{
  "best_metric": 0.4437299035369775,
  "best_model_checkpoint": "../save/jtrans-malware/checkpoint-3000",
  "epoch": 9.043927648578812,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 9.974160206718347e-05,
      "loss": 0.434,
      "step": 10
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.948320413436693e-05,
      "loss": 0.222,
      "step": 20
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.92248062015504e-05,
      "loss": 0.1538,
      "step": 30
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.896640826873386e-05,
      "loss": 0.1393,
      "step": 40
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.870801033591732e-05,
      "loss": 0.1326,
      "step": 50
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.844961240310078e-05,
      "loss": 0.1337,
      "step": 60
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.819121447028425e-05,
      "loss": 0.1312,
      "step": 70
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.793281653746771e-05,
      "loss": 0.1239,
      "step": 80
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.767441860465116e-05,
      "loss": 0.1261,
      "step": 90
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.741602067183462e-05,
      "loss": 0.1246,
      "step": 100
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.9733742331288343,
      "eval_f1": 0.09958506224066391,
      "eval_loss": 0.13609381020069122,
      "eval_precision": 0.9230769230769231,
      "eval_recall": 0.05263157894736842,
      "eval_runtime": 11.6719,
      "eval_samples_per_second": 13.965,
      "eval_steps_per_second": 0.942,
      "step": 100
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.71576227390181e-05,
      "loss": 0.1192,
      "step": 110
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.689922480620155e-05,
      "loss": 0.1341,
      "step": 120
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.664082687338501e-05,
      "loss": 0.1231,
      "step": 130
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.638242894056848e-05,
      "loss": 0.1399,
      "step": 140
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.612403100775195e-05,
      "loss": 0.1262,
      "step": 150
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.58656330749354e-05,
      "loss": 0.133,
      "step": 160
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.560723514211886e-05,
      "loss": 0.1247,
      "step": 170
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.534883720930233e-05,
      "loss": 0.129,
      "step": 180
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.509043927648579e-05,
      "loss": 0.1275,
      "step": 190
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.483204134366925e-05,
      "loss": 0.1267,
      "step": 200
    },
    {
      "epoch": 0.52,
      "eval_accuracy": 0.9720245398773006,
      "eval_f1": 0.0,
      "eval_loss": 0.12009111791849136,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 11.6034,
      "eval_samples_per_second": 14.048,
      "eval_steps_per_second": 0.948,
      "step": 200
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.457364341085272e-05,
      "loss": 0.1257,
      "step": 210
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.431524547803618e-05,
      "loss": 0.121,
      "step": 220
    },
    {
      "epoch": 0.59,
      "learning_rate": 9.405684754521964e-05,
      "loss": 0.1238,
      "step": 230
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.379844961240311e-05,
      "loss": 0.1284,
      "step": 240
    },
    {
      "epoch": 0.65,
      "learning_rate": 9.354005167958657e-05,
      "loss": 0.1206,
      "step": 250
    },
    {
      "epoch": 0.67,
      "learning_rate": 9.328165374677002e-05,
      "loss": 0.1288,
      "step": 260
    },
    {
      "epoch": 0.7,
      "learning_rate": 9.30232558139535e-05,
      "loss": 0.1151,
      "step": 270
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.276485788113696e-05,
      "loss": 0.1144,
      "step": 280
    },
    {
      "epoch": 0.75,
      "learning_rate": 9.250645994832042e-05,
      "loss": 0.13,
      "step": 290
    },
    {
      "epoch": 0.78,
      "learning_rate": 9.224806201550387e-05,
      "loss": 0.1202,
      "step": 300
    },
    {
      "epoch": 0.78,
      "eval_accuracy": 0.9734969325153374,
      "eval_f1": 0.10743801652892561,
      "eval_loss": 0.11869814246892929,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.05701754385964912,
      "eval_runtime": 11.8735,
      "eval_samples_per_second": 13.728,
      "eval_steps_per_second": 0.926,
      "step": 300
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.198966408268735e-05,
      "loss": 0.1218,
      "step": 310
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.173126614987081e-05,
      "loss": 0.1218,
      "step": 320
    },
    {
      "epoch": 0.85,
      "learning_rate": 9.147286821705426e-05,
      "loss": 0.1197,
      "step": 330
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.121447028423772e-05,
      "loss": 0.1221,
      "step": 340
    },
    {
      "epoch": 0.9,
      "learning_rate": 9.09560723514212e-05,
      "loss": 0.1223,
      "step": 350
    },
    {
      "epoch": 0.93,
      "learning_rate": 9.069767441860465e-05,
      "loss": 0.12,
      "step": 360
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.043927648578811e-05,
      "loss": 0.1113,
      "step": 370
    },
    {
      "epoch": 0.98,
      "learning_rate": 9.018087855297158e-05,
      "loss": 0.121,
      "step": 380
    },
    {
      "epoch": 1.01,
      "learning_rate": 8.992248062015505e-05,
      "loss": 0.1195,
      "step": 390
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.96640826873385e-05,
      "loss": 0.1227,
      "step": 400
    },
    {
      "epoch": 1.03,
      "eval_accuracy": 0.9734969325153374,
      "eval_f1": 0.10743801652892561,
      "eval_loss": 0.11648325622081757,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.05701754385964912,
      "eval_runtime": 11.6629,
      "eval_samples_per_second": 13.976,
      "eval_steps_per_second": 0.943,
      "step": 400
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.940568475452197e-05,
      "loss": 0.118,
      "step": 410
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.914728682170543e-05,
      "loss": 0.12,
      "step": 420
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.1223,
      "step": 430
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.863049095607236e-05,
      "loss": 0.1154,
      "step": 440
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.837209302325582e-05,
      "loss": 0.1124,
      "step": 450
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.811369509043928e-05,
      "loss": 0.1095,
      "step": 460
    },
    {
      "epoch": 1.21,
