{
  "best_metric": 0.8015357353809806,
  "best_model_checkpoint": "/tmp/logs/binary_classification_model_v3.1.5_Junction_folder10/checkpoint-4735",
  "epoch": 450.95238095238096,
  "eval_steps": 500,
  "global_step": 4735,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09523809523809523,
      "grad_norm": Infinity,
      "learning_rate": 0.0,
      "loss": 1.4012,
      "step": 1
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 105.49879455566406,
      "learning_rate": 1.8740629685157423e-10,
      "loss": 1.463,
      "step": 2
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 104.57809448242188,
      "learning_rate": 3.7481259370314846e-10,
      "loss": 1.3533,
      "step": 3
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 96.78276062011719,
      "learning_rate": 5.622188905547227e-10,
      "loss": 1.3083,
      "step": 4
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 86.87081146240234,
      "learning_rate": 7.496251874062969e-10,
      "loss": 1.3851,
      "step": 5
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 99.54174041748047,
      "learning_rate": 9.37031484257871e-10,
      "loss": 1.4052,
      "step": 6
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 105.97377014160156,
      "learning_rate": 1.1244377811094454e-09,
      "loss": 1.3674,
      "step": 7
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 117.7193374633789,
      "learning_rate": 1.3118440779610195e-09,
      "loss": 1.4592,
      "step": 8
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 110.17356872558594,
      "learning_rate": 1.4992503748125938e-09,
      "loss": 1.4495,
      "step": 9
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 109.03305053710938,
      "learning_rate": 1.6866566716641682e-09,
      "loss": 1.4223,
      "step": 10
    },
    {
      "epoch": 0.9523809523809523,
      "eval_accuracy": 0.6852753840389659,
      "eval_f1": 0.7584818861414606,
      "eval_loss": 0.6929942965507507,
      "eval_precision": 0.6820062047569804,
      "eval_recall": 0.8542746113989638,
      "eval_roc_auc": 0.7219533678756477,
      "eval_runtime": 1.202,
      "eval_samples_per_second": 2220.439,
      "eval_steps_per_second": 4.992,
      "step": 10
    },
    {
      "epoch": 1.0476190476190477,
      "grad_norm": 80.2335433959961,
      "learning_rate": 1.874062968515742e-09,
      "loss": 1.3215,
      "step": 11
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 115.97867584228516,
      "learning_rate": 2.0614692653673164e-09,
      "loss": 1.4214,
      "step": 12
    },
    {
      "epoch": 1.2380952380952381,
      "grad_norm": 105.73731231689453,
      "learning_rate": 2.2488755622188908e-09,
      "loss": 1.4118,
      "step": 13
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 103.12907409667969,
      "learning_rate": 2.436281859070465e-09,
      "loss": 1.3847,
      "step": 14
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 112.44313049316406,
      "learning_rate": 2.623688155922039e-09,
      "loss": 1.4623,
      "step": 15
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 103.6406478881836,
      "learning_rate": 2.8110944527736133e-09,
      "loss": 1.4486,
      "step": 16
    },
    {
      "epoch": 1.619047619047619,
      "grad_norm": 90.27802276611328,
      "learning_rate": 2.9985007496251877e-09,
      "loss": 1.404,
      "step": 17
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 96.4856185913086,
      "learning_rate": 3.1859070464767616e-09,
      "loss": 1.4192,
      "step": 18
    },
    {
      "epoch": 1.8095238095238095,
      "grad_norm": 101.81619262695312,
      "learning_rate": 3.3733133433283363e-09,
      "loss": 1.3883,
      "step": 19
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 113.48494720458984,
      "learning_rate": 3.5607196401799103e-09,
      "loss": 1.4136,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 98.515625,
      "learning_rate": 3.748125937031484e-09,
      "loss": 1.422,
      "step": 21
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6852753840389659,
      "eval_f1": 0.7583429228998849,
      "eval_loss": 0.6922231316566467,
      "eval_precision": 0.6821946169772257,
      "eval_recall": 0.8536269430051814,
      "eval_roc_auc": 0.7219496257915946,
      "eval_runtime": 1.2942,
      "eval_samples_per_second": 2062.333,
      "eval_steps_per_second": 4.636,
      "step": 21
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 102.9639663696289,
      "learning_rate": 3.9355322338830585e-09,
      "loss": 1.3666,
      "step": 22
    },
    {
      "epoch": 2.1904761904761907,
      "grad_norm": 106.0006332397461,
      "learning_rate": 4.122938530734633e-09,
      "loss": 1.3874,
      "step": 23
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 97.28984832763672,
      "learning_rate": 4.310344827586206e-09,
      "loss": 1.3638,
      "step": 24
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 106.2695083618164,
      "learning_rate": 4.4977511244377815e-09,
      "loss": 1.4204,
      "step": 25
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 112.76600646972656,
      "learning_rate": 4.685157421289355e-09,
      "loss": 1.3955,
      "step": 26
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 96.83776092529297,
      "learning_rate": 4.87256371814093e-09,
      "loss": 1.3999,
      "step": 27
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 87.85649108886719,
      "learning_rate": 5.059970014992504e-09,
      "loss": 1.3537,
      "step": 28
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 102.71916961669922,
      "learning_rate": 5.247376311844078e-09,
      "loss": 1.4219,
      "step": 29
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 99.2935791015625,
      "learning_rate": 5.434782608695652e-09,
      "loss": 1.3916,
      "step": 30
    },
    {
      "epoch": 2.9523809523809526,
      "grad_norm": 95.25791931152344,
      "learning_rate": 5.622188905547227e-09,
      "loss": 1.4288,
      "step": 31
    },
    {
      "epoch": 2.9523809523809526,
      "eval_accuracy": 0.6845260397152492,
      "eval_f1": 0.7574884792626728,
      "eval_loss": 0.6911234855651855,
      "eval_precision": 0.6820539419087137,
      "eval_recall": 0.8516839378238342,
      "eval_roc_auc": 0.7219237190558434,
      "eval_runtime": 1.1237,
      "eval_samples_per_second": 2375.214,
      "eval_steps_per_second": 5.34,
      "step": 31
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 104.1244125366211,
      "learning_rate": 5.8095952023988e-09,
      "loss": 1.4041,
      "step": 32
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 101.19844055175781,
      "learning_rate": 5.997001499250375e-09,
      "loss": 1.3404,
      "step": 33
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 87.8860855102539,
      "learning_rate": 6.184407796101949e-09,
      "loss": 1.4263,
      "step": 34
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 96.34718322753906,
      "learning_rate": 6.371814092953523e-09,
      "loss": 1.3771,
      "step": 35
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 98.07717895507812,
      "learning_rate": 6.5592203898050975e-09,
      "loss": 1.4294,
      "step": 36
    },
    {
      "epoch": 3.5238095238095237,
      "grad_norm": 98.57717895507812,
      "learning_rate": 6.746626686656673e-09,
      "loss": 1.3477,
      "step": 37
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 106.59417724609375,
      "learning_rate": 6.934032983508246e-09,
      "loss": 1.4278,
      "step": 38
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 101.46930694580078,
      "learning_rate": 7.1214392803598205e-09,
      "loss": 1.3988,
      "step": 39
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 115.23248291015625,
      "learning_rate": 7.308845577211394e-09,
      "loss": 1.4424,
      "step": 40
    },
    {
      "epoch": 3.9047619047619047,
      "grad_norm": 100.2016830444336,
      "learning_rate": 7.496251874062968e-09,
      "loss": 1.3932,
      "step": 41
    },
    {
      "epoch": 4.0,
      "grad_norm": 110.41011810302734,
      "learning_rate": 7.683658170914544e-09,
      "loss": 1.4236,
      "step": 42
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6856500562008243,
      "eval_f1": 0.7580040380732622,
      "eval_loss": 0.6892919540405273,
      "eval_precision": 0.6833073322932918,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.7218949337938974,
      "eval_runtime": 1.1259,
      "eval_samples_per_second": 2370.508,
      "eval_steps_per_second": 5.329,
      "step": 42
    },
    {
      "epoch": 4.095238095238095,
      "grad_norm": 86.44599914550781,
      "learning_rate": 7.871064467766117e-09,
      "loss": 1.3405,
      "step": 43
    },
    {
      "epoch": 4.190476190476191,
      "grad_norm": 89.20963287353516,
      "learning_rate": 8.058470764617691e-09,
      "loss": 1.4478,
      "step": 44
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 107.15955352783203,
      "learning_rate": 8.245877061469266e-09,
      "loss": 1.3755,
      "step": 45
    },
    {
      "epoch": 4.380952380952381,
      "grad_norm": 102.78776550292969,
      "learning_rate": 8.43328335832084e-09,
      "loss": 1.4022,
      "step": 46
    },
    {
      "epoch": 4.476190476190476,
      "grad_norm": 120.45516204833984,
      "learning_rate": 8.620689655172413e-09,
      "loss": 1.4396,
      "step": 47
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 92.11113739013672,
      "learning_rate": 8.808095952023989e-09,
      "loss": 1.3656,
      "step": 48
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 100.20201873779297,
      "learning_rate": 8.995502248875563e-09,
      "loss": 1.4059,
      "step": 49
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 88.53689575195312,
      "learning_rate": 9.182908545727137e-09,
      "loss": 1.3268,
      "step": 50
    },
    {
      "epoch": 4.857142857142857,
      "grad_norm": 106.94924926757812,
      "learning_rate": 9.37031484257871e-09,
      "loss": 1.4176,
      "step": 51
    },
    {
      "epoch": 4.9523809523809526,
      "grad_norm": 91.2048110961914,
      "learning_rate": 9.557721139430284e-09,
      "loss": 1.4304,
      "step": 52
    },
    {
      "epoch": 4.9523809523809526,
      "eval_accuracy": 0.6882727613338329,
      "eval_f1": 0.7591198610306891,
      "eval_loss": 0.6873071789741516,
      "eval_precision": 0.6863874345549739,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7219168105929764,
      "eval_runtime": 1.2348,
      "eval_samples_per_second": 2161.52,
      "eval_steps_per_second": 4.859,
      "step": 52
    },
    {
      "epoch": 5.0476190476190474,
      "grad_norm": 113.53949737548828,
      "learning_rate": 9.74512743628186e-09,
      "loss": 1.4397,
      "step": 53
    },
    {
      "epoch": 5.142857142857143,
      "grad_norm": 109.27808380126953,
      "learning_rate": 9.932533733133435e-09,
      "loss": 1.4053,
      "step": 54
    },
    {
      "epoch": 5.238095238095238,
      "grad_norm": 90.2195816040039,
      "learning_rate": 1.0119940029985007e-08,
      "loss": 1.3786,
      "step": 55
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 91.2532730102539,
      "learning_rate": 1.0307346326836582e-08,
      "loss": 1.3868,
      "step": 56
    },
    {
      "epoch": 5.428571428571429,
      "grad_norm": 109.18313598632812,
      "learning_rate": 1.0494752623688156e-08,
      "loss": 1.3365,
      "step": 57
    },
    {
      "epoch": 5.523809523809524,
      "grad_norm": 74.70710754394531,
      "learning_rate": 1.0682158920539732e-08,
      "loss": 1.4311,
      "step": 58
    },
    {
      "epoch": 5.619047619047619,
      "grad_norm": 105.76243591308594,
      "learning_rate": 1.0869565217391305e-08,
      "loss": 1.3777,
      "step": 59
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 87.60913848876953,
      "learning_rate": 1.1056971514242879e-08,
      "loss": 1.405,
      "step": 60
    },
    {
      "epoch": 5.809523809523809,
      "grad_norm": 92.69194030761719,
      "learning_rate": 1.1244377811094453e-08,
      "loss": 1.4627,
      "step": 61
    },
    {
      "epoch": 5.904761904761905,
      "grad_norm": 90.4942855834961,
      "learning_rate": 1.1431784107946028e-08,
      "loss": 1.3523,
      "step": 62
    },
    {
      "epoch": 6.0,
      "grad_norm": 87.5588150024414,
      "learning_rate": 1.16191904047976e-08,
      "loss": 1.3618,
      "step": 63
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6863994005245411,
      "eval_f1": 0.7566152951439372,
      "eval_loss": 0.6846460700035095,
      "eval_precision": 0.6865435356200528,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7218773747841105,
      "eval_runtime": 1.222,
      "eval_samples_per_second": 2184.196,
      "eval_steps_per_second": 4.91,
      "step": 63
    },
    {
      "epoch": 6.095238095238095,
      "grad_norm": 96.4197998046875,
      "learning_rate": 1.1806596701649176e-08,
      "loss": 1.4462,
      "step": 64
    },
    {
      "epoch": 6.190476190476191,
      "grad_norm": 96.55722045898438,
      "learning_rate": 1.199400299850075e-08,
      "loss": 1.45,
      "step": 65
    },
    {
      "epoch": 6.285714285714286,
      "grad_norm": 84.94829559326172,
      "learning_rate": 1.2181409295352325e-08,
      "loss": 1.3282,
      "step": 66
    },
    {
      "epoch": 6.380952380952381,
      "grad_norm": 85.26909637451172,
      "learning_rate": 1.2368815592203898e-08,
      "loss": 1.3238,
      "step": 67
    },
    {
      "epoch": 6.476190476190476,
      "grad_norm": 101.26187896728516,
      "learning_rate": 1.2556221889055474e-08,
      "loss": 1.4357,
      "step": 68
    },
    {
      "epoch": 6.571428571428571,
      "grad_norm": 78.13963317871094,
      "learning_rate": 1.2743628185907046e-08,
      "loss": 1.316,
      "step": 69
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 97.32000732421875,
      "learning_rate": 1.2931034482758622e-08,
      "loss": 1.4427,
      "step": 70
    },
    {
      "epoch": 6.761904761904762,
      "grad_norm": 93.46250915527344,
      "learning_rate": 1.3118440779610195e-08,
      "loss": 1.3841,
      "step": 71
    },
    {
      "epoch": 6.857142857142857,
      "grad_norm": 107.21956634521484,
      "learning_rate": 1.330584707646177e-08,
      "loss": 1.4041,
      "step": 72
    },
    {
      "epoch": 6.9523809523809526,
      "grad_norm": 81.90492248535156,
      "learning_rate": 1.3493253373313345e-08,
      "loss": 1.4,
      "step": 73
    },
    {
      "epoch": 6.9523809523809526,
      "eval_accuracy": 0.6856500562008243,
      "eval_f1": 0.755036496350365,
      "eval_loss": 0.6818766593933105,
      "eval_precision": 0.6874003189792663,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7218707541738629,
      "eval_runtime": 1.2794,
      "eval_samples_per_second": 2086.087,
      "eval_steps_per_second": 4.69,
      "step": 73
    },
    {
      "epoch": 7.0476190476190474,
      "grad_norm": 87.02517700195312,
      "learning_rate": 1.3680659670164918e-08,
      "loss": 1.274,
      "step": 74
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 89.8641586303711,
      "learning_rate": 1.3868065967016492e-08,
      "loss": 1.3081,
      "step": 75
    },
    {
      "epoch": 7.238095238095238,
      "grad_norm": 90.02095031738281,
      "learning_rate": 1.4055472263868065e-08,
      "loss": 1.4267,
      "step": 76
    },
    {
      "epoch": 7.333333333333333,
      "grad_norm": 86.56961059570312,
      "learning_rate": 1.4242878560719641e-08,
      "loss": 1.3516,
      "step": 77
    },
    {
      "epoch": 7.428571428571429,
      "grad_norm": 93.65435028076172,
      "learning_rate": 1.4430284857571215e-08,
      "loss": 1.393,
      "step": 78
    },
    {
      "epoch": 7.523809523809524,
      "grad_norm": 75.84454345703125,
      "learning_rate": 1.4617691154422788e-08,
      "loss": 1.3685,
      "step": 79
    },
    {
      "epoch": 7.619047619047619,
      "grad_norm": 92.14884948730469,
      "learning_rate": 1.4805097451274364e-08,
      "loss": 1.4322,
      "step": 80
    },
    {
      "epoch": 7.714285714285714,
      "grad_norm": 96.71854400634766,
      "learning_rate": 1.4992503748125937e-08,
      "loss": 1.4496,
      "step": 81
    },
    {
      "epoch": 7.809523809523809,
      "grad_norm": 76.76438903808594,
      "learning_rate": 1.517991004497751e-08,
      "loss": 1.3499,
      "step": 82
    },
    {
      "epoch": 7.904761904761905,
      "grad_norm": 85.24507904052734,
      "learning_rate": 1.536731634182909e-08,
      "loss": 1.3642,
      "step": 83
    },
    {
      "epoch": 8.0,
      "grad_norm": 94.62620544433594,
      "learning_rate": 1.555472263868066e-08,
      "loss": 1.37,
      "step": 84
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6860247283626827,
      "eval_f1": 0.7541079812206573,
      "eval_loss": 0.6784666776657104,
      "eval_precision": 0.6893776824034334,
      "eval_recall": 0.8322538860103627,
      "eval_roc_auc": 0.721849453080023,
      "eval_runtime": 1.1703,
      "eval_samples_per_second": 2280.546,
      "eval_steps_per_second": 5.127,
      "step": 84
    },
    {
      "epoch": 8.095238095238095,
      "grad_norm": 90.98119354248047,
      "learning_rate": 1.5742128935532234e-08,
      "loss": 1.3537,
      "step": 85
    },
    {
      "epoch": 8.19047619047619,
      "grad_norm": 83.05977630615234,
      "learning_rate": 1.592953523238381e-08,
      "loss": 1.3945,
      "step": 86
    },
    {
      "epoch": 8.285714285714286,
      "grad_norm": 77.79987335205078,
      "learning_rate": 1.6116941529235383e-08,
      "loss": 1.3033,
      "step": 87
    },
    {
      "epoch": 8.380952380952381,
      "grad_norm": 83.56439971923828,
      "learning_rate": 1.6304347826086957e-08,
      "loss": 1.4001,
      "step": 88
    },
    {
      "epoch": 8.476190476190476,
      "grad_norm": 65.35790252685547,
      "learning_rate": 1.649175412293853e-08,
      "loss": 1.3183,
      "step": 89
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 72.92051696777344,
      "learning_rate": 1.6679160419790106e-08,
      "loss": 1.2747,
      "step": 90
    },
    {
      "epoch": 8.666666666666666,
      "grad_norm": 92.48654174804688,
      "learning_rate": 1.686656671664168e-08,
      "loss": 1.3942,
      "step": 91
    },
    {
      "epoch": 8.761904761904763,
      "grad_norm": 92.03863525390625,
      "learning_rate": 1.7053973013493254e-08,
      "loss": 1.3977,
      "step": 92
    },
    {
      "epoch": 8.857142857142858,
      "grad_norm": 78.154296875,
      "learning_rate": 1.7241379310344825e-08,
      "loss": 1.4182,
      "step": 93
    },
    {
      "epoch": 8.952380952380953,
      "grad_norm": 77.85521697998047,
      "learning_rate": 1.7428785607196403e-08,
      "loss": 1.339,
      "step": 94
    },
    {
      "epoch": 8.952380952380953,
      "eval_accuracy": 0.6871487448482577,
      "eval_f1": 0.7541948778333823,
      "eval_loss": 0.675155758857727,
      "eval_precision": 0.6913113869400972,
      "eval_recall": 0.8296632124352331,
      "eval_roc_auc": 0.7217829591249281,
      "eval_runtime": 1.1052,
      "eval_samples_per_second": 2414.859,
      "eval_steps_per_second": 5.429,
      "step": 94
    },
    {
      "epoch": 9.047619047619047,
      "grad_norm": 75.20770263671875,
      "learning_rate": 1.7616191904047977e-08,
      "loss": 1.3267,
      "step": 95
    },
    {
      "epoch": 9.142857142857142,
      "grad_norm": 74.02423858642578,
      "learning_rate": 1.7803598200899552e-08,
      "loss": 1.4283,
      "step": 96
    },
    {
      "epoch": 9.238095238095237,
      "grad_norm": 71.2074966430664,
      "learning_rate": 1.7991004497751126e-08,
      "loss": 1.3037,
      "step": 97
    },
    {
      "epoch": 9.333333333333334,
      "grad_norm": 70.90864562988281,
      "learning_rate": 1.8178410794602697e-08,
      "loss": 1.3745,
      "step": 98
    },
    {
      "epoch": 9.428571428571429,
      "grad_norm": 93.27525329589844,
      "learning_rate": 1.8365817091454275e-08,
      "loss": 1.4259,
      "step": 99
    },
    {
      "epoch": 9.523809523809524,
      "grad_norm": 80.44207763671875,
      "learning_rate": 1.855322338830585e-08,
      "loss": 1.3235,
      "step": 100
    },
    {
      "epoch": 9.619047619047619,
      "grad_norm": 69.08979034423828,
      "learning_rate": 1.874062968515742e-08,
      "loss": 1.3567,
      "step": 101
    },
    {
      "epoch": 9.714285714285714,
      "grad_norm": 89.39695739746094,
      "learning_rate": 1.8928035982008998e-08,
      "loss": 1.4014,
      "step": 102
    },
    {
      "epoch": 9.80952380952381,
      "grad_norm": 91.82769012451172,
      "learning_rate": 1.911544227886057e-08,
      "loss": 1.4411,
      "step": 103
    },
    {
      "epoch": 9.904761904761905,
      "grad_norm": 78.9308090209961,
      "learning_rate": 1.9302848575712146e-08,
      "loss": 1.3307,
      "step": 104
    },
    {
      "epoch": 10.0,
      "grad_norm": 65.37604522705078,
      "learning_rate": 1.949025487256372e-08,
      "loss": 1.3799,
      "step": 105
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6875234170101161,
      "eval_f1": 0.7534003548196333,
      "eval_loss": 0.6712872982025146,
      "eval_precision": 0.6931447225244831,
      "eval_recall": 0.8251295336787565,
      "eval_roc_auc": 0.7217037996545768,
      "eval_runtime": 1.1823,
      "eval_samples_per_second": 2257.479,
      "eval_steps_per_second": 5.075,
      "step": 105
    },
    {
      "epoch": 10.095238095238095,
      "grad_norm": 77.4139175415039,
      "learning_rate": 1.9677661169415292e-08,
      "loss": 1.3822,
      "step": 106
    },
    {
      "epoch": 10.19047619047619,
      "grad_norm": 75.99652862548828,
      "learning_rate": 1.986506746626687e-08,
      "loss": 1.3308,
      "step": 107
    },
    {
      "epoch": 10.285714285714286,
      "grad_norm": 73.82905578613281,
      "learning_rate": 2.005247376311844e-08,
      "loss": 1.3461,
      "step": 108
    },
    {
      "epoch": 10.380952380952381,
      "grad_norm": 69.6850814819336,
      "learning_rate": 2.0239880059970015e-08,
      "loss": 1.2927,
      "step": 109
    },
    {
      "epoch": 10.476190476190476,
      "grad_norm": 83.00745391845703,
      "learning_rate": 2.0427286356821592e-08,
      "loss": 1.395,
      "step": 110
    },
    {
      "epoch": 10.571428571428571,
      "grad_norm": 76.98029327392578,
      "learning_rate": 2.0614692653673163e-08,
      "loss": 1.3835,
      "step": 111
    },
    {
      "epoch": 10.666666666666666,
      "grad_norm": 65.298828125,
      "learning_rate": 2.0802098950524738e-08,
      "loss": 1.3835,
      "step": 112
    },
    {
      "epoch": 10.761904761904763,
      "grad_norm": 65.25972747802734,
      "learning_rate": 2.0989505247376312e-08,
      "loss": 1.3321,
      "step": 113
    },
    {
      "epoch": 10.857142857142858,
      "grad_norm": 71.95514678955078,
      "learning_rate": 2.1176911544227886e-08,
      "loss": 1.3456,
      "step": 114
    },
    {
      "epoch": 10.952380952380953,
      "grad_norm": 71.81147003173828,
      "learning_rate": 2.1364317841079464e-08,
      "loss": 1.4254,
      "step": 115
    },
    {
      "epoch": 10.952380952380953,
      "eval_accuracy": 0.6878980891719745,
      "eval_f1": 0.7528923168199347,
      "eval_loss": 0.6676313281059265,
      "eval_precision": 0.6945812807881774,
      "eval_recall": 0.8218911917098446,
      "eval_roc_auc": 0.721703223949338,
      "eval_runtime": 1.459,
      "eval_samples_per_second": 1829.393,
      "eval_steps_per_second": 4.113,
      "step": 115
    },
    {
      "epoch": 11.047619047619047,
      "grad_norm": 56.503204345703125,
      "learning_rate": 2.1551724137931035e-08,
      "loss": 1.3297,
      "step": 116
    },
    {
      "epoch": 11.142857142857142,
      "grad_norm": 58.1290397644043,
      "learning_rate": 2.173913043478261e-08,
      "loss": 1.3334,
      "step": 117
    },
    {
      "epoch": 11.238095238095237,
      "grad_norm": 74.19014739990234,
      "learning_rate": 2.1926536731634184e-08,
      "loss": 1.3858,
      "step": 118
    },
    {
      "epoch": 11.333333333333334,
      "grad_norm": 74.8871841430664,
      "learning_rate": 2.2113943028485758e-08,
      "loss": 1.3663,
      "step": 119
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 78.6004867553711,
      "learning_rate": 2.2301349325337332e-08,
      "loss": 1.3541,
      "step": 120
    },
    {
      "epoch": 11.523809523809524,
      "grad_norm": 61.97282791137695,
      "learning_rate": 2.2488755622188907e-08,
      "loss": 1.3786,
      "step": 121
    },
    {
      "epoch": 11.619047619047619,
      "grad_norm": 66.15354919433594,
      "learning_rate": 2.267616191904048e-08,
      "loss": 1.3387,
      "step": 122
    },
    {
      "epoch": 11.714285714285714,
      "grad_norm": 66.2543716430664,
      "learning_rate": 2.2863568215892055e-08,
      "loss": 1.3319,
      "step": 123
    },
    {
      "epoch": 11.80952380952381,
      "grad_norm": 68.32968139648438,
      "learning_rate": 2.305097451274363e-08,
      "loss": 1.3416,
      "step": 124
    },
    {
      "epoch": 11.904761904761905,
      "grad_norm": 50.14171600341797,
      "learning_rate": 2.32383808095952e-08,
      "loss": 1.2865,
      "step": 125
    },
    {
      "epoch": 12.0,
      "grad_norm": 60.39354705810547,
      "learning_rate": 2.342578710644678e-08,
      "loss": 1.3506,
      "step": 126
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6912701386286999,
      "eval_f1": 0.75344105326152,
      "eval_loss": 0.6636635065078735,
      "eval_precision": 0.7002224694104561,
      "eval_recall": 0.8154145077720207,
      "eval_roc_auc": 0.7216375935521013,
      "eval_runtime": 1.251,
      "eval_samples_per_second": 2133.499,
      "eval_steps_per_second": 4.796,
      "step": 126
    },
    {
      "epoch": 12.095238095238095,
      "grad_norm": 68.8990249633789,
      "learning_rate": 2.3613193403298353e-08,
      "loss": 1.3287,
      "step": 127
    },
    {
      "epoch": 12.19047619047619,
      "grad_norm": 55.19594192504883,
      "learning_rate": 2.3800599700149927e-08,
      "loss": 1.3361,
      "step": 128
    },
    {
      "epoch": 12.285714285714286,
      "grad_norm": 61.52813720703125,
      "learning_rate": 2.39880059970015e-08,
      "loss": 1.3434,
      "step": 129
    },
    {
      "epoch": 12.380952380952381,
      "grad_norm": 61.1476936340332,
      "learning_rate": 2.4175412293853072e-08,
      "loss": 1.2941,
      "step": 130
    },
    {
      "epoch": 12.476190476190476,
      "grad_norm": 45.287376403808594,
      "learning_rate": 2.436281859070465e-08,
      "loss": 1.3373,
      "step": 131
    },
    {
      "epoch": 12.571428571428571,
      "grad_norm": 67.01655578613281,
      "learning_rate": 2.4550224887556224e-08,
      "loss": 1.4024,
      "step": 132
    },
    {
      "epoch": 12.666666666666666,
      "grad_norm": 51.15004348754883,
      "learning_rate": 2.4737631184407795e-08,
      "loss": 1.386,
      "step": 133
    },
    {
      "epoch": 12.761904761904763,
      "grad_norm": 54.940879821777344,
      "learning_rate": 2.492503748125937e-08,
      "loss": 1.3333,
      "step": 134
    },
    {
      "epoch": 12.857142857142858,
      "grad_norm": 50.37574768066406,
      "learning_rate": 2.5112443778110947e-08,
      "loss": 1.3527,
      "step": 135
    },
    {
      "epoch": 12.952380952380953,
      "grad_norm": 46.545135498046875,
      "learning_rate": 2.529985007496252e-08,
      "loss": 1.2964,
      "step": 136
    },
    {
      "epoch": 12.952380952380953,
      "eval_accuracy": 0.689396777819408,
      "eval_f1": 0.7503763926528154,
      "eval_loss": 0.6601486802101135,
      "eval_precision": 0.7011817670230726,
      "eval_recall": 0.8069948186528497,
      "eval_roc_auc": 0.7214781232009211,
      "eval_runtime": 1.2976,
      "eval_samples_per_second": 2056.814,
      "eval_steps_per_second": 4.624,
      "step": 136
    },
    {
      "epoch": 13.047619047619047,
      "grad_norm": 63.312862396240234,
      "learning_rate": 2.5487256371814093e-08,
      "loss": 1.3792,
      "step": 137
    },
    {
      "epoch": 13.142857142857142,
      "grad_norm": 77.56578826904297,
      "learning_rate": 2.567466266866567e-08,
      "loss": 1.4125,
      "step": 138
    },
    {
      "epoch": 13.238095238095237,
      "grad_norm": 33.94204330444336,
      "learning_rate": 2.5862068965517245e-08,
      "loss": 1.3047,
      "step": 139
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 58.11079025268555,
      "learning_rate": 2.6049475262368816e-08,
      "loss": 1.24,
      "step": 140
    },
    {
      "epoch": 13.428571428571429,
      "grad_norm": 47.870269775390625,
      "learning_rate": 2.623688155922039e-08,
      "loss": 1.3368,
      "step": 141
    },
    {
      "epoch": 13.523809523809524,
      "grad_norm": 35.01408767700195,
      "learning_rate": 2.6424287856071968e-08,
      "loss": 1.3086,
      "step": 142
    },
    {
      "epoch": 13.619047619047619,
      "grad_norm": 54.038848876953125,
      "learning_rate": 2.661169415292354e-08,
      "loss": 1.3006,
      "step": 143
    },
    {
      "epoch": 13.714285714285714,
      "grad_norm": 49.3018684387207,
      "learning_rate": 2.6799100449775113e-08,
      "loss": 1.316,
      "step": 144
    },
    {
      "epoch": 13.80952380952381,
      "grad_norm": 51.7872428894043,
      "learning_rate": 2.698650674662669e-08,
      "loss": 1.3577,
      "step": 145
    },
    {
      "epoch": 13.904761904761905,
      "grad_norm": 45.14494705200195,
      "learning_rate": 2.7173913043478262e-08,
      "loss": 1.3951,
      "step": 146
    },
    {
      "epoch": 14.0,
      "grad_norm": 39.679683685302734,
      "learning_rate": 2.7361319340329836e-08,
      "loss": 1.318,
      "step": 147
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6908954664668415,
      "eval_f1": 0.7493163172288059,
      "eval_loss": 0.656489908695221,
      "eval_precision": 0.7057813394390383,
      "eval_recall": 0.7985751295336787,
      "eval_roc_auc": 0.721395221646517,
      "eval_runtime": 1.1331,
      "eval_samples_per_second": 2355.495,
      "eval_steps_per_second": 5.295,
      "step": 147
    },
    {
      "epoch": 14.095238095238095,
      "grad_norm": 49.89531707763672,
      "learning_rate": 2.7548725637181414e-08,
      "loss": 1.3619,
      "step": 148
    },
    {
      "epoch": 14.19047619047619,
      "grad_norm": 41.59170150756836,
      "learning_rate": 2.7736131934032985e-08,
      "loss": 1.2874,
      "step": 149
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": 48.24529266357422,
      "learning_rate": 2.792353823088456e-08,
      "loss": 1.3604,
      "step": 150
    },
    {
      "epoch": 14.380952380952381,
      "grad_norm": 38.974098205566406,
      "learning_rate": 2.811094452773613e-08,
      "loss": 1.3147,
      "step": 151
    },
    {
      "epoch": 14.476190476190476,
      "grad_norm": 44.127227783203125,
      "learning_rate": 2.8298350824587708e-08,
      "loss": 1.3018,
      "step": 152
    },
    {
      "epoch": 14.571428571428571,
      "grad_norm": 27.71172332763672,
      "learning_rate": 2.8485757121439282e-08,
      "loss": 1.3879,
      "step": 153
    },
    {
      "epoch": 14.666666666666666,
      "grad_norm": 40.47721862792969,
      "learning_rate": 2.8673163418290853e-08,
      "loss": 1.3635,
      "step": 154
    },
    {
      "epoch": 14.761904761904763,
      "grad_norm": 33.5676155090332,
      "learning_rate": 2.886056971514243e-08,
      "loss": 1.2736,
      "step": 155
    },
    {
      "epoch": 14.857142857142858,
      "grad_norm": 39.84068298339844,
      "learning_rate": 2.9047976011994005e-08,
      "loss": 1.3315,
      "step": 156
    },
    {
      "epoch": 14.952380952380953,
      "grad_norm": 42.16352844238281,
      "learning_rate": 2.9235382308845576e-08,
      "loss": 1.3555,
      "step": 157
    },
    {
      "epoch": 14.952380952380953,
      "eval_accuracy": 0.692394155114275,
      "eval_f1": 0.7485451761102604,
      "eval_loss": 0.6533787846565247,
      "eval_precision": 0.7100522951772226,
      "eval_recall": 0.7914507772020726,
      "eval_roc_auc": 0.7213140472078295,
      "eval_runtime": 1.1034,
      "eval_samples_per_second": 2418.914,
      "eval_steps_per_second": 5.438,
      "step": 157
    },
    {
      "epoch": 15.047619047619047,
      "grad_norm": 39.9268798828125,
      "learning_rate": 2.942278860569715e-08,
      "loss": 1.3337,
      "step": 158
    },
    {
      "epoch": 15.142857142857142,
      "grad_norm": 55.53904724121094,
      "learning_rate": 2.9610194902548728e-08,
      "loss": 1.3204,
      "step": 159
    },
    {
      "epoch": 15.238095238095237,
      "grad_norm": 25.50678825378418,
      "learning_rate": 2.9797601199400302e-08,
      "loss": 1.2897,
      "step": 160
    },
    {
      "epoch": 15.333333333333334,
      "grad_norm": 35.85453414916992,
      "learning_rate": 2.9985007496251873e-08,
      "loss": 1.3681,
      "step": 161
    },
    {
      "epoch": 15.428571428571429,
      "grad_norm": 37.41948699951172,
      "learning_rate": 3.0172413793103454e-08,
      "loss": 1.3464,
      "step": 162
    },
    {
      "epoch": 15.523809523809524,
      "grad_norm": 37.18633270263672,
      "learning_rate": 3.035982008995502e-08,
      "loss": 1.3106,
      "step": 163
    },
    {
      "epoch": 15.619047619047619,
      "grad_norm": 21.66717529296875,
      "learning_rate": 3.0547226386806596e-08,
      "loss": 1.3386,
      "step": 164
    },
    {
      "epoch": 15.714285714285714,
      "grad_norm": 24.518701553344727,
      "learning_rate": 3.073463268365818e-08,
      "loss": 1.3262,
      "step": 165
    },
    {
      "epoch": 15.80952380952381,
      "grad_norm": 23.171676635742188,
      "learning_rate": 3.0922038980509745e-08,
      "loss": 1.2522,
      "step": 166
    },
    {
      "epoch": 15.904761904761905,
      "grad_norm": 22.128908157348633,
      "learning_rate": 3.110944527736132e-08,
      "loss": 1.3117,
      "step": 167
    },
    {
      "epoch": 16.0,
      "grad_norm": 30.98495864868164,
      "learning_rate": 3.1296851574212894e-08,
      "loss": 1.2886,
      "step": 168
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6875234170101161,
      "eval_f1": 0.7413151364764268,
      "eval_loss": 0.6505224704742432,
      "eval_precision": 0.7113095238095238,
      "eval_recall": 0.7739637305699482,
      "eval_roc_auc": 0.7212000575705239,
      "eval_runtime": 1.2094,
      "eval_samples_per_second": 2206.913,
      "eval_steps_per_second": 4.961,
      "step": 168
    },
    {
      "epoch": 16.095238095238095,
      "grad_norm": 21.845766067504883,
      "learning_rate": 3.148425787106447e-08,
      "loss": 1.2719,
      "step": 169
    },
    {
      "epoch": 16.19047619047619,
      "grad_norm": 14.632295608520508,
      "learning_rate": 3.167166416791604e-08,
      "loss": 1.3311,
      "step": 170
    },
    {
      "epoch": 16.285714285714285,
      "grad_norm": 16.54092788696289,
      "learning_rate": 3.185907046476762e-08,
      "loss": 1.3079,
      "step": 171
    },
    {
      "epoch": 16.38095238095238,
      "grad_norm": 28.98506736755371,
      "learning_rate": 3.204647676161919e-08,
      "loss": 1.2828,
      "step": 172
    },
    {
      "epoch": 16.476190476190474,
      "grad_norm": 22.530902862548828,
      "learning_rate": 3.2233883058470765e-08,
      "loss": 1.3206,
      "step": 173
    },
    {
      "epoch": 16.571428571428573,
      "grad_norm": 39.143653869628906,
      "learning_rate": 3.242128935532234e-08,
      "loss": 1.2731,
      "step": 174
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 42.83742141723633,
      "learning_rate": 3.2608695652173914e-08,
      "loss": 1.3593,
      "step": 175
    },
    {
      "epoch": 16.761904761904763,
      "grad_norm": 24.04018211364746,
      "learning_rate": 3.279610194902549e-08,
      "loss": 1.3516,
      "step": 176
    },
    {
      "epoch": 16.857142857142858,
      "grad_norm": 27.83500862121582,
      "learning_rate": 3.298350824587706e-08,
      "loss": 1.3182,
      "step": 177
    },
    {
      "epoch": 16.952380952380953,
      "grad_norm": 16.24953842163086,
      "learning_rate": 3.317091454272864e-08,
      "loss": 1.2646,
      "step": 178
    },
    {
      "epoch": 16.952380952380953,
      "eval_accuracy": 0.6860247283626827,
      "eval_f1": 0.7377972465581978,
      "eval_loss": 0.6483314633369446,
      "eval_precision": 0.7136803874092009,
      "eval_recall": 0.7636010362694301,
      "eval_roc_auc": 0.7211528497409326,
      "eval_runtime": 1.2241,
      "eval_samples_per_second": 2180.389,
      "eval_steps_per_second": 4.902,
      "step": 178
    },
    {
      "epoch": 17.047619047619047,
      "grad_norm": 18.79022789001465,
      "learning_rate": 3.335832083958021e-08,
      "loss": 1.2849,
      "step": 179
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 18.4663143157959,
      "learning_rate": 3.3545727136431786e-08,
      "loss": 1.3546,
      "step": 180
    },
    {
      "epoch": 17.238095238095237,
      "grad_norm": 22.815502166748047,
      "learning_rate": 3.373313343328336e-08,
      "loss": 1.3085,
      "step": 181
    },
    {
      "epoch": 17.333333333333332,
      "grad_norm": 23.185617446899414,
      "learning_rate": 3.3920539730134934e-08,
      "loss": 1.3128,
      "step": 182
    },
    {
      "epoch": 17.428571428571427,
      "grad_norm": 22.799545288085938,
      "learning_rate": 3.410794602698651e-08,
      "loss": 1.3307,
      "step": 183
    },
    {
      "epoch": 17.523809523809526,
      "grad_norm": 13.100482940673828,
      "learning_rate": 3.429535232383808e-08,
      "loss": 1.2815,
      "step": 184
    },
    {
      "epoch": 17.61904761904762,
      "grad_norm": 14.181185722351074,
      "learning_rate": 3.448275862068965e-08,
      "loss": 1.3237,
      "step": 185
    },
    {
      "epoch": 17.714285714285715,
      "grad_norm": 25.63428497314453,
      "learning_rate": 3.467016491754123e-08,
      "loss": 1.375,
      "step": 186
    },
    {
      "epoch": 17.80952380952381,
      "grad_norm": 14.835464477539062,
      "learning_rate": 3.4857571214392806e-08,
      "loss": 1.3288,
      "step": 187
    },
    {
      "epoch": 17.904761904761905,
      "grad_norm": 20.395618438720703,
      "learning_rate": 3.504497751124438e-08,
      "loss": 1.3057,
      "step": 188
    },
    {
      "epoch": 18.0,
      "grad_norm": 11.902831077575684,
      "learning_rate": 3.5232383808095955e-08,
      "loss": 1.2873,
      "step": 189
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6849007118771075,
      "eval_f1": 0.7347839798170924,
      "eval_loss": 0.6463755965232849,
      "eval_precision": 0.7160417947141979,
      "eval_recall": 0.7545336787564767,
      "eval_roc_auc": 0.7211459412780656,
      "eval_runtime": 1.2122,
      "eval_samples_per_second": 2201.763,
      "eval_steps_per_second": 4.95,
      "step": 189
    },
    {
      "epoch": 18.095238095238095,
      "grad_norm": 19.942028045654297,
      "learning_rate": 3.541979010494753e-08,
      "loss": 1.3181,
      "step": 190
    },
    {
      "epoch": 18.19047619047619,
      "grad_norm": 11.15404224395752,
      "learning_rate": 3.5607196401799103e-08,
      "loss": 1.187,
      "step": 191
    },
    {
      "epoch": 18.285714285714285,
      "grad_norm": 13.81049919128418,
      "learning_rate": 3.579460269865068e-08,
      "loss": 1.3189,
      "step": 192
    },
    {
      "epoch": 18.38095238095238,
      "grad_norm": 12.814773559570312,
      "learning_rate": 3.598200899550225e-08,
      "loss": 1.3253,
      "step": 193
    },
    {
      "epoch": 18.476190476190474,
      "grad_norm": 17.163558959960938,
      "learning_rate": 3.6169415292353826e-08,
      "loss": 1.3064,
      "step": 194
    },
    {
      "epoch": 18.571428571428573,
      "grad_norm": 16.707319259643555,
      "learning_rate": 3.6356821589205394e-08,
      "loss": 1.3173,
      "step": 195
    },
    {
      "epoch": 18.666666666666668,
      "grad_norm": 16.393266677856445,
      "learning_rate": 3.6544227886056975e-08,
      "loss": 1.3359,
      "step": 196
    },
    {
      "epoch": 18.761904761904763,
      "grad_norm": 11.493505477905273,
      "learning_rate": 3.673163418290855e-08,
      "loss": 1.3048,
      "step": 197
    },
    {
      "epoch": 18.857142857142858,
      "grad_norm": 11.934215545654297,
      "learning_rate": 3.691904047976012e-08,
      "loss": 1.2677,
      "step": 198
    },
    {
      "epoch": 18.952380952380953,
      "grad_norm": 15.357635498046875,
      "learning_rate": 3.71064467766117e-08,
      "loss": 1.2959,
      "step": 199
    },
    {
      "epoch": 18.952380952380953,
      "eval_accuracy": 0.6830273510678156,
      "eval_f1": 0.7321089297023432,
      "eval_loss": 0.6447131037712097,
      "eval_precision": 0.7162329615861215,
      "eval_recall": 0.7487046632124352,
      "eval_roc_auc": 0.7214608520437537,
      "eval_runtime": 1.1709,
      "eval_samples_per_second": 2279.454,
      "eval_steps_per_second": 5.124,
      "step": 199
    },
    {
      "epoch": 19.047619047619047,
      "grad_norm": 14.120625495910645,
      "learning_rate": 3.729385307346327e-08,
      "loss": 1.3118,
      "step": 200
    },
    {
      "epoch": 19.142857142857142,
      "grad_norm": 17.531734466552734,
      "learning_rate": 3.748125937031484e-08,
      "loss": 1.3112,
      "step": 201
    },
    {
      "epoch": 19.238095238095237,
      "grad_norm": 16.632896423339844,
      "learning_rate": 3.766866566716642e-08,
      "loss": 1.3308,
      "step": 202
    },
    {
      "epoch": 19.333333333333332,
      "grad_norm": 12.964826583862305,
      "learning_rate": 3.7856071964017995e-08,
      "loss": 1.2656,
      "step": 203
    },
    {
      "epoch": 19.428571428571427,
      "grad_norm": 12.439407348632812,
      "learning_rate": 3.804347826086956e-08,
      "loss": 1.2727,
      "step": 204
    },
    {
      "epoch": 19.523809523809526,
      "grad_norm": 25.103260040283203,
      "learning_rate": 3.823088455772114e-08,
      "loss": 1.3474,
      "step": 205
    },
    {
      "epoch": 19.61904761904762,
      "grad_norm": 18.016218185424805,
      "learning_rate": 3.841829085457272e-08,
      "loss": 1.3193,
      "step": 206
    },
    {
      "epoch": 19.714285714285715,
      "grad_norm": 13.59919261932373,
      "learning_rate": 3.860569715142429e-08,
      "loss": 1.2927,
      "step": 207
    },
    {
      "epoch": 19.80952380952381,
      "grad_norm": 15.137652397155762,
      "learning_rate": 3.879310344827586e-08,
      "loss": 1.3174,
      "step": 208
    },
    {
      "epoch": 19.904761904761905,
      "grad_norm": 21.939130783081055,
      "learning_rate": 3.898050974512744e-08,
      "loss": 1.2291,
      "step": 209
    },
    {
      "epoch": 20.0,
      "grad_norm": 14.990110397338867,
      "learning_rate": 3.9167916041979016e-08,
      "loss": 1.3391,
      "step": 210
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6830273510678156,
      "eval_f1": 0.7315989847715736,
      "eval_loss": 0.6429556608200073,
      "eval_precision": 0.7170398009950248,
      "eval_recall": 0.7467616580310881,
      "eval_roc_auc": 0.7219881980426022,
      "eval_runtime": 1.1317,
      "eval_samples_per_second": 2358.443,
      "eval_steps_per_second": 5.302,
      "step": 210
    },
    {
      "epoch": 20.095238095238095,
      "grad_norm": 13.944443702697754,
      "learning_rate": 3.9355322338830584e-08,
      "loss": 1.348,
      "step": 211
    },
    {
      "epoch": 20.19047619047619,
      "grad_norm": 13.567949295043945,
      "learning_rate": 3.954272863568216e-08,
      "loss": 1.3795,
      "step": 212
    },
    {
      "epoch": 20.285714285714285,
      "grad_norm": 14.929570198059082,
      "learning_rate": 3.973013493253374e-08,
      "loss": 1.2863,
      "step": 213
    },
    {
      "epoch": 20.38095238095238,
      "grad_norm": 11.928326606750488,
      "learning_rate": 3.9917541229385307e-08,
      "loss": 1.2453,
      "step": 214
    },
    {
      "epoch": 20.476190476190474,
      "grad_norm": 16.773330688476562,
      "learning_rate": 4.010494752623688e-08,
      "loss": 1.3193,
      "step": 215
    },
    {
      "epoch": 20.571428571428573,
      "grad_norm": 16.72808265686035,
      "learning_rate": 4.029235382308846e-08,
      "loss": 1.3068,
      "step": 216
    },
    {
      "epoch": 20.666666666666668,
      "grad_norm": 13.820122718811035,
      "learning_rate": 4.047976011994003e-08,
      "loss": 1.2867,
      "step": 217
    },
    {
      "epoch": 20.761904761904763,
      "grad_norm": 11.65041732788086,
      "learning_rate": 4.0667166416791604e-08,
      "loss": 1.3137,
      "step": 218
    },
    {
      "epoch": 20.857142857142858,
      "grad_norm": 11.561596870422363,
      "learning_rate": 4.0854572713643185e-08,
      "loss": 1.2622,
      "step": 219
    },
    {
      "epoch": 20.952380952380953,
      "grad_norm": 11.974090576171875,
      "learning_rate": 4.104197901049475e-08,
      "loss": 1.2637,
      "step": 220
    },
    {
      "epoch": 20.952380952380953,
      "eval_accuracy": 0.6837766953915324,
      "eval_f1": 0.7325728770595691,
      "eval_loss": 0.6413893103599548,
      "eval_precision": 0.71712158808933,
      "eval_recall": 0.7487046632124352,
      "eval_roc_auc": 0.7224211283822682,
      "eval_runtime": 1.1887,
      "eval_samples_per_second": 2245.39,
      "eval_steps_per_second": 5.048,
      "step": 220
    },
    {
      "epoch": 21.047619047619047,
      "grad_norm": 15.292080879211426,
      "learning_rate": 4.122938530734633e-08,
      "loss": 1.3156,
      "step": 221
    },
    {
      "epoch": 21.142857142857142,
      "grad_norm": 15.198674201965332,
      "learning_rate": 4.14167916041979e-08,
      "loss": 1.329,
      "step": 222
    },
    {
      "epoch": 21.238095238095237,
      "grad_norm": 18.44618034362793,
      "learning_rate": 4.1604197901049476e-08,
      "loss": 1.254,
      "step": 223
    },
    {
      "epoch": 21.333333333333332,
      "grad_norm": 25.041452407836914,
      "learning_rate": 4.179160419790105e-08,
      "loss": 1.2522,
      "step": 224
    },
    {
      "epoch": 21.428571428571427,
      "grad_norm": 16.231889724731445,
      "learning_rate": 4.1979010494752624e-08,
      "loss": 1.2916,
      "step": 225
    },
    {
      "epoch": 21.523809523809526,
      "grad_norm": 13.011600494384766,
      "learning_rate": 4.2166416791604205e-08,
      "loss": 1.3383,
      "step": 226
    },
    {
      "epoch": 21.61904761904762,
      "grad_norm": 13.868030548095703,
      "learning_rate": 4.235382308845577e-08,
      "loss": 1.3488,
      "step": 227
    },
    {
      "epoch": 21.714285714285715,
      "grad_norm": 21.20703887939453,
      "learning_rate": 4.254122938530735e-08,
      "loss": 1.2101,
      "step": 228
    },
    {
      "epoch": 21.80952380952381,
      "grad_norm": 15.180831909179688,
      "learning_rate": 4.272863568215893e-08,
      "loss": 1.3172,
      "step": 229
    },
    {
      "epoch": 21.904761904761905,
      "grad_norm": 14.952122688293457,
      "learning_rate": 4.2916041979010496e-08,
      "loss": 1.3397,
      "step": 230
    },
    {
      "epoch": 22.0,
      "grad_norm": 15.182557106018066,
      "learning_rate": 4.310344827586207e-08,
      "loss": 1.3171,
      "step": 231
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6830273510678156,
      "eval_f1": 0.7319391634980988,
      "eval_loss": 0.6397904753684998,
      "eval_precision": 0.716501240694789,
      "eval_recall": 0.7480569948186528,
      "eval_roc_auc": 0.722874496257916,
      "eval_runtime": 1.4672,
      "eval_samples_per_second": 1819.081,
      "eval_steps_per_second": 4.089,
      "step": 231
    },
    {
      "epoch": 22.095238095238095,
      "grad_norm": 19.74833106994629,
      "learning_rate": 4.3290854572713645e-08,
      "loss": 1.3591,
      "step": 232
    },
    {
      "epoch": 22.19047619047619,
      "grad_norm": 12.066325187683105,
      "learning_rate": 4.347826086956522e-08,
      "loss": 1.3181,
      "step": 233
    },
    {
      "epoch": 22.285714285714285,
      "grad_norm": 10.577549934387207,
      "learning_rate": 4.366566716641679e-08,
      "loss": 1.2408,
      "step": 234
    },
    {
      "epoch": 22.38095238095238,
      "grad_norm": 17.58034896850586,
      "learning_rate": 4.385307346326837e-08,
      "loss": 1.3027,
      "step": 235
    },
    {
      "epoch": 22.476190476190474,
      "grad_norm": 12.278691291809082,
      "learning_rate": 4.404047976011994e-08,
      "loss": 1.2905,
      "step": 236
    },
    {
      "epoch": 22.571428571428573,
      "grad_norm": 13.021527290344238,
      "learning_rate": 4.4227886056971516e-08,
      "loss": 1.2831,
      "step": 237
    },
    {
      "epoch": 22.666666666666668,
      "grad_norm": 13.841450691223145,
      "learning_rate": 4.441529235382309e-08,
      "loss": 1.3171,
      "step": 238
    },
    {
      "epoch": 22.761904761904763,
      "grad_norm": 11.652981758117676,
      "learning_rate": 4.4602698650674665e-08,
      "loss": 1.3003,
      "step": 239
    },
    {
      "epoch": 22.857142857142858,
      "grad_norm": 15.65576171875,
      "learning_rate": 4.479010494752624e-08,
      "loss": 1.3055,
      "step": 240
    },
    {
      "epoch": 22.952380952380953,
      "grad_norm": 11.00838565826416,
      "learning_rate": 4.4977511244377814e-08,
      "loss": 1.2486,
      "step": 241
    },
    {
      "epoch": 22.952380952380953,
      "eval_accuracy": 0.6826526789059573,
      "eval_f1": 0.7313669521091024,
      "eval_loss": 0.6382837295532227,
      "eval_precision": 0.7165941578620261,
      "eval_recall": 0.7467616580310881,
      "eval_roc_auc": 0.7232576280944156,
      "eval_runtime": 1.2133,
      "eval_samples_per_second": 2199.758,
      "eval_steps_per_second": 4.945,
      "step": 241
    },
    {
      "epoch": 23.047619047619047,
      "grad_norm": 13.720239639282227,
      "learning_rate": 4.516491754122939e-08,
      "loss": 1.2912,
      "step": 242
    },
    {
      "epoch": 23.142857142857142,
      "grad_norm": 14.70983600616455,
      "learning_rate": 4.535232383808096e-08,
      "loss": 1.2933,
      "step": 243
    },
    {
      "epoch": 23.238095238095237,
      "grad_norm": 14.259865760803223,
      "learning_rate": 4.5539730134932537e-08,
      "loss": 1.3368,
      "step": 244
    },
    {
      "epoch": 23.333333333333332,
      "grad_norm": 11.778265953063965,
      "learning_rate": 4.572713643178411e-08,
      "loss": 1.2697,
      "step": 245
    },
    {
      "epoch": 23.428571428571427,
      "grad_norm": 12.168619155883789,
      "learning_rate": 4.5914542728635685e-08,
      "loss": 1.3096,
      "step": 246
    },
    {
      "epoch": 23.523809523809526,
      "grad_norm": 15.17798137664795,
      "learning_rate": 4.610194902548726e-08,
      "loss": 1.2814,
      "step": 247
    },
    {
      "epoch": 23.61904761904762,
      "grad_norm": 13.276535034179688,
      "learning_rate": 4.6289355322338834e-08,
      "loss": 1.2687,
      "step": 248
    },
    {
      "epoch": 23.714285714285715,
      "grad_norm": 16.98518180847168,
      "learning_rate": 4.64767616191904e-08,
      "loss": 1.2899,
      "step": 249
    },
    {
      "epoch": 23.80952380952381,
      "grad_norm": 11.48328971862793,
      "learning_rate": 4.666416791604198e-08,
      "loss": 1.2549,
      "step": 250
    },
    {
      "epoch": 23.904761904761905,
      "grad_norm": 11.340226173400879,
      "learning_rate": 4.685157421289356e-08,
      "loss": 1.2531,
      "step": 251
    },
    {
      "epoch": 24.0,
      "grad_norm": 10.424729347229004,
      "learning_rate": 4.703898050974513e-08,
      "loss": 1.3059,
      "step": 252
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.683402023229674,
      "eval_f1": 0.7323408299018055,
      "eval_loss": 0.63660728931427,
      "eval_precision": 0.7166769993800371,
      "eval_recall": 0.7487046632124352,
      "eval_roc_auc": 0.7237892918825561,
      "eval_runtime": 1.3641,
      "eval_samples_per_second": 1956.631,
      "eval_steps_per_second": 4.399,
      "step": 252
    },
    {
      "epoch": 24.095238095238095,
      "grad_norm": 13.842662811279297,
      "learning_rate": 4.7226386806596706e-08,
      "loss": 1.2593,
      "step": 253
    },
    {
      "epoch": 24.19047619047619,
      "grad_norm": 10.153083801269531,
      "learning_rate": 4.741379310344828e-08,
      "loss": 1.2947,
      "step": 254
    },
    {
      "epoch": 24.285714285714285,
      "grad_norm": 13.279069900512695,
      "learning_rate": 4.7601199400299854e-08,
      "loss": 1.3419,
      "step": 255
    },
    {
      "epoch": 24.38095238095238,
      "grad_norm": 10.15274715423584,
      "learning_rate": 4.778860569715143e-08,
      "loss": 1.2321,
      "step": 256
    },
    {
      "epoch": 24.476190476190474,
      "grad_norm": 11.35965347290039,
      "learning_rate": 4.7976011994003e-08,
      "loss": 1.3073,
      "step": 257
    },
    {
      "epoch": 24.571428571428573,
      "grad_norm": 14.660529136657715,
      "learning_rate": 4.816341829085458e-08,
      "loss": 1.332,
      "step": 258
    },
    {
      "epoch": 24.666666666666668,
      "grad_norm": 14.38634204864502,
      "learning_rate": 4.8350824587706145e-08,
      "loss": 1.3127,
      "step": 259
    },
    {
      "epoch": 24.761904761904763,
      "grad_norm": 11.082036972045898,
      "learning_rate": 4.8538230884557726e-08,
      "loss": 1.2769,
      "step": 260
    },
    {
      "epoch": 24.857142857142858,
      "grad_norm": 13.869253158569336,
      "learning_rate": 4.87256371814093e-08,
      "loss": 1.3329,
      "step": 261
    },
    {
      "epoch": 24.952380952380953,
      "grad_norm": 10.266282081604004,
      "learning_rate": 4.8913043478260875e-08,
      "loss": 1.2312,
      "step": 262
    },
    {
      "epoch": 24.952380952380953,
      "eval_accuracy": 0.6845260397152492,
      "eval_f1": 0.7333755541481951,
      "eval_loss": 0.6350855827331543,
      "eval_precision": 0.7174721189591078,
      "eval_recall": 0.75,
      "eval_roc_auc": 0.7242173287276914,
      "eval_runtime": 1.2504,
      "eval_samples_per_second": 2134.503,
      "eval_steps_per_second": 4.798,
      "step": 262
    },
    {
      "epoch": 25.047619047619047,
      "grad_norm": 13.72497844696045,
      "learning_rate": 4.910044977511245e-08,
      "loss": 1.3405,
      "step": 263
    },
    {
      "epoch": 25.142857142857142,
      "grad_norm": 12.01894760131836,
      "learning_rate": 4.928785607196402e-08,
      "loss": 1.3145,
      "step": 264
    },
    {
      "epoch": 25.238095238095237,
      "grad_norm": 10.575872421264648,
      "learning_rate": 4.947526236881559e-08,
      "loss": 1.2693,
      "step": 265
    },
    {
      "epoch": 25.333333333333332,
      "grad_norm": 18.743549346923828,
      "learning_rate": 4.9662668665667165e-08,
      "loss": 1.2669,
      "step": 266
    },
    {
      "epoch": 25.428571428571427,
      "grad_norm": 8.173626899719238,
      "learning_rate": 4.985007496251874e-08,
      "loss": 1.2132,
      "step": 267
    },
    {
      "epoch": 25.523809523809526,
      "grad_norm": 10.303065299987793,
      "learning_rate": 5.003748125937032e-08,
      "loss": 1.2327,
      "step": 268
    },
    {
      "epoch": 25.61904761904762,
      "grad_norm": 11.798093795776367,
      "learning_rate": 5.0224887556221895e-08,
      "loss": 1.2299,
      "step": 269
    },
    {
      "epoch": 25.714285714285715,
      "grad_norm": 10.679269790649414,
      "learning_rate": 5.041229385307347e-08,
      "loss": 1.3056,
      "step": 270
    },
    {
      "epoch": 25.80952380952381,
      "grad_norm": 11.035752296447754,
      "learning_rate": 5.059970014992504e-08,
      "loss": 1.2988,
      "step": 271
    },
    {
      "epoch": 25.904761904761905,
      "grad_norm": 9.624054908752441,
      "learning_rate": 5.078710644677661e-08,
      "loss": 1.2798,
      "step": 272
    },
    {
      "epoch": 26.0,
      "grad_norm": 11.053035736083984,
      "learning_rate": 5.0974512743628186e-08,
      "loss": 1.2709,
      "step": 273
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6860247283626827,
      "eval_f1": 0.7346421785940469,
      "eval_loss": 0.6334132552146912,
      "eval_precision": 0.7187112763320942,
      "eval_recall": 0.7512953367875648,
      "eval_roc_auc": 0.7246937248128957,
      "eval_runtime": 1.1191,
      "eval_samples_per_second": 2384.911,
      "eval_steps_per_second": 5.361,
      "step": 273
    },
    {
      "epoch": 26.095238095238095,
      "grad_norm": 11.396424293518066,
      "learning_rate": 5.116191904047976e-08,
      "loss": 1.268,
      "step": 274
    },
    {
      "epoch": 26.19047619047619,
      "grad_norm": 14.449176788330078,
      "learning_rate": 5.134932533733134e-08,
      "loss": 1.3123,
      "step": 275
    },
    {
      "epoch": 26.285714285714285,
      "grad_norm": 13.00364875793457,
      "learning_rate": 5.1536731634182915e-08,
      "loss": 1.2875,
      "step": 276
    },
    {
      "epoch": 26.38095238095238,
      "grad_norm": 9.805450439453125,
      "learning_rate": 5.172413793103449e-08,
      "loss": 1.2843,
      "step": 277
    },
    {
      "epoch": 26.476190476190474,
      "grad_norm": 11.89764404296875,
      "learning_rate": 5.191154422788606e-08,
      "loss": 1.3268,
      "step": 278
    },
    {
      "epoch": 26.571428571428573,
      "grad_norm": 10.767585754394531,
      "learning_rate": 5.209895052473763e-08,
      "loss": 1.298,
      "step": 279
    },
    {
      "epoch": 26.666666666666668,
      "grad_norm": 13.058414459228516,
      "learning_rate": 5.2286356821589206e-08,
      "loss": 1.2781,
      "step": 280
    },
    {
      "epoch": 26.761904761904763,
      "grad_norm": 17.320100784301758,
      "learning_rate": 5.247376311844078e-08,
      "loss": 1.2988,
      "step": 281
    },
    {
      "epoch": 26.857142857142858,
      "grad_norm": 17.9411563873291,
      "learning_rate": 5.266116941529236e-08,
      "loss": 1.287,
      "step": 282
    },
    {
      "epoch": 26.952380952380953,
      "grad_norm": 10.667381286621094,
      "learning_rate": 5.2848575712143936e-08,
      "loss": 1.2671,
      "step": 283
    },
    {
      "epoch": 26.952380952380953,
      "eval_accuracy": 0.6867740726863995,
      "eval_f1": 0.7351077313054499,
      "eval_loss": 0.6319016218185425,
      "eval_precision": 0.7196029776674938,
      "eval_recall": 0.7512953367875648,
      "eval_roc_auc": 0.7251082325849165,
      "eval_runtime": 1.1627,
      "eval_samples_per_second": 2295.462,
      "eval_steps_per_second": 5.16,
      "step": 283
    },
    {
      "epoch": 27.047619047619047,
      "grad_norm": 12.966008186340332,
      "learning_rate": 5.30359820089955e-08,
      "loss": 1.3341,
      "step": 284
    },
    {
      "epoch": 27.142857142857142,
      "grad_norm": 12.528854370117188,
      "learning_rate": 5.322338830584708e-08,
      "loss": 1.2418,
      "step": 285
    },
    {
      "epoch": 27.238095238095237,
      "grad_norm": 13.622454643249512,
      "learning_rate": 5.341079460269865e-08,
      "loss": 1.3488,
      "step": 286
    },
    {
      "epoch": 27.333333333333332,
      "grad_norm": 16.258296966552734,
      "learning_rate": 5.3598200899550226e-08,
      "loss": 1.235,
      "step": 287
    },
    {
      "epoch": 27.428571428571427,
      "grad_norm": 9.838101387023926,
      "learning_rate": 5.3785607196401794e-08,
      "loss": 1.2687,
      "step": 288
    },
    {
      "epoch": 27.523809523809526,
      "grad_norm": 9.929924964904785,
      "learning_rate": 5.397301349325338e-08,
      "loss": 1.2249,
      "step": 289
    },
    {
      "epoch": 27.61904761904762,
      "grad_norm": 9.11926555633545,
      "learning_rate": 5.416041979010495e-08,
      "loss": 1.3112,
      "step": 290
    },
    {
      "epoch": 27.714285714285715,
      "grad_norm": 12.670866012573242,
      "learning_rate": 5.4347826086956524e-08,
      "loss": 1.3038,
      "step": 291
    },
    {
      "epoch": 27.80952380952381,
      "grad_norm": 16.6510066986084,
      "learning_rate": 5.45352323838081e-08,
      "loss": 1.3411,
      "step": 292
    },
    {
      "epoch": 27.904761904761905,
      "grad_norm": 10.850798606872559,
      "learning_rate": 5.472263868065967e-08,
      "loss": 1.3158,
      "step": 293
    },
    {
      "epoch": 28.0,
      "grad_norm": 14.082104682922363,
      "learning_rate": 5.491004497751124e-08,
      "loss": 1.321,
      "step": 294
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6878980891719745,
      "eval_f1": 0.7368088467614534,
      "eval_loss": 0.6302856802940369,
      "eval_precision": 0.7193090684762492,
      "eval_recall": 0.7551813471502591,
      "eval_roc_auc": 0.7255808865860679,
      "eval_runtime": 1.1704,
      "eval_samples_per_second": 2280.384,
      "eval_steps_per_second": 5.126,
      "step": 294
    },
    {
      "epoch": 28.095238095238095,
      "grad_norm": 12.710125923156738,
      "learning_rate": 5.509745127436283e-08,
      "loss": 1.3059,
      "step": 295
    },
    {
      "epoch": 28.19047619047619,
      "grad_norm": 12.972898483276367,
      "learning_rate": 5.52848575712144e-08,
      "loss": 1.2712,
      "step": 296
    },
    {
      "epoch": 28.285714285714285,
      "grad_norm": 10.665593147277832,
      "learning_rate": 5.547226386806597e-08,
      "loss": 1.2697,
      "step": 297
    },
    {
      "epoch": 28.38095238095238,
      "grad_norm": 10.437527656555176,
      "learning_rate": 5.5659670164917544e-08,
      "loss": 1.2156,
      "step": 298
    },
    {
      "epoch": 28.476190476190474,
      "grad_norm": 12.75148868560791,
      "learning_rate": 5.584707646176912e-08,
      "loss": 1.2533,
      "step": 299
    },
    {
      "epoch": 28.571428571428573,
      "grad_norm": 16.342567443847656,
      "learning_rate": 5.603448275862069e-08,
      "loss": 1.3416,
      "step": 300
    },
    {
      "epoch": 28.666666666666668,
      "grad_norm": 14.161049842834473,
      "learning_rate": 5.622188905547226e-08,
      "loss": 1.2625,
      "step": 301
    },
    {
      "epoch": 28.761904761904763,
      "grad_norm": 12.947792053222656,
      "learning_rate": 5.640929535232385e-08,
      "loss": 1.2499,
      "step": 302
    },
    {
      "epoch": 28.857142857142858,
      "grad_norm": 12.01698112487793,
      "learning_rate": 5.6596701649175416e-08,
      "loss": 1.3089,
      "step": 303
    },
    {
      "epoch": 28.952380952380953,
      "grad_norm": 9.873224258422852,
      "learning_rate": 5.678410794602699e-08,
      "loss": 1.2842,
      "step": 304
    },
    {
      "epoch": 28.952380952380953,
      "eval_accuracy": 0.6897714499812664,
      "eval_f1": 0.7391304347826086,
      "eval_loss": 0.6289286017417908,
      "eval_precision": 0.7196319018404908,
      "eval_recall": 0.7597150259067358,
      "eval_roc_auc": 0.7259801381692574,
      "eval_runtime": 1.2466,
      "eval_samples_per_second": 2141.024,
      "eval_steps_per_second": 4.813,
      "step": 304
    },
    {
      "epoch": 29.047619047619047,
      "grad_norm": 10.019207954406738,
      "learning_rate": 5.6971514242878564e-08,
      "loss": 1.2933,
      "step": 305
    },
    {
      "epoch": 29.142857142857142,
      "grad_norm": 12.953960418701172,
      "learning_rate": 5.715892053973014e-08,
      "loss": 1.2779,
      "step": 306
    },
    {
      "epoch": 29.238095238095237,
      "grad_norm": 11.685044288635254,
      "learning_rate": 5.7346326836581706e-08,
      "loss": 1.2449,
      "step": 307
    },
    {
      "epoch": 29.333333333333332,
      "grad_norm": 15.267889022827148,
      "learning_rate": 5.753373313343328e-08,
      "loss": 1.2968,
      "step": 308
    },
    {
      "epoch": 29.428571428571427,
      "grad_norm": 8.864400863647461,
      "learning_rate": 5.772113943028486e-08,
      "loss": 1.3027,
      "step": 309
    },
    {
      "epoch": 29.523809523809526,
      "grad_norm": 27.022933959960938,
      "learning_rate": 5.7908545727136436e-08,
      "loss": 1.3559,
      "step": 310
    },
    {
      "epoch": 29.61904761904762,
      "grad_norm": 12.17434024810791,
      "learning_rate": 5.809595202398801e-08,
      "loss": 1.3358,
      "step": 311
    },
    {
      "epoch": 29.714285714285715,
      "grad_norm": 12.910569190979004,
      "learning_rate": 5.8283358320839585e-08,
      "loss": 1.2667,
      "step": 312
    },
    {
      "epoch": 29.80952380952381,
      "grad_norm": 9.485709190368652,
      "learning_rate": 5.847076461769115e-08,
      "loss": 1.2922,
      "step": 313
    },
    {
      "epoch": 29.904761904761905,
      "grad_norm": 10.216976165771484,
      "learning_rate": 5.8658170914542727e-08,
      "loss": 1.2439,
      "step": 314
    },
    {
      "epoch": 30.0,
      "grad_norm": 13.940195083618164,
      "learning_rate": 5.88455772113943e-08,
      "loss": 1.2695,
      "step": 315
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6916448107905583,
      "eval_f1": 0.7411135577225543,
      "eval_loss": 0.6274736523628235,
      "eval_precision": 0.7204892966360856,
      "eval_recall": 0.7629533678756477,
      "eval_roc_auc": 0.7264208405296488,
      "eval_runtime": 1.324,
      "eval_samples_per_second": 2015.884,
      "eval_steps_per_second": 4.532,
      "step": 315
    },
    {
      "epoch": 30.095238095238095,
      "grad_norm": 12.867043495178223,
      "learning_rate": 5.903298350824588e-08,
      "loss": 1.2534,
      "step": 316
    },
    {
      "epoch": 30.19047619047619,
      "grad_norm": 10.215372085571289,
      "learning_rate": 5.9220389805097456e-08,
      "loss": 1.2551,
      "step": 317
    },
    {
      "epoch": 30.285714285714285,
      "grad_norm": 12.636347770690918,
      "learning_rate": 5.940779610194903e-08,
      "loss": 1.3279,
      "step": 318
    },
    {
      "epoch": 30.38095238095238,
      "grad_norm": 15.71722412109375,
      "learning_rate": 5.9595202398800605e-08,
      "loss": 1.2522,
      "step": 319
    },
    {
      "epoch": 30.476190476190474,
      "grad_norm": 13.175235748291016,
      "learning_rate": 5.978260869565218e-08,
      "loss": 1.2719,
      "step": 320
    },
    {
      "epoch": 30.571428571428573,
      "grad_norm": 12.036287307739258,
      "learning_rate": 5.997001499250375e-08,
      "loss": 1.2943,
      "step": 321
    },
    {
      "epoch": 30.666666666666668,
      "grad_norm": 10.251227378845215,
      "learning_rate": 6.015742128935533e-08,
      "loss": 1.2898,
      "step": 322
    },
    {
      "epoch": 30.761904761904763,
      "grad_norm": 10.132257461547852,
      "learning_rate": 6.034482758620691e-08,
      "loss": 1.2605,
      "step": 323
    },
    {
      "epoch": 30.857142857142858,
      "grad_norm": 13.103528022766113,
      "learning_rate": 6.053223388305848e-08,
      "loss": 1.2523,
      "step": 324
    },
    {
      "epoch": 30.952380952380953,
      "grad_norm": 10.725114822387695,
      "learning_rate": 6.071964017991004e-08,
      "loss": 1.2916,
      "step": 325
    },
    {
      "epoch": 30.952380952380953,
      "eval_accuracy": 0.6905207943049831,
      "eval_f1": 0.740251572327044,
      "eval_loss": 0.6261727809906006,
      "eval_precision": 0.719437652811736,
      "eval_recall": 0.7623056994818653,
      "eval_roc_auc": 0.7266635002878526,
      "eval_runtime": 1.1207,
      "eval_samples_per_second": 2381.643,
      "eval_steps_per_second": 5.354,
      "step": 325
    },
    {
      "epoch": 31.047619047619047,
      "grad_norm": 16.890737533569336,
      "learning_rate": 6.090704647676163e-08,
      "loss": 1.1974,
      "step": 326
    },
    {
      "epoch": 31.142857142857142,
      "grad_norm": 10.4693603515625,
      "learning_rate": 6.109445277361319e-08,
      "loss": 1.2511,
      "step": 327
    },
    {
      "epoch": 31.238095238095237,
      "grad_norm": 11.922187805175781,
      "learning_rate": 6.128185907046476e-08,
      "loss": 1.2439,
      "step": 328
    },
    {
      "epoch": 31.333333333333332,
      "grad_norm": 9.826422691345215,
      "learning_rate": 6.146926536731635e-08,
      "loss": 1.2901,
      "step": 329
    },
    {
      "epoch": 31.428571428571427,
      "grad_norm": 10.615519523620605,
      "learning_rate": 6.165667166416792e-08,
      "loss": 1.2171,
      "step": 330
    },
    {
      "epoch": 31.523809523809526,
      "grad_norm": 10.314847946166992,
      "learning_rate": 6.184407796101949e-08,
      "loss": 1.3078,
      "step": 331
    },
    {
      "epoch": 31.61904761904762,
      "grad_norm": 9.442974090576172,
      "learning_rate": 6.203148425787107e-08,
      "loss": 1.2235,
      "step": 332
    },
    {
      "epoch": 31.714285714285715,
      "grad_norm": 13.98375129699707,
      "learning_rate": 6.221889055472264e-08,
      "loss": 1.2872,
      "step": 333
    },
    {
      "epoch": 31.80952380952381,
      "grad_norm": 9.29346752166748,
      "learning_rate": 6.240629685157421e-08,
      "loss": 1.2985,
      "step": 334
    },
    {
      "epoch": 31.904761904761905,
      "grad_norm": 12.216511726379395,
      "learning_rate": 6.259370314842579e-08,
      "loss": 1.3001,
      "step": 335
    },
    {
      "epoch": 32.0,
      "grad_norm": 11.124615669250488,
      "learning_rate": 6.278110944527737e-08,
      "loss": 1.2794,
      "step": 336
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.6920194829524167,
      "eval_f1": 0.7411838790931989,
      "eval_loss": 0.6247643828392029,
      "eval_precision": 0.7212009803921569,
      "eval_recall": 0.7623056994818653,
      "eval_roc_auc": 0.727045192861255,
      "eval_runtime": 1.1903,
      "eval_samples_per_second": 2242.322,
      "eval_steps_per_second": 5.041,
      "step": 336
    },
    {
      "epoch": 32.095238095238095,
      "grad_norm": 17.693710327148438,
      "learning_rate": 6.296851574212894e-08,
      "loss": 1.2293,
      "step": 337
    },
    {
      "epoch": 32.19047619047619,
      "grad_norm": 13.685829162597656,
      "learning_rate": 6.315592203898052e-08,
      "loss": 1.2249,
      "step": 338
    },
    {
      "epoch": 32.285714285714285,
      "grad_norm": 12.157681465148926,
      "learning_rate": 6.334332833583208e-08,
      "loss": 1.3495,
      "step": 339
    },
    {
      "epoch": 32.38095238095238,
      "grad_norm": 19.257387161254883,
      "learning_rate": 6.353073463268365e-08,
      "loss": 1.2029,
      "step": 340
    },
    {
      "epoch": 32.476190476190474,
      "grad_norm": 14.897787094116211,
      "learning_rate": 6.371814092953523e-08,
      "loss": 1.2478,
      "step": 341
    },
    {
      "epoch": 32.57142857142857,
      "grad_norm": 8.744516372680664,
      "learning_rate": 6.39055472263868e-08,
      "loss": 1.2386,
      "step": 342
    },
    {
      "epoch": 32.666666666666664,
      "grad_norm": 11.148335456848145,
      "learning_rate": 6.409295352323838e-08,
      "loss": 1.2786,
      "step": 343
    },
    {
      "epoch": 32.76190476190476,
      "grad_norm": 10.281587600708008,
      "learning_rate": 6.428035982008996e-08,
      "loss": 1.3163,
      "step": 344
    },
    {
      "epoch": 32.857142857142854,
      "grad_norm": 10.477128982543945,
      "learning_rate": 6.446776611694153e-08,
      "loss": 1.2902,
      "step": 345
    },
    {
      "epoch": 32.95238095238095,
      "grad_norm": 7.760868072509766,
      "learning_rate": 6.465517241379311e-08,
      "loss": 1.2286,
      "step": 346
    },
    {
      "epoch": 32.95238095238095,
      "eval_accuracy": 0.692394155114275,
      "eval_f1": 0.7414173228346457,
      "eval_loss": 0.6235555410385132,
      "eval_precision": 0.7216431637032495,
      "eval_recall": 0.7623056994818653,
      "eval_roc_auc": 0.7272826712723086,
      "eval_runtime": 1.4945,
      "eval_samples_per_second": 1785.937,
      "eval_steps_per_second": 4.015,
      "step": 346
    },
    {
      "epoch": 33.04761904761905,
      "grad_norm": 10.051183700561523,
      "learning_rate": 6.484257871064468e-08,
      "loss": 1.285,
      "step": 347
    },
    {
      "epoch": 33.142857142857146,
      "grad_norm": 8.26945972442627,
      "learning_rate": 6.502998500749625e-08,
      "loss": 1.2618,
      "step": 348
    },
    {
      "epoch": 33.23809523809524,
      "grad_norm": 10.043110847473145,
      "learning_rate": 6.521739130434783e-08,
      "loss": 1.3071,
      "step": 349
    },
    {
      "epoch": 33.333333333333336,
      "grad_norm": 8.486072540283203,
      "learning_rate": 6.540479760119941e-08,
      "loss": 1.3003,
      "step": 350
    },
    {
      "epoch": 33.42857142857143,
      "grad_norm": 9.369478225708008,
      "learning_rate": 6.559220389805098e-08,
      "loss": 1.2488,
      "step": 351
    },
    {
      "epoch": 33.523809523809526,
      "grad_norm": 19.558025360107422,
      "learning_rate": 6.577961019490256e-08,
      "loss": 1.2199,
      "step": 352
    },
    {
      "epoch": 33.61904761904762,
      "grad_norm": 9.55553150177002,
      "learning_rate": 6.596701649175413e-08,
      "loss": 1.2679,
      "step": 353
    },
    {
      "epoch": 33.714285714285715,
      "grad_norm": 12.629579544067383,
      "learning_rate": 6.615442278860569e-08,
      "loss": 1.2395,
      "step": 354
    },
    {
      "epoch": 33.80952380952381,
      "grad_norm": 14.873830795288086,
      "learning_rate": 6.634182908545727e-08,
      "loss": 1.293,
      "step": 355
    },
    {
      "epoch": 33.904761904761905,
      "grad_norm": 9.722416877746582,
      "learning_rate": 6.652923538230886e-08,
      "loss": 1.2764,
      "step": 356
    },
    {
      "epoch": 34.0,
      "grad_norm": 9.184341430664062,
      "learning_rate": 6.671664167916042e-08,
      "loss": 1.2784,
      "step": 357
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6931434994379918,
      "eval_f1": 0.7422096317280453,
      "eval_loss": 0.6222566366195679,
      "eval_precision": 0.7219840783833436,
      "eval_recall": 0.7636010362694301,
      "eval_roc_auc": 0.7276188831318364,
      "eval_runtime": 1.1766,
      "eval_samples_per_second": 2268.491,
      "eval_steps_per_second": 5.1,
      "step": 357
    },
    {
      "epoch": 34.095238095238095,
      "grad_norm": 17.952123641967773,
      "learning_rate": 6.6904047976012e-08,
      "loss": 1.2041,
      "step": 358
    },
    {
      "epoch": 34.19047619047619,
      "grad_norm": 9.577611923217773,
      "learning_rate": 6.709145427286357e-08,
      "loss": 1.2329,
      "step": 359
    },
    {
      "epoch": 34.285714285714285,
      "grad_norm": 10.438279151916504,
      "learning_rate": 6.727886056971514e-08,
      "loss": 1.2664,
      "step": 360
    },
    {
      "epoch": 34.38095238095238,
      "grad_norm": 11.91615104675293,
      "learning_rate": 6.746626686656672e-08,
      "loss": 1.2901,
      "step": 361
    },
    {
      "epoch": 34.476190476190474,
      "grad_norm": 11.50903034210205,
      "learning_rate": 6.765367316341829e-08,
      "loss": 1.3196,
      "step": 362
    },
    {
      "epoch": 34.57142857142857,
      "grad_norm": 21.127635955810547,
      "learning_rate": 6.784107946026987e-08,
      "loss": 1.246,
      "step": 363
    },
    {
      "epoch": 34.666666666666664,
      "grad_norm": 8.733270645141602,
      "learning_rate": 6.802848575712145e-08,
      "loss": 1.2559,
      "step": 364
    },
    {
      "epoch": 34.76190476190476,
      "grad_norm": 12.190812110900879,
      "learning_rate": 6.821589205397302e-08,
      "loss": 1.2671,
      "step": 365
    },
    {
      "epoch": 34.857142857142854,
      "grad_norm": 9.869928359985352,
      "learning_rate": 6.840329835082459e-08,
      "loss": 1.2123,
      "step": 366
    },
    {
      "epoch": 34.95238095238095,
      "grad_norm": 14.302867889404297,
      "learning_rate": 6.859070464767617e-08,
      "loss": 1.2832,
      "step": 367
    },
    {
      "epoch": 34.95238095238095,
      "eval_accuracy": 0.6942675159235668,
      "eval_f1": 0.7442006269592476,
      "eval_loss": 0.6210344433784485,
      "eval_precision": 0.721142162818955,
      "eval_recall": 0.7687823834196891,
      "eval_roc_auc": 0.7280549798503166,
      "eval_runtime": 1.3197,
      "eval_samples_per_second": 2022.409,
      "eval_steps_per_second": 4.546,
      "step": 367
    },
    {
      "epoch": 35.04761904761905,
      "grad_norm": 10.143964767456055,
      "learning_rate": 6.877811094452773e-08,
      "loss": 1.2314,
      "step": 368
    },
    {
      "epoch": 35.142857142857146,
      "grad_norm": 13.52642822265625,
      "learning_rate": 6.89655172413793e-08,
      "loss": 1.3133,
      "step": 369
    },
    {
      "epoch": 35.23809523809524,
      "grad_norm": 8.791267395019531,
      "learning_rate": 6.91529235382309e-08,
      "loss": 1.2008,
      "step": 370
    },
    {
      "epoch": 35.333333333333336,
      "grad_norm": 8.982850074768066,
      "learning_rate": 6.934032983508246e-08,
      "loss": 1.221,
      "step": 371
    },
    {
      "epoch": 35.42857142857143,
      "grad_norm": 6.191827774047852,
      "learning_rate": 6.952773613193403e-08,
      "loss": 1.2471,
      "step": 372
    },
    {
      "epoch": 35.523809523809526,
      "grad_norm": 8.668267250061035,
      "learning_rate": 6.971514242878561e-08,
      "loss": 1.2409,
      "step": 373
    },
    {
      "epoch": 35.61904761904762,
      "grad_norm": 11.444587707519531,
      "learning_rate": 6.990254872563718e-08,
      "loss": 1.2306,
      "step": 374
    },
    {
      "epoch": 35.714285714285715,
      "grad_norm": 8.564613342285156,
      "learning_rate": 7.008995502248876e-08,
      "loss": 1.2479,
      "step": 375
    },
    {
      "epoch": 35.80952380952381,
      "grad_norm": 17.54589080810547,
      "learning_rate": 7.027736131934034e-08,
      "loss": 1.2814,
      "step": 376
    },
    {
      "epoch": 35.904761904761905,
      "grad_norm": 10.60114574432373,
      "learning_rate": 7.046476761619191e-08,
      "loss": 1.3122,
      "step": 377
    },
    {
      "epoch": 36.0,
      "grad_norm": 13.735464096069336,
      "learning_rate": 7.065217391304348e-08,
      "loss": 1.257,
      "step": 378
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.6946421880854252,
      "eval_f1": 0.7445941711062363,
      "eval_loss": 0.619743287563324,
      "eval_precision": 0.7213114754098361,
      "eval_recall": 0.7694300518134715,
      "eval_roc_auc": 0.7284668969487623,
      "eval_runtime": 1.1237,
      "eval_samples_per_second": 2375.245,
      "eval_steps_per_second": 5.34,
      "step": 378
    },
    {
      "epoch": 36.095238095238095,
      "grad_norm": 11.545016288757324,
      "learning_rate": 7.083958020989506e-08,
      "loss": 1.2738,
      "step": 379
    },
    {
      "epoch": 36.19047619047619,
      "grad_norm": 11.023893356323242,
      "learning_rate": 7.102698650674663e-08,
      "loss": 1.2481,
      "step": 380
    },
    {
      "epoch": 36.285714285714285,
      "grad_norm": 10.620939254760742,
      "learning_rate": 7.121439280359821e-08,
      "loss": 1.1878,
      "step": 381
    },
    {
      "epoch": 36.38095238095238,
      "grad_norm": 9.518024444580078,
      "learning_rate": 7.140179910044977e-08,
      "loss": 1.2586,
      "step": 382
    },
    {
      "epoch": 36.476190476190474,
      "grad_norm": 9.962079048156738,
      "learning_rate": 7.158920539730136e-08,
      "loss": 1.3146,
      "step": 383
    },
    {
      "epoch": 36.57142857142857,
      "grad_norm": 7.642823219299316,
      "learning_rate": 7.177661169415294e-08,
      "loss": 1.2584,
      "step": 384
    },
    {
      "epoch": 36.666666666666664,
      "grad_norm": 6.657567977905273,
      "learning_rate": 7.19640179910045e-08,
      "loss": 1.2405,
      "step": 385
    },
    {
      "epoch": 36.76190476190476,
      "grad_norm": 8.174885749816895,
      "learning_rate": 7.215142428785607e-08,
      "loss": 1.2213,
      "step": 386
    },
    {
      "epoch": 36.857142857142854,
      "grad_norm": 9.345281600952148,
      "learning_rate": 7.233883058470765e-08,
      "loss": 1.2349,
      "step": 387
    },
    {
      "epoch": 36.95238095238095,
      "grad_norm": 10.576764106750488,
      "learning_rate": 7.252623688155922e-08,
      "loss": 1.2673,
      "step": 388
    },
    {
      "epoch": 36.95238095238095,
      "eval_accuracy": 0.695391532409142,
      "eval_f1": 0.7453805198872534,
      "eval_loss": 0.6185930371284485,
      "eval_precision": 0.7216494845360825,
      "eval_recall": 0.7707253886010362,
      "eval_roc_auc": 0.7288293033966609,
      "eval_runtime": 1.1526,
      "eval_samples_per_second": 2315.706,
      "eval_steps_per_second": 5.206,
      "step": 388
    },
    {
      "epoch": 37.04761904761905,
      "grad_norm": 13.148157119750977,
      "learning_rate": 7.271364317841079e-08,
      "loss": 1.3245,
      "step": 389
    },
    {
      "epoch": 37.142857142857146,
      "grad_norm": 16.42124366760254,
      "learning_rate": 7.290104947526238e-08,
      "loss": 1.248,
      "step": 390
    },
    {
      "epoch": 37.23809523809524,
      "grad_norm": 10.700825691223145,
      "learning_rate": 7.308845577211395e-08,
      "loss": 1.278,
      "step": 391
    },
    {
      "epoch": 37.333333333333336,
      "grad_norm": 6.934027671813965,
      "learning_rate": 7.327586206896552e-08,
      "loss": 1.1932,
      "step": 392
    },
    {
      "epoch": 37.42857142857143,
      "grad_norm": 7.132205009460449,
      "learning_rate": 7.34632683658171e-08,
      "loss": 1.1896,
      "step": 393
    },
    {
      "epoch": 37.523809523809526,
      "grad_norm": 9.692222595214844,
      "learning_rate": 7.365067466266867e-08,
      "loss": 1.2578,
      "step": 394
    },
    {
      "epoch": 37.61904761904762,
      "grad_norm": 8.961304664611816,
      "learning_rate": 7.383808095952023e-08,
      "loss": 1.269,
      "step": 395
    },
    {
      "epoch": 37.714285714285715,
      "grad_norm": 17.51494598388672,
      "learning_rate": 7.402548725637182e-08,
      "loss": 1.2132,
      "step": 396
    },
    {
      "epoch": 37.80952380952381,
      "grad_norm": 10.411418914794922,
      "learning_rate": 7.42128935532234e-08,
      "loss": 1.2914,
      "step": 397
    },
    {
      "epoch": 37.904761904761905,
      "grad_norm": 32.3235969543457,
      "learning_rate": 7.440029985007496e-08,
      "loss": 1.2852,
      "step": 398
    },
    {
      "epoch": 38.0,
      "grad_norm": 11.350420951843262,
      "learning_rate": 7.458770614692654e-08,
      "loss": 1.2991,
      "step": 399
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.6927688272761334,
      "eval_f1": 0.74375,
      "eval_loss": 0.6172738671302795,
      "eval_precision": 0.7185990338164251,
      "eval_recall": 0.7707253886010362,
      "eval_roc_auc": 0.7293100172711571,
      "eval_runtime": 1.1692,
      "eval_samples_per_second": 2282.832,
      "eval_steps_per_second": 5.132,
      "step": 399
    },
    {
      "epoch": 38.095238095238095,
      "grad_norm": 10.146343231201172,
      "learning_rate": 7.477511244377811e-08,
      "loss": 1.2156,
      "step": 400
    },
    {
      "epoch": 38.19047619047619,
      "grad_norm": 9.178386688232422,
      "learning_rate": 7.496251874062968e-08,
      "loss": 1.3147,
      "step": 401
    },
    {
      "epoch": 38.285714285714285,
      "grad_norm": 16.39377212524414,
      "learning_rate": 7.514992503748126e-08,
      "loss": 1.2434,
      "step": 402
    },
    {
      "epoch": 38.38095238095238,
      "grad_norm": 8.165606498718262,
      "learning_rate": 7.533733133433284e-08,
      "loss": 1.241,
      "step": 403
    },
    {
      "epoch": 38.476190476190474,
      "grad_norm": 7.81613826751709,
      "learning_rate": 7.552473763118441e-08,
      "loss": 1.2205,
      "step": 404
    },
    {
      "epoch": 38.57142857142857,
      "grad_norm": 12.5527982711792,
      "learning_rate": 7.571214392803599e-08,
      "loss": 1.2375,
      "step": 405
    },
    {
      "epoch": 38.666666666666664,
      "grad_norm": 13.193964004516602,
      "learning_rate": 7.589955022488756e-08,
      "loss": 1.3149,
      "step": 406
    },
    {
      "epoch": 38.76190476190476,
      "grad_norm": 11.886463165283203,
      "learning_rate": 7.608695652173913e-08,
      "loss": 1.2915,
      "step": 407
    },
    {
      "epoch": 38.857142857142854,
      "grad_norm": 8.637429237365723,
      "learning_rate": 7.627436281859071e-08,
      "loss": 1.2104,
      "step": 408
    },
    {
      "epoch": 38.95238095238095,
      "grad_norm": 8.065494537353516,
      "learning_rate": 7.646176911544228e-08,
      "loss": 1.2651,
      "step": 409
    },
    {
      "epoch": 38.95238095238095,
      "eval_accuracy": 0.6935181715998502,
      "eval_f1": 0.7445346658338539,
      "eval_loss": 0.6162141561508179,
      "eval_precision": 0.7189384800965019,
      "eval_recall": 0.772020725388601,
      "eval_roc_auc": 0.7296609096142775,
      "eval_runtime": 1.2038,
      "eval_samples_per_second": 2217.099,
      "eval_steps_per_second": 4.984,
      "step": 409
    },
    {
      "epoch": 39.04761904761905,
      "grad_norm": 10.647790908813477,
      "learning_rate": 7.664917541229386e-08,
      "loss": 1.1924,
      "step": 410
    },
    {
      "epoch": 39.142857142857146,
      "grad_norm": 8.407620429992676,
      "learning_rate": 7.683658170914544e-08,
      "loss": 1.2834,
      "step": 411
    },
    {
      "epoch": 39.23809523809524,
      "grad_norm": 10.034721374511719,
      "learning_rate": 7.7023988005997e-08,
      "loss": 1.2479,
      "step": 412
    },
    {
      "epoch": 39.333333333333336,
      "grad_norm": 8.923266410827637,
      "learning_rate": 7.721139430284859e-08,
      "loss": 1.2341,
      "step": 413
    },
    {
      "epoch": 39.42857142857143,
      "grad_norm": 8.994647979736328,
      "learning_rate": 7.739880059970015e-08,
      "loss": 1.2758,
      "step": 414
    },
    {
      "epoch": 39.523809523809526,
      "grad_norm": 7.205422401428223,
      "learning_rate": 7.758620689655172e-08,
      "loss": 1.2421,
      "step": 415
    },
    {
      "epoch": 39.61904761904762,
      "grad_norm": 7.557537078857422,
      "learning_rate": 7.77736131934033e-08,
      "loss": 1.2648,
      "step": 416
    },
    {
      "epoch": 39.714285714285715,
      "grad_norm": 13.540074348449707,
      "learning_rate": 7.796101949025488e-08,
      "loss": 1.2454,
      "step": 417
    },
    {
      "epoch": 39.80952380952381,
      "grad_norm": 7.045814037322998,
      "learning_rate": 7.814842578710645e-08,
      "loss": 1.2481,
      "step": 418
    },
    {
      "epoch": 39.904761904761905,
      "grad_norm": 12.278300285339355,
      "learning_rate": 7.833583208395803e-08,
      "loss": 1.2519,
      "step": 419
    },
    {
      "epoch": 40.0,
      "grad_norm": 8.270506858825684,
      "learning_rate": 7.85232383808096e-08,
      "loss": 1.2424,
      "step": 420
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.6938928437617085,
      "eval_f1": 0.7447672602311778,
      "eval_loss": 0.6150386929512024,
      "eval_precision": 0.7193723596861799,
      "eval_recall": 0.772020725388601,
      "eval_roc_auc": 0.7299176741508346,
      "eval_runtime": 1.376,
      "eval_samples_per_second": 1939.651,
      "eval_steps_per_second": 4.36,
      "step": 420
    },
    {
      "epoch": 40.095238095238095,
      "grad_norm": 9.389835357666016,
      "learning_rate": 7.871064467766117e-08,
      "loss": 1.2773,
      "step": 421
    },
    {
      "epoch": 40.19047619047619,
      "grad_norm": 17.68068504333496,
      "learning_rate": 7.889805097451275e-08,
      "loss": 1.2939,
      "step": 422
    },
    {
      "epoch": 40.285714285714285,
      "grad_norm": 10.948050498962402,
      "learning_rate": 7.908545727136432e-08,
      "loss": 1.247,
      "step": 423
    },
    {
      "epoch": 40.38095238095238,
      "grad_norm": 10.204144477844238,
      "learning_rate": 7.92728635682159e-08,
      "loss": 1.2583,
      "step": 424
    },
    {
      "epoch": 40.476190476190474,
      "grad_norm": 8.29542350769043,
      "learning_rate": 7.946026986506748e-08,
      "loss": 1.2417,
      "step": 425
    },
    {
      "epoch": 40.57142857142857,
      "grad_norm": 10.966816902160645,
      "learning_rate": 7.964767616191905e-08,
      "loss": 1.2749,
      "step": 426
    },
    {
      "epoch": 40.666666666666664,
      "grad_norm": 7.698588848114014,
      "learning_rate": 7.983508245877061e-08,
      "loss": 1.2242,
      "step": 427
    },
    {
      "epoch": 40.76190476190476,
      "grad_norm": 13.610939979553223,
      "learning_rate": 8.00224887556222e-08,
      "loss": 1.1539,
      "step": 428
    },
    {
      "epoch": 40.857142857142854,
      "grad_norm": 9.892770767211914,
      "learning_rate": 8.020989505247376e-08,
      "loss": 1.2967,
      "step": 429
    },
    {
      "epoch": 40.95238095238095,
      "grad_norm": 8.661589622497559,
      "learning_rate": 8.039730134932534e-08,
      "loss": 1.2472,
      "step": 430
    },
    {
      "epoch": 40.95238095238095,
      "eval_accuracy": 0.692394155114275,
      "eval_f1": 0.7431967469502658,
      "eval_loss": 0.6140232682228088,
      "eval_precision": 0.7186932849364791,
      "eval_recall": 0.7694300518134715,
      "eval_roc_auc": 0.730028209556707,
      "eval_runtime": 1.1592,
      "eval_samples_per_second": 2302.479,
      "eval_steps_per_second": 5.176,
      "step": 430
    },
    {
      "epoch": 41.04761904761905,
      "grad_norm": 9.2203950881958,
      "learning_rate": 8.058470764617692e-08,
      "loss": 1.2648,
      "step": 431
    },
    {
      "epoch": 41.142857142857146,
      "grad_norm": 6.482606887817383,
      "learning_rate": 8.077211394302849e-08,
      "loss": 1.2506,
      "step": 432
    },
    {
      "epoch": 41.23809523809524,
      "grad_norm": 8.282089233398438,
      "learning_rate": 8.095952023988006e-08,
      "loss": 1.2826,
      "step": 433
    },
    {
      "epoch": 41.333333333333336,
      "grad_norm": 8.352246284484863,
      "learning_rate": 8.114692653673164e-08,
      "loss": 1.2123,
      "step": 434
    },
    {
      "epoch": 41.42857142857143,
      "grad_norm": 9.01622486114502,
      "learning_rate": 8.133433283358321e-08,
      "loss": 1.284,
      "step": 435
    },
    {
      "epoch": 41.523809523809526,
      "grad_norm": 14.874005317687988,
      "learning_rate": 8.152173913043478e-08,
      "loss": 1.2559,
      "step": 436
    },
    {
      "epoch": 41.61904761904762,
      "grad_norm": 9.86267375946045,
      "learning_rate": 8.170914542728637e-08,
      "loss": 1.234,
      "step": 437
    },
    {
      "epoch": 41.714285714285715,
      "grad_norm": 15.416117668151855,
      "learning_rate": 8.189655172413794e-08,
      "loss": 1.21,
      "step": 438
    },
    {
      "epoch": 41.80952380952381,
      "grad_norm": 7.8476881980896,
      "learning_rate": 8.20839580209895e-08,
      "loss": 1.2181,
      "step": 439
    },
    {
      "epoch": 41.904761904761905,
      "grad_norm": 7.718698024749756,
      "learning_rate": 8.227136431784109e-08,
      "loss": 1.2211,
      "step": 440
    },
    {
      "epoch": 42.0,
      "grad_norm": 8.605914115905762,
      "learning_rate": 8.245877061469265e-08,
      "loss": 1.2357,
      "step": 441
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.6908954664668415,
      "eval_f1": 0.7414603572547791,
      "eval_loss": 0.6128706932067871,
      "eval_precision": 0.7182756527018822,
      "eval_recall": 0.7661917098445595,
      "eval_roc_auc": 0.7303862982153138,
      "eval_runtime": 1.1383,
      "eval_samples_per_second": 2344.629,
      "eval_steps_per_second": 5.271,
      "step": 441
    },
    {
      "epoch": 42.095238095238095,
      "grad_norm": 15.596901893615723,
      "learning_rate": 8.264617691154423e-08,
      "loss": 1.2439,
      "step": 442
    },
    {
      "epoch": 42.19047619047619,
      "grad_norm": 10.043680191040039,
      "learning_rate": 8.28335832083958e-08,
      "loss": 1.28,
      "step": 443
    },
    {
      "epoch": 42.285714285714285,
      "grad_norm": 6.956470012664795,
      "learning_rate": 8.302098950524738e-08,
      "loss": 1.1946,
      "step": 444
    },
    {
      "epoch": 42.38095238095238,
      "grad_norm": 8.87226676940918,
      "learning_rate": 8.320839580209895e-08,
      "loss": 1.2674,
      "step": 445
    },
    {
      "epoch": 42.476190476190474,
      "grad_norm": 8.562736511230469,
      "learning_rate": 8.339580209895053e-08,
      "loss": 1.2248,
      "step": 446
    },
    {
      "epoch": 42.57142857142857,
      "grad_norm": 12.94089126586914,
      "learning_rate": 8.35832083958021e-08,
      "loss": 1.2146,
      "step": 447
    },
    {
      "epoch": 42.666666666666664,
      "grad_norm": 5.351092338562012,
      "learning_rate": 8.377061469265368e-08,
      "loss": 1.1685,
      "step": 448
    },
    {
      "epoch": 42.76190476190476,
      "grad_norm": 12.503556251525879,
      "learning_rate": 8.395802098950525e-08,
      "loss": 1.1986,
      "step": 449
    },
    {
      "epoch": 42.857142857142854,
      "grad_norm": 13.914941787719727,
      "learning_rate": 8.414542728635682e-08,
      "loss": 1.2451,
      "step": 450
    },
    {
      "epoch": 42.95238095238095,
      "grad_norm": 7.735352993011475,
      "learning_rate": 8.433283358320841e-08,
      "loss": 1.2592,
      "step": 451
    },
    {
      "epoch": 42.95238095238095,
      "eval_accuracy": 0.6920194829524167,
      "eval_f1": 0.7440846824408468,
      "eval_loss": 0.6118565201759338,
      "eval_precision": 0.7164268585131894,
      "eval_recall": 0.7739637305699482,
      "eval_roc_auc": 0.7308465745538284,
      "eval_runtime": 1.1892,
      "eval_samples_per_second": 2244.425,
      "eval_steps_per_second": 5.046,
      "step": 451
    },
    {
      "epoch": 43.04761904761905,
      "grad_norm": 13.332247734069824,
      "learning_rate": 8.452023988005998e-08,
      "loss": 1.3321,
      "step": 452
    },
    {
      "epoch": 43.142857142857146,
      "grad_norm": 6.532809257507324,
      "learning_rate": 8.470764617691155e-08,
      "loss": 1.2256,
      "step": 453
    },
    {
      "epoch": 43.23809523809524,
      "grad_norm": 6.154987812042236,
      "learning_rate": 8.489505247376313e-08,
      "loss": 1.208,
      "step": 454
    },
    {
      "epoch": 43.333333333333336,
      "grad_norm": 6.660593032836914,
      "learning_rate": 8.50824587706147e-08,
      "loss": 1.2417,
      "step": 455
    },
    {
      "epoch": 43.42857142857143,
      "grad_norm": 10.340599060058594,
      "learning_rate": 8.526986506746626e-08,
      "loss": 1.2063,
      "step": 456
    },
    {
      "epoch": 43.523809523809526,
      "grad_norm": 24.725839614868164,
      "learning_rate": 8.545727136431786e-08,
      "loss": 1.2705,
      "step": 457
    },
    {
      "epoch": 43.61904761904762,
      "grad_norm": 10.683469772338867,
      "learning_rate": 8.564467766116942e-08,
      "loss": 1.2525,
      "step": 458
    },
    {
      "epoch": 43.714285714285715,
      "grad_norm": 9.80042839050293,
      "learning_rate": 8.583208395802099e-08,
      "loss": 1.2321,
      "step": 459
    },
    {
      "epoch": 43.80952380952381,
      "grad_norm": 9.710074424743652,
      "learning_rate": 8.601949025487257e-08,
      "loss": 1.2513,
      "step": 460
    },
    {
      "epoch": 43.904761904761905,
      "grad_norm": 8.437572479248047,
      "learning_rate": 8.620689655172414e-08,
      "loss": 1.198,
      "step": 461
    },
    {
      "epoch": 44.0,
      "grad_norm": 16.720048904418945,
      "learning_rate": 8.639430284857571e-08,
      "loss": 1.1771,
      "step": 462
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.6905207943049831,
      "eval_f1": 0.74235807860262,
      "eval_loss": 0.6107428669929504,
      "eval_precision": 0.7160048134777377,
      "eval_recall": 0.7707253886010362,
      "eval_roc_auc": 0.7310915371329879,
      "eval_runtime": 1.523,
      "eval_samples_per_second": 1752.463,
      "eval_steps_per_second": 3.94,
      "step": 462
    },
    {
      "epoch": 44.095238095238095,
      "grad_norm": 11.607662200927734,
      "learning_rate": 8.658170914542729e-08,
      "loss": 1.2554,
      "step": 463
    },
    {
      "epoch": 44.19047619047619,
      "grad_norm": 10.140999794006348,
      "learning_rate": 8.676911544227887e-08,
      "loss": 1.2278,
      "step": 464
    },
    {
      "epoch": 44.285714285714285,
      "grad_norm": 8.476537704467773,
      "learning_rate": 8.695652173913044e-08,
      "loss": 1.2808,
      "step": 465
    },
    {
      "epoch": 44.38095238095238,
      "grad_norm": 6.853167533874512,
      "learning_rate": 8.714392803598202e-08,
      "loss": 1.2115,
      "step": 466
    },
    {
      "epoch": 44.476190476190474,
      "grad_norm": 7.388545513153076,
      "learning_rate": 8.733133433283359e-08,
      "loss": 1.2131,
      "step": 467
    },
    {
      "epoch": 44.57142857142857,
      "grad_norm": 12.221652030944824,
      "learning_rate": 8.751874062968515e-08,
      "loss": 1.2563,
      "step": 468
    },
    {
      "epoch": 44.666666666666664,
      "grad_norm": 8.581650733947754,
      "learning_rate": 8.770614692653674e-08,
      "loss": 1.2217,
      "step": 469
    },
    {
      "epoch": 44.76190476190476,
      "grad_norm": 10.659028053283691,
      "learning_rate": 8.78935532233883e-08,
      "loss": 1.2165,
      "step": 470
    },
    {
      "epoch": 44.857142857142854,
      "grad_norm": 7.4702935218811035,
      "learning_rate": 8.808095952023988e-08,
      "loss": 1.2186,
      "step": 471
    },
    {
      "epoch": 44.95238095238095,
      "grad_norm": 8.66907024383545,
      "learning_rate": 8.826836581709146e-08,
      "loss": 1.2805,
      "step": 472
    },
    {
      "epoch": 44.95238095238095,
      "eval_accuracy": 0.6897714499812664,
      "eval_f1": 0.74125,
      "eval_loss": 0.6098491549491882,
      "eval_precision": 0.716183574879227,
      "eval_recall": 0.7681347150259067,
      "eval_roc_auc": 0.7313713298791019,
      "eval_runtime": 1.2302,
      "eval_samples_per_second": 2169.652,
      "eval_steps_per_second": 4.877,
      "step": 472
    },
    {
      "epoch": 45.04761904761905,
      "grad_norm": 8.637537002563477,
      "learning_rate": 8.845577211394303e-08,
      "loss": 1.2485,
      "step": 473
    },
    {
      "epoch": 45.142857142857146,
      "grad_norm": 7.344876289367676,
      "learning_rate": 8.86431784107946e-08,
      "loss": 1.2742,
      "step": 474
    },
    {
      "epoch": 45.23809523809524,
      "grad_norm": 12.491889953613281,
      "learning_rate": 8.883058470764618e-08,
      "loss": 1.2503,
      "step": 475
    },
    {
      "epoch": 45.333333333333336,
      "grad_norm": 8.886862754821777,
      "learning_rate": 8.901799100449775e-08,
      "loss": 1.2116,
      "step": 476
    },
    {
      "epoch": 45.42857142857143,
      "grad_norm": 11.657227516174316,
      "learning_rate": 8.920539730134933e-08,
      "loss": 1.2066,
      "step": 477
    },
    {
      "epoch": 45.523809523809526,
      "grad_norm": 6.646770000457764,
      "learning_rate": 8.939280359820091e-08,
      "loss": 1.178,
      "step": 478
    },
    {
      "epoch": 45.61904761904762,
      "grad_norm": 12.152413368225098,
      "learning_rate": 8.958020989505248e-08,
      "loss": 1.2413,
      "step": 479
    },
    {
      "epoch": 45.714285714285715,
      "grad_norm": 10.730239868164062,
      "learning_rate": 8.976761619190406e-08,
      "loss": 1.2969,
      "step": 480
    },
    {
      "epoch": 45.80952380952381,
      "grad_norm": 8.96199893951416,
      "learning_rate": 8.995502248875563e-08,
      "loss": 1.1655,
      "step": 481
    },
    {
      "epoch": 45.904761904761905,
      "grad_norm": 11.491959571838379,
      "learning_rate": 9.01424287856072e-08,
      "loss": 1.2569,
      "step": 482
    },
    {
      "epoch": 46.0,
      "grad_norm": 9.993547439575195,
      "learning_rate": 9.032983508245878e-08,
      "loss": 1.231,
      "step": 483
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.6912701386286999,
      "eval_f1": 0.7429819089207735,
      "eval_loss": 0.6088653802871704,
      "eval_precision": 0.7166064981949458,
      "eval_recall": 0.7713730569948186,
      "eval_roc_auc": 0.7315898100172712,
      "eval_runtime": 1.2193,
      "eval_samples_per_second": 2188.898,
      "eval_steps_per_second": 4.921,
      "step": 483
    },
    {
      "epoch": 46.095238095238095,
      "grad_norm": 12.399568557739258,
      "learning_rate": 9.051724137931036e-08,
      "loss": 1.2516,
      "step": 484
    },
    {
      "epoch": 46.19047619047619,
      "grad_norm": 5.617837429046631,
      "learning_rate": 9.070464767616192e-08,
      "loss": 1.2265,
      "step": 485
    },
    {
      "epoch": 46.285714285714285,
      "grad_norm": 10.155060768127441,
      "learning_rate": 9.08920539730135e-08,
      "loss": 1.2064,
      "step": 486
    },
    {
      "epoch": 46.38095238095238,
      "grad_norm": 7.8809661865234375,
      "learning_rate": 9.107946026986507e-08,
      "loss": 1.2029,
      "step": 487
    },
    {
      "epoch": 46.476190476190474,
      "grad_norm": 7.232656478881836,
      "learning_rate": 9.126686656671664e-08,
      "loss": 1.2328,
      "step": 488
    },
    {
      "epoch": 46.57142857142857,
      "grad_norm": 6.484198093414307,
      "learning_rate": 9.145427286356822e-08,
      "loss": 1.2244,
      "step": 489
    },
    {
      "epoch": 46.666666666666664,
      "grad_norm": 9.187188148498535,
      "learning_rate": 9.164167916041979e-08,
      "loss": 1.2345,
      "step": 490
    },
    {
      "epoch": 46.76190476190476,
      "grad_norm": 6.534236431121826,
      "learning_rate": 9.182908545727137e-08,
      "loss": 1.2412,
      "step": 491
    },
    {
      "epoch": 46.857142857142854,
      "grad_norm": 5.195504188537598,
      "learning_rate": 9.201649175412295e-08,
      "loss": 1.2278,
      "step": 492
    },
    {
      "epoch": 46.95238095238095,
      "grad_norm": 7.281925201416016,
      "learning_rate": 9.220389805097452e-08,
      "loss": 1.2285,
      "step": 493
    },
    {
      "epoch": 46.95238095238095,
      "eval_accuracy": 0.6916448107905583,
      "eval_f1": 0.7444892890406706,
      "eval_loss": 0.6079878807067871,
      "eval_precision": 0.7149672033392963,
      "eval_recall": 0.7765544041450777,
      "eval_roc_auc": 0.7319453080023028,
      "eval_runtime": 1.1504,
      "eval_samples_per_second": 2319.983,
      "eval_steps_per_second": 5.215,
      "step": 493
    },
    {
      "epoch": 47.04761904761905,
      "grad_norm": 14.828407287597656,
      "learning_rate": 9.239130434782609e-08,
      "loss": 1.2821,
      "step": 494
    },
    {
      "epoch": 47.142857142857146,
      "grad_norm": 6.8681182861328125,
      "learning_rate": 9.257871064467767e-08,
      "loss": 1.2301,
      "step": 495
    },
    {
      "epoch": 47.23809523809524,
      "grad_norm": 8.041788101196289,
      "learning_rate": 9.276611694152924e-08,
      "loss": 1.1844,
      "step": 496
    },
    {
      "epoch": 47.333333333333336,
      "grad_norm": 12.36368179321289,
      "learning_rate": 9.29535232383808e-08,
      "loss": 1.2044,
      "step": 497
    },
    {
      "epoch": 47.42857142857143,
      "grad_norm": 12.137744903564453,
      "learning_rate": 9.31409295352324e-08,
      "loss": 1.2119,
      "step": 498
    },
    {
      "epoch": 47.523809523809526,
      "grad_norm": 13.399957656860352,
      "learning_rate": 9.332833583208397e-08,
      "loss": 1.2211,
      "step": 499
    },
    {
      "epoch": 47.61904761904762,
      "grad_norm": 9.091277122497559,
      "learning_rate": 9.351574212893553e-08,
      "loss": 1.193,
      "step": 500
    },
    {
      "epoch": 47.714285714285715,
      "grad_norm": 10.781776428222656,
      "learning_rate": 9.370314842578711e-08,
      "loss": 1.301,
      "step": 501
    },
    {
      "epoch": 47.80952380952381,
      "grad_norm": 7.946430683135986,
      "learning_rate": 9.389055472263868e-08,
      "loss": 1.2459,
      "step": 502
    },
    {
      "epoch": 47.904761904761905,
      "grad_norm": 17.722209930419922,
      "learning_rate": 9.407796101949026e-08,
      "loss": 1.2716,
      "step": 503
    },
    {
      "epoch": 48.0,
      "grad_norm": 10.651185989379883,
      "learning_rate": 9.426536731634183e-08,
      "loss": 1.1701,
      "step": 504
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.6920194829524167,
      "eval_f1": 0.7447204968944099,
      "eval_loss": 0.6071056723594666,
      "eval_precision": 0.7153937947494033,
      "eval_recall": 0.7765544041450777,
      "eval_roc_auc": 0.7321105354058722,
      "eval_runtime": 1.1292,
      "eval_samples_per_second": 2363.553,
      "eval_steps_per_second": 5.313,
      "step": 504
    },
    {
      "epoch": 48.095238095238095,
      "grad_norm": 8.36655044555664,
      "learning_rate": 9.445277361319341e-08,
      "loss": 1.2032,
      "step": 505
    },
    {
      "epoch": 48.19047619047619,
      "grad_norm": 6.818087577819824,
      "learning_rate": 9.464017991004498e-08,
      "loss": 1.2746,
      "step": 506
    },
    {
      "epoch": 48.285714285714285,
      "grad_norm": 7.613657474517822,
      "learning_rate": 9.482758620689656e-08,
      "loss": 1.2157,
      "step": 507
    },
    {
      "epoch": 48.38095238095238,
      "grad_norm": 5.677556991577148,
      "learning_rate": 9.501499250374813e-08,
      "loss": 1.1771,
      "step": 508
    },
    {
      "epoch": 48.476190476190474,
      "grad_norm": 7.941476345062256,
      "learning_rate": 9.520239880059971e-08,
      "loss": 1.2557,
      "step": 509
    },
    {
      "epoch": 48.57142857142857,
      "grad_norm": 6.0007805824279785,
      "learning_rate": 9.538980509745128e-08,
      "loss": 1.2346,
      "step": 510
    },
    {
      "epoch": 48.666666666666664,
      "grad_norm": 8.160831451416016,
      "learning_rate": 9.557721139430286e-08,
      "loss": 1.2416,
      "step": 511
    },
    {
      "epoch": 48.76190476190476,
      "grad_norm": 11.385088920593262,
      "learning_rate": 9.576461769115442e-08,
      "loss": 1.182,
      "step": 512
    },
    {
      "epoch": 48.857142857142854,
      "grad_norm": 10.08471965789795,
      "learning_rate": 9.5952023988006e-08,
      "loss": 1.1992,
      "step": 513
    },
    {
      "epoch": 48.95238095238095,
      "grad_norm": 10.728684425354004,
      "learning_rate": 9.613943028485757e-08,
      "loss": 1.2379,
      "step": 514
    },
    {
      "epoch": 48.95238095238095,
      "eval_accuracy": 0.6890221056575496,
      "eval_f1": 0.7409488139825219,
      "eval_loss": 0.6063689589500427,
      "eval_precision": 0.7150602409638555,
      "eval_recall": 0.7687823834196891,
      "eval_roc_auc": 0.7321614853195165,
      "eval_runtime": 1.2051,
      "eval_samples_per_second": 2214.838,
      "eval_steps_per_second": 4.979,
      "step": 514
    },
    {
      "epoch": 49.04761904761905,
      "grad_norm": 11.368154525756836,
      "learning_rate": 9.632683658170915e-08,
      "loss": 1.2628,
      "step": 515
    },
    {
      "epoch": 49.142857142857146,
      "grad_norm": 6.578487873077393,
      "learning_rate": 9.651424287856072e-08,
      "loss": 1.2107,
      "step": 516
    },
    {
      "epoch": 49.23809523809524,
      "grad_norm": 7.267766952514648,
      "learning_rate": 9.670164917541229e-08,
      "loss": 1.1805,
      "step": 517
    },
    {
      "epoch": 49.333333333333336,
      "grad_norm": 6.106053829193115,
      "learning_rate": 9.688905547226388e-08,
      "loss": 1.2076,
      "step": 518
    },
    {
      "epoch": 49.42857142857143,
      "grad_norm": 10.81891918182373,
      "learning_rate": 9.707646176911545e-08,
      "loss": 1.244,
      "step": 519
    },
    {
      "epoch": 49.523809523809526,
      "grad_norm": 7.16576623916626,
      "learning_rate": 9.726386806596702e-08,
      "loss": 1.2448,
      "step": 520
    },
    {
      "epoch": 49.61904761904762,
      "grad_norm": 7.20371675491333,
      "learning_rate": 9.74512743628186e-08,
      "loss": 1.2267,
      "step": 521
    },
    {
      "epoch": 49.714285714285715,
      "grad_norm": 7.5791730880737305,
      "learning_rate": 9.763868065967017e-08,
      "loss": 1.1871,
      "step": 522
    },
    {
      "epoch": 49.80952380952381,
      "grad_norm": 6.973799705505371,
      "learning_rate": 9.782608695652175e-08,
      "loss": 1.2649,
      "step": 523
    },
    {
      "epoch": 49.904761904761905,
      "grad_norm": 7.447197914123535,
      "learning_rate": 9.801349325337332e-08,
      "loss": 1.2246,
      "step": 524
    },
    {
      "epoch": 50.0,
      "grad_norm": 7.1552734375,
      "learning_rate": 9.82008995502249e-08,
      "loss": 1.2511,
      "step": 525
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.6875234170101161,
      "eval_f1": 0.7408328154133001,
      "eval_loss": 0.6054435968399048,
      "eval_precision": 0.7120669056152927,
      "eval_recall": 0.772020725388601,
      "eval_roc_auc": 0.7324881980426021,
      "eval_runtime": 1.2114,
      "eval_samples_per_second": 2203.31,
      "eval_steps_per_second": 4.953,
      "step": 525
    },
    {
      "epoch": 50.095238095238095,
      "grad_norm": 9.0575590133667,
      "learning_rate": 9.838830584707645e-08,
      "loss": 1.2519,
      "step": 526
    },
    {
      "epoch": 50.19047619047619,
      "grad_norm": 5.1338324546813965,
      "learning_rate": 9.857571214392805e-08,
      "loss": 1.2371,
      "step": 527
    },
    {
      "epoch": 50.285714285714285,
      "grad_norm": 13.615644454956055,
      "learning_rate": 9.876311844077963e-08,
      "loss": 1.2103,
      "step": 528
    },
    {
      "epoch": 50.38095238095238,
      "grad_norm": 9.063286781311035,
      "learning_rate": 9.895052473763118e-08,
      "loss": 1.2197,
      "step": 529
    },
    {
      "epoch": 50.476190476190474,
      "grad_norm": 7.897054195404053,
      "learning_rate": 9.913793103448278e-08,
      "loss": 1.2472,
      "step": 530
    },
    {
      "epoch": 50.57142857142857,
      "grad_norm": 8.853591918945312,
      "learning_rate": 9.932533733133433e-08,
      "loss": 1.2366,
      "step": 531
    },
    {
      "epoch": 50.666666666666664,
      "grad_norm": 5.471190929412842,
      "learning_rate": 9.951274362818591e-08,
      "loss": 1.2298,
      "step": 532
    },
    {
      "epoch": 50.76190476190476,
      "grad_norm": 7.534485816955566,
      "learning_rate": 9.970014992503748e-08,
      "loss": 1.1715,
      "step": 533
    },
    {
      "epoch": 50.857142857142854,
      "grad_norm": 4.532857894897461,
      "learning_rate": 9.988755622188906e-08,
      "loss": 1.1834,
      "step": 534
    },
    {
      "epoch": 50.95238095238095,
      "grad_norm": 6.546600341796875,
      "learning_rate": 1.0007496251874064e-07,
      "loss": 1.2442,
      "step": 535
    },
    {
      "epoch": 50.95238095238095,
      "eval_accuracy": 0.6916448107905583,
      "eval_f1": 0.7462226333641689,
      "eval_loss": 0.6045239567756653,
      "eval_precision": 0.7121836374337845,
      "eval_recall": 0.783678756476684,
      "eval_roc_auc": 0.7328635578583765,
      "eval_runtime": 1.2782,
      "eval_samples_per_second": 2088.101,
      "eval_steps_per_second": 4.694,
      "step": 535
    },
    {
      "epoch": 51.04761904761905,
      "grad_norm": 5.521002292633057,
      "learning_rate": 1.0026236881559221e-07,
      "loss": 1.2303,
      "step": 536
    },
    {
      "epoch": 51.142857142857146,
      "grad_norm": 6.953253746032715,
      "learning_rate": 1.0044977511244379e-07,
      "loss": 1.2548,
      "step": 537
    },
    {
      "epoch": 51.23809523809524,
      "grad_norm": 10.340543746948242,
      "learning_rate": 1.0063718140929536e-07,
      "loss": 1.2135,
      "step": 538
    },
    {
      "epoch": 51.333333333333336,
      "grad_norm": 11.009821891784668,
      "learning_rate": 1.0082458770614694e-07,
      "loss": 1.2175,
      "step": 539
    },
    {
      "epoch": 51.42857142857143,
      "grad_norm": 7.70527458190918,
      "learning_rate": 1.0101199400299849e-07,
      "loss": 1.2264,
      "step": 540
    },
    {
      "epoch": 51.523809523809526,
      "grad_norm": 5.6265363693237305,
      "learning_rate": 1.0119940029985007e-07,
      "loss": 1.212,
      "step": 541
    },
    {
      "epoch": 51.61904761904762,
      "grad_norm": 9.259636878967285,
      "learning_rate": 1.0138680659670167e-07,
      "loss": 1.1998,
      "step": 542
    },
    {
      "epoch": 51.714285714285715,
      "grad_norm": 7.9342851638793945,
      "learning_rate": 1.0157421289355322e-07,
      "loss": 1.1955,
      "step": 543
    },
    {
      "epoch": 51.80952380952381,
      "grad_norm": 5.805824279785156,
      "learning_rate": 1.017616191904048e-07,
      "loss": 1.207,
      "step": 544
    },
    {
      "epoch": 51.904761904761905,
      "grad_norm": 6.57551908493042,
      "learning_rate": 1.0194902548725637e-07,
      "loss": 1.2073,
      "step": 545
    },
    {
      "epoch": 52.0,
      "grad_norm": 7.8102216720581055,
      "learning_rate": 1.0213643178410795e-07,
      "loss": 1.2194,
      "step": 546
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.6908954664668415,
      "eval_f1": 0.7451343836886005,
      "eval_loss": 0.6035574078559875,
      "eval_precision": 0.7123449497932663,
      "eval_recall": 0.7810880829015544,
      "eval_roc_auc": 0.7331142774899252,
      "eval_runtime": 1.1522,
      "eval_samples_per_second": 2316.385,
      "eval_steps_per_second": 5.207,
      "step": 546
    },
    {
      "epoch": 52.095238095238095,
      "grad_norm": 7.845072269439697,
      "learning_rate": 1.0232383808095952e-07,
      "loss": 1.1608,
      "step": 547
    },
    {
      "epoch": 52.19047619047619,
      "grad_norm": 5.869110107421875,
      "learning_rate": 1.025112443778111e-07,
      "loss": 1.216,
      "step": 548
    },
    {
      "epoch": 52.285714285714285,
      "grad_norm": 8.363801002502441,
      "learning_rate": 1.0269865067466268e-07,
      "loss": 1.2384,
      "step": 549
    },
    {
      "epoch": 52.38095238095238,
      "grad_norm": 16.105985641479492,
      "learning_rate": 1.0288605697151425e-07,
      "loss": 1.2684,
      "step": 550
    },
    {
      "epoch": 52.476190476190474,
      "grad_norm": 5.838757038116455,
      "learning_rate": 1.0307346326836583e-07,
      "loss": 1.2206,
      "step": 551
    },
    {
      "epoch": 52.57142857142857,
      "grad_norm": 16.695104598999023,
      "learning_rate": 1.0326086956521738e-07,
      "loss": 1.2561,
      "step": 552
    },
    {
      "epoch": 52.666666666666664,
      "grad_norm": 5.210092544555664,
      "learning_rate": 1.0344827586206898e-07,
      "loss": 1.2346,
      "step": 553
    },
    {
      "epoch": 52.76190476190476,
      "grad_norm": 5.63034200668335,
      "learning_rate": 1.0363568215892053e-07,
      "loss": 1.1727,
      "step": 554
    },
    {
      "epoch": 52.857142857142854,
      "grad_norm": 7.307580947875977,
      "learning_rate": 1.0382308845577211e-07,
      "loss": 1.2514,
      "step": 555
    },
    {
      "epoch": 52.95238095238095,
      "grad_norm": 7.287981986999512,
      "learning_rate": 1.040104947526237e-07,
      "loss": 1.1089,
      "step": 556
    },
    {
      "epoch": 52.95238095238095,
      "eval_accuracy": 0.6916448107905583,
      "eval_f1": 0.7459092312442112,
      "eval_loss": 0.602644145488739,
      "eval_precision": 0.7126843657817109,
      "eval_recall": 0.7823834196891192,
      "eval_roc_auc": 0.7335103626943006,
      "eval_runtime": 1.1449,
      "eval_samples_per_second": 2331.155,
      "eval_steps_per_second": 5.241,
      "step": 556
    },
    {
      "epoch": 53.04761904761905,
      "grad_norm": 11.443868637084961,
      "learning_rate": 1.0419790104947526e-07,
      "loss": 1.1587,
      "step": 557
    },
    {
      "epoch": 53.142857142857146,
      "grad_norm": 7.264131546020508,
      "learning_rate": 1.0438530734632684e-07,
      "loss": 1.2418,
      "step": 558
    },
    {
      "epoch": 53.23809523809524,
      "grad_norm": 4.907642841339111,
      "learning_rate": 1.0457271364317841e-07,
      "loss": 1.1863,
      "step": 559
    },
    {
      "epoch": 53.333333333333336,
      "grad_norm": 8.334844589233398,
      "learning_rate": 1.0476011994002999e-07,
      "loss": 1.2441,
      "step": 560
    },
    {
      "epoch": 53.42857142857143,
      "grad_norm": 9.130343437194824,
      "learning_rate": 1.0494752623688156e-07,
      "loss": 1.2188,
      "step": 561
    },
    {
      "epoch": 53.523809523809526,
      "grad_norm": 6.453733921051025,
      "learning_rate": 1.0513493253373314e-07,
      "loss": 1.1792,
      "step": 562
    },
    {
      "epoch": 53.61904761904762,
      "grad_norm": 10.290508270263672,
      "learning_rate": 1.0532233883058472e-07,
      "loss": 1.2191,
      "step": 563
    },
    {
      "epoch": 53.714285714285715,
      "grad_norm": 10.130738258361816,
      "learning_rate": 1.0550974512743628e-07,
      "loss": 1.2772,
      "step": 564
    },
    {
      "epoch": 53.80952380952381,
      "grad_norm": 6.250458240509033,
      "learning_rate": 1.0569715142428787e-07,
      "loss": 1.1638,
      "step": 565
    },
    {
      "epoch": 53.904761904761905,
      "grad_norm": 8.7243013381958,
      "learning_rate": 1.0588455772113943e-07,
      "loss": 1.1861,
      "step": 566
    },
    {
      "epoch": 54.0,
      "grad_norm": 14.219977378845215,
      "learning_rate": 1.06071964017991e-07,
      "loss": 1.2299,
      "step": 567
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.6901461221431248,
      "eval_f1": 0.7443585780525502,
      "eval_loss": 0.6018810272216797,
      "eval_precision": 0.7120047309284447,
      "eval_recall": 0.7797927461139896,
      "eval_roc_auc": 0.7337489925158318,
      "eval_runtime": 1.1922,
      "eval_samples_per_second": 2238.788,
      "eval_steps_per_second": 5.033,
      "step": 567
    },
    {
      "epoch": 54.095238095238095,
      "grad_norm": 6.662168025970459,
      "learning_rate": 1.0625937031484257e-07,
      "loss": 1.1866,
      "step": 568
    },
    {
      "epoch": 54.19047619047619,
      "grad_norm": 13.390618324279785,
      "learning_rate": 1.0644677661169416e-07,
      "loss": 1.2342,
      "step": 569
    },
    {
      "epoch": 54.285714285714285,
      "grad_norm": 6.478188991546631,
      "learning_rate": 1.0663418290854574e-07,
      "loss": 1.1909,
      "step": 570
    },
    {
      "epoch": 54.38095238095238,
      "grad_norm": 9.548168182373047,
      "learning_rate": 1.068215892053973e-07,
      "loss": 1.1936,
      "step": 571
    },
    {
      "epoch": 54.476190476190474,
      "grad_norm": 5.8785295486450195,
      "learning_rate": 1.0700899550224888e-07,
      "loss": 1.2344,
      "step": 572
    },
    {
      "epoch": 54.57142857142857,
      "grad_norm": 6.237848281860352,
      "learning_rate": 1.0719640179910045e-07,
      "loss": 1.2835,
      "step": 573
    },
    {
      "epoch": 54.666666666666664,
      "grad_norm": 3.976562261581421,
      "learning_rate": 1.0738380809595203e-07,
      "loss": 1.1899,
      "step": 574
    },
    {
      "epoch": 54.76190476190476,
      "grad_norm": 12.389945983886719,
      "learning_rate": 1.0757121439280359e-07,
      "loss": 1.1579,
      "step": 575
    },
    {
      "epoch": 54.857142857142854,
      "grad_norm": 5.776051998138428,
      "learning_rate": 1.0775862068965518e-07,
      "loss": 1.2387,
      "step": 576
    },
    {
      "epoch": 54.95238095238095,
      "grad_norm": 9.520394325256348,
      "learning_rate": 1.0794602698650676e-07,
      "loss": 1.2015,
      "step": 577
    },
    {
      "epoch": 54.95238095238095,
      "eval_accuracy": 0.6905207943049831,
      "eval_f1": 0.7441140024783147,
      "eval_loss": 0.6013320684432983,
      "eval_precision": 0.7131828978622328,
      "eval_recall": 0.7778497409326425,
      "eval_roc_auc": 0.7338149107656879,
      "eval_runtime": 1.1497,
      "eval_samples_per_second": 2321.471,
      "eval_steps_per_second": 5.219,
      "step": 577
    },
    {
      "epoch": 55.04761904761905,
      "grad_norm": 11.354265213012695,
      "learning_rate": 1.0813343328335832e-07,
      "loss": 1.1656,
      "step": 578
    },
    {
      "epoch": 55.142857142857146,
      "grad_norm": 6.79288911819458,
      "learning_rate": 1.083208395802099e-07,
      "loss": 1.2565,
      "step": 579
    },
    {
      "epoch": 55.23809523809524,
      "grad_norm": 6.954402923583984,
      "learning_rate": 1.0850824587706147e-07,
      "loss": 1.2943,
      "step": 580
    },
    {
      "epoch": 55.333333333333336,
      "grad_norm": 7.525900363922119,
      "learning_rate": 1.0869565217391305e-07,
      "loss": 1.1596,
      "step": 581
    },
    {
      "epoch": 55.42857142857143,
      "grad_norm": 5.8448309898376465,
      "learning_rate": 1.0888305847076463e-07,
      "loss": 1.1933,
      "step": 582
    },
    {
      "epoch": 55.523809523809526,
      "grad_norm": 6.475961208343506,
      "learning_rate": 1.090704647676162e-07,
      "loss": 1.2261,
      "step": 583
    },
    {
      "epoch": 55.61904761904762,
      "grad_norm": 7.44426155090332,
      "learning_rate": 1.0925787106446778e-07,
      "loss": 1.1633,
      "step": 584
    },
    {
      "epoch": 55.714285714285715,
      "grad_norm": 10.208169937133789,
      "learning_rate": 1.0944527736131934e-07,
      "loss": 1.1981,
      "step": 585
    },
    {
      "epoch": 55.80952380952381,
      "grad_norm": 8.34847354888916,
      "learning_rate": 1.0963268365817093e-07,
      "loss": 1.1768,
      "step": 586
    },
    {
      "epoch": 55.904761904761905,
      "grad_norm": 5.280081748962402,
      "learning_rate": 1.0982008995502248e-07,
      "loss": 1.189,
      "step": 587
    },
    {
      "epoch": 56.0,
      "grad_norm": 8.047446250915527,
      "learning_rate": 1.1000749625187407e-07,
      "loss": 1.2522,
      "step": 588
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.692394155114275,
      "eval_f1": 0.7474623192863734,
      "eval_loss": 0.6006296873092651,
      "eval_precision": 0.7117750439367311,
      "eval_recall": 0.7869170984455959,
      "eval_roc_auc": 0.7342130109383994,
      "eval_runtime": 1.2179,
      "eval_samples_per_second": 2191.539,
      "eval_steps_per_second": 4.927,
      "step": 588
    },
    {
      "epoch": 56.095238095238095,
      "grad_norm": 5.879335403442383,
      "learning_rate": 1.1019490254872566e-07,
      "loss": 1.2014,
      "step": 589
    },
    {
      "epoch": 56.19047619047619,
      "grad_norm": 6.829493045806885,
      "learning_rate": 1.1038230884557721e-07,
      "loss": 1.1867,
      "step": 590
    },
    {
      "epoch": 56.285714285714285,
      "grad_norm": 6.574869155883789,
      "learning_rate": 1.105697151424288e-07,
      "loss": 1.1913,
      "step": 591
    },
    {
      "epoch": 56.38095238095238,
      "grad_norm": 7.857141971588135,
      "learning_rate": 1.1075712143928036e-07,
      "loss": 1.2192,
      "step": 592
    },
    {
      "epoch": 56.476190476190474,
      "grad_norm": 6.09318733215332,
      "learning_rate": 1.1094452773613194e-07,
      "loss": 1.2501,
      "step": 593
    },
    {
      "epoch": 56.57142857142857,
      "grad_norm": 7.295796871185303,
      "learning_rate": 1.1113193403298351e-07,
      "loss": 1.2479,
      "step": 594
    },
    {
      "epoch": 56.666666666666664,
      "grad_norm": 7.448188781738281,
      "learning_rate": 1.1131934032983509e-07,
      "loss": 1.1998,
      "step": 595
    },
    {
      "epoch": 56.76190476190476,
      "grad_norm": 5.921523094177246,
      "learning_rate": 1.1150674662668667e-07,
      "loss": 1.2029,
      "step": 596
    },
    {
      "epoch": 56.857142857142854,
      "grad_norm": 7.603562355041504,
      "learning_rate": 1.1169415292353824e-07,
      "loss": 1.2234,
      "step": 597
    },
    {
      "epoch": 56.95238095238095,
