{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1170,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02564102564102564,
      "grad_norm": 3.406068801879883,
      "learning_rate": 5e-06,
      "loss": 0.9168,
      "step": 10
    },
    {
      "epoch": 0.05128205128205128,
      "grad_norm": 3.5034339427948,
      "learning_rate": 5e-06,
      "loss": 0.7688,
      "step": 20
    },
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 1.0085859298706055,
      "learning_rate": 5e-06,
      "loss": 0.729,
      "step": 30
    },
    {
      "epoch": 0.10256410256410256,
      "grad_norm": 0.8697875142097473,
      "learning_rate": 5e-06,
      "loss": 0.7004,
      "step": 40
    },
    {
      "epoch": 0.1282051282051282,
      "grad_norm": 0.9098314642906189,
      "learning_rate": 5e-06,
      "loss": 0.6853,
      "step": 50
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.240673542022705,
      "learning_rate": 5e-06,
      "loss": 0.6668,
      "step": 60
    },
    {
      "epoch": 0.1794871794871795,
      "grad_norm": 1.3245731592178345,
      "learning_rate": 5e-06,
      "loss": 0.654,
      "step": 70
    },
    {
      "epoch": 0.20512820512820512,
      "grad_norm": 0.6942853331565857,
      "learning_rate": 5e-06,
      "loss": 0.6427,
      "step": 80
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 0.7867761254310608,
      "learning_rate": 5e-06,
      "loss": 0.6425,
      "step": 90
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 0.7657337784767151,
      "learning_rate": 5e-06,
      "loss": 0.6305,
      "step": 100
    },
    {
      "epoch": 0.28205128205128205,
      "grad_norm": 0.8531173467636108,
      "learning_rate": 5e-06,
      "loss": 0.6308,
      "step": 110
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.6474210023880005,
      "learning_rate": 5e-06,
      "loss": 0.6227,
      "step": 120
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.6232497692108154,
      "learning_rate": 5e-06,
      "loss": 0.6293,
      "step": 130
    },
    {
      "epoch": 0.358974358974359,
      "grad_norm": 0.6542862057685852,
      "learning_rate": 5e-06,
      "loss": 0.6217,
      "step": 140
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 0.9984610080718994,
      "learning_rate": 5e-06,
      "loss": 0.6213,
      "step": 150
    },
    {
      "epoch": 0.41025641025641024,
      "grad_norm": 0.7436656355857849,
      "learning_rate": 5e-06,
      "loss": 0.614,
      "step": 160
    },
    {
      "epoch": 0.4358974358974359,
      "grad_norm": 0.7107124328613281,
      "learning_rate": 5e-06,
      "loss": 0.6057,
      "step": 170
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.8069647550582886,
      "learning_rate": 5e-06,
      "loss": 0.6134,
      "step": 180
    },
    {
      "epoch": 0.48717948717948717,
      "grad_norm": 0.5554560422897339,
      "learning_rate": 5e-06,
      "loss": 0.6069,
      "step": 190
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 0.5725631713867188,
      "learning_rate": 5e-06,
      "loss": 0.6081,
      "step": 200
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 0.7718241214752197,
      "learning_rate": 5e-06,
      "loss": 0.6003,
      "step": 210
    },
    {
      "epoch": 0.5641025641025641,
      "grad_norm": 0.659430742263794,
      "learning_rate": 5e-06,
      "loss": 0.6044,
      "step": 220
    },
    {
      "epoch": 0.5897435897435898,
      "grad_norm": 0.5641574859619141,
      "learning_rate": 5e-06,
      "loss": 0.5988,
      "step": 230
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.613983154296875,
      "learning_rate": 5e-06,
      "loss": 0.6079,
      "step": 240
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 0.7252273559570312,
      "learning_rate": 5e-06,
      "loss": 0.5981,
      "step": 250
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.6930380463600159,
      "learning_rate": 5e-06,
      "loss": 0.5927,
      "step": 260
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 0.6013543605804443,
      "learning_rate": 5e-06,
      "loss": 0.5942,
      "step": 270
    },
    {
      "epoch": 0.717948717948718,
      "grad_norm": 0.5443430542945862,
      "learning_rate": 5e-06,
      "loss": 0.594,
      "step": 280
    },
    {
      "epoch": 0.7435897435897436,
      "grad_norm": 0.6630271077156067,
      "learning_rate": 5e-06,
      "loss": 0.5926,
      "step": 290
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.6013346314430237,
      "learning_rate": 5e-06,
      "loss": 0.5874,
      "step": 300
    },
    {
      "epoch": 0.7948717948717948,
      "grad_norm": 0.5420301556587219,
