{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9997069597069597,
  "eval_steps": 100,
  "global_step": 1706,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005860805860805861,
      "grad_norm": 2.521568536758423,
      "learning_rate": 0.00019941383352872216,
      "loss": 2.3132,
      "step": 5
    },
    {
      "epoch": 0.011721611721611722,
      "grad_norm": 2.104935884475708,
      "learning_rate": 0.00019882766705744433,
      "loss": 0.9444,
      "step": 10
    },
    {
      "epoch": 0.017582417582417582,
      "grad_norm": 1.2950881719589233,
      "learning_rate": 0.00019824150058616647,
      "loss": 0.357,
      "step": 15
    },
    {
      "epoch": 0.023443223443223443,
      "grad_norm": 1.1862170696258545,
      "learning_rate": 0.00019765533411488865,
      "loss": 0.2105,
      "step": 20
    },
    {
      "epoch": 0.029304029304029304,
      "grad_norm": 0.5247148871421814,
      "learning_rate": 0.0001970691676436108,
      "loss": 0.1087,
      "step": 25
    },
    {
      "epoch": 0.035164835164835165,
      "grad_norm": 1.2535285949707031,
      "learning_rate": 0.00019648300117233296,
      "loss": 0.1185,
      "step": 30
    },
    {
      "epoch": 0.041025641025641026,
      "grad_norm": 0.5407606959342957,
      "learning_rate": 0.0001958968347010551,
      "loss": 0.0775,
      "step": 35
    },
    {
      "epoch": 0.046886446886446886,
      "grad_norm": 1.137266993522644,
      "learning_rate": 0.00019531066822977726,
      "loss": 0.0728,
      "step": 40
    },
    {
      "epoch": 0.05274725274725275,
      "grad_norm": 0.5244052410125732,
      "learning_rate": 0.00019472450175849943,
      "loss": 0.056,
      "step": 45
    },
    {
      "epoch": 0.05860805860805861,
      "grad_norm": 0.4511496126651764,
      "learning_rate": 0.00019413833528722157,
      "loss": 0.0648,
      "step": 50
    },
    {
      "epoch": 0.06446886446886448,
      "grad_norm": 0.33913975954055786,
      "learning_rate": 0.00019355216881594375,
      "loss": 0.0479,
      "step": 55
    },
    {
      "epoch": 0.07032967032967033,
      "grad_norm": 0.354777991771698,
      "learning_rate": 0.0001929660023446659,
      "loss": 0.0758,
      "step": 60
    },
    {
      "epoch": 0.0761904761904762,
      "grad_norm": 0.3968910276889801,
      "learning_rate": 0.00019237983587338807,
      "loss": 0.0776,
      "step": 65
    },
    {
      "epoch": 0.08205128205128205,
      "grad_norm": 0.5061652660369873,
      "learning_rate": 0.0001917936694021102,
      "loss": 0.0349,
      "step": 70
    },
    {
      "epoch": 0.08791208791208792,
      "grad_norm": 0.24455250799655914,
      "learning_rate": 0.00019120750293083236,
      "loss": 0.0369,
      "step": 75
    },
    {
      "epoch": 0.09377289377289377,
      "grad_norm": 0.3959537446498871,
      "learning_rate": 0.00019062133645955453,
      "loss": 0.0513,
      "step": 80
    },
    {
      "epoch": 0.09963369963369964,
      "grad_norm": 0.4534469544887543,
      "learning_rate": 0.00019003516998827668,
      "loss": 0.0459,
      "step": 85
    },
    {
      "epoch": 0.1054945054945055,
      "grad_norm": 0.30694451928138733,
      "learning_rate": 0.00018944900351699885,
      "loss": 0.0377,
      "step": 90
    },
    {
      "epoch": 0.11135531135531136,
      "grad_norm": 0.15078052878379822,
      "learning_rate": 0.000188862837045721,
      "loss": 0.0297,
      "step": 95
    },
    {
      "epoch": 0.11721611721611722,
      "grad_norm": 0.3153330981731415,
      "learning_rate": 0.00018827667057444317,
      "loss": 0.0301,
      "step": 100
    },
    {
      "epoch": 0.11721611721611722,
      "eval_loss": 0.028654273599386215,
      "eval_runtime": 4.8016,
      "eval_samples_per_second": 5.831,
      "eval_steps_per_second": 2.916,
      "step": 100
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": 0.5777396559715271,
      "learning_rate": 0.0001876905041031653,
      "loss": 0.0515,
      "step": 105
    },
    {
      "epoch": 0.12893772893772895,
      "grad_norm": 0.3677718937397003,
      "learning_rate": 0.00018710433763188746,
      "loss": 0.0315,
      "step": 110
    },
    {
      "epoch": 0.1347985347985348,
      "grad_norm": 0.30658507347106934,
      "learning_rate": 0.0001865181711606096,
      "loss": 0.0248,
      "step": 115
    },
    {
      "epoch": 0.14065934065934066,
      "grad_norm": 0.21649648249149323,
      "learning_rate": 0.00018593200468933178,
      "loss": 0.0352,
      "step": 120
    },
    {
      "epoch": 0.14652014652014653,
      "grad_norm": 0.2877885103225708,
      "learning_rate": 0.00018534583821805395,
      "loss": 0.0456,
      "step": 125
    },
    {
      "epoch": 0.1523809523809524,
      "grad_norm": 0.1782904863357544,
      "learning_rate": 0.0001847596717467761,
      "loss": 0.0257,
      "step": 130
    },
    {
      "epoch": 0.15824175824175823,
      "grad_norm": 0.1211300641298294,
      "learning_rate": 0.00018417350527549827,
      "loss": 0.0437,
      "step": 135
    },
    {
      "epoch": 0.1641025641025641,
      "grad_norm": 0.298053503036499,
      "learning_rate": 0.0001835873388042204,
      "loss": 0.036,
      "step": 140
    },
    {
      "epoch": 0.16996336996336997,
      "grad_norm": 0.2280658483505249,
      "learning_rate": 0.00018300117233294256,
      "loss": 0.0331,
      "step": 145
    },
    {
      "epoch": 0.17582417582417584,
      "grad_norm": 0.14896267652511597,
      "learning_rate": 0.0001824150058616647,
      "loss": 0.0239,
      "step": 150
    },
    {
      "epoch": 0.18168498168498168,
      "grad_norm": 0.1769961267709732,
      "learning_rate": 0.00018182883939038688,
      "loss": 0.0375,
      "step": 155
    },
    {
      "epoch": 0.18754578754578755,
      "grad_norm": 0.19849297404289246,
      "learning_rate": 0.00018124267291910902,
      "loss": 0.0357,
      "step": 160
    },
    {
      "epoch": 0.1934065934065934,
      "grad_norm": 0.2294420450925827,
      "learning_rate": 0.0001806565064478312,
      "loss": 0.0504,
      "step": 165
    },
    {
      "epoch": 0.19926739926739928,
      "grad_norm": 0.10744224488735199,
      "learning_rate": 0.00018007033997655337,
      "loss": 0.0209,
      "step": 170
    },
    {
      "epoch": 0.20512820512820512,
      "grad_norm": 0.06066066771745682,
      "learning_rate": 0.00017948417350527551,
      "loss": 0.0175,
      "step": 175
    },
    {
      "epoch": 0.210989010989011,
      "grad_norm": 0.5421485304832458,
      "learning_rate": 0.00017889800703399766,
      "loss": 0.0398,
      "step": 180
    },
    {
      "epoch": 0.21684981684981686,
      "grad_norm": 0.1306767612695694,
      "learning_rate": 0.0001783118405627198,
      "loss": 0.0258,
      "step": 185
    },
    {
      "epoch": 0.22271062271062272,
      "grad_norm": 0.11510124802589417,
      "learning_rate": 0.00017772567409144198,
      "loss": 0.0253,
      "step": 190
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 0.20577751100063324,
      "learning_rate": 0.00017713950762016412,
      "loss": 0.0277,
      "step": 195
    },
    {
      "epoch": 0.23443223443223443,
      "grad_norm": 0.20588932931423187,
      "learning_rate": 0.0001765533411488863,
      "loss": 0.0296,
      "step": 200
    },
    {
      "epoch": 0.23443223443223443,
      "eval_loss": 0.019306689500808716,
      "eval_runtime": 4.7344,
      "eval_samples_per_second": 5.914,
      "eval_steps_per_second": 2.957,
      "step": 200
    },
    {
      "epoch": 0.2402930402930403,
      "grad_norm": 0.13713975250720978,
      "learning_rate": 0.00017596717467760847,
      "loss": 0.0372,
      "step": 205
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 0.14788508415222168,
      "learning_rate": 0.00017538100820633061,
      "loss": 0.033,
      "step": 210
    },
    {
      "epoch": 0.252014652014652,
