{
  "best_metric": 0.2718897759914398,
  "best_model_checkpoint": "./vit-large-binary-isic-patch-16/checkpoint-2100",
  "epoch": 4.0,
  "global_step": 4304,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.0001995353159851301,
      "loss": 0.6972,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019907063197026022,
      "loss": 0.5517,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019860594795539035,
      "loss": 0.4475,
      "step": 30
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019814126394052047,
      "loss": 0.4486,
      "step": 40
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019767657992565056,
      "loss": 0.3632,
      "step": 50
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019721189591078068,
      "loss": 0.6664,
      "step": 60
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019679368029739777,
      "loss": 0.7834,
      "step": 70
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001963289962825279,
      "loss": 0.6194,
      "step": 80
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019586431226765801,
      "loss": 0.6248,
      "step": 90
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001953996282527881,
      "loss": 0.5596,
      "step": 100
    },
    {
      "epoch": 0.09,
      "eval_accuracy": 0.7780625320348539,
      "eval_f1": 0.7780625320348539,
      "eval_loss": 0.4727737307548523,
      "eval_precision": 0.7780625320348539,
      "eval_recall": 0.7780625320348539,
      "eval_runtime": 82.6706,
      "eval_samples_per_second": 94.399,
      "eval_steps_per_second": 11.806,
      "step": 100
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019493494423791823,
      "loss": 0.5283,
      "step": 110
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019447026022304832,
      "loss": 0.5472,
      "step": 120
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019400557620817844,
      "loss": 0.4244,
      "step": 130
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019354089219330857,
      "loss": 0.5444,
      "step": 140
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019307620817843869,
      "loss": 0.3566,
      "step": 150
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019261152416356878,
      "loss": 0.3662,
      "step": 160
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001921468401486989,
      "loss": 0.6231,
      "step": 170
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000191682156133829,
      "loss": 0.3903,
      "step": 180
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019121747211895912,
      "loss": 0.3428,
      "step": 190
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019075278810408924,
      "loss": 0.3373,
      "step": 200
    },
    {
      "epoch": 0.19,
      "eval_accuracy": 0.8266273705791901,
      "eval_f1": 0.8266273705791901,
      "eval_loss": 0.3593936562538147,
      "eval_precision": 0.8266273705791901,
      "eval_recall": 0.8266273705791901,
      "eval_runtime": 83.3668,
      "eval_samples_per_second": 93.61,
      "eval_steps_per_second": 11.707,
      "step": 200
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019028810408921933,
      "loss": 0.3686,
      "step": 210
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018982342007434945,
      "loss": 0.4349,
      "step": 220
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018935873605947955,
      "loss": 0.3641,
      "step": 230
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001888940520446097,
      "loss": 0.3376,
      "step": 240
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001884293680297398,
      "loss": 0.4286,
      "step": 250
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001879646840148699,
      "loss": 0.3639,
      "step": 260
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001875,
      "loss": 0.3894,
      "step": 270
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001870353159851301,
      "loss": 0.3926,
      "step": 280
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00018657063197026025,
      "loss": 0.4166,
      "step": 290
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00018610594795539034,
      "loss": 0.397,
      "step": 300
    },
    {
      "epoch": 0.28,
      "eval_accuracy": 0.7694771911840081,
      "eval_f1": 0.7694771911840081,
      "eval_loss": 0.5284339785575867,
      "eval_precision": 0.7694771911840081,
      "eval_recall": 0.7694771911840081,
      "eval_runtime": 83.3061,
      "eval_samples_per_second": 93.679,
      "eval_steps_per_second": 11.716,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018564126394052046,
      "loss": 0.343,
      "step": 310
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018517657992565056,
      "loss": 0.4755,
      "step": 320
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018471189591078068,
      "loss": 0.4006,
      "step": 330
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001842472118959108,
      "loss": 0.3504,
      "step": 340
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018378252788104092,
      "loss": 0.3963,
      "step": 350
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000183317843866171,
      "loss": 0.4396,
      "step": 360
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001828531598513011,
      "loss": 0.436,
      "step": 370
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018238847583643123,
      "loss": 0.3115,
      "step": 380
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018192379182156135,
      "loss": 0.3645,
      "step": 390
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018145910780669147,
      "loss": 0.3913,
      "step": 400
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.8384161968221425,
      "eval_f1": 0.8384161968221425,
      "eval_loss": 0.3314719498157501,
      "eval_precision": 0.8384161968221425,
      "eval_recall": 0.8384161968221425,
      "eval_runtime": 84.5555,
      "eval_samples_per_second": 92.294,
      "eval_steps_per_second": 11.543,
      "step": 400
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018099442379182156,
      "loss": 0.3651,
      "step": 410
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018052973977695168,
      "loss": 0.3856,
      "step": 420
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00018006505576208178,
      "loss": 0.3303,
      "step": 430
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00017960037174721193,
      "loss": 0.2943,
      "step": 440
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017913568773234202,
      "loss": 0.4615,
      "step": 450
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00017867100371747211,
      "loss": 0.2892,
      "step": 460
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00017820631970260224,
      "loss": 0.3111,
      "step": 470
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017774163568773236,
      "loss": 0.36,
      "step": 480
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017727695167286248,
      "loss": 0.2742,
      "step": 490
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017681226765799257,
      "loss": 0.3147,
      "step": 500
    },
    {
      "epoch": 0.46,
      "eval_accuracy": 0.7778062532034854,
      "eval_f1": 0.7778062532034853,
      "eval_loss": 0.4424784183502197,
      "eval_precision": 0.7778062532034854,
      "eval_recall": 0.7778062532034854,
      "eval_runtime": 82.8072,
      "eval_samples_per_second": 94.243,
      "eval_steps_per_second": 11.786,
      "step": 500
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001763475836431227,
      "loss": 0.2909,
      "step": 510
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001758828996282528,
      "loss": 0.3201,
      "step": 520
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001754182156133829,
      "loss": 0.3369,
      "step": 530
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00017495353159851303,
      "loss": 0.2276,
      "step": 540
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017448884758364312,
      "loss": 0.3965,
      "step": 550
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017402416356877324,
      "loss": 0.2972,
      "step": 560
