{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4007212983370066,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0016028851933480265,
      "grad_norm": 776.670654296875,
      "learning_rate": 5.399999999999999e-06,
      "loss": 9.5724,
      "step": 10
    },
    {
      "epoch": 0.003205770386696053,
      "grad_norm": 193.75341796875,
      "learning_rate": 1.14e-05,
      "loss": 7.2141,
      "step": 20
    },
    {
      "epoch": 0.004808655580044079,
      "grad_norm": 52.97894287109375,
      "learning_rate": 1.74e-05,
      "loss": 4.7166,
      "step": 30
    },
    {
      "epoch": 0.006411540773392106,
      "grad_norm": 19.748123168945312,
      "learning_rate": 2.34e-05,
      "loss": 3.6157,
      "step": 40
    },
    {
      "epoch": 0.008014425966740133,
      "grad_norm": 6.880778789520264,
      "learning_rate": 2.94e-05,
      "loss": 3.1652,
      "step": 50
    },
    {
      "epoch": 0.009617311160088158,
      "grad_norm": 10.432415008544922,
      "learning_rate": 3.539999999999999e-05,
      "loss": 2.8247,
      "step": 60
    },
    {
      "epoch": 0.011220196353436185,
      "grad_norm": 13.381453514099121,
      "learning_rate": 4.14e-05,
      "loss": 2.5518,
      "step": 70
    },
    {
      "epoch": 0.012823081546784212,
      "grad_norm": 6.769928932189941,
      "learning_rate": 4.7399999999999993e-05,
      "loss": 2.4099,
      "step": 80
    },
    {
      "epoch": 0.014425966740132239,
      "grad_norm": 12.884427070617676,
      "learning_rate": 5.339999999999999e-05,
      "loss": 2.2892,
      "step": 90
    },
    {
      "epoch": 0.016028851933480266,
      "grad_norm": 7.974192142486572,
      "learning_rate": 5.94e-05,
      "loss": 2.1681,
      "step": 100
    },
    {
      "epoch": 0.017631737126828292,
      "grad_norm": 3.720524549484253,
      "learning_rate": 6.539999999999999e-05,
      "loss": 2.0774,
      "step": 110
    },
    {
      "epoch": 0.019234622320176316,
      "grad_norm": 2.466925859451294,
      "learning_rate": 7.139999999999999e-05,
      "loss": 1.9974,
      "step": 120
    },
    {
      "epoch": 0.020837507513524343,
      "grad_norm": 8.301297187805176,
      "learning_rate": 7.74e-05,
      "loss": 1.9744,
      "step": 130
    },
    {
      "epoch": 0.02244039270687237,
      "grad_norm": 2.760185480117798,
      "learning_rate": 8.34e-05,
      "loss": 1.9556,
      "step": 140
    },
    {
      "epoch": 0.024043277900220397,
      "grad_norm": 1.4739573001861572,
      "learning_rate": 8.939999999999999e-05,
      "loss": 1.8721,
      "step": 150
    },
    {
      "epoch": 0.025646163093568423,
      "grad_norm": 2.8045942783355713,
      "learning_rate": 9.539999999999999e-05,
      "loss": 1.8465,
      "step": 160
    },
    {
      "epoch": 0.02724904828691645,
      "grad_norm": 1.683029055595398,
      "learning_rate": 0.0001014,
      "loss": 1.8229,
      "step": 170
    },
    {
      "epoch": 0.028851933480264477,
      "grad_norm": 2.744515895843506,
      "learning_rate": 0.00010739999999999998,
      "loss": 1.8295,
      "step": 180
    },
    {
      "epoch": 0.030454818673612504,
      "grad_norm": 1.9398820400238037,
      "learning_rate": 0.00011339999999999999,
      "loss": 1.7885,
      "step": 190
    },
    {
      "epoch": 0.03205770386696053,
      "grad_norm": 2.0050437450408936,
      "learning_rate": 0.0001194,
      "loss": 1.7718,
      "step": 200
    },
    {
      "epoch": 0.033660589060308554,
      "grad_norm": 2.0628662109375,
      "learning_rate": 0.00012539999999999999,
      "loss": 1.7572,
      "step": 210
    },
    {
      "epoch": 0.035263474253656585,
      "grad_norm": 3.041044235229492,
      "learning_rate": 0.0001314,
      "loss": 1.7524,
      "step": 220
    },
    {
      "epoch": 0.03686635944700461,
      "grad_norm": 1.5776487588882446,
      "learning_rate": 0.0001374,
      "loss": 1.7808,
      "step": 230
    },
    {
      "epoch": 0.03846924464035263,
      "grad_norm": 2.0454773902893066,
      "learning_rate": 0.0001434,
      "loss": 1.6838,
      "step": 240
    },
    {
      "epoch": 0.04007212983370066,
      "grad_norm": 2.13191556930542,
      "learning_rate": 0.0001494,
      "loss": 1.7068,
      "step": 250
    },
    {
      "epoch": 0.041675015027048685,
      "grad_norm": 1.6366831064224243,
      "learning_rate": 0.00015539999999999998,
      "loss": 1.7058,
      "step": 260
    },
    {
      "epoch": 0.043277900220396716,
      "grad_norm": 1.694743037223816,
      "learning_rate": 0.0001614,
      "loss": 1.6746,
      "step": 270
    },
    {
      "epoch": 0.04488078541374474,
      "grad_norm": 1.8631821870803833,
      "learning_rate": 0.0001674,
      "loss": 1.6851,
      "step": 280
    },
    {
      "epoch": 0.04648367060709277,
      "grad_norm": 1.0038516521453857,
      "learning_rate": 0.00017339999999999996,
      "loss": 1.6461,
      "step": 290
    },
    {
      "epoch": 0.04808655580044079,
      "grad_norm": 1.987015962600708,
      "learning_rate": 0.00017939999999999997,
      "loss": 1.6531,
      "step": 300
