{
  "best_global_step": 7600,
  "best_metric": 0.8689859363434493,
  "best_model_checkpoint": "./runs_by_subfolder/seed_FGHI/ckpt/checkpoint-7600",
  "epoch": 3.5884297520661157,
  "eval_steps": 200,
  "global_step": 7600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.023612750885478158,
      "grad_norm": 26.885774612426758,
      "learning_rate": 9.8e-05,
      "loss": 0.7006,
      "step": 50
    },
    {
      "epoch": 0.047225501770956316,
      "grad_norm": 1.9265247583389282,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.8354,
      "step": 100
    },
    {
      "epoch": 0.07083825265643448,
      "grad_norm": 12.014982223510742,
      "learning_rate": 0.00019906577693040993,
      "loss": 0.7391,
      "step": 150
    },
    {
      "epoch": 0.09445100354191263,
      "grad_norm": 5.504847526550293,
      "learning_rate": 0.00019811248808388943,
      "loss": 0.694,
      "step": 200
    },
    {
      "epoch": 0.09445100354191263,
      "eval_accuracy": 0.7472242783123613,
      "eval_auc": 0.7627784504609225,
      "eval_cm_tn_fp_fn_tp": [
        1589,
        288,
        395,
        430
      ],
      "eval_f1": 0.5573558003888529,
      "eval_loss": 0.6126384735107422,
      "eval_prauc": 0.573211269991273,
      "eval_runtime": 205.4942,
      "eval_samples_per_second": 13.149,
      "eval_steps_per_second": 13.149,
      "step": 200
    },
    {
      "epoch": 0.1180637544273908,
      "grad_norm": 44.49080276489258,
      "learning_rate": 0.00019715919923736895,
      "loss": 0.6485,
      "step": 250
    },
    {
      "epoch": 0.14167650531286896,
      "grad_norm": 30.886829376220703,
      "learning_rate": 0.00019620591039084842,
      "loss": 0.6915,
      "step": 300
    },
    {
      "epoch": 0.1652892561983471,
      "grad_norm": 31.330167770385742,
      "learning_rate": 0.00019525262154432795,
      "loss": 0.6654,
      "step": 350
    },
    {
      "epoch": 0.18890200708382526,
      "grad_norm": 47.85934066772461,
      "learning_rate": 0.00019429933269780745,
      "loss": 0.6317,
      "step": 400
    },
    {
      "epoch": 0.18890200708382526,
      "eval_accuracy": 0.38341968911917096,
      "eval_auc": 0.7961996093056296,
      "eval_cm_tn_fp_fn_tp": [
        235,
        1642,
        24,
        801
      ],
      "eval_f1": 0.49020807833537333,
      "eval_loss": 0.9440730214118958,
      "eval_prauc": 0.6164847270416698,
      "eval_runtime": 205.1188,
      "eval_samples_per_second": 13.173,
      "eval_steps_per_second": 13.173,
      "step": 400
    },
    {
      "epoch": 0.21251475796930341,
      "grad_norm": 8.905306816101074,
      "learning_rate": 0.00019334604385128694,
      "loss": 0.7062,
      "step": 450
    },
    {
      "epoch": 0.2361275088547816,
      "grad_norm": 4.294670104980469,
      "learning_rate": 0.00019239275500476647,
      "loss": 0.7161,
      "step": 500
    },
    {
      "epoch": 0.2597402597402597,
      "grad_norm": 11.443839073181152,
      "learning_rate": 0.00019143946615824594,
      "loss": 0.6773,
      "step": 550
    },
    {
      "epoch": 0.2833530106257379,
      "grad_norm": 23.348785400390625,
      "learning_rate": 0.00019048617731172546,
      "loss": 0.6225,
      "step": 600
    },
    {
      "epoch": 0.2833530106257379,
      "eval_accuracy": 0.7468541820873427,
      "eval_auc": 0.8461351931676919,
      "eval_cm_tn_fp_fn_tp": [
        1757,
        120,
        564,
        261
      ],
      "eval_f1": 0.43283582089552236,
      "eval_loss": 0.49758678674697876,
      "eval_prauc": 0.6640447298188182,
      "eval_runtime": 205.8102,
      "eval_samples_per_second": 13.129,
      "eval_steps_per_second": 13.129,
      "step": 600
    },
    {
      "epoch": 0.3069657615112161,
      "grad_norm": 35.55908966064453,
      "learning_rate": 0.00018953288846520496,
      "loss": 0.5804,
      "step": 650
    },
    {
      "epoch": 0.3305785123966942,
      "grad_norm": 10.787803649902344,
      "learning_rate": 0.00018857959961868448,
      "loss": 0.6132,
      "step": 700
    },
    {
      "epoch": 0.3541912632821724,
      "grad_norm": 8.3643159866333,
      "learning_rate": 0.00018762631077216398,
      "loss": 0.6169,
      "step": 750
    },
    {
      "epoch": 0.3778040141676505,
      "grad_norm": 5.386566162109375,
      "learning_rate": 0.00018667302192564348,
      "loss": 0.63,
      "step": 800
    },
    {
      "epoch": 0.3778040141676505,
      "eval_accuracy": 0.7450037009622502,
      "eval_auc": 0.8587439660321919,
      "eval_cm_tn_fp_fn_tp": [
        1245,
        632,
        57,
        768
      ],
      "eval_f1": 0.6903370786516854,
      "eval_loss": 0.6075592637062073,
      "eval_prauc": 0.6781332740995903,
      "eval_runtime": 205.6994,
      "eval_samples_per_second": 13.136,
      "eval_steps_per_second": 13.136,
      "step": 800
    },
    {
      "epoch": 0.4014167650531287,
      "grad_norm": 14.160999298095703,
      "learning_rate": 0.00018571973307912297,
      "loss": 0.5996,
      "step": 850
    },
    {
      "epoch": 0.42502951593860683,
      "grad_norm": 6.024763107299805,
      "learning_rate": 0.0001847664442326025,
      "loss": 0.5722,
      "step": 900
    },
    {
      "epoch": 0.448642266824085,
      "grad_norm": 9.487037658691406,
      "learning_rate": 0.000183813155386082,
      "loss": 0.578,
      "step": 950
    },
    {
      "epoch": 0.4722550177095632,
      "grad_norm": 13.469863891601562,
      "learning_rate": 0.0001828598665395615,
      "loss": 0.6212,
      "step": 1000
