{
  "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": 0.30600793318211245,
      "learning_rate": 1.0256410256410257e-06,
      "loss": 0.9032,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.2514201879313361,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.774,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.2815342243617038,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 0.8071,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.4342429919289369,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.8463,
      "step": 15
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.40134817145365687,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.7641,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.4801336194121041,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.7841,
      "step": 25
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3662347234181133,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.8035,
      "step": 30
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3828937627126127,
      "learning_rate": 3.58974358974359e-05,
      "loss": 0.7619,
      "step": 35
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.5209991714750453,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 0.7523,
      "step": 40
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3538483930059845,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.7162,
      "step": 45
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3148565907399919,
      "learning_rate": 5.128205128205128e-05,
      "loss": 0.7221,
      "step": 50
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.32280818983272175,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 0.7327,
      "step": 55
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.34859257467971205,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.8037,
      "step": 60
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.44817375010146265,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.7524,
      "step": 65
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.42173446308945434,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.709,
      "step": 70
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3363076863733399,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.7941,
      "step": 75
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.27688030101735517,
      "learning_rate": 8.205128205128205e-05,
      "loss": 0.6823,
      "step": 80
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3723821338885567,
      "learning_rate": 8.717948717948718e-05,
      "loss": 0.7341,
      "step": 85
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3186657332784112,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.7346,
      "step": 90
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.37306291352319626,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.7743,
      "step": 95
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.28945644880254084,
      "learning_rate": 0.00010256410256410256,
      "loss": 0.7818,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.34480275462309634,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.754,
      "step": 105
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.3155913853718325,
      "learning_rate": 0.00011282051282051283,
      "loss": 0.7594,
      "step": 110
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.3059326537424213,
      "learning_rate": 0.00011794871794871796,
      "loss": 0.7158,
      "step": 115
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.27974812973149993,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.7589,
      "step": 120
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.299433207874807,
      "learning_rate": 0.00012820512820512823,
      "loss": 0.7197,
      "step": 125
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.2612132298308243,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.7397,
      "step": 130
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.2118754575979382,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.7062,
      "step": 135
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.28002913954832237,
      "learning_rate": 0.0001435897435897436,
      "loss": 0.6977,
      "step": 140
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.2764004850870581,
      "learning_rate": 0.00014871794871794872,
      "loss": 0.6647,
      "step": 145
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.3259429104445731,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.7782,
      "step": 150
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.23129511604751066,
      "learning_rate": 0.00015897435897435896,
      "loss": 0.6811,
      "step": 155
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.28280035674970594,
      "learning_rate": 0.0001641025641025641,
      "loss": 0.6948,
      "step": 160
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.30419155366348055,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.7425,
      "step": 165
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.27453468529046765,
      "learning_rate": 0.00017435897435897436,
      "loss": 0.7834,
      "step": 170
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.28946812946398637,
      "learning_rate": 0.0001794871794871795,
      "loss": 0.8073,
      "step": 175
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.29153757015873033,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.6967,
      "step": 180
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.3115374789472242,
      "learning_rate": 0.00018974358974358974,
      "loss": 0.7717,
      "step": 185
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.2437427835538451,
      "learning_rate": 0.00019487179487179487,
      "loss": 0.6544,
      "step": 190
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.28438457622449187,
      "learning_rate": 0.0002,
      "loss": 0.7664,
      "step": 195
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.284243320695823,
      "learning_rate": 0.00019999598996948235,
      "loss": 0.7218,
      "step": 200
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.32936450674202034,
      "learning_rate": 0.00019998396019953624,
      "loss": 0.7304,
      "step": 205
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.2916467132891398,
      "learning_rate": 0.0001999639116549566,
      "loss": 0.6466,
      "step": 210
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.2958357263986098,
      "learning_rate": 0.00019993584594364894,
      "loss": 0.6867,
      "step": 215
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.23925612772585692,
      "learning_rate": 0.0001998997653165004,
      "loss": 0.7199,
      "step": 220
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.31281056186446304,
      "learning_rate": 0.00019985567266719934,
      "loss": 0.807,
      "step": 225
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.32585674000063625,
      "learning_rate": 0.00019980357153200315,
      "loss": 0.749,
      "step": 230
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.28088368121647767,
      "learning_rate": 0.00019974346608945466,
      "loss": 0.7427,
      "step": 235
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.2563659676800454,
      "learning_rate": 0.00019967536116004698,
      "loss": 0.7364,
      "step": 240
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.3258833727345096,
      "learning_rate": 0.00019959926220583713,
      "loss": 0.7233,
      "step": 245
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.2723443980110157,
      "learning_rate": 0.00019951517533000764,
      "loss": 0.688,
      "step": 250
