{
  "best_metric": 0.9023026110381105,
  "best_model_checkpoint": "../save/jtrans-malware-4f-100c-1/checkpoint-4000",
  "epoch": 9.029345372460497,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 9.97742663656885e-05,
      "loss": 0.4763,
      "step": 10
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.954853273137697e-05,
      "loss": 0.2498,
      "step": 20
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.932279909706546e-05,
      "loss": 0.1516,
      "step": 30
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.909706546275395e-05,
      "loss": 0.109,
      "step": 40
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.887133182844243e-05,
      "loss": 0.0969,
      "step": 50
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.864559819413092e-05,
      "loss": 0.0897,
      "step": 60
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.841986455981941e-05,
      "loss": 0.0877,
      "step": 70
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.81941309255079e-05,
      "loss": 0.0867,
      "step": 80
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.79683972911964e-05,
      "loss": 0.0832,
      "step": 90
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.774266365688489e-05,
      "loss": 0.0876,
      "step": 100
    },
    {
      "epoch": 0.23,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.08578885346651077,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.6301920123867697,
      "eval_runtime": 3.5622,
      "eval_samples_per_second": 52.215,
      "eval_steps_per_second": 3.369,
      "step": 100
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.751693002257338e-05,
      "loss": 0.0788,
      "step": 110
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.729119638826185e-05,
      "loss": 0.0853,
      "step": 120
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.706546275395035e-05,
      "loss": 0.0888,
      "step": 130
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.683972911963884e-05,
      "loss": 0.0836,
      "step": 140
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.661399548532731e-05,
      "loss": 0.0863,
      "step": 150
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.63882618510158e-05,
      "loss": 0.0798,
      "step": 160
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.61625282167043e-05,
      "loss": 0.0823,
      "step": 170
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.593679458239279e-05,
      "loss": 0.0821,
      "step": 180
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.571106094808126e-05,
      "loss": 0.0807,
      "step": 190
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.548532731376975e-05,
      "loss": 0.0842,
      "step": 200
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.08438392728567123,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.6431196288803497,
      "eval_runtime": 3.5805,
      "eval_samples_per_second": 51.949,
      "eval_steps_per_second": 3.352,
      "step": 200
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.525959367945825e-05,
      "loss": 0.0795,
      "step": 210
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.503386004514672e-05,
      "loss": 0.0832,
      "step": 220
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.480812641083521e-05,
      "loss": 0.0905,
      "step": 230
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.45823927765237e-05,
      "loss": 0.0808,
      "step": 240
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.43566591422122e-05,
      "loss": 0.085,
      "step": 250
    },
    {
      "epoch": 0.59,
      "learning_rate": 9.413092550790069e-05,
      "loss": 0.0808,
      "step": 260
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.390519187358918e-05,
      "loss": 0.0803,
      "step": 270
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.367945823927767e-05,
      "loss": 0.0832,
      "step": 280
    },
    {
      "epoch": 0.65,
      "learning_rate": 9.345372460496615e-05,
      "loss": 0.0814,
      "step": 290
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.322799097065464e-05,
      "loss": 0.0766,
      "step": 300
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.08525288850069046,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.6508643813900336,
      "eval_runtime": 3.6819,
      "eval_samples_per_second": 50.517,
      "eval_steps_per_second": 3.259,
      "step": 300
    },
    {
      "epoch": 0.7,
      "learning_rate": 9.300225733634313e-05,
      "loss": 0.089,
      "step": 310
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.27765237020316e-05,
      "loss": 0.0885,
      "step": 320
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.25507900677201e-05,
      "loss": 0.0755,
      "step": 330
    },
    {
      "epoch": 0.77,
      "learning_rate": 9.232505643340859e-05,
      "loss": 0.0782,
      "step": 340
    },
    {
      "epoch": 0.79,
      "learning_rate": 9.209932279909706e-05,
      "loss": 0.0856,
      "step": 350
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.187358916478555e-05,
      "loss": 0.0844,
      "step": 360
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.164785553047405e-05,
      "loss": 0.0783,
      "step": 370
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.142212189616254e-05,
      "loss": 0.0832,
      "step": 380
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.119638826185101e-05,
      "loss": 0.0826,
      "step": 390
    },
    {
      "epoch": 0.9,
      "learning_rate": 9.09706546275395e-05,
      "loss": 0.0843,
      "step": 400
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.08313789963722229,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.6533535023443411,
      "eval_runtime": 3.5877,
      "eval_samples_per_second": 51.844,
      "eval_steps_per_second": 3.345,
      "step": 400
    },
    {
      "epoch": 0.93,
      "learning_rate": 9.0744920993228e-05,
      "loss": 0.0804,
      "step": 410
    },
    {
      "epoch": 0.95,
      "learning_rate": 9.051918735891649e-05,
      "loss": 0.0797,
      "step": 420
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.029345372460498e-05,
      "loss": 0.0843,
      "step": 430
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.006772009029347e-05,
      "loss": 0.0804,
      "step": 440
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.984198645598195e-05,
      "loss": 0.0846,
      "step": 450
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.961625282167044e-05,
      "loss": 0.0846,
      "step": 460
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.939051918735893e-05,
      "loss": 0.0775,
      "step": 470
    },
    {
      "epoch": 1.08,
      "learning_rate": 8.91647855530474e-05,
      "loss": 0.083,
      "step": 480
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.89390519187359e-05,
      "loss": 0.082,
      "step": 490
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.871331828442439e-05,
      "loss": 0.0843,
      "step": 500
    },
    {
      "epoch": 1.13,
      "eval_accuracy": 0.9831182795698925,
      "eval_f1": 0.07647058823529411,
      "eval_loss": 0.08325152099132538,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.03987730061349693,
      "eval_roc_auc_score": 0.6600380472843176,
      "eval_runtime": 3.5838,
      "eval_samples_per_second": 51.9,
      "eval_steps_per_second": 3.348,
      "step": 500
    },
    {
      "epoch": 1.15,
      "learning_rate": 8.848758465011288e-05,
      "loss": 0.0819,
      "step": 510
    },
    {
      "epoch": 1.17,
      "learning_rate": 8.826185101580135e-05,
      "loss": 0.075,
      "step": 520
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.803611738148985e-05,
      "loss": 0.0815,
      "step": 530
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.781038374717834e-05,
      "loss": 0.0803,
      "step": 540
