{
  "best_metric": 0.9899861346479741,
  "best_model_checkpoint": "deit-base-patch16-224-finetuned-lora-medmnistv2/checkpoint-4324",
  "epoch": 9.990749306197966,
  "eval_steps": 500,
  "global_step": 5400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 2.755105972290039,
      "learning_rate": 0.004990740740740741,
      "loss": 1.8298,
      "step": 10
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.729515790939331,
      "learning_rate": 0.004981481481481482,
      "loss": 1.2318,
      "step": 20
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.507834792137146,
      "learning_rate": 0.0049722222222222225,
      "loss": 1.0656,
      "step": 30
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.6946210861206055,
      "learning_rate": 0.004962962962962963,
      "loss": 1.0045,
      "step": 40
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.5376859903335571,
      "learning_rate": 0.004953703703703703,
      "loss": 0.952,
      "step": 50
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.783308982849121,
      "learning_rate": 0.004944444444444445,
      "loss": 0.808,
      "step": 60
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.132023811340332,
      "learning_rate": 0.004935185185185186,
      "loss": 0.7731,
      "step": 70
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.7192047834396362,
      "learning_rate": 0.004925925925925926,
      "loss": 0.8522,
      "step": 80
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.745439052581787,
      "learning_rate": 0.004916666666666666,
      "loss": 0.7773,
      "step": 90
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.608622670173645,
      "learning_rate": 0.004907407407407408,
      "loss": 0.7742,
      "step": 100
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.5263983011245728,
      "learning_rate": 0.004898148148148148,
      "loss": 0.8207,
      "step": 110
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.6577465534210205,
      "learning_rate": 0.004888888888888889,
      "loss": 0.7417,
      "step": 120
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.3107540607452393,
      "learning_rate": 0.00487962962962963,
      "loss": 0.7652,
      "step": 130
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.8946397304534912,
      "learning_rate": 0.00487037037037037,
      "loss": 0.763,
      "step": 140
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.175830125808716,
      "learning_rate": 0.004861111111111111,
      "loss": 0.7303,
      "step": 150
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.8831381797790527,
      "learning_rate": 0.004851851851851852,
      "loss": 0.7505,
      "step": 160
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.2608139514923096,
      "learning_rate": 0.004842592592592593,
      "loss": 0.6704,
      "step": 170
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.5122900009155273,
      "learning_rate": 0.004833333333333334,
      "loss": 0.7428,
      "step": 180
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.939324975013733,
      "learning_rate": 0.004824074074074074,
      "loss": 0.7112,
      "step": 190
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.1237051486968994,
      "learning_rate": 0.004814814814814814,
      "loss": 0.7419,
      "step": 200
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.0236401557922363,
      "learning_rate": 0.004805555555555556,
      "loss": 0.6617,
      "step": 210
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.503143310546875,
      "learning_rate": 0.004796296296296297,
      "loss": 0.7201,
      "step": 220
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.937041163444519,
      "learning_rate": 0.004787037037037037,
      "loss": 0.6503,
      "step": 230
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.2516043186187744,
      "learning_rate": 0.004777777777777778,
      "loss": 0.6473,
      "step": 240
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.2114853858947754,
      "learning_rate": 0.004768518518518518,
      "loss": 0.6475,
      "step": 250
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.8521231412887573,
      "learning_rate": 0.004759259259259259,
      "loss": 0.6729,
      "step": 260
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.599143147468567,
      "learning_rate": 0.00475,
      "loss": 0.6437,
      "step": 270
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.5192047357559204,
      "learning_rate": 0.004740740740740741,
      "loss": 0.6731,
      "step": 280
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.874036431312561,
      "learning_rate": 0.0047314814814814815,
      "loss": 0.707,
      "step": 290
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.582754135131836,
      "learning_rate": 0.004722222222222222,
      "loss": 0.6477,
      "step": 300
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.7915743589401245,
      "learning_rate": 0.004712962962962963,
      "loss": 0.6226,
      "step": 310
    },
    {
      "epoch": 0.59,
      "grad_norm": 2.2843360900878906,
      "learning_rate": 0.004703703703703704,
      "loss": 0.7287,
      "step": 320
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.8337290287017822,
      "learning_rate": 0.004694444444444445,
      "loss": 0.652,
      "step": 330
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.7287871837615967,
      "learning_rate": 0.0046851851851851855,
      "loss": 0.5503,
      "step": 340
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.7953327894210815,
      "learning_rate": 0.004675925925925926,
      "loss": 0.5608,
      "step": 350
    },
    {
      "epoch": 0.67,
      "grad_norm": 3.0038208961486816,
      "learning_rate": 0.004666666666666667,
      "loss": 0.7144,
      "step": 360
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.8996288776397705,
      "learning_rate": 0.004657407407407408,
      "loss": 0.6965,
      "step": 370
    },
    {
      "epoch": 0.7,
      "grad_norm": 3.314239740371704,
      "learning_rate": 0.004648148148148148,
      "loss": 0.7012,
      "step": 380
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.629176139831543,
      "learning_rate": 0.004639814814814815,
      "loss": 0.6977,
      "step": 390
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.343112826347351,
      "learning_rate": 0.004630555555555555,
      "loss": 0.7257,
      "step": 400
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.4948642253875732,
      "learning_rate": 0.004621296296296296,
      "loss": 0.7244,
      "step": 410
    },
    {
      "epoch": 0.78,
      "grad_norm": 2.248868942260742,
      "learning_rate": 0.004612037037037038,
      "loss": 0.6938,
      "step": 420
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.4094672203063965,
      "learning_rate": 0.004602777777777778,
      "loss": 0.6338,
      "step": 430
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.7461600303649902,
      "learning_rate": 0.0045935185185185185,
      "loss": 0.623,
      "step": 440
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.9258304834365845,
      "learning_rate": 0.004584259259259259,
      "loss": 0.6681,
      "step": 450
    },
    {
      "epoch": 0.85,
      "grad_norm": 2.1258721351623535,
      "learning_rate": 0.004575,
      "loss": 0.6885,
      "step": 460
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.852216362953186,
      "learning_rate": 0.004565740740740741,
      "loss": 0.6815,
      "step": 470
    },
    {
      "epoch": 0.89,
      "grad_norm": 2.523413896560669,
      "learning_rate": 0.004556481481481482,
      "loss": 0.6267,
      "step": 480
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.133388042449951,
      "learning_rate": 0.0045472222222222225,
      "loss": 0.6133,
      "step": 490
    },
    {
      "epoch": 0.93,
      "grad_norm": 2.732682228088379,
      "learning_rate": 0.004537962962962963,
      "loss": 0.7924,
      "step": 500
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.4026442766189575,
      "learning_rate": 0.004528703703703704,
      "loss": 0.5851,
      "step": 510
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.5229674577713013,
      "learning_rate": 0.004519444444444444,
      "loss": 0.6732,
      "step": 520
    },
    {
      "epoch": 0.98,
      "grad_norm": 2.2832190990448,
      "learning_rate": 0.004510185185185186,
      "loss": 0.6453,
      "step": 530
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.637460708618164,
      "learning_rate": 0.0045009259259259264,
      "loss": 0.5849,
      "step": 540
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9442304729625636,
      "eval_f1": 0.9284990016049455,
      "eval_loss": 0.18415337800979614,
      "eval_precision": 0.9449212426032028,
      "eval_recall": 0.92677489589942,
      "eval_runtime": 53.3063,
      "eval_samples_per_second": 121.768,
      "eval_steps_per_second": 7.616,
      "step": 540
    },
    {
      "epoch": 1.02,
      "grad_norm": 1.7663180828094482,
      "learning_rate": 0.004491666666666666,
      "loss": 0.6677,
      "step": 550
    },
    {
      "epoch": 1.04,
      "grad_norm": 1.3648691177368164,
      "learning_rate": 0.004482407407407407,
      "loss": 0.6105,
      "step": 560
    },
    {
      "epoch": 1.05,
      "grad_norm": 1.8574776649475098,
      "learning_rate": 0.004473148148148149,
      "loss": 0.6633,
      "step": 570
    },
    {
      "epoch": 1.07,
      "grad_norm": 2.4903650283813477,
      "learning_rate": 0.004463888888888889,
      "loss": 0.5867,
      "step": 580
    },
    {
      "epoch": 1.09,
      "grad_norm": 1.7570849657058716,
      "learning_rate": 0.0044546296296296296,
      "loss": 0.6011,
      "step": 590
    },
    {
      "epoch": 1.11,
      "grad_norm": 1.796229600906372,
      "learning_rate": 0.00444537037037037,
      "loss": 0.565,
      "step": 600
    },
    {
      "epoch": 1.13,
      "grad_norm": 1.2062430381774902,
      "learning_rate": 0.004436111111111111,
      "loss": 0.6809,
      "step": 610
    },
    {
      "epoch": 1.15,
      "grad_norm": 1.641988754272461,
      "learning_rate": 0.004426851851851852,
      "loss": 0.6778,
      "step": 620
    },
    {
      "epoch": 1.17,
      "grad_norm": 2.171339988708496,
      "learning_rate": 0.004417592592592593,
      "loss": 0.6091,
      "step": 630
    },
    {
      "epoch": 1.18,
      "grad_norm": 1.634817123413086,
      "learning_rate": 0.0044083333333333335,
      "loss": 0.5903,
      "step": 640
    },
    {
      "epoch": 1.2,
      "grad_norm": 1.8555738925933838,
      "learning_rate": 0.004399074074074074,
      "loss": 0.6029,
      "step": 650
    },
    {
      "epoch": 1.22,
      "grad_norm": 1.2326161861419678,
      "learning_rate": 0.004389814814814815,
      "loss": 0.6358,
      "step": 660
    },
    {
      "epoch": 1.24,
      "grad_norm": 1.5019110441207886,
      "learning_rate": 0.004380555555555555,
      "loss": 0.6457,
      "step": 670
    },
    {
      "epoch": 1.26,
      "grad_norm": 2.464613199234009,
      "learning_rate": 0.004371296296296297,
      "loss": 0.6165,
      "step": 680
