{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9192740412252607,
  "eval_steps": 500,
  "global_step": 11194,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008212203334154554,
      "grad_norm": 8.587085723876953,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 2.3386,
      "step": 10
    },
    {
      "epoch": 0.0016424406668309108,
      "grad_norm": 4.4222412109375,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 2.1657,
      "step": 20
    },
    {
      "epoch": 0.002463661000246366,
      "grad_norm": 3.2723898887634277,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.8691,
      "step": 30
    },
    {
      "epoch": 0.0032848813336618216,
      "grad_norm": 1.605501413345337,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.4938,
      "step": 40
    },
    {
      "epoch": 0.004106101667077277,
      "grad_norm": 1.1179364919662476,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.2218,
      "step": 50
    },
    {
      "epoch": 0.004927322000492732,
      "grad_norm": 1.5302199125289917,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.0617,
      "step": 60
    },
    {
      "epoch": 0.005748542333908187,
      "grad_norm": 0.6661491990089417,
      "learning_rate": 7.814044943820226e-06,
      "loss": 1.0192,
      "step": 70
    },
    {
      "epoch": 0.006569762667323643,
      "grad_norm": 0.6883566975593567,
      "learning_rate": 8.932022471910112e-06,
      "loss": 1.0114,
      "step": 80
    },
    {
      "epoch": 0.007390983000739098,
      "grad_norm": 0.762592613697052,
      "learning_rate": 1.005e-05,
      "loss": 0.9338,
      "step": 90
    },
    {
      "epoch": 0.008212203334154554,
      "grad_norm": 0.6071498394012451,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.913,
      "step": 100
    },
    {
      "epoch": 0.009033423667570008,
      "grad_norm": 0.6040300130844116,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.8946,
      "step": 110
    },
    {
      "epoch": 0.009854644000985464,
      "grad_norm": 0.6161038279533386,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.9095,
      "step": 120
    },
    {
      "epoch": 0.01067586433440092,
      "grad_norm": 0.6221404671669006,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.9015,
      "step": 130
    },
    {
      "epoch": 0.011497084667816375,
      "grad_norm": 0.6306547522544861,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.9068,
      "step": 140
    },
    {
      "epoch": 0.01231830500123183,
      "grad_norm": 0.5591278672218323,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.9018,
      "step": 150
    },
    {
      "epoch": 0.013139525334647286,
      "grad_norm": 0.5367937684059143,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.7905,
      "step": 160
    },
    {
      "epoch": 0.01396074566806274,
      "grad_norm": 0.5446851253509521,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.8516,
      "step": 170
    },
    {
      "epoch": 0.014781966001478197,
      "grad_norm": 0.6117768287658691,
      "learning_rate": 2e-05,
      "loss": 0.8769,
      "step": 180
    },
    {
      "epoch": 0.015603186334893653,
      "grad_norm": 0.5499733090400696,
      "learning_rate": 2e-05,
      "loss": 0.8929,
      "step": 190
    },
    {
      "epoch": 0.01642440666830911,
      "grad_norm": 0.5772490501403809,
      "learning_rate": 2e-05,
      "loss": 0.8402,
      "step": 200
    },
    {
      "epoch": 0.017245627001724564,
      "grad_norm": 0.5514219999313354,
      "learning_rate": 2e-05,
      "loss": 0.8938,
      "step": 210
    },
    {
      "epoch": 0.018066847335140017,
      "grad_norm": 0.5629882216453552,
      "learning_rate": 2e-05,
      "loss": 0.8873,
      "step": 220
    },
    {
      "epoch": 0.018888067668555473,
      "grad_norm": 0.5477422475814819,
      "learning_rate": 2e-05,
      "loss": 0.8596,
      "step": 230
    },
    {
      "epoch": 0.01970928800197093,
      "grad_norm": 0.5064891576766968,
      "learning_rate": 2e-05,
      "loss": 0.8442,
      "step": 240
    },
    {
      "epoch": 0.020530508335386385,
      "grad_norm": 1.1094902753829956,
      "learning_rate": 2e-05,
      "loss": 0.8506,
      "step": 250
    },
    {
      "epoch": 0.02135172866880184,
      "grad_norm": 0.5511003732681274,
      "learning_rate": 2e-05,
      "loss": 0.8865,
      "step": 260
    },
    {
      "epoch": 0.022172949002217297,
      "grad_norm": 0.5206192135810852,
      "learning_rate": 2e-05,
      "loss": 0.8665,
      "step": 270
    },
    {
      "epoch": 0.02299416933563275,
      "grad_norm": 0.515914797782898,
      "learning_rate": 2e-05,
      "loss": 0.8456,
      "step": 280
    },
    {
      "epoch": 0.023815389669048205,
      "grad_norm": 0.5267996788024902,
      "learning_rate": 2e-05,
      "loss": 0.8356,
      "step": 290
    },
    {
      "epoch": 0.02463661000246366,
      "grad_norm": 0.5458669066429138,
      "learning_rate": 2e-05,
      "loss": 0.836,
      "step": 300
    },
    {
      "epoch": 0.025457830335879117,
      "grad_norm": 0.525846004486084,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 310
    },
    {
      "epoch": 0.026279050669294573,
      "grad_norm": 0.5086902379989624,
      "learning_rate": 2e-05,
      "loss": 0.8287,
      "step": 320
    },
    {
      "epoch": 0.02710027100271003,
      "grad_norm": 0.4971238374710083,
      "learning_rate": 2e-05,
      "loss": 0.8497,
      "step": 330
    },
    {
      "epoch": 0.02792149133612548,
      "grad_norm": 0.5196722745895386,
      "learning_rate": 2e-05,
      "loss": 0.8592,
      "step": 340
    },
    {
      "epoch": 0.028742711669540937,
      "grad_norm": 0.5308536291122437,
      "learning_rate": 2e-05,
      "loss": 0.8503,
      "step": 350
    },
    {
      "epoch": 0.029563932002956393,
      "grad_norm": 0.516264796257019,
      "learning_rate": 2e-05,
      "loss": 0.8753,
      "step": 360
    },
    {
      "epoch": 0.03038515233637185,
      "grad_norm": 0.5216997861862183,
      "learning_rate": 2e-05,
      "loss": 0.8095,
      "step": 370
    },
    {
      "epoch": 0.031206372669787305,
      "grad_norm": 0.5078901052474976,
      "learning_rate": 2e-05,
      "loss": 0.8461,
      "step": 380
    },
    {
      "epoch": 0.03202759300320276,
      "grad_norm": 0.48943689465522766,
      "learning_rate": 2e-05,
      "loss": 0.8472,
      "step": 390
    },
    {
      "epoch": 0.03284881333661822,
      "grad_norm": 0.4968130886554718,
      "learning_rate": 2e-05,
      "loss": 0.8429,
      "step": 400
    },
    {
      "epoch": 0.03367003367003367,
      "grad_norm": 0.5077145099639893,
      "learning_rate": 2e-05,
      "loss": 0.8115,
      "step": 410
    },
    {
      "epoch": 0.03449125400344913,
      "grad_norm": 0.47043702006340027,
      "learning_rate": 2e-05,
      "loss": 0.8497,
      "step": 420
    },
    {
      "epoch": 0.03531247433686458,
      "grad_norm": 0.8502866625785828,
      "learning_rate": 2e-05,
      "loss": 0.8208,
      "step": 430
    },
    {
      "epoch": 0.036133694670280034,
      "grad_norm": 0.5105547308921814,
      "learning_rate": 2e-05,
      "loss": 0.8235,
      "step": 440
    },
    {
      "epoch": 0.03695491500369549,
      "grad_norm": 0.5259561538696289,
      "learning_rate": 2e-05,
      "loss": 0.8505,
      "step": 450
    },
    {
      "epoch": 0.037776135337110946,
      "grad_norm": 0.5170012712478638,
      "learning_rate": 2e-05,
      "loss": 0.8135,
      "step": 460
    },
    {
      "epoch": 0.038597355670526405,
      "grad_norm": 0.47574323415756226,
      "learning_rate": 2e-05,
      "loss": 0.8399,
      "step": 470
    },
    {
      "epoch": 0.03941857600394186,
      "grad_norm": 0.5556318163871765,
      "learning_rate": 2e-05,
      "loss": 0.7829,
      "step": 480
    },
    {
      "epoch": 0.04023979633735731,
      "grad_norm": 0.521508514881134,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 490
    },
    {
      "epoch": 0.04106101667077277,
      "grad_norm": 0.5001388788223267,
      "learning_rate": 2e-05,
      "loss": 0.8267,
      "step": 500
    },
    {
      "epoch": 0.04188223700418822,
      "grad_norm": 0.4890121817588806,
      "learning_rate": 2e-05,
      "loss": 0.8037,
      "step": 510
    },
    {
      "epoch": 0.04270345733760368,
      "grad_norm": 0.5156166553497314,
      "learning_rate": 2e-05,
      "loss": 0.816,
      "step": 520
    },
    {
      "epoch": 0.043524677671019134,
      "grad_norm": 0.5050928592681885,
      "learning_rate": 2e-05,
      "loss": 0.847,
      "step": 530
    },
    {
      "epoch": 0.04434589800443459,
      "grad_norm": 0.4809299111366272,
      "learning_rate": 2e-05,
      "loss": 0.8349,
      "step": 540
    },
    {
      "epoch": 0.045167118337850046,
      "grad_norm": 0.4897677004337311,
      "learning_rate": 2e-05,
      "loss": 0.8261,
      "step": 550
    },
    {
      "epoch": 0.0459883386712655,
      "grad_norm": 0.4984337389469147,
      "learning_rate": 2e-05,
      "loss": 0.8612,
      "step": 560
    },
    {
      "epoch": 0.04680955900468096,
      "grad_norm": 0.5070453882217407,
      "learning_rate": 2e-05,
      "loss": 0.8536,
      "step": 570
    },
    {
      "epoch": 0.04763077933809641,
      "grad_norm": 0.45646464824676514,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 580
    },
    {
      "epoch": 0.04845199967151187,
      "grad_norm": 0.4721323549747467,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 590
    },
    {
      "epoch": 0.04927322000492732,
      "grad_norm": 0.4687247574329376,
      "learning_rate": 2e-05,
      "loss": 0.8288,
      "step": 600
    },
    {
      "epoch": 0.050094440338342774,
      "grad_norm": 0.4447396993637085,
      "learning_rate": 2e-05,
      "loss": 0.8308,
      "step": 610
    },
    {
      "epoch": 0.050915660671758234,
      "grad_norm": 0.49292364716529846,
      "learning_rate": 2e-05,
      "loss": 0.8139,
      "step": 620
    },
    {
      "epoch": 0.051736881005173686,
      "grad_norm": 0.46408599615097046,
      "learning_rate": 2e-05,
      "loss": 0.8092,
      "step": 630
    },
    {
      "epoch": 0.052558101338589146,
      "grad_norm": 0.5292156338691711,
      "learning_rate": 2e-05,
      "loss": 0.8603,
      "step": 640
    },
    {
      "epoch": 0.0533793216720046,
      "grad_norm": 0.46266090869903564,
      "learning_rate": 2e-05,
      "loss": 0.8071,
      "step": 650
    },
    {
      "epoch": 0.05420054200542006,
      "grad_norm": 0.5302640795707703,
      "learning_rate": 2e-05,
      "loss": 0.8231,
      "step": 660
    },
    {
      "epoch": 0.05502176233883551,
      "grad_norm": 0.4566280245780945,
      "learning_rate": 2e-05,
      "loss": 0.8006,
      "step": 670
    },
    {
      "epoch": 0.05584298267225096,
      "grad_norm": 0.4889671206474304,
      "learning_rate": 2e-05,
      "loss": 0.8209,
      "step": 680
    },
    {
      "epoch": 0.05666420300566642,
      "grad_norm": 0.4641304910182953,
      "learning_rate": 2e-05,
      "loss": 0.8092,
      "step": 690
    },
    {
      "epoch": 0.057485423339081874,
      "grad_norm": 0.4148608446121216,
      "learning_rate": 2e-05,
      "loss": 0.8292,
      "step": 700
    },
    {
      "epoch": 0.058306643672497334,
      "grad_norm": 0.4525546133518219,
      "learning_rate": 2e-05,
      "loss": 0.823,
      "step": 710
    },
    {
      "epoch": 0.059127864005912786,
      "grad_norm": 0.45530539751052856,
      "learning_rate": 2e-05,
      "loss": 0.8285,
      "step": 720
    },
    {
      "epoch": 0.05994908433932824,
      "grad_norm": 0.46801939606666565,
      "learning_rate": 2e-05,
      "loss": 0.8136,
      "step": 730
    },
    {
      "epoch": 0.0607703046727437,
      "grad_norm": 0.47091159224510193,
      "learning_rate": 2e-05,
      "loss": 0.8148,
      "step": 740
    },
    {
      "epoch": 0.06159152500615915,
      "grad_norm": 0.4805493652820587,
      "learning_rate": 2e-05,
      "loss": 0.7893,
      "step": 750
    },
    {
      "epoch": 0.06241274533957461,
      "grad_norm": 0.46650201082229614,
      "learning_rate": 2e-05,
      "loss": 0.7947,
      "step": 760
    },
    {
      "epoch": 0.06323396567299007,
      "grad_norm": 0.4958495497703552,
      "learning_rate": 2e-05,
      "loss": 0.8035,
      "step": 770
    },
    {
      "epoch": 0.06405518600640552,
      "grad_norm": 0.45363372564315796,
      "learning_rate": 2e-05,
      "loss": 0.8113,
      "step": 780
    },
    {
      "epoch": 0.06487640633982097,
      "grad_norm": 0.5573371648788452,
      "learning_rate": 2e-05,
      "loss": 0.8294,
      "step": 790
    },
    {
      "epoch": 0.06569762667323643,
      "grad_norm": 0.463094025850296,
      "learning_rate": 2e-05,
      "loss": 0.8263,
      "step": 800
    },
    {
      "epoch": 0.06651884700665188,
      "grad_norm": 0.44852298498153687,
      "learning_rate": 2e-05,
      "loss": 0.8288,
      "step": 810
    },
    {
      "epoch": 0.06734006734006734,
      "grad_norm": 0.4633004367351532,
      "learning_rate": 2e-05,
      "loss": 0.8031,
      "step": 820
    },
    {
      "epoch": 0.0681612876734828,
      "grad_norm": 0.5351435542106628,
      "learning_rate": 2e-05,
      "loss": 0.7989,
      "step": 830
    },
    {
      "epoch": 0.06898250800689826,
      "grad_norm": 0.44983765482902527,
      "learning_rate": 2e-05,
      "loss": 0.7741,
      "step": 840
    },
    {
      "epoch": 0.0698037283403137,
      "grad_norm": 0.46665406227111816,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 850
    },
    {
      "epoch": 0.07062494867372916,
      "grad_norm": 0.4422172009944916,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 860
    },
    {
      "epoch": 0.07144616900714462,
      "grad_norm": 0.4459817111492157,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 870
    },
    {
      "epoch": 0.07226738934056007,
      "grad_norm": 0.45233848690986633,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 880
    },
    {
      "epoch": 0.07308860967397553,
      "grad_norm": 0.4778662919998169,
      "learning_rate": 2e-05,
      "loss": 0.8028,
      "step": 890
    },
    {
      "epoch": 0.07390983000739099,
      "grad_norm": 0.478687584400177,
      "learning_rate": 2e-05,
      "loss": 0.7836,
      "step": 900
    },
    {
      "epoch": 0.07473105034080643,
      "grad_norm": 0.45005080103874207,
      "learning_rate": 2e-05,
      "loss": 0.8078,
      "step": 910
    },
    {
      "epoch": 0.07555227067422189,
      "grad_norm": 0.461701363325119,
      "learning_rate": 2e-05,
      "loss": 0.7897,
      "step": 920
    },
    {
      "epoch": 0.07637349100763735,
      "grad_norm": 0.4510675072669983,
      "learning_rate": 2e-05,
      "loss": 0.8079,
      "step": 930
    },
    {
      "epoch": 0.07719471134105281,
      "grad_norm": 0.44589269161224365,
      "learning_rate": 2e-05,
      "loss": 0.8069,
      "step": 940
    },
    {
      "epoch": 0.07801593167446826,
      "grad_norm": 0.44631075859069824,
      "learning_rate": 2e-05,
      "loss": 0.7879,
      "step": 950
    },
    {
      "epoch": 0.07883715200788372,
      "grad_norm": 0.4562603533267975,
      "learning_rate": 2e-05,
      "loss": 0.7897,
      "step": 960
    },
    {
      "epoch": 0.07965837234129917,
      "grad_norm": 0.4626533091068268,
      "learning_rate": 2e-05,
      "loss": 0.8222,
      "step": 970
    },
    {
      "epoch": 0.08047959267471462,
      "grad_norm": 0.4357568919658661,
      "learning_rate": 2e-05,
      "loss": 0.7865,
      "step": 980
    },
    {
      "epoch": 0.08130081300813008,
      "grad_norm": 0.47860822081565857,
      "learning_rate": 2e-05,
      "loss": 0.8178,
      "step": 990
    },
    {
      "epoch": 0.08212203334154554,
      "grad_norm": 0.450439453125,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 1000
    },
    {
      "epoch": 0.082943253674961,
      "grad_norm": 0.44870707392692566,
      "learning_rate": 2e-05,
      "loss": 0.7867,
      "step": 1010
    },
    {
      "epoch": 0.08376447400837644,
      "grad_norm": 0.425043523311615,
      "learning_rate": 2e-05,
      "loss": 0.7827,
      "step": 1020
    },
    {
      "epoch": 0.0845856943417919,
      "grad_norm": 0.44783613085746765,
      "learning_rate": 2e-05,
      "loss": 0.7968,
      "step": 1030
    },
    {
      "epoch": 0.08540691467520736,
      "grad_norm": 0.44678232073783875,
      "learning_rate": 2e-05,
      "loss": 0.7987,
      "step": 1040
    },
    {
      "epoch": 0.08622813500862281,
      "grad_norm": 0.43474194407463074,
      "learning_rate": 2e-05,
      "loss": 0.8344,
      "step": 1050
    },
    {
      "epoch": 0.08704935534203827,
      "grad_norm": 0.4671313762664795,
      "learning_rate": 2e-05,
      "loss": 0.8042,
      "step": 1060
    },
    {
      "epoch": 0.08787057567545373,
      "grad_norm": 0.4562723934650421,
      "learning_rate": 2e-05,
      "loss": 0.7735,
      "step": 1070
    },
    {
      "epoch": 0.08869179600886919,
      "grad_norm": 0.42270052433013916,
      "learning_rate": 2e-05,
      "loss": 0.8089,
      "step": 1080
    },
    {
      "epoch": 0.08951301634228463,
      "grad_norm": 0.48901864886283875,
      "learning_rate": 2e-05,
      "loss": 0.8132,
      "step": 1090
    },
    {
      "epoch": 0.09033423667570009,
      "grad_norm": 0.45863649249076843,
      "learning_rate": 2e-05,
      "loss": 0.8134,
      "step": 1100
    },
    {
      "epoch": 0.09115545700911555,
      "grad_norm": 0.4497729539871216,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 1110
    },
    {
      "epoch": 0.091976677342531,
      "grad_norm": 0.44030267000198364,
      "learning_rate": 2e-05,
      "loss": 0.7768,
      "step": 1120
    },
    {
      "epoch": 0.09279789767594646,
      "grad_norm": 0.46033450961112976,
      "learning_rate": 2e-05,
      "loss": 0.7889,
      "step": 1130
    },
    {
      "epoch": 0.09361911800936192,
      "grad_norm": 0.4357089102268219,
      "learning_rate": 2e-05,
      "loss": 0.7675,
      "step": 1140
    },
    {
      "epoch": 0.09444033834277736,
      "grad_norm": 0.43551215529441833,
      "learning_rate": 2e-05,
      "loss": 0.7472,
      "step": 1150
    },
    {
      "epoch": 0.09526155867619282,
      "grad_norm": 0.4561745822429657,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 1160
    },
    {
      "epoch": 0.09608277900960828,
      "grad_norm": 0.43203088641166687,
      "learning_rate": 2e-05,
      "loss": 0.8083,
      "step": 1170
    },
    {
      "epoch": 0.09690399934302374,
      "grad_norm": 0.4373405873775482,
      "learning_rate": 2e-05,
      "loss": 0.7751,
      "step": 1180
    },
    {
      "epoch": 0.09772521967643918,
      "grad_norm": 0.4452410936355591,
      "learning_rate": 2e-05,
      "loss": 0.7902,
      "step": 1190
    },
    {
      "epoch": 0.09854644000985464,
      "grad_norm": 0.43690672516822815,
      "learning_rate": 2e-05,
      "loss": 0.7861,
      "step": 1200
    },
    {
      "epoch": 0.0993676603432701,
      "grad_norm": 0.4637196958065033,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 1210
    },
    {
      "epoch": 0.10018888067668555,
      "grad_norm": 0.45546796917915344,
      "learning_rate": 2e-05,
      "loss": 0.767,
      "step": 1220
    },
    {
      "epoch": 0.10101010101010101,
      "grad_norm": 0.461551308631897,
      "learning_rate": 2e-05,
      "loss": 0.7884,
      "step": 1230
    },
    {
      "epoch": 0.10183132134351647,
      "grad_norm": 0.4352053999900818,
      "learning_rate": 2e-05,
      "loss": 0.8028,
      "step": 1240
    },
    {
      "epoch": 0.10265254167693193,
      "grad_norm": 0.44627439975738525,
      "learning_rate": 2e-05,
      "loss": 0.815,
      "step": 1250
    },
    {
      "epoch": 0.10347376201034737,
      "grad_norm": 0.4140150547027588,
      "learning_rate": 2e-05,
      "loss": 0.7819,
      "step": 1260
    },
    {
      "epoch": 0.10429498234376283,
      "grad_norm": 0.4781087040901184,
      "learning_rate": 2e-05,
      "loss": 0.7974,
      "step": 1270
    },
    {
      "epoch": 0.10511620267717829,
      "grad_norm": 0.4421789348125458,
      "learning_rate": 2e-05,
      "loss": 0.7632,
      "step": 1280
    },
    {
      "epoch": 0.10593742301059374,
      "grad_norm": 0.44503435492515564,
      "learning_rate": 2e-05,
      "loss": 0.7974,
      "step": 1290
    },
    {
      "epoch": 0.1067586433440092,
      "grad_norm": 0.4360303282737732,
      "learning_rate": 2e-05,
      "loss": 0.7789,
      "step": 1300
    },
    {
      "epoch": 0.10757986367742466,
      "grad_norm": 0.4185614585876465,
      "learning_rate": 2e-05,
      "loss": 0.7962,
      "step": 1310
    },
    {
      "epoch": 0.10840108401084012,
      "grad_norm": 0.46108296513557434,
      "learning_rate": 2e-05,
      "loss": 0.7752,
      "step": 1320
    },
    {
      "epoch": 0.10922230434425556,
      "grad_norm": 0.46323254704475403,
      "learning_rate": 2e-05,
      "loss": 0.8015,
      "step": 1330
    },
    {
      "epoch": 0.11004352467767102,
      "grad_norm": 0.441444456577301,
      "learning_rate": 2e-05,
      "loss": 0.7934,
      "step": 1340
    },
    {
      "epoch": 0.11086474501108648,
      "grad_norm": 0.42753297090530396,
      "learning_rate": 2e-05,
      "loss": 0.7869,
      "step": 1350
    },
    {
      "epoch": 0.11168596534450193,
      "grad_norm": 0.5354635715484619,
      "learning_rate": 2e-05,
      "loss": 0.7668,
      "step": 1360
    },
    {
      "epoch": 0.11250718567791738,
      "grad_norm": 0.4485228657722473,
      "learning_rate": 2e-05,
      "loss": 0.7916,
      "step": 1370
    },
    {
      "epoch": 0.11332840601133284,
      "grad_norm": 0.4402100145816803,
      "learning_rate": 2e-05,
      "loss": 0.7785,
      "step": 1380
    },
    {
      "epoch": 0.11414962634474829,
      "grad_norm": 0.44067618250846863,
      "learning_rate": 2e-05,
      "loss": 0.796,
      "step": 1390
    },
    {
      "epoch": 0.11497084667816375,
      "grad_norm": 0.41861099004745483,