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017355947955390334,
      "loss": 0.3601,
      "step": 570
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017309479553903346,
      "loss": 0.2637,
      "step": 580
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017263011152416358,
      "loss": 0.3529,
      "step": 590
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001721654275092937,
      "loss": 0.2709,
      "step": 600
    },
    {
      "epoch": 0.56,
      "eval_accuracy": 0.8352127114300358,
      "eval_f1": 0.8352127114300358,
      "eval_loss": 0.37867271900177,
      "eval_precision": 0.8352127114300358,
      "eval_recall": 0.8352127114300358,
      "eval_runtime": 85.1389,
      "eval_samples_per_second": 91.662,
      "eval_steps_per_second": 11.464,
      "step": 600
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001717007434944238,
      "loss": 0.2653,
      "step": 610
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001712360594795539,
      "loss": 0.3102,
      "step": 620
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017077137546468404,
      "loss": 0.2868,
      "step": 630
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017030669144981413,
      "loss": 0.323,
      "step": 640
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00016984200743494425,
      "loss": 0.2363,
      "step": 650
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016937732342007435,
      "loss": 0.391,
      "step": 660
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016891263940520447,
      "loss": 0.2941,
      "step": 670
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001684479553903346,
      "loss": 0.3239,
      "step": 680
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001679832713754647,
      "loss": 0.337,
      "step": 690
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001675185873605948,
      "loss": 0.4062,
      "step": 700
    },
    {
      "epoch": 0.65,
      "eval_accuracy": 0.8193234238851871,
      "eval_f1": 0.8193234238851872,
      "eval_loss": 0.36127969622612,
      "eval_precision": 0.8193234238851871,
      "eval_recall": 0.8193234238851871,
      "eval_runtime": 82.9688,
      "eval_samples_per_second": 94.059,
      "eval_steps_per_second": 11.763,
      "step": 700
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001670539033457249,
      "loss": 0.3751,
      "step": 710
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00016658921933085502,
      "loss": 0.2838,
      "step": 720
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016612453531598514,
      "loss": 0.3701,
      "step": 730
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016565985130111526,
      "loss": 0.5011,
      "step": 740
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00016519516728624536,
      "loss": 0.2759,
      "step": 750
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016473048327137548,
      "loss": 0.1873,
      "step": 760
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016426579925650557,
      "loss": 0.3489,
      "step": 770
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016380111524163572,
      "loss": 0.3128,
      "step": 780
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001633364312267658,
      "loss": 0.3321,
      "step": 790
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001628717472118959,
      "loss": 0.3047,
      "step": 800
    },
    {
      "epoch": 0.74,
      "eval_accuracy": 0.8480266529984624,
      "eval_f1": 0.8480266529984622,
      "eval_loss": 0.3086196482181549,
      "eval_precision": 0.8480266529984624,
      "eval_recall": 0.8480266529984624,
      "eval_runtime": 83.073,
      "eval_samples_per_second": 93.942,
      "eval_steps_per_second": 11.749,
      "step": 800
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016240706319702603,
      "loss": 0.2481,
      "step": 810
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016194237918215612,
      "loss": 0.4429,
      "step": 820
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016147769516728627,
      "loss": 0.3739,
      "step": 830
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00016101301115241636,
      "loss": 0.3238,
      "step": 840
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016054832713754648,
      "loss": 0.2907,
      "step": 850
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00016008364312267658,
      "loss": 0.2156,
      "step": 860
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001596189591078067,
      "loss": 0.3718,
      "step": 870
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015915427509293682,
      "loss": 0.3116,
      "step": 880
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00015868959107806691,
      "loss": 0.3216,
      "step": 890
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015822490706319704,
      "loss": 0.3542,
      "step": 900
    },
    {
      "epoch": 0.84,
      "eval_accuracy": 0.8619938493080471,
      "eval_f1": 0.8619938493080471,
      "eval_loss": 0.32323819398880005,
      "eval_precision": 0.8619938493080471,
      "eval_recall": 0.8619938493080471,
      "eval_runtime": 83.4354,
      "eval_samples_per_second": 93.533,
      "eval_steps_per_second": 11.698,
      "step": 900
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015776022304832713,
      "loss": 0.2311,
      "step": 910
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015729553903345725,
      "loss": 0.2891,
      "step": 920
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015683085501858737,
      "loss": 0.2286,
      "step": 930
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001563661710037175,
      "loss": 0.3637,
      "step": 940
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001559014869888476,
      "loss": 0.4039,
      "step": 950
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001554368029739777,
      "loss": 0.3959,
      "step": 960
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001549721189591078,
      "loss": 0.2257,
      "step": 970
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015450743494423792,
      "loss": 0.3245,
      "step": 980
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015404275092936804,
      "loss": 0.3313,
      "step": 990
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015357806691449814,
      "loss": 0.2096,
      "step": 1000
    },
    {
      "epoch": 0.93,
      "eval_accuracy": 0.8733982573039467,
      "eval_f1": 0.8733982573039467,
      "eval_loss": 0.2980605363845825,
      "eval_precision": 0.8733982573039467,
      "eval_recall": 0.8733982573039467,
      "eval_runtime": 82.7075,
      "eval_samples_per_second": 94.357,
      "eval_steps_per_second": 11.801,
      "step": 1000
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015311338289962826,
      "loss": 0.2437,
      "step": 1010
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015264869888475838,
      "loss": 0.2107,
      "step": 1020
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001521840148698885,
      "loss": 0.2582,
      "step": 1030
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001517193308550186,
      "loss": 0.2997,
      "step": 1040
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015125464684014872,
      "loss": 0.29,
      "step": 1050
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001507899628252788,
      "loss": 0.2528,
      "step": 1060
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015032527881040893,
      "loss": 0.2573,
      "step": 1070
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00014986059479553905,
      "loss": 0.2662,
      "step": 1080
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00014939591078066915,
      "loss": 0.3553,
      "step": 1090
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014893122676579927,
      "loss": 0.2214,
      "step": 1100
    },
    {
      "epoch": 1.02,
      "eval_accuracy": 0.8622501281394157,
      "eval_f1": 0.8622501281394157,
      "eval_loss": 0.3147830665111542,
      "eval_precision": 0.8622501281394157,
      "eval_recall": 0.8622501281394157,
      "eval_runtime": 83.1093,
      "eval_samples_per_second": 93.9,
      "eval_steps_per_second": 11.744,
      "step": 1100
