{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5001340362791529,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008935751943526047,
      "grad_norm": 4.83393669128418,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 1.4311,
      "step": 10
    },
    {
      "epoch": 0.0017871503887052094,
      "grad_norm": 3.230992555618286,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.4013,
      "step": 20
    },
    {
      "epoch": 0.0026807255830578144,
      "grad_norm": 1.9909214973449707,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.1381,
      "step": 30
    },
    {
      "epoch": 0.003574300777410419,
      "grad_norm": 0.8147533535957336,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 0.9713,
      "step": 40
    },
    {
      "epoch": 0.004467875971763024,
      "grad_norm": 0.8770450949668884,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 0.8768,
      "step": 50
    },
    {
      "epoch": 0.005361451166115629,
      "grad_norm": 0.4910338222980499,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 0.8431,
      "step": 60
    },
    {
      "epoch": 0.006255026360468234,
      "grad_norm": 0.5245729684829712,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.7821,
      "step": 70
    },
    {
      "epoch": 0.007148601554820838,
      "grad_norm": 0.4985717535018921,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.8137,
      "step": 80
    },
    {
      "epoch": 0.008042176749173444,
      "grad_norm": 0.4982825517654419,
      "learning_rate": 1.005e-05,
      "loss": 0.8014,
      "step": 90
    },
    {
      "epoch": 0.008935751943526048,
      "grad_norm": 0.4962609112262726,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.8102,
      "step": 100
    },
    {
      "epoch": 0.009829327137878652,
      "grad_norm": 0.4608575701713562,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.7836,
      "step": 110
    },
    {
      "epoch": 0.010722902332231258,
      "grad_norm": 0.4405043423175812,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.7987,
      "step": 120
    },
    {
      "epoch": 0.011616477526583862,
      "grad_norm": 0.4832342565059662,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.757,
      "step": 130
    },
    {
      "epoch": 0.012510052720936467,
      "grad_norm": 0.5148149132728577,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.7922,
      "step": 140
    },
    {
      "epoch": 0.013403627915289072,
      "grad_norm": 0.496463418006897,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.8236,
      "step": 150
    },
    {
      "epoch": 0.014297203109641676,
      "grad_norm": 0.5295011401176453,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.798,
      "step": 160
    },
    {
      "epoch": 0.015190778303994281,
      "grad_norm": 0.47589513659477234,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.8071,
      "step": 170
    },
    {
      "epoch": 0.016084353498346887,
      "grad_norm": 0.42235517501831055,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 180
    },
    {
      "epoch": 0.01697792869269949,
      "grad_norm": 0.4666895270347595,
      "learning_rate": 2e-05,
      "loss": 0.7695,
      "step": 190
    },
    {
      "epoch": 0.017871503887052095,
      "grad_norm": 0.41975900530815125,
      "learning_rate": 2e-05,
      "loss": 0.7967,
      "step": 200
    },
    {
      "epoch": 0.0187650790814047,
      "grad_norm": 0.4837177097797394,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 210
    },
    {
      "epoch": 0.019658654275757304,
      "grad_norm": 0.422789603471756,
      "learning_rate": 2e-05,
      "loss": 0.7704,
      "step": 220
    },
    {
      "epoch": 0.02055222947010991,
      "grad_norm": 0.42405927181243896,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 230
    },
    {
      "epoch": 0.021445804664462515,
      "grad_norm": 0.44666311144828796,
      "learning_rate": 2e-05,
      "loss": 0.7747,
      "step": 240
    },
    {
      "epoch": 0.02233937985881512,
      "grad_norm": 0.413830041885376,
      "learning_rate": 2e-05,
      "loss": 0.7766,
      "step": 250
    },
    {
      "epoch": 0.023232955053167723,
      "grad_norm": 0.4343000054359436,
      "learning_rate": 2e-05,
      "loss": 0.8023,
      "step": 260
    },
    {
      "epoch": 0.024126530247520327,
      "grad_norm": 0.4592699110507965,
      "learning_rate": 2e-05,
      "loss": 0.7824,
      "step": 270
    },
    {
      "epoch": 0.025020105441872935,
      "grad_norm": 0.4644310474395752,
      "learning_rate": 2e-05,
      "loss": 0.7371,
      "step": 280
    },
    {
      "epoch": 0.02591368063622554,
      "grad_norm": 0.4319989085197449,
      "learning_rate": 2e-05,
      "loss": 0.7641,
      "step": 290
    },
    {
      "epoch": 0.026807255830578143,
      "grad_norm": 0.4405927360057831,
      "learning_rate": 2e-05,
      "loss": 0.7863,
      "step": 300
    },
    {
      "epoch": 0.027700831024930747,
      "grad_norm": 0.4252799451351166,
      "learning_rate": 2e-05,
      "loss": 0.7928,
      "step": 310
    },
    {
      "epoch": 0.02859440621928335,
      "grad_norm": 0.4097546637058258,
      "learning_rate": 2e-05,
      "loss": 0.7965,
      "step": 320
    },
    {
      "epoch": 0.02948798141363596,
      "grad_norm": 0.4354838728904724,
      "learning_rate": 2e-05,
      "loss": 0.7932,
      "step": 330
    },
    {
      "epoch": 0.030381556607988563,
      "grad_norm": 0.4520812928676605,
      "learning_rate": 2e-05,
      "loss": 0.793,
      "step": 340
    },
    {
      "epoch": 0.03127513180234117,
      "grad_norm": 0.5291610360145569,
      "learning_rate": 2e-05,
      "loss": 0.7466,
      "step": 350
    },
    {
      "epoch": 0.032168706996693774,
      "grad_norm": 0.442043662071228,
      "learning_rate": 2e-05,
      "loss": 0.7455,
      "step": 360
    },
    {
      "epoch": 0.03306228219104638,
      "grad_norm": 0.39435988664627075,
      "learning_rate": 2e-05,
      "loss": 0.7664,
      "step": 370
    },
    {
      "epoch": 0.03395585738539898,
      "grad_norm": 0.6453039050102234,
      "learning_rate": 2e-05,
      "loss": 0.7415,
      "step": 380
    },
    {
      "epoch": 0.03484943257975159,
      "grad_norm": 0.6479556560516357,
      "learning_rate": 2e-05,
      "loss": 0.8238,
      "step": 390
    },
    {
      "epoch": 0.03574300777410419,
      "grad_norm": 0.4728488624095917,
      "learning_rate": 2e-05,
      "loss": 0.7672,
      "step": 400
    },
    {
      "epoch": 0.036636582968456795,
      "grad_norm": 0.4299718141555786,
      "learning_rate": 2e-05,
      "loss": 0.7397,
      "step": 410
    },
    {
      "epoch": 0.0375301581628094,
      "grad_norm": 0.35715794563293457,
      "learning_rate": 2e-05,
      "loss": 0.7332,
      "step": 420
    },
    {
      "epoch": 0.038423733357162,
      "grad_norm": 0.430271178483963,
      "learning_rate": 2e-05,
      "loss": 0.7557,
      "step": 430
    },
    {
      "epoch": 0.03931730855151461,
      "grad_norm": 0.4169669449329376,
      "learning_rate": 2e-05,
      "loss": 0.7654,
      "step": 440
    },
    {
      "epoch": 0.04021088374586722,
      "grad_norm": 0.4232149124145508,
      "learning_rate": 2e-05,
      "loss": 0.7845,
      "step": 450
    },
    {
      "epoch": 0.04110445894021982,
      "grad_norm": 0.42479565739631653,
      "learning_rate": 2e-05,
      "loss": 0.7642,
      "step": 460
    },
    {
      "epoch": 0.041998034134572426,
      "grad_norm": 0.38824278116226196,
      "learning_rate": 2e-05,
      "loss": 0.7514,
      "step": 470
    },
    {
      "epoch": 0.04289160932892503,
      "grad_norm": 0.4207800626754761,
      "learning_rate": 2e-05,
      "loss": 0.7528,
      "step": 480
    },
    {
      "epoch": 0.043785184523277634,
      "grad_norm": 0.40183427929878235,
      "learning_rate": 2e-05,
      "loss": 0.7806,
      "step": 490
    },
    {
      "epoch": 0.04467875971763024,
      "grad_norm": 0.431792676448822,
      "learning_rate": 2e-05,
      "loss": 0.7755,
      "step": 500
    },
    {
      "epoch": 0.04557233491198284,
      "grad_norm": 0.4143838584423065,
      "learning_rate": 2e-05,
      "loss": 0.7539,
      "step": 510
    },
    {
      "epoch": 0.04646591010633545,
      "grad_norm": 0.44914576411247253,
      "learning_rate": 2e-05,
      "loss": 0.7638,
      "step": 520
    },
    {
      "epoch": 0.04735948530068805,
      "grad_norm": 0.3990241587162018,
      "learning_rate": 2e-05,
      "loss": 0.7566,
      "step": 530
    },
    {
      "epoch": 0.048253060495040655,
      "grad_norm": 0.4371657371520996,
      "learning_rate": 2e-05,
      "loss": 0.7551,
      "step": 540
    },
    {
      "epoch": 0.049146635689393266,
      "grad_norm": 0.45287543535232544,
      "learning_rate": 2e-05,
      "loss": 0.7469,
      "step": 550
    },
    {
      "epoch": 0.05004021088374587,
      "grad_norm": 0.3757151961326599,
      "learning_rate": 2e-05,
      "loss": 0.7494,
      "step": 560
    },
    {
      "epoch": 0.050933786078098474,
      "grad_norm": 0.5098564624786377,
      "learning_rate": 2e-05,
      "loss": 0.7648,
      "step": 570
    },
    {
      "epoch": 0.05182736127245108,
      "grad_norm": 0.43075209856033325,
      "learning_rate": 2e-05,
      "loss": 0.7296,
      "step": 580
    },
    {
      "epoch": 0.05272093646680368,
      "grad_norm": 0.46172618865966797,
      "learning_rate": 2e-05,
      "loss": 0.7439,
      "step": 590
    },
    {
      "epoch": 0.053614511661156286,
      "grad_norm": 0.4297853112220764,
      "learning_rate": 2e-05,
      "loss": 0.7748,
      "step": 600
    },
    {
      "epoch": 0.05450808685550889,
      "grad_norm": 0.39287275075912476,
      "learning_rate": 2e-05,
      "loss": 0.7353,
      "step": 610
    },
    {
      "epoch": 0.055401662049861494,
      "grad_norm": 0.3974439203739166,
      "learning_rate": 2e-05,
      "loss": 0.7451,
      "step": 620
    },
    {
      "epoch": 0.0562952372442141,
      "grad_norm": 0.39414021372795105,
      "learning_rate": 2e-05,
      "loss": 0.7739,
      "step": 630
    },
    {
      "epoch": 0.0571888124385667,
      "grad_norm": 0.42754384875297546,
      "learning_rate": 2e-05,
      "loss": 0.779,
      "step": 640
    },
    {
      "epoch": 0.05808238763291931,
      "grad_norm": 0.3786719739437103,
      "learning_rate": 2e-05,
      "loss": 0.7636,
      "step": 650
    },
    {
      "epoch": 0.05897596282727192,
      "grad_norm": 0.36745455861091614,
      "learning_rate": 2e-05,
      "loss": 0.7749,
      "step": 660
    },
    {
      "epoch": 0.05986953802162452,
      "grad_norm": 0.44898897409439087,
      "learning_rate": 2e-05,
      "loss": 0.7306,
      "step": 670
    },
    {
      "epoch": 0.060763113215977126,
      "grad_norm": 0.4336882531642914,
      "learning_rate": 2e-05,
      "loss": 0.7385,
      "step": 680
    },
    {
      "epoch": 0.06165668841032973,
      "grad_norm": 0.3571746349334717,
      "learning_rate": 2e-05,