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.4740674495697021,
      "learning_rate": 0.0043620370370370375,
      "loss": 0.6324,
      "step": 690
    },
    {
      "epoch": 1.3,
      "grad_norm": 2.2495017051696777,
      "learning_rate": 0.0043527777777777775,
      "loss": 0.6903,
      "step": 700
    },
    {
      "epoch": 1.31,
      "grad_norm": 1.0371023416519165,
      "learning_rate": 0.004343518518518519,
      "loss": 0.7259,
      "step": 710
    },
    {
      "epoch": 1.33,
      "grad_norm": 1.4419306516647339,
      "learning_rate": 0.004334259259259259,
      "loss": 0.5841,
      "step": 720
    },
    {
      "epoch": 1.35,
      "grad_norm": 2.1435110569000244,
      "learning_rate": 0.004325,
      "loss": 0.5803,
      "step": 730
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.697944164276123,
      "learning_rate": 0.004315740740740741,
      "loss": 0.5959,
      "step": 740
    },
    {
      "epoch": 1.39,
      "grad_norm": 2.058931589126587,
      "learning_rate": 0.0043064814814814814,
      "loss": 0.619,
      "step": 750
    },
    {
      "epoch": 1.41,
      "grad_norm": 1.3956432342529297,
      "learning_rate": 0.004297222222222222,
      "loss": 0.6453,
      "step": 760
    },
    {
      "epoch": 1.42,
      "grad_norm": 1.9578568935394287,
      "learning_rate": 0.004287962962962963,
      "loss": 0.5163,
      "step": 770
    },
    {
      "epoch": 1.44,
      "grad_norm": 2.150716781616211,
      "learning_rate": 0.004278703703703704,
      "loss": 0.6397,
      "step": 780
    },
    {
      "epoch": 1.46,
      "grad_norm": 2.469938278198242,
      "learning_rate": 0.004269444444444445,
      "loss": 0.6264,
      "step": 790
    },
    {
      "epoch": 1.48,
      "grad_norm": 2.8132483959198,
      "learning_rate": 0.004260185185185185,
      "loss": 0.574,
      "step": 800
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.2079416513442993,
      "learning_rate": 0.004250925925925926,
      "loss": 0.5735,
      "step": 810
    },
    {
      "epoch": 1.52,
      "grad_norm": 1.7141121625900269,
      "learning_rate": 0.004241666666666667,
      "loss": 0.6086,
      "step": 820
    },
    {
      "epoch": 1.54,
      "grad_norm": 1.2220849990844727,
      "learning_rate": 0.004232407407407408,
      "loss": 0.6525,
      "step": 830
    },
    {
      "epoch": 1.55,
      "grad_norm": 2.2290451526641846,
      "learning_rate": 0.004223148148148149,
      "loss": 0.6515,
      "step": 840
    },
    {
      "epoch": 1.57,
      "grad_norm": 1.285662293434143,
      "learning_rate": 0.0042138888888888885,
      "loss": 0.6119,
      "step": 850
    },
    {
      "epoch": 1.59,
      "grad_norm": 1.6702948808670044,
      "learning_rate": 0.00420462962962963,
      "loss": 0.5591,
      "step": 860
    },
    {
      "epoch": 1.61,
      "grad_norm": 1.3915163278579712,
      "learning_rate": 0.00419537037037037,
      "loss": 0.5742,
      "step": 870
    },
    {
      "epoch": 1.63,
      "grad_norm": 1.0722817182540894,
      "learning_rate": 0.004186111111111111,
      "loss": 0.5443,
      "step": 880
    },
    {
      "epoch": 1.65,
      "grad_norm": 1.2701308727264404,
      "learning_rate": 0.004176851851851852,
      "loss": 0.5407,
      "step": 890
    },
    {
      "epoch": 1.67,
      "grad_norm": 2.4829461574554443,
      "learning_rate": 0.0041675925925925925,
      "loss": 0.617,
      "step": 900
    },
    {
      "epoch": 1.68,
      "grad_norm": 1.6107633113861084,
      "learning_rate": 0.004158333333333333,
      "loss": 0.6352,
      "step": 910
    },
    {
      "epoch": 1.7,
      "grad_norm": 1.6750290393829346,
      "learning_rate": 0.004149074074074074,
      "loss": 0.6186,
      "step": 920
    },
    {
      "epoch": 1.72,
      "grad_norm": 3.0813207626342773,
      "learning_rate": 0.004139814814814815,
      "loss": 0.6398,
      "step": 930
    },
    {
      "epoch": 1.74,
      "grad_norm": 1.46220862865448,
      "learning_rate": 0.004130555555555556,
      "loss": 0.6248,
      "step": 940
    },
    {
      "epoch": 1.76,
      "grad_norm": 1.8916534185409546,
      "learning_rate": 0.0041212962962962965,
      "loss": 0.5828,
      "step": 950
    },
    {
      "epoch": 1.78,
      "grad_norm": 1.397637128829956,
      "learning_rate": 0.004112037037037037,
      "loss": 0.6127,
      "step": 960
    },
    {
      "epoch": 1.79,
      "grad_norm": 1.5524698495864868,
      "learning_rate": 0.004102777777777778,
      "loss": 0.5434,
      "step": 970
    },
    {
      "epoch": 1.81,
      "grad_norm": 1.357723355293274,
      "learning_rate": 0.004093518518518519,
      "loss": 0.565,
      "step": 980
    },
    {
      "epoch": 1.83,
      "grad_norm": 2.0358312129974365,
      "learning_rate": 0.004084259259259259,
      "loss": 0.6351,
      "step": 990
    },
    {
      "epoch": 1.85,
      "grad_norm": 2.148568868637085,
      "learning_rate": 0.004075,
      "loss": 0.6259,
      "step": 1000
    },
    {
      "epoch": 1.87,
      "grad_norm": 1.5996747016906738,
      "learning_rate": 0.004065740740740741,
      "loss": 0.6821,
      "step": 1010
    },
    {
      "epoch": 1.89,
      "grad_norm": 2.8041980266571045,
      "learning_rate": 0.004056481481481481,
      "loss": 0.6141,
      "step": 1020
    },
    {
      "epoch": 1.91,
      "grad_norm": 2.5319840908050537,
      "learning_rate": 0.004047222222222222,
      "loss": 0.5978,
      "step": 1030
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.356850266456604,
      "learning_rate": 0.004037962962962964,
      "loss": 0.6444,
      "step": 1040
    },
    {
      "epoch": 1.94,
      "grad_norm": 1.4726742506027222,
      "learning_rate": 0.004028703703703704,
      "loss": 0.5937,
      "step": 1050
    },
    {
      "epoch": 1.96,
      "grad_norm": 2.525775194168091,
      "learning_rate": 0.004019444444444444,
      "loss": 0.5609,
      "step": 1060
    },
    {
      "epoch": 1.98,
      "grad_norm": 2.2801449298858643,
      "learning_rate": 0.004010185185185185,
      "loss": 0.6315,
      "step": 1070
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.0325303077697754,
      "learning_rate": 0.004000925925925926,
      "loss": 0.6494,
      "step": 1080
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9499306732398706,
      "eval_f1": 0.9509024323746281,
      "eval_loss": 0.14330759644508362,
      "eval_precision": 0.9538743946507812,
      "eval_recall": 0.9510137440302636,
      "eval_runtime": 53.345,
      "eval_samples_per_second": 121.68,
      "eval_steps_per_second": 7.611,
      "step": 1081
    },
    {
      "epoch": 2.02,
      "grad_norm": 1.2894580364227295,
      "learning_rate": 0.003991666666666667,
      "loss": 0.5896,
      "step": 1090
    },
    {
      "epoch": 2.04,
      "grad_norm": 1.835950255393982,
      "learning_rate": 0.003982407407407408,
      "loss": 0.6027,
      "step": 1100
    },
    {
      "epoch": 2.05,
      "grad_norm": 2.6905369758605957,
      "learning_rate": 0.0039731481481481475,
      "loss": 0.5896,
      "step": 1110
    },
    {
      "epoch": 2.07,
      "grad_norm": 1.3313945531845093,
      "learning_rate": 0.003963888888888889,
      "loss": 0.5735,
      "step": 1120
    },
    {
      "epoch": 2.09,
      "grad_norm": 1.584228754043579,
      "learning_rate": 0.00395462962962963,
      "loss": 0.5194,
      "step": 1130
    },
    {
      "epoch": 2.11,
      "grad_norm": 1.3231537342071533,
      "learning_rate": 0.00394537037037037,
      "loss": 0.556,
      "step": 1140
    },
    {
      "epoch": 2.13,
      "grad_norm": 2.2847185134887695,
      "learning_rate": 0.003936111111111112,
      "loss": 0.5564,
      "step": 1150
    },
    {
      "epoch": 2.15,
      "grad_norm": 1.5955736637115479,
      "learning_rate": 0.003926851851851852,
      "loss": 0.6138,
      "step": 1160
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.2592884302139282,
      "learning_rate": 0.003917592592592592,
      "loss": 0.6048,
      "step": 1170
    },
    {
      "epoch": 2.18,
      "grad_norm": 2.0293960571289062,
      "learning_rate": 0.003908333333333333,
      "loss": 0.5609,
      "step": 1180
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.5888597965240479,
      "learning_rate": 0.0038990740740740743,
      "loss": 0.6054,
      "step": 1190
    },
    {
      "epoch": 2.22,
      "grad_norm": 2.2508180141448975,
      "learning_rate": 0.003889814814814815,
      "loss": 0.596,
      "step": 1200
    },
    {
      "epoch": 2.24,
      "grad_norm": 1.29166841506958,
      "learning_rate": 0.0038805555555555555,
      "loss": 0.6829,
      "step": 1210
    },
    {
      "epoch": 2.26,
      "grad_norm": 1.7769325971603394,
      "learning_rate": 0.0038712962962962967,
      "loss": 0.6646,
      "step": 1220
    },
    {
      "epoch": 2.28,
      "grad_norm": 1.0019413232803345,
      "learning_rate": 0.003862037037037037,
      "loss": 0.563,
      "step": 1230
    },
    {
      "epoch": 2.29,
      "grad_norm": 1.9900081157684326,
      "learning_rate": 0.003852777777777778,
      "loss": 0.5392,
      "step": 1240
    },
    {
      "epoch": 2.31,
      "grad_norm": 2.0797948837280273,
      "learning_rate": 0.0038435185185185182,
      "loss": 0.5434,
      "step": 1250
    },
    {
      "epoch": 2.33,
      "grad_norm": 1.445152997970581,
      "learning_rate": 0.0038342592592592595,
      "loss": 0.567,
      "step": 1260
    },
    {
      "epoch": 2.35,
      "grad_norm": 1.6301777362823486,
      "learning_rate": 0.0038250000000000003,
      "loss": 0.6221,
      "step": 1270
    },
    {
      "epoch": 2.37,
      "grad_norm": 1.7782893180847168,
      "learning_rate": 0.0038157407407407406,
      "loss": 0.5987,
      "step": 1280
    },
    {
      "epoch": 2.39,
      "grad_norm": 1.4305975437164307,
      "learning_rate": 0.0038064814814814814,
      "loss": 0.5985,
      "step": 1290
    },
    {
      "epoch": 2.41,
      "grad_norm": 1.4855763912200928,
      "learning_rate": 0.0037972222222222227,
      "loss": 0.608,
      "step": 1300
    },
    {
      "epoch": 2.42,
      "grad_norm": 1.2296168804168701,
      "learning_rate": 0.003787962962962963,
      "loss": 0.6003,
      "step": 1310
    },
    {
      "epoch": 2.44,
      "grad_norm": 1.654550313949585,
      "learning_rate": 0.003778703703703704,
      "loss": 0.5832,
      "step": 1320
    },
    {
      "epoch": 2.46,
      "grad_norm": 1.3498139381408691,
      "learning_rate": 0.0037694444444444446,
      "loss": 0.5549,
      "step": 1330
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.7986209392547607,
      "learning_rate": 0.0037601851851851854,
      "loss": 0.6364,
      "step": 1340
    },
    {
      "epoch": 2.5,
      "grad_norm": 2.0304336547851562,
      "learning_rate": 0.0037509259259259258,
      "loss": 0.5549,
      "step": 1350
    },
    {
      "epoch": 2.52,
      "grad_norm": 1.7919527292251587,
      "learning_rate": 0.0037416666666666666,
      "loss": 0.6075,
      "step": 1360
    },
    {
      "epoch": 2.53,
      "grad_norm": 0.9223549365997314,
      "learning_rate": 0.003732407407407408,
      "loss": 0.5013,
      "step": 1370
