{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9371285056509,
  "eval_steps": 500,
  "global_step": 11194,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008371703641691084,
      "grad_norm": 6.286955833435059,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 2.1571,
      "step": 10
    },
    {
      "epoch": 0.0016743407283382169,
      "grad_norm": 3.6767687797546387,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.9243,
      "step": 20
    },
    {
      "epoch": 0.0025115110925073253,
      "grad_norm": 2.151350498199463,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.7728,
      "step": 30
    },
    {
      "epoch": 0.0033486814566764337,
      "grad_norm": 1.2491344213485718,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.4703,
      "step": 40
    },
    {
      "epoch": 0.004185851820845542,
      "grad_norm": 0.9011292457580566,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.2532,
      "step": 50
    },
    {
      "epoch": 0.0050230221850146506,
      "grad_norm": 1.1828844547271729,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.1652,
      "step": 60
    },
    {
      "epoch": 0.0058601925491837585,
      "grad_norm": 0.7618897557258606,
      "learning_rate": 7.814044943820226e-06,
      "loss": 1.1584,
      "step": 70
    },
    {
      "epoch": 0.006697362913352867,
      "grad_norm": 0.6916522979736328,
      "learning_rate": 8.932022471910112e-06,
      "loss": 1.0909,
      "step": 80
    },
    {
      "epoch": 0.007534533277521975,
      "grad_norm": 0.6934717893600464,
      "learning_rate": 1.005e-05,
      "loss": 1.0744,
      "step": 90
    },
    {
      "epoch": 0.008371703641691083,
      "grad_norm": 0.7366052865982056,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 1.0488,
      "step": 100
    },
    {
      "epoch": 0.009208874005860193,
      "grad_norm": 0.6335578560829163,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 1.0163,
      "step": 110
    },
    {
      "epoch": 0.010046044370029301,
      "grad_norm": 0.6413920521736145,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 1.0387,
      "step": 120
    },
    {
      "epoch": 0.010883214734198409,
      "grad_norm": 0.652919352054596,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 1.0123,
      "step": 130
    },
    {
      "epoch": 0.011720385098367517,
      "grad_norm": 0.6183255314826965,
      "learning_rate": 1.563988764044944e-05,
      "loss": 1.0209,
      "step": 140
    },
    {
      "epoch": 0.012557555462536627,
      "grad_norm": 0.5744184851646423,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 1.0548,
      "step": 150
    },
    {
      "epoch": 0.013394725826705735,
      "grad_norm": 0.6852168440818787,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 1.0805,
      "step": 160
    },
    {
      "epoch": 0.014231896190874843,
      "grad_norm": 0.6165246963500977,
      "learning_rate": 1.89938202247191e-05,
      "loss": 1.0197,
      "step": 170
    },
    {
      "epoch": 0.01506906655504395,
      "grad_norm": 0.6286055445671082,
      "learning_rate": 2e-05,
      "loss": 0.9775,
      "step": 180
    },
    {
      "epoch": 0.01590623691921306,
      "grad_norm": 0.5891881585121155,
      "learning_rate": 2e-05,
      "loss": 1.0042,
      "step": 190
    },
    {
      "epoch": 0.016743407283382167,
      "grad_norm": 0.6101363897323608,
      "learning_rate": 2e-05,
      "loss": 1.0048,
      "step": 200
    },
    {
      "epoch": 0.017580577647551276,
      "grad_norm": 0.5848655700683594,
      "learning_rate": 2e-05,
      "loss": 0.9719,
      "step": 210
    },
    {
      "epoch": 0.018417748011720386,
      "grad_norm": 0.5625296235084534,
      "learning_rate": 2e-05,
      "loss": 0.9727,
      "step": 220
    },
    {
      "epoch": 0.019254918375889492,
      "grad_norm": 0.5747894048690796,
      "learning_rate": 2e-05,
      "loss": 0.9626,
      "step": 230
    },
    {
      "epoch": 0.020092088740058602,
      "grad_norm": 0.5566433072090149,
      "learning_rate": 2e-05,
      "loss": 0.9992,
      "step": 240
    },
    {
      "epoch": 0.020929259104227712,
      "grad_norm": 0.6630275249481201,
      "learning_rate": 2e-05,
      "loss": 0.978,
      "step": 250
    },
    {
      "epoch": 0.021766429468396818,
      "grad_norm": 0.5642217993736267,
      "learning_rate": 2e-05,
      "loss": 0.9639,
      "step": 260
    },
    {
      "epoch": 0.022603599832565928,
      "grad_norm": 0.597459614276886,
      "learning_rate": 2e-05,
      "loss": 0.9712,
      "step": 270
    },
    {
      "epoch": 0.023440770196735034,
      "grad_norm": 0.6050856709480286,
      "learning_rate": 2e-05,
      "loss": 0.9895,
      "step": 280
    },
    {
      "epoch": 0.024277940560904144,
      "grad_norm": 0.5888883471488953,
      "learning_rate": 2e-05,
      "loss": 0.9602,
      "step": 290
    },
    {
      "epoch": 0.025115110925073254,
      "grad_norm": 0.5481868982315063,
      "learning_rate": 2e-05,
      "loss": 0.9562,
      "step": 300
    },
    {
      "epoch": 0.02595228128924236,
      "grad_norm": 0.543234646320343,
      "learning_rate": 2e-05,
      "loss": 0.9804,
      "step": 310
    },
    {
      "epoch": 0.02678945165341147,
      "grad_norm": 0.5965615510940552,
      "learning_rate": 2e-05,
      "loss": 0.962,
      "step": 320
    },
    {
      "epoch": 0.02762662201758058,
      "grad_norm": 0.5193302631378174,
      "learning_rate": 2e-05,
      "loss": 0.9852,
      "step": 330
    },
    {
      "epoch": 0.028463792381749686,
      "grad_norm": 0.524551510810852,
      "learning_rate": 2e-05,
      "loss": 0.9865,
      "step": 340
    },
    {
      "epoch": 0.029300962745918795,
      "grad_norm": 0.5498046875,
      "learning_rate": 2e-05,
      "loss": 0.9505,
      "step": 350
    },
    {
      "epoch": 0.0301381331100879,
      "grad_norm": 0.5638454556465149,
      "learning_rate": 2e-05,
      "loss": 0.9537,
      "step": 360
    },
    {
      "epoch": 0.03097530347425701,
      "grad_norm": 0.5428140759468079,
      "learning_rate": 2e-05,
      "loss": 0.9562,
      "step": 370
    },
    {
      "epoch": 0.03181247383842612,
      "grad_norm": 0.5936139225959778,
      "learning_rate": 2e-05,
      "loss": 0.9566,
      "step": 380
    },
    {
      "epoch": 0.03264964420259523,
      "grad_norm": 0.5149725079536438,
      "learning_rate": 2e-05,
      "loss": 0.9279,
      "step": 390
    },
    {
      "epoch": 0.033486814566764334,
      "grad_norm": 0.5177335143089294,
      "learning_rate": 2e-05,
      "loss": 0.9404,
      "step": 400
    },
    {
      "epoch": 0.03432398493093344,
      "grad_norm": 0.5363324284553528,
      "learning_rate": 2e-05,
      "loss": 0.9536,
      "step": 410
    },
    {
      "epoch": 0.03516115529510255,
      "grad_norm": 0.5618890523910522,
      "learning_rate": 2e-05,
      "loss": 0.9827,
      "step": 420
    },
    {
      "epoch": 0.03599832565927166,
      "grad_norm": 0.5098859667778015,
      "learning_rate": 2e-05,
      "loss": 0.9344,
      "step": 430
    },
    {
      "epoch": 0.03683549602344077,
      "grad_norm": 0.5013912320137024,
      "learning_rate": 2e-05,
      "loss": 0.9398,
      "step": 440
    },
    {
      "epoch": 0.037672666387609875,
      "grad_norm": 0.4910331666469574,
      "learning_rate": 2e-05,
      "loss": 0.9394,
      "step": 450
    },
    {
      "epoch": 0.038509836751778985,
      "grad_norm": 0.5443107485771179,
      "learning_rate": 2e-05,
      "loss": 0.9467,
      "step": 460
    },
    {
      "epoch": 0.039347007115948095,
      "grad_norm": 0.48179903626441956,
      "learning_rate": 2e-05,
      "loss": 0.9304,
      "step": 470
    },
    {
      "epoch": 0.040184177480117204,
      "grad_norm": 0.5266996026039124,
      "learning_rate": 2e-05,
      "loss": 0.9093,
      "step": 480
    },
    {
      "epoch": 0.041021347844286314,
      "grad_norm": 0.5182468295097351,
      "learning_rate": 2e-05,
      "loss": 0.933,
      "step": 490
    },
    {
      "epoch": 0.041858518208455424,
      "grad_norm": 0.5562992691993713,
      "learning_rate": 2e-05,
      "loss": 0.8997,
      "step": 500
    },
    {
      "epoch": 0.04269568857262453,
      "grad_norm": 0.5871270895004272,
      "learning_rate": 2e-05,
      "loss": 0.9345,
      "step": 510
    },
    {
      "epoch": 0.043532858936793636,
      "grad_norm": 0.5355167984962463,
      "learning_rate": 2e-05,
      "loss": 0.9235,
      "step": 520
    },
    {
      "epoch": 0.044370029300962746,
      "grad_norm": 0.508417546749115,
      "learning_rate": 2e-05,
      "loss": 0.9142,
      "step": 530
    },
    {
      "epoch": 0.045207199665131856,
      "grad_norm": 0.5056772828102112,
      "learning_rate": 2e-05,
      "loss": 0.9211,
      "step": 540
    },
    {
      "epoch": 0.046044370029300966,
      "grad_norm": 0.5066829919815063,
      "learning_rate": 2e-05,
      "loss": 0.9124,
      "step": 550
    },
    {
      "epoch": 0.04688154039347007,
      "grad_norm": 0.5150319933891296,
      "learning_rate": 2e-05,
      "loss": 0.8999,
      "step": 560
    },
    {
      "epoch": 0.04771871075763918,
      "grad_norm": 0.5607603192329407,
      "learning_rate": 2e-05,
      "loss": 0.9378,
      "step": 570
    },
    {
      "epoch": 0.04855588112180829,
      "grad_norm": 0.5025616884231567,
      "learning_rate": 2e-05,
      "loss": 0.913,
      "step": 580
    },
    {
      "epoch": 0.0493930514859774,
      "grad_norm": 0.4901517629623413,
      "learning_rate": 2e-05,
      "loss": 0.936,
      "step": 590
    },
    {
      "epoch": 0.05023022185014651,
      "grad_norm": 0.5197919011116028,
      "learning_rate": 2e-05,
      "loss": 0.9501,
      "step": 600
    },
    {
      "epoch": 0.05106739221431561,
      "grad_norm": 0.5028531551361084,
      "learning_rate": 2e-05,
      "loss": 0.9094,
      "step": 610
    },
    {
      "epoch": 0.05190456257848472,
      "grad_norm": 0.48607340455055237,
      "learning_rate": 2e-05,
      "loss": 0.9494,
      "step": 620
    },
    {
      "epoch": 0.05274173294265383,
      "grad_norm": 0.5188357830047607,
      "learning_rate": 2e-05,
      "loss": 0.9326,
      "step": 630
    },
    {
      "epoch": 0.05357890330682294,
      "grad_norm": 0.5137494802474976,
      "learning_rate": 2e-05,
      "loss": 0.9286,
      "step": 640
    },
    {
      "epoch": 0.05441607367099205,
      "grad_norm": 0.49051663279533386,
      "learning_rate": 2e-05,
      "loss": 0.9201,
      "step": 650
    },
    {
      "epoch": 0.05525324403516116,
      "grad_norm": 0.5025955438613892,
      "learning_rate": 2e-05,
      "loss": 0.9509,
      "step": 660
    },
    {
      "epoch": 0.05609041439933026,
      "grad_norm": 0.4561798870563507,
      "learning_rate": 2e-05,
      "loss": 0.902,
      "step": 670
    },
    {
      "epoch": 0.05692758476349937,
      "grad_norm": 0.5147539973258972,
      "learning_rate": 2e-05,
      "loss": 0.8891,
      "step": 680
    },
    {
      "epoch": 0.05776475512766848,
      "grad_norm": 0.512950599193573,
      "learning_rate": 2e-05,
      "loss": 0.9146,
      "step": 690
    },
    {
      "epoch": 0.05860192549183759,
      "grad_norm": 0.5343301892280579,
      "learning_rate": 2e-05,
      "loss": 0.9434,
      "step": 700
    },
    {
      "epoch": 0.0594390958560067,
      "grad_norm": 0.5037513375282288,
      "learning_rate": 2e-05,
      "loss": 0.9396,
      "step": 710
    },
    {
      "epoch": 0.0602762662201758,
      "grad_norm": 0.5113139748573303,
      "learning_rate": 2e-05,
      "loss": 0.9311,
      "step": 720
    },
    {
      "epoch": 0.06111343658434491,
      "grad_norm": 0.49999576807022095,
      "learning_rate": 2e-05,
      "loss": 0.8928,
      "step": 730
    },
    {
      "epoch": 0.06195060694851402,
      "grad_norm": 0.5156519412994385,
      "learning_rate": 2e-05,
      "loss": 0.9134,
      "step": 740
    },
    {
      "epoch": 0.06278777731268313,
      "grad_norm": 0.5189371705055237,
      "learning_rate": 2e-05,
      "loss": 0.9147,
      "step": 750
    },
    {
      "epoch": 0.06362494767685224,
      "grad_norm": 0.5053024291992188,
      "learning_rate": 2e-05,
      "loss": 0.8988,
      "step": 760
    },
    {
      "epoch": 0.06446211804102135,
      "grad_norm": 0.49766141176223755,
      "learning_rate": 2e-05,
      "loss": 0.941,
      "step": 770
    },
    {
      "epoch": 0.06529928840519046,
      "grad_norm": 0.50629061460495,
      "learning_rate": 2e-05,
      "loss": 0.9235,
      "step": 780
    },
    {
      "epoch": 0.06613645876935957,
      "grad_norm": 0.4884129464626312,
      "learning_rate": 2e-05,
      "loss": 0.911,
      "step": 790
    },
    {
      "epoch": 0.06697362913352867,
      "grad_norm": 0.5056248307228088,
      "learning_rate": 2e-05,
      "loss": 0.9273,
      "step": 800
    },
    {
      "epoch": 0.06781079949769778,
      "grad_norm": 0.47344326972961426,
      "learning_rate": 2e-05,
      "loss": 0.9276,
      "step": 810
    },
    {
      "epoch": 0.06864796986186689,
      "grad_norm": 0.49335843324661255,
      "learning_rate": 2e-05,
      "loss": 0.9277,
      "step": 820
    },
    {
      "epoch": 0.069485140226036,
      "grad_norm": 0.5105158090591431,
      "learning_rate": 2e-05,
      "loss": 0.9272,
      "step": 830
    },
    {
      "epoch": 0.0703223105902051,
      "grad_norm": 0.5000097155570984,
      "learning_rate": 2e-05,
      "loss": 0.9213,
      "step": 840
    },
    {
      "epoch": 0.07115948095437422,
      "grad_norm": 0.4862711429595947,
      "learning_rate": 2e-05,
      "loss": 0.9243,
      "step": 850
    },
    {
      "epoch": 0.07199665131854333,
      "grad_norm": 0.4822566509246826,
      "learning_rate": 2e-05,
      "loss": 0.9318,
      "step": 860
    },
    {
      "epoch": 0.07283382168271244,
      "grad_norm": 0.4932926595211029,
      "learning_rate": 2e-05,
      "loss": 0.889,
      "step": 870
    },
    {
      "epoch": 0.07367099204688154,
      "grad_norm": 0.4769861400127411,
      "learning_rate": 2e-05,
      "loss": 0.9086,
      "step": 880
    },
    {
      "epoch": 0.07450816241105065,
      "grad_norm": 0.4864020049571991,
      "learning_rate": 2e-05,
      "loss": 0.9072,
      "step": 890
    },
    {
      "epoch": 0.07534533277521975,
      "grad_norm": 0.49218103289604187,
      "learning_rate": 2e-05,
      "loss": 0.9038,
      "step": 900
    },
    {
      "epoch": 0.07618250313938886,
      "grad_norm": 0.5060418844223022,
      "learning_rate": 2e-05,
      "loss": 0.9313,
      "step": 910
    },
    {
      "epoch": 0.07701967350355797,
      "grad_norm": 0.4857514500617981,
      "learning_rate": 2e-05,
      "loss": 0.8917,
      "step": 920
    },
    {
      "epoch": 0.07785684386772708,
      "grad_norm": 0.5140605568885803,
      "learning_rate": 2e-05,
      "loss": 0.9006,
      "step": 930
    },
    {
      "epoch": 0.07869401423189619,
      "grad_norm": 0.4838702380657196,
      "learning_rate": 2e-05,
      "loss": 0.9029,
      "step": 940
    },
    {
      "epoch": 0.0795311845960653,
      "grad_norm": 0.49060988426208496,
      "learning_rate": 2e-05,
      "loss": 0.9136,
      "step": 950
    },
    {
      "epoch": 0.08036835496023441,
      "grad_norm": 0.4825526773929596,
      "learning_rate": 2e-05,
      "loss": 0.9076,
      "step": 960
    },
    {
      "epoch": 0.08120552532440352,
      "grad_norm": 0.4880685806274414,
      "learning_rate": 2e-05,
      "loss": 0.8999,
      "step": 970
    },
    {
      "epoch": 0.08204269568857263,
      "grad_norm": 0.46251583099365234,
      "learning_rate": 2e-05,
      "loss": 0.8884,
      "step": 980
    },
    {
      "epoch": 0.08287986605274174,
      "grad_norm": 0.5066401958465576,
      "learning_rate": 2e-05,
      "loss": 0.916,
      "step": 990
    },
    {
      "epoch": 0.08371703641691085,
      "grad_norm": 0.47283968329429626,
      "learning_rate": 2e-05,
      "loss": 0.8951,
      "step": 1000
    },
    {
      "epoch": 0.08455420678107994,
      "grad_norm": 0.5120212435722351,
      "learning_rate": 2e-05,
      "loss": 0.9262,
      "step": 1010
    },
    {
      "epoch": 0.08539137714524905,
      "grad_norm": 0.4767346978187561,
      "learning_rate": 2e-05,
      "loss": 0.9123,
      "step": 1020
    },
    {
      "epoch": 0.08622854750941816,
      "grad_norm": 0.5002844929695129,
      "learning_rate": 2e-05,
      "loss": 0.9067,
      "step": 1030
    },
    {
      "epoch": 0.08706571787358727,
      "grad_norm": 0.4847468435764313,
      "learning_rate": 2e-05,
      "loss": 0.8907,
      "step": 1040
    },
    {
      "epoch": 0.08790288823775638,
      "grad_norm": 0.4965667426586151,
      "learning_rate": 2e-05,
      "loss": 0.8953,
      "step": 1050
    },
    {
      "epoch": 0.08874005860192549,
      "grad_norm": 0.46984976530075073,
      "learning_rate": 2e-05,
      "loss": 0.8958,
      "step": 1060
    },
    {
      "epoch": 0.0895772289660946,
      "grad_norm": 0.44189727306365967,
      "learning_rate": 2e-05,
      "loss": 0.8899,
      "step": 1070
    },
    {
      "epoch": 0.09041439933026371,
      "grad_norm": 0.4697335660457611,
      "learning_rate": 2e-05,
      "loss": 0.8977,
      "step": 1080
    },
    {
      "epoch": 0.09125156969443282,
      "grad_norm": 0.4707675576210022,
      "learning_rate": 2e-05,
      "loss": 0.9103,
      "step": 1090
    },
    {
      "epoch": 0.09208874005860193,
      "grad_norm": 0.4781976640224457,
      "learning_rate": 2e-05,
      "loss": 0.8997,
      "step": 1100
    },
    {
      "epoch": 0.09292591042277103,
      "grad_norm": 0.45466241240501404,
      "learning_rate": 2e-05,
      "loss": 0.8989,
      "step": 1110
    },
    {
      "epoch": 0.09376308078694014,
      "grad_norm": 0.45188242197036743,
      "learning_rate": 2e-05,
      "loss": 0.9157,
      "step": 1120
    },
    {
      "epoch": 0.09460025115110925,
      "grad_norm": 0.48571550846099854,
      "learning_rate": 2e-05,
      "loss": 0.8939,
      "step": 1130
    },
    {
      "epoch": 0.09543742151527836,
      "grad_norm": 0.5890725255012512,
      "learning_rate": 2e-05,
      "loss": 0.9035,
      "step": 1140
    },
    {
      "epoch": 0.09627459187944747,
      "grad_norm": 0.4557766914367676,
      "learning_rate": 2e-05,
      "loss": 0.916,
      "step": 1150
    },
    {
      "epoch": 0.09711176224361658,
      "grad_norm": 0.4784591495990753,
      "learning_rate": 2e-05,
      "loss": 0.8826,
      "step": 1160
    },
    {
      "epoch": 0.09794893260778569,
      "grad_norm": 0.501518189907074,
      "learning_rate": 2e-05,
      "loss": 0.8908,
      "step": 1170
    },
    {
      "epoch": 0.0987861029719548,
      "grad_norm": 0.4712284803390503,
      "learning_rate": 2e-05,
      "loss": 0.8822,
      "step": 1180
    },
    {
      "epoch": 0.0996232733361239,
      "grad_norm": 0.48043906688690186,
      "learning_rate": 2e-05,
      "loss": 0.9054,
      "step": 1190
    },
    {
      "epoch": 0.10046044370029301,
      "grad_norm": 0.5000447034835815,
      "learning_rate": 2e-05,
      "loss": 0.8955,
      "step": 1200
    },
    {
      "epoch": 0.10129761406446212,
      "grad_norm": 0.4876522123813629,
      "learning_rate": 2e-05,
      "loss": 0.9194,
      "step": 1210
    },
    {
      "epoch": 0.10213478442863122,
      "grad_norm": 0.47112202644348145,
      "learning_rate": 2e-05,
      "loss": 0.9175,
      "step": 1220
    },
    {
      "epoch": 0.10297195479280033,
      "grad_norm": 0.4968307614326477,
      "learning_rate": 2e-05,
      "loss": 0.9004,
      "step": 1230
    },
    {
      "epoch": 0.10380912515696944,
      "grad_norm": 0.4805414080619812,
      "learning_rate": 2e-05,
      "loss": 0.8862,
      "step": 1240
    },
    {
      "epoch": 0.10464629552113855,
      "grad_norm": 0.4705885648727417,
      "learning_rate": 2e-05,
      "loss": 0.894,
      "step": 1250
    },
    {
      "epoch": 0.10548346588530766,
      "grad_norm": 0.5161804556846619,
      "learning_rate": 2e-05,
      "loss": 0.9077,
      "step": 1260
    },
    {
      "epoch": 0.10632063624947677,
      "grad_norm": 0.4699770212173462,
      "learning_rate": 2e-05,
      "loss": 0.8988,
      "step": 1270
    },
    {
      "epoch": 0.10715780661364588,
      "grad_norm": 0.45899173617362976,
      "learning_rate": 2e-05,
      "loss": 0.8799,
      "step": 1280
    },
    {
      "epoch": 0.10799497697781499,
      "grad_norm": 0.4808332622051239,
      "learning_rate": 2e-05,
      "loss": 0.9132,
      "step": 1290
    },
    {
      "epoch": 0.1088321473419841,
      "grad_norm": 0.46036213636398315,
      "learning_rate": 2e-05,
      "loss": 0.8565,
      "step": 1300
    },
    {
      "epoch": 0.10966931770615321,
      "grad_norm": 0.46063363552093506,
      "learning_rate": 2e-05,
      "loss": 0.9342,
      "step": 1310
    },
    {
      "epoch": 0.11050648807032232,
      "grad_norm": 0.4611421525478363,
      "learning_rate": 2e-05,
      "loss": 0.8941,
      "step": 1320
    },
    {
      "epoch": 0.11134365843449141,
      "grad_norm": 0.4850688576698303,
      "learning_rate": 2e-05,
      "loss": 0.8741,
      "step": 1330
    },
    {
      "epoch": 0.11218082879866052,
      "grad_norm": 0.4596725404262543,
      "learning_rate": 2e-05,
      "loss": 0.8929,
      "step": 1340
    },
    {
      "epoch": 0.11301799916282963,
      "grad_norm": 0.46708545088768005,
      "learning_rate": 2e-05,
      "loss": 0.8806,
      "step": 1350
    },
    {
      "epoch": 0.11385516952699874,
      "grad_norm": 0.46161842346191406,
      "learning_rate": 2e-05,
      "loss": 0.8986,
      "step": 1360
    },
    {
      "epoch": 0.11469233989116785,
      "grad_norm": 0.4660196602344513,
      "learning_rate": 2e-05,
      "loss": 0.8935,
      "step": 1370
    },
    {
      "epoch": 0.11552951025533696,
      "grad_norm": 0.4572381377220154,
      "learning_rate": 2e-05,
      "loss": 0.8861,
      "step": 1380
    },
    {
      "epoch": 0.11636668061950607,
      "grad_norm": 0.45686715841293335,
      "learning_rate": 2e-05,
      "loss": 0.8809,
      "step": 1390
    },
    {
      "epoch": 0.11720385098367518,