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.2821535086258948,
      "learning_rate": 0.00019942310727637724,
      "loss": 0.7313,
      "step": 255
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.27498014739427085,
      "learning_rate": 0.00019932306542886009,
      "loss": 0.7508,
      "step": 260
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.30120772918243055,
      "learning_rate": 0.00019921505781087334,
      "loss": 0.7653,
      "step": 265
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.27174214941327507,
      "learning_rate": 0.00019909909308469398,
      "loss": 0.7749,
      "step": 270
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.345210052811732,
      "learning_rate": 0.0001989751805507637,
      "loss": 0.7369,
      "step": 275
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.28776718331960655,
      "learning_rate": 0.00019884333014694345,
      "loss": 0.7672,
      "step": 280
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.28713153044283013,
      "learning_rate": 0.00019870355244771607,
      "loss": 0.7477,
      "step": 285
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.3175349018262891,
      "learning_rate": 0.00019855585866333835,
      "loss": 0.74,
      "step": 290
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.32783729980987397,
      "learning_rate": 0.00019840026063894193,
      "loss": 0.7451,
      "step": 295
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.3607718431965085,
      "learning_rate": 0.00019823677085358335,
      "loss": 0.7144,
      "step": 300
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2896074141326796,
      "learning_rate": 0.00019806540241924317,
      "loss": 0.7222,
      "step": 305
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2935065689751967,
      "learning_rate": 0.00019788616907977441,
      "loss": 0.7462,
      "step": 310
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2946495026306362,
      "learning_rate": 0.00019769908520980034,
      "loss": 0.7109,
      "step": 315
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.299977706975683,
      "learning_rate": 0.00019750416581356146,
      "loss": 0.7247,
      "step": 320
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.3819296604570482,
      "learning_rate": 0.00019730142652371236,
      "loss": 0.6217,
      "step": 325
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.31866635645461877,
      "learning_rate": 0.0001970908836000678,
      "loss": 0.7275,
      "step": 330
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.2967755112932246,
      "learning_rate": 0.00019687255392829877,
      "loss": 0.6961,
      "step": 335
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.3418334469211197,
      "learning_rate": 0.0001966464550185782,
      "loss": 0.6983,
      "step": 340
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.31848296393708386,
      "learning_rate": 0.0001964126050041767,
      "loss": 0.6851,
      "step": 345
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.342681097635205,
      "learning_rate": 0.0001961710226400081,
      "loss": 0.7009,
      "step": 350
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.2932106673228495,
      "learning_rate": 0.00019592172730112544,
      "loss": 0.7233,
      "step": 355
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.29821544198196664,
      "learning_rate": 0.00019566473898116713,
      "loss": 0.7364,
      "step": 360
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.2924971079235035,
      "learning_rate": 0.0001954000782907532,
      "loss": 0.7185,
      "step": 365
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.2974934399315773,
      "learning_rate": 0.00019512776645583263,
      "loss": 0.7285,
      "step": 370
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.38021458880390585,
      "learning_rate": 0.00019484782531598073,
      "loss": 0.7585,
      "step": 375
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.2981430588937584,
      "learning_rate": 0.00019456027732264784,
      "loss": 0.7803,
      "step": 380
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.3611349315632082,
      "learning_rate": 0.00019426514553735848,
      "loss": 0.7574,
      "step": 385
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.28782831286015065,
      "learning_rate": 0.00019396245362986197,
      "loss": 0.7185,
      "step": 390
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.3101501221325524,
      "learning_rate": 0.00019365222587623405,
      "loss": 0.6969,
      "step": 395
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.29633803807242076,
      "learning_rate": 0.00019333448715692995,
      "loss": 0.727,
      "step": 400
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.29649195543157963,
      "learning_rate": 0.00019300926295478884,
      "loss": 0.7737,
      "step": 405
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.3270473427640853,
      "learning_rate": 0.0001926765793529902,
      "loss": 0.7077,
      "step": 410
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.31587090374683174,
      "learning_rate": 0.00019233646303296205,
      "loss": 0.7535,
      "step": 415
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.28445746401301314,
      "learning_rate": 0.00019198894127224074,
      "loss": 0.7416,
      "step": 420
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.33806783414313096,
      "learning_rate": 0.0001916340419422837,
      "loss": 0.7646,
      "step": 425
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.26875171408200743,
      "learning_rate": 0.00019127179350623372,
      "loss": 0.7042,
      "step": 430
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.2919429137669854,
      "learning_rate": 0.0001909022250166365,
      "loss": 0.7032,
      "step": 435
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3342950205904537,
      "learning_rate": 0.00019052536611311046,
      "loss": 0.7322,
      "step": 440
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.37058231737908043,
      "learning_rate": 0.00019014124701996973,
      "loss": 0.7341,
      "step": 445
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.31051296314991944,
      "learning_rate": 0.00018974989854379996,
      "loss": 0.7426,
      "step": 450
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3080588455443033,
      "learning_rate": 0.00018935135207098785,
      "loss": 0.7238,
      "step": 455
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.324352920388171,
      "learning_rate": 0.00018894563956520374,
      "loss": 0.7187,
      "step": 460
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3243888155901372,
      "learning_rate": 0.00018853279356483826,
      "loss": 0.7838,
      "step": 465
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.35477016603327965,
      "learning_rate": 0.00018811284718039256,
      "loss": 0.6414,
      "step": 470
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3590058679310689,
      "learning_rate": 0.00018768583409182305,
      "loss": 0.7181,
      "step": 475
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.26904364380570617,
      "learning_rate": 0.00018725178854584007,
      "loss": 0.7359,
      "step": 480
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.23152114902162865,
      "learning_rate": 0.00018681074535316125,
      "loss": 0.7598,
      "step": 485
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3362860964416464,
      "learning_rate": 0.00018636273988571991,
      "loss": 0.6994,
      "step": 490
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.32491680412743684,
      "learning_rate": 0.0001859078080738279,
      "loss": 0.6543,
      "step": 495