      "learning_rate": 5e-06,
      "loss": 0.5913,
      "step": 310
    },
    {
      "epoch": 0.8205128205128205,
      "grad_norm": 0.809611976146698,
      "learning_rate": 5e-06,
      "loss": 0.5917,
      "step": 320
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 0.7773135304450989,
      "learning_rate": 5e-06,
      "loss": 0.5836,
      "step": 330
    },
    {
      "epoch": 0.8717948717948718,
      "grad_norm": 0.5458276271820068,
      "learning_rate": 5e-06,
      "loss": 0.5794,
      "step": 340
    },
    {
      "epoch": 0.8974358974358975,
      "grad_norm": 0.6851891875267029,
      "learning_rate": 5e-06,
      "loss": 0.5902,
      "step": 350
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.5757733583450317,
      "learning_rate": 5e-06,
      "loss": 0.5826,
      "step": 360
    },
    {
      "epoch": 0.9487179487179487,
      "grad_norm": 0.5545377135276794,
      "learning_rate": 5e-06,
      "loss": 0.5887,
      "step": 370
    },
    {
      "epoch": 0.9743589743589743,
      "grad_norm": 0.5440164804458618,
      "learning_rate": 5e-06,
      "loss": 0.584,
      "step": 380
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5356547236442566,
      "learning_rate": 5e-06,
      "loss": 0.5779,
      "step": 390
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5792508721351624,
      "eval_runtime": 37.2797,
      "eval_samples_per_second": 281.762,
      "eval_steps_per_second": 1.127,
      "step": 390
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 0.6288334131240845,
      "learning_rate": 5e-06,
      "loss": 0.5453,
      "step": 400
    },
    {
      "epoch": 1.0512820512820513,
      "grad_norm": 0.6558266282081604,
      "learning_rate": 5e-06,
      "loss": 0.5475,
      "step": 410
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.7556455135345459,
      "learning_rate": 5e-06,
      "loss": 0.5465,
      "step": 420
    },
    {
      "epoch": 1.1025641025641026,
      "grad_norm": 0.586370050907135,
      "learning_rate": 5e-06,
      "loss": 0.5395,
      "step": 430
    },
    {
      "epoch": 1.1282051282051282,
      "grad_norm": 0.5581791996955872,
      "learning_rate": 5e-06,
      "loss": 0.5366,
      "step": 440
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 0.5495139956474304,
      "learning_rate": 5e-06,
      "loss": 0.5487,
      "step": 450
    },
    {
      "epoch": 1.1794871794871795,
      "grad_norm": 0.6099910736083984,
      "learning_rate": 5e-06,
      "loss": 0.5397,
      "step": 460
    },
    {
      "epoch": 1.205128205128205,
      "grad_norm": 0.5565390586853027,
      "learning_rate": 5e-06,
      "loss": 0.5376,
      "step": 470
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.6461251378059387,
      "learning_rate": 5e-06,
      "loss": 0.534,
      "step": 480
    },
    {
      "epoch": 1.2564102564102564,
      "grad_norm": 0.6213774681091309,
      "learning_rate": 5e-06,
      "loss": 0.5366,
      "step": 490
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 0.5515546798706055,
      "learning_rate": 5e-06,
      "loss": 0.5381,
      "step": 500
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.6342268586158752,
      "learning_rate": 5e-06,
      "loss": 0.536,
      "step": 510
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.5595047473907471,
      "learning_rate": 5e-06,
      "loss": 0.5396,
      "step": 520
    },
    {
      "epoch": 1.358974358974359,
      "grad_norm": 0.5788885951042175,
      "learning_rate": 5e-06,
      "loss": 0.5371,
      "step": 530
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.5231652855873108,
      "learning_rate": 5e-06,
      "loss": 0.5357,
      "step": 540
    },
    {
      "epoch": 1.4102564102564101,
      "grad_norm": 0.7670683860778809,
      "learning_rate": 5e-06,
      "loss": 0.5376,
      "step": 550
    },
    {
      "epoch": 1.435897435897436,
      "grad_norm": 0.5181464552879333,
      "learning_rate": 5e-06,
      "loss": 0.5321,
      "step": 560
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 0.5286003351211548,
      "learning_rate": 5e-06,
      "loss": 0.5361,
      "step": 570
    },
    {
      "epoch": 1.4871794871794872,
      "grad_norm": 0.5364812016487122,
      "learning_rate": 5e-06,
      "loss": 0.5391,
      "step": 580
    },
    {
      "epoch": 1.5128205128205128,
      "grad_norm": 0.6188532710075378,
      "learning_rate": 5e-06,
      "loss": 0.5336,
      "step": 590
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.4892689883708954,
      "learning_rate": 5e-06,
      "loss": 0.5352,
      "step": 600
    },
    {
      "epoch": 1.564102564102564,