      "learning_rate": 2e-05,
      "loss": 0.7896,
      "step": 1400
    },
    {
      "epoch": 0.11579206701157921,
      "grad_norm": 0.427713006734848,
      "learning_rate": 2e-05,
      "loss": 0.7876,
      "step": 1410
    },
    {
      "epoch": 0.11661328734499467,
      "grad_norm": 0.5142576098442078,
      "learning_rate": 2e-05,
      "loss": 0.7832,
      "step": 1420
    },
    {
      "epoch": 0.11743450767841011,
      "grad_norm": 0.4011819660663605,
      "learning_rate": 2e-05,
      "loss": 0.7894,
      "step": 1430
    },
    {
      "epoch": 0.11825572801182557,
      "grad_norm": 0.4302540421485901,
      "learning_rate": 2e-05,
      "loss": 0.7805,
      "step": 1440
    },
    {
      "epoch": 0.11907694834524103,
      "grad_norm": 0.4870634973049164,
      "learning_rate": 2e-05,
      "loss": 0.7596,
      "step": 1450
    },
    {
      "epoch": 0.11989816867865648,
      "grad_norm": 0.4451192021369934,
      "learning_rate": 2e-05,
      "loss": 0.7769,
      "step": 1460
    },
    {
      "epoch": 0.12071938901207194,
      "grad_norm": 0.42198508977890015,
      "learning_rate": 2e-05,
      "loss": 0.7746,
      "step": 1470
    },
    {
      "epoch": 0.1215406093454874,
      "grad_norm": 0.4552297294139862,
      "learning_rate": 2e-05,
      "loss": 0.8001,
      "step": 1480
    },
    {
      "epoch": 0.12236182967890286,
      "grad_norm": 0.44329190254211426,
      "learning_rate": 2e-05,
      "loss": 0.7702,
      "step": 1490
    },
    {
      "epoch": 0.1231830500123183,
      "grad_norm": 0.44024938344955444,
      "learning_rate": 2e-05,
      "loss": 0.7976,
      "step": 1500
    },
    {
      "epoch": 0.12400427034573376,
      "grad_norm": 0.4466274082660675,
      "learning_rate": 2e-05,
      "loss": 0.8065,
      "step": 1510
    },
    {
      "epoch": 0.12482549067914922,
      "grad_norm": 0.4443671703338623,
      "learning_rate": 2e-05,
      "loss": 0.7734,
      "step": 1520
    },
    {
      "epoch": 0.12564671101256467,
      "grad_norm": 0.4458763301372528,
      "learning_rate": 2e-05,
      "loss": 0.7849,
      "step": 1530
    },
    {
      "epoch": 0.12646793134598014,
      "grad_norm": 0.4324412941932678,
      "learning_rate": 2e-05,
      "loss": 0.8091,
      "step": 1540
    },
    {
      "epoch": 0.12728915167939558,
      "grad_norm": 0.474600613117218,
      "learning_rate": 2e-05,
      "loss": 0.7841,
      "step": 1550
    },
    {
      "epoch": 0.12811037201281103,
      "grad_norm": 0.42612481117248535,
      "learning_rate": 2e-05,
      "loss": 0.779,
      "step": 1560
    },
    {
      "epoch": 0.1289315923462265,
      "grad_norm": 0.41486653685569763,
      "learning_rate": 2e-05,
      "loss": 0.7862,
      "step": 1570
    },
    {
      "epoch": 0.12975281267964195,
      "grad_norm": 0.4324989318847656,
      "learning_rate": 2e-05,
      "loss": 0.7953,
      "step": 1580
    },
    {
      "epoch": 0.1305740330130574,
      "grad_norm": 0.4577152729034424,
      "learning_rate": 2e-05,
      "loss": 0.7666,
      "step": 1590
    },
    {
      "epoch": 0.13139525334647287,
      "grad_norm": 0.44246548414230347,
      "learning_rate": 2e-05,
      "loss": 0.7608,
      "step": 1600
    },
    {
      "epoch": 0.1322164736798883,
      "grad_norm": 0.4186859130859375,
      "learning_rate": 2e-05,
      "loss": 0.7604,
      "step": 1610
    },
    {
      "epoch": 0.13303769401330376,
      "grad_norm": 0.42295071482658386,
      "learning_rate": 2e-05,
      "loss": 0.7466,
      "step": 1620
    },
    {
      "epoch": 0.13385891434671923,
      "grad_norm": 0.425224632024765,
      "learning_rate": 2e-05,
      "loss": 0.7704,
      "step": 1630
    },
    {
      "epoch": 0.13468013468013468,
      "grad_norm": 0.4620305895805359,
      "learning_rate": 2e-05,
      "loss": 0.7699,
      "step": 1640
    },
    {
      "epoch": 0.13550135501355012,
      "grad_norm": 0.4204477369785309,
      "learning_rate": 2e-05,
      "loss": 0.8106,
      "step": 1650
    },
    {
      "epoch": 0.1363225753469656,
      "grad_norm": 0.41886085271835327,
      "learning_rate": 2e-05,
      "loss": 0.7764,
      "step": 1660
    },
    {
      "epoch": 0.13714379568038104,
      "grad_norm": 0.45777595043182373,
      "learning_rate": 2e-05,
      "loss": 0.7689,
      "step": 1670
    },
    {
      "epoch": 0.13796501601379652,
      "grad_norm": 0.41233789920806885,
      "learning_rate": 2e-05,
      "loss": 0.7701,
      "step": 1680
    },
    {
      "epoch": 0.13878623634721196,
      "grad_norm": 0.44005653262138367,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 1690
    },
    {
      "epoch": 0.1396074566806274,
      "grad_norm": 0.47396785020828247,
      "learning_rate": 2e-05,
      "loss": 0.7398,
      "step": 1700
    },
    {
      "epoch": 0.14042867701404288,
      "grad_norm": 0.43871358036994934,
      "learning_rate": 2e-05,
      "loss": 0.7747,
      "step": 1710
    },
    {
      "epoch": 0.14124989734745833,
      "grad_norm": 0.421408474445343,
      "learning_rate": 2e-05,
      "loss": 0.7905,
      "step": 1720
    },
    {
      "epoch": 0.14207111768087377,
      "grad_norm": 0.43005624413490295,
      "learning_rate": 2e-05,
      "loss": 0.7609,
      "step": 1730
    },
    {
      "epoch": 0.14289233801428924,
      "grad_norm": 0.4219532012939453,
      "learning_rate": 2e-05,
      "loss": 0.7609,
      "step": 1740
    },
    {
      "epoch": 0.1437135583477047,
      "grad_norm": 0.4508216083049774,
      "learning_rate": 2e-05,
      "loss": 0.6405,
      "step": 1750
    },
    {
      "epoch": 0.14453477868112014,
      "grad_norm": 0.4121701121330261,
      "learning_rate": 2e-05,
      "loss": 0.7763,
      "step": 1760
    },
    {
      "epoch": 0.1453559990145356,
      "grad_norm": 0.4090173542499542,
      "learning_rate": 2e-05,
      "loss": 0.7645,
      "step": 1770
    },
    {
      "epoch": 0.14617721934795105,
      "grad_norm": 0.43645748496055603,
      "learning_rate": 2e-05,
      "loss": 0.8122,
      "step": 1780
    },
    {
      "epoch": 0.1469984396813665,
      "grad_norm": 1.1069415807724,
      "learning_rate": 2e-05,
      "loss": 0.767,
      "step": 1790
    },
    {
      "epoch": 0.14781966001478197,
      "grad_norm": 0.4456346035003662,
      "learning_rate": 2e-05,
      "loss": 0.7763,
      "step": 1800
    },
    {
      "epoch": 0.14864088034819742,
      "grad_norm": 0.42877396941185,
      "learning_rate": 2e-05,
      "loss": 0.7624,
      "step": 1810
    },
    {
      "epoch": 0.14946210068161286,
      "grad_norm": 0.4232544004917145,
      "learning_rate": 2e-05,
      "loss": 0.7606,
      "step": 1820
    },
    {
      "epoch": 0.15028332101502834,
      "grad_norm": 0.421718031167984,
      "learning_rate": 2e-05,
      "loss": 0.7834,
      "step": 1830
    },
    {
      "epoch": 0.15110454134844378,
      "grad_norm": 0.44949644804000854,
      "learning_rate": 2e-05,
      "loss": 0.778,
      "step": 1840
    },
    {
      "epoch": 0.15192576168185926,
      "grad_norm": 0.430152028799057,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 1850
    },
    {
      "epoch": 0.1527469820152747,
      "grad_norm": 0.48783883452415466,
      "learning_rate": 2e-05,
      "loss": 0.768,
      "step": 1860
    },
    {
      "epoch": 0.15356820234869015,
      "grad_norm": 0.405388206243515,
      "learning_rate": 2e-05,
      "loss": 0.7677,
      "step": 1870
    },
    {
      "epoch": 0.15438942268210562,
      "grad_norm": 0.4099145531654358,
      "learning_rate": 2e-05,
      "loss": 0.7448,
      "step": 1880
    },
    {
      "epoch": 0.15521064301552107,
      "grad_norm": 0.4216562509536743,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 1890
    },
    {
      "epoch": 0.1560318633489365,
      "grad_norm": 0.42413970828056335,
      "learning_rate": 2e-05,
      "loss": 0.7534,
      "step": 1900
    },
    {
      "epoch": 0.15685308368235198,
      "grad_norm": 0.43225663900375366,
      "learning_rate": 2e-05,
      "loss": 0.7739,
      "step": 1910
    },
    {
      "epoch": 0.15767430401576743,
      "grad_norm": 0.4282965362071991,
      "learning_rate": 2e-05,
      "loss": 0.7698,
      "step": 1920
    },
    {
      "epoch": 0.15849552434918288,
      "grad_norm": 0.49937933683395386,
      "learning_rate": 2e-05,
      "loss": 0.7814,
      "step": 1930
    },
    {
      "epoch": 0.15931674468259835,
      "grad_norm": 0.4623638093471527,
      "learning_rate": 2e-05,
      "loss": 0.7455,
      "step": 1940
    },
    {
      "epoch": 0.1601379650160138,
      "grad_norm": 0.40429919958114624,
      "learning_rate": 2e-05,
      "loss": 0.7624,
      "step": 1950
    },
    {
      "epoch": 0.16095918534942924,
      "grad_norm": 0.4180993139743805,
      "learning_rate": 2e-05,
      "loss": 0.812,
      "step": 1960
    },
    {
      "epoch": 0.1617804056828447,
      "grad_norm": 0.4380674362182617,
      "learning_rate": 2e-05,
      "loss": 0.7812,
      "step": 1970
    },
    {
      "epoch": 0.16260162601626016,
      "grad_norm": 0.4467218518257141,
      "learning_rate": 2e-05,
      "loss": 0.777,
      "step": 1980
    },
    {
      "epoch": 0.1634228463496756,
      "grad_norm": 0.6018455624580383,
      "learning_rate": 2e-05,
      "loss": 0.8078,
      "step": 1990
    },
    {
      "epoch": 0.16424406668309108,
      "grad_norm": 0.42848190665245056,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 2000
    },
    {
      "epoch": 0.16506528701650652,
      "grad_norm": 0.434704065322876,
      "learning_rate": 2e-05,
      "loss": 0.7626,
      "step": 2010
    },
    {
      "epoch": 0.165886507349922,
      "grad_norm": 0.4402262270450592,
      "learning_rate": 2e-05,
      "loss": 0.7705,
      "step": 2020
    },
    {
      "epoch": 0.16670772768333744,
      "grad_norm": 0.42635026574134827,
      "learning_rate": 2e-05,
      "loss": 0.7682,
      "step": 2030
    },
    {
      "epoch": 0.1675289480167529,
      "grad_norm": 0.4290219843387604,
      "learning_rate": 2e-05,
      "loss": 0.7966,
      "step": 2040
    },
    {
      "epoch": 0.16835016835016836,
      "grad_norm": 0.4775022268295288,
      "learning_rate": 2e-05,
      "loss": 0.7724,
      "step": 2050
    },
    {
      "epoch": 0.1691713886835838,
      "grad_norm": 0.43119269609451294,
      "learning_rate": 2e-05,
      "loss": 0.7801,
      "step": 2060
    },
    {
      "epoch": 0.16999260901699925,
      "grad_norm": 0.5889827013015747,
      "learning_rate": 2e-05,
      "loss": 0.8041,
      "step": 2070
    },
    {
      "epoch": 0.17081382935041473,
      "grad_norm": 0.4321953356266022,
      "learning_rate": 2e-05,
      "loss": 0.8019,
      "step": 2080
    },
    {
      "epoch": 0.17163504968383017,
      "grad_norm": 0.428431898355484,
      "learning_rate": 2e-05,
      "loss": 0.7751,
      "step": 2090
    },
    {
      "epoch": 0.17245627001724562,
      "grad_norm": 0.402853399515152,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 2100
    },
    {
      "epoch": 0.1732774903506611,
      "grad_norm": 0.4419548511505127,
      "learning_rate": 2e-05,
      "loss": 0.7806,
      "step": 2110
    },
    {
      "epoch": 0.17409871068407654,
      "grad_norm": 0.44852688908576965,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 2120
    },
    {
      "epoch": 0.17491993101749198,
      "grad_norm": 0.4230371415615082,
      "learning_rate": 2e-05,
      "loss": 0.7914,
      "step": 2130
    },
    {
      "epoch": 0.17574115135090745,
      "grad_norm": 0.3861851394176483,
      "learning_rate": 2e-05,
      "loss": 0.7233,
      "step": 2140
    },
    {
      "epoch": 0.1765623716843229,
      "grad_norm": 0.41707614064216614,
      "learning_rate": 2e-05,
      "loss": 0.7569,
      "step": 2150
    },
    {
      "epoch": 0.17738359201773837,
      "grad_norm": 0.4453151226043701,
      "learning_rate": 2e-05,
      "loss": 0.758,
      "step": 2160
    },
    {
      "epoch": 0.17820481235115382,
      "grad_norm": 0.433058500289917,
      "learning_rate": 2e-05,
      "loss": 0.8025,
      "step": 2170
    },
    {
      "epoch": 0.17902603268456926,
      "grad_norm": 0.5075773596763611,
      "learning_rate": 2e-05,
      "loss": 0.7356,
      "step": 2180
    },
    {
      "epoch": 0.17984725301798474,
      "grad_norm": 0.4086076617240906,
      "learning_rate": 2e-05,
      "loss": 0.7379,
      "step": 2190
    },
    {
      "epoch": 0.18066847335140018,
      "grad_norm": 0.4531570076942444,
      "learning_rate": 2e-05,
      "loss": 0.7911,
      "step": 2200
    },
    {
      "epoch": 0.18148969368481563,
      "grad_norm": 0.41858381032943726,
      "learning_rate": 2e-05,
      "loss": 0.7672,
      "step": 2210
    },
    {
      "epoch": 0.1823109140182311,
      "grad_norm": 0.4220307767391205,
      "learning_rate": 2e-05,
      "loss": 0.7534,
      "step": 2220
    },
    {
      "epoch": 0.18313213435164655,
      "grad_norm": 0.4243931472301483,
      "learning_rate": 2e-05,
      "loss": 0.7547,
      "step": 2230
    },
    {
      "epoch": 0.183953354685062,
      "grad_norm": 0.4126180410385132,
      "learning_rate": 2e-05,
      "loss": 0.7489,
      "step": 2240
    },
    {
      "epoch": 0.18477457501847747,
      "grad_norm": 0.42353975772857666,
      "learning_rate": 2e-05,
      "loss": 0.7961,
      "step": 2250
    },
    {
      "epoch": 0.1855957953518929,
      "grad_norm": 0.4082265794277191,
      "learning_rate": 2e-05,
      "loss": 0.757,
      "step": 2260
    },
    {
      "epoch": 0.18641701568530836,
      "grad_norm": 0.407305508852005,
      "learning_rate": 2e-05,
      "loss": 0.7543,
      "step": 2270
    },
    {
      "epoch": 0.18723823601872383,
      "grad_norm": 0.42294174432754517,
      "learning_rate": 2e-05,
      "loss": 0.7583,
      "step": 2280
    },
    {
      "epoch": 0.18805945635213928,
      "grad_norm": 0.39620959758758545,
      "learning_rate": 2e-05,
      "loss": 0.7566,
      "step": 2290
    },
    {
      "epoch": 0.18888067668555472,
      "grad_norm": 0.4146956205368042,
      "learning_rate": 2e-05,
      "loss": 0.7802,
      "step": 2300
    },
    {
      "epoch": 0.1897018970189702,
      "grad_norm": 0.43170496821403503,
      "learning_rate": 2e-05,
      "loss": 0.7559,
      "step": 2310
    },
    {
      "epoch": 0.19052311735238564,
      "grad_norm": 0.4103269577026367,
      "learning_rate": 2e-05,
      "loss": 0.7575,
      "step": 2320
    },
    {
      "epoch": 0.1913443376858011,
      "grad_norm": 0.39337730407714844,
      "learning_rate": 2e-05,
      "loss": 0.7466,
      "step": 2330
    },
    {
      "epoch": 0.19216555801921656,
      "grad_norm": 0.4346054196357727,
      "learning_rate": 2e-05,
      "loss": 0.7906,
      "step": 2340
    },
    {
      "epoch": 0.192986778352632,
      "grad_norm": 0.42356476187705994,
      "learning_rate": 2e-05,
      "loss": 0.7628,
      "step": 2350
    },
    {
      "epoch": 0.19380799868604748,
      "grad_norm": 0.377941370010376,
      "learning_rate": 2e-05,
      "loss": 0.7514,
      "step": 2360
    },
    {
      "epoch": 0.19462921901946292,
      "grad_norm": 0.4733028709888458,
      "learning_rate": 2e-05,
      "loss": 0.7463,
      "step": 2370
    },
    {
      "epoch": 0.19545043935287837,
      "grad_norm": 0.4126057028770447,
      "learning_rate": 2e-05,
      "loss": 0.7892,
      "step": 2380
    },
    {
      "epoch": 0.19627165968629384,
      "grad_norm": 0.4060257375240326,
      "learning_rate": 2e-05,
      "loss": 0.7651,
      "step": 2390
    },
    {
      "epoch": 0.1970928800197093,
      "grad_norm": 0.41388368606567383,
      "learning_rate": 2e-05,
      "loss": 0.7562,
      "step": 2400
    },
    {
      "epoch": 0.19791410035312473,
      "grad_norm": 0.42862391471862793,
      "learning_rate": 2e-05,
      "loss": 0.7811,
      "step": 2410
    },
    {
      "epoch": 0.1987353206865402,
      "grad_norm": 0.4284455180168152,
      "learning_rate": 2e-05,
      "loss": 0.7721,
      "step": 2420
    },
    {
      "epoch": 0.19955654101995565,
      "grad_norm": 0.4654240310192108,
      "learning_rate": 2e-05,
      "loss": 0.7944,
      "step": 2430
    },
    {
      "epoch": 0.2003777613533711,
      "grad_norm": 0.38943159580230713,
      "learning_rate": 2e-05,
      "loss": 0.7601,
      "step": 2440
    },
    {
      "epoch": 0.20119898168678657,
      "grad_norm": 0.40954166650772095,
      "learning_rate": 2e-05,
      "loss": 0.7495,
      "step": 2450
    },
    {
      "epoch": 0.20202020202020202,
      "grad_norm": 0.3856504261493683,
      "learning_rate": 2e-05,
      "loss": 0.7451,
      "step": 2460
    },
    {
      "epoch": 0.20284142235361746,
      "grad_norm": 0.39670035243034363,
      "learning_rate": 2e-05,
      "loss": 0.7394,
      "step": 2470
    },
    {
      "epoch": 0.20366264268703294,
      "grad_norm": 0.39305543899536133,
      "learning_rate": 2e-05,
      "loss": 0.7679,
      "step": 2480
    },
    {
      "epoch": 0.20448386302044838,
      "grad_norm": 0.39719080924987793,
      "learning_rate": 2e-05,
      "loss": 0.784,
      "step": 2490
    },
    {
      "epoch": 0.20530508335386385,
      "grad_norm": 0.41948506236076355,
      "learning_rate": 2e-05,
      "loss": 0.7668,
      "step": 2500
    },
    {
      "epoch": 0.2061263036872793,
      "grad_norm": 0.42262619733810425,
      "learning_rate": 2e-05,
      "loss": 0.7699,
      "step": 2510
    },
    {
      "epoch": 0.20694752402069475,
      "grad_norm": 0.40783119201660156,
      "learning_rate": 2e-05,
      "loss": 0.7699,
      "step": 2520
    },
    {
      "epoch": 0.20776874435411022,
      "grad_norm": 0.4184160828590393,
      "learning_rate": 2e-05,
      "loss": 0.8074,
      "step": 2530
    },
    {
      "epoch": 0.20858996468752566,
      "grad_norm": 0.40542080998420715,
      "learning_rate": 2e-05,
      "loss": 0.7752,
      "step": 2540
    },
    {
      "epoch": 0.2094111850209411,
      "grad_norm": 0.4655381143093109,
      "learning_rate": 2e-05,
      "loss": 0.7454,
      "step": 2550
    },
    {
      "epoch": 0.21023240535435658,
      "grad_norm": 0.4105184078216553,
      "learning_rate": 2e-05,
      "loss": 0.7944,
      "step": 2560
    },
    {
      "epoch": 0.21105362568777203,
      "grad_norm": 0.4311380088329315,
      "learning_rate": 2e-05,
      "loss": 0.7735,
      "step": 2570
    },
    {
      "epoch": 0.21187484602118747,
      "grad_norm": 0.4171161651611328,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 2580
    },
    {
      "epoch": 0.21269606635460295,
      "grad_norm": 0.40367957949638367,
      "learning_rate": 2e-05,
      "loss": 0.7538,
      "step": 2590
    },
    {
      "epoch": 0.2135172866880184,
      "grad_norm": 0.4587738513946533,
      "learning_rate": 2e-05,
      "loss": 0.7854,
      "step": 2600
    },
    {
      "epoch": 0.21433850702143384,
      "grad_norm": 0.42091241478919983,
      "learning_rate": 2e-05,
      "loss": 0.7742,
      "step": 2610
    },
    {
      "epoch": 0.2151597273548493,
      "grad_norm": 0.41562601923942566,
      "learning_rate": 2e-05,
      "loss": 0.7517,
      "step": 2620
    },
    {
      "epoch": 0.21598094768826476,
      "grad_norm": 0.4034254848957062,
      "learning_rate": 2e-05,
      "loss": 0.7631,
      "step": 2630
    },
    {
      "epoch": 0.21680216802168023,
      "grad_norm": 0.4134150743484497,
      "learning_rate": 2e-05,
      "loss": 0.7744,
      "step": 2640
    },
    {
      "epoch": 0.21762338835509568,
      "grad_norm": 0.42591631412506104,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 2650
    },
    {
      "epoch": 0.21844460868851112,
      "grad_norm": 0.4130219519138336,
      "learning_rate": 2e-05,
      "loss": 0.7387,
      "step": 2660
    },
    {
      "epoch": 0.2192658290219266,
      "grad_norm": 0.43417590856552124,
      "learning_rate": 2e-05,
      "loss": 0.7353,
      "step": 2670
    },
    {
      "epoch": 0.22008704935534204,
      "grad_norm": 0.41847848892211914,
      "learning_rate": 2e-05,
      "loss": 0.7423,
      "step": 2680
    },
    {
      "epoch": 0.22090826968875749,
      "grad_norm": 0.3885892629623413,
      "learning_rate": 2e-05,
      "loss": 0.756,
      "step": 2690
    },
    {
      "epoch": 0.22172949002217296,
      "grad_norm": 0.413310170173645,
      "learning_rate": 2e-05,
      "loss": 0.7752,
      "step": 2700
    },
    {
      "epoch": 0.2225507103555884,
      "grad_norm": 0.4031926393508911,
      "learning_rate": 2e-05,
      "loss": 0.7549,
      "step": 2710
    },
    {
      "epoch": 0.22337193068900385,
      "grad_norm": 0.42383497953414917,
      "learning_rate": 2e-05,
      "loss": 0.7453,
      "step": 2720
    },
    {
      "epoch": 0.22419315102241932,
      "grad_norm": 0.4209596514701843,
      "learning_rate": 2e-05,
      "loss": 0.7491,
      "step": 2730
    },
    {
      "epoch": 0.22501437135583477,
      "grad_norm": 0.4406531751155853,
      "learning_rate": 2e-05,
      "loss": 0.7408,
      "step": 2740
    },
    {
      "epoch": 0.22583559168925021,
      "grad_norm": 0.40509483218193054,
      "learning_rate": 2e-05,
      "loss": 0.7781,
      "step": 2750
    },
    {
      "epoch": 0.2266568120226657,
      "grad_norm": 0.4356958866119385,
      "learning_rate": 2e-05,
      "loss": 0.783,
      "step": 2760
    },
    {
      "epoch": 0.22747803235608113,
      "grad_norm": 0.41241705417633057,
      "learning_rate": 2e-05,
      "loss": 0.7687,
      "step": 2770
    },
    {
      "epoch": 0.22829925268949658,
      "grad_norm": 0.4022531807422638,
      "learning_rate": 2e-05,
      "loss": 0.7667,
      "step": 2780
    },
    {
      "epoch": 0.22912047302291205,
      "grad_norm": 0.41086339950561523,
      "learning_rate": 2e-05,
      "loss": 0.7418,
      "step": 2790
    },
    {
      "epoch": 0.2299416933563275,
      "grad_norm": 0.3777279853820801,
      "learning_rate": 2e-05,
      "loss": 0.7624,
      "step": 2800
    },
    {
