{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.45764513491414555,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008176614881439084,
      "grad_norm": 10.534998893737793,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 2.2924,
      "step": 10
    },
    {
      "epoch": 0.001635322976287817,
      "grad_norm": 6.318049430847168,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 2.3794,
      "step": 20
    },
    {
      "epoch": 0.0024529844644317253,
      "grad_norm": 2.999676465988159,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.9011,
      "step": 30
    },
    {
      "epoch": 0.003270645952575634,
      "grad_norm": 1.3771625757217407,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.5254,
      "step": 40
    },
    {
      "epoch": 0.004088307440719542,
      "grad_norm": 0.8931247591972351,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.226,
      "step": 50
    },
    {
      "epoch": 0.004905968928863451,
      "grad_norm": 1.1269806623458862,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.0716,
      "step": 60
    },
    {
      "epoch": 0.005723630417007359,
      "grad_norm": 1.387283205986023,
      "learning_rate": 7.814044943820226e-06,
      "loss": 1.011,
      "step": 70
    },
    {
      "epoch": 0.006541291905151268,
      "grad_norm": 0.6601500511169434,
      "learning_rate": 8.932022471910112e-06,
      "loss": 1.0061,
      "step": 80
    },
    {
      "epoch": 0.007358953393295176,
      "grad_norm": 0.6297673583030701,
      "learning_rate": 1.005e-05,
      "loss": 0.9876,
      "step": 90
    },
    {
      "epoch": 0.008176614881439084,
      "grad_norm": 0.5978390574455261,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.93,
      "step": 100
    },
    {
      "epoch": 0.008994276369582993,
      "grad_norm": 0.6085056662559509,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.911,
      "step": 110
    },
    {
      "epoch": 0.009811937857726901,
      "grad_norm": 0.630516767501831,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.9181,
      "step": 120
    },
    {
      "epoch": 0.01062959934587081,
      "grad_norm": 0.659934937953949,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.8963,
      "step": 130
    },
    {
      "epoch": 0.011447260834014717,
      "grad_norm": 0.5893965363502502,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.9024,
      "step": 140
    },
    {
      "epoch": 0.012264922322158627,
      "grad_norm": 0.5959360003471375,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.926,
      "step": 150
    },
    {
      "epoch": 0.013082583810302535,
      "grad_norm": 0.5704044699668884,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.8832,
      "step": 160
    },
    {
      "epoch": 0.013900245298446443,
      "grad_norm": 0.6047025322914124,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.9009,
      "step": 170
    },
    {
      "epoch": 0.014717906786590351,
      "grad_norm": 0.5899404883384705,
      "learning_rate": 2e-05,
      "loss": 0.8876,
      "step": 180
    },
    {
      "epoch": 0.01553556827473426,
      "grad_norm": 0.5657159090042114,
      "learning_rate": 2e-05,
      "loss": 0.8758,
      "step": 190
    },
    {
      "epoch": 0.016353229762878167,
      "grad_norm": 0.5136110186576843,
      "learning_rate": 2e-05,
      "loss": 0.892,
      "step": 200
    },
    {
      "epoch": 0.017170891251022075,
      "grad_norm": 0.5496647357940674,
      "learning_rate": 2e-05,
      "loss": 0.9069,
      "step": 210
    },
    {
      "epoch": 0.017988552739165987,
      "grad_norm": 0.5497331023216248,
      "learning_rate": 2e-05,
      "loss": 0.863,
      "step": 220
    },
    {
      "epoch": 0.018806214227309895,
      "grad_norm": 0.5021716952323914,
      "learning_rate": 2e-05,
      "loss": 0.8952,
      "step": 230
    },
    {
      "epoch": 0.019623875715453803,
      "grad_norm": 0.5393562912940979,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 240
    },
    {
      "epoch": 0.02044153720359771,
      "grad_norm": 0.5259848833084106,
      "learning_rate": 2e-05,
      "loss": 0.8904,
      "step": 250
    },
    {
      "epoch": 0.02125919869174162,
      "grad_norm": 0.562362790107727,
      "learning_rate": 2e-05,
      "loss": 0.922,
      "step": 260
    },
    {
      "epoch": 0.022076860179885527,
      "grad_norm": 0.514157235622406,
      "learning_rate": 2e-05,
      "loss": 0.8547,
      "step": 270
    },
    {
      "epoch": 0.022894521668029435,
      "grad_norm": 0.5613844394683838,
      "learning_rate": 2e-05,
      "loss": 0.8867,
      "step": 280
    },
    {
      "epoch": 0.023712183156173343,
      "grad_norm": 0.5137860774993896,
      "learning_rate": 2e-05,
      "loss": 0.864,
      "step": 290
    },
    {
      "epoch": 0.024529844644317254,
      "grad_norm": 0.5275009274482727,
      "learning_rate": 2e-05,
      "loss": 0.927,
      "step": 300
    },
    {
      "epoch": 0.025347506132461162,
      "grad_norm": 0.5223109722137451,
      "learning_rate": 2e-05,
      "loss": 0.8691,
      "step": 310
    },
    {
      "epoch": 0.02616516762060507,
      "grad_norm": 0.5527063012123108,
      "learning_rate": 2e-05,
      "loss": 0.8648,
      "step": 320
    },
    {
      "epoch": 0.02698282910874898,
      "grad_norm": 0.49740317463874817,
      "learning_rate": 2e-05,
      "loss": 0.8486,
      "step": 330
    },
    {
      "epoch": 0.027800490596892886,
      "grad_norm": 0.49503129720687866,
      "learning_rate": 2e-05,
      "loss": 0.8395,
      "step": 340
    },
    {
      "epoch": 0.028618152085036794,
      "grad_norm": 0.5311452150344849,
      "learning_rate": 2e-05,
      "loss": 0.8358,
      "step": 350
    },
    {
      "epoch": 0.029435813573180702,
      "grad_norm": 0.5130277872085571,
      "learning_rate": 2e-05,
      "loss": 0.8554,
      "step": 360
    },
    {
      "epoch": 0.03025347506132461,
      "grad_norm": 0.46086397767066956,
      "learning_rate": 2e-05,
      "loss": 0.8545,
      "step": 370
    },
    {
      "epoch": 0.03107113654946852,
      "grad_norm": 0.49420565366744995,
      "learning_rate": 2e-05,
      "loss": 0.8401,
      "step": 380
    },
    {
      "epoch": 0.03188879803761243,
      "grad_norm": 0.4804873764514923,
      "learning_rate": 2e-05,
      "loss": 0.8533,
      "step": 390
    },
    {
      "epoch": 0.032706459525756335,
      "grad_norm": 0.4856545329093933,
      "learning_rate": 2e-05,
      "loss": 0.8172,
      "step": 400
    },
    {
      "epoch": 0.033524121013900246,
      "grad_norm": 0.5700660347938538,
      "learning_rate": 2e-05,
      "loss": 0.87,
      "step": 410
    },
    {
      "epoch": 0.03434178250204415,
      "grad_norm": 0.48742008209228516,
      "learning_rate": 2e-05,
      "loss": 0.8653,
      "step": 420
    },
    {
      "epoch": 0.03515944399018806,
      "grad_norm": 0.5448684096336365,
      "learning_rate": 2e-05,
      "loss": 0.8574,
      "step": 430
    },
    {
      "epoch": 0.035977105478331974,
      "grad_norm": 0.9145104289054871,
      "learning_rate": 2e-05,
      "loss": 0.8529,
      "step": 440
    },
    {
      "epoch": 0.03679476696647588,
      "grad_norm": 0.6051157712936401,
      "learning_rate": 2e-05,
      "loss": 0.8492,
      "step": 450
    },
    {
      "epoch": 0.03761242845461979,
      "grad_norm": 0.5123623013496399,
      "learning_rate": 2e-05,
      "loss": 0.8689,
      "step": 460
    },
    {
      "epoch": 0.038430089942763694,
      "grad_norm": 0.49921512603759766,
      "learning_rate": 2e-05,
      "loss": 0.8126,
      "step": 470
    },
    {
      "epoch": 0.039247751430907606,
      "grad_norm": 0.5155230760574341,
      "learning_rate": 2e-05,
      "loss": 0.7817,
      "step": 480
    },
    {
      "epoch": 0.04006541291905151,
      "grad_norm": 0.46784713864326477,
      "learning_rate": 2e-05,
      "loss": 0.8324,
      "step": 490
    },
    {
      "epoch": 0.04088307440719542,
      "grad_norm": 0.4902730882167816,
      "learning_rate": 2e-05,
      "loss": 0.8196,
      "step": 500
    },
    {
      "epoch": 0.041700735895339326,
      "grad_norm": 0.47778692841529846,
      "learning_rate": 2e-05,
      "loss": 0.8537,
      "step": 510
    },
    {
      "epoch": 0.04251839738348324,
      "grad_norm": 0.46416184306144714,
      "learning_rate": 2e-05,
      "loss": 0.815,
      "step": 520
    },
    {
      "epoch": 0.04333605887162715,
      "grad_norm": 0.4957575798034668,
      "learning_rate": 2e-05,
      "loss": 0.8317,
      "step": 530
    },
    {
      "epoch": 0.044153720359771054,
      "grad_norm": 0.4955309331417084,
      "learning_rate": 2e-05,
      "loss": 0.8053,
      "step": 540
    },
    {
      "epoch": 0.044971381847914965,
      "grad_norm": 0.47801926732063293,
      "learning_rate": 2e-05,
      "loss": 0.8318,
      "step": 550
    },
    {
      "epoch": 0.04578904333605887,
      "grad_norm": 0.4526337683200836,
      "learning_rate": 2e-05,
      "loss": 0.8297,
      "step": 560
    },
    {
      "epoch": 0.04660670482420278,
      "grad_norm": 0.5001497864723206,
      "learning_rate": 2e-05,
      "loss": 0.7966,
      "step": 570
    },
    {
      "epoch": 0.047424366312346686,
      "grad_norm": 0.4969994127750397,
      "learning_rate": 2e-05,
      "loss": 0.849,
      "step": 580
    },
    {
      "epoch": 0.0482420278004906,
      "grad_norm": 0.4873654246330261,
      "learning_rate": 2e-05,
      "loss": 0.8098,
      "step": 590
    },
    {
      "epoch": 0.04905968928863451,
      "grad_norm": 0.46147921681404114,
      "learning_rate": 2e-05,
      "loss": 0.8462,
      "step": 600
    },
    {
      "epoch": 0.04987735077677841,
      "grad_norm": 0.48419851064682007,
      "learning_rate": 2e-05,
      "loss": 0.7878,
      "step": 610
    },
    {
      "epoch": 0.050695012264922325,
      "grad_norm": 0.48362842202186584,
      "learning_rate": 2e-05,
      "loss": 0.8309,
      "step": 620
    },
    {
      "epoch": 0.05151267375306623,
      "grad_norm": 0.4815925657749176,
      "learning_rate": 2e-05,
      "loss": 0.8409,
      "step": 630
    },
    {
      "epoch": 0.05233033524121014,
      "grad_norm": 0.5199002027511597,
      "learning_rate": 2e-05,
      "loss": 0.8685,
      "step": 640
    },
    {
      "epoch": 0.053147996729354045,
      "grad_norm": 0.4724765419960022,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 650
    },
    {
      "epoch": 0.05396565821749796,
      "grad_norm": 0.5170514583587646,
      "learning_rate": 2e-05,
      "loss": 0.8361,
      "step": 660
    },
    {
      "epoch": 0.05478331970564186,
      "grad_norm": 0.5067883133888245,
      "learning_rate": 2e-05,
      "loss": 0.8425,
      "step": 670
    },
    {
      "epoch": 0.05560098119378577,
      "grad_norm": 0.43147748708724976,
      "learning_rate": 2e-05,
      "loss": 0.8048,
      "step": 680
    },
    {
      "epoch": 0.056418642681929684,
      "grad_norm": 0.508711576461792,
      "learning_rate": 2e-05,
      "loss": 0.848,
