{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.45963702061263034,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008212203334154554,
      "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": 0.35734832286834717,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.6148,
      "step": 5090
    },
    {
      "epoch": 0.4188223700418822,
      "grad_norm": 0.4095503091812134,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.7596,
      "step": 5100
    },
    {
      "epoch": 0.4196435903752977,
      "grad_norm": 0.38448286056518555,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.7182,
      "step": 5110
    },
    {
      "epoch": 0.42046481070871317,
      "grad_norm": 0.3865455985069275,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.713,
      "step": 5120
    },
    {
      "epoch": 0.4212860310421286,
      "grad_norm": 0.38850587606430054,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.7444,
      "step": 5130
    },
    {
      "epoch": 0.42210725137554406,
      "grad_norm": 0.4044715464115143,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.6964,
      "step": 5140
    },
    {
      "epoch": 0.42292847170895953,
      "grad_norm": 0.38730761408805847,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.7064,
      "step": 5150
    },
    {
      "epoch": 0.42374969204237495,
      "grad_norm": 0.3933960199356079,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.7303,
      "step": 5160
    },
    {
      "epoch": 0.4245709123757904,
      "grad_norm": 0.39135101437568665,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.7558,
      "step": 5170
    },
    {
      "epoch": 0.4253921327092059,
      "grad_norm": 0.4046337306499481,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.7544,
      "step": 5180
    },
    {
      "epoch": 0.4262133530426213,
      "grad_norm": 0.38508665561676025,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.7189,
      "step": 5190
    },
    {
      "epoch": 0.4270345733760368,
      "grad_norm": 0.37780576944351196,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.7359,
      "step": 5200
    },
    {
      "epoch": 0.42785579370945226,
      "grad_norm": 0.4109767973423004,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.7147,
      "step": 5210
    },
    {
      "epoch": 0.4286770140428677,
      "grad_norm": 0.3744688630104065,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.7294,
      "step": 5220
    },
    {
      "epoch": 0.42949823437628315,
      "grad_norm": 0.431018203496933,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.6863,
      "step": 5230
    },
    {
      "epoch": 0.4303194547096986,
      "grad_norm": 0.39931178092956543,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.7129,
      "step": 5240
    },
    {
      "epoch": 0.43114067504311404,
      "grad_norm": 0.3699224889278412,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.7402,
      "step": 5250
    },
    {
      "epoch": 0.4319618953765295,
      "grad_norm": 0.3900527060031891,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.7182,
      "step": 5260
    },
    {
      "epoch": 0.432783115709945,
      "grad_norm": 0.3892708718776703,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.7153,
      "step": 5270
    },
    {
      "epoch": 0.43360433604336046,
      "grad_norm": 0.3766044080257416,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.7099,
      "step": 5280
    },
    {
      "epoch": 0.4344255563767759,
      "grad_norm": 0.3806709349155426,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.7106,
      "step": 5290
    },
    {
      "epoch": 0.43524677671019135,
      "grad_norm": 0.40260937809944153,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.7318,
      "step": 5300
    },
    {
      "epoch": 0.4360679970436068,
      "grad_norm": 0.4003444314002991,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.7098,
      "step": 5310
    },
    {
      "epoch": 0.43688921737702224,
      "grad_norm": 0.409837007522583,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.707,
      "step": 5320
    },
    {
      "epoch": 0.4377104377104377,
      "grad_norm": 0.37916991114616394,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.7218,
      "step": 5330
    },
    {
      "epoch": 0.4385316580438532,
      "grad_norm": 0.4117047190666199,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.7343,
      "step": 5340
    },
    {
      "epoch": 0.4393528783772686,
      "grad_norm": 0.43916305899620056,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.7001,
      "step": 5350
    },
    {
      "epoch": 0.4401740987106841,
      "grad_norm": 0.4038277864456177,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.7507,
      "step": 5360
    },
    {
      "epoch": 0.44099531904409955,
      "grad_norm": 0.38642510771751404,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.7157,
      "step": 5370
    },
    {
      "epoch": 0.44181653937751497,
      "grad_norm": 0.3973623216152191,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.6975,
      "step": 5380
    },
    {
      "epoch": 0.44263775971093045,
      "grad_norm": 0.3986726403236389,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.7133,
      "step": 5390
    },
    {
      "epoch": 0.4434589800443459,
      "grad_norm": 0.3730117082595825,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.7181,
      "step": 5400
    },
    {
      "epoch": 0.44428020037776134,
      "grad_norm": 0.361395925283432,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.6952,
      "step": 5410
    },
    {
      "epoch": 0.4451014207111768,
      "grad_norm": 0.4060004949569702,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.7367,
      "step": 5420
    },
    {
      "epoch": 0.4459226410445923,
      "grad_norm": 0.373687744140625,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.7083,
      "step": 5430
    },
    {
      "epoch": 0.4467438613780077,
      "grad_norm": 0.3722589313983917,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.6822,
      "step": 5440
    },
    {
      "epoch": 0.4475650817114232,
      "grad_norm": 0.37433305382728577,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.6943,
      "step": 5450
    },
    {
      "epoch": 0.44838630204483865,
      "grad_norm": 0.37297913432121277,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.7278,
      "step": 5460
    },
    {
      "epoch": 0.44920752237825406,
      "grad_norm": 0.3839002549648285,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.7349,
      "step": 5470
    },
    {
      "epoch": 0.45002874271166954,
      "grad_norm": 0.3785195052623749,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.7289,
      "step": 5480
    },
    {
      "epoch": 0.450849963045085,
      "grad_norm": 0.38314253091812134,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.6978,
      "step": 5490
    },
    {
      "epoch": 0.45167118337850043,
      "grad_norm": 0.36921361088752747,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.7092,
      "step": 5500
    },
    {
      "epoch": 0.4524924037119159,
      "grad_norm": 0.3840564489364624,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.7257,
      "step": 5510
    },
    {
      "epoch": 0.4533136240453314,
      "grad_norm": 0.3610815107822418,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.7189,
      "step": 5520
    },
    {
      "epoch": 0.4541348443787468,
      "grad_norm": 0.37645575404167175,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.7016,
      "step": 5530
    },
    {
      "epoch": 0.45495606471216227,
      "grad_norm": 0.3659215569496155,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.7127,
      "step": 5540
    },
    {
      "epoch": 0.45577728504557774,
      "grad_norm": 0.36679306626319885,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.6859,
      "step": 5550
    },
    {
      "epoch": 0.45659850537899316,
      "grad_norm": 0.3753034174442291,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.7195,
      "step": 5560
    },
    {
      "epoch": 0.45741972571240863,
      "grad_norm": 0.37745463848114014,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.6855,
      "step": 5570
    },
    {
      "epoch": 0.4582409460458241,
      "grad_norm": 0.38777971267700195,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.7172,
      "step": 5580
    },
    {
      "epoch": 0.4590621663792395,
      "grad_norm": 0.3852372169494629,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.7356,
      "step": 5590
    },
    {
      "epoch": 0.45963702061263034,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.024309312528385654,
      "train_runtime": 1765.7464,
      "train_samples_per_second": 38.037,
      "train_steps_per_second": 3.17
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}