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.2765388519594696,
      "learning_rate": 0.00018544598640329432,
      "loss": 0.7357,
      "step": 500
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.2816315947273805,
      "learning_rate": 0.00018497731191249894,
      "loss": 0.7661,
      "step": 505
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.27513815571955674,
      "learning_rate": 0.000184501822189422,
      "loss": 0.7409,
      "step": 510
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.29971326936735054,
      "learning_rate": 0.00018401955536862948,
      "loss": 0.7458,
      "step": 515
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.27499944559694983,
      "learning_rate": 0.0001835305501282148,
      "loss": 0.7617,
      "step": 520
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.28113335320963195,
      "learning_rate": 0.00018303484568669667,
      "loss": 0.7064,
      "step": 525
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.27519607108872957,
      "learning_rate": 0.00018253248179987388,
      "loss": 0.6662,
      "step": 530
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.3411668902700722,
      "learning_rate": 0.0001820234987576368,
      "loss": 0.708,
      "step": 535
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3466535243515036,
      "learning_rate": 0.00018150793738073602,
      "loss": 0.7017,
      "step": 540
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3006025390063751,
      "learning_rate": 0.00018098583901750867,
      "loss": 0.7195,
      "step": 545
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3342265196706073,
      "learning_rate": 0.00018045724554056214,
      "loss": 0.7165,
      "step": 550
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.35097995781505315,
      "learning_rate": 0.0001799221993434159,
      "loss": 0.7512,
      "step": 555
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.402025689464666,
      "learning_rate": 0.00017938074333710157,
      "loss": 0.7223,
      "step": 560
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.2848036827365996,
      "learning_rate": 0.00017883292094672128,
      "loss": 0.7156,
      "step": 565
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.3265548244089928,
      "learning_rate": 0.00017827877610796514,
      "loss": 0.7012,
      "step": 570
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.34382331081020817,
      "learning_rate": 0.00017771835326358743,
      "loss": 0.7054,
      "step": 575
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.24702163585830686,
      "learning_rate": 0.00017715169735984233,
      "loss": 0.6616,
      "step": 580
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.27778376017681095,
      "learning_rate": 0.0001765788538428792,
      "loss": 0.7787,
      "step": 585
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.38075564289830116,
      "learning_rate": 0.00017599986865509767,
      "loss": 0.7178,
      "step": 590
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.2704261360967656,
      "learning_rate": 0.00017541478823146327,
      "loss": 0.7506,
      "step": 595
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.29902224786080056,
      "learning_rate": 0.00017482365949578302,
      "loss": 0.7168,
      "step": 600
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3348731188034772,
      "learning_rate": 0.00017422652985694237,
      "loss": 0.7585,
      "step": 605
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.2985058741202778,
      "learning_rate": 0.00017362344720510278,
      "loss": 0.7011,
      "step": 610
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.31163972031224013,
      "learning_rate": 0.00017301445990786102,
      "loss": 0.7651,
      "step": 615
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.30230243003700485,
      "learning_rate": 0.00017239961680637,
      "loss": 0.7307,
      "step": 620
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.285713779713542,
      "learning_rate": 0.0001717789672114218,
      "loss": 0.8099,
      "step": 625
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.30987091480193457,
      "learning_rate": 0.0001711525608994927,
      "loss": 0.6981,
      "step": 630
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3621562136720511,
      "learning_rate": 0.00017052044810875126,
      "loss": 0.7656,
      "step": 635
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.35038216227613783,
      "learning_rate": 0.00016988267953502913,
      "loss": 0.7123,
      "step": 640
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.32472931615304945,
      "learning_rate": 0.00016923930632775516,
      "loss": 0.6981,
      "step": 645
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.35508029722752593,
      "learning_rate": 0.00016859038008585326,
      "loss": 0.7184,
      "step": 650
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3017654786452634,
      "learning_rate": 0.0001679359528536041,
      "loss": 0.6841,
      "step": 655
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3154279848347468,
      "learning_rate": 0.00016727607711647114,
      "loss": 0.7285,
      "step": 660
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3329741850112974,
      "learning_rate": 0.00016661080579689132,
      "loss": 0.7049,
      "step": 665
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3908307369485803,
      "learning_rate": 0.0001659401922500304,
      "loss": 0.713,
      "step": 670
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3207974085098528,
      "learning_rate": 0.00016526429025950424,
      "loss": 0.6761,
      "step": 675
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3293679898468649,
      "learning_rate": 0.00016458315403306502,
      "loss": 0.7624,
      "step": 680
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.32802837822586844,
      "learning_rate": 0.0001638968381982538,
      "loss": 0.7035,
      "step": 685
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.2918059849954773,
      "learning_rate": 0.0001632053977980194,
      "loss": 0.7579,
      "step": 690
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.42725236485175366,
      "learning_rate": 0.000162508888286304,
      "loss": 0.7311,
      "step": 695
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.29870401033734395,
      "learning_rate": 0.00016180736552359553,
      "loss": 0.7566,
      "step": 700
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3262514365628641,
      "learning_rate": 0.00016110088577244773,
      "loss": 0.7315,
      "step": 705
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3299951029943434,
      "learning_rate": 0.00016038950569296785,
      "loss": 0.7052,
      "step": 710
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3030534481188404,
      "learning_rate": 0.00015967328233827249,
      "loss": 0.7569,
      "step": 715
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3502796609404742,
      "learning_rate": 0.00015895227314991178,
      "loss": 0.6723,
      "step": 720
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3135691965969973,
      "learning_rate": 0.00015822653595326275,
      "loss": 0.7017,
      "step": 725
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3287968913852606,
      "learning_rate": 0.00015749612895289152,
      "loss": 0.7068,
      "step": 730
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.2937490795256513,
      "learning_rate": 0.00015676111072788527,
      "loss": 0.7017,
      "step": 735
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3516674743601905,
      "learning_rate": 0.00015602154022715435,
      "loss": 0.7325,
      "step": 740
