{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4998213966779782,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000893016610108948,
      "grad_norm": 3.7484283447265625,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 1.4058,
      "step": 10
    },
    {
      "epoch": 0.001786033220217896,
      "grad_norm": 1.9000362157821655,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.3045,
      "step": 20
    },
    {
      "epoch": 0.002679049830326844,
      "grad_norm": 2.4037389755249023,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.1338,
      "step": 30
    },
    {
      "epoch": 0.003572066440435792,
      "grad_norm": 1.0070688724517822,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.0082,
      "step": 40
    },
    {
      "epoch": 0.00446508305054474,
      "grad_norm": 0.6030862927436829,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 0.8595,
      "step": 50
    },
    {
      "epoch": 0.005358099660653688,
      "grad_norm": 0.6932407021522522,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 0.7768,
      "step": 60
    },
    {
      "epoch": 0.006251116270762636,
      "grad_norm": 0.5126384496688843,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.7961,
      "step": 70
    },
    {
      "epoch": 0.007144132880871584,
      "grad_norm": 0.5326073169708252,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.777,
      "step": 80
    },
    {
      "epoch": 0.008037149490980533,
      "grad_norm": 0.4840937554836273,
      "learning_rate": 1.005e-05,
      "loss": 0.8092,
      "step": 90
    },
    {
      "epoch": 0.00893016610108948,
      "grad_norm": 0.4524643123149872,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.7689,
      "step": 100
    },
    {
      "epoch": 0.009823182711198428,
      "grad_norm": 0.4209403097629547,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.7862,
      "step": 110
    },
    {
      "epoch": 0.010716199321307376,
      "grad_norm": 0.47385457158088684,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.7922,
      "step": 120
    },
    {
      "epoch": 0.011609215931416325,
      "grad_norm": 0.4067447781562805,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.7441,
      "step": 130
    },
    {
      "epoch": 0.012502232541525273,
      "grad_norm": 0.467846542596817,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.8082,
      "step": 140
    },
    {
      "epoch": 0.01339524915163422,
      "grad_norm": 0.5435770153999329,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.8008,
      "step": 150
    },
    {
      "epoch": 0.014288265761743168,
      "grad_norm": 0.4837871789932251,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.7348,
      "step": 160
    },
    {
      "epoch": 0.015181282371852116,
      "grad_norm": 0.44955676794052124,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.7796,
      "step": 170
    },
    {
      "epoch": 0.016074298981961065,
      "grad_norm": 0.4254562258720398,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 180
    },
    {
      "epoch": 0.01696731559207001,
      "grad_norm": 0.4308949410915375,
      "learning_rate": 2e-05,
      "loss": 0.7623,
      "step": 190
    },
    {
      "epoch": 0.01786033220217896,
      "grad_norm": 0.4657016098499298,
      "learning_rate": 2e-05,
      "loss": 0.741,
      "step": 200
    },
    {
      "epoch": 0.01875334881228791,
      "grad_norm": 0.4654051661491394,
      "learning_rate": 2e-05,
      "loss": 0.7624,
      "step": 210
    },
    {
      "epoch": 0.019646365422396856,
      "grad_norm": 0.4155106544494629,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 220
    },
    {
      "epoch": 0.020539382032505805,
      "grad_norm": 0.47818872332572937,
      "learning_rate": 2e-05,
      "loss": 0.7431,
      "step": 230
    },
    {
      "epoch": 0.02143239864261475,
      "grad_norm": 0.43436357378959656,
      "learning_rate": 2e-05,
      "loss": 0.7559,
      "step": 240
    },
    {
      "epoch": 0.0223254152527237,
      "grad_norm": 0.4438256621360779,
      "learning_rate": 2e-05,
      "loss": 0.7874,
      "step": 250
    },
    {
      "epoch": 0.02321843186283265,
      "grad_norm": 0.42037931084632874,
      "learning_rate": 2e-05,
      "loss": 0.7776,
      "step": 260
    },
    {
      "epoch": 0.024111448472941596,
      "grad_norm": 0.4763539433479309,
      "learning_rate": 2e-05,
      "loss": 0.7769,
      "step": 270
    },
    {
      "epoch": 0.025004465083050546,
      "grad_norm": 0.4746975898742676,
      "learning_rate": 2e-05,
      "loss": 0.7761,
      "step": 280
    },
    {
      "epoch": 0.02589748169315949,
      "grad_norm": 0.4424160122871399,
      "learning_rate": 2e-05,
      "loss": 0.7603,
      "step": 290
    },
    {
      "epoch": 0.02679049830326844,
      "grad_norm": 0.4469321668148041,
      "learning_rate": 2e-05,
      "loss": 0.756,
      "step": 300
    },
    {
      "epoch": 0.02768351491337739,
      "grad_norm": 0.42572203278541565,
      "learning_rate": 2e-05,
      "loss": 0.7527,
      "step": 310
    },
    {
      "epoch": 0.028576531523486336,
      "grad_norm": 0.37948864698410034,
      "learning_rate": 2e-05,
      "loss": 0.6629,
      "step": 320
    },
    {
      "epoch": 0.029469548133595286,
      "grad_norm": 0.4720393717288971,
      "learning_rate": 2e-05,
      "loss": 0.7598,
      "step": 330
    },
    {
      "epoch": 0.030362564743704232,
      "grad_norm": 0.4645335078239441,
      "learning_rate": 2e-05,
      "loss": 0.7635,
      "step": 340
    },
    {
      "epoch": 0.03125558135381318,
      "grad_norm": 0.4660659432411194,
      "learning_rate": 2e-05,
      "loss": 0.77,
      "step": 350
    },
    {
      "epoch": 0.03214859796392213,
      "grad_norm": 0.47986748814582825,
      "learning_rate": 2e-05,
      "loss": 0.7442,
      "step": 360
    },
    {
      "epoch": 0.03304161457403108,
      "grad_norm": 0.4416871964931488,
      "learning_rate": 2e-05,
      "loss": 0.7316,
      "step": 370
    },
    {
      "epoch": 0.03393463118414002,
      "grad_norm": 0.3982086181640625,
      "learning_rate": 2e-05,
      "loss": 0.7805,
      "step": 380
    },
    {
      "epoch": 0.034827647794248975,
      "grad_norm": 0.4491710662841797,
      "learning_rate": 2e-05,
      "loss": 0.743,
      "step": 390
    },
    {
      "epoch": 0.03572066440435792,
      "grad_norm": 0.4009242653846741,
      "learning_rate": 2e-05,
      "loss": 0.7401,
      "step": 400
    },
    {
      "epoch": 0.03661368101446687,
      "grad_norm": 0.3790673017501831,
      "learning_rate": 2e-05,
      "loss": 0.7317,
      "step": 410
    },
    {
      "epoch": 0.03750669762457582,
      "grad_norm": 0.38693296909332275,
      "learning_rate": 2e-05,
      "loss": 0.7373,
      "step": 420
    },
    {
      "epoch": 0.038399714234684766,
      "grad_norm": 0.4592219293117523,
      "learning_rate": 2e-05,
      "loss": 0.7421,
      "step": 430
    },
    {
      "epoch": 0.03929273084479371,
      "grad_norm": 0.48258858919143677,
      "learning_rate": 2e-05,
      "loss": 0.7904,
      "step": 440
    },
    {
      "epoch": 0.04018574745490266,
      "grad_norm": 0.8017710447311401,
      "learning_rate": 2e-05,
      "loss": 0.7501,
      "step": 450
    },
    {
      "epoch": 0.04107876406501161,
      "grad_norm": 0.41467684507369995,
      "learning_rate": 2e-05,
      "loss": 0.7129,
      "step": 460
    },
    {
      "epoch": 0.04197178067512056,
      "grad_norm": 0.490904301404953,
      "learning_rate": 2e-05,
      "loss": 0.7671,
      "step": 470
    },
    {
      "epoch": 0.0428647972852295,
      "grad_norm": 0.4521087110042572,
      "learning_rate": 2e-05,
      "loss": 0.7397,
      "step": 480
    },
    {
      "epoch": 0.043757813895338456,
      "grad_norm": 0.4126375913619995,
      "learning_rate": 2e-05,
      "loss": 0.7391,
      "step": 490
    },
    {
      "epoch": 0.0446508305054474,
      "grad_norm": 0.4476797878742218,
      "learning_rate": 2e-05,
      "loss": 0.7796,
      "step": 500
    },
    {
      "epoch": 0.04554384711555635,
      "grad_norm": 0.43812263011932373,
      "learning_rate": 2e-05,
      "loss": 0.7352,
      "step": 510
    },
    {
      "epoch": 0.0464368637256653,
      "grad_norm": 0.48256680369377136,
      "learning_rate": 2e-05,
      "loss": 0.755,
      "step": 520
    },
    {
      "epoch": 0.047329880335774246,
      "grad_norm": 0.39438191056251526,
      "learning_rate": 2e-05,
      "loss": 0.7438,
      "step": 530
    },
    {
      "epoch": 0.04822289694588319,
      "grad_norm": 0.4146062731742859,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 540
    },
    {
      "epoch": 0.04911591355599214,
      "grad_norm": 0.41129109263420105,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 550
    },
    {
      "epoch": 0.05000893016610109,
      "grad_norm": 0.38930240273475647,
      "learning_rate": 2e-05,
      "loss": 0.7625,
      "step": 560
    },
    {
      "epoch": 0.05090194677621004,
      "grad_norm": 0.48423701524734497,
      "learning_rate": 2e-05,
      "loss": 0.7631,
      "step": 570
    },
    {
      "epoch": 0.05179496338631898,
      "grad_norm": 0.40557992458343506,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 580
    },
    {
      "epoch": 0.052687979996427936,
      "grad_norm": 0.4137013554573059,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 590
    },
    {
      "epoch": 0.05358099660653688,
      "grad_norm": 0.45856934785842896,
      "learning_rate": 2e-05,
      "loss": 0.7721,
      "step": 600
    },
    {
      "epoch": 0.05447401321664583,
      "grad_norm": 0.4169819951057434,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 610
    },
    {
      "epoch": 0.05536702982675478,
      "grad_norm": 0.38590312004089355,
      "learning_rate": 2e-05,
      "loss": 0.7374,
      "step": 620
    },
    {
      "epoch": 0.05626004643686373,
      "grad_norm": 0.40914642810821533,
      "learning_rate": 2e-05,
      "loss": 0.7495,
      "step": 630
    },
    {
      "epoch": 0.05715306304697267,
      "grad_norm": 0.41702547669410706,
      "learning_rate": 2e-05,
      "loss": 0.7398,
      "step": 640
    },
    {
      "epoch": 0.05804607965708162,
      "grad_norm": 0.39613404870033264,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 650
    },
    {
      "epoch": 0.05893909626719057,
      "grad_norm": 0.39417654275894165,
      "learning_rate": 2e-05,
      "loss": 0.7308,
      "step": 660
    },
    {
      "epoch": 0.05983211287729952,
      "grad_norm": 0.3835000693798065,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 670
    },
    {
      "epoch": 0.060725129487408464,
      "grad_norm": 0.40290987491607666,
      "learning_rate": 2e-05,
      "loss": 0.7252,
      "step": 680
    },
    {
      "epoch": 0.061618146097517416,
      "grad_norm": 0.5707727074623108,
      "learning_rate": 2e-05,
      "loss": 0.7349,
      "step": 690
    },
    {
