{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1179,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02544529262086514,
      "grad_norm": 3.0892276763916016,
      "learning_rate": 5e-06,
      "loss": 0.8937,
      "step": 10
    },
    {
      "epoch": 0.05089058524173028,
      "grad_norm": 1.0776044130325317,
      "learning_rate": 5e-06,
      "loss": 0.7655,
      "step": 20
    },
    {
      "epoch": 0.07633587786259542,
      "grad_norm": 2.4046175479888916,
      "learning_rate": 5e-06,
      "loss": 0.7156,
      "step": 30
    },
    {
      "epoch": 0.10178117048346055,
      "grad_norm": 1.0319740772247314,
      "learning_rate": 5e-06,
      "loss": 0.6884,
      "step": 40
    },
    {
      "epoch": 0.1272264631043257,
      "grad_norm": 0.8365700840950012,
      "learning_rate": 5e-06,
      "loss": 0.6666,
      "step": 50
    },
    {
      "epoch": 0.15267175572519084,
      "grad_norm": 1.0443907976150513,
      "learning_rate": 5e-06,
      "loss": 0.6517,
      "step": 60
    },
    {
      "epoch": 0.178117048346056,
      "grad_norm": 0.6737371683120728,
      "learning_rate": 5e-06,
      "loss": 0.6476,
      "step": 70
    },
    {
      "epoch": 0.2035623409669211,
      "grad_norm": 0.7133864760398865,
      "learning_rate": 5e-06,
      "loss": 0.6313,
      "step": 80
    },
    {
      "epoch": 0.22900763358778625,
      "grad_norm": 0.873466432094574,
      "learning_rate": 5e-06,
      "loss": 0.6223,
      "step": 90
    },
    {
      "epoch": 0.2544529262086514,
      "grad_norm": 0.8450167179107666,
      "learning_rate": 5e-06,
      "loss": 0.6252,
      "step": 100
    },
    {
      "epoch": 0.27989821882951654,
      "grad_norm": 0.5868242383003235,
      "learning_rate": 5e-06,
      "loss": 0.6153,
      "step": 110
    },
    {
      "epoch": 0.3053435114503817,
      "grad_norm": 0.6943073868751526,
      "learning_rate": 5e-06,
      "loss": 0.6155,
      "step": 120
    },
    {
      "epoch": 0.33078880407124683,
      "grad_norm": 0.568712592124939,
      "learning_rate": 5e-06,
      "loss": 0.6086,
      "step": 130
    },
    {
      "epoch": 0.356234096692112,
      "grad_norm": 0.5304349660873413,
      "learning_rate": 5e-06,
      "loss": 0.6046,
      "step": 140
    },
    {
      "epoch": 0.3816793893129771,
      "grad_norm": 0.5914269089698792,
      "learning_rate": 5e-06,
      "loss": 0.6014,
      "step": 150
    },
    {
      "epoch": 0.4071246819338422,
      "grad_norm": 0.6255616545677185,
      "learning_rate": 5e-06,
      "loss": 0.6059,
      "step": 160
    },
    {
      "epoch": 0.43256997455470736,
      "grad_norm": 0.6703746318817139,
      "learning_rate": 5e-06,
      "loss": 0.595,
      "step": 170
    },
    {
      "epoch": 0.4580152671755725,
      "grad_norm": 0.6239282488822937,
      "learning_rate": 5e-06,
      "loss": 0.5961,
      "step": 180
    },
    {
      "epoch": 0.48346055979643765,
      "grad_norm": 0.5409078598022461,
      "learning_rate": 5e-06,
      "loss": 0.5936,
      "step": 190
    },
    {
      "epoch": 0.5089058524173028,
      "grad_norm": 0.47156909108161926,
      "learning_rate": 5e-06,
      "loss": 0.5871,
      "step": 200
    },
    {
      "epoch": 0.5343511450381679,
      "grad_norm": 0.6108508110046387,
      "learning_rate": 5e-06,
      "loss": 0.5936,
      "step": 210
    },
    {
      "epoch": 0.5597964376590331,
      "grad_norm": 0.510125994682312,
      "learning_rate": 5e-06,
      "loss": 0.5804,
      "step": 220
    },
    {
      "epoch": 0.5852417302798982,
      "grad_norm": 0.532505214214325,
      "learning_rate": 5e-06,
      "loss": 0.583,
      "step": 230
    },
    {
      "epoch": 0.6106870229007634,
      "grad_norm": 0.732284426689148,
      "learning_rate": 5e-06,
      "loss": 0.5817,
      "step": 240
    },
    {
      "epoch": 0.6361323155216285,