    },
    {
      "epoch": 0.4722550177095632,
      "eval_accuracy": 0.7757216876387861,
      "eval_auc": 0.8602437803716438,
      "eval_cm_tn_fp_fn_tp": [
        1488,
        389,
        217,
        608
      ],
      "eval_f1": 0.6673984632272228,
      "eval_loss": 0.4853891432285309,
      "eval_prauc": 0.6973230007282509,
      "eval_runtime": 205.8828,
      "eval_samples_per_second": 13.124,
      "eval_steps_per_second": 13.124,
      "step": 1000
    },
    {
      "epoch": 0.49586776859504134,
      "grad_norm": 13.80494213104248,
      "learning_rate": 0.000181906577693041,
      "loss": 0.5901,
      "step": 1050
    },
    {
      "epoch": 0.5194805194805194,
      "grad_norm": 12.014090538024902,
      "learning_rate": 0.0001809532888465205,
      "loss": 0.5893,
      "step": 1100
    },
    {
      "epoch": 0.5430932703659976,
      "grad_norm": 8.504700660705566,
      "learning_rate": 0.00018,
      "loss": 0.5986,
      "step": 1150
    },
    {
      "epoch": 0.5667060212514758,
      "grad_norm": 11.56762409210205,
      "learning_rate": 0.0001790467111534795,
      "loss": 0.5968,
      "step": 1200
    },
    {
      "epoch": 0.5667060212514758,
      "eval_accuracy": 0.7479644707623982,
      "eval_auc": 0.8540575709142572,
      "eval_cm_tn_fp_fn_tp": [
        1721,
        156,
        525,
        300
      ],
      "eval_f1": 0.468384074941452,
      "eval_loss": 0.48603448271751404,
      "eval_prauc": 0.6729216596362579,
      "eval_runtime": 205.5076,
      "eval_samples_per_second": 13.148,
      "eval_steps_per_second": 13.148,
      "step": 1200
    },
    {
      "epoch": 0.5903187721369539,
      "grad_norm": 17.351064682006836,
      "learning_rate": 0.00017809342230695903,
      "loss": 0.5722,
      "step": 1250
    },
    {
      "epoch": 0.6139315230224321,
      "grad_norm": 19.315166473388672,
      "learning_rate": 0.00017714013346043853,
      "loss": 0.5721,
      "step": 1300
    },
    {
      "epoch": 0.6375442739079102,
      "grad_norm": 18.946020126342773,
      "learning_rate": 0.00017618684461391803,
      "loss": 0.543,
      "step": 1350
    },
    {
      "epoch": 0.6611570247933884,
      "grad_norm": 31.721004486083984,
      "learning_rate": 0.00017523355576739752,
      "loss": 0.5481,
      "step": 1400
    },
    {
      "epoch": 0.6611570247933884,
      "eval_accuracy": 0.7934863064396743,
      "eval_auc": 0.8765441307050257,
      "eval_cm_tn_fp_fn_tp": [
        1443,
        434,
        124,
        701
      ],
      "eval_f1": 0.7153061224489796,
      "eval_loss": 0.5041375160217285,
      "eval_prauc": 0.7236062359198674,
      "eval_runtime": 206.2016,
      "eval_samples_per_second": 13.104,
      "eval_steps_per_second": 13.104,
      "step": 1400
    },
    {
      "epoch": 0.6847697756788665,
      "grad_norm": 19.903593063354492,
      "learning_rate": 0.00017428026692087705,
      "loss": 0.5268,
      "step": 1450
    },
    {
      "epoch": 0.7083825265643447,
      "grad_norm": 10.583416938781738,
      "learning_rate": 0.00017332697807435655,
      "loss": 0.5398,
      "step": 1500
    },
    {
      "epoch": 0.731995277449823,
      "grad_norm": 6.75493860244751,
      "learning_rate": 0.00017237368922783604,
      "loss": 0.5395,
      "step": 1550
    },
    {
      "epoch": 0.755608028335301,
      "grad_norm": 17.679969787597656,
      "learning_rate": 0.00017142040038131554,
      "loss": 0.5518,
      "step": 1600
    },
    {
      "epoch": 0.755608028335301,
      "eval_accuracy": 0.6717246484085863,
      "eval_auc": 0.8371243602783294,
      "eval_cm_tn_fp_fn_tp": [
        1073,
        804,
        83,
        742
      ],
      "eval_f1": 0.625896246309574,
      "eval_loss": 0.6220740079879761,
      "eval_prauc": 0.6838386610949146,
      "eval_runtime": 205.6529,
      "eval_samples_per_second": 13.139,
      "eval_steps_per_second": 13.139,
      "step": 1600
    },
    {
      "epoch": 0.7792207792207793,
      "grad_norm": 18.73421859741211,
      "learning_rate": 0.00017046711153479504,
      "loss": 0.5626,
      "step": 1650
    },
    {
      "epoch": 0.8028335301062574,
      "grad_norm": 5.534360885620117,
      "learning_rate": 0.00016951382268827456,
      "loss": 0.5257,
      "step": 1700
    },
    {
      "epoch": 0.8264462809917356,
      "grad_norm": 35.911190032958984,
      "learning_rate": 0.00016856053384175406,
      "loss": 0.5162,
      "step": 1750
    },
    {
      "epoch": 0.8500590318772137,
      "grad_norm": 10.834784507751465,
      "learning_rate": 0.00016760724499523358,
      "loss": 0.5308,
      "step": 1800
    },
    {
      "epoch": 0.8500590318772137,
      "eval_accuracy": 0.7901554404145078,
      "eval_auc": 0.8671716633570656,
      "eval_cm_tn_fp_fn_tp": [
        1528,
        349,
        218,
        607
      ],
      "eval_f1": 0.6816395283548569,
      "eval_loss": 0.4776131510734558,
      "eval_prauc": 0.7076089276124425,
      "eval_runtime": 205.4984,
      "eval_samples_per_second": 13.149,
      "eval_steps_per_second": 13.149,
      "step": 1800
    },
    {
      "epoch": 0.8736717827626919,
      "grad_norm": 5.949131488800049,
      "learning_rate": 0.00016665395614871305,
      "loss": 0.4887,
      "step": 1850
    },
    {
      "epoch": 0.89728453364817,
      "grad_norm": 18.94248390197754,
      "learning_rate": 0.00016570066730219258,
      "loss": 0.474,
      "step": 1900
    },
    {
      "epoch": 0.9208972845336482,
      "grad_norm": 23.43456268310547,