      "grad_norm": 0.5404101014137268,
      "learning_rate": 5e-06,
      "loss": 0.5321,
      "step": 610
    },
    {
      "epoch": 1.5897435897435899,
      "grad_norm": 0.5570105314254761,
      "learning_rate": 5e-06,
      "loss": 0.5341,
      "step": 620
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.6178380846977234,
      "learning_rate": 5e-06,
      "loss": 0.5321,
      "step": 630
    },
    {
      "epoch": 1.641025641025641,
      "grad_norm": 0.8673184514045715,
      "learning_rate": 5e-06,
      "loss": 0.5216,
      "step": 640
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.6193057298660278,
      "learning_rate": 5e-06,
      "loss": 0.5312,
      "step": 650
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.5737960934638977,
      "learning_rate": 5e-06,
      "loss": 0.5317,
      "step": 660
    },
    {
      "epoch": 1.717948717948718,
      "grad_norm": 0.6953139305114746,
      "learning_rate": 5e-06,
      "loss": 0.5297,
      "step": 670
    },
    {
      "epoch": 1.7435897435897436,
      "grad_norm": 0.5698896050453186,
      "learning_rate": 5e-06,
      "loss": 0.5248,
      "step": 680
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 0.4578630328178406,
      "learning_rate": 5e-06,
      "loss": 0.5251,
      "step": 690
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 0.4654785692691803,
      "learning_rate": 5e-06,
      "loss": 0.5308,
      "step": 700
    },
    {
      "epoch": 1.8205128205128205,
      "grad_norm": 0.500807523727417,
      "learning_rate": 5e-06,
      "loss": 0.5259,
      "step": 710
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.5273367762565613,
      "learning_rate": 5e-06,
      "loss": 0.5286,
      "step": 720
    },
    {
      "epoch": 1.8717948717948718,
      "grad_norm": 0.49950945377349854,
      "learning_rate": 5e-06,
      "loss": 0.5292,
      "step": 730
    },
    {
      "epoch": 1.8974358974358974,
      "grad_norm": 0.46231916546821594,
      "learning_rate": 5e-06,
      "loss": 0.5243,
      "step": 740
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.5422210693359375,
      "learning_rate": 5e-06,
      "loss": 0.5241,
      "step": 750
    },
    {
      "epoch": 1.9487179487179487,
      "grad_norm": 0.569016695022583,
      "learning_rate": 5e-06,
      "loss": 0.5237,
      "step": 760
    },
    {
      "epoch": 1.9743589743589745,
      "grad_norm": 0.5401661396026611,
      "learning_rate": 5e-06,
      "loss": 0.5289,
      "step": 770
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6024136543273926,
      "learning_rate": 5e-06,
      "loss": 0.5262,
      "step": 780
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5533561706542969,
      "eval_runtime": 36.5672,
      "eval_samples_per_second": 287.252,
      "eval_steps_per_second": 1.149,
      "step": 780
    },
    {
      "epoch": 2.0256410256410255,
      "grad_norm": 0.5916836857795715,
      "learning_rate": 5e-06,
      "loss": 0.4847,
      "step": 790
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 0.6934983730316162,
      "learning_rate": 5e-06,
      "loss": 0.4906,
      "step": 800
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.546357274055481,
      "learning_rate": 5e-06,
      "loss": 0.4857,
      "step": 810
    },
    {
      "epoch": 2.1025641025641026,
      "grad_norm": 0.7491461634635925,
      "learning_rate": 5e-06,
      "loss": 0.4861,
      "step": 820
    },
    {
      "epoch": 2.128205128205128,
      "grad_norm": 0.5870935320854187,
      "learning_rate": 5e-06,
      "loss": 0.4847,
      "step": 830
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.6286914348602295,
      "learning_rate": 5e-06,
      "loss": 0.4895,
      "step": 840
    },
    {
      "epoch": 2.1794871794871793,
      "grad_norm": 0.515387773513794,
      "learning_rate": 5e-06,
      "loss": 0.4897,
      "step": 850
    },
    {
      "epoch": 2.2051282051282053,
      "grad_norm": 0.5456680655479431,
      "learning_rate": 5e-06,
      "loss": 0.4861,
      "step": 860
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 0.49457597732543945,
      "learning_rate": 5e-06,
      "loss": 0.484,
      "step": 870
    },
    {
      "epoch": 2.2564102564102564,
      "grad_norm": 0.5255215167999268,
      "learning_rate": 5e-06,
      "loss": 0.4936,
      "step": 880
    },
    {
      "epoch": 2.282051282051282,
      "grad_norm": 0.5122373700141907,
      "learning_rate": 5e-06,
      "loss": 0.4915,
      "step": 890
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.5085675120353699,
      "learning_rate": 5e-06,
      "loss": 0.4894,
      "step": 900
