1.228,
      "step": 63423
    },
    {
      "epoch": 3.0233034701170256,
      "grad_norm": 0.1809176504611969,
      "learning_rate": 8.613965952956515e-05,
      "loss": 1.446,
      "step": 63430
    },
    {
      "epoch": 3.0238454112825375,
      "grad_norm": 0.20341655611991882,
      "learning_rate": 8.61282239376255e-05,
      "loss": 1.4551,
      "step": 63440
    },
    {
      "epoch": 3.02438735244805,
      "grad_norm": 0.1895596832036972,
      "learning_rate": 8.611678448965661e-05,
      "loss": 1.4487,
      "step": 63450
    },
    {
      "epoch": 3.024929293613562,
      "grad_norm": 0.2253957986831665,
      "learning_rate": 8.610534118707556e-05,
      "loss": 1.4559,
      "step": 63460
    },
    {
      "epoch": 3.0254712347790744,
      "grad_norm": 0.2183631956577301,
      "learning_rate": 8.609389403129988e-05,
      "loss": 1.4415,
      "step": 63470
    },
    {
      "epoch": 3.0260131759445863,
      "grad_norm": 0.3052237629890442,
      "learning_rate": 8.608244302374762e-05,
      "loss": 1.4455,
      "step": 63480
    },
    {
      "epoch": 3.0265551171100986,
      "grad_norm": 0.28673261404037476,
      "learning_rate": 8.607098816583725e-05,
      "loss": 1.4617,
      "step": 63490
    },
    {
      "epoch": 3.027097058275611,
      "grad_norm": 0.20791488885879517,
      "learning_rate": 8.605952945898777e-05,
      "loss": 1.4532,
      "step": 63500
    },
    {
      "epoch": 3.0276389994411232,
      "grad_norm": 0.34028980135917664,
      "learning_rate": 8.604806690461863e-05,
      "loss": 1.4535,
      "step": 63510
    },
    {
      "epoch": 3.0276389994411232,
      "eval_loss": 2.4497005939483643,
      "eval_runtime": 21.9833,
      "eval_samples_per_second": 227.446,
      "eval_steps_per_second": 1.228,
      "step": 63510
    },
    {
      "epoch": 3.0281809406066356,
      "grad_norm": 0.22710201144218445,
      "learning_rate": 8.603660050414974e-05,
      "loss": 1.4425,
      "step": 63520
    },
    {
      "epoch": 3.0287228817721474,
      "grad_norm": 0.17469532787799835,
      "learning_rate": 8.602513025900155e-05,
      "loss": 1.4445,
      "step": 63530
    },
    {
      "epoch": 3.0292648229376598,
      "grad_norm": 0.21239697933197021,
      "learning_rate": 8.601365617059491e-05,
      "loss": 1.4371,
      "step": 63540
    },
    {
      "epoch": 3.029806764103172,
      "grad_norm": 0.1866355538368225,
      "learning_rate": 8.60021782403512e-05,
      "loss": 1.4426,
      "step": 63550
    },
    {
      "epoch": 3.0303487052686844,
      "grad_norm": 0.175773486495018,
      "learning_rate": 8.599069646969223e-05,
      "loss": 1.4477,
      "step": 63560
    },
    {
      "epoch": 3.0308906464341967,
      "grad_norm": 0.2518947124481201,
      "learning_rate": 8.597921086004035e-05,
      "loss": 1.4465,
      "step": 63570
    },
    {
      "epoch": 3.0314325875997086,
      "grad_norm": 0.3670835494995117,
      "learning_rate": 8.596772141281833e-05,
      "loss": 1.4487,
      "step": 63580
    },
    {
      "epoch": 3.031974528765221,
      "grad_norm": 0.3510816991329193,
      "learning_rate": 8.59562281294494e-05,
      "loss": 1.4518,
      "step": 63590
    },
    {
      "epoch": 3.0323538875810794,
      "eval_loss": 2.4546306133270264,
      "eval_runtime": 22.1411,
      "eval_samples_per_second": 225.824,
      "eval_steps_per_second": 1.219,
      "step": 63597
    },
    {
      "epoch": 3.032516469930733,
      "grad_norm": 0.3301527500152588,
      "learning_rate": 8.594473101135734e-05,
      "loss": 1.4655,
      "step": 63600
    },
    {
      "epoch": 3.0330584110962455,
      "grad_norm": 0.2271365225315094,
      "learning_rate": 8.593323005996638e-05,
      "loss": 1.4479,
      "step": 63610
    },
    {
      "epoch": 3.0336003522617574,
      "grad_norm": 0.3536682724952698,
      "learning_rate": 8.592172527670114e-05,
      "loss": 1.4523,
      "step": 63620
    },
    {
      "epoch": 3.0341422934272697,
      "grad_norm": 0.4474346935749054,
      "learning_rate": 8.591021666298684e-05,
      "loss": 1.4426,
      "step": 63630
    },
    {
      "epoch": 3.034684234592782,
      "grad_norm": 0.41748547554016113,
      "learning_rate": 8.589870422024909e-05,
      "loss": 1.4496,
      "step": 63640
    },
    {
      "epoch": 3.0352261757582943,
      "grad_norm": 0.39396533370018005,
      "learning_rate": 8.5887187949914e-05,
      "loss": 1.4458,
      "step": 63650
    },
    {
      "epoch": 3.0357681169238067,
      "grad_norm": 0.30107173323631287,
      "learning_rate": 8.58756678534082e-05,
      "loss": 1.4475,
      "step": 63660
    },
    {
      "epoch": 3.0363100580893185,
      "grad_norm": 0.21877449750900269,
      "learning_rate": 8.586414393215869e-05,
      "loss": 1.4491,
      "step": 63670
    },
    {
      "epoch": 3.036851999254831,
      "grad_norm": 0.3609618842601776,
      "learning_rate": 8.585261618759303e-05,
      "loss": 1.4532,
      "step": 63680
    },
    {
      "epoch": 3.037068775721036,
      "eval_loss": 2.4435439109802246,
      "eval_runtime": 21.984,
      "eval_samples_per_second": 227.438,
      "eval_steps_per_second": 1.228,
      "step": 63684
    },
    {
      "epoch": 3.037393940420343,
      "grad_norm": 0.21839331090450287,
      "learning_rate": 8.584108462113922e-05,
      "loss": 1.4429,
      "step": 63690
    },
    {
      "epoch": 3.0379358815858555,
      "grad_norm": 0.24262754619121552,
      "learning_rate": 8.582954923422578e-05,
      "loss": 1.4498,
      "step": 63700
    },
    {
      "epoch": 3.0384778227513674,
      "grad_norm": 0.18923676013946533,
      "learning_rate": 8.581801002828159e-05,
      "loss": 1.4449,
      "step": 63710
    },
    {
      "epoch": 3.0390197639168797,
      "grad_norm": 0.2811407446861267,
      "learning_rate": 8.580646700473614e-05,
      "loss": 1.4465,
      "step": 63720
    },
    {
      "epoch": 3.039561705082392,
      "grad_norm": 0.2581160366535187,
      "learning_rate": 8.579492016501929e-05,
      "loss": 1.4538,
      "step": 63730
    },
    {
      "epoch": 3.0401036462479043,
      "grad_norm": 0.4067544639110565,
      "learning_rate": 8.578336951056145e-05,
      "loss": 1.4401,
      "step": 63740
    },
    {
      "epoch": 3.0406455874134166,
      "grad_norm": 0.2783118784427643,
      "learning_rate": 8.577181504279342e-05,
      "loss": 1.4468,
      "step": 63750
    },
    {
      "epoch": 3.0411875285789285,
      "grad_norm": 0.2964227795600891,
      "learning_rate": 8.576025676314654e-05,
      "loss": 1.4491,
      "step": 63760
    },
    {
      "epoch": 3.041729469744441,
      "grad_norm": 0.2709335684776306,
      "learning_rate": 8.57486946730526e-05,
      "loss": 1.4501,
      "step": 63770
    },
    {
      "epoch": 3.041783663860992,
      "eval_loss": 2.4320621490478516,
      "eval_runtime": 21.9837,
      "eval_samples_per_second": 227.441,
      "eval_steps_per_second": 1.228,
      "step": 63771
    },
    {
      "epoch": 3.042271410909953,
      "grad_norm": 0.31969088315963745,
      "learning_rate": 8.573712877394387e-05,
      "loss": 1.4542,
      "step": 63780
    },
    {
      "epoch": 3.0428133520754654,
      "grad_norm": 0.43009239435195923,
      "learning_rate": 8.572555906725309e-05,
      "loss": 1.4478,
      "step": 63790
    },
    {
      "epoch": 3.0433552932409773,
      "grad_norm": 0.24113118648529053,
      "learning_rate": 8.571398555441344e-05,
      "loss": 1.4368,
      "step": 63800
    },
    {
      "epoch": 3.0438972344064896,
      "grad_norm": 0.30968958139419556,
      "learning_rate": 8.570240823685858e-05,
      "loss": 1.451,
      "step": 63810
    },
    {
      "epoch": 3.044439175572002,
      "grad_norm": 0.2378184050321579,
      "learning_rate": 8.569082711602271e-05,
      "loss": 1.443,
      "step": 63820
    },
    {
      "epoch": 3.0449811167375143,
      "grad_norm": 0.19775545597076416,
      "learning_rate": 8.567924219334042e-05,
      "loss": 1.452,
      "step": 63830
    },
    {
      "epoch": 3.0455230579030266,
      "grad_norm": 0.17699836194515228,
      "learning_rate": 8.566765347024679e-05,
      "loss": 1.452,
      "step": 63840
    },
    {
      "epoch": 3.0460649990685384,
      "grad_norm": 0.1834602653980255,
      "learning_rate": 8.565606094817741e-05,
      "loss": 1.4525,
      "step": 63850
    },
    {
      "epoch": 3.046498552000948,
      "eval_loss": 2.435060977935791,
      "eval_runtime": 21.9853,
      "eval_samples_per_second": 227.424,
      "eval_steps_per_second": 1.228,
      "step": 63858
    },
    {
      "epoch": 3.0466069402340508,
      "grad_norm": 0.2576271891593933,
      "learning_rate": 8.56444646285683e-05,
      "loss": 1.4452,
      "step": 63860
    },
    {
      "epoch": 3.047148881399563,
      "grad_norm": 0.20896773040294647,
      "learning_rate": 8.563286451285595e-05,
      "loss": 1.4548,
      "step": 63870
    },
    {
      "epoch": 3.0476908225650754,
      "grad_norm": 0.2274290919303894,
      "learning_rate": 8.562126060247733e-05,
      "loss": 1.4541,
      "step": 63880
    },
    {
      "epoch": 3.0482327637305877,
      "grad_norm": 0.2640198767185211,
      "learning_rate": 8.560965289886987e-05,
      "loss": 1.4511,
      "step": 63890
    },
    {
      "epoch": 3.0487747048960996,
      "grad_norm": 0.20334435999393463,
      "learning_rate": 8.559804140347156e-05,
      "loss": 1.4497,
      "step": 63900
    },
    {
      "epoch": 3.049316646061612,
      "grad_norm": 0.28822678327560425,
      "learning_rate": 8.558642611772069e-05,
      "loss": 1.4519,
      "step": 63910
    },
    {
      "epoch": 3.049858587227124,
      "grad_norm": 0.30483877658843994,
      "learning_rate": 8.557480704305614e-05,
      "loss": 1.4519,
      "step": 63920
    },
    {
      "epoch": 3.0504005283926365,
      "grad_norm": 0.29864802956581116,
      "learning_rate": 8.556318418091724e-05,
      "loss": 1.449,
      "step": 63930
    },
    {
      "epoch": 3.0509424695581484,
      "grad_norm": 0.43397045135498047,
      "learning_rate": 8.555155753274379e-05,
      "loss": 1.4485,
      "step": 63940
    },
    {
      "epoch": 3.051213440140905,
      "eval_loss": 2.4513673782348633,
      "eval_runtime": 21.9827,
      "eval_samples_per_second": 227.451,
      "eval_steps_per_second": 1.228,
      "step": 63945
    },
    {
      "epoch": 3.0514844107236607,
      "grad_norm": 0.3487168550491333,
      "learning_rate": 8.553992709997602e-05,
      "loss": 1.4545,
      "step": 63950
    },
    {
      "epoch": 3.052026351889173,
      "grad_norm": 0.18993450701236725,
      "learning_rate": 8.552829288405467e-05,
      "loss": 1.4491,
      "step": 63960
    },
    {
      "epoch": 3.0525682930546854,
      "grad_norm": 0.22048941254615784,
      "learning_rate": 8.551665488642096e-05,
      "loss": 1.442,
      "step": 63970
    },
    {
      "epoch": 3.0531102342201977,
      "grad_norm": 0.2803003191947937,
      "learning_rate": 8.55050131085165e-05,
      "loss": 1.4496,
      "step": 63980
    },
    {
      "epoch": 3.0536521753857095,
      "grad_norm": 0.43568509817123413,
      "learning_rate": 8.549336755178347e-05,
      "loss": 1.4516,
      "step": 63990
    },
    {
      "epoch": 3.054194116551222,
      "grad_norm": 0.24292457103729248,
      "learning_rate": 8.548171821766448e-05,
      "loss": 1.4408,
      "step": 64000
    },
    {
      "epoch": 3.054736057716734,
      "grad_norm": 0.1853036731481552,
      "learning_rate": 8.547006510760254e-05,
      "loss": 1.4507,
      "step": 64010
    },
    {
      "epoch": 3.0552779988822465,
      "grad_norm": 0.2446528673171997,
      "learning_rate": 8.545840822304125e-05,
      "loss": 1.4424,
      "step": 64020
    },
    {
      "epoch": 3.0558199400477584,
      "grad_norm": 0.17944113910198212,
      "learning_rate": 8.544674756542457e-05,
      "loss": 1.455,
      "step": 64030
    },
    {
      "epoch": 3.055928328280861,
      "eval_loss": 2.433335304260254,
      "eval_runtime": 21.9868,
      "eval_samples_per_second": 227.409,
      "eval_steps_per_second": 1.228,
      "step": 64032
    },
    {
      "epoch": 3.0563618812132707,
      "grad_norm": 0.3536889851093292,
      "learning_rate": 8.5435083136197e-05,
      "loss": 1.4485,
      "step": 64040
    },
    {
      "epoch": 3.056903822378783,
      "grad_norm": 0.1827295571565628,
      "learning_rate": 8.542341493680345e-05,
      "loss": 1.4464,
      "step": 64050
    },
    {
      "epoch": 3.0574457635442953,
      "grad_norm": 0.1838574856519699,
      "learning_rate": 8.541174296868935e-05,
      "loss": 1.4464,
      "step": 64060
    },
    {
      "epoch": 3.0579877047098076,
      "grad_norm": 0.2586890757083893,
      "learning_rate": 8.540006723330057e-05,
      "loss": 1.4497,
      "step": 64070
    },
    {
      "epoch": 3.0585296458753195,
      "grad_norm": 0.3454444408416748,
      "learning_rate": 8.538838773208344e-05,
      "loss": 1.4462,
      "step": 64080
    },
    {
      "epoch": 3.059071587040832,
      "grad_norm": 0.27531322836875916,
      "learning_rate": 8.537670446648477e-05,
      "loss": 1.4482,
      "step": 64090
    },
    {
      "epoch": 3.059613528206344,
      "grad_norm": 0.2372499257326126,
      "learning_rate": 8.536501743795183e-05,
      "loss": 1.4447,
      "step": 64100
    },
    {
      "epoch": 3.0601554693718565,
      "grad_norm": 0.20991750061511993,
      "learning_rate": 8.535332664793237e-05,
      "loss": 1.4462,
      "step": 64110
    },
    {
      "epoch": 3.0606432164208175,
      "eval_loss": 2.4356119632720947,
      "eval_runtime": 21.9868,
      "eval_samples_per_second": 227.409,
      "eval_steps_per_second": 1.228,
      "step": 64119
    },
    {
      "epoch": 3.0606974105373683,
      "grad_norm": 0.23890291154384613,
      "learning_rate": 8.534163209787459e-05,
      "loss": 1.4582,
      "step": 64120
    },
    {
      "epoch": 3.0612393517028806,
      "grad_norm": 0.25897133350372314,
      "learning_rate": 8.532993378922716e-05,
      "loss": 1.4452,
      "step": 64130
    },
    {
      "epoch": 3.061781292868393,
      "grad_norm": 0.2815474569797516,
      "learning_rate": 8.53182317234392e-05,
      "loss": 1.4509,
      "step": 64140
    },
    {
      "epoch": 3.0623232340339053,
      "grad_norm": 0.22710613906383514,
      "learning_rate": 8.530652590196033e-05,
      "loss": 1.4503,
      "step": 64150
    },
    {
      "epoch": 3.0628651751994176,
      "grad_norm": 0.26313063502311707,
      "learning_rate": 8.529481632624059e-05,
      "loss": 1.4433,
      "step": 64160
    },
    {
      "epoch": 3.0634071163649295,
      "grad_norm": 0.17818008363246918,
      "learning_rate": 8.528310299773055e-05,
      "loss": 1.4435,
      "step": 64170
    },
    {
      "epoch": 3.0639490575304418,
      "grad_norm": 0.1890927255153656,
      "learning_rate": 8.527138591788118e-05,
      "loss": 1.4366,
      "step": 64180
    },
    {
      "epoch": 3.064490998695954,
      "grad_norm": 0.27579453587532043,
      "learning_rate": 8.525966508814396e-05,
      "loss": 1.4499,
      "step": 64190
    },
    {
      "epoch": 3.0650329398614664,
      "grad_norm": 0.17526105046272278,
      "learning_rate": 8.524794050997079e-05,
      "loss": 1.4415,
      "step": 64200
    },
    {
      "epoch": 3.0653581045607736,
      "eval_loss": 2.440138101577759,
      "eval_runtime": 21.9841,
      "eval_samples_per_second": 227.437,
      "eval_steps_per_second": 1.228,
      "step": 64206
    },
    {
      "epoch": 3.0655748810269783,
      "grad_norm": 0.23076623678207397,
      "learning_rate": 8.523621218481407e-05,
      "loss": 1.4393,
      "step": 64210
    },
    {
      "epoch": 3.0661168221924906,
      "grad_norm": 0.29393690824508667,
      "learning_rate": 8.522448011412665e-05,
      "loss": 1.4491,
      "step": 64220
    },
    {
      "epoch": 3.066658763358003,
      "grad_norm": 0.16649694740772247,
      "learning_rate": 8.521274429936187e-05,
      "loss": 1.4364,
      "step": 64230
    },
    {
      "epoch": 3.0672007045235152,
      "grad_norm": 0.23092569410800934,
      "learning_rate": 8.520100474197348e-05,
      "loss": 1.444,
      "step": 64240
    },
    {
      "epoch": 3.0677426456890275,
      "grad_norm": 0.24326132237911224,
      "learning_rate": 8.518926144341577e-05,
      "loss": 1.4569,
      "step": 64250
    },
    {
      "epoch": 3.0682845868545394,
      "grad_norm": 0.27092161774635315,
      "learning_rate": 8.51775144051434e-05,
      "loss": 1.443,
      "step": 64260
    },
    {
      "epoch": 3.0688265280200517,
      "grad_norm": 0.3469485640525818,
      "learning_rate": 8.516576362861159e-05,
      "loss": 1.4355,
      "step": 64270
    },
    {
      "epoch": 3.069368469185564,
      "grad_norm": 0.22915403544902802,
      "learning_rate": 8.515400911527592e-05,
      "loss": 1.4623,
      "step": 64280
    },
    {
      "epoch": 3.0699104103510764,
      "grad_norm": 0.2217090129852295,
      "learning_rate": 8.514225086659253e-05,
      "loss": 1.4463,
      "step": 64290
    },
    {
      "epoch": 3.0700729927007298,
      "eval_loss": 2.4342753887176514,
      "eval_runtime": 21.983,
      "eval_samples_per_second": 227.448,
      "eval_steps_per_second": 1.228,
      "step": 64293
    },
    {
      "epoch": 3.0704523515165887,
      "grad_norm": 0.3960467278957367,
      "learning_rate": 8.513048888401795e-05,
      "loss": 1.4428,
      "step": 64300
    },
    {
      "epoch": 3.0709942926821006,
      "grad_norm": 0.20913895964622498,
      "learning_rate": 8.511872316900925e-05,
      "loss": 1.4455,
      "step": 64310
    },
    {
      "epoch": 3.071536233847613,
      "grad_norm": 0.3125520944595337,
      "learning_rate": 8.510695372302385e-05,
      "loss": 1.4572,
      "step": 64320
    },
    {
      "epoch": 3.072078175013125,
      "grad_norm": 0.3635540306568146,
      "learning_rate": 8.509518054751976e-05,
      "loss": 1.452,
      "step": 64330
    },
    {
      "epoch": 3.0726201161786375,
      "grad_norm": 0.2593584358692169,
      "learning_rate": 8.508340364395536e-05,
      "loss": 1.4375,
      "step": 64340
    },
    {
      "epoch": 3.0731620573441494,
      "grad_norm": 0.22683855891227722,
      "learning_rate": 8.507162301378952e-05,
      "loss": 1.4461,
      "step": 64350
    },
    {
      "epoch": 3.0737039985096617,
      "grad_norm": 0.18193486332893372,
      "learning_rate": 8.505983865848158e-05,
      "loss": 1.4501,
      "step": 64360
    },
    {
      "epoch": 3.074245939675174,
      "grad_norm": 0.35355544090270996,
      "learning_rate": 8.504805057949132e-05,
      "loss": 1.444,
      "step": 64370
    },
    {
      "epoch": 3.0747878808406863,
      "grad_norm": 0.16228243708610535,
      "learning_rate": 8.503625877827904e-05,
      "loss": 1.4489,
      "step": 64380
    },
    {
      "epoch": 3.0747878808406863,
      "eval_loss": 2.4377474784851074,
      "eval_runtime": 21.9803,
      "eval_samples_per_second": 227.477,
      "eval_steps_per_second": 1.228,
      "step": 64380
    },
    {
      "epoch": 3.0753298220061986,
      "grad_norm": 0.26819851994514465,
      "learning_rate": 8.50244632563054e-05,
      "loss": 1.4487,
      "step": 64390
    },
    {
      "epoch": 3.0758717631717105,
      "grad_norm": 0.21382486820220947,
      "learning_rate": 8.501266401503164e-05,
      "loss": 1.4461,
      "step": 64400
    },
    {
      "epoch": 3.076413704337223,
      "grad_norm": 0.20011401176452637,
      "learning_rate": 8.500086105591935e-05,
      "loss": 1.4511,
      "step": 64410
    },
    {
      "epoch": 3.076955645502735,
      "grad_norm": 0.2521159052848816,
      "learning_rate": 8.498905438043061e-05,
      "loss": 1.4421,
      "step": 64420
    },
    {
      "epoch": 3.0774975866682475,
      "grad_norm": 0.27458107471466064,
      "learning_rate": 8.497724399002805e-05,
      "loss": 1.4556,
      "step": 64430
    },
    {
      "epoch": 3.0780395278337593,
      "grad_norm": 0.17625468969345093,
      "learning_rate": 8.496542988617463e-05,
      "loss": 1.4562,
      "step": 64440
    },
    {
      "epoch": 3.0785814689992717,
      "grad_norm": 0.23053233325481415,
      "learning_rate": 8.495361207033387e-05,
      "loss": 1.4438,
      "step": 64450
    },
    {
      "epoch": 3.079123410164784,
      "grad_norm": 0.20668716728687286,
      "learning_rate": 8.494179054396968e-05,
      "loss": 1.4421,
      "step": 64460
    },
    {
      "epoch": 3.0795027689806425,
      "eval_loss": 2.4462552070617676,
      "eval_runtime": 208.061,
      "eval_samples_per_second": 24.031,
      "eval_steps_per_second": 0.13,
      "step": 64467
    },
    {
      "epoch": 3.0796653513302963,
      "grad_norm": 0.2157292515039444,
      "learning_rate": 8.492996530854646e-05,
      "loss": 1.441,
      "step": 64470
    },
    {
      "epoch": 3.0802072924958086,
      "grad_norm": 0.1837170124053955,
      "learning_rate": 8.491813636552911e-05,
      "loss": 1.4421,
      "step": 64480
    },
    {
      "epoch": 3.0807492336613205,
      "grad_norm": 0.2104165405035019,
      "learning_rate": 8.490630371638291e-05,
      "loss": 1.4516,
      "step": 64490
    },
    {
      "epoch": 3.081291174826833,
      "grad_norm": 0.24644435942173004,
      "learning_rate": 8.489446736257365e-05,
      "loss": 1.4451,
      "step": 64500
    },
    {
      "epoch": 3.081833115992345,
      "grad_norm": 0.18532828986644745,
      "learning_rate": 8.488262730556754e-05,
      "loss": 1.4551,
      "step": 64510
    },
    {
      "epoch": 3.0823750571578574,
      "grad_norm": 0.20310962200164795,
      "learning_rate": 8.487078354683132e-05,
      "loss": 1.4506,
      "step": 64520
    },
    {
      "epoch": 3.0829169983233697,
      "grad_norm": 0.22750219702720642,
      "learning_rate": 8.48589360878321e-05,
      "loss": 1.4364,
      "step": 64530
    },
    {
      "epoch": 3.0834589394888816,
      "grad_norm": 0.23471009731292725,
      "learning_rate": 8.484708493003753e-05,
      "loss": 1.455,
      "step": 64540
    },
    {
      "epoch": 3.084000880654394,
      "grad_norm": 0.17168253660202026,
      "learning_rate": 8.483523007491565e-05,
      "loss": 1.4408,
      "step": 64550
    },
    {
      "epoch": 3.084217657120599,
      "eval_loss": 2.4336390495300293,
      "eval_runtime": 21.9861,
      "eval_samples_per_second": 227.416,
      "eval_steps_per_second": 1.228,
      "step": 64554
    },
    {
      "epoch": 3.0845428218199062,
      "grad_norm": 0.21957682073116302,
      "learning_rate": 8.4823371523935e-05,
      "loss": 1.4431,
      "step": 64560
    },
    {
      "epoch": 3.0850847629854186,
      "grad_norm": 0.25939157605171204,
      "learning_rate": 8.481150927856458e-05,
      "loss": 1.4492,
      "step": 64570
    },
    {
      "epoch": 3.0856267041509304,
      "grad_norm": 0.22554215788841248,
      "learning_rate": 8.479964334027381e-05,
      "loss": 1.4551,
      "step": 64580
    },
    {
      "epoch": 3.0861686453164427,
      "grad_norm": 0.20122438669204712,
      "learning_rate": 8.478777371053259e-05,
      "loss": 1.4445,
      "step": 64590
    },
    {
      "epoch": 3.086710586481955,
      "grad_norm": 0.18250145018100739,
      "learning_rate": 8.477590039081131e-05,
      "loss": 1.4475,
      "step": 64600
    },
    {
      "epoch": 3.0872525276474674,
      "grad_norm": 0.18288695812225342,
      "learning_rate": 8.476402338258073e-05,
      "loss": 1.4484,
      "step": 64610
    },
    {
      "epoch": 3.0877944688129797,
      "grad_norm": 0.19832782447338104,
      "learning_rate": 8.475214268731219e-05,
      "loss": 1.4513,
      "step": 64620
    },
    {
      "epoch": 3.0883364099784916,
      "grad_norm": 0.38973701000213623,
      "learning_rate": 8.474025830647737e-05,
      "loss": 1.442,
      "step": 64630
    },
    {
      "epoch": 3.088878351144004,
      "grad_norm": 0.3162194490432739,
      "learning_rate": 8.472837024154847e-05,
      "loss": 1.4464,
      "step": 64640
    },
    {
      "epoch": 3.088932545260555,
      "eval_loss": 2.423926830291748,
      "eval_runtime": 21.9851,
      "eval_samples_per_second": 227.426,
      "eval_steps_per_second": 1.228,
      "step": 64641
    },
    {
      "epoch": 3.089420292309516,
      "grad_norm": 0.1702577918767929,
      "learning_rate": 8.471647849399815e-05,
      "loss": 1.4542,
      "step": 64650
    },
    {
      "epoch": 3.0899622334750285,
      "grad_norm": 0.315075546503067,
      "learning_rate": 8.470458306529946e-05,
      "loss": 1.4416,
      "step": 64660
    },
    {
      "epoch": 3.0905041746405404,
      "grad_norm": 0.25835493206977844,
      "learning_rate": 8.4692683956926e-05,
      "loss": 1.436,
      "step": 64670
    },
    {
      "epoch": 3.0910461158060527,
      "grad_norm": 0.6356673240661621,
      "learning_rate": 8.468078117035176e-05,
      "loss": 1.4472,
      "step": 64680
    },
    {
      "epoch": 3.091588056971565,
      "grad_norm": 0.3583969175815582,
      "learning_rate": 8.466887470705121e-05,
      "loss": 1.4487,
      "step": 64690
    },
    {
      "epoch": 3.0921299981370773,
      "grad_norm": 0.21938195824623108,
      "learning_rate": 8.465696456849928e-05,
      "loss": 1.4513,
      "step": 64700
    },
    {
      "epoch": 3.0926719393025897,
      "grad_norm": 0.26139944791793823,
      "learning_rate": 8.464505075617133e-05,
      "loss": 1.4505,
      "step": 64710
    },
    {
      "epoch": 3.0932138804681015,
      "grad_norm": 0.2597460150718689,
      "learning_rate": 8.46331332715432e-05,
      "loss": 1.4483,
      "step": 64720
    },
    {
      "epoch": 3.0936474334005113,
      "eval_loss": 2.4286162853240967,
      "eval_runtime": 21.6785,
      "eval_samples_per_second": 230.643,
      "eval_steps_per_second": 1.245,
      "step": 64728
    },
    {
      "epoch": 3.093755821633614,
      "grad_norm": 0.21105916798114777,
      "learning_rate": 8.462121211609117e-05,
      "loss": 1.4437,
      "step": 64730
    },
    {
      "epoch": 3.094297762799126,
      "grad_norm": 0.1735154688358307,
      "learning_rate": 8.4609287291292e-05,
      "loss": 1.4413,
      "step": 64740
    },
    {
      "epoch": 3.0948397039646385,
      "grad_norm": 0.1659323126077652,
      "learning_rate": 8.459735879862286e-05,
      "loss": 1.4488,
      "step": 64750
    },
    {
      "epoch": 3.0953816451301503,
      "grad_norm": 0.19731490314006805,
      "learning_rate": 8.45854266395614e-05,
      "loss": 1.4478,
      "step": 64760
    },
    {
      "epoch": 3.0959235862956627,
      "grad_norm": 0.19939690828323364,
      "learning_rate": 8.457349081558576e-05,
      "loss": 1.4378,
      "step": 64770
    },
    {
      "epoch": 3.096465527461175,
      "grad_norm": 0.24715344607830048,
      "learning_rate": 8.456155132817443e-05,
      "loss": 1.4477,
      "step": 64780
    },
    {
      "epoch": 3.0970074686266873,
      "grad_norm": 0.20696981251239777,
      "learning_rate": 8.45496081788065e-05,
      "loss": 1.4452,
      "step": 64790
    },
    {
      "epoch": 3.0975494097921996,
      "grad_norm": 0.19315387308597565,
      "learning_rate": 8.45376613689614e-05,
      "loss": 1.4469,
      "step": 64800
    },
    {
      "epoch": 3.0980913509577115,
      "grad_norm": 0.23861078917980194,
      "learning_rate": 8.452571090011905e-05,
      "loss": 1.4624,
      "step": 64810
    },
    {
      "epoch": 3.098362321540468,
      "eval_loss": 2.4268784523010254,
      "eval_runtime": 22.037,
      "eval_samples_per_second": 226.891,
      "eval_steps_per_second": 1.225,
      "step": 64815
    },
    {
      "epoch": 3.098633292123224,
      "grad_norm": 0.1983337700366974,
      "learning_rate": 8.45137567737598e-05,
      "loss": 1.4448,
      "step": 64820
    },
    {
      "epoch": 3.099175233288736,
      "grad_norm": 0.2253299057483673,
      "learning_rate": 8.450179899136451e-05,
      "loss": 1.4471,
      "step": 64830
    },
    {
      "epoch": 3.0997171744542484,
      "grad_norm": 0.2381051778793335,
      "learning_rate": 8.448983755441447e-05,
      "loss": 1.4461,
      "step": 64840
    },
    {
      "epoch": 3.1002591156197603,
      "grad_norm": 0.23918847739696503,
      "learning_rate": 8.447787246439135e-05,
      "loss": 1.4571,
      "step": 64850
    },
    {
      "epoch": 3.1008010567852726,
      "grad_norm": 0.17684470117092133,
      "learning_rate": 8.446590372277738e-05,
      "loss": 1.4455,
      "step": 64860
    },
    {
      "epoch": 3.101342997950785,
      "grad_norm": 0.1668468713760376,
      "learning_rate": 8.445393133105519e-05,
      "loss": 1.4437,
      "step": 64870
    },
    {
      "epoch": 3.1018849391162973,
      "grad_norm": 0.18969334661960602,
      "learning_rate": 8.444195529070785e-05,
      "loss": 1.4375,
      "step": 64880
    },
    {
      "epoch": 3.1024268802818096,
      "grad_norm": 0.1837807595729828,
      "learning_rate": 8.442997560321894e-05,
      "loss": 1.4407,
      "step": 64890
    },
    {
      "epoch": 3.1029688214473214,
      "grad_norm": 0.22503748536109924,
      "learning_rate": 8.44179922700724e-05,
      "loss": 1.4465,
      "step": 64900
    },
    {
      "epoch": 3.103077209680424,
      "eval_loss": 2.4285776615142822,
      "eval_runtime": 21.9808,
      "eval_samples_per_second": 227.472,
      "eval_steps_per_second": 1.228,
      "step": 64902
    },
    {
      "epoch": 3.1035107626128338,
      "grad_norm": 0.20925331115722656,
      "learning_rate": 8.44060052927527e-05,
      "loss": 1.4417,
      "step": 64910
    },
    {
      "epoch": 3.104052703778346,
      "grad_norm": 0.36361682415008545,
      "learning_rate": 8.439401467274474e-05,
      "loss": 1.4505,
      "step": 64920
    },
    {
      "epoch": 3.1045946449438584,
      "grad_norm": 0.2141452133655548,
      "learning_rate": 8.438202041153385e-05,
      "loss": 1.4502,
      "step": 64930
    },
    {
      "epoch": 3.1051365861093707,
      "grad_norm": 0.22404509782791138,
      "learning_rate": 8.437002251060585e-05,
      "loss": 1.4467,
      "step": 64940
    },
    {
      "epoch": 3.1056785272748826,
      "grad_norm": 0.26203832030296326,
      "learning_rate": 8.435802097144696e-05,
      "loss": 1.4514,
      "step": 64950
    },
    {
      "epoch": 3.106220468440395,
      "grad_norm": 0.265337198972702,
      "learning_rate": 8.434601579554389e-05,
      "loss": 1.4438,
      "step": 64960
    },
    {
      "epoch": 3.106762409605907,
      "grad_norm": 0.1931227445602417,
      "learning_rate": 8.433400698438381e-05,
      "loss": 1.4518,
      "step": 64970
    },
    {
      "epoch": 3.1073043507714195,
      "grad_norm": 0.2169187366962433,
      "learning_rate": 8.432199453945427e-05,
      "loss": 1.4367,
      "step": 64980
    },
    {
      "epoch": 3.1077920978203806,
      "eval_loss": 2.420952796936035,
      "eval_runtime": 21.9789,
      "eval_samples_per_second": 227.491,
      "eval_steps_per_second": 1.228,
      "step": 64989
    },
    {
      "epoch": 3.1078462919369314,
      "grad_norm": 0.2237025499343872,
      "learning_rate": 8.430997846224338e-05,
      "loss": 1.4437,
      "step": 64990
    },
    {
      "epoch": 3.1083882331024437,
      "grad_norm": 0.18218779563903809,
      "learning_rate": 8.42979587542396e-05,
      "loss": 1.442,
      "step": 65000
    },
    {
      "epoch": 3.108930174267956,
      "grad_norm": 0.2928388714790344,
      "learning_rate": 8.428593541693188e-05,
      "loss": 1.4455,
      "step": 65010
    },
    {
      "epoch": 3.1094721154334684,
      "grad_norm": 0.22536158561706543,
      "learning_rate": 8.427390845180963e-05,
      "loss": 1.4388,
      "step": 65020
    },
    {
      "epoch": 3.1100140565989807,
      "grad_norm": 0.3034621477127075,
      "learning_rate": 8.426187786036269e-05,
      "loss": 1.4458,
      "step": 65030
    },
    {
      "epoch": 3.1105559977644925,
      "grad_norm": 0.21954983472824097,
      "learning_rate": 8.424984364408138e-05,
      "loss": 1.4349,
      "step": 65040
    },
    {
      "epoch": 3.111097938930005,
      "grad_norm": 0.18350431323051453,
      "learning_rate": 8.423780580445642e-05,
      "loss": 1.4489,
      "step": 65050
    },
    {
      "epoch": 3.111639880095517,
      "grad_norm": 0.19277605414390564,
      "learning_rate": 8.4225764342979e-05,
      "loss": 1.4507,
      "step": 65060
    },
    {
      "epoch": 3.1121818212610295,
      "grad_norm": 0.25510692596435547,
      "learning_rate": 8.42137192611408e-05,
      "loss": 1.4467,
      "step": 65070
    },
    {
      "epoch": 3.1125069859603367,
      "eval_loss": 2.427020788192749,
      "eval_runtime": 21.9917,
      "eval_samples_per_second": 227.358,
      "eval_steps_per_second": 1.228,
      "step": 65076
    },
    {
      "epoch": 3.1127237624265414,
      "grad_norm": 0.2290118932723999,
      "learning_rate": 8.42016705604339e-05,
      "loss": 1.4456,
      "step": 65080
    },
    {
      "epoch": 3.1132657035920537,
      "grad_norm": 0.19887521862983704,
      "learning_rate": 8.41896182423508e-05,
      "loss": 1.4479,
      "step": 65090
    },
    {
      "epoch": 3.113807644757566,
      "grad_norm": 0.17075598239898682,
      "learning_rate": 8.417756230838455e-05,
      "loss": 1.443,
      "step": 65100
    },
    {
      "epoch": 3.1143495859230783,
      "grad_norm": 0.2500585913658142,
      "learning_rate": 8.416550276002853e-05,
      "loss": 1.44,
      "step": 65110
    },
    {
      "epoch": 3.1148915270885906,
      "grad_norm": 0.2130533754825592,
      "learning_rate": 8.415343959877668e-05,
      "loss": 1.4478,
      "step": 65120
    },
    {
      "epoch": 3.1154334682541025,
      "grad_norm": 0.3005780279636383,
      "learning_rate": 8.41413728261233e-05,
      "loss": 1.4438,
      "step": 65130
    },
    {
      "epoch": 3.115975409419615,
      "grad_norm": 0.23416976630687714,
      "learning_rate": 8.412930244356316e-05,
      "loss": 1.4435,
      "step": 65140
    },
    {
      "epoch": 3.116517350585127,
      "grad_norm": 0.26363202929496765,
      "learning_rate": 8.41172284525915e-05,
      "loss": 1.4469,
      "step": 65150
    },
    {
      "epoch": 3.1170592917506394,
      "grad_norm": 0.20009048283100128,
      "learning_rate": 8.4105150854704e-05,
      "loss": 1.4412,
      "step": 65160
    },
    {
      "epoch": 3.117221874100293,
      "eval_loss": 2.424807071685791,
      "eval_runtime": 21.9875,
      "eval_samples_per_second": 227.402,
      "eval_steps_per_second": 1.228,
      "step": 65163
    },
    {
      "epoch": 3.1176012329161518,
      "grad_norm": 0.2043033242225647,
      "learning_rate": 8.409306965139677e-05,
      "loss": 1.4525,
      "step": 65170
    },
    {
      "epoch": 3.1181431740816636,
      "grad_norm": 0.27456438541412354,
      "learning_rate": 8.408098484416639e-05,
      "loss": 1.4432,
      "step": 65180
    },
    {
      "epoch": 3.118685115247176,
      "grad_norm": 0.4368920624256134,
      "learning_rate": 8.406889643450984e-05,
      "loss": 1.4449,
      "step": 65190
    },
    {
      "epoch": 3.1192270564126883,
      "grad_norm": 0.3747996389865875,
      "learning_rate": 8.405680442392464e-05,
      "loss": 1.4441,
      "step": 65200
    },
    {
      "epoch": 3.1197689975782006,
      "grad_norm": 0.3094594478607178,
      "learning_rate": 8.404470881390863e-05,
      "loss": 1.4518,
      "step": 65210
    },
    {
      "epoch": 3.1203109387437125,
      "grad_norm": 0.30118393898010254,
      "learning_rate": 8.40326096059602e-05,
      "loss": 1.4454,
      "step": 65220
    },
    {
      "epoch": 3.1208528799092248,
      "grad_norm": 0.20605798065662384,
      "learning_rate": 8.402050680157816e-05,
      "loss": 1.4495,
      "step": 65230
    },
    {
      "epoch": 3.121394821074737,
      "grad_norm": 0.21210525929927826,
      "learning_rate": 8.400840040226172e-05,
      "loss": 1.4417,
      "step": 65240
    },
    {
      "epoch": 3.1219367622402494,
      "grad_norm": 0.17496547102928162,
      "learning_rate": 8.399629040951057e-05,
      "loss": 1.4494,
      "step": 65250
    },
    {
      "epoch": 3.1219367622402494,
      "eval_loss": 2.42055082321167,
      "eval_runtime": 21.5444,
      "eval_samples_per_second": 232.079,
      "eval_steps_per_second": 1.253,
      "step": 65250
    },
    {
      "epoch": 3.1224787034057617,
      "grad_norm": 0.26712679862976074,
      "learning_rate": 8.398417682482486e-05,
      "loss": 1.4512,
      "step": 65260
    },
    {
      "epoch": 3.1230206445712736,
      "grad_norm": 0.2958221733570099,
      "learning_rate": 8.397205964970515e-05,
      "loss": 1.4454,
      "step": 65270
    },
    {
      "epoch": 3.123562585736786,
      "grad_norm": 0.2679630219936371,
      "learning_rate": 8.39599388856525e-05,
      "loss": 1.4397,
      "step": 65280
    },
    {
      "epoch": 3.1241045269022982,
      "grad_norm": 0.16526180505752563,
      "learning_rate": 8.394781453416832e-05,
      "loss": 1.4498,
      "step": 65290
    },
    {
      "epoch": 3.1246464680678105,
      "grad_norm": 0.20681805908679962,
      "learning_rate": 8.393568659675458e-05,
      "loss": 1.4472,
      "step": 65300
    },
    {
      "epoch": 3.1251884092333224,
      "grad_norm": 0.19305351376533508,
      "learning_rate": 8.392355507491361e-05,
      "loss": 1.4432,
      "step": 65310
    },
    {
      "epoch": 3.1257303503988347,
      "grad_norm": 0.21879535913467407,
      "learning_rate": 8.391141997014819e-05,
      "loss": 1.4487,
      "step": 65320
    },
    {
      "epoch": 3.126272291564347,
      "grad_norm": 0.22270238399505615,
      "learning_rate": 8.389928128396161e-05,
      "loss": 1.4372,
      "step": 65330
    },
    {
      "epoch": 3.1266516503802055,
      "eval_loss": 2.4269917011260986,
      "eval_runtime": 21.979,
      "eval_samples_per_second": 227.49,
      "eval_steps_per_second": 1.228,
      "step": 65337
    },
    {
      "epoch": 3.1268142327298594,
      "grad_norm": 0.2375665158033371,
      "learning_rate": 8.388713901785753e-05,
      "loss": 1.4447,
      "step": 65340
    },
    {
      "epoch": 3.1273561738953717,
      "grad_norm": 0.30421313643455505,
      "learning_rate": 8.387499317334007e-05,
      "loss": 1.4416,
      "step": 65350
    },
    {
      "epoch": 3.1278981150608836,
      "grad_norm": 0.3451899290084839,
      "learning_rate": 8.386284375191381e-05,
      "loss": 1.4467,
      "step": 65360
    },
    {
      "epoch": 3.128440056226396,
      "grad_norm": 0.26824066042900085,
      "learning_rate": 8.385069075508379e-05,
      "loss": 1.4496,
      "step": 65370
    },
    {
      "epoch": 3.128981997391908,
      "grad_norm": 0.17959991097450256,
      "learning_rate": 8.383853418435546e-05,
      "loss": 1.4531,
      "step": 65380
    },
    {
      "epoch": 3.1295239385574205,
      "grad_norm": 0.18542522192001343,
      "learning_rate": 8.38263740412347e-05,
      "loss": 1.4485,
      "step": 65390
    },
    {
      "epoch": 3.1300658797229324,
      "grad_norm": 0.1759880632162094,
      "learning_rate": 8.38142103272279e-05,
      "loss": 1.4377,
      "step": 65400
    },
    {
      "epoch": 3.1306078208884447,
      "grad_norm": 0.24404051899909973,
      "learning_rate": 8.380204304384181e-05,
      "loss": 1.4517,
      "step": 65410
    },
    {
      "epoch": 3.131149762053957,
      "grad_norm": 0.18275314569473267,
      "learning_rate": 8.37898721925837e-05,
      "loss": 1.4604,
      "step": 65420
    },
    {
      "epoch": 3.131366538520162,
      "eval_loss": 2.424156665802002,
      "eval_runtime": 21.9886,
      "eval_samples_per_second": 227.39,
      "eval_steps_per_second": 1.228,
      "step": 65424
    },
    {
      "epoch": 3.1316917032194693,
      "grad_norm": 0.22716642916202545,
      "learning_rate": 8.377769777496118e-05,
      "loss": 1.4573,
      "step": 65430
    },
    {
      "epoch": 3.1322336443849816,
      "grad_norm": 0.2442425787448883,
      "learning_rate": 8.376551979248242e-05,
      "loss": 1.4396,
      "step": 65440
    },
    {
      "epoch": 3.1327755855504935,
      "grad_norm": 0.25765150785446167,
      "learning_rate": 8.375333824665594e-05,
      "loss": 1.4532,
      "step": 65450
    },
    {
      "epoch": 3.133317526716006,
      "grad_norm": 0.22376912832260132,
      "learning_rate": 8.374115313899077e-05,
      "loss": 1.4431,
      "step": 65460
    },
    {
      "epoch": 3.133859467881518,
      "grad_norm": 0.36618417501449585,
      "learning_rate": 8.372896447099634e-05,
      "loss": 1.4448,
      "step": 65470
    },
    {
      "epoch": 3.1344014090470305,
      "grad_norm": 0.35767969489097595,
      "learning_rate": 8.371677224418248e-05,
      "loss": 1.4337,
      "step": 65480
    },
    {
      "epoch": 3.1349433502125423,
      "grad_norm": 0.48897460103034973,
      "learning_rate": 8.370457646005957e-05,
      "loss": 1.4542,
      "step": 65490
    },
    {
      "epoch": 3.1354852913780547,
      "grad_norm": 0.39159682393074036,
      "learning_rate": 8.369237712013835e-05,
      "loss": 1.4522,
      "step": 65500
    },
    {
      "epoch": 3.136027232543567,
      "grad_norm": 0.36611253023147583,
      "learning_rate": 8.368017422593003e-05,
      "loss": 1.441,
      "step": 65510
    },
    {
      "epoch": 3.1360814266601182,
      "eval_loss": 2.427338123321533,
      "eval_runtime": 22.8212,
      "eval_samples_per_second": 219.095,
      "eval_steps_per_second": 1.183,
      "step": 65511
    },
    {
      "epoch": 3.1365691737090793,
      "grad_norm": 0.26878681778907776,
      "learning_rate": 8.366796777894624e-05,
      "loss": 1.4446,
      "step": 65520
    },
    {
      "epoch": 3.1371111148745916,
      "grad_norm": 0.2367294430732727,
      "learning_rate": 8.365575778069907e-05,
      "loss": 1.4538,
      "step": 65530
    },
    {
      "epoch": 3.1376530560401035,
      "grad_norm": 0.29069700837135315,
      "learning_rate": 8.364354423270102e-05,
      "loss": 1.4405,
      "step": 65540
    },
    {
      "epoch": 3.138194997205616,
      "grad_norm": 0.22299885749816895,
      "learning_rate": 8.363132713646509e-05,
      "loss": 1.4406,
      "step": 65550
    },
    {
      "epoch": 3.138736938371128,
      "grad_norm": 0.2363450676202774,
      "learning_rate": 8.361910649350465e-05,
      "loss": 1.4323,
      "step": 65560
    },
    {
      "epoch": 3.1392788795366404,
      "grad_norm": 0.2508092224597931,
      "learning_rate": 8.360688230533356e-05,
      "loss": 1.4348,
      "step": 65570
    },
    {
      "epoch": 3.1398208207021527,
      "grad_norm": 0.20643532276153564,
      "learning_rate": 8.359465457346607e-05,
      "loss": 1.4358,
      "step": 65580
    },
    {
      "epoch": 3.1403627618676646,
      "grad_norm": 0.22416460514068604,
      "learning_rate": 8.358242329941692e-05,
      "loss": 1.4501,
      "step": 65590
    },
    {
      "epoch": 3.1407963148000744,
      "eval_loss": 2.4230735301971436,
      "eval_runtime": 23.1943,
      "eval_samples_per_second": 215.571,
      "eval_steps_per_second": 1.164,
      "step": 65598
    },
    {
      "epoch": 3.140904703033177,
      "grad_norm": 0.3181062936782837,
      "learning_rate": 8.357018848470128e-05,
      "loss": 1.4362,
      "step": 65600
    },
    {
      "epoch": 3.1414466441986892,
      "grad_norm": 0.23214562237262726,
      "learning_rate": 8.35579501308347e-05,
      "loss": 1.4405,
      "step": 65610
    },
    {
      "epoch": 3.1419885853642016,
      "grad_norm": 0.35660114884376526,
      "learning_rate": 8.354570823933327e-05,
      "loss": 1.4464,
      "step": 65620
    },
    {
      "epoch": 3.1425305265297134,
      "grad_norm": 0.38225075602531433,
      "learning_rate": 8.353346281171343e-05,
      "loss": 1.4411,
      "step": 65630
    },
    {
      "epoch": 3.1430724676952257,
      "grad_norm": 0.2181611806154251,
      "learning_rate": 8.35212138494921e-05,
      "loss": 1.442,
      "step": 65640
    },
    {
      "epoch": 3.143614408860738,
      "grad_norm": 0.23916418850421906,
      "learning_rate": 8.35089613541866e-05,
      "loss": 1.4485,
      "step": 65650
    },
    {
      "epoch": 3.1441563500262504,
      "grad_norm": 0.19111785292625427,
      "learning_rate": 8.349670532731478e-05,
      "loss": 1.4476,
      "step": 65660
    },
    {
      "epoch": 3.1446982911917627,
      "grad_norm": 0.31324857473373413,
      "learning_rate": 8.34844457703948e-05,
      "loss": 1.4374,
      "step": 65670
    },
    {
      "epoch": 3.1452402323572746,
      "grad_norm": 0.3937076926231384,
      "learning_rate": 8.347218268494535e-05,
      "loss": 1.4406,
      "step": 65680
    },
    {
      "epoch": 3.145511202940031,
      "eval_loss": 2.4279704093933105,
      "eval_runtime": 22.8461,
      "eval_samples_per_second": 218.856,
      "eval_steps_per_second": 1.182,
      "step": 65685
    },
    {
      "epoch": 3.145782173522787,
      "grad_norm": 0.27373912930488586,
      "learning_rate": 8.345991607248553e-05,
      "loss": 1.4465,
      "step": 65690
    },
    {
      "epoch": 3.146324114688299,
      "grad_norm": 0.16764716804027557,
      "learning_rate": 8.344764593453485e-05,
      "loss": 1.4447,
      "step": 65700
    },
    {
      "epoch": 3.1468660558538115,
      "grad_norm": 0.21356230974197388,
      "learning_rate": 8.343537227261332e-05,
      "loss": 1.4484,
      "step": 65710
    },
    {
      "epoch": 3.1474079970193234,
      "grad_norm": 0.17575284838676453,
      "learning_rate": 8.342309508824132e-05,
      "loss": 1.4387,
      "step": 65720
    },
    {
      "epoch": 3.1479499381848357,
      "grad_norm": 0.20323576033115387,
      "learning_rate": 8.34108143829397e-05,
      "loss": 1.4325,
      "step": 65730
    },
    {
      "epoch": 3.148491879350348,
      "grad_norm": 0.30600157380104065,
      "learning_rate": 8.339853015822974e-05,
      "loss": 1.4434,
      "step": 65740
    },
    {
      "epoch": 3.1490338205158603,
      "grad_norm": 0.22637711465358734,
      "learning_rate": 8.338624241563316e-05,
      "loss": 1.4399,
      "step": 65750
    },
    {
      "epoch": 3.1495757616813727,
      "grad_norm": 0.17684820294380188,
      "learning_rate": 8.33739511566721e-05,
      "loss": 1.4412,
      "step": 65760
    },
    {
      "epoch": 3.1501177028468845,
      "grad_norm": 0.20868733525276184,
      "learning_rate": 8.336165638286916e-05,
      "loss": 1.4387,
      "step": 65770
    },
    {
      "epoch": 3.150226091079987,
      "eval_loss": 2.425462245941162,
      "eval_runtime": 22.0377,
      "eval_samples_per_second": 226.884,
      "eval_steps_per_second": 1.225,
      "step": 65772
    },
    {
      "epoch": 3.150659644012397,
      "grad_norm": 0.23098771274089813,
      "learning_rate": 8.334935809574738e-05,
      "loss": 1.4372,
      "step": 65780
    },
    {
      "epoch": 3.151201585177909,
      "grad_norm": 0.17905355989933014,
      "learning_rate": 8.33370562968302e-05,
      "loss": 1.4408,
      "step": 65790
    },
    {
      "epoch": 3.1517435263434215,
      "grad_norm": 0.21767891943454742,
      "learning_rate": 8.332475098764149e-05,
      "loss": 1.4326,
      "step": 65800
    },
    {
      "epoch": 3.152285467508934,
      "grad_norm": 0.2932087779045105,
      "learning_rate": 8.331244216970561e-05,
      "loss": 1.4417,
      "step": 65810
    },
    {
      "epoch": 3.1528274086744457,
      "grad_norm": 0.2508069574832916,
      "learning_rate": 8.330012984454732e-05,
      "loss": 1.4398,
      "step": 65820
    },
    {
      "epoch": 3.153369349839958,
      "grad_norm": 0.20580008625984192,
      "learning_rate": 8.32878140136918e-05,
      "loss": 1.4417,
      "step": 65830
    },
    {
      "epoch": 3.1539112910054703,
      "grad_norm": 0.222931370139122,
      "learning_rate": 8.327549467866472e-05,
      "loss": 1.4447,
      "step": 65840
    },
    {
      "epoch": 3.1544532321709826,
      "grad_norm": 0.17937925457954407,
      "learning_rate": 8.32631718409921e-05,
      "loss": 1.4547,
      "step": 65850
    },
    {
      "epoch": 3.154940979219943,
      "eval_loss": 2.4275054931640625,
      "eval_runtime": 21.995,
      "eval_samples_per_second": 227.324,
      "eval_steps_per_second": 1.228,
      "step": 65859
    },
    {
      "epoch": 3.1549951733364945,
      "grad_norm": 0.27297332882881165,
      "learning_rate": 8.325084550220046e-05,
      "loss": 1.4369,
      "step": 65860
    },
    {
      "epoch": 3.155537114502007,
      "grad_norm": 0.2378816157579422,
      "learning_rate": 8.323851566381672e-05,
      "loss": 1.4453,
      "step": 65870
    },
    {
      "epoch": 3.156079055667519,
      "grad_norm": 0.30926477909088135,
      "learning_rate": 8.322618232736827e-05,
      "loss": 1.4455,
      "step": 65880
    },
    {
      "epoch": 3.1566209968330314,
      "grad_norm": 0.19149400293827057,
      "learning_rate": 8.32138454943829e-05,
      "loss": 1.4509,
      "step": 65890
    },
    {
      "epoch": 3.1571629379985433,
      "grad_norm": 0.18199436366558075,
      "learning_rate": 8.320150516638884e-05,
      "loss": 1.4574,
      "step": 65900
    },
    {
      "epoch": 3.1577048791640556,
      "grad_norm": 0.3855782747268677,
      "learning_rate": 8.318916134491477e-05,
      "loss": 1.4557,
      "step": 65910
    },
    {
      "epoch": 3.158246820329568,
      "grad_norm": 0.29194483160972595,
      "learning_rate": 8.317681403148978e-05,
      "loss": 1.4476,
      "step": 65920
    },
    {
      "epoch": 3.1587887614950803,
      "grad_norm": 0.3535638451576233,
      "learning_rate": 8.316446322764338e-05,
      "loss": 1.4398,
      "step": 65930
    },
    {
      "epoch": 3.1593307026605926,
      "grad_norm": 0.36056360602378845,
      "learning_rate": 8.315210893490556e-05,
      "loss": 1.44,
      "step": 65940
    },
    {
      "epoch": 3.1596558673599,
      "eval_loss": 2.4279699325561523,
      "eval_runtime": 21.9818,
      "eval_samples_per_second": 227.461,
      "eval_steps_per_second": 1.228,
      "step": 65946
    },
    {
      "epoch": 3.1598726438261044,
      "grad_norm": 0.17839354276657104,
      "learning_rate": 8.313975115480671e-05,
      "loss": 1.4455,
      "step": 65950
    },
    {
      "epoch": 3.1604145849916168,
      "grad_norm": 0.21037691831588745,
      "learning_rate": 8.312738988887766e-05,
      "loss": 1.4449,
      "step": 65960
    },
    {
      "epoch": 3.160956526157129,
      "grad_norm": 0.25134557485580444,
      "learning_rate": 8.311502513864966e-05,
      "loss": 1.4403,
      "step": 65970
    },
    {
      "epoch": 3.1614984673226414,
      "grad_norm": 0.17751184105873108,
      "learning_rate": 8.310265690565443e-05,
      "loss": 1.4473,
      "step": 65980
    },
    {
      "epoch": 3.1620404084881537,
      "grad_norm": 0.1860496550798416,
      "learning_rate": 8.309028519142406e-05,
      "loss": 1.4408,
      "step": 65990
    },
    {
      "epoch": 3.1625823496536656,
      "grad_norm": 0.22845673561096191,
      "learning_rate": 8.30779099974911e-05,
      "loss": 1.4459,
      "step": 66000
    },
    {
      "epoch": 3.163124290819178,
      "grad_norm": 0.4244441092014313,
      "learning_rate": 8.306553132538856e-05,
      "loss": 1.4389,
      "step": 66010
    },
    {
      "epoch": 3.16366623198469,
      "grad_norm": 0.3140546679496765,
      "learning_rate": 8.305314917664985e-05,
      "loss": 1.4415,
      "step": 66020
    },
    {
      "epoch": 3.1642081731502025,
      "grad_norm": 0.34219682216644287,
      "learning_rate": 8.304076355280883e-05,
      "loss": 1.4586,
      "step": 66030
    },
    {
      "epoch": 3.164370755499856,
      "eval_loss": 2.432257890701294,
      "eval_runtime": 21.9773,
      "eval_samples_per_second": 227.508,
      "eval_steps_per_second": 1.229,
      "step": 66033
    },
    {
      "epoch": 3.1647501143157144,
      "grad_norm": 0.17820283770561218,
      "learning_rate": 8.302837445539974e-05,
      "loss": 1.4485,
      "step": 66040
    },
    {
      "epoch": 3.1652920554812267,
      "grad_norm": 0.22140522301197052,
      "learning_rate": 8.301598188595732e-05,
      "loss": 1.4522,
      "step": 66050
    },
    {
      "epoch": 3.165833996646739,
      "grad_norm": 0.24713575839996338,
      "learning_rate": 8.300358584601671e-05,
      "loss": 1.4491,
      "step": 66060
    },
    {
      "epoch": 3.1663759378122514,
      "grad_norm": 0.32958516478538513,
      "learning_rate": 8.299118633711344e-05,
      "loss": 1.44,
      "step": 66070
    },
    {
      "epoch": 3.1669178789777637,
      "grad_norm": 0.24236957728862762,
      "learning_rate": 8.297878336078354e-05,
      "loss": 1.4442,
      "step": 66080
    },
    {
      "epoch": 3.1674598201432755,
      "grad_norm": 0.2900751531124115,
      "learning_rate": 8.296637691856342e-05,
      "loss": 1.4535,
      "step": 66090
    },
    {
      "epoch": 3.168001761308788,
      "grad_norm": 0.2060800939798355,
      "learning_rate": 8.295396701198996e-05,
      "loss": 1.4472,
      "step": 66100
    },
    {
      "epoch": 3.1685437024743,
      "grad_norm": 0.45042285323143005,
      "learning_rate": 8.294155364260045e-05,
      "loss": 1.4494,
      "step": 66110
    },
    {
      "epoch": 3.1690856436398125,
      "grad_norm": 0.21350212395191193,
      "learning_rate": 8.292913681193254e-05,
      "loss": 1.447,
      "step": 66120
    },
    {
      "epoch": 3.1690856436398125,
      "eval_loss": 2.430042266845703,
      "eval_runtime": 21.5428,
      "eval_samples_per_second": 232.096,
      "eval_steps_per_second": 1.253,
      "step": 66120
    },
    {
      "epoch": 3.1696275848053244,
      "grad_norm": 0.19495175778865814,
      "learning_rate": 8.291671652152445e-05,
      "loss": 1.448,
      "step": 66130
    },
    {
      "epoch": 3.1701695259708367,
      "grad_norm": 0.22730892896652222,
      "learning_rate": 8.290429277291471e-05,
      "loss": 1.4446,
      "step": 66140
    },
    {
      "epoch": 3.170711467136349,
      "grad_norm": 0.23822692036628723,
      "learning_rate": 8.289186556764233e-05,
      "loss": 1.4448,
      "step": 66150
    },
    {
      "epoch": 3.1712534083018613,
      "grad_norm": 0.21188586950302124,
      "learning_rate": 8.287943490724673e-05,
      "loss": 1.441,
      "step": 66160
    },
    {
      "epoch": 3.1717953494673736,
      "grad_norm": 0.279965877532959,
      "learning_rate": 8.286700079326777e-05,
      "loss": 1.4367,
      "step": 66170
    },
    {
      "epoch": 3.1723372906328855,
      "grad_norm": 0.16774581372737885,
      "learning_rate": 8.285456322724577e-05,
      "loss": 1.4426,
      "step": 66180
    },
    {
      "epoch": 3.172879231798398,
      "grad_norm": 0.2504456341266632,
      "learning_rate": 8.284212221072139e-05,
      "loss": 1.4505,
      "step": 66190
    },
    {
      "epoch": 3.17342117296391,
      "grad_norm": 0.17107921838760376,
      "learning_rate": 8.282967774523579e-05,
      "loss": 1.4483,
      "step": 66200
    },
    {
      "epoch": 3.1738005317797686,
      "eval_loss": 2.4283711910247803,
      "eval_runtime": 21.9868,
      "eval_samples_per_second": 227.409,
      "eval_steps_per_second": 1.228,
      "step": 66207
    },
    {
      "epoch": 3.1739631141294224,
      "grad_norm": 0.2413926124572754,
      "learning_rate": 8.281722983233054e-05,
      "loss": 1.4405,
      "step": 66210
    },
    {
      "epoch": 3.1745050552949348,
      "grad_norm": 0.22221823036670685,
      "learning_rate": 8.280477847354763e-05,
      "loss": 1.4362,
      "step": 66220
    },
    {
      "epoch": 3.1750469964604466,
      "grad_norm": 0.2293311357498169,
      "learning_rate": 8.279232367042946e-05,
      "loss": 1.4406,
      "step": 66230
    },
    {
      "epoch": 3.175588937625959,
      "grad_norm": 0.22770948708057404,
      "learning_rate": 8.27798654245189e-05,
      "loss": 1.4537,
      "step": 66240
    },
    {
      "epoch": 3.1761308787914713,
      "grad_norm": 0.2080654501914978,
      "learning_rate": 8.276740373735922e-05,
      "loss": 1.4435,
      "step": 66250
    },
    {
      "epoch": 3.1766728199569836,
      "grad_norm": 0.23802123963832855,
      "learning_rate": 8.275493861049414e-05,
      "loss": 1.4486,
      "step": 66260
    },
    {
      "epoch": 3.1772147611224955,
      "grad_norm": 0.18610985577106476,
      "learning_rate": 8.274247004546775e-05,
      "loss": 1.4385,
      "step": 66270
    },
    {
      "epoch": 3.1777567022880078,
      "grad_norm": 0.2491571456193924,
      "learning_rate": 8.272999804382461e-05,
      "loss": 1.4416,
      "step": 66280
    },
    {
      "epoch": 3.17829864345352,
      "grad_norm": 0.41166606545448303,
      "learning_rate": 8.271752260710972e-05,
      "loss": 1.4539,
      "step": 66290
    },
    {
      "epoch": 3.1785154199197247,
      "eval_loss": 2.4275481700897217,
      "eval_runtime": 21.9832,
      "eval_samples_per_second": 227.446,
      "eval_steps_per_second": 1.228,
      "step": 66294
    },
    {
      "epoch": 3.1788405846190324,
      "grad_norm": 0.2705402672290802,
      "learning_rate": 8.270504373686846e-05,
      "loss": 1.4452,
      "step": 66300
    },
    {
      "epoch": 3.1793825257845447,
      "grad_norm": 0.16571412980556488,
      "learning_rate": 8.269256143464666e-05,
      "loss": 1.4402,
      "step": 66310
    },
    {
      "epoch": 3.1799244669500566,
      "grad_norm": 0.17406509816646576,
      "learning_rate": 8.268007570199058e-05,
      "loss": 1.4416,
      "step": 66320
    },
    {
      "epoch": 3.180466408115569,
      "grad_norm": 0.26004037261009216,
      "learning_rate": 8.26675865404469e-05,
      "loss": 1.4407,
      "step": 66330
    },
    {
      "epoch": 3.1810083492810812,
      "grad_norm": 0.18530094623565674,
      "learning_rate": 8.265509395156272e-05,
      "loss": 1.4352,
      "step": 66340
    },
    {
      "epoch": 3.1815502904465935,
      "grad_norm": 0.22706040740013123,
      "learning_rate": 8.264259793688555e-05,
      "loss": 1.4528,
      "step": 66350
    },
    {
      "epoch": 3.1820922316121054,
      "grad_norm": 0.3016396462917328,
      "learning_rate": 8.263009849796338e-05,
      "loss": 1.4534,
      "step": 66360
    },
    {
      "epoch": 3.1826341727776177,
      "grad_norm": 0.2243020385503769,
      "learning_rate": 8.261759563634458e-05,
      "loss": 1.4415,
      "step": 66370
    },
    {
      "epoch": 3.18317611394313,
      "grad_norm": 0.18285973370075226,
      "learning_rate": 8.260508935357791e-05,
      "loss": 1.4299,
      "step": 66380
    },
    {
      "epoch": 3.1832303080596813,
      "eval_loss": 2.425734043121338,
      "eval_runtime": 21.9789,
      "eval_samples_per_second": 227.491,
      "eval_steps_per_second": 1.228,
      "step": 66381
    },
    {
      "epoch": 3.1837180551086424,
      "grad_norm": 0.2271660417318344,
      "learning_rate": 8.259257965121263e-05,
      "loss": 1.4353,
      "step": 66390
    },
    {
      "epoch": 3.1842599962741547,
      "grad_norm": 0.1780574768781662,
      "learning_rate": 8.258006653079838e-05,
      "loss": 1.4335,
      "step": 66400
    },
    {
      "epoch": 3.1848019374396666,
      "grad_norm": 0.2370023876428604,
      "learning_rate": 8.256754999388522e-05,
      "loss": 1.4438,
      "step": 66410
    },
    {
      "epoch": 3.185343878605179,
      "grad_norm": 0.34876930713653564,
      "learning_rate": 8.255503004202365e-05,
      "loss": 1.4481,
      "step": 66420
    },
    {
      "epoch": 3.185885819770691,
      "grad_norm": 0.19301725924015045,
      "learning_rate": 8.254250667676461e-05,
      "loss": 1.4412,
      "step": 66430
    },
    {
      "epoch": 3.1864277609362035,
      "grad_norm": 0.22976385056972504,
      "learning_rate": 8.252997989965942e-05,
      "loss": 1.4483,
      "step": 66440
    },
    {
      "epoch": 3.186969702101716,
      "grad_norm": 0.17878347635269165,
      "learning_rate": 8.251744971225982e-05,
      "loss": 1.4393,
      "step": 66450
    },
    {
      "epoch": 3.1875116432672277,
      "grad_norm": 0.2477036416530609,
      "learning_rate": 8.250491611611803e-05,
      "loss": 1.4437,
      "step": 66460
    },
    {
      "epoch": 3.1879451961996375,
      "eval_loss": 2.425063371658325,
      "eval_runtime": 21.9818,
      "eval_samples_per_second": 227.46,
      "eval_steps_per_second": 1.228,
      "step": 66468
    },
    {
      "epoch": 3.18805358443274,
      "grad_norm": 0.26957419514656067,
      "learning_rate": 8.249237911278665e-05,
      "loss": 1.4446,
      "step": 66470
    },
    {
      "epoch": 3.1885955255982523,
      "grad_norm": 0.19030284881591797,
      "learning_rate": 8.24798387038187e-05,
      "loss": 1.4414,
      "step": 66480
    },
    {
      "epoch": 3.1891374667637646,
      "grad_norm": 0.18507073819637299,
      "learning_rate": 8.246729489076763e-05,
      "loss": 1.4501,
      "step": 66490
    },
    {
      "epoch": 3.1896794079292765,
      "grad_norm": 0.17628097534179688,
      "learning_rate": 8.245474767518734e-05,
      "loss": 1.444,
      "step": 66500
    },
    {
      "epoch": 3.190221349094789,
      "grad_norm": 0.259546160697937,
      "learning_rate": 8.24421970586321e-05,
      "loss": 1.44,
      "step": 66510
    },
    {
      "epoch": 3.190763290260301,
      "grad_norm": 0.22335298359394073,
      "learning_rate": 8.242964304265662e-05,
      "loss": 1.4356,
      "step": 66520
    },
    {
      "epoch": 3.1913052314258135,
      "grad_norm": 0.22630515694618225,
      "learning_rate": 8.241708562881604e-05,
      "loss": 1.4469,
      "step": 66530
    },
    {
      "epoch": 3.1918471725913253,
      "grad_norm": 0.2385028600692749,
      "learning_rate": 8.240452481866595e-05,
      "loss": 1.4392,
      "step": 66540
    },
    {
      "epoch": 3.1923891137568376,
      "grad_norm": 0.22308200597763062,
      "learning_rate": 8.239196061376229e-05,
      "loss": 1.4402,
      "step": 66550
    },
    {
      "epoch": 3.192660084339594,
      "eval_loss": 2.420628309249878,
      "eval_runtime": 21.9793,
      "eval_samples_per_second": 227.487,
      "eval_steps_per_second": 1.228,
      "step": 66555
    },
    {
      "epoch": 3.19293105492235,
      "grad_norm": 0.388051837682724,
      "learning_rate": 8.237939301566148e-05,
      "loss": 1.4462,
      "step": 66560
    },
    {
      "epoch": 3.1934729960878623,
      "grad_norm": 0.18918316066265106,
      "learning_rate": 8.236682202592032e-05,
      "loss": 1.4424,
      "step": 66570
    },
    {
      "epoch": 3.1940149372533746,
      "grad_norm": 0.392870157957077,
      "learning_rate": 8.235424764609607e-05,
      "loss": 1.447,
      "step": 66580
    },
    {
      "epoch": 3.1945568784188865,
      "grad_norm": 0.46847355365753174,
      "learning_rate": 8.234166987774639e-05,
      "loss": 1.4523,
      "step": 66590
    },
    {
      "epoch": 3.195098819584399,
      "grad_norm": 0.23610202968120575,
      "learning_rate": 8.232908872242932e-05,
      "loss": 1.4373,
      "step": 66600
    },
    {
      "epoch": 3.195640760749911,
      "grad_norm": 0.39547818899154663,
      "learning_rate": 8.231650418170343e-05,
      "loss": 1.4522,
      "step": 66610
    },
    {
      "epoch": 3.1961827019154234,
      "grad_norm": 0.3930874168872833,
      "learning_rate": 8.230391625712759e-05,
      "loss": 1.4393,
      "step": 66620
    },
    {
      "epoch": 3.1967246430809357,
      "grad_norm": 0.2097681313753128,
      "learning_rate": 8.229132495026113e-05,
      "loss": 1.4432,
      "step": 66630
    },
    {
      "epoch": 3.1972665842464476,
      "grad_norm": 0.19665232300758362,
      "learning_rate": 8.227873026266383e-05,
      "loss": 1.436,
      "step": 66640
    },
    {
      "epoch": 3.19737497247955,
      "eval_loss": 2.4246163368225098,
      "eval_runtime": 21.9864,
      "eval_samples_per_second": 227.414,
      "eval_steps_per_second": 1.228,
      "step": 66642
    },
    {
      "epoch": 3.19780852541196,
      "grad_norm": 0.2388431578874588,
      "learning_rate": 8.226613219589583e-05,
      "loss": 1.4458,
      "step": 66650
    },
    {
      "epoch": 3.1983504665774722,
      "grad_norm": 0.22420534491539001,
      "learning_rate": 8.225353075151781e-05,
      "loss": 1.4418,
      "step": 66660
    },
    {
      "epoch": 3.1988924077429846,
      "grad_norm": 0.22993390262126923,
      "learning_rate": 8.224092593109068e-05,
      "loss": 1.4308,
      "step": 66670
    },
    {
      "epoch": 3.1994343489084964,
      "grad_norm": 0.21597042679786682,
      "learning_rate": 8.222831773617592e-05,
      "loss": 1.4508,
      "step": 66680
    },
    {
      "epoch": 3.1999762900740087,
      "grad_norm": 0.21598871052265167,
      "learning_rate": 8.221570616833538e-05,
      "loss": 1.4428,
      "step": 66690
    },
    {
      "epoch": 3.200518231239521,
      "grad_norm": 0.20474159717559814,
      "learning_rate": 8.220309122913132e-05,
      "loss": 1.4461,
      "step": 66700
    },
    {
      "epoch": 3.2010601724050334,
      "grad_norm": 0.2574845850467682,
      "learning_rate": 8.219047292012642e-05,
      "loss": 1.4409,
      "step": 66710
    },
    {
      "epoch": 3.2016021135705457,
      "grad_norm": 0.17870694398880005,
      "learning_rate": 8.217785124288378e-05,
      "loss": 1.4424,
      "step": 66720
    },
    {
      "epoch": 3.2020898606195063,
      "eval_loss": 2.422410726547241,
      "eval_runtime": 21.9819,
      "eval_samples_per_second": 227.46,
      "eval_steps_per_second": 1.228,
      "step": 66729
    },
    {
      "epoch": 3.2021440547360576,
      "grad_norm": 0.2235659956932068,
      "learning_rate": 8.216522619896693e-05,
      "loss": 1.4361,
      "step": 66730
    },
    {
      "epoch": 3.20268599590157,
      "grad_norm": 0.16267363727092743,
      "learning_rate": 8.215259778993979e-05,
      "loss": 1.4447,
      "step": 66740
    },
    {
      "epoch": 3.203227937067082,
