{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9358749268455815,
  "eval_steps": 500,
  "global_step": 11194,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000836050497450046,
      "grad_norm": 5.8952789306640625,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 1.5346,
      "step": 10
    },
    {
      "epoch": 0.001672100994900092,
      "grad_norm": 2.239838123321533,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.3887,
      "step": 20
    },
    {
      "epoch": 0.002508151492350138,
      "grad_norm": 1.8720221519470215,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.2524,
      "step": 30
    },
    {
      "epoch": 0.003344201989800184,
      "grad_norm": 1.1449085474014282,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.0681,
      "step": 40
    },
    {
      "epoch": 0.0041802524872502295,
      "grad_norm": 1.360563039779663,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 0.8906,
      "step": 50
    },
    {
      "epoch": 0.005016302984700276,
      "grad_norm": 0.8737388849258423,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 0.8052,
      "step": 60
    },
    {
      "epoch": 0.005852353482150322,
      "grad_norm": 0.50184565782547,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.7827,
      "step": 70
    },
    {
      "epoch": 0.006688403979600368,
      "grad_norm": 0.4593908190727234,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.7758,
      "step": 80
    },
    {
      "epoch": 0.007524454477050414,
      "grad_norm": 0.5259537100791931,
      "learning_rate": 1.005e-05,
      "loss": 0.7545,
      "step": 90
    },
    {
      "epoch": 0.008360504974500459,
      "grad_norm": 0.5117310285568237,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.7779,
      "step": 100
    },
    {
      "epoch": 0.009196555471950505,
      "grad_norm": 0.46054765582084656,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.7623,
      "step": 110
    },
    {
      "epoch": 0.010032605969400551,
      "grad_norm": 0.6909846067428589,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.7352,
      "step": 120
    },
    {
      "epoch": 0.010868656466850597,
      "grad_norm": 0.4003996253013611,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.7429,
      "step": 130
    },
    {
      "epoch": 0.011704706964300643,
      "grad_norm": 0.47065919637680054,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.7416,
      "step": 140
    },
    {
      "epoch": 0.01254075746175069,
      "grad_norm": 0.5040171146392822,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.7821,
      "step": 150
    },
    {
      "epoch": 0.013376807959200735,
      "grad_norm": 0.396039217710495,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.6167,
      "step": 160
    },
    {
      "epoch": 0.014212858456650782,
      "grad_norm": 0.42775487899780273,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.703,
      "step": 170
    },
    {
      "epoch": 0.015048908954100828,
      "grad_norm": 0.5157128572463989,
      "learning_rate": 2e-05,
      "loss": 0.7219,
      "step": 180
    },
    {
      "epoch": 0.015884959451550874,
      "grad_norm": 0.5191335082054138,
      "learning_rate": 2e-05,
      "loss": 0.6895,
      "step": 190
    },
    {
      "epoch": 0.016721009949000918,
      "grad_norm": 0.44552481174468994,
      "learning_rate": 2e-05,
      "loss": 0.7773,
      "step": 200
    },
    {
      "epoch": 0.017557060446450966,
      "grad_norm": 0.3954372704029083,
      "learning_rate": 2e-05,
      "loss": 0.7441,
      "step": 210
    },
    {
      "epoch": 0.01839311094390101,
      "grad_norm": 0.45830389857292175,
      "learning_rate": 2e-05,
      "loss": 0.7504,
      "step": 220
    },
    {
      "epoch": 0.019229161441351058,
      "grad_norm": 0.4249721169471741,
      "learning_rate": 2e-05,
      "loss": 0.7186,
      "step": 230
    },
    {
      "epoch": 0.020065211938801102,
      "grad_norm": 0.4513215720653534,
      "learning_rate": 2e-05,
      "loss": 0.745,
      "step": 240
    },
    {
      "epoch": 0.02090126243625115,
      "grad_norm": 0.43894580006599426,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 250
    },
    {
      "epoch": 0.021737312933701194,
      "grad_norm": 7.123316287994385,
      "learning_rate": 2e-05,
      "loss": 0.7635,
      "step": 260
    },
    {
      "epoch": 0.022573363431151242,
      "grad_norm": 0.4213659465312958,
      "learning_rate": 2e-05,
      "loss": 0.7474,
      "step": 270
    },
    {
      "epoch": 0.023409413928601287,
      "grad_norm": 0.4028798043727875,
      "learning_rate": 2e-05,
      "loss": 0.7168,
      "step": 280
    },
    {
      "epoch": 0.024245464426051334,
      "grad_norm": 0.3978917598724365,
      "learning_rate": 2e-05,
      "loss": 0.7272,
      "step": 290
    },
    {
      "epoch": 0.02508151492350138,
      "grad_norm": 0.45915454626083374,
      "learning_rate": 2e-05,
      "loss": 0.7791,
      "step": 300
    },
    {
      "epoch": 0.025917565420951427,
      "grad_norm": 0.44984596967697144,
      "learning_rate": 2e-05,
      "loss": 0.7934,
      "step": 310
    },
    {
      "epoch": 0.02675361591840147,
      "grad_norm": 0.43878060579299927,
      "learning_rate": 2e-05,
      "loss": 0.7644,
      "step": 320
    },
    {
      "epoch": 0.02758966641585152,
      "grad_norm": 0.4639083743095398,
      "learning_rate": 2e-05,
      "loss": 0.7801,
      "step": 330
    },
    {
      "epoch": 0.028425716913301563,
      "grad_norm": 0.43591466546058655,
      "learning_rate": 2e-05,
      "loss": 0.7261,
      "step": 340
    },
    {
      "epoch": 0.02926176741075161,
      "grad_norm": 0.5230687260627747,
      "learning_rate": 2e-05,
      "loss": 0.7732,
      "step": 350
    },
    {
      "epoch": 0.030097817908201655,
      "grad_norm": 0.4230197072029114,
      "learning_rate": 2e-05,
      "loss": 0.7576,
      "step": 360
    },
    {
      "epoch": 0.030933868405651703,
      "grad_norm": 0.44197097420692444,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 370
    },
    {
      "epoch": 0.03176991890310175,
      "grad_norm": 0.4144206643104553,
      "learning_rate": 2e-05,
      "loss": 0.7316,
      "step": 380
    },
    {
      "epoch": 0.03260596940055179,
      "grad_norm": 0.4325674772262573,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 390
    },
    {
      "epoch": 0.033442019898001836,
      "grad_norm": 0.3876762390136719,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 400
    },
    {
      "epoch": 0.03427807039545189,
      "grad_norm": 0.4181121587753296,
      "learning_rate": 2e-05,
      "loss": 0.7415,
      "step": 410
    },
    {
      "epoch": 0.03511412089290193,
      "grad_norm": 0.6723706722259521,
      "learning_rate": 2e-05,
      "loss": 0.7408,
      "step": 420
    },
    {
      "epoch": 0.035950171390351976,
      "grad_norm": 0.42723605036735535,
      "learning_rate": 2e-05,
      "loss": 0.7381,
      "step": 430
    },
    {
      "epoch": 0.03678622188780202,
      "grad_norm": 0.40303024649620056,
      "learning_rate": 2e-05,
      "loss": 0.7108,
      "step": 440
    },
    {
      "epoch": 0.03762227238525207,
      "grad_norm": 0.4335564076900482,
      "learning_rate": 2e-05,
      "loss": 0.7368,
      "step": 450
    },
    {
      "epoch": 0.038458322882702116,
      "grad_norm": 0.4370777904987335,
      "learning_rate": 2e-05,
      "loss": 0.7813,
      "step": 460
    },
    {
      "epoch": 0.03929437338015216,
      "grad_norm": 0.4104174077510834,
      "learning_rate": 2e-05,
      "loss": 0.7359,
      "step": 470
    },
    {
      "epoch": 0.040130423877602205,
      "grad_norm": 0.4259282946586609,
      "learning_rate": 2e-05,
      "loss": 0.7206,
      "step": 480
    },
    {
      "epoch": 0.040966474375052256,
      "grad_norm": 0.39744478464126587,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 490
    },
    {
      "epoch": 0.0418025248725023,
      "grad_norm": 0.42015591263771057,
      "learning_rate": 2e-05,
      "loss": 0.7416,
      "step": 500
    },
    {
      "epoch": 0.042638575369952345,
      "grad_norm": 0.379894882440567,
      "learning_rate": 2e-05,
      "loss": 0.7315,
      "step": 510
    },
    {
      "epoch": 0.04347462586740239,
      "grad_norm": 0.4095825254917145,
      "learning_rate": 2e-05,
      "loss": 0.7185,
      "step": 520
    },
    {
      "epoch": 0.04431067636485244,
      "grad_norm": 0.4044616222381592,
      "learning_rate": 2e-05,
      "loss": 0.7443,
      "step": 530
    },
    {
      "epoch": 0.045146726862302484,
      "grad_norm": 0.4145281910896301,
      "learning_rate": 2e-05,
      "loss": 0.7009,
      "step": 540
    },
    {
      "epoch": 0.04598277735975253,
      "grad_norm": 0.6660082936286926,
      "learning_rate": 2e-05,
      "loss": 0.7139,
      "step": 550
    },
    {
      "epoch": 0.04681882785720257,
      "grad_norm": 0.39196470379829407,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 560
    },
    {
      "epoch": 0.047654878354652624,
      "grad_norm": 0.39314568042755127,
      "learning_rate": 2e-05,
      "loss": 0.7562,
      "step": 570
    },
    {
      "epoch": 0.04849092885210267,
      "grad_norm": 0.4184809923171997,
      "learning_rate": 2e-05,
      "loss": 0.7281,
      "step": 580
    },
    {
      "epoch": 0.04932697934955271,
      "grad_norm": 0.4390700161457062,
      "learning_rate": 2e-05,
      "loss": 0.7206,
      "step": 590
    },
    {
      "epoch": 0.05016302984700276,
      "grad_norm": 0.39948198199272156,
      "learning_rate": 2e-05,
      "loss": 0.7311,
      "step": 600
    },
    {
      "epoch": 0.0509990803444528,
      "grad_norm": 0.3739895820617676,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 610
    },
    {
      "epoch": 0.05183513084190285,
      "grad_norm": 0.37271761894226074,
      "learning_rate": 2e-05,
      "loss": 0.7092,
      "step": 620
    },
    {
      "epoch": 0.0526711813393529,
      "grad_norm": 0.404062420129776,
      "learning_rate": 2e-05,
      "loss": 0.6953,
      "step": 630
    },
    {
      "epoch": 0.05350723183680294,
      "grad_norm": 0.4164421260356903,
      "learning_rate": 2e-05,
      "loss": 0.7431,
      "step": 640
    },
    {
      "epoch": 0.054343282334252986,
      "grad_norm": 0.4303136169910431,
      "learning_rate": 2e-05,
      "loss": 0.7299,
      "step": 650
    },
    {
      "epoch": 0.05517933283170304,
      "grad_norm": 0.4234829843044281,
      "learning_rate": 2e-05,
      "loss": 0.7442,
      "step": 660
    },
    {
      "epoch": 0.05601538332915308,
      "grad_norm": 0.3876648545265198,
      "learning_rate": 2e-05,
      "loss": 0.7146,
      "step": 670
    },
    {
      "epoch": 0.056851433826603126,
      "grad_norm": 0.4167863726615906,
      "learning_rate": 2e-05,
      "loss": 0.7098,
      "step": 680
    },
    {
      "epoch": 0.05768748432405317,
      "grad_norm": 0.44938138127326965,
      "learning_rate": 2e-05,
      "loss": 0.7457,
      "step": 690
    },
    {
      "epoch": 0.05852353482150322,
      "grad_norm": 0.4265012741088867,
      "learning_rate": 2e-05,
      "loss": 0.7091,
      "step": 700
    },
    {
      "epoch": 0.059359585318953266,
      "grad_norm": 0.42332151532173157,
      "learning_rate": 2e-05,
      "loss": 0.7499,
      "step": 710
    },
    {
      "epoch": 0.06019563581640331,
      "grad_norm": 0.4157474637031555,
      "learning_rate": 2e-05,
      "loss": 0.6786,
      "step": 720
    },
    {
      "epoch": 0.061031686313853355,
      "grad_norm": 0.370686799287796,
      "learning_rate": 2e-05,
      "loss": 0.7323,
      "step": 730
    },
    {
      "epoch": 0.061867736811303406,
      "grad_norm": 0.3953830599784851,
      "learning_rate": 2e-05,
      "loss": 0.7309,
      "step": 740
    },
    {
      "epoch": 0.06270378730875345,
      "grad_norm": 0.40635451674461365,
      "learning_rate": 2e-05,
      "loss": 0.732,
      "step": 750
    },
    {
      "epoch": 0.0635398378062035,
      "grad_norm": 0.4116203486919403,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 760
    },
    {
      "epoch": 0.06437588830365354,
      "grad_norm": 0.3634946048259735,
      "learning_rate": 2e-05,
      "loss": 0.679,
      "step": 770
    },
    {
      "epoch": 0.06521193880110358,
      "grad_norm": 0.4082890450954437,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 780
    },
    {
      "epoch": 0.06604798929855363,
      "grad_norm": 0.4118865728378296,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 790
    },
    {
      "epoch": 0.06688403979600367,
      "grad_norm": 0.40019503235816956,
      "learning_rate": 2e-05,
      "loss": 0.7234,
      "step": 800
    },
    {
      "epoch": 0.06772009029345373,
      "grad_norm": 0.41497135162353516,
      "learning_rate": 2e-05,
      "loss": 0.7026,
      "step": 810
    },
    {
      "epoch": 0.06855614079090377,
      "grad_norm": 0.37976041436195374,
      "learning_rate": 2e-05,
      "loss": 0.6984,
      "step": 820
    },
    {
      "epoch": 0.06939219128835382,
      "grad_norm": 0.4239822030067444,
      "learning_rate": 2e-05,
      "loss": 0.7123,
      "step": 830
    },
    {
      "epoch": 0.07022824178580386,
      "grad_norm": 0.40763989090919495,
      "learning_rate": 2e-05,
      "loss": 0.6923,
      "step": 840
    },
    {
      "epoch": 0.07106429228325391,
      "grad_norm": 0.40808385610580444,
      "learning_rate": 2e-05,
      "loss": 0.7095,
      "step": 850
    },
    {
      "epoch": 0.07190034278070395,
      "grad_norm": 0.44947150349617004,
      "learning_rate": 2e-05,
      "loss": 0.7188,
      "step": 860
    },
    {
      "epoch": 0.072736393278154,
      "grad_norm": 0.3837183713912964,
      "learning_rate": 2e-05,
      "loss": 0.7442,
      "step": 870
    },
    {
      "epoch": 0.07357244377560404,
      "grad_norm": 0.4867953956127167,
      "learning_rate": 2e-05,
      "loss": 0.717,
      "step": 880
    },
    {
      "epoch": 0.0744084942730541,
      "grad_norm": 0.4862518310546875,
      "learning_rate": 2e-05,
      "loss": 0.6994,
      "step": 890
    },
    {
      "epoch": 0.07524454477050414,
      "grad_norm": 0.4381377398967743,
      "learning_rate": 2e-05,
      "loss": 0.7172,
      "step": 900
    },
    {
      "epoch": 0.07608059526795419,
      "grad_norm": 0.44232502579689026,
      "learning_rate": 2e-05,
      "loss": 0.7078,
      "step": 910
    },
    {
      "epoch": 0.07691664576540423,
      "grad_norm": 0.39795035123825073,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 920
    },
    {
      "epoch": 0.07775269626285428,
      "grad_norm": 0.3873278796672821,
      "learning_rate": 2e-05,
      "loss": 0.7302,
      "step": 930
    },
    {
      "epoch": 0.07858874676030432,
      "grad_norm": 0.4043973684310913,
      "learning_rate": 2e-05,
      "loss": 0.7497,
      "step": 940
    },
    {
      "epoch": 0.07942479725775436,
      "grad_norm": 0.43075886368751526,
      "learning_rate": 2e-05,
      "loss": 0.726,
      "step": 950
    },
    {
      "epoch": 0.08026084775520441,
      "grad_norm": 0.4056244492530823,
      "learning_rate": 2e-05,
      "loss": 0.6807,
      "step": 960
    },
    {
      "epoch": 0.08109689825265445,
      "grad_norm": 0.4083726406097412,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 970
    },
    {
      "epoch": 0.08193294875010451,
      "grad_norm": 0.4576243460178375,
      "learning_rate": 2e-05,
      "loss": 0.6996,
      "step": 980
    },
    {
      "epoch": 0.08276899924755456,
      "grad_norm": 0.41480258107185364,
      "learning_rate": 2e-05,
      "loss": 0.6787,
      "step": 990
    },
    {
      "epoch": 0.0836050497450046,
      "grad_norm": 0.38216501474380493,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 1000
    },
    {
      "epoch": 0.08444110024245464,
      "grad_norm": 0.37291425466537476,
      "learning_rate": 2e-05,
      "loss": 0.7013,
      "step": 1010
    },
    {
      "epoch": 0.08527715073990469,
      "grad_norm": 0.44588473439216614,
      "learning_rate": 2e-05,
      "loss": 0.7362,
      "step": 1020
    },
    {
      "epoch": 0.08611320123735473,
      "grad_norm": 0.38740336894989014,
      "learning_rate": 2e-05,
      "loss": 0.7268,
      "step": 1030
    },
    {
      "epoch": 0.08694925173480478,
      "grad_norm": 0.3775883913040161,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 1040
    },
    {
      "epoch": 0.08778530223225482,
      "grad_norm": 0.4709722399711609,
      "learning_rate": 2e-05,
      "loss": 0.731,
      "step": 1050
    },
    {
      "epoch": 0.08862135272970488,
      "grad_norm": 0.4146833121776581,
      "learning_rate": 2e-05,
      "loss": 0.7448,
      "step": 1060
    },
    {
      "epoch": 0.08945740322715492,
      "grad_norm": 0.39623764157295227,
      "learning_rate": 2e-05,
      "loss": 0.6927,
      "step": 1070
    },
    {
      "epoch": 0.09029345372460497,
      "grad_norm": 0.3726375102996826,
      "learning_rate": 2e-05,
      "loss": 0.7264,
      "step": 1080
    },
    {
      "epoch": 0.09112950422205501,
      "grad_norm": 0.3898238241672516,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 1090
    },
    {
      "epoch": 0.09196555471950506,
      "grad_norm": 0.3660300076007843,
      "learning_rate": 2e-05,
      "loss": 0.7464,
      "step": 1100
    },
    {
      "epoch": 0.0928016052169551,
      "grad_norm": 0.43982595205307007,
      "learning_rate": 2e-05,
      "loss": 0.7031,
      "step": 1110
    },
    {
      "epoch": 0.09363765571440515,
      "grad_norm": 0.41887497901916504,
      "learning_rate": 2e-05,
      "loss": 0.6931,
      "step": 1120
    },
    {
      "epoch": 0.09447370621185519,
      "grad_norm": 0.3669204115867615,
      "learning_rate": 2e-05,
      "loss": 0.7256,
      "step": 1130
    },
    {
      "epoch": 0.09530975670930525,
      "grad_norm": 0.39729389548301697,
      "learning_rate": 2e-05,
      "loss": 0.7253,
      "step": 1140
    },
    {
      "epoch": 0.0961458072067553,
      "grad_norm": 0.3818066418170929,
      "learning_rate": 2e-05,
      "loss": 0.6842,
      "step": 1150
    },
    {
      "epoch": 0.09698185770420534,
      "grad_norm": 0.3896878659725189,
      "learning_rate": 2e-05,
      "loss": 0.6928,
      "step": 1160
    },
    {
      "epoch": 0.09781790820165538,
      "grad_norm": 0.37976863980293274,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 1170
    },
    {
      "epoch": 0.09865395869910543,
      "grad_norm": 0.37471631169319153,
      "learning_rate": 2e-05,
      "loss": 0.6839,
      "step": 1180
    },
    {
      "epoch": 0.09949000919655547,
      "grad_norm": 0.3750368356704712,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 1190
    },
    {
      "epoch": 0.10032605969400551,
      "grad_norm": 0.37220221757888794,
      "learning_rate": 2e-05,
      "loss": 0.7175,
      "step": 1200
    },
    {
      "epoch": 0.10116211019145556,
      "grad_norm": 0.407367080450058,
      "learning_rate": 2e-05,
      "loss": 0.706,
      "step": 1210
    },
    {
      "epoch": 0.1019981606889056,
      "grad_norm": 0.35999026894569397,
      "learning_rate": 2e-05,
      "loss": 0.7197,
      "step": 1220
    },
    {
      "epoch": 0.10283421118635566,
      "grad_norm": 0.36594730615615845,
      "learning_rate": 2e-05,
      "loss": 0.6944,
      "step": 1230
    },
    {
      "epoch": 0.1036702616838057,
      "grad_norm": 0.3911300301551819,
      "learning_rate": 2e-05,
      "loss": 0.7117,
      "step": 1240
    },
    {
      "epoch": 0.10450631218125575,
      "grad_norm": 0.4042055010795593,
      "learning_rate": 2e-05,
      "loss": 0.7157,
      "step": 1250
    },
    {
      "epoch": 0.1053423626787058,
      "grad_norm": 0.4203055500984192,
      "learning_rate": 2e-05,
      "loss": 0.7426,
      "step": 1260
    },
    {
      "epoch": 0.10617841317615584,
      "grad_norm": 0.3811381459236145,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 1270
    },
    {
      "epoch": 0.10701446367360588,
      "grad_norm": 0.35855522751808167,
      "learning_rate": 2e-05,
      "loss": 0.7526,
      "step": 1280
    },
    {
      "epoch": 0.10785051417105593,
      "grad_norm": 0.3886844217777252,
      "learning_rate": 2e-05,
      "loss": 0.7052,
      "step": 1290
    },
    {
      "epoch": 0.10868656466850597,
      "grad_norm": 0.3775457739830017,
      "learning_rate": 2e-05,
      "loss": 0.71,
      "step": 1300
    },
    {
      "epoch": 0.10952261516595603,
      "grad_norm": 0.4011971652507782,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 1310
    },
    {
      "epoch": 0.11035866566340607,
      "grad_norm": 0.40073534846305847,
      "learning_rate": 2e-05,
      "loss": 0.7294,
      "step": 1320
    },
    {
      "epoch": 0.11119471616085612,
      "grad_norm": 0.6666777729988098,
      "learning_rate": 2e-05,
      "loss": 0.7135,
      "step": 1330
    },
    {
      "epoch": 0.11203076665830616,
      "grad_norm": 0.40671786665916443,
      "learning_rate": 2e-05,
      "loss": 0.7055,
      "step": 1340
    },
    {
      "epoch": 0.11286681715575621,
      "grad_norm": 0.35212790966033936,
      "learning_rate": 2e-05,
      "loss": 0.7068,
      "step": 1350
    },
    {
      "epoch": 0.11370286765320625,
      "grad_norm": 0.39304518699645996,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 1360
    },
    {
      "epoch": 0.1145389181506563,
      "grad_norm": 0.40921252965927124,
      "learning_rate": 2e-05,
      "loss": 0.7149,
      "step": 1370
    },
    {
      "epoch": 0.11537496864810634,
      "grad_norm": 0.3641948401927948,
      "learning_rate": 2e-05,
      "loss": 0.6947,
      "step": 1380
    },
    {
      "epoch": 0.11621101914555639,
      "grad_norm": 0.3776053786277771,
      "learning_rate": 2e-05,
      "loss": 0.7098,
      "step": 1390
    },
    {
      "epoch": 0.11704706964300644,