      "learning_rate": 8.785529715762275e-05,
      "loss": 0.1159,
      "step": 470
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.759689922480621e-05,
      "loss": 0.1229,
      "step": 480
    },
    {
      "epoch": 1.27,
      "learning_rate": 8.733850129198967e-05,
      "loss": 0.1185,
      "step": 490
    },
    {
      "epoch": 1.29,
      "learning_rate": 8.708010335917312e-05,
      "loss": 0.1139,
      "step": 500
    },
    {
      "epoch": 1.29,
      "eval_accuracy": 0.9734969325153374,
      "eval_f1": 0.10743801652892561,
      "eval_loss": 0.11325845867395401,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.05701754385964912,
      "eval_runtime": 11.9509,
      "eval_samples_per_second": 13.639,
      "eval_steps_per_second": 0.92,
      "step": 500
    },
    {
      "epoch": 1.32,
      "learning_rate": 8.68217054263566e-05,
      "loss": 0.1167,
      "step": 510
    },
    {
      "epoch": 1.34,
      "learning_rate": 8.656330749354006e-05,
      "loss": 0.119,
      "step": 520
    },
    {
      "epoch": 1.37,
      "learning_rate": 8.630490956072352e-05,
      "loss": 0.1115,
      "step": 530
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.604651162790697e-05,
      "loss": 0.1108,
      "step": 540
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.578811369509044e-05,
      "loss": 0.1194,
      "step": 550
    },
    {
      "epoch": 1.45,
      "learning_rate": 8.552971576227391e-05,
      "loss": 0.1147,
      "step": 560
    },
    {
      "epoch": 1.47,
      "learning_rate": 8.527131782945736e-05,
      "loss": 0.1171,
      "step": 570
    },
    {
      "epoch": 1.5,
      "learning_rate": 8.501291989664083e-05,
      "loss": 0.1218,
      "step": 580
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.475452196382429e-05,
      "loss": 0.1211,
      "step": 590
    },
    {
      "epoch": 1.55,
      "learning_rate": 8.449612403100775e-05,
      "loss": 0.1102,
      "step": 600
    },
    {
      "epoch": 1.55,
      "eval_accuracy": 0.9734969325153374,
      "eval_f1": 0.10743801652892561,
      "eval_loss": 0.10692530870437622,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.05701754385964912,
      "eval_runtime": 11.9673,
      "eval_samples_per_second": 13.62,
      "eval_steps_per_second": 0.919,
      "step": 600
    },
    {
      "epoch": 1.58,
      "learning_rate": 8.423772609819122e-05,
      "loss": 0.1061,
      "step": 610
    },
    {
      "epoch": 1.6,
      "learning_rate": 8.397932816537468e-05,
      "loss": 0.1062,
      "step": 620
    },
    {
      "epoch": 1.63,
      "learning_rate": 8.372093023255814e-05,
      "loss": 0.1148,
      "step": 630
    },
    {
      "epoch": 1.65,
      "learning_rate": 8.34625322997416e-05,
      "loss": 0.1114,
      "step": 640
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.320413436692507e-05,
      "loss": 0.1117,
      "step": 650
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.294573643410853e-05,
      "loss": 0.109,
      "step": 660
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.2687338501292e-05,
      "loss": 0.1042,
      "step": 670
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.242894056847546e-05,
      "loss": 0.1153,
      "step": 680
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.217054263565892e-05,
      "loss": 0.1026,
      "step": 690
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.191214470284238e-05,
      "loss": 0.1102,
      "step": 700
    },
    {
      "epoch": 1.81,
      "eval_accuracy": 0.9734969325153374,
      "eval_f1": 0.10743801652892561,
      "eval_loss": 0.10341155529022217,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.05701754385964912,
      "eval_runtime": 11.7838,
      "eval_samples_per_second": 13.833,
      "eval_steps_per_second": 0.933,
      "step": 700
    },
    {
      "epoch": 1.83,
      "learning_rate": 8.165374677002583e-05,
      "loss": 0.1002,
      "step": 710
    },
    {
      "epoch": 1.86,
      "learning_rate": 8.139534883720931e-05,
      "loss": 0.1159,
      "step": 720
    },
    {
      "epoch": 1.89,
      "learning_rate": 8.113695090439277e-05,
      "loss": 0.1098,
      "step": 730
    },
    {
      "epoch": 1.91,
      "learning_rate": 8.087855297157622e-05,
      "loss": 0.1057,
      "step": 740
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.062015503875969e-05,
      "loss": 0.1024,
      "step": 750
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.036175710594316e-05,
      "loss": 0.1013,
      "step": 760
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.010335917312663e-05,
      "loss": 0.0976,
      "step": 770
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.984496124031008e-05,
      "loss": 0.1043,
      "step": 780
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.958656330749354e-05,
      "loss": 0.1005,
      "step": 790
    },
    {
      "epoch": 2.07,
      "learning_rate": 7.932816537467702e-05,
      "loss": 0.0986,
      "step": 800
    },
    {
      "epoch": 2.07,
      "eval_accuracy": 0.9738650306748466,
      "eval_f1": 0.13061224489795917,
      "eval_loss": 0.10085967928171158,
      "eval_precision": 0.9411764705882353,
      "eval_recall": 0.07017543859649122,
      "eval_runtime": 11.946,
      "eval_samples_per_second": 13.645,
      "eval_steps_per_second": 0.921,
      "step": 800
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.906976744186047e-05,
      "loss": 0.0978,
      "step": 810
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.881136950904393e-05,
      "loss": 0.0943,
      "step": 820
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.855297157622739e-05,
      "loss": 0.0916,
      "step": 830
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.829457364341086e-05,
      "loss": 0.1007,
      "step": 840
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.803617571059432e-05,
      "loss": 0.1053,
      "step": 850
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.777777777777778e-05,
      "loss": 0.0992,
      "step": 860
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.751937984496124e-05,
      "loss": 0.1038,
      "step": 870
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.726098191214471e-05,
      "loss": 0.0982,
      "step": 880
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.700258397932817e-05,
      "loss": 0.1031,
      "step": 890
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.674418604651163e-05,
      "loss": 0.0964,
      "step": 900
    },
    {
      "epoch": 2.33,
      "eval_accuracy": 0.9749693251533742,
      "eval_f1": 0.20930232558139536,
      "eval_loss": 0.09663821756839752,
      "eval_precision": 0.9,
      "eval_recall": 0.11842105263157894,
      "eval_runtime": 11.6402,
      "eval_samples_per_second": 14.003,