      "loss": 0.7209,
      "step": 690
    },
    {
      "epoch": 0.06255026360468234,
      "grad_norm": 0.4294300079345703,
      "learning_rate": 2e-05,
      "loss": 0.7694,
      "step": 700
    },
    {
      "epoch": 0.06344383879903494,
      "grad_norm": 0.3911939859390259,
      "learning_rate": 2e-05,
      "loss": 0.7571,
      "step": 710
    },
    {
      "epoch": 0.06433741399338755,
      "grad_norm": 0.44994938373565674,
      "learning_rate": 2e-05,
      "loss": 0.7509,
      "step": 720
    },
    {
      "epoch": 0.06523098918774015,
      "grad_norm": 0.43563222885131836,
      "learning_rate": 2e-05,
      "loss": 0.7727,
      "step": 730
    },
    {
      "epoch": 0.06612456438209276,
      "grad_norm": 0.4137369394302368,
      "learning_rate": 2e-05,
      "loss": 0.7598,
      "step": 740
    },
    {
      "epoch": 0.06701813957644535,
      "grad_norm": 0.3853653073310852,
      "learning_rate": 2e-05,
      "loss": 0.7585,
      "step": 750
    },
    {
      "epoch": 0.06791171477079797,
      "grad_norm": 0.3681003153324127,
      "learning_rate": 2e-05,
      "loss": 0.7394,
      "step": 760
    },
    {
      "epoch": 0.06880528996515056,
      "grad_norm": 0.3928927183151245,
      "learning_rate": 2e-05,
      "loss": 0.7484,
      "step": 770
    },
    {
      "epoch": 0.06969886515950317,
      "grad_norm": 0.387007474899292,
      "learning_rate": 2e-05,
      "loss": 0.7262,
      "step": 780
    },
    {
      "epoch": 0.07059244035385577,
      "grad_norm": 0.46645933389663696,
      "learning_rate": 2e-05,
      "loss": 0.7281,
      "step": 790
    },
    {
      "epoch": 0.07148601554820838,
      "grad_norm": 0.4278911352157593,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 800
    },
    {
      "epoch": 0.07237959074256099,
      "grad_norm": 0.39950302243232727,
      "learning_rate": 2e-05,
      "loss": 0.7012,
      "step": 810
    },
    {
      "epoch": 0.07327316593691359,
      "grad_norm": 0.3807908892631531,
      "learning_rate": 2e-05,
      "loss": 0.7324,
      "step": 820
    },
    {
      "epoch": 0.0741667411312662,
      "grad_norm": 0.4291982352733612,
      "learning_rate": 2e-05,
      "loss": 0.7244,
      "step": 830
    },
    {
      "epoch": 0.0750603163256188,
      "grad_norm": 0.40053242444992065,
      "learning_rate": 2e-05,
      "loss": 0.7638,
      "step": 840
    },
    {
      "epoch": 0.07595389151997141,
      "grad_norm": 0.4022897481918335,
      "learning_rate": 2e-05,
      "loss": 0.753,
      "step": 850
    },
    {
      "epoch": 0.076847466714324,
      "grad_norm": 0.5166751146316528,
      "learning_rate": 2e-05,
      "loss": 0.7418,
      "step": 860
    },
    {
      "epoch": 0.07774104190867662,
      "grad_norm": 0.4001097083091736,
      "learning_rate": 2e-05,
      "loss": 0.7557,
      "step": 870
    },
    {
      "epoch": 0.07863461710302921,
      "grad_norm": 0.37985655665397644,
      "learning_rate": 2e-05,
      "loss": 0.7429,
      "step": 880
    },
    {
      "epoch": 0.07952819229738183,
      "grad_norm": 0.3874037563800812,
      "learning_rate": 2e-05,
      "loss": 0.7409,
      "step": 890
    },
    {
      "epoch": 0.08042176749173444,
      "grad_norm": 0.40059277415275574,
      "learning_rate": 2e-05,
      "loss": 0.7468,
      "step": 900
    },
    {
      "epoch": 0.08131534268608703,
      "grad_norm": 0.3915744423866272,
      "learning_rate": 2e-05,
      "loss": 0.7425,
      "step": 910
    },
    {
      "epoch": 0.08220891788043964,
      "grad_norm": 0.4296461343765259,
      "learning_rate": 2e-05,
      "loss": 0.7538,
      "step": 920
    },
    {
      "epoch": 0.08310249307479224,
      "grad_norm": 0.39975130558013916,
      "learning_rate": 2e-05,
      "loss": 0.7751,
      "step": 930
    },
    {
      "epoch": 0.08399606826914485,
      "grad_norm": 0.4234040379524231,
      "learning_rate": 2e-05,
      "loss": 0.7217,
      "step": 940
    },
    {
      "epoch": 0.08488964346349745,
      "grad_norm": 0.7239550352096558,
      "learning_rate": 2e-05,
      "loss": 0.7523,
      "step": 950
    },
    {
      "epoch": 0.08578321865785006,
      "grad_norm": 0.40926212072372437,
      "learning_rate": 2e-05,
      "loss": 0.7073,
      "step": 960
    },
    {
      "epoch": 0.08667679385220266,
      "grad_norm": 0.42253273725509644,
      "learning_rate": 2e-05,
      "loss": 0.7645,
      "step": 970
    },
    {
      "epoch": 0.08757036904655527,
      "grad_norm": 0.42166000604629517,
      "learning_rate": 2e-05,
      "loss": 0.7206,
      "step": 980
    },
    {
      "epoch": 0.08846394424090787,
      "grad_norm": 0.4257868826389313,
      "learning_rate": 2e-05,
      "loss": 0.7336,
      "step": 990
    },
    {
      "epoch": 0.08935751943526048,
      "grad_norm": 0.3908492922782898,
      "learning_rate": 2e-05,
      "loss": 0.7297,
      "step": 1000
    },
    {
      "epoch": 0.09025109462961309,
      "grad_norm": 0.38620543479919434,
      "learning_rate": 2e-05,
      "loss": 0.7451,
      "step": 1010
    },
    {
      "epoch": 0.09114466982396568,
      "grad_norm": 0.40934133529663086,
      "learning_rate": 2e-05,
      "loss": 0.7403,
      "step": 1020
    },
    {
      "epoch": 0.0920382450183183,
      "grad_norm": 0.435519814491272,
      "learning_rate": 2e-05,
      "loss": 0.7144,
      "step": 1030
    },
    {
      "epoch": 0.0929318202126709,
      "grad_norm": 0.46169203519821167,
      "learning_rate": 2e-05,
      "loss": 0.7317,
      "step": 1040
    },
    {
      "epoch": 0.0938253954070235,
      "grad_norm": 0.37340661883354187,
      "learning_rate": 2e-05,
      "loss": 0.7714,
      "step": 1050
    },
    {
      "epoch": 0.0947189706013761,
      "grad_norm": 0.41289299726486206,
      "learning_rate": 2e-05,
      "loss": 0.7351,
      "step": 1060
    },
    {
      "epoch": 0.09561254579572871,
      "grad_norm": 0.4187810719013214,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 1070
    },
    {
      "epoch": 0.09650612099008131,
      "grad_norm": 0.41718336939811707,
      "learning_rate": 2e-05,
      "loss": 0.7864,
      "step": 1080
    },
    {
      "epoch": 0.09739969618443392,
      "grad_norm": 0.4145309329032898,
      "learning_rate": 2e-05,
      "loss": 0.7505,
      "step": 1090
    },
    {
      "epoch": 0.09829327137878653,
      "grad_norm": 0.4168350100517273,
      "learning_rate": 2e-05,
      "loss": 0.7359,
      "step": 1100
    },
    {
      "epoch": 0.09918684657313913,
      "grad_norm": 0.39129555225372314,
      "learning_rate": 2e-05,
      "loss": 0.7321,
      "step": 1110
    },
    {
      "epoch": 0.10008042176749174,
      "grad_norm": 0.39075788855552673,
      "learning_rate": 2e-05,
      "loss": 0.7625,
      "step": 1120
    },
    {
      "epoch": 0.10097399696184434,
      "grad_norm": 0.4408645033836365,
      "learning_rate": 2e-05,
      "loss": 0.7388,
      "step": 1130
    },
    {
      "epoch": 0.10186757215619695,
      "grad_norm": 0.3980259597301483,
      "learning_rate": 2e-05,
      "loss": 0.7537,
      "step": 1140
    },
    {
      "epoch": 0.10276114735054954,
      "grad_norm": 0.3763454556465149,
      "learning_rate": 2e-05,
      "loss": 0.7488,
      "step": 1150
    },
    {
      "epoch": 0.10365472254490216,
      "grad_norm": 0.37730491161346436,
      "learning_rate": 2e-05,
      "loss": 0.7465,
      "step": 1160
    },
    {
      "epoch": 0.10454829773925475,
      "grad_norm": 0.37538275122642517,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 1170
    },
    {
      "epoch": 0.10544187293360736,
      "grad_norm": 0.397089421749115,
      "learning_rate": 2e-05,
      "loss": 0.769,
      "step": 1180
    },
    {
      "epoch": 0.10633544812795996,
      "grad_norm": 0.3705765902996063,
      "learning_rate": 2e-05,
      "loss": 0.7385,
      "step": 1190
    },
    {
      "epoch": 0.10722902332231257,
      "grad_norm": 0.3545670807361603,
      "learning_rate": 2e-05,
      "loss": 0.7408,
      "step": 1200
    },
    {
      "epoch": 0.10812259851666518,
      "grad_norm": 0.4176211357116699,
      "learning_rate": 2e-05,
      "loss": 0.7409,
      "step": 1210
    },
    {
      "epoch": 0.10901617371101778,
      "grad_norm": 0.36956384778022766,
      "learning_rate": 2e-05,
      "loss": 0.7246,
      "step": 1220
    },
    {
      "epoch": 0.10990974890537039,
      "grad_norm": 0.3938690423965454,
      "learning_rate": 2e-05,
      "loss": 0.7292,
      "step": 1230
    },
    {
      "epoch": 0.11080332409972299,
      "grad_norm": 0.39145007729530334,
      "learning_rate": 2e-05,
      "loss": 0.7285,
      "step": 1240
    },
    {
      "epoch": 0.1116968992940756,
      "grad_norm": 0.38112127780914307,
      "learning_rate": 2e-05,
      "loss": 0.7551,
      "step": 1250
    },
    {
      "epoch": 0.1125904744884282,
      "grad_norm": 0.3316250741481781,
      "learning_rate": 2e-05,
      "loss": 0.7323,
      "step": 1260
    },
    {
      "epoch": 0.11348404968278081,
      "grad_norm": 0.4049963057041168,
      "learning_rate": 2e-05,
      "loss": 0.7329,
      "step": 1270
    },
    {
      "epoch": 0.1143776248771334,
      "grad_norm": 0.48271071910858154,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 1280
    },
    {
      "epoch": 0.11527120007148602,
      "grad_norm": 0.37332484126091003,
      "learning_rate": 2e-05,
      "loss": 0.7045,
      "step": 1290
    },
    {
      "epoch": 0.11616477526583863,
      "grad_norm": 0.3976444602012634,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 1300
    },
    {
      "epoch": 0.11705835046019122,
      "grad_norm": 0.39804327487945557,
      "learning_rate": 2e-05,
      "loss": 0.7406,
      "step": 1310
    },
    {
      "epoch": 0.11795192565454383,
      "grad_norm": 0.4065830707550049,
      "learning_rate": 2e-05,
      "loss": 0.7332,
      "step": 1320
    },
    {
      "epoch": 0.11884550084889643,
      "grad_norm": 0.38445714116096497,
      "learning_rate": 2e-05,
      "loss": 0.7277,
      "step": 1330
    },
    {
      "epoch": 0.11973907604324904,
      "grad_norm": 0.3987727761268616,
      "learning_rate": 2e-05,
      "loss": 0.6999,
      "step": 1340
    },
    {
      "epoch": 0.12063265123760164,
      "grad_norm": 0.37083423137664795,
      "learning_rate": 2e-05,
      "loss": 0.7425,
      "step": 1350
    },
    {
      "epoch": 0.12152622643195425,
      "grad_norm": 0.4784857928752899,
      "learning_rate": 2e-05,
      "loss": 0.7522,
      "step": 1360
    },
    {
      "epoch": 0.12241980162630685,
      "grad_norm": 0.39088118076324463,
      "learning_rate": 2e-05,
      "loss": 0.7366,
      "step": 1370
    },
    {
      "epoch": 0.12331337682065946,
      "grad_norm": 0.39330583810806274,
      "learning_rate": 2e-05,
      "loss": 0.7293,
      "step": 1380
    },
    {
      "epoch": 0.12420695201501206,
      "grad_norm": 0.3860457241535187,
      "learning_rate": 2e-05,
      "loss": 0.7143,