      "grad_norm": 8.197514533996582,
      "learning_rate": 1.1188155922038982e-07,
      "loss": 1.1689,
      "step": 598
    },
    {
      "epoch": 56.95238095238095,
      "eval_accuracy": 0.6931434994379918,
      "eval_f1": 0.7483870967741936,
      "eval_loss": 0.5999482870101929,
      "eval_precision": 0.711864406779661,
      "eval_recall": 0.788860103626943,
      "eval_roc_auc": 0.7343552101324122,
      "eval_runtime": 1.2535,
      "eval_samples_per_second": 2129.228,
      "eval_steps_per_second": 4.787,
      "step": 598
    },
    {
      "epoch": 57.04761904761905,
      "grad_norm": 5.4424147605896,
      "learning_rate": 1.1206896551724139e-07,
      "loss": 1.2533,
      "step": 599
    },
    {
      "epoch": 57.142857142857146,
      "grad_norm": 6.09983491897583,
      "learning_rate": 1.1225637181409297e-07,
      "loss": 1.2196,
      "step": 600
    },
    {
      "epoch": 57.23809523809524,
      "grad_norm": 10.144179344177246,
      "learning_rate": 1.1244377811094452e-07,
      "loss": 1.1214,
      "step": 601
    },
    {
      "epoch": 57.333333333333336,
      "grad_norm": 8.143068313598633,
      "learning_rate": 1.126311844077961e-07,
      "loss": 1.2736,
      "step": 602
    },
    {
      "epoch": 57.42857142857143,
      "grad_norm": 4.57938289642334,
      "learning_rate": 1.128185907046477e-07,
      "loss": 1.1829,
      "step": 603
    },
    {
      "epoch": 57.523809523809526,
      "grad_norm": 8.159676551818848,
      "learning_rate": 1.1300599700149925e-07,
      "loss": 1.2055,
      "step": 604
    },
    {
      "epoch": 57.61904761904762,
      "grad_norm": 5.003109455108643,
      "learning_rate": 1.1319340329835083e-07,
      "loss": 1.1579,
      "step": 605
    },
    {
      "epoch": 57.714285714285715,
      "grad_norm": 5.263125419616699,
      "learning_rate": 1.133808095952024e-07,
      "loss": 1.18,
      "step": 606
    },
    {
      "epoch": 57.80952380952381,
      "grad_norm": 5.333617687225342,
      "learning_rate": 1.1356821589205398e-07,
      "loss": 1.234,
      "step": 607
    },
    {
      "epoch": 57.904761904761905,
      "grad_norm": 7.104213237762451,
      "learning_rate": 1.1375562218890555e-07,
      "loss": 1.2407,
      "step": 608
    },
    {
      "epoch": 58.0,
      "grad_norm": 5.938223361968994,
      "learning_rate": 1.1394302848575713e-07,
      "loss": 1.2129,
      "step": 609
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.6920194829524167,
      "eval_f1": 0.7472324723247232,
      "eval_loss": 0.5991713404655457,
      "eval_precision": 0.7113583138173302,
      "eval_recall": 0.7869170984455959,
      "eval_roc_auc": 0.7346292458261371,
      "eval_runtime": 1.1169,
      "eval_samples_per_second": 2389.62,
      "eval_steps_per_second": 5.372,
      "step": 609
    },
    {
      "epoch": 58.095238095238095,
      "grad_norm": 5.953458309173584,
      "learning_rate": 1.1413043478260871e-07,
      "loss": 1.231,
      "step": 610
    },
    {
      "epoch": 58.19047619047619,
      "grad_norm": 9.69005012512207,
      "learning_rate": 1.1431784107946028e-07,
      "loss": 1.1305,
      "step": 611
    },
    {
      "epoch": 58.285714285714285,
      "grad_norm": 3.9009530544281006,
      "learning_rate": 1.1450524737631186e-07,
      "loss": 1.1851,
      "step": 612
    },
    {
      "epoch": 58.38095238095238,
      "grad_norm": 8.352299690246582,
      "learning_rate": 1.1469265367316341e-07,
      "loss": 1.2033,
      "step": 613
    },
    {
      "epoch": 58.476190476190474,
      "grad_norm": 7.871908664703369,
      "learning_rate": 1.1488005997001501e-07,
      "loss": 1.2383,
      "step": 614
    },
    {
      "epoch": 58.57142857142857,
      "grad_norm": 5.937413692474365,
      "learning_rate": 1.1506746626686656e-07,
      "loss": 1.2332,
      "step": 615
    },
    {
      "epoch": 58.666666666666664,
      "grad_norm": 7.5808820724487305,
      "learning_rate": 1.1525487256371814e-07,
      "loss": 1.2052,
      "step": 616
    },
    {
      "epoch": 58.76190476190476,
      "grad_norm": 10.876705169677734,
      "learning_rate": 1.1544227886056972e-07,
      "loss": 1.2613,
      "step": 617
    },
    {
      "epoch": 58.857142857142854,
      "grad_norm": 6.522879123687744,
      "learning_rate": 1.1562968515742129e-07,
      "loss": 1.2303,
      "step": 618
    },
    {
      "epoch": 58.95238095238095,
      "grad_norm": 4.543294906616211,
      "learning_rate": 1.1581709145427287e-07,
      "loss": 1.1908,
      "step": 619
    },
    {
      "epoch": 58.95238095238095,
      "eval_accuracy": 0.692394155114275,
      "eval_f1": 0.7477726574500768,
      "eval_loss": 0.5984930396080017,
      "eval_precision": 0.7112799532437172,
      "eval_recall": 0.7882124352331606,
      "eval_roc_auc": 0.7348506044905009,
      "eval_runtime": 1.1584,
      "eval_samples_per_second": 2304.056,
      "eval_steps_per_second": 5.18,
      "step": 619
    },
    {
      "epoch": 59.04761904761905,
      "grad_norm": 6.806978702545166,
      "learning_rate": 1.1600449775112444e-07,
      "loss": 1.2054,
      "step": 620
    },
    {
      "epoch": 59.142857142857146,
      "grad_norm": 4.517679214477539,
      "learning_rate": 1.1619190404797602e-07,
      "loss": 1.2155,
      "step": 621
    },
    {
      "epoch": 59.23809523809524,
      "grad_norm": 8.569692611694336,
      "learning_rate": 1.1637931034482758e-07,
      "loss": 1.2062,
      "step": 622
    },
    {
      "epoch": 59.333333333333336,
      "grad_norm": 7.021420478820801,
      "learning_rate": 1.1656671664167917e-07,
      "loss": 1.1733,
      "step": 623
    },
    {
      "epoch": 59.42857142857143,
      "grad_norm": 6.8625288009643555,
      "learning_rate": 1.1675412293853075e-07,
      "loss": 1.2997,
      "step": 624
    },
    {
      "epoch": 59.523809523809526,
      "grad_norm": 6.276980876922607,
      "learning_rate": 1.169415292353823e-07,
      "loss": 1.1977,
      "step": 625
    },
    {
      "epoch": 59.61904761904762,
      "grad_norm": 4.541833400726318,
      "learning_rate": 1.171289355322339e-07,
      "loss": 1.1695,
      "step": 626
    },
    {
      "epoch": 59.714285714285715,
      "grad_norm": 6.162714004516602,
      "learning_rate": 1.1731634182908545e-07,
      "loss": 1.2158,
      "step": 627
    },
    {
      "epoch": 59.80952380952381,
      "grad_norm": 9.742534637451172,
      "learning_rate": 1.1750374812593703e-07,
      "loss": 1.1878,
      "step": 628
    },
    {
      "epoch": 59.904761904761905,
      "grad_norm": 11.55949878692627,
      "learning_rate": 1.176911544227886e-07,
      "loss": 1.2577,
      "step": 629
    },
    {
      "epoch": 60.0,
      "grad_norm": 7.8720502853393555,
      "learning_rate": 1.1787856071964018e-07,
      "loss": 1.1847,
      "step": 630
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.6916448107905583,
      "eval_f1": 0.7474685486345505,
      "eval_loss": 0.5978415012359619,
      "eval_precision": 0.710204081632653,
      "eval_recall": 0.788860103626943,
      "eval_roc_auc": 0.7350463442717328,
      "eval_runtime": 1.1763,
      "eval_samples_per_second": 2269.063,
      "eval_steps_per_second": 5.101,
      "step": 630
    },
    {
      "epoch": 60.095238095238095,
      "grad_norm": 10.057273864746094,
      "learning_rate": 1.1806596701649176e-07,
      "loss": 1.241,
      "step": 631
    },
    {
      "epoch": 60.19047619047619,
      "grad_norm": 5.35072660446167,
      "learning_rate": 1.1825337331334333e-07,
      "loss": 1.1841,
      "step": 632
    },
    {
      "epoch": 60.285714285714285,
      "grad_norm": 8.153692245483398,
      "learning_rate": 1.1844077961019491e-07,
      "loss": 1.2294,
      "step": 633
    },
    {
      "epoch": 60.38095238095238,
      "grad_norm": 4.509757995605469,
      "learning_rate": 1.1862818590704648e-07,
      "loss": 1.2288,
      "step": 634
    },
    {
      "epoch": 60.476190476190474,
      "grad_norm": 9.07571029663086,
      "learning_rate": 1.1881559220389806e-07,
      "loss": 1.2248,
      "step": 635
    },
    {
      "epoch": 60.57142857142857,
      "grad_norm": 8.7220458984375,
      "learning_rate": 1.1900299850074964e-07,
      "loss": 1.2432,
      "step": 636
    },
    {
      "epoch": 60.666666666666664,
      "grad_norm": 8.08939266204834,
      "learning_rate": 1.1919040479760121e-07,
      "loss": 1.1833,
      "step": 637
    },
    {
      "epoch": 60.76190476190476,
      "grad_norm": 4.108371257781982,
      "learning_rate": 1.1937781109445278e-07,
      "loss": 1.19,
      "step": 638
    },
    {
      "epoch": 60.857142857142854,
      "grad_norm": 11.294686317443848,
      "learning_rate": 1.1956521739130436e-07,
      "loss": 1.1802,
      "step": 639
    },
    {
      "epoch": 60.95238095238095,
      "grad_norm": 3.7819859981536865,
      "learning_rate": 1.1975262368815594e-07,
      "loss": 1.1992,
      "step": 640
    },
    {
      "epoch": 60.95238095238095,
      "eval_accuracy": 0.6927688272761334,
      "eval_f1": 0.7495418448381185,
      "eval_loss": 0.5971837043762207,
      "eval_precision": 0.7092485549132947,
      "eval_recall": 0.7946891191709845,
      "eval_roc_auc": 0.7353765112262521,
      "eval_runtime": 1.2087,
      "eval_samples_per_second": 2208.082,
      "eval_steps_per_second": 4.964,
      "step": 640
    },
    {
      "epoch": 61.04761904761905,
      "grad_norm": 5.13773775100708,
      "learning_rate": 1.199400299850075e-07,
      "loss": 1.2065,
      "step": 641
    },
    {
      "epoch": 61.142857142857146,
      "grad_norm": 5.876140594482422,
      "learning_rate": 1.2012743628185907e-07,
      "loss": 1.2267,
      "step": 642
    },
    {
      "epoch": 61.23809523809524,
      "grad_norm": 7.470527172088623,
      "learning_rate": 1.2031484257871066e-07,
      "loss": 1.2262,
      "step": 643
    },
    {
      "epoch": 61.333333333333336,
      "grad_norm": 5.522604942321777,
      "learning_rate": 1.205022488755622e-07,
      "loss": 1.1942,
      "step": 644
    },
    {
      "epoch": 61.42857142857143,
      "grad_norm": 4.152938365936279,
      "learning_rate": 1.2068965517241382e-07,
      "loss": 1.2191,
      "step": 645
    },
    {
      "epoch": 61.523809523809526,
      "grad_norm": 8.6640043258667,
      "learning_rate": 1.2087706146926537e-07,
      "loss": 1.2091,
      "step": 646
    },
    {
      "epoch": 61.61904761904762,
      "grad_norm": 7.228600978851318,
      "learning_rate": 1.2106446776611695e-07,
      "loss": 1.2186,
      "step": 647
    },
    {
      "epoch": 61.714285714285715,
      "grad_norm": 8.163618087768555,
      "learning_rate": 1.212518740629685e-07,
      "loss": 1.1775,
      "step": 648
    },
    {
      "epoch": 61.80952380952381,
      "grad_norm": 7.274651050567627,
      "learning_rate": 1.214392803598201e-07,
      "loss": 1.17,
      "step": 649
    },
    {
      "epoch": 61.904761904761905,
      "grad_norm": 4.588978290557861,
      "learning_rate": 1.2162668665667167e-07,
      "loss": 1.1833,
      "step": 650
    },
    {
      "epoch": 62.0,
      "grad_norm": 7.658914566040039,
      "learning_rate": 1.2181409295352325e-07,
      "loss": 1.1698,
      "step": 651
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.6920194829524167,
      "eval_f1": 0.7480073574494175,
      "eval_loss": 0.59659343957901,
      "eval_precision": 0.710128055878929,
      "eval_recall": 0.7901554404145078,
      "eval_roc_auc": 0.7355400115141049,
      "eval_runtime": 1.2505,
      "eval_samples_per_second": 2134.361,
      "eval_steps_per_second": 4.798,
      "step": 651
    },
    {
      "epoch": 62.095238095238095,
      "grad_norm": 6.7331743240356445,
      "learning_rate": 1.2200149925037483e-07,
      "loss": 1.1716,
      "step": 652
    },
    {
      "epoch": 62.19047619047619,
      "grad_norm": 6.890786647796631,
      "learning_rate": 1.2218890554722639e-07,
      "loss": 1.2546,
      "step": 653
    },
    {
      "epoch": 62.285714285714285,
      "grad_norm": 6.733562469482422,
      "learning_rate": 1.2237631184407797e-07,
      "loss": 1.2633,
      "step": 654
    },
    {
      "epoch": 62.38095238095238,
      "grad_norm": 4.396595001220703,
      "learning_rate": 1.2256371814092952e-07,
      "loss": 1.1832,
      "step": 655
    },
    {
      "epoch": 62.476190476190474,
      "grad_norm": 5.2550764083862305,
      "learning_rate": 1.227511244377811e-07,
      "loss": 1.2088,
      "step": 656
    },
    {
      "epoch": 62.57142857142857,
      "grad_norm": 8.098922729492188,
      "learning_rate": 1.229385307346327e-07,
      "loss": 1.1956,
      "step": 657
    },
    {
      "epoch": 62.666666666666664,
      "grad_norm": 8.616472244262695,
      "learning_rate": 1.2312593703148426e-07,
      "loss": 1.1821,
      "step": 658
    },
    {
      "epoch": 62.76190476190476,
      "grad_norm": 3.985522508621216,
      "learning_rate": 1.2331334332833585e-07,
      "loss": 1.1895,
      "step": 659
    },
    {
      "epoch": 62.857142857142854,
      "grad_norm": 8.401481628417969,
      "learning_rate": 1.235007496251874e-07,
      "loss": 1.2246,
      "step": 660
    },
    {
      "epoch": 62.95238095238095,
      "grad_norm": 7.483948230743408,
      "learning_rate": 1.2368815592203898e-07,
      "loss": 1.1683,
      "step": 661
    },
    {
      "epoch": 62.95238095238095,
      "eval_accuracy": 0.692394155114275,
      "eval_f1": 0.7499238501370697,
      "eval_loss": 0.5961089134216309,
      "eval_precision": 0.7078780908568143,
      "eval_recall": 0.7972797927461139,
      "eval_roc_auc": 0.7358785261945884,
      "eval_runtime": 1.1192,
      "eval_samples_per_second": 2384.799,
      "eval_steps_per_second": 5.361,
      "step": 661
    },
    {
      "epoch": 63.04761904761905,
      "grad_norm": 7.974298477172852,
      "learning_rate": 1.2387556221889056e-07,
      "loss": 1.2418,
      "step": 662
    },
    {
      "epoch": 63.142857142857146,
      "grad_norm": 4.714601516723633,
      "learning_rate": 1.2406296851574214e-07,
      "loss": 1.1899,
      "step": 663
    },
    {
      "epoch": 63.23809523809524,
      "grad_norm": 8.871092796325684,
      "learning_rate": 1.2425037481259372e-07,
      "loss": 1.1732,
      "step": 664
    },
    {
      "epoch": 63.333333333333336,
      "grad_norm": 6.921295642852783,
      "learning_rate": 1.2443778110944528e-07,
      "loss": 1.1797,
      "step": 665
    },
    {
      "epoch": 63.42857142857143,
      "grad_norm": 6.838317394256592,
      "learning_rate": 1.2462518740629686e-07,
      "loss": 1.2006,
      "step": 666
    },
    {
      "epoch": 63.523809523809526,
      "grad_norm": 9.975998878479004,
      "learning_rate": 1.2481259370314841e-07,
      "loss": 1.1928,
      "step": 667
    },
    {
      "epoch": 63.61904761904762,
      "grad_norm": 4.443400859832764,
      "learning_rate": 1.2500000000000002e-07,
      "loss": 1.1637,
      "step": 668
    },
    {
      "epoch": 63.714285714285715,
      "grad_norm": 6.422091960906982,
      "learning_rate": 1.2518740629685158e-07,
      "loss": 1.2192,
      "step": 669
    },
    {
      "epoch": 63.80952380952381,
      "grad_norm": 8.040130615234375,
      "learning_rate": 1.2537481259370316e-07,
      "loss": 1.2418,
      "step": 670
    },
    {
      "epoch": 63.904761904761905,
      "grad_norm": 4.976798057556152,
      "learning_rate": 1.2556221889055474e-07,
      "loss": 1.1433,
      "step": 671
    },
    {
      "epoch": 64.0,
      "grad_norm": 5.444014549255371,
      "learning_rate": 1.257496251874063e-07,
      "loss": 1.2515,
      "step": 672
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.6931434994379918,
      "eval_f1": 0.7518933656467737,
      "eval_loss": 0.5956265330314636,
      "eval_precision": 0.7063175867956745,
      "eval_recall": 0.8037564766839378,
      "eval_roc_auc": 0.7361352907311456,
      "eval_runtime": 1.1898,
      "eval_samples_per_second": 2243.242,
      "eval_steps_per_second": 5.043,
      "step": 672
    },
    {
      "epoch": 64.0952380952381,
      "grad_norm": 7.590255260467529,
      "learning_rate": 1.2593703148425787e-07,
      "loss": 1.2581,
      "step": 673
    },
    {
      "epoch": 64.19047619047619,
      "grad_norm": 4.764690399169922,
      "learning_rate": 1.2612443778110945e-07,
      "loss": 1.2219,
      "step": 674
    },
    {
      "epoch": 64.28571428571429,
      "grad_norm": 4.292867660522461,
      "learning_rate": 1.2631184407796103e-07,
      "loss": 1.1789,
      "step": 675
    },
    {
      "epoch": 64.38095238095238,
      "grad_norm": 8.631563186645508,
      "learning_rate": 1.264992503748126e-07,
      "loss": 1.2549,
      "step": 676
    },
    {
      "epoch": 64.47619047619048,
      "grad_norm": 5.184398174285889,
      "learning_rate": 1.2668665667166417e-07,
      "loss": 1.2146,
      "step": 677
    },
    {
      "epoch": 64.57142857142857,
      "grad_norm": 9.575491905212402,
      "learning_rate": 1.2687406296851575e-07,
      "loss": 1.1306,
      "step": 678
    },
    {
      "epoch": 64.66666666666667,
      "grad_norm": 4.342453479766846,
      "learning_rate": 1.270614692653673e-07,
      "loss": 1.182,
      "step": 679
    },
    {
      "epoch": 64.76190476190476,
      "grad_norm": 4.187139987945557,
      "learning_rate": 1.272488755622189e-07,
      "loss": 1.1806,
      "step": 680
    },
    {
      "epoch": 64.85714285714286,
      "grad_norm": 7.976365089416504,
      "learning_rate": 1.2743628185907047e-07,
      "loss": 1.1707,
      "step": 681
    },
    {
      "epoch": 64.95238095238095,
      "grad_norm": 4.439985275268555,
      "learning_rate": 1.2762368815592205e-07,
      "loss": 1.1925,
      "step": 682
    },
    {
      "epoch": 64.95238095238095,
      "eval_accuracy": 0.6931434994379918,
      "eval_f1": 0.7515923566878981,
      "eval_loss": 0.5950964689254761,
      "eval_precision": 0.7067883628066173,
      "eval_recall": 0.802461139896373,
      "eval_roc_auc": 0.736240932642487,
      "eval_runtime": 1.1681,
      "eval_samples_per_second": 2284.886,
      "eval_steps_per_second": 5.136,
      "step": 682
    },
    {
      "epoch": 65.04761904761905,
      "grad_norm": 13.692363739013672,
      "learning_rate": 1.278110944527736e-07,
      "loss": 1.2215,
      "step": 683
    },
    {
      "epoch": 65.14285714285714,
      "grad_norm": 12.23557186126709,
      "learning_rate": 1.2799850074962518e-07,
      "loss": 1.2143,
      "step": 684
    },
    {
      "epoch": 65.23809523809524,
      "grad_norm": 5.822898864746094,
      "learning_rate": 1.2818590704647676e-07,
      "loss": 1.2465,
      "step": 685
    },
    {
      "epoch": 65.33333333333333,
      "grad_norm": 8.577857971191406,
      "learning_rate": 1.2837331334332835e-07,
      "loss": 1.1758,
      "step": 686
    },
    {
      "epoch": 65.42857142857143,
      "grad_norm": 5.592923641204834,
      "learning_rate": 1.2856071964017993e-07,
      "loss": 1.229,
      "step": 687
    },
    {
      "epoch": 65.52380952380952,
      "grad_norm": 8.428314208984375,
      "learning_rate": 1.2874812593703148e-07,
      "loss": 1.1457,
      "step": 688
    },
    {
      "epoch": 65.61904761904762,
      "grad_norm": 4.13578462600708,
      "learning_rate": 1.2893553223388306e-07,
      "loss": 1.221,
      "step": 689
    },
    {
      "epoch": 65.71428571428571,
      "grad_norm": 10.429786682128906,
      "learning_rate": 1.2912293853073464e-07,
      "loss": 1.2281,
      "step": 690
    },
    {
      "epoch": 65.80952380952381,
      "grad_norm": 6.562197685241699,
      "learning_rate": 1.2931034482758622e-07,
      "loss": 1.2364,
      "step": 691
    },
    {
      "epoch": 65.9047619047619,
      "grad_norm": 5.115170001983643,
      "learning_rate": 1.294977511244378e-07,
      "loss": 1.1427,
      "step": 692
    },
    {
      "epoch": 66.0,
      "grad_norm": 5.716797351837158,
      "learning_rate": 1.2968515742128936e-07,
      "loss": 1.1304,
      "step": 693
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.6920194829524167,
      "eval_f1": 0.7510599636583889,
      "eval_loss": 0.5946346521377563,
      "eval_precision": 0.7053469852104665,
      "eval_recall": 0.8031088082901554,
      "eval_roc_auc": 0.7363713298791019,
      "eval_runtime": 1.4654,
      "eval_samples_per_second": 1821.307,
      "eval_steps_per_second": 4.094,
      "step": 693
    },
    {
      "epoch": 66.0952380952381,
      "grad_norm": 4.564248085021973,
      "learning_rate": 1.2987256371814094e-07,
      "loss": 1.1607,
      "step": 694
    },
    {
      "epoch": 66.19047619047619,
      "grad_norm": 5.520058631896973,
      "learning_rate": 1.300599700149925e-07,
      "loss": 1.1834,
      "step": 695
    },
    {
      "epoch": 66.28571428571429,
      "grad_norm": 7.891077041625977,
      "learning_rate": 1.3024737631184408e-07,
      "loss": 1.241,
      "step": 696
    },
    {
      "epoch": 66.38095238095238,
      "grad_norm": 3.926841974258423,
      "learning_rate": 1.3043478260869566e-07,
      "loss": 1.1777,
      "step": 697
    },
    {
      "epoch": 66.47619047619048,
      "grad_norm": 7.784186840057373,
      "learning_rate": 1.3062218890554724e-07,
      "loss": 1.2522,
      "step": 698
    },
    {
      "epoch": 66.57142857142857,
      "grad_norm": 4.014712810516357,
      "learning_rate": 1.3080959520239882e-07,
      "loss": 1.1957,
      "step": 699
    },
    {
      "epoch": 66.66666666666667,
      "grad_norm": 3.500624656677246,
      "learning_rate": 1.3099700149925037e-07,
      "loss": 1.1759,
      "step": 700
    },
    {
      "epoch": 66.76190476190476,
      "grad_norm": 4.814046859741211,
      "learning_rate": 1.3118440779610195e-07,
      "loss": 1.199,
      "step": 701
    },
    {
      "epoch": 66.85714285714286,
      "grad_norm": 7.344492435455322,
      "learning_rate": 1.313718140929535e-07,
      "loss": 1.2273,
      "step": 702
    },
    {
      "epoch": 66.95238095238095,
      "grad_norm": 3.8098535537719727,
      "learning_rate": 1.3155922038980512e-07,
      "loss": 1.1647,
      "step": 703
    },
    {
      "epoch": 66.95238095238095,
      "eval_accuracy": 0.6938928437617085,
      "eval_f1": 0.7529482915028727,
      "eval_loss": 0.5941770672798157,
      "eval_precision": 0.706182643221781,
      "eval_recall": 0.8063471502590673,
      "eval_roc_auc": 0.736562176165803,
      "eval_runtime": 1.2309,
      "eval_samples_per_second": 2168.262,
      "eval_steps_per_second": 4.874,
      "step": 703
    },
    {
      "epoch": 67.04761904761905,
      "grad_norm": 6.632507801055908,
      "learning_rate": 1.317466266866567e-07,
      "loss": 1.1796,
      "step": 704
    },
    {
      "epoch": 67.14285714285714,
      "grad_norm": 4.132903575897217,
      "learning_rate": 1.3193403298350825e-07,
      "loss": 1.1704,
      "step": 705
    },
    {
      "epoch": 67.23809523809524,
      "grad_norm": 6.37913179397583,
      "learning_rate": 1.3212143928035983e-07,
      "loss": 1.208,
      "step": 706
    },
    {
      "epoch": 67.33333333333333,
      "grad_norm": 7.594150543212891,
      "learning_rate": 1.3230884557721139e-07,
      "loss": 1.159,
      "step": 707
    },
    {
      "epoch": 67.42857142857143,
      "grad_norm": 6.472071170806885,
      "learning_rate": 1.3249625187406297e-07,
      "loss": 1.1981,
      "step": 708
    },
    {
      "epoch": 67.52380952380952,
      "grad_norm": 8.726730346679688,
      "learning_rate": 1.3268365817091455e-07,
      "loss": 1.23,
      "step": 709
    },
    {
      "epoch": 67.61904761904762,
      "grad_norm": 6.169523239135742,
      "learning_rate": 1.3287106446776613e-07,
      "loss": 1.2393,
      "step": 710
    },
    {
      "epoch": 67.71428571428571,
      "grad_norm": 7.1220879554748535,
      "learning_rate": 1.330584707646177e-07,
      "loss": 1.2379,
      "step": 711
    },
    {
      "epoch": 67.80952380952381,
      "grad_norm": 4.605555534362793,
      "learning_rate": 1.3324587706146927e-07,
      "loss": 1.1859,
      "step": 712
    },
    {
      "epoch": 67.9047619047619,
      "grad_norm": 3.750711679458618,
      "learning_rate": 1.3343328335832085e-07,
      "loss": 1.1814,
      "step": 713
    },
    {
      "epoch": 68.0,
      "grad_norm": 9.115017890930176,
      "learning_rate": 1.3362068965517243e-07,
      "loss": 1.1628,
      "step": 714
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.6927688272761334,
      "eval_f1": 0.7512135922330098,
      "eval_loss": 0.5937598943710327,
      "eval_precision": 0.70662100456621,
      "eval_recall": 0.8018134715025906,
      "eval_roc_auc": 0.736656303972366,
      "eval_runtime": 1.2369,
      "eval_samples_per_second": 2157.854,
      "eval_steps_per_second": 4.851,
      "step": 714
    },
    {
      "epoch": 68.0952380952381,
      "grad_norm": 6.957290172576904,
      "learning_rate": 1.33808095952024e-07,
      "loss": 1.1832,
      "step": 715
    },
    {
      "epoch": 68.19047619047619,
      "grad_norm": 6.766869068145752,
      "learning_rate": 1.3399550224887556e-07,
      "loss": 1.2354,
      "step": 716
    },
    {
      "epoch": 68.28571428571429,
      "grad_norm": 4.803560733795166,
      "learning_rate": 1.3418290854572714e-07,
      "loss": 1.1683,
      "step": 717
    },
    {
      "epoch": 68.38095238095238,
      "grad_norm": 6.3927693367004395,
      "learning_rate": 1.3437031484257872e-07,
      "loss": 1.2238,
      "step": 718
    },
    {
      "epoch": 68.47619047619048,
      "grad_norm": 4.72044563293457,
      "learning_rate": 1.3455772113943028e-07,
      "loss": 1.1838,
      "step": 719
    },
    {
      "epoch": 68.57142857142857,
      "grad_norm": 3.594869375228882,
      "learning_rate": 1.3474512743628186e-07,
      "loss": 1.2327,
      "step": 720
    },
    {
      "epoch": 68.66666666666667,
      "grad_norm": 8.74575138092041,
      "learning_rate": 1.3493253373313344e-07,
      "loss": 1.1613,
      "step": 721
    },
    {
      "epoch": 68.76190476190476,
      "grad_norm": 4.941747665405273,
      "learning_rate": 1.3511994002998502e-07,
      "loss": 1.208,
      "step": 722
    },
    {
      "epoch": 68.85714285714286,
      "grad_norm": 4.247956275939941,
      "learning_rate": 1.3530734632683658e-07,
      "loss": 1.2211,
      "step": 723
    },
    {
      "epoch": 68.95238095238095,
      "grad_norm": 13.995353698730469,
      "learning_rate": 1.3549475262368816e-07,
      "loss": 1.1874,
      "step": 724
    },
    {
      "epoch": 68.95238095238095,
      "eval_accuracy": 0.6931434994379918,
      "eval_f1": 0.7514415781487102,
      "eval_loss": 0.5934665203094482,
      "eval_precision": 0.7070245573957739,
      "eval_recall": 0.8018134715025906,
      "eval_roc_auc": 0.7367440990213011,
      "eval_runtime": 1.1208,
      "eval_samples_per_second": 2381.392,
      "eval_steps_per_second": 5.353,
      "step": 724
    },
    {
      "epoch": 69.04761904761905,
      "grad_norm": 5.009316444396973,
      "learning_rate": 1.3568215892053974e-07,
      "loss": 1.2182,
      "step": 725
    },
    {
      "epoch": 69.14285714285714,
      "grad_norm": 9.797475814819336,
      "learning_rate": 1.3586956521739132e-07,
      "loss": 1.1913,
      "step": 726
    },
    {
      "epoch": 69.23809523809524,
      "grad_norm": 3.616032600402832,
      "learning_rate": 1.360569715142429e-07,
      "loss": 1.1511,
      "step": 727
    },
    {
      "epoch": 69.33333333333333,
      "grad_norm": 7.884753704071045,
      "learning_rate": 1.3624437781109445e-07,
      "loss": 1.1546,
      "step": 728
    },
    {
      "epoch": 69.42857142857143,
      "grad_norm": 6.181896686553955,
      "learning_rate": 1.3643178410794604e-07,
      "loss": 1.2139,
      "step": 729
    },
    {
      "epoch": 69.52380952380952,
      "grad_norm": 5.938560485839844,
      "learning_rate": 1.366191904047976e-07,
      "loss": 1.1997,
      "step": 730
    },
    {
      "epoch": 69.61904761904762,
      "grad_norm": 7.971124172210693,
      "learning_rate": 1.3680659670164917e-07,
      "loss": 1.2043,
      "step": 731
    },
    {
      "epoch": 69.71428571428571,
      "grad_norm": 5.756801128387451,
      "learning_rate": 1.3699400299850075e-07,
      "loss": 1.2044,
      "step": 732
    },
    {
      "epoch": 69.80952380952381,
      "grad_norm": 4.534045696258545,
      "learning_rate": 1.3718140929535233e-07,
      "loss": 1.1576,
      "step": 733
    },
    {
      "epoch": 69.9047619047619,
      "grad_norm": 4.456146240234375,
      "learning_rate": 1.3736881559220391e-07,
      "loss": 1.2251,
      "step": 734
    },
    {
      "epoch": 70.0,
      "grad_norm": 4.985422134399414,
      "learning_rate": 1.3755622188905547e-07,
      "loss": 1.1776,
      "step": 735
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.6935181715998502,
      "eval_f1": 0.752720677146312,
      "eval_loss": 0.5932142734527588,
      "eval_precision": 0.70578231292517,
      "eval_recall": 0.8063471502590673,
      "eval_roc_auc": 0.7369044329303398,
      "eval_runtime": 1.1959,
      "eval_samples_per_second": 2231.765,
      "eval_steps_per_second": 5.017,
      "step": 735
    },
    {
      "epoch": 70.0952380952381,
      "grad_norm": 4.807669162750244,
      "learning_rate": 1.3774362818590705e-07,
      "loss": 1.2149,
      "step": 736
    },
    {
      "epoch": 70.19047619047619,
      "grad_norm": 5.5707502365112305,
      "learning_rate": 1.379310344827586e-07,
      "loss": 1.1785,
      "step": 737
    },
    {
      "epoch": 70.28571428571429,
      "grad_norm": 4.975805759429932,
      "learning_rate": 1.381184407796102e-07,
      "loss": 1.1759,
      "step": 738
    },
    {
      "epoch": 70.38095238095238,
      "grad_norm": 9.631810188293457,
      "learning_rate": 1.383058470764618e-07,
      "loss": 1.1972,
      "step": 739
    },
    {
      "epoch": 70.47619047619048,
      "grad_norm": 7.148436546325684,
      "learning_rate": 1.3849325337331335e-07,
      "loss": 1.1757,
      "step": 740
    },
    {
      "epoch": 70.57142857142857,
      "grad_norm": 4.956615924835205,
      "learning_rate": 1.3868065967016493e-07,
      "loss": 1.2066,
      "step": 741
    },
    {
      "epoch": 70.66666666666667,
      "grad_norm": 3.861035108566284,
      "learning_rate": 1.3886806596701648e-07,
      "loss": 1.1936,
      "step": 742
    },
    {
      "epoch": 70.76190476190476,
      "grad_norm": 8.290705680847168,
      "learning_rate": 1.3905547226386806e-07,
      "loss": 1.1542,
      "step": 743
    },
    {
      "epoch": 70.85714285714286,
      "grad_norm": 4.794528484344482,
      "learning_rate": 1.3924287856071967e-07,
      "loss": 1.1657,
      "step": 744
    },
    {
      "epoch": 70.95238095238095,
      "grad_norm": 10.255331039428711,
      "learning_rate": 1.3943028485757122e-07,
      "loss": 1.291,
      "step": 745
    },
    {
      "epoch": 70.95238095238095,
      "eval_accuracy": 0.6946421880854252,
      "eval_f1": 0.7538508003624282,
      "eval_loss": 0.5930056571960449,
      "eval_precision": 0.7062818336162988,
      "eval_recall": 0.8082901554404145,
      "eval_roc_auc": 0.7369087507196316,
      "eval_runtime": 1.2213,
      "eval_samples_per_second": 2185.328,
      "eval_steps_per_second": 4.913,
      "step": 745
    },
    {
      "epoch": 71.04761904761905,
      "grad_norm": 6.958563804626465,
      "learning_rate": 1.396176911544228e-07,
      "loss": 1.228,
      "step": 746
    },
    {
      "epoch": 71.14285714285714,
      "grad_norm": 7.86414098739624,
      "learning_rate": 1.3980509745127436e-07,
      "loss": 1.1859,
      "step": 747
    },
    {
      "epoch": 71.23809523809524,
      "grad_norm": 4.688377380371094,
      "learning_rate": 1.3999250374812594e-07,
      "loss": 1.1815,
      "step": 748
    },
    {
      "epoch": 71.33333333333333,
      "grad_norm": 6.568027496337891,
      "learning_rate": 1.4017991004497752e-07,
      "loss": 1.1951,
      "step": 749
    },
    {
      "epoch": 71.42857142857143,
      "grad_norm": 9.667662620544434,
      "learning_rate": 1.403673163418291e-07,
      "loss": 1.1897,
      "step": 750
    },
    {
      "epoch": 71.52380952380952,
      "grad_norm": 3.651684045791626,
      "learning_rate": 1.4055472263868068e-07,
      "loss": 1.2159,
      "step": 751
    },
    {
      "epoch": 71.61904761904762,
      "grad_norm": 4.854371070861816,
      "learning_rate": 1.4074212893553224e-07,
      "loss": 1.2089,
      "step": 752
    },
    {
      "epoch": 71.71428571428571,
      "grad_norm": 5.8277153968811035,
      "learning_rate": 1.4092953523238382e-07,
      "loss": 1.1586,
      "step": 753
    },
    {
      "epoch": 71.80952380952381,
      "grad_norm": 8.932456016540527,
      "learning_rate": 1.4111694152923537e-07,
      "loss": 1.1964,
      "step": 754
    },
    {
      "epoch": 71.9047619047619,
      "grad_norm": 7.3312482833862305,
      "learning_rate": 1.4130434782608695e-07,
      "loss": 1.2107,
      "step": 755
    },
    {
      "epoch": 72.0,
      "grad_norm": 4.927225589752197,
      "learning_rate": 1.4149175412293854e-07,
      "loss": 1.1468,
      "step": 756
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.6961408767328587,
      "eval_f1": 0.7550588945937783,
      "eval_loss": 0.5926088690757751,
      "eval_precision": 0.7074136955291455,
      "eval_recall": 0.8095854922279793,
      "eval_roc_auc": 0.7370814622913068,
      "eval_runtime": 1.2983,
      "eval_samples_per_second": 2055.808,
      "eval_steps_per_second": 4.622,
      "step": 756
    },
    {
      "epoch": 72.0952380952381,
      "grad_norm": 4.906456470489502,
      "learning_rate": 1.4167916041979012e-07,
      "loss": 1.1828,
      "step": 757
    },
    {
      "epoch": 72.19047619047619,
      "grad_norm": 3.9079153537750244,
      "learning_rate": 1.418665667166417e-07,
      "loss": 1.2224,
      "step": 758
    },
    {
      "epoch": 72.28571428571429,
      "grad_norm": 6.527932167053223,
      "learning_rate": 1.4205397301349325e-07,
      "loss": 1.1986,
      "step": 759
    },
    {
      "epoch": 72.38095238095238,
      "grad_norm": 4.133279800415039,
      "learning_rate": 1.4224137931034483e-07,
      "loss": 1.1761,
      "step": 760
    },
    {
      "epoch": 72.47619047619048,
      "grad_norm": 4.781145095825195,
      "learning_rate": 1.4242878560719641e-07,
      "loss": 1.1717,
      "step": 761
    },
    {
      "epoch": 72.57142857142857,
      "grad_norm": 3.2274749279022217,
      "learning_rate": 1.42616191904048e-07,
      "loss": 1.1883,
      "step": 762
    },
    {
      "epoch": 72.66666666666667,
      "grad_norm": 4.167154312133789,
      "learning_rate": 1.4280359820089955e-07,
      "loss": 1.1845,
      "step": 763
    },
    {
      "epoch": 72.76190476190476,
      "grad_norm": 3.8813252449035645,
      "learning_rate": 1.4299100449775113e-07,
      "loss": 1.183,
      "step": 764
    },
    {
      "epoch": 72.85714285714286,
      "grad_norm": 6.702694892883301,
      "learning_rate": 1.431784107946027e-07,
      "loss": 1.2002,
      "step": 765
    },
    {
      "epoch": 72.95238095238095,
      "grad_norm": 5.558817386627197,
      "learning_rate": 1.4336581709145427e-07,
      "loss": 1.2337,
      "step": 766
    },
    {
      "epoch": 72.95238095238095,
      "eval_accuracy": 0.6972648932184339,
      "eval_f1": 0.756185878092939,
      "eval_loss": 0.5922711491584778,
      "eval_precision": 0.707909604519774,
      "eval_recall": 0.8115284974093264,
      "eval_roc_auc": 0.7372495682210709,
      "eval_runtime": 1.2573,
      "eval_samples_per_second": 2122.807,
      "eval_steps_per_second": 4.772,
      "step": 766
    },
    {
      "epoch": 73.04761904761905,
      "grad_norm": 3.9507691860198975,
      "learning_rate": 1.4355322338830587e-07,
      "loss": 1.1799,
      "step": 767
    },
    {
      "epoch": 73.14285714285714,
      "grad_norm": 4.559001445770264,
      "learning_rate": 1.4374062968515743e-07,
      "loss": 1.1582,
      "step": 768
    },
    {
      "epoch": 73.23809523809524,
      "grad_norm": 3.7043728828430176,
      "learning_rate": 1.43928035982009e-07,
      "loss": 1.1928,
      "step": 769
    },
    {
      "epoch": 73.33333333333333,
      "grad_norm": 4.907166957855225,
      "learning_rate": 1.4411544227886056e-07,
      "loss": 1.1562,
      "step": 770
    },
    {
      "epoch": 73.42857142857143,
      "grad_norm": 10.487092971801758,
      "learning_rate": 1.4430284857571214e-07,
      "loss": 1.2349,
      "step": 771
    },
    {
      "epoch": 73.52380952380952,
      "grad_norm": 5.899962425231934,
      "learning_rate": 1.4449025487256373e-07,
      "loss": 1.2333,
      "step": 772
    },
    {
      "epoch": 73.61904761904762,
      "grad_norm": 6.258727073669434,
      "learning_rate": 1.446776611694153e-07,
      "loss": 1.1558,
      "step": 773
    },
    {
      "epoch": 73.71428571428571,
      "grad_norm": 4.086488723754883,
      "learning_rate": 1.448650674662669e-07,
      "loss": 1.221,
      "step": 774
    },
    {
      "epoch": 73.80952380952381,
      "grad_norm": 6.162347793579102,
      "learning_rate": 1.4505247376311844e-07,
      "loss": 1.2575,
      "step": 775
    },
    {
      "epoch": 73.9047619047619,
      "grad_norm": 4.086270809173584,
      "learning_rate": 1.4523988005997002e-07,
      "loss": 1.1659,
      "step": 776
    },
    {
      "epoch": 74.0,
      "grad_norm": 5.596920967102051,
      "learning_rate": 1.4542728635682158e-07,
      "loss": 1.1919,
      "step": 777
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.6965155488947171,
      "eval_f1": 0.7554347826086957,
      "eval_loss": 0.5918785333633423,
      "eval_precision": 0.707579185520362,
      "eval_recall": 0.8102331606217616,
      "eval_roc_auc": 0.7373598157743234,
      "eval_runtime": 1.1713,
      "eval_samples_per_second": 2278.735,
      "eval_steps_per_second": 5.123,
      "step": 777
    },
    {
      "epoch": 74.0952380952381,
      "grad_norm": 6.950702667236328,
      "learning_rate": 1.4561469265367316e-07,
      "loss": 1.2012,
      "step": 778
    },
    {
      "epoch": 74.19047619047619,
      "grad_norm": 6.536509037017822,
      "learning_rate": 1.4580209895052477e-07,
      "loss": 1.1679,
      "step": 779
    },
    {
      "epoch": 74.28571428571429,
      "grad_norm": 3.573523998260498,
      "learning_rate": 1.4598950524737632e-07,
      "loss": 1.2136,
      "step": 780
    },
    {
      "epoch": 74.38095238095238,
      "grad_norm": 5.456609725952148,
      "learning_rate": 1.461769115442279e-07,
      "loss": 1.21,
      "step": 781
    },
    {
      "epoch": 74.47619047619048,
      "grad_norm": 6.9601922035217285,
      "learning_rate": 1.4636431784107946e-07,
      "loss": 1.1881,
      "step": 782
    },
    {
      "epoch": 74.57142857142857,
      "grad_norm": 9.396449089050293,
      "learning_rate": 1.4655172413793104e-07,
      "loss": 1.2234,
      "step": 783
    },
    {
      "epoch": 74.66666666666667,
      "grad_norm": 7.922398567199707,
      "learning_rate": 1.4673913043478262e-07,
      "loss": 1.1364,
      "step": 784
    },
    {
      "epoch": 74.76190476190476,
      "grad_norm": 7.5325093269348145,
      "learning_rate": 1.469265367316342e-07,
      "loss": 1.189,
      "step": 785
    },
    {
      "epoch": 74.85714285714286,
      "grad_norm": 5.111273765563965,
      "learning_rate": 1.4711394302848578e-07,
      "loss": 1.2269,
      "step": 786
    },
    {
      "epoch": 74.95238095238095,
      "grad_norm": 3.6373484134674072,
      "learning_rate": 1.4730134932533733e-07,
      "loss": 1.1915,
      "step": 787
    },
    {
      "epoch": 74.95238095238095,
      "eval_accuracy": 0.6976395653802923,
      "eval_f1": 0.7565610859728507,
      "eval_loss": 0.5917426943778992,
      "eval_precision": 0.7080745341614907,
      "eval_recall": 0.8121761658031088,
      "eval_roc_auc": 0.7373359240069084,
      "eval_runtime": 1.1623,
      "eval_samples_per_second": 2296.364,
      "eval_steps_per_second": 5.162,
      "step": 787
    },
    {
      "epoch": 75.04761904761905,
      "grad_norm": 3.7085328102111816,
      "learning_rate": 1.4748875562218891e-07,
      "loss": 1.2211,
      "step": 788
    },
    {
      "epoch": 75.14285714285714,
      "grad_norm": 5.547911643981934,
      "learning_rate": 1.4767616191904047e-07,
      "loss": 1.2192,
      "step": 789
    },
    {
      "epoch": 75.23809523809524,
      "grad_norm": 5.634171485900879,
      "learning_rate": 1.4786356821589205e-07,
      "loss": 1.2382,
      "step": 790
    },
    {
      "epoch": 75.33333333333333,
      "grad_norm": 4.337423801422119,
      "learning_rate": 1.4805097451274363e-07,
      "loss": 1.1841,
      "step": 791
    },
    {
      "epoch": 75.42857142857143,
      "grad_norm": 3.4552102088928223,
      "learning_rate": 1.482383808095952e-07,
      "loss": 1.1838,
      "step": 792
    },
    {
      "epoch": 75.52380952380952,
      "grad_norm": 4.098034381866455,
      "learning_rate": 1.484257871064468e-07,
      "loss": 1.1774,
      "step": 793
    },
    {
      "epoch": 75.61904761904762,
      "grad_norm": 3.889285087585449,
      "learning_rate": 1.4861319340329835e-07,
      "loss": 1.1794,
      "step": 794
    },
    {
      "epoch": 75.71428571428571,
      "grad_norm": 6.518879413604736,
      "learning_rate": 1.4880059970014993e-07,
      "loss": 1.1664,
      "step": 795
    },
    {
      "epoch": 75.80952380952381,
      "grad_norm": 5.728705883026123,
      "learning_rate": 1.489880059970015e-07,
      "loss": 1.1829,
      "step": 796
    },
    {
      "epoch": 75.9047619047619,
      "grad_norm": 6.4669389724731445,
      "learning_rate": 1.491754122938531e-07,
      "loss": 1.1957,
      "step": 797
    },
    {
      "epoch": 76.0,
      "grad_norm": 3.6122305393218994,
      "learning_rate": 1.4936281859070467e-07,
      "loss": 1.1849,
      "step": 798
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.6980142375421506,
      "eval_f1": 0.7572289156626506,
      "eval_loss": 0.5914971232414246,
      "eval_precision": 0.7077702702702703,
      "eval_recall": 0.814119170984456,
      "eval_roc_auc": 0.7374999999999999,
      "eval_runtime": 1.2208,
      "eval_samples_per_second": 2186.252,
      "eval_steps_per_second": 4.915,
      "step": 798
    },
    {
      "epoch": 76.0952380952381,
      "grad_norm": 5.138617038726807,
      "learning_rate": 1.4955022488755623e-07,
      "loss": 1.1999,
      "step": 799
    },
    {
      "epoch": 76.19047619047619,
      "grad_norm": 4.243399143218994,
      "learning_rate": 1.497376311844078e-07,
      "loss": 1.1639,
      "step": 800
    },
    {
      "epoch": 76.28571428571429,
      "grad_norm": 3.2358734607696533,
      "learning_rate": 1.4992503748125936e-07,
      "loss": 1.1862,
      "step": 801
    },
    {
      "epoch": 76.38095238095238,
      "grad_norm": 3.2951650619506836,
      "learning_rate": 1.5011244377811097e-07,
      "loss": 1.2081,
      "step": 802
    },
    {
      "epoch": 76.47619047619048,
      "grad_norm": 7.856222629547119,
      "learning_rate": 1.5029985007496252e-07,
      "loss": 1.1956,
      "step": 803
    },
    {
      "epoch": 76.57142857142857,
      "grad_norm": 5.153473377227783,
      "learning_rate": 1.504872563718141e-07,
      "loss": 1.1984,
      "step": 804
    },
    {
      "epoch": 76.66666666666667,
      "grad_norm": 5.476777076721191,
      "learning_rate": 1.5067466266866568e-07,
      "loss": 1.1677,
      "step": 805
    },
    {
      "epoch": 76.76190476190476,
      "grad_norm": 3.60621976852417,
      "learning_rate": 1.5086206896551724e-07,
      "loss": 1.1979,
      "step": 806
    },
    {
      "epoch": 76.85714285714286,
      "grad_norm": 11.043159484863281,
      "learning_rate": 1.5104947526236882e-07,
      "loss": 1.2544,
      "step": 807
    },
    {
      "epoch": 76.95238095238095,
      "grad_norm": 4.6011433601379395,
      "learning_rate": 1.512368815592204e-07,
      "loss": 1.1528,
      "step": 808
    },
    {
      "epoch": 76.95238095238095,
      "eval_accuracy": 0.6995129261895842,
      "eval_f1": 0.7600239377618193,
      "eval_loss": 0.5912597179412842,
      "eval_precision": 0.7063403781979978,
      "eval_recall": 0.822538860103627,
      "eval_roc_auc": 0.7378198042602189,
      "eval_runtime": 1.1765,
      "eval_samples_per_second": 2268.536,
      "eval_steps_per_second": 5.1,
      "step": 808
    },
    {
      "epoch": 77.04761904761905,
      "grad_norm": 7.566778659820557,
      "learning_rate": 1.5142428785607198e-07,
      "loss": 1.1806,
      "step": 809
    },
    {
      "epoch": 77.14285714285714,
      "grad_norm": 4.038211345672607,
      "learning_rate": 1.5161169415292354e-07,
      "loss": 1.174,
      "step": 810
    },
    {
      "epoch": 77.23809523809524,
      "grad_norm": 4.832150459289551,
      "learning_rate": 1.5179910044977512e-07,
      "loss": 1.1962,
      "step": 811
    },
    {
      "epoch": 77.33333333333333,
      "grad_norm": 3.9250776767730713,
      "learning_rate": 1.519865067466267e-07,
      "loss": 1.1681,
      "step": 812
    },
    {
      "epoch": 77.42857142857143,
      "grad_norm": 5.859079360961914,
      "learning_rate": 1.5217391304347825e-07,
      "loss": 1.1581,
      "step": 813
    },
    {
      "epoch": 77.52380952380952,
      "grad_norm": 5.626636981964111,
      "learning_rate": 1.5236131934032986e-07,
      "loss": 1.1526,
      "step": 814
    },
    {
      "epoch": 77.61904761904762,
      "grad_norm": 2.9289772510528564,
      "learning_rate": 1.5254872563718141e-07,
      "loss": 1.1899,
      "step": 815
    },
    {
      "epoch": 77.71428571428571,
      "grad_norm": 3.3265268802642822,
      "learning_rate": 1.52736131934033e-07,
      "loss": 1.1873,
      "step": 816
    },
    {
      "epoch": 77.80952380952381,
      "grad_norm": 3.9200990200042725,
      "learning_rate": 1.5292353823088455e-07,
      "loss": 1.1851,
      "step": 817
    },
    {
      "epoch": 77.9047619047619,
      "grad_norm": 4.96051549911499,
      "learning_rate": 1.5311094452773613e-07,
      "loss": 1.209,
      "step": 818
    },
    {
      "epoch": 78.0,
      "grad_norm": 4.609013557434082,
      "learning_rate": 1.532983508245877e-07,
      "loss": 1.2238,
      "step": 819
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.6991382540277258,
      "eval_f1": 0.7584962406015038,
      "eval_loss": 0.5908147096633911,
      "eval_precision": 0.708029197080292,
      "eval_recall": 0.8167098445595855,
      "eval_roc_auc": 0.7379674726540012,
      "eval_runtime": 1.328,
      "eval_samples_per_second": 2009.856,
      "eval_steps_per_second": 4.518,
      "step": 819
    },
    {
      "epoch": 78.0952380952381,
      "grad_norm": 6.584000587463379,
      "learning_rate": 1.534857571214393e-07,
      "loss": 1.197,
      "step": 820
    },
    {
      "epoch": 78.19047619047619,
      "grad_norm": 4.348421573638916,
      "learning_rate": 1.5367316341829087e-07,
      "loss": 1.1639,
      "step": 821
    },
    {
      "epoch": 78.28571428571429,
      "grad_norm": 5.262395858764648,
      "learning_rate": 1.5386056971514243e-07,
      "loss": 1.2254,
      "step": 822
    },
    {
      "epoch": 78.38095238095238,
      "grad_norm": 4.977214813232422,
      "learning_rate": 1.54047976011994e-07,
      "loss": 1.1838,
      "step": 823
    },
    {
      "epoch": 78.47619047619048,
      "grad_norm": 6.306005001068115,
      "learning_rate": 1.5423538230884556e-07,
      "loss": 1.1657,
      "step": 824
    },
    {
      "epoch": 78.57142857142857,
      "grad_norm": 4.172593593597412,
      "learning_rate": 1.5442278860569717e-07,
      "loss": 1.1989,
      "step": 825
    },
    {
      "epoch": 78.66666666666667,
      "grad_norm": 4.104514122009277,
      "learning_rate": 1.5461019490254875e-07,
      "loss": 1.1957,
      "step": 826
    },
    {
      "epoch": 78.76190476190476,
      "grad_norm": 11.191329002380371,
      "learning_rate": 1.547976011994003e-07,
      "loss": 1.1628,
      "step": 827
    },
    {
      "epoch": 78.85714285714286,
      "grad_norm": 3.7192697525024414,
      "learning_rate": 1.549850074962519e-07,
      "loss": 1.1801,
      "step": 828
    },
    {
      "epoch": 78.95238095238095,
      "grad_norm": 4.374116897583008,
      "learning_rate": 1.5517241379310344e-07,
      "loss": 1.1667,
      "step": 829
    },
    {
      "epoch": 78.95238095238095,
      "eval_accuracy": 0.6980142375421506,
      "eval_f1": 0.758248350329934,
      "eval_loss": 0.590340793132782,
      "eval_precision": 0.7061452513966481,
      "eval_recall": 0.8186528497409327,
      "eval_roc_auc": 0.7383906160046056,
      "eval_runtime": 1.1274,
      "eval_samples_per_second": 2367.487,
      "eval_steps_per_second": 5.322,
      "step": 829
    },
    {
      "epoch": 79.04761904761905,
      "grad_norm": 6.362501621246338,
      "learning_rate": 1.5535982008995502e-07,
      "loss": 1.1866,
      "step": 830
    },
    {
      "epoch": 79.14285714285714,
      "grad_norm": 7.911940097808838,
      "learning_rate": 1.555472263868066e-07,
      "loss": 1.1772,
      "step": 831
    },
    {
      "epoch": 79.23809523809524,
      "grad_norm": 7.175947666168213,
      "learning_rate": 1.5573463268365819e-07,
      "loss": 1.1564,
      "step": 832
    },
    {
      "epoch": 79.33333333333333,
      "grad_norm": 10.416701316833496,
      "learning_rate": 1.5592203898050977e-07,
      "loss": 1.2142,
      "step": 833
    },
    {
      "epoch": 79.42857142857143,
      "grad_norm": 8.933462142944336,
      "learning_rate": 1.5610944527736132e-07,
      "loss": 1.2047,
      "step": 834
    },
    {
      "epoch": 79.52380952380952,
      "grad_norm": 4.097228527069092,
      "learning_rate": 1.562968515742129e-07,
      "loss": 1.1948,
      "step": 835
    },
    {
      "epoch": 79.61904761904762,
      "grad_norm": 3.429769277572632,
      "learning_rate": 1.5648425787106446e-07,
      "loss": 1.1977,
      "step": 836
    },
    {
      "epoch": 79.71428571428571,
      "grad_norm": 3.96598744392395,
      "learning_rate": 1.5667166416791606e-07,
      "loss": 1.1859,
      "step": 837
    },
    {
      "epoch": 79.80952380952381,
      "grad_norm": 2.8495140075683594,
      "learning_rate": 1.5685907046476762e-07,
      "loss": 1.1776,
      "step": 838
    },
    {
      "epoch": 79.9047619047619,
      "grad_norm": 5.458662509918213,
      "learning_rate": 1.570464767616192e-07,
      "loss": 1.1592,
      "step": 839
    },
    {
      "epoch": 80.0,
      "grad_norm": 4.803214073181152,
      "learning_rate": 1.5723388305847078e-07,
      "loss": 1.2071,
      "step": 840
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.7002622705133008,
      "eval_f1": 0.7593261131167268,
      "eval_loss": 0.5900521874427795,
      "eval_precision": 0.7089887640449438,
      "eval_recall": 0.8173575129533679,
      "eval_roc_auc": 0.7384001151410478,
      "eval_runtime": 1.2067,
      "eval_samples_per_second": 2211.756,
      "eval_steps_per_second": 4.972,
      "step": 840
    },
    {
      "epoch": 80.0952380952381,
      "grad_norm": 3.455930709838867,
      "learning_rate": 1.5742128935532233e-07,
      "loss": 1.1699,
      "step": 841
    },
    {
      "epoch": 80.19047619047619,
      "grad_norm": 4.584237575531006,
      "learning_rate": 1.5760869565217392e-07,
      "loss": 1.2239,
      "step": 842
    },
    {
      "epoch": 80.28571428571429,
      "grad_norm": 2.777982711791992,
      "learning_rate": 1.577961019490255e-07,
      "loss": 1.1637,
      "step": 843
    },
    {
      "epoch": 80.38095238095238,
      "grad_norm": 4.465492248535156,
      "learning_rate": 1.5798350824587708e-07,
      "loss": 1.195,
      "step": 844
    },
    {
      "epoch": 80.47619047619048,
      "grad_norm": 4.4649481773376465,
      "learning_rate": 1.5817091454272863e-07,
      "loss": 1.1838,
      "step": 845
    },
    {
      "epoch": 80.57142857142857,
      "grad_norm": 4.6688337326049805,
      "learning_rate": 1.583583208395802e-07,
      "loss": 1.2007,
      "step": 846
    },
    {
      "epoch": 80.66666666666667,
      "grad_norm": 2.963181257247925,
      "learning_rate": 1.585457271364318e-07,
      "loss": 1.1718,
      "step": 847
    },
    {
      "epoch": 80.76190476190476,
      "grad_norm": 6.0897979736328125,
      "learning_rate": 1.5873313343328337e-07,
      "loss": 1.2046,
      "step": 848
    },
    {
      "epoch": 80.85714285714286,
      "grad_norm": 3.63608717918396,
      "learning_rate": 1.5892053973013496e-07,
      "loss": 1.1933,
      "step": 849
    },
    {
      "epoch": 80.95238095238095,
      "grad_norm": 3.738607883453369,
      "learning_rate": 1.591079460269865e-07,
      "loss": 1.2184,
      "step": 850
    },
    {
      "epoch": 80.95238095238095,
      "eval_accuracy": 0.6991382540277258,
      "eval_f1": 0.758205359831376,
      "eval_loss": 0.5898799300193787,
      "eval_precision": 0.7084974676420934,
      "eval_recall": 0.8154145077720207,
      "eval_roc_auc": 0.7383805411629245,
      "eval_runtime": 1.1853,
      "eval_samples_per_second": 2251.813,
      "eval_steps_per_second": 5.062,
      "step": 850
    },
    {
      "epoch": 81.04761904761905,
      "grad_norm": 4.331661701202393,
      "learning_rate": 1.592953523238381e-07,
      "loss": 1.1899,
      "step": 851
    },
    {
      "epoch": 81.14285714285714,
      "grad_norm": 5.725168704986572,
      "learning_rate": 1.5948275862068967e-07,
      "loss": 1.1629,
      "step": 852
    },
    {
      "epoch": 81.23809523809524,
      "grad_norm": 6.6885504722595215,
      "learning_rate": 1.5967016491754123e-07,
      "loss": 1.1414,
      "step": 853
    },
    {
      "epoch": 81.33333333333333,
      "grad_norm": 4.884006977081299,
      "learning_rate": 1.598575712143928e-07,
      "loss": 1.2167,
      "step": 854
    },
    {
      "epoch": 81.42857142857143,
      "grad_norm": 4.00887393951416,
      "learning_rate": 1.600449775112444e-07,
      "loss": 1.194,
      "step": 855
    },
    {
      "epoch": 81.52380952380952,
      "grad_norm": 9.518308639526367,
      "learning_rate": 1.6023238380809597e-07,
      "loss": 1.2198,
      "step": 856
    },
    {
      "epoch": 81.61904761904762,
      "grad_norm": 4.420541286468506,
      "learning_rate": 1.6041979010494752e-07,
      "loss": 1.1623,
      "step": 857
    },
    {
      "epoch": 81.71428571428571,
      "grad_norm": 4.352540016174316,
      "learning_rate": 1.606071964017991e-07,
      "loss": 1.1953,
      "step": 858
    },
    {
      "epoch": 81.80952380952381,
      "grad_norm": 4.800368309020996,
      "learning_rate": 1.6079460269865069e-07,
      "loss": 1.2255,
      "step": 859
    },
    {
      "epoch": 81.9047619047619,
      "grad_norm": 9.488035202026367,
      "learning_rate": 1.6098200899550227e-07,
      "loss": 1.1621,
      "step": 860
    },
    {
      "epoch": 82.0,
      "grad_norm": 2.658348321914673,
      "learning_rate": 1.6116941529235385e-07,
      "loss": 1.1856,
      "step": 861
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.6995129261895842,
      "eval_f1": 0.7597363690832835,
      "eval_loss": 0.5896615386009216,
      "eval_precision": 0.7068004459308808,
      "eval_recall": 0.8212435233160622,
      "eval_roc_auc": 0.7387081174438688,
      "eval_runtime": 1.3138,
      "eval_samples_per_second": 2031.445,
      "eval_steps_per_second": 4.567,
      "step": 861
    },
    {
      "epoch": 82.0952380952381,
      "grad_norm": 4.096118927001953,
      "learning_rate": 1.613568215892054e-07,
      "loss": 1.168,
      "step": 862
    },
    {
      "epoch": 82.19047619047619,
      "grad_norm": 3.9486894607543945,
      "learning_rate": 1.6154422788605698e-07,
      "loss": 1.2038,
      "step": 863
    },
    {
      "epoch": 82.28571428571429,
      "grad_norm": 7.675639629364014,
      "learning_rate": 1.6173163418290854e-07,
      "loss": 1.2139,
      "step": 864
    },
    {
      "epoch": 82.38095238095238,
      "grad_norm": 3.3430981636047363,
      "learning_rate": 1.6191904047976012e-07,
      "loss": 1.1643,
      "step": 865
    },
    {
      "epoch": 82.47619047619048,
      "grad_norm": 4.858800888061523,
      "learning_rate": 1.621064467766117e-07,
      "loss": 1.2191,
      "step": 866
    },
    {
      "epoch": 82.57142857142857,
      "grad_norm": 3.3145360946655273,
      "learning_rate": 1.6229385307346328e-07,
      "loss": 1.1903,
      "step": 867
    },
    {
      "epoch": 82.66666666666667,
      "grad_norm": 2.669579267501831,
      "learning_rate": 1.6248125937031486e-07,
      "loss": 1.1716,
      "step": 868
    },
    {
      "epoch": 82.76190476190476,
      "grad_norm": 5.312630653381348,
      "learning_rate": 1.6266866566716642e-07,
      "loss": 1.1918,
      "step": 869
    },
    {
      "epoch": 82.85714285714286,
      "grad_norm": 9.223489761352539,
      "learning_rate": 1.62856071964018e-07,
      "loss": 1.185,
      "step": 870
    },
    {
      "epoch": 82.95238095238095,
      "grad_norm": 6.911379337310791,
      "learning_rate": 1.6304347826086955e-07,
      "loss": 1.1429,
      "step": 871
    },
    {
      "epoch": 82.95238095238095,
      "eval_accuracy": 0.6995129261895842,
      "eval_f1": 0.7604540023894862,
      "eval_loss": 0.5893028974533081,
      "eval_precision": 0.7056541019955654,
      "eval_recall": 0.8244818652849741,
      "eval_roc_auc": 0.7389502014968335,
      "eval_runtime": 1.1376,
      "eval_samples_per_second": 2346.142,
      "eval_steps_per_second": 5.274,
      "step": 871
    },
    {
      "epoch": 83.04761904761905,
      "grad_norm": 4.5483174324035645,
      "learning_rate": 1.6323088455772116e-07,
      "loss": 1.2073,
      "step": 872
    },
    {
      "epoch": 83.14285714285714,
      "grad_norm": 6.0994062423706055,
      "learning_rate": 1.6341829085457274e-07,
      "loss": 1.1722,
      "step": 873
    },
    {
      "epoch": 83.23809523809524,
      "grad_norm": 3.3759844303131104,
      "learning_rate": 1.636056971514243e-07,
      "loss": 1.1758,
      "step": 874
    },
    {
      "epoch": 83.33333333333333,
      "grad_norm": 4.4953227043151855,
      "learning_rate": 1.6379310344827587e-07,
      "loss": 1.2061,
      "step": 875
    },
    {
      "epoch": 83.42857142857143,
      "grad_norm": 4.761415958404541,
      "learning_rate": 1.6398050974512743e-07,
      "loss": 1.1749,
      "step": 876
    },
    {
      "epoch": 83.52380952380952,
      "grad_norm": 9.842004776000977,
      "learning_rate": 1.64167916041979e-07,
      "loss": 1.2147,
      "step": 877
    },
    {
      "epoch": 83.61904761904762,
      "grad_norm": 4.910158634185791,
      "learning_rate": 1.643553223388306e-07,
      "loss": 1.1608,
      "step": 878
    },
    {
      "epoch": 83.71428571428571,
      "grad_norm": 3.1456174850463867,
      "learning_rate": 1.6454272863568217e-07,
      "loss": 1.1959,
      "step": 879
    },
    {
      "epoch": 83.80952380952381,
      "grad_norm": 3.7636923789978027,
      "learning_rate": 1.6473013493253375e-07,
      "loss": 1.169,
      "step": 880
    },
    {
      "epoch": 83.9047619047619,
      "grad_norm": 2.942305564880371,
      "learning_rate": 1.649175412293853e-07,
      "loss": 1.1658,
      "step": 881
    },
    {
      "epoch": 84.0,
      "grad_norm": 4.029232025146484,
      "learning_rate": 1.651049475262369e-07,
      "loss": 1.1984,
      "step": 882
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.6987635818658674,
      "eval_f1": 0.7591372079089275,
      "eval_loss": 0.5888885259628296,
      "eval_precision": 0.7062430323299889,
      "eval_recall": 0.8205958549222798,
      "eval_roc_auc": 0.739222797927461,
      "eval_runtime": 1.1564,
      "eval_samples_per_second": 2308.117,
      "eval_steps_per_second": 5.189,
      "step": 882
    },
    {
      "epoch": 84.0952380952381,
      "grad_norm": 5.96453857421875,
      "learning_rate": 1.6529235382308847e-07,
      "loss": 1.16,
      "step": 883
    },
    {
      "epoch": 84.19047619047619,
      "grad_norm": 3.663802146911621,
      "learning_rate": 1.6547976011994005e-07,
      "loss": 1.192,
      "step": 884
    },
    {
      "epoch": 84.28571428571429,
      "grad_norm": 6.514193534851074,
      "learning_rate": 1.656671664167916e-07,
      "loss": 1.1417,
      "step": 885
    },
    {
      "epoch": 84.38095238095238,
      "grad_norm": 2.786970376968384,
      "learning_rate": 1.6585457271364319e-07,
      "loss": 1.1705,
      "step": 886
    },
    {
      "epoch": 84.47619047619048,
      "grad_norm": 4.917025566101074,
      "learning_rate": 1.6604197901049477e-07,
      "loss": 1.1819,
      "step": 887
    },
    {
      "epoch": 84.57142857142857,
      "grad_norm": 7.882221221923828,
      "learning_rate": 1.6622938530734632e-07,
      "loss": 1.1895,
      "step": 888
    },
    {
      "epoch": 84.66666666666667,
      "grad_norm": 3.4152791500091553,
      "learning_rate": 1.664167916041979e-07,
      "loss": 1.1616,
      "step": 889
    },
    {
      "epoch": 84.76190476190476,
      "grad_norm": 4.9652485847473145,
      "learning_rate": 1.6660419790104948e-07,
      "loss": 1.1608,
      "step": 890
    },
    {
      "epoch": 84.85714285714286,
      "grad_norm": 4.204701900482178,
      "learning_rate": 1.6679160419790106e-07,
      "loss": 1.1934,
      "step": 891
    },
    {
      "epoch": 84.95238095238095,
      "grad_norm": 6.280852794647217,
      "learning_rate": 1.6697901049475262e-07,
      "loss": 1.1434,
      "step": 892
    },
    {
      "epoch": 84.95238095238095,
      "eval_accuracy": 0.701386286998876,
      "eval_f1": 0.7617339312406577,
      "eval_loss": 0.5886748433113098,
      "eval_precision": 0.7073847862298723,
      "eval_recall": 0.8251295336787565,
      "eval_roc_auc": 0.7393137593552102,
      "eval_runtime": 1.2673,
      "eval_samples_per_second": 2106.034,
      "eval_steps_per_second": 4.734,
      "step": 892
    },
    {
      "epoch": 85.04761904761905,
      "grad_norm": 3.2578413486480713,
      "learning_rate": 1.671664167916042e-07,
      "loss": 1.1724,
      "step": 893
    },
    {
      "epoch": 85.14285714285714,
      "grad_norm": 7.428417682647705,
      "learning_rate": 1.6735382308845578e-07,
      "loss": 1.2087,
      "step": 894
    },
    {
      "epoch": 85.23809523809524,
      "grad_norm": 6.177584171295166,
      "learning_rate": 1.6754122938530736e-07,
      "loss": 1.2465,
      "step": 895
    },
    {
      "epoch": 85.33333333333333,
      "grad_norm": 3.587831497192383,
      "learning_rate": 1.6772863568215894e-07,
      "loss": 1.2009,
      "step": 896
    },
    {
      "epoch": 85.42857142857143,
      "grad_norm": 5.211238384246826,
      "learning_rate": 1.679160419790105e-07,
      "loss": 1.1999,
      "step": 897
    },
    {
      "epoch": 85.52380952380952,
      "grad_norm": 3.0934226512908936,
      "learning_rate": 1.6810344827586208e-07,
      "loss": 1.19,
      "step": 898
    },
    {
      "epoch": 85.61904761904762,
      "grad_norm": 10.23063850402832,
      "learning_rate": 1.6829085457271363e-07,
      "loss": 1.1526,
      "step": 899
    },
    {
      "epoch": 85.71428571428571,
      "grad_norm": 3.947057008743286,
      "learning_rate": 1.684782608695652e-07,
      "loss": 1.155,
      "step": 900
    },
    {
      "epoch": 85.80952380952381,
      "grad_norm": 3.20998477935791,
      "learning_rate": 1.6866566716641682e-07,
      "loss": 1.1781,
      "step": 901
    },
    {
      "epoch": 85.9047619047619,
      "grad_norm": 4.5016093254089355,
      "learning_rate": 1.6885307346326838e-07,
      "loss": 1.1671,
      "step": 902
    },
    {
      "epoch": 86.0,
      "grad_norm": 4.235739707946777,
      "learning_rate": 1.6904047976011996e-07,
      "loss": 1.2128,
      "step": 903
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.6998875983514425,
      "eval_f1": 0.7603948549207299,
      "eval_loss": 0.5884115695953369,
      "eval_precision": 0.7065036131183992,
      "eval_recall": 0.8231865284974094,
      "eval_roc_auc": 0.7395020149683362,
      "eval_runtime": 1.2146,
      "eval_samples_per_second": 2197.493,
      "eval_steps_per_second": 4.94,
      "step": 903
    },
    {
      "epoch": 86.0952380952381,
      "grad_norm": 3.380007266998291,
      "learning_rate": 1.692278860569715e-07,
      "loss": 1.1776,
      "step": 904
    },
    {
      "epoch": 86.19047619047619,
      "grad_norm": 3.720041036605835,
      "learning_rate": 1.694152923538231e-07,
      "loss": 1.1642,
      "step": 905
    },
    {
      "epoch": 86.28571428571429,
      "grad_norm": 7.0945234298706055,
      "learning_rate": 1.6960269865067467e-07,
      "loss": 1.2604,
      "step": 906
    },
    {
      "epoch": 86.38095238095238,
      "grad_norm": 5.684896469116211,
      "learning_rate": 1.6979010494752625e-07,
      "loss": 1.1482,
      "step": 907
    },
    {
      "epoch": 86.47619047619048,
      "grad_norm": 5.6247029304504395,
      "learning_rate": 1.6997751124437783e-07,
      "loss": 1.1838,
      "step": 908
    },
    {
      "epoch": 86.57142857142857,
      "grad_norm": 7.488827705383301,
      "learning_rate": 1.701649175412294e-07,
      "loss": 1.1694,
      "step": 909
    },
    {
      "epoch": 86.66666666666667,
      "grad_norm": 3.2675230503082275,
      "learning_rate": 1.7035232383808097e-07,
      "loss": 1.1989,
      "step": 910
    },
    {
      "epoch": 86.76190476190476,
      "grad_norm": 5.990199565887451,
      "learning_rate": 1.7053973013493252e-07,
      "loss": 1.2253,
      "step": 911
    },
    {
      "epoch": 86.85714285714286,
      "grad_norm": 4.462316513061523,
      "learning_rate": 1.707271364317841e-07,
      "loss": 1.1602,
      "step": 912
    },
    {
      "epoch": 86.95238095238095,
      "grad_norm": 6.762936592102051,
      "learning_rate": 1.709145427286357e-07,
      "loss": 1.1828,
      "step": 913
    },
    {
      "epoch": 86.95238095238095,
      "eval_accuracy": 0.7021356313225927,
      "eval_f1": 0.7624738571855393,
      "eval_loss": 0.5883358120918274,
      "eval_precision": 0.7077093732667776,
      "eval_recall": 0.8264248704663213,
      "eval_roc_auc": 0.7396272308578008,
      "eval_runtime": 1.3325,
      "eval_samples_per_second": 2002.931,
      "eval_steps_per_second": 4.503,
      "step": 913
    },
    {
      "epoch": 87.04761904761905,
      "grad_norm": 3.4856576919555664,
      "learning_rate": 1.7110194902548727e-07,
      "loss": 1.1861,
      "step": 914
    },
    {
      "epoch": 87.14285714285714,
      "grad_norm": 5.72012996673584,
      "learning_rate": 1.7128935532233885e-07,
      "loss": 1.1541,
      "step": 915
    },
    {
      "epoch": 87.23809523809524,
      "grad_norm": 2.709678888320923,
      "learning_rate": 1.714767616191904e-07,
      "loss": 1.2146,
      "step": 916
    },
    {
      "epoch": 87.33333333333333,
      "grad_norm": 3.5865318775177,
      "learning_rate": 1.7166416791604198e-07,
      "loss": 1.2038,
      "step": 917
    },
    {
      "epoch": 87.42857142857143,
      "grad_norm": 3.717111349105835,
      "learning_rate": 1.7185157421289356e-07,
      "loss": 1.152,
      "step": 918
    },
    {
      "epoch": 87.52380952380952,
      "grad_norm": 5.886796474456787,
      "learning_rate": 1.7203898050974515e-07,
      "loss": 1.2045,
      "step": 919
    },
    {
      "epoch": 87.61904761904762,
      "grad_norm": 5.211839199066162,
      "learning_rate": 1.7222638680659673e-07,
      "loss": 1.1633,
      "step": 920
    },
    {
      "epoch": 87.71428571428571,
      "grad_norm": 4.2755842208862305,
      "learning_rate": 1.7241379310344828e-07,
      "loss": 1.1599,
      "step": 921
    },
    {
      "epoch": 87.80952380952381,
      "grad_norm": 5.807846546173096,
      "learning_rate": 1.7260119940029986e-07,
      "loss": 1.2609,
      "step": 922
    },
    {
      "epoch": 87.9047619047619,
      "grad_norm": 6.997453212738037,
      "learning_rate": 1.7278860569715142e-07,
      "loss": 1.1704,
      "step": 923
    },
    {
      "epoch": 88.0,
      "grad_norm": 6.106979846954346,
      "learning_rate": 1.72976011994003e-07,
      "loss": 1.162,
      "step": 924
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.7006369426751592,
      "eval_f1": 0.7602760276027603,
      "eval_loss": 0.5882436037063599,
      "eval_precision": 0.7082168809390721,
      "eval_recall": 0.8205958549222798,
      "eval_roc_auc": 0.739506908462867,
      "eval_runtime": 1.1639,
      "eval_samples_per_second": 2293.21,
      "eval_steps_per_second": 5.155,
      "step": 924
    },
    {
      "epoch": 88.0952380952381,
      "grad_norm": 2.476592779159546,
      "learning_rate": 1.7316341829085458e-07,
      "loss": 1.1604,
      "step": 925
    },
    {
      "epoch": 88.19047619047619,
      "grad_norm": 4.160835266113281,
      "learning_rate": 1.7335082458770616e-07,
      "loss": 1.1682,
      "step": 926
    },
    {
      "epoch": 88.28571428571429,
      "grad_norm": 5.423745155334473,
      "learning_rate": 1.7353823088455774e-07,
      "loss": 1.1931,
      "step": 927
    },
    {
      "epoch": 88.38095238095238,
      "grad_norm": 3.820173740386963,
      "learning_rate": 1.737256371814093e-07,
      "loss": 1.1994,
      "step": 928
    },
    {
      "epoch": 88.47619047619048,
      "grad_norm": 4.19473934173584,
      "learning_rate": 1.7391304347826088e-07,
      "loss": 1.211,
      "step": 929
    },
    {
      "epoch": 88.57142857142857,
      "grad_norm": 5.371788501739502,
      "learning_rate": 1.7410044977511246e-07,
      "loss": 1.1899,
      "step": 930
    },
    {
      "epoch": 88.66666666666667,
      "grad_norm": 7.475896835327148,
      "learning_rate": 1.7428785607196404e-07,
      "loss": 1.1901,
      "step": 931
    },
    {
      "epoch": 88.76190476190476,
      "grad_norm": 2.701341390609741,
      "learning_rate": 1.744752623688156e-07,
      "loss": 1.1609,
      "step": 932
    },
    {
      "epoch": 88.85714285714286,
      "grad_norm": 8.771926879882812,
      "learning_rate": 1.7466266866566717e-07,
      "loss": 1.2532,
      "step": 933
    },
    {
      "epoch": 88.95238095238095,
      "grad_norm": 3.1728227138519287,
      "learning_rate": 1.7485007496251875e-07,
      "loss": 1.1807,
      "step": 934
    },
    {
      "epoch": 88.95238095238095,
      "eval_accuracy": 0.704383664293743,
      "eval_f1": 0.7652484379648914,
      "eval_loss": 0.5880123972892761,
      "eval_precision": 0.7077600440286186,
      "eval_recall": 0.832901554404145,
      "eval_roc_auc": 0.7398719055843408,
      "eval_runtime": 1.4168,
      "eval_samples_per_second": 1883.871,
      "eval_steps_per_second": 4.235,
      "step": 934
    },
    {
      "epoch": 89.04761904761905,
      "grad_norm": 4.075814723968506,
      "learning_rate": 1.750374812593703e-07,
      "loss": 1.1809,
      "step": 935
    },
    {
      "epoch": 89.14285714285714,
      "grad_norm": 3.444646120071411,
      "learning_rate": 1.7522488755622192e-07,
      "loss": 1.1393,
      "step": 936
    },
    {
      "epoch": 89.23809523809524,
      "grad_norm": 8.94544792175293,
      "learning_rate": 1.7541229385307347e-07,
      "loss": 1.2097,
      "step": 937
    },
    {
      "epoch": 89.33333333333333,
      "grad_norm": 4.065417766571045,
      "learning_rate": 1.7559970014992505e-07,
      "loss": 1.199,
      "step": 938
    },
    {
      "epoch": 89.42857142857143,
      "grad_norm": 7.621069431304932,
      "learning_rate": 1.757871064467766e-07,
      "loss": 1.1531,
      "step": 939
    },
    {
      "epoch": 89.52380952380952,
      "grad_norm": 3.296189069747925,
      "learning_rate": 1.7597451274362819e-07,
      "loss": 1.172,
      "step": 940
    },
    {
      "epoch": 89.61904761904762,
      "grad_norm": 2.939818859100342,
      "learning_rate": 1.7616191904047977e-07,
      "loss": 1.1395,
      "step": 941
    },
    {
      "epoch": 89.71428571428571,
      "grad_norm": 3.8071563243865967,
      "learning_rate": 1.7634932533733135e-07,
      "loss": 1.1599,
      "step": 942
    },
    {
      "epoch": 89.80952380952381,
      "grad_norm": 5.116278171539307,
      "learning_rate": 1.7653673163418293e-07,
      "loss": 1.2588,
      "step": 943
    },
    {
      "epoch": 89.9047619047619,
      "grad_norm": 3.6566922664642334,
      "learning_rate": 1.7672413793103448e-07,
      "loss": 1.1885,
      "step": 944
    },
    {
      "epoch": 90.0,
      "grad_norm": 4.694438457489014,
      "learning_rate": 1.7691154422788606e-07,
      "loss": 1.2195,
      "step": 945
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.7040089921318846,
      "eval_f1": 0.7633313361294188,
      "eval_loss": 0.5877653956413269,
      "eval_precision": 0.7101449275362319,
      "eval_recall": 0.8251295336787565,
      "eval_roc_auc": 0.7399142199194012,
      "eval_runtime": 1.2109,
      "eval_samples_per_second": 2204.135,
      "eval_steps_per_second": 4.955,
      "step": 945
    },
    {
      "epoch": 90.0952380952381,
      "grad_norm": 5.013833522796631,
      "learning_rate": 1.7709895052473762e-07,
      "loss": 1.1197,
      "step": 946
    },
    {
      "epoch": 90.19047619047619,
      "grad_norm": 6.55540657043457,
      "learning_rate": 1.772863568215892e-07,
      "loss": 1.1313,
      "step": 947
    },
    {
      "epoch": 90.28571428571429,
      "grad_norm": 3.288936138153076,
      "learning_rate": 1.774737631184408e-07,
      "loss": 1.1893,
      "step": 948
    },
    {
      "epoch": 90.38095238095238,
      "grad_norm": 4.897750377655029,
      "learning_rate": 1.7766116941529236e-07,
      "loss": 1.1797,
      "step": 949
    },
    {
      "epoch": 90.47619047619048,
      "grad_norm": 5.5732903480529785,
      "learning_rate": 1.7784857571214394e-07,
      "loss": 1.2339,
      "step": 950
    },
    {
      "epoch": 90.57142857142857,
      "grad_norm": 5.4378767013549805,
      "learning_rate": 1.780359820089955e-07,
      "loss": 1.2078,
      "step": 951
    },
    {
      "epoch": 90.66666666666667,
      "grad_norm": 4.127724647521973,
      "learning_rate": 1.7822338830584708e-07,
      "loss": 1.1732,
      "step": 952
    },
    {
      "epoch": 90.76190476190476,
      "grad_norm": 4.829543590545654,
      "learning_rate": 1.7841079460269866e-07,
      "loss": 1.2367,
      "step": 953
    },
    {
      "epoch": 90.85714285714286,
      "grad_norm": 3.430927038192749,
      "learning_rate": 1.7859820089955024e-07,
      "loss": 1.1363,
      "step": 954
    },
    {
      "epoch": 90.95238095238095,
      "grad_norm": 3.5088884830474854,
      "learning_rate": 1.7878560719640182e-07,
      "loss": 1.1859,
      "step": 955
    },
    {
      "epoch": 90.95238095238095,
      "eval_accuracy": 0.7047583364556014,
      "eval_f1": 0.7649164677804295,
      "eval_loss": 0.5876271724700928,
      "eval_precision": 0.709070796460177,
      "eval_recall": 0.8303108808290155,
      "eval_roc_auc": 0.740163788140472,
      "eval_runtime": 1.2241,
      "eval_samples_per_second": 2180.311,
      "eval_steps_per_second": 4.901,
      "step": 955
    },
    {
      "epoch": 91.04761904761905,
      "grad_norm": 4.03278923034668,
      "learning_rate": 1.7897301349325338e-07,
      "loss": 1.1953,
      "step": 956
    },
    {
      "epoch": 91.14285714285714,
      "grad_norm": 4.45944356918335,
      "learning_rate": 1.7916041979010496e-07,
      "loss": 1.1832,
      "step": 957
    },
    {
      "epoch": 91.23809523809524,
      "grad_norm": 5.994672775268555,
      "learning_rate": 1.793478260869565e-07,
      "loss": 1.1691,
      "step": 958
    },
    {
      "epoch": 91.33333333333333,
      "grad_norm": 5.616822719573975,
      "learning_rate": 1.7953523238380812e-07,
      "loss": 1.1598,
      "step": 959
    },
    {
      "epoch": 91.42857142857143,
      "grad_norm": 5.643009662628174,
      "learning_rate": 1.7972263868065967e-07,
      "loss": 1.1954,
      "step": 960
    },
    {
      "epoch": 91.52380952380952,
      "grad_norm": 3.8720247745513916,
      "learning_rate": 1.7991004497751125e-07,
      "loss": 1.1931,
      "step": 961
    },
    {
      "epoch": 91.61904761904762,
      "grad_norm": 4.31902551651001,
      "learning_rate": 1.8009745127436284e-07,
      "loss": 1.1827,
      "step": 962
    },
    {
      "epoch": 91.71428571428571,
      "grad_norm": 4.202095031738281,
      "learning_rate": 1.802848575712144e-07,
      "loss": 1.159,
      "step": 963
    },
    {
      "epoch": 91.80952380952381,
      "grad_norm": 5.559122562408447,
      "learning_rate": 1.8047226386806597e-07,
      "loss": 1.1767,
      "step": 964
    },
    {
      "epoch": 91.9047619047619,
      "grad_norm": 2.7115461826324463,
      "learning_rate": 1.8065967016491755e-07,
      "loss": 1.1845,
      "step": 965
    },
    {
      "epoch": 92.0,
      "grad_norm": 3.8898439407348633,
      "learning_rate": 1.8084707646176913e-07,
      "loss": 1.1827,
      "step": 966
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.7055076807793181,
      "eval_f1": 0.7655131264916468,
      "eval_loss": 0.5875535607337952,
      "eval_precision": 0.7096238938053098,
      "eval_recall": 0.8309585492227979,
      "eval_roc_auc": 0.7402328727691423,
      "eval_runtime": 1.2339,
      "eval_samples_per_second": 2162.998,
      "eval_steps_per_second": 4.862,
      "step": 966
    },
    {
      "epoch": 92.0952380952381,
      "grad_norm": 4.475981712341309,
      "learning_rate": 1.8103448275862071e-07,
      "loss": 1.226,
      "step": 967
    },
    {
      "epoch": 92.19047619047619,
      "grad_norm": 12.51380729675293,
      "learning_rate": 1.8122188905547227e-07,
      "loss": 1.1773,
      "step": 968
    },
    {
      "epoch": 92.28571428571429,
      "grad_norm": 3.1769425868988037,
      "learning_rate": 1.8140929535232385e-07,
      "loss": 1.2155,
      "step": 969
    },
    {
      "epoch": 92.38095238095238,
      "grad_norm": 6.938444137573242,
      "learning_rate": 1.815967016491754e-07,
      "loss": 1.1839,
      "step": 970
    },
    {
      "epoch": 92.47619047619048,
      "grad_norm": 5.644467353820801,
      "learning_rate": 1.81784107946027e-07,
      "loss": 1.1665,
      "step": 971
    },
    {
      "epoch": 92.57142857142857,
      "grad_norm": 4.443206310272217,
      "learning_rate": 1.8197151424287857e-07,
      "loss": 1.1653,
      "step": 972
    },
    {
      "epoch": 92.66666666666667,
      "grad_norm": 4.741915702819824,
      "learning_rate": 1.8215892053973015e-07,
      "loss": 1.1543,
      "step": 973
    },
    {
      "epoch": 92.76190476190476,
      "grad_norm": 3.8203067779541016,
      "learning_rate": 1.8234632683658173e-07,
      "loss": 1.1586,
      "step": 974
    },
    {
      "epoch": 92.85714285714286,
      "grad_norm": 7.701940536499023,
      "learning_rate": 1.8253373313343328e-07,
      "loss": 1.1927,
      "step": 975
    },
    {
      "epoch": 92.95238095238095,
      "grad_norm": 4.172492027282715,
      "learning_rate": 1.8272113943028486e-07,
      "loss": 1.234,
      "step": 976
    },
    {
      "epoch": 92.95238095238095,
      "eval_accuracy": 0.7055076807793181,
      "eval_f1": 0.7649521531100478,
      "eval_loss": 0.5876275300979614,
      "eval_precision": 0.7105555555555556,
      "eval_recall": 0.8283678756476683,
      "eval_roc_auc": 0.7400630397236615,
      "eval_runtime": 1.1616,
      "eval_samples_per_second": 2297.727,
      "eval_steps_per_second": 5.165,
      "step": 976
    },
    {
      "epoch": 93.04761904761905,
      "grad_norm": 3.0559241771698,
      "learning_rate": 1.8290854572713644e-07,
      "loss": 1.1455,
      "step": 977
    },
    {
      "epoch": 93.14285714285714,
      "grad_norm": 3.401667833328247,
      "learning_rate": 1.8309595202398802e-07,
      "loss": 1.1993,
      "step": 978
    },
    {
      "epoch": 93.23809523809524,
      "grad_norm": 4.739910125732422,
      "learning_rate": 1.8328335832083958e-07,
      "loss": 1.2022,
      "step": 979
    },
    {
      "epoch": 93.33333333333333,
      "grad_norm": 5.978082180023193,
      "learning_rate": 1.8347076461769116e-07,
      "loss": 1.225,
      "step": 980
    },
    {
      "epoch": 93.42857142857143,
      "grad_norm": 3.4651079177856445,
      "learning_rate": 1.8365817091454274e-07,
      "loss": 1.1836,
      "step": 981
    },
    {
      "epoch": 93.52380952380952,
      "grad_norm": 3.436234951019287,
      "learning_rate": 1.8384557721139432e-07,
      "loss": 1.2121,
      "step": 982
    },
    {
      "epoch": 93.61904761904762,
      "grad_norm": 2.9314863681793213,
      "learning_rate": 1.840329835082459e-07,
      "loss": 1.1723,
      "step": 983
    },
    {
      "epoch": 93.71428571428571,
      "grad_norm": 9.982884407043457,
      "learning_rate": 1.8422038980509746e-07,
      "loss": 1.186,
      "step": 984
    },
    {
      "epoch": 93.80952380952381,
      "grad_norm": 3.900665283203125,
      "learning_rate": 1.8440779610194904e-07,
      "loss": 1.164,
      "step": 985
    },
    {
      "epoch": 93.9047619047619,
      "grad_norm": 3.7296035289764404,
      "learning_rate": 1.845952023988006e-07,
      "loss": 1.1702,
      "step": 986
    },
    {
      "epoch": 94.0,
      "grad_norm": 2.609611988067627,
      "learning_rate": 1.8478260869565217e-07,
      "loss": 1.1407,
      "step": 987
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.7088797302360434,
      "eval_f1": 0.7701863354037267,
      "eval_loss": 0.58750981092453,
      "eval_precision": 0.708764289602613,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7405644789867587,
      "eval_runtime": 1.1414,
      "eval_samples_per_second": 2338.278,
      "eval_steps_per_second": 5.257,
      "step": 987
    },
    {
      "epoch": 94.0952380952381,
      "grad_norm": 5.930230617523193,
      "learning_rate": 1.8497001499250375e-07,
      "loss": 1.183,
      "step": 988
    },
    {
      "epoch": 94.19047619047619,
      "grad_norm": 3.943761110305786,
      "learning_rate": 1.8515742128935534e-07,
      "loss": 1.1921,
      "step": 989
    },
    {
      "epoch": 94.28571428571429,
      "grad_norm": 9.60708999633789,
      "learning_rate": 1.8534482758620692e-07,
      "loss": 1.2425,
      "step": 990
    },
    {
      "epoch": 94.38095238095238,
      "grad_norm": 3.7731473445892334,
      "learning_rate": 1.8553223388305847e-07,
      "loss": 1.1399,
      "step": 991
    },
    {
      "epoch": 94.47619047619048,
      "grad_norm": 6.896683216094971,
      "learning_rate": 1.8571964017991005e-07,
      "loss": 1.239,
      "step": 992
    },
    {
      "epoch": 94.57142857142857,
      "grad_norm": 4.011725425720215,
      "learning_rate": 1.859070464767616e-07,
      "loss": 1.1741,
      "step": 993
    },
    {
      "epoch": 94.66666666666667,
      "grad_norm": 2.2900426387786865,
      "learning_rate": 1.8609445277361321e-07,
      "loss": 1.1527,
      "step": 994
    },
    {
      "epoch": 94.76190476190476,
      "grad_norm": 4.162290096282959,
      "learning_rate": 1.862818590704648e-07,
      "loss": 1.1454,
      "step": 995
    },
    {
      "epoch": 94.85714285714286,
      "grad_norm": 4.327588081359863,
      "learning_rate": 1.8646926536731635e-07,
      "loss": 1.1874,
      "step": 996
    },
    {
      "epoch": 94.95238095238095,
      "grad_norm": 3.36289119720459,
      "learning_rate": 1.8665667166416793e-07,
      "loss": 1.1501,
      "step": 997
    },
    {
      "epoch": 94.95238095238095,
      "eval_accuracy": 0.7070063694267515,
      "eval_f1": 0.767814726840855,
      "eval_loss": 0.5870710015296936,
      "eval_precision": 0.7088815789473685,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7406669545192861,
      "eval_runtime": 1.2238,
      "eval_samples_per_second": 2180.834,
      "eval_steps_per_second": 4.903,
      "step": 997
    },
    {
      "epoch": 95.04761904761905,
      "grad_norm": 4.248078346252441,
      "learning_rate": 1.8684407796101948e-07,
      "loss": 1.2222,
      "step": 998
    },
    {
      "epoch": 95.14285714285714,
      "grad_norm": 5.069729328155518,
      "learning_rate": 1.8703148425787107e-07,
      "loss": 1.1808,
      "step": 999
    },
    {
      "epoch": 95.23809523809524,
      "grad_norm": 8.28036880493164,
      "learning_rate": 1.8721889055472265e-07,
      "loss": 1.2253,
      "step": 1000
    },
    {
      "epoch": 95.33333333333333,
      "grad_norm": 6.096401214599609,
      "learning_rate": 1.8740629685157423e-07,
      "loss": 1.1584,
      "step": 1001
    },
    {
      "epoch": 95.42857142857143,
      "grad_norm": 4.412257194519043,
      "learning_rate": 1.875937031484258e-07,
      "loss": 1.2083,
      "step": 1002
    },
    {
      "epoch": 95.52380952380952,
      "grad_norm": 4.404515743255615,
      "learning_rate": 1.8778110944527736e-07,
      "loss": 1.1689,
      "step": 1003
    },
    {
      "epoch": 95.61904761904762,
      "grad_norm": 3.2303812503814697,
      "learning_rate": 1.8796851574212894e-07,
      "loss": 1.1695,
      "step": 1004
    },
    {
      "epoch": 95.71428571428571,
      "grad_norm": 3.9076950550079346,
      "learning_rate": 1.8815592203898052e-07,
      "loss": 1.1857,
      "step": 1005
    },
    {
      "epoch": 95.80952380952381,
      "grad_norm": 4.176351547241211,
      "learning_rate": 1.883433283358321e-07,
      "loss": 1.1781,
      "step": 1006
    },
    {
      "epoch": 95.9047619047619,
      "grad_norm": 3.8240013122558594,
      "learning_rate": 1.8853073463268366e-07,
      "loss": 1.1576,
      "step": 1007
    },
    {
      "epoch": 96.0,
      "grad_norm": 2.9544076919555664,
      "learning_rate": 1.8871814092953524e-07,
      "loss": 1.1761,
      "step": 1008
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.7066316972648932,
      "eval_f1": 0.7667560321715817,
      "eval_loss": 0.5868284702301025,
      "eval_precision": 0.709873138444567,
      "eval_recall": 0.8335492227979274,
      "eval_roc_auc": 0.740759067357513,
      "eval_runtime": 1.2315,
      "eval_samples_per_second": 2167.19,
      "eval_steps_per_second": 4.872,
      "step": 1008
    },
    {
      "epoch": 96.0952380952381,
      "grad_norm": 6.389705657958984,
      "learning_rate": 1.8890554722638682e-07,
      "loss": 1.2402,
      "step": 1009
    },
    {
      "epoch": 96.19047619047619,
      "grad_norm": 5.934082508087158,
      "learning_rate": 1.8909295352323838e-07,
      "loss": 1.1749,
      "step": 1010
    },
    {
      "epoch": 96.28571428571429,
      "grad_norm": 3.007215976715088,
      "learning_rate": 1.8928035982008996e-07,
      "loss": 1.1969,
      "step": 1011
    },
    {
      "epoch": 96.38095238095238,
      "grad_norm": 6.05702018737793,
      "learning_rate": 1.8946776611694154e-07,
      "loss": 1.1679,
      "step": 1012
    },
    {
      "epoch": 96.47619047619048,
      "grad_norm": 6.701571941375732,
      "learning_rate": 1.8965517241379312e-07,
      "loss": 1.1731,
      "step": 1013
    },
    {
      "epoch": 96.57142857142857,
      "grad_norm": 4.297491073608398,
      "learning_rate": 1.8984257871064467e-07,
      "loss": 1.1646,
      "step": 1014
    },
    {
      "epoch": 96.66666666666667,
      "grad_norm": 5.01285982131958,
      "learning_rate": 1.9002998500749625e-07,
      "loss": 1.189,
      "step": 1015
    },
    {
      "epoch": 96.76190476190476,
      "grad_norm": 2.779428482055664,
      "learning_rate": 1.9021739130434784e-07,
      "loss": 1.1289,
      "step": 1016
    },
    {
      "epoch": 96.85714285714286,
      "grad_norm": 2.592012882232666,
      "learning_rate": 1.9040479760119942e-07,
      "loss": 1.1656,
      "step": 1017
    },
    {
      "epoch": 96.95238095238095,
      "grad_norm": 6.2044291496276855,
      "learning_rate": 1.90592203898051e-07,
      "loss": 1.1965,
      "step": 1018
    },
    {
      "epoch": 96.95238095238095,
      "eval_accuracy": 0.7062570251030349,
      "eval_f1": 0.7665276950565812,
      "eval_loss": 0.5867811441421509,
      "eval_precision": 0.7094818081587652,
      "eval_recall": 0.8335492227979274,
      "eval_roc_auc": 0.7409081750143928,
      "eval_runtime": 1.3479,
      "eval_samples_per_second": 1980.164,
      "eval_steps_per_second": 4.451,
      "step": 1018
    },
    {
      "epoch": 97.04761904761905,
      "grad_norm": 2.637268304824829,
      "learning_rate": 1.9077961019490255e-07,
      "loss": 1.1588,
      "step": 1019
    },
    {
      "epoch": 97.14285714285714,
      "grad_norm": 7.145625591278076,
      "learning_rate": 1.9096701649175413e-07,
      "loss": 1.1809,
      "step": 1020
    },
    {
      "epoch": 97.23809523809524,
      "grad_norm": 3.879804849624634,
      "learning_rate": 1.9115442278860571e-07,
      "loss": 1.1216,
      "step": 1021
    },
    {
      "epoch": 97.33333333333333,
      "grad_norm": 3.9333603382110596,
      "learning_rate": 1.9134182908545727e-07,
      "loss": 1.2074,
      "step": 1022
    },
    {
      "epoch": 97.42857142857143,
      "grad_norm": 3.357753038406372,
      "learning_rate": 1.9152923538230885e-07,
      "loss": 1.2006,
      "step": 1023
    },
    {
      "epoch": 97.52380952380952,
      "grad_norm": 7.069590091705322,
      "learning_rate": 1.9171664167916043e-07,
      "loss": 1.2018,
      "step": 1024
    },
    {
      "epoch": 97.61904761904762,
      "grad_norm": 3.987957715988159,
      "learning_rate": 1.91904047976012e-07,
      "loss": 1.1415,
      "step": 1025
    },
    {
      "epoch": 97.71428571428571,
      "grad_norm": 4.330837249755859,
      "learning_rate": 1.9209145427286357e-07,
      "loss": 1.1873,
      "step": 1026
    },
    {
      "epoch": 97.80952380952381,
      "grad_norm": 2.9018101692199707,
      "learning_rate": 1.9227886056971515e-07,
      "loss": 1.1942,
      "step": 1027
    },
    {
      "epoch": 97.9047619047619,
      "grad_norm": 3.1628873348236084,
      "learning_rate": 1.9246626686656673e-07,
      "loss": 1.141,
      "step": 1028
    },
    {
      "epoch": 98.0,
      "grad_norm": 6.722561359405518,
      "learning_rate": 1.926536731634183e-07,
      "loss": 1.237,
      "step": 1029
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.7096290745597602,
      "eval_f1": 0.7706422018348624,
      "eval_loss": 0.586707592010498,
      "eval_precision": 0.7095367847411445,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7411977547495682,
      "eval_runtime": 1.1565,
      "eval_samples_per_second": 2307.874,
      "eval_steps_per_second": 5.188,
      "step": 1029
    },
    {
      "epoch": 98.0952380952381,
      "grad_norm": 4.370731830596924,
      "learning_rate": 1.928410794602699e-07,
      "loss": 1.1843,
      "step": 1030
    },
    {
      "epoch": 98.19047619047619,
      "grad_norm": 2.9520230293273926,
      "learning_rate": 1.9302848575712144e-07,
      "loss": 1.1888,
      "step": 1031
    },
    {
      "epoch": 98.28571428571429,
      "grad_norm": 3.1140427589416504,
      "learning_rate": 1.9321589205397303e-07,
      "loss": 1.1886,
      "step": 1032
    },
    {
      "epoch": 98.38095238095238,
      "grad_norm": 2.5182907581329346,
      "learning_rate": 1.9340329835082458e-07,
      "loss": 1.1701,
      "step": 1033
    },
    {
      "epoch": 98.47619047619048,
      "grad_norm": 7.141319274902344,
      "learning_rate": 1.9359070464767616e-07,
      "loss": 1.1674,
      "step": 1034
    },
    {
      "epoch": 98.57142857142857,
      "grad_norm": 3.4200496673583984,
      "learning_rate": 1.9377811094452777e-07,
      "loss": 1.1621,
      "step": 1035
    },
    {
      "epoch": 98.66666666666667,
      "grad_norm": 3.106863260269165,
      "learning_rate": 1.9396551724137932e-07,
      "loss": 1.1857,
      "step": 1036
    },
    {
      "epoch": 98.76190476190476,
      "grad_norm": 6.592114448547363,
      "learning_rate": 1.941529235382309e-07,
      "loss": 1.2144,
      "step": 1037
    },
    {
      "epoch": 98.85714285714286,
      "grad_norm": 4.127084255218506,
      "learning_rate": 1.9434032983508246e-07,
      "loss": 1.195,
      "step": 1038
    },
    {
      "epoch": 98.95238095238095,
      "grad_norm": 3.0792510509490967,
      "learning_rate": 1.9452773613193404e-07,
      "loss": 1.1611,
      "step": 1039
    },
    {
      "epoch": 98.95238095238095,
      "eval_accuracy": 0.7088797302360434,
      "eval_f1": 0.769504598042124,
      "eval_loss": 0.5864457488059998,
      "eval_precision": 0.7099069512862616,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7414237190558434,
      "eval_runtime": 1.141,
      "eval_samples_per_second": 2339.082,
      "eval_steps_per_second": 5.258,
      "step": 1039
    },
    {
      "epoch": 99.04761904761905,
      "grad_norm": 4.860541820526123,
      "learning_rate": 1.9471514242878562e-07,
      "loss": 1.2126,
      "step": 1040
    },
    {
      "epoch": 99.14285714285714,
      "grad_norm": 5.0624566078186035,
      "learning_rate": 1.949025487256372e-07,
      "loss": 1.1765,
      "step": 1041
    },
    {
      "epoch": 99.23809523809524,
      "grad_norm": 3.882685422897339,
      "learning_rate": 1.9508995502248878e-07,
      "loss": 1.1533,
      "step": 1042
    },
    {
      "epoch": 99.33333333333333,
      "grad_norm": 3.081310749053955,
      "learning_rate": 1.9527736131934034e-07,
      "loss": 1.1557,
      "step": 1043
    },
    {
      "epoch": 99.42857142857143,
      "grad_norm": 3.0742149353027344,
      "learning_rate": 1.954647676161919e-07,
      "loss": 1.1964,
      "step": 1044
    },
    {
      "epoch": 99.52380952380952,
      "grad_norm": 2.9772236347198486,
      "learning_rate": 1.956521739130435e-07,
      "loss": 1.1767,
      "step": 1045
    },
    {
      "epoch": 99.61904761904762,
      "grad_norm": 3.278563976287842,
      "learning_rate": 1.9583958020989505e-07,
      "loss": 1.186,
      "step": 1046
    },
    {
      "epoch": 99.71428571428571,
      "grad_norm": 2.069175958633423,
      "learning_rate": 1.9602698650674663e-07,
      "loss": 1.1763,
      "step": 1047
    },
    {
      "epoch": 99.80952380952381,
      "grad_norm": 5.378757953643799,
      "learning_rate": 1.9621439280359824e-07,
      "loss": 1.2126,
      "step": 1048
    },
    {
      "epoch": 99.9047619047619,
      "grad_norm": 7.8624749183654785,
      "learning_rate": 1.964017991004498e-07,
      "loss": 1.1484,
      "step": 1049
    },
    {
      "epoch": 100.0,
      "grad_norm": 3.2688686847686768,
      "learning_rate": 1.9658920539730135e-07,
      "loss": 1.1375,
      "step": 1050
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.7077557137504683,
      "eval_f1": 0.768270944741533,
      "eval_loss": 0.5862920880317688,
      "eval_precision": 0.7096597145993414,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7416292458261371,
      "eval_runtime": 1.208,
      "eval_samples_per_second": 2209.442,
      "eval_steps_per_second": 4.967,
      "step": 1050
    },
    {
      "epoch": 100.0952380952381,
      "grad_norm": 4.440683364868164,
      "learning_rate": 1.967766116941529e-07,
      "loss": 1.184,
      "step": 1051
    },
    {
      "epoch": 100.19047619047619,
      "grad_norm": 2.988774061203003,
      "learning_rate": 1.969640179910045e-07,
      "loss": 1.1936,
      "step": 1052
    },
    {
      "epoch": 100.28571428571429,
      "grad_norm": 3.475858688354492,
      "learning_rate": 1.971514242878561e-07,
      "loss": 1.166,
      "step": 1053
    },
    {
      "epoch": 100.38095238095238,
      "grad_norm": 2.3044967651367188,
      "learning_rate": 1.9733883058470765e-07,
      "loss": 1.1795,
      "step": 1054
    },
    {
      "epoch": 100.47619047619048,
      "grad_norm": 2.6562399864196777,
      "learning_rate": 1.9752623688155925e-07,
      "loss": 1.1644,
      "step": 1055
    },
    {
      "epoch": 100.57142857142857,
      "grad_norm": 3.209141254425049,
      "learning_rate": 1.977136431784108e-07,
      "loss": 1.174,
      "step": 1056
    },
    {
      "epoch": 100.66666666666667,
      "grad_norm": 6.200804710388184,
      "learning_rate": 1.9790104947526236e-07,
      "loss": 1.1441,
      "step": 1057
    },
    {
      "epoch": 100.76190476190476,
      "grad_norm": 3.0925281047821045,
      "learning_rate": 1.9808845577211394e-07,
      "loss": 1.2014,
      "step": 1058
    },
    {
      "epoch": 100.85714285714286,
      "grad_norm": 3.244074821472168,
      "learning_rate": 1.9827586206896555e-07,
      "loss": 1.1827,
      "step": 1059
    },
    {
      "epoch": 100.95238095238095,
      "grad_norm": 4.83599853515625,
      "learning_rate": 1.984632683658171e-07,
      "loss": 1.187,
      "step": 1060
    },
    {
      "epoch": 100.95238095238095,
      "eval_accuracy": 0.7085050580741851,
      "eval_f1": 0.769139465875371,
      "eval_loss": 0.5860828757286072,
      "eval_precision": 0.7097480832420592,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7419314910765689,
      "eval_runtime": 1.2084,
      "eval_samples_per_second": 2208.654,
      "eval_steps_per_second": 4.965,
      "step": 1060
    },
    {
      "epoch": 101.04761904761905,
      "grad_norm": 8.952556610107422,
      "learning_rate": 1.9865067466266866e-07,
      "loss": 1.2129,
      "step": 1061
    },
    {
      "epoch": 101.14285714285714,
      "grad_norm": 7.221254348754883,
      "learning_rate": 1.9883808095952027e-07,
      "loss": 1.1905,
      "step": 1062
    },
    {
      "epoch": 101.23809523809524,
      "grad_norm": 4.311481952667236,
      "learning_rate": 1.9902548725637182e-07,
      "loss": 1.2263,
      "step": 1063
    },
    {
      "epoch": 101.33333333333333,
      "grad_norm": 3.539215087890625,
      "learning_rate": 1.992128935532234e-07,
      "loss": 1.1666,
      "step": 1064
    },
    {
      "epoch": 101.42857142857143,
      "grad_norm": 3.2866811752319336,
      "learning_rate": 1.9940029985007496e-07,
      "loss": 1.2134,
      "step": 1065
    },
    {
      "epoch": 101.52380952380952,
      "grad_norm": 3.0386905670166016,
      "learning_rate": 1.9958770614692657e-07,
      "loss": 1.168,
      "step": 1066
    },
    {
      "epoch": 101.61904761904762,
      "grad_norm": 6.335067272186279,
      "learning_rate": 1.9977511244377812e-07,
      "loss": 1.1881,
      "step": 1067
    },
    {
      "epoch": 101.71428571428571,
      "grad_norm": 11.900758743286133,
      "learning_rate": 1.9996251874062967e-07,
      "loss": 1.1134,
      "step": 1068
    },
    {
      "epoch": 101.80952380952381,
      "grad_norm": 3.2636590003967285,
      "learning_rate": 2.0014992503748128e-07,
      "loss": 1.2082,
      "step": 1069
    },
    {
      "epoch": 101.9047619047619,
      "grad_norm": 8.951799392700195,
      "learning_rate": 2.0033733133433286e-07,
      "loss": 1.1496,
      "step": 1070
    },
    {
      "epoch": 102.0,
      "grad_norm": 4.132380962371826,
      "learning_rate": 2.0052473763118442e-07,
      "loss": 1.1816,
      "step": 1071
    },
    {
      "epoch": 102.0,
      "eval_accuracy": 0.7081303859123267,
      "eval_f1": 0.7689113022841887,
      "eval_loss": 0.5860377550125122,
      "eval_precision": 0.7093596059113301,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7419801381692573,
      "eval_runtime": 1.2551,
      "eval_samples_per_second": 2126.441,
      "eval_steps_per_second": 4.78,
      "step": 1071
    },
    {
      "epoch": 102.0952380952381,
      "grad_norm": 2.5746874809265137,
      "learning_rate": 2.0071214392803597e-07,
      "loss": 1.1736,
      "step": 1072
    },
    {
      "epoch": 102.19047619047619,
      "grad_norm": 4.014669895172119,
      "learning_rate": 2.0089955022488758e-07,
      "loss": 1.1717,
      "step": 1073
    },
    {
      "epoch": 102.28571428571429,
      "grad_norm": 4.1914520263671875,
      "learning_rate": 2.0108695652173913e-07,
      "loss": 1.1752,
      "step": 1074
    },
    {
      "epoch": 102.38095238095238,
      "grad_norm": 4.514261245727539,
      "learning_rate": 2.0127436281859071e-07,
      "loss": 1.2236,
      "step": 1075
    },
    {
      "epoch": 102.47619047619048,
      "grad_norm": 3.665982961654663,
      "learning_rate": 2.014617691154423e-07,
      "loss": 1.1594,
      "step": 1076
    },
    {
      "epoch": 102.57142857142857,
      "grad_norm": 2.9374074935913086,
      "learning_rate": 2.0164917541229388e-07,
      "loss": 1.183,
      "step": 1077
    },
    {
      "epoch": 102.66666666666667,
      "grad_norm": 12.804777145385742,
      "learning_rate": 2.0183658170914543e-07,
      "loss": 1.2346,
      "step": 1078
    },
    {
      "epoch": 102.76190476190476,
      "grad_norm": 2.884666919708252,
      "learning_rate": 2.0202398800599699e-07,
      "loss": 1.1582,
      "step": 1079
    },
    {
      "epoch": 102.85714285714286,
      "grad_norm": 3.5441818237304688,
      "learning_rate": 2.022113943028486e-07,
      "loss": 1.15,
      "step": 1080
    },
    {
      "epoch": 102.95238095238095,
      "grad_norm": 3.3858766555786133,
      "learning_rate": 2.0239880059970015e-07,
      "loss": 1.1679,
      "step": 1081
    },
    {
      "epoch": 102.95238095238095,
      "eval_accuracy": 0.7100037467216186,
      "eval_f1": 0.7710059171597633,
      "eval_loss": 0.5860357284545898,
      "eval_precision": 0.7096949891067538,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7421658031088082,
      "eval_runtime": 1.3927,
      "eval_samples_per_second": 1916.353,
      "eval_steps_per_second": 4.308,
      "step": 1081
    },
    {
      "epoch": 103.04761904761905,
      "grad_norm": 5.704709053039551,
      "learning_rate": 2.0258620689655173e-07,
      "loss": 1.1737,
      "step": 1082
    },
    {
      "epoch": 103.14285714285714,
      "grad_norm": 5.600817680358887,
      "learning_rate": 2.0277361319340334e-07,
      "loss": 1.2036,
      "step": 1083
    },
    {
      "epoch": 103.23809523809524,
      "grad_norm": 4.802742958068848,
      "learning_rate": 2.029610194902549e-07,
      "loss": 1.2017,
      "step": 1084
    },
    {
      "epoch": 103.33333333333333,
      "grad_norm": 3.191215991973877,
      "learning_rate": 2.0314842578710645e-07,
      "loss": 1.1863,
      "step": 1085
    },
    {
      "epoch": 103.42857142857143,
      "grad_norm": 3.547861337661743,
      "learning_rate": 2.0333583208395803e-07,
      "loss": 1.2072,
      "step": 1086
    },
    {
      "epoch": 103.52380952380952,
      "grad_norm": 3.1179792881011963,
      "learning_rate": 2.035232383808096e-07,
      "loss": 1.1715,
      "step": 1087
    },
    {
      "epoch": 103.61904761904762,
      "grad_norm": 3.455887794494629,
      "learning_rate": 2.037106446776612e-07,
      "loss": 1.1727,
      "step": 1088
    },
    {
      "epoch": 103.71428571428571,
      "grad_norm": 3.4433748722076416,
      "learning_rate": 2.0389805097451274e-07,
      "loss": 1.2025,
      "step": 1089
    },
    {
      "epoch": 103.80952380952381,
      "grad_norm": 3.5270094871520996,
      "learning_rate": 2.0408545727136435e-07,
      "loss": 1.1437,
      "step": 1090
    },
    {
      "epoch": 103.9047619047619,
      "grad_norm": 5.01560640335083,
      "learning_rate": 2.042728635682159e-07,
      "loss": 1.1163,
      "step": 1091
    },
    {
      "epoch": 104.0,
      "grad_norm": 2.7155611515045166,
      "learning_rate": 2.0446026986506746e-07,
      "loss": 1.1798,
      "step": 1092
    },
    {
      "epoch": 104.0,
      "eval_accuracy": 0.710378418883477,
      "eval_f1": 0.770962962962963,
      "eval_loss": 0.5859765410423279,
      "eval_precision": 0.7105406881485528,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7421905584340817,
      "eval_runtime": 1.1491,
      "eval_samples_per_second": 2322.73,
      "eval_steps_per_second": 5.222,
      "step": 1092
    },
    {
      "epoch": 104.0952380952381,
      "grad_norm": 2.4141287803649902,
      "learning_rate": 2.0464767616191904e-07,
      "loss": 1.169,
      "step": 1093
    },
    {
      "epoch": 104.19047619047619,
      "grad_norm": 5.855286121368408,
      "learning_rate": 2.0483508245877065e-07,
      "loss": 1.1054,
      "step": 1094
    },
    {
      "epoch": 104.28571428571429,
      "grad_norm": 7.8229498863220215,
      "learning_rate": 2.050224887556222e-07,
      "loss": 1.1633,
      "step": 1095
    },
    {
      "epoch": 104.38095238095238,
      "grad_norm": 3.975707769393921,
      "learning_rate": 2.0520989505247376e-07,
      "loss": 1.2149,
      "step": 1096
    },
    {
      "epoch": 104.47619047619048,
      "grad_norm": 3.858734369277954,
      "learning_rate": 2.0539730134932536e-07,
      "loss": 1.2171,
      "step": 1097
    },
    {
      "epoch": 104.57142857142857,
      "grad_norm": 6.441440105438232,
      "learning_rate": 2.0558470764617692e-07,
      "loss": 1.1842,
      "step": 1098
    },
    {
      "epoch": 104.66666666666667,
      "grad_norm": 5.97361946105957,
      "learning_rate": 2.057721139430285e-07,
      "loss": 1.1458,
      "step": 1099
    },
    {
      "epoch": 104.76190476190476,
      "grad_norm": 5.021552562713623,
      "learning_rate": 2.0595952023988005e-07,
      "loss": 1.2462,
      "step": 1100
    },
    {
      "epoch": 104.85714285714286,
      "grad_norm": 9.252442359924316,
      "learning_rate": 2.0614692653673166e-07,
      "loss": 1.1586,
      "step": 1101
    },
    {
      "epoch": 104.95238095238095,
      "grad_norm": 4.486863613128662,
      "learning_rate": 2.0633433283358322e-07,
      "loss": 1.188,
      "step": 1102
    },
    {
      "epoch": 104.95238095238095,
      "eval_accuracy": 0.7081303859123267,
      "eval_f1": 0.768361581920904,
      "eval_loss": 0.5858822464942932,
      "eval_precision": 0.7102803738317757,
      "eval_recall": 0.8367875647668394,
      "eval_roc_auc": 0.7421781807714452,
      "eval_runtime": 1.176,
      "eval_samples_per_second": 2269.511,
      "eval_steps_per_second": 5.102,
      "step": 1102
    },
    {
      "epoch": 105.04761904761905,
      "grad_norm": 5.8441362380981445,
      "learning_rate": 2.0652173913043477e-07,
      "loss": 1.1511,
      "step": 1103
    },
    {
      "epoch": 105.14285714285714,
      "grad_norm": 6.34725284576416,
      "learning_rate": 2.0670914542728638e-07,
      "loss": 1.1635,
      "step": 1104
    },
    {
      "epoch": 105.23809523809524,
      "grad_norm": 2.6412837505340576,
      "learning_rate": 2.0689655172413796e-07,
      "loss": 1.2041,
      "step": 1105
    },
    {
      "epoch": 105.33333333333333,
      "grad_norm": 2.169490098953247,
      "learning_rate": 2.070839580209895e-07,
      "loss": 1.1938,
      "step": 1106
    },
    {
      "epoch": 105.42857142857143,
      "grad_norm": 3.8251116275787354,
      "learning_rate": 2.0727136431784107e-07,
      "loss": 1.2071,
      "step": 1107
    },
    {
      "epoch": 105.52380952380952,
      "grad_norm": 5.872073650360107,
      "learning_rate": 2.0745877061469267e-07,
      "loss": 1.0972,
      "step": 1108
    },
    {
      "epoch": 105.61904761904762,
      "grad_norm": 3.1469690799713135,
      "learning_rate": 2.0764617691154423e-07,
      "loss": 1.1162,
      "step": 1109
    },
    {
      "epoch": 105.71428571428571,
      "grad_norm": 2.4707601070404053,
      "learning_rate": 2.078335832083958e-07,
      "loss": 1.1861,
      "step": 1110
    },
    {
      "epoch": 105.80952380952381,
      "grad_norm": 4.59213924407959,
      "learning_rate": 2.080209895052474e-07,
      "loss": 1.1789,
      "step": 1111
    },
    {
      "epoch": 105.9047619047619,
      "grad_norm": 7.704277515411377,
      "learning_rate": 2.0820839580209897e-07,
      "loss": 1.2179,
      "step": 1112
    },
    {
      "epoch": 106.0,
      "grad_norm": 3.2231431007385254,
      "learning_rate": 2.0839580209895053e-07,
      "loss": 1.1869,
      "step": 1113
    },
    {
      "epoch": 106.0,
      "eval_accuracy": 0.7126264518546273,
      "eval_f1": 0.7735459108355477,
      "eval_loss": 0.585536777973175,
      "eval_precision": 0.7107976125881714,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7428284398388025,
      "eval_runtime": 1.1995,
      "eval_samples_per_second": 2225.008,
      "eval_steps_per_second": 5.002,
      "step": 1113
    },
    {
      "epoch": 106.0952380952381,
      "grad_norm": 5.697627067565918,
      "learning_rate": 2.0858320839580208e-07,
      "loss": 1.1273,
      "step": 1114
    },
    {
      "epoch": 106.19047619047619,
      "grad_norm": 10.429737091064453,
      "learning_rate": 2.087706146926537e-07,
      "loss": 1.1915,
      "step": 1115
    },
    {
      "epoch": 106.28571428571429,
      "grad_norm": 3.9276132583618164,
      "learning_rate": 2.0895802098950527e-07,
      "loss": 1.1987,
      "step": 1116
    },
    {
      "epoch": 106.38095238095238,
      "grad_norm": 8.287455558776855,
      "learning_rate": 2.0914542728635682e-07,
      "loss": 1.2169,
      "step": 1117
    },
    {
      "epoch": 106.47619047619048,
      "grad_norm": 3.018214464187622,
      "learning_rate": 2.0933283358320843e-07,
      "loss": 1.1799,
      "step": 1118
    },
    {
      "epoch": 106.57142857142857,
      "grad_norm": 2.187699794769287,
      "learning_rate": 2.0952023988005999e-07,
      "loss": 1.1894,
      "step": 1119
    },
    {
      "epoch": 106.66666666666667,
      "grad_norm": 3.7179007530212402,
      "learning_rate": 2.0970764617691154e-07,
      "loss": 1.1598,
      "step": 1120
    },
    {
      "epoch": 106.76190476190476,
      "grad_norm": 10.05657958984375,
      "learning_rate": 2.0989505247376312e-07,
      "loss": 1.1403,
      "step": 1121
    },
    {
      "epoch": 106.85714285714286,
      "grad_norm": 2.488464117050171,
      "learning_rate": 2.100824587706147e-07,
      "loss": 1.1659,
      "step": 1122
    },
    {
      "epoch": 106.95238095238095,
      "grad_norm": 2.6016857624053955,
      "learning_rate": 2.1026986506746628e-07,
      "loss": 1.1786,
      "step": 1123
    },
    {
      "epoch": 106.95238095238095,
      "eval_accuracy": 0.7111277632071937,
      "eval_f1": 0.7715555555555556,
      "eval_loss": 0.5852952599525452,
      "eval_precision": 0.7110868377935554,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7429594127806562,
      "eval_runtime": 1.2724,
      "eval_samples_per_second": 2097.676,
      "eval_steps_per_second": 4.716,
      "step": 1123
    },
    {
      "epoch": 107.04761904761905,
      "grad_norm": 12.497701644897461,
      "learning_rate": 2.1045727136431784e-07,
      "loss": 1.2145,
      "step": 1124
    },
    {
      "epoch": 107.14285714285714,
      "grad_norm": 3.5568621158599854,
      "learning_rate": 2.1064467766116944e-07,
      "loss": 1.2061,
      "step": 1125
    },
    {
      "epoch": 107.23809523809524,
      "grad_norm": 9.025201797485352,
      "learning_rate": 2.10832083958021e-07,
      "loss": 1.1153,
      "step": 1126
    },
    {
      "epoch": 107.33333333333333,
      "grad_norm": 5.04897928237915,
      "learning_rate": 2.1101949025487255e-07,
      "loss": 1.1678,
      "step": 1127
    },
    {
      "epoch": 107.42857142857143,
      "grad_norm": 3.4108526706695557,
      "learning_rate": 2.1120689655172413e-07,
      "loss": 1.1833,
      "step": 1128
    },
    {
      "epoch": 107.52380952380952,
      "grad_norm": 2.397312641143799,
      "learning_rate": 2.1139430284857574e-07,
      "loss": 1.1684,
      "step": 1129
    },
    {
      "epoch": 107.61904761904762,
      "grad_norm": 3.07975697517395,
      "learning_rate": 2.115817091454273e-07,
      "loss": 1.2202,
      "step": 1130
    },
    {
      "epoch": 107.71428571428571,
      "grad_norm": 3.9994544982910156,
      "learning_rate": 2.1176911544227885e-07,
      "loss": 1.1658,
      "step": 1131
    },
    {
      "epoch": 107.80952380952381,
      "grad_norm": 3.7984204292297363,
      "learning_rate": 2.1195652173913046e-07,
      "loss": 1.1601,
      "step": 1132
    },
    {
      "epoch": 107.9047619047619,
      "grad_norm": 3.4608612060546875,
      "learning_rate": 2.12143928035982e-07,
      "loss": 1.17,
      "step": 1133
    },
    {
      "epoch": 108.0,
      "grad_norm": 3.1976664066314697,
      "learning_rate": 2.123313343328336e-07,
      "loss": 1.1929,
      "step": 1134
    },
    {
      "epoch": 108.0,
      "eval_accuracy": 0.7115024353690521,
      "eval_f1": 0.771513353115727,
      "eval_loss": 0.5852454900741577,
      "eval_precision": 0.7119386637458927,
      "eval_recall": 0.8419689119170984,
      "eval_roc_auc": 0.7429919401266551,
      "eval_runtime": 1.1644,
      "eval_samples_per_second": 2292.254,
      "eval_steps_per_second": 5.153,
      "step": 1134
    },
    {
      "epoch": 108.0952380952381,
      "grad_norm": 5.28352689743042,
      "learning_rate": 2.1251874062968515e-07,
      "loss": 1.1692,
      "step": 1135
    },
    {
      "epoch": 108.19047619047619,
      "grad_norm": 5.02522611618042,
      "learning_rate": 2.1270614692653676e-07,
      "loss": 1.165,
      "step": 1136
    },
    {
      "epoch": 108.28571428571429,
      "grad_norm": 3.9818742275238037,
      "learning_rate": 2.128935532233883e-07,
      "loss": 1.1638,
      "step": 1137
    },
    {
      "epoch": 108.38095238095238,
      "grad_norm": 4.110209941864014,
      "learning_rate": 2.1308095952023986e-07,
      "loss": 1.197,
      "step": 1138
    },
    {
      "epoch": 108.47619047619048,
      "grad_norm": 4.014791965484619,
      "learning_rate": 2.1326836581709147e-07,
      "loss": 1.2021,
      "step": 1139
    },
    {
      "epoch": 108.57142857142857,
      "grad_norm": 3.4760916233062744,
      "learning_rate": 2.1345577211394305e-07,
      "loss": 1.2079,
      "step": 1140
    },
    {
      "epoch": 108.66666666666667,
      "grad_norm": 2.6710264682769775,
      "learning_rate": 2.136431784107946e-07,
      "loss": 1.1863,
      "step": 1141
    },
    {
      "epoch": 108.76190476190476,
      "grad_norm": 3.0839014053344727,
      "learning_rate": 2.1383058470764616e-07,
      "loss": 1.161,
      "step": 1142
    },
    {
      "epoch": 108.85714285714286,
      "grad_norm": 7.811803340911865,
      "learning_rate": 2.1401799100449777e-07,
      "loss": 1.156,
      "step": 1143
    },
    {
      "epoch": 108.95238095238095,
      "grad_norm": 2.7280216217041016,
      "learning_rate": 2.1420539730134932e-07,
      "loss": 1.1514,
      "step": 1144
    },
    {
      "epoch": 108.95238095238095,
      "eval_accuracy": 0.7122517796927689,
      "eval_f1": 0.7722419928825622,
      "eval_loss": 0.5852576494216919,
      "eval_precision": 0.712253829321663,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7431142774899251,
      "eval_runtime": 1.1663,
      "eval_samples_per_second": 2288.452,
      "eval_steps_per_second": 5.145,
      "step": 1144
    },
    {
      "epoch": 109.04761904761905,
      "grad_norm": 3.729548454284668,
      "learning_rate": 2.143928035982009e-07,
      "loss": 1.2153,
      "step": 1145
    },
    {
      "epoch": 109.14285714285714,
      "grad_norm": 4.262019157409668,
      "learning_rate": 2.145802098950525e-07,
      "loss": 1.1541,
      "step": 1146
    },
    {
      "epoch": 109.23809523809524,
      "grad_norm": 4.890219688415527,
      "learning_rate": 2.1476761619190407e-07,
      "loss": 1.1869,
      "step": 1147
    },
    {
      "epoch": 109.33333333333333,
      "grad_norm": 4.600557804107666,
      "learning_rate": 2.1495502248875562e-07,
      "loss": 1.1686,
      "step": 1148
    },
    {
      "epoch": 109.42857142857143,
      "grad_norm": 3.3701367378234863,
      "learning_rate": 2.1514242878560718e-07,
      "loss": 1.1191,
      "step": 1149
    },
    {
      "epoch": 109.52380952380952,
      "grad_norm": 3.143719434738159,
      "learning_rate": 2.1532983508245878e-07,
      "loss": 1.1985,
      "step": 1150
    },
    {
      "epoch": 109.61904761904762,
      "grad_norm": 2.748405694961548,
      "learning_rate": 2.1551724137931036e-07,
      "loss": 1.1981,
      "step": 1151
    },
    {
      "epoch": 109.71428571428571,
      "grad_norm": 3.2317254543304443,
      "learning_rate": 2.1570464767616192e-07,
      "loss": 1.1714,
      "step": 1152
    },
    {
      "epoch": 109.80952380952381,
      "grad_norm": 3.7985424995422363,
      "learning_rate": 2.1589205397301353e-07,
      "loss": 1.2185,
      "step": 1153
    },
    {
      "epoch": 109.9047619047619,
      "grad_norm": 3.4442954063415527,
      "learning_rate": 2.1607946026986508e-07,
      "loss": 1.1895,
      "step": 1154
    },
    {
      "epoch": 110.0,
      "grad_norm": 7.063841342926025,
      "learning_rate": 2.1626686656671664e-07,
      "loss": 1.1805,
      "step": 1155
    },
    {
      "epoch": 110.0,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.772821270310192,
      "eval_loss": 0.5851633548736572,
      "eval_precision": 0.710483432916893,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.743503742084053,
      "eval_runtime": 1.2251,
      "eval_samples_per_second": 2178.533,
      "eval_steps_per_second": 4.897,
      "step": 1155
    },
    {
      "epoch": 110.0952380952381,
      "grad_norm": 4.239227771759033,
      "learning_rate": 2.1645427286356824e-07,
      "loss": 1.1938,
      "step": 1156
    },
    {
      "epoch": 110.19047619047619,
      "grad_norm": 3.055314302444458,
      "learning_rate": 2.166416791604198e-07,
      "loss": 1.1715,
      "step": 1157
    },
    {
      "epoch": 110.28571428571429,
      "grad_norm": 4.560515880584717,
      "learning_rate": 2.1682908545727138e-07,
      "loss": 1.1353,
      "step": 1158
    },
    {
      "epoch": 110.38095238095238,
      "grad_norm": 3.283205509185791,
      "learning_rate": 2.1701649175412293e-07,
      "loss": 1.1551,
      "step": 1159
    },
    {
      "epoch": 110.47619047619048,
      "grad_norm": 3.285792589187622,
      "learning_rate": 2.1720389805097454e-07,
      "loss": 1.2105,
      "step": 1160
    },
    {
      "epoch": 110.57142857142857,
      "grad_norm": 7.5094380378723145,
      "learning_rate": 2.173913043478261e-07,
      "loss": 1.1471,
      "step": 1161
    },
    {
      "epoch": 110.66666666666667,
      "grad_norm": 4.275049209594727,
      "learning_rate": 2.1757871064467765e-07,
      "loss": 1.2339,
      "step": 1162
    },
    {
      "epoch": 110.76190476190476,
      "grad_norm": 4.221767902374268,
      "learning_rate": 2.1776611694152926e-07,
      "loss": 1.162,
      "step": 1163
    },
    {
      "epoch": 110.85714285714286,
      "grad_norm": 6.287664413452148,
      "learning_rate": 2.1795352323838084e-07,
      "loss": 1.1657,
      "step": 1164
    },
    {
      "epoch": 110.95238095238095,
      "grad_norm": 4.596139907836914,
      "learning_rate": 2.181409295352324e-07,
      "loss": 1.1674,
      "step": 1165
    },
    {
      "epoch": 110.95238095238095,
      "eval_accuracy": 0.710378418883477,
      "eval_f1": 0.7704187704187704,
      "eval_loss": 0.5850911140441895,
      "eval_precision": 0.7114646187602852,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7432691421991939,
      "eval_runtime": 1.2191,
      "eval_samples_per_second": 2189.248,
      "eval_steps_per_second": 4.922,
      "step": 1165
    },
    {
      "epoch": 111.04761904761905,
      "grad_norm": 3.9053878784179688,
      "learning_rate": 2.1832833583208395e-07,
      "loss": 1.1676,
      "step": 1166
    },
    {
      "epoch": 111.14285714285714,
      "grad_norm": 2.1512110233306885,
      "learning_rate": 2.1851574212893555e-07,
      "loss": 1.1689,
      "step": 1167
    },
    {
      "epoch": 111.23809523809524,
      "grad_norm": 4.1650590896606445,
      "learning_rate": 2.187031484257871e-07,
      "loss": 1.2255,
      "step": 1168
    },
    {
      "epoch": 111.33333333333333,
      "grad_norm": 3.7853543758392334,
      "learning_rate": 2.188905547226387e-07,
      "loss": 1.1473,
      "step": 1169
    },
    {
      "epoch": 111.42857142857143,
      "grad_norm": 5.21444034576416,
      "learning_rate": 2.190779610194903e-07,
      "loss": 1.1541,
      "step": 1170
    },
    {
      "epoch": 111.52380952380952,
      "grad_norm": 2.9562485218048096,
      "learning_rate": 2.1926536731634185e-07,
      "loss": 1.1349,
      "step": 1171
    },
    {
      "epoch": 111.61904761904762,
      "grad_norm": 3.4188034534454346,
      "learning_rate": 2.194527736131934e-07,
      "loss": 1.2242,
      "step": 1172
    },
    {
      "epoch": 111.71428571428571,
      "grad_norm": 2.437049627304077,
      "learning_rate": 2.1964017991004496e-07,
      "loss": 1.1651,
      "step": 1173
    },
    {
      "epoch": 111.80952380952381,
      "grad_norm": 4.054041385650635,
      "learning_rate": 2.1982758620689657e-07,
      "loss": 1.2115,
      "step": 1174
    },
    {
      "epoch": 111.9047619047619,
      "grad_norm": 5.592970371246338,
      "learning_rate": 2.2001499250374815e-07,
      "loss": 1.1969,
      "step": 1175
    },
    {
      "epoch": 112.0,
      "grad_norm": 3.1093966960906982,
      "learning_rate": 2.202023988005997e-07,
      "loss": 1.1868,
      "step": 1176
    },
    {
      "epoch": 112.0,
      "eval_accuracy": 0.7111277632071937,
      "eval_f1": 0.7726334414626954,
      "eval_loss": 0.5847890377044678,
      "eval_precision": 0.7092582566323768,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7440662061024754,
      "eval_runtime": 1.5656,
      "eval_samples_per_second": 1704.743,
      "eval_steps_per_second": 3.832,
      "step": 1176
    },
    {
      "epoch": 112.0952380952381,
      "grad_norm": 2.705421209335327,
      "learning_rate": 2.203898050974513e-07,
      "loss": 1.1709,
      "step": 1177
    },
    {
      "epoch": 112.19047619047619,
      "grad_norm": 2.8298819065093994,
      "learning_rate": 2.2057721139430286e-07,
      "loss": 1.1828,
      "step": 1178
    },
    {
      "epoch": 112.28571428571429,
      "grad_norm": 4.289648056030273,
      "learning_rate": 2.2076461769115442e-07,
      "loss": 1.1973,
      "step": 1179
    },
    {
      "epoch": 112.38095238095238,
      "grad_norm": 5.3589911460876465,
      "learning_rate": 2.20952023988006e-07,
      "loss": 1.1765,
      "step": 1180
    },
    {
      "epoch": 112.47619047619048,
      "grad_norm": 4.928300380706787,
      "learning_rate": 2.211394302848576e-07,
      "loss": 1.1323,
      "step": 1181
    },
    {
      "epoch": 112.57142857142857,
      "grad_norm": 3.292466878890991,
      "learning_rate": 2.2132683658170916e-07,
      "loss": 1.1578,
      "step": 1182
    },
    {
      "epoch": 112.66666666666667,
      "grad_norm": 3.1881377696990967,
      "learning_rate": 2.2151424287856072e-07,
      "loss": 1.2117,
      "step": 1183
    },
    {
      "epoch": 112.76190476190476,
      "grad_norm": 3.145686626434326,
      "learning_rate": 2.2170164917541232e-07,
      "loss": 1.1578,
      "step": 1184
    },
    {
      "epoch": 112.85714285714286,
      "grad_norm": 2.949483633041382,
      "learning_rate": 2.2188905547226388e-07,
      "loss": 1.1966,
      "step": 1185
    },
    {
      "epoch": 112.95238095238095,
      "grad_norm": 2.7736291885375977,
      "learning_rate": 2.2207646176911546e-07,
      "loss": 1.1991,
      "step": 1186
    },
    {
      "epoch": 112.95238095238095,
      "eval_accuracy": 0.710378418883477,
      "eval_f1": 0.7702823179791977,
      "eval_loss": 0.5846362113952637,
      "eval_precision": 0.7116968698517299,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7440405872193436,
      "eval_runtime": 1.1484,
      "eval_samples_per_second": 2324.094,
      "eval_steps_per_second": 5.225,
      "step": 1186
    },
    {
      "epoch": 113.04761904761905,
      "grad_norm": 2.5959534645080566,
      "learning_rate": 2.2226386806596701e-07,
      "loss": 1.1329,
      "step": 1187
    },
    {
      "epoch": 113.14285714285714,
      "grad_norm": 5.067058086395264,
      "learning_rate": 2.2245127436281862e-07,
      "loss": 1.1448,
      "step": 1188
    },
    {
      "epoch": 113.23809523809524,
      "grad_norm": 3.178851842880249,
      "learning_rate": 2.2263868065967018e-07,