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00014846654275092936,
      "loss": 0.2162,
      "step": 1110
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00014800185873605948,
      "loss": 0.2749,
      "step": 1120
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001475371747211896,
      "loss": 0.2001,
      "step": 1130
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001470724907063197,
      "loss": 0.2367,
      "step": 1140
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00014660780669144982,
      "loss": 0.2775,
      "step": 1150
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001461431226765799,
      "loss": 0.2673,
      "step": 1160
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014567843866171006,
      "loss": 0.2109,
      "step": 1170
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00014521375464684016,
      "loss": 0.2578,
      "step": 1180
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014474907063197028,
      "loss": 0.2407,
      "step": 1190
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014428438661710037,
      "loss": 0.2646,
      "step": 1200
    },
    {
      "epoch": 1.12,
      "eval_accuracy": 0.8591747821629934,
      "eval_f1": 0.8591747821629934,
      "eval_loss": 0.31926098465919495,
      "eval_precision": 0.8591747821629934,
      "eval_recall": 0.8591747821629934,
      "eval_runtime": 83.2933,
      "eval_samples_per_second": 93.693,
      "eval_steps_per_second": 11.718,
      "step": 1200
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001438197026022305,
      "loss": 0.2858,
      "step": 1210
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001433550185873606,
      "loss": 0.2316,
      "step": 1220
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001428903345724907,
      "loss": 0.1852,
      "step": 1230
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014242565055762083,
      "loss": 0.3961,
      "step": 1240
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014196096654275092,
      "loss": 0.3315,
      "step": 1250
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00014149628252788104,
      "loss": 0.2146,
      "step": 1260
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014103159851301116,
      "loss": 0.2684,
      "step": 1270
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014056691449814129,
      "loss": 0.2854,
      "step": 1280
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00014010223048327138,
      "loss": 0.249,
      "step": 1290
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001396375464684015,
      "loss": 0.2464,
      "step": 1300
    },
    {
      "epoch": 1.21,
      "eval_accuracy": 0.8347001537672988,
      "eval_f1": 0.8347001537672987,
      "eval_loss": 0.43240851163864136,
      "eval_precision": 0.8347001537672988,
      "eval_recall": 0.8347001537672988,
      "eval_runtime": 83.0735,
      "eval_samples_per_second": 93.941,
      "eval_steps_per_second": 11.749,
      "step": 1300
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001391728624535316,
      "loss": 0.2569,
      "step": 1310
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013870817843866172,
      "loss": 0.2072,
      "step": 1320
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013824349442379184,
      "loss": 0.2483,
      "step": 1330
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00013777881040892193,
      "loss": 0.2149,
      "step": 1340
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00013731412639405205,
      "loss": 0.2426,
      "step": 1350
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013684944237918215,
      "loss": 0.2066,
      "step": 1360
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001363847583643123,
      "loss": 0.2625,
      "step": 1370
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001359200743494424,
      "loss": 0.2567,
      "step": 1380
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001354553903345725,
      "loss": 0.2833,
      "step": 1390
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001349907063197026,
      "loss": 0.2769,
      "step": 1400
    },
    {
      "epoch": 1.3,
      "eval_accuracy": 0.871604305484367,
      "eval_f1": 0.871604305484367,
      "eval_loss": 0.28323572874069214,
      "eval_precision": 0.871604305484367,
      "eval_recall": 0.871604305484367,
      "eval_runtime": 82.6504,
      "eval_samples_per_second": 94.422,
      "eval_steps_per_second": 11.809,
      "step": 1400
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013452602230483272,
      "loss": 0.2412,
      "step": 1410
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013406133828996284,
      "loss": 0.1847,
      "step": 1420
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013359665427509294,
      "loss": 0.2542,
      "step": 1430
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00013313197026022306,
      "loss": 0.2447,
      "step": 1440
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013266728624535315,
      "loss": 0.2718,
      "step": 1450
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013220260223048327,
      "loss": 0.1971,
      "step": 1460
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001317379182156134,
      "loss": 0.2202,
      "step": 1470
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013127323420074352,
      "loss": 0.2028,
      "step": 1480
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001308085501858736,
      "loss": 0.1725,
      "step": 1490
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001303438661710037,
      "loss": 0.2726,
      "step": 1500
    },
    {
      "epoch": 1.39,
      "eval_accuracy": 0.8704510507432086,
      "eval_f1": 0.8704510507432086,
      "eval_loss": 0.28377220034599304,
      "eval_precision": 0.8704510507432086,
      "eval_recall": 0.8704510507432086,
      "eval_runtime": 83.2434,
      "eval_samples_per_second": 93.749,
      "eval_steps_per_second": 11.725,
      "step": 1500
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00012987918215613383,
      "loss": 0.2709,
      "step": 1510
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012941449814126395,
      "loss": 0.2107,
      "step": 1520
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012894981412639407,
      "loss": 0.1519,
      "step": 1530
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012848513011152416,
      "loss": 0.2049,
      "step": 1540
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00012802044609665428,
      "loss": 0.3251,
      "step": 1550
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001275557620817844,
      "loss": 0.2611,
      "step": 1560
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00012709107806691453,
      "loss": 0.2678,
      "step": 1570
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00012662639405204462,
      "loss": 0.2036,
      "step": 1580
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001261617100371747,
      "loss": 0.1938,
      "step": 1590
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012569702602230483,
      "loss": 0.3334,
      "step": 1600
    },
    {
      "epoch": 1.49,
      "eval_accuracy": 0.8494361865709892,
      "eval_f1": 0.8494361865709892,
      "eval_loss": 0.32918739318847656,
      "eval_precision": 0.8494361865709892,
      "eval_recall": 0.8494361865709892,
      "eval_runtime": 81.5591,
      "eval_samples_per_second": 95.685,
      "eval_steps_per_second": 11.967,
      "step": 1600
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012523234200743496,
      "loss": 0.2787,
      "step": 1610
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012476765799256508,
      "loss": 0.2801,
      "step": 1620
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012430297397769517,
      "loss": 0.3583,
      "step": 1630
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001238382899628253,
      "loss": 0.2256,
      "step": 1640
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012337360594795539,
      "loss": 0.2427,
      "step": 1650
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001229089219330855,
      "loss": 0.3166,
      "step": 1660