      "grad_norm": 0.2372630536556244,
      "learning_rate": 0.00017479484173505276,
      "loss": 0.0281,
      "step": 215
    },
    {
      "epoch": 0.2578754578754579,
      "grad_norm": 0.2357954978942871,
      "learning_rate": 0.0001742086752637749,
      "loss": 0.0295,
      "step": 220
    },
    {
      "epoch": 0.26373626373626374,
      "grad_norm": 0.2537606358528137,
      "learning_rate": 0.00017362250879249708,
      "loss": 0.036,
      "step": 225
    },
    {
      "epoch": 0.2695970695970696,
      "grad_norm": 0.2222289741039276,
      "learning_rate": 0.00017303634232121922,
      "loss": 0.0402,
      "step": 230
    },
    {
      "epoch": 0.2754578754578755,
      "grad_norm": 0.19696177542209625,
      "learning_rate": 0.0001724501758499414,
      "loss": 0.025,
      "step": 235
    },
    {
      "epoch": 0.2813186813186813,
      "grad_norm": 0.08915109932422638,
      "learning_rate": 0.00017186400937866357,
      "loss": 0.0292,
      "step": 240
    },
    {
      "epoch": 0.28717948717948716,
      "grad_norm": 0.39625948667526245,
      "learning_rate": 0.00017127784290738572,
      "loss": 0.0324,
      "step": 245
    },
    {
      "epoch": 0.29304029304029305,
      "grad_norm": 0.05654177442193031,
      "learning_rate": 0.00017069167643610786,
      "loss": 0.0384,
      "step": 250
    },
    {
      "epoch": 0.2989010989010989,
      "grad_norm": 0.23707075417041779,
      "learning_rate": 0.00017010550996483,
      "loss": 0.0311,
      "step": 255
    },
    {
      "epoch": 0.3047619047619048,
      "grad_norm": 0.2619571387767792,
      "learning_rate": 0.00016951934349355218,
      "loss": 0.0275,
      "step": 260
    },
    {
      "epoch": 0.31062271062271063,
      "grad_norm": 0.11028550565242767,
      "learning_rate": 0.00016893317702227432,
      "loss": 0.0194,
      "step": 265
    },
    {
      "epoch": 0.31648351648351647,
      "grad_norm": 0.26710912585258484,
      "learning_rate": 0.0001683470105509965,
      "loss": 0.0318,
      "step": 270
    },
    {
      "epoch": 0.32234432234432236,
      "grad_norm": 0.20064710080623627,
      "learning_rate": 0.00016776084407971864,
      "loss": 0.0517,
      "step": 275
    },
    {
      "epoch": 0.3282051282051282,
      "grad_norm": 0.06760745495557785,
      "learning_rate": 0.00016717467760844082,
      "loss": 0.0223,
      "step": 280
    },
    {
      "epoch": 0.33406593406593404,
      "grad_norm": 0.14518442749977112,
      "learning_rate": 0.00016658851113716296,
      "loss": 0.0216,
      "step": 285
    },
    {
      "epoch": 0.33992673992673994,
      "grad_norm": 0.35427016019821167,
      "learning_rate": 0.0001660023446658851,
      "loss": 0.0268,
      "step": 290
    },
    {
      "epoch": 0.3457875457875458,
      "grad_norm": 0.14693213999271393,
      "learning_rate": 0.00016541617819460728,
      "loss": 0.0246,
      "step": 295
    },
    {
      "epoch": 0.3516483516483517,
      "grad_norm": 0.2351713478565216,
      "learning_rate": 0.00016483001172332943,
      "loss": 0.0399,
      "step": 300
    },
    {
      "epoch": 0.3516483516483517,
      "eval_loss": 0.02264154888689518,
      "eval_runtime": 4.8013,
      "eval_samples_per_second": 5.832,
      "eval_steps_per_second": 2.916,
      "step": 300
    },
    {
      "epoch": 0.3575091575091575,
      "grad_norm": 0.16985514760017395,
      "learning_rate": 0.0001642438452520516,
      "loss": 0.0243,
      "step": 305
    },
    {
      "epoch": 0.36336996336996336,
      "grad_norm": 0.199642613530159,
      "learning_rate": 0.00016365767878077374,
      "loss": 0.0329,
      "step": 310
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 0.05538804084062576,
      "learning_rate": 0.00016307151230949592,
      "loss": 0.0474,
      "step": 315
    },
    {
      "epoch": 0.3750915750915751,
      "grad_norm": 0.1542443037033081,
      "learning_rate": 0.00016248534583821806,
      "loss": 0.0232,
      "step": 320
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.16808335483074188,
      "learning_rate": 0.0001618991793669402,
      "loss": 0.0272,
      "step": 325
    },
    {
      "epoch": 0.3868131868131868,
      "grad_norm": 0.20816853642463684,
      "learning_rate": 0.00016131301289566238,
      "loss": 0.0375,
      "step": 330
    },
    {
      "epoch": 0.39267399267399267,
      "grad_norm": 0.10338038206100464,
      "learning_rate": 0.00016072684642438453,
      "loss": 0.0233,
      "step": 335
    },
    {
      "epoch": 0.39853479853479856,
      "grad_norm": 0.06522126495838165,
      "learning_rate": 0.0001601406799531067,
      "loss": 0.0288,
      "step": 340
    },
    {
      "epoch": 0.4043956043956044,
      "grad_norm": 0.11830403655767441,
      "learning_rate": 0.00015955451348182884,
      "loss": 0.0164,
      "step": 345
    },
    {
      "epoch": 0.41025641025641024,
      "grad_norm": 0.2777006924152374,
      "learning_rate": 0.00015896834701055102,
      "loss": 0.0345,
      "step": 350
    },
    {
      "epoch": 0.41611721611721614,
      "grad_norm": 0.1277918815612793,
      "learning_rate": 0.00015838218053927316,
      "loss": 0.0229,
      "step": 355
    },
    {
      "epoch": 0.421978021978022,
      "grad_norm": 0.09861145913600922,
      "learning_rate": 0.0001577960140679953,
      "loss": 0.0181,
      "step": 360
    },
    {
      "epoch": 0.4278388278388278,
      "grad_norm": 0.08698171377182007,
      "learning_rate": 0.00015720984759671748,
      "loss": 0.0365,
      "step": 365
    },
    {
      "epoch": 0.4336996336996337,
      "grad_norm": 0.23488883674144745,
      "learning_rate": 0.00015662368112543963,
      "loss": 0.0352,
      "step": 370
    },
    {
      "epoch": 0.43956043956043955,
      "grad_norm": 0.05140375718474388,
      "learning_rate": 0.0001560375146541618,
      "loss": 0.0235,
      "step": 375
    },
    {
      "epoch": 0.44542124542124545,
      "grad_norm": 0.15759135782718658,
      "learning_rate": 0.00015545134818288395,
      "loss": 0.0236,
      "step": 380
    },
    {
      "epoch": 0.4512820512820513,
      "grad_norm": 0.07523085922002792,
      "learning_rate": 0.00015486518171160612,
      "loss": 0.0218,
      "step": 385
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 0.1892630159854889,
      "learning_rate": 0.00015427901524032826,
      "loss": 0.0264,
      "step": 390
    },
    {
      "epoch": 0.463003663003663,
      "grad_norm": 0.12749487161636353,
      "learning_rate": 0.0001536928487690504,
      "loss": 0.0414,
      "step": 395
    },
    {
      "epoch": 0.46886446886446886,
      "grad_norm": 0.11519593745470047,
      "learning_rate": 0.00015310668229777258,
      "loss": 0.0173,
      "step": 400
    },
    {
      "epoch": 0.46886446886446886,
      "eval_loss": 0.01765686459839344,
      "eval_runtime": 4.8182,
      "eval_samples_per_second": 5.811,
      "eval_steps_per_second": 2.906,
      "step": 400
    },
    {
      "epoch": 0.4747252747252747,
      "grad_norm": 0.14240577816963196,
      "learning_rate": 0.00015252051582649473,
      "loss": 0.0214,
      "step": 405
    },
    {
      "epoch": 0.4805860805860806,
      "grad_norm": 0.11283282935619354,
      "learning_rate": 0.0001519343493552169,
      "loss": 0.0266,
      "step": 410
    },
    {
      "epoch": 0.48644688644688644,
      "grad_norm": 0.16279707849025726,
      "learning_rate": 0.00015134818288393905,
      "loss": 0.0274,
      "step": 415
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 0.07319923490285873,
      "learning_rate": 0.00015076201641266122,
      "loss": 0.0216,
      "step": 420
    },
    {
      "epoch": 0.4981684981684982,
      "grad_norm": 0.13532328605651855,
      "learning_rate": 0.00015017584994138336,
      "loss": 0.0308,
      "step": 425
    },
    {
      "epoch": 0.504029304029304,
      "grad_norm": 0.12896127998828888,