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.33502124348161944,
      "learning_rate": 0.0001552774767647043,
      "loss": 0.694,
      "step": 745
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.37201228720271595,
      "learning_rate": 0.0001545289800148789,
      "loss": 0.7172,
      "step": 750
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3295048584114821,
      "learning_rate": 0.0001537761100075744,
      "loss": 0.7251,
      "step": 755
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.27702734134470053,
      "learning_rate": 0.00015301892712342482,
      "loss": 0.7649,
      "step": 760
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3539124471673018,
      "learning_rate": 0.00015225749208895968,
      "loss": 0.7572,
      "step": 765
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.30059385325607507,
      "learning_rate": 0.0001514918659717335,
      "loss": 0.7684,
      "step": 770
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.26506863063677677,
      "learning_rate": 0.00015072211017542813,
      "loss": 0.7333,
      "step": 775
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.2873361753567141,
      "learning_rate": 0.00014994828643492827,
      "loss": 0.7016,
      "step": 780
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.312388337814943,
      "learning_rate": 0.00014917045681137026,
      "loss": 0.717,
      "step": 785
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.30241864824677817,
      "learning_rate": 0.0001483886836871646,
      "loss": 0.6728,
      "step": 790
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3144095290225978,
      "learning_rate": 0.00014760302976099304,
      "loss": 0.7116,
      "step": 795
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3661687740278934,
      "learning_rate": 0.00014681355804278001,
      "loss": 0.706,
      "step": 800
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.26075524783800497,
      "learning_rate": 0.00014602033184863913,
      "loss": 0.7115,
      "step": 805
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3308386961158724,
      "learning_rate": 0.00014522341479579533,
      "loss": 0.7968,
      "step": 810
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.2603812719663815,
      "learning_rate": 0.00014442287079748263,
      "loss": 0.7291,
      "step": 815
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4338654790783011,
      "learning_rate": 0.00014361876405781832,
      "loss": 0.7073,
      "step": 820
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.38970393579567186,
      "learning_rate": 0.00014281115906665374,
      "loss": 0.7694,
      "step": 825
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3684501022609005,
      "learning_rate": 0.00014200012059440207,
      "loss": 0.7024,
      "step": 830
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.33557966231835323,
      "learning_rate": 0.00014118571368684383,
      "loss": 0.7781,
      "step": 835
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3074027678129374,
      "learning_rate": 0.00014036800365991008,
      "loss": 0.6559,
      "step": 840
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.36181309050726196,
      "learning_rate": 0.00013954705609444404,
      "loss": 0.7231,
      "step": 845
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3228631583048502,
      "learning_rate": 0.00013872293683094152,
      "loss": 0.7473,
      "step": 850
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.34101064635279904,
      "learning_rate": 0.00013789571196427055,
      "loss": 0.7207,
      "step": 855
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3863033671752472,
      "learning_rate": 0.00013706544783837022,
      "loss": 0.6917,
      "step": 860
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3322990594993244,
      "learning_rate": 0.00013623221104093025,
      "loss": 0.7813,
      "step": 865
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.2994766911720971,
      "learning_rate": 0.00013539606839805036,
      "loss": 0.7211,
      "step": 870
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.359729930933566,
      "learning_rate": 0.00013455708696888085,
      "loss": 0.7332,
      "step": 875
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.2846394069315403,
      "learning_rate": 0.00013371533404024438,
      "loss": 0.7424,
      "step": 880
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.31303595211825364,
      "learning_rate": 0.00013287087712123962,
      "loss": 0.7539,
      "step": 885
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3394022151204218,
      "learning_rate": 0.00013202378393782692,
      "loss": 0.7176,
      "step": 890
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.45987531025131423,
      "learning_rate": 0.00013117412242739655,
      "loss": 0.6813,
      "step": 895
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.32789566992648417,
      "learning_rate": 0.00013032196073332027,
      "loss": 0.7617,
      "step": 900
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.31944612772807995,
      "learning_rate": 0.00012946736719948607,
      "loss": 0.7283,
      "step": 905
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.2682045705602512,
      "learning_rate": 0.000128610410364817,
      "loss": 0.6878,
      "step": 910
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.32648100192685736,
      "learning_rate": 0.00012775115895777417,
      "loss": 0.7074,
      "step": 915
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3990622736391209,
      "learning_rate": 0.00012688968189084493,
      "loss": 0.6879,
      "step": 920
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3269394019516448,
      "learning_rate": 0.00012602604825501587,
      "loss": 0.7281,
      "step": 925
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2775723422331167,
      "learning_rate": 0.00012516032731423165,
      "loss": 0.7146,
      "step": 930
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2899589945929131,
      "learning_rate": 0.00012429258849984014,
      "loss": 0.7314,
      "step": 935
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.34015486313072996,
      "learning_rate": 0.00012342290140502388,
      "loss": 0.7329,
      "step": 940
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3154317960889019,
      "learning_rate": 0.00012255133577921868,
      "loss": 0.7303,
      "step": 945
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.36345240707288057,
      "learning_rate": 0.0001216779615225197,
      "loss": 0.6922,
      "step": 950
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.32115987953591585,
      "learning_rate": 0.00012080284868007541,
      "loss": 0.7131,
      "step": 955
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.315835892374506,
      "learning_rate": 0.0001199260674364699,
      "loss": 0.7479,
      "step": 960
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.29547841451849666,
      "learning_rate": 0.00011904768811009405,
      "loss": 0.7544,
      "step": 965
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.24461772178347557,
      "learning_rate": 0.00011816778114750593,
      "loss": 0.7273,
      "step": 970
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.27250938619200016,
      "learning_rate": 0.00011728641711778103,
      "loss": 0.7488,
      "step": 975
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.28021149787423305,
      "learning_rate": 0.00011640366670685248,
      "loss": 0.6688,
      "step": 980
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2963856674451753,
      "learning_rate": 0.00011551960071184195,
      "loss": 0.8196,
      "step": 985
