{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.46502160186108343,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008308408109006315,
      "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": 0.4199659526348114,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.8744,
      "step": 5090
    },
    {
      "epoch": 0.423728813559322,
      "grad_norm": 0.4632173776626587,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.8516,
      "step": 5100
    },
    {
      "epoch": 0.42455965437022264,
      "grad_norm": 0.6056671738624573,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.8085,
      "step": 5110
    },
    {
      "epoch": 0.4253904951811233,
      "grad_norm": 0.5153040289878845,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.8267,
      "step": 5120
    },
    {
      "epoch": 0.42622133599202394,
      "grad_norm": 0.4063456654548645,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.8217,
      "step": 5130
    },
    {
      "epoch": 0.42705217680292457,
      "grad_norm": 0.4757425785064697,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.864,
      "step": 5140
    },
    {
      "epoch": 0.4278830176138252,
      "grad_norm": 0.4486253261566162,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.8373,
      "step": 5150
    },
    {
      "epoch": 0.4287138584247258,
      "grad_norm": 0.4408472180366516,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.8541,
      "step": 5160
    },
    {
      "epoch": 0.42954469923562644,
      "grad_norm": 0.442969411611557,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.8385,
      "step": 5170
    },
    {
      "epoch": 0.43037554004652706,
      "grad_norm": 0.45393094420433044,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.836,
      "step": 5180
    },
    {
      "epoch": 0.43120638085742774,
      "grad_norm": 0.4256947338581085,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.8451,
      "step": 5190
    },
    {
      "epoch": 0.43203722166832836,
      "grad_norm": 0.42153313755989075,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.847,
      "step": 5200
    },
    {
      "epoch": 0.432868062479229,
      "grad_norm": 0.4386671185493469,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.8351,
      "step": 5210
    },
    {
      "epoch": 0.4336989032901296,
      "grad_norm": 0.41105470061302185,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.846,
      "step": 5220
    },
    {
      "epoch": 0.43452974410103024,
      "grad_norm": 0.420073002576828,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.8524,
      "step": 5230
    },
    {
      "epoch": 0.43536058491193086,
      "grad_norm": 0.7505195140838623,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.8382,
      "step": 5240
    },
    {
      "epoch": 0.4361914257228315,
      "grad_norm": 0.4421382546424866,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.8453,
      "step": 5250
    },
    {
      "epoch": 0.43702226653373216,
      "grad_norm": 0.4234689772129059,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.8471,
      "step": 5260
    },
    {
      "epoch": 0.4378531073446328,
      "grad_norm": 1.849616289138794,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.8538,
      "step": 5270
    },
    {
      "epoch": 0.4386839481555334,
      "grad_norm": 0.42529231309890747,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.8543,
      "step": 5280
    },
    {
      "epoch": 0.43951478896643403,
      "grad_norm": 0.45581308007240295,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.8657,
      "step": 5290
    },
    {
      "epoch": 0.44034562977733466,
      "grad_norm": 0.463362455368042,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.8477,
      "step": 5300
    },
    {
      "epoch": 0.4411764705882353,
      "grad_norm": 0.4080484211444855,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.8389,
      "step": 5310
    },
    {
      "epoch": 0.4420073113991359,
      "grad_norm": 0.39870691299438477,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.855,
      "step": 5320
    },
    {
      "epoch": 0.4428381522100366,
      "grad_norm": 0.4225146770477295,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.8669,
      "step": 5330
    },
    {
      "epoch": 0.4436689930209372,
      "grad_norm": 0.7861130237579346,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.8226,
      "step": 5340
    },
    {
      "epoch": 0.44449983383183783,
      "grad_norm": 0.3933430016040802,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.8465,
      "step": 5350
    },
    {
      "epoch": 0.44533067464273846,
      "grad_norm": 0.4345282018184662,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.8442,
      "step": 5360
    },
    {
      "epoch": 0.4461615154536391,
      "grad_norm": 0.4222422242164612,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.8413,
      "step": 5370
    },
    {
      "epoch": 0.4469923562645397,
      "grad_norm": 0.41591107845306396,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.8118,
      "step": 5380
    },
    {
      "epoch": 0.4478231970754403,
      "grad_norm": 0.42406073212623596,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.8384,
      "step": 5390
    },
    {
      "epoch": 0.44865403788634095,
      "grad_norm": 0.4009072184562683,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.8268,
      "step": 5400
    },
    {
      "epoch": 0.44948487869724163,
      "grad_norm": 0.5106752514839172,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.7985,
      "step": 5410
    },
    {
      "epoch": 0.45031571950814225,
      "grad_norm": 0.4096420109272003,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.8794,
      "step": 5420
    },
    {
      "epoch": 0.4511465603190429,
      "grad_norm": 0.4227890074253082,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.8016,
      "step": 5430
    },
    {
      "epoch": 0.4519774011299435,
      "grad_norm": 0.42503228783607483,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.8449,
      "step": 5440
    },
    {
      "epoch": 0.4528082419408441,
      "grad_norm": 0.46982717514038086,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.7996,
      "step": 5450
    },
    {
      "epoch": 0.45363908275174475,
      "grad_norm": 0.3979046940803528,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.8414,
      "step": 5460
    },
    {
      "epoch": 0.4544699235626454,
      "grad_norm": 0.4202454090118408,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.8285,
      "step": 5470
    },
    {
      "epoch": 0.45530076437354605,
      "grad_norm": 0.45515576004981995,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.8416,
      "step": 5480
    },
    {
      "epoch": 0.4561316051844467,
      "grad_norm": 0.431459903717041,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.8307,
      "step": 5490
    },
    {
      "epoch": 0.4569624459953473,
      "grad_norm": 0.43237003684043884,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.8361,
      "step": 5500
    },
    {
      "epoch": 0.4577932868062479,
      "grad_norm": 0.4045921266078949,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.8301,
      "step": 5510
    },
    {
      "epoch": 0.45862412761714855,
      "grad_norm": 0.4192327558994293,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.8396,
      "step": 5520
    },
    {
      "epoch": 0.45945496842804917,
      "grad_norm": 0.41404488682746887,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.7986,
      "step": 5530
    },
    {
      "epoch": 0.4602858092389498,
      "grad_norm": 0.40141066908836365,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.8813,
      "step": 5540
    },
    {
      "epoch": 0.4611166500498505,
      "grad_norm": 0.39575648307800293,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.8071,
      "step": 5550
    },
    {
      "epoch": 0.4619474908607511,
      "grad_norm": 0.4322142004966736,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.8416,
      "step": 5560
    },
    {
      "epoch": 0.4627783316716517,
      "grad_norm": 0.4215182363986969,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.8258,
      "step": 5570
    },
    {
      "epoch": 0.46360917248255235,
      "grad_norm": 0.40274325013160706,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.8321,
      "step": 5580
    },
    {
      "epoch": 0.46444001329345297,
      "grad_norm": 0.4037381112575531,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.8257,
      "step": 5590
    },
    {
      "epoch": 0.46502160186108343,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.028428952822157884,
      "train_runtime": 1768.2408,
      "train_samples_per_second": 37.984,
      "train_steps_per_second": 3.165
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}