{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1949,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.06121898619475577,
      "learning_rate": 1.0256410256410257e-06,
      "loss": 0.9032,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.05076189465509454,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.7741,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.056309939667137404,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 0.807,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.08652735131470117,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.8461,
      "step": 15
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.07961187993196545,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.7646,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.09626619309485744,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.7847,
      "step": 25
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.07333598901062204,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.804,
      "step": 30
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.07656413646264956,
      "learning_rate": 3.58974358974359e-05,
      "loss": 0.762,
      "step": 35
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.10274445509894202,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 0.7525,
      "step": 40
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.06931022616141319,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.7161,
      "step": 45
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.06249108360478095,
      "learning_rate": 5.128205128205128e-05,
      "loss": 0.7223,
      "step": 50
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.06488575342654651,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 0.7326,
      "step": 55
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.07291924815316946,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.8038,
      "step": 60
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.060040360077683416,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.7521,
      "step": 65
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.08523783674260317,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.709,
      "step": 70
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.06703200831653133,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.7942,
      "step": 75
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.05463611621688078,
      "learning_rate": 8.205128205128205e-05,
      "loss": 0.6818,
      "step": 80
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.07432808646419857,
      "learning_rate": 8.717948717948718e-05,
      "loss": 0.7341,
      "step": 85
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.06238952277649466,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.7342,
      "step": 90
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.07425456689389622,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.7742,
      "step": 95
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.0605404650874565,
      "learning_rate": 0.00010256410256410256,
      "loss": 0.7809,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.0694984420567863,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.7539,
      "step": 105
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.06656828962914285,
      "learning_rate": 0.00011282051282051283,
      "loss": 0.7593,
      "step": 110
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.05957346631757917,
      "learning_rate": 0.00011794871794871796,
      "loss": 0.7155,
      "step": 115
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.05662071050147848,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.7591,
      "step": 120
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.06578556046938931,
      "learning_rate": 0.00012820512820512823,
      "loss": 0.7192,
      "step": 125
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.051881403101396946,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.7401,
      "step": 130
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.042026782084728216,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.7068,
      "step": 135
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.0557128886654579,
      "learning_rate": 0.0001435897435897436,
      "loss": 0.6984,
      "step": 140
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.05570176969330806,
      "learning_rate": 0.00014871794871794872,
      "loss": 0.6639,
      "step": 145
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.06656654996242936,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.7783,
      "step": 150
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.045655704141469376,
      "learning_rate": 0.00015897435897435896,
      "loss": 0.681,
      "step": 155
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.05653995551779284,
      "learning_rate": 0.0001641025641025641,
      "loss": 0.6951,
      "step": 160
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.06359181018131281,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.7425,
      "step": 165
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.05324062210646119,
      "learning_rate": 0.00017435897435897436,
      "loss": 0.7827,
      "step": 170
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.05722015806927473,
      "learning_rate": 0.0001794871794871795,
      "loss": 0.8072,
      "step": 175
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.057554172233914125,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.6969,
      "step": 180
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.06236094601265461,
      "learning_rate": 0.00018974358974358974,
      "loss": 0.7717,
      "step": 185
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.04706829300587947,
      "learning_rate": 0.00019487179487179487,
      "loss": 0.6552,
      "step": 190
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.05907857575128907,
      "learning_rate": 0.0002,
      "loss": 0.7665,
      "step": 195
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.05666486553086404,
      "learning_rate": 0.00019999598996948235,
      "loss": 0.722,
      "step": 200
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.06751342355465378,
      "learning_rate": 0.00019998396019953624,
      "loss": 0.7304,
      "step": 205
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.05870214883989128,
      "learning_rate": 0.0001999639116549566,
      "loss": 0.6464,
      "step": 210
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.06856502375216533,
      "learning_rate": 0.00019993584594364894,
      "loss": 0.6875,
      "step": 215
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.04809000181014521,
      "learning_rate": 0.0001998997653165004,
      "loss": 0.7196,
      "step": 220
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.06330068013492544,
      "learning_rate": 0.00019985567266719934,
      "loss": 0.8071,
      "step": 225
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.06501358267774308,
      "learning_rate": 0.00019980357153200315,
      "loss": 0.7484,
      "step": 230
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.056442995828752306,
      "learning_rate": 0.00019974346608945466,
      "loss": 0.7425,
      "step": 235
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.05136163609317852,
      "learning_rate": 0.00019967536116004698,
      "loss": 0.7365,
      "step": 240
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.06539335014303507,
      "learning_rate": 0.00019959926220583713,
      "loss": 0.7225,
      "step": 245
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.05450780975464187,
      "learning_rate": 0.00019951517533000764,
      "loss": 0.6874,