      "step": 1390
    },
    {
      "epoch": 0.12510052720936468,
      "grad_norm": 0.38415059447288513,
      "learning_rate": 2e-05,
      "loss": 0.6834,
      "step": 1400
    },
    {
      "epoch": 0.12599410240371728,
      "grad_norm": 0.39965566992759705,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 1410
    },
    {
      "epoch": 0.12688767759806988,
      "grad_norm": 0.38492318987846375,
      "learning_rate": 2e-05,
      "loss": 0.7547,
      "step": 1420
    },
    {
      "epoch": 0.12778125279242247,
      "grad_norm": 0.41795384883880615,
      "learning_rate": 2e-05,
      "loss": 0.7267,
      "step": 1430
    },
    {
      "epoch": 0.1286748279867751,
      "grad_norm": 0.3837740123271942,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 1440
    },
    {
      "epoch": 0.1295684031811277,
      "grad_norm": 0.35027605295181274,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 1450
    },
    {
      "epoch": 0.1304619783754803,
      "grad_norm": 0.36292174458503723,
      "learning_rate": 2e-05,
      "loss": 0.7441,
      "step": 1460
    },
    {
      "epoch": 0.1313555535698329,
      "grad_norm": 0.3984721899032593,
      "learning_rate": 2e-05,
      "loss": 0.7197,
      "step": 1470
    },
    {
      "epoch": 0.13224912876418551,
      "grad_norm": 0.4132763147354126,
      "learning_rate": 2e-05,
      "loss": 0.7457,
      "step": 1480
    },
    {
      "epoch": 0.1331427039585381,
      "grad_norm": 0.3790658712387085,
      "learning_rate": 2e-05,
      "loss": 0.7274,
      "step": 1490
    },
    {
      "epoch": 0.1340362791528907,
      "grad_norm": 0.3779606819152832,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 1500
    },
    {
      "epoch": 0.13492985434724333,
      "grad_norm": 0.9285029172897339,
      "learning_rate": 2e-05,
      "loss": 0.7231,
      "step": 1510
    },
    {
      "epoch": 0.13582342954159593,
      "grad_norm": 0.3995453119277954,
      "learning_rate": 2e-05,
      "loss": 0.7022,
      "step": 1520
    },
    {
      "epoch": 0.13671700473594853,
      "grad_norm": 0.4919504225254059,
      "learning_rate": 2e-05,
      "loss": 0.7461,
      "step": 1530
    },
    {
      "epoch": 0.13761057993030112,
      "grad_norm": 0.37025532126426697,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 1540
    },
    {
      "epoch": 0.13850415512465375,
      "grad_norm": 0.3523809611797333,
      "learning_rate": 2e-05,
      "loss": 0.7124,
      "step": 1550
    },
    {
      "epoch": 0.13939773031900635,
      "grad_norm": 0.4186336398124695,
      "learning_rate": 2e-05,
      "loss": 0.724,
      "step": 1560
    },
    {
      "epoch": 0.14029130551335894,
      "grad_norm": 0.39147084951400757,
      "learning_rate": 2e-05,
      "loss": 0.7444,
      "step": 1570
    },
    {
      "epoch": 0.14118488070771154,
      "grad_norm": 0.3777700364589691,
      "learning_rate": 2e-05,
      "loss": 0.7383,
      "step": 1580
    },
    {
      "epoch": 0.14207845590206417,
      "grad_norm": 0.41446128487586975,
      "learning_rate": 2e-05,
      "loss": 0.735,
      "step": 1590
    },
    {
      "epoch": 0.14297203109641676,
      "grad_norm": 0.38765832781791687,
      "learning_rate": 2e-05,
      "loss": 0.7147,
      "step": 1600
    },
    {
      "epoch": 0.14386560629076936,
      "grad_norm": 0.5072250962257385,
      "learning_rate": 2e-05,
      "loss": 0.7163,
      "step": 1610
    },
    {
      "epoch": 0.14475918148512198,
      "grad_norm": 0.3880365490913391,
      "learning_rate": 2e-05,
      "loss": 0.7141,
      "step": 1620
    },
    {
      "epoch": 0.14565275667947458,
      "grad_norm": 0.35895800590515137,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 1630
    },
    {
      "epoch": 0.14654633187382718,
      "grad_norm": 0.35093289613723755,
      "learning_rate": 2e-05,
      "loss": 0.7359,
      "step": 1640
    },
    {
      "epoch": 0.14743990706817978,
      "grad_norm": 0.3536919951438904,
      "learning_rate": 2e-05,
      "loss": 0.739,
      "step": 1650
    },
    {
      "epoch": 0.1483334822625324,
      "grad_norm": 0.6556247472763062,
      "learning_rate": 2e-05,
      "loss": 0.7185,
      "step": 1660
    },
    {
      "epoch": 0.149227057456885,
      "grad_norm": 0.3924124538898468,
      "learning_rate": 2e-05,
      "loss": 0.747,
      "step": 1670
    },
    {
      "epoch": 0.1501206326512376,
      "grad_norm": 0.8066271543502808,
      "learning_rate": 2e-05,
      "loss": 0.7584,
      "step": 1680
    },
    {
      "epoch": 0.1510142078455902,
      "grad_norm": 0.4922969341278076,
      "learning_rate": 2e-05,
      "loss": 0.7044,
      "step": 1690
    },
    {
      "epoch": 0.15190778303994282,
      "grad_norm": 0.34611326456069946,
      "learning_rate": 2e-05,
      "loss": 0.7443,
      "step": 1700
    },
    {
      "epoch": 0.15280135823429541,
      "grad_norm": 0.3866153657436371,
      "learning_rate": 2e-05,
      "loss": 0.725,
      "step": 1710
    },
    {
      "epoch": 0.153694933428648,
      "grad_norm": 0.36212798953056335,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 1720
    },
    {
      "epoch": 0.15458850862300064,
      "grad_norm": 0.3953538239002228,
      "learning_rate": 2e-05,
      "loss": 0.7347,
      "step": 1730
    },
    {
      "epoch": 0.15548208381735323,
      "grad_norm": 0.4102666676044464,
      "learning_rate": 2e-05,
      "loss": 0.736,
      "step": 1740
    },
    {
      "epoch": 0.15637565901170583,
      "grad_norm": 0.4497690498828888,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 1750
    },
    {
      "epoch": 0.15726923420605843,
      "grad_norm": 0.4439946711063385,
      "learning_rate": 2e-05,
      "loss": 0.7347,
      "step": 1760
    },
    {
      "epoch": 0.15816280940041105,
      "grad_norm": 1.074998140335083,
      "learning_rate": 2e-05,
      "loss": 0.7432,
      "step": 1770
    },
    {
      "epoch": 0.15905638459476365,
      "grad_norm": 0.3673039972782135,
      "learning_rate": 2e-05,
      "loss": 0.7465,
      "step": 1780
    },
    {
      "epoch": 0.15994995978911625,
      "grad_norm": 0.3478325605392456,
      "learning_rate": 2e-05,
      "loss": 0.7225,
      "step": 1790
    },
    {
      "epoch": 0.16084353498346887,
      "grad_norm": 0.3802620470523834,
      "learning_rate": 2e-05,
      "loss": 0.6983,
      "step": 1800
    },
    {
      "epoch": 0.16173711017782147,
      "grad_norm": 0.38965746760368347,
      "learning_rate": 2e-05,
      "loss": 0.712,
      "step": 1810
    },
    {
      "epoch": 0.16263068537217407,
      "grad_norm": 0.3660908043384552,
      "learning_rate": 2e-05,
      "loss": 0.7198,
      "step": 1820
    },
    {
      "epoch": 0.16352426056652666,
      "grad_norm": 0.38565513491630554,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 1830
    },
    {
      "epoch": 0.1644178357608793,
      "grad_norm": 0.49884864687919617,
      "learning_rate": 2e-05,
      "loss": 0.7378,
      "step": 1840
    },
    {
      "epoch": 0.16531141095523189,
      "grad_norm": 0.3706195652484894,
      "learning_rate": 2e-05,
      "loss": 0.7051,
      "step": 1850
    },
    {
      "epoch": 0.16620498614958448,
      "grad_norm": 0.3866501450538635,
      "learning_rate": 2e-05,
      "loss": 0.7321,
      "step": 1860
    },
    {
      "epoch": 0.16709856134393708,
      "grad_norm": 0.38853004574775696,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 1870
    },
    {
      "epoch": 0.1679921365382897,
      "grad_norm": 0.42117682099342346,
      "learning_rate": 2e-05,
      "loss": 0.7241,
      "step": 1880
    },
    {
      "epoch": 0.1688857117326423,
      "grad_norm": 0.4492304027080536,
      "learning_rate": 2e-05,
      "loss": 0.7349,
      "step": 1890
    },
    {
      "epoch": 0.1697792869269949,
      "grad_norm": 0.4259665608406067,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 1900
    },
    {
      "epoch": 0.17067286212134752,
      "grad_norm": 0.3557761609554291,
      "learning_rate": 2e-05,
      "loss": 0.6986,
      "step": 1910
    },
    {
      "epoch": 0.17156643731570012,
      "grad_norm": 0.35731881856918335,
      "learning_rate": 2e-05,
      "loss": 0.7351,
      "step": 1920
    },
    {
      "epoch": 0.17246001251005272,
      "grad_norm": 0.38917088508605957,
      "learning_rate": 2e-05,
      "loss": 0.721,
      "step": 1930
    },
    {
      "epoch": 0.17335358770440532,
      "grad_norm": 0.39296889305114746,
      "learning_rate": 2e-05,
      "loss": 0.7457,
      "step": 1940
    },
    {
      "epoch": 0.17424716289875794,
      "grad_norm": 0.3701498508453369,
      "learning_rate": 2e-05,
      "loss": 0.7699,
      "step": 1950
    },
    {
      "epoch": 0.17514073809311054,
      "grad_norm": 0.3800022006034851,
      "learning_rate": 2e-05,
      "loss": 0.7222,
      "step": 1960
    },
    {
      "epoch": 0.17603431328746313,
      "grad_norm": 0.3638252913951874,
      "learning_rate": 2e-05,
      "loss": 0.6871,
      "step": 1970
    },
    {
      "epoch": 0.17692788848181573,
      "grad_norm": 0.38935279846191406,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 1980
    },
    {
      "epoch": 0.17782146367616836,
      "grad_norm": 0.40048500895500183,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 1990
    },
    {
      "epoch": 0.17871503887052095,
      "grad_norm": 0.41115978360176086,
      "learning_rate": 2e-05,
      "loss": 0.7409,
      "step": 2000
    },
    {
      "epoch": 0.17960861406487355,
      "grad_norm": 0.39649268984794617,
      "learning_rate": 2e-05,
      "loss": 0.7172,
      "step": 2010
    },
    {
      "epoch": 0.18050218925922618,
      "grad_norm": 0.5061083436012268,
      "learning_rate": 2e-05,
      "loss": 0.7264,
      "step": 2020
    },
    {
      "epoch": 0.18139576445357877,
      "grad_norm": 0.38063013553619385,
      "learning_rate": 2e-05,
      "loss": 0.7434,
      "step": 2030
    },
    {
      "epoch": 0.18228933964793137,
      "grad_norm": 0.4168822169303894,
      "learning_rate": 2e-05,
      "loss": 0.7063,
      "step": 2040
    },
    {
      "epoch": 0.18318291484228397,
      "grad_norm": 0.36266323924064636,
      "learning_rate": 2e-05,
      "loss": 0.7401,
      "step": 2050
    },
    {
      "epoch": 0.1840764900366366,
      "grad_norm": 0.37227675318717957,
      "learning_rate": 2e-05,
      "loss": 0.7144,
      "step": 2060
    },
    {
      "epoch": 0.1849700652309892,
      "grad_norm": 0.39245229959487915,
      "learning_rate": 2e-05,
      "loss": 0.7084,
      "step": 2070
    },
    {
      "epoch": 0.1858636404253418,
      "grad_norm": 0.37118712067604065,
      "learning_rate": 2e-05,
      "loss": 0.7217,
      "step": 2080
    },
    {
      "epoch": 0.1867572156196944,
      "grad_norm": 0.3615480661392212,
      "learning_rate": 2e-05,
      "loss": 0.7143,
      "step": 2090
    },
    {
      "epoch": 0.187650790814047,
      "grad_norm": 0.4019179046154022,
