{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 17.02127659574468,
  "eval_steps": 20,
  "global_step": 3200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 0.0004995563442768412,
      "loss": 1.3139,
      "step": 10
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004986690328305235,
      "loss": 1.2812,
      "step": 20
    },
    {
      "epoch": 0.11,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8499402403831482,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 125.9807,
      "eval_samples_per_second": 11.907,
      "eval_steps_per_second": 0.373,
      "step": 20
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004977817213842058,
      "loss": 0.832,
      "step": 30
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004968944099378882,
      "loss": 0.8156,
      "step": 40
    },
    {
      "epoch": 0.21,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8848057985305786,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 46.3365,
      "eval_samples_per_second": 32.372,
      "eval_steps_per_second": 1.014,
      "step": 40
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004960070984915705,
      "loss": 0.7475,
      "step": 50
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004951197870452529,
      "loss": 0.7478,
      "step": 60
    },
    {
      "epoch": 0.32,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7691774964332581,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0551,
      "eval_samples_per_second": 247.726,
      "eval_steps_per_second": 7.762,
      "step": 60
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004942324755989353,
      "loss": 0.7663,
      "step": 70
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004933451641526176,
      "loss": 0.6034,
      "step": 80
    },
    {
      "epoch": 0.43,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7553095817565918,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.025,
      "eval_samples_per_second": 248.962,
      "eval_steps_per_second": 7.801,
      "step": 80
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004924578527063,
      "loss": 0.9511,
      "step": 90
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004915705412599822,
      "loss": 0.7548,
      "step": 100
    },
    {
      "epoch": 0.53,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7768574357032776,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9796,
      "eval_samples_per_second": 250.851,
      "eval_steps_per_second": 7.86,
      "step": 100
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004906832298136646,
      "loss": 0.858,
      "step": 110
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004897959183673469,
      "loss": 0.8452,
      "step": 120
    },
    {
      "epoch": 0.64,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.745326817035675,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1179,
      "eval_samples_per_second": 245.183,
      "eval_steps_per_second": 7.682,
      "step": 120
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004889086069210293,
      "loss": 0.8152,
      "step": 130
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00048802129547471164,
      "loss": 0.8972,
      "step": 140
    },
    {
      "epoch": 0.74,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7610827088356018,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0456,
      "eval_samples_per_second": 248.113,
      "eval_steps_per_second": 7.774,
      "step": 140
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.000487133984028394,
      "loss": 0.7226,
      "step": 150
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0004862466725820763,
      "loss": 0.9377,
      "step": 160
    },
    {
      "epoch": 0.85,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7714303135871887,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9995,
      "eval_samples_per_second": 250.022,
      "eval_steps_per_second": 7.834,
      "step": 160
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0004853593611357587,
      "loss": 0.594,
      "step": 170
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00048447204968944104,
      "loss": 1.0968,
      "step": 180
    },
    {
      "epoch": 0.96,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.9954016804695129,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9871,
      "eval_samples_per_second": 250.537,
      "eval_steps_per_second": 7.85,
      "step": 180
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048358473824312333,
      "loss": 0.9644,
      "step": 190
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00048269742679680566,
      "loss": 0.7502,
      "step": 200
    },
    {
      "epoch": 1.06,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7544593214988708,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0077,
      "eval_samples_per_second": 249.679,
      "eval_steps_per_second": 7.823,
      "step": 200
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.000481810115350488,
      "loss": 0.7231,
      "step": 210
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0004809228039041704,
      "loss": 0.7902,
      "step": 220
    },
    {
      "epoch": 1.17,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7758567929267883,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0531,
      "eval_samples_per_second": 247.806,
      "eval_steps_per_second": 7.765,
      "step": 220
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00048003549245785273,
      "loss": 0.7591,
      "step": 230