      "grad_norm": 0.624311089515686,
      "learning_rate": 5e-06,
      "loss": 0.5821,
      "step": 250
    },
    {
      "epoch": 0.6615776081424937,
      "grad_norm": 0.6264485120773315,
      "learning_rate": 5e-06,
      "loss": 0.5805,
      "step": 260
    },
    {
      "epoch": 0.6870229007633588,
      "grad_norm": 0.5334043502807617,
      "learning_rate": 5e-06,
      "loss": 0.5818,
      "step": 270
    },
    {
      "epoch": 0.712468193384224,
      "grad_norm": 0.5711916089057922,
      "learning_rate": 5e-06,
      "loss": 0.5768,
      "step": 280
    },
    {
      "epoch": 0.7379134860050891,
      "grad_norm": 0.4914567172527313,
      "learning_rate": 5e-06,
      "loss": 0.5733,
      "step": 290
    },
    {
      "epoch": 0.7633587786259542,
      "grad_norm": 0.4886472523212433,
      "learning_rate": 5e-06,
      "loss": 0.5827,
      "step": 300
    },
    {
      "epoch": 0.7888040712468194,
      "grad_norm": 0.5745688080787659,
      "learning_rate": 5e-06,
      "loss": 0.5756,
      "step": 310
    },
    {
      "epoch": 0.8142493638676844,
      "grad_norm": 0.6722710132598877,
      "learning_rate": 5e-06,
      "loss": 0.5722,
      "step": 320
    },
    {
      "epoch": 0.8396946564885496,
      "grad_norm": 0.5262378454208374,
      "learning_rate": 5e-06,
      "loss": 0.5704,
      "step": 330
    },
    {
      "epoch": 0.8651399491094147,
      "grad_norm": 0.5720959305763245,
      "learning_rate": 5e-06,
      "loss": 0.5739,
      "step": 340
    },
    {
      "epoch": 0.8905852417302799,
      "grad_norm": 0.525618314743042,
      "learning_rate": 5e-06,
      "loss": 0.5648,
      "step": 350
    },
    {
      "epoch": 0.916030534351145,
      "grad_norm": 0.5551468133926392,
      "learning_rate": 5e-06,
      "loss": 0.5726,
      "step": 360
    },
    {
      "epoch": 0.9414758269720102,
      "grad_norm": 0.5689135193824768,
      "learning_rate": 5e-06,
      "loss": 0.5644,
      "step": 370
    },
    {
      "epoch": 0.9669211195928753,
      "grad_norm": 0.5880911946296692,
      "learning_rate": 5e-06,
      "loss": 0.5628,
      "step": 380
    },
    {
      "epoch": 0.9923664122137404,
      "grad_norm": 0.5270939469337463,
      "learning_rate": 5e-06,
      "loss": 0.5649,
      "step": 390
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5690909624099731,
      "eval_runtime": 35.8687,
      "eval_samples_per_second": 294.602,
      "eval_steps_per_second": 1.171,
      "step": 393
    },
    {
      "epoch": 1.0178117048346056,
      "grad_norm": 0.6128649115562439,
      "learning_rate": 5e-06,
      "loss": 0.5446,
      "step": 400
    },
    {
      "epoch": 1.0432569974554706,
      "grad_norm": 0.6446624398231506,
      "learning_rate": 5e-06,
      "loss": 0.5287,
      "step": 410
    },
    {
      "epoch": 1.0687022900763359,
      "grad_norm": 0.6626042127609253,
      "learning_rate": 5e-06,
      "loss": 0.528,
      "step": 420
    },
    {
      "epoch": 1.094147582697201,
      "grad_norm": 0.6851757168769836,
      "learning_rate": 5e-06,
      "loss": 0.5275,
      "step": 430
    },
    {
      "epoch": 1.1195928753180662,
      "grad_norm": 0.5607665181159973,
      "learning_rate": 5e-06,
      "loss": 0.5317,
      "step": 440
    },
    {
      "epoch": 1.1450381679389312,
      "grad_norm": 0.5421668887138367,
      "learning_rate": 5e-06,
      "loss": 0.5279,
      "step": 450
    },
    {
      "epoch": 1.1704834605597965,
      "grad_norm": 0.5087480545043945,
      "learning_rate": 5e-06,
      "loss": 0.5323,
      "step": 460
    },
    {
      "epoch": 1.1959287531806615,
      "grad_norm": 0.6041454672813416,
      "learning_rate": 5e-06,
      "loss": 0.5221,
      "step": 470
    },
    {
      "epoch": 1.2213740458015268,
      "grad_norm": 0.5389483571052551,
      "learning_rate": 5e-06,