      "eval_steps_per_second": 0.945,
      "step": 900
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.648578811369508e-05,
      "loss": 0.0991,
      "step": 910
    },
    {
      "epoch": 2.38,
      "learning_rate": 7.622739018087856e-05,
      "loss": 0.0971,
      "step": 920
    },
    {
      "epoch": 2.4,
      "learning_rate": 7.596899224806202e-05,
      "loss": 0.093,
      "step": 930
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.571059431524549e-05,
      "loss": 0.0965,
      "step": 940
    },
    {
      "epoch": 2.45,
      "learning_rate": 7.545219638242894e-05,
      "loss": 0.0949,
      "step": 950
    },
    {
      "epoch": 2.48,
      "learning_rate": 7.519379844961241e-05,
      "loss": 0.0941,
      "step": 960
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.493540051679588e-05,
      "loss": 0.0985,
      "step": 970
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.467700258397933e-05,
      "loss": 0.0999,
      "step": 980
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.441860465116279e-05,
      "loss": 0.1096,
      "step": 990
    },
    {
      "epoch": 2.58,
      "learning_rate": 7.416020671834627e-05,
      "loss": 0.0994,
      "step": 1000
    },
    {
      "epoch": 2.58,
      "eval_accuracy": 0.9750920245398773,
      "eval_f1": 0.20392156862745098,
      "eval_loss": 0.094399094581604,
      "eval_precision": 0.9629629629629629,
      "eval_recall": 0.11403508771929824,
      "eval_runtime": 11.5607,
      "eval_samples_per_second": 14.1,
      "eval_steps_per_second": 0.952,
      "step": 1000
    },
    {
      "epoch": 2.61,
      "learning_rate": 7.390180878552973e-05,
      "loss": 0.1052,
      "step": 1010
    },
    {
      "epoch": 2.64,
      "learning_rate": 7.364341085271318e-05,
      "loss": 0.0974,
      "step": 1020
    },
    {
      "epoch": 2.66,
      "learning_rate": 7.338501291989664e-05,
      "loss": 0.099,
      "step": 1030
    },
    {
      "epoch": 2.69,
      "learning_rate": 7.31266149870801e-05,
      "loss": 0.0999,
      "step": 1040
    },
    {
      "epoch": 2.71,
      "learning_rate": 7.286821705426357e-05,
      "loss": 0.1039,
      "step": 1050
    },
    {
      "epoch": 2.74,
      "learning_rate": 7.260981912144703e-05,
      "loss": 0.0935,
      "step": 1060
    },
    {
      "epoch": 2.76,
      "learning_rate": 7.23514211886305e-05,
      "loss": 0.1,
      "step": 1070
    },
    {
      "epoch": 2.79,
      "learning_rate": 7.209302325581396e-05,
      "loss": 0.0921,
      "step": 1080
    },
    {
      "epoch": 2.82,
      "learning_rate": 7.183462532299742e-05,
      "loss": 0.0941,
      "step": 1090
    },
    {
      "epoch": 2.84,
      "learning_rate": 7.157622739018088e-05,
      "loss": 0.0913,
      "step": 1100
    },
    {
      "epoch": 2.84,
      "eval_accuracy": 0.9757055214723926,
      "eval_f1": 0.23846153846153845,
      "eval_loss": 0.09476516395807266,
      "eval_precision": 0.96875,
      "eval_recall": 0.13596491228070176,
      "eval_runtime": 11.7524,
      "eval_samples_per_second": 13.87,
      "eval_steps_per_second": 0.936,
      "step": 1100
    },
    {
      "epoch": 2.87,
      "learning_rate": 7.131782945736435e-05,
      "loss": 0.0944,
      "step": 1110
    },
    {
      "epoch": 2.89,
      "learning_rate": 7.105943152454781e-05,
      "loss": 0.0946,
      "step": 1120
    },
    {
      "epoch": 2.92,
      "learning_rate": 7.080103359173127e-05,
      "loss": 0.0842,
      "step": 1130
    },
    {
      "epoch": 2.95,
      "learning_rate": 7.054263565891474e-05,
      "loss": 0.0896,
      "step": 1140
    },
    {
      "epoch": 2.97,
      "learning_rate": 7.028423772609819e-05,
      "loss": 0.0977,
      "step": 1150
    },
    {
      "epoch": 3.0,
      "learning_rate": 7.002583979328165e-05,
      "loss": 0.1078,
      "step": 1160
    },
    {
      "epoch": 3.02,
      "learning_rate": 6.976744186046513e-05,
      "loss": 0.0854,
      "step": 1170
    },
    {
      "epoch": 3.05,
      "learning_rate": 6.950904392764859e-05,
      "loss": 0.0872,
      "step": 1180
    },
    {
      "epoch": 3.07,
      "learning_rate": 6.925064599483204e-05,
      "loss": 0.1017,
      "step": 1190
    },
    {
      "epoch": 3.1,
      "learning_rate": 6.89922480620155e-05,
      "loss": 0.0964,
      "step": 1200
    },
    {
      "epoch": 3.1,
      "eval_accuracy": 0.9737423312883435,
      "eval_f1": 0.2014925373134328,
      "eval_loss": 0.09359467029571533,
      "eval_precision": 0.675,
      "eval_recall": 0.11842105263157894,
      "eval_runtime": 11.5484,
      "eval_samples_per_second": 14.115,
      "eval_steps_per_second": 0.953,
      "step": 1200
    },
    {
      "epoch": 3.13,
      "learning_rate": 6.873385012919898e-05,
      "loss": 0.092,
      "step": 1210
    },
    {
      "epoch": 3.15,
      "learning_rate": 6.847545219638243e-05,
      "loss": 0.0919,
      "step": 1220
    },
    {
      "epoch": 3.18,
      "learning_rate": 6.821705426356589e-05,
      "loss": 0.0872,
      "step": 1230
    },
    {
      "epoch": 3.2,
      "learning_rate": 6.795865633074935e-05,
      "loss": 0.0822,
      "step": 1240
    },
    {
      "epoch": 3.23,
      "learning_rate": 6.770025839793283e-05,
      "loss": 0.0931,
      "step": 1250
    },
    {
      "epoch": 3.26,
      "learning_rate": 6.744186046511628e-05,
      "loss": 0.0878,
      "step": 1260
    },
    {
      "epoch": 3.28,
      "learning_rate": 6.718346253229974e-05,
      "loss": 0.0896,
      "step": 1270
    },
    {
      "epoch": 3.31,
      "learning_rate": 6.69250645994832e-05,
      "loss": 0.0917,
      "step": 1280
    },
    {
      "epoch": 3.33,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0949,
      "step": 1290
    },
    {
      "epoch": 3.36,
      "learning_rate": 6.640826873385013e-05,
      "loss": 0.1,
      "step": 1300
    },
    {
      "epoch": 3.36,
      "eval_accuracy": 0.9759509202453988,
      "eval_f1": 0.2740740740740741,
      "eval_loss": 0.08845529705286026,
      "eval_precision": 0.8809523809523809,
      "eval_recall": 0.16228070175438597,
      "eval_runtime": 11.5355,
      "eval_samples_per_second": 14.13,
      "eval_steps_per_second": 0.954,
      "step": 1300
    },
    {
      "epoch": 3.39,
      "learning_rate": 6.61498708010336e-05,
      "loss": 0.092,
      "step": 1310
    },
    {
      "epoch": 3.41,
      "learning_rate": 6.589147286821705e-05,
      "loss": 0.09,
      "step": 1320
    },
    {
      "epoch": 3.44,
      "learning_rate": 6.563307493540052e-05,
      "loss": 0.0844,
      "step": 1330
    },
    {
      "epoch": 3.46,
      "learning_rate": 6.537467700258399e-05,
      "loss": 0.087,
      "step": 1340
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.511627906976745e-05,