      "grad_norm": 0.40454384684562683,
      "learning_rate": 2e-05,
      "loss": 0.6876,
      "step": 1400
    },
    {
      "epoch": 0.11788312014045649,
      "grad_norm": 0.3694128692150116,
      "learning_rate": 2e-05,
      "loss": 0.7112,
      "step": 1410
    },
    {
      "epoch": 0.11871917063790653,
      "grad_norm": 0.4031735956668854,
      "learning_rate": 2e-05,
      "loss": 0.6743,
      "step": 1420
    },
    {
      "epoch": 0.11955522113535658,
      "grad_norm": 0.36599618196487427,
      "learning_rate": 2e-05,
      "loss": 0.7177,
      "step": 1430
    },
    {
      "epoch": 0.12039127163280662,
      "grad_norm": 0.3794446587562561,
      "learning_rate": 2e-05,
      "loss": 0.7672,
      "step": 1440
    },
    {
      "epoch": 0.12122732213025667,
      "grad_norm": 0.42689165472984314,
      "learning_rate": 2e-05,
      "loss": 0.6943,
      "step": 1450
    },
    {
      "epoch": 0.12206337262770671,
      "grad_norm": 0.39176973700523376,
      "learning_rate": 2e-05,
      "loss": 0.7454,
      "step": 1460
    },
    {
      "epoch": 0.12289942312515675,
      "grad_norm": 0.38585641980171204,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 1470
    },
    {
      "epoch": 0.12373547362260681,
      "grad_norm": 0.3828059434890747,
      "learning_rate": 2e-05,
      "loss": 0.7162,
      "step": 1480
    },
    {
      "epoch": 0.12457152412005686,
      "grad_norm": 0.36190828680992126,
      "learning_rate": 2e-05,
      "loss": 0.7189,
      "step": 1490
    },
    {
      "epoch": 0.1254075746175069,
      "grad_norm": 0.35898563265800476,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 1500
    },
    {
      "epoch": 0.12624362511495693,
      "grad_norm": 0.3842126429080963,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 1510
    },
    {
      "epoch": 0.127079675612407,
      "grad_norm": 0.38555908203125,
      "learning_rate": 2e-05,
      "loss": 0.6886,
      "step": 1520
    },
    {
      "epoch": 0.12791572610985705,
      "grad_norm": 0.41499239206314087,
      "learning_rate": 2e-05,
      "loss": 0.7005,
      "step": 1530
    },
    {
      "epoch": 0.12875177660730708,
      "grad_norm": 0.3923279345035553,
      "learning_rate": 2e-05,
      "loss": 0.6807,
      "step": 1540
    },
    {
      "epoch": 0.12958782710475714,
      "grad_norm": 0.38119590282440186,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 1550
    },
    {
      "epoch": 0.13042387760220717,
      "grad_norm": 0.44171908497810364,
      "learning_rate": 2e-05,
      "loss": 0.711,
      "step": 1560
    },
    {
      "epoch": 0.13125992809965722,
      "grad_norm": 0.3561314642429352,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 1570
    },
    {
      "epoch": 0.13209597859710726,
      "grad_norm": 0.44831010699272156,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 1580
    },
    {
      "epoch": 0.1329320290945573,
      "grad_norm": 0.42015114426612854,
      "learning_rate": 2e-05,
      "loss": 0.7072,
      "step": 1590
    },
    {
      "epoch": 0.13376807959200734,
      "grad_norm": 0.4430758059024811,
      "learning_rate": 2e-05,
      "loss": 0.7187,
      "step": 1600
    },
    {
      "epoch": 0.1346041300894574,
      "grad_norm": 0.37224727869033813,
      "learning_rate": 2e-05,
      "loss": 0.7045,
      "step": 1610
    },
    {
      "epoch": 0.13544018058690746,
      "grad_norm": 0.3727637529373169,
      "learning_rate": 2e-05,
      "loss": 0.7151,
      "step": 1620
    },
    {
      "epoch": 0.1362762310843575,
      "grad_norm": 0.36329200863838196,
      "learning_rate": 2e-05,
      "loss": 0.7359,
      "step": 1630
    },
    {
      "epoch": 0.13711228158180755,
      "grad_norm": 0.36346569657325745,
      "learning_rate": 2e-05,
      "loss": 0.689,
      "step": 1640
    },
    {
      "epoch": 0.13794833207925758,
      "grad_norm": 0.3709527254104614,
      "learning_rate": 2e-05,
      "loss": 0.7124,
      "step": 1650
    },
    {
      "epoch": 0.13878438257670764,
      "grad_norm": 0.41566887497901917,
      "learning_rate": 2e-05,
      "loss": 0.708,
      "step": 1660
    },
    {
      "epoch": 0.13962043307415767,
      "grad_norm": 0.3450981378555298,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 1670
    },
    {
      "epoch": 0.14045648357160773,
      "grad_norm": 0.3517451584339142,
      "learning_rate": 2e-05,
      "loss": 0.7031,
      "step": 1680
    },
    {
      "epoch": 0.14129253406905778,
      "grad_norm": 0.400630384683609,
      "learning_rate": 2e-05,
      "loss": 0.7124,
      "step": 1690
    },
    {
      "epoch": 0.14212858456650782,
      "grad_norm": 0.4013160765171051,
      "learning_rate": 2e-05,
      "loss": 0.7085,
      "step": 1700
    },
    {
      "epoch": 0.14296463506395787,
      "grad_norm": 0.3802119791507721,
      "learning_rate": 2e-05,
      "loss": 0.7056,
      "step": 1710
    },
    {
      "epoch": 0.1438006855614079,
      "grad_norm": 0.4005148112773895,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 1720
    },
    {
      "epoch": 0.14463673605885796,
      "grad_norm": 0.36928942799568176,
      "learning_rate": 2e-05,
      "loss": 0.6813,
      "step": 1730
    },
    {
      "epoch": 0.145472786556308,
      "grad_norm": 0.37812256813049316,
      "learning_rate": 2e-05,
      "loss": 0.7049,
      "step": 1740
    },
    {
      "epoch": 0.14630883705375805,
      "grad_norm": 0.3583225607872009,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 1750
    },
    {
      "epoch": 0.14714488755120808,
      "grad_norm": 0.3970085680484772,
      "learning_rate": 2e-05,
      "loss": 0.6873,
      "step": 1760
    },
    {
      "epoch": 0.14798093804865814,
      "grad_norm": 0.38219332695007324,
      "learning_rate": 2e-05,
      "loss": 0.6963,
      "step": 1770
    },
    {
      "epoch": 0.1488169885461082,
      "grad_norm": 0.362723171710968,
      "learning_rate": 2e-05,
      "loss": 0.7162,
      "step": 1780
    },
    {
      "epoch": 0.14965303904355823,
      "grad_norm": 0.43519818782806396,
      "learning_rate": 2e-05,
      "loss": 0.7274,
      "step": 1790
    },
    {
      "epoch": 0.1504890895410083,
      "grad_norm": 0.36672765016555786,
      "learning_rate": 2e-05,
      "loss": 0.746,
      "step": 1800
    },
    {
      "epoch": 0.15132514003845832,
      "grad_norm": 0.34433141350746155,
      "learning_rate": 2e-05,
      "loss": 0.6847,
      "step": 1810
    },
    {
      "epoch": 0.15216119053590837,
      "grad_norm": 0.3767019510269165,
      "learning_rate": 2e-05,
      "loss": 0.7217,
      "step": 1820
    },
    {
      "epoch": 0.1529972410333584,
      "grad_norm": 0.3689115047454834,
      "learning_rate": 2e-05,
      "loss": 0.7378,
      "step": 1830
    },
    {
      "epoch": 0.15383329153080846,
      "grad_norm": 0.36576899886131287,
      "learning_rate": 2e-05,
      "loss": 0.7198,
      "step": 1840
    },
    {
      "epoch": 0.1546693420282585,
      "grad_norm": 0.4920649230480194,
      "learning_rate": 2e-05,
      "loss": 0.6827,
      "step": 1850
    },
    {
      "epoch": 0.15550539252570855,
      "grad_norm": 0.35931605100631714,
      "learning_rate": 2e-05,
      "loss": 0.7097,
      "step": 1860
    },
    {
      "epoch": 0.1563414430231586,
      "grad_norm": 0.36113813519477844,
      "learning_rate": 2e-05,
      "loss": 0.7307,
      "step": 1870
    },
    {
      "epoch": 0.15717749352060864,
      "grad_norm": 0.3809521496295929,
      "learning_rate": 2e-05,
      "loss": 0.7038,
      "step": 1880
    },
    {
      "epoch": 0.1580135440180587,
      "grad_norm": 0.4045011103153229,
      "learning_rate": 2e-05,
      "loss": 0.7031,
      "step": 1890
    },
    {
      "epoch": 0.15884959451550873,
      "grad_norm": 0.3887285590171814,
      "learning_rate": 2e-05,
      "loss": 0.6739,
      "step": 1900
    },
    {
      "epoch": 0.1596856450129588,
      "grad_norm": 0.4463697671890259,
      "learning_rate": 2e-05,
      "loss": 0.7147,
      "step": 1910
    },
    {
      "epoch": 0.16052169551040882,
      "grad_norm": 0.3879319131374359,
      "learning_rate": 2e-05,
      "loss": 0.7375,
      "step": 1920
    },
    {
      "epoch": 0.16135774600785888,
      "grad_norm": 0.35314497351646423,
      "learning_rate": 2e-05,
      "loss": 0.6811,
      "step": 1930
    },
    {
      "epoch": 0.1621937965053089,
      "grad_norm": 0.4394693970680237,
      "learning_rate": 2e-05,
      "loss": 0.7156,
      "step": 1940
    },
    {
      "epoch": 0.16302984700275897,
      "grad_norm": 0.3700949549674988,
      "learning_rate": 2e-05,
      "loss": 0.7054,
      "step": 1950
    },
    {
      "epoch": 0.16386589750020902,
      "grad_norm": 0.3599170446395874,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 1960
    },
    {
      "epoch": 0.16470194799765905,
      "grad_norm": 0.41550520062446594,
      "learning_rate": 2e-05,
      "loss": 0.6874,
      "step": 1970
    },
    {
      "epoch": 0.1655379984951091,
      "grad_norm": 0.4028865694999695,
      "learning_rate": 2e-05,
      "loss": 0.7222,
      "step": 1980
    },
    {
      "epoch": 0.16637404899255914,
      "grad_norm": 0.3849325478076935,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 1990
    },
    {
      "epoch": 0.1672100994900092,
      "grad_norm": 0.34337013959884644,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 2000
    },
    {
      "epoch": 0.16804614998745923,
      "grad_norm": 0.3867923319339752,
      "learning_rate": 2e-05,
      "loss": 0.6965,
      "step": 2010
    },
    {
      "epoch": 0.1688822004849093,
      "grad_norm": 0.3923274874687195,
      "learning_rate": 2e-05,
      "loss": 0.7098,
      "step": 2020
    },
    {
      "epoch": 0.16971825098235935,
      "grad_norm": 0.3679618239402771,
      "learning_rate": 2e-05,
      "loss": 0.6978,
      "step": 2030
    },
    {
      "epoch": 0.17055430147980938,
      "grad_norm": 0.39310866594314575,
      "learning_rate": 2e-05,
      "loss": 0.71,
      "step": 2040
    },
    {
      "epoch": 0.17139035197725944,
      "grad_norm": 0.3967050313949585,
      "learning_rate": 2e-05,
      "loss": 0.7243,
      "step": 2050
    },
    {
      "epoch": 0.17222640247470947,
      "grad_norm": 0.40121224522590637,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 2060
    },
    {
      "epoch": 0.17306245297215953,
      "grad_norm": 0.38088318705558777,
      "learning_rate": 2e-05,
      "loss": 0.6401,
      "step": 2070
    },
    {
      "epoch": 0.17389850346960956,
      "grad_norm": 0.3435222804546356,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 2080
    },
    {
      "epoch": 0.1747345539670596,
      "grad_norm": 0.4220814108848572,
      "learning_rate": 2e-05,
      "loss": 0.7141,
      "step": 2090
    },
    {
      "epoch": 0.17557060446450964,
      "grad_norm": 0.35260170698165894,
      "learning_rate": 2e-05,
      "loss": 0.6713,
      "step": 2100
    },
    {
      "epoch": 0.1764066549619597,
      "grad_norm": 0.40899789333343506,
      "learning_rate": 2e-05,
      "loss": 0.7129,
      "step": 2110
    },
    {
      "epoch": 0.17724270545940976,
      "grad_norm": 0.4075130522251129,
      "learning_rate": 2e-05,
      "loss": 0.6915,
      "step": 2120
    },
    {
      "epoch": 0.1780787559568598,
      "grad_norm": 0.4192414879798889,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 2130
    },
    {
      "epoch": 0.17891480645430985,
      "grad_norm": 0.38824033737182617,
      "learning_rate": 2e-05,
      "loss": 0.7058,
      "step": 2140
    },
    {
      "epoch": 0.17975085695175988,
      "grad_norm": 0.3785799443721771,
      "learning_rate": 2e-05,
      "loss": 0.7003,
      "step": 2150
    },
    {
      "epoch": 0.18058690744920994,
      "grad_norm": 0.3589932322502136,
      "learning_rate": 2e-05,
      "loss": 0.6703,
      "step": 2160
    },
    {
      "epoch": 0.18142295794665997,
      "grad_norm": 0.36105579137802124,
      "learning_rate": 2e-05,
      "loss": 0.6972,
      "step": 2170
    },
    {
      "epoch": 0.18225900844411003,
      "grad_norm": 0.3652353584766388,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 2180
    },
    {
      "epoch": 0.18309505894156006,
      "grad_norm": 0.40131404995918274,
      "learning_rate": 2e-05,
      "loss": 0.7082,
      "step": 2190
    },
    {
      "epoch": 0.18393110943901012,
      "grad_norm": 0.40471386909484863,
      "learning_rate": 2e-05,
      "loss": 0.696,
      "step": 2200
    },
    {
      "epoch": 0.18476715993646017,
      "grad_norm": 0.3407021462917328,
      "learning_rate": 2e-05,
      "loss": 0.653,
      "step": 2210
    },
    {
      "epoch": 0.1856032104339102,
      "grad_norm": 0.37151047587394714,
      "learning_rate": 2e-05,
      "loss": 0.6928,
      "step": 2220
    },
    {
      "epoch": 0.18643926093136026,
      "grad_norm": 0.3697320818901062,
      "learning_rate": 2e-05,
      "loss": 0.692,
      "step": 2230
    },
    {
      "epoch": 0.1872753114288103,
      "grad_norm": 0.3968718349933624,
      "learning_rate": 2e-05,
      "loss": 0.6909,
      "step": 2240
    },
    {
      "epoch": 0.18811136192626035,
      "grad_norm": 0.3623242676258087,
      "learning_rate": 2e-05,
      "loss": 0.7195,
      "step": 2250
    },
    {
      "epoch": 0.18894741242371038,
      "grad_norm": 0.3716718852519989,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 2260
    },
    {
      "epoch": 0.18978346292116044,
      "grad_norm": 0.38747113943099976,
      "learning_rate": 2e-05,
      "loss": 0.7033,
      "step": 2270
    },
    {
      "epoch": 0.1906195134186105,
      "grad_norm": 0.3699060380458832,
      "learning_rate": 2e-05,
      "loss": 0.7035,
      "step": 2280
    },
    {
      "epoch": 0.19145556391606053,
      "grad_norm": 0.3485587239265442,
      "learning_rate": 2e-05,
      "loss": 0.6982,
      "step": 2290
    },
    {
      "epoch": 0.1922916144135106,
      "grad_norm": 0.35775020718574524,
      "learning_rate": 2e-05,
      "loss": 0.6977,
      "step": 2300
    },
    {
      "epoch": 0.19312766491096062,
      "grad_norm": 0.386114239692688,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 2310
    },
    {
      "epoch": 0.19396371540841068,
      "grad_norm": 0.4111344814300537,
      "learning_rate": 2e-05,
      "loss": 0.7134,
      "step": 2320
    },
    {
      "epoch": 0.1947997659058607,
      "grad_norm": 0.3693266212940216,
      "learning_rate": 2e-05,
      "loss": 0.6793,
      "step": 2330
    },
    {
      "epoch": 0.19563581640331076,
      "grad_norm": 0.3926129639148712,
      "learning_rate": 2e-05,
      "loss": 0.6856,
      "step": 2340
    },
    {
      "epoch": 0.1964718669007608,
      "grad_norm": 0.38753512501716614,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 2350
    },
    {
      "epoch": 0.19730791739821085,
      "grad_norm": 0.3788638710975647,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 2360
    },
    {
      "epoch": 0.1981439678956609,
      "grad_norm": 0.3338569104671478,
      "learning_rate": 2e-05,
      "loss": 0.6775,
      "step": 2370
    },
    {
      "epoch": 0.19898001839311094,
      "grad_norm": 0.366208016872406,
      "learning_rate": 2e-05,
      "loss": 0.7282,
      "step": 2380
    },
    {
      "epoch": 0.199816068890561,
      "grad_norm": 0.379518985748291,
      "learning_rate": 2e-05,
      "loss": 0.6781,
      "step": 2390
    },
    {
      "epoch": 0.20065211938801103,
      "grad_norm": 0.3841893970966339,
      "learning_rate": 2e-05,
      "loss": 0.7026,
      "step": 2400
    },
    {
      "epoch": 0.2014881698854611,
      "grad_norm": 0.3684847950935364,
      "learning_rate": 2e-05,
      "loss": 0.6626,
      "step": 2410
    },
    {
      "epoch": 0.20232422038291112,
      "grad_norm": 0.3933838903903961,
      "learning_rate": 2e-05,
      "loss": 0.7178,
      "step": 2420
    },
    {
      "epoch": 0.20316027088036118,
      "grad_norm": 0.3518088757991791,
      "learning_rate": 2e-05,
      "loss": 0.6663,
      "step": 2430
    },
    {
      "epoch": 0.2039963213778112,
      "grad_norm": 0.383081316947937,
      "learning_rate": 2e-05,
      "loss": 0.6947,
      "step": 2440
    },
    {
      "epoch": 0.20483237187526127,
      "grad_norm": 0.34842392802238464,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 2450
    },
    {
      "epoch": 0.20566842237271132,
      "grad_norm": 0.36848798394203186,
      "learning_rate": 2e-05,
      "loss": 0.6873,
      "step": 2460
    },
    {
      "epoch": 0.20650447287016135,
      "grad_norm": 0.36847376823425293,
      "learning_rate": 2e-05,
      "loss": 0.73,
      "step": 2470
    },
    {
      "epoch": 0.2073405233676114,
      "grad_norm": 0.4002387523651123,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 2480
    },
    {
      "epoch": 0.20817657386506144,
      "grad_norm": 0.383114755153656,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 2490
    },
    {
      "epoch": 0.2090126243625115,
      "grad_norm": 0.3491422235965729,
      "learning_rate": 2e-05,
      "loss": 0.6831,
      "step": 2500
    },
    {
      "epoch": 0.20984867485996153,
      "grad_norm": 0.3453516364097595,
      "learning_rate": 2e-05,
      "loss": 0.6988,
      "step": 2510
    },
    {
      "epoch": 0.2106847253574116,
      "grad_norm": 0.3652738332748413,
      "learning_rate": 2e-05,
      "loss": 0.7116,
      "step": 2520
    },
    {
      "epoch": 0.21152077585486162,
      "grad_norm": 0.3410079777240753,
      "learning_rate": 2e-05,
      "loss": 0.7418,
      "step": 2530
    },
    {
      "epoch": 0.21235682635231168,
      "grad_norm": 0.38150179386138916,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 2540
    },
    {
      "epoch": 0.21319287684976174,
      "grad_norm": 0.33772507309913635,
      "learning_rate": 2e-05,
      "loss": 0.6822,
      "step": 2550
    },
    {
      "epoch": 0.21402892734721177,
      "grad_norm": 0.40987348556518555,
      "learning_rate": 2e-05,
      "loss": 0.7091,
      "step": 2560
    },
    {
      "epoch": 0.21486497784466183,
      "grad_norm": 0.3767969012260437,
      "learning_rate": 2e-05,
      "loss": 0.7025,
      "step": 2570
    },
    {
      "epoch": 0.21570102834211186,
      "grad_norm": 0.3843465745449066,
      "learning_rate": 2e-05,
      "loss": 0.7031,
      "step": 2580
    },
    {
      "epoch": 0.21653707883956191,
      "grad_norm": 0.363743394613266,
      "learning_rate": 2e-05,
      "loss": 0.6946,
      "step": 2590
    },
    {
      "epoch": 0.21737312933701194,
      "grad_norm": 0.3590858280658722,
      "learning_rate": 2e-05,
      "loss": 0.683,
      "step": 2600
    },
    {
      "epoch": 0.218209179834462,
      "grad_norm": 0.5465121865272522,
      "learning_rate": 2e-05,
      "loss": 0.7204,
      "step": 2610
    },
    {
      "epoch": 0.21904523033191206,
      "grad_norm": 0.37804701924324036,
      "learning_rate": 2e-05,
      "loss": 0.6918,
      "step": 2620
    },
    {
      "epoch": 0.2198812808293621,
      "grad_norm": 0.37326765060424805,
      "learning_rate": 2e-05,
      "loss": 0.6885,
      "step": 2630
    },
    {
      "epoch": 0.22071733132681215,
      "grad_norm": 0.42487889528274536,
      "learning_rate": 2e-05,
      "loss": 0.6697,
      "step": 2640
    },
    {
      "epoch": 0.22155338182426218,
      "grad_norm": 0.3364788293838501,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 2650
    },
    {
      "epoch": 0.22238943232171224,
      "grad_norm": 0.34788456559181213,
      "learning_rate": 2e-05,
      "loss": 0.6872,
      "step": 2660
    },
    {
      "epoch": 0.22322548281916227,
      "grad_norm": 0.4006475508213043,
      "learning_rate": 2e-05,
      "loss": 0.6791,
      "step": 2670
    },
    {
      "epoch": 0.22406153331661233,
      "grad_norm": 0.3868175446987152,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 2680
    },
    {
      "epoch": 0.22489758381406236,
      "grad_norm": 0.38875290751457214,
      "learning_rate": 2e-05,
      "loss": 0.6633,
      "step": 2690
    },
    {
      "epoch": 0.22573363431151242,
      "grad_norm": 0.3829600214958191,
      "learning_rate": 2e-05,
      "loss": 0.6843,
      "step": 2700
    },
    {
      "epoch": 0.22656968480896247,
      "grad_norm": 0.4310683310031891,
      "learning_rate": 2e-05,
      "loss": 0.7004,
      "step": 2710
    },
    {
      "epoch": 0.2274057353064125,
      "grad_norm": 0.36083126068115234,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 2720
    },
    {
      "epoch": 0.22824178580386256,
      "grad_norm": 0.4124043583869934,
      "learning_rate": 2e-05,
      "loss": 0.6971,
      "step": 2730
    },
    {
      "epoch": 0.2290778363013126,
      "grad_norm": 0.4539124667644501,
      "learning_rate": 2e-05,
      "loss": 0.6848,
      "step": 2740
    },
    {
      "epoch": 0.22991388679876265,
      "grad_norm": 0.37063753604888916,
      "learning_rate": 2e-05,
      "loss": 0.6925,
      "step": 2750
    },
    {
      "epoch": 0.23074993729621268,
      "grad_norm": 0.3623632788658142,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 2760
    },
    {
      "epoch": 0.23158598779366274,
      "grad_norm": 0.44299641251564026,
      "learning_rate": 2e-05,
      "loss": 0.6755,
      "step": 2770
    },
    {
      "epoch": 0.23242203829111277,
      "grad_norm": 0.356118381023407,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 2780
    },
    {
      "epoch": 0.23325808878856283,
      "grad_norm": 0.38731667399406433,
      "learning_rate": 2e-05,
      "loss": 0.6911,
      "step": 2790
    },
    {
      "epoch": 0.2340941392860129,
      "grad_norm": 0.34089842438697815,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 2800
    },
    {
      "epoch": 0.23493018978346292,
      "grad_norm": 0.3597925007343292,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 2810
