{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 3063,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0032653061224489797,
      "grad_norm": 1.763192892074585,
      "learning_rate": 0.0005,
      "loss": 6.1164,
      "step": 10
    },
    {
      "epoch": 0.006530612244897959,
      "grad_norm": 1.1696356534957886,
      "learning_rate": 0.0005,
      "loss": 5.8499,
      "step": 20
    },
    {
      "epoch": 0.009795918367346938,
      "grad_norm": 1.1040449142456055,
      "learning_rate": 0.0005,
      "loss": 5.6782,
      "step": 30
    },
    {
      "epoch": 0.013061224489795919,
      "grad_norm": 1.3148446083068848,
      "learning_rate": 0.0005,
      "loss": 5.6237,
      "step": 40
    },
    {
      "epoch": 0.0163265306122449,
      "grad_norm": 1.438623070716858,
      "learning_rate": 0.0005,
      "loss": 5.5178,
      "step": 50
    },
    {
      "epoch": 0.019591836734693877,
      "grad_norm": 1.2186000347137451,
      "learning_rate": 0.0005,
      "loss": 5.5895,
      "step": 60
    },
    {
      "epoch": 0.022857142857142857,
      "grad_norm": 1.088930606842041,
      "learning_rate": 0.0005,
      "loss": 5.4889,
      "step": 70
    },
    {
      "epoch": 0.026122448979591838,
      "grad_norm": 1.1969929933547974,
      "learning_rate": 0.0005,
      "loss": 5.3515,
      "step": 80
    },
    {
      "epoch": 0.029387755102040815,
      "grad_norm": 1.0405751466751099,
      "learning_rate": 0.0005,
      "loss": 5.2943,
      "step": 90
    },
    {
      "epoch": 0.0326530612244898,
      "grad_norm": 1.6840550899505615,
      "learning_rate": 0.0005,
      "loss": 5.1347,
      "step": 100
    },
    {
      "epoch": 0.035918367346938776,
      "grad_norm": 1.0448979139328003,
      "learning_rate": 0.0005,
      "loss": 5.2275,
      "step": 110
    },
    {
      "epoch": 0.03918367346938775,
      "grad_norm": 1.2239587306976318,
      "learning_rate": 0.0005,
      "loss": 5.1866,
      "step": 120
    },
    {
      "epoch": 0.04244897959183674,
      "grad_norm": 1.0111207962036133,
      "learning_rate": 0.0005,
      "loss": 5.1201,
      "step": 130
    },
    {
      "epoch": 0.045714285714285714,
      "grad_norm": 1.0412639379501343,
      "learning_rate": 0.0005,
      "loss": 5.0317,
      "step": 140
    },
    {
      "epoch": 0.04897959183673469,
      "grad_norm": 1.6490758657455444,
      "learning_rate": 0.0005,
      "loss": 4.8709,
      "step": 150
    },
    {
      "epoch": 0.052244897959183675,
      "grad_norm": 1.320185661315918,
      "learning_rate": 0.0005,
      "loss": 5.0482,
      "step": 160
    },
    {
      "epoch": 0.05551020408163265,
      "grad_norm": 1.23984694480896,
      "learning_rate": 0.0005,
      "loss": 5.1341,
      "step": 170
    },
    {
      "epoch": 0.05877551020408163,
      "grad_norm": 1.0305520296096802,
      "learning_rate": 0.0005,
      "loss": 4.9743,
      "step": 180
    },
    {
      "epoch": 0.062040816326530614,
      "grad_norm": 1.1690291166305542,
      "learning_rate": 0.0005,
      "loss": 4.8896,
      "step": 190
    },
    {
      "epoch": 0.0653061224489796,
      "grad_norm": 1.5160081386566162,
      "learning_rate": 0.0005,
      "loss": 4.7129,
      "step": 200
    },
    {
      "epoch": 0.06857142857142857,
      "grad_norm": 1.293174147605896,
      "learning_rate": 0.0005,
      "loss": 4.8936,
      "step": 210
    },
    {
      "epoch": 0.07183673469387755,
      "grad_norm": 0.9752352237701416,
      "learning_rate": 0.0005,
      "loss": 4.9842,
      "step": 220
    },
    {
      "epoch": 0.07510204081632653,
      "grad_norm": 1.2358481884002686,
      "learning_rate": 0.0005,
      "loss": 4.8956,
      "step": 230
    },
    {
      "epoch": 0.0783673469387755,
      "grad_norm": 1.4549710750579834,
      "learning_rate": 0.0005,
      "loss": 4.7804,
      "step": 240
    },
    {
      "epoch": 0.08163265306122448,
      "grad_norm": 1.6275160312652588,
      "learning_rate": 0.0005,
      "loss": 4.6489,
      "step": 250
    },
    {
      "epoch": 0.08489795918367347,
      "grad_norm": 1.3446155786514282,
      "learning_rate": 0.0005,
      "loss": 4.8603,
      "step": 260
    },
    {
      "epoch": 0.08816326530612245,
      "grad_norm": 1.3125636577606201,
      "learning_rate": 0.0005,
      "loss": 4.8614,
      "step": 270
    },
    {
      "epoch": 0.09142857142857143,
      "grad_norm": 1.5089269876480103,
      "learning_rate": 0.0005,
      "loss": 4.7528,
      "step": 280
    },
    {
      "epoch": 0.0946938775510204,
      "grad_norm": 1.240677833557129,
      "learning_rate": 0.0005,
      "loss": 4.6491,
      "step": 290
    },
    {
      "epoch": 0.09795918367346938,
      "grad_norm": 1.6580984592437744,
      "learning_rate": 0.0005,
      "loss": 4.6021,
      "step": 300
    },
    {
      "epoch": 0.10122448979591837,
      "grad_norm": 1.293299674987793,
      "learning_rate": 0.0005,
      "loss": 4.6556,
      "step": 310
    },
    {
      "epoch": 0.10448979591836735,
      "grad_norm": 1.0091382265090942,
      "learning_rate": 0.0005,
      "loss": 4.8806,
      "step": 320
    },
    {
      "epoch": 0.10775510204081633,
      "grad_norm": 1.2688143253326416,
      "learning_rate": 0.0005,
      "loss": 4.706,
      "step": 330
    },
    {
      "epoch": 0.1110204081632653,
      "grad_norm": 1.514876365661621,
      "learning_rate": 0.0005,
      "loss": 4.649,
      "step": 340
    },
    {
      "epoch": 0.11428571428571428,
      "grad_norm": 1.8324439525604248,
      "learning_rate": 0.0005,
      "loss": 4.4951,
      "step": 350
    },
    {
      "epoch": 0.11755102040816326,
      "grad_norm": 1.6239265203475952,
      "learning_rate": 0.0005,
      "loss": 4.5715,
      "step": 360
    },
    {
      "epoch": 0.12081632653061225,
      "grad_norm": 1.2813515663146973,
      "learning_rate": 0.0005,
      "loss": 4.8563,
      "step": 370
    },
    {
      "epoch": 0.12408163265306123,
      "grad_norm": 1.2220065593719482,
      "learning_rate": 0.0005,
      "loss": 4.6549,
      "step": 380
    },
    {
      "epoch": 0.1273469387755102,
      "grad_norm": 1.176604986190796,
      "learning_rate": 0.0005,