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.42323936972543524,
      "learning_rate": 0.00011463429003538196,
      "loss": 0.7435,
      "step": 990
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2807530558829986,
      "learning_rate": 0.000113747805679929,
      "loss": 0.7579,
      "step": 995
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3763766165258716,
      "learning_rate": 0.00011286021874206952,
      "loss": 0.6704,
      "step": 1000
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.3472841951791292,
      "learning_rate": 0.00011197160040681762,
      "loss": 0.8043,
      "step": 1005
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.27899036039154235,
      "learning_rate": 0.0001110820219419062,
      "loss": 0.7342,
      "step": 1010
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.2890345473794268,
      "learning_rate": 0.0001101915546920711,
      "loss": 0.7264,
      "step": 1015
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.2845198372692673,
      "learning_rate": 0.00010930027007332923,
      "loss": 0.672,
      "step": 1020
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.30480506610903685,
      "learning_rate": 0.00010840823956725103,
      "loss": 0.7273,
      "step": 1025
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3008478217545316,
      "learning_rate": 0.00010751553471522757,
      "loss": 0.7308,
      "step": 1030
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.2964651215060616,
      "learning_rate": 0.00010662222711273279,
      "loss": 0.6941,
      "step": 1035
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.2922042194314711,
      "learning_rate": 0.00010572838840358168,
      "loss": 0.6713,
      "step": 1040
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3154946482457206,
      "learning_rate": 0.00010483409027418425,
      "loss": 0.6925,
      "step": 1045
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.31511855490630747,
      "learning_rate": 0.00010393940444779635,
      "loss": 0.6714,
      "step": 1050
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.319949419633222,
      "learning_rate": 0.00010304440267876727,
      "loss": 0.7685,
      "step": 1055
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.33340619519598513,
      "learning_rate": 0.00010214915674678523,
      "loss": 0.7769,
      "step": 1060
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.32834216382858106,
      "learning_rate": 0.00010125373845112034,
      "loss": 0.7627,
      "step": 1065
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.35298554027692325,
      "learning_rate": 0.00010035821960486643,
      "loss": 0.7382,
      "step": 1070
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.33667375260779,
      "learning_rate": 9.946267202918157e-05,
      "loss": 0.6894,
      "step": 1075
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.40614343620033194,
      "learning_rate": 9.856716754752796e-05,
      "loss": 0.691,
      "step": 1080
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3337962004070806,
      "learning_rate": 9.767177797991155e-05,
      "loss": 0.7398,
      "step": 1085
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.27396845307294787,
      "learning_rate": 9.677657513712221e-05,
      "loss": 0.7143,
      "step": 1090
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.27225179948189643,
      "learning_rate": 9.588163081497427e-05,
      "loss": 0.7628,
      "step": 1095
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.31897496500357725,
      "learning_rate": 9.498701678854865e-05,
      "loss": 0.7231,
      "step": 1100
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3680616601189938,
      "learning_rate": 9.409280480643628e-05,
      "loss": 0.7625,
      "step": 1105
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.32776900421268723,
      "learning_rate": 9.319906658498389e-05,
      "loss": 0.7422,
      "step": 1110
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3229530750140877,
      "learning_rate": 9.230587380254237e-05,
      "loss": 0.7185,
      "step": 1115
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.26891277818159653,
      "learning_rate": 9.141329809371803e-05,
      "loss": 0.7289,
      "step": 1120
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.28191964970264505,
      "learning_rate": 9.052141104362748e-05,
      "loss": 0.6875,
      "step": 1125
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3043145079069415,
      "learning_rate": 8.963028418215653e-05,
      "loss": 0.7929,
      "step": 1130
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.2991598027119205,
      "learning_rate": 8.873998897822336e-05,
      "loss": 0.7775,
      "step": 1135
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.30191353240259794,
      "learning_rate": 8.785059683404672e-05,
      "loss": 0.6925,
      "step": 1140
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.32998588702317205,
      "learning_rate": 8.696217907941941e-05,
      "loss": 0.7445,
      "step": 1145
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3529251348165701,
      "learning_rate": 8.607480696598762e-05,
      "loss": 0.6603,
      "step": 1150
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3381612883981257,
      "learning_rate": 8.518855166153644e-05,
      "loss": 0.751,
      "step": 1155
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.31254901088918846,
      "learning_rate": 8.43034842442822e-05,
      "loss": 0.7088,
      "step": 1160
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3148505223711432,
      "learning_rate": 8.341967569717202e-05,
      "loss": 0.6678,
      "step": 1165
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3334869057549173,
      "learning_rate": 8.253719690219079e-05,
      "loss": 0.6848,
      "step": 1170
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3511805071017855,
      "learning_rate": 8.165611863467644e-05,
      "loss": 0.7051,
      "step": 1175
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.30055035546382686,
      "learning_rate": 8.077651155764387e-05,
      "loss": 0.6853,
      "step": 1180
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3152611255597378,
      "learning_rate": 7.98984462161175e-05,
      "loss": 0.7435,
      "step": 1185
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.27189411488369086,
      "learning_rate": 7.902199303147363e-05,
      "loss": 0.7363,
      "step": 1190
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3056569152174516,
      "learning_rate": 7.814722229579264e-05,
      "loss": 0.7381,
      "step": 1195
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.32534449161129786,
      "learning_rate": 7.727420416622144e-05,
      "loss": 0.7398,
      "step": 1200
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.2897183016884322,
      "learning_rate": 7.640300865934687e-05,
      "loss": 0.6671,
      "step": 1205
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.33996599065200905,
      "learning_rate": 7.553370564558032e-05,
      "loss": 0.7308,
      "step": 1210
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.2642886492574283,
      "learning_rate": 7.46663648435541e-05,
      "loss": 0.7159,
      "step": 1215
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.39197107543423354,
      "learning_rate": 7.380105581452987e-05,
      "loss": 0.7021,
      "step": 1220
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.32985129368047594,
      "learning_rate": 7.293784795681994e-05,
      "loss": 0.7636,
      "step": 1225
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.34382061739176006,
      "learning_rate": 7.207681050022132e-05,
      "loss": 0.7011,