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00047914818101153507,
      "loss": 0.817,
      "step": 240
    },
    {
      "epoch": 1.28,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.767899751663208,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1711,
      "eval_samples_per_second": 243.067,
      "eval_steps_per_second": 7.616,
      "step": 240
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0004782608695652174,
      "loss": 0.6681,
      "step": 250
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00047737355811889974,
      "loss": 0.9199,
      "step": 260
    },
    {
      "epoch": 1.38,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7462519407272339,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0331,
      "eval_samples_per_second": 248.63,
      "eval_steps_per_second": 7.79,
      "step": 260
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00047648624667258213,
      "loss": 0.7409,
      "step": 270
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0004755989352262644,
      "loss": 0.7246,
      "step": 280
    },
    {
      "epoch": 1.49,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7581822872161865,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1558,
      "eval_samples_per_second": 243.673,
      "eval_steps_per_second": 7.635,
      "step": 280
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00047471162377994675,
      "loss": 0.6876,
      "step": 290
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0004738243123336291,
      "loss": 0.7508,
      "step": 300
    },
    {
      "epoch": 1.6,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7464602589607239,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1548,
      "eval_samples_per_second": 243.711,
      "eval_steps_per_second": 7.636,
      "step": 300
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0004729370008873114,
      "loss": 0.8539,
      "step": 310
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0004720496894409938,
      "loss": 0.7011,
      "step": 320
    },
    {
      "epoch": 1.7,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.788334310054779,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1655,
      "eval_samples_per_second": 243.289,
      "eval_steps_per_second": 7.623,
      "step": 320
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00047116237799467615,
      "loss": 0.9111,
      "step": 330
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0004702750665483585,
      "loss": 0.8442,
      "step": 340
    },
    {
      "epoch": 1.81,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7656621336936951,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2177,
      "eval_samples_per_second": 241.249,
      "eval_steps_per_second": 7.559,
      "step": 340
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00046938775510204083,
      "loss": 0.7179,
      "step": 350
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00046850044365572317,
      "loss": 1.0021,
      "step": 360
    },
    {
      "epoch": 1.91,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7755117416381836,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0871,
      "eval_samples_per_second": 246.422,
      "eval_steps_per_second": 7.721,
      "step": 360
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0004676131322094055,
      "loss": 0.8331,
      "step": 370
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.00046672582076308784,
      "loss": 0.7982,
      "step": 380
    },
    {
      "epoch": 2.02,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8436231017112732,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0719,
      "eval_samples_per_second": 247.04,
      "eval_steps_per_second": 7.741,
      "step": 380
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0004658385093167702,
      "loss": 0.8408,
      "step": 390
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0004649511978704525,
      "loss": 0.8299,
      "step": 400
    },
    {
      "epoch": 2.13,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7582988142967224,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2274,
      "eval_samples_per_second": 240.873,
      "eval_steps_per_second": 7.547,
      "step": 400
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0004640638864241349,
      "loss": 0.7852,
      "step": 410
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.00046317657497781724,
      "loss": 0.8789,
      "step": 420
    },
    {
      "epoch": 2.23,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7546933889389038,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3059,
      "eval_samples_per_second": 237.872,
      "eval_steps_per_second": 7.453,
      "step": 420
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0004622892635314996,
      "loss": 0.6384,
      "step": 430
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0004614019520851819,
      "loss": 0.8853,
      "step": 440
    },
    {
      "epoch": 2.34,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7661357522010803,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3477,
      "eval_samples_per_second": 236.306,
      "eval_steps_per_second": 7.404,
      "step": 440
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0004605146406388642,
      "loss": 0.7721,
      "step": 450
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0004596273291925466,
      "loss": 0.7522,
      "step": 460
    },
    {
      "epoch": 2.45,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7484121322631836,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2677,