      "grad_norm": 0.4713088572025299,
      "learning_rate": 2e-05,
      "loss": 0.8483,
      "step": 1400
    },
    {
      "epoch": 0.11804102134784429,
      "grad_norm": 0.4591504633426666,
      "learning_rate": 2e-05,
      "loss": 0.8993,
      "step": 1410
    },
    {
      "epoch": 0.1188781917120134,
      "grad_norm": 0.4753212034702301,
      "learning_rate": 2e-05,
      "loss": 0.8871,
      "step": 1420
    },
    {
      "epoch": 0.1197153620761825,
      "grad_norm": 0.4820215404033661,
      "learning_rate": 2e-05,
      "loss": 0.8849,
      "step": 1430
    },
    {
      "epoch": 0.1205525324403516,
      "grad_norm": 0.46577829122543335,
      "learning_rate": 2e-05,
      "loss": 0.8699,
      "step": 1440
    },
    {
      "epoch": 0.12138970280452072,
      "grad_norm": 0.46643346548080444,
      "learning_rate": 2e-05,
      "loss": 0.9081,
      "step": 1450
    },
    {
      "epoch": 0.12222687316868983,
      "grad_norm": 0.4909195899963379,
      "learning_rate": 2e-05,
      "loss": 0.9072,
      "step": 1460
    },
    {
      "epoch": 0.12306404353285894,
      "grad_norm": 0.4641784131526947,
      "learning_rate": 2e-05,
      "loss": 0.8778,
      "step": 1470
    },
    {
      "epoch": 0.12390121389702805,
      "grad_norm": 0.47702017426490784,
      "learning_rate": 2e-05,
      "loss": 0.8771,
      "step": 1480
    },
    {
      "epoch": 0.12473838426119715,
      "grad_norm": 0.45714545249938965,
      "learning_rate": 2e-05,
      "loss": 0.8859,
      "step": 1490
    },
    {
      "epoch": 0.12557555462536626,
      "grad_norm": 0.45355841517448425,
      "learning_rate": 2e-05,
      "loss": 0.8945,
      "step": 1500
    },
    {
      "epoch": 0.12641272498953537,
      "grad_norm": 0.584877610206604,
      "learning_rate": 2e-05,
      "loss": 0.8901,
      "step": 1510
    },
    {
      "epoch": 0.12724989535370448,
      "grad_norm": 0.5109133124351501,
      "learning_rate": 2e-05,
      "loss": 0.8709,
      "step": 1520
    },
    {
      "epoch": 0.1280870657178736,
      "grad_norm": 0.4634550213813782,
      "learning_rate": 2e-05,
      "loss": 0.9303,
      "step": 1530
    },
    {
      "epoch": 0.1289242360820427,
      "grad_norm": 0.46686896681785583,
      "learning_rate": 2e-05,
      "loss": 0.8873,
      "step": 1540
    },
    {
      "epoch": 0.1297614064462118,
      "grad_norm": 0.46461957693099976,
      "learning_rate": 2e-05,
      "loss": 0.9004,
      "step": 1550
    },
    {
      "epoch": 0.13059857681038092,
      "grad_norm": 0.4640693664550781,
      "learning_rate": 2e-05,
      "loss": 0.9212,
      "step": 1560
    },
    {
      "epoch": 0.13143574717455003,
      "grad_norm": 0.45362356305122375,
      "learning_rate": 2e-05,
      "loss": 0.8629,
      "step": 1570
    },
    {
      "epoch": 0.13227291753871914,
      "grad_norm": 0.4508582055568695,
      "learning_rate": 2e-05,
      "loss": 0.862,
      "step": 1580
    },
    {
      "epoch": 0.13311008790288822,
      "grad_norm": 0.4616556465625763,
      "learning_rate": 2e-05,
      "loss": 0.9102,
      "step": 1590
    },
    {
      "epoch": 0.13394725826705733,
      "grad_norm": 0.47913822531700134,
      "learning_rate": 2e-05,
      "loss": 0.8826,
      "step": 1600
    },
    {
      "epoch": 0.13478442863122644,
      "grad_norm": 0.47434917092323303,
      "learning_rate": 2e-05,
      "loss": 0.8941,
      "step": 1610
    },
    {
      "epoch": 0.13562159899539555,
      "grad_norm": 0.4437176585197449,
      "learning_rate": 2e-05,
      "loss": 0.8694,
      "step": 1620
    },
    {
      "epoch": 0.13645876935956466,
      "grad_norm": 0.46696802973747253,
      "learning_rate": 2e-05,
      "loss": 0.86,
      "step": 1630
    },
    {
      "epoch": 0.13729593972373377,
      "grad_norm": 0.4766770899295807,
      "learning_rate": 2e-05,
      "loss": 0.9089,
      "step": 1640
    },
    {
      "epoch": 0.13813311008790288,
      "grad_norm": 0.4405345916748047,
      "learning_rate": 2e-05,
      "loss": 0.8723,
      "step": 1650
    },
    {
      "epoch": 0.138970280452072,
      "grad_norm": 0.4427041709423065,
      "learning_rate": 2e-05,
      "loss": 0.8725,
      "step": 1660
    },
    {
      "epoch": 0.1398074508162411,
      "grad_norm": 0.4577621817588806,
      "learning_rate": 2e-05,
      "loss": 0.8701,
      "step": 1670
    },
    {
      "epoch": 0.1406446211804102,
      "grad_norm": 0.47423696517944336,
      "learning_rate": 2e-05,
      "loss": 0.8669,
      "step": 1680
    },
    {
      "epoch": 0.14148179154457932,
      "grad_norm": 0.44417867064476013,
      "learning_rate": 2e-05,
      "loss": 0.8693,
      "step": 1690
    },
    {
      "epoch": 0.14231896190874843,
      "grad_norm": 0.45095181465148926,
      "learning_rate": 2e-05,
      "loss": 0.8832,
      "step": 1700
    },
    {
      "epoch": 0.14315613227291754,
      "grad_norm": 0.4428626000881195,
      "learning_rate": 2e-05,
      "loss": 0.8731,
      "step": 1710
    },
    {
      "epoch": 0.14399330263708665,
      "grad_norm": 0.47432005405426025,
      "learning_rate": 2e-05,
      "loss": 0.8844,
      "step": 1720
    },
    {
      "epoch": 0.14483047300125576,
      "grad_norm": 0.43629175424575806,
      "learning_rate": 2e-05,
      "loss": 0.8358,
      "step": 1730
    },
    {
      "epoch": 0.14566764336542487,
      "grad_norm": 0.4563296437263489,
      "learning_rate": 2e-05,
      "loss": 0.8826,
      "step": 1740
    },
    {
      "epoch": 0.14650481372959398,
      "grad_norm": 0.47259217500686646,
      "learning_rate": 2e-05,
      "loss": 0.8889,
      "step": 1750
    },
    {
      "epoch": 0.1473419840937631,
      "grad_norm": 0.4536290764808655,
      "learning_rate": 2e-05,
      "loss": 0.8768,
      "step": 1760
    },
    {
      "epoch": 0.1481791544579322,
      "grad_norm": 0.45183804631233215,
      "learning_rate": 2e-05,
      "loss": 0.8752,
      "step": 1770
    },
    {
      "epoch": 0.1490163248221013,
      "grad_norm": 0.46316322684288025,
      "learning_rate": 2e-05,
      "loss": 0.8614,
      "step": 1780
    },
    {
      "epoch": 0.14985349518627042,
      "grad_norm": 0.4563836455345154,
      "learning_rate": 2e-05,
      "loss": 0.8634,
      "step": 1790
    },
    {
      "epoch": 0.1506906655504395,
      "grad_norm": 0.4589375853538513,
      "learning_rate": 2e-05,
      "loss": 0.8864,
      "step": 1800
    },
    {
      "epoch": 0.1515278359146086,
      "grad_norm": 0.4381108582019806,
      "learning_rate": 2e-05,
      "loss": 0.8911,
      "step": 1810
    },
    {
      "epoch": 0.15236500627877772,
      "grad_norm": 0.4347422122955322,
      "learning_rate": 2e-05,
      "loss": 0.8715,
      "step": 1820
    },
    {
      "epoch": 0.15320217664294683,
      "grad_norm": 0.4662458598613739,
      "learning_rate": 2e-05,
      "loss": 0.8749,
      "step": 1830
    },
    {
      "epoch": 0.15403934700711594,
      "grad_norm": 0.4737149178981781,
      "learning_rate": 2e-05,
      "loss": 0.8617,
      "step": 1840
    },
    {
      "epoch": 0.15487651737128505,
      "grad_norm": 0.4526223838329315,
      "learning_rate": 2e-05,
      "loss": 0.8528,
      "step": 1850
    },
    {
      "epoch": 0.15571368773545416,
      "grad_norm": 0.44724318385124207,
      "learning_rate": 2e-05,
      "loss": 0.8786,
      "step": 1860
    },
    {
      "epoch": 0.15655085809962327,
      "grad_norm": 0.47638562321662903,
      "learning_rate": 2e-05,
      "loss": 0.8774,
      "step": 1870
    },
    {
      "epoch": 0.15738802846379238,
      "grad_norm": 0.43727222084999084,
      "learning_rate": 2e-05,
      "loss": 0.8815,
      "step": 1880
    },
    {
      "epoch": 0.1582251988279615,
      "grad_norm": 0.49028047919273376,
      "learning_rate": 2e-05,
      "loss": 0.8807,
      "step": 1890
    },
    {
      "epoch": 0.1590623691921306,
      "grad_norm": 0.4596083462238312,
      "learning_rate": 2e-05,
      "loss": 0.9151,
      "step": 1900
    },
    {
      "epoch": 0.1598995395562997,
      "grad_norm": 0.441745400428772,
      "learning_rate": 2e-05,
      "loss": 0.8605,
      "step": 1910
    },
    {
      "epoch": 0.16073670992046882,
      "grad_norm": 0.4500328600406647,
      "learning_rate": 2e-05,
      "loss": 0.8636,
      "step": 1920
    },
    {
      "epoch": 0.16157388028463793,
      "grad_norm": 0.4569918215274811,
      "learning_rate": 2e-05,
      "loss": 0.8626,
      "step": 1930
    },
    {
      "epoch": 0.16241105064880704,
      "grad_norm": 0.6225832104682922,
      "learning_rate": 2e-05,
      "loss": 0.8611,
      "step": 1940
    },
    {
      "epoch": 0.16324822101297615,
      "grad_norm": 0.5002962350845337,
      "learning_rate": 2e-05,
      "loss": 0.8811,
      "step": 1950
    },
    {
      "epoch": 0.16408539137714526,
      "grad_norm": 0.45770740509033203,
      "learning_rate": 2e-05,
      "loss": 0.8823,
      "step": 1960
    },
    {
      "epoch": 0.16492256174131437,
      "grad_norm": 0.456687331199646,
      "learning_rate": 2e-05,
      "loss": 0.8713,
      "step": 1970
    },
    {
      "epoch": 0.16575973210548348,
      "grad_norm": 0.4765065610408783,
      "learning_rate": 2e-05,
      "loss": 0.8961,
      "step": 1980
    },
    {
      "epoch": 0.16659690246965259,
      "grad_norm": 0.44931989908218384,
      "learning_rate": 2e-05,
      "loss": 0.8734,
      "step": 1990
    },
    {
      "epoch": 0.1674340728338217,
      "grad_norm": 0.4341471791267395,
      "learning_rate": 2e-05,
      "loss": 0.8465,
      "step": 2000
    },
    {
      "epoch": 0.16827124319799078,
      "grad_norm": 0.4397622346878052,
      "learning_rate": 2e-05,
      "loss": 0.8663,
      "step": 2010
    },
    {
      "epoch": 0.1691084135621599,
      "grad_norm": 0.4476166367530823,
      "learning_rate": 2e-05,
      "loss": 0.8835,
      "step": 2020
    },
    {
      "epoch": 0.169945583926329,
      "grad_norm": 0.4278470277786255,
      "learning_rate": 2e-05,
      "loss": 0.8695,
      "step": 2030
    },
    {
      "epoch": 0.1707827542904981,
      "grad_norm": 0.45106714963912964,
      "learning_rate": 2e-05,
      "loss": 0.8724,
      "step": 2040
    },
    {
      "epoch": 0.17161992465466722,
      "grad_norm": 0.44101276993751526,
      "learning_rate": 2e-05,
      "loss": 0.8607,
      "step": 2050
    },
    {
      "epoch": 0.17245709501883633,
      "grad_norm": 0.4804248511791229,
      "learning_rate": 2e-05,
      "loss": 0.8921,
      "step": 2060
    },
    {
      "epoch": 0.17329426538300544,
      "grad_norm": 0.4646872878074646,
      "learning_rate": 2e-05,
      "loss": 0.9011,
      "step": 2070
    },
    {
      "epoch": 0.17413143574717455,
      "grad_norm": 0.46392732858657837,
      "learning_rate": 2e-05,
      "loss": 0.8636,
      "step": 2080
    },
    {
      "epoch": 0.17496860611134366,
      "grad_norm": 0.4388207495212555,
      "learning_rate": 2e-05,
      "loss": 0.8762,
      "step": 2090
    },
    {
      "epoch": 0.17580577647551276,
      "grad_norm": 0.4509705901145935,
      "learning_rate": 2e-05,
      "loss": 0.8428,
      "step": 2100
    },
    {
      "epoch": 0.17664294683968187,
      "grad_norm": 0.43261757493019104,
      "learning_rate": 2e-05,
      "loss": 0.8458,
      "step": 2110
    },
    {
      "epoch": 0.17748011720385098,
      "grad_norm": 0.44039949774742126,
      "learning_rate": 2e-05,
      "loss": 0.8654,
      "step": 2120
    },
    {
      "epoch": 0.1783172875680201,
      "grad_norm": 0.44489777088165283,
      "learning_rate": 2e-05,
      "loss": 0.8619,
      "step": 2130
    },
    {
      "epoch": 0.1791544579321892,
      "grad_norm": 0.4366418421268463,
      "learning_rate": 2e-05,
      "loss": 0.8816,
      "step": 2140
    },
    {
      "epoch": 0.1799916282963583,
      "grad_norm": 0.4640413522720337,
      "learning_rate": 2e-05,
      "loss": 0.8647,
      "step": 2150
    },
    {
      "epoch": 0.18082879866052742,
      "grad_norm": 0.4382632374763489,
      "learning_rate": 2e-05,
      "loss": 0.8677,
      "step": 2160
    },
    {
      "epoch": 0.18166596902469653,
      "grad_norm": 0.4480655789375305,
      "learning_rate": 2e-05,
      "loss": 0.8655,
      "step": 2170
    },
    {
      "epoch": 0.18250313938886564,
      "grad_norm": 0.45486730337142944,
      "learning_rate": 2e-05,
      "loss": 0.8733,
      "step": 2180
    },
    {
      "epoch": 0.18334030975303475,
      "grad_norm": 0.45164886116981506,
      "learning_rate": 2e-05,
      "loss": 0.8742,
      "step": 2190
    },
    {
      "epoch": 0.18417748011720386,
      "grad_norm": 0.430537611246109,
      "learning_rate": 2e-05,
      "loss": 0.8707,
      "step": 2200
    },
    {
      "epoch": 0.18501465048137297,
      "grad_norm": 0.48182976245880127,
      "learning_rate": 2e-05,
      "loss": 0.8398,
      "step": 2210
    },
    {
      "epoch": 0.18585182084554205,
      "grad_norm": 0.44728755950927734,
      "learning_rate": 2e-05,
      "loss": 0.8579,
      "step": 2220
    },
    {
      "epoch": 0.18668899120971116,
      "grad_norm": 0.438927561044693,
      "learning_rate": 2e-05,
      "loss": 0.8746,
      "step": 2230
    },
    {
      "epoch": 0.18752616157388027,
      "grad_norm": 0.4833535850048065,
      "learning_rate": 2e-05,
      "loss": 0.8615,
      "step": 2240
    },
    {
      "epoch": 0.18836333193804938,
      "grad_norm": 0.4395700991153717,
      "learning_rate": 2e-05,
      "loss": 0.8502,
      "step": 2250
    },
    {
      "epoch": 0.1892005023022185,
      "grad_norm": 0.5781416296958923,
      "learning_rate": 2e-05,
      "loss": 0.8625,
      "step": 2260
    },
    {
      "epoch": 0.1900376726663876,
      "grad_norm": 0.42882025241851807,
      "learning_rate": 2e-05,
      "loss": 0.8533,
      "step": 2270
    },
    {
      "epoch": 0.1908748430305567,
      "grad_norm": 0.44762805104255676,
      "learning_rate": 2e-05,
      "loss": 0.8412,
      "step": 2280
    },
    {
      "epoch": 0.19171201339472582,
      "grad_norm": 0.45068269968032837,
      "learning_rate": 2e-05,
      "loss": 0.8588,
      "step": 2290
    },
    {
      "epoch": 0.19254918375889493,
      "grad_norm": 0.437988817691803,
      "learning_rate": 2e-05,
      "loss": 0.883,
      "step": 2300
    },
    {
      "epoch": 0.19338635412306404,
      "grad_norm": 0.448384165763855,
      "learning_rate": 2e-05,
      "loss": 0.8528,
      "step": 2310
    },
    {
      "epoch": 0.19422352448723315,
      "grad_norm": 0.474157452583313,
      "learning_rate": 2e-05,
      "loss": 0.8593,
      "step": 2320
    },
    {
      "epoch": 0.19506069485140226,
      "grad_norm": 0.45292016863822937,
      "learning_rate": 2e-05,
      "loss": 0.8254,
      "step": 2330
    },
    {
      "epoch": 0.19589786521557137,
      "grad_norm": 0.43578028678894043,
      "learning_rate": 2e-05,
      "loss": 0.8503,
      "step": 2340
    },
    {
      "epoch": 0.19673503557974048,
      "grad_norm": 0.43714937567710876,
      "learning_rate": 2e-05,
      "loss": 0.8308,
      "step": 2350
    },
    {
      "epoch": 0.1975722059439096,
      "grad_norm": 0.4639517068862915,
      "learning_rate": 2e-05,
      "loss": 0.8762,
      "step": 2360
    },
    {
      "epoch": 0.1984093763080787,
      "grad_norm": 0.44235238432884216,
      "learning_rate": 2e-05,
      "loss": 0.8547,
      "step": 2370
    },
    {
      "epoch": 0.1992465466722478,
      "grad_norm": 0.44393929839134216,
      "learning_rate": 2e-05,
      "loss": 0.8522,
      "step": 2380
    },
    {
      "epoch": 0.20008371703641692,
      "grad_norm": 0.46399083733558655,
      "learning_rate": 2e-05,
      "loss": 0.871,
      "step": 2390
    },
    {
      "epoch": 0.20092088740058603,
      "grad_norm": 0.43202951550483704,
      "learning_rate": 2e-05,
      "loss": 0.8465,
      "step": 2400
    },
    {
      "epoch": 0.20175805776475514,
      "grad_norm": 0.44708123803138733,
      "learning_rate": 2e-05,
      "loss": 0.8665,
      "step": 2410
    },
    {
      "epoch": 0.20259522812892425,
      "grad_norm": 0.4400118291378021,
      "learning_rate": 2e-05,
      "loss": 0.8524,
      "step": 2420
    },
    {
      "epoch": 0.20343239849309336,
      "grad_norm": 0.42689865827560425,
      "learning_rate": 2e-05,
      "loss": 0.8455,
      "step": 2430
    },
    {
      "epoch": 0.20426956885726244,
      "grad_norm": 0.45444467663764954,
      "learning_rate": 2e-05,
      "loss": 0.8492,
      "step": 2440
    },
    {
      "epoch": 0.20510673922143155,
      "grad_norm": 0.47486284375190735,
      "learning_rate": 2e-05,
      "loss": 0.8689,
      "step": 2450
    },
    {
      "epoch": 0.20594390958560066,
      "grad_norm": 0.4352690875530243,
      "learning_rate": 2e-05,
      "loss": 0.8516,
      "step": 2460
    },
    {
      "epoch": 0.20678107994976977,
      "grad_norm": 0.4403368830680847,
      "learning_rate": 2e-05,
      "loss": 0.847,
      "step": 2470
    },
    {
      "epoch": 0.20761825031393888,
      "grad_norm": 0.44160476326942444,
      "learning_rate": 2e-05,
      "loss": 0.8681,
      "step": 2480
    },
    {
      "epoch": 0.208455420678108,
      "grad_norm": 0.43098336458206177,
      "learning_rate": 2e-05,
      "loss": 0.8433,
      "step": 2490
    },
    {
      "epoch": 0.2092925910422771,
      "grad_norm": 0.46148884296417236,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 2500
    },
    {
      "epoch": 0.2101297614064462,
      "grad_norm": 0.4290294051170349,
      "learning_rate": 2e-05,
      "loss": 0.8546,
      "step": 2510
    },
    {
      "epoch": 0.21096693177061532,
      "grad_norm": 0.4311007559299469,
      "learning_rate": 2e-05,
      "loss": 0.8763,
      "step": 2520
    },
    {
      "epoch": 0.21180410213478443,
      "grad_norm": 0.45104533433914185,
      "learning_rate": 2e-05,
      "loss": 0.8489,
      "step": 2530
    },
    {
      "epoch": 0.21264127249895354,
      "grad_norm": 0.42825406789779663,
      "learning_rate": 2e-05,
      "loss": 0.8633,
      "step": 2540
    },
    {
      "epoch": 0.21347844286312265,
      "grad_norm": 0.42387330532073975,
      "learning_rate": 2e-05,
      "loss": 0.8422,
      "step": 2550
    },
    {
      "epoch": 0.21431561322729176,
      "grad_norm": 0.4319068193435669,
      "learning_rate": 2e-05,
      "loss": 0.8824,
      "step": 2560
    },
    {
      "epoch": 0.21515278359146087,
      "grad_norm": 0.4178818464279175,
      "learning_rate": 2e-05,
      "loss": 0.8442,
      "step": 2570
    },
    {
      "epoch": 0.21598995395562998,
      "grad_norm": 0.4258483946323395,
      "learning_rate": 2e-05,
      "loss": 0.8536,
      "step": 2580
    },
    {
      "epoch": 0.21682712431979909,
      "grad_norm": 0.4532221555709839,
      "learning_rate": 2e-05,
      "loss": 0.8476,
      "step": 2590
    },
    {
      "epoch": 0.2176642946839682,
      "grad_norm": 0.4477638304233551,
      "learning_rate": 2e-05,
      "loss": 0.8707,
      "step": 2600
    },
    {
      "epoch": 0.2185014650481373,
      "grad_norm": 0.4326961040496826,
      "learning_rate": 2e-05,
      "loss": 0.8726,
      "step": 2610
    },
    {
      "epoch": 0.21933863541230642,
      "grad_norm": 0.46689221262931824,
      "learning_rate": 2e-05,
      "loss": 0.8736,
      "step": 2620
    },
    {
      "epoch": 0.22017580577647552,
      "grad_norm": 0.4493032991886139,
      "learning_rate": 2e-05,
      "loss": 0.8711,
      "step": 2630
    },
    {
      "epoch": 0.22101297614064463,
      "grad_norm": 0.4334334433078766,
      "learning_rate": 2e-05,
      "loss": 0.8675,
      "step": 2640
    },
    {
      "epoch": 0.22185014650481372,
      "grad_norm": 0.4326641261577606,
      "learning_rate": 2e-05,
      "loss": 0.829,
      "step": 2650
    },
    {
      "epoch": 0.22268731686898283,
      "grad_norm": 0.4365394413471222,
      "learning_rate": 2e-05,
      "loss": 0.8503,
      "step": 2660
    },
    {
      "epoch": 0.22352448723315194,
      "grad_norm": 0.47003981471061707,
      "learning_rate": 2e-05,
      "loss": 0.8567,
      "step": 2670
    },
    {
      "epoch": 0.22436165759732105,
      "grad_norm": 0.41743093729019165,
      "learning_rate": 2e-05,
      "loss": 0.8451,
      "step": 2680
    },
    {
      "epoch": 0.22519882796149016,
      "grad_norm": 0.44406813383102417,
      "learning_rate": 2e-05,
      "loss": 0.8643,
      "step": 2690
    },
    {
      "epoch": 0.22603599832565927,
      "grad_norm": 0.42659661173820496,
      "learning_rate": 2e-05,
      "loss": 0.8687,
      "step": 2700
    },
    {
      "epoch": 0.22687316868982838,
      "grad_norm": 0.4428755044937134,
      "learning_rate": 2e-05,
      "loss": 0.8524,
      "step": 2710
    },
    {
      "epoch": 0.22771033905399748,
      "grad_norm": 0.42310386896133423,
      "learning_rate": 2e-05,
      "loss": 0.8484,
      "step": 2720
    },
    {
      "epoch": 0.2285475094181666,
      "grad_norm": 0.4624371826648712,
      "learning_rate": 2e-05,
      "loss": 0.8327,
      "step": 2730
    },
    {
      "epoch": 0.2293846797823357,
      "grad_norm": 0.43734869360923767,
      "learning_rate": 2e-05,
      "loss": 0.8518,
      "step": 2740
    },
    {
      "epoch": 0.23022185014650481,
      "grad_norm": 0.4506842792034149,
      "learning_rate": 2e-05,
      "loss": 0.8565,
      "step": 2750
    },
    {
      "epoch": 0.23105902051067392,
      "grad_norm": 0.48162660002708435,
      "learning_rate": 2e-05,
      "loss": 0.8417,
      "step": 2760
    },
    {
      "epoch": 0.23189619087484303,
      "grad_norm": 0.5661482810974121,
      "learning_rate": 2e-05,
      "loss": 0.8662,
      "step": 2770
    },
    {
      "epoch": 0.23273336123901214,
      "grad_norm": 0.42844292521476746,
      "learning_rate": 2e-05,
      "loss": 0.8342,
      "step": 2780
    },
    {
      "epoch": 0.23357053160318125,
      "grad_norm": 0.43888863921165466,
      "learning_rate": 2e-05,
      "loss": 0.8846,
      "step": 2790
    },
    {
      "epoch": 0.23440770196735036,
      "grad_norm": 0.4474732577800751,
      "learning_rate": 2e-05,
      "loss": 0.8317,
      "step": 2800
    },