      "learning_rate": 0.00016474737845567207,
      "loss": 0.5009,
      "step": 1950
    },
    {
      "epoch": 0.9445100354191264,
      "grad_norm": 10.950305938720703,
      "learning_rate": 0.0001637940896091516,
      "loss": 0.5191,
      "step": 2000
    },
    {
      "epoch": 0.9445100354191264,
      "eval_accuracy": 0.7879348630643968,
      "eval_auc": 0.854450848387982,
      "eval_cm_tn_fp_fn_tp": [
        1674,
        203,
        370,
        455
      ],
      "eval_f1": 0.6136210384356036,
      "eval_loss": 0.4846158027648926,
      "eval_prauc": 0.679086820640918,
      "eval_runtime": 205.8588,
      "eval_samples_per_second": 13.125,
      "eval_steps_per_second": 13.125,
      "step": 2000
    },
    {
      "epoch": 0.9681227863046045,
      "grad_norm": 6.395274639129639,
      "learning_rate": 0.0001628408007626311,
      "loss": 0.512,
      "step": 2050
    },
    {
      "epoch": 0.9917355371900827,
      "grad_norm": 15.790814399719238,
      "learning_rate": 0.00016188751191611057,
      "loss": 0.4957,
      "step": 2100
    },
    {
      "epoch": 1.015112160566706,
      "grad_norm": 8.16740608215332,
      "learning_rate": 0.0001609342230695901,
      "loss": 0.5373,
      "step": 2150
    },
    {
      "epoch": 1.0387249114521842,
      "grad_norm": 12.968246459960938,
      "learning_rate": 0.0001599809342230696,
      "loss": 0.4559,
      "step": 2200
    },
    {
      "epoch": 1.0387249114521842,
      "eval_accuracy": 0.7916358253145818,
      "eval_auc": 0.8525183642498506,
      "eval_cm_tn_fp_fn_tp": [
        1495,
        382,
        181,
        644
      ],
      "eval_f1": 0.6958400864397623,
      "eval_loss": 0.5206837058067322,
      "eval_prauc": 0.7212775169816144,
      "eval_runtime": 205.5639,
      "eval_samples_per_second": 13.144,
      "eval_steps_per_second": 13.144,
      "step": 2200
    },
    {
      "epoch": 1.0623376623376624,
      "grad_norm": 28.586788177490234,
      "learning_rate": 0.0001590276453765491,
      "loss": 0.5093,
      "step": 2250
    },
    {
      "epoch": 1.0859504132231406,
      "grad_norm": 27.615903854370117,
      "learning_rate": 0.0001580743565300286,
      "loss": 0.5065,
      "step": 2300
    },
    {
      "epoch": 1.1095631641086188,
      "grad_norm": 53.75819396972656,
      "learning_rate": 0.0001571210676835081,
      "loss": 0.4824,
      "step": 2350
    },
    {
      "epoch": 1.1331759149940968,
      "grad_norm": 8.528438568115234,
      "learning_rate": 0.0001561677788369876,
      "loss": 0.4322,
      "step": 2400
    },
    {
      "epoch": 1.1331759149940968,
      "eval_accuracy": 0.7616580310880829,
      "eval_auc": 0.8179680663857543,
      "eval_cm_tn_fp_fn_tp": [
        1465,
        412,
        232,
        593
      ],
      "eval_f1": 0.6480874316939891,
      "eval_loss": 0.5494776964187622,
      "eval_prauc": 0.6808013830886499,
      "eval_runtime": 206.703,
      "eval_samples_per_second": 13.072,
      "eval_steps_per_second": 13.072,
      "step": 2400
    },
    {
      "epoch": 1.156788665879575,
      "grad_norm": 29.651885986328125,
      "learning_rate": 0.00015521448999046713,
      "loss": 0.4388,
      "step": 2450
    },
    {
      "epoch": 1.1804014167650532,
      "grad_norm": 18.571435928344727,
      "learning_rate": 0.00015426120114394662,
      "loss": 0.447,
      "step": 2500
    },
    {
      "epoch": 1.2040141676505314,
      "grad_norm": 16.326656341552734,
      "learning_rate": 0.00015330791229742615,
      "loss": 0.4583,
      "step": 2550
    },
    {
      "epoch": 1.2276269185360094,
      "grad_norm": 3.8166370391845703,
      "learning_rate": 0.00015235462345090562,
      "loss": 0.4852,
      "step": 2600
    },
    {
      "epoch": 1.2276269185360094,
      "eval_accuracy": 0.8064396743153219,
      "eval_auc": 0.8901767811304305,
      "eval_cm_tn_fp_fn_tp": [
        1716,
        161,
        362,
        463
      ],
      "eval_f1": 0.6390614216701174,
      "eval_loss": 0.4528096914291382,
      "eval_prauc": 0.7433076094213714,
      "eval_runtime": 205.278,
      "eval_samples_per_second": 13.163,
      "eval_steps_per_second": 13.163,
      "step": 2600
    },
    {
      "epoch": 1.2512396694214876,
      "grad_norm": 11.237424850463867,
      "learning_rate": 0.00015140133460438512,
      "loss": 0.4496,
      "step": 2650
    },
    {
      "epoch": 1.2748524203069658,
      "grad_norm": 33.1157112121582,
      "learning_rate": 0.00015044804575786464,
      "loss": 0.4818,
      "step": 2700
    },
    {
      "epoch": 1.298465171192444,
      "grad_norm": 18.051271438598633,
      "learning_rate": 0.00014949475691134414,
      "loss": 0.4794,
      "step": 2750
    },
    {
      "epoch": 1.3220779220779222,
      "grad_norm": 8.78036880493164,
      "learning_rate": 0.00014854146806482366,
      "loss": 0.4548,
      "step": 2800
    },
    {
      "epoch": 1.3220779220779222,
      "eval_accuracy": 0.8238341968911918,
      "eval_auc": 0.87179832421175,
      "eval_cm_tn_fp_fn_tp": [
        1663,
        214,
        262,
        563
      ],
      "eval_f1": 0.7028714107365793,
      "eval_loss": 0.457581102848053,
      "eval_prauc": 0.7227392617970211,
      "eval_runtime": 205.3192,
      "eval_samples_per_second": 13.16,
      "eval_steps_per_second": 13.16,
      "step": 2800
    },
    {
      "epoch": 1.3456906729634002,
      "grad_norm": 11.54551887512207,
      "learning_rate": 0.00014758817921830316,
      "loss": 0.4363,
      "step": 2850
    },
    {