    },
    {
      "epoch": 0.23576624028091298,
      "grad_norm": 0.36081627011299133,
      "learning_rate": 2e-05,
      "loss": 0.6539,
      "step": 2820
    },
    {
      "epoch": 0.236602290778363,
      "grad_norm": 0.3887976408004761,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 2830
    },
    {
      "epoch": 0.23743834127581306,
      "grad_norm": 0.37405726313591003,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 2840
    },
    {
      "epoch": 0.2382743917732631,
      "grad_norm": 0.37012699246406555,
      "learning_rate": 2e-05,
      "loss": 0.6773,
      "step": 2850
    },
    {
      "epoch": 0.23911044227071315,
      "grad_norm": 0.39837878942489624,
      "learning_rate": 2e-05,
      "loss": 0.6733,
      "step": 2860
    },
    {
      "epoch": 0.2399464927681632,
      "grad_norm": 0.3539638817310333,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 2870
    },
    {
      "epoch": 0.24078254326561324,
      "grad_norm": 0.4221721291542053,
      "learning_rate": 2e-05,
      "loss": 0.6896,
      "step": 2880
    },
    {
      "epoch": 0.2416185937630633,
      "grad_norm": 0.35411736369132996,
      "learning_rate": 2e-05,
      "loss": 0.6895,
      "step": 2890
    },
    {
      "epoch": 0.24245464426051333,
      "grad_norm": 0.3807287812232971,
      "learning_rate": 2e-05,
      "loss": 0.6683,
      "step": 2900
    },
    {
      "epoch": 0.2432906947579634,
      "grad_norm": 0.33664026856422424,
      "learning_rate": 2e-05,
      "loss": 0.6926,
      "step": 2910
    },
    {
      "epoch": 0.24412674525541342,
      "grad_norm": 0.3574134111404419,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 2920
    },
    {
      "epoch": 0.24496279575286348,
      "grad_norm": 0.34514501690864563,
      "learning_rate": 2e-05,
      "loss": 0.6632,
      "step": 2930
    },
    {
      "epoch": 0.2457988462503135,
      "grad_norm": 0.3853206932544708,
      "learning_rate": 2e-05,
      "loss": 0.6684,
      "step": 2940
    },
    {
      "epoch": 0.24663489674776357,
      "grad_norm": 0.37129130959510803,
      "learning_rate": 2e-05,
      "loss": 0.6754,
      "step": 2950
    },
    {
      "epoch": 0.24747094724521362,
      "grad_norm": 0.36231255531311035,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 2960
    },
    {
      "epoch": 0.24830699774266365,
      "grad_norm": 0.36850860714912415,
      "learning_rate": 2e-05,
      "loss": 0.6721,
      "step": 2970
    },
    {
      "epoch": 0.2491430482401137,
      "grad_norm": 0.3462553024291992,
      "learning_rate": 2e-05,
      "loss": 0.6848,
      "step": 2980
    },
    {
      "epoch": 0.24997909873756374,
      "grad_norm": 0.3916807770729065,
      "learning_rate": 2e-05,
      "loss": 0.6859,
      "step": 2990
    },
    {
      "epoch": 0.2508151492350138,
      "grad_norm": 0.37022486329078674,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 3000
    },
    {
      "epoch": 0.25165119973246386,
      "grad_norm": 0.3758009970188141,
      "learning_rate": 2e-05,
      "loss": 0.6982,
      "step": 3010
    },
    {
      "epoch": 0.25248725022991386,
      "grad_norm": 0.3388315439224243,
      "learning_rate": 2e-05,
      "loss": 0.6803,
      "step": 3020
    },
    {
      "epoch": 0.2533233007273639,
      "grad_norm": 0.3719688057899475,
      "learning_rate": 2e-05,
      "loss": 0.701,
      "step": 3030
    },
    {
      "epoch": 0.254159351224814,
      "grad_norm": 0.38503506779670715,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 3040
    },
    {
      "epoch": 0.25499540172226404,
      "grad_norm": 0.408718466758728,
      "learning_rate": 2e-05,
      "loss": 0.6707,
      "step": 3050
    },
    {
      "epoch": 0.2558314522197141,
      "grad_norm": 0.3707769215106964,
      "learning_rate": 2e-05,
      "loss": 0.6685,
      "step": 3060
    },
    {
      "epoch": 0.2566675027171641,
      "grad_norm": 0.35878700017929077,
      "learning_rate": 2e-05,
      "loss": 0.6771,
      "step": 3070
    },
    {
      "epoch": 0.25750355321461416,
      "grad_norm": 0.40199294686317444,
      "learning_rate": 2e-05,
      "loss": 0.7029,
      "step": 3080
    },
    {
      "epoch": 0.2583396037120642,
      "grad_norm": 0.39858362078666687,
      "learning_rate": 2e-05,
      "loss": 0.6961,
      "step": 3090
    },
    {
      "epoch": 0.2591756542095143,
      "grad_norm": 0.4109138548374176,
      "learning_rate": 2e-05,
      "loss": 0.7061,
      "step": 3100
    },
    {
      "epoch": 0.2600117047069643,
      "grad_norm": 0.34975865483283997,
      "learning_rate": 2e-05,
      "loss": 0.6675,
      "step": 3110
    },
    {
      "epoch": 0.26084775520441433,
      "grad_norm": 0.35346266627311707,
      "learning_rate": 2e-05,
      "loss": 0.6397,
      "step": 3120
    },
    {
      "epoch": 0.2616838057018644,
      "grad_norm": 0.37450355291366577,
      "learning_rate": 2e-05,
      "loss": 0.6871,
      "step": 3130
    },
    {
      "epoch": 0.26251985619931445,
      "grad_norm": 0.33483919501304626,
      "learning_rate": 2e-05,
      "loss": 0.6629,
      "step": 3140
    },
    {
      "epoch": 0.2633559066967645,
      "grad_norm": 0.3381412625312805,
      "learning_rate": 2e-05,
      "loss": 0.6839,
      "step": 3150
    },
    {
      "epoch": 0.2641919571942145,
      "grad_norm": 0.34557151794433594,
      "learning_rate": 2e-05,
      "loss": 0.6498,
      "step": 3160
    },
    {
      "epoch": 0.26502800769166457,
      "grad_norm": 0.3589693009853363,
      "learning_rate": 2e-05,
      "loss": 0.6742,
      "step": 3170
    },
    {
      "epoch": 0.2658640581891146,
      "grad_norm": 0.5422878861427307,
      "learning_rate": 2e-05,
      "loss": 0.7182,
      "step": 3180
    },
    {
      "epoch": 0.2667001086865647,
      "grad_norm": 0.39269575476646423,
      "learning_rate": 2e-05,
      "loss": 0.6678,
      "step": 3190
    },
    {
      "epoch": 0.2675361591840147,
      "grad_norm": 0.3926292359828949,
      "learning_rate": 2e-05,
      "loss": 0.6907,
      "step": 3200
    },
    {
      "epoch": 0.26837220968146475,
      "grad_norm": 0.36348679661750793,
      "learning_rate": 2e-05,
      "loss": 0.6649,
      "step": 3210
    },
    {
      "epoch": 0.2692082601789148,
      "grad_norm": 0.36078062653541565,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 3220
    },
    {
      "epoch": 0.27004431067636486,
      "grad_norm": 0.43532776832580566,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 3230
    },
    {
      "epoch": 0.2708803611738149,
      "grad_norm": 0.3712572753429413,
      "learning_rate": 2e-05,
      "loss": 0.6804,
      "step": 3240
    },
    {
      "epoch": 0.2717164116712649,
      "grad_norm": 0.3930246829986572,
      "learning_rate": 2e-05,
      "loss": 0.6852,
      "step": 3250
    },
    {
      "epoch": 0.272552462168715,
      "grad_norm": 0.3693724572658539,
      "learning_rate": 2e-05,
      "loss": 0.681,
      "step": 3260
    },
    {
      "epoch": 0.27338851266616504,
      "grad_norm": 0.3628208637237549,
      "learning_rate": 2e-05,
      "loss": 0.672,
      "step": 3270
    },
    {
      "epoch": 0.2742245631636151,
      "grad_norm": 0.3537019193172455,
      "learning_rate": 2e-05,
      "loss": 0.6722,
      "step": 3280
    },
    {
      "epoch": 0.2750606136610651,
      "grad_norm": 0.40035539865493774,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 3290
    },
    {
      "epoch": 0.27589666415851516,
      "grad_norm": 0.38971859216690063,
      "learning_rate": 2e-05,
      "loss": 0.6566,
      "step": 3300
    },
    {
      "epoch": 0.2767327146559652,
      "grad_norm": 0.6136935949325562,
      "learning_rate": 2e-05,
      "loss": 0.6877,
      "step": 3310
    },
    {
      "epoch": 0.2775687651534153,
      "grad_norm": 0.3618673086166382,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 3320
    },
    {
      "epoch": 0.27840481565086533,
      "grad_norm": 0.3465752899646759,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 3330
    },
    {
      "epoch": 0.27924086614831534,
      "grad_norm": 0.3752216696739197,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 3340
    },
    {
      "epoch": 0.2800769166457654,
      "grad_norm": 0.37954431772232056,
      "learning_rate": 2e-05,
      "loss": 0.689,
      "step": 3350
    },
    {
      "epoch": 0.28091296714321545,
      "grad_norm": 0.39894333481788635,
      "learning_rate": 2e-05,
      "loss": 0.6845,
      "step": 3360
    },
    {
      "epoch": 0.2817490176406655,
      "grad_norm": 0.33840516209602356,
      "learning_rate": 2e-05,
      "loss": 0.6856,
      "step": 3370
    },
    {
      "epoch": 0.28258506813811557,
      "grad_norm": 0.3931964039802551,
      "learning_rate": 2e-05,
      "loss": 0.6862,
      "step": 3380
    },
    {
      "epoch": 0.28342111863556557,
      "grad_norm": 0.40239232778549194,
      "learning_rate": 2e-05,
      "loss": 0.6439,
      "step": 3390
    },
    {
      "epoch": 0.28425716913301563,
      "grad_norm": 0.6196666359901428,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 3400
    },
    {
      "epoch": 0.2850932196304657,
      "grad_norm": 0.4078761041164398,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 3410
    },
    {
      "epoch": 0.28592927012791575,
      "grad_norm": 0.35943377017974854,
      "learning_rate": 2e-05,
      "loss": 0.641,
      "step": 3420
    },
    {
      "epoch": 0.28676532062536575,
      "grad_norm": 0.39068886637687683,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 3430
    },
    {
      "epoch": 0.2876013711228158,
      "grad_norm": 1.7716305255889893,
      "learning_rate": 2e-05,
      "loss": 0.674,
      "step": 3440
    },
    {
      "epoch": 0.28843742162026587,
      "grad_norm": 0.38933104276657104,
      "learning_rate": 2e-05,
      "loss": 0.6772,
      "step": 3450
    },
    {
      "epoch": 0.2892734721177159,
      "grad_norm": 0.3752393424510956,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 3460
    },
    {
      "epoch": 0.290109522615166,
      "grad_norm": 0.35931167006492615,
      "learning_rate": 2e-05,
      "loss": 0.6881,
      "step": 3470
    },
    {
      "epoch": 0.290945573112616,
      "grad_norm": 0.38396894931793213,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 3480
    },
    {
      "epoch": 0.29178162361006604,
      "grad_norm": 0.37433430552482605,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 3490
    },
    {
      "epoch": 0.2926176741075161,
      "grad_norm": 0.3965727984905243,
      "learning_rate": 2e-05,
      "loss": 0.6878,
      "step": 3500
    },
    {
      "epoch": 0.29345372460496616,
      "grad_norm": 0.327019602060318,
      "learning_rate": 2e-05,
      "loss": 0.6601,
      "step": 3510
    },
    {
      "epoch": 0.29428977510241616,
      "grad_norm": 0.37980222702026367,
      "learning_rate": 2e-05,
      "loss": 0.6974,
      "step": 3520
    },
    {
      "epoch": 0.2951258255998662,
      "grad_norm": 0.4086945652961731,
      "learning_rate": 2e-05,
      "loss": 0.6966,
      "step": 3530
    },
    {
      "epoch": 0.2959618760973163,
      "grad_norm": 0.362612783908844,
      "learning_rate": 2e-05,
      "loss": 0.6942,
      "step": 3540
    },
    {
      "epoch": 0.29679792659476634,
      "grad_norm": 0.37178340554237366,
      "learning_rate": 2e-05,
      "loss": 0.6664,
      "step": 3550
    },
    {
      "epoch": 0.2976339770922164,
      "grad_norm": 0.37645795941352844,
      "learning_rate": 2e-05,
      "loss": 0.6761,
      "step": 3560
    },
    {
      "epoch": 0.2984700275896664,
      "grad_norm": 0.347457617521286,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 3570
    },
    {
      "epoch": 0.29930607808711646,
      "grad_norm": 0.3348453640937805,
      "learning_rate": 2e-05,
      "loss": 0.6509,
      "step": 3580
    },
    {
      "epoch": 0.3001421285845665,
      "grad_norm": 0.35900646448135376,
      "learning_rate": 2e-05,
      "loss": 0.6732,
      "step": 3590
    },
    {
      "epoch": 0.3009781790820166,
      "grad_norm": 0.35713493824005127,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 3600
    },
    {
      "epoch": 0.3018142295794666,
      "grad_norm": 0.3648741841316223,
      "learning_rate": 2e-05,
      "loss": 0.6575,
      "step": 3610
    },
    {
      "epoch": 0.30265028007691663,
      "grad_norm": 0.3970864415168762,
      "learning_rate": 2e-05,
      "loss": 0.6909,
      "step": 3620
    },
    {
      "epoch": 0.3034863305743667,
      "grad_norm": 0.38486719131469727,
      "learning_rate": 2e-05,
      "loss": 0.6768,
      "step": 3630
    },
    {
      "epoch": 0.30432238107181675,
      "grad_norm": 0.38870275020599365,
      "learning_rate": 2e-05,
      "loss": 0.6524,
      "step": 3640
    },
    {
      "epoch": 0.3051584315692668,
      "grad_norm": 0.4234541356563568,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 3650
    },
    {
      "epoch": 0.3059944820667168,
      "grad_norm": 0.37357190251350403,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 3660
    },
    {
      "epoch": 0.30683053256416687,
      "grad_norm": 0.39218512177467346,
      "learning_rate": 2e-05,
      "loss": 0.6627,
      "step": 3670
    },
    {
      "epoch": 0.3076665830616169,
      "grad_norm": 0.38100025057792664,
      "learning_rate": 2e-05,
      "loss": 0.6931,
      "step": 3680
    },
    {
      "epoch": 0.308502633559067,
      "grad_norm": 0.3262445330619812,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 3690
    },
    {
      "epoch": 0.309338684056517,
      "grad_norm": 0.36331281065940857,
      "learning_rate": 2e-05,
      "loss": 0.6591,
      "step": 3700
    },
    {
      "epoch": 0.31017473455396705,
      "grad_norm": 0.35214877128601074,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 3710
    },
    {
      "epoch": 0.3110107850514171,
      "grad_norm": 0.38968539237976074,
      "learning_rate": 2e-05,
      "loss": 0.6981,
      "step": 3720
    },
    {
      "epoch": 0.31184683554886716,
      "grad_norm": 0.3726252317428589,
      "learning_rate": 2e-05,
      "loss": 0.6803,
      "step": 3730
    },
    {
      "epoch": 0.3126828860463172,
      "grad_norm": 0.3784383535385132,
      "learning_rate": 2e-05,
      "loss": 0.6678,
      "step": 3740
    },
    {
      "epoch": 0.3135189365437672,
      "grad_norm": 0.3368676006793976,
      "learning_rate": 2e-05,
      "loss": 0.6521,
      "step": 3750
    },
    {
      "epoch": 0.3143549870412173,
      "grad_norm": 0.3492288589477539,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 3760
    },
    {
      "epoch": 0.31519103753866734,
      "grad_norm": 0.35406845808029175,
      "learning_rate": 2e-05,
      "loss": 0.6549,
      "step": 3770
    },
    {
      "epoch": 0.3160270880361174,
      "grad_norm": 0.39807578921318054,
      "learning_rate": 2e-05,
      "loss": 0.6746,
      "step": 3780
    },
    {
      "epoch": 0.3168631385335674,
      "grad_norm": 0.3588227927684784,
      "learning_rate": 2e-05,
      "loss": 0.6718,
      "step": 3790
    },
    {
      "epoch": 0.31769918903101746,
      "grad_norm": 0.36940997838974,
      "learning_rate": 2e-05,
      "loss": 0.6615,
      "step": 3800
    },
    {
      "epoch": 0.3185352395284675,
      "grad_norm": 0.3693230450153351,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 3810
    },
    {
      "epoch": 0.3193712900259176,
      "grad_norm": 0.3649313449859619,
      "learning_rate": 2e-05,
      "loss": 0.6929,
      "step": 3820
    },
    {
      "epoch": 0.32020734052336763,
      "grad_norm": 0.3412187099456787,
      "learning_rate": 2e-05,
      "loss": 0.7131,
      "step": 3830
    },
    {
      "epoch": 0.32104339102081764,
      "grad_norm": 0.4318503737449646,
      "learning_rate": 2e-05,
      "loss": 0.7043,
      "step": 3840
    },
    {
      "epoch": 0.3218794415182677,
      "grad_norm": 0.35563552379608154,
      "learning_rate": 2e-05,
      "loss": 0.6621,
      "step": 3850
    },
    {
      "epoch": 0.32271549201571775,
      "grad_norm": 0.3535933494567871,
      "learning_rate": 2e-05,
      "loss": 0.6662,
      "step": 3860
    },
    {
      "epoch": 0.3235515425131678,
      "grad_norm": 0.3648977279663086,
      "learning_rate": 2e-05,
      "loss": 0.6853,
      "step": 3870
    },
    {
      "epoch": 0.3243875930106178,
      "grad_norm": 0.3543550372123718,
      "learning_rate": 2e-05,
      "loss": 0.6848,
      "step": 3880
    },
    {
      "epoch": 0.3252236435080679,
      "grad_norm": 0.38484376668930054,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 3890
    },
    {
      "epoch": 0.32605969400551793,
      "grad_norm": 0.37139758467674255,
      "learning_rate": 2e-05,
      "loss": 0.6962,
      "step": 3900
    },
    {
      "epoch": 0.326895744502968,
      "grad_norm": 0.38717278838157654,
      "learning_rate": 2e-05,
      "loss": 0.7069,
      "step": 3910
    },
    {
      "epoch": 0.32773179500041805,
      "grad_norm": 0.3783646523952484,
      "learning_rate": 2e-05,
      "loss": 0.6738,
      "step": 3920
    },
    {
      "epoch": 0.32856784549786805,
      "grad_norm": 0.3423800766468048,
      "learning_rate": 2e-05,
      "loss": 0.6651,
      "step": 3930
    },
    {
      "epoch": 0.3294038959953181,
      "grad_norm": 0.3589499294757843,
      "learning_rate": 2e-05,
      "loss": 0.6574,
      "step": 3940
    },
    {
      "epoch": 0.33023994649276817,
      "grad_norm": 0.40661120414733887,
      "learning_rate": 2e-05,
      "loss": 0.6873,
      "step": 3950
    },
    {
      "epoch": 0.3310759969902182,
      "grad_norm": 0.3802325129508972,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 3960
    },
    {
      "epoch": 0.3319120474876683,
      "grad_norm": 0.3785724639892578,
      "learning_rate": 2e-05,
      "loss": 0.6507,
      "step": 3970
    },
    {
      "epoch": 0.3327480979851183,
      "grad_norm": 0.36652204394340515,
      "learning_rate": 2e-05,
      "loss": 0.7036,
      "step": 3980
    },
    {
      "epoch": 0.33358414848256834,
      "grad_norm": 0.33603420853614807,
      "learning_rate": 2e-05,
      "loss": 0.7145,
      "step": 3990
    },
    {
      "epoch": 0.3344201989800184,
      "grad_norm": 0.3536396026611328,
      "learning_rate": 2e-05,
      "loss": 0.6736,
      "step": 4000
    },
    {
      "epoch": 0.33525624947746846,
      "grad_norm": 0.4215162396430969,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 4010
    },
    {
      "epoch": 0.33609229997491846,
      "grad_norm": 0.36844271421432495,
      "learning_rate": 2e-05,
      "loss": 0.6881,
      "step": 4020
    },
    {
      "epoch": 0.3369283504723685,
      "grad_norm": 0.4516617953777313,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 4030
    },
    {
      "epoch": 0.3377644009698186,
      "grad_norm": 0.3380790054798126,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 4040
    },
    {
      "epoch": 0.33860045146726864,
      "grad_norm": 0.3842700719833374,
      "learning_rate": 2e-05,
      "loss": 0.6276,
      "step": 4050
    },
    {
      "epoch": 0.3394365019647187,
      "grad_norm": 0.34872040152549744,
      "learning_rate": 2e-05,
      "loss": 0.653,
      "step": 4060
    },
    {
      "epoch": 0.3402725524621687,
      "grad_norm": 0.3866327702999115,
      "learning_rate": 2e-05,
      "loss": 0.6716,
      "step": 4070
    },
    {
      "epoch": 0.34110860295961876,
      "grad_norm": 0.3686073124408722,
      "learning_rate": 2e-05,
      "loss": 0.6701,
      "step": 4080
    },
    {
      "epoch": 0.3419446534570688,
      "grad_norm": 0.42053115367889404,
      "learning_rate": 2e-05,
      "loss": 0.6859,
      "step": 4090
    },
    {
      "epoch": 0.3427807039545189,
      "grad_norm": 0.3751194477081299,
      "learning_rate": 2e-05,
      "loss": 0.658,
      "step": 4100
    },
    {
      "epoch": 0.3436167544519689,
      "grad_norm": 0.375248521566391,
      "learning_rate": 2e-05,
      "loss": 0.6791,
      "step": 4110
    },
    {
      "epoch": 0.34445280494941893,
      "grad_norm": 0.32522687315940857,
      "learning_rate": 2e-05,
      "loss": 0.6833,
      "step": 4120
    },
    {
      "epoch": 0.345288855446869,
      "grad_norm": 0.34771233797073364,
      "learning_rate": 2e-05,
      "loss": 0.6234,
      "step": 4130
    },
    {
      "epoch": 0.34612490594431905,
      "grad_norm": 0.3351838290691376,
      "learning_rate": 2e-05,
      "loss": 0.6574,
      "step": 4140
    },
    {
      "epoch": 0.3469609564417691,
      "grad_norm": 0.3469686210155487,
      "learning_rate": 2e-05,
      "loss": 0.6474,
      "step": 4150
    },
    {
      "epoch": 0.3477970069392191,
      "grad_norm": 0.3836895823478699,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 4160
    },
    {
      "epoch": 0.34863305743666917,
      "grad_norm": 0.3830288350582123,
      "learning_rate": 2e-05,
      "loss": 0.6745,
      "step": 4170
    },
    {
      "epoch": 0.3494691079341192,
      "grad_norm": 0.37637174129486084,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 4180
    },
    {
      "epoch": 0.3503051584315693,
      "grad_norm": 0.4341719150543213,
      "learning_rate": 2e-05,
      "loss": 0.6545,
      "step": 4190
    },
    {
      "epoch": 0.3511412089290193,
      "grad_norm": 0.42583167552948,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 4200
    },
    {
      "epoch": 0.35197725942646935,
      "grad_norm": 0.3464331030845642,
      "learning_rate": 2e-05,
      "loss": 0.659,
      "step": 4210
    },
    {
      "epoch": 0.3528133099239194,
      "grad_norm": 0.3504962921142578,
      "learning_rate": 2e-05,
      "loss": 0.6594,
      "step": 4220
    },
    {
      "epoch": 0.35364936042136946,
      "grad_norm": 0.3957189619541168,
      "learning_rate": 2e-05,
      "loss": 0.6585,