      "eval_samples_per_second": 239.321,
      "eval_steps_per_second": 7.499,
      "step": 460
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0004587400177462289,
      "loss": 0.8076,
      "step": 470
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.00045785270629991126,
      "loss": 0.9035,
      "step": 480
    },
    {
      "epoch": 2.55,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7667799592018127,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.4501,
      "eval_samples_per_second": 232.554,
      "eval_steps_per_second": 7.287,
      "step": 480
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0004569653948535936,
      "loss": 0.7544,
      "step": 490
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.000456078083407276,
      "loss": 0.8717,
      "step": 500
    },
    {
      "epoch": 2.66,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7541614770889282,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.4865,
      "eval_samples_per_second": 231.251,
      "eval_steps_per_second": 7.246,
      "step": 500
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00045519077196095833,
      "loss": 0.6823,
      "step": 510
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.00045430346051464067,
      "loss": 0.8903,
      "step": 520
    },
    {
      "epoch": 2.77,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.801760733127594,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3567,
      "eval_samples_per_second": 235.972,
      "eval_steps_per_second": 7.394,
      "step": 520
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.000453416149068323,
      "loss": 0.7565,
      "step": 530
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0004525288376220053,
      "loss": 0.7392,
      "step": 540
    },
    {
      "epoch": 2.87,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7826161980628967,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.5294,
      "eval_samples_per_second": 229.731,
      "eval_steps_per_second": 7.198,
      "step": 540
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0004516415261756877,
      "loss": 0.629,
      "step": 550
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.00045075421472937,
      "loss": 0.8626,
      "step": 560
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7489452958106995,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3521,
      "eval_samples_per_second": 236.143,
      "eval_steps_per_second": 7.399,
      "step": 560
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.00044986690328305235,
      "loss": 0.7379,
      "step": 570
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0004489795918367347,
      "loss": 0.6063,
      "step": 580
    },
    {
      "epoch": 3.09,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.9793464541435242,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3272,
      "eval_samples_per_second": 237.071,
      "eval_steps_per_second": 7.428,
      "step": 580
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.000448092280390417,
      "loss": 0.9408,
      "step": 590
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0004472049689440994,
      "loss": 0.8703,
      "step": 600
    },
    {
      "epoch": 3.19,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7475989460945129,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3253,
      "eval_samples_per_second": 237.143,
      "eval_steps_per_second": 7.43,
      "step": 600
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.00044631765749778175,
      "loss": 0.8594,
      "step": 610
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0004454303460514641,
      "loss": 0.797,
      "step": 620
    },
    {
      "epoch": 3.3,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7692281603813171,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.6464,
      "eval_samples_per_second": 225.686,
      "eval_steps_per_second": 7.071,
      "step": 620
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0004445430346051464,
      "loss": 0.8531,
      "step": 630
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.00044365572315882877,
      "loss": 0.7526,
      "step": 640
    },
    {
      "epoch": 3.4,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7548736929893494,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.5723,
      "eval_samples_per_second": 228.23,
      "eval_steps_per_second": 7.151,
      "step": 640
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0004427684117125111,
      "loss": 0.8442,
      "step": 650
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.00044188110026619344,
      "loss": 0.9005,
      "step": 660
    },
    {
      "epoch": 3.51,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8103847503662109,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.4776,
      "eval_samples_per_second": 231.566,
      "eval_steps_per_second": 7.256,
      "step": 660
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.0004409937888198758,
      "loss": 0.8343,
      "step": 670
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.0004401064773735581,
      "loss": 0.7993,
      "step": 680
    },
    {
      "epoch": 3.62,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8053144812583923,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3285,
      "eval_samples_per_second": 237.021,
      "eval_steps_per_second": 7.427,
      "step": 680
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.0004392191659272405,
      "loss": 0.8369,
      "step": 690