      "loss": 0.5296,
      "step": 480
    },
    {
      "epoch": 1.2468193384223918,
      "grad_norm": 0.5498439073562622,
      "learning_rate": 5e-06,
      "loss": 0.5236,
      "step": 490
    },
    {
      "epoch": 1.272264631043257,
      "grad_norm": 0.534608781337738,
      "learning_rate": 5e-06,
      "loss": 0.5274,
      "step": 500
    },
    {
      "epoch": 1.297709923664122,
      "grad_norm": 0.5173637866973877,
      "learning_rate": 5e-06,
      "loss": 0.5234,
      "step": 510
    },
    {
      "epoch": 1.3231552162849873,
      "grad_norm": 0.640381932258606,
      "learning_rate": 5e-06,
      "loss": 0.5191,
      "step": 520
    },
    {
      "epoch": 1.3486005089058524,
      "grad_norm": 0.636634111404419,
      "learning_rate": 5e-06,
      "loss": 0.5274,
      "step": 530
    },
    {
      "epoch": 1.3740458015267176,
      "grad_norm": 0.4825289249420166,
      "learning_rate": 5e-06,
      "loss": 0.5271,
      "step": 540
    },
    {
      "epoch": 1.3994910941475827,
      "grad_norm": 0.7268803715705872,
      "learning_rate": 5e-06,
      "loss": 0.5239,
      "step": 550
    },
    {
      "epoch": 1.424936386768448,
      "grad_norm": 0.5402505993843079,
      "learning_rate": 5e-06,
      "loss": 0.5212,
      "step": 560
    },
    {
      "epoch": 1.450381679389313,
      "grad_norm": 0.5150734186172485,
      "learning_rate": 5e-06,
      "loss": 0.5147,
      "step": 570
    },
    {
      "epoch": 1.4758269720101782,
      "grad_norm": 0.5088803768157959,
      "learning_rate": 5e-06,
      "loss": 0.5238,
      "step": 580
    },
    {
      "epoch": 1.5012722646310432,
      "grad_norm": 0.5912160277366638,
      "learning_rate": 5e-06,
      "loss": 0.5226,
      "step": 590
    },
    {
      "epoch": 1.5267175572519083,
      "grad_norm": 0.5235996842384338,
      "learning_rate": 5e-06,
      "loss": 0.5161,
      "step": 600
    },
    {
      "epoch": 1.5521628498727735,
      "grad_norm": 0.6075550317764282,
      "learning_rate": 5e-06,
      "loss": 0.5157,
      "step": 610
    },
    {
      "epoch": 1.5776081424936388,
      "grad_norm": 0.5467493534088135,
      "learning_rate": 5e-06,
      "loss": 0.5173,
      "step": 620
    },
    {
      "epoch": 1.6030534351145038,
      "grad_norm": 0.5767125487327576,
      "learning_rate": 5e-06,
      "loss": 0.518,
      "step": 630
    },
    {
      "epoch": 1.6284987277353689,
      "grad_norm": 0.6012499332427979,
      "learning_rate": 5e-06,
      "loss": 0.5136,
      "step": 640
    },
    {
      "epoch": 1.6539440203562341,
      "grad_norm": 0.5314013361930847,
      "learning_rate": 5e-06,
      "loss": 0.5175,
      "step": 650
    },
    {
      "epoch": 1.6793893129770994,
      "grad_norm": 0.6663753390312195,
      "learning_rate": 5e-06,
      "loss": 0.5211,
      "step": 660
    },
    {
      "epoch": 1.7048346055979644,
      "grad_norm": 0.6120356917381287,
      "learning_rate": 5e-06,
      "loss": 0.5228,
      "step": 670
    },
    {
      "epoch": 1.7302798982188294,
      "grad_norm": 0.6509705781936646,
      "learning_rate": 5e-06,
      "loss": 0.5177,
      "step": 680
    },
    {
      "epoch": 1.7557251908396947,
      "grad_norm": 0.5193184614181519,
      "learning_rate": 5e-06,
      "loss": 0.5147,
      "step": 690
    },
    {
      "epoch": 1.78117048346056,
      "grad_norm": 0.5160722136497498,
      "learning_rate": 5e-06,
      "loss": 0.5167,
      "step": 700
    },
    {
      "epoch": 1.806615776081425,
      "grad_norm": 0.6566229462623596,
      "learning_rate": 5e-06,
      "loss": 0.5116,
      "step": 710
    },
    {
      "epoch": 1.83206106870229,
      "grad_norm": 0.4844520688056946,
      "learning_rate": 5e-06,
      "loss": 0.5125,
      "step": 720
    },
    {
      "epoch": 1.8575063613231553,
      "grad_norm": 0.5609481334686279,
      "learning_rate": 5e-06,
