{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9300432037221669,
  "eval_steps": 500,
  "global_step": 11194,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008308408109006315,
      "grad_norm": 5.0185394287109375,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 1.7151,
      "step": 10
    },
    {
      "epoch": 0.001661681621801263,
      "grad_norm": 2.385542631149292,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.7579,
      "step": 20
    },
    {
      "epoch": 0.0024925224327018943,
      "grad_norm": 2.753164291381836,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.4674,
      "step": 30
    },
    {
      "epoch": 0.003323363243602526,
      "grad_norm": 0.8955636024475098,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.2782,
      "step": 40
    },
    {
      "epoch": 0.004154204054503157,
      "grad_norm": 0.9788411855697632,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.1572,
      "step": 50
    },
    {
      "epoch": 0.004985044865403789,
      "grad_norm": 0.5795058012008667,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.1187,
      "step": 60
    },
    {
      "epoch": 0.00581588567630442,
      "grad_norm": 0.6206412315368652,
      "learning_rate": 7.814044943820226e-06,
      "loss": 1.0551,
      "step": 70
    },
    {
      "epoch": 0.006646726487205052,
      "grad_norm": 0.604301929473877,
      "learning_rate": 8.932022471910112e-06,
      "loss": 1.0394,
      "step": 80
    },
    {
      "epoch": 0.007477567298105683,
      "grad_norm": 0.6093521118164062,
      "learning_rate": 1.005e-05,
      "loss": 1.0592,
      "step": 90
    },
    {
      "epoch": 0.008308408109006314,
      "grad_norm": 0.6659312844276428,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 1.0234,
      "step": 100
    },
    {
      "epoch": 0.009139248919906947,
      "grad_norm": 0.5740776658058167,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 1.0529,
      "step": 110
    },
    {
      "epoch": 0.009970089730807577,
      "grad_norm": 0.576384961605072,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.9948,
      "step": 120
    },
    {
      "epoch": 0.010800930541708208,
      "grad_norm": 0.953546941280365,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 1.0426,
      "step": 130
    },
    {
      "epoch": 0.01163177135260884,
      "grad_norm": 0.5870928168296814,
      "learning_rate": 1.563988764044944e-05,
      "loss": 1.0148,
      "step": 140
    },
    {
      "epoch": 0.012462612163509471,
      "grad_norm": 0.6562204360961914,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 1.0398,
      "step": 150
    },
    {
      "epoch": 0.013293452974410104,
      "grad_norm": 0.5265231132507324,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 1.0222,
      "step": 160
    },
    {
      "epoch": 0.014124293785310734,
      "grad_norm": 0.5627759099006653,
      "learning_rate": 1.89938202247191e-05,
      "loss": 1.0173,
      "step": 170
    },
    {
      "epoch": 0.014955134596211365,
      "grad_norm": 0.5194225907325745,
      "learning_rate": 2e-05,
      "loss": 0.989,
      "step": 180
    },
    {
      "epoch": 0.015785975407111998,
      "grad_norm": 0.5717377066612244,
      "learning_rate": 2e-05,
      "loss": 0.9578,
      "step": 190
    },
    {
      "epoch": 0.01661681621801263,
      "grad_norm": 0.5366178750991821,
      "learning_rate": 2e-05,
      "loss": 1.0255,
      "step": 200
    },
    {
      "epoch": 0.01744765702891326,
      "grad_norm": 0.6772796511650085,
      "learning_rate": 2e-05,
      "loss": 1.0158,
      "step": 210
    },
    {
      "epoch": 0.018278497839813893,
      "grad_norm": 0.5122703909873962,
      "learning_rate": 2e-05,
      "loss": 1.0005,
      "step": 220
    },
    {
      "epoch": 0.019109338650714524,
      "grad_norm": 0.6427125334739685,
      "learning_rate": 2e-05,
      "loss": 1.0214,
      "step": 230
    },
    {
      "epoch": 0.019940179461615155,
      "grad_norm": 0.49990859627723694,
      "learning_rate": 2e-05,
      "loss": 0.9847,
      "step": 240
    },
    {
      "epoch": 0.020771020272515785,
      "grad_norm": 0.5407299399375916,
      "learning_rate": 2e-05,
      "loss": 0.9808,
      "step": 250
    },
    {
      "epoch": 0.021601861083416416,
      "grad_norm": 0.5241397619247437,
      "learning_rate": 2e-05,
      "loss": 0.9863,
      "step": 260
    },
    {
      "epoch": 0.02243270189431705,
      "grad_norm": 0.5140271782875061,
      "learning_rate": 2e-05,
      "loss": 1.0024,
      "step": 270
    },
    {
      "epoch": 0.02326354270521768,
      "grad_norm": 0.5602551698684692,
      "learning_rate": 2e-05,
      "loss": 0.9681,
      "step": 280
    },
    {
      "epoch": 0.024094383516118312,
      "grad_norm": 0.5047415494918823,
      "learning_rate": 2e-05,
      "loss": 1.007,
      "step": 290
    },
    {
      "epoch": 0.024925224327018942,
      "grad_norm": 0.4970276653766632,
      "learning_rate": 2e-05,
      "loss": 0.9759,
      "step": 300
    },
    {
      "epoch": 0.025756065137919573,
      "grad_norm": 0.5245757102966309,
      "learning_rate": 2e-05,
      "loss": 0.9802,
      "step": 310
    },
    {
      "epoch": 0.026586905948820207,
      "grad_norm": 0.5001291036605835,
      "learning_rate": 2e-05,
      "loss": 0.9628,
      "step": 320
    },
    {
      "epoch": 0.027417746759720838,
      "grad_norm": 0.5402788519859314,
      "learning_rate": 2e-05,
      "loss": 0.9878,
      "step": 330
    },
    {
      "epoch": 0.02824858757062147,
      "grad_norm": 0.5057405829429626,
      "learning_rate": 2e-05,
      "loss": 1.017,
      "step": 340
    },
    {
      "epoch": 0.0290794283815221,
      "grad_norm": 0.5450842380523682,
      "learning_rate": 2e-05,
      "loss": 0.9815,
      "step": 350
    },
    {
      "epoch": 0.02991026919242273,
      "grad_norm": 0.6072407364845276,
      "learning_rate": 2e-05,
      "loss": 1.0245,
      "step": 360
    },
    {
      "epoch": 0.030741110003323364,
      "grad_norm": 0.5196362733840942,
      "learning_rate": 2e-05,
      "loss": 0.9699,
      "step": 370
    },
    {
      "epoch": 0.031571950814223995,
      "grad_norm": 0.5128522515296936,
      "learning_rate": 2e-05,
      "loss": 1.0091,
      "step": 380
    },
    {
      "epoch": 0.032402791625124626,
      "grad_norm": 0.6012204885482788,
      "learning_rate": 2e-05,
      "loss": 0.985,
      "step": 390
    },
    {
      "epoch": 0.03323363243602526,
      "grad_norm": 0.5065486431121826,
      "learning_rate": 2e-05,
      "loss": 0.9606,
      "step": 400
    },
    {
      "epoch": 0.03406447324692589,
      "grad_norm": 0.5270485281944275,
      "learning_rate": 2e-05,
      "loss": 0.9496,
      "step": 410
    },
    {
      "epoch": 0.03489531405782652,
      "grad_norm": 0.5859076380729675,
      "learning_rate": 2e-05,
      "loss": 0.9841,
      "step": 420
    },
    {
      "epoch": 0.03572615486872715,
      "grad_norm": 0.5001581311225891,
      "learning_rate": 2e-05,
      "loss": 0.9435,
      "step": 430
    },
    {
      "epoch": 0.036556995679627786,
      "grad_norm": 0.5033804178237915,
      "learning_rate": 2e-05,
      "loss": 1.0107,
      "step": 440
    },
    {
      "epoch": 0.03738783649052842,
      "grad_norm": 0.48901426792144775,
      "learning_rate": 2e-05,
      "loss": 0.9775,
      "step": 450
    },
    {
      "epoch": 0.03821867730142905,
      "grad_norm": 0.4957520067691803,
      "learning_rate": 2e-05,
      "loss": 0.9657,
      "step": 460
    },
    {
      "epoch": 0.03904951811232968,
      "grad_norm": 0.48940908908843994,
      "learning_rate": 2e-05,
      "loss": 0.9631,
      "step": 470
    },
    {
      "epoch": 0.03988035892323031,
      "grad_norm": 0.4692400097846985,
      "learning_rate": 2e-05,
      "loss": 0.9791,
      "step": 480
    },
    {
      "epoch": 0.04071119973413094,
      "grad_norm": 0.49479761719703674,
      "learning_rate": 2e-05,
      "loss": 0.9712,
      "step": 490
    },
    {
      "epoch": 0.04154204054503157,
      "grad_norm": 0.5517172813415527,
      "learning_rate": 2e-05,
      "loss": 0.9865,
      "step": 500
    },
    {
      "epoch": 0.0423728813559322,
      "grad_norm": 0.5271807312965393,
      "learning_rate": 2e-05,
      "loss": 0.9872,
      "step": 510
    },
    {
      "epoch": 0.04320372216683283,
      "grad_norm": 0.5044783353805542,
      "learning_rate": 2e-05,
      "loss": 0.9525,
      "step": 520
    },
    {
      "epoch": 0.04403456297773346,
      "grad_norm": 0.5325606465339661,
      "learning_rate": 2e-05,
      "loss": 0.9773,
      "step": 530
    },
    {
      "epoch": 0.0448654037886341,
      "grad_norm": 0.463705837726593,
      "learning_rate": 2e-05,
      "loss": 0.9835,
      "step": 540
    },
    {
      "epoch": 0.04569624459953473,
      "grad_norm": 0.49333709478378296,
      "learning_rate": 2e-05,
      "loss": 0.9739,
      "step": 550
    },
    {
      "epoch": 0.04652708541043536,
      "grad_norm": 0.49047335982322693,
      "learning_rate": 2e-05,
      "loss": 0.9679,
      "step": 560
    },
    {
      "epoch": 0.04735792622133599,
      "grad_norm": 0.4731309711933136,
      "learning_rate": 2e-05,
      "loss": 0.9521,
      "step": 570
    },
    {
      "epoch": 0.048188767032236623,
      "grad_norm": 0.8461064100265503,
      "learning_rate": 2e-05,
      "loss": 0.9521,
      "step": 580
    },
    {
      "epoch": 0.049019607843137254,
      "grad_norm": 0.47802215814590454,
      "learning_rate": 2e-05,
      "loss": 0.9634,
      "step": 590
    },
    {
      "epoch": 0.049850448654037885,
      "grad_norm": 0.4835042655467987,
      "learning_rate": 2e-05,
      "loss": 0.9522,
      "step": 600
    },
    {
      "epoch": 0.050681289464938516,
      "grad_norm": 0.49139562249183655,
      "learning_rate": 2e-05,
      "loss": 0.9417,
      "step": 610
    },
    {
      "epoch": 0.051512130275839146,
      "grad_norm": 0.5035521984100342,
      "learning_rate": 2e-05,
      "loss": 0.9599,
      "step": 620
    },
    {
      "epoch": 0.052342971086739784,
      "grad_norm": 0.4629312753677368,
      "learning_rate": 2e-05,
      "loss": 0.925,
      "step": 630
    },
    {
      "epoch": 0.053173811897640415,
      "grad_norm": 0.4884300231933594,
      "learning_rate": 2e-05,
      "loss": 0.9467,
      "step": 640
    },
    {
      "epoch": 0.054004652708541045,
      "grad_norm": 0.47893789410591125,
      "learning_rate": 2e-05,
      "loss": 0.9436,
      "step": 650
    },
    {
      "epoch": 0.054835493519441676,
      "grad_norm": 0.49069446325302124,
      "learning_rate": 2e-05,
      "loss": 0.9789,
      "step": 660
    },
    {
      "epoch": 0.05566633433034231,
      "grad_norm": 0.5776323676109314,
      "learning_rate": 2e-05,
      "loss": 0.9727,
      "step": 670
    },
    {
      "epoch": 0.05649717514124294,
      "grad_norm": 0.46106594800949097,
      "learning_rate": 2e-05,
      "loss": 0.945,
      "step": 680
    },
    {
      "epoch": 0.05732801595214357,
      "grad_norm": 0.4655075967311859,
      "learning_rate": 2e-05,
      "loss": 0.9593,
      "step": 690
    },
    {
      "epoch": 0.0581588567630442,
      "grad_norm": 0.5039606094360352,
      "learning_rate": 2e-05,
      "loss": 0.9565,
      "step": 700
    },
    {
      "epoch": 0.05898969757394483,
      "grad_norm": 0.43979552388191223,
      "learning_rate": 2e-05,
      "loss": 0.9728,
      "step": 710
    },
    {
      "epoch": 0.05982053838484546,
      "grad_norm": 0.5355712175369263,
      "learning_rate": 2e-05,
      "loss": 0.9588,
      "step": 720
    },
    {
      "epoch": 0.0606513791957461,
      "grad_norm": 0.48468437790870667,
      "learning_rate": 2e-05,
      "loss": 0.9721,
      "step": 730
    },
    {
      "epoch": 0.06148222000664673,
      "grad_norm": 0.6858370304107666,
      "learning_rate": 2e-05,
      "loss": 0.9605,
      "step": 740
    },
    {
      "epoch": 0.06231306081754736,
      "grad_norm": 0.4598209261894226,
      "learning_rate": 2e-05,
      "loss": 0.9432,
      "step": 750
    },
    {
      "epoch": 0.06314390162844799,
      "grad_norm": 0.46015241742134094,
      "learning_rate": 2e-05,
      "loss": 0.9338,
      "step": 760
    },
    {
      "epoch": 0.06397474243934861,
      "grad_norm": 0.5221343040466309,
      "learning_rate": 2e-05,
      "loss": 0.9755,
      "step": 770
    },
    {
      "epoch": 0.06480558325024925,
      "grad_norm": 0.4553109109401703,
      "learning_rate": 2e-05,
      "loss": 0.9429,
      "step": 780
    },
    {
      "epoch": 0.06563642406114989,
      "grad_norm": 0.47823524475097656,
      "learning_rate": 2e-05,
      "loss": 0.9784,
      "step": 790
    },
    {
      "epoch": 0.06646726487205051,
      "grad_norm": 0.49694108963012695,
      "learning_rate": 2e-05,
      "loss": 0.9785,
      "step": 800
    },
    {
      "epoch": 0.06729810568295115,
      "grad_norm": 0.5347675085067749,
      "learning_rate": 2e-05,
      "loss": 0.9018,
      "step": 810
    },
    {
      "epoch": 0.06812894649385177,
      "grad_norm": 0.45874178409576416,
      "learning_rate": 2e-05,
      "loss": 0.9555,
      "step": 820
    },
    {
      "epoch": 0.06895978730475241,
      "grad_norm": 0.49684765934944153,
      "learning_rate": 2e-05,
      "loss": 0.964,
      "step": 830
    },
    {
      "epoch": 0.06979062811565304,
      "grad_norm": 0.4911392331123352,
      "learning_rate": 2e-05,
      "loss": 0.9173,
      "step": 840
    },
    {
      "epoch": 0.07062146892655367,
      "grad_norm": 0.4836362600326538,
      "learning_rate": 2e-05,
      "loss": 0.953,
      "step": 850
    },
    {
      "epoch": 0.0714523097374543,
      "grad_norm": 0.5126013159751892,
      "learning_rate": 2e-05,
      "loss": 0.9453,
      "step": 860
    },
    {
      "epoch": 0.07228315054835494,
      "grad_norm": 0.5170702338218689,
      "learning_rate": 2e-05,
      "loss": 0.9615,
      "step": 870
    },
    {
      "epoch": 0.07311399135925557,
      "grad_norm": 0.4929334223270416,
      "learning_rate": 2e-05,
      "loss": 0.9343,
      "step": 880
    },
    {
      "epoch": 0.0739448321701562,
      "grad_norm": 0.4450336992740631,
      "learning_rate": 2e-05,
      "loss": 0.9526,
      "step": 890
    },
    {
      "epoch": 0.07477567298105683,
      "grad_norm": 0.44215553998947144,
      "learning_rate": 2e-05,
      "loss": 0.9376,
      "step": 900
    },
    {
      "epoch": 0.07560651379195746,
      "grad_norm": 0.46003344655036926,
      "learning_rate": 2e-05,
      "loss": 0.9352,
      "step": 910
    },
    {
      "epoch": 0.0764373546028581,
      "grad_norm": 0.46930357813835144,
      "learning_rate": 2e-05,
      "loss": 0.9526,
      "step": 920
    },
    {
      "epoch": 0.07726819541375872,
      "grad_norm": 0.47233858704566956,
      "learning_rate": 2e-05,
      "loss": 0.933,
      "step": 930
    },
    {
      "epoch": 0.07809903622465936,
      "grad_norm": 0.47735634446144104,
      "learning_rate": 2e-05,
      "loss": 0.9433,
      "step": 940
    },
    {
      "epoch": 0.07892987703555998,
      "grad_norm": 0.48975709080696106,
      "learning_rate": 2e-05,
      "loss": 0.9652,
      "step": 950
    },
    {
      "epoch": 0.07976071784646062,
      "grad_norm": 0.5132952332496643,
      "learning_rate": 2e-05,
      "loss": 0.9478,
      "step": 960
    },
    {
      "epoch": 0.08059155865736126,
      "grad_norm": 0.4636126458644867,
      "learning_rate": 2e-05,
      "loss": 0.9853,
      "step": 970
    },
    {
      "epoch": 0.08142239946826188,
      "grad_norm": 0.4711599349975586,
      "learning_rate": 2e-05,
      "loss": 0.9557,
      "step": 980
    },
    {
      "epoch": 0.08225324027916252,
      "grad_norm": 0.4458945393562317,
      "learning_rate": 2e-05,
      "loss": 0.9526,
      "step": 990
    },
    {
      "epoch": 0.08308408109006314,
      "grad_norm": 0.4782368242740631,
      "learning_rate": 2e-05,
      "loss": 0.919,
      "step": 1000
    },
    {
      "epoch": 0.08391492190096378,
      "grad_norm": 0.4880605638027191,
      "learning_rate": 2e-05,
      "loss": 0.9535,
      "step": 1010
    },
    {
      "epoch": 0.0847457627118644,
      "grad_norm": 0.507720410823822,
      "learning_rate": 2e-05,
      "loss": 0.958,
      "step": 1020
    },
    {
      "epoch": 0.08557660352276504,
      "grad_norm": 0.4823307394981384,
      "learning_rate": 2e-05,
      "loss": 0.9177,
      "step": 1030
    },
    {
      "epoch": 0.08640744433366566,
      "grad_norm": 0.5217606425285339,
      "learning_rate": 2e-05,
      "loss": 0.9566,
      "step": 1040
    },
    {
      "epoch": 0.0872382851445663,
      "grad_norm": 0.6853054165840149,
      "learning_rate": 2e-05,
      "loss": 0.9474,
      "step": 1050
    },
    {
      "epoch": 0.08806912595546693,
      "grad_norm": 0.4901024401187897,
      "learning_rate": 2e-05,
      "loss": 0.98,
      "step": 1060
    },
    {
      "epoch": 0.08889996676636756,
      "grad_norm": 0.5073767900466919,
      "learning_rate": 2e-05,
      "loss": 0.9348,
      "step": 1070
    },
    {
      "epoch": 0.0897308075772682,
      "grad_norm": 0.49116551876068115,
      "learning_rate": 2e-05,
      "loss": 0.9511,
      "step": 1080
    },
    {
      "epoch": 0.09056164838816883,
      "grad_norm": 0.4953039884567261,
      "learning_rate": 2e-05,
      "loss": 0.9491,
      "step": 1090
    },
    {
      "epoch": 0.09139248919906946,
      "grad_norm": 0.4982893168926239,
      "learning_rate": 2e-05,
      "loss": 0.9414,
      "step": 1100
    },
    {
      "epoch": 0.09222333000997009,
      "grad_norm": 0.5743095874786377,
      "learning_rate": 2e-05,
      "loss": 0.9465,
      "step": 1110
    },
    {
      "epoch": 0.09305417082087072,
      "grad_norm": 0.4815952479839325,
      "learning_rate": 2e-05,
      "loss": 0.9588,
      "step": 1120
    },
    {
      "epoch": 0.09388501163177135,
      "grad_norm": 0.45988714694976807,
      "learning_rate": 2e-05,
      "loss": 0.9521,
      "step": 1130
    },
    {
      "epoch": 0.09471585244267199,
      "grad_norm": 0.4645231366157532,
      "learning_rate": 2e-05,
      "loss": 0.9501,
      "step": 1140
    },
    {
      "epoch": 0.09554669325357261,
      "grad_norm": 0.6942915916442871,
      "learning_rate": 2e-05,
      "loss": 0.9516,
      "step": 1150
    },
    {
      "epoch": 0.09637753406447325,
      "grad_norm": 0.49897119402885437,
      "learning_rate": 2e-05,
      "loss": 0.9197,
      "step": 1160
    },
    {
      "epoch": 0.09720837487537388,
      "grad_norm": 0.4670521914958954,
      "learning_rate": 2e-05,
      "loss": 0.948,
      "step": 1170
    },
    {
      "epoch": 0.09803921568627451,
      "grad_norm": 0.5074694156646729,
      "learning_rate": 2e-05,
      "loss": 0.9475,
      "step": 1180
    },
    {
      "epoch": 0.09887005649717515,
      "grad_norm": 0.5443651676177979,
      "learning_rate": 2e-05,
      "loss": 0.9605,
      "step": 1190
    },
    {
      "epoch": 0.09970089730807577,
      "grad_norm": 0.450802206993103,
      "learning_rate": 2e-05,
      "loss": 0.9235,
      "step": 1200
    },
    {
      "epoch": 0.10053173811897641,
      "grad_norm": 0.45253437757492065,
      "learning_rate": 2e-05,
      "loss": 0.9272,
      "step": 1210
    },
    {
      "epoch": 0.10136257892987703,
      "grad_norm": 0.46179214119911194,
      "learning_rate": 2e-05,
      "loss": 0.9332,
      "step": 1220
    },
    {
      "epoch": 0.10219341974077767,
      "grad_norm": 0.45921969413757324,
      "learning_rate": 2e-05,
      "loss": 0.9005,
      "step": 1230
    },
    {
      "epoch": 0.10302426055167829,
      "grad_norm": 0.4624720513820648,
      "learning_rate": 2e-05,
      "loss": 0.9354,
      "step": 1240
    },
    {
      "epoch": 0.10385510136257893,
      "grad_norm": 0.4451490044593811,
      "learning_rate": 2e-05,
      "loss": 0.9387,
      "step": 1250
    },
    {
      "epoch": 0.10468594217347957,
      "grad_norm": 0.4594932496547699,
      "learning_rate": 2e-05,
      "loss": 0.9569,
      "step": 1260
    },
    {
      "epoch": 0.10551678298438019,
      "grad_norm": 0.5099719762802124,
      "learning_rate": 2e-05,
      "loss": 0.9416,
      "step": 1270
    },
    {
      "epoch": 0.10634762379528083,
      "grad_norm": 0.4535796046257019,
      "learning_rate": 2e-05,
      "loss": 0.9333,
      "step": 1280
    },
    {
      "epoch": 0.10717846460618145,
      "grad_norm": 0.4514749348163605,
      "learning_rate": 2e-05,
      "loss": 0.8909,
      "step": 1290
    },
    {
      "epoch": 0.10800930541708209,
      "grad_norm": 0.46552297472953796,
      "learning_rate": 2e-05,
      "loss": 0.9753,
      "step": 1300
    },
    {
      "epoch": 0.10884014622798271,
      "grad_norm": 0.47683244943618774,
      "learning_rate": 2e-05,
      "loss": 0.9432,
      "step": 1310
    },
    {
      "epoch": 0.10967098703888335,
      "grad_norm": 0.4612570106983185,
      "learning_rate": 2e-05,
      "loss": 0.926,
      "step": 1320
    },
    {
      "epoch": 0.11050182784978398,
      "grad_norm": 0.5698566436767578,
      "learning_rate": 2e-05,
      "loss": 0.9264,
      "step": 1330
    },
    {
      "epoch": 0.11133266866068461,
      "grad_norm": 0.4209228754043579,
      "learning_rate": 2e-05,
      "loss": 0.924,
      "step": 1340
    },
    {
      "epoch": 0.11216350947158524,
      "grad_norm": 0.4713667929172516,
      "learning_rate": 2e-05,
      "loss": 0.9605,
      "step": 1350
    },
    {
      "epoch": 0.11299435028248588,
      "grad_norm": 0.4521118104457855,
      "learning_rate": 2e-05,
      "loss": 0.9226,
      "step": 1360
    },
    {
      "epoch": 0.11382519109338651,
      "grad_norm": 0.4553276300430298,
      "learning_rate": 2e-05,
      "loss": 0.9243,
      "step": 1370
    },
    {
      "epoch": 0.11465603190428714,
      "grad_norm": 0.4790123999118805,
      "learning_rate": 2e-05,
      "loss": 0.9337,
      "step": 1380
    },
    {
      "epoch": 0.11548687271518777,
      "grad_norm": 0.5393106341362,
      "learning_rate": 2e-05,
      "loss": 0.9295,
      "step": 1390
    },
    {
      "epoch": 0.1163177135260884,