      "epoch": 0.06251116270762636,
      "grad_norm": 0.44035205245018005,
      "learning_rate": 2e-05,
      "loss": 0.7322,
      "step": 700
    },
    {
      "epoch": 0.06340417931773532,
      "grad_norm": 0.4347422420978546,
      "learning_rate": 2e-05,
      "loss": 0.7574,
      "step": 710
    },
    {
      "epoch": 0.06429719592784426,
      "grad_norm": 0.4078957140445709,
      "learning_rate": 2e-05,
      "loss": 0.7621,
      "step": 720
    },
    {
      "epoch": 0.06519021253795321,
      "grad_norm": 0.3920077979564667,
      "learning_rate": 2e-05,
      "loss": 0.7195,
      "step": 730
    },
    {
      "epoch": 0.06608322914806215,
      "grad_norm": 0.38933590054512024,
      "learning_rate": 2e-05,
      "loss": 0.7471,
      "step": 740
    },
    {
      "epoch": 0.0669762457581711,
      "grad_norm": 0.4065713584423065,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 750
    },
    {
      "epoch": 0.06786926236828004,
      "grad_norm": 0.41228511929512024,
      "learning_rate": 2e-05,
      "loss": 0.7547,
      "step": 760
    },
    {
      "epoch": 0.068762278978389,
      "grad_norm": 0.3922688364982605,
      "learning_rate": 2e-05,
      "loss": 0.7148,
      "step": 770
    },
    {
      "epoch": 0.06965529558849795,
      "grad_norm": 0.4028988778591156,
      "learning_rate": 2e-05,
      "loss": 0.7262,
      "step": 780
    },
    {
      "epoch": 0.0705483121986069,
      "grad_norm": 0.44637545943260193,
      "learning_rate": 2e-05,
      "loss": 0.7614,
      "step": 790
    },
    {
      "epoch": 0.07144132880871584,
      "grad_norm": 0.40410199761390686,
      "learning_rate": 2e-05,
      "loss": 0.7524,
      "step": 800
    },
    {
      "epoch": 0.07233434541882479,
      "grad_norm": 0.41571539640426636,
      "learning_rate": 2e-05,
      "loss": 0.7423,
      "step": 810
    },
    {
      "epoch": 0.07322736202893373,
      "grad_norm": 0.48081284761428833,
      "learning_rate": 2e-05,
      "loss": 0.7067,
      "step": 820
    },
    {
      "epoch": 0.07412037863904268,
      "grad_norm": 0.40906158089637756,
      "learning_rate": 2e-05,
      "loss": 0.7576,
      "step": 830
    },
    {
      "epoch": 0.07501339524915164,
      "grad_norm": 0.37821027636528015,
      "learning_rate": 2e-05,
      "loss": 0.7579,
      "step": 840
    },
    {
      "epoch": 0.07590641185926059,
      "grad_norm": 0.3791477382183075,
      "learning_rate": 2e-05,
      "loss": 0.7153,
      "step": 850
    },
    {
      "epoch": 0.07679942846936953,
      "grad_norm": 0.38766372203826904,
      "learning_rate": 2e-05,
      "loss": 0.7393,
      "step": 860
    },
    {
      "epoch": 0.07769244507947848,
      "grad_norm": 0.3937142491340637,
      "learning_rate": 2e-05,
      "loss": 0.7371,
      "step": 870
    },
    {
      "epoch": 0.07858546168958742,
      "grad_norm": 0.3776763677597046,
      "learning_rate": 2e-05,
      "loss": 0.7221,
      "step": 880
    },
    {
      "epoch": 0.07947847829969637,
      "grad_norm": 0.4015044867992401,
      "learning_rate": 2e-05,
      "loss": 0.723,
      "step": 890
    },
    {
      "epoch": 0.08037149490980532,
      "grad_norm": 0.43104034662246704,
      "learning_rate": 2e-05,
      "loss": 0.7621,
      "step": 900
    },
    {
      "epoch": 0.08126451151991428,
      "grad_norm": 0.3779524564743042,
      "learning_rate": 2e-05,
      "loss": 0.7332,
      "step": 910
    },
    {
      "epoch": 0.08215752813002322,
      "grad_norm": 0.3575296998023987,
      "learning_rate": 2e-05,
      "loss": 0.7575,
      "step": 920
    },
    {
      "epoch": 0.08305054474013217,
      "grad_norm": 0.43516188859939575,
      "learning_rate": 2e-05,
      "loss": 0.7601,
      "step": 930
    },
    {
      "epoch": 0.08394356135024111,
      "grad_norm": 0.3753051161766052,
      "learning_rate": 2e-05,
      "loss": 0.7269,
      "step": 940
    },
    {
      "epoch": 0.08483657796035006,
      "grad_norm": 0.3971746861934662,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 950
    },
    {
      "epoch": 0.085729594570459,
      "grad_norm": 0.41950562596321106,
      "learning_rate": 2e-05,
      "loss": 0.7804,
      "step": 960
    },
    {
      "epoch": 0.08662261118056797,
      "grad_norm": 0.3685847222805023,
      "learning_rate": 2e-05,
      "loss": 0.722,
      "step": 970
    },
    {
      "epoch": 0.08751562779067691,
      "grad_norm": 0.3798861801624298,
      "learning_rate": 2e-05,
      "loss": 0.752,
      "step": 980
    },
    {
      "epoch": 0.08840864440078586,
      "grad_norm": 0.40628236532211304,
      "learning_rate": 2e-05,
      "loss": 0.7563,
      "step": 990
    },
    {
      "epoch": 0.0893016610108948,
      "grad_norm": 0.37584495544433594,
      "learning_rate": 2e-05,
      "loss": 0.719,
      "step": 1000
    },
    {
      "epoch": 0.09019467762100375,
      "grad_norm": 0.3755060136318207,
      "learning_rate": 2e-05,
      "loss": 0.7321,
      "step": 1010
    },
    {
      "epoch": 0.0910876942311127,
      "grad_norm": 0.3935806155204773,
      "learning_rate": 2e-05,
      "loss": 0.7296,
      "step": 1020
    },
    {
      "epoch": 0.09198071084122164,
      "grad_norm": 0.4500720798969269,
      "learning_rate": 2e-05,
      "loss": 0.7681,
      "step": 1030
    },
    {
      "epoch": 0.0928737274513306,
      "grad_norm": 0.4131600856781006,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 1040
    },
    {
      "epoch": 0.09376674406143955,
      "grad_norm": 0.38126325607299805,
      "learning_rate": 2e-05,
      "loss": 0.7203,
      "step": 1050
    },
    {
      "epoch": 0.09465976067154849,
      "grad_norm": 0.438177227973938,
      "learning_rate": 2e-05,
      "loss": 0.7332,
      "step": 1060
    },
    {
      "epoch": 0.09555277728165744,
      "grad_norm": 0.3714941143989563,
      "learning_rate": 2e-05,
      "loss": 0.7107,
      "step": 1070
    },
    {
      "epoch": 0.09644579389176638,
      "grad_norm": 0.37563997507095337,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 1080
    },
    {
      "epoch": 0.09733881050187533,
      "grad_norm": 0.38376548886299133,
      "learning_rate": 2e-05,
      "loss": 0.7471,
      "step": 1090
    },
    {
      "epoch": 0.09823182711198428,
      "grad_norm": 0.4249208867549896,
      "learning_rate": 2e-05,
      "loss": 0.7632,
      "step": 1100
    },
    {
      "epoch": 0.09912484372209324,
      "grad_norm": 0.39996418356895447,
      "learning_rate": 2e-05,
      "loss": 0.7302,
      "step": 1110
    },
    {
      "epoch": 0.10001786033220218,
      "grad_norm": 0.4173353612422943,
      "learning_rate": 2e-05,
      "loss": 0.7412,
      "step": 1120
    },
    {
      "epoch": 0.10091087694231113,
      "grad_norm": 0.3551906943321228,
      "learning_rate": 2e-05,
      "loss": 0.7452,
      "step": 1130
    },
    {
      "epoch": 0.10180389355242007,
      "grad_norm": 0.4338817894458771,
      "learning_rate": 2e-05,
      "loss": 0.7549,
      "step": 1140
    },
    {
      "epoch": 0.10269691016252902,
      "grad_norm": 0.49487829208374023,
      "learning_rate": 2e-05,
      "loss": 0.7578,
      "step": 1150
    },
    {
      "epoch": 0.10358992677263797,
      "grad_norm": 0.37487703561782837,
      "learning_rate": 2e-05,
      "loss": 0.7532,
      "step": 1160
    },
    {
      "epoch": 0.10448294338274691,
      "grad_norm": 0.36092835664749146,
      "learning_rate": 2e-05,
      "loss": 0.7386,
      "step": 1170
    },
    {
      "epoch": 0.10537595999285587,
      "grad_norm": 0.3902466297149658,
      "learning_rate": 2e-05,
      "loss": 0.7287,
      "step": 1180
    },
    {
      "epoch": 0.10626897660296482,
      "grad_norm": 0.36106592416763306,
      "learning_rate": 2e-05,
      "loss": 0.7497,
      "step": 1190
    },
    {
      "epoch": 0.10716199321307376,
      "grad_norm": 0.42885327339172363,
      "learning_rate": 2e-05,
      "loss": 0.7801,
      "step": 1200
    },
    {
      "epoch": 0.10805500982318271,
      "grad_norm": 0.36564674973487854,
      "learning_rate": 2e-05,
      "loss": 0.7152,
      "step": 1210
    },
    {
      "epoch": 0.10894802643329166,
      "grad_norm": 0.34475770592689514,
      "learning_rate": 2e-05,
      "loss": 0.7162,
      "step": 1220
    },
    {
      "epoch": 0.1098410430434006,
      "grad_norm": 0.379625529050827,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 1230
    },
    {
      "epoch": 0.11073405965350956,
      "grad_norm": 0.3452484905719757,
      "learning_rate": 2e-05,
      "loss": 0.7218,
      "step": 1240
    },
    {
      "epoch": 0.11162707626361851,
      "grad_norm": 0.4212513566017151,
      "learning_rate": 2e-05,
      "loss": 0.7304,
      "step": 1250
    },
    {
      "epoch": 0.11252009287372745,
      "grad_norm": 0.40886190533638,
      "learning_rate": 2e-05,
      "loss": 0.7373,
      "step": 1260
    },
    {
      "epoch": 0.1134131094838364,
      "grad_norm": 0.3536968529224396,
      "learning_rate": 2e-05,
      "loss": 0.7323,
      "step": 1270
    },
    {
      "epoch": 0.11430612609394535,
      "grad_norm": 0.38670650124549866,
      "learning_rate": 2e-05,
      "loss": 0.741,
      "step": 1280
    },
    {
      "epoch": 0.11519914270405429,
      "grad_norm": 0.42771264910697937,
      "learning_rate": 2e-05,
      "loss": 0.7424,
      "step": 1290
    },
    {
      "epoch": 0.11609215931416324,
      "grad_norm": 0.39867374300956726,
      "learning_rate": 2e-05,
      "loss": 0.7251,
      "step": 1300
    },
    {
      "epoch": 0.1169851759242722,
      "grad_norm": 0.38744813203811646,
      "learning_rate": 2e-05,
      "loss": 0.715,
      "step": 1310
    },
    {
      "epoch": 0.11787819253438114,
      "grad_norm": 0.4313223958015442,
      "learning_rate": 2e-05,
      "loss": 0.6988,
      "step": 1320
    },
    {
      "epoch": 0.11877120914449009,
      "grad_norm": 0.3686710298061371,
      "learning_rate": 2e-05,
      "loss": 0.7162,
      "step": 1330
    },
    {
      "epoch": 0.11966422575459904,
      "grad_norm": 0.8750284314155579,
      "learning_rate": 2e-05,
      "loss": 0.7319,
      "step": 1340
    },
    {
      "epoch": 0.12055724236470798,
      "grad_norm": 0.3907322287559509,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 1350
    },
    {
      "epoch": 0.12145025897481693,
      "grad_norm": 0.355712354183197,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 1360
    },
    {
      "epoch": 0.12234327558492587,
      "grad_norm": 0.3635517954826355,
      "learning_rate": 2e-05,
      "loss": 0.7576,
      "step": 1370
    },
    {
      "epoch": 0.12323629219503483,
      "grad_norm": 0.39474037289619446,
      "learning_rate": 2e-05,
      "loss": 0.6988,
      "step": 1380
    },
    {
      "epoch": 0.12412930880514378,
      "grad_norm": 0.38797375559806824,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 1390
    },
    {
      "epoch": 0.1250223254152527,
      "grad_norm": 0.35904696583747864,