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.00043833185448092284,
      "loss": 0.7966,
      "step": 700
    },
    {
      "epoch": 3.72,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7448229193687439,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3865,
      "eval_samples_per_second": 234.871,
      "eval_steps_per_second": 7.359,
      "step": 700
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.0004374445430346052,
      "loss": 0.7612,
      "step": 710
    },
    {
      "epoch": 3.83,
      "learning_rate": 0.00043655723158828746,
      "loss": 0.7656,
      "step": 720
    },
    {
      "epoch": 3.83,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7586243748664856,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3473,
      "eval_samples_per_second": 236.32,
      "eval_steps_per_second": 7.405,
      "step": 720
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.0004356699201419698,
      "loss": 0.6469,
      "step": 730
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0004347826086956522,
      "loss": 0.6889,
      "step": 740
    },
    {
      "epoch": 3.94,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7640272378921509,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3581,
      "eval_samples_per_second": 235.92,
      "eval_steps_per_second": 7.392,
      "step": 740
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.0004338952972493345,
      "loss": 0.7086,
      "step": 750
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.00043300798580301686,
      "loss": 0.8226,
      "step": 760
    },
    {
      "epoch": 4.04,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8118994235992432,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3396,
      "eval_samples_per_second": 236.61,
      "eval_steps_per_second": 7.414,
      "step": 760
    },
    {
      "epoch": 4.1,
      "learning_rate": 0.0004321206743566992,
      "loss": 1.0159,
      "step": 770
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.0004312333629103816,
      "loss": 0.8907,
      "step": 780
    },
    {
      "epoch": 4.15,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7906585335731506,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.4149,
      "eval_samples_per_second": 233.829,
      "eval_steps_per_second": 7.327,
      "step": 780
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.00043034605146406393,
      "loss": 0.6486,
      "step": 790
    },
    {
      "epoch": 4.26,
      "learning_rate": 0.00042945874001774627,
      "loss": 0.8749,
      "step": 800
    },
    {
      "epoch": 4.26,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7677525877952576,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3008,
      "eval_samples_per_second": 238.066,
      "eval_steps_per_second": 7.459,
      "step": 800
    },
    {
      "epoch": 4.31,
      "learning_rate": 0.00042857142857142855,
      "loss": 0.6891,
      "step": 810
    },
    {
      "epoch": 4.36,
      "learning_rate": 0.0004276841171251109,
      "loss": 0.749,
      "step": 820
    },
    {
      "epoch": 4.36,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7736731767654419,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.384,
      "eval_samples_per_second": 234.964,
      "eval_steps_per_second": 7.362,
      "step": 820
    },
    {
      "epoch": 4.41,
      "learning_rate": 0.0004267968056787933,
      "loss": 0.7781,
      "step": 830
    },
    {
      "epoch": 4.47,
      "learning_rate": 0.0004259094942324756,
      "loss": 0.7333,
      "step": 840
    },
    {
      "epoch": 4.47,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8354427218437195,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1708,
      "eval_samples_per_second": 243.081,
      "eval_steps_per_second": 7.617,
      "step": 840
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.00042502218278615795,
      "loss": 1.1763,
      "step": 850
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.0004241348713398403,
      "loss": 0.9287,
      "step": 860
    },
    {
      "epoch": 4.57,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7903938889503479,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1862,
      "eval_samples_per_second": 242.475,
      "eval_steps_per_second": 7.598,
      "step": 860
    },
    {
      "epoch": 4.63,
      "learning_rate": 0.0004232475598935226,
      "loss": 0.7707,
      "step": 870
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.000422360248447205,
      "loss": 0.8231,
      "step": 880
    },
    {
      "epoch": 4.68,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7689003944396973,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0616,
      "eval_samples_per_second": 247.459,
      "eval_steps_per_second": 7.754,
      "step": 880
    },
    {
      "epoch": 4.73,
      "learning_rate": 0.00042147293700088735,
      "loss": 0.723,
      "step": 890
    },
    {
      "epoch": 4.79,
      "learning_rate": 0.00042058562555456964,
      "loss": 0.7486,
      "step": 900
    },
    {
      "epoch": 4.79,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7458215951919556,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0863,
      "eval_samples_per_second": 246.454,
      "eval_steps_per_second": 7.722,
      "step": 900
    },
    {
      "epoch": 4.84,
      "learning_rate": 0.000419698314108252,
      "loss": 0.6488,
      "step": 910
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.0004188110026619343,
      "loss": 0.8551,
      "step": 920
    },
    {