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.9208143949508667,
      "learning_rate": 0.003723148148148148,
      "loss": 0.596,
      "step": 1380
    },
    {
      "epoch": 2.57,
      "grad_norm": 1.878339171409607,
      "learning_rate": 0.003713888888888889,
      "loss": 0.5878,
      "step": 1390
    },
    {
      "epoch": 2.59,
      "grad_norm": 1.3851909637451172,
      "learning_rate": 0.0037046296296296293,
      "loss": 0.5433,
      "step": 1400
    },
    {
      "epoch": 2.61,
      "grad_norm": 1.551100730895996,
      "learning_rate": 0.0036953703703703706,
      "loss": 0.5827,
      "step": 1410
    },
    {
      "epoch": 2.63,
      "grad_norm": 1.2063732147216797,
      "learning_rate": 0.0036861111111111114,
      "loss": 0.5331,
      "step": 1420
    },
    {
      "epoch": 2.65,
      "grad_norm": 1.514872431755066,
      "learning_rate": 0.0036768518518518517,
      "loss": 0.548,
      "step": 1430
    },
    {
      "epoch": 2.66,
      "grad_norm": 1.643267035484314,
      "learning_rate": 0.003667592592592593,
      "loss": 0.598,
      "step": 1440
    },
    {
      "epoch": 2.68,
      "grad_norm": 1.5592392683029175,
      "learning_rate": 0.0036583333333333333,
      "loss": 0.4851,
      "step": 1450
    },
    {
      "epoch": 2.7,
      "grad_norm": 1.1929229497909546,
      "learning_rate": 0.003649074074074074,
      "loss": 0.5362,
      "step": 1460
    },
    {
      "epoch": 2.72,
      "grad_norm": 1.610229253768921,
      "learning_rate": 0.0036398148148148145,
      "loss": 0.5826,
      "step": 1470
    },
    {
      "epoch": 2.74,
      "grad_norm": 1.2494343519210815,
      "learning_rate": 0.0036305555555555557,
      "loss": 0.5206,
      "step": 1480
    },
    {
      "epoch": 2.76,
      "grad_norm": 1.1670695543289185,
      "learning_rate": 0.0036212962962962965,
      "loss": 0.569,
      "step": 1490
    },
    {
      "epoch": 2.78,
      "grad_norm": 1.3939517736434937,
      "learning_rate": 0.003612037037037037,
      "loss": 0.5018,
      "step": 1500
    },
    {
      "epoch": 2.79,
      "grad_norm": 1.3553017377853394,
      "learning_rate": 0.0036027777777777777,
      "loss": 0.5167,
      "step": 1510
    },
    {
      "epoch": 2.81,
      "grad_norm": 1.3546302318572998,
      "learning_rate": 0.003593518518518519,
      "loss": 0.4979,
      "step": 1520
    },
    {
      "epoch": 2.83,
      "grad_norm": 1.2396551370620728,
      "learning_rate": 0.0035842592592592593,
      "loss": 0.4837,
      "step": 1530
    },
    {
      "epoch": 2.85,
      "grad_norm": 1.103031873703003,
      "learning_rate": 0.003575,
      "loss": 0.6083,
      "step": 1540
    },
    {
      "epoch": 2.87,
      "grad_norm": 1.642699122428894,
      "learning_rate": 0.003565740740740741,
      "loss": 0.5832,
      "step": 1550
    },
    {
      "epoch": 2.89,
      "grad_norm": 1.550116777420044,
      "learning_rate": 0.0035564814814814816,
      "loss": 0.479,
      "step": 1560
    },
    {
      "epoch": 2.9,
      "grad_norm": 1.745978832244873,
      "learning_rate": 0.0035472222222222224,
      "loss": 0.5643,
      "step": 1570
    },
    {
      "epoch": 2.92,
      "grad_norm": 0.9975535273551941,
      "learning_rate": 0.003537962962962963,
      "loss": 0.5837,
      "step": 1580
    },
    {
      "epoch": 2.94,
      "grad_norm": 1.8515207767486572,
      "learning_rate": 0.003528703703703704,
      "loss": 0.5637,
      "step": 1590
    },
    {
      "epoch": 2.96,
      "grad_norm": 2.022219181060791,
      "learning_rate": 0.0035194444444444444,
      "loss": 0.5121,
      "step": 1600
    },
    {
      "epoch": 2.98,
      "grad_norm": 1.8385446071624756,
      "learning_rate": 0.003510185185185185,
      "loss": 0.5244,
      "step": 1610
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8141658306121826,
      "learning_rate": 0.0035009259259259256,
      "loss": 0.6059,
      "step": 1620
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9562471113849946,
      "eval_f1": 0.9592862858808702,
      "eval_loss": 0.1171223372220993,
      "eval_precision": 0.9658949301313363,
      "eval_recall": 0.9569053041890571,
      "eval_runtime": 53.6538,
      "eval_samples_per_second": 120.979,
      "eval_steps_per_second": 7.567,
      "step": 1621
    },
    {
      "epoch": 3.02,
      "grad_norm": 1.811785340309143,
      "learning_rate": 0.003491666666666667,
      "loss": 0.5229,
      "step": 1630
    },
    {
      "epoch": 3.03,
      "grad_norm": 0.8801301717758179,
      "learning_rate": 0.0034824074074074076,
      "loss": 0.5735,
      "step": 1640
    },
    {
      "epoch": 3.05,
      "grad_norm": 1.5100030899047852,
      "learning_rate": 0.003473148148148148,
      "loss": 0.5581,
      "step": 1650
    },
    {
      "epoch": 3.07,
      "grad_norm": 1.7307904958724976,
      "learning_rate": 0.003463888888888889,
      "loss": 0.4894,
      "step": 1660
    },
    {
      "epoch": 3.09,
      "grad_norm": 2.264219045639038,
      "learning_rate": 0.00345462962962963,
      "loss": 0.5689,
      "step": 1670
    },
    {
      "epoch": 3.11,
      "grad_norm": 1.2732802629470825,
      "learning_rate": 0.0034453703703703703,
      "loss": 0.5168,
      "step": 1680
    },
    {
      "epoch": 3.13,
      "grad_norm": 1.5013095140457153,
      "learning_rate": 0.003436111111111111,
      "loss": 0.4907,
      "step": 1690
    },
    {
      "epoch": 3.15,
      "grad_norm": 1.617488980293274,
      "learning_rate": 0.003426851851851852,
      "loss": 0.4795,
      "step": 1700
    },
    {
      "epoch": 3.16,
      "grad_norm": 1.5595793724060059,
      "learning_rate": 0.0034175925925925927,
      "loss": 0.5816,
      "step": 1710
    },
    {
      "epoch": 3.18,
      "grad_norm": 1.9643666744232178,
      "learning_rate": 0.003408333333333333,
      "loss": 0.5512,
      "step": 1720
    },
    {
      "epoch": 3.2,
      "grad_norm": 1.1487085819244385,
      "learning_rate": 0.003399074074074074,
      "loss": 0.5253,
      "step": 1730
    },
    {
      "epoch": 3.22,
      "grad_norm": 1.359946608543396,
      "learning_rate": 0.003389814814814815,
      "loss": 0.5056,
      "step": 1740
    },
    {
      "epoch": 3.24,
      "grad_norm": 1.0156477689743042,
      "learning_rate": 0.0033805555555555555,
      "loss": 0.5154,
      "step": 1750
    },
    {
      "epoch": 3.26,
      "grad_norm": 1.1285754442214966,
      "learning_rate": 0.0033712962962962963,
      "loss": 0.5344,
      "step": 1760
    },
    {
      "epoch": 3.27,
      "grad_norm": 1.5709244012832642,
      "learning_rate": 0.0033620370370370375,
      "loss": 0.5373,
      "step": 1770
    },
    {
      "epoch": 3.29,
      "grad_norm": 1.4874547719955444,
      "learning_rate": 0.003352777777777778,
      "loss": 0.4925,
      "step": 1780
    },
    {
      "epoch": 3.31,
      "grad_norm": 2.197540044784546,
      "learning_rate": 0.0033435185185185187,
      "loss": 0.4636,
      "step": 1790
    },
    {
      "epoch": 3.33,
      "grad_norm": 1.0891751050949097,
      "learning_rate": 0.003334259259259259,
      "loss": 0.5313,
      "step": 1800
    },
    {
      "epoch": 3.35,
      "grad_norm": 1.4213520288467407,
      "learning_rate": 0.0033250000000000003,
      "loss": 0.5016,
      "step": 1810
    },
    {
      "epoch": 3.37,
      "grad_norm": 1.420879602432251,
      "learning_rate": 0.0033157407407407406,
      "loss": 0.4622,
      "step": 1820
    },
    {
      "epoch": 3.39,
      "grad_norm": 0.9189732074737549,
      "learning_rate": 0.0033064814814814814,
      "loss": 0.5148,
      "step": 1830
    },
    {
      "epoch": 3.4,
      "grad_norm": 1.5656261444091797,
      "learning_rate": 0.0032972222222222227,
      "loss": 0.4722,
      "step": 1840
    },
    {
      "epoch": 3.42,
      "grad_norm": 1.112702488899231,
      "learning_rate": 0.003287962962962963,
      "loss": 0.4968,
      "step": 1850
    },
    {
      "epoch": 3.44,
      "grad_norm": 1.58233642578125,
      "learning_rate": 0.003278703703703704,
      "loss": 0.4366,
      "step": 1860
    },
    {
      "epoch": 3.46,
      "grad_norm": 1.3384062051773071,
      "learning_rate": 0.003269444444444444,
      "loss": 0.4679,
      "step": 1870
    },
    {
      "epoch": 3.48,
      "grad_norm": 1.0208711624145508,
      "learning_rate": 0.0032601851851851854,
      "loss": 0.5348,
      "step": 1880
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.9358342885971069,
      "learning_rate": 0.003250925925925926,
      "loss": 0.468,
      "step": 1890
    },
    {
      "epoch": 3.52,
      "grad_norm": 1.3694192171096802,
      "learning_rate": 0.0032416666666666666,
      "loss": 0.4936,
      "step": 1900
    },
    {
      "epoch": 3.53,
      "grad_norm": 1.218279242515564,
      "learning_rate": 0.0032324074074074074,
      "loss": 0.5086,
      "step": 1910
    },
    {
      "epoch": 3.55,
      "grad_norm": 1.1621546745300293,
      "learning_rate": 0.0032231481481481486,
      "loss": 0.4918,
      "step": 1920
    },
    {
      "epoch": 3.57,
      "grad_norm": 1.2146507501602173,
      "learning_rate": 0.003213888888888889,
      "loss": 0.488,
      "step": 1930
    },
    {
      "epoch": 3.59,
      "grad_norm": 1.1717473268508911,
      "learning_rate": 0.0032046296296296298,
      "loss": 0.5438,
      "step": 1940
    },
    {
      "epoch": 3.61,
      "grad_norm": 0.9260009527206421,
      "learning_rate": 0.0031953703703703706,
      "loss": 0.4683,
      "step": 1950
    },
    {
      "epoch": 3.63,
      "grad_norm": 1.7548483610153198,
      "learning_rate": 0.0031861111111111113,
      "loss": 0.5116,
      "step": 1960
    },
    {
      "epoch": 3.64,
      "grad_norm": 1.603769302368164,
      "learning_rate": 0.0031768518518518517,
      "loss": 0.506,
      "step": 1970
    },
    {
      "epoch": 3.66,
      "grad_norm": 1.4770747423171997,
      "learning_rate": 0.0031675925925925925,
      "loss": 0.4877,
      "step": 1980
    },
    {
      "epoch": 3.68,
      "grad_norm": 1.8213857412338257,
      "learning_rate": 0.0031583333333333337,
      "loss": 0.5239,
      "step": 1990
    },
    {
      "epoch": 3.7,
      "grad_norm": 1.074977993965149,
      "learning_rate": 0.003149074074074074,
      "loss": 0.4655,
      "step": 2000
    },
    {
      "epoch": 3.72,
      "grad_norm": 1.3410956859588623,
      "learning_rate": 0.003139814814814815,
      "loss": 0.4097,
      "step": 2010
    },
    {
      "epoch": 3.74,
      "grad_norm": 0.6797301173210144,
      "learning_rate": 0.0031305555555555553,
      "loss": 0.5193,
      "step": 2020
    },
    {
      "epoch": 3.76,
      "grad_norm": 1.1780236959457397,
      "learning_rate": 0.0031212962962962965,
      "loss": 0.438,
      "step": 2030
    },
    {
      "epoch": 3.77,
      "grad_norm": 1.4561818838119507,
      "learning_rate": 0.0031120370370370373,
      "loss": 0.5156,
      "step": 2040
    },
    {
      "epoch": 3.79,
      "grad_norm": 1.404532790184021,
      "learning_rate": 0.0031027777777777777,
      "loss": 0.4319,
      "step": 2050
    },
    {
      "epoch": 3.81,
      "grad_norm": 1.453012466430664,
