{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.46000620058905595,
  "eval_steps": 500,
  "global_step": 5935,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007750736319950396,
      "grad_norm": 1.3823052644729614,
      "learning_rate": 2.910112359550562e-06,
      "loss": 0.7897,
      "step": 10
    },
    {
      "epoch": 0.0015501472639900791,
      "grad_norm": 1.15909743309021,
      "learning_rate": 3.921348314606742e-06,
      "loss": 0.7524,
      "step": 20
    },
    {
      "epoch": 0.002325220895985119,
      "grad_norm": 0.9035636186599731,
      "learning_rate": 4.932584269662922e-06,
      "loss": 0.7419,
      "step": 30
    },
    {
      "epoch": 0.0031002945279801583,
      "grad_norm": 0.7007580995559692,
      "learning_rate": 5.943820224719102e-06,
      "loss": 0.7113,
      "step": 40
    },
    {
      "epoch": 0.0038753681599751977,
      "grad_norm": 0.7174043655395508,
      "learning_rate": 6.955056179775282e-06,
      "loss": 0.7164,
      "step": 50
    },
    {
      "epoch": 0.004650441791970238,
      "grad_norm": 0.720160961151123,
      "learning_rate": 7.966292134831462e-06,
      "loss": 0.7077,
      "step": 60
    },
    {
      "epoch": 0.005425515423965277,
      "grad_norm": 0.6832423210144043,
      "learning_rate": 8.977528089887641e-06,
      "loss": 0.6898,
      "step": 70
    },
    {
      "epoch": 0.0062005890559603165,
      "grad_norm": 0.7489743232727051,
      "learning_rate": 9.988764044943822e-06,
      "loss": 0.7083,
      "step": 80
    },
    {
      "epoch": 0.0069756626879553555,
      "grad_norm": 0.7608446478843689,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.6885,
      "step": 90
    },
    {
      "epoch": 0.007750736319950395,
      "grad_norm": 0.7493481636047363,
      "learning_rate": 1.2011235955056182e-05,
      "loss": 0.6888,
      "step": 100
    },
    {
      "epoch": 0.008525809951945434,
      "grad_norm": 0.7710257172584534,
      "learning_rate": 1.3022471910112362e-05,
      "loss": 0.6827,
      "step": 110
    },
    {
      "epoch": 0.009300883583940475,
      "grad_norm": 0.797978401184082,
      "learning_rate": 1.403370786516854e-05,
      "loss": 0.716,
      "step": 120
    },
    {
      "epoch": 0.010075957215935514,
      "grad_norm": 0.7106821537017822,
      "learning_rate": 1.504494382022472e-05,
      "loss": 0.6825,
      "step": 130
    },
    {
      "epoch": 0.010851030847930553,
      "grad_norm": 0.6821442246437073,
      "learning_rate": 1.60561797752809e-05,
      "loss": 0.699,
      "step": 140
    },
    {
      "epoch": 0.011626104479925592,
      "grad_norm": 0.7108712792396545,
      "learning_rate": 1.706741573033708e-05,
      "loss": 0.6979,
      "step": 150
    },
    {
      "epoch": 0.012401178111920633,
      "grad_norm": 0.7028000354766846,
      "learning_rate": 1.8078651685393256e-05,
      "loss": 0.6895,
      "step": 160
    },
    {
      "epoch": 0.013176251743915672,
      "grad_norm": 0.721693217754364,
      "learning_rate": 1.908988764044944e-05,
      "loss": 0.6835,
      "step": 170
    },
    {
      "epoch": 0.013951325375910711,
      "grad_norm": 0.589385449886322,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 180
    },
    {
      "epoch": 0.014726399007905752,
      "grad_norm": 1.0280730724334717,
      "learning_rate": 2e-05,
      "loss": 0.7003,
      "step": 190
    },
    {
      "epoch": 0.01550147263990079,
      "grad_norm": 0.6079630255699158,
      "learning_rate": 2e-05,
      "loss": 0.6898,
      "step": 200
    },
    {
      "epoch": 0.01627654627189583,
      "grad_norm": 0.7045310139656067,
      "learning_rate": 2e-05,
      "loss": 0.6917,
      "step": 210
    },
    {
      "epoch": 0.01705161990389087,
      "grad_norm": 0.5710526704788208,
      "learning_rate": 2e-05,
      "loss": 0.6881,
      "step": 220
    },
    {
      "epoch": 0.01782669353588591,
      "grad_norm": 0.5788965225219727,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 230
    },
    {
      "epoch": 0.01860176716788095,
      "grad_norm": 0.6106708645820618,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 240
    },
    {
      "epoch": 0.019376840799875988,
      "grad_norm": 0.561893105506897,
      "learning_rate": 2e-05,
      "loss": 0.6921,
      "step": 250
    },
    {
      "epoch": 0.02015191443187103,
      "grad_norm": 0.5690069794654846,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 260
    },
    {
      "epoch": 0.020926988063866066,
      "grad_norm": 0.4941854774951935,
      "learning_rate": 2e-05,
      "loss": 0.6717,
      "step": 270
    },
    {
      "epoch": 0.021702061695861107,
      "grad_norm": 0.5330811738967896,
      "learning_rate": 2e-05,
      "loss": 0.68,
      "step": 280
    },
    {
      "epoch": 0.022477135327856147,
      "grad_norm": 0.5373284220695496,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 290
    },
    {
      "epoch": 0.023252208959851185,
      "grad_norm": 0.4778149724006653,
      "learning_rate": 2e-05,
      "loss": 0.6953,
      "step": 300
    },
    {
      "epoch": 0.024027282591846225,
      "grad_norm": 0.509177565574646,
      "learning_rate": 2e-05,
      "loss": 0.6774,
      "step": 310
    },
    {
      "epoch": 0.024802356223841266,
      "grad_norm": 0.48862743377685547,
      "learning_rate": 2e-05,
      "loss": 0.6819,
      "step": 320
    },
    {
      "epoch": 0.025577429855836303,
      "grad_norm": 0.5163516402244568,
      "learning_rate": 2e-05,
      "loss": 0.688,
      "step": 330
    },
    {
      "epoch": 0.026352503487831344,
      "grad_norm": 0.48807498812675476,
      "learning_rate": 2e-05,
      "loss": 0.6817,
      "step": 340
    },
    {
      "epoch": 0.027127577119826385,
      "grad_norm": 0.5226020216941833,
      "learning_rate": 2e-05,
      "loss": 0.6873,
      "step": 350
    },
    {
      "epoch": 0.027902650751821422,
      "grad_norm": 0.48832374811172485,
      "learning_rate": 2e-05,
      "loss": 0.6793,
      "step": 360
    },
    {
      "epoch": 0.028677724383816463,
      "grad_norm": 0.5948355793952942,
      "learning_rate": 2e-05,
      "loss": 0.7136,
      "step": 370
    },
    {
      "epoch": 0.029452798015811504,
      "grad_norm": 0.47444790601730347,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 380
    },
    {
      "epoch": 0.03022787164780654,
      "grad_norm": 0.46937182545661926,
      "learning_rate": 2e-05,
      "loss": 0.6799,
      "step": 390
    },
    {
      "epoch": 0.03100294527980158,
      "grad_norm": 0.4924728572368622,
      "learning_rate": 2e-05,
      "loss": 0.6882,
      "step": 400
    },
    {
      "epoch": 0.03177801891179662,
      "grad_norm": 0.46873292326927185,
      "learning_rate": 2e-05,
      "loss": 0.6784,
      "step": 410
    },
    {
      "epoch": 0.03255309254379166,
      "grad_norm": 0.5081288814544678,
      "learning_rate": 2e-05,
      "loss": 0.6673,
      "step": 420
    },
    {
      "epoch": 0.0333281661757867,
      "grad_norm": 0.48152658343315125,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 430
    },
    {
      "epoch": 0.03410323980778174,
      "grad_norm": 0.5463388562202454,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 440
    },
    {
      "epoch": 0.03487831343977678,
      "grad_norm": 0.493053674697876,
      "learning_rate": 2e-05,
      "loss": 0.6837,
      "step": 450
    },
    {
      "epoch": 0.03565338707177182,
      "grad_norm": 0.4562004506587982,
      "learning_rate": 2e-05,
      "loss": 0.6805,
      "step": 460
    },
    {
      "epoch": 0.03642846070376686,
      "grad_norm": 0.45049288868904114,
      "learning_rate": 2e-05,
      "loss": 0.6853,
      "step": 470
    },
    {
      "epoch": 0.0372035343357619,
      "grad_norm": 0.44892555475234985,
      "learning_rate": 2e-05,
      "loss": 0.6939,
      "step": 480
    },
    {
      "epoch": 0.037978607967756935,
      "grad_norm": 0.45876285433769226,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 490
    },
    {
      "epoch": 0.038753681599751975,
      "grad_norm": 0.4538588523864746,
      "learning_rate": 2e-05,
      "loss": 0.6822,
      "step": 500
    },
    {
      "epoch": 0.039528755231747016,
      "grad_norm": 0.46754956245422363,
      "learning_rate": 2e-05,
      "loss": 0.6762,
      "step": 510
    },
    {
      "epoch": 0.04030382886374206,
      "grad_norm": 0.5183224081993103,
      "learning_rate": 2e-05,
      "loss": 0.6857,
      "step": 520
    },
    {
      "epoch": 0.0410789024957371,
      "grad_norm": 0.4643199145793915,
      "learning_rate": 2e-05,
      "loss": 0.6758,
      "step": 530
    },
    {
      "epoch": 0.04185397612773213,
      "grad_norm": 0.45650675892829895,
      "learning_rate": 2e-05,
      "loss": 0.6851,
      "step": 540
    },
    {
      "epoch": 0.04262904975972717,
      "grad_norm": 0.42580446600914,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 550
    },
    {
      "epoch": 0.04340412339172221,
      "grad_norm": 0.4507347047328949,
      "learning_rate": 2e-05,
      "loss": 0.6803,
      "step": 560
    },
    {
      "epoch": 0.044179197023717254,
      "grad_norm": 0.447135329246521,
      "learning_rate": 2e-05,
      "loss": 0.6877,
      "step": 570
    },
    {
      "epoch": 0.044954270655712295,
      "grad_norm": 0.4340040981769562,
      "learning_rate": 2e-05,
      "loss": 0.6622,
      "step": 580
    },
    {
      "epoch": 0.045729344287707335,
      "grad_norm": 0.4411773383617401,
      "learning_rate": 2e-05,
      "loss": 0.6895,
      "step": 590
    },
    {
      "epoch": 0.04650441791970237,
      "grad_norm": 0.4499826729297638,
      "learning_rate": 2e-05,
      "loss": 0.6898,
      "step": 600
    },
    {
      "epoch": 0.04727949155169741,
      "grad_norm": 0.4610482156276703,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 610
    },
    {
      "epoch": 0.04805456518369245,
      "grad_norm": 0.4646986126899719,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 620
    },
    {
      "epoch": 0.04882963881568749,
      "grad_norm": 0.4380117654800415,
      "learning_rate": 2e-05,
      "loss": 0.6651,
      "step": 630
    },
    {
      "epoch": 0.04960471244768253,
      "grad_norm": 0.7984209656715393,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 640
    },
    {
      "epoch": 0.050379786079677566,
      "grad_norm": 0.45534566044807434,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 650
    },
    {
      "epoch": 0.05115485971167261,
      "grad_norm": 0.43884512782096863,
      "learning_rate": 2e-05,
      "loss": 0.6748,
      "step": 660
    },
    {
      "epoch": 0.05192993334366765,
      "grad_norm": 0.45912545919418335,
      "learning_rate": 2e-05,
      "loss": 0.672,
      "step": 670
    },
    {
      "epoch": 0.05270500697566269,
      "grad_norm": 0.43527087569236755,
      "learning_rate": 2e-05,
      "loss": 0.6772,
      "step": 680
    },
    {
      "epoch": 0.05348008060765773,
      "grad_norm": 0.41790688037872314,
      "learning_rate": 2e-05,
      "loss": 0.6748,
      "step": 690
    },
    {
      "epoch": 0.05425515423965277,
      "grad_norm": 0.4713754951953888,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 700
    },
    {
      "epoch": 0.055030227871647804,
      "grad_norm": 0.44159114360809326,
      "learning_rate": 2e-05,
      "loss": 0.6666,
      "step": 710
    },
    {
      "epoch": 0.055805301503642844,
      "grad_norm": 0.4281693696975708,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 720
    },
    {
      "epoch": 0.056580375135637885,
      "grad_norm": 0.42915165424346924,
      "learning_rate": 2e-05,
      "loss": 0.6745,