      "step": 690
    },
    {
      "epoch": 0.05723630417007359,
      "grad_norm": 0.4623899459838867,
      "learning_rate": 2e-05,
      "loss": 0.826,
      "step": 700
    },
    {
      "epoch": 0.0580539656582175,
      "grad_norm": 0.48188719153404236,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 710
    },
    {
      "epoch": 0.058871627146361405,
      "grad_norm": 0.4509360194206238,
      "learning_rate": 2e-05,
      "loss": 0.8107,
      "step": 720
    },
    {
      "epoch": 0.059689288634505316,
      "grad_norm": 0.4348622262477875,
      "learning_rate": 2e-05,
      "loss": 0.7997,
      "step": 730
    },
    {
      "epoch": 0.06050695012264922,
      "grad_norm": 0.4793204963207245,
      "learning_rate": 2e-05,
      "loss": 0.8357,
      "step": 740
    },
    {
      "epoch": 0.06132461161079313,
      "grad_norm": 0.47185131907463074,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 750
    },
    {
      "epoch": 0.06214227309893704,
      "grad_norm": 0.5104201436042786,
      "learning_rate": 2e-05,
      "loss": 0.8447,
      "step": 760
    },
    {
      "epoch": 0.06295993458708095,
      "grad_norm": 0.45201629400253296,
      "learning_rate": 2e-05,
      "loss": 0.8265,
      "step": 770
    },
    {
      "epoch": 0.06377759607522486,
      "grad_norm": 1.1489161252975464,
      "learning_rate": 2e-05,
      "loss": 0.8434,
      "step": 780
    },
    {
      "epoch": 0.06459525756336877,
      "grad_norm": 0.5133305788040161,
      "learning_rate": 2e-05,
      "loss": 0.821,
      "step": 790
    },
    {
      "epoch": 0.06541291905151267,
      "grad_norm": 0.45519399642944336,
      "learning_rate": 2e-05,
      "loss": 0.8432,
      "step": 800
    },
    {
      "epoch": 0.06623058053965658,
      "grad_norm": 1.8952999114990234,
      "learning_rate": 2e-05,
      "loss": 0.7693,
      "step": 810
    },
    {
      "epoch": 0.06704824202780049,
      "grad_norm": 0.4610855281352997,
      "learning_rate": 2e-05,
      "loss": 0.8228,
      "step": 820
    },
    {
      "epoch": 0.0678659035159444,
      "grad_norm": 0.49510180950164795,
      "learning_rate": 2e-05,
      "loss": 0.8122,
      "step": 830
    },
    {
      "epoch": 0.0686835650040883,
      "grad_norm": 0.4776274561882019,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 840
    },
    {
      "epoch": 0.06950122649223221,
      "grad_norm": 0.47551512718200684,
      "learning_rate": 2e-05,
      "loss": 0.8284,
      "step": 850
    },
    {
      "epoch": 0.07031888798037612,
      "grad_norm": 0.49425825476646423,
      "learning_rate": 2e-05,
      "loss": 0.802,
      "step": 860
    },
    {
      "epoch": 0.07113654946852004,
      "grad_norm": 0.46732547879219055,
      "learning_rate": 2e-05,
      "loss": 0.822,
      "step": 870
    },
    {
      "epoch": 0.07195421095666395,
      "grad_norm": 0.44407016038894653,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 880
    },
    {
      "epoch": 0.07277187244480784,
      "grad_norm": 0.45862287282943726,
      "learning_rate": 2e-05,
      "loss": 0.8219,
      "step": 890
    },
    {
      "epoch": 0.07358953393295176,
      "grad_norm": 0.48010408878326416,
      "learning_rate": 2e-05,
      "loss": 0.815,
      "step": 900
    },
    {
      "epoch": 0.07440719542109567,
      "grad_norm": 0.4639987349510193,
      "learning_rate": 2e-05,
      "loss": 0.811,
      "step": 910
    },
    {
      "epoch": 0.07522485690923958,
      "grad_norm": 0.4830052852630615,
      "learning_rate": 2e-05,
      "loss": 0.8209,
      "step": 920
    },
    {
      "epoch": 0.07604251839738348,
      "grad_norm": 0.44569143652915955,
      "learning_rate": 2e-05,
      "loss": 0.7969,
      "step": 930
    },
    {
      "epoch": 0.07686017988552739,
      "grad_norm": 0.4402197301387787,
      "learning_rate": 2e-05,
      "loss": 0.8048,
      "step": 940
    },
    {
      "epoch": 0.0776778413736713,
      "grad_norm": 0.4393838346004486,
      "learning_rate": 2e-05,
      "loss": 0.7999,
      "step": 950
    },
    {
      "epoch": 0.07849550286181521,
      "grad_norm": 0.45926329493522644,
      "learning_rate": 2e-05,
      "loss": 0.7993,
      "step": 960
    },
    {
      "epoch": 0.07931316434995912,
      "grad_norm": 0.45618489384651184,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 970
    },
    {
      "epoch": 0.08013082583810302,
      "grad_norm": 0.4323713779449463,
      "learning_rate": 2e-05,
      "loss": 0.8128,
      "step": 980
    },
    {
      "epoch": 0.08094848732624693,
      "grad_norm": 0.4577762484550476,
      "learning_rate": 2e-05,
      "loss": 0.8158,
      "step": 990
    },
    {
      "epoch": 0.08176614881439084,
      "grad_norm": 0.44366759061813354,
      "learning_rate": 2e-05,
      "loss": 0.801,
      "step": 1000
    },
    {
      "epoch": 0.08258381030253475,
      "grad_norm": 0.43710198998451233,
      "learning_rate": 2e-05,
      "loss": 0.8178,
      "step": 1010
    },
    {
      "epoch": 0.08340147179067865,
      "grad_norm": 0.5216193795204163,
      "learning_rate": 2e-05,
      "loss": 0.8356,
      "step": 1020
    },
    {
      "epoch": 0.08421913327882256,
      "grad_norm": 0.6718442440032959,
      "learning_rate": 2e-05,
      "loss": 0.8186,
      "step": 1030
    },
    {
      "epoch": 0.08503679476696648,
      "grad_norm": 0.45956555008888245,
      "learning_rate": 2e-05,
      "loss": 0.7599,
      "step": 1040
    },
    {
      "epoch": 0.08585445625511039,
      "grad_norm": 0.4390004575252533,
      "learning_rate": 2e-05,
      "loss": 0.7939,
      "step": 1050
    },
    {
      "epoch": 0.0866721177432543,
      "grad_norm": 0.458306223154068,
      "learning_rate": 2e-05,
      "loss": 0.8154,
      "step": 1060
    },
    {
      "epoch": 0.0874897792313982,
      "grad_norm": 0.46446549892425537,
      "learning_rate": 2e-05,
      "loss": 0.8013,
      "step": 1070
    },
    {
      "epoch": 0.08830744071954211,
      "grad_norm": 0.46321365237236023,
      "learning_rate": 2e-05,
      "loss": 0.7987,
      "step": 1080
    },
    {
      "epoch": 0.08912510220768602,
      "grad_norm": 0.4468840956687927,
      "learning_rate": 2e-05,
      "loss": 0.8049,
      "step": 1090
    },
    {
      "epoch": 0.08994276369582993,
      "grad_norm": 0.4241533577442169,
      "learning_rate": 2e-05,
      "loss": 0.7683,
      "step": 1100
    },
    {
      "epoch": 0.09076042518397383,
      "grad_norm": 0.4573608934879303,
      "learning_rate": 2e-05,
      "loss": 0.7942,
      "step": 1110
    },
    {
      "epoch": 0.09157808667211774,
      "grad_norm": 0.4434383809566498,
      "learning_rate": 2e-05,
      "loss": 0.8005,
      "step": 1120
    },
    {
      "epoch": 0.09239574816026165,
      "grad_norm": 0.5078239440917969,
      "learning_rate": 2e-05,
      "loss": 0.8192,
      "step": 1130
    },
    {
      "epoch": 0.09321340964840556,
      "grad_norm": 0.4617445766925812,
      "learning_rate": 2e-05,
      "loss": 0.804,
      "step": 1140
    },
    {
      "epoch": 0.09403107113654947,
      "grad_norm": 0.43501752614974976,
      "learning_rate": 2e-05,
      "loss": 0.8172,
      "step": 1150
    },
    {
      "epoch": 0.09484873262469337,
      "grad_norm": 0.4166872501373291,
      "learning_rate": 2e-05,
      "loss": 0.8197,
      "step": 1160
    },
    {
      "epoch": 0.09566639411283728,
      "grad_norm": 0.655982494354248,
      "learning_rate": 2e-05,
      "loss": 0.789,
      "step": 1170
    },
    {
      "epoch": 0.0964840556009812,
      "grad_norm": 0.41751641035079956,
      "learning_rate": 2e-05,
      "loss": 0.7995,
      "step": 1180
    },
    {
      "epoch": 0.0973017170891251,
      "grad_norm": 0.470846027135849,
      "learning_rate": 2e-05,
      "loss": 0.8322,
      "step": 1190
    },
    {
      "epoch": 0.09811937857726902,
      "grad_norm": 0.4337841272354126,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 1200
    },
    {
      "epoch": 0.09893704006541292,
      "grad_norm": 0.47191646695137024,
      "learning_rate": 2e-05,
      "loss": 0.8039,
      "step": 1210
    },
    {
      "epoch": 0.09975470155355683,
      "grad_norm": 0.43299242854118347,
      "learning_rate": 2e-05,
      "loss": 0.8096,
      "step": 1220
    },
    {
      "epoch": 0.10057236304170074,
      "grad_norm": 0.4392620623111725,
      "learning_rate": 2e-05,
      "loss": 0.811,
      "step": 1230
    },
    {
      "epoch": 0.10139002452984465,
      "grad_norm": 0.4958798885345459,
      "learning_rate": 2e-05,
      "loss": 0.821,
      "step": 1240
    },
    {
      "epoch": 0.10220768601798855,
      "grad_norm": 0.449624240398407,
      "learning_rate": 2e-05,
      "loss": 0.8024,
      "step": 1250
    },
    {
      "epoch": 0.10302534750613246,
      "grad_norm": 0.5577616095542908,
      "learning_rate": 2e-05,
      "loss": 0.7876,
      "step": 1260
    },
    {
      "epoch": 0.10384300899427637,
      "grad_norm": 0.4583219289779663,
      "learning_rate": 2e-05,
      "loss": 0.7821,
      "step": 1270
    },
    {
      "epoch": 0.10466067048242028,
      "grad_norm": 0.4646966755390167,
      "learning_rate": 2e-05,
      "loss": 0.791,
      "step": 1280
    },
    {
      "epoch": 0.1054783319705642,
      "grad_norm": 0.4247964918613434,
      "learning_rate": 2e-05,
      "loss": 0.78,
      "step": 1290
    },
    {
      "epoch": 0.10629599345870809,
      "grad_norm": 0.46946001052856445,
      "learning_rate": 2e-05,
      "loss": 0.7948,
      "step": 1300
    },
    {
      "epoch": 0.107113654946852,
      "grad_norm": 0.43030065298080444,
      "learning_rate": 2e-05,
      "loss": 0.8059,
      "step": 1310
    },
    {
      "epoch": 0.10793131643499591,
      "grad_norm": 0.4443809986114502,
      "learning_rate": 2e-05,
      "loss": 0.7681,
      "step": 1320
    },
    {
      "epoch": 0.10874897792313983,
      "grad_norm": 0.4212033748626709,
      "learning_rate": 2e-05,
      "loss": 0.7689,
      "step": 1330
    },
    {
      "epoch": 0.10956663941128372,
      "grad_norm": 0.486328125,
      "learning_rate": 2e-05,
      "loss": 0.7612,
      "step": 1340
    },
    {
      "epoch": 0.11038430089942763,
      "grad_norm": 0.43208643794059753,
      "learning_rate": 2e-05,
      "loss": 0.8189,
      "step": 1350
    },
    {
      "epoch": 0.11120196238757155,
      "grad_norm": 0.4304050803184509,
      "learning_rate": 2e-05,
      "loss": 0.7886,
      "step": 1360
    },
    {
      "epoch": 0.11201962387571546,
      "grad_norm": 0.4358374774456024,
      "learning_rate": 2e-05,
      "loss": 0.7838,
      "step": 1370
    },
    {
      "epoch": 0.11283728536385937,
      "grad_norm": 0.4316851794719696,
      "learning_rate": 2e-05,
      "loss": 0.7945,
      "step": 1380
    },
    {
      "epoch": 0.11365494685200327,
      "grad_norm": 0.43443015217781067,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 1390
    },
    {
      "epoch": 0.11447260834014718,
      "grad_norm": 0.4291702210903168,