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.758465011286681e-05,
      "loss": 0.0821,
      "step": 550
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.73589164785553e-05,
      "loss": 0.084,
      "step": 560
    },
    {
      "epoch": 1.29,
      "learning_rate": 8.71331828442438e-05,
      "loss": 0.0862,
      "step": 570
    },
    {
      "epoch": 1.31,
      "learning_rate": 8.690744920993227e-05,
      "loss": 0.0808,
      "step": 580
    },
    {
      "epoch": 1.33,
      "learning_rate": 8.668171557562076e-05,
      "loss": 0.0772,
      "step": 590
    },
    {
      "epoch": 1.35,
      "learning_rate": 8.645598194130925e-05,
      "loss": 0.0809,
      "step": 600
    },
    {
      "epoch": 1.35,
      "eval_accuracy": 0.9831182795698925,
      "eval_f1": 0.07647058823529411,
      "eval_loss": 0.08176498860120773,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.03987730061349693,
      "eval_roc_auc_score": 0.6891173788768246,
      "eval_runtime": 3.719,
      "eval_samples_per_second": 50.014,
      "eval_steps_per_second": 3.227,
      "step": 600
    },
    {
      "epoch": 1.38,
      "learning_rate": 8.623024830699775e-05,
      "loss": 0.0802,
      "step": 610
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.600451467268624e-05,
      "loss": 0.0766,
      "step": 620
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.577878103837473e-05,
      "loss": 0.0789,
      "step": 630
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.555304740406322e-05,
      "loss": 0.0821,
      "step": 640
    },
    {
      "epoch": 1.47,
      "learning_rate": 8.53273137697517e-05,
      "loss": 0.0808,
      "step": 650
    },
    {
      "epoch": 1.49,
      "learning_rate": 8.510158013544019e-05,
      "loss": 0.0818,
      "step": 660
    },
    {
      "epoch": 1.51,
      "learning_rate": 8.487584650112868e-05,
      "loss": 0.0828,
      "step": 670
    },
    {
      "epoch": 1.53,
      "learning_rate": 8.465011286681715e-05,
      "loss": 0.0797,
      "step": 680
    },
    {
      "epoch": 1.56,
      "learning_rate": 8.442437923250564e-05,
      "loss": 0.0862,
      "step": 690
    },
    {
      "epoch": 1.58,
      "learning_rate": 8.419864559819414e-05,
      "loss": 0.0781,
      "step": 700
    },
    {
      "epoch": 1.58,
      "eval_accuracy": 0.9831182795698925,
      "eval_f1": 0.07647058823529411,
      "eval_loss": 0.08141683787107468,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.03987730061349693,
      "eval_roc_auc_score": 0.6985658325785202,
      "eval_runtime": 3.5869,
      "eval_samples_per_second": 51.856,
      "eval_steps_per_second": 3.346,
      "step": 700
    },
    {
      "epoch": 1.6,
      "learning_rate": 8.397291196388263e-05,
      "loss": 0.0792,
      "step": 710
    },
    {
      "epoch": 1.63,
      "learning_rate": 8.37471783295711e-05,
      "loss": 0.0826,
      "step": 720
    },
    {
      "epoch": 1.65,
      "learning_rate": 8.35214446952596e-05,
      "loss": 0.0799,
      "step": 730
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.329571106094809e-05,
      "loss": 0.0803,
      "step": 740
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.306997742663656e-05,
      "loss": 0.0997,
      "step": 750
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.284424379232505e-05,
      "loss": 0.0766,
      "step": 760
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.261851015801354e-05,
      "loss": 0.0823,
      "step": 770
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.239277652370204e-05,
      "loss": 0.0759,
      "step": 780
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.216704288939053e-05,
      "loss": 0.0833,
      "step": 790
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.194130925507902e-05,
      "loss": 0.0808,
      "step": 800
    },
    {
      "epoch": 1.81,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.08207544684410095,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.6866500798009307,
      "eval_runtime": 3.5737,
      "eval_samples_per_second": 52.046,
      "eval_steps_per_second": 3.358,
      "step": 800
    },
    {
      "epoch": 1.83,
      "learning_rate": 8.17155756207675e-05,
      "loss": 0.0737,
      "step": 810
    },
    {
      "epoch": 1.85,
      "learning_rate": 8.148984198645599e-05,
      "loss": 0.0816,
      "step": 820
    },
    {
      "epoch": 1.87,
      "learning_rate": 8.126410835214448e-05,
      "loss": 0.0822,
      "step": 830
    },
    {
      "epoch": 1.9,
      "learning_rate": 8.103837471783297e-05,
      "loss": 0.0819,
      "step": 840
    },
    {
      "epoch": 1.92,
      "learning_rate": 8.081264108352144e-05,
      "loss": 0.0769,
      "step": 850
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.058690744920994e-05,
      "loss": 0.0838,
      "step": 860
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.036117381489843e-05,
      "loss": 0.0836,
      "step": 870
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.01354401805869e-05,
      "loss": 0.0838,
      "step": 880
    },
    {
      "epoch": 2.01,
      "learning_rate": 7.99097065462754e-05,
      "loss": 0.0802,
      "step": 890
    },
    {
      "epoch": 2.03,
      "learning_rate": 7.968397291196389e-05,
      "loss": 0.0857,
      "step": 900
    },
    {
      "epoch": 2.03,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.08378244936466217,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.6764166259884472,
      "eval_runtime": 3.7493,
      "eval_samples_per_second": 49.609,
      "eval_steps_per_second": 3.201,
      "step": 900
    },
    {
      "epoch": 2.05,
      "learning_rate": 7.945823927765236e-05,
      "loss": 0.0789,
      "step": 910
    },
    {
      "epoch": 2.08,
      "learning_rate": 7.923250564334085e-05,
      "loss": 0.0842,
      "step": 920
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.900677200902934e-05,
      "loss": 0.0825,
      "step": 930
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.878103837471784e-05,
      "loss": 0.0782,
      "step": 940
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.855530474040633e-05,
      "loss": 0.0811,
      "step": 950
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.832957110609482e-05,
      "loss": 0.0802,
      "step": 960
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.810383747178331e-05,
      "loss": 0.0826,
      "step": 970
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.787810383747179e-05,
      "loss": 0.0874,
      "step": 980
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.765237020316028e-05,
      "loss": 0.0843,
      "step": 990
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.742663656884877e-05,
      "loss": 0.0784,
      "step": 1000
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.08410345017910004,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.6778744113967949,
      "eval_runtime": 3.5461,
      "eval_samples_per_second": 52.452,
      "eval_steps_per_second": 3.384,
      "step": 1000
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.720090293453724e-05,
      "loss": 0.0848,
      "step": 1010
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.697516930022574e-05,
      "loss": 0.0801,
      "step": 1020
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.674943566591423e-05,
      "loss": 0.0768,
      "step": 1030
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.652370203160272e-05,
      "loss": 0.0719,
      "step": 1040
    },
    {
      "epoch": 2.37,
      "learning_rate": 7.62979683972912e-05,
      "loss": 0.0802,
      "step": 1050
    },
    {
      "epoch": 2.39,
      "learning_rate": 7.607223476297969e-05,
      "loss": 0.0747,
      "step": 1060
    },
    {
      "epoch": 2.42,
      "learning_rate": 7.584650112866818e-05,
      "loss": 0.0796,
      "step": 1070
    },
    {
      "epoch": 2.44,