      "learning_rate": 0.0001495896834701055,
      "loss": 0.0129,
      "step": 430
    },
    {
      "epoch": 0.5098901098901099,
      "grad_norm": 0.03718201071023941,
      "learning_rate": 0.00014900351699882766,
      "loss": 0.0181,
      "step": 435
    },
    {
      "epoch": 0.5157509157509158,
      "grad_norm": 0.05926808714866638,
      "learning_rate": 0.00014841735052754983,
      "loss": 0.0146,
      "step": 440
    },
    {
      "epoch": 0.5216117216117216,
      "grad_norm": 0.09976931661367416,
      "learning_rate": 0.00014783118405627197,
      "loss": 0.0231,
      "step": 445
    },
    {
      "epoch": 0.5274725274725275,
      "grad_norm": 0.18460391461849213,
      "learning_rate": 0.00014724501758499415,
      "loss": 0.02,
      "step": 450
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.07457377016544342,
      "learning_rate": 0.00014665885111371632,
      "loss": 0.0209,
      "step": 455
    },
    {
      "epoch": 0.5391941391941392,
      "grad_norm": 0.03992030769586563,
      "learning_rate": 0.00014607268464243847,
      "loss": 0.0251,
      "step": 460
    },
    {
      "epoch": 0.545054945054945,
      "grad_norm": 0.196414053440094,
      "learning_rate": 0.0001454865181711606,
      "loss": 0.0344,
      "step": 465
    },
    {
      "epoch": 0.550915750915751,
      "grad_norm": 0.19978556036949158,
      "learning_rate": 0.00014490035169988276,
      "loss": 0.0159,
      "step": 470
    },
    {
      "epoch": 0.5567765567765568,
      "grad_norm": 0.11389517784118652,
      "learning_rate": 0.00014431418522860493,
      "loss": 0.0197,
      "step": 475
    },
    {
      "epoch": 0.5626373626373626,
      "grad_norm": 0.07047716528177261,
      "learning_rate": 0.00014372801875732708,
      "loss": 0.0138,
      "step": 480
    },
    {
      "epoch": 0.5684981684981685,
      "grad_norm": 0.10614708811044693,
      "learning_rate": 0.00014314185228604925,
      "loss": 0.0208,
      "step": 485
    },
    {
      "epoch": 0.5743589743589743,
      "grad_norm": 0.20569799840450287,
      "learning_rate": 0.00014255568581477142,
      "loss": 0.0203,
      "step": 490
    },
    {
      "epoch": 0.5802197802197803,
      "grad_norm": 0.19309553503990173,
      "learning_rate": 0.00014196951934349357,
      "loss": 0.0282,
      "step": 495
    },
    {
      "epoch": 0.5860805860805861,
      "grad_norm": 0.07542768865823746,
      "learning_rate": 0.0001413833528722157,
      "loss": 0.0173,
      "step": 500
    },
    {
      "epoch": 0.5860805860805861,
      "eval_loss": 0.022126102820038795,
      "eval_runtime": 4.8755,
      "eval_samples_per_second": 5.743,
      "eval_steps_per_second": 2.872,
      "step": 500
    },
    {
      "epoch": 0.591941391941392,
      "grad_norm": 0.1205630674958229,
      "learning_rate": 0.00014079718640093786,
      "loss": 0.0322,
      "step": 505
    },
    {
      "epoch": 0.5978021978021978,
      "grad_norm": 0.13761042058467865,
      "learning_rate": 0.00014021101992966003,
      "loss": 0.0203,
      "step": 510
    },
    {
      "epoch": 0.6036630036630036,
      "grad_norm": 0.08595598489046097,
      "learning_rate": 0.00013962485345838218,
      "loss": 0.0145,
      "step": 515
    },
    {
      "epoch": 0.6095238095238096,
      "grad_norm": 0.11087319999933243,
      "learning_rate": 0.00013903868698710435,
      "loss": 0.0218,
      "step": 520
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.14962054789066315,
      "learning_rate": 0.00013845252051582652,
      "loss": 0.0322,
      "step": 525
    },
    {
      "epoch": 0.6212454212454213,
      "grad_norm": 0.07573894411325455,
      "learning_rate": 0.00013786635404454867,
      "loss": 0.0275,
      "step": 530
    },
    {
      "epoch": 0.6271062271062271,
      "grad_norm": 0.069780133664608,
      "learning_rate": 0.0001372801875732708,
      "loss": 0.0235,
      "step": 535
    },
    {
      "epoch": 0.6329670329670329,
      "grad_norm": 0.07833613455295563,
      "learning_rate": 0.00013669402110199296,
      "loss": 0.0344,
      "step": 540
    },
    {
      "epoch": 0.6388278388278388,
      "grad_norm": 0.07331829518079758,
      "learning_rate": 0.00013610785463071513,
      "loss": 0.0135,
      "step": 545
    },
    {
      "epoch": 0.6446886446886447,
      "grad_norm": 0.22369089722633362,
      "learning_rate": 0.00013552168815943728,
      "loss": 0.0222,
      "step": 550
    },
    {
      "epoch": 0.6505494505494506,
      "grad_norm": 0.1465146392583847,
      "learning_rate": 0.00013493552168815945,
      "loss": 0.0307,
      "step": 555
    },
    {
      "epoch": 0.6564102564102564,
      "grad_norm": 0.06348715722560883,
      "learning_rate": 0.00013434935521688162,
      "loss": 0.0358,
      "step": 560
    },
    {
      "epoch": 0.6622710622710622,
      "grad_norm": 0.09298256784677505,
      "learning_rate": 0.00013376318874560377,
      "loss": 0.0224,
      "step": 565
    },
    {
      "epoch": 0.6681318681318681,
      "grad_norm": 0.18280836939811707,
      "learning_rate": 0.00013317702227432591,
      "loss": 0.0263,
      "step": 570
    },
    {
      "epoch": 0.673992673992674,
      "grad_norm": 0.07080171257257462,
      "learning_rate": 0.00013259085580304806,
      "loss": 0.0192,
      "step": 575
    },
    {
      "epoch": 0.6798534798534799,
      "grad_norm": 0.11019092798233032,
      "learning_rate": 0.00013200468933177023,
      "loss": 0.0211,
      "step": 580
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 0.13162659108638763,
      "learning_rate": 0.00013141852286049238,
      "loss": 0.0284,
      "step": 585
    },
    {
      "epoch": 0.6915750915750916,
      "grad_norm": 0.19205588102340698,
      "learning_rate": 0.00013083235638921455,
      "loss": 0.0245,
      "step": 590
    },
    {
      "epoch": 0.6974358974358974,
      "grad_norm": 0.0971173569560051,
      "learning_rate": 0.0001302461899179367,
      "loss": 0.0216,
      "step": 595
    },
    {
      "epoch": 0.7032967032967034,
      "grad_norm": 0.2179749310016632,
      "learning_rate": 0.00012966002344665887,
      "loss": 0.0268,
      "step": 600
    },
    {
      "epoch": 0.7032967032967034,
      "eval_loss": 0.01850169710814953,
      "eval_runtime": 4.6688,
      "eval_samples_per_second": 5.997,
      "eval_steps_per_second": 2.999,
      "step": 600
    },
    {
      "epoch": 0.7091575091575092,
      "grad_norm": 0.23079490661621094,
      "learning_rate": 0.00012907385697538101,
      "loss": 0.0269,
      "step": 605
    },
    {
      "epoch": 0.715018315018315,
      "grad_norm": 0.08325810730457306,
      "learning_rate": 0.00012848769050410316,
      "loss": 0.0211,
      "step": 610
    },
    {
      "epoch": 0.7208791208791209,
      "grad_norm": 0.05983910337090492,
      "learning_rate": 0.00012790152403282533,
      "loss": 0.0217,
      "step": 615
    },
    {
      "epoch": 0.7267399267399267,
      "grad_norm": 0.13831888139247894,
      "learning_rate": 0.00012731535756154748,
      "loss": 0.0142,
      "step": 620
    },
    {
      "epoch": 0.7326007326007326,
      "grad_norm": 0.12296965718269348,
      "learning_rate": 0.00012672919109026965,
      "loss": 0.0253,
      "step": 625
    },
    {
      "epoch": 0.7384615384615385,
      "grad_norm": 0.13777951896190643,
      "learning_rate": 0.0001261430246189918,
      "loss": 0.0214,
      "step": 630
    },
    {
      "epoch": 0.7443223443223443,
      "grad_norm": 0.12136834859848022,
      "learning_rate": 0.00012555685814771397,
      "loss": 0.0244,
      "step": 635
    },
    {
      "epoch": 0.7501831501831502,
      "grad_norm": 0.050576552748680115,
      "learning_rate": 0.00012497069167643612,
      "loss": 0.0137,
      "step": 640
    },
    {
      "epoch": 0.756043956043956,
      "grad_norm": 0.22222141921520233,
