{
  "best_metric": 1.6930413246154785,
  "best_model_checkpoint": "./FT_models/[LDH]0221_all_llama31_docs/checkpoint-500",
  "epoch": 68.63366336633663,
  "eval_steps": 500,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19801980198019803,
      "grad_norm": 0.4907771050930023,
      "learning_rate": 0.0001999995055317446,
      "loss": 1.9296,
      "step": 10
    },
    {
      "epoch": 0.39603960396039606,
      "grad_norm": 0.4170154929161072,
      "learning_rate": 0.0001999955498150411,
      "loss": 1.5182,
      "step": 20
    },
    {
      "epoch": 0.594059405940594,
      "grad_norm": 0.6515398025512695,
      "learning_rate": 0.00019998763853811184,
      "loss": 1.3992,
      "step": 30
    },
    {
      "epoch": 0.7920792079207921,
      "grad_norm": 0.6787588596343994,
      "learning_rate": 0.00019997577201390606,
      "loss": 1.3514,
      "step": 40
    },
    {
      "epoch": 0.9900990099009901,
      "grad_norm": 0.6807988286018372,
      "learning_rate": 0.0001999599507118322,
      "loss": 1.1463,
      "step": 50
    },
    {
      "epoch": 1.1782178217821782,
      "grad_norm": 0.7439409494400024,
      "learning_rate": 0.00019994017525773913,
      "loss": 0.9557,
      "step": 60
    },
    {
      "epoch": 1.3762376237623761,
      "grad_norm": 0.8682450652122498,
      "learning_rate": 0.0001999164464338918,
      "loss": 0.8612,
      "step": 70
    },
    {
      "epoch": 1.5742574257425743,
      "grad_norm": 0.9405132532119751,
      "learning_rate": 0.0001998887651789398,
      "loss": 0.713,
      "step": 80
    },
    {
      "epoch": 1.7722772277227723,
      "grad_norm": 1.0701061487197876,
      "learning_rate": 0.0001998571325878806,
      "loss": 0.6861,
      "step": 90
    },
    {
      "epoch": 1.9702970297029703,
      "grad_norm": 0.8982220888137817,
      "learning_rate": 0.00019982154991201608,
      "loss": 0.5632,
      "step": 100
    },
    {
      "epoch": 2.1584158415841586,
      "grad_norm": 0.8719994425773621,
      "learning_rate": 0.00019978201855890308,
      "loss": 0.4271,
      "step": 110
    },
    {
      "epoch": 2.3564356435643563,
      "grad_norm": 0.9812946319580078,
      "learning_rate": 0.00019973854009229763,
      "loss": 0.316,
      "step": 120
    },
    {
      "epoch": 2.5544554455445545,
      "grad_norm": 0.9043093323707581,
      "learning_rate": 0.00019969111623209323,
      "loss": 0.2683,
      "step": 130
    },
    {
      "epoch": 2.7524752475247523,
      "grad_norm": 0.7294723987579346,
      "learning_rate": 0.00019963974885425266,
      "loss": 0.2439,
      "step": 140
    },
    {
      "epoch": 2.9504950495049505,
      "grad_norm": 0.8509527444839478,
      "learning_rate": 0.00019958443999073397,
      "loss": 0.2409,
      "step": 150
    },
    {
      "epoch": 3.1386138613861387,
      "grad_norm": 0.6639506816864014,
      "learning_rate": 0.00019952519182940993,
      "loss": 0.1796,
      "step": 160
    },
    {
      "epoch": 3.3366336633663365,
      "grad_norm": 0.64533531665802,
      "learning_rate": 0.0001994620067139815,
      "loss": 0.1456,
      "step": 170
    },
    {
      "epoch": 3.5346534653465347,
      "grad_norm": 0.7491998672485352,
      "learning_rate": 0.00019939488714388524,
      "loss": 0.114,
      "step": 180
    },
    {
      "epoch": 3.7326732673267324,
      "grad_norm": 0.49455082416534424,
      "learning_rate": 0.00019932383577419432,
      "loss": 0.1028,
      "step": 190
    },
    {
      "epoch": 3.9306930693069306,
      "grad_norm": 0.5427107214927673,
      "learning_rate": 0.0001992488554155135,
      "loss": 0.1218,
      "step": 200
    },
    {
      "epoch": 4.118811881188119,
      "grad_norm": 0.3893876373767853,
      "learning_rate": 0.0001991699490338681,
      "loss": 0.0833,
      "step": 210
    },
    {
      "epoch": 4.316831683168317,
      "grad_norm": 0.4123828113079071,
      "learning_rate": 0.00019908711975058637,
      "loss": 0.0684,
      "step": 220
    },
    {
      "epoch": 4.514851485148515,
      "grad_norm": 0.5183358788490295,
      "learning_rate": 0.00019900037084217637,
      "loss": 0.0741,
      "step": 230
    },
    {
      "epoch": 4.712871287128713,
      "grad_norm": 0.3828432261943817,
      "learning_rate": 0.00019890970574019617,
      "loss": 0.0607,
      "step": 240
    },
    {
      "epoch": 4.910891089108911,
      "grad_norm": 0.30472326278686523,
      "learning_rate": 0.00019881512803111796,
      "loss": 0.0607,
      "step": 250
    },
    {
      "epoch": 5.099009900990099,
      "grad_norm": 0.3620370328426361,
      "learning_rate": 0.00019871664145618657,
      "loss": 0.0545,
      "step": 260
    },
    {
      "epoch": 5.297029702970297,
      "grad_norm": 0.3857772946357727,
      "learning_rate": 0.00019861424991127115,
      "loss": 0.0394,
      "step": 270
    },
    {
      "epoch": 5.4950495049504955,
      "grad_norm": 0.22827056050300598,
      "learning_rate": 0.00019850795744671116,
      "loss": 0.0404,
      "step": 280
    },
    {
      "epoch": 5.693069306930693,
      "grad_norm": 0.38091033697128296,
      "learning_rate": 0.00019839776826715614,
      "loss": 0.0406,
      "step": 290
    },
    {
      "epoch": 5.891089108910891,
      "grad_norm": 0.4117518663406372,
      "learning_rate": 0.00019828368673139947,
      "loss": 0.0414,
      "step": 300
    },
    {
      "epoch": 6.079207920792079,
      "grad_norm": 0.32705605030059814,
      "learning_rate": 0.00019816571735220583,
      "loss": 0.038,
      "step": 310
    },
    {
      "epoch": 6.2772277227722775,
      "grad_norm": 0.3450210988521576,
      "learning_rate": 0.0001980438647961327,
      "loss": 0.0305,
      "step": 320
    },
    {
      "epoch": 6.475247524752476,
      "grad_norm": 0.29956933856010437,
      "learning_rate": 0.00019791813388334581,
      "loss": 0.0301,
      "step": 330
    },
    {
      "epoch": 6.673267326732673,
      "grad_norm": 0.39558252692222595,
      "learning_rate": 0.00019778852958742853,
      "loss": 0.0312,
      "step": 340
    },
    {
      "epoch": 6.871287128712871,
      "grad_norm": 0.29619818925857544,
      "learning_rate": 0.00019765505703518496,
      "loss": 0.0364,
      "step": 350
    },
    {
      "epoch": 7.0594059405940595,
      "grad_norm": 0.2381938099861145,
      "learning_rate": 0.00019751772150643722,
      "loss": 0.0307,
      "step": 360
    },
    {
      "epoch": 7.257425742574258,
      "grad_norm": 0.38348841667175293,
      "learning_rate": 0.0001973765284338167,
      "loss": 0.0269,
      "step": 370
    },
    {
      "epoch": 7.455445544554456,
      "grad_norm": 0.25046518445014954,
      "learning_rate": 0.00019723148340254892,
      "loss": 0.0233,
      "step": 380
    },
    {
      "epoch": 7.653465346534653,
      "grad_norm": 0.27623361349105835,
      "learning_rate": 0.0001970825921502328,
      "loss": 0.0285,
      "step": 390
    },
    {
      "epoch": 7.851485148514851,
      "grad_norm": 0.2806844413280487,
      "learning_rate": 0.00019692986056661356,
      "loss": 0.0284,
      "step": 400
    },
    {
      "epoch": 8.03960396039604,
      "grad_norm": 0.31793248653411865,
      "learning_rate": 0.0001967732946933499,
      "loss": 0.0298,
      "step": 410
    },
    {
      "epoch": 8.237623762376238,
      "grad_norm": 0.26446977257728577,
      "learning_rate": 0.00019661290072377482,
      "loss": 0.0242,
      "step": 420
    },
    {
      "epoch": 8.435643564356436,
      "grad_norm": 0.21607118844985962,
      "learning_rate": 0.0001964486850026507,
      "loss": 0.0212,
      "step": 430
    },
    {
      "epoch": 8.633663366336634,
      "grad_norm": 0.3113400340080261,
      "learning_rate": 0.00019628065402591845,
      "loss": 0.0223,
      "step": 440
    },
    {