      "epoch": 0.23076291368974297,
      "grad_norm": 0.39642786979675293,
      "learning_rate": 2e-05,
      "loss": 0.7705,
      "step": 2810
    },
    {
      "epoch": 0.23158413402315842,
      "grad_norm": 0.4321180582046509,
      "learning_rate": 2e-05,
      "loss": 0.7424,
      "step": 2820
    },
    {
      "epoch": 0.23240535435657386,
      "grad_norm": 0.41064831614494324,
      "learning_rate": 2e-05,
      "loss": 0.7447,
      "step": 2830
    },
    {
      "epoch": 0.23322657468998934,
      "grad_norm": 0.40122440457344055,
      "learning_rate": 2e-05,
      "loss": 0.7459,
      "step": 2840
    },
    {
      "epoch": 0.23404779502340478,
      "grad_norm": 0.39732176065444946,
      "learning_rate": 2e-05,
      "loss": 0.7308,
      "step": 2850
    },
    {
      "epoch": 0.23486901535682023,
      "grad_norm": 0.4249623715877533,
      "learning_rate": 2e-05,
      "loss": 0.7522,
      "step": 2860
    },
    {
      "epoch": 0.2356902356902357,
      "grad_norm": 0.4154263734817505,
      "learning_rate": 2e-05,
      "loss": 0.7428,
      "step": 2870
    },
    {
      "epoch": 0.23651145602365115,
      "grad_norm": 0.41803911328315735,
      "learning_rate": 2e-05,
      "loss": 0.7978,
      "step": 2880
    },
    {
      "epoch": 0.2373326763570666,
      "grad_norm": 0.41938889026641846,
      "learning_rate": 2e-05,
      "loss": 0.7603,
      "step": 2890
    },
    {
      "epoch": 0.23815389669048206,
      "grad_norm": 0.3966026306152344,
      "learning_rate": 2e-05,
      "loss": 0.7213,
      "step": 2900
    },
    {
      "epoch": 0.2389751170238975,
      "grad_norm": 0.4096229076385498,
      "learning_rate": 2e-05,
      "loss": 0.7811,
      "step": 2910
    },
    {
      "epoch": 0.23979633735731296,
      "grad_norm": 0.412630558013916,
      "learning_rate": 2e-05,
      "loss": 0.723,
      "step": 2920
    },
    {
      "epoch": 0.24061755769072843,
      "grad_norm": 0.42935243248939514,
      "learning_rate": 2e-05,
      "loss": 0.7626,
      "step": 2930
    },
    {
      "epoch": 0.24143877802414387,
      "grad_norm": 0.4085247218608856,
      "learning_rate": 2e-05,
      "loss": 0.7328,
      "step": 2940
    },
    {
      "epoch": 0.24225999835755932,
      "grad_norm": 0.37739884853363037,
      "learning_rate": 2e-05,
      "loss": 0.7499,
      "step": 2950
    },
    {
      "epoch": 0.2430812186909748,
      "grad_norm": 0.3953222632408142,
      "learning_rate": 2e-05,
      "loss": 0.7425,
      "step": 2960
    },
    {
      "epoch": 0.24390243902439024,
      "grad_norm": 0.4459753632545471,
      "learning_rate": 2e-05,
      "loss": 0.7697,
      "step": 2970
    },
    {
      "epoch": 0.2447236593578057,
      "grad_norm": 0.4445836544036865,
      "learning_rate": 2e-05,
      "loss": 0.7393,
      "step": 2980
    },
    {
      "epoch": 0.24554487969122116,
      "grad_norm": 0.40826845169067383,
      "learning_rate": 2e-05,
      "loss": 0.7467,
      "step": 2990
    },
    {
      "epoch": 0.2463661000246366,
      "grad_norm": 0.3922569453716278,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 3000
    },
    {
      "epoch": 0.24718732035805208,
      "grad_norm": 0.43796128034591675,
      "learning_rate": 2e-05,
      "loss": 0.7618,
      "step": 3010
    },
    {
      "epoch": 0.24800854069146752,
      "grad_norm": 0.425777792930603,
      "learning_rate": 2e-05,
      "loss": 0.763,
      "step": 3020
    },
    {
      "epoch": 0.24882976102488297,
      "grad_norm": 0.4194261133670807,
      "learning_rate": 2e-05,
      "loss": 0.7416,
      "step": 3030
    },
    {
      "epoch": 0.24965098135829844,
      "grad_norm": 0.41821855306625366,
      "learning_rate": 2e-05,
      "loss": 0.7599,
      "step": 3040
    },
    {
      "epoch": 0.2504722016917139,
      "grad_norm": 0.3936750292778015,
      "learning_rate": 2e-05,
      "loss": 0.727,
      "step": 3050
    },
    {
      "epoch": 0.25129342202512933,
      "grad_norm": 0.3971061706542969,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 3060
    },
    {
      "epoch": 0.2521146423585448,
      "grad_norm": 0.4379628598690033,
      "learning_rate": 2e-05,
      "loss": 0.7536,
      "step": 3070
    },
    {
      "epoch": 0.2529358626919603,
      "grad_norm": 0.41062507033348083,
      "learning_rate": 2e-05,
      "loss": 0.7449,
      "step": 3080
    },
    {
      "epoch": 0.2537570830253757,
      "grad_norm": 0.3907121419906616,
      "learning_rate": 2e-05,
      "loss": 0.7439,
      "step": 3090
    },
    {
      "epoch": 0.25457830335879117,
      "grad_norm": 0.42453211545944214,
      "learning_rate": 2e-05,
      "loss": 0.7613,
      "step": 3100
    },
    {
      "epoch": 0.25539952369220664,
      "grad_norm": 0.4198872447013855,
      "learning_rate": 2e-05,
      "loss": 0.7531,
      "step": 3110
    },
    {
      "epoch": 0.25622074402562206,
      "grad_norm": 0.4108971357345581,
      "learning_rate": 2e-05,
      "loss": 0.7554,
      "step": 3120
    },
    {
      "epoch": 0.25704196435903753,
      "grad_norm": 0.39757779240608215,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 3130
    },
    {
      "epoch": 0.257863184692453,
      "grad_norm": 0.41511502861976624,
      "learning_rate": 2e-05,
      "loss": 0.7473,
      "step": 3140
    },
    {
      "epoch": 0.2586844050258684,
      "grad_norm": 0.392802357673645,
      "learning_rate": 2e-05,
      "loss": 0.741,
      "step": 3150
    },
    {
      "epoch": 0.2595056253592839,
      "grad_norm": 0.4123692810535431,
      "learning_rate": 2e-05,
      "loss": 0.7681,
      "step": 3160
    },
    {
      "epoch": 0.26032684569269937,
      "grad_norm": 0.3944304883480072,
      "learning_rate": 2e-05,
      "loss": 0.7263,
      "step": 3170
    },
    {
      "epoch": 0.2611480660261148,
      "grad_norm": 0.4104151129722595,
      "learning_rate": 2e-05,
      "loss": 0.7334,
      "step": 3180
    },
    {
      "epoch": 0.26196928635953026,
      "grad_norm": 0.42089709639549255,
      "learning_rate": 2e-05,
      "loss": 0.7243,
      "step": 3190
    },
    {
      "epoch": 0.26279050669294574,
      "grad_norm": 0.429602712392807,
      "learning_rate": 2e-05,
      "loss": 0.7553,
      "step": 3200
    },
    {
      "epoch": 0.26361172702636115,
      "grad_norm": 0.40421098470687866,
      "learning_rate": 2e-05,
      "loss": 0.7013,
      "step": 3210
    },
    {
      "epoch": 0.2644329473597766,
      "grad_norm": 0.4190472662448883,
      "learning_rate": 2e-05,
      "loss": 0.7609,
      "step": 3220
    },
    {
      "epoch": 0.2652541676931921,
      "grad_norm": 0.4069626033306122,
      "learning_rate": 2e-05,
      "loss": 0.7454,
      "step": 3230
    },
    {
      "epoch": 0.2660753880266075,
      "grad_norm": 0.43579307198524475,
      "learning_rate": 2e-05,
      "loss": 0.7794,
      "step": 3240
    },
    {
      "epoch": 0.266896608360023,
      "grad_norm": 0.4225179851055145,
      "learning_rate": 2e-05,
      "loss": 0.7444,
      "step": 3250
    },
    {
      "epoch": 0.26771782869343846,
      "grad_norm": 0.4154949486255646,
      "learning_rate": 2e-05,
      "loss": 0.7302,
      "step": 3260
    },
    {
      "epoch": 0.2685390490268539,
      "grad_norm": 0.4359280467033386,
      "learning_rate": 2e-05,
      "loss": 0.7263,
      "step": 3270
    },
    {
      "epoch": 0.26936026936026936,
      "grad_norm": 0.4006137549877167,
      "learning_rate": 2e-05,
      "loss": 0.7471,
      "step": 3280
    },
    {
      "epoch": 0.27018148969368483,
      "grad_norm": 0.42727410793304443,
      "learning_rate": 2e-05,
      "loss": 0.7359,
      "step": 3290
    },
    {
      "epoch": 0.27100271002710025,
      "grad_norm": 0.41258957982063293,
      "learning_rate": 2e-05,
      "loss": 0.7272,
      "step": 3300
    },
    {
      "epoch": 0.2718239303605157,
      "grad_norm": 0.4213484823703766,
      "learning_rate": 2e-05,
      "loss": 0.7236,
      "step": 3310
    },
    {
      "epoch": 0.2726451506939312,
      "grad_norm": 0.4244912266731262,
      "learning_rate": 2e-05,
      "loss": 0.7826,
      "step": 3320
    },
    {
      "epoch": 0.2734663710273466,
      "grad_norm": 0.3961259424686432,
      "learning_rate": 2e-05,
      "loss": 0.7328,
      "step": 3330
    },
    {
      "epoch": 0.2742875913607621,
      "grad_norm": 0.40558815002441406,
      "learning_rate": 2e-05,
      "loss": 0.793,
      "step": 3340
    },
    {
      "epoch": 0.27510881169417756,
      "grad_norm": 0.41350263357162476,
      "learning_rate": 2e-05,
      "loss": 0.7386,
      "step": 3350
    },
    {
      "epoch": 0.27593003202759303,
      "grad_norm": 0.4032664895057678,
      "learning_rate": 2e-05,
      "loss": 0.7542,
      "step": 3360
    },
    {
      "epoch": 0.27675125236100845,
      "grad_norm": 0.4187905490398407,
      "learning_rate": 2e-05,
      "loss": 0.7565,
      "step": 3370
    },
    {
      "epoch": 0.2775724726944239,
      "grad_norm": 0.4015965163707733,
      "learning_rate": 2e-05,
      "loss": 0.7796,
      "step": 3380
    },
    {
      "epoch": 0.2783936930278394,
      "grad_norm": 0.3947266638278961,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 3390
    },
    {
      "epoch": 0.2792149133612548,
      "grad_norm": 0.4074665307998657,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 3400
    },
    {
      "epoch": 0.2800361336946703,
      "grad_norm": 0.4260404109954834,
      "learning_rate": 2e-05,
      "loss": 0.7457,
      "step": 3410
    },
    {
      "epoch": 0.28085735402808576,
      "grad_norm": 0.4113856554031372,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 3420
    },
    {
      "epoch": 0.2816785743615012,
      "grad_norm": 0.4115002751350403,
      "learning_rate": 2e-05,
      "loss": 0.7862,
      "step": 3430
    },
    {
      "epoch": 0.28249979469491665,
      "grad_norm": 0.3859333395957947,
      "learning_rate": 2e-05,
      "loss": 0.7635,
      "step": 3440
    },
    {
      "epoch": 0.2833210150283321,
      "grad_norm": 0.42104247212409973,
      "learning_rate": 2e-05,
      "loss": 0.7484,
      "step": 3450
    },
    {
      "epoch": 0.28414223536174754,
      "grad_norm": 0.41294246912002563,
      "learning_rate": 2e-05,
      "loss": 0.7467,
      "step": 3460
    },
    {
      "epoch": 0.284963455695163,
      "grad_norm": 0.38498160243034363,
      "learning_rate": 2e-05,
      "loss": 0.7455,
      "step": 3470
    },
    {
      "epoch": 0.2857846760285785,
      "grad_norm": 0.5266622304916382,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 3480
    },
    {
      "epoch": 0.2866058963619939,
      "grad_norm": 0.44556841254234314,
      "learning_rate": 2e-05,
      "loss": 0.757,
      "step": 3490
    },
    {
      "epoch": 0.2874271166954094,
      "grad_norm": 0.404109925031662,
      "learning_rate": 2e-05,
      "loss": 0.7422,
      "step": 3500
    },
    {
      "epoch": 0.28824833702882485,
      "grad_norm": 0.41726675629615784,
      "learning_rate": 2e-05,
      "loss": 0.7566,
      "step": 3510
    },
    {
      "epoch": 0.28906955736224027,
      "grad_norm": 0.4020760655403137,
      "learning_rate": 2e-05,
      "loss": 0.7233,
      "step": 3520
    },
    {
      "epoch": 0.28989077769565574,
      "grad_norm": 0.3666248619556427,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 3530
    },
    {
      "epoch": 0.2907119980290712,
      "grad_norm": 0.4160138666629791,
      "learning_rate": 2e-05,
      "loss": 0.7494,
      "step": 3540
    },
    {
      "epoch": 0.29153321836248663,
      "grad_norm": 0.41128918528556824,
      "learning_rate": 2e-05,
      "loss": 0.7178,
      "step": 3550
    },
    {
      "epoch": 0.2923544386959021,
      "grad_norm": 0.4046924412250519,
      "learning_rate": 2e-05,
      "loss": 0.7597,
      "step": 3560
    },
    {
      "epoch": 0.2931756590293176,
      "grad_norm": 0.4178462028503418,
      "learning_rate": 2e-05,
      "loss": 0.7269,
      "step": 3570
    },
    {
      "epoch": 0.293996879362733,
      "grad_norm": 0.4053454101085663,
      "learning_rate": 2e-05,
      "loss": 0.7764,
      "step": 3580
    },
    {
      "epoch": 0.2948180996961485,
      "grad_norm": 0.41103944182395935,
      "learning_rate": 2e-05,
      "loss": 0.738,
      "step": 3590
    },
    {
      "epoch": 0.29563932002956395,
      "grad_norm": 0.39164984226226807,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 3600
    },
    {
      "epoch": 0.29646054036297936,
      "grad_norm": 0.40310123562812805,
      "learning_rate": 2e-05,
      "loss": 0.7531,
      "step": 3610
    },
    {
      "epoch": 0.29728176069639484,
      "grad_norm": 0.3818446099758148,
      "learning_rate": 2e-05,
      "loss": 0.7358,
      "step": 3620
    },
    {
      "epoch": 0.2981029810298103,
      "grad_norm": 0.38728272914886475,
      "learning_rate": 2e-05,
      "loss": 0.7559,
      "step": 3630
    },
    {
      "epoch": 0.29892420136322573,
      "grad_norm": 0.4156200587749481,
      "learning_rate": 2e-05,
      "loss": 0.7688,
      "step": 3640
    },
    {
      "epoch": 0.2997454216966412,
      "grad_norm": 0.38862425088882446,
      "learning_rate": 2e-05,
      "loss": 0.6984,
      "step": 3650
    },
    {
      "epoch": 0.3005666420300567,
      "grad_norm": 0.4446597397327423,
      "learning_rate": 2e-05,
      "loss": 0.7677,
      "step": 3660
    },
    {
      "epoch": 0.3013878623634721,
      "grad_norm": 0.41146260499954224,
      "learning_rate": 2e-05,
      "loss": 0.7135,
      "step": 3670
    },
    {
      "epoch": 0.30220908269688757,
      "grad_norm": 0.40543463826179504,
      "learning_rate": 2e-05,
      "loss": 0.7515,
      "step": 3680
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 0.39398670196533203,
      "learning_rate": 2e-05,
      "loss": 0.726,
      "step": 3690
    },
    {
      "epoch": 0.3038515233637185,
      "grad_norm": 0.40294191241264343,
      "learning_rate": 2e-05,
      "loss": 0.7501,
      "step": 3700
    },
    {
      "epoch": 0.30467274369713393,
      "grad_norm": 0.4296289086341858,
      "learning_rate": 2e-05,
      "loss": 0.7415,
      "step": 3710
    },
    {
      "epoch": 0.3054939640305494,
      "grad_norm": 0.4265771508216858,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 3720
    },
    {
      "epoch": 0.3063151843639649,
      "grad_norm": 0.39263609051704407,
      "learning_rate": 2e-05,
      "loss": 0.7566,
      "step": 3730
    },
    {
      "epoch": 0.3071364046973803,
      "grad_norm": 0.3951655924320221,
      "learning_rate": 2e-05,
      "loss": 0.7417,
      "step": 3740
    },
    {
      "epoch": 0.30795762503079577,
      "grad_norm": 0.4019796848297119,
      "learning_rate": 2e-05,
      "loss": 0.73,
      "step": 3750
    },
    {
      "epoch": 0.30877884536421124,
      "grad_norm": 0.43790656328201294,
      "learning_rate": 2e-05,
      "loss": 0.7316,
      "step": 3760
    },
    {
      "epoch": 0.30960006569762666,
      "grad_norm": 0.394927978515625,
      "learning_rate": 2e-05,
      "loss": 0.7432,
      "step": 3770
    },
    {
      "epoch": 0.31042128603104213,
      "grad_norm": 0.3678690195083618,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 3780
    },
    {
      "epoch": 0.3112425063644576,
      "grad_norm": 0.39598825573921204,
      "learning_rate": 2e-05,
      "loss": 0.7573,
      "step": 3790
    },
    {
      "epoch": 0.312063726697873,
      "grad_norm": 0.39488571882247925,
      "learning_rate": 2e-05,
      "loss": 0.7341,
      "step": 3800
    },
    {
      "epoch": 0.3128849470312885,
      "grad_norm": 0.39103806018829346,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 3810
    },
    {
      "epoch": 0.31370616736470397,
      "grad_norm": 0.3993924856185913,
      "learning_rate": 2e-05,
      "loss": 0.7205,
      "step": 3820
    },
    {
      "epoch": 0.3145273876981194,
      "grad_norm": 0.3883530795574188,
      "learning_rate": 2e-05,
      "loss": 0.7331,
      "step": 3830
    },
    {
      "epoch": 0.31534860803153486,
      "grad_norm": 0.40700817108154297,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 3840
    },
    {
      "epoch": 0.31616982836495033,
      "grad_norm": 0.42441993951797485,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 3850
    },
    {
      "epoch": 0.31699104869836575,
      "grad_norm": 0.37016576528549194,
      "learning_rate": 2e-05,
      "loss": 0.7346,
      "step": 3860
    },
    {
      "epoch": 0.3178122690317812,
      "grad_norm": 0.41952431201934814,
      "learning_rate": 2e-05,
      "loss": 0.7335,
      "step": 3870
    },
    {
      "epoch": 0.3186334893651967,
      "grad_norm": 0.40923693776130676,
      "learning_rate": 2e-05,
      "loss": 0.7778,
      "step": 3880
    },
    {
      "epoch": 0.3194547096986121,
      "grad_norm": 0.410786896944046,
      "learning_rate": 2e-05,
      "loss": 0.7388,
      "step": 3890
    },
    {
      "epoch": 0.3202759300320276,
      "grad_norm": 0.42773470282554626,
      "learning_rate": 2e-05,
      "loss": 0.7704,
      "step": 3900
    },
    {
      "epoch": 0.32109715036544306,
      "grad_norm": 0.42799991369247437,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 3910
    },
    {
      "epoch": 0.3219183706988585,
      "grad_norm": 0.39419862627983093,
      "learning_rate": 2e-05,
      "loss": 0.729,
      "step": 3920
    },
    {
      "epoch": 0.32273959103227395,
      "grad_norm": 0.4130425751209259,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 3930
    },
    {
      "epoch": 0.3235608113656894,
      "grad_norm": 0.36902064085006714,
      "learning_rate": 2e-05,
      "loss": 0.7039,
      "step": 3940
    },
    {
      "epoch": 0.32438203169910484,
      "grad_norm": 0.39881351590156555,
      "learning_rate": 2e-05,
      "loss": 0.7312,
      "step": 3950
    },
    {
      "epoch": 0.3252032520325203,
      "grad_norm": 0.38550812005996704,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 3960
    },
    {
      "epoch": 0.3260244723659358,
      "grad_norm": 0.37750986218452454,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 3970
    },
    {
      "epoch": 0.3268456926993512,
      "grad_norm": 0.3967691659927368,
      "learning_rate": 2e-05,
      "loss": 0.6914,
      "step": 3980
    },
    {
      "epoch": 0.3276669130327667,
      "grad_norm": 0.4260273873806,
      "learning_rate": 2e-05,
      "loss": 0.7159,
      "step": 3990
    },
    {
      "epoch": 0.32848813336618216,
      "grad_norm": 0.4173099100589752,
      "learning_rate": 2e-05,
      "loss": 0.7612,
      "step": 4000
    },
    {
      "epoch": 0.32930935369959763,
      "grad_norm": 0.4106500446796417,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 4010
    },
    {
      "epoch": 0.33013057403301305,
      "grad_norm": 0.4011145830154419,
      "learning_rate": 2e-05,
      "loss": 0.7326,
      "step": 4020
    },
    {
      "epoch": 0.3309517943664285,
      "grad_norm": 0.39498040080070496,
      "learning_rate": 2e-05,
      "loss": 0.7565,
      "step": 4030
    },
    {
      "epoch": 0.331773014699844,
      "grad_norm": 0.39776620268821716,
      "learning_rate": 2e-05,
      "loss": 0.7228,
      "step": 4040
    },
    {
      "epoch": 0.3325942350332594,
      "grad_norm": 0.404032826423645,
      "learning_rate": 2e-05,
      "loss": 0.764,
      "step": 4050
    },
    {
      "epoch": 0.3334154553666749,
      "grad_norm": 0.44440531730651855,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 4060
    },
    {
      "epoch": 0.33423667570009036,
      "grad_norm": 0.3921901285648346,
      "learning_rate": 2e-05,
      "loss": 0.712,
      "step": 4070
    },
    {
      "epoch": 0.3350578960335058,
      "grad_norm": 0.3711211681365967,
      "learning_rate": 2e-05,
      "loss": 0.7323,
      "step": 4080
    },
    {
      "epoch": 0.33587911636692125,
      "grad_norm": 0.37448689341545105,
      "learning_rate": 2e-05,
      "loss": 0.7488,
      "step": 4090
    },
    {
      "epoch": 0.3367003367003367,
      "grad_norm": 0.3964804708957672,
      "learning_rate": 2e-05,
      "loss": 0.7487,
      "step": 4100
    },
    {
      "epoch": 0.33752155703375214,
      "grad_norm": 0.39893484115600586,
      "learning_rate": 2e-05,
      "loss": 0.739,
      "step": 4110
    },
    {
      "epoch": 0.3383427773671676,
      "grad_norm": 0.4004378616809845,
      "learning_rate": 2e-05,
      "loss": 0.7432,
      "step": 4120
    },
    {
      "epoch": 0.3391639977005831,
      "grad_norm": 0.3802061378955841,
      "learning_rate": 2e-05,
      "loss": 0.7375,
      "step": 4130
    },
    {
      "epoch": 0.3399852180339985,
      "grad_norm": 0.38909509778022766,
      "learning_rate": 2e-05,
      "loss": 0.7514,
      "step": 4140
    },
    {
      "epoch": 0.340806438367414,
      "grad_norm": 0.3952394127845764,
      "learning_rate": 2e-05,
      "loss": 0.7501,
      "step": 4150
    },
    {
      "epoch": 0.34162765870082945,
      "grad_norm": 0.3679156005382538,
      "learning_rate": 2e-05,
      "loss": 0.7326,
      "step": 4160
    },
    {
      "epoch": 0.34244887903424487,
      "grad_norm": 0.38880136609077454,
      "learning_rate": 2e-05,
      "loss": 0.7215,
      "step": 4170
    },
    {
      "epoch": 0.34327009936766034,
      "grad_norm": 0.3892967104911804,
      "learning_rate": 2e-05,
      "loss": 0.7299,
      "step": 4180
    },
    {
      "epoch": 0.3440913197010758,
      "grad_norm": 0.39590567350387573,
      "learning_rate": 2e-05,
      "loss": 0.7723,
      "step": 4190
    },
    {
      "epoch": 0.34491254003449123,
      "grad_norm": 0.4128817021846771,
      "learning_rate": 2e-05,
      "loss": 0.7617,
      "step": 4200
    },
    {
      "epoch": 0.3457337603679067,
      "grad_norm": 0.3808560371398926,
      "learning_rate": 2e-05,
      "loss": 0.7648,
      "step": 4210
    },
    {
      "epoch": 0.3465549807013222,
      "grad_norm": 0.3992762565612793,
      "learning_rate": 2e-05,
      "loss": 0.7391,
      "step": 4220