      "learning_rate": 7.562076749435665e-05,
      "loss": 0.0766,
      "step": 1080
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.539503386004514e-05,
      "loss": 0.0785,
      "step": 1090
    },
    {
      "epoch": 2.48,
      "learning_rate": 7.516930022573364e-05,
      "loss": 0.0857,
      "step": 1100
    },
    {
      "epoch": 2.48,
      "eval_accuracy": 0.9823655913978495,
      "eval_f1": 0.06285714285714286,
      "eval_loss": 0.07912848889827728,
      "eval_precision": 0.4583333333333333,
      "eval_recall": 0.03374233128834356,
      "eval_roc_auc_score": 0.7252293983003106,
      "eval_runtime": 3.5575,
      "eval_samples_per_second": 52.284,
      "eval_steps_per_second": 3.373,
      "step": 1100
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.494356659142213e-05,
      "loss": 0.0797,
      "step": 1110
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.471783295711062e-05,
      "loss": 0.0749,
      "step": 1120
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.449209932279911e-05,
      "loss": 0.0756,
      "step": 1130
    },
    {
      "epoch": 2.57,
      "learning_rate": 7.42663656884876e-05,
      "loss": 0.0775,
      "step": 1140
    },
    {
      "epoch": 2.6,
      "learning_rate": 7.404063205417608e-05,
      "loss": 0.0771,
      "step": 1150
    },
    {
      "epoch": 2.62,
      "learning_rate": 7.381489841986457e-05,
      "loss": 0.0798,
      "step": 1160
    },
    {
      "epoch": 2.64,
      "learning_rate": 7.358916478555306e-05,
      "loss": 0.0798,
      "step": 1170
    },
    {
      "epoch": 2.66,
      "learning_rate": 7.336343115124154e-05,
      "loss": 0.0785,
      "step": 1180
    },
    {
      "epoch": 2.69,
      "learning_rate": 7.313769751693003e-05,
      "loss": 0.0703,
      "step": 1190
    },
    {
      "epoch": 2.71,
      "learning_rate": 7.291196388261852e-05,
      "loss": 0.0733,
      "step": 1200
    },
    {
      "epoch": 2.71,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.07782133668661118,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.7335851634055828,
      "eval_runtime": 3.5494,
      "eval_samples_per_second": 52.403,
      "eval_steps_per_second": 3.381,
      "step": 1200
    },
    {
      "epoch": 2.73,
      "learning_rate": 7.2686230248307e-05,
      "loss": 0.0729,
      "step": 1210
    },
    {
      "epoch": 2.75,
      "learning_rate": 7.246049661399549e-05,
      "loss": 0.0815,
      "step": 1220
    },
    {
      "epoch": 2.78,
      "learning_rate": 7.223476297968398e-05,
      "loss": 0.0781,
      "step": 1230
    },
    {
      "epoch": 2.8,
      "learning_rate": 7.200902934537245e-05,
      "loss": 0.0767,
      "step": 1240
    },
    {
      "epoch": 2.82,
      "learning_rate": 7.178329571106094e-05,
      "loss": 0.0748,
      "step": 1250
    },
    {
      "epoch": 2.84,
      "learning_rate": 7.155756207674944e-05,
      "loss": 0.0759,
      "step": 1260
    },
    {
      "epoch": 2.87,
      "learning_rate": 7.133182844243793e-05,
      "loss": 0.0765,
      "step": 1270
    },
    {
      "epoch": 2.89,
      "learning_rate": 7.110609480812642e-05,
      "loss": 0.0805,
      "step": 1280
    },
    {
      "epoch": 2.91,
      "learning_rate": 7.088036117381491e-05,
      "loss": 0.0785,
      "step": 1290
    },
    {
      "epoch": 2.93,
      "learning_rate": 7.06546275395034e-05,
      "loss": 0.0692,
      "step": 1300
    },
    {
      "epoch": 2.93,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.0759485736489296,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.7689336688754884,
      "eval_runtime": 3.5936,
      "eval_samples_per_second": 51.759,
      "eval_steps_per_second": 3.339,
      "step": 1300
    },
    {
      "epoch": 2.96,
      "learning_rate": 7.042889390519188e-05,
      "loss": 0.0747,
      "step": 1310
    },
    {
      "epoch": 2.98,
      "learning_rate": 7.020316027088037e-05,
      "loss": 0.0775,
      "step": 1320
    },
    {
      "epoch": 3.0,
      "learning_rate": 6.997742663656886e-05,
      "loss": 0.0704,
      "step": 1330
    },
    {
      "epoch": 3.02,
      "learning_rate": 6.975169300225734e-05,
      "loss": 0.0755,
      "step": 1340
    },
    {
      "epoch": 3.05,
      "learning_rate": 6.952595936794583e-05,
      "loss": 0.0813,
      "step": 1350
    },
    {
      "epoch": 3.07,
      "learning_rate": 6.930022573363432e-05,
      "loss": 0.0774,
      "step": 1360
    },
    {
      "epoch": 3.09,
      "learning_rate": 6.907449209932281e-05,
      "loss": 0.075,
      "step": 1370
    },
    {
      "epoch": 3.12,
      "learning_rate": 6.884875846501129e-05,
      "loss": 0.0764,
      "step": 1380
    },
    {
      "epoch": 3.14,
      "learning_rate": 6.862302483069978e-05,
      "loss": 0.0715,
      "step": 1390
    },
    {
      "epoch": 3.16,
      "learning_rate": 6.839729119638827e-05,
      "loss": 0.0748,
      "step": 1400
    },
    {
      "epoch": 3.16,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.0,
      "eval_loss": 0.07465961575508118,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc_score": 0.7781728843353157,
      "eval_runtime": 3.6973,
      "eval_samples_per_second": 50.306,
      "eval_steps_per_second": 3.246,
      "step": 1400
    },
    {
      "epoch": 3.18,
      "learning_rate": 6.817155756207674e-05,
      "loss": 0.0746,
      "step": 1410
    },
    {
      "epoch": 3.21,
      "learning_rate": 6.794582392776524e-05,
      "loss": 0.0704,
      "step": 1420
    },
    {
      "epoch": 3.23,
      "learning_rate": 6.772009029345373e-05,
      "loss": 0.0748,
      "step": 1430
    },
    {
      "epoch": 3.25,
      "learning_rate": 6.749435665914222e-05,
      "loss": 0.0685,
      "step": 1440
    },
    {
      "epoch": 3.27,
      "learning_rate": 6.726862302483071e-05,
      "loss": 0.0696,
      "step": 1450
    },
    {
      "epoch": 3.3,
      "learning_rate": 6.70428893905192e-05,
      "loss": 0.0711,
      "step": 1460
    },
    {
      "epoch": 3.32,
      "learning_rate": 6.681715575620769e-05,
      "loss": 0.0726,
      "step": 1470
    },
    {
      "epoch": 3.34,
      "learning_rate": 6.659142212189617e-05,
      "loss": 0.078,
      "step": 1480
    },
    {
      "epoch": 3.36,
      "learning_rate": 6.636568848758466e-05,
      "loss": 0.0719,
      "step": 1490
    },
    {
      "epoch": 3.39,
      "learning_rate": 6.613995485327315e-05,
      "loss": 0.0724,
      "step": 1500
    },
    {
      "epoch": 3.39,
      "eval_accuracy": 0.9824731182795698,
      "eval_f1": 0.07386363636363637,
      "eval_loss": 0.074778251349926,
      "eval_precision": 0.5,
      "eval_recall": 0.03987730061349693,
      "eval_roc_auc_score": 0.7822150515902778,
      "eval_runtime": 3.5546,
      "eval_samples_per_second": 52.326,
      "eval_steps_per_second": 3.376,
      "step": 1500
    },
    {
      "epoch": 3.41,
      "learning_rate": 6.591422121896163e-05,
      "loss": 0.0741,
      "step": 1510
    },
    {
      "epoch": 3.43,
      "learning_rate": 6.568848758465012e-05,
      "loss": 0.0684,
      "step": 1520
    },
    {
      "epoch": 3.45,
      "learning_rate": 6.546275395033861e-05,
      "loss": 0.0699,
      "step": 1530
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.523702031602708e-05,
      "loss": 0.0711,
      "step": 1540
    },
    {
      "epoch": 3.5,
      "learning_rate": 6.501128668171558e-05,
      "loss": 0.0787,
      "step": 1550
    },
    {
      "epoch": 3.52,
      "learning_rate": 6.478555304740407e-05,
      "loss": 0.0707,
      "step": 1560
    },
    {
      "epoch": 3.54,
      "learning_rate": 6.455981941309256e-05,
      "loss": 0.079,
      "step": 1570
    },
    {
      "epoch": 3.57,
      "learning_rate": 6.433408577878103e-05,
      "loss": 0.0688,
      "step": 1580
    },
    {
      "epoch": 3.59,
      "learning_rate": 6.410835214446953e-05,