      "learning_rate": 0.003093518518518519,
      "loss": 0.4678,
      "step": 2060
    },
    {
      "epoch": 3.83,
      "grad_norm": 1.7319668531417847,
      "learning_rate": 0.0030842592592592592,
      "loss": 0.4567,
      "step": 2070
    },
    {
      "epoch": 3.85,
      "grad_norm": 1.718991994857788,
      "learning_rate": 0.003075,
      "loss": 0.5631,
      "step": 2080
    },
    {
      "epoch": 3.87,
      "grad_norm": 1.4493403434753418,
      "learning_rate": 0.0030657407407407404,
      "loss": 0.5046,
      "step": 2090
    },
    {
      "epoch": 3.89,
      "grad_norm": 1.5228378772735596,
      "learning_rate": 0.0030564814814814816,
      "loss": 0.4583,
      "step": 2100
    },
    {
      "epoch": 3.9,
      "grad_norm": 1.3635989427566528,
      "learning_rate": 0.0030472222222222224,
      "loss": 0.4678,
      "step": 2110
    },
    {
      "epoch": 3.92,
      "grad_norm": 1.7252169847488403,
      "learning_rate": 0.003037962962962963,
      "loss": 0.4817,
      "step": 2120
    },
    {
      "epoch": 3.94,
      "grad_norm": 1.4143972396850586,
      "learning_rate": 0.0030287037037037036,
      "loss": 0.4663,
      "step": 2130
    },
    {
      "epoch": 3.96,
      "grad_norm": 1.108224868774414,
      "learning_rate": 0.003019444444444445,
      "loss": 0.5028,
      "step": 2140
    },
    {
      "epoch": 3.98,
      "grad_norm": 1.5944609642028809,
      "learning_rate": 0.003010185185185185,
      "loss": 0.5215,
      "step": 2150
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9836990833282471,
      "learning_rate": 0.003000925925925926,
      "loss": 0.3547,
      "step": 2160
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.966569095670929,
      "eval_f1": 0.9701817800594627,
      "eval_loss": 0.09807376563549042,
      "eval_precision": 0.970888445046177,
      "eval_recall": 0.9711737953539892,
      "eval_runtime": 53.6052,
      "eval_samples_per_second": 121.089,
      "eval_steps_per_second": 7.574,
      "step": 2162
    },
    {
      "epoch": 4.01,
      "grad_norm": 1.2644659280776978,
      "learning_rate": 0.0029916666666666668,
      "loss": 0.5066,
      "step": 2170
    },
    {
      "epoch": 4.03,
      "grad_norm": 1.5990442037582397,
      "learning_rate": 0.0029824074074074076,
      "loss": 0.5394,
      "step": 2180
    },
    {
      "epoch": 4.05,
      "grad_norm": 1.1844220161437988,
      "learning_rate": 0.0029731481481481484,
      "loss": 0.5361,
      "step": 2190
    },
    {
      "epoch": 4.07,
      "grad_norm": 1.003459095954895,
      "learning_rate": 0.0029638888888888887,
      "loss": 0.4965,
      "step": 2200
    },
    {
      "epoch": 4.09,
      "grad_norm": 1.5128463506698608,
      "learning_rate": 0.00295462962962963,
      "loss": 0.4961,
      "step": 2210
    },
    {
      "epoch": 4.11,
      "grad_norm": 1.4107407331466675,
      "learning_rate": 0.0029453703703703703,
      "loss": 0.4157,
      "step": 2220
    },
    {
      "epoch": 4.13,
      "grad_norm": 1.0188244581222534,
      "learning_rate": 0.002936111111111111,
      "loss": 0.4571,
      "step": 2230
    },
    {
      "epoch": 4.14,
      "grad_norm": 1.7941489219665527,
      "learning_rate": 0.0029268518518518515,
      "loss": 0.4037,
      "step": 2240
    },
    {
      "epoch": 4.16,
      "grad_norm": 0.703614354133606,
      "learning_rate": 0.0029175925925925927,
      "loss": 0.469,
      "step": 2250
    },
    {
      "epoch": 4.18,
      "grad_norm": 1.0244730710983276,
      "learning_rate": 0.0029083333333333335,
      "loss": 0.446,
      "step": 2260
    },
    {
      "epoch": 4.2,
      "grad_norm": 1.940670132637024,
      "learning_rate": 0.002899074074074074,
      "loss": 0.4198,
      "step": 2270
    },
    {
      "epoch": 4.22,
      "grad_norm": 1.3189454078674316,
      "learning_rate": 0.002889814814814815,
      "loss": 0.4627,
      "step": 2280
    },
    {
      "epoch": 4.24,
      "grad_norm": 1.0493321418762207,
      "learning_rate": 0.002880555555555556,
      "loss": 0.4885,
      "step": 2290
    },
    {
      "epoch": 4.26,
      "grad_norm": 1.0230928659439087,
      "learning_rate": 0.0028712962962962963,
      "loss": 0.4205,
      "step": 2300
    },
    {
      "epoch": 4.27,
      "grad_norm": 0.7459888458251953,
      "learning_rate": 0.002862037037037037,
      "loss": 0.4316,
      "step": 2310
    },
    {
      "epoch": 4.29,
      "grad_norm": 1.2074246406555176,
      "learning_rate": 0.002852777777777778,
      "loss": 0.4352,
      "step": 2320
    },
    {
      "epoch": 4.31,
      "grad_norm": 1.7306902408599854,
      "learning_rate": 0.0028435185185185187,
      "loss": 0.391,
      "step": 2330
    },
    {
      "epoch": 4.33,
      "grad_norm": 1.4361604452133179,
      "learning_rate": 0.002834259259259259,
      "loss": 0.4819,
      "step": 2340
    },
    {
      "epoch": 4.35,
      "grad_norm": 1.3152241706848145,
      "learning_rate": 0.002825,
      "loss": 0.3861,
      "step": 2350
    },
    {
      "epoch": 4.37,
      "grad_norm": 1.6985182762145996,
      "learning_rate": 0.002815740740740741,
      "loss": 0.4093,
      "step": 2360
    },
    {
      "epoch": 4.38,
      "grad_norm": 1.1082671880722046,
      "learning_rate": 0.0028064814814814814,
      "loss": 0.5325,
      "step": 2370
    },
    {
      "epoch": 4.4,
      "grad_norm": 1.39676833152771,
      "learning_rate": 0.002797222222222222,
      "loss": 0.4798,
      "step": 2380
    },
    {
      "epoch": 4.42,
      "grad_norm": 1.0717923641204834,
      "learning_rate": 0.0027879629629629634,
      "loss": 0.3892,
      "step": 2390
    },
    {
      "epoch": 4.44,
      "grad_norm": 0.82547926902771,
      "learning_rate": 0.002778703703703704,
      "loss": 0.4563,
      "step": 2400
    },
    {
      "epoch": 4.46,
      "grad_norm": 1.1817727088928223,
      "learning_rate": 0.0027694444444444446,
      "loss": 0.4213,
      "step": 2410
    },
    {
      "epoch": 4.48,
      "grad_norm": 1.127348780632019,
      "learning_rate": 0.002760185185185185,
      "loss": 0.4636,
      "step": 2420
    },
    {
      "epoch": 4.5,
      "grad_norm": 1.2709828615188599,
      "learning_rate": 0.002750925925925926,
      "loss": 0.5099,
      "step": 2430
    },
    {
      "epoch": 4.51,
      "grad_norm": 1.093279242515564,
      "learning_rate": 0.0027416666666666666,
      "loss": 0.439,
      "step": 2440
    },
    {
      "epoch": 4.53,
      "grad_norm": 0.8145557045936584,
      "learning_rate": 0.0027324074074074074,
      "loss": 0.4192,
      "step": 2450
    },
    {
      "epoch": 4.55,
      "grad_norm": 1.5339893102645874,
      "learning_rate": 0.0027231481481481477,
      "loss": 0.4591,
      "step": 2460
    },
    {
      "epoch": 4.57,
      "grad_norm": 1.2502670288085938,
      "learning_rate": 0.002713888888888889,
      "loss": 0.4177,
      "step": 2470
    },
    {
      "epoch": 4.59,
      "grad_norm": 0.951678991317749,
      "learning_rate": 0.0027046296296296297,
      "loss": 0.4206,
      "step": 2480
    },
    {
      "epoch": 4.61,
      "grad_norm": 1.7269920110702515,
      "learning_rate": 0.00269537037037037,
      "loss": 0.4318,
      "step": 2490
    },
    {
      "epoch": 4.63,
      "grad_norm": 1.3369319438934326,
      "learning_rate": 0.0026861111111111113,
      "loss": 0.3328,
      "step": 2500
    },
    {
      "epoch": 4.64,
      "grad_norm": 1.1424506902694702,
      "learning_rate": 0.002676851851851852,
      "loss": 0.4531,
      "step": 2510
    },
    {
      "epoch": 4.66,
      "grad_norm": 0.7604424953460693,
      "learning_rate": 0.0026675925925925925,
      "loss": 0.4192,
      "step": 2520
    },
    {
      "epoch": 4.68,
      "grad_norm": 1.501299500465393,
      "learning_rate": 0.0026583333333333333,
      "loss": 0.4588,
      "step": 2530
    },
    {
      "epoch": 4.7,
      "grad_norm": 0.7411832213401794,
      "learning_rate": 0.002649074074074074,
      "loss": 0.507,
      "step": 2540
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.9906969666481018,
      "learning_rate": 0.002639814814814815,
      "loss": 0.4733,
      "step": 2550
    },
    {
      "epoch": 4.74,
      "grad_norm": 1.5065710544586182,
      "learning_rate": 0.0026305555555555557,
      "loss": 0.4134,
      "step": 2560
    },
    {
      "epoch": 4.75,
      "grad_norm": 1.1822443008422852,
      "learning_rate": 0.0026212962962962965,
      "loss": 0.4283,
      "step": 2570
    },
    {
      "epoch": 4.77,
      "grad_norm": 1.1529394388198853,
      "learning_rate": 0.0026120370370370373,
      "loss": 0.4447,
      "step": 2580
    },
    {
      "epoch": 4.79,
      "grad_norm": 1.294115424156189,
      "learning_rate": 0.0026027777777777776,
      "loss": 0.4889,
      "step": 2590
    },
    {
      "epoch": 4.81,
      "grad_norm": 0.7541661858558655,
      "learning_rate": 0.0025935185185185184,
      "loss": 0.4539,
      "step": 2600
    },
    {
      "epoch": 4.83,
      "grad_norm": 1.0515060424804688,
      "learning_rate": 0.0025842592592592597,
      "loss": 0.4458,
      "step": 2610
    },
    {
      "epoch": 4.85,
      "grad_norm": 1.2407331466674805,
      "learning_rate": 0.002575,
      "loss": 0.4479,
      "step": 2620
    },
    {
      "epoch": 4.87,
      "grad_norm": 1.4178121089935303,
      "learning_rate": 0.002565740740740741,
      "loss": 0.451,
      "step": 2630
    },
    {
      "epoch": 4.88,
      "grad_norm": 0.9205562472343445,
      "learning_rate": 0.002556481481481481,
      "loss": 0.4179,
      "step": 2640
    },
    {
      "epoch": 4.9,
      "grad_norm": 1.106848120689392,
      "learning_rate": 0.0025472222222222224,
      "loss": 0.4341,
      "step": 2650
    },
    {
      "epoch": 4.92,
      "grad_norm": 0.6212519407272339,
      "learning_rate": 0.0025379629629629632,
      "loss": 0.3167,
      "step": 2660
    },
    {
      "epoch": 4.94,
      "grad_norm": 2.72360897064209,
      "learning_rate": 0.0025287037037037036,
      "loss": 0.4454,
      "step": 2670
    },
    {
      "epoch": 4.96,
      "grad_norm": 1.535974144935608,
      "learning_rate": 0.002519444444444445,
      "loss": 0.4625,
      "step": 2680
    },
    {
      "epoch": 4.98,
      "grad_norm": 0.6457435488700867,
      "learning_rate": 0.002510185185185185,
      "loss": 0.3886,
      "step": 2690
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9673438668251038,
      "learning_rate": 0.002500925925925926,
      "loss": 0.4852,
      "step": 2700
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9816669234324449,
      "eval_f1": 0.9842159244197872,
      "eval_loss": 0.0539265014231205,
      "eval_precision": 0.9847615394086703,
      "eval_recall": 0.9841734919436653,
      "eval_runtime": 53.4958,
      "eval_samples_per_second": 121.337,
      "eval_steps_per_second": 7.589,
      "step": 2702
    },
    {
      "epoch": 5.01,
      "grad_norm": 0.9143092036247253,
      "learning_rate": 0.0024916666666666668,
      "loss": 0.4017,
      "step": 2710
    },
    {
      "epoch": 5.03,
      "grad_norm": 1.2292033433914185,
      "learning_rate": 0.0024824074074074076,
      "loss": 0.3677,