    {
      "epoch": 0.23524487233151947,
      "grad_norm": 0.4584842026233673,
      "learning_rate": 2e-05,
      "loss": 0.8915,
      "step": 2810
    },
    {
      "epoch": 0.23608204269568858,
      "grad_norm": 0.43371322751045227,
      "learning_rate": 2e-05,
      "loss": 0.8526,
      "step": 2820
    },
    {
      "epoch": 0.2369192130598577,
      "grad_norm": 0.4327155351638794,
      "learning_rate": 2e-05,
      "loss": 0.8677,
      "step": 2830
    },
    {
      "epoch": 0.2377563834240268,
      "grad_norm": 0.4316923916339874,
      "learning_rate": 2e-05,
      "loss": 0.8331,
      "step": 2840
    },
    {
      "epoch": 0.2385935537881959,
      "grad_norm": 0.42735129594802856,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 2850
    },
    {
      "epoch": 0.239430724152365,
      "grad_norm": 0.4395933449268341,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 2860
    },
    {
      "epoch": 0.2402678945165341,
      "grad_norm": 0.4804455041885376,
      "learning_rate": 2e-05,
      "loss": 0.8238,
      "step": 2870
    },
    {
      "epoch": 0.2411050648807032,
      "grad_norm": 0.4475085437297821,
      "learning_rate": 2e-05,
      "loss": 0.8763,
      "step": 2880
    },
    {
      "epoch": 0.24194223524487232,
      "grad_norm": 0.4516408443450928,
      "learning_rate": 2e-05,
      "loss": 0.8444,
      "step": 2890
    },
    {
      "epoch": 0.24277940560904143,
      "grad_norm": 0.46845370531082153,
      "learning_rate": 2e-05,
      "loss": 0.8214,
      "step": 2900
    },
    {
      "epoch": 0.24361657597321054,
      "grad_norm": 0.4366689622402191,
      "learning_rate": 2e-05,
      "loss": 0.8308,
      "step": 2910
    },
    {
      "epoch": 0.24445374633737965,
      "grad_norm": 0.4363156855106354,
      "learning_rate": 2e-05,
      "loss": 0.8394,
      "step": 2920
    },
    {
      "epoch": 0.24529091670154876,
      "grad_norm": 0.4714019000530243,
      "learning_rate": 2e-05,
      "loss": 0.8688,
      "step": 2930
    },
    {
      "epoch": 0.24612808706571787,
      "grad_norm": 0.48325541615486145,
      "learning_rate": 2e-05,
      "loss": 0.8354,
      "step": 2940
    },
    {
      "epoch": 0.24696525742988698,
      "grad_norm": 0.4532882273197174,
      "learning_rate": 2e-05,
      "loss": 0.8763,
      "step": 2950
    },
    {
      "epoch": 0.2478024277940561,
      "grad_norm": 0.4276827573776245,
      "learning_rate": 2e-05,
      "loss": 0.861,
      "step": 2960
    },
    {
      "epoch": 0.2486395981582252,
      "grad_norm": 0.42689263820648193,
      "learning_rate": 2e-05,
      "loss": 0.8619,
      "step": 2970
    },
    {
      "epoch": 0.2494767685223943,
      "grad_norm": 0.4465515911579132,
      "learning_rate": 2e-05,
      "loss": 0.8722,
      "step": 2980
    },
    {
      "epoch": 0.2503139388865634,
      "grad_norm": 0.45104026794433594,
      "learning_rate": 2e-05,
      "loss": 0.8663,
      "step": 2990
    },
    {
      "epoch": 0.25115110925073253,
      "grad_norm": 0.44007986783981323,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 3000
    },
    {
      "epoch": 0.2519882796149016,
      "grad_norm": 0.43077486753463745,
      "learning_rate": 2e-05,
      "loss": 0.8545,
      "step": 3010
    },
    {
      "epoch": 0.25282544997907075,
      "grad_norm": 0.4256698787212372,
      "learning_rate": 2e-05,
      "loss": 0.8425,
      "step": 3020
    },
    {
      "epoch": 0.25366262034323983,
      "grad_norm": 0.44382596015930176,
      "learning_rate": 2e-05,
      "loss": 0.845,
      "step": 3030
    },
    {
      "epoch": 0.25449979070740897,
      "grad_norm": 0.44387444853782654,
      "learning_rate": 2e-05,
      "loss": 0.868,
      "step": 3040
    },
    {
      "epoch": 0.25533696107157805,
      "grad_norm": 0.42307358980178833,
      "learning_rate": 2e-05,
      "loss": 0.8586,
      "step": 3050
    },
    {
      "epoch": 0.2561741314357472,
      "grad_norm": 0.5496047139167786,
      "learning_rate": 2e-05,
      "loss": 0.875,
      "step": 3060
    },
    {
      "epoch": 0.25701130179991627,
      "grad_norm": 0.43474847078323364,
      "learning_rate": 2e-05,
      "loss": 0.8445,
      "step": 3070
    },
    {
      "epoch": 0.2578484721640854,
      "grad_norm": 0.4298436939716339,
      "learning_rate": 2e-05,
      "loss": 0.8354,
      "step": 3080
    },
    {
      "epoch": 0.2586856425282545,
      "grad_norm": 0.433672696352005,
      "learning_rate": 2e-05,
      "loss": 0.8307,
      "step": 3090
    },
    {
      "epoch": 0.2595228128924236,
      "grad_norm": 0.4268275797367096,
      "learning_rate": 2e-05,
      "loss": 0.8408,
      "step": 3100
    },
    {
      "epoch": 0.2603599832565927,
      "grad_norm": 0.44805261492729187,
      "learning_rate": 2e-05,
      "loss": 0.8805,
      "step": 3110
    },
    {
      "epoch": 0.26119715362076185,
      "grad_norm": 0.41279494762420654,
      "learning_rate": 2e-05,
      "loss": 0.8571,
      "step": 3120
    },
    {
      "epoch": 0.26203432398493093,
      "grad_norm": 0.4382464587688446,
      "learning_rate": 2e-05,
      "loss": 0.8408,
      "step": 3130
    },
    {
      "epoch": 0.26287149434910007,
      "grad_norm": 0.43196719884872437,
      "learning_rate": 2e-05,
      "loss": 0.8359,
      "step": 3140
    },
    {
      "epoch": 0.26370866471326915,
      "grad_norm": 0.4206157326698303,
      "learning_rate": 2e-05,
      "loss": 0.8781,
      "step": 3150
    },
    {
      "epoch": 0.2645458350774383,
      "grad_norm": 0.4439392387866974,
      "learning_rate": 2e-05,
      "loss": 0.8353,
      "step": 3160
    },
    {
      "epoch": 0.26538300544160737,
      "grad_norm": 0.43137675523757935,
      "learning_rate": 2e-05,
      "loss": 0.8543,
      "step": 3170
    },
    {
      "epoch": 0.26622017580577645,
      "grad_norm": 0.4718741178512573,
      "learning_rate": 2e-05,
      "loss": 0.8766,
      "step": 3180
    },
    {
      "epoch": 0.2670573461699456,
      "grad_norm": 0.4124031662940979,
      "learning_rate": 2e-05,
      "loss": 0.856,
      "step": 3190
    },
    {
      "epoch": 0.26789451653411467,
      "grad_norm": 0.44760680198669434,
      "learning_rate": 2e-05,
      "loss": 0.8495,
      "step": 3200
    },
    {
      "epoch": 0.2687316868982838,
      "grad_norm": 0.4445425570011139,
      "learning_rate": 2e-05,
      "loss": 0.866,
      "step": 3210
    },
    {
      "epoch": 0.2695688572624529,
      "grad_norm": 0.43041038513183594,
      "learning_rate": 2e-05,
      "loss": 0.8411,
      "step": 3220
    },
    {
      "epoch": 0.270406027626622,
      "grad_norm": 0.42874640226364136,
      "learning_rate": 2e-05,
      "loss": 0.8639,
      "step": 3230
    },
    {
      "epoch": 0.2712431979907911,
      "grad_norm": 0.4241209626197815,
      "learning_rate": 2e-05,
      "loss": 0.8662,
      "step": 3240
    },
    {
      "epoch": 0.27208036835496024,
      "grad_norm": 0.4623543322086334,
      "learning_rate": 2e-05,
      "loss": 0.8692,
      "step": 3250
    },
    {
      "epoch": 0.2729175387191293,
      "grad_norm": 0.41813695430755615,
      "learning_rate": 2e-05,
      "loss": 0.8454,
      "step": 3260
    },
    {
      "epoch": 0.27375470908329846,
      "grad_norm": 0.43213996291160583,
      "learning_rate": 2e-05,
      "loss": 0.8264,
      "step": 3270
    },
    {
      "epoch": 0.27459187944746755,
      "grad_norm": 0.43402811884880066,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 3280
    },
    {
      "epoch": 0.2754290498116367,
      "grad_norm": 0.4213794767856598,
      "learning_rate": 2e-05,
      "loss": 0.8348,
      "step": 3290
    },
    {
      "epoch": 0.27626622017580577,
      "grad_norm": 0.4604377746582031,
      "learning_rate": 2e-05,
      "loss": 0.8314,
      "step": 3300
    },
    {
      "epoch": 0.2771033905399749,
      "grad_norm": 0.41930535435676575,
      "learning_rate": 2e-05,
      "loss": 0.8619,
      "step": 3310
    },
    {
      "epoch": 0.277940560904144,
      "grad_norm": 0.44456246495246887,
      "learning_rate": 2e-05,
      "loss": 0.8373,
      "step": 3320
    },
    {
      "epoch": 0.2787777312683131,
      "grad_norm": 0.42515015602111816,
      "learning_rate": 2e-05,
      "loss": 0.8325,
      "step": 3330
    },
    {
      "epoch": 0.2796149016324822,
      "grad_norm": 0.543720543384552,
      "learning_rate": 2e-05,
      "loss": 0.8514,
      "step": 3340
    },
    {
      "epoch": 0.28045207199665134,
      "grad_norm": 0.4654465317726135,
      "learning_rate": 2e-05,
      "loss": 0.8256,
      "step": 3350
    },
    {
      "epoch": 0.2812892423608204,
      "grad_norm": 0.43853747844696045,
      "learning_rate": 2e-05,
      "loss": 0.8471,
      "step": 3360
    },
    {
      "epoch": 0.28212641272498956,
      "grad_norm": 0.4390884339809418,
      "learning_rate": 2e-05,
      "loss": 0.83,
      "step": 3370
    },
    {
      "epoch": 0.28296358308915864,
      "grad_norm": 0.427254319190979,
      "learning_rate": 2e-05,
      "loss": 0.8267,
      "step": 3380
    },
    {
      "epoch": 0.2838007534533277,
      "grad_norm": 0.5277079343795776,
      "learning_rate": 2e-05,
      "loss": 0.8306,
      "step": 3390
    },
    {
      "epoch": 0.28463792381749686,
      "grad_norm": 0.4379133880138397,
      "learning_rate": 2e-05,
      "loss": 0.8403,
      "step": 3400
    },
    {
      "epoch": 0.28547509418166594,
      "grad_norm": 0.4159203767776489,
      "learning_rate": 2e-05,
      "loss": 0.8257,
      "step": 3410
    },
    {
      "epoch": 0.2863122645458351,
      "grad_norm": 0.40820810198783875,
      "learning_rate": 2e-05,
      "loss": 0.8397,
      "step": 3420
    },
    {
      "epoch": 0.28714943491000416,
      "grad_norm": 0.4453214406967163,
      "learning_rate": 2e-05,
      "loss": 0.8553,
      "step": 3430
    },
    {
      "epoch": 0.2879866052741733,
      "grad_norm": 0.44322121143341064,
      "learning_rate": 2e-05,
      "loss": 0.8469,
      "step": 3440
    },
    {
      "epoch": 0.2888237756383424,
      "grad_norm": 0.4469110667705536,
      "learning_rate": 2e-05,
      "loss": 0.8271,
      "step": 3450
    },
    {
      "epoch": 0.2896609460025115,
      "grad_norm": 0.4392768442630768,
      "learning_rate": 2e-05,
      "loss": 0.8399,
      "step": 3460
    },
    {
      "epoch": 0.2904981163666806,
      "grad_norm": 0.4181481897830963,
      "learning_rate": 2e-05,
      "loss": 0.8604,
      "step": 3470
    },
    {
      "epoch": 0.29133528673084974,
      "grad_norm": 0.428493857383728,
      "learning_rate": 2e-05,
      "loss": 0.8402,
      "step": 3480
    },
    {
      "epoch": 0.2921724570950188,
      "grad_norm": 0.4284713566303253,
      "learning_rate": 2e-05,
      "loss": 0.8434,
      "step": 3490
    },
    {
      "epoch": 0.29300962745918796,
      "grad_norm": 0.43312734365463257,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 3500
    },
    {
      "epoch": 0.29384679782335704,
      "grad_norm": 0.4221225678920746,
      "learning_rate": 2e-05,
      "loss": 0.847,
      "step": 3510
    },
    {
      "epoch": 0.2946839681875262,
      "grad_norm": 0.4417845606803894,
      "learning_rate": 2e-05,
      "loss": 0.8526,
      "step": 3520
    },
    {
      "epoch": 0.29552113855169526,
      "grad_norm": 0.417818546295166,
      "learning_rate": 2e-05,
      "loss": 0.835,
      "step": 3530
    },
    {
      "epoch": 0.2963583089158644,
      "grad_norm": 0.4408692419528961,
      "learning_rate": 2e-05,
      "loss": 0.8325,
      "step": 3540
    },
    {
      "epoch": 0.2971954792800335,
      "grad_norm": 0.42140427231788635,
      "learning_rate": 2e-05,
      "loss": 0.86,
      "step": 3550
    },
    {
      "epoch": 0.2980326496442026,
      "grad_norm": 0.42939653992652893,
      "learning_rate": 2e-05,
      "loss": 0.848,
      "step": 3560
    },
    {
      "epoch": 0.2988698200083717,
      "grad_norm": 0.4309823513031006,
      "learning_rate": 2e-05,
      "loss": 0.8467,
      "step": 3570
    },
    {
      "epoch": 0.29970699037254084,
      "grad_norm": 0.41343390941619873,
      "learning_rate": 2e-05,
      "loss": 0.8462,
      "step": 3580
    },
    {
      "epoch": 0.3005441607367099,
      "grad_norm": 0.44925427436828613,
      "learning_rate": 2e-05,
      "loss": 0.8608,
      "step": 3590
    },
    {
      "epoch": 0.301381331100879,
      "grad_norm": 0.4298679530620575,
      "learning_rate": 2e-05,
      "loss": 0.834,
      "step": 3600
    },
    {
      "epoch": 0.30221850146504814,
      "grad_norm": 0.43456289172172546,
      "learning_rate": 2e-05,
      "loss": 0.8484,
      "step": 3610
    },
    {
      "epoch": 0.3030556718292172,
      "grad_norm": 0.441661536693573,
      "learning_rate": 2e-05,
      "loss": 0.8242,
      "step": 3620
    },
    {
      "epoch": 0.30389284219338636,
      "grad_norm": 0.4271041750907898,
      "learning_rate": 2e-05,
      "loss": 0.812,
      "step": 3630
    },
    {
      "epoch": 0.30473001255755544,
      "grad_norm": 0.41006773710250854,
      "learning_rate": 2e-05,
      "loss": 0.8424,
      "step": 3640
    },
    {
      "epoch": 0.3055671829217246,
      "grad_norm": 0.4275254011154175,
      "learning_rate": 2e-05,
      "loss": 0.8317,
      "step": 3650
    },
    {
      "epoch": 0.30640435328589366,
      "grad_norm": 0.42956051230430603,
      "learning_rate": 2e-05,
      "loss": 0.8585,
      "step": 3660
    },
    {
      "epoch": 0.3072415236500628,
      "grad_norm": 0.44132646918296814,
      "learning_rate": 2e-05,
      "loss": 0.8378,
      "step": 3670
    },
    {
      "epoch": 0.3080786940142319,
      "grad_norm": 0.45235463976860046,
      "learning_rate": 2e-05,
      "loss": 0.8573,
      "step": 3680
    },
    {
      "epoch": 0.308915864378401,
      "grad_norm": 0.4310559630393982,
      "learning_rate": 2e-05,
      "loss": 0.8248,
      "step": 3690
    },
    {
      "epoch": 0.3097530347425701,
      "grad_norm": 0.44843408465385437,
      "learning_rate": 2e-05,
      "loss": 0.8343,
      "step": 3700
    },
    {
      "epoch": 0.31059020510673924,
      "grad_norm": 0.4408901035785675,
      "learning_rate": 2e-05,
      "loss": 0.8326,
      "step": 3710
    },
    {
      "epoch": 0.3114273754709083,
      "grad_norm": 0.43724024295806885,
      "learning_rate": 2e-05,
      "loss": 0.8313,
      "step": 3720
    },
    {
      "epoch": 0.31226454583507746,
      "grad_norm": 0.41545209288597107,
      "learning_rate": 2e-05,
      "loss": 0.8299,
      "step": 3730
    },
    {
      "epoch": 0.31310171619924654,
      "grad_norm": 0.43244078755378723,
      "learning_rate": 2e-05,
      "loss": 0.8495,
      "step": 3740
    },
    {
      "epoch": 0.3139388865634157,
      "grad_norm": 0.4475247263908386,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 3750
    },
    {
      "epoch": 0.31477605692758476,
      "grad_norm": 0.4203198254108429,
      "learning_rate": 2e-05,
      "loss": 0.8505,
      "step": 3760
    },
    {
      "epoch": 0.3156132272917539,
      "grad_norm": 0.4319896101951599,
      "learning_rate": 2e-05,
      "loss": 0.8182,
      "step": 3770
    },
    {
      "epoch": 0.316450397655923,
      "grad_norm": 0.4427873194217682,
      "learning_rate": 2e-05,
      "loss": 0.8509,
      "step": 3780
    },
    {
      "epoch": 0.3172875680200921,
      "grad_norm": 0.4312664568424225,
      "learning_rate": 2e-05,
      "loss": 0.8504,
      "step": 3790
    },
    {
      "epoch": 0.3181247383842612,
      "grad_norm": 0.4734603464603424,
      "learning_rate": 2e-05,
      "loss": 0.8639,
      "step": 3800
    },
    {
      "epoch": 0.3189619087484303,
      "grad_norm": 0.42495936155319214,
      "learning_rate": 2e-05,
      "loss": 0.8335,
      "step": 3810
    },
    {
      "epoch": 0.3197990791125994,
      "grad_norm": 0.42727217078208923,
      "learning_rate": 2e-05,
      "loss": 0.8107,
      "step": 3820
    },
    {
      "epoch": 0.3206362494767685,
      "grad_norm": 0.41671445965766907,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 3830
    },
    {
      "epoch": 0.32147341984093764,
      "grad_norm": 0.41387617588043213,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 3840
    },
    {
      "epoch": 0.3223105902051067,
      "grad_norm": 0.45229795575141907,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 3850
    },
    {
      "epoch": 0.32314776056927585,
      "grad_norm": 0.453480988740921,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 3860
    },
    {
      "epoch": 0.32398493093344494,
      "grad_norm": 0.4310716390609741,
      "learning_rate": 2e-05,
      "loss": 0.8372,
      "step": 3870
    },
    {
      "epoch": 0.3248221012976141,
      "grad_norm": 0.41317251324653625,
      "learning_rate": 2e-05,
      "loss": 0.838,
      "step": 3880
    },
    {
      "epoch": 0.32565927166178316,
      "grad_norm": 0.41291454434394836,
      "learning_rate": 2e-05,
      "loss": 0.833,
      "step": 3890
    },
    {
      "epoch": 0.3264964420259523,
      "grad_norm": 0.4336530566215515,
      "learning_rate": 2e-05,
      "loss": 0.8128,
      "step": 3900
    },
    {
      "epoch": 0.3273336123901214,
      "grad_norm": 0.4162159860134125,
      "learning_rate": 2e-05,
      "loss": 0.8258,
      "step": 3910
    },
    {
      "epoch": 0.3281707827542905,
      "grad_norm": 0.43340811133384705,
      "learning_rate": 2e-05,
      "loss": 0.8643,
      "step": 3920
    },
    {
      "epoch": 0.3290079531184596,
      "grad_norm": 0.4096095561981201,
      "learning_rate": 2e-05,
      "loss": 0.8387,
      "step": 3930
    },
    {
      "epoch": 0.32984512348262873,
      "grad_norm": 0.4369688332080841,
      "learning_rate": 2e-05,
      "loss": 0.8285,
      "step": 3940
    },
    {
      "epoch": 0.3306822938467978,
      "grad_norm": 0.459893137216568,
      "learning_rate": 2e-05,
      "loss": 0.858,
      "step": 3950
    },
    {
      "epoch": 0.33151946421096695,
      "grad_norm": 0.4191811978816986,
      "learning_rate": 2e-05,
      "loss": 0.8116,
      "step": 3960
    },
    {
      "epoch": 0.33235663457513603,
      "grad_norm": 0.4309510886669159,
      "learning_rate": 2e-05,
      "loss": 0.8391,
      "step": 3970
    },
    {
      "epoch": 0.33319380493930517,
      "grad_norm": 0.4352155923843384,
      "learning_rate": 2e-05,
      "loss": 0.8508,
      "step": 3980
    },
    {
      "epoch": 0.33403097530347425,
      "grad_norm": 0.4311888515949249,
      "learning_rate": 2e-05,
      "loss": 0.8085,
      "step": 3990
    },
    {
      "epoch": 0.3348681456676434,
      "grad_norm": 0.42564472556114197,
      "learning_rate": 2e-05,
      "loss": 0.8175,
      "step": 4000
    },
    {
      "epoch": 0.3357053160318125,
      "grad_norm": 0.4221665561199188,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 4010
    },
    {
      "epoch": 0.33654248639598155,
      "grad_norm": 0.43019554018974304,
      "learning_rate": 2e-05,
      "loss": 0.8532,
      "step": 4020
    },
    {
      "epoch": 0.3373796567601507,
      "grad_norm": 0.4272821843624115,
      "learning_rate": 2e-05,
      "loss": 0.8313,
      "step": 4030
    },
    {
      "epoch": 0.3382168271243198,
      "grad_norm": 0.4159959852695465,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 4040
    },
    {
      "epoch": 0.3390539974884889,
      "grad_norm": 0.43675434589385986,
      "learning_rate": 2e-05,
      "loss": 0.8059,
      "step": 4050
    },
    {
      "epoch": 0.339891167852658,
      "grad_norm": 0.43672120571136475,
      "learning_rate": 2e-05,
      "loss": 0.8453,
      "step": 4060
    },
    {
      "epoch": 0.34072833821682713,
      "grad_norm": 0.427731990814209,
      "learning_rate": 2e-05,
      "loss": 0.8355,
      "step": 4070
    },
    {
      "epoch": 0.3415655085809962,
      "grad_norm": 0.41670680046081543,
      "learning_rate": 2e-05,
      "loss": 0.8242,
      "step": 4080
    },
    {
      "epoch": 0.34240267894516535,
      "grad_norm": 0.4309384226799011,
      "learning_rate": 2e-05,
      "loss": 0.827,
      "step": 4090
    },
    {
      "epoch": 0.34323984930933443,
      "grad_norm": 0.42762354016304016,
      "learning_rate": 2e-05,
      "loss": 0.8564,
      "step": 4100
    },
    {
      "epoch": 0.34407701967350357,
      "grad_norm": 0.4144584834575653,
      "learning_rate": 2e-05,
      "loss": 0.8427,
      "step": 4110
    },
    {
      "epoch": 0.34491419003767265,
      "grad_norm": 0.4221765398979187,
      "learning_rate": 2e-05,
      "loss": 0.8122,
      "step": 4120
    },
    {
      "epoch": 0.3457513604018418,
      "grad_norm": 0.4422992467880249,
      "learning_rate": 2e-05,
      "loss": 0.8219,
      "step": 4130
    },
    {
      "epoch": 0.34658853076601087,
      "grad_norm": 0.4365583658218384,
      "learning_rate": 2e-05,
      "loss": 0.8432,
      "step": 4140
    },
    {
      "epoch": 0.34742570113018,
      "grad_norm": 0.3962838053703308,
      "learning_rate": 2e-05,
      "loss": 0.8413,
      "step": 4150
    },
    {
      "epoch": 0.3482628714943491,
      "grad_norm": 0.43464258313179016,
      "learning_rate": 2e-05,
      "loss": 0.8494,
      "step": 4160
    },
    {
      "epoch": 0.34910004185851823,
      "grad_norm": 0.4626946747303009,
      "learning_rate": 2e-05,
      "loss": 0.8419,
      "step": 4170
    },
    {
      "epoch": 0.3499372122226873,
      "grad_norm": 0.4124692678451538,
      "learning_rate": 2e-05,
      "loss": 0.8163,
      "step": 4180
    },
    {
      "epoch": 0.35077438258685645,
      "grad_norm": 0.4296865165233612,
      "learning_rate": 2e-05,
      "loss": 0.8438,
      "step": 4190
    },
    {
      "epoch": 0.35161155295102553,
      "grad_norm": 0.40684860944747925,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 4200
    },
    {
      "epoch": 0.35244872331519467,
      "grad_norm": 0.43835723400115967,
      "learning_rate": 2e-05,
      "loss": 0.8444,
      "step": 4210
    },
    {
      "epoch": 0.35328589367936375,
      "grad_norm": 0.43864011764526367,
      "learning_rate": 2e-05,
      "loss": 0.8293,
      "step": 4220
    },
    {