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012244423791821563,
      "loss": 0.1891,
      "step": 1670
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012197955390334572,
      "loss": 0.2569,
      "step": 1680
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012151486988847584,
      "loss": 0.1854,
      "step": 1690
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012105018587360595,
      "loss": 0.2172,
      "step": 1700
    },
    {
      "epoch": 1.58,
      "eval_accuracy": 0.8635315222962583,
      "eval_f1": 0.8635315222962583,
      "eval_loss": 0.30229443311691284,
      "eval_precision": 0.8635315222962583,
      "eval_recall": 0.8635315222962583,
      "eval_runtime": 81.5723,
      "eval_samples_per_second": 95.67,
      "eval_steps_per_second": 11.965,
      "step": 1700
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012058550185873607,
      "loss": 0.1757,
      "step": 1710
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00012012081784386618,
      "loss": 0.241,
      "step": 1720
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001196561338289963,
      "loss": 0.2434,
      "step": 1730
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001191914498141264,
      "loss": 0.1893,
      "step": 1740
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0001187267657992565,
      "loss": 0.3023,
      "step": 1750
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011826208178438662,
      "loss": 0.2395,
      "step": 1760
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011779739776951673,
      "loss": 0.2023,
      "step": 1770
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011733271375464685,
      "loss": 0.2144,
      "step": 1780
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011686802973977695,
      "loss": 0.2172,
      "step": 1790
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011640334572490708,
      "loss": 0.2382,
      "step": 1800
    },
    {
      "epoch": 1.67,
      "eval_accuracy": 0.8513582778062532,
      "eval_f1": 0.8513582778062531,
      "eval_loss": 0.3191060721874237,
      "eval_precision": 0.8513582778062532,
      "eval_recall": 0.8513582778062532,
      "eval_runtime": 81.5172,
      "eval_samples_per_second": 95.734,
      "eval_steps_per_second": 11.973,
      "step": 1800
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011593866171003717,
      "loss": 0.2222,
      "step": 1810
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0001154739776951673,
      "loss": 0.2171,
      "step": 1820
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001150092936802974,
      "loss": 0.2281,
      "step": 1830
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011454460966542751,
      "loss": 0.2809,
      "step": 1840
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011407992565055763,
      "loss": 0.1798,
      "step": 1850
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011361524163568773,
      "loss": 0.3181,
      "step": 1860
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011315055762081786,
      "loss": 0.2329,
      "step": 1870
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011268587360594795,
      "loss": 0.243,
      "step": 1880
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011222118959107807,
      "loss": 0.2187,
      "step": 1890
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011175650557620818,
      "loss": 0.1616,
      "step": 1900
    },
    {
      "epoch": 1.77,
      "eval_accuracy": 0.8874935930292158,
      "eval_f1": 0.8874935930292158,
      "eval_loss": 0.30438435077667236,
      "eval_precision": 0.8874935930292158,
      "eval_recall": 0.8874935930292158,
      "eval_runtime": 82.6352,
      "eval_samples_per_second": 94.439,
      "eval_steps_per_second": 11.811,
      "step": 1900
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001112918215613383,
      "loss": 0.2711,
      "step": 1910
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011082713754646841,
      "loss": 0.2552,
      "step": 1920
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0001103624535315985,
      "loss": 0.1063,
      "step": 1930
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00010989776951672863,
      "loss": 0.1619,
      "step": 1940
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010943308550185873,
      "loss": 0.1382,
      "step": 1950
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010896840148698885,
      "loss": 0.1206,
      "step": 1960
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010850371747211896,
      "loss": 0.299,
      "step": 1970
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010803903345724908,
      "loss": 0.2133,
      "step": 1980
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00010757434944237919,
      "loss": 0.3214,
      "step": 1990
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00010710966542750931,
      "loss": 0.1527,
      "step": 2000
    },
    {
      "epoch": 1.86,
      "eval_accuracy": 0.8789082521783701,
      "eval_f1": 0.8789082521783701,
      "eval_loss": 0.29628437757492065,
      "eval_precision": 0.8789082521783701,
      "eval_recall": 0.8789082521783701,
      "eval_runtime": 82.9979,
      "eval_samples_per_second": 94.026,
      "eval_steps_per_second": 11.759,
      "step": 2000
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0001066449814126394,
      "loss": 0.2059,
      "step": 2010
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010618029739776951,
      "loss": 0.1791,
      "step": 2020
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010571561338289963,
      "loss": 0.2058,
      "step": 2030
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010525092936802974,
      "loss": 0.2465,
      "step": 2040
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010478624535315986,
      "loss": 0.1888,
      "step": 2050
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010432156133828996,
      "loss": 0.1491,
      "step": 2060
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010385687732342009,
      "loss": 0.1852,
      "step": 2070
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010339219330855019,
      "loss": 0.2082,
      "step": 2080
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010292750929368031,
      "loss": 0.2285,
      "step": 2090
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010246282527881041,
      "loss": 0.2123,
      "step": 2100
    },
    {
      "epoch": 1.95,
      "eval_accuracy": 0.8741670937980522,
      "eval_f1": 0.8741670937980522,
      "eval_loss": 0.2718897759914398,
      "eval_precision": 0.8741670937980522,
      "eval_recall": 0.8741670937980522,
      "eval_runtime": 81.531,
      "eval_samples_per_second": 95.718,
      "eval_steps_per_second": 11.971,
      "step": 2100
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010199814126394052,
      "loss": 0.1424,
      "step": 2110
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010153345724907064,
      "loss": 0.189,
      "step": 2120
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010106877323420074,
      "loss": 0.1315,
      "step": 2130
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010060408921933087,
      "loss": 0.1826,
      "step": 2140
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00010013940520446097,
      "loss": 0.1885,
      "step": 2150
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.967472118959109e-05,
      "loss": 0.2363,
      "step": 2160
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.92100371747212e-05,
      "loss": 0.179,
      "step": 2170
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.87453531598513e-05,
      "loss": 0.188,
      "step": 2180
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.828066914498142e-05,
      "loss": 0.1603,
      "step": 2190
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.781598513011153e-05,
      "loss": 0.1489,
      "step": 2200
    },
    {
      "epoch": 2.04,
      "eval_accuracy": 0.860456176319836,
      "eval_f1": 0.860456176319836,
      "eval_loss": 0.34448257088661194,
      "eval_precision": 0.860456176319836,
      "eval_recall": 0.860456176319836,
      "eval_runtime": 81.7458,
      "eval_samples_per_second": 95.467,