      "learning_rate": 2e-05,
      "loss": 0.7187,
      "step": 1400
    },
    {
      "epoch": 0.12591534202536167,
      "grad_norm": 0.4828942120075226,
      "learning_rate": 2e-05,
      "loss": 0.7326,
      "step": 1410
    },
    {
      "epoch": 0.12680835863547063,
      "grad_norm": 0.3671641945838928,
      "learning_rate": 2e-05,
      "loss": 0.7149,
      "step": 1420
    },
    {
      "epoch": 0.12770137524557956,
      "grad_norm": 0.40876835584640503,
      "learning_rate": 2e-05,
      "loss": 0.7398,
      "step": 1430
    },
    {
      "epoch": 0.12859439185568852,
      "grad_norm": 0.3652034401893616,
      "learning_rate": 2e-05,
      "loss": 0.724,
      "step": 1440
    },
    {
      "epoch": 0.12948740846579745,
      "grad_norm": 0.3712545931339264,
      "learning_rate": 2e-05,
      "loss": 0.7169,
      "step": 1450
    },
    {
      "epoch": 0.13038042507590641,
      "grad_norm": 0.39639317989349365,
      "learning_rate": 2e-05,
      "loss": 0.7623,
      "step": 1460
    },
    {
      "epoch": 0.13127344168601535,
      "grad_norm": 0.38050130009651184,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 1470
    },
    {
      "epoch": 0.1321664582961243,
      "grad_norm": 0.39309820532798767,
      "learning_rate": 2e-05,
      "loss": 0.7601,
      "step": 1480
    },
    {
      "epoch": 0.13305947490623327,
      "grad_norm": 0.3582014739513397,
      "learning_rate": 2e-05,
      "loss": 0.7489,
      "step": 1490
    },
    {
      "epoch": 0.1339524915163422,
      "grad_norm": 0.37961331009864807,
      "learning_rate": 2e-05,
      "loss": 0.7129,
      "step": 1500
    },
    {
      "epoch": 0.13484550812645116,
      "grad_norm": 0.5690667033195496,
      "learning_rate": 2e-05,
      "loss": 0.7345,
      "step": 1510
    },
    {
      "epoch": 0.1357385247365601,
      "grad_norm": 0.3793013095855713,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 1520
    },
    {
      "epoch": 0.13663154134666905,
      "grad_norm": 0.38825497031211853,
      "learning_rate": 2e-05,
      "loss": 0.7198,
      "step": 1530
    },
    {
      "epoch": 0.137524557956778,
      "grad_norm": 0.38483718037605286,
      "learning_rate": 2e-05,
      "loss": 0.716,
      "step": 1540
    },
    {
      "epoch": 0.13841757456688694,
      "grad_norm": 0.3974905014038086,
      "learning_rate": 2e-05,
      "loss": 0.7388,
      "step": 1550
    },
    {
      "epoch": 0.1393105911769959,
      "grad_norm": 0.390345960855484,
      "learning_rate": 2e-05,
      "loss": 0.729,
      "step": 1560
    },
    {
      "epoch": 0.14020360778710483,
      "grad_norm": 0.3711761236190796,
      "learning_rate": 2e-05,
      "loss": 0.7439,
      "step": 1570
    },
    {
      "epoch": 0.1410966243972138,
      "grad_norm": 0.39335206151008606,
      "learning_rate": 2e-05,
      "loss": 0.6857,
      "step": 1580
    },
    {
      "epoch": 0.14198964100732273,
      "grad_norm": 0.431006520986557,
      "learning_rate": 2e-05,
      "loss": 0.6739,
      "step": 1590
    },
    {
      "epoch": 0.14288265761743169,
      "grad_norm": 0.37578749656677246,
      "learning_rate": 2e-05,
      "loss": 0.7253,
      "step": 1600
    },
    {
      "epoch": 0.14377567422754065,
      "grad_norm": 0.36983808875083923,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 1610
    },
    {
      "epoch": 0.14466869083764958,
      "grad_norm": 0.36534303426742554,
      "learning_rate": 2e-05,
      "loss": 0.7048,
      "step": 1620
    },
    {
      "epoch": 0.14556170744775854,
      "grad_norm": 0.3659409284591675,
      "learning_rate": 2e-05,
      "loss": 0.7145,
      "step": 1630
    },
    {
      "epoch": 0.14645472405786747,
      "grad_norm": 0.4026823341846466,
      "learning_rate": 2e-05,
      "loss": 0.7084,
      "step": 1640
    },
    {
      "epoch": 0.14734774066797643,
      "grad_norm": 0.40227675437927246,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 1650
    },
    {
      "epoch": 0.14824075727808536,
      "grad_norm": 0.3589552640914917,
      "learning_rate": 2e-05,
      "loss": 0.6962,
      "step": 1660
    },
    {
      "epoch": 0.14913377388819432,
      "grad_norm": 0.40876489877700806,
      "learning_rate": 2e-05,
      "loss": 0.7581,
      "step": 1670
    },
    {
      "epoch": 0.15002679049830328,
      "grad_norm": 0.3787183463573456,
      "learning_rate": 2e-05,
      "loss": 0.712,
      "step": 1680
    },
    {
      "epoch": 0.1509198071084122,
      "grad_norm": 0.3726568818092346,
      "learning_rate": 2e-05,
      "loss": 0.7329,
      "step": 1690
    },
    {
      "epoch": 0.15181282371852117,
      "grad_norm": 0.36480188369750977,
      "learning_rate": 2e-05,
      "loss": 0.7192,
      "step": 1700
    },
    {
      "epoch": 0.1527058403286301,
      "grad_norm": 0.4598802328109741,
      "learning_rate": 2e-05,
      "loss": 0.742,
      "step": 1710
    },
    {
      "epoch": 0.15359885693873906,
      "grad_norm": 0.34519508481025696,
      "learning_rate": 2e-05,
      "loss": 0.7418,
      "step": 1720
    },
    {
      "epoch": 0.154491873548848,
      "grad_norm": 0.39450502395629883,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 1730
    },
    {
      "epoch": 0.15538489015895696,
      "grad_norm": 0.44403451681137085,
      "learning_rate": 2e-05,
      "loss": 0.7168,
      "step": 1740
    },
    {
      "epoch": 0.15627790676906592,
      "grad_norm": 0.3881482183933258,
      "learning_rate": 2e-05,
      "loss": 0.7322,
      "step": 1750
    },
    {
      "epoch": 0.15717092337917485,
      "grad_norm": 0.3961481750011444,
      "learning_rate": 2e-05,
      "loss": 0.7593,
      "step": 1760
    },
    {
      "epoch": 0.1580639399892838,
      "grad_norm": 0.38921841979026794,
      "learning_rate": 2e-05,
      "loss": 0.7356,
      "step": 1770
    },
    {
      "epoch": 0.15895695659939274,
      "grad_norm": 0.3746980130672455,
      "learning_rate": 2e-05,
      "loss": 0.7109,
      "step": 1780
    },
    {
      "epoch": 0.1598499732095017,
      "grad_norm": 0.36172837018966675,
      "learning_rate": 2e-05,
      "loss": 0.7193,
      "step": 1790
    },
    {
      "epoch": 0.16074298981961063,
      "grad_norm": 0.34081071615219116,
      "learning_rate": 2e-05,
      "loss": 0.734,
      "step": 1800
    },
    {
      "epoch": 0.1616360064297196,
      "grad_norm": 0.3822826147079468,
      "learning_rate": 2e-05,
      "loss": 0.7213,
      "step": 1810
    },
    {
      "epoch": 0.16252902303982855,
      "grad_norm": 0.39521777629852295,
      "learning_rate": 2e-05,
      "loss": 0.7071,
      "step": 1820
    },
    {
      "epoch": 0.16342203964993748,
      "grad_norm": 0.3572978973388672,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 1830
    },
    {
      "epoch": 0.16431505626004644,
      "grad_norm": 0.37090566754341125,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 1840
    },
    {
      "epoch": 0.16520807287015538,
      "grad_norm": 1.1004981994628906,
      "learning_rate": 2e-05,
      "loss": 0.7045,
      "step": 1850
    },
    {
      "epoch": 0.16610108948026434,
      "grad_norm": 0.406376451253891,
      "learning_rate": 2e-05,
      "loss": 0.7121,
      "step": 1860
    },
    {
      "epoch": 0.16699410609037327,
      "grad_norm": 0.410037100315094,
      "learning_rate": 2e-05,
      "loss": 0.6982,
      "step": 1870
    },
    {
      "epoch": 0.16788712270048223,
      "grad_norm": 0.4311705231666565,
      "learning_rate": 2e-05,
      "loss": 0.6981,
      "step": 1880
    },
    {
      "epoch": 0.1687801393105912,
      "grad_norm": 0.4563307464122772,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 1890
    },
    {
      "epoch": 0.16967315592070012,
      "grad_norm": 0.4294278621673584,
      "learning_rate": 2e-05,
      "loss": 0.7177,
      "step": 1900
    },
    {
      "epoch": 0.17056617253080908,
      "grad_norm": 0.41810211539268494,
      "learning_rate": 2e-05,
      "loss": 0.7447,
      "step": 1910
    },
    {
      "epoch": 0.171459189140918,
      "grad_norm": 0.47829675674438477,
      "learning_rate": 2e-05,
      "loss": 0.7345,
      "step": 1920
    },
    {
      "epoch": 0.17235220575102697,
      "grad_norm": 0.40736883878707886,
      "learning_rate": 2e-05,
      "loss": 0.7146,
      "step": 1930
    },
    {
      "epoch": 0.17324522236113593,
      "grad_norm": 0.43190640211105347,
      "learning_rate": 2e-05,
      "loss": 0.7641,
      "step": 1940
    },
    {
      "epoch": 0.17413823897124486,
      "grad_norm": 0.4052496552467346,
      "learning_rate": 2e-05,
      "loss": 0.6961,
      "step": 1950
    },
    {
      "epoch": 0.17503125558135382,
      "grad_norm": 0.3576786518096924,
      "learning_rate": 2e-05,
      "loss": 0.7191,
      "step": 1960
    },
    {
      "epoch": 0.17592427219146275,
      "grad_norm": 0.38320717215538025,
      "learning_rate": 2e-05,
      "loss": 0.7182,
      "step": 1970
    },
    {
      "epoch": 0.17681728880157171,
      "grad_norm": 0.33529797196388245,
      "learning_rate": 2e-05,
      "loss": 0.6998,
      "step": 1980
    },
    {
      "epoch": 0.17771030541168065,
      "grad_norm": 0.4263617992401123,
      "learning_rate": 2e-05,
      "loss": 0.7237,
      "step": 1990
    },
    {
      "epoch": 0.1786033220217896,
      "grad_norm": 0.3683549761772156,
      "learning_rate": 2e-05,
      "loss": 0.7177,
      "step": 2000
    },
    {
      "epoch": 0.17949633863189857,
      "grad_norm": 0.37825068831443787,
      "learning_rate": 2e-05,
      "loss": 0.6936,
      "step": 2010
    },
    {
      "epoch": 0.1803893552420075,
      "grad_norm": 0.38845622539520264,
      "learning_rate": 2e-05,
      "loss": 0.6886,
      "step": 2020
    },
    {
      "epoch": 0.18128237185211646,
      "grad_norm": 0.41564154624938965,
      "learning_rate": 2e-05,
      "loss": 0.6744,
      "step": 2030
    },
    {
      "epoch": 0.1821753884622254,
      "grad_norm": 0.375347763299942,
      "learning_rate": 2e-05,
      "loss": 0.6926,
      "step": 2040
    },
    {
      "epoch": 0.18306840507233435,
      "grad_norm": 0.39925751090049744,
      "learning_rate": 2e-05,
      "loss": 0.7214,
      "step": 2050
    },
    {
      "epoch": 0.18396142168244328,
      "grad_norm": 0.3923713266849518,
      "learning_rate": 2e-05,
      "loss": 0.7478,
      "step": 2060
    },
    {
      "epoch": 0.18485443829255224,
      "grad_norm": 0.33927154541015625,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 2070
    },
    {
      "epoch": 0.1857474549026612,
      "grad_norm": 0.3772149384021759,
      "learning_rate": 2e-05,
      "loss": 0.7127,
      "step": 2080
    },
    {
      "epoch": 0.18664047151277013,
      "grad_norm": 0.4021603763103485,
      "learning_rate": 2e-05,
      "loss": 0.6927,
      "step": 2090
    },
    {
      "epoch": 0.1875334881228791,
      "grad_norm": 0.400551974773407,
      "learning_rate": 2e-05,