      "epoch": 4.89,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7851412892341614,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9684,
      "eval_samples_per_second": 251.326,
      "eval_steps_per_second": 7.875,
      "step": 920
    },
    {
      "epoch": 4.95,
      "learning_rate": 0.0004179236912156167,
      "loss": 0.7437,
      "step": 930
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00041703637976929904,
      "loss": 0.7452,
      "step": 940
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7961435317993164,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0838,
      "eval_samples_per_second": 246.558,
      "eval_steps_per_second": 7.725,
      "step": 940
    },
    {
      "epoch": 5.05,
      "learning_rate": 0.0004161490683229814,
      "loss": 0.8069,
      "step": 950
    },
    {
      "epoch": 5.11,
      "learning_rate": 0.0004152617568766637,
      "loss": 0.7453,
      "step": 960
    },
    {
      "epoch": 5.11,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7462239861488342,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1042,
      "eval_samples_per_second": 245.731,
      "eval_steps_per_second": 7.7,
      "step": 960
    },
    {
      "epoch": 5.16,
      "learning_rate": 0.0004143744454303461,
      "loss": 0.5743,
      "step": 970
    },
    {
      "epoch": 5.21,
      "learning_rate": 0.00041348713398402844,
      "loss": 0.7784,
      "step": 980
    },
    {
      "epoch": 5.21,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7713678479194641,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0898,
      "eval_samples_per_second": 246.314,
      "eval_steps_per_second": 7.718,
      "step": 980
    },
    {
      "epoch": 5.27,
      "learning_rate": 0.0004125998225377107,
      "loss": 0.7916,
      "step": 990
    },
    {
      "epoch": 5.32,
      "learning_rate": 0.00041171251109139306,
      "loss": 0.7943,
      "step": 1000
    },
    {
      "epoch": 5.32,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7590846419334412,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0172,
      "eval_samples_per_second": 249.284,
      "eval_steps_per_second": 7.811,
      "step": 1000
    },
    {
      "epoch": 5.37,
      "learning_rate": 0.0004108251996450754,
      "loss": 0.7477,
      "step": 1010
    },
    {
      "epoch": 5.43,
      "learning_rate": 0.0004099378881987578,
      "loss": 0.8187,
      "step": 1020
    },
    {
      "epoch": 5.43,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7467174530029297,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0527,
      "eval_samples_per_second": 247.824,
      "eval_steps_per_second": 7.765,
      "step": 1020
    },
    {
      "epoch": 5.48,
      "learning_rate": 0.0004090505767524401,
      "loss": 0.724,
      "step": 1030
    },
    {
      "epoch": 5.53,
      "learning_rate": 0.00040816326530612246,
      "loss": 0.82,
      "step": 1040
    },
    {
      "epoch": 5.53,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7653962969779968,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0631,
      "eval_samples_per_second": 247.397,
      "eval_steps_per_second": 7.752,
      "step": 1040
    },
    {
      "epoch": 5.59,
      "learning_rate": 0.0004072759538598048,
      "loss": 0.7251,
      "step": 1050
    },
    {
      "epoch": 5.64,
      "learning_rate": 0.00040638864241348714,
      "loss": 0.8369,
      "step": 1060
    },
    {
      "epoch": 5.64,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7671139240264893,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9969,
      "eval_samples_per_second": 250.128,
      "eval_steps_per_second": 7.837,
      "step": 1060
    },
    {
      "epoch": 5.69,
      "learning_rate": 0.0004055013309671695,
      "loss": 0.8827,
      "step": 1070
    },
    {
      "epoch": 5.74,
      "learning_rate": 0.0004046140195208518,
      "loss": 0.8594,
      "step": 1080
    },
    {
      "epoch": 5.74,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7597693800926208,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1035,
      "eval_samples_per_second": 245.759,
      "eval_steps_per_second": 7.7,
      "step": 1080
    },
    {
      "epoch": 5.8,
      "learning_rate": 0.00040372670807453415,
      "loss": 0.6991,
      "step": 1090
    },
    {
      "epoch": 5.85,
      "learning_rate": 0.0004028393966282165,
      "loss": 0.763,
      "step": 1100
    },
    {
      "epoch": 5.85,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7626741528511047,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2099,
      "eval_samples_per_second": 241.55,
      "eval_steps_per_second": 7.569,
      "step": 1100
    },
    {
      "epoch": 5.9,
      "learning_rate": 0.0004019520851818989,
      "loss": 0.803,
      "step": 1110
    },
    {
      "epoch": 5.96,
      "learning_rate": 0.0004010647737355812,
      "loss": 0.8508,
      "step": 1120
    },
    {
      "epoch": 5.96,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7459200620651245,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1055,
      "eval_samples_per_second": 245.68,
      "eval_steps_per_second": 7.698,
      "step": 1120
    },
    {
      "epoch": 6.01,
      "learning_rate": 0.00040017746228926355,
      "loss": 0.655,
      "step": 1130
    },
    {
      "epoch": 6.06,
      "learning_rate": 0.0003992901508429459,
      "loss": 0.67,
      "step": 1140
    },
    {
      "epoch": 6.06,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7736058235168457,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0743,