      "eval_steps_per_second": 11.939,
      "step": 2200
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.735130111524164e-05,
      "loss": 0.2102,
      "step": 2210
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.688661710037176e-05,
      "loss": 0.1669,
      "step": 2220
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.642193308550185e-05,
      "loss": 0.1382,
      "step": 2230
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.595724907063197e-05,
      "loss": 0.1638,
      "step": 2240
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.549256505576208e-05,
      "loss": 0.1179,
      "step": 2250
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.50278810408922e-05,
      "loss": 0.0859,
      "step": 2260
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.456319702602231e-05,
      "loss": 0.1584,
      "step": 2270
    },
    {
      "epoch": 2.12,
      "learning_rate": 9.409851301115242e-05,
      "loss": 0.1141,
      "step": 2280
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.363382899628254e-05,
      "loss": 0.0965,
      "step": 2290
    },
    {
      "epoch": 2.14,
      "learning_rate": 9.316914498141265e-05,
      "loss": 0.2052,
      "step": 2300
    },
    {
      "epoch": 2.14,
      "eval_accuracy": 0.8799333675038442,
      "eval_f1": 0.8799333675038442,
      "eval_loss": 0.3297293782234192,
      "eval_precision": 0.8799333675038442,
      "eval_recall": 0.8799333675038442,
      "eval_runtime": 81.5937,
      "eval_samples_per_second": 95.645,
      "eval_steps_per_second": 11.962,
      "step": 2300
    },
    {
      "epoch": 2.15,
      "learning_rate": 9.270446096654275e-05,
      "loss": 0.1231,
      "step": 2310
    },
    {
      "epoch": 2.16,
      "learning_rate": 9.223977695167286e-05,
      "loss": 0.1772,
      "step": 2320
    },
    {
      "epoch": 2.17,
      "learning_rate": 9.177509293680297e-05,
      "loss": 0.2179,
      "step": 2330
    },
    {
      "epoch": 2.17,
      "learning_rate": 9.131040892193309e-05,
      "loss": 0.1335,
      "step": 2340
    },
    {
      "epoch": 2.18,
      "learning_rate": 9.08457249070632e-05,
      "loss": 0.182,
      "step": 2350
    },
    {
      "epoch": 2.19,
      "learning_rate": 9.038104089219332e-05,
      "loss": 0.165,
      "step": 2360
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.991635687732343e-05,
      "loss": 0.2221,
      "step": 2370
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.945167286245353e-05,
      "loss": 0.1836,
      "step": 2380
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.898698884758366e-05,
      "loss": 0.1041,
      "step": 2390
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.852230483271375e-05,
      "loss": 0.172,
      "step": 2400
    },
    {
      "epoch": 2.23,
      "eval_accuracy": 0.8833931317273194,
      "eval_f1": 0.8833931317273194,
      "eval_loss": 0.3089473843574524,
      "eval_precision": 0.8833931317273194,
      "eval_recall": 0.8833931317273194,
      "eval_runtime": 80.9982,
      "eval_samples_per_second": 96.348,
      "eval_steps_per_second": 12.05,
      "step": 2400
    },
    {
      "epoch": 2.24,
      "learning_rate": 8.805762081784387e-05,
      "loss": 0.1657,
      "step": 2410
    },
    {
      "epoch": 2.25,
      "learning_rate": 8.759293680297398e-05,
      "loss": 0.1735,
      "step": 2420
    },
    {
      "epoch": 2.26,
      "learning_rate": 8.71282527881041e-05,
      "loss": 0.1904,
      "step": 2430
    },
    {
      "epoch": 2.27,
      "learning_rate": 8.66635687732342e-05,
      "loss": 0.1191,
      "step": 2440
    },
    {
      "epoch": 2.28,
      "learning_rate": 8.619888475836431e-05,
      "loss": 0.228,
      "step": 2450
    },
    {
      "epoch": 2.29,
      "learning_rate": 8.573420074349443e-05,
      "loss": 0.1509,
      "step": 2460
    },
    {
      "epoch": 2.3,
      "learning_rate": 8.526951672862454e-05,
      "loss": 0.0814,
      "step": 2470
    },
    {
      "epoch": 2.3,
      "learning_rate": 8.480483271375465e-05,
      "loss": 0.1587,
      "step": 2480
    },
    {
      "epoch": 2.31,
      "learning_rate": 8.434014869888476e-05,
      "loss": 0.1757,
      "step": 2490
    },
    {
      "epoch": 2.32,
      "learning_rate": 8.387546468401486e-05,
      "loss": 0.1167,
      "step": 2500
    },
    {
      "epoch": 2.32,
      "eval_accuracy": 0.8763454638646848,
      "eval_f1": 0.8763454638646848,
      "eval_loss": 0.2973257899284363,
      "eval_precision": 0.8763454638646848,
      "eval_recall": 0.8763454638646848,
      "eval_runtime": 82.236,
      "eval_samples_per_second": 94.898,
      "eval_steps_per_second": 11.868,
      "step": 2500
    },
    {
      "epoch": 2.33,
      "learning_rate": 8.341078066914499e-05,
      "loss": 0.1991,
      "step": 2510
    },
    {
      "epoch": 2.34,
      "learning_rate": 8.29460966542751e-05,
      "loss": 0.1288,
      "step": 2520
    },
    {
      "epoch": 2.35,
      "learning_rate": 8.248141263940521e-05,
      "loss": 0.1247,
      "step": 2530
    },
    {
      "epoch": 2.36,
      "learning_rate": 8.20631970260223e-05,
      "loss": 0.1099,
      "step": 2540
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.159851301115243e-05,
      "loss": 0.135,
      "step": 2550
    },
    {
      "epoch": 2.38,
      "learning_rate": 8.113382899628253e-05,
      "loss": 0.1227,
      "step": 2560
    },
    {
      "epoch": 2.39,
      "learning_rate": 8.066914498141265e-05,
      "loss": 0.183,
      "step": 2570
    },
    {
      "epoch": 2.4,
      "learning_rate": 8.020446096654275e-05,
      "loss": 0.1891,
      "step": 2580
    },
    {
      "epoch": 2.41,
      "learning_rate": 7.973977695167286e-05,
      "loss": 0.054,
      "step": 2590
    },
    {
      "epoch": 2.42,
      "learning_rate": 7.927509293680298e-05,
      "loss": 0.0705,
      "step": 2600
    },
    {
      "epoch": 2.42,
      "eval_accuracy": 0.8912096360840595,
      "eval_f1": 0.8912096360840595,
      "eval_loss": 0.3585003912448883,
      "eval_precision": 0.8912096360840595,
      "eval_recall": 0.8912096360840595,
      "eval_runtime": 81.6003,
      "eval_samples_per_second": 95.637,
      "eval_steps_per_second": 11.961,
      "step": 2600
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.885687732342008e-05,
      "loss": 0.195,
      "step": 2610
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.839219330855019e-05,
      "loss": 0.1481,
      "step": 2620
    },
    {
      "epoch": 2.44,
      "learning_rate": 7.79275092936803e-05,
      "loss": 0.2028,
      "step": 2630
    },
    {
      "epoch": 2.45,
      "learning_rate": 7.746282527881042e-05,
      "loss": 0.083,
      "step": 2640
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.699814126394052e-05,
      "loss": 0.1181,
      "step": 2650
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.653345724907065e-05,
      "loss": 0.1704,
      "step": 2660
    },
    {
      "epoch": 2.48,
      "learning_rate": 7.606877323420075e-05,
      "loss": 0.144,
      "step": 2670
    },
    {
      "epoch": 2.49,
      "learning_rate": 7.560408921933086e-05,
      "loss": 0.1065,
      "step": 2680
    },
    {
      "epoch": 2.5,
      "learning_rate": 7.513940520446097e-05,
      "loss": 0.166,
      "step": 2690
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.467472118959108e-05,
      "loss": 0.212,
      "step": 2700
    },
    {
      "epoch": 2.51,
      "eval_accuracy": 0.8671194259354177,
      "eval_f1": 0.8671194259354177,
      "eval_loss": 0.40511995553970337,
      "eval_precision": 0.8671194259354177,
      "eval_recall": 0.8671194259354177,
      "eval_runtime": 81.9633,
      "eval_samples_per_second": 95.213,
      "eval_steps_per_second": 11.908,
      "step": 2700
    },
    {
      "epoch": 2.52,
      "learning_rate": 7.42100371747212e-05,
      "loss": 0.1879,
      "step": 2710
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.37453531598513e-05,
      "loss": 0.1556,
      "step": 2720
    },
    {
      "epoch": 2.54,
      "learning_rate": 7.328066914498141e-05,
      "loss": 0.1873,
      "step": 2730
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.281598513011153e-05,
      "loss": 0.1788,