      "epoch": 1.3693034238488784,
      "grad_norm": 38.003814697265625,
      "learning_rate": 0.00014663489037178266,
      "loss": 0.4663,
      "step": 2900
    },
    {
      "epoch": 1.3929161747343566,
      "grad_norm": 5.946821212768555,
      "learning_rate": 0.00014568160152526215,
      "loss": 0.4907,
      "step": 2950
    },
    {
      "epoch": 1.4165289256198348,
      "grad_norm": 5.326801776885986,
      "learning_rate": 0.00014472831267874168,
      "loss": 0.476,
      "step": 3000
    },
    {
      "epoch": 1.4165289256198348,
      "eval_accuracy": 0.7809030347890451,
      "eval_auc": 0.8492778611904878,
      "eval_cm_tn_fp_fn_tp": [
        1736,
        141,
        451,
        374
      ],
      "eval_f1": 0.5582089552238806,
      "eval_loss": 0.48823481798171997,
      "eval_prauc": 0.6924867242651588,
      "eval_runtime": 205.1913,
      "eval_samples_per_second": 13.168,
      "eval_steps_per_second": 13.168,
      "step": 3000
    },
    {
      "epoch": 1.4401416765053128,
      "grad_norm": 14.843865394592285,
      "learning_rate": 0.00014377502383222117,
      "loss": 0.4521,
      "step": 3050
    },
    {
      "epoch": 1.463754427390791,
      "grad_norm": 17.06186866760254,
      "learning_rate": 0.00014282173498570067,
      "loss": 0.4178,
      "step": 3100
    },
    {
      "epoch": 1.4873671782762692,
      "grad_norm": 15.016847610473633,
      "learning_rate": 0.00014186844613918017,
      "loss": 0.4553,
      "step": 3150
    },
    {
      "epoch": 1.5109799291617474,
      "grad_norm": 13.980330467224121,
      "learning_rate": 0.00014091515729265967,
      "loss": 0.4719,
      "step": 3200
    },
    {
      "epoch": 1.5109799291617474,
      "eval_accuracy": 0.7964470762398224,
      "eval_auc": 0.8655378505351868,
      "eval_cm_tn_fp_fn_tp": [
        1715,
        162,
        388,
        437
      ],
      "eval_f1": 0.6137640449438202,
      "eval_loss": 0.46954312920570374,
      "eval_prauc": 0.7223225301667069,
      "eval_runtime": 206.0574,
      "eval_samples_per_second": 13.113,
      "eval_steps_per_second": 13.113,
      "step": 3200
    },
    {
      "epoch": 1.5345926800472256,
      "grad_norm": 12.579852104187012,
      "learning_rate": 0.0001399618684461392,
      "loss": 0.4822,
      "step": 3250
    },
    {
      "epoch": 1.5582054309327038,
      "grad_norm": 22.1433048248291,
      "learning_rate": 0.0001390085795996187,
      "loss": 0.4499,
      "step": 3300
    },
    {
      "epoch": 1.5818181818181818,
      "grad_norm": 15.12907886505127,
      "learning_rate": 0.0001380552907530982,
      "loss": 0.4149,
      "step": 3350
    },
    {
      "epoch": 1.60543093270366,
      "grad_norm": 15.873414993286133,
      "learning_rate": 0.00013710200190657768,
      "loss": 0.4451,
      "step": 3400
    },
    {
      "epoch": 1.60543093270366,
      "eval_accuracy": 0.7927461139896373,
      "eval_auc": 0.88108070583297,
      "eval_cm_tn_fp_fn_tp": [
        1431,
        446,
        114,
        711
      ],
      "eval_f1": 0.7174571140262361,
      "eval_loss": 0.5250758528709412,
      "eval_prauc": 0.7317536363311827,
      "eval_runtime": 215.7478,
      "eval_samples_per_second": 12.524,
      "eval_steps_per_second": 12.524,
      "step": 3400
    },
    {
      "epoch": 1.629043683589138,
      "grad_norm": 12.564345359802246,
      "learning_rate": 0.0001361487130600572,
      "loss": 0.4378,
      "step": 3450
    },
    {
      "epoch": 1.6526564344746162,
      "grad_norm": 6.282104015350342,
      "learning_rate": 0.0001351954242135367,
      "loss": 0.4244,
      "step": 3500
    },
    {
      "epoch": 1.6762691853600944,
      "grad_norm": 17.35226058959961,
      "learning_rate": 0.00013424213536701623,
      "loss": 0.4026,
      "step": 3550
    },
    {
      "epoch": 1.6998819362455726,
      "grad_norm": 11.386140823364258,
      "learning_rate": 0.00013328884652049572,
      "loss": 0.4414,
      "step": 3600
    },
    {
      "epoch": 1.6998819362455726,
      "eval_accuracy": 0.8075499629903775,
      "eval_auc": 0.8929855830548424,
      "eval_cm_tn_fp_fn_tp": [
        1440,
        437,
        83,
        742
      ],
      "eval_f1": 0.7405189620758483,
      "eval_loss": 0.5384999513626099,
      "eval_prauc": 0.7435629940711642,
      "eval_runtime": 205.4048,
      "eval_samples_per_second": 13.155,
      "eval_steps_per_second": 13.155,
      "step": 3600
    },
    {
      "epoch": 1.7234946871310508,
      "grad_norm": 13.51180648803711,
      "learning_rate": 0.00013233555767397522,
      "loss": 0.455,
      "step": 3650
    },
    {
      "epoch": 1.747107438016529,
      "grad_norm": 28.12225341796875,
      "learning_rate": 0.00013138226882745472,
      "loss": 0.3575,
      "step": 3700
    },
    {
      "epoch": 1.7707201889020072,
      "grad_norm": 18.93398094177246,
      "learning_rate": 0.00013042897998093422,
      "loss": 0.39,
      "step": 3750
    },
    {
      "epoch": 1.7943329397874852,
      "grad_norm": 9.61694049835205,
      "learning_rate": 0.00012947569113441374,
      "loss": 0.4416,
      "step": 3800
    },
    {
      "epoch": 1.7943329397874852,
      "eval_accuracy": 0.8286454478164322,
      "eval_auc": 0.8747295006538479,
      "eval_cm_tn_fp_fn_tp": [
        1595,
        282,
        181,
        644
      ],
      "eval_f1": 0.7355796687607081,
      "eval_loss": 0.4878920912742615,
      "eval_prauc": 0.7247270995184248,
      "eval_runtime": 205.7666,
      "eval_samples_per_second": 13.131,