      "loss": 4.5817,
      "step": 390
    },
    {
      "epoch": 0.1306122448979592,
      "grad_norm": 1.7251240015029907,
      "learning_rate": 0.0005,
      "loss": 4.4422,
      "step": 400
    },
    {
      "epoch": 0.13387755102040816,
      "grad_norm": 1.2302827835083008,
      "learning_rate": 0.0005,
      "loss": 4.4658,
      "step": 410
    },
    {
      "epoch": 0.13714285714285715,
      "grad_norm": 1.0500273704528809,
      "learning_rate": 0.0005,
      "loss": 4.722,
      "step": 420
    },
    {
      "epoch": 0.1404081632653061,
      "grad_norm": 1.2972230911254883,
      "learning_rate": 0.0005,
      "loss": 4.6021,
      "step": 430
    },
    {
      "epoch": 0.1436734693877551,
      "grad_norm": 1.294978141784668,
      "learning_rate": 0.0005,
      "loss": 4.4754,
      "step": 440
    },
    {
      "epoch": 0.1469387755102041,
      "grad_norm": 1.6263995170593262,
      "learning_rate": 0.0005,
      "loss": 4.4881,
      "step": 450
    },
    {
      "epoch": 0.15020408163265306,
      "grad_norm": 1.2330529689788818,
      "learning_rate": 0.0005,
      "loss": 4.4441,
      "step": 460
    },
    {
      "epoch": 0.15346938775510205,
      "grad_norm": 0.8953180909156799,
      "learning_rate": 0.0005,
      "loss": 4.6281,
      "step": 470
    },
    {
      "epoch": 0.156734693877551,
      "grad_norm": 1.2303706407546997,
      "learning_rate": 0.0005,
      "loss": 4.5374,
      "step": 480
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.4355968236923218,
      "learning_rate": 0.0005,
      "loss": 4.5039,
      "step": 490
    },
    {
      "epoch": 0.16326530612244897,
      "grad_norm": 2.549136161804199,
      "learning_rate": 0.0005,
      "loss": 4.3848,
      "step": 500
    },
    {
      "epoch": 0.16326530612244897,
      "eval_loss": 4.471920967102051,
      "eval_runtime": 44.0079,
      "eval_samples_per_second": 45.446,
      "eval_steps_per_second": 11.362,
      "step": 500
    },
    {
      "epoch": 0.16653061224489796,
      "grad_norm": 1.3125051259994507,
      "learning_rate": 0.0005,
      "loss": 4.429,
      "step": 510
    },
    {
      "epoch": 0.16979591836734695,
      "grad_norm": 1.1771793365478516,
      "learning_rate": 0.0005,
      "loss": 4.6117,
      "step": 520
    },
    {
      "epoch": 0.1730612244897959,
      "grad_norm": 1.0817294120788574,
      "learning_rate": 0.0005,
      "loss": 4.5011,
      "step": 530
    },
    {
      "epoch": 0.1763265306122449,
      "grad_norm": 1.7551524639129639,
      "learning_rate": 0.0005,
      "loss": 4.4272,
      "step": 540
    },
    {
      "epoch": 0.17959183673469387,
      "grad_norm": 2.062394142150879,
      "learning_rate": 0.0005,
      "loss": 4.2264,
      "step": 550
    },
    {
      "epoch": 0.18285714285714286,
      "grad_norm": 1.150291919708252,
      "learning_rate": 0.0005,
      "loss": 4.2967,
      "step": 560
    },
    {
      "epoch": 0.18612244897959185,
      "grad_norm": 1.318264126777649,
      "learning_rate": 0.0005,
      "loss": 4.5895,
      "step": 570
    },
    {
      "epoch": 0.1893877551020408,
      "grad_norm": 1.267132043838501,
      "learning_rate": 0.0005,
      "loss": 4.4743,
      "step": 580
    },
    {
      "epoch": 0.1926530612244898,
      "grad_norm": 1.0571893453598022,
      "learning_rate": 0.0005,
      "loss": 4.3215,
      "step": 590
    },
    {
      "epoch": 0.19591836734693877,
      "grad_norm": 1.7111858129501343,
      "learning_rate": 0.0005,
      "loss": 4.3655,
      "step": 600
    },
    {
      "epoch": 0.19918367346938776,
      "grad_norm": 1.3666787147521973,
      "learning_rate": 0.0005,
      "loss": 4.2062,
      "step": 610
    },
    {
      "epoch": 0.20244897959183675,
      "grad_norm": 1.1153773069381714,
      "learning_rate": 0.0005,
      "loss": 4.5956,
      "step": 620
    },
    {
      "epoch": 0.2057142857142857,
      "grad_norm": 1.3033626079559326,
      "learning_rate": 0.0005,
      "loss": 4.3626,
      "step": 630
    },
    {
      "epoch": 0.2089795918367347,
      "grad_norm": 1.2020397186279297,
      "learning_rate": 0.0005,
      "loss": 4.2874,
      "step": 640
    },
    {
      "epoch": 0.21224489795918366,
      "grad_norm": 1.620437741279602,
      "learning_rate": 0.0005,
      "loss": 4.3094,
      "step": 650
    },
    {
      "epoch": 0.21551020408163266,
      "grad_norm": 1.3368499279022217,
      "learning_rate": 0.0005,
      "loss": 4.1476,
      "step": 660
    },
    {
      "epoch": 0.21877551020408162,
      "grad_norm": 1.3473259210586548,
      "learning_rate": 0.0005,
      "loss": 4.515,
      "step": 670
    },
    {
      "epoch": 0.2220408163265306,
      "grad_norm": 1.3251821994781494,
      "learning_rate": 0.0005,
      "loss": 4.2691,
      "step": 680
    },
    {
      "epoch": 0.2253061224489796,
      "grad_norm": 1.4138338565826416,
      "learning_rate": 0.0005,
      "loss": 4.2523,
      "step": 690
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 1.8546305894851685,
      "learning_rate": 0.0005,
      "loss": 4.2569,
      "step": 700
    },
    {
      "epoch": 0.23183673469387756,
      "grad_norm": 1.3483269214630127,
      "learning_rate": 0.0005,
      "loss": 4.0691,
      "step": 710
    },
    {
      "epoch": 0.23510204081632652,
      "grad_norm": 1.0672351121902466,
      "learning_rate": 0.0005,
      "loss": 4.4731,
      "step": 720
    },
    {
      "epoch": 0.2383673469387755,
      "grad_norm": 1.168258786201477,
      "learning_rate": 0.0005,
      "loss": 4.4081,
      "step": 730
    },
    {
      "epoch": 0.2416326530612245,
      "grad_norm": 1.1797970533370972,
      "learning_rate": 0.0005,
      "loss": 4.2101,
      "step": 740
    },
    {
      "epoch": 0.24489795918367346,
      "grad_norm": 1.774629831314087,
      "learning_rate": 0.0005,
      "loss": 4.1419,
      "step": 750
    },
    {
      "epoch": 0.24816326530612245,
      "grad_norm": 1.4174522161483765,
      "learning_rate": 0.0005,
      "loss": 4.0785,
      "step": 760
    },
    {
      "epoch": 0.25142857142857145,
      "grad_norm": 1.1591559648513794,
      "learning_rate": 0.0005,
      "loss": 4.4851,
      "step": 770
    },