      "learning_rate": 0.00012438452520515826,
      "loss": 0.0254,
      "step": 645
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.06815624237060547,
      "learning_rate": 0.00012379835873388043,
      "loss": 0.0231,
      "step": 650
    },
    {
      "epoch": 0.7677655677655678,
      "grad_norm": 0.19518345594406128,
      "learning_rate": 0.00012321219226260258,
      "loss": 0.0218,
      "step": 655
    },
    {
      "epoch": 0.7736263736263737,
      "grad_norm": 0.06349798291921616,
      "learning_rate": 0.00012262602579132475,
      "loss": 0.0265,
      "step": 660
    },
    {
      "epoch": 0.7794871794871795,
      "grad_norm": 0.09031341969966888,
      "learning_rate": 0.00012203985932004688,
      "loss": 0.0299,
      "step": 665
    },
    {
      "epoch": 0.7853479853479853,
      "grad_norm": 0.060232892632484436,
      "learning_rate": 0.00012145369284876906,
      "loss": 0.0227,
      "step": 670
    },
    {
      "epoch": 0.7912087912087912,
      "grad_norm": 0.23972396552562714,
      "learning_rate": 0.00012086752637749122,
      "loss": 0.0233,
      "step": 675
    },
    {
      "epoch": 0.7970695970695971,
      "grad_norm": 0.06141636520624161,
      "learning_rate": 0.00012028135990621336,
      "loss": 0.017,
      "step": 680
    },
    {
      "epoch": 0.802930402930403,
      "grad_norm": 0.05603253096342087,
      "learning_rate": 0.00011969519343493553,
      "loss": 0.025,
      "step": 685
    },
    {
      "epoch": 0.8087912087912088,
      "grad_norm": 0.06840907782316208,
      "learning_rate": 0.0001191090269636577,
      "loss": 0.0164,
      "step": 690
    },
    {
      "epoch": 0.8146520146520146,
      "grad_norm": 0.1270790845155716,
      "learning_rate": 0.00011852286049237984,
      "loss": 0.0237,
      "step": 695
    },
    {
      "epoch": 0.8205128205128205,
      "grad_norm": 0.03222518041729927,
      "learning_rate": 0.00011793669402110198,
      "loss": 0.0169,
      "step": 700
    },
    {
      "epoch": 0.8205128205128205,
      "eval_loss": 0.026108432561159134,
      "eval_runtime": 4.6865,
      "eval_samples_per_second": 5.975,
      "eval_steps_per_second": 2.987,
      "step": 700
    },
    {
      "epoch": 0.8263736263736263,
      "grad_norm": 0.06273896247148514,
      "learning_rate": 0.00011735052754982416,
      "loss": 0.0245,
      "step": 705
    },
    {
      "epoch": 0.8322344322344323,
      "grad_norm": 0.1422451138496399,
      "learning_rate": 0.00011676436107854632,
      "loss": 0.0218,
      "step": 710
    },
    {
      "epoch": 0.8380952380952381,
      "grad_norm": 0.07166247069835663,
      "learning_rate": 0.00011617819460726846,
      "loss": 0.0259,
      "step": 715
    },
    {
      "epoch": 0.843956043956044,
      "grad_norm": 0.13188450038433075,
      "learning_rate": 0.00011559202813599064,
      "loss": 0.0185,
      "step": 720
    },
    {
      "epoch": 0.8498168498168498,
      "grad_norm": 0.11839079111814499,
      "learning_rate": 0.0001150058616647128,
      "loss": 0.0196,
      "step": 725
    },
    {
      "epoch": 0.8556776556776556,
      "grad_norm": 0.09421879053115845,
      "learning_rate": 0.00011441969519343494,
      "loss": 0.0207,
      "step": 730
    },
    {
      "epoch": 0.8615384615384616,
      "grad_norm": 0.11722107976675034,
      "learning_rate": 0.00011383352872215709,
      "loss": 0.0286,
      "step": 735
    },
    {
      "epoch": 0.8673992673992674,
      "grad_norm": 0.07790110260248184,
      "learning_rate": 0.00011324736225087926,
      "loss": 0.0157,
      "step": 740
    },
    {
      "epoch": 0.8732600732600733,
      "grad_norm": 0.11153840273618698,
      "learning_rate": 0.00011266119577960142,
      "loss": 0.0184,
      "step": 745
    },
    {
      "epoch": 0.8791208791208791,
      "grad_norm": 0.07105362415313721,
      "learning_rate": 0.00011207502930832356,
      "loss": 0.0193,
      "step": 750
    },
    {
      "epoch": 0.884981684981685,
      "grad_norm": 0.11616308242082596,
      "learning_rate": 0.00011148886283704571,
      "loss": 0.0219,
      "step": 755
    },
    {
      "epoch": 0.8908424908424909,
      "grad_norm": 0.10045047104358673,
      "learning_rate": 0.0001109026963657679,
      "loss": 0.0177,
      "step": 760
    },
    {
      "epoch": 0.8967032967032967,
      "grad_norm": 0.07033990323543549,
      "learning_rate": 0.00011031652989449004,
      "loss": 0.0227,
      "step": 765
    },
    {
      "epoch": 0.9025641025641026,
      "grad_norm": 0.07648850232362747,
      "learning_rate": 0.00010973036342321219,
      "loss": 0.023,
      "step": 770
    },
    {
      "epoch": 0.9084249084249084,
      "grad_norm": 0.05392804369330406,
      "learning_rate": 0.00010914419695193436,
      "loss": 0.0136,
      "step": 775
    },
    {
      "epoch": 0.9142857142857143,
      "grad_norm": 0.17311276495456696,
      "learning_rate": 0.00010855803048065652,
      "loss": 0.0257,
      "step": 780
    },
    {
      "epoch": 0.9201465201465201,
      "grad_norm": 0.07022574543952942,
      "learning_rate": 0.00010797186400937866,
      "loss": 0.0282,
      "step": 785
    },
    {
      "epoch": 0.926007326007326,
      "grad_norm": 0.15858297049999237,
      "learning_rate": 0.00010738569753810081,
      "loss": 0.0219,
      "step": 790
    },
    {
      "epoch": 0.9318681318681319,
      "grad_norm": 0.06796769052743912,
      "learning_rate": 0.00010679953106682298,
      "loss": 0.0288,
      "step": 795
    },
    {
      "epoch": 0.9377289377289377,
      "grad_norm": 0.11868051439523697,
      "learning_rate": 0.00010621336459554514,
      "loss": 0.0248,
      "step": 800
    },
    {
      "epoch": 0.9377289377289377,
      "eval_loss": 0.019950957968831062,
      "eval_runtime": 4.8291,
      "eval_samples_per_second": 5.798,
      "eval_steps_per_second": 2.899,
      "step": 800
    },
    {
      "epoch": 0.9435897435897436,
      "grad_norm": 0.19787561893463135,
      "learning_rate": 0.00010562719812426729,
      "loss": 0.0249,
      "step": 805
    },
    {
      "epoch": 0.9494505494505494,
      "grad_norm": 0.06437662243843079,
      "learning_rate": 0.00010504103165298946,
      "loss": 0.0141,
      "step": 810
    },
    {
      "epoch": 0.9553113553113554,
      "grad_norm": 0.09178975969552994,
      "learning_rate": 0.00010445486518171162,
      "loss": 0.0218,
      "step": 815
    },
    {
      "epoch": 0.9611721611721612,
      "grad_norm": 0.09567834436893463,
      "learning_rate": 0.00010386869871043376,
      "loss": 0.0229,
      "step": 820
    },
    {
      "epoch": 0.967032967032967,
      "grad_norm": 0.039594829082489014,
      "learning_rate": 0.00010328253223915591,
      "loss": 0.0186,
      "step": 825
    },
    {
      "epoch": 0.9728937728937729,
      "grad_norm": 0.18495650589466095,
      "learning_rate": 0.00010269636576787808,
      "loss": 0.0237,
      "step": 830
    },
    {
      "epoch": 0.9787545787545787,
      "grad_norm": 0.1861388385295868,
      "learning_rate": 0.00010211019929660024,
      "loss": 0.0367,
      "step": 835
    },
    {
      "epoch": 0.9846153846153847,
      "grad_norm": 0.05491223558783531,
      "learning_rate": 0.00010152403282532239,
      "loss": 0.015,
      "step": 840
    },
    {
      "epoch": 0.9904761904761905,
      "grad_norm": 0.04110349714756012,
      "learning_rate": 0.00010093786635404456,
      "loss": 0.0258,
      "step": 845
    },
    {
      "epoch": 0.9963369963369964,
      "grad_norm": 0.07649147510528564,
      "learning_rate": 0.00010035169988276672,
      "loss": 0.0235,
      "step": 850
    },
    {
      "epoch": 1.0021978021978022,
      "grad_norm": 0.11078579723834991,
      "learning_rate": 9.976553341148887e-05,
      "loss": 0.0204,
      "step": 855
    },
    {
      "epoch": 1.008058608058608,