      "grad_norm": 0.4630919098854065,
      "learning_rate": 2e-05,
      "loss": 0.9283,
      "step": 1400
    },
    {
      "epoch": 0.11714855433698904,
      "grad_norm": 0.4788759648799896,
      "learning_rate": 2e-05,
      "loss": 0.9245,
      "step": 1410
    },
    {
      "epoch": 0.11797939514788966,
      "grad_norm": 0.47794008255004883,
      "learning_rate": 2e-05,
      "loss": 0.9296,
      "step": 1420
    },
    {
      "epoch": 0.1188102359587903,
      "grad_norm": 0.5339735746383667,
      "learning_rate": 2e-05,
      "loss": 0.9245,
      "step": 1430
    },
    {
      "epoch": 0.11964107676969092,
      "grad_norm": 0.429898202419281,
      "learning_rate": 2e-05,
      "loss": 0.9088,
      "step": 1440
    },
    {
      "epoch": 0.12047191758059156,
      "grad_norm": 0.5483120083808899,
      "learning_rate": 2e-05,
      "loss": 0.9283,
      "step": 1450
    },
    {
      "epoch": 0.1213027583914922,
      "grad_norm": 0.4595089554786682,
      "learning_rate": 2e-05,
      "loss": 0.9595,
      "step": 1460
    },
    {
      "epoch": 0.12213359920239282,
      "grad_norm": 0.47356319427490234,
      "learning_rate": 2e-05,
      "loss": 0.9343,
      "step": 1470
    },
    {
      "epoch": 0.12296444001329346,
      "grad_norm": 0.4519900977611542,
      "learning_rate": 2e-05,
      "loss": 0.9264,
      "step": 1480
    },
    {
      "epoch": 0.12379528082419408,
      "grad_norm": 0.4612765908241272,
      "learning_rate": 2e-05,
      "loss": 0.9281,
      "step": 1490
    },
    {
      "epoch": 0.12462612163509472,
      "grad_norm": 0.5075944066047668,
      "learning_rate": 2e-05,
      "loss": 0.9271,
      "step": 1500
    },
    {
      "epoch": 0.12545696244599536,
      "grad_norm": 0.460589200258255,
      "learning_rate": 2e-05,
      "loss": 0.9271,
      "step": 1510
    },
    {
      "epoch": 0.12628780325689598,
      "grad_norm": 0.44189098477363586,
      "learning_rate": 2e-05,
      "loss": 0.9115,
      "step": 1520
    },
    {
      "epoch": 0.1271186440677966,
      "grad_norm": 0.5025042295455933,
      "learning_rate": 2e-05,
      "loss": 0.9302,
      "step": 1530
    },
    {
      "epoch": 0.12794948487869723,
      "grad_norm": 0.43608036637306213,
      "learning_rate": 2e-05,
      "loss": 0.9365,
      "step": 1540
    },
    {
      "epoch": 0.12878032568959788,
      "grad_norm": 0.47563910484313965,
      "learning_rate": 2e-05,
      "loss": 0.9342,
      "step": 1550
    },
    {
      "epoch": 0.1296111665004985,
      "grad_norm": 0.5102983713150024,
      "learning_rate": 2e-05,
      "loss": 0.9025,
      "step": 1560
    },
    {
      "epoch": 0.13044200731139913,
      "grad_norm": 0.476362019777298,
      "learning_rate": 2e-05,
      "loss": 0.9053,
      "step": 1570
    },
    {
      "epoch": 0.13127284812229978,
      "grad_norm": 0.44051846861839294,
      "learning_rate": 2e-05,
      "loss": 0.9334,
      "step": 1580
    },
    {
      "epoch": 0.1321036889332004,
      "grad_norm": 0.49285897612571716,
      "learning_rate": 2e-05,
      "loss": 0.9103,
      "step": 1590
    },
    {
      "epoch": 0.13293452974410103,
      "grad_norm": 0.47332510352134705,
      "learning_rate": 2e-05,
      "loss": 0.927,
      "step": 1600
    },
    {
      "epoch": 0.13376537055500165,
      "grad_norm": 0.430770605802536,
      "learning_rate": 2e-05,
      "loss": 0.9081,
      "step": 1610
    },
    {
      "epoch": 0.1345962113659023,
      "grad_norm": 0.45503318309783936,
      "learning_rate": 2e-05,
      "loss": 0.9614,
      "step": 1620
    },
    {
      "epoch": 0.13542705217680293,
      "grad_norm": 0.4389674961566925,
      "learning_rate": 2e-05,
      "loss": 0.9343,
      "step": 1630
    },
    {
      "epoch": 0.13625789298770355,
      "grad_norm": 0.4448428750038147,
      "learning_rate": 2e-05,
      "loss": 0.9251,
      "step": 1640
    },
    {
      "epoch": 0.1370887337986042,
      "grad_norm": 0.4891062378883362,
      "learning_rate": 2e-05,
      "loss": 0.9332,
      "step": 1650
    },
    {
      "epoch": 0.13791957460950482,
      "grad_norm": 0.43644747138023376,
      "learning_rate": 2e-05,
      "loss": 0.9162,
      "step": 1660
    },
    {
      "epoch": 0.13875041542040545,
      "grad_norm": 0.5883463621139526,
      "learning_rate": 2e-05,
      "loss": 0.9069,
      "step": 1670
    },
    {
      "epoch": 0.13958125623130607,
      "grad_norm": 0.42841386795043945,
      "learning_rate": 2e-05,
      "loss": 0.8869,
      "step": 1680
    },
    {
      "epoch": 0.14041209704220672,
      "grad_norm": 0.43758800625801086,
      "learning_rate": 2e-05,
      "loss": 0.9047,
      "step": 1690
    },
    {
      "epoch": 0.14124293785310735,
      "grad_norm": 0.9693111181259155,
      "learning_rate": 2e-05,
      "loss": 0.9182,
      "step": 1700
    },
    {
      "epoch": 0.14207377866400797,
      "grad_norm": 0.46193361282348633,
      "learning_rate": 2e-05,
      "loss": 0.9201,
      "step": 1710
    },
    {
      "epoch": 0.1429046194749086,
      "grad_norm": 0.4599347412586212,
      "learning_rate": 2e-05,
      "loss": 0.8987,
      "step": 1720
    },
    {
      "epoch": 0.14373546028580925,
      "grad_norm": 0.5046244263648987,
      "learning_rate": 2e-05,
      "loss": 0.9489,
      "step": 1730
    },
    {
      "epoch": 0.14456630109670987,
      "grad_norm": 0.465670645236969,
      "learning_rate": 2e-05,
      "loss": 0.9007,
      "step": 1740
    },
    {
      "epoch": 0.1453971419076105,
      "grad_norm": 0.44361841678619385,
      "learning_rate": 2e-05,
      "loss": 0.9092,
      "step": 1750
    },
    {
      "epoch": 0.14622798271851115,
      "grad_norm": 0.46966323256492615,
      "learning_rate": 2e-05,
      "loss": 0.9422,
      "step": 1760
    },
    {
      "epoch": 0.14705882352941177,
      "grad_norm": 0.42061373591423035,
      "learning_rate": 2e-05,
      "loss": 0.9304,
      "step": 1770
    },
    {
      "epoch": 0.1478896643403124,
      "grad_norm": 0.46718308329582214,
      "learning_rate": 2e-05,
      "loss": 0.9496,
      "step": 1780
    },
    {
      "epoch": 0.14872050515121302,
      "grad_norm": 0.44790440797805786,
      "learning_rate": 2e-05,
      "loss": 0.8873,
      "step": 1790
    },
    {
      "epoch": 0.14955134596211367,
      "grad_norm": 0.4441417455673218,
      "learning_rate": 2e-05,
      "loss": 0.9192,
      "step": 1800
    },
    {
      "epoch": 0.1503821867730143,
      "grad_norm": 0.43312087655067444,
      "learning_rate": 2e-05,
      "loss": 0.9438,
      "step": 1810
    },
    {
      "epoch": 0.15121302758391492,
      "grad_norm": 0.46151110529899597,
      "learning_rate": 2e-05,
      "loss": 0.9246,
      "step": 1820
    },
    {
      "epoch": 0.15204386839481554,
      "grad_norm": 0.44976237416267395,
      "learning_rate": 2e-05,
      "loss": 0.9139,
      "step": 1830
    },
    {
      "epoch": 0.1528747092057162,
      "grad_norm": 0.4504854083061218,
      "learning_rate": 2e-05,
      "loss": 0.9276,
      "step": 1840
    },
    {
      "epoch": 0.15370555001661682,
      "grad_norm": 0.43953531980514526,
      "learning_rate": 2e-05,
      "loss": 0.9201,
      "step": 1850
    },
    {
      "epoch": 0.15453639082751744,
      "grad_norm": 0.4370169937610626,
      "learning_rate": 2e-05,
      "loss": 0.9017,
      "step": 1860
    },
    {
      "epoch": 0.1553672316384181,
      "grad_norm": 0.48096010088920593,
      "learning_rate": 2e-05,
      "loss": 0.9296,
      "step": 1870
    },
    {
      "epoch": 0.15619807244931871,
      "grad_norm": 0.43849658966064453,
      "learning_rate": 2e-05,
      "loss": 0.9317,
      "step": 1880
    },
    {
      "epoch": 0.15702891326021934,
      "grad_norm": 0.42286741733551025,
      "learning_rate": 2e-05,
      "loss": 0.9156,
      "step": 1890
    },
    {
      "epoch": 0.15785975407111996,
      "grad_norm": 0.47379568219184875,
      "learning_rate": 2e-05,
      "loss": 0.927,
      "step": 1900
    },
    {
      "epoch": 0.1586905948820206,
      "grad_norm": 0.49215933680534363,
      "learning_rate": 2e-05,
      "loss": 0.902,
      "step": 1910
    },
    {
      "epoch": 0.15952143569292124,
      "grad_norm": 1.1707018613815308,
      "learning_rate": 2e-05,
      "loss": 0.8982,
      "step": 1920
    },
    {
      "epoch": 0.16035227650382186,
      "grad_norm": 0.42774245142936707,
      "learning_rate": 2e-05,
      "loss": 0.9225,
      "step": 1930
    },
    {
      "epoch": 0.1611831173147225,
      "grad_norm": 0.46903279423713684,
      "learning_rate": 2e-05,
      "loss": 0.9021,
      "step": 1940
    },
    {
      "epoch": 0.16201395812562314,
      "grad_norm": 0.43232014775276184,
      "learning_rate": 2e-05,
      "loss": 0.9002,
      "step": 1950
    },
    {
      "epoch": 0.16284479893652376,
      "grad_norm": 0.4257102608680725,
      "learning_rate": 2e-05,
      "loss": 0.9045,
      "step": 1960
    },
    {
      "epoch": 0.16367563974742438,
      "grad_norm": 0.4273391664028168,
      "learning_rate": 2e-05,
      "loss": 0.9145,
      "step": 1970
    },
    {
      "epoch": 0.16450648055832504,
      "grad_norm": 0.4816047251224518,
      "learning_rate": 2e-05,
      "loss": 0.9364,
      "step": 1980
    },
    {
      "epoch": 0.16533732136922566,
      "grad_norm": 0.48251327872276306,
      "learning_rate": 2e-05,
      "loss": 0.9033,
      "step": 1990
    },
    {
      "epoch": 0.16616816218012628,
      "grad_norm": 0.4153393507003784,
      "learning_rate": 2e-05,
      "loss": 0.9192,
      "step": 2000
    },
    {
      "epoch": 0.1669990029910269,
      "grad_norm": 0.4360354542732239,
      "learning_rate": 2e-05,
      "loss": 0.9344,
      "step": 2010
    },
    {
      "epoch": 0.16782984380192756,
      "grad_norm": 0.4371736943721771,
      "learning_rate": 2e-05,
      "loss": 0.9147,
      "step": 2020
    },
    {
      "epoch": 0.16866068461282818,
      "grad_norm": 0.4477806091308594,
      "learning_rate": 2e-05,
      "loss": 0.9057,
      "step": 2030
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 0.42955541610717773,
      "learning_rate": 2e-05,
      "loss": 0.9299,
      "step": 2040
    },
    {
      "epoch": 0.17032236623462946,
      "grad_norm": 0.43824130296707153,
      "learning_rate": 2e-05,
      "loss": 0.9161,
      "step": 2050
    },
    {
      "epoch": 0.17115320704553008,
      "grad_norm": 0.4666792154312134,
      "learning_rate": 2e-05,
      "loss": 0.8963,
      "step": 2060
    },
    {
      "epoch": 0.1719840478564307,
      "grad_norm": 0.43251997232437134,
      "learning_rate": 2e-05,
      "loss": 0.9265,
      "step": 2070
    },
    {
      "epoch": 0.17281488866733133,
      "grad_norm": 0.4566524922847748,
      "learning_rate": 2e-05,
      "loss": 0.9352,
      "step": 2080
    },
    {
      "epoch": 0.17364572947823198,
      "grad_norm": 0.4343387484550476,
      "learning_rate": 2e-05,
      "loss": 0.8882,
      "step": 2090
    },
    {
      "epoch": 0.1744765702891326,
      "grad_norm": 0.42792704701423645,
      "learning_rate": 2e-05,
      "loss": 0.9164,
      "step": 2100
    },
    {
      "epoch": 0.17530741110003323,
      "grad_norm": 0.5312557816505432,
      "learning_rate": 2e-05,
      "loss": 0.9243,
      "step": 2110
    },
    {
      "epoch": 0.17613825191093385,
      "grad_norm": 0.5039758086204529,
      "learning_rate": 2e-05,
      "loss": 0.9377,
      "step": 2120
    },
    {
      "epoch": 0.1769690927218345,
      "grad_norm": 0.4619676172733307,
      "learning_rate": 2e-05,
      "loss": 0.8889,
      "step": 2130
    },
    {
      "epoch": 0.17779993353273513,
      "grad_norm": 0.43388980627059937,
      "learning_rate": 2e-05,
      "loss": 0.91,
      "step": 2140
    },
    {
      "epoch": 0.17863077434363575,
      "grad_norm": 0.4530365765094757,
      "learning_rate": 2e-05,
      "loss": 0.874,
      "step": 2150
    },
    {
      "epoch": 0.1794616151545364,
      "grad_norm": 0.4473583400249481,
      "learning_rate": 2e-05,
      "loss": 0.9083,
      "step": 2160
    },
    {
      "epoch": 0.18029245596543703,
      "grad_norm": 0.4544511139392853,
      "learning_rate": 2e-05,
      "loss": 0.8851,
      "step": 2170
    },
    {
      "epoch": 0.18112329677633765,
      "grad_norm": 0.5991601943969727,
      "learning_rate": 2e-05,
      "loss": 0.9085,
      "step": 2180
    },
    {
      "epoch": 0.18195413758723827,
      "grad_norm": 0.4439004361629486,
      "learning_rate": 2e-05,
      "loss": 0.9056,
      "step": 2190
    },
    {
      "epoch": 0.18278497839813893,
      "grad_norm": 0.4527803659439087,
      "learning_rate": 2e-05,
      "loss": 0.9233,
      "step": 2200
    },
    {
      "epoch": 0.18361581920903955,
      "grad_norm": 0.4314718544483185,
      "learning_rate": 2e-05,
      "loss": 0.8999,
      "step": 2210
    },
    {
      "epoch": 0.18444666001994017,
      "grad_norm": 0.43120718002319336,
      "learning_rate": 2e-05,
      "loss": 0.8843,
      "step": 2220
    },
    {
      "epoch": 0.18527750083084082,
      "grad_norm": 0.4640350937843323,
      "learning_rate": 2e-05,
      "loss": 0.922,
      "step": 2230
    },
    {
      "epoch": 0.18610834164174145,
      "grad_norm": 0.46143069863319397,
      "learning_rate": 2e-05,
      "loss": 0.8954,
      "step": 2240
    },
    {
      "epoch": 0.18693918245264207,
      "grad_norm": 0.44617486000061035,
      "learning_rate": 2e-05,
      "loss": 0.8752,
      "step": 2250
    },
    {
      "epoch": 0.1877700232635427,
      "grad_norm": 0.4476693868637085,
      "learning_rate": 2e-05,
      "loss": 0.9329,
      "step": 2260
    },
    {
      "epoch": 0.18860086407444335,
      "grad_norm": 0.44085007905960083,
      "learning_rate": 2e-05,
      "loss": 0.9211,
      "step": 2270
    },
    {
      "epoch": 0.18943170488534397,
      "grad_norm": 0.479470819234848,
      "learning_rate": 2e-05,
      "loss": 0.9123,
      "step": 2280
    },
    {
      "epoch": 0.1902625456962446,
      "grad_norm": 0.5018478631973267,
      "learning_rate": 2e-05,
      "loss": 0.8975,
      "step": 2290
    },
    {
      "epoch": 0.19109338650714522,
      "grad_norm": 0.39884376525878906,
      "learning_rate": 2e-05,
      "loss": 0.9007,
      "step": 2300
    },
    {
      "epoch": 0.19192422731804587,
      "grad_norm": 0.47602373361587524,
      "learning_rate": 2e-05,
      "loss": 0.8646,
      "step": 2310
    },
    {
      "epoch": 0.1927550681289465,
      "grad_norm": 0.42797547578811646,
      "learning_rate": 2e-05,
      "loss": 0.8983,
      "step": 2320
    },
    {
      "epoch": 0.19358590893984712,
      "grad_norm": 0.44223758578300476,
      "learning_rate": 2e-05,
      "loss": 0.9237,
      "step": 2330
    },
    {
      "epoch": 0.19441674975074777,
      "grad_norm": 0.44112953543663025,
      "learning_rate": 2e-05,
      "loss": 0.9164,
      "step": 2340
    },
    {
      "epoch": 0.1952475905616484,
      "grad_norm": 0.4872469902038574,
      "learning_rate": 2e-05,
      "loss": 0.8894,
      "step": 2350
    },
    {
      "epoch": 0.19607843137254902,
      "grad_norm": 0.4467613995075226,
      "learning_rate": 2e-05,
      "loss": 0.9153,
      "step": 2360
    },
    {
      "epoch": 0.19690927218344964,
      "grad_norm": 0.4924972951412201,
      "learning_rate": 2e-05,
      "loss": 0.8792,
      "step": 2370
    },
    {
      "epoch": 0.1977401129943503,
      "grad_norm": 0.40645840764045715,
      "learning_rate": 2e-05,
      "loss": 0.9119,
      "step": 2380
    },
    {
      "epoch": 0.19857095380525092,
      "grad_norm": 0.43075770139694214,
      "learning_rate": 2e-05,
      "loss": 0.928,
      "step": 2390
    },
    {
      "epoch": 0.19940179461615154,
      "grad_norm": 0.44500741362571716,
      "learning_rate": 2e-05,
      "loss": 0.9571,
      "step": 2400
    },
    {
      "epoch": 0.20023263542705216,
      "grad_norm": 0.41890957951545715,
      "learning_rate": 2e-05,
      "loss": 0.876,
      "step": 2410
    },
    {
      "epoch": 0.20106347623795282,
      "grad_norm": 0.5513232946395874,
      "learning_rate": 2e-05,
      "loss": 0.8904,
      "step": 2420
    },
    {
      "epoch": 0.20189431704885344,
      "grad_norm": 0.438161700963974,
      "learning_rate": 2e-05,
      "loss": 0.8692,
      "step": 2430
    },
    {
      "epoch": 0.20272515785975406,
      "grad_norm": 0.9411785006523132,
      "learning_rate": 2e-05,
      "loss": 0.8769,
      "step": 2440
    },
    {
      "epoch": 0.20355599867065471,
      "grad_norm": 0.4301227033138275,
      "learning_rate": 2e-05,
      "loss": 0.8652,
      "step": 2450
    },
    {
      "epoch": 0.20438683948155534,
      "grad_norm": 0.4590553641319275,
      "learning_rate": 2e-05,
      "loss": 0.9044,
      "step": 2460
    },
    {
      "epoch": 0.20521768029245596,
      "grad_norm": 0.4509306848049164,
      "learning_rate": 2e-05,
      "loss": 0.8709,
      "step": 2470
    },
    {
      "epoch": 0.20604852110335659,
      "grad_norm": 0.408542662858963,
      "learning_rate": 2e-05,
      "loss": 0.8983,
      "step": 2480
    },
    {
      "epoch": 0.20687936191425724,
      "grad_norm": 0.4729969799518585,
      "learning_rate": 2e-05,
      "loss": 0.8736,
      "step": 2490
    },
    {
      "epoch": 0.20771020272515786,
      "grad_norm": 0.464586466550827,
      "learning_rate": 2e-05,
      "loss": 0.8955,
      "step": 2500
    },
    {
      "epoch": 0.20854104353605848,
      "grad_norm": 0.8892510533332825,
      "learning_rate": 2e-05,
      "loss": 0.8964,
      "step": 2510
    },
    {
      "epoch": 0.20937188434695914,
      "grad_norm": 0.4715708792209625,
      "learning_rate": 2e-05,
      "loss": 0.8743,
      "step": 2520
    },
    {
      "epoch": 0.21020272515785976,
      "grad_norm": 0.5275328159332275,
      "learning_rate": 2e-05,
      "loss": 0.8951,
      "step": 2530
    },
    {
      "epoch": 0.21103356596876038,
      "grad_norm": 0.45450401306152344,
      "learning_rate": 2e-05,
      "loss": 0.8803,
      "step": 2540
    },
    {
      "epoch": 0.211864406779661,
      "grad_norm": 0.4248448610305786,
      "learning_rate": 2e-05,
      "loss": 0.9009,
      "step": 2550
    },
    {
      "epoch": 0.21269524759056166,
      "grad_norm": 0.46027621626853943,
      "learning_rate": 2e-05,
      "loss": 0.8797,
      "step": 2560
    },
    {
      "epoch": 0.21352608840146228,
      "grad_norm": 0.43015068769454956,
      "learning_rate": 2e-05,
      "loss": 0.8861,
      "step": 2570
    },
    {
      "epoch": 0.2143569292123629,
      "grad_norm": 0.426276832818985,
      "learning_rate": 2e-05,
      "loss": 0.9279,
      "step": 2580
    },
    {
      "epoch": 0.21518777002326353,
      "grad_norm": 0.474142462015152,
      "learning_rate": 2e-05,
      "loss": 0.9266,
      "step": 2590
    },
    {
      "epoch": 0.21601861083416418,
      "grad_norm": 0.4561924338340759,
      "learning_rate": 2e-05,
      "loss": 0.9178,
      "step": 2600
    },
    {
      "epoch": 0.2168494516450648,
      "grad_norm": 0.48385292291641235,
      "learning_rate": 2e-05,
      "loss": 0.8601,
      "step": 2610
    },
    {
      "epoch": 0.21768029245596543,
      "grad_norm": 0.43850117921829224,
      "learning_rate": 2e-05,
      "loss": 0.902,
      "step": 2620
    },
    {
      "epoch": 0.21851113326686608,
      "grad_norm": 0.40800923109054565,
      "learning_rate": 2e-05,
      "loss": 0.9076,
      "step": 2630
    },
    {
      "epoch": 0.2193419740777667,
      "grad_norm": 0.4295235872268677,
      "learning_rate": 2e-05,
      "loss": 0.9069,
      "step": 2640
    },
    {
      "epoch": 0.22017281488866733,
      "grad_norm": 0.45928657054901123,
      "learning_rate": 2e-05,
      "loss": 0.8739,
      "step": 2650
    },
    {
      "epoch": 0.22100365569956795,
      "grad_norm": 0.45654815435409546,
      "learning_rate": 2e-05,
      "loss": 0.9172,
      "step": 2660
    },
    {
      "epoch": 0.2218344965104686,
      "grad_norm": 0.43029630184173584,
      "learning_rate": 2e-05,
      "loss": 0.8795,
      "step": 2670
    },
    {
      "epoch": 0.22266533732136923,
      "grad_norm": 0.5521813035011292,
      "learning_rate": 2e-05,
      "loss": 0.9136,
      "step": 2680
    },
    {
      "epoch": 0.22349617813226985,
      "grad_norm": 0.4297541081905365,
      "learning_rate": 2e-05,
      "loss": 0.8837,
      "step": 2690
    },
    {
      "epoch": 0.22432701894317048,
      "grad_norm": 0.4344077706336975,
      "learning_rate": 2e-05,
      "loss": 0.8977,
      "step": 2700
    },
    {
      "epoch": 0.22515785975407113,
      "grad_norm": 0.4418643116950989,
      "learning_rate": 2e-05,
      "loss": 0.9101,
      "step": 2710
    },
    {
      "epoch": 0.22598870056497175,
      "grad_norm": 0.4646466374397278,
      "learning_rate": 2e-05,
      "loss": 0.9439,
      "step": 2720
    },
    {
      "epoch": 0.22681954137587237,
      "grad_norm": 0.4619177579879761,
      "learning_rate": 2e-05,
      "loss": 0.9132,
      "step": 2730
    },
    {
      "epoch": 0.22765038218677303,
      "grad_norm": 0.45407360792160034,
      "learning_rate": 2e-05,
      "loss": 0.8676,
      "step": 2740
    },
    {
      "epoch": 0.22848122299767365,
      "grad_norm": 0.5395656824111938,
      "learning_rate": 2e-05,
      "loss": 0.8616,
      "step": 2750
    },
    {
      "epoch": 0.22931206380857427,
      "grad_norm": 0.4302252531051636,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 2760
    },
    {
      "epoch": 0.2301429046194749,
      "grad_norm": 0.433845579624176,
      "learning_rate": 2e-05,
      "loss": 0.8942,
      "step": 2770
    },
    {
      "epoch": 0.23097374543037555,
      "grad_norm": 0.46152353286743164,
      "learning_rate": 2e-05,
      "loss": 0.8779,
      "step": 2780
    },
    {
      "epoch": 0.23180458624127617,
      "grad_norm": 0.48064422607421875,
      "learning_rate": 2e-05,
      "loss": 0.9175,
      "step": 2790
    },
    {
      "epoch": 0.2326354270521768,
      "grad_norm": 0.6684866547584534,
      "learning_rate": 2e-05,
      "loss": 0.9008,
      "step": 2800
    },
    {
      "epoch": 0.23346626786307745,