      "grad_norm": 0.26697710156440735,
      "learning_rate": 8.213996601736673e-05,
      "loss": 1.4388,
      "step": 66750
    },
    {
      "epoch": 3.2037698782325945,
      "grad_norm": 0.3052631616592407,
      "learning_rate": 8.21273308828125e-05,
      "loss": 1.4425,
      "step": 66760
    },
    {
      "epoch": 3.2043118193981064,
      "grad_norm": 0.22814327478408813,
      "learning_rate": 8.21146923878423e-05,
      "loss": 1.4351,
      "step": 66770
    },
    {
      "epoch": 3.2048537605636187,
      "grad_norm": 0.19500422477722168,
      "learning_rate": 8.210205053402172e-05,
      "loss": 1.4411,
      "step": 66780
    },
    {
      "epoch": 3.205395701729131,
      "grad_norm": 0.1982443779706955,
      "learning_rate": 8.20894053229168e-05,
      "loss": 1.4501,
      "step": 66790
    },
    {
      "epoch": 3.2059376428946433,
      "grad_norm": 0.2346677929162979,
      "learning_rate": 8.207675675609395e-05,
      "loss": 1.443,
      "step": 66800
    },
    {
      "epoch": 3.2064795840601557,
      "grad_norm": 0.18981340527534485,
      "learning_rate": 8.206410483512004e-05,
      "loss": 1.4277,
      "step": 66810
    },
    {
      "epoch": 3.206804748759463,
      "eval_loss": 2.4243276119232178,
      "eval_runtime": 21.9797,
      "eval_samples_per_second": 227.482,
      "eval_steps_per_second": 1.228,
      "step": 66816
    },
    {
      "epoch": 3.2070215252256675,
      "grad_norm": 0.20465369522571564,
      "learning_rate": 8.20514495615623e-05,
      "loss": 1.4445,
      "step": 66820
    },
    {
      "epoch": 3.20756346639118,
      "grad_norm": 0.18857480585575104,
      "learning_rate": 8.203879093698845e-05,
      "loss": 1.439,
      "step": 66830
    },
    {
      "epoch": 3.208105407556692,
      "grad_norm": 0.19271798431873322,
      "learning_rate": 8.202612896296657e-05,
      "loss": 1.4458,
      "step": 66840
    },
    {
      "epoch": 3.2086473487222045,
      "grad_norm": 0.42280739545822144,
      "learning_rate": 8.201346364106516e-05,
      "loss": 1.4374,
      "step": 66850
    },
    {
      "epoch": 3.209189289887717,
      "grad_norm": 0.21561777591705322,
      "learning_rate": 8.200079497285316e-05,
      "loss": 1.4565,
      "step": 66860
    },
    {
      "epoch": 3.2097312310532287,
      "grad_norm": 0.20918765664100647,
      "learning_rate": 8.198812295989991e-05,
      "loss": 1.4396,
      "step": 66870
    },
    {
      "epoch": 3.210273172218741,
      "grad_norm": 0.33416321873664856,
      "learning_rate": 8.197544760377514e-05,
      "loss": 1.4401,
      "step": 66880
    },
    {
      "epoch": 3.2108151133842533,
      "grad_norm": 0.2590186297893524,
      "learning_rate": 8.196276890604906e-05,
      "loss": 1.4388,
      "step": 66890
    },
    {
      "epoch": 3.2113570545497656,
      "grad_norm": 0.20215663313865662,
      "learning_rate": 8.195008686829222e-05,
      "loss": 1.4443,
      "step": 66900
    },
    {
      "epoch": 3.211519636899419,
      "eval_loss": 2.431623935699463,
      "eval_runtime": 21.9765,
      "eval_samples_per_second": 227.515,
      "eval_steps_per_second": 1.229,
      "step": 66903
    },
    {
      "epoch": 3.2118989957152775,
      "grad_norm": 0.3255484700202942,
      "learning_rate": 8.193740149207561e-05,
      "loss": 1.44,
      "step": 66910
    },
    {
      "epoch": 3.21244093688079,
      "grad_norm": 0.33131662011146545,
      "learning_rate": 8.192471277897068e-05,
      "loss": 1.4452,
      "step": 66920
    },
    {
      "epoch": 3.212982878046302,
      "grad_norm": 0.2151406705379486,
      "learning_rate": 8.191202073054922e-05,
      "loss": 1.4488,
      "step": 66930
    },
    {
      "epoch": 3.2135248192118144,
      "grad_norm": 0.2985169291496277,
      "learning_rate": 8.189932534838346e-05,
      "loss": 1.442,
      "step": 66940
    },
    {
      "epoch": 3.2140667603773263,
      "grad_norm": 0.21713115274906158,
      "learning_rate": 8.188662663404607e-05,
      "loss": 1.44,
      "step": 66950
    },
    {
      "epoch": 3.2146087015428386,
      "grad_norm": 0.25025808811187744,
      "learning_rate": 8.18739245891101e-05,
      "loss": 1.4394,
      "step": 66960
    },
    {
      "epoch": 3.215150642708351,
      "grad_norm": 0.4430973529815674,
      "learning_rate": 8.186121921514903e-05,
      "loss": 1.4407,
      "step": 66970
    },
    {
      "epoch": 3.2156925838738633,
      "grad_norm": 0.25356000661849976,
      "learning_rate": 8.184851051373673e-05,
      "loss": 1.4364,
      "step": 66980
    },
    {
      "epoch": 3.2162345250393756,
      "grad_norm": 0.3219984471797943,
      "learning_rate": 8.183579848644753e-05,
      "loss": 1.4377,
      "step": 66990
    },
    {
      "epoch": 3.2162345250393756,
      "eval_loss": 2.4254963397979736,
      "eval_runtime": 21.8545,
      "eval_samples_per_second": 228.785,
      "eval_steps_per_second": 1.235,
      "step": 66990
    },
    {
      "epoch": 3.2167764662048874,
      "grad_norm": 0.17054729163646698,
      "learning_rate": 8.18230831348561e-05,
      "loss": 1.4393,
      "step": 67000
    },
    {
      "epoch": 3.2173184073703998,
      "grad_norm": 0.26265397667884827,
      "learning_rate": 8.181036446053761e-05,
      "loss": 1.4521,
      "step": 67010
    },
    {
      "epoch": 3.217860348535912,
      "grad_norm": 0.3703189492225647,
      "learning_rate": 8.179764246506755e-05,
      "loss": 1.4443,
      "step": 67020
    },
    {
      "epoch": 3.2184022897014244,
      "grad_norm": 0.21587614715099335,
      "learning_rate": 8.17849171500219e-05,
      "loss": 1.4459,
      "step": 67030
    },
    {
      "epoch": 3.2189442308669367,
      "grad_norm": 0.21136698126792908,
      "learning_rate": 8.1772188516977e-05,
      "loss": 1.4417,
      "step": 67040
    },
    {
      "epoch": 3.2194861720324486,
      "grad_norm": 0.38758090138435364,
      "learning_rate": 8.175945656750962e-05,
      "loss": 1.4469,
      "step": 67050
    },
    {
      "epoch": 3.220028113197961,
      "grad_norm": 0.3118115961551666,
      "learning_rate": 8.174672130319694e-05,
      "loss": 1.4373,
      "step": 67060
    },
    {
      "epoch": 3.220570054363473,
      "grad_norm": 0.19323886930942535,
      "learning_rate": 8.173398272561654e-05,
      "loss": 1.4394,
      "step": 67070
    },
    {
      "epoch": 3.2209494131793317,
      "eval_loss": 2.42934513092041,
      "eval_runtime": 24.88,
      "eval_samples_per_second": 200.964,
      "eval_steps_per_second": 1.085,
      "step": 67077
    },
    {
      "epoch": 3.2211119955289855,
      "grad_norm": 0.22981129586696625,
      "learning_rate": 8.172124083634643e-05,
      "loss": 1.4441,
      "step": 67080
    },
    {
      "epoch": 3.2216539366944974,
      "grad_norm": 0.18165723979473114,
      "learning_rate": 8.170849563696501e-05,
      "loss": 1.4517,
      "step": 67090
    },
    {
      "epoch": 3.2221958778600097,
      "grad_norm": 0.19143837690353394,
      "learning_rate": 8.169574712905111e-05,
      "loss": 1.4327,
      "step": 67100
    },
    {
      "epoch": 3.222737819025522,
      "grad_norm": 0.3314666152000427,
      "learning_rate": 8.168299531418396e-05,
      "loss": 1.4345,
      "step": 67110
    },
    {
      "epoch": 3.2232797601910343,
      "grad_norm": 0.2969511151313782,
      "learning_rate": 8.167024019394321e-05,
      "loss": 1.4327,
      "step": 67120
    },
    {
      "epoch": 3.2238217013565467,
      "grad_norm": 0.2874842882156372,
      "learning_rate": 8.165748176990887e-05,
      "loss": 1.4512,
      "step": 67130
    },
    {
      "epoch": 3.2243636425220585,
      "grad_norm": 0.2401866316795349,
      "learning_rate": 8.164472004366145e-05,
      "loss": 1.4391,
      "step": 67140
    },
    {
      "epoch": 3.224905583687571,
      "grad_norm": 0.26989611983299255,
      "learning_rate": 8.163195501678177e-05,
      "loss": 1.4425,
      "step": 67150
    },
    {
      "epoch": 3.225447524853083,
      "grad_norm": 0.21331489086151123,
      "learning_rate": 8.161918669085113e-05,
      "loss": 1.4442,
      "step": 67160
    },
    {
      "epoch": 3.225664301319288,
      "eval_loss": 2.4356348514556885,
      "eval_runtime": 26.8873,
      "eval_samples_per_second": 185.961,
      "eval_steps_per_second": 1.004,
      "step": 67164
    },
    {
      "epoch": 3.2259894660185955,
      "grad_norm": 0.28904426097869873,
      "learning_rate": 8.160641506745123e-05,
      "loss": 1.4346,
      "step": 67170
    },
    {
      "epoch": 3.2265314071841074,
      "grad_norm": 0.3067649304866791,
      "learning_rate": 8.159364014816412e-05,
      "loss": 1.449,
      "step": 67180
    },
    {
      "epoch": 3.2270733483496197,
      "grad_norm": 0.23630201816558838,
      "learning_rate": 8.158086193457234e-05,
      "loss": 1.4384,
      "step": 67190
    },
    {
      "epoch": 3.227615289515132,
      "grad_norm": 0.21013253927230835,
      "learning_rate": 8.156808042825878e-05,
      "loss": 1.4325,
      "step": 67200
    },
    {
      "epoch": 3.2281572306806443,
      "grad_norm": 0.18870244920253754,
      "learning_rate": 8.155529563080676e-05,
      "loss": 1.4469,
      "step": 67210
    },
    {
      "epoch": 3.2286991718461566,
      "grad_norm": 0.266428142786026,
      "learning_rate": 8.154250754380002e-05,
      "loss": 1.442,
      "step": 67220
    },
    {
      "epoch": 3.2292411130116685,
      "grad_norm": 0.22033743560314178,
      "learning_rate": 8.152971616882269e-05,
      "loss": 1.4389,
      "step": 67230
    },
    {
      "epoch": 3.229783054177181,
      "grad_norm": 0.29417768120765686,
      "learning_rate": 8.151692150745928e-05,
      "loss": 1.4398,
      "step": 67240
    },
    {
      "epoch": 3.230324995342693,
      "grad_norm": 0.2595650255680084,
      "learning_rate": 8.150412356129478e-05,
      "loss": 1.4334,
      "step": 67250
    },
    {
      "epoch": 3.2303791894592444,
      "eval_loss": 2.4399569034576416,
      "eval_runtime": 22.4553,
      "eval_samples_per_second": 222.665,
      "eval_steps_per_second": 1.202,
      "step": 67251
    },
    {
      "epoch": 3.2308669365082054,
      "grad_norm": 0.16750743985176086,
      "learning_rate": 8.14913223319145e-05,
      "loss": 1.4424,
      "step": 67260
    },
    {
      "epoch": 3.2314088776737178,
      "grad_norm": 0.17600034177303314,
      "learning_rate": 8.147851782090425e-05,
      "loss": 1.4473,
      "step": 67270
    },
    {
      "epoch": 3.2319508188392296,
      "grad_norm": 0.18144869804382324,
      "learning_rate": 8.146571002985013e-05,
      "loss": 1.4402,
      "step": 67280
    },
    {
      "epoch": 3.232492760004742,
      "grad_norm": 0.19220831990242004,
      "learning_rate": 8.145289896033879e-05,
      "loss": 1.4406,
      "step": 67290
    },
    {
      "epoch": 3.2330347011702543,
      "grad_norm": 0.22534775733947754,
      "learning_rate": 8.144008461395716e-05,
      "loss": 1.4429,
      "step": 67300
    },
    {
      "epoch": 3.2335766423357666,
      "grad_norm": 0.18956543505191803,
      "learning_rate": 8.142726699229265e-05,
      "loss": 1.4538,
      "step": 67310
    },
    {
      "epoch": 3.2341185835012785,
      "grad_norm": 0.2575107514858246,
      "learning_rate": 8.141444609693302e-05,
      "loss": 1.4445,
      "step": 67320
    },
    {
      "epoch": 3.2346605246667908,
      "grad_norm": 0.37571388483047485,
      "learning_rate": 8.14016219294665e-05,
      "loss": 1.4412,
      "step": 67330
    },
    {
      "epoch": 3.2350940775992005,
      "eval_loss": 2.430121660232544,
      "eval_runtime": 24.4358,
      "eval_samples_per_second": 204.618,
      "eval_steps_per_second": 1.105,
      "step": 67338
    },
    {
      "epoch": 3.235202465832303,
      "grad_norm": 0.2669738531112671,
      "learning_rate": 8.138879449148168e-05,
      "loss": 1.4511,
      "step": 67340
    },
    {
      "epoch": 3.2357444069978154,
      "grad_norm": 0.16547146439552307,
      "learning_rate": 8.137596378456757e-05,
      "loss": 1.4397,
      "step": 67350
    },
    {
      "epoch": 3.2362863481633277,
      "grad_norm": 0.168304443359375,
      "learning_rate": 8.136312981031358e-05,
      "loss": 1.4388,
      "step": 67360
    },
    {
      "epoch": 3.2368282893288396,
      "grad_norm": 0.43757399916648865,
      "learning_rate": 8.135029257030953e-05,
      "loss": 1.4345,
      "step": 67370
    },
    {
      "epoch": 3.237370230494352,
      "grad_norm": 0.26308125257492065,
      "learning_rate": 8.133745206614561e-05,
      "loss": 1.4538,
      "step": 67380
    },
    {
      "epoch": 3.2379121716598642,
      "grad_norm": 0.18845511972904205,
      "learning_rate": 8.132460829941252e-05,
      "loss": 1.4393,
      "step": 67390
    },
    {
      "epoch": 3.2384541128253765,
      "grad_norm": 0.17935842275619507,
      "learning_rate": 8.13117612717012e-05,
      "loss": 1.437,
      "step": 67400
    },
    {
      "epoch": 3.2389960539908884,
      "grad_norm": 0.18170326948165894,
      "learning_rate": 8.129891098460316e-05,
      "loss": 1.4439,
      "step": 67410
    },
    {
      "epoch": 3.2395379951564007,
      "grad_norm": 0.2227085679769516,
      "learning_rate": 8.128605743971018e-05,
      "loss": 1.4394,
      "step": 67420
    },
    {
      "epoch": 3.239808965739157,
      "eval_loss": 2.421604633331299,
      "eval_runtime": 22.8082,
      "eval_samples_per_second": 219.219,
      "eval_steps_per_second": 1.184,
      "step": 67425
    },
    {
      "epoch": 3.240079936321913,
      "grad_norm": 0.26854372024536133,
      "learning_rate": 8.127320063861455e-05,
      "loss": 1.4398,
      "step": 67430
    },
    {
      "epoch": 3.2406218774874254,
      "grad_norm": 0.2939850389957428,
      "learning_rate": 8.126034058290887e-05,
      "loss": 1.4404,
      "step": 67440
    },
    {
      "epoch": 3.2411638186529377,
      "grad_norm": 0.20707334578037262,
      "learning_rate": 8.124747727418623e-05,
      "loss": 1.4451,
      "step": 67450
    },
    {
      "epoch": 3.2417057598184496,
      "grad_norm": 0.17989417910575867,
      "learning_rate": 8.123461071404005e-05,
      "loss": 1.4282,
      "step": 67460
    },
    {
      "epoch": 3.242247700983962,
      "grad_norm": 0.2010103315114975,
      "learning_rate": 8.122174090406418e-05,
      "loss": 1.4304,
      "step": 67470
    },
    {
      "epoch": 3.242789642149474,
      "grad_norm": 0.24504514038562775,
      "learning_rate": 8.120886784585292e-05,
      "loss": 1.4444,
      "step": 67480
    },
    {
      "epoch": 3.2433315833149865,
      "grad_norm": 0.16585427522659302,
      "learning_rate": 8.119599154100087e-05,
      "loss": 1.4337,
      "step": 67490
    },
    {
      "epoch": 3.243873524480499,
      "grad_norm": 0.2048460990190506,
      "learning_rate": 8.118311199110314e-05,
      "loss": 1.4403,
      "step": 67500
    },
    {
      "epoch": 3.2444154656460107,
      "grad_norm": 0.18525651097297668,
      "learning_rate": 8.117022919775516e-05,
      "loss": 1.4412,
      "step": 67510
    },
    {
      "epoch": 3.2445238538791132,
      "eval_loss": 2.4250948429107666,
      "eval_runtime": 24.9863,
      "eval_samples_per_second": 200.11,
      "eval_steps_per_second": 1.081,
      "step": 67512
    },
    {
      "epoch": 3.244957406811523,
      "grad_norm": 0.1921282261610031,
      "learning_rate": 8.115734316255281e-05,
      "loss": 1.4357,
      "step": 67520
    },
    {
      "epoch": 3.2454993479770353,
      "grad_norm": 0.3551950454711914,
      "learning_rate": 8.114445388709236e-05,
      "loss": 1.449,
      "step": 67530
    },
    {
      "epoch": 3.2460412891425476,
      "grad_norm": 0.26464539766311646,
      "learning_rate": 8.113156137297048e-05,
      "loss": 1.4367,
      "step": 67540
    },
    {
      "epoch": 3.2465832303080595,
      "grad_norm": 0.24829427897930145,
      "learning_rate": 8.111866562178419e-05,
      "loss": 1.4472,
      "step": 67550
    },
    {
      "epoch": 3.247125171473572,
      "grad_norm": 0.2007465362548828,
      "learning_rate": 8.110576663513105e-05,
      "loss": 1.4395,
      "step": 67560
    },
    {
      "epoch": 3.247667112639084,
      "grad_norm": 0.17801623046398163,
      "learning_rate": 8.109286441460885e-05,
      "loss": 1.4425,
      "step": 67570
    },
    {
      "epoch": 3.2482090538045965,
      "grad_norm": 0.2939807176589966,
      "learning_rate": 8.107995896181588e-05,
      "loss": 1.4434,
      "step": 67580
    },
    {
      "epoch": 3.2487509949701083,
      "grad_norm": 0.18530774116516113,
      "learning_rate": 8.106705027835083e-05,
      "loss": 1.4438,
      "step": 67590
    },
    {
      "epoch": 3.2492387420190694,
      "eval_loss": 2.4217963218688965,
      "eval_runtime": 23.4048,
      "eval_samples_per_second": 213.631,
      "eval_steps_per_second": 1.154,
      "step": 67599
    },
    {
      "epoch": 3.2492929361356206,
      "grad_norm": 0.2213825136423111,
      "learning_rate": 8.105413836581277e-05,
      "loss": 1.4492,
      "step": 67600
    },
    {
      "epoch": 3.249834877301133,
      "grad_norm": 0.2034333199262619,
      "learning_rate": 8.104122322580116e-05,
      "loss": 1.4339,
      "step": 67610
    },
    {
      "epoch": 3.2503768184666453,
      "grad_norm": 0.16729718446731567,
      "learning_rate": 8.102830485991589e-05,
      "loss": 1.4311,
      "step": 67620
    },
    {
      "epoch": 3.2509187596321576,
      "grad_norm": 0.18632495403289795,
      "learning_rate": 8.101538326975721e-05,
      "loss": 1.4385,
      "step": 67630
    },
    {
      "epoch": 3.2514607007976695,
      "grad_norm": 0.3231146037578583,
      "learning_rate": 8.100245845692579e-05,
      "loss": 1.4494,
      "step": 67640
    },
    {
      "epoch": 3.252002641963182,
      "grad_norm": 0.37084832787513733,
      "learning_rate": 8.09895304230227e-05,
      "loss": 1.4487,
      "step": 67650
    },
    {
      "epoch": 3.252544583128694,
      "grad_norm": 0.37098175287246704,
      "learning_rate": 8.097659916964943e-05,
      "loss": 1.4386,
      "step": 67660
    },
    {
      "epoch": 3.2530865242942064,
      "grad_norm": 0.2461938112974167,
      "learning_rate": 8.096366469840785e-05,
      "loss": 1.4404,
      "step": 67670
    },
    {
      "epoch": 3.2536284654597187,
      "grad_norm": 0.21850334107875824,
      "learning_rate": 8.095072701090019e-05,
      "loss": 1.4517,
      "step": 67680
    },
    {
      "epoch": 3.253953630159026,
      "eval_loss": 2.423842191696167,
      "eval_runtime": 21.9821,
      "eval_samples_per_second": 227.458,
      "eval_steps_per_second": 1.228,
      "step": 67686
    },
    {
      "epoch": 3.2541704066252306,
      "grad_norm": 0.18838092684745789,
      "learning_rate": 8.093778610872912e-05,
      "loss": 1.4467,
      "step": 67690
    },
    {
      "epoch": 3.254712347790743,
      "grad_norm": 0.21146613359451294,
      "learning_rate": 8.092484199349775e-05,
      "loss": 1.4391,
      "step": 67700
    },
    {
      "epoch": 3.2552542889562552,
      "grad_norm": 0.21148332953453064,
      "learning_rate": 8.091189466680948e-05,
      "loss": 1.4415,
      "step": 67710
    },
    {
      "epoch": 3.2557962301217676,
      "grad_norm": 0.18964509665966034,
      "learning_rate": 8.089894413026823e-05,
      "loss": 1.4444,
      "step": 67720
    },
    {
      "epoch": 3.25633817128728,
      "grad_norm": 0.21399128437042236,
      "learning_rate": 8.08859903854782e-05,
      "loss": 1.4514,
      "step": 67730
    },
    {
      "epoch": 3.2568801124527917,
      "grad_norm": 0.20721645653247833,
      "learning_rate": 8.087303343404406e-05,
      "loss": 1.4518,
      "step": 67740
    },
    {
      "epoch": 3.257422053618304,
      "grad_norm": 0.2715248763561249,
      "learning_rate": 8.086007327757088e-05,
      "loss": 1.4342,
      "step": 67750
    },
    {
      "epoch": 3.2579639947838164,
      "grad_norm": 0.19381947815418243,
      "learning_rate": 8.08471099176641e-05,
      "loss": 1.4455,
      "step": 67760
    },
    {
      "epoch": 3.2585059359493287,
      "grad_norm": 0.169798344373703,
      "learning_rate": 8.083414335592955e-05,
      "loss": 1.4354,
      "step": 67770
    },
    {
      "epoch": 3.258668518298982,
      "eval_loss": 2.425126075744629,
      "eval_runtime": 23.083,
      "eval_samples_per_second": 216.61,
      "eval_steps_per_second": 1.17,
      "step": 67773
    },
    {
      "epoch": 3.2590478771148406,
      "grad_norm": 0.17293022572994232,
      "learning_rate": 8.08211735939735e-05,
      "loss": 1.4359,
      "step": 67780
    },
    {
      "epoch": 3.259589818280353,
      "grad_norm": 0.28945857286453247,
      "learning_rate": 8.080820063340254e-05,
      "loss": 1.4333,
      "step": 67790
    },
    {
      "epoch": 3.260131759445865,
      "grad_norm": 0.18180948495864868,
      "learning_rate": 8.079522447582375e-05,
      "loss": 1.4428,
      "step": 67800
    },
    {
      "epoch": 3.2606737006113775,
      "grad_norm": 0.2261655181646347,
      "learning_rate": 8.078224512284455e-05,
      "loss": 1.4295,
      "step": 67810
    },
    {
      "epoch": 3.2612156417768894,
      "grad_norm": 0.5620535612106323,
      "learning_rate": 8.076926257607274e-05,
      "loss": 1.4416,
      "step": 67820
    },
    {
      "epoch": 3.2617575829424017,
      "grad_norm": 0.24003808200359344,
      "learning_rate": 8.075627683711658e-05,
      "loss": 1.4422,
      "step": 67830
    },
    {
      "epoch": 3.262299524107914,
      "grad_norm": 0.2081241011619568,
      "learning_rate": 8.074328790758466e-05,
      "loss": 1.4396,
      "step": 67840
    },
    {
      "epoch": 3.2628414652734263,
      "grad_norm": 0.22388330101966858,
      "learning_rate": 8.073029578908601e-05,
      "loss": 1.4407,
      "step": 67850
    },
    {
      "epoch": 3.2633834064389386,
      "grad_norm": 0.27499768137931824,
      "learning_rate": 8.071730048323002e-05,
      "loss": 1.4343,
      "step": 67860
    },
    {
      "epoch": 3.2633834064389386,
      "eval_loss": 2.4300835132598877,
      "eval_runtime": 22.2873,
      "eval_samples_per_second": 224.343,
      "eval_steps_per_second": 1.211,
      "step": 67860
    },
    {
      "epoch": 3.2639253476044505,
      "grad_norm": 0.17892539501190186,
      "learning_rate": 8.070430199162648e-05,
      "loss": 1.4281,
      "step": 67870
    },
    {
      "epoch": 3.264467288769963,
      "grad_norm": 0.24364718794822693,
      "learning_rate": 8.069130031588562e-05,
      "loss": 1.4381,
      "step": 67880
    },
    {
      "epoch": 3.265009229935475,
      "grad_norm": 0.22364751994609833,
      "learning_rate": 8.067829545761804e-05,
      "loss": 1.4383,
      "step": 67890
    },
    {
      "epoch": 3.2655511711009875,
      "grad_norm": 0.3235653340816498,
      "learning_rate": 8.066528741843468e-05,
      "loss": 1.4419,
      "step": 67900
    },
    {
      "epoch": 3.2660931122665,
      "grad_norm": 0.2588692605495453,
      "learning_rate": 8.065227619994695e-05,
      "loss": 1.4372,
      "step": 67910
    },
    {
      "epoch": 3.2666350534320117,
      "grad_norm": 0.22628279030323029,
      "learning_rate": 8.063926180376661e-05,
      "loss": 1.4345,
      "step": 67920
    },
    {
      "epoch": 3.267176994597524,
      "grad_norm": 0.29548385739326477,
      "learning_rate": 8.062624423150584e-05,
      "loss": 1.4383,
      "step": 67930
    },
    {
      "epoch": 3.2677189357630363,
      "grad_norm": 0.21970342099666595,
      "learning_rate": 8.061322348477717e-05,
      "loss": 1.4512,
      "step": 67940
    },
    {
      "epoch": 3.268098294578895,
      "eval_loss": 2.420971393585205,
      "eval_runtime": 28.5235,
      "eval_samples_per_second": 175.294,
      "eval_steps_per_second": 0.947,
      "step": 67947
    },
    {
      "epoch": 3.2682608769285486,
      "grad_norm": 0.19080619513988495,
      "learning_rate": 8.06001995651936e-05,
      "loss": 1.4366,
      "step": 67950
    },
    {
      "epoch": 3.2688028180940605,
      "grad_norm": 0.21438750624656677,
      "learning_rate": 8.058717247436845e-05,
      "loss": 1.4318,
      "step": 67960
    },
    {
      "epoch": 3.269344759259573,
      "grad_norm": 0.24113547801971436,
      "learning_rate": 8.057414221391545e-05,
      "loss": 1.443,
      "step": 67970
    },
    {
      "epoch": 3.269886700425085,
      "grad_norm": 0.22619077563285828,
      "learning_rate": 8.056110878544875e-05,
      "loss": 1.4333,
      "step": 67980
    },
    {
      "epoch": 3.2704286415905974,
      "grad_norm": 0.192766472697258,
      "learning_rate": 8.054807219058287e-05,
      "loss": 1.4479,
      "step": 67990
    },
    {
      "epoch": 3.2709705827561093,
      "grad_norm": 0.22505764663219452,
      "learning_rate": 8.053503243093275e-05,
      "loss": 1.4371,
      "step": 68000
    },
    {
      "epoch": 3.2715125239216216,
      "grad_norm": 0.18467269837856293,
      "learning_rate": 8.052198950811364e-05,
      "loss": 1.4294,
      "step": 68010
    },
    {
      "epoch": 3.272054465087134,
      "grad_norm": 0.21085874736309052,
      "learning_rate": 8.050894342374128e-05,
      "loss": 1.4519,
      "step": 68020
    },
    {
      "epoch": 3.2725964062526463,
      "grad_norm": 0.3685001730918884,
      "learning_rate": 8.049589417943176e-05,
      "loss": 1.4364,
      "step": 68030
    },
    {
      "epoch": 3.272813182718851,
      "eval_loss": 2.4268741607666016,
      "eval_runtime": 23.9572,
      "eval_samples_per_second": 208.705,
      "eval_steps_per_second": 1.127,
      "step": 68034
    },
    {
      "epoch": 3.2731383474181586,
      "grad_norm": 0.21414147317409515,
      "learning_rate": 8.048284177680158e-05,
      "loss": 1.4309,
      "step": 68040
    },
    {
      "epoch": 3.2736802885836704,
      "grad_norm": 0.23878353834152222,
      "learning_rate": 8.046978621746759e-05,
      "loss": 1.4443,
      "step": 68050
    },
    {
      "epoch": 3.2742222297491828,
      "grad_norm": 0.18432900309562683,
      "learning_rate": 8.045672750304703e-05,
      "loss": 1.429,
      "step": 68060
    },
    {
      "epoch": 3.274764170914695,
      "grad_norm": 0.18248611688613892,
      "learning_rate": 8.044366563515762e-05,
      "loss": 1.4367,
      "step": 68070
    },
    {
      "epoch": 3.2753061120802074,
      "grad_norm": 0.22367575764656067,
      "learning_rate": 8.043060061541737e-05,
      "loss": 1.4467,
      "step": 68080
    },
    {
      "epoch": 3.2758480532457197,
      "grad_norm": 0.17164652049541473,
      "learning_rate": 8.041753244544472e-05,
      "loss": 1.4351,
      "step": 68090
    },
    {
      "epoch": 3.2763899944112316,
      "grad_norm": 0.2587142288684845,
      "learning_rate": 8.04044611268585e-05,
      "loss": 1.4501,
      "step": 68100
    },
    {
      "epoch": 3.276931935576744,
      "grad_norm": 0.21978062391281128,
      "learning_rate": 8.039138666127793e-05,
      "loss": 1.4343,
      "step": 68110
    },
    {
      "epoch": 3.277473876742256,
      "grad_norm": 0.1821766346693039,
      "learning_rate": 8.037830905032264e-05,
      "loss": 1.4463,
      "step": 68120
    },
    {
      "epoch": 3.2775280708588075,
      "eval_loss": 2.42914080619812,
      "eval_runtime": 25.1974,
      "eval_samples_per_second": 198.433,
      "eval_steps_per_second": 1.072,
      "step": 68121
    },
    {
      "epoch": 3.2780158179077685,
      "grad_norm": 0.20630845427513123,
      "learning_rate": 8.036522829561259e-05,
      "loss": 1.4386,
      "step": 68130
    },
    {
      "epoch": 3.278557759073281,
      "grad_norm": 0.18165987730026245,
      "learning_rate": 8.035214439876818e-05,
      "loss": 1.4445,
      "step": 68140
    },
    {
      "epoch": 3.2790997002387927,
      "grad_norm": 0.2866304814815521,
      "learning_rate": 8.03390573614102e-05,
      "loss": 1.4429,
      "step": 68150
    },
    {
      "epoch": 3.279641641404305,
      "grad_norm": 0.4309597909450531,
      "learning_rate": 8.032596718515982e-05,
      "loss": 1.4471,
      "step": 68160
    },
    {
      "epoch": 3.2801835825698173,
      "grad_norm": 0.16174156963825226,
      "learning_rate": 8.031287387163854e-05,
      "loss": 1.4416,
      "step": 68170
    },
    {
      "epoch": 3.2807255237353297,
      "grad_norm": 0.20739299058914185,
      "learning_rate": 8.029977742246837e-05,
      "loss": 1.434,
      "step": 68180
    },
    {
      "epoch": 3.2812674649008415,
      "grad_norm": 0.30324587225914,
      "learning_rate": 8.02866778392716e-05,
      "loss": 1.4353,
      "step": 68190
    },
    {
      "epoch": 3.281809406066354,
      "grad_norm": 0.2629935145378113,
      "learning_rate": 8.027357512367097e-05,
      "loss": 1.4326,
      "step": 68200
    },
    {
      "epoch": 3.2822429589987636,
      "eval_loss": 2.420156955718994,
      "eval_runtime": 25.5157,
      "eval_samples_per_second": 195.958,
      "eval_steps_per_second": 1.058,
      "step": 68208
    },
    {
      "epoch": 3.282351347231866,
      "grad_norm": 0.23528800904750824,
      "learning_rate": 8.026046927728959e-05,
      "loss": 1.4362,
      "step": 68210
    },
    {
      "epoch": 3.2828932883973785,
      "grad_norm": 0.18353603780269623,
      "learning_rate": 8.024736030175092e-05,
      "loss": 1.4437,
      "step": 68220
    },
    {
      "epoch": 3.2834352295628904,
      "grad_norm": 0.3570035994052887,
      "learning_rate": 8.02342481986789e-05,
      "loss": 1.4461,
      "step": 68230
    },
    {
      "epoch": 3.2839771707284027,
      "grad_norm": 0.43758606910705566,
      "learning_rate": 8.022113296969773e-05,
      "loss": 1.4389,
      "step": 68240
    },
    {
      "epoch": 3.284519111893915,
      "grad_norm": 0.2364160418510437,
      "learning_rate": 8.020801461643214e-05,
      "loss": 1.4388,
      "step": 68250
    },
    {
      "epoch": 3.2850610530594273,
      "grad_norm": 0.3303424119949341,
      "learning_rate": 8.019489314050715e-05,
      "loss": 1.4482,
      "step": 68260
    },
    {
      "epoch": 3.2856029942249396,
      "grad_norm": 0.3738381266593933,
      "learning_rate": 8.018176854354815e-05,
      "loss": 1.4299,
      "step": 68270
    },
    {
      "epoch": 3.2861449353904515,
      "grad_norm": 0.1708504557609558,
      "learning_rate": 8.016864082718102e-05,
      "loss": 1.4301,
      "step": 68280
    },
    {
      "epoch": 3.286686876555964,
      "grad_norm": 0.22790126502513885,
      "learning_rate": 8.015550999303192e-05,
      "loss": 1.4363,
      "step": 68290
    },
    {
      "epoch": 3.28695784713872,
      "eval_loss": 2.4150733947753906,
      "eval_runtime": 25.2805,
      "eval_samples_per_second": 197.781,
      "eval_steps_per_second": 1.068,
      "step": 68295
    },
    {
      "epoch": 3.287228817721476,
      "grad_norm": 0.17331331968307495,
      "learning_rate": 8.014237604272744e-05,
      "loss": 1.4265,
      "step": 68300
    },
    {
      "epoch": 3.2877707588869884,
      "grad_norm": 0.2437063455581665,
      "learning_rate": 8.012923897789461e-05,
      "loss": 1.4362,
      "step": 68310
    },
    {
      "epoch": 3.2883127000525008,
      "grad_norm": 0.1904052495956421,
      "learning_rate": 8.011609880016074e-05,
      "loss": 1.4315,
      "step": 68320
    },
    {
      "epoch": 3.2888546412180126,
      "grad_norm": 0.17929191887378693,
      "learning_rate": 8.010295551115358e-05,
      "loss": 1.4352,
      "step": 68330
    },
    {
      "epoch": 3.289396582383525,
      "grad_norm": 0.24011602997779846,
      "learning_rate": 8.00898091125013e-05,
      "loss": 1.4416,
      "step": 68340
    },
    {
      "epoch": 3.2899385235490373,
      "grad_norm": 0.26743006706237793,
      "learning_rate": 8.007665960583237e-05,
      "loss": 1.4295,
      "step": 68350
    },
    {
      "epoch": 3.2904804647145496,
      "grad_norm": 0.21712139248847961,
      "learning_rate": 8.006350699277573e-05,
      "loss": 1.4409,
      "step": 68360
    },
    {
      "epoch": 3.2910224058800615,
      "grad_norm": 0.19347068667411804,
      "learning_rate": 8.005035127496068e-05,
      "loss": 1.4342,
      "step": 68370
    },
    {
      "epoch": 3.2915643470455738,
      "grad_norm": 0.2079339623451233,
      "learning_rate": 8.003719245401684e-05,
      "loss": 1.4394,
      "step": 68380
    },
    {
      "epoch": 3.2916727352786763,
      "eval_loss": 2.4259541034698486,
      "eval_runtime": 29.0409,
      "eval_samples_per_second": 172.171,
      "eval_steps_per_second": 0.93,
      "step": 68382
    },
    {
      "epoch": 3.292106288211086,
      "grad_norm": 0.17019985616207123,
      "learning_rate": 8.002403053157432e-05,
      "loss": 1.4358,
      "step": 68390
    },
    {
      "epoch": 3.2926482293765984,
      "grad_norm": 0.18729358911514282,
      "learning_rate": 8.001086550926354e-05,
      "loss": 1.431,
      "step": 68400
    },
    {
      "epoch": 3.2931901705421103,
      "grad_norm": 0.3095646798610687,
      "learning_rate": 7.999769738871533e-05,
      "loss": 1.4412,
      "step": 68410
    },
    {
      "epoch": 3.2937321117076226,
      "grad_norm": 0.17975299060344696,
      "learning_rate": 7.998452617156088e-05,
      "loss": 1.4424,
      "step": 68420
    },
    {
      "epoch": 3.294274052873135,
      "grad_norm": 0.15972350537776947,
      "learning_rate": 7.997135185943182e-05,
      "loss": 1.4329,
      "step": 68430
    },
    {
      "epoch": 3.2948159940386472,
      "grad_norm": 0.19751180708408356,
      "learning_rate": 7.995817445396009e-05,
      "loss": 1.437,
      "step": 68440
    },
    {
      "epoch": 3.2953579352041595,
      "grad_norm": 0.17910368740558624,
      "learning_rate": 7.994499395677807e-05,
      "loss": 1.4297,
      "step": 68450
    },
    {
      "epoch": 3.2958998763696714,
      "grad_norm": 0.24977631866931915,
      "learning_rate": 7.99318103695185e-05,
      "loss": 1.4352,
      "step": 68460
    },
    {
      "epoch": 3.2963876234186325,
      "eval_loss": 2.4190785884857178,
      "eval_runtime": 26.6502,
      "eval_samples_per_second": 187.616,
      "eval_steps_per_second": 1.013,
      "step": 68469
    },
    {
      "epoch": 3.2964418175351837,
      "grad_norm": 0.2186552733182907,
      "learning_rate": 7.99186236938145e-05,
      "loss": 1.4364,
      "step": 68470
    },
    {
      "epoch": 3.296983758700696,
      "grad_norm": 0.29373136162757874,
      "learning_rate": 7.990543393129959e-05,
      "loss": 1.4334,
      "step": 68480
    },
    {
      "epoch": 3.2975256998662084,
      "grad_norm": 0.17339174449443817,
      "learning_rate": 7.989224108360765e-05,
      "loss": 1.4425,
      "step": 68490
    },
    {
      "epoch": 3.2980676410317207,
      "grad_norm": 0.16692684590816498,
      "learning_rate": 7.987904515237297e-05,
      "loss": 1.445,
      "step": 68500
    },
    {
      "epoch": 3.2986095821972325,
      "grad_norm": 0.17328135669231415,
      "learning_rate": 7.986584613923017e-05,
      "loss": 1.4348,
      "step": 68510
    },
    {
      "epoch": 3.299151523362745,
      "grad_norm": 0.2851394712924957,
      "learning_rate": 7.985264404581431e-05,
      "loss": 1.45,
      "step": 68520
    },
    {
      "epoch": 3.299693464528257,
      "grad_norm": 0.18991175293922424,
      "learning_rate": 7.983943887376083e-05,
      "loss": 1.4392,
      "step": 68530
    },
    {
      "epoch": 3.3002354056937695,
      "grad_norm": 0.18527284264564514,
      "learning_rate": 7.982623062470547e-05,
      "loss": 1.4445,
      "step": 68540
    },
    {
      "epoch": 3.300777346859282,
      "grad_norm": 0.1802045851945877,
      "learning_rate": 7.981301930028446e-05,
      "loss": 1.4487,
      "step": 68550
    },
    {
      "epoch": 3.301102511558589,
      "eval_loss": 2.417846441268921,
      "eval_runtime": 26.4831,
      "eval_samples_per_second": 188.799,
      "eval_steps_per_second": 1.02,
      "step": 68556
    },
    {
      "epoch": 3.3013192880247937,
      "grad_norm": 0.15754951536655426,
      "learning_rate": 7.979980490213435e-05,
      "loss": 1.4363,
      "step": 68560
    },
    {
      "epoch": 3.301861229190306,
      "grad_norm": 0.31062695384025574,
      "learning_rate": 7.978658743189205e-05,
      "loss": 1.4359,
      "step": 68570
    },
    {
      "epoch": 3.3024031703558183,
      "grad_norm": 0.2875862717628479,
      "learning_rate": 7.977336689119495e-05,
      "loss": 1.4459,
      "step": 68580
    },
    {
      "epoch": 3.3029451115213306,
      "grad_norm": 0.18160034716129303,
      "learning_rate": 7.97601432816807e-05,
      "loss": 1.4378,
      "step": 68590
    },
    {
      "epoch": 3.3034870526868425,
      "grad_norm": 0.24987049400806427,
      "learning_rate": 7.97469166049874e-05,
      "loss": 1.4372,
      "step": 68600
    },
    {
      "epoch": 3.304028993852355,
      "grad_norm": 0.2157442718744278,
      "learning_rate": 7.973368686275353e-05,
      "loss": 1.4379,
      "step": 68610
    },
    {
      "epoch": 3.304570935017867,
      "grad_norm": 0.20178145170211792,
      "learning_rate": 7.972045405661788e-05,
      "loss": 1.4259,
      "step": 68620
    },
    {
      "epoch": 3.3051128761833795,
      "grad_norm": 0.24957998096942902,
      "learning_rate": 7.970721818821972e-05,
      "loss": 1.4474,
      "step": 68630
    },
    {
      "epoch": 3.3056548173488913,
      "grad_norm": 0.16413570940494537,
      "learning_rate": 7.969397925919863e-05,
      "loss": 1.434,
      "step": 68640
    },
    {
      "epoch": 3.305817399698545,
      "eval_loss": 2.419210433959961,
      "eval_runtime": 24.5355,
      "eval_samples_per_second": 203.786,
      "eval_steps_per_second": 1.1,
      "step": 68643
    },
    {
      "epoch": 3.3061967585144036,
      "grad_norm": 0.31925585865974426,
      "learning_rate": 7.968073727119461e-05,
      "loss": 1.4409,
      "step": 68650
    },
    {
      "epoch": 3.306738699679916,
      "grad_norm": 0.193577840924263,
      "learning_rate": 7.966749222584802e-05,
      "loss": 1.4379,
      "step": 68660
    },
    {
      "epoch": 3.3072806408454283,
      "grad_norm": 0.3227108418941498,
      "learning_rate": 7.965424412479958e-05,
      "loss": 1.4412,
      "step": 68670
    },
    {
      "epoch": 3.3078225820109406,
      "grad_norm": 0.2504270076751709,
      "learning_rate": 7.964099296969042e-05,
      "loss": 1.4256,
      "step": 68680
    },
    {
      "epoch": 3.3083645231764525,
      "grad_norm": 0.23025809228420258,
      "learning_rate": 7.962773876216202e-05,
      "loss": 1.4289,
      "step": 68690
    },
    {
      "epoch": 3.308906464341965,
      "grad_norm": 0.17569345235824585,
      "learning_rate": 7.961448150385628e-05,
      "loss": 1.4433,
      "step": 68700
    },
    {
      "epoch": 3.309448405507477,
      "grad_norm": 0.24907849729061127,
      "learning_rate": 7.960122119641542e-05,
      "loss": 1.443,
      "step": 68710
    },
    {
      "epoch": 3.3099903466729894,
      "grad_norm": 0.30965548753738403,
      "learning_rate": 7.958795784148208e-05,
      "loss": 1.4377,
      "step": 68720
    },
    {
      "epoch": 3.3105322878385017,
      "grad_norm": 0.30555593967437744,
      "learning_rate": 7.95746914406993e-05,
      "loss": 1.441,
      "step": 68730
    },
    {
      "epoch": 3.3105322878385017,
      "eval_loss": 2.420576572418213,
      "eval_runtime": 21.9481,
      "eval_samples_per_second": 227.81,
      "eval_steps_per_second": 1.23,
      "step": 68730
    },
    {
      "epoch": 3.3110742290040136,
      "grad_norm": 0.2148512899875641,
      "learning_rate": 7.956142199571042e-05,
      "loss": 1.4404,
      "step": 68740
    },
    {
      "epoch": 3.311616170169526,
      "grad_norm": 0.3420555591583252,
      "learning_rate": 7.954814950815922e-05,
      "loss": 1.4432,
      "step": 68750
    },
    {
      "epoch": 3.3121581113350382,
      "grad_norm": 0.22187262773513794,
      "learning_rate": 7.953487397968984e-05,
      "loss": 1.4425,
      "step": 68760
    },
    {
      "epoch": 3.3127000525005506,
      "grad_norm": 0.2587040662765503,
      "learning_rate": 7.95215954119468e-05,
      "loss": 1.451,
      "step": 68770
    },
    {
      "epoch": 3.313241993666063,
      "grad_norm": 0.20211346447467804,
      "learning_rate": 7.950831380657496e-05,
      "loss": 1.4466,
      "step": 68780
    },
    {
      "epoch": 3.3137839348315747,
      "grad_norm": 0.2853049337863922,
      "learning_rate": 7.949502916521962e-05,
      "loss": 1.4423,
      "step": 68790
    },
    {
      "epoch": 3.314325875997087,
      "grad_norm": 0.17778904736042023,
      "learning_rate": 7.948174148952642e-05,
      "loss": 1.4383,
      "step": 68800
    },
    {
      "epoch": 3.3148678171625994,
      "grad_norm": 0.30810484290122986,
      "learning_rate": 7.946845078114137e-05,
      "loss": 1.4441,
      "step": 68810
    },
    {
      "epoch": 3.315247175978458,
      "eval_loss": 2.4207398891448975,
      "eval_runtime": 32.3649,
      "eval_samples_per_second": 154.488,
      "eval_steps_per_second": 0.834,
      "step": 68817
    },
    {
      "epoch": 3.3154097583281117,
      "grad_norm": 0.19365157186985016,
      "learning_rate": 7.945515704171088e-05,
      "loss": 1.4428,
      "step": 68820
    },
    {
      "epoch": 3.3159516994936236,
      "grad_norm": 0.17585448920726776,
      "learning_rate": 7.944186027288169e-05,
      "loss": 1.4359,
      "step": 68830
    },
    {
      "epoch": 3.316493640659136,
      "grad_norm": 0.1955270618200302,
      "learning_rate": 7.942856047630098e-05,
      "loss": 1.4375,
      "step": 68840
    },
    {
      "epoch": 3.317035581824648,
      "grad_norm": 0.17522399127483368,
      "learning_rate": 7.941525765361624e-05,
      "loss": 1.4442,
      "step": 68850
    },
    {
      "epoch": 3.3175775229901605,
      "grad_norm": 0.20333018898963928,
      "learning_rate": 7.940195180647539e-05,
      "loss": 1.4478,
      "step": 68860
    },
    {
      "epoch": 3.3181194641556724,
      "grad_norm": 0.22244490683078766,
      "learning_rate": 7.93886429365267e-05,
      "loss": 1.428,
      "step": 68870
    },
    {
      "epoch": 3.3186614053211847,
      "grad_norm": 0.22800850868225098,
      "learning_rate": 7.93753310454188e-05,
      "loss": 1.4382,
      "step": 68880
    },
    {
      "epoch": 3.319203346486697,
      "grad_norm": 0.3180639147758484,
      "learning_rate": 7.93620161348007e-05,
      "loss": 1.4318,
      "step": 68890
    },
    {
      "epoch": 3.3197452876522093,
      "grad_norm": 0.4034172594547272,
      "learning_rate": 7.934869820632183e-05,
      "loss": 1.4457,
      "step": 68900
    },
    {
      "epoch": 3.319962064118414,
      "eval_loss": 2.4232521057128906,
      "eval_runtime": 24.2726,
      "eval_samples_per_second": 205.994,
      "eval_steps_per_second": 1.112,
      "step": 68904
    },
    {
      "epoch": 3.3202872288177216,
      "grad_norm": 0.32735925912857056,
      "learning_rate": 7.933537726163195e-05,
      "loss": 1.4403,
      "step": 68910
    },
    {
      "epoch": 3.3208291699832335,
      "grad_norm": 0.2666130065917969,
      "learning_rate": 7.932205330238118e-05,
      "loss": 1.4348,
      "step": 68920
    },
    {
      "epoch": 3.321371111148746,
      "grad_norm": 0.20888814330101013,
      "learning_rate": 7.930872633022006e-05,
      "loss": 1.4446,
      "step": 68930
    },
    {
      "epoch": 3.321913052314258,
      "grad_norm": 0.26774919033050537,
      "learning_rate": 7.929539634679941e-05,
      "loss": 1.4374,
      "step": 68940
    },
    {
      "epoch": 3.3224549934797705,
      "grad_norm": 0.18306075036525726,
      "learning_rate": 7.928206335377057e-05,
      "loss": 1.4408,
      "step": 68950
    },
    {
      "epoch": 3.322996934645283,
      "grad_norm": 0.2182762324810028,
      "learning_rate": 7.926872735278514e-05,
      "loss": 1.439,
      "step": 68960
    },
    {
      "epoch": 3.3235388758107947,
      "grad_norm": 0.23973847925662994,
      "learning_rate": 7.925538834549514e-05,
      "loss": 1.4403,
      "step": 68970
    },
    {
      "epoch": 3.324080816976307,
      "grad_norm": 0.2488727569580078,
      "learning_rate": 7.924204633355295e-05,
      "loss": 1.4386,
      "step": 68980
    },
    {
      "epoch": 3.3246227581418193,
      "grad_norm": 0.2777588963508606,
      "learning_rate": 7.922870131861127e-05,
      "loss": 1.4477,
      "step": 68990
    },
    {
      "epoch": 3.3246769522583706,
      "eval_loss": 2.42183518409729,
      "eval_runtime": 21.8322,
      "eval_samples_per_second": 229.02,
      "eval_steps_per_second": 1.237,
      "step": 68991
    },
    {
      "epoch": 3.3251646993073316,
      "grad_norm": 0.18954722583293915,
      "learning_rate": 7.921535330232328e-05,
      "loss": 1.4356,
      "step": 69000
    },
    {
      "epoch": 3.3257066404728435,
      "grad_norm": 0.2268332540988922,
      "learning_rate": 7.920200228634245e-05,
      "loss": 1.4413,
      "step": 69010
    },
    {
      "epoch": 3.326248581638356,
      "grad_norm": 0.21182286739349365,
      "learning_rate": 7.918864827232268e-05,
      "loss": 1.4295,
      "step": 69020
    },
    {
      "epoch": 3.326790522803868,
      "grad_norm": 0.16312585771083832,
      "learning_rate": 7.917529126191815e-05,
      "loss": 1.4398,
      "step": 69030
    },
    {
      "epoch": 3.3273324639693804,
      "grad_norm": 0.2508258819580078,
      "learning_rate": 7.916193125678349e-05,
      "loss": 1.431,
      "step": 69040
    },
    {
      "epoch": 3.3278744051348923,
      "grad_norm": 0.3282955586910248,
      "learning_rate": 7.914856825857371e-05,
      "loss": 1.4384,
      "step": 69050
    },
    {
      "epoch": 3.3284163463004046,
      "grad_norm": 0.20805367827415466,
      "learning_rate": 7.913520226894413e-05,
      "loss": 1.4365,
      "step": 69060
    },
    {
      "epoch": 3.328958287465917,
      "grad_norm": 0.32144641876220703,
      "learning_rate": 7.912183328955047e-05,
      "loss": 1.439,
      "step": 69070
    },
    {
      "epoch": 3.3293918403983267,
      "eval_loss": 2.420147657394409,
      "eval_runtime": 23.884,
      "eval_samples_per_second": 209.345,
      "eval_steps_per_second": 1.13,
      "step": 69078
    },
    {
      "epoch": 3.3295002286314292,
      "grad_norm": 0.27090176939964294,
      "learning_rate": 7.910846132204883e-05,
      "loss": 1.4386,
      "step": 69080
    },
    {
      "epoch": 3.3300421697969416,
      "grad_norm": 0.25403621792793274,
      "learning_rate": 7.909508636809567e-05,
      "loss": 1.4382,
      "step": 69090
    },
    {
      "epoch": 3.3305841109624534,
      "grad_norm": 0.23677849769592285,
      "learning_rate": 7.908170842934783e-05,
      "loss": 1.4414,
      "step": 69100
    },
    {
      "epoch": 3.3311260521279658,
      "grad_norm": 0.2112101912498474,
      "learning_rate": 7.90683275074625e-05,
      "loss": 1.4353,
      "step": 69110
    },
    {
      "epoch": 3.331667993293478,
      "grad_norm": 0.18061186373233795,
      "learning_rate": 7.905494360409725e-05,
      "loss": 1.4271,
      "step": 69120
    },
    {
      "epoch": 3.3322099344589904,
      "grad_norm": 0.34150585532188416,
      "learning_rate": 7.904155672091002e-05,
      "loss": 1.4275,
      "step": 69130
    },
    {
      "epoch": 3.3327518756245027,
      "grad_norm": 0.20496203005313873,
      "learning_rate": 7.902816685955912e-05,
      "loss": 1.4328,
      "step": 69140
    },
    {
      "epoch": 3.3332938167900146,
      "grad_norm": 0.43717578053474426,
      "learning_rate": 7.901477402170323e-05,
      "loss": 1.4455,
      "step": 69150
    },
    {
      "epoch": 3.333835757955527,
      "grad_norm": 0.29195889830589294,
      "learning_rate": 7.90013782090014e-05,
      "loss": 1.4471,
      "step": 69160
    },
    {
      "epoch": 3.3341067285382833,
      "eval_loss": 2.4220991134643555,
      "eval_runtime": 23.5219,
      "eval_samples_per_second": 212.568,
      "eval_steps_per_second": 1.148,
      "step": 69165
    },
    {
      "epoch": 3.334377699121039,
      "grad_norm": 0.27426832914352417,
      "learning_rate": 7.898797942311304e-05,
      "loss": 1.4368,
      "step": 69170
    },
    {
      "epoch": 3.3349196402865515,
      "grad_norm": 0.21939682960510254,
      "learning_rate": 7.897457766569793e-05,
      "loss": 1.4365,
      "step": 69180
    },
    {
      "epoch": 3.335461581452064,
      "grad_norm": 0.20194536447525024,
      "learning_rate": 7.896117293841622e-05,
      "loss": 1.4327,
      "step": 69190
    },
    {
      "epoch": 3.3360035226175757,
      "grad_norm": 0.22306720912456512,
      "learning_rate": 7.894776524292845e-05,
      "loss": 1.4425,
      "step": 69200
    },
    {
      "epoch": 3.336545463783088,
      "grad_norm": 0.547161877155304,
      "learning_rate": 7.893435458089549e-05,
      "loss": 1.4359,
      "step": 69210
    },
    {
      "epoch": 3.3370874049486003,
      "grad_norm": 0.17057354748249054,
      "learning_rate": 7.89209409539786e-05,
      "loss": 1.4307,
      "step": 69220
    },
    {
      "epoch": 3.3376293461141127,
      "grad_norm": 0.4212014973163605,
      "learning_rate": 7.890752436383939e-05,
      "loss": 1.4284,
      "step": 69230
    },
    {
      "epoch": 3.3381712872796245,
      "grad_norm": 0.272225558757782,
      "learning_rate": 7.889410481213986e-05,
      "loss": 1.4391,
      "step": 69240
    },
    {
      "epoch": 3.338713228445137,
      "grad_norm": 0.2661522626876831,
      "learning_rate": 7.888068230054236e-05,
      "loss": 1.4344,
      "step": 69250
    },
    {
      "epoch": 3.3388216166782394,
      "eval_loss": 2.4209210872650146,
      "eval_runtime": 23.9348,
      "eval_samples_per_second": 208.901,
      "eval_steps_per_second": 1.128,
      "step": 69252
    },
    {
      "epoch": 3.339255169610649,
      "grad_norm": 0.21756187081336975,
      "learning_rate": 7.886725683070963e-05,
      "loss": 1.4455,
      "step": 69260
    },
    {
      "epoch": 3.3397971107761615,
      "grad_norm": 0.1777390092611313,
      "learning_rate": 7.885382840430475e-05,
      "loss": 1.4404,
      "step": 69270
    },
    {
      "epoch": 3.3403390519416734,
      "grad_norm": 0.21643613278865814,
      "learning_rate": 7.884039702299113e-05,
      "loss": 1.4348,
      "step": 69280
    },
    {
      "epoch": 3.3408809931071857,
      "grad_norm": 0.21973682940006256,
      "learning_rate": 7.882696268843267e-05,
      "loss": 1.4467,
      "step": 69290
    },
    {
      "epoch": 3.341422934272698,
      "grad_norm": 0.2839016020298004,
      "learning_rate": 7.881352540229351e-05,
      "loss": 1.4296,
      "step": 69300
    },
    {
      "epoch": 3.3419648754382103,
      "grad_norm": 0.17551878094673157,
      "learning_rate": 7.880008516623822e-05,
      "loss": 1.4323,
      "step": 69310
    },
    {
      "epoch": 3.3425068166037226,
      "grad_norm": 0.19211049377918243,
      "learning_rate": 7.878664198193169e-05,
      "loss": 1.436,
      "step": 69320
    },
    {
      "epoch": 3.3430487577692345,
      "grad_norm": 0.2430991530418396,
      "learning_rate": 7.877319585103922e-05,
      "loss": 1.4405,
      "step": 69330
    },
    {
      "epoch": 3.3435365048181955,
      "eval_loss": 2.417872190475464,
      "eval_runtime": 28.3658,
      "eval_samples_per_second": 176.269,
      "eval_steps_per_second": 0.952,
      "step": 69339
    },
    {
      "epoch": 3.343590698934747,
      "grad_norm": 0.2997911274433136,
      "learning_rate": 7.875974677522648e-05,
      "loss": 1.4357,
      "step": 69340
    },
    {
      "epoch": 3.344132640100259,
      "grad_norm": 0.18667905032634735,
      "learning_rate": 7.874629475615946e-05,
      "loss": 1.4338,
      "step": 69350
    },
    {
      "epoch": 3.3446745812657714,
      "grad_norm": 0.20530866086483002,
      "learning_rate": 7.873283979550452e-05,
      "loss": 1.4477,
      "step": 69360
    },
    {
      "epoch": 3.3452165224312838,
      "grad_norm": 0.16726155579090118,
      "learning_rate": 7.871938189492844e-05,
      "loss": 1.4393,
      "step": 69370
    },
    {
      "epoch": 3.3457584635967956,
      "grad_norm": 0.16196750104427338,
      "learning_rate": 7.870592105609832e-05,
      "loss": 1.4274,
      "step": 69380
    },
    {
      "epoch": 3.346300404762308,
      "grad_norm": 0.18552835285663605,
      "learning_rate": 7.86924572806816e-05,
      "loss": 1.4396,
      "step": 69390
    },
    {
      "epoch": 3.3468423459278203,
      "grad_norm": 0.19698499143123627,
      "learning_rate": 7.867899057034616e-05,
      "loss": 1.4374,
      "step": 69400
    },
    {
      "epoch": 3.3473842870933326,
      "grad_norm": 0.23695054650306702,
      "learning_rate": 7.866552092676015e-05,
      "loss": 1.4411,
      "step": 69410
    },
    {
      "epoch": 3.347926228258845,
      "grad_norm": 0.2107638120651245,
      "learning_rate": 7.865204835159217e-05,
      "loss": 1.4317,
      "step": 69420
    },
    {
      "epoch": 3.348251392958152,
      "eval_loss": 2.422180652618408,
      "eval_runtime": 29.2437,
      "eval_samples_per_second": 170.977,
      "eval_steps_per_second": 0.923,
      "step": 69426
    },
    {
      "epoch": 3.3484681694243568,
      "grad_norm": 0.24402372539043427,
      "learning_rate": 7.863857284651111e-05,
      "loss": 1.4371,
      "step": 69430
    },
    {
      "epoch": 3.349010110589869,
      "grad_norm": 0.2646619379520416,
      "learning_rate": 7.862509441318627e-05,
      "loss": 1.4186,
      "step": 69440
    },
    {
      "epoch": 3.3495520517553814,
      "grad_norm": 0.3436126708984375,
      "learning_rate": 7.861161305328733e-05,
      "loss": 1.4437,
      "step": 69450
    },
    {
      "epoch": 3.3500939929208937,
      "grad_norm": 0.1728588491678238,
      "learning_rate": 7.859812876848426e-05,
      "loss": 1.4393,
      "step": 69460
    },
    {
      "epoch": 3.3506359340864056,
      "grad_norm": 0.2518586814403534,
      "learning_rate": 7.858464156044745e-05,
      "loss": 1.4365,
      "step": 69470
    },
    {
      "epoch": 3.351177875251918,
      "grad_norm": 0.2023596614599228,
      "learning_rate": 7.857115143084763e-05,
      "loss": 1.4384,
      "step": 69480
    },
    {
      "epoch": 3.35171981641743,
      "grad_norm": 0.22805948555469513,
      "learning_rate": 7.855765838135592e-05,
      "loss": 1.4287,
      "step": 69490
    },
    {
      "epoch": 3.3522617575829425,
      "grad_norm": 0.18878723680973053,
      "learning_rate": 7.854416241364376e-05,
      "loss": 1.4451,
      "step": 69500
    },
    {
      "epoch": 3.3528036987484544,
      "grad_norm": 0.22859229147434235,
      "learning_rate": 7.8530663529383e-05,
      "loss": 1.4299,
      "step": 69510
    },
    {
      "epoch": 3.3529662810981082,
      "eval_loss": 2.420332908630371,
      "eval_runtime": 22.0276,
      "eval_samples_per_second": 226.988,
      "eval_steps_per_second": 1.226,
      "step": 69513
    },
    {
      "epoch": 3.3533456399139667,
      "grad_norm": 0.23224987089633942,
      "learning_rate": 7.851716173024578e-05,
      "loss": 1.4408,
      "step": 69520
    },
    {
      "epoch": 3.353887581079479,
      "grad_norm": 0.2047203630208969,
      "learning_rate": 7.850365701790466e-05,
      "loss": 1.4358,
      "step": 69530
    },
    {
      "epoch": 3.3544295222449914,
      "grad_norm": 0.40388110280036926,
      "learning_rate": 7.849014939403256e-05,
      "loss": 1.43,
      "step": 69540
    },
    {
      "epoch": 3.3549714634105037,
      "grad_norm": 0.21261848509311676,
      "learning_rate": 7.847663886030274e-05,
      "loss": 1.4404,
      "step": 69550
    },
    {
      "epoch": 3.3555134045760155,
      "grad_norm": 0.2009463608264923,
      "learning_rate": 7.846312541838883e-05,
      "loss": 1.4355,
      "step": 69560
    },
    {
      "epoch": 3.356055345741528,
      "grad_norm": 0.22957339882850647,
      "learning_rate": 7.844960906996481e-05,
      "loss": 1.4421,
      "step": 69570
    },
    {
      "epoch": 3.35659728690704,
      "grad_norm": 0.2667635679244995,
      "learning_rate": 7.843608981670501e-05,
      "loss": 1.4367,
      "step": 69580
    },
    {
      "epoch": 3.3571392280725525,
      "grad_norm": 0.19165495038032532,
      "learning_rate": 7.842256766028416e-05,
      "loss": 1.4375,
      "step": 69590
    },
    {
      "epoch": 3.357681169238065,
      "grad_norm": 0.2126622498035431,
      "learning_rate": 7.840904260237732e-05,
      "loss": 1.4342,
      "step": 69600
    },
    {
      "epoch": 3.357681169238065,
      "eval_loss": 2.420774459838867,
      "eval_runtime": 21.9323,
      "eval_samples_per_second": 227.974,
      "eval_steps_per_second": 1.231,
      "step": 69600
    },
    {
      "epoch": 3.3582231104035767,
      "grad_norm": 0.21864204108715057,
      "learning_rate": 7.839551464465992e-05,
      "loss": 1.4502,
      "step": 69610
    },
    {
      "epoch": 3.358765051569089,
      "grad_norm": 0.1772039234638214,
      "learning_rate": 7.838198378880772e-05,
      "loss": 1.4457,
      "step": 69620
    },
    {
      "epoch": 3.3593069927346013,
      "grad_norm": 0.16376861929893494,
      "learning_rate": 7.83684500364969e-05,
      "loss": 1.4307,
      "step": 69630
    },
    {
      "epoch": 3.3598489339001136,
      "grad_norm": 0.19408194720745087,
      "learning_rate": 7.835491338940395e-05,
      "loss": 1.4434,
      "step": 69640
    },
    {
      "epoch": 3.3603908750656255,
      "grad_norm": 0.18771915137767792,
      "learning_rate": 7.834137384920572e-05,
      "loss": 1.4245,
      "step": 69650
    },
    {
      "epoch": 3.360932816231138,
      "grad_norm": 0.1942000687122345,
      "learning_rate": 7.832783141757943e-05,
      "loss": 1.4336,
      "step": 69660
    },
    {
      "epoch": 3.36147475739665,
      "grad_norm": 0.22460921108722687,
      "learning_rate": 7.831428609620267e-05,
      "loss": 1.4373,
      "step": 69670
    },
    {
      "epoch": 3.3620166985621625,
      "grad_norm": 0.182665154337883,
      "learning_rate": 7.830073788675336e-05,
      "loss": 1.439,
      "step": 69680
    },
    {
      "epoch": 3.362396057378021,
      "eval_loss": 2.4225962162017822,
      "eval_runtime": 26.9741,
      "eval_samples_per_second": 185.363,
      "eval_steps_per_second": 1.001,
      "step": 69687
    },
    {
      "epoch": 3.3625586397276743,
      "grad_norm": 0.18876463174819946,
      "learning_rate": 7.828718679090981e-05,
      "loss": 1.4367,
      "step": 69690
    },
    {
      "epoch": 3.3631005808931866,
      "grad_norm": 0.2040254920721054,
      "learning_rate": 7.827363281035067e-05,
      "loss": 1.4358,
      "step": 69700
    },
    {
      "epoch": 3.363642522058699,
      "grad_norm": 0.375626802444458,
      "learning_rate": 7.826007594675493e-05,
      "loss": 1.436,
      "step": 69710
    },
    {
      "epoch": 3.3641844632242113,
      "grad_norm": 0.1845981627702713,
      "learning_rate": 7.824651620180196e-05,
      "loss": 1.4379,
      "step": 69720
    },
    {
      "epoch": 3.3647264043897236,
      "grad_norm": 0.1836758852005005,
      "learning_rate": 7.82329535771715e-05,
      "loss": 1.4381,
      "step": 69730
    },
    {
      "epoch": 3.3652683455552355,
      "grad_norm": 0.2422439604997635,
      "learning_rate": 7.821938807454361e-05,
      "loss": 1.4424,
      "step": 69740
    },
    {
      "epoch": 3.365810286720748,
      "grad_norm": 0.2933938205242157,
      "learning_rate": 7.820581969559877e-05,
      "loss": 1.4292,
      "step": 69750
    },
    {
      "epoch": 3.36635222788626,
      "grad_norm": 0.4664517343044281,
      "learning_rate": 7.819224844201769e-05,
      "loss": 1.4266,
      "step": 69760
    },
    {
      "epoch": 3.3668941690517724,
      "grad_norm": 0.211371049284935,
      "learning_rate": 7.817867431548158e-05,
      "loss": 1.4308,
      "step": 69770
    },
    {
      "epoch": 3.367110945517977,
      "eval_loss": 2.440751791000366,
      "eval_runtime": 22.2008,
      "eval_samples_per_second": 225.217,
      "eval_steps_per_second": 1.216,
      "step": 69774
    },
    {
      "epoch": 3.3674361102172847,
      "grad_norm": 0.39874711632728577,
      "learning_rate": 7.816509731767191e-05,
      "loss": 1.4285,
      "step": 69780
    },
    {
      "epoch": 3.3679780513827966,
      "grad_norm": 0.23130638897418976,
      "learning_rate": 7.815151745027058e-05,
      "loss": 1.4322,
      "step": 69790
    },
    {
      "epoch": 3.368519992548309,
      "grad_norm": 0.18918637931346893,
      "learning_rate": 7.81379347149598e-05,
      "loss": 1.4355,
      "step": 69800
    },
    {
      "epoch": 3.3690619337138212,
      "grad_norm": 0.27294501662254333,
      "learning_rate": 7.812434911342209e-05,
      "loss": 1.4453,
      "step": 69810
    },
    {
      "epoch": 3.3696038748793335,
      "grad_norm": 0.19342243671417236,
      "learning_rate": 7.811076064734043e-05,
      "loss": 1.4399,
      "step": 69820
    },
    {
      "epoch": 3.370145816044846,
      "grad_norm": 0.2188844233751297,
      "learning_rate": 7.809716931839804e-05,
      "loss": 1.4368,
      "step": 69830
    },
    {
      "epoch": 3.3706877572103577,
      "grad_norm": 0.19374194741249084,
      "learning_rate": 7.808357512827862e-05,
      "loss": 1.4304,
      "step": 69840
    },
    {
      "epoch": 3.37122969837587,
      "grad_norm": 0.2000492513179779,
      "learning_rate": 7.806997807866614e-05,
      "loss": 1.426,
      "step": 69850
    },
    {
      "epoch": 3.3717716395413824,
      "grad_norm": 0.2283279299736023,
      "learning_rate": 7.805637817124493e-05,
      "loss": 1.4398,
      "step": 69860
    },
    {
      "epoch": 3.3718258336579336,
      "eval_loss": 2.4315900802612305,
      "eval_runtime": 24.6052,
      "eval_samples_per_second": 203.209,
      "eval_steps_per_second": 1.097,
      "step": 69861
    },
    {
      "epoch": 3.3723135807068947,
      "grad_norm": 0.21723055839538574,
      "learning_rate": 7.804277540769967e-05,
      "loss": 1.4309,
      "step": 69870
    },
    {
      "epoch": 3.3728555218724066,
      "grad_norm": 0.19172869622707367,
      "learning_rate": 7.802916978971546e-05,
      "loss": 1.4248,
      "step": 69880
    },
    {
      "epoch": 3.373397463037919,
      "grad_norm": 0.16879266500473022,
      "learning_rate": 7.801556131897769e-05,
      "loss": 1.4338,
      "step": 69890
    },
    {
      "epoch": 3.373939404203431,
      "grad_norm": 0.3514922857284546,
      "learning_rate": 7.800194999717207e-05,
      "loss": 1.4281,
      "step": 69900
    },
    {
      "epoch": 3.3744813453689435,
      "grad_norm": 0.28702405095100403,
      "learning_rate": 7.798833582598476e-05,
      "loss": 1.4389,
      "step": 69910
    },
    {
      "epoch": 3.3750232865344554,
      "grad_norm": 0.2494763731956482,
      "learning_rate": 7.797471880710223e-05,
      "loss": 1.4392,
      "step": 69920
    },
    {
      "epoch": 3.3755652276999677,
      "grad_norm": 0.2236117422580719,
      "learning_rate": 7.796109894221127e-05,
      "loss": 1.4325,
      "step": 69930
    },
    {
      "epoch": 3.37610716886548,
      "grad_norm": 0.1991259753704071,
      "learning_rate": 7.794747623299906e-05,
      "loss": 1.4294,
      "step": 69940
    },
    {
      "epoch": 3.3765407217978898,
      "eval_loss": 2.4242312908172607,
      "eval_runtime": 22.1933,
      "eval_samples_per_second": 225.294,
      "eval_steps_per_second": 1.217,
      "step": 69948
    },
    {
      "epoch": 3.3766491100309923,
      "grad_norm": 0.3077497184276581,
      "learning_rate": 7.793385068115312e-05,
      "loss": 1.4422,
      "step": 69950
    },
    {
      "epoch": 3.3771910511965046,
      "grad_norm": 0.17531952261924744,
      "learning_rate": 7.792022228836133e-05,
      "loss": 1.4326,
      "step": 69960
    },
    {
      "epoch": 3.3777329923620165,
      "grad_norm": 0.1661709100008011,
      "learning_rate": 7.790659105631192e-05,
      "loss": 1.4322,
      "step": 69970
    },
    {
      "epoch": 3.378274933527529,
      "grad_norm": 0.21228908002376556,
      "learning_rate": 7.789295698669345e-05,
      "loss": 1.4374,
      "step": 69980
    },
    {
      "epoch": 3.378816874693041,
      "grad_norm": 0.22769136726856232,
      "learning_rate": 7.787932008119487e-05,
      "loss": 1.4357,
      "step": 69990
    },
    {
      "epoch": 3.3793588158585535,
      "grad_norm": 0.16828486323356628,
      "learning_rate": 7.786568034150545e-05,
      "loss": 1.4477,
      "step": 70000
    },
    {
      "epoch": 3.379900757024066,
      "grad_norm": 0.32244718074798584,
      "learning_rate": 7.785203776931482e-05,
      "loss": 1.4328,
      "step": 70010
    },
    {
      "epoch": 3.3804426981895777,
      "grad_norm": 0.2961633801460266,
      "learning_rate": 7.783839236631294e-05,
      "loss": 1.4321,
      "step": 70020
    },
    {
      "epoch": 3.38098463935509,
      "grad_norm": 0.1998753696680069,
      "learning_rate": 7.78247441341902e-05,
      "loss": 1.4415,
      "step": 70030
    },
    {
      "epoch": 3.3812556099378464,
      "eval_loss": 2.421795606613159,
      "eval_runtime": 24.1321,
      "eval_samples_per_second": 207.193,
      "eval_steps_per_second": 1.119,
      "step": 70035
    },
    {
      "epoch": 3.3815265805206023,
      "grad_norm": 0.16955262422561646,
      "learning_rate": 7.781109307463725e-05,
      "loss": 1.4366,
      "step": 70040
    },
    {
      "epoch": 3.3820685216861146,
      "grad_norm": 0.18801350891590118,
      "learning_rate": 7.77974391893451e-05,
      "loss": 1.4322,
      "step": 70050
    },
    {
      "epoch": 3.3826104628516265,
      "grad_norm": 0.1863052397966385,
      "learning_rate": 7.778378248000517e-05,
      "loss": 1.4334,
      "step": 70060
    },
    {
      "epoch": 3.383152404017139,
      "grad_norm": 0.3966461420059204,