      "epoch": 8.831683168316832,
      "grad_norm": 0.21292166411876678,
      "learning_rate": 0.0001961088144404403,
      "loss": 0.0251,
      "step": 450
    },
    {
      "epoch": 9.01980198019802,
      "grad_norm": 0.2799708843231201,
      "learning_rate": 0.00019593317304373705,
      "loss": 0.0222,
      "step": 460
    },
    {
      "epoch": 9.217821782178218,
      "grad_norm": 0.3010728061199188,
      "learning_rate": 0.00019575373678371909,
      "loss": 0.0192,
      "step": 470
    },
    {
      "epoch": 9.415841584158416,
      "grad_norm": 0.26538291573524475,
      "learning_rate": 0.0001955705127584117,
      "loss": 0.0208,
      "step": 480
    },
    {
      "epoch": 9.613861386138614,
      "grad_norm": 0.29595261812210083,
      "learning_rate": 0.00019538350821567404,
      "loss": 0.0211,
      "step": 490
    },
    {
      "epoch": 9.811881188118813,
      "grad_norm": 0.28233084082603455,
      "learning_rate": 0.00019519273055291266,
      "loss": 0.0248,
      "step": 500
    },
    {
      "epoch": 9.811881188118813,
      "eval_loss": 1.6930413246154785,
      "eval_runtime": 78.4865,
      "eval_samples_per_second": 4.039,
      "eval_steps_per_second": 2.026,
      "step": 500
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.392619788646698,
      "learning_rate": 0.00019499818731678873,
      "loss": 0.0263,
      "step": 510
    },
    {
      "epoch": 10.198019801980198,
      "grad_norm": 0.18203061819076538,
      "learning_rate": 0.00019479988620291956,
      "loss": 0.0197,
      "step": 520
    },
    {
      "epoch": 10.396039603960396,
      "grad_norm": 0.4150699973106384,
      "learning_rate": 0.00019459783505557424,
      "loss": 0.0244,
      "step": 530
    },
    {
      "epoch": 10.594059405940595,
      "grad_norm": 0.1509980857372284,
      "learning_rate": 0.0001943920418673633,
      "loss": 0.021,
      "step": 540
    },
    {
      "epoch": 10.792079207920793,
      "grad_norm": 0.4003594219684601,
      "learning_rate": 0.0001941825147789225,
      "loss": 0.0224,
      "step": 550
    },
    {
      "epoch": 10.990099009900991,
      "grad_norm": 0.18431659042835236,
      "learning_rate": 0.00019396926207859084,
      "loss": 0.0221,
      "step": 560
    },
    {
      "epoch": 11.178217821782178,
      "grad_norm": 0.15787458419799805,
      "learning_rate": 0.00019375229220208276,
      "loss": 0.0193,
      "step": 570
    },
    {
      "epoch": 11.376237623762377,
      "grad_norm": 0.3416723608970642,
      "learning_rate": 0.0001935316137321543,
      "loss": 0.0209,
      "step": 580
    },
    {
      "epoch": 11.574257425742575,
      "grad_norm": 0.486870139837265,
      "learning_rate": 0.00019330723539826375,
      "loss": 0.021,
      "step": 590
    },
    {
      "epoch": 11.772277227722773,
      "grad_norm": 0.25933241844177246,
      "learning_rate": 0.0001930791660762262,
      "loss": 0.0198,
      "step": 600
    },
    {
      "epoch": 11.97029702970297,
      "grad_norm": 0.10881418734788895,
      "learning_rate": 0.0001928474147878626,
      "loss": 0.02,
      "step": 610
    },
    {
      "epoch": 12.158415841584159,
      "grad_norm": 0.19103562831878662,
      "learning_rate": 0.0001926119907006426,
      "loss": 0.0178,
      "step": 620
    },
    {
      "epoch": 12.356435643564357,
      "grad_norm": 0.1987970769405365,
      "learning_rate": 0.00019237290312732226,
      "loss": 0.0188,
      "step": 630
    },
    {
      "epoch": 12.554455445544555,
      "grad_norm": 0.45853590965270996,
      "learning_rate": 0.0001921301615255754,
      "loss": 0.0182,
      "step": 640
    },
    {
      "epoch": 12.752475247524753,
      "grad_norm": 0.22175024449825287,
      "learning_rate": 0.00019188377549761963,
      "loss": 0.02,
      "step": 650
    },
    {
      "epoch": 12.950495049504951,
      "grad_norm": 0.45677581429481506,
      "learning_rate": 0.00019163375478983632,
      "loss": 0.0231,
      "step": 660
    },
    {
      "epoch": 13.138613861386139,
      "grad_norm": 0.3287411630153656,
      "learning_rate": 0.00019138010929238534,
      "loss": 0.0215,
      "step": 670
    },
    {
      "epoch": 13.336633663366337,
      "grad_norm": 0.43493467569351196,
      "learning_rate": 0.0001911228490388136,
      "loss": 0.021,
      "step": 680
    },
    {
      "epoch": 13.534653465346535,
      "grad_norm": 0.15248054265975952,
      "learning_rate": 0.00019086198420565823,
      "loss": 0.0201,
      "step": 690
    },
    {
      "epoch": 13.732673267326733,
      "grad_norm": 0.17455808818340302,
      "learning_rate": 0.000190597525112044,
      "loss": 0.0225,
      "step": 700
    },
    {
      "epoch": 13.930693069306932,
      "grad_norm": 0.2868499457836151,
      "learning_rate": 0.00019032948221927524,
      "loss": 0.0195,
      "step": 710
    },
    {
      "epoch": 14.118811881188119,
      "grad_norm": 0.20362168550491333,
      "learning_rate": 0.00019005786613042185,
      "loss": 0.0211,
      "step": 720
    },
    {
      "epoch": 14.316831683168317,
      "grad_norm": 0.17784035205841064,
      "learning_rate": 0.00018978268758989991,
      "loss": 0.0217,
      "step": 730
    },
    {
      "epoch": 14.514851485148515,
      "grad_norm": 0.1722538322210312,
      "learning_rate": 0.00018950395748304678,
      "loss": 0.0178,
      "step": 740
    },
    {
      "epoch": 14.712871287128714,
      "grad_norm": 0.3050313889980316,
      "learning_rate": 0.0001892216868356904,
      "loss": 0.0252,
      "step": 750
    },
    {
      "epoch": 14.910891089108912,
      "grad_norm": 0.20032328367233276,
      "learning_rate": 0.00018893588681371303,
      "loss": 0.0191,
      "step": 760
    },
    {
      "epoch": 15.099009900990099,
      "grad_norm": 0.2856597304344177,
      "learning_rate": 0.00018864656872260985,
      "loss": 0.0196,
      "step": 770
    },
    {
      "epoch": 15.297029702970297,
      "grad_norm": 0.1927269697189331,
      "learning_rate": 0.00018835374400704154,
      "loss": 0.0177,
      "step": 780
    },
    {
      "epoch": 15.495049504950495,
      "grad_norm": 0.10624806582927704,
      "learning_rate": 0.00018805742425038145,
      "loss": 0.0187,
      "step": 790
    },
    {
      "epoch": 15.693069306930694,
      "grad_norm": 0.17643067240715027,
      "learning_rate": 0.00018775762117425777,
      "loss": 0.018,
      "step": 800
    },
    {
      "epoch": 15.891089108910892,
      "grad_norm": 0.25009414553642273,
      "learning_rate": 0.00018745434663808942,
      "loss": 0.0196,
      "step": 810
    },
    {
      "epoch": 16.07920792079208,
      "grad_norm": 0.13281401991844177,
      "learning_rate": 0.00018714761263861728,
      "loss": 0.0153,
      "step": 820
    },
    {
      "epoch": 16.277227722772277,
      "grad_norm": 0.20706744492053986,
      "learning_rate": 0.00018683743130942928,
      "loss": 0.0153,
      "step": 830
    },
    {
      "epoch": 16.475247524752476,
      "grad_norm": 0.26562607288360596,
      "learning_rate": 0.00018652381492048083,
      "loss": 0.0161,
      "step": 840
    },
    {
      "epoch": 16.673267326732674,
      "grad_norm": 0.24612267315387726,
      "learning_rate": 0.00018620677587760916,
      "loss": 0.017,
      "step": 850
    },
    {
      "epoch": 16.871287128712872,
      "grad_norm": 0.12470393627882004,
      "learning_rate": 0.00018588632672204264,
      "loss": 0.0163,
      "step": 860
    },
    {
      "epoch": 17.059405940594058,
      "grad_norm": 0.16730207204818726,
      "learning_rate": 0.00018556248012990468,
      "loss": 0.0164,
      "step": 870
    },
    {
      "epoch": 17.257425742574256,
      "grad_norm": 0.22669559717178345,
      "learning_rate": 0.0001852352489117124,
      "loss": 0.0185,
      "step": 880
    },
    {