      "epoch": 0.35412306404353283,
      "grad_norm": 0.4002416431903839,
      "learning_rate": 2e-05,
      "loss": 0.8104,
      "step": 4230
    },
    {
      "epoch": 0.35496023440770197,
      "grad_norm": 0.43456172943115234,
      "learning_rate": 2e-05,
      "loss": 0.8428,
      "step": 4240
    },
    {
      "epoch": 0.35579740477187105,
      "grad_norm": 0.43075916171073914,
      "learning_rate": 2e-05,
      "loss": 0.8571,
      "step": 4250
    },
    {
      "epoch": 0.3566345751360402,
      "grad_norm": 0.4223424196243286,
      "learning_rate": 2e-05,
      "loss": 0.821,
      "step": 4260
    },
    {
      "epoch": 0.35747174550020927,
      "grad_norm": 0.43250206112861633,
      "learning_rate": 2e-05,
      "loss": 0.8144,
      "step": 4270
    },
    {
      "epoch": 0.3583089158643784,
      "grad_norm": 0.41428571939468384,
      "learning_rate": 2e-05,
      "loss": 0.8397,
      "step": 4280
    },
    {
      "epoch": 0.3591460862285475,
      "grad_norm": 0.43673330545425415,
      "learning_rate": 2e-05,
      "loss": 0.837,
      "step": 4290
    },
    {
      "epoch": 0.3599832565927166,
      "grad_norm": 0.4234159588813782,
      "learning_rate": 2e-05,
      "loss": 0.8376,
      "step": 4300
    },
    {
      "epoch": 0.3608204269568857,
      "grad_norm": 0.4296039044857025,
      "learning_rate": 2e-05,
      "loss": 0.8396,
      "step": 4310
    },
    {
      "epoch": 0.36165759732105485,
      "grad_norm": 0.41784560680389404,
      "learning_rate": 2e-05,
      "loss": 0.798,
      "step": 4320
    },
    {
      "epoch": 0.36249476768522393,
      "grad_norm": 0.4596424400806427,
      "learning_rate": 2e-05,
      "loss": 0.8442,
      "step": 4330
    },
    {
      "epoch": 0.36333193804939307,
      "grad_norm": 0.4594140350818634,
      "learning_rate": 2e-05,
      "loss": 0.8428,
      "step": 4340
    },
    {
      "epoch": 0.36416910841356215,
      "grad_norm": 0.408284991979599,
      "learning_rate": 2e-05,
      "loss": 0.8223,
      "step": 4350
    },
    {
      "epoch": 0.3650062787777313,
      "grad_norm": 0.431534081697464,
      "learning_rate": 2e-05,
      "loss": 0.8533,
      "step": 4360
    },
    {
      "epoch": 0.36584344914190037,
      "grad_norm": 0.42122581601142883,
      "learning_rate": 2e-05,
      "loss": 0.8118,
      "step": 4370
    },
    {
      "epoch": 0.3666806195060695,
      "grad_norm": 0.4410568177700043,
      "learning_rate": 2e-05,
      "loss": 0.8503,
      "step": 4380
    },
    {
      "epoch": 0.3675177898702386,
      "grad_norm": 0.48844408988952637,
      "learning_rate": 2e-05,
      "loss": 0.8378,
      "step": 4390
    },
    {
      "epoch": 0.3683549602344077,
      "grad_norm": 0.4141460359096527,
      "learning_rate": 2e-05,
      "loss": 0.8253,
      "step": 4400
    },
    {
      "epoch": 0.3691921305985768,
      "grad_norm": 0.4506625831127167,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 4410
    },
    {
      "epoch": 0.37002930096274594,
      "grad_norm": 0.42208731174468994,
      "learning_rate": 2e-05,
      "loss": 0.7994,
      "step": 4420
    },
    {
      "epoch": 0.370866471326915,
      "grad_norm": 0.4610078036785126,
      "learning_rate": 2e-05,
      "loss": 0.8222,
      "step": 4430
    },
    {
      "epoch": 0.3717036416910841,
      "grad_norm": 0.41858646273612976,
      "learning_rate": 2e-05,
      "loss": 0.8421,
      "step": 4440
    },
    {
      "epoch": 0.37254081205525325,
      "grad_norm": 0.47338756918907166,
      "learning_rate": 2e-05,
      "loss": 0.84,
      "step": 4450
    },
    {
      "epoch": 0.3733779824194223,
      "grad_norm": 0.4165925979614258,
      "learning_rate": 2e-05,
      "loss": 0.8103,
      "step": 4460
    },
    {
      "epoch": 0.37421515278359146,
      "grad_norm": 0.41632720828056335,
      "learning_rate": 2e-05,
      "loss": 0.833,
      "step": 4470
    },
    {
      "epoch": 0.37505232314776055,
      "grad_norm": 0.4042990803718567,
      "learning_rate": 2e-05,
      "loss": 0.8246,
      "step": 4480
    },
    {
      "epoch": 0.3758894935119297,
      "grad_norm": 0.4438692033290863,
      "learning_rate": 2e-05,
      "loss": 0.8333,
      "step": 4490
    },
    {
      "epoch": 0.37672666387609877,
      "grad_norm": 0.4200449287891388,
      "learning_rate": 2e-05,
      "loss": 0.8242,
      "step": 4500
    },
    {
      "epoch": 0.3775638342402679,
      "grad_norm": 0.4096335470676422,
      "learning_rate": 2e-05,
      "loss": 0.8156,
      "step": 4510
    },
    {
      "epoch": 0.378401004604437,
      "grad_norm": 0.44866660237312317,
      "learning_rate": 2e-05,
      "loss": 0.8311,
      "step": 4520
    },
    {
      "epoch": 0.3792381749686061,
      "grad_norm": 0.4202701449394226,
      "learning_rate": 2e-05,
      "loss": 0.8331,
      "step": 4530
    },
    {
      "epoch": 0.3800753453327752,
      "grad_norm": 0.44000279903411865,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 4540
    },
    {
      "epoch": 0.38091251569694434,
      "grad_norm": 0.4213865399360657,
      "learning_rate": 2e-05,
      "loss": 0.8257,
      "step": 4550
    },
    {
      "epoch": 0.3817496860611134,
      "grad_norm": 0.4106435775756836,
      "learning_rate": 2e-05,
      "loss": 0.8158,
      "step": 4560
    },
    {
      "epoch": 0.38258685642528256,
      "grad_norm": 0.42261892557144165,
      "learning_rate": 2e-05,
      "loss": 0.8365,
      "step": 4570
    },
    {
      "epoch": 0.38342402678945164,
      "grad_norm": 0.42611584067344666,
      "learning_rate": 2e-05,
      "loss": 0.8441,
      "step": 4580
    },
    {
      "epoch": 0.3842611971536208,
      "grad_norm": 0.4328523874282837,
      "learning_rate": 2e-05,
      "loss": 0.8227,
      "step": 4590
    },
    {
      "epoch": 0.38509836751778986,
      "grad_norm": 0.4371832609176636,
      "learning_rate": 2e-05,
      "loss": 0.8143,
      "step": 4600
    },
    {
      "epoch": 0.385935537881959,
      "grad_norm": 0.44326528906822205,
      "learning_rate": 2e-05,
      "loss": 0.8204,
      "step": 4610
    },
    {
      "epoch": 0.3867727082461281,
      "grad_norm": 0.4146097004413605,
      "learning_rate": 2e-05,
      "loss": 0.8255,
      "step": 4620
    },
    {
      "epoch": 0.3876098786102972,
      "grad_norm": 0.4093981385231018,
      "learning_rate": 2e-05,
      "loss": 0.8407,
      "step": 4630
    },
    {
      "epoch": 0.3884470489744663,
      "grad_norm": 0.4069594740867615,
      "learning_rate": 2e-05,
      "loss": 0.8501,
      "step": 4640
    },
    {
      "epoch": 0.38928421933863544,
      "grad_norm": 0.42581525444984436,
      "learning_rate": 2e-05,
      "loss": 0.8113,
      "step": 4650
    },
    {
      "epoch": 0.3901213897028045,
      "grad_norm": 0.40370485186576843,
      "learning_rate": 2e-05,
      "loss": 0.8108,
      "step": 4660
    },
    {
      "epoch": 0.3909585600669736,
      "grad_norm": 0.42503565549850464,
      "learning_rate": 2e-05,
      "loss": 0.8327,
      "step": 4670
    },
    {
      "epoch": 0.39179573043114274,
      "grad_norm": 0.4641563296318054,
      "learning_rate": 2e-05,
      "loss": 0.8133,
      "step": 4680
    },
    {
      "epoch": 0.3926329007953118,
      "grad_norm": 0.4232136607170105,
      "learning_rate": 2e-05,
      "loss": 0.7952,
      "step": 4690
    },
    {
      "epoch": 0.39347007115948096,
      "grad_norm": 0.4193868935108185,
      "learning_rate": 2e-05,
      "loss": 0.8256,
      "step": 4700
    },
    {
      "epoch": 0.39430724152365004,
      "grad_norm": 0.43483099341392517,
      "learning_rate": 2e-05,
      "loss": 0.8425,
      "step": 4710
    },
    {
      "epoch": 0.3951444118878192,
      "grad_norm": 0.425460547208786,
      "learning_rate": 2e-05,
      "loss": 0.8098,
      "step": 4720
    },
    {
      "epoch": 0.39598158225198826,
      "grad_norm": 0.40391620993614197,
      "learning_rate": 2e-05,
      "loss": 0.835,
      "step": 4730
    },
    {
      "epoch": 0.3968187526161574,
      "grad_norm": 0.4231874346733093,
      "learning_rate": 2e-05,
      "loss": 0.818,
      "step": 4740
    },
    {
      "epoch": 0.3976559229803265,
      "grad_norm": 0.4210924804210663,
      "learning_rate": 2e-05,
      "loss": 0.8347,
      "step": 4750
    },
    {
      "epoch": 0.3984930933444956,
      "grad_norm": 0.39859437942504883,
      "learning_rate": 2e-05,
      "loss": 0.8397,
      "step": 4760
    },
    {
      "epoch": 0.3993302637086647,
      "grad_norm": 0.41163817048072815,
      "learning_rate": 2e-05,
      "loss": 0.8256,
      "step": 4770
    },
    {
      "epoch": 0.40016743407283384,
      "grad_norm": 0.4214293360710144,
      "learning_rate": 2e-05,
      "loss": 0.8478,
      "step": 4780
    },
    {
      "epoch": 0.4010046044370029,
      "grad_norm": 0.40330448746681213,
      "learning_rate": 2e-05,
      "loss": 0.8337,
      "step": 4790
    },
    {
      "epoch": 0.40184177480117206,
      "grad_norm": 0.4489496350288391,
      "learning_rate": 2e-05,
      "loss": 0.8351,
      "step": 4800
    },
    {
      "epoch": 0.40267894516534114,
      "grad_norm": 0.4314161241054535,
      "learning_rate": 2e-05,
      "loss": 0.801,
      "step": 4810
    },
    {
      "epoch": 0.4035161155295103,
      "grad_norm": 0.4279036223888397,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 4820
    },
    {
      "epoch": 0.40435328589367936,
      "grad_norm": 0.41456297039985657,
      "learning_rate": 2e-05,
      "loss": 0.8196,
      "step": 4830
    },
    {
      "epoch": 0.4051904562578485,
      "grad_norm": 0.41446003317832947,
      "learning_rate": 2e-05,
      "loss": 0.837,
      "step": 4840
    },
    {
      "epoch": 0.4060276266220176,
      "grad_norm": 0.428045392036438,
      "learning_rate": 2e-05,
      "loss": 0.8272,
      "step": 4850
    },
    {
      "epoch": 0.4068647969861867,
      "grad_norm": 0.42162343859672546,
      "learning_rate": 2e-05,
      "loss": 0.8537,
      "step": 4860
    },
    {
      "epoch": 0.4077019673503558,
      "grad_norm": 0.4241536259651184,
      "learning_rate": 2e-05,
      "loss": 0.8161,
      "step": 4870
    },
    {
      "epoch": 0.4085391377145249,
      "grad_norm": 0.42472705245018005,
      "learning_rate": 2e-05,
      "loss": 0.8267,
      "step": 4880
    },
    {
      "epoch": 0.409376308078694,
      "grad_norm": 0.415218710899353,
      "learning_rate": 2e-05,
      "loss": 0.844,
      "step": 4890
    },
    {
      "epoch": 0.4102134784428631,
      "grad_norm": 0.432001531124115,
      "learning_rate": 2e-05,
      "loss": 0.8059,
      "step": 4900
    },
    {
      "epoch": 0.41105064880703224,
      "grad_norm": 0.41773557662963867,
      "learning_rate": 2e-05,
      "loss": 0.818,
      "step": 4910
    },
    {
      "epoch": 0.4118878191712013,
      "grad_norm": 0.40862348675727844,
      "learning_rate": 2e-05,
      "loss": 0.8217,
      "step": 4920
    },
    {
      "epoch": 0.41272498953537046,
      "grad_norm": 0.4347500205039978,
      "learning_rate": 2e-05,
      "loss": 0.8666,
      "step": 4930
    },
    {
      "epoch": 0.41356215989953954,
      "grad_norm": 0.4370155930519104,
      "learning_rate": 2e-05,
      "loss": 0.8303,
      "step": 4940
    },
    {
      "epoch": 0.4143993302637087,
      "grad_norm": 0.42120909690856934,
      "learning_rate": 2e-05,
      "loss": 0.8132,
      "step": 4950
    },
    {
      "epoch": 0.41523650062787776,
      "grad_norm": 0.4310665428638458,
      "learning_rate": 2e-05,
      "loss": 0.8219,
      "step": 4960
    },
    {
      "epoch": 0.4160736709920469,
      "grad_norm": 0.4336986243724823,
      "learning_rate": 2e-05,
      "loss": 0.8019,
      "step": 4970
    },
    {
      "epoch": 0.416910841356216,
      "grad_norm": 0.4416171908378601,
      "learning_rate": 2e-05,
      "loss": 0.8289,
      "step": 4980
    },
    {
      "epoch": 0.4177480117203851,
      "grad_norm": 0.41987624764442444,
      "learning_rate": 2e-05,
      "loss": 0.8182,
      "step": 4990
    },
    {
      "epoch": 0.4185851820845542,
      "grad_norm": 0.43798530101776123,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 5000
    },
    {
      "epoch": 0.41942235244872333,
      "grad_norm": 0.4883551001548767,
      "learning_rate": 2e-05,
      "loss": 0.8616,
      "step": 5010
    },
    {
      "epoch": 0.4202595228128924,
      "grad_norm": 0.46935969591140747,
      "learning_rate": 2e-05,
      "loss": 0.8396,
      "step": 5020
    },
    {
      "epoch": 0.42109669317706155,
      "grad_norm": 0.4080098569393158,
      "learning_rate": 2e-05,
      "loss": 0.8062,
      "step": 5030
    },
    {
      "epoch": 0.42193386354123064,
      "grad_norm": 0.4224624037742615,
      "learning_rate": 2e-05,
      "loss": 0.7853,
      "step": 5040
    },
    {
      "epoch": 0.4227710339053998,
      "grad_norm": 0.42404913902282715,
      "learning_rate": 2e-05,
      "loss": 0.8284,
      "step": 5050
    },
    {
      "epoch": 0.42360820426956886,
      "grad_norm": 0.423594206571579,
      "learning_rate": 2e-05,
      "loss": 0.8183,
      "step": 5060
    },
    {
      "epoch": 0.424445374633738,
      "grad_norm": 0.43580183386802673,
      "learning_rate": 2e-05,
      "loss": 0.8207,
      "step": 5070
    },
    {
      "epoch": 0.4252825449979071,
      "grad_norm": 0.4515228867530823,
      "learning_rate": 2e-05,
      "loss": 0.8067,
      "step": 5080
    },
    {
      "epoch": 0.42611971536207616,
      "grad_norm": 0.43099066615104675,
      "learning_rate": 2e-05,
      "loss": 0.8382,
      "step": 5090
    },
    {
      "epoch": 0.4269568857262453,
      "grad_norm": 0.425804078578949,
      "learning_rate": 2e-05,
      "loss": 0.8264,
      "step": 5100
    },
    {
      "epoch": 0.4277940560904144,
      "grad_norm": 0.42808157205581665,
      "learning_rate": 2e-05,
      "loss": 0.8116,
      "step": 5110
    },
    {
      "epoch": 0.4286312264545835,
      "grad_norm": 0.43376073241233826,
      "learning_rate": 2e-05,
      "loss": 0.8354,
      "step": 5120
    },
    {
      "epoch": 0.4294683968187526,
      "grad_norm": 0.4108678996562958,
      "learning_rate": 2e-05,
      "loss": 0.8289,
      "step": 5130
    },
    {
      "epoch": 0.43030556718292173,
      "grad_norm": 0.41111838817596436,
      "learning_rate": 2e-05,
      "loss": 0.8209,
      "step": 5140
    },
    {
      "epoch": 0.4311427375470908,
      "grad_norm": 0.4257042407989502,
      "learning_rate": 2e-05,
      "loss": 0.8386,
      "step": 5150
    },
    {
      "epoch": 0.43197990791125995,
      "grad_norm": 0.4200092554092407,
      "learning_rate": 2e-05,
      "loss": 0.8171,
      "step": 5160
    },
    {
      "epoch": 0.43281707827542903,
      "grad_norm": 0.41734182834625244,
      "learning_rate": 2e-05,
      "loss": 0.8205,
      "step": 5170
    },
    {
      "epoch": 0.43365424863959817,
      "grad_norm": 0.4185524582862854,
      "learning_rate": 2e-05,
      "loss": 0.8096,
      "step": 5180
    },
    {
      "epoch": 0.43449141900376725,
      "grad_norm": 0.41878604888916016,
      "learning_rate": 2e-05,
      "loss": 0.8429,
      "step": 5190
    },
    {
      "epoch": 0.4353285893679364,
      "grad_norm": 0.4180535674095154,
      "learning_rate": 2e-05,
      "loss": 0.8087,
      "step": 5200
    },
    {
      "epoch": 0.4361657597321055,
      "grad_norm": 0.4270792603492737,
      "learning_rate": 2e-05,
      "loss": 0.8325,
      "step": 5210
    },
    {
      "epoch": 0.4370029300962746,
      "grad_norm": 0.41992783546447754,
      "learning_rate": 2e-05,
      "loss": 0.8016,
      "step": 5220
    },
    {
      "epoch": 0.4378401004604437,
      "grad_norm": 0.43400758504867554,
      "learning_rate": 2e-05,
      "loss": 0.8474,
      "step": 5230
    },
    {
      "epoch": 0.43867727082461283,
      "grad_norm": 0.42177486419677734,
      "learning_rate": 2e-05,
      "loss": 0.8167,
      "step": 5240
    },
    {
      "epoch": 0.4395144411887819,
      "grad_norm": 0.42733344435691833,
      "learning_rate": 2e-05,
      "loss": 0.8313,
      "step": 5250
    },
    {
      "epoch": 0.44035161155295105,
      "grad_norm": 0.41207048296928406,
      "learning_rate": 2e-05,
      "loss": 0.8239,
      "step": 5260
    },
    {
      "epoch": 0.44118878191712013,
      "grad_norm": 0.4162651598453522,
      "learning_rate": 2e-05,
      "loss": 0.8338,
      "step": 5270
    },
    {
      "epoch": 0.44202595228128927,
      "grad_norm": 0.4078151285648346,
      "learning_rate": 2e-05,
      "loss": 0.8138,
      "step": 5280
    },
    {
      "epoch": 0.44286312264545835,
      "grad_norm": 0.42526522278785706,
      "learning_rate": 2e-05,
      "loss": 0.844,
      "step": 5290
    },
    {
      "epoch": 0.44370029300962743,
      "grad_norm": 0.43256497383117676,
      "learning_rate": 2e-05,
      "loss": 0.83,
      "step": 5300
    },
    {
      "epoch": 0.44453746337379657,
      "grad_norm": 0.43939757347106934,
      "learning_rate": 2e-05,
      "loss": 0.8063,
      "step": 5310
    },
    {
      "epoch": 0.44537463373796565,
      "grad_norm": 0.40747320652008057,
      "learning_rate": 2e-05,
      "loss": 0.7945,
      "step": 5320
    },
    {
      "epoch": 0.4462118041021348,
      "grad_norm": 0.41533413529396057,
      "learning_rate": 2e-05,
      "loss": 0.8294,
      "step": 5330
    },
    {
      "epoch": 0.4470489744663039,
      "grad_norm": 0.42405378818511963,
      "learning_rate": 2e-05,
      "loss": 0.7995,
      "step": 5340
    },
    {
      "epoch": 0.447886144830473,
      "grad_norm": 0.43562015891075134,
      "learning_rate": 2e-05,
      "loss": 0.8209,
      "step": 5350
    },
    {
      "epoch": 0.4487233151946421,
      "grad_norm": 0.40605980157852173,
      "learning_rate": 2e-05,
      "loss": 0.8092,
      "step": 5360
    },
    {
      "epoch": 0.44956048555881123,
      "grad_norm": 0.4186358153820038,
      "learning_rate": 2e-05,
      "loss": 0.8253,
      "step": 5370
    },
    {
      "epoch": 0.4503976559229803,
      "grad_norm": 0.42532554268836975,
      "learning_rate": 2e-05,
      "loss": 0.8016,
      "step": 5380
    },
    {
      "epoch": 0.45123482628714945,
      "grad_norm": 0.40878286957740784,
      "learning_rate": 2e-05,
      "loss": 0.8117,
      "step": 5390
    },
    {
      "epoch": 0.45207199665131853,
      "grad_norm": 0.44343554973602295,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 5400
    },
    {
      "epoch": 0.45290916701548767,
      "grad_norm": 0.42760589718818665,
      "learning_rate": 2e-05,
      "loss": 0.7917,
      "step": 5410
    },
    {
      "epoch": 0.45374633737965675,
      "grad_norm": 0.4153249263763428,
      "learning_rate": 2e-05,
      "loss": 0.8132,
      "step": 5420
    },
    {
      "epoch": 0.4545835077438259,
      "grad_norm": 0.3924368619918823,
      "learning_rate": 2e-05,
      "loss": 0.8033,
      "step": 5430
    },
    {
      "epoch": 0.45542067810799497,
      "grad_norm": 0.4194936156272888,
      "learning_rate": 2e-05,
      "loss": 0.7979,
      "step": 5440
    },
    {
      "epoch": 0.4562578484721641,
      "grad_norm": 0.42243438959121704,
      "learning_rate": 2e-05,
      "loss": 0.8397,
      "step": 5450
    },
    {
      "epoch": 0.4570950188363332,
      "grad_norm": 0.4158000349998474,
      "learning_rate": 2e-05,
      "loss": 0.8431,
      "step": 5460
    },
    {
      "epoch": 0.4579321892005023,
      "grad_norm": 0.43687647581100464,
      "learning_rate": 2e-05,
      "loss": 0.8172,
      "step": 5470
    },
    {
      "epoch": 0.4587693595646714,
      "grad_norm": 0.4203634560108185,
      "learning_rate": 2e-05,
      "loss": 0.8396,
      "step": 5480
    },
    {
      "epoch": 0.45960652992884055,
      "grad_norm": 0.43301740288734436,
      "learning_rate": 2e-05,
      "loss": 0.8403,
      "step": 5490
    },
    {
      "epoch": 0.46044370029300963,
      "grad_norm": 0.42429298162460327,
      "learning_rate": 2e-05,
      "loss": 0.7979,
      "step": 5500
    },
    {
      "epoch": 0.4612808706571787,
      "grad_norm": 0.41749587655067444,
      "learning_rate": 2e-05,
      "loss": 0.83,
      "step": 5510
    },
    {
      "epoch": 0.46211804102134785,
      "grad_norm": 0.3986694812774658,
      "learning_rate": 2e-05,
      "loss": 0.8178,
      "step": 5520
    },
    {
      "epoch": 0.46295521138551693,
      "grad_norm": 0.4017690122127533,
      "learning_rate": 2e-05,
      "loss": 0.8268,
      "step": 5530
    },
    {
      "epoch": 0.46379238174968607,
      "grad_norm": 0.44958123564720154,
      "learning_rate": 2e-05,
      "loss": 0.8397,
      "step": 5540
    },
    {
      "epoch": 0.46462955211385515,
      "grad_norm": 0.4256006181240082,
      "learning_rate": 2e-05,
      "loss": 0.8299,
      "step": 5550
    },
    {
      "epoch": 0.4654667224780243,
      "grad_norm": 0.4168212115764618,
      "learning_rate": 2e-05,
      "loss": 0.7956,
      "step": 5560
    },
    {
      "epoch": 0.46630389284219337,
      "grad_norm": 0.4205619990825653,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 5570
    },
    {
      "epoch": 0.4671410632063625,
      "grad_norm": 0.4160117506980896,
      "learning_rate": 2e-05,
      "loss": 0.8125,
      "step": 5580
    },
    {
      "epoch": 0.4679782335705316,
      "grad_norm": 0.4162079989910126,
      "learning_rate": 2e-05,
      "loss": 0.7781,
      "step": 5590
    },
    {
      "epoch": 0.4688154039347007,
      "grad_norm": 0.40398508310317993,
      "learning_rate": 2e-05,
      "loss": 0.7982,
      "step": 5600
    },
    {
      "epoch": 0.4696525742988698,
      "grad_norm": 0.4162850081920624,
      "learning_rate": 2e-05,
      "loss": 0.8291,
      "step": 5610
    },
    {
      "epoch": 0.47048974466303894,
      "grad_norm": 0.4446522891521454,
      "learning_rate": 2e-05,
      "loss": 0.8002,
      "step": 5620
    },
    {
      "epoch": 0.471326915027208,
      "grad_norm": 0.41184085607528687,
      "learning_rate": 2e-05,
      "loss": 0.8078,
      "step": 5630
    },
    {
      "epoch": 0.47216408539137716,
      "grad_norm": 0.4306154251098633,