      "learning_rate": 7.77701229483092e-05,
      "loss": 1.4332,
      "step": 70070
    },
    {
      "epoch": 3.383694345182651,
      "grad_norm": 0.2498437464237213,
      "learning_rate": 7.775646059594924e-05,
      "loss": 1.4432,
      "step": 70080
    },
    {
      "epoch": 3.3842362863481634,
      "grad_norm": 0.22239406406879425,
      "learning_rate": 7.774279542461776e-05,
      "loss": 1.4272,
      "step": 70090
    },
    {
      "epoch": 3.3847782275136753,
      "grad_norm": 0.26137417554855347,
      "learning_rate": 7.77291274360075e-05,
      "loss": 1.4275,
      "step": 70100
    },
    {
      "epoch": 3.3853201686791876,
      "grad_norm": 0.28286683559417725,
      "learning_rate": 7.771545663181161e-05,
      "loss": 1.4328,
      "step": 70110
    },
    {
      "epoch": 3.3858621098447,
      "grad_norm": 0.22739897668361664,
      "learning_rate": 7.770178301372361e-05,
      "loss": 1.4253,
      "step": 70120
    },
    {
      "epoch": 3.3859704980778025,
      "eval_loss": 2.4276702404022217,
      "eval_runtime": 21.9823,
      "eval_samples_per_second": 227.455,
      "eval_steps_per_second": 1.228,
      "step": 70122
    },
    {
      "epoch": 3.3864040510102122,
      "grad_norm": 0.20265692472457886,
      "learning_rate": 7.768810658343724e-05,
      "loss": 1.4306,
      "step": 70130
    },
    {
      "epoch": 3.3869459921757246,
      "grad_norm": 0.2867891490459442,
      "learning_rate": 7.767442734264677e-05,
      "loss": 1.4319,
      "step": 70140
    },
    {
      "epoch": 3.3874879333412364,
      "grad_norm": 0.1854824721813202,
      "learning_rate": 7.766074529304666e-05,
      "loss": 1.4365,
      "step": 70150
    },
    {
      "epoch": 3.3880298745067488,
      "grad_norm": 0.3422788679599762,
      "learning_rate": 7.764706043633183e-05,
      "loss": 1.4379,
      "step": 70160
    },
    {
      "epoch": 3.388571815672261,
      "grad_norm": 0.17778363823890686,
      "learning_rate": 7.763337277419745e-05,
      "loss": 1.4323,
      "step": 70170
    },
    {
      "epoch": 3.3891137568377734,
      "grad_norm": 0.1999989151954651,
      "learning_rate": 7.761968230833913e-05,
      "loss": 1.4236,
      "step": 70180
    },
    {
      "epoch": 3.3896556980032857,
      "grad_norm": 0.24541445076465607,
      "learning_rate": 7.760598904045277e-05,
      "loss": 1.4302,
      "step": 70190
    },
    {
      "epoch": 3.3901976391687976,
      "grad_norm": 0.2255592793226242,
      "learning_rate": 7.759229297223463e-05,
      "loss": 1.441,
      "step": 70200
    },
    {
      "epoch": 3.3906853862177586,
      "eval_loss": 2.4177823066711426,
      "eval_runtime": 23.5061,
      "eval_samples_per_second": 212.711,
      "eval_steps_per_second": 1.149,
      "step": 70209
    },
    {
      "epoch": 3.39073958033431,
      "grad_norm": 0.19751523435115814,
      "learning_rate": 7.757859410538131e-05,
      "loss": 1.4337,
      "step": 70210
    },
    {
      "epoch": 3.391281521499822,
      "grad_norm": 0.3200652301311493,
      "learning_rate": 7.75648924415898e-05,
      "loss": 1.4221,
      "step": 70220
    },
    {
      "epoch": 3.3918234626653345,
      "grad_norm": 0.2096845954656601,
      "learning_rate": 7.755118798255738e-05,
      "loss": 1.4365,
      "step": 70230
    },
    {
      "epoch": 3.392365403830847,
      "grad_norm": 0.2815345525741577,
      "learning_rate": 7.753748072998169e-05,
      "loss": 1.4421,
      "step": 70240
    },
    {
      "epoch": 3.3929073449963587,
      "grad_norm": 0.21538645029067993,
      "learning_rate": 7.752377068556073e-05,
      "loss": 1.4339,
      "step": 70250
    },
    {
      "epoch": 3.393449286161871,
      "grad_norm": 0.1685652881860733,
      "learning_rate": 7.751005785099286e-05,
      "loss": 1.4168,
      "step": 70260
    },
    {
      "epoch": 3.3939912273273833,
      "grad_norm": 0.16810309886932373,
      "learning_rate": 7.749634222797674e-05,
      "loss": 1.437,
      "step": 70270
    },
    {
      "epoch": 3.3945331684928957,
      "grad_norm": 0.32125574350357056,
      "learning_rate": 7.748262381821143e-05,
      "loss": 1.435,
      "step": 70280
    },
    {
      "epoch": 3.3950751096584075,
      "grad_norm": 0.2670728266239166,
      "learning_rate": 7.746890262339627e-05,
      "loss": 1.4457,
      "step": 70290
    },
    {
      "epoch": 3.395400274357715,
      "eval_loss": 2.4280776977539062,
      "eval_runtime": 25.8886,
      "eval_samples_per_second": 193.135,
      "eval_steps_per_second": 1.043,
      "step": 70296
    },
    {
      "epoch": 3.39561705082392,
      "grad_norm": 0.3197952210903168,
      "learning_rate": 7.745517864523102e-05,
      "loss": 1.4286,
      "step": 70300
    },
    {
      "epoch": 3.396158991989432,
      "grad_norm": 0.19565744698047638,
      "learning_rate": 7.744145188541573e-05,
      "loss": 1.4334,
      "step": 70310
    },
    {
      "epoch": 3.3967009331549445,
      "grad_norm": 0.22424980998039246,
      "learning_rate": 7.742772234565081e-05,
      "loss": 1.432,
      "step": 70320
    },
    {
      "epoch": 3.3972428743204564,
      "grad_norm": 0.27934402227401733,
      "learning_rate": 7.741399002763702e-05,
      "loss": 1.4402,
      "step": 70330
    },
    {
      "epoch": 3.3977848154859687,
      "grad_norm": 0.16731417179107666,
      "learning_rate": 7.740025493307543e-05,
      "loss": 1.4352,
      "step": 70340
    },
    {
      "epoch": 3.398326756651481,
      "grad_norm": 0.24355627596378326,
      "learning_rate": 7.738651706366754e-05,
      "loss": 1.4329,
      "step": 70350
    },
    {
      "epoch": 3.3988686978169933,
      "grad_norm": 0.1672547459602356,
      "learning_rate": 7.73727764211151e-05,
      "loss": 1.4408,
      "step": 70360
    },
    {
      "epoch": 3.3994106389825056,
      "grad_norm": 0.21636275947093964,
      "learning_rate": 7.735903300712025e-05,
      "loss": 1.4255,
      "step": 70370
    },
    {
      "epoch": 3.3999525801480175,
      "grad_norm": 0.21118460595607758,
      "learning_rate": 7.734528682338546e-05,
      "loss": 1.4435,
      "step": 70380
    },
    {
      "epoch": 3.4001151624976713,
      "eval_loss": 2.429720163345337,
      "eval_runtime": 24.7724,
      "eval_samples_per_second": 201.837,
      "eval_steps_per_second": 1.09,
      "step": 70383
    },
    {
      "epoch": 3.40049452131353,
      "grad_norm": 0.30819687247276306,
      "learning_rate": 7.733153787161356e-05,
      "loss": 1.4255,
      "step": 70390
    },
    {
      "epoch": 3.401036462479042,
      "grad_norm": 0.20422889292240143,
      "learning_rate": 7.73177861535077e-05,
      "loss": 1.4364,
      "step": 70400
    },
    {
      "epoch": 3.4015784036445544,
      "grad_norm": 0.3542598783969879,
      "learning_rate": 7.73040316707714e-05,
      "loss": 1.4333,
      "step": 70410
    },
    {
      "epoch": 3.4021203448100668,
      "grad_norm": 0.20476211607456207,
      "learning_rate": 7.729027442510847e-05,
      "loss": 1.4459,
      "step": 70420
    },
    {
      "epoch": 3.4026622859755786,
      "grad_norm": 0.22746922075748444,
      "learning_rate": 7.727651441822312e-05,
      "loss": 1.4267,
      "step": 70430
    },
    {
      "epoch": 3.403204227141091,
      "grad_norm": 0.23665057122707367,
      "learning_rate": 7.726275165181988e-05,
      "loss": 1.4302,
      "step": 70440
    },
    {
      "epoch": 3.4037461683066033,
      "grad_norm": 0.2850460112094879,
      "learning_rate": 7.724898612760362e-05,
      "loss": 1.4315,
      "step": 70450
    },
    {
      "epoch": 3.4042881094721156,
      "grad_norm": 0.17364229261875153,
      "learning_rate": 7.723521784727956e-05,
      "loss": 1.4364,
      "step": 70460
    },
    {
      "epoch": 3.404830050637628,
      "grad_norm": 0.21966204047203064,
      "learning_rate": 7.722144681255325e-05,
      "loss": 1.4284,
      "step": 70470
    },
    {
      "epoch": 3.404830050637628,
      "eval_loss": 2.4222443103790283,
      "eval_runtime": 21.8336,
      "eval_samples_per_second": 229.005,
      "eval_steps_per_second": 1.237,
      "step": 70470
    },
    {
      "epoch": 3.4053719918031398,
      "grad_norm": 0.23222704231739044,
      "learning_rate": 7.720767302513059e-05,
      "loss": 1.4334,
      "step": 70480
    },
    {
      "epoch": 3.405913932968652,
      "grad_norm": 0.19313640892505646,
      "learning_rate": 7.719389648671779e-05,
      "loss": 1.4361,
      "step": 70490
    },
    {
      "epoch": 3.4064558741341644,
      "grad_norm": 0.17659373581409454,
      "learning_rate": 7.718011719902146e-05,
      "loss": 1.4402,
      "step": 70500
    },
    {
      "epoch": 3.4069978152996767,
      "grad_norm": 0.2190655916929245,
      "learning_rate": 7.716633516374852e-05,
      "loss": 1.4326,
      "step": 70510
    },
    {
      "epoch": 3.4075397564651886,
      "grad_norm": 0.2351011484861374,
      "learning_rate": 7.715255038260621e-05,
      "loss": 1.4198,
      "step": 70520
    },
    {
      "epoch": 3.408081697630701,
      "grad_norm": 0.2913278341293335,
      "learning_rate": 7.713876285730213e-05,
      "loss": 1.4363,
      "step": 70530
    },
    {
      "epoch": 3.408623638796213,
      "grad_norm": 0.21653717756271362,
      "learning_rate": 7.712497258954422e-05,
      "loss": 1.4284,
      "step": 70540
    },
    {
      "epoch": 3.4091655799617255,
      "grad_norm": 0.2118827849626541,
      "learning_rate": 7.711117958104077e-05,
      "loss": 1.4306,
      "step": 70550
    },
    {
      "epoch": 3.409544938777584,
      "eval_loss": 2.4244723320007324,
      "eval_runtime": 23.3419,
      "eval_samples_per_second": 214.207,
      "eval_steps_per_second": 1.157,
      "step": 70557
    },
    {
      "epoch": 3.4097075211272374,
      "grad_norm": 0.3481065630912781,
      "learning_rate": 7.70973838335004e-05,
      "loss": 1.4373,
      "step": 70560
    },
    {
      "epoch": 3.4102494622927497,
      "grad_norm": 0.16835106909275055,
      "learning_rate": 7.708358534863205e-05,
      "loss": 1.433,
      "step": 70570
    },
    {
      "epoch": 3.410791403458262,
      "grad_norm": 0.3342517912387848,
      "learning_rate": 7.706978412814501e-05,
      "loss": 1.4379,
      "step": 70580
    },
    {
      "epoch": 3.4113333446237744,
      "grad_norm": 0.19866886734962463,
      "learning_rate": 7.705598017374894e-05,
      "loss": 1.4306,
      "step": 70590
    },
    {
      "epoch": 3.4118752857892867,
      "grad_norm": 0.29506245255470276,
      "learning_rate": 7.704217348715381e-05,
      "loss": 1.4344,
      "step": 70600
    },
    {
      "epoch": 3.4124172269547985,
      "grad_norm": 0.2535003125667572,
      "learning_rate": 7.702836407006993e-05,
      "loss": 1.441,
      "step": 70610
    },
    {
      "epoch": 3.412959168120311,
      "grad_norm": 0.23304949700832367,
      "learning_rate": 7.701455192420793e-05,
      "loss": 1.429,
      "step": 70620
    },
    {
      "epoch": 3.413501109285823,
      "grad_norm": 0.2439218908548355,
      "learning_rate": 7.700073705127883e-05,
      "loss": 1.4379,
      "step": 70630
    },
    {
      "epoch": 3.4140430504513355,
      "grad_norm": 0.2127150297164917,
      "learning_rate": 7.698691945299392e-05,
      "loss": 1.4366,
      "step": 70640
    },
    {
      "epoch": 3.41425982691754,
      "eval_loss": 2.4175617694854736,
      "eval_runtime": 21.9871,
      "eval_samples_per_second": 227.406,
      "eval_steps_per_second": 1.228,
      "step": 70644
    },
    {
      "epoch": 3.414584991616848,
      "grad_norm": 0.17583894729614258,
      "learning_rate": 7.697309913106491e-05,
      "loss": 1.4296,
      "step": 70650
    },
    {
      "epoch": 3.4151269327823597,
      "grad_norm": 0.1831618696451187,
      "learning_rate": 7.695927608720376e-05,
      "loss": 1.435,
      "step": 70660
    },
    {
      "epoch": 3.415668873947872,
      "grad_norm": 0.22078917920589447,
      "learning_rate": 7.694545032312284e-05,
      "loss": 1.4256,
      "step": 70670
    },
    {
      "epoch": 3.4162108151133843,
      "grad_norm": 0.19926197826862335,
      "learning_rate": 7.69316218405348e-05,
      "loss": 1.4341,
      "step": 70680
    },
    {
      "epoch": 3.4167527562788966,
      "grad_norm": 0.20558814704418182,
      "learning_rate": 7.691779064115267e-05,
      "loss": 1.4258,
      "step": 70690
    },
    {
      "epoch": 3.4172946974444085,
      "grad_norm": 0.1772947460412979,
      "learning_rate": 7.690395672668979e-05,
      "loss": 1.4311,
      "step": 70700
    },
    {
      "epoch": 3.417836638609921,
      "grad_norm": 0.29191502928733826,
      "learning_rate": 7.689012009885986e-05,
      "loss": 1.4376,
      "step": 70710
    },
    {
      "epoch": 3.418378579775433,
      "grad_norm": 0.21399341523647308,
      "learning_rate": 7.687628075937689e-05,
      "loss": 1.4347,
      "step": 70720
    },
    {
      "epoch": 3.4189205209409455,
      "grad_norm": 0.2068105787038803,
      "learning_rate": 7.686243870995522e-05,
      "loss": 1.4269,
      "step": 70730
    },
    {
      "epoch": 3.4189747150574967,
      "eval_loss": 2.414785861968994,
      "eval_runtime": 22.7197,
      "eval_samples_per_second": 220.073,
      "eval_steps_per_second": 1.188,
      "step": 70731
    },
    {
      "epoch": 3.4194624621064573,
      "grad_norm": 0.22110241651535034,
      "learning_rate": 7.684859395230956e-05,
      "loss": 1.4331,
      "step": 70740
    },
    {
      "epoch": 3.4200044032719696,
      "grad_norm": 0.1638895571231842,
      "learning_rate": 7.683474648815496e-05,
      "loss": 1.419,
      "step": 70750
    },
    {
      "epoch": 3.420546344437482,
      "grad_norm": 0.20719309151172638,
      "learning_rate": 7.682089631920674e-05,
      "loss": 1.4246,
      "step": 70760
    },
    {
      "epoch": 3.4210882856029943,
      "grad_norm": 0.19159828126430511,
      "learning_rate": 7.680704344718063e-05,
      "loss": 1.4245,
      "step": 70770
    },
    {
      "epoch": 3.4216302267685066,
      "grad_norm": 0.29333052039146423,
      "learning_rate": 7.679318787379264e-05,
      "loss": 1.4397,
      "step": 70780
    },
    {
      "epoch": 3.4221721679340185,
      "grad_norm": 0.22254757583141327,
      "learning_rate": 7.677932960075917e-05,
      "loss": 1.4231,
      "step": 70790
    },
    {
      "epoch": 3.4227141090995308,
      "grad_norm": 0.2216396927833557,
      "learning_rate": 7.67654686297969e-05,
      "loss": 1.433,
      "step": 70800
    },
    {
      "epoch": 3.423256050265043,
      "grad_norm": 0.27683907747268677,
      "learning_rate": 7.675160496262288e-05,
      "loss": 1.4404,
      "step": 70810
    },
    {
      "epoch": 3.423689603197453,
      "eval_loss": 2.416546583175659,
      "eval_runtime": 21.9857,
      "eval_samples_per_second": 227.421,
      "eval_steps_per_second": 1.228,
      "step": 70818
    },
    {
      "epoch": 3.4237979914305554,
      "grad_norm": 0.3137832581996918,
      "learning_rate": 7.673773860095445e-05,
      "loss": 1.434,
      "step": 70820
    },
    {
      "epoch": 3.4243399325960677,
      "grad_norm": 0.1663368195295334,
      "learning_rate": 7.672386954650936e-05,
      "loss": 1.427,
      "step": 70830
    },
    {
      "epoch": 3.4248818737615796,
      "grad_norm": 0.18768776953220367,
      "learning_rate": 7.670999780100563e-05,
      "loss": 1.4314,
      "step": 70840
    },
    {
      "epoch": 3.425423814927092,
      "grad_norm": 0.19177640974521637,
      "learning_rate": 7.669612336616162e-05,
      "loss": 1.4275,
      "step": 70850
    },
    {
      "epoch": 3.4259657560926042,
      "grad_norm": 0.2142564207315445,
      "learning_rate": 7.668224624369603e-05,
      "loss": 1.4467,
      "step": 70860
    },
    {
      "epoch": 3.4265076972581165,
      "grad_norm": 0.24929867684841156,
      "learning_rate": 7.666836643532793e-05,
      "loss": 1.4374,
      "step": 70870
    },
    {
      "epoch": 3.427049638423629,
      "grad_norm": 0.20959332585334778,
      "learning_rate": 7.665448394277664e-05,
      "loss": 1.4289,
      "step": 70880
    },
    {
      "epoch": 3.4275915795891407,
      "grad_norm": 0.23564212024211884,
      "learning_rate": 7.664059876776195e-05,
      "loss": 1.4368,
      "step": 70890
    },
    {
      "epoch": 3.428133520754653,
      "grad_norm": 0.29305994510650635,
      "learning_rate": 7.662671091200378e-05,
      "loss": 1.4361,
      "step": 70900
    },
    {
      "epoch": 3.4284044913374094,
      "eval_loss": 2.4167122840881348,
      "eval_runtime": 22.6755,
      "eval_samples_per_second": 220.502,
      "eval_steps_per_second": 1.191,
      "step": 70905
    },
    {
      "epoch": 3.4286754619201654,
      "grad_norm": 0.19599109888076782,
      "learning_rate": 7.66128203772226e-05,
      "loss": 1.4286,
      "step": 70910
    },
    {
      "epoch": 3.4292174030856777,
      "grad_norm": 0.2723342478275299,
      "learning_rate": 7.659892716513904e-05,
      "loss": 1.4253,
      "step": 70920
    },
    {
      "epoch": 3.4297593442511896,
      "grad_norm": 0.25713953375816345,
      "learning_rate": 7.658503127747415e-05,
      "loss": 1.4388,
      "step": 70930
    },
    {
      "epoch": 3.430301285416702,
      "grad_norm": 0.24171973764896393,
      "learning_rate": 7.657113271594931e-05,
      "loss": 1.4361,
      "step": 70940
    },
    {
      "epoch": 3.430843226582214,
      "grad_norm": 0.17661350965499878,
      "learning_rate": 7.65572314822862e-05,
      "loss": 1.4331,
      "step": 70950
    },
    {
      "epoch": 3.4313851677477265,
      "grad_norm": 0.23326051235198975,
      "learning_rate": 7.654332757820684e-05,
      "loss": 1.4239,
      "step": 70960
    },
    {
      "epoch": 3.4319271089132384,
      "grad_norm": 0.1615256667137146,
      "learning_rate": 7.652942100543361e-05,
      "loss": 1.4351,
      "step": 70970
    },
    {
      "epoch": 3.4324690500787507,
      "grad_norm": 0.19447456300258636,
      "learning_rate": 7.651551176568916e-05,
      "loss": 1.4277,
      "step": 70980
    },
    {
      "epoch": 3.433010991244263,
      "grad_norm": 0.3392447829246521,
      "learning_rate": 7.650159986069653e-05,
      "loss": 1.4348,
      "step": 70990
    },
    {
      "epoch": 3.4331193794773656,
      "eval_loss": 2.42305850982666,
      "eval_runtime": 22.2077,
      "eval_samples_per_second": 225.147,
      "eval_steps_per_second": 1.216,
      "step": 70992
    },
    {
      "epoch": 3.4335529324097753,
      "grad_norm": 0.2070346623659134,
      "learning_rate": 7.648768529217907e-05,
      "loss": 1.4279,
      "step": 71000
    },
    {
      "epoch": 3.4340948735752876,
      "grad_norm": 0.2491535097360611,
      "learning_rate": 7.647376806186043e-05,
      "loss": 1.4386,
      "step": 71010
    },
    {
      "epoch": 3.4346368147407995,
      "grad_norm": 0.3890012502670288,
      "learning_rate": 7.645984817146464e-05,
      "loss": 1.429,
      "step": 71020
    },
    {
      "epoch": 3.435178755906312,
      "grad_norm": 0.19839666783809662,
      "learning_rate": 7.644592562271603e-05,
      "loss": 1.4271,
      "step": 71030
    },
    {
      "epoch": 3.435720697071824,
      "grad_norm": 0.24999631941318512,
      "learning_rate": 7.643200041733926e-05,
      "loss": 1.4357,
      "step": 71040
    },
    {
      "epoch": 3.4362626382373365,
      "grad_norm": 0.18307673931121826,
      "learning_rate": 7.641807255705932e-05,
      "loss": 1.4381,
      "step": 71050
    },
    {
      "epoch": 3.436804579402849,
      "grad_norm": 0.3571625351905823,
      "learning_rate": 7.640414204360154e-05,
      "loss": 1.4406,
      "step": 71060
    },
    {
      "epoch": 3.4373465205683607,
      "grad_norm": 0.23355412483215332,
      "learning_rate": 7.639020887869157e-05,
      "loss": 1.4392,
      "step": 71070
    },
    {
      "epoch": 3.4378342676173217,
      "eval_loss": 2.4152307510375977,
      "eval_runtime": 22.2007,
      "eval_samples_per_second": 225.218,
      "eval_steps_per_second": 1.216,
      "step": 71079
    },
    {
      "epoch": 3.437888461733873,
      "grad_norm": 0.20043282210826874,
      "learning_rate": 7.63762730640554e-05,
      "loss": 1.4451,
      "step": 71080
    },
    {
      "epoch": 3.4384304028993853,
      "grad_norm": 0.19627268612384796,
      "learning_rate": 7.636233460141934e-05,
      "loss": 1.4379,
      "step": 71090
    },
    {
      "epoch": 3.4389723440648976,
      "grad_norm": 0.24721965193748474,
      "learning_rate": 7.634839349251e-05,
      "loss": 1.4345,
      "step": 71100
    },
    {
      "epoch": 3.43951428523041,
      "grad_norm": 0.2560265362262726,
      "learning_rate": 7.633444973905435e-05,
      "loss": 1.4404,
      "step": 71110
    },
    {
      "epoch": 3.440056226395922,
      "grad_norm": 0.21362638473510742,
      "learning_rate": 7.63205033427797e-05,
      "loss": 1.4416,
      "step": 71120
    },
    {
      "epoch": 3.440598167561434,
      "grad_norm": 0.24023322761058807,
      "learning_rate": 7.630655430541367e-05,
      "loss": 1.4308,
      "step": 71130
    },
    {
      "epoch": 3.4411401087269464,
      "grad_norm": 0.20237214863300323,
      "learning_rate": 7.629260262868419e-05,
      "loss": 1.4432,
      "step": 71140
    },
    {
      "epoch": 3.4416820498924587,
      "grad_norm": 0.18272891640663147,
      "learning_rate": 7.627864831431955e-05,
      "loss": 1.4389,
      "step": 71150
    },
    {
      "epoch": 3.4422239910579706,
      "grad_norm": 0.2650783061981201,
      "learning_rate": 7.626469136404834e-05,
      "loss": 1.439,
      "step": 71160
    },
    {
      "epoch": 3.4425491557572783,
      "eval_loss": 2.4182510375976562,
      "eval_runtime": 22.3053,
      "eval_samples_per_second": 224.162,
      "eval_steps_per_second": 1.21,
      "step": 71166
    },
    {
      "epoch": 3.442765932223483,
      "grad_norm": 0.2870977222919464,
      "learning_rate": 7.625073177959945e-05,
      "loss": 1.4253,
      "step": 71170
    },
    {
      "epoch": 3.4433078733889952,
      "grad_norm": 0.20810657739639282,
      "learning_rate": 7.62367695627022e-05,
      "loss": 1.4484,
      "step": 71180
    },
    {
      "epoch": 3.4438498145545076,
      "grad_norm": 0.177906796336174,
      "learning_rate": 7.622280471508611e-05,
      "loss": 1.441,
      "step": 71190
    },
    {
      "epoch": 3.4443917557200194,
      "grad_norm": 0.17265500128269196,
      "learning_rate": 7.620883723848114e-05,
      "loss": 1.4263,
      "step": 71200
    },
    {
      "epoch": 3.4449336968855317,
      "grad_norm": 0.23366230726242065,
      "learning_rate": 7.61948671346175e-05,
      "loss": 1.4195,
      "step": 71210
    },
    {
      "epoch": 3.445475638051044,
      "grad_norm": 0.18871016800403595,
      "learning_rate": 7.618089440522571e-05,
      "loss": 1.4275,
      "step": 71220
    },
    {
      "epoch": 3.4460175792165564,
      "grad_norm": 0.1917314976453781,
      "learning_rate": 7.61669190520367e-05,
      "loss": 1.4325,
      "step": 71230
    },
    {
      "epoch": 3.4465595203820687,
      "grad_norm": 0.2467162013053894,
      "learning_rate": 7.615294107678165e-05,
      "loss": 1.4374,
      "step": 71240
    },
    {
      "epoch": 3.4471014615475806,
      "grad_norm": 0.1812368631362915,
      "learning_rate": 7.61389604811921e-05,
      "loss": 1.4434,
      "step": 71250
    },
    {
      "epoch": 3.4472640438972344,
      "eval_loss": 2.420837163925171,
      "eval_runtime": 23.1164,
      "eval_samples_per_second": 216.296,
      "eval_steps_per_second": 1.168,
      "step": 71253
    },
    {
      "epoch": 3.447643402713093,
      "grad_norm": 0.24200530350208282,
      "learning_rate": 7.61249772669999e-05,
      "loss": 1.4313,
      "step": 71260
    },
    {
      "epoch": 3.448185343878605,
      "grad_norm": 0.19202779233455658,
      "learning_rate": 7.61109914359372e-05,
      "loss": 1.4212,
      "step": 71270
    },
    {
      "epoch": 3.4487272850441175,
      "grad_norm": 0.34746161103248596,
      "learning_rate": 7.609700298973659e-05,
      "loss": 1.4312,
      "step": 71280
    },
    {
      "epoch": 3.44926922620963,
      "grad_norm": 0.19444984197616577,
      "learning_rate": 7.608301193013082e-05,
      "loss": 1.416,
      "step": 71290
    },
    {
      "epoch": 3.4498111673751417,
      "grad_norm": 0.17890185117721558,
      "learning_rate": 7.606901825885305e-05,
      "loss": 1.431,
      "step": 71300
    },
    {
      "epoch": 3.450353108540654,
      "grad_norm": 0.1906130313873291,
      "learning_rate": 7.605502197763678e-05,
      "loss": 1.4435,
      "step": 71310
    },
    {
      "epoch": 3.4508950497061663,
      "grad_norm": 0.2459091991186142,
      "learning_rate": 7.60410230882158e-05,
      "loss": 1.4252,
      "step": 71320
    },
    {
      "epoch": 3.4514369908716787,
      "grad_norm": 0.3785655200481415,
      "learning_rate": 7.602702159232424e-05,
      "loss": 1.4277,
      "step": 71330
    },
    {
      "epoch": 3.4519789320371905,
      "grad_norm": 0.2577289044857025,
      "learning_rate": 7.601301749169652e-05,
      "loss": 1.429,
      "step": 71340
    },
    {
      "epoch": 3.4519789320371905,
      "eval_loss": 2.4179348945617676,
      "eval_runtime": 21.9382,
      "eval_samples_per_second": 227.913,
      "eval_steps_per_second": 1.231,
      "step": 71340
    },
    {
      "epoch": 3.452520873202703,
      "grad_norm": 0.2082802802324295,
      "learning_rate": 7.599901078806744e-05,
      "loss": 1.4214,
      "step": 71350
    },
    {
      "epoch": 3.453062814368215,
      "grad_norm": 0.251081645488739,
      "learning_rate": 7.598500148317206e-05,
      "loss": 1.4394,
      "step": 71360
    },
    {
      "epoch": 3.4536047555337275,
      "grad_norm": 0.1895270198583603,
      "learning_rate": 7.597098957874582e-05,
      "loss": 1.4283,
      "step": 71370
    },
    {
      "epoch": 3.4541466966992393,
      "grad_norm": 0.252121239900589,
      "learning_rate": 7.595697507652442e-05,
      "loss": 1.4305,
      "step": 71380
    },
    {
      "epoch": 3.4546886378647517,
      "grad_norm": 0.32456740736961365,
      "learning_rate": 7.594295797824393e-05,
      "loss": 1.4383,
      "step": 71390
    },
    {
      "epoch": 3.455230579030264,
      "grad_norm": 0.22235190868377686,
      "learning_rate": 7.592893828564073e-05,
      "loss": 1.4284,
      "step": 71400
    },
    {
      "epoch": 3.4557725201957763,
      "grad_norm": 0.19792044162750244,
      "learning_rate": 7.591491600045155e-05,
      "loss": 1.4249,
      "step": 71410
    },
    {
      "epoch": 3.4563144613612886,
      "grad_norm": 0.26035961508750916,
      "learning_rate": 7.590089112441336e-05,
      "loss": 1.4245,
      "step": 71420
    },
    {
      "epoch": 3.456693820177147,
      "eval_loss": 2.4162979125976562,
      "eval_runtime": 21.987,
      "eval_samples_per_second": 227.408,
      "eval_steps_per_second": 1.228,
      "step": 71427
    },
    {
      "epoch": 3.4568564025268005,
      "grad_norm": 0.18416984379291534,
      "learning_rate": 7.58868636592635e-05,
      "loss": 1.4349,
      "step": 71430
    },
    {
      "epoch": 3.457398343692313,
      "grad_norm": 0.21309533715248108,
      "learning_rate": 7.587283360673969e-05,
      "loss": 1.4194,
      "step": 71440
    },
    {
      "epoch": 3.457940284857825,
      "grad_norm": 0.17019227147102356,
      "learning_rate": 7.585880096857985e-05,
      "loss": 1.4342,
      "step": 71450
    },
    {
      "epoch": 3.4584822260233374,
      "grad_norm": 0.16507531702518463,
      "learning_rate": 7.58447657465223e-05,
      "loss": 1.424,
      "step": 71460
    },
    {
      "epoch": 3.4590241671888498,
      "grad_norm": 0.21044619381427765,
      "learning_rate": 7.583072794230567e-05,
      "loss": 1.428,
      "step": 71470
    },
    {
      "epoch": 3.4595661083543616,
      "grad_norm": 0.2250647246837616,
      "learning_rate": 7.581668755766891e-05,
      "loss": 1.4316,
      "step": 71480
    },
    {
      "epoch": 3.460108049519874,
      "grad_norm": 0.21987579762935638,
      "learning_rate": 7.580264459435129e-05,
      "loss": 1.4411,
      "step": 71490
    },
    {
      "epoch": 3.4606499906853863,
      "grad_norm": 0.21779508888721466,
      "learning_rate": 7.578859905409234e-05,
      "loss": 1.4381,
      "step": 71500
    },
    {
      "epoch": 3.4611919318508986,
      "grad_norm": 0.3549853265285492,
      "learning_rate": 7.577455093863202e-05,
      "loss": 1.4346,
      "step": 71510
    },
    {
      "epoch": 3.4614087083171032,
      "eval_loss": 2.4182517528533936,
      "eval_runtime": 21.9784,
      "eval_samples_per_second": 227.496,
      "eval_steps_per_second": 1.228,
      "step": 71514
    },
    {
      "epoch": 3.461733873016411,
      "grad_norm": 0.46638888120651245,
      "learning_rate": 7.57605002497105e-05,
      "loss": 1.4274,
      "step": 71520
    },
    {
      "epoch": 3.4622758141819228,
      "grad_norm": 0.28370195627212524,
      "learning_rate": 7.574644698906836e-05,
      "loss": 1.4282,
      "step": 71530
    },
    {
      "epoch": 3.462817755347435,
      "grad_norm": 0.2209625244140625,
      "learning_rate": 7.573239115844644e-05,
      "loss": 1.4407,
      "step": 71540
    },
    {
      "epoch": 3.4633596965129474,
      "grad_norm": 0.2586290240287781,
      "learning_rate": 7.571833275958591e-05,
      "loss": 1.437,
      "step": 71550
    },
    {
      "epoch": 3.4639016376784597,
      "grad_norm": 0.17665983736515045,
      "learning_rate": 7.570427179422827e-05,
      "loss": 1.4312,
      "step": 71560
    },
    {
      "epoch": 3.4644435788439716,
      "grad_norm": 0.1650412678718567,
      "learning_rate": 7.569020826411532e-05,
      "loss": 1.4262,
      "step": 71570
    },
    {
      "epoch": 3.464985520009484,
      "grad_norm": 0.19921010732650757,
      "learning_rate": 7.56761421709892e-05,
      "loss": 1.4333,
      "step": 71580
    },
    {
      "epoch": 3.465527461174996,
      "grad_norm": 0.2843552827835083,
      "learning_rate": 7.566207351659236e-05,
      "loss": 1.4295,
      "step": 71590
    },
    {
      "epoch": 3.4660694023405085,
      "grad_norm": 0.3298152983188629,
      "learning_rate": 7.564800230266756e-05,
      "loss": 1.4246,
      "step": 71600
    },
    {
      "epoch": 3.46612359645706,
      "eval_loss": 2.4160807132720947,
      "eval_runtime": 21.9823,
      "eval_samples_per_second": 227.456,
      "eval_steps_per_second": 1.228,
      "step": 71601
    },
    {
      "epoch": 3.4666113435060204,
      "grad_norm": 0.41355177760124207,
      "learning_rate": 7.563392853095786e-05,
      "loss": 1.4208,
      "step": 71610
    },
    {
      "epoch": 3.4671532846715327,
      "grad_norm": 0.25345563888549805,
      "learning_rate": 7.56198522032067e-05,
      "loss": 1.4262,
      "step": 71620
    },
    {
      "epoch": 3.467695225837045,
      "grad_norm": 0.2895798981189728,
      "learning_rate": 7.560577332115777e-05,
      "loss": 1.4344,
      "step": 71630
    },
    {
      "epoch": 3.4682371670025574,
      "grad_norm": 0.18971815705299377,
      "learning_rate": 7.559169188655509e-05,
      "loss": 1.4359,
      "step": 71640
    },
    {
      "epoch": 3.4687791081680697,
      "grad_norm": 0.24166151881217957,
      "learning_rate": 7.557760790114304e-05,
      "loss": 1.4282,
      "step": 71650
    },
    {
      "epoch": 3.4693210493335815,
      "grad_norm": 0.18398047983646393,
      "learning_rate": 7.556352136666624e-05,
      "loss": 1.4319,
      "step": 71660
    },
    {
      "epoch": 3.469862990499094,
      "grad_norm": 0.1882873922586441,
      "learning_rate": 7.554943228486969e-05,
      "loss": 1.4305,
      "step": 71670
    },
    {
      "epoch": 3.470404931664606,
      "grad_norm": 0.39805862307548523,
      "learning_rate": 7.55353406574987e-05,
      "loss": 1.4383,
      "step": 71680
    },
    {
      "epoch": 3.470838484597016,
      "eval_loss": 2.422006607055664,
      "eval_runtime": 21.9837,
      "eval_samples_per_second": 227.442,
      "eval_steps_per_second": 1.228,
      "step": 71688
    },
    {
      "epoch": 3.4709468728301185,
      "grad_norm": 0.4627326726913452,
      "learning_rate": 7.552124648629887e-05,
      "loss": 1.423,
      "step": 71690
    },
    {
      "epoch": 3.471488813995631,
      "grad_norm": 0.20239268243312836,
      "learning_rate": 7.55071497730161e-05,
      "loss": 1.4374,
      "step": 71700
    },
    {
      "epoch": 3.4720307551611427,
      "grad_norm": 0.21879947185516357,
      "learning_rate": 7.549305051939665e-05,
      "loss": 1.4308,
      "step": 71710
    },
    {
      "epoch": 3.472572696326655,
      "grad_norm": 0.2617088854312897,
      "learning_rate": 7.547894872718709e-05,
      "loss": 1.4352,
      "step": 71720
    },
    {
      "epoch": 3.4731146374921673,
      "grad_norm": 0.24820886552333832,
      "learning_rate": 7.546484439813427e-05,
      "loss": 1.4237,
      "step": 71730
    },
    {
      "epoch": 3.4736565786576796,
      "grad_norm": 0.22648629546165466,
      "learning_rate": 7.545073753398537e-05,
      "loss": 1.435,
      "step": 71740
    },
    {
      "epoch": 3.474198519823192,
      "grad_norm": 0.20121285319328308,
      "learning_rate": 7.543662813648789e-05,
      "loss": 1.4264,
      "step": 71750
    },
    {
      "epoch": 3.474740460988704,
      "grad_norm": 0.32857656478881836,
      "learning_rate": 7.542251620738964e-05,
      "loss": 1.4422,
      "step": 71760
    },
    {
      "epoch": 3.475282402154216,
      "grad_norm": 0.1786399483680725,
      "learning_rate": 7.540840174843876e-05,
      "loss": 1.4308,
      "step": 71770
    },
    {
      "epoch": 3.475553372736972,
      "eval_loss": 2.418339490890503,
      "eval_runtime": 21.9871,
      "eval_samples_per_second": 227.406,
      "eval_steps_per_second": 1.228,
      "step": 71775
    },
    {
      "epoch": 3.4758243433197284,
      "grad_norm": 0.202467679977417,
      "learning_rate": 7.539428476138367e-05,
      "loss": 1.4229,
      "step": 71780
    },
    {
      "epoch": 3.4763662844852408,
      "grad_norm": 0.18343421816825867,
      "learning_rate": 7.538016524797313e-05,
      "loss": 1.4268,
      "step": 71790
    },
    {
      "epoch": 3.4769082256507526,
      "grad_norm": 0.3386618494987488,
      "learning_rate": 7.53660432099562e-05,
      "loss": 1.4319,
      "step": 71800
    },
    {
      "epoch": 3.477450166816265,
      "grad_norm": 0.2889906167984009,
      "learning_rate": 7.535191864908224e-05,
      "loss": 1.4279,
      "step": 71810
    },
    {
      "epoch": 3.4779921079817773,
      "grad_norm": 0.25574788451194763,
      "learning_rate": 7.533779156710098e-05,
      "loss": 1.4254,
      "step": 71820
    },
    {
      "epoch": 3.4785340491472896,
      "grad_norm": 0.282247930765152,
      "learning_rate": 7.532366196576238e-05,
      "loss": 1.4315,
      "step": 71830
    },
    {
      "epoch": 3.4790759903128015,
      "grad_norm": 0.22523202002048492,
      "learning_rate": 7.530952984681679e-05,
      "loss": 1.4407,
      "step": 71840
    },
    {
      "epoch": 3.4796179314783138,
      "grad_norm": 0.2516654431819916,
      "learning_rate": 7.529539521201481e-05,
      "loss": 1.4203,
      "step": 71850
    },
    {
      "epoch": 3.480159872643826,
      "grad_norm": 0.19450673460960388,
      "learning_rate": 7.528125806310737e-05,
      "loss": 1.4341,
      "step": 71860
    },
    {
      "epoch": 3.4802682608769286,
      "eval_loss": 2.420518636703491,
      "eval_runtime": 21.9826,
      "eval_samples_per_second": 227.453,
      "eval_steps_per_second": 1.228,
      "step": 71862
    },
    {
      "epoch": 3.4807018138093384,
      "grad_norm": 0.17963695526123047,
      "learning_rate": 7.526711840184576e-05,
      "loss": 1.4376,
      "step": 71870
    },
    {
      "epoch": 3.4812437549748507,
      "grad_norm": 0.1823495328426361,
      "learning_rate": 7.525297622998151e-05,
      "loss": 1.4332,
      "step": 71880
    },
    {
      "epoch": 3.4817856961403626,
      "grad_norm": 0.18945930898189545,
      "learning_rate": 7.523883154926648e-05,
      "loss": 1.4278,
      "step": 71890
    },
    {
      "epoch": 3.482327637305875,
      "grad_norm": 0.20653122663497925,
      "learning_rate": 7.522468436145288e-05,
      "loss": 1.4371,
      "step": 71900
    },
    {
      "epoch": 3.4828695784713872,
      "grad_norm": 0.17666640877723694,
      "learning_rate": 7.521053466829317e-05,
      "loss": 1.4255,
      "step": 71910
    },
    {
      "epoch": 3.4834115196368995,
      "grad_norm": 0.21840277314186096,
      "learning_rate": 7.51963824715402e-05,
      "loss": 1.4243,
      "step": 71920
    },
    {
      "epoch": 3.483953460802412,
      "grad_norm": 0.19102028012275696,
      "learning_rate": 7.518222777294703e-05,
      "loss": 1.4402,
      "step": 71930
    },
    {
      "epoch": 3.4844954019679237,
      "grad_norm": 0.17519988119602203,
      "learning_rate": 7.51680705742671e-05,
      "loss": 1.4324,
      "step": 71940
    },
    {
      "epoch": 3.4849831490168848,
      "eval_loss": 2.4200730323791504,
      "eval_runtime": 21.6386,
      "eval_samples_per_second": 231.068,
      "eval_steps_per_second": 1.248,
      "step": 71949
    },
    {
      "epoch": 3.485037343133436,
      "grad_norm": 0.18477900326251984,
      "learning_rate": 7.515391087725416e-05,
      "loss": 1.4264,
      "step": 71950
    },
    {
      "epoch": 3.4855792842989484,
      "grad_norm": 0.21089692413806915,
      "learning_rate": 7.513974868366224e-05,
      "loss": 1.4269,
      "step": 71960
    },
    {
      "epoch": 3.4861212254644607,
      "grad_norm": 0.34691062569618225,
      "learning_rate": 7.51255839952457e-05,
      "loss": 1.4232,
      "step": 71970
    },
    {
      "epoch": 3.4866631666299726,
      "grad_norm": 0.2588226795196533,
      "learning_rate": 7.511141681375917e-05,
      "loss": 1.4332,
      "step": 71980
    },
    {
      "epoch": 3.487205107795485,
      "grad_norm": 0.16295021772384644,
      "learning_rate": 7.509724714095766e-05,
      "loss": 1.4296,
      "step": 71990
    },
    {
      "epoch": 3.487747048960997,
      "grad_norm": 0.22175998985767365,
      "learning_rate": 7.508307497859641e-05,
      "loss": 1.4286,
      "step": 72000
    },
    {
      "epoch": 3.0005419411655123,
      "grad_norm": 0.28885793685913086,
      "learning_rate": 7.506890032843104e-05,
      "loss": 1.4315,
      "step": 72010
    },
    {
      "epoch": 3.0010838823310246,
      "grad_norm": 0.17862290143966675,
      "learning_rate": 7.505472319221742e-05,
      "loss": 1.4237,
      "step": 72020
    },
    {
      "epoch": 3.0016258234965365,
      "grad_norm": 0.1830783486366272,
      "learning_rate": 7.504054357171176e-05,
      "loss": 1.4236,
      "step": 72030
    },
    {
      "epoch": 3.001950988195844,
      "eval_loss": 2.42079496383667,
      "eval_runtime": 22.1932,
      "eval_samples_per_second": 225.294,
      "eval_steps_per_second": 1.217,
      "step": 72036
    },
    {
      "epoch": 3.002167764662049,
      "grad_norm": 0.22852027416229248,
      "learning_rate": 7.502636146867058e-05,
      "loss": 1.4378,
      "step": 72040
    },
    {
      "epoch": 3.002709705827561,
      "grad_norm": 0.17318575084209442,
      "learning_rate": 7.501217688485067e-05,
      "loss": 1.4238,
      "step": 72050
    },
    {
      "epoch": 3.0032516469930735,
      "grad_norm": 0.32434844970703125,
      "learning_rate": 7.499798982200917e-05,
      "loss": 1.4348,
      "step": 72060
    },
    {
      "epoch": 3.0037935881585853,
      "grad_norm": 0.3870048522949219,
      "learning_rate": 7.498380028190353e-05,
      "loss": 1.4307,
      "step": 72070
    },
    {
      "epoch": 3.0043355293240976,
      "grad_norm": 0.21093975007534027,
      "learning_rate": 7.496960826629147e-05,
      "loss": 1.4365,
      "step": 72080
    },
    {
      "epoch": 3.00487747048961,
      "grad_norm": 0.1736556887626648,
      "learning_rate": 7.495541377693103e-05,
      "loss": 1.4264,
      "step": 72090
    },
    {
      "epoch": 3.0054194116551223,
      "grad_norm": 0.23631495237350464,
      "learning_rate": 7.494121681558056e-05,
      "loss": 1.4222,
      "step": 72100
    },
    {
      "epoch": 3.0059613528206346,
      "grad_norm": 0.1855953484773636,
      "learning_rate": 7.492701738399874e-05,
      "loss": 1.4191,
      "step": 72110
    },
    {
      "epoch": 3.0065032939861465,
      "grad_norm": 0.16358767449855804,
      "learning_rate": 7.491281548394454e-05,
      "loss": 1.4286,
      "step": 72120
    },
    {
      "epoch": 3.0066658763358003,
      "eval_loss": 2.4213919639587402,
      "eval_runtime": 22.016,
      "eval_samples_per_second": 227.108,
      "eval_steps_per_second": 1.226,
      "step": 72123
    },
    {
      "epoch": 3.007045235151659,
      "grad_norm": 0.17632584273815155,
      "learning_rate": 7.489861111717718e-05,
      "loss": 1.4182,
      "step": 72130
    },
    {
      "epoch": 3.007587176317171,
      "grad_norm": 0.16463103890419006,
      "learning_rate": 7.488440428545626e-05,
      "loss": 1.4179,
      "step": 72140
    },
    {
      "epoch": 3.0081291174826834,
      "grad_norm": 0.3055421710014343,
      "learning_rate": 7.487019499054169e-05,
      "loss": 1.426,
      "step": 72150
    },
    {
      "epoch": 3.0086710586481953,
      "grad_norm": 0.27146390080451965,
      "learning_rate": 7.485598323419362e-05,
      "loss": 1.425,
      "step": 72160
    },
    {
      "epoch": 3.0092129998137076,
      "grad_norm": 0.2870032787322998,
      "learning_rate": 7.484176901817256e-05,
      "loss": 1.4395,
      "step": 72170
    },
    {
      "epoch": 3.00975494097922,
      "grad_norm": 0.18291185796260834,
      "learning_rate": 7.482755234423931e-05,
      "loss": 1.428,
      "step": 72180
    },
    {
      "epoch": 3.0102968821447322,
      "grad_norm": 0.16777105629444122,
      "learning_rate": 7.481333321415493e-05,
      "loss": 1.4148,
      "step": 72190
    },
    {
      "epoch": 3.0108388233102445,
      "grad_norm": 0.19106002151966095,
      "learning_rate": 7.479911162968087e-05,
      "loss": 1.4297,
      "step": 72200
    },
    {
      "epoch": 3.0113807644757564,
      "grad_norm": 0.22009027004241943,
      "learning_rate": 7.478488759257882e-05,
      "loss": 1.4288,
      "step": 72210
    },
    {
      "epoch": 3.0113807644757564,
      "eval_loss": 2.4199483394622803,
      "eval_runtime": 22.9392,
      "eval_samples_per_second": 217.968,
      "eval_steps_per_second": 1.177,
      "step": 72210
    },
    {
      "epoch": 3.0119227056412687,
      "grad_norm": 0.20799531042575836,
      "learning_rate": 7.477066110461078e-05,
      "loss": 1.4217,
      "step": 72220
    },
    {
      "epoch": 3.012464646806781,
      "grad_norm": 0.17034943401813507,
      "learning_rate": 7.475643216753909e-05,
      "loss": 1.4257,
      "step": 72230
    },
    {
      "epoch": 3.0130065879722934,
      "grad_norm": 0.16152597963809967,
      "learning_rate": 7.474220078312636e-05,
      "loss": 1.4231,
      "step": 72240
    },
    {
      "epoch": 3.0135485291378057,
      "grad_norm": 0.23870690166950226,
      "learning_rate": 7.47279669531355e-05,
      "loss": 1.4311,
      "step": 72250
    },
    {
      "epoch": 3.0140904703033176,
      "grad_norm": 0.19177548587322235,
      "learning_rate": 7.471373067932975e-05,
      "loss": 1.4227,
      "step": 72260
    },
    {
      "epoch": 3.01463241146883,
      "grad_norm": 0.23941627144813538,
      "learning_rate": 7.469949196347263e-05,
      "loss": 1.4267,
      "step": 72270
    },
    {
      "epoch": 3.015174352634342,
      "grad_norm": 0.1809268444776535,
      "learning_rate": 7.468525080732798e-05,
      "loss": 1.421,
      "step": 72280
    },
    {
      "epoch": 3.0157162937998545,
      "grad_norm": 0.18789643049240112,
      "learning_rate": 7.46710072126599e-05,
      "loss": 1.4302,
      "step": 72290
    },
    {
      "epoch": 3.016095652615713,
      "eval_loss": 2.415761947631836,
      "eval_runtime": 22.0228,
      "eval_samples_per_second": 227.038,
      "eval_steps_per_second": 1.226,
      "step": 72297
    },
    {
      "epoch": 3.0162582349653664,
      "grad_norm": 0.18077979981899261,
      "learning_rate": 7.465676118123287e-05,
      "loss": 1.4176,
      "step": 72300
    },
    {
      "epoch": 3.0168001761308787,
      "grad_norm": 0.4619799852371216,
      "learning_rate": 7.46425127148116e-05,
      "loss": 1.4209,
      "step": 72310
    },
    {
      "epoch": 3.017342117296391,
      "grad_norm": 0.1716815084218979,
      "learning_rate": 7.462826181516115e-05,
      "loss": 1.4313,
      "step": 72320
    },
    {
      "epoch": 3.0178840584619033,
      "grad_norm": 0.21145406365394592,
      "learning_rate": 7.461400848404681e-05,
      "loss": 1.4226,
      "step": 72330
    },
    {
      "epoch": 3.0184259996274156,
      "grad_norm": 0.19309388101100922,
      "learning_rate": 7.459975272323427e-05,
      "loss": 1.4252,
      "step": 72340
    },
    {
      "epoch": 3.0189679407929275,
      "grad_norm": 0.37424296140670776,
      "learning_rate": 7.458549453448948e-05,
      "loss": 1.4245,
      "step": 72350
    },
    {
      "epoch": 3.01950988195844,
      "grad_norm": 0.36267608404159546,
      "learning_rate": 7.457123391957863e-05,
      "loss": 1.4271,
      "step": 72360
    },
    {
      "epoch": 3.020051823123952,
      "grad_norm": 0.1854252964258194,
      "learning_rate": 7.455697088026831e-05,
      "loss": 1.4313,
      "step": 72370
    },
    {
      "epoch": 3.0205937642894645,
      "grad_norm": 0.2632382810115814,
      "learning_rate": 7.454270541832532e-05,
      "loss": 1.4254,
      "step": 72380
    },
    {
      "epoch": 3.020810540755669,
      "eval_loss": 2.422041416168213,
      "eval_runtime": 21.9894,
      "eval_samples_per_second": 227.382,
      "eval_steps_per_second": 1.228,
      "step": 72384
    },
    {
      "epoch": 3.0211357054549763,
      "grad_norm": 0.16941377520561218,
      "learning_rate": 7.452843753551687e-05,
      "loss": 1.433,
      "step": 72390
    },
    {
      "epoch": 3.0216776466204887,
      "grad_norm": 0.20150338113307953,
      "learning_rate": 7.451416723361033e-05,
      "loss": 1.4228,
      "step": 72400
    },
    {
      "epoch": 3.022219587786001,
      "grad_norm": 0.20586276054382324,
      "learning_rate": 7.449989451437347e-05,
      "loss": 1.4336,
      "step": 72410
    },
    {
      "epoch": 3.0227615289515133,
      "grad_norm": 0.180009663105011,
      "learning_rate": 7.448561937957435e-05,
      "loss": 1.4399,
      "step": 72420
    },
    {
      "epoch": 3.0233034701170256,
      "grad_norm": 0.17290925979614258,
      "learning_rate": 7.44713418309813e-05,
      "loss": 1.4349,
      "step": 72430
    },
    {
      "epoch": 3.0238454112825375,
      "grad_norm": 0.19423924386501312,
      "learning_rate": 7.445706187036295e-05,
      "loss": 1.4175,
      "step": 72440
    },
    {
      "epoch": 3.02438735244805,
      "grad_norm": 0.19001413881778717,
      "learning_rate": 7.444277949948826e-05,
      "loss": 1.4253,
      "step": 72450
    },
    {
      "epoch": 3.024929293613562,
      "grad_norm": 0.27171728014945984,
      "learning_rate": 7.442849472012648e-05,
      "loss": 1.4299,
      "step": 72460
    },
    {
      "epoch": 3.0254712347790744,
      "grad_norm": 0.24754732847213745,
      "learning_rate": 7.441420753404709e-05,
      "loss": 1.4204,
      "step": 72470
    },
    {
      "epoch": 3.0255254288956257,
      "eval_loss": 2.420412063598633,
      "eval_runtime": 21.9948,
      "eval_samples_per_second": 227.326,
      "eval_steps_per_second": 1.228,
      "step": 72471
    },
    {
      "epoch": 3.0260131759445863,
      "grad_norm": 0.18741434812545776,
      "learning_rate": 7.439991794301997e-05,
      "loss": 1.4214,
      "step": 72480
    },
    {
      "epoch": 3.0265551171100986,
      "grad_norm": 0.2331148236989975,
      "learning_rate": 7.438562594881523e-05,
      "loss": 1.4267,
      "step": 72490
    },
    {
      "epoch": 3.027097058275611,
      "grad_norm": 0.21943721175193787,
      "learning_rate": 7.437133155320333e-05,
      "loss": 1.4202,
      "step": 72500
    },
    {
      "epoch": 3.0276389994411232,
      "grad_norm": 0.1892043948173523,
      "learning_rate": 7.435703475795498e-05,
      "loss": 1.4201,
      "step": 72510
    },
    {
      "epoch": 3.0281809406066356,
      "grad_norm": 0.2600436210632324,
      "learning_rate": 7.434273556484119e-05,
      "loss": 1.4357,
      "step": 72520
    },
    {
      "epoch": 3.0287228817721474,
      "grad_norm": 0.32084086537361145,
      "learning_rate": 7.43284339756333e-05,
      "loss": 1.4274,
      "step": 72530
    },
    {
      "epoch": 3.0292648229376598,
      "grad_norm": 0.23586052656173706,
      "learning_rate": 7.431412999210292e-05,
      "loss": 1.4261,
      "step": 72540
    },
    {
      "epoch": 3.029806764103172,
      "grad_norm": 0.2831364870071411,
      "learning_rate": 7.429982361602198e-05,
      "loss": 1.4191,
      "step": 72550
    },
    {
      "epoch": 3.030240317035582,
      "eval_loss": 2.41964054107666,
      "eval_runtime": 21.992,
      "eval_samples_per_second": 227.355,
      "eval_steps_per_second": 1.228,
      "step": 72558
    },
    {
      "epoch": 3.0303487052686844,
      "grad_norm": 0.1781311184167862,
      "learning_rate": 7.428551484916269e-05,
      "loss": 1.4297,
      "step": 72560
    },
    {
      "epoch": 3.0308906464341967,
      "grad_norm": 0.34944701194763184,
      "learning_rate": 7.427120369329753e-05,
      "loss": 1.4319,
      "step": 72570
    },
    {
      "epoch": 3.0314325875997086,
      "grad_norm": 0.17898167669773102,
      "learning_rate": 7.425689015019932e-05,
      "loss": 1.4229,
      "step": 72580
    },
    {
      "epoch": 3.031974528765221,
      "grad_norm": 0.19159488379955292,
      "learning_rate": 7.42425742216412e-05,
      "loss": 1.4248,
      "step": 72590
    },
    {
      "epoch": 3.032516469930733,
      "grad_norm": 0.1828126460313797,
      "learning_rate": 7.422825590939648e-05,
      "loss": 1.4298,
      "step": 72600
    },
    {
      "epoch": 3.0330584110962455,
      "grad_norm": 0.1625342071056366,
      "learning_rate": 7.421393521523893e-05,
      "loss": 1.425,
      "step": 72610
    },
    {
      "epoch": 3.0336003522617574,
      "grad_norm": 0.1856265813112259,
      "learning_rate": 7.41996121409425e-05,
      "loss": 1.4231,
      "step": 72620
    },
    {
      "epoch": 3.0341422934272697,
      "grad_norm": 0.20504432916641235,
      "learning_rate": 7.418528668828147e-05,
      "loss": 1.4211,
      "step": 72630
    },
    {
      "epoch": 3.034684234592782,
      "grad_norm": 0.1928977072238922,
      "learning_rate": 7.417095885903043e-05,
      "loss": 1.421,
      "step": 72640
    },
    {
      "epoch": 3.034955205175538,
      "eval_loss": 2.4224603176116943,
      "eval_runtime": 21.9942,
      "eval_samples_per_second": 227.333,
      "eval_steps_per_second": 1.228,
      "step": 72645
    },
    {
      "epoch": 3.0352261757582943,
      "grad_norm": 0.26011061668395996,
      "learning_rate": 7.415662865496421e-05,
      "loss": 1.4156,
      "step": 72650
    },
    {
      "epoch": 3.0357681169238067,
      "grad_norm": 0.1762949675321579,
      "learning_rate": 7.414229607785803e-05,
      "loss": 1.4256,
      "step": 72660
    },
    {
      "epoch": 3.0363100580893185,
      "grad_norm": 0.23641474545001984,
      "learning_rate": 7.412796112948731e-05,
      "loss": 1.4283,
      "step": 72670
    },
    {
      "epoch": 3.036851999254831,
      "grad_norm": 0.2524915635585785,
      "learning_rate": 7.41136238116278e-05,
      "loss": 1.4353,
      "step": 72680
    },
    {
      "epoch": 3.037393940420343,
      "grad_norm": 0.16015665233135223,
      "learning_rate": 7.409928412605557e-05,
      "loss": 1.4262,
      "step": 72690
    },
    {
      "epoch": 3.0379358815858555,
      "grad_norm": 0.252564936876297,
      "learning_rate": 7.408494207454694e-05,
      "loss": 1.4269,
      "step": 72700
    },
    {
      "epoch": 3.0384778227513674,
      "grad_norm": 0.2023906707763672,
      "learning_rate": 7.407059765887854e-05,
      "loss": 1.4323,
      "step": 72710
    },
    {
      "epoch": 3.0390197639168797,
      "grad_norm": 0.22321388125419617,
      "learning_rate": 7.40562508808273e-05,
      "loss": 1.4271,
      "step": 72720
    },
    {
      "epoch": 3.039561705082392,
      "grad_norm": 0.26535144448280334,
      "learning_rate": 7.404190174217044e-05,
      "loss": 1.4305,
      "step": 72730
    },
    {
      "epoch": 3.0396700933154945,
      "eval_loss": 2.43088436126709,
      "eval_runtime": 21.9914,
      "eval_samples_per_second": 227.362,
      "eval_steps_per_second": 1.228,
      "step": 72732
    },
    {
      "epoch": 3.0401036462479043,
      "grad_norm": 0.20327869057655334,
      "learning_rate": 7.402755024468547e-05,
      "loss": 1.4258,
      "step": 72740
    },
    {
      "epoch": 3.0406455874134166,
      "grad_norm": 0.2531057596206665,
      "learning_rate": 7.401319639015018e-05,
      "loss": 1.4226,
      "step": 72750
    },
    {
      "epoch": 3.0411875285789285,
      "grad_norm": 0.2527235746383667,
      "learning_rate": 7.399884018034265e-05,
      "loss": 1.4343,
      "step": 72760
    },
    {
      "epoch": 3.041729469744441,
      "grad_norm": 0.16884733736515045,
      "learning_rate": 7.398448161704131e-05,
      "loss": 1.4123,
      "step": 72770
    },
    {
      "epoch": 3.042271410909953,
      "grad_norm": 0.2351938784122467,
      "learning_rate": 7.397012070202478e-05,
      "loss": 1.4316,
      "step": 72780
    },
    {
      "epoch": 3.0428133520754654,
      "grad_norm": 0.17390334606170654,
      "learning_rate": 7.395575743707208e-05,
      "loss": 1.4276,
      "step": 72790
    },
    {
      "epoch": 3.0433552932409773,
      "grad_norm": 0.23687273263931274,
      "learning_rate": 7.394139182396245e-05,
      "loss": 1.4308,
      "step": 72800
    },
    {
      "epoch": 3.0438972344064896,
      "grad_norm": 0.22024738788604736,
      "learning_rate": 7.39270238644754e-05,
      "loss": 1.4276,
      "step": 72810
    },
    {
      "epoch": 3.0443849814554507,
      "eval_loss": 2.4263041019439697,
      "eval_runtime": 21.995,
      "eval_samples_per_second": 227.324,
      "eval_steps_per_second": 1.228,
      "step": 72819
    },
    {
      "epoch": 3.044439175572002,
      "grad_norm": 0.37231162190437317,
      "learning_rate": 7.391265356039084e-05,
      "loss": 1.4351,
      "step": 72820
    },
    {
      "epoch": 3.0449811167375143,
      "grad_norm": 0.1920226663351059,
      "learning_rate": 7.389828091348888e-05,
      "loss": 1.4319,
      "step": 72830
    },
    {
      "epoch": 3.0455230579030266,
      "grad_norm": 0.16061918437480927,
      "learning_rate": 7.388390592554989e-05,
      "loss": 1.4258,
      "step": 72840
    },
    {
      "epoch": 3.0460649990685384,
      "grad_norm": 0.20206622779369354,
      "learning_rate": 7.386952859835466e-05,
      "loss": 1.4202,
      "step": 72850
    },
    {
      "epoch": 3.0466069402340508,
      "grad_norm": 0.31317874789237976,
      "learning_rate": 7.385514893368411e-05,
      "loss": 1.4229,
      "step": 72860
    },
    {
      "epoch": 3.047148881399563,
      "grad_norm": 0.17705164849758148,
      "learning_rate": 7.38407669333196e-05,
      "loss": 1.4275,
      "step": 72870
    },
    {
      "epoch": 3.0476908225650754,
      "grad_norm": 0.2397981733083725,
      "learning_rate": 7.382638259904268e-05,
      "loss": 1.4225,
      "step": 72880
    },
    {
      "epoch": 3.0482327637305877,
      "grad_norm": 0.19595515727996826,
      "learning_rate": 7.38119959326352e-05,
      "loss": 1.437,
      "step": 72890
    },
    {
      "epoch": 3.0487747048960996,
      "grad_norm": 0.18026091158390045,
      "learning_rate": 7.379760693587935e-05,
      "loss": 1.43,
      "step": 72900
    },
    {
      "epoch": 3.0490998695954072,
      "eval_loss": 2.430786371231079,
      "eval_runtime": 22.1488,
      "eval_samples_per_second": 225.746,
      "eval_steps_per_second": 1.219,
      "step": 72906
    },
    {
      "epoch": 3.049316646061612,
      "grad_norm": 0.26389095187187195,
      "learning_rate": 7.378321561055756e-05,
      "loss": 1.4251,
      "step": 72910
    },
    {
      "epoch": 3.049858587227124,
      "grad_norm": 0.33883413672447205,
      "learning_rate": 7.376882195845256e-05,
      "loss": 1.4346,
      "step": 72920
    },
    {
      "epoch": 3.0504005283926365,
      "grad_norm": 0.17455531656742096,
      "learning_rate": 7.37544259813474e-05,
      "loss": 1.4272,
      "step": 72930
    },
    {
      "epoch": 3.0509424695581484,
      "grad_norm": 0.1896173059940338,
      "learning_rate": 7.374002768102535e-05,
      "loss": 1.4286,
      "step": 72940
    },
    {
      "epoch": 3.0514844107236607,
      "grad_norm": 0.1727907508611679,
      "learning_rate": 7.372562705927006e-05,
      "loss": 1.4246,
      "step": 72950
    },
    {
      "epoch": 3.052026351889173,
      "grad_norm": 0.17617569863796234,
      "learning_rate": 7.371122411786538e-05,
      "loss": 1.433,
      "step": 72960
    },
    {
      "epoch": 3.0525682930546854,
      "grad_norm": 0.2908771336078644,
      "learning_rate": 7.369681885859548e-05,
      "loss": 1.4263,
      "step": 72970
    },
    {
      "epoch": 3.0531102342201977,
      "grad_norm": 0.2100307047367096,
      "learning_rate": 7.368241128324485e-05,
      "loss": 1.4221,
      "step": 72980
    },
    {
      "epoch": 3.0536521753857095,
      "grad_norm": 0.24599306285381317,
      "learning_rate": 7.366800139359822e-05,
      "loss": 1.4234,
      "step": 72990
    },
    {
      "epoch": 3.0538147577353634,
      "eval_loss": 2.424680709838867,
      "eval_runtime": 22.0534,
      "eval_samples_per_second": 226.723,
      "eval_steps_per_second": 1.224,
      "step": 72993
    },
    {
      "epoch": 3.054194116551222,
      "grad_norm": 0.18987296521663666,
      "learning_rate": 7.365358919144063e-05,
      "loss": 1.427,
      "step": 73000
    },
    {
      "epoch": 3.054736057716734,
      "grad_norm": 0.17865373194217682,
      "learning_rate": 7.363917467855736e-05,
      "loss": 1.426,
      "step": 73010
    },
    {
      "epoch": 3.0552779988822465,
      "grad_norm": 0.17104874551296234,
      "learning_rate": 7.362475785673409e-05,
      "loss": 1.4333,
      "step": 73020
    },
    {
      "epoch": 3.0558199400477584,
      "grad_norm": 0.17414012551307678,
      "learning_rate": 7.361033872775667e-05,
      "loss": 1.4307,
      "step": 73030
    },
    {
      "epoch": 3.0563618812132707,
      "grad_norm": 0.2933975160121918,
      "learning_rate": 7.359591729341128e-05,
      "loss": 1.4239,
      "step": 73040
    },
    {
      "epoch": 3.056903822378783,
      "grad_norm": 0.1879478394985199,
      "learning_rate": 7.358149355548439e-05,
      "loss": 1.4258,
      "step": 73050
    },
    {
      "epoch": 3.0574457635442953,
      "grad_norm": 0.18025019764900208,
      "learning_rate": 7.356706751576274e-05,
      "loss": 1.421,
      "step": 73060
    },
    {
      "epoch": 3.0579877047098076,
      "grad_norm": 0.16190853714942932,
      "learning_rate": 7.355263917603341e-05,
      "loss": 1.4229,
      "step": 73070
    },
    {
      "epoch": 3.0585296458753195,
      "grad_norm": 0.19061139225959778,
      "learning_rate": 7.353820853808365e-05,
      "loss": 1.4202,
      "step": 73080
    },
    {
      "epoch": 3.0585296458753195,
      "eval_loss": 2.4280989170074463,
      "eval_runtime": 21.9869,
      "eval_samples_per_second": 227.408,
      "eval_steps_per_second": 1.228,
      "step": 73080
    },
    {
      "epoch": 3.059071587040832,
      "grad_norm": 0.24955511093139648,
      "learning_rate": 7.352377560370113e-05,
      "loss": 1.4388,
      "step": 73090
    },
    {
      "epoch": 3.059613528206344,
      "grad_norm": 0.24170927703380585,
      "learning_rate": 7.350934037467371e-05,
      "loss": 1.4251,
      "step": 73100
    },
    {
      "epoch": 3.0601554693718565,
      "grad_norm": 0.19894978404045105,
      "learning_rate": 7.349490285278953e-05,
      "loss": 1.4229,
      "step": 73110
    },
    {
      "epoch": 3.0606974105373683,
      "grad_norm": 0.16050221025943756,
      "learning_rate": 7.348046303983713e-05,
      "loss": 1.4217,
      "step": 73120
    },
    {
      "epoch": 3.0612393517028806,
      "grad_norm": 0.20622624456882477,
      "learning_rate": 7.346602093760516e-05,
      "loss": 1.4358,
      "step": 73130
    },
    {
      "epoch": 3.061781292868393,
      "grad_norm": 0.17460785806179047,
      "learning_rate": 7.345157654788272e-05,
      "loss": 1.4205,
      "step": 73140
    },
    {
      "epoch": 3.0623232340339053,
      "grad_norm": 0.2245752215385437,
      "learning_rate": 7.343712987245908e-05,
      "loss": 1.4378,
      "step": 73150
    },
    {
      "epoch": 3.0628651751994176,
      "grad_norm": 0.2324395626783371,
      "learning_rate": 7.342268091312381e-05,
      "loss": 1.4331,
      "step": 73160
    },
    {
      "epoch": 3.063244534015276,
      "eval_loss": 2.4189603328704834,
      "eval_runtime": 22.2302,
      "eval_samples_per_second": 224.919,
      "eval_steps_per_second": 1.215,
      "step": 73167
    },
    {
      "epoch": 3.0634071163649295,
      "grad_norm": 0.16124655306339264,
      "learning_rate": 7.340822967166684e-05,
      "loss": 1.427,
      "step": 73170
    },
    {
      "epoch": 3.0639490575304418,
      "grad_norm": 0.171858012676239,
      "learning_rate": 7.339377614987827e-05,
      "loss": 1.4179,
      "step": 73180
    },
    {
      "epoch": 3.064490998695954,
      "grad_norm": 0.28253042697906494,
      "learning_rate": 7.33793203495486e-05,
      "loss": 1.4261,
      "step": 73190
    },
    {
      "epoch": 3.0650329398614664,
      "grad_norm": 0.1876271665096283,
      "learning_rate": 7.336486227246851e-05,
      "loss": 1.431,
      "step": 73200
    },
    {
      "epoch": 3.0655748810269783,
      "grad_norm": 0.2576892077922821,
      "learning_rate": 7.335040192042901e-05,
      "loss": 1.4216,
      "step": 73210
    },
    {
      "epoch": 3.0661168221924906,
      "grad_norm": 0.28171801567077637,
      "learning_rate": 7.333593929522138e-05,
      "loss": 1.4285,
      "step": 73220
    },
    {
      "epoch": 3.066658763358003,
      "grad_norm": 0.16934189200401306,
      "learning_rate": 7.33214743986372e-05,
      "loss": 1.4277,
      "step": 73230
    },
    {
      "epoch": 3.0672007045235152,
      "grad_norm": 0.17389722168445587,
      "learning_rate": 7.330700723246835e-05,
      "loss": 1.4219,
      "step": 73240
    },
    {
      "epoch": 3.0677426456890275,
      "grad_norm": 0.18714025616645813,
      "learning_rate": 7.32925377985069e-05,
      "loss": 1.4354,
      "step": 73250
    },
    {
      "epoch": 3.067959422155232,
      "eval_loss": 2.4224860668182373,
      "eval_runtime": 21.9928,
      "eval_samples_per_second": 227.347,
      "eval_steps_per_second": 1.228,
      "step": 73254
    },
    {
      "epoch": 3.0682845868545394,
      "grad_norm": 0.1830187737941742,
      "learning_rate": 7.327806609854527e-05,
      "loss": 1.4227,
      "step": 73260
    },
    {
      "epoch": 3.0688265280200517,
      "grad_norm": 0.1765843778848648,
      "learning_rate": 7.326359213437618e-05,
      "loss": 1.4215,
      "step": 73270
    },
    {
      "epoch": 3.069368469185564,
      "grad_norm": 0.19436825811862946,
      "learning_rate": 7.32491159077926e-05,
      "loss": 1.423,
      "step": 73280
    },
    {
      "epoch": 3.0699104103510764,
      "grad_norm": 0.31252750754356384,
      "learning_rate": 7.323463742058776e-05,
      "loss": 1.4223,
      "step": 73290
    },
    {
      "epoch": 3.0704523515165887,
      "grad_norm": 0.2948218286037445,
      "learning_rate": 7.322015667455521e-05,
      "loss": 1.4295,
      "step": 73300
    },
    {
      "epoch": 3.0709942926821006,
      "grad_norm": 0.1795574426651001,
      "learning_rate": 7.320567367148875e-05,
      "loss": 1.4268,
      "step": 73310
    },
    {
      "epoch": 3.071536233847613,
      "grad_norm": 0.2362855225801468,
      "learning_rate": 7.319118841318246e-05,
      "loss": 1.4224,
      "step": 73320
    },
    {
      "epoch": 3.072078175013125,
      "grad_norm": 0.22740645706653595,
      "learning_rate": 7.317670090143076e-05,
      "loss": 1.4203,
      "step": 73330
    },
    {
      "epoch": 3.0726201161786375,
      "grad_norm": 0.1676589697599411,
      "learning_rate": 7.316221113802825e-05,
      "loss": 1.4286,
      "step": 73340
    },
    {
      "epoch": 3.072674310295189,
      "eval_loss": 2.4217214584350586,
      "eval_runtime": 21.9907,
      "eval_samples_per_second": 227.369,
      "eval_steps_per_second": 1.228,
      "step": 73341
    },
    {
      "epoch": 3.0731620573441494,
      "grad_norm": 0.4173310697078705,
      "learning_rate": 7.314771912476987e-05,
      "loss": 1.4293,
      "step": 73350
    },
    {
      "epoch": 3.0737039985096617,
      "grad_norm": 0.198992520570755,
      "learning_rate": 7.313322486345085e-05,
      "loss": 1.4195,
      "step": 73360
    },
    {
      "epoch": 3.074245939675174,
      "grad_norm": 0.19896645843982697,
      "learning_rate": 7.311872835586665e-05,
      "loss": 1.4355,
      "step": 73370
    },
    {
      "epoch": 3.0747878808406863,
      "grad_norm": 0.1793803870677948,
      "learning_rate": 7.310422960381305e-05,
      "loss": 1.4228,
      "step": 73380
    },
    {
      "epoch": 3.0753298220061986,
      "grad_norm": 0.21066676080226898,
      "learning_rate": 7.308972860908609e-05,
      "loss": 1.4219,