      "step": 2740
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.235130111524164e-05,
      "loss": 0.1815,
      "step": 2750
    },
    {
      "epoch": 2.57,
      "learning_rate": 7.188661710037176e-05,
      "loss": 0.1559,
      "step": 2760
    },
    {
      "epoch": 2.57,
      "learning_rate": 7.142193308550186e-05,
      "loss": 0.1592,
      "step": 2770
    },
    {
      "epoch": 2.58,
      "learning_rate": 7.095724907063198e-05,
      "loss": 0.1014,
      "step": 2780
    },
    {
      "epoch": 2.59,
      "learning_rate": 7.049256505576208e-05,
      "loss": 0.2037,
      "step": 2790
    },
    {
      "epoch": 2.6,
      "learning_rate": 7.002788104089219e-05,
      "loss": 0.2053,
      "step": 2800
    },
    {
      "epoch": 2.6,
      "eval_accuracy": 0.8910814966683752,
      "eval_f1": 0.8910814966683752,
      "eval_loss": 0.30884426832199097,
      "eval_precision": 0.8910814966683752,
      "eval_recall": 0.8910814966683752,
      "eval_runtime": 81.4179,
      "eval_samples_per_second": 95.851,
      "eval_steps_per_second": 11.988,
      "step": 2800
    },
    {
      "epoch": 2.61,
      "learning_rate": 6.956319702602231e-05,
      "loss": 0.1541,
      "step": 2810
    },
    {
      "epoch": 2.62,
      "learning_rate": 6.909851301115242e-05,
      "loss": 0.1452,
      "step": 2820
    },
    {
      "epoch": 2.63,
      "learning_rate": 6.863382899628253e-05,
      "loss": 0.0788,
      "step": 2830
    },
    {
      "epoch": 2.64,
      "learning_rate": 6.816914498141265e-05,
      "loss": 0.1091,
      "step": 2840
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.770446096654274e-05,
      "loss": 0.1381,
      "step": 2850
    },
    {
      "epoch": 2.66,
      "learning_rate": 6.723977695167286e-05,
      "loss": 0.0573,
      "step": 2860
    },
    {
      "epoch": 2.67,
      "learning_rate": 6.677509293680297e-05,
      "loss": 0.1623,
      "step": 2870
    },
    {
      "epoch": 2.68,
      "learning_rate": 6.631040892193309e-05,
      "loss": 0.0873,
      "step": 2880
    },
    {
      "epoch": 2.69,
      "learning_rate": 6.58457249070632e-05,
      "loss": 0.1201,
      "step": 2890
    },
    {
      "epoch": 2.7,
      "learning_rate": 6.538104089219331e-05,
      "loss": 0.0718,
      "step": 2900
    },
    {
      "epoch": 2.7,
      "eval_accuracy": 0.8894156842644797,
      "eval_f1": 0.8894156842644797,
      "eval_loss": 0.3223333954811096,
      "eval_precision": 0.8894156842644797,
      "eval_recall": 0.8894156842644797,
      "eval_runtime": 81.8904,
      "eval_samples_per_second": 95.298,
      "eval_steps_per_second": 11.918,
      "step": 2900
    },
    {
      "epoch": 2.7,
      "learning_rate": 6.491635687732343e-05,
      "loss": 0.1323,
      "step": 2910
    },
    {
      "epoch": 2.71,
      "learning_rate": 6.445167286245354e-05,
      "loss": 0.1431,
      "step": 2920
    },
    {
      "epoch": 2.72,
      "learning_rate": 6.398698884758366e-05,
      "loss": 0.1032,
      "step": 2930
    },
    {
      "epoch": 2.73,
      "learning_rate": 6.352230483271375e-05,
      "loss": 0.1454,
      "step": 2940
    },
    {
      "epoch": 2.74,
      "learning_rate": 6.305762081784386e-05,
      "loss": 0.1016,
      "step": 2950
    },
    {
      "epoch": 2.75,
      "learning_rate": 6.259293680297398e-05,
      "loss": 0.1578,
      "step": 2960
    },
    {
      "epoch": 2.76,
      "learning_rate": 6.212825278810409e-05,
      "loss": 0.1418,
      "step": 2970
    },
    {
      "epoch": 2.77,
      "learning_rate": 6.166356877323421e-05,
      "loss": 0.125,
      "step": 2980
    },
    {
      "epoch": 2.78,
      "learning_rate": 6.119888475836432e-05,
      "loss": 0.1093,
      "step": 2990
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.073420074349443e-05,
      "loss": 0.0648,
      "step": 3000
    },
    {
      "epoch": 2.79,
      "eval_accuracy": 0.8776268580215274,
      "eval_f1": 0.8776268580215274,
      "eval_loss": 0.342730849981308,
      "eval_precision": 0.8776268580215274,
      "eval_recall": 0.8776268580215274,
      "eval_runtime": 81.3512,
      "eval_samples_per_second": 95.93,
      "eval_steps_per_second": 11.997,
      "step": 3000
    },
    {
      "epoch": 2.8,
      "learning_rate": 6.0269516728624545e-05,
      "loss": 0.0831,
      "step": 3010
    },
    {
      "epoch": 2.81,
      "learning_rate": 5.980483271375465e-05,
      "loss": 0.1566,
      "step": 3020
    },
    {
      "epoch": 2.82,
      "learning_rate": 5.934014869888476e-05,
      "loss": 0.1468,
      "step": 3030
    },
    {
      "epoch": 2.83,
      "learning_rate": 5.887546468401487e-05,
      "loss": 0.0703,
      "step": 3040
    },
    {
      "epoch": 2.83,
      "learning_rate": 5.841078066914498e-05,
      "loss": 0.0471,
      "step": 3050
    },
    {
      "epoch": 2.84,
      "learning_rate": 5.7946096654275096e-05,
      "loss": 0.0853,
      "step": 3060
    },
    {
      "epoch": 2.85,
      "learning_rate": 5.748141263940521e-05,
      "loss": 0.1258,
      "step": 3070
    },
    {
      "epoch": 2.86,
      "learning_rate": 5.701672862453532e-05,
      "loss": 0.1041,
      "step": 3080
    },
    {
      "epoch": 2.87,
      "learning_rate": 5.655204460966543e-05,
      "loss": 0.0977,
      "step": 3090
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.6087360594795546e-05,
      "loss": 0.0889,
      "step": 3100
    },
    {
      "epoch": 2.88,
      "eval_accuracy": 0.8880061506919529,
      "eval_f1": 0.8880061506919529,
      "eval_loss": 0.3503774404525757,
      "eval_precision": 0.8880061506919529,
      "eval_recall": 0.8880061506919529,
      "eval_runtime": 81.4671,
      "eval_samples_per_second": 95.793,
      "eval_steps_per_second": 11.98,
      "step": 3100
    },
    {
      "epoch": 2.89,
      "learning_rate": 5.562267657992565e-05,
      "loss": 0.1593,
      "step": 3110
    },
    {
      "epoch": 2.9,
      "learning_rate": 5.515799256505576e-05,
      "loss": 0.1187,
      "step": 3120
    },
    {
      "epoch": 2.91,
      "learning_rate": 5.4693308550185876e-05,
      "loss": 0.0628,
      "step": 3130
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.422862453531598e-05,
      "loss": 0.0999,
      "step": 3140
    },
    {
      "epoch": 2.93,
      "learning_rate": 5.37639405204461e-05,
      "loss": 0.1343,
      "step": 3150
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.329925650557621e-05,
      "loss": 0.1795,
      "step": 3160
    },
    {
      "epoch": 2.95,
      "learning_rate": 5.2834572490706326e-05,
      "loss": 0.1066,
      "step": 3170
    },
    {
      "epoch": 2.96,
      "learning_rate": 5.236988847583644e-05,
      "loss": 0.1175,
      "step": 3180
    },
    {
      "epoch": 2.96,
      "learning_rate": 5.190520446096655e-05,
      "loss": 0.0596,
      "step": 3190
    },
    {
      "epoch": 2.97,
      "learning_rate": 5.144052044609665e-05,
      "loss": 0.098,
      "step": 3200
    },
    {
      "epoch": 2.97,
      "eval_accuracy": 0.8769861609431061,
      "eval_f1": 0.8769861609431061,
      "eval_loss": 0.3520107567310333,
      "eval_precision": 0.8769861609431061,
      "eval_recall": 0.8769861609431061,
      "eval_runtime": 81.1441,
      "eval_samples_per_second": 96.175,
      "eval_steps_per_second": 12.028,
      "step": 3200
    },
    {
      "epoch": 2.98,
      "learning_rate": 5.097583643122676e-05,
      "loss": 0.1101,
      "step": 3210
    },
    {
      "epoch": 2.99,
      "learning_rate": 5.051115241635688e-05,
      "loss": 0.1006,
      "step": 3220
    },
    {
      "epoch": 3.0,
      "learning_rate": 5.004646840148699e-05,
      "loss": 0.0532,
      "step": 3230
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.9581784386617106e-05,
      "loss": 0.0669,
      "step": 3240
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.9117100371747213e-05,
      "loss": 0.0352,
      "step": 3250
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.865241635687733e-05,
      "loss": 0.0249,
      "step": 3260
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.8187732342007435e-05,
      "loss": 0.0715,
      "step": 3270
    },
    {
      "epoch": 3.05,
      "learning_rate": 4.772304832713755e-05,