      "loss": 1.1859,
      "step": 1189
    },
    {
      "epoch": 113.33333333333333,
      "grad_norm": 4.485405921936035,
      "learning_rate": 2.2282608695652173e-07,
      "loss": 1.1334,
      "step": 1190
    },
    {
      "epoch": 113.42857142857143,
      "grad_norm": 5.1164350509643555,
      "learning_rate": 2.2301349325337334e-07,
      "loss": 1.1733,
      "step": 1191
    },
    {
      "epoch": 113.52380952380952,
      "grad_norm": 4.061066150665283,
      "learning_rate": 2.2320089955022492e-07,
      "loss": 1.1619,
      "step": 1192
    },
    {
      "epoch": 113.61904761904762,
      "grad_norm": 3.489614725112915,
      "learning_rate": 2.2338830584707647e-07,
      "loss": 1.1891,
      "step": 1193
    },
    {
      "epoch": 113.71428571428571,
      "grad_norm": 3.178351640701294,
      "learning_rate": 2.2357571214392803e-07,
      "loss": 1.1894,
      "step": 1194
    },
    {
      "epoch": 113.80952380952381,
      "grad_norm": 3.2042295932769775,
      "learning_rate": 2.2376311844077963e-07,
      "loss": 1.1812,
      "step": 1195
    },
    {
      "epoch": 113.9047619047619,
      "grad_norm": 5.772481441497803,
      "learning_rate": 2.239505247376312e-07,
      "loss": 1.1984,
      "step": 1196
    },
    {
      "epoch": 114.0,
      "grad_norm": 6.261301040649414,
      "learning_rate": 2.2413793103448277e-07,
      "loss": 1.1562,
      "step": 1197
    },
    {
      "epoch": 114.0,
      "eval_accuracy": 0.7122517796927689,
      "eval_f1": 0.7726465364120781,
      "eval_loss": 0.5846668481826782,
      "eval_precision": 0.7115594329334788,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7440354058721933,
      "eval_runtime": 1.1159,
      "eval_samples_per_second": 2391.852,
      "eval_steps_per_second": 5.377,
      "step": 1197
    },
    {
      "epoch": 114.0952380952381,
      "grad_norm": 3.7072525024414062,
      "learning_rate": 2.2432533733133435e-07,
      "loss": 1.1596,
      "step": 1198
    },
    {
      "epoch": 114.19047619047619,
      "grad_norm": 4.127257823944092,
      "learning_rate": 2.2451274362818593e-07,
      "loss": 1.2198,
      "step": 1199
    },
    {
      "epoch": 114.28571428571429,
      "grad_norm": 5.291389465332031,
      "learning_rate": 2.247001499250375e-07,
      "loss": 1.214,
      "step": 1200
    },
    {
      "epoch": 114.38095238095238,
      "grad_norm": 4.877745628356934,
      "learning_rate": 2.2488755622188904e-07,
      "loss": 1.1677,
      "step": 1201
    },
    {
      "epoch": 114.47619047619048,
      "grad_norm": 7.776942253112793,
      "learning_rate": 2.2507496251874065e-07,
      "loss": 1.2202,
      "step": 1202
    },
    {
      "epoch": 114.57142857142857,
      "grad_norm": 5.472081661224365,
      "learning_rate": 2.252623688155922e-07,
      "loss": 1.1885,
      "step": 1203
    },
    {
      "epoch": 114.66666666666667,
      "grad_norm": 5.365962028503418,
      "learning_rate": 2.2544977511244378e-07,
      "loss": 1.115,
      "step": 1204
    },
    {
      "epoch": 114.76190476190476,
      "grad_norm": 3.9224565029144287,
      "learning_rate": 2.256371814092954e-07,
      "loss": 1.1652,
      "step": 1205
    },
    {
      "epoch": 114.85714285714286,
      "grad_norm": 4.8845133781433105,
      "learning_rate": 2.2582458770614695e-07,
      "loss": 1.1369,
      "step": 1206
    },
    {
      "epoch": 114.95238095238095,
      "grad_norm": 7.695840835571289,
      "learning_rate": 2.260119940029985e-07,
      "loss": 1.2119,
      "step": 1207
    },
    {
      "epoch": 114.95238095238095,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7720023626698169,
      "eval_loss": 0.5845794081687927,
      "eval_precision": 0.7095548317046688,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.744254749568221,
      "eval_runtime": 1.2453,
      "eval_samples_per_second": 2143.316,
      "eval_steps_per_second": 4.818,
      "step": 1207
    },
    {
      "epoch": 115.04761904761905,
      "grad_norm": 5.336812973022461,
      "learning_rate": 2.2619940029985005e-07,
      "loss": 1.216,
      "step": 1208
    },
    {
      "epoch": 115.14285714285714,
      "grad_norm": 5.411586761474609,
      "learning_rate": 2.2638680659670166e-07,
      "loss": 1.1411,
      "step": 1209
    },
    {
      "epoch": 115.23809523809524,
      "grad_norm": 3.4818155765533447,
      "learning_rate": 2.2657421289355324e-07,
      "loss": 1.2321,
      "step": 1210
    },
    {
      "epoch": 115.33333333333333,
      "grad_norm": 6.3038859367370605,
      "learning_rate": 2.267616191904048e-07,
      "loss": 1.105,
      "step": 1211
    },
    {
      "epoch": 115.42857142857143,
      "grad_norm": 3.4177987575531006,
      "learning_rate": 2.269490254872564e-07,
      "loss": 1.1846,
      "step": 1212
    },
    {
      "epoch": 115.52380952380952,
      "grad_norm": 6.946967124938965,
      "learning_rate": 2.2713643178410796e-07,
      "loss": 1.1919,
      "step": 1213
    },
    {
      "epoch": 115.61904761904762,
      "grad_norm": 3.6147685050964355,
      "learning_rate": 2.2732383808095951e-07,
      "loss": 1.1587,
      "step": 1214
    },
    {
      "epoch": 115.71428571428571,
      "grad_norm": 5.091091632843018,
      "learning_rate": 2.275112443778111e-07,
      "loss": 1.1811,
      "step": 1215
    },
    {
      "epoch": 115.80952380952381,
      "grad_norm": 2.9732019901275635,
      "learning_rate": 2.276986506746627e-07,
      "loss": 1.1818,
      "step": 1216
    },
    {
      "epoch": 115.9047619047619,
      "grad_norm": 3.7847113609313965,
      "learning_rate": 2.2788605697151426e-07,
      "loss": 1.1645,
      "step": 1217
    },
    {
      "epoch": 116.0,
      "grad_norm": 3.948373317718506,
      "learning_rate": 2.280734632683658e-07,
      "loss": 1.1802,
      "step": 1218
    },
    {
      "epoch": 116.0,
      "eval_accuracy": 0.7111277632071937,
      "eval_f1": 0.7714201008004744,
      "eval_loss": 0.5844592452049255,
      "eval_precision": 0.7113176599234554,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7443549222797927,
      "eval_runtime": 1.2081,
      "eval_samples_per_second": 2209.313,
      "eval_steps_per_second": 4.967,
      "step": 1218
    },
    {
      "epoch": 116.0952380952381,
      "grad_norm": 4.794178485870361,
      "learning_rate": 2.2826086956521742e-07,
      "loss": 1.1632,
      "step": 1219
    },
    {
      "epoch": 116.19047619047619,
      "grad_norm": 4.711289882659912,
      "learning_rate": 2.2844827586206897e-07,
      "loss": 1.2005,
      "step": 1220
    },
    {
      "epoch": 116.28571428571429,
      "grad_norm": 3.6995790004730225,
      "learning_rate": 2.2863568215892055e-07,
      "loss": 1.173,
      "step": 1221
    },
    {
      "epoch": 116.38095238095238,
      "grad_norm": 7.084836959838867,
      "learning_rate": 2.288230884557721e-07,
      "loss": 1.1971,
      "step": 1222
    },
    {
      "epoch": 116.47619047619048,
      "grad_norm": 4.868923187255859,
      "learning_rate": 2.2901049475262372e-07,
      "loss": 1.1268,
      "step": 1223
    },
    {
      "epoch": 116.57142857142857,
      "grad_norm": 3.7296676635742188,
      "learning_rate": 2.2919790104947527e-07,
      "loss": 1.1685,
      "step": 1224
    },
    {
      "epoch": 116.66666666666667,
      "grad_norm": 1.99024498462677,
      "learning_rate": 2.2938530734632683e-07,
      "loss": 1.1571,
      "step": 1225
    },
    {
      "epoch": 116.76190476190476,
      "grad_norm": 4.100002765655518,
      "learning_rate": 2.2957271364317843e-07,
      "loss": 1.1668,
      "step": 1226
    },
    {
      "epoch": 116.85714285714286,
      "grad_norm": 3.0545077323913574,
      "learning_rate": 2.2976011994003001e-07,
      "loss": 1.1952,
      "step": 1227
    },
    {
      "epoch": 116.95238095238095,
      "grad_norm": 6.123928546905518,
      "learning_rate": 2.2994752623688157e-07,
      "loss": 1.1824,
      "step": 1228
    },
    {
      "epoch": 116.95238095238095,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7718676122931442,
      "eval_loss": 0.5843604207038879,
      "eval_precision": 0.7097826086956521,
      "eval_recall": 0.8458549222797928,
      "eval_roc_auc": 0.744713298791019,
      "eval_runtime": 1.3086,
      "eval_samples_per_second": 2039.52,
      "eval_steps_per_second": 4.585,
      "step": 1228
    },
    {
      "epoch": 117.04761904761905,
      "grad_norm": 5.737049102783203,
      "learning_rate": 2.3013493253373312e-07,
      "loss": 1.2047,
      "step": 1229
    },
    {
      "epoch": 117.14285714285714,
      "grad_norm": 3.4674408435821533,
      "learning_rate": 2.3032233883058473e-07,
      "loss": 1.175,
      "step": 1230
    },
    {
      "epoch": 117.23809523809524,
      "grad_norm": 3.057267427444458,
      "learning_rate": 2.3050974512743628e-07,
      "loss": 1.1479,
      "step": 1231
    },
    {
      "epoch": 117.33333333333333,
      "grad_norm": 6.359550476074219,
      "learning_rate": 2.3069715142428787e-07,
      "loss": 1.2121,
      "step": 1232
    },
    {
      "epoch": 117.42857142857143,
      "grad_norm": 4.31020450592041,
      "learning_rate": 2.3088455772113945e-07,
      "loss": 1.2132,
      "step": 1233
    },
    {
      "epoch": 117.52380952380952,
      "grad_norm": 6.0614705085754395,
      "learning_rate": 2.3107196401799103e-07,
      "loss": 1.1611,
      "step": 1234
    },
    {
      "epoch": 117.61904761904762,
      "grad_norm": 6.543889999389648,
      "learning_rate": 2.3125937031484258e-07,
      "loss": 1.1608,
      "step": 1235
    },
    {
      "epoch": 117.71428571428571,
      "grad_norm": 2.0105361938476562,
      "learning_rate": 2.3144677661169414e-07,
      "loss": 1.1534,
      "step": 1236
    },
    {
      "epoch": 117.80952380952381,
      "grad_norm": 4.8146443367004395,
      "learning_rate": 2.3163418290854574e-07,
      "loss": 1.1649,
      "step": 1237
    },
    {
      "epoch": 117.9047619047619,
      "grad_norm": 4.572299003601074,
      "learning_rate": 2.318215892053973e-07,
      "loss": 1.1401,
      "step": 1238
    },
    {
      "epoch": 118.0,
      "grad_norm": 2.9639434814453125,
      "learning_rate": 2.3200899550224888e-07,
      "loss": 1.17,
      "step": 1239
    },
    {
      "epoch": 118.0,
      "eval_accuracy": 0.7111277632071937,
      "eval_f1": 0.7715555555555556,
      "eval_loss": 0.584344208240509,
      "eval_precision": 0.7110868377935554,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7446220495106506,
      "eval_runtime": 1.1134,
      "eval_samples_per_second": 2397.16,
      "eval_steps_per_second": 5.389,
      "step": 1239
    },
    {
      "epoch": 118.0952380952381,
      "grad_norm": 2.3465840816497803,
      "learning_rate": 2.3219640179910049e-07,
      "loss": 1.1434,
      "step": 1240
    },
    {
      "epoch": 118.19047619047619,
      "grad_norm": 2.926537275314331,
      "learning_rate": 2.3238380809595204e-07,
      "loss": 1.2079,
      "step": 1241
    },
    {
      "epoch": 118.28571428571429,
      "grad_norm": 2.285618305206299,
      "learning_rate": 2.325712143928036e-07,
      "loss": 1.1869,
      "step": 1242
    },
    {
      "epoch": 118.38095238095238,
      "grad_norm": 4.09526252746582,
      "learning_rate": 2.3275862068965515e-07,
      "loss": 1.188,
      "step": 1243
    },
    {
      "epoch": 118.47619047619048,
      "grad_norm": 2.2800137996673584,
      "learning_rate": 2.3294602698650676e-07,
      "loss": 1.199,
      "step": 1244
    },
    {
      "epoch": 118.57142857142857,
      "grad_norm": 2.6519763469696045,
      "learning_rate": 2.3313343328335834e-07,
      "loss": 1.1863,
      "step": 1245
    },
    {
      "epoch": 118.66666666666667,
      "grad_norm": 3.5764822959899902,
      "learning_rate": 2.333208395802099e-07,
      "loss": 1.2064,
      "step": 1246
    },
    {
      "epoch": 118.76190476190476,
      "grad_norm": 5.1671576499938965,
      "learning_rate": 2.335082458770615e-07,
      "loss": 1.1677,
      "step": 1247
    },
    {
      "epoch": 118.85714285714286,
      "grad_norm": 3.731257915496826,
      "learning_rate": 2.3369565217391305e-07,
      "loss": 1.1424,
      "step": 1248
    },
    {
      "epoch": 118.95238095238095,
      "grad_norm": 2.59311842918396,
      "learning_rate": 2.338830584707646e-07,
      "loss": 1.1511,
      "step": 1249
    },
    {
      "epoch": 118.95238095238095,
      "eval_accuracy": 0.7115024353690521,
      "eval_f1": 0.7720544701006513,
      "eval_loss": 0.5842468738555908,
      "eval_precision": 0.7110141766630316,
      "eval_recall": 0.844559585492228,
      "eval_roc_auc": 0.7448177892918826,
      "eval_runtime": 1.2423,
      "eval_samples_per_second": 2148.371,
      "eval_steps_per_second": 4.83,
      "step": 1249
    },
    {
      "epoch": 119.04761904761905,
      "grad_norm": 4.2373857498168945,
      "learning_rate": 2.340704647676162e-07,
      "loss": 1.1727,
      "step": 1250
    },
    {
      "epoch": 119.14285714285714,
      "grad_norm": 4.400254249572754,
      "learning_rate": 2.342578710644678e-07,
      "loss": 1.1225,
      "step": 1251
    },
    {
      "epoch": 119.23809523809524,
      "grad_norm": 4.6587419509887695,
      "learning_rate": 2.3444527736131935e-07,
      "loss": 1.1804,
      "step": 1252
    },
    {
      "epoch": 119.33333333333333,
      "grad_norm": 6.453824520111084,
      "learning_rate": 2.346326836581709e-07,
      "loss": 1.2251,
      "step": 1253
    },
    {
      "epoch": 119.42857142857143,
      "grad_norm": 4.208118915557861,
      "learning_rate": 2.3482008995502251e-07,
      "loss": 1.1708,
      "step": 1254
    },
    {
      "epoch": 119.52380952380952,
      "grad_norm": 6.1400346755981445,
      "learning_rate": 2.3500749625187407e-07,
      "loss": 1.1984,
      "step": 1255
    },
    {
      "epoch": 119.61904761904762,
      "grad_norm": 6.060001850128174,
      "learning_rate": 2.3519490254872565e-07,
      "loss": 1.1838,
      "step": 1256
    },
    {
      "epoch": 119.71428571428571,
      "grad_norm": 3.5763938426971436,
      "learning_rate": 2.353823088455772e-07,
      "loss": 1.1486,
      "step": 1257
    },
    {
      "epoch": 119.80952380952381,
      "grad_norm": 2.3884499073028564,
      "learning_rate": 2.355697151424288e-07,
      "loss": 1.1696,
      "step": 1258
    },
    {
      "epoch": 119.9047619047619,
      "grad_norm": 2.4123425483703613,
      "learning_rate": 2.3575712143928037e-07,
      "loss": 1.1633,
      "step": 1259
    },
    {
      "epoch": 120.0,
      "grad_norm": 3.5864171981811523,
      "learning_rate": 2.3594452773613192e-07,
      "loss": 1.1351,
      "step": 1260
    },
    {
      "epoch": 120.0,
      "eval_accuracy": 0.7100037467216186,
      "eval_f1": 0.7710059171597633,
      "eval_loss": 0.584304690361023,
      "eval_precision": 0.7096949891067538,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7448684513529072,
      "eval_runtime": 1.2312,
      "eval_samples_per_second": 2167.728,
      "eval_steps_per_second": 4.873,
      "step": 1260
    },
    {
      "epoch": 120.0952380952381,
      "grad_norm": 4.860226631164551,
      "learning_rate": 2.3613193403298353e-07,
      "loss": 1.1599,
      "step": 1261
    },
    {
      "epoch": 120.19047619047619,
      "grad_norm": 3.0438621044158936,
      "learning_rate": 2.363193403298351e-07,
      "loss": 1.2055,
      "step": 1262
    },
    {
      "epoch": 120.28571428571429,
      "grad_norm": 3.7721710205078125,
      "learning_rate": 2.3650674662668666e-07,
      "loss": 1.1663,
      "step": 1263
    },
    {
      "epoch": 120.38095238095238,
      "grad_norm": 2.985271692276001,
      "learning_rate": 2.3669415292353827e-07,
      "loss": 1.2082,
      "step": 1264
    },
    {
      "epoch": 120.47619047619048,
      "grad_norm": 4.9321489334106445,
      "learning_rate": 2.3688155922038983e-07,
      "loss": 1.1306,
      "step": 1265
    },
    {
      "epoch": 120.57142857142857,
      "grad_norm": 4.529792308807373,
      "learning_rate": 2.3706896551724138e-07,
      "loss": 1.1922,
      "step": 1266
    },
    {
      "epoch": 120.66666666666667,
      "grad_norm": 3.7236716747283936,
      "learning_rate": 2.3725637181409296e-07,
      "loss": 1.191,
      "step": 1267
    },
    {
      "epoch": 120.76190476190476,
      "grad_norm": 2.2532882690429688,
      "learning_rate": 2.3744377811094454e-07,
      "loss": 1.1717,
      "step": 1268
    },
    {
      "epoch": 120.85714285714286,
      "grad_norm": 1.9853754043579102,
      "learning_rate": 2.3763118440779612e-07,
      "loss": 1.1676,
      "step": 1269
    },
    {
      "epoch": 120.95238095238095,
      "grad_norm": 3.591606378555298,
      "learning_rate": 2.3781859070464768e-07,
      "loss": 1.1982,
      "step": 1270
    },
    {
      "epoch": 120.95238095238095,
      "eval_accuracy": 0.7096290745597602,
      "eval_f1": 0.7702342128668841,
      "eval_loss": 0.5842850208282471,
      "eval_precision": 0.7102241662110443,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.7447354634427173,
      "eval_runtime": 1.3384,
      "eval_samples_per_second": 1994.107,
      "eval_steps_per_second": 4.483,
      "step": 1270
    },
    {
      "epoch": 121.04761904761905,
      "grad_norm": 4.7795891761779785,
      "learning_rate": 2.3800599700149928e-07,
      "loss": 1.1261,
      "step": 1271
    },
    {
      "epoch": 121.14285714285714,
      "grad_norm": 4.506382465362549,
      "learning_rate": 2.3819340329835084e-07,
      "loss": 1.1857,
      "step": 1272
    },
    {
      "epoch": 121.23809523809524,
      "grad_norm": 6.346680641174316,
      "learning_rate": 2.3838080959520242e-07,
      "loss": 1.183,
      "step": 1273
    },
    {
      "epoch": 121.33333333333333,
      "grad_norm": 2.1210904121398926,
      "learning_rate": 2.3856821589205395e-07,
      "loss": 1.1398,
      "step": 1274
    },
    {
      "epoch": 121.42857142857143,
      "grad_norm": 8.61037540435791,
      "learning_rate": 2.3875562218890556e-07,
      "loss": 1.1491,
      "step": 1275
    },
    {
      "epoch": 121.52380952380952,
      "grad_norm": 4.891674041748047,
      "learning_rate": 2.389430284857571e-07,
      "loss": 1.1495,
      "step": 1276
    },
    {
      "epoch": 121.61904761904762,
      "grad_norm": 3.5965917110443115,
      "learning_rate": 2.391304347826087e-07,
      "loss": 1.1773,
      "step": 1277
    },
    {
      "epoch": 121.71428571428571,
      "grad_norm": 4.160679340362549,
      "learning_rate": 2.393178410794603e-07,
      "loss": 1.1941,
      "step": 1278
    },
    {
      "epoch": 121.80952380952381,
      "grad_norm": 5.527775287628174,
      "learning_rate": 2.395052473763119e-07,
      "loss": 1.1966,
      "step": 1279
    },
    {
      "epoch": 121.9047619047619,
      "grad_norm": 3.736419200897217,
      "learning_rate": 2.3969265367316343e-07,
      "loss": 1.1968,
      "step": 1280
    },
    {
      "epoch": 122.0,
      "grad_norm": 4.868208408355713,
      "learning_rate": 2.39880059970015e-07,
      "loss": 1.1798,
      "step": 1281
    },
    {
      "epoch": 122.0,
      "eval_accuracy": 0.7100037467216186,
      "eval_f1": 0.7716814159292036,
      "eval_loss": 0.5840305089950562,
      "eval_precision": 0.7085590465872156,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.745215889464594,
      "eval_runtime": 1.1693,
      "eval_samples_per_second": 2282.508,
      "eval_steps_per_second": 5.131,
      "step": 1281
    },
    {
      "epoch": 122.0952380952381,
      "grad_norm": 2.7348592281341553,
      "learning_rate": 2.400674662668666e-07,
      "loss": 1.2017,
      "step": 1282
    },
    {
      "epoch": 122.19047619047619,
      "grad_norm": 4.157247066497803,
      "learning_rate": 2.4025487256371815e-07,
      "loss": 1.1972,
      "step": 1283
    },
    {
      "epoch": 122.28571428571429,
      "grad_norm": 2.7334799766540527,
      "learning_rate": 2.404422788605697e-07,
      "loss": 1.1519,
      "step": 1284
    },
    {
      "epoch": 122.38095238095238,
      "grad_norm": 4.976047515869141,
      "learning_rate": 2.406296851574213e-07,
      "loss": 1.1174,
      "step": 1285
    },
    {
      "epoch": 122.47619047619048,
      "grad_norm": 2.340428352355957,
      "learning_rate": 2.4081709145427287e-07,
      "loss": 1.1801,
      "step": 1286
    },
    {
      "epoch": 122.57142857142857,
      "grad_norm": 5.766010284423828,
      "learning_rate": 2.410044977511244e-07,
      "loss": 1.1731,
      "step": 1287
    },
    {
      "epoch": 122.66666666666667,
      "grad_norm": 7.431637763977051,
      "learning_rate": 2.4119190404797603e-07,
      "loss": 1.1525,
      "step": 1288
    },
    {
      "epoch": 122.76190476190476,
      "grad_norm": 6.350968360900879,
      "learning_rate": 2.4137931034482764e-07,
      "loss": 1.1964,
      "step": 1289
    },
    {
      "epoch": 122.85714285714286,
      "grad_norm": 4.233700752258301,
      "learning_rate": 2.415667166416792e-07,
      "loss": 1.1762,
      "step": 1290
    },
    {
      "epoch": 122.95238095238095,
      "grad_norm": 3.7037289142608643,
      "learning_rate": 2.4175412293853074e-07,
      "loss": 1.2184,
      "step": 1291
    },
    {
      "epoch": 122.95238095238095,
      "eval_accuracy": 0.710378418883477,
      "eval_f1": 0.7717744316504281,
      "eval_loss": 0.5837914943695068,
      "eval_precision": 0.7091698317959848,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7454545192861255,
      "eval_runtime": 1.4528,
      "eval_samples_per_second": 1837.107,
      "eval_steps_per_second": 4.13,
      "step": 1291
    },
    {
      "epoch": 123.04761904761905,
      "grad_norm": 4.702028751373291,
      "learning_rate": 2.4194152923538235e-07,
      "loss": 1.1751,
      "step": 1292
    },
    {
      "epoch": 123.14285714285714,
      "grad_norm": 4.630548000335693,
      "learning_rate": 2.421289355322339e-07,
      "loss": 1.1632,
      "step": 1293
    },
    {
      "epoch": 123.23809523809524,
      "grad_norm": 3.001009702682495,
      "learning_rate": 2.4231634182908546e-07,
      "loss": 1.147,
      "step": 1294
    },
    {
      "epoch": 123.33333333333333,
      "grad_norm": 2.9855480194091797,
      "learning_rate": 2.42503748125937e-07,
      "loss": 1.1795,
      "step": 1295
    },
    {
      "epoch": 123.42857142857143,
      "grad_norm": 4.929809093475342,
      "learning_rate": 2.426911544227886e-07,
      "loss": 1.199,
      "step": 1296
    },
    {
      "epoch": 123.52380952380952,
      "grad_norm": 3.593780994415283,
      "learning_rate": 2.428785607196402e-07,
      "loss": 1.2116,
      "step": 1297
    },
    {
      "epoch": 123.61904761904762,
      "grad_norm": 3.6316239833831787,
      "learning_rate": 2.4306596701649173e-07,
      "loss": 1.2109,
      "step": 1298
    },
    {
      "epoch": 123.71428571428571,
      "grad_norm": 4.772429466247559,
      "learning_rate": 2.4325337331334334e-07,
      "loss": 1.1818,
      "step": 1299
    },
    {
      "epoch": 123.80952380952381,
      "grad_norm": 2.6498239040374756,
      "learning_rate": 2.4344077961019495e-07,
      "loss": 1.1726,
      "step": 1300
    },
    {
      "epoch": 123.9047619047619,
      "grad_norm": 4.785904407501221,
      "learning_rate": 2.436281859070465e-07,
      "loss": 1.1772,
      "step": 1301
    },
    {
      "epoch": 124.0,
      "grad_norm": 5.397092819213867,
      "learning_rate": 2.4381559220389806e-07,
      "loss": 1.1346,
      "step": 1302
    },
    {
      "epoch": 124.0,
      "eval_accuracy": 0.7092544023979018,
      "eval_f1": 0.7700059276822763,
      "eval_loss": 0.583419919013977,
      "eval_precision": 0.7098360655737705,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.7457625215889465,
      "eval_runtime": 1.1763,
      "eval_samples_per_second": 2269.028,
      "eval_steps_per_second": 5.101,
      "step": 1302
    },
    {
      "epoch": 124.0952380952381,
      "grad_norm": 3.3638992309570312,
      "learning_rate": 2.4400299850074966e-07,
      "loss": 1.1885,
      "step": 1303
    },
    {
      "epoch": 124.19047619047619,
      "grad_norm": 4.315061569213867,
      "learning_rate": 2.441904047976012e-07,
      "loss": 1.1324,
      "step": 1304
    },
    {
      "epoch": 124.28571428571429,
      "grad_norm": 1.9663842916488647,
      "learning_rate": 2.4437781109445277e-07,
      "loss": 1.1901,
      "step": 1305
    },
    {
      "epoch": 124.38095238095238,
      "grad_norm": 6.373414039611816,
      "learning_rate": 2.445652173913044e-07,
      "loss": 1.2036,
      "step": 1306
    },
    {
      "epoch": 124.47619047619048,
      "grad_norm": 4.554337978363037,
      "learning_rate": 2.4475262368815593e-07,
      "loss": 1.133,
      "step": 1307
    },
    {
      "epoch": 124.57142857142857,
      "grad_norm": 7.343845844268799,
      "learning_rate": 2.449400299850075e-07,
      "loss": 1.1857,
      "step": 1308
    },
    {
      "epoch": 124.66666666666667,
      "grad_norm": 3.1621153354644775,
      "learning_rate": 2.4512743628185904e-07,
      "loss": 1.1735,
      "step": 1309
    },
    {
      "epoch": 124.76190476190476,
      "grad_norm": 2.7307000160217285,
      "learning_rate": 2.4531484257871065e-07,
      "loss": 1.1837,
      "step": 1310
    },
    {
      "epoch": 124.85714285714286,
      "grad_norm": 4.455824851989746,
      "learning_rate": 2.455022488755622e-07,
      "loss": 1.1798,
      "step": 1311
    },
    {
      "epoch": 124.95238095238095,
      "grad_norm": 8.043598175048828,
      "learning_rate": 2.456896551724138e-07,
      "loss": 1.1953,
      "step": 1312
    },
    {
      "epoch": 124.95238095238095,
      "eval_accuracy": 0.7115024353690521,
      "eval_f1": 0.7737955346650999,
      "eval_loss": 0.5832556486129761,
      "eval_precision": 0.7080645161290322,
      "eval_recall": 0.852979274611399,
      "eval_roc_auc": 0.7464021301093839,
      "eval_runtime": 1.2072,
      "eval_samples_per_second": 2210.932,
      "eval_steps_per_second": 4.97,
      "step": 1312
    },
    {
      "epoch": 125.04761904761905,
      "grad_norm": 3.0496273040771484,
      "learning_rate": 2.458770614692654e-07,
      "loss": 1.1428,
      "step": 1313
    },
    {
      "epoch": 125.14285714285714,
      "grad_norm": 2.3517754077911377,
      "learning_rate": 2.46064467766117e-07,
      "loss": 1.1795,
      "step": 1314
    },
    {
      "epoch": 125.23809523809524,
      "grad_norm": 4.751769065856934,
      "learning_rate": 2.4625187406296853e-07,
      "loss": 1.1206,
      "step": 1315
    },
    {
      "epoch": 125.33333333333333,
      "grad_norm": 3.0632271766662598,
      "learning_rate": 2.464392803598201e-07,
      "loss": 1.196,
      "step": 1316
    },
    {
      "epoch": 125.42857142857143,
      "grad_norm": 7.766650199890137,
      "learning_rate": 2.466266866566717e-07,
      "loss": 1.2012,
      "step": 1317
    },
    {
      "epoch": 125.52380952380952,
      "grad_norm": 3.9962661266326904,
      "learning_rate": 2.4681409295352324e-07,
      "loss": 1.1835,
      "step": 1318
    },
    {
      "epoch": 125.61904761904762,
      "grad_norm": 4.209256649017334,
      "learning_rate": 2.470014992503748e-07,
      "loss": 1.1662,
      "step": 1319
    },
    {
      "epoch": 125.71428571428571,
      "grad_norm": 3.4023072719573975,
      "learning_rate": 2.471889055472264e-07,
      "loss": 1.1473,
      "step": 1320
    },
    {
      "epoch": 125.80952380952381,
      "grad_norm": 3.4137771129608154,
      "learning_rate": 2.4737631184407796e-07,
      "loss": 1.1915,
      "step": 1321
    },
    {
      "epoch": 125.9047619047619,
      "grad_norm": 3.2793519496917725,
      "learning_rate": 2.475637181409295e-07,
      "loss": 1.2126,
      "step": 1322
    },
    {
      "epoch": 126.0,
      "grad_norm": 4.163229465484619,
      "learning_rate": 2.477511244377811e-07,
      "loss": 1.126,
      "step": 1323
    },
    {
      "epoch": 126.0,
      "eval_accuracy": 0.7115024353690521,
      "eval_f1": 0.7719194312796208,
      "eval_loss": 0.582944929599762,
      "eval_precision": 0.7112445414847162,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7463943580886586,
      "eval_runtime": 1.3269,
      "eval_samples_per_second": 2011.487,
      "eval_steps_per_second": 4.522,
      "step": 1323
    },
    {
      "epoch": 126.0952380952381,
      "grad_norm": 2.1432619094848633,
      "learning_rate": 2.4793853073463273e-07,
      "loss": 1.199,
      "step": 1324
    },
    {
      "epoch": 126.19047619047619,
      "grad_norm": 3.3765182495117188,
      "learning_rate": 2.481259370314843e-07,
      "loss": 1.1606,
      "step": 1325
    },
    {
      "epoch": 126.28571428571429,
      "grad_norm": 3.6215734481811523,
      "learning_rate": 2.4831334332833584e-07,
      "loss": 1.2013,
      "step": 1326
    },
    {
      "epoch": 126.38095238095238,
      "grad_norm": 3.0442678928375244,
      "learning_rate": 2.4850074962518745e-07,
      "loss": 1.1996,
      "step": 1327
    },
    {
      "epoch": 126.47619047619048,
      "grad_norm": 3.2260961532592773,
      "learning_rate": 2.48688155922039e-07,
      "loss": 1.1508,
      "step": 1328
    },
    {
      "epoch": 126.57142857142857,
      "grad_norm": 8.212839126586914,
      "learning_rate": 2.4887556221889056e-07,
      "loss": 1.0953,
      "step": 1329
    },
    {
      "epoch": 126.66666666666667,
      "grad_norm": 3.9986352920532227,
      "learning_rate": 2.490629685157421e-07,
      "loss": 1.2293,
      "step": 1330
    },
    {
      "epoch": 126.76190476190476,
      "grad_norm": 3.904066324234009,
      "learning_rate": 2.492503748125937e-07,
      "loss": 1.1291,
      "step": 1331
    },
    {
      "epoch": 126.85714285714286,
      "grad_norm": 8.378240585327148,
      "learning_rate": 2.4943778110944527e-07,
      "loss": 1.1579,
      "step": 1332
    },
    {
      "epoch": 126.95238095238095,
      "grad_norm": 3.0266950130462646,
      "learning_rate": 2.4962518740629683e-07,
      "loss": 1.2049,
      "step": 1333
    },
    {
      "epoch": 126.95238095238095,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7713270142180095,
      "eval_loss": 0.5828430652618408,
      "eval_precision": 0.7106986899563319,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7465633275762809,
      "eval_runtime": 1.1244,
      "eval_samples_per_second": 2373.785,
      "eval_steps_per_second": 5.336,
      "step": 1333
    },
    {
      "epoch": 127.04761904761905,
      "grad_norm": 6.547031402587891,
      "learning_rate": 2.4981259370314843e-07,
      "loss": 1.1745,
      "step": 1334
    },
    {
      "epoch": 127.14285714285714,
      "grad_norm": 5.5182013511657715,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 1.1564,
      "step": 1335
    },
    {
      "epoch": 127.23809523809524,
      "grad_norm": 2.4143738746643066,
      "learning_rate": 2.501874062968516e-07,
      "loss": 1.1717,
      "step": 1336
    },
    {
      "epoch": 127.33333333333333,
      "grad_norm": 4.863409042358398,
      "learning_rate": 2.5037481259370315e-07,
      "loss": 1.1933,
      "step": 1337
    },
    {
      "epoch": 127.42857142857143,
      "grad_norm": 3.1126413345336914,
      "learning_rate": 2.5056221889055476e-07,
      "loss": 1.1717,
      "step": 1338
    },
    {
      "epoch": 127.52380952380952,
      "grad_norm": 4.216578006744385,
      "learning_rate": 2.507496251874063e-07,
      "loss": 1.2029,
      "step": 1339
    },
    {
      "epoch": 127.61904761904762,
      "grad_norm": 2.9910659790039062,
      "learning_rate": 2.5093703148425787e-07,
      "loss": 1.1934,
      "step": 1340
    },
    {
      "epoch": 127.71428571428571,
      "grad_norm": 2.8089070320129395,
      "learning_rate": 2.511244377811095e-07,
      "loss": 1.1704,
      "step": 1341
    },
    {
      "epoch": 127.80952380952381,
      "grad_norm": 5.142420291900635,
      "learning_rate": 2.5131184407796103e-07,
      "loss": 1.1208,
      "step": 1342
    },
    {
      "epoch": 127.9047619047619,
      "grad_norm": 5.560968399047852,
      "learning_rate": 2.514992503748126e-07,
      "loss": 1.1662,
      "step": 1343
    },
    {
      "epoch": 128.0,
      "grad_norm": 4.953802108764648,
      "learning_rate": 2.5168665667166414e-07,
      "loss": 1.2133,
      "step": 1344
    },
    {
      "epoch": 128.0,
      "eval_accuracy": 0.7115024353690521,
      "eval_f1": 0.7729952830188679,
      "eval_loss": 0.5828237533569336,
      "eval_precision": 0.7094155844155844,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7466491076568796,
      "eval_runtime": 1.1537,
      "eval_samples_per_second": 2313.46,
      "eval_steps_per_second": 5.201,
      "step": 1344
    },
    {
      "epoch": 128.0952380952381,
      "grad_norm": 2.2175867557525635,
      "learning_rate": 2.5187406296851575e-07,
      "loss": 1.1923,
      "step": 1345
    },
    {
      "epoch": 128.1904761904762,
      "grad_norm": 3.569380760192871,
      "learning_rate": 2.5206146926536735e-07,
      "loss": 1.1585,
      "step": 1346
    },
    {
      "epoch": 128.28571428571428,
      "grad_norm": 3.4245896339416504,
      "learning_rate": 2.522488755622189e-07,
      "loss": 1.1983,
      "step": 1347
    },
    {
      "epoch": 128.38095238095238,
      "grad_norm": 2.583385467529297,
      "learning_rate": 2.524362818590705e-07,
      "loss": 1.1682,
      "step": 1348
    },
    {
      "epoch": 128.47619047619048,
      "grad_norm": 3.7909340858459473,
      "learning_rate": 2.5262368815592207e-07,
      "loss": 1.1604,
      "step": 1349
    },
    {
      "epoch": 128.57142857142858,
      "grad_norm": 4.447571754455566,
      "learning_rate": 2.528110944527736e-07,
      "loss": 1.1556,
      "step": 1350
    },
    {
      "epoch": 128.66666666666666,
      "grad_norm": 4.511852264404297,
      "learning_rate": 2.529985007496252e-07,
      "loss": 1.1578,
      "step": 1351
    },
    {
      "epoch": 128.76190476190476,
      "grad_norm": 3.8867123126983643,
      "learning_rate": 2.531859070464768e-07,
      "loss": 1.1497,
      "step": 1352
    },
    {
      "epoch": 128.85714285714286,
      "grad_norm": 4.38187313079834,
      "learning_rate": 2.5337331334332834e-07,
      "loss": 1.1803,
      "step": 1353
    },
    {
      "epoch": 128.95238095238096,
      "grad_norm": 3.493037462234497,
      "learning_rate": 2.535607196401799e-07,
      "loss": 1.1846,
      "step": 1354
    },
    {
      "epoch": 128.95238095238096,
      "eval_accuracy": 0.710378418883477,
      "eval_f1": 0.7710986082321587,
      "eval_loss": 0.5827430486679077,
      "eval_precision": 0.7103109656301145,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7467225100748416,
      "eval_runtime": 1.1675,
      "eval_samples_per_second": 2285.99,
      "eval_steps_per_second": 5.139,
      "step": 1354
    },
    {
      "epoch": 129.04761904761904,
      "grad_norm": 6.4895548820495605,
      "learning_rate": 2.537481259370315e-07,
      "loss": 1.1259,
      "step": 1355
    },
    {
      "epoch": 129.14285714285714,
      "grad_norm": 3.8797340393066406,
      "learning_rate": 2.5393553223388306e-07,
      "loss": 1.1476,
      "step": 1356
    },
    {
      "epoch": 129.23809523809524,
      "grad_norm": 4.010172367095947,
      "learning_rate": 2.541229385307346e-07,
      "loss": 1.1496,
      "step": 1357
    },
    {
      "epoch": 129.33333333333334,
      "grad_norm": 4.8416595458984375,
      "learning_rate": 2.543103448275862e-07,
      "loss": 1.146,
      "step": 1358
    },
    {
      "epoch": 129.42857142857142,
      "grad_norm": 2.402299642562866,
      "learning_rate": 2.544977511244378e-07,
      "loss": 1.187,
      "step": 1359
    },
    {
      "epoch": 129.52380952380952,
      "grad_norm": 4.096165657043457,
      "learning_rate": 2.546851574212894e-07,
      "loss": 1.1673,
      "step": 1360
    },
    {
      "epoch": 129.61904761904762,
      "grad_norm": 6.944141864776611,
      "learning_rate": 2.5487256371814093e-07,
      "loss": 1.1702,
      "step": 1361
    },
    {
      "epoch": 129.71428571428572,
      "grad_norm": 3.276299476623535,
      "learning_rate": 2.5505997001499254e-07,
      "loss": 1.1306,
      "step": 1362
    },
    {
      "epoch": 129.8095238095238,
      "grad_norm": 9.009763717651367,
      "learning_rate": 2.552473763118441e-07,
      "loss": 1.2419,
      "step": 1363
    },
    {
      "epoch": 129.9047619047619,
      "grad_norm": 3.6103036403656006,
      "learning_rate": 2.5543478260869565e-07,
      "loss": 1.172,
      "step": 1364
    },
    {
      "epoch": 130.0,
      "grad_norm": 3.216857433319092,
      "learning_rate": 2.556221889055472e-07,
      "loss": 1.1975,
      "step": 1365
    },
    {
      "epoch": 130.0,
      "eval_accuracy": 0.7111277632071937,
      "eval_f1": 0.7726334414626954,
      "eval_loss": 0.5825395584106445,
      "eval_precision": 0.7092582566323768,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7471263672999424,
      "eval_runtime": 1.3705,
      "eval_samples_per_second": 1947.421,
      "eval_steps_per_second": 4.378,
      "step": 1365
    },
    {
      "epoch": 130.0952380952381,
      "grad_norm": 3.9114797115325928,
      "learning_rate": 2.558095952023988e-07,
      "loss": 1.1589,
      "step": 1366
    },
    {
      "epoch": 130.1904761904762,
      "grad_norm": 2.176995038986206,
      "learning_rate": 2.5599700149925037e-07,
      "loss": 1.1442,
      "step": 1367
    },
    {
      "epoch": 130.28571428571428,
      "grad_norm": 2.248365879058838,
      "learning_rate": 2.561844077961019e-07,
      "loss": 1.1856,
      "step": 1368
    },
    {
      "epoch": 130.38095238095238,
      "grad_norm": 2.4582395553588867,
      "learning_rate": 2.5637181409295353e-07,
      "loss": 1.1695,
      "step": 1369
    },
    {
      "epoch": 130.47619047619048,
      "grad_norm": 2.164710283279419,
      "learning_rate": 2.5655922038980514e-07,
      "loss": 1.1698,
      "step": 1370
    },
    {
      "epoch": 130.57142857142858,
      "grad_norm": 2.6641697883605957,
      "learning_rate": 2.567466266866567e-07,
      "loss": 1.1838,
      "step": 1371
    },
    {
      "epoch": 130.66666666666666,
      "grad_norm": 2.3906073570251465,
      "learning_rate": 2.5693403298350825e-07,
      "loss": 1.161,
      "step": 1372
    },
    {
      "epoch": 130.76190476190476,
      "grad_norm": 2.098816394805908,
      "learning_rate": 2.5712143928035985e-07,
      "loss": 1.1615,
      "step": 1373
    },
    {
      "epoch": 130.85714285714286,
      "grad_norm": 1.8485771417617798,
      "learning_rate": 2.573088455772114e-07,
      "loss": 1.1923,
      "step": 1374
    },
    {
      "epoch": 130.95238095238096,
      "grad_norm": 3.726562261581421,
      "learning_rate": 2.5749625187406296e-07,
      "loss": 1.1684,
      "step": 1375
    },
    {
      "epoch": 130.95238095238096,
      "eval_accuracy": 0.7100037467216186,
      "eval_f1": 0.7710059171597633,
      "eval_loss": 0.5824764370918274,
      "eval_precision": 0.7096949891067538,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7473465745538285,
      "eval_runtime": 1.1899,
      "eval_samples_per_second": 2242.957,
      "eval_steps_per_second": 5.042,
      "step": 1375
    },
    {
      "epoch": 131.04761904761904,
      "grad_norm": 3.8663816452026367,
      "learning_rate": 2.5768365817091457e-07,
      "loss": 1.1569,
      "step": 1376
    },
    {
      "epoch": 131.14285714285714,
      "grad_norm": 3.5459115505218506,
      "learning_rate": 2.578710644677661e-07,
      "loss": 1.1762,
      "step": 1377
    },
    {
      "epoch": 131.23809523809524,
      "grad_norm": 3.518721580505371,
      "learning_rate": 2.580584707646177e-07,
      "loss": 1.1732,
      "step": 1378
    },
    {
      "epoch": 131.33333333333334,
      "grad_norm": 4.939159393310547,
      "learning_rate": 2.582458770614693e-07,
      "loss": 1.1596,
      "step": 1379
    },
    {
      "epoch": 131.42857142857142,
      "grad_norm": 3.7521708011627197,
      "learning_rate": 2.5843328335832084e-07,
      "loss": 1.1283,
      "step": 1380
    },
    {
      "epoch": 131.52380952380952,
      "grad_norm": 2.7267091274261475,
      "learning_rate": 2.5862068965517245e-07,
      "loss": 1.1724,
      "step": 1381
    },
    {
      "epoch": 131.61904761904762,
      "grad_norm": 6.932819366455078,
      "learning_rate": 2.58808095952024e-07,
      "loss": 1.1694,
      "step": 1382
    },
    {
      "epoch": 131.71428571428572,
      "grad_norm": 2.6388416290283203,
      "learning_rate": 2.589955022488756e-07,
      "loss": 1.171,
      "step": 1383
    },
    {
      "epoch": 131.8095238095238,
      "grad_norm": 2.6707115173339844,
      "learning_rate": 2.5918290854572716e-07,
      "loss": 1.2032,
      "step": 1384
    },
    {
      "epoch": 131.9047619047619,
      "grad_norm": 2.9166555404663086,
      "learning_rate": 2.593703148425787e-07,
      "loss": 1.1955,
      "step": 1385
    },
    {
      "epoch": 132.0,
      "grad_norm": 8.50768756866455,
      "learning_rate": 2.595577211394303e-07,
      "loss": 1.1999,
      "step": 1386
    },
    {
      "epoch": 132.0,
      "eval_accuracy": 0.7122517796927689,
      "eval_f1": 0.7743830787309048,
      "eval_loss": 0.5823792815208435,
      "eval_precision": 0.7086021505376344,
      "eval_recall": 0.8536269430051814,
      "eval_roc_auc": 0.7480109383995394,
      "eval_runtime": 1.1519,
      "eval_samples_per_second": 2316.954,
      "eval_steps_per_second": 5.209,
      "step": 1386
    },
    {
      "epoch": 132.0952380952381,
      "grad_norm": 2.550859212875366,
      "learning_rate": 2.597451274362819e-07,
      "loss": 1.1689,
      "step": 1387
    },
    {
      "epoch": 132.1904761904762,
      "grad_norm": 3.9516186714172363,
      "learning_rate": 2.5993253373313343e-07,
      "loss": 1.1599,
      "step": 1388
    },
    {
      "epoch": 132.28571428571428,
      "grad_norm": 8.39937686920166,
      "learning_rate": 2.60119940029985e-07,
      "loss": 1.2044,
      "step": 1389
    },
    {
      "epoch": 132.38095238095238,
      "grad_norm": 9.259173393249512,
      "learning_rate": 2.603073463268366e-07,
      "loss": 1.1766,
      "step": 1390
    },
    {
      "epoch": 132.47619047619048,
      "grad_norm": 3.003726005554199,
      "learning_rate": 2.6049475262368815e-07,
      "loss": 1.1569,
      "step": 1391
    },
    {
      "epoch": 132.57142857142858,
      "grad_norm": 3.2033815383911133,
      "learning_rate": 2.606821589205397e-07,
      "loss": 1.1957,
      "step": 1392
    },
    {
      "epoch": 132.66666666666666,
      "grad_norm": 2.2865841388702393,
      "learning_rate": 2.608695652173913e-07,
      "loss": 1.1646,
      "step": 1393
    },
    {
      "epoch": 132.76190476190476,
      "grad_norm": 3.505711317062378,
      "learning_rate": 2.610569715142429e-07,
      "loss": 1.1766,
      "step": 1394
    },
    {
      "epoch": 132.85714285714286,
      "grad_norm": 5.692092418670654,
      "learning_rate": 2.612443778110945e-07,
      "loss": 1.1545,
      "step": 1395
    },
    {
      "epoch": 132.95238095238096,
      "grad_norm": 6.441229343414307,
      "learning_rate": 2.6143178410794603e-07,
      "loss": 1.1214,
      "step": 1396
    },
    {
      "epoch": 132.95238095238096,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7713270142180095,
      "eval_loss": 0.5824524760246277,
      "eval_precision": 0.7106986899563319,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7477320092112838,
      "eval_runtime": 1.162,
      "eval_samples_per_second": 2296.861,
      "eval_steps_per_second": 5.163,
      "step": 1396
    },
    {
      "epoch": 133.04761904761904,
      "grad_norm": 3.6968135833740234,
      "learning_rate": 2.6161919040479764e-07,
      "loss": 1.1982,
      "step": 1397
    },
    {
      "epoch": 133.14285714285714,
      "grad_norm": 2.27152156829834,
      "learning_rate": 2.618065967016492e-07,
      "loss": 1.1361,
      "step": 1398
    },
    {
      "epoch": 133.23809523809524,
      "grad_norm": 5.761772155761719,
      "learning_rate": 2.6199400299850075e-07,
      "loss": 1.2019,
      "step": 1399
    },
    {
      "epoch": 133.33333333333334,
      "grad_norm": 6.545323848724365,
      "learning_rate": 2.6218140929535235e-07,
      "loss": 1.1749,
      "step": 1400
    },
    {
      "epoch": 133.42857142857142,
      "grad_norm": 10.077860832214355,
      "learning_rate": 2.623688155922039e-07,
      "loss": 1.2102,
      "step": 1401
    },
    {
      "epoch": 133.52380952380952,
      "grad_norm": 2.281158208847046,
      "learning_rate": 2.6255622188905546e-07,
      "loss": 1.1869,
      "step": 1402
    },
    {
      "epoch": 133.61904761904762,
      "grad_norm": 3.5024521350860596,
      "learning_rate": 2.62743628185907e-07,
      "loss": 1.1421,
      "step": 1403
    },
    {
      "epoch": 133.71428571428572,
      "grad_norm": 2.298860549926758,
      "learning_rate": 2.629310344827586e-07,
      "loss": 1.1538,
      "step": 1404
    },
    {
      "epoch": 133.8095238095238,
      "grad_norm": 4.36176872253418,
      "learning_rate": 2.6311844077961023e-07,
      "loss": 1.1518,
      "step": 1405
    },
    {
      "epoch": 133.9047619047619,
      "grad_norm": 2.632678508758545,
      "learning_rate": 2.633058470764618e-07,
      "loss": 1.1745,
      "step": 1406
    },
    {
      "epoch": 134.0,
      "grad_norm": 4.977126598358154,
      "learning_rate": 2.634932533733134e-07,
      "loss": 1.1899,
      "step": 1407
    },
    {
      "epoch": 134.0,
      "eval_accuracy": 0.7115024353690521,
      "eval_f1": 0.7725930301240401,
      "eval_loss": 0.5824618339538574,
      "eval_precision": 0.7100977198697068,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7481289579735175,
      "eval_runtime": 1.2194,
      "eval_samples_per_second": 2188.826,
      "eval_steps_per_second": 4.921,
      "step": 1407
    },
    {
      "epoch": 134.0952380952381,
      "grad_norm": 3.8728256225585938,
      "learning_rate": 2.6368065967016495e-07,
      "loss": 1.1517,
      "step": 1408
    },
    {
      "epoch": 134.1904761904762,
      "grad_norm": 6.059358596801758,
      "learning_rate": 2.638680659670165e-07,
      "loss": 1.1919,
      "step": 1409
    },
    {
      "epoch": 134.28571428571428,
      "grad_norm": 4.765652656555176,
      "learning_rate": 2.6405547226386806e-07,
      "loss": 1.2029,
      "step": 1410
    },
    {
      "epoch": 134.38095238095238,
      "grad_norm": 4.513894557952881,
      "learning_rate": 2.6424287856071966e-07,
      "loss": 1.1677,
      "step": 1411
    },
    {
      "epoch": 134.47619047619048,
      "grad_norm": 5.389193534851074,
      "learning_rate": 2.644302848575712e-07,
      "loss": 1.1616,
      "step": 1412
    },
    {
      "epoch": 134.57142857142858,
      "grad_norm": 2.583965301513672,
      "learning_rate": 2.6461769115442277e-07,
      "loss": 1.1809,
      "step": 1413
    },
    {
      "epoch": 134.66666666666666,
      "grad_norm": 3.9450693130493164,
      "learning_rate": 2.648050974512744e-07,
      "loss": 1.1799,
      "step": 1414
    },
    {
      "epoch": 134.76190476190476,
      "grad_norm": 6.976175308227539,
      "learning_rate": 2.6499250374812594e-07,
      "loss": 1.1621,
      "step": 1415
    },
    {
      "epoch": 134.85714285714286,
      "grad_norm": 2.458829879760742,
      "learning_rate": 2.6517991004497754e-07,
      "loss": 1.181,
      "step": 1416
    },
    {
      "epoch": 134.95238095238096,
      "grad_norm": 3.2024972438812256,
      "learning_rate": 2.653673163418291e-07,
      "loss": 1.1247,
      "step": 1417
    },
    {
      "epoch": 134.95238095238096,
      "eval_accuracy": 0.710378418883477,
      "eval_f1": 0.7721780135573239,
      "eval_loss": 0.5821589827537537,
      "eval_precision": 0.7084910762574365,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7485892343120323,
      "eval_runtime": 1.1491,
      "eval_samples_per_second": 2322.722,
      "eval_steps_per_second": 5.222,
      "step": 1417
    },
    {
      "epoch": 135.04761904761904,
      "grad_norm": 3.9485716819763184,
      "learning_rate": 2.655547226386807e-07,
      "loss": 1.1981,
      "step": 1418
    },
    {
      "epoch": 135.14285714285714,
      "grad_norm": 3.075244665145874,
      "learning_rate": 2.6574212893553226e-07,
      "loss": 1.1506,
      "step": 1419
    },
    {
      "epoch": 135.23809523809524,
      "grad_norm": 3.1542418003082275,
      "learning_rate": 2.659295352323838e-07,
      "loss": 1.2158,
      "step": 1420
    },
    {
      "epoch": 135.33333333333334,
      "grad_norm": 2.9404492378234863,
      "learning_rate": 2.661169415292354e-07,
      "loss": 1.1874,
      "step": 1421
    },
    {
      "epoch": 135.42857142857142,
      "grad_norm": 2.844932794570923,
      "learning_rate": 2.66304347826087e-07,
      "loss": 1.1846,
      "step": 1422
    },
    {
      "epoch": 135.52380952380952,
      "grad_norm": 3.8234691619873047,
      "learning_rate": 2.6649175412293853e-07,
      "loss": 1.1572,
      "step": 1423
    },
    {
      "epoch": 135.61904761904762,
      "grad_norm": 2.9241669178009033,
      "learning_rate": 2.666791604197901e-07,
      "loss": 1.1255,
      "step": 1424
    },
    {
      "epoch": 135.71428571428572,
      "grad_norm": 6.175973892211914,
      "learning_rate": 2.668665667166417e-07,
      "loss": 1.1565,
      "step": 1425
    },
    {
      "epoch": 135.8095238095238,
      "grad_norm": 7.453120708465576,
      "learning_rate": 2.6705397301349325e-07,
      "loss": 1.1774,
      "step": 1426
    },
    {
      "epoch": 135.9047619047619,
      "grad_norm": 5.209527969360352,
      "learning_rate": 2.6724137931034485e-07,
      "loss": 1.149,
      "step": 1427
    },
    {
      "epoch": 136.0,
      "grad_norm": 7.697098255157471,
      "learning_rate": 2.674287856071964e-07,
      "loss": 1.2117,
      "step": 1428
    },
    {
      "epoch": 136.0,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.7737569873492204,
      "eval_loss": 0.5818696618080139,
      "eval_precision": 0.7088948787061995,
      "eval_recall": 0.8516839378238342,
      "eval_roc_auc": 0.7490256188831319,
      "eval_runtime": 1.1367,
      "eval_samples_per_second": 2348.011,
      "eval_steps_per_second": 5.278,
      "step": 1428
    },
    {
      "epoch": 136.0952380952381,
      "grad_norm": 6.136030673980713,
      "learning_rate": 2.67616191904048e-07,
      "loss": 1.2038,
      "step": 1429
    },
    {
      "epoch": 136.1904761904762,
      "grad_norm": 2.5705273151397705,
      "learning_rate": 2.6780359820089957e-07,
      "loss": 1.1662,
      "step": 1430
    },
    {
      "epoch": 136.28571428571428,
      "grad_norm": 5.877147674560547,
      "learning_rate": 2.679910044977511e-07,
      "loss": 1.1392,
      "step": 1431
    },
    {
      "epoch": 136.38095238095238,
      "grad_norm": 2.156723976135254,
      "learning_rate": 2.6817841079460273e-07,
      "loss": 1.1652,
      "step": 1432
    },
    {
      "epoch": 136.47619047619048,
      "grad_norm": 3.0156610012054443,
      "learning_rate": 2.683658170914543e-07,
      "loss": 1.1567,
      "step": 1433
    },
    {
      "epoch": 136.57142857142858,
      "grad_norm": 2.8673226833343506,
      "learning_rate": 2.6855322338830584e-07,
      "loss": 1.1952,
      "step": 1434
    },
    {
      "epoch": 136.66666666666666,
      "grad_norm": 3.6061689853668213,
      "learning_rate": 2.6874062968515745e-07,
      "loss": 1.1871,
      "step": 1435
    },
    {
      "epoch": 136.76190476190476,
      "grad_norm": 3.9926493167877197,
      "learning_rate": 2.68928035982009e-07,
      "loss": 1.1449,
      "step": 1436
    },
    {
      "epoch": 136.85714285714286,
      "grad_norm": 4.736005783081055,
      "learning_rate": 2.6911544227886056e-07,
      "loss": 1.2134,
      "step": 1437
    },
    {
      "epoch": 136.95238095238096,
      "grad_norm": 2.764103651046753,
      "learning_rate": 2.693028485757121e-07,
      "loss": 1.1678,
      "step": 1438
    },
    {
      "epoch": 136.95238095238096,
      "eval_accuracy": 0.7096290745597602,
      "eval_f1": 0.7698247698247698,
      "eval_loss": 0.5817188620567322,
      "eval_precision": 0.7109160724081185,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7489231433506045,
      "eval_runtime": 1.193,
      "eval_samples_per_second": 2237.207,
      "eval_steps_per_second": 5.029,
      "step": 1438
    },
    {
      "epoch": 137.04761904761904,
      "grad_norm": 3.620636224746704,
      "learning_rate": 2.694902548725637e-07,
      "loss": 1.1886,
      "step": 1439
    },
    {
      "epoch": 137.14285714285714,
      "grad_norm": 4.410121440887451,
      "learning_rate": 2.6967766116941533e-07,
      "loss": 1.1758,
      "step": 1440
    },
    {
      "epoch": 137.23809523809524,
      "grad_norm": 2.61210036277771,
      "learning_rate": 2.698650674662669e-07,
      "loss": 1.1506,
      "step": 1441
    },
    {
      "epoch": 137.33333333333334,
      "grad_norm": 3.4050190448760986,
      "learning_rate": 2.700524737631185e-07,
      "loss": 1.1437,
      "step": 1442
    },
    {
      "epoch": 137.42857142857142,
      "grad_norm": 2.654712438583374,
      "learning_rate": 2.7023988005997004e-07,
      "loss": 1.1457,
      "step": 1443
    },
    {
      "epoch": 137.52380952380952,
      "grad_norm": 3.8472094535827637,
      "learning_rate": 2.704272863568216e-07,
      "loss": 1.1954,
      "step": 1444
    },
    {
      "epoch": 137.61904761904762,
      "grad_norm": 3.686859130859375,
      "learning_rate": 2.7061469265367315e-07,
      "loss": 1.1606,
      "step": 1445
    },
    {
      "epoch": 137.71428571428572,
      "grad_norm": 2.689823627471924,
      "learning_rate": 2.7080209895052476e-07,
      "loss": 1.1483,
      "step": 1446
    },
    {
      "epoch": 137.8095238095238,
      "grad_norm": 3.9015514850616455,
      "learning_rate": 2.709895052473763e-07,
      "loss": 1.1997,
      "step": 1447
    },
    {
      "epoch": 137.9047619047619,
      "grad_norm": 3.536795139312744,
      "learning_rate": 2.7117691154422787e-07,
      "loss": 1.2187,
      "step": 1448
    },
    {
      "epoch": 138.0,
      "grad_norm": 3.393120288848877,
      "learning_rate": 2.713643178410795e-07,
      "loss": 1.1405,
      "step": 1449
    },
    {
      "epoch": 138.0,
      "eval_accuracy": 0.7100037467216186,
      "eval_f1": 0.7718160377358491,
      "eval_loss": 0.5816904902458191,
      "eval_precision": 0.7083333333333334,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7490673575129533,
      "eval_runtime": 1.2137,
      "eval_samples_per_second": 2199.026,
      "eval_steps_per_second": 4.943,
      "step": 1449
    },
    {
      "epoch": 138.0952380952381,
      "grad_norm": 6.070676803588867,
      "learning_rate": 2.7155172413793103e-07,
      "loss": 1.188,
      "step": 1450
    },
    {
      "epoch": 138.1904761904762,
      "grad_norm": 3.7084219455718994,
      "learning_rate": 2.7173913043478264e-07,
      "loss": 1.1275,
      "step": 1451
    },
    {
      "epoch": 138.28571428571428,
      "grad_norm": 3.660405397415161,
      "learning_rate": 2.719265367316342e-07,
      "loss": 1.1814,
      "step": 1452
    },
    {
      "epoch": 138.38095238095238,
      "grad_norm": 5.310576915740967,
      "learning_rate": 2.721139430284858e-07,
      "loss": 1.2151,
      "step": 1453
    },
    {
      "epoch": 138.47619047619048,
      "grad_norm": 3.6512293815612793,
      "learning_rate": 2.7230134932533735e-07,
      "loss": 1.1324,
      "step": 1454
    },
    {
      "epoch": 138.57142857142858,
      "grad_norm": 5.349870204925537,
      "learning_rate": 2.724887556221889e-07,
      "loss": 1.1598,
      "step": 1455
    },
    {
      "epoch": 138.66666666666666,
      "grad_norm": 3.535796880722046,
      "learning_rate": 2.726761619190405e-07,
      "loss": 1.2146,
      "step": 1456
    },
    {
      "epoch": 138.76190476190476,
      "grad_norm": 3.3005921840667725,
      "learning_rate": 2.7286356821589207e-07,
      "loss": 1.1762,
      "step": 1457
    },
    {
      "epoch": 138.85714285714286,
      "grad_norm": 5.350109577178955,
      "learning_rate": 2.730509745127436e-07,
      "loss": 1.1281,
      "step": 1458
    },
    {
      "epoch": 138.95238095238096,
      "grad_norm": 2.827852487564087,
      "learning_rate": 2.732383808095952e-07,
      "loss": 1.1908,
      "step": 1459
    },
    {
      "epoch": 138.95238095238096,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7720023626698169,
      "eval_loss": 0.5818772315979004,
      "eval_precision": 0.7095548317046688,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7488477259643063,
      "eval_runtime": 1.3464,
      "eval_samples_per_second": 1982.292,
      "eval_steps_per_second": 4.456,
      "step": 1459
    },
    {
      "epoch": 139.04761904761904,
      "grad_norm": 3.779970645904541,
      "learning_rate": 2.734257871064468e-07,
      "loss": 1.1934,
      "step": 1460
    },
    {
      "epoch": 139.14285714285714,
      "grad_norm": 4.349908828735352,
      "learning_rate": 2.7361319340329834e-07,
      "loss": 1.1311,
      "step": 1461
    },
    {
      "epoch": 139.23809523809524,
      "grad_norm": 6.623378753662109,
      "learning_rate": 2.7380059970014995e-07,
      "loss": 1.1791,
      "step": 1462
    },
    {
      "epoch": 139.33333333333334,
      "grad_norm": 5.597513198852539,
      "learning_rate": 2.739880059970015e-07,
      "loss": 1.1463,
      "step": 1463
    },
    {
      "epoch": 139.42857142857142,
      "grad_norm": 2.972928047180176,
      "learning_rate": 2.741754122938531e-07,
      "loss": 1.1486,
      "step": 1464
    },
    {
      "epoch": 139.52380952380952,
      "grad_norm": 5.351190567016602,
      "learning_rate": 2.7436281859070467e-07,
      "loss": 1.2003,
      "step": 1465
    },
    {
      "epoch": 139.61904761904762,
      "grad_norm": 2.219123363494873,
      "learning_rate": 2.745502248875562e-07,
      "loss": 1.1734,
      "step": 1466
    },
    {
      "epoch": 139.71428571428572,
      "grad_norm": 2.1666510105133057,
      "learning_rate": 2.7473763118440783e-07,
      "loss": 1.1937,
      "step": 1467
    },
    {
      "epoch": 139.8095238095238,
      "grad_norm": 3.52593994140625,
      "learning_rate": 2.749250374812594e-07,
      "loss": 1.1555,
      "step": 1468
    },
    {
      "epoch": 139.9047619047619,
      "grad_norm": 3.0705714225769043,
      "learning_rate": 2.7511244377811094e-07,
      "loss": 1.1667,
      "step": 1469
    },
    {
      "epoch": 140.0,
      "grad_norm": 2.8801708221435547,
      "learning_rate": 2.7529985007496254e-07,
      "loss": 1.1483,
      "step": 1470
    },
    {
      "epoch": 140.0,
      "eval_accuracy": 0.7111277632071937,
      "eval_f1": 0.7727674624226348,
      "eval_loss": 0.5820427536964417,
      "eval_precision": 0.7090319091400757,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7490020149683363,
      "eval_runtime": 1.1074,
      "eval_samples_per_second": 2410.145,
      "eval_steps_per_second": 5.418,
      "step": 1470
    },
    {
      "epoch": 140.0952380952381,
      "grad_norm": 2.7389588356018066,
      "learning_rate": 2.754872563718141e-07,
      "loss": 1.2047,
      "step": 1471
    },
    {
      "epoch": 140.1904761904762,
      "grad_norm": 2.8530631065368652,
      "learning_rate": 2.7567466266866565e-07,
      "loss": 1.19,
      "step": 1472
    },
    {
      "epoch": 140.28571428571428,
      "grad_norm": 2.755894899368286,
      "learning_rate": 2.758620689655172e-07,
      "loss": 1.1528,
      "step": 1473
    },
    {
      "epoch": 140.38095238095238,
      "grad_norm": 4.2909464836120605,
      "learning_rate": 2.760494752623688e-07,
      "loss": 1.1439,
      "step": 1474
    },
    {
      "epoch": 140.47619047619048,
      "grad_norm": 4.200039863586426,
      "learning_rate": 2.762368815592204e-07,
      "loss": 1.1652,
      "step": 1475
    },
    {
      "epoch": 140.57142857142858,
      "grad_norm": 6.071081638336182,
      "learning_rate": 2.76424287856072e-07,
      "loss": 1.1684,
      "step": 1476
    },
    {
      "epoch": 140.66666666666666,
      "grad_norm": 3.9966976642608643,
      "learning_rate": 2.766116941529236e-07,
      "loss": 1.1467,
      "step": 1477
    },
    {
      "epoch": 140.76190476190476,
      "grad_norm": 2.3274450302124023,
      "learning_rate": 2.7679910044977514e-07,
      "loss": 1.1746,
      "step": 1478
    },
    {
      "epoch": 140.85714285714286,
      "grad_norm": 5.671662330627441,
      "learning_rate": 2.769865067466267e-07,
      "loss": 1.153,
      "step": 1479
    },
    {
      "epoch": 140.95238095238096,
      "grad_norm": 4.11067533493042,
      "learning_rate": 2.7717391304347825e-07,
      "loss": 1.1295,
      "step": 1480
    },
    {
      "epoch": 140.95238095238096,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.7733569112879458,
      "eval_loss": 0.581862211227417,
      "eval_precision": 0.709572742022715,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7493468624064479,
      "eval_runtime": 1.1773,
      "eval_samples_per_second": 2267.004,
      "eval_steps_per_second": 5.096,
      "step": 1480
    },
    {
      "epoch": 141.04761904761904,
      "grad_norm": 3.680772542953491,
      "learning_rate": 2.7736131934032985e-07,
      "loss": 1.1456,
      "step": 1481
    },
    {
      "epoch": 141.14285714285714,
      "grad_norm": 3.461637258529663,
      "learning_rate": 2.775487256371814e-07,
      "loss": 1.1318,
      "step": 1482
    },
    {
      "epoch": 141.23809523809524,
      "grad_norm": 7.937395095825195,
      "learning_rate": 2.7773613193403296e-07,
      "loss": 1.1882,
      "step": 1483
    },
    {
      "epoch": 141.33333333333334,
      "grad_norm": 2.5857739448547363,
      "learning_rate": 2.7792353823088457e-07,
      "loss": 1.155,
      "step": 1484
    },
    {
      "epoch": 141.42857142857142,
      "grad_norm": 4.173889636993408,
      "learning_rate": 2.781109445277361e-07,
      "loss": 1.144,
      "step": 1485
    },
    {
      "epoch": 141.52380952380952,
      "grad_norm": 2.58803653717041,
      "learning_rate": 2.7829835082458773e-07,
      "loss": 1.1905,
      "step": 1486
    },
    {
      "epoch": 141.61904761904762,
      "grad_norm": 7.0563530921936035,
      "learning_rate": 2.7848575712143934e-07,
      "loss": 1.181,
      "step": 1487
    },
    {
      "epoch": 141.71428571428572,
      "grad_norm": 4.463258743286133,
      "learning_rate": 2.786731634182909e-07,
      "loss": 1.2167,
      "step": 1488
    },
    {
      "epoch": 141.8095238095238,
      "grad_norm": 3.7728588581085205,
      "learning_rate": 2.7886056971514245e-07,
      "loss": 1.1736,
      "step": 1489
    },
    {
      "epoch": 141.9047619047619,
      "grad_norm": 3.2672605514526367,
      "learning_rate": 2.79047976011994e-07,
      "loss": 1.1645,
      "step": 1490
    },
    {
      "epoch": 142.0,
      "grad_norm": 4.211281776428223,
      "learning_rate": 2.792353823088456e-07,
      "loss": 1.1652,
      "step": 1491
    },
    {
      "epoch": 142.0,
      "eval_accuracy": 0.7122517796927689,
      "eval_f1": 0.7741176470588236,
      "eval_loss": 0.5816420316696167,
      "eval_precision": 0.709051724137931,
      "eval_recall": 0.8523316062176166,
      "eval_roc_auc": 0.7496450777202072,
      "eval_runtime": 1.2197,
      "eval_samples_per_second": 2188.197,
      "eval_steps_per_second": 4.919,
      "step": 1491
    },
    {
      "epoch": 142.0952380952381,
      "grad_norm": 5.845593452453613,
      "learning_rate": 2.7942278860569717e-07,
      "loss": 1.1628,
      "step": 1492
    },
    {
      "epoch": 142.1904761904762,
      "grad_norm": 4.2243170738220215,
      "learning_rate": 2.796101949025487e-07,
      "loss": 1.1997,
      "step": 1493
    },
    {
      "epoch": 142.28571428571428,
      "grad_norm": 2.975741147994995,
      "learning_rate": 2.7979760119940033e-07,
      "loss": 1.1822,
      "step": 1494
    },
    {
      "epoch": 142.38095238095238,
      "grad_norm": 3.2676541805267334,
      "learning_rate": 2.799850074962519e-07,
      "loss": 1.1531,
      "step": 1495
    },
    {
      "epoch": 142.47619047619048,
      "grad_norm": 3.6477251052856445,
      "learning_rate": 2.8017241379310344e-07,
      "loss": 1.1689,
      "step": 1496
    },
    {
      "epoch": 142.57142857142858,
      "grad_norm": 3.187495231628418,
      "learning_rate": 2.8035982008995504e-07,
      "loss": 1.1464,
      "step": 1497
    },
    {
      "epoch": 142.66666666666666,
      "grad_norm": 8.654559135437012,
      "learning_rate": 2.805472263868066e-07,
      "loss": 1.2071,
      "step": 1498
    },
    {
      "epoch": 142.76190476190476,
      "grad_norm": 3.998518228530884,
      "learning_rate": 2.807346326836582e-07,
      "loss": 1.1739,
      "step": 1499
    },
    {
      "epoch": 142.85714285714286,
      "grad_norm": 2.40251088142395,
      "learning_rate": 2.8092203898050976e-07,
      "loss": 1.1866,
      "step": 1500
    },
    {
      "epoch": 142.95238095238096,
      "grad_norm": 6.565639019012451,
      "learning_rate": 2.8110944527736137e-07,
      "loss": 1.1773,
      "step": 1501
    },
    {
      "epoch": 142.95238095238096,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.7734904270986745,
      "eval_loss": 0.5813953280448914,
      "eval_precision": 0.7093462992976769,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7498618307426598,
      "eval_runtime": 1.3528,
      "eval_samples_per_second": 1972.9,
      "eval_steps_per_second": 4.435,
      "step": 1501
    },
    {
      "epoch": 143.04761904761904,
      "grad_norm": 2.458740472793579,
      "learning_rate": 2.812968515742129e-07,
      "loss": 1.1938,
      "step": 1502
    },
    {
      "epoch": 143.14285714285714,
      "grad_norm": 7.570618629455566,
      "learning_rate": 2.814842578710645e-07,
      "loss": 1.2069,
      "step": 1503
    },
    {
      "epoch": 143.23809523809524,
      "grad_norm": 6.107430934906006,
      "learning_rate": 2.8167166416791603e-07,
      "loss": 1.176,
      "step": 1504
    },
    {
      "epoch": 143.33333333333334,
      "grad_norm": 4.109244346618652,
      "learning_rate": 2.8185907046476764e-07,
      "loss": 1.1538,
      "step": 1505
    },
    {
      "epoch": 143.42857142857142,
      "grad_norm": 3.977546453475952,
      "learning_rate": 2.820464767616192e-07,
      "loss": 1.1966,
      "step": 1506
    },
    {
      "epoch": 143.52380952380952,
      "grad_norm": 5.082589626312256,
      "learning_rate": 2.8223388305847075e-07,
      "loss": 1.117,
      "step": 1507
    },
    {
      "epoch": 143.61904761904762,
      "grad_norm": 6.457265377044678,
      "learning_rate": 2.8242128935532235e-07,
      "loss": 1.146,
      "step": 1508
    },
    {
      "epoch": 143.71428571428572,
      "grad_norm": 4.3683648109436035,
      "learning_rate": 2.826086956521739e-07,
      "loss": 1.1888,
      "step": 1509
    },
    {
      "epoch": 143.8095238095238,
      "grad_norm": 2.4352548122406006,
      "learning_rate": 2.827961019490255e-07,
      "loss": 1.1526,
      "step": 1510
    },
    {
      "epoch": 143.9047619047619,
      "grad_norm": 8.339164733886719,
      "learning_rate": 2.8298350824587707e-07,
      "loss": 1.1664,
      "step": 1511
    },
    {
      "epoch": 144.0,
      "grad_norm": 3.4991281032562256,
      "learning_rate": 2.831709145427287e-07,
      "loss": 1.1515,
      "step": 1512
    },
    {
      "epoch": 144.0,
      "eval_accuracy": 0.7096290745597602,
      "eval_f1": 0.7714538484222944,
      "eval_loss": 0.5811057686805725,
      "eval_precision": 0.7081754195993503,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.750195739781232,
      "eval_runtime": 1.1549,
      "eval_samples_per_second": 2311.007,
      "eval_steps_per_second": 5.195,
      "step": 1512
    },
    {
      "epoch": 144.0952380952381,
      "grad_norm": 9.736416816711426,
      "learning_rate": 2.8335832083958023e-07,
      "loss": 1.2106,
      "step": 1513
    },
    {
      "epoch": 144.1904761904762,
      "grad_norm": 3.4234447479248047,
      "learning_rate": 2.835457271364318e-07,
      "loss": 1.164,
      "step": 1514
    },
    {
      "epoch": 144.28571428571428,
      "grad_norm": 7.636989593505859,
      "learning_rate": 2.837331334332834e-07,
      "loss": 1.2098,
      "step": 1515
    },
    {
      "epoch": 144.38095238095238,
      "grad_norm": 3.0599682331085205,
      "learning_rate": 2.8392053973013495e-07,
      "loss": 1.1746,
      "step": 1516
    },
    {
      "epoch": 144.47619047619048,
      "grad_norm": 3.2300186157226562,
      "learning_rate": 2.841079460269865e-07,
      "loss": 1.1701,
      "step": 1517
    },
    {
      "epoch": 144.57142857142858,
      "grad_norm": 2.491485118865967,
      "learning_rate": 2.8429535232383806e-07,
      "loss": 1.1558,
      "step": 1518
    },
    {
      "epoch": 144.66666666666666,
      "grad_norm": 5.057804584503174,
      "learning_rate": 2.8448275862068967e-07,
      "loss": 1.1231,
      "step": 1519
    },
    {
      "epoch": 144.76190476190476,
      "grad_norm": 3.5403995513916016,
      "learning_rate": 2.846701649175412e-07,
      "loss": 1.1828,
      "step": 1520
    },
    {
      "epoch": 144.85714285714286,
      "grad_norm": 2.31864333152771,
      "learning_rate": 2.8485757121439283e-07,
      "loss": 1.1614,
      "step": 1521
    },
    {
      "epoch": 144.95238095238096,
      "grad_norm": 6.159409999847412,
      "learning_rate": 2.8504497751124444e-07,
      "loss": 1.1686,
      "step": 1522
    },
    {
      "epoch": 144.95238095238096,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7720023626698169,
      "eval_loss": 0.5807437896728516,
      "eval_precision": 0.7095548317046688,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7504884858952218,
      "eval_runtime": 1.1783,
      "eval_samples_per_second": 2265.14,
      "eval_steps_per_second": 5.092,
      "step": 1522
    },
    {
      "epoch": 145.04761904761904,
      "grad_norm": 6.86565637588501,
      "learning_rate": 2.85232383808096e-07,
      "loss": 1.1791,
      "step": 1523
    },
    {
      "epoch": 145.14285714285714,
      "grad_norm": 3.2620322704315186,
      "learning_rate": 2.8541979010494754e-07,
      "loss": 1.1668,
      "step": 1524
    },
    {
      "epoch": 145.23809523809524,
      "grad_norm": 3.2191574573516846,
      "learning_rate": 2.856071964017991e-07,
      "loss": 1.1789,
      "step": 1525
    },
    {
      "epoch": 145.33333333333334,
      "grad_norm": 2.270550012588501,
      "learning_rate": 2.857946026986507e-07,
      "loss": 1.2023,
      "step": 1526
    },
    {
      "epoch": 145.42857142857142,
      "grad_norm": 4.476311206817627,
      "learning_rate": 2.8598200899550226e-07,
      "loss": 1.1375,
      "step": 1527
    },
    {
      "epoch": 145.52380952380952,
      "grad_norm": 3.982914447784424,
      "learning_rate": 2.861694152923538e-07,
      "loss": 1.1816,
      "step": 1528
    },
    {
      "epoch": 145.61904761904762,
      "grad_norm": 4.971301555633545,
      "learning_rate": 2.863568215892054e-07,
      "loss": 1.1739,
      "step": 1529
    },
    {
      "epoch": 145.71428571428572,
      "grad_norm": 4.544134616851807,
      "learning_rate": 2.86544227886057e-07,
      "loss": 1.1737,
      "step": 1530
    },
    {
      "epoch": 145.8095238095238,
      "grad_norm": 2.5624587535858154,
      "learning_rate": 2.8673163418290853e-07,
      "loss": 1.1578,
      "step": 1531
    },
    {
      "epoch": 145.9047619047619,
      "grad_norm": 4.983476638793945,
      "learning_rate": 2.8691904047976014e-07,
      "loss": 1.1557,
      "step": 1532
    },
    {
      "epoch": 146.0,
      "grad_norm": 4.180739402770996,
      "learning_rate": 2.8710644677661175e-07,
      "loss": 1.1512,
      "step": 1533
    },
    {
      "epoch": 146.0,
      "eval_accuracy": 0.7137504683402023,
      "eval_f1": 0.7744982290436836,
      "eval_loss": 0.5804224610328674,
      "eval_precision": 0.7114967462039046,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7510523891767416,
      "eval_runtime": 1.1812,
      "eval_samples_per_second": 2259.473,
      "eval_steps_per_second": 5.079,
      "step": 1533
    },
    {
      "epoch": 146.0952380952381,
      "grad_norm": 3.5563807487487793,
      "learning_rate": 2.872938530734633e-07,
      "loss": 1.1534,
      "step": 1534
    },
    {
      "epoch": 146.1904761904762,
      "grad_norm": 2.703789710998535,
      "learning_rate": 2.8748125937031486e-07,
      "loss": 1.1653,
      "step": 1535
    },
    {
      "epoch": 146.28571428571428,
      "grad_norm": 7.2478928565979,
      "learning_rate": 2.8766866566716646e-07,
      "loss": 1.172,
      "step": 1536
    },
    {
      "epoch": 146.38095238095238,
      "grad_norm": 6.74290132522583,
      "learning_rate": 2.87856071964018e-07,
      "loss": 1.1539,
      "step": 1537
    },
    {
      "epoch": 146.47619047619048,
      "grad_norm": 4.319555759429932,
      "learning_rate": 2.8804347826086957e-07,
      "loss": 1.1777,
      "step": 1538
    },
    {
      "epoch": 146.57142857142858,
      "grad_norm": 3.823532819747925,
      "learning_rate": 2.882308845577211e-07,
      "loss": 1.1957,
      "step": 1539
    },
    {
      "epoch": 146.66666666666666,
      "grad_norm": 5.92212438583374,
      "learning_rate": 2.8841829085457273e-07,
      "loss": 1.2175,
      "step": 1540
    },
    {
      "epoch": 146.76190476190476,
      "grad_norm": 2.487151622772217,
      "learning_rate": 2.886056971514243e-07,
      "loss": 1.1577,
      "step": 1541
    },
    {
      "epoch": 146.85714285714286,
      "grad_norm": 4.248367786407471,
      "learning_rate": 2.8879310344827584e-07,
      "loss": 1.153,
      "step": 1542
    },
    {
      "epoch": 146.95238095238096,
      "grad_norm": 5.908459663391113,
      "learning_rate": 2.8898050974512745e-07,
      "loss": 1.1667,
      "step": 1543
    },
    {
      "epoch": 146.95238095238096,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.7721481481481481,
      "eval_loss": 0.5804580450057983,
      "eval_precision": 0.7116329874385582,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7509289004029935,
      "eval_runtime": 1.3128,
      "eval_samples_per_second": 2033.01,
      "eval_steps_per_second": 4.57,
      "step": 1543
    },
    {
      "epoch": 147.04761904761904,
      "grad_norm": 3.983474016189575,
      "learning_rate": 2.89167916041979e-07,
      "loss": 1.1745,
      "step": 1544
    },
    {
      "epoch": 147.14285714285714,
      "grad_norm": 5.25045108795166,
      "learning_rate": 2.893553223388306e-07,
      "loss": 1.1795,
      "step": 1545
    },
    {
      "epoch": 147.23809523809524,
      "grad_norm": 6.433871269226074,
      "learning_rate": 2.8954272863568217e-07,
      "loss": 1.1582,
      "step": 1546
    },
    {
      "epoch": 147.33333333333334,
      "grad_norm": 2.469536542892456,
      "learning_rate": 2.897301349325338e-07,
      "loss": 1.1536,
      "step": 1547
    },
    {
      "epoch": 147.42857142857142,
      "grad_norm": 7.244877338409424,
      "learning_rate": 2.8991754122938533e-07,
      "loss": 1.1583,
      "step": 1548
    },
    {
      "epoch": 147.52380952380952,
      "grad_norm": 5.108227729797363,
      "learning_rate": 2.901049475262369e-07,
      "loss": 1.1873,
      "step": 1549
    },
    {
      "epoch": 147.61904761904762,
      "grad_norm": 4.663844585418701,
      "learning_rate": 2.902923538230885e-07,
      "loss": 1.1375,
      "step": 1550
    },
    {
      "epoch": 147.71428571428572,
      "grad_norm": 9.764266014099121,
      "learning_rate": 2.9047976011994004e-07,
      "loss": 1.1745,
      "step": 1551
    },
    {
      "epoch": 147.8095238095238,
      "grad_norm": 2.4404590129852295,
      "learning_rate": 2.906671664167916e-07,
      "loss": 1.1535,
      "step": 1552
    },
    {
      "epoch": 147.9047619047619,
      "grad_norm": 4.609780788421631,
      "learning_rate": 2.9085457271364315e-07,
      "loss": 1.1226,
      "step": 1553
    },
    {
      "epoch": 148.0,
      "grad_norm": 7.716403961181641,
      "learning_rate": 2.9104197901049476e-07,
      "loss": 1.275,
      "step": 1554
    },
    {
      "epoch": 148.0,
      "eval_accuracy": 0.7126264518546273,
      "eval_f1": 0.7746106376726418,
      "eval_loss": 0.5806289315223694,
      "eval_precision": 0.7089833243679398,
      "eval_recall": 0.8536269430051814,
      "eval_roc_auc": 0.7512006332757629,
      "eval_runtime": 1.139,
      "eval_samples_per_second": 2343.255,
      "eval_steps_per_second": 5.268,
      "step": 1554
    },
    {
      "epoch": 148.0952380952381,
      "grad_norm": 3.309723138809204,
      "learning_rate": 2.912293853073463e-07,
      "loss": 1.135,
      "step": 1555
    },
    {
      "epoch": 148.1904761904762,
      "grad_norm": 4.461905479431152,
      "learning_rate": 2.914167916041979e-07,
      "loss": 1.1913,
      "step": 1556
    },
    {
      "epoch": 148.28571428571428,
      "grad_norm": 4.881902694702148,
      "learning_rate": 2.9160419790104953e-07,
      "loss": 1.2228,
      "step": 1557
    },
    {
      "epoch": 148.38095238095238,
      "grad_norm": 4.563226699829102,
      "learning_rate": 2.917916041979011e-07,
      "loss": 1.1685,
      "step": 1558
    },
    {
      "epoch": 148.47619047619048,
      "grad_norm": 2.4159016609191895,
      "learning_rate": 2.9197901049475264e-07,
      "loss": 1.1282,
      "step": 1559
    },
    {
      "epoch": 148.57142857142858,
      "grad_norm": 2.5408880710601807,
      "learning_rate": 2.921664167916042e-07,
      "loss": 1.1528,
      "step": 1560
    },
    {
      "epoch": 148.66666666666666,
      "grad_norm": 2.2992091178894043,
      "learning_rate": 2.923538230884558e-07,
      "loss": 1.1591,
      "step": 1561
    },
    {
      "epoch": 148.76190476190476,
      "grad_norm": 3.7680423259735107,
      "learning_rate": 2.9254122938530736e-07,
      "loss": 1.1838,
      "step": 1562
    },
    {
      "epoch": 148.85714285714286,
      "grad_norm": 3.443899631500244,
      "learning_rate": 2.927286356821589e-07,
      "loss": 1.1458,
      "step": 1563
    },
    {
      "epoch": 148.95238095238096,
      "grad_norm": 3.9514212608337402,
      "learning_rate": 2.929160419790105e-07,
      "loss": 1.1756,
      "step": 1564
    },
    {
      "epoch": 148.95238095238096,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7718676122931442,
      "eval_loss": 0.5807510018348694,
      "eval_precision": 0.7097826086956521,
      "eval_recall": 0.8458549222797928,
      "eval_roc_auc": 0.7508871617731722,
      "eval_runtime": 1.1401,
      "eval_samples_per_second": 2341.037,
      "eval_steps_per_second": 5.263,
      "step": 1564
    },
    {
      "epoch": 149.04761904761904,
      "grad_norm": 3.019251823425293,
      "learning_rate": 2.9310344827586207e-07,
      "loss": 1.1705,
      "step": 1565
    },
    {
      "epoch": 149.14285714285714,
      "grad_norm": 6.118655204772949,
      "learning_rate": 2.9329085457271363e-07,
      "loss": 1.1902,
      "step": 1566
    },
    {
      "epoch": 149.23809523809524,
      "grad_norm": 5.084479331970215,
      "learning_rate": 2.9347826086956523e-07,
      "loss": 1.1595,
      "step": 1567
    },
    {
      "epoch": 149.33333333333334,
      "grad_norm": 3.2713265419006348,
      "learning_rate": 2.9366566716641684e-07,
      "loss": 1.1787,
      "step": 1568
    },
    {
      "epoch": 149.42857142857142,
      "grad_norm": 4.392632961273193,
      "learning_rate": 2.938530734632684e-07,
      "loss": 1.1612,
      "step": 1569
    },
    {
      "epoch": 149.52380952380952,
      "grad_norm": 5.749429225921631,
      "learning_rate": 2.9404047976011995e-07,
      "loss": 1.1431,
      "step": 1570
    },
    {
      "epoch": 149.61904761904762,
      "grad_norm": 3.3047690391540527,
      "learning_rate": 2.9422788605697156e-07,
      "loss": 1.1407,
      "step": 1571
    },
    {
      "epoch": 149.71428571428572,
      "grad_norm": 4.179772853851318,
      "learning_rate": 2.944152923538231e-07,
      "loss": 1.1356,
      "step": 1572
    },
    {
      "epoch": 149.8095238095238,
      "grad_norm": 6.8627448081970215,
      "learning_rate": 2.9460269865067467e-07,
      "loss": 1.2415,
      "step": 1573
    },
    {
      "epoch": 149.9047619047619,
      "grad_norm": 2.2968685626983643,
      "learning_rate": 2.947901049475262e-07,
      "loss": 1.1914,
      "step": 1574
    },
    {
      "epoch": 150.0,
      "grad_norm": 6.251117706298828,
      "learning_rate": 2.9497751124437783e-07,
      "loss": 1.1623,
      "step": 1575
    },
    {
      "epoch": 150.0,
      "eval_accuracy": 0.7133757961783439,
      "eval_f1": 0.7761194029850746,
      "eval_loss": 0.580627977848053,
      "eval_precision": 0.7079551521623064,
      "eval_recall": 0.8588082901554405,
      "eval_roc_auc": 0.7516729994242948,
      "eval_runtime": 1.1794,
      "eval_samples_per_second": 2262.925,
      "eval_steps_per_second": 5.087,
      "step": 1575
    },
    {
      "epoch": 150.0952380952381,
      "grad_norm": 4.730234146118164,
      "learning_rate": 2.951649175412294e-07,
      "loss": 1.1716,
      "step": 1576
    },
    {
      "epoch": 150.1904761904762,
      "grad_norm": 3.1547439098358154,
      "learning_rate": 2.9535232383808094e-07,
      "loss": 1.1722,
      "step": 1577
    },
    {
      "epoch": 150.28571428571428,
      "grad_norm": 5.73432731628418,
      "learning_rate": 2.9553973013493255e-07,
      "loss": 1.1863,
      "step": 1578
    },
    {
      "epoch": 150.38095238095238,
      "grad_norm": 3.081118583679199,
      "learning_rate": 2.957271364317841e-07,
      "loss": 1.1773,
      "step": 1579
    },
    {
      "epoch": 150.47619047619048,
      "grad_norm": 6.794926643371582,
      "learning_rate": 2.959145427286357e-07,
      "loss": 1.1139,
      "step": 1580
    },
    {
      "epoch": 150.57142857142858,
      "grad_norm": 2.378877639770508,
      "learning_rate": 2.9610194902548726e-07,
      "loss": 1.1698,
      "step": 1581
    },
    {
      "epoch": 150.66666666666666,
      "grad_norm": 6.612598419189453,
      "learning_rate": 2.9628935532233887e-07,
      "loss": 1.2078,
      "step": 1582
    },
    {
      "epoch": 150.76190476190476,
      "grad_norm": 3.649820327758789,
      "learning_rate": 2.964767616191904e-07,
      "loss": 1.1628,
      "step": 1583
    },
    {
      "epoch": 150.85714285714286,
      "grad_norm": 9.336405754089355,
      "learning_rate": 2.96664167916042e-07,
      "loss": 1.1628,
      "step": 1584
    },
    {
      "epoch": 150.95238095238096,
      "grad_norm": 2.9944679737091064,
      "learning_rate": 2.968515742128936e-07,
      "loss": 1.1524,
      "step": 1585
    },
    {
      "epoch": 150.95238095238096,
      "eval_accuracy": 0.7122517796927689,
      "eval_f1": 0.7735849056603774,
      "eval_loss": 0.5800833106040955,
      "eval_precision": 0.70995670995671,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.75201065054692,
      "eval_runtime": 1.193,
      "eval_samples_per_second": 2237.236,
      "eval_steps_per_second": 5.029,
      "step": 1585
    },
    {
      "epoch": 151.04761904761904,
      "grad_norm": 3.0176329612731934,
      "learning_rate": 2.9703898050974514e-07,
      "loss": 1.1367,
      "step": 1586
    },
    {
      "epoch": 151.14285714285714,
      "grad_norm": 3.8548378944396973,
      "learning_rate": 2.972263868065967e-07,
      "loss": 1.1625,
      "step": 1587
    },
    {
      "epoch": 151.23809523809524,
      "grad_norm": 3.6817703247070312,
      "learning_rate": 2.9741379310344825e-07,
      "loss": 1.1805,
      "step": 1588
    },
    {
      "epoch": 151.33333333333334,
      "grad_norm": 2.0584304332733154,
      "learning_rate": 2.9760119940029986e-07,
      "loss": 1.165,
      "step": 1589
    },
    {
      "epoch": 151.42857142857142,
      "grad_norm": 3.7080583572387695,
      "learning_rate": 2.977886056971514e-07,
      "loss": 1.1501,
      "step": 1590
    },
    {
      "epoch": 151.52380952380952,
      "grad_norm": 4.056233882904053,
      "learning_rate": 2.97976011994003e-07,
      "loss": 1.1399,
      "step": 1591
    },
    {
      "epoch": 151.61904761904762,
      "grad_norm": 3.2380008697509766,
      "learning_rate": 2.981634182908546e-07,
      "loss": 1.1614,
      "step": 1592
    },
    {
      "epoch": 151.71428571428572,
      "grad_norm": 5.078243732452393,
      "learning_rate": 2.983508245877062e-07,
      "loss": 1.1707,
      "step": 1593
    },
    {
      "epoch": 151.8095238095238,
      "grad_norm": 5.057522296905518,
      "learning_rate": 2.9853823088455773e-07,
      "loss": 1.1609,
      "step": 1594
    },
    {
      "epoch": 151.9047619047619,
      "grad_norm": 6.1785078048706055,
      "learning_rate": 2.9872563718140934e-07,
      "loss": 1.1885,
      "step": 1595
    },
    {
      "epoch": 152.0,
      "grad_norm": 4.74155330657959,
      "learning_rate": 2.989130434782609e-07,
      "loss": 1.1702,
      "step": 1596
    },
    {
      "epoch": 152.0,
      "eval_accuracy": 0.7126264518546273,
      "eval_f1": 0.7739463601532567,
      "eval_loss": 0.5797269344329834,
      "eval_precision": 0.7101135749053542,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7524715025906736,
      "eval_runtime": 1.1764,
      "eval_samples_per_second": 2268.706,
      "eval_steps_per_second": 5.1,
      "step": 1596
    },
    {
      "epoch": 152.0952380952381,
      "grad_norm": 3.2207934856414795,
      "learning_rate": 2.9910044977511245e-07,
      "loss": 1.1762,
      "step": 1597
    },
    {
      "epoch": 152.1904761904762,
      "grad_norm": 2.9277608394622803,
      "learning_rate": 2.99287856071964e-07,
      "loss": 1.1555,
      "step": 1598
    },
    {
      "epoch": 152.28571428571428,
      "grad_norm": 6.294363021850586,
      "learning_rate": 2.994752623688156e-07,
      "loss": 1.2192,
      "step": 1599
    },
    {
      "epoch": 152.38095238095238,
      "grad_norm": 2.1523854732513428,
      "learning_rate": 2.9966266866566717e-07,
      "loss": 1.1764,
      "step": 1600
    },
    {
      "epoch": 152.47619047619048,
      "grad_norm": 3.6600348949432373,
      "learning_rate": 2.998500749625187e-07,
      "loss": 1.166,
      "step": 1601
    },
    {
      "epoch": 152.57142857142858,
      "grad_norm": 4.593667507171631,
      "learning_rate": 3.0003748125937033e-07,
      "loss": 1.2047,
      "step": 1602
    },
    {
      "epoch": 152.66666666666666,
      "grad_norm": 5.261059761047363,
      "learning_rate": 3.0022488755622194e-07,
      "loss": 1.1187,
      "step": 1603
    },
    {
      "epoch": 152.76190476190476,
      "grad_norm": 4.2113213539123535,
      "learning_rate": 3.004122938530735e-07,
      "loss": 1.1643,
      "step": 1604
    },
    {
      "epoch": 152.85714285714286,
      "grad_norm": 4.283077239990234,
      "learning_rate": 3.0059970014992505e-07,
      "loss": 1.1297,
      "step": 1605
    },
    {
      "epoch": 152.95238095238096,
      "grad_norm": 4.8893961906433105,
      "learning_rate": 3.0078710644677665e-07,
      "loss": 1.1703,
      "step": 1606
    },
    {
      "epoch": 152.95238095238096,
      "eval_accuracy": 0.7126264518546273,
      "eval_f1": 0.7738130345030965,
      "eval_loss": 0.5795263648033142,
      "eval_precision": 0.7103410936654033,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7529726540011514,
      "eval_runtime": 1.1454,
      "eval_samples_per_second": 2330.141,
      "eval_steps_per_second": 5.238,
      "step": 1606
    },
    {
      "epoch": 153.04761904761904,
      "grad_norm": 4.8148722648620605,
      "learning_rate": 3.009745127436282e-07,
      "loss": 1.1685,
      "step": 1607
    },
    {
      "epoch": 153.14285714285714,
      "grad_norm": 3.824242115020752,
      "learning_rate": 3.0116191904047976e-07,
      "loss": 1.155,
      "step": 1608
    },
    {
      "epoch": 153.23809523809524,
      "grad_norm": 3.3369882106781006,
      "learning_rate": 3.0134932533733137e-07,
      "loss": 1.1803,
      "step": 1609
    },
    {
      "epoch": 153.33333333333334,
      "grad_norm": 5.317166805267334,
      "learning_rate": 3.015367316341829e-07,
      "loss": 1.1508,
      "step": 1610
    },
    {
      "epoch": 153.42857142857142,
      "grad_norm": 2.601602077484131,
      "learning_rate": 3.017241379310345e-07,
      "loss": 1.1432,
      "step": 1611
    },
    {
      "epoch": 153.52380952380952,
      "grad_norm": 4.989608287811279,
      "learning_rate": 3.0191154422788603e-07,
      "loss": 1.165,
      "step": 1612
    },
    {
      "epoch": 153.61904761904762,
      "grad_norm": 3.7194533348083496,
      "learning_rate": 3.0209895052473764e-07,
      "loss": 1.1652,
      "step": 1613
    },
    {
      "epoch": 153.71428571428572,
      "grad_norm": 2.8490591049194336,
      "learning_rate": 3.0228635682158925e-07,
      "loss": 1.1779,
      "step": 1614
    },
    {
      "epoch": 153.8095238095238,
      "grad_norm": 3.7863845825195312,
      "learning_rate": 3.024737631184408e-07,
      "loss": 1.1378,
      "step": 1615
    },
    {
      "epoch": 153.9047619047619,
      "grad_norm": 3.58205246925354,
      "learning_rate": 3.026611694152924e-07,
      "loss": 1.1786,
      "step": 1616
    },
    {
      "epoch": 154.0,
      "grad_norm": 2.69224214553833,
      "learning_rate": 3.0284857571214396e-07,
      "loss": 1.1933,
      "step": 1617
    },
    {
      "epoch": 154.0,
      "eval_accuracy": 0.7133757961783439,
      "eval_f1": 0.7741364038972542,
      "eval_loss": 0.5793300867080688,
      "eval_precision": 0.711340206185567,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7533151986183074,
      "eval_runtime": 1.2676,
      "eval_samples_per_second": 2105.495,
      "eval_steps_per_second": 4.733,
      "step": 1617
    },
    {
      "epoch": 154.0952380952381,
      "grad_norm": 8.547765731811523,
      "learning_rate": 3.030359820089955e-07,
      "loss": 1.1543,
      "step": 1618
    },
    {
      "epoch": 154.1904761904762,
      "grad_norm": 4.940852642059326,
      "learning_rate": 3.0322338830584707e-07,
      "loss": 1.1645,
      "step": 1619
    },
    {
      "epoch": 154.28571428571428,
      "grad_norm": 2.6271817684173584,
      "learning_rate": 3.034107946026987e-07,
      "loss": 1.1482,
      "step": 1620
    },
    {
      "epoch": 154.38095238095238,
      "grad_norm": 3.078421115875244,
      "learning_rate": 3.0359820089955023e-07,
      "loss": 1.1222,
      "step": 1621
    },
    {
      "epoch": 154.47619047619048,
      "grad_norm": 6.029754638671875,
      "learning_rate": 3.037856071964018e-07,
      "loss": 1.2054,
      "step": 1622
    },
    {
      "epoch": 154.57142857142858,
      "grad_norm": 4.2551188468933105,
      "learning_rate": 3.039730134932534e-07,
      "loss": 1.1916,
      "step": 1623
    },
    {
      "epoch": 154.66666666666666,
      "grad_norm": 3.1497421264648438,
      "learning_rate": 3.0416041979010495e-07,
      "loss": 1.1322,
      "step": 1624
    },
    {
      "epoch": 154.76190476190476,
      "grad_norm": 3.4132707118988037,
      "learning_rate": 3.043478260869565e-07,
      "loss": 1.1689,
      "step": 1625
    },
    {
      "epoch": 154.85714285714286,
      "grad_norm": 3.04034161567688,
      "learning_rate": 3.045352323838081e-07,
      "loss": 1.1495,
      "step": 1626
    },
    {
      "epoch": 154.95238095238096,
      "grad_norm": 3.6474931240081787,
      "learning_rate": 3.047226386806597e-07,
      "loss": 1.1827,
      "step": 1627
    },
    {
      "epoch": 154.95238095238096,
      "eval_accuracy": 0.7141251405020607,
      "eval_f1": 0.7753900500441566,
      "eval_loss": 0.5792854428291321,
      "eval_precision": 0.710739341608203,
      "eval_recall": 0.852979274611399,
      "eval_roc_auc": 0.7535290731145654,
      "eval_runtime": 1.1995,
      "eval_samples_per_second": 2225.127,
      "eval_steps_per_second": 5.002,
      "step": 1627
    },
    {
      "epoch": 155.04761904761904,
      "grad_norm": 5.743706703186035,
      "learning_rate": 3.049100449775113e-07,
      "loss": 1.1394,
      "step": 1628
    },
    {
      "epoch": 155.14285714285714,
      "grad_norm": 2.847236394882202,
      "learning_rate": 3.0509745127436283e-07,
      "loss": 1.1703,
      "step": 1629
    },
    {
      "epoch": 155.23809523809524,
      "grad_norm": 3.0275094509124756,
      "learning_rate": 3.0528485757121444e-07,
      "loss": 1.196,
      "step": 1630
    },
    {
      "epoch": 155.33333333333334,
      "grad_norm": 4.455320358276367,
      "learning_rate": 3.05472263868066e-07,
      "loss": 1.1521,
      "step": 1631
    },
    {
      "epoch": 155.42857142857142,
      "grad_norm": 4.467602252960205,
      "learning_rate": 3.0565967016491755e-07,
      "loss": 1.1466,
      "step": 1632
    },
    {
      "epoch": 155.52380952380952,
      "grad_norm": 4.637279510498047,
      "learning_rate": 3.058470764617691e-07,
      "loss": 1.1421,
      "step": 1633
    },
    {
      "epoch": 155.61904761904762,
      "grad_norm": 7.024368762969971,
      "learning_rate": 3.060344827586207e-07,
      "loss": 1.1524,
      "step": 1634
    },
    {
      "epoch": 155.71428571428572,
      "grad_norm": 2.950413942337036,
      "learning_rate": 3.0622188905547226e-07,
      "loss": 1.1879,
      "step": 1635
    },
    {
      "epoch": 155.8095238095238,
      "grad_norm": 3.2424774169921875,
      "learning_rate": 3.064092953523238e-07,
      "loss": 1.1723,
      "step": 1636
    },
    {
      "epoch": 155.9047619047619,
      "grad_norm": 3.6627702713012695,
      "learning_rate": 3.065967016491754e-07,
      "loss": 1.1706,
      "step": 1637
    },
    {
      "epoch": 156.0,
      "grad_norm": 4.732444763183594,
      "learning_rate": 3.0678410794602703e-07,
      "loss": 1.1414,
      "step": 1638
    },
    {
      "epoch": 156.0,
      "eval_accuracy": 0.7141251405020607,
      "eval_f1": 0.7744605379840378,
      "eval_loss": 0.5791009664535522,
      "eval_precision": 0.7123436650353453,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7538658606793321,
      "eval_runtime": 1.2898,
      "eval_samples_per_second": 2069.323,
      "eval_steps_per_second": 4.652,
      "step": 1638
    },
    {
      "epoch": 156.0952380952381,
      "grad_norm": 8.923810958862305,
      "learning_rate": 3.069715142428786e-07,
      "loss": 1.1463,
      "step": 1639
    },
    {
      "epoch": 156.1904761904762,
      "grad_norm": 4.258830547332764,
      "learning_rate": 3.0715892053973014e-07,
      "loss": 1.1602,
      "step": 1640
    },
    {
      "epoch": 156.28571428571428,
      "grad_norm": 5.375497341156006,
      "learning_rate": 3.0734632683658175e-07,
      "loss": 1.1538,
      "step": 1641
    },
    {
      "epoch": 156.38095238095238,
      "grad_norm": 2.5572433471679688,
      "learning_rate": 3.075337331334333e-07,
      "loss": 1.1395,
      "step": 1642
    },
    {
      "epoch": 156.47619047619048,
      "grad_norm": 4.052694797515869,
      "learning_rate": 3.0772113943028486e-07,
      "loss": 1.2135,
      "step": 1643
    },
    {
      "epoch": 156.57142857142858,
      "grad_norm": 5.803094387054443,
      "learning_rate": 3.0790854572713646e-07,
      "loss": 1.2036,
      "step": 1644
    },
    {
      "epoch": 156.66666666666666,
      "grad_norm": 4.393676280975342,
      "learning_rate": 3.08095952023988e-07,
      "loss": 1.1575,
      "step": 1645
    },
    {
      "epoch": 156.76190476190476,
      "grad_norm": 2.8207361698150635,
      "learning_rate": 3.0828335832083957e-07,
      "loss": 1.172,
      "step": 1646
    },
    {
      "epoch": 156.85714285714286,
      "grad_norm": 3.8662214279174805,
      "learning_rate": 3.0847076461769113e-07,
      "loss": 1.1986,
      "step": 1647
    },
    {
      "epoch": 156.95238095238096,
      "grad_norm": 2.2252655029296875,
      "learning_rate": 3.0865817091454274e-07,
      "loss": 1.1686,
      "step": 1648
    },
    {
      "epoch": 156.95238095238096,
      "eval_accuracy": 0.7122517796927689,
      "eval_f1": 0.7735849056603774,
      "eval_loss": 0.5791285634040833,
      "eval_precision": 0.70995670995671,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7541652274035694,
      "eval_runtime": 1.132,
      "eval_samples_per_second": 2357.715,
      "eval_steps_per_second": 5.3,
      "step": 1648
    },
    {
      "epoch": 157.04761904761904,
      "grad_norm": 5.072911262512207,
      "learning_rate": 3.0884557721139434e-07,
      "loss": 1.1583,
      "step": 1649
    },
    {
      "epoch": 157.14285714285714,
      "grad_norm": 3.5795886516571045,
      "learning_rate": 3.090329835082459e-07,
      "loss": 1.1602,
      "step": 1650
    },
    {
      "epoch": 157.23809523809524,
      "grad_norm": 2.649064064025879,
      "learning_rate": 3.092203898050975e-07,
      "loss": 1.1591,
      "step": 1651
    },
    {
      "epoch": 157.33333333333334,
      "grad_norm": 2.5507094860076904,
      "learning_rate": 3.0940779610194906e-07,
      "loss": 1.1871,
      "step": 1652
    },
    {
      "epoch": 157.42857142857142,
      "grad_norm": 4.117077827453613,
      "learning_rate": 3.095952023988006e-07,
      "loss": 1.1838,
      "step": 1653
    },
    {
      "epoch": 157.52380952380952,
      "grad_norm": 3.221205949783325,
      "learning_rate": 3.0978260869565217e-07,
      "loss": 1.1684,
      "step": 1654
    },
    {
      "epoch": 157.61904761904762,
      "grad_norm": 8.151176452636719,
      "learning_rate": 3.099700149925038e-07,
      "loss": 1.1948,
      "step": 1655
    },
    {
      "epoch": 157.71428571428572,
      "grad_norm": 3.3436615467071533,
      "learning_rate": 3.1015742128935533e-07,
      "loss": 1.182,
      "step": 1656
    },
    {
      "epoch": 157.8095238095238,
      "grad_norm": 5.598202228546143,
      "learning_rate": 3.103448275862069e-07,
      "loss": 1.1113,
      "step": 1657
    },
    {
      "epoch": 157.9047619047619,
      "grad_norm": 3.317392587661743,
      "learning_rate": 3.105322338830585e-07,
      "loss": 1.132,
      "step": 1658
    },
    {
      "epoch": 158.0,
      "grad_norm": 3.3303115367889404,
      "learning_rate": 3.1071964017991005e-07,
      "loss": 1.2094,
      "step": 1659
    },
    {
      "epoch": 158.0,
      "eval_accuracy": 0.7130011240164855,
      "eval_f1": 0.7735068007096393,
      "eval_loss": 0.5792306661605835,
      "eval_precision": 0.7116430903155604,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7539985607369027,
      "eval_runtime": 1.2239,
      "eval_samples_per_second": 2180.666,
      "eval_steps_per_second": 4.902,
      "step": 1659
    },
    {
      "epoch": 158.0952380952381,
      "grad_norm": 3.01957631111145,
      "learning_rate": 3.109070464767616e-07,
      "loss": 1.1732,
      "step": 1660
    },
    {
      "epoch": 158.1904761904762,
      "grad_norm": 4.130393981933594,
      "learning_rate": 3.110944527736132e-07,
      "loss": 1.1752,
      "step": 1661
    },
    {
      "epoch": 158.28571428571428,
      "grad_norm": 3.2094545364379883,
      "learning_rate": 3.112818590704648e-07,
      "loss": 1.1735,
      "step": 1662
    },
    {
      "epoch": 158.38095238095238,
      "grad_norm": 6.156533718109131,
      "learning_rate": 3.1146926536731637e-07,
      "loss": 1.1488,
      "step": 1663
    },
    {
      "epoch": 158.47619047619048,
      "grad_norm": 14.892172813415527,
      "learning_rate": 3.116566716641679e-07,
      "loss": 1.2048,
      "step": 1664
    },
    {
      "epoch": 158.57142857142858,
      "grad_norm": 3.576772689819336,
      "learning_rate": 3.1184407796101953e-07,
      "loss": 1.1629,
      "step": 1665
    },
    {
      "epoch": 158.66666666666666,
      "grad_norm": 5.326858997344971,
      "learning_rate": 3.120314842578711e-07,
      "loss": 1.1625,
      "step": 1666
    },
    {
      "epoch": 158.76190476190476,
      "grad_norm": 7.5055084228515625,
      "learning_rate": 3.1221889055472264e-07,
      "loss": 1.113,
      "step": 1667
    },
    {
      "epoch": 158.85714285714286,
      "grad_norm": 5.809722423553467,
      "learning_rate": 3.124062968515742e-07,
      "loss": 1.1523,
      "step": 1668
    },
    {
      "epoch": 158.95238095238096,
      "grad_norm": 5.568952560424805,
      "learning_rate": 3.125937031484258e-07,
      "loss": 1.1823,
      "step": 1669
    },
    {
      "epoch": 158.95238095238096,
      "eval_accuracy": 0.7107530910453354,
      "eval_f1": 0.7726737338044759,
      "eval_loss": 0.5794424414634705,
      "eval_precision": 0.7084233261339092,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7538808290155441,
      "eval_runtime": 1.2192,
      "eval_samples_per_second": 2189.195,
      "eval_steps_per_second": 4.921,
      "step": 1669
    },
    {
      "epoch": 159.04761904761904,
      "grad_norm": 2.202739953994751,
      "learning_rate": 3.1278110944527736e-07,
      "loss": 1.1819,
      "step": 1670
    },
    {
      "epoch": 159.14285714285714,
      "grad_norm": 3.3776562213897705,
      "learning_rate": 3.129685157421289e-07,
      "loss": 1.1367,
      "step": 1671
    },
    {
      "epoch": 159.23809523809524,
      "grad_norm": 7.199225425720215,
      "learning_rate": 3.131559220389805e-07,
      "loss": 1.144,
      "step": 1672
    },
    {
      "epoch": 159.33333333333334,
      "grad_norm": 3.1565022468566895,
      "learning_rate": 3.1334332833583213e-07,
      "loss": 1.2118,
      "step": 1673
    },
    {
      "epoch": 159.42857142857142,
      "grad_norm": 12.221293449401855,
      "learning_rate": 3.135307346326837e-07,
      "loss": 1.1934,
      "step": 1674
    },
    {
      "epoch": 159.52380952380952,
      "grad_norm": 6.696208477020264,
      "learning_rate": 3.1371814092953524e-07,
      "loss": 1.1831,
      "step": 1675
    },
    {
      "epoch": 159.61904761904762,
      "grad_norm": 3.944971799850464,
      "learning_rate": 3.1390554722638684e-07,
      "loss": 1.1751,
      "step": 1676
    },
    {
      "epoch": 159.71428571428572,
      "grad_norm": 5.407956123352051,
      "learning_rate": 3.140929535232384e-07,
      "loss": 1.1373,
      "step": 1677
    },
    {
      "epoch": 159.8095238095238,
      "grad_norm": 3.2926764488220215,
      "learning_rate": 3.1428035982008995e-07,
      "loss": 1.1734,
      "step": 1678
    },
    {
      "epoch": 159.9047619047619,
      "grad_norm": 4.692826747894287,
      "learning_rate": 3.1446776611694156e-07,
      "loss": 1.1429,
      "step": 1679
    },
    {
      "epoch": 160.0,
      "grad_norm": 5.594501495361328,
      "learning_rate": 3.146551724137931e-07,
      "loss": 1.1453,
      "step": 1680
    },
    {
      "epoch": 160.0,
      "eval_accuracy": 0.7130011240164855,
      "eval_f1": 0.7735068007096393,
      "eval_loss": 0.5794059038162231,
      "eval_precision": 0.7116430903155604,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7536447898675878,
      "eval_runtime": 1.3207,
      "eval_samples_per_second": 2020.963,
      "eval_steps_per_second": 4.543,
      "step": 1680
    },
    {
      "epoch": 160.0952380952381,
      "grad_norm": 2.961744785308838,
      "learning_rate": 3.1484257871064467e-07,
      "loss": 1.1718,
      "step": 1681
    },
    {
      "epoch": 160.1904761904762,
      "grad_norm": 4.337584972381592,
      "learning_rate": 3.150299850074962e-07,
      "loss": 1.2078,
      "step": 1682
    },
    {
      "epoch": 160.28571428571428,
      "grad_norm": 3.1441948413848877,
      "learning_rate": 3.1521739130434783e-07,
      "loss": 1.1365,
      "step": 1683
    },
    {
      "epoch": 160.38095238095238,
      "grad_norm": 10.010917663574219,
      "learning_rate": 3.1540479760119944e-07,
      "loss": 1.1623,
      "step": 1684
    },
    {
      "epoch": 160.47619047619048,
      "grad_norm": 3.047410249710083,
      "learning_rate": 3.15592203898051e-07,
      "loss": 1.1479,
      "step": 1685
    },
    {
      "epoch": 160.57142857142858,
      "grad_norm": 2.7744319438934326,
      "learning_rate": 3.157796101949026e-07,
      "loss": 1.1684,
      "step": 1686
    },
    {
      "epoch": 160.66666666666666,
      "grad_norm": 2.880967617034912,
      "learning_rate": 3.1596701649175415e-07,
      "loss": 1.1694,
      "step": 1687
    },
    {
      "epoch": 160.76190476190476,
      "grad_norm": 3.9865102767944336,
      "learning_rate": 3.161544227886057e-07,
      "loss": 1.1591,
      "step": 1688
    },
    {
      "epoch": 160.85714285714286,
      "grad_norm": 9.290629386901855,
      "learning_rate": 3.1634182908545726e-07,
      "loss": 1.225,
      "step": 1689
    },
    {
      "epoch": 160.95238095238096,
      "grad_norm": 6.068443775177002,
      "learning_rate": 3.1652923538230887e-07,
      "loss": 1.1401,
      "step": 1690
    },
    {
      "epoch": 160.95238095238096,
      "eval_accuracy": 0.7126264518546273,
      "eval_f1": 0.7735459108355477,
      "eval_loss": 0.5790919661521912,
      "eval_precision": 0.7107976125881714,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7541301093839954,
      "eval_runtime": 1.1361,
      "eval_samples_per_second": 2349.34,
      "eval_steps_per_second": 5.281,
      "step": 1690
    },
    {
      "epoch": 161.04761904761904,
      "grad_norm": 3.733458995819092,
      "learning_rate": 3.167166416791604e-07,
      "loss": 1.1656,
      "step": 1691
    },
    {
      "epoch": 161.14285714285714,
      "grad_norm": 3.664858102798462,
      "learning_rate": 3.16904047976012e-07,
      "loss": 1.1442,
      "step": 1692
    },
    {
      "epoch": 161.23809523809524,
      "grad_norm": 2.717442274093628,
      "learning_rate": 3.170914542728636e-07,
      "loss": 1.1454,
      "step": 1693
    },
    {
      "epoch": 161.33333333333334,
      "grad_norm": 3.2815489768981934,
      "learning_rate": 3.1727886056971514e-07,
      "loss": 1.1479,
      "step": 1694
    },
    {
      "epoch": 161.42857142857142,
      "grad_norm": 2.4982147216796875,
      "learning_rate": 3.1746626686656675e-07,
      "loss": 1.1653,
      "step": 1695
    },
    {
      "epoch": 161.52380952380952,
      "grad_norm": 4.64665412902832,
      "learning_rate": 3.176536731634183e-07,
      "loss": 1.1571,
      "step": 1696
    },
    {
      "epoch": 161.61904761904762,
      "grad_norm": 3.416707992553711,
      "learning_rate": 3.178410794602699e-07,
      "loss": 1.1982,
      "step": 1697
    },
    {
      "epoch": 161.71428571428572,
      "grad_norm": 4.434908390045166,
      "learning_rate": 3.1802848575712147e-07,
      "loss": 1.1707,
      "step": 1698
    },
    {
      "epoch": 161.8095238095238,
      "grad_norm": 7.077025890350342,
      "learning_rate": 3.18215892053973e-07,
      "loss": 1.1793,
      "step": 1699
    },
    {
      "epoch": 161.9047619047619,
      "grad_norm": 3.968093156814575,
      "learning_rate": 3.1840329835082463e-07,
      "loss": 1.1527,
      "step": 1700
    },
    {
      "epoch": 162.0,
      "grad_norm": 2.842799663543701,
      "learning_rate": 3.185907046476762e-07,
      "loss": 1.1789,
      "step": 1701
    },
    {
      "epoch": 162.0,
      "eval_accuracy": 0.7133757961783439,
      "eval_f1": 0.7738693467336684,
      "eval_loss": 0.5786438584327698,
      "eval_precision": 0.711799891245242,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7546603339090385,
      "eval_runtime": 1.2166,
      "eval_samples_per_second": 2193.76,
      "eval_steps_per_second": 4.932,
      "step": 1701
    },
    {
      "epoch": 162.0952380952381,
      "grad_norm": 2.8696420192718506,
      "learning_rate": 3.1877811094452774e-07,
      "loss": 1.1815,
      "step": 1702
    },
    {
      "epoch": 162.1904761904762,
      "grad_norm": 5.638867378234863,
      "learning_rate": 3.1896551724137934e-07,
      "loss": 1.1527,
      "step": 1703
    },
    {
      "epoch": 162.28571428571428,
      "grad_norm": 3.586515426635742,
      "learning_rate": 3.191529235382309e-07,
      "loss": 1.1662,
      "step": 1704
    },
    {
      "epoch": 162.38095238095238,
      "grad_norm": 3.2161989212036133,
      "learning_rate": 3.1934032983508245e-07,
      "loss": 1.2009,
      "step": 1705
    },
    {
      "epoch": 162.47619047619048,
      "grad_norm": 5.910765647888184,
      "learning_rate": 3.19527736131934e-07,
      "loss": 1.1426,
      "step": 1706
    },
    {
      "epoch": 162.57142857142858,
      "grad_norm": 4.062796592712402,
      "learning_rate": 3.197151424287856e-07,
      "loss": 1.1538,
      "step": 1707
    },
    {
      "epoch": 162.66666666666666,
      "grad_norm": 3.743438243865967,
      "learning_rate": 3.199025487256372e-07,
      "loss": 1.1472,
      "step": 1708
    },
    {
      "epoch": 162.76190476190476,
      "grad_norm": 7.2488508224487305,
      "learning_rate": 3.200899550224888e-07,
      "loss": 1.1939,
      "step": 1709
    },
    {
      "epoch": 162.85714285714286,
      "grad_norm": 3.9524292945861816,
      "learning_rate": 3.202773613193404e-07,
      "loss": 1.1322,
      "step": 1710
    },
    {
      "epoch": 162.95238095238096,
      "grad_norm": 5.516472339630127,
      "learning_rate": 3.2046476761619194e-07,
      "loss": 1.1396,
      "step": 1711
    },
    {
      "epoch": 162.95238095238096,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.7721481481481481,
      "eval_loss": 0.5783283710479736,
      "eval_precision": 0.7116329874385582,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.755014968336212,
      "eval_runtime": 1.2621,
      "eval_samples_per_second": 2114.736,
      "eval_steps_per_second": 4.754,
      "step": 1711
    },
    {
      "epoch": 163.04761904761904,
      "grad_norm": 5.999850749969482,
      "learning_rate": 3.206521739130435e-07,
      "loss": 1.1333,
      "step": 1712
    },
    {
      "epoch": 163.14285714285714,
      "grad_norm": 5.138979911804199,
      "learning_rate": 3.2083958020989505e-07,
      "loss": 1.1356,
      "step": 1713
    },
    {
      "epoch": 163.23809523809524,
      "grad_norm": 2.94199275970459,
      "learning_rate": 3.2102698650674665e-07,
      "loss": 1.206,
      "step": 1714
    },
    {
      "epoch": 163.33333333333334,
      "grad_norm": 4.085175037384033,
      "learning_rate": 3.212143928035982e-07,
      "loss": 1.1372,
      "step": 1715
    },
    {
      "epoch": 163.42857142857142,
      "grad_norm": 2.2926993370056152,
      "learning_rate": 3.2140179910044976e-07,
      "loss": 1.1509,
      "step": 1716
    },
    {
      "epoch": 163.52380952380952,
      "grad_norm": 3.0990068912506104,
      "learning_rate": 3.2158920539730137e-07,
      "loss": 1.1416,
      "step": 1717
    },
    {
      "epoch": 163.61904761904762,
      "grad_norm": 5.597672462463379,
      "learning_rate": 3.217766116941529e-07,
      "loss": 1.1744,
      "step": 1718
    },
    {
      "epoch": 163.71428571428572,
      "grad_norm": 2.897651433944702,
      "learning_rate": 3.2196401799100453e-07,
      "loss": 1.158,
      "step": 1719
    },
    {
      "epoch": 163.8095238095238,
      "grad_norm": 6.378485202789307,
      "learning_rate": 3.221514242878561e-07,
      "loss": 1.1771,
      "step": 1720
    },
    {
      "epoch": 163.9047619047619,
      "grad_norm": 9.879870414733887,
      "learning_rate": 3.223388305847077e-07,
      "loss": 1.2047,
      "step": 1721
    },
    {
      "epoch": 164.0,
      "grad_norm": 5.101013660430908,
      "learning_rate": 3.2252623688155925e-07,
      "loss": 1.1571,
      "step": 1722
    },
    {
      "epoch": 164.0,
      "eval_accuracy": 0.7126264518546273,
      "eval_f1": 0.774345395704619,
      "eval_loss": 0.5781943202018738,
      "eval_precision": 0.7094339622641509,
      "eval_recall": 0.8523316062176166,
      "eval_roc_auc": 0.7555529648819804,
      "eval_runtime": 1.3698,
      "eval_samples_per_second": 1948.521,
      "eval_steps_per_second": 4.38,
      "step": 1722
    },
    {
      "epoch": 164.0952380952381,
      "grad_norm": 2.68703556060791,
      "learning_rate": 3.227136431784108e-07,
      "loss": 1.1588,
      "step": 1723
    },
    {
      "epoch": 164.1904761904762,
      "grad_norm": 5.951897621154785,
      "learning_rate": 3.229010494752624e-07,
      "loss": 1.1672,
      "step": 1724
    },
    {
      "epoch": 164.28571428571428,
      "grad_norm": 6.9368438720703125,
      "learning_rate": 3.2308845577211397e-07,
      "loss": 1.1562,
      "step": 1725
    },
    {
      "epoch": 164.38095238095238,
      "grad_norm": 4.164420127868652,
      "learning_rate": 3.232758620689655e-07,
      "loss": 1.1505,
      "step": 1726
    },
    {
      "epoch": 164.47619047619048,
      "grad_norm": 2.9614076614379883,
      "learning_rate": 3.234632683658171e-07,
      "loss": 1.1773,
      "step": 1727
    },
    {
      "epoch": 164.57142857142858,
      "grad_norm": 4.118006706237793,
      "learning_rate": 3.236506746626687e-07,
      "loss": 1.125,
      "step": 1728
    },
    {
      "epoch": 164.66666666666666,
      "grad_norm": 6.172893524169922,
      "learning_rate": 3.2383808095952024e-07,
      "loss": 1.1511,
      "step": 1729
    },
    {
      "epoch": 164.76190476190476,
      "grad_norm": 2.903033494949341,
      "learning_rate": 3.2402548725637184e-07,
      "loss": 1.1505,
      "step": 1730
    },
    {
      "epoch": 164.85714285714286,
      "grad_norm": 4.236114025115967,
      "learning_rate": 3.242128935532234e-07,
      "loss": 1.1901,
      "step": 1731
    },
    {
      "epoch": 164.95238095238096,
      "grad_norm": 5.413357734680176,
      "learning_rate": 3.24400299850075e-07,
      "loss": 1.1835,
      "step": 1732
    },
    {
      "epoch": 164.95238095238096,
      "eval_accuracy": 0.7130011240164855,
      "eval_f1": 0.7725653206650831,
      "eval_loss": 0.5783714056015015,
      "eval_precision": 0.7132675438596491,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7549363845710997,
      "eval_runtime": 1.1442,
      "eval_samples_per_second": 2332.686,
      "eval_steps_per_second": 5.244,
      "step": 1732
    },
    {
      "epoch": 165.04761904761904,
      "grad_norm": 5.139907360076904,
      "learning_rate": 3.2458770614692656e-07,
      "loss": 1.1825,
      "step": 1733
    },
    {
      "epoch": 165.14285714285714,
      "grad_norm": 7.035548686981201,
      "learning_rate": 3.247751124437781e-07,
      "loss": 1.1158,
      "step": 1734
    },
    {
      "epoch": 165.23809523809524,
      "grad_norm": 2.559943199157715,
      "learning_rate": 3.249625187406297e-07,
      "loss": 1.165,
      "step": 1735
    },
    {
      "epoch": 165.33333333333334,
      "grad_norm": 4.389877796173096,
      "learning_rate": 3.251499250374813e-07,
      "loss": 1.1688,
      "step": 1736
    },
    {
      "epoch": 165.42857142857142,
      "grad_norm": 3.601215362548828,
      "learning_rate": 3.2533733133433283e-07,
      "loss": 1.2054,
      "step": 1737
    },
    {
      "epoch": 165.52380952380952,
      "grad_norm": 3.479674816131592,
      "learning_rate": 3.2552473763118444e-07,
      "loss": 1.1752,
      "step": 1738
    },
    {
      "epoch": 165.61904761904762,
      "grad_norm": 4.253512382507324,
      "learning_rate": 3.25712143928036e-07,
      "loss": 1.1698,
      "step": 1739
    },
    {
      "epoch": 165.71428571428572,
      "grad_norm": 6.709036350250244,
      "learning_rate": 3.2589955022488755e-07,
      "loss": 1.1617,
      "step": 1740
    },
    {
      "epoch": 165.8095238095238,
      "grad_norm": 4.812006950378418,
      "learning_rate": 3.260869565217391e-07,
      "loss": 1.2067,
      "step": 1741
    },
    {
      "epoch": 165.9047619047619,
      "grad_norm": 3.5801570415496826,
      "learning_rate": 3.262743628185907e-07,
      "loss": 1.1455,
      "step": 1742
    },
    {
      "epoch": 166.0,
      "grad_norm": 5.324729919433594,
      "learning_rate": 3.264617691154423e-07,
      "loss": 1.1339,
      "step": 1743
    },
    {
      "epoch": 166.0,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.7732232379828959,
      "eval_loss": 0.5782414674758911,
      "eval_precision": 0.7097996751488901,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7554231433506045,
      "eval_runtime": 1.2109,
      "eval_samples_per_second": 2204.106,
      "eval_steps_per_second": 4.955,
      "step": 1743
    },
    {
      "epoch": 166.0952380952381,
      "grad_norm": 6.025104522705078,
      "learning_rate": 3.2664917541229387e-07,
      "loss": 1.1132,
      "step": 1744
    },
    {
      "epoch": 166.1904761904762,
      "grad_norm": 4.197454929351807,
      "learning_rate": 3.268365817091455e-07,
      "loss": 1.2025,
      "step": 1745
    },
    {
      "epoch": 166.28571428571428,
      "grad_norm": 2.0884721279144287,
      "learning_rate": 3.2702398800599703e-07,
      "loss": 1.1522,
      "step": 1746
    },
    {
      "epoch": 166.38095238095238,
      "grad_norm": 5.595632553100586,
      "learning_rate": 3.272113943028486e-07,
      "loss": 1.1931,
      "step": 1747
    },
    {
      "epoch": 166.47619047619048,
      "grad_norm": 2.5613150596618652,
      "learning_rate": 3.2739880059970014e-07,
      "loss": 1.1487,
      "step": 1748
    },
    {
      "epoch": 166.57142857142858,
      "grad_norm": 4.494848728179932,
      "learning_rate": 3.2758620689655175e-07,
      "loss": 1.1442,
      "step": 1749
    },
    {
      "epoch": 166.66666666666666,
      "grad_norm": 4.147763729095459,
      "learning_rate": 3.277736131934033e-07,
      "loss": 1.1473,
      "step": 1750
    },
    {
      "epoch": 166.76190476190476,
      "grad_norm": 4.883570194244385,
      "learning_rate": 3.2796101949025486e-07,
      "loss": 1.1481,
      "step": 1751
    },
    {
      "epoch": 166.85714285714286,
      "grad_norm": 3.538252353668213,
      "learning_rate": 3.2814842578710647e-07,
      "loss": 1.1876,
      "step": 1752
    },
    {
      "epoch": 166.95238095238096,
      "grad_norm": 3.5304596424102783,
      "learning_rate": 3.28335832083958e-07,
      "loss": 1.1672,
      "step": 1753
    },
    {
      "epoch": 166.95238095238096,
      "eval_accuracy": 0.7122517796927689,
      "eval_f1": 0.7730496453900709,
      "eval_loss": 0.5780826807022095,
      "eval_precision": 0.7108695652173913,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7556171560161197,
      "eval_runtime": 1.23,
      "eval_samples_per_second": 2169.958,
      "eval_steps_per_second": 4.878,
      "step": 1753
    },
    {
      "epoch": 167.04761904761904,
      "grad_norm": 6.027864456176758,
      "learning_rate": 3.2852323838080963e-07,
      "loss": 1.115,
      "step": 1754
    },
    {
      "epoch": 167.14285714285714,
      "grad_norm": 4.775788307189941,
      "learning_rate": 3.287106446776612e-07,
      "loss": 1.1697,
      "step": 1755
    },
    {
      "epoch": 167.23809523809524,
      "grad_norm": 5.333404064178467,
      "learning_rate": 3.288980509745128e-07,
      "loss": 1.2077,
      "step": 1756
    },
    {
      "epoch": 167.33333333333334,
      "grad_norm": 5.563086032867432,
      "learning_rate": 3.2908545727136434e-07,
      "loss": 1.156,
      "step": 1757
    },
    {
      "epoch": 167.42857142857142,
      "grad_norm": 5.439138412475586,
      "learning_rate": 3.292728635682159e-07,
      "loss": 1.172,
      "step": 1758
    },
    {
      "epoch": 167.52380952380952,
      "grad_norm": 3.481065511703491,
      "learning_rate": 3.294602698650675e-07,
      "loss": 1.1268,
      "step": 1759
    },
    {
      "epoch": 167.61904761904762,
      "grad_norm": 2.863832473754883,
      "learning_rate": 3.2964767616191906e-07,
      "loss": 1.1691,
      "step": 1760
    },
    {
      "epoch": 167.71428571428572,
      "grad_norm": 7.240921974182129,
      "learning_rate": 3.298350824587706e-07,
      "loss": 1.2117,
      "step": 1761
    },
    {
      "epoch": 167.8095238095238,
      "grad_norm": 3.0119240283966064,
      "learning_rate": 3.3002248875562217e-07,
      "loss": 1.1734,
      "step": 1762
    },
    {
      "epoch": 167.9047619047619,
      "grad_norm": 5.17049503326416,
      "learning_rate": 3.302098950524738e-07,
      "loss": 1.125,
      "step": 1763
    },
    {
      "epoch": 168.0,
      "grad_norm": 2.893317222595215,
      "learning_rate": 3.3039730134932533e-07,
      "loss": 1.1594,
      "step": 1764
    },
    {
      "epoch": 168.0,
      "eval_accuracy": 0.7159985013113526,
      "eval_f1": 0.7767962308598351,
      "eval_loss": 0.5774693489074707,
      "eval_precision": 0.7122030237580994,
      "eval_recall": 0.8542746113989638,
      "eval_roc_auc": 0.7565512377662635,
      "eval_runtime": 1.25,
      "eval_samples_per_second": 2135.174,
      "eval_steps_per_second": 4.8,
      "step": 1764
    },
    {
      "epoch": 168.0952380952381,
      "grad_norm": 4.347754955291748,
      "learning_rate": 3.3058470764617694e-07,
      "loss": 1.1856,
      "step": 1765
    },
    {
      "epoch": 168.1904761904762,
      "grad_norm": 3.9195492267608643,
      "learning_rate": 3.307721139430285e-07,
      "loss": 1.1709,
      "step": 1766
    },
    {
      "epoch": 168.28571428571428,
      "grad_norm": 6.3589887619018555,
      "learning_rate": 3.309595202398801e-07,
      "loss": 1.1727,
      "step": 1767
    },
    {
      "epoch": 168.38095238095238,
      "grad_norm": 2.310558319091797,
      "learning_rate": 3.3114692653673166e-07,
      "loss": 1.1734,
      "step": 1768
    },
    {
      "epoch": 168.47619047619048,
      "grad_norm": 3.736511707305908,
      "learning_rate": 3.313343328335832e-07,
      "loss": 1.1803,
      "step": 1769
    },
    {
      "epoch": 168.57142857142858,
      "grad_norm": 3.2041683197021484,
      "learning_rate": 3.315217391304348e-07,
      "loss": 1.2179,
      "step": 1770
    },
    {
      "epoch": 168.66666666666666,
      "grad_norm": 2.9342873096466064,
      "learning_rate": 3.3170914542728637e-07,
      "loss": 1.1647,
      "step": 1771
    },
    {
      "epoch": 168.76190476190476,
      "grad_norm": 4.083154201507568,
      "learning_rate": 3.318965517241379e-07,
      "loss": 1.1399,
      "step": 1772
    },
    {
      "epoch": 168.85714285714286,
      "grad_norm": 3.029041290283203,
      "learning_rate": 3.3208395802098953e-07,
      "loss": 1.1488,
      "step": 1773
    },
    {
      "epoch": 168.95238095238096,
      "grad_norm": 6.299585819244385,
      "learning_rate": 3.322713643178411e-07,
      "loss": 1.1108,
      "step": 1774
    },
    {
      "epoch": 168.95238095238096,
      "eval_accuracy": 0.7152491569876358,
      "eval_f1": 0.7744807121661721,
      "eval_loss": 0.5772000551223755,
      "eval_precision": 0.7146768893756845,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.75660103626943,
      "eval_runtime": 1.159,
      "eval_samples_per_second": 2302.943,
      "eval_steps_per_second": 5.177,
      "step": 1774
    },
    {
      "epoch": 169.04761904761904,
      "grad_norm": 3.145608901977539,
      "learning_rate": 3.3245877061469264e-07,
      "loss": 1.1624,
      "step": 1775
    },
    {
      "epoch": 169.14285714285714,
      "grad_norm": 3.2308030128479004,
      "learning_rate": 3.3264617691154425e-07,
      "loss": 1.1641,
      "step": 1776
    },
    {
      "epoch": 169.23809523809524,
      "grad_norm": 2.888650417327881,
      "learning_rate": 3.328335832083958e-07,
      "loss": 1.1488,
      "step": 1777
    },
    {
      "epoch": 169.33333333333334,
      "grad_norm": 6.054225444793701,
      "learning_rate": 3.330209895052474e-07,
      "loss": 1.1517,
      "step": 1778
    },
    {
      "epoch": 169.42857142857142,
      "grad_norm": 10.32518482208252,
      "learning_rate": 3.3320839580209897e-07,
      "loss": 1.1155,
      "step": 1779
    },
    {
      "epoch": 169.52380952380952,
      "grad_norm": 4.094943523406982,
      "learning_rate": 3.333958020989506e-07,
      "loss": 1.159,
      "step": 1780
    },
    {