      "step": 250
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.05390893998675368,
      "learning_rate": 0.00019942310727637724,
      "loss": 0.7312,
      "step": 255
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.05549601275822287,
      "learning_rate": 0.00019932306542886009,
      "loss": 0.7505,
      "step": 260
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.06035962160268456,
      "learning_rate": 0.00019921505781087334,
      "loss": 0.7648,
      "step": 265
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.05407880633367139,
      "learning_rate": 0.00019909909308469398,
      "loss": 0.7745,
      "step": 270
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.07099486719812642,
      "learning_rate": 0.0001989751805507637,
      "loss": 0.7364,
      "step": 275
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.06961377188666044,
      "learning_rate": 0.00019884333014694345,
      "loss": 0.7675,
      "step": 280
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.05666459121830388,
      "learning_rate": 0.00019870355244771607,
      "loss": 0.7478,
      "step": 285
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.06243860942546019,
      "learning_rate": 0.00019855585866333835,
      "loss": 0.7405,
      "step": 290
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.06614433181862764,
      "learning_rate": 0.00019840026063894193,
      "loss": 0.746,
      "step": 295
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.07249007154385097,
      "learning_rate": 0.00019823677085358335,
      "loss": 0.7144,
      "step": 300
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.059290083038044415,
      "learning_rate": 0.00019806540241924317,
      "loss": 0.7228,
      "step": 305
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.0590976532576803,
      "learning_rate": 0.00019788616907977441,
      "loss": 0.7468,
      "step": 310
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.059034355580792085,
      "learning_rate": 0.00019769908520980034,
      "loss": 0.7107,
      "step": 315
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.06039820906529118,
      "learning_rate": 0.00019750416581356146,
      "loss": 0.7255,
      "step": 320
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.07506252340754628,
      "learning_rate": 0.00019730142652371236,
      "loss": 0.6218,
      "step": 325
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.06227640687325107,
      "learning_rate": 0.0001970908836000678,
      "loss": 0.7271,
      "step": 330
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.059379397850922444,
      "learning_rate": 0.00019687255392829877,
      "loss": 0.6962,
      "step": 335
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.06733271176615478,
      "learning_rate": 0.0001966464550185782,
      "loss": 0.6985,
      "step": 340
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.06276187431993832,
      "learning_rate": 0.0001964126050041767,
      "loss": 0.6847,
      "step": 345
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.06756039835970662,
      "learning_rate": 0.0001961710226400081,
      "loss": 0.7006,
      "step": 350
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.058664527111491456,
      "learning_rate": 0.00019592172730112544,
      "loss": 0.7226,
      "step": 355
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.05992656585552198,
      "learning_rate": 0.00019566473898116713,
      "loss": 0.7373,
      "step": 360
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.059131250218324535,
      "learning_rate": 0.0001954000782907532,
      "loss": 0.7174,
      "step": 365
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.05973639610049065,
      "learning_rate": 0.00019512776645583263,
      "loss": 0.7287,
      "step": 370
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.07529175059087607,
      "learning_rate": 0.00019484782531598073,
      "loss": 0.7584,
      "step": 375
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.05996235414061146,
      "learning_rate": 0.00019456027732264784,
      "loss": 0.7802,
      "step": 380
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.0741871606600372,
      "learning_rate": 0.00019426514553735848,
      "loss": 0.7574,
      "step": 385
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.05719814806544979,
      "learning_rate": 0.00019396245362986197,
      "loss": 0.7187,
      "step": 390
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.062306915853923,
      "learning_rate": 0.00019365222587623405,
      "loss": 0.6974,
      "step": 395
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.0578376102759499,
      "learning_rate": 0.00019333448715692995,
      "loss": 0.7263,
      "step": 400
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.05895437900134049,
      "learning_rate": 0.00019300926295478884,
      "loss": 0.7737,
      "step": 405
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.06528074688398286,
      "learning_rate": 0.0001926765793529902,
      "loss": 0.7083,
      "step": 410
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.06352416055439895,
      "learning_rate": 0.00019233646303296205,
      "loss": 0.7538,
      "step": 415
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.05931981491215686,
      "learning_rate": 0.00019198894127224074,
      "loss": 0.7418,
      "step": 420
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.06745142735169624,
      "learning_rate": 0.0001916340419422837,
      "loss": 0.7643,
      "step": 425
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.05313044781100756,
      "learning_rate": 0.00019127179350623372,
      "loss": 0.7045,
      "step": 430
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.058146422311574276,
      "learning_rate": 0.0001909022250166365,
      "loss": 0.7029,
      "step": 435
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.06465624410270512,
      "learning_rate": 0.00019052536611311046,
      "loss": 0.7329,
      "step": 440
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.07476190763880249,
      "learning_rate": 0.00019014124701996973,
      "loss": 0.7337,
      "step": 445
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.06473337914713544,
      "learning_rate": 0.00018974989854379996,
      "loss": 0.7425,
      "step": 450
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.06217891087939524,
      "learning_rate": 0.00018935135207098785,
      "loss": 0.7233,
      "step": 455
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.06538705398075274,
      "learning_rate": 0.00018894563956520374,
      "loss": 0.7181,
      "step": 460
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.06585818368422851,
      "learning_rate": 0.00018853279356483826,
      "loss": 0.7839,
      "step": 465
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.07158174416486009,
      "learning_rate": 0.00018811284718039256,
      "loss": 0.6415,
      "step": 470
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.07546822617864929,
      "learning_rate": 0.00018768583409182305,
      "loss": 0.7182,
      "step": 475
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.05416665482621128,
      "learning_rate": 0.00018725178854584007,
      "loss": 0.7367,
      "step": 480
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.04571169273365327,
      "learning_rate": 0.00018681074535316125,
      "loss": 0.7597,
      "step": 485
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.0662815718580705,
      "learning_rate": 0.00018636273988571991,
      "loss": 0.6998,
      "step": 490
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.06457843383025075,
      "learning_rate": 0.0001859078080738279,
      "loss": 0.6547,