      "step": 730
    },
    {
      "epoch": 0.057355448767632926,
      "grad_norm": 0.4454156160354614,
      "learning_rate": 2e-05,
      "loss": 0.6714,
      "step": 740
    },
    {
      "epoch": 0.05813052239962797,
      "grad_norm": 0.5239861607551575,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 750
    },
    {
      "epoch": 0.05890559603162301,
      "grad_norm": 0.44252240657806396,
      "learning_rate": 2e-05,
      "loss": 0.6731,
      "step": 760
    },
    {
      "epoch": 0.05968066966361804,
      "grad_norm": 0.5074107646942139,
      "learning_rate": 2e-05,
      "loss": 0.684,
      "step": 770
    },
    {
      "epoch": 0.06045574329561308,
      "grad_norm": 0.4552975296974182,
      "learning_rate": 2e-05,
      "loss": 0.6836,
      "step": 780
    },
    {
      "epoch": 0.06123081692760812,
      "grad_norm": 0.42174383997917175,
      "learning_rate": 2e-05,
      "loss": 0.6717,
      "step": 790
    },
    {
      "epoch": 0.06200589055960316,
      "grad_norm": 0.44547125697135925,
      "learning_rate": 2e-05,
      "loss": 0.6775,
      "step": 800
    },
    {
      "epoch": 0.0627809641915982,
      "grad_norm": 0.44405683875083923,
      "learning_rate": 2e-05,
      "loss": 0.689,
      "step": 810
    },
    {
      "epoch": 0.06355603782359324,
      "grad_norm": 0.42822331190109253,
      "learning_rate": 2e-05,
      "loss": 0.6822,
      "step": 820
    },
    {
      "epoch": 0.06433111145558829,
      "grad_norm": 0.402357280254364,
      "learning_rate": 2e-05,
      "loss": 0.6716,
      "step": 830
    },
    {
      "epoch": 0.06510618508758333,
      "grad_norm": 0.43392375111579895,
      "learning_rate": 2e-05,
      "loss": 0.6689,
      "step": 840
    },
    {
      "epoch": 0.06588125871957835,
      "grad_norm": 0.4689558744430542,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 850
    },
    {
      "epoch": 0.0666563323515734,
      "grad_norm": 0.4157777428627014,
      "learning_rate": 2e-05,
      "loss": 0.6812,
      "step": 860
    },
    {
      "epoch": 0.06743140598356843,
      "grad_norm": 0.5666890740394592,
      "learning_rate": 2e-05,
      "loss": 0.6693,
      "step": 870
    },
    {
      "epoch": 0.06820647961556348,
      "grad_norm": 0.40637901425361633,
      "learning_rate": 2e-05,
      "loss": 0.6894,
      "step": 880
    },
    {
      "epoch": 0.06898155324755852,
      "grad_norm": 0.42309945821762085,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 890
    },
    {
      "epoch": 0.06975662687955356,
      "grad_norm": 0.429792195558548,
      "learning_rate": 2e-05,
      "loss": 0.6767,
      "step": 900
    },
    {
      "epoch": 0.0705317005115486,
      "grad_norm": 0.440060555934906,
      "learning_rate": 2e-05,
      "loss": 0.6687,
      "step": 910
    },
    {
      "epoch": 0.07130677414354364,
      "grad_norm": 0.4015533924102783,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 920
    },
    {
      "epoch": 0.07208184777553868,
      "grad_norm": 0.43568137288093567,
      "learning_rate": 2e-05,
      "loss": 0.663,
      "step": 930
    },
    {
      "epoch": 0.07285692140753372,
      "grad_norm": 0.4287867844104767,
      "learning_rate": 2e-05,
      "loss": 0.677,
      "step": 940
    },
    {
      "epoch": 0.07363199503952876,
      "grad_norm": 0.43465718626976013,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 950
    },
    {
      "epoch": 0.0744070686715238,
      "grad_norm": 0.4249207675457001,
      "learning_rate": 2e-05,
      "loss": 0.6627,
      "step": 960
    },
    {
      "epoch": 0.07518214230351883,
      "grad_norm": 0.42393025755882263,
      "learning_rate": 2e-05,
      "loss": 0.6832,
      "step": 970
    },
    {
      "epoch": 0.07595721593551387,
      "grad_norm": 0.442570298910141,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 980
    },
    {
      "epoch": 0.07673228956750891,
      "grad_norm": 0.5258124470710754,
      "learning_rate": 2e-05,
      "loss": 0.6925,
      "step": 990
    },
    {
      "epoch": 0.07750736319950395,
      "grad_norm": 0.43202176690101624,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 1000
    },
    {
      "epoch": 0.07828243683149899,
      "grad_norm": 0.39281439781188965,
      "learning_rate": 2e-05,
      "loss": 0.6673,
      "step": 1010
    },
    {
      "epoch": 0.07905751046349403,
      "grad_norm": 0.7493364214897156,
      "learning_rate": 2e-05,
      "loss": 0.6652,
      "step": 1020
    },
    {
      "epoch": 0.07983258409548907,
      "grad_norm": 0.4130217730998993,
      "learning_rate": 2e-05,
      "loss": 0.6437,
      "step": 1030
    },
    {
      "epoch": 0.08060765772748411,
      "grad_norm": 0.4456067681312561,
      "learning_rate": 2e-05,
      "loss": 0.6876,
      "step": 1040
    },
    {
      "epoch": 0.08138273135947915,
      "grad_norm": 0.50323885679245,
      "learning_rate": 2e-05,
      "loss": 0.6567,
      "step": 1050
    },
    {
      "epoch": 0.0821578049914742,
      "grad_norm": 0.44397807121276855,
      "learning_rate": 2e-05,
      "loss": 0.66,
      "step": 1060
    },
    {
      "epoch": 0.08293287862346924,
      "grad_norm": 0.4491206407546997,
      "learning_rate": 2e-05,
      "loss": 0.6703,
      "step": 1070
    },
    {
      "epoch": 0.08370795225546426,
      "grad_norm": 0.41675621271133423,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 1080
    },
    {
      "epoch": 0.0844830258874593,
      "grad_norm": 0.4307570457458496,
      "learning_rate": 2e-05,
      "loss": 0.6551,
      "step": 1090
    },
    {
      "epoch": 0.08525809951945434,
      "grad_norm": 0.4013335406780243,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 1100
    },
    {
      "epoch": 0.08603317315144939,
      "grad_norm": 0.42541980743408203,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 1110
    },
    {
      "epoch": 0.08680824678344443,
      "grad_norm": 0.4157240390777588,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 1120
    },
    {
      "epoch": 0.08758332041543947,
      "grad_norm": 0.3873870074748993,
      "learning_rate": 2e-05,
      "loss": 0.6654,
      "step": 1130
    },
    {
      "epoch": 0.08835839404743451,
      "grad_norm": 0.4169309735298157,
      "learning_rate": 2e-05,
      "loss": 0.6718,
      "step": 1140
    },
    {
      "epoch": 0.08913346767942955,
      "grad_norm": 0.4192080497741699,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 1150
    },
    {
      "epoch": 0.08990854131142459,
      "grad_norm": 0.3823591470718384,
      "learning_rate": 2e-05,
      "loss": 0.676,
      "step": 1160
    },
    {
      "epoch": 0.09068361494341963,
      "grad_norm": 0.4174516797065735,
      "learning_rate": 2e-05,
      "loss": 0.6688,
      "step": 1170
    },
    {
      "epoch": 0.09145868857541467,
      "grad_norm": 0.42104941606521606,
      "learning_rate": 2e-05,
      "loss": 0.6878,
      "step": 1180
    },
    {
      "epoch": 0.0922337622074097,
      "grad_norm": 0.42700275778770447,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 1190
    },
    {
      "epoch": 0.09300883583940474,
      "grad_norm": 0.4459708034992218,
      "learning_rate": 2e-05,
      "loss": 0.6735,
      "step": 1200
    },
    {
      "epoch": 0.09378390947139978,
      "grad_norm": 0.4184398353099823,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 1210
    },
    {
      "epoch": 0.09455898310339482,
      "grad_norm": 0.40880343317985535,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 1220
    },
    {
      "epoch": 0.09533405673538986,
      "grad_norm": 0.422642320394516,
      "learning_rate": 2e-05,
      "loss": 0.6478,
      "step": 1230
    },
    {
      "epoch": 0.0961091303673849,
      "grad_norm": 0.40179723501205444,
      "learning_rate": 2e-05,
      "loss": 0.6808,
      "step": 1240
    },
    {
      "epoch": 0.09688420399937994,
      "grad_norm": 0.44401225447654724,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 1250
    },
    {
      "epoch": 0.09765927763137498,
      "grad_norm": 0.4101133346557617,
      "learning_rate": 2e-05,
      "loss": 0.6558,
      "step": 1260
    },
    {
      "epoch": 0.09843435126337002,
      "grad_norm": 0.3988134562969208,
      "learning_rate": 2e-05,
      "loss": 0.6706,
      "step": 1270
    },
    {
      "epoch": 0.09920942489536506,
      "grad_norm": 0.43161827325820923,
      "learning_rate": 2e-05,
      "loss": 0.6577,
      "step": 1280
    },
    {
      "epoch": 0.0999844985273601,
      "grad_norm": 0.4305391311645508,
      "learning_rate": 2e-05,
      "loss": 0.6735,
      "step": 1290
    },
    {
      "epoch": 0.10075957215935513,
      "grad_norm": 0.5483478307723999,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 1300
    },
    {
      "epoch": 0.10153464579135017,
      "grad_norm": 0.43918097019195557,
      "learning_rate": 2e-05,
      "loss": 0.6843,
      "step": 1310
    },
    {
      "epoch": 0.10230971942334521,
      "grad_norm": 0.4245496392250061,
      "learning_rate": 2e-05,
      "loss": 0.6784,
      "step": 1320
    },
    {
      "epoch": 0.10308479305534025,
      "grad_norm": 0.42245975136756897,
      "learning_rate": 2e-05,
      "loss": 0.6602,
      "step": 1330
    },
    {
      "epoch": 0.1038598666873353,
      "grad_norm": 0.42837628722190857,
      "learning_rate": 2e-05,
      "loss": 0.6665,
      "step": 1340
    },
    {
      "epoch": 0.10463494031933034,
      "grad_norm": 0.4032275080680847,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 1350
    },
    {
      "epoch": 0.10541001395132538,
      "grad_norm": 0.41980066895484924,
      "learning_rate": 2e-05,
      "loss": 0.6516,
      "step": 1360
    },
    {
      "epoch": 0.10618508758332042,
      "grad_norm": 0.4778800308704376,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 1370
    },
    {
      "epoch": 0.10696016121531546,
      "grad_norm": 0.4086174964904785,
      "learning_rate": 2e-05,
      "loss": 0.6706,
      "step": 1380
    },
    {
      "epoch": 0.1077352348473105,
      "grad_norm": 0.39581477642059326,
      "learning_rate": 2e-05,
      "loss": 0.6601,
      "step": 1390
    },
    {
      "epoch": 0.10851030847930554,
      "grad_norm": 0.4274316132068634,
      "learning_rate": 2e-05,
      "loss": 0.6553,
      "step": 1400
    },
    {
      "epoch": 0.10928538211130058,
      "grad_norm": 1.0265446901321411,
      "learning_rate": 2e-05,
      "loss": 0.648,
      "step": 1410
    },
    {
      "epoch": 0.11006045574329561,
      "grad_norm": 0.3895666003227234,
      "learning_rate": 2e-05,
      "loss": 0.6377,
      "step": 1420
    },
    {
      "epoch": 0.11083552937529065,
      "grad_norm": 0.40829238295555115,
      "learning_rate": 2e-05,
      "loss": 0.659,
      "step": 1430
    },
    {
      "epoch": 0.11161060300728569,
      "grad_norm": 0.417646199464798,
      "learning_rate": 2e-05,
      "loss": 0.6935,
      "step": 1440
    },
    {
      "epoch": 0.11238567663928073,
      "grad_norm": 0.4010778069496155,
      "learning_rate": 2e-05,
      "loss": 0.6758,
      "step": 1450
    },
    {
      "epoch": 0.11316075027127577,
      "grad_norm": 0.4173222780227661,
      "learning_rate": 2e-05,
      "loss": 0.6724,
      "step": 1460
    },
    {
      "epoch": 0.11393582390327081,
      "grad_norm": 0.4018092155456543,
      "learning_rate": 2e-05,
      "loss": 0.6701,
      "step": 1470
    },
    {
      "epoch": 0.11471089753526585,
      "grad_norm": 0.43614667654037476,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 1480
    },
    {
      "epoch": 0.11548597116726089,
      "grad_norm": 0.41816315054893494,