      "epoch": 169.61904761904762,
      "grad_norm": 4.512625217437744,
      "learning_rate": 3.3358320839580213e-07,
      "loss": 1.148,
      "step": 1781
    },
    {
      "epoch": 169.71428571428572,
      "grad_norm": 3.9920194149017334,
      "learning_rate": 3.337706146926537e-07,
      "loss": 1.2131,
      "step": 1782
    },
    {
      "epoch": 169.8095238095238,
      "grad_norm": 3.589287519454956,
      "learning_rate": 3.3395802098950524e-07,
      "loss": 1.1572,
      "step": 1783
    },
    {
      "epoch": 169.9047619047619,
      "grad_norm": 5.493167400360107,
      "learning_rate": 3.3414542728635684e-07,
      "loss": 1.1631,
      "step": 1784
    },
    {
      "epoch": 170.0,
      "grad_norm": 3.7833290100097656,
      "learning_rate": 3.343328335832084e-07,
      "loss": 1.1449,
      "step": 1785
    },
    {
      "epoch": 170.0,
      "eval_accuracy": 0.7178718621206445,
      "eval_f1": 0.7775480059084195,
      "eval_loss": 0.577059268951416,
      "eval_precision": 0.714828897338403,
      "eval_recall": 0.8523316062176166,
      "eval_roc_auc": 0.7568560736902705,
      "eval_runtime": 1.1623,
      "eval_samples_per_second": 2296.241,
      "eval_steps_per_second": 5.162,
      "step": 1785
    },
    {
      "epoch": 170.0952380952381,
      "grad_norm": 3.7169711589813232,
      "learning_rate": 3.3452023988005995e-07,
      "loss": 1.1807,
      "step": 1786
    },
    {
      "epoch": 170.1904761904762,
      "grad_norm": 3.6557819843292236,
      "learning_rate": 3.3470764617691156e-07,
      "loss": 1.1478,
      "step": 1787
    },
    {
      "epoch": 170.28571428571428,
      "grad_norm": 2.8102827072143555,
      "learning_rate": 3.348950524737631e-07,
      "loss": 1.1509,
      "step": 1788
    },
    {
      "epoch": 170.38095238095238,
      "grad_norm": 4.472448348999023,
      "learning_rate": 3.350824587706147e-07,
      "loss": 1.2002,
      "step": 1789
    },
    {
      "epoch": 170.47619047619048,
      "grad_norm": 3.8644332885742188,
      "learning_rate": 3.352698650674663e-07,
      "loss": 1.1361,
      "step": 1790
    },
    {
      "epoch": 170.57142857142858,
      "grad_norm": 5.075953960418701,
      "learning_rate": 3.354572713643179e-07,
      "loss": 1.1837,
      "step": 1791
    },
    {
      "epoch": 170.66666666666666,
      "grad_norm": 2.4229235649108887,
      "learning_rate": 3.3564467766116944e-07,
      "loss": 1.1228,
      "step": 1792
    },
    {
      "epoch": 170.76190476190476,
      "grad_norm": 3.281778573989868,
      "learning_rate": 3.35832083958021e-07,
      "loss": 1.1924,
      "step": 1793
    },
    {
      "epoch": 170.85714285714286,
      "grad_norm": 8.002859115600586,
      "learning_rate": 3.360194902548726e-07,
      "loss": 1.1562,
      "step": 1794
    },
    {
      "epoch": 170.95238095238096,
      "grad_norm": 3.903043270111084,
      "learning_rate": 3.3620689655172416e-07,
      "loss": 1.1253,
      "step": 1795
    },
    {
      "epoch": 170.95238095238096,
      "eval_accuracy": 0.7171225177969277,
      "eval_f1": 0.7774830533451224,
      "eval_loss": 0.5771528482437134,
      "eval_precision": 0.7133585722011898,
      "eval_recall": 0.8542746113989638,
      "eval_roc_auc": 0.7570040299366724,
      "eval_runtime": 1.2237,
      "eval_samples_per_second": 2181.027,
      "eval_steps_per_second": 4.903,
      "step": 1795
    },
    {
      "epoch": 171.04761904761904,
      "grad_norm": 3.943495035171509,
      "learning_rate": 3.363943028485757e-07,
      "loss": 1.1992,
      "step": 1796
    },
    {
      "epoch": 171.14285714285714,
      "grad_norm": 7.058687210083008,
      "learning_rate": 3.3658170914542726e-07,
      "loss": 1.1109,
      "step": 1797
    },
    {
      "epoch": 171.23809523809524,
      "grad_norm": 5.804258823394775,
      "learning_rate": 3.3676911544227887e-07,
      "loss": 1.2426,
      "step": 1798
    },
    {
      "epoch": 171.33333333333334,
      "grad_norm": 6.50364875793457,
      "learning_rate": 3.369565217391304e-07,
      "loss": 1.1299,
      "step": 1799
    },
    {
      "epoch": 171.42857142857142,
      "grad_norm": 10.158699989318848,
      "learning_rate": 3.3714392803598203e-07,
      "loss": 1.1628,
      "step": 1800
    },
    {
      "epoch": 171.52380952380952,
      "grad_norm": 4.90827751159668,
      "learning_rate": 3.3733133433283364e-07,
      "loss": 1.1583,
      "step": 1801
    },
    {
      "epoch": 171.61904761904762,
      "grad_norm": 3.285327672958374,
      "learning_rate": 3.375187406296852e-07,
      "loss": 1.1732,
      "step": 1802
    },
    {
      "epoch": 171.71428571428572,
      "grad_norm": 6.1621222496032715,
      "learning_rate": 3.3770614692653675e-07,
      "loss": 1.1565,
      "step": 1803
    },
    {
      "epoch": 171.8095238095238,
      "grad_norm": 5.6082892417907715,
      "learning_rate": 3.378935532233883e-07,
      "loss": 1.1558,
      "step": 1804
    },
    {
      "epoch": 171.9047619047619,
      "grad_norm": 6.2096052169799805,
      "learning_rate": 3.380809595202399e-07,
      "loss": 1.1366,
      "step": 1805
    },
    {
      "epoch": 172.0,
      "grad_norm": 3.6200644969940186,
      "learning_rate": 3.3826836581709147e-07,
      "loss": 1.166,
      "step": 1806
    },
    {
      "epoch": 172.0,
      "eval_accuracy": 0.7141251405020607,
      "eval_f1": 0.7736576683476714,
      "eval_loss": 0.577314555644989,
      "eval_precision": 0.7137383689107827,
      "eval_recall": 0.844559585492228,
      "eval_roc_auc": 0.7566957397812321,
      "eval_runtime": 1.327,
      "eval_samples_per_second": 2011.312,
      "eval_steps_per_second": 4.521,
      "step": 1806
    },
    {
      "epoch": 172.0952380952381,
      "grad_norm": 2.324643135070801,
      "learning_rate": 3.38455772113943e-07,
      "loss": 1.1627,
      "step": 1807
    },
    {
      "epoch": 172.1904761904762,
      "grad_norm": 7.590202808380127,
      "learning_rate": 3.3864317841079463e-07,
      "loss": 1.1201,
      "step": 1808
    },
    {
      "epoch": 172.28571428571428,
      "grad_norm": 4.35092306137085,
      "learning_rate": 3.388305847076462e-07,
      "loss": 1.1914,
      "step": 1809
    },
    {
      "epoch": 172.38095238095238,
      "grad_norm": 4.167804718017578,
      "learning_rate": 3.3901799100449774e-07,
      "loss": 1.165,
      "step": 1810
    },
    {
      "epoch": 172.47619047619048,
      "grad_norm": 4.640730381011963,
      "learning_rate": 3.3920539730134934e-07,
      "loss": 1.1631,
      "step": 1811
    },
    {
      "epoch": 172.57142857142858,
      "grad_norm": 5.027839183807373,
      "learning_rate": 3.393928035982009e-07,
      "loss": 1.2238,
      "step": 1812
    },
    {
      "epoch": 172.66666666666666,
      "grad_norm": 3.8021483421325684,
      "learning_rate": 3.395802098950525e-07,
      "loss": 1.1714,
      "step": 1813
    },
    {
      "epoch": 172.76190476190476,
      "grad_norm": 5.696237564086914,
      "learning_rate": 3.3976761619190406e-07,
      "loss": 1.1689,
      "step": 1814
    },
    {
      "epoch": 172.85714285714286,
      "grad_norm": 4.191372871398926,
      "learning_rate": 3.3995502248875567e-07,
      "loss": 1.1471,
      "step": 1815
    },
    {
      "epoch": 172.95238095238096,
      "grad_norm": 2.8746988773345947,
      "learning_rate": 3.401424287856072e-07,
      "loss": 1.1472,
      "step": 1816
    },
    {
      "epoch": 172.95238095238096,
      "eval_accuracy": 0.7152491569876358,
      "eval_f1": 0.7764705882352941,
      "eval_loss": 0.5772579312324524,
      "eval_precision": 0.7112068965517241,
      "eval_recall": 0.8549222797927462,
      "eval_roc_auc": 0.7572544617156016,
      "eval_runtime": 1.1525,
      "eval_samples_per_second": 2315.776,
      "eval_steps_per_second": 5.206,
      "step": 1816
    },
    {
      "epoch": 173.04761904761904,
      "grad_norm": 5.165085315704346,
      "learning_rate": 3.403298350824588e-07,
      "loss": 1.1126,
      "step": 1817
    },
    {
      "epoch": 173.14285714285714,
      "grad_norm": 8.25855541229248,
      "learning_rate": 3.405172413793104e-07,
      "loss": 1.1671,
      "step": 1818
    },
    {
      "epoch": 173.23809523809524,
      "grad_norm": 5.805681228637695,
      "learning_rate": 3.4070464767616194e-07,
      "loss": 1.1901,
      "step": 1819
    },
    {
      "epoch": 173.33333333333334,
      "grad_norm": 5.523385047912598,
      "learning_rate": 3.408920539730135e-07,
      "loss": 1.1544,
      "step": 1820
    },
    {
      "epoch": 173.42857142857142,
      "grad_norm": 3.4633982181549072,
      "learning_rate": 3.4107946026986505e-07,
      "loss": 1.1649,
      "step": 1821
    },
    {
      "epoch": 173.52380952380952,
      "grad_norm": 3.0569186210632324,
      "learning_rate": 3.4126686656671666e-07,
      "loss": 1.1594,
      "step": 1822
    },
    {
      "epoch": 173.61904761904762,
      "grad_norm": 11.102819442749023,
      "learning_rate": 3.414542728635682e-07,
      "loss": 1.1987,
      "step": 1823
    },
    {
      "epoch": 173.71428571428572,
      "grad_norm": 7.979638576507568,
      "learning_rate": 3.416416791604198e-07,
      "loss": 1.1861,
      "step": 1824
    },
    {
      "epoch": 173.8095238095238,
      "grad_norm": 4.374381065368652,
      "learning_rate": 3.418290854572714e-07,
      "loss": 1.1493,
      "step": 1825
    },
    {
      "epoch": 173.9047619047619,
      "grad_norm": 4.893923282623291,
      "learning_rate": 3.42016491754123e-07,
      "loss": 1.1151,
      "step": 1826
    },
    {
      "epoch": 174.0,
      "grad_norm": 5.232106685638428,
      "learning_rate": 3.4220389805097453e-07,
      "loss": 1.1677,
      "step": 1827
    },
    {
      "epoch": 174.0,
      "eval_accuracy": 0.7167478456350693,
      "eval_f1": 0.7773851590106007,
      "eval_loss": 0.5769217014312744,
      "eval_precision": 0.712742980561555,
      "eval_recall": 0.8549222797927462,
      "eval_roc_auc": 0.7575754173862982,
      "eval_runtime": 1.2067,
      "eval_samples_per_second": 2211.813,
      "eval_steps_per_second": 4.972,
      "step": 1827
    },
    {
      "epoch": 174.0952380952381,
      "grad_norm": 2.487565040588379,
      "learning_rate": 3.423913043478261e-07,
      "loss": 1.1533,
      "step": 1828
    },
    {
      "epoch": 174.1904761904762,
      "grad_norm": 2.702059030532837,
      "learning_rate": 3.425787106446777e-07,
      "loss": 1.1271,
      "step": 1829
    },
    {
      "epoch": 174.28571428571428,
      "grad_norm": 4.765565395355225,
      "learning_rate": 3.4276611694152925e-07,
      "loss": 1.1576,
      "step": 1830
    },
    {
      "epoch": 174.38095238095238,
      "grad_norm": 4.310859680175781,
      "learning_rate": 3.429535232383808e-07,
      "loss": 1.1629,
      "step": 1831
    },
    {
      "epoch": 174.47619047619048,
      "grad_norm": 2.900235652923584,
      "learning_rate": 3.431409295352324e-07,
      "loss": 1.145,
      "step": 1832
    },
    {
      "epoch": 174.57142857142858,
      "grad_norm": 5.54879903793335,
      "learning_rate": 3.4332833583208397e-07,
      "loss": 1.1828,
      "step": 1833
    },
    {
      "epoch": 174.66666666666666,
      "grad_norm": 3.8463099002838135,
      "learning_rate": 3.435157421289355e-07,
      "loss": 1.2078,
      "step": 1834
    },
    {
      "epoch": 174.76190476190476,
      "grad_norm": 4.029839992523193,
      "learning_rate": 3.4370314842578713e-07,
      "loss": 1.176,
      "step": 1835
    },
    {
      "epoch": 174.85714285714286,
      "grad_norm": 6.263148307800293,
      "learning_rate": 3.4389055472263874e-07,
      "loss": 1.1302,
      "step": 1836
    },
    {
      "epoch": 174.95238095238096,
      "grad_norm": 3.3856215476989746,
      "learning_rate": 3.440779610194903e-07,
      "loss": 1.1422,
      "step": 1837
    },
    {
      "epoch": 174.95238095238096,
      "eval_accuracy": 0.7137504683402023,
      "eval_f1": 0.7732937685459941,
      "eval_loss": 0.5764679908752441,
      "eval_precision": 0.7135815991237678,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7579099021301094,
      "eval_runtime": 1.2138,
      "eval_samples_per_second": 2198.854,
      "eval_steps_per_second": 4.943,
      "step": 1837
    },
    {
      "epoch": 175.04761904761904,
      "grad_norm": 4.67045783996582,
      "learning_rate": 3.4426536731634185e-07,
      "loss": 1.1808,
      "step": 1838
    },
    {
      "epoch": 175.14285714285714,
      "grad_norm": 5.133697986602783,
      "learning_rate": 3.4445277361319345e-07,
      "loss": 1.1475,
      "step": 1839
    },
    {
      "epoch": 175.23809523809524,
      "grad_norm": 2.8523051738739014,
      "learning_rate": 3.44640179910045e-07,
      "loss": 1.1904,
      "step": 1840
    },
    {
      "epoch": 175.33333333333334,
      "grad_norm": 4.153690814971924,
      "learning_rate": 3.4482758620689656e-07,
      "loss": 1.125,
      "step": 1841
    },
    {
      "epoch": 175.42857142857142,
      "grad_norm": 2.2273285388946533,
      "learning_rate": 3.450149925037481e-07,
      "loss": 1.1406,
      "step": 1842
    },
    {
      "epoch": 175.52380952380952,
      "grad_norm": 7.832995891571045,
      "learning_rate": 3.452023988005997e-07,
      "loss": 1.0953,
      "step": 1843
    },
    {
      "epoch": 175.61904761904762,
      "grad_norm": 5.448593616485596,
      "learning_rate": 3.453898050974513e-07,
      "loss": 1.208,
      "step": 1844
    },
    {
      "epoch": 175.71428571428572,
      "grad_norm": 5.135989665985107,
      "learning_rate": 3.4557721139430283e-07,
      "loss": 1.1912,
      "step": 1845
    },
    {
      "epoch": 175.8095238095238,
      "grad_norm": 5.489715099334717,
      "learning_rate": 3.4576461769115444e-07,
      "loss": 1.1702,
      "step": 1846
    },
    {
      "epoch": 175.9047619047619,
      "grad_norm": 4.431760311126709,
      "learning_rate": 3.45952023988006e-07,
      "loss": 1.1431,
      "step": 1847
    },
    {
      "epoch": 176.0,
      "grad_norm": 3.1945583820343018,
      "learning_rate": 3.461394302848576e-07,
      "loss": 1.1766,
      "step": 1848
    },
    {
      "epoch": 176.0,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7779740871613663,
      "eval_loss": 0.5763475894927979,
      "eval_precision": 0.7132829373650108,
      "eval_recall": 0.8555699481865285,
      "eval_roc_auc": 0.7584815774323548,
      "eval_runtime": 1.3454,
      "eval_samples_per_second": 1983.803,
      "eval_steps_per_second": 4.46,
      "step": 1848
    },
    {
      "epoch": 176.0952380952381,
      "grad_norm": 5.7543439865112305,
      "learning_rate": 3.4632683658170916e-07,
      "loss": 1.1441,
      "step": 1849
    },
    {
      "epoch": 176.1904761904762,
      "grad_norm": 2.080615520477295,
      "learning_rate": 3.4651424287856076e-07,
      "loss": 1.1376,
      "step": 1850
    },
    {
      "epoch": 176.28571428571428,
      "grad_norm": 11.54956340789795,
      "learning_rate": 3.467016491754123e-07,
      "loss": 1.2058,
      "step": 1851
    },
    {
      "epoch": 176.38095238095238,
      "grad_norm": 6.846175193786621,
      "learning_rate": 3.4688905547226387e-07,
      "loss": 1.2177,
      "step": 1852
    },
    {
      "epoch": 176.47619047619048,
      "grad_norm": 5.839255332946777,
      "learning_rate": 3.470764617691155e-07,
      "loss": 1.1688,
      "step": 1853
    },
    {
      "epoch": 176.57142857142858,
      "grad_norm": 2.9963738918304443,
      "learning_rate": 3.4726386806596703e-07,
      "loss": 1.1329,
      "step": 1854
    },
    {
      "epoch": 176.66666666666666,
      "grad_norm": 3.8140134811401367,
      "learning_rate": 3.474512743628186e-07,
      "loss": 1.1437,
      "step": 1855
    },
    {
      "epoch": 176.76190476190476,
      "grad_norm": 4.6968207359313965,
      "learning_rate": 3.4763868065967014e-07,
      "loss": 1.1591,
      "step": 1856
    },
    {
      "epoch": 176.85714285714286,
      "grad_norm": 6.985243797302246,
      "learning_rate": 3.4782608695652175e-07,
      "loss": 1.1364,
      "step": 1857
    },
    {
      "epoch": 176.95238095238096,
      "grad_norm": 4.726886749267578,
      "learning_rate": 3.480134932533733e-07,
      "loss": 1.1693,
      "step": 1858
    },
    {
      "epoch": 176.95238095238096,
      "eval_accuracy": 0.7159985013113526,
      "eval_f1": 0.7761370348493798,
      "eval_loss": 0.5765255093574524,
      "eval_precision": 0.7133550488599348,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.758339378238342,
      "eval_runtime": 1.1569,
      "eval_samples_per_second": 2307.045,
      "eval_steps_per_second": 5.186,
      "step": 1858
    },
    {
      "epoch": 177.04761904761904,
      "grad_norm": 11.07115364074707,
      "learning_rate": 3.482008995502249e-07,
      "loss": 1.1426,
      "step": 1859
    },
    {
      "epoch": 177.14285714285714,
      "grad_norm": 2.7573394775390625,
      "learning_rate": 3.483883058470765e-07,
      "loss": 1.1399,
      "step": 1860
    },
    {
      "epoch": 177.23809523809524,
      "grad_norm": 5.692921161651611,
      "learning_rate": 3.485757121439281e-07,
      "loss": 1.1824,
      "step": 1861
    },
    {
      "epoch": 177.33333333333334,
      "grad_norm": 4.1265339851379395,
      "learning_rate": 3.4876311844077963e-07,
      "loss": 1.1755,
      "step": 1862
    },
    {
      "epoch": 177.42857142857142,
      "grad_norm": 6.006172180175781,
      "learning_rate": 3.489505247376312e-07,
      "loss": 1.1776,
      "step": 1863
    },
    {
      "epoch": 177.52380952380952,
      "grad_norm": 4.846124649047852,
      "learning_rate": 3.491379310344828e-07,
      "loss": 1.1543,
      "step": 1864
    },
    {
      "epoch": 177.61904761904762,
      "grad_norm": 3.1988956928253174,
      "learning_rate": 3.4932533733133435e-07,
      "loss": 1.1367,
      "step": 1865
    },
    {
      "epoch": 177.71428571428572,
      "grad_norm": 5.961909770965576,
      "learning_rate": 3.495127436281859e-07,
      "loss": 1.1896,
      "step": 1866
    },
    {
      "epoch": 177.8095238095238,
      "grad_norm": 3.8654661178588867,
      "learning_rate": 3.497001499250375e-07,
      "loss": 1.1268,
      "step": 1867
    },
    {
      "epoch": 177.9047619047619,
      "grad_norm": 4.253927707672119,
      "learning_rate": 3.4988755622188906e-07,
      "loss": 1.1478,
      "step": 1868
    },
    {
      "epoch": 178.0,
      "grad_norm": 11.02048397064209,
      "learning_rate": 3.500749625187406e-07,
      "loss": 1.146,
      "step": 1869
    },
    {
      "epoch": 178.0,
      "eval_accuracy": 0.7156238291494942,
      "eval_f1": 0.7747105966162066,
      "eval_loss": 0.5769871473312378,
      "eval_precision": 0.7150684931506849,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7574484743811168,
      "eval_runtime": 1.1238,
      "eval_samples_per_second": 2374.888,
      "eval_steps_per_second": 5.339,
      "step": 1869
    },
    {
      "epoch": 178.0952380952381,
      "grad_norm": 4.868641376495361,
      "learning_rate": 3.502623688155922e-07,
      "loss": 1.1933,
      "step": 1870
    },
    {
      "epoch": 178.1904761904762,
      "grad_norm": 5.7350335121154785,
      "learning_rate": 3.5044977511244383e-07,
      "loss": 1.1308,
      "step": 1871
    },
    {
      "epoch": 178.28571428571428,
      "grad_norm": 2.5435409545898438,
      "learning_rate": 3.506371814092954e-07,
      "loss": 1.158,
      "step": 1872
    },
    {
      "epoch": 178.38095238095238,
      "grad_norm": 5.791324138641357,
      "learning_rate": 3.5082458770614694e-07,
      "loss": 1.167,
      "step": 1873
    },
    {
      "epoch": 178.47619047619048,
      "grad_norm": 5.436519622802734,
      "learning_rate": 3.5101199400299855e-07,
      "loss": 1.1139,
      "step": 1874
    },
    {
      "epoch": 178.57142857142858,
      "grad_norm": 3.6552703380584717,
      "learning_rate": 3.511994002998501e-07,
      "loss": 1.1787,
      "step": 1875
    },
    {
      "epoch": 178.66666666666666,
      "grad_norm": 5.235700607299805,
      "learning_rate": 3.5138680659670166e-07,
      "loss": 1.175,
      "step": 1876
    },
    {
      "epoch": 178.76190476190476,
      "grad_norm": 7.033854007720947,
      "learning_rate": 3.515742128935532e-07,
      "loss": 1.1755,
      "step": 1877
    },
    {
      "epoch": 178.85714285714286,
      "grad_norm": 6.690187931060791,
      "learning_rate": 3.517616191904048e-07,
      "loss": 1.1441,
      "step": 1878
    },
    {
      "epoch": 178.95238095238096,
      "grad_norm": 12.126110076904297,
      "learning_rate": 3.5194902548725637e-07,
      "loss": 1.189,
      "step": 1879
    },
    {
      "epoch": 178.95238095238096,
      "eval_accuracy": 0.7159985013113526,
      "eval_f1": 0.7771898883009994,
      "eval_loss": 0.5770087242126465,
      "eval_precision": 0.7115177610333692,
      "eval_recall": 0.8562176165803109,
      "eval_roc_auc": 0.757912204951065,
      "eval_runtime": 1.1826,
      "eval_samples_per_second": 2256.886,
      "eval_steps_per_second": 5.074,
      "step": 1879
    },
    {
      "epoch": 179.04761904761904,
      "grad_norm": 6.9428205490112305,
      "learning_rate": 3.5213643178410793e-07,
      "loss": 1.1226,
      "step": 1880
    },
    {
      "epoch": 179.14285714285714,
      "grad_norm": 4.363292217254639,
      "learning_rate": 3.5232383808095954e-07,
      "loss": 1.2072,
      "step": 1881
    },
    {
      "epoch": 179.23809523809524,
      "grad_norm": 4.244919776916504,
      "learning_rate": 3.5251124437781114e-07,
      "loss": 1.1241,
      "step": 1882
    },
    {
      "epoch": 179.33333333333334,
      "grad_norm": 3.1588680744171143,
      "learning_rate": 3.526986506746627e-07,
      "loss": 1.1578,
      "step": 1883
    },
    {
      "epoch": 179.42857142857142,
      "grad_norm": 3.038062334060669,
      "learning_rate": 3.5288605697151425e-07,
      "loss": 1.1442,
      "step": 1884
    },
    {
      "epoch": 179.52380952380952,
      "grad_norm": 6.6813788414001465,
      "learning_rate": 3.5307346326836586e-07,
      "loss": 1.1435,
      "step": 1885
    },
    {
      "epoch": 179.61904761904762,
      "grad_norm": 3.848639488220215,
      "learning_rate": 3.532608695652174e-07,
      "loss": 1.1718,
      "step": 1886
    },
    {
      "epoch": 179.71428571428572,
      "grad_norm": 5.688107490539551,
      "learning_rate": 3.5344827586206897e-07,
      "loss": 1.166,
      "step": 1887
    },
    {
      "epoch": 179.8095238095238,
      "grad_norm": 11.533185005187988,
      "learning_rate": 3.536356821589206e-07,
      "loss": 1.1493,
      "step": 1888
    },
    {
      "epoch": 179.9047619047619,
      "grad_norm": 11.271069526672363,
      "learning_rate": 3.5382308845577213e-07,
      "loss": 1.1494,
      "step": 1889
    },
    {
      "epoch": 180.0,
      "grad_norm": 5.525691509246826,
      "learning_rate": 3.540104947526237e-07,
      "loss": 1.189,
      "step": 1890
    },
    {
      "epoch": 180.0,
      "eval_accuracy": 0.7159985013113526,
      "eval_f1": 0.7746730083234245,
      "eval_loss": 0.5767568349838257,
      "eval_precision": 0.7159340659340659,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.75776597582038,
      "eval_runtime": 1.1887,
      "eval_samples_per_second": 2245.287,
      "eval_steps_per_second": 5.047,
      "step": 1890
    },
    {
      "epoch": 180.0952380952381,
      "grad_norm": 3.933689832687378,
      "learning_rate": 3.5419790104947524e-07,
      "loss": 1.1657,
      "step": 1891
    },
    {
      "epoch": 180.1904761904762,
      "grad_norm": 5.561554431915283,
      "learning_rate": 3.5438530734632685e-07,
      "loss": 1.1765,
      "step": 1892
    },
    {
      "epoch": 180.28571428571428,
      "grad_norm": 3.654136896133423,
      "learning_rate": 3.545727136431784e-07,
      "loss": 1.1255,
      "step": 1893
    },
    {
      "epoch": 180.38095238095238,
      "grad_norm": 6.423804759979248,
      "learning_rate": 3.5476011994003e-07,
      "loss": 1.1659,
      "step": 1894
    },
    {
      "epoch": 180.47619047619048,
      "grad_norm": 2.809739351272583,
      "learning_rate": 3.549475262368816e-07,
      "loss": 1.1434,
      "step": 1895
    },
    {
      "epoch": 180.57142857142858,
      "grad_norm": 4.307884693145752,
      "learning_rate": 3.5513493253373317e-07,
      "loss": 1.1903,
      "step": 1896
    },
    {
      "epoch": 180.66666666666666,
      "grad_norm": 5.794278144836426,
      "learning_rate": 3.553223388305847e-07,
      "loss": 1.1238,
      "step": 1897
    },
    {
      "epoch": 180.76190476190476,
      "grad_norm": 6.3453450202941895,
      "learning_rate": 3.555097451274363e-07,
      "loss": 1.1934,
      "step": 1898
    },
    {
      "epoch": 180.85714285714286,
      "grad_norm": 2.430403232574463,
      "learning_rate": 3.556971514242879e-07,
      "loss": 1.1543,
      "step": 1899
    },
    {
      "epoch": 180.95238095238096,
      "grad_norm": 10.58128833770752,
      "learning_rate": 3.5588455772113944e-07,
      "loss": 1.1845,
      "step": 1900
    },
    {
      "epoch": 180.95238095238096,
      "eval_accuracy": 0.7156238291494942,
      "eval_f1": 0.7753773305711749,
      "eval_loss": 0.5764981508255005,
      "eval_precision": 0.7138964577656676,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7583736327000575,
      "eval_runtime": 1.2787,
      "eval_samples_per_second": 2087.334,
      "eval_steps_per_second": 4.692,
      "step": 1900
    },
    {
      "epoch": 181.04761904761904,
      "grad_norm": 2.947765827178955,
      "learning_rate": 3.56071964017991e-07,
      "loss": 1.1545,
      "step": 1901
    },
    {
      "epoch": 181.14285714285714,
      "grad_norm": 4.267097473144531,
      "learning_rate": 3.562593703148426e-07,
      "loss": 1.1873,
      "step": 1902
    },
    {
      "epoch": 181.23809523809524,
      "grad_norm": 12.315367698669434,
      "learning_rate": 3.5644677661169416e-07,
      "loss": 1.116,
      "step": 1903
    },
    {
      "epoch": 181.33333333333334,
      "grad_norm": 7.7876715660095215,
      "learning_rate": 3.566341829085457e-07,
      "loss": 1.1758,
      "step": 1904
    },
    {
      "epoch": 181.42857142857142,
      "grad_norm": 7.434537410736084,
      "learning_rate": 3.568215892053973e-07,
      "loss": 1.1752,
      "step": 1905
    },
    {
      "epoch": 181.52380952380952,
      "grad_norm": 5.361802101135254,
      "learning_rate": 3.570089955022489e-07,
      "loss": 1.1933,
      "step": 1906
    },
    {
      "epoch": 181.61904761904762,
      "grad_norm": 2.9396021366119385,
      "learning_rate": 3.571964017991005e-07,
      "loss": 1.1814,
      "step": 1907
    },
    {
      "epoch": 181.71428571428572,
      "grad_norm": 9.127396583557129,
      "learning_rate": 3.5738380809595204e-07,
      "loss": 1.131,
      "step": 1908
    },
    {
      "epoch": 181.8095238095238,
      "grad_norm": 7.989266872406006,
      "learning_rate": 3.5757121439280364e-07,
      "loss": 1.1635,
      "step": 1909
    },
    {
      "epoch": 181.9047619047619,
      "grad_norm": 5.933679103851318,
      "learning_rate": 3.577586206896552e-07,
      "loss": 1.149,
      "step": 1910
    },
    {
      "epoch": 182.0,
      "grad_norm": 4.823862552642822,
      "learning_rate": 3.5794602698650675e-07,
      "loss": 1.1308,
      "step": 1911
    },
    {
      "epoch": 182.0,
      "eval_accuracy": 0.7159985013113526,
      "eval_f1": 0.7757396449704143,
      "eval_loss": 0.5764167308807373,
      "eval_precision": 0.7140522875816994,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.758552677029361,
      "eval_runtime": 1.1291,
      "eval_samples_per_second": 2363.802,
      "eval_steps_per_second": 5.314,
      "step": 1911
    },
    {
      "epoch": 182.0952380952381,
      "grad_norm": 3.4051706790924072,
      "learning_rate": 3.581334332833583e-07,
      "loss": 1.1716,
      "step": 1912
    },
    {
      "epoch": 182.1904761904762,
      "grad_norm": 3.0059852600097656,
      "learning_rate": 3.583208395802099e-07,
      "loss": 1.1393,
      "step": 1913
    },
    {
      "epoch": 182.28571428571428,
      "grad_norm": 3.3431499004364014,
      "learning_rate": 3.5850824587706147e-07,
      "loss": 1.1219,
      "step": 1914
    },
    {
      "epoch": 182.38095238095238,
      "grad_norm": 3.8559226989746094,
      "learning_rate": 3.58695652173913e-07,
      "loss": 1.1325,
      "step": 1915
    },
    {
      "epoch": 182.47619047619048,
      "grad_norm": 4.502316474914551,
      "learning_rate": 3.5888305847076463e-07,
      "loss": 1.1543,
      "step": 1916
    },
    {
      "epoch": 182.57142857142858,
      "grad_norm": 8.332260131835938,
      "learning_rate": 3.5907046476761624e-07,
      "loss": 1.1884,
      "step": 1917
    },
    {
      "epoch": 182.66666666666666,
      "grad_norm": 6.321863174438477,
      "learning_rate": 3.592578710644678e-07,
      "loss": 1.191,
      "step": 1918
    },
    {
      "epoch": 182.76190476190476,
      "grad_norm": 3.683375597000122,
      "learning_rate": 3.5944527736131935e-07,
      "loss": 1.1066,
      "step": 1919
    },
    {
      "epoch": 182.85714285714286,
      "grad_norm": 5.32153844833374,
      "learning_rate": 3.5963268365817095e-07,
      "loss": 1.1706,
      "step": 1920
    },
    {
      "epoch": 182.95238095238096,
      "grad_norm": 4.957334995269775,
      "learning_rate": 3.598200899550225e-07,
      "loss": 1.1953,
      "step": 1921
    },
    {
      "epoch": 182.95238095238096,
      "eval_accuracy": 0.7152491569876358,
      "eval_f1": 0.7760754272245138,
      "eval_loss": 0.576198160648346,
      "eval_precision": 0.7118918918918918,
      "eval_recall": 0.852979274611399,
      "eval_roc_auc": 0.7590506620610248,
      "eval_runtime": 1.2035,
      "eval_samples_per_second": 2217.773,
      "eval_steps_per_second": 4.986,
      "step": 1921
    },
    {
      "epoch": 183.04761904761904,
      "grad_norm": 6.652534484863281,
      "learning_rate": 3.6000749625187406e-07,
      "loss": 1.2154,
      "step": 1922
    },
    {
      "epoch": 183.14285714285714,
      "grad_norm": 2.489819288253784,
      "learning_rate": 3.6019490254872567e-07,
      "loss": 1.149,
      "step": 1923
    },
    {
      "epoch": 183.23809523809524,
      "grad_norm": 5.232963562011719,
      "learning_rate": 3.603823088455772e-07,
      "loss": 1.2149,
      "step": 1924
    },
    {
      "epoch": 183.33333333333334,
      "grad_norm": 5.250478267669678,
      "learning_rate": 3.605697151424288e-07,
      "loss": 1.1563,
      "step": 1925
    },
    {
      "epoch": 183.42857142857142,
      "grad_norm": 8.206533432006836,
      "learning_rate": 3.607571214392804e-07,
      "loss": 1.1054,
      "step": 1926
    },
    {
      "epoch": 183.52380952380952,
      "grad_norm": 3.369680166244507,
      "learning_rate": 3.6094452773613194e-07,
      "loss": 1.1832,
      "step": 1927
    },
    {
      "epoch": 183.61904761904762,
      "grad_norm": 4.8924078941345215,
      "learning_rate": 3.611319340329835e-07,
      "loss": 1.1753,
      "step": 1928
    },
    {
      "epoch": 183.71428571428572,
      "grad_norm": 4.222039222717285,
      "learning_rate": 3.613193403298351e-07,
      "loss": 1.1668,
      "step": 1929
    },
    {
      "epoch": 183.8095238095238,
      "grad_norm": 2.895860433578491,
      "learning_rate": 3.615067466266867e-07,
      "loss": 1.1363,
      "step": 1930
    },
    {
      "epoch": 183.9047619047619,
      "grad_norm": 5.144315719604492,
      "learning_rate": 3.6169415292353826e-07,
      "loss": 1.1378,
      "step": 1931
    },
    {
      "epoch": 184.0,
      "grad_norm": 8.487016677856445,
      "learning_rate": 3.618815592203898e-07,
      "loss": 1.1334,
      "step": 1932
    },
    {
      "epoch": 184.0,
      "eval_accuracy": 0.7167478456350693,
      "eval_f1": 0.7763313609467456,
      "eval_loss": 0.5758945941925049,
      "eval_precision": 0.7145969498910676,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7593747841105355,
      "eval_runtime": 1.2153,
      "eval_samples_per_second": 2196.076,
      "eval_steps_per_second": 4.937,
      "step": 1932
    },
    {
      "epoch": 184.0952380952381,
      "grad_norm": 6.197401523590088,
      "learning_rate": 3.6206896551724143e-07,
      "loss": 1.1115,
      "step": 1933
    },
    {
      "epoch": 184.1904761904762,
      "grad_norm": 8.295247077941895,
      "learning_rate": 3.62256371814093e-07,
      "loss": 1.1771,
      "step": 1934
    },
    {
      "epoch": 184.28571428571428,
      "grad_norm": 3.6977932453155518,
      "learning_rate": 3.6244377811094454e-07,
      "loss": 1.1489,
      "step": 1935
    },
    {
      "epoch": 184.38095238095238,
      "grad_norm": 6.017064094543457,
      "learning_rate": 3.626311844077961e-07,
      "loss": 1.1236,
      "step": 1936
    },
    {
      "epoch": 184.47619047619048,
      "grad_norm": 4.377641677856445,
      "learning_rate": 3.628185907046477e-07,
      "loss": 1.1521,
      "step": 1937
    },
    {
      "epoch": 184.57142857142858,
      "grad_norm": 12.319416999816895,
      "learning_rate": 3.6300599700149925e-07,
      "loss": 1.1957,
      "step": 1938
    },
    {
      "epoch": 184.66666666666666,
      "grad_norm": 5.208367824554443,
      "learning_rate": 3.631934032983508e-07,
      "loss": 1.1538,
      "step": 1939
    },
    {
      "epoch": 184.76190476190476,
      "grad_norm": 8.780491828918457,
      "learning_rate": 3.633808095952024e-07,
      "loss": 1.165,
      "step": 1940
    },
    {
      "epoch": 184.85714285714286,
      "grad_norm": 12.607026100158691,
      "learning_rate": 3.63568215892054e-07,
      "loss": 1.1552,
      "step": 1941
    },
    {
      "epoch": 184.95238095238096,
      "grad_norm": 7.559422492980957,
      "learning_rate": 3.637556221889056e-07,
      "loss": 1.2093,
      "step": 1942
    },
    {
      "epoch": 184.95238095238096,
      "eval_accuracy": 0.7159985013113526,
      "eval_f1": 0.7782328847279111,
      "eval_loss": 0.575957179069519,
      "eval_precision": 0.7097118463180363,
      "eval_recall": 0.8613989637305699,
      "eval_roc_auc": 0.7598085780080599,
      "eval_runtime": 1.3194,
      "eval_samples_per_second": 2022.932,
      "eval_steps_per_second": 4.548,
      "step": 1942
    },
    {
      "epoch": 185.04761904761904,
      "grad_norm": 4.233537197113037,
      "learning_rate": 3.6394302848575713e-07,
      "loss": 1.1932,
      "step": 1943
    },
    {
      "epoch": 185.14285714285714,
      "grad_norm": 7.4202752113342285,
      "learning_rate": 3.6413043478260874e-07,
      "loss": 1.1603,
      "step": 1944
    },
    {
      "epoch": 185.23809523809524,
      "grad_norm": 2.546308755874634,
      "learning_rate": 3.643178410794603e-07,
      "loss": 1.1637,
      "step": 1945
    },
    {
      "epoch": 185.33333333333334,
      "grad_norm": 2.575289011001587,
      "learning_rate": 3.6450524737631185e-07,
      "loss": 1.2095,
      "step": 1946
    },
    {
      "epoch": 185.42857142857142,
      "grad_norm": 3.5033421516418457,
      "learning_rate": 3.6469265367316345e-07,
      "loss": 1.1268,
      "step": 1947
    },
    {
      "epoch": 185.52380952380952,
      "grad_norm": 7.139731407165527,
      "learning_rate": 3.64880059970015e-07,
      "loss": 1.1341,
      "step": 1948
    },
    {
      "epoch": 185.61904761904762,
      "grad_norm": 2.049715995788574,
      "learning_rate": 3.6506746626686656e-07,
      "loss": 1.1605,
      "step": 1949
    },
    {
      "epoch": 185.71428571428572,
      "grad_norm": 3.4298887252807617,
      "learning_rate": 3.652548725637181e-07,
      "loss": 1.1748,
      "step": 1950
    },
    {
      "epoch": 185.8095238095238,
      "grad_norm": 11.024620056152344,
      "learning_rate": 3.654422788605697e-07,
      "loss": 1.1378,
      "step": 1951
    },
    {
      "epoch": 185.9047619047619,
      "grad_norm": 3.8029916286468506,
      "learning_rate": 3.6562968515742133e-07,
      "loss": 1.1897,
      "step": 1952
    },
    {
      "epoch": 186.0,
      "grad_norm": 3.457836627960205,
      "learning_rate": 3.658170914542729e-07,
      "loss": 1.1673,
      "step": 1953
    },
    {
      "epoch": 186.0,
      "eval_accuracy": 0.7156238291494942,
      "eval_f1": 0.77390527256479,
      "eval_loss": 0.5756555795669556,
      "eval_precision": 0.716492002206288,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.7596646516983304,
      "eval_runtime": 1.192,
      "eval_samples_per_second": 2239.091,
      "eval_steps_per_second": 5.034,
      "step": 1953
    },
    {
      "epoch": 186.0952380952381,
      "grad_norm": 3.491661787033081,
      "learning_rate": 3.660044977511245e-07,
      "loss": 1.1534,
      "step": 1954
    },
    {
      "epoch": 186.1904761904762,
      "grad_norm": 5.940413951873779,
      "learning_rate": 3.6619190404797605e-07,
      "loss": 1.1502,
      "step": 1955
    },
    {
      "epoch": 186.28571428571428,
      "grad_norm": 3.201045036315918,
      "learning_rate": 3.663793103448276e-07,
      "loss": 1.1731,
      "step": 1956
    },
    {
      "epoch": 186.38095238095238,
      "grad_norm": 3.6759390830993652,
      "learning_rate": 3.6656671664167916e-07,
      "loss": 1.1626,
      "step": 1957
    },
    {
      "epoch": 186.47619047619048,
      "grad_norm": 2.4864330291748047,
      "learning_rate": 3.6675412293853077e-07,
      "loss": 1.1512,
      "step": 1958
    },
    {
      "epoch": 186.57142857142858,
      "grad_norm": 3.656336545944214,
      "learning_rate": 3.669415292353823e-07,
      "loss": 1.1399,
      "step": 1959
    },
    {
      "epoch": 186.66666666666666,
      "grad_norm": 2.4883368015289307,
      "learning_rate": 3.671289355322339e-07,
      "loss": 1.1748,
      "step": 1960
    },
    {
      "epoch": 186.76190476190476,
      "grad_norm": 5.662504196166992,
      "learning_rate": 3.673163418290855e-07,
      "loss": 1.1648,
      "step": 1961
    },
    {
      "epoch": 186.85714285714286,
      "grad_norm": 3.2387423515319824,
      "learning_rate": 3.6750374812593704e-07,
      "loss": 1.1512,
      "step": 1962
    },
    {
      "epoch": 186.95238095238096,
      "grad_norm": 5.380378723144531,
      "learning_rate": 3.6769115442278864e-07,
      "loss": 1.1682,
      "step": 1963
    },
    {
      "epoch": 186.95238095238096,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7779740871613663,
      "eval_loss": 0.5752643942832947,
      "eval_precision": 0.7132829373650108,
      "eval_recall": 0.8555699481865285,
      "eval_roc_auc": 0.7605184225676453,
      "eval_runtime": 1.2174,
      "eval_samples_per_second": 2192.344,
      "eval_steps_per_second": 4.928,
      "step": 1963
    },
    {
      "epoch": 187.04761904761904,
      "grad_norm": 4.980207443237305,
      "learning_rate": 3.678785607196402e-07,
      "loss": 1.2062,
      "step": 1964
    },
    {
      "epoch": 187.14285714285714,
      "grad_norm": 3.1846394538879395,
      "learning_rate": 3.680659670164918e-07,
      "loss": 1.1531,
      "step": 1965
    },
    {
      "epoch": 187.23809523809524,
      "grad_norm": 2.922837495803833,
      "learning_rate": 3.6825337331334336e-07,
      "loss": 1.1524,
      "step": 1966
    },
    {
      "epoch": 187.33333333333334,
      "grad_norm": 3.404749870300293,
      "learning_rate": 3.684407796101949e-07,
      "loss": 1.1296,
      "step": 1967
    },
    {
      "epoch": 187.42857142857142,
      "grad_norm": 3.8887367248535156,
      "learning_rate": 3.686281859070465e-07,
      "loss": 1.1537,
      "step": 1968
    },
    {
      "epoch": 187.52380952380952,
      "grad_norm": 4.6107611656188965,
      "learning_rate": 3.688155922038981e-07,
      "loss": 1.1757,
      "step": 1969
    },
    {
      "epoch": 187.61904761904762,
      "grad_norm": 4.161106109619141,
      "learning_rate": 3.6900299850074963e-07,
      "loss": 1.1987,
      "step": 1970
    },
    {
      "epoch": 187.71428571428572,
      "grad_norm": 6.394876480102539,
      "learning_rate": 3.691904047976012e-07,
      "loss": 1.1902,
      "step": 1971
    },
    {
      "epoch": 187.8095238095238,
      "grad_norm": 8.40281867980957,
      "learning_rate": 3.693778110944528e-07,
      "loss": 1.151,
      "step": 1972
    },
    {
      "epoch": 187.9047619047619,
      "grad_norm": 5.9665937423706055,
      "learning_rate": 3.6956521739130435e-07,
      "loss": 1.1155,
      "step": 1973
    },
    {
      "epoch": 188.0,
      "grad_norm": 3.7505180835723877,
      "learning_rate": 3.697526236881559e-07,
      "loss": 1.187,
      "step": 1974
    },
    {
      "epoch": 188.0,
      "eval_accuracy": 0.7152491569876358,
      "eval_f1": 0.7740784780023782,
      "eval_loss": 0.5748009085655212,
      "eval_precision": 0.7153846153846154,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7608298791018998,
      "eval_runtime": 1.1991,
      "eval_samples_per_second": 2225.752,
      "eval_steps_per_second": 5.004,
      "step": 1974
    },
    {
      "epoch": 188.0952380952381,
      "grad_norm": 2.0834028720855713,
      "learning_rate": 3.699400299850075e-07,
      "loss": 1.1773,
      "step": 1975
    },
    {
      "epoch": 188.1904761904762,
      "grad_norm": 5.331453800201416,
      "learning_rate": 3.701274362818591e-07,
      "loss": 1.1259,
      "step": 1976
    },
    {
      "epoch": 188.28571428571428,
      "grad_norm": 5.906782150268555,
      "learning_rate": 3.7031484257871067e-07,
      "loss": 1.2037,
      "step": 1977
    },
    {
      "epoch": 188.38095238095238,
      "grad_norm": 4.825325012207031,
      "learning_rate": 3.705022488755622e-07,
      "loss": 1.1547,
      "step": 1978
    },
    {
      "epoch": 188.47619047619048,
      "grad_norm": 6.54678201675415,
      "learning_rate": 3.7068965517241383e-07,
      "loss": 1.148,
      "step": 1979
    },
    {
      "epoch": 188.57142857142858,
      "grad_norm": 4.087493419647217,
      "learning_rate": 3.708770614692654e-07,
      "loss": 1.1481,
      "step": 1980
    },
    {
      "epoch": 188.66666666666666,
      "grad_norm": 4.3248467445373535,
      "learning_rate": 3.7106446776611694e-07,
      "loss": 1.1438,
      "step": 1981
    },
    {
      "epoch": 188.76190476190476,
      "grad_norm": 2.789849281311035,
      "learning_rate": 3.7125187406296855e-07,
      "loss": 1.1641,
      "step": 1982
    },
    {
      "epoch": 188.85714285714286,
      "grad_norm": 5.392822742462158,
      "learning_rate": 3.714392803598201e-07,
      "loss": 1.1577,
      "step": 1983
    },
    {
      "epoch": 188.95238095238096,
      "grad_norm": 3.5019123554229736,
      "learning_rate": 3.7162668665667166e-07,
      "loss": 1.1604,
      "step": 1984
    },
    {
      "epoch": 188.95238095238096,
      "eval_accuracy": 0.7148744848257774,
      "eval_f1": 0.7746520580396802,
      "eval_loss": 0.5750032067298889,
      "eval_precision": 0.7135842880523732,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.760961139896373,
      "eval_runtime": 1.2975,
      "eval_samples_per_second": 2056.973,
      "eval_steps_per_second": 4.624,
      "step": 1984
    },
    {
      "epoch": 189.04761904761904,
      "grad_norm": 3.6215627193450928,
      "learning_rate": 3.718140929535232e-07,
      "loss": 1.195,
      "step": 1985
    },
    {
      "epoch": 189.14285714285714,
      "grad_norm": 5.709612846374512,
      "learning_rate": 3.720014992503748e-07,
      "loss": 1.1404,
      "step": 1986
    },
    {
      "epoch": 189.23809523809524,
      "grad_norm": 8.219498634338379,
      "learning_rate": 3.7218890554722643e-07,
      "loss": 1.1572,
      "step": 1987
    },
    {
      "epoch": 189.33333333333334,
      "grad_norm": 6.0207061767578125,
      "learning_rate": 3.72376311844078e-07,
      "loss": 1.1864,
      "step": 1988
    },
    {
      "epoch": 189.42857142857142,
      "grad_norm": 7.10655403137207,
      "learning_rate": 3.725637181409296e-07,
      "loss": 1.23,
      "step": 1989
    },
    {
      "epoch": 189.52380952380952,
      "grad_norm": 3.2644829750061035,
      "learning_rate": 3.7275112443778114e-07,
      "loss": 1.147,
      "step": 1990
    },
    {
      "epoch": 189.61904761904762,
      "grad_norm": 4.245572090148926,
      "learning_rate": 3.729385307346327e-07,
      "loss": 1.137,
      "step": 1991
    },
    {
      "epoch": 189.71428571428572,
      "grad_norm": 4.594895362854004,
      "learning_rate": 3.7312593703148425e-07,
      "loss": 1.1319,
      "step": 1992
    },
    {
      "epoch": 189.8095238095238,
      "grad_norm": 3.6682186126708984,
      "learning_rate": 3.7331334332833586e-07,
      "loss": 1.162,
      "step": 1993
    },
    {
      "epoch": 189.9047619047619,
      "grad_norm": 3.349700689315796,
      "learning_rate": 3.735007496251874e-07,
      "loss": 1.1492,
      "step": 1994
    },
    {
      "epoch": 190.0,
      "grad_norm": 4.069940090179443,
      "learning_rate": 3.7368815592203897e-07,
      "loss": 1.1501,
      "step": 1995
    },
    {
      "epoch": 190.0,
      "eval_accuracy": 0.7118771075309105,
      "eval_f1": 0.7724178751109796,
      "eval_loss": 0.5751973390579224,
      "eval_precision": 0.7111716621253406,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7608937823834199,
      "eval_runtime": 1.1744,
      "eval_samples_per_second": 2272.561,
      "eval_steps_per_second": 5.109,
      "step": 1995
    },
    {
      "epoch": 190.0952380952381,
      "grad_norm": 7.315029621124268,
      "learning_rate": 3.738755622188906e-07,
      "loss": 1.1588,
      "step": 1996
    },
    {
      "epoch": 190.1904761904762,
      "grad_norm": 3.7769927978515625,
      "learning_rate": 3.7406296851574213e-07,
      "loss": 1.1753,
      "step": 1997
    },
    {
      "epoch": 190.28571428571428,
      "grad_norm": 5.542825698852539,
      "learning_rate": 3.7425037481259374e-07,
      "loss": 1.1572,
      "step": 1998
    },
    {
      "epoch": 190.38095238095238,
      "grad_norm": 4.059653282165527,
      "learning_rate": 3.744377811094453e-07,
      "loss": 1.1302,
      "step": 1999
    },
    {
      "epoch": 190.47619047619048,
      "grad_norm": 6.358081340789795,
      "learning_rate": 3.746251874062969e-07,
      "loss": 1.1556,
      "step": 2000
    },
    {
      "epoch": 190.57142857142858,
      "grad_norm": 3.3779191970825195,
      "learning_rate": 3.7481259370314846e-07,
      "loss": 1.181,
      "step": 2001
    },
    {
      "epoch": 190.66666666666666,
      "grad_norm": 2.771360158920288,
      "learning_rate": 3.75e-07,
      "loss": 1.1661,
      "step": 2002
    },
    {
      "epoch": 190.76190476190476,
      "grad_norm": 4.921844482421875,
      "learning_rate": 3.751874062968516e-07,
      "loss": 1.1727,
      "step": 2003
    },
    {
      "epoch": 190.85714285714286,
      "grad_norm": 3.3652825355529785,
      "learning_rate": 3.7537481259370317e-07,
      "loss": 1.1663,
      "step": 2004
    },
    {
      "epoch": 190.95238095238096,
      "grad_norm": 7.287751197814941,
      "learning_rate": 3.755622188905547e-07,
      "loss": 1.1232,
      "step": 2005
    },
    {
      "epoch": 190.95238095238096,
      "eval_accuracy": 0.7137504683402023,
      "eval_f1": 0.7746312684365781,
      "eval_loss": 0.5749582052230835,
      "eval_precision": 0.7112676056338029,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7615135290731145,
      "eval_runtime": 1.1582,
      "eval_samples_per_second": 2304.46,
      "eval_steps_per_second": 5.181,
      "step": 2005
    },
    {
      "epoch": 191.04761904761904,
      "grad_norm": 3.0846879482269287,
      "learning_rate": 3.757496251874063e-07,
      "loss": 1.1737,
      "step": 2006
    },
    {
      "epoch": 191.14285714285714,
      "grad_norm": 2.902621030807495,
      "learning_rate": 3.759370314842579e-07,
      "loss": 1.1728,
      "step": 2007
    },
    {
      "epoch": 191.23809523809524,
      "grad_norm": 3.8469674587249756,
      "learning_rate": 3.7612443778110944e-07,
      "loss": 1.1821,
      "step": 2008
    },
    {
      "epoch": 191.33333333333334,
      "grad_norm": 4.116859436035156,
      "learning_rate": 3.7631184407796105e-07,
      "loss": 1.1174,
      "step": 2009
    },
    {
      "epoch": 191.42857142857142,
      "grad_norm": 7.507976055145264,
      "learning_rate": 3.764992503748126e-07,
      "loss": 1.2025,
      "step": 2010
    },
    {
      "epoch": 191.52380952380952,
      "grad_norm": 6.259975433349609,
      "learning_rate": 3.766866566716642e-07,
      "loss": 1.1568,
      "step": 2011
    },
    {
      "epoch": 191.61904761904762,
      "grad_norm": 5.180998802185059,
      "learning_rate": 3.7687406296851577e-07,
      "loss": 1.1603,
      "step": 2012
    },
    {
      "epoch": 191.71428571428572,
      "grad_norm": 3.2812700271606445,
      "learning_rate": 3.770614692653673e-07,
      "loss": 1.157,
      "step": 2013
    },
    {
      "epoch": 191.8095238095238,
      "grad_norm": 4.7818379402160645,
      "learning_rate": 3.7724887556221893e-07,
      "loss": 1.1923,
      "step": 2014
    },
    {
      "epoch": 191.9047619047619,
      "grad_norm": 8.186214447021484,
      "learning_rate": 3.774362818590705e-07,
      "loss": 1.1419,
      "step": 2015
    },
    {
      "epoch": 192.0,
      "grad_norm": 6.474642753601074,
      "learning_rate": 3.7762368815592204e-07,
      "loss": 1.1409,
      "step": 2016
    },
    {
      "epoch": 192.0,
      "eval_accuracy": 0.7137504683402023,
      "eval_f1": 0.7719402985074627,
      "eval_loss": 0.5745519995689392,
      "eval_precision": 0.7159468438538206,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7616039147956246,
      "eval_runtime": 1.1466,
      "eval_samples_per_second": 2327.785,
      "eval_steps_per_second": 5.233,
      "step": 2016
    },
    {
      "epoch": 192.0952380952381,
      "grad_norm": 6.402791976928711,
      "learning_rate": 3.7781109445277364e-07,
      "loss": 1.195,
      "step": 2017
    },
    {
      "epoch": 192.1904761904762,
      "grad_norm": 4.038552284240723,
      "learning_rate": 3.779985007496252e-07,
      "loss": 1.1443,
      "step": 2018
    },
    {
      "epoch": 192.28571428571428,
      "grad_norm": 3.914491653442383,
      "learning_rate": 3.7818590704647675e-07,
      "loss": 1.1565,
      "step": 2019
    },
    {
      "epoch": 192.38095238095238,
      "grad_norm": 4.010745525360107,
      "learning_rate": 3.783733133433283e-07,
      "loss": 1.1719,
      "step": 2020
    },
    {
      "epoch": 192.47619047619048,
      "grad_norm": 6.152637958526611,
      "learning_rate": 3.785607196401799e-07,
      "loss": 1.1406,
      "step": 2021
    },
    {
      "epoch": 192.57142857142858,
      "grad_norm": 5.661224842071533,
      "learning_rate": 3.787481259370315e-07,
      "loss": 1.1677,
      "step": 2022
    },
    {
      "epoch": 192.66666666666666,
      "grad_norm": 3.715280294418335,
      "learning_rate": 3.789355322338831e-07,
      "loss": 1.1605,
      "step": 2023
    },
    {
      "epoch": 192.76190476190476,
      "grad_norm": 4.92529296875,
      "learning_rate": 3.791229385307347e-07,
      "loss": 1.1522,
      "step": 2024
    },
    {
      "epoch": 192.85714285714286,
      "grad_norm": 4.248322486877441,
      "learning_rate": 3.7931034482758624e-07,
      "loss": 1.1564,
      "step": 2025
    },
    {
      "epoch": 192.95238095238096,
      "grad_norm": 4.95744514465332,
      "learning_rate": 3.794977511244378e-07,
      "loss": 1.1798,
      "step": 2026
    },
    {
      "epoch": 192.95238095238096,
      "eval_accuracy": 0.7182465342825028,
      "eval_f1": 0.7779090372120496,
      "eval_loss": 0.5743523836135864,
      "eval_precision": 0.7149837133550488,
      "eval_recall": 0.852979274611399,
      "eval_roc_auc": 0.7618511801957397,
      "eval_runtime": 1.239,
      "eval_samples_per_second": 2154.151,
      "eval_steps_per_second": 4.843,
      "step": 2026
    },
    {
      "epoch": 193.04761904761904,
      "grad_norm": 3.4746222496032715,
      "learning_rate": 3.7968515742128935e-07,
      "loss": 1.1368,
      "step": 2027
    },
    {
      "epoch": 193.14285714285714,
      "grad_norm": 3.15362548828125,
      "learning_rate": 3.7987256371814096e-07,
      "loss": 1.1771,
      "step": 2028
    },
    {
      "epoch": 193.23809523809524,
      "grad_norm": 5.005309104919434,
      "learning_rate": 3.800599700149925e-07,
      "loss": 1.1872,
      "step": 2029
    },
    {
      "epoch": 193.33333333333334,
      "grad_norm": 3.0793089866638184,
      "learning_rate": 3.8024737631184406e-07,
      "loss": 1.128,
      "step": 2030
    },
    {
      "epoch": 193.42857142857142,
      "grad_norm": 2.29117751121521,
      "learning_rate": 3.8043478260869567e-07,
      "loss": 1.1459,
      "step": 2031
    },
    {
      "epoch": 193.52380952380952,
      "grad_norm": 4.069672107696533,
      "learning_rate": 3.806221889055472e-07,
      "loss": 1.1598,
      "step": 2032
    },
    {
      "epoch": 193.61904761904762,
      "grad_norm": 4.900834083557129,
      "learning_rate": 3.8080959520239883e-07,
      "loss": 1.1229,
      "step": 2033
    },
    {
      "epoch": 193.71428571428572,
      "grad_norm": 3.2819442749023438,
      "learning_rate": 3.809970014992504e-07,
      "loss": 1.1383,
      "step": 2034
    },
    {
      "epoch": 193.8095238095238,
      "grad_norm": 6.044792175292969,
      "learning_rate": 3.81184407796102e-07,
      "loss": 1.1592,
      "step": 2035
    },
    {
      "epoch": 193.9047619047619,
      "grad_norm": 3.2246363162994385,
      "learning_rate": 3.8137181409295355e-07,
      "loss": 1.1529,
      "step": 2036
    },
    {
      "epoch": 194.0,
      "grad_norm": 4.810468673706055,
      "learning_rate": 3.815592203898051e-07,
      "loss": 1.1537,
      "step": 2037
    },
    {
      "epoch": 194.0,
      "eval_accuracy": 0.7152491569876358,
      "eval_f1": 0.7755463673951565,
      "eval_loss": 0.5744901299476624,
      "eval_precision": 0.7128121606948968,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7617599309153713,
      "eval_runtime": 1.1555,
      "eval_samples_per_second": 2309.841,
      "eval_steps_per_second": 5.193,
      "step": 2037
    },
    {
      "epoch": 194.0952380952381,
      "grad_norm": 3.0019352436065674,
      "learning_rate": 3.817466266866567e-07,
      "loss": 1.1464,
      "step": 2038
    },
    {
      "epoch": 194.1904761904762,
      "grad_norm": 6.727803707122803,
      "learning_rate": 3.8193403298350827e-07,
      "loss": 1.1953,
      "step": 2039
    },
    {
      "epoch": 194.28571428571428,
      "grad_norm": 6.521770000457764,
      "learning_rate": 3.821214392803598e-07,
      "loss": 1.1938,
      "step": 2040
    },
    {
      "epoch": 194.38095238095238,
      "grad_norm": 5.184866905212402,
      "learning_rate": 3.8230884557721143e-07,
      "loss": 1.1416,
      "step": 2041
    },
    {
      "epoch": 194.47619047619048,
      "grad_norm": 2.739201784133911,
      "learning_rate": 3.82496251874063e-07,
      "loss": 1.1729,
      "step": 2042
    },
    {
      "epoch": 194.57142857142858,
      "grad_norm": 4.125956058502197,
      "learning_rate": 3.8268365817091454e-07,
      "loss": 1.1504,
      "step": 2043
    },
    {
      "epoch": 194.66666666666666,
      "grad_norm": 7.283012390136719,
      "learning_rate": 3.8287106446776614e-07,
      "loss": 1.143,
      "step": 2044
    },
    {
      "epoch": 194.76190476190476,
      "grad_norm": 5.7457122802734375,
      "learning_rate": 3.830584707646177e-07,
      "loss": 1.1596,
      "step": 2045
    },
    {
      "epoch": 194.85714285714286,
      "grad_norm": 3.988215446472168,
      "learning_rate": 3.832458770614693e-07,
      "loss": 1.1372,
      "step": 2046
    },
    {
      "epoch": 194.95238095238096,
      "grad_norm": 3.1475110054016113,
      "learning_rate": 3.8343328335832086e-07,
      "loss": 1.1506,
      "step": 2047
    },
    {
      "epoch": 194.95238095238096,
      "eval_accuracy": 0.7156238291494942,
      "eval_f1": 0.774844259863542,
      "eval_loss": 0.5744240283966064,
      "eval_precision": 0.7148330596606459,
      "eval_recall": 0.8458549222797928,
      "eval_roc_auc": 0.7617547495682211,
      "eval_runtime": 1.1349,
      "eval_samples_per_second": 2351.667,
      "eval_steps_per_second": 5.287,
      "step": 2047
    },
    {
      "epoch": 195.04761904761904,
      "grad_norm": 3.0281283855438232,
      "learning_rate": 3.8362068965517247e-07,
      "loss": 1.1494,
      "step": 2048
    },
    {
      "epoch": 195.14285714285714,
      "grad_norm": 4.342933177947998,
      "learning_rate": 3.83808095952024e-07,
      "loss": 1.1916,
      "step": 2049
    },
    {
      "epoch": 195.23809523809524,
      "grad_norm": 3.2373242378234863,
      "learning_rate": 3.839955022488756e-07,
      "loss": 1.1375,
      "step": 2050
    },
    {
      "epoch": 195.33333333333334,
      "grad_norm": 3.446120262145996,
      "learning_rate": 3.8418290854572713e-07,
      "loss": 1.1064,
      "step": 2051
    },
    {
      "epoch": 195.42857142857142,
      "grad_norm": 2.7731194496154785,
      "learning_rate": 3.8437031484257874e-07,
      "loss": 1.1574,
      "step": 2052
    },
    {
      "epoch": 195.52380952380952,
      "grad_norm": 2.7649197578430176,
      "learning_rate": 3.845577211394303e-07,
      "loss": 1.1417,
      "step": 2053
    },
    {
      "epoch": 195.61904761904762,
      "grad_norm": 6.727789402008057,
      "learning_rate": 3.8474512743628185e-07,
      "loss": 1.1949,
      "step": 2054
    },
    {
      "epoch": 195.71428571428572,
      "grad_norm": 7.244948863983154,
      "learning_rate": 3.8493253373313346e-07,
      "loss": 1.1084,
      "step": 2055
    },
    {
      "epoch": 195.8095238095238,
      "grad_norm": 4.702371120452881,
      "learning_rate": 3.85119940029985e-07,
      "loss": 1.1538,
      "step": 2056
    },
    {
      "epoch": 195.9047619047619,
      "grad_norm": 7.4241557121276855,
      "learning_rate": 3.853073463268366e-07,
      "loss": 1.1966,
      "step": 2057
    },
    {
      "epoch": 196.0,
      "grad_norm": 3.8166346549987793,
      "learning_rate": 3.8549475262368817e-07,
      "loss": 1.1731,
      "step": 2058
    },
    {
      "epoch": 196.0,
      "eval_accuracy": 0.7182465342825028,
      "eval_f1": 0.7768545994065282,
      "eval_loss": 0.5741073489189148,
      "eval_precision": 0.7168674698795181,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7620805987334485,
      "eval_runtime": 1.2068,
      "eval_samples_per_second": 2211.58,
      "eval_steps_per_second": 4.972,
      "step": 2058
    },
    {
      "epoch": 196.0952380952381,
      "grad_norm": 3.794862985610962,
      "learning_rate": 3.856821589205398e-07,
      "loss": 1.1682,
      "step": 2059
    },
    {
      "epoch": 196.1904761904762,
      "grad_norm": 4.032334804534912,
      "learning_rate": 3.8586956521739133e-07,
      "loss": 1.1684,
      "step": 2060
    },
    {
      "epoch": 196.28571428571428,
      "grad_norm": 2.493825674057007,
      "learning_rate": 3.860569715142429e-07,
      "loss": 1.1872,
      "step": 2061
    },
    {
      "epoch": 196.38095238095238,
      "grad_norm": 6.8107171058654785,
      "learning_rate": 3.862443778110945e-07,
      "loss": 1.1476,
      "step": 2062
    },
    {
      "epoch": 196.47619047619048,
      "grad_norm": 5.340165615081787,
      "learning_rate": 3.8643178410794605e-07,
      "loss": 1.1115,
      "step": 2063
    },
    {
      "epoch": 196.57142857142858,
      "grad_norm": 8.906362533569336,
      "learning_rate": 3.866191904047976e-07,
      "loss": 1.1214,
      "step": 2064
    },
    {
      "epoch": 196.66666666666666,
      "grad_norm": 5.633921146392822,
      "learning_rate": 3.8680659670164916e-07,
      "loss": 1.1858,
      "step": 2065
    },
    {
      "epoch": 196.76190476190476,
      "grad_norm": 3.9064207077026367,
      "learning_rate": 3.8699400299850077e-07,
      "loss": 1.1631,
      "step": 2066
    },
    {
      "epoch": 196.85714285714286,
      "grad_norm": 3.3096764087677,
      "learning_rate": 3.871814092953523e-07,
      "loss": 1.1868,
      "step": 2067
    },
    {
      "epoch": 196.95238095238096,
      "grad_norm": 6.722177982330322,
      "learning_rate": 3.8736881559220393e-07,
      "loss": 1.1433,
      "step": 2068
    },
    {
      "epoch": 196.95238095238096,
      "eval_accuracy": 0.7171225177969277,
      "eval_f1": 0.7756315007429421,
      "eval_loss": 0.573854386806488,
      "eval_precision": 0.7166392092257001,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7624821531375935,
      "eval_runtime": 1.3412,
      "eval_samples_per_second": 1990.066,
      "eval_steps_per_second": 4.474,
      "step": 2068
    },
    {
      "epoch": 197.04761904761904,
      "grad_norm": 4.195196628570557,
      "learning_rate": 3.8755622188905554e-07,
      "loss": 1.1479,
      "step": 2069
    },
    {
      "epoch": 197.14285714285714,
      "grad_norm": 9.604462623596191,
      "learning_rate": 3.877436281859071e-07,
      "loss": 1.173,
      "step": 2070
    },
    {
      "epoch": 197.23809523809524,
      "grad_norm": 3.1296794414520264,
      "learning_rate": 3.8793103448275865e-07,
      "loss": 1.1208,
      "step": 2071
    },
    {
      "epoch": 197.33333333333334,
      "grad_norm": 4.265172004699707,
      "learning_rate": 3.881184407796102e-07,
      "loss": 1.1395,
      "step": 2072
    },
    {
      "epoch": 197.42857142857142,
      "grad_norm": 4.306982040405273,
      "learning_rate": 3.883058470764618e-07,
      "loss": 1.207,
      "step": 2073
    },
    {
      "epoch": 197.52380952380952,
      "grad_norm": 2.888141632080078,
      "learning_rate": 3.8849325337331336e-07,
      "loss": 1.1417,
      "step": 2074
    },
    {
      "epoch": 197.61904761904762,
      "grad_norm": 4.370743274688721,
      "learning_rate": 3.886806596701649e-07,
      "loss": 1.1533,
      "step": 2075
    },
    {
      "epoch": 197.71428571428572,
      "grad_norm": 8.061007499694824,
      "learning_rate": 3.888680659670165e-07,
      "loss": 1.1919,
      "step": 2076
    },
    {
      "epoch": 197.8095238095238,
      "grad_norm": 3.11177659034729,
      "learning_rate": 3.890554722638681e-07,
      "loss": 1.1521,
      "step": 2077
    },
    {
      "epoch": 197.9047619047619,
      "grad_norm": 3.6492860317230225,
      "learning_rate": 3.8924287856071963e-07,
      "loss": 1.1524,
      "step": 2078
    },
    {
      "epoch": 198.0,
      "grad_norm": 5.121402740478516,
      "learning_rate": 3.8943028485757124e-07,
      "loss": 1.1617,
      "step": 2079
    },
    {
      "epoch": 198.0,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7763938315539739,
      "eval_loss": 0.5734204053878784,
      "eval_precision": 0.7160831509846827,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.763346862406448,
      "eval_runtime": 1.1368,
      "eval_samples_per_second": 2347.743,
      "eval_steps_per_second": 5.278,
      "step": 2079
    },
    {
      "epoch": 198.0952380952381,
      "grad_norm": 5.118983745574951,
      "learning_rate": 3.896176911544228e-07,
      "loss": 1.1499,
      "step": 2080
    },
    {
      "epoch": 198.1904761904762,
      "grad_norm": 5.3729400634765625,
      "learning_rate": 3.898050974512744e-07,
      "loss": 1.1532,
      "step": 2081
    },
    {
      "epoch": 198.28571428571428,
      "grad_norm": 3.7796311378479004,
      "learning_rate": 3.8999250374812596e-07,
      "loss": 1.1294,
      "step": 2082
    },
    {
      "epoch": 198.38095238095238,
      "grad_norm": 2.8317065238952637,
      "learning_rate": 3.9017991004497756e-07,
      "loss": 1.153,
      "step": 2083
    },
    {
      "epoch": 198.47619047619048,
      "grad_norm": 6.204058647155762,
      "learning_rate": 3.903673163418291e-07,
      "loss": 1.175,
      "step": 2084
    },
    {
      "epoch": 198.57142857142858,
      "grad_norm": 4.600681781768799,
      "learning_rate": 3.9055472263868067e-07,
      "loss": 1.1812,
      "step": 2085
    },
    {
      "epoch": 198.66666666666666,
      "grad_norm": 3.409691095352173,
      "learning_rate": 3.9074212893553223e-07,
      "loss": 1.1537,
      "step": 2086
    },
    {
      "epoch": 198.76190476190476,
      "grad_norm": 5.143957138061523,
      "learning_rate": 3.909295352323838e-07,
      "loss": 1.1778,
      "step": 2087
    },
    {
      "epoch": 198.85714285714286,
      "grad_norm": 6.1059393882751465,
      "learning_rate": 3.9111694152923544e-07,
      "loss": 1.1434,
      "step": 2088
    },
    {
      "epoch": 198.95238095238096,
      "grad_norm": 5.029159069061279,
      "learning_rate": 3.91304347826087e-07,
      "loss": 1.1449,
      "step": 2089
    },
    {
      "epoch": 198.95238095238096,
      "eval_accuracy": 0.7178718621206445,
      "eval_f1": 0.7772848269742679,
      "eval_loss": 0.5733014345169067,
      "eval_precision": 0.7152966793685357,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.7635958549222799,
      "eval_runtime": 1.22,
      "eval_samples_per_second": 2187.731,
      "eval_steps_per_second": 4.918,
      "step": 2089
    },
    {
      "epoch": 199.04761904761904,
      "grad_norm": 4.662865161895752,
      "learning_rate": 3.9149175412293855e-07,
      "loss": 1.1717,
      "step": 2090
    },
    {
      "epoch": 199.14285714285714,
      "grad_norm": 4.18621826171875,
      "learning_rate": 3.916791604197901e-07,
      "loss": 1.2011,
      "step": 2091
    },
    {
      "epoch": 199.23809523809524,
      "grad_norm": 4.175061225891113,
      "learning_rate": 3.918665667166417e-07,
      "loss": 1.1323,
      "step": 2092
    },
    {
      "epoch": 199.33333333333334,
      "grad_norm": 4.9943928718566895,
      "learning_rate": 3.9205397301349327e-07,
      "loss": 1.1413,
      "step": 2093
    },
    {
      "epoch": 199.42857142857142,
      "grad_norm": 4.654736042022705,
      "learning_rate": 3.922413793103448e-07,
      "loss": 1.1301,
      "step": 2094
    },
    {
      "epoch": 199.52380952380952,
      "grad_norm": 4.7878098487854,
      "learning_rate": 3.924287856071965e-07,
      "loss": 1.1916,
      "step": 2095
    },
    {
      "epoch": 199.61904761904762,
      "grad_norm": 4.815639495849609,
      "learning_rate": 3.9261619190404804e-07,
      "loss": 1.2155,
      "step": 2096
    },
    {
      "epoch": 199.71428571428572,
      "grad_norm": 4.483081817626953,
      "learning_rate": 3.928035982008996e-07,
      "loss": 1.1616,
      "step": 2097
    },
    {
      "epoch": 199.8095238095238,
      "grad_norm": 3.760834217071533,
      "learning_rate": 3.9299100449775115e-07,
      "loss": 1.1406,
      "step": 2098
    },
    {
      "epoch": 199.9047619047619,
      "grad_norm": 4.239285945892334,
      "learning_rate": 3.931784107946027e-07,
      "loss": 1.1427,
      "step": 2099
    },
    {
      "epoch": 200.0,
      "grad_norm": 3.784797430038452,
      "learning_rate": 3.9336581709145425e-07,
      "loss": 1.1143,
      "step": 2100
    },
    {
      "epoch": 200.0,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7781047675103002,
      "eval_loss": 0.5735935568809509,
      "eval_precision": 0.7130528586839266,
      "eval_recall": 0.8562176165803109,
      "eval_roc_auc": 0.7635814622913067,
      "eval_runtime": 1.1863,
      "eval_samples_per_second": 2249.924,
      "eval_steps_per_second": 5.058,
      "step": 2100
    },
    {
      "epoch": 200.0952380952381,
      "grad_norm": 4.269444942474365,
      "learning_rate": 3.935532233883058e-07,
      "loss": 1.1332,
      "step": 2101
    },
    {
      "epoch": 200.1904761904762,
      "grad_norm": 9.206201553344727,
      "learning_rate": 3.9374062968515747e-07,
      "loss": 1.1922,
      "step": 2102
    },
    {
      "epoch": 200.28571428571428,
      "grad_norm": 4.745060443878174,
      "learning_rate": 3.93928035982009e-07,
      "loss": 1.0933,
      "step": 2103
    },
    {
      "epoch": 200.38095238095238,
      "grad_norm": 2.75970458984375,
      "learning_rate": 3.9411544227886063e-07,
      "loss": 1.209,
      "step": 2104
    },
    {
      "epoch": 200.47619047619048,
      "grad_norm": 7.039388179779053,
      "learning_rate": 3.943028485757122e-07,
      "loss": 1.1453,
      "step": 2105
    },
    {
      "epoch": 200.57142857142858,
      "grad_norm": 2.3920702934265137,
      "learning_rate": 3.9449025487256374e-07,
      "loss": 1.1373,
      "step": 2106
    },
    {
      "epoch": 200.66666666666666,
      "grad_norm": 3.3662586212158203,
      "learning_rate": 3.946776611694153e-07,
      "loss": 1.1773,
      "step": 2107
    },
    {
      "epoch": 200.76190476190476,
      "grad_norm": 3.275402784347534,
      "learning_rate": 3.9486506746626685e-07,
      "loss": 1.1561,
      "step": 2108
    },
    {
      "epoch": 200.85714285714286,
      "grad_norm": 2.953057289123535,
      "learning_rate": 3.950524737631185e-07,
      "loss": 1.1738,
      "step": 2109
    },
    {
      "epoch": 200.95238095238096,
      "grad_norm": 2.5332555770874023,
      "learning_rate": 3.9523988005997006e-07,
      "loss": 1.1239,
      "step": 2110
    },
    {
      "epoch": 200.95238095238096,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7766587677725119,
      "eval_loss": 0.573513925075531,
      "eval_precision": 0.7156113537117904,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7633212435233161,
      "eval_runtime": 1.2693,
      "eval_samples_per_second": 2102.79,
      "eval_steps_per_second": 4.727,
      "step": 2110
    },
    {
      "epoch": 201.04761904761904,
      "grad_norm": 3.6361196041107178,
      "learning_rate": 3.954272863568216e-07,
      "loss": 1.1166,
      "step": 2111
    },
    {
      "epoch": 201.14285714285714,
      "grad_norm": 1.9977929592132568,
      "learning_rate": 3.9561469265367317e-07,
      "loss": 1.1666,
      "step": 2112
    },
    {
      "epoch": 201.23809523809524,
      "grad_norm": 3.9000375270843506,
      "learning_rate": 3.9580209895052473e-07,
      "loss": 1.1772,
      "step": 2113
    },
    {
      "epoch": 201.33333333333334,
      "grad_norm": 5.172115802764893,
      "learning_rate": 3.9598950524737633e-07,
      "loss": 1.1216,
      "step": 2114
    },
    {
      "epoch": 201.42857142857142,
      "grad_norm": 4.378371238708496,
      "learning_rate": 3.961769115442279e-07,
      "loss": 1.1377,
      "step": 2115
    },
    {
      "epoch": 201.52380952380952,
      "grad_norm": 3.344632863998413,
      "learning_rate": 3.963643178410795e-07,
      "loss": 1.149,
      "step": 2116
    },
    {
      "epoch": 201.61904761904762,
      "grad_norm": 6.246377944946289,
      "learning_rate": 3.965517241379311e-07,
      "loss": 1.1658,
      "step": 2117
    },
    {
      "epoch": 201.71428571428572,
      "grad_norm": 5.464715003967285,
      "learning_rate": 3.9673913043478266e-07,
      "loss": 1.1611,
      "step": 2118
    },
    {
      "epoch": 201.8095238095238,
      "grad_norm": 8.447710990905762,
      "learning_rate": 3.969265367316342e-07,
      "loss": 1.1469,
      "step": 2119
    },
    {
      "epoch": 201.9047619047619,
      "grad_norm": 5.849113464355469,
      "learning_rate": 3.9711394302848577e-07,
      "loss": 1.1664,
      "step": 2120
    },
    {
      "epoch": 202.0,
      "grad_norm": 2.994105339050293,
      "learning_rate": 3.973013493253373e-07,
      "loss": 1.1456,
      "step": 2121
    },
    {
      "epoch": 202.0,
      "eval_accuracy": 0.7178718621206445,
      "eval_f1": 0.7755588673621461,
      "eval_loss": 0.573491096496582,
      "eval_precision": 0.718387631143015,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7629887737478411,
      "eval_runtime": 1.1257,
      "eval_samples_per_second": 2370.919,
      "eval_steps_per_second": 5.33,
      "step": 2121
    },
    {
      "epoch": 202.0952380952381,
      "grad_norm": 9.809347152709961,
      "learning_rate": 3.974887556221889e-07,
      "loss": 1.1572,
      "step": 2122
    },
    {
      "epoch": 202.1904761904762,
      "grad_norm": 3.824449300765991,
      "learning_rate": 3.9767616191904054e-07,
      "loss": 1.1721,
      "step": 2123
    },
    {
      "epoch": 202.28571428571428,
      "grad_norm": 3.7709100246429443,
      "learning_rate": 3.978635682158921e-07,
      "loss": 1.1496,
      "step": 2124
    },
    {
      "epoch": 202.38095238095238,
      "grad_norm": 8.865727424621582,
      "learning_rate": 3.9805097451274365e-07,
      "loss": 1.1732,
      "step": 2125
    },
    {
      "epoch": 202.47619047619048,
      "grad_norm": 6.136362552642822,
      "learning_rate": 3.982383808095952e-07,
      "loss": 1.1341,
      "step": 2126
    },
    {
      "epoch": 202.57142857142858,
      "grad_norm": 4.306957244873047,
      "learning_rate": 3.984257871064468e-07,
      "loss": 1.1121,
      "step": 2127
    },
    {
      "epoch": 202.66666666666666,
      "grad_norm": 4.527181148529053,
      "learning_rate": 3.9861319340329836e-07,
      "loss": 1.1743,
      "step": 2128
    },
    {
      "epoch": 202.76190476190476,
      "grad_norm": 5.983310222625732,
      "learning_rate": 3.988005997001499e-07,
      "loss": 1.1754,
      "step": 2129
    },
    {
      "epoch": 202.85714285714286,
      "grad_norm": 4.577535629272461,
      "learning_rate": 3.989880059970016e-07,
      "loss": 1.1163,
      "step": 2130
    },
    {
      "epoch": 202.95238095238096,
      "grad_norm": 5.112128257751465,
      "learning_rate": 3.9917541229385313e-07,
      "loss": 1.1802,
      "step": 2131
    },
    {
      "epoch": 202.95238095238096,
      "eval_accuracy": 0.7178718621206445,
      "eval_f1": 0.7783338239623196,
      "eval_loss": 0.5732585191726685,
      "eval_precision": 0.7134376686454398,
      "eval_recall": 0.8562176165803109,
      "eval_roc_auc": 0.7635374208405297,
      "eval_runtime": 1.2737,
      "eval_samples_per_second": 2095.482,
      "eval_steps_per_second": 4.711,
      "step": 2131
    },
    {
      "epoch": 203.04761904761904,
      "grad_norm": 5.311213970184326,
      "learning_rate": 3.993628185907047e-07,
      "loss": 1.1193,
      "step": 2132
    },
    {
      "epoch": 203.14285714285714,
      "grad_norm": 4.955393314361572,
      "learning_rate": 3.9955022488755624e-07,
      "loss": 1.1466,
      "step": 2133
    },
    {
      "epoch": 203.23809523809524,
      "grad_norm": 4.441088676452637,
      "learning_rate": 3.997376311844078e-07,
      "loss": 1.1374,
      "step": 2134
    },
    {
      "epoch": 203.33333333333334,
      "grad_norm": 7.389732837677002,
      "learning_rate": 3.9992503748125935e-07,
      "loss": 1.1618,
      "step": 2135
    },
    {
      "epoch": 203.42857142857142,
      "grad_norm": 4.27824592590332,
      "learning_rate": 4.001124437781109e-07,
      "loss": 1.1846,
      "step": 2136
    },
    {
      "epoch": 203.52380952380952,
      "grad_norm": 3.7284393310546875,
      "learning_rate": 4.0029985007496256e-07,
      "loss": 1.1396,
      "step": 2137
    },
    {
      "epoch": 203.61904761904762,
      "grad_norm": 3.8060035705566406,
      "learning_rate": 4.004872563718141e-07,
      "loss": 1.1884,
      "step": 2138
    },
    {
      "epoch": 203.71428571428572,
      "grad_norm": 5.725419521331787,
      "learning_rate": 4.006746626686657e-07,
      "loss": 1.1493,
      "step": 2139
    },
    {
      "epoch": 203.8095238095238,
      "grad_norm": 5.268664836883545,
      "learning_rate": 4.008620689655173e-07,
      "loss": 1.1579,
      "step": 2140
    },
    {
      "epoch": 203.9047619047619,
      "grad_norm": 2.8941152095794678,
      "learning_rate": 4.0104947526236884e-07,
      "loss": 1.1393,
      "step": 2141
    },
    {
      "epoch": 204.0,
      "grad_norm": 8.16684627532959,
      "learning_rate": 4.012368815592204e-07,
      "loss": 1.1049,
      "step": 2142
    },
    {
      "epoch": 204.0,
      "eval_accuracy": 0.7182465342825028,
      "eval_f1": 0.7764565992865636,
      "eval_loss": 0.5726208090782166,
      "eval_precision": 0.7175824175824176,
      "eval_recall": 0.8458549222797928,
      "eval_roc_auc": 0.7639162348877375,
      "eval_runtime": 1.2147,
      "eval_samples_per_second": 2197.252,
      "eval_steps_per_second": 4.939,
      "step": 2142
    },
    {
      "epoch": 204.0952380952381,
      "grad_norm": 6.156484603881836,
      "learning_rate": 4.0142428785607194e-07,
      "loss": 1.2162,
      "step": 2143
    },
    {
      "epoch": 204.1904761904762,
      "grad_norm": 4.6590681076049805,
      "learning_rate": 4.016116941529236e-07,
      "loss": 1.1895,
      "step": 2144
    },
    {
      "epoch": 204.28571428571428,
      "grad_norm": 4.852943420410156,
      "learning_rate": 4.0179910044977516e-07,
      "loss": 1.1787,
      "step": 2145
    },
    {
      "epoch": 204.38095238095238,
      "grad_norm": 5.074846267700195,
      "learning_rate": 4.019865067466267e-07,
      "loss": 1.1044,
      "step": 2146
    },
    {
      "epoch": 204.47619047619048,
      "grad_norm": 5.913023471832275,
      "learning_rate": 4.0217391304347827e-07,
      "loss": 1.1562,
      "step": 2147
    },
    {
      "epoch": 204.57142857142858,
      "grad_norm": 4.1200971603393555,
      "learning_rate": 4.023613193403298e-07,
      "loss": 1.1657,
      "step": 2148
    },
    {
      "epoch": 204.66666666666666,
      "grad_norm": 6.686506271362305,
      "learning_rate": 4.0254872563718143e-07,
      "loss": 1.1429,
      "step": 2149
    },
    {
      "epoch": 204.76190476190476,
      "grad_norm": 4.119016647338867,
      "learning_rate": 4.02736131934033e-07,
      "loss": 1.1531,
      "step": 2150
    },
    {
      "epoch": 204.85714285714286,
      "grad_norm": 6.244527339935303,
      "learning_rate": 4.029235382308846e-07,
      "loss": 1.1425,
      "step": 2151
    },
    {
      "epoch": 204.95238095238096,
      "grad_norm": 9.093226432800293,
      "learning_rate": 4.031109445277362e-07,
      "loss": 1.1276,
      "step": 2152
    },
    {
      "epoch": 204.95238095238096,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7767910005920663,
      "eval_loss": 0.5725477933883667,
      "eval_precision": 0.7153762268266085,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.76413298791019,
      "eval_runtime": 1.1696,
      "eval_samples_per_second": 2282.001,
      "eval_steps_per_second": 5.13,
      "step": 2152
    },
    {
      "epoch": 205.04761904761904,
      "grad_norm": 5.017244815826416,
      "learning_rate": 4.0329835082458775e-07,
      "loss": 1.1546,
      "step": 2153
    },
    {
      "epoch": 205.14285714285714,
      "grad_norm": 8.460247039794922,
      "learning_rate": 4.034857571214393e-07,
      "loss": 1.1757,
      "step": 2154
    },
    {
      "epoch": 205.23809523809524,
      "grad_norm": 4.2112908363342285,
      "learning_rate": 4.0367316341829086e-07,
      "loss": 1.1581,
      "step": 2155
    },
    {
      "epoch": 205.33333333333334,
      "grad_norm": 6.3861236572265625,
      "learning_rate": 4.038605697151424e-07,
      "loss": 1.1644,
      "step": 2156
    },
    {
      "epoch": 205.42857142857142,
      "grad_norm": 3.080798387527466,
      "learning_rate": 4.0404797601199397e-07,
      "loss": 1.1655,
      "step": 2157
    },
    {
      "epoch": 205.52380952380952,
      "grad_norm": 2.398893356323242,
      "learning_rate": 4.0423538230884563e-07,
      "loss": 1.162,
      "step": 2158
    },
    {
      "epoch": 205.61904761904762,
      "grad_norm": 3.430669069290161,
      "learning_rate": 4.044227886056972e-07,
      "loss": 1.1521,
      "step": 2159
    },
    {
      "epoch": 205.71428571428572,
      "grad_norm": 7.12416934967041,
      "learning_rate": 4.0461019490254874e-07,
      "loss": 1.1749,
      "step": 2160
    },
    {
      "epoch": 205.8095238095238,
      "grad_norm": 5.870452404022217,
      "learning_rate": 4.047976011994003e-07,
      "loss": 1.1205,
      "step": 2161
    },
    {
      "epoch": 205.9047619047619,
      "grad_norm": 4.407711982727051,
      "learning_rate": 4.049850074962519e-07,
      "loss": 1.1705,
      "step": 2162
    },
    {
      "epoch": 206.0,
      "grad_norm": 9.45398235321045,
      "learning_rate": 4.0517241379310346e-07,
      "loss": 1.0999,
      "step": 2163
    },
    {
      "epoch": 206.0,
      "eval_accuracy": 0.716373173473211,
      "eval_f1": 0.7738273080370481,
      "eval_loss": 0.5725975036621094,
      "eval_precision": 0.7182473655019412,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.7642725964306275,
      "eval_runtime": 1.1717,
      "eval_samples_per_second": 2277.879,
      "eval_steps_per_second": 5.121,
      "step": 2163
    },
    {
      "epoch": 206.0952380952381,
      "grad_norm": 3.9617202281951904,
      "learning_rate": 4.05359820089955e-07,
      "loss": 1.1539,
      "step": 2164
    },
    {
      "epoch": 206.1904761904762,
      "grad_norm": 4.726264476776123,
      "learning_rate": 4.0554722638680667e-07,
      "loss": 1.1628,
      "step": 2165
    },
    {
      "epoch": 206.28571428571428,
      "grad_norm": 4.4896697998046875,
      "learning_rate": 4.0573463268365823e-07,
      "loss": 1.1192,
      "step": 2166
    },
    {
      "epoch": 206.38095238095238,
      "grad_norm": 7.291873931884766,
      "learning_rate": 4.059220389805098e-07,
      "loss": 1.1721,
      "step": 2167
    },
    {
      "epoch": 206.47619047619048,
      "grad_norm": 7.347695827484131,
      "learning_rate": 4.0610944527736134e-07,
      "loss": 1.0977,
      "step": 2168
    },
    {
      "epoch": 206.57142857142858,
      "grad_norm": 6.2318572998046875,
      "learning_rate": 4.062968515742129e-07,
      "loss": 1.1896,
      "step": 2169
    },
    {
      "epoch": 206.66666666666666,
      "grad_norm": 12.115691184997559,
      "learning_rate": 4.0648425787106444e-07,
      "loss": 1.1579,
      "step": 2170
    },
    {
      "epoch": 206.76190476190476,
      "grad_norm": 3.2223641872406006,
      "learning_rate": 4.0667166416791605e-07,
      "loss": 1.1495,
      "step": 2171
    },
    {
      "epoch": 206.85714285714286,
      "grad_norm": 8.225313186645508,
      "learning_rate": 4.0685907046476766e-07,
      "loss": 1.1384,
      "step": 2172
    },
    {
      "epoch": 206.95238095238096,
      "grad_norm": 11.196160316467285,
      "learning_rate": 4.070464767616192e-07,
      "loss": 1.1882,
      "step": 2173
    },
    {
      "epoch": 206.95238095238096,
      "eval_accuracy": 0.7178718621206445,
      "eval_f1": 0.7780725022104332,
      "eval_loss": 0.5724543333053589,
      "eval_precision": 0.7138994050838291,
      "eval_recall": 0.8549222797927462,
      "eval_roc_auc": 0.7647561888313184,
      "eval_runtime": 1.166,
      "eval_samples_per_second": 2288.928,
      "eval_steps_per_second": 5.146,
      "step": 2173
    },
    {
      "epoch": 207.04761904761904,
      "grad_norm": 4.099627494812012,
      "learning_rate": 4.072338830584708e-07,
      "loss": 1.1492,
      "step": 2174
    },
    {
      "epoch": 207.14285714285714,
      "grad_norm": 7.178406238555908,
      "learning_rate": 4.074212893553224e-07,
      "loss": 1.1689,
      "step": 2175
    },
    {
      "epoch": 207.23809523809524,
      "grad_norm": 3.8665335178375244,
      "learning_rate": 4.0760869565217393e-07,
      "loss": 1.1435,
      "step": 2176
    },
    {
      "epoch": 207.33333333333334,
      "grad_norm": 5.064719200134277,
      "learning_rate": 4.077961019490255e-07,
      "loss": 1.1111,
      "step": 2177
    },
    {
      "epoch": 207.42857142857142,
      "grad_norm": 6.1857147216796875,
      "learning_rate": 4.0798350824587704e-07,
      "loss": 1.1411,
      "step": 2178
    },
    {
      "epoch": 207.52380952380952,
      "grad_norm": 6.367858409881592,
      "learning_rate": 4.081709145427287e-07,
      "loss": 1.1286,
      "step": 2179
    },
    {
      "epoch": 207.61904761904762,
      "grad_norm": 12.050115585327148,
      "learning_rate": 4.0835832083958025e-07,
      "loss": 1.2575,
      "step": 2180
    },
    {
      "epoch": 207.71428571428572,
      "grad_norm": 8.755186080932617,
      "learning_rate": 4.085457271364318e-07,
      "loss": 1.1639,
      "step": 2181
    },
    {
      "epoch": 207.8095238095238,
      "grad_norm": 5.138454914093018,
      "learning_rate": 4.0873313343328336e-07,
      "loss": 1.1696,
      "step": 2182
    },
    {
      "epoch": 207.9047619047619,
      "grad_norm": 5.218094348907471,
      "learning_rate": 4.089205397301349e-07,
      "loss": 1.1381,
      "step": 2183
    },
    {
      "epoch": 208.0,
      "grad_norm": 8.238438606262207,
      "learning_rate": 4.091079460269865e-07,
      "loss": 1.1741,
      "step": 2184
    },
    {
      "epoch": 208.0,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7766587677725119,
      "eval_loss": 0.5723700523376465,
      "eval_precision": 0.7156113537117904,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7646459412780657,
      "eval_runtime": 1.3691,
      "eval_samples_per_second": 1949.417,
      "eval_steps_per_second": 4.382,
      "step": 2184
    },
    {
      "epoch": 208.0952380952381,
      "grad_norm": 4.4855804443359375,
      "learning_rate": 4.092953523238381e-07,
      "loss": 1.18,
      "step": 2185
    },
    {
      "epoch": 208.1904761904762,
      "grad_norm": 4.409901142120361,
      "learning_rate": 4.094827586206897e-07,
      "loss": 1.1369,
      "step": 2186
    },
    {
      "epoch": 208.28571428571428,
      "grad_norm": 3.015300750732422,
      "learning_rate": 4.096701649175413e-07,
      "loss": 1.1921,
      "step": 2187
    },
    {
      "epoch": 208.38095238095238,
      "grad_norm": 5.867238521575928,
      "learning_rate": 4.0985757121439285e-07,
      "loss": 1.1966,
      "step": 2188
    },
    {
      "epoch": 208.47619047619048,
      "grad_norm": 9.443415641784668,
      "learning_rate": 4.100449775112444e-07,
      "loss": 1.1324,
      "step": 2189
    },
    {
      "epoch": 208.57142857142858,
      "grad_norm": 5.5111589431762695,
      "learning_rate": 4.1023238380809596e-07,
      "loss": 1.1466,
      "step": 2190
    },
    {
      "epoch": 208.66666666666666,
      "grad_norm": 10.333208084106445,
      "learning_rate": 4.104197901049475e-07,
      "loss": 1.1346,
      "step": 2191
    },
    {
      "epoch": 208.76190476190476,
      "grad_norm": 7.5379462242126465,
      "learning_rate": 4.1060719640179907e-07,
      "loss": 1.1363,
      "step": 2192
    },
    {
      "epoch": 208.85714285714286,
      "grad_norm": 3.4411845207214355,
      "learning_rate": 4.1079460269865073e-07,
      "loss": 1.136,
      "step": 2193
    },
    {
      "epoch": 208.95238095238096,
      "grad_norm": 10.017407417297363,
      "learning_rate": 4.109820089955023e-07,
      "loss": 1.1699,
      "step": 2194
    },
    {
      "epoch": 208.95238095238096,
      "eval_accuracy": 0.716373173473211,
      "eval_f1": 0.7768936044798114,
      "eval_loss": 0.5722417235374451,
      "eval_precision": 0.7128177393185505,
      "eval_recall": 0.8536269430051814,
      "eval_roc_auc": 0.7651404720782958,
      "eval_runtime": 1.1633,
      "eval_samples_per_second": 2294.263,
      "eval_steps_per_second": 5.158,
      "step": 2194
    },
    {
      "epoch": 209.04761904761904,
      "grad_norm": 8.25883960723877,
      "learning_rate": 4.1116941529235384e-07,
      "loss": 1.1607,
      "step": 2195
    },
    {
      "epoch": 209.14285714285714,
      "grad_norm": 4.923328876495361,
      "learning_rate": 4.1135682158920544e-07,
      "loss": 1.1483,
      "step": 2196
    },
    {
      "epoch": 209.23809523809524,
      "grad_norm": 3.406498670578003,
      "learning_rate": 4.11544227886057e-07,
      "loss": 1.1324,
      "step": 2197
    },
    {
      "epoch": 209.33333333333334,
      "grad_norm": 3.716740131378174,
      "learning_rate": 4.1173163418290855e-07,
      "loss": 1.1369,
      "step": 2198
    },
    {
      "epoch": 209.42857142857142,
      "grad_norm": 10.365269660949707,
      "learning_rate": 4.119190404797601e-07,
      "loss": 1.1487,
      "step": 2199
    },
    {
      "epoch": 209.52380952380952,
      "grad_norm": 2.718311309814453,
      "learning_rate": 4.1210644677661177e-07,
      "loss": 1.1306,
      "step": 2200
    },
    {
      "epoch": 209.61904761904762,
      "grad_norm": 9.89725112915039,
      "learning_rate": 4.122938530734633e-07,
      "loss": 1.1405,
      "step": 2201
    },
    {
      "epoch": 209.71428571428572,
      "grad_norm": 3.5114660263061523,
      "learning_rate": 4.124812593703149e-07,
      "loss": 1.1697,
      "step": 2202
    },
    {
      "epoch": 209.8095238095238,
      "grad_norm": 10.346630096435547,
      "learning_rate": 4.1266866566716643e-07,
      "loss": 1.2071,
      "step": 2203
    },
    {
      "epoch": 209.9047619047619,
      "grad_norm": 9.676673889160156,
      "learning_rate": 4.12856071964018e-07,
      "loss": 1.2243,
      "step": 2204
    },
    {
      "epoch": 210.0,
      "grad_norm": 7.612486839294434,
      "learning_rate": 4.1304347826086954e-07,
      "loss": 1.1319,
      "step": 2205
    },
    {
      "epoch": 210.0,
      "eval_accuracy": 0.7171225177969277,
      "eval_f1": 0.7761636525348354,
      "eval_loss": 0.5721729397773743,
      "eval_precision": 0.7156916347731,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7652625215889464,
      "eval_runtime": 1.1999,
      "eval_samples_per_second": 2224.387,
      "eval_steps_per_second": 5.0,
      "step": 2205
    },
    {
      "epoch": 210.0952380952381,
      "grad_norm": 3.142979383468628,
      "learning_rate": 4.1323088455772115e-07,
      "loss": 1.1685,
      "step": 2206
    },
    {
      "epoch": 210.1904761904762,
      "grad_norm": 4.474756240844727,
      "learning_rate": 4.1341829085457275e-07,
      "loss": 1.1643,
      "step": 2207
    },
    {
      "epoch": 210.28571428571428,
      "grad_norm": 3.771099805831909,
      "learning_rate": 4.136056971514243e-07,
      "loss": 1.1463,
      "step": 2208
    },
    {
      "epoch": 210.38095238095238,
      "grad_norm": 7.966793060302734,
      "learning_rate": 4.137931034482759e-07,
      "loss": 1.0928,
      "step": 2209
    },
    {
      "epoch": 210.47619047619048,
      "grad_norm": 3.9173672199249268,
      "learning_rate": 4.1398050974512747e-07,
      "loss": 1.2006,
      "step": 2210
    },
    {
      "epoch": 210.57142857142858,
      "grad_norm": 4.2888875007629395,
      "learning_rate": 4.14167916041979e-07,
      "loss": 1.1902,
      "step": 2211
    },
    {
      "epoch": 210.66666666666666,
      "grad_norm": 5.334024906158447,
      "learning_rate": 4.143553223388306e-07,
      "loss": 1.1327,
      "step": 2212
    },
    {
      "epoch": 210.76190476190476,
      "grad_norm": 3.1423180103302,
      "learning_rate": 4.1454272863568213e-07,
      "loss": 1.1423,
      "step": 2213
    },
    {
      "epoch": 210.85714285714286,
      "grad_norm": 3.1354405879974365,
      "learning_rate": 4.147301349325338e-07,
      "loss": 1.1604,
      "step": 2214
    },
    {
      "epoch": 210.95238095238096,
      "grad_norm": 5.210052490234375,
      "learning_rate": 4.1491754122938535e-07,
      "loss": 1.1864,
      "step": 2215
    },
    {
      "epoch": 210.95238095238096,
      "eval_accuracy": 0.7178718621206445,
      "eval_f1": 0.776092774308653,
      "eval_loss": 0.5724236965179443,
      "eval_precision": 0.7174271577789995,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7650028785261946,
      "eval_runtime": 1.1778,
      "eval_samples_per_second": 2266.118,
      "eval_steps_per_second": 5.094,
      "step": 2215
    },
    {
      "epoch": 211.04761904761904,
      "grad_norm": 7.045891761779785,
      "learning_rate": 4.151049475262369e-07,
      "loss": 1.1503,
      "step": 2216
    },
    {
      "epoch": 211.14285714285714,
      "grad_norm": 3.449984550476074,
      "learning_rate": 4.1529235382308846e-07,
      "loss": 1.1661,
      "step": 2217
    },
    {
      "epoch": 211.23809523809524,
      "grad_norm": 4.770975589752197,
      "learning_rate": 4.1547976011994e-07,
      "loss": 1.205,
      "step": 2218
    },
    {
      "epoch": 211.33333333333334,
      "grad_norm": 6.622992515563965,
      "learning_rate": 4.156671664167916e-07,
      "loss": 1.0991,
      "step": 2219
    },
    {
      "epoch": 211.42857142857142,
      "grad_norm": 5.019271373748779,
      "learning_rate": 4.158545727136432e-07,
      "loss": 1.1856,
      "step": 2220
    },
    {
      "epoch": 211.52380952380952,
      "grad_norm": 3.0830953121185303,
      "learning_rate": 4.160419790104948e-07,
      "loss": 1.1469,
      "step": 2221
    },
    {
      "epoch": 211.61904761904762,
      "grad_norm": 3.6828019618988037,
      "learning_rate": 4.162293853073464e-07,
      "loss": 1.161,
      "step": 2222
    },
    {
      "epoch": 211.71428571428572,
      "grad_norm": 3.6262388229370117,
      "learning_rate": 4.1641679160419794e-07,
      "loss": 1.1006,
      "step": 2223
    },
    {
      "epoch": 211.8095238095238,
      "grad_norm": 4.846446990966797,
      "learning_rate": 4.166041979010495e-07,
      "loss": 1.1494,
      "step": 2224
    },
    {
      "epoch": 211.9047619047619,
      "grad_norm": 3.820887804031372,
      "learning_rate": 4.1679160419790105e-07,
      "loss": 1.151,
      "step": 2225
    },
    {
      "epoch": 212.0,
      "grad_norm": 3.8886525630950928,
      "learning_rate": 4.169790104947526e-07,
      "loss": 1.1847,
      "step": 2226
    },
    {
      "epoch": 212.0,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7777122641509434,
      "eval_loss": 0.5727176666259766,
      "eval_precision": 0.7137445887445888,
      "eval_recall": 0.8542746113989638,
      "eval_roc_auc": 0.7648865860679332,
      "eval_runtime": 1.2503,
      "eval_samples_per_second": 2134.72,
      "eval_steps_per_second": 4.799,
      "step": 2226
    },
    {
      "epoch": 212.0952380952381,
      "grad_norm": 8.59073543548584,
      "learning_rate": 4.1716641679160416e-07,
      "loss": 1.1629,
      "step": 2227
    },
    {
      "epoch": 212.1904761904762,
      "grad_norm": 2.888920783996582,
      "learning_rate": 4.173538230884558e-07,
      "loss": 1.127,
      "step": 2228
    },
    {
      "epoch": 212.28571428571428,
      "grad_norm": 5.506545066833496,
      "learning_rate": 4.175412293853074e-07,
      "loss": 1.1072,
      "step": 2229
    },
    {
      "epoch": 212.38095238095238,
      "grad_norm": 4.807139873504639,
      "learning_rate": 4.1772863568215893e-07,
      "loss": 1.1414,
      "step": 2230
    },
    {
      "epoch": 212.47619047619048,
      "grad_norm": 6.0634965896606445,
      "learning_rate": 4.1791604197901054e-07,
      "loss": 1.1252,
      "step": 2231
    },
    {
      "epoch": 212.57142857142858,
      "grad_norm": 4.370108127593994,
      "learning_rate": 4.181034482758621e-07,
      "loss": 1.2013,
      "step": 2232
    },
    {
      "epoch": 212.66666666666666,
      "grad_norm": 4.020362377166748,
      "learning_rate": 4.1829085457271365e-07,
      "loss": 1.1788,
      "step": 2233
    },
    {
      "epoch": 212.76190476190476,
      "grad_norm": 4.844730854034424,
      "learning_rate": 4.184782608695652e-07,
      "loss": 1.1873,
      "step": 2234
    },
    {
      "epoch": 212.85714285714286,
      "grad_norm": 6.405012607574463,
      "learning_rate": 4.1866566716641686e-07,
      "loss": 1.1471,
      "step": 2235
    },
    {
      "epoch": 212.95238095238096,
      "grad_norm": 7.924243927001953,
      "learning_rate": 4.188530734632684e-07,
      "loss": 1.1486,
      "step": 2236
    },
    {
      "epoch": 212.95238095238096,
      "eval_accuracy": 0.7178718621206445,
      "eval_f1": 0.776624147137348,
      "eval_loss": 0.5724982023239136,
      "eval_precision": 0.7164750957854407,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7647668393782383,
      "eval_runtime": 1.1569,
      "eval_samples_per_second": 2306.976,
      "eval_steps_per_second": 5.186,
      "step": 2236
    },
    {
      "epoch": 213.04761904761904,
      "grad_norm": 7.97761344909668,
      "learning_rate": 4.1904047976011997e-07,
      "loss": 1.1542,
      "step": 2237
    },
    {
      "epoch": 213.14285714285714,
      "grad_norm": 2.720165252685547,
      "learning_rate": 4.192278860569715e-07,
      "loss": 1.1536,
      "step": 2238
    },
    {
      "epoch": 213.23809523809524,
      "grad_norm": 2.4307806491851807,
      "learning_rate": 4.194152923538231e-07,
      "loss": 1.1604,
      "step": 2239
    },
    {
      "epoch": 213.33333333333334,
      "grad_norm": 10.833077430725098,
      "learning_rate": 4.1960269865067463e-07,
      "loss": 1.1664,
      "step": 2240
    },
    {
      "epoch": 213.42857142857142,
      "grad_norm": 6.367500305175781,
      "learning_rate": 4.1979010494752624e-07,
      "loss": 1.2025,
      "step": 2241
    },
    {
      "epoch": 213.52380952380952,
      "grad_norm": 2.9368419647216797,
      "learning_rate": 4.1997751124437785e-07,
      "loss": 1.1574,
      "step": 2242
    },
    {
      "epoch": 213.61904761904762,
      "grad_norm": 4.824771881103516,
      "learning_rate": 4.201649175412294e-07,
      "loss": 1.1246,
      "step": 2243
    },
    {
      "epoch": 213.71428571428572,
      "grad_norm": 7.3083696365356445,
      "learning_rate": 4.20352323838081e-07,
      "loss": 1.1425,
      "step": 2244
    },
    {
      "epoch": 213.8095238095238,
      "grad_norm": 4.5281453132629395,
      "learning_rate": 4.2053973013493257e-07,
      "loss": 1.1185,
      "step": 2245
    },
    {
      "epoch": 213.9047619047619,
      "grad_norm": 5.061934947967529,
      "learning_rate": 4.207271364317841e-07,
      "loss": 1.1074,
      "step": 2246
    },
    {
      "epoch": 214.0,
      "grad_norm": 5.4383673667907715,
      "learning_rate": 4.209145427286357e-07,
      "loss": 1.1617,
      "step": 2247
    },
    {
      "epoch": 214.0,
      "eval_accuracy": 0.7171225177969277,
      "eval_f1": 0.775897892549718,
      "eval_loss": 0.5720090270042419,
      "eval_precision": 0.7161643835616438,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7653572251007483,
      "eval_runtime": 1.1624,
      "eval_samples_per_second": 2296.067,
      "eval_steps_per_second": 5.162,
      "step": 2247
    },
    {
      "epoch": 214.0952380952381,
      "grad_norm": 3.1440367698669434,
      "learning_rate": 4.2110194902548723e-07,
      "loss": 1.1204,
      "step": 2248
    },
    {
      "epoch": 214.1904761904762,
      "grad_norm": 3.43729567527771,
      "learning_rate": 4.212893553223389e-07,
      "loss": 1.1387,
      "step": 2249
    },
    {
      "epoch": 214.28571428571428,
      "grad_norm": 6.375138282775879,
      "learning_rate": 4.2147676161919044e-07,
      "loss": 1.1185,
      "step": 2250
    },
    {
      "epoch": 214.38095238095238,
      "grad_norm": 6.120760917663574,
      "learning_rate": 4.21664167916042e-07,
      "loss": 1.1553,
      "step": 2251
    },
    {
      "epoch": 214.47619047619048,
      "grad_norm": 8.098929405212402,
      "learning_rate": 4.2185157421289355e-07,
      "loss": 1.1866,
      "step": 2252
    },
    {
      "epoch": 214.57142857142858,
      "grad_norm": 5.230090141296387,
      "learning_rate": 4.220389805097451e-07,
      "loss": 1.2084,
      "step": 2253
    },
    {
      "epoch": 214.66666666666666,
      "grad_norm": 5.517597198486328,
      "learning_rate": 4.222263868065967e-07,
      "loss": 1.1486,
      "step": 2254
    },
    {
      "epoch": 214.76190476190476,
      "grad_norm": 7.162447452545166,
      "learning_rate": 4.2241379310344827e-07,
      "loss": 1.1461,
      "step": 2255
    },
    {
      "epoch": 214.85714285714286,
      "grad_norm": 4.703765392303467,
      "learning_rate": 4.2260119940029993e-07,
      "loss": 1.1885,
      "step": 2256
    },
    {
      "epoch": 214.95238095238096,
      "grad_norm": 3.9778876304626465,
      "learning_rate": 4.227886056971515e-07,
      "loss": 1.1438,
      "step": 2257
    },
    {
      "epoch": 214.95238095238096,
      "eval_accuracy": 0.7182465342825028,
      "eval_f1": 0.7777777777777778,
      "eval_loss": 0.57195645570755,
      "eval_precision": 0.7152173913043478,
      "eval_recall": 0.8523316062176166,
      "eval_roc_auc": 0.7654749568221071,
      "eval_runtime": 1.2576,
      "eval_samples_per_second": 2122.364,
      "eval_steps_per_second": 4.771,
      "step": 2257
    },
    {
      "epoch": 215.04761904761904,
      "grad_norm": 4.2935051918029785,
      "learning_rate": 4.2297601199400304e-07,
      "loss": 1.1877,
      "step": 2258
    },
    {
      "epoch": 215.14285714285714,
      "grad_norm": 7.638199806213379,
      "learning_rate": 4.231634182908546e-07,
      "loss": 1.116,
      "step": 2259
    },
    {
      "epoch": 215.23809523809524,
      "grad_norm": 4.469337463378906,
      "learning_rate": 4.2335082458770615e-07,
      "loss": 1.169,
      "step": 2260
    },
    {
      "epoch": 215.33333333333334,
      "grad_norm": 4.260531902313232,
      "learning_rate": 4.235382308845577e-07,
      "loss": 1.1045,
      "step": 2261
    },
    {
      "epoch": 215.42857142857142,
      "grad_norm": 5.3834638595581055,
      "learning_rate": 4.2372563718140926e-07,
      "loss": 1.0919,
      "step": 2262
    },
    {
      "epoch": 215.52380952380952,
      "grad_norm": 7.487053871154785,
      "learning_rate": 4.239130434782609e-07,
      "loss": 1.1892,
      "step": 2263
    },
    {
      "epoch": 215.61904761904762,
      "grad_norm": 8.624944686889648,
      "learning_rate": 4.2410044977511247e-07,
      "loss": 1.152,
      "step": 2264
    },
    {
      "epoch": 215.71428571428572,
      "grad_norm": 4.9617438316345215,
      "learning_rate": 4.24287856071964e-07,
      "loss": 1.1661,
      "step": 2265
    },
    {
      "epoch": 215.8095238095238,
      "grad_norm": 4.929933071136475,
      "learning_rate": 4.2447526236881563e-07,
      "loss": 1.1726,
      "step": 2266
    },
    {
      "epoch": 215.9047619047619,
      "grad_norm": 3.8353214263916016,
      "learning_rate": 4.246626686656672e-07,
      "loss": 1.1477,
      "step": 2267
    },
    {
      "epoch": 216.0,
      "grad_norm": 3.108581304550171,
      "learning_rate": 4.2485007496251874e-07,
      "loss": 1.184,
      "step": 2268
    },
    {
      "epoch": 216.0,
      "eval_accuracy": 0.716373173473211,
      "eval_f1": 0.7761017450458444,
      "eval_loss": 0.5715641975402832,
      "eval_precision": 0.7142079477408819,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7657878526194588,
      "eval_runtime": 1.2948,
      "eval_samples_per_second": 2061.335,
      "eval_steps_per_second": 4.634,
      "step": 2268
    },
    {
      "epoch": 216.0952380952381,
      "grad_norm": 6.062177658081055,
      "learning_rate": 4.250374812593703e-07,
      "loss": 1.1456,
      "step": 2269
    },
    {
      "epoch": 216.1904761904762,
      "grad_norm": 6.099572658538818,
      "learning_rate": 4.2522488755622196e-07,
      "loss": 1.1492,
      "step": 2270
    },
    {
      "epoch": 216.28571428571428,
      "grad_norm": 7.1180291175842285,
      "learning_rate": 4.254122938530735e-07,
      "loss": 1.1307,
      "step": 2271
    },
    {
      "epoch": 216.38095238095238,
      "grad_norm": 3.0259740352630615,
      "learning_rate": 4.2559970014992507e-07,
      "loss": 1.1654,
      "step": 2272
    },
    {
      "epoch": 216.47619047619048,
      "grad_norm": 4.901364803314209,
      "learning_rate": 4.257871064467766e-07,
      "loss": 1.1515,
      "step": 2273
    },
    {
      "epoch": 216.57142857142858,
      "grad_norm": 4.846545696258545,
      "learning_rate": 4.259745127436282e-07,
      "loss": 1.1758,
      "step": 2274
    },
    {
      "epoch": 216.66666666666666,
      "grad_norm": 8.551472663879395,
      "learning_rate": 4.2616191904047973e-07,
      "loss": 1.1208,
      "step": 2275
    },
    {
      "epoch": 216.76190476190476,
      "grad_norm": 4.555139541625977,
      "learning_rate": 4.2634932533733134e-07,
      "loss": 1.1495,
      "step": 2276
    },
    {
      "epoch": 216.85714285714286,
      "grad_norm": 10.477103233337402,
      "learning_rate": 4.2653673163418294e-07,
      "loss": 1.1901,
      "step": 2277
    },
    {
      "epoch": 216.95238095238096,
      "grad_norm": 8.654388427734375,
      "learning_rate": 4.267241379310345e-07,
      "loss": 1.1739,
      "step": 2278
    },
    {
      "epoch": 216.95238095238096,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7754615842763549,
      "eval_loss": 0.5713890790939331,
      "eval_precision": 0.7177508269018743,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7659024179620034,
      "eval_runtime": 1.1422,
      "eval_samples_per_second": 2336.711,
      "eval_steps_per_second": 5.253,
      "step": 2278
    },
    {
      "epoch": 217.04761904761904,
      "grad_norm": 14.373431205749512,
      "learning_rate": 4.269115442278861e-07,
      "loss": 1.1315,
      "step": 2279
    },
    {
      "epoch": 217.14285714285714,
      "grad_norm": 6.774090766906738,
      "learning_rate": 4.2709895052473766e-07,
      "loss": 1.1362,
      "step": 2280
    },
    {
      "epoch": 217.23809523809524,
      "grad_norm": 4.8597259521484375,
      "learning_rate": 4.272863568215892e-07,
      "loss": 1.1534,
      "step": 2281
    },
    {
      "epoch": 217.33333333333334,
      "grad_norm": 6.150089740753174,
      "learning_rate": 4.2747376311844077e-07,
      "loss": 1.1144,
      "step": 2282
    },
    {
      "epoch": 217.42857142857142,
      "grad_norm": 10.673328399658203,
      "learning_rate": 4.276611694152923e-07,
      "loss": 1.1573,
      "step": 2283
    },
    {
      "epoch": 217.52380952380952,
      "grad_norm": 6.223097324371338,
      "learning_rate": 4.27848575712144e-07,
      "loss": 1.1323,
      "step": 2284
    },
    {
      "epoch": 217.61904761904762,
      "grad_norm": 9.125777244567871,
      "learning_rate": 4.2803598200899554e-07,
      "loss": 1.1752,
      "step": 2285
    },
    {
      "epoch": 217.71428571428572,
      "grad_norm": 4.172426700592041,
      "learning_rate": 4.282233883058471e-07,
      "loss": 1.1602,
      "step": 2286
    },
    {
      "epoch": 217.8095238095238,
      "grad_norm": 4.502418041229248,
      "learning_rate": 4.2841079460269865e-07,
      "loss": 1.1482,
      "step": 2287
    },
    {
      "epoch": 217.9047619047619,
      "grad_norm": 13.205572128295898,
      "learning_rate": 4.285982008995502e-07,
      "loss": 1.1785,
      "step": 2288
    },
    {
      "epoch": 218.0,
      "grad_norm": 7.098892688751221,
      "learning_rate": 4.287856071964018e-07,
      "loss": 1.1286,
      "step": 2289
    },
    {
      "epoch": 218.0,
      "eval_accuracy": 0.7182465342825028,
      "eval_f1": 0.7760571768910065,
      "eval_loss": 0.5713024139404297,
      "eval_precision": 0.7183020948180816,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7658497409326425,
      "eval_runtime": 1.1913,
      "eval_samples_per_second": 2240.34,
      "eval_steps_per_second": 5.036,
      "step": 2289
    },
    {
      "epoch": 218.0952380952381,
      "grad_norm": 2.8342089653015137,
      "learning_rate": 4.2897301349325336e-07,
      "loss": 1.1516,
      "step": 2290
    },
    {
      "epoch": 218.1904761904762,
      "grad_norm": 8.66518783569336,
      "learning_rate": 4.29160419790105e-07,
      "loss": 1.1909,
      "step": 2291
    },
    {
      "epoch": 218.28571428571428,
      "grad_norm": 5.65387487411499,
      "learning_rate": 4.293478260869566e-07,
      "loss": 1.1667,
      "step": 2292
    },
    {
      "epoch": 218.38095238095238,
      "grad_norm": 2.9471826553344727,
      "learning_rate": 4.2953523238380813e-07,
      "loss": 1.1703,
      "step": 2293
    },
    {
      "epoch": 218.47619047619048,
      "grad_norm": 5.311910629272461,
      "learning_rate": 4.297226386806597e-07,
      "loss": 1.1337,
      "step": 2294
    },
    {
      "epoch": 218.57142857142858,
      "grad_norm": 3.465955972671509,
      "learning_rate": 4.2991004497751124e-07,
      "loss": 1.1064,
      "step": 2295
    },
    {
      "epoch": 218.66666666666666,
      "grad_norm": 3.5405027866363525,
      "learning_rate": 4.300974512743628e-07,
      "loss": 1.1456,
      "step": 2296
    },
    {
      "epoch": 218.76190476190476,
      "grad_norm": 4.713162899017334,
      "learning_rate": 4.3028485757121435e-07,
      "loss": 1.1944,
      "step": 2297
    },
    {
      "epoch": 218.85714285714286,
      "grad_norm": 3.6023991107940674,
      "learning_rate": 4.30472263868066e-07,
      "loss": 1.1501,
      "step": 2298
    },
    {
      "epoch": 218.95238095238096,
      "grad_norm": 4.573559761047363,
      "learning_rate": 4.3065967016491757e-07,
      "loss": 1.12,
      "step": 2299
    },
    {
      "epoch": 218.95238095238096,
      "eval_accuracy": 0.7186212064443612,
      "eval_f1": 0.7772174428952833,
      "eval_loss": 0.5710164904594421,
      "eval_precision": 0.7170224411603722,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7663013816925734,
      "eval_runtime": 1.2075,
      "eval_samples_per_second": 2210.389,
      "eval_steps_per_second": 4.969,
      "step": 2299
    },
    {
      "epoch": 219.04761904761904,
      "grad_norm": 6.432830810546875,
      "learning_rate": 4.308470764617691e-07,
      "loss": 1.1894,
      "step": 2300
    },
    {
      "epoch": 219.14285714285714,
      "grad_norm": 9.601896286010742,
      "learning_rate": 4.3103448275862073e-07,
      "loss": 1.1693,
      "step": 2301
    },
    {
      "epoch": 219.23809523809524,
      "grad_norm": 3.743082284927368,
      "learning_rate": 4.312218890554723e-07,
      "loss": 1.1238,
      "step": 2302
    },
    {
      "epoch": 219.33333333333334,
      "grad_norm": 4.7529168128967285,
      "learning_rate": 4.3140929535232384e-07,
      "loss": 1.1241,
      "step": 2303
    },
    {
      "epoch": 219.42857142857142,
      "grad_norm": 11.7463960647583,
      "learning_rate": 4.315967016491754e-07,
      "loss": 1.1282,
      "step": 2304
    },
    {
      "epoch": 219.52380952380952,
      "grad_norm": 4.231055736541748,
      "learning_rate": 4.3178410794602705e-07,
      "loss": 1.1524,
      "step": 2305
    },
    {
      "epoch": 219.61904761904762,
      "grad_norm": 3.1641948223114014,
      "learning_rate": 4.319715142428786e-07,
      "loss": 1.1592,
      "step": 2306
    },
    {
      "epoch": 219.71428571428572,
      "grad_norm": 3.844000816345215,
      "learning_rate": 4.3215892053973016e-07,
      "loss": 1.1645,
      "step": 2307
    },
    {
      "epoch": 219.8095238095238,
      "grad_norm": 10.697543144226074,
      "learning_rate": 4.323463268365817e-07,
      "loss": 1.1261,
      "step": 2308
    },
    {
      "epoch": 219.9047619047619,
      "grad_norm": 8.771400451660156,
      "learning_rate": 4.3253373313343327e-07,
      "loss": 1.1431,
      "step": 2309
    },
    {
      "epoch": 220.0,
      "grad_norm": 3.680692195892334,
      "learning_rate": 4.327211394302848e-07,
      "loss": 1.1709,
      "step": 2310
    },
    {
      "epoch": 220.0,
      "eval_accuracy": 0.7182465342825028,
      "eval_f1": 0.7761904761904762,
      "eval_loss": 0.5708158612251282,
      "eval_precision": 0.7180616740088106,
      "eval_recall": 0.844559585492228,
      "eval_roc_auc": 0.7665362694300518,
      "eval_runtime": 1.2926,
      "eval_samples_per_second": 2064.826,
      "eval_steps_per_second": 4.642,
      "step": 2310
    },
    {
      "epoch": 220.0952380952381,
      "grad_norm": 5.360668182373047,
      "learning_rate": 4.329085457271365e-07,
      "loss": 1.155,
      "step": 2311
    },
    {
      "epoch": 220.1904761904762,
      "grad_norm": 2.6076276302337646,
      "learning_rate": 4.3309595202398804e-07,
      "loss": 1.1352,
      "step": 2312
    },
    {
      "epoch": 220.28571428571428,
      "grad_norm": 2.816897392272949,
      "learning_rate": 4.332833583208396e-07,
      "loss": 1.1547,
      "step": 2313
    },
    {
      "epoch": 220.38095238095238,
      "grad_norm": 9.164076805114746,
      "learning_rate": 4.334707646176912e-07,
      "loss": 1.1507,
      "step": 2314
    },
    {
      "epoch": 220.47619047619048,
      "grad_norm": 9.430002212524414,
      "learning_rate": 4.3365817091454276e-07,
      "loss": 1.1124,
      "step": 2315
    },
    {
      "epoch": 220.57142857142858,
      "grad_norm": 8.547354698181152,
      "learning_rate": 4.338455772113943e-07,
      "loss": 1.1953,
      "step": 2316
    },
    {
      "epoch": 220.66666666666666,
      "grad_norm": 3.7488644123077393,
      "learning_rate": 4.3403298350824586e-07,
      "loss": 1.1528,
      "step": 2317
    },
    {
      "epoch": 220.76190476190476,
      "grad_norm": 3.1749069690704346,
      "learning_rate": 4.342203898050975e-07,
      "loss": 1.1311,
      "step": 2318
    },
    {
      "epoch": 220.85714285714286,
      "grad_norm": 3.318031072616577,
      "learning_rate": 4.344077961019491e-07,
      "loss": 1.1468,
      "step": 2319
    },
    {
      "epoch": 220.95238095238096,
      "grad_norm": 3.5835556983947754,
      "learning_rate": 4.3459520239880063e-07,
      "loss": 1.1506,
      "step": 2320
    },
    {
      "epoch": 220.95238095238096,
      "eval_accuracy": 0.7186212064443612,
      "eval_f1": 0.7772174428952833,
      "eval_loss": 0.5706951022148132,
      "eval_precision": 0.7170224411603722,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7666669545192862,
      "eval_runtime": 1.1476,
      "eval_samples_per_second": 2325.798,
      "eval_steps_per_second": 5.228,
      "step": 2320
    },
    {
      "epoch": 221.04761904761904,
      "grad_norm": 6.415576457977295,
      "learning_rate": 4.347826086956522e-07,
      "loss": 1.1645,
      "step": 2321
    },
    {
      "epoch": 221.14285714285714,
      "grad_norm": 6.643709659576416,
      "learning_rate": 4.3497001499250374e-07,
      "loss": 1.1505,
      "step": 2322
    },
    {
      "epoch": 221.23809523809524,
      "grad_norm": 4.2586822509765625,
      "learning_rate": 4.351574212893553e-07,
      "loss": 1.1772,
      "step": 2323
    },
    {
      "epoch": 221.33333333333334,
      "grad_norm": 4.354816913604736,
      "learning_rate": 4.353448275862069e-07,
      "loss": 1.1163,
      "step": 2324
    },
    {
      "epoch": 221.42857142857142,
      "grad_norm": 3.812659502029419,
      "learning_rate": 4.355322338830585e-07,
      "loss": 1.1893,
      "step": 2325
    },
    {
      "epoch": 221.52380952380952,
      "grad_norm": 2.639324188232422,
      "learning_rate": 4.357196401799101e-07,
      "loss": 1.1589,
      "step": 2326
    },
    {
      "epoch": 221.61904761904762,
      "grad_norm": 8.526213645935059,
      "learning_rate": 4.359070464767617e-07,
      "loss": 1.1101,
      "step": 2327
    },
    {
      "epoch": 221.71428571428572,
      "grad_norm": 8.683570861816406,
      "learning_rate": 4.3609445277361323e-07,
      "loss": 1.1543,
      "step": 2328
    },
    {
      "epoch": 221.8095238095238,
      "grad_norm": 4.670089244842529,
      "learning_rate": 4.362818590704648e-07,
      "loss": 1.1489,
      "step": 2329
    },
    {
      "epoch": 221.9047619047619,
      "grad_norm": 3.6284923553466797,
      "learning_rate": 4.3646926536731634e-07,
      "loss": 1.1382,
      "step": 2330
    },
    {
      "epoch": 222.0,
      "grad_norm": 6.06208610534668,
      "learning_rate": 4.366566716641679e-07,
      "loss": 1.1784,
      "step": 2331
    },
    {
      "epoch": 222.0,
      "eval_accuracy": 0.7186212064443612,
      "eval_f1": 0.7749475576865448,
      "eval_loss": 0.5703732371330261,
      "eval_precision": 0.7211377579475738,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7669130685089234,
      "eval_runtime": 1.4477,
      "eval_samples_per_second": 1843.605,
      "eval_steps_per_second": 4.144,
      "step": 2331
    },
    {
      "epoch": 222.0952380952381,
      "grad_norm": 5.371791362762451,
      "learning_rate": 4.3684407796101955e-07,
      "loss": 1.124,
      "step": 2332
    },
    {
      "epoch": 222.1904761904762,
      "grad_norm": 6.109635829925537,
      "learning_rate": 4.370314842578711e-07,
      "loss": 1.1417,
      "step": 2333
    },
    {
      "epoch": 222.28571428571428,
      "grad_norm": 5.1480560302734375,
      "learning_rate": 4.3721889055472266e-07,
      "loss": 1.1479,
      "step": 2334
    },
    {
      "epoch": 222.38095238095238,
      "grad_norm": 3.191652536392212,
      "learning_rate": 4.374062968515742e-07,
      "loss": 1.1171,
      "step": 2335
    },
    {
      "epoch": 222.47619047619048,
      "grad_norm": 7.5411272048950195,
      "learning_rate": 4.375937031484258e-07,
      "loss": 1.1868,
      "step": 2336
    },
    {
      "epoch": 222.57142857142858,
      "grad_norm": 4.1391401290893555,
      "learning_rate": 4.377811094452774e-07,
      "loss": 1.143,
      "step": 2337
    },
    {
      "epoch": 222.66666666666666,
      "grad_norm": 8.584721565246582,
      "learning_rate": 4.3796851574212893e-07,
      "loss": 1.1658,
      "step": 2338
    },
    {
      "epoch": 222.76190476190476,
      "grad_norm": 8.805205345153809,
      "learning_rate": 4.381559220389806e-07,
      "loss": 1.1905,
      "step": 2339
    },
    {
      "epoch": 222.85714285714286,
      "grad_norm": 4.957940101623535,
      "learning_rate": 4.3834332833583215e-07,
      "loss": 1.1784,
      "step": 2340
    },
    {
      "epoch": 222.95238095238096,
      "grad_norm": 9.811948776245117,
      "learning_rate": 4.385307346326837e-07,
      "loss": 1.1905,
      "step": 2341
    },
    {
      "epoch": 222.95238095238096,
      "eval_accuracy": 0.7189958786062196,
      "eval_f1": 0.775851763299462,
      "eval_loss": 0.570548951625824,
      "eval_precision": 0.7203107658157603,
      "eval_recall": 0.8406735751295337,
      "eval_roc_auc": 0.7670463442717328,
      "eval_runtime": 1.2238,
      "eval_samples_per_second": 2180.979,
      "eval_steps_per_second": 4.903,
      "step": 2341
    },
    {
      "epoch": 223.04761904761904,
      "grad_norm": 8.00674819946289,
      "learning_rate": 4.3871814092953526e-07,
      "loss": 1.1571,
      "step": 2342
    },
    {
      "epoch": 223.14285714285714,
      "grad_norm": 8.01071834564209,
      "learning_rate": 4.389055472263868e-07,
      "loss": 1.105,
      "step": 2343
    },
    {
      "epoch": 223.23809523809524,
      "grad_norm": 8.743189811706543,
      "learning_rate": 4.3909295352323837e-07,
      "loss": 1.1469,
      "step": 2344
    },
    {
      "epoch": 223.33333333333334,
      "grad_norm": 3.163163900375366,
      "learning_rate": 4.392803598200899e-07,
      "loss": 1.155,
      "step": 2345
    },
    {
      "epoch": 223.42857142857142,
      "grad_norm": 5.545464515686035,
      "learning_rate": 4.394677661169416e-07,
      "loss": 1.1703,
      "step": 2346
    },
    {
      "epoch": 223.52380952380952,
      "grad_norm": 6.351076126098633,
      "learning_rate": 4.3965517241379313e-07,
      "loss": 1.1415,
      "step": 2347
    },
    {
      "epoch": 223.61904761904762,
      "grad_norm": 4.026891231536865,
      "learning_rate": 4.398425787106447e-07,
      "loss": 1.1376,
      "step": 2348
    },
    {
      "epoch": 223.71428571428572,
      "grad_norm": 9.977435111999512,
      "learning_rate": 4.400299850074963e-07,
      "loss": 1.1836,
      "step": 2349
    },
    {
      "epoch": 223.8095238095238,
      "grad_norm": 3.645195722579956,
      "learning_rate": 4.4021739130434785e-07,
      "loss": 1.1816,
      "step": 2350
    },
    {
      "epoch": 223.9047619047619,
      "grad_norm": 3.4103715419769287,
      "learning_rate": 4.404047976011994e-07,
      "loss": 1.1381,
      "step": 2351
    },
    {
      "epoch": 224.0,
      "grad_norm": 3.77848744392395,
      "learning_rate": 4.4059220389805096e-07,
      "loss": 1.1195,
      "step": 2352
    },
    {
      "epoch": 224.0,
      "eval_accuracy": 0.7197452229299363,
      "eval_f1": 0.7783046828689982,
      "eval_loss": 0.5704892873764038,
      "eval_precision": 0.7174863387978142,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7676044905008634,
      "eval_runtime": 1.424,
      "eval_samples_per_second": 1874.259,
      "eval_steps_per_second": 4.213,
      "step": 2352
    },
    {
      "epoch": 224.0952380952381,
      "grad_norm": 4.599103927612305,
      "learning_rate": 4.407796101949026e-07,
      "loss": 1.1545,
      "step": 2353
    },
    {
      "epoch": 224.1904761904762,
      "grad_norm": 4.559905052185059,
      "learning_rate": 4.409670164917542e-07,
      "loss": 1.1492,
      "step": 2354
    },
    {
      "epoch": 224.28571428571428,
      "grad_norm": 4.591866493225098,
      "learning_rate": 4.4115442278860573e-07,
      "loss": 1.077,
      "step": 2355
    },
    {
      "epoch": 224.38095238095238,
      "grad_norm": 5.925033092498779,
      "learning_rate": 4.413418290854573e-07,
      "loss": 1.1578,
      "step": 2356
    },
    {
      "epoch": 224.47619047619048,
      "grad_norm": 9.684907913208008,
      "learning_rate": 4.4152923538230884e-07,
      "loss": 1.1807,
      "step": 2357
    },
    {
      "epoch": 224.57142857142858,
      "grad_norm": 4.006957530975342,
      "learning_rate": 4.4171664167916045e-07,
      "loss": 1.1713,
      "step": 2358
    },
    {
      "epoch": 224.66666666666666,
      "grad_norm": 5.357590198516846,
      "learning_rate": 4.41904047976012e-07,
      "loss": 1.1408,
      "step": 2359
    },
    {
      "epoch": 224.76190476190476,
      "grad_norm": 7.220717906951904,
      "learning_rate": 4.420914542728636e-07,
      "loss": 1.2085,
      "step": 2360
    },
    {
      "epoch": 224.85714285714286,
      "grad_norm": 3.5976977348327637,
      "learning_rate": 4.422788605697152e-07,
      "loss": 1.1425,
      "step": 2361
    },
    {
      "epoch": 224.95238095238096,
      "grad_norm": 6.680579662322998,
      "learning_rate": 4.4246626686656677e-07,
      "loss": 1.1477,
      "step": 2362
    },
    {
      "epoch": 224.95238095238096,
      "eval_accuracy": 0.7193705507680779,
      "eval_f1": 0.7772821885221528,
      "eval_loss": 0.5701494216918945,
      "eval_precision": 0.7185266630016492,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7679289004029937,
      "eval_runtime": 1.1883,
      "eval_samples_per_second": 2245.985,
      "eval_steps_per_second": 5.049,
      "step": 2362
    },
    {
      "epoch": 225.04761904761904,
      "grad_norm": 2.512918472290039,
      "learning_rate": 4.426536731634183e-07,
      "loss": 1.1161,
      "step": 2363
    },
    {
      "epoch": 225.14285714285714,
      "grad_norm": 4.432623863220215,
      "learning_rate": 4.428410794602699e-07,
      "loss": 1.141,
      "step": 2364
    },
    {
      "epoch": 225.23809523809524,
      "grad_norm": 2.8132100105285645,
      "learning_rate": 4.4302848575712143e-07,
      "loss": 1.1529,
      "step": 2365
    },
    {
      "epoch": 225.33333333333334,
      "grad_norm": 5.2681097984313965,
      "learning_rate": 4.43215892053973e-07,
      "loss": 1.1792,
      "step": 2366
    },
    {
      "epoch": 225.42857142857142,
      "grad_norm": 2.881300210952759,
      "learning_rate": 4.4340329835082465e-07,
      "loss": 1.1431,
      "step": 2367
    },
    {
      "epoch": 225.52380952380952,
      "grad_norm": 3.3978936672210693,
      "learning_rate": 4.435907046476762e-07,
      "loss": 1.1452,
      "step": 2368
    },
    {
      "epoch": 225.61904761904762,
      "grad_norm": 2.8875365257263184,
      "learning_rate": 4.4377811094452776e-07,
      "loss": 1.1367,
      "step": 2369
    },
    {
      "epoch": 225.71428571428572,
      "grad_norm": 5.989372730255127,
      "learning_rate": 4.439655172413793e-07,
      "loss": 1.1804,
      "step": 2370
    },
    {
      "epoch": 225.8095238095238,
      "grad_norm": 2.5179331302642822,
      "learning_rate": 4.441529235382309e-07,
      "loss": 1.1437,
      "step": 2371
    },
    {
      "epoch": 225.9047619047619,