    },
    {
      "epoch": 0.049689440993788817,
      "grad_norm": 1.0111745595932007,
      "learning_rate": 0.00018539999999999998,
      "loss": 1.6454,
      "step": 310
    },
    {
      "epoch": 0.05129232618713685,
      "grad_norm": 1.0822091102600098,
      "learning_rate": 0.0001914,
      "loss": 1.6277,
      "step": 320
    },
    {
      "epoch": 0.05289521138048487,
      "grad_norm": 1.1981233358383179,
      "learning_rate": 0.0001974,
      "loss": 1.6288,
      "step": 330
    },
    {
      "epoch": 0.0544980965738329,
      "grad_norm": 1.1497327089309692,
      "learning_rate": 0.00020339999999999998,
      "loss": 1.5923,
      "step": 340
    },
    {
      "epoch": 0.056100981767180924,
      "grad_norm": 1.0253567695617676,
      "learning_rate": 0.00020939999999999997,
      "loss": 1.5851,
      "step": 350
    },
    {
      "epoch": 0.057703866960528954,
      "grad_norm": 1.7407950162887573,
      "learning_rate": 0.00021539999999999998,
      "loss": 1.5868,
      "step": 360
    },
    {
      "epoch": 0.05930675215387698,
      "grad_norm": 1.2507755756378174,
      "learning_rate": 0.0002214,
      "loss": 1.6035,
      "step": 370
    },
    {
      "epoch": 0.06090963734722501,
      "grad_norm": 0.733970046043396,
      "learning_rate": 0.00022739999999999997,
      "loss": 1.5651,
      "step": 380
    },
    {
      "epoch": 0.06251252254057303,
      "grad_norm": 0.873483419418335,
      "learning_rate": 0.00023339999999999998,
      "loss": 1.5402,
      "step": 390
    },
    {
      "epoch": 0.06411540773392106,
      "grad_norm": 1.0391820669174194,
      "learning_rate": 0.0002394,
      "loss": 1.5666,
      "step": 400
    },
    {
      "epoch": 0.06571829292726908,
      "grad_norm": 0.8471903204917908,
      "learning_rate": 0.00024539999999999995,
      "loss": 1.5238,
      "step": 410
    },
    {
      "epoch": 0.06732117812061711,
      "grad_norm": 0.7101859450340271,
      "learning_rate": 0.0002514,
      "loss": 1.5225,
      "step": 420
    },
    {
      "epoch": 0.06892406331396514,
      "grad_norm": 0.7229673266410828,
      "learning_rate": 0.00025739999999999997,
      "loss": 1.5409,
      "step": 430
    },
    {
      "epoch": 0.07052694850731317,
      "grad_norm": 0.6100331544876099,
      "learning_rate": 0.00026339999999999995,
      "loss": 1.514,
      "step": 440
    },
    {
      "epoch": 0.07212983370066119,
      "grad_norm": 0.8513933420181274,
      "learning_rate": 0.0002694,
      "loss": 1.5146,
      "step": 450
    },
    {
      "epoch": 0.07373271889400922,
      "grad_norm": 0.8575087189674377,
      "learning_rate": 0.00027539999999999997,
      "loss": 1.5153,
      "step": 460
    },
    {
      "epoch": 0.07533560408735725,
      "grad_norm": 0.6953585147857666,
      "learning_rate": 0.00028139999999999996,
      "loss": 1.5056,
      "step": 470
    },
    {
      "epoch": 0.07693848928070526,
      "grad_norm": 0.7698249220848083,
      "learning_rate": 0.00028739999999999994,
      "loss": 1.475,
      "step": 480
    },
    {
      "epoch": 0.0785413744740533,
      "grad_norm": 1.200210690498352,
      "learning_rate": 0.0002934,
      "loss": 1.4969,
      "step": 490
    },
    {
      "epoch": 0.08014425966740132,
      "grad_norm": 0.7224527597427368,
      "learning_rate": 0.00029939999999999996,
      "loss": 1.4693,
      "step": 500
    },
    {
      "epoch": 0.08174714486074935,
      "grad_norm": 0.955494225025177,
      "learning_rate": 0.0003,
      "loss": 1.4744,
      "step": 510
    },
    {
      "epoch": 0.08335003005409737,
      "grad_norm": 0.545981228351593,
      "learning_rate": 0.0003,
      "loss": 1.4777,
      "step": 520
    },
    {
      "epoch": 0.0849529152474454,
      "grad_norm": 0.5107579231262207,
      "learning_rate": 0.0003,
      "loss": 1.447,
      "step": 530
    },
    {
      "epoch": 0.08655580044079343,
      "grad_norm": 0.5158474445343018,
      "learning_rate": 0.0003,
      "loss": 1.4615,
      "step": 540
    },
    {
      "epoch": 0.08815868563414145,
      "grad_norm": 0.6262283325195312,
      "learning_rate": 0.0003,
      "loss": 1.4725,
      "step": 550
    },
    {
      "epoch": 0.08976157082748948,
      "grad_norm": 0.652737021446228,
      "learning_rate": 0.0003,
      "loss": 1.4347,
      "step": 560
    },
    {
      "epoch": 0.09136445602083751,
      "grad_norm": 0.5195409655570984,
      "learning_rate": 0.0003,
      "loss": 1.4311,
      "step": 570
    },
    {
      "epoch": 0.09296734121418554,
      "grad_norm": 0.6093418598175049,
      "learning_rate": 0.0003,
      "loss": 1.4312,
      "step": 580
    },
    {
      "epoch": 0.09457022640753356,
      "grad_norm": 0.766174852848053,
      "learning_rate": 0.0003,
      "loss": 1.4556,
      "step": 590
    },
    {
      "epoch": 0.09617311160088159,
      "grad_norm": 0.7088481783866882,
      "learning_rate": 0.0003,
      "loss": 1.4386,
      "step": 600
    },
    {
      "epoch": 0.09777599679422962,
      "grad_norm": 0.5216871500015259,
      "learning_rate": 0.0003,