      "learning_rate": 2e-05,
      "loss": 0.6773,
      "step": 1490
    },
    {
      "epoch": 0.11626104479925593,
      "grad_norm": 0.39446917176246643,
      "learning_rate": 2e-05,
      "loss": 0.6608,
      "step": 1500
    },
    {
      "epoch": 0.11703611843125097,
      "grad_norm": 0.447988897562027,
      "learning_rate": 2e-05,
      "loss": 0.6654,
      "step": 1510
    },
    {
      "epoch": 0.11781119206324601,
      "grad_norm": 0.41491132974624634,
      "learning_rate": 2e-05,
      "loss": 0.6858,
      "step": 1520
    },
    {
      "epoch": 0.11858626569524104,
      "grad_norm": 0.406842976808548,
      "learning_rate": 2e-05,
      "loss": 0.6686,
      "step": 1530
    },
    {
      "epoch": 0.11936133932723608,
      "grad_norm": 0.6441549062728882,
      "learning_rate": 2e-05,
      "loss": 0.6552,
      "step": 1540
    },
    {
      "epoch": 0.12013641295923112,
      "grad_norm": 0.4140605628490448,
      "learning_rate": 2e-05,
      "loss": 0.6547,
      "step": 1550
    },
    {
      "epoch": 0.12091148659122616,
      "grad_norm": 0.41722798347473145,
      "learning_rate": 2e-05,
      "loss": 0.6582,
      "step": 1560
    },
    {
      "epoch": 0.1216865602232212,
      "grad_norm": 0.43592479825019836,
      "learning_rate": 2e-05,
      "loss": 0.6687,
      "step": 1570
    },
    {
      "epoch": 0.12246163385521625,
      "grad_norm": 0.4203198552131653,
      "learning_rate": 2e-05,
      "loss": 0.6691,
      "step": 1580
    },
    {
      "epoch": 0.12323670748721129,
      "grad_norm": 0.4950765371322632,
      "learning_rate": 2e-05,
      "loss": 0.6769,
      "step": 1590
    },
    {
      "epoch": 0.12401178111920633,
      "grad_norm": 0.4129418432712555,
      "learning_rate": 2e-05,
      "loss": 0.6696,
      "step": 1600
    },
    {
      "epoch": 0.12478685475120137,
      "grad_norm": 0.4015454053878784,
      "learning_rate": 2e-05,
      "loss": 0.6594,
      "step": 1610
    },
    {
      "epoch": 0.1255619283831964,
      "grad_norm": 0.40145838260650635,
      "learning_rate": 2e-05,
      "loss": 0.6548,
      "step": 1620
    },
    {
      "epoch": 0.12633700201519144,
      "grad_norm": 0.40511614084243774,
      "learning_rate": 2e-05,
      "loss": 0.6583,
      "step": 1630
    },
    {
      "epoch": 0.1271120756471865,
      "grad_norm": 0.4618526101112366,
      "learning_rate": 2e-05,
      "loss": 0.6596,
      "step": 1640
    },
    {
      "epoch": 0.12788714927918152,
      "grad_norm": 0.4379795789718628,
      "learning_rate": 2e-05,
      "loss": 0.6816,
      "step": 1650
    },
    {
      "epoch": 0.12866222291117657,
      "grad_norm": 0.5952076315879822,
      "learning_rate": 2e-05,
      "loss": 0.6773,
      "step": 1660
    },
    {
      "epoch": 0.1294372965431716,
      "grad_norm": 0.39318859577178955,
      "learning_rate": 2e-05,
      "loss": 0.6531,
      "step": 1670
    },
    {
      "epoch": 0.13021237017516665,
      "grad_norm": 0.4082723557949066,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 1680
    },
    {
      "epoch": 0.13098744380716168,
      "grad_norm": 0.4530302584171295,
      "learning_rate": 2e-05,
      "loss": 0.6684,
      "step": 1690
    },
    {
      "epoch": 0.1317625174391567,
      "grad_norm": 0.43130335211753845,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 1700
    },
    {
      "epoch": 0.13253759107115176,
      "grad_norm": 0.4232826828956604,
      "learning_rate": 2e-05,
      "loss": 0.6621,
      "step": 1710
    },
    {
      "epoch": 0.1333126647031468,
      "grad_norm": 0.4304817318916321,
      "learning_rate": 2e-05,
      "loss": 0.6542,
      "step": 1720
    },
    {
      "epoch": 0.13408773833514184,
      "grad_norm": 0.40232929587364197,
      "learning_rate": 2e-05,
      "loss": 0.6736,
      "step": 1730
    },
    {
      "epoch": 0.13486281196713687,
      "grad_norm": 0.39328262209892273,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 1740
    },
    {
      "epoch": 0.13563788559913192,
      "grad_norm": 0.384797602891922,
      "learning_rate": 2e-05,
      "loss": 0.6591,
      "step": 1750
    },
    {
      "epoch": 0.13641295923112695,
      "grad_norm": 0.39744696021080017,
      "learning_rate": 2e-05,
      "loss": 0.6552,
      "step": 1760
    },
    {
      "epoch": 0.137188032863122,
      "grad_norm": 0.8520597219467163,
      "learning_rate": 2e-05,
      "loss": 0.6649,
      "step": 1770
    },
    {
      "epoch": 0.13796310649511703,
      "grad_norm": 0.402743399143219,
      "learning_rate": 2e-05,
      "loss": 0.6639,
      "step": 1780
    },
    {
      "epoch": 0.1387381801271121,
      "grad_norm": 0.40198618173599243,
      "learning_rate": 2e-05,
      "loss": 0.6653,
      "step": 1790
    },
    {
      "epoch": 0.13951325375910711,
      "grad_norm": 0.41853809356689453,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 1800
    },
    {
      "epoch": 0.14028832739110214,
      "grad_norm": 1.1882494688034058,
      "learning_rate": 2e-05,
      "loss": 0.6481,
      "step": 1810
    },
    {
      "epoch": 0.1410634010230972,
      "grad_norm": 0.5747016668319702,
      "learning_rate": 2e-05,
      "loss": 0.662,
      "step": 1820
    },
    {
      "epoch": 0.14183847465509222,
      "grad_norm": 0.41559216380119324,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 1830
    },
    {
      "epoch": 0.14261354828708728,
      "grad_norm": 0.3821167051792145,
      "learning_rate": 2e-05,
      "loss": 0.6656,
      "step": 1840
    },
    {
      "epoch": 0.1433886219190823,
      "grad_norm": 0.4053109884262085,
      "learning_rate": 2e-05,
      "loss": 0.658,
      "step": 1850
    },
    {
      "epoch": 0.14416369555107736,
      "grad_norm": 0.8168757557868958,
      "learning_rate": 2e-05,
      "loss": 0.67,
      "step": 1860
    },
    {
      "epoch": 0.14493876918307239,
      "grad_norm": 0.42168760299682617,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 1870
    },
    {
      "epoch": 0.14571384281506744,
      "grad_norm": 0.42126354575157166,
      "learning_rate": 2e-05,
      "loss": 0.6661,
      "step": 1880
    },
    {
      "epoch": 0.14648891644706247,
      "grad_norm": 0.4499135911464691,
      "learning_rate": 2e-05,
      "loss": 0.6592,
      "step": 1890
    },
    {
      "epoch": 0.14726399007905752,
      "grad_norm": 0.42486631870269775,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 1900
    },
    {
      "epoch": 0.14803906371105255,
      "grad_norm": 0.3847733736038208,
      "learning_rate": 2e-05,
      "loss": 0.6635,
      "step": 1910
    },
    {
      "epoch": 0.1488141373430476,
      "grad_norm": 0.4100988507270813,
      "learning_rate": 2e-05,
      "loss": 0.6558,
      "step": 1920
    },
    {
      "epoch": 0.14958921097504263,
      "grad_norm": 0.4142761826515198,
      "learning_rate": 2e-05,
      "loss": 0.6635,
      "step": 1930
    },
    {
      "epoch": 0.15036428460703766,
      "grad_norm": 0.40710175037384033,
      "learning_rate": 2e-05,
      "loss": 0.6742,
      "step": 1940
    },
    {
      "epoch": 0.1511393582390327,
      "grad_norm": 0.4328700602054596,
      "learning_rate": 2e-05,
      "loss": 0.6715,
      "step": 1950
    },
    {
      "epoch": 0.15191443187102774,
      "grad_norm": 0.605353593826294,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 1960
    },
    {
      "epoch": 0.1526895055030228,
      "grad_norm": 0.5102190971374512,
      "learning_rate": 2e-05,
      "loss": 0.6644,
      "step": 1970
    },
    {
      "epoch": 0.15346457913501782,
      "grad_norm": 0.45153525471687317,
      "learning_rate": 2e-05,
      "loss": 0.6451,
      "step": 1980
    },
    {
      "epoch": 0.15423965276701287,
      "grad_norm": 0.4187360107898712,
      "learning_rate": 2e-05,
      "loss": 0.6549,
      "step": 1990
    },
    {
      "epoch": 0.1550147263990079,
      "grad_norm": 0.41170287132263184,
      "learning_rate": 2e-05,
      "loss": 0.6661,
      "step": 2000
    },
    {
      "epoch": 0.15578980003100296,
      "grad_norm": 0.3859499990940094,
      "learning_rate": 2e-05,
      "loss": 0.6571,
      "step": 2010
    },
    {
      "epoch": 0.15656487366299798,
      "grad_norm": 0.38852736353874207,
      "learning_rate": 2e-05,
      "loss": 0.6679,
      "step": 2020
    },
    {
      "epoch": 0.15733994729499304,
      "grad_norm": 0.4105587601661682,
      "learning_rate": 2e-05,
      "loss": 0.6436,
      "step": 2030
    },
    {
      "epoch": 0.15811502092698806,
      "grad_norm": 0.4161953628063202,
      "learning_rate": 2e-05,
      "loss": 0.6683,
      "step": 2040
    },
    {
      "epoch": 0.1588900945589831,
      "grad_norm": 0.41664355993270874,
      "learning_rate": 2e-05,
      "loss": 0.6534,
      "step": 2050
    },
    {
      "epoch": 0.15966516819097815,
      "grad_norm": 0.40171241760253906,
      "learning_rate": 2e-05,
      "loss": 0.6569,
      "step": 2060
    },
    {
      "epoch": 0.16044024182297317,
      "grad_norm": 1.195680856704712,
      "learning_rate": 2e-05,
      "loss": 0.6538,
      "step": 2070
    },
    {
      "epoch": 0.16121531545496823,
      "grad_norm": 0.446020245552063,
      "learning_rate": 2e-05,
      "loss": 0.6706,
      "step": 2080
    },
    {
      "epoch": 0.16199038908696325,
      "grad_norm": 0.4383505582809448,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 2090
    },
    {
      "epoch": 0.1627654627189583,
      "grad_norm": 0.4303628206253052,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 2100
    },
    {
      "epoch": 0.16354053635095334,
      "grad_norm": 0.46713578701019287,
      "learning_rate": 2e-05,
      "loss": 0.6757,
      "step": 2110
    },
    {
      "epoch": 0.1643156099829484,
      "grad_norm": 0.41250234842300415,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 2120
    },
    {
      "epoch": 0.16509068361494342,
      "grad_norm": 0.4887694716453552,
      "learning_rate": 2e-05,
      "loss": 0.6659,
      "step": 2130
    },
    {
      "epoch": 0.16586575724693847,
      "grad_norm": 0.43109914660453796,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 2140
    },
    {
      "epoch": 0.1666408308789335,
      "grad_norm": 0.41217032074928284,
      "learning_rate": 2e-05,
      "loss": 0.6779,
      "step": 2150
    },
    {
      "epoch": 0.16741590451092853,
      "grad_norm": 0.4060550332069397,
      "learning_rate": 2e-05,
      "loss": 0.6542,
      "step": 2160
    },
    {
      "epoch": 0.16819097814292358,
      "grad_norm": 0.4079914093017578,
      "learning_rate": 2e-05,
      "loss": 0.6702,
      "step": 2170
    },
    {
      "epoch": 0.1689660517749186,
      "grad_norm": 0.3878198266029358,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 2180
    },
    {
      "epoch": 0.16974112540691366,
      "grad_norm": 0.4579482972621918,
      "learning_rate": 2e-05,
      "loss": 0.6517,
      "step": 2190
    },
    {
      "epoch": 0.1705161990389087,
      "grad_norm": 0.41060948371887207,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 2200
    },
    {
      "epoch": 0.17129127267090374,
      "grad_norm": 0.3973774015903473,
      "learning_rate": 2e-05,
      "loss": 0.6547,
      "step": 2210
    },
    {
      "epoch": 0.17206634630289877,
      "grad_norm": 0.40613505244255066,
      "learning_rate": 2e-05,
      "loss": 0.6665,
      "step": 2220
    },
    {
      "epoch": 0.17284141993489383,
      "grad_norm": 0.44879674911499023,
      "learning_rate": 2e-05,
      "loss": 0.6603,
      "step": 2230
    },
    {
      "epoch": 0.17361649356688885,
      "grad_norm": 0.49533888697624207,