      "step": 495
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.054545024268562024,
      "learning_rate": 0.00018544598640329432,
      "loss": 0.7352,
      "step": 500
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.05705170468183904,
      "learning_rate": 0.00018497731191249894,
      "loss": 0.7663,
      "step": 505
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.05546917332133685,
      "learning_rate": 0.000184501822189422,
      "loss": 0.7412,
      "step": 510
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.061537343417600796,
      "learning_rate": 0.00018401955536862948,
      "loss": 0.7453,
      "step": 515
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.05548194532215062,
      "learning_rate": 0.0001835305501282148,
      "loss": 0.7617,
      "step": 520
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.05596980684672694,
      "learning_rate": 0.00018303484568669667,
      "loss": 0.7057,
      "step": 525
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.05553023356923259,
      "learning_rate": 0.00018253248179987388,
      "loss": 0.666,
      "step": 530
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.06815419873023405,
      "learning_rate": 0.0001820234987576368,
      "loss": 0.7085,
      "step": 535
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.07008274336459665,
      "learning_rate": 0.00018150793738073602,
      "loss": 0.7008,
      "step": 540
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.06075688816959689,
      "learning_rate": 0.00018098583901750867,
      "loss": 0.7204,
      "step": 545
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.0740518099336573,
      "learning_rate": 0.00018045724554056214,
      "loss": 0.7168,
      "step": 550
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.06931559683782222,
      "learning_rate": 0.0001799221993434159,
      "loss": 0.7513,
      "step": 555
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.0808502355660198,
      "learning_rate": 0.00017938074333710157,
      "loss": 0.7221,
      "step": 560
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.057224574717575194,
      "learning_rate": 0.00017883292094672128,
      "loss": 0.7159,
      "step": 565
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.06478263819099996,
      "learning_rate": 0.00017827877610796514,
      "loss": 0.7003,
      "step": 570
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.06759261764339991,
      "learning_rate": 0.00017771835326358743,
      "loss": 0.7051,
      "step": 575
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.054230696689679704,
      "learning_rate": 0.00017715169735984233,
      "loss": 0.6614,
      "step": 580
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.05539403398423511,
      "learning_rate": 0.0001765788538428792,
      "loss": 0.7778,
      "step": 585
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1047776677529583,
      "learning_rate": 0.00017599986865509767,
      "loss": 0.7174,
      "step": 590
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.054148533545801215,
      "learning_rate": 0.00017541478823146327,
      "loss": 0.7501,
      "step": 595
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.062010495956553884,
      "learning_rate": 0.00017482365949578302,
      "loss": 0.7168,
      "step": 600
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.06518741238538411,
      "learning_rate": 0.00017422652985694237,
      "loss": 0.7587,
      "step": 605
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.05897004105933778,
      "learning_rate": 0.00017362344720510278,
      "loss": 0.7,
      "step": 610
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.06674251786961966,
      "learning_rate": 0.00017301445990786102,
      "loss": 0.7656,
      "step": 615
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.059907265503361756,
      "learning_rate": 0.00017239961680637,
      "loss": 0.7302,
      "step": 620
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.05659492984342654,
      "learning_rate": 0.0001717789672114218,
      "loss": 0.8095,
      "step": 625
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.06211544092043524,
      "learning_rate": 0.0001711525608994927,
      "loss": 0.6989,
      "step": 630
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.07196474317183386,
      "learning_rate": 0.00017052044810875126,
      "loss": 0.7653,
      "step": 635
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.0703844878707138,
      "learning_rate": 0.00016988267953502913,
      "loss": 0.7115,
      "step": 640
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.06502783286863137,
      "learning_rate": 0.00016923930632775516,
      "loss": 0.6975,
      "step": 645
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.06625609228888715,
      "learning_rate": 0.00016859038008585326,
      "loss": 0.7186,
      "step": 650
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.060170689666281245,
      "learning_rate": 0.0001679359528536041,
      "loss": 0.6841,
      "step": 655
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.06189493863054274,
      "learning_rate": 0.00016727607711647114,
      "loss": 0.7285,
      "step": 660
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.06529829421352147,
      "learning_rate": 0.00016661080579689132,
      "loss": 0.7046,
      "step": 665
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.07643231666581908,
      "learning_rate": 0.0001659401922500304,
      "loss": 0.712,
      "step": 670
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.06356569960794868,
      "learning_rate": 0.00016526429025950424,
      "loss": 0.6761,
      "step": 675
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.0657780464611706,
      "learning_rate": 0.00016458315403306502,
      "loss": 0.7623,
      "step": 680
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.06528216463724903,
      "learning_rate": 0.0001638968381982538,
      "loss": 0.7035,
      "step": 685
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.05768226228469023,
      "learning_rate": 0.0001632053977980194,
      "loss": 0.7592,
      "step": 690
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.08439997111949711,
      "learning_rate": 0.000162508888286304,
      "loss": 0.7302,
      "step": 695
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.05921155467869476,
      "learning_rate": 0.00016180736552359553,
      "loss": 0.7561,
      "step": 700
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.06479164017299047,
      "learning_rate": 0.00016110088577244773,
      "loss": 0.7319,
      "step": 705
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.06671590691395402,
      "learning_rate": 0.00016038950569296785,
      "loss": 0.7056,
      "step": 710
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.060116735850209595,
      "learning_rate": 0.00015967328233827249,
      "loss": 0.7566,
      "step": 715
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.0671946478188747,
      "learning_rate": 0.00015895227314991178,
      "loss": 0.672,
      "step": 720
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.06337938637321384,
      "learning_rate": 0.00015822653595326275,
      "loss": 0.7015,
      "step": 725
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.05904487810677323,
      "learning_rate": 0.00015749612895289152,
      "loss": 0.7058,
      "step": 730
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.05858071309598837,
      "learning_rate": 0.00015676111072788527,
      "loss": 0.7012,
      "step": 735
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.07042516508192798,
      "learning_rate": 0.00015602154022715435,
      "loss": 0.732,
      "step": 740