      "loss": 0.0792,
      "step": 1350
    },
    {
      "epoch": 3.51,
      "learning_rate": 6.48578811369509e-05,
      "loss": 0.0868,
      "step": 1360
    },
    {
      "epoch": 3.54,
      "learning_rate": 6.459948320413438e-05,
      "loss": 0.0892,
      "step": 1370
    },
    {
      "epoch": 3.57,
      "learning_rate": 6.434108527131784e-05,
      "loss": 0.0792,
      "step": 1380
    },
    {
      "epoch": 3.59,
      "learning_rate": 6.408268733850129e-05,
      "loss": 0.0876,
      "step": 1390
    },
    {
      "epoch": 3.62,
      "learning_rate": 6.382428940568475e-05,
      "loss": 0.0982,
      "step": 1400
    },
    {
      "epoch": 3.62,
      "eval_accuracy": 0.9753374233128834,
      "eval_f1": 0.22988505747126434,
      "eval_loss": 0.09349281340837479,
      "eval_precision": 0.9090909090909091,
      "eval_recall": 0.13157894736842105,
      "eval_runtime": 11.6178,
      "eval_samples_per_second": 14.03,
      "eval_steps_per_second": 0.947,
      "step": 1400
    },
    {
      "epoch": 3.64,
      "learning_rate": 6.356589147286823e-05,
      "loss": 0.0905,
      "step": 1410
    },
    {
      "epoch": 3.67,
      "learning_rate": 6.330749354005169e-05,
      "loss": 0.0881,
      "step": 1420
    },
    {
      "epoch": 3.7,
      "learning_rate": 6.304909560723514e-05,
      "loss": 0.0849,
      "step": 1430
    },
    {
      "epoch": 3.72,
      "learning_rate": 6.27906976744186e-05,
      "loss": 0.0829,
      "step": 1440
    },
    {
      "epoch": 3.75,
      "learning_rate": 6.253229974160208e-05,
      "loss": 0.0992,
      "step": 1450
    },
    {
      "epoch": 3.77,
      "learning_rate": 6.227390180878553e-05,
      "loss": 0.0998,
      "step": 1460
    },
    {
      "epoch": 3.8,
      "learning_rate": 6.201550387596899e-05,
      "loss": 0.0966,
      "step": 1470
    },
    {
      "epoch": 3.82,
      "learning_rate": 6.175710594315246e-05,
      "loss": 0.0833,
      "step": 1480
    },
    {
      "epoch": 3.85,
      "learning_rate": 6.149870801033592e-05,
      "loss": 0.0914,
      "step": 1490
    },
    {
      "epoch": 3.88,
      "learning_rate": 6.124031007751938e-05,
      "loss": 0.0785,
      "step": 1500
    },
    {
      "epoch": 3.88,
      "eval_accuracy": 0.9759509202453988,
      "eval_f1": 0.2949640287769784,
      "eval_loss": 0.08881861716508865,
      "eval_precision": 0.82,
      "eval_recall": 0.17982456140350878,
      "eval_runtime": 11.8266,
      "eval_samples_per_second": 13.783,
      "eval_steps_per_second": 0.93,
      "step": 1500
    },
    {
      "epoch": 3.9,
      "learning_rate": 6.0981912144702846e-05,
      "loss": 0.0993,
      "step": 1510
    },
    {
      "epoch": 3.93,
      "learning_rate": 6.072351421188631e-05,
      "loss": 0.0951,
      "step": 1520
    },
    {
      "epoch": 3.95,
      "learning_rate": 6.0465116279069765e-05,
      "loss": 0.0937,
      "step": 1530
    },
    {
      "epoch": 3.98,
      "learning_rate": 6.0206718346253235e-05,
      "loss": 0.089,
      "step": 1540
    },
    {
      "epoch": 4.01,
      "learning_rate": 5.99483204134367e-05,
      "loss": 0.0891,
      "step": 1550
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.9689922480620155e-05,
      "loss": 0.0856,
      "step": 1560
    },
    {
      "epoch": 4.06,
      "learning_rate": 5.943152454780362e-05,
      "loss": 0.0832,
      "step": 1570
    },
    {
      "epoch": 4.08,
      "learning_rate": 5.917312661498709e-05,
      "loss": 0.088,
      "step": 1580
    },
    {
      "epoch": 4.11,
      "learning_rate": 5.891472868217055e-05,
      "loss": 0.0829,
      "step": 1590
    },
    {
      "epoch": 4.13,
      "learning_rate": 5.865633074935401e-05,
      "loss": 0.0859,
      "step": 1600
    },
    {
      "epoch": 4.13,
      "eval_accuracy": 0.9759509202453988,
      "eval_f1": 0.3,
      "eval_loss": 0.08691960573196411,
      "eval_precision": 0.8076923076923077,
      "eval_recall": 0.18421052631578946,
      "eval_runtime": 11.8328,
      "eval_samples_per_second": 13.775,
      "eval_steps_per_second": 0.93,
      "step": 1600
    },
    {
      "epoch": 4.16,
      "learning_rate": 5.839793281653747e-05,
      "loss": 0.0927,
      "step": 1610
    },
    {
      "epoch": 4.19,
      "learning_rate": 5.8139534883720933e-05,
      "loss": 0.0904,
      "step": 1620
    },
    {
      "epoch": 4.21,
      "learning_rate": 5.788113695090439e-05,
      "loss": 0.0891,
      "step": 1630
    },
    {
      "epoch": 4.24,
      "learning_rate": 5.762273901808786e-05,
      "loss": 0.0819,
      "step": 1640
    },
    {
      "epoch": 4.26,
      "learning_rate": 5.736434108527132e-05,
      "loss": 0.0868,
      "step": 1650
    },
    {
      "epoch": 4.29,
      "learning_rate": 5.7105943152454786e-05,
      "loss": 0.0909,
      "step": 1660
    },
    {
      "epoch": 4.32,
      "learning_rate": 5.684754521963824e-05,
      "loss": 0.0797,
      "step": 1670
    },
    {
      "epoch": 4.34,
      "learning_rate": 5.6589147286821706e-05,
      "loss": 0.0863,
      "step": 1680
    },
    {
      "epoch": 4.37,
      "learning_rate": 5.6330749354005176e-05,
      "loss": 0.0849,
      "step": 1690
    },
    {
      "epoch": 4.39,
      "learning_rate": 5.607235142118863e-05,
      "loss": 0.0794,
      "step": 1700
    },
    {
      "epoch": 4.39,
      "eval_accuracy": 0.976319018404908,
      "eval_f1": 0.3228070175438596,
      "eval_loss": 0.08520764857530594,
      "eval_precision": 0.8070175438596491,
      "eval_recall": 0.20175438596491227,
      "eval_runtime": 11.7131,
      "eval_samples_per_second": 13.916,
      "eval_steps_per_second": 0.939,
      "step": 1700
    },
    {
      "epoch": 4.42,
      "learning_rate": 5.5813953488372095e-05,
      "loss": 0.0857,
      "step": 1710
    },
    {
      "epoch": 4.44,
      "learning_rate": 5.555555555555556e-05,
      "loss": 0.0876,
      "step": 1720
    },
    {
      "epoch": 4.47,
      "learning_rate": 5.5297157622739015e-05,
      "loss": 0.0807,
      "step": 1730
    },
    {
      "epoch": 4.5,
      "learning_rate": 5.503875968992248e-05,
      "loss": 0.0903,
      "step": 1740
    },
    {
      "epoch": 4.52,
      "learning_rate": 5.478036175710595e-05,
      "loss": 0.0804,
      "step": 1750
    },
    {
      "epoch": 4.55,
      "learning_rate": 5.452196382428941e-05,
      "loss": 0.0921,
      "step": 1760
    },
    {
      "epoch": 4.57,
      "learning_rate": 5.426356589147287e-05,
      "loss": 0.0734,
      "step": 1770
    },
    {
      "epoch": 4.6,
      "learning_rate": 5.400516795865633e-05,
      "loss": 0.0778,
      "step": 1780
    },
    {
      "epoch": 4.63,
      "learning_rate": 5.37467700258398e-05,
      "loss": 0.0833,
      "step": 1790
    },
    {
      "epoch": 4.65,