      "learning_rate": 2e-05,
      "loss": 0.8028,
      "step": 1400
    },
    {
      "epoch": 0.11529026982829109,
      "grad_norm": 0.4630155861377716,
      "learning_rate": 2e-05,
      "loss": 0.7724,
      "step": 1410
    },
    {
      "epoch": 0.116107931316435,
      "grad_norm": 0.4348618686199188,
      "learning_rate": 2e-05,
      "loss": 0.7816,
      "step": 1420
    },
    {
      "epoch": 0.1169255928045789,
      "grad_norm": 0.4198819100856781,
      "learning_rate": 2e-05,
      "loss": 0.7851,
      "step": 1430
    },
    {
      "epoch": 0.11774325429272281,
      "grad_norm": 0.4539889991283417,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 1440
    },
    {
      "epoch": 0.11856091578086672,
      "grad_norm": 0.4621836245059967,
      "learning_rate": 2e-05,
      "loss": 0.7858,
      "step": 1450
    },
    {
      "epoch": 0.11937857726901063,
      "grad_norm": 0.44036760926246643,
      "learning_rate": 2e-05,
      "loss": 0.7865,
      "step": 1460
    },
    {
      "epoch": 0.12019623875715454,
      "grad_norm": 0.4228440225124359,
      "learning_rate": 2e-05,
      "loss": 0.7579,
      "step": 1470
    },
    {
      "epoch": 0.12101390024529844,
      "grad_norm": 0.435478150844574,
      "learning_rate": 2e-05,
      "loss": 0.7823,
      "step": 1480
    },
    {
      "epoch": 0.12183156173344235,
      "grad_norm": 0.4398757815361023,
      "learning_rate": 2e-05,
      "loss": 0.7727,
      "step": 1490
    },
    {
      "epoch": 0.12264922322158626,
      "grad_norm": 0.42032548785209656,
      "learning_rate": 2e-05,
      "loss": 0.7772,
      "step": 1500
    },
    {
      "epoch": 0.12346688470973018,
      "grad_norm": 1.2638468742370605,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 1510
    },
    {
      "epoch": 0.12428454619787407,
      "grad_norm": 0.4492338001728058,
      "learning_rate": 2e-05,
      "loss": 0.7976,
      "step": 1520
    },
    {
      "epoch": 0.12510220768601799,
      "grad_norm": 0.4194067120552063,
      "learning_rate": 2e-05,
      "loss": 0.7729,
      "step": 1530
    },
    {
      "epoch": 0.1259198691741619,
      "grad_norm": 0.4248548150062561,
      "learning_rate": 2e-05,
      "loss": 0.7832,
      "step": 1540
    },
    {
      "epoch": 0.1267375306623058,
      "grad_norm": 0.44248121976852417,
      "learning_rate": 2e-05,
      "loss": 0.7993,
      "step": 1550
    },
    {
      "epoch": 0.12755519215044972,
      "grad_norm": 0.4308151304721832,
      "learning_rate": 2e-05,
      "loss": 0.8292,
      "step": 1560
    },
    {
      "epoch": 0.12837285363859363,
      "grad_norm": 0.46226322650909424,
      "learning_rate": 2e-05,
      "loss": 0.7998,
      "step": 1570
    },
    {
      "epoch": 0.12919051512673754,
      "grad_norm": 0.6823664307594299,
      "learning_rate": 2e-05,
      "loss": 0.7797,
      "step": 1580
    },
    {
      "epoch": 0.13000817661488143,
      "grad_norm": 0.49192625284194946,
      "learning_rate": 2e-05,
      "loss": 0.792,
      "step": 1590
    },
    {
      "epoch": 0.13082583810302534,
      "grad_norm": 0.46564188599586487,
      "learning_rate": 2e-05,
      "loss": 0.7812,
      "step": 1600
    },
    {
      "epoch": 0.13164349959116925,
      "grad_norm": 0.4175106883049011,
      "learning_rate": 2e-05,
      "loss": 0.8088,
      "step": 1610
    },
    {
      "epoch": 0.13246116107931316,
      "grad_norm": 0.43853816390037537,
      "learning_rate": 2e-05,
      "loss": 0.7881,
      "step": 1620
    },
    {
      "epoch": 0.13327882256745707,
      "grad_norm": 0.48302924633026123,
      "learning_rate": 2e-05,
      "loss": 0.7913,
      "step": 1630
    },
    {
      "epoch": 0.13409648405560098,
      "grad_norm": 0.6604704856872559,
      "learning_rate": 2e-05,
      "loss": 0.7709,
      "step": 1640
    },
    {
      "epoch": 0.1349141455437449,
      "grad_norm": 0.523360550403595,
      "learning_rate": 2e-05,
      "loss": 0.7948,
      "step": 1650
    },
    {
      "epoch": 0.1357318070318888,
      "grad_norm": 0.41866135597229004,
      "learning_rate": 2e-05,
      "loss": 0.7374,
      "step": 1660
    },
    {
      "epoch": 0.13654946852003272,
      "grad_norm": 0.43435412645339966,
      "learning_rate": 2e-05,
      "loss": 0.7516,
      "step": 1670
    },
    {
      "epoch": 0.1373671300081766,
      "grad_norm": 0.4378770887851715,
      "learning_rate": 2e-05,
      "loss": 0.7793,
      "step": 1680
    },
    {
      "epoch": 0.1381847914963205,
      "grad_norm": 0.4642718434333801,
      "learning_rate": 2e-05,
      "loss": 0.7891,
      "step": 1690
    },
    {
      "epoch": 0.13900245298446443,
      "grad_norm": 0.4553477168083191,
      "learning_rate": 2e-05,
      "loss": 0.7693,
      "step": 1700
    },
    {
      "epoch": 0.13982011447260834,
      "grad_norm": 0.44605308771133423,
      "learning_rate": 2e-05,
      "loss": 0.7894,
      "step": 1710
    },
    {
      "epoch": 0.14063777596075225,
      "grad_norm": 0.4435379207134247,
      "learning_rate": 2e-05,
      "loss": 0.7834,
      "step": 1720
    },
    {
      "epoch": 0.14145543744889616,
      "grad_norm": 0.43862536549568176,
      "learning_rate": 2e-05,
      "loss": 0.7783,
      "step": 1730
    },
    {
      "epoch": 0.14227309893704007,
      "grad_norm": 0.3968248963356018,
      "learning_rate": 2e-05,
      "loss": 0.7753,
      "step": 1740
    },
    {
      "epoch": 0.14309076042518398,
      "grad_norm": 0.43859657645225525,
      "learning_rate": 2e-05,
      "loss": 0.8167,
      "step": 1750
    },
    {
      "epoch": 0.1439084219133279,
      "grad_norm": 0.4324052035808563,
      "learning_rate": 2e-05,
      "loss": 0.7865,
      "step": 1760
    },
    {
      "epoch": 0.14472608340147178,
      "grad_norm": 0.43473488092422485,
      "learning_rate": 2e-05,
      "loss": 0.8224,
      "step": 1770
    },
    {
      "epoch": 0.1455437448896157,
      "grad_norm": 0.42597419023513794,
      "learning_rate": 2e-05,
      "loss": 0.7985,
      "step": 1780
    },
    {
      "epoch": 0.1463614063777596,
      "grad_norm": 0.48900550603866577,
      "learning_rate": 2e-05,
      "loss": 0.7562,
      "step": 1790
    },
    {
      "epoch": 0.1471790678659035,
      "grad_norm": 0.4394257068634033,
      "learning_rate": 2e-05,
      "loss": 0.7721,
      "step": 1800
    },
    {
      "epoch": 0.14799672935404742,
      "grad_norm": 0.43295830488204956,
      "learning_rate": 2e-05,
      "loss": 0.7727,
      "step": 1810
    },
    {
      "epoch": 0.14881439084219134,
      "grad_norm": 0.4200812578201294,
      "learning_rate": 2e-05,
      "loss": 0.7866,
      "step": 1820
    },
    {
      "epoch": 0.14963205233033525,
      "grad_norm": 0.4190243184566498,
      "learning_rate": 2e-05,
      "loss": 0.7765,
      "step": 1830
    },
    {
      "epoch": 0.15044971381847916,
      "grad_norm": 0.45326995849609375,
      "learning_rate": 2e-05,
      "loss": 0.7871,
      "step": 1840
    },
    {
      "epoch": 0.15126737530662307,
      "grad_norm": 0.4055960476398468,
      "learning_rate": 2e-05,
      "loss": 0.7632,
      "step": 1850
    },
    {
      "epoch": 0.15208503679476695,
      "grad_norm": 0.48894283175468445,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 1860
    },
    {
      "epoch": 0.15290269828291086,
      "grad_norm": 0.4656607508659363,
      "learning_rate": 2e-05,
      "loss": 0.7726,
      "step": 1870
    },
    {
      "epoch": 0.15372035977105478,
      "grad_norm": 0.4367874562740326,
      "learning_rate": 2e-05,
      "loss": 0.7772,
      "step": 1880
    },
    {
      "epoch": 0.1545380212591987,
      "grad_norm": 0.4750969707965851,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 1890
    },
    {
      "epoch": 0.1553556827473426,
      "grad_norm": 0.4409693479537964,
      "learning_rate": 2e-05,
      "loss": 0.7904,
      "step": 1900
    },
    {
      "epoch": 0.1561733442354865,
      "grad_norm": 0.4396913945674896,
      "learning_rate": 2e-05,
      "loss": 0.7597,
      "step": 1910
    },
    {
      "epoch": 0.15699100572363042,
      "grad_norm": 0.41550886631011963,
      "learning_rate": 2e-05,
      "loss": 0.7812,
      "step": 1920
    },
    {
      "epoch": 0.15780866721177433,
      "grad_norm": 0.42665189504623413,
      "learning_rate": 2e-05,
      "loss": 0.7871,
      "step": 1930
    },
    {
      "epoch": 0.15862632869991825,
      "grad_norm": 0.4286573827266693,
      "learning_rate": 2e-05,
      "loss": 0.7736,
      "step": 1940
    },
    {
      "epoch": 0.15944399018806213,
      "grad_norm": 0.42506206035614014,
      "learning_rate": 2e-05,
      "loss": 0.7784,
      "step": 1950
    },
    {
      "epoch": 0.16026165167620604,
      "grad_norm": 0.46702292561531067,
      "learning_rate": 2e-05,
      "loss": 0.7703,
      "step": 1960
    },
    {
      "epoch": 0.16107931316434995,
      "grad_norm": 0.43847423791885376,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 1970
    },
    {
      "epoch": 0.16189697465249386,
      "grad_norm": 0.4711706042289734,
      "learning_rate": 2e-05,
      "loss": 0.7657,
      "step": 1980
    },
    {
      "epoch": 0.16271463614063778,
      "grad_norm": 0.5090168118476868,
      "learning_rate": 2e-05,
      "loss": 0.7592,
      "step": 1990
    },
    {
      "epoch": 0.1635322976287817,
      "grad_norm": 0.46786943078041077,
      "learning_rate": 2e-05,
      "loss": 0.7921,
      "step": 2000
    },
    {
      "epoch": 0.1643499591169256,
      "grad_norm": 0.4125913977622986,
      "learning_rate": 2e-05,
      "loss": 0.7831,
      "step": 2010
    },
    {
      "epoch": 0.1651676206050695,
      "grad_norm": 0.4442942440509796,
      "learning_rate": 2e-05,
      "loss": 0.7536,
      "step": 2020
    },
    {
      "epoch": 0.16598528209321342,
      "grad_norm": 0.4498803913593292,
      "learning_rate": 2e-05,
      "loss": 0.7825,
      "step": 2030
    },
    {
      "epoch": 0.1668029435813573,
      "grad_norm": 0.4364321529865265,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 2040
    },
    {
      "epoch": 0.16762060506950122,
      "grad_norm": 0.4316538870334625,
      "learning_rate": 2e-05,
      "loss": 0.7791,
      "step": 2050
    },
    {
      "epoch": 0.16843826655764513,
      "grad_norm": 0.45543980598449707,
      "learning_rate": 2e-05,
      "loss": 0.7882,
      "step": 2060
    },
    {
      "epoch": 0.16925592804578904,
      "grad_norm": 0.4247235953807831,
      "learning_rate": 2e-05,
      "loss": 0.7986,
      "step": 2070
    },
    {
      "epoch": 0.17007358953393295,
      "grad_norm": 0.4491972029209137,
      "learning_rate": 2e-05,
      "loss": 0.8298,
      "step": 2080
    },
    {
      "epoch": 0.17089125102207686,
      "grad_norm": 0.4386272132396698,
      "learning_rate": 2e-05,
      "loss": 0.7572,
      "step": 2090
    },
    {
      "epoch": 0.17170891251022077,
      "grad_norm": 0.4444292485713959,
      "learning_rate": 2e-05,
      "loss": 0.7801,
      "step": 2100