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.06845893159668907,
      "learning_rate": 0.0001552774767647043,
      "loss": 0.6938,
      "step": 745
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.0725751629044133,
      "learning_rate": 0.0001545289800148789,
      "loss": 0.7172,
      "step": 750
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.06616285707027898,
      "learning_rate": 0.0001537761100075744,
      "loss": 0.7247,
      "step": 755
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.055431700566913336,
      "learning_rate": 0.00015301892712342482,
      "loss": 0.7647,
      "step": 760
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.07044884291430001,
      "learning_rate": 0.00015225749208895968,
      "loss": 0.7562,
      "step": 765
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.05989702445794234,
      "learning_rate": 0.0001514918659717335,
      "loss": 0.7686,
      "step": 770
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.054216080405442194,
      "learning_rate": 0.00015072211017542813,
      "loss": 0.7333,
      "step": 775
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.05722110654549261,
      "learning_rate": 0.00014994828643492827,
      "loss": 0.7012,
      "step": 780
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.06301212828108414,
      "learning_rate": 0.00014917045681137026,
      "loss": 0.7175,
      "step": 785
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.060197780995845586,
      "learning_rate": 0.0001483886836871646,
      "loss": 0.673,
      "step": 790
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.06219148524987427,
      "learning_rate": 0.00014760302976099304,
      "loss": 0.7114,
      "step": 795
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.07450360556471726,
      "learning_rate": 0.00014681355804278001,
      "loss": 0.7053,
      "step": 800
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.05107293406160496,
      "learning_rate": 0.00014602033184863913,
      "loss": 0.7106,
      "step": 805
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.06265788071525129,
      "learning_rate": 0.00014522341479579533,
      "loss": 0.7966,
      "step": 810
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.05206720433117282,
      "learning_rate": 0.00014442287079748263,
      "loss": 0.7293,
      "step": 815
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.0757239829239859,
      "learning_rate": 0.00014361876405781832,
      "loss": 0.7085,
      "step": 820
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.07831803288446669,
      "learning_rate": 0.00014281115906665374,
      "loss": 0.7687,
      "step": 825
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.07291838892590645,
      "learning_rate": 0.00014200012059440207,
      "loss": 0.7025,
      "step": 830
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.06634284103936865,
      "learning_rate": 0.00014118571368684383,
      "loss": 0.7781,
      "step": 835
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.06089133278443612,
      "learning_rate": 0.00014036800365991008,
      "loss": 0.6563,
      "step": 840
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.071974124311568,
      "learning_rate": 0.00013954705609444404,
      "loss": 0.7234,
      "step": 845
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.0640536251178564,
      "learning_rate": 0.00013872293683094152,
      "loss": 0.7471,
      "step": 850
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.06850057235253415,
      "learning_rate": 0.00013789571196427055,
      "loss": 0.7212,
      "step": 855
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.07611987427477505,
      "learning_rate": 0.00013706544783837022,
      "loss": 0.6921,
      "step": 860
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.06558103857160827,
      "learning_rate": 0.00013623221104093025,
      "loss": 0.7803,
      "step": 865
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.060280688322475906,
      "learning_rate": 0.00013539606839805036,
      "loss": 0.7223,
      "step": 870
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.0644359499391939,
      "learning_rate": 0.00013455708696888085,
      "loss": 0.7332,
      "step": 875
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.0545048505716513,
      "learning_rate": 0.00013371533404024438,
      "loss": 0.7426,
      "step": 880
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.062408064191097595,
      "learning_rate": 0.00013287087712123962,
      "loss": 0.7533,
      "step": 885
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.06389869452396484,
      "learning_rate": 0.00013202378393782692,
      "loss": 0.7179,
      "step": 890
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.08333515596016716,
      "learning_rate": 0.00013117412242739655,
      "loss": 0.6805,
      "step": 895
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.06605322926802382,
      "learning_rate": 0.00013032196073332027,
      "loss": 0.7626,
      "step": 900
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.06438983023731973,
      "learning_rate": 0.00012946736719948607,
      "loss": 0.7285,
      "step": 905
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.05317618314846494,
      "learning_rate": 0.000128610410364817,
      "loss": 0.6886,
      "step": 910
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.06426168509188175,
      "learning_rate": 0.00012775115895777417,
      "loss": 0.7066,
      "step": 915
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.07823316157528669,
      "learning_rate": 0.00012688968189084493,
      "loss": 0.6876,
      "step": 920
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.06352925688166776,
      "learning_rate": 0.00012602604825501587,
      "loss": 0.7281,
      "step": 925
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.0560358701166056,
      "learning_rate": 0.00012516032731423165,
      "loss": 0.714,
      "step": 930
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.05967077285578719,
      "learning_rate": 0.00012429258849984014,
      "loss": 0.7313,
      "step": 935
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.06331893233925272,
      "learning_rate": 0.00012342290140502388,
      "loss": 0.7319,
      "step": 940
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.06346188515705922,
      "learning_rate": 0.00012255133577921868,
      "loss": 0.7298,
      "step": 945
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.06487969791361832,
      "learning_rate": 0.0001216779615225197,
      "loss": 0.6916,
      "step": 950
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.06459701736380498,
      "learning_rate": 0.00012080284868007541,
      "loss": 0.7131,
      "step": 955
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.0637596839504511,
      "learning_rate": 0.0001199260674364699,
      "loss": 0.7483,
      "step": 960
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.058147103713184295,
      "learning_rate": 0.00011904768811009405,
      "loss": 0.7546,
      "step": 965
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.04960680867211613,
      "learning_rate": 0.00011816778114750593,
      "loss": 0.7271,
      "step": 970
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.054982008015380404,
      "learning_rate": 0.00011728641711778103,
      "loss": 0.7482,
      "step": 975
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.05707338359759536,
      "learning_rate": 0.00011640366670685248,
      "loss": 0.6697,
      "step": 980
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.05927872290783747,
      "learning_rate": 0.00011551960071184195,
      "loss": 0.8193,
      "step": 985