    {
      "epoch": 0.2546938775510204,
      "grad_norm": 1.3229284286499023,
      "learning_rate": 0.0005,
      "loss": 4.2819,
      "step": 780
    },
    {
      "epoch": 0.25795918367346937,
      "grad_norm": 1.174056887626648,
      "learning_rate": 0.0005,
      "loss": 4.1234,
      "step": 790
    },
    {
      "epoch": 0.2612244897959184,
      "grad_norm": 1.6990975141525269,
      "learning_rate": 0.0005,
      "loss": 4.0965,
      "step": 800
    },
    {
      "epoch": 0.26448979591836735,
      "grad_norm": 1.4492194652557373,
      "learning_rate": 0.0005,
      "loss": 4.0653,
      "step": 810
    },
    {
      "epoch": 0.2677551020408163,
      "grad_norm": 0.9765328764915466,
      "learning_rate": 0.0005,
      "loss": 4.4531,
      "step": 820
    },
    {
      "epoch": 0.2710204081632653,
      "grad_norm": 1.4868978261947632,
      "learning_rate": 0.0005,
      "loss": 4.2941,
      "step": 830
    },
    {
      "epoch": 0.2742857142857143,
      "grad_norm": 1.3501718044281006,
      "learning_rate": 0.0005,
      "loss": 4.1963,
      "step": 840
    },
    {
      "epoch": 0.27755102040816326,
      "grad_norm": 1.9492390155792236,
      "learning_rate": 0.0005,
      "loss": 4.1563,
      "step": 850
    },
    {
      "epoch": 0.2808163265306122,
      "grad_norm": 1.2434947490692139,
      "learning_rate": 0.0005,
      "loss": 4.0553,
      "step": 860
    },
    {
      "epoch": 0.28408163265306124,
      "grad_norm": 1.1206755638122559,
      "learning_rate": 0.0005,
      "loss": 4.3843,
      "step": 870
    },
    {
      "epoch": 0.2873469387755102,
      "grad_norm": 1.1947944164276123,
      "learning_rate": 0.0005,
      "loss": 4.2625,
      "step": 880
    },
    {
      "epoch": 0.29061224489795917,
      "grad_norm": 1.1833423376083374,
      "learning_rate": 0.0005,
      "loss": 4.1207,
      "step": 890
    },
    {
      "epoch": 0.2938775510204082,
      "grad_norm": 1.6545424461364746,
      "learning_rate": 0.0005,
      "loss": 4.1006,
      "step": 900
    },
    {
      "epoch": 0.29714285714285715,
      "grad_norm": 1.1993494033813477,
      "learning_rate": 0.0005,
      "loss": 3.9992,
      "step": 910
    },
    {
      "epoch": 0.3004081632653061,
      "grad_norm": 1.017640233039856,
      "learning_rate": 0.0005,
      "loss": 4.3819,
      "step": 920
    },
    {
      "epoch": 0.3036734693877551,
      "grad_norm": 1.300307273864746,
      "learning_rate": 0.0005,
      "loss": 4.182,
      "step": 930
    },
    {
      "epoch": 0.3069387755102041,
      "grad_norm": 1.4982844591140747,
      "learning_rate": 0.0005,
      "loss": 4.0585,
      "step": 940
    },
    {
      "epoch": 0.31020408163265306,
      "grad_norm": 1.8079185485839844,
      "learning_rate": 0.0005,
      "loss": 4.0825,
      "step": 950
    },
    {
      "epoch": 0.313469387755102,
      "grad_norm": 1.2230055332183838,
      "learning_rate": 0.0005,
      "loss": 3.9335,
      "step": 960
    },
    {
      "epoch": 0.31673469387755104,
      "grad_norm": 1.372392177581787,
      "learning_rate": 0.0005,
      "loss": 4.3513,
      "step": 970
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.4195263385772705,
      "learning_rate": 0.0005,
      "loss": 4.1446,
      "step": 980
    },
    {
      "epoch": 0.32326530612244897,
      "grad_norm": 1.5404787063598633,
      "learning_rate": 0.0005,
      "loss": 4.0195,
      "step": 990
    },
    {
      "epoch": 0.32653061224489793,
      "grad_norm": 1.884207844734192,
      "learning_rate": 0.0005,
      "loss": 4.0313,
      "step": 1000
    },
    {
      "epoch": 0.32653061224489793,
      "eval_loss": 4.118765354156494,
      "eval_runtime": 43.1312,
      "eval_samples_per_second": 46.37,
      "eval_steps_per_second": 11.593,
      "step": 1000
    },
    {
      "epoch": 0.32979591836734695,
      "grad_norm": 1.2172777652740479,
      "learning_rate": 0.0005,
      "loss": 3.8635,
      "step": 1010
    },
    {
      "epoch": 0.3330612244897959,
      "grad_norm": 1.3385504484176636,
      "learning_rate": 0.0005,
      "loss": 4.2695,
      "step": 1020
    },
    {
      "epoch": 0.3363265306122449,
      "grad_norm": 1.2645728588104248,
      "learning_rate": 0.0005,
      "loss": 4.1298,
      "step": 1030
    },
    {
      "epoch": 0.3395918367346939,
      "grad_norm": 1.1611703634262085,
      "learning_rate": 0.0005,
      "loss": 4.0871,
      "step": 1040
    },
    {
      "epoch": 0.34285714285714286,
      "grad_norm": 1.9043705463409424,
      "learning_rate": 0.0005,
      "loss": 4.0687,
      "step": 1050
    },
    {
      "epoch": 0.3461224489795918,
      "grad_norm": 1.109143853187561,
      "learning_rate": 0.0005,
      "loss": 3.9311,
      "step": 1060
    },
    {
      "epoch": 0.34938775510204084,
      "grad_norm": 1.3506762981414795,
      "learning_rate": 0.0005,
      "loss": 4.2937,
      "step": 1070
    },
    {
      "epoch": 0.3526530612244898,
      "grad_norm": 1.5633002519607544,
      "learning_rate": 0.0005,
      "loss": 4.0552,
      "step": 1080
    },
    {
      "epoch": 0.35591836734693877,
      "grad_norm": 1.3422085046768188,
      "learning_rate": 0.0005,
      "loss": 4.1209,
      "step": 1090
    },
    {
      "epoch": 0.35918367346938773,
      "grad_norm": 1.8398027420043945,
      "learning_rate": 0.0005,
      "loss": 4.0818,
      "step": 1100
    },
    {
      "epoch": 0.36244897959183675,
      "grad_norm": 1.0788570642471313,
      "learning_rate": 0.0005,
      "loss": 3.8917,
      "step": 1110
    },
    {
      "epoch": 0.3657142857142857,
      "grad_norm": 1.3736050128936768,
      "learning_rate": 0.0005,
      "loss": 4.2967,
      "step": 1120
    },
    {
      "epoch": 0.3689795918367347,
      "grad_norm": 1.249338150024414,
      "learning_rate": 0.0005,
      "loss": 4.1155,
      "step": 1130
    },
    {
      "epoch": 0.3722448979591837,
      "grad_norm": 1.3408821821212769,
      "learning_rate": 0.0005,
      "loss": 4.0186,
      "step": 1140
    },
    {
      "epoch": 0.37551020408163266,
      "grad_norm": 1.4768186807632446,
      "learning_rate": 0.0005,
      "loss": 3.9932,
      "step": 1150
    },
    {