      "epoch": 17.455445544554454,
      "grad_norm": 0.18988390266895294,
      "learning_rate": 0.0001849046460118698,
      "loss": 0.0168,
      "step": 890
    },
    {
      "epoch": 17.653465346534652,
      "grad_norm": 0.1657964289188385,
      "learning_rate": 0.00018457068450815562,
      "loss": 0.0169,
      "step": 900
    },
    {
      "epoch": 17.85148514851485,
      "grad_norm": 0.12119423598051071,
      "learning_rate": 0.00018423337761120618,
      "loss": 0.0173,
      "step": 910
    },
    {
      "epoch": 18.03960396039604,
      "grad_norm": 0.10356303304433823,
      "learning_rate": 0.00018389273866399275,
      "loss": 0.0159,
      "step": 920
    },
    {
      "epoch": 18.237623762376238,
      "grad_norm": 0.09106975793838501,
      "learning_rate": 0.00018354878114129367,
      "loss": 0.0145,
      "step": 930
    },
    {
      "epoch": 18.435643564356436,
      "grad_norm": 0.11265785992145538,
      "learning_rate": 0.00018320151864916135,
      "loss": 0.0142,
      "step": 940
    },
    {
      "epoch": 18.633663366336634,
      "grad_norm": 0.1934558004140854,
      "learning_rate": 0.00018285096492438424,
      "loss": 0.0146,
      "step": 950
    },
    {
      "epoch": 18.831683168316832,
      "grad_norm": 0.09819146245718002,
      "learning_rate": 0.00018249713383394303,
      "loss": 0.0163,
      "step": 960
    },
    {
      "epoch": 19.019801980198018,
      "grad_norm": 0.17335668206214905,
      "learning_rate": 0.00018214003937446253,
      "loss": 0.0165,
      "step": 970
    },
    {
      "epoch": 19.217821782178216,
      "grad_norm": 0.10033983737230301,
      "learning_rate": 0.0001817796956716578,
      "loss": 0.0141,
      "step": 980
    },
    {
      "epoch": 19.415841584158414,
      "grad_norm": 0.10356291383504868,
      "learning_rate": 0.00018141611697977529,
      "loss": 0.0138,
      "step": 990
    },
    {
      "epoch": 19.613861386138613,
      "grad_norm": 0.16795016825199127,
      "learning_rate": 0.0001810493176810292,
      "loss": 0.0148,
      "step": 1000
    },
    {
      "epoch": 19.613861386138613,
      "eval_loss": 1.8084672689437866,
      "eval_runtime": 82.2773,
      "eval_samples_per_second": 3.853,
      "eval_steps_per_second": 1.932,
      "step": 1000
    },
    {
      "epoch": 19.81188118811881,
      "grad_norm": 0.23259301483631134,
      "learning_rate": 0.00018067931228503246,
      "loss": 0.016,
      "step": 1010
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.07684651017189026,
      "learning_rate": 0.00018030611542822257,
      "loss": 0.0148,
      "step": 1020
    },
    {
      "epoch": 20.198019801980198,
      "grad_norm": 0.16659528017044067,
      "learning_rate": 0.00017992974187328305,
      "loss": 0.0127,
      "step": 1030
    },
    {
      "epoch": 20.396039603960396,
      "grad_norm": 0.25504982471466064,
      "learning_rate": 0.000179550206508559,
      "loss": 0.0125,
      "step": 1040
    },
    {
      "epoch": 20.594059405940595,
      "grad_norm": 0.13706262409687042,
      "learning_rate": 0.00017916752434746856,
      "loss": 0.0139,
      "step": 1050
    },
    {
      "epoch": 20.792079207920793,
      "grad_norm": 0.09546064585447311,
      "learning_rate": 0.00017878171052790868,
      "loss": 0.0143,
      "step": 1060
    },
    {
      "epoch": 20.99009900990099,
      "grad_norm": 0.19178028404712677,
      "learning_rate": 0.00017839278031165658,
      "loss": 0.0146,
      "step": 1070
    },
    {
      "epoch": 21.178217821782177,
      "grad_norm": 0.09425017982721329,
      "learning_rate": 0.00017800074908376584,
      "loss": 0.0137,
      "step": 1080
    },
    {
      "epoch": 21.376237623762375,
      "grad_norm": 0.0950135812163353,
      "learning_rate": 0.0001776056323519579,
      "loss": 0.0142,
      "step": 1090
    },
    {
      "epoch": 21.574257425742573,
      "grad_norm": 0.11520115286111832,
      "learning_rate": 0.00017720744574600863,
      "loss": 0.0147,
      "step": 1100
    },
    {
      "epoch": 21.77227722772277,
      "grad_norm": 0.07679902017116547,
      "learning_rate": 0.00017680620501712996,
      "loss": 0.013,
      "step": 1110
    },
    {
      "epoch": 21.97029702970297,
      "grad_norm": 0.0910814180970192,
      "learning_rate": 0.00017640192603734692,
      "loss": 0.0139,
      "step": 1120
    },
    {
      "epoch": 22.15841584158416,
      "grad_norm": 0.18983718752861023,
      "learning_rate": 0.00017599462479886974,
      "loss": 0.0119,
      "step": 1130
    },
    {
      "epoch": 22.356435643564357,
      "grad_norm": 0.05227164551615715,
      "learning_rate": 0.00017558431741346122,
      "loss": 0.0121,
      "step": 1140
    },
    {
      "epoch": 22.554455445544555,
      "grad_norm": 0.10026168078184128,
      "learning_rate": 0.00017517102011179933,
      "loss": 0.0128,
      "step": 1150
    },
    {
      "epoch": 22.752475247524753,
      "grad_norm": 0.10607390105724335,
      "learning_rate": 0.00017475474924283536,
      "loss": 0.0126,
      "step": 1160
    },
    {
      "epoch": 22.95049504950495,
      "grad_norm": 0.16404001414775848,
      "learning_rate": 0.000174335521273147,
      "loss": 0.013,
      "step": 1170
    },
    {
      "epoch": 23.138613861386137,
      "grad_norm": 0.09486360847949982,
      "learning_rate": 0.00017391335278628712,
      "loss": 0.0122,
      "step": 1180
    },
    {
      "epoch": 23.336633663366335,
      "grad_norm": 0.05201767012476921,
      "learning_rate": 0.0001734882604821276,
      "loss": 0.0125,
      "step": 1190
    },
    {
      "epoch": 23.534653465346533,
      "grad_norm": 0.34572598338127136,
      "learning_rate": 0.00017306026117619889,
      "loss": 0.0168,
      "step": 1200
    },
    {
      "epoch": 23.73267326732673,
      "grad_norm": 0.22629716992378235,
      "learning_rate": 0.00017262937179902472,
      "loss": 0.0143,
      "step": 1210
    },
    {
      "epoch": 23.93069306930693,
      "grad_norm": 0.2683790922164917,
      "learning_rate": 0.00017219560939545246,
      "loss": 0.0159,
      "step": 1220
    },
    {
      "epoch": 24.11881188118812,
      "grad_norm": 0.17118693888187408,
      "learning_rate": 0.0001717589911239788,
      "loss": 0.016,
      "step": 1230
    },
    {
      "epoch": 24.316831683168317,
      "grad_norm": 0.105218805372715,
      "learning_rate": 0.00017131953425607104,
      "loss": 0.016,
      "step": 1240
    },
    {
      "epoch": 24.514851485148515,
      "grad_norm": 0.20184026658535004,
      "learning_rate": 0.00017087725617548385,
      "loss": 0.0183,
      "step": 1250
    },
    {
      "epoch": 24.712871287128714,
      "grad_norm": 0.2845208942890167,
      "learning_rate": 0.00017043217437757164,
      "loss": 0.0189,
      "step": 1260
    },
    {
      "epoch": 24.91089108910891,
      "grad_norm": 0.2577328383922577,
      "learning_rate": 0.00016998430646859654,
      "loss": 0.0196,
      "step": 1270
    },
    {
      "epoch": 25.099009900990097,
      "grad_norm": 0.22288930416107178,
      "learning_rate": 0.00016953367016503182,
      "loss": 0.0177,
      "step": 1280
    },
    {
      "epoch": 25.297029702970296,
      "grad_norm": 0.36929747462272644,
      "learning_rate": 0.00016908028329286112,
      "loss": 0.0202,
      "step": 1290
    },
    {
      "epoch": 25.495049504950494,
      "grad_norm": 0.2899525463581085,
      "learning_rate": 0.0001686241637868734,
      "loss": 0.0213,
      "step": 1300
    },
    {
      "epoch": 25.693069306930692,
      "grad_norm": 0.32868900895118713,
      "learning_rate": 0.00016816532968995328,
      "loss": 0.0182,
      "step": 1310
    },
    {
      "epoch": 25.89108910891089,
      "grad_norm": 0.3192131817340851,
      "learning_rate": 0.00016770379915236766,
      "loss": 0.0218,