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.07725069996761502,
      "learning_rate": 0.00011463429003538196,
      "loss": 0.7432,
      "step": 990
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.05725145842711473,
      "learning_rate": 0.000113747805679929,
      "loss": 0.759,
      "step": 995
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.07186182485542994,
      "learning_rate": 0.00011286021874206952,
      "loss": 0.6704,
      "step": 1000
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.0686861437931256,
      "learning_rate": 0.00011197160040681762,
      "loss": 0.8042,
      "step": 1005
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.05601004454883831,
      "learning_rate": 0.0001110820219419062,
      "loss": 0.7338,
      "step": 1010
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.05732144501117736,
      "learning_rate": 0.0001101915546920711,
      "loss": 0.7266,
      "step": 1015
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.05767273887002733,
      "learning_rate": 0.00010930027007332923,
      "loss": 0.6721,
      "step": 1020
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.06070633674818661,
      "learning_rate": 0.00010840823956725103,
      "loss": 0.7276,
      "step": 1025
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.0608684388554378,
      "learning_rate": 0.00010751553471522757,
      "loss": 0.7308,
      "step": 1030
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.058955147100918356,
      "learning_rate": 0.00010662222711273279,
      "loss": 0.6947,
      "step": 1035
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.060134707447051496,
      "learning_rate": 0.00010572838840358168,
      "loss": 0.6722,
      "step": 1040
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.06428460825842836,
      "learning_rate": 0.00010483409027418425,
      "loss": 0.6918,
      "step": 1045
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.06287339641207959,
      "learning_rate": 0.00010393940444779635,
      "loss": 0.6713,
      "step": 1050
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.0633410011658368,
      "learning_rate": 0.00010304440267876727,
      "loss": 0.7685,
      "step": 1055
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.06581703699322343,
      "learning_rate": 0.00010214915674678523,
      "loss": 0.7767,
      "step": 1060
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.07413976856973425,
      "learning_rate": 0.00010125373845112034,
      "loss": 0.762,
      "step": 1065
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.07167916408221033,
      "learning_rate": 0.00010035821960486643,
      "loss": 0.7376,
      "step": 1070
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.06708982281673678,
      "learning_rate": 9.946267202918157e-05,
      "loss": 0.6889,
      "step": 1075
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.0819220894814777,
      "learning_rate": 9.856716754752796e-05,
      "loss": 0.6908,
      "step": 1080
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.06608143578295782,
      "learning_rate": 9.767177797991155e-05,
      "loss": 0.7399,
      "step": 1085
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.05592575944392656,
      "learning_rate": 9.677657513712221e-05,
      "loss": 0.7138,
      "step": 1090
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.0546831567691992,
      "learning_rate": 9.588163081497427e-05,
      "loss": 0.7627,
      "step": 1095
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.06346039164995447,
      "learning_rate": 9.498701678854865e-05,
      "loss": 0.7236,
      "step": 1100
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.0746991316854835,
      "learning_rate": 9.409280480643628e-05,
      "loss": 0.7629,
      "step": 1105
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.06237683578790127,
      "learning_rate": 9.319906658498389e-05,
      "loss": 0.7416,
      "step": 1110
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.06387729635810044,
      "learning_rate": 9.230587380254237e-05,
      "loss": 0.7184,
      "step": 1115
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.053639300773281065,
      "learning_rate": 9.141329809371803e-05,
      "loss": 0.7293,
      "step": 1120
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.055148045352961035,
      "learning_rate": 9.052141104362748e-05,
      "loss": 0.6874,
      "step": 1125
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.06026774699282564,
      "learning_rate": 8.963028418215653e-05,
      "loss": 0.792,
      "step": 1130
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.05815501167689513,
      "learning_rate": 8.873998897822336e-05,
      "loss": 0.7768,
      "step": 1135
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.06092520719858454,
      "learning_rate": 8.785059683404672e-05,
      "loss": 0.6928,
      "step": 1140
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.06680081897157324,
      "learning_rate": 8.696217907941941e-05,
      "loss": 0.744,
      "step": 1145
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.07064620273754864,
      "learning_rate": 8.607480696598762e-05,
      "loss": 0.6606,
      "step": 1150
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.06809202255793978,
      "learning_rate": 8.518855166153644e-05,
      "loss": 0.7513,
      "step": 1155
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.06295987055363506,
      "learning_rate": 8.43034842442822e-05,
      "loss": 0.7085,
      "step": 1160
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.0628445698799043,
      "learning_rate": 8.341967569717202e-05,
      "loss": 0.668,
      "step": 1165
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.06659960321763707,
      "learning_rate": 8.253719690219079e-05,
      "loss": 0.6853,
      "step": 1170
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.0688529485097757,
      "learning_rate": 8.165611863467644e-05,
      "loss": 0.7052,
      "step": 1175
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.06023727493720994,
      "learning_rate": 8.077651155764387e-05,
      "loss": 0.6858,
      "step": 1180
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.06220450495769832,
      "learning_rate": 7.98984462161175e-05,
      "loss": 0.7433,
      "step": 1185
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.053745746371760474,
      "learning_rate": 7.902199303147363e-05,
      "loss": 0.7362,
      "step": 1190
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.061298250546248934,
      "learning_rate": 7.814722229579264e-05,
      "loss": 0.7385,
      "step": 1195
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.06481532056857399,
      "learning_rate": 7.727420416622144e-05,
      "loss": 0.7399,
      "step": 1200
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.05841856930605358,
      "learning_rate": 7.640300865934687e-05,
      "loss": 0.6672,
      "step": 1205
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.06847941130435668,
      "learning_rate": 7.553370564558032e-05,
      "loss": 0.73,
      "step": 1210
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.053093666640836194,
      "learning_rate": 7.46663648435541e-05,
      "loss": 0.7162,
      "step": 1215
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.07700372089299697,
      "learning_rate": 7.380105581452987e-05,
      "loss": 0.7027,
      "step": 1220
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.06497009385059972,
      "learning_rate": 7.293784795681994e-05,
      "loss": 0.7629,
      "step": 1225
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.07176927987751118,
      "learning_rate": 7.207681050022132e-05,
      "loss": 0.7009,
      "step": 1230