      "step": 1230
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.31403686439018674,
      "learning_rate": 7.121801250046363e-05,
      "loss": 0.8093,
      "step": 1235
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.27784647722145367,
      "learning_rate": 7.036152283367056e-05,
      "loss": 0.6889,
      "step": 1240
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3119143232239076,
      "learning_rate": 6.950741019083617e-05,
      "loss": 0.7219,
      "step": 1245
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.30282200792727687,
      "learning_rate": 6.865574307231575e-05,
      "loss": 0.7057,
      "step": 1250
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.35743126694258187,
      "learning_rate": 6.780658978233199e-05,
      "loss": 0.7106,
      "step": 1255
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3009152875810381,
      "learning_rate": 6.696001842349702e-05,
      "loss": 0.7236,
      "step": 1260
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3147441789350989,
      "learning_rate": 6.611609689135056e-05,
      "loss": 0.7807,
      "step": 1265
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.32658208617631634,
      "learning_rate": 6.527489286891459e-05,
      "loss": 0.6189,
      "step": 1270
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.32453208450466825,
      "learning_rate": 6.443647382126509e-05,
      "loss": 0.7265,
      "step": 1275
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3033249165003457,
      "learning_rate": 6.360090699012145e-05,
      "loss": 0.7151,
      "step": 1280
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.32043745847027666,
      "learning_rate": 6.27682593884535e-05,
      "loss": 0.6417,
      "step": 1285
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.29733048444645577,
      "learning_rate": 6.193859779510712e-05,
      "loss": 0.7391,
      "step": 1290
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.2832352164219465,
      "learning_rate": 6.111198874944845e-05,
      "loss": 0.7132,
      "step": 1295
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3327901932989352,
      "learning_rate": 6.0288498546027536e-05,
      "loss": 0.7687,
      "step": 1300
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3450811991465242,
      "learning_rate": 5.946819322926127e-05,
      "loss": 0.6737,
      "step": 1305
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.31294833083964624,
      "learning_rate": 5.865113858813673e-05,
      "loss": 0.7173,
      "step": 1310
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.29835317503949266,
      "learning_rate": 5.783740015093484e-05,
      "loss": 0.7228,
      "step": 1315
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.30012406084966387,
      "learning_rate": 5.702704317997492e-05,
      "loss": 0.685,
      "step": 1320
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.315312458494754,
      "learning_rate": 5.6220132666380635e-05,
      "loss": 0.6659,
      "step": 1325
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3502893529406923,
      "learning_rate": 5.541673332486773e-05,
      "loss": 0.7534,
      "step": 1330
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.2866563574117178,
      "learning_rate": 5.4616909588553674e-05,
      "loss": 0.8126,
      "step": 1335
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.2885727570352686,
      "learning_rate": 5.3820725603790346e-05,
      "loss": 0.7785,
      "step": 1340
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.2979650510099925,
      "learning_rate": 5.30282452250193e-05,
      "loss": 0.7762,
      "step": 1345
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.32229730103844956,
      "learning_rate": 5.223953200965055e-05,
      "loss": 0.7303,
      "step": 1350
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3126960391124247,
      "learning_rate": 5.145464921296537e-05,
      "loss": 0.7962,
      "step": 1355
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.34544789759267497,
      "learning_rate": 5.067365978304315e-05,
      "loss": 0.6871,
      "step": 1360
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3339123151664445,
      "learning_rate": 4.9896626355712805e-05,
      "loss": 0.6939,
      "step": 1365
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.2729865167396553,
      "learning_rate": 4.912361124952948e-05,
      "loss": 0.7153,
      "step": 1370
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.2966987002321613,
      "learning_rate": 4.835467646077656e-05,
      "loss": 0.6814,
      "step": 1375
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.27796921107480527,
      "learning_rate": 4.7589883658493296e-05,
      "loss": 0.6917,
      "step": 1380
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.33415061896123593,
      "learning_rate": 4.682929417952939e-05,
      "loss": 0.6879,
      "step": 1385
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.2657599274207207,
      "learning_rate": 4.6072969023625165e-05,
      "loss": 0.7084,
      "step": 1390
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.2963301427902277,
      "learning_rate": 4.532096884851978e-05,
      "loss": 0.7192,
      "step": 1395
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.34405748625187765,
      "learning_rate": 4.457335396508631e-05,
      "loss": 0.7191,
      "step": 1400
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.2836783169964031,
      "learning_rate": 4.383018433249464e-05,
      "loss": 0.7087,
      "step": 1405
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.31448928123984227,
      "learning_rate": 4.309151955340297e-05,
      "loss": 0.6926,
      "step": 1410
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2868854866352251,
      "learning_rate": 4.2357418869177354e-05,
      "loss": 0.7553,
      "step": 1415
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3373951207672345,
      "learning_rate": 4.162794115514078e-05,
      "loss": 0.7637,
      "step": 1420
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3238763492542821,
      "learning_rate": 4.0903144915851174e-05,
      "loss": 0.6468,
      "step": 1425
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2967660421217855,
      "learning_rate": 4.018308828040924e-05,
      "loss": 0.6687,
      "step": 1430
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.2664973906610404,
      "learning_rate": 3.946782899779667e-05,
      "loss": 0.7033,
      "step": 1435
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3344040629264455,
      "learning_rate": 3.875742443224451e-05,
      "loss": 0.7207,
      "step": 1440
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.28981326938339314,
      "learning_rate": 3.805193155863247e-05,
      "loss": 0.6746,
      "step": 1445
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.29792212124490053,
      "learning_rate": 3.7351406957919636e-05,
      "loss": 0.729,
      "step": 1450
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.322484422395136,
      "learning_rate": 3.665590681260658e-05,
      "loss": 0.7185,
      "step": 1455
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.30050768862622057,
      "learning_rate": 3.59654869022294e-05,
      "loss": 0.7665,
      "step": 1460
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.25285793655912636,
      "learning_rate": 3.5280202598886324e-05,
      "loss": 0.7648,
      "step": 1465
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3141160494292346,
      "learning_rate": 3.4600108862796796e-05,
      "loss": 0.7147,
      "step": 1470
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.30470187329804443,
      "learning_rate": 3.392526023789349e-05,