    },
    {
      "epoch": 0.17252657399836469,
      "grad_norm": 0.4270576238632202,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 2110
    },
    {
      "epoch": 0.1733442354865086,
      "grad_norm": 0.436068594455719,
      "learning_rate": 2e-05,
      "loss": 0.7444,
      "step": 2120
    },
    {
      "epoch": 0.17416189697465248,
      "grad_norm": 0.42339572310447693,
      "learning_rate": 2e-05,
      "loss": 0.7805,
      "step": 2130
    },
    {
      "epoch": 0.1749795584627964,
      "grad_norm": 0.4199734628200531,
      "learning_rate": 2e-05,
      "loss": 0.7519,
      "step": 2140
    },
    {
      "epoch": 0.1757972199509403,
      "grad_norm": 0.4218316376209259,
      "learning_rate": 2e-05,
      "loss": 0.7825,
      "step": 2150
    },
    {
      "epoch": 0.17661488143908421,
      "grad_norm": 0.417795330286026,
      "learning_rate": 2e-05,
      "loss": 0.7871,
      "step": 2160
    },
    {
      "epoch": 0.17743254292722813,
      "grad_norm": 0.4626673460006714,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 2170
    },
    {
      "epoch": 0.17825020441537204,
      "grad_norm": 0.468065470457077,
      "learning_rate": 2e-05,
      "loss": 0.7915,
      "step": 2180
    },
    {
      "epoch": 0.17906786590351595,
      "grad_norm": 0.4166199564933777,
      "learning_rate": 2e-05,
      "loss": 0.7726,
      "step": 2190
    },
    {
      "epoch": 0.17988552739165986,
      "grad_norm": 0.42800837755203247,
      "learning_rate": 2e-05,
      "loss": 0.7725,
      "step": 2200
    },
    {
      "epoch": 0.18070318887980377,
      "grad_norm": 0.4120378792285919,
      "learning_rate": 2e-05,
      "loss": 0.743,
      "step": 2210
    },
    {
      "epoch": 0.18152085036794766,
      "grad_norm": 0.5455228686332703,
      "learning_rate": 2e-05,
      "loss": 0.7318,
      "step": 2220
    },
    {
      "epoch": 0.18233851185609157,
      "grad_norm": 0.4296571612358093,
      "learning_rate": 2e-05,
      "loss": 0.7928,
      "step": 2230
    },
    {
      "epoch": 0.18315617334423548,
      "grad_norm": 0.4547067880630493,
      "learning_rate": 2e-05,
      "loss": 0.7747,
      "step": 2240
    },
    {
      "epoch": 0.1839738348323794,
      "grad_norm": 0.4168989956378937,
      "learning_rate": 2e-05,
      "loss": 0.7923,
      "step": 2250
    },
    {
      "epoch": 0.1847914963205233,
      "grad_norm": 0.44426074624061584,
      "learning_rate": 2e-05,
      "loss": 0.7748,
      "step": 2260
    },
    {
      "epoch": 0.1856091578086672,
      "grad_norm": 0.4370101988315582,
      "learning_rate": 2e-05,
      "loss": 0.7758,
      "step": 2270
    },
    {
      "epoch": 0.18642681929681112,
      "grad_norm": 0.49895063042640686,
      "learning_rate": 2e-05,
      "loss": 0.7619,
      "step": 2280
    },
    {
      "epoch": 0.18724448078495504,
      "grad_norm": 0.40498825907707214,
      "learning_rate": 2e-05,
      "loss": 0.7709,
      "step": 2290
    },
    {
      "epoch": 0.18806214227309895,
      "grad_norm": 0.4341062903404236,
      "learning_rate": 2e-05,
      "loss": 0.7862,
      "step": 2300
    },
    {
      "epoch": 0.18887980376124286,
      "grad_norm": 0.46437665820121765,
      "learning_rate": 2e-05,
      "loss": 0.8152,
      "step": 2310
    },
    {
      "epoch": 0.18969746524938674,
      "grad_norm": 0.43534669280052185,
      "learning_rate": 2e-05,
      "loss": 0.764,
      "step": 2320
    },
    {
      "epoch": 0.19051512673753065,
      "grad_norm": 0.44681885838508606,
      "learning_rate": 2e-05,
      "loss": 0.7873,
      "step": 2330
    },
    {
      "epoch": 0.19133278822567457,
      "grad_norm": 0.45268961787223816,
      "learning_rate": 2e-05,
      "loss": 0.7995,
      "step": 2340
    },
    {
      "epoch": 0.19215044971381848,
      "grad_norm": 0.41534167528152466,
      "learning_rate": 2e-05,
      "loss": 0.7722,
      "step": 2350
    },
    {
      "epoch": 0.1929681112019624,
      "grad_norm": 0.44464072585105896,
      "learning_rate": 2e-05,
      "loss": 0.7842,
      "step": 2360
    },
    {
      "epoch": 0.1937857726901063,
      "grad_norm": 0.4211314618587494,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 2370
    },
    {
      "epoch": 0.1946034341782502,
      "grad_norm": 0.44980642199516296,
      "learning_rate": 2e-05,
      "loss": 0.7635,
      "step": 2380
    },
    {
      "epoch": 0.19542109566639412,
      "grad_norm": 0.43298453092575073,
      "learning_rate": 2e-05,
      "loss": 0.7998,
      "step": 2390
    },
    {
      "epoch": 0.19623875715453803,
      "grad_norm": 0.4239964485168457,
      "learning_rate": 2e-05,
      "loss": 0.7589,
      "step": 2400
    },
    {
      "epoch": 0.19705641864268192,
      "grad_norm": 0.4066570997238159,
      "learning_rate": 2e-05,
      "loss": 0.779,
      "step": 2410
    },
    {
      "epoch": 0.19787408013082583,
      "grad_norm": 0.47745639085769653,
      "learning_rate": 2e-05,
      "loss": 0.7877,
      "step": 2420
    },
    {
      "epoch": 0.19869174161896974,
      "grad_norm": 0.4101332426071167,
      "learning_rate": 2e-05,
      "loss": 0.8011,
      "step": 2430
    },
    {
      "epoch": 0.19950940310711365,
      "grad_norm": 0.4742473065853119,
      "learning_rate": 2e-05,
      "loss": 0.7922,
      "step": 2440
    },
    {
      "epoch": 0.20032706459525756,
      "grad_norm": 0.4261021316051483,
      "learning_rate": 2e-05,
      "loss": 0.751,
      "step": 2450
    },
    {
      "epoch": 0.20114472608340148,
      "grad_norm": 0.5145176649093628,
      "learning_rate": 2e-05,
      "loss": 0.7285,
      "step": 2460
    },
    {
      "epoch": 0.2019623875715454,
      "grad_norm": 0.40921565890312195,
      "learning_rate": 2e-05,
      "loss": 0.7871,
      "step": 2470
    },
    {
      "epoch": 0.2027800490596893,
      "grad_norm": 0.44564947485923767,
      "learning_rate": 2e-05,
      "loss": 0.8052,
      "step": 2480
    },
    {
      "epoch": 0.2035977105478332,
      "grad_norm": 0.42048290371894836,
      "learning_rate": 2e-05,
      "loss": 0.8011,
      "step": 2490
    },
    {
      "epoch": 0.2044153720359771,
      "grad_norm": 0.408529669046402,
      "learning_rate": 2e-05,
      "loss": 0.7733,
      "step": 2500
    },
    {
      "epoch": 0.205233033524121,
      "grad_norm": 0.42771342396736145,
      "learning_rate": 2e-05,
      "loss": 0.7809,
      "step": 2510
    },
    {
      "epoch": 0.20605069501226492,
      "grad_norm": 0.39116600155830383,
      "learning_rate": 2e-05,
      "loss": 0.7621,
      "step": 2520
    },
    {
      "epoch": 0.20686835650040883,
      "grad_norm": 0.43292510509490967,
      "learning_rate": 2e-05,
      "loss": 0.7797,
      "step": 2530
    },
    {
      "epoch": 0.20768601798855274,
      "grad_norm": 0.40812504291534424,
      "learning_rate": 2e-05,
      "loss": 0.7545,
      "step": 2540
    },
    {
      "epoch": 0.20850367947669665,
      "grad_norm": 0.416536420583725,
      "learning_rate": 2e-05,
      "loss": 0.7432,
      "step": 2550
    },
    {
      "epoch": 0.20932134096484056,
      "grad_norm": 0.3991974890232086,
      "learning_rate": 2e-05,
      "loss": 0.784,
      "step": 2560
    },
    {
      "epoch": 0.21013900245298447,
      "grad_norm": 0.4071870744228363,
      "learning_rate": 2e-05,
      "loss": 0.7697,
      "step": 2570
    },
    {
      "epoch": 0.2109566639411284,
      "grad_norm": 0.407435804605484,
      "learning_rate": 2e-05,
      "loss": 0.7512,
      "step": 2580
    },
    {
      "epoch": 0.21177432542927227,
      "grad_norm": 0.3893747925758362,
      "learning_rate": 2e-05,
      "loss": 0.7493,
      "step": 2590
    },
    {
      "epoch": 0.21259198691741618,
      "grad_norm": 0.4174593985080719,
      "learning_rate": 2e-05,
      "loss": 0.7752,
      "step": 2600
    },
    {
      "epoch": 0.2134096484055601,
      "grad_norm": 0.4031468629837036,
      "learning_rate": 2e-05,
      "loss": 0.7418,
      "step": 2610
    },
    {
      "epoch": 0.214227309893704,
      "grad_norm": 0.44015318155288696,
      "learning_rate": 2e-05,
      "loss": 0.7472,
      "step": 2620
    },
    {
      "epoch": 0.21504497138184792,
      "grad_norm": 0.4368755519390106,
      "learning_rate": 2e-05,
      "loss": 0.7864,
      "step": 2630
    },
    {
      "epoch": 0.21586263286999183,
      "grad_norm": 0.425601065158844,
      "learning_rate": 2e-05,
      "loss": 0.8093,
      "step": 2640
    },
    {
      "epoch": 0.21668029435813574,
      "grad_norm": 0.43123167753219604,
      "learning_rate": 2e-05,
      "loss": 0.763,
      "step": 2650
    },
    {
      "epoch": 0.21749795584627965,
      "grad_norm": 0.41291922330856323,
      "learning_rate": 2e-05,
      "loss": 0.7669,
      "step": 2660
    },
    {
      "epoch": 0.21831561733442356,
      "grad_norm": 0.4302712380886078,
      "learning_rate": 2e-05,
      "loss": 0.7857,
      "step": 2670
    },
    {
      "epoch": 0.21913327882256745,
      "grad_norm": 0.41941574215888977,
      "learning_rate": 2e-05,
      "loss": 0.7444,
      "step": 2680
    },
    {
      "epoch": 0.21995094031071136,
      "grad_norm": 0.40381744503974915,
      "learning_rate": 2e-05,
      "loss": 0.7802,
      "step": 2690
    },
    {
      "epoch": 0.22076860179885527,
      "grad_norm": 0.42162802815437317,
      "learning_rate": 2e-05,
      "loss": 0.7807,
      "step": 2700
    },
    {
      "epoch": 0.22158626328699918,
      "grad_norm": 0.9290792346000671,
      "learning_rate": 2e-05,
      "loss": 0.8272,
      "step": 2710
    },
    {
      "epoch": 0.2224039247751431,
      "grad_norm": 0.3997645080089569,
      "learning_rate": 2e-05,
      "loss": 0.7506,
      "step": 2720
    },
    {
      "epoch": 0.223221586263287,
      "grad_norm": 0.4363479018211365,
      "learning_rate": 2e-05,
      "loss": 0.7878,
      "step": 2730
    },
    {
      "epoch": 0.22403924775143091,
      "grad_norm": 0.4270990490913391,
      "learning_rate": 2e-05,
      "loss": 0.774,
      "step": 2740
    },
    {
      "epoch": 0.22485690923957483,
      "grad_norm": 0.4099840521812439,
      "learning_rate": 2e-05,
      "loss": 0.7726,
      "step": 2750
    },
    {
      "epoch": 0.22567457072771874,
      "grad_norm": 0.4205203652381897,
      "learning_rate": 2e-05,
      "loss": 0.7432,
      "step": 2760
    },
    {
      "epoch": 0.22649223221586262,
      "grad_norm": 0.41790467500686646,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 2770
    },
    {
      "epoch": 0.22730989370400653,
      "grad_norm": 0.4241865277290344,
      "learning_rate": 2e-05,
      "loss": 0.7461,
      "step": 2780
    },
    {
      "epoch": 0.22812755519215044,
      "grad_norm": 0.3862425684928894,
      "learning_rate": 2e-05,
      "loss": 0.7811,
      "step": 2790
    },
    {
      "epoch": 0.22894521668029436,
      "grad_norm": 0.39432331919670105,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 2800
    },
    {
      "epoch": 0.22976287816843827,