      "loss": 1.4147,
      "step": 610
    },
    {
      "epoch": 0.09937888198757763,
      "grad_norm": 0.7307924032211304,
      "learning_rate": 0.0003,
      "loss": 1.4654,
      "step": 620
    },
    {
      "epoch": 0.10098176718092566,
      "grad_norm": 0.6216389536857605,
      "learning_rate": 0.0003,
      "loss": 1.4172,
      "step": 630
    },
    {
      "epoch": 0.1025846523742737,
      "grad_norm": 0.5258434414863586,
      "learning_rate": 0.0003,
      "loss": 1.4146,
      "step": 640
    },
    {
      "epoch": 0.10418753756762172,
      "grad_norm": 0.43452855944633484,
      "learning_rate": 0.0003,
      "loss": 1.4289,
      "step": 650
    },
    {
      "epoch": 0.10579042276096974,
      "grad_norm": 0.6089204549789429,
      "learning_rate": 0.0003,
      "loss": 1.3934,
      "step": 660
    },
    {
      "epoch": 0.10739330795431777,
      "grad_norm": 1.0554406642913818,
      "learning_rate": 0.0003,
      "loss": 1.4158,
      "step": 670
    },
    {
      "epoch": 0.1089961931476658,
      "grad_norm": 0.5392816066741943,
      "learning_rate": 0.0003,
      "loss": 1.3923,
      "step": 680
    },
    {
      "epoch": 0.11059907834101383,
      "grad_norm": 0.8444032669067383,
      "learning_rate": 0.0003,
      "loss": 1.4113,
      "step": 690
    },
    {
      "epoch": 0.11220196353436185,
      "grad_norm": 0.5817086696624756,
      "learning_rate": 0.0003,
      "loss": 1.4036,
      "step": 700
    },
    {
      "epoch": 0.11380484872770988,
      "grad_norm": 0.49225664138793945,
      "learning_rate": 0.0003,
      "loss": 1.4154,
      "step": 710
    },
    {
      "epoch": 0.11540773392105791,
      "grad_norm": 0.47414547204971313,
      "learning_rate": 0.0003,
      "loss": 1.3967,
      "step": 720
    },
    {
      "epoch": 0.11701061911440593,
      "grad_norm": 0.47163447737693787,
      "learning_rate": 0.0003,
      "loss": 1.4044,
      "step": 730
    },
    {
      "epoch": 0.11861350430775396,
      "grad_norm": 0.636263370513916,
      "learning_rate": 0.0003,
      "loss": 1.4135,
      "step": 740
    },
    {
      "epoch": 0.12021638950110199,
      "grad_norm": 0.44551926851272583,
      "learning_rate": 0.0003,
      "loss": 1.3943,
      "step": 750
    },
    {
      "epoch": 0.12181927469445002,
      "grad_norm": 0.5771851539611816,
      "learning_rate": 0.0003,
      "loss": 1.4018,
      "step": 760
    },
    {
      "epoch": 0.12342215988779803,
      "grad_norm": 0.45589786767959595,
      "learning_rate": 0.0003,
      "loss": 1.3687,
      "step": 770
    },
    {
      "epoch": 0.12502504508114606,
      "grad_norm": 0.42476892471313477,
      "learning_rate": 0.0003,
      "loss": 1.3994,
      "step": 780
    },
    {
      "epoch": 0.1266279302744941,
      "grad_norm": 0.472340852022171,
      "learning_rate": 0.0003,
      "loss": 1.3919,
      "step": 790
    },
    {
      "epoch": 0.12823081546784212,
      "grad_norm": 0.3882564306259155,
      "learning_rate": 0.0003,
      "loss": 1.3754,
      "step": 800
    },
    {
      "epoch": 0.12983370066119015,
      "grad_norm": 0.4175819158554077,
      "learning_rate": 0.0003,
      "loss": 1.3827,
      "step": 810
    },
    {
      "epoch": 0.13143658585453816,
      "grad_norm": 0.4040442705154419,
      "learning_rate": 0.0003,
      "loss": 1.369,
      "step": 820
    },
    {
      "epoch": 0.1330394710478862,
      "grad_norm": 0.5839427709579468,
      "learning_rate": 0.0003,
      "loss": 1.3525,
      "step": 830
    },
    {
      "epoch": 0.13464235624123422,
      "grad_norm": 0.4182599186897278,
      "learning_rate": 0.0003,
      "loss": 1.3555,
      "step": 840
    },
    {
      "epoch": 0.13624524143458225,
      "grad_norm": 0.41773858666419983,
      "learning_rate": 0.0003,
      "loss": 1.3715,
      "step": 850
    },
    {
      "epoch": 0.13784812662793028,
      "grad_norm": 0.45445168018341064,
      "learning_rate": 0.0003,
      "loss": 1.3678,
      "step": 860
    },
    {
      "epoch": 0.1394510118212783,
      "grad_norm": 0.4130675494670868,
      "learning_rate": 0.0003,
      "loss": 1.3522,
      "step": 870
    },
    {
      "epoch": 0.14105389701462634,
      "grad_norm": 0.5697482228279114,
      "learning_rate": 0.0003,
      "loss": 1.362,
      "step": 880
    },
    {
      "epoch": 0.14265678220797434,
      "grad_norm": 0.4686114192008972,
      "learning_rate": 0.0003,
      "loss": 1.3738,
      "step": 890
    },
    {
      "epoch": 0.14425966740132237,
      "grad_norm": 0.3764347732067108,
      "learning_rate": 0.0003,
      "loss": 1.3653,
      "step": 900
    },
    {
      "epoch": 0.1458625525946704,
      "grad_norm": 0.4642839729785919,
      "learning_rate": 0.0003,
      "loss": 1.3489,
      "step": 910
    },
    {
      "epoch": 0.14746543778801843,
      "grad_norm": 0.4343486726284027,
      "learning_rate": 0.0003,
      "loss": 1.3635,
      "step": 920
    },
    {
      "epoch": 0.14906832298136646,
      "grad_norm": 0.4216885268688202,
      "learning_rate": 0.0003,