      "loss": 0.0696,
      "step": 1590
    },
    {
      "epoch": 3.61,
      "learning_rate": 6.388261851015802e-05,
      "loss": 0.0708,
      "step": 1600
    },
    {
      "epoch": 3.61,
      "eval_accuracy": 0.9826344086021506,
      "eval_f1": 0.02416918429003021,
      "eval_loss": 0.07308202981948853,
      "eval_precision": 0.8,
      "eval_recall": 0.012269938650306749,
      "eval_roc_auc_score": 0.800050408539136,
      "eval_runtime": 3.5742,
      "eval_samples_per_second": 52.04,
      "eval_steps_per_second": 3.357,
      "step": 1600
    },
    {
      "epoch": 3.63,
      "learning_rate": 6.365688487584651e-05,
      "loss": 0.0718,
      "step": 1610
    },
    {
      "epoch": 3.66,
      "learning_rate": 6.3431151241535e-05,
      "loss": 0.067,
      "step": 1620
    },
    {
      "epoch": 3.68,
      "learning_rate": 6.320541760722349e-05,
      "loss": 0.0723,
      "step": 1630
    },
    {
      "epoch": 3.7,
      "learning_rate": 6.297968397291197e-05,
      "loss": 0.0682,
      "step": 1640
    },
    {
      "epoch": 3.72,
      "learning_rate": 6.275395033860046e-05,
      "loss": 0.072,
      "step": 1650
    },
    {
      "epoch": 3.75,
      "learning_rate": 6.252821670428895e-05,
      "loss": 0.0716,
      "step": 1660
    },
    {
      "epoch": 3.77,
      "learning_rate": 6.230248306997743e-05,
      "loss": 0.0727,
      "step": 1670
    },
    {
      "epoch": 3.79,
      "learning_rate": 6.207674943566592e-05,
      "loss": 0.0653,
      "step": 1680
    },
    {
      "epoch": 3.81,
      "learning_rate": 6.185101580135441e-05,
      "loss": 0.0768,
      "step": 1690
    },
    {
      "epoch": 3.84,
      "learning_rate": 6.16252821670429e-05,
      "loss": 0.0691,
      "step": 1700
    },
    {
      "epoch": 3.84,
      "eval_accuracy": 0.9826344086021506,
      "eval_f1": 0.02416918429003021,
      "eval_loss": 0.07123066484928131,
      "eval_precision": 0.8,
      "eval_recall": 0.012269938650306749,
      "eval_roc_auc_score": 0.8146110569107875,
      "eval_runtime": 3.6045,
      "eval_samples_per_second": 51.603,
      "eval_steps_per_second": 3.329,
      "step": 1700
    },
    {
      "epoch": 3.86,
      "learning_rate": 6.139954853273138e-05,
      "loss": 0.0751,
      "step": 1710
    },
    {
      "epoch": 3.88,
      "learning_rate": 6.117381489841987e-05,
      "loss": 0.0661,
      "step": 1720
    },
    {
      "epoch": 3.91,
      "learning_rate": 6.094808126410836e-05,
      "loss": 0.0662,
      "step": 1730
    },
    {
      "epoch": 3.93,
      "learning_rate": 6.072234762979684e-05,
      "loss": 0.0683,
      "step": 1740
    },
    {
      "epoch": 3.95,
      "learning_rate": 6.049661399548533e-05,
      "loss": 0.0675,
      "step": 1750
    },
    {
      "epoch": 3.97,
      "learning_rate": 6.027088036117382e-05,
      "loss": 0.0718,
      "step": 1760
    },
    {
      "epoch": 4.0,
      "learning_rate": 6.00451467268623e-05,
      "loss": 0.0722,
      "step": 1770
    },
    {
      "epoch": 4.02,
      "learning_rate": 5.981941309255079e-05,
      "loss": 0.0659,
      "step": 1780
    },
    {
      "epoch": 4.04,
      "learning_rate": 5.959367945823928e-05,
      "loss": 0.067,
      "step": 1790
    },
    {
      "epoch": 4.06,
      "learning_rate": 5.936794582392777e-05,
      "loss": 0.0694,
      "step": 1800
    },
    {
      "epoch": 4.06,
      "eval_accuracy": 0.9826881720430107,
      "eval_f1": 0.030120481927710843,
      "eval_loss": 0.06988376379013062,
      "eval_precision": 0.8333333333333334,
      "eval_recall": 0.015337423312883436,
      "eval_roc_auc_score": 0.8333625970318217,
      "eval_runtime": 3.5813,
      "eval_samples_per_second": 51.936,
      "eval_steps_per_second": 3.351,
      "step": 1800
    },
    {
      "epoch": 4.09,
      "learning_rate": 5.914221218961625e-05,
      "loss": 0.0701,
      "step": 1810
    },
    {
      "epoch": 4.11,
      "learning_rate": 5.891647855530474e-05,
      "loss": 0.0655,
      "step": 1820
    },
    {
      "epoch": 4.13,
      "learning_rate": 5.869074492099323e-05,
      "loss": 0.0714,
      "step": 1830
    },
    {
      "epoch": 4.15,
      "learning_rate": 5.8465011286681716e-05,
      "loss": 0.0692,
      "step": 1840
    },
    {
      "epoch": 4.18,
      "learning_rate": 5.823927765237021e-05,
      "loss": 0.0645,
      "step": 1850
    },
    {
      "epoch": 4.2,
      "learning_rate": 5.80135440180587e-05,
      "loss": 0.0665,
      "step": 1860
    },
    {
      "epoch": 4.22,
      "learning_rate": 5.7787810383747176e-05,
      "loss": 0.064,
      "step": 1870
    },
    {
      "epoch": 4.24,
      "learning_rate": 5.7562076749435666e-05,
      "loss": 0.0649,
      "step": 1880
    },
    {
      "epoch": 4.27,
      "learning_rate": 5.733634311512416e-05,
      "loss": 0.0708,
      "step": 1890
    },
    {
      "epoch": 4.29,
      "learning_rate": 5.711060948081265e-05,
      "loss": 0.0623,
      "step": 1900
    },
    {
      "epoch": 4.29,
      "eval_accuracy": 0.982741935483871,
      "eval_f1": 0.036036036036036036,
      "eval_loss": 0.07092401385307312,
      "eval_precision": 0.8571428571428571,
      "eval_recall": 0.018404907975460124,
      "eval_roc_auc_score": 0.827388270303915,
      "eval_runtime": 3.5366,
      "eval_samples_per_second": 52.593,
      "eval_steps_per_second": 3.393,
      "step": 1900
    },
    {
      "epoch": 4.31,
      "learning_rate": 5.688487584650113e-05,
      "loss": 0.0661,
      "step": 1910
    },
    {
      "epoch": 4.33,
      "learning_rate": 5.665914221218962e-05,
      "loss": 0.0686,
      "step": 1920
    },
    {
      "epoch": 4.36,
      "learning_rate": 5.6433408577878114e-05,
      "loss": 0.0666,
      "step": 1930
    },
    {
      "epoch": 4.38,
      "learning_rate": 5.620767494356659e-05,
      "loss": 0.0678,
      "step": 1940
    },
    {
      "epoch": 4.4,
      "learning_rate": 5.598194130925508e-05,
      "loss": 0.0654,
      "step": 1950
    },
    {
      "epoch": 4.42,
      "learning_rate": 5.575620767494357e-05,
      "loss": 0.0624,
      "step": 1960
    },
    {
      "epoch": 4.45,
      "learning_rate": 5.553047404063205e-05,
      "loss": 0.064,
      "step": 1970
    },
    {
      "epoch": 4.47,
      "learning_rate": 5.530474040632054e-05,
      "loss": 0.0694,
      "step": 1980
    },
    {
      "epoch": 4.49,
      "learning_rate": 5.507900677200903e-05,
      "loss": 0.0652,
      "step": 1990
    },
    {
      "epoch": 4.51,
      "learning_rate": 5.485327313769752e-05,
      "loss": 0.0647,
      "step": 2000
    },
    {
      "epoch": 4.51,
      "eval_accuracy": 0.9829032258064516,
      "eval_f1": 0.053571428571428575,
      "eval_loss": 0.06878232210874557,
      "eval_precision": 0.9,
      "eval_recall": 0.027607361963190184,
      "eval_roc_auc_score": 0.8427663158827688,
      "eval_runtime": 3.7561,
      "eval_samples_per_second": 49.52,
      "eval_steps_per_second": 3.195,
      "step": 2000
    },
    {
      "epoch": 4.54,
      "learning_rate": 5.462753950338601e-05,
      "loss": 0.0702,
      "step": 2010
    },
    {
      "epoch": 4.56,
      "learning_rate": 5.44018058690745e-05,
      "loss": 0.0703,
      "step": 2020
    },
    {
      "epoch": 4.58,
      "learning_rate": 5.417607223476299e-05,
      "loss": 0.0676,
      "step": 2030
    },
    {
      "epoch": 4.6,
      "learning_rate": 5.3950338600451466e-05,
      "loss": 0.0615,
      "step": 2040
    },
    {
      "epoch": 4.63,
      "learning_rate": 5.372460496613996e-05,
      "loss": 0.0655,
      "step": 2050
    },
    {
      "epoch": 4.65,
      "learning_rate": 5.349887133182845e-05,
      "loss": 0.0657,
      "step": 2060
    },
    {
      "epoch": 4.67,
      "learning_rate": 5.3273137697516925e-05,
      "loss": 0.0647,
      "step": 2070
    },
    {
      "epoch": 4.7,