      "grad_norm": 0.4487655460834503,
      "learning_rate": 2e-05,
      "loss": 0.7939,
      "step": 2810
    },
    {
      "epoch": 0.23058053965658218,
      "grad_norm": 0.41441550850868225,
      "learning_rate": 2e-05,
      "loss": 0.773,
      "step": 2820
    },
    {
      "epoch": 0.2313982011447261,
      "grad_norm": 0.41408640146255493,
      "learning_rate": 2e-05,
      "loss": 0.746,
      "step": 2830
    },
    {
      "epoch": 0.23221586263287,
      "grad_norm": 0.4297657310962677,
      "learning_rate": 2e-05,
      "loss": 0.7557,
      "step": 2840
    },
    {
      "epoch": 0.2330335241210139,
      "grad_norm": 0.4253140985965729,
      "learning_rate": 2e-05,
      "loss": 0.7718,
      "step": 2850
    },
    {
      "epoch": 0.2338511856091578,
      "grad_norm": 0.4623691141605377,
      "learning_rate": 2e-05,
      "loss": 0.7645,
      "step": 2860
    },
    {
      "epoch": 0.2346688470973017,
      "grad_norm": 0.425809383392334,
      "learning_rate": 2e-05,
      "loss": 0.7639,
      "step": 2870
    },
    {
      "epoch": 0.23548650858544562,
      "grad_norm": 0.4268561601638794,
      "learning_rate": 2e-05,
      "loss": 0.7501,
      "step": 2880
    },
    {
      "epoch": 0.23630417007358953,
      "grad_norm": 0.42480748891830444,
      "learning_rate": 2e-05,
      "loss": 0.737,
      "step": 2890
    },
    {
      "epoch": 0.23712183156173344,
      "grad_norm": 0.3940679430961609,
      "learning_rate": 2e-05,
      "loss": 0.7478,
      "step": 2900
    },
    {
      "epoch": 0.23793949304987735,
      "grad_norm": 0.3852456212043762,
      "learning_rate": 2e-05,
      "loss": 0.7685,
      "step": 2910
    },
    {
      "epoch": 0.23875715453802127,
      "grad_norm": 0.4131034016609192,
      "learning_rate": 2e-05,
      "loss": 0.7467,
      "step": 2920
    },
    {
      "epoch": 0.23957481602616518,
      "grad_norm": 0.44036728143692017,
      "learning_rate": 2e-05,
      "loss": 0.7303,
      "step": 2930
    },
    {
      "epoch": 0.2403924775143091,
      "grad_norm": 0.43323394656181335,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 2940
    },
    {
      "epoch": 0.24121013900245297,
      "grad_norm": 0.44328805804252625,
      "learning_rate": 2e-05,
      "loss": 0.7589,
      "step": 2950
    },
    {
      "epoch": 0.24202780049059688,
      "grad_norm": 0.41394710540771484,
      "learning_rate": 2e-05,
      "loss": 0.7534,
      "step": 2960
    },
    {
      "epoch": 0.2428454619787408,
      "grad_norm": 0.41553813219070435,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 2970
    },
    {
      "epoch": 0.2436631234668847,
      "grad_norm": 0.39981281757354736,
      "learning_rate": 2e-05,
      "loss": 0.7355,
      "step": 2980
    },
    {
      "epoch": 0.24448078495502862,
      "grad_norm": 0.42764589190483093,
      "learning_rate": 2e-05,
      "loss": 0.7612,
      "step": 2990
    },
    {
      "epoch": 0.24529844644317253,
      "grad_norm": 0.46958279609680176,
      "learning_rate": 2e-05,
      "loss": 0.7685,
      "step": 3000
    },
    {
      "epoch": 0.24611610793131644,
      "grad_norm": 0.43474486470222473,
      "learning_rate": 2e-05,
      "loss": 0.7476,
      "step": 3010
    },
    {
      "epoch": 0.24693376941946035,
      "grad_norm": 0.4077485501766205,
      "learning_rate": 2e-05,
      "loss": 0.7395,
      "step": 3020
    },
    {
      "epoch": 0.24775143090760426,
      "grad_norm": 0.4127010107040405,
      "learning_rate": 2e-05,
      "loss": 0.7449,
      "step": 3030
    },
    {
      "epoch": 0.24856909239574815,
      "grad_norm": 0.38957399129867554,
      "learning_rate": 2e-05,
      "loss": 0.75,
      "step": 3040
    },
    {
      "epoch": 0.24938675388389206,
      "grad_norm": 0.4030892848968506,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 3050
    },
    {
      "epoch": 0.25020441537203597,
      "grad_norm": 0.41952016949653625,
      "learning_rate": 2e-05,
      "loss": 0.7711,
      "step": 3060
    },
    {
      "epoch": 0.2510220768601799,
      "grad_norm": 0.4226265251636505,
      "learning_rate": 2e-05,
      "loss": 0.7614,
      "step": 3070
    },
    {
      "epoch": 0.2518397383483238,
      "grad_norm": 0.43005725741386414,
      "learning_rate": 2e-05,
      "loss": 0.7844,
      "step": 3080
    },
    {
      "epoch": 0.2526573998364677,
      "grad_norm": 0.4619760811328888,
      "learning_rate": 2e-05,
      "loss": 0.7607,
      "step": 3090
    },
    {
      "epoch": 0.2534750613246116,
      "grad_norm": 0.43441349267959595,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 3100
    },
    {
      "epoch": 0.2542927228127555,
      "grad_norm": 0.3946240246295929,
      "learning_rate": 2e-05,
      "loss": 0.7373,
      "step": 3110
    },
    {
      "epoch": 0.25511038430089944,
      "grad_norm": 0.42219096422195435,
      "learning_rate": 2e-05,
      "loss": 0.7143,
      "step": 3120
    },
    {
      "epoch": 0.2559280457890433,
      "grad_norm": 0.4020823836326599,
      "learning_rate": 2e-05,
      "loss": 0.7531,
      "step": 3130
    },
    {
      "epoch": 0.25674570727718726,
      "grad_norm": 0.4099595546722412,
      "learning_rate": 2e-05,
      "loss": 0.7517,
      "step": 3140
    },
    {
      "epoch": 0.25756336876533115,
      "grad_norm": 0.41835877299308777,
      "learning_rate": 2e-05,
      "loss": 0.7505,
      "step": 3150
    },
    {
      "epoch": 0.2583810302534751,
      "grad_norm": 0.39703795313835144,
      "learning_rate": 2e-05,
      "loss": 0.7823,
      "step": 3160
    },
    {
      "epoch": 0.25919869174161897,
      "grad_norm": 0.43623700737953186,
      "learning_rate": 2e-05,
      "loss": 0.7477,
      "step": 3170
    },
    {
      "epoch": 0.26001635322976285,
      "grad_norm": 0.41160234808921814,
      "learning_rate": 2e-05,
      "loss": 0.7454,
      "step": 3180
    },
    {
      "epoch": 0.2608340147179068,
      "grad_norm": 0.41497576236724854,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 3190
    },
    {
      "epoch": 0.2616516762060507,
      "grad_norm": 0.41480571031570435,
      "learning_rate": 2e-05,
      "loss": 0.775,
      "step": 3200
    },
    {
      "epoch": 0.2624693376941946,
      "grad_norm": 0.42725762724876404,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 3210
    },
    {
      "epoch": 0.2632869991823385,
      "grad_norm": 0.4155767560005188,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 3220
    },
    {
      "epoch": 0.26410466067048244,
      "grad_norm": 0.3951760232448578,
      "learning_rate": 2e-05,
      "loss": 0.7702,
      "step": 3230
    },
    {
      "epoch": 0.2649223221586263,
      "grad_norm": 0.40908899903297424,
      "learning_rate": 2e-05,
      "loss": 0.7464,
      "step": 3240
    },
    {
      "epoch": 0.26573998364677026,
      "grad_norm": 0.4152882397174835,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 3250
    },
    {
      "epoch": 0.26655764513491415,
      "grad_norm": 0.41987019777297974,
      "learning_rate": 2e-05,
      "loss": 0.7585,
      "step": 3260
    },
    {
      "epoch": 0.26737530662305803,
      "grad_norm": 0.3909783363342285,
      "learning_rate": 2e-05,
      "loss": 0.7568,
      "step": 3270
    },
    {
      "epoch": 0.26819296811120197,
      "grad_norm": 0.417426198720932,
      "learning_rate": 2e-05,
      "loss": 0.7924,
      "step": 3280
    },
    {
      "epoch": 0.26901062959934585,
      "grad_norm": 0.4050072729587555,
      "learning_rate": 2e-05,
      "loss": 0.7607,
      "step": 3290
    },
    {
      "epoch": 0.2698282910874898,
      "grad_norm": 0.4422713816165924,
      "learning_rate": 2e-05,
      "loss": 0.7751,
      "step": 3300
    },
    {
      "epoch": 0.2706459525756337,
      "grad_norm": 0.42104074358940125,
      "learning_rate": 2e-05,
      "loss": 0.7616,
      "step": 3310
    },
    {
      "epoch": 0.2714636140637776,
      "grad_norm": 0.40387916564941406,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 3320
    },
    {
      "epoch": 0.2722812755519215,
      "grad_norm": 0.40367233753204346,
      "learning_rate": 2e-05,
      "loss": 0.7799,
      "step": 3330
    },
    {
      "epoch": 0.27309893704006544,
      "grad_norm": 0.46548694372177124,
      "learning_rate": 2e-05,
      "loss": 0.7537,
      "step": 3340
    },
    {
      "epoch": 0.2739165985282093,
      "grad_norm": 0.451147198677063,
      "learning_rate": 2e-05,
      "loss": 0.7961,
      "step": 3350
    },
    {
      "epoch": 0.2747342600163532,
      "grad_norm": 0.44955241680145264,
      "learning_rate": 2e-05,
      "loss": 0.7614,
      "step": 3360
    },
    {
      "epoch": 0.27555192150449714,
      "grad_norm": 0.4327465295791626,
      "learning_rate": 2e-05,
      "loss": 0.7484,
      "step": 3370
    },
    {
      "epoch": 0.276369582992641,
      "grad_norm": 0.3961194157600403,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 3380
    },
    {
      "epoch": 0.27718724448078497,
      "grad_norm": 0.4116878807544708,
      "learning_rate": 2e-05,
      "loss": 0.7187,
      "step": 3390
    },
    {
      "epoch": 0.27800490596892885,
      "grad_norm": 0.40448230504989624,
      "learning_rate": 2e-05,
      "loss": 0.7718,
      "step": 3400
    },
    {
      "epoch": 0.2788225674570728,
      "grad_norm": 0.4147052764892578,
      "learning_rate": 2e-05,
      "loss": 0.7433,
      "step": 3410
    },
    {
      "epoch": 0.2796402289452167,
      "grad_norm": 0.4361591935157776,
      "learning_rate": 2e-05,
      "loss": 0.7556,
      "step": 3420
    },
    {
      "epoch": 0.2804578904333606,
      "grad_norm": 0.44823896884918213,
      "learning_rate": 2e-05,
      "loss": 0.7495,
      "step": 3430
    },
    {
      "epoch": 0.2812755519215045,
      "grad_norm": 0.3785291016101837,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 3440
    },
    {
      "epoch": 0.2820932134096484,
      "grad_norm": 0.38844063878059387,
      "learning_rate": 2e-05,
      "loss": 0.7416,
      "step": 3450
    },
    {
      "epoch": 0.2829108748977923,
      "grad_norm": 0.43286728858947754,
      "learning_rate": 2e-05,
      "loss": 0.7566,
      "step": 3460
    },
    {
      "epoch": 0.2837285363859362,
      "grad_norm": 0.4022294878959656,
      "learning_rate": 2e-05,
      "loss": 0.7529,
      "step": 3470
    },
    {
      "epoch": 0.28454619787408014,
      "grad_norm": 0.4239967167377472,
      "learning_rate": 2e-05,
      "loss": 0.7432,
      "step": 3480
    },
    {
      "epoch": 0.285363859362224,
      "grad_norm": 0.40576833486557007,
      "learning_rate": 2e-05,
      "loss": 0.7611,
      "step": 3490
    },
    {
      "epoch": 0.28618152085036797,
      "grad_norm": 0.4220907986164093,
      "learning_rate": 2e-05,
      "loss": 0.7545,
      "step": 3500
    },
    {
      "epoch": 0.28699918233851185,
      "grad_norm": 0.3993762731552124,
      "learning_rate": 2e-05,
      "loss": 0.7597,
      "step": 3510
    },
    {