      "learning_rate": 2e-05,
      "loss": 0.7897,
      "step": 5640
    },
    {
      "epoch": 0.47300125575554625,
      "grad_norm": 0.42332980036735535,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 5650
    },
    {
      "epoch": 0.4738384261197154,
      "grad_norm": 0.3883364498615265,
      "learning_rate": 2e-05,
      "loss": 0.8148,
      "step": 5660
    },
    {
      "epoch": 0.47467559648388447,
      "grad_norm": 0.4140203595161438,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 5670
    },
    {
      "epoch": 0.4755127668480536,
      "grad_norm": 0.6120416522026062,
      "learning_rate": 2e-05,
      "loss": 0.8041,
      "step": 5680
    },
    {
      "epoch": 0.4763499372122227,
      "grad_norm": 0.44696035981178284,
      "learning_rate": 2e-05,
      "loss": 0.8165,
      "step": 5690
    },
    {
      "epoch": 0.4771871075763918,
      "grad_norm": 0.472190797328949,
      "learning_rate": 2e-05,
      "loss": 0.8217,
      "step": 5700
    },
    {
      "epoch": 0.4780242779405609,
      "grad_norm": 0.40033575892448425,
      "learning_rate": 2e-05,
      "loss": 0.8349,
      "step": 5710
    },
    {
      "epoch": 0.47886144830473,
      "grad_norm": 0.4053618311882019,
      "learning_rate": 2e-05,
      "loss": 0.8375,
      "step": 5720
    },
    {
      "epoch": 0.4796986186688991,
      "grad_norm": 0.4280753433704376,
      "learning_rate": 2e-05,
      "loss": 0.8714,
      "step": 5730
    },
    {
      "epoch": 0.4805357890330682,
      "grad_norm": 0.42038729786872864,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 5740
    },
    {
      "epoch": 0.48137295939723734,
      "grad_norm": 0.40925249457359314,
      "learning_rate": 2e-05,
      "loss": 0.8169,
      "step": 5750
    },
    {
      "epoch": 0.4822101297614064,
      "grad_norm": 0.4118926525115967,
      "learning_rate": 2e-05,
      "loss": 0.8068,
      "step": 5760
    },
    {
      "epoch": 0.48304730012557556,
      "grad_norm": 0.41810524463653564,
      "learning_rate": 2e-05,
      "loss": 0.7981,
      "step": 5770
    },
    {
      "epoch": 0.48388447048974464,
      "grad_norm": 0.43730852007865906,
      "learning_rate": 2e-05,
      "loss": 0.7928,
      "step": 5780
    },
    {
      "epoch": 0.4847216408539138,
      "grad_norm": 0.40214866399765015,
      "learning_rate": 2e-05,
      "loss": 0.8125,
      "step": 5790
    },
    {
      "epoch": 0.48555881121808286,
      "grad_norm": 0.4087003767490387,
      "learning_rate": 2e-05,
      "loss": 0.8442,
      "step": 5800
    },
    {
      "epoch": 0.486395981582252,
      "grad_norm": 0.4067695736885071,
      "learning_rate": 2e-05,
      "loss": 0.801,
      "step": 5810
    },
    {
      "epoch": 0.4872331519464211,
      "grad_norm": 0.43072715401649475,
      "learning_rate": 2e-05,
      "loss": 0.8243,
      "step": 5820
    },
    {
      "epoch": 0.4880703223105902,
      "grad_norm": 0.44007259607315063,
      "learning_rate": 2e-05,
      "loss": 0.8121,
      "step": 5830
    },
    {
      "epoch": 0.4889074926747593,
      "grad_norm": 0.40092843770980835,
      "learning_rate": 2e-05,
      "loss": 0.8157,
      "step": 5840
    },
    {
      "epoch": 0.48974466303892844,
      "grad_norm": 0.41950035095214844,
      "learning_rate": 2e-05,
      "loss": 0.8107,
      "step": 5850
    },
    {
      "epoch": 0.4905818334030975,
      "grad_norm": 0.3988892734050751,
      "learning_rate": 2e-05,
      "loss": 0.7991,
      "step": 5860
    },
    {
      "epoch": 0.49141900376726666,
      "grad_norm": 0.42034777998924255,
      "learning_rate": 2e-05,
      "loss": 0.8262,
      "step": 5870
    },
    {
      "epoch": 0.49225617413143574,
      "grad_norm": 0.4446219503879547,
      "learning_rate": 2e-05,
      "loss": 0.8039,
      "step": 5880
    },
    {
      "epoch": 0.4930933444956049,
      "grad_norm": 0.45052894949913025,
      "learning_rate": 2e-05,
      "loss": 0.8248,
      "step": 5890
    },
    {
      "epoch": 0.49393051485977396,
      "grad_norm": 0.39699801802635193,
      "learning_rate": 2e-05,
      "loss": 0.8078,
      "step": 5900
    },
    {
      "epoch": 0.4947676852239431,
      "grad_norm": 0.4173450767993927,
      "learning_rate": 2e-05,
      "loss": 0.8146,
      "step": 5910
    },
    {
      "epoch": 0.4956048555881122,
      "grad_norm": 0.44117727875709534,
      "learning_rate": 2e-05,
      "loss": 0.806,
      "step": 5920
    },
    {
      "epoch": 0.49644202595228126,
      "grad_norm": 0.4214985966682434,
      "learning_rate": 2e-05,
      "loss": 0.8128,
      "step": 5930
    },
    {
      "epoch": 0.4972791963164504,
      "grad_norm": 0.4070131778717041,
      "learning_rate": 2e-05,
      "loss": 0.8167,
      "step": 5940
    },
    {
      "epoch": 0.4981163666806195,
      "grad_norm": 0.4527878761291504,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 5950
    },
    {
      "epoch": 0.4989535370447886,
      "grad_norm": 0.39434418082237244,
      "learning_rate": 2e-05,
      "loss": 0.8102,
      "step": 5960
    },
    {
      "epoch": 0.4997907074089577,
      "grad_norm": 0.4257313013076782,
      "learning_rate": 2e-05,
      "loss": 0.8143,
      "step": 5970
    },
    {
      "epoch": 0.5006278777731268,
      "grad_norm": 0.42737051844596863,
      "learning_rate": 2e-05,
      "loss": 0.7991,
      "step": 5980
    },
    {
      "epoch": 0.501465048137296,
      "grad_norm": 0.4357013404369354,
      "learning_rate": 2e-05,
      "loss": 0.8271,
      "step": 5990
    },
    {
      "epoch": 0.5023022185014651,
      "grad_norm": 0.45830410718917847,
      "learning_rate": 2e-05,
      "loss": 0.8155,
      "step": 6000
    },
    {
      "epoch": 0.5031393888656341,
      "grad_norm": 0.4182405173778534,
      "learning_rate": 2e-05,
      "loss": 0.8025,
      "step": 6010
    },
    {
      "epoch": 0.5039765592298032,
      "grad_norm": 0.4107055962085724,
      "learning_rate": 2e-05,
      "loss": 0.7942,
      "step": 6020
    },
    {
      "epoch": 0.5048137295939724,
      "grad_norm": 0.43023812770843506,
      "learning_rate": 2e-05,
      "loss": 0.8076,
      "step": 6030
    },
    {
      "epoch": 0.5056508999581415,
      "grad_norm": 0.42324554920196533,
      "learning_rate": 2e-05,
      "loss": 0.7952,
      "step": 6040
    },
    {
      "epoch": 0.5064880703223106,
      "grad_norm": 0.4048148989677429,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 6050
    },
    {
      "epoch": 0.5073252406864797,
      "grad_norm": 0.41163137555122375,
      "learning_rate": 2e-05,
      "loss": 0.8276,
      "step": 6060
    },
    {
      "epoch": 0.5081624110506489,
      "grad_norm": 0.4342605173587799,
      "learning_rate": 2e-05,
      "loss": 0.8152,
      "step": 6070
    },
    {
      "epoch": 0.5089995814148179,
      "grad_norm": 0.4276532828807831,
      "learning_rate": 2e-05,
      "loss": 0.8274,
      "step": 6080
    },
    {
      "epoch": 0.509836751778987,
      "grad_norm": 0.41123196482658386,
      "learning_rate": 2e-05,
      "loss": 0.8113,
      "step": 6090
    },
    {
      "epoch": 0.5106739221431561,
      "grad_norm": 0.43460285663604736,
      "learning_rate": 2e-05,
      "loss": 0.8093,
      "step": 6100
    },
    {
      "epoch": 0.5115110925073253,
      "grad_norm": 0.40841037034988403,
      "learning_rate": 2e-05,
      "loss": 0.8049,
      "step": 6110
    },
    {
      "epoch": 0.5123482628714944,
      "grad_norm": 0.4815380871295929,
      "learning_rate": 2e-05,
      "loss": 0.7862,
      "step": 6120
    },
    {
      "epoch": 0.5131854332356635,
      "grad_norm": 0.43933025002479553,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 6130
    },
    {
      "epoch": 0.5140226035998325,
      "grad_norm": 0.4382390081882477,
      "learning_rate": 2e-05,
      "loss": 0.7854,
      "step": 6140
    },
    {
      "epoch": 0.5148597739640016,
      "grad_norm": 0.41584649682044983,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 6150
    },
    {
      "epoch": 0.5156969443281708,
      "grad_norm": 0.3916720151901245,
      "learning_rate": 2e-05,
      "loss": 0.7897,
      "step": 6160
    },
    {
      "epoch": 0.5165341146923399,
      "grad_norm": 0.39625677466392517,
      "learning_rate": 2e-05,
      "loss": 0.7835,
      "step": 6170
    },
    {
      "epoch": 0.517371285056509,
      "grad_norm": 0.39596593379974365,
      "learning_rate": 2e-05,
      "loss": 0.8033,
      "step": 6180
    },
    {
      "epoch": 0.5182084554206781,
      "grad_norm": 0.4218946695327759,
      "learning_rate": 2e-05,
      "loss": 0.816,
      "step": 6190
    },
    {
      "epoch": 0.5190456257848473,
      "grad_norm": 0.42067819833755493,
      "learning_rate": 2e-05,
      "loss": 0.8155,
      "step": 6200
    },
    {
      "epoch": 0.5198827961490163,
      "grad_norm": 0.43086957931518555,
      "learning_rate": 2e-05,
      "loss": 0.8134,
      "step": 6210
    },
    {
      "epoch": 0.5207199665131854,
      "grad_norm": 0.4047655463218689,
      "learning_rate": 2e-05,
      "loss": 0.8141,
      "step": 6220
    },
    {
      "epoch": 0.5215571368773545,
      "grad_norm": 0.40903475880622864,
      "learning_rate": 2e-05,
      "loss": 0.7998,
      "step": 6230
    },
    {
      "epoch": 0.5223943072415237,
      "grad_norm": 0.4179118871688843,
      "learning_rate": 2e-05,
      "loss": 0.8045,
      "step": 6240
    },
    {
      "epoch": 0.5232314776056928,
      "grad_norm": 0.4107031226158142,
      "learning_rate": 2e-05,
      "loss": 0.802,
      "step": 6250
    },
    {
      "epoch": 0.5240686479698619,
      "grad_norm": 0.4047863483428955,
      "learning_rate": 2e-05,
      "loss": 0.8064,
      "step": 6260
    },
    {
      "epoch": 0.5249058183340309,
      "grad_norm": 0.40617135167121887,
      "learning_rate": 2e-05,
      "loss": 0.8308,
      "step": 6270
    },
    {
      "epoch": 0.5257429886982001,
      "grad_norm": 0.4011264145374298,
      "learning_rate": 2e-05,
      "loss": 0.8096,
      "step": 6280
    },
    {
      "epoch": 0.5265801590623692,
      "grad_norm": 0.4343249499797821,
      "learning_rate": 2e-05,
      "loss": 0.8365,
      "step": 6290
    },
    {
      "epoch": 0.5274173294265383,
      "grad_norm": 0.3885353207588196,
      "learning_rate": 2e-05,
      "loss": 0.7878,
      "step": 6300
    },
    {
      "epoch": 0.5282544997907074,
      "grad_norm": 0.42856672406196594,
      "learning_rate": 2e-05,
      "loss": 0.7963,
      "step": 6310
    },
    {
      "epoch": 0.5290916701548766,
      "grad_norm": 0.4048072397708893,
      "learning_rate": 2e-05,
      "loss": 0.8039,
      "step": 6320
    },
    {
      "epoch": 0.5299288405190457,
      "grad_norm": 0.4116542339324951,
      "learning_rate": 2e-05,
      "loss": 0.833,
      "step": 6330
    },
    {
      "epoch": 0.5307660108832147,
      "grad_norm": 0.41515734791755676,
      "learning_rate": 2e-05,
      "loss": 0.7868,
      "step": 6340
    },
    {
      "epoch": 0.5316031812473838,
      "grad_norm": 0.4509551227092743,
      "learning_rate": 2e-05,
      "loss": 0.8066,
      "step": 6350
    },
    {
      "epoch": 0.5324403516115529,
      "grad_norm": 0.41244998574256897,
      "learning_rate": 2e-05,
      "loss": 0.8047,
      "step": 6360
    },
    {
      "epoch": 0.5332775219757221,
      "grad_norm": 0.41589218378067017,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 6370
    },
    {
      "epoch": 0.5341146923398912,
      "grad_norm": 0.4403518736362457,
      "learning_rate": 2e-05,
      "loss": 0.8274,
      "step": 6380
    },
    {
      "epoch": 0.5349518627040603,
      "grad_norm": 0.4054332375526428,
      "learning_rate": 2e-05,
      "loss": 0.7944,
      "step": 6390
    },
    {
      "epoch": 0.5357890330682293,
      "grad_norm": 0.4953019320964813,
      "learning_rate": 2e-05,
      "loss": 0.8211,
      "step": 6400
    },
    {
      "epoch": 0.5366262034323985,
      "grad_norm": 0.42315933108329773,
      "learning_rate": 2e-05,
      "loss": 0.8015,
      "step": 6410
    },
    {
      "epoch": 0.5374633737965676,
      "grad_norm": 0.43524065613746643,
      "learning_rate": 2e-05,
      "loss": 0.7873,
      "step": 6420
    },
    {
      "epoch": 0.5383005441607367,
      "grad_norm": 0.4234393835067749,
      "learning_rate": 2e-05,
      "loss": 0.8054,
      "step": 6430
    },
    {
      "epoch": 0.5391377145249058,
      "grad_norm": 0.42887526750564575,
      "learning_rate": 2e-05,
      "loss": 0.7924,
      "step": 6440
    },
    {
      "epoch": 0.539974884889075,
      "grad_norm": 0.4137824475765228,
      "learning_rate": 2e-05,
      "loss": 0.8046,
      "step": 6450
    },
    {
      "epoch": 0.540812055253244,
      "grad_norm": 0.4071830213069916,
      "learning_rate": 2e-05,
      "loss": 0.7743,
      "step": 6460
    },
    {
      "epoch": 0.5416492256174131,
      "grad_norm": 0.3836536705493927,
      "learning_rate": 2e-05,
      "loss": 0.793,
      "step": 6470
    },
    {
      "epoch": 0.5424863959815822,
      "grad_norm": 0.4118112623691559,
      "learning_rate": 2e-05,
      "loss": 0.8322,
      "step": 6480
    },
    {
      "epoch": 0.5433235663457514,
      "grad_norm": 0.41501113772392273,
      "learning_rate": 2e-05,
      "loss": 0.8264,
      "step": 6490
    },
    {
      "epoch": 0.5441607367099205,
      "grad_norm": 0.42500951886177063,
      "learning_rate": 2e-05,
      "loss": 0.8021,
      "step": 6500
    },
    {
      "epoch": 0.5449979070740896,
      "grad_norm": 0.44062936305999756,
      "learning_rate": 2e-05,
      "loss": 0.7954,
      "step": 6510
    },
    {
      "epoch": 0.5458350774382587,
      "grad_norm": 0.42948347330093384,
      "learning_rate": 2e-05,
      "loss": 0.8005,
      "step": 6520
    },
    {
      "epoch": 0.5466722478024278,
      "grad_norm": 0.4122495651245117,
      "learning_rate": 2e-05,
      "loss": 0.8337,
      "step": 6530
    },
    {
      "epoch": 0.5475094181665969,
      "grad_norm": 0.43499791622161865,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 6540
    },
    {
      "epoch": 0.548346588530766,
      "grad_norm": 0.3832750916481018,
      "learning_rate": 2e-05,
      "loss": 0.8038,
      "step": 6550
    },
    {
      "epoch": 0.5491837588949351,
      "grad_norm": 0.40254923701286316,
      "learning_rate": 2e-05,
      "loss": 0.7885,
      "step": 6560
    },
    {
      "epoch": 0.5500209292591042,
      "grad_norm": 0.42910677194595337,
      "learning_rate": 2e-05,
      "loss": 0.8118,
      "step": 6570
    },
    {
      "epoch": 0.5508580996232734,
      "grad_norm": 0.4367128312587738,
      "learning_rate": 2e-05,
      "loss": 0.824,
      "step": 6580
    },
    {
      "epoch": 0.5516952699874424,
      "grad_norm": 0.4092158079147339,
      "learning_rate": 2e-05,
      "loss": 0.8083,
      "step": 6590
    },
    {
      "epoch": 0.5525324403516115,
      "grad_norm": 0.4118812382221222,
      "learning_rate": 2e-05,
      "loss": 0.7891,
      "step": 6600
    },
    {
      "epoch": 0.5533696107157806,
      "grad_norm": 0.41295745968818665,
      "learning_rate": 2e-05,
      "loss": 0.8079,
      "step": 6610
    },
    {
      "epoch": 0.5542067810799498,
      "grad_norm": 0.4511883854866028,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 6620
    },
    {
      "epoch": 0.5550439514441189,
      "grad_norm": 0.4457618296146393,
      "learning_rate": 2e-05,
      "loss": 0.8035,
      "step": 6630
    },
    {
      "epoch": 0.555881121808288,
      "grad_norm": 0.43562233448028564,
      "learning_rate": 2e-05,
      "loss": 0.8235,
      "step": 6640
    },
    {
      "epoch": 0.556718292172457,
      "grad_norm": 0.6008503437042236,
      "learning_rate": 2e-05,
      "loss": 0.8083,
      "step": 6650
    },
    {
      "epoch": 0.5575554625366262,
      "grad_norm": 0.5283923149108887,
      "learning_rate": 2e-05,
      "loss": 0.795,
      "step": 6660
    },
    {
      "epoch": 0.5583926329007953,
      "grad_norm": 0.41668012738227844,
      "learning_rate": 2e-05,
      "loss": 0.8122,
      "step": 6670
    },
    {
      "epoch": 0.5592298032649644,
      "grad_norm": 0.4084201455116272,
      "learning_rate": 2e-05,
      "loss": 0.8367,
      "step": 6680
    },
    {
      "epoch": 0.5600669736291335,
      "grad_norm": 0.40944334864616394,
      "learning_rate": 2e-05,
      "loss": 0.7965,
      "step": 6690
    },
    {
      "epoch": 0.5609041439933027,
      "grad_norm": 0.4102044999599457,
      "learning_rate": 2e-05,
      "loss": 0.8185,
      "step": 6700
    },
    {
      "epoch": 0.5617413143574718,
      "grad_norm": 0.4162530303001404,
      "learning_rate": 2e-05,
      "loss": 0.8015,
      "step": 6710
    },
    {
      "epoch": 0.5625784847216408,
      "grad_norm": 0.43799322843551636,
      "learning_rate": 2e-05,
      "loss": 0.7953,
      "step": 6720
    },
    {
      "epoch": 0.5634156550858099,
      "grad_norm": 0.4068587124347687,
      "learning_rate": 2e-05,
      "loss": 0.7875,
      "step": 6730
    },
    {
      "epoch": 0.5642528254499791,
      "grad_norm": 0.42114582657814026,
      "learning_rate": 2e-05,
      "loss": 0.8218,
      "step": 6740
    },
    {
      "epoch": 0.5650899958141482,
      "grad_norm": 0.44762396812438965,
      "learning_rate": 2e-05,
      "loss": 0.8096,
      "step": 6750
    },
    {
      "epoch": 0.5659271661783173,
      "grad_norm": 0.4178694784641266,
      "learning_rate": 2e-05,
      "loss": 0.7999,
      "step": 6760
    },
    {
      "epoch": 0.5667643365424864,
      "grad_norm": 0.4166998863220215,
      "learning_rate": 2e-05,
      "loss": 0.8066,
      "step": 6770
    },
    {
      "epoch": 0.5676015069066555,
      "grad_norm": 0.4012550115585327,
      "learning_rate": 2e-05,
      "loss": 0.8092,
      "step": 6780
    },
    {
      "epoch": 0.5684386772708246,
      "grad_norm": 2.6021780967712402,
      "learning_rate": 2e-05,
      "loss": 0.8136,
      "step": 6790
    },
    {
      "epoch": 0.5692758476349937,
      "grad_norm": 0.41423115134239197,
      "learning_rate": 2e-05,
      "loss": 0.8005,
      "step": 6800
    },
    {
      "epoch": 0.5701130179991628,
      "grad_norm": 0.41177746653556824,
      "learning_rate": 2e-05,
      "loss": 0.8011,
      "step": 6810
    },
    {
      "epoch": 0.5709501883633319,
      "grad_norm": 0.4305766224861145,
      "learning_rate": 2e-05,
      "loss": 0.8116,
      "step": 6820
    },
    {
      "epoch": 0.5717873587275011,
      "grad_norm": 0.4768018424510956,
      "learning_rate": 2e-05,
      "loss": 0.8384,
      "step": 6830
    },
    {
      "epoch": 0.5726245290916702,
      "grad_norm": 0.4288696050643921,
      "learning_rate": 2e-05,
      "loss": 0.8179,
      "step": 6840
    },
    {
      "epoch": 0.5734616994558392,
      "grad_norm": 0.41483551263809204,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 6850
    },
    {
      "epoch": 0.5742988698200083,
      "grad_norm": 0.42573094367980957,
      "learning_rate": 2e-05,
      "loss": 0.828,
      "step": 6860
    },
    {
      "epoch": 0.5751360401841775,
      "grad_norm": 0.4145154654979706,
      "learning_rate": 2e-05,
      "loss": 0.8442,
      "step": 6870
    },
    {
      "epoch": 0.5759732105483466,
      "grad_norm": 0.5237193703651428,
      "learning_rate": 2e-05,
      "loss": 0.8019,
      "step": 6880
    },
    {
      "epoch": 0.5768103809125157,
      "grad_norm": 0.41672471165657043,
      "learning_rate": 2e-05,
      "loss": 0.7891,
      "step": 6890
    },
    {
      "epoch": 0.5776475512766848,
      "grad_norm": 0.42464450001716614,
      "learning_rate": 2e-05,
      "loss": 0.826,
      "step": 6900
    },
    {
      "epoch": 0.578484721640854,
      "grad_norm": 0.41314971446990967,
      "learning_rate": 2e-05,
      "loss": 0.7757,
      "step": 6910
    },
    {
      "epoch": 0.579321892005023,
      "grad_norm": 0.42597058415412903,
      "learning_rate": 2e-05,
      "loss": 0.8363,
      "step": 6920
    },
    {
      "epoch": 0.5801590623691921,
      "grad_norm": 0.4074229300022125,
      "learning_rate": 2e-05,
      "loss": 0.7959,
      "step": 6930
    },
    {
      "epoch": 0.5809962327333612,
      "grad_norm": 0.42489832639694214,
      "learning_rate": 2e-05,
      "loss": 0.8083,
      "step": 6940
    },
    {
      "epoch": 0.5818334030975304,
      "grad_norm": 0.40630918741226196,
      "learning_rate": 2e-05,
      "loss": 0.8113,
      "step": 6950
    },
    {
      "epoch": 0.5826705734616995,
      "grad_norm": 0.4240066707134247,
      "learning_rate": 2e-05,
      "loss": 0.806,
      "step": 6960
    },
    {
      "epoch": 0.5835077438258686,
      "grad_norm": 0.374102383852005,
      "learning_rate": 2e-05,
      "loss": 0.7866,
      "step": 6970
    },
    {
      "epoch": 0.5843449141900376,
      "grad_norm": 0.40525341033935547,
      "learning_rate": 2e-05,
      "loss": 0.7824,
      "step": 6980
    },
    {
      "epoch": 0.5851820845542067,
      "grad_norm": 0.4199190139770508,
      "learning_rate": 2e-05,
      "loss": 0.7839,
      "step": 6990
    },
    {
      "epoch": 0.5860192549183759,
      "grad_norm": 0.39600563049316406,
      "learning_rate": 2e-05,
      "loss": 0.822,
      "step": 7000
    },
    {
      "epoch": 0.586856425282545,
      "grad_norm": 0.3901606798171997,
      "learning_rate": 2e-05,
      "loss": 0.7922,
      "step": 7010
    },
    {
      "epoch": 0.5876935956467141,
      "grad_norm": 0.40760451555252075,
      "learning_rate": 2e-05,
      "loss": 0.8163,
      "step": 7020
    },
    {
      "epoch": 0.5885307660108832,
      "grad_norm": 0.413845956325531,
      "learning_rate": 2e-05,
      "loss": 0.793,
      "step": 7030
    },
    {
      "epoch": 0.5893679363750524,
      "grad_norm": 0.42291495203971863,
      "learning_rate": 2e-05,
      "loss": 0.7963,
      "step": 7040
    },
    {
      "epoch": 0.5902051067392214,
      "grad_norm": 0.3969959020614624,
      "learning_rate": 2e-05,
      "loss": 0.8195,
      "step": 7050
    },
    {
      "epoch": 0.5910422771033905,
      "grad_norm": 0.4352134168148041,
      "learning_rate": 2e-05,