      "learning_rate": 2e-05,
      "loss": 0.6454,
      "step": 2240
    },
    {
      "epoch": 0.1743915671988839,
      "grad_norm": 0.5093989372253418,
      "learning_rate": 2e-05,
      "loss": 0.6702,
      "step": 2250
    },
    {
      "epoch": 0.17516664083087893,
      "grad_norm": 0.4315301477909088,
      "learning_rate": 2e-05,
      "loss": 0.6417,
      "step": 2260
    },
    {
      "epoch": 0.17594171446287396,
      "grad_norm": 0.6004558205604553,
      "learning_rate": 2e-05,
      "loss": 0.643,
      "step": 2270
    },
    {
      "epoch": 0.17671678809486902,
      "grad_norm": 0.4064202308654785,
      "learning_rate": 2e-05,
      "loss": 0.6729,
      "step": 2280
    },
    {
      "epoch": 0.17749186172686404,
      "grad_norm": 0.4044870436191559,
      "learning_rate": 2e-05,
      "loss": 0.6475,
      "step": 2290
    },
    {
      "epoch": 0.1782669353588591,
      "grad_norm": 0.38951683044433594,
      "learning_rate": 2e-05,
      "loss": 0.6509,
      "step": 2300
    },
    {
      "epoch": 0.17904200899085412,
      "grad_norm": 0.4390566647052765,
      "learning_rate": 2e-05,
      "loss": 0.6602,
      "step": 2310
    },
    {
      "epoch": 0.17981708262284918,
      "grad_norm": 0.4063456058502197,
      "learning_rate": 2e-05,
      "loss": 0.6586,
      "step": 2320
    },
    {
      "epoch": 0.1805921562548442,
      "grad_norm": 0.38408976793289185,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 2330
    },
    {
      "epoch": 0.18136722988683926,
      "grad_norm": 0.39897775650024414,
      "learning_rate": 2e-05,
      "loss": 0.6658,
      "step": 2340
    },
    {
      "epoch": 0.1821423035188343,
      "grad_norm": 1.0825657844543457,
      "learning_rate": 2e-05,
      "loss": 0.6653,
      "step": 2350
    },
    {
      "epoch": 0.18291737715082934,
      "grad_norm": 0.4177400767803192,
      "learning_rate": 2e-05,
      "loss": 0.648,
      "step": 2360
    },
    {
      "epoch": 0.18369245078282437,
      "grad_norm": 0.4209858179092407,
      "learning_rate": 2e-05,
      "loss": 0.6449,
      "step": 2370
    },
    {
      "epoch": 0.1844675244148194,
      "grad_norm": 0.39963626861572266,
      "learning_rate": 2e-05,
      "loss": 0.6507,
      "step": 2380
    },
    {
      "epoch": 0.18524259804681445,
      "grad_norm": 0.39542877674102783,
      "learning_rate": 2e-05,
      "loss": 0.6475,
      "step": 2390
    },
    {
      "epoch": 0.18601767167880948,
      "grad_norm": 0.3948878049850464,
      "learning_rate": 2e-05,
      "loss": 0.6599,
      "step": 2400
    },
    {
      "epoch": 0.18679274531080453,
      "grad_norm": 0.4023815095424652,
      "learning_rate": 2e-05,
      "loss": 0.6727,
      "step": 2410
    },
    {
      "epoch": 0.18756781894279956,
      "grad_norm": 0.5075188279151917,
      "learning_rate": 2e-05,
      "loss": 0.6585,
      "step": 2420
    },
    {
      "epoch": 0.1883428925747946,
      "grad_norm": 0.3735017776489258,
      "learning_rate": 2e-05,
      "loss": 0.6591,
      "step": 2430
    },
    {
      "epoch": 0.18911796620678964,
      "grad_norm": 0.4221167266368866,
      "learning_rate": 2e-05,
      "loss": 0.6415,
      "step": 2440
    },
    {
      "epoch": 0.1898930398387847,
      "grad_norm": 0.41844654083251953,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 2450
    },
    {
      "epoch": 0.19066811347077972,
      "grad_norm": 0.3794083297252655,
      "learning_rate": 2e-05,
      "loss": 0.6435,
      "step": 2460
    },
    {
      "epoch": 0.19144318710277478,
      "grad_norm": 0.41103601455688477,
      "learning_rate": 2e-05,
      "loss": 0.6431,
      "step": 2470
    },
    {
      "epoch": 0.1922182607347698,
      "grad_norm": 0.39791375398635864,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 2480
    },
    {
      "epoch": 0.19299333436676483,
      "grad_norm": 0.3982989192008972,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 2490
    },
    {
      "epoch": 0.19376840799875988,
      "grad_norm": 0.42089393734931946,
      "learning_rate": 2e-05,
      "loss": 0.6697,
      "step": 2500
    },
    {
      "epoch": 0.1945434816307549,
      "grad_norm": 0.4063425660133362,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 2510
    },
    {
      "epoch": 0.19531855526274997,
      "grad_norm": 0.37145042419433594,
      "learning_rate": 2e-05,
      "loss": 0.6432,
      "step": 2520
    },
    {
      "epoch": 0.196093628894745,
      "grad_norm": 0.4100053906440735,
      "learning_rate": 2e-05,
      "loss": 0.6688,
      "step": 2530
    },
    {
      "epoch": 0.19686870252674005,
      "grad_norm": 0.37570226192474365,
      "learning_rate": 2e-05,
      "loss": 0.6436,
      "step": 2540
    },
    {
      "epoch": 0.19764377615873507,
      "grad_norm": 0.404792457818985,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 2550
    },
    {
      "epoch": 0.19841884979073013,
      "grad_norm": 0.38528645038604736,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 2560
    },
    {
      "epoch": 0.19919392342272516,
      "grad_norm": 0.4154411554336548,
      "learning_rate": 2e-05,
      "loss": 0.654,
      "step": 2570
    },
    {
      "epoch": 0.1999689970547202,
      "grad_norm": 0.4468025863170624,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 2580
    },
    {
      "epoch": 0.20074407068671524,
      "grad_norm": 0.43527552485466003,
      "learning_rate": 2e-05,
      "loss": 0.6736,
      "step": 2590
    },
    {
      "epoch": 0.20151914431871026,
      "grad_norm": 0.39578428864479065,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 2600
    },
    {
      "epoch": 0.20229421795070532,
      "grad_norm": 0.4273620545864105,
      "learning_rate": 2e-05,
      "loss": 0.6502,
      "step": 2610
    },
    {
      "epoch": 0.20306929158270035,
      "grad_norm": 0.48650169372558594,
      "learning_rate": 2e-05,
      "loss": 0.6485,
      "step": 2620
    },
    {
      "epoch": 0.2038443652146954,
      "grad_norm": 0.4622916281223297,
      "learning_rate": 2e-05,
      "loss": 0.6569,
      "step": 2630
    },
    {
      "epoch": 0.20461943884669043,
      "grad_norm": 0.4230853021144867,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 2640
    },
    {
      "epoch": 0.20539451247868548,
      "grad_norm": 0.39467793703079224,
      "learning_rate": 2e-05,
      "loss": 0.6434,
      "step": 2650
    },
    {
      "epoch": 0.2061695861106805,
      "grad_norm": 0.41830822825431824,
      "learning_rate": 2e-05,
      "loss": 0.6484,
      "step": 2660
    },
    {
      "epoch": 0.20694465974267556,
      "grad_norm": 0.5037385821342468,
      "learning_rate": 2e-05,
      "loss": 0.6443,
      "step": 2670
    },
    {
      "epoch": 0.2077197333746706,
      "grad_norm": 0.38853520154953003,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 2680
    },
    {
      "epoch": 0.20849480700666564,
      "grad_norm": 0.414017915725708,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 2690
    },
    {
      "epoch": 0.20926988063866067,
      "grad_norm": 0.3972592055797577,
      "learning_rate": 2e-05,
      "loss": 0.6457,
      "step": 2700
    },
    {
      "epoch": 0.21004495427065573,
      "grad_norm": 0.3871782720088959,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 2710
    },
    {
      "epoch": 0.21082002790265075,
      "grad_norm": 0.3889021575450897,
      "learning_rate": 2e-05,
      "loss": 0.6526,
      "step": 2720
    },
    {
      "epoch": 0.21159510153464578,
      "grad_norm": 0.3956632614135742,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 2730
    },
    {
      "epoch": 0.21237017516664083,
      "grad_norm": 0.39243122935295105,
      "learning_rate": 2e-05,
      "loss": 0.6403,
      "step": 2740
    },
    {
      "epoch": 0.21314524879863586,
      "grad_norm": 0.40234375,
      "learning_rate": 2e-05,
      "loss": 0.6545,
      "step": 2750
    },
    {
      "epoch": 0.21392032243063092,
      "grad_norm": 0.37857896089553833,
      "learning_rate": 2e-05,
      "loss": 0.6445,
      "step": 2760
    },
    {
      "epoch": 0.21469539606262594,
      "grad_norm": 0.41570788621902466,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 2770
    },
    {
      "epoch": 0.215470469694621,
      "grad_norm": 0.4410853981971741,
      "learning_rate": 2e-05,
      "loss": 0.6608,
      "step": 2780
    },
    {
      "epoch": 0.21624554332661602,
      "grad_norm": 0.44700971245765686,
      "learning_rate": 2e-05,
      "loss": 0.6605,
      "step": 2790
    },
    {
      "epoch": 0.21702061695861108,
      "grad_norm": 0.40519022941589355,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 2800
    },
    {
      "epoch": 0.2177956905906061,
      "grad_norm": 0.42666929960250854,
      "learning_rate": 2e-05,
      "loss": 0.6515,
      "step": 2810
    },
    {
      "epoch": 0.21857076422260116,
      "grad_norm": 0.412100225687027,
      "learning_rate": 2e-05,
      "loss": 0.6383,
      "step": 2820
    },
    {
      "epoch": 0.2193458378545962,
      "grad_norm": 0.40622761845588684,
      "learning_rate": 2e-05,
      "loss": 0.6567,
      "step": 2830
    },
    {
      "epoch": 0.22012091148659121,
      "grad_norm": 0.41891127824783325,
      "learning_rate": 2e-05,
      "loss": 0.6531,
      "step": 2840
    },
    {
      "epoch": 0.22089598511858627,
      "grad_norm": 0.4018741548061371,
      "learning_rate": 2e-05,
      "loss": 0.6469,
      "step": 2850
    },
    {
      "epoch": 0.2216710587505813,
      "grad_norm": 0.4432869851589203,
      "learning_rate": 2e-05,
      "loss": 0.6656,
      "step": 2860
    },
    {
      "epoch": 0.22244613238257635,
      "grad_norm": 0.4150429666042328,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 2870
    },
    {
      "epoch": 0.22322120601457138,
      "grad_norm": 0.4564347267150879,
      "learning_rate": 2e-05,
      "loss": 0.6451,
      "step": 2880
    },
    {
      "epoch": 0.22399627964656643,
      "grad_norm": 0.37577444314956665,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 2890
    },
    {
      "epoch": 0.22477135327856146,
      "grad_norm": 0.39881935715675354,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 2900
    },
    {
      "epoch": 0.2255464269105565,
      "grad_norm": 0.4113995134830475,
      "learning_rate": 2e-05,
      "loss": 0.6602,
      "step": 2910
    },
    {
      "epoch": 0.22632150054255154,
      "grad_norm": 0.3787456452846527,
      "learning_rate": 2e-05,
      "loss": 0.6675,
      "step": 2920
    },
    {
      "epoch": 0.2270965741745466,
      "grad_norm": 0.4072747528553009,
      "learning_rate": 2e-05,
      "loss": 0.6665,
      "step": 2930
    },
    {
      "epoch": 0.22787164780654162,
      "grad_norm": 0.41496026515960693,
      "learning_rate": 2e-05,
      "loss": 0.6515,
      "step": 2940
    },
    {
      "epoch": 0.22864672143853665,
      "grad_norm": 0.4140514135360718,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 2950
    },
    {
      "epoch": 0.2294217950705317,
      "grad_norm": 0.4361623227596283,
      "learning_rate": 2e-05,
      "loss": 0.6522,
      "step": 2960
    },
    {
      "epoch": 0.23019686870252673,
      "grad_norm": 0.3820911645889282,
      "learning_rate": 2e-05,
      "loss": 0.6398,
      "step": 2970
    },
    {
      "epoch": 0.23097194233452178,
      "grad_norm": 0.3671109974384308,
      "learning_rate": 2e-05,
      "loss": 0.6644,
      "step": 2980
    },
    {
      "epoch": 0.2317470159665168,
      "grad_norm": 0.3904815912246704,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 2990
    },
    {
      "epoch": 0.23252208959851187,