      "grad_norm": 0.40702152252197266,
      "learning_rate": 2e-05,
      "loss": 0.8972,
      "step": 2810
    },
    {
      "epoch": 0.23429710867397807,
      "grad_norm": 0.44743430614471436,
      "learning_rate": 2e-05,
      "loss": 0.883,
      "step": 2820
    },
    {
      "epoch": 0.2351279494848787,
      "grad_norm": 0.4514210820198059,
      "learning_rate": 2e-05,
      "loss": 0.927,
      "step": 2830
    },
    {
      "epoch": 0.23595879029577932,
      "grad_norm": 0.43500110507011414,
      "learning_rate": 2e-05,
      "loss": 0.907,
      "step": 2840
    },
    {
      "epoch": 0.23678963110667997,
      "grad_norm": 0.4374711215496063,
      "learning_rate": 2e-05,
      "loss": 0.8988,
      "step": 2850
    },
    {
      "epoch": 0.2376204719175806,
      "grad_norm": 0.5247493982315063,
      "learning_rate": 2e-05,
      "loss": 0.8627,
      "step": 2860
    },
    {
      "epoch": 0.23845131272848122,
      "grad_norm": 0.45627421140670776,
      "learning_rate": 2e-05,
      "loss": 0.8707,
      "step": 2870
    },
    {
      "epoch": 0.23928215353938184,
      "grad_norm": 0.4203733205795288,
      "learning_rate": 2e-05,
      "loss": 0.8884,
      "step": 2880
    },
    {
      "epoch": 0.2401129943502825,
      "grad_norm": 0.5086071491241455,
      "learning_rate": 2e-05,
      "loss": 0.8854,
      "step": 2890
    },
    {
      "epoch": 0.24094383516118312,
      "grad_norm": 0.476149320602417,
      "learning_rate": 2e-05,
      "loss": 0.8836,
      "step": 2900
    },
    {
      "epoch": 0.24177467597208374,
      "grad_norm": 0.48206570744514465,
      "learning_rate": 2e-05,
      "loss": 0.8998,
      "step": 2910
    },
    {
      "epoch": 0.2426055167829844,
      "grad_norm": 0.43814602494239807,
      "learning_rate": 2e-05,
      "loss": 0.9083,
      "step": 2920
    },
    {
      "epoch": 0.24343635759388502,
      "grad_norm": 0.4566115438938141,
      "learning_rate": 2e-05,
      "loss": 0.8786,
      "step": 2930
    },
    {
      "epoch": 0.24426719840478564,
      "grad_norm": 0.4960267245769501,
      "learning_rate": 2e-05,
      "loss": 0.9021,
      "step": 2940
    },
    {
      "epoch": 0.24509803921568626,
      "grad_norm": 0.4436706304550171,
      "learning_rate": 2e-05,
      "loss": 0.8826,
      "step": 2950
    },
    {
      "epoch": 0.24592888002658692,
      "grad_norm": 0.42017072439193726,
      "learning_rate": 2e-05,
      "loss": 0.9081,
      "step": 2960
    },
    {
      "epoch": 0.24675972083748754,
      "grad_norm": 0.42673128843307495,
      "learning_rate": 2e-05,
      "loss": 0.8833,
      "step": 2970
    },
    {
      "epoch": 0.24759056164838816,
      "grad_norm": 0.4424765706062317,
      "learning_rate": 2e-05,
      "loss": 0.9011,
      "step": 2980
    },
    {
      "epoch": 0.2484214024592888,
      "grad_norm": 0.4783121943473816,
      "learning_rate": 2e-05,
      "loss": 0.9228,
      "step": 2990
    },
    {
      "epoch": 0.24925224327018944,
      "grad_norm": 0.4395361542701721,
      "learning_rate": 2e-05,
      "loss": 0.8867,
      "step": 3000
    },
    {
      "epoch": 0.2500830840810901,
      "grad_norm": 0.4635724723339081,
      "learning_rate": 2e-05,
      "loss": 0.9109,
      "step": 3010
    },
    {
      "epoch": 0.2509139248919907,
      "grad_norm": 0.4547848105430603,
      "learning_rate": 2e-05,
      "loss": 0.8768,
      "step": 3020
    },
    {
      "epoch": 0.25174476570289134,
      "grad_norm": 0.4364224970340729,
      "learning_rate": 2e-05,
      "loss": 0.864,
      "step": 3030
    },
    {
      "epoch": 0.25257560651379196,
      "grad_norm": 0.8048545122146606,
      "learning_rate": 2e-05,
      "loss": 0.8595,
      "step": 3040
    },
    {
      "epoch": 0.2534064473246926,
      "grad_norm": 0.4434371590614319,
      "learning_rate": 2e-05,
      "loss": 0.8689,
      "step": 3050
    },
    {
      "epoch": 0.2542372881355932,
      "grad_norm": 0.4406764507293701,
      "learning_rate": 2e-05,
      "loss": 0.862,
      "step": 3060
    },
    {
      "epoch": 0.25506812894649383,
      "grad_norm": 0.44526636600494385,
      "learning_rate": 2e-05,
      "loss": 0.87,
      "step": 3070
    },
    {
      "epoch": 0.25589896975739446,
      "grad_norm": 0.42892077565193176,
      "learning_rate": 2e-05,
      "loss": 0.9062,
      "step": 3080
    },
    {
      "epoch": 0.25672981056829514,
      "grad_norm": 0.4292849600315094,
      "learning_rate": 2e-05,
      "loss": 0.8766,
      "step": 3090
    },
    {
      "epoch": 0.25756065137919576,
      "grad_norm": 0.4798363149166107,
      "learning_rate": 2e-05,
      "loss": 0.8992,
      "step": 3100
    },
    {
      "epoch": 0.2583914921900964,
      "grad_norm": 0.4330069124698639,
      "learning_rate": 2e-05,
      "loss": 0.9048,
      "step": 3110
    },
    {
      "epoch": 0.259222333000997,
      "grad_norm": 0.45333030819892883,
      "learning_rate": 2e-05,
      "loss": 0.8766,
      "step": 3120
    },
    {
      "epoch": 0.26005317381189763,
      "grad_norm": 0.4337524175643921,
      "learning_rate": 2e-05,
      "loss": 0.882,
      "step": 3130
    },
    {
      "epoch": 0.26088401462279825,
      "grad_norm": 0.43314796686172485,
      "learning_rate": 2e-05,
      "loss": 0.8849,
      "step": 3140
    },
    {
      "epoch": 0.2617148554336989,
      "grad_norm": 0.43416863679885864,
      "learning_rate": 2e-05,
      "loss": 0.8747,
      "step": 3150
    },
    {
      "epoch": 0.26254569624459956,
      "grad_norm": 0.4168887734413147,
      "learning_rate": 2e-05,
      "loss": 0.8499,
      "step": 3160
    },
    {
      "epoch": 0.2633765370555002,
      "grad_norm": 0.40954482555389404,
      "learning_rate": 2e-05,
      "loss": 0.8653,
      "step": 3170
    },
    {
      "epoch": 0.2642073778664008,
      "grad_norm": 0.4521748423576355,
      "learning_rate": 2e-05,
      "loss": 0.9025,
      "step": 3180
    },
    {
      "epoch": 0.26503821867730143,
      "grad_norm": 0.45301973819732666,
      "learning_rate": 2e-05,
      "loss": 0.8713,
      "step": 3190
    },
    {
      "epoch": 0.26586905948820205,
      "grad_norm": 0.4402906894683838,
      "learning_rate": 2e-05,
      "loss": 0.8804,
      "step": 3200
    },
    {
      "epoch": 0.2666999002991027,
      "grad_norm": 0.4207035303115845,
      "learning_rate": 2e-05,
      "loss": 0.912,
      "step": 3210
    },
    {
      "epoch": 0.2675307411100033,
      "grad_norm": 0.45026668906211853,
      "learning_rate": 2e-05,
      "loss": 0.8981,
      "step": 3220
    },
    {
      "epoch": 0.268361581920904,
      "grad_norm": 0.4424113631248474,
      "learning_rate": 2e-05,
      "loss": 0.8791,
      "step": 3230
    },
    {
      "epoch": 0.2691924227318046,
      "grad_norm": 0.4415736496448517,
      "learning_rate": 2e-05,
      "loss": 0.8486,
      "step": 3240
    },
    {
      "epoch": 0.2700232635427052,
      "grad_norm": 0.43698400259017944,
      "learning_rate": 2e-05,
      "loss": 0.8804,
      "step": 3250
    },
    {
      "epoch": 0.27085410435360585,
      "grad_norm": 0.4405103325843811,
      "learning_rate": 2e-05,
      "loss": 0.8851,
      "step": 3260
    },
    {
      "epoch": 0.2716849451645065,
      "grad_norm": 0.4531089663505554,
      "learning_rate": 2e-05,
      "loss": 0.8882,
      "step": 3270
    },
    {
      "epoch": 0.2725157859754071,
      "grad_norm": 0.4532719850540161,
      "learning_rate": 2e-05,
      "loss": 0.8682,
      "step": 3280
    },
    {
      "epoch": 0.2733466267863077,
      "grad_norm": 0.428219735622406,
      "learning_rate": 2e-05,
      "loss": 0.8852,
      "step": 3290
    },
    {
      "epoch": 0.2741774675972084,
      "grad_norm": 0.438568651676178,
      "learning_rate": 2e-05,
      "loss": 0.8706,
      "step": 3300
    },
    {
      "epoch": 0.275008308408109,
      "grad_norm": 0.43582403659820557,
      "learning_rate": 2e-05,
      "loss": 0.8694,
      "step": 3310
    },
    {
      "epoch": 0.27583914921900965,
      "grad_norm": 0.446028470993042,
      "learning_rate": 2e-05,
      "loss": 0.8958,
      "step": 3320
    },
    {
      "epoch": 0.2766699900299103,
      "grad_norm": 0.4281398355960846,
      "learning_rate": 2e-05,
      "loss": 0.907,
      "step": 3330
    },
    {
      "epoch": 0.2775008308408109,
      "grad_norm": 0.4182804524898529,
      "learning_rate": 2e-05,
      "loss": 0.9083,
      "step": 3340
    },
    {
      "epoch": 0.2783316716517115,
      "grad_norm": 0.39218708872795105,
      "learning_rate": 2e-05,
      "loss": 0.8431,
      "step": 3350
    },
    {
      "epoch": 0.27916251246261214,
      "grad_norm": 0.4490000009536743,
      "learning_rate": 2e-05,
      "loss": 0.8698,
      "step": 3360
    },
    {
      "epoch": 0.27999335327351277,
      "grad_norm": 0.4409296214580536,
      "learning_rate": 2e-05,
      "loss": 0.9039,
      "step": 3370
    },
    {
      "epoch": 0.28082419408441345,
      "grad_norm": 0.4960165321826935,
      "learning_rate": 2e-05,
      "loss": 0.8659,
      "step": 3380
    },
    {
      "epoch": 0.28165503489531407,
      "grad_norm": 0.4306507408618927,
      "learning_rate": 2e-05,
      "loss": 0.8835,
      "step": 3390
    },
    {
      "epoch": 0.2824858757062147,
      "grad_norm": 0.4303608238697052,
      "learning_rate": 2e-05,
      "loss": 0.8628,
      "step": 3400
    },
    {
      "epoch": 0.2833167165171153,
      "grad_norm": 0.4399205446243286,
      "learning_rate": 2e-05,
      "loss": 0.8995,
      "step": 3410
    },
    {
      "epoch": 0.28414755732801594,
      "grad_norm": 0.4601419270038605,
      "learning_rate": 2e-05,
      "loss": 0.8862,
      "step": 3420
    },
    {
      "epoch": 0.28497839813891657,
      "grad_norm": 0.47692254185676575,
      "learning_rate": 2e-05,
      "loss": 0.8722,
      "step": 3430
    },
    {
      "epoch": 0.2858092389498172,
      "grad_norm": 0.4244423508644104,
      "learning_rate": 2e-05,
      "loss": 0.8964,
      "step": 3440
    },
    {
      "epoch": 0.28664007976071787,
      "grad_norm": 0.4808521270751953,
      "learning_rate": 2e-05,
      "loss": 0.8832,
      "step": 3450
    },
    {
      "epoch": 0.2874709205716185,
      "grad_norm": 0.4549659192562103,
      "learning_rate": 2e-05,
      "loss": 0.8646,
      "step": 3460
    },
    {
      "epoch": 0.2883017613825191,
      "grad_norm": 0.42053741216659546,
      "learning_rate": 2e-05,
      "loss": 0.8905,
      "step": 3470
    },
    {
      "epoch": 0.28913260219341974,
      "grad_norm": 0.48250409960746765,
      "learning_rate": 2e-05,
      "loss": 0.885,
      "step": 3480
    },
    {
      "epoch": 0.28996344300432036,
      "grad_norm": 0.41871193051338196,
      "learning_rate": 2e-05,
      "loss": 0.8407,
      "step": 3490
    },
    {
      "epoch": 0.290794283815221,
      "grad_norm": 0.45479318499565125,
      "learning_rate": 2e-05,
      "loss": 0.9334,
      "step": 3500
    },
    {
      "epoch": 0.2916251246261216,
      "grad_norm": 0.4419419765472412,
      "learning_rate": 2e-05,
      "loss": 0.9035,
      "step": 3510
    },
    {
      "epoch": 0.2924559654370223,
      "grad_norm": 0.42783603072166443,
      "learning_rate": 2e-05,
      "loss": 0.8814,
      "step": 3520
    },
    {
      "epoch": 0.2932868062479229,
      "grad_norm": 0.42276325821876526,
      "learning_rate": 2e-05,
      "loss": 0.8669,
      "step": 3530
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 0.44785454869270325,
      "learning_rate": 2e-05,
      "loss": 0.8498,
      "step": 3540
    },
    {
      "epoch": 0.29494848786972416,
      "grad_norm": 0.4347696006298065,
      "learning_rate": 2e-05,
      "loss": 0.9025,
      "step": 3550
    },
    {
      "epoch": 0.2957793286806248,
      "grad_norm": 0.44057634472846985,
      "learning_rate": 2e-05,
      "loss": 0.8653,
      "step": 3560
    },
    {
      "epoch": 0.2966101694915254,
      "grad_norm": 0.4397209584712982,
      "learning_rate": 2e-05,
      "loss": 0.8802,
      "step": 3570
    },
    {
      "epoch": 0.29744101030242603,
      "grad_norm": 0.41466838121414185,
      "learning_rate": 2e-05,
      "loss": 0.9101,
      "step": 3580
    },
    {
      "epoch": 0.2982718511133267,
      "grad_norm": 0.42472904920578003,
      "learning_rate": 2e-05,
      "loss": 0.892,
      "step": 3590
    },
    {
      "epoch": 0.29910269192422734,
      "grad_norm": 0.4503692388534546,
      "learning_rate": 2e-05,
      "loss": 0.8933,
      "step": 3600
    },
    {
      "epoch": 0.29993353273512796,
      "grad_norm": 0.4196890592575073,
      "learning_rate": 2e-05,
      "loss": 0.8599,
      "step": 3610
    },
    {
      "epoch": 0.3007643735460286,
      "grad_norm": 0.45148637890815735,
      "learning_rate": 2e-05,
      "loss": 0.8673,
      "step": 3620
    },
    {
      "epoch": 0.3015952143569292,
      "grad_norm": 0.4308314323425293,
      "learning_rate": 2e-05,
      "loss": 0.8656,
      "step": 3630
    },
    {
      "epoch": 0.30242605516782983,
      "grad_norm": 0.47956836223602295,
      "learning_rate": 2e-05,
      "loss": 0.8776,
      "step": 3640
    },
    {
      "epoch": 0.30325689597873046,
      "grad_norm": 0.4451814293861389,
      "learning_rate": 2e-05,
      "loss": 0.8829,
      "step": 3650
    },
    {
      "epoch": 0.3040877367896311,
      "grad_norm": 0.5133103728294373,
      "learning_rate": 2e-05,
      "loss": 0.8353,
      "step": 3660
    },
    {
      "epoch": 0.30491857760053176,
      "grad_norm": 0.4278886914253235,
      "learning_rate": 2e-05,
      "loss": 0.8594,
      "step": 3670
    },
    {
      "epoch": 0.3057494184114324,
      "grad_norm": 0.4697556495666504,
      "learning_rate": 2e-05,
      "loss": 0.8582,
      "step": 3680
    },
    {
      "epoch": 0.306580259222333,
      "grad_norm": 0.44366103410720825,
      "learning_rate": 2e-05,
      "loss": 0.8917,
      "step": 3690
    },
    {
      "epoch": 0.30741110003323363,
      "grad_norm": 0.4215337038040161,
      "learning_rate": 2e-05,
      "loss": 0.8478,
      "step": 3700
    },
    {
      "epoch": 0.30824194084413425,
      "grad_norm": 0.44416317343711853,
      "learning_rate": 2e-05,
      "loss": 0.8752,
      "step": 3710
    },
    {
      "epoch": 0.3090727816550349,
      "grad_norm": 0.45367932319641113,
      "learning_rate": 2e-05,
      "loss": 0.8768,
      "step": 3720
    },
    {
      "epoch": 0.3099036224659355,
      "grad_norm": 0.4384778141975403,
      "learning_rate": 2e-05,
      "loss": 0.8494,
      "step": 3730
    },
    {
      "epoch": 0.3107344632768362,
      "grad_norm": 0.41297242045402527,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 3740
    },
    {
      "epoch": 0.3115653040877368,
      "grad_norm": 0.4522661566734314,
      "learning_rate": 2e-05,
      "loss": 0.8808,
      "step": 3750
    },
    {
      "epoch": 0.31239614489863743,
      "grad_norm": 0.4485504627227783,
      "learning_rate": 2e-05,
      "loss": 0.8737,
      "step": 3760
    },
    {
      "epoch": 0.31322698570953805,
      "grad_norm": 0.4649835526943207,
      "learning_rate": 2e-05,
      "loss": 0.8838,
      "step": 3770
    },
    {
      "epoch": 0.3140578265204387,
      "grad_norm": 0.5588969588279724,
      "learning_rate": 2e-05,
      "loss": 0.9036,
      "step": 3780
    },
    {
      "epoch": 0.3148886673313393,
      "grad_norm": 0.4562835395336151,
      "learning_rate": 2e-05,
      "loss": 0.8698,
      "step": 3790
    },
    {
      "epoch": 0.3157195081422399,
      "grad_norm": 0.443407267332077,
      "learning_rate": 2e-05,
      "loss": 0.8846,
      "step": 3800
    },
    {
      "epoch": 0.3165503489531406,
      "grad_norm": 0.4531760811805725,
      "learning_rate": 2e-05,
      "loss": 0.9008,
      "step": 3810
    },
    {
      "epoch": 0.3173811897640412,
      "grad_norm": 0.41710126399993896,
      "learning_rate": 2e-05,
      "loss": 0.8412,
      "step": 3820
    },
    {
      "epoch": 0.31821203057494185,
      "grad_norm": 0.45021387934684753,
      "learning_rate": 2e-05,
      "loss": 0.8815,
      "step": 3830
    },
    {
      "epoch": 0.3190428713858425,
      "grad_norm": 0.428801566362381,
      "learning_rate": 2e-05,
      "loss": 0.8211,
      "step": 3840
    },
    {
      "epoch": 0.3198737121967431,
      "grad_norm": 0.4128269553184509,
      "learning_rate": 2e-05,
      "loss": 0.8245,
      "step": 3850
    },
    {
      "epoch": 0.3207045530076437,
      "grad_norm": 0.44466432929039,
      "learning_rate": 2e-05,
      "loss": 0.8796,
      "step": 3860
    },
    {
      "epoch": 0.32153539381854435,
      "grad_norm": 0.5281911492347717,
      "learning_rate": 2e-05,
      "loss": 0.8689,
      "step": 3870
    },
    {
      "epoch": 0.322366234629445,
      "grad_norm": 0.45108282566070557,
      "learning_rate": 2e-05,
      "loss": 0.8826,
      "step": 3880
    },
    {
      "epoch": 0.32319707544034565,
      "grad_norm": 0.45069044828414917,
      "learning_rate": 2e-05,
      "loss": 0.8827,
      "step": 3890
    },
    {
      "epoch": 0.3240279162512463,
      "grad_norm": 0.43937811255455017,
      "learning_rate": 2e-05,
      "loss": 0.8771,
      "step": 3900
    },
    {
      "epoch": 0.3248587570621469,
      "grad_norm": 0.45722702145576477,
      "learning_rate": 2e-05,
      "loss": 0.8526,
      "step": 3910
    },
    {
      "epoch": 0.3256895978730475,
      "grad_norm": 0.4165329337120056,
      "learning_rate": 2e-05,
      "loss": 0.8646,
      "step": 3920
    },
    {
      "epoch": 0.32652043868394814,
      "grad_norm": 0.45522579550743103,
      "learning_rate": 2e-05,
      "loss": 0.8832,
      "step": 3930
    },
    {
      "epoch": 0.32735127949484877,
      "grad_norm": 0.58698970079422,
      "learning_rate": 2e-05,
      "loss": 0.8523,
      "step": 3940
    },
    {
      "epoch": 0.3281821203057494,
      "grad_norm": 0.42000773549079895,
      "learning_rate": 2e-05,
      "loss": 0.8559,
      "step": 3950
    },
    {
      "epoch": 0.32901296111665007,
      "grad_norm": 0.44193536043167114,
      "learning_rate": 2e-05,
      "loss": 0.8639,
      "step": 3960
    },
    {
      "epoch": 0.3298438019275507,
      "grad_norm": 0.44770556688308716,
      "learning_rate": 2e-05,
      "loss": 0.8557,
      "step": 3970
    },
    {
      "epoch": 0.3306746427384513,
      "grad_norm": 0.4489646852016449,
      "learning_rate": 2e-05,
      "loss": 0.8442,
      "step": 3980
    },
    {
      "epoch": 0.33150548354935194,
      "grad_norm": 0.4335518181324005,
      "learning_rate": 2e-05,
      "loss": 0.8332,
      "step": 3990
    },
    {
      "epoch": 0.33233632436025257,
      "grad_norm": 0.4336153566837311,
      "learning_rate": 2e-05,
      "loss": 0.866,
      "step": 4000
    },
    {
      "epoch": 0.3331671651711532,
      "grad_norm": 0.4541057050228119,
      "learning_rate": 2e-05,
      "loss": 0.8441,
      "step": 4010
    },
    {
      "epoch": 0.3339980059820538,
      "grad_norm": 0.446132630109787,
      "learning_rate": 2e-05,
      "loss": 0.8797,
      "step": 4020
    },
    {
      "epoch": 0.3348288467929545,
      "grad_norm": 0.5124335289001465,
      "learning_rate": 2e-05,
      "loss": 0.8776,
      "step": 4030
    },
    {
      "epoch": 0.3356596876038551,
      "grad_norm": 0.4477880299091339,
      "learning_rate": 2e-05,
      "loss": 0.8597,
      "step": 4040
    },
    {
      "epoch": 0.33649052841475574,
      "grad_norm": 0.4752373993396759,
      "learning_rate": 2e-05,
      "loss": 0.8978,
      "step": 4050
    },
    {
      "epoch": 0.33732136922565636,
      "grad_norm": 0.43309518694877625,
      "learning_rate": 2e-05,
      "loss": 0.8385,
      "step": 4060
    },
    {
      "epoch": 0.338152210036557,
      "grad_norm": 0.3936338722705841,
      "learning_rate": 2e-05,
      "loss": 0.8493,
      "step": 4070
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 0.49663209915161133,
      "learning_rate": 2e-05,
      "loss": 0.8602,
      "step": 4080
    },
    {
      "epoch": 0.33981389165835824,
      "grad_norm": 0.4601741433143616,
      "learning_rate": 2e-05,
      "loss": 0.8719,
      "step": 4090
    },
    {
      "epoch": 0.3406447324692589,
      "grad_norm": 0.44826430082321167,
      "learning_rate": 2e-05,
      "loss": 0.8623,
      "step": 4100
    },
    {
      "epoch": 0.34147557328015954,
      "grad_norm": 0.429208904504776,
      "learning_rate": 2e-05,
      "loss": 0.8348,
      "step": 4110
    },
    {
      "epoch": 0.34230641409106016,
      "grad_norm": 0.45465946197509766,
      "learning_rate": 2e-05,
      "loss": 0.8644,
      "step": 4120
    },
    {
      "epoch": 0.3431372549019608,
      "grad_norm": 0.45604151487350464,
      "learning_rate": 2e-05,
      "loss": 0.8414,
      "step": 4130
    },
    {
      "epoch": 0.3439680957128614,
      "grad_norm": 0.4523800313472748,
      "learning_rate": 2e-05,
      "loss": 0.8707,
      "step": 4140
    },
    {
      "epoch": 0.34479893652376203,
      "grad_norm": 0.41325634717941284,
      "learning_rate": 2e-05,
      "loss": 0.8794,
      "step": 4150
    },
    {
      "epoch": 0.34562977733466266,
      "grad_norm": 0.4378925561904907,
      "learning_rate": 2e-05,
      "loss": 0.8544,
      "step": 4160
    },
    {
      "epoch": 0.34646061814556334,
      "grad_norm": 0.441423237323761,
      "learning_rate": 2e-05,
      "loss": 0.8573,
      "step": 4170
    },
    {
      "epoch": 0.34729145895646396,
      "grad_norm": 0.4191168546676636,
      "learning_rate": 2e-05,
      "loss": 0.8557,
      "step": 4180
    },
    {
      "epoch": 0.3481222997673646,
      "grad_norm": 0.4876786172389984,
      "learning_rate": 2e-05,
      "loss": 0.8742,
      "step": 4190
    },
    {
      "epoch": 0.3489531405782652,
      "grad_norm": 0.41200849413871765,
      "learning_rate": 2e-05,
      "loss": 0.8666,
      "step": 4200
    },
    {
      "epoch": 0.34978398138916583,
      "grad_norm": 0.47031447291374207,
      "learning_rate": 2e-05,
      "loss": 0.8715,
      "step": 4210
    },
    {
      "epoch": 0.35061482220006646,
      "grad_norm": 0.41981005668640137,
      "learning_rate": 2e-05,