      "loss": 0.1082,
      "step": 3280
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.7258364312267664e-05,
      "loss": 0.0833,
      "step": 3290
    },
    {
      "epoch": 3.07,
      "learning_rate": 4.679368029739777e-05,
      "loss": 0.1231,
      "step": 3300
    },
    {
      "epoch": 3.07,
      "eval_accuracy": 0.8799333675038442,
      "eval_f1": 0.8799333675038442,
      "eval_loss": 0.4712468087673187,
      "eval_precision": 0.8799333675038442,
      "eval_recall": 0.8799333675038442,
      "eval_runtime": 81.4378,
      "eval_samples_per_second": 95.828,
      "eval_steps_per_second": 11.985,
      "step": 3300
    },
    {
      "epoch": 3.08,
      "learning_rate": 4.632899628252788e-05,
      "loss": 0.0457,
      "step": 3310
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.586431226765799e-05,
      "loss": 0.1485,
      "step": 3320
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.539962825278811e-05,
      "loss": 0.0372,
      "step": 3330
    },
    {
      "epoch": 3.1,
      "learning_rate": 4.493494423791822e-05,
      "loss": 0.0708,
      "step": 3340
    },
    {
      "epoch": 3.11,
      "learning_rate": 4.447026022304833e-05,
      "loss": 0.1233,
      "step": 3350
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.400557620817844e-05,
      "loss": 0.0471,
      "step": 3360
    },
    {
      "epoch": 3.13,
      "learning_rate": 4.354089219330855e-05,
      "loss": 0.071,
      "step": 3370
    },
    {
      "epoch": 3.14,
      "learning_rate": 4.3076208178438666e-05,
      "loss": 0.0642,
      "step": 3380
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.261152416356878e-05,
      "loss": 0.0383,
      "step": 3390
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.214684014869889e-05,
      "loss": 0.0598,
      "step": 3400
    },
    {
      "epoch": 3.16,
      "eval_accuracy": 0.877883136852896,
      "eval_f1": 0.877883136852896,
      "eval_loss": 0.4759017527103424,
      "eval_precision": 0.877883136852896,
      "eval_recall": 0.877883136852896,
      "eval_runtime": 81.989,
      "eval_samples_per_second": 95.183,
      "eval_steps_per_second": 11.904,
      "step": 3400
    },
    {
      "epoch": 3.17,
      "learning_rate": 4.1682156133828995e-05,
      "loss": 0.0697,
      "step": 3410
    },
    {
      "epoch": 3.18,
      "learning_rate": 4.121747211895911e-05,
      "loss": 0.1165,
      "step": 3420
    },
    {
      "epoch": 3.19,
      "learning_rate": 4.0752788104089224e-05,
      "loss": 0.0887,
      "step": 3430
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.028810408921933e-05,
      "loss": 0.083,
      "step": 3440
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.9823420074349445e-05,
      "loss": 0.0697,
      "step": 3450
    },
    {
      "epoch": 3.22,
      "learning_rate": 3.935873605947955e-05,
      "loss": 0.0845,
      "step": 3460
    },
    {
      "epoch": 3.22,
      "learning_rate": 3.889405204460967e-05,
      "loss": 0.0313,
      "step": 3470
    },
    {
      "epoch": 3.23,
      "learning_rate": 3.8429368029739775e-05,
      "loss": 0.0884,
      "step": 3480
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.796468401486989e-05,
      "loss": 0.0632,
      "step": 3490
    },
    {
      "epoch": 3.25,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.0558,
      "step": 3500
    },
    {
      "epoch": 3.25,
      "eval_accuracy": 0.8798052280881599,
      "eval_f1": 0.8798052280881599,
      "eval_loss": 0.41797372698783875,
      "eval_precision": 0.8798052280881599,
      "eval_recall": 0.8798052280881599,
      "eval_runtime": 82.5775,
      "eval_samples_per_second": 94.505,
      "eval_steps_per_second": 11.819,
      "step": 3500
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.703531598513012e-05,
      "loss": 0.0388,
      "step": 3510
    },
    {
      "epoch": 3.27,
      "learning_rate": 3.6570631970260225e-05,
      "loss": 0.0197,
      "step": 3520
    },
    {
      "epoch": 3.28,
      "learning_rate": 3.610594795539033e-05,
      "loss": 0.0325,
      "step": 3530
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.564126394052045e-05,
      "loss": 0.0868,
      "step": 3540
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.517657992565056e-05,
      "loss": 0.0496,
      "step": 3550
    },
    {
      "epoch": 3.31,
      "learning_rate": 3.4711895910780676e-05,
      "loss": 0.0712,
      "step": 3560
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.424721189591078e-05,
      "loss": 0.0173,
      "step": 3570
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.378252788104089e-05,
      "loss": 0.0241,
      "step": 3580
    },
    {
      "epoch": 3.34,
      "learning_rate": 3.3317843866171005e-05,
      "loss": 0.0711,
      "step": 3590
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.285315985130112e-05,
      "loss": 0.0595,
      "step": 3600
    },
    {
      "epoch": 3.35,
      "eval_accuracy": 0.8864684777037417,
      "eval_f1": 0.8864684777037417,
      "eval_loss": 0.5599675178527832,
      "eval_precision": 0.8864684777037417,
      "eval_recall": 0.8864684777037417,
      "eval_runtime": 81.6461,
      "eval_samples_per_second": 95.583,
      "eval_steps_per_second": 11.954,
      "step": 3600
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.238847583643123e-05,
      "loss": 0.0903,
      "step": 3610
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.192379182156134e-05,
      "loss": 0.0517,
      "step": 3620
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.145910780669145e-05,
      "loss": 0.0931,
      "step": 3630
    },
    {
      "epoch": 3.38,
      "learning_rate": 3.099442379182156e-05,
      "loss": 0.028,
      "step": 3640
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.052973977695168e-05,
      "loss": 0.059,
      "step": 3650
    },
    {
      "epoch": 3.4,
      "learning_rate": 3.0065055762081785e-05,
      "loss": 0.0787,
      "step": 3660
    },
    {
      "epoch": 3.41,
      "learning_rate": 2.9600371747211896e-05,
      "loss": 0.0481,
      "step": 3670
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.913568773234201e-05,
      "loss": 0.0395,
      "step": 3680
    },
    {
      "epoch": 3.43,
      "learning_rate": 2.867100371747212e-05,
      "loss": 0.0369,
      "step": 3690
    },
    {
      "epoch": 3.44,
      "learning_rate": 2.820631970260223e-05,
      "loss": 0.0796,
      "step": 3700
    },
    {
      "epoch": 3.44,
      "eval_accuracy": 0.8922347514095336,
      "eval_f1": 0.8922347514095337,
      "eval_loss": 0.46909022331237793,
      "eval_precision": 0.8922347514095336,
      "eval_recall": 0.8922347514095336,
      "eval_runtime": 81.8248,
      "eval_samples_per_second": 95.375,
      "eval_steps_per_second": 11.928,
      "step": 3700
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.7741635687732343e-05,
      "loss": 0.0789,
      "step": 3710
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.7276951672862454e-05,
      "loss": 0.0514,
      "step": 3720
    },
    {
      "epoch": 3.47,
      "learning_rate": 2.6812267657992568e-05,
      "loss": 0.0962,
      "step": 3730
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.6347583643122676e-05,
      "loss": 0.1187,
      "step": 3740
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.5882899628252787e-05,
      "loss": 0.0257,
      "step": 3750
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.54182156133829e-05,
      "loss": 0.1486,
      "step": 3760
    },
    {
      "epoch": 3.5,
      "learning_rate": 2.4953531598513012e-05,
      "loss": 0.0984,
      "step": 3770
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.4488847583643123e-05,
      "loss": 0.0306,
      "step": 3780
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.4024163568773237e-05,
      "loss": 0.0479,
      "step": 3790
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.3559479553903345e-05,
      "loss": 0.0122,
      "step": 3800
    },
    {
      "epoch": 3.53,
      "eval_accuracy": 0.8935161455663763,
      "eval_f1": 0.8935161455663763,
      "eval_loss": 0.4117405414581299,
      "eval_precision": 0.8935161455663763,
      "eval_recall": 0.8935161455663763,