      "loss": 1.3575,
      "step": 930
    },
    {
      "epoch": 0.1506712081747145,
      "grad_norm": 0.4141499698162079,
      "learning_rate": 0.0003,
      "loss": 1.3797,
      "step": 940
    },
    {
      "epoch": 0.15227409336806252,
      "grad_norm": 0.4110085964202881,
      "learning_rate": 0.0003,
      "loss": 1.3589,
      "step": 950
    },
    {
      "epoch": 0.15387697856141053,
      "grad_norm": 0.4380963146686554,
      "learning_rate": 0.0003,
      "loss": 1.3387,
      "step": 960
    },
    {
      "epoch": 0.15547986375475856,
      "grad_norm": 0.43011143803596497,
      "learning_rate": 0.0003,
      "loss": 1.3324,
      "step": 970
    },
    {
      "epoch": 0.1570827489481066,
      "grad_norm": 0.4955873489379883,
      "learning_rate": 0.0003,
      "loss": 1.3365,
      "step": 980
    },
    {
      "epoch": 0.15868563414145462,
      "grad_norm": 0.37225353717803955,
      "learning_rate": 0.0003,
      "loss": 1.3427,
      "step": 990
    },
    {
      "epoch": 0.16028851933480265,
      "grad_norm": 0.42871296405792236,
      "learning_rate": 0.0003,
      "loss": 1.3559,
      "step": 1000
    },
    {
      "epoch": 0.16189140452815068,
      "grad_norm": 0.3852056562900543,
      "learning_rate": 0.0003,
      "loss": 1.3496,
      "step": 1010
    },
    {
      "epoch": 0.1634942897214987,
      "grad_norm": 0.3976001739501953,
      "learning_rate": 0.0003,
      "loss": 1.3107,
      "step": 1020
    },
    {
      "epoch": 0.1650971749148467,
      "grad_norm": 0.4487122893333435,
      "learning_rate": 0.0003,
      "loss": 1.3194,
      "step": 1030
    },
    {
      "epoch": 0.16670006010819474,
      "grad_norm": 0.4602479040622711,
      "learning_rate": 0.0003,
      "loss": 1.3701,
      "step": 1040
    },
    {
      "epoch": 0.16830294530154277,
      "grad_norm": 0.4249193072319031,
      "learning_rate": 0.0003,
      "loss": 1.3278,
      "step": 1050
    },
    {
      "epoch": 0.1699058304948908,
      "grad_norm": 0.42926594614982605,
      "learning_rate": 0.0003,
      "loss": 1.3428,
      "step": 1060
    },
    {
      "epoch": 0.17150871568823883,
      "grad_norm": 0.5915783643722534,
      "learning_rate": 0.0003,
      "loss": 1.3263,
      "step": 1070
    },
    {
      "epoch": 0.17311160088158686,
      "grad_norm": 0.4589625597000122,
      "learning_rate": 0.0003,
      "loss": 1.3265,
      "step": 1080
    },
    {
      "epoch": 0.1747144860749349,
      "grad_norm": 0.42623311281204224,
      "learning_rate": 0.0003,
      "loss": 1.325,
      "step": 1090
    },
    {
      "epoch": 0.1763173712682829,
      "grad_norm": 0.42230185866355896,
      "learning_rate": 0.0003,
      "loss": 1.3151,
      "step": 1100
    },
    {
      "epoch": 0.17792025646163093,
      "grad_norm": 0.3890533149242401,
      "learning_rate": 0.0003,
      "loss": 1.3161,
      "step": 1110
    },
    {
      "epoch": 0.17952314165497896,
      "grad_norm": 0.38169047236442566,
      "learning_rate": 0.0003,
      "loss": 1.3171,
      "step": 1120
    },
    {
      "epoch": 0.181126026848327,
      "grad_norm": 0.48665517568588257,
      "learning_rate": 0.0003,
      "loss": 1.3167,
      "step": 1130
    },
    {
      "epoch": 0.18272891204167502,
      "grad_norm": 0.38646721839904785,
      "learning_rate": 0.0003,
      "loss": 1.3233,
      "step": 1140
    },
    {
      "epoch": 0.18433179723502305,
      "grad_norm": 0.384014368057251,
      "learning_rate": 0.0003,
      "loss": 1.305,
      "step": 1150
    },
    {
      "epoch": 0.18593468242837108,
      "grad_norm": 0.5111443400382996,
      "learning_rate": 0.0003,
      "loss": 1.3172,
      "step": 1160
    },
    {
      "epoch": 0.18753756762171908,
      "grad_norm": 0.38314148783683777,
      "learning_rate": 0.0003,
      "loss": 1.3238,
      "step": 1170
    },
    {
      "epoch": 0.1891404528150671,
      "grad_norm": 0.3933526873588562,
      "learning_rate": 0.0003,
      "loss": 1.3103,
      "step": 1180
    },
    {
      "epoch": 0.19074333800841514,
      "grad_norm": 0.38892167806625366,
      "learning_rate": 0.0003,
      "loss": 1.3192,
      "step": 1190
    },
    {
      "epoch": 0.19234622320176317,
      "grad_norm": 0.40912216901779175,
      "learning_rate": 0.0003,
      "loss": 1.293,
      "step": 1200
    },
    {
      "epoch": 0.1939491083951112,
      "grad_norm": 0.3755427300930023,
      "learning_rate": 0.0003,
      "loss": 1.3047,
      "step": 1210
    },
    {
      "epoch": 0.19555199358845923,
      "grad_norm": 0.38369011878967285,
      "learning_rate": 0.0003,
      "loss": 1.3168,
      "step": 1220
    },
    {
      "epoch": 0.19715487878180726,
      "grad_norm": 0.3703300952911377,
      "learning_rate": 0.0003,
      "loss": 1.2966,
      "step": 1230
    },
    {
      "epoch": 0.19875776397515527,
      "grad_norm": 0.4175979495048523,
      "learning_rate": 0.0003,
      "loss": 1.3076,
      "step": 1240
    },
    {
      "epoch": 0.2003606491685033,
      "grad_norm": 0.48666366934776306,
      "learning_rate": 0.0003,