      "grad_norm": 0.08302613347768784,
      "learning_rate": 9.917936694021102e-05,
      "loss": 0.0188,
      "step": 860
    },
    {
      "epoch": 1.0139194139194139,
      "grad_norm": 0.19045108556747437,
      "learning_rate": 9.859320046893318e-05,
      "loss": 0.0226,
      "step": 865
    },
    {
      "epoch": 1.0197802197802197,
      "grad_norm": 0.04657626897096634,
      "learning_rate": 9.800703399765534e-05,
      "loss": 0.0205,
      "step": 870
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 0.10237232595682144,
      "learning_rate": 9.742086752637749e-05,
      "loss": 0.0259,
      "step": 875
    },
    {
      "epoch": 1.0315018315018314,
      "grad_norm": 0.1746947020292282,
      "learning_rate": 9.683470105509965e-05,
      "loss": 0.0229,
      "step": 880
    },
    {
      "epoch": 1.0373626373626375,
      "grad_norm": 0.1032433807849884,
      "learning_rate": 9.624853458382182e-05,
      "loss": 0.0277,
      "step": 885
    },
    {
      "epoch": 1.0432234432234433,
      "grad_norm": 0.08428288996219635,
      "learning_rate": 9.566236811254397e-05,
      "loss": 0.0161,
      "step": 890
    },
    {
      "epoch": 1.0490842490842491,
      "grad_norm": 0.16661523282527924,
      "learning_rate": 9.507620164126613e-05,
      "loss": 0.0212,
      "step": 895
    },
    {
      "epoch": 1.054945054945055,
      "grad_norm": 0.08473166078329086,
      "learning_rate": 9.449003516998827e-05,
      "loss": 0.0285,
      "step": 900
    },
    {
      "epoch": 1.054945054945055,
      "eval_loss": 0.018958253785967827,
      "eval_runtime": 4.8179,
      "eval_samples_per_second": 5.812,
      "eval_steps_per_second": 2.906,
      "step": 900
    },
    {
      "epoch": 1.0608058608058608,
      "grad_norm": 0.04149739816784859,
      "learning_rate": 9.390386869871044e-05,
      "loss": 0.0138,
      "step": 905
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 0.05616866052150726,
      "learning_rate": 9.331770222743259e-05,
      "loss": 0.0194,
      "step": 910
    },
    {
      "epoch": 1.0725274725274725,
      "grad_norm": 0.16730394959449768,
      "learning_rate": 9.273153575615475e-05,
      "loss": 0.0359,
      "step": 915
    },
    {
      "epoch": 1.0783882783882783,
      "grad_norm": 0.06336849927902222,
      "learning_rate": 9.214536928487691e-05,
      "loss": 0.0147,
      "step": 920
    },
    {
      "epoch": 1.0842490842490842,
      "grad_norm": 0.0882687047123909,
      "learning_rate": 9.155920281359907e-05,
      "loss": 0.015,
      "step": 925
    },
    {
      "epoch": 1.09010989010989,
      "grad_norm": 0.05476200208067894,
      "learning_rate": 9.097303634232123e-05,
      "loss": 0.019,
      "step": 930
    },
    {
      "epoch": 1.095970695970696,
      "grad_norm": 0.05358652025461197,
      "learning_rate": 9.038686987104337e-05,
      "loss": 0.0174,
      "step": 935
    },
    {
      "epoch": 1.101831501831502,
      "grad_norm": 0.24095569550991058,
      "learning_rate": 8.980070339976554e-05,
      "loss": 0.0293,
      "step": 940
    },
    {
      "epoch": 1.1076923076923078,
      "grad_norm": 0.06653840094804764,
      "learning_rate": 8.921453692848769e-05,
      "loss": 0.0133,
      "step": 945
    },
    {
      "epoch": 1.1135531135531136,
      "grad_norm": 0.08839567750692368,
      "learning_rate": 8.862837045720985e-05,
      "loss": 0.0208,
      "step": 950
    },
    {
      "epoch": 1.1194139194139194,
      "grad_norm": 0.032916922122240067,
      "learning_rate": 8.804220398593201e-05,
      "loss": 0.0267,
      "step": 955
    },
    {
      "epoch": 1.1252747252747253,
      "grad_norm": 0.11819420754909515,
      "learning_rate": 8.745603751465417e-05,
      "loss": 0.0384,
      "step": 960
    },
    {
      "epoch": 1.1311355311355311,
      "grad_norm": 0.06757565587759018,
      "learning_rate": 8.686987104337633e-05,
      "loss": 0.0135,
      "step": 965
    },
    {
      "epoch": 1.136996336996337,
      "grad_norm": 0.0970228835940361,
      "learning_rate": 8.628370457209847e-05,
      "loss": 0.0166,
      "step": 970
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.13436350226402283,
      "learning_rate": 8.569753810082065e-05,
      "loss": 0.016,
      "step": 975
    },
    {
      "epoch": 1.1487179487179486,
      "grad_norm": 0.10584839433431625,
      "learning_rate": 8.511137162954279e-05,
      "loss": 0.0177,
      "step": 980
    },
    {
      "epoch": 1.1545787545787545,
      "grad_norm": 0.21206024289131165,
      "learning_rate": 8.452520515826495e-05,
      "loss": 0.0325,
      "step": 985
    },
    {
      "epoch": 1.1604395604395605,
      "grad_norm": 0.04815613850951195,
      "learning_rate": 8.393903868698711e-05,
      "loss": 0.0137,
      "step": 990
    },
    {
      "epoch": 1.1663003663003664,
      "grad_norm": 0.07466138154268265,
      "learning_rate": 8.335287221570927e-05,
      "loss": 0.0173,
      "step": 995
    },
    {
      "epoch": 1.1721611721611722,
      "grad_norm": 0.09366811066865921,
      "learning_rate": 8.276670574443143e-05,
      "loss": 0.025,
      "step": 1000
    },
    {
      "epoch": 1.1721611721611722,
      "eval_loss": 0.019907595589756966,
      "eval_runtime": 4.819,
      "eval_samples_per_second": 5.81,
      "eval_steps_per_second": 2.905,
      "step": 1000
    },
    {
      "epoch": 1.178021978021978,
      "grad_norm": 0.08527784794569016,
      "learning_rate": 8.218053927315357e-05,
      "loss": 0.0208,
      "step": 1005
    },
    {
      "epoch": 1.183882783882784,
      "grad_norm": 0.08328138291835785,
      "learning_rate": 8.159437280187575e-05,
      "loss": 0.0216,
      "step": 1010
    },
    {
      "epoch": 1.1897435897435897,
      "grad_norm": 0.05000188946723938,
      "learning_rate": 8.100820633059789e-05,
      "loss": 0.0211,
      "step": 1015
    },
    {
      "epoch": 1.1956043956043956,
      "grad_norm": 0.028807902708649635,
      "learning_rate": 8.042203985932005e-05,
      "loss": 0.0096,
      "step": 1020
    },
    {
      "epoch": 1.2014652014652014,
      "grad_norm": 0.20507606863975525,
      "learning_rate": 7.983587338804221e-05,
      "loss": 0.0222,
      "step": 1025
    },
    {
      "epoch": 1.2073260073260073,
      "grad_norm": 0.04885656014084816,
      "learning_rate": 7.924970691676437e-05,
      "loss": 0.0215,
      "step": 1030
    },
    {
      "epoch": 1.213186813186813,
      "grad_norm": 0.047489382326602936,
      "learning_rate": 7.866354044548652e-05,
      "loss": 0.0178,
      "step": 1035
    },
    {
      "epoch": 1.2190476190476192,
      "grad_norm": 0.05971779301762581,
      "learning_rate": 7.807737397420867e-05,
      "loss": 0.0176,
      "step": 1040
    },
    {
      "epoch": 1.224908424908425,
      "grad_norm": 0.04695736616849899,
      "learning_rate": 7.749120750293083e-05,
      "loss": 0.0148,
      "step": 1045
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.08131909370422363,
      "learning_rate": 7.690504103165299e-05,
      "loss": 0.0201,
      "step": 1050
    },
    {
      "epoch": 1.2366300366300367,
      "grad_norm": 0.06954577565193176,
      "learning_rate": 7.631887456037515e-05,
      "loss": 0.0149,
      "step": 1055
    },
    {
      "epoch": 1.2424908424908425,
      "grad_norm": 0.054430391639471054,
      "learning_rate": 7.57327080890973e-05,
      "loss": 0.0078,
      "step": 1060
    },
    {
      "epoch": 1.2483516483516484,
      "grad_norm": 0.12231959402561188,
      "learning_rate": 7.514654161781947e-05,
      "loss": 0.0258,
      "step": 1065
    },
    {
      "epoch": 1.2542124542124542,
      "grad_norm": 0.04983118548989296,
      "learning_rate": 7.456037514654162e-05,
      "loss": 0.0189,
      "step": 1070
    },
    {
      "epoch": 1.26007326007326,
      "grad_norm": 0.11981873214244843,
      "learning_rate": 7.397420867526378e-05,