      "eval_runtime": 81.4692,
      "eval_samples_per_second": 95.791,
      "eval_steps_per_second": 11.98,
      "step": 3800
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.309479553903346e-05,
      "loss": 0.0404,
      "step": 3810
    },
    {
      "epoch": 3.55,
      "learning_rate": 2.263011152416357e-05,
      "loss": 0.0301,
      "step": 3820
    },
    {
      "epoch": 3.56,
      "learning_rate": 2.216542750929368e-05,
      "loss": 0.1052,
      "step": 3830
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.1700743494423792e-05,
      "loss": 0.0303,
      "step": 3840
    },
    {
      "epoch": 3.58,
      "learning_rate": 2.1236059479553906e-05,
      "loss": 0.0587,
      "step": 3850
    },
    {
      "epoch": 3.59,
      "learning_rate": 2.0771375464684014e-05,
      "loss": 0.0249,
      "step": 3860
    },
    {
      "epoch": 3.6,
      "learning_rate": 2.0306691449814128e-05,
      "loss": 0.0341,
      "step": 3870
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.984200743494424e-05,
      "loss": 0.0241,
      "step": 3880
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.937732342007435e-05,
      "loss": 0.029,
      "step": 3890
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.8912639405204464e-05,
      "loss": 0.0633,
      "step": 3900
    },
    {
      "epoch": 3.62,
      "eval_accuracy": 0.8956945156330087,
      "eval_f1": 0.8956945156330087,
      "eval_loss": 0.4275100827217102,
      "eval_precision": 0.8956945156330087,
      "eval_recall": 0.8956945156330087,
      "eval_runtime": 81.288,
      "eval_samples_per_second": 96.004,
      "eval_steps_per_second": 12.007,
      "step": 3900
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.844795539033457e-05,
      "loss": 0.0672,
      "step": 3910
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.7983271375464686e-05,
      "loss": 0.0166,
      "step": 3920
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.7518587360594797e-05,
      "loss": 0.1497,
      "step": 3930
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.7053903345724908e-05,
      "loss": 0.096,
      "step": 3940
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.658921933085502e-05,
      "loss": 0.0067,
      "step": 3950
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.6124535315985133e-05,
      "loss": 0.0558,
      "step": 3960
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.565985130111524e-05,
      "loss": 0.048,
      "step": 3970
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.5195167286245355e-05,
      "loss": 0.0254,
      "step": 3980
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.4730483271375464e-05,
      "loss": 0.0208,
      "step": 3990
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.4265799256505577e-05,
      "loss": 0.0659,
      "step": 4000
    },
    {
      "epoch": 3.72,
      "eval_accuracy": 0.8936442849820605,
      "eval_f1": 0.8936442849820605,
      "eval_loss": 0.42175230383872986,
      "eval_precision": 0.8936442849820605,
      "eval_recall": 0.8936442849820605,
      "eval_runtime": 82.0179,
      "eval_samples_per_second": 95.15,
      "eval_steps_per_second": 11.9,
      "step": 4000
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3801115241635688e-05,
      "loss": 0.0329,
      "step": 4010
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.33364312267658e-05,
      "loss": 0.0611,
      "step": 4020
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.2871747211895913e-05,
      "loss": 0.0311,
      "step": 4030
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.2407063197026022e-05,
      "loss": 0.0725,
      "step": 4040
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.1942379182156135e-05,
      "loss": 0.0639,
      "step": 4050
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.1477695167286246e-05,
      "loss": 0.0727,
      "step": 4060
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.1013011152416357e-05,
      "loss": 0.055,
      "step": 4070
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.054832713754647e-05,
      "loss": 0.0206,
      "step": 4080
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.0083643122676582e-05,
      "loss": 0.0532,
      "step": 4090
    },
    {
      "epoch": 3.81,
      "learning_rate": 9.618959107806693e-06,
      "loss": 0.0155,
      "step": 4100
    },
    {
      "epoch": 3.81,
      "eval_accuracy": 0.8981291645310098,
      "eval_f1": 0.8981291645310098,
      "eval_loss": 0.41891658306121826,
      "eval_precision": 0.8981291645310098,
      "eval_recall": 0.8981291645310098,
      "eval_runtime": 81.6227,
      "eval_samples_per_second": 95.611,
      "eval_steps_per_second": 11.957,
      "step": 4100
    },
    {
      "epoch": 3.82,
      "learning_rate": 9.154275092936804e-06,
      "loss": 0.0581,
      "step": 4110
    },
    {
      "epoch": 3.83,
      "learning_rate": 8.689591078066915e-06,
      "loss": 0.0174,
      "step": 4120
    },
    {
      "epoch": 3.84,
      "learning_rate": 8.224907063197027e-06,
      "loss": 0.0252,
      "step": 4130
    },
    {
      "epoch": 3.85,
      "learning_rate": 7.760223048327138e-06,
      "loss": 0.0311,
      "step": 4140
    },
    {
      "epoch": 3.86,
      "learning_rate": 7.295539033457249e-06,
      "loss": 0.06,
      "step": 4150
    },
    {
      "epoch": 3.87,
      "learning_rate": 6.830855018587361e-06,
      "loss": 0.0464,
      "step": 4160
    },
    {
      "epoch": 3.88,
      "learning_rate": 6.366171003717472e-06,
      "loss": 0.0145,
      "step": 4170
    },
    {
      "epoch": 3.88,
      "learning_rate": 5.901486988847584e-06,
      "loss": 0.0443,
      "step": 4180
    },
    {
      "epoch": 3.89,
      "learning_rate": 5.436802973977695e-06,
      "loss": 0.005,
      "step": 4190
    },
    {
      "epoch": 3.9,
      "learning_rate": 4.972118959107807e-06,
      "loss": 0.0296,
      "step": 4200
    },
    {
      "epoch": 3.9,
      "eval_accuracy": 0.8973603280369041,
      "eval_f1": 0.8973603280369042,
      "eval_loss": 0.44437190890312195,
      "eval_precision": 0.8973603280369041,
      "eval_recall": 0.8973603280369041,
      "eval_runtime": 81.4663,
      "eval_samples_per_second": 95.794,
      "eval_steps_per_second": 11.98,
      "step": 4200
    },
    {
      "epoch": 3.91,
      "learning_rate": 4.507434944237918e-06,
      "loss": 0.0035,
      "step": 4210
    },
    {
      "epoch": 3.92,
      "learning_rate": 4.04275092936803e-06,
      "loss": 0.0286,
      "step": 4220
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.578066914498141e-06,
      "loss": 0.0622,
      "step": 4230
    },
    {
      "epoch": 3.94,
      "learning_rate": 3.113382899628253e-06,
      "loss": 0.0162,
      "step": 4240
    },
    {
      "epoch": 3.95,
      "learning_rate": 2.648698884758364e-06,
      "loss": 0.0089,
      "step": 4250
    },
    {
      "epoch": 3.96,
      "learning_rate": 2.184014869888476e-06,
      "loss": 0.0463,
      "step": 4260
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.7193308550185875e-06,
      "loss": 0.0478,
      "step": 4270
    },
    {
      "epoch": 3.98,
      "learning_rate": 1.254646840148699e-06,
      "loss": 0.0022,
      "step": 4280
    },
    {
      "epoch": 3.99,
      "learning_rate": 7.899628252788104e-07,
      "loss": 0.0361,
      "step": 4290
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.2527881040892197e-07,
      "loss": 0.0703,
      "step": 4300
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.898257303946694,
      "eval_f1": 0.898257303946694,
      "eval_loss": 0.44992464780807495,
      "eval_precision": 0.898257303946694,
      "eval_recall": 0.898257303946694,
      "eval_runtime": 82.0046,
      "eval_samples_per_second": 95.165,
      "eval_steps_per_second": 11.902,
      "step": 4300
    },
    {
      "epoch": 4.0,
      "step": 4304,
      "total_flos": 1.8862011888755245e+19,
      "train_loss": 0.1975136463254254,
      "train_runtime": 5987.1509,
      "train_samples_per_second": 11.501,
      "train_steps_per_second": 0.719
    }
  ],
  "max_steps": 4304,
  "num_train_epochs": 4,
  "total_flos": 1.8862011888755245e+19,
  "trial_name": null,
  "trial_params": null
}