      "grad_norm": 2.771777391433716,
      "learning_rate": 4.4434032983508247e-07,
      "loss": 1.132,
      "step": 2372
    },
    {
      "epoch": 226.0,
      "grad_norm": 3.8657586574554443,
      "learning_rate": 4.4452773613193403e-07,
      "loss": 1.1608,
      "step": 2373
    },
    {
      "epoch": 226.0,
      "eval_accuracy": 0.7171225177969277,
      "eval_f1": 0.7760308513794126,
      "eval_loss": 0.5697022080421448,
      "eval_precision": 0.715927750410509,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.768458549222798,
      "eval_runtime": 1.1924,
      "eval_samples_per_second": 2238.258,
      "eval_steps_per_second": 5.032,
      "step": 2373
    },
    {
      "epoch": 226.0952380952381,
      "grad_norm": 6.003801345825195,
      "learning_rate": 4.447151424287857e-07,
      "loss": 1.1063,
      "step": 2374
    },
    {
      "epoch": 226.1904761904762,
      "grad_norm": 4.66921329498291,
      "learning_rate": 4.4490254872563724e-07,
      "loss": 1.145,
      "step": 2375
    },
    {
      "epoch": 226.28571428571428,
      "grad_norm": 3.6423964500427246,
      "learning_rate": 4.450899550224888e-07,
      "loss": 1.0927,
      "step": 2376
    },
    {
      "epoch": 226.38095238095238,
      "grad_norm": 5.096931457519531,
      "learning_rate": 4.4527736131934035e-07,
      "loss": 1.1127,
      "step": 2377
    },
    {
      "epoch": 226.47619047619048,
      "grad_norm": 8.15682601928711,
      "learning_rate": 4.454647676161919e-07,
      "loss": 1.1469,
      "step": 2378
    },
    {
      "epoch": 226.57142857142858,
      "grad_norm": 2.7751176357269287,
      "learning_rate": 4.4565217391304346e-07,
      "loss": 1.2006,
      "step": 2379
    },
    {
      "epoch": 226.66666666666666,
      "grad_norm": 4.12217378616333,
      "learning_rate": 4.45839580209895e-07,
      "loss": 1.162,
      "step": 2380
    },
    {
      "epoch": 226.76190476190476,
      "grad_norm": 4.791848182678223,
      "learning_rate": 4.460269865067467e-07,
      "loss": 1.154,
      "step": 2381
    },
    {
      "epoch": 226.85714285714286,
      "grad_norm": 5.367868423461914,
      "learning_rate": 4.4621439280359823e-07,
      "loss": 1.2002,
      "step": 2382
    },
    {
      "epoch": 226.95238095238096,
      "grad_norm": 2.4993393421173096,
      "learning_rate": 4.4640179910044984e-07,
      "loss": 1.1519,
      "step": 2383
    },
    {
      "epoch": 226.95238095238096,
      "eval_accuracy": 0.720494567253653,
      "eval_f1": 0.776378896882494,
      "eval_loss": 0.5695188045501709,
      "eval_precision": 0.72265625,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.7686076568796776,
      "eval_runtime": 1.1904,
      "eval_samples_per_second": 2242.195,
      "eval_steps_per_second": 5.041,
      "step": 2383
    },
    {
      "epoch": 227.04761904761904,
      "grad_norm": 5.448750019073486,
      "learning_rate": 4.465892053973014e-07,
      "loss": 1.1544,
      "step": 2384
    },
    {
      "epoch": 227.14285714285714,
      "grad_norm": 9.408559799194336,
      "learning_rate": 4.4677661169415295e-07,
      "loss": 1.1194,
      "step": 2385
    },
    {
      "epoch": 227.23809523809524,
      "grad_norm": 5.749595642089844,
      "learning_rate": 4.469640179910045e-07,
      "loss": 1.1666,
      "step": 2386
    },
    {
      "epoch": 227.33333333333334,
      "grad_norm": 3.787705183029175,
      "learning_rate": 4.4715142428785606e-07,
      "loss": 1.1724,
      "step": 2387
    },
    {
      "epoch": 227.42857142857142,
      "grad_norm": 3.9639134407043457,
      "learning_rate": 4.473388305847077e-07,
      "loss": 1.1584,
      "step": 2388
    },
    {
      "epoch": 227.52380952380952,
      "grad_norm": 3.05351185798645,
      "learning_rate": 4.4752623688155927e-07,
      "loss": 1.1573,
      "step": 2389
    },
    {
      "epoch": 227.61904761904762,
      "grad_norm": 7.798318862915039,
      "learning_rate": 4.477136431784108e-07,
      "loss": 1.1385,
      "step": 2390
    },
    {
      "epoch": 227.71428571428572,
      "grad_norm": 4.202153205871582,
      "learning_rate": 4.479010494752624e-07,
      "loss": 1.1519,
      "step": 2391
    },
    {
      "epoch": 227.8095238095238,
      "grad_norm": 9.05360221862793,
      "learning_rate": 4.4808845577211393e-07,
      "loss": 1.151,
      "step": 2392
    },
    {
      "epoch": 227.9047619047619,
      "grad_norm": 5.996427536010742,
      "learning_rate": 4.4827586206896554e-07,
      "loss": 1.1745,
      "step": 2393
    },
    {
      "epoch": 228.0,
      "grad_norm": 3.683551073074341,
      "learning_rate": 4.484632683658171e-07,
      "loss": 1.0939,
      "step": 2394
    },
    {
      "epoch": 228.0,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7759952465834818,
      "eval_loss": 0.5693243145942688,
      "eval_precision": 0.716794731064764,
      "eval_recall": 0.8458549222797928,
      "eval_roc_auc": 0.7688845710995971,
      "eval_runtime": 1.223,
      "eval_samples_per_second": 2182.303,
      "eval_steps_per_second": 4.906,
      "step": 2394
    },
    {
      "epoch": 228.0952380952381,
      "grad_norm": 4.042111396789551,
      "learning_rate": 4.486506746626687e-07,
      "loss": 1.1741,
      "step": 2395
    },
    {
      "epoch": 228.1904761904762,
      "grad_norm": 3.9594829082489014,
      "learning_rate": 4.488380809595203e-07,
      "loss": 1.1649,
      "step": 2396
    },
    {
      "epoch": 228.28571428571428,
      "grad_norm": 5.861592769622803,
      "learning_rate": 4.4902548725637186e-07,
      "loss": 1.1348,
      "step": 2397
    },
    {
      "epoch": 228.38095238095238,
      "grad_norm": 4.552924156188965,
      "learning_rate": 4.492128935532234e-07,
      "loss": 1.1528,
      "step": 2398
    },
    {
      "epoch": 228.47619047619048,
      "grad_norm": 3.354738712310791,
      "learning_rate": 4.49400299850075e-07,
      "loss": 1.158,
      "step": 2399
    },
    {
      "epoch": 228.57142857142858,
      "grad_norm": 3.973210573196411,
      "learning_rate": 4.4958770614692653e-07,
      "loss": 1.1165,
      "step": 2400
    },
    {
      "epoch": 228.66666666666666,
      "grad_norm": 8.278904914855957,
      "learning_rate": 4.497751124437781e-07,
      "loss": 1.1517,
      "step": 2401
    },
    {
      "epoch": 228.76190476190476,
      "grad_norm": 5.11875581741333,
      "learning_rate": 4.4996251874062974e-07,
      "loss": 1.1558,
      "step": 2402
    },
    {
      "epoch": 228.85714285714286,
      "grad_norm": 8.316819190979004,
      "learning_rate": 4.501499250374813e-07,
      "loss": 1.1962,
      "step": 2403
    },
    {
      "epoch": 228.95238095238096,
      "grad_norm": 4.274223327636719,
      "learning_rate": 4.5033733133433285e-07,
      "loss": 1.1209,
      "step": 2404
    },
    {
      "epoch": 228.95238095238096,
      "eval_accuracy": 0.7189958786062196,
      "eval_f1": 0.7765196662693683,
      "eval_loss": 0.5691817402839661,
      "eval_precision": 0.7190949227373068,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7688310305123778,
      "eval_runtime": 1.1439,
      "eval_samples_per_second": 2333.192,
      "eval_steps_per_second": 5.245,
      "step": 2404
    },
    {
      "epoch": 229.04761904761904,
      "grad_norm": 3.7137949466705322,
      "learning_rate": 4.505247376311844e-07,
      "loss": 1.1322,
      "step": 2405
    },
    {
      "epoch": 229.14285714285714,
      "grad_norm": 4.170365810394287,
      "learning_rate": 4.50712143928036e-07,
      "loss": 1.116,
      "step": 2406
    },
    {
      "epoch": 229.23809523809524,
      "grad_norm": 3.865349054336548,
      "learning_rate": 4.5089955022488757e-07,
      "loss": 1.1371,
      "step": 2407
    },
    {
      "epoch": 229.33333333333334,
      "grad_norm": 4.7025909423828125,
      "learning_rate": 4.510869565217391e-07,
      "loss": 1.1646,
      "step": 2408
    },
    {
      "epoch": 229.42857142857142,
      "grad_norm": 4.561888694763184,
      "learning_rate": 4.512743628185908e-07,
      "loss": 1.1553,
      "step": 2409
    },
    {
      "epoch": 229.52380952380952,
      "grad_norm": 4.621860504150391,
      "learning_rate": 4.5146176911544234e-07,
      "loss": 1.1162,
      "step": 2410
    },
    {
      "epoch": 229.61904761904762,
      "grad_norm": 6.832645893096924,
      "learning_rate": 4.516491754122939e-07,
      "loss": 1.1645,
      "step": 2411
    },
    {
      "epoch": 229.71428571428572,
      "grad_norm": 7.709583759307861,
      "learning_rate": 4.5183658170914545e-07,
      "loss": 1.1843,
      "step": 2412
    },
    {
      "epoch": 229.8095238095238,
      "grad_norm": 6.104475498199463,
      "learning_rate": 4.52023988005997e-07,
      "loss": 1.1422,
      "step": 2413
    },
    {
      "epoch": 229.9047619047619,
      "grad_norm": 5.455144882202148,
      "learning_rate": 4.5221139430284856e-07,
      "loss": 1.1822,
      "step": 2414
    },
    {
      "epoch": 230.0,
      "grad_norm": 9.39183235168457,
      "learning_rate": 4.523988005997001e-07,
      "loss": 1.2096,
      "step": 2415
    },
    {
      "epoch": 230.0,
      "eval_accuracy": 0.7174971899587861,
      "eval_f1": 0.7750596658711217,
      "eval_loss": 0.5693327188491821,
      "eval_precision": 0.7184734513274337,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.7688477259643063,
      "eval_runtime": 1.1519,
      "eval_samples_per_second": 2317.07,
      "eval_steps_per_second": 5.209,
      "step": 2415
    },
    {
      "epoch": 230.0952380952381,
      "grad_norm": 5.667901515960693,
      "learning_rate": 4.5258620689655177e-07,
      "loss": 1.1667,
      "step": 2416
    },
    {
      "epoch": 230.1904761904762,
      "grad_norm": 5.791493892669678,
      "learning_rate": 4.527736131934033e-07,
      "loss": 1.0979,
      "step": 2417
    },
    {
      "epoch": 230.28571428571428,
      "grad_norm": 3.5346291065216064,
      "learning_rate": 4.5296101949025493e-07,
      "loss": 1.1542,
      "step": 2418
    },
    {
      "epoch": 230.38095238095238,
      "grad_norm": 5.434539318084717,
      "learning_rate": 4.531484257871065e-07,
      "loss": 1.1854,
      "step": 2419
    },
    {
      "epoch": 230.47619047619048,
      "grad_norm": 5.917377948760986,
      "learning_rate": 4.5333583208395804e-07,
      "loss": 1.1524,
      "step": 2420
    },
    {
      "epoch": 230.57142857142858,
      "grad_norm": 7.932279586791992,
      "learning_rate": 4.535232383808096e-07,
      "loss": 1.1634,
      "step": 2421
    },
    {
      "epoch": 230.66666666666666,
      "grad_norm": 12.676122665405273,
      "learning_rate": 4.5371064467766115e-07,
      "loss": 1.1323,
      "step": 2422
    },
    {
      "epoch": 230.76190476190476,
      "grad_norm": 8.025681495666504,
      "learning_rate": 4.538980509745128e-07,
      "loss": 1.141,
      "step": 2423
    },
    {
      "epoch": 230.85714285714286,
      "grad_norm": 5.724260330200195,
      "learning_rate": 4.5408545727136436e-07,
      "loss": 1.1401,
      "step": 2424
    },
    {
      "epoch": 230.95238095238096,
      "grad_norm": 3.4070522785186768,
      "learning_rate": 4.542728635682159e-07,
      "loss": 1.1843,
      "step": 2425
    },
    {
      "epoch": 230.95238095238096,
      "eval_accuracy": 0.7182465342825028,
      "eval_f1": 0.7772511848341233,
      "eval_loss": 0.5696398019790649,
      "eval_precision": 0.7161572052401747,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7688304548071386,
      "eval_runtime": 1.1655,
      "eval_samples_per_second": 2289.921,
      "eval_steps_per_second": 5.148,
      "step": 2425
    },
    {
      "epoch": 231.04761904761904,
      "grad_norm": 3.4953441619873047,
      "learning_rate": 4.544602698650675e-07,
      "loss": 1.1192,
      "step": 2426
    },
    {
      "epoch": 231.14285714285714,
      "grad_norm": 5.022072792053223,
      "learning_rate": 4.5464767616191903e-07,
      "loss": 1.1471,
      "step": 2427
    },
    {
      "epoch": 231.23809523809524,
      "grad_norm": 7.191939353942871,
      "learning_rate": 4.5483508245877064e-07,
      "loss": 1.1319,
      "step": 2428
    },
    {
      "epoch": 231.33333333333334,
      "grad_norm": 7.880967140197754,
      "learning_rate": 4.550224887556222e-07,
      "loss": 1.1201,
      "step": 2429
    },
    {
      "epoch": 231.42857142857142,
      "grad_norm": 12.768455505371094,
      "learning_rate": 4.552098950524738e-07,
      "loss": 1.2004,
      "step": 2430
    },
    {
      "epoch": 231.52380952380952,
      "grad_norm": 7.712562084197998,
      "learning_rate": 4.553973013493254e-07,
      "loss": 1.1585,
      "step": 2431
    },
    {
      "epoch": 231.61904761904762,
      "grad_norm": 4.747276782989502,
      "learning_rate": 4.5558470764617696e-07,
      "loss": 1.1559,
      "step": 2432
    },
    {
      "epoch": 231.71428571428572,
      "grad_norm": 8.367605209350586,
      "learning_rate": 4.557721139430285e-07,
      "loss": 1.1298,
      "step": 2433
    },
    {
      "epoch": 231.8095238095238,
      "grad_norm": 6.592958927154541,
      "learning_rate": 4.5595952023988007e-07,
      "loss": 1.1743,
      "step": 2434
    },
    {
      "epoch": 231.9047619047619,
      "grad_norm": 4.433559417724609,
      "learning_rate": 4.561469265367316e-07,
      "loss": 1.1075,
      "step": 2435
    },
    {
      "epoch": 232.0,
      "grad_norm": 8.43216609954834,
      "learning_rate": 4.563343328335832e-07,
      "loss": 1.12,
      "step": 2436
    },
    {
      "epoch": 232.0,
      "eval_accuracy": 0.7171225177969277,
      "eval_f1": 0.774828511780495,
      "eval_loss": 0.5698753595352173,
      "eval_precision": 0.7180762852404643,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.768350604490501,
      "eval_runtime": 1.3776,
      "eval_samples_per_second": 1937.376,
      "eval_steps_per_second": 4.355,
      "step": 2436
    },
    {
      "epoch": 232.0952380952381,
      "grad_norm": 4.639498233795166,
      "learning_rate": 4.5652173913043484e-07,
      "loss": 1.131,
      "step": 2437
    },
    {
      "epoch": 232.1904761904762,
      "grad_norm": 4.520444393157959,
      "learning_rate": 4.567091454272864e-07,
      "loss": 1.1549,
      "step": 2438
    },
    {
      "epoch": 232.28571428571428,
      "grad_norm": 3.809060573577881,
      "learning_rate": 4.5689655172413795e-07,
      "loss": 1.1717,
      "step": 2439
    },
    {
      "epoch": 232.38095238095238,
      "grad_norm": 3.0320441722869873,
      "learning_rate": 4.570839580209895e-07,
      "loss": 1.1552,
      "step": 2440
    },
    {
      "epoch": 232.47619047619048,
      "grad_norm": 7.700618743896484,
      "learning_rate": 4.572713643178411e-07,
      "loss": 1.1288,
      "step": 2441
    },
    {
      "epoch": 232.57142857142858,
      "grad_norm": 8.786882400512695,
      "learning_rate": 4.5745877061469266e-07,
      "loss": 1.1354,
      "step": 2442
    },
    {
      "epoch": 232.66666666666666,
      "grad_norm": 13.372679710388184,
      "learning_rate": 4.576461769115442e-07,
      "loss": 1.1771,
      "step": 2443
    },
    {
      "epoch": 232.76190476190476,
      "grad_norm": 3.7465243339538574,
      "learning_rate": 4.578335832083959e-07,
      "loss": 1.1508,
      "step": 2444
    },
    {
      "epoch": 232.85714285714286,
      "grad_norm": 3.863992214202881,
      "learning_rate": 4.5802098950524743e-07,
      "loss": 1.1347,
      "step": 2445
    },
    {
      "epoch": 232.95238095238096,
      "grad_norm": 4.447786331176758,
      "learning_rate": 4.58208395802099e-07,
      "loss": 1.1635,
      "step": 2446
    },
    {
      "epoch": 232.95238095238096,
      "eval_accuracy": 0.7189958786062196,
      "eval_f1": 0.7781065088757396,
      "eval_loss": 0.5697467923164368,
      "eval_precision": 0.7162309368191722,
      "eval_recall": 0.8516839378238342,
      "eval_roc_auc": 0.7686997697179043,
      "eval_runtime": 1.1824,
      "eval_samples_per_second": 2257.22,
      "eval_steps_per_second": 5.074,
      "step": 2446
    },
    {
      "epoch": 233.04761904761904,
      "grad_norm": 4.46320104598999,
      "learning_rate": 4.5839580209895054e-07,
      "loss": 1.1732,
      "step": 2447
    },
    {
      "epoch": 233.14285714285714,
      "grad_norm": 6.758903980255127,
      "learning_rate": 4.585832083958021e-07,
      "loss": 1.1669,
      "step": 2448
    },
    {
      "epoch": 233.23809523809524,
      "grad_norm": 12.404580116271973,
      "learning_rate": 4.5877061469265365e-07,
      "loss": 1.1597,
      "step": 2449
    },
    {
      "epoch": 233.33333333333334,
      "grad_norm": 5.045426368713379,
      "learning_rate": 4.589580209895052e-07,
      "loss": 1.1608,
      "step": 2450
    },
    {
      "epoch": 233.42857142857142,
      "grad_norm": 5.315583229064941,
      "learning_rate": 4.5914542728635687e-07,
      "loss": 1.1361,
      "step": 2451
    },
    {
      "epoch": 233.52380952380952,
      "grad_norm": 6.253724575042725,
      "learning_rate": 4.593328335832084e-07,
      "loss": 1.1571,
      "step": 2452
    },
    {
      "epoch": 233.61904761904762,
      "grad_norm": 3.5962157249450684,
      "learning_rate": 4.5952023988006003e-07,
      "loss": 1.1451,
      "step": 2453
    },
    {
      "epoch": 233.71428571428572,
      "grad_norm": 5.8950514793396,
      "learning_rate": 4.597076461769116e-07,
      "loss": 1.137,
      "step": 2454
    },
    {
      "epoch": 233.8095238095238,
      "grad_norm": 3.1741690635681152,
      "learning_rate": 4.5989505247376314e-07,
      "loss": 1.1279,
      "step": 2455
    },
    {
      "epoch": 233.9047619047619,
      "grad_norm": 3.401339054107666,
      "learning_rate": 4.600824587706147e-07,
      "loss": 1.1507,
      "step": 2456
    },
    {
      "epoch": 234.0,
      "grad_norm": 4.254356384277344,
      "learning_rate": 4.6026986506746625e-07,
      "loss": 1.1214,
      "step": 2457
    },
    {
      "epoch": 234.0,
      "eval_accuracy": 0.7197452229299363,
      "eval_f1": 0.7759137207908927,
      "eval_loss": 0.569048285484314,
      "eval_precision": 0.721850613154961,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.7692043753598157,
      "eval_runtime": 1.1584,
      "eval_samples_per_second": 2303.977,
      "eval_steps_per_second": 5.179,
      "step": 2457
    },
    {
      "epoch": 234.0952380952381,
      "grad_norm": 4.8844146728515625,
      "learning_rate": 4.604572713643179e-07,
      "loss": 1.1469,
      "step": 2458
    },
    {
      "epoch": 234.1904761904762,
      "grad_norm": 4.128713607788086,
      "learning_rate": 4.6064467766116946e-07,
      "loss": 1.1274,
      "step": 2459
    },
    {
      "epoch": 234.28571428571428,
      "grad_norm": 3.244645118713379,
      "learning_rate": 4.60832083958021e-07,
      "loss": 1.1395,
      "step": 2460
    },
    {
      "epoch": 234.38095238095238,
      "grad_norm": 5.353634834289551,
      "learning_rate": 4.6101949025487257e-07,
      "loss": 1.1475,
      "step": 2461
    },
    {
      "epoch": 234.47619047619048,
      "grad_norm": 5.806576728820801,
      "learning_rate": 4.612068965517241e-07,
      "loss": 1.1432,
      "step": 2462
    },
    {
      "epoch": 234.57142857142858,
      "grad_norm": 13.247061729431152,
      "learning_rate": 4.6139430284857573e-07,
      "loss": 1.1144,
      "step": 2463
    },
    {
      "epoch": 234.66666666666666,
      "grad_norm": 2.7329390048980713,
      "learning_rate": 4.615817091454273e-07,
      "loss": 1.1431,
      "step": 2464
    },
    {
      "epoch": 234.76190476190476,
      "grad_norm": 3.864218235015869,
      "learning_rate": 4.617691154422789e-07,
      "loss": 1.1806,
      "step": 2465
    },
    {
      "epoch": 234.85714285714286,
      "grad_norm": 14.171060562133789,
      "learning_rate": 4.619565217391305e-07,
      "loss": 1.1975,
      "step": 2466
    },
    {
      "epoch": 234.95238095238096,
      "grad_norm": 8.89664077758789,
      "learning_rate": 4.6214392803598205e-07,
      "loss": 1.139,
      "step": 2467
    },
    {
      "epoch": 234.95238095238096,
      "eval_accuracy": 0.720494567253653,
      "eval_f1": 0.7783719548425431,
      "eval_loss": 0.5684369206428528,
      "eval_precision": 0.7189901207464325,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7698618307426597,
      "eval_runtime": 1.1725,
      "eval_samples_per_second": 2276.408,
      "eval_steps_per_second": 5.117,
      "step": 2467
    },
    {
      "epoch": 235.04761904761904,
      "grad_norm": 7.042123794555664,
      "learning_rate": 4.623313343328336e-07,
      "loss": 1.1587,
      "step": 2468
    },
    {
      "epoch": 235.14285714285714,
      "grad_norm": 6.67264986038208,
      "learning_rate": 4.6251874062968516e-07,
      "loss": 1.155,
      "step": 2469
    },
    {
      "epoch": 235.23809523809524,
      "grad_norm": 2.7996301651000977,
      "learning_rate": 4.627061469265367e-07,
      "loss": 1.1606,
      "step": 2470
    },
    {
      "epoch": 235.33333333333334,
      "grad_norm": 6.432344913482666,
      "learning_rate": 4.6289355322338827e-07,
      "loss": 1.1244,
      "step": 2471
    },
    {
      "epoch": 235.42857142857142,
      "grad_norm": 7.238835334777832,
      "learning_rate": 4.6308095952023993e-07,
      "loss": 1.1404,
      "step": 2472
    },
    {
      "epoch": 235.52380952380952,
      "grad_norm": 7.994748592376709,
      "learning_rate": 4.632683658170915e-07,
      "loss": 1.1516,
      "step": 2473
    },
    {
      "epoch": 235.61904761904762,
      "grad_norm": 9.119189262390137,
      "learning_rate": 4.6345577211394304e-07,
      "loss": 1.1525,
      "step": 2474
    },
    {
      "epoch": 235.71428571428572,
      "grad_norm": 3.28615403175354,
      "learning_rate": 4.636431784107946e-07,
      "loss": 1.1559,
      "step": 2475
    },
    {
      "epoch": 235.8095238095238,
      "grad_norm": 3.7294490337371826,
      "learning_rate": 4.638305847076462e-07,
      "loss": 1.157,
      "step": 2476
    },
    {
      "epoch": 235.9047619047619,
      "grad_norm": 7.549288272857666,
      "learning_rate": 4.6401799100449776e-07,
      "loss": 1.1369,
      "step": 2477
    },
    {
      "epoch": 236.0,
      "grad_norm": 6.290277481079102,
      "learning_rate": 4.642053973013493e-07,
      "loss": 1.0902,
      "step": 2478
    },
    {
      "epoch": 236.0,
      "eval_accuracy": 0.720494567253653,
      "eval_f1": 0.7765128819652487,
      "eval_loss": 0.5682975053787231,
      "eval_precision": 0.7224080267558528,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7701263672999424,
      "eval_runtime": 1.2309,
      "eval_samples_per_second": 2168.419,
      "eval_steps_per_second": 4.875,
      "step": 2478
    },
    {
      "epoch": 236.0952380952381,
      "grad_norm": 5.862534523010254,
      "learning_rate": 4.6439280359820097e-07,
      "loss": 1.1472,
      "step": 2479
    },
    {
      "epoch": 236.1904761904762,
      "grad_norm": 4.279983997344971,
      "learning_rate": 4.6458020989505253e-07,
      "loss": 1.1246,
      "step": 2480
    },
    {
      "epoch": 236.28571428571428,
      "grad_norm": 14.398469924926758,
      "learning_rate": 4.647676161919041e-07,
      "loss": 1.1189,
      "step": 2481
    },
    {
      "epoch": 236.38095238095238,
      "grad_norm": 3.601273536682129,
      "learning_rate": 4.6495502248875564e-07,
      "loss": 1.1493,
      "step": 2482
    },
    {
      "epoch": 236.47619047619048,
      "grad_norm": 8.719027519226074,
      "learning_rate": 4.651424287856072e-07,
      "loss": 1.1419,
      "step": 2483
    },
    {
      "epoch": 236.57142857142858,
      "grad_norm": 3.2955257892608643,
      "learning_rate": 4.6532983508245875e-07,
      "loss": 1.1642,
      "step": 2484
    },
    {
      "epoch": 236.66666666666666,
      "grad_norm": 3.568007707595825,
      "learning_rate": 4.655172413793103e-07,
      "loss": 1.1462,
      "step": 2485
    },
    {
      "epoch": 236.76190476190476,
      "grad_norm": 5.648521900177002,
      "learning_rate": 4.6570464767616196e-07,
      "loss": 1.1876,
      "step": 2486
    },
    {
      "epoch": 236.85714285714286,
      "grad_norm": 6.08394193649292,
      "learning_rate": 4.658920539730135e-07,
      "loss": 1.1548,
      "step": 2487
    },
    {
      "epoch": 236.95238095238096,
      "grad_norm": 3.1771278381347656,
      "learning_rate": 4.660794602698651e-07,
      "loss": 1.1594,
      "step": 2488
    },
    {
      "epoch": 236.95238095238096,
      "eval_accuracy": 0.7212439115773698,
      "eval_f1": 0.7785714285714286,
      "eval_loss": 0.5684502720832825,
      "eval_precision": 0.7202643171806168,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7701934369602764,
      "eval_runtime": 1.2586,
      "eval_samples_per_second": 2120.603,
      "eval_steps_per_second": 4.767,
      "step": 2488
    },
    {
      "epoch": 237.04761904761904,
      "grad_norm": 7.633711814880371,
      "learning_rate": 4.662668665667167e-07,
      "loss": 1.1075,
      "step": 2489
    },
    {
      "epoch": 237.14285714285714,
      "grad_norm": 7.139071464538574,
      "learning_rate": 4.6645427286356823e-07,
      "loss": 1.112,
      "step": 2490
    },
    {
      "epoch": 237.23809523809524,
      "grad_norm": 4.67210054397583,
      "learning_rate": 4.666416791604198e-07,
      "loss": 1.1502,
      "step": 2491
    },
    {
      "epoch": 237.33333333333334,
      "grad_norm": 5.977503299713135,
      "learning_rate": 4.6682908545727134e-07,
      "loss": 1.1348,
      "step": 2492
    },
    {
      "epoch": 237.42857142857142,
      "grad_norm": 7.3450236320495605,
      "learning_rate": 4.67016491754123e-07,
      "loss": 1.1331,
      "step": 2493
    },
    {
      "epoch": 237.52380952380952,
      "grad_norm": 12.501540184020996,
      "learning_rate": 4.6720389805097456e-07,
      "loss": 1.1568,
      "step": 2494
    },
    {
      "epoch": 237.61904761904762,
      "grad_norm": 4.337953090667725,
      "learning_rate": 4.673913043478261e-07,
      "loss": 1.1957,
      "step": 2495
    },
    {
      "epoch": 237.71428571428572,
      "grad_norm": 7.36350679397583,
      "learning_rate": 4.6757871064467766e-07,
      "loss": 1.1367,
      "step": 2496
    },
    {
      "epoch": 237.8095238095238,
      "grad_norm": 12.35847282409668,
      "learning_rate": 4.677661169415292e-07,
      "loss": 1.1597,
      "step": 2497
    },
    {
      "epoch": 237.9047619047619,
      "grad_norm": 4.463048458099365,
      "learning_rate": 4.679535232383808e-07,
      "loss": 1.178,
      "step": 2498
    },
    {
      "epoch": 238.0,
      "grad_norm": 5.635746479034424,
      "learning_rate": 4.681409295352324e-07,
      "loss": 1.1515,
      "step": 2499
    },
    {
      "epoch": 238.0,
      "eval_accuracy": 0.7208692394155114,
      "eval_f1": 0.777810915597972,
      "eval_loss": 0.568373441696167,
      "eval_precision": 0.720840243228303,
      "eval_recall": 0.844559585492228,
      "eval_roc_auc": 0.7703940702360392,
      "eval_runtime": 1.1859,
      "eval_samples_per_second": 2250.588,
      "eval_steps_per_second": 5.059,
      "step": 2499
    },
    {
      "epoch": 238.0952380952381,
      "grad_norm": 3.2132604122161865,
      "learning_rate": 4.68328335832084e-07,
      "loss": 1.1247,
      "step": 2500
    },
    {
      "epoch": 238.1904761904762,
      "grad_norm": 5.9469194412231445,
      "learning_rate": 4.685157421289356e-07,
      "loss": 1.1052,
      "step": 2501
    },
    {
      "epoch": 238.28571428571428,
      "grad_norm": 6.368658542633057,
      "learning_rate": 4.6870314842578715e-07,
      "loss": 1.159,
      "step": 2502
    },
    {
      "epoch": 238.38095238095238,
      "grad_norm": 4.543058395385742,
      "learning_rate": 4.688905547226387e-07,
      "loss": 1.1733,
      "step": 2503
    },
    {
      "epoch": 238.47619047619048,
      "grad_norm": 8.83027172088623,
      "learning_rate": 4.6907796101949026e-07,
      "loss": 1.1485,
      "step": 2504
    },
    {
      "epoch": 238.57142857142858,
      "grad_norm": 4.797172546386719,
      "learning_rate": 4.692653673163418e-07,
      "loss": 1.1442,
      "step": 2505
    },
    {
      "epoch": 238.66666666666666,
      "grad_norm": 7.15238094329834,
      "learning_rate": 4.6945277361319337e-07,
      "loss": 1.1412,
      "step": 2506
    },
    {
      "epoch": 238.76190476190476,
      "grad_norm": 9.403074264526367,
      "learning_rate": 4.6964017991004503e-07,
      "loss": 1.1514,
      "step": 2507
    },
    {
      "epoch": 238.85714285714286,
      "grad_norm": 7.695036888122559,
      "learning_rate": 4.698275862068966e-07,
      "loss": 1.1721,
      "step": 2508
    },
    {
      "epoch": 238.95238095238096,
      "grad_norm": 5.925418853759766,
      "learning_rate": 4.7001499250374814e-07,
      "loss": 1.1613,
      "step": 2509
    },
    {
      "epoch": 238.95238095238096,
      "eval_accuracy": 0.7231172723866617,
      "eval_f1": 0.7807772174428953,
      "eval_loss": 0.568254292011261,
      "eval_precision": 0.7203065134099617,
      "eval_recall": 0.8523316062176166,
      "eval_roc_auc": 0.7708552101324121,
      "eval_runtime": 1.1677,
      "eval_samples_per_second": 2285.74,
      "eval_steps_per_second": 5.138,
      "step": 2509
    },
    {
      "epoch": 239.04761904761904,
      "grad_norm": 4.571139812469482,
      "learning_rate": 4.702023988005997e-07,
      "loss": 1.1747,
      "step": 2510
    },
    {
      "epoch": 239.14285714285714,
      "grad_norm": 3.523711919784546,
      "learning_rate": 4.703898050974513e-07,
      "loss": 1.1276,
      "step": 2511
    },
    {
      "epoch": 239.23809523809524,
      "grad_norm": 3.364738941192627,
      "learning_rate": 4.7057721139430285e-07,
      "loss": 1.1753,
      "step": 2512
    },
    {
      "epoch": 239.33333333333334,
      "grad_norm": 3.596714973449707,
      "learning_rate": 4.707646176911544e-07,
      "loss": 1.15,
      "step": 2513
    },
    {
      "epoch": 239.42857142857142,
      "grad_norm": 10.352349281311035,
      "learning_rate": 4.7095202398800607e-07,
      "loss": 1.1382,
      "step": 2514
    },
    {
      "epoch": 239.52380952380952,
      "grad_norm": 10.572864532470703,
      "learning_rate": 4.711394302848576e-07,
      "loss": 1.1505,
      "step": 2515
    },
    {
      "epoch": 239.61904761904762,
      "grad_norm": 3.3369085788726807,
      "learning_rate": 4.713268365817092e-07,
      "loss": 1.136,
      "step": 2516
    },
    {
      "epoch": 239.71428571428572,
      "grad_norm": 7.171133041381836,
      "learning_rate": 4.7151424287856073e-07,
      "loss": 1.1219,
      "step": 2517
    },
    {
      "epoch": 239.8095238095238,
      "grad_norm": 7.1187238693237305,
      "learning_rate": 4.717016491754123e-07,
      "loss": 1.0975,
      "step": 2518
    },
    {
      "epoch": 239.9047619047619,
      "grad_norm": 12.227962493896484,
      "learning_rate": 4.7188905547226384e-07,
      "loss": 1.1898,
      "step": 2519
    },
    {
      "epoch": 240.0,
      "grad_norm": 3.9097506999969482,
      "learning_rate": 4.7207646176911545e-07,
      "loss": 1.1471,
      "step": 2520
    },
    {
      "epoch": 240.0,
      "eval_accuracy": 0.7219932559010865,
      "eval_f1": 0.7795603089720736,
      "eval_loss": 0.5678714513778687,
      "eval_precision": 0.7200878155872668,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7711871042026482,
      "eval_runtime": 1.2783,
      "eval_samples_per_second": 2087.875,
      "eval_steps_per_second": 4.694,
      "step": 2520
    },
    {
      "epoch": 240.0952380952381,
      "grad_norm": 7.612576961517334,
      "learning_rate": 4.7226386806596706e-07,
      "loss": 1.0953,
      "step": 2521
    },
    {
      "epoch": 240.1904761904762,
      "grad_norm": 9.153794288635254,
      "learning_rate": 4.724512743628186e-07,
      "loss": 1.1538,
      "step": 2522
    },
    {
      "epoch": 240.28571428571428,
      "grad_norm": 3.632295608520508,
      "learning_rate": 4.726386806596702e-07,
      "loss": 1.178,
      "step": 2523
    },
    {
      "epoch": 240.38095238095238,
      "grad_norm": 3.858614444732666,
      "learning_rate": 4.7282608695652177e-07,
      "loss": 1.1294,
      "step": 2524
    },
    {
      "epoch": 240.47619047619048,
      "grad_norm": 6.603476524353027,
      "learning_rate": 4.730134932533733e-07,
      "loss": 1.1404,
      "step": 2525
    },
    {
      "epoch": 240.57142857142858,
      "grad_norm": 6.5844550132751465,
      "learning_rate": 4.732008995502249e-07,
      "loss": 1.1337,
      "step": 2526
    },
    {
      "epoch": 240.66666666666666,
      "grad_norm": 6.402451038360596,
      "learning_rate": 4.7338830584707654e-07,
      "loss": 1.1919,
      "step": 2527
    },
    {
      "epoch": 240.76190476190476,
      "grad_norm": 4.313340187072754,
      "learning_rate": 4.735757121439281e-07,
      "loss": 1.1371,
      "step": 2528
    },
    {
      "epoch": 240.85714285714286,
      "grad_norm": 2.5007331371307373,
      "learning_rate": 4.7376311844077965e-07,
      "loss": 1.1315,
      "step": 2529
    },
    {
      "epoch": 240.95238095238096,
      "grad_norm": 3.1889805793762207,
      "learning_rate": 4.739505247376312e-07,
      "loss": 1.124,
      "step": 2530
    },
    {
      "epoch": 240.95238095238096,
      "eval_accuracy": 0.7223679280629449,
      "eval_f1": 0.7767399819222658,
      "eval_loss": 0.5677710175514221,
      "eval_precision": 0.7261971830985916,
      "eval_recall": 0.8348445595854922,
      "eval_roc_auc": 0.7709075993091538,
      "eval_runtime": 1.1464,
      "eval_samples_per_second": 2328.123,
      "eval_steps_per_second": 5.234,
      "step": 2530
    },
    {
      "epoch": 241.04761904761904,
      "grad_norm": 3.370150327682495,
      "learning_rate": 4.7413793103448276e-07,
      "loss": 1.1566,
      "step": 2531
    },
    {
      "epoch": 241.14285714285714,
      "grad_norm": 7.428020000457764,
      "learning_rate": 4.743253373313343e-07,
      "loss": 1.1453,
      "step": 2532
    },
    {
      "epoch": 241.23809523809524,
      "grad_norm": 4.833296775817871,
      "learning_rate": 4.745127436281859e-07,
      "loss": 1.2207,
      "step": 2533
    },
    {
      "epoch": 241.33333333333334,
      "grad_norm": 7.765406131744385,
      "learning_rate": 4.7470014992503753e-07,
      "loss": 1.1642,
      "step": 2534
    },
    {
      "epoch": 241.42857142857142,
      "grad_norm": 12.157771110534668,
      "learning_rate": 4.748875562218891e-07,
      "loss": 1.1321,
      "step": 2535
    },
    {
      "epoch": 241.52380952380952,
      "grad_norm": 4.821937084197998,
      "learning_rate": 4.750749625187407e-07,
      "loss": 1.1655,
      "step": 2536
    },
    {
      "epoch": 241.61904761904762,
      "grad_norm": 4.60735559463501,
      "learning_rate": 4.7526236881559224e-07,
      "loss": 1.1469,
      "step": 2537
    },
    {
      "epoch": 241.71428571428572,
      "grad_norm": 3.312079668045044,
      "learning_rate": 4.754497751124438e-07,
      "loss": 1.1382,
      "step": 2538
    },
    {
      "epoch": 241.8095238095238,
      "grad_norm": 6.541538238525391,
      "learning_rate": 4.7563718140929535e-07,
      "loss": 1.1609,
      "step": 2539
    },
    {
      "epoch": 241.9047619047619,
      "grad_norm": 3.9310693740844727,
      "learning_rate": 4.758245877061469e-07,
      "loss": 1.1411,
      "step": 2540
    },
    {
      "epoch": 242.0,
      "grad_norm": 7.887485980987549,
      "learning_rate": 4.7601199400299857e-07,
      "loss": 1.1084,
      "step": 2541
    },
    {
      "epoch": 242.0,
      "eval_accuracy": 0.7246159610340952,
      "eval_f1": 0.7822222222222223,
      "eval_loss": 0.5681803226470947,
      "eval_precision": 0.7209175314036046,
      "eval_recall": 0.8549222797927462,
      "eval_roc_auc": 0.7706738629821532,
      "eval_runtime": 1.1999,
      "eval_samples_per_second": 2224.376,
      "eval_steps_per_second": 5.0,
      "step": 2541
    },
    {
      "epoch": 242.0952380952381,
      "grad_norm": 3.5262506008148193,
      "learning_rate": 4.761994002998501e-07,
      "loss": 1.1124,
      "step": 2542
    },
    {
      "epoch": 242.1904761904762,
      "grad_norm": 4.377519130706787,
      "learning_rate": 4.763868065967017e-07,
      "loss": 1.1453,
      "step": 2543
    },
    {
      "epoch": 242.28571428571428,
      "grad_norm": 3.5326547622680664,
      "learning_rate": 4.7657421289355323e-07,
      "loss": 1.1554,
      "step": 2544
    },
    {
      "epoch": 242.38095238095238,
      "grad_norm": 11.76237964630127,
      "learning_rate": 4.7676161919040484e-07,
      "loss": 1.1544,
      "step": 2545
    },
    {
      "epoch": 242.47619047619048,
      "grad_norm": 10.268291473388672,
      "learning_rate": 4.769490254872563e-07,
      "loss": 1.1011,
      "step": 2546
    },
    {
      "epoch": 242.57142857142858,
      "grad_norm": 5.779625415802002,
      "learning_rate": 4.771364317841079e-07,
      "loss": 1.108,
      "step": 2547
    },
    {
      "epoch": 242.66666666666666,
      "grad_norm": 9.959461212158203,
      "learning_rate": 4.773238380809596e-07,
      "loss": 1.16,
      "step": 2548
    },
    {
      "epoch": 242.76190476190476,
      "grad_norm": 7.101002216339111,
      "learning_rate": 4.775112443778111e-07,
      "loss": 1.151,
      "step": 2549
    },
    {
      "epoch": 242.85714285714286,
      "grad_norm": 7.493264675140381,
      "learning_rate": 4.776986506746627e-07,
      "loss": 1.1914,
      "step": 2550
    },
    {
      "epoch": 242.95238095238096,
      "grad_norm": 5.301811695098877,
      "learning_rate": 4.778860569715142e-07,
      "loss": 1.15,
      "step": 2551
    },
    {
      "epoch": 242.95238095238096,
      "eval_accuracy": 0.7212439115773698,
      "eval_f1": 0.7785714285714286,
      "eval_loss": 0.5683825016021729,
      "eval_precision": 0.7202643171806168,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7700987334484745,
      "eval_runtime": 1.2247,
      "eval_samples_per_second": 2179.256,
      "eval_steps_per_second": 4.899,
      "step": 2551
    },
    {
      "epoch": 243.04761904761904,
      "grad_norm": 10.990437507629395,
      "learning_rate": 4.780734632683659e-07,
      "loss": 1.1112,
      "step": 2552
    },
    {
      "epoch": 243.14285714285714,
      "grad_norm": 6.5883660316467285,
      "learning_rate": 4.782608695652174e-07,
      "loss": 1.186,
      "step": 2553
    },
    {
      "epoch": 243.23809523809524,
      "grad_norm": 5.139063358306885,
      "learning_rate": 4.78448275862069e-07,
      "loss": 1.1898,
      "step": 2554
    },
    {
      "epoch": 243.33333333333334,
      "grad_norm": 5.769028186798096,
      "learning_rate": 4.786356821589206e-07,
      "loss": 1.1261,
      "step": 2555
    },
    {
      "epoch": 243.42857142857142,
      "grad_norm": 2.6461703777313232,
      "learning_rate": 4.788230884557722e-07,
      "loss": 1.1643,
      "step": 2556
    },
    {
      "epoch": 243.52380952380952,
      "grad_norm": 8.833353996276855,
      "learning_rate": 4.790104947526238e-07,
      "loss": 1.0969,
      "step": 2557
    },
    {
      "epoch": 243.61904761904762,
      "grad_norm": 8.292926788330078,
      "learning_rate": 4.791979010494753e-07,
      "loss": 1.1508,
      "step": 2558
    },
    {
      "epoch": 243.71428571428572,
      "grad_norm": 5.664553642272949,
      "learning_rate": 4.793853073463269e-07,
      "loss": 1.1831,
      "step": 2559
    },
    {
      "epoch": 243.8095238095238,
      "grad_norm": 3.355820655822754,
      "learning_rate": 4.795727136431784e-07,
      "loss": 1.1163,
      "step": 2560
    },
    {
      "epoch": 243.9047619047619,
      "grad_norm": 3.2066733837127686,
      "learning_rate": 4.7976011994003e-07,
      "loss": 1.1299,
      "step": 2561
    },
    {
      "epoch": 244.0,
      "grad_norm": 16.678447723388672,
      "learning_rate": 4.799475262368816e-07,
      "loss": 1.1288,
      "step": 2562
    },
    {
      "epoch": 244.0,
      "eval_accuracy": 0.7234919445485201,
      "eval_f1": 0.7806183115338883,
      "eval_loss": 0.5679145455360413,
      "eval_precision": 0.7214285714285714,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7707867012089811,
      "eval_runtime": 1.6147,
      "eval_samples_per_second": 1652.969,
      "eval_steps_per_second": 3.716,
      "step": 2562
    },
    {
      "epoch": 244.0952380952381,
      "grad_norm": 3.1166558265686035,
      "learning_rate": 4.801349325337332e-07,
      "loss": 1.1614,
      "step": 2563
    },
    {
      "epoch": 244.1904761904762,
      "grad_norm": 4.723116874694824,
      "learning_rate": 4.803223388305847e-07,
      "loss": 1.1669,
      "step": 2564
    },
    {
      "epoch": 244.28571428571428,
      "grad_norm": 5.824684143066406,
      "learning_rate": 4.805097451274363e-07,
      "loss": 1.1639,
      "step": 2565
    },
    {
      "epoch": 244.38095238095238,
      "grad_norm": 4.820675849914551,
      "learning_rate": 4.806971514242879e-07,
      "loss": 1.1597,
      "step": 2566
    },
    {
      "epoch": 244.47619047619048,
      "grad_norm": 4.829911708831787,
      "learning_rate": 4.808845577211394e-07,
      "loss": 1.177,
      "step": 2567
    },
    {
      "epoch": 244.57142857142858,
      "grad_norm": 3.4412834644317627,
      "learning_rate": 4.81071964017991e-07,
      "loss": 1.1339,
      "step": 2568
    },
    {
      "epoch": 244.66666666666666,
      "grad_norm": 5.206437110900879,
      "learning_rate": 4.812593703148426e-07,
      "loss": 1.1264,
      "step": 2569
    },
    {
      "epoch": 244.76190476190476,
      "grad_norm": 7.612485408782959,
      "learning_rate": 4.814467766116942e-07,
      "loss": 1.1412,
      "step": 2570
    },
    {
      "epoch": 244.85714285714286,
      "grad_norm": 3.95589542388916,
      "learning_rate": 4.816341829085457e-07,
      "loss": 1.1521,
      "step": 2571
    },
    {
      "epoch": 244.95238095238096,
      "grad_norm": 5.139015197753906,
      "learning_rate": 4.818215892053973e-07,
      "loss": 1.1365,
      "step": 2572
    },
    {
      "epoch": 244.95238095238096,
      "eval_accuracy": 0.7216185837392282,
      "eval_f1": 0.7805022156573117,
      "eval_loss": 0.5677750706672668,
      "eval_precision": 0.7175448126018468,
      "eval_recall": 0.8555699481865285,
      "eval_roc_auc": 0.7708963730569948,
      "eval_runtime": 1.1166,
      "eval_samples_per_second": 2390.274,
      "eval_steps_per_second": 5.373,
      "step": 2572
    },
    {
      "epoch": 245.04761904761904,
      "grad_norm": 5.804911136627197,
      "learning_rate": 4.820089955022488e-07,
      "loss": 1.1618,
      "step": 2573
    },
    {
      "epoch": 245.14285714285714,
      "grad_norm": 8.898213386535645,
      "learning_rate": 4.821964017991005e-07,
      "loss": 1.1644,
      "step": 2574
    },
    {
      "epoch": 245.23809523809524,
      "grad_norm": 5.84812593460083,
      "learning_rate": 4.823838080959521e-07,
      "loss": 1.0921,
      "step": 2575
    },
    {
      "epoch": 245.33333333333334,
      "grad_norm": 7.604754447937012,
      "learning_rate": 4.825712143928036e-07,
      "loss": 1.1487,
      "step": 2576
    },
    {
      "epoch": 245.42857142857142,
      "grad_norm": 7.634527683258057,
      "learning_rate": 4.827586206896553e-07,
      "loss": 1.1865,
      "step": 2577
    },
    {
      "epoch": 245.52380952380952,
      "grad_norm": 4.072860240936279,
      "learning_rate": 4.829460269865068e-07,
      "loss": 1.1139,
      "step": 2578
    },
    {
      "epoch": 245.61904761904762,
      "grad_norm": 8.710189819335938,
      "learning_rate": 4.831334332833584e-07,
      "loss": 1.1172,
      "step": 2579
    },
    {
      "epoch": 245.71428571428572,
      "grad_norm": 4.168026924133301,
      "learning_rate": 4.833208395802099e-07,
      "loss": 1.1178,
      "step": 2580
    },
    {
      "epoch": 245.8095238095238,
      "grad_norm": 4.758199214935303,
      "learning_rate": 4.835082458770615e-07,
      "loss": 1.1392,
      "step": 2581
    },
    {
      "epoch": 245.9047619047619,
      "grad_norm": 4.961451053619385,
      "learning_rate": 4.83695652173913e-07,
      "loss": 1.1511,
      "step": 2582
    },
    {
      "epoch": 246.0,
      "grad_norm": 3.5863423347473145,
      "learning_rate": 4.838830584707647e-07,
      "loss": 1.1492,
      "step": 2583
    },
    {
      "epoch": 246.0,
      "eval_accuracy": 0.7219932559010865,
      "eval_f1": 0.7774445110977805,
      "eval_loss": 0.5673912167549133,
      "eval_precision": 0.7240223463687151,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7707337363270006,
      "eval_runtime": 1.1796,
      "eval_samples_per_second": 2262.657,
      "eval_steps_per_second": 5.087,
      "step": 2583
    },
    {
      "epoch": 246.0952380952381,
      "grad_norm": 5.360644817352295,
      "learning_rate": 4.840704647676163e-07,
      "loss": 1.1396,
      "step": 2584
    },
    {
      "epoch": 246.1904761904762,
      "grad_norm": 5.5368523597717285,
      "learning_rate": 4.842578710644678e-07,
      "loss": 1.1432,
      "step": 2585
    },
    {
      "epoch": 246.28571428571428,
      "grad_norm": 7.85012674331665,
      "learning_rate": 4.844452773613194e-07,
      "loss": 1.142,
      "step": 2586
    },
    {
      "epoch": 246.38095238095238,
      "grad_norm": 3.8528225421905518,
      "learning_rate": 4.846326836581709e-07,
      "loss": 1.1475,
      "step": 2587
    },
    {
      "epoch": 246.47619047619048,
      "grad_norm": 4.584531307220459,
      "learning_rate": 4.848200899550225e-07,
      "loss": 1.1688,
      "step": 2588
    },
    {
      "epoch": 246.57142857142858,
      "grad_norm": 4.979501247406006,
      "learning_rate": 4.85007496251874e-07,
      "loss": 1.1626,
      "step": 2589
    },
    {
      "epoch": 246.66666666666666,
      "grad_norm": 11.004454612731934,
      "learning_rate": 4.851949025487257e-07,
      "loss": 1.1249,
      "step": 2590
    },
    {
      "epoch": 246.76190476190476,
      "grad_norm": 10.931334495544434,
      "learning_rate": 4.853823088455772e-07,
      "loss": 1.1176,
      "step": 2591
    },
    {
      "epoch": 246.85714285714286,
      "grad_norm": 6.858458995819092,
      "learning_rate": 4.855697151424288e-07,
      "loss": 1.1268,
      "step": 2592
    },
    {
      "epoch": 246.95238095238096,
      "grad_norm": 6.829026222229004,
      "learning_rate": 4.857571214392804e-07,
      "loss": 1.1773,
      "step": 2593
    },
    {
      "epoch": 246.95238095238096,
      "eval_accuracy": 0.7223679280629449,
      "eval_f1": 0.7797919762258544,
      "eval_loss": 0.5674595236778259,
      "eval_precision": 0.7204832509610104,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7707849740932642,
      "eval_runtime": 1.2422,
      "eval_samples_per_second": 2148.687,
      "eval_steps_per_second": 4.83,
      "step": 2593
    },
    {
      "epoch": 247.04761904761904,
      "grad_norm": 4.348031997680664,
      "learning_rate": 4.859445277361319e-07,
      "loss": 1.1357,
      "step": 2594
    },
    {
      "epoch": 247.14285714285714,
      "grad_norm": 2.821397542953491,
      "learning_rate": 4.861319340329835e-07,
      "loss": 1.1486,
      "step": 2595
    },
    {
      "epoch": 247.23809523809524,
      "grad_norm": 4.46021032333374,
      "learning_rate": 4.86319340329835e-07,
      "loss": 1.1585,
      "step": 2596
    },
    {
      "epoch": 247.33333333333334,
      "grad_norm": 5.806177616119385,
      "learning_rate": 4.865067466266867e-07,
      "loss": 1.1376,
      "step": 2597
    },
    {
      "epoch": 247.42857142857142,
      "grad_norm": 6.550351619720459,
      "learning_rate": 4.866941529235382e-07,
      "loss": 1.1208,
      "step": 2598
    },
    {
      "epoch": 247.52380952380952,
      "grad_norm": 5.049726963043213,
      "learning_rate": 4.868815592203899e-07,
      "loss": 1.1931,
      "step": 2599
    },
    {
      "epoch": 247.61904761904762,
      "grad_norm": 12.2033052444458,
      "learning_rate": 4.870689655172414e-07,
      "loss": 1.1471,
      "step": 2600
    },
    {
      "epoch": 247.71428571428572,
      "grad_norm": 10.302976608276367,
      "learning_rate": 4.87256371814093e-07,
      "loss": 1.1419,
      "step": 2601
    },
    {
      "epoch": 247.8095238095238,
      "grad_norm": 7.384419918060303,
      "learning_rate": 4.874437781109446e-07,
      "loss": 1.1408,
      "step": 2602
    },
    {
      "epoch": 247.9047619047619,
      "grad_norm": 7.357702255249023,
      "learning_rate": 4.876311844077961e-07,
      "loss": 1.1146,
      "step": 2603
    },
    {
      "epoch": 248.0,
      "grad_norm": 14.828198432922363,
      "learning_rate": 4.878185907046478e-07,
      "loss": 1.1099,
      "step": 2604
    },
    {
      "epoch": 248.0,
      "eval_accuracy": 0.7193705507680779,
      "eval_f1": 0.7756813417190775,
      "eval_loss": 0.5678009986877441,
      "eval_precision": 0.7214484679665738,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.7703428324697754,
      "eval_runtime": 1.1966,
      "eval_samples_per_second": 2230.494,
      "eval_steps_per_second": 5.014,
      "step": 2604
    },
    {
      "epoch": 248.0952380952381,
      "grad_norm": 7.183742523193359,
      "learning_rate": 4.880059970014993e-07,
      "loss": 1.1776,
      "step": 2605
    },
    {
      "epoch": 248.1904761904762,
      "grad_norm": 8.25802230834961,
      "learning_rate": 4.881934032983509e-07,
      "loss": 1.1201,
      "step": 2606
    },
    {
      "epoch": 248.28571428571428,
      "grad_norm": 4.379749298095703,
      "learning_rate": 4.883808095952024e-07,
      "loss": 1.1509,
      "step": 2607
    },
    {
      "epoch": 248.38095238095238,
      "grad_norm": 4.609863758087158,
      "learning_rate": 4.88568215892054e-07,
      "loss": 1.1903,
      "step": 2608
    },
    {
      "epoch": 248.47619047619048,
      "grad_norm": 5.011764049530029,
      "learning_rate": 4.887556221889055e-07,
      "loss": 1.1881,
      "step": 2609
    },
    {
      "epoch": 248.57142857142858,
      "grad_norm": 4.677763938903809,
      "learning_rate": 4.889430284857571e-07,
      "loss": 1.1537,
      "step": 2610
    },
    {
      "epoch": 248.66666666666666,
      "grad_norm": 8.4342041015625,
      "learning_rate": 4.891304347826088e-07,
      "loss": 1.1329,
      "step": 2611
    },
    {
      "epoch": 248.76190476190476,
      "grad_norm": 6.544559478759766,
      "learning_rate": 4.893178410794603e-07,
      "loss": 1.1114,
      "step": 2612
    },
    {
      "epoch": 248.85714285714286,
      "grad_norm": 4.162692070007324,
      "learning_rate": 4.895052473763119e-07,
      "loss": 1.1318,
      "step": 2613
    },
    {
      "epoch": 248.95238095238096,
      "grad_norm": 3.0155041217803955,
      "learning_rate": 4.896926536731634e-07,
      "loss": 1.1124,
      "step": 2614
    },
    {
      "epoch": 248.95238095238096,
      "eval_accuracy": 0.7219932559010865,
      "eval_f1": 0.7798219584569733,
      "eval_loss": 0.5675429105758667,
      "eval_precision": 0.71960569550931,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.7711234887737478,
      "eval_runtime": 1.1373,
      "eval_samples_per_second": 2346.798,
      "eval_steps_per_second": 5.276,
      "step": 2614
    },
    {
      "epoch": 249.04761904761904,
      "grad_norm": 3.5584659576416016,
      "learning_rate": 4.89880059970015e-07,
      "loss": 1.1268,
      "step": 2615
    },
    {
      "epoch": 249.14285714285714,
      "grad_norm": 5.403181076049805,
      "learning_rate": 4.900674662668665e-07,
      "loss": 1.1221,
      "step": 2616
    },
    {
      "epoch": 249.23809523809524,
      "grad_norm": 5.95130729675293,
      "learning_rate": 4.902548725637181e-07,
      "loss": 1.1212,
      "step": 2617
    },
    {
      "epoch": 249.33333333333334,
      "grad_norm": 2.7318475246429443,
      "learning_rate": 4.904422788605697e-07,
      "loss": 1.1541,
      "step": 2618
    },
    {
      "epoch": 249.42857142857142,
      "grad_norm": 7.354856967926025,
      "learning_rate": 4.906296851574213e-07,
      "loss": 1.139,
      "step": 2619
    },
    {
      "epoch": 249.52380952380952,
      "grad_norm": 4.743566989898682,
      "learning_rate": 4.908170914542729e-07,
      "loss": 1.1343,
      "step": 2620
    },
    {
      "epoch": 249.61904761904762,
      "grad_norm": 3.644205093383789,
      "learning_rate": 4.910044977511244e-07,
      "loss": 1.1356,
      "step": 2621
    },
    {
      "epoch": 249.71428571428572,
      "grad_norm": 3.6012938022613525,
      "learning_rate": 4.911919040479761e-07,
      "loss": 1.1496,
      "step": 2622
    },
    {
      "epoch": 249.8095238095238,
      "grad_norm": 5.7909650802612305,
      "learning_rate": 4.913793103448276e-07,
      "loss": 1.1433,
      "step": 2623
    },
    {
      "epoch": 249.9047619047619,
      "grad_norm": 5.431772708892822,
      "learning_rate": 4.915667166416792e-07,
      "loss": 1.1625,
      "step": 2624
    },
    {
      "epoch": 250.0,
      "grad_norm": 4.164554595947266,
      "learning_rate": 4.917541229385308e-07,
      "loss": 1.1322,
      "step": 2625
    },
    {
      "epoch": 250.0,
      "eval_accuracy": 0.7219932559010865,
      "eval_f1": 0.7770432692307693,
      "eval_loss": 0.5667824149131775,
      "eval_precision": 0.7247757847533632,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7719421416234887,
      "eval_runtime": 1.3328,
      "eval_samples_per_second": 2002.626,
      "eval_steps_per_second": 4.502,
      "step": 2625
    },
    {
      "epoch": 250.0952380952381,
      "grad_norm": 5.059668064117432,
      "learning_rate": 4.919415292353824e-07,
      "loss": 1.1411,
      "step": 2626
    },
    {
      "epoch": 250.1904761904762,
      "grad_norm": 4.866458892822266,
      "learning_rate": 4.92128935532234e-07,
      "loss": 1.1845,
      "step": 2627
    },
    {
      "epoch": 250.28571428571428,
      "grad_norm": 8.36328125,
      "learning_rate": 4.923163418290855e-07,
      "loss": 1.1561,
      "step": 2628
    },
    {
      "epoch": 250.38095238095238,
      "grad_norm": 2.8156042098999023,
      "learning_rate": 4.925037481259371e-07,
      "loss": 1.1703,
      "step": 2629
    },
    {
      "epoch": 250.47619047619048,
      "grad_norm": 4.026419162750244,
      "learning_rate": 4.926911544227886e-07,
      "loss": 1.1547,
      "step": 2630
    },
    {
      "epoch": 250.57142857142858,
      "grad_norm": 7.374605178833008,
      "learning_rate": 4.928785607196402e-07,
      "loss": 1.1162,
      "step": 2631
    },
    {
      "epoch": 250.66666666666666,
      "grad_norm": 4.192410945892334,
      "learning_rate": 4.930659670164918e-07,
      "loss": 1.1401,
      "step": 2632
    },
    {
      "epoch": 250.76190476190476,
      "grad_norm": 8.36020565032959,
      "learning_rate": 4.932533733133434e-07,
      "loss": 1.1354,
      "step": 2633
    },
    {
      "epoch": 250.85714285714286,
      "grad_norm": 4.311028003692627,
      "learning_rate": 4.934407796101949e-07,
      "loss": 1.1371,
      "step": 2634
    },
    {
      "epoch": 250.95238095238096,
      "grad_norm": 4.741408348083496,
      "learning_rate": 4.936281859070465e-07,
      "loss": 1.1553,
      "step": 2635
    },
    {
      "epoch": 250.95238095238096,
      "eval_accuracy": 0.7223679280629449,
      "eval_f1": 0.7796610169491526,
      "eval_loss": 0.5665171146392822,
      "eval_precision": 0.7207256734469488,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7729004029936672,
      "eval_runtime": 1.2164,
      "eval_samples_per_second": 2194.164,
      "eval_steps_per_second": 4.933,
      "step": 2635
    },
    {
      "epoch": 251.04761904761904,
      "grad_norm": 3.4054064750671387,
      "learning_rate": 4.93815592203898e-07,
      "loss": 1.1093,
      "step": 2636
    },
    {
      "epoch": 251.14285714285714,
      "grad_norm": 4.24013614654541,
      "learning_rate": 4.940029985007496e-07,
      "loss": 1.0858,
      "step": 2637
    },
    {
      "epoch": 251.23809523809524,
      "grad_norm": 6.393743515014648,
      "learning_rate": 4.941904047976012e-07,
      "loss": 1.19,
      "step": 2638
    },
    {
      "epoch": 251.33333333333334,
      "grad_norm": 3.3348259925842285,
      "learning_rate": 4.943778110944528e-07,
      "loss": 1.1088,
      "step": 2639
    },
    {
      "epoch": 251.42857142857142,
      "grad_norm": 7.557923316955566,
      "learning_rate": 4.945652173913044e-07,
      "loss": 1.1138,
      "step": 2640
    },
    {
      "epoch": 251.52380952380952,
      "grad_norm": 5.54427433013916,
      "learning_rate": 4.947526236881559e-07,
      "loss": 1.1517,
      "step": 2641
    },
    {
      "epoch": 251.61904761904762,
      "grad_norm": 5.514889717102051,
      "learning_rate": 4.949400299850075e-07,
      "loss": 1.1478,
      "step": 2642
    },
    {
      "epoch": 251.71428571428572,
      "grad_norm": 3.9975523948669434,
      "learning_rate": 4.95127436281859e-07,
      "loss": 1.1301,
      "step": 2643
    },
    {
      "epoch": 251.8095238095238,
      "grad_norm": 3.9698591232299805,
      "learning_rate": 4.953148425787107e-07,
      "loss": 1.1489,
      "step": 2644
    },
    {
      "epoch": 251.9047619047619,
      "grad_norm": 9.93479061126709,
      "learning_rate": 4.955022488755622e-07,
      "loss": 1.1834,
      "step": 2645
    },
    {
      "epoch": 252.0,
      "grad_norm": 7.301760196685791,
      "learning_rate": 4.956896551724138e-07,
      "loss": 1.1597,
      "step": 2646
    },
    {
      "epoch": 252.0,
      "eval_accuracy": 0.7197452229299363,
      "eval_f1": 0.7740181268882175,
      "eval_loss": 0.5662791728973389,
      "eval_precision": 0.7253680634201586,
      "eval_recall": 0.8296632124352331,
      "eval_roc_auc": 0.7728767990788716,
      "eval_runtime": 1.159,
      "eval_samples_per_second": 2302.753,
      "eval_steps_per_second": 5.177,
      "step": 2646
    },
    {
      "epoch": 252.0952380952381,
      "grad_norm": 9.864036560058594,
      "learning_rate": 4.958770614692655e-07,
      "loss": 1.12,
      "step": 2647
    },
    {
      "epoch": 252.1904761904762,
      "grad_norm": 7.417358875274658,
      "learning_rate": 4.96064467766117e-07,
      "loss": 1.1745,
      "step": 2648
    },
    {
      "epoch": 252.28571428571428,
      "grad_norm": 9.244400024414062,
      "learning_rate": 4.962518740629686e-07,
      "loss": 1.1274,
      "step": 2649
    },
    {
      "epoch": 252.38095238095238,
      "grad_norm": 5.600161075592041,
      "learning_rate": 4.964392803598201e-07,
      "loss": 1.1308,
      "step": 2650
    },
    {
      "epoch": 252.47619047619048,
      "grad_norm": 3.25844144821167,
      "learning_rate": 4.966266866566717e-07,
      "loss": 1.1433,
      "step": 2651
    },
    {
      "epoch": 252.57142857142858,
      "grad_norm": 5.3738203048706055,
      "learning_rate": 4.968140929535232e-07,
      "loss": 1.1348,
      "step": 2652
    },
    {
      "epoch": 252.66666666666666,
      "grad_norm": 4.872201919555664,
      "learning_rate": 4.970014992503749e-07,
      "loss": 1.1805,
      "step": 2653
    },
    {
      "epoch": 252.76190476190476,
      "grad_norm": 3.1699211597442627,
      "learning_rate": 4.971889055472264e-07,
      "loss": 1.1138,
      "step": 2654
    },
    {
      "epoch": 252.85714285714286,
      "grad_norm": 8.699714660644531,
      "learning_rate": 4.97376311844078e-07,
      "loss": 1.1406,
      "step": 2655
    },
    {
      "epoch": 252.95238095238096,
      "grad_norm": 4.888248443603516,
      "learning_rate": 4.975637181409296e-07,
      "loss": 1.1397,
      "step": 2656
    },
    {
      "epoch": 252.95238095238096,
      "eval_accuracy": 0.7208692394155114,
      "eval_f1": 0.7792592592592592,
      "eval_loss": 0.5662895441055298,
      "eval_precision": 0.718186783178591,
      "eval_recall": 0.8516839378238342,
      "eval_roc_auc": 0.77301065054692,
      "eval_runtime": 1.2162,
      "eval_samples_per_second": 2194.61,
      "eval_steps_per_second": 4.934,
      "step": 2656
    },
    {
      "epoch": 253.04761904761904,
      "grad_norm": 6.35478401184082,
      "learning_rate": 4.977511244377811e-07,
      "loss": 1.1514,
      "step": 2657
    },
    {
      "epoch": 253.14285714285714,
      "grad_norm": 8.917704582214355,
      "learning_rate": 4.979385307346327e-07,
      "loss": 1.1433,
      "step": 2658
    },
    {
      "epoch": 253.23809523809524,
      "grad_norm": 4.451404571533203,
      "learning_rate": 4.981259370314842e-07,
      "loss": 1.1446,
      "step": 2659
    },
    {
      "epoch": 253.33333333333334,
      "grad_norm": 5.947729110717773,
      "learning_rate": 4.983133433283359e-07,
      "loss": 1.1874,
      "step": 2660
    },
    {
      "epoch": 253.42857142857142,
      "grad_norm": 4.357635498046875,
      "learning_rate": 4.985007496251874e-07,
      "loss": 1.1575,
      "step": 2661
    },
    {
      "epoch": 253.52380952380952,
      "grad_norm": 2.9355976581573486,
      "learning_rate": 4.98688155922039e-07,
      "loss": 1.1786,
      "step": 2662
    },
    {
      "epoch": 253.61904761904762,
      "grad_norm": 6.246455669403076,
      "learning_rate": 4.988755622188905e-07,
      "loss": 1.1795,
      "step": 2663
    },
    {
      "epoch": 253.71428571428572,
      "grad_norm": 3.440420150756836,
      "learning_rate": 4.990629685157421e-07,
      "loss": 1.1208,
      "step": 2664
    },
    {
      "epoch": 253.8095238095238,
      "grad_norm": 7.903586387634277,
      "learning_rate": 4.992503748125937e-07,
      "loss": 1.1363,
      "step": 2665
    },
    {
      "epoch": 253.9047619047619,
      "grad_norm": 8.214106559753418,
      "learning_rate": 4.994377811094453e-07,
      "loss": 1.1367,
      "step": 2666
    },
    {
      "epoch": 254.0,
      "grad_norm": 7.515315532684326,
      "learning_rate": 4.996251874062969e-07,
      "loss": 1.103,
      "step": 2667
    },
    {
      "epoch": 254.0,
      "eval_accuracy": 0.7197452229299363,
      "eval_f1": 0.7765830346475507,
      "eval_loss": 0.5663872957229614,
      "eval_precision": 0.720620842572062,
      "eval_recall": 0.8419689119170984,
      "eval_roc_auc": 0.7728983880253311,
      "eval_runtime": 1.2631,
      "eval_samples_per_second": 2112.992,
      "eval_steps_per_second": 4.75,
      "step": 2667
    },
    {
      "epoch": 254.0952380952381,
      "grad_norm": 4.3371782302856445,
      "learning_rate": 4.998125937031484e-07,
      "loss": 1.1568,
      "step": 2668
    },
    {
      "epoch": 254.1904761904762,
      "grad_norm": 5.310108184814453,
      "learning_rate": 5.000000000000001e-07,
      "loss": 1.1713,
      "step": 2669
    },
    {
      "epoch": 254.28571428571428,
      "grad_norm": 6.331097602844238,
      "learning_rate": 5.001874062968516e-07,
      "loss": 1.16,
      "step": 2670
    },
    {
      "epoch": 254.38095238095238,
      "grad_norm": 4.3251214027404785,
      "learning_rate": 5.003748125937032e-07,
      "loss": 1.118,
      "step": 2671
    },
    {
      "epoch": 254.47619047619048,
      "grad_norm": 5.350346088409424,
      "learning_rate": 5.005622188905547e-07,
      "loss": 1.1931,
      "step": 2672
    },
    {
      "epoch": 254.57142857142858,
      "grad_norm": 4.380564212799072,
      "learning_rate": 5.007496251874063e-07,
      "loss": 1.1534,
      "step": 2673
    },
    {
      "epoch": 254.66666666666666,
      "grad_norm": 4.470727443695068,
      "learning_rate": 5.00937031484258e-07,
      "loss": 1.172,
      "step": 2674
    },
    {
      "epoch": 254.76190476190476,
      "grad_norm": 5.436304092407227,
      "learning_rate": 5.011244377811095e-07,
      "loss": 1.1059,
      "step": 2675
    },
    {
      "epoch": 254.85714285714286,
      "grad_norm": 5.644359588623047,
      "learning_rate": 5.013118440779611e-07,
      "loss": 1.1502,
      "step": 2676
    },
    {
      "epoch": 254.95238095238096,
      "grad_norm": 4.652618408203125,
      "learning_rate": 5.014992503748126e-07,
      "loss": 1.1219,
      "step": 2677
    },
    {
      "epoch": 254.95238095238096,
      "eval_accuracy": 0.7216185837392282,
      "eval_f1": 0.7799822327509623,
      "eval_loss": 0.5664767622947693,
      "eval_precision": 0.7184942716857611,
      "eval_recall": 0.852979274611399,
      "eval_roc_auc": 0.7731445020149683,
      "eval_runtime": 1.3646,
      "eval_samples_per_second": 1955.908,
      "eval_steps_per_second": 4.397,
      "step": 2677
    },
    {
      "epoch": 255.04761904761904,
      "grad_norm": 6.856327056884766,
      "learning_rate": 5.016866566716642e-07,
      "loss": 1.1468,
      "step": 2678
    },
    {
      "epoch": 255.14285714285714,
      "grad_norm": 8.316398620605469,
      "learning_rate": 5.018740629685157e-07,
      "loss": 1.1719,
      "step": 2679
    },
    {
      "epoch": 255.23809523809524,
      "grad_norm": 11.568921089172363,
      "learning_rate": 5.020614692653673e-07,
      "loss": 1.1473,
      "step": 2680
    },
    {
      "epoch": 255.33333333333334,
      "grad_norm": 2.822263240814209,
      "learning_rate": 5.02248875562219e-07,
      "loss": 1.1532,
      "step": 2681
    },
    {
      "epoch": 255.42857142857142,
      "grad_norm": 7.216583728790283,
      "learning_rate": 5.024362818590705e-07,
      "loss": 1.122,
      "step": 2682
    },
    {
      "epoch": 255.52380952380952,
      "grad_norm": 6.852657318115234,
      "learning_rate": 5.026236881559221e-07,
      "loss": 1.0884,
      "step": 2683
    },
    {
      "epoch": 255.61904761904762,
      "grad_norm": 4.248050212860107,
      "learning_rate": 5.028110944527736e-07,
      "loss": 1.1541,
      "step": 2684
    },
    {
      "epoch": 255.71428571428572,
      "grad_norm": 4.815935134887695,
      "learning_rate": 5.029985007496252e-07,
      "loss": 1.1383,
      "step": 2685
    },
    {
      "epoch": 255.8095238095238,
      "grad_norm": 4.0007476806640625,
      "learning_rate": 5.031859070464767e-07,
      "loss": 1.1343,
      "step": 2686
    },
    {
      "epoch": 255.9047619047619,
      "grad_norm": 9.387462615966797,
      "learning_rate": 5.033733133433283e-07,
      "loss": 1.2121,
      "step": 2687
    },
    {
      "epoch": 256.0,
      "grad_norm": 5.414767265319824,
      "learning_rate": 5.035607196401799e-07,
      "loss": 1.157,
      "step": 2688
    },
    {
      "epoch": 256.0,
      "eval_accuracy": 0.7234919445485201,
      "eval_f1": 0.7775768535262206,
      "eval_loss": 0.5660356283187866,
      "eval_precision": 0.7271702367531003,
      "eval_recall": 0.8354922279792746,
      "eval_roc_auc": 0.7734519286125504,
      "eval_runtime": 1.1637,
      "eval_samples_per_second": 2293.636,
      "eval_steps_per_second": 5.156,
      "step": 2688
    },
    {
      "epoch": 256.0952380952381,
      "grad_norm": 6.968070030212402,
      "learning_rate": 5.037481259370315e-07,
      "loss": 1.1234,
      "step": 2689
    },
    {
      "epoch": 256.1904761904762,
      "grad_norm": 5.248779296875,
      "learning_rate": 5.03935532233883e-07,
      "loss": 1.1436,
      "step": 2690
    },
    {
      "epoch": 256.2857142857143,
      "grad_norm": 14.160701751708984,
      "learning_rate": 5.041229385307347e-07,
      "loss": 1.1447,
      "step": 2691
    },
    {
      "epoch": 256.3809523809524,
      "grad_norm": 4.778572082519531,
      "learning_rate": 5.043103448275863e-07,
      "loss": 1.1345,
      "step": 2692
    },
    {
      "epoch": 256.4761904761905,
      "grad_norm": 3.8514914512634277,
      "learning_rate": 5.044977511244378e-07,
      "loss": 1.1497,
      "step": 2693
    },
    {
      "epoch": 256.57142857142856,
      "grad_norm": 7.788860321044922,
      "learning_rate": 5.046851574212894e-07,
      "loss": 1.1367,
      "step": 2694
    },
    {
      "epoch": 256.6666666666667,
      "grad_norm": 6.130805015563965,
      "learning_rate": 5.04872563718141e-07,
      "loss": 1.173,
      "step": 2695
    },
    {
      "epoch": 256.76190476190476,
      "grad_norm": 6.780652046203613,
      "learning_rate": 5.050599700149926e-07,
      "loss": 1.1691,
      "step": 2696
    },
    {
      "epoch": 256.85714285714283,
      "grad_norm": 6.14133358001709,
      "learning_rate": 5.052473763118441e-07,
      "loss": 1.1414,
      "step": 2697
    },
    {
      "epoch": 256.95238095238096,
      "grad_norm": 4.445262908935547,
      "learning_rate": 5.054347826086957e-07,
      "loss": 1.189,
      "step": 2698
    },
    {
      "epoch": 256.95238095238096,
      "eval_accuracy": 0.7234919445485201,
      "eval_f1": 0.7799642218246869,
      "eval_loss": 0.5659570693969727,
      "eval_precision": 0.7226519337016575,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7735716753022452,
      "eval_runtime": 1.2354,
      "eval_samples_per_second": 2160.474,
      "eval_steps_per_second": 4.857,
      "step": 2698
    },
    {
      "epoch": 257.04761904761904,
      "grad_norm": 8.566206932067871,
      "learning_rate": 5.056221889055472e-07,
      "loss": 1.107,
      "step": 2699
    },
    {
      "epoch": 257.14285714285717,
      "grad_norm": 3.2759463787078857,
      "learning_rate": 5.058095952023988e-07,
      "loss": 1.1409,
      "step": 2700
    },
    {
      "epoch": 257.23809523809524,
      "grad_norm": 4.49222993850708,
      "learning_rate": 5.059970014992504e-07,
      "loss": 1.1603,
      "step": 2701
    },
    {
      "epoch": 257.3333333333333,
      "grad_norm": 3.603001594543457,
      "learning_rate": 5.06184407796102e-07,
      "loss": 1.1433,
      "step": 2702
    },
    {
      "epoch": 257.42857142857144,
      "grad_norm": 5.98636531829834,
      "learning_rate": 5.063718140929536e-07,
      "loss": 1.1373,
      "step": 2703
    },
    {
      "epoch": 257.5238095238095,
      "grad_norm": 6.03758430480957,
      "learning_rate": 5.065592203898051e-07,
      "loss": 1.1238,
      "step": 2704
    },
    {
      "epoch": 257.6190476190476,
      "grad_norm": 6.095311641693115,
      "learning_rate": 5.067466266866567e-07,
      "loss": 1.1018,
      "step": 2705
    },
    {
      "epoch": 257.7142857142857,
      "grad_norm": 3.1190407276153564,
      "learning_rate": 5.069340329835082e-07,
      "loss": 1.1761,
      "step": 2706
    },
    {
      "epoch": 257.8095238095238,
      "grad_norm": 4.101867198944092,
      "learning_rate": 5.071214392803598e-07,
      "loss": 1.1582,
      "step": 2707
    },
    {
      "epoch": 257.9047619047619,
      "grad_norm": 4.426729679107666,
      "learning_rate": 5.073088455772113e-07,
      "loss": 1.1002,
      "step": 2708
    },
    {
      "epoch": 258.0,
      "grad_norm": 4.8217668533325195,
      "learning_rate": 5.07496251874063e-07,
      "loss": 1.1547,
      "step": 2709
    },
    {
      "epoch": 258.0,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.7850632167009703,
      "eval_loss": 0.566510796546936,
      "eval_precision": 0.7189014539579968,
      "eval_recall": 0.8646373056994818,
      "eval_roc_auc": 0.7735877950489349,
      "eval_runtime": 1.1537,
      "eval_samples_per_second": 2313.354,
      "eval_steps_per_second": 5.2,
      "step": 2709
    },
    {
      "epoch": 258.0952380952381,
      "grad_norm": 11.279150009155273,
      "learning_rate": 5.076836581709146e-07,
      "loss": 1.1437,
      "step": 2710
    },
    {
      "epoch": 258.1904761904762,
      "grad_norm": 6.843417167663574,
      "learning_rate": 5.078710644677661e-07,
      "loss": 1.1344,
      "step": 2711
    },
    {
      "epoch": 258.2857142857143,
      "grad_norm": 4.734285354614258,
      "learning_rate": 5.080584707646177e-07,
      "loss": 1.1491,
      "step": 2712
    },
    {
      "epoch": 258.3809523809524,
      "grad_norm": 6.189227104187012,
      "learning_rate": 5.082458770614692e-07,
      "loss": 1.1561,
      "step": 2713
    },
    {
      "epoch": 258.4761904761905,
      "grad_norm": 5.562397003173828,
      "learning_rate": 5.084332833583209e-07,
      "loss": 1.1525,
      "step": 2714
    },
    {
      "epoch": 258.57142857142856,
      "grad_norm": 4.941099166870117,
      "learning_rate": 5.086206896551724e-07,
      "loss": 1.1536,
      "step": 2715
    },
    {
      "epoch": 258.6666666666667,
      "grad_norm": 6.65519905090332,
      "learning_rate": 5.088080959520241e-07,
      "loss": 1.1511,
      "step": 2716
    },
    {
      "epoch": 258.76190476190476,
      "grad_norm": 3.833465576171875,
      "learning_rate": 5.089955022488757e-07,
      "loss": 1.1595,
      "step": 2717
    },
    {
      "epoch": 258.85714285714283,
      "grad_norm": 2.8721892833709717,
      "learning_rate": 5.091829085457272e-07,
      "loss": 1.1216,
      "step": 2718
    },
    {
      "epoch": 258.95238095238096,
      "grad_norm": 7.314364433288574,
      "learning_rate": 5.093703148425788e-07,
      "loss": 1.1316,
      "step": 2719
    },
    {
      "epoch": 258.95238095238096,
      "eval_accuracy": 0.7268639940052454,
      "eval_f1": 0.7823230815168707,
      "eval_loss": 0.565944254398346,
      "eval_precision": 0.7257617728531855,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.773705814622913,
      "eval_runtime": 1.1601,
      "eval_samples_per_second": 2300.601,
      "eval_steps_per_second": 5.172,
      "step": 2719
    },
    {
      "epoch": 259.04761904761904,
      "grad_norm": 4.111525535583496,
      "learning_rate": 5.095577211394303e-07,
      "loss": 1.1183,
      "step": 2720
    },
    {
      "epoch": 259.14285714285717,
      "grad_norm": 4.4706130027771,
      "learning_rate": 5.097451274362819e-07,
      "loss": 1.1062,
      "step": 2721
    },
    {
      "epoch": 259.23809523809524,
      "grad_norm": 5.0127129554748535,
      "learning_rate": 5.099325337331334e-07,
      "loss": 1.1628,
      "step": 2722
    },
    {
      "epoch": 259.3333333333333,
      "grad_norm": 5.392455577850342,
      "learning_rate": 5.101199400299851e-07,
      "loss": 1.165,
      "step": 2723
    },
    {
      "epoch": 259.42857142857144,
      "grad_norm": 11.411964416503906,
      "learning_rate": 5.103073463268366e-07,
      "loss": 1.1307,
      "step": 2724
    },
    {
      "epoch": 259.5238095238095,
      "grad_norm": 9.55789852142334,
      "learning_rate": 5.104947526236882e-07,
      "loss": 1.1287,
      "step": 2725
    },
    {
      "epoch": 259.6190476190476,
      "grad_norm": 4.320809364318848,
      "learning_rate": 5.106821589205397e-07,
      "loss": 1.1133,
      "step": 2726
    },
    {
      "epoch": 259.7142857142857,
      "grad_norm": 6.221428871154785,
      "learning_rate": 5.108695652173913e-07,
      "loss": 1.1716,
      "step": 2727
    },
    {
      "epoch": 259.8095238095238,
      "grad_norm": 6.889163494110107,
      "learning_rate": 5.110569715142429e-07,
      "loss": 1.1539,
      "step": 2728
    },
    {
      "epoch": 259.9047619047619,
      "grad_norm": 7.771426200866699,
      "learning_rate": 5.112443778110944e-07,
      "loss": 1.1633,
      "step": 2729
    },
    {
      "epoch": 260.0,
      "grad_norm": 7.223512172698975,
      "learning_rate": 5.114317841079461e-07,
      "loss": 1.1894,
      "step": 2730
    },
    {
      "epoch": 260.0,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.7823757070556713,
      "eval_loss": 0.5655758380889893,
      "eval_precision": 0.7239669421487603,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.7740702360391479,
      "eval_runtime": 1.2076,
      "eval_samples_per_second": 2210.143,
      "eval_steps_per_second": 4.968,
      "step": 2730
    },
    {
      "epoch": 260.0952380952381,
      "grad_norm": 4.196420669555664,
      "learning_rate": 5.116191904047976e-07,
      "loss": 1.1319,
      "step": 2731
    },
    {
      "epoch": 260.1904761904762,
      "grad_norm": 6.068887233734131,
      "learning_rate": 5.118065967016492e-07,
      "loss": 1.1364,
      "step": 2732
    },
    {
      "epoch": 260.2857142857143,
      "grad_norm": 5.328514099121094,
      "learning_rate": 5.119940029985007e-07,
      "loss": 1.1266,
      "step": 2733
    },
    {
      "epoch": 260.3809523809524,
      "grad_norm": 6.16312313079834,
      "learning_rate": 5.121814092953523e-07,
      "loss": 1.1274,
      "step": 2734
    },
    {
      "epoch": 260.4761904761905,
      "grad_norm": 11.435556411743164,
      "learning_rate": 5.123688155922038e-07,
      "loss": 1.2014,
      "step": 2735
    },
    {
      "epoch": 260.57142857142856,
      "grad_norm": 2.7801122665405273,
      "learning_rate": 5.125562218890555e-07,
      "loss": 1.1256,
      "step": 2736
    },
    {
      "epoch": 260.6666666666667,
      "grad_norm": 4.128719806671143,
      "learning_rate": 5.127436281859071e-07,
      "loss": 1.1606,
      "step": 2737
    },
    {
      "epoch": 260.76190476190476,
      "grad_norm": 6.106013298034668,
      "learning_rate": 5.129310344827586e-07,
      "loss": 1.1465,
      "step": 2738
    },
    {
      "epoch": 260.85714285714283,
      "grad_norm": 3.1265065670013428,
      "learning_rate": 5.131184407796103e-07,
      "loss": 1.1276,
      "step": 2739
    },
    {
      "epoch": 260.95238095238096,
      "grad_norm": 4.820446491241455,
      "learning_rate": 5.133058470764618e-07,
      "loss": 1.1364,
      "step": 2740
    },
    {
      "epoch": 260.95238095238096,
      "eval_accuracy": 0.7246159610340952,
      "eval_f1": 0.779742283488163,
      "eval_loss": 0.5656084418296814,
      "eval_precision": 0.7255995538204127,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.773786701208981,
      "eval_runtime": 1.1679,
      "eval_samples_per_second": 2285.247,
      "eval_steps_per_second": 5.137,
      "step": 2740
    },
    {
      "epoch": 261.04761904761904,
      "grad_norm": 6.337623596191406,
      "learning_rate": 5.134932533733134e-07,
      "loss": 1.1083,
      "step": 2741
    },
    {
      "epoch": 261.14285714285717,
      "grad_norm": 5.403284072875977,
      "learning_rate": 5.136806596701649e-07,
      "loss": 1.1444,
      "step": 2742
    },
    {
      "epoch": 261.23809523809524,
      "grad_norm": 6.919364929199219,
      "learning_rate": 5.138680659670165e-07,
      "loss": 1.1273,
      "step": 2743
    },
    {
      "epoch": 261.3333333333333,
      "grad_norm": 3.3351051807403564,
      "learning_rate": 5.140554722638682e-07,
      "loss": 1.1507,
      "step": 2744
    },
    {
      "epoch": 261.42857142857144,
      "grad_norm": 4.848644256591797,
      "learning_rate": 5.142428785607197e-07,
      "loss": 1.1443,
      "step": 2745
    },
    {
      "epoch": 261.5238095238095,
      "grad_norm": 7.278774261474609,
      "learning_rate": 5.144302848575713e-07,
      "loss": 1.1689,
      "step": 2746
    },
    {
      "epoch": 261.6190476190476,
      "grad_norm": 5.584431171417236,
      "learning_rate": 5.146176911544228e-07,
      "loss": 1.1741,
      "step": 2747
    },
    {
      "epoch": 261.7142857142857,
      "grad_norm": 5.8609843254089355,
      "learning_rate": 5.148050974512744e-07,
      "loss": 1.115,
      "step": 2748
    },
    {
      "epoch": 261.8095238095238,
      "grad_norm": 5.493621349334717,
      "learning_rate": 5.149925037481259e-07,
      "loss": 1.1479,
      "step": 2749
    },
    {
      "epoch": 261.9047619047619,
      "grad_norm": 8.412647247314453,
      "learning_rate": 5.151799100449776e-07,
      "loss": 1.1326,
      "step": 2750
    },
    {
      "epoch": 262.0,
      "grad_norm": 4.3439788818359375,
      "learning_rate": 5.153673163418291e-07,
      "loss": 1.1445,
      "step": 2751
    },
    {
      "epoch": 262.0,
      "eval_accuracy": 0.7246159610340952,
      "eval_f1": 0.7820930922027869,
      "eval_loss": 0.5658958554267883,
      "eval_precision": 0.7211591033351559,
      "eval_recall": 0.8542746113989638,
      "eval_roc_auc": 0.7734879101899828,
      "eval_runtime": 1.3142,
      "eval_samples_per_second": 2030.942,
      "eval_steps_per_second": 4.566,
      "step": 2751
    },
    {
      "epoch": 262.0952380952381,
      "grad_norm": 5.292582988739014,
      "learning_rate": 5.155547226386807e-07,
      "loss": 1.136,
      "step": 2752
    },
    {
      "epoch": 262.1904761904762,
      "grad_norm": 6.31008243560791,
      "learning_rate": 5.157421289355322e-07,
      "loss": 1.1214,
      "step": 2753
    },
    {
      "epoch": 262.2857142857143,
      "grad_norm": 6.329434871673584,
      "learning_rate": 5.159295352323838e-07,
      "loss": 1.1915,
      "step": 2754
    },
    {
      "epoch": 262.3809523809524,
      "grad_norm": 9.693968772888184,
      "learning_rate": 5.161169415292354e-07,
      "loss": 1.1609,
      "step": 2755
    },
    {
      "epoch": 262.4761904761905,
      "grad_norm": 5.809243679046631,
      "learning_rate": 5.163043478260869e-07,
      "loss": 1.1662,
      "step": 2756
    },
    {
      "epoch": 262.57142857142856,
      "grad_norm": 3.715272903442383,
      "learning_rate": 5.164917541229386e-07,
      "loss": 1.1101,
      "step": 2757
    },
    {
      "epoch": 262.6666666666667,
      "grad_norm": 3.8494436740875244,
      "learning_rate": 5.166791604197901e-07,
      "loss": 1.1257,
      "step": 2758
    },
    {
      "epoch": 262.76190476190476,
      "grad_norm": 4.125227928161621,
      "learning_rate": 5.168665667166417e-07,
      "loss": 1.1486,
      "step": 2759
    },
    {
      "epoch": 262.85714285714283,
      "grad_norm": 13.442261695861816,
      "learning_rate": 5.170539730134932e-07,
      "loss": 1.1654,
      "step": 2760
    },
    {
      "epoch": 262.95238095238096,
      "grad_norm": 5.7701497077941895,
      "learning_rate": 5.172413793103449e-07,
      "loss": 1.1293,
      "step": 2761
    },
    {
      "epoch": 262.95238095238096,
      "eval_accuracy": 0.7238666167103784,
      "eval_f1": 0.7779451642060862,
      "eval_loss": 0.5653892755508423,
      "eval_precision": 0.7273239436619718,
      "eval_recall": 0.836139896373057,
      "eval_roc_auc": 0.7735898100172712,
      "eval_runtime": 1.1528,
      "eval_samples_per_second": 2315.228,
      "eval_steps_per_second": 5.205,
      "step": 2761
    },
    {
      "epoch": 263.04761904761904,
      "grad_norm": 5.449621200561523,
      "learning_rate": 5.174287856071964e-07,
      "loss": 1.1372,
      "step": 2762
    },
    {
      "epoch": 263.14285714285717,
      "grad_norm": 11.872653007507324,
      "learning_rate": 5.17616191904048e-07,
      "loss": 1.1212,
      "step": 2763
    },
    {
      "epoch": 263.23809523809524,
      "grad_norm": 5.305274486541748,
      "learning_rate": 5.178035982008997e-07,
      "loss": 1.1336,
      "step": 2764
    },
    {
      "epoch": 263.3333333333333,
      "grad_norm": 8.194924354553223,
      "learning_rate": 5.179910044977512e-07,
      "loss": 1.1706,
      "step": 2765
    },
    {
      "epoch": 263.42857142857144,
      "grad_norm": 4.264476776123047,
      "learning_rate": 5.181784107946028e-07,
      "loss": 1.157,
      "step": 2766
    },
    {
      "epoch": 263.5238095238095,
      "grad_norm": 16.339935302734375,
      "learning_rate": 5.183658170914543e-07,
      "loss": 1.1812,
      "step": 2767
    },
    {
      "epoch": 263.6190476190476,
      "grad_norm": 6.533397197723389,
      "learning_rate": 5.185532233883059e-07,
      "loss": 1.1091,
      "step": 2768
    },
    {
      "epoch": 263.7142857142857,
      "grad_norm": 5.343795299530029,
      "learning_rate": 5.187406296851574e-07,
      "loss": 1.138,
      "step": 2769
    },
    {
      "epoch": 263.8095238095238,
      "grad_norm": 7.269284248352051,
      "learning_rate": 5.18928035982009e-07,
      "loss": 1.1334,
      "step": 2770
    },
    {
      "epoch": 263.9047619047619,
      "grad_norm": 6.141402721405029,
      "learning_rate": 5.191154422788607e-07,
      "loss": 1.1346,
      "step": 2771
    },
    {
      "epoch": 264.0,
      "grad_norm": 7.910759449005127,
      "learning_rate": 5.193028485757122e-07,
      "loss": 1.1253,
      "step": 2772
    },
    {
      "epoch": 264.0,
      "eval_accuracy": 0.7268639940052454,
      "eval_f1": 0.7816711590296496,
      "eval_loss": 0.5649216771125793,
      "eval_precision": 0.7270194986072424,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7744035693724812,
      "eval_runtime": 1.1907,
      "eval_samples_per_second": 2241.481,
      "eval_steps_per_second": 5.039,
      "step": 2772
    },
    {
      "epoch": 264.0952380952381,
      "grad_norm": 5.423191070556641,
      "learning_rate": 5.194902548725638e-07,
      "loss": 1.1218,
      "step": 2773
    },
    {
      "epoch": 264.1904761904762,
      "grad_norm": 4.0316243171691895,
      "learning_rate": 5.196776611694153e-07,
      "loss": 1.0887,
      "step": 2774
    },
    {
      "epoch": 264.2857142857143,
      "grad_norm": 8.018619537353516,
      "learning_rate": 5.198650674662669e-07,
      "loss": 1.1859,
      "step": 2775
    },
    {
      "epoch": 264.3809523809524,
      "grad_norm": 8.210404396057129,
      "learning_rate": 5.200524737631184e-07,
      "loss": 1.0862,
      "step": 2776
    },
    {
      "epoch": 264.4761904761905,
      "grad_norm": 11.399775505065918,
      "learning_rate": 5.2023988005997e-07,
      "loss": 1.1147,
      "step": 2777
    },
    {
      "epoch": 264.57142857142856,
      "grad_norm": 11.730525016784668,
      "learning_rate": 5.204272863568216e-07,
      "loss": 1.1992,
      "step": 2778
    },
    {
      "epoch": 264.6666666666667,
      "grad_norm": 5.436350345611572,
      "learning_rate": 5.206146926536732e-07,
      "loss": 1.115,
      "step": 2779
    },
    {
      "epoch": 264.76190476190476,
      "grad_norm": 3.6730194091796875,
      "learning_rate": 5.208020989505247e-07,
      "loss": 1.1266,
      "step": 2780
    },
    {
      "epoch": 264.85714285714283,
      "grad_norm": 10.828516006469727,
      "learning_rate": 5.209895052473763e-07,
      "loss": 1.1695,
      "step": 2781
    },
    {
      "epoch": 264.95238095238096,
      "grad_norm": 6.543045520782471,
      "learning_rate": 5.211769115442279e-07,
      "loss": 1.1905,
      "step": 2782
    },
    {
      "epoch": 264.95238095238096,
      "eval_accuracy": 0.725365305357812,
      "eval_f1": 0.7809979085748432,
      "eval_loss": 0.5646792650222778,
      "eval_precision": 0.7249029395452025,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7749300518134715,
      "eval_runtime": 1.2059,
      "eval_samples_per_second": 2213.365,
      "eval_steps_per_second": 4.976,
      "step": 2782
    },
    {
      "epoch": 265.04761904761904,
      "grad_norm": 9.305048942565918,
      "learning_rate": 5.213643178410794e-07,
      "loss": 1.1608,
      "step": 2783
    },
    {
      "epoch": 265.14285714285717,
      "grad_norm": 8.853373527526855,
      "learning_rate": 5.215517241379311e-07,
      "loss": 1.1278,
      "step": 2784
    },
    {
      "epoch": 265.23809523809524,
      "grad_norm": 4.973694801330566,
      "learning_rate": 5.217391304347826e-07,
      "loss": 1.1237,
      "step": 2785
    },
    {
      "epoch": 265.3333333333333,
      "grad_norm": 4.541196346282959,
      "learning_rate": 5.219265367316343e-07,
      "loss": 1.1741,
      "step": 2786
    },
    {
      "epoch": 265.42857142857144,
      "grad_norm": 4.542008399963379,
      "learning_rate": 5.221139430284858e-07,
      "loss": 1.1403,
      "step": 2787
    },
    {
      "epoch": 265.5238095238095,
      "grad_norm": 4.306549549102783,
      "learning_rate": 5.223013493253374e-07,
      "loss": 1.1615,
      "step": 2788
    },
    {
      "epoch": 265.6190476190476,
      "grad_norm": 11.334341049194336,
      "learning_rate": 5.22488755622189e-07,
      "loss": 1.1244,
      "step": 2789
    },
    {
      "epoch": 265.7142857142857,
      "grad_norm": 3.0584142208099365,
      "learning_rate": 5.226761619190405e-07,
      "loss": 1.1814,
      "step": 2790
    },
    {
      "epoch": 265.8095238095238,
      "grad_norm": 13.463911056518555,
      "learning_rate": 5.228635682158921e-07,
      "loss": 1.1346,
      "step": 2791
    },
    {
      "epoch": 265.9047619047619,
      "grad_norm": 5.293436527252197,
      "learning_rate": 5.230509745127437e-07,
      "loss": 1.1403,
      "step": 2792
    },
    {
      "epoch": 266.0,
      "grad_norm": 3.8362538814544678,
      "learning_rate": 5.232383808095953e-07,
      "loss": 1.1268,
      "step": 2793
    },
    {
      "epoch": 266.0,
      "eval_accuracy": 0.7249906331959536,
      "eval_f1": 0.781547619047619,
      "eval_loss": 0.5646971464157104,
      "eval_precision": 0.7230176211453745,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7751358664363845,
      "eval_runtime": 1.1638,
      "eval_samples_per_second": 2293.433,
      "eval_steps_per_second": 5.156,
      "step": 2793
    },
    {
      "epoch": 266.0952380952381,
      "grad_norm": 11.430266380310059,
      "learning_rate": 5.234257871064468e-07,
      "loss": 1.1498,
      "step": 2794
    },
    {
      "epoch": 266.1904761904762,
      "grad_norm": 14.580435752868652,
      "learning_rate": 5.236131934032984e-07,
      "loss": 1.191,
      "step": 2795
    },
    {
      "epoch": 266.2857142857143,
      "grad_norm": 9.603107452392578,
      "learning_rate": 5.238005997001499e-07,
      "loss": 1.1362,
      "step": 2796
    },
    {
      "epoch": 266.3809523809524,
      "grad_norm": 11.882341384887695,
      "learning_rate": 5.239880059970015e-07,
      "loss": 1.1268,
      "step": 2797
    },
    {
      "epoch": 266.4761904761905,
      "grad_norm": 3.629544973373413,
      "learning_rate": 5.24175412293853e-07,
      "loss": 1.1628,
      "step": 2798
    },
    {
      "epoch": 266.57142857142856,
      "grad_norm": 13.007302284240723,
      "learning_rate": 5.243628185907047e-07,
      "loss": 1.1492,
      "step": 2799
    },
    {
      "epoch": 266.6666666666667,
      "grad_norm": 5.342100620269775,
      "learning_rate": 5.245502248875563e-07,
      "loss": 1.1439,
      "step": 2800
    },
    {
      "epoch": 266.76190476190476,
      "grad_norm": 8.716962814331055,
      "learning_rate": 5.247376311844078e-07,
      "loss": 1.1493,
      "step": 2801
    },
    {
      "epoch": 266.85714285714283,
      "grad_norm": 6.74729585647583,
      "learning_rate": 5.249250374812594e-07,
      "loss": 1.1258,
      "step": 2802
    },
    {
      "epoch": 266.95238095238096,
      "grad_norm": 9.423856735229492,
      "learning_rate": 5.251124437781109e-07,
      "loss": 1.1271,
      "step": 2803
    },
    {
      "epoch": 266.95238095238096,
      "eval_accuracy": 0.7208692394155114,
      "eval_f1": 0.774992449411054,
      "eval_loss": 0.5647932887077332,
      "eval_precision": 0.7260894170911149,
      "eval_recall": 0.8309585492227979,
      "eval_roc_auc": 0.7751534254461716,
      "eval_runtime": 1.1741,
      "eval_samples_per_second": 2273.142,
      "eval_steps_per_second": 5.11,
      "step": 2803
    },
    {
      "epoch": 267.04761904761904,
      "grad_norm": 7.002502918243408,
      "learning_rate": 5.252998500749625e-07,
      "loss": 1.1664,
      "step": 2804
    },
    {
      "epoch": 267.14285714285717,
      "grad_norm": 3.9827327728271484,
      "learning_rate": 5.25487256371814e-07,
      "loss": 1.1584,
      "step": 2805
    },
    {
      "epoch": 267.23809523809524,
      "grad_norm": 9.634724617004395,
      "learning_rate": 5.256746626686657e-07,
      "loss": 1.158,
      "step": 2806
    },
    {
      "epoch": 267.3333333333333,
      "grad_norm": 8.2416353225708,
      "learning_rate": 5.258620689655172e-07,
      "loss": 1.1193,
      "step": 2807
    },
    {
      "epoch": 267.42857142857144,
      "grad_norm": 6.562640190124512,
      "learning_rate": 5.260494752623688e-07,
      "loss": 1.0989,
      "step": 2808
    },
    {
      "epoch": 267.5238095238095,
      "grad_norm": 9.472210884094238,
      "learning_rate": 5.262368815592205e-07,
      "loss": 1.2015,
      "step": 2809
    },
    {
      "epoch": 267.6190476190476,
      "grad_norm": 6.735084533691406,
      "learning_rate": 5.26424287856072e-07,
      "loss": 1.149,
      "step": 2810
    },
    {
      "epoch": 267.7142857142857,
      "grad_norm": 5.654566287994385,
      "learning_rate": 5.266116941529236e-07,
      "loss": 1.1365,
      "step": 2811
    },
    {
      "epoch": 267.8095238095238,
      "grad_norm": 8.3509521484375,
      "learning_rate": 5.267991004497751e-07,
      "loss": 1.1366,
      "step": 2812
    },
    {
      "epoch": 267.9047619047619,
      "grad_norm": 11.307655334472656,
      "learning_rate": 5.269865067466268e-07,
      "loss": 1.1122,
      "step": 2813
    },
    {
      "epoch": 268.0,
      "grad_norm": 5.473557949066162,
      "learning_rate": 5.271739130434783e-07,
      "loss": 1.115,
      "step": 2814
    },
    {
      "epoch": 268.0,
      "eval_accuracy": 0.7238666167103784,
      "eval_f1": 0.7800656520441659,
      "eval_loss": 0.5653632879257202,
      "eval_precision": 0.7232982844493636,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7745020149683361,
      "eval_runtime": 1.225,
      "eval_samples_per_second": 2178.849,
      "eval_steps_per_second": 4.898,
      "step": 2814
    },
    {
      "epoch": 268.0952380952381,
      "grad_norm": 7.1619391441345215,
      "learning_rate": 5.273613193403299e-07,
      "loss": 1.1266,
      "step": 2815
    },
    {
      "epoch": 268.1904761904762,
      "grad_norm": 6.1082329750061035,
      "learning_rate": 5.275487256371815e-07,
      "loss": 1.1865,
      "step": 2816
    },
    {
      "epoch": 268.2857142857143,
      "grad_norm": 5.075636863708496,
      "learning_rate": 5.27736131934033e-07,
      "loss": 1.1453,
      "step": 2817
    },
    {
      "epoch": 268.3809523809524,
      "grad_norm": 16.62962532043457,
      "learning_rate": 5.279235382308846e-07,
      "loss": 1.141,
      "step": 2818
    },
    {
      "epoch": 268.4761904761905,
      "grad_norm": 4.551577091217041,
      "learning_rate": 5.281109445277361e-07,
      "loss": 1.1515,
      "step": 2819
    },
    {
      "epoch": 268.57142857142856,
      "grad_norm": 7.195343494415283,
      "learning_rate": 5.282983508245878e-07,
      "loss": 1.1221,
      "step": 2820
    },
    {
      "epoch": 268.6666666666667,
      "grad_norm": 8.327423095703125,
      "learning_rate": 5.284857571214393e-07,
      "loss": 1.1538,
      "step": 2821
    },
    {
      "epoch": 268.76190476190476,
      "grad_norm": 4.540790557861328,
      "learning_rate": 5.286731634182909e-07,
      "loss": 1.1255,
      "step": 2822
    },
    {
      "epoch": 268.85714285714283,
      "grad_norm": 7.532181262969971,
      "learning_rate": 5.288605697151424e-07,
      "loss": 1.1624,
      "step": 2823
    },
    {
      "epoch": 268.95238095238096,
      "grad_norm": 7.560113906860352,
      "learning_rate": 5.29047976011994e-07,
      "loss": 1.1273,
      "step": 2824
    },
    {
      "epoch": 268.95238095238096,
      "eval_accuracy": 0.725365305357812,
      "eval_f1": 0.7820398453761522,
      "eval_loss": 0.565043568611145,
      "eval_precision": 0.7229246838922485,
      "eval_recall": 0.8516839378238342,
      "eval_roc_auc": 0.7749015544041451,
      "eval_runtime": 1.3209,
      "eval_samples_per_second": 2020.546,
      "eval_steps_per_second": 4.542,
      "step": 2824
    },
    {
      "epoch": 269.04761904761904,
      "grad_norm": 5.054904937744141,
      "learning_rate": 5.292353823088455e-07,
      "loss": 1.1444,
      "step": 2825
    },
    {
      "epoch": 269.14285714285717,
      "grad_norm": 3.899892807006836,
      "learning_rate": 5.294227886056971e-07,
      "loss": 1.1272,
      "step": 2826
    },
    {
      "epoch": 269.23809523809524,
      "grad_norm": 6.381415843963623,
      "learning_rate": 5.296101949025488e-07,
      "loss": 1.1562,
      "step": 2827
    },
    {
      "epoch": 269.3333333333333,
      "grad_norm": 3.797952175140381,
      "learning_rate": 5.297976011994003e-07,
      "loss": 1.166,
      "step": 2828
    },
    {
      "epoch": 269.42857142857144,
      "grad_norm": 6.653491020202637,
      "learning_rate": 5.299850074962519e-07,
      "loss": 1.102,
      "step": 2829
    },
    {
      "epoch": 269.5238095238095,
      "grad_norm": 6.340592384338379,
      "learning_rate": 5.301724137931034e-07,
      "loss": 1.1018,
      "step": 2830
    },
    {
      "epoch": 269.6190476190476,
      "grad_norm": 6.27230978012085,
      "learning_rate": 5.303598200899551e-07,
      "loss": 1.198,
      "step": 2831
    },
    {
      "epoch": 269.7142857142857,
      "grad_norm": 4.86538028717041,
      "learning_rate": 5.305472263868066e-07,
      "loss": 1.1142,
      "step": 2832
    },
    {
      "epoch": 269.8095238095238,
      "grad_norm": 5.868375301361084,
      "learning_rate": 5.307346326836582e-07,
      "loss": 1.1949,
      "step": 2833
    },
    {
      "epoch": 269.9047619047619,
      "grad_norm": 6.007962226867676,
      "learning_rate": 5.309220389805099e-07,
      "loss": 1.0948,
      "step": 2834
    },
    {
      "epoch": 270.0,
      "grad_norm": 5.161440372467041,
      "learning_rate": 5.311094452773614e-07,
      "loss": 1.1596,
      "step": 2835
    },
    {
      "epoch": 270.0,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.7806780678067807,
      "eval_loss": 0.5640901327133179,
      "eval_precision": 0.7272219116825042,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7755400115141048,
      "eval_runtime": 1.1751,
      "eval_samples_per_second": 2271.295,
      "eval_steps_per_second": 5.106,
      "step": 2835
    },
    {
      "epoch": 270.0952380952381,
      "grad_norm": 4.958430290222168,
      "learning_rate": 5.31296851574213e-07,
      "loss": 1.1289,
      "step": 2836
    },
    {
      "epoch": 270.1904761904762,
      "grad_norm": 6.797719955444336,
      "learning_rate": 5.314842578710645e-07,
      "loss": 1.1431,
      "step": 2837
    },
    {
      "epoch": 270.2857142857143,
      "grad_norm": 3.3344295024871826,
      "learning_rate": 5.316716641679161e-07,
      "loss": 1.1098,
      "step": 2838
    },
    {
      "epoch": 270.3809523809524,
      "grad_norm": 3.955629348754883,
      "learning_rate": 5.318590704647676e-07,
      "loss": 1.1829,
      "step": 2839
    },
    {
      "epoch": 270.4761904761905,
      "grad_norm": 5.1240620613098145,
      "learning_rate": 5.320464767616192e-07,
      "loss": 1.1741,
      "step": 2840
    },
    {
      "epoch": 270.57142857142856,
      "grad_norm": 4.2287492752075195,
      "learning_rate": 5.322338830584708e-07,
      "loss": 1.1609,
      "step": 2841
    },
    {
      "epoch": 270.6666666666667,
      "grad_norm": 5.582423210144043,
      "learning_rate": 5.324212893553224e-07,
      "loss": 1.1394,
      "step": 2842
    },
    {
      "epoch": 270.76190476190476,
      "grad_norm": 4.948960781097412,
      "learning_rate": 5.32608695652174e-07,
      "loss": 1.1455,
      "step": 2843
    },
    {
      "epoch": 270.85714285714283,
      "grad_norm": 8.221921920776367,
      "learning_rate": 5.327961019490255e-07,
      "loss": 1.1144,
      "step": 2844
    },
    {
      "epoch": 270.95238095238096,
      "grad_norm": 6.162173271179199,
      "learning_rate": 5.329835082458771e-07,
      "loss": 1.1182,
      "step": 2845
    },
    {
      "epoch": 270.95238095238096,
      "eval_accuracy": 0.7257399775196703,
      "eval_f1": 0.7795180722891566,
      "eval_loss": 0.5640262365341187,
      "eval_precision": 0.7286036036036037,
      "eval_recall": 0.8380829015544041,
      "eval_roc_auc": 0.7757751871042027,
      "eval_runtime": 1.1735,
      "eval_samples_per_second": 2274.469,
      "eval_steps_per_second": 5.113,
      "step": 2845
    },
    {
      "epoch": 271.04761904761904,
      "grad_norm": 6.911406993865967,
      "learning_rate": 5.331709145427286e-07,
      "loss": 1.1601,
      "step": 2846
    },
    {
      "epoch": 271.14285714285717,
      "grad_norm": 5.134574890136719,
      "learning_rate": 5.333583208395802e-07,
      "loss": 1.1173,
      "step": 2847
    },
    {
      "epoch": 271.23809523809524,
      "grad_norm": 3.4689102172851562,
      "learning_rate": 5.335457271364318e-07,
      "loss": 1.1417,
      "step": 2848
    },
    {
      "epoch": 271.3333333333333,
      "grad_norm": 4.0263447761535645,
      "learning_rate": 5.337331334332834e-07,
      "loss": 1.158,
      "step": 2849
    },
    {
      "epoch": 271.42857142857144,
      "grad_norm": 9.338512420654297,
      "learning_rate": 5.339205397301349e-07,
      "loss": 1.149,
      "step": 2850
    },
    {
      "epoch": 271.5238095238095,
      "grad_norm": 6.389646530151367,
      "learning_rate": 5.341079460269865e-07,
      "loss": 1.111,
      "step": 2851
    },
    {
      "epoch": 271.6190476190476,
      "grad_norm": 6.352331638336182,
      "learning_rate": 5.34295352323838e-07,
      "loss": 1.1442,
      "step": 2852
    },
    {
      "epoch": 271.7142857142857,
      "grad_norm": 6.263265132904053,
      "learning_rate": 5.344827586206897e-07,
      "loss": 1.1455,
      "step": 2853
    },
    {
      "epoch": 271.8095238095238,
      "grad_norm": 6.314106464385986,
      "learning_rate": 5.346701649175413e-07,
      "loss": 1.1446,
      "step": 2854
    },
    {
      "epoch": 271.9047619047619,
      "grad_norm": 6.959097385406494,
      "learning_rate": 5.348575712143928e-07,
      "loss": 1.1264,
      "step": 2855
    },
    {
      "epoch": 272.0,
      "grad_norm": 8.125739097595215,
      "learning_rate": 5.350449775112445e-07,
      "loss": 1.1855,
      "step": 2856
    },
    {
      "epoch": 272.0,
      "eval_accuracy": 0.7242412888722368,
      "eval_f1": 0.7789789789789789,
      "eval_loss": 0.5639445185661316,
      "eval_precision": 0.7262038073908175,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.775972941853771,
      "eval_runtime": 1.2494,
      "eval_samples_per_second": 2136.149,
      "eval_steps_per_second": 4.802,
      "step": 2856
    },
    {
      "epoch": 272.0952380952381,
      "grad_norm": 7.242920875549316,
      "learning_rate": 5.35232383808096e-07,
      "loss": 1.167,
      "step": 2857
    },
    {
      "epoch": 272.1904761904762,
      "grad_norm": 6.216794013977051,
      "learning_rate": 5.354197901049476e-07,
      "loss": 1.121,
      "step": 2858
    },
    {
      "epoch": 272.2857142857143,
      "grad_norm": 5.231599807739258,
      "learning_rate": 5.356071964017991e-07,
      "loss": 1.1327,
      "step": 2859
    },
    {
      "epoch": 272.3809523809524,
      "grad_norm": 8.307430267333984,
      "learning_rate": 5.357946026986507e-07,
      "loss": 1.1287,
      "step": 2860
    },
    {
      "epoch": 272.4761904761905,
      "grad_norm": 10.797221183776855,
      "learning_rate": 5.359820089955022e-07,
      "loss": 1.1367,
      "step": 2861
    },
    {
      "epoch": 272.57142857142856,
      "grad_norm": 4.599208354949951,
      "learning_rate": 5.361694152923539e-07,
      "loss": 1.12,
      "step": 2862
    },
    {
      "epoch": 272.6666666666667,
      "grad_norm": 4.301135063171387,
      "learning_rate": 5.363568215892055e-07,
      "loss": 1.121,
      "step": 2863
    },
    {
      "epoch": 272.76190476190476,
      "grad_norm": 3.4644644260406494,
      "learning_rate": 5.36544227886057e-07,
      "loss": 1.151,
      "step": 2864
    },
    {
      "epoch": 272.85714285714283,
      "grad_norm": 7.4475274085998535,
      "learning_rate": 5.367316341829086e-07,
      "loss": 1.1208,
      "step": 2865
    },
    {
      "epoch": 272.95238095238096,
      "grad_norm": 5.376792907714844,
      "learning_rate": 5.369190404797601e-07,
      "loss": 1.1535,
      "step": 2866
    },
    {
      "epoch": 272.95238095238096,
      "eval_accuracy": 0.7242412888722368,
      "eval_f1": 0.7804295942720764,
      "eval_loss": 0.5637869834899902,
      "eval_precision": 0.7234513274336283,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.776422855497985,
      "eval_runtime": 1.3836,
      "eval_samples_per_second": 1928.965,
      "eval_steps_per_second": 4.336,
      "step": 2866
    },
    {
      "epoch": 273.04761904761904,
      "grad_norm": 11.990949630737305,
      "learning_rate": 5.371064467766117e-07,
      "loss": 1.1783,
      "step": 2867
    },
    {
      "epoch": 273.14285714285717,
      "grad_norm": 8.884700775146484,
      "learning_rate": 5.372938530734632e-07,
      "loss": 1.1722,
      "step": 2868
    },
    {
      "epoch": 273.23809523809524,
      "grad_norm": 4.0039777755737305,
      "learning_rate": 5.374812593703149e-07,
      "loss": 1.0928,
      "step": 2869
    },
    {
      "epoch": 273.3333333333333,
      "grad_norm": 5.6337738037109375,
      "learning_rate": 5.376686656671665e-07,
      "loss": 1.1199,
      "step": 2870
    },
    {
      "epoch": 273.42857142857144,
      "grad_norm": 3.9249324798583984,
      "learning_rate": 5.37856071964018e-07,
      "loss": 1.1054,
      "step": 2871
    },
    {
      "epoch": 273.5238095238095,
      "grad_norm": 6.832759380340576,
      "learning_rate": 5.380434782608696e-07,
      "loss": 1.1588,
      "step": 2872
    },
    {
      "epoch": 273.6190476190476,
      "grad_norm": 3.857856035232544,
      "learning_rate": 5.382308845577211e-07,
      "loss": 1.1377,
      "step": 2873
    },
    {
      "epoch": 273.7142857142857,
      "grad_norm": 7.15529727935791,
      "learning_rate": 5.384182908545727e-07,
      "loss": 1.1736,
      "step": 2874
    },
    {
      "epoch": 273.8095238095238,
      "grad_norm": 6.371598243713379,
      "learning_rate": 5.386056971514242e-07,
      "loss": 1.1782,
      "step": 2875
    },
    {
      "epoch": 273.9047619047619,
      "grad_norm": 9.470749855041504,
      "learning_rate": 5.387931034482759e-07,
      "loss": 1.1703,
      "step": 2876
    },
    {
      "epoch": 274.0,
      "grad_norm": 5.563154697418213,
      "learning_rate": 5.389805097451274e-07,
      "loss": 1.1281,
      "step": 2877
    },
    {
      "epoch": 274.0,
      "eval_accuracy": 0.7238666167103784,
      "eval_f1": 0.7783458646616541,
      "eval_loss": 0.5636013150215149,
      "eval_precision": 0.7265581134194273,
      "eval_recall": 0.8380829015544041,
      "eval_roc_auc": 0.7764179620034543,
      "eval_runtime": 1.1257,
      "eval_samples_per_second": 2370.982,
      "eval_steps_per_second": 5.33,
      "step": 2877
    },
    {
      "epoch": 274.0952380952381,
      "grad_norm": 8.047298431396484,
      "learning_rate": 5.391679160419791e-07,
      "loss": 1.1498,
      "step": 2878
    },
    {
      "epoch": 274.1904761904762,
      "grad_norm": 4.311565399169922,
      "learning_rate": 5.393553223388307e-07,
      "loss": 1.1662,
      "step": 2879
    },
    {
      "epoch": 274.2857142857143,
      "grad_norm": 11.471311569213867,
      "learning_rate": 5.395427286356822e-07,
      "loss": 1.1045,
      "step": 2880
    },
    {
      "epoch": 274.3809523809524,
      "grad_norm": 10.548619270324707,
      "learning_rate": 5.397301349325338e-07,
      "loss": 1.1346,
      "step": 2881
    },
    {
      "epoch": 274.4761904761905,
      "grad_norm": 8.780191421508789,
      "learning_rate": 5.399175412293853e-07,
      "loss": 1.1225,
      "step": 2882
    },
    {
      "epoch": 274.57142857142856,
      "grad_norm": 3.8586528301239014,
      "learning_rate": 5.40104947526237e-07,
      "loss": 1.1465,
      "step": 2883
    },
    {
      "epoch": 274.6666666666667,
      "grad_norm": 5.38668966293335,
      "learning_rate": 5.402923538230885e-07,
      "loss": 1.1243,
      "step": 2884
    },
    {
      "epoch": 274.76190476190476,
      "grad_norm": 10.542987823486328,
      "learning_rate": 5.404797601199401e-07,
      "loss": 1.1485,
      "step": 2885
    },
    {
      "epoch": 274.85714285714283,
      "grad_norm": 5.546933650970459,
      "learning_rate": 5.406671664167916e-07,
      "loss": 1.1072,
      "step": 2886
    },
    {
      "epoch": 274.95238095238096,
      "grad_norm": 5.900209903717041,
      "learning_rate": 5.408545727136432e-07,
      "loss": 1.1339,
      "step": 2887
    },
    {
      "epoch": 274.95238095238096,
      "eval_accuracy": 0.7276133383289621,
      "eval_f1": 0.7839524517087667,
      "eval_loss": 0.5638872385025024,
      "eval_precision": 0.7243272926963207,
      "eval_recall": 0.8542746113989638,
      "eval_roc_auc": 0.7764245826137018,
      "eval_runtime": 1.2282,
      "eval_samples_per_second": 2173.041,
      "eval_steps_per_second": 4.885,
      "step": 2887
    },
    {
      "epoch": 275.04761904761904,
      "grad_norm": 3.780916452407837,
      "learning_rate": 5.410419790104947e-07,
      "loss": 1.1033,
      "step": 2888
    },
    {
      "epoch": 275.14285714285717,
      "grad_norm": 9.117225646972656,
      "learning_rate": 5.412293853073463e-07,
      "loss": 1.0961,
      "step": 2889
    },
    {
      "epoch": 275.23809523809524,
      "grad_norm": 6.246429920196533,
      "learning_rate": 5.41416791604198e-07,
      "loss": 1.1595,
      "step": 2890
    },
    {
      "epoch": 275.3333333333333,
      "grad_norm": 3.8772833347320557,
      "learning_rate": 5.416041979010495e-07,
      "loss": 1.1602,
      "step": 2891
    },
    {
      "epoch": 275.42857142857144,
      "grad_norm": 11.349824905395508,
      "learning_rate": 5.417916041979011e-07,
      "loss": 1.1747,
      "step": 2892
    },
    {
      "epoch": 275.5238095238095,
      "grad_norm": 7.2987542152404785,
      "learning_rate": 5.419790104947526e-07,
      "loss": 1.1192,
      "step": 2893
    },
    {
      "epoch": 275.6190476190476,
      "grad_norm": 5.3660783767700195,
      "learning_rate": 5.421664167916042e-07,
      "loss": 1.1368,
      "step": 2894
    },
    {
      "epoch": 275.7142857142857,
      "grad_norm": 7.8456010818481445,
      "learning_rate": 5.423538230884557e-07,
      "loss": 1.1483,
      "step": 2895
    },
    {
      "epoch": 275.8095238095238,
      "grad_norm": 3.772411823272705,
      "learning_rate": 5.425412293853073e-07,
      "loss": 1.1487,
      "step": 2896
    },
    {
      "epoch": 275.9047619047619,
      "grad_norm": 4.100811958312988,
      "learning_rate": 5.42728635682159e-07,
      "loss": 1.1666,
      "step": 2897
    },
    {
      "epoch": 276.0,
      "grad_norm": 12.531455039978027,
      "learning_rate": 5.429160419790105e-07,
      "loss": 1.1379,
      "step": 2898
    },
    {
      "epoch": 276.0,
      "eval_accuracy": 0.7249906331959536,
      "eval_f1": 0.7790487658037327,
      "eval_loss": 0.5640032887458801,
      "eval_precision": 0.7277840269966255,
      "eval_recall": 0.8380829015544041,
      "eval_roc_auc": 0.7759821531375934,
      "eval_runtime": 1.1919,
      "eval_samples_per_second": 2239.234,
      "eval_steps_per_second": 5.034,
      "step": 2898
    },
    {
      "epoch": 276.0952380952381,
      "grad_norm": 11.592368125915527,
      "learning_rate": 5.431034482758621e-07,
      "loss": 1.114,
      "step": 2899
    },
    {
      "epoch": 276.1904761904762,
      "grad_norm": 3.5621049404144287,
      "learning_rate": 5.432908545727136e-07,
      "loss": 1.1558,
      "step": 2900
    },
    {
      "epoch": 276.2857142857143,
      "grad_norm": 6.428314685821533,
      "learning_rate": 5.434782608695653e-07,
      "loss": 1.1089,
      "step": 2901
    },
    {
      "epoch": 276.3809523809524,
      "grad_norm": 8.273981094360352,
      "learning_rate": 5.436656671664168e-07,
      "loss": 1.1277,
      "step": 2902
    },
    {
      "epoch": 276.4761904761905,
      "grad_norm": 7.387972354888916,
      "learning_rate": 5.438530734632684e-07,
      "loss": 1.1845,
      "step": 2903
    },
    {
      "epoch": 276.57142857142856,
      "grad_norm": 4.1005706787109375,
      "learning_rate": 5.4404047976012e-07,
      "loss": 1.1568,
      "step": 2904
    },
    {
      "epoch": 276.6666666666667,
      "grad_norm": 4.488372802734375,
      "learning_rate": 5.442278860569716e-07,
      "loss": 1.1023,
      "step": 2905
    },
    {
      "epoch": 276.76190476190476,
      "grad_norm": 5.401954650878906,
      "learning_rate": 5.444152923538232e-07,
      "loss": 1.1355,
      "step": 2906
    },
    {
      "epoch": 276.85714285714283,
      "grad_norm": 7.288771629333496,
      "learning_rate": 5.446026986506747e-07,
      "loss": 1.1368,
      "step": 2907
    },
    {
      "epoch": 276.95238095238096,
      "grad_norm": 7.125945568084717,
      "learning_rate": 5.447901049475263e-07,
      "loss": 1.1767,
      "step": 2908
    },
    {
      "epoch": 276.95238095238096,
      "eval_accuracy": 0.7298613713001124,
      "eval_f1": 0.7866232613199171,
      "eval_loss": 0.564026951789856,
      "eval_precision": 0.7242506811989101,
      "eval_recall": 0.8607512953367875,
      "eval_roc_auc": 0.7765402993667242,
      "eval_runtime": 1.1584,
      "eval_samples_per_second": 2304.062,
      "eval_steps_per_second": 5.18,
      "step": 2908
    },
    {
      "epoch": 277.04761904761904,
      "grad_norm": 11.035928726196289,
      "learning_rate": 5.449775112443778e-07,
      "loss": 1.1625,
      "step": 2909
    },
    {
      "epoch": 277.14285714285717,
      "grad_norm": 4.437666893005371,
      "learning_rate": 5.451649175412294e-07,
      "loss": 1.1427,
      "step": 2910
    },
    {
      "epoch": 277.23809523809524,
      "grad_norm": 4.022305965423584,
      "learning_rate": 5.45352323838081e-07,
      "loss": 1.129,
      "step": 2911
    },
    {
      "epoch": 277.3333333333333,
      "grad_norm": 10.658130645751953,
      "learning_rate": 5.455397301349326e-07,
      "loss": 1.0895,
      "step": 2912
    },
    {
      "epoch": 277.42857142857144,
      "grad_norm": 5.271421909332275,
      "learning_rate": 5.457271364317841e-07,
      "loss": 1.1431,
      "step": 2913
    },
    {
      "epoch": 277.5238095238095,
      "grad_norm": 6.368401050567627,
      "learning_rate": 5.459145427286357e-07,
      "loss": 1.1147,
      "step": 2914
    },
    {
      "epoch": 277.6190476190476,
      "grad_norm": 3.81168532371521,
      "learning_rate": 5.461019490254873e-07,
      "loss": 1.1266,
      "step": 2915
    },
    {
      "epoch": 277.7142857142857,
      "grad_norm": 3.3784971237182617,
      "learning_rate": 5.462893553223388e-07,
      "loss": 1.1751,
      "step": 2916
    },
    {
      "epoch": 277.8095238095238,
      "grad_norm": 5.5185627937316895,
      "learning_rate": 5.464767616191904e-07,
      "loss": 1.1456,
      "step": 2917
    },
    {
      "epoch": 277.9047619047619,
      "grad_norm": 10.223770141601562,
      "learning_rate": 5.46664167916042e-07,
      "loss": 1.1173,
      "step": 2918
    },
    {
      "epoch": 278.0,
      "grad_norm": 5.04381799697876,
      "learning_rate": 5.468515742128936e-07,
      "loss": 1.1602,
      "step": 2919
    },
    {
      "epoch": 278.0,
      "eval_accuracy": 0.7234919445485201,
      "eval_f1": 0.7774427020506635,
      "eval_loss": 0.5633842945098877,
      "eval_precision": 0.7274266365688488,
      "eval_recall": 0.8348445595854922,
      "eval_roc_auc": 0.7765256188831319,
      "eval_runtime": 1.1165,
      "eval_samples_per_second": 2390.568,
      "eval_steps_per_second": 5.374,
      "step": 2919
    },
    {
      "epoch": 278.0952380952381,
      "grad_norm": 4.8481221199035645,
      "learning_rate": 5.470389805097451e-07,
      "loss": 1.1563,
      "step": 2920
    },
    {
      "epoch": 278.1904761904762,
      "grad_norm": 4.2458906173706055,
      "learning_rate": 5.472263868065967e-07,
      "loss": 1.1748,
      "step": 2921
    },
    {
      "epoch": 278.2857142857143,
      "grad_norm": 6.992273807525635,
      "learning_rate": 5.474137931034482e-07,
      "loss": 1.1342,
      "step": 2922
    },
    {
      "epoch": 278.3809523809524,
      "grad_norm": 8.638583183288574,
      "learning_rate": 5.476011994002999e-07,
      "loss": 1.1363,
      "step": 2923
    },
    {
      "epoch": 278.4761904761905,
      "grad_norm": 5.823126316070557,
      "learning_rate": 5.477886056971515e-07,
      "loss": 1.1124,
      "step": 2924
    },
    {
      "epoch": 278.57142857142856,
      "grad_norm": 8.983731269836426,
      "learning_rate": 5.47976011994003e-07,
      "loss": 1.1614,
      "step": 2925
    },
    {
      "epoch": 278.6666666666667,
      "grad_norm": 3.8761818408966064,
      "learning_rate": 5.481634182908547e-07,
      "loss": 1.1422,
      "step": 2926
    },
    {
      "epoch": 278.76190476190476,
      "grad_norm": 5.85255765914917,
      "learning_rate": 5.483508245877062e-07,
      "loss": 1.1216,
      "step": 2927
    },
    {
      "epoch": 278.85714285714283,
      "grad_norm": 5.081287384033203,
      "learning_rate": 5.485382308845578e-07,
      "loss": 1.1161,
      "step": 2928
    },
    {
      "epoch": 278.95238095238096,
      "grad_norm": 8.263365745544434,
      "learning_rate": 5.487256371814093e-07,
      "loss": 1.14,
      "step": 2929
    },
    {
      "epoch": 278.95238095238096,
      "eval_accuracy": 0.7298613713001124,
      "eval_f1": 0.7854805117524546,
      "eval_loss": 0.5636739730834961,
      "eval_precision": 0.7264722069345074,
      "eval_recall": 0.8549222797927462,
      "eval_roc_auc": 0.7763837075417386,
      "eval_runtime": 1.1999,
      "eval_samples_per_second": 2224.265,
      "eval_steps_per_second": 5.0,
      "step": 2929
    },
    {
      "epoch": 279.04761904761904,
      "grad_norm": 9.072875022888184,
      "learning_rate": 5.489130434782609e-07,
      "loss": 1.1527,
      "step": 2930
    },
    {
      "epoch": 279.14285714285717,
      "grad_norm": 6.613090991973877,
      "learning_rate": 5.491004497751124e-07,
      "loss": 1.1414,
      "step": 2931
    },
    {
      "epoch": 279.23809523809524,
      "grad_norm": 7.818994522094727,
      "learning_rate": 5.492878560719641e-07,
      "loss": 1.1396,
      "step": 2932
    },
    {
      "epoch": 279.3333333333333,
      "grad_norm": 5.4621124267578125,
      "learning_rate": 5.494752623688157e-07,
      "loss": 1.1323,
      "step": 2933
    },
    {
      "epoch": 279.42857142857144,
      "grad_norm": 11.873200416564941,
      "learning_rate": 5.496626686656672e-07,
      "loss": 1.1364,
      "step": 2934
    },
    {
      "epoch": 279.5238095238095,
      "grad_norm": 4.848647594451904,
      "learning_rate": 5.498500749625188e-07,
      "loss": 1.1477,
      "step": 2935
    },
    {
      "epoch": 279.6190476190476,
      "grad_norm": 9.585158348083496,
      "learning_rate": 5.500374812593703e-07,
      "loss": 1.0999,
      "step": 2936
    },
    {
      "epoch": 279.7142857142857,
      "grad_norm": 4.959415435791016,
      "learning_rate": 5.502248875562219e-07,
      "loss": 1.1543,
      "step": 2937
    },
    {
      "epoch": 279.8095238095238,
      "grad_norm": 3.9837114810943604,
      "learning_rate": 5.504122938530734e-07,
      "loss": 1.1479,
      "step": 2938
    },
    {
      "epoch": 279.9047619047619,
      "grad_norm": 5.842416763305664,
      "learning_rate": 5.505997001499251e-07,
      "loss": 1.116,
      "step": 2939
    },
    {
      "epoch": 280.0,
      "grad_norm": 5.95350980758667,
      "learning_rate": 5.507871064467766e-07,
      "loss": 1.1409,
      "step": 2940
    },
    {
      "epoch": 280.0,
      "eval_accuracy": 0.7234919445485201,
      "eval_f1": 0.7766343825665859,
      "eval_loss": 0.5636378526687622,
      "eval_precision": 0.7289772727272728,
      "eval_recall": 0.8309585492227979,
      "eval_roc_auc": 0.7759784110535406,
      "eval_runtime": 1.3077,
      "eval_samples_per_second": 2040.91,
      "eval_steps_per_second": 4.588,
      "step": 2940
    },
    {
      "epoch": 280.0952380952381,
      "grad_norm": 9.22157096862793,
      "learning_rate": 5.509745127436282e-07,
      "loss": 1.1084,
      "step": 2941
    },
    {
      "epoch": 280.1904761904762,
      "grad_norm": 4.287394046783447,
      "learning_rate": 5.511619190404798e-07,
      "loss": 1.1666,
      "step": 2942
    },
    {
      "epoch": 280.2857142857143,
      "grad_norm": 8.031574249267578,
      "learning_rate": 5.513493253373313e-07,
      "loss": 1.1422,
      "step": 2943
    },
    {
      "epoch": 280.3809523809524,
      "grad_norm": 6.339638710021973,
      "learning_rate": 5.515367316341829e-07,
      "loss": 1.1359,
      "step": 2944
    },
    {
      "epoch": 280.4761904761905,
      "grad_norm": 6.532580852508545,
      "learning_rate": 5.517241379310344e-07,
      "loss": 1.1241,
      "step": 2945
    },
    {
      "epoch": 280.57142857142856,
      "grad_norm": 4.235518455505371,
      "learning_rate": 5.519115442278861e-07,
      "loss": 1.1372,
      "step": 2946
    },
    {
      "epoch": 280.6666666666667,
      "grad_norm": 4.8458991050720215,
      "learning_rate": 5.520989505247376e-07,
      "loss": 1.1259,
      "step": 2947
    },
    {
      "epoch": 280.76190476190476,
      "grad_norm": 5.886928081512451,
      "learning_rate": 5.522863568215893e-07,
      "loss": 1.15,
      "step": 2948
    },
    {
      "epoch": 280.85714285714283,
      "grad_norm": 3.73628568649292,
      "learning_rate": 5.524737631184408e-07,
      "loss": 1.1483,
      "step": 2949
    },
    {
      "epoch": 280.95238095238096,
      "grad_norm": 4.2562255859375,
      "learning_rate": 5.526611694152924e-07,
      "loss": 1.1616,
      "step": 2950
    },
    {
      "epoch": 280.95238095238096,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.7804145389005708,
      "eval_loss": 0.5636350512504578,
      "eval_precision": 0.7277310924369748,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.77594156591825,
      "eval_runtime": 1.1488,
      "eval_samples_per_second": 2323.242,
      "eval_steps_per_second": 5.223,
      "step": 2950
    },
    {
      "epoch": 281.04761904761904,
      "grad_norm": 4.070923805236816,
      "learning_rate": 5.52848575712144e-07,
      "loss": 1.1298,
      "step": 2951
    },
    {
      "epoch": 281.14285714285717,
      "grad_norm": 3.9660825729370117,
      "learning_rate": 5.530359820089955e-07,
      "loss": 1.1391,
      "step": 2952
    },
    {
      "epoch": 281.23809523809524,
      "grad_norm": 7.847269058227539,
      "learning_rate": 5.532233883058472e-07,
      "loss": 1.1308,
      "step": 2953
    },
    {
      "epoch": 281.3333333333333,
      "grad_norm": 10.086249351501465,
      "learning_rate": 5.534107946026987e-07,
      "loss": 1.0882,
      "step": 2954
    },
    {
      "epoch": 281.42857142857144,
      "grad_norm": 7.166942596435547,
      "learning_rate": 5.535982008995503e-07,
      "loss": 1.1334,
      "step": 2955
    },
    {
      "epoch": 281.5238095238095,
      "grad_norm": 4.273452281951904,
      "learning_rate": 5.537856071964018e-07,
      "loss": 1.1654,
      "step": 2956
    },
    {
      "epoch": 281.6190476190476,
      "grad_norm": 5.848499298095703,
      "learning_rate": 5.539730134932534e-07,
      "loss": 1.146,
      "step": 2957
    },
    {
      "epoch": 281.7142857142857,
      "grad_norm": 3.5379650592803955,
      "learning_rate": 5.541604197901049e-07,
      "loss": 1.1066,
      "step": 2958
    },
    {
      "epoch": 281.8095238095238,
      "grad_norm": 4.1590046882629395,
      "learning_rate": 5.543478260869565e-07,
      "loss": 1.147,
      "step": 2959
    },
    {
      "epoch": 281.9047619047619,
      "grad_norm": 4.934486389160156,
      "learning_rate": 5.545352323838082e-07,
      "loss": 1.1444,
      "step": 2960
    },
    {
      "epoch": 282.0,
      "grad_norm": 9.666616439819336,
      "learning_rate": 5.547226386806597e-07,
      "loss": 1.1567,
      "step": 2961
    },
    {
      "epoch": 282.0,
      "eval_accuracy": 0.7287373548145373,
      "eval_f1": 0.7857988165680473,
      "eval_loss": 0.5639248490333557,
      "eval_precision": 0.7233115468409586,
      "eval_recall": 0.8601036269430051,
      "eval_roc_auc": 0.7761266551525619,
      "eval_runtime": 1.2174,
      "eval_samples_per_second": 2192.455,
      "eval_steps_per_second": 4.929,
      "step": 2961
    },
    {
      "epoch": 282.0952380952381,
      "grad_norm": 3.8604648113250732,
      "learning_rate": 5.549100449775113e-07,