      "loss": 0.8115,
      "step": 7060
    },
    {
      "epoch": 0.5918794474675596,
      "grad_norm": 0.41534343361854553,
      "learning_rate": 2e-05,
      "loss": 0.7952,
      "step": 7070
    },
    {
      "epoch": 0.5927166178317288,
      "grad_norm": 0.40757063031196594,
      "learning_rate": 2e-05,
      "loss": 0.7668,
      "step": 7080
    },
    {
      "epoch": 0.5935537881958979,
      "grad_norm": 0.40818482637405396,
      "learning_rate": 2e-05,
      "loss": 0.7866,
      "step": 7090
    },
    {
      "epoch": 0.594390958560067,
      "grad_norm": 0.40475544333457947,
      "learning_rate": 2e-05,
      "loss": 0.7882,
      "step": 7100
    },
    {
      "epoch": 0.595228128924236,
      "grad_norm": 0.41150614619255066,
      "learning_rate": 2e-05,
      "loss": 0.798,
      "step": 7110
    },
    {
      "epoch": 0.5960652992884052,
      "grad_norm": 0.4296150207519531,
      "learning_rate": 2e-05,
      "loss": 0.8161,
      "step": 7120
    },
    {
      "epoch": 0.5969024696525743,
      "grad_norm": 0.40952855348587036,
      "learning_rate": 2e-05,
      "loss": 0.7862,
      "step": 7130
    },
    {
      "epoch": 0.5977396400167434,
      "grad_norm": 0.44159936904907227,
      "learning_rate": 2e-05,
      "loss": 0.7977,
      "step": 7140
    },
    {
      "epoch": 0.5985768103809125,
      "grad_norm": 0.42437484860420227,
      "learning_rate": 2e-05,
      "loss": 0.8355,
      "step": 7150
    },
    {
      "epoch": 0.5994139807450817,
      "grad_norm": 0.4003716707229614,
      "learning_rate": 2e-05,
      "loss": 0.8105,
      "step": 7160
    },
    {
      "epoch": 0.6002511511092508,
      "grad_norm": 0.4303557872772217,
      "learning_rate": 2e-05,
      "loss": 0.809,
      "step": 7170
    },
    {
      "epoch": 0.6010883214734198,
      "grad_norm": 0.41689321398735046,
      "learning_rate": 2e-05,
      "loss": 0.8229,
      "step": 7180
    },
    {
      "epoch": 0.6019254918375889,
      "grad_norm": 0.404521644115448,
      "learning_rate": 2e-05,
      "loss": 0.8223,
      "step": 7190
    },
    {
      "epoch": 0.602762662201758,
      "grad_norm": 0.42582377791404724,
      "learning_rate": 2e-05,
      "loss": 0.8082,
      "step": 7200
    },
    {
      "epoch": 0.6035998325659272,
      "grad_norm": 0.4071408808231354,
      "learning_rate": 2e-05,
      "loss": 0.8004,
      "step": 7210
    },
    {
      "epoch": 0.6044370029300963,
      "grad_norm": 0.45669516921043396,
      "learning_rate": 2e-05,
      "loss": 0.779,
      "step": 7220
    },
    {
      "epoch": 0.6052741732942654,
      "grad_norm": 0.4022812247276306,
      "learning_rate": 2e-05,
      "loss": 0.7754,
      "step": 7230
    },
    {
      "epoch": 0.6061113436584344,
      "grad_norm": 0.41805559396743774,
      "learning_rate": 2e-05,
      "loss": 0.819,
      "step": 7240
    },
    {
      "epoch": 0.6069485140226036,
      "grad_norm": 0.4173201322555542,
      "learning_rate": 2e-05,
      "loss": 0.7788,
      "step": 7250
    },
    {
      "epoch": 0.6077856843867727,
      "grad_norm": 0.41724205017089844,
      "learning_rate": 2e-05,
      "loss": 0.7994,
      "step": 7260
    },
    {
      "epoch": 0.6086228547509418,
      "grad_norm": 0.42887189984321594,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 7270
    },
    {
      "epoch": 0.6094600251151109,
      "grad_norm": 0.3973146378993988,
      "learning_rate": 2e-05,
      "loss": 0.8139,
      "step": 7280
    },
    {
      "epoch": 0.6102971954792801,
      "grad_norm": 0.41664618253707886,
      "learning_rate": 2e-05,
      "loss": 0.8098,
      "step": 7290
    },
    {
      "epoch": 0.6111343658434492,
      "grad_norm": 0.4121076166629791,
      "learning_rate": 2e-05,
      "loss": 0.807,
      "step": 7300
    },
    {
      "epoch": 0.6119715362076182,
      "grad_norm": 0.40787574648857117,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 7310
    },
    {
      "epoch": 0.6128087065717873,
      "grad_norm": 0.4021184742450714,
      "learning_rate": 2e-05,
      "loss": 0.8009,
      "step": 7320
    },
    {
      "epoch": 0.6136458769359565,
      "grad_norm": 0.41523832082748413,
      "learning_rate": 2e-05,
      "loss": 0.7964,
      "step": 7330
    },
    {
      "epoch": 0.6144830473001256,
      "grad_norm": 0.4011082649230957,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 7340
    },
    {
      "epoch": 0.6153202176642947,
      "grad_norm": 0.41723012924194336,
      "learning_rate": 2e-05,
      "loss": 0.8285,
      "step": 7350
    },
    {
      "epoch": 0.6161573880284638,
      "grad_norm": 0.40085941553115845,
      "learning_rate": 2e-05,
      "loss": 0.8132,
      "step": 7360
    },
    {
      "epoch": 0.616994558392633,
      "grad_norm": 0.3842713236808777,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 7370
    },
    {
      "epoch": 0.617831728756802,
      "grad_norm": 0.40647852420806885,
      "learning_rate": 2e-05,
      "loss": 0.8241,
      "step": 7380
    },
    {
      "epoch": 0.6186688991209711,
      "grad_norm": 0.4019443988800049,
      "learning_rate": 2e-05,
      "loss": 0.786,
      "step": 7390
    },
    {
      "epoch": 0.6195060694851402,
      "grad_norm": 0.42185693979263306,
      "learning_rate": 2e-05,
      "loss": 0.8045,
      "step": 7400
    },
    {
      "epoch": 0.6203432398493093,
      "grad_norm": 0.42037999629974365,
      "learning_rate": 2e-05,
      "loss": 0.8103,
      "step": 7410
    },
    {
      "epoch": 0.6211804102134785,
      "grad_norm": 0.45487967133522034,
      "learning_rate": 2e-05,
      "loss": 0.791,
      "step": 7420
    },
    {
      "epoch": 0.6220175805776476,
      "grad_norm": 0.4355306923389435,
      "learning_rate": 2e-05,
      "loss": 0.7818,
      "step": 7430
    },
    {
      "epoch": 0.6228547509418166,
      "grad_norm": 0.4217500686645508,
      "learning_rate": 2e-05,
      "loss": 0.8281,
      "step": 7440
    },
    {
      "epoch": 0.6236919213059857,
      "grad_norm": 0.4521215260028839,
      "learning_rate": 2e-05,
      "loss": 0.8055,
      "step": 7450
    },
    {
      "epoch": 0.6245290916701549,
      "grad_norm": 0.4080929458141327,
      "learning_rate": 2e-05,
      "loss": 0.8109,
      "step": 7460
    },
    {
      "epoch": 0.625366262034324,
      "grad_norm": 0.4102347195148468,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 7470
    },
    {
      "epoch": 0.6262034323984931,
      "grad_norm": 0.4006361663341522,
      "learning_rate": 2e-05,
      "loss": 0.7886,
      "step": 7480
    },
    {
      "epoch": 0.6270406027626622,
      "grad_norm": 0.41682279109954834,
      "learning_rate": 2e-05,
      "loss": 0.796,
      "step": 7490
    },
    {
      "epoch": 0.6278777731268314,
      "grad_norm": 0.4106208384037018,
      "learning_rate": 2e-05,
      "loss": 0.7979,
      "step": 7500
    },
    {
      "epoch": 0.6287149434910004,
      "grad_norm": 0.4006294906139374,
      "learning_rate": 2e-05,
      "loss": 0.8129,
      "step": 7510
    },
    {
      "epoch": 0.6295521138551695,
      "grad_norm": 0.4153207242488861,
      "learning_rate": 2e-05,
      "loss": 0.7866,
      "step": 7520
    },
    {
      "epoch": 0.6303892842193386,
      "grad_norm": 0.4397103190422058,
      "learning_rate": 2e-05,
      "loss": 0.7829,
      "step": 7530
    },
    {
      "epoch": 0.6312264545835078,
      "grad_norm": 0.41311487555503845,
      "learning_rate": 2e-05,
      "loss": 0.8011,
      "step": 7540
    },
    {
      "epoch": 0.6320636249476769,
      "grad_norm": 0.4098183214664459,
      "learning_rate": 2e-05,
      "loss": 0.7776,
      "step": 7550
    },
    {
      "epoch": 0.632900795311846,
      "grad_norm": 0.4068772792816162,
      "learning_rate": 2e-05,
      "loss": 0.8068,
      "step": 7560
    },
    {
      "epoch": 0.633737965676015,
      "grad_norm": 0.44142770767211914,
      "learning_rate": 2e-05,
      "loss": 0.8158,
      "step": 7570
    },
    {
      "epoch": 0.6345751360401842,
      "grad_norm": 0.4156513810157776,
      "learning_rate": 2e-05,
      "loss": 0.792,
      "step": 7580
    },
    {
      "epoch": 0.6354123064043533,
      "grad_norm": 0.43381622433662415,
      "learning_rate": 2e-05,
      "loss": 0.8009,
      "step": 7590
    },
    {
      "epoch": 0.6362494767685224,
      "grad_norm": 0.43044334650039673,
      "learning_rate": 2e-05,
      "loss": 0.8247,
      "step": 7600
    },
    {
      "epoch": 0.6370866471326915,
      "grad_norm": 0.3877158463001251,
      "learning_rate": 2e-05,
      "loss": 0.7774,
      "step": 7610
    },
    {
      "epoch": 0.6379238174968606,
      "grad_norm": 0.42374610900878906,
      "learning_rate": 2e-05,
      "loss": 0.7879,
      "step": 7620
    },
    {
      "epoch": 0.6387609878610297,
      "grad_norm": 0.43404421210289,
      "learning_rate": 2e-05,
      "loss": 0.8072,
      "step": 7630
    },
    {
      "epoch": 0.6395981582251988,
      "grad_norm": 0.4450354278087616,
      "learning_rate": 2e-05,
      "loss": 0.8266,
      "step": 7640
    },
    {
      "epoch": 0.6404353285893679,
      "grad_norm": 0.41809383034706116,
      "learning_rate": 2e-05,
      "loss": 0.8094,
      "step": 7650
    },
    {
      "epoch": 0.641272498953537,
      "grad_norm": 0.41986146569252014,
      "learning_rate": 2e-05,
      "loss": 0.8101,
      "step": 7660
    },
    {
      "epoch": 0.6421096693177062,
      "grad_norm": 0.452059268951416,
      "learning_rate": 2e-05,
      "loss": 0.7946,
      "step": 7670
    },
    {
      "epoch": 0.6429468396818753,
      "grad_norm": 2.523205041885376,
      "learning_rate": 2e-05,
      "loss": 0.8418,
      "step": 7680
    },
    {
      "epoch": 0.6437840100460444,
      "grad_norm": 0.6466136574745178,
      "learning_rate": 2e-05,
      "loss": 0.7997,
      "step": 7690
    },
    {
      "epoch": 0.6446211804102134,
      "grad_norm": 0.3952144384384155,
      "learning_rate": 2e-05,
      "loss": 0.7873,
      "step": 7700
    },
    {
      "epoch": 0.6454583507743826,
      "grad_norm": 0.4083433747291565,
      "learning_rate": 2e-05,
      "loss": 0.7739,
      "step": 7710
    },
    {
      "epoch": 0.6462955211385517,
      "grad_norm": 0.4122130870819092,
      "learning_rate": 2e-05,
      "loss": 0.8083,
      "step": 7720
    },
    {
      "epoch": 0.6471326915027208,
      "grad_norm": 0.41584187746047974,
      "learning_rate": 2e-05,
      "loss": 0.8047,
      "step": 7730
    },
    {
      "epoch": 0.6479698618668899,
      "grad_norm": 0.4358825981616974,
      "learning_rate": 2e-05,
      "loss": 0.7837,
      "step": 7740
    },
    {
      "epoch": 0.6488070322310591,
      "grad_norm": 0.45423972606658936,
      "learning_rate": 2e-05,
      "loss": 0.8341,
      "step": 7750
    },
    {
      "epoch": 0.6496442025952281,
      "grad_norm": 0.4200233817100525,
      "learning_rate": 2e-05,
      "loss": 0.805,
      "step": 7760
    },
    {
      "epoch": 0.6504813729593972,
      "grad_norm": 0.43946215510368347,
      "learning_rate": 2e-05,
      "loss": 0.8104,
      "step": 7770
    },
    {
      "epoch": 0.6513185433235663,
      "grad_norm": 0.4175316393375397,
      "learning_rate": 2e-05,
      "loss": 0.7986,
      "step": 7780
    },
    {
      "epoch": 0.6521557136877355,
      "grad_norm": 0.4062187671661377,
      "learning_rate": 2e-05,
      "loss": 0.7893,
      "step": 7790
    },
    {
      "epoch": 0.6529928840519046,
      "grad_norm": 0.42061999440193176,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 7800
    },
    {
      "epoch": 0.6538300544160737,
      "grad_norm": 0.41082292795181274,
      "learning_rate": 2e-05,
      "loss": 0.7982,
      "step": 7810
    },
    {
      "epoch": 0.6546672247802428,
      "grad_norm": 0.4375016391277313,
      "learning_rate": 2e-05,
      "loss": 0.7854,
      "step": 7820
    },
    {
      "epoch": 0.6555043951444118,
      "grad_norm": 0.3991371989250183,
      "learning_rate": 2e-05,
      "loss": 0.7819,
      "step": 7830
    },
    {
      "epoch": 0.656341565508581,
      "grad_norm": 0.4160153567790985,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 7840
    },
    {
      "epoch": 0.6571787358727501,
      "grad_norm": 0.4237268567085266,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 7850
    },
    {
      "epoch": 0.6580159062369192,
      "grad_norm": 0.39804473519325256,
      "learning_rate": 2e-05,
      "loss": 0.738,
      "step": 7860
    },
    {
      "epoch": 0.6588530766010883,
      "grad_norm": 0.41910815238952637,
      "learning_rate": 2e-05,
      "loss": 0.8057,
      "step": 7870
    },
    {
      "epoch": 0.6596902469652575,
      "grad_norm": 0.40552714467048645,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 7880
    },
    {
      "epoch": 0.6605274173294265,
      "grad_norm": 0.4109598696231842,
      "learning_rate": 2e-05,
      "loss": 0.767,
      "step": 7890
    },
    {
      "epoch": 0.6613645876935956,
      "grad_norm": 0.4243473708629608,
      "learning_rate": 2e-05,
      "loss": 0.7944,
      "step": 7900
    },
    {
      "epoch": 0.6622017580577647,
      "grad_norm": 0.41516244411468506,
      "learning_rate": 2e-05,
      "loss": 0.8067,
      "step": 7910
    },
    {
      "epoch": 0.6630389284219339,
      "grad_norm": 0.40203672647476196,
      "learning_rate": 2e-05,
      "loss": 0.8237,
      "step": 7920
    },
    {
      "epoch": 0.663876098786103,
      "grad_norm": 0.4219101369380951,
      "learning_rate": 2e-05,
      "loss": 0.7938,
      "step": 7930
    },
    {
      "epoch": 0.6647132691502721,
      "grad_norm": 0.41490620374679565,
      "learning_rate": 2e-05,
      "loss": 0.8213,
      "step": 7940
    },
    {
      "epoch": 0.6655504395144412,
      "grad_norm": 0.4170083701610565,
      "learning_rate": 2e-05,
      "loss": 0.8006,
      "step": 7950
    },
    {
      "epoch": 0.6663876098786103,
      "grad_norm": 0.43775680661201477,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 7960
    },
    {
      "epoch": 0.6672247802427794,
      "grad_norm": 0.41242069005966187,
      "learning_rate": 2e-05,
      "loss": 0.7839,
      "step": 7970
    },
    {
      "epoch": 0.6680619506069485,
      "grad_norm": 0.41182929277420044,
      "learning_rate": 2e-05,
      "loss": 0.7779,
      "step": 7980
    },
    {
      "epoch": 0.6688991209711176,
      "grad_norm": 0.42064037919044495,
      "learning_rate": 2e-05,
      "loss": 0.7751,
      "step": 7990
    },
    {
      "epoch": 0.6697362913352868,
      "grad_norm": 0.3792489171028137,
      "learning_rate": 2e-05,
      "loss": 0.7881,
      "step": 8000
    },
    {
      "epoch": 0.6705734616994559,
      "grad_norm": 0.4303393065929413,
      "learning_rate": 2e-05,
      "loss": 0.7762,
      "step": 8010
    },
    {
      "epoch": 0.671410632063625,
      "grad_norm": 0.4031810760498047,
      "learning_rate": 2e-05,
      "loss": 0.795,
      "step": 8020
    },
    {
      "epoch": 0.672247802427794,
      "grad_norm": 0.45795199275016785,
      "learning_rate": 2e-05,
      "loss": 0.8014,
      "step": 8030
    },
    {
      "epoch": 0.6730849727919631,
      "grad_norm": 0.4107005000114441,
      "learning_rate": 2e-05,
      "loss": 0.7998,
      "step": 8040
    },
    {
      "epoch": 0.6739221431561323,
      "grad_norm": 0.40837547183036804,
      "learning_rate": 2e-05,
      "loss": 0.7904,
      "step": 8050
    },
    {
      "epoch": 0.6747593135203014,
      "grad_norm": 0.43157094717025757,
      "learning_rate": 2e-05,
      "loss": 0.7964,
      "step": 8060
    },
    {
      "epoch": 0.6755964838844705,
      "grad_norm": 0.4390855133533478,
      "learning_rate": 2e-05,
      "loss": 0.8061,
      "step": 8070
    },
    {
      "epoch": 0.6764336542486395,
      "grad_norm": 0.4061453342437744,
      "learning_rate": 2e-05,
      "loss": 0.805,
      "step": 8080
    },
    {
      "epoch": 0.6772708246128087,
      "grad_norm": 0.40727293491363525,
      "learning_rate": 2e-05,
      "loss": 0.7689,
      "step": 8090
    },
    {
      "epoch": 0.6781079949769778,
      "grad_norm": 0.4190889000892639,
      "learning_rate": 2e-05,
      "loss": 0.8045,
      "step": 8100
    },
    {
      "epoch": 0.6789451653411469,
      "grad_norm": 0.3897724449634552,
      "learning_rate": 2e-05,
      "loss": 0.7917,
      "step": 8110
    },
    {
      "epoch": 0.679782335705316,
      "grad_norm": 0.4452584683895111,
      "learning_rate": 2e-05,
      "loss": 0.8047,
      "step": 8120
    },
    {
      "epoch": 0.6806195060694852,
      "grad_norm": 0.4112711548805237,
      "learning_rate": 2e-05,
      "loss": 0.8014,
      "step": 8130
    },
    {
      "epoch": 0.6814566764336543,
      "grad_norm": 0.4256594181060791,
      "learning_rate": 2e-05,
      "loss": 0.7975,
      "step": 8140
    },
    {
      "epoch": 0.6822938467978233,
      "grad_norm": 0.42692509293556213,
      "learning_rate": 2e-05,
      "loss": 0.8009,
      "step": 8150
    },
    {
      "epoch": 0.6831310171619924,
      "grad_norm": 0.4026581645011902,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 8160
    },
    {
      "epoch": 0.6839681875261616,
      "grad_norm": 0.39802464842796326,
      "learning_rate": 2e-05,
      "loss": 0.805,
      "step": 8170
    },
    {
      "epoch": 0.6848053578903307,
      "grad_norm": 0.42487025260925293,
      "learning_rate": 2e-05,
      "loss": 0.7807,
      "step": 8180
    },
    {
      "epoch": 0.6856425282544998,
      "grad_norm": 0.45635178685188293,
      "learning_rate": 2e-05,
      "loss": 0.8198,
      "step": 8190
    },
    {
      "epoch": 0.6864796986186689,
      "grad_norm": 0.4110526442527771,
      "learning_rate": 2e-05,
      "loss": 0.7944,
      "step": 8200
    },
    {
      "epoch": 0.6873168689828381,
      "grad_norm": 0.405009388923645,
      "learning_rate": 2e-05,
      "loss": 0.7928,
      "step": 8210
    },
    {
      "epoch": 0.6881540393470071,
      "grad_norm": 0.40149828791618347,
      "learning_rate": 2e-05,
      "loss": 0.7956,
      "step": 8220
    },
    {
      "epoch": 0.6889912097111762,
      "grad_norm": 0.41621068120002747,
      "learning_rate": 2e-05,
      "loss": 0.7838,
      "step": 8230
    },
    {
      "epoch": 0.6898283800753453,
      "grad_norm": 0.3983251452445984,
      "learning_rate": 2e-05,
      "loss": 0.7726,
      "step": 8240
    },
    {
      "epoch": 0.6906655504395144,
      "grad_norm": 0.3939606249332428,
      "learning_rate": 2e-05,
      "loss": 0.7977,
      "step": 8250
    },
    {
      "epoch": 0.6915027208036836,
      "grad_norm": 0.5177695155143738,
      "learning_rate": 2e-05,
      "loss": 0.7987,
      "step": 8260
    },
    {
      "epoch": 0.6923398911678527,
      "grad_norm": 0.4023822546005249,
      "learning_rate": 2e-05,
      "loss": 0.7255,
      "step": 8270
    },
    {
      "epoch": 0.6931770615320217,
      "grad_norm": 0.41135936975479126,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 8280
    },
    {
      "epoch": 0.6940142318961908,
      "grad_norm": 0.42059531807899475,
      "learning_rate": 2e-05,
      "loss": 0.7982,
      "step": 8290
    },
    {
      "epoch": 0.69485140226036,
      "grad_norm": 0.4041062891483307,
      "learning_rate": 2e-05,
      "loss": 0.8245,
      "step": 8300
    },
    {
      "epoch": 0.6956885726245291,
      "grad_norm": 0.3898472785949707,
      "learning_rate": 2e-05,
      "loss": 0.7988,
      "step": 8310
    },
    {
      "epoch": 0.6965257429886982,
      "grad_norm": 0.44837409257888794,
      "learning_rate": 2e-05,
      "loss": 0.8073,
      "step": 8320
    },
    {
      "epoch": 0.6973629133528673,
      "grad_norm": 0.3889636695384979,
      "learning_rate": 2e-05,
      "loss": 0.7776,
      "step": 8330
    },
    {
      "epoch": 0.6982000837170365,
      "grad_norm": 0.3963388502597809,
      "learning_rate": 2e-05,
      "loss": 0.792,
      "step": 8340
    },
    {
      "epoch": 0.6990372540812055,
      "grad_norm": 0.41144704818725586,
      "learning_rate": 2e-05,
      "loss": 0.782,
      "step": 8350
    },
    {
      "epoch": 0.6998744244453746,
      "grad_norm": 0.4133698642253876,
      "learning_rate": 2e-05,
      "loss": 0.7792,
      "step": 8360
    },
    {
      "epoch": 0.7007115948095437,
      "grad_norm": 0.40950098633766174,
      "learning_rate": 2e-05,
      "loss": 0.7984,
      "step": 8370
    },
    {
      "epoch": 0.7015487651737129,
      "grad_norm": 0.4023144841194153,
      "learning_rate": 2e-05,
      "loss": 0.7764,
      "step": 8380
    },
    {
      "epoch": 0.702385935537882,
      "grad_norm": 0.4100695848464966,
      "learning_rate": 2e-05,
      "loss": 0.7927,
      "step": 8390
    },
    {
      "epoch": 0.7032231059020511,
      "grad_norm": 0.4084780216217041,
      "learning_rate": 2e-05,
      "loss": 0.8082,
      "step": 8400
    },
    {
      "epoch": 0.7040602762662201,
      "grad_norm": 0.4039047658443451,
      "learning_rate": 2e-05,
      "loss": 0.7962,
      "step": 8410
    },
    {
      "epoch": 0.7048974466303893,
      "grad_norm": 0.4231127202510834,
      "learning_rate": 2e-05,
      "loss": 0.7822,
      "step": 8420
    },
    {
      "epoch": 0.7057346169945584,
      "grad_norm": 0.4030919075012207,
      "learning_rate": 2e-05,
      "loss": 0.7768,
      "step": 8430
    },
    {
      "epoch": 0.7065717873587275,
      "grad_norm": 0.40938398241996765,
      "learning_rate": 2e-05,
      "loss": 0.776,
      "step": 8440
    },
    {
      "epoch": 0.7074089577228966,
      "grad_norm": 0.39689889550209045,
      "learning_rate": 2e-05,
      "loss": 0.8094,
      "step": 8450
    },
    {
      "epoch": 0.7082461280870657,
      "grad_norm": 0.40689119696617126,
      "learning_rate": 2e-05,
      "loss": 0.8068,
      "step": 8460
    },
    {
      "epoch": 0.7090832984512349,
      "grad_norm": 0.40702974796295166,
      "learning_rate": 2e-05,
      "loss": 0.7885,
      "step": 8470
    },
    {
      "epoch": 0.7099204688154039,
      "grad_norm": 0.3854145407676697,
      "learning_rate": 2e-05,
      "loss": 0.7857,
      "step": 8480
    },
    {
      "epoch": 0.710757639179573,