      "grad_norm": 0.40013980865478516,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 3000
    },
    {
      "epoch": 0.2332971632305069,
      "grad_norm": 0.42152348160743713,
      "learning_rate": 2e-05,
      "loss": 0.6436,
      "step": 3010
    },
    {
      "epoch": 0.23407223686250195,
      "grad_norm": 0.38032928109169006,
      "learning_rate": 2e-05,
      "loss": 0.6434,
      "step": 3020
    },
    {
      "epoch": 0.23484731049449697,
      "grad_norm": 0.39556047320365906,
      "learning_rate": 2e-05,
      "loss": 0.6457,
      "step": 3030
    },
    {
      "epoch": 0.23562238412649203,
      "grad_norm": 0.40398159623146057,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 3040
    },
    {
      "epoch": 0.23639745775848706,
      "grad_norm": 0.4593210518360138,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 3050
    },
    {
      "epoch": 0.23717253139048208,
      "grad_norm": 0.42830073833465576,
      "learning_rate": 2e-05,
      "loss": 0.654,
      "step": 3060
    },
    {
      "epoch": 0.23794760502247714,
      "grad_norm": 0.4068194329738617,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 3070
    },
    {
      "epoch": 0.23872267865447216,
      "grad_norm": 2.078359603881836,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 3080
    },
    {
      "epoch": 0.23949775228646722,
      "grad_norm": 0.6029475927352905,
      "learning_rate": 2e-05,
      "loss": 0.6542,
      "step": 3090
    },
    {
      "epoch": 0.24027282591846225,
      "grad_norm": 0.39002111554145813,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 3100
    },
    {
      "epoch": 0.2410478995504573,
      "grad_norm": 0.39298468828201294,
      "learning_rate": 2e-05,
      "loss": 0.6605,
      "step": 3110
    },
    {
      "epoch": 0.24182297318245233,
      "grad_norm": 0.4210423529148102,
      "learning_rate": 2e-05,
      "loss": 0.6351,
      "step": 3120
    },
    {
      "epoch": 0.24259804681444738,
      "grad_norm": 0.9378540515899658,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 3130
    },
    {
      "epoch": 0.2433731204464424,
      "grad_norm": 0.4258650243282318,
      "learning_rate": 2e-05,
      "loss": 0.6649,
      "step": 3140
    },
    {
      "epoch": 0.24414819407843746,
      "grad_norm": 0.40488994121551514,
      "learning_rate": 2e-05,
      "loss": 0.665,
      "step": 3150
    },
    {
      "epoch": 0.2449232677104325,
      "grad_norm": 0.4172995686531067,
      "learning_rate": 2e-05,
      "loss": 0.6691,
      "step": 3160
    },
    {
      "epoch": 0.24569834134242752,
      "grad_norm": 0.3930172920227051,
      "learning_rate": 2e-05,
      "loss": 0.6351,
      "step": 3170
    },
    {
      "epoch": 0.24647341497442257,
      "grad_norm": 0.4278660714626312,
      "learning_rate": 2e-05,
      "loss": 0.6396,
      "step": 3180
    },
    {
      "epoch": 0.2472484886064176,
      "grad_norm": 0.40976083278656006,
      "learning_rate": 2e-05,
      "loss": 0.646,
      "step": 3190
    },
    {
      "epoch": 0.24802356223841265,
      "grad_norm": 0.4093548357486725,
      "learning_rate": 2e-05,
      "loss": 0.6448,
      "step": 3200
    },
    {
      "epoch": 0.24879863587040768,
      "grad_norm": 0.39366021752357483,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 3210
    },
    {
      "epoch": 0.24957370950240274,
      "grad_norm": 0.45590394735336304,
      "learning_rate": 2e-05,
      "loss": 0.6461,
      "step": 3220
    },
    {
      "epoch": 0.2503487831343978,
      "grad_norm": 0.4243846833705902,
      "learning_rate": 2e-05,
      "loss": 0.6582,
      "step": 3230
    },
    {
      "epoch": 0.2511238567663928,
      "grad_norm": 0.42751291394233704,
      "learning_rate": 2e-05,
      "loss": 0.6521,
      "step": 3240
    },
    {
      "epoch": 0.25189893039838784,
      "grad_norm": 0.3897493779659271,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 3250
    },
    {
      "epoch": 0.25267400403038287,
      "grad_norm": 0.3933364450931549,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 3260
    },
    {
      "epoch": 0.25344907766237795,
      "grad_norm": 0.3864021599292755,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 3270
    },
    {
      "epoch": 0.254224151294373,
      "grad_norm": 0.3909198045730591,
      "learning_rate": 2e-05,
      "loss": 0.6508,
      "step": 3280
    },
    {
      "epoch": 0.254999224926368,
      "grad_norm": 0.3932095170021057,
      "learning_rate": 2e-05,
      "loss": 0.6326,
      "step": 3290
    },
    {
      "epoch": 0.25577429855836303,
      "grad_norm": 0.3967517912387848,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 3300
    },
    {
      "epoch": 0.25654937219035806,
      "grad_norm": 0.4061006009578705,
      "learning_rate": 2e-05,
      "loss": 0.6433,
      "step": 3310
    },
    {
      "epoch": 0.25732444582235314,
      "grad_norm": 0.41564321517944336,
      "learning_rate": 2e-05,
      "loss": 0.6555,
      "step": 3320
    },
    {
      "epoch": 0.25809951945434817,
      "grad_norm": 0.4872203767299652,
      "learning_rate": 2e-05,
      "loss": 0.627,
      "step": 3330
    },
    {
      "epoch": 0.2588745930863432,
      "grad_norm": 0.41196078062057495,
      "learning_rate": 2e-05,
      "loss": 0.6336,
      "step": 3340
    },
    {
      "epoch": 0.2596496667183382,
      "grad_norm": 0.4241744875907898,
      "learning_rate": 2e-05,
      "loss": 0.6524,
      "step": 3350
    },
    {
      "epoch": 0.2604247403503333,
      "grad_norm": 0.40693116188049316,
      "learning_rate": 2e-05,
      "loss": 0.6443,
      "step": 3360
    },
    {
      "epoch": 0.26119981398232833,
      "grad_norm": 0.4028782546520233,
      "learning_rate": 2e-05,
      "loss": 0.6482,
      "step": 3370
    },
    {
      "epoch": 0.26197488761432336,
      "grad_norm": 0.44793808460235596,
      "learning_rate": 2e-05,
      "loss": 0.6681,
      "step": 3380
    },
    {
      "epoch": 0.2627499612463184,
      "grad_norm": 0.40758317708969116,
      "learning_rate": 2e-05,
      "loss": 0.658,
      "step": 3390
    },
    {
      "epoch": 0.2635250348783134,
      "grad_norm": 0.4020330607891083,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 3400
    },
    {
      "epoch": 0.2643001085103085,
      "grad_norm": 0.4066820442676544,
      "learning_rate": 2e-05,
      "loss": 0.6361,
      "step": 3410
    },
    {
      "epoch": 0.2650751821423035,
      "grad_norm": 0.3967213034629822,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 3420
    },
    {
      "epoch": 0.26585025577429855,
      "grad_norm": 0.41118374466896057,
      "learning_rate": 2e-05,
      "loss": 0.6421,
      "step": 3430
    },
    {
      "epoch": 0.2666253294062936,
      "grad_norm": 0.3936464190483093,
      "learning_rate": 2e-05,
      "loss": 0.6377,
      "step": 3440
    },
    {
      "epoch": 0.26740040303828866,
      "grad_norm": 0.38485974073410034,
      "learning_rate": 2e-05,
      "loss": 0.6641,
      "step": 3450
    },
    {
      "epoch": 0.2681754766702837,
      "grad_norm": 0.40260979533195496,
      "learning_rate": 2e-05,
      "loss": 0.6302,
      "step": 3460
    },
    {
      "epoch": 0.2689505503022787,
      "grad_norm": 0.37042638659477234,
      "learning_rate": 2e-05,
      "loss": 0.6387,
      "step": 3470
    },
    {
      "epoch": 0.26972562393427374,
      "grad_norm": 0.4179660975933075,
      "learning_rate": 2e-05,
      "loss": 0.6452,
      "step": 3480
    },
    {
      "epoch": 0.2705006975662688,
      "grad_norm": 0.41497689485549927,
      "learning_rate": 2e-05,
      "loss": 0.6442,
      "step": 3490
    },
    {
      "epoch": 0.27127577119826385,
      "grad_norm": 0.413665235042572,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 3500
    },
    {
      "epoch": 0.2720508448302589,
      "grad_norm": 0.48886892199516296,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 3510
    },
    {
      "epoch": 0.2728259184622539,
      "grad_norm": 0.41618990898132324,
      "learning_rate": 2e-05,
      "loss": 0.6443,
      "step": 3520
    },
    {
      "epoch": 0.27360099209424893,
      "grad_norm": 0.3849734961986542,
      "learning_rate": 2e-05,
      "loss": 0.6454,
      "step": 3530
    },
    {
      "epoch": 0.274376065726244,
      "grad_norm": 0.4238608777523041,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 3540
    },
    {
      "epoch": 0.27515113935823904,
      "grad_norm": 0.4051784574985504,
      "learning_rate": 2e-05,
      "loss": 0.6725,
      "step": 3550
    },
    {
      "epoch": 0.27592621299023407,
      "grad_norm": 0.40739062428474426,
      "learning_rate": 2e-05,
      "loss": 0.6454,
      "step": 3560
    },
    {
      "epoch": 0.2767012866222291,
      "grad_norm": 0.39898690581321716,
      "learning_rate": 2e-05,
      "loss": 0.6628,
      "step": 3570
    },
    {
      "epoch": 0.2774763602542242,
      "grad_norm": 0.39826616644859314,
      "learning_rate": 2e-05,
      "loss": 0.6479,
      "step": 3580
    },
    {
      "epoch": 0.2782514338862192,
      "grad_norm": 0.45955926179885864,
      "learning_rate": 2e-05,
      "loss": 0.6449,
      "step": 3590
    },
    {
      "epoch": 0.27902650751821423,
      "grad_norm": 0.41046255826950073,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 3600
    },
    {
      "epoch": 0.27980158115020926,
      "grad_norm": 0.41490477323532104,
      "learning_rate": 2e-05,
      "loss": 0.6344,
      "step": 3610
    },
    {
      "epoch": 0.2805766547822043,
      "grad_norm": 0.418863445520401,
      "learning_rate": 2e-05,
      "loss": 0.6572,
      "step": 3620
    },
    {
      "epoch": 0.28135172841419936,
      "grad_norm": 0.3723398745059967,
      "learning_rate": 2e-05,
      "loss": 0.6395,
      "step": 3630
    },
    {
      "epoch": 0.2821268020461944,
      "grad_norm": 0.4250670373439789,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 3640
    },
    {
      "epoch": 0.2829018756781894,
      "grad_norm": 0.41790127754211426,
      "learning_rate": 2e-05,
      "loss": 0.6462,
      "step": 3650
    },
    {
      "epoch": 0.28367694931018445,
      "grad_norm": 0.3801466226577759,
      "learning_rate": 2e-05,
      "loss": 0.6553,
      "step": 3660
    },
    {
      "epoch": 0.2844520229421795,
      "grad_norm": 0.3941139280796051,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 3670
    },
    {
      "epoch": 0.28522709657417455,
      "grad_norm": 0.3839934766292572,
      "learning_rate": 2e-05,
      "loss": 0.6357,
      "step": 3680
    },
    {
      "epoch": 0.2860021702061696,
      "grad_norm": 0.5973071455955505,
      "learning_rate": 2e-05,
      "loss": 0.6594,
      "step": 3690
    },
    {
      "epoch": 0.2867772438381646,
      "grad_norm": 0.42879819869995117,
      "learning_rate": 2e-05,
      "loss": 0.6435,
      "step": 3700
    },
    {
      "epoch": 0.2875523174701597,
      "grad_norm": 0.4851306676864624,
      "learning_rate": 2e-05,
      "loss": 0.6534,
      "step": 3710
    },
    {
      "epoch": 0.2883273911021547,
      "grad_norm": 0.3869592547416687,
      "learning_rate": 2e-05,
      "loss": 0.6456,
      "step": 3720
    },
    {
      "epoch": 0.28910246473414974,
      "grad_norm": 0.42536336183547974,
      "learning_rate": 2e-05,
      "loss": 0.6413,
      "step": 3730
    },
    {
      "epoch": 0.28987753836614477,
      "grad_norm": 0.44545114040374756,
      "learning_rate": 2e-05,
      "loss": 0.6489,
      "step": 3740
    },
    {
      "epoch": 0.2906526119981398,
      "grad_norm": 0.44202905893325806,
      "learning_rate": 2e-05,
      "loss": 0.671,