      "loss": 0.7256,
      "step": 2100
    },
    {
      "epoch": 0.18842650473298803,
      "grad_norm": 0.41825857758522034,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 2110
    },
    {
      "epoch": 0.18931952134309699,
      "grad_norm": 0.3584500551223755,
      "learning_rate": 2e-05,
      "loss": 0.7141,
      "step": 2120
    },
    {
      "epoch": 0.19021253795320592,
      "grad_norm": 0.8151934146881104,
      "learning_rate": 2e-05,
      "loss": 0.6834,
      "step": 2130
    },
    {
      "epoch": 0.19110555456331488,
      "grad_norm": 0.36020809412002563,
      "learning_rate": 2e-05,
      "loss": 0.7175,
      "step": 2140
    },
    {
      "epoch": 0.19199857117342384,
      "grad_norm": 0.3733537495136261,
      "learning_rate": 2e-05,
      "loss": 0.7156,
      "step": 2150
    },
    {
      "epoch": 0.19289158778353277,
      "grad_norm": 0.4166247844696045,
      "learning_rate": 2e-05,
      "loss": 0.7636,
      "step": 2160
    },
    {
      "epoch": 0.19378460439364173,
      "grad_norm": 0.3874996602535248,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 2170
    },
    {
      "epoch": 0.19467762100375066,
      "grad_norm": 0.3566342294216156,
      "learning_rate": 2e-05,
      "loss": 0.7453,
      "step": 2180
    },
    {
      "epoch": 0.19557063761385962,
      "grad_norm": 0.35978227853775024,
      "learning_rate": 2e-05,
      "loss": 0.7152,
      "step": 2190
    },
    {
      "epoch": 0.19646365422396855,
      "grad_norm": 0.34371668100357056,
      "learning_rate": 2e-05,
      "loss": 0.7258,
      "step": 2200
    },
    {
      "epoch": 0.1973566708340775,
      "grad_norm": 0.36636587977409363,
      "learning_rate": 2e-05,
      "loss": 0.7109,
      "step": 2210
    },
    {
      "epoch": 0.19824968744418647,
      "grad_norm": 0.3694961667060852,
      "learning_rate": 2e-05,
      "loss": 0.7266,
      "step": 2220
    },
    {
      "epoch": 0.1991427040542954,
      "grad_norm": 0.3831530809402466,
      "learning_rate": 2e-05,
      "loss": 0.7165,
      "step": 2230
    },
    {
      "epoch": 0.20003572066440437,
      "grad_norm": 0.4752442538738251,
      "learning_rate": 2e-05,
      "loss": 0.7054,
      "step": 2240
    },
    {
      "epoch": 0.2009287372745133,
      "grad_norm": 0.3605307638645172,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 2250
    },
    {
      "epoch": 0.20182175388462226,
      "grad_norm": 0.3741367757320404,
      "learning_rate": 2e-05,
      "loss": 0.7164,
      "step": 2260
    },
    {
      "epoch": 0.2027147704947312,
      "grad_norm": 0.44306308031082153,
      "learning_rate": 2e-05,
      "loss": 0.7054,
      "step": 2270
    },
    {
      "epoch": 0.20360778710484015,
      "grad_norm": 0.35885170102119446,
      "learning_rate": 2e-05,
      "loss": 0.7193,
      "step": 2280
    },
    {
      "epoch": 0.2045008037149491,
      "grad_norm": 0.42225345969200134,
      "learning_rate": 2e-05,
      "loss": 0.7167,
      "step": 2290
    },
    {
      "epoch": 0.20539382032505804,
      "grad_norm": 0.36834093928337097,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 2300
    },
    {
      "epoch": 0.206286836935167,
      "grad_norm": 0.37331831455230713,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 2310
    },
    {
      "epoch": 0.20717985354527593,
      "grad_norm": 0.33582958579063416,
      "learning_rate": 2e-05,
      "loss": 0.7447,
      "step": 2320
    },
    {
      "epoch": 0.2080728701553849,
      "grad_norm": 0.3406867980957031,
      "learning_rate": 2e-05,
      "loss": 0.6888,
      "step": 2330
    },
    {
      "epoch": 0.20896588676549382,
      "grad_norm": 0.7537785768508911,
      "learning_rate": 2e-05,
      "loss": 0.7387,
      "step": 2340
    },
    {
      "epoch": 0.20985890337560278,
      "grad_norm": 0.3987394869327545,
      "learning_rate": 2e-05,
      "loss": 0.6995,
      "step": 2350
    },
    {
      "epoch": 0.21075191998571174,
      "grad_norm": 0.3712831437587738,
      "learning_rate": 2e-05,
      "loss": 0.6763,
      "step": 2360
    },
    {
      "epoch": 0.21164493659582068,
      "grad_norm": 0.3862318992614746,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 2370
    },
    {
      "epoch": 0.21253795320592964,
      "grad_norm": 0.35739755630493164,
      "learning_rate": 2e-05,
      "loss": 0.7143,
      "step": 2380
    },
    {
      "epoch": 0.21343096981603857,
      "grad_norm": 0.37976863980293274,
      "learning_rate": 2e-05,
      "loss": 0.721,
      "step": 2390
    },
    {
      "epoch": 0.21432398642614753,
      "grad_norm": 0.3507783114910126,
      "learning_rate": 2e-05,
      "loss": 0.7008,
      "step": 2400
    },
    {
      "epoch": 0.2152170030362565,
      "grad_norm": 0.37717995047569275,
      "learning_rate": 2e-05,
      "loss": 0.7123,
      "step": 2410
    },
    {
      "epoch": 0.21611001964636542,
      "grad_norm": 0.5691996216773987,
      "learning_rate": 2e-05,
      "loss": 0.7418,
      "step": 2420
    },
    {
      "epoch": 0.21700303625647438,
      "grad_norm": 0.38756728172302246,
      "learning_rate": 2e-05,
      "loss": 0.7104,
      "step": 2430
    },
    {
      "epoch": 0.2178960528665833,
      "grad_norm": 0.554707407951355,
      "learning_rate": 2e-05,
      "loss": 0.68,
      "step": 2440
    },
    {
      "epoch": 0.21878906947669227,
      "grad_norm": 0.3568492829799652,
      "learning_rate": 2e-05,
      "loss": 0.6845,
      "step": 2450
    },
    {
      "epoch": 0.2196820860868012,
      "grad_norm": 0.3751198947429657,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 2460
    },
    {
      "epoch": 0.22057510269691016,
      "grad_norm": 0.39954832196235657,
      "learning_rate": 2e-05,
      "loss": 0.7364,
      "step": 2470
    },
    {
      "epoch": 0.22146811930701912,
      "grad_norm": 0.37244030833244324,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 2480
    },
    {
      "epoch": 0.22236113591712806,
      "grad_norm": 0.38988977670669556,
      "learning_rate": 2e-05,
      "loss": 0.7324,
      "step": 2490
    },
    {
      "epoch": 0.22325415252723702,
      "grad_norm": 0.3926638662815094,
      "learning_rate": 2e-05,
      "loss": 0.7181,
      "step": 2500
    },
    {
      "epoch": 0.22414716913734595,
      "grad_norm": 0.361802339553833,
      "learning_rate": 2e-05,
      "loss": 0.6984,
      "step": 2510
    },
    {
      "epoch": 0.2250401857474549,
      "grad_norm": 0.38118746876716614,
      "learning_rate": 2e-05,
      "loss": 0.7376,
      "step": 2520
    },
    {
      "epoch": 0.22593320235756384,
      "grad_norm": 0.39917808771133423,
      "learning_rate": 2e-05,
      "loss": 0.7166,
      "step": 2530
    },
    {
      "epoch": 0.2268262189676728,
      "grad_norm": 0.39203181862831116,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 2540
    },
    {
      "epoch": 0.22771923557778176,
      "grad_norm": 0.82660973072052,
      "learning_rate": 2e-05,
      "loss": 0.6914,
      "step": 2550
    },
    {
      "epoch": 0.2286122521878907,
      "grad_norm": 0.3741285502910614,
      "learning_rate": 2e-05,
      "loss": 0.7014,
      "step": 2560
    },
    {
      "epoch": 0.22950526879799965,
      "grad_norm": 0.3964354395866394,
      "learning_rate": 2e-05,
      "loss": 0.6982,
      "step": 2570
    },
    {
      "epoch": 0.23039828540810858,
      "grad_norm": 0.44241467118263245,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 2580
    },
    {
      "epoch": 0.23129130201821754,
      "grad_norm": 0.36581575870513916,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 2590
    },
    {
      "epoch": 0.23218431862832647,
      "grad_norm": 0.3831479549407959,
      "learning_rate": 2e-05,
      "loss": 0.7266,
      "step": 2600
    },
    {
      "epoch": 0.23307733523843543,
      "grad_norm": 0.36546680331230164,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 2610
    },
    {
      "epoch": 0.2339703518485444,
      "grad_norm": 0.3919263780117035,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 2620
    },
    {
      "epoch": 0.23486336845865333,
      "grad_norm": 0.37016761302948,
      "learning_rate": 2e-05,
      "loss": 0.7009,
      "step": 2630
    },
    {
      "epoch": 0.2357563850687623,
      "grad_norm": 0.37380245327949524,
      "learning_rate": 2e-05,
      "loss": 0.7025,
      "step": 2640
    },
    {
      "epoch": 0.23664940167887122,
      "grad_norm": 0.647921085357666,
      "learning_rate": 2e-05,
      "loss": 0.709,
      "step": 2650
    },
    {
      "epoch": 0.23754241828898018,
      "grad_norm": 0.3720865845680237,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 2660
    },
    {
      "epoch": 0.2384354348990891,
      "grad_norm": 0.38328051567077637,
      "learning_rate": 2e-05,
      "loss": 0.7039,
      "step": 2670
    },
    {
      "epoch": 0.23932845150919807,
      "grad_norm": 0.46826261281967163,
      "learning_rate": 2e-05,
      "loss": 0.7047,
      "step": 2680
    },
    {
      "epoch": 0.24022146811930703,
      "grad_norm": 0.37531983852386475,
      "learning_rate": 2e-05,
      "loss": 0.7185,
      "step": 2690
    },
    {
      "epoch": 0.24111448472941596,
      "grad_norm": 0.3817111551761627,
      "learning_rate": 2e-05,
      "loss": 0.7202,
      "step": 2700
    },
    {
      "epoch": 0.24200750133952492,
      "grad_norm": 0.4027620851993561,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 2710
    },
    {
      "epoch": 0.24290051794963385,
      "grad_norm": 0.3872319459915161,
      "learning_rate": 2e-05,
      "loss": 0.7163,
      "step": 2720
    },
    {
      "epoch": 0.2437935345597428,
      "grad_norm": 0.37678542733192444,
      "learning_rate": 2e-05,
      "loss": 0.6815,
      "step": 2730
    },
    {
      "epoch": 0.24468655116985175,
      "grad_norm": 0.35330837965011597,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 2740
    },
    {
      "epoch": 0.2455795677799607,
      "grad_norm": 0.3747386634349823,
      "learning_rate": 2e-05,
      "loss": 0.6878,
      "step": 2750
    },
    {
      "epoch": 0.24647258439006967,
      "grad_norm": 0.41168853640556335,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 2760
    },
    {
      "epoch": 0.2473656010001786,
      "grad_norm": 0.32120653986930847,
      "learning_rate": 2e-05,
      "loss": 0.6855,
      "step": 2770
    },
    {
      "epoch": 0.24825861761028756,
      "grad_norm": 0.3390073776245117,
      "learning_rate": 2e-05,
      "loss": 0.7101,
      "step": 2780
    },
    {
      "epoch": 0.2491516342203965,
      "grad_norm": 0.37074482440948486,
      "learning_rate": 2e-05,
      "loss": 0.6986,
      "step": 2790
    },
    {
      "epoch": 0.2500446508305054,
      "grad_norm": 0.3548227548599243,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 2800
    },
    {
      "epoch": 0.2509376674406144,
      "grad_norm": 0.3685583174228668,
      "learning_rate": 2e-05,