      "epoch": 0.3787755102040816,
      "grad_norm": 1.274889349937439,
      "learning_rate": 0.0005,
      "loss": 3.7688,
      "step": 1160
    },
    {
      "epoch": 0.3820408163265306,
      "grad_norm": 1.17601478099823,
      "learning_rate": 0.0005,
      "loss": 4.1915,
      "step": 1170
    },
    {
      "epoch": 0.3853061224489796,
      "grad_norm": 1.2624982595443726,
      "learning_rate": 0.0005,
      "loss": 4.0991,
      "step": 1180
    },
    {
      "epoch": 0.38857142857142857,
      "grad_norm": 1.4248754978179932,
      "learning_rate": 0.0005,
      "loss": 4.0275,
      "step": 1190
    },
    {
      "epoch": 0.39183673469387753,
      "grad_norm": 1.8577288389205933,
      "learning_rate": 0.0005,
      "loss": 3.9708,
      "step": 1200
    },
    {
      "epoch": 0.39510204081632655,
      "grad_norm": 1.2153960466384888,
      "learning_rate": 0.0005,
      "loss": 3.8195,
      "step": 1210
    },
    {
      "epoch": 0.3983673469387755,
      "grad_norm": 1.238071084022522,
      "learning_rate": 0.0005,
      "loss": 4.3136,
      "step": 1220
    },
    {
      "epoch": 0.4016326530612245,
      "grad_norm": 1.4334009885787964,
      "learning_rate": 0.0005,
      "loss": 4.1319,
      "step": 1230
    },
    {
      "epoch": 0.4048979591836735,
      "grad_norm": 1.3810005187988281,
      "learning_rate": 0.0005,
      "loss": 4.0287,
      "step": 1240
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 2.1688148975372314,
      "learning_rate": 0.0005,
      "loss": 4.0158,
      "step": 1250
    },
    {
      "epoch": 0.4114285714285714,
      "grad_norm": 1.2777838706970215,
      "learning_rate": 0.0005,
      "loss": 3.8439,
      "step": 1260
    },
    {
      "epoch": 0.4146938775510204,
      "grad_norm": 1.1235041618347168,
      "learning_rate": 0.0005,
      "loss": 4.2695,
      "step": 1270
    },
    {
      "epoch": 0.4179591836734694,
      "grad_norm": 1.2477576732635498,
      "learning_rate": 0.0005,
      "loss": 4.0986,
      "step": 1280
    },
    {
      "epoch": 0.42122448979591837,
      "grad_norm": 1.2333978414535522,
      "learning_rate": 0.0005,
      "loss": 4.007,
      "step": 1290
    },
    {
      "epoch": 0.42448979591836733,
      "grad_norm": 2.194262981414795,
      "learning_rate": 0.0005,
      "loss": 3.9411,
      "step": 1300
    },
    {
      "epoch": 0.42775510204081635,
      "grad_norm": 1.084908366203308,
      "learning_rate": 0.0005,
      "loss": 3.6208,
      "step": 1310
    },
    {
      "epoch": 0.4310204081632653,
      "grad_norm": 1.275455117225647,
      "learning_rate": 0.0005,
      "loss": 4.2136,
      "step": 1320
    },
    {
      "epoch": 0.4342857142857143,
      "grad_norm": 1.4374445676803589,
      "learning_rate": 0.0005,
      "loss": 4.0137,
      "step": 1330
    },
    {
      "epoch": 0.43755102040816324,
      "grad_norm": 1.3499337434768677,
      "learning_rate": 0.0005,
      "loss": 3.8929,
      "step": 1340
    },
    {
      "epoch": 0.44081632653061226,
      "grad_norm": 1.9091603755950928,
      "learning_rate": 0.0005,
      "loss": 3.8582,
      "step": 1350
    },
    {
      "epoch": 0.4440816326530612,
      "grad_norm": 1.0902653932571411,
      "learning_rate": 0.0005,
      "loss": 3.6813,
      "step": 1360
    },
    {
      "epoch": 0.4473469387755102,
      "grad_norm": 1.1447824239730835,
      "learning_rate": 0.0005,
      "loss": 4.236,
      "step": 1370
    },
    {
      "epoch": 0.4506122448979592,
      "grad_norm": 1.4679241180419922,
      "learning_rate": 0.0005,
      "loss": 3.9569,
      "step": 1380
    },
    {
      "epoch": 0.45387755102040817,
      "grad_norm": 1.7417553663253784,
      "learning_rate": 0.0005,
      "loss": 3.8949,
      "step": 1390
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 1.7916682958602905,
      "learning_rate": 0.0005,
      "loss": 3.8576,
      "step": 1400
    },
    {
      "epoch": 0.46040816326530615,
      "grad_norm": 1.053566813468933,
      "learning_rate": 0.0005,
      "loss": 3.6635,
      "step": 1410
    },
    {
      "epoch": 0.4636734693877551,
      "grad_norm": 1.0085692405700684,
      "learning_rate": 0.0005,
      "loss": 4.11,
      "step": 1420
    },
    {
      "epoch": 0.4669387755102041,
      "grad_norm": 1.3383585214614868,
      "learning_rate": 0.0005,
      "loss": 3.9825,
      "step": 1430
    },
    {
      "epoch": 0.47020408163265304,
      "grad_norm": 1.3227241039276123,
      "learning_rate": 0.0005,
      "loss": 3.8703,
      "step": 1440
    },
    {
      "epoch": 0.47346938775510206,
      "grad_norm": 2.0333919525146484,
      "learning_rate": 0.0005,
      "loss": 3.8328,
      "step": 1450
    },
    {
      "epoch": 0.476734693877551,
      "grad_norm": 1.2266664505004883,
      "learning_rate": 0.0005,
      "loss": 3.6192,
      "step": 1460
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.3952326774597168,
      "learning_rate": 0.0005,
      "loss": 4.149,
      "step": 1470
    },
    {
      "epoch": 0.483265306122449,
      "grad_norm": 1.4793697595596313,
      "learning_rate": 0.0005,
      "loss": 4.0945,
      "step": 1480
    },
    {
      "epoch": 0.48653061224489796,
      "grad_norm": 1.4412100315093994,
      "learning_rate": 0.0005,
      "loss": 3.8816,
      "step": 1490
    },
    {
      "epoch": 0.4897959183673469,
      "grad_norm": 1.8379197120666504,
      "learning_rate": 0.0005,
      "loss": 3.766,
      "step": 1500
    },
    {
      "epoch": 0.4897959183673469,
      "eval_loss": 3.902669906616211,
      "eval_runtime": 43.0604,
      "eval_samples_per_second": 46.446,
      "eval_steps_per_second": 11.612,
      "step": 1500
    },
    {
      "epoch": 0.4930612244897959,
      "grad_norm": 1.1561907529830933,
      "learning_rate": 0.0005,
      "loss": 3.5564,
      "step": 1510
    },
    {
      "epoch": 0.4963265306122449,
      "grad_norm": 1.1232249736785889,
      "learning_rate": 0.0005,
      "loss": 4.1061,
      "step": 1520
    },
    {
      "epoch": 0.4995918367346939,
      "grad_norm": 1.1983751058578491,
      "learning_rate": 0.0005,
      "loss": 4.0355,
      "step": 1530
    },
    {
      "epoch": 0.5028571428571429,