      "step": 3750
    },
    {
      "epoch": 0.2914276856301349,
      "grad_norm": 0.4026680588722229,
      "learning_rate": 2e-05,
      "loss": 0.6521,
      "step": 3760
    },
    {
      "epoch": 0.2922027592621299,
      "grad_norm": 0.4531676471233368,
      "learning_rate": 2e-05,
      "loss": 0.66,
      "step": 3770
    },
    {
      "epoch": 0.29297783289412493,
      "grad_norm": 0.5140746235847473,
      "learning_rate": 2e-05,
      "loss": 0.6237,
      "step": 3780
    },
    {
      "epoch": 0.29375290652611996,
      "grad_norm": 0.6336513161659241,
      "learning_rate": 2e-05,
      "loss": 0.6452,
      "step": 3790
    },
    {
      "epoch": 0.29452798015811504,
      "grad_norm": 0.39968904852867126,
      "learning_rate": 2e-05,
      "loss": 0.6503,
      "step": 3800
    },
    {
      "epoch": 0.29530305379011007,
      "grad_norm": 0.3805924952030182,
      "learning_rate": 2e-05,
      "loss": 0.6596,
      "step": 3810
    },
    {
      "epoch": 0.2960781274221051,
      "grad_norm": 0.38822469115257263,
      "learning_rate": 2e-05,
      "loss": 0.6417,
      "step": 3820
    },
    {
      "epoch": 0.2968532010541001,
      "grad_norm": 0.39380943775177,
      "learning_rate": 2e-05,
      "loss": 0.6461,
      "step": 3830
    },
    {
      "epoch": 0.2976282746860952,
      "grad_norm": 0.40815481543540955,
      "learning_rate": 2e-05,
      "loss": 0.6459,
      "step": 3840
    },
    {
      "epoch": 0.29840334831809023,
      "grad_norm": 0.40621674060821533,
      "learning_rate": 2e-05,
      "loss": 0.6506,
      "step": 3850
    },
    {
      "epoch": 0.29917842195008526,
      "grad_norm": 0.3831629157066345,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 3860
    },
    {
      "epoch": 0.2999534955820803,
      "grad_norm": 0.5715486407279968,
      "learning_rate": 2e-05,
      "loss": 0.6342,
      "step": 3870
    },
    {
      "epoch": 0.3007285692140753,
      "grad_norm": 0.3794034719467163,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 3880
    },
    {
      "epoch": 0.3015036428460704,
      "grad_norm": 0.4201430380344391,
      "learning_rate": 2e-05,
      "loss": 0.647,
      "step": 3890
    },
    {
      "epoch": 0.3022787164780654,
      "grad_norm": 0.3711002469062805,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 3900
    },
    {
      "epoch": 0.30305379011006045,
      "grad_norm": 0.4040549397468567,
      "learning_rate": 2e-05,
      "loss": 0.6677,
      "step": 3910
    },
    {
      "epoch": 0.3038288637420555,
      "grad_norm": 0.4129764139652252,
      "learning_rate": 2e-05,
      "loss": 0.635,
      "step": 3920
    },
    {
      "epoch": 0.30460393737405056,
      "grad_norm": 0.41761329770088196,
      "learning_rate": 2e-05,
      "loss": 0.6571,
      "step": 3930
    },
    {
      "epoch": 0.3053790110060456,
      "grad_norm": 0.39217743277549744,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 3940
    },
    {
      "epoch": 0.3061540846380406,
      "grad_norm": 0.4051911234855652,
      "learning_rate": 2e-05,
      "loss": 0.6575,
      "step": 3950
    },
    {
      "epoch": 0.30692915827003564,
      "grad_norm": 0.38253021240234375,
      "learning_rate": 2e-05,
      "loss": 0.621,
      "step": 3960
    },
    {
      "epoch": 0.30770423190203067,
      "grad_norm": 0.3943375051021576,
      "learning_rate": 2e-05,
      "loss": 0.6657,
      "step": 3970
    },
    {
      "epoch": 0.30847930553402575,
      "grad_norm": 0.38000527024269104,
      "learning_rate": 2e-05,
      "loss": 0.6324,
      "step": 3980
    },
    {
      "epoch": 0.3092543791660208,
      "grad_norm": 0.40028852224349976,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 3990
    },
    {
      "epoch": 0.3100294527980158,
      "grad_norm": 0.39642465114593506,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 4000
    },
    {
      "epoch": 0.31080452643001083,
      "grad_norm": 0.4065065383911133,
      "learning_rate": 2e-05,
      "loss": 0.6367,
      "step": 4010
    },
    {
      "epoch": 0.3115796000620059,
      "grad_norm": 0.4016914367675781,
      "learning_rate": 2e-05,
      "loss": 0.6426,
      "step": 4020
    },
    {
      "epoch": 0.31235467369400094,
      "grad_norm": 0.3893164396286011,
      "learning_rate": 2e-05,
      "loss": 0.6348,
      "step": 4030
    },
    {
      "epoch": 0.31312974732599597,
      "grad_norm": 0.3809949457645416,
      "learning_rate": 2e-05,
      "loss": 0.6472,
      "step": 4040
    },
    {
      "epoch": 0.313904820957991,
      "grad_norm": 0.3741115629673004,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 4050
    },
    {
      "epoch": 0.3146798945899861,
      "grad_norm": 0.4036007523536682,
      "learning_rate": 2e-05,
      "loss": 0.6506,
      "step": 4060
    },
    {
      "epoch": 0.3154549682219811,
      "grad_norm": 0.4299529790878296,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 4070
    },
    {
      "epoch": 0.31623004185397613,
      "grad_norm": 0.42603036761283875,
      "learning_rate": 2e-05,
      "loss": 0.6531,
      "step": 4080
    },
    {
      "epoch": 0.31700511548597116,
      "grad_norm": 0.3851945400238037,
      "learning_rate": 2e-05,
      "loss": 0.6213,
      "step": 4090
    },
    {
      "epoch": 0.3177801891179662,
      "grad_norm": 0.4368724524974823,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 4100
    },
    {
      "epoch": 0.31855526274996127,
      "grad_norm": 1.2080000638961792,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 4110
    },
    {
      "epoch": 0.3193303363819563,
      "grad_norm": 0.38193827867507935,
      "learning_rate": 2e-05,
      "loss": 0.6368,
      "step": 4120
    },
    {
      "epoch": 0.3201054100139513,
      "grad_norm": 0.42428210377693176,
      "learning_rate": 2e-05,
      "loss": 0.6374,
      "step": 4130
    },
    {
      "epoch": 0.32088048364594635,
      "grad_norm": 0.42469552159309387,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 4140
    },
    {
      "epoch": 0.32165555727794143,
      "grad_norm": 0.39727783203125,
      "learning_rate": 2e-05,
      "loss": 0.6362,
      "step": 4150
    },
    {
      "epoch": 0.32243063090993646,
      "grad_norm": 0.3932878375053406,
      "learning_rate": 2e-05,
      "loss": 0.6376,
      "step": 4160
    },
    {
      "epoch": 0.3232057045419315,
      "grad_norm": 0.38844063878059387,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 4170
    },
    {
      "epoch": 0.3239807781739265,
      "grad_norm": 0.4405646324157715,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 4180
    },
    {
      "epoch": 0.32475585180592154,
      "grad_norm": 0.3897880017757416,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 4190
    },
    {
      "epoch": 0.3255309254379166,
      "grad_norm": 0.3960212767124176,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 4200
    },
    {
      "epoch": 0.32630599906991165,
      "grad_norm": 0.4208945631980896,
      "learning_rate": 2e-05,
      "loss": 0.6492,
      "step": 4210
    },
    {
      "epoch": 0.32708107270190667,
      "grad_norm": 0.41353705525398254,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 4220
    },
    {
      "epoch": 0.3278561463339017,
      "grad_norm": 0.41692233085632324,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 4230
    },
    {
      "epoch": 0.3286312199658968,
      "grad_norm": 0.6839759349822998,
      "learning_rate": 2e-05,
      "loss": 0.6324,
      "step": 4240
    },
    {
      "epoch": 0.3294062935978918,
      "grad_norm": 0.40540701150894165,
      "learning_rate": 2e-05,
      "loss": 0.6239,
      "step": 4250
    },
    {
      "epoch": 0.33018136722988684,
      "grad_norm": 0.3870887756347656,
      "learning_rate": 2e-05,
      "loss": 0.6316,
      "step": 4260
    },
    {
      "epoch": 0.33095644086188186,
      "grad_norm": 0.6416301727294922,
      "learning_rate": 2e-05,
      "loss": 0.6442,
      "step": 4270
    },
    {
      "epoch": 0.33173151449387694,
      "grad_norm": 0.48203006386756897,
      "learning_rate": 2e-05,
      "loss": 0.639,
      "step": 4280
    },
    {
      "epoch": 0.33250658812587197,
      "grad_norm": 0.39743801951408386,
      "learning_rate": 2e-05,
      "loss": 0.6509,
      "step": 4290
    },
    {
      "epoch": 0.333281661757867,
      "grad_norm": 0.3943183422088623,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 4300
    },
    {
      "epoch": 0.334056735389862,
      "grad_norm": 0.44573351740837097,
      "learning_rate": 2e-05,
      "loss": 0.6365,
      "step": 4310
    },
    {
      "epoch": 0.33483180902185705,
      "grad_norm": 0.3921007812023163,
      "learning_rate": 2e-05,
      "loss": 0.6439,
      "step": 4320
    },
    {
      "epoch": 0.33560688265385213,
      "grad_norm": 0.3896481394767761,
      "learning_rate": 2e-05,
      "loss": 0.662,
      "step": 4330
    },
    {
      "epoch": 0.33638195628584716,
      "grad_norm": 0.39618930220603943,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 4340
    },
    {
      "epoch": 0.3371570299178422,
      "grad_norm": 0.3820051848888397,
      "learning_rate": 2e-05,
      "loss": 0.6293,
      "step": 4350
    },
    {
      "epoch": 0.3379321035498372,
      "grad_norm": 0.3976047933101654,
      "learning_rate": 2e-05,
      "loss": 0.6283,
      "step": 4360
    },
    {
      "epoch": 0.3387071771818323,
      "grad_norm": 0.3893871307373047,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 4370
    },
    {
      "epoch": 0.3394822508138273,
      "grad_norm": 0.39987680315971375,
      "learning_rate": 2e-05,
      "loss": 0.6473,
      "step": 4380
    },
    {
      "epoch": 0.34025732444582235,
      "grad_norm": 0.3805384635925293,
      "learning_rate": 2e-05,
      "loss": 0.6485,
      "step": 4390
    },
    {
      "epoch": 0.3410323980778174,
      "grad_norm": 0.38630494475364685,
      "learning_rate": 2e-05,
      "loss": 0.6291,
      "step": 4400
    },
    {
      "epoch": 0.3418074717098124,
      "grad_norm": 0.3626739978790283,
      "learning_rate": 2e-05,
      "loss": 0.6456,
      "step": 4410
    },
    {
      "epoch": 0.3425825453418075,
      "grad_norm": 0.44230303168296814,
      "learning_rate": 2e-05,
      "loss": 0.631,
      "step": 4420
    },
    {
      "epoch": 0.3433576189738025,
      "grad_norm": 0.38728994131088257,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 4430
    },
    {
      "epoch": 0.34413269260579754,
      "grad_norm": 0.3877067267894745,
      "learning_rate": 2e-05,
      "loss": 0.6304,
      "step": 4440
    },
    {
      "epoch": 0.34490776623779257,
      "grad_norm": 0.39325466752052307,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 4450
    },
    {
      "epoch": 0.34568283986978765,
      "grad_norm": 0.38017114996910095,
      "learning_rate": 2e-05,
      "loss": 0.649,
      "step": 4460
    },
    {
      "epoch": 0.3464579135017827,
      "grad_norm": 0.41399797797203064,
      "learning_rate": 2e-05,
      "loss": 0.6425,
      "step": 4470
    },
    {
      "epoch": 0.3472329871337777,
      "grad_norm": 0.3768306076526642,
      "learning_rate": 2e-05,
      "loss": 0.6473,
      "step": 4480
    },
    {
      "epoch": 0.34800806076577273,
      "grad_norm": 0.43770062923431396,
      "learning_rate": 2e-05,
      "loss": 0.6564,
      "step": 4490
    },
    {
      "epoch": 0.3487831343977678,
      "grad_norm": 0.3727886974811554,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 4500
    },
    {
      "epoch": 0.34955820802976284,
      "grad_norm": 0.40988942980766296,
      "learning_rate": 2e-05,
      "loss": 0.626,