      "learning_rate": 2e-05,
      "loss": 0.7449,
      "step": 2100
    },
    {
      "epoch": 0.1885443660083996,
      "grad_norm": 0.38081419467926025,
      "learning_rate": 2e-05,
      "loss": 0.724,
      "step": 2110
    },
    {
      "epoch": 0.1894379412027522,
      "grad_norm": 0.39533179998397827,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 2120
    },
    {
      "epoch": 0.19033151639710483,
      "grad_norm": 0.4144379794597626,
      "learning_rate": 2e-05,
      "loss": 0.7177,
      "step": 2130
    },
    {
      "epoch": 0.19122509159145742,
      "grad_norm": 0.38099774718284607,
      "learning_rate": 2e-05,
      "loss": 0.7035,
      "step": 2140
    },
    {
      "epoch": 0.19211866678581002,
      "grad_norm": 0.36082664132118225,
      "learning_rate": 2e-05,
      "loss": 0.729,
      "step": 2150
    },
    {
      "epoch": 0.19301224198016262,
      "grad_norm": 0.38504886627197266,
      "learning_rate": 2e-05,
      "loss": 0.7049,
      "step": 2160
    },
    {
      "epoch": 0.19390581717451524,
      "grad_norm": 0.3863053619861603,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 2170
    },
    {
      "epoch": 0.19479939236886784,
      "grad_norm": 0.3692690432071686,
      "learning_rate": 2e-05,
      "loss": 0.7298,
      "step": 2180
    },
    {
      "epoch": 0.19569296756322044,
      "grad_norm": 0.3638010621070862,
      "learning_rate": 2e-05,
      "loss": 0.7256,
      "step": 2190
    },
    {
      "epoch": 0.19658654275757306,
      "grad_norm": 0.3824363648891449,
      "learning_rate": 2e-05,
      "loss": 0.7376,
      "step": 2200
    },
    {
      "epoch": 0.19748011795192566,
      "grad_norm": 0.35318830609321594,
      "learning_rate": 2e-05,
      "loss": 0.736,
      "step": 2210
    },
    {
      "epoch": 0.19837369314627826,
      "grad_norm": 0.35539188981056213,
      "learning_rate": 2e-05,
      "loss": 0.7027,
      "step": 2220
    },
    {
      "epoch": 0.19926726834063085,
      "grad_norm": 0.414533406496048,
      "learning_rate": 2e-05,
      "loss": 0.6909,
      "step": 2230
    },
    {
      "epoch": 0.20016084353498348,
      "grad_norm": 0.39531072974205017,
      "learning_rate": 2e-05,
      "loss": 0.7301,
      "step": 2240
    },
    {
      "epoch": 0.20105441872933608,
      "grad_norm": 0.3679538071155548,
      "learning_rate": 2e-05,
      "loss": 0.7203,
      "step": 2250
    },
    {
      "epoch": 0.20194799392368867,
      "grad_norm": 0.37164831161499023,
      "learning_rate": 2e-05,
      "loss": 0.725,
      "step": 2260
    },
    {
      "epoch": 0.20284156911804127,
      "grad_norm": 0.3633818030357361,
      "learning_rate": 2e-05,
      "loss": 0.7192,
      "step": 2270
    },
    {
      "epoch": 0.2037351443123939,
      "grad_norm": 0.3682442307472229,
      "learning_rate": 2e-05,
      "loss": 0.7099,
      "step": 2280
    },
    {
      "epoch": 0.2046287195067465,
      "grad_norm": 0.34997084736824036,
      "learning_rate": 2e-05,
      "loss": 0.7543,
      "step": 2290
    },
    {
      "epoch": 0.2055222947010991,
      "grad_norm": 0.403976172208786,
      "learning_rate": 2e-05,
      "loss": 0.7511,
      "step": 2300
    },
    {
      "epoch": 0.20641586989545171,
      "grad_norm": 0.41093239188194275,
      "learning_rate": 2e-05,
      "loss": 0.6794,
      "step": 2310
    },
    {
      "epoch": 0.2073094450898043,
      "grad_norm": 0.38257455825805664,
      "learning_rate": 2e-05,
      "loss": 0.7263,
      "step": 2320
    },
    {
      "epoch": 0.2082030202841569,
      "grad_norm": 0.3526884615421295,
      "learning_rate": 2e-05,
      "loss": 0.7185,
      "step": 2330
    },
    {
      "epoch": 0.2090965954785095,
      "grad_norm": 0.3658798336982727,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 2340
    },
    {
      "epoch": 0.20999017067286213,
      "grad_norm": 0.3977317214012146,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 2350
    },
    {
      "epoch": 0.21088374586721473,
      "grad_norm": 0.386541485786438,
      "learning_rate": 2e-05,
      "loss": 0.7196,
      "step": 2360
    },
    {
      "epoch": 0.21177732106156733,
      "grad_norm": 0.39826154708862305,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 2370
    },
    {
      "epoch": 0.21267089625591992,
      "grad_norm": 0.4097355902194977,
      "learning_rate": 2e-05,
      "loss": 0.7117,
      "step": 2380
    },
    {
      "epoch": 0.21356447145027255,
      "grad_norm": 0.3464928865432739,
      "learning_rate": 2e-05,
      "loss": 0.7126,
      "step": 2390
    },
    {
      "epoch": 0.21445804664462514,
      "grad_norm": 0.397562712430954,
      "learning_rate": 2e-05,
      "loss": 0.7197,
      "step": 2400
    },
    {
      "epoch": 0.21535162183897774,
      "grad_norm": 0.35849079489707947,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 2410
    },
    {
      "epoch": 0.21624519703333037,
      "grad_norm": 0.5133382081985474,
      "learning_rate": 2e-05,
      "loss": 0.7267,
      "step": 2420
    },
    {
      "epoch": 0.21713877222768296,
      "grad_norm": 0.37571123242378235,
      "learning_rate": 2e-05,
      "loss": 0.7223,
      "step": 2430
    },
    {
      "epoch": 0.21803234742203556,
      "grad_norm": 0.3847047686576843,
      "learning_rate": 2e-05,
      "loss": 0.7139,
      "step": 2440
    },
    {
      "epoch": 0.21892592261638816,
      "grad_norm": 0.3535878360271454,
      "learning_rate": 2e-05,
      "loss": 0.6956,
      "step": 2450
    },
    {
      "epoch": 0.21981949781074078,
      "grad_norm": 0.3725936710834503,
      "learning_rate": 2e-05,
      "loss": 0.7107,
      "step": 2460
    },
    {
      "epoch": 0.22071307300509338,
      "grad_norm": 0.4022907316684723,
      "learning_rate": 2e-05,
      "loss": 0.7524,
      "step": 2470
    },
    {
      "epoch": 0.22160664819944598,
      "grad_norm": 0.3813494145870209,
      "learning_rate": 2e-05,
      "loss": 0.7229,
      "step": 2480
    },
    {
      "epoch": 0.2225002233937986,
      "grad_norm": 0.38350072503089905,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 2490
    },
    {
      "epoch": 0.2233937985881512,
      "grad_norm": 0.3981742858886719,
      "learning_rate": 2e-05,
      "loss": 0.7362,
      "step": 2500
    },
    {
      "epoch": 0.2242873737825038,
      "grad_norm": 0.3835667371749878,
      "learning_rate": 2e-05,
      "loss": 0.7107,
      "step": 2510
    },
    {
      "epoch": 0.2251809489768564,
      "grad_norm": 0.372981458902359,
      "learning_rate": 2e-05,
      "loss": 0.7333,
      "step": 2520
    },
    {
      "epoch": 0.22607452417120902,
      "grad_norm": 0.35782164335250854,
      "learning_rate": 2e-05,
      "loss": 0.6753,
      "step": 2530
    },
    {
      "epoch": 0.22696809936556162,
      "grad_norm": 0.36539599299430847,
      "learning_rate": 2e-05,
      "loss": 0.6941,
      "step": 2540
    },
    {
      "epoch": 0.2278616745599142,
      "grad_norm": 0.3650493919849396,
      "learning_rate": 2e-05,
      "loss": 0.7003,
      "step": 2550
    },
    {
      "epoch": 0.2287552497542668,
      "grad_norm": 0.3873494863510132,
      "learning_rate": 2e-05,
      "loss": 0.7172,
      "step": 2560
    },
    {
      "epoch": 0.22964882494861943,
      "grad_norm": 0.3694900870323181,
      "learning_rate": 2e-05,
      "loss": 0.7238,
      "step": 2570
    },
    {
      "epoch": 0.23054240014297203,
      "grad_norm": 0.3531987965106964,
      "learning_rate": 2e-05,
      "loss": 0.7374,
      "step": 2580
    },
    {
      "epoch": 0.23143597533732463,
      "grad_norm": 0.3426517844200134,
      "learning_rate": 2e-05,
      "loss": 0.7285,
      "step": 2590
    },
    {
      "epoch": 0.23232955053167725,
      "grad_norm": 0.3928007185459137,
      "learning_rate": 2e-05,
      "loss": 0.7108,
      "step": 2600
    },
    {
      "epoch": 0.23322312572602985,
      "grad_norm": 0.37236160039901733,
      "learning_rate": 2e-05,
      "loss": 0.7158,
      "step": 2610
    },
    {
      "epoch": 0.23411670092038245,
      "grad_norm": 0.38145866990089417,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 2620
    },
    {
      "epoch": 0.23501027611473504,
      "grad_norm": 0.3722963333129883,
      "learning_rate": 2e-05,
      "loss": 0.7169,
      "step": 2630
    },
    {
      "epoch": 0.23590385130908767,
      "grad_norm": 0.42692744731903076,
      "learning_rate": 2e-05,
      "loss": 0.7264,
      "step": 2640
    },
    {
      "epoch": 0.23679742650344027,
      "grad_norm": 0.37643131613731384,
      "learning_rate": 2e-05,
      "loss": 0.7307,
      "step": 2650
    },
    {
      "epoch": 0.23769100169779286,
      "grad_norm": 0.3665594756603241,
      "learning_rate": 2e-05,
      "loss": 0.7002,
      "step": 2660
    },
    {
      "epoch": 0.23858457689214546,
      "grad_norm": 0.4624324142932892,
      "learning_rate": 2e-05,
      "loss": 0.7116,
      "step": 2670
    },
    {
      "epoch": 0.2394781520864981,
      "grad_norm": 0.3787866234779358,
      "learning_rate": 2e-05,
      "loss": 0.7084,
      "step": 2680
    },
    {
      "epoch": 0.24037172728085068,
      "grad_norm": 0.3793836236000061,
      "learning_rate": 2e-05,
      "loss": 0.7268,
      "step": 2690
    },
    {
      "epoch": 0.24126530247520328,
      "grad_norm": 0.3420543670654297,
      "learning_rate": 2e-05,
      "loss": 0.7109,
      "step": 2700
    },
    {
      "epoch": 0.2421588776695559,
      "grad_norm": 0.3823523223400116,
      "learning_rate": 2e-05,
      "loss": 0.7257,
      "step": 2710
    },
    {
      "epoch": 0.2430524528639085,
      "grad_norm": 0.35775208473205566,
      "learning_rate": 2e-05,
      "loss": 0.6907,
      "step": 2720
    },
    {
      "epoch": 0.2439460280582611,
      "grad_norm": 0.36065220832824707,
      "learning_rate": 2e-05,
      "loss": 0.7154,
      "step": 2730
    },
    {
      "epoch": 0.2448396032526137,
      "grad_norm": 0.3732667565345764,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 2740
    },
    {
      "epoch": 0.24573317844696632,
      "grad_norm": 0.34359318017959595,
      "learning_rate": 2e-05,
      "loss": 0.7244,
      "step": 2750
    },
    {
      "epoch": 0.24662675364131892,
      "grad_norm": 0.35983023047447205,
      "learning_rate": 2e-05,
      "loss": 0.6753,
      "step": 2760
    },
    {
      "epoch": 0.24752032883567152,
      "grad_norm": 0.37808793783187866,
      "learning_rate": 2e-05,
      "loss": 0.7379,
      "step": 2770
    },
    {
      "epoch": 0.2484139040300241,
      "grad_norm": 0.38024666905403137,
      "learning_rate": 2e-05,
      "loss": 0.7081,
      "step": 2780
    },
    {
      "epoch": 0.24930747922437674,
      "grad_norm": 0.3341760039329529,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 2790
    },
    {
      "epoch": 0.25020105441872936,
      "grad_norm": 0.363836407661438,
      "learning_rate": 2e-05,
      "loss": 0.6929,
      "step": 2800