      "epoch": 0.2878168438266558,
      "grad_norm": 0.4114720821380615,
      "learning_rate": 2e-05,
      "loss": 0.736,
      "step": 3520
    },
    {
      "epoch": 0.28863450531479967,
      "grad_norm": 0.43589070439338684,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 3530
    },
    {
      "epoch": 0.28945216680294356,
      "grad_norm": 0.4150513708591461,
      "learning_rate": 2e-05,
      "loss": 0.7698,
      "step": 3540
    },
    {
      "epoch": 0.2902698282910875,
      "grad_norm": 0.39021164178848267,
      "learning_rate": 2e-05,
      "loss": 0.7489,
      "step": 3550
    },
    {
      "epoch": 0.2910874897792314,
      "grad_norm": 0.5082252621650696,
      "learning_rate": 2e-05,
      "loss": 0.7374,
      "step": 3560
    },
    {
      "epoch": 0.2919051512673753,
      "grad_norm": 0.43987151980400085,
      "learning_rate": 2e-05,
      "loss": 0.7452,
      "step": 3570
    },
    {
      "epoch": 0.2927228127555192,
      "grad_norm": 0.4348721504211426,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 3580
    },
    {
      "epoch": 0.29354047424366314,
      "grad_norm": 0.3936716616153717,
      "learning_rate": 2e-05,
      "loss": 0.7405,
      "step": 3590
    },
    {
      "epoch": 0.294358135731807,
      "grad_norm": 0.38866978883743286,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 3600
    },
    {
      "epoch": 0.29517579721995096,
      "grad_norm": 0.4239301085472107,
      "learning_rate": 2e-05,
      "loss": 0.7557,
      "step": 3610
    },
    {
      "epoch": 0.29599345870809485,
      "grad_norm": 0.4139637351036072,
      "learning_rate": 2e-05,
      "loss": 0.7682,
      "step": 3620
    },
    {
      "epoch": 0.29681112019623873,
      "grad_norm": 0.4086167812347412,
      "learning_rate": 2e-05,
      "loss": 0.7413,
      "step": 3630
    },
    {
      "epoch": 0.29762878168438267,
      "grad_norm": 0.3776358962059021,
      "learning_rate": 2e-05,
      "loss": 0.7483,
      "step": 3640
    },
    {
      "epoch": 0.29844644317252655,
      "grad_norm": 0.39760732650756836,
      "learning_rate": 2e-05,
      "loss": 0.746,
      "step": 3650
    },
    {
      "epoch": 0.2992641046606705,
      "grad_norm": 0.37500932812690735,
      "learning_rate": 2e-05,
      "loss": 0.7874,
      "step": 3660
    },
    {
      "epoch": 0.3000817661488144,
      "grad_norm": 0.43193116784095764,
      "learning_rate": 2e-05,
      "loss": 0.7651,
      "step": 3670
    },
    {
      "epoch": 0.3008994276369583,
      "grad_norm": 0.41836270689964294,
      "learning_rate": 2e-05,
      "loss": 0.7507,
      "step": 3680
    },
    {
      "epoch": 0.3017170891251022,
      "grad_norm": 0.3975755572319031,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 3690
    },
    {
      "epoch": 0.30253475061324614,
      "grad_norm": 0.4004629850387573,
      "learning_rate": 2e-05,
      "loss": 0.7406,
      "step": 3700
    },
    {
      "epoch": 0.30335241210139,
      "grad_norm": 0.41910433769226074,
      "learning_rate": 2e-05,
      "loss": 0.8037,
      "step": 3710
    },
    {
      "epoch": 0.3041700735895339,
      "grad_norm": 0.4036235511302948,
      "learning_rate": 2e-05,
      "loss": 0.7465,
      "step": 3720
    },
    {
      "epoch": 0.30498773507767785,
      "grad_norm": 0.41243213415145874,
      "learning_rate": 2e-05,
      "loss": 0.7469,
      "step": 3730
    },
    {
      "epoch": 0.30580539656582173,
      "grad_norm": 0.45436981320381165,
      "learning_rate": 2e-05,
      "loss": 0.7575,
      "step": 3740
    },
    {
      "epoch": 0.30662305805396567,
      "grad_norm": 0.3992025554180145,
      "learning_rate": 2e-05,
      "loss": 0.7409,
      "step": 3750
    },
    {
      "epoch": 0.30744071954210955,
      "grad_norm": 0.3761412501335144,
      "learning_rate": 2e-05,
      "loss": 0.7385,
      "step": 3760
    },
    {
      "epoch": 0.3082583810302535,
      "grad_norm": 0.450571745634079,
      "learning_rate": 2e-05,
      "loss": 0.7623,
      "step": 3770
    },
    {
      "epoch": 0.3090760425183974,
      "grad_norm": 0.43576136231422424,
      "learning_rate": 2e-05,
      "loss": 0.7465,
      "step": 3780
    },
    {
      "epoch": 0.3098937040065413,
      "grad_norm": 0.38754603266716003,
      "learning_rate": 2e-05,
      "loss": 0.7577,
      "step": 3790
    },
    {
      "epoch": 0.3107113654946852,
      "grad_norm": 0.41827720403671265,
      "learning_rate": 2e-05,
      "loss": 0.7554,
      "step": 3800
    },
    {
      "epoch": 0.3115290269828291,
      "grad_norm": 0.4002350866794586,
      "learning_rate": 2e-05,
      "loss": 0.7514,
      "step": 3810
    },
    {
      "epoch": 0.312346688470973,
      "grad_norm": 0.3866146504878998,
      "learning_rate": 2e-05,
      "loss": 0.7646,
      "step": 3820
    },
    {
      "epoch": 0.3131643499591169,
      "grad_norm": 0.4048006236553192,
      "learning_rate": 2e-05,
      "loss": 0.7727,
      "step": 3830
    },
    {
      "epoch": 0.31398201144726084,
      "grad_norm": 0.4159805476665497,
      "learning_rate": 2e-05,
      "loss": 0.7536,
      "step": 3840
    },
    {
      "epoch": 0.31479967293540473,
      "grad_norm": 0.41745713353157043,
      "learning_rate": 2e-05,
      "loss": 0.7335,
      "step": 3850
    },
    {
      "epoch": 0.31561733442354867,
      "grad_norm": 0.37902170419692993,
      "learning_rate": 2e-05,
      "loss": 0.7598,
      "step": 3860
    },
    {
      "epoch": 0.31643499591169255,
      "grad_norm": 0.40632471442222595,
      "learning_rate": 2e-05,
      "loss": 0.7628,
      "step": 3870
    },
    {
      "epoch": 0.3172526573998365,
      "grad_norm": 0.42005592584609985,
      "learning_rate": 2e-05,
      "loss": 0.753,
      "step": 3880
    },
    {
      "epoch": 0.3180703188879804,
      "grad_norm": 0.4179593324661255,
      "learning_rate": 2e-05,
      "loss": 0.7511,
      "step": 3890
    },
    {
      "epoch": 0.31888798037612426,
      "grad_norm": 0.4019879996776581,
      "learning_rate": 2e-05,
      "loss": 0.7428,
      "step": 3900
    },
    {
      "epoch": 0.3197056418642682,
      "grad_norm": 0.3838173747062683,
      "learning_rate": 2e-05,
      "loss": 0.7218,
      "step": 3910
    },
    {
      "epoch": 0.3205233033524121,
      "grad_norm": 0.3869258761405945,
      "learning_rate": 2e-05,
      "loss": 0.739,
      "step": 3920
    },
    {
      "epoch": 0.321340964840556,
      "grad_norm": 0.3965376317501068,
      "learning_rate": 2e-05,
      "loss": 0.7557,
      "step": 3930
    },
    {
      "epoch": 0.3221586263286999,
      "grad_norm": 0.3937940001487732,
      "learning_rate": 2e-05,
      "loss": 0.7434,
      "step": 3940
    },
    {
      "epoch": 0.32297628781684384,
      "grad_norm": 0.4084051847457886,
      "learning_rate": 2e-05,
      "loss": 0.7375,
      "step": 3950
    },
    {
      "epoch": 0.3237939493049877,
      "grad_norm": 0.38900405168533325,
      "learning_rate": 2e-05,
      "loss": 0.7534,
      "step": 3960
    },
    {
      "epoch": 0.32461161079313167,
      "grad_norm": 0.397067129611969,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 3970
    },
    {
      "epoch": 0.32542927228127555,
      "grad_norm": 0.4044935405254364,
      "learning_rate": 2e-05,
      "loss": 0.7681,
      "step": 3980
    },
    {
      "epoch": 0.32624693376941943,
      "grad_norm": 0.4075620770454407,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 3990
    },
    {
      "epoch": 0.3270645952575634,
      "grad_norm": 0.37622272968292236,
      "learning_rate": 2e-05,
      "loss": 0.7515,
      "step": 4000
    },
    {
      "epoch": 0.32788225674570726,
      "grad_norm": 0.39748191833496094,
      "learning_rate": 2e-05,
      "loss": 0.7801,
      "step": 4010
    },
    {
      "epoch": 0.3286999182338512,
      "grad_norm": 0.39921826124191284,
      "learning_rate": 2e-05,
      "loss": 0.7668,
      "step": 4020
    },
    {
      "epoch": 0.3295175797219951,
      "grad_norm": 0.3993796706199646,
      "learning_rate": 2e-05,
      "loss": 0.7588,
      "step": 4030
    },
    {
      "epoch": 0.330335241210139,
      "grad_norm": 0.40211594104766846,
      "learning_rate": 2e-05,
      "loss": 0.7293,
      "step": 4040
    },
    {
      "epoch": 0.3311529026982829,
      "grad_norm": 0.3940093517303467,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 4050
    },
    {
      "epoch": 0.33197056418642684,
      "grad_norm": 0.41875386238098145,
      "learning_rate": 2e-05,
      "loss": 0.7188,
      "step": 4060
    },
    {
      "epoch": 0.3327882256745707,
      "grad_norm": 0.4149841368198395,
      "learning_rate": 2e-05,
      "loss": 0.7494,
      "step": 4070
    },
    {
      "epoch": 0.3336058871627146,
      "grad_norm": 0.40230244398117065,
      "learning_rate": 2e-05,
      "loss": 0.7384,
      "step": 4080
    },
    {
      "epoch": 0.33442354865085855,
      "grad_norm": 0.5382981896400452,
      "learning_rate": 2e-05,
      "loss": 0.7529,
      "step": 4090
    },
    {
      "epoch": 0.33524121013900243,
      "grad_norm": 0.3823828101158142,
      "learning_rate": 2e-05,
      "loss": 0.7814,
      "step": 4100
    },
    {
      "epoch": 0.33605887162714637,
      "grad_norm": 0.4252546429634094,
      "learning_rate": 2e-05,
      "loss": 0.7531,
      "step": 4110
    },
    {
      "epoch": 0.33687653311529026,
      "grad_norm": 0.40745359659194946,
      "learning_rate": 2e-05,
      "loss": 0.7765,
      "step": 4120
    },
    {
      "epoch": 0.3376941946034342,
      "grad_norm": 0.3941667675971985,
      "learning_rate": 2e-05,
      "loss": 0.7566,
      "step": 4130
    },
    {
      "epoch": 0.3385118560915781,
      "grad_norm": 0.40058091282844543,
      "learning_rate": 2e-05,
      "loss": 0.7448,
      "step": 4140
    },
    {
      "epoch": 0.339329517579722,
      "grad_norm": 0.41285571455955505,
      "learning_rate": 2e-05,
      "loss": 0.7231,
      "step": 4150
    },
    {
      "epoch": 0.3401471790678659,
      "grad_norm": 0.408279150724411,
      "learning_rate": 2e-05,
      "loss": 0.7582,
      "step": 4160
    },
    {
      "epoch": 0.3409648405560098,
      "grad_norm": 0.3870287835597992,
      "learning_rate": 2e-05,
      "loss": 0.7542,
      "step": 4170
    },
    {
      "epoch": 0.3417825020441537,
      "grad_norm": 0.4457457363605499,
      "learning_rate": 2e-05,
      "loss": 0.7583,
      "step": 4180
    },
    {
      "epoch": 0.3426001635322976,
      "grad_norm": 0.38335496187210083,
      "learning_rate": 2e-05,
      "loss": 0.7482,
      "step": 4190
    },
    {
      "epoch": 0.34341782502044155,
      "grad_norm": 0.432076632976532,
      "learning_rate": 2e-05,
      "loss": 0.7669,
      "step": 4200
    },
    {
      "epoch": 0.34423548650858543,
      "grad_norm": 0.4170612096786499,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 4210
    },
    {
      "epoch": 0.34505314799672937,
      "grad_norm": 0.39313265681266785,
      "learning_rate": 2e-05,
      "loss": 0.7247,