      "eval_steps_per_second": 13.131,
      "step": 3800
    },
    {
      "epoch": 1.8179456906729634,
      "grad_norm": 5.901403427124023,
      "learning_rate": 0.00012852240228789324,
      "loss": 0.4254,
      "step": 3850
    },
    {
      "epoch": 1.8415584415584414,
      "grad_norm": 8.924416542053223,
      "learning_rate": 0.00012756911344137273,
      "loss": 0.3969,
      "step": 3900
    },
    {
      "epoch": 1.8651711924439196,
      "grad_norm": 21.536972045898438,
      "learning_rate": 0.00012661582459485223,
      "loss": 0.4095,
      "step": 3950
    },
    {
      "epoch": 1.8887839433293978,
      "grad_norm": 23.68355941772461,
      "learning_rate": 0.00012566253574833176,
      "loss": 0.4698,
      "step": 4000
    },
    {
      "epoch": 1.8887839433293978,
      "eval_accuracy": 0.8201332346410066,
      "eval_auc": 0.8902213396619364,
      "eval_cm_tn_fp_fn_tp": [
        1686,
        191,
        295,
        530
      ],
      "eval_f1": 0.685640362225097,
      "eval_loss": 0.4424063563346863,
      "eval_prauc": 0.7390274557405583,
      "eval_runtime": 205.8289,
      "eval_samples_per_second": 13.127,
      "eval_steps_per_second": 13.127,
      "step": 4000
    },
    {
      "epoch": 1.912396694214876,
      "grad_norm": 8.193197250366211,
      "learning_rate": 0.00012470924690181125,
      "loss": 0.3865,
      "step": 4050
    },
    {
      "epoch": 1.9360094451003542,
      "grad_norm": 29.619186401367188,
      "learning_rate": 0.00012375595805529078,
      "loss": 0.4474,
      "step": 4100
    },
    {
      "epoch": 1.9596221959858324,
      "grad_norm": 5.884322166442871,
      "learning_rate": 0.00012280266920877025,
      "loss": 0.4187,
      "step": 4150
    },
    {
      "epoch": 1.9832349468713106,
      "grad_norm": 24.293256759643555,
      "learning_rate": 0.00012184938036224977,
      "loss": 0.3788,
      "step": 4200
    },
    {
      "epoch": 1.9832349468713106,
      "eval_accuracy": 0.7982975573649149,
      "eval_auc": 0.8759923152677549,
      "eval_cm_tn_fp_fn_tp": [
        1472,
        405,
        140,
        685
      ],
      "eval_f1": 0.7154046997389034,
      "eval_loss": 0.5157374739646912,
      "eval_prauc": 0.7657810063630972,
      "eval_runtime": 205.5207,
      "eval_samples_per_second": 13.147,
      "eval_steps_per_second": 13.147,
      "step": 4200
    },
    {
      "epoch": 2.0066115702479337,
      "grad_norm": 26.29973602294922,
      "learning_rate": 0.00012089609151572927,
      "loss": 0.4213,
      "step": 4250
    },
    {
      "epoch": 2.030224321133412,
      "grad_norm": 14.445331573486328,
      "learning_rate": 0.0001199428026692088,
      "loss": 0.3619,
      "step": 4300
    },
    {
      "epoch": 2.05383707201889,
      "grad_norm": 3.6922478675842285,
      "learning_rate": 0.00011898951382268828,
      "loss": 0.3313,
      "step": 4350
    },
    {
      "epoch": 2.0774498229043683,
      "grad_norm": 10.860296249389648,
      "learning_rate": 0.00011803622497616777,
      "loss": 0.3829,
      "step": 4400
    },
    {
      "epoch": 2.0774498229043683,
      "eval_accuracy": 0.5817912657290896,
      "eval_auc": 0.8061184029963997,
      "eval_cm_tn_fp_fn_tp": [
        837,
        1040,
        90,
        735
      ],
      "eval_f1": 0.5653846153846154,
      "eval_loss": 0.7727663516998291,
      "eval_prauc": 0.683895614914998,
      "eval_runtime": 205.7518,
      "eval_samples_per_second": 13.132,
      "eval_steps_per_second": 13.132,
      "step": 4400
    },
    {
      "epoch": 2.1010625737898465,
      "grad_norm": 8.952884674072266,
      "learning_rate": 0.00011708293612964728,
      "loss": 0.3882,
      "step": 4450
    },
    {
      "epoch": 2.1246753246753247,
      "grad_norm": 12.174997329711914,
      "learning_rate": 0.00011612964728312678,
      "loss": 0.3689,
      "step": 4500
    },
    {
      "epoch": 2.148288075560803,
      "grad_norm": 4.450479984283447,
      "learning_rate": 0.0001151763584366063,
      "loss": 0.358,
      "step": 4550
    },
    {
      "epoch": 2.171900826446281,
      "grad_norm": 12.854631423950195,
      "learning_rate": 0.00011422306959008579,
      "loss": 0.3776,
      "step": 4600
    },
    {
      "epoch": 2.171900826446281,
      "eval_accuracy": 0.7416728349370837,
      "eval_auc": 0.8377194426954683,
      "eval_cm_tn_fp_fn_tp": [
        1378,
        499,
        199,
        626
      ],
      "eval_f1": 0.642051282051282,
      "eval_loss": 0.5387515425682068,
      "eval_prauc": 0.7232880191562454,
      "eval_runtime": 206.0363,
      "eval_samples_per_second": 13.114,
      "eval_steps_per_second": 13.114,
      "step": 4600
    },
    {
      "epoch": 2.1955135773317593,
      "grad_norm": 5.861292839050293,
      "learning_rate": 0.00011326978074356531,
      "loss": 0.3664,
      "step": 4650
    },
    {
      "epoch": 2.2191263282172375,
      "grad_norm": 7.848627090454102,
      "learning_rate": 0.00011231649189704481,
      "loss": 0.365,
      "step": 4700
    },
    {
      "epoch": 2.2427390791027153,
      "grad_norm": 12.896380424499512,
      "learning_rate": 0.00011136320305052432,
      "loss": 0.3786,
      "step": 4750
    },
    {
      "epoch": 2.2663518299881935,
      "grad_norm": 6.248437404632568,
      "learning_rate": 0.00011040991420400382,
      "loss": 0.3626,
      "step": 4800
    },
    {
      "epoch": 2.2663518299881935,
      "eval_accuracy": 0.8279052553663953,
      "eval_auc": 0.8455788572996885,