      "learning_rate": 5.348837209302326e-05,
      "loss": 0.0835,
      "step": 1800
    },
    {
      "epoch": 4.65,
      "eval_accuracy": 0.9773006134969325,
      "eval_f1": 0.37288135593220334,
      "eval_loss": 0.08421062678098679,
      "eval_precision": 0.8208955223880597,
      "eval_recall": 0.2412280701754386,
      "eval_runtime": 11.9185,
      "eval_samples_per_second": 13.676,
      "eval_steps_per_second": 0.923,
      "step": 1800
    },
    {
      "epoch": 4.68,
      "learning_rate": 5.322997416020672e-05,
      "loss": 0.0774,
      "step": 1810
    },
    {
      "epoch": 4.7,
      "learning_rate": 5.297157622739018e-05,
      "loss": 0.0829,
      "step": 1820
    },
    {
      "epoch": 4.73,
      "learning_rate": 5.271317829457365e-05,
      "loss": 0.0821,
      "step": 1830
    },
    {
      "epoch": 4.75,
      "learning_rate": 5.24547803617571e-05,
      "loss": 0.0827,
      "step": 1840
    },
    {
      "epoch": 4.78,
      "learning_rate": 5.219638242894057e-05,
      "loss": 0.0836,
      "step": 1850
    },
    {
      "epoch": 4.81,
      "learning_rate": 5.1937984496124036e-05,
      "loss": 0.0767,
      "step": 1860
    },
    {
      "epoch": 4.83,
      "learning_rate": 5.167958656330749e-05,
      "loss": 0.0769,
      "step": 1870
    },
    {
      "epoch": 4.86,
      "learning_rate": 5.1421188630490955e-05,
      "loss": 0.0823,
      "step": 1880
    },
    {
      "epoch": 4.88,
      "learning_rate": 5.1162790697674425e-05,
      "loss": 0.0833,
      "step": 1890
    },
    {
      "epoch": 4.91,
      "learning_rate": 5.090439276485789e-05,
      "loss": 0.0831,
      "step": 1900
    },
    {
      "epoch": 4.91,
      "eval_accuracy": 0.9775460122699386,
      "eval_f1": 0.37542662116040953,
      "eval_loss": 0.0831713080406189,
      "eval_precision": 0.8461538461538461,
      "eval_recall": 0.2412280701754386,
      "eval_runtime": 11.6419,
      "eval_samples_per_second": 14.001,
      "eval_steps_per_second": 0.945,
      "step": 1900
    },
    {
      "epoch": 4.94,
      "learning_rate": 5.0645994832041345e-05,
      "loss": 0.0885,
      "step": 1910
    },
    {
      "epoch": 4.96,
      "learning_rate": 5.038759689922481e-05,
      "loss": 0.0838,
      "step": 1920
    },
    {
      "epoch": 4.99,
      "learning_rate": 5.012919896640828e-05,
      "loss": 0.0797,
      "step": 1930
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.9870801033591734e-05,
      "loss": 0.0734,
      "step": 1940
    },
    {
      "epoch": 5.04,
      "learning_rate": 4.96124031007752e-05,
      "loss": 0.075,
      "step": 1950
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.935400516795866e-05,
      "loss": 0.0721,
      "step": 1960
    },
    {
      "epoch": 5.09,
      "learning_rate": 4.9095607235142123e-05,
      "loss": 0.0746,
      "step": 1970
    },
    {
      "epoch": 5.12,
      "learning_rate": 4.883720930232558e-05,
      "loss": 0.0678,
      "step": 1980
    },
    {
      "epoch": 5.14,
      "learning_rate": 4.857881136950905e-05,
      "loss": 0.077,
      "step": 1990
    },
    {
      "epoch": 5.17,
      "learning_rate": 4.8320413436692506e-05,
      "loss": 0.0763,
      "step": 2000
    },
    {
      "epoch": 5.17,
      "eval_accuracy": 0.9771779141104294,
      "eval_f1": 0.3673469387755102,
      "eval_loss": 0.08154629170894623,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.23684210526315788,
      "eval_runtime": 11.9474,
      "eval_samples_per_second": 13.643,
      "eval_steps_per_second": 0.921,
      "step": 2000
    },
    {
      "epoch": 5.19,
      "learning_rate": 4.8062015503875976e-05,
      "loss": 0.0803,
      "step": 2010
    },
    {
      "epoch": 5.22,
      "learning_rate": 4.780361757105943e-05,
      "loss": 0.086,
      "step": 2020
    },
    {
      "epoch": 5.25,
      "learning_rate": 4.7545219638242896e-05,
      "loss": 0.0766,
      "step": 2030
    },
    {
      "epoch": 5.27,
      "learning_rate": 4.728682170542636e-05,
      "loss": 0.0832,
      "step": 2040
    },
    {
      "epoch": 5.3,
      "learning_rate": 4.702842377260982e-05,
      "loss": 0.0817,
      "step": 2050
    },
    {
      "epoch": 5.32,
      "learning_rate": 4.6770025839793285e-05,
      "loss": 0.0825,
      "step": 2060
    },
    {
      "epoch": 5.35,
      "learning_rate": 4.651162790697675e-05,
      "loss": 0.0751,
      "step": 2070
    },
    {
      "epoch": 5.37,
      "learning_rate": 4.625322997416021e-05,
      "loss": 0.0719,
      "step": 2080
    },
    {
      "epoch": 5.4,
      "learning_rate": 4.5994832041343674e-05,
      "loss": 0.0826,
      "step": 2090
    },
    {
      "epoch": 5.43,
      "learning_rate": 4.573643410852713e-05,
      "loss": 0.0784,
      "step": 2100
    },
    {
      "epoch": 5.43,
      "eval_accuracy": 0.9773006134969325,
      "eval_f1": 0.3642611683848797,
      "eval_loss": 0.08145377784967422,
      "eval_precision": 0.8412698412698413,
      "eval_recall": 0.2324561403508772,
      "eval_runtime": 11.7404,
      "eval_samples_per_second": 13.884,
      "eval_steps_per_second": 0.937,
      "step": 2100
    },
    {
      "epoch": 5.45,
      "learning_rate": 4.54780361757106e-05,
      "loss": 0.0888,
      "step": 2110
    },
    {
      "epoch": 5.48,
      "learning_rate": 4.521963824289406e-05,
      "loss": 0.0743,
      "step": 2120
    },
    {
      "epoch": 5.5,
      "learning_rate": 4.496124031007753e-05,
      "loss": 0.0678,
      "step": 2130
    },
    {
      "epoch": 5.53,
      "learning_rate": 4.4702842377260983e-05,
      "loss": 0.0737,
      "step": 2140
    },
    {
      "epoch": 5.56,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.0747,
      "step": 2150
    },
    {
      "epoch": 5.58,
      "learning_rate": 4.418604651162791e-05,
      "loss": 0.0735,
      "step": 2160
    },
    {
      "epoch": 5.61,
      "learning_rate": 4.392764857881137e-05,
      "loss": 0.075,
      "step": 2170
    },
    {
      "epoch": 5.63,
      "learning_rate": 4.3669250645994836e-05,
      "loss": 0.0726,
      "step": 2180
    },
    {
      "epoch": 5.66,
      "learning_rate": 4.34108527131783e-05,
      "loss": 0.0701,
      "step": 2190
    },
    {
      "epoch": 5.68,
      "learning_rate": 4.315245478036176e-05,
      "loss": 0.073,
      "step": 2200
    },
    {
      "epoch": 5.68,
      "eval_accuracy": 0.9769325153374233,
      "eval_f1": 0.38562091503267976,
      "eval_loss": 0.08116535097360611,
      "eval_precision": 0.7564102564102564,
      "eval_recall": 0.25877192982456143,
      "eval_runtime": 11.6668,
      "eval_samples_per_second": 13.971,
      "eval_steps_per_second": 0.943,
      "step": 2200
    },
    {
      "epoch": 5.71,
      "learning_rate": 4.289405684754522e-05,
      "loss": 0.0725,