      "loss": 0.7048,
      "step": 2810
    },
    {
      "epoch": 0.25183068405072334,
      "grad_norm": 0.3529677391052246,
      "learning_rate": 2e-05,
      "loss": 0.7216,
      "step": 2820
    },
    {
      "epoch": 0.2527237006608323,
      "grad_norm": 0.3592994809150696,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 2830
    },
    {
      "epoch": 0.25361671727094126,
      "grad_norm": 0.36516258120536804,
      "learning_rate": 2e-05,
      "loss": 0.701,
      "step": 2840
    },
    {
      "epoch": 0.2545097338810502,
      "grad_norm": 0.3898088037967682,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 2850
    },
    {
      "epoch": 0.2554027504911591,
      "grad_norm": 0.34210461378097534,
      "learning_rate": 2e-05,
      "loss": 0.6915,
      "step": 2860
    },
    {
      "epoch": 0.25629576710126806,
      "grad_norm": 0.34576666355133057,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 2870
    },
    {
      "epoch": 0.25718878371137704,
      "grad_norm": 0.3502766788005829,
      "learning_rate": 2e-05,
      "loss": 0.7146,
      "step": 2880
    },
    {
      "epoch": 0.258081800321486,
      "grad_norm": 0.37955331802368164,
      "learning_rate": 2e-05,
      "loss": 0.7165,
      "step": 2890
    },
    {
      "epoch": 0.2589748169315949,
      "grad_norm": 0.3612077832221985,
      "learning_rate": 2e-05,
      "loss": 0.7198,
      "step": 2900
    },
    {
      "epoch": 0.2598678335417039,
      "grad_norm": 0.3505285084247589,
      "learning_rate": 2e-05,
      "loss": 0.7248,
      "step": 2910
    },
    {
      "epoch": 0.26076085015181283,
      "grad_norm": 0.3673901855945587,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 2920
    },
    {
      "epoch": 0.26165386676192176,
      "grad_norm": 0.37945684790611267,
      "learning_rate": 2e-05,
      "loss": 0.6932,
      "step": 2930
    },
    {
      "epoch": 0.2625468833720307,
      "grad_norm": 0.3634501099586487,
      "learning_rate": 2e-05,
      "loss": 0.6931,
      "step": 2940
    },
    {
      "epoch": 0.2634398999821397,
      "grad_norm": 0.3780888020992279,
      "learning_rate": 2e-05,
      "loss": 0.7335,
      "step": 2950
    },
    {
      "epoch": 0.2643329165922486,
      "grad_norm": 0.3801252245903015,
      "learning_rate": 2e-05,
      "loss": 0.6956,
      "step": 2960
    },
    {
      "epoch": 0.26522593320235754,
      "grad_norm": 0.36999383568763733,
      "learning_rate": 2e-05,
      "loss": 0.7009,
      "step": 2970
    },
    {
      "epoch": 0.26611894981246653,
      "grad_norm": 0.34486889839172363,
      "learning_rate": 2e-05,
      "loss": 0.6622,
      "step": 2980
    },
    {
      "epoch": 0.26701196642257546,
      "grad_norm": 0.39070191979408264,
      "learning_rate": 2e-05,
      "loss": 0.6658,
      "step": 2990
    },
    {
      "epoch": 0.2679049830326844,
      "grad_norm": 0.37735554575920105,
      "learning_rate": 2e-05,
      "loss": 0.7351,
      "step": 3000
    },
    {
      "epoch": 0.26879799964279333,
      "grad_norm": 0.45589423179626465,
      "learning_rate": 2e-05,
      "loss": 0.7251,
      "step": 3010
    },
    {
      "epoch": 0.2696910162529023,
      "grad_norm": 0.36697840690612793,
      "learning_rate": 2e-05,
      "loss": 0.7083,
      "step": 3020
    },
    {
      "epoch": 0.27058403286301125,
      "grad_norm": 0.6844993233680725,
      "learning_rate": 2e-05,
      "loss": 0.7378,
      "step": 3030
    },
    {
      "epoch": 0.2714770494731202,
      "grad_norm": 0.36845651268959045,
      "learning_rate": 2e-05,
      "loss": 0.7181,
      "step": 3040
    },
    {
      "epoch": 0.27237006608322917,
      "grad_norm": 0.3911682069301605,
      "learning_rate": 2e-05,
      "loss": 0.7009,
      "step": 3050
    },
    {
      "epoch": 0.2732630826933381,
      "grad_norm": 0.3813776671886444,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 3060
    },
    {
      "epoch": 0.27415609930344703,
      "grad_norm": 0.39326590299606323,
      "learning_rate": 2e-05,
      "loss": 0.7043,
      "step": 3070
    },
    {
      "epoch": 0.275049115913556,
      "grad_norm": 0.3827492594718933,
      "learning_rate": 2e-05,
      "loss": 0.7136,
      "step": 3080
    },
    {
      "epoch": 0.27594213252366495,
      "grad_norm": 0.3630823493003845,
      "learning_rate": 2e-05,
      "loss": 0.6918,
      "step": 3090
    },
    {
      "epoch": 0.2768351491337739,
      "grad_norm": 0.3365762233734131,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 3100
    },
    {
      "epoch": 0.2777281657438828,
      "grad_norm": 0.39159443974494934,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 3110
    },
    {
      "epoch": 0.2786211823539918,
      "grad_norm": 0.3930347263813019,
      "learning_rate": 2e-05,
      "loss": 0.7083,
      "step": 3120
    },
    {
      "epoch": 0.27951419896410074,
      "grad_norm": 0.35313093662261963,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 3130
    },
    {
      "epoch": 0.28040721557420967,
      "grad_norm": 0.33984461426734924,
      "learning_rate": 2e-05,
      "loss": 0.6683,
      "step": 3140
    },
    {
      "epoch": 0.28130023218431865,
      "grad_norm": 0.40427762269973755,
      "learning_rate": 2e-05,
      "loss": 0.7342,
      "step": 3150
    },
    {
      "epoch": 0.2821932487944276,
      "grad_norm": 0.3635815978050232,
      "learning_rate": 2e-05,
      "loss": 0.6897,
      "step": 3160
    },
    {
      "epoch": 0.2830862654045365,
      "grad_norm": 0.41092535853385925,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 3170
    },
    {
      "epoch": 0.28397928201464545,
      "grad_norm": 0.3885297179222107,
      "learning_rate": 2e-05,
      "loss": 0.6929,
      "step": 3180
    },
    {
      "epoch": 0.28487229862475444,
      "grad_norm": 0.3909347653388977,
      "learning_rate": 2e-05,
      "loss": 0.6872,
      "step": 3190
    },
    {
      "epoch": 0.28576531523486337,
      "grad_norm": 0.34606102108955383,
      "learning_rate": 2e-05,
      "loss": 0.673,
      "step": 3200
    },
    {
      "epoch": 0.2866583318449723,
      "grad_norm": 0.3511039614677429,
      "learning_rate": 2e-05,
      "loss": 0.7187,
      "step": 3210
    },
    {
      "epoch": 0.2875513484550813,
      "grad_norm": 0.48329827189445496,
      "learning_rate": 2e-05,
      "loss": 0.7078,
      "step": 3220
    },
    {
      "epoch": 0.2884443650651902,
      "grad_norm": 0.36455851793289185,
      "learning_rate": 2e-05,
      "loss": 0.6885,
      "step": 3230
    },
    {
      "epoch": 0.28933738167529915,
      "grad_norm": 0.36680030822753906,
      "learning_rate": 2e-05,
      "loss": 0.6744,
      "step": 3240
    },
    {
      "epoch": 0.2902303982854081,
      "grad_norm": 0.39824068546295166,
      "learning_rate": 2e-05,
      "loss": 0.7061,
      "step": 3250
    },
    {
      "epoch": 0.2911234148955171,
      "grad_norm": 0.3565004765987396,
      "learning_rate": 2e-05,
      "loss": 0.6894,
      "step": 3260
    },
    {
      "epoch": 0.292016431505626,
      "grad_norm": 0.34405532479286194,
      "learning_rate": 2e-05,
      "loss": 0.6807,
      "step": 3270
    },
    {
      "epoch": 0.29290944811573494,
      "grad_norm": 0.3882642090320587,
      "learning_rate": 2e-05,
      "loss": 0.696,
      "step": 3280
    },
    {
      "epoch": 0.2938024647258439,
      "grad_norm": 0.3908495008945465,
      "learning_rate": 2e-05,
      "loss": 0.6932,
      "step": 3290
    },
    {
      "epoch": 0.29469548133595286,
      "grad_norm": 0.39699220657348633,
      "learning_rate": 2e-05,
      "loss": 0.7252,
      "step": 3300
    },
    {
      "epoch": 0.2955884979460618,
      "grad_norm": 0.37945181131362915,
      "learning_rate": 2e-05,
      "loss": 0.7179,
      "step": 3310
    },
    {
      "epoch": 0.2964815145561707,
      "grad_norm": 0.3482162058353424,
      "learning_rate": 2e-05,
      "loss": 0.6994,
      "step": 3320
    },
    {
      "epoch": 0.2973745311662797,
      "grad_norm": 0.3984408378601074,
      "learning_rate": 2e-05,
      "loss": 0.7088,
      "step": 3330
    },
    {
      "epoch": 0.29826754777638864,
      "grad_norm": 0.3412688076496124,
      "learning_rate": 2e-05,
      "loss": 0.6745,
      "step": 3340
    },
    {
      "epoch": 0.2991605643864976,
      "grad_norm": 0.34985795617103577,
      "learning_rate": 2e-05,
      "loss": 0.7055,
      "step": 3350
    },
    {
      "epoch": 0.30005358099660656,
      "grad_norm": 0.3310694694519043,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 3360
    },
    {
      "epoch": 0.3009465976067155,
      "grad_norm": 0.34946319460868835,
      "learning_rate": 2e-05,
      "loss": 0.6766,
      "step": 3370
    },
    {
      "epoch": 0.3018396142168244,
      "grad_norm": 0.36742743849754333,
      "learning_rate": 2e-05,
      "loss": 0.6916,
      "step": 3380
    },
    {
      "epoch": 0.30273263082693336,
      "grad_norm": 0.3899418115615845,
      "learning_rate": 2e-05,
      "loss": 0.7216,
      "step": 3390
    },
    {
      "epoch": 0.30362564743704235,
      "grad_norm": 0.3610823154449463,
      "learning_rate": 2e-05,
      "loss": 0.7164,
      "step": 3400
    },
    {
      "epoch": 0.3045186640471513,
      "grad_norm": 0.36853116750717163,
      "learning_rate": 2e-05,
      "loss": 0.6832,
      "step": 3410
    },
    {
      "epoch": 0.3054116806572602,
      "grad_norm": 0.3777369260787964,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 3420
    },
    {
      "epoch": 0.3063046972673692,
      "grad_norm": 0.37218400835990906,
      "learning_rate": 2e-05,
      "loss": 0.6983,
      "step": 3430
    },
    {
      "epoch": 0.30719771387747813,
      "grad_norm": 0.39339882135391235,
      "learning_rate": 2e-05,
      "loss": 0.7022,
      "step": 3440
    },
    {
      "epoch": 0.30809073048758706,
      "grad_norm": 0.36650851368904114,
      "learning_rate": 2e-05,
      "loss": 0.6761,
      "step": 3450
    },
    {
      "epoch": 0.308983747097696,
      "grad_norm": 0.3942379951477051,
      "learning_rate": 2e-05,
      "loss": 0.6723,
      "step": 3460
    },
    {
      "epoch": 0.309876763707805,
      "grad_norm": 0.3716135621070862,
      "learning_rate": 2e-05,
      "loss": 0.7155,
      "step": 3470
    },
    {
      "epoch": 0.3107697803179139,
      "grad_norm": 0.3720799684524536,
      "learning_rate": 2e-05,
      "loss": 0.6939,
      "step": 3480
    },
    {
      "epoch": 0.31166279692802284,
      "grad_norm": 0.35542234778404236,
      "learning_rate": 2e-05,
      "loss": 0.6962,
      "step": 3490
    },
    {
      "epoch": 0.31255581353813183,
      "grad_norm": 0.41491395235061646,
      "learning_rate": 2e-05,
      "loss": 0.7012,
      "step": 3500
    },
    {
      "epoch": 0.31344883014824076,
      "grad_norm": 0.35415154695510864,
      "learning_rate": 2e-05,
      "loss": 0.6697,
      "step": 3510
    },
    {
      "epoch": 0.3143418467583497,