    },
    {
      "epoch": 0.25109462961308193,
      "grad_norm": 0.36290687322616577,
      "learning_rate": 2e-05,
      "loss": 0.7013,
      "step": 2810
    },
    {
      "epoch": 0.25198820480743456,
      "grad_norm": 0.3314644396305084,
      "learning_rate": 2e-05,
      "loss": 0.7216,
      "step": 2820
    },
    {
      "epoch": 0.2528817800017871,
      "grad_norm": 0.3978954255580902,
      "learning_rate": 2e-05,
      "loss": 0.7301,
      "step": 2830
    },
    {
      "epoch": 0.25377535519613975,
      "grad_norm": 0.37203192710876465,
      "learning_rate": 2e-05,
      "loss": 0.7025,
      "step": 2840
    },
    {
      "epoch": 0.2546689303904924,
      "grad_norm": 0.3674238622188568,
      "learning_rate": 2e-05,
      "loss": 0.6771,
      "step": 2850
    },
    {
      "epoch": 0.25556250558484495,
      "grad_norm": 0.398285448551178,
      "learning_rate": 2e-05,
      "loss": 0.7265,
      "step": 2860
    },
    {
      "epoch": 0.25645608077919757,
      "grad_norm": 0.4424970746040344,
      "learning_rate": 2e-05,
      "loss": 0.7293,
      "step": 2870
    },
    {
      "epoch": 0.2573496559735502,
      "grad_norm": 0.34791824221611023,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 2880
    },
    {
      "epoch": 0.25824323116790276,
      "grad_norm": 0.3702598214149475,
      "learning_rate": 2e-05,
      "loss": 0.7207,
      "step": 2890
    },
    {
      "epoch": 0.2591368063622554,
      "grad_norm": 0.36925578117370605,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 2900
    },
    {
      "epoch": 0.260030381556608,
      "grad_norm": 0.3745940923690796,
      "learning_rate": 2e-05,
      "loss": 0.7172,
      "step": 2910
    },
    {
      "epoch": 0.2609239567509606,
      "grad_norm": 0.3436862528324127,
      "learning_rate": 2e-05,
      "loss": 0.6852,
      "step": 2920
    },
    {
      "epoch": 0.2618175319453132,
      "grad_norm": 0.40740537643432617,
      "learning_rate": 2e-05,
      "loss": 0.6863,
      "step": 2930
    },
    {
      "epoch": 0.2627111071396658,
      "grad_norm": 0.398599773645401,
      "learning_rate": 2e-05,
      "loss": 0.7181,
      "step": 2940
    },
    {
      "epoch": 0.2636046823340184,
      "grad_norm": 0.37414512038230896,
      "learning_rate": 2e-05,
      "loss": 0.6983,
      "step": 2950
    },
    {
      "epoch": 0.26449825752837103,
      "grad_norm": 0.39800015091896057,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 2960
    },
    {
      "epoch": 0.2653918327227236,
      "grad_norm": 0.37266287207603455,
      "learning_rate": 2e-05,
      "loss": 0.715,
      "step": 2970
    },
    {
      "epoch": 0.2662854079170762,
      "grad_norm": 0.35028383135795593,
      "learning_rate": 2e-05,
      "loss": 0.7252,
      "step": 2980
    },
    {
      "epoch": 0.26717898311142885,
      "grad_norm": 0.36741185188293457,
      "learning_rate": 2e-05,
      "loss": 0.7095,
      "step": 2990
    },
    {
      "epoch": 0.2680725583057814,
      "grad_norm": 0.3469706177711487,
      "learning_rate": 2e-05,
      "loss": 0.7302,
      "step": 3000
    },
    {
      "epoch": 0.26896613350013404,
      "grad_norm": 0.38094988465309143,
      "learning_rate": 2e-05,
      "loss": 0.7099,
      "step": 3010
    },
    {
      "epoch": 0.26985970869448667,
      "grad_norm": 0.34936171770095825,
      "learning_rate": 2e-05,
      "loss": 0.6844,
      "step": 3020
    },
    {
      "epoch": 0.27075328388883924,
      "grad_norm": 0.3987492620944977,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 3030
    },
    {
      "epoch": 0.27164685908319186,
      "grad_norm": 0.39074254035949707,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 3040
    },
    {
      "epoch": 0.27254043427754443,
      "grad_norm": 0.4020327925682068,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 3050
    },
    {
      "epoch": 0.27343400947189705,
      "grad_norm": 0.6894024014472961,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 3060
    },
    {
      "epoch": 0.2743275846662497,
      "grad_norm": 0.41694730520248413,
      "learning_rate": 2e-05,
      "loss": 0.7251,
      "step": 3070
    },
    {
      "epoch": 0.27522115986060225,
      "grad_norm": 0.3496346175670624,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 3080
    },
    {
      "epoch": 0.2761147350549549,
      "grad_norm": 0.35392478108406067,
      "learning_rate": 2e-05,
      "loss": 0.7099,
      "step": 3090
    },
    {
      "epoch": 0.2770083102493075,
      "grad_norm": 0.3742842674255371,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 3100
    },
    {
      "epoch": 0.27790188544366007,
      "grad_norm": 0.36130213737487793,
      "learning_rate": 2e-05,
      "loss": 0.7179,
      "step": 3110
    },
    {
      "epoch": 0.2787954606380127,
      "grad_norm": 0.36318132281303406,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 3120
    },
    {
      "epoch": 0.2796890358323653,
      "grad_norm": 0.40958553552627563,
      "learning_rate": 2e-05,
      "loss": 0.7042,
      "step": 3130
    },
    {
      "epoch": 0.2805826110267179,
      "grad_norm": 0.3563935458660126,
      "learning_rate": 2e-05,
      "loss": 0.7358,
      "step": 3140
    },
    {
      "epoch": 0.2814761862210705,
      "grad_norm": 0.3450090289115906,
      "learning_rate": 2e-05,
      "loss": 0.6818,
      "step": 3150
    },
    {
      "epoch": 0.2823697614154231,
      "grad_norm": 0.40280184149742126,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 3160
    },
    {
      "epoch": 0.2832633366097757,
      "grad_norm": 0.37556108832359314,
      "learning_rate": 2e-05,
      "loss": 0.7182,
      "step": 3170
    },
    {
      "epoch": 0.28415691180412833,
      "grad_norm": 0.41327449679374695,
      "learning_rate": 2e-05,
      "loss": 0.6926,
      "step": 3180
    },
    {
      "epoch": 0.2850504869984809,
      "grad_norm": 0.33954551815986633,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 3190
    },
    {
      "epoch": 0.2859440621928335,
      "grad_norm": 0.3450734317302704,
      "learning_rate": 2e-05,
      "loss": 0.7083,
      "step": 3200
    },
    {
      "epoch": 0.28683763738718615,
      "grad_norm": 0.3419007360935211,
      "learning_rate": 2e-05,
      "loss": 0.7117,
      "step": 3210
    },
    {
      "epoch": 0.2877312125815387,
      "grad_norm": 0.4342067837715149,
      "learning_rate": 2e-05,
      "loss": 0.6925,
      "step": 3220
    },
    {
      "epoch": 0.28862478777589134,
      "grad_norm": 0.39862027764320374,
      "learning_rate": 2e-05,
      "loss": 0.7058,
      "step": 3230
    },
    {
      "epoch": 0.28951836297024397,
      "grad_norm": 0.37627285718917847,
      "learning_rate": 2e-05,
      "loss": 0.7113,
      "step": 3240
    },
    {
      "epoch": 0.29041193816459654,
      "grad_norm": 0.36114147305488586,
      "learning_rate": 2e-05,
      "loss": 0.7084,
      "step": 3250
    },
    {
      "epoch": 0.29130551335894916,
      "grad_norm": 0.3913547098636627,
      "learning_rate": 2e-05,
      "loss": 0.7155,
      "step": 3260
    },
    {
      "epoch": 0.29219908855330173,
      "grad_norm": 0.48569539189338684,
      "learning_rate": 2e-05,
      "loss": 0.6851,
      "step": 3270
    },
    {
      "epoch": 0.29309266374765436,
      "grad_norm": 0.35713067650794983,
      "learning_rate": 2e-05,
      "loss": 0.6896,
      "step": 3280
    },
    {
      "epoch": 0.293986238942007,
      "grad_norm": 0.3882346451282501,
      "learning_rate": 2e-05,
      "loss": 0.7099,
      "step": 3290
    },
    {
      "epoch": 0.29487981413635955,
      "grad_norm": 0.4093131124973297,
      "learning_rate": 2e-05,
      "loss": 0.7114,
      "step": 3300
    },
    {
      "epoch": 0.2957733893307122,
      "grad_norm": 0.3757636845111847,
      "learning_rate": 2e-05,
      "loss": 0.7258,
      "step": 3310
    },
    {
      "epoch": 0.2966669645250648,
      "grad_norm": 0.4188380837440491,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 3320
    },
    {
      "epoch": 0.29756053971941737,
      "grad_norm": 0.4249805510044098,
      "learning_rate": 2e-05,
      "loss": 0.6834,
      "step": 3330
    },
    {
      "epoch": 0.29845411491377,
      "grad_norm": 0.37131360173225403,
      "learning_rate": 2e-05,
      "loss": 0.6813,
      "step": 3340
    },
    {
      "epoch": 0.2993476901081226,
      "grad_norm": 0.4095926582813263,
      "learning_rate": 2e-05,
      "loss": 0.6989,
      "step": 3350
    },
    {
      "epoch": 0.3002412653024752,
      "grad_norm": 0.40671223402023315,
      "learning_rate": 2e-05,
      "loss": 0.6926,
      "step": 3360
    },
    {
      "epoch": 0.3011348404968278,
      "grad_norm": 0.3677366077899933,
      "learning_rate": 2e-05,
      "loss": 0.6928,
      "step": 3370
    },
    {
      "epoch": 0.3020284156911804,
      "grad_norm": 0.38548967242240906,
      "learning_rate": 2e-05,
      "loss": 0.7158,
      "step": 3380
    },
    {
      "epoch": 0.302921990885533,
      "grad_norm": 0.38187336921691895,
      "learning_rate": 2e-05,
      "loss": 0.7225,
      "step": 3390
    },
    {
      "epoch": 0.30381556607988564,
      "grad_norm": 0.3772178590297699,
      "learning_rate": 2e-05,
      "loss": 0.6714,
      "step": 3400
    },
    {
      "epoch": 0.3047091412742382,
      "grad_norm": 0.36621707677841187,
      "learning_rate": 2e-05,
      "loss": 0.7108,
      "step": 3410
    },
    {
      "epoch": 0.30560271646859083,
      "grad_norm": 0.3519044816493988,
      "learning_rate": 2e-05,
      "loss": 0.6995,
      "step": 3420
    },
    {
      "epoch": 0.30649629166294345,
      "grad_norm": 0.362652987241745,
      "learning_rate": 2e-05,
      "loss": 0.7082,
      "step": 3430
    },
    {
      "epoch": 0.307389866857296,
      "grad_norm": 0.3681955635547638,
      "learning_rate": 2e-05,
      "loss": 0.7242,
      "step": 3440
    },
    {
      "epoch": 0.30828344205164865,
      "grad_norm": 0.36930564045906067,
      "learning_rate": 2e-05,
      "loss": 0.731,
      "step": 3450
    },
    {
      "epoch": 0.3091770172460013,
      "grad_norm": 0.36028867959976196,
      "learning_rate": 2e-05,
      "loss": 0.7353,
      "step": 3460
    },
    {
      "epoch": 0.31007059244035384,
      "grad_norm": 0.3412564992904663,
      "learning_rate": 2e-05,
      "loss": 0.6911,
      "step": 3470
    },
    {
      "epoch": 0.31096416763470647,
      "grad_norm": 0.3616200387477875,
      "learning_rate": 2e-05,
      "loss": 0.6709,
      "step": 3480
    },
    {
      "epoch": 0.3118577428290591,
      "grad_norm": 0.3621264100074768,
      "learning_rate": 2e-05,
      "loss": 0.6842,
      "step": 3490
    },
    {
      "epoch": 0.31275131802341166,
      "grad_norm": 0.38668254017829895,
      "learning_rate": 2e-05,
      "loss": 0.7577,
      "step": 3500
    },
    {
      "epoch": 0.3136448932177643,
      "grad_norm": 0.40664446353912354,
      "learning_rate": 2e-05,