    },
    {
      "epoch": 0.3473762010347376,
      "grad_norm": 0.40010175108909607,
      "learning_rate": 2e-05,
      "loss": 0.7182,
      "step": 4230
    },
    {
      "epoch": 0.34819742136815307,
      "grad_norm": 0.37894871830940247,
      "learning_rate": 2e-05,
      "loss": 0.7328,
      "step": 4240
    },
    {
      "epoch": 0.34901864170156854,
      "grad_norm": 0.4118288457393646,
      "learning_rate": 2e-05,
      "loss": 0.7314,
      "step": 4250
    },
    {
      "epoch": 0.34983986203498396,
      "grad_norm": 0.3993299901485443,
      "learning_rate": 2e-05,
      "loss": 0.7637,
      "step": 4260
    },
    {
      "epoch": 0.35066108236839943,
      "grad_norm": 0.4058755040168762,
      "learning_rate": 2e-05,
      "loss": 0.7551,
      "step": 4270
    },
    {
      "epoch": 0.3514823027018149,
      "grad_norm": 0.4023033678531647,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 4280
    },
    {
      "epoch": 0.3523035230352303,
      "grad_norm": 0.3920232653617859,
      "learning_rate": 2e-05,
      "loss": 0.734,
      "step": 4290
    },
    {
      "epoch": 0.3531247433686458,
      "grad_norm": 0.38466572761535645,
      "learning_rate": 2e-05,
      "loss": 0.6928,
      "step": 4300
    },
    {
      "epoch": 0.3539459637020613,
      "grad_norm": 0.380645215511322,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 4310
    },
    {
      "epoch": 0.35476718403547675,
      "grad_norm": 0.4126320779323578,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 4320
    },
    {
      "epoch": 0.35558840436889216,
      "grad_norm": 0.35835152864456177,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 4330
    },
    {
      "epoch": 0.35640962470230764,
      "grad_norm": 0.388376384973526,
      "learning_rate": 2e-05,
      "loss": 0.7191,
      "step": 4340
    },
    {
      "epoch": 0.3572308450357231,
      "grad_norm": 0.4169785976409912,
      "learning_rate": 2e-05,
      "loss": 0.7264,
      "step": 4350
    },
    {
      "epoch": 0.35805206536913853,
      "grad_norm": 0.3863677978515625,
      "learning_rate": 2e-05,
      "loss": 0.747,
      "step": 4360
    },
    {
      "epoch": 0.358873285702554,
      "grad_norm": 0.3928108811378479,
      "learning_rate": 2e-05,
      "loss": 0.7242,
      "step": 4370
    },
    {
      "epoch": 0.3596945060359695,
      "grad_norm": 0.3956344723701477,
      "learning_rate": 2e-05,
      "loss": 0.7346,
      "step": 4380
    },
    {
      "epoch": 0.3605157263693849,
      "grad_norm": 0.4082147181034088,
      "learning_rate": 2e-05,
      "loss": 0.7504,
      "step": 4390
    },
    {
      "epoch": 0.36133694670280037,
      "grad_norm": 0.3981010317802429,
      "learning_rate": 2e-05,
      "loss": 0.7594,
      "step": 4400
    },
    {
      "epoch": 0.36215816703621584,
      "grad_norm": 0.3951975703239441,
      "learning_rate": 2e-05,
      "loss": 0.719,
      "step": 4410
    },
    {
      "epoch": 0.36297938736963126,
      "grad_norm": 0.42050042748451233,
      "learning_rate": 2e-05,
      "loss": 0.7175,
      "step": 4420
    },
    {
      "epoch": 0.36380060770304673,
      "grad_norm": 0.41786742210388184,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 4430
    },
    {
      "epoch": 0.3646218280364622,
      "grad_norm": 0.3928516209125519,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 4440
    },
    {
      "epoch": 0.3654430483698776,
      "grad_norm": 0.4063854515552521,
      "learning_rate": 2e-05,
      "loss": 0.7139,
      "step": 4450
    },
    {
      "epoch": 0.3662642687032931,
      "grad_norm": 0.3987753391265869,
      "learning_rate": 2e-05,
      "loss": 0.714,
      "step": 4460
    },
    {
      "epoch": 0.36708548903670857,
      "grad_norm": 0.39688828587532043,
      "learning_rate": 2e-05,
      "loss": 0.7436,
      "step": 4470
    },
    {
      "epoch": 0.367906709370124,
      "grad_norm": 0.3865756392478943,
      "learning_rate": 2e-05,
      "loss": 0.7323,
      "step": 4480
    },
    {
      "epoch": 0.36872792970353946,
      "grad_norm": 0.3768831491470337,
      "learning_rate": 2e-05,
      "loss": 0.761,
      "step": 4490
    },
    {
      "epoch": 0.36954915003695493,
      "grad_norm": 0.39193153381347656,
      "learning_rate": 2e-05,
      "loss": 0.7152,
      "step": 4500
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 0.40094587206840515,
      "learning_rate": 2e-05,
      "loss": 0.7258,
      "step": 4510
    },
    {
      "epoch": 0.3711915907037858,
      "grad_norm": 0.39501965045928955,
      "learning_rate": 2e-05,
      "loss": 0.7179,
      "step": 4520
    },
    {
      "epoch": 0.3720128110372013,
      "grad_norm": 0.393291711807251,
      "learning_rate": 2e-05,
      "loss": 0.7345,
      "step": 4530
    },
    {
      "epoch": 0.3728340313706167,
      "grad_norm": 0.395773708820343,
      "learning_rate": 2e-05,
      "loss": 0.7121,
      "step": 4540
    },
    {
      "epoch": 0.3736552517040322,
      "grad_norm": 0.41279321908950806,
      "learning_rate": 2e-05,
      "loss": 0.7447,
      "step": 4550
    },
    {
      "epoch": 0.37447647203744766,
      "grad_norm": 0.39585351943969727,
      "learning_rate": 2e-05,
      "loss": 0.7578,
      "step": 4560
    },
    {
      "epoch": 0.3752976923708631,
      "grad_norm": 0.40861478447914124,
      "learning_rate": 2e-05,
      "loss": 0.7081,
      "step": 4570
    },
    {
      "epoch": 0.37611891270427855,
      "grad_norm": 0.39568030834198,
      "learning_rate": 2e-05,
      "loss": 0.7104,
      "step": 4580
    },
    {
      "epoch": 0.376940133037694,
      "grad_norm": 0.4339118003845215,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 4590
    },
    {
      "epoch": 0.37776135337110944,
      "grad_norm": 0.4051828384399414,
      "learning_rate": 2e-05,
      "loss": 0.7267,
      "step": 4600
    },
    {
      "epoch": 0.3785825737045249,
      "grad_norm": 0.4043790102005005,
      "learning_rate": 2e-05,
      "loss": 0.7513,
      "step": 4610
    },
    {
      "epoch": 0.3794037940379404,
      "grad_norm": 0.3923429548740387,
      "learning_rate": 2e-05,
      "loss": 0.7102,
      "step": 4620
    },
    {
      "epoch": 0.3802250143713558,
      "grad_norm": 0.4058830440044403,
      "learning_rate": 2e-05,
      "loss": 0.7415,
      "step": 4630
    },
    {
      "epoch": 0.3810462347047713,
      "grad_norm": 0.3877350389957428,
      "learning_rate": 2e-05,
      "loss": 0.7502,
      "step": 4640
    },
    {
      "epoch": 0.38186745503818675,
      "grad_norm": 0.4066460132598877,
      "learning_rate": 2e-05,
      "loss": 0.7486,
      "step": 4650
    },
    {
      "epoch": 0.3826886753716022,
      "grad_norm": 0.39090630412101746,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 4660
    },
    {
      "epoch": 0.38350989570501764,
      "grad_norm": 0.4208731949329376,
      "learning_rate": 2e-05,
      "loss": 0.7326,
      "step": 4670
    },
    {
      "epoch": 0.3843311160384331,
      "grad_norm": 0.3768414855003357,
      "learning_rate": 2e-05,
      "loss": 0.7176,
      "step": 4680
    },
    {
      "epoch": 0.3851523363718486,
      "grad_norm": 0.38560089468955994,
      "learning_rate": 2e-05,
      "loss": 0.7298,
      "step": 4690
    },
    {
      "epoch": 0.385973556705264,
      "grad_norm": 0.3999074697494507,
      "learning_rate": 2e-05,
      "loss": 0.7342,
      "step": 4700
    },
    {
      "epoch": 0.3867947770386795,
      "grad_norm": 0.3930422067642212,
      "learning_rate": 2e-05,
      "loss": 0.7167,
      "step": 4710
    },
    {
      "epoch": 0.38761599737209496,
      "grad_norm": 0.3955245316028595,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 4720
    },
    {
      "epoch": 0.3884372177055104,
      "grad_norm": 0.4193728268146515,
      "learning_rate": 2e-05,
      "loss": 0.7228,
      "step": 4730
    },
    {
      "epoch": 0.38925843803892585,
      "grad_norm": 0.40219366550445557,
      "learning_rate": 2e-05,
      "loss": 0.7191,
      "step": 4740
    },
    {
      "epoch": 0.3900796583723413,
      "grad_norm": 0.3611651360988617,
      "learning_rate": 2e-05,
      "loss": 0.7213,
      "step": 4750
    },
    {
      "epoch": 0.39090087870575674,
      "grad_norm": 0.3882848024368286,
      "learning_rate": 2e-05,
      "loss": 0.7355,
      "step": 4760
    },
    {
      "epoch": 0.3917220990391722,
      "grad_norm": 0.3923344910144806,
      "learning_rate": 2e-05,
      "loss": 0.7366,
      "step": 4770
    },
    {
      "epoch": 0.3925433193725877,
      "grad_norm": 0.4193645715713501,
      "learning_rate": 2e-05,
      "loss": 0.7675,
      "step": 4780
    },
    {
      "epoch": 0.3933645397060031,
      "grad_norm": 0.3620869815349579,
      "learning_rate": 2e-05,
      "loss": 0.7006,
      "step": 4790
    },
    {
      "epoch": 0.3941857600394186,
      "grad_norm": 0.3968870937824249,
      "learning_rate": 2e-05,
      "loss": 0.7243,
      "step": 4800
    },
    {
      "epoch": 0.39500698037283405,
      "grad_norm": 0.405109703540802,
      "learning_rate": 2e-05,
      "loss": 0.7425,
      "step": 4810
    },
    {
      "epoch": 0.39582820070624947,
      "grad_norm": 0.3702356815338135,
      "learning_rate": 2e-05,
      "loss": 0.7506,
      "step": 4820
    },
    {
      "epoch": 0.39664942103966494,
      "grad_norm": 0.39436784386634827,
      "learning_rate": 2e-05,
      "loss": 0.727,
      "step": 4830
    },
    {
      "epoch": 0.3974706413730804,
      "grad_norm": 0.4006347954273224,
      "learning_rate": 2e-05,
      "loss": 0.7173,
      "step": 4840
    },
    {
      "epoch": 0.39829186170649583,
      "grad_norm": 0.38862812519073486,
      "learning_rate": 2e-05,
      "loss": 0.7253,
      "step": 4850
    },
    {
      "epoch": 0.3991130820399113,
      "grad_norm": 0.39549189805984497,
      "learning_rate": 2e-05,
      "loss": 0.7454,
      "step": 4860
    },
    {
      "epoch": 0.3999343023733268,
      "grad_norm": 0.3807682693004608,
      "learning_rate": 2e-05,
      "loss": 0.7217,
      "step": 4870
    },
    {
      "epoch": 0.4007555227067422,
      "grad_norm": 0.42229798436164856,
      "learning_rate": 2e-05,
      "loss": 0.7529,
      "step": 4880
    },
    {
      "epoch": 0.40157674304015767,
      "grad_norm": 0.4186745285987854,
      "learning_rate": 2e-05,
      "loss": 0.7246,
      "step": 4890
    },
    {
      "epoch": 0.40239796337357314,
      "grad_norm": 0.3918178379535675,
      "learning_rate": 2e-05,
      "loss": 0.7472,
      "step": 4900
    },
    {
      "epoch": 0.40321918370698856,
      "grad_norm": 0.39102402329444885,
      "learning_rate": 2e-05,
      "loss": 0.7304,
      "step": 4910
    },
    {
      "epoch": 0.40404040404040403,
      "grad_norm": 0.40069061517715454,
      "learning_rate": 2e-05,
      "loss": 0.7277,
      "step": 4920
    },
    {
      "epoch": 0.4048616243738195,
      "grad_norm": 0.37603867053985596,
      "learning_rate": 2e-05,
      "loss": 0.6822,
      "step": 4930
    },
    {
      "epoch": 0.4056828447072349,
      "grad_norm": 0.36831676959991455,
      "learning_rate": 2e-05,
      "loss": 0.7208,
      "step": 4940
    },
    {
      "epoch": 0.4065040650406504,
      "grad_norm": 0.3840339481830597,
      "learning_rate": 2e-05,
      "loss": 0.7429,
      "step": 4950
    },
    {
      "epoch": 0.40732528537406587,
      "grad_norm": 0.38564029335975647,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 4960
    },
    {
      "epoch": 0.40814650570748134,
      "grad_norm": 0.3892430365085602,
      "learning_rate": 2e-05,
      "loss": 0.7324,
      "step": 4970
    },
    {
      "epoch": 0.40896772604089676,
      "grad_norm": 0.38965925574302673,
      "learning_rate": 2e-05,
      "loss": 0.7067,
      "step": 4980
    },
    {
      "epoch": 0.40978894637431224,
      "grad_norm": 0.4817599058151245,
      "learning_rate": 2e-05,
      "loss": 0.7347,
      "step": 4990
    },
    {
      "epoch": 0.4106101667077277,
      "grad_norm": 0.36471059918403625,
      "learning_rate": 2e-05,
      "loss": 0.7254,
      "step": 5000
    },
    {
      "epoch": 0.4114313870411431,
      "grad_norm": 0.3769645690917969,
      "learning_rate": 2e-05,
      "loss": 0.7357,
      "step": 5010
    },
    {
      "epoch": 0.4122526073745586,
      "grad_norm": 0.3883325457572937,
      "learning_rate": 2e-05,
      "loss": 0.7479,
      "step": 5020
    },
    {
      "epoch": 0.4130738277079741,
      "grad_norm": 0.3702440857887268,
      "learning_rate": 2e-05,
      "loss": 0.723,
      "step": 5030
    },
    {
      "epoch": 0.4138950480413895,
      "grad_norm": 0.39130136370658875,
      "learning_rate": 2e-05,
      "loss": 0.7339,
      "step": 5040
    },
    {
      "epoch": 0.41471626837480496,
      "grad_norm": 0.35221561789512634,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 5050
    },
    {
      "epoch": 0.41553748870822044,
      "grad_norm": 0.40981602668762207,
      "learning_rate": 2e-05,
      "loss": 0.737,
      "step": 5060
    },
    {
      "epoch": 0.41635870904163585,
      "grad_norm": 0.3841480612754822,
      "learning_rate": 2e-05,
      "loss": 0.7429,
      "step": 5070
    },
    {
      "epoch": 0.41717992937505133,
      "grad_norm": 0.4221237599849701,
      "learning_rate": 2e-05,
      "loss": 0.7173,
      "step": 5080
    },
    {
      "epoch": 0.4180011497084668,
      "grad_norm": 0.35533520579338074,
      "learning_rate": 2e-05,
      "loss": 0.7399,
      "step": 5090
    },
    {
      "epoch": 0.4188223700418822,
      "grad_norm": 0.43763330578804016,
      "learning_rate": 2e-05,
      "loss": 0.7596,
      "step": 5100
    },
    {
      "epoch": 0.4196435903752977,
      "grad_norm": 0.39289507269859314,
      "learning_rate": 2e-05,
      "loss": 0.7182,
      "step": 5110
    },
    {
      "epoch": 0.42046481070871317,
      "grad_norm": 0.3874718248844147,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 5120
    },
    {
      "epoch": 0.4212860310421286,
      "grad_norm": 0.3871784508228302,
      "learning_rate": 2e-05,
      "loss": 0.7446,
      "step": 5130
    },
    {
      "epoch": 0.42210725137554406,
      "grad_norm": 0.4105471670627594,
      "learning_rate": 2e-05,
      "loss": 0.6966,
      "step": 5140
    },
    {
      "epoch": 0.42292847170895953,
      "grad_norm": 0.4011363387107849,
      "learning_rate": 2e-05,
      "loss": 0.7067,
      "step": 5150
    },
    {
      "epoch": 0.42374969204237495,
      "grad_norm": 0.41625866293907166,
      "learning_rate": 2e-05,
      "loss": 0.7308,
      "step": 5160
    },
    {
      "epoch": 0.4245709123757904,
      "grad_norm": 0.4366151988506317,
      "learning_rate": 2e-05,
      "loss": 0.7565,
      "step": 5170
    },
    {
      "epoch": 0.4253921327092059,
      "grad_norm": 0.39051303267478943,
      "learning_rate": 2e-05,
      "loss": 0.7552,
      "step": 5180
    },
    {
      "epoch": 0.4262133530426213,
      "grad_norm": 0.38735073804855347,
      "learning_rate": 2e-05,
      "loss": 0.7201,
      "step": 5190
    },
    {
      "epoch": 0.4270345733760368,
      "grad_norm": 0.37748709321022034,
      "learning_rate": 2e-05,
      "loss": 0.7373,
      "step": 5200
    },
    {
      "epoch": 0.42785579370945226,
      "grad_norm": 0.4150856137275696,
      "learning_rate": 2e-05,
      "loss": 0.7165,
      "step": 5210
    },
    {
      "epoch": 0.4286770140428677,
      "grad_norm": 0.383572518825531,
      "learning_rate": 2e-05,
      "loss": 0.7315,
      "step": 5220
    },
    {
      "epoch": 0.42949823437628315,
      "grad_norm": 0.38993608951568604,
      "learning_rate": 2e-05,
      "loss": 0.6888,
      "step": 5230
    },
    {
      "epoch": 0.4303194547096986,
      "grad_norm": 0.40374061465263367,
      "learning_rate": 2e-05,
      "loss": 0.7159,
      "step": 5240
    },
    {
      "epoch": 0.43114067504311404,
      "grad_norm": 0.3692219853401184,
      "learning_rate": 2e-05,
      "loss": 0.7167,
      "step": 5250
    },
    {
      "epoch": 0.4319618953765295,
      "grad_norm": 0.4148082435131073,
      "learning_rate": 2e-05,
      "loss": 0.7223,
      "step": 5260
    },
    {
      "epoch": 0.432783115709945,
      "grad_norm": 0.38898903131484985,
      "learning_rate": 2e-05,
      "loss": 0.7197,
      "step": 5270
    },
    {
      "epoch": 0.43360433604336046,
      "grad_norm": 0.37857526540756226,
      "learning_rate": 2e-05,
      "loss": 0.7147,
      "step": 5280
    },
    {
      "epoch": 0.4344255563767759,
      "grad_norm": 0.38649624586105347,
      "learning_rate": 2e-05,
      "loss": 0.7165,
      "step": 5290
    },
    {
      "epoch": 0.43524677671019135,
      "grad_norm": 0.39216694235801697,
      "learning_rate": 2e-05,
      "loss": 0.7381,
      "step": 5300
    },
    {
      "epoch": 0.4360679970436068,
      "grad_norm": 0.3968610167503357,
      "learning_rate": 2e-05,
      "loss": 0.7164,
      "step": 5310
    },
    {
      "epoch": 0.43688921737702224,
      "grad_norm": 0.4204171299934387,
      "learning_rate": 2e-05,
      "loss": 0.714,
      "step": 5320
    },
    {
      "epoch": 0.4377104377104377,
      "grad_norm": 0.3884009122848511,
      "learning_rate": 2e-05,
      "loss": 0.73,
      "step": 5330
    },
    {
      "epoch": 0.4385316580438532,
      "grad_norm": 0.4209270179271698,
      "learning_rate": 2e-05,
      "loss": 0.7437,
      "step": 5340
    },
    {
      "epoch": 0.4393528783772686,
      "grad_norm": 0.3883524537086487,
      "learning_rate": 2e-05,
      "loss": 0.7093,
      "step": 5350
    },
    {
      "epoch": 0.4401740987106841,
      "grad_norm": 0.4279435873031616,
      "learning_rate": 2e-05,
      "loss": 0.7611,
      "step": 5360
    },
    {
      "epoch": 0.44099531904409955,
      "grad_norm": 0.3872007429599762,
      "learning_rate": 2e-05,
      "loss": 0.7266,
      "step": 5370
    },
    {
      "epoch": 0.44181653937751497,
      "grad_norm": 0.4002803862094879,
      "learning_rate": 2e-05,
      "loss": 0.7091,
      "step": 5380
    },
    {
      "epoch": 0.44263775971093045,
      "grad_norm": 0.41259849071502686,
      "learning_rate": 2e-05,
      "loss": 0.7258,
      "step": 5390
    },
    {
      "epoch": 0.4434589800443459,
      "grad_norm": 0.38737940788269043,
      "learning_rate": 2e-05,
      "loss": 0.7309,
      "step": 5400
    },
    {
      "epoch": 0.44428020037776134,
      "grad_norm": 0.3710504174232483,
      "learning_rate": 2e-05,
      "loss": 0.7409,
      "step": 5410
    },
    {
      "epoch": 0.4451014207111768,
      "grad_norm": 0.4141605496406555,
      "learning_rate": 2e-05,
      "loss": 0.7507,
      "step": 5420
    },
    {
      "epoch": 0.4459226410445923,
      "grad_norm": 0.38736847043037415,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 5430
    },
    {
      "epoch": 0.4467438613780077,
      "grad_norm": 0.38136526942253113,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 5440
    },
    {
      "epoch": 0.4475650817114232,
      "grad_norm": 0.37488067150115967,
      "learning_rate": 2e-05,
      "loss": 0.7099,
      "step": 5450
    },
    {
      "epoch": 0.44838630204483865,
      "grad_norm": 0.3839886486530304,
      "learning_rate": 2e-05,
      "loss": 0.7421,
      "step": 5460
    },
    {
      "epoch": 0.44920752237825406,
      "grad_norm": 0.4002284109592438,
      "learning_rate": 2e-05,
      "loss": 0.7508,
      "step": 5470
    },
    {
      "epoch": 0.45002874271166954,
      "grad_norm": 0.3969842493534088,
      "learning_rate": 2e-05,
      "loss": 0.7451,
      "step": 5480
    },
    {
      "epoch": 0.450849963045085,
      "grad_norm": 0.3897230327129364,
      "learning_rate": 2e-05,
      "loss": 0.7149,
      "step": 5490
    },
    {
      "epoch": 0.45167118337850043,
      "grad_norm": 0.3789793848991394,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 5500
    },
    {
      "epoch": 0.4524924037119159,
      "grad_norm": 0.39217594265937805,
      "learning_rate": 2e-05,
      "loss": 0.743,
      "step": 5510
    },
    {
      "epoch": 0.4533136240453314,
      "grad_norm": 0.37494194507598877,
      "learning_rate": 2e-05,
      "loss": 0.7361,
      "step": 5520
    },
    {
      "epoch": 0.4541348443787468,
      "grad_norm": 0.3946336507797241,
      "learning_rate": 2e-05,
      "loss": 0.7157,
      "step": 5530
    },
    {
      "epoch": 0.45495606471216227,
      "grad_norm": 0.389522910118103,
      "learning_rate": 2e-05,
      "loss": 0.731,
      "step": 5540
    },
    {
      "epoch": 0.45577728504557774,
      "grad_norm": 0.3856944739818573,
      "learning_rate": 2e-05,
      "loss": 0.7037,
      "step": 5550
    },
    {
      "epoch": 0.45659850537899316,
      "grad_norm": 0.3818206787109375,
      "learning_rate": 2e-05,
      "loss": 0.7374,
      "step": 5560
    },
    {
      "epoch": 0.45741972571240863,
      "grad_norm": 0.39970195293426514,
      "learning_rate": 2e-05,
      "loss": 0.7269,
      "step": 5570
    },
    {
      "epoch": 0.4582409460458241,
      "grad_norm": 0.41426146030426025,
      "learning_rate": 2e-05,
      "loss": 0.7353,
      "step": 5580
    },
    {
      "epoch": 0.4590621663792395,
      "grad_norm": 0.4082249104976654,
      "learning_rate": 2e-05,
      "loss": 0.7537,
      "step": 5590
    },
    {
      "epoch": 0.459883386712655,
      "grad_norm": 0.4322008788585663,
      "learning_rate": 2e-05,
      "loss": 0.725,
      "step": 5600
    },
    {
      "epoch": 0.46070460704607047,
      "grad_norm": 0.40711671113967896,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 5610
    },
    {
      "epoch": 0.46152582737948594,
      "grad_norm": 0.36283454298973083,
      "learning_rate": 2e-05,
      "loss": 0.6883,
      "step": 5620
    },
    {
      "epoch": 0.46234704771290136,
      "grad_norm": 0.40201500058174133,
      "learning_rate": 2e-05,
      "loss": 0.7212,
      "step": 5630
    },
    {
      "epoch": 0.46316826804631683,
      "grad_norm": 0.3914097845554352,