      "step": 2210
    },
    {
      "epoch": 5.74,
      "learning_rate": 4.263565891472868e-05,
      "loss": 0.087,
      "step": 2220
    },
    {
      "epoch": 5.76,
      "learning_rate": 4.2377260981912145e-05,
      "loss": 0.0849,
      "step": 2230
    },
    {
      "epoch": 5.79,
      "learning_rate": 4.211886304909561e-05,
      "loss": 0.0696,
      "step": 2240
    },
    {
      "epoch": 5.81,
      "learning_rate": 4.186046511627907e-05,
      "loss": 0.0848,
      "step": 2250
    },
    {
      "epoch": 5.84,
      "learning_rate": 4.1602067183462534e-05,
      "loss": 0.0798,
      "step": 2260
    },
    {
      "epoch": 5.87,
      "learning_rate": 4.1343669250646e-05,
      "loss": 0.0751,
      "step": 2270
    },
    {
      "epoch": 5.89,
      "learning_rate": 4.108527131782946e-05,
      "loss": 0.0773,
      "step": 2280
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.082687338501292e-05,
      "loss": 0.0848,
      "step": 2290
    },
    {
      "epoch": 5.94,
      "learning_rate": 4.056847545219639e-05,
      "loss": 0.0801,
      "step": 2300
    },
    {
      "epoch": 5.94,
      "eval_accuracy": 0.9776687116564418,
      "eval_f1": 0.3591549295774648,
      "eval_loss": 0.07930972427129745,
      "eval_precision": 0.9107142857142857,
      "eval_recall": 0.2236842105263158,
      "eval_runtime": 11.5705,
      "eval_samples_per_second": 14.088,
      "eval_steps_per_second": 0.951,
      "step": 2300
    },
    {
      "epoch": 5.97,
      "learning_rate": 4.0310077519379843e-05,
      "loss": 0.0853,
      "step": 2310
    },
    {
      "epoch": 5.99,
      "learning_rate": 4.005167958656331e-05,
      "loss": 0.0789,
      "step": 2320
    },
    {
      "epoch": 6.02,
      "learning_rate": 3.979328165374677e-05,
      "loss": 0.0724,
      "step": 2330
    },
    {
      "epoch": 6.05,
      "learning_rate": 3.953488372093023e-05,
      "loss": 0.0754,
      "step": 2340
    },
    {
      "epoch": 6.07,
      "learning_rate": 3.9276485788113696e-05,
      "loss": 0.0769,
      "step": 2350
    },
    {
      "epoch": 6.1,
      "learning_rate": 3.901808785529716e-05,
      "loss": 0.0665,
      "step": 2360
    },
    {
      "epoch": 6.12,
      "learning_rate": 3.875968992248062e-05,
      "loss": 0.067,
      "step": 2370
    },
    {
      "epoch": 6.15,
      "learning_rate": 3.8501291989664085e-05,
      "loss": 0.0682,
      "step": 2380
    },
    {
      "epoch": 6.18,
      "learning_rate": 3.824289405684754e-05,
      "loss": 0.0741,
      "step": 2390
    },
    {
      "epoch": 6.2,
      "learning_rate": 3.798449612403101e-05,
      "loss": 0.077,
      "step": 2400
    },
    {
      "epoch": 6.2,
      "eval_accuracy": 0.9774233128834355,
      "eval_f1": 0.3825503355704698,
      "eval_loss": 0.07888315618038177,
      "eval_precision": 0.8142857142857143,
      "eval_recall": 0.25,
      "eval_runtime": 11.8196,
      "eval_samples_per_second": 13.791,
      "eval_steps_per_second": 0.931,
      "step": 2400
    },
    {
      "epoch": 6.23,
      "learning_rate": 3.772609819121447e-05,
      "loss": 0.0685,
      "step": 2410
    },
    {
      "epoch": 6.25,
      "learning_rate": 3.746770025839794e-05,
      "loss": 0.0752,
      "step": 2420
    },
    {
      "epoch": 6.28,
      "learning_rate": 3.7209302325581394e-05,
      "loss": 0.0704,
      "step": 2430
    },
    {
      "epoch": 6.3,
      "learning_rate": 3.6950904392764864e-05,
      "loss": 0.0755,
      "step": 2440
    },
    {
      "epoch": 6.33,
      "learning_rate": 3.669250645994832e-05,
      "loss": 0.0726,
      "step": 2450
    },
    {
      "epoch": 6.36,
      "learning_rate": 3.6434108527131784e-05,
      "loss": 0.0622,
      "step": 2460
    },
    {
      "epoch": 6.38,
      "learning_rate": 3.617571059431525e-05,
      "loss": 0.0742,
      "step": 2470
    },
    {
      "epoch": 6.41,
      "learning_rate": 3.591731266149871e-05,
      "loss": 0.0813,
      "step": 2480
    },
    {
      "epoch": 6.43,
      "learning_rate": 3.565891472868217e-05,
      "loss": 0.0724,
      "step": 2490
    },
    {
      "epoch": 6.46,
      "learning_rate": 3.5400516795865637e-05,
      "loss": 0.0749,
      "step": 2500
    },
    {
      "epoch": 6.46,
      "eval_accuracy": 0.9779141104294479,
      "eval_f1": 0.41558441558441556,
      "eval_loss": 0.077651746571064,
      "eval_precision": 0.8,
      "eval_recall": 0.2807017543859649,
      "eval_runtime": 12.0214,
      "eval_samples_per_second": 13.559,
      "eval_steps_per_second": 0.915,
      "step": 2500
    },
    {
      "epoch": 6.49,
      "learning_rate": 3.514211886304909e-05,
      "loss": 0.0667,
      "step": 2510
    },
    {
      "epoch": 6.51,
      "learning_rate": 3.488372093023256e-05,
      "loss": 0.0646,
      "step": 2520
    },
    {
      "epoch": 6.54,
      "learning_rate": 3.462532299741602e-05,
      "loss": 0.0744,
      "step": 2530
    },
    {
      "epoch": 6.56,
      "learning_rate": 3.436692506459949e-05,
      "loss": 0.0815,
      "step": 2540
    },
    {
      "epoch": 6.59,
      "learning_rate": 3.4108527131782945e-05,
      "loss": 0.0629,
      "step": 2550
    },
    {
      "epoch": 6.61,
      "learning_rate": 3.3850129198966415e-05,
      "loss": 0.0729,
      "step": 2560
    },
    {
      "epoch": 6.64,
      "learning_rate": 3.359173126614987e-05,
      "loss": 0.0874,
      "step": 2570
    },
    {
      "epoch": 6.67,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0725,
      "step": 2580
    },
    {
      "epoch": 6.69,
      "learning_rate": 3.30749354005168e-05,
      "loss": 0.0693,
      "step": 2590
    },
    {
      "epoch": 6.72,
      "learning_rate": 3.281653746770026e-05,
      "loss": 0.0704,
      "step": 2600
    },
    {
      "epoch": 6.72,
      "eval_accuracy": 0.9786503067484662,
      "eval_f1": 0.423841059602649,
      "eval_loss": 0.07763181626796722,
      "eval_precision": 0.8648648648648649,
      "eval_recall": 0.2807017543859649,
      "eval_runtime": 11.5898,
      "eval_samples_per_second": 14.064,
      "eval_steps_per_second": 0.949,
      "step": 2600
    },
    {
      "epoch": 6.74,
      "learning_rate": 3.2558139534883724e-05,
      "loss": 0.0812,
      "step": 2610
    },
    {
      "epoch": 6.77,
      "learning_rate": 3.229974160206719e-05,
      "loss": 0.0693,
      "step": 2620
    },
    {
      "epoch": 6.8,
      "learning_rate": 3.2041343669250644e-05,
      "loss": 0.0687,
      "step": 2630
    },
    {
      "epoch": 6.82,
      "learning_rate": 3.1782945736434114e-05,
      "loss": 0.0698,
      "step": 2640
    },
    {
      "epoch": 6.85,
      "learning_rate": 3.152454780361757e-05,
      "loss": 0.0688,
      "step": 2650
    },
    {
      "epoch": 6.87,