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.5925104022026062,
      "learning_rate": 5e-06,
      "loss": 0.4873,
      "step": 910
    },
    {
      "epoch": 2.358974358974359,
      "grad_norm": 0.519869327545166,
      "learning_rate": 5e-06,
      "loss": 0.4882,
      "step": 920
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 0.550400972366333,
      "learning_rate": 5e-06,
      "loss": 0.486,
      "step": 930
    },
    {
      "epoch": 2.41025641025641,
      "grad_norm": 0.6560305953025818,
      "learning_rate": 5e-06,
      "loss": 0.4963,
      "step": 940
    },
    {
      "epoch": 2.435897435897436,
      "grad_norm": 0.5110371112823486,
      "learning_rate": 5e-06,
      "loss": 0.4855,
      "step": 950
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.48929715156555176,
      "learning_rate": 5e-06,
      "loss": 0.4859,
      "step": 960
    },
    {
      "epoch": 2.4871794871794872,
      "grad_norm": 0.5958628058433533,
      "learning_rate": 5e-06,
      "loss": 0.4884,
      "step": 970
    },
    {
      "epoch": 2.5128205128205128,
      "grad_norm": 0.604121208190918,
      "learning_rate": 5e-06,
      "loss": 0.4858,
      "step": 980
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 0.5479825735092163,
      "learning_rate": 5e-06,
      "loss": 0.4864,
      "step": 990
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 0.5345059633255005,
      "learning_rate": 5e-06,
      "loss": 0.4859,
      "step": 1000
    },
    {
      "epoch": 2.58974358974359,
      "grad_norm": 0.519960880279541,
      "learning_rate": 5e-06,
      "loss": 0.4867,
      "step": 1010
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.5771586298942566,
      "learning_rate": 5e-06,
      "loss": 0.4921,
      "step": 1020
    },
    {
      "epoch": 2.641025641025641,
      "grad_norm": 0.5097467303276062,
      "learning_rate": 5e-06,
      "loss": 0.4898,
      "step": 1030
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.5330426096916199,
      "learning_rate": 5e-06,
      "loss": 0.487,
      "step": 1040
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.6157095432281494,
      "learning_rate": 5e-06,
      "loss": 0.4896,
      "step": 1050
    },
    {
      "epoch": 2.717948717948718,
      "grad_norm": 0.5218502283096313,
      "learning_rate": 5e-06,
      "loss": 0.4883,
      "step": 1060
    },
    {
      "epoch": 2.7435897435897436,
      "grad_norm": 0.620424747467041,
      "learning_rate": 5e-06,
      "loss": 0.4895,
      "step": 1070
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.5703057646751404,
      "learning_rate": 5e-06,
      "loss": 0.4854,
      "step": 1080
    },
    {
      "epoch": 2.7948717948717947,
      "grad_norm": 0.49105432629585266,
      "learning_rate": 5e-06,
      "loss": 0.4929,
      "step": 1090
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 0.5956032872200012,
      "learning_rate": 5e-06,
      "loss": 0.4847,
      "step": 1100
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.6310620307922363,
      "learning_rate": 5e-06,
      "loss": 0.4824,
      "step": 1110
    },
    {
      "epoch": 2.871794871794872,
      "grad_norm": 0.5469298958778381,
      "learning_rate": 5e-06,
      "loss": 0.4843,
      "step": 1120
    },
    {
      "epoch": 2.8974358974358974,
      "grad_norm": 0.6324126720428467,
      "learning_rate": 5e-06,
      "loss": 0.4822,
      "step": 1130
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.5963416695594788,
      "learning_rate": 5e-06,
      "loss": 0.4856,
      "step": 1140
    },
    {
      "epoch": 2.948717948717949,
      "grad_norm": 0.5227168202400208,
      "learning_rate": 5e-06,
      "loss": 0.4888,
      "step": 1150
    },
    {
      "epoch": 2.9743589743589745,
      "grad_norm": 0.4714514911174774,
      "learning_rate": 5e-06,
      "loss": 0.4913,
      "step": 1160
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5580222010612488,
      "learning_rate": 5e-06,
      "loss": 0.4851,
      "step": 1170
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5471954345703125,
      "eval_runtime": 35.7738,
      "eval_samples_per_second": 293.622,
      "eval_steps_per_second": 1.174,
      "step": 1170
    },
    {
      "epoch": 3.0,
      "step": 1170,
      "total_flos": 5.524377893797154e+19,
      "train_loss": 0.5494362928928473,
      "train_runtime": 8709.0296,
      "train_samples_per_second": 68.748,
      "train_steps_per_second": 0.134
    }
  ],
  "logging_steps": 10,
  "max_steps": 1170,
  "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": 5.524377893797154e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}