      "loss": 1.3078,
      "step": 1250
    },
    {
      "epoch": 0.20196353436185133,
      "grad_norm": 0.4616610109806061,
      "learning_rate": 0.0003,
      "loss": 1.3139,
      "step": 1260
    },
    {
      "epoch": 0.20356641955519936,
      "grad_norm": 0.4073954224586487,
      "learning_rate": 0.0003,
      "loss": 1.2989,
      "step": 1270
    },
    {
      "epoch": 0.2051693047485474,
      "grad_norm": 0.3653002083301544,
      "learning_rate": 0.0003,
      "loss": 1.3107,
      "step": 1280
    },
    {
      "epoch": 0.20677218994189542,
      "grad_norm": 0.3824729323387146,
      "learning_rate": 0.0003,
      "loss": 1.3152,
      "step": 1290
    },
    {
      "epoch": 0.20837507513524345,
      "grad_norm": 0.4418122470378876,
      "learning_rate": 0.0003,
      "loss": 1.295,
      "step": 1300
    },
    {
      "epoch": 0.20997796032859145,
      "grad_norm": 0.3720099925994873,
      "learning_rate": 0.0003,
      "loss": 1.2732,
      "step": 1310
    },
    {
      "epoch": 0.21158084552193948,
      "grad_norm": 0.3385481536388397,
      "learning_rate": 0.0003,
      "loss": 1.2976,
      "step": 1320
    },
    {
      "epoch": 0.2131837307152875,
      "grad_norm": 0.4389696717262268,
      "learning_rate": 0.0003,
      "loss": 1.3176,
      "step": 1330
    },
    {
      "epoch": 0.21478661590863554,
      "grad_norm": 0.39441773295402527,
      "learning_rate": 0.0003,
      "loss": 1.2758,
      "step": 1340
    },
    {
      "epoch": 0.21638950110198357,
      "grad_norm": 0.4222624599933624,
      "learning_rate": 0.0003,
      "loss": 1.3175,
      "step": 1350
    },
    {
      "epoch": 0.2179923862953316,
      "grad_norm": 0.3878946900367737,
      "learning_rate": 0.0003,
      "loss": 1.3077,
      "step": 1360
    },
    {
      "epoch": 0.21959527148867963,
      "grad_norm": 0.3654826879501343,
      "learning_rate": 0.0003,
      "loss": 1.3031,
      "step": 1370
    },
    {
      "epoch": 0.22119815668202766,
      "grad_norm": 0.4034998416900635,
      "learning_rate": 0.0003,
      "loss": 1.3089,
      "step": 1380
    },
    {
      "epoch": 0.22280104187537567,
      "grad_norm": 0.45924675464630127,
      "learning_rate": 0.0003,
      "loss": 1.2925,
      "step": 1390
    },
    {
      "epoch": 0.2244039270687237,
      "grad_norm": 0.37337183952331543,
      "learning_rate": 0.0003,
      "loss": 1.3007,
      "step": 1400
    },
    {
      "epoch": 0.22600681226207173,
      "grad_norm": 0.41120049357414246,
      "learning_rate": 0.0003,
      "loss": 1.2839,
      "step": 1410
    },
    {
      "epoch": 0.22760969745541976,
      "grad_norm": 0.4049991965293884,
      "learning_rate": 0.0003,
      "loss": 1.2923,
      "step": 1420
    },
    {
      "epoch": 0.2292125826487678,
      "grad_norm": 0.4364527463912964,
      "learning_rate": 0.0003,
      "loss": 1.2924,
      "step": 1430
    },
    {
      "epoch": 0.23081546784211582,
      "grad_norm": 0.39752981066703796,
      "learning_rate": 0.0003,
      "loss": 1.2989,
      "step": 1440
    },
    {
      "epoch": 0.23241835303546385,
      "grad_norm": 0.3233504891395569,
      "learning_rate": 0.0003,
      "loss": 1.2956,
      "step": 1450
    },
    {
      "epoch": 0.23402123822881185,
      "grad_norm": 0.3570205569267273,
      "learning_rate": 0.0003,
      "loss": 1.2837,
      "step": 1460
    },
    {
      "epoch": 0.23562412342215988,
      "grad_norm": 0.3920878767967224,
      "learning_rate": 0.0003,
      "loss": 1.2704,
      "step": 1470
    },
    {
      "epoch": 0.2372270086155079,
      "grad_norm": 0.6673392057418823,
      "learning_rate": 0.0003,
      "loss": 1.2949,
      "step": 1480
    },
    {
      "epoch": 0.23882989380885594,
      "grad_norm": 0.3967534899711609,
      "learning_rate": 0.0003,
      "loss": 1.3099,
      "step": 1490
    },
    {
      "epoch": 0.24043277900220397,
      "grad_norm": 0.375952810049057,
      "learning_rate": 0.0003,
      "loss": 1.2909,
      "step": 1500
    },
    {
      "epoch": 0.242035664195552,
      "grad_norm": 0.4173837900161743,
      "learning_rate": 0.0003,
      "loss": 1.2933,
      "step": 1510
    },
    {
      "epoch": 0.24363854938890003,
      "grad_norm": 0.35205432772636414,
      "learning_rate": 0.0003,
      "loss": 1.2771,
      "step": 1520
    },
    {
      "epoch": 0.24524143458224804,
      "grad_norm": 0.3854178786277771,
      "learning_rate": 0.0003,
      "loss": 1.2817,
      "step": 1530
    },
    {
      "epoch": 0.24684431977559607,
      "grad_norm": 0.3283289670944214,
      "learning_rate": 0.0003,
      "loss": 1.2934,
      "step": 1540
    },
    {
      "epoch": 0.2484472049689441,
      "grad_norm": 0.37120121717453003,
      "learning_rate": 0.0003,
      "loss": 1.2678,
      "step": 1550
    },
    {
      "epoch": 0.2500500901622921,
      "grad_norm": 0.40129926800727844,
      "learning_rate": 0.0003,
      "loss": 1.2887,
      "step": 1560
    },
    {
      "epoch": 0.25165297535564013,
      "grad_norm": 0.3538423776626587,
      "learning_rate": 0.0003,