      "loss": 0.6919,
      "step": 3510
    },
    {
      "epoch": 0.31453846841211686,
      "grad_norm": 0.4402414858341217,
      "learning_rate": 2e-05,
      "loss": 0.7077,
      "step": 3520
    },
    {
      "epoch": 0.3154320436064695,
      "grad_norm": 0.3952922821044922,
      "learning_rate": 2e-05,
      "loss": 0.6773,
      "step": 3530
    },
    {
      "epoch": 0.3163256188008221,
      "grad_norm": 0.4013434946537018,
      "learning_rate": 2e-05,
      "loss": 0.6899,
      "step": 3540
    },
    {
      "epoch": 0.3172191939951747,
      "grad_norm": 0.37860897183418274,
      "learning_rate": 2e-05,
      "loss": 0.695,
      "step": 3550
    },
    {
      "epoch": 0.3181127691895273,
      "grad_norm": 0.397117555141449,
      "learning_rate": 2e-05,
      "loss": 0.6925,
      "step": 3560
    },
    {
      "epoch": 0.3190063443838799,
      "grad_norm": 0.46799492835998535,
      "learning_rate": 2e-05,
      "loss": 0.7116,
      "step": 3570
    },
    {
      "epoch": 0.3198999195782325,
      "grad_norm": 0.34963753819465637,
      "learning_rate": 2e-05,
      "loss": 0.7115,
      "step": 3580
    },
    {
      "epoch": 0.3207934947725851,
      "grad_norm": 0.411518931388855,
      "learning_rate": 2e-05,
      "loss": 0.6943,
      "step": 3590
    },
    {
      "epoch": 0.32168706996693774,
      "grad_norm": 0.36255308985710144,
      "learning_rate": 2e-05,
      "loss": 0.7075,
      "step": 3600
    },
    {
      "epoch": 0.3225806451612903,
      "grad_norm": 0.3666975796222687,
      "learning_rate": 2e-05,
      "loss": 0.7209,
      "step": 3610
    },
    {
      "epoch": 0.32347422035564294,
      "grad_norm": 0.3676561713218689,
      "learning_rate": 2e-05,
      "loss": 0.6981,
      "step": 3620
    },
    {
      "epoch": 0.3243677955499955,
      "grad_norm": 0.417124480009079,
      "learning_rate": 2e-05,
      "loss": 0.7004,
      "step": 3630
    },
    {
      "epoch": 0.32526137074434813,
      "grad_norm": 0.3794247806072235,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 3640
    },
    {
      "epoch": 0.32615494593870076,
      "grad_norm": 0.40678223967552185,
      "learning_rate": 2e-05,
      "loss": 0.7257,
      "step": 3650
    },
    {
      "epoch": 0.3270485211330533,
      "grad_norm": 0.37268680334091187,
      "learning_rate": 2e-05,
      "loss": 0.6853,
      "step": 3660
    },
    {
      "epoch": 0.32794209632740595,
      "grad_norm": 0.38034191727638245,
      "learning_rate": 2e-05,
      "loss": 0.6743,
      "step": 3670
    },
    {
      "epoch": 0.3288356715217586,
      "grad_norm": 0.35265952348709106,
      "learning_rate": 2e-05,
      "loss": 0.7162,
      "step": 3680
    },
    {
      "epoch": 0.32972924671611115,
      "grad_norm": 0.36596548557281494,
      "learning_rate": 2e-05,
      "loss": 0.6653,
      "step": 3690
    },
    {
      "epoch": 0.33062282191046377,
      "grad_norm": 0.37113913893699646,
      "learning_rate": 2e-05,
      "loss": 0.6983,
      "step": 3700
    },
    {
      "epoch": 0.3315163971048164,
      "grad_norm": 0.4150978624820709,
      "learning_rate": 2e-05,
      "loss": 0.7174,
      "step": 3710
    },
    {
      "epoch": 0.33240997229916897,
      "grad_norm": 0.34214451909065247,
      "learning_rate": 2e-05,
      "loss": 0.7131,
      "step": 3720
    },
    {
      "epoch": 0.3333035474935216,
      "grad_norm": 0.6437499523162842,
      "learning_rate": 2e-05,
      "loss": 0.694,
      "step": 3730
    },
    {
      "epoch": 0.33419712268787416,
      "grad_norm": 0.33215585350990295,
      "learning_rate": 2e-05,
      "loss": 0.7026,
      "step": 3740
    },
    {
      "epoch": 0.3350906978822268,
      "grad_norm": 0.36878666281700134,
      "learning_rate": 2e-05,
      "loss": 0.714,
      "step": 3750
    },
    {
      "epoch": 0.3359842730765794,
      "grad_norm": 0.36607626080513,
      "learning_rate": 2e-05,
      "loss": 0.7004,
      "step": 3760
    },
    {
      "epoch": 0.336877848270932,
      "grad_norm": 0.41994553804397583,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 3770
    },
    {
      "epoch": 0.3377714234652846,
      "grad_norm": 0.3406926691532135,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 3780
    },
    {
      "epoch": 0.33866499865963723,
      "grad_norm": 0.3424694538116455,
      "learning_rate": 2e-05,
      "loss": 0.7139,
      "step": 3790
    },
    {
      "epoch": 0.3395585738539898,
      "grad_norm": 0.3926262855529785,
      "learning_rate": 2e-05,
      "loss": 0.6655,
      "step": 3800
    },
    {
      "epoch": 0.3404521490483424,
      "grad_norm": 0.40683773159980774,
      "learning_rate": 2e-05,
      "loss": 0.7186,
      "step": 3810
    },
    {
      "epoch": 0.34134572424269505,
      "grad_norm": 0.3801291286945343,
      "learning_rate": 2e-05,
      "loss": 0.6809,
      "step": 3820
    },
    {
      "epoch": 0.3422392994370476,
      "grad_norm": 0.3531859517097473,
      "learning_rate": 2e-05,
      "loss": 0.684,
      "step": 3830
    },
    {
      "epoch": 0.34313287463140024,
      "grad_norm": 0.4508625268936157,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 3840
    },
    {
      "epoch": 0.3440264498257528,
      "grad_norm": 0.3657115399837494,
      "learning_rate": 2e-05,
      "loss": 0.6673,
      "step": 3850
    },
    {
      "epoch": 0.34492002502010544,
      "grad_norm": 0.3592742085456848,
      "learning_rate": 2e-05,
      "loss": 0.6633,
      "step": 3860
    },
    {
      "epoch": 0.34581360021445806,
      "grad_norm": 0.36773285269737244,
      "learning_rate": 2e-05,
      "loss": 0.6801,
      "step": 3870
    },
    {
      "epoch": 0.34670717540881063,
      "grad_norm": 0.37082305550575256,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 3880
    },
    {
      "epoch": 0.34760075060316326,
      "grad_norm": 0.3258555233478546,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 3890
    },
    {
      "epoch": 0.3484943257975159,
      "grad_norm": 0.4185699224472046,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 3900
    },
    {
      "epoch": 0.34938790099186845,
      "grad_norm": 0.41331052780151367,
      "learning_rate": 2e-05,
      "loss": 0.6871,
      "step": 3910
    },
    {
      "epoch": 0.3502814761862211,
      "grad_norm": 0.36346831917762756,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 3920
    },
    {
      "epoch": 0.3511750513805737,
      "grad_norm": 0.3878091871738434,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 3930
    },
    {
      "epoch": 0.35206862657492627,
      "grad_norm": 0.3615358769893646,
      "learning_rate": 2e-05,
      "loss": 0.6847,
      "step": 3940
    },
    {
      "epoch": 0.3529622017692789,
      "grad_norm": 0.3688424229621887,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 3950
    },
    {
      "epoch": 0.35385577696363146,
      "grad_norm": 0.40755635499954224,
      "learning_rate": 2e-05,
      "loss": 0.6959,
      "step": 3960
    },
    {
      "epoch": 0.3547493521579841,
      "grad_norm": 0.3542814552783966,
      "learning_rate": 2e-05,
      "loss": 0.688,
      "step": 3970
    },
    {
      "epoch": 0.3556429273523367,
      "grad_norm": 0.3793947994709015,
      "learning_rate": 2e-05,
      "loss": 0.7086,
      "step": 3980
    },
    {
      "epoch": 0.3565365025466893,
      "grad_norm": 0.4207400381565094,
      "learning_rate": 2e-05,
      "loss": 0.7057,
      "step": 3990
    },
    {
      "epoch": 0.3574300777410419,
      "grad_norm": 0.373004287481308,
      "learning_rate": 2e-05,
      "loss": 0.68,
      "step": 4000
    },
    {
      "epoch": 0.35832365293539453,
      "grad_norm": 0.4065035879611969,
      "learning_rate": 2e-05,
      "loss": 0.6917,
      "step": 4010
    },
    {
      "epoch": 0.3592172281297471,
      "grad_norm": 0.3997182846069336,
      "learning_rate": 2e-05,
      "loss": 0.6863,
      "step": 4020
    },
    {
      "epoch": 0.3601108033240997,
      "grad_norm": 0.3566669225692749,
      "learning_rate": 2e-05,
      "loss": 0.6734,
      "step": 4030
    },
    {
      "epoch": 0.36100437851845235,
      "grad_norm": 0.3704060912132263,
      "learning_rate": 2e-05,
      "loss": 0.6712,
      "step": 4040
    },
    {
      "epoch": 0.3618979537128049,
      "grad_norm": 0.35917559266090393,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 4050
    },
    {
      "epoch": 0.36279152890715755,
      "grad_norm": 0.3765454888343811,
      "learning_rate": 2e-05,
      "loss": 0.6951,
      "step": 4060
    },
    {
      "epoch": 0.3636851041015101,
      "grad_norm": 0.3873068690299988,
      "learning_rate": 2e-05,
      "loss": 0.6772,
      "step": 4070
    },
    {
      "epoch": 0.36457867929586274,
      "grad_norm": 0.45908427238464355,
      "learning_rate": 2e-05,
      "loss": 0.6765,
      "step": 4080
    },
    {
      "epoch": 0.36547225449021536,
      "grad_norm": 0.3917163610458374,
      "learning_rate": 2e-05,
      "loss": 0.6787,
      "step": 4090
    },
    {
      "epoch": 0.36636582968456793,
      "grad_norm": 0.39318227767944336,
      "learning_rate": 2e-05,
      "loss": 0.6915,
      "step": 4100
    },
    {
      "epoch": 0.36725940487892056,
      "grad_norm": 0.4183279573917389,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 4110
    },
    {
      "epoch": 0.3681529800732732,
      "grad_norm": 0.3996908664703369,
      "learning_rate": 2e-05,
      "loss": 0.7152,
      "step": 4120
    },
    {
      "epoch": 0.36904655526762575,
      "grad_norm": 0.38322579860687256,
      "learning_rate": 2e-05,
      "loss": 0.6713,
      "step": 4130
    },
    {
      "epoch": 0.3699401304619784,
      "grad_norm": 0.417958527803421,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 4140
    },
    {
      "epoch": 0.370833705656331,
      "grad_norm": 0.386135458946228,
      "learning_rate": 2e-05,
      "loss": 0.6773,
      "step": 4150
    },
    {
      "epoch": 0.3717272808506836,
      "grad_norm": 0.3646849989891052,
      "learning_rate": 2e-05,
      "loss": 0.6784,
      "step": 4160
    },
    {
      "epoch": 0.3726208560450362,
      "grad_norm": 0.38974642753601074,
      "learning_rate": 2e-05,
      "loss": 0.6961,
      "step": 4170
    },
    {
      "epoch": 0.3735144312393888,
      "grad_norm": 0.36325088143348694,
      "learning_rate": 2e-05,
      "loss": 0.6646,
      "step": 4180
    },
    {
      "epoch": 0.3744080064337414,
      "grad_norm": 0.3774178624153137,
      "learning_rate": 2e-05,
      "loss": 0.6642,
      "step": 4190
    },
    {
      "epoch": 0.375301581628094,
      "grad_norm": 0.3733026385307312,
      "learning_rate": 2e-05,
      "loss": 0.6835,
      "step": 4200
    },
    {
      "epoch": 0.3761951568224466,
      "grad_norm": 0.4059685170650482,
      "learning_rate": 2e-05,
      "loss": 0.7016,
      "step": 4210
    },
    {
      "epoch": 0.3770887320167992,
      "grad_norm": 0.3935318887233734,