      "loss": 0.5134,
      "step": 730
    },
    {
      "epoch": 1.8829516539440203,
      "grad_norm": 0.4838588535785675,
      "learning_rate": 5e-06,
      "loss": 0.5111,
      "step": 740
    },
    {
      "epoch": 1.9083969465648853,
      "grad_norm": 0.5089361667633057,
      "learning_rate": 5e-06,
      "loss": 0.5147,
      "step": 750
    },
    {
      "epoch": 1.9338422391857506,
      "grad_norm": 0.4935343861579895,
      "learning_rate": 5e-06,
      "loss": 0.5092,
      "step": 760
    },
    {
      "epoch": 1.9592875318066159,
      "grad_norm": 0.5261263251304626,
      "learning_rate": 5e-06,
      "loss": 0.5092,
      "step": 770
    },
    {
      "epoch": 1.984732824427481,
      "grad_norm": 0.5327720642089844,
      "learning_rate": 5e-06,
      "loss": 0.5093,
      "step": 780
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5446056127548218,
      "eval_runtime": 35.403,
      "eval_samples_per_second": 298.478,
      "eval_steps_per_second": 1.186,
      "step": 786
    },
    {
      "epoch": 2.010178117048346,
      "grad_norm": 0.7140676975250244,
      "learning_rate": 5e-06,
      "loss": 0.4964,
      "step": 790
    },
    {
      "epoch": 2.035623409669211,
      "grad_norm": 0.5526481866836548,
      "learning_rate": 5e-06,
      "loss": 0.4699,
      "step": 800
    },
    {
      "epoch": 2.0610687022900764,
      "grad_norm": 0.8097620010375977,
      "learning_rate": 5e-06,
      "loss": 0.4747,
      "step": 810
    },
    {
      "epoch": 2.0865139949109412,
      "grad_norm": 0.5774556994438171,
      "learning_rate": 5e-06,
      "loss": 0.4731,
      "step": 820
    },
    {
      "epoch": 2.1119592875318065,
      "grad_norm": 0.7799936532974243,
      "learning_rate": 5e-06,
      "loss": 0.481,
      "step": 830
    },
    {
      "epoch": 2.1374045801526718,
      "grad_norm": 0.7324852347373962,
      "learning_rate": 5e-06,
      "loss": 0.472,
      "step": 840
    },
    {
      "epoch": 2.162849872773537,
      "grad_norm": 0.6810862421989441,
      "learning_rate": 5e-06,
      "loss": 0.4788,
      "step": 850
    },
    {
      "epoch": 2.188295165394402,
      "grad_norm": 0.6535518169403076,
      "learning_rate": 5e-06,
      "loss": 0.4787,
      "step": 860
    },
    {
      "epoch": 2.213740458015267,
      "grad_norm": 0.5570102334022522,
      "learning_rate": 5e-06,
      "loss": 0.4785,
      "step": 870
    },
    {
      "epoch": 2.2391857506361323,
      "grad_norm": 0.5471156239509583,
      "learning_rate": 5e-06,
      "loss": 0.4817,
      "step": 880
    },
    {
      "epoch": 2.2646310432569976,
      "grad_norm": 0.5434932112693787,
      "learning_rate": 5e-06,
      "loss": 0.4732,
      "step": 890
    },
    {
      "epoch": 2.2900763358778624,
      "grad_norm": 0.5234562754631042,
      "learning_rate": 5e-06,
      "loss": 0.4742,
      "step": 900
    },
    {
      "epoch": 2.3155216284987277,
      "grad_norm": 0.5137863159179688,
      "learning_rate": 5e-06,
      "loss": 0.4725,
      "step": 910
    },
    {
      "epoch": 2.340966921119593,
      "grad_norm": 0.5251743197441101,
      "learning_rate": 5e-06,
      "loss": 0.4793,
      "step": 920
    },
    {
      "epoch": 2.366412213740458,
      "grad_norm": 0.5063709020614624,
      "learning_rate": 5e-06,
      "loss": 0.4735,
      "step": 930
    },
    {
      "epoch": 2.391857506361323,
      "grad_norm": 0.5757549405097961,
      "learning_rate": 5e-06,
      "loss": 0.4728,
      "step": 940
    },
    {
      "epoch": 2.4173027989821882,
      "grad_norm": 0.48364320397377014,
      "learning_rate": 5e-06,
      "loss": 0.4785,
      "step": 950
    },
    {
      "epoch": 2.4427480916030535,
      "grad_norm": 0.5600482225418091,
      "learning_rate": 5e-06,
      "loss": 0.4748,
      "step": 960
    },
    {
      "epoch": 2.4681933842239188,