      "step": 1320
    },
    {
      "epoch": 26.07920792079208,
      "grad_norm": 0.26361629366874695,
      "learning_rate": 0.00016723959043104728,
      "loss": 0.0179,
      "step": 1330
    },
    {
      "epoch": 26.277227722772277,
      "grad_norm": 0.3538498282432556,
      "learning_rate": 0.00016677272188886483,
      "loss": 0.0168,
      "step": 1340
    },
    {
      "epoch": 26.475247524752476,
      "grad_norm": 0.19713373482227325,
      "learning_rate": 0.00016630321199390867,
      "loss": 0.0173,
      "step": 1350
    },
    {
      "epoch": 26.673267326732674,
      "grad_norm": 0.29515835642814636,
      "learning_rate": 0.00016583107931875192,
      "loss": 0.0177,
      "step": 1360
    },
    {
      "epoch": 26.871287128712872,
      "grad_norm": 0.09990435093641281,
      "learning_rate": 0.00016535634253971794,
      "loss": 0.0167,
      "step": 1370
    },
    {
      "epoch": 27.059405940594058,
      "grad_norm": 0.308371365070343,
      "learning_rate": 0.00016487902043614173,
      "loss": 0.0165,
      "step": 1380
    },
    {
      "epoch": 27.257425742574256,
      "grad_norm": 0.2830803394317627,
      "learning_rate": 0.00016439913188962685,
      "loss": 0.0155,
      "step": 1390
    },
    {
      "epoch": 27.455445544554454,
      "grad_norm": 0.21548472344875336,
      "learning_rate": 0.0001639166958832985,
      "loss": 0.0163,
      "step": 1400
    },
    {
      "epoch": 27.653465346534652,
      "grad_norm": 0.1809612214565277,
      "learning_rate": 0.00016343173150105278,
      "loss": 0.0146,
      "step": 1410
    },
    {
      "epoch": 27.85148514851485,
      "grad_norm": 0.2266589254140854,
      "learning_rate": 0.0001629442579268016,
      "loss": 0.0144,
      "step": 1420
    },
    {
      "epoch": 28.03960396039604,
      "grad_norm": 0.06419487297534943,
      "learning_rate": 0.0001624542944437139,
      "loss": 0.0139,
      "step": 1430
    },
    {
      "epoch": 28.237623762376238,
      "grad_norm": 0.15811343491077423,
      "learning_rate": 0.00016196186043345288,
      "loss": 0.0136,
      "step": 1440
    },
    {
      "epoch": 28.435643564356436,
      "grad_norm": 0.15876223146915436,
      "learning_rate": 0.00016146697537540924,
      "loss": 0.0132,
      "step": 1450
    },
    {
      "epoch": 28.633663366336634,
      "grad_norm": 0.14463020861148834,
      "learning_rate": 0.0001609696588459307,
      "loss": 0.0129,
      "step": 1460
    },
    {
      "epoch": 28.831683168316832,
      "grad_norm": 0.2847503423690796,
      "learning_rate": 0.00016046993051754756,
      "loss": 0.0139,
      "step": 1470
    },
    {
      "epoch": 29.019801980198018,
      "grad_norm": 0.26266077160835266,
      "learning_rate": 0.0001599678101581945,
      "loss": 0.015,
      "step": 1480
    },
    {
      "epoch": 29.217821782178216,
      "grad_norm": 0.11117064952850342,
      "learning_rate": 0.00015946331763042867,
      "loss": 0.012,
      "step": 1490
    },
    {
      "epoch": 29.415841584158414,
      "grad_norm": 0.041199155151844025,
      "learning_rate": 0.00015895647289064396,
      "loss": 0.0113,
      "step": 1500
    },
    {
      "epoch": 29.415841584158414,
      "eval_loss": 1.8528623580932617,
      "eval_runtime": 75.9027,
      "eval_samples_per_second": 4.176,
      "eval_steps_per_second": 2.095,
      "step": 1500
    },
    {
      "epoch": 29.613861386138613,
      "grad_norm": 0.4097837507724762,
      "learning_rate": 0.0001584472959882815,
      "loss": 0.014,
      "step": 1510
    },
    {
      "epoch": 29.81188118811881,
      "grad_norm": 0.10821210592985153,
      "learning_rate": 0.0001579358070650367,
      "loss": 0.0129,
      "step": 1520
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.11605525761842728,
      "learning_rate": 0.00015742202635406235,
      "loss": 0.0129,
      "step": 1530
    },
    {
      "epoch": 30.198019801980198,
      "grad_norm": 0.21412257850170135,
      "learning_rate": 0.0001569059741791684,
      "loss": 0.011,
      "step": 1540
    },
    {
      "epoch": 30.396039603960396,
      "grad_norm": 0.0895719975233078,
      "learning_rate": 0.0001563876709540178,
      "loss": 0.0129,
      "step": 1550
    },
    {
      "epoch": 30.594059405940595,
      "grad_norm": 0.18187634646892548,
      "learning_rate": 0.00015586713718131922,
      "loss": 0.0127,
      "step": 1560
    },
    {
      "epoch": 30.792079207920793,
      "grad_norm": 0.10705845803022385,
      "learning_rate": 0.0001553443934520159,
      "loss": 0.014,
      "step": 1570
    },
    {
      "epoch": 30.99009900990099,
      "grad_norm": 0.1223723515868187,
      "learning_rate": 0.00015481946044447099,
      "loss": 0.0136,
      "step": 1580
    },
    {
      "epoch": 31.178217821782177,
      "grad_norm": 0.046719685196876526,
      "learning_rate": 0.00015429235892364994,
      "loss": 0.0122,
      "step": 1590
    },
    {
      "epoch": 31.376237623762375,
      "grad_norm": 0.06917575001716614,
      "learning_rate": 0.00015376310974029873,
      "loss": 0.0112,
      "step": 1600
    },
    {
      "epoch": 31.574257425742573,
      "grad_norm": 0.045560333877801895,
      "learning_rate": 0.0001532317338301192,
      "loss": 0.0113,
      "step": 1610
    },
    {
      "epoch": 31.77227722772277,
      "grad_norm": 0.06091497838497162,
      "learning_rate": 0.00015269825221294098,
      "loss": 0.0117,
      "step": 1620
    },
    {
      "epoch": 31.97029702970297,
      "grad_norm": 0.07725334167480469,
      "learning_rate": 0.0001521626859918898,
      "loss": 0.0124,
      "step": 1630
    },
    {
      "epoch": 32.15841584158416,
      "grad_norm": 0.0351332388818264,
      "learning_rate": 0.00015162505635255287,
      "loss": 0.0106,
      "step": 1640
    },
    {
      "epoch": 32.35643564356435,
      "grad_norm": 0.0352192148566246,
      "learning_rate": 0.0001510853845621409,
      "loss": 0.0104,
      "step": 1650
    },
    {
      "epoch": 32.554455445544555,
      "grad_norm": 0.03620712086558342,
      "learning_rate": 0.00015054369196864644,
      "loss": 0.011,
      "step": 1660
    },
    {
      "epoch": 32.75247524752475,
      "grad_norm": 0.06429705768823624,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.0116,
      "step": 1670
    },
    {
      "epoch": 32.95049504950495,
      "grad_norm": 0.03938416391611099,
      "learning_rate": 0.0001494543301632219,
      "loss": 0.0116,
      "step": 1680
    },
    {
      "epoch": 33.13861386138614,
      "grad_norm": 0.03617365285754204,
      "learning_rate": 0.0001489067040435717,
      "loss": 0.0109,
      "step": 1690
    },
    {
      "epoch": 33.336633663366335,
      "grad_norm": 0.09898613393306732,
      "learning_rate": 0.00014835714330369446,
      "loss": 0.0108,
      "step": 1700
    },
    {
      "epoch": 33.53465346534654,
      "grad_norm": 0.07661855220794678,
      "learning_rate": 0.0001478056696827636,
      "loss": 0.0115,
      "step": 1710
    },
    {
      "epoch": 33.73267326732673,
      "grad_norm": 0.0875670462846756,
      "learning_rate": 0.00014725230499562119,
      "loss": 0.0112,
      "step": 1720
    },
    {
      "epoch": 33.93069306930693,
      "grad_norm": 0.045731861144304276,
      "learning_rate": 0.00014669707113191483,
      "loss": 0.012,
      "step": 1730
    },
    {
      "epoch": 34.118811881188115,
      "grad_norm": 0.03665975108742714,
      "learning_rate": 0.00014613999005523174,
      "loss": 0.0108,
      "step": 1740
    },
    {
      "epoch": 34.31683168316832,
      "grad_norm": 0.07179899513721466,
      "learning_rate": 0.00014558108380223012,
      "loss": 0.0106,
      "step": 1750
    },
    {
      "epoch": 34.51485148514851,
      "grad_norm": 0.04096843674778938,