      "step": 4230
    },
    {
      "epoch": 0.3544854109188195,
      "grad_norm": 0.35942843556404114,
      "learning_rate": 2e-05,
      "loss": 0.6663,
      "step": 4240
    },
    {
      "epoch": 0.3553214614162695,
      "grad_norm": 0.36357617378234863,
      "learning_rate": 2e-05,
      "loss": 0.6762,
      "step": 4250
    },
    {
      "epoch": 0.3561575119137196,
      "grad_norm": 0.3632410764694214,
      "learning_rate": 2e-05,
      "loss": 0.6707,
      "step": 4260
    },
    {
      "epoch": 0.35699356241116964,
      "grad_norm": 0.40080034732818604,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 4270
    },
    {
      "epoch": 0.3578296129086197,
      "grad_norm": 0.37345850467681885,
      "learning_rate": 2e-05,
      "loss": 0.6821,
      "step": 4280
    },
    {
      "epoch": 0.3586656634060697,
      "grad_norm": 0.3551454246044159,
      "learning_rate": 2e-05,
      "loss": 0.6923,
      "step": 4290
    },
    {
      "epoch": 0.35950171390351976,
      "grad_norm": 0.41821813583374023,
      "learning_rate": 2e-05,
      "loss": 0.6756,
      "step": 4300
    },
    {
      "epoch": 0.3603377644009698,
      "grad_norm": 0.35665106773376465,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 4310
    },
    {
      "epoch": 0.3611738148984199,
      "grad_norm": 0.32754582166671753,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 4320
    },
    {
      "epoch": 0.36200986539586993,
      "grad_norm": 0.36815065145492554,
      "learning_rate": 2e-05,
      "loss": 0.6808,
      "step": 4330
    },
    {
      "epoch": 0.36284591589331994,
      "grad_norm": 0.41788262128829956,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 4340
    },
    {
      "epoch": 0.36368196639077,
      "grad_norm": 0.3677462041378021,
      "learning_rate": 2e-05,
      "loss": 0.6819,
      "step": 4350
    },
    {
      "epoch": 0.36451801688822005,
      "grad_norm": 0.3276795446872711,
      "learning_rate": 2e-05,
      "loss": 0.6403,
      "step": 4360
    },
    {
      "epoch": 0.3653540673856701,
      "grad_norm": 0.3425849378108978,
      "learning_rate": 2e-05,
      "loss": 0.6592,
      "step": 4370
    },
    {
      "epoch": 0.3661901178831201,
      "grad_norm": 0.34393274784088135,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 4380
    },
    {
      "epoch": 0.3670261683805702,
      "grad_norm": 0.36734405159950256,
      "learning_rate": 2e-05,
      "loss": 0.6722,
      "step": 4390
    },
    {
      "epoch": 0.36786221887802023,
      "grad_norm": 0.3715624213218689,
      "learning_rate": 2e-05,
      "loss": 0.6579,
      "step": 4400
    },
    {
      "epoch": 0.3686982693754703,
      "grad_norm": 0.4003128111362457,
      "learning_rate": 2e-05,
      "loss": 0.7004,
      "step": 4410
    },
    {
      "epoch": 0.36953431987292035,
      "grad_norm": 0.3671753406524658,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 4420
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 0.3676854372024536,
      "learning_rate": 2e-05,
      "loss": 0.6676,
      "step": 4430
    },
    {
      "epoch": 0.3712064208678204,
      "grad_norm": 0.34763482213020325,
      "learning_rate": 2e-05,
      "loss": 0.6921,
      "step": 4440
    },
    {
      "epoch": 0.37204247136527047,
      "grad_norm": 0.38204512000083923,
      "learning_rate": 2e-05,
      "loss": 0.6733,
      "step": 4450
    },
    {
      "epoch": 0.3728785218627205,
      "grad_norm": 0.36626923084259033,
      "learning_rate": 2e-05,
      "loss": 0.6402,
      "step": 4460
    },
    {
      "epoch": 0.3737145723601705,
      "grad_norm": 0.3997361361980438,
      "learning_rate": 2e-05,
      "loss": 0.6596,
      "step": 4470
    },
    {
      "epoch": 0.3745506228576206,
      "grad_norm": 0.3861956298351288,
      "learning_rate": 2e-05,
      "loss": 0.6494,
      "step": 4480
    },
    {
      "epoch": 0.37538667335507064,
      "grad_norm": 0.3729952871799469,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 4490
    },
    {
      "epoch": 0.3762227238525207,
      "grad_norm": 0.37843218445777893,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 4500
    },
    {
      "epoch": 0.37705877434997076,
      "grad_norm": 0.3461994230747223,
      "learning_rate": 2e-05,
      "loss": 0.6882,
      "step": 4510
    },
    {
      "epoch": 0.37789482484742076,
      "grad_norm": 0.37727099657058716,
      "learning_rate": 2e-05,
      "loss": 0.6639,
      "step": 4520
    },
    {
      "epoch": 0.3787308753448708,
      "grad_norm": 0.3723204433917999,
      "learning_rate": 2e-05,
      "loss": 0.6924,
      "step": 4530
    },
    {
      "epoch": 0.3795669258423209,
      "grad_norm": 0.34893015027046204,
      "learning_rate": 2e-05,
      "loss": 0.6693,
      "step": 4540
    },
    {
      "epoch": 0.38040297633977094,
      "grad_norm": 0.37652260065078735,
      "learning_rate": 2e-05,
      "loss": 0.6659,
      "step": 4550
    },
    {
      "epoch": 0.381239026837221,
      "grad_norm": 0.3943141996860504,
      "learning_rate": 2e-05,
      "loss": 0.6502,
      "step": 4560
    },
    {
      "epoch": 0.382075077334671,
      "grad_norm": 0.39593082666397095,
      "learning_rate": 2e-05,
      "loss": 0.6285,
      "step": 4570
    },
    {
      "epoch": 0.38291112783212106,
      "grad_norm": 0.3677999675273895,
      "learning_rate": 2e-05,
      "loss": 0.6215,
      "step": 4580
    },
    {
      "epoch": 0.3837471783295711,
      "grad_norm": 0.44067198038101196,
      "learning_rate": 2e-05,
      "loss": 0.659,
      "step": 4590
    },
    {
      "epoch": 0.3845832288270212,
      "grad_norm": 0.3503727316856384,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 4600
    },
    {
      "epoch": 0.3854192793244712,
      "grad_norm": 0.3658464252948761,
      "learning_rate": 2e-05,
      "loss": 0.6668,
      "step": 4610
    },
    {
      "epoch": 0.38625532982192123,
      "grad_norm": 0.379948228597641,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 4620
    },
    {
      "epoch": 0.3870913803193713,
      "grad_norm": 0.3512934446334839,
      "learning_rate": 2e-05,
      "loss": 0.6856,
      "step": 4630
    },
    {
      "epoch": 0.38792743081682135,
      "grad_norm": 0.3343440592288971,
      "learning_rate": 2e-05,
      "loss": 0.6485,
      "step": 4640
    },
    {
      "epoch": 0.3887634813142714,
      "grad_norm": 0.3770824372768402,
      "learning_rate": 2e-05,
      "loss": 0.6745,
      "step": 4650
    },
    {
      "epoch": 0.3895995318117214,
      "grad_norm": 0.3582046627998352,
      "learning_rate": 2e-05,
      "loss": 0.6691,
      "step": 4660
    },
    {
      "epoch": 0.39043558230917147,
      "grad_norm": 0.3880390226840973,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 4670
    },
    {
      "epoch": 0.3912716328066215,
      "grad_norm": 0.35768410563468933,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 4680
    },
    {
      "epoch": 0.3921076833040716,
      "grad_norm": 0.3596102297306061,
      "learning_rate": 2e-05,
      "loss": 0.6683,
      "step": 4690
    },
    {
      "epoch": 0.3929437338015216,
      "grad_norm": 0.3419645428657532,
      "learning_rate": 2e-05,
      "loss": 0.6551,
      "step": 4700
    },
    {
      "epoch": 0.39377978429897165,
      "grad_norm": 0.3907841145992279,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 4710
    },
    {
      "epoch": 0.3946158347964217,
      "grad_norm": 0.35143643617630005,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 4720
    },
    {
      "epoch": 0.39545188529387176,
      "grad_norm": 0.3897535800933838,
      "learning_rate": 2e-05,
      "loss": 0.6844,
      "step": 4730
    },
    {
      "epoch": 0.3962879357913218,
      "grad_norm": 0.3715405762195587,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 4740
    },
    {
      "epoch": 0.3971239862887718,
      "grad_norm": 0.3395850360393524,
      "learning_rate": 2e-05,
      "loss": 0.6814,
      "step": 4750
    },
    {
      "epoch": 0.3979600367862219,
      "grad_norm": 0.3568040728569031,
      "learning_rate": 2e-05,
      "loss": 0.666,
      "step": 4760
    },
    {
      "epoch": 0.39879608728367194,
      "grad_norm": 0.6625901460647583,
      "learning_rate": 2e-05,
      "loss": 0.6961,
      "step": 4770
    },
    {
      "epoch": 0.399632137781122,
      "grad_norm": 0.41476544737815857,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 4780
    },
    {
      "epoch": 0.400468188278572,
      "grad_norm": 0.3590489327907562,
      "learning_rate": 2e-05,
      "loss": 0.613,
      "step": 4790
    },
    {
      "epoch": 0.40130423877602206,
      "grad_norm": 0.37327614426612854,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 4800
    },
    {
      "epoch": 0.4021402892734721,
      "grad_norm": 0.3871711194515228,
      "learning_rate": 2e-05,
      "loss": 0.6481,
      "step": 4810
    },
    {
      "epoch": 0.4029763397709222,
      "grad_norm": 0.37193727493286133,
      "learning_rate": 2e-05,
      "loss": 0.6478,
      "step": 4820
    },
    {
      "epoch": 0.40381239026837223,
      "grad_norm": 0.3502122163772583,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 4830
    },
    {
      "epoch": 0.40464844076582224,
      "grad_norm": 0.39542460441589355,
      "learning_rate": 2e-05,
      "loss": 0.6797,
      "step": 4840
    },
    {
      "epoch": 0.4054844912632723,
      "grad_norm": 0.3589038550853729,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 4850
    },
    {
      "epoch": 0.40632054176072235,
      "grad_norm": 0.38512277603149414,
      "learning_rate": 2e-05,
      "loss": 0.6179,
      "step": 4860
    },
    {
      "epoch": 0.4071565922581724,
      "grad_norm": 0.4002944231033325,
      "learning_rate": 2e-05,
      "loss": 0.6633,
      "step": 4870
    },
    {
      "epoch": 0.4079926427556224,
      "grad_norm": 0.36734187602996826,
      "learning_rate": 2e-05,
      "loss": 0.6271,
      "step": 4880
    },
    {
      "epoch": 0.4088286932530725,
      "grad_norm": 0.3817862868309021,
      "learning_rate": 2e-05,
      "loss": 0.6877,
      "step": 4890
    },
    {
      "epoch": 0.40966474375052253,
      "grad_norm": 0.3771727979183197,
      "learning_rate": 2e-05,
      "loss": 0.6649,
      "step": 4900
    },
    {
      "epoch": 0.4105007942479726,
      "grad_norm": 0.3870513439178467,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 4910
    },
    {
      "epoch": 0.41133684474542265,
      "grad_norm": 0.34439635276794434,
      "learning_rate": 2e-05,
      "loss": 0.6159,
      "step": 4920
    },
    {
      "epoch": 0.41217289524287265,
      "grad_norm": 0.3439784049987793,
      "learning_rate": 2e-05,
      "loss": 0.6122,
      "step": 4930
    },
    {
      "epoch": 0.4130089457403227,
      "grad_norm": 0.32294797897338867,
      "learning_rate": 2e-05,
      "loss": 0.6493,
      "step": 4940
    },
    {
      "epoch": 0.41384499623777277,
      "grad_norm": 0.3561195433139801,
      "learning_rate": 2e-05,
      "loss": 0.6806,
      "step": 4950
    },
    {
      "epoch": 0.4146810467352228,
      "grad_norm": 0.37559956312179565,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 4960
    },
    {
      "epoch": 0.4155170972326728,
      "grad_norm": 0.37067362666130066,
      "learning_rate": 2e-05,
      "loss": 0.6345,
      "step": 4970
    },
    {
      "epoch": 0.4163531477301229,
      "grad_norm": 0.37078824639320374,
      "learning_rate": 2e-05,
      "loss": 0.6718,
      "step": 4980
    },
    {
      "epoch": 0.41718919822757294,
      "grad_norm": 0.3785903751850128,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 4990
    },
    {
      "epoch": 0.418025248725023,
      "grad_norm": 0.3805888295173645,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 5000
    },
    {
      "epoch": 0.41886129922247306,
      "grad_norm": 0.3531053066253662,
      "learning_rate": 2e-05,
      "loss": 0.6455,
      "step": 5010
    },
    {
      "epoch": 0.41969734971992306,
      "grad_norm": 0.36093053221702576,
      "learning_rate": 2e-05,
      "loss": 0.6551,
      "step": 5020
    },
    {
      "epoch": 0.4205334002173731,
      "grad_norm": 0.35770487785339355,
      "learning_rate": 2e-05,
      "loss": 0.64,
      "step": 5030
    },
    {
      "epoch": 0.4213694507148232,
      "grad_norm": 0.39595308899879456,
      "learning_rate": 2e-05,
      "loss": 0.6729,
      "step": 5040
    },
    {
      "epoch": 0.42220550121227324,
      "grad_norm": 0.3384452760219574,
      "learning_rate": 2e-05,
      "loss": 0.6496,
      "step": 5050
    },
    {
      "epoch": 0.42304155170972324,
      "grad_norm": 0.34772634506225586,
      "learning_rate": 2e-05,
      "loss": 0.6019,
      "step": 5060
    },
    {
      "epoch": 0.4238776022071733,
      "grad_norm": 0.34704867005348206,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 5070
    },
    {
      "epoch": 0.42471365270462336,
      "grad_norm": 0.3721390664577484,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 5080
    },
    {
      "epoch": 0.4255497032020734,
      "grad_norm": 0.37306642532348633,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 5090
    },
    {
      "epoch": 0.4263857536995235,
      "grad_norm": 0.33520811796188354,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 5100
    },
    {
      "epoch": 0.4272218041969735,
      "grad_norm": 0.36285078525543213,
      "learning_rate": 2e-05,
      "loss": 0.6699,
      "step": 5110
    },
    {
      "epoch": 0.42805785469442353,
      "grad_norm": 0.37109819054603577,
      "learning_rate": 2e-05,
      "loss": 0.662,
      "step": 5120
    },
    {
      "epoch": 0.4288939051918736,
      "grad_norm": 0.3391190767288208,
      "learning_rate": 2e-05,
      "loss": 0.6532,
      "step": 5130
    },
    {
      "epoch": 0.42972995568932365,
      "grad_norm": 0.3734348714351654,
      "learning_rate": 2e-05,
      "loss": 0.6513,
      "step": 5140
    },
    {
      "epoch": 0.4305660061867737,
      "grad_norm": 0.40317806601524353,
      "learning_rate": 2e-05,
      "loss": 0.6844,
      "step": 5150
    },
    {
      "epoch": 0.4314020566842237,
      "grad_norm": 0.37194934487342834,
      "learning_rate": 2e-05,
      "loss": 0.6671,
      "step": 5160
    },
    {
      "epoch": 0.43223810718167377,
      "grad_norm": 0.36219897866249084,
      "learning_rate": 2e-05,
      "loss": 0.6609,
      "step": 5170
    },
    {
      "epoch": 0.43307415767912383,
      "grad_norm": 0.3716292083263397,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 5180
    },
    {
      "epoch": 0.4339102081765739,
      "grad_norm": 0.3661538064479828,
      "learning_rate": 2e-05,
      "loss": 0.651,
      "step": 5190
    },
    {
      "epoch": 0.4347462586740239,
      "grad_norm": 0.38041263818740845,
      "learning_rate": 2e-05,
      "loss": 0.658,
      "step": 5200
    },
    {
      "epoch": 0.43558230917147395,
      "grad_norm": 0.36274105310440063,
      "learning_rate": 2e-05,
      "loss": 0.6768,
      "step": 5210
    },
    {
      "epoch": 0.436418359668924,
      "grad_norm": 0.44696131348609924,
      "learning_rate": 2e-05,
      "loss": 0.6671,
      "step": 5220
    },
    {
      "epoch": 0.43725441016637406,
      "grad_norm": 0.3828553855419159,
      "learning_rate": 2e-05,
      "loss": 0.6605,
      "step": 5230
    },
    {
      "epoch": 0.4380904606638241,
      "grad_norm": 0.39655470848083496,
      "learning_rate": 2e-05,
      "loss": 0.6615,
      "step": 5240
    },
    {
      "epoch": 0.4389265111612741,
      "grad_norm": 0.367038756608963,
      "learning_rate": 2e-05,
      "loss": 0.6446,
      "step": 5250
    },
    {
      "epoch": 0.4397625616587242,
      "grad_norm": 0.35305437445640564,
      "learning_rate": 2e-05,
      "loss": 0.6462,
      "step": 5260
    },
    {
      "epoch": 0.44059861215617424,
      "grad_norm": 0.3451675474643707,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 5270
    },
    {
      "epoch": 0.4414346626536243,
      "grad_norm": 0.3524886667728424,
      "learning_rate": 2e-05,
      "loss": 0.6244,
      "step": 5280
    },
    {
      "epoch": 0.4422707131510743,
      "grad_norm": 0.35340991616249084,
      "learning_rate": 2e-05,
      "loss": 0.68,
      "step": 5290
    },
    {
      "epoch": 0.44310676364852436,
      "grad_norm": 0.3325699269771576,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 5300
    },
    {
      "epoch": 0.4439428141459744,
      "grad_norm": 0.351138710975647,
      "learning_rate": 2e-05,
      "loss": 0.6536,
      "step": 5310
    },
    {
      "epoch": 0.4447788646434245,
      "grad_norm": 0.33755823969841003,
      "learning_rate": 2e-05,
      "loss": 0.6472,
      "step": 5320
    },
    {
      "epoch": 0.44561491514087453,
      "grad_norm": 0.37768250703811646,
      "learning_rate": 2e-05,
      "loss": 0.6467,
      "step": 5330
    },
    {
      "epoch": 0.44645096563832454,
      "grad_norm": 0.35582277178764343,
      "learning_rate": 2e-05,
      "loss": 0.7014,
      "step": 5340
    },
    {
      "epoch": 0.4472870161357746,
      "grad_norm": 0.3603978455066681,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 5350
    },
    {
      "epoch": 0.44812306663322465,
      "grad_norm": 0.3806731402873993,
      "learning_rate": 2e-05,
      "loss": 0.6631,
      "step": 5360
    },
    {
      "epoch": 0.4489591171306747,
      "grad_norm": 0.3463365137577057,
      "learning_rate": 2e-05,
      "loss": 0.6663,
      "step": 5370
    },
    {
      "epoch": 0.4497951676281247,
      "grad_norm": 0.3559459447860718,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 5380
    },
    {
      "epoch": 0.4506312181255748,
      "grad_norm": 0.48492664098739624,
      "learning_rate": 2e-05,
      "loss": 0.664,
      "step": 5390
    },
    {
      "epoch": 0.45146726862302483,
      "grad_norm": 0.37034884095191956,
      "learning_rate": 2e-05,
      "loss": 0.6347,
      "step": 5400
    },
    {
      "epoch": 0.4523033191204749,
      "grad_norm": 0.38452285528182983,
      "learning_rate": 2e-05,
      "loss": 0.6592,
      "step": 5410
    },
    {
      "epoch": 0.45313936961792495,
      "grad_norm": 0.38597428798675537,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 5420
    },
    {
      "epoch": 0.45397542011537495,
      "grad_norm": 0.3872252404689789,
      "learning_rate": 2e-05,
      "loss": 0.6507,
      "step": 5430
    },
    {
      "epoch": 0.454811470612825,
      "grad_norm": 0.3807690143585205,
      "learning_rate": 2e-05,
      "loss": 0.6907,
      "step": 5440
    },
    {
      "epoch": 0.45564752111027507,
      "grad_norm": 0.37120118737220764,
      "learning_rate": 2e-05,
      "loss": 0.6974,
      "step": 5450
    },
    {
      "epoch": 0.4564835716077251,
      "grad_norm": 0.3623567223548889,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 5460
    },
    {
      "epoch": 0.4573196221051751,
      "grad_norm": 0.4010092616081238,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 5470
    },
    {
      "epoch": 0.4581556726026252,
      "grad_norm": 0.39036762714385986,
      "learning_rate": 2e-05,
      "loss": 0.6749,
      "step": 5480
    },
    {
      "epoch": 0.45899172310007524,
      "grad_norm": 0.3458063304424286,
      "learning_rate": 2e-05,
      "loss": 0.6487,
      "step": 5490
    },
    {
      "epoch": 0.4598277735975253,
      "grad_norm": 0.38566914200782776,
      "learning_rate": 2e-05,
      "loss": 0.6842,
      "step": 5500
    },
    {
      "epoch": 0.46066382409497536,
      "grad_norm": 0.3714940845966339,
      "learning_rate": 2e-05,
      "loss": 0.6476,
      "step": 5510
    },
    {
      "epoch": 0.46149987459242536,
      "grad_norm": 0.3972676396369934,
      "learning_rate": 2e-05,
      "loss": 0.648,
      "step": 5520
    },
    {
      "epoch": 0.4623359250898754,
      "grad_norm": 0.3655273914337158,
      "learning_rate": 2e-05,
      "loss": 0.6492,
      "step": 5530
    },
    {
      "epoch": 0.4631719755873255,
      "grad_norm": 0.47826796770095825,
      "learning_rate": 2e-05,
      "loss": 0.6475,
      "step": 5540
    },
    {
      "epoch": 0.46400802608477554,
      "grad_norm": 0.35415518283843994,
      "learning_rate": 2e-05,
      "loss": 0.6411,
      "step": 5550
    },
    {
      "epoch": 0.46484407658222554,
      "grad_norm": 0.35670533776283264,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 5560
    },
    {
      "epoch": 0.4656801270796756,
      "grad_norm": 0.35704949498176575,
      "learning_rate": 2e-05,
      "loss": 0.5959,
      "step": 5570
    },
    {
      "epoch": 0.46651617757712566,
      "grad_norm": 0.3651845157146454,
      "learning_rate": 2e-05,
      "loss": 0.6493,
      "step": 5580
    },
    {
      "epoch": 0.4673522280745757,
      "grad_norm": 0.3384692370891571,
      "learning_rate": 2e-05,
      "loss": 0.6456,
      "step": 5590
    },
    {
      "epoch": 0.4681882785720258,
      "grad_norm": 0.3535986840724945,
      "learning_rate": 2e-05,
      "loss": 0.6336,
      "step": 5600
    },
    {
      "epoch": 0.4690243290694758,
      "grad_norm": 0.43083256483078003,
      "learning_rate": 2e-05,
      "loss": 0.6213,
      "step": 5610
    },
    {
      "epoch": 0.46986037956692583,
      "grad_norm": 0.35124170780181885,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 5620
    },
    {
      "epoch": 0.4706964300643759,
      "grad_norm": 0.3564167320728302,
      "learning_rate": 2e-05,
      "loss": 0.6551,
      "step": 5630
    },
    {
      "epoch": 0.47153248056182595,
      "grad_norm": 0.3742994964122772,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 5640
    },
    {
      "epoch": 0.47236853105927595,
      "grad_norm": 0.35218045115470886,
      "learning_rate": 2e-05,
      "loss": 0.6793,
      "step": 5650
    },
    {
      "epoch": 0.473204581556726,