      "eval_cm_tn_fp_fn_tp": [
        1674,
        203,
        262,
        563
      ],
      "eval_f1": 0.7077309868007542,
      "eval_loss": 0.4897433817386627,
      "eval_prauc": 0.7210946777965951,
      "eval_runtime": 205.9923,
      "eval_samples_per_second": 13.117,
      "eval_steps_per_second": 13.117,
      "step": 4800
    },
    {
      "epoch": 2.2899645808736717,
      "grad_norm": 7.61652135848999,
      "learning_rate": 0.00010945662535748333,
      "loss": 0.3542,
      "step": 4850
    },
    {
      "epoch": 2.31357733175915,
      "grad_norm": 22.21363067626953,
      "learning_rate": 0.00010850333651096283,
      "loss": 0.3865,
      "step": 4900
    },
    {
      "epoch": 2.337190082644628,
      "grad_norm": 25.892425537109375,
      "learning_rate": 0.00010755004766444232,
      "loss": 0.3833,
      "step": 4950
    },
    {
      "epoch": 2.3608028335301063,
      "grad_norm": 21.37782096862793,
      "learning_rate": 0.00010659675881792184,
      "loss": 0.3794,
      "step": 5000
    },
    {
      "epoch": 2.3608028335301063,
      "eval_accuracy": 0.7953367875647669,
      "eval_auc": 0.8713559677757867,
      "eval_cm_tn_fp_fn_tp": [
        1451,
        426,
        127,
        698
      ],
      "eval_f1": 0.7162647511544382,
      "eval_loss": 0.5515428781509399,
      "eval_prauc": 0.7465915400991109,
      "eval_runtime": 205.5994,
      "eval_samples_per_second": 13.142,
      "eval_steps_per_second": 13.142,
      "step": 5000
    },
    {
      "epoch": 2.3844155844155845,
      "grad_norm": 12.759713172912598,
      "learning_rate": 0.00010564346997140133,
      "loss": 0.3997,
      "step": 5050
    },
    {
      "epoch": 2.4080283353010628,
      "grad_norm": 3.6114132404327393,
      "learning_rate": 0.00010469018112488084,
      "loss": 0.3441,
      "step": 5100
    },
    {
      "epoch": 2.4316410861865405,
      "grad_norm": 6.879336357116699,
      "learning_rate": 0.00010373689227836034,
      "loss": 0.3947,
      "step": 5150
    },
    {
      "epoch": 2.4552538370720187,
      "grad_norm": 5.560421466827393,
      "learning_rate": 0.00010278360343183986,
      "loss": 0.3459,
      "step": 5200
    },
    {
      "epoch": 2.4552538370720187,
      "eval_accuracy": 0.6565507031828275,
      "eval_auc": 0.8567033144443906,
      "eval_cm_tn_fp_fn_tp": [
        1046,
        831,
        97,
        728
      ],
      "eval_f1": 0.610738255033557,
      "eval_loss": 0.6395134925842285,
      "eval_prauc": 0.7554881183404748,
      "eval_runtime": 205.5903,
      "eval_samples_per_second": 13.143,
      "eval_steps_per_second": 13.143,
      "step": 5200
    },
    {
      "epoch": 2.478866587957497,
      "grad_norm": 10.426353454589844,
      "learning_rate": 0.00010183031458531935,
      "loss": 0.3801,
      "step": 5250
    },
    {
      "epoch": 2.502479338842975,
      "grad_norm": 8.82383918762207,
      "learning_rate": 0.00010087702573879887,
      "loss": 0.3337,
      "step": 5300
    },
    {
      "epoch": 2.5260920897284533,
      "grad_norm": 8.333550453186035,
      "learning_rate": 9.992373689227837e-05,
      "loss": 0.3902,
      "step": 5350
    },
    {
      "epoch": 2.5497048406139315,
      "grad_norm": 6.590660572052002,
      "learning_rate": 9.897044804575787e-05,
      "loss": 0.3751,
      "step": 5400
    },
    {
      "epoch": 2.5497048406139315,
      "eval_accuracy": 0.8323464100666174,
      "eval_auc": 0.8878594146042201,
      "eval_cm_tn_fp_fn_tp": [
        1575,
        302,
        151,
        674
      ],
      "eval_f1": 0.7484730705163798,
      "eval_loss": 0.47527438402175903,
      "eval_prauc": 0.7917505188586466,
      "eval_runtime": 205.3611,
      "eval_samples_per_second": 13.157,
      "eval_steps_per_second": 13.157,
      "step": 5400
    },
    {
      "epoch": 2.5733175914994098,
      "grad_norm": 19.815631866455078,
      "learning_rate": 9.801715919923738e-05,
      "loss": 0.3209,
      "step": 5450
    },
    {
      "epoch": 2.596930342384888,
      "grad_norm": 12.452373504638672,
      "learning_rate": 9.706387035271687e-05,
      "loss": 0.3154,
      "step": 5500
    },
    {
      "epoch": 2.620543093270366,
      "grad_norm": 3.750296115875244,
      "learning_rate": 9.611058150619639e-05,
      "loss": 0.3763,
      "step": 5550
    },
    {
      "epoch": 2.6441558441558444,
      "grad_norm": 23.37812042236328,
      "learning_rate": 9.51572926596759e-05,
      "loss": 0.364,
      "step": 5600
    },
    {
      "epoch": 2.6441558441558444,
      "eval_accuracy": 0.6032568467801629,
      "eval_auc": 0.8076388821620575,
      "eval_cm_tn_fp_fn_tp": [
        932,
        945,
        127,
        698
      ],
      "eval_f1": 0.5656401944894651,
      "eval_loss": 0.6932260394096375,
      "eval_prauc": 0.7096068675591821,
      "eval_runtime": 205.4319,
      "eval_samples_per_second": 13.153,
      "eval_steps_per_second": 13.153,
      "step": 5600
    },
    {
      "epoch": 2.6677685950413226,
      "grad_norm": 22.82330894470215,
      "learning_rate": 9.420400381315539e-05,
      "loss": 0.3923,
      "step": 5650
    },
    {
      "epoch": 2.6913813459268003,
      "grad_norm": 16.2337703704834,
      "learning_rate": 9.32507149666349e-05,
      "loss": 0.3306,
      "step": 5700
    },
    {
      "epoch": 2.7149940968122785,
      "grad_norm": 7.920469284057617,
      "learning_rate": 9.229742612011439e-05,
      "loss": 0.3558,
      "step": 5750
    },
    {
      "epoch": 2.7386068476977568,