      "step": 4220
    },
    {
      "epoch": 0.34587080948487325,
      "grad_norm": 0.40817657113075256,
      "learning_rate": 2e-05,
      "loss": 0.7545,
      "step": 4230
    },
    {
      "epoch": 0.3466884709730172,
      "grad_norm": 0.392434298992157,
      "learning_rate": 2e-05,
      "loss": 0.7402,
      "step": 4240
    },
    {
      "epoch": 0.3475061324611611,
      "grad_norm": 0.42975276708602905,
      "learning_rate": 2e-05,
      "loss": 0.7636,
      "step": 4250
    },
    {
      "epoch": 0.34832379394930496,
      "grad_norm": 0.3929069936275482,
      "learning_rate": 2e-05,
      "loss": 0.732,
      "step": 4260
    },
    {
      "epoch": 0.3491414554374489,
      "grad_norm": 0.389665424823761,
      "learning_rate": 2e-05,
      "loss": 0.7468,
      "step": 4270
    },
    {
      "epoch": 0.3499591169255928,
      "grad_norm": 0.38716766238212585,
      "learning_rate": 2e-05,
      "loss": 0.7405,
      "step": 4280
    },
    {
      "epoch": 0.3507767784137367,
      "grad_norm": 0.3947966992855072,
      "learning_rate": 2e-05,
      "loss": 0.7251,
      "step": 4290
    },
    {
      "epoch": 0.3515944399018806,
      "grad_norm": 0.37235280871391296,
      "learning_rate": 2e-05,
      "loss": 0.729,
      "step": 4300
    },
    {
      "epoch": 0.35241210139002455,
      "grad_norm": 0.4839659035205841,
      "learning_rate": 2e-05,
      "loss": 0.7582,
      "step": 4310
    },
    {
      "epoch": 0.35322976287816843,
      "grad_norm": 0.41242969036102295,
      "learning_rate": 2e-05,
      "loss": 0.7599,
      "step": 4320
    },
    {
      "epoch": 0.35404742436631237,
      "grad_norm": 0.42157986760139465,
      "learning_rate": 2e-05,
      "loss": 0.7902,
      "step": 4330
    },
    {
      "epoch": 0.35486508585445625,
      "grad_norm": 0.39237549901008606,
      "learning_rate": 2e-05,
      "loss": 0.7306,
      "step": 4340
    },
    {
      "epoch": 0.35568274734260014,
      "grad_norm": 0.4073483943939209,
      "learning_rate": 2e-05,
      "loss": 0.7423,
      "step": 4350
    },
    {
      "epoch": 0.3565004088307441,
      "grad_norm": 0.37807080149650574,
      "learning_rate": 2e-05,
      "loss": 0.7605,
      "step": 4360
    },
    {
      "epoch": 0.35731807031888796,
      "grad_norm": 0.4259825348854065,
      "learning_rate": 2e-05,
      "loss": 0.7368,
      "step": 4370
    },
    {
      "epoch": 0.3581357318070319,
      "grad_norm": 0.39345216751098633,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 4380
    },
    {
      "epoch": 0.3589533932951758,
      "grad_norm": 0.4019021689891815,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 4390
    },
    {
      "epoch": 0.3597710547833197,
      "grad_norm": 0.3999948799610138,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 4400
    },
    {
      "epoch": 0.3605887162714636,
      "grad_norm": 0.4193735420703888,
      "learning_rate": 2e-05,
      "loss": 0.7567,
      "step": 4410
    },
    {
      "epoch": 0.36140637775960754,
      "grad_norm": 0.40287792682647705,
      "learning_rate": 2e-05,
      "loss": 0.7506,
      "step": 4420
    },
    {
      "epoch": 0.36222403924775143,
      "grad_norm": 0.4220403730869293,
      "learning_rate": 2e-05,
      "loss": 0.7585,
      "step": 4430
    },
    {
      "epoch": 0.3630417007358953,
      "grad_norm": 0.3977868854999542,
      "learning_rate": 2e-05,
      "loss": 0.7393,
      "step": 4440
    },
    {
      "epoch": 0.36385936222403925,
      "grad_norm": 0.5432819128036499,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 4450
    },
    {
      "epoch": 0.36467702371218313,
      "grad_norm": 0.4258385896682739,
      "learning_rate": 2e-05,
      "loss": 0.7682,
      "step": 4460
    },
    {
      "epoch": 0.3654946852003271,
      "grad_norm": 0.3945724666118622,
      "learning_rate": 2e-05,
      "loss": 0.75,
      "step": 4470
    },
    {
      "epoch": 0.36631234668847096,
      "grad_norm": 0.41011112928390503,
      "learning_rate": 2e-05,
      "loss": 0.7485,
      "step": 4480
    },
    {
      "epoch": 0.3671300081766149,
      "grad_norm": 0.3658716082572937,
      "learning_rate": 2e-05,
      "loss": 0.719,
      "step": 4490
    },
    {
      "epoch": 0.3679476696647588,
      "grad_norm": 0.38978058099746704,
      "learning_rate": 2e-05,
      "loss": 0.7624,
      "step": 4500
    },
    {
      "epoch": 0.3687653311529027,
      "grad_norm": 0.41422396898269653,
      "learning_rate": 2e-05,
      "loss": 0.7292,
      "step": 4510
    },
    {
      "epoch": 0.3695829926410466,
      "grad_norm": 0.6165766716003418,
      "learning_rate": 2e-05,
      "loss": 0.7435,
      "step": 4520
    },
    {
      "epoch": 0.3704006541291905,
      "grad_norm": 0.39596688747406006,
      "learning_rate": 2e-05,
      "loss": 0.7539,
      "step": 4530
    },
    {
      "epoch": 0.3712183156173344,
      "grad_norm": 0.4293264150619507,
      "learning_rate": 2e-05,
      "loss": 0.7695,
      "step": 4540
    },
    {
      "epoch": 0.3720359771054783,
      "grad_norm": 0.5381836295127869,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 4550
    },
    {
      "epoch": 0.37285363859362225,
      "grad_norm": 0.42194926738739014,
      "learning_rate": 2e-05,
      "loss": 0.7436,
      "step": 4560
    },
    {
      "epoch": 0.37367130008176613,
      "grad_norm": 0.3940383791923523,
      "learning_rate": 2e-05,
      "loss": 0.7405,
      "step": 4570
    },
    {
      "epoch": 0.3744889615699101,
      "grad_norm": 0.3936174511909485,
      "learning_rate": 2e-05,
      "loss": 0.7148,
      "step": 4580
    },
    {
      "epoch": 0.37530662305805396,
      "grad_norm": 0.4172099232673645,
      "learning_rate": 2e-05,
      "loss": 0.7578,
      "step": 4590
    },
    {
      "epoch": 0.3761242845461979,
      "grad_norm": 0.3652750849723816,
      "learning_rate": 2e-05,
      "loss": 0.7628,
      "step": 4600
    },
    {
      "epoch": 0.3769419460343418,
      "grad_norm": 0.40293166041374207,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 4610
    },
    {
      "epoch": 0.3777596075224857,
      "grad_norm": 0.4645305871963501,
      "learning_rate": 2e-05,
      "loss": 0.7235,
      "step": 4620
    },
    {
      "epoch": 0.3785772690106296,
      "grad_norm": 0.47156858444213867,
      "learning_rate": 2e-05,
      "loss": 0.7149,
      "step": 4630
    },
    {
      "epoch": 0.3793949304987735,
      "grad_norm": 0.3718200922012329,
      "learning_rate": 2e-05,
      "loss": 0.7276,
      "step": 4640
    },
    {
      "epoch": 0.3802125919869174,
      "grad_norm": 0.38712579011917114,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 4650
    },
    {
      "epoch": 0.3810302534750613,
      "grad_norm": 0.38667309284210205,
      "learning_rate": 2e-05,
      "loss": 0.7565,
      "step": 4660
    },
    {
      "epoch": 0.38184791496320525,
      "grad_norm": 0.3759092092514038,
      "learning_rate": 2e-05,
      "loss": 0.764,
      "step": 4670
    },
    {
      "epoch": 0.38266557645134913,
      "grad_norm": 0.39515963196754456,
      "learning_rate": 2e-05,
      "loss": 0.7443,
      "step": 4680
    },
    {
      "epoch": 0.38348323793949307,
      "grad_norm": 0.41865164041519165,
      "learning_rate": 2e-05,
      "loss": 0.7549,
      "step": 4690
    },
    {
      "epoch": 0.38430089942763695,
      "grad_norm": 0.4113163352012634,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 4700
    },
    {
      "epoch": 0.3851185609157809,
      "grad_norm": 0.39225131273269653,
      "learning_rate": 2e-05,
      "loss": 0.7493,
      "step": 4710
    },
    {
      "epoch": 0.3859362224039248,
      "grad_norm": 0.39372774958610535,
      "learning_rate": 2e-05,
      "loss": 0.7385,
      "step": 4720
    },
    {
      "epoch": 0.38675388389206866,
      "grad_norm": 0.3898671567440033,
      "learning_rate": 2e-05,
      "loss": 0.7735,
      "step": 4730
    },
    {
      "epoch": 0.3875715453802126,
      "grad_norm": 0.415476530790329,
      "learning_rate": 2e-05,
      "loss": 0.7246,
      "step": 4740
    },
    {
      "epoch": 0.3883892068683565,
      "grad_norm": 0.39973869919776917,
      "learning_rate": 2e-05,
      "loss": 0.7139,
      "step": 4750
    },
    {
      "epoch": 0.3892068683565004,
      "grad_norm": 0.3770519495010376,
      "learning_rate": 2e-05,
      "loss": 0.7483,
      "step": 4760
    },
    {
      "epoch": 0.3900245298446443,
      "grad_norm": 0.4224013388156891,
      "learning_rate": 2e-05,
      "loss": 0.7415,
      "step": 4770
    },
    {
      "epoch": 0.39084219133278825,
      "grad_norm": 0.43009740114212036,
      "learning_rate": 2e-05,
      "loss": 0.7763,
      "step": 4780
    },
    {
      "epoch": 0.39165985282093213,
      "grad_norm": 0.3691262900829315,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 4790
    },
    {
      "epoch": 0.39247751430907607,
      "grad_norm": 0.42145824432373047,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 4800
    },
    {
      "epoch": 0.39329517579721995,
      "grad_norm": 0.40462708473205566,
      "learning_rate": 2e-05,
      "loss": 0.7571,
      "step": 4810
    },
    {
      "epoch": 0.39411283728536384,
      "grad_norm": 0.3951694369316101,
      "learning_rate": 2e-05,
      "loss": 0.7414,
      "step": 4820
    },
    {
      "epoch": 0.3949304987735078,
      "grad_norm": 0.3983229696750641,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 4830
    },
    {
      "epoch": 0.39574816026165166,
      "grad_norm": 0.4490920901298523,
      "learning_rate": 2e-05,
      "loss": 0.7582,
      "step": 4840
    },
    {
      "epoch": 0.3965658217497956,
      "grad_norm": 0.39146313071250916,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 4850
    },
    {
      "epoch": 0.3973834832379395,
      "grad_norm": 0.4129073917865753,
      "learning_rate": 2e-05,
      "loss": 0.7246,
      "step": 4860
    },
    {
      "epoch": 0.3982011447260834,
      "grad_norm": 0.41237565875053406,
      "learning_rate": 2e-05,
      "loss": 0.7214,
      "step": 4870
    },
    {
      "epoch": 0.3990188062142273,
      "grad_norm": 0.4975462257862091,
      "learning_rate": 2e-05,
      "loss": 0.7427,
      "step": 4880
    },
    {
      "epoch": 0.39983646770237125,
      "grad_norm": 0.444686621427536,
      "learning_rate": 2e-05,
      "loss": 0.7753,
      "step": 4890
    },
    {
      "epoch": 0.40065412919051513,
      "grad_norm": 0.40740424394607544,
      "learning_rate": 2e-05,
      "loss": 0.7414,
      "step": 4900
    },
    {
      "epoch": 0.401471790678659,
      "grad_norm": 0.4233196973800659,
      "learning_rate": 2e-05,
      "loss": 0.7496,
      "step": 4910
    },
    {
      "epoch": 0.40228945216680295,
      "grad_norm": 0.39530307054519653,
      "learning_rate": 2e-05,
      "loss": 0.769,
      "step": 4920
    },
    {
      "epoch": 0.40310711365494684,
      "grad_norm": 0.37782007455825806,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 4930