      "loss": 1.1145,
      "step": 2962
    },
    {
      "epoch": 282.1904761904762,
      "grad_norm": 11.779272079467773,
      "learning_rate": 5.550974512743628e-07,
      "loss": 1.1494,
      "step": 2963
    },
    {
      "epoch": 282.2857142857143,
      "grad_norm": 9.37829875946045,
      "learning_rate": 5.552848575712144e-07,
      "loss": 1.1446,
      "step": 2964
    },
    {
      "epoch": 282.3809523809524,
      "grad_norm": 8.775849342346191,
      "learning_rate": 5.554722638680659e-07,
      "loss": 1.1425,
      "step": 2965
    },
    {
      "epoch": 282.4761904761905,
      "grad_norm": 3.7040064334869385,
      "learning_rate": 5.556596701649176e-07,
      "loss": 1.139,
      "step": 2966
    },
    {
      "epoch": 282.57142857142856,
      "grad_norm": 9.283699989318848,
      "learning_rate": 5.558470764617691e-07,
      "loss": 1.1133,
      "step": 2967
    },
    {
      "epoch": 282.6666666666667,
      "grad_norm": 4.366463661193848,
      "learning_rate": 5.560344827586207e-07,
      "loss": 1.1191,
      "step": 2968
    },
    {
      "epoch": 282.76190476190476,
      "grad_norm": 9.383438110351562,
      "learning_rate": 5.562218890554723e-07,
      "loss": 1.1277,
      "step": 2969
    },
    {
      "epoch": 282.85714285714283,
      "grad_norm": 10.565701484680176,
      "learning_rate": 5.564092953523238e-07,
      "loss": 1.155,
      "step": 2970
    },
    {
      "epoch": 282.95238095238096,
      "grad_norm": 4.4187397956848145,
      "learning_rate": 5.565967016491755e-07,
      "loss": 1.1471,
      "step": 2971
    },
    {
      "epoch": 282.95238095238096,
      "eval_accuracy": 0.7276133383289621,
      "eval_f1": 0.7812217875413783,
      "eval_loss": 0.563374936580658,
      "eval_precision": 0.7296233839235525,
      "eval_recall": 0.8406735751295337,
      "eval_roc_auc": 0.7761419113413933,
      "eval_runtime": 1.1967,
      "eval_samples_per_second": 2230.318,
      "eval_steps_per_second": 5.014,
      "step": 2971
    },
    {
      "epoch": 283.04761904761904,
      "grad_norm": 5.452326774597168,
      "learning_rate": 5.56784107946027e-07,
      "loss": 1.1872,
      "step": 2972
    },
    {
      "epoch": 283.14285714285717,
      "grad_norm": 11.2903413772583,
      "learning_rate": 5.569715142428787e-07,
      "loss": 1.1473,
      "step": 2973
    },
    {
      "epoch": 283.23809523809524,
      "grad_norm": 4.844279766082764,
      "learning_rate": 5.571589205397302e-07,
      "loss": 1.1261,
      "step": 2974
    },
    {
      "epoch": 283.3333333333333,
      "grad_norm": 5.5100321769714355,
      "learning_rate": 5.573463268365818e-07,
      "loss": 1.0917,
      "step": 2975
    },
    {
      "epoch": 283.42857142857144,
      "grad_norm": 6.915529251098633,
      "learning_rate": 5.575337331334333e-07,
      "loss": 1.1668,
      "step": 2976
    },
    {
      "epoch": 283.5238095238095,
      "grad_norm": 5.634005546569824,
      "learning_rate": 5.577211394302849e-07,
      "loss": 1.0749,
      "step": 2977
    },
    {
      "epoch": 283.6190476190476,
      "grad_norm": 10.577078819274902,
      "learning_rate": 5.579085457271365e-07,
      "loss": 1.1603,
      "step": 2978
    },
    {
      "epoch": 283.7142857142857,
      "grad_norm": 4.12070369720459,
      "learning_rate": 5.58095952023988e-07,
      "loss": 1.1099,
      "step": 2979
    },
    {
      "epoch": 283.8095238095238,
      "grad_norm": 5.871093273162842,
      "learning_rate": 5.582833583208397e-07,
      "loss": 1.1818,
      "step": 2980
    },
    {
      "epoch": 283.9047619047619,
      "grad_norm": 6.47518253326416,
      "learning_rate": 5.584707646176912e-07,
      "loss": 1.1224,
      "step": 2981
    },
    {
      "epoch": 284.0,
      "grad_norm": 6.158949375152588,
      "learning_rate": 5.586581709145428e-07,
      "loss": 1.1205,
      "step": 2982
    },
    {
      "epoch": 284.0,
      "eval_accuracy": 0.729486699138254,
      "eval_f1": 0.7837028160575195,
      "eval_loss": 0.5626216530799866,
      "eval_precision": 0.7290969899665551,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7773434081750143,
      "eval_runtime": 1.2058,
      "eval_samples_per_second": 2213.511,
      "eval_steps_per_second": 4.976,
      "step": 2982
    },
    {
      "epoch": 284.0952380952381,
      "grad_norm": 5.4894118309021,
      "learning_rate": 5.588455772113943e-07,
      "loss": 1.1201,
      "step": 2983
    },
    {
      "epoch": 284.1904761904762,
      "grad_norm": 5.079287528991699,
      "learning_rate": 5.590329835082459e-07,
      "loss": 1.1071,
      "step": 2984
    },
    {
      "epoch": 284.2857142857143,
      "grad_norm": 7.196651458740234,
      "learning_rate": 5.592203898050974e-07,
      "loss": 1.1238,
      "step": 2985
    },
    {
      "epoch": 284.3809523809524,
      "grad_norm": 5.972583293914795,
      "learning_rate": 5.59407796101949e-07,
      "loss": 1.1456,
      "step": 2986
    },
    {
      "epoch": 284.4761904761905,
      "grad_norm": 4.000961780548096,
      "learning_rate": 5.595952023988007e-07,
      "loss": 1.1411,
      "step": 2987
    },
    {
      "epoch": 284.57142857142856,
      "grad_norm": 6.014395713806152,
      "learning_rate": 5.597826086956522e-07,
      "loss": 1.1582,
      "step": 2988
    },
    {
      "epoch": 284.6666666666667,
      "grad_norm": 4.829216003417969,
      "learning_rate": 5.599700149925038e-07,
      "loss": 1.147,
      "step": 2989
    },
    {
      "epoch": 284.76190476190476,
      "grad_norm": 4.213350772857666,
      "learning_rate": 5.601574212893553e-07,
      "loss": 1.1444,
      "step": 2990
    },
    {
      "epoch": 284.85714285714283,
      "grad_norm": 5.931327819824219,
      "learning_rate": 5.603448275862069e-07,
      "loss": 1.1281,
      "step": 2991
    },
    {
      "epoch": 284.95238095238096,
      "grad_norm": 4.410636901855469,
      "learning_rate": 5.605322338830584e-07,
      "loss": 1.105,
      "step": 2992
    },
    {
      "epoch": 284.95238095238096,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.7797529376318169,
      "eval_loss": 0.5622423887252808,
      "eval_precision": 0.7290140845070423,
      "eval_recall": 0.8380829015544041,
      "eval_roc_auc": 0.7776154289004029,
      "eval_runtime": 1.1496,
      "eval_samples_per_second": 2321.654,
      "eval_steps_per_second": 5.219,
      "step": 2992
    },
    {
      "epoch": 285.04761904761904,
      "grad_norm": 3.58267879486084,
      "learning_rate": 5.607196401799101e-07,
      "loss": 1.1236,
      "step": 2993
    },
    {
      "epoch": 285.14285714285717,
      "grad_norm": 5.702927112579346,
      "learning_rate": 5.609070464767616e-07,
      "loss": 1.1681,
      "step": 2994
    },
    {
      "epoch": 285.23809523809524,
      "grad_norm": 5.016921043395996,
      "learning_rate": 5.610944527736132e-07,
      "loss": 1.1307,
      "step": 2995
    },
    {
      "epoch": 285.3333333333333,
      "grad_norm": 9.380377769470215,
      "learning_rate": 5.612818590704649e-07,
      "loss": 1.156,
      "step": 2996
    },
    {
      "epoch": 285.42857142857144,
      "grad_norm": 5.757893085479736,
      "learning_rate": 5.614692653673164e-07,
      "loss": 1.13,
      "step": 2997
    },
    {
      "epoch": 285.5238095238095,
      "grad_norm": 5.188559055328369,
      "learning_rate": 5.61656671664168e-07,
      "loss": 1.1178,
      "step": 2998
    },
    {
      "epoch": 285.6190476190476,
      "grad_norm": 6.849817276000977,
      "learning_rate": 5.618440779610195e-07,
      "loss": 1.1294,
      "step": 2999
    },
    {
      "epoch": 285.7142857142857,
      "grad_norm": 11.583452224731445,
      "learning_rate": 5.620314842578711e-07,
      "loss": 1.1146,
      "step": 3000
    },
    {
      "epoch": 285.8095238095238,
      "grad_norm": 8.750665664672852,
      "learning_rate": 5.622188905547227e-07,
      "loss": 1.1744,
      "step": 3001
    },
    {
      "epoch": 285.9047619047619,
      "grad_norm": 5.649927616119385,
      "learning_rate": 5.624062968515743e-07,
      "loss": 1.1339,
      "step": 3002
    },
    {
      "epoch": 286.0,
      "grad_norm": 4.693404674530029,
      "learning_rate": 5.625937031484258e-07,
      "loss": 1.1333,
      "step": 3003
    },
    {
      "epoch": 286.0,
      "eval_accuracy": 0.726489321843387,
      "eval_f1": 0.7803850782190133,
      "eval_loss": 0.5620488524436951,
      "eval_precision": 0.7286516853932584,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7778422567645366,
      "eval_runtime": 1.1855,
      "eval_samples_per_second": 2251.445,
      "eval_steps_per_second": 5.061,
      "step": 3003
    },
    {
      "epoch": 286.0952380952381,
      "grad_norm": 3.090787887573242,
      "learning_rate": 5.627811094452774e-07,
      "loss": 1.1242,
      "step": 3004
    },
    {
      "epoch": 286.1904761904762,
      "grad_norm": 3.8368513584136963,
      "learning_rate": 5.62968515742129e-07,
      "loss": 1.08,
      "step": 3005
    },
    {
      "epoch": 286.2857142857143,
      "grad_norm": 9.99586296081543,
      "learning_rate": 5.631559220389805e-07,
      "loss": 1.1767,
      "step": 3006
    },
    {
      "epoch": 286.3809523809524,
      "grad_norm": 3.3602445125579834,
      "learning_rate": 5.633433283358321e-07,
      "loss": 1.1401,
      "step": 3007
    },
    {
      "epoch": 286.4761904761905,
      "grad_norm": 9.76583480834961,
      "learning_rate": 5.635307346326837e-07,
      "loss": 1.1104,
      "step": 3008
    },
    {
      "epoch": 286.57142857142856,
      "grad_norm": 8.486311912536621,
      "learning_rate": 5.637181409295353e-07,
      "loss": 1.166,
      "step": 3009
    },
    {
      "epoch": 286.6666666666667,
      "grad_norm": 5.658336639404297,
      "learning_rate": 5.639055472263868e-07,
      "loss": 1.1606,
      "step": 3010
    },
    {
      "epoch": 286.76190476190476,
      "grad_norm": 6.460410118103027,
      "learning_rate": 5.640929535232384e-07,
      "loss": 1.1226,
      "step": 3011
    },
    {
      "epoch": 286.85714285714283,
      "grad_norm": 5.95920991897583,
      "learning_rate": 5.642803598200899e-07,
      "loss": 1.1116,
      "step": 3012
    },
    {
      "epoch": 286.95238095238096,
      "grad_norm": 8.102746963500977,
      "learning_rate": 5.644677661169415e-07,
      "loss": 1.163,
      "step": 3013
    },
    {
      "epoch": 286.95238095238096,
      "eval_accuracy": 0.7268639940052454,
      "eval_f1": 0.7808836789900812,
      "eval_loss": 0.5618669986724854,
      "eval_precision": 0.7285473920358946,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.7780023028209557,
      "eval_runtime": 1.3258,
      "eval_samples_per_second": 2013.07,
      "eval_steps_per_second": 4.525,
      "step": 3013
    },
    {
      "epoch": 287.04761904761904,
      "grad_norm": 3.968151330947876,
      "learning_rate": 5.64655172413793e-07,
      "loss": 1.1229,
      "step": 3014
    },
    {
      "epoch": 287.14285714285717,
      "grad_norm": 4.948294639587402,
      "learning_rate": 5.648425787106447e-07,
      "loss": 1.1114,
      "step": 3015
    },
    {
      "epoch": 287.23809523809524,
      "grad_norm": 8.754914283752441,
      "learning_rate": 5.650299850074963e-07,
      "loss": 1.1497,
      "step": 3016
    },
    {
      "epoch": 287.3333333333333,
      "grad_norm": 4.025392532348633,
      "learning_rate": 5.652173913043478e-07,
      "loss": 1.1172,
      "step": 3017
    },
    {
      "epoch": 287.42857142857144,
      "grad_norm": 11.300790786743164,
      "learning_rate": 5.654047976011995e-07,
      "loss": 1.1366,
      "step": 3018
    },
    {
      "epoch": 287.5238095238095,
      "grad_norm": 7.936151027679443,
      "learning_rate": 5.65592203898051e-07,
      "loss": 1.1396,
      "step": 3019
    },
    {
      "epoch": 287.6190476190476,
      "grad_norm": 8.417755126953125,
      "learning_rate": 5.657796101949026e-07,
      "loss": 1.1312,
      "step": 3020
    },
    {
      "epoch": 287.7142857142857,
      "grad_norm": 8.80263614654541,
      "learning_rate": 5.659670164917541e-07,
      "loss": 1.071,
      "step": 3021
    },
    {
      "epoch": 287.8095238095238,
      "grad_norm": 6.077959060668945,
      "learning_rate": 5.661544227886058e-07,
      "loss": 1.1265,
      "step": 3022
    },
    {
      "epoch": 287.9047619047619,
      "grad_norm": 7.415537357330322,
      "learning_rate": 5.663418290854574e-07,
      "loss": 1.1485,
      "step": 3023
    },
    {
      "epoch": 288.0,
      "grad_norm": 11.430869102478027,
      "learning_rate": 5.665292353823089e-07,
      "loss": 1.2005,
      "step": 3024
    },
    {
      "epoch": 288.0,
      "eval_accuracy": 0.7279880104908205,
      "eval_f1": 0.7821128451380552,
      "eval_loss": 0.5615506172180176,
      "eval_precision": 0.7287472035794184,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7781177317213587,
      "eval_runtime": 1.186,
      "eval_samples_per_second": 2250.509,
      "eval_steps_per_second": 5.059,
      "step": 3024
    },
    {
      "epoch": 288.0952380952381,
      "grad_norm": 5.465104579925537,
      "learning_rate": 5.667166416791605e-07,
      "loss": 1.1319,
      "step": 3025
    },
    {
      "epoch": 288.1904761904762,
      "grad_norm": 8.500192642211914,
      "learning_rate": 5.66904047976012e-07,
      "loss": 1.1595,
      "step": 3026
    },
    {
      "epoch": 288.2857142857143,
      "grad_norm": 8.348478317260742,
      "learning_rate": 5.670914542728636e-07,
      "loss": 1.109,
      "step": 3027
    },
    {
      "epoch": 288.3809523809524,
      "grad_norm": 10.712888717651367,
      "learning_rate": 5.672788605697151e-07,
      "loss": 1.156,
      "step": 3028
    },
    {
      "epoch": 288.4761904761905,
      "grad_norm": 4.471301078796387,
      "learning_rate": 5.674662668665668e-07,
      "loss": 1.1707,
      "step": 3029
    },
    {
      "epoch": 288.57142857142856,
      "grad_norm": 5.227294921875,
      "learning_rate": 5.676536731634183e-07,
      "loss": 1.1179,
      "step": 3030
    },
    {
      "epoch": 288.6666666666667,
      "grad_norm": 7.055893421173096,
      "learning_rate": 5.678410794602699e-07,
      "loss": 1.1221,
      "step": 3031
    },
    {
      "epoch": 288.76190476190476,
      "grad_norm": 14.00888729095459,
      "learning_rate": 5.680284857571215e-07,
      "loss": 1.1227,
      "step": 3032
    },
    {
      "epoch": 288.85714285714283,
      "grad_norm": 13.468306541442871,
      "learning_rate": 5.68215892053973e-07,
      "loss": 1.1306,
      "step": 3033
    },
    {
      "epoch": 288.95238095238096,
      "grad_norm": 3.5932788848876953,
      "learning_rate": 5.684032983508246e-07,
      "loss": 1.1477,
      "step": 3034
    },
    {
      "epoch": 288.95238095238096,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.7801503759398496,
      "eval_loss": 0.5614120960235596,
      "eval_precision": 0.7282425603593486,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7784884858952217,
      "eval_runtime": 1.1599,
      "eval_samples_per_second": 2301.009,
      "eval_steps_per_second": 5.173,
      "step": 3034
    },
    {
      "epoch": 289.04761904761904,
      "grad_norm": 5.253028392791748,
      "learning_rate": 5.685907046476761e-07,
      "loss": 1.1977,
      "step": 3035
    },
    {
      "epoch": 289.14285714285717,
      "grad_norm": 7.134814739227295,
      "learning_rate": 5.687781109445278e-07,
      "loss": 1.1047,
      "step": 3036
    },
    {
      "epoch": 289.23809523809524,
      "grad_norm": 6.438389778137207,
      "learning_rate": 5.689655172413793e-07,
      "loss": 1.1218,
      "step": 3037
    },
    {
      "epoch": 289.3333333333333,
      "grad_norm": 5.677706718444824,
      "learning_rate": 5.691529235382309e-07,
      "loss": 1.1593,
      "step": 3038
    },
    {
      "epoch": 289.42857142857144,
      "grad_norm": 5.320911884307861,
      "learning_rate": 5.693403298350824e-07,
      "loss": 1.1046,
      "step": 3039
    },
    {
      "epoch": 289.5238095238095,
      "grad_norm": 6.542576313018799,
      "learning_rate": 5.695277361319341e-07,
      "loss": 1.1368,
      "step": 3040
    },
    {
      "epoch": 289.6190476190476,
      "grad_norm": 9.399479866027832,
      "learning_rate": 5.697151424287857e-07,
      "loss": 1.1047,
      "step": 3041
    },
    {
      "epoch": 289.7142857142857,
      "grad_norm": 7.715976238250732,
      "learning_rate": 5.699025487256372e-07,
      "loss": 1.1477,
      "step": 3042
    },
    {
      "epoch": 289.8095238095238,
      "grad_norm": 7.544461727142334,
      "learning_rate": 5.700899550224889e-07,
      "loss": 1.1429,
      "step": 3043
    },
    {
      "epoch": 289.9047619047619,
      "grad_norm": 11.521993637084961,
      "learning_rate": 5.702773613193404e-07,
      "loss": 1.1868,
      "step": 3044
    },
    {
      "epoch": 290.0,
      "grad_norm": 4.851450443267822,
      "learning_rate": 5.70464767616192e-07,
      "loss": 1.1391,
      "step": 3045
    },
    {
      "epoch": 290.0,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.7798855766335441,
      "eval_loss": 0.5616016983985901,
      "eval_precision": 0.7287563308947664,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.7783494530800231,
      "eval_runtime": 1.1982,
      "eval_samples_per_second": 2227.525,
      "eval_steps_per_second": 5.008,
      "step": 3045
    },
    {
      "epoch": 290.0952380952381,
      "grad_norm": 5.691206455230713,
      "learning_rate": 5.706521739130435e-07,
      "loss": 1.1532,
      "step": 3046
    },
    {
      "epoch": 290.1904761904762,
      "grad_norm": 8.848892211914062,
      "learning_rate": 5.708395802098951e-07,
      "loss": 1.1201,
      "step": 3047
    },
    {
      "epoch": 290.2857142857143,
      "grad_norm": 4.961413860321045,
      "learning_rate": 5.710269865067466e-07,
      "loss": 1.1733,
      "step": 3048
    },
    {
      "epoch": 290.3809523809524,
      "grad_norm": 4.45950984954834,
      "learning_rate": 5.712143928035982e-07,
      "loss": 1.1741,
      "step": 3049
    },
    {
      "epoch": 290.4761904761905,
      "grad_norm": 3.6861424446105957,
      "learning_rate": 5.714017991004499e-07,
      "loss": 1.1357,
      "step": 3050
    },
    {
      "epoch": 290.57142857142856,
      "grad_norm": 4.187479019165039,
      "learning_rate": 5.715892053973014e-07,
      "loss": 1.1444,
      "step": 3051
    },
    {
      "epoch": 290.6666666666667,
      "grad_norm": 5.746031761169434,
      "learning_rate": 5.71776611694153e-07,
      "loss": 1.134,
      "step": 3052
    },
    {
      "epoch": 290.76190476190476,
      "grad_norm": 5.604963779449463,
      "learning_rate": 5.719640179910045e-07,
      "loss": 1.1561,
      "step": 3053
    },
    {
      "epoch": 290.85714285714283,
      "grad_norm": 5.199735641479492,
      "learning_rate": 5.721514242878561e-07,
      "loss": 1.1173,
      "step": 3054
    },
    {
      "epoch": 290.95238095238096,
      "grad_norm": 4.521169662475586,
      "learning_rate": 5.723388305847076e-07,
      "loss": 1.0963,
      "step": 3055
    },
    {
      "epoch": 290.95238095238096,
      "eval_accuracy": 0.7298613713001124,
      "eval_f1": 0.7832882476705741,
      "eval_loss": 0.5621520280838013,
      "eval_precision": 0.730790802019069,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7781056419113415,
      "eval_runtime": 1.3525,
      "eval_samples_per_second": 1973.311,
      "eval_steps_per_second": 4.436,
      "step": 3055
    },
    {
      "epoch": 291.04761904761904,
      "grad_norm": 4.686976432800293,
      "learning_rate": 5.725262368815592e-07,
      "loss": 1.1488,
      "step": 3056
    },
    {
      "epoch": 291.14285714285717,
      "grad_norm": 4.308133125305176,
      "learning_rate": 5.727136431784108e-07,
      "loss": 1.088,
      "step": 3057
    },
    {
      "epoch": 291.23809523809524,
      "grad_norm": 9.826374053955078,
      "learning_rate": 5.729010494752624e-07,
      "loss": 1.1471,
      "step": 3058
    },
    {
      "epoch": 291.3333333333333,
      "grad_norm": 5.6400980949401855,
      "learning_rate": 5.73088455772114e-07,
      "loss": 1.1539,
      "step": 3059
    },
    {
      "epoch": 291.42857142857144,
      "grad_norm": 4.731324672698975,
      "learning_rate": 5.732758620689655e-07,
      "loss": 1.1177,
      "step": 3060
    },
    {
      "epoch": 291.5238095238095,
      "grad_norm": 7.6967644691467285,
      "learning_rate": 5.734632683658171e-07,
      "loss": 1.1073,
      "step": 3061
    },
    {
      "epoch": 291.6190476190476,
      "grad_norm": 3.6323401927948,
      "learning_rate": 5.736506746626686e-07,
      "loss": 1.182,
      "step": 3062
    },
    {
      "epoch": 291.7142857142857,
      "grad_norm": 4.018256664276123,
      "learning_rate": 5.738380809595203e-07,
      "loss": 1.1464,
      "step": 3063
    },
    {
      "epoch": 291.8095238095238,
      "grad_norm": 4.508745193481445,
      "learning_rate": 5.740254872563718e-07,
      "loss": 1.1568,
      "step": 3064
    },
    {
      "epoch": 291.9047619047619,
      "grad_norm": 6.37939453125,
      "learning_rate": 5.742128935532235e-07,
      "loss": 1.1368,
      "step": 3065
    },
    {
      "epoch": 292.0,
      "grad_norm": 3.512845039367676,
      "learning_rate": 5.74400299850075e-07,
      "loss": 1.1478,
      "step": 3066
    },
    {
      "epoch": 292.0,
      "eval_accuracy": 0.7287373548145373,
      "eval_f1": 0.7827130852340937,
      "eval_loss": 0.5623800158500671,
      "eval_precision": 0.7293064876957495,
      "eval_recall": 0.844559585492228,
      "eval_roc_auc": 0.7777728842832469,
      "eval_runtime": 1.1205,
      "eval_samples_per_second": 2381.874,
      "eval_steps_per_second": 5.355,
      "step": 3066
    },
    {
      "epoch": 292.0952380952381,
      "grad_norm": 10.951603889465332,
      "learning_rate": 5.745877061469266e-07,
      "loss": 1.161,
      "step": 3067
    },
    {
      "epoch": 292.1904761904762,
      "grad_norm": 13.926806449890137,
      "learning_rate": 5.747751124437782e-07,
      "loss": 1.165,
      "step": 3068
    },
    {
      "epoch": 292.2857142857143,
      "grad_norm": 3.765690565109253,
      "learning_rate": 5.749625187406297e-07,
      "loss": 1.1571,
      "step": 3069
    },
    {
      "epoch": 292.3809523809524,
      "grad_norm": 8.859353065490723,
      "learning_rate": 5.751499250374813e-07,
      "loss": 1.1286,
      "step": 3070
    },
    {
      "epoch": 292.4761904761905,
      "grad_norm": 6.424526214599609,
      "learning_rate": 5.753373313343329e-07,
      "loss": 1.1304,
      "step": 3071
    },
    {
      "epoch": 292.57142857142856,
      "grad_norm": 10.029023170471191,
      "learning_rate": 5.755247376311845e-07,
      "loss": 1.0804,
      "step": 3072
    },
    {
      "epoch": 292.6666666666667,
      "grad_norm": 3.650876045227051,
      "learning_rate": 5.75712143928036e-07,
      "loss": 1.1577,
      "step": 3073
    },
    {
      "epoch": 292.76190476190476,
      "grad_norm": 6.428129196166992,
      "learning_rate": 5.758995502248876e-07,
      "loss": 1.137,
      "step": 3074
    },
    {
      "epoch": 292.85714285714283,
      "grad_norm": 7.669662952423096,
      "learning_rate": 5.760869565217391e-07,
      "loss": 1.1369,
      "step": 3075
    },
    {
      "epoch": 292.95238095238096,
      "grad_norm": 7.731568813323975,
      "learning_rate": 5.762743628185907e-07,
      "loss": 1.1473,
      "step": 3076
    },
    {
      "epoch": 292.95238095238096,
      "eval_accuracy": 0.7306107156238292,
      "eval_f1": 0.7856929955290611,
      "eval_loss": 0.5623287558555603,
      "eval_precision": 0.7277747101049145,
      "eval_recall": 0.8536269430051814,
      "eval_roc_auc": 0.7782962003454232,
      "eval_runtime": 1.2639,
      "eval_samples_per_second": 2111.777,
      "eval_steps_per_second": 4.747,
      "step": 3076
    },
    {
      "epoch": 293.04761904761904,
      "grad_norm": 8.484759330749512,
      "learning_rate": 5.764617691154423e-07,
      "loss": 1.1324,
      "step": 3077
    },
    {
      "epoch": 293.14285714285717,
      "grad_norm": 9.59839916229248,
      "learning_rate": 5.766491754122939e-07,
      "loss": 1.1433,
      "step": 3078
    },
    {
      "epoch": 293.23809523809524,
      "grad_norm": 6.049525737762451,
      "learning_rate": 5.768365817091455e-07,
      "loss": 1.1578,
      "step": 3079
    },
    {
      "epoch": 293.3333333333333,
      "grad_norm": 6.255111217498779,
      "learning_rate": 5.77023988005997e-07,
      "loss": 1.1264,
      "step": 3080
    },
    {
      "epoch": 293.42857142857144,
      "grad_norm": 4.266199111938477,
      "learning_rate": 5.772113943028486e-07,
      "loss": 1.0949,
      "step": 3081
    },
    {
      "epoch": 293.5238095238095,
      "grad_norm": 8.692477226257324,
      "learning_rate": 5.773988005997001e-07,
      "loss": 1.1688,
      "step": 3082
    },
    {
      "epoch": 293.6190476190476,
      "grad_norm": 4.867812633514404,
      "learning_rate": 5.775862068965517e-07,
      "loss": 1.1337,
      "step": 3083
    },
    {
      "epoch": 293.7142857142857,
      "grad_norm": 4.503742218017578,
      "learning_rate": 5.777736131934032e-07,
      "loss": 1.1476,
      "step": 3084
    },
    {
      "epoch": 293.8095238095238,
      "grad_norm": 10.558121681213379,
      "learning_rate": 5.779610194902549e-07,
      "loss": 1.1134,
      "step": 3085
    },
    {
      "epoch": 293.9047619047619,
      "grad_norm": 11.817536354064941,
      "learning_rate": 5.781484257871065e-07,
      "loss": 1.1337,
      "step": 3086
    },
    {
      "epoch": 294.0,
      "grad_norm": 8.518043518066406,
      "learning_rate": 5.78335832083958e-07,
      "loss": 1.1348,
      "step": 3087
    },
    {
      "epoch": 294.0,
      "eval_accuracy": 0.7249906331959536,
      "eval_f1": 0.7779794313369631,
      "eval_loss": 0.5618783235549927,
      "eval_precision": 0.7298524404086265,
      "eval_recall": 0.832901554404145,
      "eval_roc_auc": 0.7786315486470926,
      "eval_runtime": 1.1578,
      "eval_samples_per_second": 2305.293,
      "eval_steps_per_second": 5.182,
      "step": 3087
    },
    {
      "epoch": 294.0952380952381,
      "grad_norm": 5.27886962890625,
      "learning_rate": 5.785232383808097e-07,
      "loss": 1.1217,
      "step": 3088
    },
    {
      "epoch": 294.1904761904762,
      "grad_norm": 3.7267980575561523,
      "learning_rate": 5.787106446776612e-07,
      "loss": 1.1421,
      "step": 3089
    },
    {
      "epoch": 294.2857142857143,
      "grad_norm": 5.075734615325928,
      "learning_rate": 5.788980509745128e-07,
      "loss": 1.1312,
      "step": 3090
    },
    {
      "epoch": 294.3809523809524,
      "grad_norm": 11.15335464477539,
      "learning_rate": 5.790854572713643e-07,
      "loss": 1.1175,
      "step": 3091
    },
    {
      "epoch": 294.4761904761905,
      "grad_norm": 8.802132606506348,
      "learning_rate": 5.79272863568216e-07,
      "loss": 1.1491,
      "step": 3092
    },
    {
      "epoch": 294.57142857142856,
      "grad_norm": 6.129073619842529,
      "learning_rate": 5.794602698650675e-07,
      "loss": 1.0718,
      "step": 3093
    },
    {
      "epoch": 294.6666666666667,
      "grad_norm": 5.770191192626953,
      "learning_rate": 5.796476761619191e-07,
      "loss": 1.1447,
      "step": 3094
    },
    {
      "epoch": 294.76190476190476,
      "grad_norm": 5.671845436096191,
      "learning_rate": 5.798350824587707e-07,
      "loss": 1.1716,
      "step": 3095
    },
    {
      "epoch": 294.85714285714283,
      "grad_norm": 15.143241882324219,
      "learning_rate": 5.800224887556222e-07,
      "loss": 1.1416,
      "step": 3096
    },
    {
      "epoch": 294.95238095238096,
      "grad_norm": 15.241469383239746,
      "learning_rate": 5.802098950524738e-07,
      "loss": 1.1278,
      "step": 3097
    },
    {
      "epoch": 294.95238095238096,
      "eval_accuracy": 0.732484076433121,
      "eval_f1": 0.7898763978811065,
      "eval_loss": 0.5626599788665771,
      "eval_precision": 0.7238403451995685,
      "eval_recall": 0.8691709844559585,
      "eval_roc_auc": 0.7789355210132412,
      "eval_runtime": 1.1853,
      "eval_samples_per_second": 2251.769,
      "eval_steps_per_second": 5.062,
      "step": 3097
    },
    {
      "epoch": 295.04761904761904,
      "grad_norm": 14.951590538024902,
      "learning_rate": 5.803973013493253e-07,
      "loss": 1.1323,
      "step": 3098
    },
    {
      "epoch": 295.14285714285717,
      "grad_norm": 15.76667308807373,
      "learning_rate": 5.80584707646177e-07,
      "loss": 1.1233,
      "step": 3099
    },
    {
      "epoch": 295.23809523809524,
      "grad_norm": 10.656340599060059,
      "learning_rate": 5.807721139430285e-07,
      "loss": 1.1226,
      "step": 3100
    },
    {
      "epoch": 295.3333333333333,
      "grad_norm": 7.614379405975342,
      "learning_rate": 5.809595202398801e-07,
      "loss": 1.1035,
      "step": 3101
    },
    {
      "epoch": 295.42857142857144,
      "grad_norm": 5.129384994506836,
      "learning_rate": 5.811469265367316e-07,
      "loss": 1.1489,
      "step": 3102
    },
    {
      "epoch": 295.5238095238095,
      "grad_norm": 9.847796440124512,
      "learning_rate": 5.813343328335832e-07,
      "loss": 1.1712,
      "step": 3103
    },
    {
      "epoch": 295.6190476190476,
      "grad_norm": 5.071105003356934,
      "learning_rate": 5.815217391304348e-07,
      "loss": 1.1127,
      "step": 3104
    },
    {
      "epoch": 295.7142857142857,
      "grad_norm": 7.032314777374268,
      "learning_rate": 5.817091454272863e-07,
      "loss": 1.1381,
      "step": 3105
    },
    {
      "epoch": 295.8095238095238,
      "grad_norm": 21.643314361572266,
      "learning_rate": 5.81896551724138e-07,
      "loss": 1.1751,
      "step": 3106
    },
    {
      "epoch": 295.9047619047619,
      "grad_norm": 16.88426399230957,
      "learning_rate": 5.820839580209895e-07,
      "loss": 1.1669,
      "step": 3107
    },
    {
      "epoch": 296.0,
      "grad_norm": 14.358475685119629,
      "learning_rate": 5.822713643178411e-07,
      "loss": 1.1254,
      "step": 3108
    },
    {
      "epoch": 296.0,
      "eval_accuracy": 0.7234919445485201,
      "eval_f1": 0.7743119266055046,
      "eval_loss": 0.5619354844093323,
      "eval_precision": 0.7334878331402086,
      "eval_recall": 0.8199481865284974,
      "eval_roc_auc": 0.7782605066206102,
      "eval_runtime": 1.162,
      "eval_samples_per_second": 2296.94,
      "eval_steps_per_second": 5.164,
      "step": 3108
    },
    {
      "epoch": 296.0952380952381,
      "grad_norm": 5.653419017791748,
      "learning_rate": 5.824587706146926e-07,
      "loss": 1.0971,
      "step": 3109
    },
    {
      "epoch": 296.1904761904762,
      "grad_norm": 11.617242813110352,
      "learning_rate": 5.826461769115443e-07,
      "loss": 1.1437,
      "step": 3110
    },
    {
      "epoch": 296.2857142857143,
      "grad_norm": 15.939972877502441,
      "learning_rate": 5.828335832083958e-07,
      "loss": 1.1009,
      "step": 3111
    },
    {
      "epoch": 296.3809523809524,
      "grad_norm": 15.541411399841309,
      "learning_rate": 5.830209895052474e-07,
      "loss": 1.1665,
      "step": 3112
    },
    {
      "epoch": 296.4761904761905,
      "grad_norm": 4.3177056312561035,
      "learning_rate": 5.832083958020991e-07,
      "loss": 1.1738,
      "step": 3113
    },
    {
      "epoch": 296.57142857142856,
      "grad_norm": 9.515015602111816,
      "learning_rate": 5.833958020989506e-07,
      "loss": 1.1184,
      "step": 3114
    },
    {
      "epoch": 296.6666666666667,
      "grad_norm": 19.697540283203125,
      "learning_rate": 5.835832083958022e-07,
      "loss": 1.2232,
      "step": 3115
    },
    {
      "epoch": 296.76190476190476,
      "grad_norm": 3.511566638946533,
      "learning_rate": 5.837706146926537e-07,
      "loss": 1.1091,
      "step": 3116
    },
    {
      "epoch": 296.85714285714283,
      "grad_norm": 6.693878650665283,
      "learning_rate": 5.839580209895053e-07,
      "loss": 1.0979,
      "step": 3117
    },
    {
      "epoch": 296.95238095238096,
      "grad_norm": 8.09457778930664,
      "learning_rate": 5.841454272863568e-07,
      "loss": 1.1329,
      "step": 3118
    },
    {
      "epoch": 296.95238095238096,
      "eval_accuracy": 0.7298613713001124,
      "eval_f1": 0.7861168792643133,
      "eval_loss": 0.5612280964851379,
      "eval_precision": 0.7252326217843459,
      "eval_recall": 0.8581606217616581,
      "eval_roc_auc": 0.7798862982153137,
      "eval_runtime": 1.1797,
      "eval_samples_per_second": 2262.461,
      "eval_steps_per_second": 5.086,
      "step": 3118
    },
    {
      "epoch": 297.04761904761904,
      "grad_norm": 7.095411777496338,
      "learning_rate": 5.843328335832084e-07,
      "loss": 1.1483,
      "step": 3119
    },
    {
      "epoch": 297.14285714285717,
      "grad_norm": 10.408041954040527,
      "learning_rate": 5.8452023988006e-07,
      "loss": 1.1275,
      "step": 3120
    },
    {
      "epoch": 297.23809523809524,
      "grad_norm": 5.680013179779053,
      "learning_rate": 5.847076461769116e-07,
      "loss": 1.1233,
      "step": 3121
    },
    {
      "epoch": 297.3333333333333,
      "grad_norm": 5.734152317047119,
      "learning_rate": 5.848950524737632e-07,
      "loss": 1.1404,
      "step": 3122
    },
    {
      "epoch": 297.42857142857144,
      "grad_norm": 6.248361110687256,
      "learning_rate": 5.850824587706147e-07,
      "loss": 1.177,
      "step": 3123
    },
    {
      "epoch": 297.5238095238095,
      "grad_norm": 8.506665229797363,
      "learning_rate": 5.852698650674663e-07,
      "loss": 1.1269,
      "step": 3124
    },
    {
      "epoch": 297.6190476190476,
      "grad_norm": 4.191070079803467,
      "learning_rate": 5.854572713643178e-07,
      "loss": 1.1508,
      "step": 3125
    },
    {
      "epoch": 297.7142857142857,
      "grad_norm": 12.647905349731445,
      "learning_rate": 5.856446776611694e-07,
      "loss": 1.0912,
      "step": 3126
    },
    {
      "epoch": 297.8095238095238,
      "grad_norm": 5.739502429962158,
      "learning_rate": 5.85832083958021e-07,
      "loss": 1.1881,
      "step": 3127
    },
    {
      "epoch": 297.9047619047619,
      "grad_norm": 9.85162353515625,
      "learning_rate": 5.860194902548726e-07,
      "loss": 1.1236,
      "step": 3128
    },
    {
      "epoch": 298.0,
      "grad_norm": 15.419437408447266,
      "learning_rate": 5.862068965517241e-07,
      "loss": 1.0916,
      "step": 3129
    },
    {
      "epoch": 298.0,
      "eval_accuracy": 0.7257399775196703,
      "eval_f1": 0.7765567765567766,
      "eval_loss": 0.5605422854423523,
      "eval_precision": 0.7344110854503464,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.780507484168106,
      "eval_runtime": 1.2568,
      "eval_samples_per_second": 2123.632,
      "eval_steps_per_second": 4.774,
      "step": 3129
    },
    {
      "epoch": 298.0952380952381,
      "grad_norm": 3.829314947128296,
      "learning_rate": 5.863943028485757e-07,
      "loss": 1.1854,
      "step": 3130
    },
    {
      "epoch": 298.1904761904762,
      "grad_norm": 6.155840873718262,
      "learning_rate": 5.865817091454273e-07,
      "loss": 1.1357,
      "step": 3131
    },
    {
      "epoch": 298.2857142857143,
      "grad_norm": 11.116759300231934,
      "learning_rate": 5.867691154422788e-07,
      "loss": 1.0738,
      "step": 3132
    },
    {
      "epoch": 298.3809523809524,
      "grad_norm": 8.681174278259277,
      "learning_rate": 5.869565217391305e-07,
      "loss": 1.1234,
      "step": 3133
    },
    {
      "epoch": 298.4761904761905,
      "grad_norm": 4.055141448974609,
      "learning_rate": 5.87143928035982e-07,
      "loss": 1.1191,
      "step": 3134
    },
    {
      "epoch": 298.57142857142856,
      "grad_norm": 5.928338527679443,
      "learning_rate": 5.873313343328337e-07,
      "loss": 1.1314,
      "step": 3135
    },
    {
      "epoch": 298.6666666666667,
      "grad_norm": 6.639645576477051,
      "learning_rate": 5.875187406296852e-07,
      "loss": 1.152,
      "step": 3136
    },
    {
      "epoch": 298.76190476190476,
      "grad_norm": 5.529913902282715,
      "learning_rate": 5.877061469265368e-07,
      "loss": 1.0963,
      "step": 3137
    },
    {
      "epoch": 298.85714285714283,
      "grad_norm": 18.109577178955078,
      "learning_rate": 5.878935532233883e-07,
      "loss": 1.1149,
      "step": 3138
    },
    {
      "epoch": 298.95238095238096,
      "grad_norm": 7.9466872215271,
      "learning_rate": 5.880809595202399e-07,
      "loss": 1.1287,
      "step": 3139
    },
    {
      "epoch": 298.95238095238096,
      "eval_accuracy": 0.7302360434619708,
      "eval_f1": 0.787359716479622,
      "eval_loss": 0.560848593711853,
      "eval_precision": 0.7236699239956569,
      "eval_recall": 0.8633419689119171,
      "eval_roc_auc": 0.7808442717328729,
      "eval_runtime": 1.1353,
      "eval_samples_per_second": 2350.957,
      "eval_steps_per_second": 5.285,
      "step": 3139
    },
    {
      "epoch": 299.04761904761904,
      "grad_norm": 6.386238098144531,
      "learning_rate": 5.882683658170915e-07,
      "loss": 1.0765,
      "step": 3140
    },
    {
      "epoch": 299.14285714285717,
      "grad_norm": 6.555985927581787,
      "learning_rate": 5.884557721139431e-07,
      "loss": 1.14,
      "step": 3141
    },
    {
      "epoch": 299.23809523809524,
      "grad_norm": 6.371258735656738,
      "learning_rate": 5.886431784107947e-07,
      "loss": 1.1133,
      "step": 3142
    },
    {
      "epoch": 299.3333333333333,
      "grad_norm": 3.4509472846984863,
      "learning_rate": 5.888305847076462e-07,
      "loss": 1.1415,
      "step": 3143
    },
    {
      "epoch": 299.42857142857144,
      "grad_norm": 5.557456016540527,
      "learning_rate": 5.890179910044978e-07,
      "loss": 1.0862,
      "step": 3144
    },
    {
      "epoch": 299.5238095238095,
      "grad_norm": 10.220240592956543,
      "learning_rate": 5.892053973013493e-07,
      "loss": 1.1722,
      "step": 3145
    },
    {
      "epoch": 299.6190476190476,
      "grad_norm": 4.97195291519165,
      "learning_rate": 5.893928035982009e-07,
      "loss": 1.1453,
      "step": 3146
    },
    {
      "epoch": 299.7142857142857,
      "grad_norm": 4.511751174926758,
      "learning_rate": 5.895802098950524e-07,
      "loss": 1.1307,
      "step": 3147
    },
    {
      "epoch": 299.8095238095238,
      "grad_norm": 10.958107948303223,
      "learning_rate": 5.897676161919041e-07,
      "loss": 1.0468,
      "step": 3148
    },
    {
      "epoch": 299.9047619047619,
      "grad_norm": 5.010954856872559,
      "learning_rate": 5.899550224887557e-07,
      "loss": 1.1802,
      "step": 3149
    },
    {
      "epoch": 300.0,
      "grad_norm": 7.055208683013916,
      "learning_rate": 5.901424287856072e-07,
      "loss": 1.2058,
      "step": 3150
    },
    {
      "epoch": 300.0,
      "eval_accuracy": 0.7283626826526789,
      "eval_f1": 0.7820859633303276,
      "eval_loss": 0.5605196952819824,
      "eval_precision": 0.7296690970274817,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7803629821531376,
      "eval_runtime": 1.2717,
      "eval_samples_per_second": 2098.823,
      "eval_steps_per_second": 4.718,
      "step": 3150
    },
    {
      "epoch": 300.0952380952381,
      "grad_norm": 5.877634525299072,
      "learning_rate": 5.903298350824588e-07,
      "loss": 1.151,
      "step": 3151
    },
    {
      "epoch": 300.1904761904762,
      "grad_norm": 10.211005210876465,
      "learning_rate": 5.905172413793103e-07,
      "loss": 1.1311,
      "step": 3152
    },
    {
      "epoch": 300.2857142857143,
      "grad_norm": 4.744696140289307,
      "learning_rate": 5.907046476761619e-07,
      "loss": 1.1364,
      "step": 3153
    },
    {
      "epoch": 300.3809523809524,
      "grad_norm": 8.057610511779785,
      "learning_rate": 5.908920539730134e-07,
      "loss": 1.1331,
      "step": 3154
    },
    {
      "epoch": 300.4761904761905,
      "grad_norm": 4.582040786743164,
      "learning_rate": 5.910794602698651e-07,
      "loss": 1.0779,
      "step": 3155
    },
    {
      "epoch": 300.57142857142856,
      "grad_norm": 5.584441184997559,
      "learning_rate": 5.912668665667166e-07,
      "loss": 1.1095,
      "step": 3156
    },
    {
      "epoch": 300.6666666666667,
      "grad_norm": 5.46194314956665,
      "learning_rate": 5.914542728635682e-07,
      "loss": 1.1512,
      "step": 3157
    },
    {
      "epoch": 300.76190476190476,
      "grad_norm": 5.296677112579346,
      "learning_rate": 5.916416791604199e-07,
      "loss": 1.1492,
      "step": 3158
    },
    {
      "epoch": 300.85714285714283,
      "grad_norm": 4.495828151702881,
      "learning_rate": 5.918290854572714e-07,
      "loss": 1.1227,
      "step": 3159
    },
    {
      "epoch": 300.95238095238096,
      "grad_norm": 9.015779495239258,
      "learning_rate": 5.92016491754123e-07,
      "loss": 1.1717,
      "step": 3160
    },
    {
      "epoch": 300.95238095238096,
      "eval_accuracy": 0.7291120269763957,
      "eval_f1": 0.7841146610928635,
      "eval_loss": 0.5596321225166321,
      "eval_precision": 0.7274238227146814,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7817251007484168,
      "eval_runtime": 1.1674,
      "eval_samples_per_second": 2286.256,
      "eval_steps_per_second": 5.14,
      "step": 3160
    },
    {
      "epoch": 301.04761904761904,
      "grad_norm": 6.904094696044922,
      "learning_rate": 5.922038980509745e-07,
      "loss": 1.0834,
      "step": 3161
    },
    {
      "epoch": 301.14285714285717,
      "grad_norm": 8.023409843444824,
      "learning_rate": 5.923913043478262e-07,
      "loss": 1.1702,
      "step": 3162
    },
    {
      "epoch": 301.23809523809524,
      "grad_norm": 6.607522487640381,
      "learning_rate": 5.925787106446777e-07,
      "loss": 1.1145,
      "step": 3163
    },
    {
      "epoch": 301.3333333333333,
      "grad_norm": 6.27717399597168,
      "learning_rate": 5.927661169415293e-07,
      "loss": 1.1473,
      "step": 3164
    },
    {
      "epoch": 301.42857142857144,
      "grad_norm": 13.253412246704102,
      "learning_rate": 5.929535232383808e-07,
      "loss": 1.1361,
      "step": 3165
    },
    {
      "epoch": 301.5238095238095,
      "grad_norm": 6.1444807052612305,
      "learning_rate": 5.931409295352324e-07,
      "loss": 1.1103,
      "step": 3166
    },
    {
      "epoch": 301.6190476190476,
      "grad_norm": 4.564396381378174,
      "learning_rate": 5.93328335832084e-07,
      "loss": 1.1062,
      "step": 3167
    },
    {
      "epoch": 301.7142857142857,
      "grad_norm": 14.411751747131348,
      "learning_rate": 5.935157421289355e-07,
      "loss": 1.1852,
      "step": 3168
    },
    {
      "epoch": 301.8095238095238,
      "grad_norm": 6.965888977050781,
      "learning_rate": 5.937031484257872e-07,
      "loss": 1.1626,
      "step": 3169
    },
    {
      "epoch": 301.9047619047619,
      "grad_norm": 10.08370304107666,
      "learning_rate": 5.938905547226387e-07,
      "loss": 1.1333,
      "step": 3170
    },
    {
      "epoch": 302.0,
      "grad_norm": 6.450741291046143,
      "learning_rate": 5.940779610194903e-07,
      "loss": 1.1493,
      "step": 3171
    },
    {
      "epoch": 302.0,
      "eval_accuracy": 0.7234919445485201,
      "eval_f1": 0.7759562841530054,
      "eval_loss": 0.5597299933433533,
      "eval_precision": 0.7302857142857143,
      "eval_recall": 0.8277202072538861,
      "eval_roc_auc": 0.7810618883131837,
      "eval_runtime": 1.2015,
      "eval_samples_per_second": 2221.468,
      "eval_steps_per_second": 4.994,
      "step": 3171
    },
    {
      "epoch": 302.0952380952381,
      "grad_norm": 6.088359832763672,
      "learning_rate": 5.942653673163418e-07,
      "loss": 1.1616,
      "step": 3172
    },
    {
      "epoch": 302.1904761904762,
      "grad_norm": 8.05422306060791,
      "learning_rate": 5.944527736131934e-07,
      "loss": 1.1455,
      "step": 3173
    },
    {
      "epoch": 302.2857142857143,
      "grad_norm": 16.03872299194336,
      "learning_rate": 5.946401799100449e-07,
      "loss": 1.1452,
      "step": 3174
    },
    {
      "epoch": 302.3809523809524,
      "grad_norm": 8.73220443725586,
      "learning_rate": 5.948275862068965e-07,
      "loss": 1.1488,
      "step": 3175
    },
    {
      "epoch": 302.4761904761905,
      "grad_norm": 10.711005210876465,
      "learning_rate": 5.950149925037482e-07,
      "loss": 1.1715,
      "step": 3176
    },
    {
      "epoch": 302.57142857142856,
      "grad_norm": 12.020049095153809,
      "learning_rate": 5.952023988005997e-07,
      "loss": 1.1498,
      "step": 3177
    },
    {
      "epoch": 302.6666666666667,
      "grad_norm": 7.707366943359375,
      "learning_rate": 5.953898050974513e-07,
      "loss": 1.1374,
      "step": 3178
    },
    {
      "epoch": 302.76190476190476,
      "grad_norm": 6.902293682098389,
      "learning_rate": 5.955772113943028e-07,
      "loss": 1.1004,
      "step": 3179
    },
    {
      "epoch": 302.85714285714283,
      "grad_norm": 9.250264167785645,
      "learning_rate": 5.957646176911545e-07,
      "loss": 1.093,
      "step": 3180
    },
    {
      "epoch": 302.95238095238096,
      "grad_norm": 5.168294429779053,
      "learning_rate": 5.95952023988006e-07,
      "loss": 1.1286,
      "step": 3181
    },
    {
      "epoch": 302.95238095238096,
      "eval_accuracy": 0.7328587485949793,
      "eval_f1": 0.789861479516652,
      "eval_loss": 0.5604127645492554,
      "eval_precision": 0.7247160627366144,
      "eval_recall": 0.8678756476683938,
      "eval_roc_auc": 0.7815261945883708,
      "eval_runtime": 1.174,
      "eval_samples_per_second": 2273.494,
      "eval_steps_per_second": 5.111,
      "step": 3181
    },
    {
      "epoch": 303.04761904761904,
      "grad_norm": 6.529985427856445,
      "learning_rate": 5.961394302848576e-07,
      "loss": 1.122,
      "step": 3182
    },
    {
      "epoch": 303.14285714285717,
      "grad_norm": 10.856659889221191,
      "learning_rate": 5.963268365817093e-07,
      "loss": 1.1605,
      "step": 3183
    },
    {
      "epoch": 303.23809523809524,
      "grad_norm": 27.985136032104492,
      "learning_rate": 5.965142428785608e-07,
      "loss": 1.1823,
      "step": 3184
    },
    {
      "epoch": 303.3333333333333,
      "grad_norm": 11.863192558288574,
      "learning_rate": 5.967016491754124e-07,
      "loss": 1.0658,
      "step": 3185
    },
    {
      "epoch": 303.42857142857144,
      "grad_norm": 11.549949645996094,
      "learning_rate": 5.968890554722639e-07,
      "loss": 1.1329,
      "step": 3186
    },
    {
      "epoch": 303.5238095238095,
      "grad_norm": 27.801267623901367,
      "learning_rate": 5.970764617691155e-07,
      "loss": 1.187,
      "step": 3187
    },
    {
      "epoch": 303.6190476190476,
      "grad_norm": 12.57270622253418,
      "learning_rate": 5.97263868065967e-07,
      "loss": 1.1489,
      "step": 3188
    },
    {
      "epoch": 303.7142857142857,
      "grad_norm": 4.093427658081055,
      "learning_rate": 5.974512743628187e-07,
      "loss": 1.0962,
      "step": 3189
    },
    {
      "epoch": 303.8095238095238,
      "grad_norm": 4.537065029144287,
      "learning_rate": 5.976386806596702e-07,
      "loss": 1.1486,
      "step": 3190
    },
    {
      "epoch": 303.9047619047619,
      "grad_norm": 5.669375896453857,
      "learning_rate": 5.978260869565218e-07,
      "loss": 1.1027,
      "step": 3191
    },
    {
      "epoch": 304.0,
      "grad_norm": 5.12407922744751,
      "learning_rate": 5.980134932533733e-07,
      "loss": 1.1195,
      "step": 3192
    },
    {
      "epoch": 304.0,
      "eval_accuracy": 0.7276133383289621,
      "eval_f1": 0.778150747635032,
      "eval_loss": 0.5602059364318848,
      "eval_precision": 0.7357184073860358,
      "eval_recall": 0.8257772020725389,
      "eval_roc_auc": 0.7804870466321243,
      "eval_runtime": 1.1956,
      "eval_samples_per_second": 2232.4,
      "eval_steps_per_second": 5.019,
      "step": 3192
    },
    {
      "epoch": 304.0952380952381,
      "grad_norm": 4.144861698150635,
      "learning_rate": 5.982008995502249e-07,
      "loss": 1.1244,
      "step": 3193
    },
    {
      "epoch": 304.1904761904762,
      "grad_norm": 4.079629421234131,
      "learning_rate": 5.983883058470765e-07,
      "loss": 1.1458,
      "step": 3194
    },
    {
      "epoch": 304.2857142857143,
      "grad_norm": 18.062515258789062,
      "learning_rate": 5.98575712143928e-07,
      "loss": 1.1121,
      "step": 3195
    },
    {
      "epoch": 304.3809523809524,
      "grad_norm": 8.738750457763672,
      "learning_rate": 5.987631184407797e-07,
      "loss": 1.1344,
      "step": 3196
    },
    {
      "epoch": 304.4761904761905,
      "grad_norm": 7.128112316131592,
      "learning_rate": 5.989505247376312e-07,
      "loss": 1.1471,
      "step": 3197
    },
    {
      "epoch": 304.57142857142856,
      "grad_norm": 6.939492225646973,
      "learning_rate": 5.991379310344828e-07,
      "loss": 1.1812,
      "step": 3198
    },
    {
      "epoch": 304.6666666666667,
      "grad_norm": 4.064919948577881,
      "learning_rate": 5.993253373313343e-07,
      "loss": 1.1323,
      "step": 3199
    },
    {
      "epoch": 304.76190476190476,
      "grad_norm": 12.939764022827148,
      "learning_rate": 5.995127436281859e-07,
      "loss": 1.1151,
      "step": 3200
    },
    {
      "epoch": 304.85714285714283,
      "grad_norm": 4.833240509033203,
      "learning_rate": 5.997001499250374e-07,
      "loss": 1.1176,
      "step": 3201
    },
    {
      "epoch": 304.95238095238096,
      "grad_norm": 7.5120673179626465,
      "learning_rate": 5.998875562218891e-07,
      "loss": 1.1406,
      "step": 3202
    },
    {
      "epoch": 304.95238095238096,
      "eval_accuracy": 0.7336080929186961,
      "eval_f1": 0.7900797165633304,
      "eval_loss": 0.5607642531394958,
      "eval_precision": 0.7259902333152469,
      "eval_recall": 0.866580310880829,
      "eval_roc_auc": 0.7805192861255037,
      "eval_runtime": 1.3776,
      "eval_samples_per_second": 1937.392,
      "eval_steps_per_second": 4.355,
      "step": 3202
    },
    {
      "epoch": 305.04761904761904,
      "grad_norm": 12.696020126342773,
      "learning_rate": 6.000749625187407e-07,
      "loss": 1.1232,
      "step": 3203
    },
    {
      "epoch": 305.14285714285717,
      "grad_norm": 17.43992042541504,
      "learning_rate": 6.002623688155922e-07,
      "loss": 1.1179,
      "step": 3204
    },
    {
      "epoch": 305.23809523809524,
      "grad_norm": 5.639843940734863,
      "learning_rate": 6.004497751124439e-07,
      "loss": 1.0978,
      "step": 3205
    },
    {
      "epoch": 305.3333333333333,
      "grad_norm": 25.095577239990234,
      "learning_rate": 6.006371814092954e-07,
      "loss": 1.1403,
      "step": 3206
    },
    {
      "epoch": 305.42857142857144,
      "grad_norm": 10.123703956604004,
      "learning_rate": 6.00824587706147e-07,
      "loss": 1.102,
      "step": 3207
    },
    {
      "epoch": 305.5238095238095,
      "grad_norm": 6.3739824295043945,
      "learning_rate": 6.010119940029985e-07,
      "loss": 1.1449,
      "step": 3208
    },
    {
      "epoch": 305.6190476190476,
      "grad_norm": 6.56973123550415,
      "learning_rate": 6.011994002998501e-07,
      "loss": 1.1216,
      "step": 3209
    },
    {
      "epoch": 305.7142857142857,
      "grad_norm": 7.129731178283691,
      "learning_rate": 6.013868065967018e-07,
      "loss": 1.1629,
      "step": 3210
    },
    {
      "epoch": 305.8095238095238,
      "grad_norm": 3.856766700744629,
      "learning_rate": 6.015742128935533e-07,
      "loss": 1.1298,
      "step": 3211
    },
    {
      "epoch": 305.9047619047619,
      "grad_norm": 11.901754379272461,
      "learning_rate": 6.017616191904049e-07,
      "loss": 1.1581,
      "step": 3212
    },
    {
      "epoch": 306.0,
      "grad_norm": 6.4798054695129395,
      "learning_rate": 6.019490254872564e-07,
      "loss": 1.1326,
      "step": 3213
    },
    {
      "epoch": 306.0,
      "eval_accuracy": 0.7279880104908205,
      "eval_f1": 0.7779816513761468,
      "eval_loss": 0.5608794689178467,
      "eval_precision": 0.7369640787949016,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.7791991940126656,
      "eval_runtime": 1.1268,
      "eval_samples_per_second": 2368.694,
      "eval_steps_per_second": 5.325,
      "step": 3213
    },
    {
      "epoch": 306.0952380952381,
      "grad_norm": 11.762500762939453,
      "learning_rate": 6.02136431784108e-07,
      "loss": 1.1089,
      "step": 3214
    },
    {
      "epoch": 306.1904761904762,
      "grad_norm": 6.889684200286865,
      "learning_rate": 6.023238380809595e-07,
      "loss": 1.1317,
      "step": 3215
    },
    {
      "epoch": 306.2857142857143,
      "grad_norm": 8.107551574707031,
      "learning_rate": 6.025112443778111e-07,
      "loss": 1.1479,
      "step": 3216
    },
    {
      "epoch": 306.3809523809524,
      "grad_norm": 13.955403327941895,
      "learning_rate": 6.026986506746627e-07,
      "loss": 1.1003,
      "step": 3217
    },
    {
      "epoch": 306.4761904761905,
      "grad_norm": 5.956514358520508,
      "learning_rate": 6.028860569715143e-07,
      "loss": 1.1478,
      "step": 3218
    },
    {
      "epoch": 306.57142857142856,
      "grad_norm": 6.5395917892456055,
      "learning_rate": 6.030734632683658e-07,
      "loss": 1.1259,
      "step": 3219
    },
    {
      "epoch": 306.6666666666667,
      "grad_norm": 13.178661346435547,
      "learning_rate": 6.032608695652174e-07,
      "loss": 1.1703,
      "step": 3220
    },
    {
      "epoch": 306.76190476190476,
      "grad_norm": 8.879888534545898,
      "learning_rate": 6.03448275862069e-07,
      "loss": 1.1012,
      "step": 3221
    },
    {
      "epoch": 306.85714285714283,
      "grad_norm": 8.496637344360352,
      "learning_rate": 6.036356821589205e-07,
      "loss": 1.1188,
      "step": 3222
    },
    {
      "epoch": 306.95238095238096,
      "grad_norm": 12.254331588745117,
      "learning_rate": 6.038230884557721e-07,
      "loss": 1.1522,
      "step": 3223
    },
    {
      "epoch": 306.95238095238096,
      "eval_accuracy": 0.7328587485949793,
      "eval_f1": 0.7889908256880734,
      "eval_loss": 0.5601345896720886,
      "eval_precision": 0.7264305177111716,
      "eval_recall": 0.8633419689119171,
      "eval_roc_auc": 0.7810305123776626,
      "eval_runtime": 1.1948,
      "eval_samples_per_second": 2233.805,
      "eval_steps_per_second": 5.022,
      "step": 3223
    },
    {
      "epoch": 307.04761904761904,
      "grad_norm": 6.876726150512695,
      "learning_rate": 6.040104947526237e-07,
      "loss": 1.1352,
      "step": 3224
    },
    {
      "epoch": 307.14285714285717,
      "grad_norm": 6.194924354553223,
      "learning_rate": 6.041979010494753e-07,
      "loss": 1.1511,
      "step": 3225
    },
    {
      "epoch": 307.23809523809524,
      "grad_norm": 9.343500137329102,
      "learning_rate": 6.043853073463268e-07,
      "loss": 1.0997,
      "step": 3226
    },
    {
      "epoch": 307.3333333333333,
      "grad_norm": 3.359732151031494,
      "learning_rate": 6.045727136431785e-07,
      "loss": 1.1152,
      "step": 3227
    },
    {
      "epoch": 307.42857142857144,
      "grad_norm": 4.9511542320251465,
      "learning_rate": 6.0476011994003e-07,
      "loss": 1.1046,
      "step": 3228
    },
    {
      "epoch": 307.5238095238095,
      "grad_norm": 10.57309627532959,
      "learning_rate": 6.049475262368816e-07,
      "loss": 1.1745,
      "step": 3229
    },
    {
      "epoch": 307.6190476190476,
      "grad_norm": 5.695428848266602,
      "learning_rate": 6.051349325337332e-07,
      "loss": 1.1544,
      "step": 3230
    },
    {
      "epoch": 307.7142857142857,
      "grad_norm": 7.539062023162842,
      "learning_rate": 6.053223388305848e-07,
      "loss": 1.1166,
      "step": 3231
    },
    {
      "epoch": 307.8095238095238,
      "grad_norm": 5.299892425537109,
      "learning_rate": 6.055097451274364e-07,
      "loss": 1.144,
      "step": 3232
    },
    {
      "epoch": 307.9047619047619,
      "grad_norm": 9.088667869567871,
      "learning_rate": 6.056971514242879e-07,
      "loss": 1.1555,
      "step": 3233
    },
    {
      "epoch": 308.0,
      "grad_norm": 8.569293022155762,
      "learning_rate": 6.058845577211395e-07,
      "loss": 1.1036,
      "step": 3234
    },
    {
      "epoch": 308.0,
      "eval_accuracy": 0.7317347321094043,
      "eval_f1": 0.785242951409718,
      "eval_loss": 0.5596063137054443,
      "eval_precision": 0.7312849162011174,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7812645365572826,
      "eval_runtime": 1.2018,
      "eval_samples_per_second": 2220.77,
      "eval_steps_per_second": 4.992,
      "step": 3234
    },
    {
      "epoch": 308.0952380952381,
      "grad_norm": 4.837582111358643,
      "learning_rate": 6.06071964017991e-07,
      "loss": 1.1471,
      "step": 3235
    },
    {
      "epoch": 308.1904761904762,
      "grad_norm": 4.442275524139404,
      "learning_rate": 6.062593703148426e-07,
      "loss": 1.1361,
      "step": 3236
    },
    {
      "epoch": 308.2857142857143,
      "grad_norm": 11.222310066223145,
      "learning_rate": 6.064467766116941e-07,
      "loss": 1.1192,
      "step": 3237
    },
    {
      "epoch": 308.3809523809524,
      "grad_norm": 9.199163436889648,
      "learning_rate": 6.066341829085458e-07,
      "loss": 1.1205,
      "step": 3238
    },
    {
      "epoch": 308.4761904761905,
      "grad_norm": 18.25674057006836,
      "learning_rate": 6.068215892053974e-07,
      "loss": 1.0867,
      "step": 3239
    },
    {
      "epoch": 308.57142857142856,
      "grad_norm": 5.885474681854248,
      "learning_rate": 6.070089955022489e-07,
      "loss": 1.1149,
      "step": 3240
    },
    {
      "epoch": 308.6666666666667,
      "grad_norm": 4.786909580230713,
      "learning_rate": 6.071964017991005e-07,
      "loss": 1.1215,
      "step": 3241
    },
    {
      "epoch": 308.76190476190476,
      "grad_norm": 9.002260208129883,
      "learning_rate": 6.07383808095952e-07,
      "loss": 1.1238,
      "step": 3242
    },
    {
      "epoch": 308.85714285714283,
      "grad_norm": 6.030719757080078,
      "learning_rate": 6.075712143928036e-07,
      "loss": 1.1225,
      "step": 3243
    },
    {
      "epoch": 308.95238095238096,
      "grad_norm": 7.590175151824951,
      "learning_rate": 6.077586206896551e-07,
      "loss": 1.1342,
      "step": 3244
    },
    {
      "epoch": 308.95238095238096,
      "eval_accuracy": 0.7276133383289621,
      "eval_f1": 0.7798970632758099,
      "eval_loss": 0.5592300295829773,
      "eval_precision": 0.7322342239909039,
      "eval_recall": 0.8341968911917098,
      "eval_roc_auc": 0.7817251007484167,
      "eval_runtime": 1.1711,
      "eval_samples_per_second": 2278.968,
      "eval_steps_per_second": 5.123,
      "step": 3244
    },
    {
      "epoch": 309.04761904761904,
      "grad_norm": 4.260223865509033,
      "learning_rate": 6.079460269865068e-07,
      "loss": 1.1425,
      "step": 3245
    },
    {
      "epoch": 309.14285714285717,
      "grad_norm": 8.455692291259766,
      "learning_rate": 6.081334332833583e-07,
      "loss": 1.0991,
      "step": 3246
    },
    {
      "epoch": 309.23809523809524,
      "grad_norm": 6.05549955368042,
      "learning_rate": 6.083208395802099e-07,
      "loss": 1.1541,
      "step": 3247
    },
    {
      "epoch": 309.3333333333333,
      "grad_norm": 5.0326995849609375,
      "learning_rate": 6.085082458770615e-07,
      "loss": 1.1085,
      "step": 3248
    },
    {
      "epoch": 309.42857142857144,
      "grad_norm": 4.912335395812988,
      "learning_rate": 6.08695652173913e-07,
      "loss": 1.1493,
      "step": 3249
    },
    {
      "epoch": 309.5238095238095,
      "grad_norm": 10.788124084472656,
      "learning_rate": 6.088830584707647e-07,
      "loss": 1.1418,
      "step": 3250
    },
    {
      "epoch": 309.6190476190476,
      "grad_norm": 5.081390380859375,
      "learning_rate": 6.090704647676162e-07,
      "loss": 1.1463,
      "step": 3251
    },
    {
      "epoch": 309.7142857142857,
      "grad_norm": 5.906666278839111,
      "learning_rate": 6.092578710644679e-07,
      "loss": 1.1376,
      "step": 3252
    },
    {
      "epoch": 309.8095238095238,
      "grad_norm": 10.533183097839355,
      "learning_rate": 6.094452773613194e-07,
      "loss": 1.1166,
      "step": 3253
    },
    {
      "epoch": 309.9047619047619,
      "grad_norm": 9.069061279296875,
      "learning_rate": 6.09632683658171e-07,
      "loss": 1.1509,
      "step": 3254
    },
    {
      "epoch": 310.0,
      "grad_norm": 6.89107084274292,
      "learning_rate": 6.098200899550225e-07,
      "loss": 1.1169,
      "step": 3255
    },
    {
      "epoch": 310.0,
      "eval_accuracy": 0.729486699138254,
      "eval_f1": 0.7844776119402985,
      "eval_loss": 0.5588894486427307,
      "eval_precision": 0.7275747508305648,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.7822990788716178,
      "eval_runtime": 1.168,
      "eval_samples_per_second": 2285.1,
      "eval_steps_per_second": 5.137,
      "step": 3255
    },
    {
      "epoch": 310.0952380952381,
      "grad_norm": 8.010951042175293,
      "learning_rate": 6.100074962518741e-07,
      "loss": 1.1057,
      "step": 3256
    },
    {
      "epoch": 310.1904761904762,
      "grad_norm": 4.460325717926025,
      "learning_rate": 6.101949025487257e-07,
      "loss": 1.141,
      "step": 3257
    },
    {
      "epoch": 310.2857142857143,
      "grad_norm": 12.403966903686523,
      "learning_rate": 6.103823088455772e-07,
      "loss": 1.12,
      "step": 3258
    },
    {
      "epoch": 310.3809523809524,
      "grad_norm": 4.519659996032715,
      "learning_rate": 6.105697151424289e-07,
      "loss": 1.1342,
      "step": 3259
    },
    {
      "epoch": 310.4761904761905,
      "grad_norm": 9.70462703704834,
      "learning_rate": 6.107571214392804e-07,
      "loss": 1.1626,
      "step": 3260
    },
    {
      "epoch": 310.57142857142856,
      "grad_norm": 6.432512283325195,
      "learning_rate": 6.10944527736132e-07,
      "loss": 1.1871,
      "step": 3261
    },
    {
      "epoch": 310.6666666666667,
      "grad_norm": 5.8990254402160645,
      "learning_rate": 6.111319340329835e-07,
      "loss": 1.1867,
      "step": 3262
    },
    {
      "epoch": 310.76190476190476,
      "grad_norm": 11.098644256591797,
      "learning_rate": 6.113193403298351e-07,
      "loss": 1.1311,
      "step": 3263
    },
    {
      "epoch": 310.85714285714283,
      "grad_norm": 13.900815963745117,
      "learning_rate": 6.115067466266866e-07,
      "loss": 1.081,
      "step": 3264
    },
    {
      "epoch": 310.95238095238096,
      "grad_norm": 4.705795764923096,
      "learning_rate": 6.116941529235382e-07,
      "loss": 1.137,
      "step": 3265
    },
    {
      "epoch": 310.95238095238096,
      "eval_accuracy": 0.726489321843387,
      "eval_f1": 0.7802528597230584,
      "eval_loss": 0.5585808753967285,
      "eval_precision": 0.7289088863892014,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7824746689694876,
      "eval_runtime": 1.5471,
      "eval_samples_per_second": 1725.177,
      "eval_steps_per_second": 3.878,
      "step": 3265
    },
    {
      "epoch": 311.04761904761904,
      "grad_norm": 4.762983798980713,
      "learning_rate": 6.118815592203899e-07,
      "loss": 1.1286,
      "step": 3266
    },
    {
      "epoch": 311.14285714285717,
      "grad_norm": 4.509739398956299,
      "learning_rate": 6.120689655172414e-07,
      "loss": 1.0949,
      "step": 3267
    },
    {
      "epoch": 311.23809523809524,
      "grad_norm": 8.220304489135742,
      "learning_rate": 6.12256371814093e-07,
      "loss": 1.1219,
      "step": 3268
    },
    {
      "epoch": 311.3333333333333,
      "grad_norm": 4.448649883270264,
      "learning_rate": 6.124437781109445e-07,
      "loss": 1.1405,
      "step": 3269
    },
    {
      "epoch": 311.42857142857144,
      "grad_norm": 4.820764064788818,
      "learning_rate": 6.126311844077961e-07,
      "loss": 1.1275,
      "step": 3270
    },
    {
      "epoch": 311.5238095238095,
      "grad_norm": 5.165571212768555,
      "learning_rate": 6.128185907046476e-07,
      "loss": 1.0832,
      "step": 3271
    },
    {
      "epoch": 311.6190476190476,
      "grad_norm": 5.083733081817627,
      "learning_rate": 6.130059970014993e-07,
      "loss": 1.1247,
      "step": 3272
    },
    {
      "epoch": 311.7142857142857,
      "grad_norm": 6.846983909606934,
      "learning_rate": 6.131934032983508e-07,
      "loss": 1.1273,
      "step": 3273
    },
    {
      "epoch": 311.8095238095238,
      "grad_norm": 14.690434455871582,
      "learning_rate": 6.133808095952024e-07,
      "loss": 1.1779,
      "step": 3274
    },
    {
      "epoch": 311.9047619047619,
      "grad_norm": 4.378728866577148,
      "learning_rate": 6.135682158920541e-07,
      "loss": 1.1281,
      "step": 3275
    },
    {
      "epoch": 312.0,
      "grad_norm": 5.9867658615112305,
      "learning_rate": 6.137556221889056e-07,
      "loss": 1.1285,
      "step": 3276
    },
    {
      "epoch": 312.0,
      "eval_accuracy": 0.725365305357812,
      "eval_f1": 0.7787503773015394,
      "eval_loss": 0.5584920644760132,
      "eval_precision": 0.7292255511588468,
      "eval_recall": 0.8354922279792746,
      "eval_roc_auc": 0.7823529073114566,
      "eval_runtime": 1.2509,
      "eval_samples_per_second": 2133.686,
      "eval_steps_per_second": 4.797,
      "step": 3276
    },
    {
      "epoch": 312.0952380952381,
      "grad_norm": 6.125533580780029,
      "learning_rate": 6.139430284857572e-07,
      "loss": 1.1553,
      "step": 3277
    },
    {
      "epoch": 312.1904761904762,
      "grad_norm": 11.429598808288574,
      "learning_rate": 6.141304347826087e-07,
      "loss": 1.1018,
      "step": 3278
    },
    {
      "epoch": 312.2857142857143,
      "grad_norm": 13.404115676879883,
      "learning_rate": 6.143178410794603e-07,
      "loss": 1.1201,
      "step": 3279
    },
    {
      "epoch": 312.3809523809524,
      "grad_norm": 6.066341400146484,
      "learning_rate": 6.145052473763119e-07,
      "loss": 1.1588,
      "step": 3280
    },
    {
      "epoch": 312.4761904761905,
      "grad_norm": 5.60275411605835,
      "learning_rate": 6.146926536731635e-07,
      "loss": 1.1359,
      "step": 3281
    },
    {
      "epoch": 312.57142857142856,
      "grad_norm": 2.7833516597747803,
      "learning_rate": 6.14880059970015e-07,
      "loss": 1.1629,
      "step": 3282
    },
    {
      "epoch": 312.6666666666667,
      "grad_norm": 14.339527130126953,
      "learning_rate": 6.150674662668666e-07,
      "loss": 1.1026,
      "step": 3283
    },
    {
      "epoch": 312.76190476190476,
      "grad_norm": 10.421635627746582,
      "learning_rate": 6.152548725637182e-07,
      "loss": 1.1388,
      "step": 3284
    },
    {
      "epoch": 312.85714285714283,
      "grad_norm": 6.532027244567871,
      "learning_rate": 6.154422788605697e-07,
      "loss": 1.1199,
      "step": 3285
    },
    {
      "epoch": 312.95238095238096,
      "grad_norm": 10.787590980529785,
      "learning_rate": 6.156296851574213e-07,
      "loss": 1.1507,
      "step": 3286
    },
    {
      "epoch": 312.95238095238096,
      "eval_accuracy": 0.7261146496815286,
      "eval_f1": 0.777879064114251,
      "eval_loss": 0.5586705207824707,
      "eval_precision": 0.7326846021751574,
      "eval_recall": 0.8290155440414507,
      "eval_roc_auc": 0.7821442141623489,
      "eval_runtime": 1.214,
      "eval_samples_per_second": 2198.608,
      "eval_steps_per_second": 4.943,
      "step": 3286
    },
    {
      "epoch": 313.04761904761904,
      "grad_norm": 15.449256896972656,
      "learning_rate": 6.158170914542729e-07,
      "loss": 1.1215,
      "step": 3287
    },
    {
      "epoch": 313.14285714285717,
      "grad_norm": 9.633466720581055,
      "learning_rate": 6.160044977511245e-07,
      "loss": 1.1711,
      "step": 3288
    },
    {
      "epoch": 313.23809523809524,
      "grad_norm": 5.2529191970825195,
      "learning_rate": 6.16191904047976e-07,
      "loss": 1.1441,
      "step": 3289
    },
    {
      "epoch": 313.3333333333333,
      "grad_norm": 4.596034526824951,
      "learning_rate": 6.163793103448276e-07,
      "loss": 1.1165,
      "step": 3290
    },
    {
      "epoch": 313.42857142857144,
      "grad_norm": 4.310565948486328,
      "learning_rate": 6.165667166416791e-07,
      "loss": 1.1557,
      "step": 3291
    },
    {
      "epoch": 313.5238095238095,
      "grad_norm": 5.154080867767334,
      "learning_rate": 6.167541229385307e-07,
      "loss": 1.1244,
      "step": 3292
    },
    {
      "epoch": 313.6190476190476,
      "grad_norm": 6.325362205505371,
      "learning_rate": 6.169415292353823e-07,
      "loss": 1.0947,
      "step": 3293
    },
    {
      "epoch": 313.7142857142857,
      "grad_norm": 10.20145320892334,
      "learning_rate": 6.171289355322339e-07,
      "loss": 1.1397,
      "step": 3294
    },
    {
      "epoch": 313.8095238095238,
      "grad_norm": 5.5285844802856445,
      "learning_rate": 6.173163418290855e-07,
      "loss": 1.182,
      "step": 3295
    },
    {
      "epoch": 313.9047619047619,
      "grad_norm": 16.499744415283203,
      "learning_rate": 6.17503748125937e-07,
      "loss": 1.1293,
      "step": 3296
    },
    {
      "epoch": 314.0,
      "grad_norm": 12.139077186584473,
      "learning_rate": 6.176911544227887e-07,
      "loss": 1.1397,
      "step": 3297
    },
    {
      "epoch": 314.0,
      "eval_accuracy": 0.7272386661671038,
      "eval_f1": 0.7799274486094316,
      "eval_loss": 0.5583422183990479,
      "eval_precision": 0.7312925170068028,
      "eval_recall": 0.8354922279792746,
      "eval_roc_auc": 0.7827049510650548,
      "eval_runtime": 1.1746,
      "eval_samples_per_second": 2272.288,
      "eval_steps_per_second": 5.108,
      "step": 3297
    },
    {
      "epoch": 314.0952380952381,
      "grad_norm": 7.897889614105225,
      "learning_rate": 6.178785607196402e-07,
      "loss": 1.1594,
      "step": 3298
    },
    {
      "epoch": 314.1904761904762,
      "grad_norm": 6.885706901550293,
      "learning_rate": 6.180659670164918e-07,
      "loss": 1.1127,
      "step": 3299
    },
    {
      "epoch": 314.2857142857143,
      "grad_norm": 4.971257209777832,
      "learning_rate": 6.182533733133433e-07,
      "loss": 1.1407,
      "step": 3300
    },
    {
      "epoch": 314.3809523809524,
      "grad_norm": 7.690533638000488,
      "learning_rate": 6.18440779610195e-07,
      "loss": 1.1584,
      "step": 3301
    },
    {
      "epoch": 314.4761904761905,
      "grad_norm": 5.1898274421691895,
      "learning_rate": 6.186281859070466e-07,
      "loss": 1.1201,
      "step": 3302
    },
    {
      "epoch": 314.57142857142856,
      "grad_norm": 3.992206573486328,
      "learning_rate": 6.188155922038981e-07,
      "loss": 1.1321,
      "step": 3303
    },
    {
      "epoch": 314.6666666666667,
      "grad_norm": 7.562021255493164,
      "learning_rate": 6.190029985007497e-07,
      "loss": 1.0823,
      "step": 3304
    },
    {
      "epoch": 314.76190476190476,
      "grad_norm": 9.269352912902832,
      "learning_rate": 6.191904047976012e-07,
      "loss": 1.1033,
      "step": 3305
    },
    {
      "epoch": 314.85714285714283,
      "grad_norm": 11.719595909118652,
      "learning_rate": 6.193778110944528e-07,
      "loss": 1.1579,
      "step": 3306
    },
    {
      "epoch": 314.95238095238096,
      "grad_norm": 5.1851372718811035,
      "learning_rate": 6.195652173913043e-07,
      "loss": 1.1304,
      "step": 3307
    },
    {
      "epoch": 314.95238095238096,
      "eval_accuracy": 0.732484076433121,
      "eval_f1": 0.7872467222884386,
      "eval_loss": 0.5586660504341125,
      "eval_precision": 0.7290286975717439,
      "eval_recall": 0.8555699481865285,
      "eval_roc_auc": 0.7826174438687391,
      "eval_runtime": 1.1578,
      "eval_samples_per_second": 2305.332,
      "eval_steps_per_second": 5.182,
      "step": 3307
    },
    {
      "epoch": 315.04761904761904,
      "grad_norm": 10.430843353271484,
      "learning_rate": 6.19752623688156e-07,
      "loss": 1.0927,
      "step": 3308
    },
    {
      "epoch": 315.14285714285717,
      "grad_norm": 3.1264262199401855,
      "learning_rate": 6.199400299850076e-07,
      "loss": 1.1322,
      "step": 3309
    },
    {
      "epoch": 315.23809523809524,
      "grad_norm": 3.9145965576171875,
      "learning_rate": 6.201274362818591e-07,
      "loss": 1.138,
      "step": 3310
    },
    {
      "epoch": 315.3333333333333,
      "grad_norm": 3.780506134033203,
      "learning_rate": 6.203148425787107e-07,
      "loss": 1.146,
      "step": 3311
    },
    {
      "epoch": 315.42857142857144,
      "grad_norm": 2.8768436908721924,
      "learning_rate": 6.205022488755622e-07,
      "loss": 1.1304,
      "step": 3312
    },
    {
      "epoch": 315.5238095238095,
      "grad_norm": 4.919506549835205,
      "learning_rate": 6.206896551724138e-07,
      "loss": 1.1322,
      "step": 3313
    },
    {
      "epoch": 315.6190476190476,
      "grad_norm": 7.0171709060668945,
      "learning_rate": 6.208770614692653e-07,
      "loss": 1.0799,
      "step": 3314
    },
    {
      "epoch": 315.7142857142857,
      "grad_norm": 8.8346586227417,
      "learning_rate": 6.21064467766117e-07,
      "loss": 1.1269,
      "step": 3315
    },
    {
      "epoch": 315.8095238095238,
      "grad_norm": 11.773017883300781,
      "learning_rate": 6.212518740629685e-07,
      "loss": 1.1579,
      "step": 3316
    },
    {
      "epoch": 315.9047619047619,
      "grad_norm": 4.387534141540527,
      "learning_rate": 6.214392803598201e-07,
      "loss": 1.1322,
      "step": 3317
    },
    {
      "epoch": 316.0,
      "grad_norm": 4.816924571990967,
      "learning_rate": 6.216266866566716e-07,
      "loss": 1.1106,
      "step": 3318
    },
    {
      "epoch": 316.0,
      "eval_accuracy": 0.7279880104908205,
      "eval_f1": 0.7806646525679758,
      "eval_loss": 0.5587853193283081,
      "eval_precision": 0.7315968289920725,
      "eval_recall": 0.8367875647668394,
      "eval_roc_auc": 0.7817772020725389,
      "eval_runtime": 1.3043,
      "eval_samples_per_second": 2046.246,
      "eval_steps_per_second": 4.6,
      "step": 3318
    },
    {
      "epoch": 316.0952380952381,
      "grad_norm": 5.566372871398926,
      "learning_rate": 6.218140929535232e-07,
      "loss": 1.1183,
      "step": 3319
    },
    {
      "epoch": 316.1904761904762,
      "grad_norm": 6.3373942375183105,
      "learning_rate": 6.220014992503749e-07,
      "loss": 1.0724,
      "step": 3320
    },
    {
      "epoch": 316.2857142857143,
      "grad_norm": 7.2872233390808105,
      "learning_rate": 6.221889055472264e-07,
      "loss": 1.1399,
      "step": 3321
    },
    {
      "epoch": 316.3809523809524,
      "grad_norm": 7.598955154418945,
      "learning_rate": 6.223763118440781e-07,
      "loss": 1.1447,
      "step": 3322
    },
    {
      "epoch": 316.4761904761905,
      "grad_norm": 7.6095991134643555,
      "learning_rate": 6.225637181409296e-07,
      "loss": 1.1293,
      "step": 3323
    },
    {
      "epoch": 316.57142857142856,
      "grad_norm": 6.845755577087402,
      "learning_rate": 6.227511244377812e-07,
      "loss": 1.1348,
      "step": 3324
    },
    {
      "epoch": 316.6666666666667,
      "grad_norm": 3.707514762878418,
      "learning_rate": 6.229385307346327e-07,
      "loss": 1.125,
      "step": 3325
    },
    {
      "epoch": 316.76190476190476,
      "grad_norm": 8.177051544189453,
      "learning_rate": 6.231259370314843e-07,
      "loss": 1.1184,
      "step": 3326
    },
    {
      "epoch": 316.85714285714283,
      "grad_norm": 9.464016914367676,
      "learning_rate": 6.233133433283358e-07,
      "loss": 1.0847,
      "step": 3327
    },
    {
      "epoch": 316.95238095238096,
      "grad_norm": 9.221125602722168,
      "learning_rate": 6.235007496251874e-07,
      "loss": 1.1479,
      "step": 3328
    },
    {
      "epoch": 316.95238095238096,
      "eval_accuracy": 0.7272386661671038,
      "eval_f1": 0.7809867629362214,
      "eval_loss": 0.5587996244430542,
      "eval_precision": 0.7292134831460674,
      "eval_recall": 0.8406735751295337,
      "eval_roc_auc": 0.7818117443868737,
      "eval_runtime": 1.1582,
      "eval_samples_per_second": 2304.423,
      "eval_steps_per_second": 5.18,
      "step": 3328
    },
    {
      "epoch": 317.04761904761904,
      "grad_norm": 4.902645111083984,
      "learning_rate": 6.236881559220391e-07,
      "loss": 1.1764,
      "step": 3329
    },
    {
      "epoch": 317.14285714285717,
      "grad_norm": 4.456246852874756,
      "learning_rate": 6.238755622188906e-07,
      "loss": 1.112,
      "step": 3330
    },
    {
      "epoch": 317.23809523809524,
      "grad_norm": 11.598137855529785,
      "learning_rate": 6.240629685157422e-07,
      "loss": 1.0803,
      "step": 3331
    },
    {
      "epoch": 317.3333333333333,
      "grad_norm": 5.1275529861450195,
      "learning_rate": 6.242503748125937e-07,
      "loss": 1.1288,
      "step": 3332
    },
    {
      "epoch": 317.42857142857144,
      "grad_norm": 11.168167114257812,
      "learning_rate": 6.244377811094453e-07,
      "loss": 1.1478,
      "step": 3333
    },
    {
      "epoch": 317.5238095238095,
      "grad_norm": 4.918898105621338,
      "learning_rate": 6.246251874062968e-07,
      "loss": 1.1507,
      "step": 3334
    },
    {
      "epoch": 317.6190476190476,
      "grad_norm": 11.758259773254395,
      "learning_rate": 6.248125937031484e-07,
      "loss": 1.1648,
      "step": 3335
    },
    {
      "epoch": 317.7142857142857,
      "grad_norm": 3.0417706966400146,
      "learning_rate": 6.25e-07,
      "loss": 1.1398,
      "step": 3336
    },
    {
      "epoch": 317.8095238095238,
      "grad_norm": 9.044265747070312,
      "learning_rate": 6.251874062968516e-07,
      "loss": 1.128,
      "step": 3337
    },
    {
      "epoch": 317.9047619047619,
      "grad_norm": 6.247106075286865,
      "learning_rate": 6.253748125937032e-07,
      "loss": 1.13,
      "step": 3338
    },
    {
      "epoch": 318.0,
      "grad_norm": 11.326921463012695,
      "learning_rate": 6.255622188905547e-07,
      "loss": 1.0818,
      "step": 3339
    },
    {
      "epoch": 318.0,
      "eval_accuracy": 0.7279880104908205,
      "eval_f1": 0.7795992714025501,
      "eval_loss": 0.5579902529716492,
      "eval_precision": 0.7337142857142858,
      "eval_recall": 0.8316062176165803,
      "eval_roc_auc": 0.7827248128957974,
      "eval_runtime": 1.1928,
      "eval_samples_per_second": 2237.596,
      "eval_steps_per_second": 5.03,
      "step": 3339
    },
    {
      "epoch": 318.0952380952381,
      "grad_norm": 8.263168334960938,
      "learning_rate": 6.257496251874063e-07,
      "loss": 1.1696,
      "step": 3340
    },
    {
      "epoch": 318.1904761904762,
      "grad_norm": 5.750796794891357,
      "learning_rate": 6.259370314842578e-07,
      "loss": 1.1072,
      "step": 3341
    },
    {
      "epoch": 318.2857142857143,
      "grad_norm": 3.7694246768951416,
      "learning_rate": 6.261244377811095e-07,
      "loss": 1.1296,
      "step": 3342
    },
    {
      "epoch": 318.3809523809524,
      "grad_norm": 3.5072813034057617,
      "learning_rate": 6.26311844077961e-07,
      "loss": 1.1201,
      "step": 3343
    },
    {
      "epoch": 318.4761904761905,
      "grad_norm": 4.239100933074951,
      "learning_rate": 6.264992503748126e-07,
      "loss": 1.1158,
      "step": 3344
    },
    {
      "epoch": 318.57142857142856,
      "grad_norm": 9.425958633422852,
      "learning_rate": 6.266866566716643e-07,
      "loss": 1.0952,
      "step": 3345
    },
    {
      "epoch": 318.6666666666667,
      "grad_norm": 5.783169269561768,
      "learning_rate": 6.268740629685158e-07,
      "loss": 1.1676,
      "step": 3346
    },
    {
      "epoch": 318.76190476190476,
      "grad_norm": 4.611095428466797,
      "learning_rate": 6.270614692653674e-07,
      "loss": 1.0811,
      "step": 3347
    },
    {
      "epoch": 318.85714285714283,
      "grad_norm": 10.244481086730957,
      "learning_rate": 6.272488755622189e-07,
      "loss": 1.0756,
      "step": 3348
    },
    {
      "epoch": 318.95238095238096,
      "grad_norm": 15.405163764953613,
      "learning_rate": 6.274362818590705e-07,
      "loss": 1.161,
      "step": 3349
    },
    {
      "epoch": 318.95238095238096,
      "eval_accuracy": 0.7328587485949793,
      "eval_f1": 0.7859501651155809,
      "eval_loss": 0.5577879548072815,
      "eval_precision": 0.7325125909345271,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.783233448474381,
      "eval_runtime": 1.2591,
      "eval_samples_per_second": 2119.693,
      "eval_steps_per_second": 4.765,
      "step": 3349
    },
    {
      "epoch": 319.04761904761904,
      "grad_norm": 6.132430076599121,
      "learning_rate": 6.276236881559221e-07,
      "loss": 1.176,
      "step": 3350
    },
    {
      "epoch": 319.14285714285717,
      "grad_norm": 6.559388160705566,
      "learning_rate": 6.278110944527737e-07,
      "loss": 1.158,
      "step": 3351
    },
    {
      "epoch": 319.23809523809524,
      "grad_norm": 7.797235488891602,
      "learning_rate": 6.279985007496252e-07,
      "loss": 1.1378,
      "step": 3352
    },
    {
      "epoch": 319.3333333333333,
      "grad_norm": 6.024532794952393,
      "learning_rate": 6.281859070464768e-07,
      "loss": 1.1724,
      "step": 3353
    },
    {
      "epoch": 319.42857142857144,
      "grad_norm": 6.553297996520996,
      "learning_rate": 6.283733133433283e-07,
      "loss": 1.1502,
      "step": 3354
    },
    {
      "epoch": 319.5238095238095,
      "grad_norm": 5.962327480316162,
      "learning_rate": 6.285607196401799e-07,
      "loss": 1.1233,
      "step": 3355
    },
    {
      "epoch": 319.6190476190476,
      "grad_norm": 5.1203837394714355,
      "learning_rate": 6.287481259370315e-07,
      "loss": 1.1058,
      "step": 3356
    },
    {
      "epoch": 319.7142857142857,
      "grad_norm": 5.673627853393555,
      "learning_rate": 6.289355322338831e-07,
      "loss": 1.1309,
      "step": 3357
    },
    {
      "epoch": 319.8095238095238,
      "grad_norm": 6.080491065979004,
      "learning_rate": 6.291229385307347e-07,
      "loss": 1.1293,
      "step": 3358
    },
    {
      "epoch": 319.9047619047619,
      "grad_norm": 10.415179252624512,
      "learning_rate": 6.293103448275862e-07,
      "loss": 1.1229,
      "step": 3359
    },
    {
      "epoch": 320.0,
      "grad_norm": 6.787259578704834,
      "learning_rate": 6.294977511244378e-07,
      "loss": 1.092,
      "step": 3360
    },
    {
      "epoch": 320.0,
      "eval_accuracy": 0.7321094042712627,
      "eval_f1": 0.7858640311470501,
      "eval_loss": 0.5576049089431763,
      "eval_precision": 0.7309192200557103,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.7837668393782383,
      "eval_runtime": 1.3345,
      "eval_samples_per_second": 2000.012,
      "eval_steps_per_second": 4.496,
      "step": 3360
    },
    {
      "epoch": 320.0952380952381,
      "grad_norm": 10.310934066772461,
      "learning_rate": 6.296851574212893e-07,
      "loss": 1.1127,
      "step": 3361
    },
    {
      "epoch": 320.1904761904762,
      "grad_norm": 10.253336906433105,
      "learning_rate": 6.298725637181409e-07,
      "loss": 1.1832,
      "step": 3362
    },
    {
      "epoch": 320.2857142857143,
      "grad_norm": 7.179525852203369,
      "learning_rate": 6.300599700149924e-07,
      "loss": 1.111,
      "step": 3363
    },
    {
      "epoch": 320.3809523809524,
      "grad_norm": 6.53006649017334,
      "learning_rate": 6.302473763118441e-07,
      "loss": 1.1105,
      "step": 3364
    },
    {
      "epoch": 320.4761904761905,
      "grad_norm": 7.752871513366699,
      "learning_rate": 6.304347826086957e-07,
      "loss": 1.1257,
      "step": 3365
    },
    {
      "epoch": 320.57142857142856,
      "grad_norm": 9.54736042022705,
      "learning_rate": 6.306221889055472e-07,
      "loss": 1.1221,
      "step": 3366
    },
    {
      "epoch": 320.6666666666667,
      "grad_norm": 12.933712005615234,
      "learning_rate": 6.308095952023989e-07,
      "loss": 1.1537,
      "step": 3367
    },
    {
      "epoch": 320.76190476190476,
      "grad_norm": 14.127771377563477,
      "learning_rate": 6.309970014992504e-07,
      "loss": 1.131,
      "step": 3368
    },
    {
      "epoch": 320.85714285714283,
      "grad_norm": 10.824623107910156,
      "learning_rate": 6.31184407796102e-07,
      "loss": 1.1564,
      "step": 3369
    },
    {
      "epoch": 320.95238095238096,
      "grad_norm": 7.595580577850342,
      "learning_rate": 6.313718140929535e-07,
      "loss": 1.114,
      "step": 3370
    },
    {
      "epoch": 320.95238095238096,
      "eval_accuracy": 0.7328587485949793,
      "eval_f1": 0.7854348480288895,
      "eval_loss": 0.5570670962333679,
      "eval_precision": 0.7335581787521079,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7843957973517559,
      "eval_runtime": 1.1426,
      "eval_samples_per_second": 2335.936,
      "eval_steps_per_second": 5.251,
      "step": 3370
    },
    {
      "epoch": 321.04761904761904,
      "grad_norm": 4.967525005340576,
      "learning_rate": 6.315592203898052e-07,
      "loss": 1.1011,
      "step": 3371
    },
    {
      "epoch": 321.14285714285717,
      "grad_norm": 6.865474224090576,
      "learning_rate": 6.317466266866568e-07,
      "loss": 1.1482,
      "step": 3372
    },
    {
      "epoch": 321.23809523809524,
      "grad_norm": 5.231659412384033,
      "learning_rate": 6.319340329835083e-07,
      "loss": 1.1513,
      "step": 3373
    },
    {
      "epoch": 321.3333333333333,
      "grad_norm": 8.249855995178223,
      "learning_rate": 6.321214392803599e-07,
      "loss": 1.1368,
      "step": 3374
    },
    {
      "epoch": 321.42857142857144,
      "grad_norm": 5.477447032928467,
      "learning_rate": 6.323088455772114e-07,
      "loss": 1.1297,
      "step": 3375
    },
    {
      "epoch": 321.5238095238095,
      "grad_norm": 7.059687614440918,
      "learning_rate": 6.32496251874063e-07,
      "loss": 1.089,
      "step": 3376
    },
    {
      "epoch": 321.6190476190476,
      "grad_norm": 7.401088714599609,
      "learning_rate": 6.326836581709145e-07,
      "loss": 1.1318,
      "step": 3377
    },
    {
      "epoch": 321.7142857142857,
      "grad_norm": 7.332993984222412,
      "learning_rate": 6.328710644677662e-07,
      "loss": 1.1427,
      "step": 3378
    },
    {
      "epoch": 321.8095238095238,
      "grad_norm": 5.104832649230957,
      "learning_rate": 6.330584707646177e-07,
      "loss": 1.129,
      "step": 3379
    },
    {
      "epoch": 321.9047619047619,
      "grad_norm": 5.148171424865723,
      "learning_rate": 6.332458770614693e-07,
      "loss": 1.1143,
      "step": 3380
    },
    {
      "epoch": 322.0,
      "grad_norm": 12.024181365966797,
      "learning_rate": 6.334332833583208e-07,
      "loss": 1.1546,
      "step": 3381
    },
    {
      "epoch": 322.0,
      "eval_accuracy": 0.729486699138254,
      "eval_f1": 0.7810794420861128,
      "eval_loss": 0.5574794411659241,
      "eval_precision": 0.734321550741163,
      "eval_recall": 0.8341968911917098,
      "eval_roc_auc": 0.7835354058721935,
      "eval_runtime": 1.1773,
      "eval_samples_per_second": 2266.969,
      "eval_steps_per_second": 5.096,
      "step": 3381
    },
    {
      "epoch": 322.0952380952381,
      "grad_norm": 8.503469467163086,
      "learning_rate": 6.336206896551724e-07,
      "loss": 1.1533,
      "step": 3382
    },
    {
      "epoch": 322.1904761904762,
      "grad_norm": 13.482050895690918,
      "learning_rate": 6.33808095952024e-07,
      "loss": 1.0889,
      "step": 3383
    },
    {
      "epoch": 322.2857142857143,
      "grad_norm": 3.6773922443389893,
      "learning_rate": 6.339955022488755e-07,
      "loss": 1.1106,
      "step": 3384
    },
    {
      "epoch": 322.3809523809524,
      "grad_norm": 9.904181480407715,
      "learning_rate": 6.341829085457272e-07,
      "loss": 1.1195,
      "step": 3385
    },
    {
      "epoch": 322.4761904761905,
      "grad_norm": 11.262557029724121,
      "learning_rate": 6.343703148425787e-07,
      "loss": 1.0854,
      "step": 3386
    },
    {
      "epoch": 322.57142857142856,
      "grad_norm": 8.886178016662598,
      "learning_rate": 6.345577211394303e-07,
      "loss": 1.1319,
      "step": 3387
    },
    {
      "epoch": 322.6666666666667,
      "grad_norm": 10.291092872619629,
      "learning_rate": 6.347451274362818e-07,
      "loss": 1.1216,
      "step": 3388
    },
    {
      "epoch": 322.76190476190476,
      "grad_norm": 8.690281867980957,
      "learning_rate": 6.349325337331335e-07,
      "loss": 1.1387,
      "step": 3389
    },
    {
      "epoch": 322.85714285714283,
      "grad_norm": 5.215178966522217,
      "learning_rate": 6.35119940029985e-07,
      "loss": 1.1351,
      "step": 3390
    },
    {
      "epoch": 322.95238095238096,
      "grad_norm": 3.838747501373291,
      "learning_rate": 6.353073463268366e-07,
      "loss": 1.1343,
      "step": 3391
    },
    {
      "epoch": 322.95238095238096,
      "eval_accuracy": 0.729486699138254,
      "eval_f1": 0.782922429344558,
      "eval_loss": 0.5572158694267273,
      "eval_precision": 0.7306397306397306,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7839916522740358,
      "eval_runtime": 1.2145,
      "eval_samples_per_second": 2197.652,
      "eval_steps_per_second": 4.94,
      "step": 3391
    },
    {
      "epoch": 323.04761904761904,
      "grad_norm": 5.711175441741943,
      "learning_rate": 6.354947526236883e-07,
      "loss": 1.1285,
      "step": 3392
    },
    {
      "epoch": 323.14285714285717,
      "grad_norm": 7.926575183868408,
      "learning_rate": 6.356821589205398e-07,
      "loss": 1.1537,
      "step": 3393
    },
    {
      "epoch": 323.23809523809524,
      "grad_norm": 8.844377517700195,
      "learning_rate": 6.358695652173914e-07,
      "loss": 1.1409,
      "step": 3394
    },
    {
      "epoch": 323.3333333333333,
      "grad_norm": 13.982290267944336,
      "learning_rate": 6.360569715142429e-07,
      "loss": 1.1452,
      "step": 3395
    },
    {
      "epoch": 323.42857142857144,
      "grad_norm": 6.307914733886719,
      "learning_rate": 6.362443778110945e-07,
      "loss": 1.0981,
      "step": 3396
    },
    {
      "epoch": 323.5238095238095,
      "grad_norm": 3.3946173191070557,
      "learning_rate": 6.36431784107946e-07,
      "loss": 1.0955,
      "step": 3397
    },
    {
      "epoch": 323.6190476190476,
      "grad_norm": 13.919370651245117,
      "learning_rate": 6.366191904047976e-07,
      "loss": 1.1081,
      "step": 3398
    },
    {
      "epoch": 323.7142857142857,
      "grad_norm": 9.541321754455566,
      "learning_rate": 6.368065967016493e-07,
      "loss": 1.159,
      "step": 3399
    },
    {
      "epoch": 323.8095238095238,
      "grad_norm": 7.30950927734375,
      "learning_rate": 6.369940029985008e-07,
      "loss": 1.1319,
      "step": 3400
    },
    {
      "epoch": 323.9047619047619,
      "grad_norm": 4.72685432434082,
      "learning_rate": 6.371814092953524e-07,
      "loss": 1.1306,
      "step": 3401
    },
    {
      "epoch": 324.0,
      "grad_norm": 7.5063090324401855,
      "learning_rate": 6.373688155922039e-07,
      "loss": 1.1321,
      "step": 3402
    },
    {
      "epoch": 324.0,
      "eval_accuracy": 0.7306107156238292,
      "eval_f1": 0.7824508320726172,
      "eval_loss": 0.5564565658569336,
      "eval_precision": 0.7342419080068143,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7845097869890616,
      "eval_runtime": 1.2683,
      "eval_samples_per_second": 2104.47,
      "eval_steps_per_second": 4.731,
      "step": 3402
    },
    {
      "epoch": 324.0952380952381,
      "grad_norm": 11.140090942382812,
      "learning_rate": 6.375562218890555e-07,
      "loss": 1.1572,
      "step": 3403
    },
    {
      "epoch": 324.1904761904762,
      "grad_norm": 4.548201560974121,
      "learning_rate": 6.37743628185907e-07,
      "loss": 1.1466,
      "step": 3404
    },
    {
      "epoch": 324.2857142857143,
      "grad_norm": 10.859220504760742,
      "learning_rate": 6.379310344827587e-07,
      "loss": 1.0967,
      "step": 3405
    },
    {
      "epoch": 324.3809523809524,
      "grad_norm": 6.48252534866333,
      "learning_rate": 6.381184407796102e-07,
      "loss": 1.1105,
      "step": 3406
    },
    {
      "epoch": 324.4761904761905,
      "grad_norm": 5.144731044769287,
      "learning_rate": 6.383058470764618e-07,
      "loss": 1.1343,
      "step": 3407
    },
    {
      "epoch": 324.57142857142856,
      "grad_norm": 8.819598197937012,
      "learning_rate": 6.384932533733134e-07,
      "loss": 1.1031,
      "step": 3408
    },
    {
      "epoch": 324.6666666666667,
      "grad_norm": 3.8726017475128174,
      "learning_rate": 6.386806596701649e-07,
      "loss": 1.1425,
      "step": 3409
    },
    {
      "epoch": 324.76190476190476,
      "grad_norm": 9.518330574035645,
      "learning_rate": 6.388680659670165e-07,
      "loss": 1.0801,
      "step": 3410
    },
    {
      "epoch": 324.85714285714283,
      "grad_norm": 10.46760368347168,
      "learning_rate": 6.39055472263868e-07,
      "loss": 1.1716,
      "step": 3411
    },
    {
      "epoch": 324.95238095238096,
      "grad_norm": 6.5744524002075195,
      "learning_rate": 6.392428785607197e-07,
      "loss": 1.1494,
      "step": 3412
    },
    {
      "epoch": 324.95238095238096,
      "eval_accuracy": 0.7287373548145373,
      "eval_f1": 0.781664656212304,
      "eval_loss": 0.5566065311431885,
      "eval_precision": 0.7313769751693002,
      "eval_recall": 0.8393782383419689,
      "eval_roc_auc": 0.7841865284974092,
      "eval_runtime": 1.1265,
      "eval_samples_per_second": 2369.223,
      "eval_steps_per_second": 5.326,
      "step": 3412
    },
    {
      "epoch": 325.04761904761904,
      "grad_norm": 5.198469161987305,
      "learning_rate": 6.394302848575712e-07,
      "loss": 1.1435,
      "step": 3413
    },
    {
      "epoch": 325.14285714285717,
      "grad_norm": 7.31497049331665,
      "learning_rate": 6.396176911544229e-07,
      "loss": 1.1152,
      "step": 3414
    },
    {
      "epoch": 325.23809523809524,
      "grad_norm": 3.6078991889953613,
      "learning_rate": 6.398050974512744e-07,
      "loss": 1.1313,
      "step": 3415
    },
    {
      "epoch": 325.3333333333333,
      "grad_norm": 15.786949157714844,
      "learning_rate": 6.39992503748126e-07,
      "loss": 1.1381,
      "step": 3416
    },
    {
      "epoch": 325.42857142857144,
      "grad_norm": 7.494317054748535,
      "learning_rate": 6.401799100449776e-07,
      "loss": 1.1206,
      "step": 3417
    },
    {
      "epoch": 325.5238095238095,
      "grad_norm": 6.337263584136963,
      "learning_rate": 6.403673163418291e-07,
      "loss": 1.1413,
      "step": 3418
    },
    {
      "epoch": 325.6190476190476,
      "grad_norm": 9.863685607910156,
      "learning_rate": 6.405547226386808e-07,
      "loss": 1.1607,
      "step": 3419
    },
    {
      "epoch": 325.7142857142857,
      "grad_norm": 5.73906946182251,
      "learning_rate": 6.407421289355323e-07,
      "loss": 1.091,
      "step": 3420
    },
    {
      "epoch": 325.8095238095238,
      "grad_norm": 5.083852291107178,
      "learning_rate": 6.409295352323839e-07,
      "loss": 1.1101,
      "step": 3421
    },
    {
      "epoch": 325.9047619047619,
      "grad_norm": 5.438633441925049,
      "learning_rate": 6.411169415292354e-07,
      "loss": 1.0855,
      "step": 3422
    },
    {
      "epoch": 326.0,
      "grad_norm": 17.124401092529297,
      "learning_rate": 6.41304347826087e-07,
      "loss": 1.1609,
      "step": 3423
    },
    {
      "epoch": 326.0,
      "eval_accuracy": 0.7309853877856876,
      "eval_f1": 0.7850299401197605,
      "eval_loss": 0.5572195649147034,
      "eval_precision": 0.7299554565701559,
      "eval_recall": 0.8490932642487047,
      "eval_roc_auc": 0.7838831318364998,
      "eval_runtime": 1.1314,
      "eval_samples_per_second": 2359.053,
      "eval_steps_per_second": 5.303,
      "step": 3423
    },
    {
      "epoch": 326.0952380952381,
      "grad_norm": 5.153423309326172,
      "learning_rate": 6.414917541229385e-07,
      "loss": 1.0732,
      "step": 3424
    },
    {
      "epoch": 326.1904761904762,
      "grad_norm": 5.137463092803955,
      "learning_rate": 6.416791604197901e-07,
      "loss": 1.1078,
      "step": 3425
    },
    {
      "epoch": 326.2857142857143,
      "grad_norm": 6.32768440246582,
      "learning_rate": 6.418665667166418e-07,
      "loss": 1.159,
      "step": 3426
    },
    {
      "epoch": 326.3809523809524,
      "grad_norm": 6.702006816864014,
      "learning_rate": 6.420539730134933e-07,
      "loss": 1.1379,
      "step": 3427
    },
    {
      "epoch": 326.4761904761905,
      "grad_norm": 5.366683483123779,
      "learning_rate": 6.422413793103449e-07,
      "loss": 1.1531,
      "step": 3428
    },
    {
      "epoch": 326.57142857142856,
      "grad_norm": 6.240462779998779,
      "learning_rate": 6.424287856071964e-07,
      "loss": 1.1293,
      "step": 3429
    },
    {
      "epoch": 326.6666666666667,
      "grad_norm": 6.877244472503662,
      "learning_rate": 6.42616191904048e-07,
      "loss": 1.1515,
      "step": 3430
    },
    {
      "epoch": 326.76190476190476,
      "grad_norm": 5.66732120513916,
      "learning_rate": 6.428035982008995e-07,
      "loss": 1.1392,
      "step": 3431
    },
    {
      "epoch": 326.85714285714283,
      "grad_norm": 4.892309188842773,
      "learning_rate": 6.429910044977511e-07,
      "loss": 1.0918,
      "step": 3432
    },
    {
      "epoch": 326.95238095238096,
      "grad_norm": 7.027184963226318,
      "learning_rate": 6.431784107946027e-07,
      "loss": 1.164,
      "step": 3433
    },
    {
      "epoch": 326.95238095238096,
      "eval_accuracy": 0.732484076433121,
      "eval_f1": 0.7851985559566786,
      "eval_loss": 0.5570408701896667,
      "eval_precision": 0.7331460674157303,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7840287852619459,
      "eval_runtime": 1.1261,
      "eval_samples_per_second": 2370.097,
      "eval_steps_per_second": 5.328,
      "step": 3433
    },
    {
      "epoch": 327.04761904761904,
      "grad_norm": 7.328666687011719,
      "learning_rate": 6.433658170914543e-07,
      "loss": 1.1519,
      "step": 3434
    },
    {
      "epoch": 327.14285714285717,
      "grad_norm": 8.158414840698242,
      "learning_rate": 6.435532233883059e-07,
      "loss": 1.1626,
      "step": 3435
    },
    {
      "epoch": 327.23809523809524,
      "grad_norm": 7.522820949554443,
      "learning_rate": 6.437406296851574e-07,
      "loss": 1.1562,
      "step": 3436
    },
    {
      "epoch": 327.3333333333333,
      "grad_norm": 5.471183776855469,
      "learning_rate": 6.439280359820091e-07,
      "loss": 1.1115,
      "step": 3437
    },
    {
      "epoch": 327.42857142857144,
      "grad_norm": 13.796281814575195,
      "learning_rate": 6.441154422788606e-07,
      "loss": 1.0954,
      "step": 3438
    },
    {
      "epoch": 327.5238095238095,
      "grad_norm": 9.086668014526367,
      "learning_rate": 6.443028485757122e-07,
      "loss": 1.1177,
      "step": 3439
    },
    {
      "epoch": 327.6190476190476,
      "grad_norm": 6.380118370056152,
      "learning_rate": 6.444902548725638e-07,
      "loss": 1.164,
      "step": 3440
    },
    {
      "epoch": 327.7142857142857,
      "grad_norm": 10.345011711120605,
      "learning_rate": 6.446776611694154e-07,
      "loss": 1.1433,
      "step": 3441
    },
    {
      "epoch": 327.8095238095238,
      "grad_norm": 4.250133991241455,
      "learning_rate": 6.448650674662669e-07,
      "loss": 1.1391,
      "step": 3442
    },
    {
      "epoch": 327.9047619047619,
      "grad_norm": 7.399500370025635,
      "learning_rate": 6.450524737631185e-07,
      "loss": 1.1499,
      "step": 3443
    },
    {
      "epoch": 328.0,
      "grad_norm": 5.87099027633667,
      "learning_rate": 6.4523988005997e-07,
      "loss": 1.1016,
      "step": 3444
    },
    {
      "epoch": 328.0,
      "eval_accuracy": 0.7343574372424129,
      "eval_f1": 0.7850863898150955,
      "eval_loss": 0.5564801692962646,
      "eval_precision": 0.7378917378917379,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.784690846286701,
      "eval_runtime": 1.1322,
      "eval_samples_per_second": 2357.292,
      "eval_steps_per_second": 5.299,
      "step": 3444
    },
    {
      "epoch": 328.0952380952381,
      "grad_norm": 5.727064609527588,
      "learning_rate": 6.454272863568216e-07,
      "loss": 1.1292,
      "step": 3445
    },
    {
      "epoch": 328.1904761904762,
      "grad_norm": 4.016811370849609,
      "learning_rate": 6.456146926536732e-07,
      "loss": 1.1085,
      "step": 3446
    },
    {
      "epoch": 328.2857142857143,
      "grad_norm": 6.0340352058410645,
      "learning_rate": 6.458020989505248e-07,
      "loss": 1.0875,
      "step": 3447
    },
    {
      "epoch": 328.3809523809524,
      "grad_norm": 4.349868297576904,
      "learning_rate": 6.459895052473764e-07,
      "loss": 1.1237,
      "step": 3448
    },
    {
      "epoch": 328.4761904761905,
      "grad_norm": 9.41535472869873,
      "learning_rate": 6.461769115442279e-07,
      "loss": 1.1675,
      "step": 3449
    },
    {
      "epoch": 328.57142857142856,
      "grad_norm": 10.782812118530273,
      "learning_rate": 6.463643178410795e-07,
      "loss": 1.1496,
      "step": 3450
    },
    {
      "epoch": 328.6666666666667,
      "grad_norm": 4.990645408630371,
      "learning_rate": 6.46551724137931e-07,
      "loss": 1.1289,
      "step": 3451
    },
    {
      "epoch": 328.76190476190476,
      "grad_norm": 5.7273850440979,
      "learning_rate": 6.467391304347826e-07,
      "loss": 1.1303,
      "step": 3452
    },
    {
      "epoch": 328.85714285714283,
      "grad_norm": 3.955021858215332,
      "learning_rate": 6.469265367316341e-07,
      "loss": 1.1059,
      "step": 3453
    },
    {
      "epoch": 328.95238095238096,
      "grad_norm": 6.905742645263672,
      "learning_rate": 6.471139430284858e-07,
      "loss": 1.1573,
      "step": 3454
    },
    {
      "epoch": 328.95238095238096,
      "eval_accuracy": 0.7351067815661296,
      "eval_f1": 0.7885133113969488,
      "eval_loss": 0.5563375949859619,
      "eval_precision": 0.7326292384658143,
      "eval_recall": 0.8536269430051814,
      "eval_roc_auc": 0.7854332181922854,
      "eval_runtime": 1.1084,
      "eval_samples_per_second": 2408.01,
      "eval_steps_per_second": 5.413,
      "step": 3454
    },
    {
      "epoch": 329.04761904761904,
      "grad_norm": 14.890825271606445,
      "learning_rate": 6.473013493253374e-07,
      "loss": 1.1772,
      "step": 3455
    },
    {
      "epoch": 329.14285714285717,
      "grad_norm": 11.016300201416016,
      "learning_rate": 6.474887556221889e-07,
      "loss": 1.0925,
      "step": 3456
    },
    {
      "epoch": 329.23809523809524,
      "grad_norm": 4.1203227043151855,
      "learning_rate": 6.476761619190405e-07,
      "loss": 1.093,
      "step": 3457
    },
    {
      "epoch": 329.3333333333333,
      "grad_norm": 5.667579174041748,
      "learning_rate": 6.47863568215892e-07,
      "loss": 1.1552,
      "step": 3458
    },
    {
      "epoch": 329.42857142857144,
      "grad_norm": 9.598355293273926,
      "learning_rate": 6.480509745127437e-07,
      "loss": 1.1022,
      "step": 3459
    },
    {
      "epoch": 329.5238095238095,
      "grad_norm": 15.727906227111816,
      "learning_rate": 6.482383808095952e-07,
      "loss": 1.1737,
      "step": 3460
    },
    {
      "epoch": 329.6190476190476,
      "grad_norm": 5.195409774780273,
      "learning_rate": 6.484257871064468e-07,
      "loss": 1.1394,
      "step": 3461
    },
    {
      "epoch": 329.7142857142857,
      "grad_norm": 8.702081680297852,
      "learning_rate": 6.486131934032985e-07,
      "loss": 1.1333,
      "step": 3462
    },
    {
      "epoch": 329.8095238095238,
      "grad_norm": 5.054378986358643,
      "learning_rate": 6.4880059970015e-07,
      "loss": 1.1154,
      "step": 3463
    },
    {
      "epoch": 329.9047619047619,
      "grad_norm": 5.856356143951416,
      "learning_rate": 6.489880059970016e-07,
      "loss": 1.1472,
      "step": 3464
    },
    {
      "epoch": 330.0,
      "grad_norm": 6.509719371795654,
      "learning_rate": 6.491754122938531e-07,
      "loss": 1.1341,
      "step": 3465
    },
    {
      "epoch": 330.0,
      "eval_accuracy": 0.7336080929186961,
      "eval_f1": 0.784480145498636,
      "eval_loss": 0.5562961101531982,
      "eval_precision": 0.7373219373219373,
      "eval_recall": 0.8380829015544041,
      "eval_roc_auc": 0.7849631548647092,
      "eval_runtime": 1.1231,
      "eval_samples_per_second": 2376.361,
      "eval_steps_per_second": 5.342,
      "step": 3465
    },
    {
      "epoch": 330.0952380952381,
      "grad_norm": 8.819149017333984,
      "learning_rate": 6.493628185907047e-07,
      "loss": 1.1193,
      "step": 3466
    },
    {
      "epoch": 330.1904761904762,
      "grad_norm": 5.841251373291016,
      "learning_rate": 6.495502248875562e-07,
      "loss": 1.1278,
      "step": 3467
    },
    {
      "epoch": 330.2857142857143,
      "grad_norm": 7.35856294631958,
      "learning_rate": 6.497376311844079e-07,
      "loss": 1.1653,
      "step": 3468
    },
    {
      "epoch": 330.3809523809524,
      "grad_norm": 10.730827331542969,
      "learning_rate": 6.499250374812594e-07,
      "loss": 1.098,
      "step": 3469
    },
    {
      "epoch": 330.4761904761905,
      "grad_norm": 13.916901588439941,
      "learning_rate": 6.50112443778111e-07,
      "loss": 1.1301,
      "step": 3470
    },
    {
      "epoch": 330.57142857142856,
      "grad_norm": 12.171894073486328,
      "learning_rate": 6.502998500749626e-07,
      "loss": 1.1151,
      "step": 3471
    },
    {
      "epoch": 330.6666666666667,
      "grad_norm": 4.601052284240723,
      "learning_rate": 6.504872563718141e-07,
      "loss": 1.1187,
      "step": 3472
    },
    {
      "epoch": 330.76190476190476,
      "grad_norm": 6.772599220275879,
      "learning_rate": 6.506746626686657e-07,
      "loss": 1.1054,
      "step": 3473
    },
    {
      "epoch": 330.85714285714283,
      "grad_norm": 15.050695419311523,
      "learning_rate": 6.508620689655172e-07,
      "loss": 1.12,
      "step": 3474
    },
    {
      "epoch": 330.95238095238096,
      "grad_norm": 4.695240020751953,
      "learning_rate": 6.510494752623689e-07,
      "loss": 1.1394,
      "step": 3475
    },
    {
      "epoch": 330.95238095238096,
      "eval_accuracy": 0.735481453727988,
      "eval_f1": 0.7876052948255114,
      "eval_loss": 0.5557202100753784,
      "eval_precision": 0.7353932584269663,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7858235463442717,
      "eval_runtime": 1.1291,
      "eval_samples_per_second": 2363.921,
      "eval_steps_per_second": 5.314,
      "step": 3475
    },
    {
      "epoch": 331.04761904761904,
      "grad_norm": 8.036359786987305,
      "learning_rate": 6.512368815592204e-07,
      "loss": 1.1233,
      "step": 3476
    },
    {
      "epoch": 331.14285714285717,
      "grad_norm": 7.4813995361328125,
      "learning_rate": 6.51424287856072e-07,
      "loss": 1.1522,
      "step": 3477
    },
    {
      "epoch": 331.23809523809524,
      "grad_norm": 6.448642253875732,
      "learning_rate": 6.516116941529235e-07,
      "loss": 1.1023,
      "step": 3478
    },
    {
      "epoch": 331.3333333333333,
      "grad_norm": 5.55362606048584,
      "learning_rate": 6.517991004497751e-07,
      "loss": 1.1096,
      "step": 3479
    },
    {
      "epoch": 331.42857142857144,
      "grad_norm": 4.546058177947998,
      "learning_rate": 6.519865067466266e-07,
      "loss": 1.137,
      "step": 3480
    },
    {
      "epoch": 331.5238095238095,
      "grad_norm": 4.071013450622559,
      "learning_rate": 6.521739130434782e-07,
      "loss": 1.147,
      "step": 3481
    },
    {
      "epoch": 331.6190476190476,
      "grad_norm": 5.781118392944336,
      "learning_rate": 6.523613193403299e-07,
      "loss": 1.1107,
      "step": 3482
    },
    {
      "epoch": 331.7142857142857,
      "grad_norm": 5.479564189910889,
      "learning_rate": 6.525487256371814e-07,
      "loss": 1.141,
      "step": 3483
    },
    {
      "epoch": 331.8095238095238,
      "grad_norm": 4.575967311859131,
      "learning_rate": 6.527361319340331e-07,
      "loss": 1.1133,
      "step": 3484
    },
    {
      "epoch": 331.9047619047619,
      "grad_norm": 6.548676490783691,
      "learning_rate": 6.529235382308846e-07,
      "loss": 1.0983,
      "step": 3485
    },
    {
      "epoch": 332.0,
      "grad_norm": 15.194136619567871,
      "learning_rate": 6.531109445277362e-07,
      "loss": 1.1231,
      "step": 3486
    },
    {
      "epoch": 332.0,
      "eval_accuracy": 0.7366054702135632,
      "eval_f1": 0.7865168539325843,
      "eval_loss": 0.5558101534843445,
      "eval_precision": 0.740423098913665,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.7858981001727117,
      "eval_runtime": 1.1449,
      "eval_samples_per_second": 2331.269,
      "eval_steps_per_second": 5.241,
      "step": 3486
    },
    {
      "epoch": 332.0952380952381,
      "grad_norm": 5.636200904846191,
      "learning_rate": 6.532983508245877e-07,
      "loss": 1.1396,
      "step": 3487
    },
    {
      "epoch": 332.1904761904762,
      "grad_norm": 13.208602905273438,
      "learning_rate": 6.534857571214393e-07,
      "loss": 1.1486,
      "step": 3488
    },
    {
      "epoch": 332.2857142857143,
      "grad_norm": 8.23987865447998,
      "learning_rate": 6.53673163418291e-07,
      "loss": 1.0966,
      "step": 3489
    },
    {
      "epoch": 332.3809523809524,
      "grad_norm": 6.295897006988525,
      "learning_rate": 6.538605697151425e-07,
      "loss": 1.1478,
      "step": 3490
    },
    {
      "epoch": 332.4761904761905,
      "grad_norm": 4.879843711853027,
      "learning_rate": 6.540479760119941e-07,
      "loss": 1.0825,
      "step": 3491
    },
    {
      "epoch": 332.57142857142856,
      "grad_norm": 2.976145029067993,
      "learning_rate": 6.542353823088456e-07,
      "loss": 1.1143,
      "step": 3492
    },
    {
      "epoch": 332.6666666666667,
      "grad_norm": 15.933645248413086,
      "learning_rate": 6.544227886056972e-07,
      "loss": 1.1543,
      "step": 3493
    },
    {
      "epoch": 332.76190476190476,
      "grad_norm": 4.494801998138428,
      "learning_rate": 6.546101949025487e-07,
      "loss": 1.089,
      "step": 3494
    },
    {
      "epoch": 332.85714285714283,
      "grad_norm": 5.822324275970459,
      "learning_rate": 6.547976011994003e-07,
      "loss": 1.1265,
      "step": 3495
    },
    {
      "epoch": 332.95238095238096,
      "grad_norm": 8.572771072387695,
      "learning_rate": 6.549850074962519e-07,
      "loss": 1.125,
      "step": 3496
    },
    {
      "epoch": 332.95238095238096,
      "eval_accuracy": 0.7373548145372799,
      "eval_f1": 0.7905587092919032,
      "eval_loss": 0.5562298893928528,
      "eval_precision": 0.7337770382695508,
      "eval_recall": 0.8568652849740933,
      "eval_roc_auc": 0.7858097294185377,
      "eval_runtime": 1.1423,
      "eval_samples_per_second": 2336.425,
      "eval_steps_per_second": 5.252,
      "step": 3496
    },
    {
      "epoch": 333.04761904761904,
      "grad_norm": 11.127102851867676,
      "learning_rate": 6.551724137931035e-07,
      "loss": 1.2003,
      "step": 3497
    },
    {
      "epoch": 333.14285714285717,
      "grad_norm": 5.642254829406738,
      "learning_rate": 6.553598200899551e-07,
      "loss": 1.0894,
      "step": 3498
    },
    {
      "epoch": 333.23809523809524,
      "grad_norm": 6.612725257873535,
      "learning_rate": 6.555472263868066e-07,
      "loss": 1.1221,
      "step": 3499
    },
    {
      "epoch": 333.3333333333333,
      "grad_norm": 17.96694564819336,
      "learning_rate": 6.557346326836582e-07,
      "loss": 1.1726,
      "step": 3500
    },
    {
      "epoch": 333.42857142857144,
      "grad_norm": 3.2642133235931396,
      "learning_rate": 6.559220389805097e-07,
      "loss": 1.1022,
      "step": 3501
    },
    {
      "epoch": 333.5238095238095,
      "grad_norm": 6.946366786956787,
      "learning_rate": 6.561094452773613e-07,
      "loss": 1.1372,
      "step": 3502
    },
    {
      "epoch": 333.6190476190476,
      "grad_norm": 6.836911201477051,
      "learning_rate": 6.562968515742129e-07,
      "loss": 1.0981,
      "step": 3503
    },
    {
      "epoch": 333.7142857142857,
      "grad_norm": 6.185560703277588,
      "learning_rate": 6.564842578710645e-07,
      "loss": 1.1081,
      "step": 3504
    },
    {
      "epoch": 333.8095238095238,
      "grad_norm": 11.273193359375,
      "learning_rate": 6.56671664167916e-07,
      "loss": 1.121,
      "step": 3505
    },
    {
      "epoch": 333.9047619047619,
      "grad_norm": 5.383264064788818,
      "learning_rate": 6.568590704647676e-07,
      "loss": 1.1328,
      "step": 3506
    },
    {
      "epoch": 334.0,
      "grad_norm": 8.786903381347656,
      "learning_rate": 6.570464767616193e-07,
      "loss": 1.1291,
      "step": 3507
    },
    {
      "epoch": 334.0,
      "eval_accuracy": 0.7347321094042712,
      "eval_f1": 0.7858439201451906,
      "eval_loss": 0.5563549399375916,
      "eval_precision": 0.7372304199772985,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.784830742659758,
      "eval_runtime": 1.1532,
      "eval_samples_per_second": 2314.338,
      "eval_steps_per_second": 5.203,
      "step": 3507
    },
    {
      "epoch": 334.0952380952381,
      "grad_norm": 8.342055320739746,
      "learning_rate": 6.572338830584708e-07,
      "loss": 1.1618,
      "step": 3508
    },
    {
      "epoch": 334.1904761904762,
      "grad_norm": 5.02683162689209,
      "learning_rate": 6.574212893553224e-07,
      "loss": 1.1333,
      "step": 3509
    },
    {
      "epoch": 334.2857142857143,
      "grad_norm": 12.340161323547363,
      "learning_rate": 6.57608695652174e-07,
      "loss": 1.1035,
      "step": 3510
    },
    {
      "epoch": 334.3809523809524,
      "grad_norm": 4.038184642791748,
      "learning_rate": 6.577961019490256e-07,
      "loss": 1.1365,
      "step": 3511
    },
    {
      "epoch": 334.4761904761905,
      "grad_norm": 5.561829566955566,
      "learning_rate": 6.579835082458771e-07,
      "loss": 1.0896,
      "step": 3512
    },
    {
      "epoch": 334.57142857142856,
      "grad_norm": 6.000752925872803,
      "learning_rate": 6.581709145427287e-07,
      "loss": 1.108,
      "step": 3513
    },
    {
      "epoch": 334.6666666666667,
      "grad_norm": 7.689496994018555,
      "learning_rate": 6.583583208395802e-07,
      "loss": 1.1406,
      "step": 3514
    },
    {
      "epoch": 334.76190476190476,
      "grad_norm": 8.802385330200195,
      "learning_rate": 6.585457271364318e-07,
      "loss": 1.1199,
      "step": 3515
    },
    {
      "epoch": 334.85714285714283,
      "grad_norm": 5.754329681396484,
      "learning_rate": 6.587331334332834e-07,
      "loss": 1.1222,
      "step": 3516
    },
    {
      "epoch": 334.95238095238096,
      "grad_norm": 5.224997043609619,
      "learning_rate": 6.58920539730135e-07,
      "loss": 1.0855,
      "step": 3517
    },
    {
      "epoch": 334.95238095238096,
      "eval_accuracy": 0.7388535031847133,
      "eval_f1": 0.792621243677477,
      "eval_loss": 0.5562059283256531,
      "eval_precision": 0.7330764997248211,
      "eval_recall": 0.8626943005181347,
      "eval_roc_auc": 0.7856744386873922,
      "eval_runtime": 1.1406,
      "eval_samples_per_second": 2339.928,
      "eval_steps_per_second": 5.26,
      "step": 3517
    },
    {
      "epoch": 335.04761904761904,
      "grad_norm": 9.480507850646973,
      "learning_rate": 6.591079460269866e-07,
      "loss": 1.146,
      "step": 3518
    },
    {
      "epoch": 335.14285714285717,
      "grad_norm": 9.278295516967773,
      "learning_rate": 6.592953523238381e-07,
      "loss": 1.1548,
      "step": 3519
    },
    {
      "epoch": 335.23809523809524,
      "grad_norm": 3.4852707386016846,
      "learning_rate": 6.594827586206897e-07,
      "loss": 1.1137,
      "step": 3520
    },
    {
      "epoch": 335.3333333333333,
      "grad_norm": 16.312044143676758,
      "learning_rate": 6.596701649175412e-07,
      "loss": 1.0929,
      "step": 3521
    },
    {
      "epoch": 335.42857142857144,
      "grad_norm": 10.9599027633667,
      "learning_rate": 6.598575712143928e-07,
      "loss": 1.1378,
      "step": 3522
    },
    {
      "epoch": 335.5238095238095,
      "grad_norm": 4.634957313537598,
      "learning_rate": 6.600449775112443e-07,
      "loss": 1.1423,
      "step": 3523
    },
    {
      "epoch": 335.6190476190476,
      "grad_norm": 5.783304214477539,
      "learning_rate": 6.60232383808096e-07,
      "loss": 1.103,
      "step": 3524
    },
    {
      "epoch": 335.7142857142857,
      "grad_norm": 14.14034652709961,
      "learning_rate": 6.604197901049476e-07,
      "loss": 1.1274,
      "step": 3525
    },
    {
      "epoch": 335.8095238095238,
      "grad_norm": 7.4911789894104,
      "learning_rate": 6.606071964017991e-07,
      "loss": 1.1116,
      "step": 3526
    },
    {
      "epoch": 335.9047619047619,
      "grad_norm": 7.270888328552246,
      "learning_rate": 6.607946026986507e-07,
      "loss": 1.1475,
      "step": 3527
    },
    {
      "epoch": 336.0,
      "grad_norm": 10.573918342590332,
      "learning_rate": 6.609820089955022e-07,
      "loss": 1.1337,
      "step": 3528
    },
    {
      "epoch": 336.0,
      "eval_accuracy": 0.7309853877856876,
      "eval_f1": 0.780562347188264,
      "eval_loss": 0.5544623732566833,
      "eval_precision": 0.7390046296296297,
      "eval_recall": 0.8270725388601037,
      "eval_roc_auc": 0.7872259643062751,
      "eval_runtime": 1.1055,
      "eval_samples_per_second": 2414.265,
      "eval_steps_per_second": 5.427,
      "step": 3528
    },
    {
      "epoch": 336.0952380952381,
      "grad_norm": 14.302084922790527,
      "learning_rate": 6.611694152923539e-07,
      "loss": 1.1146,
      "step": 3529
    },
    {
      "epoch": 336.1904761904762,
      "grad_norm": 7.01728630065918,
      "learning_rate": 6.613568215892054e-07,
      "loss": 1.1553,
      "step": 3530
    },
    {
      "epoch": 336.2857142857143,
      "grad_norm": 8.606803894042969,
      "learning_rate": 6.61544227886057e-07,
      "loss": 1.0986,
      "step": 3531
    },
    {
      "epoch": 336.3809523809524,
      "grad_norm": 5.691741466522217,
      "learning_rate": 6.617316341829086e-07,
      "loss": 1.1194,
      "step": 3532
    },
    {
      "epoch": 336.4761904761905,
      "grad_norm": 5.615129470825195,
      "learning_rate": 6.619190404797602e-07,
      "loss": 1.0961,
      "step": 3533
    },
    {
      "epoch": 336.57142857142856,
      "grad_norm": 7.926955223083496,
      "learning_rate": 6.621064467766118e-07,
      "loss": 1.1396,
      "step": 3534
    },
    {
      "epoch": 336.6666666666667,
      "grad_norm": 3.7907376289367676,
      "learning_rate": 6.622938530734633e-07,
      "loss": 1.0845,
      "step": 3535
    },
    {
      "epoch": 336.76190476190476,
      "grad_norm": 9.664169311523438,
      "learning_rate": 6.624812593703149e-07,
      "loss": 1.2008,
      "step": 3536
    },
    {
      "epoch": 336.85714285714283,
      "grad_norm": 5.354334831237793,
      "learning_rate": 6.626686656671664e-07,
      "loss": 1.1075,
      "step": 3537
    },
    {
      "epoch": 336.95238095238096,
      "grad_norm": 11.554585456848145,
      "learning_rate": 6.628560719640181e-07,
      "loss": 1.1256,
      "step": 3538
    },
    {
      "epoch": 336.95238095238096,
      "eval_accuracy": 0.7418508804795804,
      "eval_f1": 0.7940209267563527,
      "eval_loss": 0.5547006726264954,
      "eval_precision": 0.7373681288173237,
      "eval_recall": 0.8601036269430051,
      "eval_roc_auc": 0.7875762809441567,
      "eval_runtime": 1.1087,
      "eval_samples_per_second": 2407.42,
      "eval_steps_per_second": 5.412,
      "step": 3538
    },
    {
      "epoch": 337.04761904761904,
      "grad_norm": 3.42289137840271,
      "learning_rate": 6.630434782608696e-07,
      "loss": 1.0927,
      "step": 3539
    },
    {
      "epoch": 337.14285714285717,
      "grad_norm": 12.857282638549805,
      "learning_rate": 6.632308845577212e-07,
      "loss": 1.128,
      "step": 3540
    },
    {
      "epoch": 337.23809523809524,
      "grad_norm": 16.507972717285156,
      "learning_rate": 6.634182908545727e-07,
      "loss": 1.1832,
      "step": 3541
    },
    {
      "epoch": 337.3333333333333,
      "grad_norm": 5.046204090118408,
      "learning_rate": 6.636056971514243e-07,
      "loss": 1.0964,
      "step": 3542
    },
    {
      "epoch": 337.42857142857144,
      "grad_norm": 7.036007404327393,
      "learning_rate": 6.637931034482759e-07,
      "loss": 1.1487,
      "step": 3543
    },
    {
      "epoch": 337.5238095238095,
      "grad_norm": 7.588108539581299,
      "learning_rate": 6.639805097451274e-07,
      "loss": 1.1212,
      "step": 3544
    },
    {
      "epoch": 337.6190476190476,
      "grad_norm": 12.190917015075684,
      "learning_rate": 6.641679160419791e-07,
      "loss": 1.103,
      "step": 3545
    },
    {
      "epoch": 337.7142857142857,
      "grad_norm": 6.134925365447998,
      "learning_rate": 6.643553223388306e-07,
      "loss": 1.0935,
      "step": 3546
    },
    {
      "epoch": 337.8095238095238,
      "grad_norm": 24.103914260864258,
      "learning_rate": 6.645427286356822e-07,
      "loss": 1.1485,
      "step": 3547
    },
    {
      "epoch": 337.9047619047619,
      "grad_norm": 12.370758056640625,
      "learning_rate": 6.647301349325337e-07,
      "loss": 1.1255,
      "step": 3548
    },
    {
      "epoch": 338.0,
      "grad_norm": 9.194156646728516,
      "learning_rate": 6.649175412293853e-07,
      "loss": 1.1161,
      "step": 3549
    },
    {
      "epoch": 338.0,
      "eval_accuracy": 0.7351067815661296,
      "eval_f1": 0.7858224780369585,
      "eval_loss": 0.5546090006828308,
      "eval_precision": 0.7381900967558338,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7870918249856074,
      "eval_runtime": 1.1211,
      "eval_samples_per_second": 2380.612,
      "eval_steps_per_second": 5.352,
      "step": 3549
    },
    {
      "epoch": 338.0952380952381,
      "grad_norm": 12.278610229492188,
      "learning_rate": 6.651049475262368e-07,
      "loss": 1.131,
      "step": 3550
    },
    {
      "epoch": 338.1904761904762,
      "grad_norm": 5.793935775756836,
      "learning_rate": 6.652923538230885e-07,
      "loss": 1.1346,
      "step": 3551
    },
    {
      "epoch": 338.2857142857143,
      "grad_norm": 7.647890567779541,
      "learning_rate": 6.654797601199401e-07,
      "loss": 1.1406,
      "step": 3552
    },
    {
      "epoch": 338.3809523809524,
      "grad_norm": 15.229239463806152,
      "learning_rate": 6.656671664167916e-07,
      "loss": 1.1486,
      "step": 3553
    },
    {
      "epoch": 338.4761904761905,
      "grad_norm": 8.464210510253906,
      "learning_rate": 6.658545727136433e-07,
      "loss": 1.1245,
      "step": 3554
    },
    {
      "epoch": 338.57142857142856,
      "grad_norm": 10.347994804382324,
      "learning_rate": 6.660419790104948e-07,
      "loss": 1.1417,
      "step": 3555
    },
    {