      "learning_rate": 0.00014502037448176734,
      "loss": 0.0112,
      "step": 1760
    },
    {
      "epoch": 34.71287128712871,
      "grad_norm": 0.03616851940751076,
      "learning_rate": 0.00014445788427402528,
      "loss": 0.0112,
      "step": 1770
    },
    {
      "epoch": 34.91089108910891,
      "grad_norm": 0.2000083476305008,
      "learning_rate": 0.00014389363542963306,
      "loss": 0.0119,
      "step": 1780
    },
    {
      "epoch": 35.0990099009901,
      "grad_norm": 0.03873177245259285,
      "learning_rate": 0.00014332765026878687,
      "loss": 0.0108,
      "step": 1790
    },
    {
      "epoch": 35.2970297029703,
      "grad_norm": 0.030958373099565506,
      "learning_rate": 0.00014275995118036693,
      "loss": 0.0099,
      "step": 1800
    },
    {
      "epoch": 35.495049504950494,
      "grad_norm": 0.03003741428256035,
      "learning_rate": 0.00014219056062105193,
      "loss": 0.01,
      "step": 1810
    },
    {
      "epoch": 35.693069306930695,
      "grad_norm": 0.03531872481107712,
      "learning_rate": 0.00014161950111443077,
      "loss": 0.0111,
      "step": 1820
    },
    {
      "epoch": 35.89108910891089,
      "grad_norm": 0.041126642376184464,
      "learning_rate": 0.0001410467952501114,
      "loss": 0.0115,
      "step": 1830
    },
    {
      "epoch": 36.07920792079208,
      "grad_norm": 0.028950873762369156,
      "learning_rate": 0.00014047246568282736,
      "loss": 0.0106,
      "step": 1840
    },
    {
      "epoch": 36.277227722772274,
      "grad_norm": 0.07969832420349121,
      "learning_rate": 0.00013989653513154165,
      "loss": 0.01,
      "step": 1850
    },
    {
      "epoch": 36.475247524752476,
      "grad_norm": 0.03239164501428604,
      "learning_rate": 0.0001393190263785479,
      "loss": 0.0104,
      "step": 1860
    },
    {
      "epoch": 36.67326732673267,
      "grad_norm": 0.088913194835186,
      "learning_rate": 0.00013873996226856933,
      "loss": 0.0112,
      "step": 1870
    },
    {
      "epoch": 36.87128712871287,
      "grad_norm": 0.34416264295578003,
      "learning_rate": 0.00013815936570785487,
      "loss": 0.0127,
      "step": 1880
    },
    {
      "epoch": 37.05940594059406,
      "grad_norm": 0.14783936738967896,
      "learning_rate": 0.00013757725966327322,
      "loss": 0.0119,
      "step": 1890
    },
    {
      "epoch": 37.257425742574256,
      "grad_norm": 0.09111394733190536,
      "learning_rate": 0.00013699366716140435,
      "loss": 0.0116,
      "step": 1900
    },
    {
      "epoch": 37.45544554455446,
      "grad_norm": 0.10149216651916504,
      "learning_rate": 0.0001364086112876284,
      "loss": 0.0113,
      "step": 1910
    },
    {
      "epoch": 37.65346534653465,
      "grad_norm": 0.058421943336725235,
      "learning_rate": 0.00013582211518521273,
      "loss": 0.0113,
      "step": 1920
    },
    {
      "epoch": 37.851485148514854,
      "grad_norm": 0.11640463024377823,
      "learning_rate": 0.00013523420205439646,
      "loss": 0.012,
      "step": 1930
    },
    {
      "epoch": 38.039603960396036,
      "grad_norm": 0.07603046298027039,
      "learning_rate": 0.00013464489515147238,
      "loss": 0.0114,
      "step": 1940
    },
    {
      "epoch": 38.23762376237624,
      "grad_norm": 0.03392176330089569,
      "learning_rate": 0.00013405421778786737,
      "loss": 0.0108,
      "step": 1950
    },
    {
      "epoch": 38.43564356435643,
      "grad_norm": 0.2335938662290573,
      "learning_rate": 0.00013346219332922016,
      "loss": 0.0108,
      "step": 1960
    },
    {
      "epoch": 38.633663366336634,
      "grad_norm": 0.07149430364370346,
      "learning_rate": 0.0001328688451944569,
      "loss": 0.0113,
      "step": 1970
    },
    {
      "epoch": 38.83168316831683,
      "grad_norm": 0.15134957432746887,
      "learning_rate": 0.00013227419685486492,
      "loss": 0.0121,
      "step": 1980
    },
    {
      "epoch": 39.01980198019802,
      "grad_norm": 0.031247029080986977,
      "learning_rate": 0.0001316782718331643,
      "loss": 0.011,
      "step": 1990
    },
    {
      "epoch": 39.21782178217822,
      "grad_norm": 0.03347298502922058,
      "learning_rate": 0.00013108109370257712,
      "loss": 0.011,
      "step": 2000
    },
    {
      "epoch": 39.21782178217822,
      "eval_loss": 1.9810045957565308,
      "eval_runtime": 76.3823,
      "eval_samples_per_second": 4.15,
      "eval_steps_per_second": 2.082,
      "step": 2000
    },
    {
      "epoch": 39.415841584158414,
      "grad_norm": 0.09714417904615402,
      "learning_rate": 0.00013048268608589533,
      "loss": 0.0116,
      "step": 2010
    },
    {
      "epoch": 39.613861386138616,
      "grad_norm": 0.05139007791876793,
      "learning_rate": 0.00012988307265454597,
      "loss": 0.0111,
      "step": 2020
    },
    {
      "epoch": 39.81188118811881,
      "grad_norm": 0.04015783220529556,
      "learning_rate": 0.00012928227712765504,
      "loss": 0.0117,
      "step": 2030
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.20379693806171417,
      "learning_rate": 0.00012868032327110904,
      "loss": 0.0128,
      "step": 2040
    },
    {
      "epoch": 40.198019801980195,
      "grad_norm": 0.1887517124414444,
      "learning_rate": 0.00012807723489661495,
      "loss": 0.0119,
      "step": 2050
    },
    {
      "epoch": 40.396039603960396,
      "grad_norm": 0.0888703465461731,
      "learning_rate": 0.0001274730358607583,
      "loss": 0.0136,
      "step": 2060
    },
    {
      "epoch": 40.59405940594059,
      "grad_norm": 0.16852861642837524,
      "learning_rate": 0.00012686775006405946,
      "loss": 0.0124,
      "step": 2070
    },
    {
      "epoch": 40.79207920792079,
      "grad_norm": 0.0944506973028183,
      "learning_rate": 0.0001262614014500282,
      "loss": 0.0134,
      "step": 2080
    },
    {
      "epoch": 40.99009900990099,
      "grad_norm": 0.09365811944007874,
      "learning_rate": 0.00012565401400421651,
      "loss": 0.0137,
      "step": 2090
    },
    {
      "epoch": 41.17821782178218,
      "grad_norm": 0.15938587486743927,
      "learning_rate": 0.00012504561175326985,
      "loss": 0.0124,
      "step": 2100
    },
    {
      "epoch": 41.37623762376238,
      "grad_norm": 0.19702400267124176,
      "learning_rate": 0.0001244362187639767,
      "loss": 0.0135,
      "step": 2110
    },
    {
      "epoch": 41.57425742574257,
      "grad_norm": 0.21812361478805542,
      "learning_rate": 0.0001238258591423165,
      "loss": 0.0132,
      "step": 2120
    },
    {
      "epoch": 41.772277227722775,
      "grad_norm": 0.1647273451089859,
      "learning_rate": 0.00012321455703250616,
      "loss": 0.0121,
      "step": 2130
    },
    {
      "epoch": 41.97029702970297,
      "grad_norm": 0.047169461846351624,
      "learning_rate": 0.0001226023366160449,
      "loss": 0.0123,
      "step": 2140
    },
    {
      "epoch": 42.15841584158416,
      "grad_norm": 0.10308068245649338,
      "learning_rate": 0.00012198922211075778,
      "loss": 0.0113,
      "step": 2150
    },
    {
      "epoch": 42.35643564356435,
      "grad_norm": 0.0610012523829937,
      "learning_rate": 0.00012137523776983757,
      "loss": 0.0131,
      "step": 2160
    },
    {
      "epoch": 42.554455445544555,
      "grad_norm": 0.12856794893741608,
      "learning_rate": 0.00012076040788088554,
      "loss": 0.0116,
      "step": 2170
    },
    {
      "epoch": 42.75247524752475,
      "grad_norm": 0.09634555876255035,
      "learning_rate": 0.00012014475676495052,
      "loss": 0.0127,
      "step": 2180
    },
    {
      "epoch": 42.95049504950495,
      "grad_norm": 0.21406149864196777,
      "learning_rate": 0.000119528308775567,
      "loss": 0.0124,
      "step": 2190
    },
    {
      "epoch": 43.13861386138614,