      "grad_norm": 0.4248461425304413,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 3520
    },
    {
      "epoch": 0.31523486336845863,
      "grad_norm": 0.42884528636932373,
      "learning_rate": 2e-05,
      "loss": 0.7165,
      "step": 3530
    },
    {
      "epoch": 0.3161278799785676,
      "grad_norm": 0.4538537263870239,
      "learning_rate": 2e-05,
      "loss": 0.7215,
      "step": 3540
    },
    {
      "epoch": 0.31702089658867655,
      "grad_norm": 0.39300262928009033,
      "learning_rate": 2e-05,
      "loss": 0.6708,
      "step": 3550
    },
    {
      "epoch": 0.3179139131987855,
      "grad_norm": 0.3674524426460266,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 3560
    },
    {
      "epoch": 0.31880692980889447,
      "grad_norm": 0.3366934359073639,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 3570
    },
    {
      "epoch": 0.3196999464190034,
      "grad_norm": 0.46228548884391785,
      "learning_rate": 2e-05,
      "loss": 0.6933,
      "step": 3580
    },
    {
      "epoch": 0.32059296302911233,
      "grad_norm": 0.3779584765434265,
      "learning_rate": 2e-05,
      "loss": 0.668,
      "step": 3590
    },
    {
      "epoch": 0.32148597963922126,
      "grad_norm": 0.3647182583808899,
      "learning_rate": 2e-05,
      "loss": 0.6798,
      "step": 3600
    },
    {
      "epoch": 0.32237899624933025,
      "grad_norm": 0.35239359736442566,
      "learning_rate": 2e-05,
      "loss": 0.6978,
      "step": 3610
    },
    {
      "epoch": 0.3232720128594392,
      "grad_norm": 0.34447768330574036,
      "learning_rate": 2e-05,
      "loss": 0.6887,
      "step": 3620
    },
    {
      "epoch": 0.3241650294695481,
      "grad_norm": 0.5083383321762085,
      "learning_rate": 2e-05,
      "loss": 0.7006,
      "step": 3630
    },
    {
      "epoch": 0.3250580460796571,
      "grad_norm": 0.37443387508392334,
      "learning_rate": 2e-05,
      "loss": 0.683,
      "step": 3640
    },
    {
      "epoch": 0.32595106268976604,
      "grad_norm": 0.36775442957878113,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 3650
    },
    {
      "epoch": 0.32684407929987497,
      "grad_norm": 0.43107321858406067,
      "learning_rate": 2e-05,
      "loss": 0.6962,
      "step": 3660
    },
    {
      "epoch": 0.3277370959099839,
      "grad_norm": 0.3437502682209015,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 3670
    },
    {
      "epoch": 0.3286301125200929,
      "grad_norm": 0.3835112452507019,
      "learning_rate": 2e-05,
      "loss": 0.679,
      "step": 3680
    },
    {
      "epoch": 0.3295231291302018,
      "grad_norm": 0.3627762794494629,
      "learning_rate": 2e-05,
      "loss": 0.6925,
      "step": 3690
    },
    {
      "epoch": 0.33041614574031075,
      "grad_norm": 0.36223524808883667,
      "learning_rate": 2e-05,
      "loss": 0.707,
      "step": 3700
    },
    {
      "epoch": 0.33130916235041974,
      "grad_norm": 0.380140095949173,
      "learning_rate": 2e-05,
      "loss": 0.6735,
      "step": 3710
    },
    {
      "epoch": 0.33220217896052867,
      "grad_norm": 0.3285742998123169,
      "learning_rate": 2e-05,
      "loss": 0.6674,
      "step": 3720
    },
    {
      "epoch": 0.3330951955706376,
      "grad_norm": 0.39423131942749023,
      "learning_rate": 2e-05,
      "loss": 0.7027,
      "step": 3730
    },
    {
      "epoch": 0.33398821218074654,
      "grad_norm": 0.3929256498813629,
      "learning_rate": 2e-05,
      "loss": 0.7024,
      "step": 3740
    },
    {
      "epoch": 0.3348812287908555,
      "grad_norm": 0.3465358018875122,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 3750
    },
    {
      "epoch": 0.33577424540096446,
      "grad_norm": 0.4118633568286896,
      "learning_rate": 2e-05,
      "loss": 0.7036,
      "step": 3760
    },
    {
      "epoch": 0.3366672620110734,
      "grad_norm": 0.3681686818599701,
      "learning_rate": 2e-05,
      "loss": 0.7157,
      "step": 3770
    },
    {
      "epoch": 0.3375602786211824,
      "grad_norm": 0.4659908413887024,
      "learning_rate": 2e-05,
      "loss": 0.6902,
      "step": 3780
    },
    {
      "epoch": 0.3384532952312913,
      "grad_norm": 0.35910534858703613,
      "learning_rate": 2e-05,
      "loss": 0.7047,
      "step": 3790
    },
    {
      "epoch": 0.33934631184140024,
      "grad_norm": 0.38420364260673523,
      "learning_rate": 2e-05,
      "loss": 0.6894,
      "step": 3800
    },
    {
      "epoch": 0.34023932845150917,
      "grad_norm": 0.4698358178138733,
      "learning_rate": 2e-05,
      "loss": 0.6686,
      "step": 3810
    },
    {
      "epoch": 0.34113234506161816,
      "grad_norm": 0.33641931414604187,
      "learning_rate": 2e-05,
      "loss": 0.7071,
      "step": 3820
    },
    {
      "epoch": 0.3420253616717271,
      "grad_norm": 0.39272478222846985,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 3830
    },
    {
      "epoch": 0.342918378281836,
      "grad_norm": 0.36490029096603394,
      "learning_rate": 2e-05,
      "loss": 0.6921,
      "step": 3840
    },
    {
      "epoch": 0.343811394891945,
      "grad_norm": 0.34400805830955505,
      "learning_rate": 2e-05,
      "loss": 0.6763,
      "step": 3850
    },
    {
      "epoch": 0.34470441150205394,
      "grad_norm": 0.3541272282600403,
      "learning_rate": 2e-05,
      "loss": 0.7135,
      "step": 3860
    },
    {
      "epoch": 0.3455974281121629,
      "grad_norm": 0.34691932797431946,
      "learning_rate": 2e-05,
      "loss": 0.6809,
      "step": 3870
    },
    {
      "epoch": 0.34649044472227186,
      "grad_norm": 0.3615210950374603,
      "learning_rate": 2e-05,
      "loss": 0.6811,
      "step": 3880
    },
    {
      "epoch": 0.3473834613323808,
      "grad_norm": 0.37400543689727783,
      "learning_rate": 2e-05,
      "loss": 0.6974,
      "step": 3890
    },
    {
      "epoch": 0.3482764779424897,
      "grad_norm": 0.41673043370246887,
      "learning_rate": 2e-05,
      "loss": 0.7509,
      "step": 3900
    },
    {
      "epoch": 0.34916949455259866,
      "grad_norm": 0.4203324615955353,
      "learning_rate": 2e-05,
      "loss": 0.7048,
      "step": 3910
    },
    {
      "epoch": 0.35006251116270765,
      "grad_norm": 0.3847067058086395,
      "learning_rate": 2e-05,
      "loss": 0.6853,
      "step": 3920
    },
    {
      "epoch": 0.3509555277728166,
      "grad_norm": 0.3331018388271332,
      "learning_rate": 2e-05,
      "loss": 0.6173,
      "step": 3930
    },
    {
      "epoch": 0.3518485443829255,
      "grad_norm": 0.38789236545562744,
      "learning_rate": 2e-05,
      "loss": 0.6686,
      "step": 3940
    },
    {
      "epoch": 0.3527415609930345,
      "grad_norm": 0.34841862320899963,
      "learning_rate": 2e-05,
      "loss": 0.6713,
      "step": 3950
    },
    {
      "epoch": 0.35363457760314343,
      "grad_norm": 0.36600327491760254,
      "learning_rate": 2e-05,
      "loss": 0.6757,
      "step": 3960
    },
    {
      "epoch": 0.35452759421325236,
      "grad_norm": 0.35073408484458923,
      "learning_rate": 2e-05,
      "loss": 0.7055,
      "step": 3970
    },
    {
      "epoch": 0.3554206108233613,
      "grad_norm": 0.33693766593933105,
      "learning_rate": 2e-05,
      "loss": 0.6924,
      "step": 3980
    },
    {
      "epoch": 0.3563136274334703,
      "grad_norm": 0.3566618263721466,
      "learning_rate": 2e-05,
      "loss": 0.6803,
      "step": 3990
    },
    {
      "epoch": 0.3572066440435792,
      "grad_norm": 0.3572796881198883,
      "learning_rate": 2e-05,
      "loss": 0.6805,
      "step": 4000
    },
    {
      "epoch": 0.35809966065368815,
      "grad_norm": 0.40160486102104187,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 4010
    },
    {
      "epoch": 0.35899267726379713,
      "grad_norm": 0.36571747064590454,
      "learning_rate": 2e-05,
      "loss": 0.6738,
      "step": 4020
    },
    {
      "epoch": 0.35988569387390607,
      "grad_norm": 0.34542420506477356,
      "learning_rate": 2e-05,
      "loss": 0.6895,
      "step": 4030
    },
    {
      "epoch": 0.360778710484015,
      "grad_norm": 0.3404637575149536,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 4040
    },
    {
      "epoch": 0.36167172709412393,
      "grad_norm": 0.4743908643722534,
      "learning_rate": 2e-05,
      "loss": 0.6849,
      "step": 4050
    },
    {
      "epoch": 0.3625647437042329,
      "grad_norm": 0.3539956510066986,
      "learning_rate": 2e-05,
      "loss": 0.6772,
      "step": 4060
    },
    {
      "epoch": 0.36345776031434185,
      "grad_norm": 0.35375577211380005,
      "learning_rate": 2e-05,
      "loss": 0.6803,
      "step": 4070
    },
    {
      "epoch": 0.3643507769244508,
      "grad_norm": 0.37531569600105286,
      "learning_rate": 2e-05,
      "loss": 0.6856,
      "step": 4080
    },
    {
      "epoch": 0.36524379353455977,
      "grad_norm": 0.3987765312194824,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 4090
    },
    {
      "epoch": 0.3661368101446687,
      "grad_norm": 0.3849702775478363,
      "learning_rate": 2e-05,
      "loss": 0.7117,
      "step": 4100
    },
    {
      "epoch": 0.36702982675477763,
      "grad_norm": 0.364816278219223,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 4110
    },
    {
      "epoch": 0.36792284336488656,
      "grad_norm": 0.372311532497406,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 4120
    },
    {
      "epoch": 0.36881585997499555,
      "grad_norm": 0.33558011054992676,
      "learning_rate": 2e-05,
      "loss": 0.6663,
      "step": 4130
    },
    {
      "epoch": 0.3697088765851045,
      "grad_norm": 0.36152487993240356,
      "learning_rate": 2e-05,
      "loss": 0.7104,
      "step": 4140
    },
    {
      "epoch": 0.3706018931952134,
      "grad_norm": 0.3753606379032135,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 4150
    },
    {
      "epoch": 0.3714949098053224,
      "grad_norm": 0.36609184741973877,
      "learning_rate": 2e-05,
      "loss": 0.6831,
      "step": 4160
    },
    {
      "epoch": 0.37238792641543134,
      "grad_norm": 0.34632691740989685,
      "learning_rate": 2e-05,
      "loss": 0.6892,
      "step": 4170
    },
    {
      "epoch": 0.37328094302554027,
      "grad_norm": 0.3701765239238739,
      "learning_rate": 2e-05,
      "loss": 0.6784,
      "step": 4180
    },
    {
      "epoch": 0.3741739596356492,
      "grad_norm": 0.40695920586586,
      "learning_rate": 2e-05,
      "loss": 0.668,
      "step": 4190
    },
    {
      "epoch": 0.3750669762457582,
      "grad_norm": 0.3674159348011017,
      "learning_rate": 2e-05,
      "loss": 0.6903,
      "step": 4200
    },
    {
      "epoch": 0.3759599928558671,
      "grad_norm": 0.32752901315689087,
      "learning_rate": 2e-05,
      "loss": 0.6563,
      "step": 4210
    },
    {
      "epoch": 0.37685300946597605,
      "grad_norm": 0.3936874568462372,
      "learning_rate": 2e-05,
      "loss": 0.6835,
      "step": 4220