    },
    {
      "epoch": 0.4039247751430908,
      "grad_norm": 0.3919887840747833,
      "learning_rate": 2e-05,
      "loss": 0.7376,
      "step": 4940
    },
    {
      "epoch": 0.40474243663123466,
      "grad_norm": 0.4067898690700531,
      "learning_rate": 2e-05,
      "loss": 0.7607,
      "step": 4950
    },
    {
      "epoch": 0.4055600981193786,
      "grad_norm": 0.39617830514907837,
      "learning_rate": 2e-05,
      "loss": 0.7676,
      "step": 4960
    },
    {
      "epoch": 0.4063777596075225,
      "grad_norm": 0.427827388048172,
      "learning_rate": 2e-05,
      "loss": 0.7547,
      "step": 4970
    },
    {
      "epoch": 0.4071954210956664,
      "grad_norm": 0.39542731642723083,
      "learning_rate": 2e-05,
      "loss": 0.7444,
      "step": 4980
    },
    {
      "epoch": 0.4080130825838103,
      "grad_norm": 0.386767715215683,
      "learning_rate": 2e-05,
      "loss": 0.7404,
      "step": 4990
    },
    {
      "epoch": 0.4088307440719542,
      "grad_norm": 0.38438013195991516,
      "learning_rate": 2e-05,
      "loss": 0.7154,
      "step": 5000
    },
    {
      "epoch": 0.40964840556009813,
      "grad_norm": 0.41563791036605835,
      "learning_rate": 2e-05,
      "loss": 0.7221,
      "step": 5010
    },
    {
      "epoch": 0.410466067048242,
      "grad_norm": 0.3918059468269348,
      "learning_rate": 2e-05,
      "loss": 0.7136,
      "step": 5020
    },
    {
      "epoch": 0.41128372853638595,
      "grad_norm": 0.37461262941360474,
      "learning_rate": 2e-05,
      "loss": 0.7057,
      "step": 5030
    },
    {
      "epoch": 0.41210139002452983,
      "grad_norm": 0.4226076006889343,
      "learning_rate": 2e-05,
      "loss": 0.7531,
      "step": 5040
    },
    {
      "epoch": 0.4129190515126738,
      "grad_norm": 0.3859594464302063,
      "learning_rate": 2e-05,
      "loss": 0.746,
      "step": 5050
    },
    {
      "epoch": 0.41373671300081766,
      "grad_norm": 0.4011143445968628,
      "learning_rate": 2e-05,
      "loss": 0.7418,
      "step": 5060
    },
    {
      "epoch": 0.4145543744889616,
      "grad_norm": 0.3988434076309204,
      "learning_rate": 2e-05,
      "loss": 0.7307,
      "step": 5070
    },
    {
      "epoch": 0.4153720359771055,
      "grad_norm": 0.4151720106601715,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 5080
    },
    {
      "epoch": 0.41618969746524936,
      "grad_norm": 0.3734935224056244,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.7821,
      "step": 5090
    },
    {
      "epoch": 0.4170073589533933,
      "grad_norm": 0.4272422194480896,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.7402,
      "step": 5100
    },
    {
      "epoch": 0.4178250204415372,
      "grad_norm": 0.385078102350235,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.7154,
      "step": 5110
    },
    {
      "epoch": 0.4186426819296811,
      "grad_norm": 0.42083486914634705,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.7324,
      "step": 5120
    },
    {
      "epoch": 0.419460343417825,
      "grad_norm": 0.40142926573753357,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.7481,
      "step": 5130
    },
    {
      "epoch": 0.42027800490596895,
      "grad_norm": 0.5385702252388,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.7286,
      "step": 5140
    },
    {
      "epoch": 0.42109566639411283,
      "grad_norm": 0.3724869191646576,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.7117,
      "step": 5150
    },
    {
      "epoch": 0.4219133278822568,
      "grad_norm": 0.3998297154903412,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.7397,
      "step": 5160
    },
    {
      "epoch": 0.42273098937040066,
      "grad_norm": 0.3887345790863037,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.7385,
      "step": 5170
    },
    {
      "epoch": 0.42354865085854454,
      "grad_norm": 0.43496760725975037,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.7748,
      "step": 5180
    },
    {
      "epoch": 0.4243663123466885,
      "grad_norm": 0.42485687136650085,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.7697,
      "step": 5190
    },
    {
      "epoch": 0.42518397383483236,
      "grad_norm": 0.37596938014030457,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.7214,
      "step": 5200
    },
    {
      "epoch": 0.4260016353229763,
      "grad_norm": 0.3709232807159424,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.7723,
      "step": 5210
    },
    {
      "epoch": 0.4268192968111202,
      "grad_norm": 0.386949747800827,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.7623,
      "step": 5220
    },
    {
      "epoch": 0.4276369582992641,
      "grad_norm": 0.3908058702945709,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.7262,
      "step": 5230
    },
    {
      "epoch": 0.428454619787408,
      "grad_norm": 0.3770569860935211,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.724,
      "step": 5240
    },
    {
      "epoch": 0.42927228127555195,
      "grad_norm": 0.3810878396034241,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.7329,
      "step": 5250
    },
    {
      "epoch": 0.43008994276369583,
      "grad_norm": 0.390196830034256,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.7556,
      "step": 5260
    },
    {
      "epoch": 0.4309076042518397,
      "grad_norm": 0.3797683119773865,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.7174,
      "step": 5270
    },
    {
      "epoch": 0.43172526573998365,
      "grad_norm": 0.3873342275619507,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.7292,
      "step": 5280
    },
    {
      "epoch": 0.43254292722812754,
      "grad_norm": 0.3752491772174835,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.6927,
      "step": 5290
    },
    {
      "epoch": 0.4333605887162715,
      "grad_norm": 0.38635361194610596,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.7271,
      "step": 5300
    },
    {
      "epoch": 0.43417825020441536,
      "grad_norm": 0.39057794213294983,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.7638,
      "step": 5310
    },
    {
      "epoch": 0.4349959116925593,
      "grad_norm": 0.3876177966594696,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.7271,
      "step": 5320
    },
    {
      "epoch": 0.4358135731807032,
      "grad_norm": 0.37395739555358887,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.7146,
      "step": 5330
    },
    {
      "epoch": 0.4366312346688471,
      "grad_norm": 0.37287428975105286,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.7254,
      "step": 5340
    },
    {
      "epoch": 0.437448896156991,
      "grad_norm": 0.40743494033813477,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.7358,
      "step": 5350
    },
    {
      "epoch": 0.4382665576451349,
      "grad_norm": 0.37617412209510803,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.6951,
      "step": 5360
    },
    {
      "epoch": 0.43908421913327883,
      "grad_norm": 0.3787813186645508,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.7084,
      "step": 5370
    },
    {
      "epoch": 0.4399018806214227,
      "grad_norm": 0.47615402936935425,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.7507,
      "step": 5380
    },
    {
      "epoch": 0.44071954210956665,
      "grad_norm": 0.4045373201370239,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.7524,
      "step": 5390
    },
    {
      "epoch": 0.44153720359771054,
      "grad_norm": 0.40211862325668335,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.7389,
      "step": 5400
    },
    {
      "epoch": 0.4423548650858545,
      "grad_norm": 0.38277652859687805,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.7087,
      "step": 5410
    },
    {
      "epoch": 0.44317252657399836,
      "grad_norm": 0.38409224152565,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.7008,
      "step": 5420
    },
    {
      "epoch": 0.4439901880621423,
      "grad_norm": 0.3716842830181122,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.7484,
      "step": 5430
    },
    {
      "epoch": 0.4448078495502862,
      "grad_norm": 0.3625447452068329,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.713,
      "step": 5440
    },
    {
      "epoch": 0.44562551103843007,
      "grad_norm": 0.4007038176059723,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.6926,
      "step": 5450
    },
    {
      "epoch": 0.446443172526574,
      "grad_norm": 0.3814511001110077,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.7195,
      "step": 5460
    },
    {
      "epoch": 0.4472608340147179,
      "grad_norm": 0.3985978662967682,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.7074,
      "step": 5470
    },
    {
      "epoch": 0.44807849550286183,
      "grad_norm": 0.41598832607269287,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.7298,
      "step": 5480
    },
    {
      "epoch": 0.4488961569910057,
      "grad_norm": 0.38021478056907654,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.7068,
      "step": 5490
    },
    {
      "epoch": 0.44971381847914965,
      "grad_norm": 0.3673088550567627,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.7124,
      "step": 5500
    },
    {
      "epoch": 0.45053147996729354,
      "grad_norm": 0.39006367325782776,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.7023,
      "step": 5510
    },
    {
      "epoch": 0.4513491414554375,
      "grad_norm": 0.35301074385643005,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.6792,
      "step": 5520
    },
    {
      "epoch": 0.45216680294358136,
      "grad_norm": 0.42610907554626465,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.7489,
      "step": 5530
    },
    {
      "epoch": 0.45298446443172524,
      "grad_norm": 0.38084253668785095,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.7101,
      "step": 5540
    },
    {
      "epoch": 0.4538021259198692,
      "grad_norm": 0.3803776502609253,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.689,
      "step": 5550
    },
    {
      "epoch": 0.45461978740801307,
      "grad_norm": 0.39923131465911865,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.6942,
      "step": 5560
    },
    {
      "epoch": 0.455437448896157,
      "grad_norm": 0.37746211886405945,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.7202,
      "step": 5570
    },
    {
      "epoch": 0.4562551103843009,
      "grad_norm": 0.3752861022949219,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.7011,
      "step": 5580
    },
    {
      "epoch": 0.4570727718724448,
      "grad_norm": 0.35417813062667847,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.7378,
      "step": 5590
    },
    {
      "epoch": 0.45764513491414555,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.02447158063077833,
      "train_runtime": 1764.166,
      "train_samples_per_second": 38.071,
      "train_steps_per_second": 3.173
    }
  ],
  "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
}