      "grad_norm": 1.4547094106674194,
      "learning_rate": 0.0005,
      "loss": 3.9718,
      "step": 1540
    },
    {
      "epoch": 0.5061224489795918,
      "grad_norm": 2.0942959785461426,
      "learning_rate": 0.0005,
      "loss": 3.8531,
      "step": 1550
    },
    {
      "epoch": 0.5093877551020408,
      "grad_norm": 1.4254790544509888,
      "learning_rate": 0.0005,
      "loss": 3.6136,
      "step": 1560
    },
    {
      "epoch": 0.5126530612244898,
      "grad_norm": 1.3434633016586304,
      "learning_rate": 0.0005,
      "loss": 4.0861,
      "step": 1570
    },
    {
      "epoch": 0.5159183673469387,
      "grad_norm": 1.2875784635543823,
      "learning_rate": 0.0005,
      "loss": 3.9535,
      "step": 1580
    },
    {
      "epoch": 0.5191836734693878,
      "grad_norm": 1.324489951133728,
      "learning_rate": 0.0005,
      "loss": 3.7898,
      "step": 1590
    },
    {
      "epoch": 0.5224489795918368,
      "grad_norm": 1.9661375284194946,
      "learning_rate": 0.0005,
      "loss": 3.8496,
      "step": 1600
    },
    {
      "epoch": 0.5257142857142857,
      "grad_norm": 1.1303874254226685,
      "learning_rate": 0.0005,
      "loss": 3.5894,
      "step": 1610
    },
    {
      "epoch": 0.5289795918367347,
      "grad_norm": 1.2739813327789307,
      "learning_rate": 0.0005,
      "loss": 4.0466,
      "step": 1620
    },
    {
      "epoch": 0.5322448979591837,
      "grad_norm": 1.3381832838058472,
      "learning_rate": 0.0005,
      "loss": 3.9463,
      "step": 1630
    },
    {
      "epoch": 0.5355102040816326,
      "grad_norm": 1.3114221096038818,
      "learning_rate": 0.0005,
      "loss": 3.8021,
      "step": 1640
    },
    {
      "epoch": 0.5387755102040817,
      "grad_norm": 1.9492048025131226,
      "learning_rate": 0.0005,
      "loss": 3.7187,
      "step": 1650
    },
    {
      "epoch": 0.5420408163265306,
      "grad_norm": 1.1877042055130005,
      "learning_rate": 0.0005,
      "loss": 3.5491,
      "step": 1660
    },
    {
      "epoch": 0.5453061224489796,
      "grad_norm": 1.1403299570083618,
      "learning_rate": 0.0005,
      "loss": 4.0365,
      "step": 1670
    },
    {
      "epoch": 0.5485714285714286,
      "grad_norm": 1.369149088859558,
      "learning_rate": 0.0005,
      "loss": 3.8219,
      "step": 1680
    },
    {
      "epoch": 0.5518367346938775,
      "grad_norm": 1.3963128328323364,
      "learning_rate": 0.0005,
      "loss": 3.9018,
      "step": 1690
    },
    {
      "epoch": 0.5551020408163265,
      "grad_norm": 1.9626106023788452,
      "learning_rate": 0.0005,
      "loss": 3.8125,
      "step": 1700
    },
    {
      "epoch": 0.5583673469387755,
      "grad_norm": 1.3681797981262207,
      "learning_rate": 0.0005,
      "loss": 3.5254,
      "step": 1710
    },
    {
      "epoch": 0.5616326530612245,
      "grad_norm": 1.3346383571624756,
      "learning_rate": 0.0005,
      "loss": 4.1034,
      "step": 1720
    },
    {
      "epoch": 0.5648979591836735,
      "grad_norm": 1.2607356309890747,
      "learning_rate": 0.0005,
      "loss": 3.8995,
      "step": 1730
    },
    {
      "epoch": 0.5681632653061225,
      "grad_norm": 1.4595521688461304,
      "learning_rate": 0.0005,
      "loss": 3.7525,
      "step": 1740
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 2.0250532627105713,
      "learning_rate": 0.0005,
      "loss": 3.8421,
      "step": 1750
    },
    {
      "epoch": 0.5746938775510204,
      "grad_norm": 1.277748942375183,
      "learning_rate": 0.0005,
      "loss": 3.6177,
      "step": 1760
    },
    {
      "epoch": 0.5779591836734694,
      "grad_norm": 1.3413293361663818,
      "learning_rate": 0.0005,
      "loss": 4.0805,
      "step": 1770
    },
    {
      "epoch": 0.5812244897959183,
      "grad_norm": 1.1629762649536133,
      "learning_rate": 0.0005,
      "loss": 3.9218,
      "step": 1780
    },
    {
      "epoch": 0.5844897959183674,
      "grad_norm": 1.3503767251968384,
      "learning_rate": 0.0005,
      "loss": 3.7097,
      "step": 1790
    },
    {
      "epoch": 0.5877551020408164,
      "grad_norm": 1.966658353805542,
      "learning_rate": 0.0005,
      "loss": 3.8113,
      "step": 1800
    },
    {
      "epoch": 0.5910204081632653,
      "grad_norm": 1.1706182956695557,
      "learning_rate": 0.0005,
      "loss": 3.6205,
      "step": 1810
    },
    {
      "epoch": 0.5942857142857143,
      "grad_norm": 1.095694661140442,
      "learning_rate": 0.0005,
      "loss": 3.9975,
      "step": 1820
    },
    {
      "epoch": 0.5975510204081632,
      "grad_norm": 1.3853328227996826,
      "learning_rate": 0.0005,
      "loss": 3.8375,
      "step": 1830
    },
    {
      "epoch": 0.6008163265306122,
      "grad_norm": 1.396588683128357,
      "learning_rate": 0.0005,
      "loss": 3.8137,
      "step": 1840
    },
    {
      "epoch": 0.6040816326530613,
      "grad_norm": 1.8910547494888306,
      "learning_rate": 0.0005,
      "loss": 3.8027,
      "step": 1850
    },
    {
      "epoch": 0.6073469387755102,
      "grad_norm": 1.1377391815185547,
      "learning_rate": 0.0005,
      "loss": 3.4935,
      "step": 1860
    },
    {
      "epoch": 0.6106122448979592,
      "grad_norm": 1.1120401620864868,
      "learning_rate": 0.0005,
      "loss": 4.0356,
      "step": 1870
    },
    {
      "epoch": 0.6138775510204082,
      "grad_norm": 1.2026475667953491,
      "learning_rate": 0.0005,
      "loss": 3.8108,
      "step": 1880
    },
    {
      "epoch": 0.6171428571428571,
      "grad_norm": 1.4205538034439087,
      "learning_rate": 0.0005,
      "loss": 3.731,
      "step": 1890
    },
    {
      "epoch": 0.6204081632653061,
      "grad_norm": 1.925189733505249,
      "learning_rate": 0.0005,
      "loss": 3.7237,
      "step": 1900
    },
    {
      "epoch": 0.6236734693877551,
      "grad_norm": 1.1621805429458618,
      "learning_rate": 0.0005,
      "loss": 3.5831,
      "step": 1910
    },
    {
      "epoch": 0.626938775510204,
      "grad_norm": 1.2125190496444702,
      "learning_rate": 0.0005,
      "loss": 3.9736,
      "step": 1920
    },
    {
      "epoch": 0.6302040816326531,
      "grad_norm": 1.4116473197937012,