      "learning_rate": 5.3047404063205416e-05,
      "loss": 0.0618,
      "step": 2080
    },
    {
      "epoch": 4.72,
      "learning_rate": 5.282167042889391e-05,
      "loss": 0.064,
      "step": 2090
    },
    {
      "epoch": 4.74,
      "learning_rate": 5.259593679458239e-05,
      "loss": 0.0643,
      "step": 2100
    },
    {
      "epoch": 4.74,
      "eval_accuracy": 0.9828494623655915,
      "eval_f1": 0.053412462908011875,
      "eval_loss": 0.0678834468126297,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.027607361963190184,
      "eval_roc_auc_score": 0.850500325313849,
      "eval_runtime": 3.5589,
      "eval_samples_per_second": 52.263,
      "eval_steps_per_second": 3.372,
      "step": 2100
    },
    {
      "epoch": 4.76,
      "learning_rate": 5.237020316027088e-05,
      "loss": 0.0673,
      "step": 2110
    },
    {
      "epoch": 4.79,
      "learning_rate": 5.214446952595937e-05,
      "loss": 0.0615,
      "step": 2120
    },
    {
      "epoch": 4.81,
      "learning_rate": 5.1918735891647864e-05,
      "loss": 0.0652,
      "step": 2130
    },
    {
      "epoch": 4.83,
      "learning_rate": 5.169300225733634e-05,
      "loss": 0.069,
      "step": 2140
    },
    {
      "epoch": 4.85,
      "learning_rate": 5.146726862302483e-05,
      "loss": 0.0676,
      "step": 2150
    },
    {
      "epoch": 4.88,
      "learning_rate": 5.124153498871332e-05,
      "loss": 0.0635,
      "step": 2160
    },
    {
      "epoch": 4.9,
      "learning_rate": 5.101580135440181e-05,
      "loss": 0.0676,
      "step": 2170
    },
    {
      "epoch": 4.92,
      "learning_rate": 5.07900677200903e-05,
      "loss": 0.0649,
      "step": 2180
    },
    {
      "epoch": 4.94,
      "learning_rate": 5.056433408577879e-05,
      "loss": 0.0679,
      "step": 2190
    },
    {
      "epoch": 4.97,
      "learning_rate": 5.0338600451467266e-05,
      "loss": 0.0603,
      "step": 2200
    },
    {
      "epoch": 4.97,
      "eval_accuracy": 0.9828494623655915,
      "eval_f1": 0.053412462908011875,
      "eval_loss": 0.06742475926876068,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.027607361963190184,
      "eval_roc_auc_score": 0.8569804999694494,
      "eval_runtime": 3.5629,
      "eval_samples_per_second": 52.204,
      "eval_steps_per_second": 3.368,
      "step": 2200
    },
    {
      "epoch": 4.99,
      "learning_rate": 5.011286681715576e-05,
      "loss": 0.0623,
      "step": 2210
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.988713318284425e-05,
      "loss": 0.0613,
      "step": 2220
    },
    {
      "epoch": 5.03,
      "learning_rate": 4.966139954853273e-05,
      "loss": 0.0611,
      "step": 2230
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.9435665914221216e-05,
      "loss": 0.0592,
      "step": 2240
    },
    {
      "epoch": 5.08,
      "learning_rate": 4.920993227990971e-05,
      "loss": 0.0642,
      "step": 2250
    },
    {
      "epoch": 5.1,
      "learning_rate": 4.89841986455982e-05,
      "loss": 0.061,
      "step": 2260
    },
    {
      "epoch": 5.12,
      "learning_rate": 4.875846501128669e-05,
      "loss": 0.0619,
      "step": 2270
    },
    {
      "epoch": 5.15,
      "learning_rate": 4.853273137697517e-05,
      "loss": 0.0592,
      "step": 2280
    },
    {
      "epoch": 5.17,
      "learning_rate": 4.830699774266366e-05,
      "loss": 0.0595,
      "step": 2290
    },
    {
      "epoch": 5.19,
      "learning_rate": 4.808126410835215e-05,
      "loss": 0.0644,
      "step": 2300
    },
    {
      "epoch": 5.19,
      "eval_accuracy": 0.9829569892473118,
      "eval_f1": 0.05934718100890207,
      "eval_loss": 0.06638684123754501,
      "eval_precision": 0.9090909090909091,
      "eval_recall": 0.03067484662576687,
      "eval_roc_auc_score": 0.8633127894336452,
      "eval_runtime": 3.5721,
      "eval_samples_per_second": 52.071,
      "eval_steps_per_second": 3.359,
      "step": 2300
    },
    {
      "epoch": 5.21,
      "learning_rate": 4.785553047404063e-05,
      "loss": 0.0603,
      "step": 2310
    },
    {
      "epoch": 5.24,
      "learning_rate": 4.762979683972912e-05,
      "loss": 0.0621,
      "step": 2320
    },
    {
      "epoch": 5.26,
      "learning_rate": 4.740406320541761e-05,
      "loss": 0.0598,
      "step": 2330
    },
    {
      "epoch": 5.28,
      "learning_rate": 4.71783295711061e-05,
      "loss": 0.066,
      "step": 2340
    },
    {
      "epoch": 5.3,
      "learning_rate": 4.695259593679459e-05,
      "loss": 0.0579,
      "step": 2350
    },
    {
      "epoch": 5.33,
      "learning_rate": 4.672686230248307e-05,
      "loss": 0.0613,
      "step": 2360
    },
    {
      "epoch": 5.35,
      "learning_rate": 4.6501128668171564e-05,
      "loss": 0.0594,
      "step": 2370
    },
    {
      "epoch": 5.37,
      "learning_rate": 4.627539503386005e-05,
      "loss": 0.0632,
      "step": 2380
    },
    {
      "epoch": 5.4,
      "learning_rate": 4.604966139954853e-05,
      "loss": 0.0662,
      "step": 2390
    },
    {
      "epoch": 5.42,
      "learning_rate": 4.582392776523702e-05,
      "loss": 0.0651,
      "step": 2400
    },
    {
      "epoch": 5.42,
      "eval_accuracy": 0.9830107526881721,
      "eval_f1": 0.07058823529411765,
      "eval_loss": 0.06618800014257431,
      "eval_precision": 0.8571428571428571,
      "eval_recall": 0.03680981595092025,
      "eval_roc_auc_score": 0.8666242594829492,
      "eval_runtime": 3.7032,
      "eval_samples_per_second": 50.226,
      "eval_steps_per_second": 3.24,
      "step": 2400
    },
    {
      "epoch": 5.44,
      "learning_rate": 4.559819413092551e-05,
      "loss": 0.0647,
      "step": 2410
    },
    {
      "epoch": 5.46,
      "learning_rate": 4.5372460496614e-05,
      "loss": 0.0613,
      "step": 2420
    },
    {
      "epoch": 5.49,
      "learning_rate": 4.514672686230249e-05,
      "loss": 0.065,
      "step": 2430
    },
    {
      "epoch": 5.51,
      "learning_rate": 4.492099322799097e-05,
      "loss": 0.0587,
      "step": 2440
    },
    {
      "epoch": 5.53,
      "learning_rate": 4.4695259593679463e-05,
      "loss": 0.0617,
      "step": 2450
    },
    {
      "epoch": 5.55,
      "learning_rate": 4.446952595936795e-05,
      "loss": 0.0643,
      "step": 2460
    },
    {
      "epoch": 5.58,
      "learning_rate": 4.424379232505644e-05,
      "loss": 0.0591,
      "step": 2470
    },
    {
      "epoch": 5.6,
      "learning_rate": 4.401805869074492e-05,
      "loss": 0.0643,
      "step": 2480
    },
    {
      "epoch": 5.62,
      "learning_rate": 4.379232505643341e-05,
      "loss": 0.0606,
      "step": 2490
    },
    {
      "epoch": 5.64,
      "learning_rate": 4.35665914221219e-05,
      "loss": 0.0617,
      "step": 2500
    },
    {
      "epoch": 5.64,
      "eval_accuracy": 0.9833870967741936,
      "eval_f1": 0.16711590296495957,
      "eval_loss": 0.06510370969772339,
      "eval_precision": 0.6888888888888889,
      "eval_recall": 0.0950920245398773,
      "eval_roc_auc_score": 0.869350399608952,
      "eval_runtime": 3.6683,
      "eval_samples_per_second": 50.704,
      "eval_steps_per_second": 3.271,
      "step": 2500
    },
    {
      "epoch": 5.67,
      "learning_rate": 4.334085778781038e-05,
      "loss": 0.0686,
      "step": 2510
    },
    {
      "epoch": 5.69,
      "learning_rate": 4.311512415349887e-05,
      "loss": 0.0595,
      "step": 2520
    },
    {
      "epoch": 5.71,
      "learning_rate": 4.2889390519187363e-05,
      "loss": 0.0604,
      "step": 2530
    },
    {
      "epoch": 5.73,
      "learning_rate": 4.266365688487585e-05,
      "loss": 0.0586,
      "step": 2540
    },
    {
      "epoch": 5.76,
      "learning_rate": 4.243792325056434e-05,
      "loss": 0.0623,
      "step": 2550
    },
    {
      "epoch": 5.78,
      "learning_rate": 4.221218961625282e-05,
      "loss": 0.06,
      "step": 2560
    },
    {