      "grad_norm": 0.43457910418510437,
      "learning_rate": 2e-05,
      "loss": 0.7895,
      "step": 8490
    },
    {
      "epoch": 0.7115948095437421,
      "grad_norm": 0.46188443899154663,
      "learning_rate": 2e-05,
      "loss": 0.8002,
      "step": 8500
    },
    {
      "epoch": 0.7124319799079113,
      "grad_norm": 0.4108119606971741,
      "learning_rate": 2e-05,
      "loss": 0.7892,
      "step": 8510
    },
    {
      "epoch": 0.7132691502720804,
      "grad_norm": 0.414360374212265,
      "learning_rate": 2e-05,
      "loss": 0.797,
      "step": 8520
    },
    {
      "epoch": 0.7141063206362495,
      "grad_norm": 0.4302920699119568,
      "learning_rate": 2e-05,
      "loss": 0.7898,
      "step": 8530
    },
    {
      "epoch": 0.7149434910004185,
      "grad_norm": 0.3963889479637146,
      "learning_rate": 2e-05,
      "loss": 0.7676,
      "step": 8540
    },
    {
      "epoch": 0.7157806613645877,
      "grad_norm": 0.4531528055667877,
      "learning_rate": 2e-05,
      "loss": 0.7848,
      "step": 8550
    },
    {
      "epoch": 0.7166178317287568,
      "grad_norm": 0.3999135494232178,
      "learning_rate": 2e-05,
      "loss": 0.8049,
      "step": 8560
    },
    {
      "epoch": 0.7174550020929259,
      "grad_norm": 0.42383256554603577,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 8570
    },
    {
      "epoch": 0.718292172457095,
      "grad_norm": 0.5073035955429077,
      "learning_rate": 2e-05,
      "loss": 0.7903,
      "step": 8580
    },
    {
      "epoch": 0.7191293428212642,
      "grad_norm": 0.41070666909217834,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 8590
    },
    {
      "epoch": 0.7199665131854333,
      "grad_norm": 0.3859638571739197,
      "learning_rate": 2e-05,
      "loss": 0.7806,
      "step": 8600
    },
    {
      "epoch": 0.7208036835496023,
      "grad_norm": 0.3983578681945801,
      "learning_rate": 2e-05,
      "loss": 0.7748,
      "step": 8610
    },
    {
      "epoch": 0.7216408539137714,
      "grad_norm": 0.4087788760662079,
      "learning_rate": 2e-05,
      "loss": 0.7948,
      "step": 8620
    },
    {
      "epoch": 0.7224780242779406,
      "grad_norm": 0.40372100472450256,
      "learning_rate": 2e-05,
      "loss": 0.7897,
      "step": 8630
    },
    {
      "epoch": 0.7233151946421097,
      "grad_norm": 0.39190176129341125,
      "learning_rate": 2e-05,
      "loss": 0.7933,
      "step": 8640
    },
    {
      "epoch": 0.7241523650062788,
      "grad_norm": 0.40031641721725464,
      "learning_rate": 2e-05,
      "loss": 0.7847,
      "step": 8650
    },
    {
      "epoch": 0.7249895353704479,
      "grad_norm": 0.4024011492729187,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 8660
    },
    {
      "epoch": 0.7258267057346169,
      "grad_norm": 0.40884527564048767,
      "learning_rate": 2e-05,
      "loss": 0.8207,
      "step": 8670
    },
    {
      "epoch": 0.7266638760987861,
      "grad_norm": 0.41317152976989746,
      "learning_rate": 2e-05,
      "loss": 0.7992,
      "step": 8680
    },
    {
      "epoch": 0.7275010464629552,
      "grad_norm": 0.411294162273407,
      "learning_rate": 2e-05,
      "loss": 0.8038,
      "step": 8690
    },
    {
      "epoch": 0.7283382168271243,
      "grad_norm": 0.41191062331199646,
      "learning_rate": 2e-05,
      "loss": 0.8015,
      "step": 8700
    },
    {
      "epoch": 0.7291753871912934,
      "grad_norm": 0.40348121523857117,
      "learning_rate": 2e-05,
      "loss": 0.7764,
      "step": 8710
    },
    {
      "epoch": 0.7300125575554626,
      "grad_norm": 0.3791161775588989,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 8720
    },
    {
      "epoch": 0.7308497279196317,
      "grad_norm": 0.40631720423698425,
      "learning_rate": 2e-05,
      "loss": 0.7711,
      "step": 8730
    },
    {
      "epoch": 0.7316868982838007,
      "grad_norm": 0.4089135229587555,
      "learning_rate": 2e-05,
      "loss": 0.8064,
      "step": 8740
    },
    {
      "epoch": 0.7325240686479698,
      "grad_norm": 0.42777562141418457,
      "learning_rate": 2e-05,
      "loss": 0.8047,
      "step": 8750
    },
    {
      "epoch": 0.733361239012139,
      "grad_norm": 0.3952829837799072,
      "learning_rate": 2e-05,
      "loss": 0.7775,
      "step": 8760
    },
    {
      "epoch": 0.7341984093763081,
      "grad_norm": 0.4024270474910736,
      "learning_rate": 2e-05,
      "loss": 0.7829,
      "step": 8770
    },
    {
      "epoch": 0.7350355797404772,
      "grad_norm": 0.42851728200912476,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 8780
    },
    {
      "epoch": 0.7358727501046463,
      "grad_norm": 0.40699300169944763,
      "learning_rate": 2e-05,
      "loss": 0.7965,
      "step": 8790
    },
    {
      "epoch": 0.7367099204688154,
      "grad_norm": 0.4464581310749054,
      "learning_rate": 2e-05,
      "loss": 0.7929,
      "step": 8800
    },
    {
      "epoch": 0.7375470908329845,
      "grad_norm": 0.39706119894981384,
      "learning_rate": 2e-05,
      "loss": 0.7883,
      "step": 8810
    },
    {
      "epoch": 0.7383842611971536,
      "grad_norm": 0.401327908039093,
      "learning_rate": 2e-05,
      "loss": 0.7968,
      "step": 8820
    },
    {
      "epoch": 0.7392214315613227,
      "grad_norm": 0.4093974232673645,
      "learning_rate": 2e-05,
      "loss": 0.7979,
      "step": 8830
    },
    {
      "epoch": 0.7400586019254919,
      "grad_norm": 0.41130492091178894,
      "learning_rate": 2e-05,
      "loss": 0.792,
      "step": 8840
    },
    {
      "epoch": 0.740895772289661,
      "grad_norm": 0.4187884032726288,
      "learning_rate": 2e-05,
      "loss": 0.7823,
      "step": 8850
    },
    {
      "epoch": 0.74173294265383,
      "grad_norm": 0.4005022644996643,
      "learning_rate": 2e-05,
      "loss": 0.8212,
      "step": 8860
    },
    {
      "epoch": 0.7425701130179991,
      "grad_norm": 0.389448881149292,
      "learning_rate": 2e-05,
      "loss": 0.8194,
      "step": 8870
    },
    {
      "epoch": 0.7434072833821682,
      "grad_norm": 0.42556115984916687,
      "learning_rate": 2e-05,
      "loss": 0.8042,
      "step": 8880
    },
    {
      "epoch": 0.7442444537463374,
      "grad_norm": 0.3984662592411041,
      "learning_rate": 2e-05,
      "loss": 0.781,
      "step": 8890
    },
    {
      "epoch": 0.7450816241105065,
      "grad_norm": 0.38468483090400696,
      "learning_rate": 2e-05,
      "loss": 0.79,
      "step": 8900
    },
    {
      "epoch": 0.7459187944746756,
      "grad_norm": 0.4322845935821533,
      "learning_rate": 2e-05,
      "loss": 0.8347,
      "step": 8910
    },
    {
      "epoch": 0.7467559648388447,
      "grad_norm": 0.4150550663471222,
      "learning_rate": 2e-05,
      "loss": 0.7694,
      "step": 8920
    },
    {
      "epoch": 0.7475931352030138,
      "grad_norm": 0.4345274269580841,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 8930
    },
    {
      "epoch": 0.7484303055671829,
      "grad_norm": 0.40091168880462646,
      "learning_rate": 2e-05,
      "loss": 0.7826,
      "step": 8940
    },
    {
      "epoch": 0.749267475931352,
      "grad_norm": 0.4115679860115051,
      "learning_rate": 2e-05,
      "loss": 0.7728,
      "step": 8950
    },
    {
      "epoch": 0.7501046462955211,
      "grad_norm": 0.40723174810409546,
      "learning_rate": 2e-05,
      "loss": 0.7924,
      "step": 8960
    },
    {
      "epoch": 0.7509418166596903,
      "grad_norm": 0.3973191976547241,
      "learning_rate": 2e-05,
      "loss": 0.7811,
      "step": 8970
    },
    {
      "epoch": 0.7517789870238594,
      "grad_norm": 0.4895854890346527,
      "learning_rate": 2e-05,
      "loss": 0.8063,
      "step": 8980
    },
    {
      "epoch": 0.7526161573880285,
      "grad_norm": 0.39921241998672485,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 8990
    },
    {
      "epoch": 0.7534533277521975,
      "grad_norm": 0.4079987704753876,
      "learning_rate": 2e-05,
      "loss": 0.8073,
      "step": 9000
    },
    {
      "epoch": 0.7542904981163667,
      "grad_norm": 0.4144892394542694,
      "learning_rate": 2e-05,
      "loss": 0.7792,
      "step": 9010
    },
    {
      "epoch": 0.7551276684805358,
      "grad_norm": 0.4095786213874817,
      "learning_rate": 2e-05,
      "loss": 0.7646,
      "step": 9020
    },
    {
      "epoch": 0.7559648388447049,
      "grad_norm": 0.40863126516342163,
      "learning_rate": 2e-05,
      "loss": 0.7855,
      "step": 9030
    },
    {
      "epoch": 0.756802009208874,
      "grad_norm": 0.4344610571861267,
      "learning_rate": 2e-05,
      "loss": 0.7784,
      "step": 9040
    },
    {
      "epoch": 0.7576391795730432,
      "grad_norm": 0.4246126413345337,
      "learning_rate": 2e-05,
      "loss": 0.784,
      "step": 9050
    },
    {
      "epoch": 0.7584763499372122,
      "grad_norm": 0.4100709855556488,
      "learning_rate": 2e-05,
      "loss": 0.773,
      "step": 9060
    },
    {
      "epoch": 0.7593135203013813,
      "grad_norm": 0.40372657775878906,
      "learning_rate": 2e-05,
      "loss": 0.8044,
      "step": 9070
    },
    {
      "epoch": 0.7601506906655504,
      "grad_norm": 0.4075734615325928,
      "learning_rate": 2e-05,
      "loss": 0.7881,
      "step": 9080
    },
    {
      "epoch": 0.7609878610297196,
      "grad_norm": 0.4120265245437622,
      "learning_rate": 2e-05,
      "loss": 0.7865,
      "step": 9090
    },
    {
      "epoch": 0.7618250313938887,
      "grad_norm": 0.4023255407810211,
      "learning_rate": 2e-05,
      "loss": 0.8112,
      "step": 9100
    },
    {
      "epoch": 0.7626622017580578,
      "grad_norm": 0.420223206281662,
      "learning_rate": 2e-05,
      "loss": 0.8097,
      "step": 9110
    },
    {
      "epoch": 0.7634993721222268,
      "grad_norm": 0.4341658055782318,
      "learning_rate": 2e-05,
      "loss": 0.8013,
      "step": 9120
    },
    {
      "epoch": 0.7643365424863959,
      "grad_norm": 0.40918242931365967,
      "learning_rate": 2e-05,
      "loss": 0.7849,
      "step": 9130
    },
    {
      "epoch": 0.7651737128505651,
      "grad_norm": 0.41013771295547485,
      "learning_rate": 2e-05,
      "loss": 0.7714,
      "step": 9140
    },
    {
      "epoch": 0.7660108832147342,
      "grad_norm": 0.4326823651790619,
      "learning_rate": 2e-05,
      "loss": 0.7888,
      "step": 9150
    },
    {
      "epoch": 0.7668480535789033,
      "grad_norm": 0.41754764318466187,
      "learning_rate": 2e-05,
      "loss": 0.7863,
      "step": 9160
    },
    {
      "epoch": 0.7676852239430724,
      "grad_norm": 0.41990208625793457,
      "learning_rate": 2e-05,
      "loss": 0.7957,
      "step": 9170
    },
    {
      "epoch": 0.7685223943072416,
      "grad_norm": 0.41873452067375183,
      "learning_rate": 2e-05,
      "loss": 0.7948,
      "step": 9180
    },
    {
      "epoch": 0.7693595646714106,
      "grad_norm": 0.43389397859573364,
      "learning_rate": 2e-05,
      "loss": 0.7877,
      "step": 9190
    },
    {
      "epoch": 0.7701967350355797,
      "grad_norm": 0.4370642900466919,
      "learning_rate": 2e-05,
      "loss": 0.7955,
      "step": 9200
    },
    {
      "epoch": 0.7710339053997488,
      "grad_norm": 0.5423875451087952,
      "learning_rate": 2e-05,
      "loss": 0.7878,
      "step": 9210
    },
    {
      "epoch": 0.771871075763918,
      "grad_norm": 0.4270927608013153,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 9220
    },
    {
      "epoch": 0.7727082461280871,
      "grad_norm": 0.39846527576446533,
      "learning_rate": 2e-05,
      "loss": 0.7554,
      "step": 9230
    },
    {
      "epoch": 0.7735454164922562,
      "grad_norm": 0.3837721347808838,
      "learning_rate": 2e-05,
      "loss": 0.7803,
      "step": 9240
    },
    {
      "epoch": 0.7743825868564252,
      "grad_norm": 0.41136160492897034,
      "learning_rate": 2e-05,
      "loss": 0.7932,
      "step": 9250
    },
    {
      "epoch": 0.7752197572205944,
      "grad_norm": 0.428602397441864,
      "learning_rate": 2e-05,
      "loss": 0.7876,
      "step": 9260
    },
    {
      "epoch": 0.7760569275847635,
      "grad_norm": 0.3869684338569641,
      "learning_rate": 2e-05,
      "loss": 0.7729,
      "step": 9270
    },
    {
      "epoch": 0.7768940979489326,
      "grad_norm": 0.38985365629196167,
      "learning_rate": 2e-05,
      "loss": 0.7826,
      "step": 9280
    },
    {
      "epoch": 0.7777312683131017,
      "grad_norm": 0.4979279935359955,
      "learning_rate": 2e-05,
      "loss": 0.7795,
      "step": 9290
    },
    {
      "epoch": 0.7785684386772709,
      "grad_norm": 0.3896363377571106,
      "learning_rate": 2e-05,
      "loss": 0.7621,
      "step": 9300
    },
    {
      "epoch": 0.77940560904144,
      "grad_norm": 0.4167052209377289,
      "learning_rate": 2e-05,
      "loss": 0.7758,
      "step": 9310
    },
    {
      "epoch": 0.780242779405609,
      "grad_norm": 0.42102763056755066,
      "learning_rate": 2e-05,
      "loss": 0.7881,
      "step": 9320
    },
    {
      "epoch": 0.7810799497697781,
      "grad_norm": 0.40131065249443054,
      "learning_rate": 2e-05,
      "loss": 0.7889,
      "step": 9330
    },
    {
      "epoch": 0.7819171201339472,
      "grad_norm": 0.40912577509880066,
      "learning_rate": 2e-05,
      "loss": 0.7869,
      "step": 9340
    },
    {
      "epoch": 0.7827542904981164,
      "grad_norm": 0.41473227739334106,
      "learning_rate": 2e-05,
      "loss": 0.8177,
      "step": 9350
    },
    {
      "epoch": 0.7835914608622855,
      "grad_norm": 0.40218645334243774,
      "learning_rate": 2e-05,
      "loss": 0.8122,
      "step": 9360
    },
    {
      "epoch": 0.7844286312264546,
      "grad_norm": 0.39744991064071655,
      "learning_rate": 2e-05,
      "loss": 0.7784,
      "step": 9370
    },
    {
      "epoch": 0.7852658015906236,
      "grad_norm": 0.5038439035415649,
      "learning_rate": 2e-05,
      "loss": 0.778,
      "step": 9380
    },
    {
      "epoch": 0.7861029719547928,
      "grad_norm": 0.4194819927215576,
      "learning_rate": 2e-05,
      "loss": 0.7995,
      "step": 9390
    },
    {
      "epoch": 0.7869401423189619,
      "grad_norm": 0.4059963524341583,
      "learning_rate": 2e-05,
      "loss": 0.8061,
      "step": 9400
    },
    {
      "epoch": 0.787777312683131,
      "grad_norm": 0.5077357292175293,
      "learning_rate": 2e-05,
      "loss": 0.7769,
      "step": 9410
    },
    {
      "epoch": 0.7886144830473001,
      "grad_norm": 0.4049496054649353,
      "learning_rate": 2e-05,
      "loss": 0.7848,
      "step": 9420
    },
    {
      "epoch": 0.7894516534114693,
      "grad_norm": 0.4718087613582611,
      "learning_rate": 2e-05,
      "loss": 0.7849,
      "step": 9430
    },
    {
      "epoch": 0.7902888237756384,
      "grad_norm": 0.3873784840106964,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 9440
    },
    {
      "epoch": 0.7911259941398074,
      "grad_norm": 0.4217713475227356,
      "learning_rate": 2e-05,
      "loss": 0.7811,
      "step": 9450
    },
    {
      "epoch": 0.7919631645039765,
      "grad_norm": 0.4054142236709595,
      "learning_rate": 2e-05,
      "loss": 0.7573,
      "step": 9460
    },
    {
      "epoch": 0.7928003348681457,
      "grad_norm": 0.4045252203941345,
      "learning_rate": 2e-05,
      "loss": 0.75,
      "step": 9470
    },
    {
      "epoch": 0.7936375052323148,
      "grad_norm": 0.4143926501274109,
      "learning_rate": 2e-05,
      "loss": 0.7849,
      "step": 9480
    },
    {
      "epoch": 0.7944746755964839,
      "grad_norm": 0.42893701791763306,
      "learning_rate": 2e-05,
      "loss": 0.7774,
      "step": 9490
    },
    {
      "epoch": 0.795311845960653,
      "grad_norm": 0.4567032754421234,
      "learning_rate": 2e-05,
      "loss": 0.7776,
      "step": 9500
    },
    {
      "epoch": 0.7961490163248222,
      "grad_norm": 0.4202049672603607,
      "learning_rate": 2e-05,
      "loss": 0.7882,
      "step": 9510
    },
    {
      "epoch": 0.7969861866889912,
      "grad_norm": 0.4143880605697632,
      "learning_rate": 2e-05,
      "loss": 0.8067,
      "step": 9520
    },
    {
      "epoch": 0.7978233570531603,
      "grad_norm": 0.3921278715133667,
      "learning_rate": 2e-05,
      "loss": 0.7943,
      "step": 9530
    },
    {
      "epoch": 0.7986605274173294,
      "grad_norm": 0.4028029143810272,
      "learning_rate": 2e-05,
      "loss": 0.7557,
      "step": 9540
    },
    {
      "epoch": 0.7994976977814985,
      "grad_norm": 0.4280000925064087,
      "learning_rate": 2e-05,
      "loss": 0.785,
      "step": 9550
    },
    {
      "epoch": 0.8003348681456677,
      "grad_norm": 0.4287838935852051,
      "learning_rate": 2e-05,
      "loss": 0.7584,
      "step": 9560
    },
    {
      "epoch": 0.8011720385098368,
      "grad_norm": 0.40632376074790955,
      "learning_rate": 2e-05,
      "loss": 0.7867,
      "step": 9570
    },
    {
      "epoch": 0.8020092088740058,
      "grad_norm": 0.4311842620372772,
      "learning_rate": 2e-05,
      "loss": 0.7586,
      "step": 9580
    },
    {
      "epoch": 0.8028463792381749,
      "grad_norm": 0.40805384516716003,
      "learning_rate": 2e-05,
      "loss": 0.7919,
      "step": 9590
    },
    {
      "epoch": 0.8036835496023441,
      "grad_norm": 0.39368927478790283,
      "learning_rate": 2e-05,
      "loss": 0.79,
      "step": 9600
    },
    {
      "epoch": 0.8045207199665132,
      "grad_norm": 0.3926279842853546,
      "learning_rate": 2e-05,
      "loss": 0.8016,
      "step": 9610
    },
    {
      "epoch": 0.8053578903306823,
      "grad_norm": 0.41362622380256653,
      "learning_rate": 2e-05,
      "loss": 0.7824,
      "step": 9620
    },
    {
      "epoch": 0.8061950606948514,
      "grad_norm": 0.4133247435092926,
      "learning_rate": 2e-05,
      "loss": 0.7778,
      "step": 9630
    },
    {
      "epoch": 0.8070322310590206,
      "grad_norm": 0.49495041370391846,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 9640
    },
    {
      "epoch": 0.8078694014231896,
      "grad_norm": 0.4665575623512268,
      "learning_rate": 2e-05,
      "loss": 0.772,
      "step": 9650
    },
    {
      "epoch": 0.8087065717873587,
      "grad_norm": 0.40782490372657776,
      "learning_rate": 2e-05,
      "loss": 0.7785,
      "step": 9660
    },
    {
      "epoch": 0.8095437421515278,
      "grad_norm": 0.41323399543762207,
      "learning_rate": 2e-05,
      "loss": 0.774,
      "step": 9670
    },
    {
      "epoch": 0.810380912515697,
      "grad_norm": 0.42448052763938904,
      "learning_rate": 2e-05,
      "loss": 0.8047,
      "step": 9680
    },
    {
      "epoch": 0.8112180828798661,
      "grad_norm": 0.41691261529922485,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 9690
    },
    {
      "epoch": 0.8120552532440352,
      "grad_norm": 0.4008970856666565,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 9700
    },
    {
      "epoch": 0.8128924236082042,
      "grad_norm": 0.40141957998275757,
      "learning_rate": 2e-05,
      "loss": 0.7794,
      "step": 9710
    },
    {
      "epoch": 0.8137295939723734,
      "grad_norm": 0.40745553374290466,
      "learning_rate": 2e-05,
      "loss": 0.7965,
      "step": 9720
    },
    {
      "epoch": 0.8145667643365425,
      "grad_norm": 0.41115742921829224,
      "learning_rate": 2e-05,
      "loss": 0.8228,
      "step": 9730
    },
    {
      "epoch": 0.8154039347007116,
      "grad_norm": 0.4241759181022644,
      "learning_rate": 2e-05,
      "loss": 0.7605,
      "step": 9740
    },
    {
      "epoch": 0.8162411050648807,
      "grad_norm": 0.4267658293247223,
      "learning_rate": 2e-05,
      "loss": 0.7546,
      "step": 9750
    },
    {
      "epoch": 0.8170782754290498,
      "grad_norm": 0.4186568856239319,
      "learning_rate": 2e-05,
      "loss": 0.7814,
      "step": 9760
    },
    {
      "epoch": 0.817915445793219,
      "grad_norm": 0.4376239776611328,
      "learning_rate": 2e-05,
      "loss": 0.7921,
      "step": 9770
    },
    {
      "epoch": 0.818752616157388,
      "grad_norm": 0.4061380624771118,
      "learning_rate": 2e-05,
      "loss": 0.7888,
      "step": 9780
    },
    {
      "epoch": 0.8195897865215571,
      "grad_norm": 0.4069521427154541,
      "learning_rate": 2e-05,
      "loss": 0.7936,
      "step": 9790
    },
    {
      "epoch": 0.8204269568857262,
      "grad_norm": 0.4433954060077667,
      "learning_rate": 2e-05,
      "loss": 0.7975,
      "step": 9800
    },
    {
      "epoch": 0.8212641272498954,
      "grad_norm": 0.38397154211997986,
      "learning_rate": 2e-05,
      "loss": 0.7896,
      "step": 9810
    },
    {
      "epoch": 0.8221012976140645,
      "grad_norm": 0.5234093070030212,
      "learning_rate": 2e-05,
      "loss": 0.7971,
      "step": 9820
    },
    {
      "epoch": 0.8229384679782336,
      "grad_norm": 0.4187336564064026,
      "learning_rate": 2e-05,
      "loss": 0.7839,
      "step": 9830
    },
    {
      "epoch": 0.8237756383424026,
      "grad_norm": 0.42921149730682373,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 9840
    },
    {
      "epoch": 0.8246128087065718,
      "grad_norm": 0.4375268518924713,
      "learning_rate": 2e-05,
      "loss": 0.7671,
      "step": 9850
    },
    {
      "epoch": 0.8254499790707409,
      "grad_norm": 0.3993733525276184,
      "learning_rate": 2e-05,
      "loss": 0.8139,
      "step": 9860
    },
    {
      "epoch": 0.82628714943491,
      "grad_norm": 0.5366401076316833,
      "learning_rate": 2e-05,
      "loss": 0.7884,
      "step": 9870
    },
    {
      "epoch": 0.8271243197990791,
      "grad_norm": 0.40350258350372314,
      "learning_rate": 2e-05,
      "loss": 0.7773,
      "step": 9880
    },
    {
      "epoch": 0.8279614901632483,
      "grad_norm": 0.43162432312965393,
      "learning_rate": 2e-05,
      "loss": 0.7802,
      "step": 9890
    },
    {
      "epoch": 0.8287986605274174,
      "grad_norm": 0.41010573506355286,
      "learning_rate": 2e-05,
      "loss": 0.7971,
      "step": 9900
    },
    {
      "epoch": 0.8296358308915864,
      "grad_norm": 0.405447781085968,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 9910