      "learning_rate": 0.0005,
      "loss": 3.8547,
      "step": 1930
    },
    {
      "epoch": 0.6334693877551021,
      "grad_norm": 1.5884093046188354,
      "learning_rate": 0.0005,
      "loss": 3.7234,
      "step": 1940
    },
    {
      "epoch": 0.636734693877551,
      "grad_norm": 1.875867486000061,
      "learning_rate": 0.0005,
      "loss": 3.7336,
      "step": 1950
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.1436996459960938,
      "learning_rate": 0.0005,
      "loss": 3.4685,
      "step": 1960
    },
    {
      "epoch": 0.643265306122449,
      "grad_norm": 1.4208112955093384,
      "learning_rate": 0.0005,
      "loss": 4.0489,
      "step": 1970
    },
    {
      "epoch": 0.6465306122448979,
      "grad_norm": 1.3270920515060425,
      "learning_rate": 0.0005,
      "loss": 3.8098,
      "step": 1980
    },
    {
      "epoch": 0.649795918367347,
      "grad_norm": 1.6381317377090454,
      "learning_rate": 0.0005,
      "loss": 3.6228,
      "step": 1990
    },
    {
      "epoch": 0.6530612244897959,
      "grad_norm": 2.0222079753875732,
      "learning_rate": 0.0005,
      "loss": 3.678,
      "step": 2000
    },
    {
      "epoch": 0.6530612244897959,
      "eval_loss": 3.763411521911621,
      "eval_runtime": 43.2803,
      "eval_samples_per_second": 46.21,
      "eval_steps_per_second": 11.553,
      "step": 2000
    },
    {
      "epoch": 0.6563265306122449,
      "grad_norm": 1.211680293083191,
      "learning_rate": 0.0005,
      "loss": 3.5285,
      "step": 2010
    },
    {
      "epoch": 0.6595918367346939,
      "grad_norm": 1.27492094039917,
      "learning_rate": 0.0005,
      "loss": 3.977,
      "step": 2020
    },
    {
      "epoch": 0.6628571428571428,
      "grad_norm": 1.4437798261642456,
      "learning_rate": 0.0005,
      "loss": 3.8207,
      "step": 2030
    },
    {
      "epoch": 0.6661224489795918,
      "grad_norm": 1.3550224304199219,
      "learning_rate": 0.0005,
      "loss": 3.6058,
      "step": 2040
    },
    {
      "epoch": 0.6693877551020408,
      "grad_norm": 2.0937228202819824,
      "learning_rate": 0.0005,
      "loss": 3.6986,
      "step": 2050
    },
    {
      "epoch": 0.6726530612244898,
      "grad_norm": 1.0755195617675781,
      "learning_rate": 0.0005,
      "loss": 3.4733,
      "step": 2060
    },
    {
      "epoch": 0.6759183673469388,
      "grad_norm": 1.157943606376648,
      "learning_rate": 0.0005,
      "loss": 3.9306,
      "step": 2070
    },
    {
      "epoch": 0.6791836734693878,
      "grad_norm": 1.1048219203948975,
      "learning_rate": 0.0005,
      "loss": 3.7846,
      "step": 2080
    },
    {
      "epoch": 0.6824489795918367,
      "grad_norm": 1.385745882987976,
      "learning_rate": 0.0005,
      "loss": 3.804,
      "step": 2090
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 1.8955817222595215,
      "learning_rate": 0.0005,
      "loss": 3.6585,
      "step": 2100
    },
    {
      "epoch": 0.6889795918367347,
      "grad_norm": 1.0233700275421143,
      "learning_rate": 0.0005,
      "loss": 3.4444,
      "step": 2110
    },
    {
      "epoch": 0.6922448979591836,
      "grad_norm": 1.180027723312378,
      "learning_rate": 0.0005,
      "loss": 3.9621,
      "step": 2120
    },
    {
      "epoch": 0.6955102040816327,
      "grad_norm": 1.2139476537704468,
      "learning_rate": 0.0005,
      "loss": 3.785,
      "step": 2130
    },
    {
      "epoch": 0.6987755102040817,
      "grad_norm": 1.285551905632019,
      "learning_rate": 0.0005,
      "loss": 3.7749,
      "step": 2140
    },
    {
      "epoch": 0.7020408163265306,
      "grad_norm": 1.9963250160217285,
      "learning_rate": 0.0005,
      "loss": 3.6162,
      "step": 2150
    },
    {
      "epoch": 0.7053061224489796,
      "grad_norm": 1.216117262840271,
      "learning_rate": 0.0005,
      "loss": 3.5505,
      "step": 2160
    },
    {
      "epoch": 0.7085714285714285,
      "grad_norm": 1.181820034980774,
      "learning_rate": 0.0005,
      "loss": 4.0493,
      "step": 2170
    },
    {
      "epoch": 0.7118367346938775,
      "grad_norm": 1.181449294090271,
      "learning_rate": 0.0005,
      "loss": 3.7653,
      "step": 2180
    },
    {
      "epoch": 0.7151020408163266,
      "grad_norm": 1.350578784942627,
      "learning_rate": 0.0005,
      "loss": 3.7466,
      "step": 2190
    },
    {
      "epoch": 0.7183673469387755,
      "grad_norm": 2.2694125175476074,
      "learning_rate": 0.0005,
      "loss": 3.6147,
      "step": 2200
    },
    {
      "epoch": 0.7216326530612245,
      "grad_norm": 1.4255095720291138,
      "learning_rate": 0.0005,
      "loss": 3.5044,
      "step": 2210
    },
    {
      "epoch": 0.7248979591836735,
      "grad_norm": 1.037293791770935,
      "learning_rate": 0.0005,
      "loss": 3.8757,
      "step": 2220
    },
    {
      "epoch": 0.7281632653061224,
      "grad_norm": 1.4334427118301392,
      "learning_rate": 0.0005,
      "loss": 3.856,
      "step": 2230
    },
    {
      "epoch": 0.7314285714285714,
      "grad_norm": 1.3529642820358276,
      "learning_rate": 0.0005,
      "loss": 3.6443,
      "step": 2240
    },
    {
      "epoch": 0.7346938775510204,
      "grad_norm": 2.0167791843414307,
      "learning_rate": 0.0005,
      "loss": 3.6821,
      "step": 2250
    },
    {
      "epoch": 0.7379591836734694,
      "grad_norm": 1.011146068572998,
      "learning_rate": 0.0005,
      "loss": 3.2779,
      "step": 2260
    },
    {
      "epoch": 0.7412244897959184,
      "grad_norm": 1.0214072465896606,
      "learning_rate": 0.0005,
      "loss": 3.8809,
      "step": 2270
    },
    {
      "epoch": 0.7444897959183674,
      "grad_norm": 1.3080716133117676,
      "learning_rate": 0.0005,
      "loss": 3.7988,
      "step": 2280
    },
    {
      "epoch": 0.7477551020408163,
      "grad_norm": 1.4071499109268188,
      "learning_rate": 0.0005,
      "loss": 3.7096,
      "step": 2290
    },
    {
      "epoch": 0.7510204081632653,
      "grad_norm": 2.372915744781494,
      "learning_rate": 0.0005,
      "loss": 3.6902,
      "step": 2300
    },
    {
      "epoch": 0.7542857142857143,
      "grad_norm": 1.1310540437698364,
      "learning_rate": 0.0005,
      "loss": 3.5033,