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.06323784869312263,
      "learning_rate": 7.121801250046363e-05,
      "loss": 0.8098,
      "step": 1235
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.05550749509584078,
      "learning_rate": 7.036152283367056e-05,
      "loss": 0.6892,
      "step": 1240
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.06233653087775129,
      "learning_rate": 6.950741019083617e-05,
      "loss": 0.722,
      "step": 1245
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.061296742852639696,
      "learning_rate": 6.865574307231575e-05,
      "loss": 0.7057,
      "step": 1250
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.07218037431449478,
      "learning_rate": 6.780658978233199e-05,
      "loss": 0.7104,
      "step": 1255
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.06031232298569737,
      "learning_rate": 6.696001842349702e-05,
      "loss": 0.7238,
      "step": 1260
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.06261408542212123,
      "learning_rate": 6.611609689135056e-05,
      "loss": 0.7804,
      "step": 1265
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.06509293878957043,
      "learning_rate": 6.527489286891459e-05,
      "loss": 0.6191,
      "step": 1270
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.0637053230023183,
      "learning_rate": 6.443647382126509e-05,
      "loss": 0.7272,
      "step": 1275
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.060890338775236155,
      "learning_rate": 6.360090699012145e-05,
      "loss": 0.7158,
      "step": 1280
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.06390720535215949,
      "learning_rate": 6.27682593884535e-05,
      "loss": 0.6418,
      "step": 1285
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.059941265719477915,
      "learning_rate": 6.193859779510712e-05,
      "loss": 0.7396,
      "step": 1290
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.057335301960845145,
      "learning_rate": 6.111198874944845e-05,
      "loss": 0.7136,
      "step": 1295
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.0669688076887658,
      "learning_rate": 6.0288498546027536e-05,
      "loss": 0.7683,
      "step": 1300
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.06887004197989798,
      "learning_rate": 5.946819322926127e-05,
      "loss": 0.6736,
      "step": 1305
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.06281493310941981,
      "learning_rate": 5.865113858813673e-05,
      "loss": 0.7169,
      "step": 1310
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.05868856490220455,
      "learning_rate": 5.783740015093484e-05,
      "loss": 0.7227,
      "step": 1315
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.059464256939025885,
      "learning_rate": 5.702704317997492e-05,
      "loss": 0.6855,
      "step": 1320
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.06386543918702894,
      "learning_rate": 5.6220132666380635e-05,
      "loss": 0.6664,
      "step": 1325
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.06630497213003285,
      "learning_rate": 5.541673332486773e-05,
      "loss": 0.7539,
      "step": 1330
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0572397341438381,
      "learning_rate": 5.4616909588553674e-05,
      "loss": 0.8131,
      "step": 1335
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.0554406728774194,
      "learning_rate": 5.3820725603790346e-05,
      "loss": 0.7787,
      "step": 1340
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.05900113915228878,
      "learning_rate": 5.30282452250193e-05,
      "loss": 0.7764,
      "step": 1345
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.06385527269768287,
      "learning_rate": 5.223953200965055e-05,
      "loss": 0.7301,
      "step": 1350
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.06245137688685591,
      "learning_rate": 5.145464921296537e-05,
      "loss": 0.7958,
      "step": 1355
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.0684043207080321,
      "learning_rate": 5.067365978304315e-05,
      "loss": 0.6871,
      "step": 1360
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.06711154333394342,
      "learning_rate": 4.9896626355712805e-05,
      "loss": 0.694,
      "step": 1365
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.05283989756729982,
      "learning_rate": 4.912361124952948e-05,
      "loss": 0.7149,
      "step": 1370
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.060600777598165734,
      "learning_rate": 4.835467646077656e-05,
      "loss": 0.6808,
      "step": 1375
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.05563364488701937,
      "learning_rate": 4.7589883658493296e-05,
      "loss": 0.6914,
      "step": 1380
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.06530550198410516,
      "learning_rate": 4.682929417952939e-05,
      "loss": 0.6881,
      "step": 1385
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.052698731359847616,
      "learning_rate": 4.6072969023625165e-05,
      "loss": 0.7082,
      "step": 1390
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.05983275498927436,
      "learning_rate": 4.532096884851978e-05,
      "loss": 0.7188,
      "step": 1395
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.06612287041729081,
      "learning_rate": 4.457335396508631e-05,
      "loss": 0.7189,
      "step": 1400
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.05677308006082468,
      "learning_rate": 4.383018433249464e-05,
      "loss": 0.7081,
      "step": 1405
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.06237896853204969,
      "learning_rate": 4.309151955340297e-05,
      "loss": 0.6921,
      "step": 1410
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.05812595825532448,
      "learning_rate": 4.2357418869177354e-05,
      "loss": 0.7548,
      "step": 1415
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.06741213496816625,
      "learning_rate": 4.162794115514078e-05,
      "loss": 0.7643,
      "step": 1420
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.0631021793158054,
      "learning_rate": 4.0903144915851174e-05,
      "loss": 0.6469,
      "step": 1425
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.05831509904707476,
      "learning_rate": 4.018308828040924e-05,
      "loss": 0.6687,
      "step": 1430
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.0542248638635785,
      "learning_rate": 3.946782899779667e-05,
      "loss": 0.7034,
      "step": 1435
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.06625784513238857,
      "learning_rate": 3.875742443224451e-05,
      "loss": 0.721,
      "step": 1440
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.05870827407396205,
      "learning_rate": 3.805193155863247e-05,
      "loss": 0.675,
      "step": 1445
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.059612966400709386,
      "learning_rate": 3.7351406957919636e-05,
      "loss": 0.7293,
      "step": 1450
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.06211176018305973,
      "learning_rate": 3.665590681260658e-05,
      "loss": 0.7181,
      "step": 1455
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.05866568540322711,
      "learning_rate": 3.59654869022294e-05,
      "loss": 0.7666,
      "step": 1460
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.0501465470248372,
      "learning_rate": 3.5280202598886324e-05,
      "loss": 0.7647,
      "step": 1465
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.060705521344698324,
      "learning_rate": 3.4600108862796796e-05,
      "loss": 0.7136,
      "step": 1470
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.06092141310578522,
      "learning_rate": 3.392526023789349e-05,
      "loss": 0.7384,
      "step": 1475