      "grad_norm": 0.5102260708808899,
      "learning_rate": 5e-06,
      "loss": 0.4752,
      "step": 970
    },
    {
      "epoch": 2.4936386768447836,
      "grad_norm": 0.642202615737915,
      "learning_rate": 5e-06,
      "loss": 0.4743,
      "step": 980
    },
    {
      "epoch": 2.519083969465649,
      "grad_norm": 0.7704691886901855,
      "learning_rate": 5e-06,
      "loss": 0.4743,
      "step": 990
    },
    {
      "epoch": 2.544529262086514,
      "grad_norm": 0.6885623931884766,
      "learning_rate": 5e-06,
      "loss": 0.4742,
      "step": 1000
    },
    {
      "epoch": 2.569974554707379,
      "grad_norm": 0.5846525430679321,
      "learning_rate": 5e-06,
      "loss": 0.4734,
      "step": 1010
    },
    {
      "epoch": 2.595419847328244,
      "grad_norm": 0.66885906457901,
      "learning_rate": 5e-06,
      "loss": 0.4758,
      "step": 1020
    },
    {
      "epoch": 2.6208651399491094,
      "grad_norm": 0.5633515119552612,
      "learning_rate": 5e-06,
      "loss": 0.474,
      "step": 1030
    },
    {
      "epoch": 2.6463104325699747,
      "grad_norm": 0.5127110481262207,
      "learning_rate": 5e-06,
      "loss": 0.4792,
      "step": 1040
    },
    {
      "epoch": 2.67175572519084,
      "grad_norm": 0.6211147308349609,
      "learning_rate": 5e-06,
      "loss": 0.4721,
      "step": 1050
    },
    {
      "epoch": 2.6972010178117047,
      "grad_norm": 0.700775146484375,
      "learning_rate": 5e-06,
      "loss": 0.4731,
      "step": 1060
    },
    {
      "epoch": 2.72264631043257,
      "grad_norm": 0.5291451215744019,
      "learning_rate": 5e-06,
      "loss": 0.475,
      "step": 1070
    },
    {
      "epoch": 2.7480916030534353,
      "grad_norm": 0.5429026484489441,
      "learning_rate": 5e-06,
      "loss": 0.4822,
      "step": 1080
    },
    {
      "epoch": 2.7735368956743,
      "grad_norm": 0.5198021531105042,
      "learning_rate": 5e-06,
      "loss": 0.4738,
      "step": 1090
    },
    {
      "epoch": 2.7989821882951653,
      "grad_norm": 0.4582120478153229,
      "learning_rate": 5e-06,
      "loss": 0.4709,
      "step": 1100
    },
    {
      "epoch": 2.8244274809160306,
      "grad_norm": 0.47587454319000244,
      "learning_rate": 5e-06,
      "loss": 0.4692,
      "step": 1110
    },
    {
      "epoch": 2.849872773536896,
      "grad_norm": 0.5221173167228699,
      "learning_rate": 5e-06,
      "loss": 0.4767,
      "step": 1120
    },
    {
      "epoch": 2.875318066157761,
      "grad_norm": 0.5617064833641052,
      "learning_rate": 5e-06,
      "loss": 0.478,
      "step": 1130
    },
    {
      "epoch": 2.900763358778626,
      "grad_norm": 0.5362435579299927,
      "learning_rate": 5e-06,
      "loss": 0.4745,
      "step": 1140
    },
    {
      "epoch": 2.926208651399491,
      "grad_norm": 0.5266700387001038,
      "learning_rate": 5e-06,
      "loss": 0.4725,
      "step": 1150
    },
    {
      "epoch": 2.9516539440203564,
      "grad_norm": 0.5272775292396545,
      "learning_rate": 5e-06,
      "loss": 0.4744,
      "step": 1160
    },
    {
      "epoch": 2.9770992366412212,
      "grad_norm": 0.6244516372680664,
      "learning_rate": 5e-06,
      "loss": 0.4766,
      "step": 1170
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5409571528434753,
      "eval_runtime": 35.9061,
      "eval_samples_per_second": 294.296,
      "eval_steps_per_second": 1.17,
      "step": 1179
    },
    {
      "epoch": 3.0,
      "step": 1179,
      "total_flos": 5.566873103382662e+19,
      "train_loss": 0.5357507543102779,
      "train_runtime": 8300.7415,
      "train_samples_per_second": 72.558,
      "train_steps_per_second": 0.142
    }
  ],
  "logging_steps": 10,
  "max_steps": 1179,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.566873103382662e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}