      "grad_norm": 0.03979460150003433,
      "learning_rate": 0.00011891108829779165,
      "loss": 0.0116,
      "step": 2200
    },
    {
      "epoch": 43.336633663366335,
      "grad_norm": 0.05667963996529579,
      "learning_rate": 0.00011829311974723867,
      "loss": 0.0112,
      "step": 2210
    },
    {
      "epoch": 43.53465346534654,
      "grad_norm": 0.07588782906532288,
      "learning_rate": 0.00011767442756911417,
      "loss": 0.0114,
      "step": 2220
    },
    {
      "epoch": 43.73267326732673,
      "grad_norm": 0.06999919563531876,
      "learning_rate": 0.00011705503623724898,
      "loss": 0.0118,
      "step": 2230
    },
    {
      "epoch": 43.93069306930693,
      "grad_norm": 0.08128093928098679,
      "learning_rate": 0.00011643497025313061,
      "loss": 0.0118,
      "step": 2240
    },
    {
      "epoch": 44.118811881188115,
      "grad_norm": 0.03330959379673004,
      "learning_rate": 0.0001158142541449341,
      "loss": 0.0109,
      "step": 2250
    },
    {
      "epoch": 44.31683168316832,
      "grad_norm": 0.060158368200063705,
      "learning_rate": 0.0001151929124665516,
      "loss": 0.0109,
      "step": 2260
    },
    {
      "epoch": 44.51485148514851,
      "grad_norm": 0.1863548904657364,
      "learning_rate": 0.00011457096979662114,
      "loss": 0.011,
      "step": 2270
    },
    {
      "epoch": 44.71287128712871,
      "grad_norm": 0.03855551779270172,
      "learning_rate": 0.00011394845073755455,
      "loss": 0.0113,
      "step": 2280
    },
    {
      "epoch": 44.91089108910891,
      "grad_norm": 0.06027248874306679,
      "learning_rate": 0.00011332537991456398,
      "loss": 0.0111,
      "step": 2290
    },
    {
      "epoch": 45.0990099009901,
      "grad_norm": 0.1333802491426468,
      "learning_rate": 0.00011270178197468789,
      "loss": 0.0112,
      "step": 2300
    },
    {
      "epoch": 45.2970297029703,
      "grad_norm": 0.07712238281965256,
      "learning_rate": 0.00011207768158581613,
      "loss": 0.0099,
      "step": 2310
    },
    {
      "epoch": 45.495049504950494,
      "grad_norm": 0.03724438324570656,
      "learning_rate": 0.00011145310343571411,
      "loss": 0.011,
      "step": 2320
    },
    {
      "epoch": 45.693069306930695,
      "grad_norm": 0.0400845967233181,
      "learning_rate": 0.0001108280722310462,
      "loss": 0.011,
      "step": 2330
    },
    {
      "epoch": 45.89108910891089,
      "grad_norm": 0.041330672800540924,
      "learning_rate": 0.00011020261269639842,
      "loss": 0.0108,
      "step": 2340
    },
    {
      "epoch": 46.07920792079208,
      "grad_norm": 0.03175568953156471,
      "learning_rate": 0.00010957674957330042,
      "loss": 0.0102,
      "step": 2350
    },
    {
      "epoch": 46.277227722772274,
      "grad_norm": 0.027505742385983467,
      "learning_rate": 0.00010895050761924668,
      "loss": 0.0098,
      "step": 2360
    },
    {
      "epoch": 46.475247524752476,
      "grad_norm": 0.031385671347379684,
      "learning_rate": 0.00010832391160671729,
      "loss": 0.0103,
      "step": 2370
    },
    {
      "epoch": 46.67326732673267,
      "grad_norm": 0.03100833296775818,
      "learning_rate": 0.00010769698632219794,
      "loss": 0.0101,
      "step": 2380
    },
    {
      "epoch": 46.87128712871287,
      "grad_norm": 0.04194655641913414,
      "learning_rate": 0.00010706975656519946,
      "loss": 0.0105,
      "step": 2390
    },
    {
      "epoch": 47.05940594059406,
      "grad_norm": 0.17276820540428162,
      "learning_rate": 0.00010644224714727681,
      "loss": 0.0105,
      "step": 2400
    },
    {
      "epoch": 47.257425742574256,
      "grad_norm": 0.032278869301080704,
      "learning_rate": 0.00010581448289104758,
      "loss": 0.0094,
      "step": 2410
    },
    {
      "epoch": 47.45544554455446,
      "grad_norm": 0.03324766084551811,
      "learning_rate": 0.00010518648862921012,
      "loss": 0.01,
      "step": 2420
    },
    {
      "epoch": 47.65346534653465,
      "grad_norm": 0.03345724940299988,
      "learning_rate": 0.00010455828920356115,
      "loss": 0.0103,
      "step": 2430
    },
    {
      "epoch": 47.851485148514854,
      "grad_norm": 0.034674759954214096,
      "learning_rate": 0.00010392990946401313,
      "loss": 0.0105,
      "step": 2440
    },
    {
      "epoch": 48.039603960396036,
      "grad_norm": 0.024850964546203613,
      "learning_rate": 0.00010330137426761135,
      "loss": 0.0109,
      "step": 2450
    },
    {
      "epoch": 48.23762376237624,
      "grad_norm": 0.0305289626121521,
      "learning_rate": 0.00010267270847755048,
      "loss": 0.0096,
      "step": 2460
    },
    {
      "epoch": 48.43564356435643,
      "grad_norm": 0.026470104232430458,
      "learning_rate": 0.00010204393696219117,
      "loss": 0.0099,
      "step": 2470
    },
    {
      "epoch": 48.633663366336634,
      "grad_norm": 0.035170264542102814,
      "learning_rate": 0.00010141508459407623,
      "loss": 0.0105,
      "step": 2480
    },
    {
      "epoch": 48.83168316831683,
      "grad_norm": 0.03382629156112671,
      "learning_rate": 0.00010078617624894684,
      "loss": 0.0107,
      "step": 2490
    },
    {
      "epoch": 49.01980198019802,
      "grad_norm": 0.02712225168943405,
      "learning_rate": 0.00010015723680475846,
      "loss": 0.0104,
      "step": 2500
    },
    {
      "epoch": 49.01980198019802,
      "eval_loss": 2.0715720653533936,
      "eval_runtime": 61.2747,
      "eval_samples_per_second": 5.173,
      "eval_steps_per_second": 2.595,
      "step": 2500
    },
    {
      "epoch": 49.21782178217822,
      "grad_norm": 0.2509881854057312,
      "learning_rate": 9.95282911406968e-05,
      "loss": 0.0105,
      "step": 2510
    },
    {
      "epoch": 49.415841584158414,
      "grad_norm": 0.03043426387012005,
      "learning_rate": 9.889936413619356e-05,
      "loss": 0.0098,
      "step": 2520
    },
    {
      "epoch": 49.613861386138616,
      "grad_norm": 0.03412380814552307,
      "learning_rate": 9.827048066994225e-05,
      "loss": 0.0115,
      "step": 2530
    },
    {
      "epoch": 49.81188118811881,
      "grad_norm": 0.20148912072181702,
      "learning_rate": 9.764166561891432e-05,
      "loss": 0.0116,
      "step": 2540
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.26765531301498413,
      "learning_rate": 9.70129438573747e-05,
      "loss": 0.0118,
      "step": 2550
    },
    {
      "epoch": 50.198019801980195,
      "grad_norm": 0.0966796949505806,
      "learning_rate": 9.63843402558981e-05,
      "loss": 0.0111,
      "step": 2560
    },
    {
      "epoch": 50.396039603960396,
      "grad_norm": 0.11040028184652328,
      "learning_rate": 9.57558796803852e-05,
      "loss": 0.01,
      "step": 2570
    },
    {
      "epoch": 50.59405940594059,
      "grad_norm": 0.1609763503074646,
      "learning_rate": 9.512758699107879e-05,
      "loss": 0.011,
      "step": 2580
    },
    {
      "epoch": 50.79207920792079,
      "grad_norm": 0.07201898843050003,
      "learning_rate": 9.449948704158071e-05,
      "loss": 0.0106,
      "step": 2590
    },
    {
      "epoch": 50.99009900990099,
      "grad_norm": 0.06254687905311584,
      "learning_rate": 9.38716046778684e-05,
      "loss": 0.011,
      "step": 2600
    },
    {
      "epoch": 51.17821782178218,
      "grad_norm": 0.03031199239194393,
      "learning_rate": 9.324396473731217e-05,
      "loss": 0.0103,
      "step": 2610
    },
    {
      "epoch": 51.37623762376238,
      "grad_norm": 0.05816170200705528,
      "learning_rate": 9.261659204769284e-05,
      "loss": 0.0098,
      "step": 2620
    },
    {
      "epoch": 51.57425742574257,
      "grad_norm": 0.14442995190620422,
      "learning_rate": 9.198951142621929e-05,
      "loss": 0.0107,
      "step": 2630
    },
    {