      "learning_rate": 2e-05,
      "loss": 0.7217,
      "step": 5640
    },
    {
      "epoch": 0.4639894883797323,
      "grad_norm": 0.3976267874240875,
      "learning_rate": 2e-05,
      "loss": 0.7404,
      "step": 5650
    },
    {
      "epoch": 0.4648107087131477,
      "grad_norm": 0.4136793911457062,
      "learning_rate": 2e-05,
      "loss": 0.7254,
      "step": 5660
    },
    {
      "epoch": 0.4656319290465632,
      "grad_norm": 0.3906679153442383,
      "learning_rate": 2e-05,
      "loss": 0.7423,
      "step": 5670
    },
    {
      "epoch": 0.46645314937997867,
      "grad_norm": 0.38211697340011597,
      "learning_rate": 2e-05,
      "loss": 0.7091,
      "step": 5680
    },
    {
      "epoch": 0.4672743697133941,
      "grad_norm": 0.3781614303588867,
      "learning_rate": 2e-05,
      "loss": 0.7401,
      "step": 5690
    },
    {
      "epoch": 0.46809559004680956,
      "grad_norm": 0.39131051301956177,
      "learning_rate": 2e-05,
      "loss": 0.7467,
      "step": 5700
    },
    {
      "epoch": 0.46891681038022504,
      "grad_norm": 0.380958616733551,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 5710
    },
    {
      "epoch": 0.46973803071364045,
      "grad_norm": 0.4028722643852234,
      "learning_rate": 2e-05,
      "loss": 0.7265,
      "step": 5720
    },
    {
      "epoch": 0.4705592510470559,
      "grad_norm": 0.40671348571777344,
      "learning_rate": 2e-05,
      "loss": 0.7497,
      "step": 5730
    },
    {
      "epoch": 0.4713804713804714,
      "grad_norm": 0.4059866964817047,
      "learning_rate": 2e-05,
      "loss": 0.7501,
      "step": 5740
    },
    {
      "epoch": 0.4722016917138868,
      "grad_norm": 0.41695985198020935,
      "learning_rate": 2e-05,
      "loss": 0.7305,
      "step": 5750
    },
    {
      "epoch": 0.4730229120473023,
      "grad_norm": 0.4036257863044739,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 5760
    },
    {
      "epoch": 0.47384413238071776,
      "grad_norm": 0.36903679370880127,
      "learning_rate": 2e-05,
      "loss": 0.7292,
      "step": 5770
    },
    {
      "epoch": 0.4746653527141332,
      "grad_norm": 0.3649754226207733,
      "learning_rate": 2e-05,
      "loss": 0.7378,
      "step": 5780
    },
    {
      "epoch": 0.47548657304754866,
      "grad_norm": 0.38199976086616516,
      "learning_rate": 2e-05,
      "loss": 0.725,
      "step": 5790
    },
    {
      "epoch": 0.47630779338096413,
      "grad_norm": 0.39438048005104065,
      "learning_rate": 2e-05,
      "loss": 0.7052,
      "step": 5800
    },
    {
      "epoch": 0.47712901371437955,
      "grad_norm": 0.40481430292129517,
      "learning_rate": 2e-05,
      "loss": 0.7255,
      "step": 5810
    },
    {
      "epoch": 0.477950234047795,
      "grad_norm": 0.3679243326187134,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 5820
    },
    {
      "epoch": 0.4787714543812105,
      "grad_norm": 0.4031321406364441,
      "learning_rate": 2e-05,
      "loss": 0.7241,
      "step": 5830
    },
    {
      "epoch": 0.4795926747146259,
      "grad_norm": 0.4169681966304779,
      "learning_rate": 2e-05,
      "loss": 0.7298,
      "step": 5840
    },
    {
      "epoch": 0.4804138950480414,
      "grad_norm": 0.35515090823173523,
      "learning_rate": 2e-05,
      "loss": 0.7294,
      "step": 5850
    },
    {
      "epoch": 0.48123511538145686,
      "grad_norm": 0.433719277381897,
      "learning_rate": 2e-05,
      "loss": 0.732,
      "step": 5860
    },
    {
      "epoch": 0.4820563357148723,
      "grad_norm": 0.4044643044471741,
      "learning_rate": 2e-05,
      "loss": 0.7193,
      "step": 5870
    },
    {
      "epoch": 0.48287755604828775,
      "grad_norm": 0.35301753878593445,
      "learning_rate": 2e-05,
      "loss": 0.694,
      "step": 5880
    },
    {
      "epoch": 0.4836987763817032,
      "grad_norm": 0.37355056405067444,
      "learning_rate": 2e-05,
      "loss": 0.717,
      "step": 5890
    },
    {
      "epoch": 0.48451999671511864,
      "grad_norm": 0.48972272872924805,
      "learning_rate": 2e-05,
      "loss": 0.7206,
      "step": 5900
    },
    {
      "epoch": 0.4853412170485341,
      "grad_norm": 0.3847252428531647,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 5910
    },
    {
      "epoch": 0.4861624373819496,
      "grad_norm": 0.40424826741218567,
      "learning_rate": 2e-05,
      "loss": 0.7172,
      "step": 5920
    },
    {
      "epoch": 0.48698365771536506,
      "grad_norm": 0.38980022072792053,
      "learning_rate": 2e-05,
      "loss": 0.7218,
      "step": 5930
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 0.7591944932937622,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 5940
    },
    {
      "epoch": 0.48862609838219595,
      "grad_norm": 0.3887224495410919,
      "learning_rate": 2e-05,
      "loss": 0.6944,
      "step": 5950
    },
    {
      "epoch": 0.4894473187156114,
      "grad_norm": 0.39599037170410156,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 5960
    },
    {
      "epoch": 0.49026853904902684,
      "grad_norm": 0.43290939927101135,
      "learning_rate": 2e-05,
      "loss": 0.7314,
      "step": 5970
    },
    {
      "epoch": 0.4910897593824423,
      "grad_norm": 0.3959241509437561,
      "learning_rate": 2e-05,
      "loss": 0.7302,
      "step": 5980
    },
    {
      "epoch": 0.4919109797158578,
      "grad_norm": 0.40210312604904175,
      "learning_rate": 2e-05,
      "loss": 0.6998,
      "step": 5990
    },
    {
      "epoch": 0.4927322000492732,
      "grad_norm": 0.38321635127067566,
      "learning_rate": 2e-05,
      "loss": 0.7199,
      "step": 6000
    },
    {
      "epoch": 0.4935534203826887,
      "grad_norm": 0.3560730516910553,
      "learning_rate": 2e-05,
      "loss": 0.714,
      "step": 6010
    },
    {
      "epoch": 0.49437464071610415,
      "grad_norm": 0.40549057722091675,
      "learning_rate": 2e-05,
      "loss": 0.7386,
      "step": 6020
    },
    {
      "epoch": 0.49519586104951957,
      "grad_norm": 0.37178122997283936,
      "learning_rate": 2e-05,
      "loss": 0.7194,
      "step": 6030
    },
    {
      "epoch": 0.49601708138293504,
      "grad_norm": 0.37464916706085205,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 6040
    },
    {
      "epoch": 0.4968383017163505,
      "grad_norm": 0.40350523591041565,
      "learning_rate": 2e-05,
      "loss": 0.6873,
      "step": 6050
    },
    {
      "epoch": 0.49765952204976593,
      "grad_norm": 0.3710397481918335,
      "learning_rate": 2e-05,
      "loss": 0.6776,
      "step": 6060
    },
    {
      "epoch": 0.4984807423831814,
      "grad_norm": 0.3905718922615051,
      "learning_rate": 2e-05,
      "loss": 0.7133,
      "step": 6070
    },
    {
      "epoch": 0.4993019627165969,
      "grad_norm": 0.40968436002731323,
      "learning_rate": 2e-05,
      "loss": 0.7059,
      "step": 6080
    },
    {
      "epoch": 0.5001231830500124,
      "grad_norm": 0.4127086102962494,
      "learning_rate": 2e-05,
      "loss": 0.7237,
      "step": 6090
    },
    {
      "epoch": 0.5009444033834278,
      "grad_norm": 0.372730553150177,
      "learning_rate": 2e-05,
      "loss": 0.7119,
      "step": 6100
    },
    {
      "epoch": 0.5017656237168432,
      "grad_norm": 0.3668278157711029,
      "learning_rate": 2e-05,
      "loss": 0.7261,
      "step": 6110
    },
    {
      "epoch": 0.5025868440502587,
      "grad_norm": 0.3914894759654999,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 6120
    },
    {
      "epoch": 0.5034080643836741,
      "grad_norm": 0.35023725032806396,
      "learning_rate": 2e-05,
      "loss": 0.7329,
      "step": 6130
    },
    {
      "epoch": 0.5042292847170896,
      "grad_norm": 0.39226359128952026,
      "learning_rate": 2e-05,
      "loss": 0.7457,
      "step": 6140
    },
    {
      "epoch": 0.5050505050505051,
      "grad_norm": 0.388748437166214,
      "learning_rate": 2e-05,
      "loss": 0.7299,
      "step": 6150
    },
    {
      "epoch": 0.5058717253839206,
      "grad_norm": 0.42357906699180603,
      "learning_rate": 2e-05,
      "loss": 0.7077,
      "step": 6160
    },
    {
      "epoch": 0.5066929457173359,
      "grad_norm": 0.353646844625473,
      "learning_rate": 2e-05,
      "loss": 0.7035,
      "step": 6170
    },
    {
      "epoch": 0.5075141660507514,
      "grad_norm": 0.43806391954421997,
      "learning_rate": 2e-05,
      "loss": 0.722,
      "step": 6180
    },
    {
      "epoch": 0.5083353863841669,
      "grad_norm": 0.40449821949005127,
      "learning_rate": 2e-05,
      "loss": 0.6993,
      "step": 6190
    },
    {
      "epoch": 0.5091566067175823,
      "grad_norm": 0.3555327355861664,
      "learning_rate": 2e-05,
      "loss": 0.7388,
      "step": 6200
    },
    {
      "epoch": 0.5099778270509978,
      "grad_norm": 0.392312228679657,
      "learning_rate": 2e-05,
      "loss": 0.7095,
      "step": 6210
    },
    {
      "epoch": 0.5107990473844133,
      "grad_norm": 0.38466712832450867,
      "learning_rate": 2e-05,
      "loss": 0.727,
      "step": 6220
    },
    {
      "epoch": 0.5116202677178286,
      "grad_norm": 0.37945228815078735,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 6230
    },
    {
      "epoch": 0.5124414880512441,
      "grad_norm": 0.3828352391719818,
      "learning_rate": 2e-05,
      "loss": 0.7016,
      "step": 6240
    },
    {
      "epoch": 0.5132627083846596,
      "grad_norm": 0.39961621165275574,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 6250
    },
    {
      "epoch": 0.5140839287180751,
      "grad_norm": 0.3910387456417084,
      "learning_rate": 2e-05,
      "loss": 0.7117,
      "step": 6260
    },
    {
      "epoch": 0.5149051490514905,
      "grad_norm": 0.4011975824832916,
      "learning_rate": 2e-05,
      "loss": 0.7205,
      "step": 6270
    },
    {
      "epoch": 0.515726369384906,
      "grad_norm": 0.3868503272533417,
      "learning_rate": 2e-05,
      "loss": 0.7225,
      "step": 6280
    },
    {
      "epoch": 0.5165475897183214,
      "grad_norm": 0.40181854367256165,
      "learning_rate": 2e-05,
      "loss": 0.7081,
      "step": 6290
    },
    {
      "epoch": 0.5173688100517368,
      "grad_norm": 0.3640654385089874,
      "learning_rate": 2e-05,
      "loss": 0.7183,
      "step": 6300
    },
    {
      "epoch": 0.5181900303851523,
      "grad_norm": 0.4075981676578522,
      "learning_rate": 2e-05,
      "loss": 0.7206,
      "step": 6310
    },
    {
      "epoch": 0.5190112507185678,
      "grad_norm": 0.3861958384513855,
      "learning_rate": 2e-05,
      "loss": 0.7227,
      "step": 6320
    },
    {
      "epoch": 0.5198324710519833,
      "grad_norm": 0.3960513770580292,
      "learning_rate": 2e-05,
      "loss": 0.7125,
      "step": 6330
    },
    {
      "epoch": 0.5206536913853987,
      "grad_norm": 0.39452090859413147,
      "learning_rate": 2e-05,
      "loss": 0.7399,
      "step": 6340
    },
    {
      "epoch": 0.5214749117188141,
      "grad_norm": 0.3818000257015228,
      "learning_rate": 2e-05,
      "loss": 0.7019,
      "step": 6350
    },
    {
      "epoch": 0.5222961320522296,
      "grad_norm": 0.3916766941547394,
      "learning_rate": 2e-05,
      "loss": 0.7299,
      "step": 6360
    },
    {
      "epoch": 0.523117352385645,
      "grad_norm": 0.36682823300361633,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 6370
    },
    {
      "epoch": 0.5239385727190605,
      "grad_norm": 0.39577019214630127,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 6380
    },
    {
      "epoch": 0.524759793052476,
      "grad_norm": 0.40615084767341614,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 6390
    },
    {
      "epoch": 0.5255810133858915,
      "grad_norm": 0.40829741954803467,
      "learning_rate": 2e-05,
      "loss": 0.7121,
      "step": 6400
    },
    {
      "epoch": 0.526402233719307,
      "grad_norm": 0.3768143653869629,
      "learning_rate": 2e-05,
      "loss": 0.7235,
      "step": 6410
    },
    {
      "epoch": 0.5272234540527223,
      "grad_norm": 0.405173659324646,
      "learning_rate": 2e-05,
      "loss": 0.7628,
      "step": 6420
    },
    {
      "epoch": 0.5280446743861378,
      "grad_norm": 0.3852141797542572,
      "learning_rate": 2e-05,
      "loss": 0.7028,
      "step": 6430
    },
    {
      "epoch": 0.5288658947195533,
      "grad_norm": 0.42622965574264526,
      "learning_rate": 2e-05,
      "loss": 0.7285,
      "step": 6440
    },
    {
      "epoch": 0.5296871150529687,
      "grad_norm": 0.38349831104278564,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 6450
    },
    {
      "epoch": 0.5305083353863842,
      "grad_norm": 0.39456772804260254,
      "learning_rate": 2e-05,
      "loss": 0.7215,
      "step": 6460
    },
    {
      "epoch": 0.5313295557197997,
      "grad_norm": 0.36512288451194763,
      "learning_rate": 2e-05,
      "loss": 0.6936,
      "step": 6470
    },
    {
      "epoch": 0.532150776053215,
      "grad_norm": 0.4081200659275055,
      "learning_rate": 2e-05,
      "loss": 0.7342,
      "step": 6480
    },
    {
      "epoch": 0.5329719963866305,
      "grad_norm": 0.49716460704803467,
      "learning_rate": 2e-05,
      "loss": 0.6937,
      "step": 6490
    },
    {
      "epoch": 0.533793216720046,
      "grad_norm": 0.3695957362651825,
      "learning_rate": 2e-05,
      "loss": 0.7477,
      "step": 6500
    },
    {
      "epoch": 0.5346144370534615,
      "grad_norm": 0.4099523425102234,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 6510
    },
    {
      "epoch": 0.5354356573868769,
      "grad_norm": 0.3978554308414459,
      "learning_rate": 2e-05,
      "loss": 0.7694,
      "step": 6520
    },
    {
      "epoch": 0.5362568777202924,
      "grad_norm": 0.3913761377334595,
      "learning_rate": 2e-05,
      "loss": 0.7443,
      "step": 6530
    },
    {
      "epoch": 0.5370780980537078,
      "grad_norm": 0.3978048861026764,
      "learning_rate": 2e-05,
      "loss": 0.7274,
      "step": 6540
    },
    {
      "epoch": 0.5378993183871232,
      "grad_norm": 0.6967421174049377,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 6550
    },
    {
      "epoch": 0.5387205387205387,
      "grad_norm": 0.40671467781066895,
      "learning_rate": 2e-05,
      "loss": 0.7234,
      "step": 6560
    },
    {
      "epoch": 0.5395417590539542,
      "grad_norm": 0.3947722911834717,
      "learning_rate": 2e-05,
      "loss": 0.7203,
      "step": 6570
    },
    {
      "epoch": 0.5403629793873697,
      "grad_norm": 0.38039320707321167,
      "learning_rate": 2e-05,
      "loss": 0.6916,
      "step": 6580
    },
    {
      "epoch": 0.5411841997207851,
      "grad_norm": 0.39534735679626465,
      "learning_rate": 2e-05,
      "loss": 0.7102,
      "step": 6590
    },
    {
      "epoch": 0.5420054200542005,
      "grad_norm": 0.3808237612247467,
      "learning_rate": 2e-05,
      "loss": 0.6842,
      "step": 6600
    },
    {
      "epoch": 0.542826640387616,
      "grad_norm": 0.38083505630493164,
      "learning_rate": 2e-05,
      "loss": 0.7068,
      "step": 6610
    },
    {
      "epoch": 0.5436478607210314,
      "grad_norm": 0.3666287064552307,
      "learning_rate": 2e-05,
      "loss": 0.6921,
      "step": 6620
    },
    {
      "epoch": 0.5444690810544469,
      "grad_norm": 0.39683595299720764,
      "learning_rate": 2e-05,
      "loss": 0.7218,
      "step": 6630
    },
    {
      "epoch": 0.5452903013878624,
      "grad_norm": 0.36132293939590454,
      "learning_rate": 2e-05,
      "loss": 0.7416,
      "step": 6640
    },
    {
      "epoch": 0.5461115217212779,
      "grad_norm": 0.38175541162490845,
      "learning_rate": 2e-05,
      "loss": 0.736,
      "step": 6650
    },
    {
      "epoch": 0.5469327420546932,
      "grad_norm": 0.40198060870170593,
      "learning_rate": 2e-05,
      "loss": 0.725,
      "step": 6660
    },
    {
      "epoch": 0.5477539623881087,
      "grad_norm": 0.397853285074234,
      "learning_rate": 2e-05,
      "loss": 0.6902,
      "step": 6670
    },
    {
      "epoch": 0.5485751827215242,
      "grad_norm": 0.40711256861686707,
      "learning_rate": 2e-05,
      "loss": 0.732,
      "step": 6680
    },
    {
      "epoch": 0.5493964030549396,
      "grad_norm": 0.4229613244533539,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 6690
    },
    {
      "epoch": 0.5502176233883551,
      "grad_norm": 0.3844345510005951,
      "learning_rate": 2e-05,
      "loss": 0.711,
      "step": 6700
    },
    {
      "epoch": 0.5510388437217706,
      "grad_norm": 0.37241098284721375,
      "learning_rate": 2e-05,
      "loss": 0.7181,
      "step": 6710
    },
    {
      "epoch": 0.5518600640551861,
      "grad_norm": 0.4058673679828644,
      "learning_rate": 2e-05,
      "loss": 0.738,
      "step": 6720
    },
    {
      "epoch": 0.5526812843886014,
      "grad_norm": 0.3739365041255951,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 6730
    },
    {
      "epoch": 0.5535025047220169,
      "grad_norm": 0.3855057656764984,
      "learning_rate": 2e-05,
      "loss": 0.7332,
      "step": 6740
    },
    {
      "epoch": 0.5543237250554324,
      "grad_norm": 0.3911612927913666,
      "learning_rate": 2e-05,
      "loss": 0.7215,
      "step": 6750
    },
    {
      "epoch": 0.5551449453888478,
      "grad_norm": 0.37556013464927673,
      "learning_rate": 2e-05,
      "loss": 0.7045,
      "step": 6760
    },
    {
      "epoch": 0.5559661657222633,
      "grad_norm": 0.3648332953453064,
      "learning_rate": 2e-05,
      "loss": 0.7202,
      "step": 6770
    },
    {
      "epoch": 0.5567873860556788,
      "grad_norm": 0.3752780556678772,
      "learning_rate": 2e-05,
      "loss": 0.6943,
      "step": 6780
    },
    {
      "epoch": 0.5576086063890942,
      "grad_norm": 0.4007888734340668,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 6790
    },
    {
      "epoch": 0.5584298267225096,
      "grad_norm": 0.40781888365745544,
      "learning_rate": 2e-05,
      "loss": 0.7277,
      "step": 6800
    },
    {
      "epoch": 0.5592510470559251,
      "grad_norm": 0.40122586488723755,
      "learning_rate": 2e-05,
      "loss": 0.7383,
      "step": 6810
    },
    {
      "epoch": 0.5600722673893406,
      "grad_norm": 0.36612218618392944,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 6820
    },
    {
      "epoch": 0.560893487722756,
      "grad_norm": 0.48450422286987305,
      "learning_rate": 2e-05,
      "loss": 0.6972,
      "step": 6830
    },
    {
      "epoch": 0.5617147080561715,
      "grad_norm": 0.4043528139591217,
      "learning_rate": 2e-05,
      "loss": 0.7517,
      "step": 6840
    },
    {
      "epoch": 0.5625359283895869,
      "grad_norm": 0.37291401624679565,
      "learning_rate": 2e-05,
      "loss": 0.7148,
      "step": 6850
    },
    {
      "epoch": 0.5633571487230024,
      "grad_norm": 0.4111575484275818,
      "learning_rate": 2e-05,
      "loss": 0.6972,
      "step": 6860
    },
    {
      "epoch": 0.5641783690564178,
      "grad_norm": 0.4368439316749573,
      "learning_rate": 2e-05,
      "loss": 0.7476,
      "step": 6870
    },
    {
      "epoch": 0.5649995893898333,
      "grad_norm": 0.38704952597618103,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 6880
    },
    {
      "epoch": 0.5658208097232488,
      "grad_norm": 0.36741289496421814,
      "learning_rate": 2e-05,
      "loss": 0.7428,
      "step": 6890
    },
    {
      "epoch": 0.5666420300566642,
      "grad_norm": 0.3772566616535187,
      "learning_rate": 2e-05,
      "loss": 0.7092,
      "step": 6900
    },
    {
      "epoch": 0.5674632503900796,
      "grad_norm": 0.38634201884269714,
      "learning_rate": 2e-05,
      "loss": 0.7009,
      "step": 6910
    },
    {
      "epoch": 0.5682844707234951,
      "grad_norm": 0.3704381585121155,
      "learning_rate": 2e-05,
      "loss": 0.7326,
      "step": 6920
    },
    {
      "epoch": 0.5691056910569106,
      "grad_norm": 0.3998047709465027,
      "learning_rate": 2e-05,
      "loss": 0.7625,
      "step": 6930
    },
    {
      "epoch": 0.569926911390326,
      "grad_norm": 0.7627277970314026,
      "learning_rate": 2e-05,
      "loss": 0.7067,
      "step": 6940
    },
    {
      "epoch": 0.5707481317237415,
      "grad_norm": 0.37870973348617554,
      "learning_rate": 2e-05,
      "loss": 0.7042,
      "step": 6950
    },
    {
      "epoch": 0.571569352057157,
      "grad_norm": 0.35699641704559326,
      "learning_rate": 2e-05,
      "loss": 0.7203,
      "step": 6960
    },
    {
      "epoch": 0.5723905723905723,
      "grad_norm": 0.406017005443573,
      "learning_rate": 2e-05,
      "loss": 0.7114,
      "step": 6970
    },
    {
      "epoch": 0.5732117927239878,
      "grad_norm": 0.3750627934932709,
      "learning_rate": 2e-05,
      "loss": 0.7105,
      "step": 6980
    },
    {
      "epoch": 0.5740330130574033,
      "grad_norm": 0.3883343040943146,
      "learning_rate": 2e-05,
      "loss": 0.7088,
      "step": 6990
    },
    {
      "epoch": 0.5748542333908188,
      "grad_norm": 0.38125044107437134,
      "learning_rate": 2e-05,
      "loss": 0.7205,
      "step": 7000
    },
    {
      "epoch": 0.5756754537242342,
      "grad_norm": 0.3819754421710968,
      "learning_rate": 2e-05,
      "loss": 0.7253,
      "step": 7010
    },
    {
      "epoch": 0.5764966740576497,
      "grad_norm": 0.3695208430290222,
      "learning_rate": 2e-05,
      "loss": 0.738,
      "step": 7020
    },
    {
      "epoch": 0.5773178943910651,
      "grad_norm": 0.3863351345062256,
      "learning_rate": 2e-05,
      "loss": 0.7063,
      "step": 7030
    },
    {
      "epoch": 0.5781391147244805,
      "grad_norm": 0.38179731369018555,
      "learning_rate": 2e-05,
      "loss": 0.7074,
      "step": 7040
    },
    {
      "epoch": 0.578960335057896,
      "grad_norm": 0.37154340744018555,
      "learning_rate": 2e-05,
      "loss": 0.6975,
      "step": 7050
    },
    {
      "epoch": 0.5797815553913115,
      "grad_norm": 0.39739131927490234,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 7060