      "grad_norm": 4.530557632446289,
      "learning_rate": 9.13441372735939e-05,
      "loss": 0.3215,
      "step": 5800
    },
    {
      "epoch": 2.7386068476977568,
      "eval_accuracy": 0.8497409326424871,
      "eval_auc": 0.917995188970149,
      "eval_cm_tn_fp_fn_tp": [
        1621,
        256,
        150,
        675
      ],
      "eval_f1": 0.7687927107061503,
      "eval_loss": 0.4086146354675293,
      "eval_prauc": 0.8212032649082162,
      "eval_runtime": 205.5797,
      "eval_samples_per_second": 13.143,
      "eval_steps_per_second": 13.143,
      "step": 5800
    },
    {
      "epoch": 2.762219598583235,
      "grad_norm": 17.762514114379883,
      "learning_rate": 9.039084842707341e-05,
      "loss": 0.3463,
      "step": 5850
    },
    {
      "epoch": 2.785832349468713,
      "grad_norm": 12.965508460998535,
      "learning_rate": 8.94375595805529e-05,
      "loss": 0.3463,
      "step": 5900
    },
    {
      "epoch": 2.8094451003541914,
      "grad_norm": 5.069027423858643,
      "learning_rate": 8.848427073403242e-05,
      "loss": 0.2931,
      "step": 5950
    },
    {
      "epoch": 2.8330578512396696,
      "grad_norm": 14.7559814453125,
      "learning_rate": 8.753098188751191e-05,
      "loss": 0.3302,
      "step": 6000
    },
    {
      "epoch": 2.8330578512396696,
      "eval_accuracy": 0.7339008142116951,
      "eval_auc": 0.8722855620671284,
      "eval_cm_tn_fp_fn_tp": [
        1287,
        590,
        129,
        696
      ],
      "eval_f1": 0.659403126480341,
      "eval_loss": 0.5873863697052002,
      "eval_prauc": 0.7796160497762147,
      "eval_runtime": 205.1472,
      "eval_samples_per_second": 13.171,
      "eval_steps_per_second": 13.171,
      "step": 6000
    },
    {
      "epoch": 2.8566706021251473,
      "grad_norm": 8.567463874816895,
      "learning_rate": 8.657769304099142e-05,
      "loss": 0.3125,
      "step": 6050
    },
    {
      "epoch": 2.8802833530106255,
      "grad_norm": 21.14914321899414,
      "learning_rate": 8.562440419447094e-05,
      "loss": 0.3503,
      "step": 6100
    },
    {
      "epoch": 2.9038961038961038,
      "grad_norm": 12.390323638916016,
      "learning_rate": 8.467111534795043e-05,
      "loss": 0.3415,
      "step": 6150
    },
    {
      "epoch": 2.927508854781582,
      "grad_norm": 5.121053695678711,
      "learning_rate": 8.371782650142994e-05,
      "loss": 0.3266,
      "step": 6200
    },
    {
      "epoch": 2.927508854781582,
      "eval_accuracy": 0.8412287194670615,
      "eval_auc": 0.9269827739300303,
      "eval_cm_tn_fp_fn_tp": [
        1639,
        238,
        191,
        634
      ],
      "eval_f1": 0.7472009428403065,
      "eval_loss": 0.384371817111969,
      "eval_prauc": 0.8382389930976012,
      "eval_runtime": 205.3874,
      "eval_samples_per_second": 13.156,
      "eval_steps_per_second": 13.156,
      "step": 6200
    },
    {
      "epoch": 2.95112160566706,
      "grad_norm": 18.484142303466797,
      "learning_rate": 8.276453765490944e-05,
      "loss": 0.3488,
      "step": 6250
    },
    {
      "epoch": 2.9747343565525384,
      "grad_norm": 37.039432525634766,
      "learning_rate": 8.181124880838894e-05,
      "loss": 0.3479,
      "step": 6300
    },
    {
      "epoch": 2.9983471074380166,
      "grad_norm": 20.839073181152344,
      "learning_rate": 8.085795996186845e-05,
      "loss": 0.3464,
      "step": 6350
    },
    {
      "epoch": 3.02172373081464,
      "grad_norm": 9.81174373626709,
      "learning_rate": 7.990467111534795e-05,
      "loss": 0.2552,
      "step": 6400
    },
    {
      "epoch": 3.02172373081464,
      "eval_accuracy": 0.8290155440414507,
      "eval_auc": 0.8928280137550249,
      "eval_cm_tn_fp_fn_tp": [
        1508,
        369,
        93,
        732
      ],
      "eval_f1": 0.7601246105919003,
      "eval_loss": 0.5704883933067322,
      "eval_prauc": 0.7964650981389155,
      "eval_runtime": 205.4254,
      "eval_samples_per_second": 13.153,
      "eval_steps_per_second": 13.153,
      "step": 6400
    },
    {
      "epoch": 3.0453364817001183,
      "grad_norm": 14.28138542175293,
      "learning_rate": 7.895138226882746e-05,
      "loss": 0.2651,
      "step": 6450
    },
    {
      "epoch": 3.068949232585596,
      "grad_norm": 15.993682861328125,
      "learning_rate": 7.799809342230697e-05,
      "loss": 0.2934,
      "step": 6500
    },
    {
      "epoch": 3.0925619834710742,
      "grad_norm": 7.407653331756592,
      "learning_rate": 7.704480457578646e-05,
      "loss": 0.2698,
      "step": 6550
    },
    {
      "epoch": 3.1161747343565525,
      "grad_norm": 2.8148610591888428,
      "learning_rate": 7.609151572926597e-05,
      "loss": 0.2951,
      "step": 6600
    },
    {
      "epoch": 3.1161747343565525,
      "eval_accuracy": 0.8016284233900814,
      "eval_auc": 0.868587203952148,
      "eval_cm_tn_fp_fn_tp": [
        1473,
        404,
        132,
        693
      ],
      "eval_f1": 0.7211238293444329,
      "eval_loss": 0.542561948299408,
      "eval_prauc": 0.7605646587301521,
      "eval_runtime": 205.5379,
      "eval_samples_per_second": 13.146,
      "eval_steps_per_second": 13.146,
      "step": 6600
    },
    {
      "epoch": 3.1397874852420307,
      "grad_norm": 8.826736450195312,
      "learning_rate": 7.513822688274547e-05,
      "loss": 0.3013,
      "step": 6650
    },
    {
      "epoch": 3.163400236127509,
      "grad_norm": 6.963262557983398,
      "learning_rate": 7.418493803622498e-05,
      "loss": 0.3069,
      "step": 6700