      "grad_norm": 0.35776063799858093,
      "learning_rate": 2e-05,
      "loss": 0.6247,
      "step": 5660
    },
    {
      "epoch": 0.47404063205417607,
      "grad_norm": 0.38124266266822815,
      "learning_rate": 2e-05,
      "loss": 0.6549,
      "step": 5670
    },
    {
      "epoch": 0.47487668255162613,
      "grad_norm": 0.3873339593410492,
      "learning_rate": 2e-05,
      "loss": 0.6863,
      "step": 5680
    },
    {
      "epoch": 0.4757127330490762,
      "grad_norm": 0.3914729356765747,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 5690
    },
    {
      "epoch": 0.4765487835465262,
      "grad_norm": 0.3846113681793213,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 5700
    },
    {
      "epoch": 0.47738483404397625,
      "grad_norm": 0.34895122051239014,
      "learning_rate": 2e-05,
      "loss": 0.6256,
      "step": 5710
    },
    {
      "epoch": 0.4782208845414263,
      "grad_norm": 0.3587910532951355,
      "learning_rate": 2e-05,
      "loss": 0.6425,
      "step": 5720
    },
    {
      "epoch": 0.47905693503887636,
      "grad_norm": 0.4189320504665375,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 5730
    },
    {
      "epoch": 0.4798929855363264,
      "grad_norm": 0.4043200612068176,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 5740
    },
    {
      "epoch": 0.4807290360337764,
      "grad_norm": 0.3713650703430176,
      "learning_rate": 2e-05,
      "loss": 0.6655,
      "step": 5750
    },
    {
      "epoch": 0.4815650865312265,
      "grad_norm": 0.35589706897735596,
      "learning_rate": 2e-05,
      "loss": 0.6614,
      "step": 5760
    },
    {
      "epoch": 0.48240113702867654,
      "grad_norm": 0.36650487780570984,
      "learning_rate": 2e-05,
      "loss": 0.6799,
      "step": 5770
    },
    {
      "epoch": 0.4832371875261266,
      "grad_norm": 0.362053245306015,
      "learning_rate": 2e-05,
      "loss": 0.6925,
      "step": 5780
    },
    {
      "epoch": 0.4840732380235766,
      "grad_norm": 0.46135276556015015,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 5790
    },
    {
      "epoch": 0.48490928852102666,
      "grad_norm": 0.5090813040733337,
      "learning_rate": 2e-05,
      "loss": 0.6348,
      "step": 5800
    },
    {
      "epoch": 0.4857453390184767,
      "grad_norm": 0.3569628596305847,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 5810
    },
    {
      "epoch": 0.4865813895159268,
      "grad_norm": 0.34732237458229065,
      "learning_rate": 2e-05,
      "loss": 0.6279,
      "step": 5820
    },
    {
      "epoch": 0.48741744001337683,
      "grad_norm": 0.36834198236465454,
      "learning_rate": 2e-05,
      "loss": 0.6679,
      "step": 5830
    },
    {
      "epoch": 0.48825349051082684,
      "grad_norm": 0.37787771224975586,
      "learning_rate": 2e-05,
      "loss": 0.6599,
      "step": 5840
    },
    {
      "epoch": 0.4890895410082769,
      "grad_norm": 0.37708190083503723,
      "learning_rate": 2e-05,
      "loss": 0.6639,
      "step": 5850
    },
    {
      "epoch": 0.48992559150572695,
      "grad_norm": 0.34140101075172424,
      "learning_rate": 2e-05,
      "loss": 0.6578,
      "step": 5860
    },
    {
      "epoch": 0.490761642003177,
      "grad_norm": 0.36427590250968933,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 5870
    },
    {
      "epoch": 0.491597692500627,
      "grad_norm": 0.3398895859718323,
      "learning_rate": 2e-05,
      "loss": 0.618,
      "step": 5880
    },
    {
      "epoch": 0.4924337429980771,
      "grad_norm": 0.3468286991119385,
      "learning_rate": 2e-05,
      "loss": 0.6538,
      "step": 5890
    },
    {
      "epoch": 0.49326979349552713,
      "grad_norm": 0.38035982847213745,
      "learning_rate": 2e-05,
      "loss": 0.6661,
      "step": 5900
    },
    {
      "epoch": 0.4941058439929772,
      "grad_norm": 0.3798688054084778,
      "learning_rate": 2e-05,
      "loss": 0.654,
      "step": 5910
    },
    {
      "epoch": 0.49494189449042725,
      "grad_norm": 0.36911895871162415,
      "learning_rate": 2e-05,
      "loss": 0.6472,
      "step": 5920
    },
    {
      "epoch": 0.49577794498787725,
      "grad_norm": 0.3868490159511566,
      "learning_rate": 2e-05,
      "loss": 0.6648,
      "step": 5930
    },
    {
      "epoch": 0.4966139954853273,
      "grad_norm": 0.35158196091651917,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 5940
    },
    {
      "epoch": 0.49745004598277737,
      "grad_norm": 0.3511255979537964,
      "learning_rate": 2e-05,
      "loss": 0.6578,
      "step": 5950
    },
    {
      "epoch": 0.4982860964802274,
      "grad_norm": 0.32360202074050903,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 5960
    },
    {
      "epoch": 0.49912214697767743,
      "grad_norm": 0.37470629811286926,
      "learning_rate": 2e-05,
      "loss": 0.6621,
      "step": 5970
    },
    {
      "epoch": 0.4999581974751275,
      "grad_norm": 0.3609738349914551,
      "learning_rate": 2e-05,
      "loss": 0.6999,
      "step": 5980
    },
    {
      "epoch": 0.5007942479725775,
      "grad_norm": 0.3503910005092621,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 5990
    },
    {
      "epoch": 0.5016302984700276,
      "grad_norm": 0.33237144351005554,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 6000
    },
    {
      "epoch": 0.5024663489674777,
      "grad_norm": 0.35927101969718933,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 6010
    },
    {
      "epoch": 0.5033023994649277,
      "grad_norm": 0.3422349989414215,
      "learning_rate": 2e-05,
      "loss": 0.6601,
      "step": 6020
    },
    {
      "epoch": 0.5041384499623778,
      "grad_norm": 0.41844573616981506,
      "learning_rate": 2e-05,
      "loss": 0.6174,
      "step": 6030
    },
    {
      "epoch": 0.5049745004598277,
      "grad_norm": 0.3745769262313843,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 6040
    },
    {
      "epoch": 0.5058105509572778,
      "grad_norm": 0.4188084900379181,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 6050
    },
    {
      "epoch": 0.5066466014547278,
      "grad_norm": 0.3455961048603058,
      "learning_rate": 2e-05,
      "loss": 0.6555,
      "step": 6060
    },
    {
      "epoch": 0.5074826519521779,
      "grad_norm": 0.38813912868499756,
      "learning_rate": 2e-05,
      "loss": 0.6445,
      "step": 6070
    },
    {
      "epoch": 0.508318702449628,
      "grad_norm": 0.36088070273399353,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 6080
    },
    {
      "epoch": 0.509154752947078,
      "grad_norm": 0.3604501187801361,
      "learning_rate": 2e-05,
      "loss": 0.6794,
      "step": 6090
    },
    {
      "epoch": 0.5099908034445281,
      "grad_norm": 0.36417829990386963,
      "learning_rate": 2e-05,
      "loss": 0.6645,
      "step": 6100
    },
    {
      "epoch": 0.5108268539419781,
      "grad_norm": 0.3451578617095947,
      "learning_rate": 2e-05,
      "loss": 0.6233,
      "step": 6110
    },
    {
      "epoch": 0.5116629044394282,
      "grad_norm": 0.3554096817970276,
      "learning_rate": 2e-05,
      "loss": 0.665,
      "step": 6120
    },
    {
      "epoch": 0.5124989549368781,
      "grad_norm": 0.35005977749824524,
      "learning_rate": 2e-05,
      "loss": 0.6469,
      "step": 6130
    },
    {
      "epoch": 0.5133350054343282,
      "grad_norm": 0.3871210515499115,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 6140
    },
    {
      "epoch": 0.5141710559317783,
      "grad_norm": 0.3464312255382538,
      "learning_rate": 2e-05,
      "loss": 0.6522,
      "step": 6150
    },
    {
      "epoch": 0.5150071064292283,
      "grad_norm": 0.39144864678382874,
      "learning_rate": 2e-05,
      "loss": 0.6291,
      "step": 6160
    },
    {
      "epoch": 0.5158431569266784,
      "grad_norm": 3.0951991081237793,
      "learning_rate": 2e-05,
      "loss": 0.6418,
      "step": 6170
    },
    {
      "epoch": 0.5166792074241284,
      "grad_norm": 0.3611271381378174,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 6180
    },
    {
      "epoch": 0.5175152579215785,
      "grad_norm": 0.34425824880599976,
      "learning_rate": 2e-05,
      "loss": 0.6269,
      "step": 6190
    },
    {
      "epoch": 0.5183513084190285,
      "grad_norm": 0.36573079228401184,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 6200
    },
    {
      "epoch": 0.5191873589164786,
      "grad_norm": 0.3877670466899872,
      "learning_rate": 2e-05,
      "loss": 0.6453,
      "step": 6210
    },
    {
      "epoch": 0.5200234094139286,
      "grad_norm": 0.3406931459903717,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 6220
    },
    {
      "epoch": 0.5208594599113786,
      "grad_norm": 0.3717559278011322,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 6230
    },
    {
      "epoch": 0.5216955104088287,
      "grad_norm": 0.3714001476764679,
      "learning_rate": 2e-05,
      "loss": 0.6494,
      "step": 6240
    },
    {
      "epoch": 0.5225315609062787,
      "grad_norm": 0.3619757294654846,
      "learning_rate": 2e-05,
      "loss": 0.684,
      "step": 6250
    },
    {
      "epoch": 0.5233676114037288,
      "grad_norm": 0.5626235008239746,
      "learning_rate": 2e-05,
      "loss": 0.6781,
      "step": 6260
    },
    {
      "epoch": 0.5242036619011788,
      "grad_norm": 0.4112389385700226,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 6270
    },
    {
      "epoch": 0.5250397123986289,
      "grad_norm": 0.35102036595344543,
      "learning_rate": 2e-05,
      "loss": 0.649,
      "step": 6280
    },
    {
      "epoch": 0.525875762896079,
      "grad_norm": 0.38397350907325745,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 6290
    },
    {
      "epoch": 0.526711813393529,
      "grad_norm": 0.33434274792671204,
      "learning_rate": 2e-05,
      "loss": 0.6539,
      "step": 6300
    },
    {
      "epoch": 0.527547863890979,
      "grad_norm": 0.36729466915130615,
      "learning_rate": 2e-05,
      "loss": 0.6675,
      "step": 6310
    },
    {
      "epoch": 0.528383914388429,
      "grad_norm": 0.43102672696113586,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 6320
    },
    {
      "epoch": 0.5292199648858791,
      "grad_norm": 0.34917303919792175,
      "learning_rate": 2e-05,
      "loss": 0.6566,
      "step": 6330
    },
    {
      "epoch": 0.5300560153833291,
      "grad_norm": 0.3342210054397583,
      "learning_rate": 2e-05,
      "loss": 0.6225,
      "step": 6340
    },
    {
      "epoch": 0.5308920658807792,
      "grad_norm": 0.3741057515144348,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 6350
    },
    {
      "epoch": 0.5317281163782293,
      "grad_norm": 0.3196578919887543,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 6360
    },
    {
      "epoch": 0.5325641668756793,
      "grad_norm": 0.3425282835960388,
      "learning_rate": 2e-05,
      "loss": 0.6556,
      "step": 6370
    },
    {
      "epoch": 0.5334002173731294,
      "grad_norm": 0.3632437586784363,
      "learning_rate": 2e-05,
      "loss": 0.647,
      "step": 6380
    },
    {
      "epoch": 0.5342362678705794,
      "grad_norm": 0.36416107416152954,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 6390
    },
    {
      "epoch": 0.5350723183680294,
      "grad_norm": 0.35709765553474426,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 6400
    },
    {
      "epoch": 0.5359083688654794,
      "grad_norm": 0.3619385063648224,
      "learning_rate": 2e-05,
      "loss": 0.6294,
      "step": 6410
    },
    {
      "epoch": 0.5367444193629295,
      "grad_norm": 0.38470229506492615,
      "learning_rate": 2e-05,
      "loss": 0.629,
      "step": 6420
    },
    {
      "epoch": 0.5375804698603796,
      "grad_norm": 0.32936784625053406,
      "learning_rate": 2e-05,
      "loss": 0.6454,
      "step": 6430
    },
    {
      "epoch": 0.5384165203578296,
      "grad_norm": 0.37622007727622986,
      "learning_rate": 2e-05,
      "loss": 0.6579,
      "step": 6440
    },
    {
      "epoch": 0.5392525708552797,
      "grad_norm": 0.4482446312904358,
      "learning_rate": 2e-05,
      "loss": 0.6649,
      "step": 6450
    },
    {
      "epoch": 0.5400886213527297,
      "grad_norm": 0.3462783694267273,
      "learning_rate": 2e-05,
      "loss": 0.6325,
      "step": 6460
    },
    {
      "epoch": 0.5409246718501798,
      "grad_norm": 0.35818323493003845,
      "learning_rate": 2e-05,
      "loss": 0.6042,
      "step": 6470
    },
    {
      "epoch": 0.5417607223476298,
      "grad_norm": 0.3384532034397125,
      "learning_rate": 2e-05,
      "loss": 0.6293,
      "step": 6480
    },
    {
      "epoch": 0.5425967728450798,
      "grad_norm": 0.3677841126918793,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 6490
    },
    {
      "epoch": 0.5434328233425298,
      "grad_norm": 0.31180340051651,
      "learning_rate": 2e-05,
      "loss": 0.6545,
      "step": 6500
    },
    {
      "epoch": 0.5442688738399799,
      "grad_norm": 0.34328535199165344,
      "learning_rate": 2e-05,
      "loss": 0.6377,
      "step": 6510
    },
    {
      "epoch": 0.54510492433743,
      "grad_norm": 0.45215195417404175,
      "learning_rate": 2e-05,
      "loss": 0.5667,
      "step": 6520
    },
    {
      "epoch": 0.54594097483488,
      "grad_norm": 0.3589378297328949,
      "learning_rate": 2e-05,
      "loss": 0.647,
      "step": 6530
    },
    {
      "epoch": 0.5467770253323301,
      "grad_norm": 0.356719046831131,
      "learning_rate": 2e-05,
      "loss": 0.6654,
      "step": 6540
    },
    {
      "epoch": 0.5476130758297801,
      "grad_norm": 0.3706061840057373,
      "learning_rate": 2e-05,
      "loss": 0.6574,
      "step": 6550
    },
    {
      "epoch": 0.5484491263272302,
      "grad_norm": 0.37721896171569824,
      "learning_rate": 2e-05,
      "loss": 0.6412,
      "step": 6560
    },
    {
      "epoch": 0.5492851768246803,
      "grad_norm": 0.3717316687107086,
      "learning_rate": 2e-05,
      "loss": 0.6301,
      "step": 6570
    },
    {
      "epoch": 0.5501212273221302,
      "grad_norm": 0.3622138500213623,
      "learning_rate": 2e-05,
      "loss": 0.6317,
      "step": 6580
    },
    {
      "epoch": 0.5509572778195803,
      "grad_norm": 0.373122900724411,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 6590
    },
    {
      "epoch": 0.5517933283170303,
      "grad_norm": 0.38592952489852905,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 6600
    },
    {
      "epoch": 0.5526293788144804,
      "grad_norm": 0.3523613512516022,
      "learning_rate": 2e-05,
      "loss": 0.642,
      "step": 6610
    },
    {
      "epoch": 0.5534654293119304,
      "grad_norm": 0.3676135838031769,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 6620
    },
    {
      "epoch": 0.5543014798093805,
      "grad_norm": 0.34784504771232605,
      "learning_rate": 2e-05,
      "loss": 0.6579,
      "step": 6630
    },
    {
      "epoch": 0.5551375303068306,
      "grad_norm": 0.32641980051994324,
      "learning_rate": 2e-05,
      "loss": 0.6133,
      "step": 6640
    },
    {
      "epoch": 0.5559735808042806,
      "grad_norm": 0.346467524766922,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 6650
    },
    {
      "epoch": 0.5568096313017307,
      "grad_norm": 0.3682650923728943,
      "learning_rate": 2e-05,
      "loss": 0.6602,
      "step": 6660
    },
    {
      "epoch": 0.5576456817991807,
      "grad_norm": 0.36967113614082336,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 6670
    },
    {
      "epoch": 0.5584817322966307,
      "grad_norm": 0.3759212791919708,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 6680
    },
    {
      "epoch": 0.5593177827940807,
      "grad_norm": 0.38480356335639954,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 6690
    },
    {
      "epoch": 0.5601538332915308,
      "grad_norm": 0.3424999415874481,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 6700
    },
    {
      "epoch": 0.5609898837889808,
      "grad_norm": 0.3380453884601593,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 6710
    },
    {
      "epoch": 0.5618259342864309,
      "grad_norm": 0.372394323348999,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 6720
    },
    {
      "epoch": 0.562661984783881,
      "grad_norm": 0.32958224415779114,
      "learning_rate": 2e-05,
      "loss": 0.6435,
      "step": 6730
    },
    {
      "epoch": 0.563498035281331,
      "grad_norm": 1.1284234523773193,
      "learning_rate": 2e-05,
      "loss": 0.7042,
      "step": 6740
    },
    {
      "epoch": 0.5643340857787811,
      "grad_norm": 0.3774583339691162,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 6750
    },
    {
      "epoch": 0.5651701362762311,
      "grad_norm": 1.3705381155014038,
      "learning_rate": 2e-05,
      "loss": 0.6363,
      "step": 6760
    },
    {
      "epoch": 0.5660061867736811,
      "grad_norm": 0.35694143176078796,
      "learning_rate": 2e-05,
      "loss": 0.6769,
      "step": 6770
    },
    {
      "epoch": 0.5668422372711311,
      "grad_norm": 0.34649670124053955,
      "learning_rate": 2e-05,
      "loss": 0.6252,
      "step": 6780
    },
    {
      "epoch": 0.5676782877685812,
      "grad_norm": 0.3721681535243988,
      "learning_rate": 2e-05,
      "loss": 0.6395,
      "step": 6790
    },
    {
      "epoch": 0.5685143382660313,
      "grad_norm": 0.3782765865325928,
      "learning_rate": 2e-05,
      "loss": 0.6545,
      "step": 6800
    },
    {
      "epoch": 0.5693503887634813,
      "grad_norm": 0.3562970757484436,
      "learning_rate": 2e-05,
      "loss": 0.674,
      "step": 6810
    },
    {
      "epoch": 0.5701864392609314,
      "grad_norm": 0.3524180054664612,
      "learning_rate": 2e-05,
      "loss": 0.6194,
      "step": 6820
    },
    {
      "epoch": 0.5710224897583814,
      "grad_norm": 0.35739973187446594,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 6830
    },
    {
      "epoch": 0.5718585402558315,
      "grad_norm": 0.37304675579071045,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 6840
    },
    {
      "epoch": 0.5726945907532816,
      "grad_norm": 0.33797687292099,
      "learning_rate": 2e-05,
      "loss": 0.6396,
      "step": 6850
    },
    {
      "epoch": 0.5735306412507315,
      "grad_norm": 0.3608052134513855,
      "learning_rate": 2e-05,
      "loss": 0.6587,
      "step": 6860
    },
    {
      "epoch": 0.5743666917481816,
      "grad_norm": 0.35808244347572327,
      "learning_rate": 2e-05,
      "loss": 0.6712,
      "step": 6870
    },
    {
      "epoch": 0.5752027422456316,
      "grad_norm": 0.3602573275566101,
      "learning_rate": 2e-05,
      "loss": 0.6318,
      "step": 6880
    },
    {
      "epoch": 0.5760387927430817,
      "grad_norm": 0.35034722089767456,
      "learning_rate": 2e-05,
      "loss": 0.6586,
      "step": 6890
    },
    {
      "epoch": 0.5768748432405317,
      "grad_norm": 0.37191247940063477,
      "learning_rate": 2e-05,
      "loss": 0.6357,
      "step": 6900
    },
    {
      "epoch": 0.5777108937379818,
      "grad_norm": 0.33553212881088257,
      "learning_rate": 2e-05,
      "loss": 0.6246,
      "step": 6910
    },
    {
      "epoch": 0.5785469442354318,
      "grad_norm": 0.36928558349609375,
      "learning_rate": 2e-05,
      "loss": 0.6591,
      "step": 6920
    },
    {
      "epoch": 0.5793829947328819,
      "grad_norm": 0.3528483211994171,
      "learning_rate": 2e-05,
      "loss": 0.6576,
      "step": 6930
    },
    {
      "epoch": 0.580219045230332,
      "grad_norm": 0.3438965380191803,
      "learning_rate": 2e-05,
      "loss": 0.6437,
      "step": 6940
    },
    {
      "epoch": 0.5810550957277819,
      "grad_norm": 0.3749878704547882,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 6950
    },
    {
      "epoch": 0.581891146225232,
      "grad_norm": 0.3737831115722656,
      "learning_rate": 2e-05,
      "loss": 0.6108,
      "step": 6960
    },
    {
      "epoch": 0.582727196722682,
      "grad_norm": 0.3983328342437744,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 6970
    },
    {
      "epoch": 0.5835632472201321,
      "grad_norm": 0.37412458658218384,
      "learning_rate": 2e-05,
      "loss": 0.628,
      "step": 6980
    },
    {
      "epoch": 0.5843992977175821,
      "grad_norm": 0.3427202105522156,
      "learning_rate": 2e-05,
      "loss": 0.6433,
      "step": 6990
    },
    {
      "epoch": 0.5852353482150322,
      "grad_norm": 0.400119811296463,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 7000
    },
    {
      "epoch": 0.5860713987124823,
      "grad_norm": 0.3864668607711792,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 7010
    },
    {
      "epoch": 0.5869074492099323,
      "grad_norm": 0.44247785210609436,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 7020
    },
    {
      "epoch": 0.5877434997073824,
      "grad_norm": 0.34956371784210205,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 7030
    },
    {
      "epoch": 0.5885795502048323,
      "grad_norm": 0.4014231264591217,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 7040
    },
    {
      "epoch": 0.5894156007022824,
      "grad_norm": 0.3661552667617798,
      "learning_rate": 2e-05,
      "loss": 0.6507,
      "step": 7050
    },
    {
      "epoch": 0.5902516511997324,
      "grad_norm": 0.3759726881980896,
      "learning_rate": 2e-05,
      "loss": 0.6277,
      "step": 7060
    },
    {
      "epoch": 0.5910877016971825,
      "grad_norm": 0.40143024921417236,
      "learning_rate": 2e-05,
      "loss": 0.627,
      "step": 7070
    },
    {
      "epoch": 0.5919237521946326,
      "grad_norm": 0.356636106967926,
      "learning_rate": 2e-05,
      "loss": 0.6167,
      "step": 7080
    },
    {
      "epoch": 0.5927598026920826,
      "grad_norm": 0.3682616353034973,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 7090