      "epoch": 5.8,
      "learning_rate": 4.198645598194131e-05,
      "loss": 0.0602,
      "step": 2570
    },
    {
      "epoch": 5.82,
      "learning_rate": 4.17607223476298e-05,
      "loss": 0.0636,
      "step": 2580
    },
    {
      "epoch": 5.85,
      "learning_rate": 4.153498871331828e-05,
      "loss": 0.0621,
      "step": 2590
    },
    {
      "epoch": 5.87,
      "learning_rate": 4.130925507900677e-05,
      "loss": 0.0618,
      "step": 2600
    },
    {
      "epoch": 5.87,
      "eval_accuracy": 0.983225806451613,
      "eval_f1": 0.08771929824561404,
      "eval_loss": 0.06429748982191086,
      "eval_precision": 0.9375,
      "eval_recall": 0.046012269938650305,
      "eval_roc_auc_score": 0.8721356266672755,
      "eval_runtime": 3.5388,
      "eval_samples_per_second": 52.561,
      "eval_steps_per_second": 3.391,
      "step": 2600
    },
    {
      "epoch": 5.89,
      "learning_rate": 4.108352144469526e-05,
      "loss": 0.0598,
      "step": 2610
    },
    {
      "epoch": 5.91,
      "learning_rate": 4.085778781038375e-05,
      "loss": 0.0642,
      "step": 2620
    },
    {
      "epoch": 5.94,
      "learning_rate": 4.063205417607224e-05,
      "loss": 0.0584,
      "step": 2630
    },
    {
      "epoch": 5.96,
      "learning_rate": 4.040632054176072e-05,
      "loss": 0.057,
      "step": 2640
    },
    {
      "epoch": 5.98,
      "learning_rate": 4.018058690744921e-05,
      "loss": 0.0607,
      "step": 2650
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.99548532731377e-05,
      "loss": 0.0595,
      "step": 2660
    },
    {
      "epoch": 6.03,
      "learning_rate": 3.972911963882618e-05,
      "loss": 0.0573,
      "step": 2670
    },
    {
      "epoch": 6.05,
      "learning_rate": 3.950338600451467e-05,
      "loss": 0.0625,
      "step": 2680
    },
    {
      "epoch": 6.07,
      "learning_rate": 3.927765237020316e-05,
      "loss": 0.0544,
      "step": 2690
    },
    {
      "epoch": 6.09,
      "learning_rate": 3.9051918735891654e-05,
      "loss": 0.0613,
      "step": 2700
    },
    {
      "epoch": 6.09,
      "eval_accuracy": 0.9831720430107527,
      "eval_f1": 0.15633423180592992,
      "eval_loss": 0.06485240906476974,
      "eval_precision": 0.6444444444444445,
      "eval_recall": 0.08895705521472393,
      "eval_roc_auc_score": 0.8776753622935398,
      "eval_runtime": 3.5663,
      "eval_samples_per_second": 52.155,
      "eval_steps_per_second": 3.365,
      "step": 2700
    },
    {
      "epoch": 6.12,
      "learning_rate": 3.882618510158014e-05,
      "loss": 0.0582,
      "step": 2710
    },
    {
      "epoch": 6.14,
      "learning_rate": 3.860045146726862e-05,
      "loss": 0.0548,
      "step": 2720
    },
    {
      "epoch": 6.16,
      "learning_rate": 3.837471783295711e-05,
      "loss": 0.0586,
      "step": 2730
    },
    {
      "epoch": 6.19,
      "learning_rate": 3.81489841986456e-05,
      "loss": 0.0611,
      "step": 2740
    },
    {
      "epoch": 6.21,
      "learning_rate": 3.792325056433409e-05,
      "loss": 0.0585,
      "step": 2750
    },
    {
      "epoch": 6.23,
      "learning_rate": 3.769751693002257e-05,
      "loss": 0.0572,
      "step": 2760
    },
    {
      "epoch": 6.25,
      "learning_rate": 3.747178329571106e-05,
      "loss": 0.0561,
      "step": 2770
    },
    {
      "epoch": 6.28,
      "learning_rate": 3.7246049661399554e-05,
      "loss": 0.0626,
      "step": 2780
    },
    {
      "epoch": 6.3,
      "learning_rate": 3.702031602708804e-05,
      "loss": 0.0543,
      "step": 2790
    },
    {
      "epoch": 6.32,
      "learning_rate": 3.679458239277653e-05,
      "loss": 0.061,
      "step": 2800
    },
    {
      "epoch": 6.32,
      "eval_accuracy": 0.9834408602150537,
      "eval_f1": 0.11494252873563218,
      "eval_loss": 0.06335639208555222,
      "eval_precision": 0.9090909090909091,
      "eval_recall": 0.06134969325153374,
      "eval_roc_auc_score": 0.8828241505749896,
      "eval_runtime": 3.5523,
      "eval_samples_per_second": 52.361,
      "eval_steps_per_second": 3.378,
      "step": 2800
    },
    {
      "epoch": 6.34,
      "learning_rate": 3.656884875846501e-05,
      "loss": 0.0554,
      "step": 2810
    },
    {
      "epoch": 6.37,
      "learning_rate": 3.63431151241535e-05,
      "loss": 0.0605,
      "step": 2820
    },
    {
      "epoch": 6.39,
      "learning_rate": 3.611738148984199e-05,
      "loss": 0.0579,
      "step": 2830
    },
    {
      "epoch": 6.41,
      "learning_rate": 3.589164785553047e-05,
      "loss": 0.0587,
      "step": 2840
    },
    {
      "epoch": 6.43,
      "learning_rate": 3.566591422121896e-05,
      "loss": 0.0558,
      "step": 2850
    },
    {
      "epoch": 6.46,
      "learning_rate": 3.5440180586907454e-05,
      "loss": 0.0551,
      "step": 2860
    },
    {
      "epoch": 6.48,
      "learning_rate": 3.521444695259594e-05,
      "loss": 0.0585,
      "step": 2870
    },
    {
      "epoch": 6.5,
      "learning_rate": 3.498871331828443e-05,
      "loss": 0.0601,
      "step": 2880
    },
    {
      "epoch": 6.52,
      "learning_rate": 3.476297968397291e-05,
      "loss": 0.061,
      "step": 2890
    },
    {
      "epoch": 6.55,
      "learning_rate": 3.4537246049661404e-05,
      "loss": 0.0566,
      "step": 2900
    },
    {
      "epoch": 6.55,
      "eval_accuracy": 0.983494623655914,
      "eval_f1": 0.13031161473087816,
      "eval_loss": 0.062783382833004,
      "eval_precision": 0.8518518518518519,
      "eval_recall": 0.0705521472392638,
      "eval_roc_auc_score": 0.8880796982000643,
      "eval_runtime": 3.7939,
      "eval_samples_per_second": 49.026,
      "eval_steps_per_second": 3.163,
      "step": 2900
    },
    {
      "epoch": 6.57,
      "learning_rate": 3.431151241534989e-05,
      "loss": 0.0603,
      "step": 2910
    },
    {
      "epoch": 6.59,
      "learning_rate": 3.408577878103837e-05,
      "loss": 0.0601,
      "step": 2920
    },
    {
      "epoch": 6.61,
      "learning_rate": 3.386004514672686e-05,
      "loss": 0.0599,
      "step": 2930
    },
    {
      "epoch": 6.64,
      "learning_rate": 3.3634311512415354e-05,
      "loss": 0.0572,
      "step": 2940
    },
    {
      "epoch": 6.66,
      "learning_rate": 3.3408577878103845e-05,
      "loss": 0.0562,
      "step": 2950
    },
    {
      "epoch": 6.68,
      "learning_rate": 3.318284424379233e-05,
      "loss": 0.0522,
      "step": 2960
    },
    {
      "epoch": 6.7,
      "learning_rate": 3.295711060948081e-05,
      "loss": 0.0575,
      "step": 2970
    },
    {
      "epoch": 6.73,
      "learning_rate": 3.2731376975169304e-05,
      "loss": 0.0541,
      "step": 2980
    },
    {
      "epoch": 6.75,
      "learning_rate": 3.250564334085779e-05,
      "loss": 0.0591,
      "step": 2990
    },
    {
      "epoch": 6.77,
      "learning_rate": 3.227990970654628e-05,
      "loss": 0.0565,
      "step": 3000
    },
    {
      "epoch": 6.77,
      "eval_accuracy": 0.9833870967741936,
      "eval_f1": 0.1146131805157593,
      "eval_loss": 0.06268052756786346,
      "eval_precision": 0.8695652173913043,
      "eval_recall": 0.06134969325153374,
      "eval_roc_auc_score": 0.8874571871531579,
      "eval_runtime": 3.5722,
      "eval_samples_per_second": 52.068,
      "eval_steps_per_second": 3.359,
      "step": 3000
    },
    {
      "epoch": 6.79,
      "learning_rate": 3.205417607223476e-05,
      "loss": 0.0578,
      "step": 3010
    },
    {
      "epoch": 6.82,
      "learning_rate": 3.1828442437923254e-05,
      "loss": 0.0567,
      "step": 3020
    },
    {
      "epoch": 6.84,
      "learning_rate": 3.1602708803611745e-05,
      "loss": 0.055,
      "step": 3030
    },
    {
      "epoch": 6.86,
      "learning_rate": 3.137697516930023e-05,
      "loss": 0.0575,
      "step": 3040
    },
    {
      "epoch": 6.88,
      "learning_rate": 3.115124153498871e-05,