      "loss": 0.7388,
      "step": 1475
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.24228009923888186,
      "learning_rate": 3.325571084744803e-05,
      "loss": 0.6818,
      "step": 1480
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3368859264963583,
      "learning_rate": 3.259151438973024e-05,
      "loss": 0.7783,
      "step": 1485
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2697250268367978,
      "learning_rate": 3.1932724133701344e-05,
      "loss": 0.6815,
      "step": 1490
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4784067864730423,
      "learning_rate": 3.1279392914742046e-05,
      "loss": 0.7335,
      "step": 1495
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.29605034730472274,
      "learning_rate": 3.06315731304148e-05,
      "loss": 0.7612,
      "step": 1500
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.2757121861782341,
      "learning_rate": 2.998931673626175e-05,
      "loss": 0.6684,
      "step": 1505
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3083842334352063,
      "learning_rate": 2.935267524163774e-05,
      "loss": 0.7583,
      "step": 1510
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.3031689388678502,
      "learning_rate": 2.872169970557913e-05,
      "loss": 0.7543,
      "step": 1515
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.32727819432412475,
      "learning_rate": 2.8096440732709083e-05,
      "loss": 0.7734,
      "step": 1520
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.34377977870392656,
      "learning_rate": 2.7476948469178887e-05,
      "loss": 0.7244,
      "step": 1525
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.45571873834918486,
      "learning_rate": 2.6863272598646106e-05,
      "loss": 0.712,
      "step": 1530
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3036717760073397,
      "learning_rate": 2.625546233829016e-05,
      "loss": 0.7521,
      "step": 1535
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.30159506875916664,
      "learning_rate": 2.5653566434864928e-05,
      "loss": 0.7285,
      "step": 1540
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.345473195665642,
      "learning_rate": 2.5057633160789184e-05,
      "loss": 0.7181,
      "step": 1545
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2827667094521633,
      "learning_rate": 2.446771031027527e-05,
      "loss": 0.7738,
      "step": 1550
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.314078179653926,
      "learning_rate": 2.3883845195495878e-05,
      "loss": 0.7834,
      "step": 1555
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2955404908029541,
      "learning_rate": 2.330608464278953e-05,
      "loss": 0.735,
      "step": 1560
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2939404179942727,
      "learning_rate": 2.273447498890521e-05,
      "loss": 0.779,
      "step": 1565
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.2809810980876618,
      "learning_rate": 2.2169062077286075e-05,
      "loss": 0.6827,
      "step": 1570
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.28200573272884943,
      "learning_rate": 2.1609891254392678e-05,
      "loss": 0.6767,
      "step": 1575
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.35315193902923053,
      "learning_rate": 2.1057007366066373e-05,
      "loss": 0.7594,
      "step": 1580
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.3420677513120873,
      "learning_rate": 2.0510454753932395e-05,
      "loss": 0.681,
      "step": 1585
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.2618256294434834,
      "learning_rate": 1.9970277251843862e-05,
      "loss": 0.713,
      "step": 1590
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.28782137253735823,
      "learning_rate": 1.9436518182366158e-05,
      "loss": 0.7238,
      "step": 1595
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.27412934807247086,
      "learning_rate": 1.8909220353302392e-05,
      "loss": 0.7298,
      "step": 1600
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3125955140731189,
      "learning_rate": 1.838842605426031e-05,
      "loss": 0.6707,
      "step": 1605
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.26822454387889916,
      "learning_rate": 1.7874177053260598e-05,
      "loss": 0.6648,
      "step": 1610
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4034842521444172,
      "learning_rate": 1.736651459338695e-05,
      "loss": 0.6764,
      "step": 1615
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.30150614696267586,
      "learning_rate": 1.6865479389478545e-05,
      "loss": 0.7237,
      "step": 1620
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.2659911283631938,
      "learning_rate": 1.6371111624864543e-05,
      "loss": 0.7153,
      "step": 1625
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3653406248899525,
      "learning_rate": 1.5883450948141377e-05,
      "loss": 0.7007,
      "step": 1630
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.29005529325753615,
      "learning_rate": 1.540253646999299e-05,
      "loss": 0.742,
      "step": 1635
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.31746439087800815,
      "learning_rate": 1.4928406760054059e-05,
      "loss": 0.7122,
      "step": 1640
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3081458592929589,
      "learning_rate": 1.4461099843816684e-05,
      "loss": 0.7,
      "step": 1645
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.25746343675572125,
      "learning_rate": 1.4000653199580782e-05,
      "loss": 0.7595,
      "step": 1650
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.2928014845363881,
      "learning_rate": 1.3547103755448287e-05,
      "loss": 0.7417,
      "step": 1655
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.26724061619267486,
      "learning_rate": 1.3100487886361379e-05,
      "loss": 0.775,
      "step": 1660
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.2537157052459514,
      "learning_rate": 1.266084141118542e-05,
      "loss": 0.7076,
      "step": 1665
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.30196857922544756,
      "learning_rate": 1.2228199589835999e-05,
      "loss": 0.7321,
      "step": 1670
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.32518134307597746,
      "learning_rate": 1.1802597120451286e-05,
      "loss": 0.7296,
      "step": 1675
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3444957897997851,
      "learning_rate": 1.1384068136609105e-05,
      "loss": 0.7512,
      "step": 1680
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.28186990977804877,
      "learning_rate": 1.0972646204589377e-05,
      "loss": 0.6829,
      "step": 1685
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.2877542591637029,
      "learning_rate": 1.0568364320682178e-05,
      "loss": 0.752,
      "step": 1690
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.3068050447445267,
      "learning_rate": 1.0171254908541372e-05,
      "loss": 0.6702,
      "step": 1695
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.27396201685633986,
      "learning_rate": 9.781349816584162e-06,
      "loss": 0.7043,
      "step": 1700
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.2829868027709237,
      "learning_rate": 9.398680315436903e-06,
      "loss": 0.7255,
      "step": 1705
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.32779056821655844,
      "learning_rate": 9.023277095427173e-06,
      "loss": 0.6768,
      "step": 1710
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.2989710892101726,
      "learning_rate": 8.655170264122303e-06,
      "loss": 0.7468,
      "step": 1715
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.2796957331135978,