    },
    {
      "epoch": 0.580602775724727,
      "grad_norm": 0.3788069486618042,
      "learning_rate": 2e-05,
      "loss": 0.7322,
      "step": 7070
    },
    {
      "epoch": 0.5814239960581424,
      "grad_norm": 0.3721085488796234,
      "learning_rate": 2e-05,
      "loss": 0.7205,
      "step": 7080
    },
    {
      "epoch": 0.5822452163915579,
      "grad_norm": 0.7252041101455688,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 7090
    },
    {
      "epoch": 0.5830664367249733,
      "grad_norm": 0.3837658166885376,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 7100
    },
    {
      "epoch": 0.5838876570583887,
      "grad_norm": 0.4083505868911743,
      "learning_rate": 2e-05,
      "loss": 0.7263,
      "step": 7110
    },
    {
      "epoch": 0.5847088773918042,
      "grad_norm": 0.37903696298599243,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 7120
    },
    {
      "epoch": 0.5855300977252197,
      "grad_norm": 0.403678297996521,
      "learning_rate": 2e-05,
      "loss": 0.7605,
      "step": 7130
    },
    {
      "epoch": 0.5863513180586352,
      "grad_norm": 1.5854135751724243,
      "learning_rate": 2e-05,
      "loss": 0.7253,
      "step": 7140
    },
    {
      "epoch": 0.5871725383920506,
      "grad_norm": 0.366970032453537,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 7150
    },
    {
      "epoch": 0.587993758725466,
      "grad_norm": 0.4028233289718628,
      "learning_rate": 2e-05,
      "loss": 0.7534,
      "step": 7160
    },
    {
      "epoch": 0.5888149790588815,
      "grad_norm": 0.4038342535495758,
      "learning_rate": 2e-05,
      "loss": 0.6927,
      "step": 7170
    },
    {
      "epoch": 0.589636199392297,
      "grad_norm": 0.3847525417804718,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 7180
    },
    {
      "epoch": 0.5904574197257124,
      "grad_norm": 0.4157886803150177,
      "learning_rate": 2e-05,
      "loss": 0.6897,
      "step": 7190
    },
    {
      "epoch": 0.5912786400591279,
      "grad_norm": 0.39904317259788513,
      "learning_rate": 2e-05,
      "loss": 0.6933,
      "step": 7200
    },
    {
      "epoch": 0.5920998603925434,
      "grad_norm": 0.3921680748462677,
      "learning_rate": 2e-05,
      "loss": 0.7052,
      "step": 7210
    },
    {
      "epoch": 0.5929210807259587,
      "grad_norm": 0.39005833864212036,
      "learning_rate": 2e-05,
      "loss": 0.7256,
      "step": 7220
    },
    {
      "epoch": 0.5937423010593742,
      "grad_norm": 0.38722363114356995,
      "learning_rate": 2e-05,
      "loss": 0.7117,
      "step": 7230
    },
    {
      "epoch": 0.5945635213927897,
      "grad_norm": 0.40944042801856995,
      "learning_rate": 2e-05,
      "loss": 0.7076,
      "step": 7240
    },
    {
      "epoch": 0.5953847417262051,
      "grad_norm": 0.4114813208580017,
      "learning_rate": 2e-05,
      "loss": 0.7307,
      "step": 7250
    },
    {
      "epoch": 0.5962059620596206,
      "grad_norm": 0.39944928884506226,
      "learning_rate": 2e-05,
      "loss": 0.6933,
      "step": 7260
    },
    {
      "epoch": 0.5970271823930361,
      "grad_norm": 0.36242783069610596,
      "learning_rate": 2e-05,
      "loss": 0.6945,
      "step": 7270
    },
    {
      "epoch": 0.5978484027264515,
      "grad_norm": 0.42097729444503784,
      "learning_rate": 2e-05,
      "loss": 0.7113,
      "step": 7280
    },
    {
      "epoch": 0.5986696230598669,
      "grad_norm": 0.3832961618900299,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 7290
    },
    {
      "epoch": 0.5994908433932824,
      "grad_norm": 0.3991835415363312,
      "learning_rate": 2e-05,
      "loss": 0.6848,
      "step": 7300
    },
    {
      "epoch": 0.6003120637266979,
      "grad_norm": 0.36446985602378845,
      "learning_rate": 2e-05,
      "loss": 0.7366,
      "step": 7310
    },
    {
      "epoch": 0.6011332840601133,
      "grad_norm": 0.37723127007484436,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 7320
    },
    {
      "epoch": 0.6019545043935288,
      "grad_norm": 0.37297922372817993,
      "learning_rate": 2e-05,
      "loss": 0.6982,
      "step": 7330
    },
    {
      "epoch": 0.6027757247269442,
      "grad_norm": 0.4184410274028778,
      "learning_rate": 2e-05,
      "loss": 0.7188,
      "step": 7340
    },
    {
      "epoch": 0.6035969450603597,
      "grad_norm": 0.3828622102737427,
      "learning_rate": 2e-05,
      "loss": 0.71,
      "step": 7350
    },
    {
      "epoch": 0.6044181653937751,
      "grad_norm": 0.38385945558547974,
      "learning_rate": 2e-05,
      "loss": 0.7315,
      "step": 7360
    },
    {
      "epoch": 0.6052393857271906,
      "grad_norm": 0.36622339487075806,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 7370
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 0.4158931374549866,
      "learning_rate": 2e-05,
      "loss": 0.6953,
      "step": 7380
    },
    {
      "epoch": 0.6068818263940216,
      "grad_norm": 0.3827923536300659,
      "learning_rate": 2e-05,
      "loss": 0.7129,
      "step": 7390
    },
    {
      "epoch": 0.607703046727437,
      "grad_norm": 0.41263407468795776,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 7400
    },
    {
      "epoch": 0.6085242670608524,
      "grad_norm": 0.41697055101394653,
      "learning_rate": 2e-05,
      "loss": 0.724,
      "step": 7410
    },
    {
      "epoch": 0.6093454873942679,
      "grad_norm": 0.3679683804512024,
      "learning_rate": 2e-05,
      "loss": 0.7187,
      "step": 7420
    },
    {
      "epoch": 0.6101667077276833,
      "grad_norm": 0.37618783116340637,
      "learning_rate": 2e-05,
      "loss": 0.7191,
      "step": 7430
    },
    {
      "epoch": 0.6109879280610988,
      "grad_norm": 0.3684156537055969,
      "learning_rate": 2e-05,
      "loss": 0.7186,
      "step": 7440
    },
    {
      "epoch": 0.6118091483945143,
      "grad_norm": 0.3685910105705261,
      "learning_rate": 2e-05,
      "loss": 0.7072,
      "step": 7450
    },
    {
      "epoch": 0.6126303687279298,
      "grad_norm": 0.38204318284988403,
      "learning_rate": 2e-05,
      "loss": 0.7059,
      "step": 7460
    },
    {
      "epoch": 0.6134515890613451,
      "grad_norm": 0.40431588888168335,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 7470
    },
    {
      "epoch": 0.6142728093947606,
      "grad_norm": 0.39317214488983154,
      "learning_rate": 2e-05,
      "loss": 0.727,
      "step": 7480
    },
    {
      "epoch": 0.6150940297281761,
      "grad_norm": 0.39114850759506226,
      "learning_rate": 2e-05,
      "loss": 0.7151,
      "step": 7490
    },
    {
      "epoch": 0.6159152500615915,
      "grad_norm": 0.37656769156455994,
      "learning_rate": 2e-05,
      "loss": 0.709,
      "step": 7500
    },
    {
      "epoch": 0.616736470395007,
      "grad_norm": 0.40597954392433167,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 7510
    },
    {
      "epoch": 0.6175576907284225,
      "grad_norm": 0.43603280186653137,
      "learning_rate": 2e-05,
      "loss": 0.7345,
      "step": 7520
    },
    {
      "epoch": 0.6183789110618378,
      "grad_norm": 0.4854944348335266,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 7530
    },
    {
      "epoch": 0.6192001313952533,
      "grad_norm": 0.37075719237327576,
      "learning_rate": 2e-05,
      "loss": 0.7238,
      "step": 7540
    },
    {
      "epoch": 0.6200213517286688,
      "grad_norm": 0.3832797110080719,
      "learning_rate": 2e-05,
      "loss": 0.6814,
      "step": 7550
    },
    {
      "epoch": 0.6208425720620843,
      "grad_norm": 0.3982393145561218,
      "learning_rate": 2e-05,
      "loss": 0.711,
      "step": 7560
    },
    {
      "epoch": 0.6216637923954997,
      "grad_norm": 0.37549376487731934,
      "learning_rate": 2e-05,
      "loss": 0.7057,
      "step": 7570
    },
    {
      "epoch": 0.6224850127289152,
      "grad_norm": 0.41434717178344727,
      "learning_rate": 2e-05,
      "loss": 0.7302,
      "step": 7580
    },
    {
      "epoch": 0.6233062330623306,
      "grad_norm": 0.5706821084022522,
      "learning_rate": 2e-05,
      "loss": 0.7414,
      "step": 7590
    },
    {
      "epoch": 0.624127453395746,
      "grad_norm": 0.37827548384666443,
      "learning_rate": 2e-05,
      "loss": 0.7032,
      "step": 7600
    },
    {
      "epoch": 0.6249486737291615,
      "grad_norm": 0.38562992215156555,
      "learning_rate": 2e-05,
      "loss": 0.712,
      "step": 7610
    },
    {
      "epoch": 0.625769894062577,
      "grad_norm": 0.4023134112358093,
      "learning_rate": 2e-05,
      "loss": 0.7215,
      "step": 7620
    },
    {
      "epoch": 0.6265911143959925,
      "grad_norm": 0.40137407183647156,
      "learning_rate": 2e-05,
      "loss": 0.7155,
      "step": 7630
    },
    {
      "epoch": 0.6274123347294079,
      "grad_norm": 0.39676281809806824,
      "learning_rate": 2e-05,
      "loss": 0.7079,
      "step": 7640
    },
    {
      "epoch": 0.6282335550628233,
      "grad_norm": 0.3825543224811554,
      "learning_rate": 2e-05,
      "loss": 0.7052,
      "step": 7650
    },
    {
      "epoch": 0.6290547753962388,
      "grad_norm": 0.4177508056163788,
      "learning_rate": 2e-05,
      "loss": 0.7265,
      "step": 7660
    },
    {
      "epoch": 0.6298759957296542,
      "grad_norm": 0.4121154546737671,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 7670
    },
    {
      "epoch": 0.6306972160630697,
      "grad_norm": 0.379709392786026,
      "learning_rate": 2e-05,
      "loss": 0.7076,
      "step": 7680
    },
    {
      "epoch": 0.6315184363964852,
      "grad_norm": 0.3706896901130676,
      "learning_rate": 2e-05,
      "loss": 0.6805,
      "step": 7690
    },
    {
      "epoch": 0.6323396567299007,
      "grad_norm": 0.3662567734718323,
      "learning_rate": 2e-05,
      "loss": 0.7054,
      "step": 7700
    },
    {
      "epoch": 0.6331608770633161,
      "grad_norm": 0.3617132008075714,
      "learning_rate": 2e-05,
      "loss": 0.6729,
      "step": 7710
    },
    {
      "epoch": 0.6339820973967315,
      "grad_norm": 0.36475327610969543,
      "learning_rate": 2e-05,
      "loss": 0.7157,
      "step": 7720
    },
    {
      "epoch": 0.634803317730147,
      "grad_norm": 0.3757360577583313,
      "learning_rate": 2e-05,
      "loss": 0.7078,
      "step": 7730
    },
    {
      "epoch": 0.6356245380635624,
      "grad_norm": 0.3991989493370056,
      "learning_rate": 2e-05,
      "loss": 0.7153,
      "step": 7740
    },
    {
      "epoch": 0.6364457583969779,
      "grad_norm": 0.39735257625579834,
      "learning_rate": 2e-05,
      "loss": 0.6878,
      "step": 7750
    },
    {
      "epoch": 0.6372669787303934,
      "grad_norm": 0.380898118019104,
      "learning_rate": 2e-05,
      "loss": 0.6874,
      "step": 7760
    },
    {
      "epoch": 0.6380881990638089,
      "grad_norm": 0.38472551107406616,
      "learning_rate": 2e-05,
      "loss": 0.7237,
      "step": 7770
    },
    {
      "epoch": 0.6389094193972242,
      "grad_norm": 0.38063904643058777,
      "learning_rate": 2e-05,
      "loss": 0.7049,
      "step": 7780
    },
    {
      "epoch": 0.6397306397306397,
      "grad_norm": 0.3798016607761383,
      "learning_rate": 2e-05,
      "loss": 0.7113,
      "step": 7790
    },
    {
      "epoch": 0.6405518600640552,
      "grad_norm": 0.46780911087989807,
      "learning_rate": 2e-05,
      "loss": 0.7221,
      "step": 7800
    },
    {
      "epoch": 0.6413730803974707,
      "grad_norm": 0.3765014708042145,
      "learning_rate": 2e-05,
      "loss": 0.7261,
      "step": 7810
    },
    {
      "epoch": 0.6421943007308861,
      "grad_norm": 0.38984543085098267,
      "learning_rate": 2e-05,
      "loss": 0.7173,
      "step": 7820
    },
    {
      "epoch": 0.6430155210643016,
      "grad_norm": 0.3748830258846283,
      "learning_rate": 2e-05,
      "loss": 0.7198,
      "step": 7830
    },
    {
      "epoch": 0.643836741397717,
      "grad_norm": 0.3756381869316101,
      "learning_rate": 2e-05,
      "loss": 0.6887,
      "step": 7840
    },
    {
      "epoch": 0.6446579617311324,
      "grad_norm": 0.37065398693084717,
      "learning_rate": 2e-05,
      "loss": 0.6682,
      "step": 7850
    },
    {
      "epoch": 0.6454791820645479,
      "grad_norm": 0.3698570132255554,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 7860
    },
    {
      "epoch": 0.6463004023979634,
      "grad_norm": 0.3837618827819824,
      "learning_rate": 2e-05,
      "loss": 0.6731,
      "step": 7870
    },
    {
      "epoch": 0.6471216227313789,
      "grad_norm": 0.42657235264778137,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 7880
    },
    {
      "epoch": 0.6479428430647943,
      "grad_norm": 0.4023689925670624,
      "learning_rate": 2e-05,
      "loss": 0.7055,
      "step": 7890
    },
    {
      "epoch": 0.6487640633982097,
      "grad_norm": 0.41045817732810974,
      "learning_rate": 2e-05,
      "loss": 0.7063,
      "step": 7900
    },
    {
      "epoch": 0.6495852837316252,
      "grad_norm": 0.3867765963077545,
      "learning_rate": 2e-05,
      "loss": 0.672,
      "step": 7910
    },
    {
      "epoch": 0.6504065040650406,
      "grad_norm": 0.44488370418548584,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 7920
    },
    {
      "epoch": 0.6512277243984561,
      "grad_norm": 0.38452184200286865,
      "learning_rate": 2e-05,
      "loss": 0.7162,
      "step": 7930
    },
    {
      "epoch": 0.6520489447318716,
      "grad_norm": 0.39899757504463196,
      "learning_rate": 2e-05,
      "loss": 0.7249,
      "step": 7940
    },
    {
      "epoch": 0.6528701650652871,
      "grad_norm": 0.4000389575958252,
      "learning_rate": 2e-05,
      "loss": 0.7108,
      "step": 7950
    },
    {
      "epoch": 0.6536913853987024,
      "grad_norm": 0.41105762124061584,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 7960
    },
    {
      "epoch": 0.6545126057321179,
      "grad_norm": 0.37135010957717896,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 7970
    },
    {
      "epoch": 0.6553338260655334,
      "grad_norm": 0.38258442282676697,
      "learning_rate": 2e-05,
      "loss": 0.7149,
      "step": 7980
    },
    {
      "epoch": 0.6561550463989488,
      "grad_norm": 0.4085336923599243,
      "learning_rate": 2e-05,
      "loss": 0.696,
      "step": 7990
    },
    {
      "epoch": 0.6569762667323643,
      "grad_norm": 0.3857642710208893,
      "learning_rate": 2e-05,
      "loss": 0.6988,
      "step": 8000
    },
    {
      "epoch": 0.6577974870657798,
      "grad_norm": 0.38432857394218445,
      "learning_rate": 2e-05,
      "loss": 0.72,
      "step": 8010
    },
    {
      "epoch": 0.6586187073991953,
      "grad_norm": 0.39867103099823,
      "learning_rate": 2e-05,
      "loss": 0.7186,
      "step": 8020
    },
    {
      "epoch": 0.6594399277326106,
      "grad_norm": 0.35693949460983276,
      "learning_rate": 2e-05,
      "loss": 0.6741,
      "step": 8030
    },
    {
      "epoch": 0.6602611480660261,
      "grad_norm": 0.4515802562236786,
      "learning_rate": 2e-05,
      "loss": 0.688,
      "step": 8040
    },
    {
      "epoch": 0.6610823683994416,
      "grad_norm": 0.3710004687309265,
      "learning_rate": 2e-05,
      "loss": 0.7223,
      "step": 8050
    },
    {
      "epoch": 0.661903588732857,
      "grad_norm": 0.4026776850223541,
      "learning_rate": 2e-05,
      "loss": 0.6658,
      "step": 8060
    },
    {
      "epoch": 0.6627248090662725,
      "grad_norm": 0.3823258876800537,
      "learning_rate": 2e-05,
      "loss": 0.7269,
      "step": 8070
    },
    {
      "epoch": 0.663546029399688,
      "grad_norm": 0.3764406144618988,
      "learning_rate": 2e-05,
      "loss": 0.7083,
      "step": 8080
    },
    {
      "epoch": 0.6643672497331033,
      "grad_norm": 0.42511385679244995,
      "learning_rate": 2e-05,
      "loss": 0.7177,
      "step": 8090
    },
    {
      "epoch": 0.6651884700665188,
      "grad_norm": 0.38207945227622986,
      "learning_rate": 2e-05,
      "loss": 0.7246,
      "step": 8100
    },
    {
      "epoch": 0.6660096903999343,
      "grad_norm": 0.3671714961528778,
      "learning_rate": 2e-05,
      "loss": 0.7041,
      "step": 8110
    },
    {
      "epoch": 0.6668309107333498,
      "grad_norm": 0.3716355860233307,
      "learning_rate": 2e-05,
      "loss": 0.6963,
      "step": 8120
    },
    {
      "epoch": 0.6676521310667652,
      "grad_norm": 0.367675244808197,
      "learning_rate": 2e-05,
      "loss": 0.6635,
      "step": 8130
    },
    {
      "epoch": 0.6684733514001807,
      "grad_norm": 0.35553988814353943,
      "learning_rate": 2e-05,
      "loss": 0.7367,
      "step": 8140
    },
    {
      "epoch": 0.6692945717335961,
      "grad_norm": 0.3716179430484772,
      "learning_rate": 2e-05,
      "loss": 0.7284,
      "step": 8150
    },
    {
      "epoch": 0.6701157920670116,
      "grad_norm": 0.3633224070072174,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 8160
    },
    {
      "epoch": 0.670937012400427,
      "grad_norm": 0.3933897316455841,
      "learning_rate": 2e-05,
      "loss": 0.7344,
      "step": 8170
    },
    {
      "epoch": 0.6717582327338425,
      "grad_norm": 0.3754507005214691,
      "learning_rate": 2e-05,
      "loss": 0.7017,
      "step": 8180
    },
    {
      "epoch": 0.672579453067258,
      "grad_norm": 0.4060342609882355,
      "learning_rate": 2e-05,
      "loss": 0.7297,
      "step": 8190
    },
    {
      "epoch": 0.6734006734006734,
      "grad_norm": 0.39697763323783875,
      "learning_rate": 2e-05,
      "loss": 0.7178,
      "step": 8200
    },
    {
      "epoch": 0.6742218937340888,
      "grad_norm": 0.3887491226196289,
      "learning_rate": 2e-05,
      "loss": 0.7219,
      "step": 8210
    },
    {
      "epoch": 0.6750431140675043,
      "grad_norm": 0.39436063170433044,
      "learning_rate": 2e-05,
      "loss": 0.7053,
      "step": 8220
    },
    {
      "epoch": 0.6758643344009198,
      "grad_norm": 0.3961201310157776,
      "learning_rate": 2e-05,
      "loss": 0.6915,
      "step": 8230
    },
    {
      "epoch": 0.6766855547343352,
      "grad_norm": 0.3756488263607025,
      "learning_rate": 2e-05,
      "loss": 0.7402,
      "step": 8240
    },
    {
      "epoch": 0.6775067750677507,
      "grad_norm": 0.4189121127128601,
      "learning_rate": 2e-05,
      "loss": 0.698,
      "step": 8250
    },
    {
      "epoch": 0.6783279954011662,
      "grad_norm": 0.41733887791633606,
      "learning_rate": 2e-05,
      "loss": 0.6945,
      "step": 8260
    },
    {
      "epoch": 0.6791492157345815,
      "grad_norm": 0.4145960509777069,
      "learning_rate": 2e-05,
      "loss": 0.6997,
      "step": 8270
    },
    {
      "epoch": 0.679970436067997,
      "grad_norm": 0.38918983936309814,
      "learning_rate": 2e-05,
      "loss": 0.7209,
      "step": 8280
    },
    {
      "epoch": 0.6807916564014125,
      "grad_norm": 0.359842449426651,
      "learning_rate": 2e-05,
      "loss": 0.7358,
      "step": 8290
    },
    {
      "epoch": 0.681612876734828,
      "grad_norm": 0.39814630150794983,
      "learning_rate": 2e-05,
      "loss": 0.6766,
      "step": 8300
    },
    {
      "epoch": 0.6824340970682434,
      "grad_norm": 0.39943304657936096,
      "learning_rate": 2e-05,
      "loss": 0.6934,
      "step": 8310
    },
    {
      "epoch": 0.6832553174016589,
      "grad_norm": 0.40420347452163696,
      "learning_rate": 2e-05,
      "loss": 0.7069,
      "step": 8320
    },
    {
      "epoch": 0.6840765377350744,
      "grad_norm": 0.3845677971839905,
      "learning_rate": 2e-05,
      "loss": 0.6981,
      "step": 8330
    },
    {
      "epoch": 0.6848977580684897,
      "grad_norm": 0.3590324819087982,
      "learning_rate": 2e-05,
      "loss": 0.7058,
      "step": 8340
    },
    {
      "epoch": 0.6857189784019052,
      "grad_norm": 0.36812108755111694,
      "learning_rate": 2e-05,
      "loss": 0.7186,
      "step": 8350
    },
    {
      "epoch": 0.6865401987353207,
      "grad_norm": 0.3829042315483093,
      "learning_rate": 2e-05,
      "loss": 0.7168,
      "step": 8360
    },
    {
      "epoch": 0.6873614190687362,
      "grad_norm": 0.41038575768470764,
      "learning_rate": 2e-05,
      "loss": 0.754,
      "step": 8370
    },
    {
      "epoch": 0.6881826394021516,
      "grad_norm": 0.36874204874038696,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 8380
    },
    {
      "epoch": 0.6890038597355671,
      "grad_norm": 0.41837987303733826,
      "learning_rate": 2e-05,
      "loss": 0.7196,
      "step": 8390
    },
    {
      "epoch": 0.6898250800689825,
      "grad_norm": 0.4153330624103546,
      "learning_rate": 2e-05,
      "loss": 0.7078,
      "step": 8400
    },
    {
      "epoch": 0.6906463004023979,
      "grad_norm": 0.3460937738418579,
      "learning_rate": 2e-05,
      "loss": 0.7008,
      "step": 8410
    },
    {
      "epoch": 0.6914675207358134,
      "grad_norm": 0.34645378589630127,
      "learning_rate": 2e-05,
      "loss": 0.6935,
      "step": 8420
    },
    {
      "epoch": 0.6922887410692289,
      "grad_norm": 0.4237283170223236,
      "learning_rate": 2e-05,
      "loss": 0.7241,
      "step": 8430
    },
    {
      "epoch": 0.6931099614026444,
      "grad_norm": 0.3922346234321594,
      "learning_rate": 2e-05,
      "loss": 0.7214,
      "step": 8440
    },
    {
      "epoch": 0.6939311817360598,
      "grad_norm": 0.3586946129798889,
      "learning_rate": 2e-05,
      "loss": 0.7144,
      "step": 8450
    },
    {
      "epoch": 0.6947524020694752,
      "grad_norm": 0.4045044779777527,
      "learning_rate": 2e-05,
      "loss": 0.708,
      "step": 8460
    },
    {
      "epoch": 0.6955736224028907,
      "grad_norm": 0.41124197840690613,
      "learning_rate": 2e-05,
      "loss": 0.7036,
      "step": 8470
    },
    {
      "epoch": 0.6963948427363061,
      "grad_norm": 0.3983819782733917,
      "learning_rate": 2e-05,
      "loss": 0.7582,
      "step": 8480
    },
    {