      "step": 73390
    },
    {
      "epoch": 3.0758717631717105,
      "grad_norm": 0.18298877775669098,
      "learning_rate": 7.30752253734821e-05,
      "loss": 1.4309,
      "step": 73400
    },
    {
      "epoch": 3.076413704337223,
      "grad_norm": 0.202785462141037,
      "learning_rate": 7.306071989879762e-05,
      "loss": 1.423,
      "step": 73410
    },
    {
      "epoch": 3.076955645502735,
      "grad_norm": 0.17988531291484833,
      "learning_rate": 7.304621218682961e-05,
      "loss": 1.4266,
      "step": 73420
    },
    {
      "epoch": 3.077389198435145,
      "eval_loss": 2.4217171669006348,
      "eval_runtime": 22.0012,
      "eval_samples_per_second": 227.26,
      "eval_steps_per_second": 1.227,
      "step": 73428
    },
    {
      "epoch": 3.0774975866682475,
      "grad_norm": 0.29091671109199524,
      "learning_rate": 7.303170223937518e-05,
      "loss": 1.4327,
      "step": 73430
    },
    {
      "epoch": 3.0780395278337593,
      "grad_norm": 0.1689477562904358,
      "learning_rate": 7.301719005823175e-05,
      "loss": 1.4226,
      "step": 73440
    },
    {
      "epoch": 3.0785814689992717,
      "grad_norm": 0.21074718236923218,
      "learning_rate": 7.300267564519703e-05,
      "loss": 1.4233,
      "step": 73450
    },
    {
      "epoch": 3.079123410164784,
      "grad_norm": 0.18468187749385834,
      "learning_rate": 7.298815900206904e-05,
      "loss": 1.4186,
      "step": 73460
    },
    {
      "epoch": 3.0796653513302963,
      "grad_norm": 0.16234566271305084,
      "learning_rate": 7.2973640130646e-05,
      "loss": 1.42,
      "step": 73470
    },
    {
      "epoch": 3.0802072924958086,
      "grad_norm": 0.20264677703380585,
      "learning_rate": 7.295911903272643e-05,
      "loss": 1.4322,
      "step": 73480
    },
    {
      "epoch": 3.0807492336613205,
      "grad_norm": 0.20148052275180817,
      "learning_rate": 7.29445957101092e-05,
      "loss": 1.42,
      "step": 73490
    },
    {
      "epoch": 3.081291174826833,
      "grad_norm": 0.1632404625415802,
      "learning_rate": 7.293007016459333e-05,
      "loss": 1.4228,
      "step": 73500
    },
    {
      "epoch": 3.081833115992345,
      "grad_norm": 0.28163808584213257,
      "learning_rate": 7.291554239797823e-05,
      "loss": 1.4194,
      "step": 73510
    },
    {
      "epoch": 3.082104086575101,
      "eval_loss": 2.420088768005371,
      "eval_runtime": 21.9934,
      "eval_samples_per_second": 227.341,
      "eval_steps_per_second": 1.228,
      "step": 73515
    },
    {
      "epoch": 3.0823750571578574,
      "grad_norm": 0.2194111943244934,
      "learning_rate": 7.290101241206354e-05,
      "loss": 1.4236,
      "step": 73520
    },
    {
      "epoch": 3.0829169983233697,
      "grad_norm": 0.3260360658168793,
      "learning_rate": 7.288648020864911e-05,
      "loss": 1.436,
      "step": 73530
    },
    {
      "epoch": 3.0834589394888816,
      "grad_norm": 0.21551568806171417,
      "learning_rate": 7.287194578953518e-05,
      "loss": 1.423,
      "step": 73540
    },
    {
      "epoch": 3.084000880654394,
      "grad_norm": 0.2473076581954956,
      "learning_rate": 7.28574091565222e-05,
      "loss": 1.4256,
      "step": 73550
    },
    {
      "epoch": 3.0845428218199062,
      "grad_norm": 0.24069726467132568,
      "learning_rate": 7.28428703114109e-05,
      "loss": 1.4311,
      "step": 73560
    },
    {
      "epoch": 3.0850847629854186,
      "grad_norm": 0.24383601546287537,
      "learning_rate": 7.28283292560023e-05,
      "loss": 1.4249,
      "step": 73570
    },
    {
      "epoch": 3.0856267041509304,
      "grad_norm": 0.17306429147720337,
      "learning_rate": 7.281378599209768e-05,
      "loss": 1.4223,
      "step": 73580
    },
    {
      "epoch": 3.0861686453164427,
      "grad_norm": 0.3249399960041046,
      "learning_rate": 7.279924052149858e-05,
      "loss": 1.4173,
      "step": 73590
    },
    {
      "epoch": 3.086710586481955,
      "grad_norm": 0.33424919843673706,
      "learning_rate": 7.278469284600684e-05,
      "loss": 1.4187,
      "step": 73600
    },
    {
      "epoch": 3.0868189747150576,
      "eval_loss": 2.4198853969573975,
      "eval_runtime": 21.9949,
      "eval_samples_per_second": 227.326,
      "eval_steps_per_second": 1.228,
      "step": 73602
    },
    {
      "epoch": 3.0872525276474674,
      "grad_norm": 0.25549885630607605,
      "learning_rate": 7.277014296742457e-05,
      "loss": 1.4192,
      "step": 73610
    },
    {
      "epoch": 3.0877944688129797,
      "grad_norm": 0.3509141206741333,
      "learning_rate": 7.275559088755414e-05,
      "loss": 1.4217,
      "step": 73620
    },
    {
      "epoch": 3.0883364099784916,
      "grad_norm": 0.1701222062110901,
      "learning_rate": 7.27410366081982e-05,
      "loss": 1.4255,
      "step": 73630
    },
    {
      "epoch": 3.088878351144004,
      "grad_norm": 0.4111487865447998,
      "learning_rate": 7.272648013115965e-05,
      "loss": 1.4136,
      "step": 73640
    },
    {
      "epoch": 3.089420292309516,
      "grad_norm": 0.292900949716568,
      "learning_rate": 7.271192145824172e-05,
      "loss": 1.4301,
      "step": 73650
    },
    {
      "epoch": 3.0899622334750285,
      "grad_norm": 0.24990233778953552,
      "learning_rate": 7.269736059124785e-05,
      "loss": 1.4301,
      "step": 73660
    },
    {
      "epoch": 3.0905041746405404,
      "grad_norm": 0.2537482678890228,
      "learning_rate": 7.26827975319818e-05,
      "loss": 1.4215,
      "step": 73670
    },
    {
      "epoch": 3.0910461158060527,
      "grad_norm": 0.23013080656528473,
      "learning_rate": 7.266823228224754e-05,
      "loss": 1.4313,
      "step": 73680
    },
    {
      "epoch": 3.0915338628550137,
      "eval_loss": 2.418518304824829,
      "eval_runtime": 21.9932,
      "eval_samples_per_second": 227.343,
      "eval_steps_per_second": 1.228,
      "step": 73689
    },
    {
      "epoch": 3.091588056971565,
      "grad_norm": 0.20117320120334625,
      "learning_rate": 7.265366484384938e-05,
      "loss": 1.4179,
      "step": 73690
    },
    {
      "epoch": 3.0921299981370773,
      "grad_norm": 0.25527241826057434,
      "learning_rate": 7.263909521859186e-05,
      "loss": 1.4265,
      "step": 73700
    },
    {
      "epoch": 3.0926719393025897,
      "grad_norm": 0.1841808259487152,
      "learning_rate": 7.262452340827981e-05,
      "loss": 1.4271,
      "step": 73710
    },
    {
      "epoch": 3.0932138804681015,
      "grad_norm": 0.16479361057281494,
      "learning_rate": 7.260994941471832e-05,
      "loss": 1.4119,
      "step": 73720
    },
    {
      "epoch": 3.093755821633614,
      "grad_norm": 0.23891477286815643,
      "learning_rate": 7.259537323971276e-05,
      "loss": 1.424,
      "step": 73730
    },
    {
      "epoch": 3.094297762799126,
      "grad_norm": 0.22275440394878387,
      "learning_rate": 7.258079488506874e-05,
      "loss": 1.4334,
      "step": 73740
    },
    {
      "epoch": 3.0948397039646385,
      "grad_norm": 0.19591204822063446,
      "learning_rate": 7.256621435259218e-05,
      "loss": 1.4379,
      "step": 73750
    },
    {
      "epoch": 3.0953816451301503,
      "grad_norm": 0.19313859939575195,
      "learning_rate": 7.255163164408926e-05,
      "loss": 1.4185,
      "step": 73760
    },
    {
      "epoch": 3.0959235862956627,
      "grad_norm": 0.23939813673496246,
      "learning_rate": 7.25370467613664e-05,
      "loss": 1.4255,
      "step": 73770
    },
    {
      "epoch": 3.0962487509949703,
      "eval_loss": 2.417570114135742,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.323,
      "eval_steps_per_second": 1.228,
      "step": 73776
    },
    {
      "epoch": 3.096465527461175,
      "grad_norm": 0.23741766810417175,
      "learning_rate": 7.252245970623035e-05,
      "loss": 1.4347,
      "step": 73780
    },
    {
      "epoch": 3.0970074686266873,
      "grad_norm": 0.17643536627292633,
      "learning_rate": 7.250787048048804e-05,
      "loss": 1.4208,
      "step": 73790
    },
    {
      "epoch": 3.0975494097921996,
      "grad_norm": 0.1725606471300125,
      "learning_rate": 7.249327908594675e-05,
      "loss": 1.4172,
      "step": 73800
    },
    {
      "epoch": 3.0980913509577115,
      "grad_norm": 0.17194822430610657,
      "learning_rate": 7.247868552441402e-05,
      "loss": 1.4279,
      "step": 73810
    },
    {
      "epoch": 3.098633292123224,
      "grad_norm": 0.17184831202030182,
      "learning_rate": 7.246408979769759e-05,
      "loss": 1.4215,
      "step": 73820
    },
    {
      "epoch": 3.099175233288736,
      "grad_norm": 0.15591196715831757,
      "learning_rate": 7.244949190760553e-05,
      "loss": 1.413,
      "step": 73830
    },
    {
      "epoch": 3.0997171744542484,
      "grad_norm": 0.2347261756658554,
      "learning_rate": 7.243489185594618e-05,
      "loss": 1.4282,
      "step": 73840
    },
    {
      "epoch": 3.1002591156197603,
      "grad_norm": 0.17219695448875427,
      "learning_rate": 7.242028964452812e-05,
      "loss": 1.4218,
      "step": 73850
    },
    {
      "epoch": 3.1008010567852726,
      "grad_norm": 0.19845928251743317,
      "learning_rate": 7.240568527516022e-05,
      "loss": 1.4226,
      "step": 73860
    },
    {
      "epoch": 3.1009636391349265,
      "eval_loss": 2.4213979244232178,
      "eval_runtime": 21.9911,
      "eval_samples_per_second": 227.365,
      "eval_steps_per_second": 1.228,
      "step": 73863
    },
    {
      "epoch": 3.101342997950785,
      "grad_norm": 0.18098603188991547,
      "learning_rate": 7.239107874965158e-05,
      "loss": 1.4348,
      "step": 73870
    },
    {
      "epoch": 3.1018849391162973,
      "grad_norm": 0.19403134286403656,
      "learning_rate": 7.237647006981162e-05,
      "loss": 1.4321,
      "step": 73880
    },
    {
      "epoch": 3.1024268802818096,
      "grad_norm": 0.2447945922613144,
      "learning_rate": 7.236185923745e-05,
      "loss": 1.4228,
      "step": 73890
    },
    {
      "epoch": 3.1029688214473214,
      "grad_norm": 0.20486268401145935,
      "learning_rate": 7.234724625437661e-05,
      "loss": 1.4183,
      "step": 73900
    },
    {
      "epoch": 3.1035107626128338,
      "grad_norm": 0.18675167858600616,
      "learning_rate": 7.233263112240168e-05,
      "loss": 1.4293,
      "step": 73910
    },
    {
      "epoch": 3.104052703778346,
      "grad_norm": 0.3332042396068573,
      "learning_rate": 7.231801384333567e-05,
      "loss": 1.4271,
      "step": 73920
    },
    {
      "epoch": 3.1045946449438584,
      "grad_norm": 0.20681266486644745,
      "learning_rate": 7.230339441898928e-05,
      "loss": 1.4215,
      "step": 73930
    },
    {
      "epoch": 3.1051365861093707,
      "grad_norm": 0.2513537108898163,
      "learning_rate": 7.228877285117352e-05,
      "loss": 1.4296,
      "step": 73940
    },
    {
      "epoch": 3.1056785272748826,
      "grad_norm": 0.17482328414916992,
      "learning_rate": 7.227414914169961e-05,
      "loss": 1.4263,
      "step": 73950
    },
    {
      "epoch": 3.1056785272748826,
      "eval_loss": 2.4194869995117188,
      "eval_runtime": 21.9853,
      "eval_samples_per_second": 227.425,
      "eval_steps_per_second": 1.228,
      "step": 73950
    },
    {
      "epoch": 3.106220468440395,
      "grad_norm": 0.2170400321483612,
      "learning_rate": 7.225952329237913e-05,
      "loss": 1.4152,
      "step": 73960
    },
    {
      "epoch": 3.106762409605907,
      "grad_norm": 0.22855044901371002,
      "learning_rate": 7.224489530502384e-05,
      "loss": 1.4359,
      "step": 73970
    },
    {
      "epoch": 3.1073043507714195,
      "grad_norm": 0.18862365186214447,
      "learning_rate": 7.223026518144578e-05,
      "loss": 1.4132,
      "step": 73980
    },
    {
      "epoch": 3.1078462919369314,
      "grad_norm": 0.21314261853694916,
      "learning_rate": 7.221563292345728e-05,
      "loss": 1.4295,
      "step": 73990
    },
    {
      "epoch": 3.1083882331024437,
      "grad_norm": 0.3071286976337433,
      "learning_rate": 7.220099853287091e-05,
      "loss": 1.4306,
      "step": 74000
    },
    {
      "epoch": 3.108930174267956,
      "grad_norm": 0.20251207053661346,
      "learning_rate": 7.218636201149952e-05,
      "loss": 1.4239,
      "step": 74010
    },
    {
      "epoch": 3.1094721154334684,
      "grad_norm": 0.18129009008407593,
      "learning_rate": 7.217172336115622e-05,
      "loss": 1.4233,
      "step": 74020
    },
    {
      "epoch": 3.1100140565989807,
      "grad_norm": 0.21674220263957977,
      "learning_rate": 7.21570825836544e-05,
      "loss": 1.426,
      "step": 74030
    },
    {
      "epoch": 3.110393415414839,
      "eval_loss": 2.4165384769439697,
      "eval_runtime": 21.9918,
      "eval_samples_per_second": 227.357,
      "eval_steps_per_second": 1.228,
      "step": 74037
    },
    {
      "epoch": 3.1105559977644925,
      "grad_norm": 0.23506887257099152,
      "learning_rate": 7.214243968080766e-05,
      "loss": 1.4339,
      "step": 74040
    },
    {
      "epoch": 3.111097938930005,
      "grad_norm": 0.19796548783779144,
      "learning_rate": 7.212779465442993e-05,
      "loss": 1.4343,
      "step": 74050
    },
    {
      "epoch": 3.111639880095517,
      "grad_norm": 0.22716623544692993,
      "learning_rate": 7.211314750633534e-05,
      "loss": 1.4137,
      "step": 74060
    },
    {
      "epoch": 3.1121818212610295,
      "grad_norm": 0.3029038608074188,
      "learning_rate": 7.209849823833835e-05,
      "loss": 1.4262,
      "step": 74070
    },
    {
      "epoch": 3.1127237624265414,
      "grad_norm": 0.175888329744339,
      "learning_rate": 7.208384685225364e-05,
      "loss": 1.4147,
      "step": 74080
    },
    {
      "epoch": 3.1132657035920537,
      "grad_norm": 0.28215131163597107,
      "learning_rate": 7.206919334989613e-05,
      "loss": 1.4238,
      "step": 74090
    },
    {
      "epoch": 3.113807644757566,
      "grad_norm": 0.26978573203086853,
      "learning_rate": 7.205453773308107e-05,
      "loss": 1.4235,
      "step": 74100
    },
    {
      "epoch": 3.1143495859230783,
      "grad_norm": 0.16594207286834717,
      "learning_rate": 7.20398800036239e-05,
      "loss": 1.4189,
      "step": 74110
    },
    {
      "epoch": 3.1148915270885906,
      "grad_norm": 0.22124223411083221,
      "learning_rate": 7.20252201633404e-05,
      "loss": 1.4344,
      "step": 74120
    },
    {
      "epoch": 3.1151083035547953,
      "eval_loss": 2.4167044162750244,
      "eval_runtime": 21.9938,
      "eval_samples_per_second": 227.337,
      "eval_steps_per_second": 1.228,
      "step": 74124
    },
    {
      "epoch": 3.1154334682541025,
      "grad_norm": 0.16019515693187714,
      "learning_rate": 7.201055821404653e-05,
      "loss": 1.4233,
      "step": 74130
    },
    {
      "epoch": 3.115975409419615,
      "grad_norm": 0.19272254407405853,
      "learning_rate": 7.199589415755856e-05,
      "loss": 1.4239,
      "step": 74140
    },
    {
      "epoch": 3.116517350585127,
      "grad_norm": 0.3636123538017273,
      "learning_rate": 7.198122799569302e-05,
      "loss": 1.4233,
      "step": 74150
    },
    {
      "epoch": 3.1170592917506394,
      "grad_norm": 0.31708648800849915,
      "learning_rate": 7.196655973026667e-05,
      "loss": 1.4282,
      "step": 74160
    },
    {
      "epoch": 3.1176012329161518,
      "grad_norm": 0.19407716393470764,
      "learning_rate": 7.195188936309657e-05,
      "loss": 1.4302,
      "step": 74170
    },
    {
      "epoch": 3.1181431740816636,
      "grad_norm": 0.17568087577819824,
      "learning_rate": 7.193721689600002e-05,
      "loss": 1.4249,
      "step": 74180
    },
    {
      "epoch": 3.118685115247176,
      "grad_norm": 0.36815229058265686,
      "learning_rate": 7.192254233079455e-05,
      "loss": 1.4195,
      "step": 74190
    },
    {
      "epoch": 3.1192270564126883,
      "grad_norm": 0.21112428605556488,
      "learning_rate": 7.190786566929802e-05,
      "loss": 1.4215,
      "step": 74200
    },
    {
      "epoch": 3.1197689975782006,
      "grad_norm": 0.19633443653583527,
      "learning_rate": 7.189318691332851e-05,
      "loss": 1.4301,
      "step": 74210
    },
    {
      "epoch": 3.119823191694752,
      "eval_loss": 2.416978359222412,
      "eval_runtime": 21.9898,
      "eval_samples_per_second": 227.378,
      "eval_steps_per_second": 1.228,
      "step": 74211
    },
    {
      "epoch": 3.1203109387437125,
      "grad_norm": 0.32902562618255615,
      "learning_rate": 7.187850606470435e-05,
      "loss": 1.4162,
      "step": 74220
    },
    {
      "epoch": 3.1208528799092248,
      "grad_norm": 0.19823166728019714,
      "learning_rate": 7.186382312524411e-05,
      "loss": 1.4167,
      "step": 74230
    },
    {
      "epoch": 3.121394821074737,
      "grad_norm": 0.1761886328458786,
      "learning_rate": 7.18491380967667e-05,
      "loss": 1.4312,
      "step": 74240
    },
    {
      "epoch": 3.1219367622402494,
      "grad_norm": 0.3041189908981323,
      "learning_rate": 7.18344509810912e-05,
      "loss": 1.4233,
      "step": 74250
    },
    {
      "epoch": 3.1224787034057617,
      "grad_norm": 0.39153873920440674,
      "learning_rate": 7.181976178003701e-05,
      "loss": 1.4268,
      "step": 74260
    },
    {
      "epoch": 3.1230206445712736,
      "grad_norm": 0.3814428448677063,
      "learning_rate": 7.180507049542376e-05,
      "loss": 1.4169,
      "step": 74270
    },
    {
      "epoch": 3.123562585736786,
      "grad_norm": 0.29164746403694153,
      "learning_rate": 7.179037712907131e-05,
      "loss": 1.4271,
      "step": 74280
    },
    {
      "epoch": 3.1241045269022982,
      "grad_norm": 0.20582854747772217,
      "learning_rate": 7.177568168279987e-05,
      "loss": 1.4168,
      "step": 74290
    },
    {
      "epoch": 3.124538079834708,
      "eval_loss": 2.4158596992492676,
      "eval_runtime": 22.0314,
      "eval_samples_per_second": 226.949,
      "eval_steps_per_second": 1.226,
      "step": 74298
    },
    {
      "epoch": 3.1246464680678105,
      "grad_norm": 0.18478041887283325,
      "learning_rate": 7.17609841584298e-05,
      "loss": 1.4319,
      "step": 74300
    },
    {
      "epoch": 3.1251884092333224,
      "grad_norm": 0.2710546851158142,
      "learning_rate": 7.174628455778178e-05,
      "loss": 1.4217,
      "step": 74310
    },
    {
      "epoch": 3.1257303503988347,
      "grad_norm": 0.18582426011562347,
      "learning_rate": 7.173158288267674e-05,
      "loss": 1.427,
      "step": 74320
    },
    {
      "epoch": 3.126272291564347,
      "grad_norm": 0.1683790534734726,
      "learning_rate": 7.171687913493586e-05,
      "loss": 1.4333,
      "step": 74330
    },
    {
      "epoch": 3.1268142327298594,
      "grad_norm": 0.16057170927524567,
      "learning_rate": 7.170217331638056e-05,
      "loss": 1.4262,
      "step": 74340
    },
    {
      "epoch": 3.1273561738953717,
      "grad_norm": 0.19682690501213074,
      "learning_rate": 7.168746542883254e-05,
      "loss": 1.4231,
      "step": 74350
    },
    {
      "epoch": 3.1278981150608836,
      "grad_norm": 0.2511323094367981,
      "learning_rate": 7.167275547411376e-05,
      "loss": 1.4138,
      "step": 74360
    },
    {
      "epoch": 3.128440056226396,
      "grad_norm": 0.2534506320953369,
      "learning_rate": 7.16580434540464e-05,
      "loss": 1.4225,
      "step": 74370
    },
    {
      "epoch": 3.128981997391908,
      "grad_norm": 0.26025325059890747,
      "learning_rate": 7.164332937045295e-05,
      "loss": 1.425,
      "step": 74380
    },
    {
      "epoch": 3.129252967974664,
      "eval_loss": 2.41723370552063,
      "eval_runtime": 22.0929,
      "eval_samples_per_second": 226.317,
      "eval_steps_per_second": 1.222,
      "step": 74385
    },
    {
      "epoch": 3.1295239385574205,
      "grad_norm": 0.3324603736400604,
      "learning_rate": 7.162861322515613e-05,
      "loss": 1.43,
      "step": 74390
    },
    {
      "epoch": 3.1300658797229324,
      "grad_norm": 0.21369896829128265,
      "learning_rate": 7.161389501997889e-05,
      "loss": 1.4211,
      "step": 74400
    },
    {
      "epoch": 3.1306078208884447,
      "grad_norm": 0.5364189743995667,
      "learning_rate": 7.159917475674447e-05,
      "loss": 1.424,
      "step": 74410
    },
    {
      "epoch": 3.131149762053957,
      "grad_norm": 0.29370662569999695,
      "learning_rate": 7.158445243727632e-05,
      "loss": 1.4254,
      "step": 74420
    },
    {
      "epoch": 3.1316917032194693,
      "grad_norm": 0.4463825821876526,
      "learning_rate": 7.156972806339823e-05,
      "loss": 1.4264,
      "step": 74430
    },
    {
      "epoch": 3.1322336443849816,
      "grad_norm": 0.181520015001297,
      "learning_rate": 7.155500163693417e-05,
      "loss": 1.422,
      "step": 74440
    },
    {
      "epoch": 3.1327755855504935,
      "grad_norm": 0.18724457919597626,
      "learning_rate": 7.154027315970838e-05,
      "loss": 1.4197,
      "step": 74450
    },
    {
      "epoch": 3.133317526716006,
      "grad_norm": 0.19526837766170502,
      "learning_rate": 7.152554263354535e-05,
      "loss": 1.4262,
      "step": 74460
    },
    {
      "epoch": 3.133859467881518,
      "grad_norm": 0.21833248436450958,
      "learning_rate": 7.151081006026984e-05,
      "loss": 1.4281,
      "step": 74470
    },
    {
      "epoch": 3.1339678561146207,
      "eval_loss": 2.4146862030029297,
      "eval_runtime": 22.0465,
      "eval_samples_per_second": 226.794,
      "eval_steps_per_second": 1.225,
      "step": 74472
    },
    {
      "epoch": 3.1344014090470305,
      "grad_norm": 0.22797681391239166,
      "learning_rate": 7.149607544170687e-05,
      "loss": 1.4222,
      "step": 74480
    },
    {
      "epoch": 3.1349433502125423,
      "grad_norm": 0.16637133061885834,
      "learning_rate": 7.14813387796817e-05,
      "loss": 1.4116,
      "step": 74490
    },
    {
      "epoch": 3.1354852913780547,
      "grad_norm": 0.28121089935302734,
      "learning_rate": 7.146660007601984e-05,
      "loss": 1.417,
      "step": 74500
    },
    {
      "epoch": 3.136027232543567,
      "grad_norm": 0.19828984141349792,
      "learning_rate": 7.145185933254705e-05,
      "loss": 1.4322,
      "step": 74510
    },
    {
      "epoch": 3.1365691737090793,
      "grad_norm": 0.2876355051994324,
      "learning_rate": 7.143711655108935e-05,
      "loss": 1.435,
      "step": 74520
    },
    {
      "epoch": 3.1371111148745916,
      "grad_norm": 0.1621021181344986,
      "learning_rate": 7.1422371733473e-05,
      "loss": 1.4288,
      "step": 74530
    },
    {
      "epoch": 3.1376530560401035,
      "grad_norm": 0.3681042492389679,
      "learning_rate": 7.140762488152457e-05,
      "loss": 1.429,
      "step": 74540
    },
    {
      "epoch": 3.138194997205616,
      "grad_norm": 0.22536900639533997,
      "learning_rate": 7.139287599707078e-05,
      "loss": 1.4203,
      "step": 74550
    },
    {
      "epoch": 3.138682744254577,
      "eval_loss": 2.4159655570983887,
      "eval_runtime": 22.1157,
      "eval_samples_per_second": 226.084,
      "eval_steps_per_second": 1.221,
      "step": 74559
    },
    {
      "epoch": 3.138736938371128,
      "grad_norm": 0.18401791155338287,
      "learning_rate": 7.137812508193868e-05,
      "loss": 1.4321,
      "step": 74560
    },
    {
      "epoch": 3.1392788795366404,
      "grad_norm": 0.38875266909599304,
      "learning_rate": 7.136337213795556e-05,
      "loss": 1.4213,
      "step": 74570
    },
    {
      "epoch": 3.1398208207021527,
      "grad_norm": 0.28335729241371155,
      "learning_rate": 7.134861716694894e-05,
      "loss": 1.4259,
      "step": 74580
    },
    {
      "epoch": 3.1403627618676646,
      "grad_norm": 0.4349845051765442,
      "learning_rate": 7.13338601707466e-05,
      "loss": 1.4315,
      "step": 74590
    },
    {
      "epoch": 3.140904703033177,
      "grad_norm": 0.2121918946504593,
      "learning_rate": 7.131910115117659e-05,
      "loss": 1.4343,
      "step": 74600
    },
    {
      "epoch": 3.1414466441986892,
      "grad_norm": 0.281886488199234,
      "learning_rate": 7.130434011006716e-05,
      "loss": 1.4218,
      "step": 74610
    },
    {
      "epoch": 3.1419885853642016,
      "grad_norm": 0.1884024739265442,
      "learning_rate": 7.128957704924689e-05,
      "loss": 1.4322,
      "step": 74620
    },
    {
      "epoch": 3.1425305265297134,
      "grad_norm": 0.3853340446949005,
      "learning_rate": 7.127481197054452e-05,
      "loss": 1.4274,
      "step": 74630
    },
    {
      "epoch": 3.1430724676952257,
      "grad_norm": 0.18657207489013672,
      "learning_rate": 7.12600448757891e-05,
      "loss": 1.4131,
      "step": 74640
    },
    {
      "epoch": 3.1433976323945334,
      "eval_loss": 2.421818733215332,
      "eval_runtime": 22.0955,
      "eval_samples_per_second": 226.29,
      "eval_steps_per_second": 1.222,
      "step": 74646
    },
    {
      "epoch": 3.143614408860738,
      "grad_norm": 0.17503514885902405,
      "learning_rate": 7.124527576680993e-05,
      "loss": 1.425,
      "step": 74650
    },
    {
      "epoch": 3.1441563500262504,
      "grad_norm": 0.17085334658622742,
      "learning_rate": 7.123050464543652e-05,
      "loss": 1.4221,
      "step": 74660
    },
    {
      "epoch": 3.1446982911917627,
      "grad_norm": 0.19825150072574615,
      "learning_rate": 7.121573151349863e-05,
      "loss": 1.4195,
      "step": 74670
    },
    {
      "epoch": 3.1452402323572746,
      "grad_norm": 0.29708579182624817,
      "learning_rate": 7.120095637282636e-05,
      "loss": 1.4341,
      "step": 74680
    },
    {
      "epoch": 3.145782173522787,
      "grad_norm": 0.2100587636232376,
      "learning_rate": 7.118617922524996e-05,
      "loss": 1.4208,
      "step": 74690
    },
    {
      "epoch": 3.146324114688299,
      "grad_norm": 0.2220253199338913,
      "learning_rate": 7.117140007259993e-05,
      "loss": 1.4305,
      "step": 74700
    },
    {
      "epoch": 3.1468660558538115,
      "grad_norm": 0.17661844193935394,
      "learning_rate": 7.115661891670706e-05,
      "loss": 1.4279,
      "step": 74710
    },
    {
      "epoch": 3.1474079970193234,
      "grad_norm": 0.19139379262924194,
      "learning_rate": 7.11418357594024e-05,
      "loss": 1.4318,
      "step": 74720
    },
    {
      "epoch": 3.1479499381848357,
      "grad_norm": 0.17980216443538666,
      "learning_rate": 7.11270506025172e-05,
      "loss": 1.4247,
      "step": 74730
    },
    {
      "epoch": 3.1481125205344895,
      "eval_loss": 2.416665554046631,
      "eval_runtime": 22.0909,
      "eval_samples_per_second": 226.337,
      "eval_steps_per_second": 1.222,
      "step": 74733
    },
    {
      "epoch": 3.148491879350348,
      "grad_norm": 0.24544402956962585,
      "learning_rate": 7.111226344788298e-05,
      "loss": 1.4119,
      "step": 74740
    },
    {
      "epoch": 3.1490338205158603,
      "grad_norm": 0.26411178708076477,
      "learning_rate": 7.109747429733154e-05,
      "loss": 1.4213,
      "step": 74750
    },
    {
      "epoch": 3.1495757616813727,
      "grad_norm": 0.2932426333427429,
      "learning_rate": 7.108268315269485e-05,
      "loss": 1.4356,
      "step": 74760
    },
    {
      "epoch": 3.1501177028468845,
      "grad_norm": 0.35350051522254944,
      "learning_rate": 7.106789001580521e-05,
      "loss": 1.4253,
      "step": 74770
    },
    {
      "epoch": 3.150659644012397,
      "grad_norm": 0.1805257648229599,
      "learning_rate": 7.10530948884951e-05,
      "loss": 1.4194,
      "step": 74780
    },
    {
      "epoch": 3.151201585177909,
      "grad_norm": 0.20616401731967926,
      "learning_rate": 7.10382977725973e-05,
      "loss": 1.4175,
      "step": 74790
    },
    {
      "epoch": 3.1517435263434215,
      "grad_norm": 0.16186383366584778,
      "learning_rate": 7.10234986699448e-05,
      "loss": 1.4185,
      "step": 74800
    },
    {
      "epoch": 3.152285467508934,
      "grad_norm": 0.31783926486968994,
      "learning_rate": 7.100869758237085e-05,
      "loss": 1.4254,
      "step": 74810
    },
    {
      "epoch": 3.1528274086744457,
      "grad_norm": 0.20256149768829346,
      "learning_rate": 7.099389451170897e-05,
      "loss": 1.4193,
      "step": 74820
    },
    {
      "epoch": 3.1528274086744457,
      "eval_loss": 2.4210216999053955,
      "eval_runtime": 22.0823,
      "eval_samples_per_second": 226.426,
      "eval_steps_per_second": 1.223,
      "step": 74820
    },
    {
      "epoch": 3.153369349839958,
      "grad_norm": 0.169488787651062,
      "learning_rate": 7.097908945979284e-05,
      "loss": 1.4263,
      "step": 74830
    },
    {
      "epoch": 3.1539112910054703,
      "grad_norm": 0.1810005009174347,
      "learning_rate": 7.09642824284565e-05,
      "loss": 1.4223,
      "step": 74840
    },
    {
      "epoch": 3.1544532321709826,
      "grad_norm": 0.21549201011657715,
      "learning_rate": 7.094947341953417e-05,
      "loss": 1.4189,
      "step": 74850
    },
    {
      "epoch": 3.1549951733364945,
      "grad_norm": 0.274050772190094,
      "learning_rate": 7.09346624348603e-05,
      "loss": 1.4256,
      "step": 74860
    },
    {
      "epoch": 3.155537114502007,
      "grad_norm": 0.21615564823150635,
      "learning_rate": 7.091984947626964e-05,
      "loss": 1.4145,
      "step": 74870
    },
    {
      "epoch": 3.156079055667519,
      "grad_norm": 0.17899556457996368,
      "learning_rate": 7.090503454559715e-05,
      "loss": 1.4263,
      "step": 74880
    },
    {
      "epoch": 3.1566209968330314,
      "grad_norm": 0.2249787449836731,
      "learning_rate": 7.089021764467803e-05,
      "loss": 1.4118,
      "step": 74890
    },
    {
      "epoch": 3.1571629379985433,
      "grad_norm": 0.17908982932567596,
      "learning_rate": 7.087539877534772e-05,
      "loss": 1.4206,
      "step": 74900
    },
    {
      "epoch": 3.1575422968144022,
      "eval_loss": 2.4155755043029785,
      "eval_runtime": 21.9967,
      "eval_samples_per_second": 227.307,
      "eval_steps_per_second": 1.227,
      "step": 74907
    },
    {
      "epoch": 3.1577048791640556,
      "grad_norm": 0.17990335822105408,
      "learning_rate": 7.086057793944195e-05,
      "loss": 1.4283,
      "step": 74910
    },
    {
      "epoch": 3.158246820329568,
      "grad_norm": 0.19940659403800964,
      "learning_rate": 7.084575513879664e-05,
      "loss": 1.4323,
      "step": 74920
    },
    {
      "epoch": 3.1587887614950803,
      "grad_norm": 0.1748953014612198,
      "learning_rate": 7.083093037524798e-05,
      "loss": 1.4191,
      "step": 74930
    },
    {
      "epoch": 3.1593307026605926,
      "grad_norm": 0.16563956439495087,
      "learning_rate": 7.08161036506324e-05,
      "loss": 1.4231,
      "step": 74940
    },
    {
      "epoch": 3.1598726438261044,
      "grad_norm": 0.21732689440250397,
      "learning_rate": 7.080127496678655e-05,
      "loss": 1.4266,
      "step": 74950
    },
    {
      "epoch": 3.1604145849916168,
      "grad_norm": 0.17851050198078156,
      "learning_rate": 7.078644432554738e-05,
      "loss": 1.4195,
      "step": 74960
    },
    {
      "epoch": 3.160956526157129,
      "grad_norm": 0.2197248935699463,
      "learning_rate": 7.077161172875202e-05,
      "loss": 1.4252,
      "step": 74970
    },
    {
      "epoch": 3.1614984673226414,
      "grad_norm": 0.2679365575313568,
      "learning_rate": 7.075677717823787e-05,
      "loss": 1.4283,
      "step": 74980
    },
    {
      "epoch": 3.1620404084881537,
      "grad_norm": 0.18566496670246124,
      "learning_rate": 7.074194067584256e-05,
      "loss": 1.4232,
      "step": 74990
    },
    {
      "epoch": 3.1622571849543584,
      "eval_loss": 2.4319992065429688,
      "eval_runtime": 21.9934,
      "eval_samples_per_second": 227.341,
      "eval_steps_per_second": 1.228,
      "step": 74994
    },
    {
      "epoch": 3.1625823496536656,
      "grad_norm": 0.28876206278800964,
      "learning_rate": 7.072710222340399e-05,
      "loss": 1.4202,
      "step": 75000
    },
    {
      "epoch": 3.163124290819178,
      "grad_norm": 0.22490562498569489,
      "learning_rate": 7.07122618227603e-05,
      "loss": 1.424,
      "step": 75010
    },
    {
      "epoch": 3.16366623198469,
      "grad_norm": 0.17399635910987854,
      "learning_rate": 7.069741947574981e-05,
      "loss": 1.4218,
      "step": 75020
    },
    {
      "epoch": 3.1642081731502025,
      "grad_norm": 0.2350047528743744,
      "learning_rate": 7.068257518421116e-05,
      "loss": 1.4214,
      "step": 75030
    },
    {
      "epoch": 3.1647501143157144,
      "grad_norm": 0.5656566619873047,
      "learning_rate": 7.066772894998316e-05,
      "loss": 1.4225,
      "step": 75040
    },
    {
      "epoch": 3.1652920554812267,
      "grad_norm": 0.22501088678836823,
      "learning_rate": 7.065288077490493e-05,
      "loss": 1.4295,
      "step": 75050
    },
    {
      "epoch": 3.165833996646739,
      "grad_norm": 0.2622585594654083,
      "learning_rate": 7.06380306608158e-05,
      "loss": 1.411,
      "step": 75060
    },
    {
      "epoch": 3.1663759378122514,
      "grad_norm": 0.2026064693927765,
      "learning_rate": 7.062317860955529e-05,
      "loss": 1.4129,
      "step": 75070
    },
    {
      "epoch": 3.1669178789777637,
      "grad_norm": 0.16883380711078644,
      "learning_rate": 7.060832462296329e-05,
      "loss": 1.4127,
      "step": 75080
    },
    {
      "epoch": 3.1669720730943145,
      "eval_loss": 2.4277634620666504,
      "eval_runtime": 22.0408,
      "eval_samples_per_second": 226.852,
      "eval_steps_per_second": 1.225,
      "step": 75081
    },
    {
      "epoch": 3.1674598201432755,
      "grad_norm": 0.17658884823322296,
      "learning_rate": 7.059346870287978e-05,
      "loss": 1.4141,
      "step": 75090
    },
    {
      "epoch": 3.168001761308788,
      "grad_norm": 0.16324864327907562,
      "learning_rate": 7.057861085114506e-05,
      "loss": 1.4321,
      "step": 75100
    },
    {
      "epoch": 3.1685437024743,
      "grad_norm": 0.17185302078723907,
      "learning_rate": 7.056375106959967e-05,
      "loss": 1.4238,
      "step": 75110
    },
    {
      "epoch": 3.1690856436398125,
      "grad_norm": 0.23366904258728027,
      "learning_rate": 7.054888936008437e-05,
      "loss": 1.4175,
      "step": 75120
    },
    {
      "epoch": 3.1696275848053244,
      "grad_norm": 0.23290032148361206,
      "learning_rate": 7.053402572444017e-05,
      "loss": 1.4307,
      "step": 75130
    },
    {
      "epoch": 3.1701695259708367,
      "grad_norm": 0.19138121604919434,
      "learning_rate": 7.05191601645083e-05,
      "loss": 1.4317,
      "step": 75140
    },
    {
      "epoch": 3.170711467136349,
      "grad_norm": 0.1849122941493988,
      "learning_rate": 7.050429268213023e-05,
      "loss": 1.431,
      "step": 75150
    },
    {
      "epoch": 3.1712534083018613,
      "grad_norm": 0.2830689549446106,
      "learning_rate": 7.04894232791477e-05,
      "loss": 1.413,
      "step": 75160
    },
    {
      "epoch": 3.171686961234271,
      "eval_loss": 2.426955223083496,
      "eval_runtime": 22.0447,
      "eval_samples_per_second": 226.812,
      "eval_steps_per_second": 1.225,
      "step": 75168
    },
    {
      "epoch": 3.1717953494673736,
      "grad_norm": 0.21067143976688385,
      "learning_rate": 7.047455195740268e-05,
      "loss": 1.4298,
      "step": 75170
    },
    {
      "epoch": 3.1723372906328855,
      "grad_norm": 0.2845075726509094,
      "learning_rate": 7.045967871873734e-05,
      "loss": 1.4164,
      "step": 75180
    },
    {
      "epoch": 3.172879231798398,
      "grad_norm": 0.2222658097743988,
      "learning_rate": 7.044480356499412e-05,
      "loss": 1.4213,
      "step": 75190
    },
    {
      "epoch": 3.17342117296391,
      "grad_norm": 0.2342739850282669,
      "learning_rate": 7.042992649801568e-05,
      "loss": 1.425,
      "step": 75200
    },
    {
      "epoch": 3.1739631141294224,
      "grad_norm": 0.16035151481628418,
      "learning_rate": 7.041504751964494e-05,
      "loss": 1.4337,
      "step": 75210
    },
    {
      "epoch": 3.1745050552949348,
      "grad_norm": 0.1617584526538849,
      "learning_rate": 7.040016663172505e-05,
      "loss": 1.4214,
      "step": 75220
    },
    {
      "epoch": 3.1750469964604466,
      "grad_norm": 0.18865042924880981,
      "learning_rate": 7.038528383609936e-05,
      "loss": 1.4262,
      "step": 75230
    },
    {
      "epoch": 3.175588937625959,
      "grad_norm": 0.15919503569602966,
      "learning_rate": 7.037039913461152e-05,
      "loss": 1.4259,
      "step": 75240
    },
    {
      "epoch": 3.1761308787914713,
      "grad_norm": 0.18104073405265808,
      "learning_rate": 7.035551252910535e-05,
      "loss": 1.4228,
      "step": 75250
    },
    {
      "epoch": 3.176401849374227,
      "eval_loss": 2.4324724674224854,
      "eval_runtime": 22.1726,
      "eval_samples_per_second": 225.504,
      "eval_steps_per_second": 1.218,
      "step": 75255
    },
    {
      "epoch": 3.1766728199569836,
      "grad_norm": 0.24971145391464233,
      "learning_rate": 7.034062402142497e-05,
      "loss": 1.4224,
      "step": 75260
    },
    {
      "epoch": 3.1772147611224955,
      "grad_norm": 0.22338011860847473,
      "learning_rate": 7.032573361341469e-05,
      "loss": 1.4086,
      "step": 75270
    },
    {
      "epoch": 3.1777567022880078,
      "grad_norm": 0.2989422082901001,
      "learning_rate": 7.031084130691905e-05,
      "loss": 1.4188,
      "step": 75280
    },
    {
      "epoch": 3.17829864345352,
      "grad_norm": 0.38781970739364624,
      "learning_rate": 7.029594710378288e-05,
      "loss": 1.4263,
      "step": 75290
    },
    {
      "epoch": 3.1788405846190324,
      "grad_norm": 0.27515822649002075,
      "learning_rate": 7.028105100585118e-05,
      "loss": 1.4258,
      "step": 75300
    },
    {
      "epoch": 3.1793825257845447,
      "grad_norm": 0.24076929688453674,
      "learning_rate": 7.026615301496923e-05,
      "loss": 1.4264,
      "step": 75310
    },
    {
      "epoch": 3.1799244669500566,
      "grad_norm": 0.1977168768644333,
      "learning_rate": 7.025125313298253e-05,
      "loss": 1.4267,
      "step": 75320
    },
    {
      "epoch": 3.180466408115569,
      "grad_norm": 0.20575913786888123,
      "learning_rate": 7.02363513617368e-05,
      "loss": 1.4178,
      "step": 75330
    },
    {
      "epoch": 3.1810083492810812,
      "grad_norm": 0.25656020641326904,
      "learning_rate": 7.022144770307801e-05,
      "loss": 1.4224,
      "step": 75340
    },
    {
      "epoch": 3.181116737514184,
      "eval_loss": 2.4162216186523438,
      "eval_runtime": 21.9936,
      "eval_samples_per_second": 227.339,
      "eval_steps_per_second": 1.228,
      "step": 75342
    },
    {
      "epoch": 3.1815502904465935,
      "grad_norm": 0.20052428543567657,
      "learning_rate": 7.020654215885236e-05,
      "loss": 1.4175,
      "step": 75350
    },
    {
      "epoch": 3.1820922316121054,
      "grad_norm": 0.2730656564235687,
      "learning_rate": 7.01916347309063e-05,
      "loss": 1.427,
      "step": 75360
    },
    {
      "epoch": 3.1826341727776177,
      "grad_norm": 0.19795359671115875,
      "learning_rate": 7.017672542108648e-05,
      "loss": 1.4233,
      "step": 75370
    },
    {
      "epoch": 3.18317611394313,
      "grad_norm": 0.25802189111709595,
      "learning_rate": 7.016181423123981e-05,
      "loss": 1.4181,
      "step": 75380
    },
    {
      "epoch": 3.1837180551086424,
      "grad_norm": 0.2566634714603424,
      "learning_rate": 7.01469011632134e-05,
      "loss": 1.4277,
      "step": 75390
    },
    {
      "epoch": 3.1842599962741547,
      "grad_norm": 0.18765607476234436,
      "learning_rate": 7.013198621885465e-05,
      "loss": 1.4108,
      "step": 75400
    },
    {
      "epoch": 3.1848019374396666,
      "grad_norm": 0.2734335660934448,
      "learning_rate": 7.011706940001113e-05,
      "loss": 1.4238,
      "step": 75410
    },
    {
      "epoch": 3.185343878605179,
      "grad_norm": 0.2739078402519226,
      "learning_rate": 7.010215070853071e-05,
      "loss": 1.4204,
      "step": 75420
    },
    {
      "epoch": 3.18583162565414,
      "eval_loss": 2.4123969078063965,
      "eval_runtime": 21.9921,
      "eval_samples_per_second": 227.354,
      "eval_steps_per_second": 1.228,
      "step": 75429
    },
    {
      "epoch": 3.185885819770691,
      "grad_norm": 0.18177048861980438,
      "learning_rate": 7.008723014626142e-05,
      "loss": 1.4138,
      "step": 75430
    },
    {
      "epoch": 3.1864277609362035,
      "grad_norm": 0.2982783615589142,
      "learning_rate": 7.007230771505155e-05,
      "loss": 1.4217,
      "step": 75440
    },
    {
      "epoch": 3.186969702101716,
      "grad_norm": 0.24831917881965637,
      "learning_rate": 7.005738341674964e-05,
      "loss": 1.4239,
      "step": 75450
    },
    {
      "epoch": 3.1875116432672277,
      "grad_norm": 0.20293238759040833,
      "learning_rate": 7.004245725320445e-05,
      "loss": 1.4171,
      "step": 75460
    },
    {
      "epoch": 3.18805358443274,
      "grad_norm": 0.27422091364860535,
      "learning_rate": 7.002752922626496e-05,
      "loss": 1.4212,
      "step": 75470
    },
    {
      "epoch": 3.1885955255982523,
      "grad_norm": 0.19502641260623932,
      "learning_rate": 7.001259933778041e-05,
      "loss": 1.4212,
      "step": 75480
    },
    {
      "epoch": 3.1891374667637646,
      "grad_norm": 0.1884526014328003,
      "learning_rate": 6.999766758960019e-05,
      "loss": 1.4176,
      "step": 75490
    },
    {
      "epoch": 3.1896794079292765,
      "grad_norm": 0.1656298190355301,
      "learning_rate": 6.998273398357406e-05,
      "loss": 1.429,
      "step": 75500
    },
    {
      "epoch": 3.190221349094789,
      "grad_norm": 0.17676104605197906,
      "learning_rate": 6.996779852155189e-05,
      "loss": 1.423,
      "step": 75510
    },
    {
      "epoch": 3.190546513794096,
      "eval_loss": 2.412719488143921,
      "eval_runtime": 21.9959,
      "eval_samples_per_second": 227.315,
      "eval_steps_per_second": 1.228,
      "step": 75516
    },
    {
      "epoch": 3.190763290260301,
      "grad_norm": 0.3444202244281769,
      "learning_rate": 6.99528612053838e-05,
      "loss": 1.4231,
      "step": 75520
    },
    {
      "epoch": 3.1913052314258135,
      "grad_norm": 0.20383748412132263,
      "learning_rate": 6.99379220369202e-05,
      "loss": 1.4305,
      "step": 75530
    },
    {
      "epoch": 3.1918471725913253,
      "grad_norm": 0.2849511206150055,
      "learning_rate": 6.992298101801167e-05,
      "loss": 1.4264,
      "step": 75540
    },
    {
      "epoch": 3.1923891137568376,
      "grad_norm": 0.20015236735343933,
      "learning_rate": 6.990803815050903e-05,
      "loss": 1.4288,
      "step": 75550
    },
    {
      "epoch": 3.19293105492235,
      "grad_norm": 0.19224634766578674,
      "learning_rate": 6.989309343626337e-05,
      "loss": 1.4244,
      "step": 75560
    },
    {
      "epoch": 3.1934729960878623,
      "grad_norm": 0.16140466928482056,
      "learning_rate": 6.987814687712592e-05,
      "loss": 1.4274,
      "step": 75570
    },
    {
      "epoch": 3.1940149372533746,
      "grad_norm": 0.1749192178249359,
      "learning_rate": 6.986319847494826e-05,
      "loss": 1.4347,
      "step": 75580
    },
    {
      "epoch": 3.1945568784188865,
      "grad_norm": 0.16757184267044067,
      "learning_rate": 6.98482482315821e-05,
      "loss": 1.419,
      "step": 75590
    },
    {
      "epoch": 3.195098819584399,
      "grad_norm": 0.15887753665447235,
      "learning_rate": 6.98332961488794e-05,
      "loss": 1.4307,
      "step": 75600
    },
    {
      "epoch": 3.1952614019340526,
      "eval_loss": 2.424544095993042,
      "eval_runtime": 21.9927,
      "eval_samples_per_second": 227.348,
      "eval_steps_per_second": 1.228,
      "step": 75603
    },
    {
      "epoch": 3.195640760749911,
      "grad_norm": 0.3752150237560272,
      "learning_rate": 6.98183422286924e-05,
      "loss": 1.421,
      "step": 75610
    },
    {
      "epoch": 3.1961827019154234,
      "grad_norm": 0.18618738651275635,
      "learning_rate": 6.980338647287347e-05,
      "loss": 1.4293,
      "step": 75620
    },
    {
      "epoch": 3.1967246430809357,
      "grad_norm": 0.19021612405776978,
      "learning_rate": 6.97884288832753e-05,
      "loss": 1.4218,
      "step": 75630
    },
    {
      "epoch": 3.1972665842464476,
      "grad_norm": 0.1741076409816742,
      "learning_rate": 6.977346946175078e-05,
      "loss": 1.4211,
      "step": 75640
    },
    {
      "epoch": 3.19780852541196,
      "grad_norm": 0.1816461980342865,
      "learning_rate": 6.975850821015298e-05,
      "loss": 1.4124,
      "step": 75650
    },
    {
      "epoch": 3.1983504665774722,
      "grad_norm": 0.22760295867919922,
      "learning_rate": 6.974354513033528e-05,
      "loss": 1.4251,
      "step": 75660
    },
    {
      "epoch": 3.1988924077429846,
      "grad_norm": 0.255193293094635,
      "learning_rate": 6.972858022415119e-05,
      "loss": 1.4269,
      "step": 75670
    },
    {
      "epoch": 3.1994343489084964,
      "grad_norm": 0.1877882033586502,
      "learning_rate": 6.971361349345455e-05,
      "loss": 1.4093,
      "step": 75680
    },
    {
      "epoch": 3.1999762900740087,
      "grad_norm": 0.17674243450164795,
      "learning_rate": 6.969864494009934e-05,
      "loss": 1.4263,
      "step": 75690
    },
    {
      "epoch": 3.1999762900740087,
      "eval_loss": 2.420743942260742,
      "eval_runtime": 21.9872,
      "eval_samples_per_second": 227.405,
      "eval_steps_per_second": 1.228,
      "step": 75690
    },
    {
      "epoch": 3.200518231239521,
      "grad_norm": 0.1850854456424713,
      "learning_rate": 6.968367456593983e-05,
      "loss": 1.4203,
      "step": 75700
    },
    {
      "epoch": 3.2010601724050334,
      "grad_norm": 0.25063949823379517,
      "learning_rate": 6.966870237283045e-05,
      "loss": 1.4245,
      "step": 75710
    },
    {
      "epoch": 3.2016021135705457,
      "grad_norm": 0.23362627625465393,
      "learning_rate": 6.96537283626259e-05,
      "loss": 1.4203,
      "step": 75720
    },
    {
      "epoch": 3.2021440547360576,
      "grad_norm": 0.17846883833408356,
      "learning_rate": 6.96387525371811e-05,
      "loss": 1.4251,
      "step": 75730
    },
    {
      "epoch": 3.20268599590157,
      "grad_norm": 0.2213391810655594,
      "learning_rate": 6.96237748983512e-05,
      "loss": 1.4155,
      "step": 75740
    },
    {
      "epoch": 3.203227937067082,
      "grad_norm": 0.1812654733657837,
      "learning_rate": 6.960879544799156e-05,
      "loss": 1.4178,
      "step": 75750
    },
    {
      "epoch": 3.2037698782325945,
      "grad_norm": 0.18986418843269348,
      "learning_rate": 6.959381418795775e-05,
      "loss": 1.4217,
      "step": 75760
    },
    {
      "epoch": 3.2043118193981064,
      "grad_norm": 0.15823641419410706,
      "learning_rate": 6.957883112010563e-05,
      "loss": 1.4217,
      "step": 75770
    },
    {
      "epoch": 3.2046911782139653,
      "eval_loss": 2.4173123836517334,
      "eval_runtime": 21.9872,
      "eval_samples_per_second": 227.405,
      "eval_steps_per_second": 1.228,
      "step": 75777
    },
    {
      "epoch": 3.2048537605636187,
      "grad_norm": 0.25696510076522827,
      "learning_rate": 6.956384624629117e-05,
      "loss": 1.4149,
      "step": 75780
    },
    {
      "epoch": 3.205395701729131,
      "grad_norm": 0.18557502329349518,
      "learning_rate": 6.954885956837067e-05,
      "loss": 1.4202,
      "step": 75790
    },
    {
      "epoch": 3.2059376428946433,
      "grad_norm": 0.2925853133201599,
      "learning_rate": 6.953387108820064e-05,
      "loss": 1.4157,
      "step": 75800
    },
    {
      "epoch": 3.2064795840601557,
      "grad_norm": 0.17210768163204193,
      "learning_rate": 6.951888080763772e-05,
      "loss": 1.4136,
      "step": 75810
    },
    {
      "epoch": 3.2070215252256675,
      "grad_norm": 0.17783254384994507,
      "learning_rate": 6.950388872853891e-05,
      "loss": 1.4264,
      "step": 75820
    },
    {
      "epoch": 3.20756346639118,
      "grad_norm": 0.1816641092300415,
      "learning_rate": 6.948889485276132e-05,
      "loss": 1.4121,
      "step": 75830
    },
    {
      "epoch": 3.208105407556692,
      "grad_norm": 0.25515398383140564,
      "learning_rate": 6.947389918216234e-05,
      "loss": 1.4282,
      "step": 75840
    },
    {
      "epoch": 3.2086473487222045,
      "grad_norm": 0.16730128228664398,
      "learning_rate": 6.94589017185996e-05,
      "loss": 1.4233,
      "step": 75850
    },
    {
      "epoch": 3.209189289887717,
      "grad_norm": 0.2866290509700775,
      "learning_rate": 6.944390246393085e-05,
      "loss": 1.4215,
      "step": 75860
    },
    {
      "epoch": 3.2094060663539214,
      "eval_loss": 2.418275833129883,
      "eval_runtime": 21.9933,
      "eval_samples_per_second": 227.342,
      "eval_steps_per_second": 1.228,
      "step": 75864
    },
    {
      "epoch": 3.2097312310532287,
      "grad_norm": 0.21577335894107819,
      "learning_rate": 6.942890142001418e-05,
      "loss": 1.4179,
      "step": 75870
    },
    {
      "epoch": 3.210273172218741,
      "grad_norm": 0.20668751001358032,
      "learning_rate": 6.941389858870785e-05,
      "loss": 1.4216,
      "step": 75880
    },
    {
      "epoch": 3.2108151133842533,
      "grad_norm": 0.23815388977527618,
      "learning_rate": 6.939889397187034e-05,
      "loss": 1.4147,
      "step": 75890
    },
    {
      "epoch": 3.2113570545497656,
      "grad_norm": 0.18798255920410156,
      "learning_rate": 6.938388757136036e-05,
      "loss": 1.4145,
      "step": 75900
    },
    {
      "epoch": 3.2118989957152775,
      "grad_norm": 0.24447955191135406,
      "learning_rate": 6.936887938903684e-05,
      "loss": 1.43,
      "step": 75910
    },
    {
      "epoch": 3.21244093688079,
      "grad_norm": 0.288022518157959,
      "learning_rate": 6.935386942675892e-05,
      "loss": 1.4171,
      "step": 75920
    },
    {
      "epoch": 3.212982878046302,
      "grad_norm": 0.22881217300891876,
      "learning_rate": 6.933885768638598e-05,
      "loss": 1.4228,
      "step": 75930
    },
    {
      "epoch": 3.2135248192118144,
      "grad_norm": 0.20432133972644806,
      "learning_rate": 6.932384416977759e-05,
      "loss": 1.4242,
      "step": 75940
    },
    {
      "epoch": 3.2140667603773263,
      "grad_norm": 0.21099141240119934,
      "learning_rate": 6.930882887879359e-05,
      "loss": 1.4182,
      "step": 75950
    },
    {
      "epoch": 3.2141209544938776,
      "eval_loss": 2.428762674331665,
      "eval_runtime": 21.9941,
      "eval_samples_per_second": 227.334,
      "eval_steps_per_second": 1.228,
      "step": 75951
    },
    {
      "epoch": 3.2146087015428386,
      "grad_norm": 0.19293814897537231,
      "learning_rate": 6.929381181529399e-05,
      "loss": 1.4111,
      "step": 75960
    },
    {
      "epoch": 3.215150642708351,
      "grad_norm": 0.209365114569664,
      "learning_rate": 6.927879298113901e-05,
      "loss": 1.4083,
      "step": 75970
    },
    {
      "epoch": 3.2156925838738633,
      "grad_norm": 0.20382948219776154,
      "learning_rate": 6.926377237818917e-05,
      "loss": 1.4229,
      "step": 75980
    },
    {
      "epoch": 3.2162345250393756,
      "grad_norm": 0.18336911499500275,
      "learning_rate": 6.924875000830513e-05,
      "loss": 1.4244,
      "step": 75990
    },
    {
      "epoch": 3.2167764662048874,
      "grad_norm": 0.18866483867168427,
      "learning_rate": 6.92337258733478e-05,
      "loss": 1.4241,
      "step": 76000
    },
    {
      "epoch": 3.2173184073703998,
      "grad_norm": 0.21788087487220764,
      "learning_rate": 6.921869997517834e-05,
      "loss": 1.4319,
      "step": 76010
    },
    {
      "epoch": 3.217860348535912,
      "grad_norm": 0.1925356537103653,
      "learning_rate": 6.9203672315658e-05,
      "loss": 1.4204,
      "step": 76020
    },
    {
      "epoch": 3.2184022897014244,
      "grad_norm": 0.24953441321849823,
      "learning_rate": 6.918864289664845e-05,
      "loss": 1.4209,
      "step": 76030
    },
    {
      "epoch": 3.218835842633834,
      "eval_loss": 2.4290771484375,
      "eval_runtime": 21.9942,
      "eval_samples_per_second": 227.333,
      "eval_steps_per_second": 1.228,
      "step": 76038
    },
    {
      "epoch": 3.2189442308669367,
      "grad_norm": 0.18474015593528748,
      "learning_rate": 6.917361172001139e-05,
      "loss": 1.4223,
      "step": 76040
    },
    {
      "epoch": 3.2194861720324486,
      "grad_norm": 0.27110907435417175,
      "learning_rate": 6.915857878760885e-05,
      "loss": 1.4256,
      "step": 76050
    },
    {
      "epoch": 3.220028113197961,
      "grad_norm": 0.19559454917907715,
      "learning_rate": 6.914354410130305e-05,
      "loss": 1.4299,
      "step": 76060
    },
    {
      "epoch": 3.220570054363473,
      "grad_norm": 0.16140873730182648,
      "learning_rate": 6.912850766295641e-05,
      "loss": 1.4068,
      "step": 76070
    },
    {
      "epoch": 3.2211119955289855,
      "grad_norm": 0.17691193521022797,
      "learning_rate": 6.911346947443157e-05,
      "loss": 1.4216,
      "step": 76080
    },
    {
      "epoch": 3.2216539366944974,
      "grad_norm": 0.173243448138237,
      "learning_rate": 6.909842953759142e-05,
      "loss": 1.4172,
      "step": 76090
    },
    {
      "epoch": 3.2221958778600097,
      "grad_norm": 0.3001802861690521,
      "learning_rate": 6.908338785429901e-05,
      "loss": 1.4205,
      "step": 76100
    },
    {
      "epoch": 3.222737819025522,
      "grad_norm": 0.21718403697013855,
      "learning_rate": 6.906834442641768e-05,
      "loss": 1.4201,
      "step": 76110
    },
    {
      "epoch": 3.2232797601910343,
      "grad_norm": 0.19174130260944366,
      "learning_rate": 6.905329925581089e-05,
      "loss": 1.4161,
      "step": 76120
    },
    {
      "epoch": 3.2235507307737903,
      "eval_loss": 2.4190409183502197,
      "eval_runtime": 21.9914,
      "eval_samples_per_second": 227.361,
      "eval_steps_per_second": 1.228,
      "step": 76125
    },
    {
      "epoch": 3.2238217013565467,
      "grad_norm": 0.28363704681396484,
      "learning_rate": 6.903825234434243e-05,
      "loss": 1.4222,
      "step": 76130
    },
    {
      "epoch": 3.2243636425220585,
      "grad_norm": 0.21950308978557587,
      "learning_rate": 6.902320369387621e-05,
      "loss": 1.4202,
      "step": 76140
    },
    {
      "epoch": 3.224905583687571,
      "grad_norm": 0.22058185935020447,
      "learning_rate": 6.90081533062764e-05,
      "loss": 1.4147,
      "step": 76150
    },
    {
      "epoch": 3.225447524853083,
      "grad_norm": 0.354383260011673,
      "learning_rate": 6.899310118340736e-05,
      "loss": 1.4167,
      "step": 76160
    },
    {
      "epoch": 3.2259894660185955,
      "grad_norm": 0.3099483251571655,
      "learning_rate": 6.897804732713371e-05,
      "loss": 1.4339,
      "step": 76170
    },
    {
      "epoch": 3.2265314071841074,
      "grad_norm": 0.33063191175460815,
      "learning_rate": 6.896299173932023e-05,
      "loss": 1.4213,
      "step": 76180
    },
    {
      "epoch": 3.2270733483496197,
      "grad_norm": 0.17252486944198608,
      "learning_rate": 6.894793442183199e-05,
      "loss": 1.4295,
      "step": 76190
    },
    {
      "epoch": 3.227615289515132,
      "grad_norm": 0.20971451699733734,
      "learning_rate": 6.893287537653417e-05,
      "loss": 1.4133,
      "step": 76200
    },
    {
      "epoch": 3.2281572306806443,
      "grad_norm": 0.16854190826416016,
      "learning_rate": 6.891781460529223e-05,
      "loss": 1.4208,
      "step": 76210
    },
    {
      "epoch": 3.228265618913747,
      "eval_loss": 2.4187822341918945,
      "eval_runtime": 21.9892,
      "eval_samples_per_second": 227.385,
      "eval_steps_per_second": 1.228,
      "step": 76212
    },
    {
      "epoch": 3.2286991718461566,
      "grad_norm": 0.1864662617444992,
      "learning_rate": 6.890275210997185e-05,
      "loss": 1.4251,
      "step": 76220
    },
    {
      "epoch": 3.2292411130116685,
      "grad_norm": 0.15556700527668,
      "learning_rate": 6.888768789243889e-05,
      "loss": 1.4134,
      "step": 76230
    },
    {
      "epoch": 3.229783054177181,
      "grad_norm": 0.17359033226966858,
      "learning_rate": 6.887262195455946e-05,
      "loss": 1.4146,
      "step": 76240
    },
    {
      "epoch": 3.230324995342693,
      "grad_norm": 0.29231011867523193,
      "learning_rate": 6.885755429819985e-05,
      "loss": 1.4274,
      "step": 76250
    },
    {
      "epoch": 3.2308669365082054,
      "grad_norm": 0.16642163693904877,
      "learning_rate": 6.884248492522656e-05,
      "loss": 1.4075,
      "step": 76260
    },
    {
      "epoch": 3.2314088776737178,
      "grad_norm": 0.4484625458717346,
      "learning_rate": 6.882741383750635e-05,
      "loss": 1.4122,
      "step": 76270
    },
    {
      "epoch": 3.2319508188392296,
      "grad_norm": 0.14720787107944489,
      "learning_rate": 6.881234103690616e-05,
      "loss": 1.4155,
      "step": 76280
    },
    {
      "epoch": 3.232492760004742,
      "grad_norm": 0.19290123879909515,
      "learning_rate": 6.87972665252931e-05,
      "loss": 1.4226,
      "step": 76290
    },
    {
      "epoch": 3.232980507053703,
      "eval_loss": 2.4077162742614746,
      "eval_runtime": 21.993,
      "eval_samples_per_second": 227.345,
      "eval_steps_per_second": 1.228,
      "step": 76299
    },
    {
      "epoch": 3.2330347011702543,
      "grad_norm": 0.18437008559703827,
      "learning_rate": 6.878219030453459e-05,
      "loss": 1.4208,
      "step": 76300
    },
    {
      "epoch": 3.2335766423357666,
      "grad_norm": 0.1884869635105133,
      "learning_rate": 6.876711237649816e-05,
      "loss": 1.416,
      "step": 76310
    },
    {
      "epoch": 3.2341185835012785,
      "grad_norm": 0.19494092464447021,
      "learning_rate": 6.875203274305163e-05,
      "loss": 1.4133,
      "step": 76320
    },
    {
      "epoch": 3.2346605246667908,
      "grad_norm": 0.16780586540699005,
      "learning_rate": 6.8736951406063e-05,
      "loss": 1.4222,
      "step": 76330
    },
    {
      "epoch": 3.235202465832303,
      "grad_norm": 0.24331122636795044,
      "learning_rate": 6.872186836740046e-05,
      "loss": 1.4253,
      "step": 76340
    },
    {
      "epoch": 3.2357444069978154,
      "grad_norm": 0.2462978959083557,
      "learning_rate": 6.870678362893243e-05,
      "loss": 1.4215,
      "step": 76350
    },
    {
      "epoch": 3.2362863481633277,
      "grad_norm": 0.3127591907978058,
      "learning_rate": 6.869169719252756e-05,
      "loss": 1.4199,
      "step": 76360
    },
    {
      "epoch": 3.2368282893288396,
      "grad_norm": 0.19799236953258514,
      "learning_rate": 6.867660906005467e-05,
      "loss": 1.4101,
      "step": 76370
    },
    {
      "epoch": 3.237370230494352,
      "grad_norm": 0.4551452696323395,
      "learning_rate": 6.866151923338286e-05,
      "loss": 1.4229,
      "step": 76380
    },
    {
      "epoch": 3.237695395193659,
      "eval_loss": 2.4275617599487305,
      "eval_runtime": 21.99,
      "eval_samples_per_second": 227.376,
      "eval_steps_per_second": 1.228,
      "step": 76386
    },
    {
      "epoch": 3.2379121716598642,
      "grad_norm": 0.19475868344306946,
      "learning_rate": 6.864642771438136e-05,
      "loss": 1.4142,
      "step": 76390
    },
    {
      "epoch": 3.2384541128253765,
      "grad_norm": 0.5060222148895264,
      "learning_rate": 6.863133450491961e-05,
      "loss": 1.4153,
      "step": 76400
    },
    {
      "epoch": 3.2389960539908884,
      "grad_norm": 0.2045438289642334,
      "learning_rate": 6.861623960686734e-05,
      "loss": 1.413,
      "step": 76410
    },
    {
      "epoch": 3.2395379951564007,
      "grad_norm": 0.24009686708450317,
      "learning_rate": 6.860114302209443e-05,
      "loss": 1.4246,
      "step": 76420
    },
    {
      "epoch": 3.240079936321913,
      "grad_norm": 0.1824415624141693,
      "learning_rate": 6.858604475247097e-05,
      "loss": 1.4248,
      "step": 76430
    },
    {
      "epoch": 3.2406218774874254,
      "grad_norm": 0.19345873594284058,
      "learning_rate": 6.857094479986726e-05,
      "loss": 1.4301,
      "step": 76440
    },
    {
      "epoch": 3.2411638186529377,
      "grad_norm": 0.17767898738384247,
      "learning_rate": 6.855584316615384e-05,
      "loss": 1.4193,
      "step": 76450
    },
    {
      "epoch": 3.2417057598184496,
      "grad_norm": 0.24474495649337769,
      "learning_rate": 6.854073985320141e-05,
      "loss": 1.4279,
      "step": 76460
    },
    {
      "epoch": 3.242247700983962,
      "grad_norm": 0.2586842179298401,
      "learning_rate": 6.852563486288093e-05,
      "loss": 1.4149,
      "step": 76470
    },
    {
      "epoch": 3.2424102833336157,
      "eval_loss": 2.4281370639801025,
      "eval_runtime": 21.9931,
      "eval_samples_per_second": 227.344,
      "eval_steps_per_second": 1.228,
      "step": 76473
    },
    {
      "epoch": 3.242789642149474,
      "grad_norm": 0.20308630168437958,
      "learning_rate": 6.85105281970635e-05,
      "loss": 1.4204,
      "step": 76480
    },
    {
      "epoch": 3.2433315833149865,
      "grad_norm": 0.17210350930690765,
      "learning_rate": 6.849541985762053e-05,
      "loss": 1.4308,
      "step": 76490
    },
    {
      "epoch": 3.243873524480499,
      "grad_norm": 0.351959228515625,
      "learning_rate": 6.848030984642351e-05,
      "loss": 1.4203,
      "step": 76500
    },
    {
      "epoch": 3.2444154656460107,
      "grad_norm": 0.22208164632320404,
      "learning_rate": 6.846519816534423e-05,
      "loss": 1.4174,
      "step": 76510
    },
    {
      "epoch": 3.244957406811523,
      "grad_norm": 0.3489255905151367,
      "learning_rate": 6.845008481625468e-05,
      "loss": 1.4243,
      "step": 76520
    },
    {
      "epoch": 3.2454993479770353,
      "grad_norm": 0.20681335031986237,
      "learning_rate": 6.843496980102702e-05,
      "loss": 1.4338,
      "step": 76530
    },
    {
      "epoch": 3.2460412891425476,
      "grad_norm": 0.20921902358531952,
      "learning_rate": 6.841985312153362e-05,
      "loss": 1.4329,
      "step": 76540
    },
    {
      "epoch": 3.2465832303080595,
      "grad_norm": 0.2758723199367523,
      "learning_rate": 6.840473477964707e-05,
      "loss": 1.4197,
      "step": 76550
    },
    {
      "epoch": 3.247125171473572,
      "grad_norm": 0.19785663485527039,
      "learning_rate": 6.838961477724018e-05,
      "loss": 1.4155,
      "step": 76560
    },
    {
      "epoch": 3.247125171473572,
      "eval_loss": 2.4455416202545166,
      "eval_runtime": 21.9858,
      "eval_samples_per_second": 227.419,
      "eval_steps_per_second": 1.228,
      "step": 76560
    },
    {
      "epoch": 3.247667112639084,
      "grad_norm": 0.16675598919391632,
      "learning_rate": 6.837449311618595e-05,
      "loss": 1.4148,
      "step": 76570
    },
    {
      "epoch": 3.2482090538045965,
      "grad_norm": 0.2686634957790375,
      "learning_rate": 6.835936979835757e-05,
      "loss": 1.4247,
      "step": 76580
    },
    {
      "epoch": 3.2487509949701083,
      "grad_norm": 0.16704584658145905,
      "learning_rate": 6.834424482562846e-05,
      "loss": 1.412,
      "step": 76590
    },
    {
      "epoch": 3.2492929361356206,
      "grad_norm": 0.2601518929004669,
      "learning_rate": 6.832911819987224e-05,
      "loss": 1.4192,
      "step": 76600
    },
    {
      "epoch": 3.249834877301133,
      "grad_norm": 0.21344710886478424,
      "learning_rate": 6.831398992296273e-05,
      "loss": 1.4149,
      "step": 76610
    },
    {
      "epoch": 3.2503768184666453,
      "grad_norm": 0.267600953578949,
      "learning_rate": 6.829885999677395e-05,
      "loss": 1.4142,
      "step": 76620
    },
    {
      "epoch": 3.2509187596321576,
      "grad_norm": 0.1773858517408371,
      "learning_rate": 6.828372842318013e-05,
      "loss": 1.4171,
      "step": 76630
    },
    {
      "epoch": 3.2514607007976695,
      "grad_norm": 0.1974133402109146,
      "learning_rate": 6.826859520405572e-05,
      "loss": 1.4088,
      "step": 76640
    },
    {
      "epoch": 3.2518400596135284,
      "eval_loss": 2.4362905025482178,
      "eval_runtime": 21.9908,
      "eval_samples_per_second": 227.368,
      "eval_steps_per_second": 1.228,
      "step": 76647
    },
    {
      "epoch": 3.252002641963182,
      "grad_norm": 0.3224498927593231,
      "learning_rate": 6.825346034127531e-05,
      "loss": 1.417,
      "step": 76650
    },
    {
      "epoch": 3.252544583128694,
      "grad_norm": 0.32483381032943726,
      "learning_rate": 6.823832383671379e-05,
      "loss": 1.4267,
      "step": 76660
    },
    {
      "epoch": 3.2530865242942064,
      "grad_norm": 0.1879141479730606,
      "learning_rate": 6.82231856922462e-05,
      "loss": 1.4178,
      "step": 76670
    },
    {
      "epoch": 3.2536284654597187,
      "grad_norm": 0.21414776146411896,
      "learning_rate": 6.820804590974776e-05,
      "loss": 1.4175,
      "step": 76680
    },
    {
      "epoch": 3.2541704066252306,
      "grad_norm": 0.19383347034454346,
      "learning_rate": 6.819290449109395e-05,
      "loss": 1.4178,
      "step": 76690
    },
    {
      "epoch": 3.254712347790743,
      "grad_norm": 0.17373976111412048,
      "learning_rate": 6.81777614381604e-05,
      "loss": 1.4182,
      "step": 76700
    },
    {
      "epoch": 3.2552542889562552,
      "grad_norm": 0.18311308324337006,
      "learning_rate": 6.816261675282297e-05,
      "loss": 1.4162,
      "step": 76710
    },
    {
      "epoch": 3.2557962301217676,