      "step": 2720
    },
    {
      "epoch": 5.05,
      "grad_norm": 1.2112282514572144,
      "learning_rate": 0.0024731481481481484,
      "loss": 0.5256,
      "step": 2730
    },
    {
      "epoch": 5.07,
      "grad_norm": 1.3062794208526611,
      "learning_rate": 0.0024638888888888887,
      "loss": 0.45,
      "step": 2740
    },
    {
      "epoch": 5.09,
      "grad_norm": 1.280093789100647,
      "learning_rate": 0.0024546296296296295,
      "loss": 0.4144,
      "step": 2750
    },
    {
      "epoch": 5.11,
      "grad_norm": 0.9576690196990967,
      "learning_rate": 0.0024453703703703703,
      "loss": 0.4183,
      "step": 2760
    },
    {
      "epoch": 5.12,
      "grad_norm": 1.3858096599578857,
      "learning_rate": 0.002436111111111111,
      "loss": 0.4774,
      "step": 2770
    },
    {
      "epoch": 5.14,
      "grad_norm": 0.9120557904243469,
      "learning_rate": 0.002426851851851852,
      "loss": 0.4282,
      "step": 2780
    },
    {
      "epoch": 5.16,
      "grad_norm": 0.9812218546867371,
      "learning_rate": 0.0024175925925925927,
      "loss": 0.4386,
      "step": 2790
    },
    {
      "epoch": 5.18,
      "grad_norm": 0.8847913146018982,
      "learning_rate": 0.0024083333333333335,
      "loss": 0.3769,
      "step": 2800
    },
    {
      "epoch": 5.2,
      "grad_norm": 0.6346982717514038,
      "learning_rate": 0.002399074074074074,
      "loss": 0.404,
      "step": 2810
    },
    {
      "epoch": 5.22,
      "grad_norm": 1.937452793121338,
      "learning_rate": 0.002389814814814815,
      "loss": 0.4098,
      "step": 2820
    },
    {
      "epoch": 5.24,
      "grad_norm": 0.6616402864456177,
      "learning_rate": 0.0023805555555555555,
      "loss": 0.4038,
      "step": 2830
    },
    {
      "epoch": 5.25,
      "grad_norm": 1.248590111732483,
      "learning_rate": 0.0023712962962962963,
      "loss": 0.4208,
      "step": 2840
    },
    {
      "epoch": 5.27,
      "grad_norm": 1.5833156108856201,
      "learning_rate": 0.002362037037037037,
      "loss": 0.3919,
      "step": 2850
    },
    {
      "epoch": 5.29,
      "grad_norm": 1.1102122068405151,
      "learning_rate": 0.002352777777777778,
      "loss": 0.4135,
      "step": 2860
    },
    {
      "epoch": 5.31,
      "grad_norm": 1.2193430662155151,
      "learning_rate": 0.0023435185185185187,
      "loss": 0.4343,
      "step": 2870
    },
    {
      "epoch": 5.33,
      "grad_norm": 1.5274019241333008,
      "learning_rate": 0.0023342592592592594,
      "loss": 0.397,
      "step": 2880
    },
    {
      "epoch": 5.35,
      "grad_norm": 1.0362614393234253,
      "learning_rate": 0.0023250000000000002,
      "loss": 0.3789,
      "step": 2890
    },
    {
      "epoch": 5.37,
      "grad_norm": 1.0950071811676025,
      "learning_rate": 0.0023157407407407406,
      "loss": 0.4146,
      "step": 2900
    },
    {
      "epoch": 5.38,
      "grad_norm": 1.6324676275253296,
      "learning_rate": 0.002306481481481482,
      "loss": 0.4602,
      "step": 2910
    },
    {
      "epoch": 5.4,
      "grad_norm": 1.0304063558578491,
      "learning_rate": 0.002297222222222222,
      "loss": 0.3736,
      "step": 2920
    },
    {
      "epoch": 5.42,
      "grad_norm": 0.8070071339607239,
      "learning_rate": 0.002287962962962963,
      "loss": 0.4444,
      "step": 2930
    },
    {
      "epoch": 5.44,
      "grad_norm": 1.5730412006378174,
      "learning_rate": 0.002278703703703704,
      "loss": 0.4247,
      "step": 2940
    },
    {
      "epoch": 5.46,
      "grad_norm": 0.8822392821311951,
      "learning_rate": 0.0022694444444444446,
      "loss": 0.3523,
      "step": 2950
    },
    {
      "epoch": 5.48,
      "grad_norm": 1.6417633295059204,
      "learning_rate": 0.002260185185185185,
      "loss": 0.428,
      "step": 2960
    },
    {
      "epoch": 5.49,
      "grad_norm": 1.0473852157592773,
      "learning_rate": 0.002250925925925926,
      "loss": 0.3916,
      "step": 2970
    },
    {
      "epoch": 5.51,
      "grad_norm": 1.1455719470977783,
      "learning_rate": 0.0022416666666666665,
      "loss": 0.3337,
      "step": 2980
    },
    {
      "epoch": 5.53,
      "grad_norm": 0.9637939929962158,
      "learning_rate": 0.0022324074074074073,
      "loss": 0.4168,
      "step": 2990
    },
    {
      "epoch": 5.55,
      "grad_norm": 1.650855541229248,
      "learning_rate": 0.002223148148148148,
      "loss": 0.3526,
      "step": 3000
    },
    {
      "epoch": 5.57,
      "grad_norm": 0.8603165745735168,
      "learning_rate": 0.002213888888888889,
      "loss": 0.3343,
      "step": 3010
    },
    {
      "epoch": 5.59,
      "grad_norm": 0.9890497922897339,
      "learning_rate": 0.0022046296296296297,
      "loss": 0.3577,
      "step": 3020
    },
    {
      "epoch": 5.61,
      "grad_norm": 1.0613206624984741,
      "learning_rate": 0.0021953703703703705,
      "loss": 0.4288,
      "step": 3030
    },
    {
      "epoch": 5.62,
      "grad_norm": 0.8673573136329651,
      "learning_rate": 0.0021861111111111113,
      "loss": 0.3787,
      "step": 3040
    },
    {
      "epoch": 5.64,
      "grad_norm": 0.9730037450790405,
      "learning_rate": 0.0021768518518518517,
      "loss": 0.3901,
      "step": 3050
    },
    {
      "epoch": 5.66,
      "grad_norm": 0.9955042004585266,
      "learning_rate": 0.0021675925925925925,
      "loss": 0.3267,
      "step": 3060
    },
    {
      "epoch": 5.68,
      "grad_norm": 1.3361483812332153,
      "learning_rate": 0.0021583333333333333,
      "loss": 0.3821,
      "step": 3070
    },
    {
      "epoch": 5.7,
      "grad_norm": 1.1696761846542358,
      "learning_rate": 0.002149074074074074,
      "loss": 0.4164,
      "step": 3080
    },
    {
      "epoch": 5.72,
      "grad_norm": 0.9910469055175781,
      "learning_rate": 0.002139814814814815,
      "loss": 0.3488,
      "step": 3090
    },
    {
      "epoch": 5.74,
      "grad_norm": 1.1704002618789673,
      "learning_rate": 0.0021305555555555557,
      "loss": 0.3773,
      "step": 3100
    },
    {
      "epoch": 5.75,
      "grad_norm": 1.9228627681732178,
      "learning_rate": 0.0021212962962962965,
      "loss": 0.3609,
      "step": 3110
    },
    {
      "epoch": 5.77,
      "grad_norm": 1.2759217023849487,
      "learning_rate": 0.002112037037037037,
      "loss": 0.339,
      "step": 3120
    },
    {
      "epoch": 5.79,
      "grad_norm": 1.3884625434875488,
      "learning_rate": 0.002102777777777778,
      "loss": 0.3427,
      "step": 3130
    },
    {
      "epoch": 5.81,
      "grad_norm": 1.0535517930984497,
      "learning_rate": 0.0020935185185185184,
      "loss": 0.3684,
      "step": 3140
    },
    {
      "epoch": 5.83,
      "grad_norm": 0.7472217082977295,
      "learning_rate": 0.0020842592592592592,
      "loss": 0.3315,
      "step": 3150
    },
    {
      "epoch": 5.85,
      "grad_norm": 0.9054597020149231,
      "learning_rate": 0.002075,
      "loss": 0.4055,
      "step": 3160
    },
    {
      "epoch": 5.86,
      "grad_norm": 0.9936026930809021,
      "learning_rate": 0.002065740740740741,
      "loss": 0.3684,
      "step": 3170
    },
    {
      "epoch": 5.88,
      "grad_norm": 0.924675464630127,
      "learning_rate": 0.002056481481481481,
      "loss": 0.4467,
      "step": 3180
    },
    {
      "epoch": 5.9,
      "grad_norm": 0.9270643591880798,
      "learning_rate": 0.0020472222222222224,
      "loss": 0.4563,
      "step": 3190
    },
    {
      "epoch": 5.92,
      "grad_norm": 1.1928519010543823,
      "learning_rate": 0.002037962962962963,
      "loss": 0.436,
      "step": 3200
    },
    {
      "epoch": 5.94,
      "grad_norm": 0.7696529030799866,
      "learning_rate": 0.0020287037037037036,
      "loss": 0.3837,
      "step": 3210
    },
    {
      "epoch": 5.96,
      "grad_norm": 0.787181556224823,
      "learning_rate": 0.0020194444444444444,
      "loss": 0.3855,
      "step": 3220
    },
    {
      "epoch": 5.98,
      "grad_norm": 0.9738120436668396,
      "learning_rate": 0.002010185185185185,
      "loss": 0.4042,
      "step": 3230
    },
    {
      "epoch": 5.99,
      "grad_norm": 0.7975121140480042,
      "learning_rate": 0.002000925925925926,
      "loss": 0.406,
      "step": 3240
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9748883068864582,
      "eval_f1": 0.9767778940213332,
      "eval_loss": 0.08178560435771942,
      "eval_precision": 0.9793037526964313,
      "eval_recall": 0.9752347424170356,
      "eval_runtime": 52.9882,
      "eval_samples_per_second": 122.499,
      "eval_steps_per_second": 7.662,
      "step": 3243
    },
    {
      "epoch": 6.01,
      "grad_norm": 1.21721351146698,
      "learning_rate": 0.0019916666666666668,
      "loss": 0.3252,
      "step": 3250
    },
    {
      "epoch": 6.03,
      "grad_norm": 1.1153512001037598,
      "learning_rate": 0.0019824074074074076,
      "loss": 0.364,
      "step": 3260
    },
    {
      "epoch": 6.05,
      "grad_norm": 0.9769272208213806,
      "learning_rate": 0.001973148148148148,
      "loss": 0.3886,
      "step": 3270
    },
    {
      "epoch": 6.07,
      "grad_norm": 1.260012149810791,
      "learning_rate": 0.001963888888888889,
      "loss": 0.3462,
      "step": 3280
    },
    {
      "epoch": 6.09,
      "grad_norm": 1.2162351608276367,
      "learning_rate": 0.0019546296296296295,
      "loss": 0.3472,
      "step": 3290
    },
    {
      "epoch": 6.11,
      "grad_norm": 1.2838939428329468,
      "learning_rate": 0.0019453703703703703,
      "loss": 0.3427,
      "step": 3300
    },
    {
      "epoch": 6.12,
      "grad_norm": 0.850970447063446,
      "learning_rate": 0.0019361111111111113,
      "loss": 0.3793,
      "step": 3310
    },
    {
      "epoch": 6.14,
      "grad_norm": 0.7692118883132935,
      "learning_rate": 0.001926851851851852,
      "loss": 0.3029,
      "step": 3320
    },
    {
      "epoch": 6.16,
      "grad_norm": 1.337117314338684,
      "learning_rate": 0.0019175925925925927,
      "loss": 0.3526,
      "step": 3330
    },
    {
      "epoch": 6.18,
      "grad_norm": 1.0097956657409668,
      "learning_rate": 0.0019083333333333333,
      "loss": 0.3544,
      "step": 3340
    },
    {
      "epoch": 6.2,
      "grad_norm": 1.1597776412963867,
      "learning_rate": 0.001899074074074074,
      "loss": 0.3266,
      "step": 3350
    },
    {
      "epoch": 6.22,
      "grad_norm": 0.8050183653831482,
      "learning_rate": 0.0018898148148148149,
      "loss": 0.352,
      "step": 3360
    },
    {
      "epoch": 6.23,
      "grad_norm": 1.197143316268921,
      "learning_rate": 0.0018805555555555557,
      "loss": 0.3538,
      "step": 3370
    },
    {
      "epoch": 6.25,
      "grad_norm": 1.5818617343902588,
      "learning_rate": 0.0018712962962962962,
      "loss": 0.3441,
      "step": 3380
    },
    {
      "epoch": 6.27,
      "grad_norm": 0.6569982767105103,
      "learning_rate": 0.001862037037037037,
      "loss": 0.3436,
      "step": 3390
    },
    {
      "epoch": 6.29,
      "grad_norm": 0.7630707025527954,
      "learning_rate": 0.0018527777777777778,
      "loss": 0.3789,
      "step": 3400
    },
    {
      "epoch": 6.31,