    },
    {
      "epoch": 0.5935958531895327,
      "grad_norm": 0.3548130393028259,
      "learning_rate": 2e-05,
      "loss": 0.6637,
      "step": 7100
    },
    {
      "epoch": 0.5944319036869827,
      "grad_norm": 0.3500625789165497,
      "learning_rate": 2e-05,
      "loss": 0.6127,
      "step": 7110
    },
    {
      "epoch": 0.5952679541844328,
      "grad_norm": 0.3438291549682617,
      "learning_rate": 2e-05,
      "loss": 0.6674,
      "step": 7120
    },
    {
      "epoch": 0.5961040046818827,
      "grad_norm": 0.3671638071537018,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 7130
    },
    {
      "epoch": 0.5969400551793328,
      "grad_norm": 0.3384701907634735,
      "learning_rate": 2e-05,
      "loss": 0.6019,
      "step": 7140
    },
    {
      "epoch": 0.5977761056767829,
      "grad_norm": 0.3600935935974121,
      "learning_rate": 2e-05,
      "loss": 0.6519,
      "step": 7150
    },
    {
      "epoch": 0.5986121561742329,
      "grad_norm": 0.344492644071579,
      "learning_rate": 2e-05,
      "loss": 0.5987,
      "step": 7160
    },
    {
      "epoch": 0.599448206671683,
      "grad_norm": 0.37941044569015503,
      "learning_rate": 2e-05,
      "loss": 0.6615,
      "step": 7170
    },
    {
      "epoch": 0.600284257169133,
      "grad_norm": 0.36006686091423035,
      "learning_rate": 2e-05,
      "loss": 0.6256,
      "step": 7180
    },
    {
      "epoch": 0.6011203076665831,
      "grad_norm": 0.35914772748947144,
      "learning_rate": 2e-05,
      "loss": 0.6211,
      "step": 7190
    },
    {
      "epoch": 0.6019563581640331,
      "grad_norm": 0.3418668806552887,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 7200
    },
    {
      "epoch": 0.6027924086614832,
      "grad_norm": 0.34393617510795593,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 7210
    },
    {
      "epoch": 0.6036284591589332,
      "grad_norm": 0.3628663420677185,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 7220
    },
    {
      "epoch": 0.6044645096563832,
      "grad_norm": 0.38836830854415894,
      "learning_rate": 2e-05,
      "loss": 0.6643,
      "step": 7230
    },
    {
      "epoch": 0.6053005601538333,
      "grad_norm": 0.4073199927806854,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 7240
    },
    {
      "epoch": 0.6061366106512833,
      "grad_norm": 0.40124839544296265,
      "learning_rate": 2e-05,
      "loss": 0.646,
      "step": 7250
    },
    {
      "epoch": 0.6069726611487334,
      "grad_norm": 0.3892489969730377,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 7260
    },
    {
      "epoch": 0.6078087116461834,
      "grad_norm": 0.382063627243042,
      "learning_rate": 2e-05,
      "loss": 0.6367,
      "step": 7270
    },
    {
      "epoch": 0.6086447621436335,
      "grad_norm": 0.37543946504592896,
      "learning_rate": 2e-05,
      "loss": 0.6326,
      "step": 7280
    },
    {
      "epoch": 0.6094808126410836,
      "grad_norm": 0.368030309677124,
      "learning_rate": 2e-05,
      "loss": 0.6599,
      "step": 7290
    },
    {
      "epoch": 0.6103168631385336,
      "grad_norm": 0.3848540186882019,
      "learning_rate": 2e-05,
      "loss": 0.649,
      "step": 7300
    },
    {
      "epoch": 0.6111529136359836,
      "grad_norm": 0.39881232380867004,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 7310
    },
    {
      "epoch": 0.6119889641334336,
      "grad_norm": 0.34004178643226624,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 7320
    },
    {
      "epoch": 0.6128250146308837,
      "grad_norm": 0.5286120176315308,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 7330
    },
    {
      "epoch": 0.6136610651283337,
      "grad_norm": 0.3932904005050659,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 7340
    },
    {
      "epoch": 0.6144971156257838,
      "grad_norm": 0.38622966408729553,
      "learning_rate": 2e-05,
      "loss": 0.6279,
      "step": 7350
    },
    {
      "epoch": 0.6153331661232339,
      "grad_norm": 0.32701432704925537,
      "learning_rate": 2e-05,
      "loss": 0.6298,
      "step": 7360
    },
    {
      "epoch": 0.6161692166206839,
      "grad_norm": 0.4505709707736969,
      "learning_rate": 2e-05,
      "loss": 0.65,
      "step": 7370
    },
    {
      "epoch": 0.617005267118134,
      "grad_norm": 0.3811368942260742,
      "learning_rate": 2e-05,
      "loss": 0.6472,
      "step": 7380
    },
    {
      "epoch": 0.617841317615584,
      "grad_norm": 0.40746355056762695,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 7390
    },
    {
      "epoch": 0.618677368113034,
      "grad_norm": 0.3638688623905182,
      "learning_rate": 2e-05,
      "loss": 0.6138,
      "step": 7400
    },
    {
      "epoch": 0.619513418610484,
      "grad_norm": 0.34692806005477905,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 7410
    },
    {
      "epoch": 0.6203494691079341,
      "grad_norm": 0.3281863033771515,
      "learning_rate": 2e-05,
      "loss": 0.6113,
      "step": 7420
    },
    {
      "epoch": 0.6211855196053842,
      "grad_norm": 0.34765759110450745,
      "learning_rate": 2e-05,
      "loss": 0.6179,
      "step": 7430
    },
    {
      "epoch": 0.6220215701028342,
      "grad_norm": 0.3845807611942291,
      "learning_rate": 2e-05,
      "loss": 0.6552,
      "step": 7440
    },
    {
      "epoch": 0.6228576206002843,
      "grad_norm": 0.330185204744339,
      "learning_rate": 2e-05,
      "loss": 0.6312,
      "step": 7450
    },
    {
      "epoch": 0.6236936710977343,
      "grad_norm": 0.39414265751838684,
      "learning_rate": 2e-05,
      "loss": 0.6174,
      "step": 7460
    },
    {
      "epoch": 0.6245297215951844,
      "grad_norm": 0.3710842430591583,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 7470
    },
    {
      "epoch": 0.6253657720926344,
      "grad_norm": 0.3537895381450653,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 7480
    },
    {
      "epoch": 0.6262018225900844,
      "grad_norm": 0.34117355942726135,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 7490
    },
    {
      "epoch": 0.6270378730875344,
      "grad_norm": 0.39136168360710144,
      "learning_rate": 2e-05,
      "loss": 0.6751,
      "step": 7500
    },
    {
      "epoch": 0.6278739235849845,
      "grad_norm": 0.34444117546081543,
      "learning_rate": 2e-05,
      "loss": 0.656,
      "step": 7510
    },
    {
      "epoch": 0.6287099740824346,
      "grad_norm": 0.33393093943595886,
      "learning_rate": 2e-05,
      "loss": 0.6259,
      "step": 7520
    },
    {
      "epoch": 0.6295460245798846,
      "grad_norm": 0.35947027802467346,
      "learning_rate": 2e-05,
      "loss": 0.6505,
      "step": 7530
    },
    {
      "epoch": 0.6303820750773347,
      "grad_norm": 0.37965118885040283,
      "learning_rate": 2e-05,
      "loss": 0.6612,
      "step": 7540
    },
    {
      "epoch": 0.6312181255747847,
      "grad_norm": 0.36189302802085876,
      "learning_rate": 2e-05,
      "loss": 0.623,
      "step": 7550
    },
    {
      "epoch": 0.6320541760722348,
      "grad_norm": 0.6667383909225464,
      "learning_rate": 2e-05,
      "loss": 0.627,
      "step": 7560
    },
    {
      "epoch": 0.6328902265696849,
      "grad_norm": 0.33587297797203064,
      "learning_rate": 2e-05,
      "loss": 0.6493,
      "step": 7570
    },
    {
      "epoch": 0.6337262770671348,
      "grad_norm": 0.345249205827713,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 7580
    },
    {
      "epoch": 0.6345623275645849,
      "grad_norm": 0.33790817856788635,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 7590
    },
    {
      "epoch": 0.6353983780620349,
      "grad_norm": 0.36260828375816345,
      "learning_rate": 2e-05,
      "loss": 0.6786,
      "step": 7600
    },
    {
      "epoch": 0.636234428559485,
      "grad_norm": 0.3199288547039032,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 7610
    },
    {
      "epoch": 0.637070479056935,
      "grad_norm": 0.3785347044467926,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 7620
    },
    {
      "epoch": 0.6379065295543851,
      "grad_norm": 0.37974804639816284,
      "learning_rate": 2e-05,
      "loss": 0.6376,
      "step": 7630
    },
    {
      "epoch": 0.6387425800518352,
      "grad_norm": 0.37508928775787354,
      "learning_rate": 2e-05,
      "loss": 0.5929,
      "step": 7640
    },
    {
      "epoch": 0.6395786305492852,
      "grad_norm": 0.38905012607574463,
      "learning_rate": 2e-05,
      "loss": 0.6236,
      "step": 7650
    },
    {
      "epoch": 0.6404146810467353,
      "grad_norm": 0.35950303077697754,
      "learning_rate": 2e-05,
      "loss": 0.6484,
      "step": 7660
    },
    {
      "epoch": 0.6412507315441852,
      "grad_norm": 0.33383458852767944,
      "learning_rate": 2e-05,
      "loss": 0.6132,
      "step": 7670
    },
    {
      "epoch": 0.6420867820416353,
      "grad_norm": 0.5303345918655396,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 7680
    },
    {
      "epoch": 0.6429228325390853,
      "grad_norm": 0.5810869932174683,
      "learning_rate": 2e-05,
      "loss": 0.6493,
      "step": 7690
    },
    {
      "epoch": 0.6437588830365354,
      "grad_norm": 0.34757936000823975,
      "learning_rate": 2e-05,
      "loss": 0.6366,
      "step": 7700
    },
    {
      "epoch": 0.6445949335339854,
      "grad_norm": 0.33055534958839417,
      "learning_rate": 2e-05,
      "loss": 0.5967,
      "step": 7710
    },
    {
      "epoch": 0.6454309840314355,
      "grad_norm": 0.38666558265686035,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 7720
    },
    {
      "epoch": 0.6462670345288856,
      "grad_norm": 0.3405543565750122,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 7730
    },
    {
      "epoch": 0.6471030850263356,
      "grad_norm": 0.36779025197029114,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 7740
    },
    {
      "epoch": 0.6479391355237857,
      "grad_norm": 0.3702530562877655,
      "learning_rate": 2e-05,
      "loss": 0.6534,
      "step": 7750
    },
    {
      "epoch": 0.6487751860212356,
      "grad_norm": 0.369750440120697,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 7760
    },
    {
      "epoch": 0.6496112365186857,
      "grad_norm": 0.3632996380329132,
      "learning_rate": 2e-05,
      "loss": 0.6181,
      "step": 7770
    },
    {
      "epoch": 0.6504472870161357,
      "grad_norm": 0.46269580721855164,
      "learning_rate": 2e-05,
      "loss": 0.6374,
      "step": 7780
    },
    {
      "epoch": 0.6512833375135858,
      "grad_norm": 0.36560311913490295,
      "learning_rate": 2e-05,
      "loss": 0.6059,
      "step": 7790
    },
    {
      "epoch": 0.6521193880110359,
      "grad_norm": 0.36217010021209717,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 7800
    },
    {
      "epoch": 0.6529554385084859,
      "grad_norm": 0.34714728593826294,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 7810
    },
    {
      "epoch": 0.653791489005936,
      "grad_norm": 0.35189947485923767,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 7820
    },
    {
      "epoch": 0.654627539503386,
      "grad_norm": 0.33983752131462097,
      "learning_rate": 2e-05,
      "loss": 0.6185,
      "step": 7830
    },
    {
      "epoch": 0.6554635900008361,
      "grad_norm": 0.39625221490859985,
      "learning_rate": 2e-05,
      "loss": 0.6197,
      "step": 7840
    },
    {
      "epoch": 0.656299640498286,
      "grad_norm": 0.35485056042671204,
      "learning_rate": 2e-05,
      "loss": 0.6513,
      "step": 7850
    },
    {
      "epoch": 0.6571356909957361,
      "grad_norm": 0.3543018400669098,
      "learning_rate": 2e-05,
      "loss": 0.6317,
      "step": 7860
    },
    {
      "epoch": 0.6579717414931862,
      "grad_norm": 0.36564669013023376,
      "learning_rate": 2e-05,
      "loss": 0.6724,
      "step": 7870
    },
    {
      "epoch": 0.6588077919906362,
      "grad_norm": 0.3575376570224762,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 7880
    },
    {
      "epoch": 0.6596438424880863,
      "grad_norm": 0.353441447019577,
      "learning_rate": 2e-05,
      "loss": 0.6129,
      "step": 7890
    },
    {
      "epoch": 0.6604798929855363,
      "grad_norm": 0.3676687180995941,
      "learning_rate": 2e-05,
      "loss": 0.6493,
      "step": 7900
    },
    {
      "epoch": 0.6613159434829864,
      "grad_norm": 0.3700011074542999,
      "learning_rate": 2e-05,
      "loss": 0.6545,
      "step": 7910
    },
    {
      "epoch": 0.6621519939804364,
      "grad_norm": 0.344151109457016,
      "learning_rate": 2e-05,
      "loss": 0.6153,
      "step": 7920
    },
    {
      "epoch": 0.6629880444778865,
      "grad_norm": 0.3995813727378845,
      "learning_rate": 2e-05,
      "loss": 0.6203,
      "step": 7930
    },
    {
      "epoch": 0.6638240949753366,
      "grad_norm": 0.343106746673584,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 7940
    },
    {
      "epoch": 0.6646601454727865,
      "grad_norm": 0.37969765067100525,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 7950
    },
    {
      "epoch": 0.6654961959702366,
      "grad_norm": 0.39228555560112,
      "learning_rate": 2e-05,
      "loss": 0.6376,
      "step": 7960
    },
    {
      "epoch": 0.6663322464676866,
      "grad_norm": 0.34169524908065796,
      "learning_rate": 2e-05,
      "loss": 0.6334,
      "step": 7970
    },
    {
      "epoch": 0.6671682969651367,
      "grad_norm": 0.363491952419281,
      "learning_rate": 2e-05,
      "loss": 0.649,
      "step": 7980
    },
    {
      "epoch": 0.6680043474625867,
      "grad_norm": 0.39255693554878235,
      "learning_rate": 2e-05,
      "loss": 0.6242,
      "step": 7990
    },
    {
      "epoch": 0.6688403979600368,
      "grad_norm": 0.35665518045425415,
      "learning_rate": 2e-05,
      "loss": 0.6105,
      "step": 8000
    },
    {
      "epoch": 0.6696764484574869,
      "grad_norm": 0.3387342691421509,
      "learning_rate": 2e-05,
      "loss": 0.6433,
      "step": 8010
    },
    {
      "epoch": 0.6705124989549369,
      "grad_norm": 0.3877888321876526,
      "learning_rate": 2e-05,
      "loss": 0.6001,
      "step": 8020
    },
    {
      "epoch": 0.671348549452387,
      "grad_norm": 0.34042635560035706,
      "learning_rate": 2e-05,
      "loss": 0.6602,
      "step": 8030
    },
    {
      "epoch": 0.6721845999498369,
      "grad_norm": 0.38062193989753723,
      "learning_rate": 2e-05,
      "loss": 0.6341,
      "step": 8040
    },
    {
      "epoch": 0.673020650447287,
      "grad_norm": 0.4432688355445862,
      "learning_rate": 2e-05,
      "loss": 0.6399,
      "step": 8050
    },
    {
      "epoch": 0.673856700944737,
      "grad_norm": 0.3779374659061432,
      "learning_rate": 2e-05,
      "loss": 0.6599,
      "step": 8060
    },
    {
      "epoch": 0.6746927514421871,
      "grad_norm": 0.35633039474487305,
      "learning_rate": 2e-05,
      "loss": 0.5967,
      "step": 8070
    },
    {
      "epoch": 0.6755288019396372,
      "grad_norm": 0.35728421807289124,
      "learning_rate": 2e-05,
      "loss": 0.621,
      "step": 8080
    },
    {
      "epoch": 0.6763648524370872,
      "grad_norm": 0.3876197636127472,
      "learning_rate": 2e-05,
      "loss": 0.6723,
      "step": 8090
    },
    {
      "epoch": 0.6772009029345373,
      "grad_norm": 0.34664300084114075,
      "learning_rate": 2e-05,
      "loss": 0.5912,
      "step": 8100
    },
    {
      "epoch": 0.6780369534319873,
      "grad_norm": 0.3978821337223053,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 8110
    },
    {
      "epoch": 0.6788730039294374,
      "grad_norm": 0.37592893838882446,
      "learning_rate": 2e-05,
      "loss": 0.6264,
      "step": 8120
    },
    {
      "epoch": 0.6797090544268873,
      "grad_norm": 0.354134738445282,
      "learning_rate": 2e-05,
      "loss": 0.6626,
      "step": 8130
    },
    {
      "epoch": 0.6805451049243374,
      "grad_norm": 0.36554190516471863,
      "learning_rate": 2e-05,
      "loss": 0.6257,
      "step": 8140
    },
    {
      "epoch": 0.6813811554217875,
      "grad_norm": 0.34598568081855774,
      "learning_rate": 2e-05,
      "loss": 0.6476,
      "step": 8150
    },
    {
      "epoch": 0.6822172059192375,
      "grad_norm": 0.39437681436538696,
      "learning_rate": 2e-05,
      "loss": 0.6053,
      "step": 8160
    },
    {
      "epoch": 0.6830532564166876,
      "grad_norm": 0.3681078255176544,
      "learning_rate": 2e-05,
      "loss": 0.6722,
      "step": 8170
    },
    {
      "epoch": 0.6838893069141376,
      "grad_norm": 0.35879725217819214,
      "learning_rate": 2e-05,
      "loss": 0.6171,
      "step": 8180
    },
    {
      "epoch": 0.6847253574115877,
      "grad_norm": 0.3910785913467407,
      "learning_rate": 2e-05,
      "loss": 0.6451,
      "step": 8190
    },
    {
      "epoch": 0.6855614079090377,
      "grad_norm": 0.3515801727771759,
      "learning_rate": 2e-05,
      "loss": 0.6232,
      "step": 8200
    },
    {
      "epoch": 0.6863974584064878,
      "grad_norm": 0.37083208560943604,
      "learning_rate": 2e-05,
      "loss": 0.609,
      "step": 8210
    },
    {
      "epoch": 0.6872335089039378,
      "grad_norm": 0.37073031067848206,
      "learning_rate": 2e-05,
      "loss": 0.6301,
      "step": 8220
    },
    {
      "epoch": 0.6880695594013878,
      "grad_norm": 0.39060622453689575,
      "learning_rate": 2e-05,
      "loss": 0.6565,
      "step": 8230
    },
    {
      "epoch": 0.6889056098988379,
      "grad_norm": 0.34797486662864685,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 8240
    },
    {
      "epoch": 0.6897416603962879,
      "grad_norm": 0.3574109971523285,
      "learning_rate": 2e-05,
      "loss": 0.6223,
      "step": 8250
    },
    {
      "epoch": 0.690577710893738,
      "grad_norm": 0.31913992762565613,
      "learning_rate": 2e-05,
      "loss": 0.6159,
      "step": 8260
    },
    {
      "epoch": 0.691413761391188,
      "grad_norm": 0.37921833992004395,
      "learning_rate": 2e-05,
      "loss": 0.6606,
      "step": 8270
    },
    {
      "epoch": 0.6922498118886381,
      "grad_norm": 0.35300713777542114,
      "learning_rate": 2e-05,
      "loss": 0.5873,
      "step": 8280
    },
    {
      "epoch": 0.6930858623860882,
      "grad_norm": 0.33214089274406433,
      "learning_rate": 2e-05,
      "loss": 0.6041,
      "step": 8290
    },
    {
      "epoch": 0.6939219128835382,
      "grad_norm": 0.36767029762268066,
      "learning_rate": 2e-05,
      "loss": 0.6556,
      "step": 8300
    },
    {
      "epoch": 0.6947579633809882,
      "grad_norm": 0.3897394835948944,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 8310
    },
    {
      "epoch": 0.6955940138784382,
      "grad_norm": 0.327609658241272,
      "learning_rate": 2e-05,
      "loss": 0.6426,
      "step": 8320
    },
    {
      "epoch": 0.6964300643758883,
      "grad_norm": 0.39035043120384216,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 8330
    },
    {
      "epoch": 0.6972661148733383,
      "grad_norm": 0.3671237528324127,
      "learning_rate": 2e-05,
      "loss": 0.6348,
      "step": 8340
    },
    {
      "epoch": 0.6981021653707884,
      "grad_norm": 0.3611215054988861,
      "learning_rate": 2e-05,
      "loss": 0.644,
      "step": 8350
    },
    {
      "epoch": 0.6989382158682385,
      "grad_norm": 0.32708102464675903,
      "learning_rate": 2e-05,
      "loss": 0.6452,
      "step": 8360
    },
    {
      "epoch": 0.6997742663656885,
      "grad_norm": 0.39373376965522766,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 8370
    },
    {
      "epoch": 0.7006103168631386,
      "grad_norm": 0.34182775020599365,
      "learning_rate": 2e-05,
      "loss": 0.6536,
      "step": 8380
    },
    {
      "epoch": 0.7014463673605886,
      "grad_norm": 0.34567663073539734,
      "learning_rate": 2e-05,
      "loss": 0.6097,
      "step": 8390
    },
    {
      "epoch": 0.7022824178580386,
      "grad_norm": 0.37950968742370605,
      "learning_rate": 2e-05,
      "loss": 0.5947,
      "step": 8400
    },
    {
      "epoch": 0.7031184683554886,
      "grad_norm": 0.32786867022514343,
      "learning_rate": 2e-05,
      "loss": 0.5929,
      "step": 8410
    },
    {
      "epoch": 0.7039545188529387,
      "grad_norm": 0.37132781744003296,
      "learning_rate": 2e-05,
      "loss": 0.6229,
      "step": 8420
    },
    {
      "epoch": 0.7047905693503888,
      "grad_norm": 0.33797481656074524,
      "learning_rate": 2e-05,
      "loss": 0.6365,
      "step": 8430
    },
    {
      "epoch": 0.7056266198478388,
      "grad_norm": 0.40617161989212036,
      "learning_rate": 2e-05,
      "loss": 0.6384,
      "step": 8440
    },
    {
      "epoch": 0.7064626703452889,
      "grad_norm": 0.37914934754371643,
      "learning_rate": 2e-05,
      "loss": 0.6392,
      "step": 8450
    },
    {
      "epoch": 0.7072987208427389,
      "grad_norm": 0.3597451150417328,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 8460
    },
    {
      "epoch": 0.708134771340189,
      "grad_norm": 0.4551459848880768,
      "learning_rate": 2e-05,
      "loss": 0.6216,
      "step": 8470
    },
    {
      "epoch": 0.708970821837639,
      "grad_norm": 0.3727883994579315,
      "learning_rate": 2e-05,
      "loss": 0.6125,
      "step": 8480
    },
    {
      "epoch": 0.709806872335089,
      "grad_norm": 0.38276681303977966,
      "learning_rate": 2e-05,
      "loss": 0.6247,
      "step": 8490
    },
    {
      "epoch": 0.710642922832539,
      "grad_norm": 0.34124183654785156,
      "learning_rate": 2e-05,
      "loss": 0.6055,
      "step": 8500
    },
    {
      "epoch": 0.7114789733299891,
      "grad_norm": 0.353573739528656,
      "learning_rate": 2e-05,
      "loss": 0.6054,
      "step": 8510
    },
    {
      "epoch": 0.7123150238274392,