      "grad_norm": 0.1700367033481598,
      "learning_rate": 6.814747043695772e-05,
      "loss": 1.4163,
      "step": 76720
    },
    {
      "epoch": 3.25633817128728,
      "grad_norm": 0.4533218443393707,
      "learning_rate": 6.813232249244093e-05,
      "loss": 1.4266,
      "step": 76730
    },
    {
      "epoch": 3.2565549477534845,
      "eval_loss": 2.439307689666748,
      "eval_runtime": 21.9896,
      "eval_samples_per_second": 227.38,
      "eval_steps_per_second": 1.228,
      "step": 76734
    },
    {
      "epoch": 3.2568801124527917,
      "grad_norm": 0.2232131063938141,
      "learning_rate": 6.811717292114904e-05,
      "loss": 1.4208,
      "step": 76740
    },
    {
      "epoch": 3.257422053618304,
      "grad_norm": 0.24600377678871155,
      "learning_rate": 6.810202172495873e-05,
      "loss": 1.4128,
      "step": 76750
    },
    {
      "epoch": 3.2579639947838164,
      "grad_norm": 0.1848914921283722,
      "learning_rate": 6.808686890574683e-05,
      "loss": 1.4298,
      "step": 76760
    },
    {
      "epoch": 3.2585059359493287,
      "grad_norm": 0.2184385061264038,
      "learning_rate": 6.807171446539042e-05,
      "loss": 1.407,
      "step": 76770
    },
    {
      "epoch": 3.2590478771148406,
      "grad_norm": 0.18162274360656738,
      "learning_rate": 6.80565584057668e-05,
      "loss": 1.4158,
      "step": 76780
    },
    {
      "epoch": 3.259589818280353,
      "grad_norm": 0.18968087434768677,
      "learning_rate": 6.804140072875338e-05,
      "loss": 1.416,
      "step": 76790
    },
    {
      "epoch": 3.260131759445865,
      "grad_norm": 0.19787006080150604,
      "learning_rate": 6.802624143622789e-05,
      "loss": 1.4232,
      "step": 76800
    },
    {
      "epoch": 3.2606737006113775,
      "grad_norm": 0.19424383342266083,
      "learning_rate": 6.801108053006815e-05,
      "loss": 1.4144,
      "step": 76810
    },
    {
      "epoch": 3.2612156417768894,
      "grad_norm": 0.23365655541419983,
      "learning_rate": 6.799591801215223e-05,
      "loss": 1.4212,
      "step": 76820
    },
    {
      "epoch": 3.2612698358934407,
      "eval_loss": 2.4371750354766846,
      "eval_runtime": 21.9951,
      "eval_samples_per_second": 227.324,
      "eval_steps_per_second": 1.228,
      "step": 76821
    },
    {
      "epoch": 3.2617575829424017,
      "grad_norm": 0.20863786339759827,
      "learning_rate": 6.798075388435845e-05,
      "loss": 1.4189,
      "step": 76830
    },
    {
      "epoch": 3.262299524107914,
      "grad_norm": 0.18672865629196167,
      "learning_rate": 6.79655881485652e-05,
      "loss": 1.4129,
      "step": 76840
    },
    {
      "epoch": 3.2628414652734263,
      "grad_norm": 0.3191605508327484,
      "learning_rate": 6.795042080665117e-05,
      "loss": 1.4204,
      "step": 76850
    },
    {
      "epoch": 3.2633834064389386,
      "grad_norm": 0.1636675000190735,
      "learning_rate": 6.793525186049527e-05,
      "loss": 1.422,
      "step": 76860
    },
    {
      "epoch": 3.2639253476044505,
      "grad_norm": 0.2027965933084488,
      "learning_rate": 6.792008131197651e-05,
      "loss": 1.4215,
      "step": 76870
    },
    {
      "epoch": 3.264467288769963,
      "grad_norm": 0.1797472983598709,
      "learning_rate": 6.790490916297419e-05,
      "loss": 1.4197,
      "step": 76880
    },
    {
      "epoch": 3.265009229935475,
      "grad_norm": 0.2633003294467926,
      "learning_rate": 6.788973541536772e-05,
      "loss": 1.418,
      "step": 76890
    },
    {
      "epoch": 3.2655511711009875,
      "grad_norm": 0.30197060108184814,
      "learning_rate": 6.78745600710368e-05,
      "loss": 1.4259,
      "step": 76900
    },
    {
      "epoch": 3.2659847240333972,
      "eval_loss": 2.4383397102355957,
      "eval_runtime": 21.995,
      "eval_samples_per_second": 227.325,
      "eval_steps_per_second": 1.228,
      "step": 76908
    },
    {
      "epoch": 3.2660931122665,
      "grad_norm": 0.21342779695987701,
      "learning_rate": 6.785938313186128e-05,
      "loss": 1.4201,
      "step": 76910
    },
    {
      "epoch": 3.2666350534320117,
      "grad_norm": 0.18133488297462463,
      "learning_rate": 6.784420459972122e-05,
      "loss": 1.4201,
      "step": 76920
    },
    {
      "epoch": 3.267176994597524,
      "grad_norm": 0.2161795049905777,
      "learning_rate": 6.782902447649684e-05,
      "loss": 1.4147,
      "step": 76930
    },
    {
      "epoch": 3.2677189357630363,
      "grad_norm": 0.22516366839408875,
      "learning_rate": 6.78138427640686e-05,
      "loss": 1.4197,
      "step": 76940
    },
    {
      "epoch": 3.2682608769285486,
      "grad_norm": 0.20277120172977448,
      "learning_rate": 6.779865946431716e-05,
      "loss": 1.4132,
      "step": 76950
    },
    {
      "epoch": 3.2688028180940605,
      "grad_norm": 0.2785748243331909,
      "learning_rate": 6.778347457912335e-05,
      "loss": 1.4275,
      "step": 76960
    },
    {
      "epoch": 3.269344759259573,
      "grad_norm": 0.2544669210910797,
      "learning_rate": 6.776828811036821e-05,
      "loss": 1.4225,
      "step": 76970
    },
    {
      "epoch": 3.269886700425085,
      "grad_norm": 0.16335433721542358,
      "learning_rate": 6.775310005993297e-05,
      "loss": 1.4102,
      "step": 76980
    },
    {
      "epoch": 3.2704286415905974,
      "grad_norm": 0.249395951628685,
      "learning_rate": 6.773791042969907e-05,
      "loss": 1.4151,
      "step": 76990
    },
    {
      "epoch": 3.2706996121733534,
      "eval_loss": 2.4316024780273438,
      "eval_runtime": 21.9962,
      "eval_samples_per_second": 227.312,
      "eval_steps_per_second": 1.227,
      "step": 76995
    },
    {
      "epoch": 3.2709705827561093,
      "grad_norm": 0.17765724658966064,
      "learning_rate": 6.772271922154814e-05,
      "loss": 1.4182,
      "step": 77000
    },
    {
      "epoch": 3.2715125239216216,
      "grad_norm": 0.2077697366476059,
      "learning_rate": 6.770752643736196e-05,
      "loss": 1.4264,
      "step": 77010
    },
    {
      "epoch": 3.272054465087134,
      "grad_norm": 0.18246911466121674,
      "learning_rate": 6.769233207902261e-05,
      "loss": 1.4242,
      "step": 77020
    },
    {
      "epoch": 3.2725964062526463,
      "grad_norm": 0.3925098478794098,
      "learning_rate": 6.767713614841223e-05,
      "loss": 1.4246,
      "step": 77030
    },
    {
      "epoch": 3.2731383474181586,
      "grad_norm": 0.21297501027584076,
      "learning_rate": 6.766193864741327e-05,
      "loss": 1.4159,
      "step": 77040
    },
    {
      "epoch": 3.2736802885836704,
      "grad_norm": 0.19493433833122253,
      "learning_rate": 6.764673957790834e-05,
      "loss": 1.4209,
      "step": 77050
    },
    {
      "epoch": 3.2742222297491828,
      "grad_norm": 0.26771730184555054,
      "learning_rate": 6.763153894178022e-05,
      "loss": 1.4178,
      "step": 77060
    },
    {
      "epoch": 3.274764170914695,
      "grad_norm": 0.2222745716571808,
      "learning_rate": 6.761633674091187e-05,
      "loss": 1.4288,
      "step": 77070
    },
    {
      "epoch": 3.2753061120802074,
      "grad_norm": 0.27417925000190735,
      "learning_rate": 6.760113297718653e-05,
      "loss": 1.4272,
      "step": 77080
    },
    {
      "epoch": 3.2754145003133095,
      "eval_loss": 2.4333410263061523,
      "eval_runtime": 21.988,
      "eval_samples_per_second": 227.397,
      "eval_steps_per_second": 1.228,
      "step": 77082
    },
    {
      "epoch": 3.2758480532457197,
      "grad_norm": 0.4127950966358185,
      "learning_rate": 6.758592765248752e-05,
      "loss": 1.4098,
      "step": 77090
    },
    {
      "epoch": 3.2763899944112316,
      "grad_norm": 0.2694561779499054,
      "learning_rate": 6.757072076869845e-05,
      "loss": 1.4175,
      "step": 77100
    },
    {
      "epoch": 3.276931935576744,
      "grad_norm": 0.4011683166027069,
      "learning_rate": 6.755551232770306e-05,
      "loss": 1.4237,
      "step": 77110
    },
    {
      "epoch": 3.277473876742256,
      "grad_norm": 0.20086224377155304,
      "learning_rate": 6.754030233138533e-05,
      "loss": 1.4171,
      "step": 77120
    },
    {
      "epoch": 3.2780158179077685,
      "grad_norm": 0.2316787838935852,
      "learning_rate": 6.752509078162938e-05,
      "loss": 1.4164,
      "step": 77130
    },
    {
      "epoch": 3.278557759073281,
      "grad_norm": 0.259022980928421,
      "learning_rate": 6.750987768031954e-05,
      "loss": 1.4197,
      "step": 77140
    },
    {
      "epoch": 3.2790997002387927,
      "grad_norm": 0.42053425312042236,
      "learning_rate": 6.749466302934042e-05,
      "loss": 1.4367,
      "step": 77150
    },
    {
      "epoch": 3.279641641404305,
      "grad_norm": 0.1725689321756363,
      "learning_rate": 6.747944683057666e-05,
      "loss": 1.4162,
      "step": 77160
    },
    {
      "epoch": 3.280129388453266,
      "eval_loss": 2.433736801147461,
      "eval_runtime": 21.9879,
      "eval_samples_per_second": 227.397,
      "eval_steps_per_second": 1.228,
      "step": 77169
    },
    {
      "epoch": 3.2801835825698173,
      "grad_norm": 0.183704674243927,
      "learning_rate": 6.746422908591318e-05,
      "loss": 1.413,
      "step": 77170
    },
    {
      "epoch": 3.2807255237353297,
      "grad_norm": 0.2713957130908966,
      "learning_rate": 6.744900979723515e-05,
      "loss": 1.4109,
      "step": 77180
    },
    {
      "epoch": 3.2812674649008415,
      "grad_norm": 0.18868058919906616,
      "learning_rate": 6.743378896642781e-05,
      "loss": 1.4193,
      "step": 77190
    },
    {
      "epoch": 3.281809406066354,
      "grad_norm": 0.22179383039474487,
      "learning_rate": 6.741856659537669e-05,
      "loss": 1.4177,
      "step": 77200
    },
    {
      "epoch": 3.282351347231866,
      "grad_norm": 0.18624810874462128,
      "learning_rate": 6.740334268596746e-05,
      "loss": 1.4239,
      "step": 77210
    },
    {
      "epoch": 3.2828932883973785,
      "grad_norm": 0.2454560250043869,
      "learning_rate": 6.738811724008598e-05,
      "loss": 1.425,
      "step": 77220
    },
    {
      "epoch": 3.2834352295628904,
      "grad_norm": 0.26571282744407654,
      "learning_rate": 6.737289025961835e-05,
      "loss": 1.4224,
      "step": 77230
    },
    {
      "epoch": 3.2839771707284027,
      "grad_norm": 0.19458989799022675,
      "learning_rate": 6.735766174645075e-05,
      "loss": 1.4143,
      "step": 77240
    },
    {
      "epoch": 3.284519111893915,
      "grad_norm": 0.2220762073993683,
      "learning_rate": 6.734243170246968e-05,
      "loss": 1.4226,
      "step": 77250
    },
    {
      "epoch": 3.284844276593222,
      "eval_loss": 2.4338366985321045,
      "eval_runtime": 22.2711,
      "eval_samples_per_second": 224.507,
      "eval_steps_per_second": 1.212,
      "step": 77256
    },
    {
      "epoch": 3.2850610530594273,
      "grad_norm": 0.1980799287557602,
      "learning_rate": 6.732720012956175e-05,
      "loss": 1.4256,
      "step": 77260
    },
    {
      "epoch": 3.2856029942249396,
      "grad_norm": 0.2740095555782318,
      "learning_rate": 6.731196702961381e-05,
      "loss": 1.4159,
      "step": 77270
    },
    {
      "epoch": 3.2861449353904515,
      "grad_norm": 0.16144628822803497,
      "learning_rate": 6.729673240451283e-05,
      "loss": 1.4139,
      "step": 77280
    },
    {
      "epoch": 3.286686876555964,
      "grad_norm": 0.20206600427627563,
      "learning_rate": 6.728149625614602e-05,
      "loss": 1.4197,
      "step": 77290
    },
    {
      "epoch": 3.287228817721476,
      "grad_norm": 0.2360743135213852,
      "learning_rate": 6.726625858640078e-05,
      "loss": 1.422,
      "step": 77300
    },
    {
      "epoch": 3.2877707588869884,
      "grad_norm": 0.2006218433380127,
      "learning_rate": 6.72510193971647e-05,
      "loss": 1.4153,
      "step": 77310
    },
    {
      "epoch": 3.2883127000525008,
      "grad_norm": 0.2214326560497284,
      "learning_rate": 6.72357786903255e-05,
      "loss": 1.4235,
      "step": 77320
    },
    {
      "epoch": 3.2888546412180126,
      "grad_norm": 0.183137446641922,
      "learning_rate": 6.722053646777116e-05,
      "loss": 1.419,
      "step": 77330
    },
    {
      "epoch": 3.289396582383525,
      "grad_norm": 0.18729977309703827,
      "learning_rate": 6.720529273138983e-05,
      "loss": 1.4225,
      "step": 77340
    },
    {
      "epoch": 3.2895591647331788,
      "eval_loss": 2.433459758758545,
      "eval_runtime": 21.993,
      "eval_samples_per_second": 227.345,
      "eval_steps_per_second": 1.228,
      "step": 77343
    },
    {
      "epoch": 3.2899385235490373,
      "grad_norm": 0.19643135368824005,
      "learning_rate": 6.719004748306982e-05,
      "loss": 1.4253,
      "step": 77350
    },
    {
      "epoch": 3.2904804647145496,
      "grad_norm": 0.21529223024845123,
      "learning_rate": 6.717480072469967e-05,
      "loss": 1.4184,
      "step": 77360
    },
    {
      "epoch": 3.2910224058800615,
      "grad_norm": 0.1587986797094345,
      "learning_rate": 6.715955245816804e-05,
      "loss": 1.4235,
      "step": 77370
    },
    {
      "epoch": 3.2915643470455738,
      "grad_norm": 0.23578478395938873,
      "learning_rate": 6.714430268536384e-05,
      "loss": 1.4172,
      "step": 77380
    },
    {
      "epoch": 3.292106288211086,
      "grad_norm": 0.33754247426986694,
      "learning_rate": 6.712905140817616e-05,
      "loss": 1.4207,
      "step": 77390
    },
    {
      "epoch": 3.2926482293765984,
      "grad_norm": 0.20548036694526672,
      "learning_rate": 6.711379862849426e-05,
      "loss": 1.4004,
      "step": 77400
    },
    {
      "epoch": 3.2931901705421103,
      "grad_norm": 0.1700463443994522,
      "learning_rate": 6.709854434820757e-05,
      "loss": 1.419,
      "step": 77410
    },
    {
      "epoch": 3.2937321117076226,
      "grad_norm": 0.381428062915802,
      "learning_rate": 6.708328856920574e-05,
      "loss": 1.4164,
      "step": 77420
    },
    {
      "epoch": 3.294274052873135,
      "grad_norm": 0.19036360085010529,
      "learning_rate": 6.706803129337856e-05,
      "loss": 1.4059,
      "step": 77430
    },
    {
      "epoch": 3.294274052873135,
      "eval_loss": 2.4337708950042725,
      "eval_runtime": 22.2936,
      "eval_samples_per_second": 224.279,
      "eval_steps_per_second": 1.211,
      "step": 77430
    },
    {
      "epoch": 3.2948159940386472,
      "grad_norm": 0.2967754900455475,
      "learning_rate": 6.705277252261608e-05,
      "loss": 1.4183,
      "step": 77440
    },
    {
      "epoch": 3.2953579352041595,
      "grad_norm": 0.18681344389915466,
      "learning_rate": 6.703751225880847e-05,
      "loss": 1.4228,
      "step": 77450
    },
    {
      "epoch": 3.2958998763696714,
      "grad_norm": 0.15400539338588715,
      "learning_rate": 6.70222505038461e-05,
      "loss": 1.4108,
      "step": 77460
    },
    {
      "epoch": 3.2964418175351837,
      "grad_norm": 0.16726654767990112,
      "learning_rate": 6.700698725961952e-05,
      "loss": 1.4053,
      "step": 77470
    },
    {
      "epoch": 3.296983758700696,
      "grad_norm": 0.2558479309082031,
      "learning_rate": 6.699172252801948e-05,
      "loss": 1.4146,
      "step": 77480
    },
    {
      "epoch": 3.2975256998662084,
      "grad_norm": 0.16007401049137115,
      "learning_rate": 6.697645631093694e-05,
      "loss": 1.4116,
      "step": 77490
    },
    {
      "epoch": 3.2980676410317207,
      "grad_norm": 0.23393070697784424,
      "learning_rate": 6.696118861026297e-05,
      "loss": 1.4125,
      "step": 77500
    },
    {
      "epoch": 3.2986095821972325,
      "grad_norm": 0.18870221078395844,
      "learning_rate": 6.69459194278889e-05,
      "loss": 1.4135,
      "step": 77510
    },
    {
      "epoch": 3.298988941013091,
      "eval_loss": 2.4385015964508057,
      "eval_runtime": 21.9917,
      "eval_samples_per_second": 227.358,
      "eval_steps_per_second": 1.228,
      "step": 77517
    },
    {
      "epoch": 3.299151523362745,
      "grad_norm": 0.2689937651157379,
      "learning_rate": 6.69306487657062e-05,
      "loss": 1.4227,
      "step": 77520
    },
    {
      "epoch": 3.299693464528257,
      "grad_norm": 0.28837496042251587,
      "learning_rate": 6.691537662560651e-05,
      "loss": 1.4243,
      "step": 77530
    },
    {
      "epoch": 3.3002354056937695,
      "grad_norm": 0.18641532957553864,
      "learning_rate": 6.69001030094817e-05,
      "loss": 1.4236,
      "step": 77540
    },
    {
      "epoch": 3.300777346859282,
      "grad_norm": 0.18917597830295563,
      "learning_rate": 6.68848279192238e-05,
      "loss": 1.4141,
      "step": 77550
    },
    {
      "epoch": 3.3013192880247937,
      "grad_norm": 0.201051265001297,
      "learning_rate": 6.6869551356725e-05,
      "loss": 1.422,
      "step": 77560
    },
    {
      "epoch": 3.301861229190306,
      "grad_norm": 0.18273517489433289,
      "learning_rate": 6.685427332387774e-05,
      "loss": 1.4275,
      "step": 77570
    },
    {
      "epoch": 3.3024031703558183,
      "grad_norm": 0.2597062587738037,
      "learning_rate": 6.683899382257454e-05,
      "loss": 1.426,
      "step": 77580
    },
    {
      "epoch": 3.3029451115213306,
      "grad_norm": 0.33226585388183594,
      "learning_rate": 6.682371285470819e-05,
      "loss": 1.4262,
      "step": 77590
    },
    {
      "epoch": 3.3034870526868425,
      "grad_norm": 0.22054164111614227,
      "learning_rate": 6.680843042217165e-05,
      "loss": 1.4183,
      "step": 77600
    },
    {
      "epoch": 3.3037038291530476,
      "eval_loss": 2.4387526512145996,
      "eval_runtime": 21.991,
      "eval_samples_per_second": 227.366,
      "eval_steps_per_second": 1.228,
      "step": 77604
    },
    {
      "epoch": 3.304028993852355,
      "grad_norm": 0.19982264935970306,
      "learning_rate": 6.679314652685798e-05,
      "loss": 1.4194,
      "step": 77610
    },
    {
      "epoch": 3.304570935017867,
      "grad_norm": 0.3592228293418884,
      "learning_rate": 6.677786117066054e-05,
      "loss": 1.4139,
      "step": 77620
    },
    {
      "epoch": 3.3051128761833795,
      "grad_norm": 0.27388885617256165,
      "learning_rate": 6.676257435547279e-05,
      "loss": 1.4126,
      "step": 77630
    },
    {
      "epoch": 3.3056548173488913,
      "grad_norm": 0.165022611618042,
      "learning_rate": 6.674728608318839e-05,
      "loss": 1.426,
      "step": 77640
    },
    {
      "epoch": 3.3061967585144036,
      "grad_norm": 0.28118664026260376,
      "learning_rate": 6.67319963557012e-05,
      "loss": 1.4054,
      "step": 77650
    },
    {
      "epoch": 3.306738699679916,
      "grad_norm": 0.16331221163272858,
      "learning_rate": 6.671670517490525e-05,
      "loss": 1.419,
      "step": 77660
    },
    {
      "epoch": 3.3072806408454283,
      "grad_norm": 0.16730548441410065,
      "learning_rate": 6.67014125426947e-05,
      "loss": 1.4121,
      "step": 77670
    },
    {
      "epoch": 3.3078225820109406,
      "grad_norm": 0.22583581507205963,
      "learning_rate": 6.668611846096397e-05,
      "loss": 1.4194,
      "step": 77680
    },
    {
      "epoch": 3.3083645231764525,
      "grad_norm": 0.195210799574852,
      "learning_rate": 6.667082293160766e-05,
      "loss": 1.4119,
      "step": 77690
    },
    {
      "epoch": 3.3084187172930037,
      "eval_loss": 2.4286701679229736,
      "eval_runtime": 21.9887,
      "eval_samples_per_second": 227.39,
      "eval_steps_per_second": 1.228,
      "step": 77691
    },
    {
      "epoch": 3.308906464341965,
      "grad_norm": 0.18108178675174713,
      "learning_rate": 6.665552595652043e-05,
      "loss": 1.4121,
      "step": 77700
    },
    {
      "epoch": 3.309448405507477,
      "grad_norm": 0.2247098833322525,
      "learning_rate": 6.664022753759728e-05,
      "loss": 1.4101,
      "step": 77710
    },
    {
      "epoch": 3.3099903466729894,
      "grad_norm": 0.3786323666572571,
      "learning_rate": 6.662492767673325e-05,
      "loss": 1.4119,
      "step": 77720
    },
    {
      "epoch": 3.3105322878385017,
      "grad_norm": 0.19558481872081757,
      "learning_rate": 6.660962637582366e-05,
      "loss": 1.4074,
      "step": 77730
    },
    {
      "epoch": 3.3110742290040136,
      "grad_norm": 0.2246645838022232,
      "learning_rate": 6.659432363676397e-05,
      "loss": 1.4203,
      "step": 77740
    },
    {
      "epoch": 3.311616170169526,
      "grad_norm": 0.18004010617733002,
      "learning_rate": 6.65790194614498e-05,
      "loss": 1.4246,
      "step": 77750
    },
    {
      "epoch": 3.3121581113350382,
      "grad_norm": 0.19459094107151031,
      "learning_rate": 6.656371385177697e-05,
      "loss": 1.4134,
      "step": 77760
    },
    {
      "epoch": 3.3127000525005506,
      "grad_norm": 0.16068406403064728,
      "learning_rate": 6.654840680964148e-05,
      "loss": 1.4124,
      "step": 77770
    },
    {
      "epoch": 3.3131336054329603,
      "eval_loss": 2.4155197143554688,
      "eval_runtime": 21.9967,
      "eval_samples_per_second": 227.307,
      "eval_steps_per_second": 1.227,
      "step": 77778
    },
    {
      "epoch": 3.313241993666063,
      "grad_norm": 0.19042399525642395,
      "learning_rate": 6.653309833693947e-05,
      "loss": 1.4129,
      "step": 77780
    },
    {
      "epoch": 3.3137839348315747,
      "grad_norm": 0.17300648987293243,
      "learning_rate": 6.651778843556734e-05,
      "loss": 1.4206,
      "step": 77790
    },
    {
      "epoch": 3.314325875997087,
      "grad_norm": 0.23803412914276123,
      "learning_rate": 6.650247710742156e-05,
      "loss": 1.4187,
      "step": 77800
    },
    {
      "epoch": 3.3148678171625994,
      "grad_norm": 0.26673057675361633,
      "learning_rate": 6.648716435439887e-05,
      "loss": 1.4223,
      "step": 77810
    },
    {
      "epoch": 3.3154097583281117,
      "grad_norm": 0.25232914090156555,
      "learning_rate": 6.647185017839612e-05,
      "loss": 1.4211,
      "step": 77820
    },
    {
      "epoch": 3.3159516994936236,
      "grad_norm": 0.22854040563106537,
      "learning_rate": 6.645653458131037e-05,
      "loss": 1.4249,
      "step": 77830
    },
    {
      "epoch": 3.316493640659136,
      "grad_norm": 0.1900588870048523,
      "learning_rate": 6.644121756503888e-05,
      "loss": 1.4148,
      "step": 77840
    },
    {
      "epoch": 3.317035581824648,
      "grad_norm": 0.23278219997882843,
      "learning_rate": 6.642589913147902e-05,
      "loss": 1.4233,
      "step": 77850
    },
    {
      "epoch": 3.3175775229901605,
      "grad_norm": 0.2536775767803192,
      "learning_rate": 6.641057928252837e-05,
      "loss": 1.412,
      "step": 77860
    },
    {
      "epoch": 3.3178484935729164,
      "eval_loss": 2.4168598651885986,
      "eval_runtime": 21.9945,
      "eval_samples_per_second": 227.329,
      "eval_steps_per_second": 1.228,
      "step": 77865
    },
    {
      "epoch": 3.3181194641556724,
      "grad_norm": 0.16754235327243805,
      "learning_rate": 6.63952580200847e-05,
      "loss": 1.4147,
      "step": 77870
    },
    {
      "epoch": 3.3186614053211847,
      "grad_norm": 0.1937492936849594,
      "learning_rate": 6.637993534604595e-05,
      "loss": 1.4192,
      "step": 77880
    },
    {
      "epoch": 3.319203346486697,
      "grad_norm": 0.17745104432106018,
      "learning_rate": 6.636461126231022e-05,
      "loss": 1.4204,
      "step": 77890
    },
    {
      "epoch": 3.3197452876522093,
      "grad_norm": 0.21982987225055695,
      "learning_rate": 6.634928577077577e-05,
      "loss": 1.41,
      "step": 77900
    },
    {
      "epoch": 3.3202872288177216,
      "grad_norm": 0.2929491102695465,
      "learning_rate": 6.633395887334108e-05,
      "loss": 1.4172,
      "step": 77910
    },
    {
      "epoch": 3.3208291699832335,
      "grad_norm": 0.17290639877319336,
      "learning_rate": 6.631863057190479e-05,
      "loss": 1.4262,
      "step": 77920
    },
    {
      "epoch": 3.321371111148746,
      "grad_norm": 0.18496091663837433,
      "learning_rate": 6.630330086836565e-05,
      "loss": 1.4092,
      "step": 77930
    },
    {
      "epoch": 3.321913052314258,
      "grad_norm": 0.3094109892845154,
      "learning_rate": 6.62879697646227e-05,
      "loss": 1.422,
      "step": 77940
    },
    {
      "epoch": 3.3224549934797705,
      "grad_norm": 0.21480301022529602,
      "learning_rate": 6.627263726257506e-05,
      "loss": 1.4147,
      "step": 77950
    },
    {
      "epoch": 3.3225633817128726,
      "eval_loss": 2.4209840297698975,
      "eval_runtime": 21.9901,
      "eval_samples_per_second": 227.376,
      "eval_steps_per_second": 1.228,
      "step": 77952
    },
    {
      "epoch": 3.322996934645283,
      "grad_norm": 0.19507268071174622,
      "learning_rate": 6.625730336412204e-05,
      "loss": 1.4109,
      "step": 77960
    },
    {
      "epoch": 3.3235388758107947,
      "grad_norm": 0.21125446259975433,
      "learning_rate": 6.624196807116317e-05,
      "loss": 1.4112,
      "step": 77970
    },
    {
      "epoch": 3.324080816976307,
      "grad_norm": 0.2264462113380432,
      "learning_rate": 6.62266313855981e-05,
      "loss": 1.4279,
      "step": 77980
    },
    {
      "epoch": 3.3246227581418193,
      "grad_norm": 0.2783096432685852,
      "learning_rate": 6.621129330932668e-05,
      "loss": 1.4212,
      "step": 77990
    },
    {
      "epoch": 3.3251646993073316,
      "grad_norm": 0.17773371934890747,
      "learning_rate": 6.61959538442489e-05,
      "loss": 1.4149,
      "step": 78000
    },
    {
      "epoch": 3.3257066404728435,
      "grad_norm": 0.1886042207479477,
      "learning_rate": 6.618061299226497e-05,
      "loss": 1.4196,
      "step": 78010
    },
    {
      "epoch": 3.326248581638356,
      "grad_norm": 0.2692718505859375,
      "learning_rate": 6.616527075527527e-05,
      "loss": 1.4199,
      "step": 78020
    },
    {
      "epoch": 3.326790522803868,
      "grad_norm": 0.1580006331205368,
      "learning_rate": 6.61499271351803e-05,
      "loss": 1.4211,
      "step": 78030
    },
    {
      "epoch": 3.327278269852829,
      "eval_loss": 2.4329586029052734,
      "eval_runtime": 21.9893,
      "eval_samples_per_second": 227.384,
      "eval_steps_per_second": 1.228,
      "step": 78039
    },
    {
      "epoch": 3.3273324639693804,
      "grad_norm": 0.18620441854000092,
      "learning_rate": 6.613458213388073e-05,
      "loss": 1.4193,
      "step": 78040
    },
    {
      "epoch": 3.3278744051348923,
      "grad_norm": 0.20754049718379974,
      "learning_rate": 6.61192357532775e-05,
      "loss": 1.4075,
      "step": 78050
    },
    {
      "epoch": 3.3284163463004046,
      "grad_norm": 0.19851942360401154,
      "learning_rate": 6.61038879952716e-05,
      "loss": 1.4155,
      "step": 78060
    },
    {
      "epoch": 3.328958287465917,
      "grad_norm": 0.1629553735256195,
      "learning_rate": 6.608853886176426e-05,
      "loss": 1.4237,
      "step": 78070
    },
    {
      "epoch": 3.3295002286314292,
      "grad_norm": 0.15505240857601166,
      "learning_rate": 6.607318835465689e-05,
      "loss": 1.4255,
      "step": 78080
    },
    {
      "epoch": 3.3300421697969416,
      "grad_norm": 0.3402341902256012,
      "learning_rate": 6.6057836475851e-05,
      "loss": 1.4163,
      "step": 78090
    },
    {
      "epoch": 3.3305841109624534,
      "grad_norm": 0.1934230774641037,
      "learning_rate": 6.604248322724835e-05,
      "loss": 1.4173,
      "step": 78100
    },
    {
      "epoch": 3.3311260521279658,
      "grad_norm": 0.17497047781944275,
      "learning_rate": 6.602712861075082e-05,
      "loss": 1.4166,
      "step": 78110
    },
    {
      "epoch": 3.331667993293478,
      "grad_norm": 0.19935786724090576,
      "learning_rate": 6.601177262826046e-05,
      "loss": 1.4221,
      "step": 78120
    },
    {
      "epoch": 3.3319931579927853,
      "eval_loss": 2.4348928928375244,
      "eval_runtime": 21.9935,
      "eval_samples_per_second": 227.34,
      "eval_steps_per_second": 1.228,
      "step": 78126
    },
    {
      "epoch": 3.3322099344589904,
      "grad_norm": 0.17210763692855835,
      "learning_rate": 6.599641528167952e-05,
      "loss": 1.4111,
      "step": 78130
    },
    {
      "epoch": 3.3327518756245027,
      "grad_norm": 0.175918847322464,
      "learning_rate": 6.598105657291041e-05,
      "loss": 1.4183,
      "step": 78140
    },
    {
      "epoch": 3.3332938167900146,
      "grad_norm": 0.1733727604150772,
      "learning_rate": 6.596569650385567e-05,
      "loss": 1.4179,
      "step": 78150
    },
    {
      "epoch": 3.333835757955527,
      "grad_norm": 0.16511158645153046,
      "learning_rate": 6.595033507641806e-05,
      "loss": 1.415,
      "step": 78160
    },
    {
      "epoch": 3.334377699121039,
      "grad_norm": 0.3177291750907898,
      "learning_rate": 6.593497229250048e-05,
      "loss": 1.4095,
      "step": 78170
    },
    {
      "epoch": 3.3349196402865515,
      "grad_norm": 0.172699436545372,
      "learning_rate": 6.591960815400601e-05,
      "loss": 1.4035,
      "step": 78180
    },
    {
      "epoch": 3.335461581452064,
      "grad_norm": 0.28057149052619934,
      "learning_rate": 6.590424266283791e-05,
      "loss": 1.4162,
      "step": 78190
    },
    {
      "epoch": 3.3360035226175757,
      "grad_norm": 0.23236146569252014,
      "learning_rate": 6.588887582089955e-05,
      "loss": 1.417,
      "step": 78200
    },
    {
      "epoch": 3.336545463783088,
      "grad_norm": 0.2135874629020691,
      "learning_rate": 6.587350763009452e-05,
      "loss": 1.4192,
      "step": 78210
    },
    {
      "epoch": 3.336708046132742,
      "eval_loss": 2.4323973655700684,
      "eval_runtime": 21.9907,
      "eval_samples_per_second": 227.369,
      "eval_steps_per_second": 1.228,
      "step": 78213
    },
    {
      "epoch": 3.3370874049486003,
      "grad_norm": 0.4779936671257019,
      "learning_rate": 6.585813809232659e-05,
      "loss": 1.415,
      "step": 78220
    },
    {
      "epoch": 3.3376293461141127,
      "grad_norm": 0.22560039162635803,
      "learning_rate": 6.584276720949964e-05,
      "loss": 1.4238,
      "step": 78230
    },
    {
      "epoch": 3.3381712872796245,
      "grad_norm": 0.20752812922000885,
      "learning_rate": 6.582739498351778e-05,
      "loss": 1.4179,
      "step": 78240
    },
    {
      "epoch": 3.338713228445137,
      "grad_norm": 0.16067805886268616,
      "learning_rate": 6.581202141628522e-05,
      "loss": 1.4103,
      "step": 78250
    },
    {
      "epoch": 3.339255169610649,
      "grad_norm": 0.18570862710475922,
      "learning_rate": 6.579664650970638e-05,
      "loss": 1.4131,
      "step": 78260
    },
    {
      "epoch": 3.3397971107761615,
      "grad_norm": 0.2738288938999176,
      "learning_rate": 6.578127026568587e-05,
      "loss": 1.4215,
      "step": 78270
    },
    {
      "epoch": 3.3403390519416734,
      "grad_norm": 0.26719287037849426,
      "learning_rate": 6.57658926861284e-05,
      "loss": 1.4188,
      "step": 78280
    },
    {
      "epoch": 3.3408809931071857,
      "grad_norm": 0.23562686145305634,
      "learning_rate": 6.575051377293888e-05,
      "loss": 1.4081,
      "step": 78290
    },
    {
      "epoch": 3.341422934272698,
      "grad_norm": 0.17355681955814362,
      "learning_rate": 6.573513352802239e-05,
      "loss": 1.4196,
      "step": 78300
    },
    {
      "epoch": 3.341422934272698,
      "eval_loss": 2.430081605911255,
      "eval_runtime": 21.985,
      "eval_samples_per_second": 227.428,
      "eval_steps_per_second": 1.228,
      "step": 78300
    },
    {
      "epoch": 3.3419648754382103,
      "grad_norm": 0.19452211260795593,
      "learning_rate": 6.571975195328416e-05,
      "loss": 1.4239,
      "step": 78310
    },
    {
      "epoch": 3.3425068166037226,
      "grad_norm": 0.20565609633922577,
      "learning_rate": 6.570436905062961e-05,
      "loss": 1.4145,
      "step": 78320
    },
    {
      "epoch": 3.3430487577692345,
      "grad_norm": 0.163151815533638,
      "learning_rate": 6.568898482196428e-05,
      "loss": 1.4153,
      "step": 78330
    },
    {
      "epoch": 3.343590698934747,
      "grad_norm": 0.17055930197238922,
      "learning_rate": 6.567359926919394e-05,
      "loss": 1.4203,
      "step": 78340
    },
    {
      "epoch": 3.344132640100259,
      "grad_norm": 0.17260503768920898,
      "learning_rate": 6.565821239422444e-05,
      "loss": 1.4126,
      "step": 78350
    },
    {
      "epoch": 3.3446745812657714,
      "grad_norm": 0.19056035578250885,
      "learning_rate": 6.564282419896187e-05,
      "loss": 1.4296,
      "step": 78360
    },
    {
      "epoch": 3.3452165224312838,
      "grad_norm": 0.1595456898212433,
      "learning_rate": 6.562743468531244e-05,
      "loss": 1.4219,
      "step": 78370
    },
    {
      "epoch": 3.3457584635967956,
      "grad_norm": 0.16452792286872864,
      "learning_rate": 6.561204385518257e-05,
      "loss": 1.4136,
      "step": 78380
    },
    {
      "epoch": 3.346137822412654,
      "eval_loss": 2.4075536727905273,
      "eval_runtime": 21.9892,
      "eval_samples_per_second": 227.384,
      "eval_steps_per_second": 1.228,
      "step": 78387
    },
    {
      "epoch": 3.346300404762308,
      "grad_norm": 0.3267938196659088,
      "learning_rate": 6.559665171047876e-05,
      "loss": 1.4088,
      "step": 78390
    },
    {
      "epoch": 3.3468423459278203,
      "grad_norm": 0.24453267455101013,
      "learning_rate": 6.558125825310772e-05,
      "loss": 1.4215,
      "step": 78400
    },
    {
      "epoch": 3.3473842870933326,
      "grad_norm": 0.20705412328243256,
      "learning_rate": 6.556586348497637e-05,
      "loss": 1.4251,
      "step": 78410
    },
    {
      "epoch": 3.347926228258845,
      "grad_norm": 0.2045924812555313,
      "learning_rate": 6.555046740799173e-05,
      "loss": 1.4172,
      "step": 78420
    },
    {
      "epoch": 3.3484681694243568,
      "grad_norm": 0.1746375858783722,
      "learning_rate": 6.553507002406099e-05,
      "loss": 1.4219,
      "step": 78430
    },
    {
      "epoch": 3.349010110589869,
      "grad_norm": 0.23190824687480927,
      "learning_rate": 6.551967133509149e-05,
      "loss": 1.4169,
      "step": 78440
    },
    {
      "epoch": 3.3495520517553814,
      "grad_norm": 0.21592742204666138,
      "learning_rate": 6.550427134299079e-05,
      "loss": 1.415,
      "step": 78450
    },
    {
      "epoch": 3.3500939929208937,
      "grad_norm": 0.18433372676372528,
      "learning_rate": 6.548887004966658e-05,
      "loss": 1.415,
      "step": 78460
    },
    {
      "epoch": 3.3506359340864056,
      "grad_norm": 0.1842111349105835,
      "learning_rate": 6.547346745702666e-05,
      "loss": 1.4273,
      "step": 78470
    },
    {
      "epoch": 3.3508527105526107,
      "eval_loss": 2.4247753620147705,
      "eval_runtime": 21.9917,
      "eval_samples_per_second": 227.358,
      "eval_steps_per_second": 1.228,
      "step": 78474
    },
    {
      "epoch": 3.351177875251918,
      "grad_norm": 0.18961256742477417,
      "learning_rate": 6.545806356697908e-05,
      "loss": 1.4107,
      "step": 78480
    },
    {
      "epoch": 3.35171981641743,
      "grad_norm": 0.15932081639766693,
      "learning_rate": 6.544265838143197e-05,
      "loss": 1.4176,
      "step": 78490
    },
    {
      "epoch": 3.3522617575829425,
      "grad_norm": 0.1957116425037384,
      "learning_rate": 6.54272519022937e-05,
      "loss": 1.4084,
      "step": 78500
    },
    {
      "epoch": 3.3528036987484544,
      "grad_norm": 0.2836982011795044,
      "learning_rate": 6.541184413147273e-05,
      "loss": 1.4135,
      "step": 78510
    },
    {
      "epoch": 3.3533456399139667,
      "grad_norm": 0.19204333424568176,
      "learning_rate": 6.53964350708777e-05,
      "loss": 1.4181,
      "step": 78520
    },
    {
      "epoch": 3.353887581079479,
      "grad_norm": 0.173109233379364,
      "learning_rate": 6.538102472241745e-05,
      "loss": 1.4161,
      "step": 78530
    },
    {
      "epoch": 3.3544295222449914,
      "grad_norm": 0.24930110573768616,
      "learning_rate": 6.536561308800091e-05,
      "loss": 1.4201,
      "step": 78540
    },
    {
      "epoch": 3.3549714634105037,
      "grad_norm": 0.1809731125831604,
      "learning_rate": 6.535020016953723e-05,
      "loss": 1.4107,
      "step": 78550
    },
    {
      "epoch": 3.3555134045760155,
      "grad_norm": 0.18697808682918549,
      "learning_rate": 6.53347859689357e-05,
      "loss": 1.415,
      "step": 78560
    },
    {
      "epoch": 3.355567598692567,
      "eval_loss": 2.433493137359619,
      "eval_runtime": 21.9938,
      "eval_samples_per_second": 227.337,
      "eval_steps_per_second": 1.228,
      "step": 78561
    },
    {
      "epoch": 3.356055345741528,
      "grad_norm": 0.24766592681407928,
      "learning_rate": 6.531937048810573e-05,
      "loss": 1.4146,
      "step": 78570
    },
    {
      "epoch": 3.35659728690704,
      "grad_norm": 0.16486966609954834,
      "learning_rate": 6.530395372895697e-05,
      "loss": 1.4123,
      "step": 78580
    },
    {
      "epoch": 3.3571392280725525,
      "grad_norm": 0.20207463204860687,
      "learning_rate": 6.528853569339913e-05,
      "loss": 1.4197,
      "step": 78590
    },
    {
      "epoch": 3.357681169238065,
      "grad_norm": 0.17322023212909698,
      "learning_rate": 6.527311638334218e-05,
      "loss": 1.4166,
      "step": 78600
    },
    {
      "epoch": 3.3582231104035767,
      "grad_norm": 0.18780970573425293,
      "learning_rate": 6.525769580069617e-05,
      "loss": 1.4212,
      "step": 78610
    },
    {
      "epoch": 3.358765051569089,
      "grad_norm": 0.2877364456653595,
      "learning_rate": 6.524227394737135e-05,
      "loss": 1.4195,
      "step": 78620
    },
    {
      "epoch": 3.3593069927346013,
      "grad_norm": 0.2212965041399002,
      "learning_rate": 6.522685082527807e-05,
      "loss": 1.4172,
      "step": 78630
    },
    {
      "epoch": 3.3598489339001136,
      "grad_norm": 0.2948903441429138,
      "learning_rate": 6.521142643632692e-05,
      "loss": 1.4171,
      "step": 78640
    },
    {
      "epoch": 3.3602824868325234,
      "eval_loss": 2.4318501949310303,
      "eval_runtime": 24.8893,
      "eval_samples_per_second": 200.89,
      "eval_steps_per_second": 1.085,
      "step": 78648
    },
    {
      "epoch": 3.3603908750656255,
      "grad_norm": 0.1802278459072113,
      "learning_rate": 6.51960007824286e-05,
      "loss": 1.4057,
      "step": 78650
    },
    {
      "epoch": 3.360932816231138,
      "grad_norm": 0.2904506027698517,
      "learning_rate": 6.518057386549398e-05,
      "loss": 1.4219,
      "step": 78660
    },
    {
      "epoch": 3.36147475739665,
      "grad_norm": 0.3286694586277008,
      "learning_rate": 6.516514568743407e-05,
      "loss": 1.4142,
      "step": 78670
    },
    {
      "epoch": 3.3620166985621625,
      "grad_norm": 0.1851622760295868,
      "learning_rate": 6.514971625016004e-05,
      "loss": 1.4056,
      "step": 78680
    },
    {
      "epoch": 3.3625586397276743,
      "grad_norm": 0.1889377385377884,
      "learning_rate": 6.513428555558321e-05,
      "loss": 1.4122,
      "step": 78690
    },
    {
      "epoch": 3.3631005808931866,
      "grad_norm": 0.15249784290790558,
      "learning_rate": 6.51188536056151e-05,
      "loss": 1.4211,
      "step": 78700
    },
    {
      "epoch": 3.363642522058699,
      "grad_norm": 0.3704091012477875,
      "learning_rate": 6.510342040216733e-05,
      "loss": 1.4221,
      "step": 78710
    },
    {
      "epoch": 3.3641844632242113,
      "grad_norm": 0.2278876155614853,
      "learning_rate": 6.508798594715172e-05,
      "loss": 1.4273,
      "step": 78720
    },
    {
      "epoch": 3.3647264043897236,
      "grad_norm": 0.21338103711605072,
      "learning_rate": 6.507255024248019e-05,
      "loss": 1.4185,
      "step": 78730
    },
    {
      "epoch": 3.3649973749724795,
      "eval_loss": 2.418487548828125,
      "eval_runtime": 21.9902,
      "eval_samples_per_second": 227.374,
      "eval_steps_per_second": 1.228,
      "step": 78735
    },
    {
      "epoch": 3.3652683455552355,
      "grad_norm": 0.3524332642555237,
      "learning_rate": 6.505711329006488e-05,
      "loss": 1.4218,
      "step": 78740
    },
    {
      "epoch": 3.365810286720748,
      "grad_norm": 0.1828557401895523,
      "learning_rate": 6.504167509181804e-05,
      "loss": 1.4172,
      "step": 78750
    },
    {
      "epoch": 3.36635222788626,
      "grad_norm": 0.21107490360736847,
      "learning_rate": 6.502623564965206e-05,
      "loss": 1.4145,
      "step": 78760
    },
    {
      "epoch": 3.3668941690517724,
      "grad_norm": 0.26644033193588257,
      "learning_rate": 6.501079496547957e-05,
      "loss": 1.4113,
      "step": 78770
    },
    {
      "epoch": 3.3674361102172847,
      "grad_norm": 0.15477254986763,
      "learning_rate": 6.499535304121324e-05,
      "loss": 1.4176,
      "step": 78780
    },
    {
      "epoch": 3.3679780513827966,
      "grad_norm": 0.1923697292804718,
      "learning_rate": 6.497990987876598e-05,
      "loss": 1.4192,
      "step": 78790
    },
    {
      "epoch": 3.368519992548309,
      "grad_norm": 0.18154162168502808,
      "learning_rate": 6.496446548005082e-05,
      "loss": 1.4184,
      "step": 78800
    },
    {
      "epoch": 3.3690619337138212,
      "grad_norm": 0.24058213829994202,
      "learning_rate": 6.494901984698093e-05,
      "loss": 1.4109,
      "step": 78810
    },
    {
      "epoch": 3.3696038748793335,
      "grad_norm": 0.1790485382080078,
      "learning_rate": 6.493357298146965e-05,
      "loss": 1.4162,
      "step": 78820
    },
    {
      "epoch": 3.3697122631124357,
      "eval_loss": 2.420318126678467,
      "eval_runtime": 21.9925,
      "eval_samples_per_second": 227.35,
      "eval_steps_per_second": 1.228,
      "step": 78822
    },
    {
      "epoch": 3.370145816044846,
      "grad_norm": 0.1824759691953659,
      "learning_rate": 6.491812488543049e-05,
      "loss": 1.4181,
      "step": 78830
    },
    {
      "epoch": 3.3706877572103577,
      "grad_norm": 0.27409785985946655,
      "learning_rate": 6.490267556077706e-05,
      "loss": 1.4215,
      "step": 78840
    },
    {
      "epoch": 3.37122969837587,
      "grad_norm": 0.17618605494499207,
      "learning_rate": 6.488722500942321e-05,
      "loss": 1.4215,
      "step": 78850
    },
    {
      "epoch": 3.3717716395413824,
      "grad_norm": 0.21089158952236176,
      "learning_rate": 6.487177323328282e-05,
      "loss": 1.4105,
      "step": 78860
    },
    {
      "epoch": 3.3723135807068947,
      "grad_norm": 0.1864428073167801,
      "learning_rate": 6.485632023427003e-05,
      "loss": 1.4122,
      "step": 78870
    },
    {
      "epoch": 3.3728555218724066,
      "grad_norm": 0.21700777113437653,
      "learning_rate": 6.484086601429907e-05,
      "loss": 1.4169,
      "step": 78880
    },
    {
      "epoch": 3.373397463037919,
      "grad_norm": 0.1815451681613922,
      "learning_rate": 6.482541057528437e-05,
      "loss": 1.428,
      "step": 78890
    },
    {
      "epoch": 3.373939404203431,
      "grad_norm": 0.23216386139392853,
      "learning_rate": 6.480995391914046e-05,
      "loss": 1.4166,
      "step": 78900
    },
    {
      "epoch": 3.3744271512523922,
      "eval_loss": 2.411614418029785,
      "eval_runtime": 21.9934,
      "eval_samples_per_second": 227.341,
      "eval_steps_per_second": 1.228,
      "step": 78909
    },
    {
      "epoch": 3.3744813453689435,
      "grad_norm": 0.1651027947664261,
      "learning_rate": 6.479449604778206e-05,
      "loss": 1.4183,
      "step": 78910
    },
    {
      "epoch": 3.3750232865344554,
      "grad_norm": 0.20378755033016205,
      "learning_rate": 6.477903696312398e-05,
      "loss": 1.4167,
      "step": 78920
    },
    {
      "epoch": 3.3755652276999677,
      "grad_norm": 0.19732166826725006,
      "learning_rate": 6.476357666708129e-05,
      "loss": 1.4083,
      "step": 78930
    },
    {
      "epoch": 3.37610716886548,
      "grad_norm": 0.23895850777626038,
      "learning_rate": 6.47481151615691e-05,
      "loss": 1.4173,
      "step": 78940
    },
    {
      "epoch": 3.3766491100309923,
      "grad_norm": 0.15536242723464966,
      "learning_rate": 6.473265244850273e-05,
      "loss": 1.4142,
      "step": 78950
    },
    {
      "epoch": 3.3771910511965046,
      "grad_norm": 0.23873640596866608,
      "learning_rate": 6.471718852979762e-05,
      "loss": 1.4163,
      "step": 78960
    },
    {
      "epoch": 3.3777329923620165,
      "grad_norm": 0.3809368908405304,
      "learning_rate": 6.47017234073694e-05,
      "loss": 1.404,
      "step": 78970
    },
    {
      "epoch": 3.378274933527529,
      "grad_norm": 0.19484516978263855,
      "learning_rate": 6.468625708313378e-05,
      "loss": 1.4137,
      "step": 78980
    },
    {
      "epoch": 3.378816874693041,
      "grad_norm": 0.23201866447925568,
      "learning_rate": 6.46707895590067e-05,
      "loss": 1.4246,
      "step": 78990
    },
    {
      "epoch": 3.3791420393923484,
      "eval_loss": 2.4239916801452637,
      "eval_runtime": 21.9918,
      "eval_samples_per_second": 227.358,
      "eval_steps_per_second": 1.228,
      "step": 78996
    },
    {
      "epoch": 3.3793588158585535,
      "grad_norm": 0.17255160212516785,
      "learning_rate": 6.46553208369042e-05,
      "loss": 1.4182,
      "step": 79000
    },
    {
      "epoch": 3.379900757024066,
      "grad_norm": 0.148577019572258,
      "learning_rate": 6.463985091874248e-05,
      "loss": 1.4222,
      "step": 79010
    },
    {
      "epoch": 3.3804426981895777,
      "grad_norm": 0.17819197475910187,
      "learning_rate": 6.462437980643786e-05,
      "loss": 1.4207,
      "step": 79020
    },
    {
      "epoch": 3.38098463935509,
      "grad_norm": 0.18190248310565948,
      "learning_rate": 6.460890750190686e-05,
      "loss": 1.4113,
      "step": 79030
    },
    {
      "epoch": 3.3815265805206023,
      "grad_norm": 0.17424237728118896,
      "learning_rate": 6.459343400706612e-05,
      "loss": 1.4333,
      "step": 79040
    },
    {
      "epoch": 3.3820685216861146,
      "grad_norm": 0.16150598227977753,
      "learning_rate": 6.457795932383242e-05,
      "loss": 1.4171,
      "step": 79050
    },
    {
      "epoch": 3.3826104628516265,
      "grad_norm": 0.1738124042749405,
      "learning_rate": 6.456248345412272e-05,
      "loss": 1.4096,
      "step": 79060
    },
    {
      "epoch": 3.383152404017139,
      "grad_norm": 0.21694537997245789,
      "learning_rate": 6.454700639985408e-05,
      "loss": 1.4149,
      "step": 79070
    },
    {
      "epoch": 3.383694345182651,
      "grad_norm": 0.170954167842865,
      "learning_rate": 6.453152816294373e-05,
      "loss": 1.4234,
      "step": 79080
    },
    {
      "epoch": 3.383856927532305,
      "eval_loss": 2.4151031970977783,
      "eval_runtime": 21.9982,
      "eval_samples_per_second": 227.291,
      "eval_steps_per_second": 1.227,
      "step": 79083
    },
    {
      "epoch": 3.3842362863481634,
      "grad_norm": 0.2635592818260193,
      "learning_rate": 6.451604874530905e-05,
      "loss": 1.4245,
      "step": 79090
    },
    {
      "epoch": 3.3847782275136753,
      "grad_norm": 0.2918653190135956,
      "learning_rate": 6.450056814886756e-05,
      "loss": 1.4131,
      "step": 79100
    },
    {
      "epoch": 3.3853201686791876,
      "grad_norm": 0.3705879747867584,
      "learning_rate": 6.448508637553695e-05,
      "loss": 1.427,
      "step": 79110
    },
    {
      "epoch": 3.3858621098447,
      "grad_norm": 0.18804314732551575,
      "learning_rate": 6.446960342723503e-05,
      "loss": 1.4223,
      "step": 79120
    },
    {
      "epoch": 3.3864040510102122,
      "grad_norm": 0.15354295074939728,
      "learning_rate": 6.445411930587971e-05,
      "loss": 1.4193,
      "step": 79130
    },
    {
      "epoch": 3.3869459921757246,
      "grad_norm": 0.2109123021364212,
      "learning_rate": 6.443863401338917e-05,
      "loss": 1.4122,
      "step": 79140
    },
    {
      "epoch": 3.3874879333412364,
      "grad_norm": 0.2178303450345993,
      "learning_rate": 6.442314755168162e-05,
      "loss": 1.4152,
      "step": 79150
    },
    {
      "epoch": 3.3880298745067488,
      "grad_norm": 0.24825222790241241,
      "learning_rate": 6.440765992267546e-05,
      "loss": 1.4112,
      "step": 79160
    },
    {
      "epoch": 3.388571815672261,
      "grad_norm": 0.18957504630088806,
      "learning_rate": 6.439217112828924e-05,
      "loss": 1.4217,
      "step": 79170
    },
    {
      "epoch": 3.388571815672261,
      "eval_loss": 2.423619270324707,
      "eval_runtime": 21.99,
      "eval_samples_per_second": 227.376,
      "eval_steps_per_second": 1.228,
      "step": 79170
    },
    {
      "epoch": 3.3891137568377734,
      "grad_norm": 0.16275407373905182,
      "learning_rate": 6.437668117044162e-05,
      "loss": 1.4085,
      "step": 79180
    },
    {
      "epoch": 3.3896556980032857,
      "grad_norm": 0.26474809646606445,
      "learning_rate": 6.436119005105145e-05,
      "loss": 1.4169,
      "step": 79190
    },
    {
      "epoch": 3.3901976391687976,
      "grad_norm": 0.3433353900909424,
      "learning_rate": 6.434569777203772e-05,
      "loss": 1.4049,
      "step": 79200
    },
    {
      "epoch": 3.39073958033431,
      "grad_norm": 0.31058382987976074,
      "learning_rate": 6.433020433531951e-05,
      "loss": 1.4222,
      "step": 79210
    },
    {
      "epoch": 3.391281521499822,
      "grad_norm": 0.1867647022008896,
      "learning_rate": 6.43147097428161e-05,
      "loss": 1.4368,
      "step": 79220
    },
    {
      "epoch": 3.3918234626653345,
      "grad_norm": 0.38746219873428345,
      "learning_rate": 6.429921399644687e-05,
      "loss": 1.4072,
      "step": 79230
    },
    {
      "epoch": 3.392365403830847,
      "grad_norm": 0.21169978380203247,
      "learning_rate": 6.42837170981314e-05,
      "loss": 1.4154,
      "step": 79240
    },
    {
      "epoch": 3.3929073449963587,
      "grad_norm": 0.17615120112895966,
      "learning_rate": 6.426821904978936e-05,
      "loss": 1.4117,
      "step": 79250
    },
    {
      "epoch": 3.393286703812217,
      "eval_loss": 2.415933609008789,
      "eval_runtime": 21.9936,
      "eval_samples_per_second": 227.339,
      "eval_steps_per_second": 1.228,
      "step": 79257
    },
    {
      "epoch": 3.393449286161871,
      "grad_norm": 0.2107549011707306,
      "learning_rate": 6.425271985334059e-05,
      "loss": 1.4119,
      "step": 79260
    },
    {
      "epoch": 3.3939912273273833,
      "grad_norm": 0.30386942625045776,
      "learning_rate": 6.423721951070502e-05,
      "loss": 1.408,
      "step": 79270
    },
    {
      "epoch": 3.3945331684928957,
      "grad_norm": 0.17193304002285004,
      "learning_rate": 6.422171802380283e-05,
      "loss": 1.4055,
      "step": 79280
    },
    {
      "epoch": 3.3950751096584075,
      "grad_norm": 0.30357927083969116,
      "learning_rate": 6.420621539455426e-05,
      "loss": 1.411,
      "step": 79290
    },
    {
      "epoch": 3.39561705082392,
      "grad_norm": 0.20240160822868347,
      "learning_rate": 6.419071162487969e-05,
      "loss": 1.4088,
      "step": 79300
    },
    {
      "epoch": 3.396158991989432,
      "grad_norm": 0.18283316493034363,
      "learning_rate": 6.417520671669964e-05,
      "loss": 1.4094,
      "step": 79310
    },
    {
      "epoch": 3.3967009331549445,
      "grad_norm": 0.16190268099308014,
      "learning_rate": 6.415970067193483e-05,
      "loss": 1.4148,
      "step": 79320
    },
    {
      "epoch": 3.3972428743204564,
      "grad_norm": 0.22923493385314941,
      "learning_rate": 6.414419349250608e-05,
      "loss": 1.409,
      "step": 79330
    },
    {
      "epoch": 3.3977848154859687,
      "grad_norm": 0.3721964359283447,
      "learning_rate": 6.412868518033432e-05,
      "loss": 1.4198,
      "step": 79340
    },
    {
      "epoch": 3.3980015919521738,
      "eval_loss": 2.421030044555664,
      "eval_runtime": 21.9926,
      "eval_samples_per_second": 227.349,
      "eval_steps_per_second": 1.228,
      "step": 79344
    },
    {
      "epoch": 3.398326756651481,
      "grad_norm": 0.392916738986969,
      "learning_rate": 6.41131757373407e-05,
      "loss": 1.4077,
      "step": 79350
    },
    {
      "epoch": 3.3988686978169933,
      "grad_norm": 0.5418851375579834,
      "learning_rate": 6.409766516544642e-05,
      "loss": 1.4154,
      "step": 79360
    },
    {
      "epoch": 3.3994106389825056,
      "grad_norm": 0.20894989371299744,
      "learning_rate": 6.408215346657287e-05,
      "loss": 1.4091,
      "step": 79370
    },
    {
      "epoch": 3.3999525801480175,
      "grad_norm": 0.2258988916873932,
      "learning_rate": 6.406664064264159e-05,
      "loss": 1.4104,
      "step": 79380
    },
    {
      "epoch": 3.40049452131353,
      "grad_norm": 0.1968545764684677,
      "learning_rate": 6.405112669557424e-05,
      "loss": 1.4136,
      "step": 79390
    },
    {
      "epoch": 3.401036462479042,
      "grad_norm": 0.2986617684364319,
      "learning_rate": 6.403561162729257e-05,
      "loss": 1.4032,
      "step": 79400
    },
    {
      "epoch": 3.4015784036445544,
      "grad_norm": 0.16921205818653107,
      "learning_rate": 6.402009543971861e-05,
      "loss": 1.4121,
      "step": 79410
    },
    {
      "epoch": 3.4021203448100668,
      "grad_norm": 0.21860603988170624,
      "learning_rate": 6.400457813477435e-05,
      "loss": 1.418,
      "step": 79420
    },
    {
      "epoch": 3.4026622859755786,
      "grad_norm": 0.15775620937347412,
      "learning_rate": 6.398905971438207e-05,
      "loss": 1.4144,
      "step": 79430
    },
    {
      "epoch": 3.40271648009213,
      "eval_loss": 2.416477680206299,
      "eval_runtime": 21.9934,
      "eval_samples_per_second": 227.341,
      "eval_steps_per_second": 1.228,
      "step": 79431
    },
    {
      "epoch": 3.403204227141091,
      "grad_norm": 0.16970393061637878,
      "learning_rate": 6.39735401804641e-05,
      "loss": 1.4145,
      "step": 79440
    },
    {
      "epoch": 3.4037461683066033,
      "grad_norm": 0.21157051622867584,
      "learning_rate": 6.395801953494292e-05,
      "loss": 1.4297,
      "step": 79450
    },
    {
      "epoch": 3.4042881094721156,
      "grad_norm": 0.24515606462955475,
      "learning_rate": 6.394249777974118e-05,
      "loss": 1.4207,
      "step": 79460
    },
    {
      "epoch": 3.404830050637628,
      "grad_norm": 0.20401237905025482,
      "learning_rate": 6.392697491678163e-05,
      "loss": 1.4104,
      "step": 79470
    },
    {
      "epoch": 3.4053719918031398,
      "grad_norm": 0.2022889405488968,
      "learning_rate": 6.391145094798718e-05,
      "loss": 1.4191,
      "step": 79480
    },
    {
      "epoch": 3.405913932968652,
      "grad_norm": 0.2641688287258148,
      "learning_rate": 6.389592587528089e-05,
      "loss": 1.4175,
      "step": 79490
    },
    {
      "epoch": 3.4064558741341644,
      "grad_norm": 0.30329614877700806,
      "learning_rate": 6.388039970058591e-05,
      "loss": 1.4097,
      "step": 79500
    },
    {
      "epoch": 3.4069978152996767,
      "grad_norm": 0.19015826284885406,
      "learning_rate": 6.386487242582559e-05,
      "loss": 1.4183,
      "step": 79510
    },
    {
      "epoch": 3.4074313682320865,
      "eval_loss": 2.4201693534851074,
      "eval_runtime": 22.0994,
      "eval_samples_per_second": 226.25,
      "eval_steps_per_second": 1.222,
      "step": 79518
    },
    {
      "epoch": 3.4075397564651886,
      "grad_norm": 0.2884151041507721,
      "learning_rate": 6.384934405292335e-05,
      "loss": 1.4081,
      "step": 79520
    },
    {
      "epoch": 3.408081697630701,
      "grad_norm": 0.2025626003742218,
      "learning_rate": 6.383381458380279e-05,
      "loss": 1.408,
      "step": 79530
    },
    {
      "epoch": 3.408623638796213,
      "grad_norm": 0.20512153208255768,
      "learning_rate": 6.381828402038763e-05,
      "loss": 1.414,
      "step": 79540
    },
    {
      "epoch": 3.4091655799617255,
      "grad_norm": 0.23173965513706207,
      "learning_rate": 6.380275236460174e-05,
      "loss": 1.4234,
      "step": 79550
    },
    {
      "epoch": 3.4097075211272374,
      "grad_norm": 0.21376845240592957,
      "learning_rate": 6.378721961836908e-05,
      "loss": 1.4114,
      "step": 79560
    },
    {
      "epoch": 3.4102494622927497,
      "grad_norm": 0.25625619292259216,
      "learning_rate": 6.377168578361383e-05,
      "loss": 1.4096,
      "step": 79570
    },
    {
      "epoch": 3.410791403458262,
      "grad_norm": 0.1973353773355484,
      "learning_rate": 6.375615086226022e-05,
      "loss": 1.4196,
      "step": 79580
    },
    {
      "epoch": 3.4113333446237744,
      "grad_norm": 0.21326379477977753,
      "learning_rate": 6.374061485623266e-05,
      "loss": 1.4115,
      "step": 79590
    },
    {
      "epoch": 3.4118752857892867,
      "grad_norm": 0.16970664262771606,
      "learning_rate": 6.372507776745567e-05,
      "loss": 1.4198,
      "step": 79600
    },
    {
      "epoch": 3.4121462563720426,
      "eval_loss": 2.4215962886810303,
      "eval_runtime": 21.994,
      "eval_samples_per_second": 227.335,
      "eval_steps_per_second": 1.228,
      "step": 79605
    },
    {
      "epoch": 3.4124172269547985,
      "grad_norm": 0.15813182294368744,
      "learning_rate": 6.370953959785393e-05,
      "loss": 1.4056,
      "step": 79610
    },
    {
      "epoch": 3.412959168120311,
      "grad_norm": 0.3504032492637634,
      "learning_rate": 6.369400034935224e-05,
      "loss": 1.4224,
      "step": 79620
    },
    {
      "epoch": 3.413501109285823,
      "grad_norm": 0.4499680995941162,
      "learning_rate": 6.367846002387552e-05,
      "loss": 1.4056,
      "step": 79630
    },
    {
      "epoch": 3.4140430504513355,
      "grad_norm": 0.2306382656097412,
      "learning_rate": 6.366291862334887e-05,
      "loss": 1.416,
      "step": 79640
    },
    {
      "epoch": 3.414584991616848,
      "grad_norm": 0.1756628006696701,
      "learning_rate": 6.364737614969747e-05,
      "loss": 1.4121,
      "step": 79650
    },
    {
      "epoch": 3.4151269327823597,
      "grad_norm": 0.20633266866207123,
      "learning_rate": 6.363183260484665e-05,
      "loss": 1.4095,
      "step": 79660
    },
    {
      "epoch": 3.415668873947872,
      "grad_norm": 0.17313161492347717,
      "learning_rate": 6.361628799072187e-05,
      "loss": 1.4148,
      "step": 79670
    },
    {
      "epoch": 3.4162108151133843,
      "grad_norm": 0.16287872195243835,
      "learning_rate": 6.360074230924877e-05,
      "loss": 1.399,
      "step": 79680
    },
    {
      "epoch": 3.4167527562788966,
      "grad_norm": 0.18536217510700226,
      "learning_rate": 6.358519556235302e-05,
      "loss": 1.4155,
      "step": 79690
    },
    {
      "epoch": 3.4168611445119987,
      "eval_loss": 2.429548501968384,
      "eval_runtime": 21.9964,
      "eval_samples_per_second": 227.31,
      "eval_steps_per_second": 1.227,
      "step": 79692
    },
    {
      "epoch": 3.4172946974444085,
      "grad_norm": 0.3025163412094116,
      "learning_rate": 6.356964775196055e-05,
      "loss": 1.4113,
      "step": 79700
    },
    {
      "epoch": 3.417836638609921,
      "grad_norm": 0.30600860714912415,
      "learning_rate": 6.35540988799973e-05,
      "loss": 1.4125,
      "step": 79710
    },
    {
      "epoch": 3.418378579775433,
      "grad_norm": 0.2034313678741455,
      "learning_rate": 6.353854894838942e-05,
      "loss": 1.4084,
      "step": 79720
    },
    {
      "epoch": 3.4189205209409455,
      "grad_norm": 0.16151131689548492,
      "learning_rate": 6.352299795906317e-05,
      "loss": 1.4079,
      "step": 79730
    },
    {
      "epoch": 3.4194624621064573,
      "grad_norm": 0.3311507999897003,
      "learning_rate": 6.350744591394494e-05,
      "loss": 1.4156,
      "step": 79740
    },
    {
      "epoch": 3.4200044032719696,
      "grad_norm": 0.21057330071926117,
      "learning_rate": 6.349189281496124e-05,
      "loss": 1.4149,
      "step": 79750
    },
    {
      "epoch": 3.420546344437482,
      "grad_norm": 0.2549402117729187,
      "learning_rate": 6.347633866403873e-05,
      "loss": 1.4086,
      "step": 79760
    },
    {
      "epoch": 3.4210882856029943,
      "grad_norm": 0.2768961787223816,
      "learning_rate": 6.346078346310417e-05,
      "loss": 1.4118,
      "step": 79770
    },
    {
      "epoch": 3.4215760326519553,
      "eval_loss": 2.4365317821502686,
      "eval_runtime": 21.997,
      "eval_samples_per_second": 227.304,
      "eval_steps_per_second": 1.227,
      "step": 79779
    },
    {
      "epoch": 3.4216302267685066,
      "grad_norm": 0.16737370193004608,
      "learning_rate": 6.344522721408453e-05,
      "loss": 1.416,
      "step": 79780
    },
    {
      "epoch": 3.4221721679340185,
      "grad_norm": 0.1741742491722107,
      "learning_rate": 6.342966991890677e-05,
      "loss": 1.4094,
      "step": 79790
    },
    {
      "epoch": 3.4227141090995308,
      "grad_norm": 0.19367846846580505,
      "learning_rate": 6.341411157949812e-05,
      "loss": 1.4087,
      "step": 79800
    },
    {
      "epoch": 3.423256050265043,
      "grad_norm": 0.17661771178245544,
      "learning_rate": 6.339855219778586e-05,
      "loss": 1.4114,
      "step": 79810
    },
    {
      "epoch": 3.4237979914305554,
      "grad_norm": 0.1794527918100357,
      "learning_rate": 6.338299177569739e-05,
      "loss": 1.4256,
      "step": 79820
    },
    {
      "epoch": 3.4243399325960677,
      "grad_norm": 0.16882766783237457,
      "learning_rate": 6.336743031516031e-05,
      "loss": 1.4138,
      "step": 79830
    },
    {
      "epoch": 3.4248818737615796,
      "grad_norm": 0.2601432204246521,
      "learning_rate": 6.33518678181023e-05,
      "loss": 1.4253,
      "step": 79840
    },
    {
      "epoch": 3.425423814927092,
      "grad_norm": 0.15983355045318604,
      "learning_rate": 6.333630428645116e-05,
      "loss": 1.4191,
      "step": 79850
    },
    {
      "epoch": 3.4259657560926042,
      "grad_norm": 0.2166626751422882,
      "learning_rate": 6.332073972213482e-05,
      "loss": 1.4176,
      "step": 79860
    },
    {
      "epoch": 3.4262909207919114,
      "eval_loss": 2.441222906112671,
      "eval_runtime": 21.995,
      "eval_samples_per_second": 227.324,
      "eval_steps_per_second": 1.228,
      "step": 79866
    },
    {
      "epoch": 3.4265076972581165,
      "grad_norm": 0.27949678897857666,
      "learning_rate": 6.330517412708138e-05,
      "loss": 1.4055,
      "step": 79870
    },
    {
      "epoch": 3.427049638423629,
      "grad_norm": 0.23983129858970642,
      "learning_rate": 6.328960750321903e-05,
      "loss": 1.4154,
      "step": 79880
    },
    {
      "epoch": 3.4275915795891407,
      "grad_norm": 0.18861855566501617,
      "learning_rate": 6.32740398524761e-05,
      "loss": 1.4094,
      "step": 79890
    },
    {
      "epoch": 3.428133520754653,
      "grad_norm": 0.2114916443824768,
      "learning_rate": 6.325847117678102e-05,
      "loss": 1.4138,
      "step": 79900
    },
    {
      "epoch": 3.4286754619201654,
      "grad_norm": 0.22669847309589386,
      "learning_rate": 6.32429014780624e-05,
      "loss": 1.4116,
      "step": 79910
    },
    {
      "epoch": 3.4292174030856777,
      "grad_norm": 0.18799784779548645,
      "learning_rate": 6.322733075824891e-05,
      "loss": 1.4199,
      "step": 79920
    },
    {
      "epoch": 3.4297593442511896,
      "grad_norm": 0.2161915898323059,
      "learning_rate": 6.321175901926941e-05,
      "loss": 1.4138,
      "step": 79930
    },
    {
      "epoch": 3.430301285416702,
      "grad_norm": 0.17232954502105713,
      "learning_rate": 6.319618626305288e-05,
      "loss": 1.4154,
      "step": 79940
    },
    {
      "epoch": 3.430843226582214,
      "grad_norm": 0.20791089534759521,
      "learning_rate": 6.318061249152835e-05,
      "loss": 1.4172,
      "step": 79950
    },
    {
      "epoch": 3.431005808931868,
      "eval_loss": 2.4264652729034424,
      "eval_runtime": 21.9927,
      "eval_samples_per_second": 227.349,
      "eval_steps_per_second": 1.228,
      "step": 79953
    },
    {
      "epoch": 3.4313851677477265,
      "grad_norm": 0.22153067588806152,
      "learning_rate": 6.316503770662508e-05,
      "loss": 1.4078,
      "step": 79960
    },
    {
      "epoch": 3.4319271089132384,
      "grad_norm": 0.17650219798088074,
      "learning_rate": 6.314946191027238e-05,
      "loss": 1.4162,
      "step": 79970
    },
    {
      "epoch": 3.4324690500787507,
      "grad_norm": 0.24956724047660828,
      "learning_rate": 6.313388510439972e-05,
      "loss": 1.4138,
      "step": 79980
    },
    {
      "epoch": 3.433010991244263,
      "grad_norm": 0.17035385966300964,
      "learning_rate": 6.311830729093669e-05,
      "loss": 1.4043,
      "step": 79990
    },
    {
      "epoch": 3.4335529324097753,
      "grad_norm": 0.16698023676872253,
      "learning_rate": 6.3102728471813e-05,
      "loss": 1.4169,
      "step": 80000
    },
    {
      "epoch": 3.4340948735752876,
      "grad_norm": 0.19518424570560455,
      "learning_rate": 6.308714864895847e-05,
      "loss": 1.4078,
      "step": 80010
    },
    {
      "epoch": 3.4346368147407995,
      "grad_norm": 0.27307751774787903,
      "learning_rate": 6.30715678243031e-05,
      "loss": 1.4151,
      "step": 80020
    },
    {
      "epoch": 3.435178755906312,
      "grad_norm": 0.2107393890619278,
      "learning_rate": 6.305598599977694e-05,
      "loss": 1.418,
      "step": 80030
    },
    {
      "epoch": 3.435720697071824,
      "grad_norm": 0.20705653727054596,
      "learning_rate": 6.304040317731022e-05,
      "loss": 1.4223,