      "grad_norm": 0.8499374389648438,
      "learning_rate": 0.0018435185185185186,
      "loss": 0.3351,
      "step": 3410
    },
    {
      "epoch": 6.33,
      "grad_norm": 0.8909932374954224,
      "learning_rate": 0.0018342592592592594,
      "loss": 0.2914,
      "step": 3420
    },
    {
      "epoch": 6.35,
      "grad_norm": 0.9815020561218262,
      "learning_rate": 0.001825,
      "loss": 0.3234,
      "step": 3430
    },
    {
      "epoch": 6.36,
      "grad_norm": 0.9451941251754761,
      "learning_rate": 0.0018157407407407408,
      "loss": 0.3016,
      "step": 3440
    },
    {
      "epoch": 6.38,
      "grad_norm": 1.1062078475952148,
      "learning_rate": 0.0018064814814814814,
      "loss": 0.3243,
      "step": 3450
    },
    {
      "epoch": 6.4,
      "grad_norm": 1.119948387145996,
      "learning_rate": 0.0017972222222222224,
      "loss": 0.4004,
      "step": 3460
    },
    {
      "epoch": 6.42,
      "grad_norm": 1.0116353034973145,
      "learning_rate": 0.001787962962962963,
      "loss": 0.3816,
      "step": 3470
    },
    {
      "epoch": 6.44,
      "grad_norm": 1.270749568939209,
      "learning_rate": 0.0017787037037037038,
      "loss": 0.3961,
      "step": 3480
    },
    {
      "epoch": 6.46,
      "grad_norm": 0.7347724437713623,
      "learning_rate": 0.0017694444444444444,
      "loss": 0.3183,
      "step": 3490
    },
    {
      "epoch": 6.48,
      "grad_norm": 1.3908417224884033,
      "learning_rate": 0.0017601851851851852,
      "loss": 0.3717,
      "step": 3500
    },
    {
      "epoch": 6.49,
      "grad_norm": 1.0049611330032349,
      "learning_rate": 0.0017509259259259262,
      "loss": 0.3804,
      "step": 3510
    },
    {
      "epoch": 6.51,
      "grad_norm": 0.8181151747703552,
      "learning_rate": 0.0017416666666666668,
      "loss": 0.3683,
      "step": 3520
    },
    {
      "epoch": 6.53,
      "grad_norm": 1.0311352014541626,
      "learning_rate": 0.0017324074074074076,
      "loss": 0.3813,
      "step": 3530
    },
    {
      "epoch": 6.55,
      "grad_norm": 0.6164132356643677,
      "learning_rate": 0.0017231481481481481,
      "loss": 0.342,
      "step": 3540
    },
    {
      "epoch": 6.57,
      "grad_norm": 0.7826759219169617,
      "learning_rate": 0.001713888888888889,
      "loss": 0.2858,
      "step": 3550
    },
    {
      "epoch": 6.59,
      "grad_norm": 0.6946287155151367,
      "learning_rate": 0.0017046296296296295,
      "loss": 0.3245,
      "step": 3560
    },
    {
      "epoch": 6.6,
      "grad_norm": 1.3883477449417114,
      "learning_rate": 0.0016953703703703705,
      "loss": 0.3492,
      "step": 3570
    },
    {
      "epoch": 6.62,
      "grad_norm": 1.0445375442504883,
      "learning_rate": 0.001686111111111111,
      "loss": 0.2873,
      "step": 3580
    },
    {
      "epoch": 6.64,
      "grad_norm": 1.155643105506897,
      "learning_rate": 0.001676851851851852,
      "loss": 0.3771,
      "step": 3590
    },
    {
      "epoch": 6.66,
      "grad_norm": 0.8850951790809631,
      "learning_rate": 0.0016675925925925925,
      "loss": 0.3412,
      "step": 3600
    },
    {
      "epoch": 6.68,
      "grad_norm": 0.8685129880905151,
      "learning_rate": 0.0016583333333333333,
      "loss": 0.3489,
      "step": 3610
    },
    {
      "epoch": 6.7,
      "grad_norm": 1.20561683177948,
      "learning_rate": 0.0016490740740740743,
      "loss": 0.3253,
      "step": 3620
    },
    {
      "epoch": 6.72,
      "grad_norm": 0.8918961882591248,
      "learning_rate": 0.0016398148148148149,
      "loss": 0.3279,
      "step": 3630
    },
    {
      "epoch": 6.73,
      "grad_norm": 1.1090009212493896,
      "learning_rate": 0.0016305555555555557,
      "loss": 0.2887,
      "step": 3640
    },
    {
      "epoch": 6.75,
      "grad_norm": 1.1889126300811768,
      "learning_rate": 0.0016212962962962962,
      "loss": 0.3391,
      "step": 3650
    },
    {
      "epoch": 6.77,
      "grad_norm": 0.7774761915206909,
      "learning_rate": 0.001612037037037037,
      "loss": 0.354,
      "step": 3660
    },
    {
      "epoch": 6.79,
      "grad_norm": 0.9942853450775146,
      "learning_rate": 0.0016027777777777776,
      "loss": 0.4198,
      "step": 3670
    },
    {
      "epoch": 6.81,
      "grad_norm": 0.5383427143096924,
      "learning_rate": 0.0015935185185185186,
      "loss": 0.2784,
      "step": 3680
    },
    {
      "epoch": 6.83,
      "grad_norm": 1.0659656524658203,
      "learning_rate": 0.0015842592592592592,
      "loss": 0.3369,
      "step": 3690
    },
    {
      "epoch": 6.85,
      "grad_norm": 1.0167818069458008,
      "learning_rate": 0.001575,
      "loss": 0.3465,
      "step": 3700
    },
    {
      "epoch": 6.86,
      "grad_norm": 0.9056137204170227,
      "learning_rate": 0.0015657407407407408,
      "loss": 0.3111,
      "step": 3710
    },
    {
      "epoch": 6.88,
      "grad_norm": 0.9691112041473389,
      "learning_rate": 0.0015564814814814816,
      "loss": 0.2901,
      "step": 3720
    },
    {
      "epoch": 6.9,
      "grad_norm": 1.0727977752685547,
      "learning_rate": 0.0015472222222222224,
      "loss": 0.3614,
      "step": 3730
    },
    {
      "epoch": 6.92,
      "grad_norm": 0.8301049470901489,
      "learning_rate": 0.001537962962962963,
      "loss": 0.3251,
      "step": 3740
    },
    {
      "epoch": 6.94,
      "grad_norm": 0.6486156582832336,
      "learning_rate": 0.0015287037037037038,
      "loss": 0.3135,
      "step": 3750
    },
    {
      "epoch": 6.96,
      "grad_norm": 1.2031179666519165,
      "learning_rate": 0.0015194444444444444,
      "loss": 0.3371,
      "step": 3760
    },
    {
      "epoch": 6.98,
      "grad_norm": 1.1333447694778442,
      "learning_rate": 0.0015101851851851854,
      "loss": 0.3242,
      "step": 3770
    },
    {
      "epoch": 6.99,
      "grad_norm": 1.0811736583709717,
      "learning_rate": 0.001500925925925926,
      "loss": 0.3074,
      "step": 3780
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.966569095670929,
      "eval_f1": 0.9782929900155292,
      "eval_loss": 0.1289132982492447,
      "eval_precision": 0.9815165094060785,
      "eval_recall": 0.977802007605107,
      "eval_runtime": 52.9121,
      "eval_samples_per_second": 122.675,
      "eval_steps_per_second": 7.673,
      "step": 3783
    },
    {
      "epoch": 7.01,
      "grad_norm": 0.5184229016304016,
      "learning_rate": 0.0014916666666666667,
      "loss": 0.3532,
      "step": 3790
    },
    {
      "epoch": 7.03,
      "grad_norm": 0.9822409152984619,
      "learning_rate": 0.0014824074074074073,
      "loss": 0.3701,
      "step": 3800
    },
    {
      "epoch": 7.05,
      "grad_norm": 0.6723494529724121,
      "learning_rate": 0.0014731481481481481,
      "loss": 0.3553,
      "step": 3810
    },
    {
      "epoch": 7.07,
      "grad_norm": 1.0363885164260864,
      "learning_rate": 0.0014638888888888891,
      "loss": 0.2526,
      "step": 3820
    },
    {
      "epoch": 7.09,
      "grad_norm": 0.9155850410461426,
      "learning_rate": 0.0014546296296296297,
      "loss": 0.3045,
      "step": 3830
    },
    {
      "epoch": 7.1,
      "grad_norm": 1.412392497062683,
      "learning_rate": 0.0014453703703703705,
      "loss": 0.2868,
      "step": 3840
    },
    {
      "epoch": 7.12,
      "grad_norm": 0.9707776308059692,
      "learning_rate": 0.001436111111111111,
      "loss": 0.3537,
      "step": 3850
    },
    {
      "epoch": 7.14,
      "grad_norm": 1.2895119190216064,
      "learning_rate": 0.001426851851851852,
      "loss": 0.2961,
      "step": 3860
    },
    {
      "epoch": 7.16,
      "grad_norm": 0.8239867687225342,
      "learning_rate": 0.0014175925925925925,
      "loss": 0.2836,
      "step": 3870
    },
    {
      "epoch": 7.18,
      "grad_norm": 1.1721490621566772,
      "learning_rate": 0.0014083333333333335,
      "loss": 0.3331,
      "step": 3880
    },
    {
      "epoch": 7.2,
      "grad_norm": 0.6656754612922668,
      "learning_rate": 0.001399074074074074,
      "loss": 0.3221,
      "step": 3890
    },
    {
      "epoch": 7.22,
      "grad_norm": 0.8847745656967163,
      "learning_rate": 0.0013898148148148149,
      "loss": 0.3361,
      "step": 3900
    },
    {
      "epoch": 7.23,
      "grad_norm": 1.1900001764297485,
      "learning_rate": 0.0013805555555555554,
      "loss": 0.278,
      "step": 3910
    },
    {
      "epoch": 7.25,
      "grad_norm": 0.6611624956130981,
      "learning_rate": 0.0013712962962962962,
      "loss": 0.3078,
      "step": 3920
    },
    {
      "epoch": 7.27,
      "grad_norm": 0.8776438236236572,
      "learning_rate": 0.0013620370370370373,
      "loss": 0.3361,
      "step": 3930
    },
    {
      "epoch": 7.29,
      "grad_norm": 0.8815858960151672,
      "learning_rate": 0.0013527777777777778,
      "loss": 0.2727,
      "step": 3940
    },
    {
      "epoch": 7.31,
      "grad_norm": 0.8579031825065613,
      "learning_rate": 0.0013435185185185186,
      "loss": 0.2954,
      "step": 3950
    },
    {
      "epoch": 7.33,
      "grad_norm": 0.9951643943786621,
      "learning_rate": 0.0013342592592592592,
      "loss": 0.3186,
      "step": 3960
    },
    {
      "epoch": 7.35,
      "grad_norm": 0.7032579779624939,
      "learning_rate": 0.001325,
      "loss": 0.3374,
      "step": 3970
    },
    {
      "epoch": 7.36,
      "grad_norm": 0.8467724919319153,
      "learning_rate": 0.0013157407407407406,
      "loss": 0.3459,
      "step": 3980
    },
    {
      "epoch": 7.38,
      "grad_norm": 1.0427024364471436,
      "learning_rate": 0.0013064814814814816,
      "loss": 0.3223,
      "step": 3990
    },
    {
      "epoch": 7.4,
      "grad_norm": 1.081141710281372,
      "learning_rate": 0.0012972222222222222,
      "loss": 0.3421,
      "step": 4000
    },
    {
      "epoch": 7.42,
      "grad_norm": 1.1036946773529053,
      "learning_rate": 0.001287962962962963,
      "loss": 0.3357,
      "step": 4010
    },
    {
      "epoch": 7.44,
      "grad_norm": 0.47026312351226807,
      "learning_rate": 0.0012787037037037038,
      "loss": 0.3413,
      "step": 4020
    },
    {
      "epoch": 7.46,
      "grad_norm": 0.6322646141052246,
      "learning_rate": 0.0012694444444444444,
      "loss": 0.3384,
      "step": 4030
    },
    {
      "epoch": 7.47,
      "grad_norm": 1.1694568395614624,
      "learning_rate": 0.0012601851851851854,
      "loss": 0.2964,
      "step": 4040
    },
    {
      "epoch": 7.49,
      "grad_norm": 0.6389946341514587,
      "learning_rate": 0.001250925925925926,
      "loss": 0.2363,
      "step": 4050
    },
    {
      "epoch": 7.51,
      "grad_norm": 0.8997761011123657,
      "learning_rate": 0.0012416666666666667,
      "loss": 0.3002,
      "step": 4060
    },
    {
      "epoch": 7.53,
      "grad_norm": 1.0065665245056152,
      "learning_rate": 0.0012324074074074073,
      "loss": 0.2559,
      "step": 4070
    },
    {
      "epoch": 7.55,
      "grad_norm": 1.022987723350525,
      "learning_rate": 0.0012231481481481483,
      "loss": 0.3338,
      "step": 4080
    },
    {
      "epoch": 7.57,
      "grad_norm": 1.0679833889007568,