      "grad_norm": 0.35015061497688293,
      "learning_rate": 2e-05,
      "loss": 0.6108,
      "step": 8520
    },
    {
      "epoch": 0.7131510743248892,
      "grad_norm": 0.36228853464126587,
      "learning_rate": 2e-05,
      "loss": 0.6332,
      "step": 8530
    },
    {
      "epoch": 0.7139871248223393,
      "grad_norm": 0.35383185744285583,
      "learning_rate": 2e-05,
      "loss": 0.596,
      "step": 8540
    },
    {
      "epoch": 0.7148231753197893,
      "grad_norm": 0.34764471650123596,
      "learning_rate": 2e-05,
      "loss": 0.6033,
      "step": 8550
    },
    {
      "epoch": 0.7156592258172394,
      "grad_norm": 0.36279934644699097,
      "learning_rate": 2e-05,
      "loss": 0.6334,
      "step": 8560
    },
    {
      "epoch": 0.7164952763146895,
      "grad_norm": 0.40890178084373474,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 8570
    },
    {
      "epoch": 0.7173313268121394,
      "grad_norm": 0.39055535197257996,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 8580
    },
    {
      "epoch": 0.7181673773095895,
      "grad_norm": 0.3766123354434967,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 8590
    },
    {
      "epoch": 0.7190034278070395,
      "grad_norm": 0.3660091161727905,
      "learning_rate": 2e-05,
      "loss": 0.6036,
      "step": 8600
    },
    {
      "epoch": 0.7198394783044896,
      "grad_norm": 0.3255775570869446,
      "learning_rate": 2e-05,
      "loss": 0.6151,
      "step": 8610
    },
    {
      "epoch": 0.7206755288019396,
      "grad_norm": 0.3614455461502075,
      "learning_rate": 2e-05,
      "loss": 0.6326,
      "step": 8620
    },
    {
      "epoch": 0.7215115792993897,
      "grad_norm": 0.3693804442882538,
      "learning_rate": 2e-05,
      "loss": 0.63,
      "step": 8630
    },
    {
      "epoch": 0.7223476297968398,
      "grad_norm": 0.4027959406375885,
      "learning_rate": 2e-05,
      "loss": 0.6115,
      "step": 8640
    },
    {
      "epoch": 0.7231836802942898,
      "grad_norm": 0.42743971943855286,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 8650
    },
    {
      "epoch": 0.7240197307917399,
      "grad_norm": 0.4095340073108673,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 8660
    },
    {
      "epoch": 0.7248557812891898,
      "grad_norm": 0.4437895715236664,
      "learning_rate": 2e-05,
      "loss": 0.6281,
      "step": 8670
    },
    {
      "epoch": 0.7256918317866399,
      "grad_norm": 0.35897666215896606,
      "learning_rate": 2e-05,
      "loss": 0.6269,
      "step": 8680
    },
    {
      "epoch": 0.7265278822840899,
      "grad_norm": 0.35686805844306946,
      "learning_rate": 2e-05,
      "loss": 0.6117,
      "step": 8690
    },
    {
      "epoch": 0.72736393278154,
      "grad_norm": 0.37403956055641174,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 8700
    },
    {
      "epoch": 0.72819998327899,
      "grad_norm": 0.3633500933647156,
      "learning_rate": 2e-05,
      "loss": 0.6152,
      "step": 8710
    },
    {
      "epoch": 0.7290360337764401,
      "grad_norm": 0.3864735960960388,
      "learning_rate": 2e-05,
      "loss": 0.6405,
      "step": 8720
    },
    {
      "epoch": 0.7298720842738902,
      "grad_norm": 0.35216525197029114,
      "learning_rate": 2e-05,
      "loss": 0.6109,
      "step": 8730
    },
    {
      "epoch": 0.7307081347713402,
      "grad_norm": 0.36769869923591614,
      "learning_rate": 2e-05,
      "loss": 0.5939,
      "step": 8740
    },
    {
      "epoch": 0.7315441852687903,
      "grad_norm": 0.40076515078544617,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 8750
    },
    {
      "epoch": 0.7323802357662402,
      "grad_norm": 0.38828063011169434,
      "learning_rate": 2e-05,
      "loss": 0.635,
      "step": 8760
    },
    {
      "epoch": 0.7332162862636903,
      "grad_norm": 0.4033960700035095,
      "learning_rate": 2e-05,
      "loss": 0.647,
      "step": 8770
    },
    {
      "epoch": 0.7340523367611403,
      "grad_norm": 0.3574748933315277,
      "learning_rate": 2e-05,
      "loss": 0.6295,
      "step": 8780
    },
    {
      "epoch": 0.7348883872585904,
      "grad_norm": 0.35619619488716125,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 8790
    },
    {
      "epoch": 0.7357244377560405,
      "grad_norm": 0.3646388053894043,
      "learning_rate": 2e-05,
      "loss": 0.6086,
      "step": 8800
    },
    {
      "epoch": 0.7365604882534905,
      "grad_norm": 0.38466766476631165,
      "learning_rate": 2e-05,
      "loss": 0.6084,
      "step": 8810
    },
    {
      "epoch": 0.7373965387509406,
      "grad_norm": 0.4121417701244354,
      "learning_rate": 2e-05,
      "loss": 0.6403,
      "step": 8820
    },
    {
      "epoch": 0.7382325892483906,
      "grad_norm": 0.35711365938186646,
      "learning_rate": 2e-05,
      "loss": 0.6179,
      "step": 8830
    },
    {
      "epoch": 0.7390686397458407,
      "grad_norm": 0.3566686809062958,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 8840
    },
    {
      "epoch": 0.7399046902432906,
      "grad_norm": 0.34277498722076416,
      "learning_rate": 2e-05,
      "loss": 0.6172,
      "step": 8850
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 0.40746399760246277,
      "learning_rate": 2e-05,
      "loss": 0.6101,
      "step": 8860
    },
    {
      "epoch": 0.7415767912381908,
      "grad_norm": 0.37263932824134827,
      "learning_rate": 2e-05,
      "loss": 0.6126,
      "step": 8870
    },
    {
      "epoch": 0.7424128417356408,
      "grad_norm": 0.43366503715515137,
      "learning_rate": 2e-05,
      "loss": 0.6569,
      "step": 8880
    },
    {
      "epoch": 0.7432488922330909,
      "grad_norm": 0.3612329363822937,
      "learning_rate": 2e-05,
      "loss": 0.6138,
      "step": 8890
    },
    {
      "epoch": 0.7440849427305409,
      "grad_norm": 0.3808938264846802,
      "learning_rate": 2e-05,
      "loss": 0.6506,
      "step": 8900
    },
    {
      "epoch": 0.744920993227991,
      "grad_norm": 0.40133222937583923,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 8910
    },
    {
      "epoch": 0.745757043725441,
      "grad_norm": 0.3133492171764374,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 8920
    },
    {
      "epoch": 0.7465930942228911,
      "grad_norm": 0.3125399947166443,
      "learning_rate": 2e-05,
      "loss": 0.6223,
      "step": 8930
    },
    {
      "epoch": 0.747429144720341,
      "grad_norm": 0.38604459166526794,
      "learning_rate": 2e-05,
      "loss": 0.6342,
      "step": 8940
    },
    {
      "epoch": 0.7482651952177911,
      "grad_norm": 0.35728520154953003,
      "learning_rate": 2e-05,
      "loss": 0.6239,
      "step": 8950
    },
    {
      "epoch": 0.7491012457152412,
      "grad_norm": 0.37228596210479736,
      "learning_rate": 2e-05,
      "loss": 0.6231,
      "step": 8960
    },
    {
      "epoch": 0.7499372962126912,
      "grad_norm": 0.3389810025691986,
      "learning_rate": 2e-05,
      "loss": 0.6185,
      "step": 8970
    },
    {
      "epoch": 0.7507733467101413,
      "grad_norm": 0.3627842366695404,
      "learning_rate": 2e-05,
      "loss": 0.6029,
      "step": 8980
    },
    {
      "epoch": 0.7516093972075913,
      "grad_norm": 0.3359067142009735,
      "learning_rate": 2e-05,
      "loss": 0.6133,
      "step": 8990
    },
    {
      "epoch": 0.7524454477050414,
      "grad_norm": 0.37864169478416443,
      "learning_rate": 2e-05,
      "loss": 0.6121,
      "step": 9000
    },
    {
      "epoch": 0.7532814982024915,
      "grad_norm": 0.34371766448020935,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 9010
    },
    {
      "epoch": 0.7541175486999415,
      "grad_norm": 0.3812473714351654,
      "learning_rate": 2e-05,
      "loss": 0.6046,
      "step": 9020
    },
    {
      "epoch": 0.7549535991973915,
      "grad_norm": 0.3579816222190857,
      "learning_rate": 2e-05,
      "loss": 0.619,
      "step": 9030
    },
    {
      "epoch": 0.7557896496948415,
      "grad_norm": 0.3689136505126953,
      "learning_rate": 2e-05,
      "loss": 0.5896,
      "step": 9040
    },
    {
      "epoch": 0.7566257001922916,
      "grad_norm": 0.3529175817966461,
      "learning_rate": 2e-05,
      "loss": 0.6189,
      "step": 9050
    },
    {
      "epoch": 0.7574617506897416,
      "grad_norm": 0.36426064372062683,
      "learning_rate": 2e-05,
      "loss": 0.6449,
      "step": 9060
    },
    {
      "epoch": 0.7582978011871917,
      "grad_norm": 0.3498426675796509,
      "learning_rate": 2e-05,
      "loss": 0.667,
      "step": 9070
    },
    {
      "epoch": 0.7591338516846418,
      "grad_norm": 0.6616926193237305,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 9080
    },
    {
      "epoch": 0.7599699021820918,
      "grad_norm": 0.36647289991378784,
      "learning_rate": 2e-05,
      "loss": 0.6031,
      "step": 9090
    },
    {
      "epoch": 0.7608059526795419,
      "grad_norm": 0.3397458791732788,
      "learning_rate": 2e-05,
      "loss": 0.6092,
      "step": 9100
    },
    {
      "epoch": 0.7616420031769919,
      "grad_norm": 0.38301193714141846,
      "learning_rate": 2e-05,
      "loss": 0.6368,
      "step": 9110
    },
    {
      "epoch": 0.762478053674442,
      "grad_norm": 0.3724907636642456,
      "learning_rate": 2e-05,
      "loss": 0.6664,
      "step": 9120
    },
    {
      "epoch": 0.7633141041718919,
      "grad_norm": 0.36040452122688293,
      "learning_rate": 2e-05,
      "loss": 0.6431,
      "step": 9130
    },
    {
      "epoch": 0.764150154669342,
      "grad_norm": 0.4844878613948822,
      "learning_rate": 2e-05,
      "loss": 0.6439,
      "step": 9140
    },
    {
      "epoch": 0.764986205166792,
      "grad_norm": 0.36638975143432617,
      "learning_rate": 2e-05,
      "loss": 0.6266,
      "step": 9150
    },
    {
      "epoch": 0.7658222556642421,
      "grad_norm": 0.344577431678772,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 9160
    },
    {
      "epoch": 0.7666583061616922,
      "grad_norm": 0.39315271377563477,
      "learning_rate": 2e-05,
      "loss": 0.6433,
      "step": 9170
    },
    {
      "epoch": 0.7674943566591422,
      "grad_norm": 0.4230540990829468,
      "learning_rate": 2e-05,
      "loss": 0.5911,
      "step": 9180
    },
    {
      "epoch": 0.7683304071565923,
      "grad_norm": 0.32354632019996643,
      "learning_rate": 2e-05,
      "loss": 0.6264,
      "step": 9190
    },
    {
      "epoch": 0.7691664576540423,
      "grad_norm": 0.3388705849647522,
      "learning_rate": 2e-05,
      "loss": 0.6013,
      "step": 9200
    },
    {
      "epoch": 0.7700025081514924,
      "grad_norm": 0.34815776348114014,
      "learning_rate": 2e-05,
      "loss": 0.6383,
      "step": 9210
    },
    {
      "epoch": 0.7708385586489424,
      "grad_norm": 0.34967976808547974,
      "learning_rate": 2e-05,
      "loss": 0.6413,
      "step": 9220
    },
    {
      "epoch": 0.7716746091463924,
      "grad_norm": 0.36641407012939453,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 9230
    },
    {
      "epoch": 0.7725106596438425,
      "grad_norm": 0.3494429290294647,
      "learning_rate": 2e-05,
      "loss": 0.6189,
      "step": 9240
    },
    {
      "epoch": 0.7733467101412925,
      "grad_norm": 0.3621671497821808,
      "learning_rate": 2e-05,
      "loss": 0.6416,
      "step": 9250
    },
    {
      "epoch": 0.7741827606387426,
      "grad_norm": 0.3531359136104584,
      "learning_rate": 2e-05,
      "loss": 0.6346,
      "step": 9260
    },
    {
      "epoch": 0.7750188111361926,
      "grad_norm": 0.36367154121398926,
      "learning_rate": 2e-05,
      "loss": 0.6435,
      "step": 9270
    },
    {
      "epoch": 0.7758548616336427,
      "grad_norm": 0.3705047070980072,
      "learning_rate": 2e-05,
      "loss": 0.6298,
      "step": 9280
    },
    {
      "epoch": 0.7766909121310928,
      "grad_norm": 0.3508097231388092,
      "learning_rate": 2e-05,
      "loss": 0.607,
      "step": 9290
    },
    {
      "epoch": 0.7775269626285428,
      "grad_norm": 0.38197091221809387,
      "learning_rate": 2e-05,
      "loss": 0.6517,
      "step": 9300
    },
    {
      "epoch": 0.7783630131259928,
      "grad_norm": 0.6630399823188782,
      "learning_rate": 2e-05,
      "loss": 0.6135,
      "step": 9310
    },
    {
      "epoch": 0.7791990636234428,
      "grad_norm": 0.3597387373447418,
      "learning_rate": 2e-05,
      "loss": 0.6267,
      "step": 9320
    },
    {
      "epoch": 0.7800351141208929,
      "grad_norm": 0.37296104431152344,
      "learning_rate": 2e-05,
      "loss": 0.6507,
      "step": 9330
    },
    {
      "epoch": 0.7808711646183429,
      "grad_norm": 0.3904138207435608,
      "learning_rate": 2e-05,
      "loss": 0.6373,
      "step": 9340
    },
    {
      "epoch": 0.781707215115793,
      "grad_norm": 0.3662303388118744,
      "learning_rate": 2e-05,
      "loss": 0.6479,
      "step": 9350
    },
    {
      "epoch": 0.782543265613243,
      "grad_norm": 0.36050650477409363,
      "learning_rate": 2e-05,
      "loss": 0.6422,
      "step": 9360
    },
    {
      "epoch": 0.7833793161106931,
      "grad_norm": 0.34825584292411804,
      "learning_rate": 2e-05,
      "loss": 0.6,
      "step": 9370
    },
    {
      "epoch": 0.7842153666081432,
      "grad_norm": 0.3853682577610016,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 9380
    },
    {
      "epoch": 0.7850514171055932,
      "grad_norm": 0.3203522264957428,
      "learning_rate": 2e-05,
      "loss": 0.6247,
      "step": 9390
    },
    {
      "epoch": 0.7858874676030432,
      "grad_norm": 0.3575230538845062,
      "learning_rate": 2e-05,
      "loss": 0.5887,
      "step": 9400
    },
    {
      "epoch": 0.7867235181004932,
      "grad_norm": 0.3778257966041565,
      "learning_rate": 2e-05,
      "loss": 0.6583,
      "step": 9410
    },
    {
      "epoch": 0.7875595685979433,
      "grad_norm": 0.35051748156547546,
      "learning_rate": 2e-05,
      "loss": 0.6023,
      "step": 9420
    },
    {
      "epoch": 0.7883956190953934,
      "grad_norm": 0.36740484833717346,
      "learning_rate": 2e-05,
      "loss": 0.6079,
      "step": 9430
    },
    {
      "epoch": 0.7892316695928434,
      "grad_norm": 0.390663743019104,
      "learning_rate": 2e-05,
      "loss": 0.6366,
      "step": 9440
    },
    {
      "epoch": 0.7900677200902935,
      "grad_norm": 0.38424763083457947,
      "learning_rate": 2e-05,
      "loss": 0.6337,
      "step": 9450
    },
    {
      "epoch": 0.7909037705877435,
      "grad_norm": 0.5296206474304199,
      "learning_rate": 2e-05,
      "loss": 0.6091,
      "step": 9460
    },
    {
      "epoch": 0.7917398210851936,
      "grad_norm": 0.39744457602500916,
      "learning_rate": 2e-05,
      "loss": 0.6062,
      "step": 9470
    },
    {
      "epoch": 0.7925758715826436,
      "grad_norm": 0.3354249894618988,
      "learning_rate": 2e-05,
      "loss": 0.6188,
      "step": 9480
    },
    {
      "epoch": 0.7934119220800936,
      "grad_norm": 0.3593561053276062,
      "learning_rate": 2e-05,
      "loss": 0.6156,
      "step": 9490
    },
    {
      "epoch": 0.7942479725775436,
      "grad_norm": 0.3381028473377228,
      "learning_rate": 2e-05,
      "loss": 0.6153,
      "step": 9500
    },
    {
      "epoch": 0.7950840230749937,
      "grad_norm": 0.3624657988548279,
      "learning_rate": 2e-05,
      "loss": 0.6164,
      "step": 9510
    },
    {
      "epoch": 0.7959200735724438,
      "grad_norm": 0.42993080615997314,
      "learning_rate": 2e-05,
      "loss": 0.653,
      "step": 9520
    },
    {
      "epoch": 0.7967561240698938,
      "grad_norm": 0.36111199855804443,
      "learning_rate": 2e-05,
      "loss": 0.5918,
      "step": 9530
    },
    {
      "epoch": 0.7975921745673439,
      "grad_norm": 0.37998372316360474,
      "learning_rate": 2e-05,
      "loss": 0.5861,
      "step": 9540
    },
    {
      "epoch": 0.7984282250647939,
      "grad_norm": 0.4306682050228119,
      "learning_rate": 2e-05,
      "loss": 0.6329,
      "step": 9550
    },
    {
      "epoch": 0.799264275562244,
      "grad_norm": 0.6317639350891113,
      "learning_rate": 2e-05,
      "loss": 0.6347,
      "step": 9560
    },
    {
      "epoch": 0.8001003260596941,
      "grad_norm": 0.3609706163406372,
      "learning_rate": 2e-05,
      "loss": 0.633,
      "step": 9570
    },
    {
      "epoch": 0.800936376557144,
      "grad_norm": 0.3257679045200348,
      "learning_rate": 2e-05,
      "loss": 0.6232,
      "step": 9580
    },
    {
      "epoch": 0.8017724270545941,
      "grad_norm": 0.35587966442108154,
      "learning_rate": 2e-05,
      "loss": 0.6263,
      "step": 9590
    },
    {
      "epoch": 0.8026084775520441,
      "grad_norm": 0.409853994846344,
      "learning_rate": 2e-05,
      "loss": 0.6431,
      "step": 9600
    },
    {
      "epoch": 0.8034445280494942,
      "grad_norm": 0.3718413710594177,
      "learning_rate": 2e-05,
      "loss": 0.5955,
      "step": 9610
    },
    {
      "epoch": 0.8042805785469442,
      "grad_norm": 0.34535589814186096,
      "learning_rate": 2e-05,
      "loss": 0.6023,
      "step": 9620
    },
    {
      "epoch": 0.8051166290443943,
      "grad_norm": 0.3348459005355835,
      "learning_rate": 2e-05,
      "loss": 0.5886,
      "step": 9630
    },
    {
      "epoch": 0.8059526795418444,
      "grad_norm": 0.3702199161052704,
      "learning_rate": 2e-05,
      "loss": 0.6204,
      "step": 9640
    },
    {
      "epoch": 0.8067887300392944,
      "grad_norm": 0.3866741359233856,
      "learning_rate": 2e-05,
      "loss": 0.6251,
      "step": 9650
    },
    {
      "epoch": 0.8076247805367445,
      "grad_norm": 0.3488486111164093,
      "learning_rate": 2e-05,
      "loss": 0.6183,
      "step": 9660
    },
    {
      "epoch": 0.8084608310341944,
      "grad_norm": 0.3778485059738159,
      "learning_rate": 2e-05,
      "loss": 0.6368,
      "step": 9670
    },
    {
      "epoch": 0.8092968815316445,
      "grad_norm": 0.35843271017074585,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 9680
    },
    {
      "epoch": 0.8101329320290945,
      "grad_norm": 0.3854733407497406,
      "learning_rate": 2e-05,
      "loss": 0.6349,
      "step": 9690
    },
    {
      "epoch": 0.8109689825265446,
      "grad_norm": 0.3558465540409088,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 9700
    },
    {
      "epoch": 0.8118050330239946,
      "grad_norm": 0.348766028881073,
      "learning_rate": 2e-05,
      "loss": 0.6203,
      "step": 9710
    },
    {
      "epoch": 0.8126410835214447,
      "grad_norm": 0.3723942041397095,
      "learning_rate": 2e-05,
      "loss": 0.6255,
      "step": 9720
    },
    {
      "epoch": 0.8134771340188948,
      "grad_norm": 0.3663458526134491,
      "learning_rate": 2e-05,
      "loss": 0.5809,
      "step": 9730
    },
    {
      "epoch": 0.8143131845163448,
      "grad_norm": 0.34567496180534363,
      "learning_rate": 2e-05,
      "loss": 0.6695,
      "step": 9740
    },
    {
      "epoch": 0.8151492350137949,
      "grad_norm": 0.518709123134613,
      "learning_rate": 2e-05,
      "loss": 0.5724,
      "step": 9750
    },
    {
      "epoch": 0.8159852855112448,
      "grad_norm": 0.4326245188713074,
      "learning_rate": 2e-05,
      "loss": 0.5908,
      "step": 9760
    },
    {
      "epoch": 0.8168213360086949,
      "grad_norm": 0.37337279319763184,
      "learning_rate": 2e-05,
      "loss": 0.614,
      "step": 9770
    },
    {
      "epoch": 0.817657386506145,
      "grad_norm": 0.3601550757884979,
      "learning_rate": 2e-05,
      "loss": 0.6241,
      "step": 9780
    },
    {
      "epoch": 0.818493437003595,
      "grad_norm": 0.3535217046737671,
      "learning_rate": 2e-05,
      "loss": 0.598,
      "step": 9790
    },
    {
      "epoch": 0.8193294875010451,
      "grad_norm": 0.35097426176071167,
      "learning_rate": 2e-05,
      "loss": 0.6059,
      "step": 9800
    },
    {
      "epoch": 0.8201655379984951,
      "grad_norm": 0.36741960048675537,
      "learning_rate": 2e-05,
      "loss": 0.6158,
      "step": 9810
    },
    {
      "epoch": 0.8210015884959452,
      "grad_norm": 0.351399689912796,
      "learning_rate": 2e-05,
      "loss": 0.6411,
      "step": 9820
    },
    {
      "epoch": 0.8218376389933952,
      "grad_norm": 0.37758225202560425,
      "learning_rate": 2e-05,
      "loss": 0.6146,
      "step": 9830
    },
    {
      "epoch": 0.8226736894908453,
      "grad_norm": 0.35171523690223694,
      "learning_rate": 2e-05,
      "loss": 0.6096,
      "step": 9840
    },
    {
      "epoch": 0.8235097399882952,
      "grad_norm": 0.33236876130104065,
      "learning_rate": 2e-05,
      "loss": 0.5933,
      "step": 9850
    },
    {
      "epoch": 0.8243457904857453,
      "grad_norm": 0.36356234550476074,
      "learning_rate": 2e-05,
      "loss": 0.5985,
      "step": 9860
    },
    {
      "epoch": 0.8251818409831954,
      "grad_norm": 0.3597632348537445,
      "learning_rate": 2e-05,
      "loss": 0.6102,
      "step": 9870
    },
    {
      "epoch": 0.8260178914806454,
      "grad_norm": 0.3631537854671478,
      "learning_rate": 2e-05,
      "loss": 0.6016,
      "step": 9880
    },
    {
      "epoch": 0.8268539419780955,
      "grad_norm": 0.3433435261249542,
      "learning_rate": 2e-05,
      "loss": 0.6018,
      "step": 9890
    },
    {
      "epoch": 0.8276899924755455,
      "grad_norm": 0.38099363446235657,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 9900
    },
    {
      "epoch": 0.8285260429729956,
      "grad_norm": 0.37323102355003357,
      "learning_rate": 2e-05,
      "loss": 0.6187,
      "step": 9910
    },
    {
      "epoch": 0.8293620934704456,
      "grad_norm": 0.3748221695423126,
      "learning_rate": 2e-05,
      "loss": 0.5887,
      "step": 9920
    },
    {
      "epoch": 0.8301981439678957,
      "grad_norm": 0.3512956202030182,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 9930