      "learning_rate": 2e-05,
      "loss": 0.7123,
      "step": 4220
    },
    {
      "epoch": 0.37798230721115184,
      "grad_norm": 0.3663133382797241,
      "learning_rate": 2e-05,
      "loss": 0.6721,
      "step": 4230
    },
    {
      "epoch": 0.3788758824055044,
      "grad_norm": 0.3684714138507843,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 4240
    },
    {
      "epoch": 0.37976945759985703,
      "grad_norm": 0.37538403272628784,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 4250
    },
    {
      "epoch": 0.38066303279420965,
      "grad_norm": 0.4001816511154175,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 4260
    },
    {
      "epoch": 0.3815566079885622,
      "grad_norm": 0.3811742663383484,
      "learning_rate": 2e-05,
      "loss": 0.6877,
      "step": 4270
    },
    {
      "epoch": 0.38245018318291485,
      "grad_norm": 0.46673205494880676,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 4280
    },
    {
      "epoch": 0.3833437583772675,
      "grad_norm": 0.37563374638557434,
      "learning_rate": 2e-05,
      "loss": 0.6742,
      "step": 4290
    },
    {
      "epoch": 0.38423733357162004,
      "grad_norm": 6.816838264465332,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 4300
    },
    {
      "epoch": 0.38513090876597267,
      "grad_norm": 0.36754029989242554,
      "learning_rate": 2e-05,
      "loss": 0.6976,
      "step": 4310
    },
    {
      "epoch": 0.38602448396032524,
      "grad_norm": 0.3981974720954895,
      "learning_rate": 2e-05,
      "loss": 0.6844,
      "step": 4320
    },
    {
      "epoch": 0.38691805915467786,
      "grad_norm": 0.6864988803863525,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 4330
    },
    {
      "epoch": 0.3878116343490305,
      "grad_norm": 0.38520216941833496,
      "learning_rate": 2e-05,
      "loss": 0.7092,
      "step": 4340
    },
    {
      "epoch": 0.38870520954338306,
      "grad_norm": 0.3747800886631012,
      "learning_rate": 2e-05,
      "loss": 0.6966,
      "step": 4350
    },
    {
      "epoch": 0.3895987847377357,
      "grad_norm": 0.36001715064048767,
      "learning_rate": 2e-05,
      "loss": 0.6479,
      "step": 4360
    },
    {
      "epoch": 0.3904923599320883,
      "grad_norm": 0.35367342829704285,
      "learning_rate": 2e-05,
      "loss": 0.6809,
      "step": 4370
    },
    {
      "epoch": 0.3913859351264409,
      "grad_norm": 0.39095789194107056,
      "learning_rate": 2e-05,
      "loss": 0.6878,
      "step": 4380
    },
    {
      "epoch": 0.3922795103207935,
      "grad_norm": 0.3698689043521881,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 4390
    },
    {
      "epoch": 0.3931730855151461,
      "grad_norm": 0.37408924102783203,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 4400
    },
    {
      "epoch": 0.3940666607094987,
      "grad_norm": 0.36290866136550903,
      "learning_rate": 2e-05,
      "loss": 0.6966,
      "step": 4410
    },
    {
      "epoch": 0.3949602359038513,
      "grad_norm": 0.364419549703598,
      "learning_rate": 2e-05,
      "loss": 0.743,
      "step": 4420
    },
    {
      "epoch": 0.3958538110982039,
      "grad_norm": 0.35897934436798096,
      "learning_rate": 2e-05,
      "loss": 0.6659,
      "step": 4430
    },
    {
      "epoch": 0.3967473862925565,
      "grad_norm": 0.34913840889930725,
      "learning_rate": 2e-05,
      "loss": 0.6977,
      "step": 4440
    },
    {
      "epoch": 0.39764096148690914,
      "grad_norm": 0.36154356598854065,
      "learning_rate": 2e-05,
      "loss": 0.7255,
      "step": 4450
    },
    {
      "epoch": 0.3985345366812617,
      "grad_norm": 0.4215991497039795,
      "learning_rate": 2e-05,
      "loss": 0.6895,
      "step": 4460
    },
    {
      "epoch": 0.39942811187561433,
      "grad_norm": 0.519006073474884,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 4470
    },
    {
      "epoch": 0.40032168706996696,
      "grad_norm": 0.36556845903396606,
      "learning_rate": 2e-05,
      "loss": 0.679,
      "step": 4480
    },
    {
      "epoch": 0.4012152622643195,
      "grad_norm": 0.3858962059020996,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 4490
    },
    {
      "epoch": 0.40210883745867215,
      "grad_norm": 0.3571610152721405,
      "learning_rate": 2e-05,
      "loss": 0.6791,
      "step": 4500
    },
    {
      "epoch": 0.4030024126530248,
      "grad_norm": 0.3609004318714142,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 4510
    },
    {
      "epoch": 0.40389598784737735,
      "grad_norm": 0.37380915880203247,
      "learning_rate": 2e-05,
      "loss": 0.6339,
      "step": 4520
    },
    {
      "epoch": 0.40478956304172997,
      "grad_norm": 0.49198201298713684,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 4530
    },
    {
      "epoch": 0.40568313823608254,
      "grad_norm": 0.3402051329612732,
      "learning_rate": 2e-05,
      "loss": 0.69,
      "step": 4540
    },
    {
      "epoch": 0.40657671343043517,
      "grad_norm": 0.40370455384254456,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 4550
    },
    {
      "epoch": 0.4074702886247878,
      "grad_norm": 0.3506670594215393,
      "learning_rate": 2e-05,
      "loss": 0.6948,
      "step": 4560
    },
    {
      "epoch": 0.40836386381914036,
      "grad_norm": 0.363017201423645,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 4570
    },
    {
      "epoch": 0.409257439013493,
      "grad_norm": 0.38264966011047363,
      "learning_rate": 2e-05,
      "loss": 0.7156,
      "step": 4580
    },
    {
      "epoch": 0.4101510142078456,
      "grad_norm": 0.3690856099128723,
      "learning_rate": 2e-05,
      "loss": 0.6768,
      "step": 4590
    },
    {
      "epoch": 0.4110445894021982,
      "grad_norm": 0.599280595779419,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 4600
    },
    {
      "epoch": 0.4119381645965508,
      "grad_norm": 0.3705596923828125,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 4610
    },
    {
      "epoch": 0.41283173979090343,
      "grad_norm": 0.36947357654571533,
      "learning_rate": 2e-05,
      "loss": 0.6918,
      "step": 4620
    },
    {
      "epoch": 0.413725314985256,
      "grad_norm": 0.38991454243659973,
      "learning_rate": 2e-05,
      "loss": 0.6834,
      "step": 4630
    },
    {
      "epoch": 0.4146188901796086,
      "grad_norm": 0.3343438506126404,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 4640
    },
    {
      "epoch": 0.4155124653739612,
      "grad_norm": 0.3949652314186096,
      "learning_rate": 2e-05,
      "loss": 0.6711,
      "step": 4650
    },
    {
      "epoch": 0.4164060405683138,
      "grad_norm": 0.37002232670783997,
      "learning_rate": 2e-05,
      "loss": 0.6903,
      "step": 4660
    },
    {
      "epoch": 0.41729961576266644,
      "grad_norm": 0.383247971534729,
      "learning_rate": 2e-05,
      "loss": 0.6945,
      "step": 4670
    },
    {
      "epoch": 0.418193190957019,
      "grad_norm": 0.38935351371765137,
      "learning_rate": 2e-05,
      "loss": 0.6894,
      "step": 4680
    },
    {
      "epoch": 0.41908676615137164,
      "grad_norm": 0.38259515166282654,
      "learning_rate": 2e-05,
      "loss": 0.6986,
      "step": 4690
    },
    {
      "epoch": 0.41998034134572426,
      "grad_norm": 0.348149836063385,
      "learning_rate": 2e-05,
      "loss": 0.6872,
      "step": 4700
    },
    {
      "epoch": 0.42087391654007683,
      "grad_norm": 0.36275845766067505,
      "learning_rate": 2e-05,
      "loss": 0.6837,
      "step": 4710
    },
    {
      "epoch": 0.42176749173442946,
      "grad_norm": 0.4239656329154968,
      "learning_rate": 2e-05,
      "loss": 0.7012,
      "step": 4720
    },
    {
      "epoch": 0.4226610669287821,
      "grad_norm": 0.3658117651939392,
      "learning_rate": 2e-05,
      "loss": 0.7016,
      "step": 4730
    },
    {
      "epoch": 0.42355464212313465,
      "grad_norm": 0.377859503030777,
      "learning_rate": 2e-05,
      "loss": 0.6828,
      "step": 4740
    },
    {
      "epoch": 0.4244482173174873,
      "grad_norm": 0.38955992460250854,
      "learning_rate": 2e-05,
      "loss": 0.6773,
      "step": 4750
    },
    {
      "epoch": 0.42534179251183984,
      "grad_norm": 0.3484950065612793,
      "learning_rate": 2e-05,
      "loss": 0.7054,
      "step": 4760
    },
    {
      "epoch": 0.42623536770619247,
      "grad_norm": 0.3950771689414978,
      "learning_rate": 2e-05,
      "loss": 0.6854,
      "step": 4770
    },
    {
      "epoch": 0.4271289429005451,
      "grad_norm": 0.384011447429657,
      "learning_rate": 2e-05,
      "loss": 0.6921,
      "step": 4780
    },
    {
      "epoch": 0.42802251809489766,
      "grad_norm": 0.3784692883491516,
      "learning_rate": 2e-05,
      "loss": 0.6886,
      "step": 4790
    },
    {
      "epoch": 0.4289160932892503,
      "grad_norm": 0.3680237829685211,
      "learning_rate": 2e-05,
      "loss": 0.6835,
      "step": 4800
    },
    {
      "epoch": 0.4298096684836029,
      "grad_norm": 0.34581565856933594,
      "learning_rate": 2e-05,
      "loss": 0.6733,
      "step": 4810
    },
    {
      "epoch": 0.4307032436779555,
      "grad_norm": 0.35307255387306213,
      "learning_rate": 2e-05,
      "loss": 0.6833,
      "step": 4820
    },
    {
      "epoch": 0.4315968188723081,
      "grad_norm": 0.3662859797477722,
      "learning_rate": 2e-05,
      "loss": 0.681,
      "step": 4830
    },
    {
      "epoch": 0.43249039406666073,
      "grad_norm": 0.4486631155014038,
      "learning_rate": 2e-05,
      "loss": 0.6872,
      "step": 4840
    },
    {
      "epoch": 0.4333839692610133,
      "grad_norm": 0.3616761565208435,
      "learning_rate": 2e-05,
      "loss": 0.648,
      "step": 4850
    },
    {
      "epoch": 0.4342775444553659,
      "grad_norm": 0.3854536712169647,
      "learning_rate": 2e-05,
      "loss": 0.6601,
      "step": 4860
    },
    {
      "epoch": 0.4351711196497185,
      "grad_norm": 0.3570791780948639,
      "learning_rate": 2e-05,
      "loss": 0.7085,
      "step": 4870
    },
    {
      "epoch": 0.4360646948440711,
      "grad_norm": 0.3633802831172943,
      "learning_rate": 2e-05,
      "loss": 0.6874,
      "step": 4880
    },
    {
      "epoch": 0.43695827003842375,
      "grad_norm": 0.3554231524467468,
      "learning_rate": 2e-05,
      "loss": 0.6654,
      "step": 4890
    },
    {
      "epoch": 0.4378518452327763,
      "grad_norm": 0.5552303194999695,
      "learning_rate": 2e-05,
      "loss": 0.6903,
      "step": 4900
    },
    {
      "epoch": 0.43874542042712894,
      "grad_norm": 0.38484469056129456,
      "learning_rate": 2e-05,
      "loss": 0.6819,
      "step": 4910
    },
    {
      "epoch": 0.43963899562148157,
      "grad_norm": 0.4366731345653534,
      "learning_rate": 2e-05,
      "loss": 0.6987,
      "step": 4920
    },
    {
      "epoch": 0.44053257081583413,
      "grad_norm": 0.3423648774623871,
      "learning_rate": 2e-05,
      "loss": 0.6741,