      "step": 2310
    },
    {
      "epoch": 0.7575510204081632,
      "grad_norm": 1.1746577024459839,
      "learning_rate": 0.0005,
      "loss": 3.9845,
      "step": 2320
    },
    {
      "epoch": 0.7608163265306123,
      "grad_norm": 1.3103886842727661,
      "learning_rate": 0.0005,
      "loss": 3.7691,
      "step": 2330
    },
    {
      "epoch": 0.7640816326530612,
      "grad_norm": 1.3824633359909058,
      "learning_rate": 0.0005,
      "loss": 3.5941,
      "step": 2340
    },
    {
      "epoch": 0.7673469387755102,
      "grad_norm": 2.100325345993042,
      "learning_rate": 0.0005,
      "loss": 3.6041,
      "step": 2350
    },
    {
      "epoch": 0.7706122448979592,
      "grad_norm": 1.1913516521453857,
      "learning_rate": 0.0005,
      "loss": 3.3779,
      "step": 2360
    },
    {
      "epoch": 0.7738775510204081,
      "grad_norm": 0.9938827753067017,
      "learning_rate": 0.0005,
      "loss": 3.8768,
      "step": 2370
    },
    {
      "epoch": 0.7771428571428571,
      "grad_norm": 1.2203476428985596,
      "learning_rate": 0.0005,
      "loss": 3.6631,
      "step": 2380
    },
    {
      "epoch": 0.7804081632653062,
      "grad_norm": 1.2797229290008545,
      "learning_rate": 0.0005,
      "loss": 3.682,
      "step": 2390
    },
    {
      "epoch": 0.7836734693877551,
      "grad_norm": 1.820462703704834,
      "learning_rate": 0.0005,
      "loss": 3.63,
      "step": 2400
    },
    {
      "epoch": 0.7869387755102041,
      "grad_norm": 1.147829532623291,
      "learning_rate": 0.0005,
      "loss": 3.3512,
      "step": 2410
    },
    {
      "epoch": 0.7902040816326531,
      "grad_norm": 1.2591899633407593,
      "learning_rate": 0.0005,
      "loss": 3.8567,
      "step": 2420
    },
    {
      "epoch": 0.793469387755102,
      "grad_norm": 1.2597836256027222,
      "learning_rate": 0.0005,
      "loss": 3.7401,
      "step": 2430
    },
    {
      "epoch": 0.796734693877551,
      "grad_norm": 1.3258607387542725,
      "learning_rate": 0.0005,
      "loss": 3.5487,
      "step": 2440
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.17008376121521,
      "learning_rate": 0.0005,
      "loss": 3.5872,
      "step": 2450
    },
    {
      "epoch": 0.803265306122449,
      "grad_norm": 1.0918580293655396,
      "learning_rate": 0.0005,
      "loss": 3.3497,
      "step": 2460
    },
    {
      "epoch": 0.806530612244898,
      "grad_norm": 1.1970785856246948,
      "learning_rate": 0.0005,
      "loss": 3.9383,
      "step": 2470
    },
    {
      "epoch": 0.809795918367347,
      "grad_norm": 1.2330458164215088,
      "learning_rate": 0.0005,
      "loss": 3.7035,
      "step": 2480
    },
    {
      "epoch": 0.8130612244897959,
      "grad_norm": 1.3976151943206787,
      "learning_rate": 0.0005,
      "loss": 3.6041,
      "step": 2490
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 1.8973948955535889,
      "learning_rate": 0.0005,
      "loss": 3.5651,
      "step": 2500
    },
    {
      "epoch": 0.8163265306122449,
      "eval_loss": 3.65913724899292,
      "eval_runtime": 42.7312,
      "eval_samples_per_second": 46.804,
      "eval_steps_per_second": 11.701,
      "step": 2500
    },
    {
      "epoch": 0.8195918367346938,
      "grad_norm": 1.126738429069519,
      "learning_rate": 0.0005,
      "loss": 3.3749,
      "step": 2510
    },
    {
      "epoch": 0.8228571428571428,
      "grad_norm": 1.2251478433609009,
      "learning_rate": 0.0005,
      "loss": 3.884,
      "step": 2520
    },
    {
      "epoch": 0.8261224489795919,
      "grad_norm": 1.219794750213623,
      "learning_rate": 0.0005,
      "loss": 3.6404,
      "step": 2530
    },
    {
      "epoch": 0.8293877551020408,
      "grad_norm": 1.3533365726470947,
      "learning_rate": 0.0005,
      "loss": 3.5655,
      "step": 2540
    },
    {
      "epoch": 0.8326530612244898,
      "grad_norm": 1.7649017572402954,
      "learning_rate": 0.0005,
      "loss": 3.6185,
      "step": 2550
    },
    {
      "epoch": 0.8359183673469388,
      "grad_norm": 1.0244803428649902,
      "learning_rate": 0.0005,
      "loss": 3.4428,
      "step": 2560
    },
    {
      "epoch": 0.8391836734693877,
      "grad_norm": 1.3756012916564941,
      "learning_rate": 0.0005,
      "loss": 3.8133,
      "step": 2570
    },
    {
      "epoch": 0.8424489795918367,
      "grad_norm": 1.4523794651031494,
      "learning_rate": 0.0005,
      "loss": 3.6916,
      "step": 2580
    },
    {
      "epoch": 0.8457142857142858,
      "grad_norm": 1.4527126550674438,
      "learning_rate": 0.0005,
      "loss": 3.5812,
      "step": 2590
    },
    {
      "epoch": 0.8489795918367347,
      "grad_norm": 1.7975162267684937,
      "learning_rate": 0.0005,
      "loss": 3.6506,
      "step": 2600
    },
    {
      "epoch": 0.8522448979591837,
      "grad_norm": 1.29179048538208,
      "learning_rate": 0.0005,
      "loss": 3.4787,
      "step": 2610
    },
    {
      "epoch": 0.8555102040816327,
      "grad_norm": 1.2570093870162964,
      "learning_rate": 0.0005,
      "loss": 3.933,
      "step": 2620
    },
    {
      "epoch": 0.8587755102040816,
      "grad_norm": 1.350013017654419,
      "learning_rate": 0.0005,
      "loss": 3.6734,
      "step": 2630
    },
    {
      "epoch": 0.8620408163265306,
      "grad_norm": 1.4130898714065552,
      "learning_rate": 0.0005,
      "loss": 3.4979,
      "step": 2640
    },
    {
      "epoch": 0.8653061224489796,
      "grad_norm": 2.0199756622314453,
      "learning_rate": 0.0005,
      "loss": 3.5747,
      "step": 2650
    },
    {
      "epoch": 0.8685714285714285,
      "grad_norm": 1.0159718990325928,
      "learning_rate": 0.0005,
      "loss": 3.4062,
      "step": 2660
    },
    {
      "epoch": 0.8718367346938776,
      "grad_norm": 1.2518935203552246,
      "learning_rate": 0.0005,
      "loss": 3.7682,
      "step": 2670
    },
    {
      "epoch": 0.8751020408163265,
      "grad_norm": 1.2634557485580444,
      "learning_rate": 0.0005,
      "loss": 3.7365,
      "step": 2680
    },
    {
      "epoch": 0.8783673469387755,
      "grad_norm": 1.2956312894821167,
      "learning_rate": 0.0005,
      "loss": 3.6749,
      "step": 2690