      "learning_rate": 3.126614987080104e-05,
      "loss": 0.0742,
      "step": 2660
    },
    {
      "epoch": 6.9,
      "learning_rate": 3.1007751937984497e-05,
      "loss": 0.0773,
      "step": 2670
    },
    {
      "epoch": 6.93,
      "learning_rate": 3.074935400516796e-05,
      "loss": 0.0714,
      "step": 2680
    },
    {
      "epoch": 6.95,
      "learning_rate": 3.0490956072351423e-05,
      "loss": 0.0678,
      "step": 2690
    },
    {
      "epoch": 6.98,
      "learning_rate": 3.0232558139534883e-05,
      "loss": 0.0718,
      "step": 2700
    },
    {
      "epoch": 6.98,
      "eval_accuracy": 0.9776687116564418,
      "eval_f1": 0.389261744966443,
      "eval_loss": 0.07732577621936798,
      "eval_precision": 0.8285714285714286,
      "eval_recall": 0.2543859649122807,
      "eval_runtime": 11.7112,
      "eval_samples_per_second": 13.918,
      "eval_steps_per_second": 0.939,
      "step": 2700
    },
    {
      "epoch": 7.0,
      "learning_rate": 2.997416020671835e-05,
      "loss": 0.0723,
      "step": 2710
    },
    {
      "epoch": 7.03,
      "learning_rate": 2.971576227390181e-05,
      "loss": 0.0669,
      "step": 2720
    },
    {
      "epoch": 7.05,
      "learning_rate": 2.9457364341085275e-05,
      "loss": 0.0705,
      "step": 2730
    },
    {
      "epoch": 7.08,
      "learning_rate": 2.9198966408268735e-05,
      "loss": 0.0697,
      "step": 2740
    },
    {
      "epoch": 7.11,
      "learning_rate": 2.8940568475452195e-05,
      "loss": 0.0723,
      "step": 2750
    },
    {
      "epoch": 7.13,
      "learning_rate": 2.868217054263566e-05,
      "loss": 0.0693,
      "step": 2760
    },
    {
      "epoch": 7.16,
      "learning_rate": 2.842377260981912e-05,
      "loss": 0.0653,
      "step": 2770
    },
    {
      "epoch": 7.18,
      "learning_rate": 2.8165374677002588e-05,
      "loss": 0.0651,
      "step": 2780
    },
    {
      "epoch": 7.21,
      "learning_rate": 2.7906976744186048e-05,
      "loss": 0.0714,
      "step": 2790
    },
    {
      "epoch": 7.24,
      "learning_rate": 2.7648578811369507e-05,
      "loss": 0.0624,
      "step": 2800
    },
    {
      "epoch": 7.24,
      "eval_accuracy": 0.978159509202454,
      "eval_f1": 0.3945578231292517,
      "eval_loss": 0.0775144025683403,
      "eval_precision": 0.8787878787878788,
      "eval_recall": 0.2543859649122807,
      "eval_runtime": 11.801,
      "eval_samples_per_second": 13.812,
      "eval_steps_per_second": 0.932,
      "step": 2800
    },
    {
      "epoch": 7.26,
      "learning_rate": 2.7390180878552974e-05,
      "loss": 0.0596,
      "step": 2810
    },
    {
      "epoch": 7.29,
      "learning_rate": 2.7131782945736434e-05,
      "loss": 0.0712,
      "step": 2820
    },
    {
      "epoch": 7.31,
      "learning_rate": 2.68733850129199e-05,
      "loss": 0.0605,
      "step": 2830
    },
    {
      "epoch": 7.34,
      "learning_rate": 2.661498708010336e-05,
      "loss": 0.0665,
      "step": 2840
    },
    {
      "epoch": 7.36,
      "learning_rate": 2.6356589147286826e-05,
      "loss": 0.0651,
      "step": 2850
    },
    {
      "epoch": 7.39,
      "learning_rate": 2.6098191214470286e-05,
      "loss": 0.0677,
      "step": 2860
    },
    {
      "epoch": 7.42,
      "learning_rate": 2.5839793281653746e-05,
      "loss": 0.0715,
      "step": 2870
    },
    {
      "epoch": 7.44,
      "learning_rate": 2.5581395348837212e-05,
      "loss": 0.0691,
      "step": 2880
    },
    {
      "epoch": 7.47,
      "learning_rate": 2.5322997416020672e-05,
      "loss": 0.0689,
      "step": 2890
    },
    {
      "epoch": 7.49,
      "learning_rate": 2.506459948320414e-05,
      "loss": 0.0718,
      "step": 2900
    },
    {
      "epoch": 7.49,
      "eval_accuracy": 0.9780368098159509,
      "eval_f1": 0.4013377926421405,
      "eval_loss": 0.07592039555311203,
      "eval_precision": 0.8450704225352113,
      "eval_recall": 0.2631578947368421,
      "eval_runtime": 11.7263,
      "eval_samples_per_second": 13.9,
      "eval_steps_per_second": 0.938,
      "step": 2900
    },
    {
      "epoch": 7.52,
      "learning_rate": 2.48062015503876e-05,
      "loss": 0.0668,
      "step": 2910
    },
    {
      "epoch": 7.55,
      "learning_rate": 2.4547803617571062e-05,
      "loss": 0.0664,
      "step": 2920
    },
    {
      "epoch": 7.57,
      "learning_rate": 2.4289405684754525e-05,
      "loss": 0.0645,
      "step": 2930
    },
    {
      "epoch": 7.6,
      "learning_rate": 2.4031007751937988e-05,
      "loss": 0.0763,
      "step": 2940
    },
    {
      "epoch": 7.62,
      "learning_rate": 2.3772609819121448e-05,
      "loss": 0.0677,
      "step": 2950
    },
    {
      "epoch": 7.65,
      "learning_rate": 2.351421188630491e-05,
      "loss": 0.0679,
      "step": 2960
    },
    {
      "epoch": 7.67,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 0.0722,
      "step": 2970
    },
    {
      "epoch": 7.7,
      "learning_rate": 2.2997416020671837e-05,
      "loss": 0.0655,
      "step": 2980
    },
    {
      "epoch": 7.73,
      "learning_rate": 2.27390180878553e-05,
      "loss": 0.0702,
      "step": 2990
    },
    {
      "epoch": 7.75,
      "learning_rate": 2.2480620155038764e-05,
      "loss": 0.0671,
      "step": 3000
    },
    {
      "epoch": 7.75,
      "eval_accuracy": 0.9787730061349693,
      "eval_f1": 0.4437299035369775,
      "eval_loss": 0.07497038692235947,
      "eval_precision": 0.8313253012048193,
      "eval_recall": 0.3026315789473684,
      "eval_runtime": 11.8994,
      "eval_samples_per_second": 13.698,
      "eval_steps_per_second": 0.924,
      "step": 3000
    },
    {
      "epoch": 7.78,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.0681,
      "step": 3010
    },
    {
      "epoch": 7.8,
      "learning_rate": 2.1963824289405686e-05,
      "loss": 0.0687,
      "step": 3020
    },
    {
      "epoch": 7.83,
      "learning_rate": 2.170542635658915e-05,
      "loss": 0.07,
      "step": 3030
    },
    {
      "epoch": 7.86,
      "learning_rate": 2.144702842377261e-05,
      "loss": 0.0702,
      "step": 3040
    },
    {
      "epoch": 7.88,
      "learning_rate": 2.1188630490956073e-05,
      "loss": 0.0717,
      "step": 3050
    },
    {
      "epoch": 7.91,
      "learning_rate": 2.0930232558139536e-05,
      "loss": 0.0646,
      "step": 3060
    },
    {
      "epoch": 7.93,
      "learning_rate": 2.0671834625323e-05,
      "loss": 0.0706,
      "step": 3070
    },
    {
      "epoch": 7.96,
      "learning_rate": 2.041343669250646e-05,
      "loss": 0.0649,
      "step": 3080
    },
    {
      "epoch": 7.98,
      "learning_rate": 2.0155038759689922e-05,
      "loss": 0.0663,
      "step": 3090
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.9896640826873385e-05,
      "loss": 0.0554,