    },
    {
      "epoch": 3.187012987012987,
      "grad_norm": 4.87594747543335,
      "learning_rate": 7.32316491897045e-05,
      "loss": 0.3033,
      "step": 6750
    },
    {
      "epoch": 3.2106257378984653,
      "grad_norm": 5.288126468658447,
      "learning_rate": 7.227836034318399e-05,
      "loss": 0.2986,
      "step": 6800
    },
    {
      "epoch": 3.2106257378984653,
      "eval_accuracy": 0.8123612139156181,
      "eval_auc": 0.8975941621865969,
      "eval_cm_tn_fp_fn_tp": [
        1458,
        419,
        88,
        737
      ],
      "eval_f1": 0.7440686521958607,
      "eval_loss": 0.5542622208595276,
      "eval_prauc": 0.7907006312220203,
      "eval_runtime": 207.6065,
      "eval_samples_per_second": 13.015,
      "eval_steps_per_second": 13.015,
      "step": 6800
    },
    {
      "epoch": 3.2342384887839435,
      "grad_norm": 14.439066886901855,
      "learning_rate": 7.132507149666349e-05,
      "loss": 0.2635,
      "step": 6850
    },
    {
      "epoch": 3.2578512396694217,
      "grad_norm": 14.048096656799316,
      "learning_rate": 7.0371782650143e-05,
      "loss": 0.3471,
      "step": 6900
    },
    {
      "epoch": 3.2814639905548995,
      "grad_norm": 6.724935054779053,
      "learning_rate": 6.94184938036225e-05,
      "loss": 0.3099,
      "step": 6950
    },
    {
      "epoch": 3.3050767414403777,
      "grad_norm": 12.807244300842285,
      "learning_rate": 6.8465204957102e-05,
      "loss": 0.305,
      "step": 7000
    },
    {
      "epoch": 3.3050767414403777,
      "eval_accuracy": 0.7834937083641746,
      "eval_auc": 0.8916016854748874,
      "eval_cm_tn_fp_fn_tp": [
        1390,
        487,
        98,
        727
      ],
      "eval_f1": 0.7130946542422756,
      "eval_loss": 0.5494717359542847,
      "eval_prauc": 0.8068916778086699,
      "eval_runtime": 205.3346,
      "eval_samples_per_second": 13.159,
      "eval_steps_per_second": 13.159,
      "step": 7000
    },
    {
      "epoch": 3.328689492325856,
      "grad_norm": 8.32547664642334,
      "learning_rate": 6.75119161105815e-05,
      "loss": 0.2779,
      "step": 7050
    },
    {
      "epoch": 3.352302243211334,
      "grad_norm": 16.320039749145508,
      "learning_rate": 6.655862726406101e-05,
      "loss": 0.2863,
      "step": 7100
    },
    {
      "epoch": 3.3759149940968123,
      "grad_norm": 24.96271324157715,
      "learning_rate": 6.560533841754052e-05,
      "loss": 0.2718,
      "step": 7150
    },
    {
      "epoch": 3.3995277449822905,
      "grad_norm": 13.728142738342285,
      "learning_rate": 6.465204957102002e-05,
      "loss": 0.274,
      "step": 7200
    },
    {
      "epoch": 3.3995277449822905,
      "eval_accuracy": 0.8341968911917098,
      "eval_auc": 0.889711176764986,
      "eval_cm_tn_fp_fn_tp": [
        1607,
        270,
        178,
        647
      ],
      "eval_f1": 0.7428243398392652,
      "eval_loss": 0.46856218576431274,
      "eval_prauc": 0.7844383258552488,
      "eval_runtime": 204.9634,
      "eval_samples_per_second": 13.183,
      "eval_steps_per_second": 13.183,
      "step": 7200
    },
    {
      "epoch": 3.4231404958677687,
      "grad_norm": 13.499300956726074,
      "learning_rate": 6.369876072449953e-05,
      "loss": 0.2856,
      "step": 7250
    },
    {
      "epoch": 3.446753246753247,
      "grad_norm": 10.063815116882324,
      "learning_rate": 6.274547187797903e-05,
      "loss": 0.2559,
      "step": 7300
    },
    {
      "epoch": 3.4703659976387247,
      "grad_norm": 7.386045932769775,
      "learning_rate": 6.179218303145854e-05,
      "loss": 0.2955,
      "step": 7350
    },
    {
      "epoch": 3.493978748524203,
      "grad_norm": 9.75947380065918,
      "learning_rate": 6.0838894184938045e-05,
      "loss": 0.2836,
      "step": 7400
    },
    {
      "epoch": 3.493978748524203,
      "eval_accuracy": 0.8623242042931162,
      "eval_auc": 0.9092749551993026,
      "eval_cm_tn_fp_fn_tp": [
        1641,
        236,
        136,
        689
      ],
      "eval_f1": 0.7874285714285715,
      "eval_loss": 0.4439694583415985,
      "eval_prauc": 0.8166282575862323,
      "eval_runtime": 205.4193,
      "eval_samples_per_second": 13.154,
      "eval_steps_per_second": 13.154,
      "step": 7400
    },
    {
      "epoch": 3.517591499409681,
      "grad_norm": 9.995803833007812,
      "learning_rate": 5.988560533841754e-05,
      "loss": 0.2545,
      "step": 7450
    },
    {
      "epoch": 3.5412042502951593,
      "grad_norm": 8.205814361572266,
      "learning_rate": 5.8932316491897046e-05,
      "loss": 0.2634,
      "step": 7500
    },
    {
      "epoch": 3.5648170011806375,
      "grad_norm": 31.923660278320312,
      "learning_rate": 5.797902764537655e-05,
      "loss": 0.3174,
      "step": 7550
    },
    {
      "epoch": 3.5884297520661157,
      "grad_norm": 8.342251777648926,
      "learning_rate": 5.7025738798856054e-05,
      "loss": 0.2858,
      "step": 7600
    },
    {
      "epoch": 3.5884297520661157,
      "eval_accuracy": 0.8689859363434493,
      "eval_auc": 0.9157908332122504,
      "eval_cm_tn_fp_fn_tp": [
        1707,
        170,
        184,
        641
      ],
      "eval_f1": 0.7836185819070904,
      "eval_loss": 0.3948521614074707,
      "eval_prauc": 0.8201110541924611,
      "eval_runtime": 205.2094,
      "eval_samples_per_second": 13.167,
      "eval_steps_per_second": 13.167,
      "step": 7600
    }
  ],
  "logging_steps": 50,
  "max_steps": 10590,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}