    },
    {
      "epoch": 0.8816326530612245,
      "grad_norm": 1.8405163288116455,
      "learning_rate": 0.0005,
      "loss": 3.5182,
      "step": 2700
    },
    {
      "epoch": 0.8848979591836734,
      "grad_norm": 1.1379750967025757,
      "learning_rate": 0.0005,
      "loss": 3.376,
      "step": 2710
    },
    {
      "epoch": 0.8881632653061224,
      "grad_norm": 1.3452417850494385,
      "learning_rate": 0.0005,
      "loss": 3.8657,
      "step": 2720
    },
    {
      "epoch": 0.8914285714285715,
      "grad_norm": 1.1698819398880005,
      "learning_rate": 0.0005,
      "loss": 3.633,
      "step": 2730
    },
    {
      "epoch": 0.8946938775510204,
      "grad_norm": 1.42051100730896,
      "learning_rate": 0.0005,
      "loss": 3.5665,
      "step": 2740
    },
    {
      "epoch": 0.8979591836734694,
      "grad_norm": 1.9082551002502441,
      "learning_rate": 0.0005,
      "loss": 3.5884,
      "step": 2750
    },
    {
      "epoch": 0.9012244897959184,
      "grad_norm": 1.3937710523605347,
      "learning_rate": 0.0005,
      "loss": 3.3543,
      "step": 2760
    },
    {
      "epoch": 0.9044897959183673,
      "grad_norm": 1.2848858833312988,
      "learning_rate": 0.0005,
      "loss": 3.8298,
      "step": 2770
    },
    {
      "epoch": 0.9077551020408163,
      "grad_norm": 1.2677395343780518,
      "learning_rate": 0.0005,
      "loss": 3.6608,
      "step": 2780
    },
    {
      "epoch": 0.9110204081632653,
      "grad_norm": 1.372312307357788,
      "learning_rate": 0.0005,
      "loss": 3.5296,
      "step": 2790
    },
    {
      "epoch": 0.9142857142857143,
      "grad_norm": 1.928770899772644,
      "learning_rate": 0.0005,
      "loss": 3.5219,
      "step": 2800
    },
    {
      "epoch": 0.9175510204081633,
      "grad_norm": 1.1431926488876343,
      "learning_rate": 0.0005,
      "loss": 3.3378,
      "step": 2810
    },
    {
      "epoch": 0.9208163265306123,
      "grad_norm": 1.5195988416671753,
      "learning_rate": 0.0005,
      "loss": 3.8191,
      "step": 2820
    },
    {
      "epoch": 0.9240816326530612,
      "grad_norm": 1.3690383434295654,
      "learning_rate": 0.0005,
      "loss": 3.5557,
      "step": 2830
    },
    {
      "epoch": 0.9273469387755102,
      "grad_norm": 1.473240852355957,
      "learning_rate": 0.0005,
      "loss": 3.5422,
      "step": 2840
    },
    {
      "epoch": 0.9306122448979591,
      "grad_norm": 2.0330464839935303,
      "learning_rate": 0.0005,
      "loss": 3.6382,
      "step": 2850
    },
    {
      "epoch": 0.9338775510204081,
      "grad_norm": 1.0458343029022217,
      "learning_rate": 0.0005,
      "loss": 3.3883,
      "step": 2860
    },
    {
      "epoch": 0.9371428571428572,
      "grad_norm": 1.1635581254959106,
      "learning_rate": 0.0005,
      "loss": 3.8158,
      "step": 2870
    },
    {
      "epoch": 0.9404081632653061,
      "grad_norm": 1.3448679447174072,
      "learning_rate": 0.0005,
      "loss": 3.6389,
      "step": 2880
    },
    {
      "epoch": 0.9436734693877551,
      "grad_norm": 1.3173009157180786,
      "learning_rate": 0.0005,
      "loss": 3.5386,
      "step": 2890
    },
    {
      "epoch": 0.9469387755102041,
      "grad_norm": 2.144378662109375,
      "learning_rate": 0.0005,
      "loss": 3.5892,
      "step": 2900
    },
    {
      "epoch": 0.950204081632653,
      "grad_norm": 1.1590373516082764,
      "learning_rate": 0.0005,
      "loss": 3.397,
      "step": 2910
    },
    {
      "epoch": 0.953469387755102,
      "grad_norm": 1.253893494606018,
      "learning_rate": 0.0005,
      "loss": 3.8054,
      "step": 2920
    },
    {
      "epoch": 0.9567346938775511,
      "grad_norm": 1.215649962425232,
      "learning_rate": 0.0005,
      "loss": 3.612,
      "step": 2930
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.4085159301757812,
      "learning_rate": 0.0005,
      "loss": 3.5056,
      "step": 2940
    },
    {
      "epoch": 0.963265306122449,
      "grad_norm": 1.908341646194458,
      "learning_rate": 0.0005,
      "loss": 3.4516,
      "step": 2950
    },
    {
      "epoch": 0.966530612244898,
      "grad_norm": 1.1971689462661743,
      "learning_rate": 0.0005,
      "loss": 3.3546,
      "step": 2960
    },
    {
      "epoch": 0.9697959183673469,
      "grad_norm": 1.2646092176437378,
      "learning_rate": 0.0005,
      "loss": 3.8237,
      "step": 2970
    },
    {
      "epoch": 0.9730612244897959,
      "grad_norm": 1.3261162042617798,
      "learning_rate": 0.0005,
      "loss": 3.6123,
      "step": 2980
    },
    {
      "epoch": 0.976326530612245,
      "grad_norm": 1.4025673866271973,
      "learning_rate": 0.0005,
      "loss": 3.5225,
      "step": 2990
    },
    {
      "epoch": 0.9795918367346939,
      "grad_norm": 1.9254060983657837,
      "learning_rate": 0.0005,
      "loss": 3.5353,
      "step": 3000
    },
    {
      "epoch": 0.9795918367346939,
      "eval_loss": 3.575631618499756,
      "eval_runtime": 42.8247,
      "eval_samples_per_second": 46.702,
      "eval_steps_per_second": 11.675,
      "step": 3000
    },
    {
      "epoch": 0.9828571428571429,
      "grad_norm": 1.1695542335510254,
      "learning_rate": 0.0005,
      "loss": 3.1419,
      "step": 3010
    },
    {
      "epoch": 0.9861224489795918,
      "grad_norm": 1.1839579343795776,
      "learning_rate": 0.0005,
      "loss": 3.7104,
      "step": 3020
    },
    {
      "epoch": 0.9893877551020408,
      "grad_norm": 1.3528823852539062,
      "learning_rate": 0.0005,
      "loss": 3.6772,
      "step": 3030
    },
    {
      "epoch": 0.9926530612244898,
      "grad_norm": 1.42439603805542,
      "learning_rate": 0.0005,
      "loss": 3.544,
      "step": 3040
    },
    {
      "epoch": 0.9959183673469387,
      "grad_norm": 2.387221574783325,
      "learning_rate": 0.0005,
      "loss": 3.536,
      "step": 3050
    },
    {
      "epoch": 0.9991836734693877,
      "grad_norm": 1.4861875772476196,
      "learning_rate": 0.0005,
      "loss": 3.5749,
      "step": 3060
    }
  ],
  "logging_steps": 10,
  "max_steps": 3063,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 28910317824000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}