      "step": 3100
    },
    {
      "epoch": 8.01,
      "eval_accuracy": 0.9790184049079754,
      "eval_f1": 0.45714285714285713,
      "eval_loss": 0.07580507546663284,
      "eval_precision": 0.8275862068965517,
      "eval_recall": 0.3157894736842105,
      "eval_runtime": 11.6291,
      "eval_samples_per_second": 14.017,
      "eval_steps_per_second": 0.946,
      "step": 3100
    },
    {
      "epoch": 8.04,
      "learning_rate": 1.9638242894056848e-05,
      "loss": 0.0562,
      "step": 3110
    },
    {
      "epoch": 8.06,
      "learning_rate": 1.937984496124031e-05,
      "loss": 0.061,
      "step": 3120
    },
    {
      "epoch": 8.09,
      "learning_rate": 1.912144702842377e-05,
      "loss": 0.0642,
      "step": 3130
    },
    {
      "epoch": 8.11,
      "learning_rate": 1.8863049095607234e-05,
      "loss": 0.0581,
      "step": 3140
    },
    {
      "epoch": 8.14,
      "learning_rate": 1.8604651162790697e-05,
      "loss": 0.0611,
      "step": 3150
    },
    {
      "epoch": 8.17,
      "learning_rate": 1.834625322997416e-05,
      "loss": 0.0683,
      "step": 3160
    },
    {
      "epoch": 8.19,
      "learning_rate": 1.8087855297157624e-05,
      "loss": 0.0635,
      "step": 3170
    },
    {
      "epoch": 8.22,
      "learning_rate": 1.7829457364341087e-05,
      "loss": 0.0617,
      "step": 3180
    },
    {
      "epoch": 8.24,
      "learning_rate": 1.7571059431524546e-05,
      "loss": 0.0656,
      "step": 3190
    },
    {
      "epoch": 8.27,
      "learning_rate": 1.731266149870801e-05,
      "loss": 0.0533,
      "step": 3200
    },
    {
      "epoch": 8.27,
      "eval_accuracy": 0.9787730061349693,
      "eval_f1": 0.4507936507936508,
      "eval_loss": 0.07549890875816345,
      "eval_precision": 0.8160919540229885,
      "eval_recall": 0.31140350877192985,
      "eval_runtime": 11.6344,
      "eval_samples_per_second": 14.01,
      "eval_steps_per_second": 0.945,
      "step": 3200
    },
    {
      "epoch": 8.29,
      "learning_rate": 1.7054263565891473e-05,
      "loss": 0.061,
      "step": 3210
    },
    {
      "epoch": 8.32,
      "learning_rate": 1.6795865633074936e-05,
      "loss": 0.0727,
      "step": 3220
    },
    {
      "epoch": 8.35,
      "learning_rate": 1.65374677002584e-05,
      "loss": 0.057,
      "step": 3230
    },
    {
      "epoch": 8.37,
      "learning_rate": 1.6279069767441862e-05,
      "loss": 0.0692,
      "step": 3240
    },
    {
      "epoch": 8.4,
      "learning_rate": 1.6020671834625322e-05,
      "loss": 0.0709,
      "step": 3250
    },
    {
      "epoch": 8.42,
      "learning_rate": 1.5762273901808785e-05,
      "loss": 0.055,
      "step": 3260
    },
    {
      "epoch": 8.45,
      "learning_rate": 1.5503875968992248e-05,
      "loss": 0.0626,
      "step": 3270
    },
    {
      "epoch": 8.48,
      "learning_rate": 1.5245478036175711e-05,
      "loss": 0.0696,
      "step": 3280
    },
    {
      "epoch": 8.5,
      "learning_rate": 1.4987080103359175e-05,
      "loss": 0.063,
      "step": 3290
    },
    {
      "epoch": 8.53,
      "learning_rate": 1.4728682170542638e-05,
      "loss": 0.063,
      "step": 3300
    },
    {
      "epoch": 8.53,
      "eval_accuracy": 0.9790184049079754,
      "eval_f1": 0.4605678233438486,
      "eval_loss": 0.07512389868497849,
      "eval_precision": 0.8202247191011236,
      "eval_recall": 0.3201754385964912,
      "eval_runtime": 11.9167,
      "eval_samples_per_second": 13.678,
      "eval_steps_per_second": 0.923,
      "step": 3300
    },
    {
      "epoch": 8.55,
      "learning_rate": 1.4470284237726097e-05,
      "loss": 0.0664,
      "step": 3310
    },
    {
      "epoch": 8.58,
      "learning_rate": 1.421188630490956e-05,
      "loss": 0.0717,
      "step": 3320
    },
    {
      "epoch": 8.6,
      "learning_rate": 1.3953488372093024e-05,
      "loss": 0.0632,
      "step": 3330
    },
    {
      "epoch": 8.63,
      "learning_rate": 1.3695090439276487e-05,
      "loss": 0.0645,
      "step": 3340
    },
    {
      "epoch": 8.66,
      "learning_rate": 1.343669250645995e-05,
      "loss": 0.062,
      "step": 3350
    },
    {
      "epoch": 8.68,
      "learning_rate": 1.3178294573643413e-05,
      "loss": 0.0628,
      "step": 3360
    },
    {
      "epoch": 8.71,
      "learning_rate": 1.2919896640826873e-05,
      "loss": 0.0647,
      "step": 3370
    },
    {
      "epoch": 8.73,
      "learning_rate": 1.2661498708010336e-05,
      "loss": 0.0675,
      "step": 3380
    },
    {
      "epoch": 8.76,
      "learning_rate": 1.24031007751938e-05,
      "loss": 0.0563,
      "step": 3390
    },
    {
      "epoch": 8.79,
      "learning_rate": 1.2144702842377262e-05,
      "loss": 0.0605,
      "step": 3400
    },
    {
      "epoch": 8.79,
      "eval_accuracy": 0.9790184049079754,
      "eval_f1": 0.4605678233438486,
      "eval_loss": 0.07484734058380127,
      "eval_precision": 0.8202247191011236,
      "eval_recall": 0.3201754385964912,
      "eval_runtime": 12.0314,
      "eval_samples_per_second": 13.548,
      "eval_steps_per_second": 0.914,
      "step": 3400
    },
    {
      "epoch": 8.81,
      "learning_rate": 1.1886304909560724e-05,
      "loss": 0.0623,
      "step": 3410
    },
    {
      "epoch": 8.84,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 0.0666,
      "step": 3420
    },
    {
      "epoch": 8.86,
      "learning_rate": 1.136950904392765e-05,
      "loss": 0.0601,
      "step": 3430
    },
    {
      "epoch": 8.89,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0758,
      "step": 3440
    },
    {
      "epoch": 8.91,
      "learning_rate": 1.0852713178294575e-05,
      "loss": 0.0687,
      "step": 3450
    },
    {
      "epoch": 8.94,
      "learning_rate": 1.0594315245478036e-05,
      "loss": 0.0569,
      "step": 3460
    },
    {
      "epoch": 8.97,
      "learning_rate": 1.03359173126615e-05,
      "loss": 0.0714,
      "step": 3470
    },
    {
      "epoch": 8.99,
      "learning_rate": 1.0077519379844961e-05,
      "loss": 0.0742,
      "step": 3480
    },
    {
      "epoch": 9.02,
      "learning_rate": 9.819121447028424e-06,
      "loss": 0.0665,
      "step": 3490
    },
    {
      "epoch": 9.04,
      "learning_rate": 9.560723514211885e-06,
      "loss": 0.0592,
      "step": 3500
    },
    {
      "epoch": 9.04,
      "eval_accuracy": 0.9787730061349693,
      "eval_f1": 0.4327868852459017,
      "eval_loss": 0.07429111003875732,
      "eval_precision": 0.8571428571428571,
      "eval_recall": 0.2894736842105263,
      "eval_runtime": 11.4554,
      "eval_samples_per_second": 14.229,
      "eval_steps_per_second": 0.96,
      "step": 3500
    }
  ],
  "max_steps": 3870,
  "num_train_epochs": 10,
  "total_flos": 1.1850942195223757e+17,
  "trial_name": null,
  "trial_params": null
}