      "step": 4930
    },
    {
      "epoch": 0.44142614601018676,
      "grad_norm": 0.3649511933326721,
      "learning_rate": 2e-05,
      "loss": 0.6907,
      "step": 4940
    },
    {
      "epoch": 0.4423197212045394,
      "grad_norm": 0.3819730281829834,
      "learning_rate": 2e-05,
      "loss": 0.6709,
      "step": 4950
    },
    {
      "epoch": 0.44321329639889195,
      "grad_norm": 0.365536630153656,
      "learning_rate": 2e-05,
      "loss": 0.7045,
      "step": 4960
    },
    {
      "epoch": 0.4441068715932446,
      "grad_norm": 0.37038612365722656,
      "learning_rate": 2e-05,
      "loss": 0.6797,
      "step": 4970
    },
    {
      "epoch": 0.4450004467875972,
      "grad_norm": 0.3828907608985901,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 4980
    },
    {
      "epoch": 0.4458940219819498,
      "grad_norm": 0.38271892070770264,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 4990
    },
    {
      "epoch": 0.4467875971763024,
      "grad_norm": 0.3896707594394684,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 5000
    },
    {
      "epoch": 0.44768117237065497,
      "grad_norm": 0.38165053725242615,
      "learning_rate": 2e-05,
      "loss": 0.656,
      "step": 5010
    },
    {
      "epoch": 0.4485747475650076,
      "grad_norm": 0.3564092218875885,
      "learning_rate": 2e-05,
      "loss": 0.6423,
      "step": 5020
    },
    {
      "epoch": 0.4494683227593602,
      "grad_norm": 0.3507450819015503,
      "learning_rate": 2e-05,
      "loss": 0.6583,
      "step": 5030
    },
    {
      "epoch": 0.4503618979537128,
      "grad_norm": 0.36991018056869507,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 5040
    },
    {
      "epoch": 0.4512554731480654,
      "grad_norm": 0.37723660469055176,
      "learning_rate": 2e-05,
      "loss": 0.6693,
      "step": 5050
    },
    {
      "epoch": 0.45214904834241804,
      "grad_norm": 0.3848140239715576,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 5060
    },
    {
      "epoch": 0.4530426235367706,
      "grad_norm": 0.3824312388896942,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 5070
    },
    {
      "epoch": 0.45393619873112323,
      "grad_norm": 0.3876231908798218,
      "learning_rate": 2e-05,
      "loss": 0.6932,
      "step": 5080
    },
    {
      "epoch": 0.45482977392547586,
      "grad_norm": 0.3508111238479614,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.6498,
      "step": 5090
    },
    {
      "epoch": 0.4557233491198284,
      "grad_norm": 0.3386197090148926,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.68,
      "step": 5100
    },
    {
      "epoch": 0.45661692431418105,
      "grad_norm": 0.4007887840270996,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.7035,
      "step": 5110
    },
    {
      "epoch": 0.4575104995085336,
      "grad_norm": 0.3901657164096832,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.6765,
      "step": 5120
    },
    {
      "epoch": 0.45840407470288624,
      "grad_norm": 0.3588799834251404,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.6723,
      "step": 5130
    },
    {
      "epoch": 0.45929764989723887,
      "grad_norm": 0.3526768386363983,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.6794,
      "step": 5140
    },
    {
      "epoch": 0.46019122509159144,
      "grad_norm": 0.4044317901134491,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.6683,
      "step": 5150
    },
    {
      "epoch": 0.46108480028594406,
      "grad_norm": 0.3659527003765106,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.6854,
      "step": 5160
    },
    {
      "epoch": 0.4619783754802967,
      "grad_norm": 0.44135573506355286,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.6903,
      "step": 5170
    },
    {
      "epoch": 0.46287195067464926,
      "grad_norm": 0.35972270369529724,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.7004,
      "step": 5180
    },
    {
      "epoch": 0.4637655258690019,
      "grad_norm": 0.3828246593475342,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.6794,
      "step": 5190
    },
    {
      "epoch": 0.4646591010633545,
      "grad_norm": 0.3737556040287018,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.6959,
      "step": 5200
    },
    {
      "epoch": 0.4655526762577071,
      "grad_norm": 0.3815043270587921,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.6857,
      "step": 5210
    },
    {
      "epoch": 0.4664462514520597,
      "grad_norm": 0.4178937077522278,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.6817,
      "step": 5220
    },
    {
      "epoch": 0.46733982664641227,
      "grad_norm": 0.39337319135665894,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.6854,
      "step": 5230
    },
    {
      "epoch": 0.4682334018407649,
      "grad_norm": 0.3589783310890198,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.6677,
      "step": 5240
    },
    {
      "epoch": 0.4691269770351175,
      "grad_norm": 0.42616841197013855,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.6627,
      "step": 5250
    },
    {
      "epoch": 0.4700205522294701,
      "grad_norm": 0.3791368305683136,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.6828,
      "step": 5260
    },
    {
      "epoch": 0.4709141274238227,
      "grad_norm": 0.32623526453971863,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.6639,
      "step": 5270
    },
    {
      "epoch": 0.47180770261817534,
      "grad_norm": 0.36976712942123413,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.6691,
      "step": 5280
    },
    {
      "epoch": 0.4727012778125279,
      "grad_norm": 0.3780418932437897,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.6633,
      "step": 5290
    },
    {
      "epoch": 0.47359485300688053,
      "grad_norm": 0.39815905690193176,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.6562,
      "step": 5300
    },
    {
      "epoch": 0.47448842820123316,
      "grad_norm": 0.3759981095790863,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.6616,
      "step": 5310
    },
    {
      "epoch": 0.47538200339558573,
      "grad_norm": 0.36337733268737793,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.6828,
      "step": 5320
    },
    {
      "epoch": 0.47627557858993835,
      "grad_norm": 0.3758830428123474,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.6554,
      "step": 5330
    },
    {
      "epoch": 0.4771691537842909,
      "grad_norm": 0.3267016112804413,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.6674,
      "step": 5340
    },
    {
      "epoch": 0.47806272897864355,
      "grad_norm": 0.344720721244812,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.672,
      "step": 5350
    },
    {
      "epoch": 0.4789563041729962,
      "grad_norm": 0.40488025546073914,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.6812,
      "step": 5360
    },
    {
      "epoch": 0.47984987936734874,
      "grad_norm": 0.34486478567123413,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.6694,
      "step": 5370
    },
    {
      "epoch": 0.48074345456170137,
      "grad_norm": 0.3502098619937897,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.6549,
      "step": 5380
    },
    {
      "epoch": 0.481637029756054,
      "grad_norm": 0.34294840693473816,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.6612,
      "step": 5390
    },
    {
      "epoch": 0.48253060495040656,
      "grad_norm": 0.35749590396881104,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.6558,
      "step": 5400
    },
    {
      "epoch": 0.4834241801447592,
      "grad_norm": 0.3789016902446747,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.7208,
      "step": 5410
    },
    {
      "epoch": 0.4843177553391118,
      "grad_norm": 0.37972861528396606,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.6595,
      "step": 5420
    },
    {
      "epoch": 0.4852113305334644,
      "grad_norm": 0.3548266589641571,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.6953,
      "step": 5430
    },
    {
      "epoch": 0.486104905727817,
      "grad_norm": 0.34426432847976685,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.6506,
      "step": 5440
    },
    {
      "epoch": 0.4869984809221696,
      "grad_norm": 0.35695916414260864,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.6927,
      "step": 5450
    },
    {
      "epoch": 0.4878920561165222,
      "grad_norm": 0.3873395323753357,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.6562,
      "step": 5460
    },
    {
      "epoch": 0.4887856313108748,
      "grad_norm": 0.32732948660850525,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.6553,
      "step": 5470
    },
    {
      "epoch": 0.4896792065052274,
      "grad_norm": 0.34798330068588257,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.6598,
      "step": 5480
    },
    {
      "epoch": 0.49057278169958,
      "grad_norm": 0.34739646315574646,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.6593,
      "step": 5490
    },
    {
      "epoch": 0.49146635689393264,
      "grad_norm": 0.32307547330856323,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.6842,
      "step": 5500
    },
    {
      "epoch": 0.4923599320882852,
      "grad_norm": 0.4031292796134949,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.6447,
      "step": 5510
    },
    {
      "epoch": 0.49325350728263784,
      "grad_norm": 0.3543015122413635,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.6764,
      "step": 5520
    },
    {
      "epoch": 0.49414708247699046,
      "grad_norm": 0.33218860626220703,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.6682,
      "step": 5530
    },
    {
      "epoch": 0.49504065767134303,
      "grad_norm": 0.340327650308609,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.6605,
      "step": 5540
    },
    {
      "epoch": 0.49593423286569566,
      "grad_norm": 0.3881992697715759,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.6533,
      "step": 5550
    },
    {
      "epoch": 0.4968278080600482,
      "grad_norm": 0.33611172437667847,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.6685,
      "step": 5560
    },
    {
      "epoch": 0.49772138325440085,
      "grad_norm": 0.3338543176651001,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.6259,
      "step": 5570
    },
    {
      "epoch": 0.4986149584487535,
      "grad_norm": 0.36683836579322815,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6632,
      "step": 5580
    },
    {
      "epoch": 0.49950853364310605,
      "grad_norm": 0.39136382937431335,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.6699,
      "step": 5590
    },
    {
      "epoch": 0.5001340362791529,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.02263869637780175,
      "train_runtime": 1779.8859,
      "train_samples_per_second": 37.735,
      "train_steps_per_second": 3.145
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}