    },
    {
      "epoch": 0.37774602607608504,
      "grad_norm": 0.3815043568611145,
      "learning_rate": 2e-05,
      "loss": 0.7096,
      "step": 4230
    },
    {
      "epoch": 0.37863904268619397,
      "grad_norm": 0.3459341824054718,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 4240
    },
    {
      "epoch": 0.3795320592963029,
      "grad_norm": 0.39455604553222656,
      "learning_rate": 2e-05,
      "loss": 0.6611,
      "step": 4250
    },
    {
      "epoch": 0.38042507590641184,
      "grad_norm": 0.43168190121650696,
      "learning_rate": 2e-05,
      "loss": 0.6748,
      "step": 4260
    },
    {
      "epoch": 0.3813180925165208,
      "grad_norm": 0.35696497559547424,
      "learning_rate": 2e-05,
      "loss": 0.6423,
      "step": 4270
    },
    {
      "epoch": 0.38221110912662976,
      "grad_norm": 0.36734700202941895,
      "learning_rate": 2e-05,
      "loss": 0.673,
      "step": 4280
    },
    {
      "epoch": 0.3831041257367387,
      "grad_norm": 0.3984774053096771,
      "learning_rate": 2e-05,
      "loss": 0.6811,
      "step": 4290
    },
    {
      "epoch": 0.3839971423468477,
      "grad_norm": 0.33982765674591064,
      "learning_rate": 2e-05,
      "loss": 0.6909,
      "step": 4300
    },
    {
      "epoch": 0.3848901589569566,
      "grad_norm": 0.37864384055137634,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 4310
    },
    {
      "epoch": 0.38578317556706554,
      "grad_norm": 0.36823466420173645,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 4320
    },
    {
      "epoch": 0.38667619217717447,
      "grad_norm": 0.40874573588371277,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 4330
    },
    {
      "epoch": 0.38756920878728346,
      "grad_norm": 0.43925756216049194,
      "learning_rate": 2e-05,
      "loss": 0.6699,
      "step": 4340
    },
    {
      "epoch": 0.3884622253973924,
      "grad_norm": 0.4438115656375885,
      "learning_rate": 2e-05,
      "loss": 0.7074,
      "step": 4350
    },
    {
      "epoch": 0.3893552420075013,
      "grad_norm": 0.36280399560928345,
      "learning_rate": 2e-05,
      "loss": 0.6643,
      "step": 4360
    },
    {
      "epoch": 0.3902482586176103,
      "grad_norm": 0.3471735715866089,
      "learning_rate": 2e-05,
      "loss": 0.6921,
      "step": 4370
    },
    {
      "epoch": 0.39114127522771924,
      "grad_norm": 0.3418101370334625,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 4380
    },
    {
      "epoch": 0.3920342918378282,
      "grad_norm": 0.3582875430583954,
      "learning_rate": 2e-05,
      "loss": 0.7202,
      "step": 4390
    },
    {
      "epoch": 0.3929273084479371,
      "grad_norm": 0.37227901816368103,
      "learning_rate": 2e-05,
      "loss": 0.6704,
      "step": 4400
    },
    {
      "epoch": 0.3938203250580461,
      "grad_norm": 0.3621814548969269,
      "learning_rate": 2e-05,
      "loss": 0.7006,
      "step": 4410
    },
    {
      "epoch": 0.394713341668155,
      "grad_norm": 0.35347166657447815,
      "learning_rate": 2e-05,
      "loss": 0.6776,
      "step": 4420
    },
    {
      "epoch": 0.39560635827826396,
      "grad_norm": 0.42193514108657837,
      "learning_rate": 2e-05,
      "loss": 0.6937,
      "step": 4430
    },
    {
      "epoch": 0.39649937488837295,
      "grad_norm": 0.3808785676956177,
      "learning_rate": 2e-05,
      "loss": 0.6377,
      "step": 4440
    },
    {
      "epoch": 0.3973923914984819,
      "grad_norm": 0.37089216709136963,
      "learning_rate": 2e-05,
      "loss": 0.67,
      "step": 4450
    },
    {
      "epoch": 0.3982854081085908,
      "grad_norm": 0.3868139982223511,
      "learning_rate": 2e-05,
      "loss": 0.6919,
      "step": 4460
    },
    {
      "epoch": 0.39917842471869974,
      "grad_norm": 0.37381669878959656,
      "learning_rate": 2e-05,
      "loss": 0.7009,
      "step": 4470
    },
    {
      "epoch": 0.40007144132880873,
      "grad_norm": 0.33514586091041565,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 4480
    },
    {
      "epoch": 0.40096445793891766,
      "grad_norm": 0.34450769424438477,
      "learning_rate": 2e-05,
      "loss": 0.6847,
      "step": 4490
    },
    {
      "epoch": 0.4018574745490266,
      "grad_norm": 0.3657180368900299,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 4500
    },
    {
      "epoch": 0.4027504911591356,
      "grad_norm": 0.38325607776641846,
      "learning_rate": 2e-05,
      "loss": 0.6751,
      "step": 4510
    },
    {
      "epoch": 0.4036435077692445,
      "grad_norm": 0.3803079128265381,
      "learning_rate": 2e-05,
      "loss": 0.6819,
      "step": 4520
    },
    {
      "epoch": 0.40453652437935345,
      "grad_norm": 0.4088026285171509,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 4530
    },
    {
      "epoch": 0.4054295409894624,
      "grad_norm": 0.3512919843196869,
      "learning_rate": 2e-05,
      "loss": 0.6587,
      "step": 4540
    },
    {
      "epoch": 0.40632255759957137,
      "grad_norm": 0.34363865852355957,
      "learning_rate": 2e-05,
      "loss": 0.6863,
      "step": 4550
    },
    {
      "epoch": 0.4072155742096803,
      "grad_norm": 0.34950166940689087,
      "learning_rate": 2e-05,
      "loss": 0.6724,
      "step": 4560
    },
    {
      "epoch": 0.40810859081978923,
      "grad_norm": 0.394996702671051,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 4570
    },
    {
      "epoch": 0.4090016074298982,
      "grad_norm": 0.4020107090473175,
      "learning_rate": 2e-05,
      "loss": 0.6739,
      "step": 4580
    },
    {
      "epoch": 0.40989462404000715,
      "grad_norm": 0.5412101745605469,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 4590
    },
    {
      "epoch": 0.4107876406501161,
      "grad_norm": 0.33643534779548645,
      "learning_rate": 2e-05,
      "loss": 0.6516,
      "step": 4600
    },
    {
      "epoch": 0.411680657260225,
      "grad_norm": 0.3697690963745117,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 4610
    },
    {
      "epoch": 0.412573673870334,
      "grad_norm": 0.33381566405296326,
      "learning_rate": 2e-05,
      "loss": 0.6622,
      "step": 4620
    },
    {
      "epoch": 0.41346669048044293,
      "grad_norm": 0.3048006594181061,
      "learning_rate": 2e-05,
      "loss": 0.6491,
      "step": 4630
    },
    {
      "epoch": 0.41435970709055187,
      "grad_norm": 0.35339999198913574,
      "learning_rate": 2e-05,
      "loss": 0.6826,
      "step": 4640
    },
    {
      "epoch": 0.41525272370066085,
      "grad_norm": 0.40563100576400757,
      "learning_rate": 2e-05,
      "loss": 0.6453,
      "step": 4650
    },
    {
      "epoch": 0.4161457403107698,
      "grad_norm": 0.380941241979599,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 4660
    },
    {
      "epoch": 0.4170387569208787,
      "grad_norm": 0.3585035800933838,
      "learning_rate": 2e-05,
      "loss": 0.6627,
      "step": 4670
    },
    {
      "epoch": 0.41793177353098765,
      "grad_norm": 0.3435126841068268,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 4680
    },
    {
      "epoch": 0.41882479014109664,
      "grad_norm": 0.3623499572277069,
      "learning_rate": 2e-05,
      "loss": 0.6498,
      "step": 4690
    },
    {
      "epoch": 0.41971780675120557,
      "grad_norm": 0.3764245808124542,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 4700
    },
    {
      "epoch": 0.4206108233613145,
      "grad_norm": 0.3425867557525635,
      "learning_rate": 2e-05,
      "loss": 0.6717,
      "step": 4710
    },
    {
      "epoch": 0.4215038399714235,
      "grad_norm": 0.3577650189399719,
      "learning_rate": 2e-05,
      "loss": 0.6627,
      "step": 4720
    },
    {
      "epoch": 0.4223968565815324,
      "grad_norm": 0.36620646715164185,
      "learning_rate": 2e-05,
      "loss": 0.6828,
      "step": 4730
    },
    {
      "epoch": 0.42328987319164135,
      "grad_norm": 0.36670416593551636,
      "learning_rate": 2e-05,
      "loss": 0.6876,
      "step": 4740
    },
    {
      "epoch": 0.42418288980175034,
      "grad_norm": 0.3564507067203522,
      "learning_rate": 2e-05,
      "loss": 0.7026,
      "step": 4750
    },
    {
      "epoch": 0.4250759064118593,
      "grad_norm": 0.4080749452114105,
      "learning_rate": 2e-05,
      "loss": 0.6925,
      "step": 4760
    },
    {
      "epoch": 0.4259689230219682,
      "grad_norm": 0.3509368598461151,
      "learning_rate": 2e-05,
      "loss": 0.6667,
      "step": 4770
    },
    {
      "epoch": 0.42686193963207714,
      "grad_norm": 0.3736545443534851,
      "learning_rate": 2e-05,
      "loss": 0.6771,
      "step": 4780
    },
    {
      "epoch": 0.4277549562421861,
      "grad_norm": 0.3350389301776886,
      "learning_rate": 2e-05,
      "loss": 0.6885,
      "step": 4790
    },
    {
      "epoch": 0.42864797285229506,
      "grad_norm": 0.3636820614337921,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 4800
    },
    {
      "epoch": 0.429540989462404,
      "grad_norm": 0.3685387969017029,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 4810
    },
    {
      "epoch": 0.430434006072513,
      "grad_norm": 0.39025336503982544,
      "learning_rate": 2e-05,
      "loss": 0.6726,
      "step": 4820
    },
    {
      "epoch": 0.4313270226826219,
      "grad_norm": 0.3786081075668335,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 4830
    },
    {
      "epoch": 0.43222003929273084,
      "grad_norm": 0.36511704325675964,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 4840
    },
    {
      "epoch": 0.43311305590283977,
      "grad_norm": 0.32465630769729614,
      "learning_rate": 2e-05,
      "loss": 0.6582,
      "step": 4850
    },
    {
      "epoch": 0.43400607251294876,
      "grad_norm": 0.36687803268432617,
      "learning_rate": 2e-05,
      "loss": 0.6927,
      "step": 4860
    },
    {
      "epoch": 0.4348990891230577,
      "grad_norm": 0.35513487458229065,
      "learning_rate": 2e-05,
      "loss": 0.6798,
      "step": 4870
    },
    {
      "epoch": 0.4357921057331666,
      "grad_norm": 0.38180556893348694,
      "learning_rate": 2e-05,
      "loss": 0.66,
      "step": 4880
    },
    {
      "epoch": 0.4366851223432756,
      "grad_norm": 0.39429476857185364,
      "learning_rate": 2e-05,
      "loss": 0.6777,
      "step": 4890
    },
    {
      "epoch": 0.43757813895338454,
      "grad_norm": 0.38248512148857117,
      "learning_rate": 2e-05,
      "loss": 0.6685,
      "step": 4900
    },
    {
      "epoch": 0.4384711555634935,
      "grad_norm": 0.3432442545890808,
      "learning_rate": 2e-05,
      "loss": 0.6747,
      "step": 4910
    },
    {
      "epoch": 0.4393641721736024,
      "grad_norm": 0.35280707478523254,
      "learning_rate": 2e-05,
      "loss": 0.6851,
      "step": 4920
    },
    {
      "epoch": 0.4402571887837114,
      "grad_norm": 0.3475905954837799,
      "learning_rate": 2e-05,
      "loss": 0.6685,
      "step": 4930
    },