      "epoch": 0.6972160630697216,
      "grad_norm": 0.3881699740886688,
      "learning_rate": 2e-05,
      "loss": 0.7206,
      "step": 8490
    },
    {
      "epoch": 0.6980372834031371,
      "grad_norm": 0.3912210762500763,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 8500
    },
    {
      "epoch": 0.6988585037365526,
      "grad_norm": 0.37621235847473145,
      "learning_rate": 2e-05,
      "loss": 0.6805,
      "step": 8510
    },
    {
      "epoch": 0.6996797240699679,
      "grad_norm": 0.3832293152809143,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 8520
    },
    {
      "epoch": 0.7005009444033834,
      "grad_norm": 0.37290263175964355,
      "learning_rate": 2e-05,
      "loss": 0.6985,
      "step": 8530
    },
    {
      "epoch": 0.7013221647367989,
      "grad_norm": 0.5809956789016724,
      "learning_rate": 2e-05,
      "loss": 0.7069,
      "step": 8540
    },
    {
      "epoch": 0.7021433850702143,
      "grad_norm": 0.37460023164749146,
      "learning_rate": 2e-05,
      "loss": 0.6898,
      "step": 8550
    },
    {
      "epoch": 0.7029646054036298,
      "grad_norm": 0.3906730115413666,
      "learning_rate": 2e-05,
      "loss": 0.7234,
      "step": 8560
    },
    {
      "epoch": 0.7037858257370453,
      "grad_norm": 0.38863933086395264,
      "learning_rate": 2e-05,
      "loss": 0.6971,
      "step": 8570
    },
    {
      "epoch": 0.7046070460704607,
      "grad_norm": 0.39751768112182617,
      "learning_rate": 2e-05,
      "loss": 0.7219,
      "step": 8580
    },
    {
      "epoch": 0.7054282664038761,
      "grad_norm": 0.4489176571369171,
      "learning_rate": 2e-05,
      "loss": 0.6933,
      "step": 8590
    },
    {
      "epoch": 0.7062494867372916,
      "grad_norm": 0.3742929697036743,
      "learning_rate": 2e-05,
      "loss": 0.6761,
      "step": 8600
    },
    {
      "epoch": 0.7070707070707071,
      "grad_norm": 0.39749348163604736,
      "learning_rate": 2e-05,
      "loss": 0.7014,
      "step": 8610
    },
    {
      "epoch": 0.7078919274041225,
      "grad_norm": 0.40697088837623596,
      "learning_rate": 2e-05,
      "loss": 0.7114,
      "step": 8620
    },
    {
      "epoch": 0.708713147737538,
      "grad_norm": 0.3836103081703186,
      "learning_rate": 2e-05,
      "loss": 0.6989,
      "step": 8630
    },
    {
      "epoch": 0.7095343680709535,
      "grad_norm": 0.36759936809539795,
      "learning_rate": 2e-05,
      "loss": 0.6833,
      "step": 8640
    },
    {
      "epoch": 0.7103555884043689,
      "grad_norm": 0.37002986669540405,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 8650
    },
    {
      "epoch": 0.7111768087377843,
      "grad_norm": 0.39321574568748474,
      "learning_rate": 2e-05,
      "loss": 0.6801,
      "step": 8660
    },
    {
      "epoch": 0.7119980290711998,
      "grad_norm": 0.38775667548179626,
      "learning_rate": 2e-05,
      "loss": 0.6678,
      "step": 8670
    },
    {
      "epoch": 0.7128192494046153,
      "grad_norm": 0.39464637637138367,
      "learning_rate": 2e-05,
      "loss": 0.6985,
      "step": 8680
    },
    {
      "epoch": 0.7136404697380307,
      "grad_norm": 0.4122486710548401,
      "learning_rate": 2e-05,
      "loss": 0.6911,
      "step": 8690
    },
    {
      "epoch": 0.7144616900714462,
      "grad_norm": 0.38463208079338074,
      "learning_rate": 2e-05,
      "loss": 0.7173,
      "step": 8700
    },
    {
      "epoch": 0.7152829104048616,
      "grad_norm": 0.35905569791793823,
      "learning_rate": 2e-05,
      "loss": 0.7127,
      "step": 8710
    },
    {
      "epoch": 0.7161041307382771,
      "grad_norm": 0.38859298825263977,
      "learning_rate": 2e-05,
      "loss": 0.6857,
      "step": 8720
    },
    {
      "epoch": 0.7169253510716925,
      "grad_norm": 0.364869087934494,
      "learning_rate": 2e-05,
      "loss": 0.6976,
      "step": 8730
    },
    {
      "epoch": 0.717746571405108,
      "grad_norm": 0.4024193286895752,
      "learning_rate": 2e-05,
      "loss": 0.7242,
      "step": 8740
    },
    {
      "epoch": 0.7185677917385235,
      "grad_norm": 0.3685600757598877,
      "learning_rate": 2e-05,
      "loss": 0.6989,
      "step": 8750
    },
    {
      "epoch": 0.719389012071939,
      "grad_norm": 0.3493841588497162,
      "learning_rate": 2e-05,
      "loss": 0.6758,
      "step": 8760
    },
    {
      "epoch": 0.7202102324053543,
      "grad_norm": 0.490925669670105,
      "learning_rate": 2e-05,
      "loss": 0.7258,
      "step": 8770
    },
    {
      "epoch": 0.7210314527387698,
      "grad_norm": 0.3758874535560608,
      "learning_rate": 2e-05,
      "loss": 0.7015,
      "step": 8780
    },
    {
      "epoch": 0.7218526730721853,
      "grad_norm": 0.38454100489616394,
      "learning_rate": 2e-05,
      "loss": 0.6993,
      "step": 8790
    },
    {
      "epoch": 0.7226738934056007,
      "grad_norm": 0.3699193000793457,
      "learning_rate": 2e-05,
      "loss": 0.7126,
      "step": 8800
    },
    {
      "epoch": 0.7234951137390162,
      "grad_norm": 0.38351812958717346,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 8810
    },
    {
      "epoch": 0.7243163340724317,
      "grad_norm": 0.407650887966156,
      "learning_rate": 2e-05,
      "loss": 0.72,
      "step": 8820
    },
    {
      "epoch": 0.725137554405847,
      "grad_norm": 0.4000674784183502,
      "learning_rate": 2e-05,
      "loss": 0.7216,
      "step": 8830
    },
    {
      "epoch": 0.7259587747392625,
      "grad_norm": 0.3942640423774719,
      "learning_rate": 2e-05,
      "loss": 0.7034,
      "step": 8840
    },
    {
      "epoch": 0.726779995072678,
      "grad_norm": 0.3915316164493561,
      "learning_rate": 2e-05,
      "loss": 0.7015,
      "step": 8850
    },
    {
      "epoch": 0.7276012154060935,
      "grad_norm": 0.38977187871932983,
      "learning_rate": 2e-05,
      "loss": 0.6974,
      "step": 8860
    },
    {
      "epoch": 0.7284224357395089,
      "grad_norm": 0.36537256836891174,
      "learning_rate": 2e-05,
      "loss": 0.7006,
      "step": 8870
    },
    {
      "epoch": 0.7292436560729244,
      "grad_norm": 0.40082910656929016,
      "learning_rate": 2e-05,
      "loss": 0.6926,
      "step": 8880
    },
    {
      "epoch": 0.7300648764063398,
      "grad_norm": 0.4046599864959717,
      "learning_rate": 2e-05,
      "loss": 0.6981,
      "step": 8890
    },
    {
      "epoch": 0.7308860967397552,
      "grad_norm": 0.36027175188064575,
      "learning_rate": 2e-05,
      "loss": 0.703,
      "step": 8900
    },
    {
      "epoch": 0.7317073170731707,
      "grad_norm": 0.3701125681400299,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 8910
    },
    {
      "epoch": 0.7325285374065862,
      "grad_norm": 0.36572128534317017,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 8920
    },
    {
      "epoch": 0.7333497577400017,
      "grad_norm": 0.37830886244773865,
      "learning_rate": 2e-05,
      "loss": 0.6813,
      "step": 8930
    },
    {
      "epoch": 0.7341709780734171,
      "grad_norm": 0.36629819869995117,
      "learning_rate": 2e-05,
      "loss": 0.7017,
      "step": 8940
    },
    {
      "epoch": 0.7349921984068326,
      "grad_norm": 0.3819725513458252,
      "learning_rate": 2e-05,
      "loss": 0.7256,
      "step": 8950
    },
    {
      "epoch": 0.735813418740248,
      "grad_norm": 0.3755948841571808,
      "learning_rate": 2e-05,
      "loss": 0.7113,
      "step": 8960
    },
    {
      "epoch": 0.7366346390736634,
      "grad_norm": 0.3997722268104553,
      "learning_rate": 2e-05,
      "loss": 0.692,
      "step": 8970
    },
    {
      "epoch": 0.7374558594070789,
      "grad_norm": 0.37592390179634094,
      "learning_rate": 2e-05,
      "loss": 0.6964,
      "step": 8980
    },
    {
      "epoch": 0.7382770797404944,
      "grad_norm": 0.39605703949928284,
      "learning_rate": 2e-05,
      "loss": 0.713,
      "step": 8990
    },
    {
      "epoch": 0.7390983000739099,
      "grad_norm": 0.38445740938186646,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 9000
    },
    {
      "epoch": 0.7399195204073253,
      "grad_norm": 0.4233265221118927,
      "learning_rate": 2e-05,
      "loss": 0.7268,
      "step": 9010
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 0.39834946393966675,
      "learning_rate": 2e-05,
      "loss": 0.7167,
      "step": 9020
    },
    {
      "epoch": 0.7415619610741562,
      "grad_norm": 0.4134651720523834,
      "learning_rate": 2e-05,
      "loss": 0.6956,
      "step": 9030
    },
    {
      "epoch": 0.7423831814075716,
      "grad_norm": 0.36564573645591736,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 9040
    },
    {
      "epoch": 0.7432044017409871,
      "grad_norm": 0.40667298436164856,
      "learning_rate": 2e-05,
      "loss": 0.6943,
      "step": 9050
    },
    {
      "epoch": 0.7440256220744026,
      "grad_norm": 0.3698982000350952,
      "learning_rate": 2e-05,
      "loss": 0.7012,
      "step": 9060
    },
    {
      "epoch": 0.7448468424078181,
      "grad_norm": 0.39954498410224915,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 9070
    },
    {
      "epoch": 0.7456680627412334,
      "grad_norm": 0.38851791620254517,
      "learning_rate": 2e-05,
      "loss": 0.7185,
      "step": 9080
    },
    {
      "epoch": 0.7464892830746489,
      "grad_norm": 0.3608302175998688,
      "learning_rate": 2e-05,
      "loss": 0.6944,
      "step": 9090
    },
    {
      "epoch": 0.7473105034080644,
      "grad_norm": 0.35260623693466187,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 9100
    },
    {
      "epoch": 0.7481317237414798,
      "grad_norm": 0.3996467888355255,
      "learning_rate": 2e-05,
      "loss": 0.7135,
      "step": 9110
    },
    {
      "epoch": 0.7489529440748953,
      "grad_norm": 0.3852183520793915,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 9120
    },
    {
      "epoch": 0.7497741644083108,
      "grad_norm": 0.4694134294986725,
      "learning_rate": 2e-05,
      "loss": 0.7039,
      "step": 9130
    },
    {
      "epoch": 0.7505953847417262,
      "grad_norm": 0.3672221004962921,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 9140
    },
    {
      "epoch": 0.7514166050751416,
      "grad_norm": 0.36200034618377686,
      "learning_rate": 2e-05,
      "loss": 0.6998,
      "step": 9150
    },
    {
      "epoch": 0.7522378254085571,
      "grad_norm": 0.3780507743358612,
      "learning_rate": 2e-05,
      "loss": 0.6842,
      "step": 9160
    },
    {
      "epoch": 0.7530590457419726,
      "grad_norm": 0.39795684814453125,
      "learning_rate": 2e-05,
      "loss": 0.6999,
      "step": 9170
    },
    {
      "epoch": 0.753880266075388,
      "grad_norm": 0.37800705432891846,
      "learning_rate": 2e-05,
      "loss": 0.7023,
      "step": 9180
    },
    {
      "epoch": 0.7547014864088035,
      "grad_norm": 0.3718207776546478,
      "learning_rate": 2e-05,
      "loss": 0.6961,
      "step": 9190
    },
    {
      "epoch": 0.7555227067422189,
      "grad_norm": 0.3968837857246399,
      "learning_rate": 2e-05,
      "loss": 0.6993,
      "step": 9200
    },
    {
      "epoch": 0.7563439270756344,
      "grad_norm": 0.39101776480674744,
      "learning_rate": 2e-05,
      "loss": 0.7112,
      "step": 9210
    },
    {
      "epoch": 0.7571651474090498,
      "grad_norm": 0.42074018716812134,
      "learning_rate": 2e-05,
      "loss": 0.6963,
      "step": 9220
    },
    {
      "epoch": 0.7579863677424653,
      "grad_norm": 0.41898706555366516,
      "learning_rate": 2e-05,
      "loss": 0.7136,
      "step": 9230
    },
    {
      "epoch": 0.7588075880758808,
      "grad_norm": 0.37843969464302063,
      "learning_rate": 2e-05,
      "loss": 0.7023,
      "step": 9240
    },
    {
      "epoch": 0.7596288084092963,
      "grad_norm": 0.3924032747745514,
      "learning_rate": 2e-05,
      "loss": 0.6614,
      "step": 9250
    },
    {
      "epoch": 0.7604500287427116,
      "grad_norm": 0.36112943291664124,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 9260
    },
    {
      "epoch": 0.7612712490761271,
      "grad_norm": 0.38380053639411926,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 9270
    },
    {
      "epoch": 0.7620924694095426,
      "grad_norm": 0.4086054265499115,
      "learning_rate": 2e-05,
      "loss": 0.7189,
      "step": 9280
    },
    {
      "epoch": 0.762913689742958,
      "grad_norm": 0.3852984607219696,
      "learning_rate": 2e-05,
      "loss": 0.7069,
      "step": 9290
    },
    {
      "epoch": 0.7637349100763735,
      "grad_norm": 0.3730805814266205,
      "learning_rate": 2e-05,
      "loss": 0.6611,
      "step": 9300
    },
    {
      "epoch": 0.764556130409789,
      "grad_norm": 0.3919762074947357,
      "learning_rate": 2e-05,
      "loss": 0.6698,
      "step": 9310
    },
    {
      "epoch": 0.7653773507432045,
      "grad_norm": 0.41542431712150574,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 9320
    },
    {
      "epoch": 0.7661985710766198,
      "grad_norm": 0.4254397749900818,
      "learning_rate": 2e-05,
      "loss": 0.7002,
      "step": 9330
    },
    {
      "epoch": 0.7670197914100353,
      "grad_norm": 0.4249665141105652,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 9340
    },
    {
      "epoch": 0.7678410117434508,
      "grad_norm": 0.3867211639881134,
      "learning_rate": 2e-05,
      "loss": 0.7033,
      "step": 9350
    },
    {
      "epoch": 0.7686622320768662,
      "grad_norm": 0.38667991757392883,
      "learning_rate": 2e-05,
      "loss": 0.7044,
      "step": 9360
    },
    {
      "epoch": 0.7694834524102817,
      "grad_norm": 0.3845638930797577,
      "learning_rate": 2e-05,
      "loss": 0.6729,
      "step": 9370
    },
    {
      "epoch": 0.7703046727436972,
      "grad_norm": 0.4128522574901581,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 9380
    },
    {
      "epoch": 0.7711258930771125,
      "grad_norm": 0.3756697177886963,
      "learning_rate": 2e-05,
      "loss": 0.6763,
      "step": 9390
    },
    {
      "epoch": 0.771947113410528,
      "grad_norm": 0.38811859488487244,
      "learning_rate": 2e-05,
      "loss": 0.6763,
      "step": 9400
    },
    {
      "epoch": 0.7727683337439435,
      "grad_norm": 0.3732301592826843,
      "learning_rate": 2e-05,
      "loss": 0.6738,
      "step": 9410
    },
    {
      "epoch": 0.773589554077359,
      "grad_norm": 0.3953258991241455,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 9420
    },
    {
      "epoch": 0.7744107744107744,
      "grad_norm": 0.4001927971839905,
      "learning_rate": 2e-05,
      "loss": 0.6912,
      "step": 9430
    },
    {
      "epoch": 0.7752319947441899,
      "grad_norm": 0.383240669965744,
      "learning_rate": 2e-05,
      "loss": 0.7075,
      "step": 9440
    },
    {
      "epoch": 0.7760532150776053,
      "grad_norm": 0.3774302005767822,
      "learning_rate": 2e-05,
      "loss": 0.7115,
      "step": 9450
    },
    {
      "epoch": 0.7768744354110207,
      "grad_norm": 0.377055823802948,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 9460
    },
    {
      "epoch": 0.7776956557444362,
      "grad_norm": 0.3846104145050049,
      "learning_rate": 2e-05,
      "loss": 0.7124,
      "step": 9470
    },
    {
      "epoch": 0.7785168760778517,
      "grad_norm": 0.37625259160995483,
      "learning_rate": 2e-05,
      "loss": 0.6701,
      "step": 9480
    },
    {
      "epoch": 0.7793380964112672,
      "grad_norm": 0.3686918616294861,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 9490
    },
    {
      "epoch": 0.7801593167446826,
      "grad_norm": 0.37710243463516235,
      "learning_rate": 2e-05,
      "loss": 0.7005,
      "step": 9500
    },
    {
      "epoch": 0.780980537078098,
      "grad_norm": 0.3910396695137024,
      "learning_rate": 2e-05,
      "loss": 0.735,
      "step": 9510
    },
    {
      "epoch": 0.7818017574115135,
      "grad_norm": 0.38238707184791565,
      "learning_rate": 2e-05,
      "loss": 0.6963,
      "step": 9520
    },
    {
      "epoch": 0.782622977744929,
      "grad_norm": 0.40025103092193604,
      "learning_rate": 2e-05,
      "loss": 0.6913,
      "step": 9530
    },
    {
      "epoch": 0.7834441980783444,
      "grad_norm": 0.3963851034641266,
      "learning_rate": 2e-05,
      "loss": 0.6685,
      "step": 9540
    },
    {
      "epoch": 0.7842654184117599,
      "grad_norm": 0.3912496268749237,
      "learning_rate": 2e-05,
      "loss": 0.72,
      "step": 9550
    },
    {
      "epoch": 0.7850866387451754,
      "grad_norm": 0.39743468165397644,
      "learning_rate": 2e-05,
      "loss": 0.7207,
      "step": 9560
    },
    {
      "epoch": 0.7859078590785907,
      "grad_norm": 0.362433522939682,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 9570
    },
    {
      "epoch": 0.7867290794120062,
      "grad_norm": 0.39104869961738586,
      "learning_rate": 2e-05,
      "loss": 0.6951,
      "step": 9580
    },
    {
      "epoch": 0.7875502997454217,
      "grad_norm": 0.40056341886520386,
      "learning_rate": 2e-05,
      "loss": 0.6953,
      "step": 9590
    },
    {
      "epoch": 0.7883715200788372,
      "grad_norm": 0.38114219903945923,
      "learning_rate": 2e-05,
      "loss": 0.6802,
      "step": 9600
    },
    {
      "epoch": 0.7891927404122526,
      "grad_norm": 0.36840707063674927,
      "learning_rate": 2e-05,
      "loss": 0.6929,
      "step": 9610
    },
    {
      "epoch": 0.7900139607456681,
      "grad_norm": 0.37456604838371277,
      "learning_rate": 2e-05,
      "loss": 0.6956,
      "step": 9620
    },
    {
      "epoch": 0.7908351810790836,
      "grad_norm": 0.5365979075431824,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 9630
    },
    {
      "epoch": 0.7916564014124989,
      "grad_norm": 0.4239230155944824,
      "learning_rate": 2e-05,
      "loss": 0.7177,
      "step": 9640
    },
    {
      "epoch": 0.7924776217459144,
      "grad_norm": 0.3884694278240204,
      "learning_rate": 2e-05,
      "loss": 0.6874,
      "step": 9650
    },
    {
      "epoch": 0.7932988420793299,
      "grad_norm": 0.36684051156044006,
      "learning_rate": 2e-05,
      "loss": 0.6895,
      "step": 9660
    },
    {
      "epoch": 0.7941200624127454,
      "grad_norm": 0.38291963934898376,
      "learning_rate": 2e-05,
      "loss": 0.7097,
      "step": 9670
    },
    {
      "epoch": 0.7949412827461608,
      "grad_norm": 0.3479839861392975,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 9680
    },
    {
      "epoch": 0.7957625030795763,
      "grad_norm": 0.38909345865249634,
      "learning_rate": 2e-05,
      "loss": 0.7036,
      "step": 9690
    },
    {
      "epoch": 0.7965837234129917,
      "grad_norm": 0.3804406523704529,
      "learning_rate": 2e-05,
      "loss": 0.6599,
      "step": 9700
    },
    {
      "epoch": 0.7974049437464071,
      "grad_norm": 0.3979552984237671,
      "learning_rate": 2e-05,
      "loss": 0.7181,
      "step": 9710
    },
    {
      "epoch": 0.7982261640798226,
      "grad_norm": 0.36993950605392456,
      "learning_rate": 2e-05,
      "loss": 0.6944,
      "step": 9720
    },
    {
      "epoch": 0.7990473844132381,
      "grad_norm": 0.3777221441268921,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 9730
    },
    {
      "epoch": 0.7998686047466536,
      "grad_norm": 0.3976139724254608,
      "learning_rate": 2e-05,
      "loss": 0.674,
      "step": 9740
    },
    {
      "epoch": 0.800689825080069,
      "grad_norm": 0.38253989815711975,
      "learning_rate": 2e-05,
      "loss": 0.6553,
      "step": 9750
    },
    {
      "epoch": 0.8015110454134844,
      "grad_norm": 0.3812165856361389,
      "learning_rate": 2e-05,
      "loss": 0.6947,
      "step": 9760
    },
    {
      "epoch": 0.8023322657468999,
      "grad_norm": 0.40065863728523254,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 9770
    },
    {
      "epoch": 0.8031534860803153,
      "grad_norm": 0.4094850718975067,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 9780
    },
    {
      "epoch": 0.8039747064137308,
      "grad_norm": 0.3824274241924286,
      "learning_rate": 2e-05,
      "loss": 0.7127,
      "step": 9790
    },
    {
      "epoch": 0.8047959267471463,
      "grad_norm": 0.42886489629745483,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 9800
    },
    {
      "epoch": 0.8056171470805618,
      "grad_norm": 0.35409966111183167,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 9810
    },
    {
      "epoch": 0.8064383674139771,
      "grad_norm": 0.3598063290119171,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 9820
    },
    {
      "epoch": 0.8072595877473926,
      "grad_norm": 0.3542577922344208,
      "learning_rate": 2e-05,
      "loss": 0.6988,
      "step": 9830
    },
    {
      "epoch": 0.8080808080808081,
      "grad_norm": 0.4234031140804291,
      "learning_rate": 2e-05,
      "loss": 0.7023,
      "step": 9840
    },
    {
      "epoch": 0.8089020284142235,
      "grad_norm": 0.4130217730998993,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 9850
    },
    {
      "epoch": 0.809723248747639,
      "grad_norm": 0.39673882722854614,
      "learning_rate": 2e-05,
      "loss": 0.6855,
      "step": 9860
    },
    {
      "epoch": 0.8105444690810545,
      "grad_norm": 0.37097981572151184,
      "learning_rate": 2e-05,
      "loss": 0.721,
      "step": 9870
    },
    {
      "epoch": 0.8113656894144698,
      "grad_norm": 0.40268978476524353,
      "learning_rate": 2e-05,
      "loss": 0.683,
      "step": 9880
    },
    {
      "epoch": 0.8121869097478853,
      "grad_norm": 0.36814814805984497,
      "learning_rate": 2e-05,
      "loss": 0.705,
      "step": 9890
    },
    {
      "epoch": 0.8130081300813008,
      "grad_norm": 0.38488298654556274,
      "learning_rate": 2e-05,
      "loss": 0.7287,
      "step": 9900
    },
    {
      "epoch": 0.8138293504147163,
      "grad_norm": 0.3959543704986572,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 9910
    },
    {
      "epoch": 0.8146505707481317,