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.04920843840426036,
      "learning_rate": 3.325571084744803e-05,
      "loss": 0.6823,
      "step": 1480
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.06578821588742692,
      "learning_rate": 3.259151438973024e-05,
      "loss": 0.7787,
      "step": 1485
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.054652544113093456,
      "learning_rate": 3.1932724133701344e-05,
      "loss": 0.6821,
      "step": 1490
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.06096008553815832,
      "learning_rate": 3.1279392914742046e-05,
      "loss": 0.735,
      "step": 1495
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.058962269998868645,
      "learning_rate": 3.06315731304148e-05,
      "loss": 0.7616,
      "step": 1500
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.05523256517584398,
      "learning_rate": 2.998931673626175e-05,
      "loss": 0.6681,
      "step": 1505
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.05959688310829463,
      "learning_rate": 2.935267524163774e-05,
      "loss": 0.7581,
      "step": 1510
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.05982538851896185,
      "learning_rate": 2.872169970557913e-05,
      "loss": 0.7541,
      "step": 1515
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.06494218833414687,
      "learning_rate": 2.8096440732709083e-05,
      "loss": 0.7742,
      "step": 1520
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.06859915764079565,
      "learning_rate": 2.7476948469178887e-05,
      "loss": 0.7247,
      "step": 1525
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.09010453444524248,
      "learning_rate": 2.6863272598646106e-05,
      "loss": 0.712,
      "step": 1530
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.06068755169221279,
      "learning_rate": 2.625546233829016e-05,
      "loss": 0.7516,
      "step": 1535
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.06062575490061368,
      "learning_rate": 2.5653566434864928e-05,
      "loss": 0.7284,
      "step": 1540
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.06813879053211623,
      "learning_rate": 2.5057633160789184e-05,
      "loss": 0.7177,
      "step": 1545
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.05549352383131599,
      "learning_rate": 2.446771031027527e-05,
      "loss": 0.7735,
      "step": 1550
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.06167056235702363,
      "learning_rate": 2.3883845195495878e-05,
      "loss": 0.7829,
      "step": 1555
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.05941849303346335,
      "learning_rate": 2.330608464278953e-05,
      "loss": 0.7349,
      "step": 1560
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.059930008570328284,
      "learning_rate": 2.273447498890521e-05,
      "loss": 0.7793,
      "step": 1565
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.05579818006374695,
      "learning_rate": 2.2169062077286075e-05,
      "loss": 0.6831,
      "step": 1570
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.05556533610425669,
      "learning_rate": 2.1609891254392678e-05,
      "loss": 0.6766,
      "step": 1575
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.07117318253440096,
      "learning_rate": 2.1057007366066373e-05,
      "loss": 0.7592,
      "step": 1580
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.06880636339513395,
      "learning_rate": 2.0510454753932395e-05,
      "loss": 0.6818,
      "step": 1585
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.05185758649755237,
      "learning_rate": 1.9970277251843862e-05,
      "loss": 0.7131,
      "step": 1590
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.0569426799635938,
      "learning_rate": 1.9436518182366158e-05,
      "loss": 0.723,
      "step": 1595
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.05730020536410127,
      "learning_rate": 1.8909220353302392e-05,
      "loss": 0.7303,
      "step": 1600
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.06163290829765527,
      "learning_rate": 1.838842605426031e-05,
      "loss": 0.6702,
      "step": 1605
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.05358008544806708,
      "learning_rate": 1.7874177053260598e-05,
      "loss": 0.6647,
      "step": 1610
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.08348745178054953,
      "learning_rate": 1.736651459338695e-05,
      "loss": 0.6768,
      "step": 1615
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.059233304270911794,
      "learning_rate": 1.6865479389478545e-05,
      "loss": 0.7239,
      "step": 1620
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.0530792952405111,
      "learning_rate": 1.6371111624864543e-05,
      "loss": 0.7146,
      "step": 1625
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.07089338712434175,
      "learning_rate": 1.5883450948141377e-05,
      "loss": 0.7008,
      "step": 1630
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.05743797721150877,
      "learning_rate": 1.540253646999299e-05,
      "loss": 0.7422,
      "step": 1635
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.054938710195774586,
      "learning_rate": 1.4928406760054059e-05,
      "loss": 0.7121,
      "step": 1640
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.06182870818089227,
      "learning_rate": 1.4461099843816684e-05,
      "loss": 0.7006,
      "step": 1645
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.051352577478825545,
      "learning_rate": 1.4000653199580782e-05,
      "loss": 0.759,
      "step": 1650
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.059641862900702905,
      "learning_rate": 1.3547103755448287e-05,
      "loss": 0.742,
      "step": 1655
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.05397773348884262,
      "learning_rate": 1.3100487886361379e-05,
      "loss": 0.7741,
      "step": 1660
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.05135238686196935,
      "learning_rate": 1.266084141118542e-05,
      "loss": 0.7079,
      "step": 1665
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.057288371085909104,
      "learning_rate": 1.2228199589835999e-05,
      "loss": 0.7327,
      "step": 1670
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.06493631809483912,
      "learning_rate": 1.1802597120451286e-05,
      "loss": 0.7295,
      "step": 1675
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.06770101954166549,
      "learning_rate": 1.1384068136609105e-05,
      "loss": 0.7514,
      "step": 1680
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.05371862338343528,
      "learning_rate": 1.0972646204589377e-05,
      "loss": 0.6835,
      "step": 1685
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.05759804328762846,
      "learning_rate": 1.0568364320682178e-05,
      "loss": 0.7513,
      "step": 1690
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.06129923897917773,
      "learning_rate": 1.0171254908541372e-05,
      "loss": 0.6705,
      "step": 1695
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.053581168501307445,
      "learning_rate": 9.781349816584162e-06,
      "loss": 0.7042,
      "step": 1700
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.05603939945818747,
      "learning_rate": 9.398680315436903e-06,
      "loss": 0.7258,
      "step": 1705
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.06584992700683939,
      "learning_rate": 9.023277095427173e-06,
      "loss": 0.6769,
      "step": 1710
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.059968649425522685,
      "learning_rate": 8.655170264122303e-06,
      "loss": 0.7463,
      "step": 1715
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.05488029646190535,
      "learning_rate": 8.294389343914899e-06,
      "loss": 0.722,