      "epoch": 51.772277227722775,
      "grad_norm": 0.10595759749412537,
      "learning_rate": 9.136274767854716e-05,
      "loss": 0.0105,
      "step": 2640
    },
    {
      "epoch": 51.97029702970297,
      "grad_norm": 0.029600875452160835,
      "learning_rate": 9.07363255977973e-05,
      "loss": 0.0108,
      "step": 2650
    },
    {
      "epoch": 52.15841584158416,
      "grad_norm": 0.08049221336841583,
      "learning_rate": 9.011026996357503e-05,
      "loss": 0.0097,
      "step": 2660
    },
    {
      "epoch": 52.35643564356435,
      "grad_norm": 0.026186183094978333,
      "learning_rate": 8.948460554099018e-05,
      "loss": 0.0103,
      "step": 2670
    },
    {
      "epoch": 52.554455445544555,
      "grad_norm": 0.17796003818511963,
      "learning_rate": 8.885935707967716e-05,
      "loss": 0.0108,
      "step": 2680
    },
    {
      "epoch": 52.75247524752475,
      "grad_norm": 0.09132158011198044,
      "learning_rate": 8.823454931281616e-05,
      "loss": 0.0105,
      "step": 2690
    },
    {
      "epoch": 52.95049504950495,
      "grad_norm": 0.028834670782089233,
      "learning_rate": 8.76102069561545e-05,
      "loss": 0.0113,
      "step": 2700
    },
    {
      "epoch": 53.13861386138614,
      "grad_norm": 0.21559055149555206,
      "learning_rate": 8.698635470702923e-05,
      "loss": 0.0145,
      "step": 2710
    },
    {
      "epoch": 53.336633663366335,
      "grad_norm": 0.1729169338941574,
      "learning_rate": 8.636301724339004e-05,
      "loss": 0.0123,
      "step": 2720
    },
    {
      "epoch": 53.53465346534654,
      "grad_norm": 0.05573170632123947,
      "learning_rate": 8.574021922282292e-05,
      "loss": 0.0109,
      "step": 2730
    },
    {
      "epoch": 53.73267326732673,
      "grad_norm": 0.033353183418512344,
      "learning_rate": 8.511798528157512e-05,
      "loss": 0.011,
      "step": 2740
    },
    {
      "epoch": 53.93069306930693,
      "grad_norm": 0.03874226659536362,
      "learning_rate": 8.449634003358022e-05,
      "loss": 0.0116,
      "step": 2750
    },
    {
      "epoch": 54.118811881188115,
      "grad_norm": 0.029603777453303337,
      "learning_rate": 8.387530806948476e-05,
      "loss": 0.0101,
      "step": 2760
    },
    {
      "epoch": 54.31683168316832,
      "grad_norm": 0.11726613342761993,
      "learning_rate": 8.325491395567541e-05,
      "loss": 0.0109,
      "step": 2770
    },
    {
      "epoch": 54.51485148514851,
      "grad_norm": 0.04111700505018234,
      "learning_rate": 8.263518223330697e-05,
      "loss": 0.0106,
      "step": 2780
    },
    {
      "epoch": 54.71287128712871,
      "grad_norm": 0.06498590111732483,
      "learning_rate": 8.201613741733203e-05,
      "loss": 0.0115,
      "step": 2790
    },
    {
      "epoch": 54.91089108910891,
      "grad_norm": 0.03625292330980301,
      "learning_rate": 8.13978039955308e-05,
      "loss": 0.0145,
      "step": 2800
    },
    {
      "epoch": 55.0990099009901,
      "grad_norm": 0.10530784726142883,
      "learning_rate": 8.078020642754274e-05,
      "loss": 0.0115,
      "step": 2810
    },
    {
      "epoch": 55.2970297029703,
      "grad_norm": 0.059211425483226776,
      "learning_rate": 8.016336914389874e-05,
      "loss": 0.0098,
      "step": 2820
    },
    {
      "epoch": 55.495049504950494,
      "grad_norm": 0.029871392995119095,
      "learning_rate": 7.954731654505491e-05,
      "loss": 0.0105,
      "step": 2830
    },
    {
      "epoch": 55.693069306930695,
      "grad_norm": 0.034974247217178345,
      "learning_rate": 7.89320730004274e-05,
      "loss": 0.0104,
      "step": 2840
    },
    {
      "epoch": 55.89108910891089,
      "grad_norm": 0.04234467074275017,
      "learning_rate": 7.831766284742807e-05,
      "loss": 0.012,
      "step": 2850
    },
    {
      "epoch": 56.07920792079208,
      "grad_norm": 0.02388872392475605,
      "learning_rate": 7.77041103905023e-05,
      "loss": 0.0104,
      "step": 2860
    },
    {
      "epoch": 56.277227722772274,
      "grad_norm": 0.03674217313528061,
      "learning_rate": 7.709143990016702e-05,
      "loss": 0.0099,
      "step": 2870
    },
    {
      "epoch": 56.475247524752476,
      "grad_norm": 0.028430873528122902,
      "learning_rate": 7.6479675612051e-05,
      "loss": 0.0102,
      "step": 2880
    },
    {
      "epoch": 56.67326732673267,
      "grad_norm": 0.034957416355609894,
      "learning_rate": 7.586884172593609e-05,
      "loss": 0.0101,
      "step": 2890
    },
    {
      "epoch": 56.87128712871287,
      "grad_norm": 0.03257445991039276,
      "learning_rate": 7.525896240479976e-05,
      "loss": 0.0102,
      "step": 2900
    },
    {
      "epoch": 57.05940594059406,
      "grad_norm": 0.025701675564050674,
      "learning_rate": 7.465006177385953e-05,
      "loss": 0.0101,
      "step": 2910
    },
    {
      "epoch": 57.257425742574256,
      "grad_norm": 0.02833356149494648,
      "learning_rate": 7.404216391961847e-05,
      "loss": 0.0095,
      "step": 2920
    },
    {
      "epoch": 57.45544554455446,
      "grad_norm": 0.02513338252902031,
      "learning_rate": 7.343529288891239e-05,
      "loss": 0.0096,
      "step": 2930
    },
    {
      "epoch": 57.65346534653465,
      "grad_norm": 0.02715817466378212,
      "learning_rate": 7.282947268795877e-05,
      "loss": 0.01,
      "step": 2940
    },
    {
      "epoch": 57.851485148514854,
      "grad_norm": 0.030358904972672462,
      "learning_rate": 7.222472728140695e-05,
      "loss": 0.0101,
      "step": 2950
    },
    {
      "epoch": 58.039603960396036,
      "grad_norm": 0.02921466901898384,
      "learning_rate": 7.162108059139032e-05,
      "loss": 0.01,
      "step": 2960
    },
    {
      "epoch": 58.23762376237624,
      "grad_norm": 0.025500759482383728,
      "learning_rate": 7.101855649657991e-05,
      "loss": 0.0093,
      "step": 2970
    },
    {
      "epoch": 58.43564356435643,
      "grad_norm": 0.03553071618080139,
      "learning_rate": 7.041717883123977e-05,
      "loss": 0.0095,
      "step": 2980
    },
    {
      "epoch": 58.633663366336634,
      "grad_norm": 0.03326829522848129,
      "learning_rate": 6.981697138428434e-05,
      "loss": 0.0097,
      "step": 2990
    },
    {
      "epoch": 58.83168316831683,
      "grad_norm": 0.03296545520424843,
      "learning_rate": 6.921795789833723e-05,
      "loss": 0.0102,
      "step": 3000
    },
    {
      "epoch": 58.83168316831683,
      "eval_loss": 2.1854326725006104,
      "eval_runtime": 63.8492,
      "eval_samples_per_second": 4.965,
      "eval_steps_per_second": 2.49,
      "step": 3000
    },
    {
      "epoch": 59.01980198019802,
      "grad_norm": 0.03292296826839447,
      "learning_rate": 6.862016206879216e-05,
      "loss": 0.0101,
      "step": 3010
    },
    {
      "epoch": 59.21782178217822,
      "grad_norm": 0.03038034774363041,
      "learning_rate": 6.802360754287547e-05,
      "loss": 0.0092,
      "step": 3020
    },
    {
      "epoch": 59.415841584158414,
      "grad_norm": 0.031039321795105934,
      "learning_rate": 6.742831791871096e-05,
      "loss": 0.0095,
      "step": 3030
    },
    {
      "epoch": 59.613861386138616,
      "grad_norm": 0.034637823700904846,
      "learning_rate": 6.683431674438612e-05,
      "loss": 0.0096,
      "step": 3040
    },
    {
      "epoch": 59.81188118811881,
      "grad_norm": 0.034620340913534164,
      "learning_rate": 6.624162751702076e-05,
      "loss": 0.01,
      "step": 3050
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.04842129349708557,
      "learning_rate": 6.565027368183769e-05,
      "loss": 0.0102,
      "step": 3060
    },
    {
      "epoch": 60.198019801980195,
      "grad_norm": 0.027831487357616425,
      "learning_rate": 6.506027863123492e-05,
      "loss": 0.0092,
      "step": 3070