      "step": 4510
    },
    {
      "epoch": 0.35033328166175787,
      "grad_norm": 0.45901086926460266,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 4520
    },
    {
      "epoch": 0.3511083552937529,
      "grad_norm": 0.4200774133205414,
      "learning_rate": 2e-05,
      "loss": 0.6368,
      "step": 4530
    },
    {
      "epoch": 0.3518834289257479,
      "grad_norm": 0.42148029804229736,
      "learning_rate": 2e-05,
      "loss": 0.6433,
      "step": 4540
    },
    {
      "epoch": 0.352658502557743,
      "grad_norm": 0.39140892028808594,
      "learning_rate": 2e-05,
      "loss": 0.6254,
      "step": 4550
    },
    {
      "epoch": 0.35343357618973803,
      "grad_norm": 0.41720694303512573,
      "learning_rate": 2e-05,
      "loss": 0.6399,
      "step": 4560
    },
    {
      "epoch": 0.35420864982173306,
      "grad_norm": 0.5104743242263794,
      "learning_rate": 2e-05,
      "loss": 0.6277,
      "step": 4570
    },
    {
      "epoch": 0.3549837234537281,
      "grad_norm": 0.422849178314209,
      "learning_rate": 2e-05,
      "loss": 0.6163,
      "step": 4580
    },
    {
      "epoch": 0.35575879708572317,
      "grad_norm": 0.38357770442962646,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 4590
    },
    {
      "epoch": 0.3565338707177182,
      "grad_norm": 0.3941023051738739,
      "learning_rate": 2e-05,
      "loss": 0.6307,
      "step": 4600
    },
    {
      "epoch": 0.3573089443497132,
      "grad_norm": 0.4773350954055786,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 4610
    },
    {
      "epoch": 0.35808401798170825,
      "grad_norm": 0.4011412560939789,
      "learning_rate": 2e-05,
      "loss": 0.6364,
      "step": 4620
    },
    {
      "epoch": 0.35885909161370333,
      "grad_norm": 0.3978007733821869,
      "learning_rate": 2e-05,
      "loss": 0.6334,
      "step": 4630
    },
    {
      "epoch": 0.35963416524569836,
      "grad_norm": 0.4080989956855774,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 4640
    },
    {
      "epoch": 0.3604092388776934,
      "grad_norm": 0.3811449110507965,
      "learning_rate": 2e-05,
      "loss": 0.6404,
      "step": 4650
    },
    {
      "epoch": 0.3611843125096884,
      "grad_norm": 0.41419267654418945,
      "learning_rate": 2e-05,
      "loss": 0.6292,
      "step": 4660
    },
    {
      "epoch": 0.36195938614168344,
      "grad_norm": 0.3951331377029419,
      "learning_rate": 2e-05,
      "loss": 0.6456,
      "step": 4670
    },
    {
      "epoch": 0.3627344597736785,
      "grad_norm": 0.38848650455474854,
      "learning_rate": 2e-05,
      "loss": 0.6473,
      "step": 4680
    },
    {
      "epoch": 0.36350953340567355,
      "grad_norm": 0.4406234622001648,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 4690
    },
    {
      "epoch": 0.3642846070376686,
      "grad_norm": 0.38930660486221313,
      "learning_rate": 2e-05,
      "loss": 0.6249,
      "step": 4700
    },
    {
      "epoch": 0.3650596806696636,
      "grad_norm": 0.522161066532135,
      "learning_rate": 2e-05,
      "loss": 0.6325,
      "step": 4710
    },
    {
      "epoch": 0.3658347543016587,
      "grad_norm": 0.4170997440814972,
      "learning_rate": 2e-05,
      "loss": 0.6517,
      "step": 4720
    },
    {
      "epoch": 0.3666098279336537,
      "grad_norm": 0.3969718813896179,
      "learning_rate": 2e-05,
      "loss": 0.6425,
      "step": 4730
    },
    {
      "epoch": 0.36738490156564874,
      "grad_norm": 0.4235055148601532,
      "learning_rate": 2e-05,
      "loss": 0.6391,
      "step": 4740
    },
    {
      "epoch": 0.36815997519764376,
      "grad_norm": 0.47189322113990784,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 4750
    },
    {
      "epoch": 0.3689350488296388,
      "grad_norm": 0.3948569595813751,
      "learning_rate": 2e-05,
      "loss": 0.6255,
      "step": 4760
    },
    {
      "epoch": 0.36971012246163387,
      "grad_norm": 0.4018995463848114,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 4770
    },
    {
      "epoch": 0.3704851960936289,
      "grad_norm": 0.4135935604572296,
      "learning_rate": 2e-05,
      "loss": 0.63,
      "step": 4780
    },
    {
      "epoch": 0.3712602697256239,
      "grad_norm": 0.44257670640945435,
      "learning_rate": 2e-05,
      "loss": 0.64,
      "step": 4790
    },
    {
      "epoch": 0.37203534335761895,
      "grad_norm": 0.40565165877342224,
      "learning_rate": 2e-05,
      "loss": 0.6294,
      "step": 4800
    },
    {
      "epoch": 0.37281041698961404,
      "grad_norm": 0.38942795991897583,
      "learning_rate": 2e-05,
      "loss": 0.6536,
      "step": 4810
    },
    {
      "epoch": 0.37358549062160906,
      "grad_norm": 0.42309367656707764,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 4820
    },
    {
      "epoch": 0.3743605642536041,
      "grad_norm": 0.44575944542884827,
      "learning_rate": 2e-05,
      "loss": 0.6513,
      "step": 4830
    },
    {
      "epoch": 0.3751356378855991,
      "grad_norm": 0.36773574352264404,
      "learning_rate": 2e-05,
      "loss": 0.6192,
      "step": 4840
    },
    {
      "epoch": 0.3759107115175942,
      "grad_norm": 0.4058752655982971,
      "learning_rate": 2e-05,
      "loss": 0.6378,
      "step": 4850
    },
    {
      "epoch": 0.3766857851495892,
      "grad_norm": 0.3874194324016571,
      "learning_rate": 2e-05,
      "loss": 0.6252,
      "step": 4860
    },
    {
      "epoch": 0.37746085878158425,
      "grad_norm": 0.3904745280742645,
      "learning_rate": 2e-05,
      "loss": 0.6341,
      "step": 4870
    },
    {
      "epoch": 0.3782359324135793,
      "grad_norm": 0.3832138478755951,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 4880
    },
    {
      "epoch": 0.3790110060455743,
      "grad_norm": 0.4503752887248993,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 4890
    },
    {
      "epoch": 0.3797860796775694,
      "grad_norm": 0.38909709453582764,
      "learning_rate": 2e-05,
      "loss": 0.635,
      "step": 4900
    },
    {
      "epoch": 0.3805611533095644,
      "grad_norm": 0.5325521230697632,
      "learning_rate": 2e-05,
      "loss": 0.6374,
      "step": 4910
    },
    {
      "epoch": 0.38133622694155944,
      "grad_norm": 0.39805978536605835,
      "learning_rate": 2e-05,
      "loss": 0.6487,
      "step": 4920
    },
    {
      "epoch": 0.38211130057355447,
      "grad_norm": 0.3982967734336853,
      "learning_rate": 2e-05,
      "loss": 0.6307,
      "step": 4930
    },
    {
      "epoch": 0.38288637420554955,
      "grad_norm": 0.37822240591049194,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 4940
    },
    {
      "epoch": 0.3836614478375446,
      "grad_norm": 0.37436559796333313,
      "learning_rate": 2e-05,
      "loss": 0.6263,
      "step": 4950
    },
    {
      "epoch": 0.3844365214695396,
      "grad_norm": 0.5863552093505859,
      "learning_rate": 2e-05,
      "loss": 0.6331,
      "step": 4960
    },
    {
      "epoch": 0.38521159510153463,
      "grad_norm": 0.4198673963546753,
      "learning_rate": 2e-05,
      "loss": 0.6341,
      "step": 4970
    },
    {
      "epoch": 0.38598666873352966,
      "grad_norm": 0.3894166052341461,
      "learning_rate": 2e-05,
      "loss": 0.6318,
      "step": 4980
    },
    {
      "epoch": 0.38676174236552474,
      "grad_norm": 0.37254706025123596,
      "learning_rate": 2e-05,
      "loss": 0.6347,
      "step": 4990
    },
    {
      "epoch": 0.38753681599751977,
      "grad_norm": 0.3797888159751892,
      "learning_rate": 2e-05,
      "loss": 0.6163,
      "step": 5000
    },
    {
      "epoch": 0.3883118896295148,
      "grad_norm": 0.3750082552433014,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 5010
    },
    {
      "epoch": 0.3890869632615098,
      "grad_norm": 0.4142601191997528,
      "learning_rate": 2e-05,
      "loss": 0.6403,
      "step": 5020
    },
    {
      "epoch": 0.3898620368935049,
      "grad_norm": 0.37522315979003906,
      "learning_rate": 2e-05,
      "loss": 0.6245,
      "step": 5030
    },
    {
      "epoch": 0.39063711052549993,
      "grad_norm": 0.3764517903327942,
      "learning_rate": 2e-05,
      "loss": 0.6287,
      "step": 5040
    },
    {
      "epoch": 0.39141218415749496,
      "grad_norm": 0.3717632591724396,
      "learning_rate": 2e-05,
      "loss": 0.6169,
      "step": 5050
    },
    {
      "epoch": 0.39218725778949,
      "grad_norm": 0.3888686001300812,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 5060
    },
    {
      "epoch": 0.39296233142148507,
      "grad_norm": 0.38612762093544006,
      "learning_rate": 2e-05,
      "loss": 0.6307,
      "step": 5070
    },
    {
      "epoch": 0.3937374050534801,
      "grad_norm": 0.39416858553886414,
      "learning_rate": 2e-05,
      "loss": 0.6188,
      "step": 5080
    },
    {
      "epoch": 0.3945124786854751,
      "grad_norm": 0.9586396217346191,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 5090
    },
    {
      "epoch": 0.39528755231747015,
      "grad_norm": 0.38882219791412354,
      "learning_rate": 2e-05,
      "loss": 0.6317,
      "step": 5100
    },
    {
      "epoch": 0.3960626259494652,
      "grad_norm": 0.386748731136322,
      "learning_rate": 2e-05,
      "loss": 0.6179,
      "step": 5110
    },
    {
      "epoch": 0.39683769958146026,
      "grad_norm": 0.6027523279190063,
      "learning_rate": 2e-05,
      "loss": 0.623,
      "step": 5120
    },
    {
      "epoch": 0.3976127732134553,
      "grad_norm": 0.36918896436691284,
      "learning_rate": 2e-05,
      "loss": 0.6436,
      "step": 5130
    },
    {
      "epoch": 0.3983878468454503,
      "grad_norm": 0.38792452216148376,
      "learning_rate": 2e-05,
      "loss": 0.6204,
      "step": 5140
    },
    {
      "epoch": 0.39916292047744534,
      "grad_norm": 0.48190125823020935,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 5150
    },
    {
      "epoch": 0.3999379941094404,
      "grad_norm": 0.4005718529224396,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 5160
    },
    {
      "epoch": 0.40071306774143545,
      "grad_norm": 0.4082946479320526,
      "learning_rate": 2e-05,
      "loss": 0.6268,
      "step": 5170
    },
    {
      "epoch": 0.4014881413734305,
      "grad_norm": 0.4058220088481903,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 5180
    },
    {
      "epoch": 0.4022632150054255,
      "grad_norm": 0.36767181754112244,
      "learning_rate": 2e-05,
      "loss": 0.6376,
      "step": 5190
    },
    {
      "epoch": 0.4030382886374205,
      "grad_norm": 0.4004252851009369,
      "learning_rate": 2e-05,
      "loss": 0.6579,
      "step": 5200
    },
    {
      "epoch": 0.4038133622694156,
      "grad_norm": 0.3943789005279541,
      "learning_rate": 2e-05,
      "loss": 0.6486,
      "step": 5210
    },
    {
      "epoch": 0.40458843590141064,
      "grad_norm": 0.6699329018592834,
      "learning_rate": 2e-05,
      "loss": 0.6265,
      "step": 5220
    },
    {
      "epoch": 0.40536350953340566,
      "grad_norm": 0.4446292519569397,
      "learning_rate": 2e-05,
      "loss": 0.6232,
      "step": 5230
    },
    {
      "epoch": 0.4061385831654007,
      "grad_norm": 0.373099148273468,
      "learning_rate": 2e-05,
      "loss": 0.6116,
      "step": 5240
    },
    {
      "epoch": 0.4069136567973958,
      "grad_norm": 0.453997403383255,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 5250
    },
    {
      "epoch": 0.4076887304293908,
      "grad_norm": 0.40322959423065186,
      "learning_rate": 2e-05,
      "loss": 0.6292,
      "step": 5260
    },
    {
      "epoch": 0.4084638040613858,