      "step": 1720
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.05526234423387249,
      "learning_rate": 7.940963269654922e-06,
      "loss": 0.7629,
      "step": 1725
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.05556936448236485,
      "learning_rate": 7.594920386329252e-06,
      "loss": 0.7161,
      "step": 1730
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.057128352068576076,
      "learning_rate": 7.256288446788362e-06,
      "loss": 0.6614,
      "step": 1735
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.06678671968173212,
      "learning_rate": 6.925094609520455e-06,
      "loss": 0.7391,
      "step": 1740
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.0551309681633079,
      "learning_rate": 6.601365436473439e-06,
      "loss": 0.7053,
      "step": 1745
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.0660459237833797,
      "learning_rate": 6.2851268909245865e-06,
      "loss": 0.6966,
      "step": 1750
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.06028870099044076,
      "learning_rate": 5.976404335398256e-06,
      "loss": 0.7425,
      "step": 1755
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.04676509027459768,
      "learning_rate": 5.675222529631841e-06,
      "loss": 0.7178,
      "step": 1760
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.05560159391722265,
      "learning_rate": 5.381605628590003e-06,
      "loss": 0.7285,
      "step": 1765
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.06016761021331081,
      "learning_rate": 5.095577180527378e-06,
      "loss": 0.7333,
      "step": 1770
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.0607495003253662,
      "learning_rate": 4.817160125100106e-06,
      "loss": 0.7747,
      "step": 1775
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.05994300070228749,
      "learning_rate": 4.546376791525975e-06,
      "loss": 0.7357,
      "step": 1780
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.052705940192527105,
      "learning_rate": 4.2832488967935795e-06,
      "loss": 0.7319,
      "step": 1785
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.05605139500885904,
      "learning_rate": 4.02779754392072e-06,
      "loss": 0.7196,
      "step": 1790
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.06832445824454762,
      "learning_rate": 3.780043220261764e-06,
      "loss": 0.7419,
      "step": 1795
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.07505662282091412,
      "learning_rate": 3.540005795864709e-06,
      "loss": 0.6612,
      "step": 1800
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.06215112837654046,
      "learning_rate": 3.3077045218775192e-06,
      "loss": 0.7493,
      "step": 1805
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.06733733726055094,
      "learning_rate": 3.0831580290041184e-06,
      "loss": 0.7542,
      "step": 1810
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.04850593633489853,
      "learning_rate": 2.8663843260103074e-06,
      "loss": 0.7319,
      "step": 1815
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.07012766389455326,
      "learning_rate": 2.6574007982793857e-06,
      "loss": 0.7144,
      "step": 1820
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.06128302394909496,
      "learning_rate": 2.456224206417812e-06,
      "loss": 0.6819,
      "step": 1825
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.06568885594871854,
      "learning_rate": 2.262870684911045e-06,
      "loss": 0.7466,
      "step": 1830
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.05638801020800785,
      "learning_rate": 2.0773557408295343e-06,
      "loss": 0.7104,
      "step": 1835
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.05606985820795626,
      "learning_rate": 1.8996942525850047e-06,
      "loss": 0.7032,
      "step": 1840
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.055316558262847634,
      "learning_rate": 1.7299004687372665e-06,
      "loss": 0.6945,
      "step": 1845
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.0589945616388343,
      "learning_rate": 1.5679880068514174e-06,
      "loss": 0.6811,
      "step": 1850
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.0667279053464632,
      "learning_rate": 1.4139698524057165e-06,
      "loss": 0.7469,
      "step": 1855
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.06272671335327658,
      "learning_rate": 1.2678583577501624e-06,
      "loss": 0.7279,
      "step": 1860
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.06410423650132317,
      "learning_rate": 1.1296652411158182e-06,
      "loss": 0.724,
      "step": 1865
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.0554322802504446,
      "learning_rate": 9.994015856749527e-07,
      "loss": 0.6884,
      "step": 1870
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.05363867115392121,
      "learning_rate": 8.770778386522627e-07,
      "loss": 0.6692,
      "step": 1875
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.06290861069536181,
      "learning_rate": 7.627038104869199e-07,
      "loss": 0.7241,
      "step": 1880
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.058034813454390606,
      "learning_rate": 6.562886740457797e-07,
      "loss": 0.7169,
      "step": 1885
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.0569479748353595,
      "learning_rate": 5.578409638877457e-07,
      "loss": 0.6471,
      "step": 1890
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.06498145548324454,
      "learning_rate": 4.6736857557925227e-07,
      "loss": 0.7028,
      "step": 1895
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.058277777474149445,
      "learning_rate": 3.8487876506106966e-07,
      "loss": 0.784,
      "step": 1900
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.06195969485337902,
      "learning_rate": 3.1037814806634815e-07,
      "loss": 0.6927,
      "step": 1905
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.05456887165409484,
      "learning_rate": 2.43872699590042e-07,
      "loss": 0.6691,
      "step": 1910
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.05499949679669644,
      "learning_rate": 1.8536775340970425e-07,
      "loss": 0.7706,
      "step": 1915
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.06327430290778817,
      "learning_rate": 1.348680016577397e-07,
      "loss": 0.7543,
      "step": 1920
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.06143220825866882,
      "learning_rate": 9.237749444505062e-08,
      "loss": 0.7157,
      "step": 1925
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.057790877621458125,
      "learning_rate": 5.7899639536251883e-08,
      "loss": 0.7225,
      "step": 1930
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.05824054504791697,
      "learning_rate": 3.143720207635648e-08,
      "loss": 0.7305,
      "step": 1935
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.0708892592599129,
      "learning_rate": 1.299230436898613e-08,
      "loss": 0.7639,
      "step": 1940
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.061086419354873724,
      "learning_rate": 2.566425706218567e-09,
      "loss": 0.6812,
      "step": 1945
    },
    {
      "epoch": 1.0,
      "eval_loss": NaN,
      "eval_runtime": 1970.507,
      "eval_samples_per_second": 3.518,
      "eval_steps_per_second": 0.879,
      "step": 1949
    },
    {
      "epoch": 1.0,
      "step": 1949,
      "total_flos": 1.2135540828143616e+16,
      "train_loss": 0.7257849163982182,
      "train_runtime": 18158.5302,
      "train_samples_per_second": 3.434,
      "train_steps_per_second": 0.107
    }
  ],
  "logging_steps": 5,
  "max_steps": 1949,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 1.2135540828143616e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}