      "loss": 0.0156,
      "step": 1075
    },
    {
      "epoch": 1.2659340659340659,
      "grad_norm": 0.03829724341630936,
      "learning_rate": 7.338804220398593e-05,
      "loss": 0.0162,
      "step": 1080
    },
    {
      "epoch": 1.2717948717948717,
      "grad_norm": 0.1572490632534027,
      "learning_rate": 7.28018757327081e-05,
      "loss": 0.0188,
      "step": 1085
    },
    {
      "epoch": 1.2776556776556776,
      "grad_norm": 0.122514508664608,
      "learning_rate": 7.221570926143025e-05,
      "loss": 0.0229,
      "step": 1090
    },
    {
      "epoch": 1.2835164835164834,
      "grad_norm": 0.06537042558193207,
      "learning_rate": 7.16295427901524e-05,
      "loss": 0.0222,
      "step": 1095
    },
    {
      "epoch": 1.2893772893772895,
      "grad_norm": 0.1269371509552002,
      "learning_rate": 7.104337631887457e-05,
      "loss": 0.0272,
      "step": 1100
    },
    {
      "epoch": 1.2893772893772895,
      "eval_loss": 0.015397748909890652,
      "eval_runtime": 4.7526,
      "eval_samples_per_second": 5.892,
      "eval_steps_per_second": 2.946,
      "step": 1100
    },
    {
      "epoch": 1.2952380952380953,
      "grad_norm": 0.07507819682359695,
      "learning_rate": 7.045720984759672e-05,
      "loss": 0.0256,
      "step": 1105
    },
    {
      "epoch": 1.3010989010989011,
      "grad_norm": 0.04198193550109863,
      "learning_rate": 6.987104337631888e-05,
      "loss": 0.0102,
      "step": 1110
    },
    {
      "epoch": 1.306959706959707,
      "grad_norm": 0.053751297295093536,
      "learning_rate": 6.928487690504104e-05,
      "loss": 0.0141,
      "step": 1115
    },
    {
      "epoch": 1.3128205128205128,
      "grad_norm": 0.12076237797737122,
      "learning_rate": 6.86987104337632e-05,
      "loss": 0.0165,
      "step": 1120
    },
    {
      "epoch": 1.3186813186813187,
      "grad_norm": 0.0769004300236702,
      "learning_rate": 6.811254396248535e-05,
      "loss": 0.0191,
      "step": 1125
    },
    {
      "epoch": 1.3245421245421245,
      "grad_norm": 0.08021704852581024,
      "learning_rate": 6.75263774912075e-05,
      "loss": 0.0253,
      "step": 1130
    },
    {
      "epoch": 1.3304029304029303,
      "grad_norm": 0.09786754846572876,
      "learning_rate": 6.694021101992967e-05,
      "loss": 0.0191,
      "step": 1135
    },
    {
      "epoch": 1.3362637362637364,
      "grad_norm": 0.06878714263439178,
      "learning_rate": 6.635404454865182e-05,
      "loss": 0.0326,
      "step": 1140
    },
    {
      "epoch": 1.3421245421245422,
      "grad_norm": 0.11297193914651871,
      "learning_rate": 6.576787807737398e-05,
      "loss": 0.0185,
      "step": 1145
    },
    {
      "epoch": 1.347985347985348,
      "grad_norm": 0.10731342434883118,
      "learning_rate": 6.518171160609614e-05,
      "loss": 0.0168,
      "step": 1150
    },
    {
      "epoch": 1.353846153846154,
      "grad_norm": 0.08888328820466995,
      "learning_rate": 6.45955451348183e-05,
      "loss": 0.0182,
      "step": 1155
    },
    {
      "epoch": 1.3597069597069598,
      "grad_norm": 0.1666301190853119,
      "learning_rate": 6.400937866354045e-05,
      "loss": 0.0254,
      "step": 1160
    },
    {
      "epoch": 1.3655677655677656,
      "grad_norm": 0.1334419697523117,
      "learning_rate": 6.34232121922626e-05,
      "loss": 0.0215,
      "step": 1165
    },
    {
      "epoch": 1.3714285714285714,
      "grad_norm": 0.05616243556141853,
      "learning_rate": 6.283704572098477e-05,
      "loss": 0.0156,
      "step": 1170
    },
    {
      "epoch": 1.3772893772893773,
      "grad_norm": 0.1660885214805603,
      "learning_rate": 6.225087924970692e-05,
      "loss": 0.0241,
      "step": 1175
    },
    {
      "epoch": 1.3831501831501831,
      "grad_norm": 0.09245380759239197,
      "learning_rate": 6.166471277842908e-05,
      "loss": 0.0222,
      "step": 1180
    },
    {
      "epoch": 1.389010989010989,
      "grad_norm": 0.08635041117668152,
      "learning_rate": 6.107854630715122e-05,
      "loss": 0.0203,
      "step": 1185
    },
    {
      "epoch": 1.3948717948717948,
      "grad_norm": 0.07752135396003723,
      "learning_rate": 6.049237983587339e-05,
      "loss": 0.0216,
      "step": 1190
    },
    {
      "epoch": 1.4007326007326006,
      "grad_norm": 0.10603225976228714,
      "learning_rate": 5.990621336459554e-05,
      "loss": 0.0205,
      "step": 1195
    },
    {
      "epoch": 1.4065934065934065,
      "grad_norm": 0.04343140870332718,
      "learning_rate": 5.932004689331771e-05,
      "loss": 0.0105,
      "step": 1200
    },
    {
      "epoch": 1.4065934065934065,
      "eval_loss": 0.015952473506331444,
      "eval_runtime": 4.7686,
      "eval_samples_per_second": 5.872,
      "eval_steps_per_second": 2.936,
      "step": 1200
    },
    {
      "epoch": 1.4124542124542123,
      "grad_norm": 0.12541887164115906,
      "learning_rate": 5.873388042203987e-05,
      "loss": 0.0217,
      "step": 1205
    },
    {
      "epoch": 1.4183150183150184,
      "grad_norm": 0.05417335778474808,
      "learning_rate": 5.814771395076202e-05,
      "loss": 0.0205,
      "step": 1210
    },
    {
      "epoch": 1.4241758241758242,
      "grad_norm": 0.04461506009101868,
      "learning_rate": 5.756154747948418e-05,
      "loss": 0.0211,
      "step": 1215
    },
    {
      "epoch": 1.43003663003663,
      "grad_norm": 0.09423286467790604,
      "learning_rate": 5.697538100820633e-05,
      "loss": 0.0238,
      "step": 1220
    },
    {
      "epoch": 1.435897435897436,
      "grad_norm": 0.050094570964574814,
      "learning_rate": 5.638921453692849e-05,
      "loss": 0.0163,
      "step": 1225
    },
    {
      "epoch": 1.4417582417582417,
      "grad_norm": 0.13104532659053802,
      "learning_rate": 5.580304806565064e-05,
      "loss": 0.0219,
      "step": 1230
    },
    {
      "epoch": 1.4476190476190476,
      "grad_norm": 0.08238503336906433,
      "learning_rate": 5.52168815943728e-05,
      "loss": 0.0126,
      "step": 1235
    },
    {
      "epoch": 1.4534798534798534,
      "grad_norm": 0.1029452383518219,
      "learning_rate": 5.463071512309497e-05,
      "loss": 0.0247,
      "step": 1240
    },
    {
      "epoch": 1.4593406593406593,
      "grad_norm": 0.05564792454242706,
      "learning_rate": 5.404454865181712e-05,
      "loss": 0.0212,
      "step": 1245
    },
    {
      "epoch": 1.4652014652014653,
      "grad_norm": 0.08589282631874084,
      "learning_rate": 5.345838218053928e-05,
      "loss": 0.0184,
      "step": 1250
    },
    {
      "epoch": 1.4710622710622712,
      "grad_norm": 0.15644195675849915,
      "learning_rate": 5.287221570926143e-05,
      "loss": 0.0165,
      "step": 1255
    },
    {
      "epoch": 1.476923076923077,
      "grad_norm": 0.11301274597644806,
      "learning_rate": 5.228604923798359e-05,
      "loss": 0.0322,
      "step": 1260
    },
    {
      "epoch": 1.4827838827838828,
      "grad_norm": 0.044553741812705994,
      "learning_rate": 5.1699882766705743e-05,
      "loss": 0.0183,
      "step": 1265
    },
    {
      "epoch": 1.4886446886446887,
      "grad_norm": 0.06141185760498047,
      "learning_rate": 5.11137162954279e-05,
      "loss": 0.0102,
      "step": 1270
    },
    {
      "epoch": 1.4945054945054945,
      "grad_norm": 0.08107537031173706,
      "learning_rate": 5.0527549824150055e-05,
      "loss": 0.0215,
      "step": 1275
    },
    {
      "epoch": 1.5003663003663004,
      "grad_norm": 0.06017793342471123,
      "learning_rate": 4.9941383352872214e-05,
      "loss": 0.0176,
      "step": 1280
    },
    {
      "epoch": 1.5062271062271062,
      "grad_norm": 0.051033902913331985,
      "learning_rate": 4.9355216881594373e-05,
      "loss": 0.0149,
      "step": 1285
    },
    {
      "epoch": 1.512087912087912,
      "grad_norm": 0.124452143907547,