      "loss": 0.8514,
      "step": 4220
    },
    {
      "epoch": 0.3514456630109671,
      "grad_norm": 0.44124385714530945,
      "learning_rate": 2e-05,
      "loss": 0.8582,
      "step": 4230
    },
    {
      "epoch": 0.3522765038218677,
      "grad_norm": 0.46428897976875305,
      "learning_rate": 2e-05,
      "loss": 0.8527,
      "step": 4240
    },
    {
      "epoch": 0.3531073446327684,
      "grad_norm": 0.43465644121170044,
      "learning_rate": 2e-05,
      "loss": 0.8599,
      "step": 4250
    },
    {
      "epoch": 0.353938185443669,
      "grad_norm": 0.5337467789649963,
      "learning_rate": 2e-05,
      "loss": 0.8367,
      "step": 4260
    },
    {
      "epoch": 0.35476902625456963,
      "grad_norm": 0.4140731990337372,
      "learning_rate": 2e-05,
      "loss": 0.8897,
      "step": 4270
    },
    {
      "epoch": 0.35559986706547025,
      "grad_norm": 0.6828712821006775,
      "learning_rate": 2e-05,
      "loss": 0.8594,
      "step": 4280
    },
    {
      "epoch": 0.3564307078763709,
      "grad_norm": 0.39642083644866943,
      "learning_rate": 2e-05,
      "loss": 0.8438,
      "step": 4290
    },
    {
      "epoch": 0.3572615486872715,
      "grad_norm": 0.47175735235214233,
      "learning_rate": 2e-05,
      "loss": 0.8516,
      "step": 4300
    },
    {
      "epoch": 0.3580923894981721,
      "grad_norm": 0.3988470137119293,
      "learning_rate": 2e-05,
      "loss": 0.8402,
      "step": 4310
    },
    {
      "epoch": 0.3589232303090728,
      "grad_norm": 0.4466840326786041,
      "learning_rate": 2e-05,
      "loss": 0.8456,
      "step": 4320
    },
    {
      "epoch": 0.35975407111997343,
      "grad_norm": 0.4467006325721741,
      "learning_rate": 2e-05,
      "loss": 0.8785,
      "step": 4330
    },
    {
      "epoch": 0.36058491193087405,
      "grad_norm": 0.43134477734565735,
      "learning_rate": 2e-05,
      "loss": 0.8198,
      "step": 4340
    },
    {
      "epoch": 0.3614157527417747,
      "grad_norm": 0.41258856654167175,
      "learning_rate": 2e-05,
      "loss": 0.8609,
      "step": 4350
    },
    {
      "epoch": 0.3622465935526753,
      "grad_norm": 0.4408530592918396,
      "learning_rate": 2e-05,
      "loss": 0.8265,
      "step": 4360
    },
    {
      "epoch": 0.3630774343635759,
      "grad_norm": 0.4380934238433838,
      "learning_rate": 2e-05,
      "loss": 0.8628,
      "step": 4370
    },
    {
      "epoch": 0.36390827517447655,
      "grad_norm": 0.43469446897506714,
      "learning_rate": 2e-05,
      "loss": 0.8568,
      "step": 4380
    },
    {
      "epoch": 0.3647391159853772,
      "grad_norm": 0.8235448002815247,
      "learning_rate": 2e-05,
      "loss": 0.8623,
      "step": 4390
    },
    {
      "epoch": 0.36556995679627785,
      "grad_norm": 0.43067529797554016,
      "learning_rate": 2e-05,
      "loss": 0.8517,
      "step": 4400
    },
    {
      "epoch": 0.3664007976071785,
      "grad_norm": 0.426994651556015,
      "learning_rate": 2e-05,
      "loss": 0.8417,
      "step": 4410
    },
    {
      "epoch": 0.3672316384180791,
      "grad_norm": 0.4426823556423187,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 4420
    },
    {
      "epoch": 0.3680624792289797,
      "grad_norm": 0.4469177722930908,
      "learning_rate": 2e-05,
      "loss": 0.866,
      "step": 4430
    },
    {
      "epoch": 0.36889332003988035,
      "grad_norm": 0.4340355098247528,
      "learning_rate": 2e-05,
      "loss": 0.8658,
      "step": 4440
    },
    {
      "epoch": 0.36972416085078097,
      "grad_norm": 0.5006338953971863,
      "learning_rate": 2e-05,
      "loss": 0.8565,
      "step": 4450
    },
    {
      "epoch": 0.37055500166168165,
      "grad_norm": 0.44370195269584656,
      "learning_rate": 2e-05,
      "loss": 0.8668,
      "step": 4460
    },
    {
      "epoch": 0.3713858424725823,
      "grad_norm": 0.4340578019618988,
      "learning_rate": 2e-05,
      "loss": 0.8598,
      "step": 4470
    },
    {
      "epoch": 0.3722166832834829,
      "grad_norm": 0.41382598876953125,
      "learning_rate": 2e-05,
      "loss": 0.8349,
      "step": 4480
    },
    {
      "epoch": 0.3730475240943835,
      "grad_norm": 0.44862350821495056,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 4490
    },
    {
      "epoch": 0.37387836490528414,
      "grad_norm": 0.6148986220359802,
      "learning_rate": 2e-05,
      "loss": 0.8736,
      "step": 4500
    },
    {
      "epoch": 0.37470920571618477,
      "grad_norm": 0.44492316246032715,
      "learning_rate": 2e-05,
      "loss": 0.835,
      "step": 4510
    },
    {
      "epoch": 0.3755400465270854,
      "grad_norm": 0.40822532773017883,
      "learning_rate": 2e-05,
      "loss": 0.8488,
      "step": 4520
    },
    {
      "epoch": 0.376370887337986,
      "grad_norm": 0.43661847710609436,
      "learning_rate": 2e-05,
      "loss": 0.8681,
      "step": 4530
    },
    {
      "epoch": 0.3772017281488867,
      "grad_norm": 0.4171736240386963,
      "learning_rate": 2e-05,
      "loss": 0.8557,
      "step": 4540
    },
    {
      "epoch": 0.3780325689597873,
      "grad_norm": 0.43618085980415344,
      "learning_rate": 2e-05,
      "loss": 0.8011,
      "step": 4550
    },
    {
      "epoch": 0.37886340977068794,
      "grad_norm": 0.45428308844566345,
      "learning_rate": 2e-05,
      "loss": 0.8433,
      "step": 4560
    },
    {
      "epoch": 0.37969425058158857,
      "grad_norm": 0.4392106235027313,
      "learning_rate": 2e-05,
      "loss": 0.8754,
      "step": 4570
    },
    {
      "epoch": 0.3805250913924892,
      "grad_norm": 0.46240901947021484,
      "learning_rate": 2e-05,
      "loss": 0.8549,
      "step": 4580
    },
    {
      "epoch": 0.3813559322033898,
      "grad_norm": 0.48745596408843994,
      "learning_rate": 2e-05,
      "loss": 0.8406,
      "step": 4590
    },
    {
      "epoch": 0.38218677301429044,
      "grad_norm": 0.4288219213485718,
      "learning_rate": 2e-05,
      "loss": 0.8763,
      "step": 4600
    },
    {
      "epoch": 0.3830176138251911,
      "grad_norm": 0.4440419673919678,
      "learning_rate": 2e-05,
      "loss": 0.8471,
      "step": 4610
    },
    {
      "epoch": 0.38384845463609174,
      "grad_norm": 0.47463056445121765,
      "learning_rate": 2e-05,
      "loss": 0.8634,
      "step": 4620
    },
    {
      "epoch": 0.38467929544699236,
      "grad_norm": 0.40552422404289246,
      "learning_rate": 2e-05,
      "loss": 0.8499,
      "step": 4630
    },
    {
      "epoch": 0.385510136257893,
      "grad_norm": 0.4957606792449951,
      "learning_rate": 2e-05,
      "loss": 0.8603,
      "step": 4640
    },
    {
      "epoch": 0.3863409770687936,
      "grad_norm": 0.42502152919769287,
      "learning_rate": 2e-05,
      "loss": 0.849,
      "step": 4650
    },
    {
      "epoch": 0.38717181787969424,
      "grad_norm": 0.42831119894981384,
      "learning_rate": 2e-05,
      "loss": 0.8047,
      "step": 4660
    },
    {
      "epoch": 0.38800265869059486,
      "grad_norm": 0.40206557512283325,
      "learning_rate": 2e-05,
      "loss": 0.8325,
      "step": 4670
    },
    {
      "epoch": 0.38883349950149554,
      "grad_norm": 0.3957065939903259,
      "learning_rate": 2e-05,
      "loss": 0.8507,
      "step": 4680
    },
    {
      "epoch": 0.38966434031239616,
      "grad_norm": 1.109291911125183,
      "learning_rate": 2e-05,
      "loss": 0.8539,
      "step": 4690
    },
    {
      "epoch": 0.3904951811232968,
      "grad_norm": 0.42580950260162354,
      "learning_rate": 2e-05,
      "loss": 0.8549,
      "step": 4700
    },
    {
      "epoch": 0.3913260219341974,
      "grad_norm": 0.4696245491504669,
      "learning_rate": 2e-05,
      "loss": 0.861,
      "step": 4710
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 0.45528268814086914,
      "learning_rate": 2e-05,
      "loss": 0.8828,
      "step": 4720
    },
    {
      "epoch": 0.39298770355599866,
      "grad_norm": 0.42587751150131226,
      "learning_rate": 2e-05,
      "loss": 0.8806,
      "step": 4730
    },
    {
      "epoch": 0.3938185443668993,
      "grad_norm": 0.41026389598846436,
      "learning_rate": 2e-05,
      "loss": 0.804,
      "step": 4740
    },
    {
      "epoch": 0.39464938517779996,
      "grad_norm": 0.44500595331192017,
      "learning_rate": 2e-05,
      "loss": 0.8713,
      "step": 4750
    },
    {
      "epoch": 0.3954802259887006,
      "grad_norm": 0.4440081715583801,
      "learning_rate": 2e-05,
      "loss": 0.8304,
      "step": 4760
    },
    {
      "epoch": 0.3963110667996012,
      "grad_norm": 0.42570561170578003,
      "learning_rate": 2e-05,
      "loss": 0.8559,
      "step": 4770
    },
    {
      "epoch": 0.39714190761050183,
      "grad_norm": 0.43887215852737427,
      "learning_rate": 2e-05,
      "loss": 0.8491,
      "step": 4780
    },
    {
      "epoch": 0.39797274842140246,
      "grad_norm": 0.4313008189201355,
      "learning_rate": 2e-05,
      "loss": 0.8575,
      "step": 4790
    },
    {
      "epoch": 0.3988035892323031,
      "grad_norm": 0.45144444704055786,
      "learning_rate": 2e-05,
      "loss": 0.8232,
      "step": 4800
    },
    {
      "epoch": 0.3996344300432037,
      "grad_norm": 0.4506703019142151,
      "learning_rate": 2e-05,
      "loss": 0.8469,
      "step": 4810
    },
    {
      "epoch": 0.4004652708541043,
      "grad_norm": 0.5311499238014221,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 4820
    },
    {
      "epoch": 0.401296111665005,
      "grad_norm": 0.42850205302238464,
      "learning_rate": 2e-05,
      "loss": 0.8724,
      "step": 4830
    },
    {
      "epoch": 0.40212695247590563,
      "grad_norm": 0.44533830881118774,
      "learning_rate": 2e-05,
      "loss": 0.8722,
      "step": 4840
    },
    {
      "epoch": 0.40295779328680625,
      "grad_norm": 0.4404807388782501,
      "learning_rate": 2e-05,
      "loss": 0.8198,
      "step": 4850
    },
    {
      "epoch": 0.4037886340977069,
      "grad_norm": 0.6561824679374695,
      "learning_rate": 2e-05,
      "loss": 0.8595,
      "step": 4860
    },
    {
      "epoch": 0.4046194749086075,
      "grad_norm": 0.4359351098537445,
      "learning_rate": 2e-05,
      "loss": 0.8537,
      "step": 4870
    },
    {
      "epoch": 0.4054503157195081,
      "grad_norm": 0.42711949348449707,
      "learning_rate": 2e-05,
      "loss": 0.8761,
      "step": 4880
    },
    {
      "epoch": 0.40628115653040875,
      "grad_norm": 0.42845919728279114,
      "learning_rate": 2e-05,
      "loss": 0.837,
      "step": 4890
    },
    {
      "epoch": 0.40711199734130943,
      "grad_norm": 0.46975457668304443,
      "learning_rate": 2e-05,
      "loss": 0.8659,
      "step": 4900
    },
    {
      "epoch": 0.40794283815221005,
      "grad_norm": 0.45880773663520813,
      "learning_rate": 2e-05,
      "loss": 0.8214,
      "step": 4910
    },
    {
      "epoch": 0.4087736789631107,
      "grad_norm": 0.5536867380142212,
      "learning_rate": 2e-05,
      "loss": 0.8034,
      "step": 4920
    },
    {
      "epoch": 0.4096045197740113,
      "grad_norm": 0.5042794346809387,
      "learning_rate": 2e-05,
      "loss": 0.9134,
      "step": 4930
    },
    {
      "epoch": 0.4104353605849119,
      "grad_norm": 0.46999669075012207,
      "learning_rate": 2e-05,
      "loss": 0.8568,
      "step": 4940
    },
    {
      "epoch": 0.41126620139581255,
      "grad_norm": 0.46141448616981506,
      "learning_rate": 2e-05,
      "loss": 0.8624,
      "step": 4950
    },
    {
      "epoch": 0.41209704220671317,
      "grad_norm": 0.4304838180541992,
      "learning_rate": 2e-05,
      "loss": 0.8539,
      "step": 4960
    },
    {
      "epoch": 0.41292788301761385,
      "grad_norm": 0.659095823764801,
      "learning_rate": 2e-05,
      "loss": 0.8586,
      "step": 4970
    },
    {
      "epoch": 0.4137587238285145,
      "grad_norm": 0.4245087206363678,
      "learning_rate": 2e-05,
      "loss": 0.8245,
      "step": 4980
    },
    {
      "epoch": 0.4145895646394151,
      "grad_norm": 0.4179368317127228,
      "learning_rate": 2e-05,
      "loss": 0.8294,
      "step": 4990
    },
    {
      "epoch": 0.4154204054503157,
      "grad_norm": 0.45675382018089294,
      "learning_rate": 2e-05,
      "loss": 0.8128,
      "step": 5000
    },
    {
      "epoch": 0.41625124626121635,
      "grad_norm": 0.42730990052223206,
      "learning_rate": 2e-05,
      "loss": 0.8588,
      "step": 5010
    },
    {
      "epoch": 0.41708208707211697,
      "grad_norm": 0.4077673852443695,
      "learning_rate": 2e-05,
      "loss": 0.8814,
      "step": 5020
    },
    {
      "epoch": 0.4179129278830176,
      "grad_norm": 0.4508398771286011,
      "learning_rate": 2e-05,
      "loss": 0.8884,
      "step": 5030
    },
    {
      "epoch": 0.4187437686939183,
      "grad_norm": 0.44231757521629333,
      "learning_rate": 2e-05,
      "loss": 0.8469,
      "step": 5040
    },
    {
      "epoch": 0.4195746095048189,
      "grad_norm": 0.5866214036941528,
      "learning_rate": 2e-05,
      "loss": 0.837,
      "step": 5050
    },
    {
      "epoch": 0.4204054503157195,
      "grad_norm": 0.4508447051048279,
      "learning_rate": 2e-05,
      "loss": 0.8685,
      "step": 5060
    },
    {
      "epoch": 0.42123629112662014,
      "grad_norm": 0.4590610861778259,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 5070
    },
    {
      "epoch": 0.42206713193752077,
      "grad_norm": 0.4136819541454315,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 5080
    },
    {
      "epoch": 0.4228979727484214,
      "grad_norm": 0.42763930559158325,
      "learning_rate": 2e-05,
      "loss": 0.8338,
      "step": 5090
    },
    {
      "epoch": 0.423728813559322,
      "grad_norm": 0.44829559326171875,
      "learning_rate": 2e-05,
      "loss": 0.8516,
      "step": 5100
    },
    {
      "epoch": 0.42455965437022264,
      "grad_norm": 0.47433751821517944,
      "learning_rate": 2e-05,
      "loss": 0.8085,
      "step": 5110
    },
    {
      "epoch": 0.4253904951811233,
      "grad_norm": 0.44706180691719055,
      "learning_rate": 2e-05,
      "loss": 0.8268,
      "step": 5120
    },
    {
      "epoch": 0.42622133599202394,
      "grad_norm": 0.48117440938949585,
      "learning_rate": 2e-05,
      "loss": 0.8219,
      "step": 5130
    },
    {
      "epoch": 0.42705217680292457,
      "grad_norm": 0.4610225558280945,
      "learning_rate": 2e-05,
      "loss": 0.8643,
      "step": 5140
    },
    {
      "epoch": 0.4278830176138252,
      "grad_norm": 0.45085418224334717,
      "learning_rate": 2e-05,
      "loss": 0.8376,
      "step": 5150
    },
    {
      "epoch": 0.4287138584247258,
      "grad_norm": 0.4544501304626465,
      "learning_rate": 2e-05,
      "loss": 0.8546,
      "step": 5160
    },
    {
      "epoch": 0.42954469923562644,
      "grad_norm": 0.5397998094558716,
      "learning_rate": 2e-05,
      "loss": 0.8393,
      "step": 5170
    },
    {
      "epoch": 0.43037554004652706,
      "grad_norm": 0.45278045535087585,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 5180
    },
    {
      "epoch": 0.43120638085742774,
      "grad_norm": 0.4161198139190674,
      "learning_rate": 2e-05,
      "loss": 0.8463,
      "step": 5190
    },
    {
      "epoch": 0.43203722166832836,
      "grad_norm": 0.41818422079086304,
      "learning_rate": 2e-05,
      "loss": 0.8484,
      "step": 5200
    },
    {
      "epoch": 0.432868062479229,
      "grad_norm": 0.447200745344162,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 5210
    },
    {
      "epoch": 0.4336989032901296,
      "grad_norm": 0.5460422039031982,
      "learning_rate": 2e-05,
      "loss": 0.8483,
      "step": 5220
    },
    {
      "epoch": 0.43452974410103024,
      "grad_norm": 0.4227336645126343,
      "learning_rate": 2e-05,
      "loss": 0.8547,
      "step": 5230
    },
    {
      "epoch": 0.43536058491193086,
      "grad_norm": 0.4918797016143799,
      "learning_rate": 2e-05,
      "loss": 0.8409,
      "step": 5240
    },
    {
      "epoch": 0.4361914257228315,
      "grad_norm": 0.46162185072898865,
      "learning_rate": 2e-05,
      "loss": 0.8413,
      "step": 5250
    },
    {
      "epoch": 0.43702226653373216,
      "grad_norm": 0.4302051067352295,
      "learning_rate": 2e-05,
      "loss": 0.8508,
      "step": 5260
    },
    {
      "epoch": 0.4378531073446328,
      "grad_norm": 1.755716323852539,
      "learning_rate": 2e-05,
      "loss": 0.8584,
      "step": 5270
    },
    {
      "epoch": 0.4386839481555334,
      "grad_norm": 0.4296351671218872,
      "learning_rate": 2e-05,
      "loss": 0.8591,
      "step": 5280
    },
    {
      "epoch": 0.43951478896643403,
      "grad_norm": 0.4252161979675293,
      "learning_rate": 2e-05,
      "loss": 0.8709,
      "step": 5290
    },
    {
      "epoch": 0.44034562977733466,
      "grad_norm": 0.4581077992916107,
      "learning_rate": 2e-05,
      "loss": 0.8533,
      "step": 5300
    },
    {
      "epoch": 0.4411764705882353,
      "grad_norm": 0.417143315076828,
      "learning_rate": 2e-05,
      "loss": 0.8453,
      "step": 5310
    },
    {
      "epoch": 0.4420073113991359,
      "grad_norm": 0.41382282972335815,
      "learning_rate": 2e-05,
      "loss": 0.8617,
      "step": 5320
    },
    {
      "epoch": 0.4428381522100366,
      "grad_norm": 0.42758339643478394,
      "learning_rate": 2e-05,
      "loss": 0.8744,
      "step": 5330
    },
    {
      "epoch": 0.4436689930209372,
      "grad_norm": 0.4212242662906647,
      "learning_rate": 2e-05,
      "loss": 0.8299,
      "step": 5340
    },
    {
      "epoch": 0.44449983383183783,
      "grad_norm": 0.4060473144054413,
      "learning_rate": 2e-05,
      "loss": 0.8548,
      "step": 5350
    },
    {
      "epoch": 0.44533067464273846,
      "grad_norm": 0.44596362113952637,
      "learning_rate": 2e-05,
      "loss": 0.8528,
      "step": 5360
    },
    {
      "epoch": 0.4461615154536391,
      "grad_norm": 0.4341607391834259,
      "learning_rate": 2e-05,
      "loss": 0.8505,
      "step": 5370
    },
    {
      "epoch": 0.4469923562645397,
      "grad_norm": 0.42946138978004456,
      "learning_rate": 2e-05,
      "loss": 0.8222,
      "step": 5380
    },
    {
      "epoch": 0.4478231970754403,
      "grad_norm": 0.4204101264476776,
      "learning_rate": 2e-05,
      "loss": 0.8492,
      "step": 5390
    },
    {
      "epoch": 0.44865403788634095,
      "grad_norm": 0.40662965178489685,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 5400
    },
    {
      "epoch": 0.44948487869724163,
      "grad_norm": 0.4639713764190674,
      "learning_rate": 2e-05,
      "loss": 0.8413,
      "step": 5410
    },
    {
      "epoch": 0.45031571950814225,
      "grad_norm": 0.42852407693862915,
      "learning_rate": 2e-05,
      "loss": 0.8917,
      "step": 5420
    },
    {
      "epoch": 0.4511465603190429,
      "grad_norm": 0.42004239559173584,
      "learning_rate": 2e-05,
      "loss": 0.8129,
      "step": 5430
    },
    {
      "epoch": 0.4519774011299435,
      "grad_norm": 0.564906895160675,
      "learning_rate": 2e-05,
      "loss": 0.857,
      "step": 5440
    },
    {
      "epoch": 0.4528082419408441,
      "grad_norm": 0.4663470387458801,
      "learning_rate": 2e-05,
      "loss": 0.8494,
      "step": 5450
    },
    {
      "epoch": 0.45363908275174475,
      "grad_norm": 0.43713387846946716,
      "learning_rate": 2e-05,
      "loss": 0.8545,
      "step": 5460
    },
    {
      "epoch": 0.4544699235626454,
      "grad_norm": 0.4284684956073761,
      "learning_rate": 2e-05,
      "loss": 0.8418,
      "step": 5470
    },
    {
      "epoch": 0.45530076437354605,
      "grad_norm": 0.4469447135925293,
      "learning_rate": 2e-05,
      "loss": 0.8563,
      "step": 5480
    },
    {
      "epoch": 0.4561316051844467,
      "grad_norm": 0.43780601024627686,
      "learning_rate": 2e-05,
      "loss": 0.8447,
      "step": 5490
    },
    {
      "epoch": 0.4569624459953473,
      "grad_norm": 0.43914082646369934,
      "learning_rate": 2e-05,
      "loss": 0.8496,
      "step": 5500
    },
    {
      "epoch": 0.4577932868062479,
      "grad_norm": 0.5562605261802673,
      "learning_rate": 2e-05,
      "loss": 0.8452,
      "step": 5510
    },
    {
      "epoch": 0.45862412761714855,
      "grad_norm": 0.4536174535751343,
      "learning_rate": 2e-05,
      "loss": 0.8529,
      "step": 5520
    },
    {
      "epoch": 0.45945496842804917,
      "grad_norm": 0.4398854076862335,
      "learning_rate": 2e-05,
      "loss": 0.8136,
      "step": 5530
    },
    {
      "epoch": 0.4602858092389498,
      "grad_norm": 0.4348509609699249,
      "learning_rate": 2e-05,
      "loss": 0.8935,
      "step": 5540
    },
    {
      "epoch": 0.4611166500498505,
      "grad_norm": 0.42363253235816956,
      "learning_rate": 2e-05,
      "loss": 0.822,
      "step": 5550
    },
    {
      "epoch": 0.4619474908607511,
      "grad_norm": 0.46056774258613586,
      "learning_rate": 2e-05,
      "loss": 0.854,
      "step": 5560
    },
    {
      "epoch": 0.4627783316716517,
      "grad_norm": 0.4273340404033661,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 5570
    },
    {
      "epoch": 0.46360917248255235,
      "grad_norm": 0.45253318548202515,
      "learning_rate": 2e-05,
      "loss": 0.8465,
      "step": 5580
    },
    {
      "epoch": 0.46444001329345297,
      "grad_norm": 0.41663315892219543,
      "learning_rate": 2e-05,
      "loss": 0.841,
      "step": 5590
    },
    {
      "epoch": 0.4652708541043536,
      "grad_norm": 0.4364968240261078,
      "learning_rate": 2e-05,
      "loss": 0.8776,
      "step": 5600
    },
    {
      "epoch": 0.4661016949152542,
      "grad_norm": 0.42386746406555176,
      "learning_rate": 2e-05,
      "loss": 0.8308,
      "step": 5610
    },
    {
      "epoch": 0.4669325357261549,
      "grad_norm": 0.4498637020587921,
      "learning_rate": 2e-05,
      "loss": 0.8198,
      "step": 5620
    },
    {
      "epoch": 0.4677633765370555,
      "grad_norm": 0.4568334221839905,
      "learning_rate": 2e-05,
      "loss": 0.8274,
      "step": 5630
    },
    {
      "epoch": 0.46859421734795614,
      "grad_norm": 0.4365353286266327,