      "loss": 0.0549,
      "step": 3050
    },
    {
      "epoch": 6.91,
      "learning_rate": 3.0925507900677204e-05,
      "loss": 0.0581,
      "step": 3060
    },
    {
      "epoch": 6.93,
      "learning_rate": 3.069977426636569e-05,
      "loss": 0.0564,
      "step": 3070
    },
    {
      "epoch": 6.95,
      "learning_rate": 3.047404063205418e-05,
      "loss": 0.0546,
      "step": 3080
    },
    {
      "epoch": 6.98,
      "learning_rate": 3.0248306997742666e-05,
      "loss": 0.0533,
      "step": 3090
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.002257336343115e-05,
      "loss": 0.0601,
      "step": 3100
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9833333333333333,
      "eval_f1": 0.10404624277456648,
      "eval_loss": 0.06219353899359703,
      "eval_precision": 0.9,
      "eval_recall": 0.05521472392638037,
      "eval_roc_auc_score": 0.8896865438240391,
      "eval_runtime": 3.5532,
      "eval_samples_per_second": 52.348,
      "eval_steps_per_second": 3.377,
      "step": 3100
    },
    {
      "epoch": 7.02,
      "learning_rate": 2.979683972911964e-05,
      "loss": 0.0547,
      "step": 3110
    },
    {
      "epoch": 7.04,
      "learning_rate": 2.9571106094808125e-05,
      "loss": 0.0577,
      "step": 3120
    },
    {
      "epoch": 7.07,
      "learning_rate": 2.9345372460496616e-05,
      "loss": 0.0535,
      "step": 3130
    },
    {
      "epoch": 7.09,
      "learning_rate": 2.9119638826185104e-05,
      "loss": 0.0554,
      "step": 3140
    },
    {
      "epoch": 7.11,
      "learning_rate": 2.8893905191873588e-05,
      "loss": 0.0556,
      "step": 3150
    },
    {
      "epoch": 7.13,
      "learning_rate": 2.866817155756208e-05,
      "loss": 0.0552,
      "step": 3160
    },
    {
      "epoch": 7.16,
      "learning_rate": 2.8442437923250566e-05,
      "loss": 0.0572,
      "step": 3170
    },
    {
      "epoch": 7.18,
      "learning_rate": 2.8216704288939057e-05,
      "loss": 0.0557,
      "step": 3180
    },
    {
      "epoch": 7.2,
      "learning_rate": 2.799097065462754e-05,
      "loss": 0.0567,
      "step": 3190
    },
    {
      "epoch": 7.22,
      "learning_rate": 2.7765237020316025e-05,
      "loss": 0.0568,
      "step": 3200
    },
    {
      "epoch": 7.22,
      "eval_accuracy": 0.9837096774193549,
      "eval_f1": 0.141643059490085,
      "eval_loss": 0.06185278296470642,
      "eval_precision": 0.9259259259259259,
      "eval_recall": 0.07668711656441718,
      "eval_roc_auc_score": 0.8928906670176073,
      "eval_runtime": 3.5918,
      "eval_samples_per_second": 51.785,
      "eval_steps_per_second": 3.341,
      "step": 3200
    },
    {
      "epoch": 7.25,
      "learning_rate": 2.7539503386004516e-05,
      "loss": 0.0537,
      "step": 3210
    },
    {
      "epoch": 7.27,
      "learning_rate": 2.7313769751693004e-05,
      "loss": 0.0525,
      "step": 3220
    },
    {
      "epoch": 7.29,
      "learning_rate": 2.7088036117381494e-05,
      "loss": 0.053,
      "step": 3230
    },
    {
      "epoch": 7.31,
      "learning_rate": 2.686230248306998e-05,
      "loss": 0.0558,
      "step": 3240
    },
    {
      "epoch": 7.34,
      "learning_rate": 2.6636568848758463e-05,
      "loss": 0.0584,
      "step": 3250
    },
    {
      "epoch": 7.36,
      "learning_rate": 2.6410835214446954e-05,
      "loss": 0.0519,
      "step": 3260
    },
    {
      "epoch": 7.38,
      "learning_rate": 2.618510158013544e-05,
      "loss": 0.0512,
      "step": 3270
    },
    {
      "epoch": 7.4,
      "learning_rate": 2.5959367945823932e-05,
      "loss": 0.055,
      "step": 3280
    },
    {
      "epoch": 7.43,
      "learning_rate": 2.5733634311512416e-05,
      "loss": 0.0515,
      "step": 3290
    },
    {
      "epoch": 7.45,
      "learning_rate": 2.5507900677200903e-05,
      "loss": 0.052,
      "step": 3300
    },
    {
      "epoch": 7.45,
      "eval_accuracy": 0.9837096774193549,
      "eval_f1": 0.141643059490085,
      "eval_loss": 0.06193934381008148,
      "eval_precision": 0.9259259259259259,
      "eval_recall": 0.07668711656441718,
      "eval_roc_auc_score": 0.8928445892820334,
      "eval_runtime": 3.5473,
      "eval_samples_per_second": 52.434,
      "eval_steps_per_second": 3.383,
      "step": 3300
    },
    {
      "epoch": 7.47,
      "learning_rate": 2.5282167042889394e-05,
      "loss": 0.0561,
      "step": 3310
    },
    {
      "epoch": 7.49,
      "learning_rate": 2.505643340857788e-05,
      "loss": 0.0601,
      "step": 3320
    },
    {
      "epoch": 7.52,
      "learning_rate": 2.4830699774266366e-05,
      "loss": 0.0545,
      "step": 3330
    },
    {
      "epoch": 7.54,
      "learning_rate": 2.4604966139954853e-05,
      "loss": 0.0542,
      "step": 3340
    },
    {
      "epoch": 7.56,
      "learning_rate": 2.4379232505643344e-05,
      "loss": 0.0528,
      "step": 3350
    },
    {
      "epoch": 7.58,
      "learning_rate": 2.415349887133183e-05,
      "loss": 0.0565,
      "step": 3360
    },
    {
      "epoch": 7.61,
      "learning_rate": 2.3927765237020316e-05,
      "loss": 0.0541,
      "step": 3370
    },
    {
      "epoch": 7.63,
      "learning_rate": 2.3702031602708803e-05,
      "loss": 0.0551,
      "step": 3380
    },
    {
      "epoch": 7.65,
      "learning_rate": 2.3476297968397294e-05,
      "loss": 0.0576,
      "step": 3390
    },
    {
      "epoch": 7.67,
      "learning_rate": 2.3250564334085782e-05,
      "loss": 0.0509,
      "step": 3400
    },
    {
      "epoch": 7.67,
      "eval_accuracy": 0.9836021505376344,
      "eval_f1": 0.14084507042253522,
      "eval_loss": 0.06077234819531441,
      "eval_precision": 0.8620689655172413,
      "eval_recall": 0.07668711656441718,
      "eval_roc_auc_score": 0.8993380249252851,
      "eval_runtime": 3.6981,
      "eval_samples_per_second": 50.296,
      "eval_steps_per_second": 3.245,
      "step": 3400
    },
    {
      "epoch": 7.7,
      "learning_rate": 2.3024830699774266e-05,
      "loss": 0.0518,
      "step": 3410
    },
    {
      "epoch": 7.72,
      "learning_rate": 2.2799097065462753e-05,
      "loss": 0.0523,
      "step": 3420
    },
    {
      "epoch": 7.74,
      "learning_rate": 2.2573363431151244e-05,
      "loss": 0.0515,
      "step": 3430
    },
    {
      "epoch": 7.77,
      "learning_rate": 2.2347629796839732e-05,
      "loss": 0.0537,
      "step": 3440
    },
    {
      "epoch": 7.79,
      "learning_rate": 2.212189616252822e-05,
      "loss": 0.0503,
      "step": 3450
    },
    {
      "epoch": 7.81,
      "learning_rate": 2.1896162528216703e-05,
      "loss": 0.0491,
      "step": 3460
    },
    {
      "epoch": 7.83,
      "learning_rate": 2.167042889390519e-05,
      "loss": 0.049,
      "step": 3470
    },
    {
      "epoch": 7.86,
      "learning_rate": 2.1444695259593682e-05,
      "loss": 0.0554,
      "step": 3480
    },
    {
      "epoch": 7.88,
      "learning_rate": 2.121896162528217e-05,
      "loss": 0.0531,
      "step": 3490
    },
    {
      "epoch": 7.9,
      "learning_rate": 2.0993227990970657e-05,
      "loss": 0.0532,
      "step": 3500
    },
    {
      "epoch": 7.9,
      "eval_accuracy": 0.983494623655914,
      "eval_f1": 0.12034383954154729,
      "eval_loss": 0.06059078127145767,
      "eval_precision": 0.9130434782608695,
      "eval_recall": 0.06441717791411043,
      "eval_roc_auc_score": 0.8959880308675506,
      "eval_runtime": 3.7079,
      "eval_samples_per_second": 50.163,
      "eval_steps_per_second": 3.236,
      "step": 3500
    },
    {
      "epoch": 7.92,
      "learning_rate": 2.076749435665914e-05,
      "loss": 0.0533,
      "step": 3510
    },
    {
      "epoch": 7.95,
      "learning_rate": 2.054176072234763e-05,
      "loss": 0.0541,
      "step": 3520
    },
    {
      "epoch": 7.97,
      "learning_rate": 2.031602708803612e-05,
      "loss": 0.0546,
      "step": 3530