      "grad_norm": 0.3951454758644104,
      "learning_rate": 2e-05,
      "loss": 0.6211,
      "step": 5270
    },
    {
      "epoch": 0.40923887769338085,
      "grad_norm": 0.38401883840560913,
      "learning_rate": 2e-05,
      "loss": 0.6469,
      "step": 5280
    },
    {
      "epoch": 0.41001395132537594,
      "grad_norm": 0.4328628182411194,
      "learning_rate": 2e-05,
      "loss": 0.6204,
      "step": 5290
    },
    {
      "epoch": 0.41078902495737096,
      "grad_norm": 0.40740108489990234,
      "learning_rate": 2e-05,
      "loss": 0.6354,
      "step": 5300
    },
    {
      "epoch": 0.411564098589366,
      "grad_norm": 0.41443267464637756,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 5310
    },
    {
      "epoch": 0.412339172221361,
      "grad_norm": 0.36498749256134033,
      "learning_rate": 2e-05,
      "loss": 0.6337,
      "step": 5320
    },
    {
      "epoch": 0.41311424585335604,
      "grad_norm": 0.43135717511177063,
      "learning_rate": 2e-05,
      "loss": 0.6352,
      "step": 5330
    },
    {
      "epoch": 0.4138893194853511,
      "grad_norm": 0.7326414585113525,
      "learning_rate": 2e-05,
      "loss": 0.6414,
      "step": 5340
    },
    {
      "epoch": 0.41466439311734615,
      "grad_norm": 0.3837507665157318,
      "learning_rate": 1.999194520388485e-05,
      "loss": 0.6311,
      "step": 5350
    },
    {
      "epoch": 0.4154394667493412,
      "grad_norm": 0.39215368032455444,
      "learning_rate": 1.9959247303157763e-05,
      "loss": 0.6161,
      "step": 5360
    },
    {
      "epoch": 0.4162145403813362,
      "grad_norm": 0.46202924847602844,
      "learning_rate": 1.990149424459817e-05,
      "loss": 0.6389,
      "step": 5370
    },
    {
      "epoch": 0.4169896140133313,
      "grad_norm": 0.3741549551486969,
      "learning_rate": 1.9818847538648688e-05,
      "loss": 0.6292,
      "step": 5380
    },
    {
      "epoch": 0.4177646876453263,
      "grad_norm": 0.4218798875808716,
      "learning_rate": 1.971153831256721e-05,
      "loss": 0.6252,
      "step": 5390
    },
    {
      "epoch": 0.41853976127732134,
      "grad_norm": 0.3944324254989624,
      "learning_rate": 1.9579866664063435e-05,
      "loss": 0.6206,
      "step": 5400
    },
    {
      "epoch": 0.41931483490931637,
      "grad_norm": 0.43683046102523804,
      "learning_rate": 1.9424200822054782e-05,
      "loss": 0.6298,
      "step": 5410
    },
    {
      "epoch": 0.42008990854131145,
      "grad_norm": 0.3944668769836426,
      "learning_rate": 1.9244976116888626e-05,
      "loss": 0.6398,
      "step": 5420
    },
    {
      "epoch": 0.4208649821733065,
      "grad_norm": 0.4029880166053772,
      "learning_rate": 1.904269376291071e-05,
      "loss": 0.6539,
      "step": 5430
    },
    {
      "epoch": 0.4216400558053015,
      "grad_norm": 0.402089387178421,
      "learning_rate": 1.88179194567844e-05,
      "loss": 0.6266,
      "step": 5440
    },
    {
      "epoch": 0.42241512943729653,
      "grad_norm": 0.4242640733718872,
      "learning_rate": 1.8571281795480632e-05,
      "loss": 0.6418,
      "step": 5450
    },
    {
      "epoch": 0.42319020306929156,
      "grad_norm": 0.3908100128173828,
      "learning_rate": 1.8303470518362808e-05,
      "loss": 0.6373,
      "step": 5460
    },
    {
      "epoch": 0.42396527670128664,
      "grad_norm": 0.44044214487075806,
      "learning_rate": 1.801523457828271e-05,
      "loss": 0.6283,
      "step": 5470
    },
    {
      "epoch": 0.42474035033328167,
      "grad_norm": 0.40724480152130127,
      "learning_rate": 1.7707380047081796e-05,
      "loss": 0.6333,
      "step": 5480
    },
    {
      "epoch": 0.4255154239652767,
      "grad_norm": 0.397793710231781,
      "learning_rate": 1.7380767861355252e-05,
      "loss": 0.6167,
      "step": 5490
    },
    {
      "epoch": 0.4262904975972717,
      "grad_norm": 0.3732137382030487,
      "learning_rate": 1.703631141478299e-05,
      "loss": 0.6236,
      "step": 5500
    },
    {
      "epoch": 0.4270655712292668,
      "grad_norm": 0.37707749009132385,
      "learning_rate": 1.6674974003760705e-05,
      "loss": 0.6212,
      "step": 5510
    },
    {
      "epoch": 0.42784064486126183,
      "grad_norm": 0.36281004548072815,
      "learning_rate": 1.6297766133474556e-05,
      "loss": 0.6292,
      "step": 5520
    },
    {
      "epoch": 0.42861571849325686,
      "grad_norm": 0.38236552476882935,
      "learning_rate": 1.5905742691953186e-05,
      "loss": 0.5982,
      "step": 5530
    },
    {
      "epoch": 0.4293907921252519,
      "grad_norm": 0.4031112790107727,
      "learning_rate": 1.55e-05,
      "loss": 0.6329,
      "step": 5540
    },
    {
      "epoch": 0.4301658657572469,
      "grad_norm": 0.45905470848083496,
      "learning_rate": 1.5081672745255862e-05,
      "loss": 0.6362,
      "step": 5550
    },
    {
      "epoch": 0.430940939389242,
      "grad_norm": 0.5283321142196655,
      "learning_rate": 1.4651930808966206e-05,
      "loss": 0.6264,
      "step": 5560
    },
    {
      "epoch": 0.431716013021237,
      "grad_norm": 0.4312293529510498,
      "learning_rate": 1.4211975994326848e-05,
      "loss": 0.6196,
      "step": 5570
    },
    {
      "epoch": 0.43249108665323205,
      "grad_norm": 0.38796475529670715,
      "learning_rate": 1.376303866555779e-05,
      "loss": 0.6332,
      "step": 5580
    },
    {
      "epoch": 0.4332661602852271,
      "grad_norm": 0.3728007972240448,
      "learning_rate": 1.3306374307104164e-05,
      "loss": 0.6276,
      "step": 5590
    },
    {
      "epoch": 0.43404123391722216,
      "grad_norm": 0.38708752393722534,
      "learning_rate": 1.2843260012586718e-05,
      "loss": 0.6207,
      "step": 5600
    },
    {
      "epoch": 0.4348163075492172,
      "grad_norm": 0.37163254618644714,
      "learning_rate": 1.2374990913320726e-05,
      "loss": 0.6379,
      "step": 5610
    },
    {
      "epoch": 0.4355913811812122,
      "grad_norm": 0.3740164637565613,
      "learning_rate": 1.1902876556391247e-05,
      "loss": 0.6162,
      "step": 5620
    },
    {
      "epoch": 0.43636645481320724,
      "grad_norm": 0.3969056308269501,
      "learning_rate": 1.1428237242413685e-05,
      "loss": 0.6249,
      "step": 5630
    },
    {
      "epoch": 0.4371415284452023,
      "grad_norm": 0.392935186624527,
      "learning_rate": 1.095240033322132e-05,
      "loss": 0.6355,
      "step": 5640
    },
    {
      "epoch": 0.43791660207719735,
      "grad_norm": 0.39747709035873413,
      "learning_rate": 1.047669653980572e-05,
      "loss": 0.6355,
      "step": 5650
    },
    {
      "epoch": 0.4386916757091924,
      "grad_norm": 0.5514155030250549,
      "learning_rate": 1.0002456200890902e-05,
      "loss": 0.599,
      "step": 5660
    },
    {
      "epoch": 0.4394667493411874,
      "grad_norm": 0.4582012891769409,
      "learning_rate": 9.531005562548668e-06,
      "loss": 0.6125,
      "step": 5670
    },
    {
      "epoch": 0.44024182297318243,
      "grad_norm": 0.3846968114376068,
      "learning_rate": 9.06366306925925e-06,
      "loss": 0.6362,
      "step": 5680
    },
    {
      "epoch": 0.4410168966051775,
      "grad_norm": 0.38858526945114136,
      "learning_rate": 8.601735676789686e-06,
      "loss": 0.6182,
      "step": 5690
    },
    {
      "epoch": 0.44179197023717254,
      "grad_norm": 0.4033548831939697,
      "learning_rate": 8.14651519720118e-06,
      "loss": 0.6116,
      "step": 5700
    },
    {
      "epoch": 0.44256704386916756,
      "grad_norm": 0.3851390779018402,
      "learning_rate": 7.699274686206866e-06,
      "loss": 0.6039,
      "step": 5710
    },
    {
      "epoch": 0.4433421175011626,
      "grad_norm": 0.366301953792572,
      "learning_rate": 7.261264882983024e-06,
      "loss": 0.6339,
      "step": 5720
    },
    {
      "epoch": 0.4441171911331577,
      "grad_norm": 0.3673529326915741,
      "learning_rate": 6.8337107123900386e-06,
      "loss": 0.6329,
      "step": 5730
    },
    {
      "epoch": 0.4448922647651527,
      "grad_norm": 0.4046303927898407,
      "learning_rate": 6.417807859384881e-06,
      "loss": 0.6259,
      "step": 5740
    },
    {
      "epoch": 0.4456673383971477,
      "grad_norm": 0.4655354619026184,
      "learning_rate": 6.014719425205009e-06,
      "loss": 0.6203,
      "step": 5750
    },
    {
      "epoch": 0.44644241202914275,
      "grad_norm": 0.3795314133167267,
      "learning_rate": 5.6255726746749275e-06,
      "loss": 0.6186,
      "step": 5760
    },
    {
      "epoch": 0.4472174856611378,
      "grad_norm": 0.400132954120636,
      "learning_rate": 5.251455883731728e-06,
      "loss": 0.6053,
      "step": 5770
    },
    {
      "epoch": 0.44799255929313286,
      "grad_norm": 0.39265620708465576,
      "learning_rate": 4.893415295985812e-06,
      "loss": 0.6358,
      "step": 5780
    },
    {
      "epoch": 0.4487676329251279,
      "grad_norm": 0.38653743267059326,
      "learning_rate": 4.552452196827885e-06,
      "loss": 0.6254,
      "step": 5790
    },
    {
      "epoch": 0.4495427065571229,
      "grad_norm": 0.42161431908607483,
      "learning_rate": 4.229520113264785e-06,
      "loss": 0.6198,
      "step": 5800
    },
    {
      "epoch": 0.45031778018911794,
      "grad_norm": 0.4898611903190613,
      "learning_rate": 3.925522147314915e-06,
      "loss": 0.6327,
      "step": 5810
    },
    {
      "epoch": 0.451092853821113,
      "grad_norm": 0.38622191548347473,
      "learning_rate": 3.641308450420675e-06,
      "loss": 0.6214,
      "step": 5820
    },
    {
      "epoch": 0.45186792745310805,
      "grad_norm": 0.42542681097984314,
      "learning_rate": 3.37767384594087e-06,
      "loss": 0.6085,
      "step": 5830
    },
    {
      "epoch": 0.4526430010851031,
      "grad_norm": 0.3988613188266754,
      "learning_rate": 3.1353556063719357e-06,
      "loss": 0.6289,
      "step": 5840
    },
    {
      "epoch": 0.4534180747170981,
      "grad_norm": 0.4273814260959625,
      "learning_rate": 2.9150313915141474e-06,
      "loss": 0.6189,
      "step": 5850
    },
    {
      "epoch": 0.4541931483490932,
      "grad_norm": 0.3907577693462372,
      "learning_rate": 2.717317353348891e-06,
      "loss": 0.6178,
      "step": 5860
    },
    {
      "epoch": 0.4549682219810882,
      "grad_norm": 0.37921684980392456,
      "learning_rate": 2.5427664129268253e-06,
      "loss": 0.6268,
      "step": 5870
    },
    {
      "epoch": 0.45574329561308324,
      "grad_norm": 0.3783731460571289,
      "learning_rate": 2.391866714085717e-06,
      "loss": 0.6291,
      "step": 5880
    },
    {
      "epoch": 0.45651836924507827,
      "grad_norm": 0.42365536093711853,
      "learning_rate": 2.2650402583222753e-06,
      "loss": 0.6278,
      "step": 5890
    },
    {
      "epoch": 0.4572934428770733,
      "grad_norm": 0.3801100254058838,
      "learning_rate": 2.1626417246356398e-06,
      "loss": 0.628,
      "step": 5900
    },
    {
      "epoch": 0.4580685165090684,
      "grad_norm": 0.38439080119132996,
      "learning_rate": 2.084957477642894e-06,
      "loss": 0.6203,
      "step": 5910
    },
    {
      "epoch": 0.4588435901410634,
      "grad_norm": 0.3800904452800751,
      "learning_rate": 2.0322047667405147e-06,
      "loss": 0.614,
      "step": 5920
    },
    {
      "epoch": 0.45961866377305843,
      "grad_norm": 0.5425535440444946,
      "learning_rate": 2.0045311185513342e-06,
      "loss": 0.6142,
      "step": 5930
    }
  ],
  "logging_steps": 10,
  "max_steps": 5935,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2035,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6494000762504348e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}