    },
    {
      "epoch": 0.8304730012557555,
      "grad_norm": 0.4126070737838745,
      "learning_rate": 2e-05,
      "loss": 0.7961,
      "step": 9920
    },
    {
      "epoch": 0.8313101716199247,
      "grad_norm": 0.4141429364681244,
      "learning_rate": 2e-05,
      "loss": 0.7813,
      "step": 9930
    },
    {
      "epoch": 0.8321473419840938,
      "grad_norm": 0.40787264704704285,
      "learning_rate": 2e-05,
      "loss": 0.7879,
      "step": 9940
    },
    {
      "epoch": 0.8329845123482629,
      "grad_norm": 0.39907410740852356,
      "learning_rate": 2e-05,
      "loss": 0.8006,
      "step": 9950
    },
    {
      "epoch": 0.833821682712432,
      "grad_norm": 0.44457390904426575,
      "learning_rate": 2e-05,
      "loss": 0.7759,
      "step": 9960
    },
    {
      "epoch": 0.834658853076601,
      "grad_norm": 0.42804864048957825,
      "learning_rate": 2e-05,
      "loss": 0.7453,
      "step": 9970
    },
    {
      "epoch": 0.8354960234407702,
      "grad_norm": 0.4655700922012329,
      "learning_rate": 2e-05,
      "loss": 0.7882,
      "step": 9980
    },
    {
      "epoch": 0.8363331938049393,
      "grad_norm": 0.3962029218673706,
      "learning_rate": 2e-05,
      "loss": 0.7998,
      "step": 9990
    },
    {
      "epoch": 0.8371703641691084,
      "grad_norm": 0.4372413754463196,
      "learning_rate": 2e-05,
      "loss": 0.7724,
      "step": 10000
    },
    {
      "epoch": 0.8380075345332775,
      "grad_norm": 0.3969194293022156,
      "learning_rate": 2e-05,
      "loss": 0.7931,
      "step": 10010
    },
    {
      "epoch": 0.8388447048974467,
      "grad_norm": 0.43875735998153687,
      "learning_rate": 2e-05,
      "loss": 0.7797,
      "step": 10020
    },
    {
      "epoch": 0.8396818752616158,
      "grad_norm": 0.400695264339447,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 10030
    },
    {
      "epoch": 0.8405190456257848,
      "grad_norm": 0.39386802911758423,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 10040
    },
    {
      "epoch": 0.8413562159899539,
      "grad_norm": 0.39776623249053955,
      "learning_rate": 2e-05,
      "loss": 0.8022,
      "step": 10050
    },
    {
      "epoch": 0.8421933863541231,
      "grad_norm": 0.4134213924407959,
      "learning_rate": 2e-05,
      "loss": 0.78,
      "step": 10060
    },
    {
      "epoch": 0.8430305567182922,
      "grad_norm": 0.40812039375305176,
      "learning_rate": 2e-05,
      "loss": 0.7824,
      "step": 10070
    },
    {
      "epoch": 0.8438677270824613,
      "grad_norm": 0.39007607102394104,
      "learning_rate": 2e-05,
      "loss": 0.7851,
      "step": 10080
    },
    {
      "epoch": 0.8447048974466304,
      "grad_norm": 0.41147395968437195,
      "learning_rate": 2e-05,
      "loss": 0.7999,
      "step": 10090
    },
    {
      "epoch": 0.8455420678107995,
      "grad_norm": 0.41889333724975586,
      "learning_rate": 2e-05,
      "loss": 0.7703,
      "step": 10100
    },
    {
      "epoch": 0.8463792381749686,
      "grad_norm": 0.6516678333282471,
      "learning_rate": 2e-05,
      "loss": 0.7892,
      "step": 10110
    },
    {
      "epoch": 0.8472164085391377,
      "grad_norm": 0.40248796343803406,
      "learning_rate": 2e-05,
      "loss": 0.8,
      "step": 10120
    },
    {
      "epoch": 0.8480535789033068,
      "grad_norm": 0.38247135281562805,
      "learning_rate": 2e-05,
      "loss": 0.7642,
      "step": 10130
    },
    {
      "epoch": 0.848890749267476,
      "grad_norm": 0.39540186524391174,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 10140
    },
    {
      "epoch": 0.8497279196316451,
      "grad_norm": 0.4390556216239929,
      "learning_rate": 2e-05,
      "loss": 0.7911,
      "step": 10150
    },
    {
      "epoch": 0.8505650899958141,
      "grad_norm": 0.39682328701019287,
      "learning_rate": 2e-05,
      "loss": 0.7579,
      "step": 10160
    },
    {
      "epoch": 0.8514022603599832,
      "grad_norm": 0.38294512033462524,
      "learning_rate": 2e-05,
      "loss": 0.7479,
      "step": 10170
    },
    {
      "epoch": 0.8522394307241523,
      "grad_norm": 0.3970990478992462,
      "learning_rate": 1.999957358086795e-05,
      "loss": 0.7135,
      "step": 10180
    },
    {
      "epoch": 0.8530766010883215,
      "grad_norm": 0.39725810289382935,
      "learning_rate": 1.9991993813020483e-05,
      "loss": 0.7939,
      "step": 10190
    },
    {
      "epoch": 0.8539137714524906,
      "grad_norm": 0.4502103924751282,
      "learning_rate": 1.9974946373028628e-05,
      "loss": 0.7719,
      "step": 10200
    },
    {
      "epoch": 0.8547509418166597,
      "grad_norm": 0.40650278329849243,
      "learning_rate": 1.994844749501764e-05,
      "loss": 0.7779,
      "step": 10210
    },
    {
      "epoch": 0.8555881121808288,
      "grad_norm": 0.39621585607528687,
      "learning_rate": 1.9912522413632988e-05,
      "loss": 0.7736,
      "step": 10220
    },
    {
      "epoch": 0.856425282544998,
      "grad_norm": 0.4057925343513489,
      "learning_rate": 1.9867205340009656e-05,
      "loss": 0.7818,
      "step": 10230
    },
    {
      "epoch": 0.857262452909167,
      "grad_norm": 0.37848860025405884,
      "learning_rate": 1.9812539429193164e-05,
      "loss": 0.7846,
      "step": 10240
    },
    {
      "epoch": 0.8580996232733361,
      "grad_norm": 0.3956529498100281,
      "learning_rate": 1.9748576739043423e-05,
      "loss": 0.7505,
      "step": 10250
    },
    {
      "epoch": 0.8589367936375052,
      "grad_norm": 0.40521538257598877,
      "learning_rate": 1.967537818066047e-05,
      "loss": 0.7935,
      "step": 10260
    },
    {
      "epoch": 0.8597739640016744,
      "grad_norm": 0.4332334101200104,
      "learning_rate": 1.959301346037941e-05,
      "loss": 0.78,
      "step": 10270
    },
    {
      "epoch": 0.8606111343658435,
      "grad_norm": 0.4038090109825134,
      "learning_rate": 1.9501561013389646e-05,
      "loss": 0.7875,
      "step": 10280
    },
    {
      "epoch": 0.8614483047300125,
      "grad_norm": 0.3999965190887451,
      "learning_rate": 1.9401107929041795e-05,
      "loss": 0.7946,
      "step": 10290
    },
    {
      "epoch": 0.8622854750941816,
      "grad_norm": 0.4089524447917938,
      "learning_rate": 1.9291749867913225e-05,
      "loss": 0.7804,
      "step": 10300
    },
    {
      "epoch": 0.8631226454583508,
      "grad_norm": 0.40620413422584534,
      "learning_rate": 1.91735909707114e-05,
      "loss": 0.7797,
      "step": 10310
    },
    {
      "epoch": 0.8639598158225199,
      "grad_norm": 0.4112906754016876,
      "learning_rate": 1.9046743759101545e-05,
      "loss": 0.7676,
      "step": 10320
    },
    {
      "epoch": 0.864796986186689,
      "grad_norm": 0.4237734377384186,
      "learning_rate": 1.8911329028553324e-05,
      "loss": 0.7958,
      "step": 10330
    },
    {
      "epoch": 0.8656341565508581,
      "grad_norm": 0.46994486451148987,
      "learning_rate": 1.876747573330835e-05,
      "loss": 0.7696,
      "step": 10340
    },
    {
      "epoch": 0.8664713269150273,
      "grad_norm": 0.4085690379142761,
      "learning_rate": 1.8615320863578244e-05,
      "loss": 0.7825,
      "step": 10350
    },
    {
      "epoch": 0.8673084972791963,
      "grad_norm": 0.40162667632102966,
      "learning_rate": 1.8455009315090036e-05,
      "loss": 0.7944,
      "step": 10360
    },
    {
      "epoch": 0.8681456676433654,
      "grad_norm": 0.4280097186565399,
      "learning_rate": 1.8286693751103294e-05,
      "loss": 0.7349,
      "step": 10370
    },
    {
      "epoch": 0.8689828380075345,
      "grad_norm": 0.3991081118583679,
      "learning_rate": 1.811053445703024e-05,
      "loss": 0.7748,
      "step": 10380
    },
    {
      "epoch": 0.8698200083717036,
      "grad_norm": 0.4076809287071228,
      "learning_rate": 1.7926699187797375e-05,
      "loss": 0.7649,
      "step": 10390
    },
    {
      "epoch": 0.8706571787358728,
      "grad_norm": 0.40122753381729126,
      "learning_rate": 1.7735363008093995e-05,
      "loss": 0.7891,
      "step": 10400
    },
    {
      "epoch": 0.8714943491000419,
      "grad_norm": 0.41624024510383606,
      "learning_rate": 1.7536708125659577e-05,
      "loss": 0.7736,
      "step": 10410
    },
    {
      "epoch": 0.872331519464211,
      "grad_norm": 0.44014501571655273,
      "learning_rate": 1.733092371776905e-05,
      "loss": 0.7828,
      "step": 10420
    },
    {
      "epoch": 0.87316868982838,
      "grad_norm": 0.4193737804889679,
      "learning_rate": 1.7118205751080915e-05,
      "loss": 0.7596,
      "step": 10430
    },
    {
      "epoch": 0.8740058601925492,
      "grad_norm": 0.48501110076904297,
      "learning_rate": 1.6898756795019935e-05,
      "loss": 0.7781,
      "step": 10440
    },
    {
      "epoch": 0.8748430305567183,
      "grad_norm": 0.42646265029907227,
      "learning_rate": 1.667278582887211e-05,
      "loss": 0.804,
      "step": 10450
    },
    {
      "epoch": 0.8756802009208874,
      "grad_norm": 0.4148460626602173,
      "learning_rate": 1.6440508042775545e-05,
      "loss": 0.7825,
      "step": 10460
    },
    {
      "epoch": 0.8765173712850565,
      "grad_norm": 0.4158240258693695,
      "learning_rate": 1.6202144632796796e-05,
      "loss": 0.7619,
      "step": 10470
    },
    {
      "epoch": 0.8773545416492257,
      "grad_norm": 0.3982948362827301,
      "learning_rate": 1.595792259028783e-05,
      "loss": 0.7745,
      "step": 10480
    },
    {
      "epoch": 0.8781917120133947,
      "grad_norm": 0.4331926703453064,
      "learning_rate": 1.570807448572419e-05,
      "loss": 0.7814,
      "step": 10490
    },
    {
      "epoch": 0.8790288823775638,
      "grad_norm": 0.40132394433021545,
      "learning_rate": 1.5452838247230197e-05,
      "loss": 0.74,
      "step": 10500
    },
    {
      "epoch": 0.8798660527417329,
      "grad_norm": 0.4001438021659851,
      "learning_rate": 1.5192456934002106e-05,
      "loss": 0.777,
      "step": 10510
    },
    {
      "epoch": 0.8807032231059021,
      "grad_norm": 0.38074469566345215,
      "learning_rate": 1.492717850484502e-05,
      "loss": 0.7599,
      "step": 10520
    },
    {
      "epoch": 0.8815403934700712,
      "grad_norm": 0.3825223743915558,
      "learning_rate": 1.4657255582043904e-05,
      "loss": 0.7787,
      "step": 10530
    },
    {
      "epoch": 0.8823775638342403,
      "grad_norm": 0.39425063133239746,
      "learning_rate": 1.4382945210793663e-05,
      "loss": 0.7627,
      "step": 10540
    },
    {
      "epoch": 0.8832147341984093,
      "grad_norm": 0.38767021894454956,
      "learning_rate": 1.4104508614417267e-05,
      "loss": 0.7734,
      "step": 10550
    },
    {
      "epoch": 0.8840519045625785,
      "grad_norm": 0.39165788888931274,
      "learning_rate": 1.382221094560512e-05,
      "loss": 0.7509,
      "step": 10560
    },
    {
      "epoch": 0.8848890749267476,
      "grad_norm": 0.38623759150505066,
      "learning_rate": 1.3536321033912506e-05,
      "loss": 0.7608,
      "step": 10570
    },
    {
      "epoch": 0.8857262452909167,
      "grad_norm": 0.37547439336776733,
      "learning_rate": 1.3247111129755609e-05,
      "loss": 0.758,
      "step": 10580
    },
    {
      "epoch": 0.8865634156550858,
      "grad_norm": 0.7190909385681152,
      "learning_rate": 1.2954856645149868e-05,
      "loss": 0.7798,
      "step": 10590
    },
    {
      "epoch": 0.8874005860192549,
      "grad_norm": 0.40854576230049133,
      "learning_rate": 1.2659835891437553e-05,
      "loss": 0.7657,
      "step": 10600
    },
    {
      "epoch": 0.8882377563834241,
      "grad_norm": 0.3927224576473236,
      "learning_rate": 1.236232981425436e-05,
      "loss": 0.7531,
      "step": 10610
    },
    {
      "epoch": 0.8890749267475931,
      "grad_norm": 0.44055208563804626,
      "learning_rate": 1.2062621725987402e-05,
      "loss": 0.7719,
      "step": 10620
    },
    {
      "epoch": 0.8899120971117622,
      "grad_norm": 0.43005824089050293,
      "learning_rate": 1.1760997035979318e-05,
      "loss": 0.7503,
      "step": 10630
    },
    {
      "epoch": 0.8907492674759313,
      "grad_norm": 0.40167272090911865,
      "learning_rate": 1.1457742978735514e-05,
      "loss": 0.7687,
      "step": 10640
    },
    {
      "epoch": 0.8915864378401005,
      "grad_norm": 0.39120522141456604,
      "learning_rate": 1.1153148340393298e-05,
      "loss": 0.7518,
      "step": 10650
    },
    {
      "epoch": 0.8924236082042696,
      "grad_norm": 0.4053162932395935,
      "learning_rate": 1.0847503183713424e-05,
      "loss": 0.7669,
      "step": 10660
    },
    {
      "epoch": 0.8932607785684387,
      "grad_norm": 0.42379456758499146,
      "learning_rate": 1.0541098571855896e-05,
      "loss": 0.7812,
      "step": 10670
    },
    {
      "epoch": 0.8940979489326077,
      "grad_norm": 0.38055798411369324,
      "learning_rate": 1.0234226291203123e-05,
      "loss": 0.7406,
      "step": 10680
    },
    {
      "epoch": 0.8949351192967769,
      "grad_norm": 0.41027969121932983,
      "learning_rate": 9.927178573494333e-06,
      "loss": 0.7632,
      "step": 10690
    },
    {
      "epoch": 0.895772289660946,
      "grad_norm": 0.4155137538909912,
      "learning_rate": 9.620247817535902e-06,
      "loss": 0.7494,
      "step": 10700
    },
    {
      "epoch": 0.8966094600251151,
      "grad_norm": 0.39632943272590637,
      "learning_rate": 9.313726310752581e-06,
      "loss": 0.7609,
      "step": 10710
    },
    {
      "epoch": 0.8974466303892842,
      "grad_norm": 0.4075375497341156,
      "learning_rate": 9.00790595084477e-06,
      "loss": 0.7572,
      "step": 10720
    },
    {
      "epoch": 0.8982838007534534,
      "grad_norm": 0.3957861363887787,
      "learning_rate": 8.703077967816944e-06,
      "loss": 0.7493,
      "step": 10730
    },
    {
      "epoch": 0.8991209711176225,
      "grad_norm": 0.39743772149086,
      "learning_rate": 8.39953264664192e-06,
      "loss": 0.7564,
      "step": 10740
    },
    {
      "epoch": 0.8999581414817915,
      "grad_norm": 0.3956974148750305,
      "learning_rate": 8.097559050825058e-06,
      "loss": 0.7712,
      "step": 10750
    },
    {
      "epoch": 0.9007953118459606,
      "grad_norm": 0.3853384554386139,
      "learning_rate": 7.79744474713166e-06,
      "loss": 0.7689,
      "step": 10760
    },
    {
      "epoch": 0.9016324822101298,
      "grad_norm": 0.41287317872047424,
      "learning_rate": 7.499475531739696e-06,
      "loss": 0.7715,
      "step": 10770
    },
    {
      "epoch": 0.9024696525742989,
      "grad_norm": 0.39758217334747314,
      "learning_rate": 7.203935158078662e-06,
      "loss": 0.7691,
      "step": 10780
    },
    {
      "epoch": 0.903306822938468,
      "grad_norm": 0.3888950049877167,
      "learning_rate": 6.911105066613709e-06,
      "loss": 0.7568,
      "step": 10790
    },
    {
      "epoch": 0.9041439933026371,
      "grad_norm": 0.4125528633594513,
      "learning_rate": 6.621264116832394e-06,
      "loss": 0.7525,
      "step": 10800
    },
    {
      "epoch": 0.9049811636668061,
      "grad_norm": 0.38883379101753235,
      "learning_rate": 6.334688321689274e-06,
      "loss": 0.7742,
      "step": 10810
    },
    {
      "epoch": 0.9058183340309753,
      "grad_norm": 0.38019153475761414,
      "learning_rate": 6.051650584761237e-06,
      "loss": 0.7632,
      "step": 10820
    },
    {
      "epoch": 0.9066555043951444,
      "grad_norm": 0.3962251543998718,
      "learning_rate": 5.772420440363841e-06,
      "loss": 0.7474,
      "step": 10830
    },
    {
      "epoch": 0.9074926747593135,
      "grad_norm": 0.39763355255126953,
      "learning_rate": 5.497263796876207e-06,
      "loss": 0.7768,
      "step": 10840
    },
    {
      "epoch": 0.9083298451234826,
      "grad_norm": 0.4058111608028412,
      "learning_rate": 5.226442683518813e-06,
      "loss": 0.7762,
      "step": 10850
    },
    {
      "epoch": 0.9091670154876518,
      "grad_norm": 0.3862776458263397,
      "learning_rate": 4.960215000825432e-06,
      "loss": 0.7332,
      "step": 10860
    },
    {
      "epoch": 0.9100041858518209,
      "grad_norm": 0.4039154350757599,
      "learning_rate": 4.698834275046711e-06,
      "loss": 0.7571,
      "step": 10870
    },
    {
      "epoch": 0.9108413562159899,
      "grad_norm": 0.3955695629119873,
      "learning_rate": 4.442549416719423e-06,
      "loss": 0.7757,
      "step": 10880
    },
    {
      "epoch": 0.911678526580159,
      "grad_norm": 0.37649402022361755,
      "learning_rate": 4.1916044836311565e-06,
      "loss": 0.7136,
      "step": 10890
    },
    {
      "epoch": 0.9125156969443282,
      "grad_norm": 0.38480621576309204,
      "learning_rate": 3.94623844840626e-06,
      "loss": 0.7309,
      "step": 10900
    },
    {
      "epoch": 0.9133528673084973,
      "grad_norm": 0.39539921283721924,
      "learning_rate": 3.7066849709343467e-06,
      "loss": 0.7568,
      "step": 10910
    },
    {
      "epoch": 0.9141900376726664,
      "grad_norm": 0.4061337411403656,
      "learning_rate": 3.473172175858059e-06,
      "loss": 0.7854,
      "step": 10920
    },
    {
      "epoch": 0.9150272080368355,
      "grad_norm": 0.41324687004089355,
      "learning_rate": 3.2459224353320046e-06,
      "loss": 0.7561,
      "step": 10930
    },
    {
      "epoch": 0.9158643784010047,
      "grad_norm": 0.3875410854816437,
      "learning_rate": 3.025152157259735e-06,
      "loss": 0.7958,
      "step": 10940
    },
    {
      "epoch": 0.9167015487651737,
      "grad_norm": 0.3907521963119507,
      "learning_rate": 2.811071579210425e-06,
      "loss": 0.7244,
      "step": 10950
    },
    {
      "epoch": 0.9175387191293428,
      "grad_norm": 0.43058648705482483,
      "learning_rate": 2.603884568211513e-06,
      "loss": 0.7545,
      "step": 10960
    },
    {
      "epoch": 0.9183758894935119,
      "grad_norm": 0.38542991876602173,
      "learning_rate": 2.4037884266079434e-06,
      "loss": 0.7692,
      "step": 10970
    },
    {
      "epoch": 0.9192130598576811,
      "grad_norm": 0.4383396804332733,
      "learning_rate": 2.2109737041728955e-06,
      "loss": 0.733,
      "step": 10980
    },
    {
      "epoch": 0.9200502302218502,
      "grad_norm": 0.40467047691345215,
      "learning_rate": 2.0256240166489454e-06,
      "loss": 0.7522,
      "step": 10990
    },
    {
      "epoch": 0.9208874005860193,
      "grad_norm": 0.4085136651992798,
      "learning_rate": 1.847915870892413e-06,
      "loss": 0.7495,
      "step": 11000
    },
    {
      "epoch": 0.9217245709501883,
      "grad_norm": 0.3844274580478668,
      "learning_rate": 1.6780184967874715e-06,
      "loss": 0.7277,
      "step": 11010
    },
    {
      "epoch": 0.9225617413143574,
      "grad_norm": 0.3880290687084198,
      "learning_rate": 1.5160936860900294e-06,
      "loss": 0.7565,
      "step": 11020
    },
    {
      "epoch": 0.9233989116785266,
      "grad_norm": 0.36231300234794617,
      "learning_rate": 1.3622956383548762e-06,
      "loss": 0.7359,
      "step": 11030
    },
    {
      "epoch": 0.9242360820426957,
      "grad_norm": 0.40010949969291687,
      "learning_rate": 1.216770814092834e-06,
      "loss": 0.7102,
      "step": 11040
    },
    {
      "epoch": 0.9250732524068648,
      "grad_norm": 0.3719671368598938,
      "learning_rate": 1.0796577952977013e-06,
      "loss": 0.7589,
      "step": 11050
    },
    {
      "epoch": 0.9259104227710339,
      "grad_norm": 0.3980219066143036,
      "learning_rate": 9.510871534758705e-07,
      "loss": 0.7253,
      "step": 11060
    },
    {
      "epoch": 0.926747593135203,
      "grad_norm": 0.40069815516471863,
      "learning_rate": 8.311813253042334e-07,
      "loss": 0.7694,
      "step": 11070
    },
    {
      "epoch": 0.9275847634993721,
      "grad_norm": 0.42037340998649597,
      "learning_rate": 7.200544960348308e-07,
      "loss": 0.7574,
      "step": 11080
    },
    {
      "epoch": 0.9284219338635412,
      "grad_norm": 0.42481547594070435,
      "learning_rate": 6.178124907572545e-07,
      "loss": 0.7293,
      "step": 11090
    },
    {
      "epoch": 0.9292591042277103,
      "grad_norm": 0.4148436486721039,
      "learning_rate": 5.245526736223528e-07,
      "loss": 0.742,
      "step": 11100
    },
    {
      "epoch": 0.9300962745918795,
      "grad_norm": 0.3991701006889343,
      "learning_rate": 4.403638551232205e-07,
      "loss": 0.7685,
      "step": 11110
    },
    {
      "epoch": 0.9309334449560486,
      "grad_norm": 0.4135954678058624,
      "learning_rate": 3.6532620752175616e-07,
      "loss": 0.7576,
      "step": 11120
    },
    {
      "epoch": 0.9317706153202177,
      "grad_norm": 0.3953368365764618,
      "learning_rate": 2.9951118850133247e-07,
      "loss": 0.7757,
      "step": 11130
    },
    {
      "epoch": 0.9326077856843867,
      "grad_norm": 0.4290615916252136,
      "learning_rate": 2.429814731182774e-07,
      "loss": 0.7426,
      "step": 11140
    },
    {
      "epoch": 0.9334449560485559,
      "grad_norm": 0.39115163683891296,
      "learning_rate": 1.957908941169813e-07,
      "loss": 0.7622,
      "step": 11150
    },
    {
      "epoch": 0.934282126412725,
      "grad_norm": 0.3906017243862152,
      "learning_rate": 1.5798439066545254e-07,
      "loss": 0.7511,
      "step": 11160
    },
    {
      "epoch": 0.9351192967768941,
      "grad_norm": 0.3649786114692688,
      "learning_rate": 1.2959796556014471e-07,
      "loss": 0.7684,
      "step": 11170
    },
    {
      "epoch": 0.9359564671410632,
      "grad_norm": 0.382324755191803,
      "learning_rate": 1.1065865094081884e-07,
      "loss": 0.7334,
      "step": 11180
    },
    {
      "epoch": 0.9367936375052324,
      "grad_norm": 0.39119040966033936,
      "learning_rate": 1.011844825480734e-07,
      "loss": 0.7435,
      "step": 11190
    },
    {
      "epoch": 0.9371285056509,
      "step": 11194,
      "total_flos": 4.547514943786923e+19,
      "train_loss": 0.004281530875232745,
      "train_runtime": 630.7432,
      "train_samples_per_second": 212.968,
      "train_steps_per_second": 17.747
    }
  ],
  "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
}