      "grad_norm": 0.3803964853286743,
      "learning_rate": 2e-05,
      "loss": 0.7037,
      "step": 9920
    },
    {
      "epoch": 0.8154717910815472,
      "grad_norm": 0.40059012174606323,
      "learning_rate": 2e-05,
      "loss": 0.7003,
      "step": 9930
    },
    {
      "epoch": 0.8162930114149627,
      "grad_norm": 0.38866984844207764,
      "learning_rate": 2e-05,
      "loss": 0.6793,
      "step": 9940
    },
    {
      "epoch": 0.817114231748378,
      "grad_norm": 0.39701634645462036,
      "learning_rate": 2e-05,
      "loss": 0.695,
      "step": 9950
    },
    {
      "epoch": 0.8179354520817935,
      "grad_norm": 0.3775525689125061,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 9960
    },
    {
      "epoch": 0.818756672415209,
      "grad_norm": 0.375573992729187,
      "learning_rate": 2e-05,
      "loss": 0.6709,
      "step": 9970
    },
    {
      "epoch": 0.8195778927486245,
      "grad_norm": 0.36390286684036255,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 9980
    },
    {
      "epoch": 0.8203991130820399,
      "grad_norm": 0.3717043995857239,
      "learning_rate": 2e-05,
      "loss": 0.6799,
      "step": 9990
    },
    {
      "epoch": 0.8212203334154554,
      "grad_norm": 0.3817295432090759,
      "learning_rate": 2e-05,
      "loss": 0.7041,
      "step": 10000
    },
    {
      "epoch": 0.8220415537488708,
      "grad_norm": 0.3390219807624817,
      "learning_rate": 2e-05,
      "loss": 0.6815,
      "step": 10010
    },
    {
      "epoch": 0.8228627740822863,
      "grad_norm": 0.4109913408756256,
      "learning_rate": 2e-05,
      "loss": 0.7414,
      "step": 10020
    },
    {
      "epoch": 0.8236839944157017,
      "grad_norm": 0.38847294449806213,
      "learning_rate": 2e-05,
      "loss": 0.6885,
      "step": 10030
    },
    {
      "epoch": 0.8245052147491172,
      "grad_norm": 0.3859111964702606,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 10040
    },
    {
      "epoch": 0.8253264350825327,
      "grad_norm": 0.40540292859077454,
      "learning_rate": 2e-05,
      "loss": 0.7022,
      "step": 10050
    },
    {
      "epoch": 0.8261476554159481,
      "grad_norm": 0.3919416666030884,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 10060
    },
    {
      "epoch": 0.8269688757493635,
      "grad_norm": 0.3847336769104004,
      "learning_rate": 2e-05,
      "loss": 0.674,
      "step": 10070
    },
    {
      "epoch": 0.827790096082779,
      "grad_norm": 0.4472779929637909,
      "learning_rate": 2e-05,
      "loss": 0.6877,
      "step": 10080
    },
    {
      "epoch": 0.8286113164161945,
      "grad_norm": 0.3762985169887543,
      "learning_rate": 2e-05,
      "loss": 0.6866,
      "step": 10090
    },
    {
      "epoch": 0.8294325367496099,
      "grad_norm": 0.35252645611763,
      "learning_rate": 2e-05,
      "loss": 0.683,
      "step": 10100
    },
    {
      "epoch": 0.8302537570830254,
      "grad_norm": 0.37249138951301575,
      "learning_rate": 2e-05,
      "loss": 0.7095,
      "step": 10110
    },
    {
      "epoch": 0.8310749774164409,
      "grad_norm": 0.39134690165519714,
      "learning_rate": 2e-05,
      "loss": 0.6966,
      "step": 10120
    },
    {
      "epoch": 0.8318961977498562,
      "grad_norm": 0.348389595746994,
      "learning_rate": 2e-05,
      "loss": 0.707,
      "step": 10130
    },
    {
      "epoch": 0.8327174180832717,
      "grad_norm": 0.39450663328170776,
      "learning_rate": 2e-05,
      "loss": 0.6758,
      "step": 10140
    },
    {
      "epoch": 0.8335386384166872,
      "grad_norm": 0.36050963401794434,
      "learning_rate": 2e-05,
      "loss": 0.6853,
      "step": 10150
    },
    {
      "epoch": 0.8343598587501027,
      "grad_norm": 0.3787517547607422,
      "learning_rate": 2e-05,
      "loss": 0.6829,
      "step": 10160
    },
    {
      "epoch": 0.8351810790835181,
      "grad_norm": 0.39590832591056824,
      "learning_rate": 2e-05,
      "loss": 0.7101,
      "step": 10170
    },
    {
      "epoch": 0.8360022994169336,
      "grad_norm": 0.3791876435279846,
      "learning_rate": 1.999957358086795e-05,
      "loss": 0.6551,
      "step": 10180
    },
    {
      "epoch": 0.836823519750349,
      "grad_norm": 0.3683079183101654,
      "learning_rate": 1.9991993813020483e-05,
      "loss": 0.7334,
      "step": 10190
    },
    {
      "epoch": 0.8376447400837644,
      "grad_norm": 0.40464428067207336,
      "learning_rate": 1.9974946373028628e-05,
      "loss": 0.698,
      "step": 10200
    },
    {
      "epoch": 0.8384659604171799,
      "grad_norm": 0.3870455324649811,
      "learning_rate": 1.994844749501764e-05,
      "loss": 0.714,
      "step": 10210
    },
    {
      "epoch": 0.8392871807505954,
      "grad_norm": 0.3815152645111084,
      "learning_rate": 1.9912522413632988e-05,
      "loss": 0.6796,
      "step": 10220
    },
    {
      "epoch": 0.8401084010840109,
      "grad_norm": 0.3830770254135132,
      "learning_rate": 1.9867205340009656e-05,
      "loss": 0.7021,
      "step": 10230
    },
    {
      "epoch": 0.8409296214174263,
      "grad_norm": 0.37634575366973877,
      "learning_rate": 1.9812539429193164e-05,
      "loss": 0.6814,
      "step": 10240
    },
    {
      "epoch": 0.8417508417508418,
      "grad_norm": 0.4080638289451599,
      "learning_rate": 1.9748576739043423e-05,
      "loss": 0.7002,
      "step": 10250
    },
    {
      "epoch": 0.8425720620842572,
      "grad_norm": 0.363120436668396,
      "learning_rate": 1.967537818066047e-05,
      "loss": 0.7157,
      "step": 10260
    },
    {
      "epoch": 0.8433932824176726,
      "grad_norm": 0.4047775864601135,
      "learning_rate": 1.959301346037941e-05,
      "loss": 0.6961,
      "step": 10270
    },
    {
      "epoch": 0.8442145027510881,
      "grad_norm": 0.40301206707954407,
      "learning_rate": 1.9501561013389646e-05,
      "loss": 0.7178,
      "step": 10280
    },
    {
      "epoch": 0.8450357230845036,
      "grad_norm": 0.3606620132923126,
      "learning_rate": 1.9401107929041795e-05,
      "loss": 0.695,
      "step": 10290
    },
    {
      "epoch": 0.8458569434179191,
      "grad_norm": 0.35096991062164307,
      "learning_rate": 1.9291749867913225e-05,
      "loss": 0.6532,
      "step": 10300
    },
    {
      "epoch": 0.8466781637513345,
      "grad_norm": 0.3759719133377075,
      "learning_rate": 1.91735909707114e-05,
      "loss": 0.6899,
      "step": 10310
    },
    {
      "epoch": 0.8474993840847499,
      "grad_norm": 0.388521283864975,
      "learning_rate": 1.9046743759101545e-05,
      "loss": 0.6793,
      "step": 10320
    },
    {
      "epoch": 0.8483206044181654,
      "grad_norm": 0.36107343435287476,
      "learning_rate": 1.8911329028553324e-05,
      "loss": 0.6951,
      "step": 10330
    },
    {
      "epoch": 0.8491418247515808,
      "grad_norm": 0.38687455654144287,
      "learning_rate": 1.876747573330835e-05,
      "loss": 0.6963,
      "step": 10340
    },
    {
      "epoch": 0.8499630450849963,
      "grad_norm": 0.3718600869178772,
      "learning_rate": 1.8615320863578244e-05,
      "loss": 0.7023,
      "step": 10350
    },
    {
      "epoch": 0.8507842654184118,
      "grad_norm": 0.38907334208488464,
      "learning_rate": 1.8455009315090036e-05,
      "loss": 0.7168,
      "step": 10360
    },
    {
      "epoch": 0.8516054857518273,
      "grad_norm": 0.34664687514305115,
      "learning_rate": 1.8286693751103294e-05,
      "loss": 0.6603,
      "step": 10370
    },
    {
      "epoch": 0.8524267060852426,
      "grad_norm": 0.3432936668395996,
      "learning_rate": 1.811053445703024e-05,
      "loss": 0.7289,
      "step": 10380
    },
    {
      "epoch": 0.8532479264186581,
      "grad_norm": 0.3848487138748169,
      "learning_rate": 1.7926699187797375e-05,
      "loss": 0.6738,
      "step": 10390
    },
    {
      "epoch": 0.8540691467520736,
      "grad_norm": 0.38828855752944946,
      "learning_rate": 1.7735363008093995e-05,
      "loss": 0.6769,
      "step": 10400
    },
    {
      "epoch": 0.854890367085489,
      "grad_norm": 0.39348649978637695,
      "learning_rate": 1.7536708125659577e-05,
      "loss": 0.6894,
      "step": 10410
    },
    {
      "epoch": 0.8557115874189045,
      "grad_norm": 0.34964853525161743,
      "learning_rate": 1.733092371776905e-05,
      "loss": 0.686,
      "step": 10420
    },
    {
      "epoch": 0.85653280775232,
      "grad_norm": 0.3784288465976715,
      "learning_rate": 1.7118205751080915e-05,
      "loss": 0.6924,
      "step": 10430
    },
    {
      "epoch": 0.8573540280857354,
      "grad_norm": 0.410057008266449,
      "learning_rate": 1.6898756795019935e-05,
      "loss": 0.6927,
      "step": 10440
    },
    {
      "epoch": 0.8581752484191508,
      "grad_norm": 0.36901912093162537,
      "learning_rate": 1.667278582887211e-05,
      "loss": 0.6862,
      "step": 10450
    },
    {
      "epoch": 0.8589964687525663,
      "grad_norm": 0.3652726113796234,
      "learning_rate": 1.6440508042775545e-05,
      "loss": 0.6763,
      "step": 10460
    },
    {
      "epoch": 0.8598176890859818,
      "grad_norm": 0.35899096727371216,
      "learning_rate": 1.6202144632796796e-05,
      "loss": 0.6763,
      "step": 10470
    },
    {
      "epoch": 0.8606389094193972,
      "grad_norm": 0.4083310067653656,
      "learning_rate": 1.595792259028783e-05,
      "loss": 0.7072,
      "step": 10480
    },
    {
      "epoch": 0.8614601297528127,
      "grad_norm": 0.34672120213508606,
      "learning_rate": 1.570807448572419e-05,
      "loss": 0.663,
      "step": 10490
    },
    {
      "epoch": 0.8622813500862281,
      "grad_norm": 0.40855562686920166,
      "learning_rate": 1.5452838247230197e-05,
      "loss": 0.6632,
      "step": 10500
    },
    {
      "epoch": 0.8631025704196436,
      "grad_norm": 0.362225204706192,
      "learning_rate": 1.5192456934002106e-05,
      "loss": 0.6739,
      "step": 10510
    },
    {
      "epoch": 0.863923790753059,
      "grad_norm": 0.3722764551639557,
      "learning_rate": 1.492717850484502e-05,
      "loss": 0.6954,
      "step": 10520
    },
    {
      "epoch": 0.8647450110864745,
      "grad_norm": 0.4095909595489502,
      "learning_rate": 1.4657255582043904e-05,
      "loss": 0.722,
      "step": 10530
    },
    {
      "epoch": 0.86556623141989,
      "grad_norm": 0.3980634808540344,
      "learning_rate": 1.4382945210793663e-05,
      "loss": 0.6899,
      "step": 10540
    },
    {
      "epoch": 0.8663874517533054,
      "grad_norm": 0.36202120780944824,
      "learning_rate": 1.4104508614417267e-05,
      "loss": 0.7062,
      "step": 10550
    },
    {
      "epoch": 0.8672086720867209,
      "grad_norm": 0.36383169889450073,
      "learning_rate": 1.382221094560512e-05,
      "loss": 0.6728,
      "step": 10560
    },
    {
      "epoch": 0.8680298924201363,
      "grad_norm": 0.38219350576400757,
      "learning_rate": 1.3536321033912506e-05,
      "loss": 0.6953,
      "step": 10570
    },
    {
      "epoch": 0.8688511127535518,
      "grad_norm": 0.3433822691440582,
      "learning_rate": 1.3247111129755609e-05,
      "loss": 0.684,
      "step": 10580
    },
    {
      "epoch": 0.8696723330869672,
      "grad_norm": 0.3876646161079407,
      "learning_rate": 1.2954856645149868e-05,
      "loss": 0.6778,
      "step": 10590
    },
    {
      "epoch": 0.8704935534203827,
      "grad_norm": 0.4364592134952545,
      "learning_rate": 1.2659835891437553e-05,
      "loss": 0.7096,
      "step": 10600
    },
    {
      "epoch": 0.8713147737537982,
      "grad_norm": 0.36235591769218445,
      "learning_rate": 1.236232981425436e-05,
      "loss": 0.6937,
      "step": 10610
    },
    {
      "epoch": 0.8721359940872137,
      "grad_norm": 0.3763750493526459,
      "learning_rate": 1.2062621725987402e-05,
      "loss": 0.677,
      "step": 10620
    },
    {
      "epoch": 0.872957214420629,
      "grad_norm": 0.4048139154911041,
      "learning_rate": 1.1760997035979318e-05,
      "loss": 0.6566,
      "step": 10630
    },
    {
      "epoch": 0.8737784347540445,
      "grad_norm": 0.3555852770805359,
      "learning_rate": 1.1457742978735514e-05,
      "loss": 0.6969,
      "step": 10640
    },
    {
      "epoch": 0.87459965508746,
      "grad_norm": 0.36527928709983826,
      "learning_rate": 1.1153148340393298e-05,
      "loss": 0.7135,
      "step": 10650
    },
    {
      "epoch": 0.8754208754208754,
      "grad_norm": 0.3571285903453827,
      "learning_rate": 1.0847503183713424e-05,
      "loss": 0.6929,
      "step": 10660
    },
    {
      "epoch": 0.8762420957542909,
      "grad_norm": 0.393326073884964,
      "learning_rate": 1.0541098571855896e-05,
      "loss": 0.6952,
      "step": 10670
    },
    {
      "epoch": 0.8770633160877064,
      "grad_norm": 0.3636074960231781,
      "learning_rate": 1.0234226291203123e-05,
      "loss": 0.7094,
      "step": 10680
    },
    {
      "epoch": 0.8778845364211217,
      "grad_norm": 0.3841739594936371,
      "learning_rate": 9.927178573494333e-06,
      "loss": 0.6513,
      "step": 10690
    },
    {
      "epoch": 0.8787057567545372,
      "grad_norm": 0.37740558385849,
      "learning_rate": 9.620247817535902e-06,
      "loss": 0.6652,
      "step": 10700
    },
    {
      "epoch": 0.8795269770879527,
      "grad_norm": 0.35061416029930115,
      "learning_rate": 9.313726310752581e-06,
      "loss": 0.6802,
      "step": 10710
    },
    {
      "epoch": 0.8803481974213682,
      "grad_norm": 0.3645026385784149,
      "learning_rate": 9.00790595084477e-06,
      "loss": 0.6718,
      "step": 10720
    },
    {
      "epoch": 0.8811694177547836,
      "grad_norm": 0.39668869972229004,
      "learning_rate": 8.703077967816944e-06,
      "loss": 0.6685,
      "step": 10730
    },
    {
      "epoch": 0.8819906380881991,
      "grad_norm": 0.3538977801799774,
      "learning_rate": 8.39953264664192e-06,
      "loss": 0.6696,
      "step": 10740
    },
    {
      "epoch": 0.8828118584216145,
      "grad_norm": 0.3528667986392975,
      "learning_rate": 8.097559050825058e-06,
      "loss": 0.6708,
      "step": 10750
    },
    {
      "epoch": 0.8836330787550299,
      "grad_norm": 0.3845233917236328,
      "learning_rate": 7.79744474713166e-06,
      "loss": 0.6787,
      "step": 10760
    },
    {
      "epoch": 0.8844542990884454,
      "grad_norm": 0.3850906193256378,
      "learning_rate": 7.499475531739696e-06,
      "loss": 0.6829,
      "step": 10770
    },
    {
      "epoch": 0.8852755194218609,
      "grad_norm": 0.36130765080451965,
      "learning_rate": 7.203935158078662e-06,
      "loss": 0.69,
      "step": 10780
    },
    {
      "epoch": 0.8860967397552764,
      "grad_norm": 0.3746684193611145,
      "learning_rate": 6.911105066613709e-06,
      "loss": 0.6686,
      "step": 10790
    },
    {
      "epoch": 0.8869179600886918,
      "grad_norm": 0.3975910544395447,
      "learning_rate": 6.621264116832394e-06,
      "loss": 0.6848,
      "step": 10800
    },
    {
      "epoch": 0.8877391804221072,
      "grad_norm": 0.3766115605831146,
      "learning_rate": 6.334688321689274e-06,
      "loss": 0.6626,
      "step": 10810
    },
    {
      "epoch": 0.8885604007555227,
      "grad_norm": 0.3779941201210022,
      "learning_rate": 6.051650584761237e-06,
      "loss": 0.7057,
      "step": 10820
    },
    {
      "epoch": 0.8893816210889381,
      "grad_norm": 0.3698243498802185,
      "learning_rate": 5.772420440363841e-06,
      "loss": 0.6738,
      "step": 10830
    },
    {
      "epoch": 0.8902028414223536,
      "grad_norm": 0.3637429177761078,
      "learning_rate": 5.497263796876207e-06,
      "loss": 0.6656,
      "step": 10840
    },
    {
      "epoch": 0.8910240617557691,
      "grad_norm": 0.3436393439769745,
      "learning_rate": 5.226442683518813e-06,
      "loss": 0.6739,
      "step": 10850
    },
    {
      "epoch": 0.8918452820891846,
      "grad_norm": 0.36265721917152405,
      "learning_rate": 4.960215000825432e-06,
      "loss": 0.6813,
      "step": 10860
    },
    {
      "epoch": 0.8926665024226,
      "grad_norm": 0.3536547124385834,
      "learning_rate": 4.698834275046711e-06,
      "loss": 0.642,
      "step": 10870
    },
    {
      "epoch": 0.8934877227560154,
      "grad_norm": 0.37773627042770386,
      "learning_rate": 4.442549416719423e-06,
      "loss": 0.6817,
      "step": 10880
    },
    {
      "epoch": 0.8943089430894309,
      "grad_norm": 0.3661157488822937,
      "learning_rate": 4.1916044836311565e-06,
      "loss": 0.6869,
      "step": 10890
    },
    {
      "epoch": 0.8951301634228463,
      "grad_norm": 0.38924604654312134,
      "learning_rate": 3.94623844840626e-06,
      "loss": 0.6888,
      "step": 10900
    },
    {
      "epoch": 0.8959513837562618,
      "grad_norm": 0.3579263985157013,
      "learning_rate": 3.7066849709343467e-06,
      "loss": 0.6651,
      "step": 10910
    },
    {
      "epoch": 0.8967726040896773,
      "grad_norm": 0.36342766880989075,
      "learning_rate": 3.473172175858059e-06,
      "loss": 0.6714,
      "step": 10920
    },
    {
      "epoch": 0.8975938244230928,
      "grad_norm": 0.396899551153183,
      "learning_rate": 3.2459224353320046e-06,
      "loss": 0.688,
      "step": 10930
    },
    {
      "epoch": 0.8984150447565081,
      "grad_norm": 0.3511744439601898,
      "learning_rate": 3.025152157259735e-06,
      "loss": 0.6701,
      "step": 10940
    },
    {
      "epoch": 0.8992362650899236,
      "grad_norm": 0.36920204758644104,
      "learning_rate": 2.811071579210425e-06,
      "loss": 0.6786,
      "step": 10950
    },
    {
      "epoch": 0.9000574854233391,
      "grad_norm": 0.42121604084968567,
      "learning_rate": 2.603884568211513e-06,
      "loss": 0.6743,
      "step": 10960
    },
    {
      "epoch": 0.9008787057567545,
      "grad_norm": 0.3432384133338928,
      "learning_rate": 2.4037884266079434e-06,
      "loss": 0.6579,
      "step": 10970
    },
    {
      "epoch": 0.90169992609017,
      "grad_norm": 0.35237541794776917,
      "learning_rate": 2.2109737041728955e-06,
      "loss": 0.6638,
      "step": 10980
    },
    {
      "epoch": 0.9025211464235855,
      "grad_norm": 0.365495890378952,
      "learning_rate": 2.0256240166489454e-06,
      "loss": 0.7033,
      "step": 10990
    },
    {
      "epoch": 0.9033423667570009,
      "grad_norm": 0.368088036775589,
      "learning_rate": 1.847915870892413e-06,
      "loss": 0.6866,
      "step": 11000
    },
    {
      "epoch": 0.9041635870904163,
      "grad_norm": 0.4236345589160919,
      "learning_rate": 1.6780184967874715e-06,
      "loss": 0.6755,
      "step": 11010
    },
    {
      "epoch": 0.9049848074238318,
      "grad_norm": 0.3528173863887787,
      "learning_rate": 1.5160936860900294e-06,
      "loss": 0.6534,
      "step": 11020
    },
    {
      "epoch": 0.9058060277572473,
      "grad_norm": 0.3741580545902252,
      "learning_rate": 1.3622956383548762e-06,
      "loss": 0.6632,
      "step": 11030
    },
    {
      "epoch": 0.9066272480906628,
      "grad_norm": 0.36561042070388794,
      "learning_rate": 1.216770814092834e-06,
      "loss": 0.6675,
      "step": 11040
    },
    {
      "epoch": 0.9074484684240782,
      "grad_norm": 0.3518229126930237,
      "learning_rate": 1.0796577952977013e-06,
      "loss": 0.6686,
      "step": 11050
    },
    {
      "epoch": 0.9082696887574936,
      "grad_norm": 0.363115519285202,
      "learning_rate": 9.510871534758705e-07,
      "loss": 0.6558,
      "step": 11060
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.3565114736557007,
      "learning_rate": 8.311813253042334e-07,
      "loss": 0.6642,
      "step": 11070
    },
    {
      "epoch": 0.9099121294243245,
      "grad_norm": 0.36958521604537964,
      "learning_rate": 7.200544960348308e-07,
      "loss": 0.6738,
      "step": 11080
    },
    {
      "epoch": 0.91073334975774,
      "grad_norm": 0.34969067573547363,
      "learning_rate": 6.178124907572545e-07,
      "loss": 0.6628,
      "step": 11090
    },
    {
      "epoch": 0.9115545700911555,
      "grad_norm": 0.3693374991416931,
      "learning_rate": 5.245526736223528e-07,
      "loss": 0.6782,
      "step": 11100
    },
    {
      "epoch": 0.912375790424571,
      "grad_norm": 1.0517891645431519,
      "learning_rate": 4.403638551232205e-07,
      "loss": 0.6531,
      "step": 11110
    },
    {
      "epoch": 0.9131970107579863,
      "grad_norm": 0.34900400042533875,
      "learning_rate": 3.6532620752175616e-07,
      "loss": 0.6689,
      "step": 11120
    },
    {
      "epoch": 0.9140182310914018,
      "grad_norm": 0.3487856090068817,
      "learning_rate": 2.9951118850133247e-07,
      "loss": 0.6706,
      "step": 11130
    },
    {
      "epoch": 0.9148394514248173,
      "grad_norm": 0.3808301091194153,
      "learning_rate": 2.429814731182774e-07,
      "loss": 0.6787,
      "step": 11140
    },
    {
      "epoch": 0.9156606717582327,
      "grad_norm": 0.3798769414424896,
      "learning_rate": 1.957908941169813e-07,
      "loss": 0.6784,
      "step": 11150
    },
    {
      "epoch": 0.9164818920916482,
      "grad_norm": 0.37236061692237854,
      "learning_rate": 1.5798439066545254e-07,
      "loss": 0.6738,
      "step": 11160
    },
    {
      "epoch": 0.9173031124250637,
      "grad_norm": 0.4031178057193756,
      "learning_rate": 1.2959796556014471e-07,
      "loss": 0.6637,
      "step": 11170
    },
    {
      "epoch": 0.918124332758479,
      "grad_norm": 0.37646543979644775,
      "learning_rate": 1.1065865094081884e-07,
      "loss": 0.6593,
      "step": 11180
    },
    {
      "epoch": 0.9189455530918945,
      "grad_norm": 0.38803064823150635,
      "learning_rate": 1.011844825480734e-07,
      "loss": 0.6594,
      "step": 11190
    },
    {
      "epoch": 0.9192740412252607,
      "step": 11194,
      "total_flos": 4.547514943786923e+19,
      "train_loss": 0.0038491510292749777,
      "train_runtime": 609.1733,
      "train_samples_per_second": 220.509,
      "train_steps_per_second": 18.376
    }
  ],
  "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
}