      "loss": 1.2613,
      "step": 1570
    },
    {
      "epoch": 0.2532558605489882,
      "grad_norm": 0.3907589316368103,
      "learning_rate": 0.0003,
      "loss": 1.2929,
      "step": 1580
    },
    {
      "epoch": 0.2548587457423362,
      "grad_norm": 0.40157246589660645,
      "learning_rate": 0.0003,
      "loss": 1.2925,
      "step": 1590
    },
    {
      "epoch": 0.25646163093568425,
      "grad_norm": 0.4843584895133972,
      "learning_rate": 0.0003,
      "loss": 1.2655,
      "step": 1600
    },
    {
      "epoch": 0.25806451612903225,
      "grad_norm": 0.36788925528526306,
      "learning_rate": 0.0003,
      "loss": 1.2906,
      "step": 1610
    },
    {
      "epoch": 0.2596674013223803,
      "grad_norm": 0.37938061356544495,
      "learning_rate": 0.0003,
      "loss": 1.2812,
      "step": 1620
    },
    {
      "epoch": 0.2612702865157283,
      "grad_norm": 1.707719326019287,
      "learning_rate": 0.0003,
      "loss": 1.2794,
      "step": 1630
    },
    {
      "epoch": 0.2628731717090763,
      "grad_norm": 0.3634716272354126,
      "learning_rate": 0.0003,
      "loss": 1.2969,
      "step": 1640
    },
    {
      "epoch": 0.2644760569024244,
      "grad_norm": 0.4803577959537506,
      "learning_rate": 0.0003,
      "loss": 1.2798,
      "step": 1650
    },
    {
      "epoch": 0.2660789420957724,
      "grad_norm": 0.3735354244709015,
      "learning_rate": 0.0003,
      "loss": 1.2886,
      "step": 1660
    },
    {
      "epoch": 0.26768182728912043,
      "grad_norm": 0.4396728575229645,
      "learning_rate": 0.0003,
      "loss": 1.2823,
      "step": 1670
    },
    {
      "epoch": 0.26928471248246844,
      "grad_norm": 0.4243973195552826,
      "learning_rate": 0.0003,
      "loss": 1.256,
      "step": 1680
    },
    {
      "epoch": 0.2708875976758165,
      "grad_norm": 0.4132402539253235,
      "learning_rate": 0.0003,
      "loss": 1.2922,
      "step": 1690
    },
    {
      "epoch": 0.2724904828691645,
      "grad_norm": 0.4023185968399048,
      "learning_rate": 0.0003,
      "loss": 1.2756,
      "step": 1700
    },
    {
      "epoch": 0.2740933680625125,
      "grad_norm": 0.5241405367851257,
      "learning_rate": 0.0003,
      "loss": 1.2811,
      "step": 1710
    },
    {
      "epoch": 0.27569625325586056,
      "grad_norm": 0.499695748090744,
      "learning_rate": 0.0003,
      "loss": 1.2771,
      "step": 1720
    },
    {
      "epoch": 0.27729913844920856,
      "grad_norm": 0.4254522919654846,
      "learning_rate": 0.0003,
      "loss": 1.2753,
      "step": 1730
    },
    {
      "epoch": 0.2789020236425566,
      "grad_norm": 0.4126877784729004,
      "learning_rate": 0.0003,
      "loss": 1.2732,
      "step": 1740
    },
    {
      "epoch": 0.2805049088359046,
      "grad_norm": 0.4045843482017517,
      "learning_rate": 0.0003,
      "loss": 1.2514,
      "step": 1750
    },
    {
      "epoch": 0.2821077940292527,
      "grad_norm": 0.3707495331764221,
      "learning_rate": 0.0003,
      "loss": 1.2886,
      "step": 1760
    },
    {
      "epoch": 0.2837106792226007,
      "grad_norm": 0.35994088649749756,
      "learning_rate": 0.0003,
      "loss": 1.2592,
      "step": 1770
    },
    {
      "epoch": 0.2853135644159487,
      "grad_norm": 0.3238936960697174,
      "learning_rate": 0.0003,
      "loss": 1.2746,
      "step": 1780
    },
    {
      "epoch": 0.28691644960929674,
      "grad_norm": 0.35237905383110046,
      "learning_rate": 0.0003,
      "loss": 1.2568,
      "step": 1790
    },
    {
      "epoch": 0.28851933480264474,
      "grad_norm": 0.42206379771232605,
      "learning_rate": 0.0003,
      "loss": 1.2609,
      "step": 1800
    },
    {
      "epoch": 0.2901222199959928,
      "grad_norm": 0.33885058760643005,
      "learning_rate": 0.0003,
      "loss": 1.2711,
      "step": 1810
    },
    {
      "epoch": 0.2917251051893408,
      "grad_norm": 0.4460236430168152,
      "learning_rate": 0.0003,
      "loss": 1.2631,
      "step": 1820
    },
    {
      "epoch": 0.29332799038268886,
      "grad_norm": 0.49559280276298523,
      "learning_rate": 0.0003,
      "loss": 1.2793,
      "step": 1830
    },
    {
      "epoch": 0.29493087557603687,
      "grad_norm": 0.36408957839012146,
      "learning_rate": 0.0003,
      "loss": 1.2963,
      "step": 1840
    },
    {
      "epoch": 0.29653376076938487,
      "grad_norm": 0.38527244329452515,
      "learning_rate": 0.0003,
      "loss": 1.2634,
      "step": 1850
    },
    {
      "epoch": 0.2981366459627329,
      "grad_norm": 0.3253726661205292,
      "learning_rate": 0.0003,
      "loss": 1.2681,
      "step": 1860
    },
    {
      "epoch": 0.29973953115608093,
      "grad_norm": 0.3659113943576813,
      "learning_rate": 0.0003,
      "loss": 1.2682,
      "step": 1870
    },
    {
      "epoch": 0.301342416349429,
      "grad_norm": 0.35942474007606506,
      "learning_rate": 0.0003,
      "loss": 1.2598,
      "step": 1880
    },
    {
      "epoch": 0.302945301542777,
      "grad_norm": 0.3608860373497009,
      "learning_rate": 0.0003,