      "learning_rate": 0.001213888888888889,
      "loss": 0.284,
      "step": 4090
    },
    {
      "epoch": 7.59,
      "grad_norm": 0.607288122177124,
      "learning_rate": 0.0012046296296296297,
      "loss": 0.2509,
      "step": 4100
    },
    {
      "epoch": 7.6,
      "grad_norm": 0.5742871165275574,
      "learning_rate": 0.0011953703703703705,
      "loss": 0.2537,
      "step": 4110
    },
    {
      "epoch": 7.62,
      "grad_norm": 1.2387137413024902,
      "learning_rate": 0.001186111111111111,
      "loss": 0.2565,
      "step": 4120
    },
    {
      "epoch": 7.64,
      "grad_norm": 1.174902319908142,
      "learning_rate": 0.0011768518518518519,
      "loss": 0.2997,
      "step": 4130
    },
    {
      "epoch": 7.66,
      "grad_norm": 0.9305229783058167,
      "learning_rate": 0.0011675925925925927,
      "loss": 0.3119,
      "step": 4140
    },
    {
      "epoch": 7.68,
      "grad_norm": 0.8137575387954712,
      "learning_rate": 0.0011583333333333333,
      "loss": 0.3212,
      "step": 4150
    },
    {
      "epoch": 7.7,
      "grad_norm": 0.8960317373275757,
      "learning_rate": 0.001149074074074074,
      "loss": 0.2598,
      "step": 4160
    },
    {
      "epoch": 7.72,
      "grad_norm": 0.9751859903335571,
      "learning_rate": 0.0011398148148148149,
      "loss": 0.2888,
      "step": 4170
    },
    {
      "epoch": 7.73,
      "grad_norm": 1.0055298805236816,
      "learning_rate": 0.0011305555555555557,
      "loss": 0.3002,
      "step": 4180
    },
    {
      "epoch": 7.75,
      "grad_norm": 0.8062968254089355,
      "learning_rate": 0.0011212962962962965,
      "loss": 0.3344,
      "step": 4190
    },
    {
      "epoch": 7.77,
      "grad_norm": 0.95457923412323,
      "learning_rate": 0.001112037037037037,
      "loss": 0.3063,
      "step": 4200
    },
    {
      "epoch": 7.79,
      "grad_norm": 1.3791453838348389,
      "learning_rate": 0.0011027777777777778,
      "loss": 0.317,
      "step": 4210
    },
    {
      "epoch": 7.81,
      "grad_norm": 0.874679684638977,
      "learning_rate": 0.0010935185185185186,
      "loss": 0.3217,
      "step": 4220
    },
    {
      "epoch": 7.83,
      "grad_norm": 0.5453481078147888,
      "learning_rate": 0.0010842592592592592,
      "loss": 0.2493,
      "step": 4230
    },
    {
      "epoch": 7.84,
      "grad_norm": 1.0427902936935425,
      "learning_rate": 0.001075,
      "loss": 0.2792,
      "step": 4240
    },
    {
      "epoch": 7.86,
      "grad_norm": 1.1764284372329712,
      "learning_rate": 0.0010657407407407408,
      "loss": 0.2898,
      "step": 4250
    },
    {
      "epoch": 7.88,
      "grad_norm": 0.6771812438964844,
      "learning_rate": 0.0010564814814814814,
      "loss": 0.373,
      "step": 4260
    },
    {
      "epoch": 7.9,
      "grad_norm": 0.8747534155845642,
      "learning_rate": 0.0010472222222222222,
      "loss": 0.268,
      "step": 4270
    },
    {
      "epoch": 7.92,
      "grad_norm": 1.2883553504943848,
      "learning_rate": 0.001037962962962963,
      "loss": 0.2717,
      "step": 4280
    },
    {
      "epoch": 7.94,
      "grad_norm": 0.7051785588264465,
      "learning_rate": 0.0010287037037037038,
      "loss": 0.3066,
      "step": 4290
    },
    {
      "epoch": 7.96,
      "grad_norm": 0.7398721575737,
      "learning_rate": 0.0010194444444444446,
      "loss": 0.2995,
      "step": 4300
    },
    {
      "epoch": 7.97,
      "grad_norm": 1.145930290222168,
      "learning_rate": 0.0010101851851851851,
      "loss": 0.3045,
      "step": 4310
    },
    {
      "epoch": 7.99,
      "grad_norm": 1.0502551794052124,
      "learning_rate": 0.001000925925925926,
      "loss": 0.2679,
      "step": 4320
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9899861346479741,
      "eval_f1": 0.9912451005847,
      "eval_loss": 0.031062940135598183,
      "eval_precision": 0.9916388059857394,
      "eval_recall": 0.9909443527430603,
      "eval_runtime": 52.8878,
      "eval_samples_per_second": 122.731,
      "eval_steps_per_second": 7.677,
      "step": 4324
    },
    {
      "epoch": 8.01,
      "grad_norm": 1.0673542022705078,
      "learning_rate": 0.0009916666666666667,
      "loss": 0.3395,
      "step": 4330
    },
    {
      "epoch": 8.03,
      "grad_norm": 0.7167376279830933,
      "learning_rate": 0.0009824074074074073,
      "loss": 0.2648,
      "step": 4340
    },
    {
      "epoch": 8.05,
      "grad_norm": 0.7471597194671631,
      "learning_rate": 0.0009731481481481481,
      "loss": 0.3121,
      "step": 4350
    },
    {
      "epoch": 8.07,
      "grad_norm": 1.22334623336792,
      "learning_rate": 0.0009638888888888889,
      "loss": 0.2867,
      "step": 4360
    },
    {
      "epoch": 8.09,
      "grad_norm": 1.0853790044784546,
      "learning_rate": 0.0009546296296296296,
      "loss": 0.2639,
      "step": 4370
    },
    {
      "epoch": 8.1,
      "grad_norm": 0.9974374771118164,
      "learning_rate": 0.0009453703703703703,
      "loss": 0.2679,
      "step": 4380
    },
    {
      "epoch": 8.12,
      "grad_norm": 0.5487973093986511,
      "learning_rate": 0.0009361111111111112,
      "loss": 0.2552,
      "step": 4390
    },
    {
      "epoch": 8.14,
      "grad_norm": 1.0462454557418823,
      "learning_rate": 0.0009268518518518519,
      "loss": 0.2371,
      "step": 4400
    },
    {
      "epoch": 8.16,
      "grad_norm": 0.9754990935325623,
      "learning_rate": 0.0009175925925925927,
      "loss": 0.2792,
      "step": 4410
    },
    {
      "epoch": 8.18,
      "grad_norm": 1.371444821357727,
      "learning_rate": 0.0009083333333333334,
      "loss": 0.2508,
      "step": 4420
    },
    {
      "epoch": 8.2,
      "grad_norm": 1.4113060235977173,
      "learning_rate": 0.0008990740740740741,
      "loss": 0.2875,
      "step": 4430
    },
    {
      "epoch": 8.21,
      "grad_norm": 0.7861378788948059,
      "learning_rate": 0.0008898148148148149,
      "loss": 0.3163,
      "step": 4440
    },
    {
      "epoch": 8.23,
      "grad_norm": 0.6464210152626038,
      "learning_rate": 0.0008805555555555555,
      "loss": 0.2845,
      "step": 4450
    },
    {
      "epoch": 8.25,
      "grad_norm": 1.1672011613845825,
      "learning_rate": 0.0008712962962962962,
      "loss": 0.2807,
      "step": 4460
    },
    {
      "epoch": 8.27,
      "grad_norm": 0.8441992402076721,
      "learning_rate": 0.000862037037037037,
      "loss": 0.3093,
      "step": 4470
    },
    {
      "epoch": 8.29,
      "grad_norm": 0.6474927663803101,
      "learning_rate": 0.0008527777777777777,
      "loss": 0.2609,
      "step": 4480
    },
    {
      "epoch": 8.31,
      "grad_norm": 0.5043164491653442,
      "learning_rate": 0.0008435185185185186,
      "loss": 0.3138,
      "step": 4490
    },
    {
      "epoch": 8.33,
      "grad_norm": 0.9660760164260864,
      "learning_rate": 0.0008342592592592593,
      "loss": 0.3032,
      "step": 4500
    },
    {
      "epoch": 8.34,
      "grad_norm": 1.2470887899398804,
      "learning_rate": 0.0008250000000000001,
      "loss": 0.3015,
      "step": 4510
    },
    {
      "epoch": 8.36,
      "grad_norm": 1.4032260179519653,
      "learning_rate": 0.0008157407407407408,
      "loss": 0.3132,
      "step": 4520
    },
    {
      "epoch": 8.38,
      "grad_norm": 0.9354479312896729,
      "learning_rate": 0.0008064814814814815,
      "loss": 0.2862,
      "step": 4530
    },
    {
      "epoch": 8.4,
      "grad_norm": 0.8781515955924988,
      "learning_rate": 0.0007972222222222223,
      "loss": 0.2636,
      "step": 4540
    },
    {
      "epoch": 8.42,
      "grad_norm": 0.8106943964958191,
      "learning_rate": 0.000787962962962963,
      "loss": 0.3125,
      "step": 4550
    },
    {
      "epoch": 8.44,
      "grad_norm": 0.7355839610099792,
      "learning_rate": 0.0007787037037037037,
      "loss": 0.2896,
      "step": 4560
    },
    {
      "epoch": 8.46,
      "grad_norm": 0.6883030533790588,
      "learning_rate": 0.0007694444444444445,
      "loss": 0.2534,
      "step": 4570
    },
    {
      "epoch": 8.47,
      "grad_norm": 0.7917162179946899,
      "learning_rate": 0.0007601851851851851,
      "loss": 0.2627,
      "step": 4580
    },
    {
      "epoch": 8.49,
      "grad_norm": 0.8958062529563904,
      "learning_rate": 0.0007509259259259258,
      "loss": 0.2573,
      "step": 4590
    },
    {
      "epoch": 8.51,
      "grad_norm": 0.8534148335456848,
      "learning_rate": 0.0007416666666666667,
      "loss": 0.2715,
      "step": 4600
    },
    {
      "epoch": 8.53,
      "grad_norm": 0.8119780421257019,
      "learning_rate": 0.0007324074074074074,
      "loss": 0.2455,
      "step": 4610
    },
    {
      "epoch": 8.55,
      "grad_norm": 1.1178597211837769,
      "learning_rate": 0.0007231481481481482,
      "loss": 0.2798,
      "step": 4620
    },
    {
      "epoch": 8.57,
      "grad_norm": 0.8982829451560974,
      "learning_rate": 0.0007138888888888889,
      "loss": 0.2816,
      "step": 4630
    },
    {
      "epoch": 8.58,
      "grad_norm": 0.8414255976676941,
      "learning_rate": 0.0007046296296296296,
      "loss": 0.263,
      "step": 4640
    },
    {
      "epoch": 8.6,
      "grad_norm": 0.6894321441650391,
      "learning_rate": 0.0006953703703703704,
      "loss": 0.2551,
      "step": 4650
    },
    {
      "epoch": 8.62,
      "grad_norm": 0.9369879364967346,
      "learning_rate": 0.0006861111111111111,
      "loss": 0.234,
      "step": 4660
    },
    {
      "epoch": 8.64,
      "grad_norm": 0.7685711979866028,
      "learning_rate": 0.0006768518518518519,
      "loss": 0.2669,
      "step": 4670
    },
    {
      "epoch": 8.66,
      "grad_norm": 0.866146981716156,
      "learning_rate": 0.0006675925925925926,
      "loss": 0.2613,
      "step": 4680
    },
    {
      "epoch": 8.68,
      "grad_norm": 0.7694200873374939,
      "learning_rate": 0.0006583333333333333,
      "loss": 0.3076,
      "step": 4690
    },
    {
      "epoch": 8.7,
      "grad_norm": 0.7113105058670044,
      "learning_rate": 0.0006490740740740742,
      "loss": 0.2818,
      "step": 4700
    },
    {
      "epoch": 8.71,
      "grad_norm": 0.6957096457481384,
      "learning_rate": 0.0006398148148148148,
      "loss": 0.2123,
      "step": 4710
    },
    {
      "epoch": 8.73,
      "grad_norm": 0.7736373543739319,
      "learning_rate": 0.0006305555555555556,
      "loss": 0.2896,
      "step": 4720
    },
    {
      "epoch": 8.75,
      "grad_norm": 0.7972449660301208,
      "learning_rate": 0.0006212962962962963,
      "loss": 0.2606,
      "step": 4730
    },
    {
      "epoch": 8.77,
      "grad_norm": 1.4942100048065186,
      "learning_rate": 0.000612037037037037,
      "loss": 0.2688,
      "step": 4740
    },
    {
      "epoch": 8.79,
      "grad_norm": 0.5527027249336243,
      "learning_rate": 0.0006027777777777778,
      "loss": 0.2311,
      "step": 4750
    },
    {
      "epoch": 8.81,
      "grad_norm": 0.7232396006584167,
      "learning_rate": 0.0005935185185185185,
      "loss": 0.2102,
      "step": 4760
    },
    {
      "epoch": 8.83,
      "grad_norm": 0.8936424851417542,
      "learning_rate": 0.0005842592592592592,