      "step": 80040
    },
    {
      "epoch": 3.435720697071824,
      "eval_loss": 2.445282459259033,
      "eval_runtime": 21.9862,
      "eval_samples_per_second": 227.416,
      "eval_steps_per_second": 1.228,
      "step": 80040
    },
    {
      "epoch": 3.4362626382373365,
      "grad_norm": 0.1651361584663391,
      "learning_rate": 6.302481935883325e-05,
      "loss": 1.4108,
      "step": 80050
    },
    {
      "epoch": 3.436804579402849,
      "grad_norm": 0.16168753802776337,
      "learning_rate": 6.300923454627649e-05,
      "loss": 1.4176,
      "step": 80060
    },
    {
      "epoch": 3.4373465205683607,
      "grad_norm": 0.18073947727680206,
      "learning_rate": 6.299364874157054e-05,
      "loss": 1.3954,
      "step": 80070
    },
    {
      "epoch": 3.437888461733873,
      "grad_norm": 0.2578413784503937,
      "learning_rate": 6.297806194664609e-05,
      "loss": 1.4147,
      "step": 80080
    },
    {
      "epoch": 3.4384304028993853,
      "grad_norm": 0.16976307332515717,
      "learning_rate": 6.296247416343396e-05,
      "loss": 1.4134,
      "step": 80090
    },
    {
      "epoch": 3.4389723440648976,
      "grad_norm": 0.16870634257793427,
      "learning_rate": 6.29468853938651e-05,
      "loss": 1.4163,
      "step": 80100
    },
    {
      "epoch": 3.43951428523041,
      "grad_norm": 0.23185555636882782,
      "learning_rate": 6.293129563987057e-05,
      "loss": 1.4076,
      "step": 80110
    },
    {
      "epoch": 3.440056226395922,
      "grad_norm": 0.15721189975738525,
      "learning_rate": 6.291570490338159e-05,
      "loss": 1.412,
      "step": 80120
    },
    {
      "epoch": 3.4404355852117803,
      "eval_loss": 2.428380250930786,
      "eval_runtime": 21.9892,
      "eval_samples_per_second": 227.384,
      "eval_steps_per_second": 1.228,
      "step": 80127
    },
    {
      "epoch": 3.440598167561434,
      "grad_norm": 0.2671518623828888,
      "learning_rate": 6.290011318632945e-05,
      "loss": 1.4068,
      "step": 80130
    },
    {
      "epoch": 3.4411401087269464,
      "grad_norm": 0.1867525279521942,
      "learning_rate": 6.288452049064558e-05,
      "loss": 1.4163,
      "step": 80140
    },
    {
      "epoch": 3.4416820498924587,
      "grad_norm": 0.17865586280822754,
      "learning_rate": 6.286892681826154e-05,
      "loss": 1.409,
      "step": 80150
    },
    {
      "epoch": 3.4422239910579706,
      "grad_norm": 0.19090549647808075,
      "learning_rate": 6.285333217110901e-05,
      "loss": 1.4124,
      "step": 80160
    },
    {
      "epoch": 3.442765932223483,
      "grad_norm": 0.17862895131111145,
      "learning_rate": 6.283773655111979e-05,
      "loss": 1.4123,
      "step": 80170
    },
    {
      "epoch": 3.4433078733889952,
      "grad_norm": 0.17863786220550537,
      "learning_rate": 6.28221399602258e-05,
      "loss": 1.4109,
      "step": 80180
    },
    {
      "epoch": 3.4438498145545076,
      "grad_norm": 0.18190784752368927,
      "learning_rate": 6.280654240035906e-05,
      "loss": 1.4188,
      "step": 80190
    },
    {
      "epoch": 3.4443917557200194,
      "grad_norm": 0.21105967462062836,
      "learning_rate": 6.279094387345173e-05,
      "loss": 1.4069,
      "step": 80200
    },
    {
      "epoch": 3.4449336968855317,
      "grad_norm": 0.24444468319416046,
      "learning_rate": 6.277534438143612e-05,
      "loss": 1.4072,
      "step": 80210
    },
    {
      "epoch": 3.445150473351737,
      "eval_loss": 2.4381296634674072,
      "eval_runtime": 21.9948,
      "eval_samples_per_second": 227.326,
      "eval_steps_per_second": 1.228,
      "step": 80214
    },
    {
      "epoch": 3.445475638051044,
      "grad_norm": 0.16191159188747406,
      "learning_rate": 6.27597439262446e-05,
      "loss": 1.4072,
      "step": 80220
    },
    {
      "epoch": 3.4460175792165564,
      "grad_norm": 0.19114407896995544,
      "learning_rate": 6.27441425098097e-05,
      "loss": 1.4165,
      "step": 80230
    },
    {
      "epoch": 3.4465595203820687,
      "grad_norm": 0.18071117997169495,
      "learning_rate": 6.272854013406403e-05,
      "loss": 1.4204,
      "step": 80240
    },
    {
      "epoch": 3.4471014615475806,
      "grad_norm": 0.2497524917125702,
      "learning_rate": 6.271293680094037e-05,
      "loss": 1.4179,
      "step": 80250
    },
    {
      "epoch": 3.447643402713093,
      "grad_norm": 0.17332278192043304,
      "learning_rate": 6.269733251237159e-05,
      "loss": 1.4071,
      "step": 80260
    },
    {
      "epoch": 3.448185343878605,
      "grad_norm": 0.24143235385417938,
      "learning_rate": 6.268172727029065e-05,
      "loss": 1.4087,
      "step": 80270
    },
    {
      "epoch": 3.4487272850441175,
      "grad_norm": 0.18241146206855774,
      "learning_rate": 6.266612107663072e-05,
      "loss": 1.4128,
      "step": 80280
    },
    {
      "epoch": 3.44926922620963,
      "grad_norm": 0.386482834815979,
      "learning_rate": 6.265051393332498e-05,
      "loss": 1.398,
      "step": 80290
    },
    {
      "epoch": 3.4498111673751417,
      "grad_norm": 0.22998222708702087,
      "learning_rate": 6.26349058423068e-05,
      "loss": 1.4189,
      "step": 80300
    },
    {
      "epoch": 3.449865361491693,
      "eval_loss": 2.4301164150238037,
      "eval_runtime": 21.9925,
      "eval_samples_per_second": 227.35,
      "eval_steps_per_second": 1.228,
      "step": 80301
    },
    {
      "epoch": 3.450353108540654,
      "grad_norm": 0.37004634737968445,
      "learning_rate": 6.261929680550963e-05,
      "loss": 1.4088,
      "step": 80310
    },
    {
      "epoch": 3.4508950497061663,
      "grad_norm": 0.19532030820846558,
      "learning_rate": 6.260368682486704e-05,
      "loss": 1.4071,
      "step": 80320
    },
    {
      "epoch": 3.4514369908716787,
      "grad_norm": 0.1866229772567749,
      "learning_rate": 6.258807590231275e-05,
      "loss": 1.4214,
      "step": 80330
    },
    {
      "epoch": 3.4519789320371905,
      "grad_norm": 0.18280602991580963,
      "learning_rate": 6.257246403978056e-05,
      "loss": 1.4061,
      "step": 80340
    },
    {
      "epoch": 3.452520873202703,
      "grad_norm": 0.2126053422689438,
      "learning_rate": 6.255685123920437e-05,
      "loss": 1.4077,
      "step": 80350
    },
    {
      "epoch": 3.453062814368215,
      "grad_norm": 0.1738966852426529,
      "learning_rate": 6.25412375025183e-05,
      "loss": 1.4177,
      "step": 80360
    },
    {
      "epoch": 3.4536047555337275,
      "grad_norm": 0.23334649205207825,
      "learning_rate": 6.252562283165645e-05,
      "loss": 1.412,
      "step": 80370
    },
    {
      "epoch": 3.4541466966992393,
      "grad_norm": 0.20518024265766144,
      "learning_rate": 6.251000722855312e-05,
      "loss": 1.4076,
      "step": 80380
    },
    {
      "epoch": 3.4545802496316496,
      "eval_loss": 2.419971227645874,
      "eval_runtime": 22.0015,
      "eval_samples_per_second": 227.257,
      "eval_steps_per_second": 1.227,
      "step": 80388
    },
    {
      "epoch": 3.4546886378647517,
      "grad_norm": 0.23325951397418976,
      "learning_rate": 6.249439069514269e-05,
      "loss": 1.4109,
      "step": 80390
    },
    {
      "epoch": 3.455230579030264,
      "grad_norm": 0.20998698472976685,
      "learning_rate": 6.247877323335967e-05,
      "loss": 1.4102,
      "step": 80400
    },
    {
      "epoch": 3.4557725201957763,
      "grad_norm": 0.17500971257686615,
      "learning_rate": 6.246315484513873e-05,
      "loss": 1.4089,
      "step": 80410
    },
    {
      "epoch": 3.4563144613612886,
      "grad_norm": 0.19114546477794647,
      "learning_rate": 6.244753553241455e-05,
      "loss": 1.4072,
      "step": 80420
    },
    {
      "epoch": 3.4568564025268005,
      "grad_norm": 0.28155219554901123,
      "learning_rate": 6.243191529712199e-05,
      "loss": 1.4249,
      "step": 80430
    },
    {
      "epoch": 3.457398343692313,
      "grad_norm": 0.18383963406085968,
      "learning_rate": 6.241629414119603e-05,
      "loss": 1.4128,
      "step": 80440
    },
    {
      "epoch": 3.457940284857825,
      "grad_norm": 0.2947709262371063,
      "learning_rate": 6.240067206657177e-05,
      "loss": 1.4164,
      "step": 80450
    },
    {
      "epoch": 3.4584822260233374,
      "grad_norm": 0.26993468403816223,
      "learning_rate": 6.238504907518437e-05,
      "loss": 1.4158,
      "step": 80460
    },
    {
      "epoch": 3.4590241671888498,
      "grad_norm": 0.1699138730764389,
      "learning_rate": 6.236942516896915e-05,
      "loss": 1.4134,
      "step": 80470
    },
    {
      "epoch": 3.4592951377716057,
      "eval_loss": 2.409397840499878,
      "eval_runtime": 21.9932,
      "eval_samples_per_second": 227.343,
      "eval_steps_per_second": 1.228,
      "step": 80475
    },
    {
      "epoch": 3.4595661083543616,
      "grad_norm": 0.1866215318441391,
      "learning_rate": 6.235380034986154e-05,
      "loss": 1.4173,
      "step": 80480
    },
    {
      "epoch": 3.460108049519874,
      "grad_norm": 0.1942782998085022,
      "learning_rate": 6.233817461979707e-05,
      "loss": 1.4136,
      "step": 80490
    },
    {
      "epoch": 3.4606499906853863,
      "grad_norm": 0.23292642831802368,
      "learning_rate": 6.23225479807114e-05,
      "loss": 1.4177,
      "step": 80500
    },
    {
      "epoch": 3.4611919318508986,
      "grad_norm": 0.19163425266742706,
      "learning_rate": 6.230692043454027e-05,
      "loss": 1.4147,
      "step": 80510
    },
    {
      "epoch": 3.461733873016411,
      "grad_norm": 0.18934902548789978,
      "learning_rate": 6.229129198321955e-05,
      "loss": 1.4062,
      "step": 80520
    },
    {
      "epoch": 3.4622758141819228,
      "grad_norm": 0.16462522745132446,
      "learning_rate": 6.227566262868523e-05,
      "loss": 1.419,
      "step": 80530
    },
    {
      "epoch": 3.462817755347435,
      "grad_norm": 0.16511207818984985,
      "learning_rate": 6.226003237287343e-05,
      "loss": 1.4138,
      "step": 80540
    },
    {
      "epoch": 3.4633596965129474,
      "grad_norm": 0.23615549504756927,
      "learning_rate": 6.224440121772034e-05,
      "loss": 1.4121,
      "step": 80550
    },
    {
      "epoch": 3.4639016376784597,
      "grad_norm": 0.19872863590717316,
      "learning_rate": 6.222876916516225e-05,
      "loss": 1.4056,
      "step": 80560
    },
    {
      "epoch": 3.464010025911562,
      "eval_loss": 2.413600444793701,
      "eval_runtime": 21.9886,
      "eval_samples_per_second": 227.391,
      "eval_steps_per_second": 1.228,
      "step": 80562
    },
    {
      "epoch": 3.4644435788439716,
      "grad_norm": 0.2045847475528717,
      "learning_rate": 6.221313621713566e-05,
      "loss": 1.4146,
      "step": 80570
    },
    {
      "epoch": 3.464985520009484,
      "grad_norm": 0.3229454755783081,
      "learning_rate": 6.219750237557704e-05,
      "loss": 1.4039,
      "step": 80580
    },
    {
      "epoch": 3.465527461174996,
      "grad_norm": 0.20860926806926727,
      "learning_rate": 6.218186764242308e-05,
      "loss": 1.414,
      "step": 80590
    },
    {
      "epoch": 3.4660694023405085,
      "grad_norm": 0.22892901301383972,
      "learning_rate": 6.216623201961054e-05,
      "loss": 1.4056,
      "step": 80600
    },
    {
      "epoch": 3.4666113435060204,
      "grad_norm": 0.20201945304870605,
      "learning_rate": 6.215059550907632e-05,
      "loss": 1.4123,
      "step": 80610
    },
    {
      "epoch": 3.4671532846715327,
      "grad_norm": 0.21173055469989777,
      "learning_rate": 6.213495811275733e-05,
      "loss": 1.4081,
      "step": 80620
    },
    {
      "epoch": 3.467695225837045,
      "grad_norm": 0.20124772191047668,
      "learning_rate": 6.211931983259073e-05,
      "loss": 1.4108,
      "step": 80630
    },
    {
      "epoch": 3.4682371670025574,
      "grad_norm": 0.2871626019477844,
      "learning_rate": 6.210368067051368e-05,
      "loss": 1.4115,
      "step": 80640
    },
    {
      "epoch": 3.4687249140515184,
      "eval_loss": 2.4178643226623535,
      "eval_runtime": 21.9935,
      "eval_samples_per_second": 227.34,
      "eval_steps_per_second": 1.228,
      "step": 80649
    },
    {
      "epoch": 3.4687791081680697,
      "grad_norm": 0.17180606722831726,
      "learning_rate": 6.208804062846353e-05,
      "loss": 1.4096,
      "step": 80650
    },
    {
      "epoch": 3.4693210493335815,
      "grad_norm": 0.250835657119751,
      "learning_rate": 6.207239970837767e-05,
      "loss": 1.4025,
      "step": 80660
    },
    {
      "epoch": 3.469862990499094,
      "grad_norm": 0.20965339243412018,
      "learning_rate": 6.205675791219365e-05,
      "loss": 1.423,
      "step": 80670
    },
    {
      "epoch": 3.470404931664606,
      "grad_norm": 0.1662900745868683,
      "learning_rate": 6.204111524184907e-05,
      "loss": 1.406,
      "step": 80680
    },
    {
      "epoch": 3.4709468728301185,
      "grad_norm": 0.16801480948925018,
      "learning_rate": 6.202547169928173e-05,
      "loss": 1.4126,
      "step": 80690
    },
    {
      "epoch": 3.471488813995631,
      "grad_norm": 0.1683754026889801,
      "learning_rate": 6.200982728642945e-05,
      "loss": 1.4136,
      "step": 80700
    },
    {
      "epoch": 3.4720307551611427,
      "grad_norm": 0.17398451268672943,
      "learning_rate": 6.19941820052302e-05,
      "loss": 1.396,
      "step": 80710
    },
    {
      "epoch": 3.472572696326655,
      "grad_norm": 0.25240159034729004,
      "learning_rate": 6.197853585762204e-05,
      "loss": 1.4273,
      "step": 80720
    },
    {
      "epoch": 3.4731146374921673,
      "grad_norm": 0.2683483064174652,
      "learning_rate": 6.196288884554315e-05,
      "loss": 1.4114,
      "step": 80730
    },
    {
      "epoch": 3.4734398021914745,
      "eval_loss": 2.4308483600616455,
      "eval_runtime": 21.9921,
      "eval_samples_per_second": 227.355,
      "eval_steps_per_second": 1.228,
      "step": 80736
    },
    {
      "epoch": 3.4736565786576796,
      "grad_norm": 0.1892116814851761,
      "learning_rate": 6.194724097093184e-05,
      "loss": 1.4226,
      "step": 80740
    },
    {
      "epoch": 3.474198519823192,
      "grad_norm": 0.22284260392189026,
      "learning_rate": 6.193159223572647e-05,
      "loss": 1.4117,
      "step": 80750
    },
    {
      "epoch": 3.474740460988704,
      "grad_norm": 0.32861995697021484,
      "learning_rate": 6.191594264186556e-05,
      "loss": 1.4114,
      "step": 80760
    },
    {
      "epoch": 3.475282402154216,
      "grad_norm": 0.16212435066699982,
      "learning_rate": 6.190029219128769e-05,
      "loss": 1.4031,
      "step": 80770
    },
    {
      "epoch": 3.4758243433197284,
      "grad_norm": 0.16907572746276855,
      "learning_rate": 6.188464088593157e-05,
      "loss": 1.4082,
      "step": 80780
    },
    {
      "epoch": 3.4763662844852408,
      "grad_norm": 0.16995957493782043,
      "learning_rate": 6.186898872773605e-05,
      "loss": 1.4242,
      "step": 80790
    },
    {
      "epoch": 3.4769082256507526,
      "grad_norm": 0.21154280006885529,
      "learning_rate": 6.185333571864001e-05,
      "loss": 1.4095,
      "step": 80800
    },
    {
      "epoch": 3.477450166816265,
      "grad_norm": 0.22190575301647186,
      "learning_rate": 6.183768186058252e-05,
      "loss": 1.4173,
      "step": 80810
    },
    {
      "epoch": 3.4779921079817773,
      "grad_norm": 0.1600603610277176,
      "learning_rate": 6.182202715550266e-05,
      "loss": 1.4161,
      "step": 80820
    },
    {
      "epoch": 3.478154690331431,
      "eval_loss": 2.4302024841308594,
      "eval_runtime": 21.996,
      "eval_samples_per_second": 227.314,
      "eval_steps_per_second": 1.227,
      "step": 80823
    },
    {
      "epoch": 3.4785340491472896,
      "grad_norm": 0.15820740163326263,
      "learning_rate": 6.18063716053397e-05,
      "loss": 1.4155,
      "step": 80830
    },
    {
      "epoch": 3.4790759903128015,
      "grad_norm": 0.20332497358322144,
      "learning_rate": 6.179071521203299e-05,
      "loss": 1.4101,
      "step": 80840
    },
    {
      "epoch": 3.4796179314783138,
      "grad_norm": 0.23405461013317108,
      "learning_rate": 6.177505797752195e-05,
      "loss": 1.4245,
      "step": 80850
    },
    {
      "epoch": 3.480159872643826,
      "grad_norm": 0.36314642429351807,
      "learning_rate": 6.175939990374613e-05,
      "loss": 1.415,
      "step": 80860
    },
    {
      "epoch": 3.4807018138093384,
      "grad_norm": 0.3759239614009857,
      "learning_rate": 6.174374099264522e-05,
      "loss": 1.4123,
      "step": 80870
    },
    {
      "epoch": 3.4812437549748507,
      "grad_norm": 0.18687903881072998,
      "learning_rate": 6.172808124615895e-05,
      "loss": 1.4108,
      "step": 80880
    },
    {
      "epoch": 3.4817856961403626,
      "grad_norm": 0.27162638306617737,
      "learning_rate": 6.171242066622718e-05,
      "loss": 1.4069,
      "step": 80890
    },
    {
      "epoch": 3.482327637305875,
      "grad_norm": 0.16891661286354065,
      "learning_rate": 6.16967592547899e-05,
      "loss": 1.4246,
      "step": 80900
    },
    {
      "epoch": 3.4828695784713872,
      "grad_norm": 0.24053658545017242,
      "learning_rate": 6.168109701378712e-05,
      "loss": 1.4098,
      "step": 80910
    },
    {
      "epoch": 3.4828695784713872,
      "eval_loss": 2.433274269104004,
      "eval_runtime": 22.0653,
      "eval_samples_per_second": 226.601,
      "eval_steps_per_second": 1.224,
      "step": 80910
    },
    {
      "epoch": 3.4834115196368995,
      "grad_norm": 0.16287986934185028,
      "learning_rate": 6.166543394515906e-05,
      "loss": 1.4179,
      "step": 80920
    },
    {
      "epoch": 3.483953460802412,
      "grad_norm": 0.2310248166322708,
      "learning_rate": 6.1649770050846e-05,
      "loss": 1.4039,
      "step": 80930
    },
    {
      "epoch": 3.4844954019679237,
      "grad_norm": 0.1746436357498169,
      "learning_rate": 6.16341053327883e-05,
      "loss": 1.4139,
      "step": 80940
    },
    {
      "epoch": 3.485037343133436,
      "grad_norm": 0.17406900227069855,
      "learning_rate": 6.16184397929264e-05,
      "loss": 1.4117,
      "step": 80950
    },
    {
      "epoch": 3.4855792842989484,
      "grad_norm": 0.17554210126399994,
      "learning_rate": 6.160277343320095e-05,
      "loss": 1.4133,
      "step": 80960
    },
    {
      "epoch": 3.4861212254644607,
      "grad_norm": 0.1602802872657776,
      "learning_rate": 6.158710625555257e-05,
      "loss": 1.4056,
      "step": 80970
    },
    {
      "epoch": 3.4866631666299726,
      "grad_norm": 0.19458161294460297,
      "learning_rate": 6.157143826192207e-05,
      "loss": 1.4136,
      "step": 80980
    },
    {
      "epoch": 3.487205107795485,
      "grad_norm": 0.23610615730285645,
      "learning_rate": 6.155576945425032e-05,
      "loss": 1.414,
      "step": 80990
    },
    {
      "epoch": 3.4875844666113434,
      "eval_loss": 2.429168701171875,
      "eval_runtime": 22.2262,
      "eval_samples_per_second": 224.96,
      "eval_steps_per_second": 1.215,
      "step": 80997
    },
    {
      "epoch": 3.487747048960997,
      "grad_norm": 0.18065819144248962,
      "learning_rate": 6.154009983447834e-05,
      "loss": 1.4076,
      "step": 81000
    },
    {
      "epoch": 4.000541941165512,
      "grad_norm": 0.2180778533220291,
      "learning_rate": 6.152442940454717e-05,
      "loss": 1.3974,
      "step": 81010
    },
    {
      "epoch": 4.001083882331025,
      "grad_norm": 0.20014463365077972,
      "learning_rate": 6.1508758166398e-05,
      "loss": 1.4054,
      "step": 81020
    },
    {
      "epoch": 4.0016258234965365,
      "grad_norm": 0.1851683109998703,
      "learning_rate": 6.149308612197213e-05,
      "loss": 1.4141,
      "step": 81030
    },
    {
      "epoch": 4.002167764662049,
      "grad_norm": 0.1777936816215515,
      "learning_rate": 6.147741327321095e-05,
      "loss": 1.4072,
      "step": 81040
    },
    {
      "epoch": 4.002709705827561,
      "grad_norm": 0.17153799533843994,
      "learning_rate": 6.146173962205594e-05,
      "loss": 1.4019,
      "step": 81050
    },
    {
      "epoch": 4.003251646993073,
      "grad_norm": 0.18811866641044617,
      "learning_rate": 6.144606517044869e-05,
      "loss": 1.4112,
      "step": 81060
    },
    {
      "epoch": 4.003793588158586,
      "grad_norm": 0.23102541267871857,
      "learning_rate": 6.143038992033083e-05,
      "loss": 1.4195,
      "step": 81070
    },
    {
      "epoch": 4.004335529324098,
      "grad_norm": 0.2683873176574707,
      "learning_rate": 6.141471387364423e-05,
      "loss": 1.3947,
      "step": 81080
    },
    {
      "epoch": 4.004552305790303,
      "eval_loss": 2.4371650218963623,
      "eval_runtime": 22.4585,
      "eval_samples_per_second": 222.633,
      "eval_steps_per_second": 1.202,
      "step": 81084
    },
    {
      "epoch": 4.00487747048961,
      "grad_norm": 0.15493124723434448,
      "learning_rate": 6.139903703233069e-05,
      "loss": 1.4126,
      "step": 81090
    },
    {
      "epoch": 4.005419411655122,
      "grad_norm": 0.17667770385742188,
      "learning_rate": 6.138335939833225e-05,
      "loss": 1.403,
      "step": 81100
    },
    {
      "epoch": 4.005961352820634,
      "grad_norm": 0.1961461454629898,
      "learning_rate": 6.136768097359096e-05,
      "loss": 1.4091,
      "step": 81110
    },
    {
      "epoch": 4.006503293986147,
      "grad_norm": 0.15184424817562103,
      "learning_rate": 6.135200176004897e-05,
      "loss": 1.4149,
      "step": 81120
    },
    {
      "epoch": 4.007045235151659,
      "grad_norm": 0.18130755424499512,
      "learning_rate": 6.133632175964861e-05,
      "loss": 1.4037,
      "step": 81130
    },
    {
      "epoch": 4.007587176317171,
      "grad_norm": 0.1635865420103073,
      "learning_rate": 6.132064097433222e-05,
      "loss": 1.4111,
      "step": 81140
    },
    {
      "epoch": 4.008129117482683,
      "grad_norm": 0.22441507875919342,
      "learning_rate": 6.130495940604225e-05,
      "loss": 1.4135,
      "step": 81150
    },
    {
      "epoch": 4.008671058648195,
      "grad_norm": 0.30155861377716064,
      "learning_rate": 6.128927705672129e-05,
      "loss": 1.4027,
      "step": 81160
    },
    {
      "epoch": 4.009212999813708,
      "grad_norm": 0.3162151873111725,
      "learning_rate": 6.127359392831198e-05,
      "loss": 1.4073,
      "step": 81170
    },
    {
      "epoch": 4.009267193930259,
      "eval_loss": 2.4355599880218506,
      "eval_runtime": 21.9746,
      "eval_samples_per_second": 227.536,
      "eval_steps_per_second": 1.229,
      "step": 81171
    },
    {
      "epoch": 4.00975494097922,
      "grad_norm": 0.22945579886436462,
      "learning_rate": 6.12579100227571e-05,
      "loss": 1.4077,
      "step": 81180
    },
    {
      "epoch": 4.010296882144732,
      "grad_norm": 0.25422438979148865,
      "learning_rate": 6.124222534199952e-05,
      "loss": 1.4,
      "step": 81190
    },
    {
      "epoch": 4.0108388233102445,
      "grad_norm": 0.21404403448104858,
      "learning_rate": 6.122653988798214e-05,
      "loss": 1.4092,
      "step": 81200
    },
    {
      "epoch": 4.011380764475756,
      "grad_norm": 0.17690123617649078,
      "learning_rate": 6.121085366264802e-05,
      "loss": 1.4084,
      "step": 81210
    },
    {
      "epoch": 4.011922705641269,
      "grad_norm": 0.18206678330898285,
      "learning_rate": 6.119516666794034e-05,
      "loss": 1.4086,
      "step": 81220
    },
    {
      "epoch": 4.012464646806781,
      "grad_norm": 0.20577572286128998,
      "learning_rate": 6.11794789058023e-05,
      "loss": 1.4019,
      "step": 81230
    },
    {
      "epoch": 4.013006587972293,
      "grad_norm": 0.16592557728290558,
      "learning_rate": 6.116379037817724e-05,
      "loss": 1.4091,
      "step": 81240
    },
    {
      "epoch": 4.013548529137806,
      "grad_norm": 0.1908210813999176,
      "learning_rate": 6.114810108700857e-05,
      "loss": 1.4164,
      "step": 81250
    },
    {
      "epoch": 4.013982082070215,
      "eval_loss": 2.4322612285614014,
      "eval_runtime": 21.9747,
      "eval_samples_per_second": 227.534,
      "eval_steps_per_second": 1.229,
      "step": 81258
    },
    {
      "epoch": 4.014090470303318,
      "grad_norm": 0.1619943529367447,
      "learning_rate": 6.113241103423986e-05,
      "loss": 1.4089,
      "step": 81260
    },
    {
      "epoch": 4.01463241146883,
      "grad_norm": 0.17974944412708282,
      "learning_rate": 6.11167202218147e-05,
      "loss": 1.4134,
      "step": 81270
    },
    {
      "epoch": 4.015174352634342,
      "grad_norm": 0.1860845535993576,
      "learning_rate": 6.110102865167677e-05,
      "loss": 1.4062,
      "step": 81280
    },
    {
      "epoch": 4.015716293799854,
      "grad_norm": 0.16427412629127502,
      "learning_rate": 6.108533632576992e-05,
      "loss": 1.3899,
      "step": 81290
    },
    {
      "epoch": 4.016258234965367,
      "grad_norm": 0.17551249265670776,
      "learning_rate": 6.106964324603801e-05,
      "loss": 1.4103,
      "step": 81300
    },
    {
      "epoch": 4.016800176130879,
      "grad_norm": 0.20582075417041779,
      "learning_rate": 6.105394941442503e-05,
      "loss": 1.4072,
      "step": 81310
    },
    {
      "epoch": 4.017342117296391,
      "grad_norm": 0.1872161328792572,
      "learning_rate": 6.10382548328751e-05,
      "loss": 1.4127,
      "step": 81320
    },
    {
      "epoch": 4.017884058461903,
      "grad_norm": 0.18607717752456665,
      "learning_rate": 6.102255950333239e-05,
      "loss": 1.4179,
      "step": 81330
    },
    {
      "epoch": 4.018425999627415,
      "grad_norm": 0.16646206378936768,
      "learning_rate": 6.100686342774115e-05,
      "loss": 1.4067,
      "step": 81340
    },
    {
      "epoch": 4.018696970210171,
      "eval_loss": 2.431658983230591,
      "eval_runtime": 21.9738,
      "eval_samples_per_second": 227.544,
      "eval_steps_per_second": 1.229,
      "step": 81345
    },
    {
      "epoch": 4.018967940792928,
      "grad_norm": 0.33718639612197876,
      "learning_rate": 6.0991166608045745e-05,
      "loss": 1.4178,
      "step": 81350
    },
    {
      "epoch": 4.01950988195844,
      "grad_norm": 0.32710033655166626,
      "learning_rate": 6.097546904619061e-05,
      "loss": 1.4108,
      "step": 81360
    },
    {
      "epoch": 4.020051823123952,
      "grad_norm": 0.21620182693004608,
      "learning_rate": 6.095977074412034e-05,
      "loss": 1.4062,
      "step": 81370
    },
    {
      "epoch": 4.0205937642894645,
      "grad_norm": 0.2620053291320801,
      "learning_rate": 6.0944071703779536e-05,
      "loss": 1.4072,
      "step": 81380
    },
    {
      "epoch": 4.021135705454976,
      "grad_norm": 0.16901114583015442,
      "learning_rate": 6.092837192711294e-05,
      "loss": 1.4053,
      "step": 81390
    },
    {
      "epoch": 4.021677646620489,
      "grad_norm": 0.2617866098880768,
      "learning_rate": 6.091267141606537e-05,
      "loss": 1.4131,
      "step": 81400
    },
    {
      "epoch": 4.022219587786001,
      "grad_norm": 0.218822181224823,
      "learning_rate": 6.089697017258171e-05,
      "loss": 1.4063,
      "step": 81410
    },
    {
      "epoch": 4.022761528951513,
      "grad_norm": 0.167058527469635,
      "learning_rate": 6.088126819860701e-05,
      "loss": 1.4009,
      "step": 81420
    },
    {
      "epoch": 4.023303470117026,
      "grad_norm": 0.1802741438150406,
      "learning_rate": 6.086556549608634e-05,
      "loss": 1.4077,
      "step": 81430
    },
    {
      "epoch": 4.023411858350128,
      "eval_loss": 2.4219465255737305,
      "eval_runtime": 21.9753,
      "eval_samples_per_second": 227.528,
      "eval_steps_per_second": 1.229,
      "step": 81432
    },
    {
      "epoch": 4.0238454112825375,
      "grad_norm": 0.1666201651096344,
      "learning_rate": 6.084986206696487e-05,
      "loss": 1.4031,
      "step": 81440
    },
    {
      "epoch": 4.02438735244805,
      "grad_norm": 0.19865860044956207,
      "learning_rate": 6.08341579131879e-05,
      "loss": 1.3978,
      "step": 81450
    },
    {
      "epoch": 4.024929293613562,
      "grad_norm": 0.1948530673980713,
      "learning_rate": 6.081845303670077e-05,
      "loss": 1.4148,
      "step": 81460
    },
    {
      "epoch": 4.025471234779074,
      "grad_norm": 0.1879536211490631,
      "learning_rate": 6.0802747439448935e-05,
      "loss": 1.411,
      "step": 81470
    },
    {
      "epoch": 4.026013175944587,
      "grad_norm": 0.20297576487064362,
      "learning_rate": 6.078704112337795e-05,
      "loss": 1.406,
      "step": 81480
    },
    {
      "epoch": 4.026555117110099,
      "grad_norm": 0.17799319326877594,
      "learning_rate": 6.077133409043342e-05,
      "loss": 1.4034,
      "step": 81490
    },
    {
      "epoch": 4.027097058275611,
      "grad_norm": 0.23970834910869598,
      "learning_rate": 6.075562634256109e-05,
      "loss": 1.4028,
      "step": 81500
    },
    {
      "epoch": 4.027638999441123,
      "grad_norm": 0.28241363167762756,
      "learning_rate": 6.073991788170675e-05,
      "loss": 1.4104,
      "step": 81510
    },
    {
      "epoch": 4.028126746490084,
      "eval_loss": 2.4167044162750244,
      "eval_runtime": 21.9669,
      "eval_samples_per_second": 227.615,
      "eval_steps_per_second": 1.229,
      "step": 81519
    },
    {
      "epoch": 4.028180940606635,
      "grad_norm": 0.22441983222961426,
      "learning_rate": 6.072420870981631e-05,
      "loss": 1.4092,
      "step": 81520
    },
    {
      "epoch": 4.028722881772148,
      "grad_norm": 0.2237686812877655,
      "learning_rate": 6.070849882883576e-05,
      "loss": 1.4094,
      "step": 81530
    },
    {
      "epoch": 4.02926482293766,
      "grad_norm": 0.17908205091953278,
      "learning_rate": 6.069278824071114e-05,
      "loss": 1.4032,
      "step": 81540
    },
    {
      "epoch": 4.029806764103172,
      "grad_norm": 0.18530701100826263,
      "learning_rate": 6.0677076947388635e-05,
      "loss": 1.4038,
      "step": 81550
    },
    {
      "epoch": 4.030348705268684,
      "grad_norm": 0.17356233298778534,
      "learning_rate": 6.066136495081448e-05,
      "loss": 1.4042,
      "step": 81560
    },
    {
      "epoch": 4.030890646434196,
      "grad_norm": 0.1672087162733078,
      "learning_rate": 6.0645652252935005e-05,
      "loss": 1.408,
      "step": 81570
    },
    {
      "epoch": 4.031432587599709,
      "grad_norm": 0.24931927025318146,
      "learning_rate": 6.0629938855696655e-05,
      "loss": 1.4034,
      "step": 81580
    },
    {
      "epoch": 4.031974528765221,
      "grad_norm": 0.297122061252594,
      "learning_rate": 6.061422476104592e-05,
      "loss": 1.4126,
      "step": 81590
    },
    {
      "epoch": 4.032516469930733,
      "grad_norm": 0.1879911869764328,
      "learning_rate": 6.0598509970929396e-05,
      "loss": 1.4148,
      "step": 81600
    },
    {
      "epoch": 4.03284163463004,
      "eval_loss": 2.4190409183502197,
      "eval_runtime": 21.9863,
      "eval_samples_per_second": 227.414,
      "eval_steps_per_second": 1.228,
      "step": 81606
    },
    {
      "epoch": 4.0330584110962455,
      "grad_norm": 0.16557785868644714,
      "learning_rate": 6.058279448729376e-05,
      "loss": 1.3992,
      "step": 81610
    },
    {
      "epoch": 4.033600352261757,
      "grad_norm": 0.21484564244747162,
      "learning_rate": 6.056707831208579e-05,
      "loss": 1.4228,
      "step": 81620
    },
    {
      "epoch": 4.03414229342727,
      "grad_norm": 0.21425080299377441,
      "learning_rate": 6.055136144725232e-05,
      "loss": 1.4058,
      "step": 81630
    },
    {
      "epoch": 4.034684234592782,
      "grad_norm": 0.16896581649780273,
      "learning_rate": 6.0535643894740304e-05,
      "loss": 1.4159,
      "step": 81640
    },
    {
      "epoch": 4.035226175758294,
      "grad_norm": 0.20105527341365814,
      "learning_rate": 6.0519925656496746e-05,
      "loss": 1.4198,
      "step": 81650
    },
    {
      "epoch": 4.035768116923807,
      "grad_norm": 0.18627329170703888,
      "learning_rate": 6.050420673446876e-05,
      "loss": 1.3985,
      "step": 81660
    },
    {
      "epoch": 4.0363100580893185,
      "grad_norm": 0.1602090746164322,
      "learning_rate": 6.048848713060354e-05,
      "loss": 1.4063,
      "step": 81670
    },
    {
      "epoch": 4.036851999254831,
      "grad_norm": 0.27577728033065796,
      "learning_rate": 6.0472766846848384e-05,
      "loss": 1.4078,
      "step": 81680
    },
    {
      "epoch": 4.037393940420343,
      "grad_norm": 0.24706071615219116,
      "learning_rate": 6.045704588515062e-05,
      "loss": 1.4088,
      "step": 81690
    },
    {
      "epoch": 4.0375565227699965,
      "eval_loss": 2.4332613945007324,
      "eval_runtime": 21.9896,
      "eval_samples_per_second": 227.38,
      "eval_steps_per_second": 1.228,
      "step": 81693
    },
    {
      "epoch": 4.037935881585855,
      "grad_norm": 0.18013040721416473,
      "learning_rate": 6.0441324247457686e-05,
      "loss": 1.4039,
      "step": 81700
    },
    {
      "epoch": 4.038477822751368,
      "grad_norm": 0.2291117012500763,
      "learning_rate": 6.042560193571714e-05,
      "loss": 1.4175,
      "step": 81710
    },
    {
      "epoch": 4.03901976391688,
      "grad_norm": 0.24475926160812378,
      "learning_rate": 6.04098789518766e-05,
      "loss": 1.4029,
      "step": 81720
    },
    {
      "epoch": 4.039561705082392,
      "grad_norm": 0.31280967593193054,
      "learning_rate": 6.039415529788372e-05,
      "loss": 1.4068,
      "step": 81730
    },
    {
      "epoch": 4.040103646247904,
      "grad_norm": 0.274687796831131,
      "learning_rate": 6.037843097568631e-05,
      "loss": 1.4094,
      "step": 81740
    },
    {
      "epoch": 4.040645587413416,
      "grad_norm": 0.1743050217628479,
      "learning_rate": 6.036270598723222e-05,
      "loss": 1.4009,
      "step": 81750
    },
    {
      "epoch": 4.041187528578929,
      "grad_norm": 0.17927388846874237,
      "learning_rate": 6.0346980334469386e-05,
      "loss": 1.4118,
      "step": 81760
    },
    {
      "epoch": 4.041729469744441,
      "grad_norm": 0.2120426744222641,
      "learning_rate": 6.0331254019345864e-05,
      "loss": 1.3993,
      "step": 81770
    },
    {
      "epoch": 4.042271410909953,
      "grad_norm": 0.2474319189786911,
      "learning_rate": 6.0315527043809726e-05,
      "loss": 1.4069,
      "step": 81780
    },
    {
      "epoch": 4.042271410909953,
      "eval_loss": 2.4317309856414795,
      "eval_runtime": 21.9711,
      "eval_samples_per_second": 227.572,
      "eval_steps_per_second": 1.229,
      "step": 81780
    },
    {
      "epoch": 4.042813352075465,
      "grad_norm": 0.22434458136558533,
      "learning_rate": 6.0299799409809175e-05,
      "loss": 1.4062,
      "step": 81790
    },
    {
      "epoch": 4.043355293240977,
      "grad_norm": 0.16757360100746155,
      "learning_rate": 6.028407111929248e-05,
      "loss": 1.4082,
      "step": 81800
    },
    {
      "epoch": 4.04389723440649,
      "grad_norm": 0.23129674792289734,
      "learning_rate": 6.0268342174207994e-05,
      "loss": 1.3953,
      "step": 81810
    },
    {
      "epoch": 4.044439175572002,
      "grad_norm": 0.21264296770095825,
      "learning_rate": 6.025261257650416e-05,
      "loss": 1.4025,
      "step": 81820
    },
    {
      "epoch": 4.044981116737514,
      "grad_norm": 0.20318235456943512,
      "learning_rate": 6.0236882328129495e-05,
      "loss": 1.4144,
      "step": 81830
    },
    {
      "epoch": 4.045523057903027,
      "grad_norm": 0.15029390156269073,
      "learning_rate": 6.022115143103256e-05,
      "loss": 1.4105,
      "step": 81840
    },
    {
      "epoch": 4.0460649990685384,
      "grad_norm": 0.23690800368785858,
      "learning_rate": 6.020541988716207e-05,
      "loss": 1.4113,
      "step": 81850
    },
    {
      "epoch": 4.046606940234051,
      "grad_norm": 0.16659775376319885,
      "learning_rate": 6.018968769846676e-05,
      "loss": 1.4208,
      "step": 81860
    },
    {
      "epoch": 4.04698629904991,
      "eval_loss": 2.436213731765747,
      "eval_runtime": 21.9755,
      "eval_samples_per_second": 227.526,
      "eval_steps_per_second": 1.229,
      "step": 81867
    },
    {
      "epoch": 4.047148881399563,
      "grad_norm": 0.25907376408576965,
      "learning_rate": 6.017395486689547e-05,
      "loss": 1.4063,
      "step": 81870
    },
    {
      "epoch": 4.047690822565075,
      "grad_norm": 0.15979789197444916,
      "learning_rate": 6.015822139439712e-05,
      "loss": 1.4152,
      "step": 81880
    },
    {
      "epoch": 4.048232763730588,
      "grad_norm": 0.1618310660123825,
      "learning_rate": 6.014248728292068e-05,
      "loss": 1.4019,
      "step": 81890
    },
    {
      "epoch": 4.0487747048961,
      "grad_norm": 0.22831520438194275,
      "learning_rate": 6.0126752534415255e-05,
      "loss": 1.4113,
      "step": 81900
    },
    {
      "epoch": 4.049316646061612,
      "grad_norm": 0.24901092052459717,
      "learning_rate": 6.011101715082997e-05,
      "loss": 1.4011,
      "step": 81910
    },
    {
      "epoch": 4.049858587227124,
      "grad_norm": 0.18291537463665009,
      "learning_rate": 6.009528113411409e-05,
      "loss": 1.4053,
      "step": 81920
    },
    {
      "epoch": 4.050400528392636,
      "grad_norm": 0.3508703410625458,
      "learning_rate": 6.007954448621691e-05,
      "loss": 1.4052,
      "step": 81930
    },
    {
      "epoch": 4.050942469558149,
      "grad_norm": 0.22802935540676117,
      "learning_rate": 6.0063807209087784e-05,
      "loss": 1.4107,
      "step": 81940
    },
    {
      "epoch": 4.051484410723661,
      "grad_norm": 0.3096991777420044,
      "learning_rate": 6.004806930467621e-05,
      "loss": 1.4167,
      "step": 81950
    },
    {
      "epoch": 4.051701187189866,
      "eval_loss": 2.434500217437744,
      "eval_runtime": 21.9752,
      "eval_samples_per_second": 227.529,
      "eval_steps_per_second": 1.229,
      "step": 81954
    },
    {
      "epoch": 4.052026351889173,
      "grad_norm": 0.2774633765220642,
      "learning_rate": 6.0032330774931754e-05,
      "loss": 1.4139,
      "step": 81960
    },
    {
      "epoch": 4.052568293054685,
      "grad_norm": 0.16032350063323975,
      "learning_rate": 6.0016591621803986e-05,
      "loss": 1.4121,
      "step": 81970
    },
    {
      "epoch": 4.053110234220197,
      "grad_norm": 0.26607418060302734,
      "learning_rate": 6.000085184724265e-05,
      "loss": 1.4051,
      "step": 81980
    },
    {
      "epoch": 4.05365217538571,
      "grad_norm": 0.26877331733703613,
      "learning_rate": 5.998511145319748e-05,
      "loss": 1.4075,
      "step": 81990
    },
    {
      "epoch": 4.054194116551222,
      "grad_norm": 0.18772627413272858,
      "learning_rate": 5.996937044161835e-05,
      "loss": 1.41,
      "step": 82000
    },
    {
      "epoch": 4.054736057716734,
      "grad_norm": 0.17956435680389404,
      "learning_rate": 5.9953628814455184e-05,
      "loss": 1.4223,
      "step": 82010
    },
    {
      "epoch": 4.0552779988822465,
      "grad_norm": 0.38991832733154297,
      "learning_rate": 5.9937886573657986e-05,
      "loss": 1.4082,
      "step": 82020
    },
    {
      "epoch": 4.055819940047758,
      "grad_norm": 0.20310132205486298,
      "learning_rate": 5.9922143721176846e-05,
      "loss": 1.4081,
      "step": 82030
    },
    {
      "epoch": 4.056361881213271,
      "grad_norm": 0.3149273693561554,
      "learning_rate": 5.990640025896189e-05,
      "loss": 1.4048,
      "step": 82040
    },
    {
      "epoch": 4.056416075329822,
      "eval_loss": 2.435964822769165,
      "eval_runtime": 21.972,
      "eval_samples_per_second": 227.562,
      "eval_steps_per_second": 1.229,
      "step": 82041
    },
    {
      "epoch": 4.056903822378783,
      "grad_norm": 0.16741731762886047,
      "learning_rate": 5.989065618896339e-05,
      "loss": 1.4019,
      "step": 82050
    },
    {
      "epoch": 4.057445763544295,
      "grad_norm": 0.19802165031433105,
      "learning_rate": 5.987491151313164e-05,
      "loss": 1.4076,
      "step": 82060
    },
    {
      "epoch": 4.057987704709808,
      "grad_norm": 0.17228524386882782,
      "learning_rate": 5.9859166233417016e-05,
      "loss": 1.4031,
      "step": 82070
    },
    {
      "epoch": 4.0585296458753195,
      "grad_norm": 0.21210512518882751,
      "learning_rate": 5.984342035176996e-05,
      "loss": 1.4075,
      "step": 82080
    },
    {
      "epoch": 4.059071587040832,
      "grad_norm": 0.2215586155653,
      "learning_rate": 5.982767387014102e-05,
      "loss": 1.3966,
      "step": 82090
    },
    {
      "epoch": 4.059613528206344,
      "grad_norm": 0.19042372703552246,
      "learning_rate": 5.98119267904808e-05,
      "loss": 1.4053,
      "step": 82100
    },
    {
      "epoch": 4.060155469371856,
      "grad_norm": 0.2051682025194168,
      "learning_rate": 5.979617911473999e-05,
      "loss": 1.4144,
      "step": 82110
    },
    {
      "epoch": 4.060697410537369,
      "grad_norm": 0.20389145612716675,
      "learning_rate": 5.978043084486934e-05,
      "loss": 1.4136,
      "step": 82120
    },
    {
      "epoch": 4.061130963469778,
      "eval_loss": 2.4284167289733887,
      "eval_runtime": 21.9746,
      "eval_samples_per_second": 227.536,
      "eval_steps_per_second": 1.229,
      "step": 82128
    },
    {
      "epoch": 4.061239351702881,
      "grad_norm": 0.18237143754959106,
      "learning_rate": 5.9764681982819656e-05,
      "loss": 1.4024,
      "step": 82130
    },
    {
      "epoch": 4.061781292868393,
      "grad_norm": 0.17889219522476196,
      "learning_rate": 5.974893253054186e-05,
      "loss": 1.4037,
      "step": 82140
    },
    {
      "epoch": 4.062323234033905,
      "grad_norm": 0.23920495808124542,
      "learning_rate": 5.973318248998693e-05,
      "loss": 1.4037,
      "step": 82150
    },
    {
      "epoch": 4.062865175199417,
      "grad_norm": 0.18914079666137695,
      "learning_rate": 5.971743186310589e-05,
      "loss": 1.41,
      "step": 82160
    },
    {
      "epoch": 4.06340711636493,
      "grad_norm": 0.21431829035282135,
      "learning_rate": 5.970168065184987e-05,
      "loss": 1.4089,
      "step": 82170
    },
    {
      "epoch": 4.063949057530442,
      "grad_norm": 0.16956184804439545,
      "learning_rate": 5.968592885817007e-05,
      "loss": 1.409,
      "step": 82180
    },
    {
      "epoch": 4.064490998695954,
      "grad_norm": 0.23388327658176422,
      "learning_rate": 5.967017648401775e-05,
      "loss": 1.4165,
      "step": 82190
    },
    {
      "epoch": 4.065032939861466,
      "grad_norm": 0.22132398188114166,
      "learning_rate": 5.965442353134424e-05,
      "loss": 1.3967,
      "step": 82200
    },
    {
      "epoch": 4.065574881026978,
      "grad_norm": 0.32104483246803284,
      "learning_rate": 5.963867000210094e-05,
      "loss": 1.4003,
      "step": 82210
    },
    {
      "epoch": 4.065845851609734,
      "eval_loss": 2.434609889984131,
      "eval_runtime": 21.9734,
      "eval_samples_per_second": 227.548,
      "eval_steps_per_second": 1.229,
      "step": 82215
    },
    {
      "epoch": 4.066116822192491,
      "grad_norm": 0.27653661370277405,
      "learning_rate": 5.962291589823935e-05,
      "loss": 1.4099,
      "step": 82220
    },
    {
      "epoch": 4.066658763358003,
      "grad_norm": 0.18210026621818542,
      "learning_rate": 5.960716122171102e-05,
      "loss": 1.4126,
      "step": 82230
    },
    {
      "epoch": 4.067200704523515,
      "grad_norm": 0.20089492201805115,
      "learning_rate": 5.959140597446753e-05,
      "loss": 1.4042,
      "step": 82240
    },
    {
      "epoch": 4.0677426456890275,
      "grad_norm": 0.20778271555900574,
      "learning_rate": 5.957565015846063e-05,
      "loss": 1.4037,
      "step": 82250
    },
    {
      "epoch": 4.068284586854539,
      "grad_norm": 0.31842777132987976,
      "learning_rate": 5.955989377564203e-05,
      "loss": 1.4034,
      "step": 82260
    },
    {
      "epoch": 4.068826528020052,
      "grad_norm": 0.1979900300502777,
      "learning_rate": 5.95441368279636e-05,
      "loss": 1.391,
      "step": 82270
    },
    {
      "epoch": 4.069368469185564,
      "grad_norm": 0.1853083223104477,
      "learning_rate": 5.952837931737723e-05,
      "loss": 1.417,
      "step": 82280
    },
    {
      "epoch": 4.069910410351076,
      "grad_norm": 0.18772120773792267,
      "learning_rate": 5.951262124583489e-05,
      "loss": 1.4061,
      "step": 82290
    },
    {
      "epoch": 4.070452351516589,
      "grad_norm": 0.207114115357399,
      "learning_rate": 5.9496862615288615e-05,
      "loss": 1.4112,
      "step": 82300
    },
    {
      "epoch": 4.070560739749691,
      "eval_loss": 2.4208648204803467,
      "eval_runtime": 21.9732,
      "eval_samples_per_second": 227.549,
      "eval_steps_per_second": 1.229,
      "step": 82302
    },
    {
      "epoch": 4.070994292682101,
      "grad_norm": 0.21243199706077576,
      "learning_rate": 5.948110342769054e-05,
      "loss": 1.4098,
      "step": 82310
    },
    {
      "epoch": 4.071536233847613,
      "grad_norm": 0.18174535036087036,
      "learning_rate": 5.946534368499281e-05,
      "loss": 1.3921,
      "step": 82320
    },
    {
      "epoch": 4.072078175013125,
      "grad_norm": 0.195078507065773,
      "learning_rate": 5.944958338914769e-05,
      "loss": 1.3993,
      "step": 82330
    },
    {
      "epoch": 4.072620116178637,
      "grad_norm": 0.1927531659603119,
      "learning_rate": 5.943382254210751e-05,
      "loss": 1.4068,
      "step": 82340
    },
    {
      "epoch": 4.07316205734415,
      "grad_norm": 0.3011392652988434,
      "learning_rate": 5.941806114582464e-05,
      "loss": 1.4216,
      "step": 82350
    },
    {
      "epoch": 4.073703998509662,
      "grad_norm": 0.22039276361465454,
      "learning_rate": 5.940229920225154e-05,
      "loss": 1.4045,
      "step": 82360
    },
    {
      "epoch": 4.0742459396751745,
      "grad_norm": 0.1818740963935852,
      "learning_rate": 5.9386536713340726e-05,
      "loss": 1.4074,
      "step": 82370
    },
    {
      "epoch": 4.074787880840686,
      "grad_norm": 0.1722990870475769,
      "learning_rate": 5.9370773681044776e-05,
      "loss": 1.4028,
      "step": 82380
    },
    {
      "epoch": 4.075275627889647,
      "eval_loss": 2.426165819168091,
      "eval_runtime": 21.9727,
      "eval_samples_per_second": 227.555,
      "eval_steps_per_second": 1.229,
      "step": 82389
    },
    {
      "epoch": 4.075329822006198,
      "grad_norm": 0.17619258165359497,
      "learning_rate": 5.935501010731637e-05,
      "loss": 1.4043,
      "step": 82390
    },
    {
      "epoch": 4.075871763171711,
      "grad_norm": 0.1646595150232315,
      "learning_rate": 5.93392459941082e-05,
      "loss": 1.4092,
      "step": 82400
    },
    {
      "epoch": 4.076413704337223,
      "grad_norm": 0.2252381443977356,
      "learning_rate": 5.932348134337311e-05,
      "loss": 1.4022,
      "step": 82410
    },
    {
      "epoch": 4.076955645502735,
      "grad_norm": 0.23596051335334778,
      "learning_rate": 5.9307716157063895e-05,
      "loss": 1.4146,
      "step": 82420
    },
    {
      "epoch": 4.0774975866682475,
      "grad_norm": 0.20619599521160126,
      "learning_rate": 5.9291950437133515e-05,
      "loss": 1.3984,
      "step": 82430
    },
    {
      "epoch": 4.078039527833759,
      "grad_norm": 0.2613559663295746,
      "learning_rate": 5.927618418553495e-05,
      "loss": 1.4036,
      "step": 82440
    },
    {
      "epoch": 4.078581468999272,
      "grad_norm": 0.26218506693840027,
      "learning_rate": 5.9260417404221245e-05,
      "loss": 1.4058,
      "step": 82450
    },
    {
      "epoch": 4.079123410164784,
      "grad_norm": 0.17391842603683472,
      "learning_rate": 5.924465009514554e-05,
      "loss": 1.4002,
      "step": 82460
    },
    {
      "epoch": 4.079665351330296,
      "grad_norm": 0.2079029232263565,
      "learning_rate": 5.9228882260261e-05,
      "loss": 1.4079,
      "step": 82470
    },
    {
      "epoch": 4.0799905160296035,
      "eval_loss": 2.4345686435699463,
      "eval_runtime": 21.9671,
      "eval_samples_per_second": 227.614,
      "eval_steps_per_second": 1.229,
      "step": 82476
    },
    {
      "epoch": 4.080207292495809,
      "grad_norm": 0.16714847087860107,
      "learning_rate": 5.9213113901520875e-05,
      "loss": 1.4065,
      "step": 82480
    },
    {
      "epoch": 4.0807492336613205,
      "grad_norm": 0.19766834378242493,
      "learning_rate": 5.9197345020878515e-05,
      "loss": 1.4094,
      "step": 82490
    },
    {
      "epoch": 4.081291174826833,
      "grad_norm": 0.26485520601272583,
      "learning_rate": 5.918157562028726e-05,
      "loss": 1.4029,
      "step": 82500
    },
    {
      "epoch": 4.081833115992345,
      "grad_norm": 0.25532785058021545,
      "learning_rate": 5.9165805701700595e-05,
      "loss": 1.3951,
      "step": 82510
    },
    {
      "epoch": 4.082375057157857,
      "grad_norm": 0.28087395429611206,
      "learning_rate": 5.915003526707198e-05,
      "loss": 1.4069,
      "step": 82520
    },
    {
      "epoch": 4.08291699832337,
      "grad_norm": 0.306506872177124,
      "learning_rate": 5.9134264318355025e-05,
      "loss": 1.3996,
      "step": 82530
    },
    {
      "epoch": 4.083458939488882,
      "grad_norm": 0.2110404223203659,
      "learning_rate": 5.911849285750335e-05,
      "loss": 1.418,
      "step": 82540
    },
    {
      "epoch": 4.084000880654394,
      "grad_norm": 0.22511455416679382,
      "learning_rate": 5.910272088647067e-05,
      "loss": 1.4133,
      "step": 82550
    },
    {
      "epoch": 4.084542821819906,
      "grad_norm": 0.17213839292526245,
      "learning_rate": 5.9086948407210727e-05,
      "loss": 1.4103,
      "step": 82560
    },
    {
      "epoch": 4.08470540416956,
      "eval_loss": 2.431107521057129,
      "eval_runtime": 21.9786,
      "eval_samples_per_second": 227.494,
      "eval_steps_per_second": 1.228,
      "step": 82563
    },
    {
      "epoch": 4.085084762985418,
      "grad_norm": 0.26532599329948425,
      "learning_rate": 5.907117542167737e-05,
      "loss": 1.4042,
      "step": 82570
    },
    {
      "epoch": 4.085626704150931,
      "grad_norm": 0.16096952557563782,
      "learning_rate": 5.905540193182446e-05,
      "loss": 1.4038,
      "step": 82580
    },
    {
      "epoch": 4.086168645316443,
      "grad_norm": 0.1794336587190628,
      "learning_rate": 5.903962793960599e-05,
      "loss": 1.4073,
      "step": 82590
    },
    {
      "epoch": 4.086710586481955,
      "grad_norm": 0.17377042770385742,
      "learning_rate": 5.902385344697594e-05,
      "loss": 1.3989,
      "step": 82600
    },
    {
      "epoch": 4.087252527647467,
      "grad_norm": 0.16201172769069672,
      "learning_rate": 5.9008078455888394e-05,
      "loss": 1.3959,
      "step": 82610
    },
    {
      "epoch": 4.087794468812979,
      "grad_norm": 0.27610713243484497,
      "learning_rate": 5.899230296829748e-05,
      "loss": 1.3948,
      "step": 82620
    },
    {
      "epoch": 4.088336409978492,
      "grad_norm": 0.25443559885025024,
      "learning_rate": 5.897652698615741e-05,
      "loss": 1.396,
      "step": 82630
    },
    {
      "epoch": 4.088878351144004,
      "grad_norm": 0.24116799235343933,
      "learning_rate": 5.896075051142246e-05,
      "loss": 1.404,
      "step": 82640
    },
    {
      "epoch": 4.089420292309516,
      "grad_norm": 0.20483553409576416,
      "learning_rate": 5.894497354604692e-05,
      "loss": 1.405,
      "step": 82650
    },
    {
      "epoch": 4.089420292309516,
      "eval_loss": 2.4280037879943848,
      "eval_runtime": 21.9762,
      "eval_samples_per_second": 227.519,
      "eval_steps_per_second": 1.229,
      "step": 82650
    },
    {
      "epoch": 4.0899622334750285,
      "grad_norm": 0.26821398735046387,
      "learning_rate": 5.892919609198517e-05,
      "loss": 1.4138,
      "step": 82660
    },
    {
      "epoch": 4.09050417464054,
      "grad_norm": 0.1844187080860138,
      "learning_rate": 5.891341815119168e-05,
      "loss": 1.4063,
      "step": 82670
    },
    {
      "epoch": 4.091046115806053,
      "grad_norm": 0.3293144702911377,
      "learning_rate": 5.8897639725620956e-05,
      "loss": 1.4092,
      "step": 82680
    },
    {
      "epoch": 4.091588056971565,
      "grad_norm": 0.20657339692115784,
      "learning_rate": 5.888186081722752e-05,
      "loss": 1.3962,
      "step": 82690
    },
    {
      "epoch": 4.092129998137077,
      "grad_norm": 0.16567468643188477,
      "learning_rate": 5.8866081427966036e-05,
      "loss": 1.4108,
      "step": 82700
    },
    {
      "epoch": 4.09267193930259,
      "grad_norm": 0.18625999987125397,
      "learning_rate": 5.885030155979116e-05,
      "loss": 1.4004,
      "step": 82710
    },
    {
      "epoch": 4.0932138804681015,
      "grad_norm": 0.16223831474781036,
      "learning_rate": 5.8834521214657635e-05,
      "loss": 1.4017,
      "step": 82720
    },
    {
      "epoch": 4.093755821633614,
      "grad_norm": 0.1961243897676468,
      "learning_rate": 5.881874039452029e-05,
      "loss": 1.4039,
      "step": 82730
    },
    {
      "epoch": 4.094135180449473,
      "eval_loss": 2.4336957931518555,
      "eval_runtime": 21.6769,
      "eval_samples_per_second": 230.66,
      "eval_steps_per_second": 1.246,
      "step": 82737
    },
    {
      "epoch": 4.094297762799126,
      "grad_norm": 0.20036591589450836,
      "learning_rate": 5.880295910133394e-05,
      "loss": 1.4055,
      "step": 82740
    },
    {
      "epoch": 4.094839703964638,
      "grad_norm": 0.20665618777275085,
      "learning_rate": 5.8787177337053555e-05,
      "loss": 1.4043,
      "step": 82750
    },
    {
      "epoch": 4.095381645130151,
      "grad_norm": 0.18086576461791992,
      "learning_rate": 5.8771395103634065e-05,
      "loss": 1.41,
      "step": 82760
    },
    {
      "epoch": 4.095923586295663,
      "grad_norm": 0.15253938734531403,
      "learning_rate": 5.8755612403030524e-05,
      "loss": 1.4057,
      "step": 82770
    },
    {
      "epoch": 4.096465527461175,
      "grad_norm": 0.17018210887908936,
      "learning_rate": 5.873982923719804e-05,
      "loss": 1.4059,
      "step": 82780
    },
    {
      "epoch": 4.097007468626687,
      "grad_norm": 0.1573159396648407,
      "learning_rate": 5.872404560809173e-05,
      "loss": 1.4166,
      "step": 82790
    },
    {
      "epoch": 4.097549409792199,
      "grad_norm": 0.2046526074409485,
      "learning_rate": 5.870826151766683e-05,
      "loss": 1.4067,
      "step": 82800
    },
    {
      "epoch": 4.098091350957712,
      "grad_norm": 0.1698278784751892,
      "learning_rate": 5.869247696787857e-05,
      "loss": 1.4116,
      "step": 82810
    },
    {
      "epoch": 4.098633292123224,
      "grad_norm": 0.24143236875534058,
      "learning_rate": 5.867669196068231e-05,
      "loss": 1.3992,
      "step": 82820
    },
    {
      "epoch": 4.098850068589429,
      "eval_loss": 2.4264564514160156,
      "eval_runtime": 21.9736,
      "eval_samples_per_second": 227.546,
      "eval_steps_per_second": 1.229,
      "step": 82824
    },
    {
      "epoch": 4.099175233288736,
      "grad_norm": 0.16382378339767456,
      "learning_rate": 5.866090649803342e-05,
      "loss": 1.4064,
      "step": 82830
    },
    {
      "epoch": 4.099717174454248,
      "grad_norm": 0.18778187036514282,
      "learning_rate": 5.864512058188733e-05,
      "loss": 1.398,
      "step": 82840
    },
    {
      "epoch": 4.10025911561976,
      "grad_norm": 0.1958286166191101,
      "learning_rate": 5.862933421419952e-05,
      "loss": 1.4076,
      "step": 82850
    },
    {
      "epoch": 4.100801056785273,
      "grad_norm": 0.18923796713352203,
      "learning_rate": 5.861354739692553e-05,
      "loss": 1.4033,
      "step": 82860
    },
    {
      "epoch": 4.101342997950785,
      "grad_norm": 0.155537948012352,
      "learning_rate": 5.859776013202098e-05,
      "loss": 1.4019,
      "step": 82870
    },
    {
      "epoch": 4.101884939116297,
      "grad_norm": 0.2815153896808624,
      "learning_rate": 5.858197242144155e-05,
      "loss": 1.4076,
      "step": 82880
    },
    {
      "epoch": 4.10242688028181,
      "grad_norm": 0.20242029428482056,
      "learning_rate": 5.856618426714291e-05,
      "loss": 1.3973,
      "step": 82890
    },
    {
      "epoch": 4.1029688214473214,
      "grad_norm": 0.18118534982204437,
      "learning_rate": 5.855039567108084e-05,
      "loss": 1.4034,
      "step": 82900
    },
    {
      "epoch": 4.103510762612834,
      "grad_norm": 0.18179315328598022,
      "learning_rate": 5.853460663521117e-05,
      "loss": 1.3943,
      "step": 82910
    },
    {
      "epoch": 4.103564956729385,
      "eval_loss": 2.43251633644104,
      "eval_runtime": 21.9701,
      "eval_samples_per_second": 227.582,
      "eval_steps_per_second": 1.229,
      "step": 82911
    },
    {
      "epoch": 4.104052703778346,
      "grad_norm": 0.3180215358734131,
      "learning_rate": 5.851881716148979e-05,
      "loss": 1.3989,
      "step": 82920
    },
    {
      "epoch": 4.104594644943858,
      "grad_norm": 0.22346888482570648,
      "learning_rate": 5.850302725187261e-05,
      "loss": 1.4041,
      "step": 82930
    },
    {
      "epoch": 4.105136586109371,
      "grad_norm": 0.1704285740852356,
      "learning_rate": 5.8487236908315635e-05,
      "loss": 1.4034,
      "step": 82940
    },
    {
      "epoch": 4.105678527274883,
      "grad_norm": 0.1817091554403305,
      "learning_rate": 5.8471446132774864e-05,
      "loss": 1.402,
      "step": 82950
    },
    {
      "epoch": 4.106220468440395,
      "grad_norm": 0.16778716444969177,
      "learning_rate": 5.8455654927206436e-05,
      "loss": 1.399,
      "step": 82960
    },
    {
      "epoch": 4.106762409605907,
      "grad_norm": 0.20462718605995178,
      "learning_rate": 5.8439863293566476e-05,
      "loss": 1.4,
      "step": 82970
    },
    {
      "epoch": 4.107304350771419,
      "grad_norm": 0.16164131462574005,
      "learning_rate": 5.842407123381118e-05,
      "loss": 1.4022,
      "step": 82980
    },
    {
      "epoch": 4.107846291936932,
      "grad_norm": 0.17389550805091858,
      "learning_rate": 5.8408278749896816e-05,
      "loss": 1.4129,
      "step": 82990
    },
    {
      "epoch": 4.108279844869341,
      "eval_loss": 2.4385242462158203,
      "eval_runtime": 21.9887,
      "eval_samples_per_second": 227.39,
      "eval_steps_per_second": 1.228,
      "step": 82998
    },
    {
      "epoch": 4.108388233102444,
      "grad_norm": 0.3119494616985321,
      "learning_rate": 5.8392485843779676e-05,
      "loss": 1.41,
      "step": 83000
    },
    {
      "epoch": 4.108930174267956,
      "grad_norm": 0.21673525869846344,
      "learning_rate": 5.83766925174161e-05,
      "loss": 1.4092,
      "step": 83010
    },
    {
      "epoch": 4.109472115433468,
      "grad_norm": 0.17000888288021088,
      "learning_rate": 5.836089877276254e-05,
      "loss": 1.4118,
      "step": 83020
    },
    {
      "epoch": 4.11001405659898,
      "grad_norm": 0.1890566647052765,
      "learning_rate": 5.83451046117754e-05,
      "loss": 1.4017,
      "step": 83030
    },
    {
      "epoch": 4.110555997764493,
      "grad_norm": 0.21135641634464264,
      "learning_rate": 5.832931003641127e-05,
      "loss": 1.393,
      "step": 83040
    },
    {
      "epoch": 4.111097938930005,
      "grad_norm": 0.15873472392559052,
      "learning_rate": 5.8313515048626634e-05,
      "loss": 1.4016,
      "step": 83050
    },
    {
      "epoch": 4.111639880095517,
      "grad_norm": 0.24823595583438873,
      "learning_rate": 5.8297719650378136e-05,
      "loss": 1.41,
      "step": 83060
    },
    {
      "epoch": 4.1121818212610295,
      "grad_norm": 0.18484194576740265,
      "learning_rate": 5.828192384362245e-05,
      "loss": 1.4021,
      "step": 83070
    },
    {
      "epoch": 4.112723762426541,
      "grad_norm": 0.26047736406326294,
      "learning_rate": 5.826612763031632e-05,
      "loss": 1.4074,
      "step": 83080
    },
    {
      "epoch": 4.112994733009297,
      "eval_loss": 2.4378349781036377,
      "eval_runtime": 21.9749,
      "eval_samples_per_second": 227.532,
      "eval_steps_per_second": 1.229,
      "step": 83085
    },
    {
      "epoch": 4.113265703592054,
      "grad_norm": 0.38496696949005127,
      "learning_rate": 5.825033101241644e-05,
      "loss": 1.4136,
      "step": 83090
    },
    {
      "epoch": 4.113807644757566,
      "grad_norm": 0.19482018053531647,
      "learning_rate": 5.823453399187967e-05,
      "loss": 1.4096,
      "step": 83100
    },
    {
      "epoch": 4.114349585923078,
      "grad_norm": 0.17476175725460052,
      "learning_rate": 5.821873657066288e-05,
      "loss": 1.4118,
      "step": 83110
    },
    {
      "epoch": 4.114891527088591,
      "grad_norm": 0.19466517865657806,
      "learning_rate": 5.820293875072298e-05,
      "loss": 1.4088,
      "step": 83120
    },
    {
      "epoch": 4.1154334682541025,
      "grad_norm": 0.20809796452522278,
      "learning_rate": 5.818714053401695e-05,
      "loss": 1.4084,
      "step": 83130
    },
    {
      "epoch": 4.115975409419615,
      "grad_norm": 0.16254140436649323,
      "learning_rate": 5.8171341922501755e-05,
      "loss": 1.4097,
      "step": 83140
    },
    {
      "epoch": 4.116517350585127,
      "grad_norm": 0.15851789712905884,
      "learning_rate": 5.8155542918134496e-05,
      "loss": 1.4123,
      "step": 83150
    },
    {
      "epoch": 4.117059291750639,
      "grad_norm": 0.29275140166282654,
      "learning_rate": 5.8139743522872306e-05,
      "loss": 1.4053,
      "step": 83160
    },
    {
      "epoch": 4.117601232916152,
      "grad_norm": 0.2825930714607239,
      "learning_rate": 5.812394373867229e-05,
      "loss": 1.4093,
      "step": 83170
    },
    {
      "epoch": 4.117709621149254,
      "eval_loss": 2.4347915649414062,
      "eval_runtime": 21.9741,
      "eval_samples_per_second": 227.54,
      "eval_steps_per_second": 1.229,
      "step": 83172
    },
    {
      "epoch": 4.118143174081664,
      "grad_norm": 0.24042131006717682,
      "learning_rate": 5.81081435674917e-05,
      "loss": 1.4135,
      "step": 83180
    },
    {
      "epoch": 4.118685115247176,
      "grad_norm": 0.22210003435611725,
      "learning_rate": 5.809234301128779e-05,
      "loss": 1.4164,
      "step": 83190
    },
    {
      "epoch": 4.119227056412688,
      "grad_norm": 0.15581659972667694,
      "learning_rate": 5.807654207201784e-05,
      "loss": 1.3988,
      "step": 83200
    },
    {
      "epoch": 4.1197689975782,
      "grad_norm": 0.19380082190036774,
      "learning_rate": 5.8060740751639217e-05,
      "loss": 1.4111,
      "step": 83210
    },
    {
      "epoch": 4.120310938743713,
      "grad_norm": 0.1942562609910965,
      "learning_rate": 5.8044939052109315e-05,
      "loss": 1.4082,
      "step": 83220
    },
    {
      "epoch": 4.120852879909225,
      "grad_norm": 0.25676843523979187,
      "learning_rate": 5.802913697538559e-05,
      "loss": 1.4086,
      "step": 83230
    },
    {
      "epoch": 4.121394821074737,
      "grad_norm": 0.21797262132167816,
      "learning_rate": 5.8013334523425536e-05,
      "loss": 1.408,
      "step": 83240
    },
    {
      "epoch": 4.121936762240249,
      "grad_norm": 0.18944242596626282,
      "learning_rate": 5.799753169818666e-05,
      "loss": 1.4046,
      "step": 83250
    },
    {
      "epoch": 4.1224245092892104,
      "eval_loss": 2.430614709854126,
      "eval_runtime": 21.9688,
      "eval_samples_per_second": 227.596,
      "eval_steps_per_second": 1.229,
      "step": 83259
    },
    {
      "epoch": 4.122478703405761,
      "grad_norm": 0.24879737198352814,
      "learning_rate": 5.798172850162658e-05,
      "loss": 1.4057,
      "step": 83260
    },
    {
      "epoch": 4.123020644571274,
      "grad_norm": 0.16706986725330353,
      "learning_rate": 5.796592493570292e-05,
      "loss": 1.4101,
      "step": 83270
    },
    {
      "epoch": 4.123562585736786,
      "grad_norm": 0.16859695315361023,
      "learning_rate": 5.795012100237337e-05,
      "loss": 1.3993,
      "step": 83280
    },
    {
      "epoch": 4.124104526902298,
      "grad_norm": 0.17874988913536072,
      "learning_rate": 5.7934316703595625e-05,
      "loss": 1.4068,
      "step": 83290
    },
    {
      "epoch": 4.1246464680678105,
      "grad_norm": 0.2080361247062683,
      "learning_rate": 5.791851204132745e-05,
      "loss": 1.4041,
      "step": 83300
    },
    {
      "epoch": 4.125188409233322,
      "grad_norm": 0.20870813727378845,
      "learning_rate": 5.79027070175267e-05,
      "loss": 1.4055,
      "step": 83310
    },
    {
      "epoch": 4.125730350398835,
      "grad_norm": 0.17612333595752716,
      "learning_rate": 5.7886901634151205e-05,
      "loss": 1.4071,
      "step": 83320
    },
    {
      "epoch": 4.126272291564347,
      "grad_norm": 0.18745100498199463,
      "learning_rate": 5.787109589315887e-05,
      "loss": 1.4132,
      "step": 83330
    },
    {
      "epoch": 4.126814232729859,
      "grad_norm": 0.2577305734157562,
      "learning_rate": 5.7855289796507627e-05,
      "loss": 1.4039,
      "step": 83340
    },
    {
      "epoch": 4.127139397429167,
      "eval_loss": 2.428053855895996,
      "eval_runtime": 21.9759,
      "eval_samples_per_second": 227.521,
      "eval_steps_per_second": 1.229,
      "step": 83346
    },
    {
      "epoch": 4.127356173895372,
      "grad_norm": 0.1890375018119812,
      "learning_rate": 5.783948334615548e-05,
      "loss": 1.3949,
      "step": 83350
    },
    {
      "epoch": 4.1278981150608836,