      "loss": 1.252,
      "step": 1890
    },
    {
      "epoch": 0.30454818673612505,
      "grad_norm": 0.4459868371486664,
      "learning_rate": 0.0003,
      "loss": 1.2659,
      "step": 1900
    },
    {
      "epoch": 0.30615107192947305,
      "grad_norm": 0.36545702815055847,
      "learning_rate": 0.0003,
      "loss": 1.2613,
      "step": 1910
    },
    {
      "epoch": 0.30775395712282105,
      "grad_norm": 0.419463574886322,
      "learning_rate": 0.0003,
      "loss": 1.2688,
      "step": 1920
    },
    {
      "epoch": 0.3093568423161691,
      "grad_norm": 0.3596417307853699,
      "learning_rate": 0.0003,
      "loss": 1.2729,
      "step": 1930
    },
    {
      "epoch": 0.3109597275095171,
      "grad_norm": 0.3644123673439026,
      "learning_rate": 0.0003,
      "loss": 1.2707,
      "step": 1940
    },
    {
      "epoch": 0.3125626127028652,
      "grad_norm": 0.5335139632225037,
      "learning_rate": 0.0003,
      "loss": 1.2569,
      "step": 1950
    },
    {
      "epoch": 0.3141654978962132,
      "grad_norm": 0.37530243396759033,
      "learning_rate": 0.0003,
      "loss": 1.25,
      "step": 1960
    },
    {
      "epoch": 0.31576838308956123,
      "grad_norm": 0.3539182245731354,
      "learning_rate": 0.0003,
      "loss": 1.2352,
      "step": 1970
    },
    {
      "epoch": 0.31737126828290924,
      "grad_norm": 0.34524521231651306,
      "learning_rate": 0.0003,
      "loss": 1.2452,
      "step": 1980
    },
    {
      "epoch": 0.31897415347625724,
      "grad_norm": 0.39482155442237854,
      "learning_rate": 0.0003,
      "loss": 1.2797,
      "step": 1990
    },
    {
      "epoch": 0.3205770386696053,
      "grad_norm": 0.5608816742897034,
      "learning_rate": 0.0003,
      "loss": 1.2427,
      "step": 2000
    },
    {
      "epoch": 0.3221799238629533,
      "grad_norm": 0.4138280153274536,
      "learning_rate": 0.0003,
      "loss": 1.2838,
      "step": 2010
    },
    {
      "epoch": 0.32378280905630136,
      "grad_norm": 0.4490892291069031,
      "learning_rate": 0.0003,
      "loss": 1.2663,
      "step": 2020
    },
    {
      "epoch": 0.32538569424964936,
      "grad_norm": 0.3887448310852051,
      "learning_rate": 0.0003,
      "loss": 1.2512,
      "step": 2030
    },
    {
      "epoch": 0.3269885794429974,
      "grad_norm": 0.34303218126296997,
      "learning_rate": 0.0003,
      "loss": 1.2526,
      "step": 2040
    },
    {
      "epoch": 0.3285914646363454,
      "grad_norm": 0.34844136238098145,
      "learning_rate": 0.0003,
      "loss": 1.2748,
      "step": 2050
    },
    {
      "epoch": 0.3301943498296934,
      "grad_norm": 0.3607269525527954,
      "learning_rate": 0.0003,
      "loss": 1.2682,
      "step": 2060
    },
    {
      "epoch": 0.3317972350230415,
      "grad_norm": 0.3467775583267212,
      "learning_rate": 0.0003,
      "loss": 1.253,
      "step": 2070
    },
    {
      "epoch": 0.3334001202163895,
      "grad_norm": 0.39770326018333435,
      "learning_rate": 0.0003,
      "loss": 1.2309,
      "step": 2080
    },
    {
      "epoch": 0.33500300540973754,
      "grad_norm": 0.3290434777736664,
      "learning_rate": 0.0003,
      "loss": 1.2534,
      "step": 2090
    },
    {
      "epoch": 0.33660589060308554,
      "grad_norm": 0.4442046284675598,
      "learning_rate": 0.0003,
      "loss": 1.2594,
      "step": 2100
    },
    {
      "epoch": 0.3382087757964336,
      "grad_norm": 0.3657543957233429,
      "learning_rate": 0.0003,
      "loss": 1.2605,
      "step": 2110
    },
    {
      "epoch": 0.3398116609897816,
      "grad_norm": 0.5030764937400818,
      "learning_rate": 0.0003,
      "loss": 1.2521,
      "step": 2120
    },
    {
      "epoch": 0.3414145461831296,
      "grad_norm": 0.33297237753868103,
      "learning_rate": 0.0003,
      "loss": 1.2601,
      "step": 2130
    },
    {
      "epoch": 0.34301743137647767,
      "grad_norm": 0.38466474413871765,
      "learning_rate": 0.0003,
      "loss": 1.2518,
      "step": 2140
    },
    {
      "epoch": 0.34462031656982567,
      "grad_norm": 0.5619605183601379,
      "learning_rate": 0.0003,
      "loss": 1.2417,
      "step": 2150
    },
    {
      "epoch": 0.3462232017631737,
      "grad_norm": 0.3368251621723175,
      "learning_rate": 0.0003,
      "loss": 1.2534,
      "step": 2160
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.3819466829299927,
      "learning_rate": 0.0003,
      "loss": 1.2437,
      "step": 2170
    },
    {
      "epoch": 0.3494289721498698,
      "grad_norm": 0.3725869059562683,
      "learning_rate": 0.0003,
      "loss": 1.2668,
      "step": 2180
    },
    {
      "epoch": 0.3510318573432178,
      "grad_norm": 0.5592432618141174,
      "learning_rate": 0.0003,
      "loss": 1.2382,
      "step": 2190
    },
    {
      "epoch": 0.3526347425365658,
      "grad_norm": 0.37191301584243774,
      "learning_rate": 0.0003,
      "loss": 1.2469,
      "step": 2200
    },
    {
      "epoch": 0.35423762772991385,
      "grad_norm": 0.3820264935493469,
      "learning_rate": 0.0003,