      "learning_rate": 2e-05,
      "loss": 0.8244,
      "step": 5640
    },
    {
      "epoch": 0.46942505815885677,
      "grad_norm": 0.45040884613990784,
      "learning_rate": 2e-05,
      "loss": 0.8237,
      "step": 5650
    },
    {
      "epoch": 0.4702558989697574,
      "grad_norm": 0.42971035838127136,
      "learning_rate": 2e-05,
      "loss": 0.8665,
      "step": 5660
    },
    {
      "epoch": 0.471086739780658,
      "grad_norm": 0.42937299609184265,
      "learning_rate": 2e-05,
      "loss": 0.886,
      "step": 5670
    },
    {
      "epoch": 0.47191758059155864,
      "grad_norm": 0.5421526432037354,
      "learning_rate": 2e-05,
      "loss": 0.8153,
      "step": 5680
    },
    {
      "epoch": 0.47274842140245926,
      "grad_norm": 0.42067772150039673,
      "learning_rate": 2e-05,
      "loss": 0.8376,
      "step": 5690
    },
    {
      "epoch": 0.47357926221335994,
      "grad_norm": 0.4846099019050598,
      "learning_rate": 2e-05,
      "loss": 0.8315,
      "step": 5700
    },
    {
      "epoch": 0.47441010302426057,
      "grad_norm": 0.4187130331993103,
      "learning_rate": 2e-05,
      "loss": 0.8305,
      "step": 5710
    },
    {
      "epoch": 0.4752409438351612,
      "grad_norm": 0.4232124090194702,
      "learning_rate": 2e-05,
      "loss": 0.8365,
      "step": 5720
    },
    {
      "epoch": 0.4760717846460618,
      "grad_norm": 0.48799261450767517,
      "learning_rate": 2e-05,
      "loss": 0.8254,
      "step": 5730
    },
    {
      "epoch": 0.47690262545696244,
      "grad_norm": 0.4123813509941101,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 5740
    },
    {
      "epoch": 0.47773346626786306,
      "grad_norm": 0.434916615486145,
      "learning_rate": 2e-05,
      "loss": 0.8256,
      "step": 5750
    },
    {
      "epoch": 0.4785643070787637,
      "grad_norm": 0.5103340148925781,
      "learning_rate": 2e-05,
      "loss": 0.8421,
      "step": 5760
    },
    {
      "epoch": 0.47939514788966436,
      "grad_norm": 0.4282371699810028,
      "learning_rate": 2e-05,
      "loss": 0.8278,
      "step": 5770
    },
    {
      "epoch": 0.480225988700565,
      "grad_norm": 0.43314841389656067,
      "learning_rate": 2e-05,
      "loss": 0.827,
      "step": 5780
    },
    {
      "epoch": 0.4810568295114656,
      "grad_norm": 0.44658035039901733,
      "learning_rate": 2e-05,
      "loss": 0.8125,
      "step": 5790
    },
    {
      "epoch": 0.48188767032236623,
      "grad_norm": 0.39747485518455505,
      "learning_rate": 2e-05,
      "loss": 0.7777,
      "step": 5800
    },
    {
      "epoch": 0.48271851113326686,
      "grad_norm": 0.44355490803718567,
      "learning_rate": 2e-05,
      "loss": 0.8433,
      "step": 5810
    },
    {
      "epoch": 0.4835493519441675,
      "grad_norm": 0.42593395709991455,
      "learning_rate": 2e-05,
      "loss": 0.8135,
      "step": 5820
    },
    {
      "epoch": 0.4843801927550681,
      "grad_norm": 0.47251594066619873,
      "learning_rate": 2e-05,
      "loss": 0.8487,
      "step": 5830
    },
    {
      "epoch": 0.4852110335659688,
      "grad_norm": 0.4496149718761444,
      "learning_rate": 2e-05,
      "loss": 0.8321,
      "step": 5840
    },
    {
      "epoch": 0.4860418743768694,
      "grad_norm": 0.4519672393798828,
      "learning_rate": 2e-05,
      "loss": 0.8386,
      "step": 5850
    },
    {
      "epoch": 0.48687271518777003,
      "grad_norm": 0.4655514061450958,
      "learning_rate": 2e-05,
      "loss": 0.8476,
      "step": 5860
    },
    {
      "epoch": 0.48770355599867066,
      "grad_norm": 0.44724467396736145,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 5870
    },
    {
      "epoch": 0.4885343968095713,
      "grad_norm": 0.42496272921562195,
      "learning_rate": 2e-05,
      "loss": 0.8338,
      "step": 5880
    },
    {
      "epoch": 0.4893652376204719,
      "grad_norm": 0.42337357997894287,
      "learning_rate": 2e-05,
      "loss": 0.8234,
      "step": 5890
    },
    {
      "epoch": 0.49019607843137253,
      "grad_norm": 0.5500143766403198,
      "learning_rate": 2e-05,
      "loss": 0.8415,
      "step": 5900
    },
    {
      "epoch": 0.4910269192422732,
      "grad_norm": 0.4203953146934509,
      "learning_rate": 2e-05,
      "loss": 0.8158,
      "step": 5910
    },
    {
      "epoch": 0.49185776005317383,
      "grad_norm": 0.4144504964351654,
      "learning_rate": 2e-05,
      "loss": 0.8122,
      "step": 5920
    },
    {
      "epoch": 0.49268860086407446,
      "grad_norm": 0.43117639422416687,
      "learning_rate": 2e-05,
      "loss": 0.8525,
      "step": 5930
    },
    {
      "epoch": 0.4935194416749751,
      "grad_norm": 0.41700372099876404,
      "learning_rate": 2e-05,
      "loss": 0.8496,
      "step": 5940
    },
    {
      "epoch": 0.4943502824858757,
      "grad_norm": 0.4200904071331024,
      "learning_rate": 2e-05,
      "loss": 0.7951,
      "step": 5950
    },
    {
      "epoch": 0.4951811232967763,
      "grad_norm": 0.4982404112815857,
      "learning_rate": 2e-05,
      "loss": 0.7957,
      "step": 5960
    },
    {
      "epoch": 0.49601196410767695,
      "grad_norm": 0.4292984902858734,
      "learning_rate": 2e-05,
      "loss": 0.822,
      "step": 5970
    },
    {
      "epoch": 0.4968428049185776,
      "grad_norm": 0.43484193086624146,
      "learning_rate": 2e-05,
      "loss": 0.8561,
      "step": 5980
    },
    {
      "epoch": 0.49767364572947825,
      "grad_norm": 0.45673108100891113,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 5990
    },
    {
      "epoch": 0.4985044865403789,
      "grad_norm": 0.42748504877090454,
      "learning_rate": 2e-05,
      "loss": 0.8183,
      "step": 6000
    },
    {
      "epoch": 0.4993353273512795,
      "grad_norm": 0.426548033952713,
      "learning_rate": 2e-05,
      "loss": 0.8093,
      "step": 6010
    },
    {
      "epoch": 0.5001661681621802,
      "grad_norm": 0.4277189373970032,
      "learning_rate": 2e-05,
      "loss": 0.8413,
      "step": 6020
    },
    {
      "epoch": 0.5009970089730807,
      "grad_norm": 0.4033356010913849,
      "learning_rate": 2e-05,
      "loss": 0.8469,
      "step": 6030
    },
    {
      "epoch": 0.5018278497839814,
      "grad_norm": 0.4260394275188446,
      "learning_rate": 2e-05,
      "loss": 0.8036,
      "step": 6040
    },
    {
      "epoch": 0.502658690594882,
      "grad_norm": 0.4196097254753113,
      "learning_rate": 2e-05,
      "loss": 0.832,
      "step": 6050
    },
    {
      "epoch": 0.5034895314057827,
      "grad_norm": 0.4278629422187805,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 6060
    },
    {
      "epoch": 0.5043203722166832,
      "grad_norm": 0.4983876347541809,
      "learning_rate": 2e-05,
      "loss": 0.8455,
      "step": 6070
    },
    {
      "epoch": 0.5051512130275839,
      "grad_norm": 0.41838744282722473,
      "learning_rate": 2e-05,
      "loss": 0.8448,
      "step": 6080
    },
    {
      "epoch": 0.5059820538384845,
      "grad_norm": 0.47586187720298767,
      "learning_rate": 2e-05,
      "loss": 0.8192,
      "step": 6090
    },
    {
      "epoch": 0.5068128946493852,
      "grad_norm": 0.43562379479408264,
      "learning_rate": 2e-05,
      "loss": 0.8343,
      "step": 6100
    },
    {
      "epoch": 0.5076437354602858,
      "grad_norm": 0.4252253770828247,
      "learning_rate": 2e-05,
      "loss": 0.7958,
      "step": 6110
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 0.4949187934398651,
      "learning_rate": 2e-05,
      "loss": 0.838,
      "step": 6120
    },
    {
      "epoch": 0.5093054170820871,
      "grad_norm": 0.4182480275630951,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 6130
    },
    {
      "epoch": 0.5101362578929877,
      "grad_norm": 0.42631253600120544,
      "learning_rate": 2e-05,
      "loss": 0.8469,
      "step": 6140
    },
    {
      "epoch": 0.5109670987038883,
      "grad_norm": 0.4129374623298645,
      "learning_rate": 2e-05,
      "loss": 0.8501,
      "step": 6150
    },
    {
      "epoch": 0.5117979395147889,
      "grad_norm": 0.42338091135025024,
      "learning_rate": 2e-05,
      "loss": 0.8085,
      "step": 6160
    },
    {
      "epoch": 0.5126287803256896,
      "grad_norm": 0.43417391180992126,
      "learning_rate": 2e-05,
      "loss": 0.8409,
      "step": 6170
    },
    {
      "epoch": 0.5134596211365903,
      "grad_norm": 0.43058478832244873,
      "learning_rate": 2e-05,
      "loss": 0.8586,
      "step": 6180
    },
    {
      "epoch": 0.5142904619474908,
      "grad_norm": 0.4551044702529907,
      "learning_rate": 2e-05,
      "loss": 0.8021,
      "step": 6190
    },
    {
      "epoch": 0.5151213027583915,
      "grad_norm": 0.43171536922454834,
      "learning_rate": 2e-05,
      "loss": 0.8179,
      "step": 6200
    },
    {
      "epoch": 0.5159521435692921,
      "grad_norm": 0.4427269995212555,
      "learning_rate": 2e-05,
      "loss": 0.8708,
      "step": 6210
    },
    {
      "epoch": 0.5167829843801928,
      "grad_norm": 0.46175330877304077,
      "learning_rate": 2e-05,
      "loss": 0.8349,
      "step": 6220
    },
    {
      "epoch": 0.5176138251910933,
      "grad_norm": 0.44279390573501587,
      "learning_rate": 2e-05,
      "loss": 0.8177,
      "step": 6230
    },
    {
      "epoch": 0.518444666001994,
      "grad_norm": 0.4267238974571228,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 6240
    },
    {
      "epoch": 0.5192755068128947,
      "grad_norm": 0.430918425321579,
      "learning_rate": 2e-05,
      "loss": 0.8056,
      "step": 6250
    },
    {
      "epoch": 0.5201063476237953,
      "grad_norm": 0.5073515772819519,
      "learning_rate": 2e-05,
      "loss": 0.8137,
      "step": 6260
    },
    {
      "epoch": 0.5209371884346959,
      "grad_norm": 0.5549708008766174,
      "learning_rate": 2e-05,
      "loss": 0.8063,
      "step": 6270
    },
    {
      "epoch": 0.5217680292455965,
      "grad_norm": 0.4234621226787567,
      "learning_rate": 2e-05,
      "loss": 0.8282,
      "step": 6280
    },
    {
      "epoch": 0.5225988700564972,
      "grad_norm": 0.4826675355434418,
      "learning_rate": 2e-05,
      "loss": 0.8175,
      "step": 6290
    },
    {
      "epoch": 0.5234297108673978,
      "grad_norm": 0.45421767234802246,
      "learning_rate": 2e-05,
      "loss": 0.8131,
      "step": 6300
    },
    {
      "epoch": 0.5242605516782984,
      "grad_norm": 0.4507942199707031,
      "learning_rate": 2e-05,
      "loss": 0.7988,
      "step": 6310
    },
    {
      "epoch": 0.5250913924891991,
      "grad_norm": 0.41987136006355286,
      "learning_rate": 2e-05,
      "loss": 0.8189,
      "step": 6320
    },
    {
      "epoch": 0.5259222333000997,
      "grad_norm": 0.44727593660354614,
      "learning_rate": 2e-05,
      "loss": 0.8211,
      "step": 6330
    },
    {
      "epoch": 0.5267530741110004,
      "grad_norm": 0.4125182628631592,
      "learning_rate": 2e-05,
      "loss": 0.8534,
      "step": 6340
    },
    {
      "epoch": 0.5275839149219009,
      "grad_norm": 0.4269663691520691,
      "learning_rate": 2e-05,
      "loss": 0.8119,
      "step": 6350
    },
    {
      "epoch": 0.5284147557328016,
      "grad_norm": 0.44187474250793457,
      "learning_rate": 2e-05,
      "loss": 0.8375,
      "step": 6360
    },
    {
      "epoch": 0.5292455965437022,
      "grad_norm": 0.43037286400794983,
      "learning_rate": 2e-05,
      "loss": 0.8266,
      "step": 6370
    },
    {
      "epoch": 0.5300764373546029,
      "grad_norm": 0.42437753081321716,
      "learning_rate": 2e-05,
      "loss": 0.8275,
      "step": 6380
    },
    {
      "epoch": 0.5309072781655035,
      "grad_norm": 0.437866747379303,
      "learning_rate": 2e-05,
      "loss": 0.8106,
      "step": 6390
    },
    {
      "epoch": 0.5317381189764041,
      "grad_norm": 0.4367234408855438,
      "learning_rate": 2e-05,
      "loss": 0.8353,
      "step": 6400
    },
    {
      "epoch": 0.5325689597873048,
      "grad_norm": 0.4163818955421448,
      "learning_rate": 2e-05,
      "loss": 0.8392,
      "step": 6410
    },
    {
      "epoch": 0.5333998005982054,
      "grad_norm": 0.38560304045677185,
      "learning_rate": 2e-05,
      "loss": 0.8187,
      "step": 6420
    },
    {
      "epoch": 0.534230641409106,
      "grad_norm": 0.7415143251419067,
      "learning_rate": 2e-05,
      "loss": 0.8001,
      "step": 6430
    },
    {
      "epoch": 0.5350614822200066,
      "grad_norm": 0.425771027803421,
      "learning_rate": 2e-05,
      "loss": 0.8146,
      "step": 6440
    },
    {
      "epoch": 0.5358923230309073,
      "grad_norm": 0.4066011309623718,
      "learning_rate": 2e-05,
      "loss": 0.8324,
      "step": 6450
    },
    {
      "epoch": 0.536723163841808,
      "grad_norm": 0.42413318157196045,
      "learning_rate": 2e-05,
      "loss": 0.8391,
      "step": 6460
    },
    {
      "epoch": 0.5375540046527085,
      "grad_norm": 0.47565126419067383,
      "learning_rate": 2e-05,
      "loss": 0.8222,
      "step": 6470
    },
    {
      "epoch": 0.5383848454636092,
      "grad_norm": 0.46630623936653137,
      "learning_rate": 2e-05,
      "loss": 0.8321,
      "step": 6480
    },
    {
      "epoch": 0.5392156862745098,
      "grad_norm": 0.4913094937801361,
      "learning_rate": 2e-05,
      "loss": 0.7825,
      "step": 6490
    },
    {
      "epoch": 0.5400465270854105,
      "grad_norm": 0.5438485145568848,
      "learning_rate": 2e-05,
      "loss": 0.845,
      "step": 6500
    },
    {
      "epoch": 0.540877367896311,
      "grad_norm": 0.449980229139328,
      "learning_rate": 2e-05,
      "loss": 0.8213,
      "step": 6510
    },
    {
      "epoch": 0.5417082087072117,
      "grad_norm": 0.42854467034339905,
      "learning_rate": 2e-05,
      "loss": 0.8484,
      "step": 6520
    },
    {
      "epoch": 0.5425390495181124,
      "grad_norm": 0.451889306306839,
      "learning_rate": 2e-05,
      "loss": 0.8182,
      "step": 6530
    },
    {
      "epoch": 0.543369890329013,
      "grad_norm": 0.42612698674201965,
      "learning_rate": 2e-05,
      "loss": 0.8477,
      "step": 6540
    },
    {
      "epoch": 0.5442007311399136,
      "grad_norm": 0.4758527874946594,
      "learning_rate": 2e-05,
      "loss": 0.8192,
      "step": 6550
    },
    {
      "epoch": 0.5450315719508142,
      "grad_norm": 0.4272933602333069,
      "learning_rate": 2e-05,
      "loss": 0.8399,
      "step": 6560
    },
    {
      "epoch": 0.5458624127617149,
      "grad_norm": 0.4214387834072113,
      "learning_rate": 2e-05,
      "loss": 0.8267,
      "step": 6570
    },
    {
      "epoch": 0.5466932535726154,
      "grad_norm": 0.455354243516922,
      "learning_rate": 2e-05,
      "loss": 0.8131,
      "step": 6580
    },
    {
      "epoch": 0.5475240943835161,
      "grad_norm": 0.5176498293876648,
      "learning_rate": 2e-05,
      "loss": 0.8404,
      "step": 6590
    },
    {
      "epoch": 0.5483549351944168,
      "grad_norm": 0.4349859356880188,
      "learning_rate": 2e-05,
      "loss": 0.8197,
      "step": 6600
    },
    {
      "epoch": 0.5491857760053174,
      "grad_norm": 0.43980085849761963,
      "learning_rate": 2e-05,
      "loss": 0.8117,
      "step": 6610
    },
    {
      "epoch": 0.550016616816218,
      "grad_norm": 0.44528377056121826,
      "learning_rate": 2e-05,
      "loss": 0.8043,
      "step": 6620
    },
    {
      "epoch": 0.5508474576271186,
      "grad_norm": 0.40468040108680725,
      "learning_rate": 2e-05,
      "loss": 0.8239,
      "step": 6630
    },
    {
      "epoch": 0.5516782984380193,
      "grad_norm": 0.4249822199344635,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 6640
    },
    {
      "epoch": 0.5525091392489199,
      "grad_norm": 0.4094141721725464,
      "learning_rate": 2e-05,
      "loss": 0.8024,
      "step": 6650
    },
    {
      "epoch": 0.5533399800598205,
      "grad_norm": 0.4321598708629608,
      "learning_rate": 2e-05,
      "loss": 0.823,
      "step": 6660
    },
    {
      "epoch": 0.5541708208707211,
      "grad_norm": 0.43565601110458374,
      "learning_rate": 2e-05,
      "loss": 0.836,
      "step": 6670
    },
    {
      "epoch": 0.5550016616816218,
      "grad_norm": 0.615647554397583,
      "learning_rate": 2e-05,
      "loss": 0.8192,
      "step": 6680
    },
    {
      "epoch": 0.5558325024925225,
      "grad_norm": 0.4016720652580261,
      "learning_rate": 2e-05,
      "loss": 0.8037,
      "step": 6690
    },
    {
      "epoch": 0.556663343303423,
      "grad_norm": 0.427377849817276,
      "learning_rate": 2e-05,
      "loss": 0.7993,
      "step": 6700
    },
    {
      "epoch": 0.5574941841143237,
      "grad_norm": 0.4193289577960968,
      "learning_rate": 2e-05,
      "loss": 0.8318,
      "step": 6710
    },
    {
      "epoch": 0.5583250249252243,
      "grad_norm": 0.4635801613330841,
      "learning_rate": 2e-05,
      "loss": 0.8364,
      "step": 6720
    },
    {
      "epoch": 0.559155865736125,
      "grad_norm": 0.39860624074935913,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 6730
    },
    {
      "epoch": 0.5599867065470255,
      "grad_norm": 0.44901591539382935,
      "learning_rate": 2e-05,
      "loss": 0.8043,
      "step": 6740
    },
    {
      "epoch": 0.5608175473579262,
      "grad_norm": 0.5117014646530151,
      "learning_rate": 2e-05,
      "loss": 0.8115,
      "step": 6750
    },
    {
      "epoch": 0.5616483881688269,
      "grad_norm": 0.4724797010421753,
      "learning_rate": 2e-05,
      "loss": 0.8172,
      "step": 6760
    },
    {
      "epoch": 0.5624792289797275,
      "grad_norm": 0.45245063304901123,
      "learning_rate": 2e-05,
      "loss": 0.811,
      "step": 6770
    },
    {
      "epoch": 0.5633100697906281,
      "grad_norm": 0.46593913435935974,
      "learning_rate": 2e-05,
      "loss": 0.8281,
      "step": 6780
    },
    {
      "epoch": 0.5641409106015287,
      "grad_norm": 0.42442384362220764,
      "learning_rate": 2e-05,
      "loss": 0.8179,
      "step": 6790
    },
    {
      "epoch": 0.5649717514124294,
      "grad_norm": 0.37842094898223877,
      "learning_rate": 2e-05,
      "loss": 0.8031,
      "step": 6800
    },
    {
      "epoch": 0.56580259222333,
      "grad_norm": 0.43062350153923035,
      "learning_rate": 2e-05,
      "loss": 0.8138,
      "step": 6810
    },
    {
      "epoch": 0.5666334330342306,
      "grad_norm": 0.5517492890357971,
      "learning_rate": 2e-05,
      "loss": 0.7942,
      "step": 6820
    },
    {
      "epoch": 0.5674642738451313,
      "grad_norm": 0.44958561658859253,
      "learning_rate": 2e-05,
      "loss": 0.7831,
      "step": 6830
    },
    {
      "epoch": 0.5682951146560319,
      "grad_norm": 0.4615941643714905,
      "learning_rate": 2e-05,
      "loss": 0.8432,
      "step": 6840
    },
    {
      "epoch": 0.5691259554669326,
      "grad_norm": 0.42235204577445984,
      "learning_rate": 2e-05,
      "loss": 0.8541,
      "step": 6850
    },
    {
      "epoch": 0.5699567962778331,
      "grad_norm": 0.44494616985321045,
      "learning_rate": 2e-05,
      "loss": 0.8492,
      "step": 6860
    },
    {
      "epoch": 0.5707876370887338,
      "grad_norm": 0.44087088108062744,
      "learning_rate": 2e-05,
      "loss": 0.7963,
      "step": 6870
    },
    {
      "epoch": 0.5716184778996344,
      "grad_norm": 0.4105639159679413,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 6880
    },
    {
      "epoch": 0.5724493187105351,
      "grad_norm": 0.4544210731983185,
      "learning_rate": 2e-05,
      "loss": 0.8117,
      "step": 6890
    },
    {
      "epoch": 0.5732801595214357,
      "grad_norm": 0.40693262219429016,
      "learning_rate": 2e-05,
      "loss": 0.8413,
      "step": 6900
    },
    {
      "epoch": 0.5741110003323363,
      "grad_norm": 0.4064671993255615,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 6910
    },
    {
      "epoch": 0.574941841143237,
      "grad_norm": 0.42901811003685,
      "learning_rate": 2e-05,
      "loss": 0.8326,
      "step": 6920
    },
    {
      "epoch": 0.5757726819541376,
      "grad_norm": 0.44339072704315186,
      "learning_rate": 2e-05,
      "loss": 0.8145,
      "step": 6930
    },
    {
      "epoch": 0.5766035227650382,
      "grad_norm": 0.42099547386169434,
      "learning_rate": 2e-05,
      "loss": 0.802,
      "step": 6940
    },
    {
      "epoch": 0.5774343635759388,
      "grad_norm": 0.4330797791481018,
      "learning_rate": 2e-05,
      "loss": 0.7933,
      "step": 6950
    },
    {
      "epoch": 0.5782652043868395,
      "grad_norm": 0.4343934655189514,
      "learning_rate": 2e-05,
      "loss": 0.8388,
      "step": 6960
    },
    {
      "epoch": 0.5790960451977402,
      "grad_norm": 0.43999940156936646,
      "learning_rate": 2e-05,
      "loss": 0.7939,
      "step": 6970
    },
    {
      "epoch": 0.5799268860086407,
      "grad_norm": 0.46327370405197144,
      "learning_rate": 2e-05,
      "loss": 0.8285,
      "step": 6980
    },
    {
      "epoch": 0.5807577268195414,
      "grad_norm": 0.4223586320877075,
      "learning_rate": 2e-05,
      "loss": 0.8504,
      "step": 6990
    },
    {
      "epoch": 0.581588567630442,
      "grad_norm": 0.44147083163261414,
      "learning_rate": 2e-05,
      "loss": 0.8161,
      "step": 7000
    },
    {
      "epoch": 0.5824194084413427,
      "grad_norm": 0.43002596497535706,
      "learning_rate": 2e-05,
      "loss": 0.8704,
      "step": 7010
    },
    {
      "epoch": 0.5832502492522432,
      "grad_norm": 0.45780664682388306,
      "learning_rate": 2e-05,
      "loss": 0.8144,
      "step": 7020
    },
    {
      "epoch": 0.5840810900631439,
      "grad_norm": 0.42484480142593384,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 7030
    },
    {
      "epoch": 0.5849119308740446,
      "grad_norm": 0.4648698568344116,
      "learning_rate": 2e-05,
      "loss": 0.8336,
      "step": 7040
    },
    {
      "epoch": 0.5857427716849452,
      "grad_norm": 0.4718359410762787,
      "learning_rate": 2e-05,
      "loss": 0.8465,
      "step": 7050
    },
    {
      "epoch": 0.5865736124958458,
      "grad_norm": 0.42302218079566956,
      "learning_rate": 2e-05,
      "loss": 0.8084,