      "learning_rate": 8.294389343914899e-06,
      "loss": 0.7217,
      "step": 1720
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.27621286820275476,
      "learning_rate": 7.940963269654922e-06,
      "loss": 0.7626,
      "step": 1725
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.2742060936706595,
      "learning_rate": 7.594920386329252e-06,
      "loss": 0.7164,
      "step": 1730
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.2921152376136818,
      "learning_rate": 7.256288446788362e-06,
      "loss": 0.6612,
      "step": 1735
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.33165308692413886,
      "learning_rate": 6.925094609520455e-06,
      "loss": 0.7388,
      "step": 1740
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.27930971680755107,
      "learning_rate": 6.601365436473439e-06,
      "loss": 0.7051,
      "step": 1745
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.3194120865785679,
      "learning_rate": 6.2851268909245865e-06,
      "loss": 0.6962,
      "step": 1750
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.29609984935200595,
      "learning_rate": 5.976404335398256e-06,
      "loss": 0.7428,
      "step": 1755
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.23923907800913505,
      "learning_rate": 5.675222529631841e-06,
      "loss": 0.7178,
      "step": 1760
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.27969743641481126,
      "learning_rate": 5.381605628590003e-06,
      "loss": 0.7283,
      "step": 1765
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.29880789893065274,
      "learning_rate": 5.095577180527378e-06,
      "loss": 0.7332,
      "step": 1770
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.3008466962712365,
      "learning_rate": 4.817160125100106e-06,
      "loss": 0.7748,
      "step": 1775
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.30259996008890583,
      "learning_rate": 4.546376791525975e-06,
      "loss": 0.7358,
      "step": 1780
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.2672834495144112,
      "learning_rate": 4.2832488967935795e-06,
      "loss": 0.7307,
      "step": 1785
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.28548529856286414,
      "learning_rate": 4.02779754392072e-06,
      "loss": 0.7199,
      "step": 1790
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.3415043081858787,
      "learning_rate": 3.780043220261764e-06,
      "loss": 0.742,
      "step": 1795
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.3764610960034005,
      "learning_rate": 3.540005795864709e-06,
      "loss": 0.661,
      "step": 1800
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.3181188064604727,
      "learning_rate": 3.3077045218775192e-06,
      "loss": 0.7495,
      "step": 1805
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.3395535338338037,
      "learning_rate": 3.0831580290041184e-06,
      "loss": 0.7544,
      "step": 1810
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.24570446974307603,
      "learning_rate": 2.8663843260103074e-06,
      "loss": 0.7318,
      "step": 1815
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.3575242711795699,
      "learning_rate": 2.6574007982793857e-06,
      "loss": 0.7146,
      "step": 1820
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.3082511563069773,
      "learning_rate": 2.456224206417812e-06,
      "loss": 0.6818,
      "step": 1825
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.32198732939485364,
      "learning_rate": 2.262870684911045e-06,
      "loss": 0.7466,
      "step": 1830
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.2784226133533578,
      "learning_rate": 2.0773557408295343e-06,
      "loss": 0.7111,
      "step": 1835
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.28054093291710575,
      "learning_rate": 1.8996942525850047e-06,
      "loss": 0.7029,
      "step": 1840
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.2785851691775617,
      "learning_rate": 1.7299004687372665e-06,
      "loss": 0.6944,
      "step": 1845
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.30742157292880246,
      "learning_rate": 1.5679880068514174e-06,
      "loss": 0.6809,
      "step": 1850
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.33283084468222157,
      "learning_rate": 1.4139698524057165e-06,
      "loss": 0.7465,
      "step": 1855
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.31480469510269476,
      "learning_rate": 1.2678583577501624e-06,
      "loss": 0.7278,
      "step": 1860
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.3219911221559245,
      "learning_rate": 1.1296652411158182e-06,
      "loss": 0.7241,
      "step": 1865
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2799499517285736,
      "learning_rate": 9.994015856749527e-07,
      "loss": 0.6885,
      "step": 1870
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2786247050141377,
      "learning_rate": 8.770778386522627e-07,
      "loss": 0.6693,
      "step": 1875
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.31648747823824064,
      "learning_rate": 7.627038104869199e-07,
      "loss": 0.7239,
      "step": 1880
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.29448005284725853,
      "learning_rate": 6.562886740457797e-07,
      "loss": 0.7175,
      "step": 1885
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.28834475252273023,
      "learning_rate": 5.578409638877457e-07,
      "loss": 0.6461,
      "step": 1890
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.3253714736777368,
      "learning_rate": 4.6736857557925227e-07,
      "loss": 0.7028,
      "step": 1895
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.28770138704276665,
      "learning_rate": 3.8487876506106966e-07,
      "loss": 0.7843,
      "step": 1900
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.32499191549488704,
      "learning_rate": 3.1037814806634815e-07,
      "loss": 0.692,
      "step": 1905
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.2753241236710541,
      "learning_rate": 2.43872699590042e-07,
      "loss": 0.6691,
      "step": 1910
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.2733251707609935,
      "learning_rate": 1.8536775340970425e-07,
      "loss": 0.7697,
      "step": 1915
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.396509374591397,
      "learning_rate": 1.348680016577397e-07,
      "loss": 0.7544,
      "step": 1920
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.3059532145113707,
      "learning_rate": 9.237749444505062e-08,
      "loss": 0.7164,
      "step": 1925
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.2874674894374334,
      "learning_rate": 5.7899639536251883e-08,
      "loss": 0.7214,
      "step": 1930
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.29188035610372354,
      "learning_rate": 3.143720207635648e-08,
      "loss": 0.7303,
      "step": 1935
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.351168193020288,
      "learning_rate": 1.299230436898613e-08,
      "loss": 0.7646,
      "step": 1940
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.30619779353074095,
      "learning_rate": 2.566425706218567e-09,
      "loss": 0.6817,
      "step": 1945
    },
    {
      "epoch": 1.0,
      "eval_loss": NaN,
      "eval_runtime": 1609.9449,
      "eval_samples_per_second": 4.306,
      "eval_steps_per_second": 1.076,
      "step": 1949
    },
    {
      "epoch": 1.0,
      "step": 1949,
      "total_flos": 1.2135540828143616e+16,
      "train_loss": 0.7257996720494583,
      "train_runtime": 19082.4327,
      "train_samples_per_second": 3.268,
      "train_steps_per_second": 0.102
    }
  ],
  "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
}