      "epoch": 338.6666666666667,
      "grad_norm": 5.064682483673096,
      "learning_rate": 6.662293853073464e-07,
      "loss": 1.0578,
      "step": 3556
    },
    {
      "epoch": 338.76190476190476,
      "grad_norm": 8.694058418273926,
      "learning_rate": 6.664167916041979e-07,
      "loss": 1.1247,
      "step": 3557
    },
    {
      "epoch": 338.85714285714283,
      "grad_norm": 10.907423973083496,
      "learning_rate": 6.666041979010495e-07,
      "loss": 1.1499,
      "step": 3558
    },
    {
      "epoch": 338.95238095238096,
      "grad_norm": 6.98323917388916,
      "learning_rate": 6.667916041979011e-07,
      "loss": 1.0894,
      "step": 3559
    },
    {
      "epoch": 338.95238095238096,
      "eval_accuracy": 0.7369801423754215,
      "eval_f1": 0.7874015748031497,
      "eval_loss": 0.5539066791534424,
      "eval_precision": 0.7394766780432309,
      "eval_recall": 0.8419689119170984,
      "eval_roc_auc": 0.7878603914795625,
      "eval_runtime": 1.1949,
      "eval_samples_per_second": 2233.618,
      "eval_steps_per_second": 5.021,
      "step": 3559
    },
    {
      "epoch": 339.04761904761904,
      "grad_norm": 6.979728698730469,
      "learning_rate": 6.669790104947527e-07,
      "loss": 1.1115,
      "step": 3560
    },
    {
      "epoch": 339.14285714285717,
      "grad_norm": 4.081621170043945,
      "learning_rate": 6.671664167916043e-07,
      "loss": 1.1271,
      "step": 3561
    },
    {
      "epoch": 339.23809523809524,
      "grad_norm": 13.691429138183594,
      "learning_rate": 6.673538230884558e-07,
      "loss": 1.1255,
      "step": 3562
    },
    {
      "epoch": 339.3333333333333,
      "grad_norm": 11.593656539916992,
      "learning_rate": 6.675412293853074e-07,
      "loss": 1.1027,
      "step": 3563
    },
    {
      "epoch": 339.42857142857144,
      "grad_norm": 6.386806488037109,
      "learning_rate": 6.677286356821589e-07,
      "loss": 1.0918,
      "step": 3564
    },
    {
      "epoch": 339.5238095238095,
      "grad_norm": 3.813530445098877,
      "learning_rate": 6.679160419790105e-07,
      "loss": 1.1615,
      "step": 3565
    },
    {
      "epoch": 339.6190476190476,
      "grad_norm": 5.865270137786865,
      "learning_rate": 6.681034482758621e-07,
      "loss": 1.1287,
      "step": 3566
    },
    {
      "epoch": 339.7142857142857,
      "grad_norm": 5.257119178771973,
      "learning_rate": 6.682908545727137e-07,
      "loss": 1.1735,
      "step": 3567
    },
    {
      "epoch": 339.8095238095238,
      "grad_norm": 13.466306686401367,
      "learning_rate": 6.684782608695652e-07,
      "loss": 1.1191,
      "step": 3568
    },
    {
      "epoch": 339.9047619047619,
      "grad_norm": 20.444292068481445,
      "learning_rate": 6.686656671664168e-07,
      "loss": 1.1079,
      "step": 3569
    },
    {
      "epoch": 340.0,
      "grad_norm": 6.399350643157959,
      "learning_rate": 6.688530734632684e-07,
      "loss": 1.1475,
      "step": 3570
    },
    {
      "epoch": 340.0,
      "eval_accuracy": 0.7321094042712627,
      "eval_f1": 0.781010719754977,
      "eval_loss": 0.5544794201850891,
      "eval_precision": 0.7408483439860546,
      "eval_recall": 0.8257772020725389,
      "eval_roc_auc": 0.7868966609096142,
      "eval_runtime": 1.1362,
      "eval_samples_per_second": 2349.049,
      "eval_steps_per_second": 5.281,
      "step": 3570
    },
    {
      "epoch": 340.0952380952381,
      "grad_norm": 17.85746192932129,
      "learning_rate": 6.690404797601199e-07,
      "loss": 1.0988,
      "step": 3571
    },
    {
      "epoch": 340.1904761904762,
      "grad_norm": 12.25684928894043,
      "learning_rate": 6.692278860569715e-07,
      "loss": 1.1331,
      "step": 3572
    },
    {
      "epoch": 340.2857142857143,
      "grad_norm": 8.942669868469238,
      "learning_rate": 6.694152923538231e-07,
      "loss": 1.1431,
      "step": 3573
    },
    {
      "epoch": 340.3809523809524,
      "grad_norm": 6.340849876403809,
      "learning_rate": 6.696026986506747e-07,
      "loss": 1.0891,
      "step": 3574
    },
    {
      "epoch": 340.4761904761905,
      "grad_norm": 5.250206470489502,
      "learning_rate": 6.697901049475262e-07,
      "loss": 1.1616,
      "step": 3575
    },
    {
      "epoch": 340.57142857142856,
      "grad_norm": 6.328611850738525,
      "learning_rate": 6.699775112443779e-07,
      "loss": 1.1095,
      "step": 3576
    },
    {
      "epoch": 340.6666666666667,
      "grad_norm": 9.236186981201172,
      "learning_rate": 6.701649175412294e-07,
      "loss": 1.0791,
      "step": 3577
    },
    {
      "epoch": 340.76190476190476,
      "grad_norm": 11.165750503540039,
      "learning_rate": 6.70352323838081e-07,
      "loss": 1.1567,
      "step": 3578
    },
    {
      "epoch": 340.85714285714283,
      "grad_norm": 8.74531364440918,
      "learning_rate": 6.705397301349326e-07,
      "loss": 1.108,
      "step": 3579
    },
    {
      "epoch": 340.95238095238096,
      "grad_norm": 8.133259773254395,
      "learning_rate": 6.707271364317842e-07,
      "loss": 1.1202,
      "step": 3580
    },
    {
      "epoch": 340.95238095238096,
      "eval_accuracy": 0.7339827650805545,
      "eval_f1": 0.786273329319687,
      "eval_loss": 0.5549090504646301,
      "eval_precision": 0.734533183352081,
      "eval_recall": 0.8458549222797928,
      "eval_roc_auc": 0.7867956246401843,
      "eval_runtime": 1.1413,
      "eval_samples_per_second": 2338.49,
      "eval_steps_per_second": 5.257,
      "step": 3580
    },
    {
      "epoch": 341.04761904761904,
      "grad_norm": 12.086969375610352,
      "learning_rate": 6.709145427286358e-07,
      "loss": 1.1356,
      "step": 3581
    },
    {
      "epoch": 341.14285714285717,
      "grad_norm": 9.834478378295898,
      "learning_rate": 6.711019490254873e-07,
      "loss": 1.1098,
      "step": 3582
    },
    {
      "epoch": 341.23809523809524,
      "grad_norm": 8.839277267456055,
      "learning_rate": 6.712893553223389e-07,
      "loss": 1.1024,
      "step": 3583
    },
    {
      "epoch": 341.3333333333333,
      "grad_norm": 12.858485221862793,
      "learning_rate": 6.714767616191904e-07,
      "loss": 1.0935,
      "step": 3584
    },
    {
      "epoch": 341.42857142857144,
      "grad_norm": 21.6302490234375,
      "learning_rate": 6.71664167916042e-07,
      "loss": 1.1222,
      "step": 3585
    },
    {
      "epoch": 341.5238095238095,
      "grad_norm": 23.848730087280273,
      "learning_rate": 6.718515742128935e-07,
      "loss": 1.1004,
      "step": 3586
    },
    {
      "epoch": 341.6190476190476,
      "grad_norm": 13.283079147338867,
      "learning_rate": 6.720389805097452e-07,
      "loss": 1.1245,
      "step": 3587
    },
    {
      "epoch": 341.7142857142857,
      "grad_norm": 15.620110511779785,
      "learning_rate": 6.722263868065968e-07,
      "loss": 1.1212,
      "step": 3588
    },
    {
      "epoch": 341.8095238095238,
      "grad_norm": 4.007311820983887,
      "learning_rate": 6.724137931034483e-07,
      "loss": 1.1378,
      "step": 3589
    },
    {
      "epoch": 341.9047619047619,
      "grad_norm": 6.650176525115967,
      "learning_rate": 6.726011994002999e-07,
      "loss": 1.15,
      "step": 3590
    },
    {
      "epoch": 342.0,
      "grad_norm": 12.86173152923584,
      "learning_rate": 6.727886056971514e-07,
      "loss": 1.1555,
      "step": 3591
    },
    {
      "epoch": 342.0,
      "eval_accuracy": 0.738478831022855,
      "eval_f1": 0.7922619047619047,
      "eval_loss": 0.5554071068763733,
      "eval_precision": 0.7329295154185022,
      "eval_recall": 0.8620466321243523,
      "eval_roc_auc": 0.7867576280944157,
      "eval_runtime": 1.1368,
      "eval_samples_per_second": 2347.863,
      "eval_steps_per_second": 5.278,
      "step": 3591
    },
    {
      "epoch": 342.0952380952381,
      "grad_norm": 17.999942779541016,
      "learning_rate": 6.72976011994003e-07,
      "loss": 1.1372,
      "step": 3592
    },
    {
      "epoch": 342.1904761904762,
      "grad_norm": 11.610297203063965,
      "learning_rate": 6.731634182908545e-07,
      "loss": 1.0924,
      "step": 3593
    },
    {
      "epoch": 342.2857142857143,
      "grad_norm": 5.744530200958252,
      "learning_rate": 6.733508245877062e-07,
      "loss": 1.1296,
      "step": 3594
    },
    {
      "epoch": 342.3809523809524,
      "grad_norm": 4.968858242034912,
      "learning_rate": 6.735382308845577e-07,
      "loss": 1.1053,
      "step": 3595
    },
    {
      "epoch": 342.4761904761905,
      "grad_norm": 5.93127965927124,
      "learning_rate": 6.737256371814093e-07,
      "loss": 1.118,
      "step": 3596
    },
    {
      "epoch": 342.57142857142856,
      "grad_norm": 5.766799449920654,
      "learning_rate": 6.739130434782609e-07,
      "loss": 1.1342,
      "step": 3597
    },
    {
      "epoch": 342.6666666666667,
      "grad_norm": 9.84838581085205,
      "learning_rate": 6.741004497751124e-07,
      "loss": 1.1653,
      "step": 3598
    },
    {
      "epoch": 342.76190476190476,
      "grad_norm": 10.662291526794434,
      "learning_rate": 6.742878560719641e-07,
      "loss": 1.1093,
      "step": 3599
    },
    {
      "epoch": 342.85714285714283,
      "grad_norm": 4.422891139984131,
      "learning_rate": 6.744752623688156e-07,
      "loss": 1.1379,
      "step": 3600
    },
    {
      "epoch": 342.95238095238096,
      "grad_norm": 11.94595718383789,
      "learning_rate": 6.746626686656673e-07,
      "loss": 1.0888,
      "step": 3601
    },
    {
      "epoch": 342.95238095238096,
      "eval_accuracy": 0.732484076433121,
      "eval_f1": 0.7808471454880295,
      "eval_loss": 0.5546474456787109,
      "eval_precision": 0.7421236872812136,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.7872820955670696,
      "eval_runtime": 1.1459,
      "eval_samples_per_second": 2329.114,
      "eval_steps_per_second": 5.236,
      "step": 3601
    },
    {
      "epoch": 343.04761904761904,
      "grad_norm": 15.905829429626465,
      "learning_rate": 6.748500749625188e-07,
      "loss": 1.1039,
      "step": 3602
    },
    {
      "epoch": 343.14285714285717,
      "grad_norm": 16.905872344970703,
      "learning_rate": 6.750374812593704e-07,
      "loss": 1.0695,
      "step": 3603
    },
    {
      "epoch": 343.23809523809524,
      "grad_norm": 4.414461612701416,
      "learning_rate": 6.752248875562219e-07,
      "loss": 1.14,
      "step": 3604
    },
    {
      "epoch": 343.3333333333333,
      "grad_norm": 4.829150676727295,
      "learning_rate": 6.754122938530735e-07,
      "loss": 1.1487,
      "step": 3605
    },
    {
      "epoch": 343.42857142857144,
      "grad_norm": 7.950586318969727,
      "learning_rate": 6.755997001499251e-07,
      "loss": 1.0979,
      "step": 3606
    },
    {
      "epoch": 343.5238095238095,
      "grad_norm": 4.235756874084473,
      "learning_rate": 6.757871064467766e-07,
      "loss": 1.0673,
      "step": 3607
    },
    {
      "epoch": 343.6190476190476,
      "grad_norm": 9.007696151733398,
      "learning_rate": 6.759745127436283e-07,
      "loss": 1.1128,
      "step": 3608
    },
    {
      "epoch": 343.7142857142857,
      "grad_norm": 6.327154159545898,
      "learning_rate": 6.761619190404798e-07,
      "loss": 1.1134,
      "step": 3609
    },
    {
      "epoch": 343.8095238095238,
      "grad_norm": 8.22981071472168,
      "learning_rate": 6.763493253373314e-07,
      "loss": 1.1625,
      "step": 3610
    },
    {
      "epoch": 343.9047619047619,
      "grad_norm": 6.529686450958252,
      "learning_rate": 6.765367316341829e-07,
      "loss": 1.1357,
      "step": 3611
    },
    {
      "epoch": 344.0,
      "grad_norm": 9.798868179321289,
      "learning_rate": 6.767241379310345e-07,
      "loss": 1.1103,
      "step": 3612
    },
    {
      "epoch": 344.0,
      "eval_accuracy": 0.7347321094042712,
      "eval_f1": 0.7828220858895706,
      "eval_loss": 0.5543044805526733,
      "eval_precision": 0.7435897435897436,
      "eval_recall": 0.8264248704663213,
      "eval_roc_auc": 0.7875420264824409,
      "eval_runtime": 1.1258,
      "eval_samples_per_second": 2370.761,
      "eval_steps_per_second": 5.33,
      "step": 3612
    },
    {
      "epoch": 344.0952380952381,
      "grad_norm": 6.461381912231445,
      "learning_rate": 6.76911544227886e-07,
      "loss": 1.1304,
      "step": 3613
    },
    {
      "epoch": 344.1904761904762,
      "grad_norm": 6.234090328216553,
      "learning_rate": 6.770989505247376e-07,
      "loss": 1.1294,
      "step": 3614
    },
    {
      "epoch": 344.2857142857143,
      "grad_norm": 7.588049411773682,
      "learning_rate": 6.772863568215893e-07,
      "loss": 1.0847,
      "step": 3615
    },
    {
      "epoch": 344.3809523809524,
      "grad_norm": 18.067716598510742,
      "learning_rate": 6.774737631184408e-07,
      "loss": 1.1469,
      "step": 3616
    },
    {
      "epoch": 344.4761904761905,
      "grad_norm": 9.160544395446777,
      "learning_rate": 6.776611694152924e-07,
      "loss": 1.1314,
      "step": 3617
    },
    {
      "epoch": 344.57142857142856,
      "grad_norm": 12.811017990112305,
      "learning_rate": 6.778485757121439e-07,
      "loss": 1.1341,
      "step": 3618
    },
    {
      "epoch": 344.6666666666667,
      "grad_norm": 3.5876166820526123,
      "learning_rate": 6.780359820089955e-07,
      "loss": 1.1237,
      "step": 3619
    },
    {
      "epoch": 344.76190476190476,
      "grad_norm": 8.06754207611084,
      "learning_rate": 6.78223388305847e-07,
      "loss": 1.1162,
      "step": 3620
    },
    {
      "epoch": 344.85714285714283,
      "grad_norm": 10.948036193847656,
      "learning_rate": 6.784107946026987e-07,
      "loss": 1.1154,
      "step": 3621
    },
    {
      "epoch": 344.95238095238096,
      "grad_norm": 5.318874835968018,
      "learning_rate": 6.785982008995502e-07,
      "loss": 1.1006,
      "step": 3622
    },
    {
      "epoch": 344.95238095238096,
      "eval_accuracy": 0.7388535031847133,
      "eval_f1": 0.7920023873470606,
      "eval_loss": 0.554513156414032,
      "eval_precision": 0.7343663530713891,
      "eval_recall": 0.8594559585492227,
      "eval_roc_auc": 0.7877369027058146,
      "eval_runtime": 1.1197,
      "eval_samples_per_second": 2383.667,
      "eval_steps_per_second": 5.359,
      "step": 3622
    },
    {
      "epoch": 345.04761904761904,
      "grad_norm": 12.399185180664062,
      "learning_rate": 6.787856071964018e-07,
      "loss": 1.0919,
      "step": 3623
    },
    {
      "epoch": 345.14285714285717,
      "grad_norm": 20.321130752563477,
      "learning_rate": 6.789730134932535e-07,
      "loss": 1.1898,
      "step": 3624
    },
    {
      "epoch": 345.23809523809524,
      "grad_norm": 14.606483459472656,
      "learning_rate": 6.79160419790105e-07,
      "loss": 1.12,
      "step": 3625
    },
    {
      "epoch": 345.3333333333333,
      "grad_norm": 13.469183921813965,
      "learning_rate": 6.793478260869566e-07,
      "loss": 1.1608,
      "step": 3626
    },
    {
      "epoch": 345.42857142857144,
      "grad_norm": 8.02731990814209,
      "learning_rate": 6.795352323838081e-07,
      "loss": 1.1208,
      "step": 3627
    },
    {
      "epoch": 345.5238095238095,
      "grad_norm": 7.375977993011475,
      "learning_rate": 6.797226386806598e-07,
      "loss": 1.1089,
      "step": 3628
    },
    {
      "epoch": 345.6190476190476,
      "grad_norm": 9.051599502563477,
      "learning_rate": 6.799100449775113e-07,
      "loss": 1.0553,
      "step": 3629
    },
    {
      "epoch": 345.7142857142857,
      "grad_norm": 9.57223892211914,
      "learning_rate": 6.800974512743629e-07,
      "loss": 1.0911,
      "step": 3630
    },
    {
      "epoch": 345.8095238095238,
      "grad_norm": 6.9309916496276855,
      "learning_rate": 6.802848575712144e-07,
      "loss": 1.1275,
      "step": 3631
    },
    {
      "epoch": 345.9047619047619,
      "grad_norm": 14.309074401855469,
      "learning_rate": 6.80472263868066e-07,
      "loss": 1.1293,
      "step": 3632
    },
    {
      "epoch": 346.0,
      "grad_norm": 4.155423641204834,
      "learning_rate": 6.806596701649176e-07,
      "loss": 1.0834,
      "step": 3633
    },
    {
      "epoch": 346.0,
      "eval_accuracy": 0.735481453727988,
      "eval_f1": 0.7876052948255114,
      "eval_loss": 0.5538960099220276,
      "eval_precision": 0.7353932584269663,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7879467472654001,
      "eval_runtime": 1.1295,
      "eval_samples_per_second": 2362.961,
      "eval_steps_per_second": 5.312,
      "step": 3633
    },
    {
      "epoch": 346.0952380952381,
      "grad_norm": 11.9675931930542,
      "learning_rate": 6.808470764617691e-07,
      "loss": 1.1013,
      "step": 3634
    },
    {
      "epoch": 346.1904761904762,
      "grad_norm": 9.897497177124023,
      "learning_rate": 6.810344827586208e-07,
      "loss": 1.1574,
      "step": 3635
    },
    {
      "epoch": 346.2857142857143,
      "grad_norm": 13.217548370361328,
      "learning_rate": 6.812218890554723e-07,
      "loss": 1.1247,
      "step": 3636
    },
    {
      "epoch": 346.3809523809524,
      "grad_norm": 10.568078994750977,
      "learning_rate": 6.814092953523239e-07,
      "loss": 1.1355,
      "step": 3637
    },
    {
      "epoch": 346.4761904761905,
      "grad_norm": 9.215120315551758,
      "learning_rate": 6.815967016491754e-07,
      "loss": 1.1323,
      "step": 3638
    },
    {
      "epoch": 346.57142857142856,
      "grad_norm": 7.759387493133545,
      "learning_rate": 6.81784107946027e-07,
      "loss": 1.1929,
      "step": 3639
    },
    {
      "epoch": 346.6666666666667,
      "grad_norm": 6.659882068634033,
      "learning_rate": 6.819715142428785e-07,
      "loss": 1.086,
      "step": 3640
    },
    {
      "epoch": 346.76190476190476,
      "grad_norm": 19.305633544921875,
      "learning_rate": 6.821589205397301e-07,
      "loss": 1.078,
      "step": 3641
    },
    {
      "epoch": 346.85714285714283,
      "grad_norm": 6.204228401184082,
      "learning_rate": 6.823463268365818e-07,
      "loss": 1.1404,
      "step": 3642
    },
    {
      "epoch": 346.95238095238096,
      "grad_norm": 7.152632713317871,
      "learning_rate": 6.825337331334333e-07,
      "loss": 1.1183,
      "step": 3643
    },
    {
      "epoch": 346.95238095238096,
      "eval_accuracy": 0.7362307980517048,
      "eval_f1": 0.7858880778588808,
      "eval_loss": 0.5532867908477783,
      "eval_precision": 0.7408256880733946,
      "eval_recall": 0.8367875647668394,
      "eval_roc_auc": 0.7884700633275763,
      "eval_runtime": 1.1328,
      "eval_samples_per_second": 2356.206,
      "eval_steps_per_second": 5.297,
      "step": 3643
    },
    {
      "epoch": 347.04761904761904,
      "grad_norm": 22.47641372680664,
      "learning_rate": 6.827211394302849e-07,
      "loss": 1.0886,
      "step": 3644
    },
    {
      "epoch": 347.14285714285717,
      "grad_norm": 20.005538940429688,
      "learning_rate": 6.829085457271364e-07,
      "loss": 1.1646,
      "step": 3645
    },
    {
      "epoch": 347.23809523809524,
      "grad_norm": 8.337130546569824,
      "learning_rate": 6.830959520239881e-07,
      "loss": 1.0712,
      "step": 3646
    },
    {
      "epoch": 347.3333333333333,
      "grad_norm": 15.485507011413574,
      "learning_rate": 6.832833583208396e-07,
      "loss": 1.1199,
      "step": 3647
    },
    {
      "epoch": 347.42857142857144,
      "grad_norm": 17.480194091796875,
      "learning_rate": 6.834707646176912e-07,
      "loss": 1.1515,
      "step": 3648
    },
    {
      "epoch": 347.5238095238095,
      "grad_norm": 9.325078964233398,
      "learning_rate": 6.836581709145429e-07,
      "loss": 1.1247,
      "step": 3649
    },
    {
      "epoch": 347.6190476190476,
      "grad_norm": 7.568593502044678,
      "learning_rate": 6.838455772113944e-07,
      "loss": 1.1163,
      "step": 3650
    },
    {
      "epoch": 347.7142857142857,
      "grad_norm": 8.339293479919434,
      "learning_rate": 6.84032983508246e-07,
      "loss": 1.1626,
      "step": 3651
    },
    {
      "epoch": 347.8095238095238,
      "grad_norm": 11.512581825256348,
      "learning_rate": 6.842203898050975e-07,
      "loss": 1.1538,
      "step": 3652
    },
    {
      "epoch": 347.9047619047619,
      "grad_norm": 6.716801166534424,
      "learning_rate": 6.844077961019491e-07,
      "loss": 1.1551,
      "step": 3653
    },
    {
      "epoch": 348.0,
      "grad_norm": 10.907341957092285,
      "learning_rate": 6.845952023988006e-07,
      "loss": 1.0937,
      "step": 3654
    },
    {
      "epoch": 348.0,
      "eval_accuracy": 0.7343574372424129,
      "eval_f1": 0.7840389887298203,
      "eval_loss": 0.5535434484481812,
      "eval_precision": 0.7400805060379528,
      "eval_recall": 0.8335492227979274,
      "eval_roc_auc": 0.7883514680483592,
      "eval_runtime": 1.1299,
      "eval_samples_per_second": 2362.186,
      "eval_steps_per_second": 5.31,
      "step": 3654
    },
    {
      "epoch": 348.0952380952381,
      "grad_norm": 10.432865142822266,
      "learning_rate": 6.847826086956522e-07,
      "loss": 1.1648,
      "step": 3655
    },
    {
      "epoch": 348.1904761904762,
      "grad_norm": 9.500462532043457,
      "learning_rate": 6.849700149925038e-07,
      "loss": 1.0846,
      "step": 3656
    },
    {
      "epoch": 348.2857142857143,
      "grad_norm": 6.565241813659668,
      "learning_rate": 6.851574212893554e-07,
      "loss": 1.096,
      "step": 3657
    },
    {
      "epoch": 348.3809523809524,
      "grad_norm": 6.4096221923828125,
      "learning_rate": 6.853448275862069e-07,
      "loss": 1.1165,
      "step": 3658
    },
    {
      "epoch": 348.4761904761905,
      "grad_norm": 5.140103816986084,
      "learning_rate": 6.855322338830585e-07,
      "loss": 1.0618,
      "step": 3659
    },
    {
      "epoch": 348.57142857142856,
      "grad_norm": 5.91832160949707,
      "learning_rate": 6.857196401799101e-07,
      "loss": 1.1687,
      "step": 3660
    },
    {
      "epoch": 348.6666666666667,
      "grad_norm": 5.8545379638671875,
      "learning_rate": 6.859070464767616e-07,
      "loss": 1.1406,
      "step": 3661
    },
    {
      "epoch": 348.76190476190476,
      "grad_norm": 14.378029823303223,
      "learning_rate": 6.860944527736132e-07,
      "loss": 1.141,
      "step": 3662
    },
    {
      "epoch": 348.85714285714283,
      "grad_norm": 7.654487133026123,
      "learning_rate": 6.862818590704648e-07,
      "loss": 1.1031,
      "step": 3663
    },
    {
      "epoch": 348.95238095238096,
      "grad_norm": 13.655937194824219,
      "learning_rate": 6.864692653673164e-07,
      "loss": 1.1128,
      "step": 3664
    },
    {
      "epoch": 348.95238095238096,
      "eval_accuracy": 0.738478831022855,
      "eval_f1": 0.7913927077106994,
      "eval_loss": 0.5536009669303894,
      "eval_precision": 0.7347391786903441,
      "eval_recall": 0.8575129533678757,
      "eval_roc_auc": 0.7887458261370177,
      "eval_runtime": 1.1331,
      "eval_samples_per_second": 2355.514,
      "eval_steps_per_second": 5.295,
      "step": 3664
    },
    {
      "epoch": 349.04761904761904,
      "grad_norm": 16.01888084411621,
      "learning_rate": 6.866566716641679e-07,
      "loss": 1.1669,
      "step": 3665
    },
    {
      "epoch": 349.14285714285717,
      "grad_norm": 7.082530975341797,
      "learning_rate": 6.868440779610195e-07,
      "loss": 1.0966,
      "step": 3666
    },
    {
      "epoch": 349.23809523809524,
      "grad_norm": 5.911025524139404,
      "learning_rate": 6.87031484257871e-07,
      "loss": 1.0857,
      "step": 3667
    },
    {
      "epoch": 349.3333333333333,
      "grad_norm": 3.9199273586273193,
      "learning_rate": 6.872188905547226e-07,
      "loss": 1.1304,
      "step": 3668
    },
    {
      "epoch": 349.42857142857144,
      "grad_norm": 8.267579078674316,
      "learning_rate": 6.874062968515743e-07,
      "loss": 1.1276,
      "step": 3669
    },
    {
      "epoch": 349.5238095238095,
      "grad_norm": 11.254096984863281,
      "learning_rate": 6.875937031484258e-07,
      "loss": 1.1101,
      "step": 3670
    },
    {
      "epoch": 349.6190476190476,
      "grad_norm": 12.040755271911621,
      "learning_rate": 6.877811094452775e-07,
      "loss": 1.0977,
      "step": 3671
    },
    {
      "epoch": 349.7142857142857,
      "grad_norm": 8.342375755310059,
      "learning_rate": 6.87968515742129e-07,
      "loss": 1.1664,
      "step": 3672
    },
    {
      "epoch": 349.8095238095238,
      "grad_norm": 6.742152214050293,
      "learning_rate": 6.881559220389806e-07,
      "loss": 1.0981,
      "step": 3673
    },
    {
      "epoch": 349.9047619047619,
      "grad_norm": 5.508918285369873,
      "learning_rate": 6.883433283358321e-07,
      "loss": 1.1162,
      "step": 3674
    },
    {
      "epoch": 350.0,
      "grad_norm": 6.856598377227783,
      "learning_rate": 6.885307346326837e-07,
      "loss": 1.149,
      "step": 3675
    },
    {
      "epoch": 350.0,
      "eval_accuracy": 0.7358561258898464,
      "eval_f1": 0.7887323943661971,
      "eval_loss": 0.553325891494751,
      "eval_precision": 0.7339654210819855,
      "eval_recall": 0.8523316062176166,
      "eval_roc_auc": 0.7887636729994243,
      "eval_runtime": 1.1307,
      "eval_samples_per_second": 2360.581,
      "eval_steps_per_second": 5.307,
      "step": 3675
    },
    {
      "epoch": 350.0952380952381,
      "grad_norm": 7.425978183746338,
      "learning_rate": 6.887181409295352e-07,
      "loss": 1.1165,
      "step": 3676
    },
    {
      "epoch": 350.1904761904762,
      "grad_norm": 11.71827220916748,
      "learning_rate": 6.889055472263869e-07,
      "loss": 1.1188,
      "step": 3677
    },
    {
      "epoch": 350.2857142857143,
      "grad_norm": 12.371844291687012,
      "learning_rate": 6.890929535232385e-07,
      "loss": 1.102,
      "step": 3678
    },
    {
      "epoch": 350.3809523809524,
      "grad_norm": 13.872154235839844,
      "learning_rate": 6.8928035982009e-07,
      "loss": 1.1576,
      "step": 3679
    },
    {
      "epoch": 350.4761904761905,
      "grad_norm": 15.714462280273438,
      "learning_rate": 6.894677661169416e-07,
      "loss": 1.0988,
      "step": 3680
    },
    {
      "epoch": 350.57142857142856,
      "grad_norm": 10.854438781738281,
      "learning_rate": 6.896551724137931e-07,
      "loss": 1.1247,
      "step": 3681
    },
    {
      "epoch": 350.6666666666667,
      "grad_norm": 4.670620918273926,
      "learning_rate": 6.898425787106447e-07,
      "loss": 1.1434,
      "step": 3682
    },
    {
      "epoch": 350.76190476190476,
      "grad_norm": 6.489286422729492,
      "learning_rate": 6.900299850074962e-07,
      "loss": 1.1209,
      "step": 3683
    },
    {
      "epoch": 350.85714285714283,
      "grad_norm": 5.2290358543396,
      "learning_rate": 6.902173913043479e-07,
      "loss": 1.1203,
      "step": 3684
    },
    {
      "epoch": 350.95238095238096,
      "grad_norm": 16.88265037536621,
      "learning_rate": 6.904047976011994e-07,
      "loss": 1.1177,
      "step": 3685
    },
    {
      "epoch": 350.95238095238096,
      "eval_accuracy": 0.732484076433121,
      "eval_f1": 0.7811158798283262,
      "eval_loss": 0.5532238483428955,
      "eval_precision": 0.7415599534342259,
      "eval_recall": 0.8251295336787565,
      "eval_roc_auc": 0.7884602763385147,
      "eval_runtime": 1.1127,
      "eval_samples_per_second": 2398.725,
      "eval_steps_per_second": 5.392,
      "step": 3685
    },
    {
      "epoch": 351.04761904761904,
      "grad_norm": 5.788422107696533,
      "learning_rate": 6.90592203898051e-07,
      "loss": 1.0999,
      "step": 3686
    },
    {
      "epoch": 351.14285714285717,
      "grad_norm": 15.770282745361328,
      "learning_rate": 6.907796101949026e-07,
      "loss": 1.1257,
      "step": 3687
    },
    {
      "epoch": 351.23809523809524,
      "grad_norm": 4.182605266571045,
      "learning_rate": 6.909670164917541e-07,
      "loss": 1.1125,
      "step": 3688
    },
    {
      "epoch": 351.3333333333333,
      "grad_norm": 7.102372169494629,
      "learning_rate": 6.911544227886057e-07,
      "loss": 1.1161,
      "step": 3689
    },
    {
      "epoch": 351.42857142857144,
      "grad_norm": 7.176307201385498,
      "learning_rate": 6.913418290854572e-07,
      "loss": 1.144,
      "step": 3690
    },
    {
      "epoch": 351.5238095238095,
      "grad_norm": 7.356261253356934,
      "learning_rate": 6.915292353823089e-07,
      "loss": 1.1582,
      "step": 3691
    },
    {
      "epoch": 351.6190476190476,
      "grad_norm": 7.973656177520752,
      "learning_rate": 6.917166416791604e-07,
      "loss": 1.1195,
      "step": 3692
    },
    {
      "epoch": 351.7142857142857,
      "grad_norm": 6.8437018394470215,
      "learning_rate": 6.91904047976012e-07,
      "loss": 1.1236,
      "step": 3693
    },
    {
      "epoch": 351.8095238095238,
      "grad_norm": 6.552664279937744,
      "learning_rate": 6.920914542728636e-07,
      "loss": 1.1384,
      "step": 3694
    },
    {
      "epoch": 351.9047619047619,
      "grad_norm": 8.255615234375,
      "learning_rate": 6.922788605697152e-07,
      "loss": 1.1148,
      "step": 3695
    },
    {
      "epoch": 352.0,
      "grad_norm": 8.523663520812988,
      "learning_rate": 6.924662668665668e-07,
      "loss": 1.107,
      "step": 3696
    },
    {
      "epoch": 352.0,
      "eval_accuracy": 0.7298613713001124,
      "eval_f1": 0.7791730474732006,
      "eval_loss": 0.5536131262779236,
      "eval_precision": 0.7391051714119697,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.7881154289004031,
      "eval_runtime": 1.1313,
      "eval_samples_per_second": 2359.162,
      "eval_steps_per_second": 5.303,
      "step": 3696
    },
    {
      "epoch": 352.0952380952381,
      "grad_norm": 6.894546031951904,
      "learning_rate": 6.926536731634183e-07,
      "loss": 1.1187,
      "step": 3697
    },
    {
      "epoch": 352.1904761904762,
      "grad_norm": 12.019125938415527,
      "learning_rate": 6.9284107946027e-07,
      "loss": 1.1303,
      "step": 3698
    },
    {
      "epoch": 352.2857142857143,
      "grad_norm": 6.342089653015137,
      "learning_rate": 6.930284857571215e-07,
      "loss": 1.1312,
      "step": 3699
    },
    {
      "epoch": 352.3809523809524,
      "grad_norm": 5.598708152770996,
      "learning_rate": 6.932158920539731e-07,
      "loss": 1.1504,
      "step": 3700
    },
    {
      "epoch": 352.4761904761905,
      "grad_norm": 9.867156982421875,
      "learning_rate": 6.934032983508246e-07,
      "loss": 1.127,
      "step": 3701
    },
    {
      "epoch": 352.57142857142856,
      "grad_norm": 3.7137577533721924,
      "learning_rate": 6.935907046476762e-07,
      "loss": 1.0934,
      "step": 3702
    },
    {
      "epoch": 352.6666666666667,
      "grad_norm": 5.500702857971191,
      "learning_rate": 6.937781109445277e-07,
      "loss": 1.1383,
      "step": 3703
    },
    {
      "epoch": 352.76190476190476,
      "grad_norm": 7.868194103240967,
      "learning_rate": 6.939655172413793e-07,
      "loss": 1.0965,
      "step": 3704
    },
    {
      "epoch": 352.85714285714283,
      "grad_norm": 8.515414237976074,
      "learning_rate": 6.94152923538231e-07,
      "loss": 1.1162,
      "step": 3705
    },
    {
      "epoch": 352.95238095238096,
      "grad_norm": 6.582380771636963,
      "learning_rate": 6.943403298350825e-07,
      "loss": 1.1019,
      "step": 3706
    },
    {
      "epoch": 352.95238095238096,
      "eval_accuracy": 0.7362307980517048,
      "eval_f1": 0.7908496732026143,
      "eval_loss": 0.5542948842048645,
      "eval_precision": 0.7305159165751921,
      "eval_recall": 0.8620466321243523,
      "eval_roc_auc": 0.7884173862982153,
      "eval_runtime": 1.1274,
      "eval_samples_per_second": 2367.475,
      "eval_steps_per_second": 5.322,
      "step": 3706
    },
    {
      "epoch": 353.04761904761904,
      "grad_norm": 6.112668991088867,
      "learning_rate": 6.945277361319341e-07,
      "loss": 1.1165,
      "step": 3707
    },
    {
      "epoch": 353.14285714285717,
      "grad_norm": 15.119726181030273,
      "learning_rate": 6.947151424287856e-07,
      "loss": 1.1496,
      "step": 3708
    },
    {
      "epoch": 353.23809523809524,
      "grad_norm": 11.367457389831543,
      "learning_rate": 6.949025487256372e-07,
      "loss": 1.1206,
      "step": 3709
    },
    {
      "epoch": 353.3333333333333,
      "grad_norm": 6.386621952056885,
      "learning_rate": 6.950899550224887e-07,
      "loss": 1.1265,
      "step": 3710
    },
    {
      "epoch": 353.42857142857144,
      "grad_norm": 9.647610664367676,
      "learning_rate": 6.952773613193403e-07,
      "loss": 1.1095,
      "step": 3711
    },
    {
      "epoch": 353.5238095238095,
      "grad_norm": 10.258861541748047,
      "learning_rate": 6.95464767616192e-07,
      "loss": 1.0473,
      "step": 3712
    },
    {
      "epoch": 353.6190476190476,
      "grad_norm": 7.602921485900879,
      "learning_rate": 6.956521739130435e-07,
      "loss": 1.0915,
      "step": 3713
    },
    {
      "epoch": 353.7142857142857,
      "grad_norm": 10.235103607177734,
      "learning_rate": 6.958395802098951e-07,
      "loss": 1.1615,
      "step": 3714
    },
    {
      "epoch": 353.8095238095238,
      "grad_norm": 18.607894897460938,
      "learning_rate": 6.960269865067466e-07,
      "loss": 1.1101,
      "step": 3715
    },
    {
      "epoch": 353.9047619047619,
      "grad_norm": 5.014978408813477,
      "learning_rate": 6.962143928035983e-07,
      "loss": 1.1291,
      "step": 3716
    },
    {
      "epoch": 354.0,
      "grad_norm": 4.992445945739746,
      "learning_rate": 6.964017991004498e-07,
      "loss": 1.1391,
      "step": 3717
    },
    {
      "epoch": 354.0,
      "eval_accuracy": 0.7317347321094043,
      "eval_f1": 0.7800982800982801,
      "eval_loss": 0.5533689260482788,
      "eval_precision": 0.741822429906542,
      "eval_recall": 0.822538860103627,
      "eval_roc_auc": 0.7882397812320091,
      "eval_runtime": 1.1277,
      "eval_samples_per_second": 2366.68,
      "eval_steps_per_second": 5.32,
      "step": 3717
    },
    {
      "epoch": 354.0952380952381,
      "grad_norm": 3.7810394763946533,
      "learning_rate": 6.965892053973014e-07,
      "loss": 1.1567,
      "step": 3718
    },
    {
      "epoch": 354.1904761904762,
      "grad_norm": 15.194648742675781,
      "learning_rate": 6.96776611694153e-07,
      "loss": 1.1036,
      "step": 3719
    },
    {
      "epoch": 354.2857142857143,
      "grad_norm": 18.574241638183594,
      "learning_rate": 6.969640179910046e-07,
      "loss": 1.047,
      "step": 3720
    },
    {
      "epoch": 354.3809523809524,
      "grad_norm": 6.198830604553223,
      "learning_rate": 6.971514242878561e-07,
      "loss": 1.1338,
      "step": 3721
    },
    {
      "epoch": 354.4761904761905,
      "grad_norm": 6.480565547943115,
      "learning_rate": 6.973388305847077e-07,
      "loss": 1.0989,
      "step": 3722
    },
    {
      "epoch": 354.57142857142856,
      "grad_norm": 7.632514953613281,
      "learning_rate": 6.975262368815593e-07,
      "loss": 1.0984,
      "step": 3723
    },
    {
      "epoch": 354.6666666666667,
      "grad_norm": 9.1426362991333,
      "learning_rate": 6.977136431784108e-07,
      "loss": 1.1029,
      "step": 3724
    },
    {
      "epoch": 354.76190476190476,
      "grad_norm": 6.421456336975098,
      "learning_rate": 6.979010494752624e-07,
      "loss": 1.1504,
      "step": 3725
    },
    {
      "epoch": 354.85714285714283,
      "grad_norm": 15.901803970336914,
      "learning_rate": 6.98088455772114e-07,
      "loss": 1.1431,
      "step": 3726
    },
    {
      "epoch": 354.95238095238096,
      "grad_norm": 6.8048014640808105,
      "learning_rate": 6.982758620689656e-07,
      "loss": 1.1198,
      "step": 3727
    },
    {
      "epoch": 354.95238095238096,
      "eval_accuracy": 0.7362307980517048,
      "eval_f1": 0.7880794701986755,
      "eval_loss": 0.5533745884895325,
      "eval_precision": 0.7362204724409449,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7882170408750719,
      "eval_runtime": 1.4072,
      "eval_samples_per_second": 1896.614,
      "eval_steps_per_second": 4.264,
      "step": 3727
    },
    {
      "epoch": 355.04761904761904,
      "grad_norm": 7.465842247009277,
      "learning_rate": 6.984632683658171e-07,
      "loss": 1.1824,
      "step": 3728
    },
    {
      "epoch": 355.14285714285717,
      "grad_norm": 6.699007034301758,
      "learning_rate": 6.986506746626687e-07,
      "loss": 1.1086,
      "step": 3729
    },
    {
      "epoch": 355.23809523809524,
      "grad_norm": 5.756167411804199,
      "learning_rate": 6.988380809595202e-07,
      "loss": 1.0881,
      "step": 3730
    },
    {
      "epoch": 355.3333333333333,
      "grad_norm": 11.109375953674316,
      "learning_rate": 6.990254872563718e-07,
      "loss": 1.0663,
      "step": 3731
    },
    {
      "epoch": 355.42857142857144,
      "grad_norm": 6.402453899383545,
      "learning_rate": 6.992128935532234e-07,
      "loss": 1.1266,
      "step": 3732
    },
    {
      "epoch": 355.5238095238095,
      "grad_norm": 5.234026908874512,
      "learning_rate": 6.99400299850075e-07,
      "loss": 1.089,
      "step": 3733
    },
    {
      "epoch": 355.6190476190476,
      "grad_norm": 7.56530237197876,
      "learning_rate": 6.995877061469266e-07,
      "loss": 1.1751,
      "step": 3734
    },
    {
      "epoch": 355.7142857142857,
      "grad_norm": 4.21050500869751,
      "learning_rate": 6.997751124437781e-07,
      "loss": 1.1163,
      "step": 3735
    },
    {
      "epoch": 355.8095238095238,
      "grad_norm": 8.78940200805664,
      "learning_rate": 6.999625187406297e-07,
      "loss": 1.0849,
      "step": 3736
    },
    {
      "epoch": 355.9047619047619,
      "grad_norm": 10.318490028381348,
      "learning_rate": 7.001499250374812e-07,
      "loss": 1.1286,
      "step": 3737
    },
    {
      "epoch": 356.0,
      "grad_norm": 7.933201313018799,
      "learning_rate": 7.003373313343329e-07,
      "loss": 1.1523,
      "step": 3738
    },
    {
      "epoch": 356.0,
      "eval_accuracy": 0.7388535031847133,
      "eval_f1": 0.7896166616359794,
      "eval_loss": 0.5528306365013123,
      "eval_precision": 0.7394007914075749,
      "eval_recall": 0.8471502590673575,
      "eval_roc_auc": 0.7888730569948186,
      "eval_runtime": 1.1105,
      "eval_samples_per_second": 2403.431,
      "eval_steps_per_second": 5.403,
      "step": 3738
    },
    {
      "epoch": 356.0952380952381,
      "grad_norm": 6.1067070960998535,
      "learning_rate": 7.005247376311844e-07,
      "loss": 1.1268,
      "step": 3739
    },
    {
      "epoch": 356.1904761904762,
      "grad_norm": 10.120503425598145,
      "learning_rate": 7.00712143928036e-07,
      "loss": 1.1142,
      "step": 3740
    },
    {
      "epoch": 356.2857142857143,
      "grad_norm": 4.93829870223999,
      "learning_rate": 7.008995502248877e-07,
      "loss": 1.0877,
      "step": 3741
    },
    {
      "epoch": 356.3809523809524,
      "grad_norm": 10.851411819458008,
      "learning_rate": 7.010869565217392e-07,
      "loss": 1.1183,
      "step": 3742
    },
    {
      "epoch": 356.4761904761905,
      "grad_norm": 12.106622695922852,
      "learning_rate": 7.012743628185908e-07,
      "loss": 1.1104,
      "step": 3743
    },
    {
      "epoch": 356.57142857142856,
      "grad_norm": 7.220052719116211,
      "learning_rate": 7.014617691154423e-07,
      "loss": 1.1312,
      "step": 3744
    },
    {
      "epoch": 356.6666666666667,
      "grad_norm": 5.6121015548706055,
      "learning_rate": 7.016491754122939e-07,
      "loss": 1.1463,
      "step": 3745
    },
    {
      "epoch": 356.76190476190476,
      "grad_norm": 4.8981146812438965,
      "learning_rate": 7.018365817091454e-07,
      "loss": 1.1288,
      "step": 3746
    },
    {
      "epoch": 356.85714285714283,
      "grad_norm": 5.686845779418945,
      "learning_rate": 7.020239880059971e-07,
      "loss": 1.115,
      "step": 3747
    },
    {
      "epoch": 356.95238095238096,
      "grad_norm": 4.765906810760498,
      "learning_rate": 7.022113943028487e-07,
      "loss": 1.1609,
      "step": 3748
    },
    {
      "epoch": 356.95238095238096,
      "eval_accuracy": 0.735481453727988,
      "eval_f1": 0.7838334353949785,
      "eval_loss": 0.5525299310684204,
      "eval_precision": 0.743321718931475,
      "eval_recall": 0.8290155440414507,
      "eval_roc_auc": 0.7889815774323548,
      "eval_runtime": 1.1157,
      "eval_samples_per_second": 2392.192,
      "eval_steps_per_second": 5.378,
      "step": 3748
    },
    {
      "epoch": 357.04761904761904,
      "grad_norm": 11.675453186035156,
      "learning_rate": 7.023988005997002e-07,
      "loss": 1.0912,
      "step": 3749
    },
    {
      "epoch": 357.14285714285717,
      "grad_norm": 5.98292875289917,
      "learning_rate": 7.025862068965518e-07,
      "loss": 1.1203,
      "step": 3750
    },
    {
      "epoch": 357.23809523809524,
      "grad_norm": 4.8235554695129395,
      "learning_rate": 7.027736131934033e-07,
      "loss": 1.1239,
      "step": 3751
    },
    {
      "epoch": 357.3333333333333,
      "grad_norm": 7.256765842437744,
      "learning_rate": 7.029610194902549e-07,
      "loss": 1.1065,
      "step": 3752
    },
    {
      "epoch": 357.42857142857144,
      "grad_norm": 9.69433879852295,
      "learning_rate": 7.031484257871064e-07,
      "loss": 1.157,
      "step": 3753
    },
    {
      "epoch": 357.5238095238095,
      "grad_norm": 6.654829025268555,
      "learning_rate": 7.033358320839581e-07,
      "loss": 1.1326,
      "step": 3754
    },
    {
      "epoch": 357.6190476190476,
      "grad_norm": 6.713947772979736,
      "learning_rate": 7.035232383808096e-07,
      "loss": 1.0654,
      "step": 3755
    },
    {
      "epoch": 357.7142857142857,
      "grad_norm": 11.10183334350586,
      "learning_rate": 7.037106446776612e-07,
      "loss": 1.1352,
      "step": 3756
    },
    {
      "epoch": 357.8095238095238,
      "grad_norm": 6.993966102600098,
      "learning_rate": 7.038980509745127e-07,
      "loss": 1.1249,
      "step": 3757
    },
    {
      "epoch": 357.9047619047619,
      "grad_norm": 11.608559608459473,
      "learning_rate": 7.040854572713643e-07,
      "loss": 1.0958,
      "step": 3758
    },
    {
      "epoch": 358.0,
      "grad_norm": 12.34938907623291,
      "learning_rate": 7.042728635682159e-07,
      "loss": 1.1323,
      "step": 3759
    },
    {
      "epoch": 358.0,
      "eval_accuracy": 0.7392281753465717,
      "eval_f1": 0.7896009673518742,
      "eval_loss": 0.5525399446487427,
      "eval_precision": 0.7403628117913832,
      "eval_recall": 0.8458549222797928,
      "eval_roc_auc": 0.7890290731145653,
      "eval_runtime": 1.1201,
      "eval_samples_per_second": 2382.802,
      "eval_steps_per_second": 5.357,
      "step": 3759
    },
    {
      "epoch": 358.0952380952381,
      "grad_norm": 17.054271697998047,
      "learning_rate": 7.044602698650674e-07,
      "loss": 1.0948,
      "step": 3760
    },
    {
      "epoch": 358.1904761904762,
      "grad_norm": 17.62253761291504,
      "learning_rate": 7.046476761619191e-07,
      "loss": 1.1152,
      "step": 3761
    },
    {
      "epoch": 358.2857142857143,
      "grad_norm": 7.233613014221191,
      "learning_rate": 7.048350824587706e-07,
      "loss": 1.0497,
      "step": 3762
    },
    {
      "epoch": 358.3809523809524,
      "grad_norm": 19.385459899902344,
      "learning_rate": 7.050224887556223e-07,
      "loss": 1.1506,
      "step": 3763
    },
    {
      "epoch": 358.4761904761905,
      "grad_norm": 10.535709381103516,
      "learning_rate": 7.052098950524738e-07,
      "loss": 1.1232,
      "step": 3764
    },
    {
      "epoch": 358.57142857142856,
      "grad_norm": 11.408620834350586,
      "learning_rate": 7.053973013493254e-07,
      "loss": 1.1645,
      "step": 3765
    },
    {
      "epoch": 358.6666666666667,
      "grad_norm": 17.90571403503418,
      "learning_rate": 7.05584707646177e-07,
      "loss": 1.113,
      "step": 3766
    },
    {
      "epoch": 358.76190476190476,
      "grad_norm": 5.536513805389404,
      "learning_rate": 7.057721139430285e-07,
      "loss": 1.1213,
      "step": 3767
    },
    {
      "epoch": 358.85714285714283,
      "grad_norm": 8.541691780090332,
      "learning_rate": 7.059595202398802e-07,
      "loss": 1.1596,
      "step": 3768
    },
    {
      "epoch": 358.95238095238096,
      "grad_norm": 7.569893836975098,
      "learning_rate": 7.061469265367317e-07,
      "loss": 1.1625,
      "step": 3769
    },
    {
      "epoch": 358.95238095238096,
      "eval_accuracy": 0.7369801423754215,
      "eval_f1": 0.7846625766871166,
      "eval_loss": 0.5524272322654724,
      "eval_precision": 0.7453379953379954,
      "eval_recall": 0.8283678756476683,
      "eval_roc_auc": 0.7891099597006334,
      "eval_runtime": 1.123,
      "eval_samples_per_second": 2376.639,
      "eval_steps_per_second": 5.343,
      "step": 3769
    },
    {
      "epoch": 359.04761904761904,
      "grad_norm": 6.949024200439453,
      "learning_rate": 7.063343328335833e-07,
      "loss": 1.1192,
      "step": 3770
    },
    {
      "epoch": 359.14285714285717,
      "grad_norm": 5.126923084259033,
      "learning_rate": 7.065217391304348e-07,
      "loss": 1.1206,
      "step": 3771
    },
    {
      "epoch": 359.23809523809524,
      "grad_norm": 8.701910972595215,
      "learning_rate": 7.067091454272864e-07,
      "loss": 1.1306,
      "step": 3772
    },
    {
      "epoch": 359.3333333333333,
      "grad_norm": 15.019597053527832,
      "learning_rate": 7.068965517241379e-07,
      "loss": 1.1335,
      "step": 3773
    },
    {
      "epoch": 359.42857142857144,
      "grad_norm": 11.442471504211426,
      "learning_rate": 7.070839580209895e-07,
      "loss": 1.121,
      "step": 3774
    },
    {
      "epoch": 359.5238095238095,
      "grad_norm": 7.127864837646484,
      "learning_rate": 7.072713643178412e-07,
      "loss": 1.1044,
      "step": 3775
    },
    {
      "epoch": 359.6190476190476,
      "grad_norm": 13.844635963439941,
      "learning_rate": 7.074587706146927e-07,
      "loss": 1.1022,
      "step": 3776
    },
    {
      "epoch": 359.7142857142857,
      "grad_norm": 8.636882781982422,
      "learning_rate": 7.076461769115443e-07,
      "loss": 1.1134,
      "step": 3777
    },
    {
      "epoch": 359.8095238095238,
      "grad_norm": 18.998241424560547,
      "learning_rate": 7.078335832083958e-07,
      "loss": 1.121,
      "step": 3778
    },
    {
      "epoch": 359.9047619047619,
      "grad_norm": 15.878497123718262,
      "learning_rate": 7.080209895052474e-07,
      "loss": 1.1422,
      "step": 3779
    },
    {
      "epoch": 360.0,
      "grad_norm": 4.503548622131348,
      "learning_rate": 7.082083958020989e-07,
      "loss": 1.1042,
      "step": 3780
    },
    {
      "epoch": 360.0,
      "eval_accuracy": 0.741476208317722,
      "eval_f1": 0.7927927927927928,
      "eval_loss": 0.5527340769767761,
      "eval_precision": 0.7390817469204927,
      "eval_recall": 0.8549222797927462,
      "eval_roc_auc": 0.7891531375935521,
      "eval_runtime": 1.1228,
      "eval_samples_per_second": 2377.189,
      "eval_steps_per_second": 5.344,
      "step": 3780
    },
    {
      "epoch": 360.0952380952381,
      "grad_norm": 18.075942993164062,
      "learning_rate": 7.083958020989505e-07,
      "loss": 1.1838,
      "step": 3781
    },
    {
      "epoch": 360.1904761904762,
      "grad_norm": 5.373623371124268,
      "learning_rate": 7.085832083958021e-07,
      "loss": 1.1442,
      "step": 3782
    },
    {
      "epoch": 360.2857142857143,
      "grad_norm": 4.5523681640625,
      "learning_rate": 7.087706146926537e-07,
      "loss": 1.1005,
      "step": 3783
    },
    {
      "epoch": 360.3809523809524,
      "grad_norm": 10.179972648620605,
      "learning_rate": 7.089580209895052e-07,
      "loss": 1.1009,
      "step": 3784
    },
    {
      "epoch": 360.4761904761905,
      "grad_norm": 9.946412086486816,
      "learning_rate": 7.091454272863568e-07,
      "loss": 1.1263,
      "step": 3785
    },
    {
      "epoch": 360.57142857142856,
      "grad_norm": 6.647387981414795,
      "learning_rate": 7.093328335832085e-07,
      "loss": 1.133,
      "step": 3786
    },
    {
      "epoch": 360.6666666666667,
      "grad_norm": 6.924274921417236,
      "learning_rate": 7.0952023988006e-07,
      "loss": 1.1009,
      "step": 3787
    },
    {
      "epoch": 360.76190476190476,
      "grad_norm": 6.803415298461914,
      "learning_rate": 7.097076461769116e-07,
      "loss": 1.0819,
      "step": 3788
    },
    {
      "epoch": 360.85714285714283,
      "grad_norm": 8.356952667236328,
      "learning_rate": 7.098950524737632e-07,
      "loss": 1.1109,
      "step": 3789
    },
    {
      "epoch": 360.95238095238096,
      "grad_norm": 7.981040000915527,
      "learning_rate": 7.100824587706148e-07,
      "loss": 1.1094,
      "step": 3790
    },
    {
      "epoch": 360.95238095238096,
      "eval_accuracy": 0.7377294866991383,
      "eval_f1": 0.7872340425531915,
      "eval_loss": 0.5522446632385254,
      "eval_precision": 0.7416953035509737,
      "eval_recall": 0.8387305699481865,
      "eval_roc_auc": 0.7897343120322395,
      "eval_runtime": 1.1203,
      "eval_samples_per_second": 2382.456,
      "eval_steps_per_second": 5.356,
      "step": 3790
    },
    {
      "epoch": 361.04761904761904,
      "grad_norm": 10.223261833190918,
      "learning_rate": 7.102698650674663e-07,
      "loss": 1.0982,
      "step": 3791
    },
    {
      "epoch": 361.14285714285717,
      "grad_norm": 4.465803623199463,
      "learning_rate": 7.104572713643179e-07,
      "loss": 1.0778,
      "step": 3792
    },
    {
      "epoch": 361.23809523809524,
      "grad_norm": 8.075662612915039,
      "learning_rate": 7.106446776611694e-07,
      "loss": 1.1723,
      "step": 3793
    },
    {
      "epoch": 361.3333333333333,
      "grad_norm": 10.181119918823242,
      "learning_rate": 7.10832083958021e-07,
      "loss": 1.1023,
      "step": 3794
    },
    {
      "epoch": 361.42857142857144,
      "grad_norm": 6.193162441253662,
      "learning_rate": 7.110194902548726e-07,
      "loss": 1.123,
      "step": 3795
    },
    {
      "epoch": 361.5238095238095,
      "grad_norm": 3.2716386318206787,
      "learning_rate": 7.112068965517242e-07,
      "loss": 1.1325,
      "step": 3796
    },
    {
      "epoch": 361.6190476190476,
      "grad_norm": 6.4619832038879395,
      "learning_rate": 7.113943028485758e-07,
      "loss": 1.1629,
      "step": 3797
    },
    {
      "epoch": 361.7142857142857,
      "grad_norm": 6.093001365661621,
      "learning_rate": 7.115817091454273e-07,
      "loss": 1.092,
      "step": 3798
    },
    {
      "epoch": 361.8095238095238,
      "grad_norm": 14.338656425476074,
      "learning_rate": 7.117691154422789e-07,
      "loss": 1.1473,
      "step": 3799
    },
    {
      "epoch": 361.9047619047619,
      "grad_norm": 11.990729331970215,
      "learning_rate": 7.119565217391304e-07,
      "loss": 1.1122,
      "step": 3800
    },
    {
      "epoch": 362.0,
      "grad_norm": 4.780630111694336,
      "learning_rate": 7.12143928035982e-07,
      "loss": 1.0641,
      "step": 3801
    },
    {
      "epoch": 362.0,
      "eval_accuracy": 0.7373548145372799,
      "eval_f1": 0.7872534142640364,
      "eval_loss": 0.5522888898849487,
      "eval_precision": 0.7407195888063963,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7897772020725388,
      "eval_runtime": 1.1352,
      "eval_samples_per_second": 2351.214,
      "eval_steps_per_second": 5.286,
      "step": 3801
    },
    {
      "epoch": 362.0952380952381,
      "grad_norm": 9.136756896972656,
      "learning_rate": 7.123313343328335e-07,
      "loss": 1.0569,
      "step": 3802
    },
    {
      "epoch": 362.1904761904762,
      "grad_norm": 17.731273651123047,
      "learning_rate": 7.125187406296852e-07,
      "loss": 1.1098,
      "step": 3803
    },
    {
      "epoch": 362.2857142857143,
      "grad_norm": 5.175982475280762,
      "learning_rate": 7.127061469265368e-07,
      "loss": 1.0944,
      "step": 3804
    },
    {
      "epoch": 362.3809523809524,
      "grad_norm": 10.310859680175781,
      "learning_rate": 7.128935532233883e-07,
      "loss": 1.0819,
      "step": 3805
    },
    {
      "epoch": 362.4761904761905,
      "grad_norm": 5.00278377532959,
      "learning_rate": 7.130809595202399e-07,
      "loss": 1.1317,
      "step": 3806
    },
    {
      "epoch": 362.57142857142856,
      "grad_norm": 9.666382789611816,
      "learning_rate": 7.132683658170914e-07,
      "loss": 1.1537,
      "step": 3807
    },
    {
      "epoch": 362.6666666666667,
      "grad_norm": 5.540799617767334,
      "learning_rate": 7.134557721139431e-07,
      "loss": 1.1462,
      "step": 3808
    },
    {
      "epoch": 362.76190476190476,
      "grad_norm": 9.392330169677734,
      "learning_rate": 7.136431784107946e-07,
      "loss": 1.1592,
      "step": 3809
    },
    {
      "epoch": 362.85714285714283,
      "grad_norm": 10.634110450744629,
      "learning_rate": 7.138305847076462e-07,
      "loss": 1.1258,
      "step": 3810
    },
    {
      "epoch": 362.95238095238096,
      "grad_norm": 6.859245300292969,
      "learning_rate": 7.140179910044979e-07,
      "loss": 1.1117,
      "step": 3811
    },
    {
      "epoch": 362.95238095238096,
      "eval_accuracy": 0.741476208317722,
      "eval_f1": 0.7915407854984894,
      "eval_loss": 0.5519759654998779,
      "eval_precision": 0.7417893544733862,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7899614277489926,
      "eval_runtime": 1.1356,
      "eval_samples_per_second": 2350.256,
      "eval_steps_per_second": 5.283,
      "step": 3811
    },
    {
      "epoch": 363.04761904761904,
      "grad_norm": 10.381455421447754,
      "learning_rate": 7.142053973013494e-07,
      "loss": 1.0988,
      "step": 3812
    },
    {
      "epoch": 363.14285714285717,
      "grad_norm": 9.13387680053711,
      "learning_rate": 7.14392803598201e-07,
      "loss": 1.1698,
      "step": 3813
    },
    {
      "epoch": 363.23809523809524,
      "grad_norm": 12.999701499938965,
      "learning_rate": 7.145802098950525e-07,
      "loss": 1.1325,
      "step": 3814
    },
    {
      "epoch": 363.3333333333333,
      "grad_norm": 7.6583781242370605,
      "learning_rate": 7.147676161919041e-07,
      "loss": 1.1042,
      "step": 3815
    },
    {
      "epoch": 363.42857142857144,
      "grad_norm": 17.781267166137695,
      "learning_rate": 7.149550224887556e-07,
      "loss": 1.1218,
      "step": 3816
    },
    {
      "epoch": 363.5238095238095,
      "grad_norm": 7.954900741577148,
      "learning_rate": 7.151424287856073e-07,
      "loss": 1.1074,
      "step": 3817
    },
    {
      "epoch": 363.6190476190476,
      "grad_norm": 9.188843727111816,
      "learning_rate": 7.153298350824588e-07,
      "loss": 1.0778,
      "step": 3818
    },
    {
      "epoch": 363.7142857142857,
      "grad_norm": 10.288064002990723,
      "learning_rate": 7.155172413793104e-07,
      "loss": 1.1257,
      "step": 3819
    },
    {
      "epoch": 363.8095238095238,
      "grad_norm": 5.548375606536865,
      "learning_rate": 7.15704647676162e-07,
      "loss": 1.1675,
      "step": 3820
    },
    {
      "epoch": 363.9047619047619,
      "grad_norm": 9.2520170211792,
      "learning_rate": 7.158920539730135e-07,
      "loss": 1.0806,
      "step": 3821
    },
    {
      "epoch": 364.0,
      "grad_norm": 12.628405570983887,
      "learning_rate": 7.160794602698651e-07,
      "loss": 1.0937,
      "step": 3822
    },
    {
      "epoch": 364.0,
      "eval_accuracy": 0.7381041588609967,
      "eval_f1": 0.7883742052679382,
      "eval_loss": 0.552045464515686,
      "eval_precision": 0.740193291642979,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7898848589522165,
      "eval_runtime": 1.1396,
      "eval_samples_per_second": 2342.127,
      "eval_steps_per_second": 5.265,
      "step": 3822
    },
    {
      "epoch": 364.0952380952381,
      "grad_norm": 7.857677936553955,
      "learning_rate": 7.162668665667166e-07,
      "loss": 1.1142,
      "step": 3823
    },
    {
      "epoch": 364.1904761904762,
      "grad_norm": 6.276848316192627,
      "learning_rate": 7.164542728635683e-07,
      "loss": 1.1138,
      "step": 3824
    },
    {
      "epoch": 364.2857142857143,
      "grad_norm": 11.035969734191895,
      "learning_rate": 7.166416791604198e-07,
      "loss": 1.1548,
      "step": 3825
    },
    {
      "epoch": 364.3809523809524,
      "grad_norm": 9.335686683654785,
      "learning_rate": 7.168290854572714e-07,
      "loss": 1.0909,
      "step": 3826
    },
    {
      "epoch": 364.4761904761905,
      "grad_norm": 6.2040324211120605,
      "learning_rate": 7.170164917541229e-07,
      "loss": 1.0907,
      "step": 3827
    },
    {
      "epoch": 364.57142857142856,
      "grad_norm": 7.269261360168457,
      "learning_rate": 7.172038980509745e-07,
      "loss": 1.0912,
      "step": 3828
    },
    {
      "epoch": 364.6666666666667,
      "grad_norm": 3.7643415927886963,
      "learning_rate": 7.17391304347826e-07,
      "loss": 1.1181,
      "step": 3829
    },
    {
      "epoch": 364.76190476190476,
      "grad_norm": 5.428904056549072,
      "learning_rate": 7.175787106446777e-07,
      "loss": 1.1256,
      "step": 3830
    },
    {
      "epoch": 364.85714285714283,
      "grad_norm": 11.517545700073242,
      "learning_rate": 7.177661169415293e-07,
      "loss": 1.0952,
      "step": 3831
    },
    {
      "epoch": 364.95238095238096,
      "grad_norm": 9.582964897155762,
      "learning_rate": 7.179535232383808e-07,
      "loss": 1.1242,
      "step": 3832
    },
    {
      "epoch": 364.95238095238096,
      "eval_accuracy": 0.7358561258898464,
      "eval_f1": 0.7831436481082744,
      "eval_loss": 0.5519841313362122,
      "eval_precision": 0.7457527826596367,
      "eval_recall": 0.8244818652849741,
      "eval_roc_auc": 0.7898721934369602,
      "eval_runtime": 1.1201,
      "eval_samples_per_second": 2382.779,
      "eval_steps_per_second": 5.357,
      "step": 3832
    },
    {
      "epoch": 365.04761904761904,
      "grad_norm": 5.688808441162109,
      "learning_rate": 7.181409295352325e-07,
      "loss": 1.1182,
      "step": 3833
    },
    {
      "epoch": 365.14285714285717,
      "grad_norm": 8.307438850402832,
      "learning_rate": 7.18328335832084e-07,
      "loss": 1.0878,
      "step": 3834
    },
    {
      "epoch": 365.23809523809524,
      "grad_norm": 7.110036373138428,
      "learning_rate": 7.185157421289356e-07,
      "loss": 1.1538,
      "step": 3835
    },
    {
      "epoch": 365.3333333333333,
      "grad_norm": 6.807857990264893,
      "learning_rate": 7.187031484257871e-07,
      "loss": 1.1287,
      "step": 3836
    },
    {
      "epoch": 365.42857142857144,
      "grad_norm": 8.348471641540527,
      "learning_rate": 7.188905547226387e-07,
      "loss": 1.0766,
      "step": 3837
    },
    {
      "epoch": 365.5238095238095,
      "grad_norm": 6.609596252441406,
      "learning_rate": 7.190779610194904e-07,
      "loss": 1.1401,
      "step": 3838
    },
    {
      "epoch": 365.6190476190476,
      "grad_norm": 8.325094223022461,
      "learning_rate": 7.192653673163419e-07,
      "loss": 1.121,
      "step": 3839
    },
    {
      "epoch": 365.7142857142857,
      "grad_norm": 5.06738805770874,
      "learning_rate": 7.194527736131935e-07,
      "loss": 1.0984,
      "step": 3840
    },
    {
      "epoch": 365.8095238095238,
      "grad_norm": 8.403528213500977,
      "learning_rate": 7.19640179910045e-07,
      "loss": 1.103,
      "step": 3841
    },
    {
      "epoch": 365.9047619047619,
      "grad_norm": 11.857973098754883,
      "learning_rate": 7.198275862068966e-07,
      "loss": 1.147,
      "step": 3842
    },
    {
      "epoch": 366.0,
      "grad_norm": 9.966644287109375,
      "learning_rate": 7.200149925037481e-07,
      "loss": 1.1497,
      "step": 3843
    },
    {
      "epoch": 366.0,
      "eval_accuracy": 0.7418508804795804,
      "eval_f1": 0.7934032983508246,
      "eval_loss": 0.5516533255577087,
      "eval_precision": 0.7386934673366834,
      "eval_recall": 0.8568652849740933,
      "eval_roc_auc": 0.7909896373056996,
      "eval_runtime": 1.1908,
      "eval_samples_per_second": 2241.381,
      "eval_steps_per_second": 5.039,
      "step": 3843
    },
    {
      "epoch": 366.0952380952381,
      "grad_norm": 5.4201459884643555,
      "learning_rate": 7.202023988005998e-07,
      "loss": 1.1294,
      "step": 3844
    },
    {
      "epoch": 366.1904761904762,
      "grad_norm": 6.1927876472473145,
      "learning_rate": 7.203898050974513e-07,
      "loss": 1.0901,
      "step": 3845
    },
    {
      "epoch": 366.2857142857143,
      "grad_norm": 15.932611465454102,
      "learning_rate": 7.205772113943029e-07,
      "loss": 1.1438,
      "step": 3846
    },
    {
      "epoch": 366.3809523809524,
      "grad_norm": 7.462088108062744,
      "learning_rate": 7.207646176911544e-07,
      "loss": 1.1717,
      "step": 3847
    },
    {
      "epoch": 366.4761904761905,
      "grad_norm": 9.691133499145508,
      "learning_rate": 7.20952023988006e-07,
      "loss": 1.1742,
      "step": 3848
    },
    {
      "epoch": 366.57142857142856,
      "grad_norm": 6.052723407745361,
      "learning_rate": 7.211394302848576e-07,
      "loss": 1.1219,
      "step": 3849
    },
    {
      "epoch": 366.6666666666667,
      "grad_norm": 8.719459533691406,
      "learning_rate": 7.213268365817091e-07,
      "loss": 1.0821,
      "step": 3850
    },
    {
      "epoch": 366.76190476190476,
      "grad_norm": 5.203175067901611,
      "learning_rate": 7.215142428785608e-07,
      "loss": 1.1109,
      "step": 3851
    },
    {
      "epoch": 366.85714285714283,
      "grad_norm": 11.557815551757812,
      "learning_rate": 7.217016491754123e-07,
      "loss": 1.1294,
      "step": 3852
    },
    {
      "epoch": 366.95238095238096,
      "grad_norm": 7.829832077026367,
      "learning_rate": 7.218890554722639e-07,
      "loss": 1.1205,
      "step": 3853
    },
    {
      "epoch": 366.95238095238096,
      "eval_accuracy": 0.7407268639940052,
      "eval_f1": 0.7896656534954407,
      "eval_loss": 0.5511266589164734,
      "eval_precision": 0.7439862542955327,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.7909928036845135,
      "eval_runtime": 1.1232,
      "eval_samples_per_second": 2376.163,
      "eval_steps_per_second": 5.342,
      "step": 3853
    },
    {
      "epoch": 367.04761904761904,
      "grad_norm": 7.368420124053955,
      "learning_rate": 7.220764617691154e-07,
      "loss": 1.103,
      "step": 3854
    },
    {
      "epoch": 367.14285714285717,
      "grad_norm": 7.851127624511719,
      "learning_rate": 7.22263868065967e-07,
      "loss": 1.1219,
      "step": 3855
    },
    {
      "epoch": 367.23809523809524,
      "grad_norm": 5.373974323272705,
      "learning_rate": 7.224512743628187e-07,
      "loss": 1.096,
      "step": 3856
    },
    {
      "epoch": 367.3333333333333,
      "grad_norm": 7.843284606933594,
      "learning_rate": 7.226386806596702e-07,
      "loss": 1.1196,
      "step": 3857
    },
    {
      "epoch": 367.42857142857144,
      "grad_norm": 12.174184799194336,
      "learning_rate": 7.228260869565219e-07,
      "loss": 1.0599,
      "step": 3858
    },
    {
      "epoch": 367.5238095238095,
      "grad_norm": 8.143187522888184,
      "learning_rate": 7.230134932533734e-07,
      "loss": 1.1317,
      "step": 3859
    },
    {
      "epoch": 367.6190476190476,
      "grad_norm": 6.931474208831787,
      "learning_rate": 7.23200899550225e-07,
      "loss": 1.1271,
      "step": 3860
    },
    {
      "epoch": 367.7142857142857,
      "grad_norm": 5.616233825683594,
      "learning_rate": 7.233883058470765e-07,
      "loss": 1.1407,
      "step": 3861
    },
    {
      "epoch": 367.8095238095238,
      "grad_norm": 9.154975891113281,
      "learning_rate": 7.235757121439281e-07,
      "loss": 1.1845,
      "step": 3862
    },
    {
      "epoch": 367.9047619047619,
      "grad_norm": 14.478100776672363,
      "learning_rate": 7.237631184407796e-07,
      "loss": 1.0533,
      "step": 3863
    },
    {
      "epoch": 368.0,
      "grad_norm": 10.775654792785645,
      "learning_rate": 7.239505247376312e-07,
      "loss": 1.1425,
      "step": 3864
    },
    {
      "epoch": 368.0,
      "eval_accuracy": 0.7347321094042712,
      "eval_f1": 0.7820197044334976,
      "eval_loss": 0.5520932674407959,
      "eval_precision": 0.7453051643192489,
      "eval_recall": 0.822538860103627,
      "eval_roc_auc": 0.7896453655728267,
      "eval_runtime": 1.1296,
      "eval_samples_per_second": 2362.879,
      "eval_steps_per_second": 5.312,
      "step": 3864
    },
    {
      "epoch": 368.0952380952381,
      "grad_norm": 13.183884620666504,
      "learning_rate": 7.241379310344829e-07,
      "loss": 1.1199,
      "step": 3865
    },
    {
      "epoch": 368.1904761904762,
      "grad_norm": 9.267333030700684,
      "learning_rate": 7.243253373313344e-07,
      "loss": 1.085,
      "step": 3866
    },
    {
      "epoch": 368.2857142857143,
      "grad_norm": 10.788604736328125,
      "learning_rate": 7.24512743628186e-07,
      "loss": 1.1305,
      "step": 3867
    },
    {
      "epoch": 368.3809523809524,
      "grad_norm": 7.998301029205322,
      "learning_rate": 7.247001499250375e-07,
      "loss": 1.1015,
      "step": 3868
    },
    {
      "epoch": 368.4761904761905,
      "grad_norm": 6.51740837097168,
      "learning_rate": 7.248875562218891e-07,
      "loss": 1.1298,
      "step": 3869
    },
    {
      "epoch": 368.57142857142856,
      "grad_norm": 7.447991847991943,
      "learning_rate": 7.250749625187406e-07,
      "loss": 1.1301,
      "step": 3870
    },
    {
      "epoch": 368.6666666666667,
      "grad_norm": 7.863955974578857,
      "learning_rate": 7.252623688155922e-07,
      "loss": 1.0831,
      "step": 3871
    },
    {
      "epoch": 368.76190476190476,
      "grad_norm": 10.221150398254395,
      "learning_rate": 7.254497751124438e-07,
      "loss": 1.0993,
      "step": 3872
    },
    {
      "epoch": 368.85714285714283,
      "grad_norm": 14.498579978942871,
      "learning_rate": 7.256371814092954e-07,
      "loss": 1.1215,
      "step": 3873
    },
    {
      "epoch": 368.95238095238096,
      "grad_norm": 9.742353439331055,
      "learning_rate": 7.25824587706147e-07,
      "loss": 1.1466,
      "step": 3874
    },
    {
      "epoch": 368.95238095238096,
      "eval_accuracy": 0.7429748969651555,
      "eval_f1": 0.794241151769646,
      "eval_loss": 0.5526332855224609,
      "eval_precision": 0.7396648044692737,
      "eval_recall": 0.8575129533678757,
      "eval_roc_auc": 0.789238341968912,
      "eval_runtime": 1.1202,
      "eval_samples_per_second": 2382.668,
      "eval_steps_per_second": 5.356,
      "step": 3874
    },
    {
      "epoch": 369.04761904761904,
      "grad_norm": 5.24822998046875,
      "learning_rate": 7.260119940029985e-07,
      "loss": 1.1057,
      "step": 3875
    },
    {
      "epoch": 369.14285714285717,
      "grad_norm": 6.557324409484863,
      "learning_rate": 7.261994002998501e-07,
      "loss": 1.1194,
      "step": 3876
    },
    {
      "epoch": 369.23809523809524,
      "grad_norm": 10.006333351135254,
      "learning_rate": 7.263868065967016e-07,
      "loss": 1.0907,
      "step": 3877
    },
    {
      "epoch": 369.3333333333333,
      "grad_norm": 15.545181274414062,
      "learning_rate": 7.265742128935533e-07,
      "loss": 1.1058,
      "step": 3878
    },
    {
      "epoch": 369.42857142857144,
      "grad_norm": 12.693238258361816,
      "learning_rate": 7.267616191904048e-07,
      "loss": 1.1443,
      "step": 3879
    },
    {
      "epoch": 369.5238095238095,
      "grad_norm": 16.02750015258789,
      "learning_rate": 7.269490254872564e-07,
      "loss": 1.099,
      "step": 3880
    },
    {
      "epoch": 369.6190476190476,
      "grad_norm": 14.570963859558105,
      "learning_rate": 7.27136431784108e-07,
      "loss": 1.1059,
      "step": 3881
    },
    {
      "epoch": 369.7142857142857,
      "grad_norm": 17.04131317138672,
      "learning_rate": 7.273238380809596e-07,
      "loss": 1.0468,
      "step": 3882
    },
    {
      "epoch": 369.8095238095238,
      "grad_norm": 7.516332626342773,
      "learning_rate": 7.275112443778112e-07,
      "loss": 1.1619,
      "step": 3883
    },
    {
      "epoch": 369.9047619047619,
      "grad_norm": 7.201144218444824,
      "learning_rate": 7.276986506746627e-07,
      "loss": 1.1273,
      "step": 3884
    },
    {
      "epoch": 370.0,
      "grad_norm": 5.620227336883545,
      "learning_rate": 7.278860569715143e-07,
      "loss": 1.0963,
      "step": 3885
    },
    {
      "epoch": 370.0,
      "eval_accuracy": 0.7388535031847133,
      "eval_f1": 0.7882102704345184,
      "eval_loss": 0.5518689751625061,
      "eval_precision": 0.7424155695477962,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.789812320092113,
      "eval_runtime": 1.1569,
      "eval_samples_per_second": 2306.989,
      "eval_steps_per_second": 5.186,
      "step": 3885
    },
    {
      "epoch": 370.0952380952381,
      "grad_norm": 6.75601863861084,
      "learning_rate": 7.280734632683659e-07,
      "loss": 1.1112,
      "step": 3886
    },
    {
      "epoch": 370.1904761904762,
      "grad_norm": 5.5591020584106445,
      "learning_rate": 7.282608695652175e-07,
      "loss": 1.0847,
      "step": 3887
    },
    {
      "epoch": 370.2857142857143,
      "grad_norm": 9.115145683288574,
      "learning_rate": 7.28448275862069e-07,
      "loss": 1.1349,
      "step": 3888
    },
    {
      "epoch": 370.3809523809524,
      "grad_norm": 5.112926959991455,
      "learning_rate": 7.286356821589206e-07,
      "loss": 1.1822,
      "step": 3889
    },
    {
      "epoch": 370.4761904761905,
      "grad_norm": 6.981180191040039,
      "learning_rate": 7.288230884557721e-07,
      "loss": 1.1226,
      "step": 3890
    },
    {
      "epoch": 370.57142857142856,
      "grad_norm": 14.1383695602417,
      "learning_rate": 7.290104947526237e-07,
      "loss": 1.1291,
      "step": 3891
    },
    {
      "epoch": 370.6666666666667,
      "grad_norm": 16.603281021118164,
      "learning_rate": 7.291979010494752e-07,
      "loss": 1.0763,
      "step": 3892
    },
    {
      "epoch": 370.76190476190476,
      "grad_norm": 6.563464641571045,
      "learning_rate": 7.293853073463269e-07,
      "loss": 1.1482,
      "step": 3893
    },
    {
      "epoch": 370.85714285714283,
      "grad_norm": 13.651692390441895,
      "learning_rate": 7.295727136431785e-07,
      "loss": 1.104,
      "step": 3894
    },
    {
      "epoch": 370.95238095238096,
      "grad_norm": 5.060080528259277,
      "learning_rate": 7.2976011994003e-07,
      "loss": 1.1399,
      "step": 3895
    },
    {
      "epoch": 370.95238095238096,
      "eval_accuracy": 0.7381041588609967,
      "eval_f1": 0.7893943959023803,
      "eval_loss": 0.5518258810043335,
      "eval_precision": 0.7380281690140845,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7901482440990213,
      "eval_runtime": 1.1343,
      "eval_samples_per_second": 2352.985,
      "eval_steps_per_second": 5.29,
      "step": 3895
    },
    {
      "epoch": 371.04761904761904,
      "grad_norm": 6.809076309204102,
      "learning_rate": 7.299475262368816e-07,
      "loss": 1.1489,
      "step": 3896
    },
    {
      "epoch": 371.14285714285717,
      "grad_norm": 6.76948356628418,
      "learning_rate": 7.301349325337331e-07,
      "loss": 1.0803,
      "step": 3897
    },
    {
      "epoch": 371.23809523809524,
      "grad_norm": 5.986545562744141,
      "learning_rate": 7.303223388305847e-07,
      "loss": 1.1069,
      "step": 3898
    },
    {
      "epoch": 371.3333333333333,
      "grad_norm": 11.000617980957031,
      "learning_rate": 7.305097451274362e-07,
      "loss": 1.1055,
      "step": 3899
    },
    {
      "epoch": 371.42857142857144,
      "grad_norm": 8.57867431640625,
      "learning_rate": 7.306971514242879e-07,
      "loss": 1.096,
      "step": 3900
    },
    {
      "epoch": 371.5238095238095,
      "grad_norm": 9.18748950958252,
      "learning_rate": 7.308845577211395e-07,
      "loss": 1.1128,
      "step": 3901
    },
    {
      "epoch": 371.6190476190476,
      "grad_norm": 10.393960952758789,
      "learning_rate": 7.31071964017991e-07,
      "loss": 1.1361,
      "step": 3902
    },
    {
      "epoch": 371.7142857142857,
      "grad_norm": 10.242207527160645,
      "learning_rate": 7.312593703148427e-07,
      "loss": 1.1515,
      "step": 3903
    },
    {
      "epoch": 371.8095238095238,
      "grad_norm": 6.255610466003418,
      "learning_rate": 7.314467766116942e-07,
      "loss": 1.0784,
      "step": 3904
    },
    {
      "epoch": 371.9047619047619,
      "grad_norm": 8.978156089782715,
      "learning_rate": 7.316341829085458e-07,
      "loss": 1.158,
      "step": 3905
    },
    {
      "epoch": 372.0,
      "grad_norm": 11.08545970916748,
      "learning_rate": 7.318215892053973e-07,
      "loss": 1.0757,
      "step": 3906
    },
    {
      "epoch": 372.0,
      "eval_accuracy": 0.7336080929186961,
      "eval_f1": 0.7808936825885978,
      "eval_loss": 0.5521113276481628,
      "eval_precision": 0.7448559670781894,
      "eval_recall": 0.8205958549222798,
      "eval_roc_auc": 0.7896076568796776,
      "eval_runtime": 1.1181,
      "eval_samples_per_second": 2387.001,
      "eval_steps_per_second": 5.366,
      "step": 3906
    },
    {
      "epoch": 372.0952380952381,
      "grad_norm": 5.756834030151367,
      "learning_rate": 7.32008995502249e-07,
      "loss": 1.1166,
      "step": 3907
    },
    {
      "epoch": 372.1904761904762,
      "grad_norm": 6.470827102661133,
      "learning_rate": 7.321964017991005e-07,
      "loss": 1.1019,
      "step": 3908
    },
    {
      "epoch": 372.2857142857143,
      "grad_norm": 9.52391242980957,
      "learning_rate": 7.323838080959521e-07,
      "loss": 1.1114,
      "step": 3909
    },
    {
      "epoch": 372.3809523809524,
      "grad_norm": 17.73433494567871,
      "learning_rate": 7.325712143928037e-07,
      "loss": 1.1026,
      "step": 3910
    },
    {
      "epoch": 372.4761904761905,
      "grad_norm": 6.577383041381836,
      "learning_rate": 7.327586206896552e-07,
      "loss": 1.0955,
      "step": 3911
    },
    {
      "epoch": 372.57142857142856,
      "grad_norm": 8.080799102783203,
      "learning_rate": 7.329460269865068e-07,
      "loss": 1.1049,
      "step": 3912
    },
    {
      "epoch": 372.6666666666667,
      "grad_norm": 16.410451889038086,
      "learning_rate": 7.331334332833583e-07,
      "loss": 1.1505,
      "step": 3913
    },
    {
      "epoch": 372.76190476190476,
      "grad_norm": 11.473193168640137,
      "learning_rate": 7.3332083958021e-07,
      "loss": 1.1487,
      "step": 3914
    },
    {
      "epoch": 372.85714285714283,
      "grad_norm": 6.292839050292969,
      "learning_rate": 7.335082458770615e-07,
      "loss": 1.0933,
      "step": 3915
    },
    {
      "epoch": 372.95238095238096,
      "grad_norm": 11.717289924621582,
      "learning_rate": 7.336956521739131e-07,
      "loss": 1.1058,
      "step": 3916
    },
    {
      "epoch": 372.95238095238096,
      "eval_accuracy": 0.7403521918321468,
      "eval_f1": 0.7896813353566009,
      "eval_loss": 0.5519756078720093,
      "eval_precision": 0.7430039977155911,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7897651122625216,
      "eval_runtime": 1.1323,
      "eval_samples_per_second": 2357.052,
      "eval_steps_per_second": 5.299,
      "step": 3916
    },
    {
      "epoch": 373.04761904761904,
      "grad_norm": 9.770853996276855,
      "learning_rate": 7.338830584707646e-07,
      "loss": 1.1188,
      "step": 3917
    },
    {
      "epoch": 373.14285714285717,
      "grad_norm": 13.120342254638672,
      "learning_rate": 7.340704647676162e-07,
      "loss": 1.104,
      "step": 3918
    },
    {
      "epoch": 373.23809523809524,
      "grad_norm": 5.9968461990356445,
      "learning_rate": 7.342578710644677e-07,
      "loss": 1.1355,
      "step": 3919
    },
    {
      "epoch": 373.3333333333333,
      "grad_norm": 6.1175127029418945,
      "learning_rate": 7.344452773613193e-07,
      "loss": 1.1082,
      "step": 3920
    },
    {
      "epoch": 373.42857142857144,
      "grad_norm": 11.14828109741211,
      "learning_rate": 7.34632683658171e-07,
      "loss": 1.1189,
      "step": 3921
    },
    {
      "epoch": 373.5238095238095,
      "grad_norm": 7.950910568237305,
      "learning_rate": 7.348200899550225e-07,
      "loss": 1.1224,
      "step": 3922
    },
    {
      "epoch": 373.6190476190476,
      "grad_norm": 6.772651672363281,
      "learning_rate": 7.350074962518741e-07,
      "loss": 1.1048,
      "step": 3923
    },
    {
      "epoch": 373.7142857142857,
      "grad_norm": 8.026163101196289,
      "learning_rate": 7.351949025487256e-07,
      "loss": 1.1246,
      "step": 3924
    },
    {
      "epoch": 373.8095238095238,
      "grad_norm": 5.476925849914551,
      "learning_rate": 7.353823088455773e-07,
      "loss": 1.0943,
      "step": 3925
    },
    {
      "epoch": 373.9047619047619,
      "grad_norm": 10.513056755065918,
      "learning_rate": 7.355697151424288e-07,
      "loss": 1.1546,
      "step": 3926
    },
    {
      "epoch": 374.0,
      "grad_norm": 6.3184051513671875,
      "learning_rate": 7.357571214392804e-07,
      "loss": 1.0826,
      "step": 3927
    },
    {
      "epoch": 374.0,
      "eval_accuracy": 0.738478831022855,
      "eval_f1": 0.7883565797453002,
      "eval_loss": 0.5509572625160217,
      "eval_precision": 0.7411630558722919,
      "eval_recall": 0.8419689119170984,
      "eval_roc_auc": 0.7910938399539436,
      "eval_runtime": 1.128,
      "eval_samples_per_second": 2366.157,
      "eval_steps_per_second": 5.319,
      "step": 3927
    },
    {
      "epoch": 374.0952380952381,
      "grad_norm": 11.123753547668457,
      "learning_rate": 7.359445277361321e-07,
      "loss": 1.1463,
      "step": 3928
    },
    {
      "epoch": 374.1904761904762,
      "grad_norm": 6.112797737121582,
      "learning_rate": 7.361319340329836e-07,
      "loss": 1.0887,
      "step": 3929
    },
    {
      "epoch": 374.2857142857143,
      "grad_norm": 4.011114597320557,
      "learning_rate": 7.363193403298352e-07,
      "loss": 1.1321,
      "step": 3930
    },
    {
      "epoch": 374.3809523809524,
      "grad_norm": 6.350623607635498,
      "learning_rate": 7.365067466266867e-07,
      "loss": 1.1348,
      "step": 3931
    },
    {
      "epoch": 374.4761904761905,
      "grad_norm": 9.323946952819824,
      "learning_rate": 7.366941529235383e-07,
      "loss": 1.0847,
      "step": 3932
    },
    {
      "epoch": 374.57142857142856,
      "grad_norm": 5.713606834411621,
      "learning_rate": 7.368815592203898e-07,
      "loss": 1.1499,
      "step": 3933
    },
    {
      "epoch": 374.6666666666667,
      "grad_norm": 9.828307151794434,
      "learning_rate": 7.370689655172414e-07,
      "loss": 1.1031,
      "step": 3934
    },
    {
      "epoch": 374.76190476190476,
      "grad_norm": 12.144657135009766,
      "learning_rate": 7.37256371814093e-07,
      "loss": 1.1269,
      "step": 3935
    },
    {
      "epoch": 374.85714285714283,
      "grad_norm": 7.657914161682129,
      "learning_rate": 7.374437781109446e-07,
      "loss": 1.1093,
      "step": 3936
    },
    {
      "epoch": 374.95238095238096,
      "grad_norm": 9.174505233764648,
      "learning_rate": 7.376311844077962e-07,
      "loss": 1.0883,
      "step": 3937
    },
    {
      "epoch": 374.95238095238096,
      "eval_accuracy": 0.7418508804795804,
      "eval_f1": 0.7916540671303296,
      "eval_loss": 0.5501177310943604,
      "eval_precision": 0.742484401588202,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7922084052964882,
      "eval_runtime": 1.1426,
      "eval_samples_per_second": 2335.804,
      "eval_steps_per_second": 5.251,
      "step": 3937
    },
    {
      "epoch": 375.04761904761904,
      "grad_norm": 11.815903663635254,
      "learning_rate": 7.378185907046477e-07,
      "loss": 1.1139,
      "step": 3938
    },
    {
      "epoch": 375.14285714285717,
      "grad_norm": 7.327791690826416,
      "learning_rate": 7.380059970014993e-07,
      "loss": 1.1217,
      "step": 3939
    },
    {
      "epoch": 375.23809523809524,
      "grad_norm": 8.6361083984375,
      "learning_rate": 7.381934032983508e-07,
      "loss": 1.1498,
      "step": 3940
    },
    {
      "epoch": 375.3333333333333,
      "grad_norm": 9.45090389251709,
      "learning_rate": 7.383808095952024e-07,
      "loss": 1.1283,
      "step": 3941
    },
    {
      "epoch": 375.42857142857144,
      "grad_norm": 8.466257095336914,
      "learning_rate": 7.38568215892054e-07,
      "loss": 1.1063,
      "step": 3942
    },
    {
      "epoch": 375.5238095238095,
      "grad_norm": 5.573893070220947,
      "learning_rate": 7.387556221889056e-07,
      "loss": 1.0902,
      "step": 3943
    },
    {
      "epoch": 375.6190476190476,
      "grad_norm": 5.684384346008301,
      "learning_rate": 7.389430284857571e-07,
      "loss": 1.1213,
      "step": 3944
    },
    {
      "epoch": 375.7142857142857,
      "grad_norm": 4.658603191375732,
      "learning_rate": 7.391304347826087e-07,
      "loss": 1.1028,
      "step": 3945
    },
    {
      "epoch": 375.8095238095238,
      "grad_norm": 8.962539672851562,
      "learning_rate": 7.393178410794602e-07,
      "loss": 1.1042,
      "step": 3946
    },
    {
      "epoch": 375.9047619047619,
      "grad_norm": 7.740326881408691,
      "learning_rate": 7.395052473763118e-07,
      "loss": 1.1129,
      "step": 3947
    },
    {
      "epoch": 376.0,
      "grad_norm": 6.349332332611084,
      "learning_rate": 7.396926536731635e-07,
      "loss": 1.09,
      "step": 3948
    },
    {
      "epoch": 376.0,
      "eval_accuracy": 0.7418508804795804,
      "eval_f1": 0.792156862745098,
      "eval_loss": 0.5506932735443115,
      "eval_precision": 0.7413890457368718,
      "eval_recall": 0.8503886010362695,
      "eval_roc_auc": 0.7915322394933794,
      "eval_runtime": 1.1262,
      "eval_samples_per_second": 2369.989,
      "eval_steps_per_second": 5.328,
      "step": 3948
    },
    {
      "epoch": 376.0952380952381,
      "grad_norm": 5.533696174621582,
      "learning_rate": 7.39880059970015e-07,
      "loss": 1.1238,
      "step": 3949
    },
    {
      "epoch": 376.1904761904762,
      "grad_norm": 9.323712348937988,
      "learning_rate": 7.400674662668667e-07,
      "loss": 1.1281,
      "step": 3950
    },
    {
      "epoch": 376.2857142857143,
      "grad_norm": 13.140124320983887,
      "learning_rate": 7.402548725637182e-07,
      "loss": 1.0889,
      "step": 3951
    },
    {
      "epoch": 376.3809523809524,
      "grad_norm": 5.626778602600098,
      "learning_rate": 7.404422788605698e-07,
      "loss": 1.1441,
      "step": 3952
    },
    {
      "epoch": 376.4761904761905,
      "grad_norm": 10.561690330505371,
      "learning_rate": 7.406296851574213e-07,
      "loss": 1.0828,
      "step": 3953
    },
    {
      "epoch": 376.57142857142856,
      "grad_norm": 5.1880621910095215,
      "learning_rate": 7.408170914542729e-07,
      "loss": 1.1144,
      "step": 3954
    },
    {
      "epoch": 376.6666666666667,
      "grad_norm": 10.816292762756348,
      "learning_rate": 7.410044977511245e-07,
      "loss": 1.1355,
      "step": 3955
    },
    {
      "epoch": 376.76190476190476,
      "grad_norm": 11.029444694519043,
      "learning_rate": 7.411919040479761e-07,
      "loss": 1.1215,
      "step": 3956
    },
    {
      "epoch": 376.85714285714283,
      "grad_norm": 5.887256145477295,
      "learning_rate": 7.413793103448277e-07,
      "loss": 1.1361,
      "step": 3957
    },
    {
      "epoch": 376.95238095238096,
      "grad_norm": 8.500890731811523,
      "learning_rate": 7.415667166416792e-07,
      "loss": 1.1084,
      "step": 3958
    },
    {
      "epoch": 376.95238095238096,
      "eval_accuracy": 0.7422255526414387,
      "eval_f1": 0.7920193470374849,
      "eval_loss": 0.5507376790046692,
      "eval_precision": 0.7426303854875284,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7916801957397812,
      "eval_runtime": 1.134,
      "eval_samples_per_second": 2353.674,
      "eval_steps_per_second": 5.291,
      "step": 3958
    },
    {
      "epoch": 377.04761904761904,
      "grad_norm": 7.278326988220215,
      "learning_rate": 7.417541229385308e-07,
      "loss": 1.1019,
      "step": 3959
    },
    {
      "epoch": 377.14285714285717,
      "grad_norm": 6.423912525177002,
      "learning_rate": 7.419415292353823e-07,
      "loss": 1.1087,
      "step": 3960
    },
    {
      "epoch": 377.23809523809524,
      "grad_norm": 11.042447090148926,
      "learning_rate": 7.421289355322339e-07,
      "loss": 1.1223,
      "step": 3961
    },
    {
      "epoch": 377.3333333333333,
      "grad_norm": 14.736066818237305,
      "learning_rate": 7.423163418290854e-07,
      "loss": 1.099,
      "step": 3962
    },
    {
      "epoch": 377.42857142857144,
      "grad_norm": 8.444828987121582,
      "learning_rate": 7.425037481259371e-07,
      "loss": 1.1219,
      "step": 3963
    },
    {
      "epoch": 377.5238095238095,
      "grad_norm": 8.200831413269043,
      "learning_rate": 7.426911544227887e-07,
      "loss": 1.1237,
      "step": 3964
    },
    {
      "epoch": 377.6190476190476,
      "grad_norm": 12.050019264221191,
      "learning_rate": 7.428785607196402e-07,
      "loss": 1.0873,
      "step": 3965
    },
    {
      "epoch": 377.7142857142857,
      "grad_norm": 6.6788177490234375,
      "learning_rate": 7.430659670164918e-07,
      "loss": 1.108,
      "step": 3966
    },
    {
      "epoch": 377.8095238095238,
      "grad_norm": 10.5377779006958,
      "learning_rate": 7.432533733133433e-07,
      "loss": 1.1078,
      "step": 3967
    },
    {
      "epoch": 377.9047619047619,
      "grad_norm": 7.445323944091797,
      "learning_rate": 7.434407796101949e-07,
      "loss": 1.1545,
      "step": 3968
    },
    {
      "epoch": 378.0,
      "grad_norm": 9.764788627624512,
      "learning_rate": 7.436281859070464e-07,
      "loss": 1.1151,
      "step": 3969
    },
    {
      "epoch": 378.0,
      "eval_accuracy": 0.7392281753465717,
      "eval_f1": 0.7878048780487805,
      "eval_loss": 0.5502727627754211,
      "eval_precision": 0.7442396313364056,
      "eval_recall": 0.8367875647668394,
      "eval_roc_auc": 0.7921421991940126,
      "eval_runtime": 1.1304,
      "eval_samples_per_second": 2361.214,
      "eval_steps_per_second": 5.308,
      "step": 3969
    },
    {
      "epoch": 378.0952380952381,
      "grad_norm": 14.01278305053711,
      "learning_rate": 7.438155922038981e-07,
      "loss": 1.1209,
      "step": 3970
    },
    {
      "epoch": 378.1904761904762,
      "grad_norm": 5.303046226501465,
      "learning_rate": 7.440029985007496e-07,
      "loss": 1.1224,
      "step": 3971
    },
    {
      "epoch": 378.2857142857143,
      "grad_norm": 6.776137351989746,
      "learning_rate": 7.441904047976012e-07,
      "loss": 1.0761,
      "step": 3972
    },
    {
      "epoch": 378.3809523809524,
      "grad_norm": 5.707252025604248,
      "learning_rate": 7.443778110944529e-07,
      "loss": 1.1331,
      "step": 3973
    },
    {
      "epoch": 378.4761904761905,
      "grad_norm": 15.999380111694336,
      "learning_rate": 7.445652173913044e-07,
      "loss": 1.0887,
      "step": 3974
    },
    {
      "epoch": 378.57142857142856,
      "grad_norm": 11.217973709106445,
      "learning_rate": 7.44752623688156e-07,
      "loss": 1.1119,
      "step": 3975
    },
    {
      "epoch": 378.6666666666667,
      "grad_norm": 4.672736167907715,
      "learning_rate": 7.449400299850075e-07,
      "loss": 1.1536,
      "step": 3976
    },
    {
      "epoch": 378.76190476190476,
      "grad_norm": 12.972167015075684,
      "learning_rate": 7.451274362818592e-07,
      "loss": 1.1258,
      "step": 3977
    },
    {
      "epoch": 378.85714285714283,
      "grad_norm": 5.687925338745117,
      "learning_rate": 7.453148425787107e-07,
      "loss": 1.0886,
      "step": 3978
    },
    {
      "epoch": 378.95238095238096,
      "grad_norm": 6.426712989807129,
      "learning_rate": 7.455022488755623e-07,
      "loss": 1.0961,
      "step": 3979
    },
    {
      "epoch": 378.95238095238096,
      "eval_accuracy": 0.7399775196702885,
      "eval_f1": 0.7904589371980676,
      "eval_loss": 0.5504778623580933,
      "eval_precision": 0.7403846153846154,
      "eval_recall": 0.8477979274611399,
      "eval_roc_auc": 0.7919576856649396,
      "eval_runtime": 1.1251,
      "eval_samples_per_second": 2372.226,
      "eval_steps_per_second": 5.333,
      "step": 3979
    },
    {
      "epoch": 379.04761904761904,
      "grad_norm": 11.194128036499023,
      "learning_rate": 7.456896551724138e-07,
      "loss": 1.0735,
      "step": 3980
    },
    {
      "epoch": 379.14285714285717,
      "grad_norm": 8.991393089294434,
      "learning_rate": 7.458770614692654e-07,
      "loss": 1.1403,
      "step": 3981
    },
    {
      "epoch": 379.23809523809524,
      "grad_norm": 9.715619087219238,
      "learning_rate": 7.46064467766117e-07,
      "loss": 1.1153,
      "step": 3982
    },
    {
      "epoch": 379.3333333333333,
      "grad_norm": 11.953075408935547,
      "learning_rate": 7.462518740629685e-07,
      "loss": 1.1342,
      "step": 3983
    },
    {
      "epoch": 379.42857142857144,
      "grad_norm": 6.715610027313232,
      "learning_rate": 7.464392803598202e-07,
      "loss": 1.1291,
      "step": 3984
    },
    {
      "epoch": 379.5238095238095,
      "grad_norm": 7.38981294631958,
      "learning_rate": 7.466266866566717e-07,
      "loss": 1.0811,
      "step": 3985
    },
    {
      "epoch": 379.6190476190476,
      "grad_norm": 7.233059406280518,
      "learning_rate": 7.468140929535233e-07,
      "loss": 1.0664,
      "step": 3986
    },
    {
      "epoch": 379.7142857142857,
      "grad_norm": 5.6119561195373535,
      "learning_rate": 7.470014992503748e-07,
      "loss": 1.1032,
      "step": 3987
    },
    {
      "epoch": 379.8095238095238,
      "grad_norm": 12.48303508758545,
      "learning_rate": 7.471889055472264e-07,
      "loss": 1.0959,
      "step": 3988
    },
    {
      "epoch": 379.9047619047619,
      "grad_norm": 6.563333988189697,
      "learning_rate": 7.473763118440779e-07,
      "loss": 1.1254,
      "step": 3989
    },
    {
      "epoch": 380.0,
      "grad_norm": 8.098648071289062,
      "learning_rate": 7.475637181409295e-07,
      "loss": 1.136,
      "step": 3990
    },
    {
      "epoch": 380.0,
      "eval_accuracy": 0.7381041588609967,
      "eval_f1": 0.7887579329102448,
      "eval_loss": 0.5500577688217163,
      "eval_precision": 0.7393767705382436,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7920089234312033,
      "eval_runtime": 1.1123,
      "eval_samples_per_second": 2399.447,
      "eval_steps_per_second": 5.394,
      "step": 3990
    },
    {
      "epoch": 380.0952380952381,
      "grad_norm": 6.370793342590332,
      "learning_rate": 7.477511244377812e-07,
      "loss": 1.107,
      "step": 3991
    },
    {
      "epoch": 380.1904761904762,
      "grad_norm": 5.635412693023682,
      "learning_rate": 7.479385307346327e-07,
      "loss": 1.098,
      "step": 3992
    },
    {
      "epoch": 380.2857142857143,
      "grad_norm": 9.10843563079834,
      "learning_rate": 7.481259370314843e-07,
      "loss": 1.1219,
      "step": 3993
    },
    {
      "epoch": 380.3809523809524,
      "grad_norm": 8.240547180175781,
      "learning_rate": 7.483133433283358e-07,
      "loss": 1.1008,
      "step": 3994
    },
    {
      "epoch": 380.4761904761905,
      "grad_norm": 8.950239181518555,
      "learning_rate": 7.485007496251875e-07,
      "loss": 1.1599,
      "step": 3995
    },
    {
      "epoch": 380.57142857142856,
      "grad_norm": 5.558511257171631,
      "learning_rate": 7.48688155922039e-07,
      "loss": 1.1444,
      "step": 3996
    },
    {
      "epoch": 380.6666666666667,
      "grad_norm": 6.004672050476074,
      "learning_rate": 7.488755622188906e-07,
      "loss": 1.1108,
      "step": 3997
    },
    {
      "epoch": 380.76190476190476,
      "grad_norm": 6.5706963539123535,
      "learning_rate": 7.490629685157422e-07,
      "loss": 1.1451,
      "step": 3998
    },
    {
      "epoch": 380.85714285714283,
      "grad_norm": 4.188346862792969,
      "learning_rate": 7.492503748125938e-07,
      "loss": 1.1113,
      "step": 3999
    },
    {
      "epoch": 380.95238095238096,
      "grad_norm": 17.704849243164062,
      "learning_rate": 7.494377811094454e-07,
      "loss": 1.0749,
      "step": 4000
    },
    {
      "epoch": 380.95238095238096,
      "eval_accuracy": 0.7358561258898464,
      "eval_f1": 0.7808517252098228,
      "eval_loss": 0.5495111346244812,
      "eval_precision": 0.7507471607890018,
      "eval_recall": 0.8134715025906736,
      "eval_roc_auc": 0.7926370178468625,
      "eval_runtime": 1.1209,
      "eval_samples_per_second": 2381.206,
      "eval_steps_per_second": 5.353,
      "step": 4000
    },
    {
      "epoch": 381.04761904761904,
      "grad_norm": 6.359234809875488,
      "learning_rate": 7.496251874062969e-07,
      "loss": 1.0951,
      "step": 4001
    },
    {
      "epoch": 381.14285714285717,
      "grad_norm": 15.279829025268555,
      "learning_rate": 7.498125937031485e-07,
      "loss": 1.122,
      "step": 4002
    },
    {
      "epoch": 381.23809523809524,
      "grad_norm": 4.818670749664307,
      "learning_rate": 7.5e-07,
      "loss": 1.1443,
      "step": 4003
    },
    {
      "epoch": 381.3333333333333,
      "grad_norm": 5.658208847045898,
      "learning_rate": 7.501874062968516e-07,
      "loss": 1.1217,
      "step": 4004
    },
    {
      "epoch": 381.42857142857144,
      "grad_norm": 14.737176895141602,
      "learning_rate": 7.503748125937032e-07,
      "loss": 1.1419,
      "step": 4005
    },
    {
      "epoch": 381.5238095238095,
      "grad_norm": 5.5258965492248535,
      "learning_rate": 7.505622188905548e-07,
      "loss": 1.1137,
      "step": 4006
    },
    {
      "epoch": 381.6190476190476,
      "grad_norm": 9.513168334960938,
      "learning_rate": 7.507496251874063e-07,
      "loss": 1.1272,
      "step": 4007
    },
    {
      "epoch": 381.7142857142857,
      "grad_norm": 10.469417572021484,
      "learning_rate": 7.509370314842579e-07,
      "loss": 1.0777,
      "step": 4008
    },
    {
      "epoch": 381.8095238095238,
      "grad_norm": 8.181276321411133,
      "learning_rate": 7.511244377811095e-07,
      "loss": 1.0905,
      "step": 4009
    },
    {
      "epoch": 381.9047619047619,
      "grad_norm": 6.987445831298828,
      "learning_rate": 7.51311844077961e-07,
      "loss": 1.0617,
      "step": 4010
    },
    {
      "epoch": 382.0,
      "grad_norm": 21.09636116027832,
      "learning_rate": 7.514992503748126e-07,
      "loss": 1.1038,
      "step": 4011
    },
    {
      "epoch": 382.0,
      "eval_accuracy": 0.7452229299363057,
      "eval_f1": 0.7980997624703088,
      "eval_loss": 0.5505495667457581,
      "eval_precision": 0.7368421052631579,
      "eval_recall": 0.8704663212435233,
      "eval_roc_auc": 0.7933411053540588,
      "eval_runtime": 1.1316,
      "eval_samples_per_second": 2358.552,
      "eval_steps_per_second": 5.302,
      "step": 4011
    },
    {
      "epoch": 382.0952380952381,
      "grad_norm": Infinity,
      "learning_rate": 7.514992503748126e-07,
      "loss": 1.1442,
      "step": 4012
    },
    {
      "epoch": 382.1904761904762,
      "grad_norm": 13.094733238220215,
      "learning_rate": 7.516866566716642e-07,
      "loss": 1.0918,
      "step": 4013
    },
    {
      "epoch": 382.2857142857143,
      "grad_norm": 23.418006896972656,
      "learning_rate": 7.518740629685158e-07,
      "loss": 1.2109,
      "step": 4014
    },
    {
      "epoch": 382.3809523809524,
      "grad_norm": 15.063214302062988,
      "learning_rate": 7.520614692653673e-07,
      "loss": 1.1248,
      "step": 4015
    },
    {
      "epoch": 382.4761904761905,
      "grad_norm": 10.05310344696045,
      "learning_rate": 7.522488755622189e-07,
      "loss": 1.0842,
      "step": 4016
    },
    {
      "epoch": 382.57142857142856,
      "grad_norm": 4.695971965789795,
      "learning_rate": 7.524362818590704e-07,
      "loss": 1.1224,
      "step": 4017
    },
    {
      "epoch": 382.6666666666667,
      "grad_norm": 9.64921760559082,
      "learning_rate": 7.526236881559221e-07,
      "loss": 1.0769,
      "step": 4018
    },
    {
      "epoch": 382.76190476190476,
      "grad_norm": 10.011499404907227,
      "learning_rate": 7.528110944527737e-07,
      "loss": 1.0853,
      "step": 4019
    },
    {
      "epoch": 382.85714285714283,
      "grad_norm": 6.553365230560303,
      "learning_rate": 7.529985007496252e-07,
      "loss": 1.1405,
      "step": 4020
    },
    {
      "epoch": 382.95238095238096,
      "grad_norm": 13.643135070800781,
      "learning_rate": 7.531859070464769e-07,
      "loss": 1.1103,
      "step": 4021
    },
    {
      "epoch": 382.95238095238096,
      "eval_accuracy": 0.7366054702135632,
      "eval_f1": 0.7828236021007106,
      "eval_loss": 0.549675464630127,
      "eval_precision": 0.748375664500886,
      "eval_recall": 0.8205958549222798,
      "eval_roc_auc": 0.7923963730569947,
      "eval_runtime": 1.1358,
      "eval_samples_per_second": 2349.91,
      "eval_steps_per_second": 5.283,
      "step": 4021
    },
    {
      "epoch": 383.04761904761904,
      "grad_norm": 6.7152018547058105,
      "learning_rate": 7.533733133433284e-07,
      "loss": 1.182,
      "step": 4022
    },
    {
      "epoch": 383.14285714285717,
      "grad_norm": 4.382837772369385,
      "learning_rate": 7.5356071964018e-07,
      "loss": 1.1322,
      "step": 4023
    },
    {
      "epoch": 383.23809523809524,
      "grad_norm": 11.175541877746582,
      "learning_rate": 7.537481259370315e-07,
      "loss": 1.0832,
      "step": 4024
    },
    {
      "epoch": 383.3333333333333,
      "grad_norm": 13.932941436767578,
      "learning_rate": 7.539355322338831e-07,
      "loss": 1.1331,
      "step": 4025
    },
    {
      "epoch": 383.42857142857144,
      "grad_norm": 6.4109015464782715,
      "learning_rate": 7.541229385307346e-07,
      "loss": 1.1014,
      "step": 4026
    },
    {
      "epoch": 383.5238095238095,
      "grad_norm": 4.632279872894287,
      "learning_rate": 7.543103448275863e-07,
      "loss": 1.1187,
      "step": 4027
    },
    {
      "epoch": 383.6190476190476,
      "grad_norm": 7.028567790985107,
      "learning_rate": 7.544977511244379e-07,
      "loss": 1.0876,
      "step": 4028
    },
    {
      "epoch": 383.7142857142857,
      "grad_norm": 13.858692169189453,
      "learning_rate": 7.546851574212894e-07,
      "loss": 1.1304,
      "step": 4029
    },
    {
      "epoch": 383.8095238095238,
      "grad_norm": 20.205524444580078,
      "learning_rate": 7.54872563718141e-07,
      "loss": 1.1553,
      "step": 4030
    },
    {
      "epoch": 383.9047619047619,
      "grad_norm": 9.382784843444824,
      "learning_rate": 7.550599700149925e-07,
      "loss": 1.1131,
      "step": 4031
    },
    {
      "epoch": 384.0,
      "grad_norm": 4.756422996520996,
      "learning_rate": 7.552473763118441e-07,
      "loss": 1.1189,
      "step": 4032
    },
    {
      "epoch": 384.0,
      "eval_accuracy": 0.7377294866991383,
      "eval_f1": 0.7863247863247863,
      "eval_loss": 0.5493919849395752,
      "eval_precision": 0.74364896073903,
      "eval_recall": 0.8341968911917098,
      "eval_roc_auc": 0.7924628670120898,
      "eval_runtime": 1.1398,
      "eval_samples_per_second": 2341.643,
      "eval_steps_per_second": 5.264,
      "step": 4032
    },
    {
      "epoch": 384.0952380952381,
      "grad_norm": 5.865123271942139,
      "learning_rate": 7.554347826086956e-07,
      "loss": 1.1313,
      "step": 4033
    },
    {
      "epoch": 384.1904761904762,
      "grad_norm": 5.505659580230713,
      "learning_rate": 7.556221889055473e-07,
      "loss": 1.1503,
      "step": 4034
    },
    {
      "epoch": 384.2857142857143,
      "grad_norm": 7.910295009613037,
      "learning_rate": 7.558095952023988e-07,
      "loss": 1.1272,
      "step": 4035
    },
    {
      "epoch": 384.3809523809524,
      "grad_norm": 16.181732177734375,
      "learning_rate": 7.559970014992504e-07,
      "loss": 1.112,
      "step": 4036
    },
    {
      "epoch": 384.4761904761905,
      "grad_norm": 21.702638626098633,
      "learning_rate": 7.56184407796102e-07,
      "loss": 1.0762,
      "step": 4037
    },
    {
      "epoch": 384.57142857142856,
      "grad_norm": 15.46338939666748,
      "learning_rate": 7.563718140929535e-07,
      "loss": 1.1196,
      "step": 4038
    },
    {
      "epoch": 384.6666666666667,
      "grad_norm": 5.850086212158203,
      "learning_rate": 7.565592203898051e-07,
      "loss": 1.1457,
      "step": 4039
    },
    {
      "epoch": 384.76190476190476,
      "grad_norm": 10.61465835571289,
      "learning_rate": 7.567466266866566e-07,
      "loss": 1.0985,
      "step": 4040
    },
    {
      "epoch": 384.85714285714283,
      "grad_norm": 9.136896133422852,
      "learning_rate": 7.569340329835083e-07,
      "loss": 1.0866,
      "step": 4041
    },
    {
      "epoch": 384.95238095238096,
      "grad_norm": 10.092803955078125,
      "learning_rate": 7.571214392803598e-07,
      "loss": 1.1282,
      "step": 4042
    },
    {
      "epoch": 384.95238095238096,
      "eval_accuracy": 0.7426002248032971,
      "eval_f1": 0.7913756453082296,
      "eval_loss": 0.5494597554206848,
      "eval_precision": 0.7449971412235563,
      "eval_recall": 0.8439119170984456,
      "eval_roc_auc": 0.7925371329879102,
      "eval_runtime": 1.1433,
      "eval_samples_per_second": 2334.451,
      "eval_steps_per_second": 5.248,
      "step": 4042
    },
    {
      "epoch": 385.04761904761904,
      "grad_norm": 6.845860481262207,
      "learning_rate": 7.573088455772114e-07,
      "loss": 1.1061,
      "step": 4043
    },
    {
      "epoch": 385.14285714285717,
      "grad_norm": 14.234851837158203,
      "learning_rate": 7.57496251874063e-07,
      "loss": 1.1518,
      "step": 4044
    },
    {
      "epoch": 385.23809523809524,
      "grad_norm": 12.365203857421875,
      "learning_rate": 7.576836581709146e-07,
      "loss": 1.0977,
      "step": 4045
    },
    {
      "epoch": 385.3333333333333,
      "grad_norm": 13.45079231262207,
      "learning_rate": 7.578710644677662e-07,
      "loss": 1.1092,
      "step": 4046
    },
    {
      "epoch": 385.42857142857144,
      "grad_norm": 12.979645729064941,
      "learning_rate": 7.580584707646177e-07,
      "loss": 1.1114,
      "step": 4047
    },
    {
      "epoch": 385.5238095238095,
      "grad_norm": 12.906009674072266,
      "learning_rate": 7.582458770614694e-07,
      "loss": 1.1234,
      "step": 4048
    },
    {
      "epoch": 385.6190476190476,
      "grad_norm": 23.5379695892334,
      "learning_rate": 7.584332833583209e-07,
      "loss": 1.0786,
      "step": 4049
    },
    {
      "epoch": 385.7142857142857,
      "grad_norm": 9.682621002197266,
      "learning_rate": 7.586206896551725e-07,
      "loss": 1.1018,
      "step": 4050
    },
    {
      "epoch": 385.8095238095238,
      "grad_norm": 7.639532566070557,
      "learning_rate": 7.58808095952024e-07,
      "loss": 1.1342,
      "step": 4051
    },
    {
      "epoch": 385.9047619047619,
      "grad_norm": 8.08156967163086,
      "learning_rate": 7.589955022488756e-07,
      "loss": 1.1269,
      "step": 4052
    },
    {
      "epoch": 386.0,
      "grad_norm": 5.800262928009033,
      "learning_rate": 7.591829085457271e-07,
      "loss": 1.0908,
      "step": 4053
    },
    {
      "epoch": 386.0,
      "eval_accuracy": 0.7426002248032971,
      "eval_f1": 0.7941264608930176,
      "eval_loss": 0.5496708750724792,
      "eval_precision": 0.7389849414389291,
      "eval_recall": 0.8581606217616581,
      "eval_roc_auc": 0.7928500287852621,
      "eval_runtime": 1.1583,
      "eval_samples_per_second": 2304.314,
      "eval_steps_per_second": 5.18,
      "step": 4053
    },
    {
      "epoch": 386.0952380952381,
      "grad_norm": 7.999423980712891,
      "learning_rate": 7.593703148425787e-07,
      "loss": 1.1201,
      "step": 4054
    },
    {
      "epoch": 386.1904761904762,
      "grad_norm": 8.465465545654297,
      "learning_rate": 7.595577211394304e-07,
      "loss": 1.1421,
      "step": 4055
    },
    {
      "epoch": 386.2857142857143,
      "grad_norm": 7.850752353668213,
      "learning_rate": 7.597451274362819e-07,
      "loss": 1.0854,
      "step": 4056
    },
    {
      "epoch": 386.3809523809524,
      "grad_norm": 4.966638565063477,
      "learning_rate": 7.599325337331335e-07,
      "loss": 1.132,
      "step": 4057
    },
    {
      "epoch": 386.4761904761905,
      "grad_norm": 6.809133529663086,
      "learning_rate": 7.60119940029985e-07,
      "loss": 1.096,
      "step": 4058
    },
    {
      "epoch": 386.57142857142856,
      "grad_norm": 5.391140460968018,
      "learning_rate": 7.603073463268366e-07,
      "loss": 1.1183,
      "step": 4059
    },
    {
      "epoch": 386.6666666666667,
      "grad_norm": 8.04463005065918,
      "learning_rate": 7.604947526236881e-07,
      "loss": 1.1102,
      "step": 4060
    },
    {
      "epoch": 386.76190476190476,
      "grad_norm": 5.56711483001709,
      "learning_rate": 7.606821589205397e-07,
      "loss": 1.1237,
      "step": 4061
    },
    {
      "epoch": 386.85714285714283,
      "grad_norm": 11.3066987991333,
      "learning_rate": 7.608695652173913e-07,
      "loss": 1.1338,
      "step": 4062
    },
    {
      "epoch": 386.95238095238096,
      "grad_norm": 14.796104431152344,
      "learning_rate": 7.610569715142429e-07,
      "loss": 1.0928,
      "step": 4063
    },
    {
      "epoch": 386.95238095238096,
      "eval_accuracy": 0.7366054702135632,
      "eval_f1": 0.782554902567275,
      "eval_loss": 0.5493862628936768,
      "eval_precision": 0.7489638839550029,
      "eval_recall": 0.819300518134715,
      "eval_roc_auc": 0.792762809441566,
      "eval_runtime": 1.1154,
      "eval_samples_per_second": 2392.873,
      "eval_steps_per_second": 5.379,
      "step": 4063
    },
    {
      "epoch": 387.04761904761904,
      "grad_norm": 15.751404762268066,
      "learning_rate": 7.612443778110945e-07,
      "loss": 1.0897,
      "step": 4064
    },
    {
      "epoch": 387.14285714285717,
      "grad_norm": 4.9107985496521,
      "learning_rate": 7.61431784107946e-07,
      "loss": 1.1318,
      "step": 4065
    },
    {
      "epoch": 387.23809523809524,
      "grad_norm": 9.686596870422363,
      "learning_rate": 7.616191904047977e-07,
      "loss": 1.0834,
      "step": 4066
    },
    {
      "epoch": 387.3333333333333,
      "grad_norm": 12.998448371887207,
      "learning_rate": 7.618065967016492e-07,
      "loss": 1.1354,
      "step": 4067
    },
    {
      "epoch": 387.42857142857144,
      "grad_norm": 6.6138153076171875,
      "learning_rate": 7.619940029985008e-07,
      "loss": 1.0676,
      "step": 4068
    },
    {
      "epoch": 387.5238095238095,
      "grad_norm": 12.226428985595703,
      "learning_rate": 7.621814092953524e-07,
      "loss": 1.1026,
      "step": 4069
    },
    {
      "epoch": 387.6190476190476,
      "grad_norm": 15.720524787902832,
      "learning_rate": 7.62368815592204e-07,
      "loss": 1.0988,
      "step": 4070
    },
    {
      "epoch": 387.7142857142857,
      "grad_norm": 22.66543197631836,
      "learning_rate": 7.625562218890555e-07,
      "loss": 1.1942,
      "step": 4071
    },
    {
      "epoch": 387.8095238095238,
      "grad_norm": 9.127182960510254,
      "learning_rate": 7.627436281859071e-07,
      "loss": 1.1264,
      "step": 4072
    },
    {
      "epoch": 387.9047619047619,
      "grad_norm": 6.516150951385498,
      "learning_rate": 7.629310344827587e-07,
      "loss": 1.1258,
      "step": 4073
    },
    {
      "epoch": 388.0,
      "grad_norm": 8.39951229095459,
      "learning_rate": 7.631184407796102e-07,
      "loss": 1.1152,
      "step": 4074
    },
    {
      "epoch": 388.0,
      "eval_accuracy": 0.7392281753465717,
      "eval_f1": 0.7861094038106945,
      "eval_loss": 0.5489965081214905,
      "eval_precision": 0.747953216374269,
      "eval_recall": 0.8283678756476683,
      "eval_roc_auc": 0.7930201496833621,
      "eval_runtime": 1.1205,
      "eval_samples_per_second": 2381.989,
      "eval_steps_per_second": 5.355,
      "step": 4074
    },
    {
      "epoch": 388.0952380952381,
      "grad_norm": 17.57448387145996,
      "learning_rate": 7.633058470764619e-07,
      "loss": 1.1187,
      "step": 4075
    },
    {
      "epoch": 388.1904761904762,
      "grad_norm": 9.762043952941895,
      "learning_rate": 7.634932533733134e-07,
      "loss": 1.075,
      "step": 4076
    },
    {
      "epoch": 388.2857142857143,
      "grad_norm": 8.012761116027832,
      "learning_rate": 7.63680659670165e-07,
      "loss": 1.1115,
      "step": 4077
    },
    {
      "epoch": 388.3809523809524,
      "grad_norm": 7.436084747314453,
      "learning_rate": 7.638680659670165e-07,
      "loss": 1.1101,
      "step": 4078
    },
    {
      "epoch": 388.4761904761905,
      "grad_norm": 11.756044387817383,
      "learning_rate": 7.640554722638681e-07,
      "loss": 1.1194,
      "step": 4079
    },
    {
      "epoch": 388.57142857142856,
      "grad_norm": 6.643121719360352,
      "learning_rate": 7.642428785607196e-07,
      "loss": 1.0684,
      "step": 4080
    },
    {
      "epoch": 388.6666666666667,
      "grad_norm": 7.203343868255615,
      "learning_rate": 7.644302848575712e-07,
      "loss": 1.0863,
      "step": 4081
    },
    {
      "epoch": 388.76190476190476,
      "grad_norm": 20.424135208129883,
      "learning_rate": 7.646176911544229e-07,
      "loss": 1.1182,
      "step": 4082
    },
    {
      "epoch": 388.85714285714283,
      "grad_norm": 24.31191062927246,
      "learning_rate": 7.648050974512744e-07,
      "loss": 1.15,
      "step": 4083
    },
    {
      "epoch": 388.95238095238096,
      "grad_norm": 14.023303031921387,
      "learning_rate": 7.64992503748126e-07,
      "loss": 1.145,
      "step": 4084
    },
    {
      "epoch": 388.95238095238096,
      "eval_accuracy": 0.744473585612589,
      "eval_f1": 0.7951951951951952,
      "eval_loss": 0.5493708848953247,
      "eval_precision": 0.7413213885778276,
      "eval_recall": 0.8575129533678757,
      "eval_roc_auc": 0.7932363270005757,
      "eval_runtime": 1.1349,
      "eval_samples_per_second": 2351.699,
      "eval_steps_per_second": 5.287,
      "step": 4084
    },
    {
      "epoch": 389.04761904761904,
      "grad_norm": 9.623614311218262,
      "learning_rate": 7.651799100449775e-07,
      "loss": 1.1326,
      "step": 4085
    },
    {
      "epoch": 389.14285714285717,
      "grad_norm": 8.110107421875,
      "learning_rate": 7.653673163418291e-07,
      "loss": 1.079,
      "step": 4086
    },
    {
      "epoch": 389.23809523809524,
      "grad_norm": 6.075845241546631,
      "learning_rate": 7.655547226386806e-07,
      "loss": 1.0716,
      "step": 4087
    },
    {
      "epoch": 389.3333333333333,
      "grad_norm": 9.143749237060547,
      "learning_rate": 7.657421289355323e-07,
      "loss": 1.0784,
      "step": 4088
    },
    {
      "epoch": 389.42857142857144,
      "grad_norm": 16.901330947875977,
      "learning_rate": 7.659295352323838e-07,
      "loss": 1.1384,
      "step": 4089
    },
    {
      "epoch": 389.5238095238095,
      "grad_norm": 24.563400268554688,
      "learning_rate": 7.661169415292354e-07,
      "loss": 1.0713,
      "step": 4090
    },
    {
      "epoch": 389.6190476190476,
      "grad_norm": 14.202386856079102,
      "learning_rate": 7.663043478260871e-07,
      "loss": 1.1105,
      "step": 4091
    },
    {
      "epoch": 389.7142857142857,
      "grad_norm": 4.999241828918457,
      "learning_rate": 7.664917541229386e-07,
      "loss": 1.1261,
      "step": 4092
    },
    {
      "epoch": 389.8095238095238,
      "grad_norm": 8.44494342803955,
      "learning_rate": 7.666791604197902e-07,
      "loss": 1.1392,
      "step": 4093
    },
    {
      "epoch": 389.9047619047619,
      "grad_norm": 11.428085327148438,
      "learning_rate": 7.668665667166417e-07,
      "loss": 1.1268,
      "step": 4094
    },
    {
      "epoch": 390.0,
      "grad_norm": 8.754639625549316,
      "learning_rate": 7.670539730134933e-07,
      "loss": 1.166,
      "step": 4095
    },
    {
      "epoch": 390.0,
      "eval_accuracy": 0.7396028475084301,
      "eval_f1": 0.7871362940275651,
      "eval_loss": 0.5485778450965881,
      "eval_precision": 0.7466589192330041,
      "eval_recall": 0.8322538860103627,
      "eval_roc_auc": 0.7935293609671848,
      "eval_runtime": 1.1308,
      "eval_samples_per_second": 2360.28,
      "eval_steps_per_second": 5.306,
      "step": 4095
    },
    {
      "epoch": 390.0952380952381,
      "grad_norm": 6.919116020202637,
      "learning_rate": 7.672413793103449e-07,
      "loss": 1.0923,
      "step": 4096
    },
    {
      "epoch": 390.1904761904762,
      "grad_norm": 9.43787670135498,
      "learning_rate": 7.674287856071965e-07,
      "loss": 1.1345,
      "step": 4097
    },
    {
      "epoch": 390.2857142857143,
      "grad_norm": 5.613615989685059,
      "learning_rate": 7.67616191904048e-07,
      "loss": 1.0736,
      "step": 4098
    },
    {
      "epoch": 390.3809523809524,
      "grad_norm": 9.838118553161621,
      "learning_rate": 7.678035982008996e-07,
      "loss": 1.1262,
      "step": 4099
    },
    {
      "epoch": 390.4761904761905,
      "grad_norm": 22.736331939697266,
      "learning_rate": 7.679910044977512e-07,
      "loss": 1.1702,
      "step": 4100
    },
    {
      "epoch": 390.57142857142856,
      "grad_norm": 11.852968215942383,
      "learning_rate": 7.681784107946027e-07,
      "loss": 1.1134,
      "step": 4101
    },
    {
      "epoch": 390.6666666666667,
      "grad_norm": 5.733023166656494,
      "learning_rate": 7.683658170914543e-07,
      "loss": 1.094,
      "step": 4102
    },
    {
      "epoch": 390.76190476190476,
      "grad_norm": 8.226594924926758,
      "learning_rate": 7.685532233883059e-07,
      "loss": 1.1004,
      "step": 4103
    },
    {
      "epoch": 390.85714285714283,
      "grad_norm": 8.334565162658691,
      "learning_rate": 7.687406296851575e-07,
      "loss": 1.1379,
      "step": 4104
    },
    {
      "epoch": 390.95238095238096,
      "grad_norm": 7.8795976638793945,
      "learning_rate": 7.68928035982009e-07,
      "loss": 1.0851,
      "step": 4105
    },
    {
      "epoch": 390.95238095238096,
      "eval_accuracy": 0.7403521918321468,
      "eval_f1": 0.7841793833696667,
      "eval_loss": 0.5487409234046936,
      "eval_precision": 0.755248950209958,
      "eval_recall": 0.8154145077720207,
      "eval_roc_auc": 0.7933934945308002,
      "eval_runtime": 1.1417,
      "eval_samples_per_second": 2337.756,
      "eval_steps_per_second": 5.255,
      "step": 4105
    },
    {
      "epoch": 391.04761904761904,
      "grad_norm": 24.32604217529297,
      "learning_rate": 7.691154422788606e-07,
      "loss": 1.0997,
      "step": 4106
    },
    {
      "epoch": 391.14285714285717,
      "grad_norm": 21.518884658813477,
      "learning_rate": 7.693028485757121e-07,
      "loss": 1.1185,
      "step": 4107
    },
    {
      "epoch": 391.23809523809524,
      "grad_norm": 12.176133155822754,
      "learning_rate": 7.694902548725637e-07,
      "loss": 1.1042,
      "step": 4108
    },
    {
      "epoch": 391.3333333333333,
      "grad_norm": 5.470086574554443,
      "learning_rate": 7.696776611694153e-07,
      "loss": 1.0876,
      "step": 4109
    },
    {
      "epoch": 391.42857142857144,
      "grad_norm": 8.903295516967773,
      "learning_rate": 7.698650674662669e-07,
      "loss": 1.119,
      "step": 4110
    },
    {
      "epoch": 391.5238095238095,
      "grad_norm": 6.113744258880615,
      "learning_rate": 7.700524737631185e-07,
      "loss": 1.1801,
      "step": 4111
    },
    {
      "epoch": 391.6190476190476,
      "grad_norm": 6.159307479858398,
      "learning_rate": 7.7023988005997e-07,
      "loss": 1.1089,
      "step": 4112
    },
    {
      "epoch": 391.7142857142857,
      "grad_norm": 16.8972110748291,
      "learning_rate": 7.704272863568217e-07,
      "loss": 1.1127,
      "step": 4113
    },
    {
      "epoch": 391.8095238095238,
      "grad_norm": 12.682228088378906,
      "learning_rate": 7.706146926536732e-07,
      "loss": 1.1253,
      "step": 4114
    },
    {
      "epoch": 391.9047619047619,
      "grad_norm": 12.118672370910645,
      "learning_rate": 7.708020989505248e-07,
      "loss": 1.083,
      "step": 4115
    },
    {
      "epoch": 392.0,
      "grad_norm": 6.608816146850586,
      "learning_rate": 7.709895052473763e-07,
      "loss": 1.0857,
      "step": 4116
    },
    {
      "epoch": 392.0,
      "eval_accuracy": 0.7392281753465717,
      "eval_f1": 0.7851851851851852,
      "eval_loss": 0.5484685301780701,
      "eval_precision": 0.75,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.7933667242371906,
      "eval_runtime": 1.1364,
      "eval_samples_per_second": 2348.638,
      "eval_steps_per_second": 5.28,
      "step": 4116
    },
    {
      "epoch": 392.0952380952381,
      "grad_norm": 7.915292739868164,
      "learning_rate": 7.71176911544228e-07,
      "loss": 1.1346,
      "step": 4117
    },
    {
      "epoch": 392.1904761904762,
      "grad_norm": 20.730438232421875,
      "learning_rate": 7.713643178410796e-07,
      "loss": 1.0698,
      "step": 4118
    },
    {
      "epoch": 392.2857142857143,
      "grad_norm": 9.250012397766113,
      "learning_rate": 7.715517241379311e-07,
      "loss": 1.1287,
      "step": 4119
    },
    {
      "epoch": 392.3809523809524,
      "grad_norm": 10.762714385986328,
      "learning_rate": 7.717391304347827e-07,
      "loss": 1.0905,
      "step": 4120
    },
    {
      "epoch": 392.4761904761905,
      "grad_norm": 24.7636775970459,
      "learning_rate": 7.719265367316342e-07,
      "loss": 1.1358,
      "step": 4121
    },
    {
      "epoch": 392.57142857142856,
      "grad_norm": 7.282887935638428,
      "learning_rate": 7.721139430284858e-07,
      "loss": 1.0824,
      "step": 4122
    },
    {
      "epoch": 392.6666666666667,
      "grad_norm": 3.8522496223449707,
      "learning_rate": 7.723013493253373e-07,
      "loss": 1.1061,
      "step": 4123
    },
    {
      "epoch": 392.76190476190476,
      "grad_norm": 5.173220634460449,
      "learning_rate": 7.72488755622189e-07,
      "loss": 1.0511,
      "step": 4124
    },
    {
      "epoch": 392.85714285714283,
      "grad_norm": 7.5248918533325195,
      "learning_rate": 7.726761619190405e-07,
      "loss": 1.0936,
      "step": 4125
    },
    {
      "epoch": 392.95238095238096,
      "grad_norm": 7.896293640136719,
      "learning_rate": 7.728635682158921e-07,
      "loss": 1.0633,
      "step": 4126
    },
    {
      "epoch": 392.95238095238096,
      "eval_accuracy": 0.744473585612589,
      "eval_f1": 0.7947019867549668,
      "eval_loss": 0.5479307770729065,
      "eval_precision": 0.7424071991001124,
      "eval_recall": 0.8549222797927462,
      "eval_roc_auc": 0.7944922279792745,
      "eval_runtime": 1.1524,
      "eval_samples_per_second": 2316.088,
      "eval_steps_per_second": 5.207,
      "step": 4126
    },
    {
      "epoch": 393.04761904761904,
      "grad_norm": 14.139813423156738,
      "learning_rate": 7.730509745127437e-07,
      "loss": 1.1619,
      "step": 4127
    },
    {
      "epoch": 393.14285714285717,
      "grad_norm": 9.308013916015625,
      "learning_rate": 7.732383808095952e-07,
      "loss": 1.1096,
      "step": 4128
    },
    {
      "epoch": 393.23809523809524,
      "grad_norm": 5.986069679260254,
      "learning_rate": 7.734257871064468e-07,
      "loss": 1.1388,
      "step": 4129
    },
    {
      "epoch": 393.3333333333333,
      "grad_norm": 12.865239143371582,
      "learning_rate": 7.736131934032983e-07,
      "loss": 1.1361,
      "step": 4130
    },
    {
      "epoch": 393.42857142857144,
      "grad_norm": 8.613828659057617,
      "learning_rate": 7.7380059970015e-07,
      "loss": 1.0821,
      "step": 4131
    },
    {
      "epoch": 393.5238095238095,
      "grad_norm": 11.044524192810059,
      "learning_rate": 7.739880059970015e-07,
      "loss": 1.1162,
      "step": 4132
    },
    {
      "epoch": 393.6190476190476,
      "grad_norm": 5.985708236694336,
      "learning_rate": 7.741754122938531e-07,
      "loss": 1.1143,
      "step": 4133
    },
    {
      "epoch": 393.7142857142857,
      "grad_norm": 4.828469753265381,
      "learning_rate": 7.743628185907046e-07,
      "loss": 1.0967,
      "step": 4134
    },
    {
      "epoch": 393.8095238095238,
      "grad_norm": 9.598358154296875,
      "learning_rate": 7.745502248875562e-07,
      "loss": 1.1205,
      "step": 4135
    },
    {
      "epoch": 393.9047619047619,
      "grad_norm": 6.418941497802734,
      "learning_rate": 7.747376311844079e-07,
      "loss": 1.1402,
      "step": 4136
    },
    {
      "epoch": 394.0,
      "grad_norm": 14.198280334472656,
      "learning_rate": 7.749250374812594e-07,
      "loss": 1.1459,
      "step": 4137
    },
    {
      "epoch": 394.0,
      "eval_accuracy": 0.7440989134507306,
      "eval_f1": 0.7920852359208523,
      "eval_loss": 0.5479085445404053,
      "eval_precision": 0.7472716829408386,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7939928036845136,
      "eval_runtime": 1.1257,
      "eval_samples_per_second": 2370.9,
      "eval_steps_per_second": 5.33,
      "step": 4137
    },
    {
      "epoch": 394.0952380952381,
      "grad_norm": 7.425207614898682,
      "learning_rate": 7.751124437781111e-07,
      "loss": 1.0908,
      "step": 4138
    },
    {
      "epoch": 394.1904761904762,
      "grad_norm": 7.119057655334473,
      "learning_rate": 7.752998500749626e-07,
      "loss": 1.0966,
      "step": 4139
    },
    {
      "epoch": 394.2857142857143,
      "grad_norm": 10.171456336975098,
      "learning_rate": 7.754872563718142e-07,
      "loss": 1.1169,
      "step": 4140
    },
    {
      "epoch": 394.3809523809524,
      "grad_norm": 8.68658447265625,
      "learning_rate": 7.756746626686657e-07,
      "loss": 1.145,
      "step": 4141
    },
    {
      "epoch": 394.4761904761905,
      "grad_norm": 6.64039421081543,
      "learning_rate": 7.758620689655173e-07,
      "loss": 1.127,
      "step": 4142
    },
    {
      "epoch": 394.57142857142856,
      "grad_norm": 18.38219451904297,
      "learning_rate": 7.760494752623688e-07,
      "loss": 1.1243,
      "step": 4143
    },
    {
      "epoch": 394.6666666666667,
      "grad_norm": 12.123787879943848,
      "learning_rate": 7.762368815592204e-07,
      "loss": 1.1077,
      "step": 4144
    },
    {
      "epoch": 394.76190476190476,
      "grad_norm": 18.44883918762207,
      "learning_rate": 7.764242878560721e-07,
      "loss": 1.0949,
      "step": 4145
    },
    {
      "epoch": 394.85714285714283,
      "grad_norm": 6.585038661956787,
      "learning_rate": 7.766116941529236e-07,
      "loss": 1.1109,
      "step": 4146
    },
    {
      "epoch": 394.95238095238096,
      "grad_norm": 8.92080020904541,
      "learning_rate": 7.767991004497752e-07,
      "loss": 1.0813,
      "step": 4147
    },
    {
      "epoch": 394.95238095238096,
      "eval_accuracy": 0.7429748969651555,