      "learning_rate": 4.876905041031653e-05,
      "loss": 0.0209,
      "step": 1290
    },
    {
      "epoch": 1.5179487179487179,
      "grad_norm": 0.1616523712873459,
      "learning_rate": 4.8182883939038685e-05,
      "loss": 0.0192,
      "step": 1295
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.07067764550447464,
      "learning_rate": 4.759671746776085e-05,
      "loss": 0.0279,
      "step": 1300
    },
    {
      "epoch": 1.5238095238095237,
      "eval_loss": 0.017229218035936356,
      "eval_runtime": 4.6702,
      "eval_samples_per_second": 5.996,
      "eval_steps_per_second": 2.998,
      "step": 1300
    },
    {
      "epoch": 1.5296703296703296,
      "grad_norm": 0.06332267820835114,
      "learning_rate": 4.7010550996483003e-05,
      "loss": 0.0253,
      "step": 1305
    },
    {
      "epoch": 1.5355311355311354,
      "grad_norm": 0.08032066375017166,
      "learning_rate": 4.642438452520516e-05,
      "loss": 0.0128,
      "step": 1310
    },
    {
      "epoch": 1.5413919413919412,
      "grad_norm": 0.11456907540559769,
      "learning_rate": 4.5838218053927315e-05,
      "loss": 0.0127,
      "step": 1315
    },
    {
      "epoch": 1.5472527472527473,
      "grad_norm": 0.19653138518333435,
      "learning_rate": 4.5252051582649474e-05,
      "loss": 0.0236,
      "step": 1320
    },
    {
      "epoch": 1.5531135531135531,
      "grad_norm": 0.08195839822292328,
      "learning_rate": 4.4665885111371633e-05,
      "loss": 0.0173,
      "step": 1325
    },
    {
      "epoch": 1.558974358974359,
      "grad_norm": 0.11376089602708817,
      "learning_rate": 4.4079718640093786e-05,
      "loss": 0.0115,
      "step": 1330
    },
    {
      "epoch": 1.5648351648351648,
      "grad_norm": 0.055264201015233994,
      "learning_rate": 4.3493552168815945e-05,
      "loss": 0.0198,
      "step": 1335
    },
    {
      "epoch": 1.5706959706959707,
      "grad_norm": 0.13038881123065948,
      "learning_rate": 4.2907385697538104e-05,
      "loss": 0.0231,
      "step": 1340
    },
    {
      "epoch": 1.5765567765567765,
      "grad_norm": 0.0317939892411232,
      "learning_rate": 4.2321219226260263e-05,
      "loss": 0.0083,
      "step": 1345
    },
    {
      "epoch": 1.5824175824175826,
      "grad_norm": 0.151336207985878,
      "learning_rate": 4.1735052754982416e-05,
      "loss": 0.0225,
      "step": 1350
    },
    {
      "epoch": 1.5882783882783884,
      "grad_norm": 0.07817093282938004,
      "learning_rate": 4.1148886283704575e-05,
      "loss": 0.0226,
      "step": 1355
    },
    {
      "epoch": 1.5941391941391942,
      "grad_norm": 0.1341279149055481,
      "learning_rate": 4.056271981242673e-05,
      "loss": 0.0263,
      "step": 1360
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.06353727728128433,
      "learning_rate": 3.997655334114889e-05,
      "loss": 0.0198,
      "step": 1365
    },
    {
      "epoch": 1.605860805860806,
      "grad_norm": 0.11177901178598404,
      "learning_rate": 3.9390386869871046e-05,
      "loss": 0.0172,
      "step": 1370
    },
    {
      "epoch": 1.6117216117216118,
      "grad_norm": 0.047024596482515335,
      "learning_rate": 3.88042203985932e-05,
      "loss": 0.0207,
      "step": 1375
    },
    {
      "epoch": 1.6175824175824176,
      "grad_norm": 0.04343528300523758,
      "learning_rate": 3.8218053927315364e-05,
      "loss": 0.0214,
      "step": 1380
    },
    {
      "epoch": 1.6234432234432234,
      "grad_norm": 0.08330193161964417,
      "learning_rate": 3.763188745603752e-05,
      "loss": 0.0286,
      "step": 1385
    },
    {
      "epoch": 1.6293040293040293,
      "grad_norm": 0.0811009556055069,
      "learning_rate": 3.7045720984759676e-05,
      "loss": 0.0148,
      "step": 1390
    },
    {
      "epoch": 1.6351648351648351,
      "grad_norm": 0.1049441322684288,
      "learning_rate": 3.645955451348183e-05,
      "loss": 0.0184,
      "step": 1395
    },
    {
      "epoch": 1.641025641025641,
      "grad_norm": 0.11944428086280823,
      "learning_rate": 3.587338804220399e-05,
      "loss": 0.0122,
      "step": 1400
    },
    {
      "epoch": 1.641025641025641,
      "eval_loss": 0.017561230808496475,
      "eval_runtime": 4.685,
      "eval_samples_per_second": 5.977,
      "eval_steps_per_second": 2.988,
      "step": 1400
    },
    {
      "epoch": 1.6468864468864468,
      "grad_norm": 0.14023366570472717,
      "learning_rate": 3.528722157092615e-05,
      "loss": 0.0178,
      "step": 1405
    },
    {
      "epoch": 1.6527472527472526,
      "grad_norm": 0.14057691395282745,
      "learning_rate": 3.47010550996483e-05,
      "loss": 0.0268,
      "step": 1410
    },
    {
      "epoch": 1.6586080586080585,
      "grad_norm": 0.1253061443567276,
      "learning_rate": 3.411488862837046e-05,
      "loss": 0.0266,
      "step": 1415
    },
    {
      "epoch": 1.6644688644688643,
      "grad_norm": 0.03431854769587517,
      "learning_rate": 3.352872215709262e-05,
      "loss": 0.02,
      "step": 1420
    },
    {
      "epoch": 1.6703296703296702,
      "grad_norm": 0.13929079473018646,
      "learning_rate": 3.294255568581478e-05,
      "loss": 0.0226,
      "step": 1425
    },
    {
      "epoch": 1.6761904761904762,
      "grad_norm": 0.06429693102836609,
      "learning_rate": 3.235638921453693e-05,
      "loss": 0.0225,
      "step": 1430
    },
    {
      "epoch": 1.682051282051282,
      "grad_norm": 0.029311953112483025,
      "learning_rate": 3.177022274325909e-05,
      "loss": 0.0161,
      "step": 1435
    },
    {
      "epoch": 1.687912087912088,
      "grad_norm": 0.04346455633640289,
      "learning_rate": 3.118405627198124e-05,
      "loss": 0.0155,
      "step": 1440
    },
    {
      "epoch": 1.6937728937728938,
      "grad_norm": 0.09009824693202972,
      "learning_rate": 3.05978898007034e-05,
      "loss": 0.0153,
      "step": 1445
    },
    {
      "epoch": 1.6996336996336996,
      "grad_norm": 0.071926549077034,
      "learning_rate": 3.0011723329425556e-05,
      "loss": 0.0136,
      "step": 1450
    },
    {
      "epoch": 1.7054945054945057,
      "grad_norm": 0.06461833417415619,
      "learning_rate": 2.9425556858147718e-05,
      "loss": 0.0237,
      "step": 1455
    },
    {
      "epoch": 1.7113553113553115,
      "grad_norm": 0.039929524064064026,
      "learning_rate": 2.8839390386869874e-05,
      "loss": 0.0187,
      "step": 1460
    },
    {
      "epoch": 1.7172161172161173,
      "grad_norm": 0.0534372515976429,
      "learning_rate": 2.825322391559203e-05,
      "loss": 0.0192,
      "step": 1465
    },
    {
      "epoch": 1.7230769230769232,
      "grad_norm": 0.1467376947402954,
      "learning_rate": 2.7667057444314186e-05,
      "loss": 0.0203,
      "step": 1470
    },
    {
      "epoch": 1.728937728937729,
      "grad_norm": 0.0830003172159195,
      "learning_rate": 2.7080890973036345e-05,
      "loss": 0.0188,
      "step": 1475
    },
    {
      "epoch": 1.7347985347985349,
      "grad_norm": 0.07220768928527832,
      "learning_rate": 2.64947245017585e-05,
      "loss": 0.0118,
      "step": 1480
    },
    {
      "epoch": 1.7406593406593407,
      "grad_norm": 0.0751115009188652,
      "learning_rate": 2.5908558030480656e-05,
      "loss": 0.0156,
      "step": 1485
    },
    {
      "epoch": 1.7465201465201465,
      "grad_norm": 0.07690921425819397,
      "learning_rate": 2.5322391559202812e-05,
      "loss": 0.0347,
      "step": 1490
    },
    {
      "epoch": 1.7523809523809524,
      "grad_norm": 0.05416159704327583,
      "learning_rate": 2.473622508792497e-05,
      "loss": 0.0167,
      "step": 1495
    },
    {
      "epoch": 1.7582417582417582,
      "grad_norm": 0.0676250010728836,
      "learning_rate": 2.4150058616647127e-05,
      "loss": 0.0205,
      "step": 1500
    },
    {
      "epoch": 1.7582417582417582,
      "eval_loss": 0.019522378221154213,