      "loss": 0.2185,
      "step": 4770
    },
    {
      "epoch": 8.84,
      "grad_norm": 0.5523704290390015,
      "learning_rate": 0.000575,
      "loss": 0.2443,
      "step": 4780
    },
    {
      "epoch": 8.86,
      "grad_norm": 0.6891120672225952,
      "learning_rate": 0.0005657407407407408,
      "loss": 0.2722,
      "step": 4790
    },
    {
      "epoch": 8.88,
      "grad_norm": 0.8618466854095459,
      "learning_rate": 0.0005564814814814815,
      "loss": 0.2782,
      "step": 4800
    },
    {
      "epoch": 8.9,
      "grad_norm": 0.601006269454956,
      "learning_rate": 0.0005472222222222223,
      "loss": 0.3289,
      "step": 4810
    },
    {
      "epoch": 8.92,
      "grad_norm": 0.5365596413612366,
      "learning_rate": 0.000537962962962963,
      "loss": 0.2511,
      "step": 4820
    },
    {
      "epoch": 8.94,
      "grad_norm": 1.0211495161056519,
      "learning_rate": 0.0005287037037037038,
      "loss": 0.2964,
      "step": 4830
    },
    {
      "epoch": 8.95,
      "grad_norm": 0.838763952255249,
      "learning_rate": 0.0005194444444444444,
      "loss": 0.2373,
      "step": 4840
    },
    {
      "epoch": 8.97,
      "grad_norm": 0.9094264507293701,
      "learning_rate": 0.0005101851851851852,
      "loss": 0.2447,
      "step": 4850
    },
    {
      "epoch": 8.99,
      "grad_norm": 0.5573397874832153,
      "learning_rate": 0.0005009259259259259,
      "loss": 0.2439,
      "step": 4860
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9850562317054383,
      "eval_f1": 0.9881355002597147,
      "eval_loss": 0.05773022025823593,
      "eval_precision": 0.9885823118322556,
      "eval_recall": 0.9880245106891419,
      "eval_runtime": 52.821,
      "eval_samples_per_second": 122.887,
      "eval_steps_per_second": 7.686,
      "step": 4864
    },
    {
      "epoch": 9.01,
      "grad_norm": 0.5047478675842285,
      "learning_rate": 0.0004916666666666666,
      "loss": 0.2301,
      "step": 4870
    },
    {
      "epoch": 9.03,
      "grad_norm": 0.8074342012405396,
      "learning_rate": 0.00048240740740740747,
      "loss": 0.2168,
      "step": 4880
    },
    {
      "epoch": 9.05,
      "grad_norm": 0.9957315921783447,
      "learning_rate": 0.00047314814814814816,
      "loss": 0.2546,
      "step": 4890
    },
    {
      "epoch": 9.07,
      "grad_norm": 1.1737314462661743,
      "learning_rate": 0.0004638888888888889,
      "loss": 0.2526,
      "step": 4900
    },
    {
      "epoch": 9.08,
      "grad_norm": 0.47173938155174255,
      "learning_rate": 0.00045462962962962964,
      "loss": 0.281,
      "step": 4910
    },
    {
      "epoch": 9.1,
      "grad_norm": 0.7715136408805847,
      "learning_rate": 0.00044537037037037033,
      "loss": 0.2679,
      "step": 4920
    },
    {
      "epoch": 9.12,
      "grad_norm": 1.0114617347717285,
      "learning_rate": 0.00043611111111111113,
      "loss": 0.2568,
      "step": 4930
    },
    {
      "epoch": 9.14,
      "grad_norm": 0.7581719160079956,
      "learning_rate": 0.00042685185185185187,
      "loss": 0.1955,
      "step": 4940
    },
    {
      "epoch": 9.16,
      "grad_norm": 0.9538590908050537,
      "learning_rate": 0.0004175925925925926,
      "loss": 0.2764,
      "step": 4950
    },
    {
      "epoch": 9.18,
      "grad_norm": 0.9304882884025574,
      "learning_rate": 0.00040833333333333336,
      "loss": 0.2225,
      "step": 4960
    },
    {
      "epoch": 9.2,
      "grad_norm": 1.7399593591690063,
      "learning_rate": 0.00039907407407407404,
      "loss": 0.2194,
      "step": 4970
    },
    {
      "epoch": 9.21,
      "grad_norm": 1.2996737957000732,
      "learning_rate": 0.0003898148148148148,
      "loss": 0.2552,
      "step": 4980
    },
    {
      "epoch": 9.23,
      "grad_norm": 0.9532219767570496,
      "learning_rate": 0.0003805555555555556,
      "loss": 0.2371,
      "step": 4990
    },
    {
      "epoch": 9.25,
      "grad_norm": 0.5695217251777649,
      "learning_rate": 0.0003712962962962963,
      "loss": 0.1966,
      "step": 5000
    },
    {
      "epoch": 9.27,
      "grad_norm": 0.8924893140792847,
      "learning_rate": 0.000362037037037037,
      "loss": 0.2367,
      "step": 5010
    },
    {
      "epoch": 9.29,
      "grad_norm": 0.4599965214729309,
      "learning_rate": 0.00035277777777777776,
      "loss": 0.2338,
      "step": 5020
    },
    {
      "epoch": 9.31,
      "grad_norm": 1.544420838356018,
      "learning_rate": 0.0003435185185185185,
      "loss": 0.2474,
      "step": 5030
    },
    {
      "epoch": 9.32,
      "grad_norm": 0.7134206295013428,
      "learning_rate": 0.0003342592592592593,
      "loss": 0.2406,
      "step": 5040
    },
    {
      "epoch": 9.34,
      "grad_norm": 0.510735273361206,
      "learning_rate": 0.00032500000000000004,
      "loss": 0.2181,
      "step": 5050
    },
    {
      "epoch": 9.36,
      "grad_norm": 0.7918537855148315,
      "learning_rate": 0.00031574074074074073,
      "loss": 0.2582,
      "step": 5060
    },
    {
      "epoch": 9.38,
      "grad_norm": 1.1104782819747925,
      "learning_rate": 0.00030648148148148147,
      "loss": 0.2531,
      "step": 5070
    },
    {
      "epoch": 9.4,
      "grad_norm": 0.8858848810195923,
      "learning_rate": 0.0002972222222222222,
      "loss": 0.256,
      "step": 5080
    },
    {
      "epoch": 9.42,
      "grad_norm": 0.9814337491989136,
      "learning_rate": 0.00028796296296296296,
      "loss": 0.2301,
      "step": 5090
    },
    {
      "epoch": 9.44,
      "grad_norm": 0.8091614246368408,
      "learning_rate": 0.0002787037037037037,
      "loss": 0.243,
      "step": 5100
    },
    {
      "epoch": 9.45,
      "grad_norm": 0.8779675960540771,
      "learning_rate": 0.00026944444444444444,
      "loss": 0.2248,
      "step": 5110
    },
    {
      "epoch": 9.47,
      "grad_norm": 0.7793891429901123,
      "learning_rate": 0.0002601851851851852,
      "loss": 0.2431,
      "step": 5120
    },
    {
      "epoch": 9.49,
      "grad_norm": 0.7594252824783325,
      "learning_rate": 0.0002509259259259259,
      "loss": 0.2766,
      "step": 5130
    },
    {
      "epoch": 9.51,
      "grad_norm": 0.7500623464584351,
      "learning_rate": 0.00024166666666666667,
      "loss": 0.2608,
      "step": 5140
    },
    {
      "epoch": 9.53,
      "grad_norm": 1.316781997680664,
      "learning_rate": 0.00023240740740740744,
      "loss": 0.2924,
      "step": 5150
    },
    {
      "epoch": 9.55,
      "grad_norm": 0.6205342411994934,
      "learning_rate": 0.00022314814814814815,
      "loss": 0.2192,
      "step": 5160
    },
    {
      "epoch": 9.57,
      "grad_norm": 0.4915425181388855,
      "learning_rate": 0.0002138888888888889,
      "loss": 0.2284,
      "step": 5170
    },
    {
      "epoch": 9.58,
      "grad_norm": 0.5837624669075012,
      "learning_rate": 0.00020462962962962964,
      "loss": 0.2481,
      "step": 5180
    },
    {
      "epoch": 9.6,
      "grad_norm": 1.0567339658737183,
      "learning_rate": 0.00019537037037037038,
      "loss": 0.2289,
      "step": 5190
    },
    {
      "epoch": 9.62,
      "grad_norm": 0.548608660697937,
      "learning_rate": 0.0001861111111111111,
      "loss": 0.2723,
      "step": 5200
    },
    {
      "epoch": 9.64,
      "grad_norm": 0.46330851316452026,
      "learning_rate": 0.00017685185185185187,
      "loss": 0.2482,
      "step": 5210
    },
    {
      "epoch": 9.66,
      "grad_norm": 0.5314125418663025,
      "learning_rate": 0.00016759259259259258,
      "loss": 0.2305,
      "step": 5220
    },
    {
      "epoch": 9.68,
      "grad_norm": 1.2373839616775513,
      "learning_rate": 0.00015833333333333335,
      "loss": 0.2508,
      "step": 5230
    },
    {
      "epoch": 9.69,
      "grad_norm": 0.63730788230896,
      "learning_rate": 0.00014907407407407407,
      "loss": 0.2366,
      "step": 5240
    },
    {
      "epoch": 9.71,
      "grad_norm": 0.8189606666564941,
      "learning_rate": 0.0001398148148148148,
      "loss": 0.2473,
      "step": 5250
    },
    {
      "epoch": 9.73,
      "grad_norm": 0.4223102629184723,
      "learning_rate": 0.00013055555555555558,
      "loss": 0.1921,
      "step": 5260
    },
    {
      "epoch": 9.75,
      "grad_norm": 0.8108687996864319,
      "learning_rate": 0.0001212962962962963,
      "loss": 0.211,
      "step": 5270
    },
    {
      "epoch": 9.77,
      "grad_norm": 1.0115864276885986,
      "learning_rate": 0.00011203703703703704,
      "loss": 0.2199,
      "step": 5280
    },
    {
      "epoch": 9.79,
      "grad_norm": 0.578581690788269,
      "learning_rate": 0.00010277777777777778,
      "loss": 0.2079,
      "step": 5290
    },
    {
      "epoch": 9.81,
      "grad_norm": 0.7213098406791687,
      "learning_rate": 9.351851851851852e-05,
      "loss": 0.2709,
      "step": 5300
    },
    {
      "epoch": 9.82,
      "grad_norm": 0.9571123123168945,
      "learning_rate": 8.425925925925925e-05,
      "loss": 0.2925,
      "step": 5310
    },
    {
      "epoch": 9.84,
      "grad_norm": 1.0856187343597412,
      "learning_rate": 7.5e-05,
      "loss": 0.2708,
      "step": 5320
    },
    {
      "epoch": 9.86,
      "grad_norm": 0.4601670801639557,
      "learning_rate": 6.574074074074075e-05,
      "loss": 0.1895,
      "step": 5330
    },
    {
      "epoch": 9.88,
      "grad_norm": 0.7248308062553406,
      "learning_rate": 5.648148148148148e-05,
      "loss": 0.2639,
      "step": 5340
    },
    {
      "epoch": 9.9,
      "grad_norm": 0.8658019304275513,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.2721,
      "step": 5350
    },
    {
      "epoch": 9.92,
      "grad_norm": 0.877480685710907,
      "learning_rate": 3.7962962962962964e-05,
      "loss": 0.2026,
      "step": 5360
    },
    {
      "epoch": 9.94,
      "grad_norm": 0.7867253422737122,
      "learning_rate": 2.8703703703703703e-05,
      "loss": 0.2431,
      "step": 5370
    },
    {
      "epoch": 9.95,
      "grad_norm": 0.5923830270767212,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.2207,
      "step": 5380
    },
    {
      "epoch": 9.97,
      "grad_norm": 0.9413737058639526,
      "learning_rate": 1.0185185185185185e-05,
      "loss": 0.223,
      "step": 5390
    },
    {
      "epoch": 9.99,
      "grad_norm": 0.630954921245575,
      "learning_rate": 9.259259259259259e-07,
      "loss": 0.2169,
      "step": 5400
    },
    {
      "epoch": 9.99,
      "eval_accuracy": 0.9835156370358958,
      "eval_f1": 0.9881773422996485,
      "eval_loss": 0.0720474049448967,
      "eval_precision": 0.9887528399741967,
      "eval_recall": 0.988172197902048,
      "eval_runtime": 52.921,
      "eval_samples_per_second": 122.654,
      "eval_steps_per_second": 7.672,
      "step": 5400
    },
    {
      "epoch": 9.99,
      "step": 5400,
      "total_flos": 2.6962411056639013e+19,
      "train_loss": 0.442498734637543,
      "train_runtime": 5918.7974,
      "train_samples_per_second": 58.426,
      "train_steps_per_second": 0.912
    }
  ],
  "logging_steps": 10,
  "max_steps": 5400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.6962411056639013e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}