    {
      "epoch": 0.4411502053938203,
      "grad_norm": 0.3824846148490906,
      "learning_rate": 2e-05,
      "loss": 0.6698,
      "step": 4940
    },
    {
      "epoch": 0.44204322200392926,
      "grad_norm": 0.3793005049228668,
      "learning_rate": 2e-05,
      "loss": 0.6754,
      "step": 4950
    },
    {
      "epoch": 0.44293623861403825,
      "grad_norm": 0.36466237902641296,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 4960
    },
    {
      "epoch": 0.4438292552241472,
      "grad_norm": 0.3465329110622406,
      "learning_rate": 2e-05,
      "loss": 0.664,
      "step": 4970
    },
    {
      "epoch": 0.4447222718342561,
      "grad_norm": 0.3506736159324646,
      "learning_rate": 2e-05,
      "loss": 0.6785,
      "step": 4980
    },
    {
      "epoch": 0.44561528844436504,
      "grad_norm": 0.34552523493766785,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 4990
    },
    {
      "epoch": 0.44650830505447403,
      "grad_norm": 0.38686060905456543,
      "learning_rate": 2e-05,
      "loss": 0.6207,
      "step": 5000
    },
    {
      "epoch": 0.44740132166458296,
      "grad_norm": 0.3621189296245575,
      "learning_rate": 2e-05,
      "loss": 0.6745,
      "step": 5010
    },
    {
      "epoch": 0.4482943382746919,
      "grad_norm": 0.36412402987480164,
      "learning_rate": 2e-05,
      "loss": 0.6785,
      "step": 5020
    },
    {
      "epoch": 0.4491873548848009,
      "grad_norm": 0.3481839895248413,
      "learning_rate": 2e-05,
      "loss": 0.6755,
      "step": 5030
    },
    {
      "epoch": 0.4500803714949098,
      "grad_norm": 0.3802565336227417,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 5040
    },
    {
      "epoch": 0.45097338810501875,
      "grad_norm": 0.3811277449131012,
      "learning_rate": 2e-05,
      "loss": 0.6751,
      "step": 5050
    },
    {
      "epoch": 0.4518664047151277,
      "grad_norm": 0.36518847942352295,
      "learning_rate": 2e-05,
      "loss": 0.6939,
      "step": 5060
    },
    {
      "epoch": 0.45275942132523667,
      "grad_norm": 0.3774931728839874,
      "learning_rate": 2e-05,
      "loss": 0.6897,
      "step": 5070
    },
    {
      "epoch": 0.4536524379353456,
      "grad_norm": 0.355311781167984,
      "learning_rate": 2e-05,
      "loss": 0.6793,
      "step": 5080
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 0.35002586245536804,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.6198,
      "step": 5090
    },
    {
      "epoch": 0.4554384711555635,
      "grad_norm": 0.40497156977653503,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.6884,
      "step": 5100
    },
    {
      "epoch": 0.45633148776567245,
      "grad_norm": 0.689021110534668,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.6795,
      "step": 5110
    },
    {
      "epoch": 0.4572245043757814,
      "grad_norm": 0.4345141053199768,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.6518,
      "step": 5120
    },
    {
      "epoch": 0.4581175209858903,
      "grad_norm": 0.3613136410713196,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.6736,
      "step": 5130
    },
    {
      "epoch": 0.4590105375959993,
      "grad_norm": 0.3315771222114563,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.6715,
      "step": 5140
    },
    {
      "epoch": 0.45990355420610823,
      "grad_norm": 0.4435660243034363,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.6607,
      "step": 5150
    },
    {
      "epoch": 0.46079657081621717,
      "grad_norm": 0.36641597747802734,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.6532,
      "step": 5160
    },
    {
      "epoch": 0.46168958742632615,
      "grad_norm": 0.38577964901924133,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.6618,
      "step": 5170
    },
    {
      "epoch": 0.4625826040364351,
      "grad_norm": 0.37097883224487305,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.6369,
      "step": 5180
    },
    {
      "epoch": 0.463475620646544,
      "grad_norm": 0.40079164505004883,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.6715,
      "step": 5190
    },
    {
      "epoch": 0.46436863725665295,
      "grad_norm": 0.4228552579879761,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.6815,
      "step": 5200
    },
    {
      "epoch": 0.46526165386676194,
      "grad_norm": 0.38120079040527344,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.6733,
      "step": 5210
    },
    {
      "epoch": 0.46615467047687087,
      "grad_norm": 0.34104806184768677,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.6639,
      "step": 5220
    },
    {
      "epoch": 0.4670476870869798,
      "grad_norm": 0.37392836809158325,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.6796,
      "step": 5230
    },
    {
      "epoch": 0.4679407036970888,
      "grad_norm": 0.4016932547092438,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.6646,
      "step": 5240
    },
    {
      "epoch": 0.4688337203071977,
      "grad_norm": 0.43677380681037903,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.6736,
      "step": 5250
    },
    {
      "epoch": 0.46972673691730665,
      "grad_norm": 0.35742026567459106,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.6468,
      "step": 5260
    },
    {
      "epoch": 0.4706197535274156,
      "grad_norm": 0.3267393112182617,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.6759,
      "step": 5270
    },
    {
      "epoch": 0.4715127701375246,
      "grad_norm": 0.3282828629016876,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.6529,
      "step": 5280
    },
    {
      "epoch": 0.4724057867476335,
      "grad_norm": 0.3450550436973572,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.6943,
      "step": 5290
    },
    {
      "epoch": 0.47329880335774244,
      "grad_norm": 0.3606621026992798,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.6467,
      "step": 5300
    },
    {
      "epoch": 0.4741918199678514,
      "grad_norm": 0.36301174759864807,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.6584,
      "step": 5310
    },
    {
      "epoch": 0.47508483657796036,
      "grad_norm": 0.3804932236671448,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.6622,
      "step": 5320
    },
    {
      "epoch": 0.4759778531880693,
      "grad_norm": 0.3563631474971771,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.6537,
      "step": 5330
    },
    {
      "epoch": 0.4768708697981782,
      "grad_norm": 0.3642588257789612,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.6538,
      "step": 5340
    },
    {
      "epoch": 0.4777638864082872,
      "grad_norm": 0.35379135608673096,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.6589,
      "step": 5350
    },
    {
      "epoch": 0.47865690301839614,
      "grad_norm": 0.39452049136161804,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.6834,
      "step": 5360
    },
    {
      "epoch": 0.4795499196285051,
      "grad_norm": 0.343686044216156,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.6586,
      "step": 5370
    },
    {
      "epoch": 0.48044293623861406,
      "grad_norm": 0.3464377820491791,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.6434,
      "step": 5380
    },
    {
      "epoch": 0.481335952848723,
      "grad_norm": 0.3426789939403534,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.6773,
      "step": 5390
    },
    {
      "epoch": 0.4822289694588319,
      "grad_norm": 0.3540111780166626,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.6444,
      "step": 5400
    },
    {
      "epoch": 0.48312198606894086,
      "grad_norm": 0.31409138441085815,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.6313,
      "step": 5410
    },
    {
      "epoch": 0.48401500267904984,
      "grad_norm": 0.376188725233078,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.6544,
      "step": 5420
    },
    {
      "epoch": 0.4849080192891588,
      "grad_norm": 0.3917011618614197,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.6619,
      "step": 5430
    },
    {
      "epoch": 0.4858010358992677,
      "grad_norm": 0.35530489683151245,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.675,
      "step": 5440
    },
    {
      "epoch": 0.4866940525093767,
      "grad_norm": 0.3248516321182251,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.6555,
      "step": 5450
    },
    {
      "epoch": 0.4875870691194856,
      "grad_norm": 0.3662690222263336,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.6536,
      "step": 5460
    },
    {
      "epoch": 0.48848008572959456,
      "grad_norm": 0.3396593928337097,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.6767,
      "step": 5470
    },
    {
      "epoch": 0.4893731023397035,
      "grad_norm": 0.36480018496513367,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.6498,
      "step": 5480
    },
    {
      "epoch": 0.4902661189498125,
      "grad_norm": 0.32998889684677124,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.6487,
      "step": 5490
    },
    {
      "epoch": 0.4911591355599214,
      "grad_norm": 0.32447898387908936,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.6463,
      "step": 5500
    },
    {
      "epoch": 0.49205215217003034,
      "grad_norm": 0.30881306529045105,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.6655,
      "step": 5510
    },
    {
      "epoch": 0.49294516878013933,
      "grad_norm": 0.3724258542060852,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.6542,
      "step": 5520
    },
    {
      "epoch": 0.49383818539024826,
      "grad_norm": 0.32063809037208557,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.6438,
      "step": 5530
    },
    {
      "epoch": 0.4947312020003572,
      "grad_norm": 0.3529360890388489,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.6615,
      "step": 5540
    },
    {
      "epoch": 0.4956242186104661,
      "grad_norm": 0.3483227789402008,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.6562,
      "step": 5550
    },
    {
      "epoch": 0.4965172352205751,
      "grad_norm": 0.3479405641555786,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.6586,
      "step": 5560
    },
    {
      "epoch": 0.49741025183068405,
      "grad_norm": 0.3425529897212982,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.6778,
      "step": 5570
    },
    {
      "epoch": 0.498303268440793,
      "grad_norm": 0.35957732796669006,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6688,
      "step": 5580
    },
    {
      "epoch": 0.49919628505090197,
      "grad_norm": 0.3722885251045227,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.6544,
      "step": 5590
    },
    {
      "epoch": 0.4998213966779782,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.022465601677934634,
      "train_runtime": 1758.0063,
      "train_samples_per_second": 38.205,
      "train_steps_per_second": 3.184
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}