      "step": 7060
    },
    {
      "epoch": 0.5874044533067464,
      "grad_norm": 0.44642651081085205,
      "learning_rate": 2e-05,
      "loss": 0.7972,
      "step": 7070
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.4293728172779083,
      "learning_rate": 2e-05,
      "loss": 0.805,
      "step": 7080
    },
    {
      "epoch": 0.5890661349285476,
      "grad_norm": 0.44513246417045593,
      "learning_rate": 2e-05,
      "loss": 0.8515,
      "step": 7090
    },
    {
      "epoch": 0.5898969757394483,
      "grad_norm": 0.46737828850746155,
      "learning_rate": 2e-05,
      "loss": 0.7984,
      "step": 7100
    },
    {
      "epoch": 0.590727816550349,
      "grad_norm": 0.43008360266685486,
      "learning_rate": 2e-05,
      "loss": 0.8404,
      "step": 7110
    },
    {
      "epoch": 0.5915586573612496,
      "grad_norm": 0.46414613723754883,
      "learning_rate": 2e-05,
      "loss": 0.8098,
      "step": 7120
    },
    {
      "epoch": 0.5923894981721503,
      "grad_norm": 0.4354031980037689,
      "learning_rate": 2e-05,
      "loss": 0.8268,
      "step": 7130
    },
    {
      "epoch": 0.5932203389830508,
      "grad_norm": 0.4487757086753845,
      "learning_rate": 2e-05,
      "loss": 0.8019,
      "step": 7140
    },
    {
      "epoch": 0.5940511797939515,
      "grad_norm": 0.4118978977203369,
      "learning_rate": 2e-05,
      "loss": 0.824,
      "step": 7150
    },
    {
      "epoch": 0.5948820206048521,
      "grad_norm": 0.4308334290981293,
      "learning_rate": 2e-05,
      "loss": 0.8043,
      "step": 7160
    },
    {
      "epoch": 0.5957128614157527,
      "grad_norm": 0.3989607095718384,
      "learning_rate": 2e-05,
      "loss": 0.8027,
      "step": 7170
    },
    {
      "epoch": 0.5965437022266534,
      "grad_norm": 0.4389350414276123,
      "learning_rate": 2e-05,
      "loss": 0.7853,
      "step": 7180
    },
    {
      "epoch": 0.597374543037554,
      "grad_norm": 0.4360221326351166,
      "learning_rate": 2e-05,
      "loss": 0.8192,
      "step": 7190
    },
    {
      "epoch": 0.5982053838484547,
      "grad_norm": 0.6470382809638977,
      "learning_rate": 2e-05,
      "loss": 0.8405,
      "step": 7200
    },
    {
      "epoch": 0.5990362246593552,
      "grad_norm": 0.42739352583885193,
      "learning_rate": 2e-05,
      "loss": 0.837,
      "step": 7210
    },
    {
      "epoch": 0.5998670654702559,
      "grad_norm": 0.4412316381931305,
      "learning_rate": 2e-05,
      "loss": 0.823,
      "step": 7220
    },
    {
      "epoch": 0.6006979062811565,
      "grad_norm": 0.43037042021751404,
      "learning_rate": 2e-05,
      "loss": 0.8108,
      "step": 7230
    },
    {
      "epoch": 0.6015287470920572,
      "grad_norm": 0.39930081367492676,
      "learning_rate": 2e-05,
      "loss": 0.7923,
      "step": 7240
    },
    {
      "epoch": 0.6023595879029577,
      "grad_norm": 0.4298480451107025,
      "learning_rate": 2e-05,
      "loss": 0.8041,
      "step": 7250
    },
    {
      "epoch": 0.6031904287138584,
      "grad_norm": 0.456412672996521,
      "learning_rate": 2e-05,
      "loss": 0.7795,
      "step": 7260
    },
    {
      "epoch": 0.6040212695247591,
      "grad_norm": 0.4314030408859253,
      "learning_rate": 2e-05,
      "loss": 0.7876,
      "step": 7270
    },
    {
      "epoch": 0.6048521103356597,
      "grad_norm": 0.429942786693573,
      "learning_rate": 2e-05,
      "loss": 0.8081,
      "step": 7280
    },
    {
      "epoch": 0.6056829511465603,
      "grad_norm": 0.41611841320991516,
      "learning_rate": 2e-05,
      "loss": 0.8169,
      "step": 7290
    },
    {
      "epoch": 0.6065137919574609,
      "grad_norm": 0.4366942346096039,
      "learning_rate": 2e-05,
      "loss": 0.803,
      "step": 7300
    },
    {
      "epoch": 0.6073446327683616,
      "grad_norm": 0.4280656576156616,
      "learning_rate": 2e-05,
      "loss": 0.828,
      "step": 7310
    },
    {
      "epoch": 0.6081754735792622,
      "grad_norm": 0.4328523576259613,
      "learning_rate": 2e-05,
      "loss": 0.8396,
      "step": 7320
    },
    {
      "epoch": 0.6090063143901628,
      "grad_norm": 0.43899762630462646,
      "learning_rate": 2e-05,
      "loss": 0.8013,
      "step": 7330
    },
    {
      "epoch": 0.6098371552010635,
      "grad_norm": 0.45236921310424805,
      "learning_rate": 2e-05,
      "loss": 0.7983,
      "step": 7340
    },
    {
      "epoch": 0.6106679960119641,
      "grad_norm": 0.42701658606529236,
      "learning_rate": 2e-05,
      "loss": 0.8414,
      "step": 7350
    },
    {
      "epoch": 0.6114988368228648,
      "grad_norm": 0.455506831407547,
      "learning_rate": 2e-05,
      "loss": 0.8254,
      "step": 7360
    },
    {
      "epoch": 0.6123296776337653,
      "grad_norm": 0.42559343576431274,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 7370
    },
    {
      "epoch": 0.613160518444666,
      "grad_norm": 0.4917125701904297,
      "learning_rate": 2e-05,
      "loss": 0.8048,
      "step": 7380
    },
    {
      "epoch": 0.6139913592555666,
      "grad_norm": 0.4501971900463104,
      "learning_rate": 2e-05,
      "loss": 0.8014,
      "step": 7390
    },
    {
      "epoch": 0.6148222000664673,
      "grad_norm": 0.46539753675460815,
      "learning_rate": 2e-05,
      "loss": 0.8149,
      "step": 7400
    },
    {
      "epoch": 0.6156530408773679,
      "grad_norm": 0.4451500475406647,
      "learning_rate": 2e-05,
      "loss": 0.8297,
      "step": 7410
    },
    {
      "epoch": 0.6164838816882685,
      "grad_norm": 0.43894651532173157,
      "learning_rate": 2e-05,
      "loss": 0.7858,
      "step": 7420
    },
    {
      "epoch": 0.6173147224991692,
      "grad_norm": 0.527115523815155,
      "learning_rate": 2e-05,
      "loss": 0.8014,
      "step": 7430
    },
    {
      "epoch": 0.6181455633100698,
      "grad_norm": 0.41039714217185974,
      "learning_rate": 2e-05,
      "loss": 0.8167,
      "step": 7440
    },
    {
      "epoch": 0.6189764041209704,
      "grad_norm": 0.45890313386917114,
      "learning_rate": 2e-05,
      "loss": 0.7999,
      "step": 7450
    },
    {
      "epoch": 0.619807244931871,
      "grad_norm": 0.38796520233154297,
      "learning_rate": 2e-05,
      "loss": 0.8083,
      "step": 7460
    },
    {
      "epoch": 0.6206380857427717,
      "grad_norm": 0.4411107003688812,
      "learning_rate": 2e-05,
      "loss": 0.8102,
      "step": 7470
    },
    {
      "epoch": 0.6214689265536724,
      "grad_norm": 0.5067929029464722,
      "learning_rate": 2e-05,
      "loss": 0.7968,
      "step": 7480
    },
    {
      "epoch": 0.6222997673645729,
      "grad_norm": 0.42519643902778625,
      "learning_rate": 2e-05,
      "loss": 0.7621,
      "step": 7490
    },
    {
      "epoch": 0.6231306081754736,
      "grad_norm": 0.4314921796321869,
      "learning_rate": 2e-05,
      "loss": 0.8334,
      "step": 7500
    },
    {
      "epoch": 0.6239614489863742,
      "grad_norm": 0.42537370324134827,
      "learning_rate": 2e-05,
      "loss": 0.7709,
      "step": 7510
    },
    {
      "epoch": 0.6247922897972749,
      "grad_norm": 0.42143356800079346,
      "learning_rate": 2e-05,
      "loss": 0.8258,
      "step": 7520
    },
    {
      "epoch": 0.6256231306081754,
      "grad_norm": 0.41784682869911194,
      "learning_rate": 2e-05,
      "loss": 0.7779,
      "step": 7530
    },
    {
      "epoch": 0.6264539714190761,
      "grad_norm": 0.4196126163005829,
      "learning_rate": 2e-05,
      "loss": 0.816,
      "step": 7540
    },
    {
      "epoch": 0.6272848122299768,
      "grad_norm": 0.44915974140167236,
      "learning_rate": 2e-05,
      "loss": 0.8192,
      "step": 7550
    },
    {
      "epoch": 0.6281156530408774,
      "grad_norm": 0.41347000002861023,
      "learning_rate": 2e-05,
      "loss": 0.7783,
      "step": 7560
    },
    {
      "epoch": 0.628946493851778,
      "grad_norm": 0.4378238022327423,
      "learning_rate": 2e-05,
      "loss": 0.8133,
      "step": 7570
    },
    {
      "epoch": 0.6297773346626786,
      "grad_norm": 0.43245336413383484,
      "learning_rate": 2e-05,
      "loss": 0.7916,
      "step": 7580
    },
    {
      "epoch": 0.6306081754735793,
      "grad_norm": 0.4301047623157501,
      "learning_rate": 2e-05,
      "loss": 0.8252,
      "step": 7590
    },
    {
      "epoch": 0.6314390162844798,
      "grad_norm": 0.42061856389045715,
      "learning_rate": 2e-05,
      "loss": 0.797,
      "step": 7600
    },
    {
      "epoch": 0.6322698570953805,
      "grad_norm": 0.45449310541152954,
      "learning_rate": 2e-05,
      "loss": 0.8082,
      "step": 7610
    },
    {
      "epoch": 0.6331006979062812,
      "grad_norm": 0.4517688453197479,
      "learning_rate": 2e-05,
      "loss": 0.8129,
      "step": 7620
    },
    {
      "epoch": 0.6339315387171818,
      "grad_norm": 0.4215247631072998,
      "learning_rate": 2e-05,
      "loss": 0.7993,
      "step": 7630
    },
    {
      "epoch": 0.6347623795280825,
      "grad_norm": 0.4146602749824524,
      "learning_rate": 2e-05,
      "loss": 0.8297,
      "step": 7640
    },
    {
      "epoch": 0.635593220338983,
      "grad_norm": 0.4455009996891022,
      "learning_rate": 2e-05,
      "loss": 0.8284,
      "step": 7650
    },
    {
      "epoch": 0.6364240611498837,
      "grad_norm": 0.46344560384750366,
      "learning_rate": 2e-05,
      "loss": 0.8139,
      "step": 7660
    },
    {
      "epoch": 0.6372549019607843,
      "grad_norm": 0.4297507107257843,
      "learning_rate": 2e-05,
      "loss": 0.8158,
      "step": 7670
    },
    {
      "epoch": 0.638085742771685,
      "grad_norm": 0.4096633195877075,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 7680
    },
    {
      "epoch": 0.6389165835825856,
      "grad_norm": 0.45133984088897705,
      "learning_rate": 2e-05,
      "loss": 0.8151,
      "step": 7690
    },
    {
      "epoch": 0.6397474243934862,
      "grad_norm": 0.49049293994903564,
      "learning_rate": 2e-05,
      "loss": 0.8262,
      "step": 7700
    },
    {
      "epoch": 0.6405782652043869,
      "grad_norm": 0.4415861964225769,
      "learning_rate": 2e-05,
      "loss": 0.8366,
      "step": 7710
    },
    {
      "epoch": 0.6414091060152874,
      "grad_norm": 0.42402544617652893,
      "learning_rate": 2e-05,
      "loss": 0.784,
      "step": 7720
    },
    {
      "epoch": 0.6422399468261881,
      "grad_norm": 0.46036624908447266,
      "learning_rate": 2e-05,
      "loss": 0.8137,
      "step": 7730
    },
    {
      "epoch": 0.6430707876370887,
      "grad_norm": 0.4457029104232788,
      "learning_rate": 2e-05,
      "loss": 0.8102,
      "step": 7740
    },
    {
      "epoch": 0.6439016284479894,
      "grad_norm": 0.43899351358413696,
      "learning_rate": 2e-05,
      "loss": 0.802,
      "step": 7750
    },
    {
      "epoch": 0.64473246925889,
      "grad_norm": 0.4341202974319458,
      "learning_rate": 2e-05,
      "loss": 0.8054,
      "step": 7760
    },
    {
      "epoch": 0.6455633100697906,
      "grad_norm": 0.4517613351345062,
      "learning_rate": 2e-05,
      "loss": 0.8128,
      "step": 7770
    },
    {
      "epoch": 0.6463941508806913,
      "grad_norm": 0.4076000452041626,
      "learning_rate": 2e-05,
      "loss": 0.8133,
      "step": 7780
    },
    {
      "epoch": 0.6472249916915919,
      "grad_norm": 0.4477522671222687,
      "learning_rate": 2e-05,
      "loss": 0.7851,
      "step": 7790
    },
    {
      "epoch": 0.6480558325024925,
      "grad_norm": 0.4105316996574402,
      "learning_rate": 2e-05,
      "loss": 0.8162,
      "step": 7800
    },
    {
      "epoch": 0.6488866733133931,
      "grad_norm": 0.5287365913391113,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 7810
    },
    {
      "epoch": 0.6497175141242938,
      "grad_norm": 0.43843433260917664,
      "learning_rate": 2e-05,
      "loss": 0.816,
      "step": 7820
    },
    {
      "epoch": 0.6505483549351944,
      "grad_norm": 0.41933509707450867,
      "learning_rate": 2e-05,
      "loss": 0.7896,
      "step": 7830
    },
    {
      "epoch": 0.651379195746095,
      "grad_norm": 1.3448495864868164,
      "learning_rate": 2e-05,
      "loss": 0.7623,
      "step": 7840
    },
    {
      "epoch": 0.6522100365569957,
      "grad_norm": 0.49356237053871155,
      "learning_rate": 2e-05,
      "loss": 0.8039,
      "step": 7850
    },
    {
      "epoch": 0.6530408773678963,
      "grad_norm": 0.45392218232154846,
      "learning_rate": 2e-05,
      "loss": 0.7749,
      "step": 7860
    },
    {
      "epoch": 0.653871718178797,
      "grad_norm": 0.4620184004306793,
      "learning_rate": 2e-05,
      "loss": 0.8233,
      "step": 7870
    },
    {
      "epoch": 0.6547025589896975,
      "grad_norm": 0.47395840287208557,
      "learning_rate": 2e-05,
      "loss": 0.806,
      "step": 7880
    },
    {
      "epoch": 0.6555333998005982,
      "grad_norm": 0.41487324237823486,
      "learning_rate": 2e-05,
      "loss": 0.7565,
      "step": 7890
    },
    {
      "epoch": 0.6563642406114988,
      "grad_norm": 0.4387260973453522,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 7900
    },
    {
      "epoch": 0.6571950814223995,
      "grad_norm": 0.46874308586120605,
      "learning_rate": 2e-05,
      "loss": 0.8079,
      "step": 7910
    },
    {
      "epoch": 0.6580259222333001,
      "grad_norm": 0.41408103704452515,
      "learning_rate": 2e-05,
      "loss": 0.7996,
      "step": 7920
    },
    {
      "epoch": 0.6588567630442007,
      "grad_norm": 0.41814014315605164,
      "learning_rate": 2e-05,
      "loss": 0.7812,
      "step": 7930
    },
    {
      "epoch": 0.6596876038551014,
      "grad_norm": 0.47311773896217346,
      "learning_rate": 2e-05,
      "loss": 0.7641,
      "step": 7940
    },
    {
      "epoch": 0.660518444666002,
      "grad_norm": 0.43022632598876953,
      "learning_rate": 2e-05,
      "loss": 0.8089,
      "step": 7950
    },
    {
      "epoch": 0.6613492854769026,
      "grad_norm": 0.4499478340148926,
      "learning_rate": 2e-05,
      "loss": 0.8061,
      "step": 7960
    },
    {
      "epoch": 0.6621801262878032,
      "grad_norm": 0.45977261662483215,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 7970
    },
    {
      "epoch": 0.6630109670987039,
      "grad_norm": 0.623633623123169,
      "learning_rate": 2e-05,
      "loss": 0.8114,
      "step": 7980
    },
    {
      "epoch": 0.6638418079096046,
      "grad_norm": 0.424549400806427,
      "learning_rate": 2e-05,
      "loss": 0.8281,
      "step": 7990
    },
    {
      "epoch": 0.6646726487205051,
      "grad_norm": 0.5123676061630249,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 8000
    },
    {
      "epoch": 0.6655034895314058,
      "grad_norm": 0.46598541736602783,
      "learning_rate": 2e-05,
      "loss": 0.7619,
      "step": 8010
    },
    {
      "epoch": 0.6663343303423064,
      "grad_norm": 0.4552728235721588,
      "learning_rate": 2e-05,
      "loss": 0.8081,
      "step": 8020
    },
    {
      "epoch": 0.6671651711532071,
      "grad_norm": 0.4489738643169403,
      "learning_rate": 2e-05,
      "loss": 0.8048,
      "step": 8030
    },
    {
      "epoch": 0.6679960119641076,
      "grad_norm": 0.41899433732032776,
      "learning_rate": 2e-05,
      "loss": 0.7779,
      "step": 8040
    },
    {
      "epoch": 0.6688268527750083,
      "grad_norm": 0.41635802388191223,
      "learning_rate": 2e-05,
      "loss": 0.7391,
      "step": 8050
    },
    {
      "epoch": 0.669657693585909,
      "grad_norm": 0.46255946159362793,
      "learning_rate": 2e-05,
      "loss": 0.7841,
      "step": 8060
    },
    {
      "epoch": 0.6704885343968096,
      "grad_norm": 0.4314276874065399,
      "learning_rate": 2e-05,
      "loss": 0.7971,
      "step": 8070
    },
    {
      "epoch": 0.6713193752077102,
      "grad_norm": 0.4265425503253937,
      "learning_rate": 2e-05,
      "loss": 0.7854,
      "step": 8080
    },
    {
      "epoch": 0.6721502160186108,
      "grad_norm": 0.4680528938770294,
      "learning_rate": 2e-05,
      "loss": 0.7725,
      "step": 8090
    },
    {
      "epoch": 0.6729810568295115,
      "grad_norm": 0.4211062788963318,
      "learning_rate": 2e-05,
      "loss": 0.7868,
      "step": 8100
    },
    {
      "epoch": 0.673811897640412,
      "grad_norm": 0.42495059967041016,
      "learning_rate": 2e-05,
      "loss": 0.7849,
      "step": 8110
    },
    {
      "epoch": 0.6746427384513127,
      "grad_norm": 0.46682438254356384,
      "learning_rate": 2e-05,
      "loss": 0.8247,
      "step": 8120
    },
    {
      "epoch": 0.6754735792622134,
      "grad_norm": 0.4211519658565521,
      "learning_rate": 2e-05,
      "loss": 0.8022,
      "step": 8130
    },
    {
      "epoch": 0.676304420073114,
      "grad_norm": 0.45310255885124207,
      "learning_rate": 2e-05,
      "loss": 0.7741,
      "step": 8140
    },
    {
      "epoch": 0.6771352608840147,
      "grad_norm": 0.48029085993766785,
      "learning_rate": 2e-05,
      "loss": 0.7731,
      "step": 8150
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 0.4800824820995331,
      "learning_rate": 2e-05,
      "loss": 0.8273,
      "step": 8160
    },
    {
      "epoch": 0.6787969425058159,
      "grad_norm": 0.39679795503616333,
      "learning_rate": 2e-05,
      "loss": 0.7922,
      "step": 8170
    },
    {
      "epoch": 0.6796277833167165,
      "grad_norm": 0.4627399742603302,
      "learning_rate": 2e-05,
      "loss": 0.7984,
      "step": 8180
    },
    {
      "epoch": 0.6804586241276172,
      "grad_norm": 0.407001793384552,
      "learning_rate": 2e-05,
      "loss": 0.793,
      "step": 8190
    },
    {
      "epoch": 0.6812894649385178,
      "grad_norm": 0.44780170917510986,
      "learning_rate": 2e-05,
      "loss": 0.7732,
      "step": 8200
    },
    {
      "epoch": 0.6821203057494184,
      "grad_norm": 0.4543469548225403,
      "learning_rate": 2e-05,
      "loss": 0.8063,
      "step": 8210
    },
    {
      "epoch": 0.6829511465603191,
      "grad_norm": 0.40996164083480835,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 8220
    },
    {
      "epoch": 0.6837819873712196,
      "grad_norm": 0.41195347905158997,
      "learning_rate": 2e-05,
      "loss": 0.7953,
      "step": 8230
    },
    {
      "epoch": 0.6846128281821203,
      "grad_norm": 0.4411318898200989,
      "learning_rate": 2e-05,
      "loss": 0.7936,
      "step": 8240
    },
    {
      "epoch": 0.6854436689930209,
      "grad_norm": 0.42686304450035095,
      "learning_rate": 2e-05,
      "loss": 0.8047,
      "step": 8250
    },
    {
      "epoch": 0.6862745098039216,
      "grad_norm": 0.43443116545677185,
      "learning_rate": 2e-05,
      "loss": 0.7984,
      "step": 8260
    },
    {
      "epoch": 0.6871053506148223,
      "grad_norm": 0.42949527502059937,
      "learning_rate": 2e-05,
      "loss": 0.7864,
      "step": 8270
    },
    {
      "epoch": 0.6879361914257228,
      "grad_norm": 0.42888936400413513,
      "learning_rate": 2e-05,
      "loss": 0.8141,
      "step": 8280
    },
    {
      "epoch": 0.6887670322366235,
      "grad_norm": 0.4309579133987427,
      "learning_rate": 2e-05,
      "loss": 0.8026,
      "step": 8290
    },
    {
      "epoch": 0.6895978730475241,
      "grad_norm": 0.49009355902671814,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 8300
    },
    {
      "epoch": 0.6904287138584247,
      "grad_norm": 0.5142233967781067,
      "learning_rate": 2e-05,
      "loss": 0.817,
      "step": 8310
    },
    {
      "epoch": 0.6912595546693253,
      "grad_norm": 0.4679347574710846,
      "learning_rate": 2e-05,
      "loss": 0.7966,
      "step": 8320
    },
    {
      "epoch": 0.692090395480226,
      "grad_norm": 0.4385068118572235,
      "learning_rate": 2e-05,
      "loss": 0.8081,
      "step": 8330
    },
    {
      "epoch": 0.6929212362911267,
      "grad_norm": 0.4350045919418335,
      "learning_rate": 2e-05,
      "loss": 0.8126,
      "step": 8340
    },
    {
      "epoch": 0.6937520771020272,
      "grad_norm": 0.4239192605018616,
      "learning_rate": 2e-05,
      "loss": 0.7787,
      "step": 8350
    },
    {
      "epoch": 0.6945829179129279,
      "grad_norm": 0.4061550498008728,
      "learning_rate": 2e-05,
      "loss": 0.7932,
      "step": 8360
    },
    {
      "epoch": 0.6954137587238285,
      "grad_norm": 0.43523818254470825,
      "learning_rate": 2e-05,
      "loss": 0.7908,
      "step": 8370
    },
    {
      "epoch": 0.6962445995347292,
      "grad_norm": 0.4558693468570709,
      "learning_rate": 2e-05,
      "loss": 0.8273,
      "step": 8380
    },
    {
      "epoch": 0.6970754403456297,
      "grad_norm": 0.43846675753593445,
      "learning_rate": 2e-05,
      "loss": 0.8021,
      "step": 8390
    },
    {
      "epoch": 0.6979062811565304,
      "grad_norm": 0.45141398906707764,
      "learning_rate": 2e-05,
      "loss": 0.8191,
      "step": 8400
    },
    {
      "epoch": 0.698737121967431,
      "grad_norm": 0.44449442625045776,
      "learning_rate": 2e-05,
      "loss": 0.7878,
      "step": 8410
    },
    {
      "epoch": 0.6995679627783317,
      "grad_norm": 0.4176330864429474,
      "learning_rate": 2e-05,
      "loss": 0.7806,
      "step": 8420
    },
    {
      "epoch": 0.7003988035892323,
      "grad_norm": 0.4313443899154663,
      "learning_rate": 2e-05,
      "loss": 0.8125,
      "step": 8430
    },
    {
      "epoch": 0.7012296444001329,
      "grad_norm": 0.4139100909233093,
      "learning_rate": 2e-05,
      "loss": 0.7838,
      "step": 8440
    },
    {
      "epoch": 0.7020604852110336,
      "grad_norm": 0.42824503779411316,
      "learning_rate": 2e-05,
      "loss": 0.804,
      "step": 8450
    },
    {
      "epoch": 0.7028913260219342,
      "grad_norm": 0.45131051540374756,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 8460
    },
    {
      "epoch": 0.7037221668328348,
      "grad_norm": 0.4639829397201538,
      "learning_rate": 2e-05,
      "loss": 0.8131,
      "step": 8470
    },
    {
      "epoch": 0.7045530076437354,
      "grad_norm": 0.4468565881252289,
      "learning_rate": 2e-05,
      "loss": 0.8242,
      "step": 8480
    },
    {
      "epoch": 0.7053838484546361,