    },
    {
      "epoch": 60.396039603960396,
      "grad_norm": 0.028759174048900604,
      "learning_rate": 6.447166570386063e-05,
      "loss": 0.0095,
      "step": 3080
    },
    {
      "epoch": 60.59405940594059,
      "grad_norm": 0.030392073094844818,
      "learning_rate": 6.388445818368991e-05,
      "loss": 0.0096,
      "step": 3090
    },
    {
      "epoch": 60.79207920792079,
      "grad_norm": 0.0333111509680748,
      "learning_rate": 6.329867929910347e-05,
      "loss": 0.01,
      "step": 3100
    },
    {
      "epoch": 60.99009900990099,
      "grad_norm": 0.030943069607019424,
      "learning_rate": 6.271435222196916e-05,
      "loss": 0.0099,
      "step": 3110
    },
    {
      "epoch": 61.17821782178218,
      "grad_norm": 0.027012720704078674,
      "learning_rate": 6.213150006672499e-05,
      "loss": 0.0093,
      "step": 3120
    },
    {
      "epoch": 61.37623762376238,
      "grad_norm": 0.02961895428597927,
      "learning_rate": 6.15501458894651e-05,
      "loss": 0.0094,
      "step": 3130
    },
    {
      "epoch": 61.57425742574257,
      "grad_norm": 0.02908298186957836,
      "learning_rate": 6.097031268702746e-05,
      "loss": 0.0095,
      "step": 3140
    },
    {
      "epoch": 61.772277227722775,
      "grad_norm": 0.03335884213447571,
      "learning_rate": 6.039202339608432e-05,
      "loss": 0.0098,
      "step": 3150
    },
    {
      "epoch": 61.97029702970297,
      "grad_norm": 0.03192156180739403,
      "learning_rate": 5.981530089223489e-05,
      "loss": 0.01,
      "step": 3160
    },
    {
      "epoch": 62.15841584158416,
      "grad_norm": 0.02803504653275013,
      "learning_rate": 5.924016798910037e-05,
      "loss": 0.0094,
      "step": 3170
    },
    {
      "epoch": 62.35643564356435,
      "grad_norm": 0.03200392797589302,
      "learning_rate": 5.866664743742162e-05,
      "loss": 0.0093,
      "step": 3180
    },
    {
      "epoch": 62.554455445544555,
      "grad_norm": 0.03177599608898163,
      "learning_rate": 5.809476192415905e-05,
      "loss": 0.0097,
      "step": 3190
    },
    {
      "epoch": 62.75247524752475,
      "grad_norm": 0.02948221005499363,
      "learning_rate": 5.752453407159522e-05,
      "loss": 0.0098,
      "step": 3200
    },
    {
      "epoch": 62.95049504950495,
      "grad_norm": 0.02995140664279461,
      "learning_rate": 5.69559864364402e-05,
      "loss": 0.0099,
      "step": 3210
    },
    {
      "epoch": 63.13861386138614,
      "grad_norm": 0.029279688373208046,
      "learning_rate": 5.6389141508938903e-05,
      "loss": 0.0092,
      "step": 3220
    },
    {
      "epoch": 63.336633663366335,
      "grad_norm": 0.025498950853943825,
      "learning_rate": 5.5824021711981686e-05,
      "loss": 0.0093,
      "step": 3230
    },
    {
      "epoch": 63.53465346534654,
      "grad_norm": 0.03137169033288956,
      "learning_rate": 5.5260649400217326e-05,
      "loss": 0.0095,
      "step": 3240
    },
    {
      "epoch": 63.73267326732673,
      "grad_norm": 0.028380287811160088,
      "learning_rate": 5.469904685916861e-05,
      "loss": 0.0098,
      "step": 3250
    },
    {
      "epoch": 63.93069306930693,
      "grad_norm": 0.031060099601745605,
      "learning_rate": 5.4139236304350935e-05,
      "loss": 0.0099,
      "step": 3260
    },
    {
      "epoch": 64.11881188118812,
      "grad_norm": 0.025437017902731895,
      "learning_rate": 5.3581239880393375e-05,
      "loss": 0.0094,
      "step": 3270
    },
    {
      "epoch": 64.31683168316832,
      "grad_norm": 0.025531411170959473,
      "learning_rate": 5.302507966016295e-05,
      "loss": 0.0091,
      "step": 3280
    },
    {
      "epoch": 64.51485148514851,
      "grad_norm": 0.03445703908801079,
      "learning_rate": 5.247077764389099e-05,
      "loss": 0.0095,
      "step": 3290
    },
    {
      "epoch": 64.7128712871287,
      "grad_norm": 0.03112807497382164,
      "learning_rate": 5.191835575830352e-05,
      "loss": 0.0098,
      "step": 3300
    },
    {
      "epoch": 64.91089108910892,
      "grad_norm": 0.032531820237636566,
      "learning_rate": 5.136783585575336e-05,
      "loss": 0.0099,
      "step": 3310
    },
    {
      "epoch": 65.0990099009901,
      "grad_norm": 0.028146883472800255,
      "learning_rate": 5.081923971335582e-05,
      "loss": 0.0095,
      "step": 3320
    },
    {
      "epoch": 65.29702970297029,
      "grad_norm": 0.03265933692455292,
      "learning_rate": 5.0272589032127594e-05,
      "loss": 0.0091,
      "step": 3330
    },
    {
      "epoch": 65.4950495049505,
      "grad_norm": 0.026051949709653854,
      "learning_rate": 4.972790543612783e-05,
      "loss": 0.0094,
      "step": 3340
    },
    {
      "epoch": 65.6930693069307,
      "grad_norm": 0.031889524310827255,
      "learning_rate": 4.918521047160308e-05,
      "loss": 0.0096,
      "step": 3350
    },
    {
      "epoch": 65.89108910891089,
      "grad_norm": 0.032141778618097305,
      "learning_rate": 4.864452560613485e-05,
      "loss": 0.0098,
      "step": 3360
    },
    {
      "epoch": 66.07920792079207,
      "grad_norm": 0.026572005823254585,
      "learning_rate": 4.810587222779043e-05,
      "loss": 0.0096,
      "step": 3370
    },
    {
      "epoch": 66.27722772277228,
      "grad_norm": 0.026687586680054665,
      "learning_rate": 4.756927164427685e-05,
      "loss": 0.0091,
      "step": 3380
    },
    {
      "epoch": 66.47524752475248,
      "grad_norm": 0.029472634196281433,
      "learning_rate": 4.703474508209793e-05,
      "loss": 0.0093,
      "step": 3390
    },
    {
      "epoch": 66.67326732673267,
      "grad_norm": 0.03171619027853012,
      "learning_rate": 4.650231368571486e-05,
      "loss": 0.0096,
      "step": 3400
    },
    {
      "epoch": 66.87128712871286,
      "grad_norm": 0.02731896936893463,
      "learning_rate": 4.597199851670932e-05,
      "loss": 0.0098,
      "step": 3410
    },
    {
      "epoch": 67.05940594059406,
      "grad_norm": 0.028085224330425262,
      "learning_rate": 4.54438205529508e-05,
      "loss": 0.0095,
      "step": 3420
    },
    {
      "epoch": 67.25742574257426,
      "grad_norm": 0.033895280212163925,
      "learning_rate": 4.491780068776663e-05,
      "loss": 0.0092,
      "step": 3430
    },
    {
      "epoch": 67.45544554455445,
      "grad_norm": 0.030533580109477043,
      "learning_rate": 4.4393959729115244e-05,
      "loss": 0.0094,
      "step": 3440
    },
    {
      "epoch": 67.65346534653466,
      "grad_norm": 0.032808952033519745,
      "learning_rate": 4.387231839876349e-05,
      "loss": 0.0095,
      "step": 3450
    },
    {
      "epoch": 67.85148514851485,
      "grad_norm": 0.02919154055416584,
      "learning_rate": 4.335289733146665e-05,
      "loss": 0.0097,
      "step": 3460
    },
    {
      "epoch": 68.03960396039604,
      "grad_norm": 0.02805473655462265,
      "learning_rate": 4.283571707415214e-05,
      "loss": 0.0098,
      "step": 3470
    },
    {
      "epoch": 68.23762376237623,
      "grad_norm": 0.02642083168029785,
      "learning_rate": 4.2320798085107036e-05,
      "loss": 0.0091,
      "step": 3480
    },
    {
      "epoch": 68.43564356435644,
      "grad_norm": 0.02746916376054287,
      "learning_rate": 4.18081607331685e-05,
      "loss": 0.0092,
      "step": 3490
    },
    {
      "epoch": 68.63366336633663,
      "grad_norm": 0.030231185257434845,
      "learning_rate": 4.129782529691815e-05,
      "loss": 0.0095,
      "step": 3500
    },
    {
      "epoch": 68.63366336633663,
      "eval_loss": 2.408777952194214,
      "eval_runtime": 62.4857,
      "eval_samples_per_second": 5.073,
      "eval_steps_per_second": 2.545,
      "step": 3500
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0035978035769508e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}