      "grad_norm": 0.34618306159973145,
      "learning_rate": 5.782367654406047e-05,
      "loss": 1.4095,
      "step": 83360
    },
    {
      "epoch": 4.128440056226396,
      "grad_norm": 0.19756010174751282,
      "learning_rate": 5.780786939218068e-05,
      "loss": 1.4154,
      "step": 83370
    },
    {
      "epoch": 4.128981997391908,
      "grad_norm": 0.2689013183116913,
      "learning_rate": 5.77920618924742e-05,
      "loss": 1.4067,
      "step": 83380
    },
    {
      "epoch": 4.12952393855742,
      "grad_norm": 0.17390170693397522,
      "learning_rate": 5.777625404689924e-05,
      "loss": 1.4065,
      "step": 83390
    },
    {
      "epoch": 4.130065879722933,
      "grad_norm": 0.2956944704055786,
      "learning_rate": 5.7760445857413946e-05,
      "loss": 1.4047,
      "step": 83400
    },
    {
      "epoch": 4.130607820888445,
      "grad_norm": 0.2633947432041168,
      "learning_rate": 5.774463732597662e-05,
      "loss": 1.4067,
      "step": 83410
    },
    {
      "epoch": 4.131149762053957,
      "grad_norm": 0.17089660465717316,
      "learning_rate": 5.772882845454556e-05,
      "loss": 1.4171,
      "step": 83420
    },
    {
      "epoch": 4.131691703219469,
      "grad_norm": 0.18730990588665009,
      "learning_rate": 5.771301924507905e-05,
      "loss": 1.4067,
      "step": 83430
    },
    {
      "epoch": 4.131854285569123,
      "eval_loss": 2.4326930046081543,
      "eval_runtime": 21.9721,
      "eval_samples_per_second": 227.562,
      "eval_steps_per_second": 1.229,
      "step": 83433
    },
    {
      "epoch": 4.132233644384981,
      "grad_norm": 0.16217251121997833,
      "learning_rate": 5.7697209699535514e-05,
      "loss": 1.4039,
      "step": 83440
    },
    {
      "epoch": 4.132775585550494,
      "grad_norm": 0.16589990258216858,
      "learning_rate": 5.768139981987334e-05,
      "loss": 1.3967,
      "step": 83450
    },
    {
      "epoch": 4.133317526716006,
      "grad_norm": 0.1889846920967102,
      "learning_rate": 5.7665589608051e-05,
      "loss": 1.4068,
      "step": 83460
    },
    {
      "epoch": 4.133859467881518,
      "grad_norm": 0.18602542579174042,
      "learning_rate": 5.7649779066026985e-05,
      "loss": 1.4024,
      "step": 83470
    },
    {
      "epoch": 4.1344014090470305,
      "grad_norm": 0.24643942713737488,
      "learning_rate": 5.7633968195759855e-05,
      "loss": 1.4069,
      "step": 83480
    },
    {
      "epoch": 4.134943350212542,
      "grad_norm": 0.18482206761837006,
      "learning_rate": 5.7618156999208186e-05,
      "loss": 1.4186,
      "step": 83490
    },
    {
      "epoch": 4.135485291378055,
      "grad_norm": 0.1943228840827942,
      "learning_rate": 5.7602345478330577e-05,
      "loss": 1.4047,
      "step": 83500
    },
    {
      "epoch": 4.136027232543567,
      "grad_norm": 0.2460983842611313,
      "learning_rate": 5.7586533635085705e-05,
      "loss": 1.4117,
      "step": 83510
    },
    {
      "epoch": 4.136569173709079,
      "grad_norm": 0.22592556476593018,
      "learning_rate": 5.75707214714323e-05,
      "loss": 1.407,
      "step": 83520
    },
    {
      "epoch": 4.136569173709079,
      "eval_loss": 2.431594133377075,
      "eval_runtime": 21.9667,
      "eval_samples_per_second": 227.617,
      "eval_steps_per_second": 1.229,
      "step": 83520
    },
    {
      "epoch": 4.137111114874592,
      "grad_norm": 0.22698943316936493,
      "learning_rate": 5.7554908989329084e-05,
      "loss": 1.409,
      "step": 83530
    },
    {
      "epoch": 4.1376530560401035,
      "grad_norm": 0.1876407265663147,
      "learning_rate": 5.7539096190734806e-05,
      "loss": 1.4023,
      "step": 83540
    },
    {
      "epoch": 4.138194997205616,
      "grad_norm": 0.17315153777599335,
      "learning_rate": 5.752328307760833e-05,
      "loss": 1.3965,
      "step": 83550
    },
    {
      "epoch": 4.138736938371128,
      "grad_norm": 0.1867133527994156,
      "learning_rate": 5.7507469651908486e-05,
      "loss": 1.414,
      "step": 83560
    },
    {
      "epoch": 4.13927887953664,
      "grad_norm": 0.2372438907623291,
      "learning_rate": 5.74916559155942e-05,
      "loss": 1.41,
      "step": 83570
    },
    {
      "epoch": 4.139820820702153,
      "grad_norm": 0.16649603843688965,
      "learning_rate": 5.7475841870624416e-05,
      "loss": 1.3959,
      "step": 83580
    },
    {
      "epoch": 4.140362761867665,
      "grad_norm": 0.2065044343471527,
      "learning_rate": 5.746002751895807e-05,
      "loss": 1.4078,
      "step": 83590
    },
    {
      "epoch": 4.140904703033177,
      "grad_norm": 0.20945148169994354,
      "learning_rate": 5.744421286255418e-05,
      "loss": 1.3964,
      "step": 83600
    },
    {
      "epoch": 4.141284061849036,
      "eval_loss": 2.4306061267852783,
      "eval_runtime": 21.9679,
      "eval_samples_per_second": 227.605,
      "eval_steps_per_second": 1.229,
      "step": 83607
    },
    {
      "epoch": 4.141446644198689,
      "grad_norm": 0.21159766614437103,
      "learning_rate": 5.7428397903371845e-05,
      "loss": 1.4085,
      "step": 83610
    },
    {
      "epoch": 4.141988585364201,
      "grad_norm": 0.1735406070947647,
      "learning_rate": 5.741258264337009e-05,
      "loss": 1.4062,
      "step": 83620
    },
    {
      "epoch": 4.142530526529714,
      "grad_norm": 0.21971061825752258,
      "learning_rate": 5.739676708450809e-05,
      "loss": 1.4088,
      "step": 83630
    },
    {
      "epoch": 4.143072467695226,
      "grad_norm": 0.1740388572216034,
      "learning_rate": 5.7380951228744985e-05,
      "loss": 1.4059,
      "step": 83640
    },
    {
      "epoch": 4.1436144088607385,
      "grad_norm": 0.21105538308620453,
      "learning_rate": 5.736513507803999e-05,
      "loss": 1.4035,
      "step": 83650
    },
    {
      "epoch": 4.14415635002625,
      "grad_norm": 0.21012961864471436,
      "learning_rate": 5.7349318634352325e-05,
      "loss": 1.4016,
      "step": 83660
    },
    {
      "epoch": 4.144698291191762,
      "grad_norm": 0.21430379152297974,
      "learning_rate": 5.7333501899641254e-05,
      "loss": 1.399,
      "step": 83670
    },
    {
      "epoch": 4.145240232357275,
      "grad_norm": 0.16836793720722198,
      "learning_rate": 5.731768487586611e-05,
      "loss": 1.3952,
      "step": 83680
    },
    {
      "epoch": 4.145782173522787,
      "grad_norm": 0.22241011261940002,
      "learning_rate": 5.730186756498622e-05,
      "loss": 1.4089,
      "step": 83690
    },
    {
      "epoch": 4.145998949988992,
      "eval_loss": 2.433333396911621,
      "eval_runtime": 21.9717,
      "eval_samples_per_second": 227.566,
      "eval_steps_per_second": 1.229,
      "step": 83694
    },
    {
      "epoch": 4.146324114688299,
      "grad_norm": 0.34341859817504883,
      "learning_rate": 5.728604996896096e-05,
      "loss": 1.4097,
      "step": 83700
    },
    {
      "epoch": 4.1468660558538115,
      "grad_norm": 0.4072984755039215,
      "learning_rate": 5.7270232089749775e-05,
      "loss": 1.398,
      "step": 83710
    },
    {
      "epoch": 4.147407997019323,
      "grad_norm": 0.21958887577056885,
      "learning_rate": 5.725441392931209e-05,
      "loss": 1.4012,
      "step": 83720
    },
    {
      "epoch": 4.147949938184836,
      "grad_norm": 0.18960371613502502,
      "learning_rate": 5.7238595489607396e-05,
      "loss": 1.4089,
      "step": 83730
    },
    {
      "epoch": 4.148491879350348,
      "grad_norm": 0.16692198812961578,
      "learning_rate": 5.722277677259521e-05,
      "loss": 1.4043,
      "step": 83740
    },
    {
      "epoch": 4.14903382051586,
      "grad_norm": 0.1643580198287964,
      "learning_rate": 5.720695778023508e-05,
      "loss": 1.395,
      "step": 83750
    },
    {
      "epoch": 4.149575761681373,
      "grad_norm": 0.23333659768104553,
      "learning_rate": 5.7191138514486605e-05,
      "loss": 1.4017,
      "step": 83760
    },
    {
      "epoch": 4.1501177028468845,
      "grad_norm": 0.19941531121730804,
      "learning_rate": 5.7175318977309424e-05,
      "loss": 1.4022,
      "step": 83770
    },
    {
      "epoch": 4.150659644012397,
      "grad_norm": 0.19911114871501923,
      "learning_rate": 5.7159499170663144e-05,
      "loss": 1.4112,
      "step": 83780
    },
    {
      "epoch": 4.150713838128948,
      "eval_loss": 2.4322268962860107,
      "eval_runtime": 21.9745,
      "eval_samples_per_second": 227.537,
      "eval_steps_per_second": 1.229,
      "step": 83781
    },
    {
      "epoch": 4.151201585177909,
      "grad_norm": 0.18389646708965302,
      "learning_rate": 5.7143679096507494e-05,
      "loss": 1.402,
      "step": 83790
    },
    {
      "epoch": 4.151743526343421,
      "grad_norm": 0.1758451610803604,
      "learning_rate": 5.712785875680218e-05,
      "loss": 1.4039,
      "step": 83800
    },
    {
      "epoch": 4.152285467508934,
      "grad_norm": 0.16958458721637726,
      "learning_rate": 5.711203815350696e-05,
      "loss": 1.4069,
      "step": 83810
    },
    {
      "epoch": 4.152827408674446,
      "grad_norm": 0.16103295981884003,
      "learning_rate": 5.709621728858164e-05,
      "loss": 1.404,
      "step": 83820
    },
    {
      "epoch": 4.153369349839958,
      "grad_norm": 0.17550979554653168,
      "learning_rate": 5.7080396163986004e-05,
      "loss": 1.402,
      "step": 83830
    },
    {
      "epoch": 4.15391129100547,
      "grad_norm": 0.2050134539604187,
      "learning_rate": 5.706457478167992e-05,
      "loss": 1.4078,
      "step": 83840
    },
    {
      "epoch": 4.154453232170982,
      "grad_norm": 0.1583109349012375,
      "learning_rate": 5.704875314362331e-05,
      "loss": 1.414,
      "step": 83850
    },
    {
      "epoch": 4.154995173336495,
      "grad_norm": 0.20555706322193146,
      "learning_rate": 5.703293125177602e-05,
      "loss": 1.4025,
      "step": 83860
    },
    {
      "epoch": 4.155428726268904,
      "eval_loss": 2.4321482181549072,
      "eval_runtime": 21.9695,
      "eval_samples_per_second": 227.588,
      "eval_steps_per_second": 1.229,
      "step": 83868
    },
    {
      "epoch": 4.155537114502007,
      "grad_norm": 0.2936631739139557,
      "learning_rate": 5.701710910809805e-05,
      "loss": 1.4003,
      "step": 83870
    },
    {
      "epoch": 4.156079055667519,
      "grad_norm": 0.24286337196826935,
      "learning_rate": 5.700128671454935e-05,
      "loss": 1.4065,
      "step": 83880
    },
    {
      "epoch": 4.156620996833031,
      "grad_norm": 0.3643859624862671,
      "learning_rate": 5.6985464073089944e-05,
      "loss": 1.3969,
      "step": 83890
    },
    {
      "epoch": 4.157162937998543,
      "grad_norm": 0.3168624937534332,
      "learning_rate": 5.696964118567988e-05,
      "loss": 1.3999,
      "step": 83900
    },
    {
      "epoch": 4.157704879164056,
      "grad_norm": 0.24165990948677063,
      "learning_rate": 5.6953818054279206e-05,
      "loss": 1.3952,
      "step": 83910
    },
    {
      "epoch": 4.158246820329568,
      "grad_norm": 0.16725140810012817,
      "learning_rate": 5.693799468084804e-05,
      "loss": 1.4162,
      "step": 83920
    },
    {
      "epoch": 4.15878876149508,
      "grad_norm": 0.2385065108537674,
      "learning_rate": 5.6922171067346495e-05,
      "loss": 1.3966,
      "step": 83930
    },
    {
      "epoch": 4.159330702660593,
      "grad_norm": 0.23973079025745392,
      "learning_rate": 5.690634721573475e-05,
      "loss": 1.409,
      "step": 83940
    },
    {
      "epoch": 4.159872643826104,
      "grad_norm": 0.18828116357326508,
      "learning_rate": 5.6890523127972993e-05,
      "loss": 1.3998,
      "step": 83950
    },
    {
      "epoch": 4.16014361440886,
      "eval_loss": 2.4352917671203613,
      "eval_runtime": 21.9698,
      "eval_samples_per_second": 227.585,
      "eval_steps_per_second": 1.229,
      "step": 83955
    },
    {
      "epoch": 4.160414584991617,
      "grad_norm": 0.1964593529701233,
      "learning_rate": 5.687469880602143e-05,
      "loss": 1.4101,
      "step": 83960
    },
    {
      "epoch": 4.160956526157129,
      "grad_norm": 0.2196066826581955,
      "learning_rate": 5.685887425184033e-05,
      "loss": 1.4071,
      "step": 83970
    },
    {
      "epoch": 4.161498467322641,
      "grad_norm": 0.16051217913627625,
      "learning_rate": 5.684304946738995e-05,
      "loss": 1.411,
      "step": 83980
    },
    {
      "epoch": 4.162040408488154,
      "grad_norm": 0.23995092511177063,
      "learning_rate": 5.68272244546306e-05,
      "loss": 1.4054,
      "step": 83990
    },
    {
      "epoch": 4.162582349653666,
      "grad_norm": 0.35630932450294495,
      "learning_rate": 5.6811399215522634e-05,
      "loss": 1.4047,
      "step": 84000
    },
    {
      "epoch": 4.163124290819178,
      "grad_norm": 0.26559507846832275,
      "learning_rate": 5.6795573752026386e-05,
      "loss": 1.4162,
      "step": 84010
    },
    {
      "epoch": 4.16366623198469,
      "grad_norm": 0.28465500473976135,
      "learning_rate": 5.6779748066102254e-05,
      "loss": 1.4168,
      "step": 84020
    },
    {
      "epoch": 4.164208173150202,
      "grad_norm": 0.3914095163345337,
      "learning_rate": 5.676392215971066e-05,
      "loss": 1.4054,
      "step": 84030
    },
    {
      "epoch": 4.164750114315715,
      "grad_norm": 0.2624354958534241,
      "learning_rate": 5.6748096034812047e-05,
      "loss": 1.4156,
      "step": 84040
    },
    {
      "epoch": 4.164858502548817,
      "eval_loss": 2.431173324584961,
      "eval_runtime": 21.9748,
      "eval_samples_per_second": 227.534,
      "eval_steps_per_second": 1.229,
      "step": 84042
    },
    {
      "epoch": 4.165292055481227,
      "grad_norm": 0.25440993905067444,
      "learning_rate": 5.6732269693366894e-05,
      "loss": 1.4029,
      "step": 84050
    },
    {
      "epoch": 4.1658339966467395,
      "grad_norm": 0.1714615821838379,
      "learning_rate": 5.6716443137335695e-05,
      "loss": 1.3996,
      "step": 84060
    },
    {
      "epoch": 4.166375937812251,
      "grad_norm": 0.3408913016319275,
      "learning_rate": 5.670061636867896e-05,
      "loss": 1.4094,
      "step": 84070
    },
    {
      "epoch": 4.166917878977763,
      "grad_norm": 0.25224050879478455,
      "learning_rate": 5.6684789389357254e-05,
      "loss": 1.4019,
      "step": 84080
    },
    {
      "epoch": 4.167459820143276,
      "grad_norm": 0.18869130313396454,
      "learning_rate": 5.666896220133116e-05,
      "loss": 1.398,
      "step": 84090
    },
    {
      "epoch": 4.168001761308788,
      "grad_norm": 0.23559992015361786,
      "learning_rate": 5.665313480656127e-05,
      "loss": 1.3896,
      "step": 84100
    },
    {
      "epoch": 4.1685437024743,
      "grad_norm": 0.33458632230758667,
      "learning_rate": 5.663730720700822e-05,
      "loss": 1.4115,
      "step": 84110
    },
    {
      "epoch": 4.1690856436398125,
      "grad_norm": 0.24740028381347656,
      "learning_rate": 5.662147940463265e-05,
      "loss": 1.4121,
      "step": 84120
    },
    {
      "epoch": 4.1695733906887735,
      "eval_loss": 2.432617664337158,
      "eval_runtime": 21.9739,
      "eval_samples_per_second": 227.543,
      "eval_steps_per_second": 1.229,
      "step": 84129
    },
    {
      "epoch": 4.169627584805324,
      "grad_norm": 0.15262192487716675,
      "learning_rate": 5.6605651401395265e-05,
      "loss": 1.3928,
      "step": 84130
    },
    {
      "epoch": 4.170169525970837,
      "grad_norm": 0.16818031668663025,
      "learning_rate": 5.658982319925675e-05,
      "loss": 1.4084,
      "step": 84140
    },
    {
      "epoch": 4.170711467136349,
      "grad_norm": 0.16727863252162933,
      "learning_rate": 5.6573994800177844e-05,
      "loss": 1.3928,
      "step": 84150
    },
    {
      "epoch": 4.171253408301861,
      "grad_norm": 0.21401916444301605,
      "learning_rate": 5.655816620611929e-05,
      "loss": 1.3994,
      "step": 84160
    },
    {
      "epoch": 4.171795349467374,
      "grad_norm": 0.21131601929664612,
      "learning_rate": 5.6542337419041866e-05,
      "loss": 1.4001,
      "step": 84170
    },
    {
      "epoch": 4.1723372906328855,
      "grad_norm": 0.17307646572589874,
      "learning_rate": 5.652650844090639e-05,
      "loss": 1.3984,
      "step": 84180
    },
    {
      "epoch": 4.172879231798398,
      "grad_norm": 0.16799761354923248,
      "learning_rate": 5.651067927367367e-05,
      "loss": 1.405,
      "step": 84190
    },
    {
      "epoch": 4.17342117296391,
      "grad_norm": 0.20099224150180817,
      "learning_rate": 5.649484991930456e-05,
      "loss": 1.4075,
      "step": 84200
    },
    {
      "epoch": 4.173963114129422,
      "grad_norm": 0.28969597816467285,
      "learning_rate": 5.647902037975994e-05,
      "loss": 1.4005,
      "step": 84210
    },
    {
      "epoch": 4.17428827882873,
      "eval_loss": 2.434582471847534,
      "eval_runtime": 21.9717,
      "eval_samples_per_second": 227.565,
      "eval_steps_per_second": 1.229,
      "step": 84216
    },
    {
      "epoch": 4.174505055294935,
      "grad_norm": 0.272678017616272,
      "learning_rate": 5.646319065700068e-05,
      "loss": 1.4166,
      "step": 84220
    },
    {
      "epoch": 4.175046996460447,
      "grad_norm": 0.1885080486536026,
      "learning_rate": 5.644736075298772e-05,
      "loss": 1.4171,
      "step": 84230
    },
    {
      "epoch": 4.175588937625959,
      "grad_norm": 0.1917075365781784,
      "learning_rate": 5.643153066968201e-05,
      "loss": 1.3912,
      "step": 84240
    },
    {
      "epoch": 4.176130878791471,
      "grad_norm": 0.1733640879392624,
      "learning_rate": 5.641570040904448e-05,
      "loss": 1.4086,
      "step": 84250
    },
    {
      "epoch": 4.176672819956983,
      "grad_norm": 0.21975889801979065,
      "learning_rate": 5.6399869973036136e-05,
      "loss": 1.4118,
      "step": 84260
    },
    {
      "epoch": 4.177214761122496,
      "grad_norm": 0.1629042774438858,
      "learning_rate": 5.638403936361797e-05,
      "loss": 1.4043,
      "step": 84270
    },
    {
      "epoch": 4.177756702288008,
      "grad_norm": 0.16598139703273773,
      "learning_rate": 5.6368208582751026e-05,
      "loss": 1.413,
      "step": 84280
    },
    {
      "epoch": 4.17829864345352,
      "grad_norm": 0.16502542793750763,
      "learning_rate": 5.6352377632396347e-05,
      "loss": 1.404,
      "step": 84290
    },
    {
      "epoch": 4.178840584619032,
      "grad_norm": 0.16018883883953094,
      "learning_rate": 5.633654651451499e-05,
      "loss": 1.3957,
      "step": 84300
    },
    {
      "epoch": 4.179003166968686,
      "eval_loss": 2.4341461658477783,
      "eval_runtime": 21.9742,
      "eval_samples_per_second": 227.539,
      "eval_steps_per_second": 1.229,
      "step": 84303
    },
    {
      "epoch": 4.179382525784544,
      "grad_norm": 0.1671213060617447,
      "learning_rate": 5.632071523106805e-05,
      "loss": 1.4043,
      "step": 84310
    },
    {
      "epoch": 4.179924466950057,
      "grad_norm": 0.35982993245124817,
      "learning_rate": 5.630488378401665e-05,
      "loss": 1.3999,
      "step": 84320
    },
    {
      "epoch": 4.180466408115569,
      "grad_norm": 0.3010498881340027,
      "learning_rate": 5.62890521753219e-05,
      "loss": 1.4013,
      "step": 84330
    },
    {
      "epoch": 4.181008349281081,
      "grad_norm": 0.27373287081718445,
      "learning_rate": 5.627322040694497e-05,
      "loss": 1.3974,
      "step": 84340
    },
    {
      "epoch": 4.1815502904465935,
      "grad_norm": 0.21938002109527588,
      "learning_rate": 5.6257388480847026e-05,
      "loss": 1.4011,
      "step": 84350
    },
    {
      "epoch": 4.182092231612105,
      "grad_norm": 0.184707373380661,
      "learning_rate": 5.6241556398989246e-05,
      "loss": 1.3999,
      "step": 84360
    },
    {
      "epoch": 4.182634172777618,
      "grad_norm": 0.15879768133163452,
      "learning_rate": 5.622572416333286e-05,
      "loss": 1.4061,
      "step": 84370
    },
    {
      "epoch": 4.18317611394313,
      "grad_norm": 0.16178719699382782,
      "learning_rate": 5.620989177583908e-05,
      "loss": 1.4106,
      "step": 84380
    },
    {
      "epoch": 4.183718055108642,
      "grad_norm": 0.18774659931659698,
      "learning_rate": 5.619405923846916e-05,
      "loss": 1.4041,
      "step": 84390
    },
    {
      "epoch": 4.183718055108642,
      "eval_loss": 2.4340009689331055,
      "eval_runtime": 21.9661,
      "eval_samples_per_second": 227.623,
      "eval_steps_per_second": 1.229,
      "step": 84390
    },
    {
      "epoch": 4.184259996274155,
      "grad_norm": 0.29684072732925415,
      "learning_rate": 5.617822655318438e-05,
      "loss": 1.4182,
      "step": 84400
    },
    {
      "epoch": 4.1848019374396666,
      "grad_norm": 0.24184374511241913,
      "learning_rate": 5.616239372194599e-05,
      "loss": 1.406,
      "step": 84410
    },
    {
      "epoch": 4.185343878605179,
      "grad_norm": 0.17345279455184937,
      "learning_rate": 5.614656074671532e-05,
      "loss": 1.4158,
      "step": 84420
    },
    {
      "epoch": 4.185885819770691,
      "grad_norm": 0.18434324860572815,
      "learning_rate": 5.613072762945369e-05,
      "loss": 1.415,
      "step": 84430
    },
    {
      "epoch": 4.186427760936203,
      "grad_norm": 0.16675621271133423,
      "learning_rate": 5.6114894372122415e-05,
      "loss": 1.4093,
      "step": 84440
    },
    {
      "epoch": 4.186969702101716,
      "grad_norm": 0.18245235085487366,
      "learning_rate": 5.6099060976682883e-05,
      "loss": 1.4074,
      "step": 84450
    },
    {
      "epoch": 4.187511643267228,
      "grad_norm": 0.18735003471374512,
      "learning_rate": 5.608322744509644e-05,
      "loss": 1.403,
      "step": 84460
    },
    {
      "epoch": 4.1880535844327405,
      "grad_norm": 0.22122140228748322,
      "learning_rate": 5.606739377932447e-05,
      "loss": 1.3978,
      "step": 84470
    },
    {
      "epoch": 4.188432943248599,
      "eval_loss": 2.4321837425231934,
      "eval_runtime": 21.9752,
      "eval_samples_per_second": 227.529,
      "eval_steps_per_second": 1.229,
      "step": 84477
    },
    {
      "epoch": 4.188595525598252,
      "grad_norm": 0.19624446332454681,
      "learning_rate": 5.605155998132843e-05,
      "loss": 1.4006,
      "step": 84480
    },
    {
      "epoch": 4.189137466763764,
      "grad_norm": 0.2357572615146637,
      "learning_rate": 5.603572605306967e-05,
      "loss": 1.3976,
      "step": 84490
    },
    {
      "epoch": 4.189679407929277,
      "grad_norm": 0.1909378468990326,
      "learning_rate": 5.6019891996509676e-05,
      "loss": 1.3962,
      "step": 84500
    },
    {
      "epoch": 4.190221349094789,
      "grad_norm": 0.2646555006504059,
      "learning_rate": 5.600405781360989e-05,
      "loss": 1.4053,
      "step": 84510
    },
    {
      "epoch": 4.190763290260301,
      "grad_norm": 0.16853564977645874,
      "learning_rate": 5.598822350633177e-05,
      "loss": 1.4133,
      "step": 84520
    },
    {
      "epoch": 4.1913052314258135,
      "grad_norm": 0.1681780219078064,
      "learning_rate": 5.5972389076636833e-05,
      "loss": 1.4132,
      "step": 84530
    },
    {
      "epoch": 4.191847172591325,
      "grad_norm": 0.15586897730827332,
      "learning_rate": 5.595655452648655e-05,
      "loss": 1.4015,
      "step": 84540
    },
    {
      "epoch": 4.192389113756838,
      "grad_norm": 0.1730291098356247,
      "learning_rate": 5.594071985784244e-05,
      "loss": 1.3961,
      "step": 84550
    },
    {
      "epoch": 4.19293105492235,
      "grad_norm": 0.16029633581638336,
      "learning_rate": 5.5924885072666045e-05,
      "loss": 1.4024,
      "step": 84560
    },
    {
      "epoch": 4.193147831388555,
      "eval_loss": 2.4340431690216064,
      "eval_runtime": 21.9997,
      "eval_samples_per_second": 227.276,
      "eval_steps_per_second": 1.227,
      "step": 84564
    },
    {
      "epoch": 4.193472996087862,
      "grad_norm": 0.2048177421092987,
      "learning_rate": 5.5909050172918896e-05,
      "loss": 1.403,
      "step": 84570
    },
    {
      "epoch": 4.194014937253375,
      "grad_norm": 0.1830647587776184,
      "learning_rate": 5.589321516056256e-05,
      "loss": 1.4039,
      "step": 84580
    },
    {
      "epoch": 4.1945568784188865,
      "grad_norm": 0.18727242946624756,
      "learning_rate": 5.587738003755861e-05,
      "loss": 1.4054,
      "step": 84590
    },
    {
      "epoch": 4.195098819584399,
      "grad_norm": 0.2537229657173157,
      "learning_rate": 5.5861544805868624e-05,
      "loss": 1.4062,
      "step": 84600
    },
    {
      "epoch": 4.195640760749911,
      "grad_norm": 0.23978114128112793,
      "learning_rate": 5.584570946745422e-05,
      "loss": 1.4057,
      "step": 84610
    },
    {
      "epoch": 4.196182701915423,
      "grad_norm": 0.19209878146648407,
      "learning_rate": 5.5829874024276995e-05,
      "loss": 1.409,
      "step": 84620
    },
    {
      "epoch": 4.196724643080936,
      "grad_norm": 0.22839245200157166,
      "learning_rate": 5.581403847829857e-05,
      "loss": 1.4135,
      "step": 84630
    },
    {
      "epoch": 4.197266584246448,
      "grad_norm": 0.1937345266342163,
      "learning_rate": 5.5798202831480605e-05,
      "loss": 1.3996,
      "step": 84640
    },
    {
      "epoch": 4.19780852541196,
      "grad_norm": 0.23591110110282898,
      "learning_rate": 5.5782367085784725e-05,
      "loss": 1.4079,
      "step": 84650
    },
    {
      "epoch": 4.197862719528511,
      "eval_loss": 2.4296658039093018,
      "eval_runtime": 22.048,
      "eval_samples_per_second": 226.778,
      "eval_steps_per_second": 1.225,
      "step": 84651
    },
    {
      "epoch": 4.198350466577472,
      "grad_norm": 0.17365294694900513,
      "learning_rate": 5.5766531243172616e-05,
      "loss": 1.3987,
      "step": 84660
    },
    {
      "epoch": 4.198892407742984,
      "grad_norm": 0.20040321350097656,
      "learning_rate": 5.575069530560594e-05,
      "loss": 1.4116,
      "step": 84670
    },
    {
      "epoch": 4.199434348908497,
      "grad_norm": 0.1579209417104721,
      "learning_rate": 5.573485927504639e-05,
      "loss": 1.4013,
      "step": 84680
    },
    {
      "epoch": 4.199976290074009,
      "grad_norm": 0.2494509518146515,
      "learning_rate": 5.5719023153455674e-05,
      "loss": 1.4098,
      "step": 84690
    },
    {
      "epoch": 4.200518231239521,
      "grad_norm": 0.18554720282554626,
      "learning_rate": 5.5703186942795484e-05,
      "loss": 1.4108,
      "step": 84700
    },
    {
      "epoch": 4.201060172405033,
      "grad_norm": 0.1679755598306656,
      "learning_rate": 5.5687350645027544e-05,
      "loss": 1.4001,
      "step": 84710
    },
    {
      "epoch": 4.201602113570545,
      "grad_norm": 0.17022711038589478,
      "learning_rate": 5.56715142621136e-05,
      "loss": 1.3902,
      "step": 84720
    },
    {
      "epoch": 4.202144054736058,
      "grad_norm": 0.17805786430835724,
      "learning_rate": 5.5655677796015374e-05,
      "loss": 1.4025,
      "step": 84730
    },
    {
      "epoch": 4.202577607668467,
      "eval_loss": 2.4276254177093506,
      "eval_runtime": 22.04,
      "eval_samples_per_second": 226.86,
      "eval_steps_per_second": 1.225,
      "step": 84738
    },
    {
      "epoch": 4.20268599590157,
      "grad_norm": 0.18491090834140778,
      "learning_rate": 5.563984124869463e-05,
      "loss": 1.4003,
      "step": 84740
    },
    {
      "epoch": 4.203227937067082,
      "grad_norm": 0.1828235238790512,
      "learning_rate": 5.5624004622113125e-05,
      "loss": 1.4063,
      "step": 84750
    },
    {
      "epoch": 4.2037698782325945,
      "grad_norm": 0.160418301820755,
      "learning_rate": 5.560816791823264e-05,
      "loss": 1.4013,
      "step": 84760
    },
    {
      "epoch": 4.204311819398106,
      "grad_norm": 0.17551931738853455,
      "learning_rate": 5.559233113901498e-05,
      "loss": 1.4134,
      "step": 84770
    },
    {
      "epoch": 4.204853760563619,
      "grad_norm": 0.18492551147937775,
      "learning_rate": 5.557649428642189e-05,
      "loss": 1.4118,
      "step": 84780
    },
    {
      "epoch": 4.205395701729131,
      "grad_norm": 0.1690473109483719,
      "learning_rate": 5.556065736241518e-05,
      "loss": 1.4027,
      "step": 84790
    },
    {
      "epoch": 4.205937642894643,
      "grad_norm": 0.15582244098186493,
      "learning_rate": 5.5544820368956674e-05,
      "loss": 1.3997,
      "step": 84800
    },
    {
      "epoch": 4.206479584060156,
      "grad_norm": 0.21982191503047943,
      "learning_rate": 5.55289833080082e-05,
      "loss": 1.3989,
      "step": 84810
    },
    {
      "epoch": 4.2070215252256675,
      "grad_norm": 0.1695873886346817,
      "learning_rate": 5.551314618153156e-05,
      "loss": 1.4016,
      "step": 84820
    },
    {
      "epoch": 4.2072924958084235,
      "eval_loss": 2.426410436630249,
      "eval_runtime": 22.0191,
      "eval_samples_per_second": 227.076,
      "eval_steps_per_second": 1.226,
      "step": 84825
    },
    {
      "epoch": 4.20756346639118,
      "grad_norm": 0.15506160259246826,
      "learning_rate": 5.5497308991488606e-05,
      "loss": 1.4008,
      "step": 84830
    },
    {
      "epoch": 4.208105407556692,
      "grad_norm": 0.21147039532661438,
      "learning_rate": 5.548147173984116e-05,
      "loss": 1.3972,
      "step": 84840
    },
    {
      "epoch": 4.208647348722204,
      "grad_norm": 0.3284474313259125,
      "learning_rate": 5.546563442855108e-05,
      "loss": 1.405,
      "step": 84850
    },
    {
      "epoch": 4.209189289887717,
      "grad_norm": 0.2625653147697449,
      "learning_rate": 5.544979705958025e-05,
      "loss": 1.4091,
      "step": 84860
    },
    {
      "epoch": 4.209731231053229,
      "grad_norm": 0.24892371892929077,
      "learning_rate": 5.543395963489051e-05,
      "loss": 1.4066,
      "step": 84870
    },
    {
      "epoch": 4.210273172218741,
      "grad_norm": 0.26100948452949524,
      "learning_rate": 5.541812215644373e-05,
      "loss": 1.4048,
      "step": 84880
    },
    {
      "epoch": 4.210815113384253,
      "grad_norm": 0.18004588782787323,
      "learning_rate": 5.5402284626201794e-05,
      "loss": 1.3915,
      "step": 84890
    },
    {
      "epoch": 4.211357054549765,
      "grad_norm": 0.21767373383045197,
      "learning_rate": 5.538644704612658e-05,
      "loss": 1.4005,
      "step": 84900
    },
    {
      "epoch": 4.211898995715278,
      "grad_norm": 0.18447516858577728,
      "learning_rate": 5.537060941817999e-05,
      "loss": 1.4034,
      "step": 84910
    },
    {
      "epoch": 4.2120073839483805,
      "eval_loss": 2.4239137172698975,
      "eval_runtime": 21.9695,
      "eval_samples_per_second": 227.588,
      "eval_steps_per_second": 1.229,
      "step": 84912
    },
    {
      "epoch": 4.21244093688079,
      "grad_norm": 0.1832926869392395,
      "learning_rate": 5.535477174432391e-05,
      "loss": 1.4034,
      "step": 84920
    },
    {
      "epoch": 4.212982878046302,
      "grad_norm": 0.1812276393175125,
      "learning_rate": 5.533893402652025e-05,
      "loss": 1.392,
      "step": 84930
    },
    {
      "epoch": 4.213524819211814,
      "grad_norm": 0.18913184106349945,
      "learning_rate": 5.5323096266730914e-05,
      "loss": 1.4024,
      "step": 84940
    },
    {
      "epoch": 4.214066760377326,
      "grad_norm": 0.22599507868289948,
      "learning_rate": 5.530725846691781e-05,
      "loss": 1.4047,
      "step": 84950
    },
    {
      "epoch": 4.214608701542839,
      "grad_norm": 0.1804644763469696,
      "learning_rate": 5.529142062904286e-05,
      "loss": 1.4025,
      "step": 84960
    },
    {
      "epoch": 4.215150642708351,
      "grad_norm": 0.19616399705410004,
      "learning_rate": 5.527558275506799e-05,
      "loss": 1.3948,
      "step": 84970
    },
    {
      "epoch": 4.215692583873863,
      "grad_norm": 0.22985543310642242,
      "learning_rate": 5.5259744846955145e-05,
      "loss": 1.3937,
      "step": 84980
    },
    {
      "epoch": 4.216234525039376,
      "grad_norm": 0.20359660685062408,
      "learning_rate": 5.52439069066662e-05,
      "loss": 1.3941,
      "step": 84990
    },
    {
      "epoch": 4.216722272088337,
      "eval_loss": 2.4288182258605957,
      "eval_runtime": 21.9691,
      "eval_samples_per_second": 227.592,
      "eval_steps_per_second": 1.229,
      "step": 84999
    },
    {
      "epoch": 4.216776466204887,
      "grad_norm": 0.18419791758060455,
      "learning_rate": 5.5228068936163134e-05,
      "loss": 1.4054,
      "step": 85000
    },
    {
      "epoch": 4.2173184073704,
      "grad_norm": 0.17802853882312775,
      "learning_rate": 5.52122309374079e-05,
      "loss": 1.4009,
      "step": 85010
    },
    {
      "epoch": 4.217860348535912,
      "grad_norm": 0.16491647064685822,
      "learning_rate": 5.519639291236241e-05,
      "loss": 1.4079,
      "step": 85020
    },
    {
      "epoch": 4.218402289701424,
      "grad_norm": 0.26249879598617554,
      "learning_rate": 5.518055486298862e-05,
      "loss": 1.4104,
      "step": 85030
    },
    {
      "epoch": 4.218944230866937,
      "grad_norm": 0.16674381494522095,
      "learning_rate": 5.516471679124846e-05,
      "loss": 1.4072,
      "step": 85040
    },
    {
      "epoch": 4.219486172032449,
      "grad_norm": 0.1718483567237854,
      "learning_rate": 5.514887869910391e-05,
      "loss": 1.4052,
      "step": 85050
    },
    {
      "epoch": 4.220028113197961,
      "grad_norm": 0.18503984808921814,
      "learning_rate": 5.5133040588516925e-05,
      "loss": 1.3951,
      "step": 85060
    },
    {
      "epoch": 4.220570054363473,
      "grad_norm": 0.25337323546409607,
      "learning_rate": 5.511720246144945e-05,
      "loss": 1.399,
      "step": 85070
    },
    {
      "epoch": 4.221111995528985,
      "grad_norm": 0.17536914348602295,
      "learning_rate": 5.510136431986342e-05,
      "loss": 1.401,
      "step": 85080
    },
    {
      "epoch": 4.221437160228293,
      "eval_loss": 2.4300763607025146,
      "eval_runtime": 21.9731,
      "eval_samples_per_second": 227.551,
      "eval_steps_per_second": 1.229,
      "step": 85086
    },
    {
      "epoch": 4.221653936694498,
      "grad_norm": 0.25776609778404236,
      "learning_rate": 5.5085526165720835e-05,
      "loss": 1.4017,
      "step": 85090
    },
    {
      "epoch": 4.22219587786001,
      "grad_norm": 0.23579788208007812,
      "learning_rate": 5.5069688000983635e-05,
      "loss": 1.3972,
      "step": 85100
    },
    {
      "epoch": 4.222737819025522,
      "grad_norm": 0.23605017364025116,
      "learning_rate": 5.505384982761379e-05,
      "loss": 1.4034,
      "step": 85110
    },
    {
      "epoch": 4.223279760191034,
      "grad_norm": 0.23162227869033813,
      "learning_rate": 5.503801164757327e-05,
      "loss": 1.403,
      "step": 85120
    },
    {
      "epoch": 4.223821701356546,
      "grad_norm": 0.21863441169261932,
      "learning_rate": 5.502217346282401e-05,
      "loss": 1.4017,
      "step": 85130
    },
    {
      "epoch": 4.224363642522059,
      "grad_norm": 0.1740538477897644,
      "learning_rate": 5.500633527532802e-05,
      "loss": 1.4107,
      "step": 85140
    },
    {
      "epoch": 4.224905583687571,
      "grad_norm": 0.1974310427904129,
      "learning_rate": 5.499049708704723e-05,
      "loss": 1.4013,
      "step": 85150
    },
    {
      "epoch": 4.225447524853083,
      "grad_norm": 0.24513813853263855,
      "learning_rate": 5.497465889994362e-05,
      "loss": 1.4011,
      "step": 85160
    },
    {
      "epoch": 4.2259894660185955,
      "grad_norm": 0.28687775135040283,
      "learning_rate": 5.495882071597915e-05,
      "loss": 1.4033,
      "step": 85170
    },
    {
      "epoch": 4.226152048368249,
      "eval_loss": 2.429248332977295,
      "eval_runtime": 21.97,
      "eval_samples_per_second": 227.583,
      "eval_steps_per_second": 1.229,
      "step": 85173
    },
    {
      "epoch": 4.226531407184107,
      "grad_norm": 0.17238466441631317,
      "learning_rate": 5.4942982537115826e-05,
      "loss": 1.4076,
      "step": 85180
    },
    {
      "epoch": 4.22707334834962,
      "grad_norm": 0.1770007312297821,
      "learning_rate": 5.4927144365315544e-05,
      "loss": 1.4096,
      "step": 85190
    },
    {
      "epoch": 4.227615289515132,
      "grad_norm": 0.1884445995092392,
      "learning_rate": 5.49113062025403e-05,
      "loss": 1.3935,
      "step": 85200
    },
    {
      "epoch": 4.228157230680644,
      "grad_norm": 0.17585532367229462,
      "learning_rate": 5.489546805075209e-05,
      "loss": 1.4014,
      "step": 85210
    },
    {
      "epoch": 4.228699171846157,
      "grad_norm": 0.15574230253696442,
      "learning_rate": 5.487962991191281e-05,
      "loss": 1.4228,
      "step": 85220
    },
    {
      "epoch": 4.2292411130116685,
      "grad_norm": 0.1755153387784958,
      "learning_rate": 5.486379178798446e-05,
      "loss": 1.4031,
      "step": 85230
    },
    {
      "epoch": 4.229783054177181,
      "grad_norm": 0.22659145295619965,
      "learning_rate": 5.484795368092901e-05,
      "loss": 1.4086,
      "step": 85240
    },
    {
      "epoch": 4.230324995342693,
      "grad_norm": 0.2872506082057953,
      "learning_rate": 5.483211559270838e-05,
      "loss": 1.4091,
      "step": 85250
    },
    {
      "epoch": 4.230866936508205,
      "grad_norm": 0.21537162363529205,
      "learning_rate": 5.4816277525284544e-05,
      "loss": 1.4028,
      "step": 85260
    },
    {
      "epoch": 4.230866936508205,
      "eval_loss": 2.42655086517334,
      "eval_runtime": 21.9673,
      "eval_samples_per_second": 227.611,
      "eval_steps_per_second": 1.229,
      "step": 85260
    },
    {
      "epoch": 4.231408877673718,
      "grad_norm": 0.17939117550849915,
      "learning_rate": 5.480043948061947e-05,
      "loss": 1.4065,
      "step": 85270
    },
    {
      "epoch": 4.23195081883923,
      "grad_norm": 0.5240494012832642,
      "learning_rate": 5.4784601460675064e-05,
      "loss": 1.4144,
      "step": 85280
    },
    {
      "epoch": 4.232492760004742,
      "grad_norm": 0.18183523416519165,
      "learning_rate": 5.47687634674133e-05,
      "loss": 1.3938,
      "step": 85290
    },
    {
      "epoch": 4.233034701170254,
      "grad_norm": 0.16505871713161469,
      "learning_rate": 5.4752925502796105e-05,
      "loss": 1.3924,
      "step": 85300
    },
    {
      "epoch": 4.233576642335766,
      "grad_norm": 0.20003126561641693,
      "learning_rate": 5.473708756878545e-05,
      "loss": 1.4059,
      "step": 85310
    },
    {
      "epoch": 4.234118583501279,
      "grad_norm": 0.21324165165424347,
      "learning_rate": 5.472124966734322e-05,
      "loss": 1.4002,
      "step": 85320
    },
    {
      "epoch": 4.234660524666791,
      "grad_norm": 0.32130008935928345,
      "learning_rate": 5.4705411800431386e-05,
      "loss": 1.4071,
      "step": 85330
    },
    {
      "epoch": 4.2352024658323035,
      "grad_norm": 0.18662355840206146,
      "learning_rate": 5.468957397001186e-05,
      "loss": 1.3962,
      "step": 85340
    },
    {
      "epoch": 4.235581824648162,
      "eval_loss": 2.428666830062866,
      "eval_runtime": 21.9762,
      "eval_samples_per_second": 227.519,
      "eval_steps_per_second": 1.229,
      "step": 85347
    },
    {
      "epoch": 4.235744406997815,
      "grad_norm": 0.271916002035141,
      "learning_rate": 5.467373617804655e-05,
      "loss": 1.4044,
      "step": 85350
    },
    {
      "epoch": 4.236286348163327,
      "grad_norm": 0.27290457487106323,
      "learning_rate": 5.465789842649739e-05,
      "loss": 1.3982,
      "step": 85360
    },
    {
      "epoch": 4.23682828932884,
      "grad_norm": 0.25049006938934326,
      "learning_rate": 5.4642060717326305e-05,
      "loss": 1.4102,
      "step": 85370
    },
    {
      "epoch": 4.237370230494352,
      "grad_norm": 0.17757458984851837,
      "learning_rate": 5.462622305249517e-05,
      "loss": 1.4,
      "step": 85380
    },
    {
      "epoch": 4.237912171659864,
      "grad_norm": 0.16723614931106567,
      "learning_rate": 5.461038543396589e-05,
      "loss": 1.4082,
      "step": 85390
    },
    {
      "epoch": 4.2384541128253765,
      "grad_norm": 0.14943601191043854,
      "learning_rate": 5.4594547863700396e-05,
      "loss": 1.4074,
      "step": 85400
    },
    {
      "epoch": 4.238996053990888,
      "grad_norm": 0.18632861971855164,
      "learning_rate": 5.457871034366053e-05,
      "loss": 1.4057,
      "step": 85410
    },
    {
      "epoch": 4.239537995156401,
      "grad_norm": 0.16720369458198547,
      "learning_rate": 5.456287287580821e-05,
      "loss": 1.3995,
      "step": 85420
    },
    {
      "epoch": 4.240079936321913,
      "grad_norm": 0.23148633539676666,
      "learning_rate": 5.45470354621053e-05,
      "loss": 1.3956,
      "step": 85430
    },
    {
      "epoch": 4.240296712788118,
      "eval_loss": 2.4289402961730957,
      "eval_runtime": 21.9722,
      "eval_samples_per_second": 227.561,
      "eval_steps_per_second": 1.229,
      "step": 85434
    },
    {
      "epoch": 4.240621877487425,
      "grad_norm": 0.3228211998939514,
      "learning_rate": 5.453119810451366e-05,
      "loss": 1.3988,
      "step": 85440
    },
    {
      "epoch": 4.241163818652938,
      "grad_norm": 0.28075844049453735,
      "learning_rate": 5.451536080499518e-05,
      "loss": 1.3984,
      "step": 85450
    },
    {
      "epoch": 4.2417057598184496,
      "grad_norm": 0.23300626873970032,
      "learning_rate": 5.449952356551168e-05,
      "loss": 1.3963,
      "step": 85460
    },
    {
      "epoch": 4.242247700983962,
      "grad_norm": 0.2622208595275879,
      "learning_rate": 5.448368638802506e-05,
      "loss": 1.3987,
      "step": 85470
    },
    {
      "epoch": 4.242789642149474,
      "grad_norm": 0.15895061194896698,
      "learning_rate": 5.4467849274497094e-05,
      "loss": 1.4072,
      "step": 85480
    },
    {
      "epoch": 4.243331583314986,
      "grad_norm": 0.17281389236450195,
      "learning_rate": 5.4452012226889646e-05,
      "loss": 1.4066,
      "step": 85490
    },
    {
      "epoch": 4.243873524480499,
      "grad_norm": 0.2625819444656372,
      "learning_rate": 5.443617524716458e-05,
      "loss": 1.4131,
      "step": 85500
    },
    {
      "epoch": 4.244415465646011,
      "grad_norm": 0.20588402450084686,
      "learning_rate": 5.442033833728365e-05,
      "loss": 1.4123,
      "step": 85510
    },
    {
      "epoch": 4.2449574068115234,
      "grad_norm": 0.22198638319969177,
      "learning_rate": 5.440450149920869e-05,
      "loss": 1.4101,
      "step": 85520
    },
    {
      "epoch": 4.245011600928074,
      "eval_loss": 2.4276344776153564,
      "eval_runtime": 21.9945,
      "eval_samples_per_second": 227.33,
      "eval_steps_per_second": 1.228,
      "step": 85521
    },
    {
      "epoch": 4.245499347977035,
      "grad_norm": 0.16215014457702637,
      "learning_rate": 5.438866473490152e-05,
      "loss": 1.3984,
      "step": 85530
    },
    {
      "epoch": 4.246041289142547,
      "grad_norm": 0.1525203287601471,
      "learning_rate": 5.4372828046323885e-05,
      "loss": 1.3939,
      "step": 85540
    },
    {
      "epoch": 4.24658323030806,
      "grad_norm": 0.16212084889411926,
      "learning_rate": 5.43569914354376e-05,
      "loss": 1.4041,
      "step": 85550
    },
    {
      "epoch": 4.247125171473572,
      "grad_norm": 0.17924129962921143,
      "learning_rate": 5.434115490420443e-05,
      "loss": 1.3977,
      "step": 85560
    },
    {
      "epoch": 4.247667112639084,
      "grad_norm": 0.22136323153972626,
      "learning_rate": 5.432531845458612e-05,
      "loss": 1.4001,
      "step": 85570
    },
    {
      "epoch": 4.2482090538045965,
      "grad_norm": 0.26103144884109497,
      "learning_rate": 5.4309482088544416e-05,
      "loss": 1.405,
      "step": 85580
    },
    {
      "epoch": 4.248750994970108,
      "grad_norm": 0.23051413893699646,
      "learning_rate": 5.429364580804111e-05,
      "loss": 1.4005,
      "step": 85590
    },
    {
      "epoch": 4.249292936135621,
      "grad_norm": 0.19414301216602325,
      "learning_rate": 5.427780961503787e-05,
      "loss": 1.4039,
      "step": 85600
    },
    {
      "epoch": 4.24972648906803,
      "eval_loss": 2.4262514114379883,
      "eval_runtime": 21.973,
      "eval_samples_per_second": 227.551,
      "eval_steps_per_second": 1.229,
      "step": 85608
    },
    {
      "epoch": 4.249834877301133,
      "grad_norm": 0.1806284487247467,
      "learning_rate": 5.426197351149645e-05,
      "loss": 1.4023,
      "step": 85610
    },
    {
      "epoch": 4.250376818466645,
      "grad_norm": 0.1894899308681488,
      "learning_rate": 5.424613749937852e-05,
      "loss": 1.3989,
      "step": 85620
    },
    {
      "epoch": 4.250918759632158,
      "grad_norm": 0.1757279932498932,
      "learning_rate": 5.423030158064584e-05,
      "loss": 1.4181,
      "step": 85630
    },
    {
      "epoch": 4.2514607007976695,
      "grad_norm": 0.345478892326355,
      "learning_rate": 5.421446575726004e-05,
      "loss": 1.4023,
      "step": 85640
    },
    {
      "epoch": 4.252002641963182,
      "grad_norm": 0.36112889647483826,
      "learning_rate": 5.419863003118281e-05,
      "loss": 1.4054,
      "step": 85650
    },
    {
      "epoch": 4.252544583128694,
      "grad_norm": 0.1516331434249878,
      "learning_rate": 5.418279440437582e-05,
      "loss": 1.4045,
      "step": 85660
    },
    {
      "epoch": 4.253086524294206,
      "grad_norm": 0.23912928998470306,
      "learning_rate": 5.416695887880071e-05,
      "loss": 1.4001,
      "step": 85670
    },
    {
      "epoch": 4.253628465459719,
      "grad_norm": 0.27274298667907715,
      "learning_rate": 5.41511234564191e-05,
      "loss": 1.4137,
      "step": 85680
    },
    {
      "epoch": 4.254170406625231,
      "grad_norm": 0.25264525413513184,
      "learning_rate": 5.4135288139192664e-05,
      "loss": 1.4097,
      "step": 85690
    },
    {
      "epoch": 4.2544413772079865,
      "eval_loss": 2.4301393032073975,
      "eval_runtime": 21.9696,
      "eval_samples_per_second": 227.587,
      "eval_steps_per_second": 1.229,
      "step": 85695
    },
    {
      "epoch": 4.254712347790743,
      "grad_norm": 0.1786414533853531,
      "learning_rate": 5.411945292908296e-05,
      "loss": 1.4005,
      "step": 85700
    },
    {
      "epoch": 4.255254288956255,
      "grad_norm": 0.17254769802093506,
      "learning_rate": 5.41036178280516e-05,
      "loss": 1.4008,
      "step": 85710
    },
    {
      "epoch": 4.255796230121767,
      "grad_norm": 0.31122511625289917,
      "learning_rate": 5.408778283806018e-05,
      "loss": 1.4009,
      "step": 85720
    },
    {
      "epoch": 4.25633817128728,
      "grad_norm": 0.2594967484474182,
      "learning_rate": 5.407194796107027e-05,
      "loss": 1.4123,
      "step": 85730
    },
    {
      "epoch": 4.256880112452792,
      "grad_norm": 0.26264312863349915,
      "learning_rate": 5.405611319904339e-05,
      "loss": 1.39,
      "step": 85740
    },
    {
      "epoch": 4.2574220536183045,
      "grad_norm": 0.18427279591560364,
      "learning_rate": 5.404027855394113e-05,
      "loss": 1.3978,
      "step": 85750
    },
    {
      "epoch": 4.257963994783816,
      "grad_norm": 0.2185283601284027,
      "learning_rate": 5.4024444027725006e-05,
      "loss": 1.4082,
      "step": 85760
    },
    {
      "epoch": 4.258505935949328,
      "grad_norm": 0.16532769799232483,
      "learning_rate": 5.40086096223565e-05,
      "loss": 1.4063,
      "step": 85770
    },
    {
      "epoch": 4.259047877114841,
      "grad_norm": 0.17376570403575897,
      "learning_rate": 5.399277533979712e-05,
      "loss": 1.4046,
      "step": 85780
    },
    {
      "epoch": 4.259156265347944,
      "eval_loss": 2.4298863410949707,
      "eval_runtime": 21.973,
      "eval_samples_per_second": 227.552,
      "eval_steps_per_second": 1.229,
      "step": 85782
    },
    {
      "epoch": 4.259589818280353,
      "grad_norm": 0.17817267775535583,
      "learning_rate": 5.397694118200839e-05,
      "loss": 1.4041,
      "step": 85790
    },
    {
      "epoch": 4.260131759445865,
      "grad_norm": 0.21518279612064362,
      "learning_rate": 5.396110715095173e-05,
      "loss": 1.3957,
      "step": 85800
    },
    {
      "epoch": 4.2606737006113775,
      "grad_norm": 0.1868053376674652,
      "learning_rate": 5.3945273248588604e-05,
      "loss": 1.3996,
      "step": 85810
    },
    {
      "epoch": 4.261215641776889,
      "grad_norm": 0.16047891974449158,
      "learning_rate": 5.392943947688045e-05,
      "loss": 1.4125,
      "step": 85820
    },
    {
      "epoch": 4.261757582942402,
      "grad_norm": 0.16372744739055634,
      "learning_rate": 5.3913605837788686e-05,
      "loss": 1.399,
      "step": 85830
    },
    {
      "epoch": 4.262299524107914,
      "grad_norm": 0.15229913592338562,
      "learning_rate": 5.3897772333274696e-05,
      "loss": 1.4013,
      "step": 85840
    },
    {
      "epoch": 4.262841465273426,
      "grad_norm": 0.2408658266067505,
      "learning_rate": 5.3881938965299916e-05,
      "loss": 1.4027,
      "step": 85850
    },
    {
      "epoch": 4.263383406438939,
      "grad_norm": 0.1828664392232895,
      "learning_rate": 5.386610573582567e-05,
      "loss": 1.398,
      "step": 85860
    },
    {
      "epoch": 4.2638711534879,
      "eval_loss": 2.43064022064209,
      "eval_runtime": 21.9716,
      "eval_samples_per_second": 227.567,
      "eval_steps_per_second": 1.229,
      "step": 85869
    },
    {
      "epoch": 4.2639253476044505,
      "grad_norm": 0.24171289801597595,
      "learning_rate": 5.385027264681332e-05,
      "loss": 1.4029,
      "step": 85870
    },
    {
      "epoch": 4.264467288769963,
      "grad_norm": 0.19249555468559265,
      "learning_rate": 5.38344397002242e-05,
      "loss": 1.4006,
      "step": 85880
    },
    {
      "epoch": 4.265009229935475,
      "grad_norm": 0.1631373018026352,
      "learning_rate": 5.381860689801963e-05,
      "loss": 1.4004,
      "step": 85890
    },
    {
      "epoch": 4.265551171100987,
      "grad_norm": 0.2803032398223877,
      "learning_rate": 5.38027742421609e-05,
      "loss": 1.4041,
      "step": 85900
    },
    {
      "epoch": 4.2660931122665,
      "grad_norm": 0.1907922625541687,
      "learning_rate": 5.378694173460932e-05,
      "loss": 1.4021,
      "step": 85910
    },
    {
      "epoch": 4.266635053432012,
      "grad_norm": 0.182045578956604,
      "learning_rate": 5.377110937732612e-05,
      "loss": 1.4019,
      "step": 85920
    },
    {
      "epoch": 4.267176994597524,
      "grad_norm": 0.17628154158592224,
      "learning_rate": 5.3755277172272556e-05,
      "loss": 1.3939,
      "step": 85930
    },
    {
      "epoch": 4.267718935763036,
      "grad_norm": 0.18110013008117676,
      "learning_rate": 5.3739445121409846e-05,
      "loss": 1.4047,
      "step": 85940
    },
    {
      "epoch": 4.268260876928548,
      "grad_norm": 0.2522640824317932,
      "learning_rate": 5.372361322669922e-05,
      "loss": 1.4082,
      "step": 85950
    },
    {
      "epoch": 4.268586041627856,
      "eval_loss": 2.4356658458709717,
      "eval_runtime": 21.9696,
      "eval_samples_per_second": 227.588,
      "eval_steps_per_second": 1.229,
      "step": 85956
    },
    {
      "epoch": 4.268802818094061,
      "grad_norm": 0.17792746424674988,
      "learning_rate": 5.370778149010184e-05,
      "loss": 1.3923,
      "step": 85960
    },
    {
      "epoch": 4.269344759259573,
      "grad_norm": 0.20900298655033112,
      "learning_rate": 5.3691949913578865e-05,
      "loss": 1.3991,
      "step": 85970
    },
    {
      "epoch": 4.269886700425085,
      "grad_norm": 0.18769441545009613,
      "learning_rate": 5.3676118499091466e-05,
      "loss": 1.4003,
      "step": 85980
    },
    {
      "epoch": 4.270428641590597,
      "grad_norm": 0.20186538994312286,
      "learning_rate": 5.366028724860076e-05,
      "loss": 1.3982,
      "step": 85990
    },
    {
      "epoch": 4.270970582756109,
      "grad_norm": 0.15831774473190308,
      "learning_rate": 5.3644456164067847e-05,
      "loss": 1.4019,
      "step": 86000
    },
    {
      "epoch": 4.271512523921622,
      "grad_norm": 0.20776121318340302,
      "learning_rate": 5.362862524745382e-05,
      "loss": 1.3977,
      "step": 86010
    },
    {
      "epoch": 4.272054465087134,
      "grad_norm": 0.1834535449743271,
      "learning_rate": 5.3612794500719745e-05,
      "loss": 1.4064,
      "step": 86020
    },
    {
      "epoch": 4.272596406252646,
      "grad_norm": 0.18609929084777832,
      "learning_rate": 5.3596963925826657e-05,
      "loss": 1.4038,
      "step": 86030
    },
    {
      "epoch": 4.273138347418159,
      "grad_norm": 0.19254687428474426,
      "learning_rate": 5.3581133524735614e-05,
      "loss": 1.3849,
      "step": 86040
    },
    {
      "epoch": 4.273300929767812,
      "eval_loss": 2.434671640396118,
      "eval_runtime": 21.6596,
      "eval_samples_per_second": 230.845,
      "eval_steps_per_second": 1.247,
      "step": 86043
    },
    {
      "epoch": 4.27368028858367,
      "grad_norm": 0.1640702337026596,
      "learning_rate": 5.356530329940757e-05,
      "loss": 1.4098,
      "step": 86050
    },
    {
      "epoch": 4.274222229749183,
      "grad_norm": 0.20824840664863586,
      "learning_rate": 5.3549473251803514e-05,
      "loss": 1.4025,
      "step": 86060
    },
    {
      "epoch": 4.274764170914695,
      "grad_norm": 0.18921475112438202,
      "learning_rate": 5.353364338388442e-05,
      "loss": 1.399,
      "step": 86070
    },
    {
      "epoch": 4.275306112080207,
      "grad_norm": 0.27768686413764954,
      "learning_rate": 5.351781369761124e-05,
      "loss": 1.407,
      "step": 86080
    },
    {
      "epoch": 4.27584805324572,
      "grad_norm": 0.2548810839653015,
      "learning_rate": 5.350198419494484e-05,
      "loss": 1.4034,
      "step": 86090
    },
    {
      "epoch": 4.276389994411232,
      "grad_norm": 0.1735634207725525,
      "learning_rate": 5.348615487784614e-05,
      "loss": 1.4001,
      "step": 86100
    },
    {
      "epoch": 4.276931935576744,
      "grad_norm": 0.16922864317893982,
      "learning_rate": 5.3470325748276004e-05,
      "loss": 1.3943,
      "step": 86110
    },
    {
      "epoch": 4.277473876742256,
      "grad_norm": 0.20345307886600494,
      "learning_rate": 5.3454496808195264e-05,
      "loss": 1.41,
      "step": 86120
    },
    {
      "epoch": 4.278015817907768,
      "grad_norm": 0.20915158092975616,
      "learning_rate": 5.343866805956476e-05,
      "loss": 1.4036,
      "step": 86130
    },
    {
      "epoch": 4.278015817907768,
      "eval_loss": 2.4313769340515137,
      "eval_runtime": 21.964,
      "eval_samples_per_second": 227.645,
      "eval_steps_per_second": 1.229,
      "step": 86130
    },
    {
      "epoch": 4.278557759073281,
      "grad_norm": 0.2022494673728943,
      "learning_rate": 5.342283950434529e-05,
      "loss": 1.3995,
      "step": 86140
    },
    {
      "epoch": 4.279099700238793,
      "grad_norm": 0.1643056720495224,
      "learning_rate": 5.3407011144497596e-05,
      "loss": 1.4006,
      "step": 86150
    },
    {
      "epoch": 4.2796416414043055,
      "grad_norm": 0.23833966255187988,
      "learning_rate": 5.339118298198245e-05,
      "loss": 1.3962,
      "step": 86160
    },
    {
      "epoch": 4.280183582569817,
      "grad_norm": 0.22995562851428986,
      "learning_rate": 5.337535501876057e-05,
      "loss": 1.4049,
      "step": 86170
    },
    {
      "epoch": 4.280725523735329,
      "grad_norm": 0.29235386848449707,
      "learning_rate": 5.335952725679265e-05,
      "loss": 1.3905,
      "step": 86180
    },
    {
      "epoch": 4.281267464900842,
      "grad_norm": 0.3634209632873535,
      "learning_rate": 5.334369969803937e-05,
      "loss": 1.4187,
      "step": 86190
    },
    {
      "epoch": 4.281809406066354,
      "grad_norm": 0.2359488159418106,
      "learning_rate": 5.33278723444614e-05,
      "loss": 1.3963,
      "step": 86200
    },
    {
      "epoch": 4.282351347231867,
      "grad_norm": 0.19134867191314697,
      "learning_rate": 5.331204519801933e-05,
      "loss": 1.4091,
      "step": 86210
    },
    {
      "epoch": 4.282730706047725,
      "eval_loss": 2.434406280517578,
      "eval_runtime": 21.9736,
      "eval_samples_per_second": 227.546,
      "eval_steps_per_second": 1.229,
      "step": 86217
    },
    {
      "epoch": 4.2828932883973785,
      "grad_norm": 0.22415363788604736,
      "learning_rate": 5.329621826067377e-05,
      "loss": 1.4002,
      "step": 86220
    },
    {
      "epoch": 4.28343522956289,
      "grad_norm": 0.22894415259361267,
      "learning_rate": 5.32803915343853e-05,
      "loss": 1.4073,
      "step": 86230
    },
    {
      "epoch": 4.283977170728403,
      "grad_norm": 0.21755249798297882,
      "learning_rate": 5.3264565021114474e-05,
      "loss": 1.3983,
      "step": 86240
    },
    {
      "epoch": 4.284519111893915,
      "grad_norm": 0.22237862646579742,
      "learning_rate": 5.324873872282179e-05,
      "loss": 1.41,
      "step": 86250
    },
    {
      "epoch": 4.285061053059427,
      "grad_norm": 0.2396155744791031,
      "learning_rate": 5.323291264146774e-05,
      "loss": 1.3943,
      "step": 86260
    },
    {
      "epoch": 4.28560299422494,
      "grad_norm": 0.1976693719625473,
      "learning_rate": 5.321708677901282e-05,
      "loss": 1.4141,
      "step": 86270
    },
    {
      "epoch": 4.2861449353904515,
      "grad_norm": 0.23698921501636505,
      "learning_rate": 5.320126113741742e-05,
      "loss": 1.3918,
      "step": 86280
    },
    {
      "epoch": 4.286686876555964,
      "grad_norm": 0.2332429438829422,
      "learning_rate": 5.3185435718641993e-05,
      "loss": 1.3988,
      "step": 86290
    },
    {
      "epoch": 4.287228817721476,
      "grad_norm": 0.18696242570877075,
      "learning_rate": 5.3169610524646916e-05,
      "loss": 1.4141,
      "step": 86300
    },
    {
      "epoch": 4.287445594187681,
      "eval_loss": 2.429657459259033,
      "eval_runtime": 21.9725,
      "eval_samples_per_second": 227.558,
      "eval_steps_per_second": 1.229,
      "step": 86304
    },
    {
      "epoch": 4.287770758886988,
      "grad_norm": 0.15365861356258392,
      "learning_rate": 5.315378555739254e-05,
      "loss": 1.4071,
      "step": 86310
    },
    {
      "epoch": 4.288312700052501,
      "grad_norm": 0.16417507827281952,
      "learning_rate": 5.313796081883918e-05,
      "loss": 1.3956,
      "step": 86320
    },
    {
      "epoch": 4.288854641218013,
      "grad_norm": 0.19014696776866913,
      "learning_rate": 5.3122136310947194e-05,
      "loss": 1.4021,
      "step": 86330
    },
    {
      "epoch": 4.289396582383525,
      "grad_norm": 0.20190609991550446,
      "learning_rate": 5.3106312035676766e-05,
      "loss": 1.3925,
      "step": 86340
    },
    {
      "epoch": 4.289938523549037,
      "grad_norm": 0.16114814579486847,
      "learning_rate": 5.3090487994988184e-05,
      "loss": 1.3964,
      "step": 86350
    },
    {
      "epoch": 4.290480464714549,
      "grad_norm": 0.3137826919555664,
      "learning_rate": 5.307466419084166e-05,
      "loss": 1.4145,
      "step": 86360
    },
    {
      "epoch": 4.291022405880062,
      "grad_norm": 0.23774784803390503,
      "learning_rate": 5.3058840625197394e-05,
      "loss": 1.4024,
      "step": 86370
    },
    {
      "epoch": 4.291564347045574,
      "grad_norm": 0.23315760493278503,
      "learning_rate": 5.304301730001552e-05,
      "loss": 1.3947,
      "step": 86380
    },
    {
      "epoch": 4.292106288211086,
      "grad_norm": 0.34956422448158264,
      "learning_rate": 5.302719421725615e-05,
      "loss": 1.3983,
      "step": 86390
    },
    {
      "epoch": 4.292160482327637,
      "eval_loss": 2.433197498321533,
      "eval_runtime": 21.9709,
      "eval_samples_per_second": 227.574,
      "eval_steps_per_second": 1.229,
      "step": 86391
    },
    {
      "epoch": 4.292648229376598,
      "grad_norm": 0.17017148435115814,
      "learning_rate": 5.30113713788794e-05,
      "loss": 1.3998,
      "step": 86400
    },
    {
      "epoch": 4.29319017054211,
      "grad_norm": 0.17241883277893066,
      "learning_rate": 5.2995548786845336e-05,
      "loss": 1.3969,
      "step": 86410
    },
    {
      "epoch": 4.293732111707623,
      "grad_norm": 0.19140473008155823,
      "learning_rate": 5.2979726443113985e-05,
      "loss": 1.3943,
      "step": 86420
    },
    {
      "epoch": 4.294274052873135,
      "grad_norm": 0.18566745519638062,
      "learning_rate": 5.296390434964537e-05,
      "loss": 1.4065,
      "step": 86430
    },
    {
      "epoch": 4.294815994038647,
      "grad_norm": 0.19149713218212128,
      "learning_rate": 5.294808250839942e-05,
      "loss": 1.3848,
      "step": 86440
    },
    {
      "epoch": 4.2953579352041595,
      "grad_norm": 0.2127964198589325,
      "learning_rate": 5.2932260921336106e-05,
      "loss": 1.4022,
      "step": 86450
    },
    {
      "epoch": 4.295899876369671,
      "grad_norm": 0.17330385744571686,
      "learning_rate": 5.291643959041533e-05,
      "loss": 1.4034,
      "step": 86460
    },
    {
      "epoch": 4.296441817535184,
      "grad_norm": 0.21223121881484985,
      "learning_rate": 5.290061851759698e-05,
      "loss": 1.4036,
      "step": 86470
    },
    {
      "epoch": 4.2968753704675935,
      "eval_loss": 2.4309399127960205,
      "eval_runtime": 21.9752,
      "eval_samples_per_second": 227.529,
      "eval_steps_per_second": 1.229,
      "step": 86478
    },
    {
      "epoch": 4.296983758700696,
      "grad_norm": 0.2610706388950348,
      "learning_rate": 5.288479770484088e-05,
      "loss": 1.4027,
      "step": 86480
    },
    {
      "epoch": 4.297525699866208,
      "grad_norm": 0.159620001912117,
      "learning_rate": 5.2868977154106866e-05,
      "loss": 1.4052,
      "step": 86490
    },
    {
      "epoch": 4.298067641031721,
      "grad_norm": 0.2618952691555023,
      "learning_rate": 5.28531568673547e-05,
      "loss": 1.4027,
      "step": 86500
    },
    {
      "epoch": 4.2986095821972325,
      "grad_norm": 0.2610114812850952,
      "learning_rate": 5.283733684654415e-05,
      "loss": 1.4044,
      "step": 86510
    },
    {
      "epoch": 4.299151523362745,
      "grad_norm": 0.22828315198421478,
      "learning_rate": 5.282151709363492e-05,
      "loss": 1.4014,
      "step": 86520
    },
    {
      "epoch": 4.299693464528257,
      "grad_norm": 0.1908428966999054,
      "learning_rate": 5.2805697610586716e-05,
      "loss": 1.4008,
      "step": 86530
    },
    {
      "epoch": 4.300235405693769,
      "grad_norm": 0.21566815674304962,
      "learning_rate": 5.278987839935913e-05,
      "loss": 1.4012,
      "step": 86540
    },
    {
      "epoch": 4.300777346859282,
      "grad_norm": 0.23819507658481598,
      "learning_rate": 5.277405946191183e-05,
      "loss": 1.4001,
      "step": 86550
    },
    {
      "epoch": 4.301319288024794,
      "grad_norm": 0.19423484802246094,
      "learning_rate": 5.275824080020439e-05,
      "loss": 1.4052,
      "step": 86560
    },
    {
      "epoch": 4.30159025860755,
      "eval_loss": 2.4245753288269043,
      "eval_runtime": 21.9738,
      "eval_samples_per_second": 227.543,
      "eval_steps_per_second": 1.229,
      "step": 86565
    },
    {
      "epoch": 4.3018612291903064,
      "grad_norm": 0.19236032664775848,
      "learning_rate": 5.2742422416196325e-05,
      "loss": 1.3929,
      "step": 86570
    },
    {
      "epoch": 4.302403170355818,
      "grad_norm": 0.1677989512681961,
      "learning_rate": 5.272660431184717e-05,
      "loss": 1.3995,
      "step": 86580
    },
    {
      "epoch": 4.30294511152133,
      "grad_norm": 0.19151169061660767,
      "learning_rate": 5.2710786489116416e-05,
      "loss": 1.4002,
      "step": 86590
    },
    {
      "epoch": 4.303487052686843,
      "grad_norm": 0.1588166505098343,
      "learning_rate": 5.2694968949963485e-05,
      "loss": 1.4059,
      "step": 86600
    },
    {
      "epoch": 4.304028993852355,
      "grad_norm": 0.1966225802898407,
      "learning_rate": 5.267915169634779e-05,
      "loss": 1.4075,
      "step": 86610
    },
    {
      "epoch": 4.304570935017868,
      "grad_norm": 0.16342392563819885,
      "learning_rate": 5.266333473022873e-05,
      "loss": 1.4083,
      "step": 86620
    },
    {
      "epoch": 4.3051128761833795,
      "grad_norm": 0.1621856838464737,
      "learning_rate": 5.26475180535656e-05,
      "loss": 1.3987,
      "step": 86630
    },
    {
      "epoch": 4.305654817348891,
      "grad_norm": 0.17513424158096313,
      "learning_rate": 5.2631701668317726e-05,
      "loss": 1.4009,
      "step": 86640
    },
    {
      "epoch": 4.306196758514404,
      "grad_norm": 0.20303316414356232,
      "learning_rate": 5.261588557644437e-05,
      "loss": 1.3863,
      "step": 86650
    },
    {
      "epoch": 4.306305146747507,
      "eval_loss": 2.4262466430664062,
      "eval_runtime": 21.9658,
      "eval_samples_per_second": 227.627,
      "eval_steps_per_second": 1.229,
      "step": 86652
    },
    {
      "epoch": 4.306738699679916,
      "grad_norm": 0.1640668660402298,
      "learning_rate": 5.260006977990474e-05,
      "loss": 1.4091,
      "step": 86660
    },
    {
      "epoch": 4.307280640845428,
      "grad_norm": 0.16020847856998444,
      "learning_rate": 5.258425428065805e-05,
      "loss": 1.4089,
      "step": 86670
    },
    {
      "epoch": 4.307822582010941,
      "grad_norm": 0.20218387246131897,
      "learning_rate": 5.256843908066346e-05,
      "loss": 1.3933,
      "step": 86680
    },
    {