      "eval_f1": 0.7914893617021277,
      "eval_loss": 0.5480291843414307,
      "eval_precision": 0.7457044673539519,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7937679907887162,
      "eval_runtime": 1.1425,
      "eval_samples_per_second": 2336.026,
      "eval_steps_per_second": 5.251,
      "step": 4147
    },
    {
      "epoch": 395.04761904761904,
      "grad_norm": 25.43185043334961,
      "learning_rate": 7.769865067466267e-07,
      "loss": 1.1358,
      "step": 4148
    },
    {
      "epoch": 395.14285714285717,
      "grad_norm": 14.379817008972168,
      "learning_rate": 7.771739130434783e-07,
      "loss": 1.1052,
      "step": 4149
    },
    {
      "epoch": 395.23809523809524,
      "grad_norm": 10.084162712097168,
      "learning_rate": 7.773613193403298e-07,
      "loss": 1.1024,
      "step": 4150
    },
    {
      "epoch": 395.3333333333333,
      "grad_norm": 11.979368209838867,
      "learning_rate": 7.775487256371814e-07,
      "loss": 1.1147,
      "step": 4151
    },
    {
      "epoch": 395.42857142857144,
      "grad_norm": 12.168645858764648,
      "learning_rate": 7.77736131934033e-07,
      "loss": 1.1247,
      "step": 4152
    },
    {
      "epoch": 395.5238095238095,
      "grad_norm": 9.172712326049805,
      "learning_rate": 7.779235382308846e-07,
      "loss": 1.1047,
      "step": 4153
    },
    {
      "epoch": 395.6190476190476,
      "grad_norm": 7.719291687011719,
      "learning_rate": 7.781109445277362e-07,
      "loss": 1.1277,
      "step": 4154
    },
    {
      "epoch": 395.7142857142857,
      "grad_norm": 7.303157806396484,
      "learning_rate": 7.782983508245877e-07,
      "loss": 1.0759,
      "step": 4155
    },
    {
      "epoch": 395.8095238095238,
      "grad_norm": 17.161909103393555,
      "learning_rate": 7.784857571214393e-07,
      "loss": 1.0705,
      "step": 4156
    },
    {
      "epoch": 395.9047619047619,
      "grad_norm": 6.004673957824707,
      "learning_rate": 7.786731634182908e-07,
      "loss": 1.1455,
      "step": 4157
    },
    {
      "epoch": 396.0,
      "grad_norm": 8.70738697052002,
      "learning_rate": 7.788605697151425e-07,
      "loss": 1.1343,
      "step": 4158
    },
    {
      "epoch": 396.0,
      "eval_accuracy": 0.7411015361558636,
      "eval_f1": 0.7878415719987719,
      "eval_loss": 0.5477091670036316,
      "eval_precision": 0.7489784004670169,
      "eval_recall": 0.8309585492227979,
      "eval_roc_auc": 0.7940800230282095,
      "eval_runtime": 1.1309,
      "eval_samples_per_second": 2360.138,
      "eval_steps_per_second": 5.306,
      "step": 4158
    },
    {
      "epoch": 396.0952380952381,
      "grad_norm": 7.129392147064209,
      "learning_rate": 7.79047976011994e-07,
      "loss": 1.1285,
      "step": 4159
    },
    {
      "epoch": 396.1904761904762,
      "grad_norm": 7.756862163543701,
      "learning_rate": 7.792353823088456e-07,
      "loss": 1.1352,
      "step": 4160
    },
    {
      "epoch": 396.2857142857143,
      "grad_norm": 7.004748344421387,
      "learning_rate": 7.794227886056972e-07,
      "loss": 1.0852,
      "step": 4161
    },
    {
      "epoch": 396.3809523809524,
      "grad_norm": 8.161051750183105,
      "learning_rate": 7.796101949025488e-07,
      "loss": 1.0557,
      "step": 4162
    },
    {
      "epoch": 396.4761904761905,
      "grad_norm": 12.619900703430176,
      "learning_rate": 7.797976011994004e-07,
      "loss": 1.0826,
      "step": 4163
    },
    {
      "epoch": 396.57142857142856,
      "grad_norm": 19.46121597290039,
      "learning_rate": 7.799850074962519e-07,
      "loss": 1.139,
      "step": 4164
    },
    {
      "epoch": 396.6666666666667,
      "grad_norm": 6.766005516052246,
      "learning_rate": 7.801724137931035e-07,
      "loss": 1.0997,
      "step": 4165
    },
    {
      "epoch": 396.76190476190476,
      "grad_norm": 5.906949520111084,
      "learning_rate": 7.803598200899551e-07,
      "loss": 1.0896,
      "step": 4166
    },
    {
      "epoch": 396.85714285714283,
      "grad_norm": 7.1702165603637695,
      "learning_rate": 7.805472263868067e-07,
      "loss": 1.1346,
      "step": 4167
    },
    {
      "epoch": 396.95238095238096,
      "grad_norm": 10.30286979675293,
      "learning_rate": 7.807346326836582e-07,
      "loss": 1.0979,
      "step": 4168
    },
    {
      "epoch": 396.95238095238096,
      "eval_accuracy": 0.7399775196702885,
      "eval_f1": 0.7868550368550369,
      "eval_loss": 0.5477451682090759,
      "eval_precision": 0.7482476635514018,
      "eval_recall": 0.8296632124352331,
      "eval_roc_auc": 0.7941706966033392,
      "eval_runtime": 1.1281,
      "eval_samples_per_second": 2365.986,
      "eval_steps_per_second": 5.319,
      "step": 4168
    },
    {
      "epoch": 397.04761904761904,
      "grad_norm": 19.294652938842773,
      "learning_rate": 7.809220389805098e-07,
      "loss": 1.1442,
      "step": 4169
    },
    {
      "epoch": 397.14285714285717,
      "grad_norm": 6.437745094299316,
      "learning_rate": 7.811094452773613e-07,
      "loss": 1.127,
      "step": 4170
    },
    {
      "epoch": 397.23809523809524,
      "grad_norm": 11.683369636535645,
      "learning_rate": 7.81296851574213e-07,
      "loss": 1.0871,
      "step": 4171
    },
    {
      "epoch": 397.3333333333333,
      "grad_norm": 13.600419998168945,
      "learning_rate": 7.814842578710645e-07,
      "loss": 1.1089,
      "step": 4172
    },
    {
      "epoch": 397.42857142857144,
      "grad_norm": 8.484980583190918,
      "learning_rate": 7.816716641679161e-07,
      "loss": 1.1061,
      "step": 4173
    },
    {
      "epoch": 397.5238095238095,
      "grad_norm": 6.674485206604004,
      "learning_rate": 7.818590704647676e-07,
      "loss": 1.1898,
      "step": 4174
    },
    {
      "epoch": 397.6190476190476,
      "grad_norm": 7.062256336212158,
      "learning_rate": 7.820464767616192e-07,
      "loss": 1.0903,
      "step": 4175
    },
    {
      "epoch": 397.7142857142857,
      "grad_norm": 6.853922367095947,
      "learning_rate": 7.822338830584709e-07,
      "loss": 1.0625,
      "step": 4176
    },
    {
      "epoch": 397.8095238095238,
      "grad_norm": 12.509185791015625,
      "learning_rate": 7.824212893553223e-07,
      "loss": 1.1676,
      "step": 4177
    },
    {
      "epoch": 397.9047619047619,
      "grad_norm": 7.87623929977417,
      "learning_rate": 7.82608695652174e-07,
      "loss": 1.1061,
      "step": 4178
    },
    {
      "epoch": 398.0,
      "grad_norm": 7.967573642730713,
      "learning_rate": 7.827961019490254e-07,
      "loss": 1.1038,
      "step": 4179
    },
    {
      "epoch": 398.0,
      "eval_accuracy": 0.7422255526414387,
      "eval_f1": 0.7892156862745098,
      "eval_loss": 0.547892689704895,
      "eval_precision": 0.7488372093023256,
      "eval_recall": 0.8341968911917098,
      "eval_roc_auc": 0.7940877950489349,
      "eval_runtime": 1.1298,
      "eval_samples_per_second": 2362.435,
      "eval_steps_per_second": 5.311,
      "step": 4179
    },
    {
      "epoch": 398.0952380952381,
      "grad_norm": 7.890987396240234,
      "learning_rate": 7.829835082458771e-07,
      "loss": 1.1294,
      "step": 4180
    },
    {
      "epoch": 398.1904761904762,
      "grad_norm": 6.791111946105957,
      "learning_rate": 7.831709145427287e-07,
      "loss": 1.1611,
      "step": 4181
    },
    {
      "epoch": 398.2857142857143,
      "grad_norm": 18.181640625,
      "learning_rate": 7.833583208395802e-07,
      "loss": 1.1183,
      "step": 4182
    },
    {
      "epoch": 398.3809523809524,
      "grad_norm": 11.313365936279297,
      "learning_rate": 7.835457271364319e-07,
      "loss": 1.1054,
      "step": 4183
    },
    {
      "epoch": 398.4761904761905,
      "grad_norm": 9.285812377929688,
      "learning_rate": 7.837331334332834e-07,
      "loss": 1.1179,
      "step": 4184
    },
    {
      "epoch": 398.57142857142856,
      "grad_norm": 15.523584365844727,
      "learning_rate": 7.83920539730135e-07,
      "loss": 1.0533,
      "step": 4185
    },
    {
      "epoch": 398.6666666666667,
      "grad_norm": 10.66666030883789,
      "learning_rate": 7.841079460269865e-07,
      "loss": 1.1006,
      "step": 4186
    },
    {
      "epoch": 398.76190476190476,
      "grad_norm": 5.271409511566162,
      "learning_rate": 7.842953523238382e-07,
      "loss": 1.1081,
      "step": 4187
    },
    {
      "epoch": 398.85714285714283,
      "grad_norm": 7.439558029174805,
      "learning_rate": 7.844827586206896e-07,
      "loss": 1.0676,
      "step": 4188
    },
    {
      "epoch": 398.95238095238096,
      "grad_norm": 6.201678276062012,
      "learning_rate": 7.846701649175413e-07,
      "loss": 1.1458,
      "step": 4189
    },
    {
      "epoch": 398.95238095238096,
      "eval_accuracy": 0.7448482577744474,
      "eval_f1": 0.7965342097400657,
      "eval_loss": 0.5484153032302856,
      "eval_precision": 0.7393233499722685,
      "eval_recall": 0.8633419689119171,
      "eval_roc_auc": 0.794457973517559,
      "eval_runtime": 1.1373,
      "eval_samples_per_second": 2346.777,
      "eval_steps_per_second": 5.276,
      "step": 4189
    },
    {
      "epoch": 399.04761904761904,
      "grad_norm": 5.896444320678711,
      "learning_rate": 7.84857571214393e-07,
      "loss": 1.0803,
      "step": 4190
    },
    {
      "epoch": 399.14285714285717,
      "grad_norm": 19.47102165222168,
      "learning_rate": 7.850449775112444e-07,
      "loss": 1.1401,
      "step": 4191
    },
    {
      "epoch": 399.23809523809524,
      "grad_norm": 14.513976097106934,
      "learning_rate": 7.852323838080961e-07,
      "loss": 1.1022,
      "step": 4192
    },
    {
      "epoch": 399.3333333333333,
      "grad_norm": 19.547645568847656,
      "learning_rate": 7.854197901049475e-07,
      "loss": 1.1015,
      "step": 4193
    },
    {
      "epoch": 399.42857142857144,
      "grad_norm": 6.47927713394165,
      "learning_rate": 7.856071964017992e-07,
      "loss": 1.0758,
      "step": 4194
    },
    {
      "epoch": 399.5238095238095,
      "grad_norm": 11.972545623779297,
      "learning_rate": 7.857946026986506e-07,
      "loss": 1.1458,
      "step": 4195
    },
    {
      "epoch": 399.6190476190476,
      "grad_norm": 6.737449645996094,
      "learning_rate": 7.859820089955023e-07,
      "loss": 1.1259,
      "step": 4196
    },
    {
      "epoch": 399.7142857142857,
      "grad_norm": 4.990560531616211,
      "learning_rate": 7.86169415292354e-07,
      "loss": 1.1201,
      "step": 4197
    },
    {
      "epoch": 399.8095238095238,
      "grad_norm": 14.076717376708984,
      "learning_rate": 7.863568215892054e-07,
      "loss": 1.1131,
      "step": 4198
    },
    {
      "epoch": 399.9047619047619,
      "grad_norm": 13.812651634216309,
      "learning_rate": 7.865442278860571e-07,
      "loss": 1.0948,
      "step": 4199
    },
    {
      "epoch": 400.0,
      "grad_norm": 11.414024353027344,
      "learning_rate": 7.867316341829085e-07,
      "loss": 1.1089,
      "step": 4200
    },
    {
      "epoch": 400.0,
      "eval_accuracy": 0.7418508804795804,
      "eval_f1": 0.7879347491535857,
      "eval_loss": 0.547219455242157,
      "eval_precision": 0.750733137829912,
      "eval_recall": 0.8290155440414507,
      "eval_roc_auc": 0.7948862982153139,
      "eval_runtime": 1.1302,
      "eval_samples_per_second": 2361.519,
      "eval_steps_per_second": 5.309,
      "step": 4200
    },
    {
      "epoch": 400.0952380952381,
      "grad_norm": 5.199063777923584,
      "learning_rate": 7.869190404797602e-07,
      "loss": 1.1062,
      "step": 4201
    },
    {
      "epoch": 400.1904761904762,
      "grad_norm": 7.194316387176514,
      "learning_rate": 7.871064467766116e-07,
      "loss": 1.108,
      "step": 4202
    },
    {
      "epoch": 400.2857142857143,
      "grad_norm": 15.215425491333008,
      "learning_rate": 7.872938530734633e-07,
      "loss": 1.0974,
      "step": 4203
    },
    {
      "epoch": 400.3809523809524,
      "grad_norm": 6.132933616638184,
      "learning_rate": 7.874812593703149e-07,
      "loss": 1.1083,
      "step": 4204
    },
    {
      "epoch": 400.4761904761905,
      "grad_norm": 7.98865270614624,
      "learning_rate": 7.876686656671665e-07,
      "loss": 1.0969,
      "step": 4205
    },
    {
      "epoch": 400.57142857142856,
      "grad_norm": 21.348203659057617,
      "learning_rate": 7.87856071964018e-07,
      "loss": 1.1038,
      "step": 4206
    },
    {
      "epoch": 400.6666666666667,
      "grad_norm": 21.51778793334961,
      "learning_rate": 7.880434782608696e-07,
      "loss": 1.1275,
      "step": 4207
    },
    {
      "epoch": 400.76190476190476,
      "grad_norm": 4.526026248931885,
      "learning_rate": 7.882308845577213e-07,
      "loss": 1.1037,
      "step": 4208
    },
    {
      "epoch": 400.85714285714283,
      "grad_norm": 11.087535858154297,
      "learning_rate": 7.884182908545727e-07,
      "loss": 1.1356,
      "step": 4209
    },
    {
      "epoch": 400.95238095238096,
      "grad_norm": 11.458879470825195,
      "learning_rate": 7.886056971514244e-07,
      "loss": 1.1229,
      "step": 4210
    },
    {
      "epoch": 400.95238095238096,
      "eval_accuracy": 0.7388535031847133,
      "eval_f1": 0.7829336655247586,
      "eval_loss": 0.5476679801940918,
      "eval_precision": 0.7540491901619676,
      "eval_recall": 0.814119170984456,
      "eval_roc_auc": 0.7943388025331031,
      "eval_runtime": 1.1127,
      "eval_samples_per_second": 2398.776,
      "eval_steps_per_second": 5.393,
      "step": 4210
    },
    {
      "epoch": 401.04761904761904,
      "grad_norm": 9.395782470703125,
      "learning_rate": 7.88793103448276e-07,
      "loss": 1.1361,
      "step": 4211
    },
    {
      "epoch": 401.14285714285717,
      "grad_norm": 22.677854537963867,
      "learning_rate": 7.889805097451275e-07,
      "loss": 1.1372,
      "step": 4212
    },
    {
      "epoch": 401.23809523809524,
      "grad_norm": 15.018343925476074,
      "learning_rate": 7.891679160419791e-07,
      "loss": 1.0888,
      "step": 4213
    },
    {
      "epoch": 401.3333333333333,
      "grad_norm": 5.283858776092529,
      "learning_rate": 7.893553223388306e-07,
      "loss": 1.1102,
      "step": 4214
    },
    {
      "epoch": 401.42857142857144,
      "grad_norm": 9.966854095458984,
      "learning_rate": 7.895427286356822e-07,
      "loss": 1.0927,
      "step": 4215
    },
    {
      "epoch": 401.5238095238095,
      "grad_norm": 5.786252498626709,
      "learning_rate": 7.897301349325337e-07,
      "loss": 1.1023,
      "step": 4216
    },
    {
      "epoch": 401.6190476190476,
      "grad_norm": 5.132158279418945,
      "learning_rate": 7.899175412293854e-07,
      "loss": 1.1067,
      "step": 4217
    },
    {
      "epoch": 401.7142857142857,
      "grad_norm": 7.377092361450195,
      "learning_rate": 7.90104947526237e-07,
      "loss": 1.0825,
      "step": 4218
    },
    {
      "epoch": 401.8095238095238,
      "grad_norm": 13.699070930480957,
      "learning_rate": 7.902923538230885e-07,
      "loss": 1.1531,
      "step": 4219
    },
    {
      "epoch": 401.9047619047619,
      "grad_norm": 12.673673629760742,
      "learning_rate": 7.904797601199401e-07,
      "loss": 1.0908,
      "step": 4220
    },
    {
      "epoch": 402.0,
      "grad_norm": 9.16825008392334,
      "learning_rate": 7.906671664167916e-07,
      "loss": 1.1487,
      "step": 4221
    },
    {
      "epoch": 402.0,
      "eval_accuracy": 0.747470962907456,
      "eval_f1": 0.7962515114873036,
      "eval_loss": 0.5470756888389587,
      "eval_precision": 0.7465986394557823,
      "eval_recall": 0.852979274611399,
      "eval_roc_auc": 0.7952337363270007,
      "eval_runtime": 1.1274,
      "eval_samples_per_second": 2367.488,
      "eval_steps_per_second": 5.322,
      "step": 4221
    },
    {
      "epoch": 402.0952380952381,
      "grad_norm": 11.666342735290527,
      "learning_rate": 7.908545727136432e-07,
      "loss": 1.115,
      "step": 4222
    },
    {
      "epoch": 402.1904761904762,
      "grad_norm": 7.8304033279418945,
      "learning_rate": 7.910419790104947e-07,
      "loss": 1.0772,
      "step": 4223
    },
    {
      "epoch": 402.2857142857143,
      "grad_norm": 8.050092697143555,
      "learning_rate": 7.912293853073463e-07,
      "loss": 1.1208,
      "step": 4224
    },
    {
      "epoch": 402.3809523809524,
      "grad_norm": 7.431957721710205,
      "learning_rate": 7.91416791604198e-07,
      "loss": 1.139,
      "step": 4225
    },
    {
      "epoch": 402.4761904761905,
      "grad_norm": 7.459015369415283,
      "learning_rate": 7.916041979010495e-07,
      "loss": 1.1413,
      "step": 4226
    },
    {
      "epoch": 402.57142857142856,
      "grad_norm": 8.06148624420166,
      "learning_rate": 7.917916041979011e-07,
      "loss": 1.0962,
      "step": 4227
    },
    {
      "epoch": 402.6666666666667,
      "grad_norm": 6.986510753631592,
      "learning_rate": 7.919790104947527e-07,
      "loss": 1.097,
      "step": 4228
    },
    {
      "epoch": 402.76190476190476,
      "grad_norm": 7.007997035980225,
      "learning_rate": 7.921664167916042e-07,
      "loss": 1.1002,
      "step": 4229
    },
    {
      "epoch": 402.85714285714283,
      "grad_norm": 6.520593166351318,
      "learning_rate": 7.923538230884558e-07,
      "loss": 1.1174,
      "step": 4230
    },
    {
      "epoch": 402.95238095238096,
      "grad_norm": 6.7104105949401855,
      "learning_rate": 7.925412293853074e-07,
      "loss": 1.0596,
      "step": 4231
    },
    {
      "epoch": 402.95238095238096,
      "eval_accuracy": 0.7463469464218808,
      "eval_f1": 0.7957767722473604,
      "eval_loss": 0.5472564697265625,
      "eval_precision": 0.7447769621682665,
      "eval_recall": 0.8542746113989638,
      "eval_roc_auc": 0.7951096718480137,
      "eval_runtime": 1.1411,
      "eval_samples_per_second": 2339.006,
      "eval_steps_per_second": 5.258,
      "step": 4231
    },
    {
      "epoch": 403.04761904761904,
      "grad_norm": 16.685976028442383,
      "learning_rate": 7.92728635682159e-07,
      "loss": 1.0923,
      "step": 4232
    },
    {
      "epoch": 403.14285714285717,
      "grad_norm": 7.973769187927246,
      "learning_rate": 7.929160419790105e-07,
      "loss": 1.0894,
      "step": 4233
    },
    {
      "epoch": 403.23809523809524,
      "grad_norm": 16.108617782592773,
      "learning_rate": 7.931034482758622e-07,
      "loss": 1.0602,
      "step": 4234
    },
    {
      "epoch": 403.3333333333333,
      "grad_norm": 12.90723991394043,
      "learning_rate": 7.932908545727137e-07,
      "loss": 1.1349,
      "step": 4235
    },
    {
      "epoch": 403.42857142857144,
      "grad_norm": 16.323835372924805,
      "learning_rate": 7.934782608695653e-07,
      "loss": 1.1329,
      "step": 4236
    },
    {
      "epoch": 403.5238095238095,
      "grad_norm": 4.875527858734131,
      "learning_rate": 7.936656671664168e-07,
      "loss": 1.0819,
      "step": 4237
    },
    {
      "epoch": 403.6190476190476,
      "grad_norm": 6.662182331085205,
      "learning_rate": 7.938530734632684e-07,
      "loss": 1.0866,
      "step": 4238
    },
    {
      "epoch": 403.7142857142857,
      "grad_norm": 8.983207702636719,
      "learning_rate": 7.940404797601201e-07,
      "loss": 1.0948,
      "step": 4239
    },
    {
      "epoch": 403.8095238095238,
      "grad_norm": 8.16940689086914,
      "learning_rate": 7.942278860569715e-07,
      "loss": 1.1415,
      "step": 4240
    },
    {
      "epoch": 403.9047619047619,
      "grad_norm": 7.25838565826416,
      "learning_rate": 7.944152923538232e-07,
      "loss": 1.1255,
      "step": 4241
    },
    {
      "epoch": 404.0,
      "grad_norm": 8.486621856689453,
      "learning_rate": 7.946026986506746e-07,
      "loss": 1.1461,
      "step": 4242
    },
    {
      "epoch": 404.0,
      "eval_accuracy": 0.7392281753465717,
      "eval_f1": 0.7835820895522388,
      "eval_loss": 0.5473288893699646,
      "eval_precision": 0.7535885167464115,
      "eval_recall": 0.8160621761658031,
      "eval_roc_auc": 0.7946073690270582,
      "eval_runtime": 1.1341,
      "eval_samples_per_second": 2353.457,
      "eval_steps_per_second": 5.291,
      "step": 4242
    },
    {
      "epoch": 404.0952380952381,
      "grad_norm": 10.87182331085205,
      "learning_rate": 7.947901049475263e-07,
      "loss": 1.1152,
      "step": 4243
    },
    {
      "epoch": 404.1904761904762,
      "grad_norm": 11.432137489318848,
      "learning_rate": 7.949775112443778e-07,
      "loss": 1.0998,
      "step": 4244
    },
    {
      "epoch": 404.2857142857143,
      "grad_norm": 5.492263317108154,
      "learning_rate": 7.951649175412294e-07,
      "loss": 1.1035,
      "step": 4245
    },
    {
      "epoch": 404.3809523809524,
      "grad_norm": 13.131122589111328,
      "learning_rate": 7.953523238380811e-07,
      "loss": 1.1386,
      "step": 4246
    },
    {
      "epoch": 404.4761904761905,
      "grad_norm": 6.534399509429932,
      "learning_rate": 7.955397301349325e-07,
      "loss": 1.1258,
      "step": 4247
    },
    {
      "epoch": 404.57142857142856,
      "grad_norm": 9.083715438842773,
      "learning_rate": 7.957271364317842e-07,
      "loss": 1.126,
      "step": 4248
    },
    {
      "epoch": 404.6666666666667,
      "grad_norm": 5.49713659286499,
      "learning_rate": 7.959145427286356e-07,
      "loss": 1.0675,
      "step": 4249
    },
    {
      "epoch": 404.76190476190476,
      "grad_norm": 14.592151641845703,
      "learning_rate": 7.961019490254873e-07,
      "loss": 1.0971,
      "step": 4250
    },
    {
      "epoch": 404.85714285714283,
      "grad_norm": 16.717880249023438,
      "learning_rate": 7.962893553223388e-07,
      "loss": 1.1313,
      "step": 4251
    },
    {
      "epoch": 404.95238095238096,
      "grad_norm": 8.905058860778809,
      "learning_rate": 7.964767616191904e-07,
      "loss": 1.1062,
      "step": 4252
    },
    {
      "epoch": 404.95238095238096,
      "eval_accuracy": 0.7437242412888723,
      "eval_f1": 0.7926015767131595,
      "eval_loss": 0.5475287437438965,
      "eval_precision": 0.7451539338654504,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7942599309153713,
      "eval_runtime": 1.1444,
      "eval_samples_per_second": 2332.262,
      "eval_steps_per_second": 5.243,
      "step": 4252
    },
    {
      "epoch": 405.04761904761904,
      "grad_norm": 9.620675086975098,
      "learning_rate": 7.966641679160421e-07,
      "loss": 1.1391,
      "step": 4253
    },
    {
      "epoch": 405.14285714285717,
      "grad_norm": 8.57576847076416,
      "learning_rate": 7.968515742128936e-07,
      "loss": 1.1224,
      "step": 4254
    },
    {
      "epoch": 405.23809523809524,
      "grad_norm": 6.662265300750732,
      "learning_rate": 7.970389805097452e-07,
      "loss": 1.0391,
      "step": 4255
    },
    {
      "epoch": 405.3333333333333,
      "grad_norm": 11.818034172058105,
      "learning_rate": 7.972263868065967e-07,
      "loss": 1.0941,
      "step": 4256
    },
    {
      "epoch": 405.42857142857144,
      "grad_norm": 5.025628566741943,
      "learning_rate": 7.974137931034484e-07,
      "loss": 1.1243,
      "step": 4257
    },
    {
      "epoch": 405.5238095238095,
      "grad_norm": 10.131697654724121,
      "learning_rate": 7.976011994002998e-07,
      "loss": 1.0803,
      "step": 4258
    },
    {
      "epoch": 405.6190476190476,
      "grad_norm": 6.461528778076172,
      "learning_rate": 7.977886056971515e-07,
      "loss": 1.1283,
      "step": 4259
    },
    {
      "epoch": 405.7142857142857,
      "grad_norm": 10.693177223205566,
      "learning_rate": 7.979760119940032e-07,
      "loss": 1.0796,
      "step": 4260
    },
    {
      "epoch": 405.8095238095238,
      "grad_norm": 8.95005989074707,
      "learning_rate": 7.981634182908546e-07,
      "loss": 1.1086,
      "step": 4261
    },
    {
      "epoch": 405.9047619047619,
      "grad_norm": 10.91205883026123,
      "learning_rate": 7.983508245877063e-07,
      "loss": 1.1541,
      "step": 4262
    },
    {
      "epoch": 406.0,
      "grad_norm": 6.45164680480957,
      "learning_rate": 7.985382308845577e-07,
      "loss": 1.1022,
      "step": 4263
    },
    {
      "epoch": 406.0,
      "eval_accuracy": 0.7455976020981641,
      "eval_f1": 0.7938050409960522,
      "eval_loss": 0.5474799275398254,
      "eval_precision": 0.747284162378502,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7942668393782384,
      "eval_runtime": 1.148,
      "eval_samples_per_second": 2324.828,
      "eval_steps_per_second": 5.226,
      "step": 4263
    },
    {
      "epoch": 406.0952380952381,
      "grad_norm": 14.972555160522461,
      "learning_rate": 7.987256371814094e-07,
      "loss": 1.1507,
      "step": 4264
    },
    {
      "epoch": 406.1904761904762,
      "grad_norm": 16.497968673706055,
      "learning_rate": 7.989130434782608e-07,
      "loss": 1.1423,
      "step": 4265
    },
    {
      "epoch": 406.2857142857143,
      "grad_norm": 11.75539493560791,
      "learning_rate": 7.991004497751125e-07,
      "loss": 1.123,
      "step": 4266
    },
    {
      "epoch": 406.3809523809524,
      "grad_norm": 5.47906494140625,
      "learning_rate": 7.992878560719641e-07,
      "loss": 1.1137,
      "step": 4267
    },
    {
      "epoch": 406.4761904761905,
      "grad_norm": 9.136211395263672,
      "learning_rate": 7.994752623688156e-07,
      "loss": 1.1199,
      "step": 4268
    },
    {
      "epoch": 406.57142857142856,
      "grad_norm": 7.102169036865234,
      "learning_rate": 7.996626686656673e-07,
      "loss": 1.0876,
      "step": 4269
    },
    {
      "epoch": 406.6666666666667,
      "grad_norm": 16.571800231933594,
      "learning_rate": 7.998500749625187e-07,
      "loss": 1.0672,
      "step": 4270
    },
    {
      "epoch": 406.76190476190476,
      "grad_norm": 8.28835678100586,
      "learning_rate": 8.000374812593704e-07,
      "loss": 1.1015,
      "step": 4271
    },
    {
      "epoch": 406.85714285714283,
      "grad_norm": 10.805572509765625,
      "learning_rate": 8.002248875562218e-07,
      "loss": 1.0577,
      "step": 4272
    },
    {
      "epoch": 406.95238095238096,
      "grad_norm": 6.612346649169922,
      "learning_rate": 8.004122938530735e-07,
      "loss": 1.0792,
      "step": 4273
    },
    {
      "epoch": 406.95238095238096,
      "eval_accuracy": 0.7448482577744474,
      "eval_f1": 0.7915518824609734,
      "eval_loss": 0.5470107793807983,
      "eval_precision": 0.7504352872896112,
      "eval_recall": 0.8374352331606217,
      "eval_roc_auc": 0.7948126079447323,
      "eval_runtime": 1.1468,
      "eval_samples_per_second": 2327.419,
      "eval_steps_per_second": 5.232,
      "step": 4273
    },
    {
      "epoch": 407.04761904761904,
      "grad_norm": 7.979783535003662,
      "learning_rate": 8.005997001499251e-07,
      "loss": 1.106,
      "step": 4274
    },
    {
      "epoch": 407.14285714285717,
      "grad_norm": 5.28836727142334,
      "learning_rate": 8.007871064467767e-07,
      "loss": 1.0894,
      "step": 4275
    },
    {
      "epoch": 407.23809523809524,
      "grad_norm": 20.07858657836914,
      "learning_rate": 8.009745127436282e-07,
      "loss": 1.1641,
      "step": 4276
    },
    {
      "epoch": 407.3333333333333,
      "grad_norm": 7.941175937652588,
      "learning_rate": 8.011619190404798e-07,
      "loss": 1.0691,
      "step": 4277
    },
    {
      "epoch": 407.42857142857144,
      "grad_norm": 10.082222938537598,
      "learning_rate": 8.013493253373315e-07,
      "loss": 1.0746,
      "step": 4278
    },
    {
      "epoch": 407.5238095238095,
      "grad_norm": 16.410083770751953,
      "learning_rate": 8.015367316341829e-07,
      "loss": 1.1093,
      "step": 4279
    },
    {
      "epoch": 407.6190476190476,
      "grad_norm": 7.074474811553955,
      "learning_rate": 8.017241379310346e-07,
      "loss": 1.1178,
      "step": 4280
    },
    {
      "epoch": 407.7142857142857,
      "grad_norm": 4.897011756896973,
      "learning_rate": 8.019115442278862e-07,
      "loss": 1.1151,
      "step": 4281
    },
    {
      "epoch": 407.8095238095238,
      "grad_norm": 6.866818428039551,
      "learning_rate": 8.020989505247377e-07,
      "loss": 1.094,
      "step": 4282
    },
    {
      "epoch": 407.9047619047619,
      "grad_norm": 11.228714942932129,
      "learning_rate": 8.022863568215893e-07,
      "loss": 1.1042,
      "step": 4283
    },
    {
      "epoch": 408.0,
      "grad_norm": 18.426788330078125,
      "learning_rate": 8.024737631184408e-07,
      "loss": 1.1237,
      "step": 4284
    },
    {
      "epoch": 408.0,
      "eval_accuracy": 0.7470962907455976,
      "eval_f1": 0.7943953700883338,
      "eval_loss": 0.5461246371269226,
      "eval_precision": 0.7498562392179413,
      "eval_recall": 0.844559585492228,
      "eval_roc_auc": 0.7961093839953942,
      "eval_runtime": 1.1489,
      "eval_samples_per_second": 2323.052,
      "eval_steps_per_second": 5.222,
      "step": 4284
    },
    {
      "epoch": 408.0952380952381,
      "grad_norm": 9.016111373901367,
      "learning_rate": 8.026611694152924e-07,
      "loss": 1.0989,
      "step": 4285
    },
    {
      "epoch": 408.1904761904762,
      "grad_norm": 8.912129402160645,
      "learning_rate": 8.028485757121439e-07,
      "loss": 1.1514,
      "step": 4286
    },
    {
      "epoch": 408.2857142857143,
      "grad_norm": 8.221604347229004,
      "learning_rate": 8.030359820089955e-07,
      "loss": 1.0928,
      "step": 4287
    },
    {
      "epoch": 408.3809523809524,
      "grad_norm": 12.56308650970459,
      "learning_rate": 8.032233883058472e-07,
      "loss": 1.1431,
      "step": 4288
    },
    {
      "epoch": 408.4761904761905,
      "grad_norm": 11.391637802124023,
      "learning_rate": 8.034107946026987e-07,
      "loss": 1.0881,
      "step": 4289
    },
    {
      "epoch": 408.57142857142856,
      "grad_norm": 9.031922340393066,
      "learning_rate": 8.035982008995503e-07,
      "loss": 1.0857,
      "step": 4290
    },
    {
      "epoch": 408.6666666666667,
      "grad_norm": 16.605451583862305,
      "learning_rate": 8.037856071964018e-07,
      "loss": 1.0703,
      "step": 4291
    },
    {
      "epoch": 408.76190476190476,
      "grad_norm": 7.212552547454834,
      "learning_rate": 8.039730134932534e-07,
      "loss": 1.1281,
      "step": 4292
    },
    {
      "epoch": 408.85714285714283,
      "grad_norm": 6.889387607574463,
      "learning_rate": 8.041604197901049e-07,
      "loss": 1.0902,
      "step": 4293
    },
    {
      "epoch": 408.95238095238096,
      "grad_norm": 16.509017944335938,
      "learning_rate": 8.043478260869565e-07,
      "loss": 1.1056,
      "step": 4294
    },
    {
      "epoch": 408.95238095238096,
      "eval_accuracy": 0.7459722742600224,
      "eval_f1": 0.796028880866426,
      "eval_loss": 0.5461532473564148,
      "eval_precision": 0.7432584269662922,
      "eval_recall": 0.8568652849740933,
      "eval_roc_auc": 0.7970866436384572,
      "eval_runtime": 1.1196,
      "eval_samples_per_second": 2383.949,
      "eval_steps_per_second": 5.359,
      "step": 4294
    },
    {
      "epoch": 409.04761904761904,
      "grad_norm": 13.46784782409668,
      "learning_rate": 8.045352323838082e-07,
      "loss": 1.0993,
      "step": 4295
    },
    {
      "epoch": 409.14285714285717,
      "grad_norm": 9.369694709777832,
      "learning_rate": 8.047226386806596e-07,
      "loss": 1.0908,
      "step": 4296
    },
    {
      "epoch": 409.23809523809524,
      "grad_norm": 10.482138633728027,
      "learning_rate": 8.049100449775113e-07,
      "loss": 1.0897,
      "step": 4297
    },
    {
      "epoch": 409.3333333333333,
      "grad_norm": 4.546848773956299,
      "learning_rate": 8.050974512743629e-07,
      "loss": 1.1082,
      "step": 4298
    },
    {
      "epoch": 409.42857142857144,
      "grad_norm": 8.517866134643555,
      "learning_rate": 8.052848575712144e-07,
      "loss": 1.1483,
      "step": 4299
    },
    {
      "epoch": 409.5238095238095,
      "grad_norm": 7.181638717651367,
      "learning_rate": 8.05472263868066e-07,
      "loss": 1.0889,
      "step": 4300
    },
    {
      "epoch": 409.6190476190476,
      "grad_norm": 7.7330403327941895,
      "learning_rate": 8.056596701649176e-07,
      "loss": 1.1655,
      "step": 4301
    },
    {
      "epoch": 409.7142857142857,
      "grad_norm": 14.187042236328125,
      "learning_rate": 8.058470764617692e-07,
      "loss": 1.1368,
      "step": 4302
    },
    {
      "epoch": 409.8095238095238,
      "grad_norm": 16.997217178344727,
      "learning_rate": 8.060344827586207e-07,
      "loss": 1.0779,
      "step": 4303
    },
    {
      "epoch": 409.9047619047619,
      "grad_norm": 15.993823051452637,
      "learning_rate": 8.062218890554724e-07,
      "loss": 1.1321,
      "step": 4304
    },
    {
      "epoch": 410.0,
      "grad_norm": 7.269989013671875,
      "learning_rate": 8.064092953523238e-07,
      "loss": 1.1147,
      "step": 4305
    },
    {
      "epoch": 410.0,
      "eval_accuracy": 0.738478831022855,
      "eval_f1": 0.7837670384138785,
      "eval_loss": 0.545691728591919,
      "eval_precision": 0.751187648456057,
      "eval_recall": 0.819300518134715,
      "eval_roc_auc": 0.7966246401842256,
      "eval_runtime": 1.1265,
      "eval_samples_per_second": 2369.19,
      "eval_steps_per_second": 5.326,
      "step": 4305
    },
    {
      "epoch": 410.0952380952381,
      "grad_norm": 14.141743659973145,
      "learning_rate": 8.065967016491755e-07,
      "loss": 1.0654,
      "step": 4306
    },
    {
      "epoch": 410.1904761904762,
      "grad_norm": 16.728107452392578,
      "learning_rate": 8.06784107946027e-07,
      "loss": 1.0393,
      "step": 4307
    },
    {
      "epoch": 410.2857142857143,
      "grad_norm": 13.319567680358887,
      "learning_rate": 8.069715142428786e-07,
      "loss": 1.1547,
      "step": 4308
    },
    {
      "epoch": 410.3809523809524,
      "grad_norm": 12.622332572937012,
      "learning_rate": 8.071589205397303e-07,
      "loss": 1.0834,
      "step": 4309
    },
    {
      "epoch": 410.4761904761905,
      "grad_norm": 15.735387802124023,
      "learning_rate": 8.073463268365817e-07,
      "loss": 1.1014,
      "step": 4310
    },
    {
      "epoch": 410.57142857142856,
      "grad_norm": 16.525028228759766,
      "learning_rate": 8.075337331334334e-07,
      "loss": 1.0989,
      "step": 4311
    },
    {
      "epoch": 410.6666666666667,
      "grad_norm": 17.96881866455078,
      "learning_rate": 8.077211394302848e-07,
      "loss": 1.1008,
      "step": 4312
    },
    {
      "epoch": 410.76190476190476,
      "grad_norm": 6.438241958618164,
      "learning_rate": 8.079085457271365e-07,
      "loss": 1.1317,
      "step": 4313
    },
    {
      "epoch": 410.85714285714283,
      "grad_norm": 5.570372581481934,
      "learning_rate": 8.080959520239879e-07,
      "loss": 1.1357,
      "step": 4314
    },
    {
      "epoch": 410.95238095238096,
      "grad_norm": 11.380789756774902,
      "learning_rate": 8.082833583208396e-07,
      "loss": 1.0883,
      "step": 4315
    },
    {
      "epoch": 410.95238095238096,
      "eval_accuracy": 0.7381041588609967,
      "eval_f1": 0.7823108066023046,
      "eval_loss": 0.5462737679481506,
      "eval_precision": 0.7534493101379725,
      "eval_recall": 0.8134715025906736,
      "eval_roc_auc": 0.7960748416810592,
      "eval_runtime": 1.1341,
      "eval_samples_per_second": 2353.328,
      "eval_steps_per_second": 5.29,
      "step": 4315
    },
    {
      "epoch": 411.04761904761904,
      "grad_norm": 8.623936653137207,
      "learning_rate": 8.084707646176913e-07,
      "loss": 1.1474,
      "step": 4316
    },
    {
      "epoch": 411.14285714285717,
      "grad_norm": 19.895549774169922,
      "learning_rate": 8.086581709145427e-07,
      "loss": 1.1781,
      "step": 4317
    },
    {
      "epoch": 411.23809523809524,
      "grad_norm": 5.159582614898682,
      "learning_rate": 8.088455772113944e-07,
      "loss": 1.1296,
      "step": 4318
    },
    {
      "epoch": 411.3333333333333,
      "grad_norm": 23.74869155883789,
      "learning_rate": 8.090329835082458e-07,
      "loss": 1.1161,
      "step": 4319
    },
    {
      "epoch": 411.42857142857144,
      "grad_norm": 5.957592964172363,
      "learning_rate": 8.092203898050975e-07,
      "loss": 1.1283,
      "step": 4320
    },
    {
      "epoch": 411.5238095238095,
      "grad_norm": 21.57670021057129,
      "learning_rate": 8.09407796101949e-07,
      "loss": 1.0866,
      "step": 4321
    },
    {
      "epoch": 411.6190476190476,
      "grad_norm": 9.695438385009766,
      "learning_rate": 8.095952023988006e-07,
      "loss": 1.0912,
      "step": 4322
    },
    {
      "epoch": 411.7142857142857,
      "grad_norm": 7.644775390625,
      "learning_rate": 8.097826086956523e-07,
      "loss": 1.0915,
      "step": 4323
    },
    {
      "epoch": 411.8095238095238,
      "grad_norm": 12.165261268615723,
      "learning_rate": 8.099700149925038e-07,
      "loss": 1.1117,
      "step": 4324
    },
    {
      "epoch": 411.9047619047619,
      "grad_norm": 10.229143142700195,
      "learning_rate": 8.101574212893555e-07,
      "loss": 1.0971,
      "step": 4325
    },
    {
      "epoch": 412.0,
      "grad_norm": 10.323266983032227,
      "learning_rate": 8.103448275862069e-07,
      "loss": 1.0797,
      "step": 4326
    },
    {
      "epoch": 412.0,
      "eval_accuracy": 0.7426002248032971,
      "eval_f1": 0.7890696960393,
      "eval_loss": 0.5459049344062805,
      "eval_precision": 0.7501459427904261,
      "eval_recall": 0.8322538860103627,
      "eval_roc_auc": 0.7969559585492227,
      "eval_runtime": 1.1416,
      "eval_samples_per_second": 2337.93,
      "eval_steps_per_second": 5.256,
      "step": 4326
    },
    {
      "epoch": 412.0952380952381,
      "grad_norm": 8.321845054626465,
      "learning_rate": 8.105322338830586e-07,
      "loss": 1.1491,
      "step": 4327
    },
    {
      "epoch": 412.1904761904762,
      "grad_norm": 7.4247941970825195,
      "learning_rate": 8.1071964017991e-07,
      "loss": 1.138,
      "step": 4328
    },
    {
      "epoch": 412.2857142857143,
      "grad_norm": 19.407428741455078,
      "learning_rate": 8.109070464767617e-07,
      "loss": 1.0456,
      "step": 4329
    },
    {
      "epoch": 412.3809523809524,
      "grad_norm": 7.326526641845703,
      "learning_rate": 8.110944527736133e-07,
      "loss": 1.1331,
      "step": 4330
    },
    {
      "epoch": 412.4761904761905,
      "grad_norm": 7.689526081085205,
      "learning_rate": 8.112818590704648e-07,
      "loss": 1.1223,
      "step": 4331
    },
    {
      "epoch": 412.57142857142856,
      "grad_norm": 24.991323471069336,
      "learning_rate": 8.114692653673165e-07,
      "loss": 1.1007,
      "step": 4332
    },
    {
      "epoch": 412.6666666666667,
      "grad_norm": 9.8153076171875,
      "learning_rate": 8.116566716641679e-07,
      "loss": 1.1181,
      "step": 4333
    },
    {
      "epoch": 412.76190476190476,
      "grad_norm": 16.506303787231445,
      "learning_rate": 8.118440779610196e-07,
      "loss": 1.1094,
      "step": 4334
    },
    {
      "epoch": 412.85714285714283,
      "grad_norm": 11.549041748046875,
      "learning_rate": 8.12031484257871e-07,
      "loss": 1.0732,
      "step": 4335
    },
    {
      "epoch": 412.95238095238096,
      "grad_norm": 16.197805404663086,
      "learning_rate": 8.122188905547227e-07,
      "loss": 1.1104,
      "step": 4336
    },
    {
      "epoch": 412.95238095238096,
      "eval_accuracy": 0.7470962907455976,
      "eval_f1": 0.796870297923563,
      "eval_loss": 0.5461461544036865,
      "eval_precision": 0.7442383361439011,
      "eval_recall": 0.8575129533678757,
      "eval_roc_auc": 0.7971735751295337,
      "eval_runtime": 1.1432,
      "eval_samples_per_second": 2334.652,
      "eval_steps_per_second": 5.248,
      "step": 4336
    },
    {
      "epoch": 413.04761904761904,
      "grad_norm": 13.857231140136719,
      "learning_rate": 8.124062968515743e-07,
      "loss": 1.1105,
      "step": 4337
    },
    {
      "epoch": 413.14285714285717,
      "grad_norm": 16.94660758972168,
      "learning_rate": 8.125937031484258e-07,
      "loss": 1.1081,
      "step": 4338
    },
    {
      "epoch": 413.23809523809524,
      "grad_norm": 6.674830436706543,
      "learning_rate": 8.127811094452774e-07,
      "loss": 1.1017,
      "step": 4339
    },
    {
      "epoch": 413.3333333333333,
      "grad_norm": 9.114222526550293,
      "learning_rate": 8.129685157421289e-07,
      "loss": 1.0578,
      "step": 4340
    },
    {
      "epoch": 413.42857142857144,
      "grad_norm": 7.696568012237549,
      "learning_rate": 8.131559220389805e-07,
      "loss": 1.0933,
      "step": 4341
    },
    {
      "epoch": 413.5238095238095,
      "grad_norm": 11.424352645874023,
      "learning_rate": 8.133433283358321e-07,
      "loss": 1.1116,
      "step": 4342
    },
    {
      "epoch": 413.6190476190476,
      "grad_norm": 29.731801986694336,
      "learning_rate": 8.135307346326837e-07,
      "loss": 1.1092,
      "step": 4343
    },
    {
      "epoch": 413.7142857142857,
      "grad_norm": 15.900996208190918,
      "learning_rate": 8.137181409295353e-07,
      "loss": 1.0991,
      "step": 4344
    },
    {
      "epoch": 413.8095238095238,
      "grad_norm": 11.894193649291992,
      "learning_rate": 8.139055472263869e-07,
      "loss": 1.1348,
      "step": 4345
    },
    {
      "epoch": 413.9047619047619,
      "grad_norm": 5.878718376159668,
      "learning_rate": 8.140929535232384e-07,
      "loss": 1.1219,
      "step": 4346
    },
    {
      "epoch": 414.0,
      "grad_norm": 7.070927619934082,
      "learning_rate": 8.1428035982009e-07,
      "loss": 1.0884,
      "step": 4347
    },
    {
      "epoch": 414.0,
      "eval_accuracy": 0.7437242412888723,
      "eval_f1": 0.7934782608695652,
      "eval_loss": 0.5461050868034363,
      "eval_precision": 0.7432126696832579,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.7969974093264248,
      "eval_runtime": 1.1418,
      "eval_samples_per_second": 2337.604,
      "eval_steps_per_second": 5.255,
      "step": 4347
    },
    {
      "epoch": 414.0952380952381,
      "grad_norm": 8.964851379394531,
      "learning_rate": 8.144677661169416e-07,
      "loss": 1.1322,
      "step": 4348
    },
    {
      "epoch": 414.1904761904762,
      "grad_norm": 7.7236714363098145,
      "learning_rate": 8.146551724137931e-07,
      "loss": 1.0882,
      "step": 4349
    },
    {
      "epoch": 414.2857142857143,
      "grad_norm": 14.21401596069336,
      "learning_rate": 8.148425787106448e-07,
      "loss": 1.1519,
      "step": 4350
    },
    {
      "epoch": 414.3809523809524,
      "grad_norm": 11.415388107299805,
      "learning_rate": 8.150299850074964e-07,
      "loss": 1.0953,
      "step": 4351
    },
    {
      "epoch": 414.4761904761905,
      "grad_norm": 14.541751861572266,
      "learning_rate": 8.152173913043479e-07,
      "loss": 1.1454,
      "step": 4352
    },
    {
      "epoch": 414.57142857142856,
      "grad_norm": 10.25801944732666,
      "learning_rate": 8.154047976011995e-07,
      "loss": 1.1093,
      "step": 4353
    },
    {
      "epoch": 414.6666666666667,
      "grad_norm": 10.318373680114746,
      "learning_rate": 8.15592203898051e-07,
      "loss": 1.0564,
      "step": 4354
    },
    {
      "epoch": 414.76190476190476,
      "grad_norm": 6.321846961975098,
      "learning_rate": 8.157796101949026e-07,
      "loss": 1.0697,
      "step": 4355
    },
    {
      "epoch": 414.85714285714283,
      "grad_norm": 12.549822807312012,
      "learning_rate": 8.159670164917541e-07,
      "loss": 1.1069,
      "step": 4356
    },
    {
      "epoch": 414.95238095238096,
      "grad_norm": 7.117936134338379,
      "learning_rate": 8.161544227886057e-07,
      "loss": 1.149,
      "step": 4357
    },
    {
      "epoch": 414.95238095238096,
      "eval_accuracy": 0.7403521918321468,
      "eval_f1": 0.7841793833696667,
      "eval_loss": 0.5456610918045044,
      "eval_precision": 0.755248950209958,
      "eval_recall": 0.8154145077720207,
      "eval_roc_auc": 0.7966873920552677,
      "eval_runtime": 1.124,
      "eval_samples_per_second": 2374.473,
      "eval_steps_per_second": 5.338,
      "step": 4357
    },
    {
      "epoch": 415.04761904761904,
      "grad_norm": 11.925171852111816,
      "learning_rate": 8.163418290854574e-07,
      "loss": 1.1399,
      "step": 4358
    },
    {
      "epoch": 415.14285714285717,
      "grad_norm": 7.809358596801758,
      "learning_rate": 8.165292353823088e-07,
      "loss": 1.1266,
      "step": 4359
    },
    {
      "epoch": 415.23809523809524,
      "grad_norm": 8.68516731262207,
      "learning_rate": 8.167166416791605e-07,
      "loss": 1.1213,
      "step": 4360
    },
    {
      "epoch": 415.3333333333333,
      "grad_norm": 12.828359603881836,
      "learning_rate": 8.16904047976012e-07,
      "loss": 1.1063,
      "step": 4361
    },
    {
      "epoch": 415.42857142857144,
      "grad_norm": 9.938231468200684,
      "learning_rate": 8.170914542728636e-07,
      "loss": 1.1034,
      "step": 4362
    },
    {
      "epoch": 415.5238095238095,
      "grad_norm": 14.429758071899414,
      "learning_rate": 8.172788605697151e-07,
      "loss": 1.1455,
      "step": 4363
    },
    {
      "epoch": 415.6190476190476,
      "grad_norm": 6.0715718269348145,
      "learning_rate": 8.174662668665667e-07,
      "loss": 1.1066,
      "step": 4364
    },
    {
      "epoch": 415.7142857142857,
      "grad_norm": 10.693117141723633,
      "learning_rate": 8.176536731634184e-07,
      "loss": 1.0846,
      "step": 4365
    },
    {
      "epoch": 415.8095238095238,
      "grad_norm": 6.945085525512695,
      "learning_rate": 8.178410794602698e-07,
      "loss": 1.0902,
      "step": 4366
    },
    {
      "epoch": 415.9047619047619,
      "grad_norm": 7.514898777008057,
      "learning_rate": 8.180284857571215e-07,
      "loss": 1.0867,
      "step": 4367
    },
    {
      "epoch": 416.0,
      "grad_norm": 5.8616509437561035,
      "learning_rate": 8.18215892053973e-07,
      "loss": 1.0946,
      "step": 4368
    },
    {
      "epoch": 416.0,
      "eval_accuracy": 0.741476208317722,
      "eval_f1": 0.7889908256880734,
      "eval_loss": 0.5456025004386902,
      "eval_precision": 0.7473928157589803,
      "eval_recall": 0.8354922279792746,
      "eval_roc_auc": 0.7965768566493956,
      "eval_runtime": 1.1384,
      "eval_samples_per_second": 2344.428,
      "eval_steps_per_second": 5.27,
      "step": 4368
    },
    {
      "epoch": 416.0952380952381,
      "grad_norm": 4.330708980560303,
      "learning_rate": 8.184032983508246e-07,
      "loss": 1.1099,
      "step": 4369
    },
    {
      "epoch": 416.1904761904762,
      "grad_norm": 8.46364688873291,
      "learning_rate": 8.185907046476762e-07,
      "loss": 1.1266,
      "step": 4370
    },
    {
      "epoch": 416.2857142857143,
      "grad_norm": 14.600200653076172,
      "learning_rate": 8.187781109445278e-07,
      "loss": 1.0867,
      "step": 4371
    },
    {
      "epoch": 416.3809523809524,
      "grad_norm": 3.619706392288208,
      "learning_rate": 8.189655172413794e-07,
      "loss": 1.1126,
      "step": 4372
    },
    {
      "epoch": 416.4761904761905,
      "grad_norm": 6.472954750061035,
      "learning_rate": 8.191529235382309e-07,
      "loss": 1.1137,
      "step": 4373
    },
    {
      "epoch": 416.57142857142856,
      "grad_norm": 9.510734558105469,
      "learning_rate": 8.193403298350826e-07,
      "loss": 1.1469,
      "step": 4374
    },
    {
      "epoch": 416.6666666666667,
      "grad_norm": 15.471990585327148,
      "learning_rate": 8.19527736131934e-07,
      "loss": 1.0597,
      "step": 4375
    },
    {
      "epoch": 416.76190476190476,
      "grad_norm": 6.931447505950928,
      "learning_rate": 8.197151424287857e-07,
      "loss": 1.1144,
      "step": 4376
    },
    {
      "epoch": 416.85714285714283,
      "grad_norm": 10.197463035583496,
      "learning_rate": 8.199025487256371e-07,
      "loss": 1.0994,
      "step": 4377
    },
    {
      "epoch": 416.95238095238096,
      "grad_norm": 5.985307693481445,
      "learning_rate": 8.200899550224888e-07,
      "loss": 1.0616,
      "step": 4378
    },
    {
      "epoch": 416.95238095238096,
      "eval_accuracy": 0.7426002248032971,
      "eval_f1": 0.7894575543977934,
      "eval_loss": 0.5450717210769653,
      "eval_precision": 0.7492728330424665,
      "eval_recall": 0.8341968911917098,
      "eval_roc_auc": 0.7973819804260218,
      "eval_runtime": 1.1568,
      "eval_samples_per_second": 2307.26,
      "eval_steps_per_second": 5.187,
      "step": 4378
    },
    {
      "epoch": 417.04761904761904,
      "grad_norm": 6.391446113586426,
      "learning_rate": 8.202773613193405e-07,
      "loss": 1.0323,
      "step": 4379
    },
    {
      "epoch": 417.14285714285717,
      "grad_norm": 12.014970779418945,
      "learning_rate": 8.204647676161919e-07,
      "loss": 1.1424,
      "step": 4380
    },
    {
      "epoch": 417.23809523809524,
      "grad_norm": 10.252963066101074,
      "learning_rate": 8.206521739130436e-07,
      "loss": 1.1437,
      "step": 4381
    },
    {
      "epoch": 417.3333333333333,
      "grad_norm": 10.97370433807373,
      "learning_rate": 8.20839580209895e-07,
      "loss": 1.0734,
      "step": 4382
    },
    {
      "epoch": 417.42857142857144,
      "grad_norm": 5.32800817489624,
      "learning_rate": 8.210269865067467e-07,
      "loss": 1.137,
      "step": 4383
    },
    {
      "epoch": 417.5238095238095,
      "grad_norm": 9.901985168457031,
      "learning_rate": 8.212143928035981e-07,
      "loss": 1.1143,
      "step": 4384
    },
    {
      "epoch": 417.6190476190476,
      "grad_norm": 13.600570678710938,
      "learning_rate": 8.214017991004498e-07,
      "loss": 1.101,
      "step": 4385
    },
    {
      "epoch": 417.7142857142857,
      "grad_norm": 6.229227066040039,
      "learning_rate": 8.215892053973015e-07,
      "loss": 1.1173,
      "step": 4386
    },
    {
      "epoch": 417.8095238095238,
      "grad_norm": 11.53397274017334,
      "learning_rate": 8.217766116941529e-07,
      "loss": 1.0949,
      "step": 4387
    },
    {
      "epoch": 417.9047619047619,
      "grad_norm": 8.325032234191895,
      "learning_rate": 8.219640179910046e-07,
      "loss": 1.1058,
      "step": 4388
    },
    {
      "epoch": 418.0,
      "grad_norm": 11.090065002441406,
      "learning_rate": 8.22151424287856e-07,
      "loss": 1.0834,
      "step": 4389
    },
    {
      "epoch": 418.0,
      "eval_accuracy": 0.741476208317722,
      "eval_f1": 0.7866419294990723,
      "eval_loss": 0.545009434223175,
      "eval_precision": 0.7526627218934911,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.7973802533103053,
      "eval_runtime": 1.1366,
      "eval_samples_per_second": 2348.205,
      "eval_steps_per_second": 5.279,
      "step": 4389
    },
    {
      "epoch": 418.0952380952381,
      "grad_norm": 10.562765121459961,
      "learning_rate": 8.223388305847077e-07,
      "loss": 1.099,
      "step": 4390
    },
    {
      "epoch": 418.1904761904762,
      "grad_norm": 10.48797607421875,
      "learning_rate": 8.225262368815592e-07,
      "loss": 1.1413,
      "step": 4391
    },
    {
      "epoch": 418.2857142857143,
      "grad_norm": 11.87863540649414,
      "learning_rate": 8.227136431784109e-07,
      "loss": 1.0856,
      "step": 4392
    },
    {
      "epoch": 418.3809523809524,
      "grad_norm": 16.91488265991211,
      "learning_rate": 8.229010494752624e-07,
      "loss": 1.0922,
      "step": 4393
    },
    {
      "epoch": 418.4761904761905,
      "grad_norm": 12.223773956298828,
      "learning_rate": 8.23088455772114e-07,
      "loss": 1.1175,
      "step": 4394
    },
    {
      "epoch": 418.57142857142856,
      "grad_norm": 10.999683380126953,
      "learning_rate": 8.232758620689657e-07,
      "loss": 1.0696,
      "step": 4395
    },
    {
      "epoch": 418.6666666666667,
      "grad_norm": 6.884599685668945,
      "learning_rate": 8.234632683658171e-07,
      "loss": 1.1307,
      "step": 4396
    },
    {
      "epoch": 418.76190476190476,
      "grad_norm": 9.62916374206543,
      "learning_rate": 8.236506746626688e-07,
      "loss": 1.0341,
      "step": 4397
    },
    {
      "epoch": 418.85714285714283,
      "grad_norm": 11.067511558532715,
      "learning_rate": 8.238380809595202e-07,
      "loss": 1.1207,
      "step": 4398
    },
    {
      "epoch": 418.95238095238096,
      "grad_norm": 12.294548988342285,
      "learning_rate": 8.240254872563719e-07,
      "loss": 1.1302,
      "step": 4399
    },
    {
      "epoch": 418.95238095238096,
      "eval_accuracy": 0.7429748969651555,
      "eval_f1": 0.7911084043848965,
      "eval_loss": 0.5447846055030823,
      "eval_precision": 0.746551724137931,
      "eval_recall": 0.841321243523316,
      "eval_roc_auc": 0.798289291882556,
      "eval_runtime": 1.1598,
      "eval_samples_per_second": 2301.243,
      "eval_steps_per_second": 5.173,
      "step": 4399
    },
    {
      "epoch": 419.04761904761904,
      "grad_norm": 11.96934700012207,
      "learning_rate": 8.242128935532235e-07,
      "loss": 1.0869,
      "step": 4400
    },
    {
      "epoch": 419.14285714285717,
      "grad_norm": 7.476302146911621,
      "learning_rate": 8.24400299850075e-07,
      "loss": 1.0941,
      "step": 4401
    },
    {
      "epoch": 419.23809523809524,
      "grad_norm": 6.855079174041748,
      "learning_rate": 8.245877061469266e-07,
      "loss": 1.0963,
      "step": 4402
    },
    {
      "epoch": 419.3333333333333,
      "grad_norm": 8.352704048156738,
      "learning_rate": 8.247751124437781e-07,
      "loss": 1.1144,
      "step": 4403
    },
    {
      "epoch": 419.42857142857144,
      "grad_norm": 9.372175216674805,
      "learning_rate": 8.249625187406298e-07,
      "loss": 1.0919,
      "step": 4404
    },
    {
      "epoch": 419.5238095238095,
      "grad_norm": 13.681797981262207,
      "learning_rate": 8.251499250374812e-07,
      "loss": 1.1232,
      "step": 4405
    },
    {
      "epoch": 419.6190476190476,
      "grad_norm": 5.431125640869141,
      "learning_rate": 8.253373313343329e-07,
      "loss": 1.1242,
      "step": 4406
    },
    {
      "epoch": 419.7142857142857,
      "grad_norm": 5.640235424041748,
      "learning_rate": 8.255247376311845e-07,
      "loss": 1.1382,
      "step": 4407
    },
    {
      "epoch": 419.8095238095238,
      "grad_norm": 11.201006889343262,
      "learning_rate": 8.25712143928036e-07,
      "loss": 1.0872,
      "step": 4408
    },
    {
      "epoch": 419.9047619047619,
      "grad_norm": 10.526140213012695,
      "learning_rate": 8.258995502248876e-07,
      "loss": 1.112,
      "step": 4409
    },
    {
      "epoch": 420.0,
      "grad_norm": 9.951897621154785,
      "learning_rate": 8.260869565217391e-07,
      "loss": 1.1077,
      "step": 4410
    },
    {
      "epoch": 420.0,
      "eval_accuracy": 0.7437242412888723,
      "eval_f1": 0.7923497267759563,
      "eval_loss": 0.5443801879882812,
      "eval_precision": 0.7457142857142857,
      "eval_recall": 0.8452072538860104,
      "eval_roc_auc": 0.7985854922279794,
      "eval_runtime": 1.1406,
      "eval_samples_per_second": 2339.901,
      "eval_steps_per_second": 5.26,
      "step": 4410
    },
    {
      "epoch": 420.0952380952381,
      "grad_norm": 4.912606716156006,
      "learning_rate": 8.262743628185907e-07,
      "loss": 1.0917,
      "step": 4411
    },
    {
      "epoch": 420.1904761904762,
      "grad_norm": 11.934670448303223,
      "learning_rate": 8.264617691154423e-07,
      "loss": 1.1055,
      "step": 4412
    },
    {
      "epoch": 420.2857142857143,
      "grad_norm": 14.591500282287598,
      "learning_rate": 8.266491754122938e-07,
      "loss": 1.1455,
      "step": 4413
    },
    {
      "epoch": 420.3809523809524,
      "grad_norm": 7.827000617980957,
      "learning_rate": 8.268365817091455e-07,
      "loss": 1.0873,
      "step": 4414
    },
    {
      "epoch": 420.4761904761905,
      "grad_norm": 8.400473594665527,
      "learning_rate": 8.270239880059971e-07,
      "loss": 1.0894,
      "step": 4415
    },
    {
      "epoch": 420.57142857142856,
      "grad_norm": 7.976455211639404,
      "learning_rate": 8.272113943028486e-07,
      "loss": 1.0916,
      "step": 4416
    },
    {
      "epoch": 420.6666666666667,
      "grad_norm": 9.351739883422852,
      "learning_rate": 8.273988005997002e-07,
      "loss": 1.1784,
      "step": 4417
    },
    {
      "epoch": 420.76190476190476,
      "grad_norm": 17.383878707885742,
      "learning_rate": 8.275862068965518e-07,
      "loss": 1.0647,
      "step": 4418
    },
    {
      "epoch": 420.85714285714283,
      "grad_norm": 14.991252899169922,
      "learning_rate": 8.277736131934033e-07,
      "loss": 1.125,
      "step": 4419
    },
    {
      "epoch": 420.95238095238096,
      "grad_norm": 31.164175033569336,
      "learning_rate": 8.279610194902549e-07,
      "loss": 1.1136,
      "step": 4420
    },
    {
      "epoch": 420.95238095238096,
      "eval_accuracy": 0.7347321094042712,
      "eval_f1": 0.776937618147448,
      "eval_loss": 0.5451502203941345,
      "eval_precision": 0.756441717791411,
      "eval_recall": 0.7985751295336787,
      "eval_roc_auc": 0.797415371329879,
      "eval_runtime": 1.1407,
      "eval_samples_per_second": 2339.842,
      "eval_steps_per_second": 5.26,
      "step": 4420
    },
    {
      "epoch": 421.04761904761904,
      "grad_norm": 17.54259490966797,
      "learning_rate": 8.281484257871066e-07,
      "loss": 1.1021,
      "step": 4421
    },
    {
      "epoch": 421.14285714285717,
      "grad_norm": 6.919532299041748,
      "learning_rate": 8.28335832083958e-07,
      "loss": 1.1348,
      "step": 4422
    },
    {
      "epoch": 421.23809523809524,
      "grad_norm": 8.13316822052002,
      "learning_rate": 8.285232383808097e-07,
      "loss": 1.1717,
      "step": 4423
    },
    {
      "epoch": 421.3333333333333,
      "grad_norm": 14.149221420288086,
      "learning_rate": 8.287106446776612e-07,
      "loss": 1.1057,
      "step": 4424
    },
    {
      "epoch": 421.42857142857144,
      "grad_norm": 12.222851753234863,
      "learning_rate": 8.288980509745128e-07,
      "loss": 1.1559,
      "step": 4425
    },
    {
      "epoch": 421.5238095238095,
      "grad_norm": 13.207738876342773,
      "learning_rate": 8.290854572713643e-07,
      "loss": 1.0409,
      "step": 4426
    },
    {
      "epoch": 421.6190476190476,
      "grad_norm": 10.126866340637207,
      "learning_rate": 8.292728635682159e-07,
      "loss": 1.0834,
      "step": 4427
    },
    {
      "epoch": 421.7142857142857,
      "grad_norm": 12.194635391235352,
      "learning_rate": 8.294602698650676e-07,
      "loss": 1.1281,
      "step": 4428
    },
    {
      "epoch": 421.8095238095238,
      "grad_norm": 11.794801712036133,
      "learning_rate": 8.29647676161919e-07,
      "loss": 1.0945,
      "step": 4429
    },
    {
      "epoch": 421.9047619047619,
      "grad_norm": 4.966232776641846,
      "learning_rate": 8.298350824587707e-07,
      "loss": 1.0807,
      "step": 4430
    },
    {
      "epoch": 422.0,
      "grad_norm": 12.192557334899902,
      "learning_rate": 8.300224887556221e-07,
      "loss": 1.0746,
      "step": 4431
    },
    {
      "epoch": 422.0,
      "eval_accuracy": 0.744473585612589,
      "eval_f1": 0.7902829028290282,
      "eval_loss": 0.5442536473274231,
      "eval_precision": 0.7523419203747073,
      "eval_recall": 0.8322538860103627,
      "eval_roc_auc": 0.7985938399539436,
      "eval_runtime": 1.1093,
      "eval_samples_per_second": 2406.062,
      "eval_steps_per_second": 5.409,
      "step": 4431
    },
    {
      "epoch": 422.0952380952381,
      "grad_norm": 10.80482292175293,
      "learning_rate": 8.302098950524738e-07,
      "loss": 1.1118,
      "step": 4432
    },
    {
      "epoch": 422.1904761904762,
      "grad_norm": 7.838009357452393,
      "learning_rate": 8.303973013493253e-07,
      "loss": 1.1194,
      "step": 4433
    },
    {
      "epoch": 422.2857142857143,
      "grad_norm": 10.171510696411133,
      "learning_rate": 8.305847076461769e-07,
      "loss": 1.068,
      "step": 4434
    },
    {
      "epoch": 422.3809523809524,
      "grad_norm": 6.738454818725586,
      "learning_rate": 8.307721139430286e-07,
      "loss": 1.1022,
      "step": 4435
    },
    {
      "epoch": 422.4761904761905,
      "grad_norm": 14.316563606262207,
      "learning_rate": 8.3095952023988e-07,
      "loss": 1.0298,
      "step": 4436
    },
    {
      "epoch": 422.57142857142856,
      "grad_norm": 16.27669334411621,
      "learning_rate": 8.311469265367317e-07,
      "loss": 1.0936,
      "step": 4437
    },
    {
      "epoch": 422.6666666666667,
      "grad_norm": 14.908007621765137,
      "learning_rate": 8.313343328335832e-07,
      "loss": 1.1261,
      "step": 4438
    },
    {
      "epoch": 422.76190476190476,
      "grad_norm": 6.022855281829834,
      "learning_rate": 8.315217391304348e-07,
      "loss": 1.0819,
      "step": 4439
    },
    {
      "epoch": 422.85714285714283,
      "grad_norm": 21.759950637817383,
      "learning_rate": 8.317091454272863e-07,
      "loss": 1.1399,
      "step": 4440
    },
    {
      "epoch": 422.95238095238096,
      "grad_norm": 15.305691719055176,
      "learning_rate": 8.31896551724138e-07,
      "loss": 1.0986,
      "step": 4441
    },
    {
      "epoch": 422.95238095238096,
      "eval_accuracy": 0.7459722742600224,
      "eval_f1": 0.79491833030853,
      "eval_loss": 0.5446916818618774,
      "eval_precision": 0.7457434733257662,
      "eval_recall": 0.8510362694300518,
      "eval_roc_auc": 0.797884571099597,
      "eval_runtime": 1.1207,
      "eval_samples_per_second": 2381.599,
      "eval_steps_per_second": 5.354,
      "step": 4441
    },
    {
      "epoch": 423.04761904761904,
      "grad_norm": 10.44428825378418,
      "learning_rate": 8.320839580209896e-07,
      "loss": 1.0861,
      "step": 4442
    },
    {
      "epoch": 423.14285714285717,
      "grad_norm": 9.53567886352539,
      "learning_rate": 8.322713643178411e-07,
      "loss": 1.0823,
      "step": 4443
    },
    {
      "epoch": 423.23809523809524,
      "grad_norm": 8.419621467590332,
      "learning_rate": 8.324587706146928e-07,
      "loss": 1.1237,
      "step": 4444
    },
    {
      "epoch": 423.3333333333333,
      "grad_norm": 10.934845924377441,
      "learning_rate": 8.326461769115442e-07,
      "loss": 1.1218,
      "step": 4445
    },
    {
      "epoch": 423.42857142857144,
      "grad_norm": 14.968563079833984,
      "learning_rate": 8.328335832083959e-07,
      "loss": 1.0985,
      "step": 4446
    },
    {
      "epoch": 423.5238095238095,
      "grad_norm": 7.92880392074585,
      "learning_rate": 8.330209895052473e-07,
      "loss": 1.1011,
      "step": 4447
    },
    {
      "epoch": 423.6190476190476,
      "grad_norm": 18.306106567382812,
      "learning_rate": 8.33208395802099e-07,
      "loss": 1.1,
      "step": 4448
    },
    {
      "epoch": 423.7142857142857,
      "grad_norm": 9.438417434692383,
      "learning_rate": 8.333958020989507e-07,
      "loss": 1.0866,
      "step": 4449
    },
    {
      "epoch": 423.8095238095238,
      "grad_norm": 9.805127143859863,
      "learning_rate": 8.335832083958021e-07,
      "loss": 1.0854,
      "step": 4450
    },
    {
      "epoch": 423.9047619047619,
      "grad_norm": 7.226143836975098,
      "learning_rate": 8.337706146926538e-07,
      "loss": 1.1316,
      "step": 4451
    },
    {
      "epoch": 424.0,
      "grad_norm": 14.099864959716797,
      "learning_rate": 8.339580209895052e-07,
      "loss": 1.0947,
      "step": 4452
    },
    {
      "epoch": 424.0,
      "eval_accuracy": 0.7452229299363057,
      "eval_f1": 0.7928092626447288,
      "eval_loss": 0.544590950012207,
      "eval_precision": 0.7485615650172612,
      "eval_recall": 0.8426165803108808,
      "eval_roc_auc": 0.7976683937823834,
      "eval_runtime": 1.129,
      "eval_samples_per_second": 2364.033,
      "eval_steps_per_second": 5.314,
      "step": 4452
    },
    {
      "epoch": 424.0952380952381,
      "grad_norm": 14.035150527954102,
      "learning_rate": 8.341454272863569e-07,
      "loss": 1.0665,
      "step": 4453
    },
    {
      "epoch": 424.1904761904762,
      "grad_norm": 17.778974533081055,
      "learning_rate": 8.343328335832083e-07,
      "loss": 1.1228,
      "step": 4454
    },
    {
      "epoch": 424.2857142857143,
      "grad_norm": 6.747130870819092,
      "learning_rate": 8.3452023988006e-07,
      "loss": 1.1226,
      "step": 4455
    },
    {
      "epoch": 424.3809523809524,
      "grad_norm": 12.075554847717285,
      "learning_rate": 8.347076461769116e-07,
      "loss": 1.113,
      "step": 4456
    },
    {
      "epoch": 424.4761904761905,
      "grad_norm": 13.99946403503418,
      "learning_rate": 8.348950524737631e-07,
      "loss": 1.1179,
      "step": 4457
    },
    {
      "epoch": 424.57142857142856,
      "grad_norm": 13.580896377563477,
      "learning_rate": 8.350824587706148e-07,
      "loss": 1.0727,
      "step": 4458
    },
    {
      "epoch": 424.6666666666667,
      "grad_norm": 6.982471466064453,
      "learning_rate": 8.352698650674662e-07,
      "loss": 1.1384,
      "step": 4459
    },
    {
      "epoch": 424.76190476190476,
      "grad_norm": 10.41710090637207,
      "learning_rate": 8.354572713643179e-07,
      "loss": 1.0804,
      "step": 4460
    },
    {
      "epoch": 424.85714285714283,
      "grad_norm": 9.565821647644043,
      "learning_rate": 8.356446776611694e-07,
      "loss": 1.1133,
      "step": 4461
    },
    {
      "epoch": 424.95238095238096,
      "grad_norm": 11.481152534484863,
      "learning_rate": 8.358320839580211e-07,
      "loss": 1.0711,
      "step": 4462
    },
    {
      "epoch": 424.95238095238096,
      "eval_accuracy": 0.7455976020981641,
      "eval_f1": 0.7921640648913376,
      "eval_loss": 0.5441731214523315,
      "eval_precision": 0.751015670342426,
      "eval_recall": 0.8380829015544041,
      "eval_roc_auc": 0.7982130109383995,
      "eval_runtime": 1.1957,
      "eval_samples_per_second": 2232.225,
      "eval_steps_per_second": 5.018,
      "step": 4462
    },
    {
      "epoch": 425.04761904761904,
      "grad_norm": 5.38265323638916,
      "learning_rate": 8.360194902548726e-07,
      "loss": 1.1212,
      "step": 4463
    },
    {
      "epoch": 425.14285714285717,
      "grad_norm": 17.884220123291016,
      "learning_rate": 8.362068965517242e-07,
      "loss": 1.1283,
      "step": 4464
    },
    {
      "epoch": 425.23809523809524,
      "grad_norm": 9.577488899230957,
      "learning_rate": 8.363943028485758e-07,
      "loss": 1.121,
      "step": 4465
    },
    {
      "epoch": 425.3333333333333,
      "grad_norm": 23.120994567871094,
      "learning_rate": 8.365817091454273e-07,
      "loss": 1.171,
      "step": 4466
    },
    {
      "epoch": 425.42857142857144,
      "grad_norm": 10.079218864440918,
      "learning_rate": 8.36769115442279e-07,
      "loss": 1.0982,
      "step": 4467
    },
    {
      "epoch": 425.5238095238095,
      "grad_norm": 5.675240516662598,
      "learning_rate": 8.369565217391304e-07,
      "loss": 1.1029,
      "step": 4468
    },
    {
      "epoch": 425.6190476190476,
      "grad_norm": 13.191267013549805,
      "learning_rate": 8.371439280359821e-07,
      "loss": 1.1224,
      "step": 4469
    },
    {
      "epoch": 425.7142857142857,
      "grad_norm": 17.76279640197754,
      "learning_rate": 8.373313343328337e-07,
      "loss": 1.0952,
      "step": 4470
    },
    {
      "epoch": 425.8095238095238,
      "grad_norm": 17.341856002807617,
      "learning_rate": 8.375187406296852e-07,
      "loss": 1.0476,
      "step": 4471
    },
    {
      "epoch": 425.9047619047619,
      "grad_norm": 22.54288673400879,
      "learning_rate": 8.377061469265368e-07,
      "loss": 1.0452,
      "step": 4472
    },
    {
      "epoch": 426.0,
      "grad_norm": 8.028871536254883,
      "learning_rate": 8.378935532233883e-07,
      "loss": 1.0902,
      "step": 4473
    },
    {
      "epoch": 426.0,
      "eval_accuracy": 0.7452229299363057,
      "eval_f1": 0.792302993280391,
      "eval_loss": 0.544516384601593,
      "eval_precision": 0.7497109826589595,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7977124352331606,
      "eval_runtime": 1.143,
      "eval_samples_per_second": 2335.014,
      "eval_steps_per_second": 5.249,
      "step": 4473
    },
    {
      "epoch": 426.0952380952381,
      "grad_norm": 7.671505928039551,
      "learning_rate": 8.380809595202399e-07,
      "loss": 1.0819,
      "step": 4474
    },
    {
      "epoch": 426.1904761904762,
      "grad_norm": 14.16700267791748,
      "learning_rate": 8.382683658170914e-07,
      "loss": 1.1159,
      "step": 4475
    },
    {
      "epoch": 426.2857142857143,
      "grad_norm": 15.45769214630127,
      "learning_rate": 8.38455772113943e-07,
      "loss": 1.1012,
      "step": 4476
    },
    {
      "epoch": 426.3809523809524,
      "grad_norm": 21.72220802307129,
      "learning_rate": 8.386431784107947e-07,
      "loss": 1.0954,
      "step": 4477
    },
    {
      "epoch": 426.4761904761905,
      "grad_norm": 7.532237529754639,
      "learning_rate": 8.388305847076462e-07,
      "loss": 1.0694,
      "step": 4478
    },
    {
      "epoch": 426.57142857142856,
      "grad_norm": 10.898053169250488,
      "learning_rate": 8.390179910044978e-07,
      "loss": 1.0699,
      "step": 4479
    },
    {
      "epoch": 426.6666666666667,
      "grad_norm": 5.27651309967041,
      "learning_rate": 8.392053973013493e-07,
      "loss": 1.1184,
      "step": 4480
    },
    {
      "epoch": 426.76190476190476,
      "grad_norm": 9.029277801513672,
      "learning_rate": 8.393928035982009e-07,
      "loss": 1.1861,
      "step": 4481
    },
    {
      "epoch": 426.85714285714283,
      "grad_norm": 9.089925765991211,
      "learning_rate": 8.395802098950525e-07,
      "loss": 1.1061,
      "step": 4482
    },
    {
      "epoch": 426.95238095238096,
      "grad_norm": 21.758668899536133,
      "learning_rate": 8.39767616191904e-07,
      "loss": 1.0619,
      "step": 4483
    },
    {
      "epoch": 426.95238095238096,
      "eval_accuracy": 0.7403521918321468,
      "eval_f1": 0.7832342821395057,
      "eval_loss": 0.5450884699821472,
      "eval_precision": 0.7574107683000605,
      "eval_recall": 0.810880829015544,
      "eval_roc_auc": 0.7967820955670697,
      "eval_runtime": 1.1327,
      "eval_samples_per_second": 2356.249,
      "eval_steps_per_second": 5.297,
      "step": 4483
    },
    {
      "epoch": 427.04761904761904,
      "grad_norm": 11.016739845275879,
      "learning_rate": 8.399550224887557e-07,
      "loss": 1.0744,
      "step": 4484
    },
    {
      "epoch": 427.14285714285717,
      "grad_norm": 7.590450763702393,
      "learning_rate": 8.401424287856073e-07,
      "loss": 1.1195,
      "step": 4485
    },
    {
      "epoch": 427.23809523809524,
      "grad_norm": 7.918571949005127,
      "learning_rate": 8.403298350824588e-07,
      "loss": 1.1003,
      "step": 4486
    },
    {
      "epoch": 427.3333333333333,
      "grad_norm": 9.06821060180664,
      "learning_rate": 8.405172413793104e-07,
      "loss": 1.1264,
      "step": 4487
    },
    {
      "epoch": 427.42857142857144,
      "grad_norm": 7.186644554138184,
      "learning_rate": 8.40704647676162e-07,
      "loss": 1.1383,
      "step": 4488
    },
    {
      "epoch": 427.5238095238095,
      "grad_norm": 7.61638069152832,
      "learning_rate": 8.408920539730135e-07,
      "loss": 1.1264,
      "step": 4489
    },
    {
      "epoch": 427.6190476190476,
      "grad_norm": 5.264971733093262,
      "learning_rate": 8.410794602698651e-07,
      "loss": 1.0935,
      "step": 4490
    },
    {
      "epoch": 427.7142857142857,
      "grad_norm": 5.830763339996338,
      "learning_rate": 8.412668665667168e-07,
      "loss": 1.0946,
      "step": 4491
    },
    {
      "epoch": 427.8095238095238,
      "grad_norm": 6.984022617340088,
      "learning_rate": 8.414542728635682e-07,
      "loss": 1.0678,
      "step": 4492
    },
    {
      "epoch": 427.9047619047619,
      "grad_norm": 9.952226638793945,
      "learning_rate": 8.416416791604199e-07,
      "loss": 1.1374,
      "step": 4493
    },
    {
      "epoch": 428.0,
      "grad_norm": 8.996271133422852,
      "learning_rate": 8.418290854572713e-07,
      "loss": 1.0751,
      "step": 4494
    },
    {
      "epoch": 428.0,
      "eval_accuracy": 0.744473585612589,
      "eval_f1": 0.7924528301886793,
      "eval_loss": 0.5440096855163574,
      "eval_precision": 0.7474167623421355,
      "eval_recall": 0.8432642487046632,
      "eval_roc_auc": 0.7981594703511802,
      "eval_runtime": 1.1385,
      "eval_samples_per_second": 2344.319,
      "eval_steps_per_second": 5.27,
      "step": 4494
    },
    {
      "epoch": 428.0952380952381,
      "grad_norm": 8.368547439575195,
      "learning_rate": 8.42016491754123e-07,
      "loss": 1.0567,
      "step": 4495
    },
    {
      "epoch": 428.1904761904762,
      "grad_norm": 17.677316665649414,
      "learning_rate": 8.422038980509745e-07,
      "loss": 1.1367,
      "step": 4496
    },
    {
      "epoch": 428.2857142857143,
      "grad_norm": 9.510296821594238,
      "learning_rate": 8.423913043478261e-07,
      "loss": 1.1055,
      "step": 4497
    },
    {
      "epoch": 428.3809523809524,
      "grad_norm": 9.15112018585205,
      "learning_rate": 8.425787106446778e-07,
      "loss": 1.0695,
      "step": 4498
    },
    {
      "epoch": 428.4761904761905,
      "grad_norm": 7.10412073135376,
      "learning_rate": 8.427661169415292e-07,
      "loss": 1.0799,
      "step": 4499
    },
    {
      "epoch": 428.57142857142856,
      "grad_norm": 17.0451602935791,
      "learning_rate": 8.429535232383809e-07,
      "loss": 1.0909,
      "step": 4500
    },
    {
      "epoch": 428.6666666666667,
      "grad_norm": 9.14710521697998,
      "learning_rate": 8.431409295352323e-07,
      "loss": 1.1133,
      "step": 4501
    },
    {
      "epoch": 428.76190476190476,
      "grad_norm": 5.92765474319458,
      "learning_rate": 8.43328335832084e-07,
      "loss": 1.1242,
      "step": 4502
    },
    {
      "epoch": 428.85714285714283,
      "grad_norm": 9.075368881225586,
      "learning_rate": 8.435157421289354e-07,
      "loss": 1.1234,
      "step": 4503
    },
    {
      "epoch": 428.95238095238096,
      "grad_norm": 7.065749168395996,
      "learning_rate": 8.437031484257871e-07,
      "loss": 1.139,
      "step": 4504
    },
    {
      "epoch": 428.95238095238096,
      "eval_accuracy": 0.7463469464218808,
      "eval_f1": 0.793029654539896,
      "eval_loss": 0.5434408187866211,
      "eval_precision": 0.7510133178922987,
      "eval_recall": 0.8400259067357513,
      "eval_roc_auc": 0.7986295336787564,
      "eval_runtime": 1.1676,
      "eval_samples_per_second": 2285.891,
      "eval_steps_per_second": 5.139,
      "step": 4504
    },
    {
      "epoch": 429.04761904761904,
      "grad_norm": 11.20965576171875,
      "learning_rate": 8.438905547226388e-07,
      "loss": 1.0627,
      "step": 4505
    },
    {
      "epoch": 429.14285714285717,
      "grad_norm": 7.081179618835449,
      "learning_rate": 8.440779610194902e-07,
      "loss": 1.1284,
      "step": 4506
    },
    {
      "epoch": 429.23809523809524,
      "grad_norm": 8.498674392700195,
      "learning_rate": 8.442653673163419e-07,
      "loss": 1.0396,
      "step": 4507
    },
    {
      "epoch": 429.3333333333333,
      "grad_norm": 18.132099151611328,
      "learning_rate": 8.444527736131934e-07,
      "loss": 1.1265,
      "step": 4508
    },
    {
      "epoch": 429.42857142857144,
      "grad_norm": 6.378214359283447,
      "learning_rate": 8.44640179910045e-07,
      "loss": 1.115,
      "step": 4509
    },
    {
      "epoch": 429.5238095238095,
      "grad_norm": 14.451581001281738,
      "learning_rate": 8.448275862068965e-07,
      "loss": 1.1077,
      "step": 4510
    },
    {
      "epoch": 429.6190476190476,
      "grad_norm": 7.9214091300964355,
      "learning_rate": 8.450149925037482e-07,
      "loss": 1.0987,
      "step": 4511
    },
    {
      "epoch": 429.7142857142857,
      "grad_norm": 16.34021759033203,
      "learning_rate": 8.452023988005999e-07,
      "loss": 1.0802,
      "step": 4512
    },
    {
      "epoch": 429.8095238095238,
      "grad_norm": 15.271219253540039,
      "learning_rate": 8.453898050974513e-07,
      "loss": 1.0665,
      "step": 4513
    },
    {
      "epoch": 429.9047619047619,
      "grad_norm": 11.202034950256348,
      "learning_rate": 8.45577211394303e-07,
      "loss": 1.0668,
      "step": 4514
    },
    {
      "epoch": 430.0,
      "grad_norm": 8.156551361083984,
      "learning_rate": 8.457646176911544e-07,
      "loss": 1.1641,
      "step": 4515
    },
    {
      "epoch": 430.0,
      "eval_accuracy": 0.7440989134507306,
      "eval_f1": 0.7883483111248838,
      "eval_loss": 0.5432211756706238,
      "eval_precision": 0.7557932263814616,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.7988059873344847,
      "eval_runtime": 1.1164,
      "eval_samples_per_second": 2390.72,
      "eval_steps_per_second": 5.374,
      "step": 4515
    },
    {
      "epoch": 430.0952380952381,
      "grad_norm": 7.362122535705566,
      "learning_rate": 8.459520239880061e-07,
      "loss": 1.0955,
      "step": 4516
    },
    {
      "epoch": 430.1904761904762,
      "grad_norm": 8.211528778076172,
      "learning_rate": 8.461394302848575e-07,
      "loss": 1.1396,
      "step": 4517
    },
    {
      "epoch": 430.2857142857143,
      "grad_norm": 8.908727645874023,
      "learning_rate": 8.463268365817092e-07,
      "loss": 1.0972,
      "step": 4518
    },
    {
      "epoch": 430.3809523809524,
      "grad_norm": 18.641643524169922,
      "learning_rate": 8.465142428785608e-07,
      "loss": 1.1222,
      "step": 4519
    },
    {
      "epoch": 430.4761904761905,
      "grad_norm": 22.6841983795166,
      "learning_rate": 8.467016491754123e-07,
      "loss": 1.1085,
      "step": 4520
    },
    {
      "epoch": 430.57142857142856,
      "grad_norm": 14.652667999267578,
      "learning_rate": 8.46889055472264e-07,
      "loss": 1.0766,
      "step": 4521
    },
    {
      "epoch": 430.6666666666667,
      "grad_norm": 8.085213661193848,
      "learning_rate": 8.470764617691154e-07,
      "loss": 1.0738,
      "step": 4522
    },
    {
      "epoch": 430.76190476190476,
      "grad_norm": 6.663705825805664,
      "learning_rate": 8.472638680659671e-07,
      "loss": 1.1414,
      "step": 4523
    },
    {
      "epoch": 430.85714285714283,
      "grad_norm": 13.765937805175781,
      "learning_rate": 8.474512743628185e-07,
      "loss": 1.1053,
      "step": 4524
    },
    {
      "epoch": 430.95238095238096,
      "grad_norm": 16.29475975036621,
      "learning_rate": 8.476386806596702e-07,
      "loss": 1.0983,
      "step": 4525
    },
    {
      "epoch": 430.95238095238096,
      "eval_accuracy": 0.7388535031847133,
      "eval_f1": 0.780886513674945,
      "eval_loss": 0.5443218350410461,
      "eval_precision": 0.7587049480757483,
      "eval_recall": 0.8044041450777202,
      "eval_roc_auc": 0.7975322394933794,
      "eval_runtime": 1.1195,
      "eval_samples_per_second": 2384.206,
      "eval_steps_per_second": 5.36,
      "step": 4525
    },
    {
      "epoch": 431.04761904761904,
      "grad_norm": 30.667348861694336,
      "learning_rate": 8.478260869565218e-07,
      "loss": 1.1,
      "step": 4526
    },
    {
      "epoch": 431.14285714285717,
      "grad_norm": 8.329144477844238,
      "learning_rate": 8.480134932533733e-07,
      "loss": 1.0965,
      "step": 4527
    },
    {
      "epoch": 431.23809523809524,
      "grad_norm": 16.07992172241211,
      "learning_rate": 8.482008995502249e-07,
      "loss": 1.0884,
      "step": 4528
    },
    {
      "epoch": 431.3333333333333,
      "grad_norm": 9.440814018249512,
      "learning_rate": 8.483883058470765e-07,
      "loss": 1.0635,
      "step": 4529
    },
    {
      "epoch": 431.42857142857144,
      "grad_norm": 15.835526466369629,
      "learning_rate": 8.48575712143928e-07,
      "loss": 1.1016,
      "step": 4530
    },
    {
      "epoch": 431.5238095238095,
      "grad_norm": 13.049686431884766,
      "learning_rate": 8.487631184407796e-07,
      "loss": 1.0904,
      "step": 4531
    },
    {
      "epoch": 431.6190476190476,
      "grad_norm": 14.782801628112793,
      "learning_rate": 8.489505247376313e-07,
      "loss": 1.1082,
      "step": 4532
    },
    {
      "epoch": 431.7142857142857,
      "grad_norm": 22.376850128173828,
      "learning_rate": 8.491379310344828e-07,
      "loss": 1.1032,
      "step": 4533
    },
    {
      "epoch": 431.8095238095238,
      "grad_norm": 20.260921478271484,
      "learning_rate": 8.493253373313344e-07,
      "loss": 1.1155,
      "step": 4534
    },
    {
      "epoch": 431.9047619047619,
      "grad_norm": 7.1398725509643555,
      "learning_rate": 8.49512743628186e-07,
      "loss": 1.1355,
      "step": 4535
    },
    {
      "epoch": 432.0,
      "grad_norm": 12.543313980102539,
      "learning_rate": 8.497001499250375e-07,
      "loss": 1.0893,
      "step": 4536
    },
    {
      "epoch": 432.0,
      "eval_accuracy": 0.7429748969651555,
      "eval_f1": 0.785892634207241,
      "eval_loss": 0.543362021446228,
      "eval_precision": 0.7584337349397591,
      "eval_recall": 0.8154145077720207,
      "eval_roc_auc": 0.7986922855497987,
      "eval_runtime": 1.1357,
      "eval_samples_per_second": 2350.082,
      "eval_steps_per_second": 5.283,
      "step": 4536
    },
    {
      "epoch": 432.0952380952381,
      "grad_norm": 5.253561019897461,
      "learning_rate": 8.498875562218891e-07,
      "loss": 1.1291,
      "step": 4537
    },
    {
      "epoch": 432.1904761904762,
      "grad_norm": 7.003561973571777,
      "learning_rate": 8.500749625187406e-07,
      "loss": 1.1181,
      "step": 4538
    },
    {
      "epoch": 432.2857142857143,
      "grad_norm": 29.183019638061523,
      "learning_rate": 8.502623688155923e-07,
      "loss": 1.0722,
      "step": 4539
    },
    {
      "epoch": 432.3809523809524,
      "grad_norm": 16.94145393371582,
      "learning_rate": 8.504497751124439e-07,
      "loss": 1.0844,
      "step": 4540
    },
    {
      "epoch": 432.4761904761905,
      "grad_norm": 23.304765701293945,
      "learning_rate": 8.506371814092954e-07,
      "loss": 1.0513,
      "step": 4541
    },
    {
      "epoch": 432.57142857142856,
      "grad_norm": 13.36349105834961,
      "learning_rate": 8.50824587706147e-07,
      "loss": 1.1252,
      "step": 4542
    },
    {
      "epoch": 432.6666666666667,
      "grad_norm": 5.048804759979248,
      "learning_rate": 8.510119940029985e-07,
      "loss": 1.1111,
      "step": 4543
    },
    {
      "epoch": 432.76190476190476,
      "grad_norm": 10.21939468383789,
      "learning_rate": 8.511994002998501e-07,
      "loss": 1.1129,
      "step": 4544
    },
    {
      "epoch": 432.85714285714283,
      "grad_norm": 8.018023490905762,
      "learning_rate": 8.513868065967016e-07,
      "loss": 1.0731,
      "step": 4545
    },
    {
      "epoch": 432.95238095238096,
      "grad_norm": 7.383289813995361,
      "learning_rate": 8.515742128935532e-07,
      "loss": 1.0969,
      "step": 4546
    },
    {
      "epoch": 432.95238095238096,
      "eval_accuracy": 0.7463469464218808,
      "eval_f1": 0.794287450622911,
      "eval_loss": 0.5438169240951538,
      "eval_precision": 0.7481396680022896,
      "eval_recall": 0.8465025906735751,
      "eval_roc_auc": 0.7986373056994819,
      "eval_runtime": 1.1242,
      "eval_samples_per_second": 2374.237,
      "eval_steps_per_second": 5.337,
      "step": 4546
    },
    {
      "epoch": 433.04761904761904,
      "grad_norm": 12.807202339172363,
      "learning_rate": 8.517616191904049e-07,
      "loss": 1.1247,
      "step": 4547
    },
    {
      "epoch": 433.14285714285717,
      "grad_norm": 25.507034301757812,
      "learning_rate": 8.519490254872564e-07,
      "loss": 1.1461,
      "step": 4548
    },
    {
      "epoch": 433.23809523809524,
      "grad_norm": 16.608102798461914,
      "learning_rate": 8.52136431784108e-07,
      "loss": 1.0899,
      "step": 4549
    },
    {
      "epoch": 433.3333333333333,
      "grad_norm": 6.582379341125488,
      "learning_rate": 8.523238380809595e-07,
      "loss": 1.085,
      "step": 4550
    },
    {
      "epoch": 433.42857142857144,
      "grad_norm": 7.415794849395752,
      "learning_rate": 8.525112443778111e-07,
      "loss": 1.0937,
      "step": 4551
    },
    {
      "epoch": 433.5238095238095,
      "grad_norm": 9.852500915527344,
      "learning_rate": 8.526986506746627e-07,
      "loss": 1.0917,
      "step": 4552
    },
    {
      "epoch": 433.6190476190476,
      "grad_norm": 7.424929618835449,
      "learning_rate": 8.528860569715142e-07,
      "loss": 1.1468,
      "step": 4553
    },
    {
      "epoch": 433.7142857142857,
      "grad_norm": 11.212461471557617,
      "learning_rate": 8.530734632683659e-07,
      "loss": 1.1566,
      "step": 4554
    },
    {
      "epoch": 433.8095238095238,
      "grad_norm": 7.593562126159668,
      "learning_rate": 8.532608695652174e-07,
      "loss": 1.0882,
      "step": 4555
    },
    {
      "epoch": 433.9047619047619,
      "grad_norm": 29.836612701416016,
      "learning_rate": 8.53448275862069e-07,
      "loss": 1.0821,
      "step": 4556
    },
    {
      "epoch": 434.0,
      "grad_norm": 21.95094108581543,
      "learning_rate": 8.536356821589206e-07,
      "loss": 1.0802,
      "step": 4557
    },
    {
      "epoch": 434.0,
      "eval_accuracy": 0.7429748969651555,
      "eval_f1": 0.785892634207241,
      "eval_loss": 0.5434971451759338,
      "eval_precision": 0.7584337349397591,
      "eval_recall": 0.8154145077720207,
      "eval_roc_auc": 0.798743523316062,
      "eval_runtime": 1.1244,
      "eval_samples_per_second": 2373.812,
      "eval_steps_per_second": 5.336,
      "step": 4557
    },
    {
      "epoch": 434.0952380952381,
      "grad_norm": 9.611329078674316,
      "learning_rate": 8.538230884557722e-07,
      "loss": 1.127,
      "step": 4558
    },
    {
      "epoch": 434.1904761904762,
      "grad_norm": 7.204107761383057,
      "learning_rate": 8.540104947526237e-07,
      "loss": 1.1249,
      "step": 4559
    },
    {
      "epoch": 434.2857142857143,
      "grad_norm": 7.247961044311523,
      "learning_rate": 8.541979010494753e-07,
      "loss": 1.0857,
      "step": 4560
    },
    {
      "epoch": 434.3809523809524,
      "grad_norm": 6.680117607116699,
      "learning_rate": 8.54385307346327e-07,
      "loss": 1.0866,
      "step": 4561
    },
    {
      "epoch": 434.4761904761905,
      "grad_norm": 11.561103820800781,
      "learning_rate": 8.545727136431784e-07,
      "loss": 1.1357,
      "step": 4562
    },
    {
      "epoch": 434.57142857142856,
      "grad_norm": 12.856839179992676,
      "learning_rate": 8.547601199400301e-07,
      "loss": 1.0724,
      "step": 4563
    },
    {
      "epoch": 434.6666666666667,
      "grad_norm": 31.07172203063965,
      "learning_rate": 8.549475262368815e-07,
      "loss": 1.1404,
      "step": 4564
    },
    {
      "epoch": 434.76190476190476,
      "grad_norm": 5.869296550750732,
      "learning_rate": 8.551349325337332e-07,
      "loss": 1.0578,
      "step": 4565
    },
    {
      "epoch": 434.85714285714283,
      "grad_norm": 6.1051926612854,
      "learning_rate": 8.553223388305846e-07,
      "loss": 1.1019,
      "step": 4566
    },
    {
      "epoch": 434.95238095238096,
      "grad_norm": 7.054297924041748,
      "learning_rate": 8.555097451274363e-07,
      "loss": 1.075,
      "step": 4567
    },
    {
      "epoch": 434.95238095238096,
      "eval_accuracy": 0.7452229299363057,
      "eval_f1": 0.7881619937694704,
      "eval_loss": 0.5423450469970703,
      "eval_precision": 0.7593037214885955,
      "eval_recall": 0.819300518134715,
      "eval_roc_auc": 0.7997233736327,
      "eval_runtime": 1.1364,
      "eval_samples_per_second": 2348.591,
      "eval_steps_per_second": 5.28,
      "step": 4567
    },
    {
      "epoch": 435.04761904761904,
      "grad_norm": 6.471446990966797,
      "learning_rate": 8.55697151424288e-07,
      "loss": 1.119,
      "step": 4568
    },
    {
      "epoch": 435.14285714285717,
      "grad_norm": 13.11718463897705,
      "learning_rate": 8.558845577211394e-07,
      "loss": 1.0647,
      "step": 4569
    },
    {
      "epoch": 435.23809523809524,
      "grad_norm": 19.984201431274414,
      "learning_rate": 8.560719640179911e-07,
      "loss": 1.0537,
      "step": 4570
    },
    {
      "epoch": 435.3333333333333,
      "grad_norm": 15.48215389251709,
      "learning_rate": 8.562593703148425e-07,
      "loss": 1.0928,
      "step": 4571
    },
    {
      "epoch": 435.42857142857144,
      "grad_norm": 6.437912940979004,
      "learning_rate": 8.564467766116942e-07,
      "loss": 1.0896,
      "step": 4572
    },
    {
      "epoch": 435.5238095238095,
      "grad_norm": 7.394294738769531,
      "learning_rate": 8.566341829085456e-07,
      "loss": 1.0828,
      "step": 4573
    },
    {
      "epoch": 435.6190476190476,
      "grad_norm": 17.219388961791992,
      "learning_rate": 8.568215892053973e-07,
      "loss": 1.1149,
      "step": 4574
    },
    {
      "epoch": 435.7142857142857,
      "grad_norm": 11.16395378112793,
      "learning_rate": 8.57008995502249e-07,
      "loss": 1.1232,
      "step": 4575
    },
    {
      "epoch": 435.8095238095238,
      "grad_norm": 31.06517791748047,
      "learning_rate": 8.571964017991004e-07,
      "loss": 1.1299,
      "step": 4576
    },
    {
      "epoch": 435.9047619047619,
      "grad_norm": 23.21164894104004,
      "learning_rate": 8.573838080959521e-07,
      "loss": 1.1382,
      "step": 4577
    },
    {
      "epoch": 436.0,
      "grad_norm": 17.702280044555664,
      "learning_rate": 8.575712143928036e-07,
      "loss": 1.1218,
      "step": 4578
    },
    {
      "epoch": 436.0,
      "eval_accuracy": 0.7478456350693143,
      "eval_f1": 0.7922198209323865,
      "eval_loss": 0.5420184135437012,
      "eval_precision": 0.7569321533923303,
      "eval_recall": 0.8309585492227979,
      "eval_roc_auc": 0.8004461715601612,
      "eval_runtime": 1.1459,
      "eval_samples_per_second": 2329.125,
      "eval_steps_per_second": 5.236,
      "step": 4578
    },
    {
      "epoch": 436.0952380952381,
      "grad_norm": 9.291200637817383,
      "learning_rate": 8.577586206896553e-07,
      "loss": 1.1037,
      "step": 4579
    },
    {
      "epoch": 436.1904761904762,
      "grad_norm": 7.206141471862793,
      "learning_rate": 8.579460269865067e-07,
      "loss": 1.1025,
      "step": 4580
    },
    {
      "epoch": 436.2857142857143,
      "grad_norm": 11.635160446166992,
      "learning_rate": 8.581334332833584e-07,
      "loss": 1.0913,
      "step": 4581
    },
    {
      "epoch": 436.3809523809524,
      "grad_norm": 16.53832244873047,
      "learning_rate": 8.5832083958021e-07,
      "loss": 1.0808,
      "step": 4582
    },
    {
      "epoch": 436.4761904761905,
      "grad_norm": 9.663097381591797,
      "learning_rate": 8.585082458770615e-07,
      "loss": 1.1139,
      "step": 4583
    },
    {
      "epoch": 436.57142857142856,
      "grad_norm": 8.36888313293457,
      "learning_rate": 8.586956521739132e-07,
      "loss": 1.1056,
      "step": 4584
    },
    {
      "epoch": 436.6666666666667,
      "grad_norm": 9.114745140075684,
      "learning_rate": 8.588830584707646e-07,
      "loss": 1.1214,
      "step": 4585
    },
    {
      "epoch": 436.76190476190476,
      "grad_norm": 6.442534923553467,
      "learning_rate": 8.590704647676163e-07,
      "loss": 1.1052,
      "step": 4586
    },
    {
      "epoch": 436.85714285714283,
      "grad_norm": 15.638298034667969,
      "learning_rate": 8.592578710644677e-07,
      "loss": 1.0966,
      "step": 4587
    },
    {
      "epoch": 436.95238095238096,
      "grad_norm": 8.274981498718262,
      "learning_rate": 8.594452773613194e-07,
      "loss": 1.081,
      "step": 4588
    },
    {
      "epoch": 436.95238095238096,
      "eval_accuracy": 0.7440989134507306,
      "eval_f1": 0.7911953531030266,
      "eval_loss": 0.5425230860710144,
      "eval_precision": 0.7492762015055009,
      "eval_recall": 0.8380829015544041,
      "eval_roc_auc": 0.8002530224525044,
      "eval_runtime": 1.1221,
      "eval_samples_per_second": 2378.65,
      "eval_steps_per_second": 5.347,
      "step": 4588
    },
    {
      "epoch": 437.04761904761904,
      "grad_norm": 12.262890815734863,
      "learning_rate": 8.59632683658171e-07,
      "loss": 1.1309,
      "step": 4589
    },
    {
      "epoch": 437.14285714285717,
      "grad_norm": 7.5080342292785645,
      "learning_rate": 8.598200899550225e-07,
      "loss": 1.0975,
      "step": 4590
    },
    {
      "epoch": 437.23809523809524,
      "grad_norm": 13.298624992370605,
      "learning_rate": 8.600074962518741e-07,
      "loss": 1.0747,
      "step": 4591
    },
    {
      "epoch": 437.3333333333333,
      "grad_norm": 16.601577758789062,
      "learning_rate": 8.601949025487256e-07,
      "loss": 1.1014,
      "step": 4592
    },
    {
      "epoch": 437.42857142857144,
      "grad_norm": 20.987707138061523,
      "learning_rate": 8.603823088455773e-07,
      "loss": 1.1211,
      "step": 4593
    },
    {
      "epoch": 437.5238095238095,
      "grad_norm": 6.50589656829834,
      "learning_rate": 8.605697151424287e-07,
      "loss": 1.1063,
      "step": 4594
    },
    {
      "epoch": 437.6190476190476,
      "grad_norm": 6.247185707092285,
      "learning_rate": 8.607571214392804e-07,
      "loss": 1.1114,
      "step": 4595
    },
    {
      "epoch": 437.7142857142857,
      "grad_norm": 10.65212345123291,
      "learning_rate": 8.60944527736132e-07,
      "loss": 1.1128,
      "step": 4596
    },
    {
      "epoch": 437.8095238095238,
      "grad_norm": 10.363444328308105,
      "learning_rate": 8.611319340329835e-07,
      "loss": 1.0817,
      "step": 4597
    },
    {
      "epoch": 437.9047619047619,
      "grad_norm": 17.171010971069336,
      "learning_rate": 8.613193403298351e-07,
      "loss": 1.1228,
      "step": 4598
    },
    {
      "epoch": 438.0,
      "grad_norm": 12.242009162902832,
      "learning_rate": 8.615067466266867e-07,
      "loss": 1.0522,
      "step": 4599
    },
    {
      "epoch": 438.0,
      "eval_accuracy": 0.7418508804795804,
      "eval_f1": 0.787673343605547,
      "eval_loss": 0.543170154094696,
      "eval_precision": 0.7513227513227513,
      "eval_recall": 0.8277202072538861,
      "eval_roc_auc": 0.7993408175014393,
      "eval_runtime": 1.1738,
      "eval_samples_per_second": 2273.718,
      "eval_steps_per_second": 5.111,
      "step": 4599
    },
    {
      "epoch": 438.0952380952381,
      "grad_norm": 17.506868362426758,
      "learning_rate": 8.616941529235382e-07,
      "loss": 1.0765,
      "step": 4600
    },
    {
      "epoch": 438.1904761904762,
      "grad_norm": 13.496466636657715,
      "learning_rate": 8.618815592203898e-07,
      "loss": 1.0902,
      "step": 4601
    },
    {
      "epoch": 438.2857142857143,
      "grad_norm": 15.746648788452148,
      "learning_rate": 8.620689655172415e-07,
      "loss": 1.0835,
      "step": 4602
    },
    {
      "epoch": 438.3809523809524,
      "grad_norm": 16.976329803466797,
      "learning_rate": 8.62256371814093e-07,
      "loss": 1.1295,
      "step": 4603
    },
    {
      "epoch": 438.4761904761905,
      "grad_norm": 17.543289184570312,
      "learning_rate": 8.624437781109446e-07,
      "loss": 1.0789,
      "step": 4604
    },
    {
      "epoch": 438.57142857142856,
      "grad_norm": 10.029759407043457,
      "learning_rate": 8.626311844077962e-07,
      "loss": 1.069,
      "step": 4605
    },
    {
      "epoch": 438.6666666666667,
      "grad_norm": 7.242961883544922,
      "learning_rate": 8.628185907046477e-07,
      "loss": 1.0864,
      "step": 4606
    },
    {
      "epoch": 438.76190476190476,
      "grad_norm": 18.78436279296875,
      "learning_rate": 8.630059970014993e-07,
      "loss": 1.1065,
      "step": 4607
    },
    {
      "epoch": 438.85714285714283,
      "grad_norm": 9.78477668762207,
      "learning_rate": 8.631934032983508e-07,
      "loss": 1.148,
      "step": 4608
    },
    {
      "epoch": 438.95238095238096,
      "grad_norm": 5.987895965576172,
      "learning_rate": 8.633808095952024e-07,
      "loss": 1.09,
      "step": 4609
    },
    {
      "epoch": 438.95238095238096,
      "eval_accuracy": 0.7411015361558636,
      "eval_f1": 0.784399375975039,
      "eval_loss": 0.5427181124687195,
      "eval_precision": 0.7567730282962071,
      "eval_recall": 0.814119170984456,
      "eval_roc_auc": 0.7998529073114565,
      "eval_runtime": 1.4168,
      "eval_samples_per_second": 1883.875,
      "eval_steps_per_second": 4.235,
      "step": 4609
    },
    {
      "epoch": 439.04761904761904,
      "grad_norm": 9.344883918762207,
      "learning_rate": 8.635682158920541e-07,
      "loss": 1.139,
      "step": 4610
    },
    {
      "epoch": 439.14285714285717,
      "grad_norm": 16.81941795349121,
      "learning_rate": 8.637556221889056e-07,
      "loss": 1.1187,
      "step": 4611
    },
    {
      "epoch": 439.23809523809524,
      "grad_norm": 6.715560436248779,
      "learning_rate": 8.639430284857572e-07,
      "loss": 1.1438,
      "step": 4612
    },
    {
      "epoch": 439.3333333333333,
      "grad_norm": 12.95703411102295,
      "learning_rate": 8.641304347826087e-07,
      "loss": 1.0913,
      "step": 4613
    },
    {
      "epoch": 439.42857142857144,
      "grad_norm": 8.936210632324219,
      "learning_rate": 8.643178410794603e-07,
      "loss": 1.0851,
      "step": 4614
    },
    {
      "epoch": 439.5238095238095,
      "grad_norm": 6.878790855407715,
      "learning_rate": 8.64505247376312e-07,
      "loss": 1.078,
      "step": 4615
    },
    {
      "epoch": 439.6190476190476,
      "grad_norm": 5.259520530700684,
      "learning_rate": 8.646926536731634e-07,
      "loss": 1.1036,
      "step": 4616
    },
    {
      "epoch": 439.7142857142857,
      "grad_norm": 10.157668113708496,
      "learning_rate": 8.648800599700151e-07,
      "loss": 1.0713,
      "step": 4617
    },
    {
      "epoch": 439.8095238095238,
      "grad_norm": 9.646769523620605,
      "learning_rate": 8.650674662668665e-07,
      "loss": 1.1216,
      "step": 4618
    },
    {
      "epoch": 439.9047619047619,
      "grad_norm": 14.076932907104492,
      "learning_rate": 8.652548725637182e-07,
      "loss": 1.0955,
      "step": 4619
    },
    {
      "epoch": 440.0,
      "grad_norm": 7.416194915771484,
      "learning_rate": 8.654422788605696e-07,
      "loss": 1.1111,
      "step": 4620
    },
    {
      "epoch": 440.0,
      "eval_accuracy": 0.7418508804795804,
      "eval_f1": 0.7871485943775101,
      "eval_loss": 0.542918860912323,
      "eval_precision": 0.7525103366804489,
      "eval_recall": 0.8251295336787565,
      "eval_roc_auc": 0.7997084052964881,
      "eval_runtime": 1.1251,
      "eval_samples_per_second": 2372.254,
      "eval_steps_per_second": 5.333,
      "step": 4620
    },
    {
      "epoch": 440.0952380952381,
      "grad_norm": 9.420711517333984,
      "learning_rate": 8.656296851574213e-07,
      "loss": 1.0713,
      "step": 4621
    },
    {
      "epoch": 440.1904761904762,
      "grad_norm": 10.850831031799316,
      "learning_rate": 8.65817091454273e-07,
      "loss": 1.1064,
      "step": 4622
    },
    {
      "epoch": 440.2857142857143,
      "grad_norm": 11.357560157775879,
      "learning_rate": 8.660044977511244e-07,
      "loss": 1.1203,
      "step": 4623
    },
    {
      "epoch": 440.3809523809524,
      "grad_norm": 7.194850921630859,
      "learning_rate": 8.661919040479761e-07,
      "loss": 1.0921,
      "step": 4624
    },
    {
      "epoch": 440.4761904761905,
      "grad_norm": 18.304948806762695,
      "learning_rate": 8.663793103448276e-07,
      "loss": 1.0903,
      "step": 4625
    },
    {
      "epoch": 440.57142857142856,
      "grad_norm": 10.864583969116211,
      "learning_rate": 8.665667166416792e-07,
      "loss": 1.1132,
      "step": 4626
    },
    {
      "epoch": 440.6666666666667,
      "grad_norm": 16.42337989807129,
      "learning_rate": 8.667541229385307e-07,
      "loss": 1.1125,
      "step": 4627
    },
    {
      "epoch": 440.76190476190476,
      "grad_norm": 14.132763862609863,
      "learning_rate": 8.669415292353824e-07,
      "loss": 1.1365,
      "step": 4628
    },
    {
      "epoch": 440.85714285714283,
      "grad_norm": 10.242207527160645,
      "learning_rate": 8.67128935532234e-07,
      "loss": 1.0925,
      "step": 4629
    },
    {
      "epoch": 440.95238095238096,
      "grad_norm": 23.153173446655273,
      "learning_rate": 8.673163418290855e-07,
      "loss": 1.0465,
      "step": 4630
    },
    {
      "epoch": 440.95238095238096,
      "eval_accuracy": 0.7411015361558636,
      "eval_f1": 0.7820876695048881,
      "eval_loss": 0.5431788563728333,
      "eval_precision": 0.762138905961893,
      "eval_recall": 0.8031088082901554,
      "eval_roc_auc": 0.7999170984455958,
      "eval_runtime": 1.1195,
      "eval_samples_per_second": 2384.095,
      "eval_steps_per_second": 5.36,
      "step": 4630
    },
    {
      "epoch": 441.04761904761904,
      "grad_norm": 25.745071411132812,
      "learning_rate": 8.675037481259372e-07,
      "loss": 1.08,
      "step": 4631
    },
    {
      "epoch": 441.14285714285717,
      "grad_norm": 18.603288650512695,
      "learning_rate": 8.676911544227886e-07,
      "loss": 1.0773,
      "step": 4632
    },
    {
      "epoch": 441.23809523809524,
      "grad_norm": 14.742155075073242,
      "learning_rate": 8.678785607196403e-07,
      "loss": 1.1537,
      "step": 4633
    },
    {
      "epoch": 441.3333333333333,
      "grad_norm": 14.234055519104004,
      "learning_rate": 8.680659670164917e-07,
      "loss": 1.0868,
      "step": 4634
    },
    {
      "epoch": 441.42857142857144,
      "grad_norm": 5.549522399902344,
      "learning_rate": 8.682533733133434e-07,
      "loss": 1.1686,
      "step": 4635
    },
    {
      "epoch": 441.5238095238095,
      "grad_norm": 9.679647445678711,
      "learning_rate": 8.68440779610195e-07,
      "loss": 1.1148,
      "step": 4636
    },
    {
      "epoch": 441.6190476190476,
      "grad_norm": 14.539051055908203,
      "learning_rate": 8.686281859070465e-07,
      "loss": 1.1136,
      "step": 4637
    },
    {
      "epoch": 441.7142857142857,
      "grad_norm": 15.705877304077148,
      "learning_rate": 8.688155922038982e-07,
      "loss": 1.0616,
      "step": 4638
    },
    {
      "epoch": 441.8095238095238,
      "grad_norm": 5.833678245544434,
      "learning_rate": 8.690029985007496e-07,
      "loss": 1.0744,
      "step": 4639
    },
    {
      "epoch": 441.9047619047619,
      "grad_norm": 6.723447322845459,
      "learning_rate": 8.691904047976013e-07,
      "loss": 1.0927,
      "step": 4640
    },
    {
      "epoch": 442.0,
      "grad_norm": 8.890020370483398,
      "learning_rate": 8.693778110944527e-07,
      "loss": 1.0845,
      "step": 4641
    },
    {
      "epoch": 442.0,
      "eval_accuracy": 0.7448482577744474,
      "eval_f1": 0.7902679396365876,
      "eval_loss": 0.5423680543899536,
      "eval_precision": 0.7533763945977686,
      "eval_recall": 0.8309585492227979,
      "eval_roc_auc": 0.8005368451352908,
      "eval_runtime": 1.1212,
      "eval_samples_per_second": 2380.501,
      "eval_steps_per_second": 5.351,
      "step": 4641
    },
    {
      "epoch": 442.0952380952381,
      "grad_norm": 12.400238037109375,
      "learning_rate": 8.695652173913044e-07,
      "loss": 1.0651,
      "step": 4642
    },
    {
      "epoch": 442.1904761904762,
      "grad_norm": 14.389338493347168,
      "learning_rate": 8.69752623688156e-07,
      "loss": 1.163,
      "step": 4643
    },
    {
      "epoch": 442.2857142857143,
      "grad_norm": 11.256134033203125,
      "learning_rate": 8.699400299850075e-07,
      "loss": 1.0795,
      "step": 4644
    },
    {
      "epoch": 442.3809523809524,
      "grad_norm": 7.7698974609375,
      "learning_rate": 8.701274362818591e-07,
      "loss": 1.0783,
      "step": 4645
    },
    {
      "epoch": 442.4761904761905,
      "grad_norm": 8.148468971252441,
      "learning_rate": 8.703148425787106e-07,
      "loss": 1.1104,
      "step": 4646
    },
    {
      "epoch": 442.57142857142856,
      "grad_norm": 6.954012393951416,
      "learning_rate": 8.705022488755623e-07,
      "loss": 1.151,
      "step": 4647
    },
    {
      "epoch": 442.6666666666667,
      "grad_norm": 6.079837322235107,
      "learning_rate": 8.706896551724138e-07,
      "loss": 1.1116,
      "step": 4648
    },
    {
      "epoch": 442.76190476190476,
      "grad_norm": 25.121978759765625,
      "learning_rate": 8.708770614692655e-07,
      "loss": 1.1644,
      "step": 4649
    },
    {
      "epoch": 442.85714285714283,
      "grad_norm": 10.784706115722656,
      "learning_rate": 8.71064467766117e-07,
      "loss": 1.0589,
      "step": 4650
    },
    {
      "epoch": 442.95238095238096,
      "grad_norm": 16.868013381958008,
      "learning_rate": 8.712518740629686e-07,
      "loss": 1.0715,
      "step": 4651
    },
    {
      "epoch": 442.95238095238096,
      "eval_accuracy": 0.744473585612589,
      "eval_f1": 0.7870081199250468,
      "eval_loss": 0.5424505472183228,
      "eval_precision": 0.7599517490952955,
      "eval_recall": 0.8160621761658031,
      "eval_roc_auc": 0.8000805987334484,
      "eval_runtime": 1.1447,
      "eval_samples_per_second": 2331.708,
      "eval_steps_per_second": 5.242,
      "step": 4651
    },
    {
      "epoch": 443.04761904761904,
      "grad_norm": 14.400278091430664,
      "learning_rate": 8.714392803598202e-07,
      "loss": 1.1153,
      "step": 4652
    },
    {
      "epoch": 443.14285714285717,
      "grad_norm": 7.858428955078125,
      "learning_rate": 8.716266866566717e-07,
      "loss": 1.1209,
      "step": 4653
    },
    {
      "epoch": 443.23809523809524,
      "grad_norm": 16.608057022094727,
      "learning_rate": 8.718140929535233e-07,
      "loss": 1.1302,
      "step": 4654
    },
    {
      "epoch": 443.3333333333333,
      "grad_norm": 15.479148864746094,
      "learning_rate": 8.720014992503748e-07,
      "loss": 1.0628,
      "step": 4655
    },
    {
      "epoch": 443.42857142857144,
      "grad_norm": 8.721065521240234,
      "learning_rate": 8.721889055472265e-07,
      "loss": 1.1089,
      "step": 4656
    },
    {
      "epoch": 443.5238095238095,
      "grad_norm": 16.358375549316406,
      "learning_rate": 8.723763118440781e-07,
      "loss": 1.1184,
      "step": 4657
    },
    {
      "epoch": 443.6190476190476,
      "grad_norm": 8.80472469329834,
      "learning_rate": 8.725637181409296e-07,
      "loss": 1.0479,
      "step": 4658
    },
    {
      "epoch": 443.7142857142857,
      "grad_norm": 7.0978569984436035,
      "learning_rate": 8.727511244377812e-07,
      "loss": 1.1234,
      "step": 4659
    },
    {
      "epoch": 443.8095238095238,
      "grad_norm": 9.775753021240234,
      "learning_rate": 8.729385307346327e-07,
      "loss": 1.1586,
      "step": 4660
    },
    {
      "epoch": 443.9047619047619,
      "grad_norm": 6.316388130187988,
      "learning_rate": 8.731259370314843e-07,
      "loss": 1.1405,
      "step": 4661
    },
    {
      "epoch": 444.0,
      "grad_norm": 8.489418983459473,
      "learning_rate": 8.733133433283358e-07,
      "loss": 1.0751,
      "step": 4662
    },
    {
      "epoch": 444.0,
      "eval_accuracy": 0.7467216185837392,
      "eval_f1": 0.7949029126213593,
      "eval_loss": 0.5433393716812134,
      "eval_precision": 0.747716894977169,
      "eval_recall": 0.8484455958549223,
      "eval_roc_auc": 0.7994668969487622,
      "eval_runtime": 1.1107,
      "eval_samples_per_second": 2402.964,
      "eval_steps_per_second": 5.402,
      "step": 4662
    },
    {
      "epoch": 444.0952380952381,
      "grad_norm": 13.219255447387695,
      "learning_rate": 8.735007496251874e-07,
      "loss": 1.1428,
      "step": 4663
    },
    {
      "epoch": 444.1904761904762,
      "grad_norm": 11.721603393554688,
      "learning_rate": 8.736881559220391e-07,
      "loss": 1.0618,
      "step": 4664
    },
    {
      "epoch": 444.2857142857143,
      "grad_norm": 11.433424949645996,
      "learning_rate": 8.738755622188906e-07,
      "loss": 1.0972,
      "step": 4665
    },
    {
      "epoch": 444.3809523809524,
      "grad_norm": 16.360595703125,
      "learning_rate": 8.740629685157422e-07,
      "loss": 1.1192,
      "step": 4666
    },
    {
      "epoch": 444.4761904761905,
      "grad_norm": 6.362143516540527,
      "learning_rate": 8.742503748125937e-07,
      "loss": 1.0295,
      "step": 4667
    },
    {
      "epoch": 444.57142857142856,
      "grad_norm": 21.919862747192383,
      "learning_rate": 8.744377811094453e-07,
      "loss": 1.1246,
      "step": 4668
    },
    {
      "epoch": 444.6666666666667,
      "grad_norm": 10.915406227111816,
      "learning_rate": 8.746251874062969e-07,
      "loss": 1.1065,
      "step": 4669
    },
    {
      "epoch": 444.76190476190476,
      "grad_norm": 13.651249885559082,
      "learning_rate": 8.748125937031484e-07,
      "loss": 1.1167,
      "step": 4670
    },
    {
      "epoch": 444.85714285714283,
      "grad_norm": 7.456640720367432,
      "learning_rate": 8.750000000000001e-07,
      "loss": 1.1062,
      "step": 4671
    },
    {
      "epoch": 444.95238095238096,
      "grad_norm": 15.28005313873291,
      "learning_rate": 8.751874062968516e-07,
      "loss": 1.0744,
      "step": 4672
    },
    {
      "epoch": 444.95238095238096,
      "eval_accuracy": 0.7407268639940052,
      "eval_f1": 0.7825267127592709,
      "eval_loss": 0.5431521534919739,
      "eval_precision": 0.76007326007326,
      "eval_recall": 0.8063471502590673,
      "eval_roc_auc": 0.7993408175014393,
      "eval_runtime": 1.1292,
      "eval_samples_per_second": 2363.639,
      "eval_steps_per_second": 5.314,
      "step": 4672
    },
    {
      "epoch": 445.04761904761904,
      "grad_norm": 9.96605110168457,
      "learning_rate": 8.753748125937032e-07,
      "loss": 1.1101,
      "step": 4673
    },
    {
      "epoch": 445.14285714285717,
      "grad_norm": 13.393614768981934,
      "learning_rate": 8.755622188905548e-07,
      "loss": 1.0562,
      "step": 4674
    },
    {
      "epoch": 445.23809523809524,
      "grad_norm": 9.072484970092773,
      "learning_rate": 8.757496251874064e-07,
      "loss": 1.0918,
      "step": 4675
    },
    {
      "epoch": 445.3333333333333,
      "grad_norm": 11.837943077087402,
      "learning_rate": 8.759370314842579e-07,
      "loss": 1.0745,
      "step": 4676
    },
    {
      "epoch": 445.42857142857144,
      "grad_norm": 10.07509994506836,
      "learning_rate": 8.761244377811095e-07,
      "loss": 1.0584,
      "step": 4677
    },
    {
      "epoch": 445.5238095238095,
      "grad_norm": 10.085092544555664,
      "learning_rate": 8.763118440779612e-07,
      "loss": 1.1189,
      "step": 4678
    },
    {
      "epoch": 445.6190476190476,
      "grad_norm": 8.869654655456543,
      "learning_rate": 8.764992503748126e-07,
      "loss": 1.0947,
      "step": 4679
    },
    {
      "epoch": 445.7142857142857,
      "grad_norm": 13.115325927734375,
      "learning_rate": 8.766866566716643e-07,
      "loss": 1.1693,
      "step": 4680
    },
    {
      "epoch": 445.8095238095238,
      "grad_norm": 9.584236145019531,
      "learning_rate": 8.768740629685157e-07,
      "loss": 1.1207,
      "step": 4681
    },
    {
      "epoch": 445.9047619047619,
      "grad_norm": 6.502092361450195,
      "learning_rate": 8.770614692653674e-07,
      "loss": 1.1005,
      "step": 4682
    },
    {
      "epoch": 446.0,
      "grad_norm": 23.737815856933594,
      "learning_rate": 8.772488755622189e-07,
      "loss": 1.0992,
      "step": 4683
    },
    {
      "epoch": 446.0,
      "eval_accuracy": 0.7493443237167479,
      "eval_f1": 0.7921714818266542,
      "eval_loss": 0.5422975420951843,
      "eval_precision": 0.7611940298507462,
      "eval_recall": 0.8257772020725389,
      "eval_roc_auc": 0.7999453080023028,
      "eval_runtime": 1.1243,
      "eval_samples_per_second": 2373.991,
      "eval_steps_per_second": 5.337,
      "step": 4683
    },
    {
      "epoch": 446.0952380952381,
      "grad_norm": 8.786201477050781,
      "learning_rate": 8.774362818590705e-07,
      "loss": 1.0786,
      "step": 4684
    },
    {
      "epoch": 446.1904761904762,
      "grad_norm": 7.7905449867248535,
      "learning_rate": 8.776236881559222e-07,
      "loss": 1.1466,
      "step": 4685
    },
    {
      "epoch": 446.2857142857143,
      "grad_norm": 29.007137298583984,
      "learning_rate": 8.778110944527736e-07,
      "loss": 1.0394,
      "step": 4686
    },
    {
      "epoch": 446.3809523809524,
      "grad_norm": 13.108345031738281,
      "learning_rate": 8.779985007496253e-07,
      "loss": 1.1375,
      "step": 4687
    },
    {
      "epoch": 446.4761904761905,
      "grad_norm": 6.3110833168029785,
      "learning_rate": 8.781859070464767e-07,
      "loss": 1.1257,
      "step": 4688
    },
    {
      "epoch": 446.57142857142856,
      "grad_norm": 14.847665786743164,
      "learning_rate": 8.783733133433284e-07,
      "loss": 1.0791,
      "step": 4689
    },
    {
      "epoch": 446.6666666666667,
      "grad_norm": 6.850281715393066,
      "learning_rate": 8.785607196401798e-07,
      "loss": 1.0899,
      "step": 4690
    },
    {
      "epoch": 446.76190476190476,
      "grad_norm": 7.7451558113098145,
      "learning_rate": 8.787481259370315e-07,
      "loss": 1.0919,
      "step": 4691
    },
    {
      "epoch": 446.85714285714283,
      "grad_norm": 7.732067108154297,
      "learning_rate": 8.789355322338832e-07,
      "loss": 1.0963,
      "step": 4692
    },
    {
      "epoch": 446.95238095238096,
      "grad_norm": 8.77704906463623,
      "learning_rate": 8.791229385307346e-07,
      "loss": 1.0686,
      "step": 4693
    },
    {
      "epoch": 446.95238095238096,
      "eval_accuracy": 0.7455976020981641,
      "eval_f1": 0.78932671424139,
      "eval_loss": 0.5424996614456177,
      "eval_precision": 0.7575938058368076,
      "eval_recall": 0.8238341968911918,
      "eval_roc_auc": 0.800234599884859,
      "eval_runtime": 1.1375,
      "eval_samples_per_second": 2346.357,
      "eval_steps_per_second": 5.275,
      "step": 4693
    },
    {
      "epoch": 447.04761904761904,
      "grad_norm": 8.154366493225098,
      "learning_rate": 8.793103448275863e-07,
      "loss": 1.0908,
      "step": 4694
    },
    {
      "epoch": 447.14285714285717,
      "grad_norm": 10.25619125366211,
      "learning_rate": 8.794977511244378e-07,
      "loss": 1.1514,
      "step": 4695
    },
    {
      "epoch": 447.23809523809524,
      "grad_norm": 6.668560028076172,
      "learning_rate": 8.796851574212894e-07,
      "loss": 1.1213,
      "step": 4696
    },
    {
      "epoch": 447.3333333333333,
      "grad_norm": 16.608366012573242,
      "learning_rate": 8.798725637181409e-07,
      "loss": 1.1177,
      "step": 4697
    },
    {
      "epoch": 447.42857142857144,
      "grad_norm": 19.34969139099121,
      "learning_rate": 8.800599700149926e-07,
      "loss": 1.0898,
      "step": 4698
    },
    {
      "epoch": 447.5238095238095,
      "grad_norm": 15.329229354858398,
      "learning_rate": 8.802473763118443e-07,
      "loss": 1.0809,
      "step": 4699
    },
    {
      "epoch": 447.6190476190476,
      "grad_norm": 6.398341178894043,
      "learning_rate": 8.804347826086957e-07,
      "loss": 1.0798,
      "step": 4700
    },
    {
      "epoch": 447.7142857142857,
      "grad_norm": 4.657569885253906,
      "learning_rate": 8.806221889055474e-07,
      "loss": 1.0742,
      "step": 4701
    },
    {
      "epoch": 447.8095238095238,
      "grad_norm": 11.059224128723145,
      "learning_rate": 8.808095952023988e-07,
      "loss": 1.0735,
      "step": 4702
    },
    {
      "epoch": 447.9047619047619,
      "grad_norm": 6.843538284301758,
      "learning_rate": 8.809970014992505e-07,
      "loss": 1.0878,
      "step": 4703
    },
    {
      "epoch": 448.0,
      "grad_norm": 16.965524673461914,
      "learning_rate": 8.811844077961019e-07,
      "loss": 1.1474,
      "step": 4704
    },
    {
      "epoch": 448.0,
      "eval_accuracy": 0.7452229299363057,
      "eval_f1": 0.7941888619854721,
      "eval_loss": 0.5430591702461243,
      "eval_precision": 0.7454545454545455,
      "eval_recall": 0.8497409326424871,
      "eval_roc_auc": 0.8003316062176166,
      "eval_runtime": 1.1346,
      "eval_samples_per_second": 2352.38,
      "eval_steps_per_second": 5.288,
      "step": 4704
    },
    {
      "epoch": 448.0952380952381,
      "grad_norm": 7.548295497894287,
      "learning_rate": 8.813718140929536e-07,
      "loss": 1.0681,
      "step": 4705
    },
    {
      "epoch": 448.1904761904762,
      "grad_norm": 11.476109504699707,
      "learning_rate": 8.815592203898052e-07,
      "loss": 1.115,
      "step": 4706
    },
    {
      "epoch": 448.2857142857143,
      "grad_norm": 14.373614311218262,
      "learning_rate": 8.817466266866567e-07,
      "loss": 1.1139,
      "step": 4707
    },
    {
      "epoch": 448.3809523809524,
      "grad_norm": 7.564189434051514,
      "learning_rate": 8.819340329835083e-07,
      "loss": 1.07,
      "step": 4708
    },
    {
      "epoch": 448.4761904761905,
      "grad_norm": 7.844211578369141,
      "learning_rate": 8.821214392803598e-07,
      "loss": 1.1186,
      "step": 4709
    },
    {
      "epoch": 448.57142857142856,
      "grad_norm": 14.013773918151855,
      "learning_rate": 8.823088455772115e-07,
      "loss": 1.1451,
      "step": 4710
    },
    {
      "epoch": 448.6666666666667,
      "grad_norm": 17.679235458374023,
      "learning_rate": 8.824962518740629e-07,
      "loss": 1.0677,
      "step": 4711
    },
    {
      "epoch": 448.76190476190476,
      "grad_norm": 18.348920822143555,
      "learning_rate": 8.826836581709146e-07,
      "loss": 1.1012,
      "step": 4712
    },
    {
      "epoch": 448.85714285714283,
      "grad_norm": 24.86972999572754,
      "learning_rate": 8.828710644677662e-07,
      "loss": 1.109,
      "step": 4713
    },
    {
      "epoch": 448.95238095238096,
      "grad_norm": 19.28120994567871,
      "learning_rate": 8.830584707646177e-07,
      "loss": 1.0822,
      "step": 4714
    },
    {
      "epoch": 448.95238095238096,
      "eval_accuracy": 0.7433495691270139,
      "eval_f1": 0.7857366280888333,
      "eval_loss": 0.5428649187088013,
      "eval_precision": 0.7598306110102844,
      "eval_recall": 0.8134715025906736,
      "eval_roc_auc": 0.7996755900978698,
      "eval_runtime": 1.1342,
      "eval_samples_per_second": 2353.18,
      "eval_steps_per_second": 5.29,
      "step": 4714
    },
    {
      "epoch": 449.04761904761904,
      "grad_norm": 13.300694465637207,
      "learning_rate": 8.832458770614693e-07,
      "loss": 1.1411,
      "step": 4715
    },
    {
      "epoch": 449.14285714285717,
      "grad_norm": 6.260472774505615,
      "learning_rate": 8.834332833583209e-07,
      "loss": 1.0958,
      "step": 4716
    },
    {
      "epoch": 449.23809523809524,
      "grad_norm": 15.211816787719727,
      "learning_rate": 8.836206896551724e-07,
      "loss": 1.0312,
      "step": 4717
    },
    {
      "epoch": 449.3333333333333,
      "grad_norm": 11.604256629943848,
      "learning_rate": 8.83808095952024e-07,
      "loss": 1.1285,
      "step": 4718
    },
    {
      "epoch": 449.42857142857144,
      "grad_norm": 29.347837448120117,
      "learning_rate": 8.839955022488757e-07,
      "loss": 1.0759,
      "step": 4719
    },
    {
      "epoch": 449.5238095238095,
      "grad_norm": 21.008678436279297,
      "learning_rate": 8.841829085457272e-07,
      "loss": 1.1071,
      "step": 4720
    },
    {
      "epoch": 449.6190476190476,
      "grad_norm": 25.13567543029785,
      "learning_rate": 8.843703148425788e-07,
      "loss": 1.0805,
      "step": 4721
    },
    {
      "epoch": 449.7142857142857,
      "grad_norm": 17.31960678100586,
      "learning_rate": 8.845577211394304e-07,
      "loss": 1.1092,
      "step": 4722
    },
    {
      "epoch": 449.8095238095238,
      "grad_norm": 28.314422607421875,
      "learning_rate": 8.847451274362819e-07,
      "loss": 1.146,
      "step": 4723
    },
    {
      "epoch": 449.9047619047619,
      "grad_norm": 20.834762573242188,
      "learning_rate": 8.849325337331335e-07,
      "loss": 1.1115,
      "step": 4724
    },
    {
      "epoch": 450.0,
      "grad_norm": 6.921756267547607,
      "learning_rate": 8.85119940029985e-07,
      "loss": 1.1164,
      "step": 4725
    },
    {
      "epoch": 450.0,
      "eval_accuracy": 0.741476208317722,
      "eval_f1": 0.783427495291902,
      "eval_loss": 0.5428313612937927,
      "eval_precision": 0.7600487210718636,
      "eval_recall": 0.8082901554404145,
      "eval_roc_auc": 0.7998077144502015,
      "eval_runtime": 1.437,
      "eval_samples_per_second": 1857.39,
      "eval_steps_per_second": 4.175,
      "step": 4725
    },
    {
      "epoch": 450.0952380952381,
      "grad_norm": 8.552742004394531,
      "learning_rate": 8.853073463268366e-07,
      "loss": 1.1103,
      "step": 4726
    },
    {
      "epoch": 450.1904761904762,
      "grad_norm": 27.3271541595459,
      "learning_rate": 8.854947526236883e-07,
      "loss": 1.1035,
      "step": 4727
    },
    {
      "epoch": 450.2857142857143,
      "grad_norm": 21.7481689453125,
      "learning_rate": 8.856821589205398e-07,
      "loss": 1.1017,
      "step": 4728
    },
    {
      "epoch": 450.3809523809524,
      "grad_norm": 17.90006446838379,
      "learning_rate": 8.858695652173914e-07,
      "loss": 1.0786,
      "step": 4729
    },
    {
      "epoch": 450.4761904761905,
      "grad_norm": 16.27367401123047,
      "learning_rate": 8.860569715142429e-07,
      "loss": 1.1039,
      "step": 4730
    },
    {
      "epoch": 450.57142857142856,
      "grad_norm": 7.353328227996826,
      "learning_rate": 8.862443778110945e-07,
      "loss": 1.0386,
      "step": 4731
    },
    {
      "epoch": 450.6666666666667,
      "grad_norm": 10.046563148498535,
      "learning_rate": 8.86431784107946e-07,
      "loss": 1.0887,
      "step": 4732
    },
    {
      "epoch": 450.76190476190476,
      "grad_norm": 5.69588041305542,
      "learning_rate": 8.866191904047976e-07,
      "loss": 1.1012,
      "step": 4733
    },
    {
      "epoch": 450.85714285714283,
      "grad_norm": 10.160133361816406,
      "learning_rate": 8.868065967016493e-07,
      "loss": 1.0874,
      "step": 4734
    },
    {
      "epoch": 450.95238095238096,
      "grad_norm": 23.668718338012695,
      "learning_rate": 8.869940029985007e-07,
      "loss": 1.1325,
      "step": 4735
    },
    {
      "epoch": 450.95238095238096,
      "eval_accuracy": 0.7482203072311727,
      "eval_f1": 0.8015357353809806,
      "eval_loss": 0.5460059642791748,
      "eval_precision": 0.7366992399565689,
      "eval_recall": 0.8788860103626943,
      "eval_roc_auc": 0.7993002302820955,
      "eval_runtime": 1.104,
      "eval_samples_per_second": 2417.465,
      "eval_steps_per_second": 5.435,
      "step": 4735
    }
  ],
  "logging_steps": 1,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 500,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8974993938336973e+17,
  "train_batch_size": 512,
  "trial_name": null,
  "trial_params": null
}