    },
    {
      "epoch": 7.99,
      "learning_rate": 2.0090293453724607e-05,
      "loss": 0.0545,
      "step": 3540
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.986455981941309e-05,
      "loss": 0.0458,
      "step": 3550
    },
    {
      "epoch": 8.04,
      "learning_rate": 1.963882618510158e-05,
      "loss": 0.0547,
      "step": 3560
    },
    {
      "epoch": 8.06,
      "learning_rate": 1.941309255079007e-05,
      "loss": 0.0527,
      "step": 3570
    },
    {
      "epoch": 8.08,
      "learning_rate": 1.9187358916478557e-05,
      "loss": 0.0519,
      "step": 3580
    },
    {
      "epoch": 8.1,
      "learning_rate": 1.8961625282167044e-05,
      "loss": 0.0562,
      "step": 3590
    },
    {
      "epoch": 8.13,
      "learning_rate": 1.873589164785553e-05,
      "loss": 0.0482,
      "step": 3600
    },
    {
      "epoch": 8.13,
      "eval_accuracy": 0.9838172043010752,
      "eval_f1": 0.21409921671018278,
      "eval_loss": 0.060027554631233215,
      "eval_precision": 0.7192982456140351,
      "eval_recall": 0.12576687116564417,
      "eval_roc_auc_score": 0.8988576078789738,
      "eval_runtime": 3.5656,
      "eval_samples_per_second": 52.165,
      "eval_steps_per_second": 3.365,
      "step": 3600
    },
    {
      "epoch": 8.15,
      "learning_rate": 1.851015801354402e-05,
      "loss": 0.0512,
      "step": 3610
    },
    {
      "epoch": 8.17,
      "learning_rate": 1.8284424379232507e-05,
      "loss": 0.0513,
      "step": 3620
    },
    {
      "epoch": 8.19,
      "learning_rate": 1.8058690744920994e-05,
      "loss": 0.0533,
      "step": 3630
    },
    {
      "epoch": 8.22,
      "learning_rate": 1.783295711060948e-05,
      "loss": 0.0523,
      "step": 3640
    },
    {
      "epoch": 8.24,
      "learning_rate": 1.760722347629797e-05,
      "loss": 0.0541,
      "step": 3650
    },
    {
      "epoch": 8.26,
      "learning_rate": 1.7381489841986457e-05,
      "loss": 0.0477,
      "step": 3660
    },
    {
      "epoch": 8.28,
      "learning_rate": 1.7155756207674944e-05,
      "loss": 0.051,
      "step": 3670
    },
    {
      "epoch": 8.31,
      "learning_rate": 1.693002257336343e-05,
      "loss": 0.053,
      "step": 3680
    },
    {
      "epoch": 8.33,
      "learning_rate": 1.6704288939051922e-05,
      "loss": 0.0523,
      "step": 3690
    },
    {
      "epoch": 8.35,
      "learning_rate": 1.6478555304740406e-05,
      "loss": 0.0477,
      "step": 3700
    },
    {
      "epoch": 8.35,
      "eval_accuracy": 0.9838172043010752,
      "eval_f1": 0.16155988857938716,
      "eval_loss": 0.06003206595778465,
      "eval_precision": 0.8787878787878788,
      "eval_recall": 0.08895705521472393,
      "eval_roc_auc_score": 0.9000536146766569,
      "eval_runtime": 3.5688,
      "eval_samples_per_second": 52.119,
      "eval_steps_per_second": 3.362,
      "step": 3700
    },
    {
      "epoch": 8.37,
      "learning_rate": 1.6252821670428894e-05,
      "loss": 0.0562,
      "step": 3710
    },
    {
      "epoch": 8.4,
      "learning_rate": 1.602708803611738e-05,
      "loss": 0.0501,
      "step": 3720
    },
    {
      "epoch": 8.42,
      "learning_rate": 1.5801354401805872e-05,
      "loss": 0.0587,
      "step": 3730
    },
    {
      "epoch": 8.44,
      "learning_rate": 1.5575620767494356e-05,
      "loss": 0.049,
      "step": 3740
    },
    {
      "epoch": 8.47,
      "learning_rate": 1.5349887133182844e-05,
      "loss": 0.0498,
      "step": 3750
    },
    {
      "epoch": 8.49,
      "learning_rate": 1.5124153498871333e-05,
      "loss": 0.049,
      "step": 3760
    },
    {
      "epoch": 8.51,
      "learning_rate": 1.489841986455982e-05,
      "loss": 0.0521,
      "step": 3770
    },
    {
      "epoch": 8.53,
      "learning_rate": 1.4672686230248308e-05,
      "loss": 0.0503,
      "step": 3780
    },
    {
      "epoch": 8.56,
      "learning_rate": 1.4446952595936794e-05,
      "loss": 0.0487,
      "step": 3790
    },
    {
      "epoch": 8.58,
      "learning_rate": 1.4221218961625283e-05,
      "loss": 0.0446,
      "step": 3800
    },
    {
      "epoch": 8.58,
      "eval_accuracy": 0.9838709677419355,
      "eval_f1": 0.21465968586387432,
      "eval_loss": 0.05991381034255028,
      "eval_precision": 0.7321428571428571,
      "eval_recall": 0.12576687116564417,
      "eval_roc_auc_score": 0.9004402815760901,
      "eval_runtime": 3.6747,
      "eval_samples_per_second": 50.617,
      "eval_steps_per_second": 3.266,
      "step": 3800
    },
    {
      "epoch": 8.6,
      "learning_rate": 1.399548532731377e-05,
      "loss": 0.0555,
      "step": 3810
    },
    {
      "epoch": 8.62,
      "learning_rate": 1.3769751693002258e-05,
      "loss": 0.0533,
      "step": 3820
    },
    {
      "epoch": 8.65,
      "learning_rate": 1.3544018058690747e-05,
      "loss": 0.054,
      "step": 3830
    },
    {
      "epoch": 8.67,
      "learning_rate": 1.3318284424379231e-05,
      "loss": 0.0496,
      "step": 3840
    },
    {
      "epoch": 8.69,
      "learning_rate": 1.309255079006772e-05,
      "loss": 0.0526,
      "step": 3850
    },
    {
      "epoch": 8.71,
      "learning_rate": 1.2866817155756208e-05,
      "loss": 0.048,
      "step": 3860
    },
    {
      "epoch": 8.74,
      "learning_rate": 1.2641083521444697e-05,
      "loss": 0.0518,
      "step": 3870
    },
    {
      "epoch": 8.76,
      "learning_rate": 1.2415349887133183e-05,
      "loss": 0.0496,
      "step": 3880
    },
    {
      "epoch": 8.78,
      "learning_rate": 1.2189616252821672e-05,
      "loss": 0.0509,
      "step": 3890
    },
    {
      "epoch": 8.8,
      "learning_rate": 1.1963882618510158e-05,
      "loss": 0.0541,
      "step": 3900
    },
    {
      "epoch": 8.8,
      "eval_accuracy": 0.983763440860215,
      "eval_f1": 0.1611111111111111,
      "eval_loss": 0.05945482477545738,
      "eval_precision": 0.8529411764705882,
      "eval_recall": 0.08895705521472393,
      "eval_roc_auc_score": 0.9011602357031445,
      "eval_runtime": 3.5636,
      "eval_samples_per_second": 52.194,
      "eval_steps_per_second": 3.367,
      "step": 3900
    },
    {
      "epoch": 8.83,
      "learning_rate": 1.1738148984198647e-05,
      "loss": 0.053,
      "step": 3910
    },
    {
      "epoch": 8.85,
      "learning_rate": 1.1512415349887133e-05,
      "loss": 0.0496,
      "step": 3920
    },
    {
      "epoch": 8.87,
      "learning_rate": 1.1286681715575622e-05,
      "loss": 0.0529,
      "step": 3930
    },
    {
      "epoch": 8.89,
      "learning_rate": 1.106094808126411e-05,
      "loss": 0.0507,
      "step": 3940
    },
    {
      "epoch": 8.92,
      "learning_rate": 1.0835214446952595e-05,
      "loss": 0.052,
      "step": 3950
    },
    {
      "epoch": 8.94,
      "learning_rate": 1.0609480812641085e-05,
      "loss": 0.0486,
      "step": 3960
    },
    {
      "epoch": 8.96,
      "learning_rate": 1.038374717832957e-05,
      "loss": 0.0493,
      "step": 3970
    },
    {
      "epoch": 8.98,
      "learning_rate": 1.015801354401806e-05,
      "loss": 0.0549,
      "step": 3980
    },
    {
      "epoch": 9.01,
      "learning_rate": 9.932279909706545e-06,
      "loss": 0.0469,
      "step": 3990
    },
    {
      "epoch": 9.03,
      "learning_rate": 9.706546275395035e-06,
      "loss": 0.0494,
      "step": 4000
    },
    {
      "epoch": 9.03,
      "eval_accuracy": 0.9837096774193549,
      "eval_f1": 0.15598885793871864,
      "eval_loss": 0.059342943131923676,
      "eval_precision": 0.8484848484848485,
      "eval_recall": 0.08588957055214724,
      "eval_roc_auc_score": 0.9023026110381105,
      "eval_runtime": 3.5491,
      "eval_samples_per_second": 52.408,
      "eval_steps_per_second": 3.381,
      "step": 4000
    }
  ],
  "max_steps": 4430,
  "num_train_epochs": 10,
  "total_flos": 3.392129028902093e+16,
  "trial_name": null,
  "trial_params": null
}