      "epoch": 4.3083645231764525,
      "grad_norm": 0.2258821278810501,
      "learning_rate": 5.2552624181880086e-05,
      "loss": 1.406,
      "step": 86690
    },
    {
      "epoch": 4.308906464341965,
      "grad_norm": 0.2921973168849945,
      "learning_rate": 5.253680958626699e-05,
      "loss": 1.3927,
      "step": 86700
    },
    {
      "epoch": 4.309448405507477,
      "grad_norm": 0.29356056451797485,
      "learning_rate": 5.252099529578323e-05,
      "loss": 1.398,
      "step": 86710
    },
    {
      "epoch": 4.309990346672989,
      "grad_norm": 0.16703596711158752,
      "learning_rate": 5.250518131238783e-05,
      "loss": 1.4071,
      "step": 86720
    },
    {
      "epoch": 4.310532287838502,
      "grad_norm": 0.1876375526189804,
      "learning_rate": 5.248936763803972e-05,
      "loss": 1.4027,
      "step": 86730
    },
    {
      "epoch": 4.311020034887463,
      "eval_loss": 2.4287726879119873,
      "eval_runtime": 21.9767,
      "eval_samples_per_second": 227.513,
      "eval_steps_per_second": 1.229,
      "step": 86739
    },
    {
      "epoch": 4.311074229004014,
      "grad_norm": 0.20849046111106873,
      "learning_rate": 5.2473554274697846e-05,
      "loss": 1.3907,
      "step": 86740
    },
    {
      "epoch": 4.311616170169526,
      "grad_norm": 0.24068142473697662,
      "learning_rate": 5.24577412243211e-05,
      "loss": 1.3999,
      "step": 86750
    },
    {
      "epoch": 4.312158111335038,
      "grad_norm": 0.3270260691642761,
      "learning_rate": 5.244192848886834e-05,
      "loss": 1.4041,
      "step": 86760
    },
    {
      "epoch": 4.31270005250055,
      "grad_norm": 0.2979717552661896,
      "learning_rate": 5.242611607029836e-05,
      "loss": 1.4006,
      "step": 86770
    },
    {
      "epoch": 4.313241993666063,
      "grad_norm": 0.16698554158210754,
      "learning_rate": 5.241030397056996e-05,
      "loss": 1.4058,
      "step": 86780
    },
    {
      "epoch": 4.313783934831575,
      "grad_norm": 0.16404001414775848,
      "learning_rate": 5.2394492191641854e-05,
      "loss": 1.4078,
      "step": 86790
    },
    {
      "epoch": 4.314325875997087,
      "grad_norm": 0.16933251917362213,
      "learning_rate": 5.237868073547274e-05,
      "loss": 1.3915,
      "step": 86800
    },
    {
      "epoch": 4.314867817162599,
      "grad_norm": 0.18435190618038177,
      "learning_rate": 5.2362869604021304e-05,
      "loss": 1.3983,
      "step": 86810
    },
    {
      "epoch": 4.315409758328111,
      "grad_norm": 0.1849733144044876,
      "learning_rate": 5.2347058799246104e-05,
      "loss": 1.3964,
      "step": 86820
    },
    {
      "epoch": 4.315734923027419,
      "eval_loss": 2.4305026531219482,
      "eval_runtime": 21.9708,
      "eval_samples_per_second": 227.574,
      "eval_steps_per_second": 1.229,
      "step": 86826
    },
    {
      "epoch": 4.315951699493624,
      "grad_norm": 0.2581421732902527,
      "learning_rate": 5.233124832310574e-05,
      "loss": 1.3981,
      "step": 86830
    },
    {
      "epoch": 4.316493640659136,
      "grad_norm": 0.19301174581050873,
      "learning_rate": 5.2315438177558754e-05,
      "loss": 1.3909,
      "step": 86840
    },
    {
      "epoch": 4.317035581824648,
      "grad_norm": 0.3168368637561798,
      "learning_rate": 5.229962836456364e-05,
      "loss": 1.4017,
      "step": 86850
    },
    {
      "epoch": 4.3175775229901605,
      "grad_norm": 0.30351701378822327,
      "learning_rate": 5.2283818886078827e-05,
      "loss": 1.3929,
      "step": 86860
    },
    {
      "epoch": 4.318119464155672,
      "grad_norm": 0.2167241871356964,
      "learning_rate": 5.226800974406274e-05,
      "loss": 1.4048,
      "step": 86870
    },
    {
      "epoch": 4.318661405321185,
      "grad_norm": 0.24982495605945587,
      "learning_rate": 5.225220094047375e-05,
      "loss": 1.4042,
      "step": 86880
    },
    {
      "epoch": 4.319203346486697,
      "grad_norm": 0.15455639362335205,
      "learning_rate": 5.2236392477270165e-05,
      "loss": 1.4019,
      "step": 86890
    },
    {
      "epoch": 4.319745287652209,
      "grad_norm": 0.20189005136489868,
      "learning_rate": 5.22205843564103e-05,
      "loss": 1.3887,
      "step": 86900
    },
    {
      "epoch": 4.320287228817722,
      "grad_norm": 0.17087921500205994,
      "learning_rate": 5.2204776579852375e-05,
      "loss": 1.403,
      "step": 86910
    },
    {
      "epoch": 4.320449811167375,
      "eval_loss": 2.429542303085327,
      "eval_runtime": 21.9702,
      "eval_samples_per_second": 227.581,
      "eval_steps_per_second": 1.229,
      "step": 86913
    },
    {
      "epoch": 4.3208291699832335,
      "grad_norm": 0.19037020206451416,
      "learning_rate": 5.218896914955459e-05,
      "loss": 1.3954,
      "step": 86920
    },
    {
      "epoch": 4.321371111148746,
      "grad_norm": 0.29596278071403503,
      "learning_rate": 5.217316206747509e-05,
      "loss": 1.3994,
      "step": 86930
    },
    {
      "epoch": 4.321913052314258,
      "grad_norm": 0.2806200385093689,
      "learning_rate": 5.2157355335572024e-05,
      "loss": 1.4008,
      "step": 86940
    },
    {
      "epoch": 4.32245499347977,
      "grad_norm": 0.19318336248397827,
      "learning_rate": 5.214154895580342e-05,
      "loss": 1.4017,
      "step": 86950
    },
    {
      "epoch": 4.322996934645283,
      "grad_norm": 0.266025185585022,
      "learning_rate": 5.2125742930127316e-05,
      "loss": 1.3989,
      "step": 86960
    },
    {
      "epoch": 4.323538875810795,
      "grad_norm": 0.20118410885334015,
      "learning_rate": 5.210993726050173e-05,
      "loss": 1.4009,
      "step": 86970
    },
    {
      "epoch": 4.324080816976307,
      "grad_norm": 0.3275969922542572,
      "learning_rate": 5.209413194888453e-05,
      "loss": 1.3998,
      "step": 86980
    },
    {
      "epoch": 4.324622758141819,
      "grad_norm": 0.22696785628795624,
      "learning_rate": 5.2078326997233665e-05,
      "loss": 1.4047,
      "step": 86990
    },
    {
      "epoch": 4.325164699307331,
      "grad_norm": 0.1600492149591446,
      "learning_rate": 5.206252240750695e-05,
      "loss": 1.392,
      "step": 87000
    },
    {
      "epoch": 4.325164699307331,
      "eval_loss": 2.4212775230407715,
      "eval_runtime": 21.9652,
      "eval_samples_per_second": 227.633,
      "eval_steps_per_second": 1.229,
      "step": 87000
    },
    {
      "epoch": 4.325706640472844,
      "grad_norm": 0.17092597484588623,
      "learning_rate": 5.204671818166225e-05,
      "loss": 1.3931,
      "step": 87010
    },
    {
      "epoch": 4.326248581638356,
      "grad_norm": 0.3140299320220947,
      "learning_rate": 5.203091432165724e-05,
      "loss": 1.3986,
      "step": 87020
    },
    {
      "epoch": 4.3267905228038686,
      "grad_norm": 0.2115614265203476,
      "learning_rate": 5.201511082944968e-05,
      "loss": 1.4102,
      "step": 87030
    },
    {
      "epoch": 4.32733246396938,
      "grad_norm": 0.15958638489246368,
      "learning_rate": 5.199930770699725e-05,
      "loss": 1.4012,
      "step": 87040
    },
    {
      "epoch": 4.327874405134892,
      "grad_norm": 0.15782247483730316,
      "learning_rate": 5.198350495625753e-05,
      "loss": 1.4026,
      "step": 87050
    },
    {
      "epoch": 4.328416346300405,
      "grad_norm": 0.32305723428726196,
      "learning_rate": 5.1967702579188125e-05,
      "loss": 1.3988,
      "step": 87060
    },
    {
      "epoch": 4.328958287465917,
      "grad_norm": 0.23694272339344025,
      "learning_rate": 5.1951900577746584e-05,
      "loss": 1.4041,
      "step": 87070
    },
    {
      "epoch": 4.329500228631429,
      "grad_norm": 0.21021704375743866,
      "learning_rate": 5.1936098953890335e-05,
      "loss": 1.3879,
      "step": 87080
    },
    {
      "epoch": 4.329879587447287,
      "eval_loss": 2.4319331645965576,
      "eval_runtime": 21.9684,
      "eval_samples_per_second": 227.6,
      "eval_steps_per_second": 1.229,
      "step": 87087
    },
    {
      "epoch": 4.330042169796942,
      "grad_norm": 0.2358713448047638,
      "learning_rate": 5.192029770957685e-05,
      "loss": 1.3971,
      "step": 87090
    },
    {
      "epoch": 4.330584110962453,
      "grad_norm": 0.19418524205684662,
      "learning_rate": 5.1904496846763536e-05,
      "loss": 1.4053,
      "step": 87100
    },
    {
      "epoch": 4.331126052127966,
      "grad_norm": 0.21557055413722992,
      "learning_rate": 5.1888696367407696e-05,
      "loss": 1.408,
      "step": 87110
    },
    {
      "epoch": 4.331667993293478,
      "grad_norm": 0.15955239534378052,
      "learning_rate": 5.1872896273466645e-05,
      "loss": 1.4037,
      "step": 87120
    },
    {
      "epoch": 4.33220993445899,
      "grad_norm": 0.24472442269325256,
      "learning_rate": 5.1857096566897614e-05,
      "loss": 1.3886,
      "step": 87130
    },
    {
      "epoch": 4.332751875624503,
      "grad_norm": 0.24180065095424652,
      "learning_rate": 5.184129724965784e-05,
      "loss": 1.3925,
      "step": 87140
    },
    {
      "epoch": 4.333293816790015,
      "grad_norm": 0.3072583079338074,
      "learning_rate": 5.1825498323704423e-05,
      "loss": 1.4073,
      "step": 87150
    },
    {
      "epoch": 4.333835757955527,
      "grad_norm": 0.36256688833236694,
      "learning_rate": 5.1809699790994505e-05,
      "loss": 1.4036,
      "step": 87160
    },
    {
      "epoch": 4.334377699121039,
      "grad_norm": 0.3161357641220093,
      "learning_rate": 5.179390165348514e-05,
      "loss": 1.3945,
      "step": 87170
    },
    {
      "epoch": 4.334594475587244,
      "eval_loss": 2.4241182804107666,
      "eval_runtime": 21.9706,
      "eval_samples_per_second": 227.577,
      "eval_steps_per_second": 1.229,
      "step": 87174
    },
    {
      "epoch": 4.334919640286551,
      "grad_norm": 0.1909865289926529,
      "learning_rate": 5.177810391313329e-05,
      "loss": 1.4059,
      "step": 87180
    },
    {
      "epoch": 4.335461581452064,
      "grad_norm": 0.2220858484506607,
      "learning_rate": 5.176230657189596e-05,
      "loss": 1.4005,
      "step": 87190
    },
    {
      "epoch": 4.336003522617576,
      "grad_norm": 0.23918281495571136,
      "learning_rate": 5.1746509631730035e-05,
      "loss": 1.3936,
      "step": 87200
    },
    {
      "epoch": 4.336545463783088,
      "grad_norm": 0.25223907828330994,
      "learning_rate": 5.173071309459236e-05,
      "loss": 1.4003,
      "step": 87210
    },
    {
      "epoch": 4.3370874049486,
      "grad_norm": 0.1692775934934616,
      "learning_rate": 5.171491696243976e-05,
      "loss": 1.4004,
      "step": 87220
    },
    {
      "epoch": 4.337629346114112,
      "grad_norm": 0.22520475089550018,
      "learning_rate": 5.1699121237228984e-05,
      "loss": 1.4074,
      "step": 87230
    },
    {
      "epoch": 4.338171287279625,
      "grad_norm": 0.23879382014274597,
      "learning_rate": 5.168332592091673e-05,
      "loss": 1.3965,
      "step": 87240
    },
    {
      "epoch": 4.338713228445137,
      "grad_norm": 0.18485189974308014,
      "learning_rate": 5.166753101545967e-05,
      "loss": 1.4024,
      "step": 87250
    },
    {
      "epoch": 4.339255169610649,
      "grad_norm": 0.2033102959394455,
      "learning_rate": 5.165173652281441e-05,
      "loss": 1.4067,
      "step": 87260
    },
    {
      "epoch": 4.3393093637272,
      "eval_loss": 2.4278345108032227,
      "eval_runtime": 21.973,
      "eval_samples_per_second": 227.552,
      "eval_steps_per_second": 1.229,
      "step": 87261
    },
    {
      "epoch": 4.3397971107761615,
      "grad_norm": 0.16319087147712708,
      "learning_rate": 5.163594244493748e-05,
      "loss": 1.4038,
      "step": 87270
    },
    {
      "epoch": 4.340339051941673,
      "grad_norm": 0.18227989971637726,
      "learning_rate": 5.1620148783785385e-05,
      "loss": 1.4148,
      "step": 87280
    },
    {
      "epoch": 4.340880993107186,
      "grad_norm": 0.17856183648109436,
      "learning_rate": 5.1604355541314586e-05,
      "loss": 1.4062,
      "step": 87290
    },
    {
      "epoch": 4.341422934272698,
      "grad_norm": 0.21139036118984222,
      "learning_rate": 5.15885627194815e-05,
      "loss": 1.399,
      "step": 87300
    },
    {
      "epoch": 4.34196487543821,
      "grad_norm": 0.3597468435764313,
      "learning_rate": 5.157277032024245e-05,
      "loss": 1.3997,
      "step": 87310
    },
    {
      "epoch": 4.342506816603723,
      "grad_norm": 0.18101923167705536,
      "learning_rate": 5.155697834555372e-05,
      "loss": 1.3952,
      "step": 87320
    },
    {
      "epoch": 4.3430487577692345,
      "grad_norm": 0.20467126369476318,
      "learning_rate": 5.154118679737158e-05,
      "loss": 1.4021,
      "step": 87330
    },
    {
      "epoch": 4.343590698934747,
      "grad_norm": 0.2586832046508789,
      "learning_rate": 5.152539567765219e-05,
      "loss": 1.4073,
      "step": 87340
    },
    {
      "epoch": 4.344024251867157,
      "eval_loss": 2.429697036743164,
      "eval_runtime": 21.9753,
      "eval_samples_per_second": 227.528,
      "eval_steps_per_second": 1.229,
      "step": 87348
    },
    {
      "epoch": 4.344132640100259,
      "grad_norm": 0.20841944217681885,
      "learning_rate": 5.150960498835169e-05,
      "loss": 1.3959,
      "step": 87350
    },
    {
      "epoch": 4.344674581265771,
      "grad_norm": 0.15651927888393402,
      "learning_rate": 5.149381473142621e-05,
      "loss": 1.4009,
      "step": 87360
    },
    {
      "epoch": 4.345216522431284,
      "grad_norm": 0.22933441400527954,
      "learning_rate": 5.147802490883169e-05,
      "loss": 1.3964,
      "step": 87370
    },
    {
      "epoch": 4.345758463596796,
      "grad_norm": 0.2737453281879425,
      "learning_rate": 5.146223552252416e-05,
      "loss": 1.3928,
      "step": 87380
    },
    {
      "epoch": 4.346300404762308,
      "grad_norm": 0.22717493772506714,
      "learning_rate": 5.144644657445957e-05,
      "loss": 1.3937,
      "step": 87390
    },
    {
      "epoch": 4.34684234592782,
      "grad_norm": 0.16106519103050232,
      "learning_rate": 5.143065806659373e-05,
      "loss": 1.4036,
      "step": 87400
    },
    {
      "epoch": 4.347384287093332,
      "grad_norm": 0.2053665816783905,
      "learning_rate": 5.141487000088245e-05,
      "loss": 1.4002,
      "step": 87410
    },
    {
      "epoch": 4.347926228258845,
      "grad_norm": 0.14804090559482574,
      "learning_rate": 5.139908237928155e-05,
      "loss": 1.3886,
      "step": 87420
    },
    {
      "epoch": 4.348468169424357,
      "grad_norm": 0.2843748927116394,
      "learning_rate": 5.138329520374666e-05,
      "loss": 1.4034,
      "step": 87430
    },
    {
      "epoch": 4.348739140007113,
      "eval_loss": 2.4298503398895264,
      "eval_runtime": 21.9707,
      "eval_samples_per_second": 227.576,
      "eval_steps_per_second": 1.229,
      "step": 87435
    },
    {
      "epoch": 4.3490101105898695,
      "grad_norm": 0.15882396697998047,
      "learning_rate": 5.1367508476233474e-05,
      "loss": 1.3942,
      "step": 87440
    },
    {
      "epoch": 4.349552051755381,
      "grad_norm": 0.23637013137340546,
      "learning_rate": 5.135172219869755e-05,
      "loss": 1.4045,
      "step": 87450
    },
    {
      "epoch": 4.350093992920893,
      "grad_norm": 0.20774583518505096,
      "learning_rate": 5.1335936373094475e-05,
      "loss": 1.3984,
      "step": 87460
    },
    {
      "epoch": 4.350635934086406,
      "grad_norm": 0.21993571519851685,
      "learning_rate": 5.132015100137966e-05,
      "loss": 1.4071,
      "step": 87470
    },
    {
      "epoch": 4.351177875251918,
      "grad_norm": 0.16270707547664642,
      "learning_rate": 5.130436608550856e-05,
      "loss": 1.3982,
      "step": 87480
    },
    {
      "epoch": 4.35171981641743,
      "grad_norm": 0.16741152107715607,
      "learning_rate": 5.128858162743658e-05,
      "loss": 1.3972,
      "step": 87490
    },
    {
      "epoch": 4.3522617575829425,
      "grad_norm": 0.19168831408023834,
      "learning_rate": 5.1272797629118976e-05,
      "loss": 1.4038,
      "step": 87500
    },
    {
      "epoch": 4.352803698748454,
      "grad_norm": 0.18986192345619202,
      "learning_rate": 5.1257014092511e-05,
      "loss": 1.4015,
      "step": 87510
    },
    {
      "epoch": 4.353345639913967,
      "grad_norm": 0.15182600915431976,
      "learning_rate": 5.12412310195679e-05,
      "loss": 1.3979,
      "step": 87520
    },
    {
      "epoch": 4.35345402814707,
      "eval_loss": 2.431755542755127,
      "eval_runtime": 21.9949,
      "eval_samples_per_second": 227.325,
      "eval_steps_per_second": 1.228,
      "step": 87522
    },
    {
      "epoch": 4.353887581079479,
      "grad_norm": 0.23063354194164276,
      "learning_rate": 5.122544841224476e-05,
      "loss": 1.3934,
      "step": 87530
    },
    {
      "epoch": 4.354429522244991,
      "grad_norm": 0.2470499873161316,
      "learning_rate": 5.120966627249669e-05,
      "loss": 1.4009,
      "step": 87540
    },
    {
      "epoch": 4.354971463410504,
      "grad_norm": 0.15349188446998596,
      "learning_rate": 5.119388460227872e-05,
      "loss": 1.3953,
      "step": 87550
    },
    {
      "epoch": 4.3555134045760155,
      "grad_norm": 0.2129431664943695,
      "learning_rate": 5.1178103403545794e-05,
      "loss": 1.3964,
      "step": 87560
    },
    {
      "epoch": 4.356055345741528,
      "grad_norm": 0.16350790858268738,
      "learning_rate": 5.116232267825279e-05,
      "loss": 1.3981,
      "step": 87570
    },
    {
      "epoch": 4.35659728690704,
      "grad_norm": 0.19389735162258148,
      "learning_rate": 5.114654242835465e-05,
      "loss": 1.4019,
      "step": 87580
    },
    {
      "epoch": 4.357139228072552,
      "grad_norm": 0.2006743848323822,
      "learning_rate": 5.113076265580606e-05,
      "loss": 1.3975,
      "step": 87590
    },
    {
      "epoch": 4.357681169238065,
      "grad_norm": 0.1707877218723297,
      "learning_rate": 5.111498336256181e-05,
      "loss": 1.403,
      "step": 87600
    },
    {
      "epoch": 4.358168916287026,
      "eval_loss": 2.4279067516326904,
      "eval_runtime": 21.9698,
      "eval_samples_per_second": 227.585,
      "eval_steps_per_second": 1.229,
      "step": 87609
    },
    {
      "epoch": 4.358223110403577,
      "grad_norm": 0.2142249345779419,
      "learning_rate": 5.109920455057655e-05,
      "loss": 1.403,
      "step": 87610
    },
    {
      "epoch": 4.358765051569089,
      "grad_norm": 0.1643165647983551,
      "learning_rate": 5.10834262218049e-05,
      "loss": 1.3975,
      "step": 87620
    },
    {
      "epoch": 4.359306992734601,
      "grad_norm": 0.20058096945285797,
      "learning_rate": 5.106764837820141e-05,
      "loss": 1.4049,
      "step": 87630
    },
    {
      "epoch": 4.359848933900113,
      "grad_norm": 0.17463991045951843,
      "learning_rate": 5.1051871021720546e-05,
      "loss": 1.4036,
      "step": 87640
    },
    {
      "epoch": 4.360390875065626,
      "grad_norm": 0.20207591354846954,
      "learning_rate": 5.103609415431678e-05,
      "loss": 1.4025,
      "step": 87650
    },
    {
      "epoch": 4.360932816231138,
      "grad_norm": 0.1925676167011261,
      "learning_rate": 5.102031777794446e-05,
      "loss": 1.3931,
      "step": 87660
    },
    {
      "epoch": 4.36147475739665,
      "grad_norm": 0.2542779743671417,
      "learning_rate": 5.100454189455787e-05,
      "loss": 1.3996,
      "step": 87670
    },
    {
      "epoch": 4.3620166985621625,
      "grad_norm": 0.21328918635845184,
      "learning_rate": 5.0988766506111316e-05,
      "loss": 1.398,
      "step": 87680
    },
    {
      "epoch": 4.362558639727674,
      "grad_norm": 0.23086091876029968,
      "learning_rate": 5.097299161455893e-05,
      "loss": 1.4014,
      "step": 87690
    },
    {
      "epoch": 4.362883804426982,
      "eval_loss": 2.4321718215942383,
      "eval_runtime": 21.9729,
      "eval_samples_per_second": 227.553,
      "eval_steps_per_second": 1.229,
      "step": 87696
    },
    {
      "epoch": 4.363100580893187,
      "grad_norm": 0.15713347494602203,
      "learning_rate": 5.095721722185487e-05,
      "loss": 1.3932,
      "step": 87700
    },
    {
      "epoch": 4.363642522058699,
      "grad_norm": 0.16743972897529602,
      "learning_rate": 5.0941443329953185e-05,
      "loss": 1.399,
      "step": 87710
    },
    {
      "epoch": 4.364184463224211,
      "grad_norm": 0.15407240390777588,
      "learning_rate": 5.0925669940807885e-05,
      "loss": 1.4004,
      "step": 87720
    },
    {
      "epoch": 4.364726404389724,
      "grad_norm": 0.1776094287633896,
      "learning_rate": 5.090989705637289e-05,
      "loss": 1.3983,
      "step": 87730
    },
    {
      "epoch": 4.3652683455552355,
      "grad_norm": 0.16441775858402252,
      "learning_rate": 5.0894124678602116e-05,
      "loss": 1.3959,
      "step": 87740
    },
    {
      "epoch": 4.365810286720748,
      "grad_norm": 0.21521979570388794,
      "learning_rate": 5.0878352809449336e-05,
      "loss": 1.3909,
      "step": 87750
    },
    {
      "epoch": 4.36635222788626,
      "grad_norm": 0.1725737303495407,
      "learning_rate": 5.086258145086831e-05,
      "loss": 1.3974,
      "step": 87760
    },
    {
      "epoch": 4.366894169051772,
      "grad_norm": 0.1918652206659317,
      "learning_rate": 5.084681060481271e-05,
      "loss": 1.3925,
      "step": 87770
    },
    {
      "epoch": 4.367436110217285,
      "grad_norm": 0.16911828517913818,
      "learning_rate": 5.083104027323623e-05,
      "loss": 1.391,
      "step": 87780
    },
    {
      "epoch": 4.367598692566938,
      "eval_loss": 2.4334969520568848,
      "eval_runtime": 21.9702,
      "eval_samples_per_second": 227.581,
      "eval_steps_per_second": 1.229,
      "step": 87783
    },
    {
      "epoch": 4.367978051382797,
      "grad_norm": 0.16556398570537567,
      "learning_rate": 5.081527045809236e-05,
      "loss": 1.3964,
      "step": 87790
    },
    {
      "epoch": 4.368519992548309,
      "grad_norm": 0.17277011275291443,
      "learning_rate": 5.0799501161334606e-05,
      "loss": 1.3978,
      "step": 87800
    },
    {
      "epoch": 4.369061933713821,
      "grad_norm": 0.17669574916362762,
      "learning_rate": 5.078373238491642e-05,
      "loss": 1.4088,
      "step": 87810
    },
    {
      "epoch": 4.369603874879333,
      "grad_norm": 0.19774076342582703,
      "learning_rate": 5.076796413079116e-05,
      "loss": 1.4003,
      "step": 87820
    },
    {
      "epoch": 4.370145816044846,
      "grad_norm": 0.22697632014751434,
      "learning_rate": 5.075219640091212e-05,
      "loss": 1.3944,
      "step": 87830
    },
    {
      "epoch": 4.370687757210358,
      "grad_norm": 0.19095872342586517,
      "learning_rate": 5.0736429197232574e-05,
      "loss": 1.3905,
      "step": 87840
    },
    {
      "epoch": 4.3712296983758705,
      "grad_norm": 0.2845691740512848,
      "learning_rate": 5.072066252170565e-05,
      "loss": 1.4046,
      "step": 87850
    },
    {
      "epoch": 4.371771639541382,
      "grad_norm": 0.20539964735507965,
      "learning_rate": 5.070489637628447e-05,
      "loss": 1.3944,
      "step": 87860
    },
    {
      "epoch": 4.372313580706894,
      "grad_norm": 0.22066408395767212,
      "learning_rate": 5.06891307629221e-05,
      "loss": 1.3861,
      "step": 87870
    },
    {
      "epoch": 4.372313580706894,
      "eval_loss": 2.430873394012451,
      "eval_runtime": 22.02,
      "eval_samples_per_second": 227.067,
      "eval_steps_per_second": 1.226,
      "step": 87870
    },
    {
      "epoch": 4.372855521872407,
      "grad_norm": 0.25216028094291687,
      "learning_rate": 5.067336568357147e-05,
      "loss": 1.3944,
      "step": 87880
    },
    {
      "epoch": 4.373397463037919,
      "grad_norm": 0.16091406345367432,
      "learning_rate": 5.065760114018553e-05,
      "loss": 1.4024,
      "step": 87890
    },
    {
      "epoch": 4.373939404203432,
      "grad_norm": 0.19635522365570068,
      "learning_rate": 5.0641837134717095e-05,
      "loss": 1.409,
      "step": 87900
    },
    {
      "epoch": 4.3744813453689435,
      "grad_norm": 0.24364635348320007,
      "learning_rate": 5.062607366911897e-05,
      "loss": 1.3902,
      "step": 87910
    },
    {
      "epoch": 4.375023286534455,
      "grad_norm": 0.24988219141960144,
      "learning_rate": 5.0610310745343836e-05,
      "loss": 1.4047,
      "step": 87920
    },
    {
      "epoch": 4.375565227699968,
      "grad_norm": 0.18812812864780426,
      "learning_rate": 5.0594548365344354e-05,
      "loss": 1.3898,
      "step": 87930
    },
    {
      "epoch": 4.37610716886548,
      "grad_norm": 0.3591182231903076,
      "learning_rate": 5.057878653107311e-05,
      "loss": 1.4025,
      "step": 87940
    },
    {
      "epoch": 4.376649110030992,
      "grad_norm": 0.23308515548706055,
      "learning_rate": 5.0563025244482574e-05,
      "loss": 1.4003,
      "step": 87950
    },
    {
      "epoch": 4.37702846884685,
      "eval_loss": 2.4251065254211426,
      "eval_runtime": 21.9772,
      "eval_samples_per_second": 227.509,
      "eval_steps_per_second": 1.229,
      "step": 87957
    },
    {
      "epoch": 4.377191051196505,
      "grad_norm": 0.17095594108104706,
      "learning_rate": 5.054726450752521e-05,
      "loss": 1.4007,
      "step": 87960
    },
    {
      "epoch": 4.3777329923620165,
      "grad_norm": 0.19432257115840912,
      "learning_rate": 5.05315043221534e-05,
      "loss": 1.4043,
      "step": 87970
    },
    {
      "epoch": 4.378274933527529,
      "grad_norm": 0.18219472467899323,
      "learning_rate": 5.051574469031942e-05,
      "loss": 1.4022,
      "step": 87980
    },
    {
      "epoch": 4.378816874693041,
      "grad_norm": 0.17361074686050415,
      "learning_rate": 5.049998561397552e-05,
      "loss": 1.3977,
      "step": 87990
    },
    {
      "epoch": 4.379358815858553,
      "grad_norm": 0.21151654422283173,
      "learning_rate": 5.0484227095073865e-05,
      "loss": 1.3887,
      "step": 88000
    },
    {
      "epoch": 4.379900757024066,
      "grad_norm": 0.16382147371768951,
      "learning_rate": 5.0468469135566554e-05,
      "loss": 1.3901,
      "step": 88010
    },
    {
      "epoch": 4.380442698189578,
      "grad_norm": 0.1616712063550949,
      "learning_rate": 5.045271173740562e-05,
      "loss": 1.4068,
      "step": 88020
    },
    {
      "epoch": 4.38098463935509,
      "grad_norm": 0.17013293504714966,
      "learning_rate": 5.043695490254302e-05,
      "loss": 1.3819,
      "step": 88030
    },
    {
      "epoch": 4.381526580520602,
      "grad_norm": 0.1871418058872223,
      "learning_rate": 5.0421198632930624e-05,
      "loss": 1.4056,
      "step": 88040
    },
    {
      "epoch": 4.381743356986807,
      "eval_loss": 2.4317269325256348,
      "eval_runtime": 21.9684,
      "eval_samples_per_second": 227.6,
      "eval_steps_per_second": 1.229,
      "step": 88044
    },
    {
      "epoch": 4.382068521686114,
      "grad_norm": 0.2265293151140213,
      "learning_rate": 5.0405442930520253e-05,
      "loss": 1.3987,
      "step": 88050
    },
    {
      "epoch": 4.382610462851627,
      "grad_norm": 0.26006123423576355,
      "learning_rate": 5.0389687797263664e-05,
      "loss": 1.397,
      "step": 88060
    },
    {
      "epoch": 4.383152404017139,
      "grad_norm": 0.19830361008644104,
      "learning_rate": 5.037393323511256e-05,
      "loss": 1.421,
      "step": 88070
    },
    {
      "epoch": 4.383694345182651,
      "grad_norm": 0.2708401083946228,
      "learning_rate": 5.03581792460185e-05,
      "loss": 1.3956,
      "step": 88080
    },
    {
      "epoch": 4.384236286348163,
      "grad_norm": 0.18328110873699188,
      "learning_rate": 5.034242583193305e-05,
      "loss": 1.3926,
      "step": 88090
    },
    {
      "epoch": 4.384778227513675,
      "grad_norm": 0.19660696387290955,
      "learning_rate": 5.032667299480767e-05,
      "loss": 1.386,
      "step": 88100
    },
    {
      "epoch": 4.385320168679188,
      "grad_norm": 0.21113744378089905,
      "learning_rate": 5.0310920736593737e-05,
      "loss": 1.4048,
      "step": 88110
    },
    {
      "epoch": 4.3858621098447,
      "grad_norm": 0.22649426758289337,
      "learning_rate": 5.029516905924258e-05,
      "loss": 1.3993,
      "step": 88120
    },
    {
      "epoch": 4.386404051010212,
      "grad_norm": 0.16807417571544647,
      "learning_rate": 5.02794179647055e-05,
      "loss": 1.3882,
      "step": 88130
    },
    {
      "epoch": 4.3864582451267635,
      "eval_loss": 2.411639451980591,
      "eval_runtime": 21.9652,
      "eval_samples_per_second": 227.633,
      "eval_steps_per_second": 1.229,
      "step": 88131
    },
    {
      "epoch": 4.386945992175725,
      "grad_norm": 0.18633247911930084,
      "learning_rate": 5.026366745493359e-05,
      "loss": 1.4067,
      "step": 88140
    },
    {
      "epoch": 4.387487933341236,
      "grad_norm": 0.16578739881515503,
      "learning_rate": 5.0247917531877985e-05,
      "loss": 1.3962,
      "step": 88150
    },
    {
      "epoch": 4.388029874506749,
      "grad_norm": 0.23297452926635742,
      "learning_rate": 5.0232168197489746e-05,
      "loss": 1.4044,
      "step": 88160
    },
    {
      "epoch": 4.388571815672261,
      "grad_norm": 0.24161213636398315,
      "learning_rate": 5.02164194537198e-05,
      "loss": 1.4081,
      "step": 88170
    },
    {
      "epoch": 4.389113756837773,
      "grad_norm": 0.20169183611869812,
      "learning_rate": 5.020067130251904e-05,
      "loss": 1.3996,
      "step": 88180
    },
    {
      "epoch": 4.389655698003286,
      "grad_norm": 0.21233811974525452,
      "learning_rate": 5.01849237458383e-05,
      "loss": 1.4046,
      "step": 88190
    },
    {
      "epoch": 4.390197639168798,
      "grad_norm": 0.15420041978359222,
      "learning_rate": 5.016917678562828e-05,
      "loss": 1.4007,
      "step": 88200
    },
    {
      "epoch": 4.39073958033431,
      "grad_norm": 0.1760631501674652,
      "learning_rate": 5.0153430423839676e-05,
      "loss": 1.3998,
      "step": 88210
    },
    {
      "epoch": 4.39117313326672,
      "eval_loss": 2.422487258911133,
      "eval_runtime": 21.9702,
      "eval_samples_per_second": 227.581,
      "eval_steps_per_second": 1.229,
      "step": 88218
    },
    {
      "epoch": 4.391281521499822,
      "grad_norm": 0.18712304532527924,
      "learning_rate": 5.013768466242307e-05,
      "loss": 1.403,
      "step": 88220
    },
    {
      "epoch": 4.391823462665334,
      "grad_norm": 0.1596558541059494,
      "learning_rate": 5.0121939503328985e-05,
      "loss": 1.3987,
      "step": 88230
    },
    {
      "epoch": 4.392365403830847,
      "grad_norm": 0.1593562513589859,
      "learning_rate": 5.010619494850785e-05,
      "loss": 1.4045,
      "step": 88240
    },
    {
      "epoch": 4.392907344996359,
      "grad_norm": 0.18083150684833527,
      "learning_rate": 5.0090450999910035e-05,
      "loss": 1.3972,
      "step": 88250
    },
    {
      "epoch": 4.3934492861618715,
      "grad_norm": 0.16882987320423126,
      "learning_rate": 5.007470765948584e-05,
      "loss": 1.4041,
      "step": 88260
    },
    {
      "epoch": 4.393991227327383,
      "grad_norm": 0.17636427283287048,
      "learning_rate": 5.005896492918547e-05,
      "loss": 1.3923,
      "step": 88270
    },
    {
      "epoch": 4.394533168492895,
      "grad_norm": 0.16823971271514893,
      "learning_rate": 5.004322281095907e-05,
      "loss": 1.395,
      "step": 88280
    },
    {
      "epoch": 4.395075109658408,
      "grad_norm": 0.1983443945646286,
      "learning_rate": 5.002748130675672e-05,
      "loss": 1.3915,
      "step": 88290
    },
    {
      "epoch": 4.39561705082392,
      "grad_norm": 0.16008202731609344,
      "learning_rate": 5.001174041852839e-05,
      "loss": 1.3959,
      "step": 88300
    },
    {
      "epoch": 4.395888021406676,
      "eval_loss": 2.420740842819214,
      "eval_runtime": 21.9717,
      "eval_samples_per_second": 227.565,
      "eval_steps_per_second": 1.229,
      "step": 88305
    },
    {
      "epoch": 4.396158991989433,
      "grad_norm": 0.1860068142414093,
      "learning_rate": 4.9996000148224e-05,
      "loss": 1.3967,
      "step": 88310
    },
    {
      "epoch": 4.3967009331549445,
      "grad_norm": 0.2595568001270294,
      "learning_rate": 4.99802604977934e-05,
      "loss": 1.3915,
      "step": 88320
    },
    {
      "epoch": 4.397242874320456,
      "grad_norm": 0.16008210182189941,
      "learning_rate": 4.996452146918632e-05,
      "loss": 1.396,
      "step": 88330
    },
    {
      "epoch": 4.397784815485969,
      "grad_norm": 0.16884075105190277,
      "learning_rate": 4.994878306435245e-05,
      "loss": 1.3935,
      "step": 88340
    },
    {
      "epoch": 4.398326756651481,
      "grad_norm": 0.17851825058460236,
      "learning_rate": 4.993304528524143e-05,
      "loss": 1.3962,
      "step": 88350
    },
    {
      "epoch": 4.398868697816993,
      "grad_norm": 0.16581206023693085,
      "learning_rate": 4.9917308133802745e-05,
      "loss": 1.399,
      "step": 88360
    },
    {
      "epoch": 4.399410638982506,
      "grad_norm": 0.2509933412075043,
      "learning_rate": 4.9901571611985855e-05,
      "loss": 1.3986,
      "step": 88370
    },
    {
      "epoch": 4.3999525801480175,
      "grad_norm": 0.1800008863210678,
      "learning_rate": 4.988583572174015e-05,
      "loss": 1.4032,
      "step": 88380
    },
    {
      "epoch": 4.40049452131353,
      "grad_norm": 0.19821348786354065,
      "learning_rate": 4.987010046501491e-05,
      "loss": 1.3933,
      "step": 88390
    },
    {
      "epoch": 4.400602909546633,
      "eval_loss": 2.4244353771209717,
      "eval_runtime": 22.0054,
      "eval_samples_per_second": 227.217,
      "eval_steps_per_second": 1.227,
      "step": 88392
    },
    {
      "epoch": 4.401036462479042,
      "grad_norm": 0.18530848622322083,
      "learning_rate": 4.9854365843759354e-05,
      "loss": 1.405,
      "step": 88400
    },
    {
      "epoch": 4.401578403644554,
      "grad_norm": 0.1724652349948883,
      "learning_rate": 4.983863185992261e-05,
      "loss": 1.3891,
      "step": 88410
    },
    {
      "epoch": 4.402120344810067,
      "grad_norm": 0.2382229119539261,
      "learning_rate": 4.982289851545376e-05,
      "loss": 1.3989,
      "step": 88420
    },
    {
      "epoch": 4.402662285975579,
      "grad_norm": 0.22667750716209412,
      "learning_rate": 4.980716581230176e-05,
      "loss": 1.3992,
      "step": 88430
    },
    {
      "epoch": 4.403204227141091,
      "grad_norm": 0.2111603170633316,
      "learning_rate": 4.9791433752415494e-05,
      "loss": 1.3973,
      "step": 88440
    },
    {
      "epoch": 4.403746168306603,
      "grad_norm": 0.16338708996772766,
      "learning_rate": 4.977570233774382e-05,
      "loss": 1.4002,
      "step": 88450
    },
    {
      "epoch": 4.404288109472115,
      "grad_norm": 0.27522462606430054,
      "learning_rate": 4.9759971570235454e-05,
      "loss": 1.4011,
      "step": 88460
    },
    {
      "epoch": 4.404830050637628,
      "grad_norm": 0.2958241105079651,
      "learning_rate": 4.9744241451839056e-05,
      "loss": 1.4013,
      "step": 88470
    },
    {
      "epoch": 4.405317797686589,
      "eval_loss": 2.416937828063965,
      "eval_runtime": 22.0681,
      "eval_samples_per_second": 226.571,
      "eval_steps_per_second": 1.223,
      "step": 88479
    },
    {
      "epoch": 4.40537199180314,
      "grad_norm": 0.1971891224384308,
      "learning_rate": 4.9728511984503223e-05,
      "loss": 1.3898,
      "step": 88480
    },
    {
      "epoch": 4.405913932968652,
      "grad_norm": 0.18956254422664642,
      "learning_rate": 4.971278317017642e-05,
      "loss": 1.3957,
      "step": 88490
    },
    {
      "epoch": 4.406455874134164,
      "grad_norm": 0.2962309420108795,
      "learning_rate": 4.969705501080709e-05,
      "loss": 1.3972,
      "step": 88500
    },
    {
      "epoch": 4.406997815299676,
      "grad_norm": 0.2366226315498352,
      "learning_rate": 4.968132750834358e-05,
      "loss": 1.4011,
      "step": 88510
    },
    {
      "epoch": 4.407539756465189,
      "grad_norm": 0.25303101539611816,
      "learning_rate": 4.9665600664734104e-05,
      "loss": 1.3946,
      "step": 88520
    },
    {
      "epoch": 4.408081697630701,
      "grad_norm": 0.1666572242975235,
      "learning_rate": 4.964987448192686e-05,
      "loss": 1.3924,
      "step": 88530
    },
    {
      "epoch": 4.408623638796213,
      "grad_norm": 0.21928876638412476,
      "learning_rate": 4.9634148961869945e-05,
      "loss": 1.3978,
      "step": 88540
    },
    {
      "epoch": 4.4091655799617255,
      "grad_norm": 0.15686646103858948,
      "learning_rate": 4.961842410651135e-05,
      "loss": 1.3989,
      "step": 88550
    },
    {
      "epoch": 4.409707521127237,
      "grad_norm": 0.2927147448062897,
      "learning_rate": 4.960269991779902e-05,
      "loss": 1.3994,
      "step": 88560
    },
    {
      "epoch": 4.410032685826545,
      "eval_loss": 2.4300734996795654,
      "eval_runtime": 22.0915,
      "eval_samples_per_second": 226.331,
      "eval_steps_per_second": 1.222,
      "step": 88566
    },
    {
      "epoch": 4.41024946229275,
      "grad_norm": 0.16612303256988525,
      "learning_rate": 4.958697639768078e-05,
      "loss": 1.3973,
      "step": 88570
    },
    {
      "epoch": 4.410791403458262,
      "grad_norm": 0.2341911494731903,
      "learning_rate": 4.957125354810444e-05,
      "loss": 1.399,
      "step": 88580
    },
    {
      "epoch": 4.411333344623774,
      "grad_norm": 0.2245527058839798,
      "learning_rate": 4.9555531371017604e-05,
      "loss": 1.4014,
      "step": 88590
    },
    {
      "epoch": 4.411875285789287,
      "grad_norm": 0.25589320063591003,
      "learning_rate": 4.9539809868367906e-05,
      "loss": 1.395,
      "step": 88600
    },
    {
      "epoch": 4.4124172269547985,
      "grad_norm": 0.19590523838996887,
      "learning_rate": 4.952408904210288e-05,
      "loss": 1.3888,
      "step": 88610
    },
    {
      "epoch": 4.412959168120311,
      "grad_norm": 0.1683030128479004,
      "learning_rate": 4.950836889416991e-05,
      "loss": 1.3929,
      "step": 88620
    },
    {
      "epoch": 4.413501109285823,
      "grad_norm": 0.20308978855609894,
      "learning_rate": 4.949264942651637e-05,
      "loss": 1.4006,
      "step": 88630
    },
    {
      "epoch": 4.414043050451335,
      "grad_norm": 0.167439803481102,
      "learning_rate": 4.9476930641089506e-05,
      "loss": 1.3978,
      "step": 88640
    },
    {
      "epoch": 4.414584991616848,
      "grad_norm": 0.2057618796825409,
      "learning_rate": 4.9461212539836486e-05,
      "loss": 1.3981,
      "step": 88650
    },
    {
      "epoch": 4.414747573966501,
      "eval_loss": 2.4309983253479004,
      "eval_runtime": 21.9716,
      "eval_samples_per_second": 227.566,
      "eval_steps_per_second": 1.229,
      "step": 88653
    },
    {
      "epoch": 4.41512693278236,
      "grad_norm": 0.17505063116550446,
      "learning_rate": 4.944549512470441e-05,
      "loss": 1.3983,
      "step": 88660
    },
    {
      "epoch": 4.415668873947872,
      "grad_norm": 0.16853992640972137,
      "learning_rate": 4.942977839764028e-05,
      "loss": 1.3883,
      "step": 88670
    },
    {
      "epoch": 4.416210815113384,
      "grad_norm": 0.3285401165485382,
      "learning_rate": 4.941406236059104e-05,
      "loss": 1.3933,
      "step": 88680
    },
    {
      "epoch": 4.416752756278896,
      "grad_norm": 0.22175738215446472,
      "learning_rate": 4.9398347015503474e-05,
      "loss": 1.3954,
      "step": 88690
    },
    {
      "epoch": 4.417294697444409,
      "grad_norm": 0.17914403975009918,
      "learning_rate": 4.938263236432438e-05,
      "loss": 1.3919,
      "step": 88700
    },
    {
      "epoch": 4.417836638609921,
      "grad_norm": 0.2186654806137085,
      "learning_rate": 4.936691840900041e-05,
      "loss": 1.3968,
      "step": 88710
    },
    {
      "epoch": 4.418378579775434,
      "grad_norm": 0.2118908166885376,
      "learning_rate": 4.935120515147811e-05,
      "loss": 1.3969,
      "step": 88720
    },
    {
      "epoch": 4.4189205209409455,
      "grad_norm": 0.2079256772994995,
      "learning_rate": 4.9335492593704e-05,
      "loss": 1.3966,
      "step": 88730
    },
    {
      "epoch": 4.419462462106457,
      "grad_norm": 0.17602074146270752,
      "learning_rate": 4.931978073762448e-05,
      "loss": 1.404,
      "step": 88740
    },
    {
      "epoch": 4.419462462106457,
      "eval_loss": 2.4278478622436523,
      "eval_runtime": 21.9718,
      "eval_samples_per_second": 227.565,
      "eval_steps_per_second": 1.229,
      "step": 88740
    },
    {
      "epoch": 4.42000440327197,
      "grad_norm": 0.16428543627262115,
      "learning_rate": 4.930406958518584e-05,
      "loss": 1.415,
      "step": 88750
    },
    {
      "epoch": 4.420546344437482,
      "grad_norm": 0.18856200575828552,
      "learning_rate": 4.928835913833435e-05,
      "loss": 1.4063,
      "step": 88760
    },
    {
      "epoch": 4.421088285602994,
      "grad_norm": 0.22589999437332153,
      "learning_rate": 4.9272649399016134e-05,
      "loss": 1.4013,
      "step": 88770
    },
    {
      "epoch": 4.421630226768507,
      "grad_norm": 0.17148156464099884,
      "learning_rate": 4.925694036917723e-05,
      "loss": 1.3959,
      "step": 88780
    },
    {
      "epoch": 4.4221721679340185,
      "grad_norm": 0.19456231594085693,
      "learning_rate": 4.924123205076362e-05,
      "loss": 1.3995,
      "step": 88790
    },
    {
      "epoch": 4.422714109099531,
      "grad_norm": 0.18404072523117065,
      "learning_rate": 4.92255244457212e-05,
      "loss": 1.3852,
      "step": 88800
    },
    {
      "epoch": 4.423256050265043,
      "grad_norm": 0.16325946152210236,
      "learning_rate": 4.920981755599573e-05,
      "loss": 1.399,
      "step": 88810
    },
    {
      "epoch": 4.423797991430555,
      "grad_norm": 0.16499677300453186,
      "learning_rate": 4.9194111383532914e-05,
      "loss": 1.3996,
      "step": 88820
    },
    {
      "epoch": 4.4241773502464135,
      "eval_loss": 2.4304182529449463,
      "eval_runtime": 21.9655,
      "eval_samples_per_second": 227.63,
      "eval_steps_per_second": 1.229,
      "step": 88827
    },
    {
      "epoch": 4.424339932596068,
      "grad_norm": 0.20651055872440338,
      "learning_rate": 4.917840593027838e-05,
      "loss": 1.3967,
      "step": 88830
    },
    {
      "epoch": 4.42488187376158,
      "grad_norm": 0.16554446518421173,
      "learning_rate": 4.9162701198177655e-05,
      "loss": 1.3991,
      "step": 88840
    },
    {
      "epoch": 4.425423814927092,
      "grad_norm": 0.19531118869781494,
      "learning_rate": 4.914699718917615e-05,
      "loss": 1.3937,
      "step": 88850
    },
    {
      "epoch": 4.425965756092604,
      "grad_norm": 0.23367935419082642,
      "learning_rate": 4.913129390521922e-05,
      "loss": 1.3936,
      "step": 88860
    },
    {
      "epoch": 4.426507697258116,
      "grad_norm": 0.2578764855861664,
      "learning_rate": 4.911559134825213e-05,
      "loss": 1.3988,
      "step": 88870
    },
    {
      "epoch": 4.427049638423629,
      "grad_norm": 0.2369229793548584,
      "learning_rate": 4.9099889520220034e-05,
      "loss": 1.3805,
      "step": 88880
    },
    {
      "epoch": 4.427591579589141,
      "grad_norm": 0.20222489535808563,
      "learning_rate": 4.9084188423068e-05,
      "loss": 1.3985,
      "step": 88890
    },
    {
      "epoch": 4.428133520754653,
      "grad_norm": 0.1944282501935959,
      "learning_rate": 4.9068488058741044e-05,
      "loss": 1.3978,
      "step": 88900
    },
    {
      "epoch": 4.428675461920165,
      "grad_norm": 0.2579196095466614,
      "learning_rate": 4.905278842918402e-05,
      "loss": 1.4046,
      "step": 88910
    },
    {
      "epoch": 4.4288922383863705,
      "eval_loss": 2.4295578002929688,
      "eval_runtime": 21.969,
      "eval_samples_per_second": 227.593,
      "eval_steps_per_second": 1.229,
      "step": 88914
    },
    {
      "epoch": 4.429217403085677,
      "grad_norm": 0.21468320488929749,
      "learning_rate": 4.903708953634174e-05,
      "loss": 1.3975,
      "step": 88920
    },
    {
      "epoch": 4.42975934425119,
      "grad_norm": 0.35410991311073303,
      "learning_rate": 4.902139138215893e-05,
      "loss": 1.3927,
      "step": 88930
    },
    {
      "epoch": 4.430301285416702,
      "grad_norm": 0.3901987373828888,
      "learning_rate": 4.900569396858019e-05,
      "loss": 1.4055,
      "step": 88940
    },
    {
      "epoch": 4.430843226582214,
      "grad_norm": 0.20195499062538147,
      "learning_rate": 4.898999729755006e-05,
      "loss": 1.3868,
      "step": 88950
    },
    {
      "epoch": 4.4313851677477265,
      "grad_norm": 0.28525593876838684,
      "learning_rate": 4.8974301371012986e-05,
      "loss": 1.4099,
      "step": 88960
    },
    {
      "epoch": 4.431927108913238,
      "grad_norm": 0.17980363965034485,
      "learning_rate": 4.895860619091327e-05,
      "loss": 1.3877,
      "step": 88970
    },
    {
      "epoch": 4.432469050078751,
      "grad_norm": 0.18058229982852936,
      "learning_rate": 4.8942911759195196e-05,
      "loss": 1.3908,
      "step": 88980
    },
    {
      "epoch": 4.433010991244263,
      "grad_norm": 0.17202942073345184,
      "learning_rate": 4.892721807780293e-05,
      "loss": 1.399,
      "step": 88990
    },
    {
      "epoch": 4.433552932409775,
      "grad_norm": 0.24401873350143433,
      "learning_rate": 4.891152514868052e-05,
      "loss": 1.386,
      "step": 89000
    },
    {
      "epoch": 4.433607126526327,
      "eval_loss": 2.4285366535186768,
      "eval_runtime": 21.9665,
      "eval_samples_per_second": 227.619,
      "eval_steps_per_second": 1.229,
      "step": 89001
    },
    {
      "epoch": 4.434094873575288,
      "grad_norm": 0.1815049946308136,
      "learning_rate": 4.889583297377194e-05,
      "loss": 1.398,
      "step": 89010
    },
    {
      "epoch": 4.4346368147407995,
      "grad_norm": 0.2082160860300064,
      "learning_rate": 4.8880141555021055e-05,
      "loss": 1.4029,
      "step": 89020
    },
    {
      "epoch": 4.435178755906312,
      "grad_norm": 0.20171265304088593,
      "learning_rate": 4.8864450894371683e-05,
      "loss": 1.397,
      "step": 89030
    },
    {
      "epoch": 4.435720697071824,
      "grad_norm": 0.2109455019235611,
      "learning_rate": 4.8848760993767497e-05,
      "loss": 1.3919,
      "step": 89040
    },
    {
      "epoch": 4.436262638237336,
      "grad_norm": 0.2744022011756897,
      "learning_rate": 4.883307185515207e-05,
      "loss": 1.4098,
      "step": 89050
    },
    {
      "epoch": 4.436804579402849,
      "grad_norm": 0.1699230968952179,
      "learning_rate": 4.881738348046897e-05,
      "loss": 1.3943,
      "step": 89060
    },
    {
      "epoch": 4.437346520568361,
      "grad_norm": 0.16242730617523193,
      "learning_rate": 4.880169587166151e-05,
      "loss": 1.4005,
      "step": 89070
    },
    {
      "epoch": 4.437888461733873,
      "grad_norm": 0.14713962376117706,
      "learning_rate": 4.8786009030673084e-05,
      "loss": 1.3952,
      "step": 89080
    },
    {
      "epoch": 4.438322014666283,
      "eval_loss": 2.425018548965454,
      "eval_runtime": 21.9685,
      "eval_samples_per_second": 227.599,
      "eval_steps_per_second": 1.229,
      "step": 89088
    },
    {
      "epoch": 4.438430402899385,
      "grad_norm": 0.1697177290916443,
      "learning_rate": 4.877032295944689e-05,
      "loss": 1.4059,
      "step": 89090
    },
    {
      "epoch": 4.438972344064897,
      "grad_norm": 0.18118594586849213,
      "learning_rate": 4.875463765992603e-05,
      "loss": 1.3885,
      "step": 89100
    },
    {
      "epoch": 4.43951428523041,
      "grad_norm": 0.3551163375377655,
      "learning_rate": 4.8738953134053535e-05,
      "loss": 1.3941,
      "step": 89110
    },
    {
      "epoch": 4.440056226395922,
      "grad_norm": 0.18483349680900574,
      "learning_rate": 4.872326938377235e-05,
      "loss": 1.3873,
      "step": 89120
    },
    {
      "epoch": 4.4405981675614346,
      "grad_norm": 0.22286327183246613,
      "learning_rate": 4.870758641102531e-05,
      "loss": 1.3998,
      "step": 89130
    },
    {
      "epoch": 4.441140108726946,
      "grad_norm": 0.2854524850845337,
      "learning_rate": 4.8691904217755126e-05,
      "loss": 1.3987,
      "step": 89140
    },
    {
      "epoch": 4.441682049892458,
      "grad_norm": 0.37562283873558044,
      "learning_rate": 4.867622280590447e-05,
      "loss": 1.3821,
      "step": 89150
    },
    {
      "epoch": 4.442223991057971,
      "grad_norm": 0.20074434578418732,
      "learning_rate": 4.866054217741589e-05,
      "loss": 1.3955,
      "step": 89160
    },
    {
      "epoch": 4.442765932223483,
      "grad_norm": 0.1584572046995163,
      "learning_rate": 4.86448623342318e-05,
      "loss": 1.3925,
      "step": 89170
    },
    {
      "epoch": 4.443036902806239,
      "eval_loss": 2.4282519817352295,
      "eval_runtime": 21.9667,
      "eval_samples_per_second": 227.618,
      "eval_steps_per_second": 1.229,
      "step": 89175
    },
    {
      "epoch": 4.443307873388995,
      "grad_norm": 0.18110856413841248,
      "learning_rate": 4.8629183278294584e-05,
      "loss": 1.4015,
      "step": 89180
    },
    {
      "epoch": 4.443849814554508,
      "grad_norm": 0.1557188332080841,
      "learning_rate": 4.86135050115465e-05,
      "loss": 1.4022,
      "step": 89190
    },
    {
      "epoch": 4.444391755720019,
      "grad_norm": 0.15835806727409363,
      "learning_rate": 4.859782753592968e-05,
      "loss": 1.3943,
      "step": 89200
    },
    {
      "epoch": 4.444933696885532,
      "grad_norm": 0.24644139409065247,
      "learning_rate": 4.858215085338617e-05,
      "loss": 1.4002,
      "step": 89210
    },
    {
      "epoch": 4.445475638051044,
      "grad_norm": 0.3007756471633911,
      "learning_rate": 4.856647496585797e-05,
      "loss": 1.3948,
      "step": 89220
    },
    {
      "epoch": 4.446017579216556,
      "grad_norm": 0.15734322369098663,
      "learning_rate": 4.8550799875286904e-05,
      "loss": 1.4099,
      "step": 89230
    },
    {
      "epoch": 4.446559520382069,
      "grad_norm": 0.19473432004451752,
      "learning_rate": 4.853512558361475e-05,
      "loss": 1.4047,
      "step": 89240
    },
    {
      "epoch": 4.447101461547581,
      "grad_norm": 0.17770782113075256,
      "learning_rate": 4.8519452092783193e-05,
      "loss": 1.3881,
      "step": 89250
    },
    {
      "epoch": 4.447643402713093,
      "grad_norm": 0.19613268971443176,
      "learning_rate": 4.850377940473375e-05,
      "loss": 1.4063,
      "step": 89260
    },
    {
      "epoch": 4.447751790946195,
      "eval_loss": 2.4215874671936035,
      "eval_runtime": 21.964,
      "eval_samples_per_second": 227.645,
      "eval_steps_per_second": 1.229,
      "step": 89262
    },
    {
      "epoch": 4.448185343878605,
      "grad_norm": 0.23467159271240234,
      "learning_rate": 4.848810752140791e-05,
      "loss": 1.3918,
      "step": 89270
    },
    {
      "epoch": 4.448727285044117,
      "grad_norm": 0.1882312148809433,
      "learning_rate": 4.847243644474707e-05,
      "loss": 1.3882,
      "step": 89280
    },
    {
      "epoch": 4.44926922620963,
      "grad_norm": 0.16581347584724426,
      "learning_rate": 4.8456766176692435e-05,
      "loss": 1.3909,
      "step": 89290
    },
    {
      "epoch": 4.449811167375142,
      "grad_norm": 0.2854909300804138,
      "learning_rate": 4.84410967191852e-05,
      "loss": 1.4066,
      "step": 89300
    },
    {
      "epoch": 4.4503531085406545,
      "grad_norm": 0.28027182817459106,
      "learning_rate": 4.842542807416644e-05,
      "loss": 1.4093,
      "step": 89310
    },
    {
      "epoch": 4.450895049706166,
      "grad_norm": 0.2496391236782074,
      "learning_rate": 4.840976024357709e-05,
      "loss": 1.3993,
      "step": 89320
    },
    {
      "epoch": 4.451436990871678,
      "grad_norm": 0.17088313400745392,
      "learning_rate": 4.839409322935804e-05,
      "loss": 1.3956,
      "step": 89330
    },
    {
      "epoch": 4.451978932037191,
      "grad_norm": 0.20805992186069489,
      "learning_rate": 4.837842703345003e-05,
      "loss": 1.3964,
      "step": 89340
    },
    {
      "epoch": 4.452466679086152,
      "eval_loss": 2.4300498962402344,
      "eval_runtime": 21.9632,
      "eval_samples_per_second": 227.654,
      "eval_steps_per_second": 1.229,
      "step": 89349
    },
    {
      "epoch": 4.452520873202703,
      "grad_norm": 0.22051380574703217,
      "learning_rate": 4.8362761657793756e-05,
      "loss": 1.3976,
      "step": 89350
    },
    {
      "epoch": 4.453062814368215,
      "grad_norm": 0.20481519401073456,
      "learning_rate": 4.834709710432972e-05,
      "loss": 1.3972,
      "step": 89360
    },
    {
      "epoch": 4.4536047555337275,
      "grad_norm": 0.17479705810546875,
      "learning_rate": 4.8331433374998426e-05,
      "loss": 1.4061,
      "step": 89370
    },
    {
      "epoch": 4.454146696699239,
      "grad_norm": 0.16444361209869385,
      "learning_rate": 4.831577047174023e-05,
      "loss": 1.3974,
      "step": 89380
    },
    {
      "epoch": 4.454688637864752,
      "grad_norm": 0.18008925020694733,
      "learning_rate": 4.830010839649535e-05,
      "loss": 1.4009,
      "step": 89390
    },
    {
      "epoch": 4.455230579030264,
      "grad_norm": 0.17722876369953156,
      "learning_rate": 4.828444715120395e-05,
      "loss": 1.397,
      "step": 89400
    },
    {
      "epoch": 4.455772520195776,
      "grad_norm": 0.21487845480442047,
      "learning_rate": 4.826878673780609e-05,
      "loss": 1.398,
      "step": 89410
    },
    {
      "epoch": 4.456314461361289,
      "grad_norm": 0.16381223499774933,
      "learning_rate": 4.8253127158241693e-05,
      "loss": 1.4073,
      "step": 89420
    },
    {
      "epoch": 4.4568564025268005,
      "grad_norm": 0.20386900007724762,
      "learning_rate": 4.823746841445062e-05,
      "loss": 1.3958,
      "step": 89430
    },
    {
      "epoch": 4.457181567226108,
      "eval_loss": 2.4307987689971924,
      "eval_runtime": 21.974,
      "eval_samples_per_second": 227.542,
      "eval_steps_per_second": 1.229,
      "step": 89436
    },
    {
      "epoch": 4.457398343692313,
      "grad_norm": 0.2596319019794464,
      "learning_rate": 4.82218105083726e-05,
      "loss": 1.4007,
      "step": 89440
    },
    {
      "epoch": 4.457940284857825,
      "grad_norm": 0.1717374473810196,
      "learning_rate": 4.820615344194728e-05,
      "loss": 1.3932,
      "step": 89450
    },
    {
      "epoch": 4.458482226023337,
      "grad_norm": 0.16155289113521576,
      "learning_rate": 4.819049721711415e-05,
      "loss": 1.3987,
      "step": 89460
    },
    {
      "epoch": 4.45902416718885,
      "grad_norm": 0.19415195286273956,
      "learning_rate": 4.8174841835812665e-05,
      "loss": 1.3963,
      "step": 89470
    },
    {
      "epoch": 4.459566108354362,
      "grad_norm": 0.20656591653823853,
      "learning_rate": 4.815918729998217e-05,
      "loss": 1.3838,
      "step": 89480
    },
    {
      "epoch": 4.460108049519874,
      "grad_norm": 0.33386117219924927,
      "learning_rate": 4.814353361156182e-05,
      "loss": 1.3849,
      "step": 89490
    },
    {
      "epoch": 4.460649990685386,
      "grad_norm": 0.21893806755542755,
      "learning_rate": 4.8127880772490764e-05,
      "loss": 1.3819,
      "step": 89500
    },
    {
      "epoch": 4.461191931850898,
      "grad_norm": 0.22136323153972626,
      "learning_rate": 4.811222878470801e-05,
      "loss": 1.3927,
      "step": 89510
    },
    {
      "epoch": 4.461733873016411,
      "grad_norm": 0.40589383244514465,
      "learning_rate": 4.809657765015245e-05,
      "loss": 1.4007,
      "step": 89520
    },
    {
      "epoch": 4.461896455366064,
      "eval_loss": 2.4328644275665283,
      "eval_runtime": 21.9657,
      "eval_samples_per_second": 227.627,
      "eval_steps_per_second": 1.229,
      "step": 89523
    },
    {
      "epoch": 4.462275814181923,
      "grad_norm": 0.2112714648246765,
      "learning_rate": 4.808092737076287e-05,
      "loss": 1.3999,
      "step": 89530
    },
    {
      "epoch": 4.4628177553474355,
      "grad_norm": 0.22988109290599823,
      "learning_rate": 4.8065277948477996e-05,
      "loss": 1.4017,
      "step": 89540
    },
    {
      "epoch": 4.463359696512947,
      "grad_norm": 0.2842319905757904,
      "learning_rate": 4.804962938523636e-05,
      "loss": 1.3933,
      "step": 89550
    },
    {
      "epoch": 4.463901637678459,
      "grad_norm": 0.16226890683174133,
      "learning_rate": 4.803398168297645e-05,
      "loss": 1.4046,
      "step": 89560
    },
    {
      "epoch": 4.464443578843972,
      "grad_norm": 0.18163561820983887,
      "learning_rate": 4.801833484363667e-05,
      "loss": 1.3999,
      "step": 89570
    },
    {
      "epoch": 4.464985520009484,
      "grad_norm": 0.1625688374042511,
      "learning_rate": 4.8002688869155246e-05,
      "loss": 1.3945,
      "step": 89580
    },
    {
      "epoch": 4.465527461174997,
      "grad_norm": 0.15731951594352722,
      "learning_rate": 4.798704376147034e-05,
      "loss": 1.3903,
      "step": 89590
    },
    {
      "epoch": 4.4660694023405085,
      "grad_norm": 0.2153928130865097,
      "learning_rate": 4.797139952252e-05,
      "loss": 1.4023,
      "step": 89600
    },
    {
      "epoch": 4.46661134350602,
      "grad_norm": 0.2910442650318146,
      "learning_rate": 4.795575615424219e-05,
      "loss": 1.3847,
      "step": 89610
    },
    {
      "epoch": 4.46661134350602,
      "eval_loss": 2.4339542388916016,
      "eval_runtime": 21.954,
      "eval_samples_per_second": 227.749,
      "eval_steps_per_second": 1.23,
      "step": 89610
    },
    {
      "epoch": 4.467153284671533,
      "grad_norm": 0.17365294694900513,
      "learning_rate": 4.794011365857472e-05,
      "loss": 1.3971,
      "step": 89620
    },
    {
      "epoch": 4.467695225837045,
      "grad_norm": 0.3318266272544861,
      "learning_rate": 4.7924472037455304e-05,
      "loss": 1.3983,
      "step": 89630
    },
    {
      "epoch": 4.468237167002557,
      "grad_norm": 0.26000094413757324,
      "learning_rate": 4.790883129282161e-05,
      "loss": 1.3969,
      "step": 89640
    },
    {
      "epoch": 4.46877910816807,
      "grad_norm": 0.1824985295534134,
      "learning_rate": 4.789319142661107e-05,
      "loss": 1.3838,
      "step": 89650
    },
    {
      "epoch": 4.4693210493335815,
      "grad_norm": 0.1758066862821579,
      "learning_rate": 4.787755244076112e-05,
      "loss": 1.3934,
      "step": 89660
    },
    {
      "epoch": 4.469862990499094,
      "grad_norm": 0.17066720128059387,
      "learning_rate": 4.786191433720909e-05,
      "loss": 1.4048,
      "step": 89670
    },
    {
      "epoch": 4.470404931664606,
      "grad_norm": 0.1667661815881729,
      "learning_rate": 4.784627711789209e-05,
      "loss": 1.3922,
      "step": 89680
    },
    {
      "epoch": 4.470946872830118,
      "grad_norm": 0.1883508414030075,
      "learning_rate": 4.7830640784747225e-05,
      "loss": 1.3911,
      "step": 89690
    },
    {
      "epoch": 4.4713262316459765,
      "eval_loss": 2.428875207901001,
      "eval_runtime": 21.9652,
      "eval_samples_per_second": 227.633,
      "eval_steps_per_second": 1.229,
      "step": 89697
    },
    {
      "epoch": 4.471488813995631,
      "grad_norm": 0.21949131786823273,
      "learning_rate": 4.781500533971147e-05,
      "loss": 1.3919,
      "step": 89700
    },
    {
      "epoch": 4.472030755161143,
      "grad_norm": 0.19700855016708374,
      "learning_rate": 4.779937078472164e-05,
      "loss": 1.4053,
      "step": 89710
    },
    {
      "epoch": 4.472572696326655,
      "grad_norm": 0.17879433929920197,
      "learning_rate": 4.7783737121714504e-05,
      "loss": 1.3874,
      "step": 89720
    },
    {
      "epoch": 4.473114637492167,
      "grad_norm": 0.16354680061340332,
      "learning_rate": 4.77681043526267e-05,
      "loss": 1.4007,
      "step": 89730
    },
    {
      "epoch": 4.473656578657679,
      "grad_norm": 0.23582887649536133,
      "learning_rate": 4.7752472479394706e-05,
      "loss": 1.4049,
      "step": 89740
    },
    {
      "epoch": 4.474198519823192,
      "grad_norm": 0.1751241385936737,
      "learning_rate": 4.7736841503954956e-05,
      "loss": 1.3926,
      "step": 89750
    },
    {
      "epoch": 4.474740460988704,
      "grad_norm": 0.2157001942396164,
      "learning_rate": 4.772121142824374e-05,
      "loss": 1.3967,
      "step": 89760
    },
    {
      "epoch": 4.475282402154216,
      "grad_norm": 0.2224288284778595,
      "learning_rate": 4.770558225419728e-05,
      "loss": 1.4,
      "step": 89770
    },
    {
      "epoch": 4.4758243433197284,
      "grad_norm": 0.1734488606452942,
      "learning_rate": 4.7689953983751604e-05,
      "loss": 1.3875,
      "step": 89780
    },
    {
      "epoch": 4.4760411197859336,
      "eval_loss": 2.4124958515167236,
      "eval_runtime": 21.9708,
      "eval_samples_per_second": 227.574,
      "eval_steps_per_second": 1.229,
      "step": 89784
    },
    {
      "epoch": 4.47636628448524,
      "grad_norm": 0.17843768000602722,
      "learning_rate": 4.767432661884269e-05,
      "loss": 1.3945,
      "step": 89790
    },
    {
      "epoch": 4.476908225650753,
      "grad_norm": 0.165896475315094,
      "learning_rate": 4.76587001614064e-05,
      "loss": 1.4107,
      "step": 89800
    },
    {
      "epoch": 4.477450166816265,
      "grad_norm": 0.19189293682575226,
      "learning_rate": 4.7643074613378455e-05,
      "loss": 1.3998,
      "step": 89810
    },
    {
      "epoch": 4.477992107981777,
      "grad_norm": 0.26775965094566345,
      "learning_rate": 4.7627449976694486e-05,
      "loss": 1.3911,
      "step": 89820
    },
    {
      "epoch": 4.47853404914729,
      "grad_norm": 0.2671341896057129,
      "learning_rate": 4.761182625329001e-05,
      "loss": 1.4019,
      "step": 89830
    },
    {
      "epoch": 4.4790759903128015,
      "grad_norm": 0.16249962151050568,
      "learning_rate": 4.7596203445100416e-05,
      "loss": 1.3971,
      "step": 89840
    },
    {
      "epoch": 4.479617931478314,
      "grad_norm": 0.273316353559494,
      "learning_rate": 4.7580581554060986e-05,
      "loss": 1.39,
      "step": 89850
    },
    {
      "epoch": 4.480159872643826,
      "grad_norm": 0.19733589887619019,
      "learning_rate": 4.7564960582106933e-05,
      "loss": 1.4028,
      "step": 89860
    },
    {
      "epoch": 4.480701813809338,
      "grad_norm": 0.17823106050491333,
      "learning_rate": 4.754934053117325e-05,
      "loss": 1.4006,
      "step": 89870
    },
    {
      "epoch": 4.48075600792589,
      "eval_loss": 2.4114816188812256,
      "eval_runtime": 21.9671,
      "eval_samples_per_second": 227.613,
      "eval_steps_per_second": 1.229,
      "step": 89871
    },
    {
      "epoch": 4.481243754974851,
      "grad_norm": 0.18662744760513306,
      "learning_rate": 4.753372140319492e-05,
      "loss": 1.3969,
      "step": 89880
    },
    {
      "epoch": 4.481785696140363,
      "grad_norm": 0.23681077361106873,
      "learning_rate": 4.751810320010678e-05,
      "loss": 1.4017,
      "step": 89890
    },
    {
      "epoch": 4.482327637305875,
      "grad_norm": 0.3316483795642853,
      "learning_rate": 4.750248592384352e-05,
      "loss": 1.3945,
      "step": 89900
    },
    {
      "epoch": 4.482869578471387,
      "grad_norm": 0.261080801486969,
      "learning_rate": 4.748686957633975e-05,
      "loss": 1.3913,
      "step": 89910
    },
    {
      "epoch": 4.483411519636899,
      "grad_norm": 0.1946885585784912,
      "learning_rate": 4.747125415952998e-05,
      "loss": 1.3939,
      "step": 89920
    },
    {
      "epoch": 4.483953460802412,
      "grad_norm": 0.2138025015592575,
      "learning_rate": 4.745563967534855e-05,
      "loss": 1.3982,
      "step": 89930
    },
    {
      "epoch": 4.484495401967924,
      "grad_norm": 0.2956826984882355,
      "learning_rate": 4.744002612572972e-05,
      "loss": 1.3999,
      "step": 89940
    },
    {
      "epoch": 4.4850373431334365,
      "grad_norm": 0.30752211809158325,
      "learning_rate": 4.742441351260761e-05,
      "loss": 1.3892,
      "step": 89950
    },
    {
      "epoch": 4.485470896065846,
      "eval_loss": 2.4178147315979004,
      "eval_runtime": 21.9672,
      "eval_samples_per_second": 227.613,
      "eval_steps_per_second": 1.229,
      "step": 89958
    },
    {
      "epoch": 4.485579284298948,
      "grad_norm": 0.2102116495370865,
      "learning_rate": 4.7408801837916306e-05,
      "loss": 1.4062,
      "step": 89960
    },
    {
      "epoch": 4.48612122546446,
      "grad_norm": 0.3934997320175171,
      "learning_rate": 4.7393191103589643e-05,
      "loss": 1.3999,
      "step": 89970
    },
    {
      "epoch": 4.486663166629973,
      "grad_norm": 0.2734357416629791,
      "learning_rate": 4.737758131156144e-05,
      "loss": 1.3997,
      "step": 89980
    },
    {
      "epoch": 4.487205107795485,
      "grad_norm": 0.1576622873544693,
      "learning_rate": 4.736197246376538e-05,
      "loss": 1.3969,
      "step": 89990
    },
    {
      "epoch": 4.487747048960998,
      "grad_norm": 0.18201720714569092,
      "learning_rate": 4.7346364562135e-05,
      "loss": 1.3923,
      "step": 90000
    }
  ],
  "logging_steps": 10,
  "max_steps": 92260,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8386602891165565e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}