      "loss": 1.2503,
      "step": 2210
    },
    {
      "epoch": 0.35584051292326185,
      "grad_norm": 0.39766186475753784,
      "learning_rate": 0.0003,
      "loss": 1.236,
      "step": 2220
    },
    {
      "epoch": 0.3574433981166099,
      "grad_norm": 0.3511972725391388,
      "learning_rate": 0.0003,
      "loss": 1.2758,
      "step": 2230
    },
    {
      "epoch": 0.3590462833099579,
      "grad_norm": 0.3821338415145874,
      "learning_rate": 0.0003,
      "loss": 1.2609,
      "step": 2240
    },
    {
      "epoch": 0.360649168503306,
      "grad_norm": 0.536491334438324,
      "learning_rate": 0.0003,
      "loss": 1.2343,
      "step": 2250
    },
    {
      "epoch": 0.362252053696654,
      "grad_norm": 0.4457603991031647,
      "learning_rate": 0.0003,
      "loss": 1.2382,
      "step": 2260
    },
    {
      "epoch": 0.363854938890002,
      "grad_norm": 0.3784075975418091,
      "learning_rate": 0.0003,
      "loss": 1.2496,
      "step": 2270
    },
    {
      "epoch": 0.36545782408335004,
      "grad_norm": 0.3726901710033417,
      "learning_rate": 0.0003,
      "loss": 1.2509,
      "step": 2280
    },
    {
      "epoch": 0.36706070927669804,
      "grad_norm": 0.3271140158176422,
      "learning_rate": 0.0003,
      "loss": 1.2544,
      "step": 2290
    },
    {
      "epoch": 0.3686635944700461,
      "grad_norm": 0.4778403341770172,
      "learning_rate": 0.0003,
      "loss": 1.2436,
      "step": 2300
    },
    {
      "epoch": 0.3702664796633941,
      "grad_norm": 0.348621129989624,
      "learning_rate": 0.0003,
      "loss": 1.2574,
      "step": 2310
    },
    {
      "epoch": 0.37186936485674216,
      "grad_norm": 0.3956087827682495,
      "learning_rate": 0.0003,
      "loss": 1.232,
      "step": 2320
    },
    {
      "epoch": 0.37347225005009016,
      "grad_norm": 0.39291661977767944,
      "learning_rate": 0.0003,
      "loss": 1.224,
      "step": 2330
    },
    {
      "epoch": 0.37507513524343816,
      "grad_norm": 0.36051276326179504,
      "learning_rate": 0.0003,
      "loss": 1.2623,
      "step": 2340
    },
    {
      "epoch": 0.3766780204367862,
      "grad_norm": 0.40664902329444885,
      "learning_rate": 0.0003,
      "loss": 1.2336,
      "step": 2350
    },
    {
      "epoch": 0.3782809056301342,
      "grad_norm": 0.5174034833908081,
      "learning_rate": 0.0003,
      "loss": 1.2134,
      "step": 2360
    },
    {
      "epoch": 0.3798837908234823,
      "grad_norm": 0.3292120695114136,
      "learning_rate": 0.0003,
      "loss": 1.2684,
      "step": 2370
    },
    {
      "epoch": 0.3814866760168303,
      "grad_norm": 0.410427063703537,
      "learning_rate": 0.0003,
      "loss": 1.24,
      "step": 2380
    },
    {
      "epoch": 0.38308956121017834,
      "grad_norm": 0.3474559783935547,
      "learning_rate": 0.0003,
      "loss": 1.2555,
      "step": 2390
    },
    {
      "epoch": 0.38469244640352634,
      "grad_norm": 0.4369237422943115,
      "learning_rate": 0.0003,
      "loss": 1.2514,
      "step": 2400
    },
    {
      "epoch": 0.38629533159687435,
      "grad_norm": 0.3765268325805664,
      "learning_rate": 0.0003,
      "loss": 1.2618,
      "step": 2410
    },
    {
      "epoch": 0.3878982167902224,
      "grad_norm": 0.3871352970600128,
      "learning_rate": 0.0003,
      "loss": 1.2421,
      "step": 2420
    },
    {
      "epoch": 0.3895011019835704,
      "grad_norm": 0.3790617883205414,
      "learning_rate": 0.0003,
      "loss": 1.2567,
      "step": 2430
    },
    {
      "epoch": 0.39110398717691847,
      "grad_norm": 0.43595749139785767,
      "learning_rate": 0.0003,
      "loss": 1.253,
      "step": 2440
    },
    {
      "epoch": 0.39270687237026647,
      "grad_norm": 0.36090561747550964,
      "learning_rate": 0.0003,
      "loss": 1.2455,
      "step": 2450
    },
    {
      "epoch": 0.3943097575636145,
      "grad_norm": 0.34247252345085144,
      "learning_rate": 0.0003,
      "loss": 1.253,
      "step": 2460
    },
    {
      "epoch": 0.39591264275696253,
      "grad_norm": 0.47529610991477966,
      "learning_rate": 0.0003,
      "loss": 1.2319,
      "step": 2470
    },
    {
      "epoch": 0.39751552795031053,
      "grad_norm": 0.3489156663417816,
      "learning_rate": 0.0003,
      "loss": 1.2038,
      "step": 2480
    },
    {
      "epoch": 0.3991184131436586,
      "grad_norm": 0.35935741662979126,
      "learning_rate": 0.0003,
      "loss": 1.2409,
      "step": 2490
    },
    {
      "epoch": 0.4007212983370066,
      "grad_norm": 0.34685826301574707,
      "learning_rate": 0.0003,
      "loss": 1.2399,
      "step": 2500
    },
    {
      "epoch": 0.4007212983370066,
      "step": 2500,
      "total_flos": 6.105286991832678e+19,
      "train_loss": 1.4753759937286377,
      "train_runtime": 99730.4637,
      "train_samples_per_second": 6.417,
      "train_steps_per_second": 0.025
    }
  ],
  "logging_steps": 10,
  "max_steps": 2500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.105286991832678e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}