      "eval_runtime": 4.6942,
      "eval_samples_per_second": 5.965,
      "eval_steps_per_second": 2.982,
      "step": 1500
    },
    {
      "epoch": 1.764102564102564,
      "grad_norm": 0.08909754455089569,
      "learning_rate": 2.3563892145369286e-05,
      "loss": 0.0249,
      "step": 1505
    },
    {
      "epoch": 1.76996336996337,
      "grad_norm": 0.042161233723163605,
      "learning_rate": 2.2977725674091442e-05,
      "loss": 0.0117,
      "step": 1510
    },
    {
      "epoch": 1.7758241758241757,
      "grad_norm": 0.07136218249797821,
      "learning_rate": 2.23915592028136e-05,
      "loss": 0.0212,
      "step": 1515
    },
    {
      "epoch": 1.7816849816849816,
      "grad_norm": 0.14128735661506653,
      "learning_rate": 2.1805392731535757e-05,
      "loss": 0.0189,
      "step": 1520
    },
    {
      "epoch": 1.7875457875457874,
      "grad_norm": 0.05959760770201683,
      "learning_rate": 2.1219226260257916e-05,
      "loss": 0.0119,
      "step": 1525
    },
    {
      "epoch": 1.7934065934065933,
      "grad_norm": 0.038479190319776535,
      "learning_rate": 2.0633059788980072e-05,
      "loss": 0.013,
      "step": 1530
    },
    {
      "epoch": 1.7992673992673993,
      "grad_norm": 0.09512809664011002,
      "learning_rate": 2.0046893317702228e-05,
      "loss": 0.0148,
      "step": 1535
    },
    {
      "epoch": 1.8051282051282052,
      "grad_norm": 0.14848454296588898,
      "learning_rate": 1.9460726846424384e-05,
      "loss": 0.019,
      "step": 1540
    },
    {
      "epoch": 1.810989010989011,
      "grad_norm": 0.10240516811609268,
      "learning_rate": 1.8874560375146543e-05,
      "loss": 0.017,
      "step": 1545
    },
    {
      "epoch": 1.8168498168498168,
      "grad_norm": 0.09345954656600952,
      "learning_rate": 1.82883939038687e-05,
      "loss": 0.0237,
      "step": 1550
    },
    {
      "epoch": 1.8227106227106227,
      "grad_norm": 0.03815275430679321,
      "learning_rate": 1.7702227432590858e-05,
      "loss": 0.0188,
      "step": 1555
    },
    {
      "epoch": 1.8285714285714287,
      "grad_norm": 0.027827398851513863,
      "learning_rate": 1.7116060961313014e-05,
      "loss": 0.0183,
      "step": 1560
    },
    {
      "epoch": 1.8344322344322346,
      "grad_norm": 0.08811303228139877,
      "learning_rate": 1.6529894490035173e-05,
      "loss": 0.0191,
      "step": 1565
    },
    {
      "epoch": 1.8402930402930404,
      "grad_norm": 0.03119056299328804,
      "learning_rate": 1.594372801875733e-05,
      "loss": 0.0183,
      "step": 1570
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.09752997010946274,
      "learning_rate": 1.5357561547479485e-05,
      "loss": 0.0161,
      "step": 1575
    },
    {
      "epoch": 1.852014652014652,
      "grad_norm": 0.0855243131518364,
      "learning_rate": 1.477139507620164e-05,
      "loss": 0.015,
      "step": 1580
    },
    {
      "epoch": 1.857875457875458,
      "grad_norm": 0.08388842642307281,
      "learning_rate": 1.41852286049238e-05,
      "loss": 0.0148,
      "step": 1585
    },
    {
      "epoch": 1.8637362637362638,
      "grad_norm": 0.10147551447153091,
      "learning_rate": 1.3599062133645957e-05,
      "loss": 0.0154,
      "step": 1590
    },
    {
      "epoch": 1.8695970695970696,
      "grad_norm": 0.0457012839615345,
      "learning_rate": 1.3012895662368113e-05,
      "loss": 0.0186,
      "step": 1595
    },
    {
      "epoch": 1.8754578754578755,
      "grad_norm": 0.03654688224196434,
      "learning_rate": 1.242672919109027e-05,
      "loss": 0.0321,
      "step": 1600
    },
    {
      "epoch": 1.8754578754578755,
      "eval_loss": 0.017443044111132622,
      "eval_runtime": 4.7108,
      "eval_samples_per_second": 5.944,
      "eval_steps_per_second": 2.972,
      "step": 1600
    },
    {
      "epoch": 1.8813186813186813,
      "grad_norm": 0.07887323200702667,
      "learning_rate": 1.1840562719812428e-05,
      "loss": 0.0142,
      "step": 1605
    },
    {
      "epoch": 1.8871794871794871,
      "grad_norm": 0.11328335106372833,
      "learning_rate": 1.1254396248534585e-05,
      "loss": 0.025,
      "step": 1610
    },
    {
      "epoch": 1.893040293040293,
      "grad_norm": 0.09318089485168457,
      "learning_rate": 1.0668229777256741e-05,
      "loss": 0.0204,
      "step": 1615
    },
    {
      "epoch": 1.8989010989010988,
      "grad_norm": 0.06992164254188538,
      "learning_rate": 1.0082063305978899e-05,
      "loss": 0.0135,
      "step": 1620
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.14927181601524353,
      "learning_rate": 9.495896834701056e-06,
      "loss": 0.0245,
      "step": 1625
    },
    {
      "epoch": 1.9106227106227105,
      "grad_norm": 0.11408836394548416,
      "learning_rate": 8.909730363423214e-06,
      "loss": 0.0161,
      "step": 1630
    },
    {
      "epoch": 1.9164835164835163,
      "grad_norm": 0.06911155581474304,
      "learning_rate": 8.32356389214537e-06,
      "loss": 0.0154,
      "step": 1635
    },
    {
      "epoch": 1.9223443223443224,
      "grad_norm": 0.11621779948472977,
      "learning_rate": 7.737397420867527e-06,
      "loss": 0.0144,
      "step": 1640
    },
    {
      "epoch": 1.9282051282051282,
      "grad_norm": 0.046058397740125656,
      "learning_rate": 7.151230949589684e-06,
      "loss": 0.0093,
      "step": 1645
    },
    {
      "epoch": 1.934065934065934,
      "grad_norm": 0.11228576302528381,
      "learning_rate": 6.565064478311841e-06,
      "loss": 0.022,
      "step": 1650
    },
    {
      "epoch": 1.93992673992674,
      "grad_norm": 0.1315338909626007,
      "learning_rate": 5.978898007033998e-06,
      "loss": 0.0193,
      "step": 1655
    },
    {
      "epoch": 1.9457875457875458,
      "grad_norm": 0.040056392550468445,
      "learning_rate": 5.3927315357561546e-06,
      "loss": 0.0132,
      "step": 1660
    },
    {
      "epoch": 1.9516483516483516,
      "grad_norm": 0.10741738229990005,
      "learning_rate": 4.806565064478312e-06,
      "loss": 0.0352,
      "step": 1665
    },
    {
      "epoch": 1.9575091575091577,
      "grad_norm": 0.059029560536146164,
      "learning_rate": 4.220398593200469e-06,
      "loss": 0.019,
      "step": 1670
    },
    {
      "epoch": 1.9633699633699635,
      "grad_norm": 0.06789711117744446,
      "learning_rate": 3.6342321219226262e-06,
      "loss": 0.0189,
      "step": 1675
    },
    {
      "epoch": 1.9692307692307693,
      "grad_norm": 0.02918117679655552,
      "learning_rate": 3.0480656506447833e-06,
      "loss": 0.0259,
      "step": 1680
    },
    {
      "epoch": 1.9750915750915752,
      "grad_norm": 0.08073403686285019,
      "learning_rate": 2.4618991793669404e-06,
      "loss": 0.0286,
      "step": 1685
    },
    {
      "epoch": 1.980952380952381,
      "grad_norm": 0.1617717295885086,
      "learning_rate": 1.8757327080890972e-06,
      "loss": 0.0191,
      "step": 1690
    },
    {
      "epoch": 1.9868131868131869,
      "grad_norm": 0.06613462418317795,
      "learning_rate": 1.2895662368112545e-06,
      "loss": 0.0128,
      "step": 1695
    },
    {
      "epoch": 1.9926739926739927,
      "grad_norm": 0.08398256450891495,
      "learning_rate": 7.033997655334116e-07,
      "loss": 0.0118,
      "step": 1700
    },
    {
      "epoch": 1.9926739926739927,
      "eval_loss": 0.018166696652770042,
      "eval_runtime": 4.6959,
      "eval_samples_per_second": 5.963,
      "eval_steps_per_second": 2.981,
      "step": 1700
    },
    {
      "epoch": 1.9985347985347985,
      "grad_norm": 0.0773661658167839,
      "learning_rate": 1.1723329425556858e-07,
      "loss": 0.0146,
      "step": 1705
    }
  ],
  "logging_steps": 5,
  "max_steps": 1706,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.764499308335456e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}