      "grad_norm": 0.4297695755958557,
      "learning_rate": 2e-05,
      "loss": 0.8076,
      "step": 8490
    },
    {
      "epoch": 0.7062146892655368,
      "grad_norm": 0.4435446858406067,
      "learning_rate": 2e-05,
      "loss": 0.7961,
      "step": 8500
    },
    {
      "epoch": 0.7070455300764373,
      "grad_norm": 0.41391634941101074,
      "learning_rate": 2e-05,
      "loss": 0.764,
      "step": 8510
    },
    {
      "epoch": 0.707876370887338,
      "grad_norm": 0.4349653124809265,
      "learning_rate": 2e-05,
      "loss": 0.7721,
      "step": 8520
    },
    {
      "epoch": 0.7087072116982386,
      "grad_norm": 0.43902266025543213,
      "learning_rate": 2e-05,
      "loss": 0.7723,
      "step": 8530
    },
    {
      "epoch": 0.7095380525091393,
      "grad_norm": 0.44854775071144104,
      "learning_rate": 2e-05,
      "loss": 0.7487,
      "step": 8540
    },
    {
      "epoch": 0.7103688933200398,
      "grad_norm": 0.4160611927509308,
      "learning_rate": 2e-05,
      "loss": 0.8002,
      "step": 8550
    },
    {
      "epoch": 0.7111997341309405,
      "grad_norm": 0.46068301796913147,
      "learning_rate": 2e-05,
      "loss": 0.8182,
      "step": 8560
    },
    {
      "epoch": 0.7120305749418412,
      "grad_norm": 0.44864851236343384,
      "learning_rate": 2e-05,
      "loss": 0.8048,
      "step": 8570
    },
    {
      "epoch": 0.7128614157527418,
      "grad_norm": 0.43809598684310913,
      "learning_rate": 2e-05,
      "loss": 0.8128,
      "step": 8580
    },
    {
      "epoch": 0.7136922565636424,
      "grad_norm": 0.45387062430381775,
      "learning_rate": 2e-05,
      "loss": 0.797,
      "step": 8590
    },
    {
      "epoch": 0.714523097374543,
      "grad_norm": 0.4038558900356293,
      "learning_rate": 2e-05,
      "loss": 0.7828,
      "step": 8600
    },
    {
      "epoch": 0.7153539381854437,
      "grad_norm": 0.4200127422809601,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 8610
    },
    {
      "epoch": 0.7161847789963443,
      "grad_norm": 0.40347054600715637,
      "learning_rate": 2e-05,
      "loss": 0.7916,
      "step": 8620
    },
    {
      "epoch": 0.7170156198072449,
      "grad_norm": 0.4313322901725769,
      "learning_rate": 2e-05,
      "loss": 0.7781,
      "step": 8630
    },
    {
      "epoch": 0.7178464606181456,
      "grad_norm": 0.4249413311481476,
      "learning_rate": 2e-05,
      "loss": 0.7655,
      "step": 8640
    },
    {
      "epoch": 0.7186773014290462,
      "grad_norm": 0.4298620820045471,
      "learning_rate": 2e-05,
      "loss": 0.8008,
      "step": 8650
    },
    {
      "epoch": 0.7195081422399469,
      "grad_norm": 0.46679291129112244,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 8660
    },
    {
      "epoch": 0.7203389830508474,
      "grad_norm": 0.42921367287635803,
      "learning_rate": 2e-05,
      "loss": 0.8205,
      "step": 8670
    },
    {
      "epoch": 0.7211698238617481,
      "grad_norm": 0.43988123536109924,
      "learning_rate": 2e-05,
      "loss": 0.7728,
      "step": 8680
    },
    {
      "epoch": 0.7220006646726487,
      "grad_norm": 0.45344874262809753,
      "learning_rate": 2e-05,
      "loss": 0.789,
      "step": 8690
    },
    {
      "epoch": 0.7228315054835494,
      "grad_norm": 0.4028540551662445,
      "learning_rate": 2e-05,
      "loss": 0.777,
      "step": 8700
    },
    {
      "epoch": 0.72366234629445,
      "grad_norm": 0.4108804762363434,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 8710
    },
    {
      "epoch": 0.7244931871053506,
      "grad_norm": 0.4333289861679077,
      "learning_rate": 2e-05,
      "loss": 0.7697,
      "step": 8720
    },
    {
      "epoch": 0.7253240279162513,
      "grad_norm": 0.4833255410194397,
      "learning_rate": 2e-05,
      "loss": 0.8162,
      "step": 8730
    },
    {
      "epoch": 0.7261548687271518,
      "grad_norm": 0.4875302314758301,
      "learning_rate": 2e-05,
      "loss": 0.8345,
      "step": 8740
    },
    {
      "epoch": 0.7269857095380525,
      "grad_norm": 0.5643032193183899,
      "learning_rate": 2e-05,
      "loss": 0.8033,
      "step": 8750
    },
    {
      "epoch": 0.7278165503489531,
      "grad_norm": 0.46173685789108276,
      "learning_rate": 2e-05,
      "loss": 0.7733,
      "step": 8760
    },
    {
      "epoch": 0.7286473911598538,
      "grad_norm": 0.47278663516044617,
      "learning_rate": 2e-05,
      "loss": 0.7683,
      "step": 8770
    },
    {
      "epoch": 0.7294782319707545,
      "grad_norm": 0.42943331599235535,
      "learning_rate": 2e-05,
      "loss": 0.8272,
      "step": 8780
    },
    {
      "epoch": 0.730309072781655,
      "grad_norm": 0.4042111933231354,
      "learning_rate": 2e-05,
      "loss": 0.7852,
      "step": 8790
    },
    {
      "epoch": 0.7311399135925557,
      "grad_norm": 0.40895432233810425,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 8800
    },
    {
      "epoch": 0.7319707544034563,
      "grad_norm": 0.4059562087059021,
      "learning_rate": 2e-05,
      "loss": 0.7955,
      "step": 8810
    },
    {
      "epoch": 0.732801595214357,
      "grad_norm": 0.43332988023757935,
      "learning_rate": 2e-05,
      "loss": 0.7762,
      "step": 8820
    },
    {
      "epoch": 0.7336324360252575,
      "grad_norm": 0.429324209690094,
      "learning_rate": 2e-05,
      "loss": 0.7609,
      "step": 8830
    },
    {
      "epoch": 0.7344632768361582,
      "grad_norm": 0.4494699239730835,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 8840
    },
    {
      "epoch": 0.7352941176470589,
      "grad_norm": 0.43718820810317993,
      "learning_rate": 2e-05,
      "loss": 0.776,
      "step": 8850
    },
    {
      "epoch": 0.7361249584579594,
      "grad_norm": 0.4247332215309143,
      "learning_rate": 2e-05,
      "loss": 0.8008,
      "step": 8860
    },
    {
      "epoch": 0.7369557992688601,
      "grad_norm": 0.5780029892921448,
      "learning_rate": 2e-05,
      "loss": 0.774,
      "step": 8870
    },
    {
      "epoch": 0.7377866400797607,
      "grad_norm": 0.43193677067756653,
      "learning_rate": 2e-05,
      "loss": 0.7944,
      "step": 8880
    },
    {
      "epoch": 0.7386174808906614,
      "grad_norm": 0.45509257912635803,
      "learning_rate": 2e-05,
      "loss": 0.8026,
      "step": 8890
    },
    {
      "epoch": 0.7394483217015619,
      "grad_norm": 0.505480945110321,
      "learning_rate": 2e-05,
      "loss": 0.7997,
      "step": 8900
    },
    {
      "epoch": 0.7402791625124626,
      "grad_norm": 0.4052199423313141,
      "learning_rate": 2e-05,
      "loss": 0.7589,
      "step": 8910
    },
    {
      "epoch": 0.7411100033233633,
      "grad_norm": 0.4426860213279724,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 8920
    },
    {
      "epoch": 0.7419408441342639,
      "grad_norm": 0.4378291368484497,
      "learning_rate": 2e-05,
      "loss": 0.8104,
      "step": 8930
    },
    {
      "epoch": 0.7427716849451645,
      "grad_norm": 0.438697874546051,
      "learning_rate": 2e-05,
      "loss": 0.7872,
      "step": 8940
    },
    {
      "epoch": 0.7436025257560651,
      "grad_norm": 0.48882898688316345,
      "learning_rate": 2e-05,
      "loss": 0.7967,
      "step": 8950
    },
    {
      "epoch": 0.7444333665669658,
      "grad_norm": 0.4368053674697876,
      "learning_rate": 2e-05,
      "loss": 0.7519,
      "step": 8960
    },
    {
      "epoch": 0.7452642073778664,
      "grad_norm": 0.4114673137664795,
      "learning_rate": 2e-05,
      "loss": 0.801,
      "step": 8970
    },
    {
      "epoch": 0.746095048188767,
      "grad_norm": 0.44325071573257446,
      "learning_rate": 2e-05,
      "loss": 0.7793,
      "step": 8980
    },
    {
      "epoch": 0.7469258889996676,
      "grad_norm": 0.45806875824928284,
      "learning_rate": 2e-05,
      "loss": 0.7909,
      "step": 8990
    },
    {
      "epoch": 0.7477567298105683,
      "grad_norm": 0.4482547342777252,
      "learning_rate": 2e-05,
      "loss": 0.7983,
      "step": 9000
    },
    {
      "epoch": 0.748587570621469,
      "grad_norm": 0.4476955533027649,
      "learning_rate": 2e-05,
      "loss": 0.8197,
      "step": 9010
    },
    {
      "epoch": 0.7494184114323695,
      "grad_norm": 0.4792895019054413,
      "learning_rate": 2e-05,
      "loss": 0.7714,
      "step": 9020
    },
    {
      "epoch": 0.7502492522432702,
      "grad_norm": 0.43752655386924744,
      "learning_rate": 2e-05,
      "loss": 0.8078,
      "step": 9030
    },
    {
      "epoch": 0.7510800930541708,
      "grad_norm": 0.4342658519744873,
      "learning_rate": 2e-05,
      "loss": 0.7731,
      "step": 9040
    },
    {
      "epoch": 0.7519109338650715,
      "grad_norm": 0.44235146045684814,
      "learning_rate": 2e-05,
      "loss": 0.834,
      "step": 9050
    },
    {
      "epoch": 0.752741774675972,
      "grad_norm": 0.4021388292312622,
      "learning_rate": 2e-05,
      "loss": 0.7697,
      "step": 9060
    },
    {
      "epoch": 0.7535726154868727,
      "grad_norm": 0.4525681734085083,
      "learning_rate": 2e-05,
      "loss": 0.8216,
      "step": 9070
    },
    {
      "epoch": 0.7544034562977734,
      "grad_norm": 0.4465608596801758,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 9080
    },
    {
      "epoch": 0.755234297108674,
      "grad_norm": 0.44454774260520935,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 9090
    },
    {
      "epoch": 0.7560651379195746,
      "grad_norm": 0.4895791709423065,
      "learning_rate": 2e-05,
      "loss": 0.7847,
      "step": 9100
    },
    {
      "epoch": 0.7568959787304752,
      "grad_norm": 0.4609822630882263,
      "learning_rate": 2e-05,
      "loss": 0.7934,
      "step": 9110
    },
    {
      "epoch": 0.7577268195413759,
      "grad_norm": 0.4564991593360901,
      "learning_rate": 2e-05,
      "loss": 0.8063,
      "step": 9120
    },
    {
      "epoch": 0.7585576603522765,
      "grad_norm": 0.40496355295181274,
      "learning_rate": 2e-05,
      "loss": 0.7713,
      "step": 9130
    },
    {
      "epoch": 0.7593885011631771,
      "grad_norm": 0.436903715133667,
      "learning_rate": 2e-05,
      "loss": 0.811,
      "step": 9140
    },
    {
      "epoch": 0.7602193419740778,
      "grad_norm": 0.4454230070114136,
      "learning_rate": 2e-05,
      "loss": 0.8164,
      "step": 9150
    },
    {
      "epoch": 0.7610501827849784,
      "grad_norm": 0.45278969407081604,
      "learning_rate": 2e-05,
      "loss": 0.7644,
      "step": 9160
    },
    {
      "epoch": 0.7618810235958791,
      "grad_norm": 0.4294947385787964,
      "learning_rate": 2e-05,
      "loss": 0.7731,
      "step": 9170
    },
    {
      "epoch": 0.7627118644067796,
      "grad_norm": 0.4218863844871521,
      "learning_rate": 2e-05,
      "loss": 0.8067,
      "step": 9180
    },
    {
      "epoch": 0.7635427052176803,
      "grad_norm": 0.3931097090244293,
      "learning_rate": 2e-05,
      "loss": 0.7775,
      "step": 9190
    },
    {
      "epoch": 0.7643735460285809,
      "grad_norm": 1.5174784660339355,
      "learning_rate": 2e-05,
      "loss": 0.7716,
      "step": 9200
    },
    {
      "epoch": 0.7652043868394816,
      "grad_norm": 0.40769028663635254,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 9210
    },
    {
      "epoch": 0.7660352276503822,
      "grad_norm": 0.43936997652053833,
      "learning_rate": 2e-05,
      "loss": 0.7931,
      "step": 9220
    },
    {
      "epoch": 0.7668660684612828,
      "grad_norm": 0.42210590839385986,
      "learning_rate": 2e-05,
      "loss": 0.7932,
      "step": 9230
    },
    {
      "epoch": 0.7676969092721835,
      "grad_norm": 0.42505377531051636,
      "learning_rate": 2e-05,
      "loss": 0.7804,
      "step": 9240
    },
    {
      "epoch": 0.768527750083084,
      "grad_norm": 0.42359426617622375,
      "learning_rate": 2e-05,
      "loss": 0.8094,
      "step": 9250
    },
    {
      "epoch": 0.7693585908939847,
      "grad_norm": 0.43289321660995483,
      "learning_rate": 2e-05,
      "loss": 0.819,
      "step": 9260
    },
    {
      "epoch": 0.7701894317048853,
      "grad_norm": 0.4250054657459259,
      "learning_rate": 2e-05,
      "loss": 0.7786,
      "step": 9270
    },
    {
      "epoch": 0.771020272515786,
      "grad_norm": 0.4287359118461609,
      "learning_rate": 2e-05,
      "loss": 0.7705,
      "step": 9280
    },
    {
      "epoch": 0.7718511133266867,
      "grad_norm": 0.41876670718193054,
      "learning_rate": 2e-05,
      "loss": 0.7896,
      "step": 9290
    },
    {
      "epoch": 0.7726819541375872,
      "grad_norm": 0.44837886095046997,
      "learning_rate": 2e-05,
      "loss": 0.8049,
      "step": 9300
    },
    {
      "epoch": 0.7735127949484879,
      "grad_norm": 0.4287818670272827,
      "learning_rate": 2e-05,
      "loss": 0.7641,
      "step": 9310
    },
    {
      "epoch": 0.7743436357593885,
      "grad_norm": 0.43126338720321655,
      "learning_rate": 2e-05,
      "loss": 0.7978,
      "step": 9320
    },
    {
      "epoch": 0.7751744765702892,
      "grad_norm": 0.4334515631198883,
      "learning_rate": 2e-05,
      "loss": 0.8107,
      "step": 9330
    },
    {
      "epoch": 0.7760053173811897,
      "grad_norm": 0.4678569436073303,
      "learning_rate": 2e-05,
      "loss": 0.7564,
      "step": 9340
    },
    {
      "epoch": 0.7768361581920904,
      "grad_norm": 0.46585744619369507,
      "learning_rate": 2e-05,
      "loss": 0.7838,
      "step": 9350
    },
    {
      "epoch": 0.7776669990029911,
      "grad_norm": 0.4153449237346649,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 9360
    },
    {
      "epoch": 0.7784978398138916,
      "grad_norm": 0.5249794125556946,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 9370
    },
    {
      "epoch": 0.7793286806247923,
      "grad_norm": 0.42529407143592834,
      "learning_rate": 2e-05,
      "loss": 0.8215,
      "step": 9380
    },
    {
      "epoch": 0.7801595214356929,
      "grad_norm": 0.4424443542957306,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 9390
    },
    {
      "epoch": 0.7809903622465936,
      "grad_norm": 0.43509283661842346,
      "learning_rate": 2e-05,
      "loss": 0.7928,
      "step": 9400
    },
    {
      "epoch": 0.7818212030574941,
      "grad_norm": 0.4458855092525482,
      "learning_rate": 2e-05,
      "loss": 0.7968,
      "step": 9410
    },
    {
      "epoch": 0.7826520438683948,
      "grad_norm": 0.5069590210914612,
      "learning_rate": 2e-05,
      "loss": 0.7818,
      "step": 9420
    },
    {
      "epoch": 0.7834828846792955,
      "grad_norm": 0.432020902633667,
      "learning_rate": 2e-05,
      "loss": 0.77,
      "step": 9430
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 0.4301707148551941,
      "learning_rate": 2e-05,
      "loss": 0.7853,
      "step": 9440
    },
    {
      "epoch": 0.7851445663010967,
      "grad_norm": 0.4317227900028229,
      "learning_rate": 2e-05,
      "loss": 0.8016,
      "step": 9450
    },
    {
      "epoch": 0.7859754071119973,
      "grad_norm": 0.46192315220832825,
      "learning_rate": 2e-05,
      "loss": 0.7807,
      "step": 9460
    },
    {
      "epoch": 0.786806247922898,
      "grad_norm": 0.4214765429496765,
      "learning_rate": 2e-05,
      "loss": 0.7828,
      "step": 9470
    },
    {
      "epoch": 0.7876370887337986,
      "grad_norm": 0.41663551330566406,
      "learning_rate": 2e-05,
      "loss": 0.7837,
      "step": 9480
    },
    {
      "epoch": 0.7884679295446992,
      "grad_norm": 0.4648173749446869,
      "learning_rate": 2e-05,
      "loss": 0.7847,
      "step": 9490
    },
    {
      "epoch": 0.7892987703555999,
      "grad_norm": 0.42017054557800293,
      "learning_rate": 2e-05,
      "loss": 0.7912,
      "step": 9500
    },
    {
      "epoch": 0.7901296111665005,
      "grad_norm": 0.4277028441429138,
      "learning_rate": 2e-05,
      "loss": 0.7863,
      "step": 9510
    },
    {
      "epoch": 0.7909604519774012,
      "grad_norm": 0.42275169491767883,
      "learning_rate": 2e-05,
      "loss": 0.7886,
      "step": 9520
    },
    {
      "epoch": 0.7917912927883017,
      "grad_norm": 0.42866823077201843,
      "learning_rate": 2e-05,
      "loss": 0.7654,
      "step": 9530
    },
    {
      "epoch": 0.7926221335992024,
      "grad_norm": 0.44645795226097107,
      "learning_rate": 2e-05,
      "loss": 0.7637,
      "step": 9540
    },
    {
      "epoch": 0.793452974410103,
      "grad_norm": 0.4942845106124878,
      "learning_rate": 2e-05,
      "loss": 0.8154,
      "step": 9550
    },
    {
      "epoch": 0.7942838152210037,
      "grad_norm": 0.41457462310791016,
      "learning_rate": 2e-05,
      "loss": 0.7873,
      "step": 9560
    },
    {
      "epoch": 0.7951146560319042,
      "grad_norm": 0.4006474018096924,
      "learning_rate": 2e-05,
      "loss": 0.7858,
      "step": 9570
    },
    {
      "epoch": 0.7959454968428049,
      "grad_norm": 0.5160494446754456,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 9580
    },
    {
      "epoch": 0.7967763376537056,
      "grad_norm": 0.428366482257843,
      "learning_rate": 2e-05,
      "loss": 0.7615,
      "step": 9590
    },
    {
      "epoch": 0.7976071784646062,
      "grad_norm": 0.4609552025794983,
      "learning_rate": 2e-05,
      "loss": 0.7859,
      "step": 9600
    },
    {
      "epoch": 0.7984380192755068,
      "grad_norm": 0.4424075782299042,
      "learning_rate": 2e-05,
      "loss": 0.7776,
      "step": 9610
    },
    {
      "epoch": 0.7992688600864074,
      "grad_norm": 0.46239057183265686,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 9620
    },
    {
      "epoch": 0.8000997008973081,
      "grad_norm": 0.4523656964302063,
      "learning_rate": 2e-05,
      "loss": 0.7674,
      "step": 9630
    },
    {
      "epoch": 0.8009305417082087,
      "grad_norm": 0.4444538950920105,
      "learning_rate": 2e-05,
      "loss": 0.7924,
      "step": 9640
    },
    {
      "epoch": 0.8017613825191093,
      "grad_norm": 0.40881505608558655,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 9650
    },
    {
      "epoch": 0.80259222333001,
      "grad_norm": 0.4172127842903137,
      "learning_rate": 2e-05,
      "loss": 0.7742,
      "step": 9660
    },
    {
      "epoch": 0.8034230641409106,
      "grad_norm": 0.5496156811714172,
      "learning_rate": 2e-05,
      "loss": 0.8282,
      "step": 9670
    },
    {
      "epoch": 0.8042539049518113,
      "grad_norm": 0.43811842799186707,
      "learning_rate": 2e-05,
      "loss": 0.7803,
      "step": 9680
    },
    {
      "epoch": 0.8050847457627118,
      "grad_norm": 0.43851104378700256,
      "learning_rate": 2e-05,
      "loss": 0.7176,
      "step": 9690
    },
    {
      "epoch": 0.8059155865736125,
      "grad_norm": 0.4535624086856842,
      "learning_rate": 2e-05,
      "loss": 0.8137,
      "step": 9700
    },
    {
      "epoch": 0.8067464273845131,
      "grad_norm": 0.4240114986896515,
      "learning_rate": 2e-05,
      "loss": 0.8135,
      "step": 9710
    },
    {
      "epoch": 0.8075772681954138,
      "grad_norm": 0.42399826645851135,
      "learning_rate": 2e-05,
      "loss": 0.7933,
      "step": 9720
    },
    {
      "epoch": 0.8084081090063144,
      "grad_norm": 0.42481452226638794,
      "learning_rate": 2e-05,
      "loss": 0.7726,
      "step": 9730
    },
    {
      "epoch": 0.809238949817215,
      "grad_norm": 0.4716450572013855,
      "learning_rate": 2e-05,
      "loss": 0.7758,
      "step": 9740
    },
    {
      "epoch": 0.8100697906281157,
      "grad_norm": 0.40037110447883606,
      "learning_rate": 2e-05,
      "loss": 0.7766,
      "step": 9750
    },
    {
      "epoch": 0.8109006314390163,
      "grad_norm": 0.4124835431575775,
      "learning_rate": 2e-05,
      "loss": 0.7546,
      "step": 9760
    },
    {
      "epoch": 0.8117314722499169,
      "grad_norm": 0.4324896037578583,
      "learning_rate": 2e-05,
      "loss": 0.7504,
      "step": 9770
    },
    {
      "epoch": 0.8125623130608175,
      "grad_norm": 0.4292924702167511,
      "learning_rate": 2e-05,
      "loss": 0.8172,
      "step": 9780
    },
    {
      "epoch": 0.8133931538717182,
      "grad_norm": 0.4836083948612213,
      "learning_rate": 2e-05,
      "loss": 0.7742,
      "step": 9790
    },
    {
      "epoch": 0.8142239946826189,
      "grad_norm": 0.431721031665802,
      "learning_rate": 2e-05,
      "loss": 0.7814,
      "step": 9800
    },
    {
      "epoch": 0.8150548354935194,
      "grad_norm": 0.4619964063167572,
      "learning_rate": 2e-05,
      "loss": 0.7841,
      "step": 9810
    },
    {
      "epoch": 0.8158856763044201,
      "grad_norm": 0.46127644181251526,
      "learning_rate": 2e-05,
      "loss": 0.789,
      "step": 9820
    },
    {
      "epoch": 0.8167165171153207,
      "grad_norm": 0.46011772751808167,
      "learning_rate": 2e-05,
      "loss": 0.8089,
      "step": 9830
    },
    {
      "epoch": 0.8175473579262214,
      "grad_norm": 0.43971768021583557,
      "learning_rate": 2e-05,
      "loss": 0.7882,
      "step": 9840
    },
    {
      "epoch": 0.8183781987371219,
      "grad_norm": 0.4247598350048065,
      "learning_rate": 2e-05,
      "loss": 0.7776,
      "step": 9850
    },
    {
      "epoch": 0.8192090395480226,
      "grad_norm": 0.46960535645484924,
      "learning_rate": 2e-05,
      "loss": 0.7808,
      "step": 9860
    },
    {
      "epoch": 0.8200398803589233,
      "grad_norm": 0.5120652318000793,
      "learning_rate": 2e-05,
      "loss": 0.7836,
      "step": 9870
    },
    {
      "epoch": 0.8208707211698238,
      "grad_norm": 0.44843247532844543,
      "learning_rate": 2e-05,
      "loss": 0.8079,
      "step": 9880
    },
    {
      "epoch": 0.8217015619807245,
      "grad_norm": 0.42369988560676575,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 9890
    },
    {
      "epoch": 0.8225324027916251,
      "grad_norm": 0.5601595044136047,
      "learning_rate": 2e-05,
      "loss": 0.7434,
      "step": 9900
    },
    {
      "epoch": 0.8233632436025258,
      "grad_norm": 0.45025619864463806,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 9910
    },
    {
      "epoch": 0.8241940844134263,