    },
    {
      "epoch": 0.8310341944653457,
      "grad_norm": 0.3790229260921478,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 9940
    },
    {
      "epoch": 0.8318702449627957,
      "grad_norm": 0.348441481590271,
      "learning_rate": 2e-05,
      "loss": 0.6162,
      "step": 9950
    },
    {
      "epoch": 0.8327062954602458,
      "grad_norm": 0.3607241213321686,
      "learning_rate": 2e-05,
      "loss": 0.6238,
      "step": 9960
    },
    {
      "epoch": 0.8335423459576958,
      "grad_norm": 0.3722582459449768,
      "learning_rate": 2e-05,
      "loss": 0.6193,
      "step": 9970
    },
    {
      "epoch": 0.8343783964551459,
      "grad_norm": 0.38648438453674316,
      "learning_rate": 2e-05,
      "loss": 0.5946,
      "step": 9980
    },
    {
      "epoch": 0.835214446952596,
      "grad_norm": 0.351920485496521,
      "learning_rate": 2e-05,
      "loss": 0.5912,
      "step": 9990
    },
    {
      "epoch": 0.836050497450046,
      "grad_norm": 0.3653066158294678,
      "learning_rate": 2e-05,
      "loss": 0.5968,
      "step": 10000
    },
    {
      "epoch": 0.8368865479474961,
      "grad_norm": 0.3266824781894684,
      "learning_rate": 2e-05,
      "loss": 0.5985,
      "step": 10010
    },
    {
      "epoch": 0.8377225984449461,
      "grad_norm": 0.3750948905944824,
      "learning_rate": 2e-05,
      "loss": 0.5682,
      "step": 10020
    },
    {
      "epoch": 0.8385586489423961,
      "grad_norm": 0.35894718766212463,
      "learning_rate": 2e-05,
      "loss": 0.6259,
      "step": 10030
    },
    {
      "epoch": 0.8393946994398461,
      "grad_norm": 0.4020705819129944,
      "learning_rate": 2e-05,
      "loss": 0.6368,
      "step": 10040
    },
    {
      "epoch": 0.8402307499372962,
      "grad_norm": 0.39233365654945374,
      "learning_rate": 2e-05,
      "loss": 0.5765,
      "step": 10050
    },
    {
      "epoch": 0.8410668004347462,
      "grad_norm": 0.3552066385746002,
      "learning_rate": 2e-05,
      "loss": 0.6191,
      "step": 10060
    },
    {
      "epoch": 0.8419028509321963,
      "grad_norm": 0.39734166860580444,
      "learning_rate": 2e-05,
      "loss": 0.6242,
      "step": 10070
    },
    {
      "epoch": 0.8427389014296464,
      "grad_norm": 0.3577880263328552,
      "learning_rate": 2e-05,
      "loss": 0.625,
      "step": 10080
    },
    {
      "epoch": 0.8435749519270964,
      "grad_norm": 0.3468751013278961,
      "learning_rate": 2e-05,
      "loss": 0.6482,
      "step": 10090
    },
    {
      "epoch": 0.8444110024245465,
      "grad_norm": 0.3543683588504791,
      "learning_rate": 2e-05,
      "loss": 0.6022,
      "step": 10100
    },
    {
      "epoch": 0.8452470529219965,
      "grad_norm": 0.36623185873031616,
      "learning_rate": 2e-05,
      "loss": 0.6145,
      "step": 10110
    },
    {
      "epoch": 0.8460831034194465,
      "grad_norm": 0.3870335817337036,
      "learning_rate": 2e-05,
      "loss": 0.6268,
      "step": 10120
    },
    {
      "epoch": 0.8469191539168965,
      "grad_norm": 0.40605655312538147,
      "learning_rate": 2e-05,
      "loss": 0.6178,
      "step": 10130
    },
    {
      "epoch": 0.8477552044143466,
      "grad_norm": 0.3315054178237915,
      "learning_rate": 2e-05,
      "loss": 0.6234,
      "step": 10140
    },
    {
      "epoch": 0.8485912549117967,
      "grad_norm": 0.38885757327079773,
      "learning_rate": 2e-05,
      "loss": 0.6415,
      "step": 10150
    },
    {
      "epoch": 0.8494273054092467,
      "grad_norm": 0.3674159049987793,
      "learning_rate": 2e-05,
      "loss": 0.6202,
      "step": 10160
    },
    {
      "epoch": 0.8502633559066968,
      "grad_norm": 0.35860976576805115,
      "learning_rate": 2e-05,
      "loss": 0.647,
      "step": 10170
    },
    {
      "epoch": 0.8510994064041468,
      "grad_norm": 0.36236777901649475,
      "learning_rate": 1.999957358086795e-05,
      "loss": 0.6507,
      "step": 10180
    },
    {
      "epoch": 0.8519354569015969,
      "grad_norm": 0.3825151026248932,
      "learning_rate": 1.9991993813020483e-05,
      "loss": 0.6449,
      "step": 10190
    },
    {
      "epoch": 0.852771507399047,
      "grad_norm": 0.37441548705101013,
      "learning_rate": 1.9974946373028628e-05,
      "loss": 0.6238,
      "step": 10200
    },
    {
      "epoch": 0.8536075578964969,
      "grad_norm": 0.38678252696990967,
      "learning_rate": 1.994844749501764e-05,
      "loss": 0.6098,
      "step": 10210
    },
    {
      "epoch": 0.854443608393947,
      "grad_norm": 0.32180315256118774,
      "learning_rate": 1.9912522413632988e-05,
      "loss": 0.6202,
      "step": 10220
    },
    {
      "epoch": 0.855279658891397,
      "grad_norm": 0.36865687370300293,
      "learning_rate": 1.9867205340009656e-05,
      "loss": 0.6119,
      "step": 10230
    },
    {
      "epoch": 0.8561157093888471,
      "grad_norm": 0.38033464550971985,
      "learning_rate": 1.9812539429193164e-05,
      "loss": 0.631,
      "step": 10240
    },
    {
      "epoch": 0.8569517598862971,
      "grad_norm": 0.3575597107410431,
      "learning_rate": 1.9748576739043423e-05,
      "loss": 0.5869,
      "step": 10250
    },
    {
      "epoch": 0.8577878103837472,
      "grad_norm": 0.394417405128479,
      "learning_rate": 1.967537818066047e-05,
      "loss": 0.6089,
      "step": 10260
    },
    {
      "epoch": 0.8586238608811972,
      "grad_norm": 0.3664516806602478,
      "learning_rate": 1.959301346037941e-05,
      "loss": 0.6175,
      "step": 10270
    },
    {
      "epoch": 0.8594599113786473,
      "grad_norm": 0.3721388280391693,
      "learning_rate": 1.9501561013389646e-05,
      "loss": 0.6279,
      "step": 10280
    },
    {
      "epoch": 0.8602959618760974,
      "grad_norm": 0.3725232183933258,
      "learning_rate": 1.9401107929041795e-05,
      "loss": 0.6819,
      "step": 10290
    },
    {
      "epoch": 0.8611320123735474,
      "grad_norm": 0.36451494693756104,
      "learning_rate": 1.9291749867913225e-05,
      "loss": 0.6168,
      "step": 10300
    },
    {
      "epoch": 0.8619680628709974,
      "grad_norm": 0.32133838534355164,
      "learning_rate": 1.91735909707114e-05,
      "loss": 0.6547,
      "step": 10310
    },
    {
      "epoch": 0.8628041133684474,
      "grad_norm": 0.4033767580986023,
      "learning_rate": 1.9046743759101545e-05,
      "loss": 0.591,
      "step": 10320
    },
    {
      "epoch": 0.8636401638658975,
      "grad_norm": 0.3832809031009674,
      "learning_rate": 1.8911329028553324e-05,
      "loss": 0.6233,
      "step": 10330
    },
    {
      "epoch": 0.8644762143633475,
      "grad_norm": 0.3499000668525696,
      "learning_rate": 1.876747573330835e-05,
      "loss": 0.6356,
      "step": 10340
    },
    {
      "epoch": 0.8653122648607976,
      "grad_norm": 0.3311673402786255,
      "learning_rate": 1.8615320863578244e-05,
      "loss": 0.6195,
      "step": 10350
    },
    {
      "epoch": 0.8661483153582477,
      "grad_norm": 0.3482300639152527,
      "learning_rate": 1.8455009315090036e-05,
      "loss": 0.6053,
      "step": 10360
    },
    {
      "epoch": 0.8669843658556977,
      "grad_norm": 0.438000351190567,
      "learning_rate": 1.8286693751103294e-05,
      "loss": 0.6403,
      "step": 10370
    },
    {
      "epoch": 0.8678204163531478,
      "grad_norm": 0.3795582950115204,
      "learning_rate": 1.811053445703024e-05,
      "loss": 0.6455,
      "step": 10380
    },
    {
      "epoch": 0.8686564668505978,
      "grad_norm": 0.359438955783844,
      "learning_rate": 1.7926699187797375e-05,
      "loss": 0.6234,
      "step": 10390
    },
    {
      "epoch": 0.8694925173480478,
      "grad_norm": 0.37603017687797546,
      "learning_rate": 1.7735363008093995e-05,
      "loss": 0.5977,
      "step": 10400
    },
    {
      "epoch": 0.8703285678454978,
      "grad_norm": 0.35322943329811096,
      "learning_rate": 1.7536708125659577e-05,
      "loss": 0.6265,
      "step": 10410
    },
    {
      "epoch": 0.8711646183429479,
      "grad_norm": 0.39322182536125183,
      "learning_rate": 1.733092371776905e-05,
      "loss": 0.6332,
      "step": 10420
    },
    {
      "epoch": 0.872000668840398,
      "grad_norm": 0.3669787645339966,
      "learning_rate": 1.7118205751080915e-05,
      "loss": 0.6119,
      "step": 10430
    },
    {
      "epoch": 0.872836719337848,
      "grad_norm": 0.34573134779930115,
      "learning_rate": 1.6898756795019935e-05,
      "loss": 0.6347,
      "step": 10440
    },
    {
      "epoch": 0.8736727698352981,
      "grad_norm": 0.40765970945358276,
      "learning_rate": 1.667278582887211e-05,
      "loss": 0.619,
      "step": 10450
    },
    {
      "epoch": 0.8745088203327481,
      "grad_norm": 0.37119197845458984,
      "learning_rate": 1.6440508042775545e-05,
      "loss": 0.6112,
      "step": 10460
    },
    {
      "epoch": 0.8753448708301982,
      "grad_norm": 0.39677056670188904,
      "learning_rate": 1.6202144632796796e-05,
      "loss": 0.6241,
      "step": 10470
    },
    {
      "epoch": 0.8761809213276482,
      "grad_norm": 0.3530145585536957,
      "learning_rate": 1.595792259028783e-05,
      "loss": 0.6377,
      "step": 10480
    },
    {
      "epoch": 0.8770169718250982,
      "grad_norm": 0.3644874393939972,
      "learning_rate": 1.570807448572419e-05,
      "loss": 0.6047,
      "step": 10490
    },
    {
      "epoch": 0.8778530223225482,
      "grad_norm": 0.7080325484275818,
      "learning_rate": 1.5452838247230197e-05,
      "loss": 0.5916,
      "step": 10500
    },
    {
      "epoch": 0.8786890728199983,
      "grad_norm": 0.376627117395401,
      "learning_rate": 1.5192456934002106e-05,
      "loss": 0.6039,
      "step": 10510
    },
    {
      "epoch": 0.8795251233174484,
      "grad_norm": 0.3434585928916931,
      "learning_rate": 1.492717850484502e-05,
      "loss": 0.6142,
      "step": 10520
    },
    {
      "epoch": 0.8803611738148984,
      "grad_norm": 0.37058815360069275,
      "learning_rate": 1.4657255582043904e-05,
      "loss": 0.6049,
      "step": 10530
    },
    {
      "epoch": 0.8811972243123485,
      "grad_norm": 0.34882307052612305,
      "learning_rate": 1.4382945210793663e-05,
      "loss": 0.6261,
      "step": 10540
    },
    {
      "epoch": 0.8820332748097985,
      "grad_norm": 0.4276929497718811,
      "learning_rate": 1.4104508614417267e-05,
      "loss": 0.6298,
      "step": 10550
    },
    {
      "epoch": 0.8828693253072486,
      "grad_norm": 0.3642962574958801,
      "learning_rate": 1.382221094560512e-05,
      "loss": 0.6003,
      "step": 10560
    },
    {
      "epoch": 0.8837053758046987,
      "grad_norm": 0.38834086060523987,
      "learning_rate": 1.3536321033912506e-05,
      "loss": 0.6229,
      "step": 10570
    },
    {
      "epoch": 0.8845414263021486,
      "grad_norm": 0.3640705943107605,
      "learning_rate": 1.3247111129755609e-05,
      "loss": 0.5761,
      "step": 10580
    },
    {
      "epoch": 0.8853774767995987,
      "grad_norm": 0.3399708867073059,
      "learning_rate": 1.2954856645149868e-05,
      "loss": 0.5688,
      "step": 10590
    },
    {
      "epoch": 0.8862135272970487,
      "grad_norm": 0.3548114597797394,
      "learning_rate": 1.2659835891437553e-05,
      "loss": 0.603,
      "step": 10600
    },
    {
      "epoch": 0.8870495777944988,
      "grad_norm": 0.35191482305526733,
      "learning_rate": 1.236232981425436e-05,
      "loss": 0.5997,
      "step": 10610
    },
    {
      "epoch": 0.8878856282919488,
      "grad_norm": 0.4111185371875763,
      "learning_rate": 1.2062621725987402e-05,
      "loss": 0.6292,
      "step": 10620
    },
    {
      "epoch": 0.8887216787893989,
      "grad_norm": 0.7909184098243713,
      "learning_rate": 1.1760997035979318e-05,
      "loss": 0.598,
      "step": 10630
    },
    {
      "epoch": 0.889557729286849,
      "grad_norm": 0.4025023579597473,
      "learning_rate": 1.1457742978735514e-05,
      "loss": 0.5966,
      "step": 10640
    },
    {
      "epoch": 0.890393779784299,
      "grad_norm": 0.35843297839164734,
      "learning_rate": 1.1153148340393298e-05,
      "loss": 0.6118,
      "step": 10650
    },
    {
      "epoch": 0.8912298302817491,
      "grad_norm": 0.3677820563316345,
      "learning_rate": 1.0847503183713424e-05,
      "loss": 0.6077,
      "step": 10660
    },
    {
      "epoch": 0.892065880779199,
      "grad_norm": 0.892987072467804,
      "learning_rate": 1.0541098571855896e-05,
      "loss": 0.6201,
      "step": 10670
    },
    {
      "epoch": 0.8929019312766491,
      "grad_norm": 0.36586469411849976,
      "learning_rate": 1.0234226291203123e-05,
      "loss": 0.6157,
      "step": 10680
    },
    {
      "epoch": 0.8937379817740991,
      "grad_norm": 0.3509763181209564,
      "learning_rate": 9.927178573494333e-06,
      "loss": 0.5763,
      "step": 10690
    },
    {
      "epoch": 0.8945740322715492,
      "grad_norm": 0.3472200930118561,
      "learning_rate": 9.620247817535902e-06,
      "loss": 0.6036,
      "step": 10700
    },
    {
      "epoch": 0.8954100827689992,
      "grad_norm": 0.37201228737831116,
      "learning_rate": 9.313726310752581e-06,
      "loss": 0.6053,
      "step": 10710
    },
    {
      "epoch": 0.8962461332664493,
      "grad_norm": 0.3612881600856781,
      "learning_rate": 9.00790595084477e-06,
      "loss": 0.5874,
      "step": 10720
    },
    {
      "epoch": 0.8970821837638994,
      "grad_norm": 0.4058479368686676,
      "learning_rate": 8.703077967816944e-06,
      "loss": 0.5969,
      "step": 10730
    },
    {
      "epoch": 0.8979182342613494,
      "grad_norm": 0.46108946204185486,
      "learning_rate": 8.39953264664192e-06,
      "loss": 0.6078,
      "step": 10740
    },
    {
      "epoch": 0.8987542847587995,
      "grad_norm": 0.3479207754135132,
      "learning_rate": 8.097559050825058e-06,
      "loss": 0.5851,
      "step": 10750
    },
    {
      "epoch": 0.8995903352562494,
      "grad_norm": 0.3858983516693115,
      "learning_rate": 7.79744474713166e-06,
      "loss": 0.6054,
      "step": 10760
    },
    {
      "epoch": 0.9004263857536995,
      "grad_norm": 0.3158455193042755,
      "learning_rate": 7.499475531739696e-06,
      "loss": 0.5939,
      "step": 10770
    },
    {
      "epoch": 0.9012624362511495,
      "grad_norm": 0.43052423000335693,
      "learning_rate": 7.203935158078662e-06,
      "loss": 0.6223,
      "step": 10780
    },
    {
      "epoch": 0.9020984867485996,
      "grad_norm": 0.3475760221481323,
      "learning_rate": 6.911105066613709e-06,
      "loss": 0.6075,
      "step": 10790
    },
    {
      "epoch": 0.9029345372460497,
      "grad_norm": 0.3658798038959503,
      "learning_rate": 6.621264116832394e-06,
      "loss": 0.6168,
      "step": 10800
    },
    {
      "epoch": 0.9037705877434997,
      "grad_norm": 0.396554172039032,
      "learning_rate": 6.334688321689274e-06,
      "loss": 0.5786,
      "step": 10810
    },
    {
      "epoch": 0.9046066382409498,
      "grad_norm": 0.35564687848091125,
      "learning_rate": 6.051650584761237e-06,
      "loss": 0.5502,
      "step": 10820
    },
    {
      "epoch": 0.9054426887383998,
      "grad_norm": 0.35927101969718933,
      "learning_rate": 5.772420440363841e-06,
      "loss": 0.5748,
      "step": 10830
    },
    {
      "epoch": 0.9062787392358499,
      "grad_norm": 0.38214701414108276,
      "learning_rate": 5.497263796876207e-06,
      "loss": 0.6163,
      "step": 10840
    },
    {
      "epoch": 0.9071147897332998,
      "grad_norm": 0.3848903179168701,
      "learning_rate": 5.226442683518813e-06,
      "loss": 0.6087,
      "step": 10850
    },
    {
      "epoch": 0.9079508402307499,
      "grad_norm": 0.3296385705471039,
      "learning_rate": 4.960215000825432e-06,
      "loss": 0.5999,
      "step": 10860
    },
    {
      "epoch": 0.9087868907282,
      "grad_norm": 0.36416271328926086,
      "learning_rate": 4.698834275046711e-06,
      "loss": 0.5849,
      "step": 10870
    },
    {
      "epoch": 0.90962294122565,
      "grad_norm": 0.3372238576412201,
      "learning_rate": 4.442549416719423e-06,
      "loss": 0.5931,
      "step": 10880
    },
    {
      "epoch": 0.9104589917231001,
      "grad_norm": 0.6357288360595703,
      "learning_rate": 4.1916044836311565e-06,
      "loss": 0.5961,
      "step": 10890
    },
    {
      "epoch": 0.9112950422205501,
      "grad_norm": 0.35085994005203247,
      "learning_rate": 3.94623844840626e-06,
      "loss": 0.6,
      "step": 10900
    },
    {
      "epoch": 0.9121310927180002,
      "grad_norm": 0.31522876024246216,
      "learning_rate": 3.7066849709343467e-06,
      "loss": 0.5849,
      "step": 10910
    },
    {
      "epoch": 0.9129671432154502,
      "grad_norm": 0.37488582730293274,
      "learning_rate": 3.473172175858059e-06,
      "loss": 0.605,
      "step": 10920
    },
    {
      "epoch": 0.9138031937129003,
      "grad_norm": 0.35212236642837524,
      "learning_rate": 3.2459224353320046e-06,
      "loss": 0.5832,
      "step": 10930
    },
    {
      "epoch": 0.9146392442103503,
      "grad_norm": 0.3319855034351349,
      "learning_rate": 3.025152157259735e-06,
      "loss": 0.5735,
      "step": 10940
    },
    {
      "epoch": 0.9154752947078003,
      "grad_norm": 0.3578665256500244,
      "learning_rate": 2.811071579210425e-06,
      "loss": 0.5993,
      "step": 10950
    },
    {
      "epoch": 0.9163113452052504,
      "grad_norm": 0.35180723667144775,
      "learning_rate": 2.603884568211513e-06,
      "loss": 0.6115,
      "step": 10960
    },
    {
      "epoch": 0.9171473957027004,
      "grad_norm": 0.3415590822696686,
      "learning_rate": 2.4037884266079434e-06,
      "loss": 0.5965,
      "step": 10970
    },
    {
      "epoch": 0.9179834462001505,
      "grad_norm": 0.34383273124694824,
      "learning_rate": 2.2109737041728955e-06,
      "loss": 0.6105,
      "step": 10980
    },
    {
      "epoch": 0.9188194966976005,
      "grad_norm": 0.37161368131637573,
      "learning_rate": 2.0256240166489454e-06,
      "loss": 0.6029,
      "step": 10990
    },
    {
      "epoch": 0.9196555471950506,
      "grad_norm": 0.37726929783821106,
      "learning_rate": 1.847915870892413e-06,
      "loss": 0.6072,
      "step": 11000
    },
    {
      "epoch": 0.9204915976925007,
      "grad_norm": 0.36305373907089233,
      "learning_rate": 1.6780184967874715e-06,
      "loss": 0.6049,
      "step": 11010
    },
    {
      "epoch": 0.9213276481899507,
      "grad_norm": 0.3347192108631134,
      "learning_rate": 1.5160936860900294e-06,
      "loss": 0.6245,
      "step": 11020
    },
    {
      "epoch": 0.9221636986874007,
      "grad_norm": 0.35888564586639404,
      "learning_rate": 1.3622956383548762e-06,
      "loss": 0.6292,
      "step": 11030
    },
    {
      "epoch": 0.9229997491848507,
      "grad_norm": 0.33701983094215393,
      "learning_rate": 1.216770814092834e-06,
      "loss": 0.5771,
      "step": 11040
    },
    {
      "epoch": 0.9238357996823008,
      "grad_norm": 0.36267757415771484,
      "learning_rate": 1.0796577952977013e-06,
      "loss": 0.5634,
      "step": 11050
    },
    {
      "epoch": 0.9246718501797508,
      "grad_norm": 0.3506619334220886,
      "learning_rate": 9.510871534758705e-07,
      "loss": 0.6342,
      "step": 11060
    },
    {
      "epoch": 0.9255079006772009,
      "grad_norm": 0.38692036271095276,
      "learning_rate": 8.311813253042334e-07,
      "loss": 0.6264,
      "step": 11070
    },
    {
      "epoch": 0.926343951174651,
      "grad_norm": 0.35754019021987915,
      "learning_rate": 7.200544960348308e-07,
      "loss": 0.6094,
      "step": 11080
    },
    {
      "epoch": 0.927180001672101,
      "grad_norm": 0.35391393303871155,
      "learning_rate": 6.178124907572545e-07,
      "loss": 0.5929,
      "step": 11090
    },
    {
      "epoch": 0.9280160521695511,
      "grad_norm": 0.35461267828941345,
      "learning_rate": 5.245526736223528e-07,
      "loss": 0.5709,
      "step": 11100
    },
    {
      "epoch": 0.9288521026670011,
      "grad_norm": 0.3442363739013672,
      "learning_rate": 4.403638551232205e-07,
      "loss": 0.5823,
      "step": 11110
    },
    {
      "epoch": 0.9296881531644511,
      "grad_norm": 0.3770408630371094,
      "learning_rate": 3.6532620752175616e-07,
      "loss": 0.5861,
      "step": 11120
    },
    {
      "epoch": 0.9305242036619011,
      "grad_norm": 0.32698795199394226,
      "learning_rate": 2.9951118850133247e-07,
      "loss": 0.5853,
      "step": 11130
    },
    {
      "epoch": 0.9313602541593512,
      "grad_norm": 0.36496493220329285,
      "learning_rate": 2.429814731182774e-07,
      "loss": 0.6138,
      "step": 11140
    },
    {
      "epoch": 0.9321963046568013,
      "grad_norm": 0.36672013998031616,
      "learning_rate": 1.957908941169813e-07,
      "loss": 0.5562,
      "step": 11150
    },
    {
      "epoch": 0.9330323551542513,
      "grad_norm": 0.3472349941730499,
      "learning_rate": 1.5798439066545254e-07,
      "loss": 0.6069,
      "step": 11160
    },
    {
      "epoch": 0.9338684056517014,
      "grad_norm": 0.34626901149749756,
      "learning_rate": 1.2959796556014471e-07,
      "loss": 0.5976,
      "step": 11170
    },
    {
      "epoch": 0.9347044561491514,
      "grad_norm": 0.3652532398700714,
      "learning_rate": 1.1065865094081884e-07,
      "loss": 0.5878,
      "step": 11180
    },
    {
      "epoch": 0.9355405066466015,
      "grad_norm": 0.3446880280971527,
      "learning_rate": 1.011844825480734e-07,
      "loss": 0.5792,
      "step": 11190
    },
    {
      "epoch": 0.9358749268455815,
      "step": 11194,
      "total_flos": 4.547514943786923e+19,
      "train_loss": 0.0033432948966313413,
      "train_runtime": 611.3742,
      "train_samples_per_second": 219.715,
      "train_steps_per_second": 18.31
    }
  ],
  "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
}