      "grad_norm": 0.4423345923423767,
      "learning_rate": 2e-05,
      "loss": 0.7713,
      "step": 9920
    },
    {
      "epoch": 0.825024925224327,
      "grad_norm": 0.45578640699386597,
      "learning_rate": 2e-05,
      "loss": 0.7888,
      "step": 9930
    },
    {
      "epoch": 0.8258557660352277,
      "grad_norm": 0.4350617527961731,
      "learning_rate": 2e-05,
      "loss": 0.773,
      "step": 9940
    },
    {
      "epoch": 0.8266866068461283,
      "grad_norm": 0.4306197166442871,
      "learning_rate": 2e-05,
      "loss": 0.7549,
      "step": 9950
    },
    {
      "epoch": 0.827517447657029,
      "grad_norm": 0.4322715699672699,
      "learning_rate": 2e-05,
      "loss": 0.7638,
      "step": 9960
    },
    {
      "epoch": 0.8283482884679295,
      "grad_norm": 0.3968597650527954,
      "learning_rate": 2e-05,
      "loss": 0.7757,
      "step": 9970
    },
    {
      "epoch": 0.8291791292788302,
      "grad_norm": 0.42919445037841797,
      "learning_rate": 2e-05,
      "loss": 0.7684,
      "step": 9980
    },
    {
      "epoch": 0.8300099700897308,
      "grad_norm": 0.5138984322547913,
      "learning_rate": 2e-05,
      "loss": 0.8066,
      "step": 9990
    },
    {
      "epoch": 0.8308408109006314,
      "grad_norm": 0.4389176368713379,
      "learning_rate": 2e-05,
      "loss": 0.7942,
      "step": 10000
    },
    {
      "epoch": 0.8316716517115321,
      "grad_norm": 0.45706477761268616,
      "learning_rate": 2e-05,
      "loss": 0.8076,
      "step": 10010
    },
    {
      "epoch": 0.8325024925224327,
      "grad_norm": 0.44627833366394043,
      "learning_rate": 2e-05,
      "loss": 0.7781,
      "step": 10020
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.4373287260532379,
      "learning_rate": 2e-05,
      "loss": 0.7763,
      "step": 10030
    },
    {
      "epoch": 0.8341641741442339,
      "grad_norm": 0.42560192942619324,
      "learning_rate": 2e-05,
      "loss": 0.7578,
      "step": 10040
    },
    {
      "epoch": 0.8349950149551346,
      "grad_norm": 0.4728098511695862,
      "learning_rate": 2e-05,
      "loss": 0.7829,
      "step": 10050
    },
    {
      "epoch": 0.8358258557660352,
      "grad_norm": 0.472596675157547,
      "learning_rate": 2e-05,
      "loss": 0.7887,
      "step": 10060
    },
    {
      "epoch": 0.8366566965769359,
      "grad_norm": 0.4332342743873596,
      "learning_rate": 2e-05,
      "loss": 0.7725,
      "step": 10070
    },
    {
      "epoch": 0.8374875373878365,
      "grad_norm": 0.4289112389087677,
      "learning_rate": 2e-05,
      "loss": 0.8137,
      "step": 10080
    },
    {
      "epoch": 0.8383183781987371,
      "grad_norm": 0.45944467186927795,
      "learning_rate": 2e-05,
      "loss": 0.7896,
      "step": 10090
    },
    {
      "epoch": 0.8391492190096378,
      "grad_norm": 0.4143383204936981,
      "learning_rate": 2e-05,
      "loss": 0.7866,
      "step": 10100
    },
    {
      "epoch": 0.8399800598205384,
      "grad_norm": 0.47001898288726807,
      "learning_rate": 2e-05,
      "loss": 0.7623,
      "step": 10110
    },
    {
      "epoch": 0.840810900631439,
      "grad_norm": 0.4037744998931885,
      "learning_rate": 2e-05,
      "loss": 0.7953,
      "step": 10120
    },
    {
      "epoch": 0.8416417414423396,
      "grad_norm": 0.45539841055870056,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 10130
    },
    {
      "epoch": 0.8424725822532403,
      "grad_norm": 0.43888700008392334,
      "learning_rate": 2e-05,
      "loss": 0.8126,
      "step": 10140
    },
    {
      "epoch": 0.8433034230641409,
      "grad_norm": 0.4929424524307251,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 10150
    },
    {
      "epoch": 0.8441342638750415,
      "grad_norm": 0.43418776988983154,
      "learning_rate": 2e-05,
      "loss": 0.7652,
      "step": 10160
    },
    {
      "epoch": 0.8449651046859422,
      "grad_norm": 0.466757595539093,
      "learning_rate": 2e-05,
      "loss": 0.7714,
      "step": 10170
    },
    {
      "epoch": 0.8457959454968428,
      "grad_norm": 0.4594809114933014,
      "learning_rate": 1.999957358086795e-05,
      "loss": 0.7953,
      "step": 10180
    },
    {
      "epoch": 0.8466267863077435,
      "grad_norm": 0.41535869240760803,
      "learning_rate": 1.9991993813020483e-05,
      "loss": 0.7818,
      "step": 10190
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 0.487265408039093,
      "learning_rate": 1.9974946373028628e-05,
      "loss": 0.7831,
      "step": 10200
    },
    {
      "epoch": 0.8482884679295447,
      "grad_norm": 0.4355158507823944,
      "learning_rate": 1.994844749501764e-05,
      "loss": 0.803,
      "step": 10210
    },
    {
      "epoch": 0.8491193087404453,
      "grad_norm": 0.44257068634033203,
      "learning_rate": 1.9912522413632988e-05,
      "loss": 0.8303,
      "step": 10220
    },
    {
      "epoch": 0.849950149551346,
      "grad_norm": 0.45267757773399353,
      "learning_rate": 1.9867205340009656e-05,
      "loss": 0.7943,
      "step": 10230
    },
    {
      "epoch": 0.8507809903622466,
      "grad_norm": 0.46746423840522766,
      "learning_rate": 1.9812539429193164e-05,
      "loss": 0.8083,
      "step": 10240
    },
    {
      "epoch": 0.8516118311731472,
      "grad_norm": 0.44131624698638916,
      "learning_rate": 1.9748576739043423e-05,
      "loss": 0.8147,
      "step": 10250
    },
    {
      "epoch": 0.8524426719840479,
      "grad_norm": 0.6074473261833191,
      "learning_rate": 1.967537818066047e-05,
      "loss": 0.7871,
      "step": 10260
    },
    {
      "epoch": 0.8532735127949485,
      "grad_norm": 0.6140527725219727,
      "learning_rate": 1.959301346037941e-05,
      "loss": 0.7957,
      "step": 10270
    },
    {
      "epoch": 0.8541043536058491,
      "grad_norm": 0.43889448046684265,
      "learning_rate": 1.9501561013389646e-05,
      "loss": 0.7811,
      "step": 10280
    },
    {
      "epoch": 0.8549351944167497,
      "grad_norm": 0.48842984437942505,
      "learning_rate": 1.9401107929041795e-05,
      "loss": 0.7661,
      "step": 10290
    },
    {
      "epoch": 0.8557660352276504,
      "grad_norm": 0.4347827136516571,
      "learning_rate": 1.9291749867913225e-05,
      "loss": 0.775,
      "step": 10300
    },
    {
      "epoch": 0.8565968760385511,
      "grad_norm": 0.432992547750473,
      "learning_rate": 1.91735909707114e-05,
      "loss": 0.8087,
      "step": 10310
    },
    {
      "epoch": 0.8574277168494516,
      "grad_norm": 0.43306925892829895,
      "learning_rate": 1.9046743759101545e-05,
      "loss": 0.7911,
      "step": 10320
    },
    {
      "epoch": 0.8582585576603523,
      "grad_norm": 0.4238404929637909,
      "learning_rate": 1.8911329028553324e-05,
      "loss": 0.7817,
      "step": 10330
    },
    {
      "epoch": 0.8590893984712529,
      "grad_norm": 0.45451483130455017,
      "learning_rate": 1.876747573330835e-05,
      "loss": 0.7444,
      "step": 10340
    },
    {
      "epoch": 0.8599202392821536,
      "grad_norm": 0.42847272753715515,
      "learning_rate": 1.8615320863578244e-05,
      "loss": 0.7637,
      "step": 10350
    },
    {
      "epoch": 0.8607510800930541,
      "grad_norm": 0.6048896312713623,
      "learning_rate": 1.8455009315090036e-05,
      "loss": 0.7866,
      "step": 10360
    },
    {
      "epoch": 0.8615819209039548,
      "grad_norm": 0.49839094281196594,
      "learning_rate": 1.8286693751103294e-05,
      "loss": 0.7767,
      "step": 10370
    },
    {
      "epoch": 0.8624127617148555,
      "grad_norm": 0.4350515604019165,
      "learning_rate": 1.811053445703024e-05,
      "loss": 0.7745,
      "step": 10380
    },
    {
      "epoch": 0.863243602525756,
      "grad_norm": 0.4588352143764496,
      "learning_rate": 1.7926699187797375e-05,
      "loss": 0.7872,
      "step": 10390
    },
    {
      "epoch": 0.8640744433366567,
      "grad_norm": 0.4227766990661621,
      "learning_rate": 1.7735363008093995e-05,
      "loss": 0.7774,
      "step": 10400
    },
    {
      "epoch": 0.8649052841475573,
      "grad_norm": 0.4322856068611145,
      "learning_rate": 1.7536708125659577e-05,
      "loss": 0.781,
      "step": 10410
    },
    {
      "epoch": 0.865736124958458,
      "grad_norm": 0.4472157955169678,
      "learning_rate": 1.733092371776905e-05,
      "loss": 0.7574,
      "step": 10420
    },
    {
      "epoch": 0.8665669657693585,
      "grad_norm": 0.44070038199424744,
      "learning_rate": 1.7118205751080915e-05,
      "loss": 0.8039,
      "step": 10430
    },
    {
      "epoch": 0.8673978065802592,
      "grad_norm": 0.4375724494457245,
      "learning_rate": 1.6898756795019935e-05,
      "loss": 0.7484,
      "step": 10440
    },
    {
      "epoch": 0.8682286473911599,
      "grad_norm": 0.44008868932724,
      "learning_rate": 1.667278582887211e-05,
      "loss": 0.7215,
      "step": 10450
    },
    {
      "epoch": 0.8690594882020605,
      "grad_norm": 0.43110916018486023,
      "learning_rate": 1.6440508042775545e-05,
      "loss": 0.7865,
      "step": 10460
    },
    {
      "epoch": 0.8698903290129611,
      "grad_norm": 0.4576956033706665,
      "learning_rate": 1.6202144632796796e-05,
      "loss": 0.7443,
      "step": 10470
    },
    {
      "epoch": 0.8707211698238617,
      "grad_norm": 0.3973090946674347,
      "learning_rate": 1.595792259028783e-05,
      "loss": 0.7492,
      "step": 10480
    },
    {
      "epoch": 0.8715520106347624,
      "grad_norm": 0.4181090295314789,
      "learning_rate": 1.570807448572419e-05,
      "loss": 0.78,
      "step": 10490
    },
    {
      "epoch": 0.872382851445663,
      "grad_norm": 0.42761266231536865,
      "learning_rate": 1.5452838247230197e-05,
      "loss": 0.7545,
      "step": 10500
    },
    {
      "epoch": 0.8732136922565636,
      "grad_norm": 0.44214800000190735,
      "learning_rate": 1.5192456934002106e-05,
      "loss": 0.7807,
      "step": 10510
    },
    {
      "epoch": 0.8740445330674643,
      "grad_norm": 0.41227635741233826,
      "learning_rate": 1.492717850484502e-05,
      "loss": 0.7753,
      "step": 10520
    },
    {
      "epoch": 0.8748753738783649,
      "grad_norm": 0.42174994945526123,
      "learning_rate": 1.4657255582043904e-05,
      "loss": 0.779,
      "step": 10530
    },
    {
      "epoch": 0.8757062146892656,
      "grad_norm": 0.43045905232429504,
      "learning_rate": 1.4382945210793663e-05,
      "loss": 0.7991,
      "step": 10540
    },
    {
      "epoch": 0.8765370555001661,
      "grad_norm": 0.42041224241256714,
      "learning_rate": 1.4104508614417267e-05,
      "loss": 0.7699,
      "step": 10550
    },
    {
      "epoch": 0.8773678963110668,
      "grad_norm": 0.43115219473838806,
      "learning_rate": 1.382221094560512e-05,
      "loss": 0.7949,
      "step": 10560
    },
    {
      "epoch": 0.8781987371219674,
      "grad_norm": 0.46832332015037537,
      "learning_rate": 1.3536321033912506e-05,
      "loss": 0.8004,
      "step": 10570
    },
    {
      "epoch": 0.8790295779328681,
      "grad_norm": 0.417580246925354,
      "learning_rate": 1.3247111129755609e-05,
      "loss": 0.7814,
      "step": 10580
    },
    {
      "epoch": 0.8798604187437687,
      "grad_norm": 0.5604673624038696,
      "learning_rate": 1.2954856645149868e-05,
      "loss": 0.7662,
      "step": 10590
    },
    {
      "epoch": 0.8806912595546693,
      "grad_norm": 0.4227937161922455,
      "learning_rate": 1.2659835891437553e-05,
      "loss": 0.7545,
      "step": 10600
    },
    {
      "epoch": 0.88152210036557,
      "grad_norm": 0.41911235451698303,
      "learning_rate": 1.236232981425436e-05,
      "loss": 0.755,
      "step": 10610
    },
    {
      "epoch": 0.8823529411764706,
      "grad_norm": 0.41674908995628357,
      "learning_rate": 1.2062621725987402e-05,
      "loss": 0.7555,
      "step": 10620
    },
    {
      "epoch": 0.8831837819873712,
      "grad_norm": 0.4613277018070221,
      "learning_rate": 1.1760997035979318e-05,
      "loss": 0.7836,
      "step": 10630
    },
    {
      "epoch": 0.8840146227982718,
      "grad_norm": 0.4316408336162567,
      "learning_rate": 1.1457742978735514e-05,
      "loss": 0.7783,
      "step": 10640
    },
    {
      "epoch": 0.8848454636091725,
      "grad_norm": 0.42209306359291077,
      "learning_rate": 1.1153148340393298e-05,
      "loss": 0.7856,
      "step": 10650
    },
    {
      "epoch": 0.8856763044200732,
      "grad_norm": 0.43788138031959534,
      "learning_rate": 1.0847503183713424e-05,
      "loss": 0.747,
      "step": 10660
    },
    {
      "epoch": 0.8865071452309737,
      "grad_norm": 0.44516319036483765,
      "learning_rate": 1.0541098571855896e-05,
      "loss": 0.7407,
      "step": 10670
    },
    {
      "epoch": 0.8873379860418744,
      "grad_norm": 0.421673059463501,
      "learning_rate": 1.0234226291203123e-05,
      "loss": 0.7605,
      "step": 10680
    },
    {
      "epoch": 0.888168826852775,
      "grad_norm": 0.4195078909397125,
      "learning_rate": 9.927178573494333e-06,
      "loss": 0.7629,
      "step": 10690
    },
    {
      "epoch": 0.8889996676636757,
      "grad_norm": 0.5545461773872375,
      "learning_rate": 9.620247817535902e-06,
      "loss": 0.7433,
      "step": 10700
    },
    {
      "epoch": 0.8898305084745762,
      "grad_norm": 0.43663299083709717,
      "learning_rate": 9.313726310752581e-06,
      "loss": 0.7268,
      "step": 10710
    },
    {
      "epoch": 0.8906613492854769,
      "grad_norm": 0.42937150597572327,
      "learning_rate": 9.00790595084477e-06,
      "loss": 0.7855,
      "step": 10720
    },
    {
      "epoch": 0.8914921900963775,
      "grad_norm": 0.39997634291648865,
      "learning_rate": 8.703077967816944e-06,
      "loss": 0.7678,
      "step": 10730
    },
    {
      "epoch": 0.8923230309072782,
      "grad_norm": 0.40996453166007996,
      "learning_rate": 8.39953264664192e-06,
      "loss": 0.7377,
      "step": 10740
    },
    {
      "epoch": 0.8931538717181788,
      "grad_norm": 0.4170096218585968,
      "learning_rate": 8.097559050825058e-06,
      "loss": 0.7629,
      "step": 10750
    },
    {
      "epoch": 0.8939847125290794,
      "grad_norm": 0.4183298945426941,
      "learning_rate": 7.79744474713166e-06,
      "loss": 0.7537,
      "step": 10760
    },
    {
      "epoch": 0.8948155533399801,
      "grad_norm": 0.4273609519004822,
      "learning_rate": 7.499475531739696e-06,
      "loss": 0.7844,
      "step": 10770
    },
    {
      "epoch": 0.8956463941508807,
      "grad_norm": 0.4475758373737335,
      "learning_rate": 7.203935158078662e-06,
      "loss": 0.7626,
      "step": 10780
    },
    {
      "epoch": 0.8964772349617813,
      "grad_norm": 0.42410147190093994,
      "learning_rate": 6.911105066613709e-06,
      "loss": 0.7199,
      "step": 10790
    },
    {
      "epoch": 0.8973080757726819,
      "grad_norm": 0.40266868472099304,
      "learning_rate": 6.621264116832394e-06,
      "loss": 0.7719,
      "step": 10800
    },
    {
      "epoch": 0.8981389165835826,
      "grad_norm": 0.47475606203079224,
      "learning_rate": 6.334688321689274e-06,
      "loss": 0.7562,
      "step": 10810
    },
    {
      "epoch": 0.8989697573944833,
      "grad_norm": 0.48869559168815613,
      "learning_rate": 6.051650584761237e-06,
      "loss": 0.7644,
      "step": 10820
    },
    {
      "epoch": 0.8998005982053838,
      "grad_norm": 0.4059005081653595,
      "learning_rate": 5.772420440363841e-06,
      "loss": 0.7649,
      "step": 10830
    },
    {
      "epoch": 0.9006314390162845,
      "grad_norm": 0.3998551368713379,
      "learning_rate": 5.497263796876207e-06,
      "loss": 0.7641,
      "step": 10840
    },
    {
      "epoch": 0.9014622798271851,
      "grad_norm": 0.4275914132595062,
      "learning_rate": 5.226442683518813e-06,
      "loss": 0.7708,
      "step": 10850
    },
    {
      "epoch": 0.9022931206380858,
      "grad_norm": 0.4379135072231293,
      "learning_rate": 4.960215000825432e-06,
      "loss": 0.7539,
      "step": 10860
    },
    {
      "epoch": 0.9031239614489863,
      "grad_norm": 0.416816771030426,
      "learning_rate": 4.698834275046711e-06,
      "loss": 0.7238,
      "step": 10870
    },
    {
      "epoch": 0.903954802259887,
      "grad_norm": 0.41467005014419556,
      "learning_rate": 4.442549416719423e-06,
      "loss": 0.7625,
      "step": 10880
    },
    {
      "epoch": 0.9047856430707877,
      "grad_norm": 0.43065670132637024,
      "learning_rate": 4.1916044836311565e-06,
      "loss": 0.7726,
      "step": 10890
    },
    {
      "epoch": 0.9056164838816883,
      "grad_norm": 0.41847673058509827,
      "learning_rate": 3.94623844840626e-06,
      "loss": 0.7564,
      "step": 10900
    },
    {
      "epoch": 0.9064473246925889,
      "grad_norm": 0.4037667512893677,
      "learning_rate": 3.7066849709343467e-06,
      "loss": 0.7785,
      "step": 10910
    },
    {
      "epoch": 0.9072781655034895,
      "grad_norm": 0.4458758234977722,
      "learning_rate": 3.473172175858059e-06,
      "loss": 0.7445,
      "step": 10920
    },
    {
      "epoch": 0.9081090063143902,
      "grad_norm": 0.42063114047050476,
      "learning_rate": 3.2459224353320046e-06,
      "loss": 0.7448,
      "step": 10930
    },
    {
      "epoch": 0.9089398471252907,
      "grad_norm": 0.42065006494522095,
      "learning_rate": 3.025152157259735e-06,
      "loss": 0.764,
      "step": 10940
    },
    {
      "epoch": 0.9097706879361914,
      "grad_norm": 0.3943175673484802,
      "learning_rate": 2.811071579210425e-06,
      "loss": 0.7604,
      "step": 10950
    },
    {
      "epoch": 0.9106015287470921,
      "grad_norm": 0.40539565682411194,
      "learning_rate": 2.603884568211513e-06,
      "loss": 0.7337,
      "step": 10960
    },
    {
      "epoch": 0.9114323695579927,
      "grad_norm": 0.4015915095806122,
      "learning_rate": 2.4037884266079434e-06,
      "loss": 0.7338,
      "step": 10970
    },
    {
      "epoch": 0.9122632103688934,
      "grad_norm": 0.45153671503067017,
      "learning_rate": 2.2109737041728955e-06,
      "loss": 0.717,
      "step": 10980
    },
    {
      "epoch": 0.9130940511797939,
      "grad_norm": 0.40950143337249756,
      "learning_rate": 2.0256240166489454e-06,
      "loss": 0.7476,
      "step": 10990
    },
    {
      "epoch": 0.9139248919906946,
      "grad_norm": 0.415046751499176,
      "learning_rate": 1.847915870892413e-06,
      "loss": 0.756,
      "step": 11000
    },
    {
      "epoch": 0.9147557328015952,
      "grad_norm": 0.441194087266922,
      "learning_rate": 1.6780184967874715e-06,
      "loss": 0.7489,
      "step": 11010
    },
    {
      "epoch": 0.9155865736124958,
      "grad_norm": 0.41226470470428467,
      "learning_rate": 1.5160936860900294e-06,
      "loss": 0.7562,
      "step": 11020
    },
    {
      "epoch": 0.9164174144233965,
      "grad_norm": 0.4707246720790863,
      "learning_rate": 1.3622956383548762e-06,
      "loss": 0.7779,
      "step": 11030
    },
    {
      "epoch": 0.9172482552342971,
      "grad_norm": 0.40057510137557983,
      "learning_rate": 1.216770814092834e-06,
      "loss": 0.7241,
      "step": 11040
    },
    {
      "epoch": 0.9180790960451978,
      "grad_norm": 0.43560075759887695,
      "learning_rate": 1.0796577952977013e-06,
      "loss": 0.697,
      "step": 11050
    },
    {
      "epoch": 0.9189099368560983,
      "grad_norm": 0.4083419740200043,
      "learning_rate": 9.510871534758705e-07,
      "loss": 0.7551,
      "step": 11060
    },
    {
      "epoch": 0.919740777666999,
      "grad_norm": 0.41108858585357666,
      "learning_rate": 8.311813253042334e-07,
      "loss": 0.7378,
      "step": 11070
    },
    {
      "epoch": 0.9205716184778996,
      "grad_norm": 0.43034932017326355,
      "learning_rate": 7.200544960348308e-07,
      "loss": 0.7305,
      "step": 11080
    },
    {
      "epoch": 0.9214024592888003,
      "grad_norm": 0.4227776825428009,
      "learning_rate": 6.178124907572545e-07,
      "loss": 0.7315,
      "step": 11090
    },
    {
      "epoch": 0.922233300099701,
      "grad_norm": 0.43823912739753723,
      "learning_rate": 5.245526736223528e-07,
      "loss": 0.7317,
      "step": 11100
    },
    {
      "epoch": 0.9230641409106015,
      "grad_norm": 0.4270930588245392,
      "learning_rate": 4.403638551232205e-07,
      "loss": 0.7297,
      "step": 11110
    },
    {
      "epoch": 0.9238949817215022,
      "grad_norm": 0.41591307520866394,
      "learning_rate": 3.6532620752175616e-07,
      "loss": 0.7698,
      "step": 11120
    },
    {
      "epoch": 0.9247258225324028,
      "grad_norm": 0.3919279873371124,
      "learning_rate": 2.9951118850133247e-07,
      "loss": 0.7752,
      "step": 11130
    },
    {
      "epoch": 0.9255566633433034,
      "grad_norm": 1.2267820835113525,
      "learning_rate": 2.429814731182774e-07,
      "loss": 0.7476,
      "step": 11140
    },
    {
      "epoch": 0.926387504154204,
      "grad_norm": 0.3922807574272156,
      "learning_rate": 1.957908941169813e-07,
      "loss": 0.7455,
      "step": 11150
    },
    {
      "epoch": 0.9272183449651047,
      "grad_norm": 0.4089207649230957,
      "learning_rate": 1.5798439066545254e-07,
      "loss": 0.783,
      "step": 11160
    },
    {
      "epoch": 0.9280491857760054,
      "grad_norm": 0.43610185384750366,
      "learning_rate": 1.2959796556014471e-07,
      "loss": 0.7335,
      "step": 11170
    },
    {
      "epoch": 0.9288800265869059,
      "grad_norm": 0.40243199467658997,
      "learning_rate": 1.1065865094081884e-07,
      "loss": 0.7654,
      "step": 11180
    },
    {
      "epoch": 0.9297108673978066,
      "grad_norm": 0.41354572772979736,
      "learning_rate": 1.011844825480734e-07,
      "loss": 0.761,
      "step": 11190
    },
    {
      "epoch": 0.9300432037221669,
      "step": 11194,
      "total_flos": 4.547514943786923e+19,
      "train_loss": 0.0042809755307596795,
      "train_runtime": 607.2641,
      "train_samples_per_second": 221.202,
      "train_steps_per_second": 18.433
    }
  ],
  "logging_steps": 10,
  "max_steps": 11194,
  "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": 4.547514943786923e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}