      "eval_samples_per_second": 246.941,
      "eval_steps_per_second": 7.737,
      "step": 1140
    },
    {
      "epoch": 6.12,
      "learning_rate": 0.0003984028393966282,
      "loss": 0.9277,
      "step": 1150
    },
    {
      "epoch": 6.17,
      "learning_rate": 0.00039751552795031056,
      "loss": 0.7358,
      "step": 1160
    },
    {
      "epoch": 6.17,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7588059902191162,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0157,
      "eval_samples_per_second": 249.349,
      "eval_steps_per_second": 7.813,
      "step": 1160
    },
    {
      "epoch": 6.22,
      "learning_rate": 0.0003966282165039929,
      "loss": 0.6581,
      "step": 1170
    },
    {
      "epoch": 6.28,
      "learning_rate": 0.00039574090505767524,
      "loss": 0.6768,
      "step": 1180
    },
    {
      "epoch": 6.28,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7461269497871399,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0429,
      "eval_samples_per_second": 248.226,
      "eval_steps_per_second": 7.778,
      "step": 1180
    },
    {
      "epoch": 6.33,
      "learning_rate": 0.0003948535936113576,
      "loss": 0.6814,
      "step": 1190
    },
    {
      "epoch": 6.38,
      "learning_rate": 0.0003939662821650399,
      "loss": 0.7112,
      "step": 1200
    },
    {
      "epoch": 6.38,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7580690383911133,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.987,
      "eval_samples_per_second": 250.543,
      "eval_steps_per_second": 7.85,
      "step": 1200
    },
    {
      "epoch": 6.44,
      "learning_rate": 0.0003930789707187223,
      "loss": 0.78,
      "step": 1210
    },
    {
      "epoch": 6.49,
      "learning_rate": 0.00039219165927240464,
      "loss": 0.7991,
      "step": 1220
    },
    {
      "epoch": 6.49,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7527822256088257,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0389,
      "eval_samples_per_second": 248.39,
      "eval_steps_per_second": 7.783,
      "step": 1220
    },
    {
      "epoch": 6.54,
      "learning_rate": 0.000391304347826087,
      "loss": 0.833,
      "step": 1230
    },
    {
      "epoch": 6.6,
      "learning_rate": 0.0003904170363797693,
      "loss": 0.9258,
      "step": 1240
    },
    {
      "epoch": 6.6,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7517044544219971,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9942,
      "eval_samples_per_second": 250.24,
      "eval_steps_per_second": 7.841,
      "step": 1240
    },
    {
      "epoch": 6.65,
      "learning_rate": 0.00038952972493345165,
      "loss": 0.6751,
      "step": 1250
    },
    {
      "epoch": 6.7,
      "learning_rate": 0.000388642413487134,
      "loss": 0.7574,
      "step": 1260
    },
    {
      "epoch": 6.7,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7456061244010925,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1978,
      "eval_samples_per_second": 242.023,
      "eval_steps_per_second": 7.583,
      "step": 1260
    },
    {
      "epoch": 6.76,
      "learning_rate": 0.0003877551020408163,
      "loss": 0.8179,
      "step": 1270
    },
    {
      "epoch": 6.81,
      "learning_rate": 0.00038686779059449866,
      "loss": 0.7691,
      "step": 1280
    },
    {
      "epoch": 6.81,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7492356896400452,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0678,
      "eval_samples_per_second": 247.208,
      "eval_steps_per_second": 7.746,
      "step": 1280
    },
    {
      "epoch": 6.86,
      "learning_rate": 0.000385980479148181,
      "loss": 0.7528,
      "step": 1290
    },
    {
      "epoch": 6.91,
      "learning_rate": 0.0003850931677018634,
      "loss": 0.5853,
      "step": 1300
    },
    {
      "epoch": 6.91,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7996741533279419,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9805,
      "eval_samples_per_second": 250.815,
      "eval_steps_per_second": 7.859,
      "step": 1300
    },
    {
      "epoch": 6.97,
      "learning_rate": 0.0003842058562555457,
      "loss": 1.0142,
      "step": 1310
    },
    {
      "epoch": 7.02,
      "learning_rate": 0.00038331854480922806,
      "loss": 0.9782,
      "step": 1320
    },
    {
      "epoch": 7.02,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7786191701889038,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1561,
      "eval_samples_per_second": 243.66,
      "eval_steps_per_second": 7.635,
      "step": 1320
    },
    {
      "epoch": 7.07,
      "learning_rate": 0.0003824312333629104,
      "loss": 0.8774,
      "step": 1330
    },
    {
      "epoch": 7.13,
      "learning_rate": 0.0003815439219165927,
      "loss": 0.73,
      "step": 1340
    },
    {
      "epoch": 7.13,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7597090005874634,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.252,
      "eval_samples_per_second": 239.925,
      "eval_steps_per_second": 7.518,
      "step": 1340
    },
    {
      "epoch": 7.18,
      "learning_rate": 0.0003806566104702751,
      "loss": 0.6523,
      "step": 1350
    },
    {
      "epoch": 7.23,
      "learning_rate": 0.0003797692990239574,
      "loss": 0.5814,
      "step": 1360
    },
    {
      "epoch": 7.23,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8069555759429932,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0437,
      "eval_samples_per_second": 248.194,
      "eval_steps_per_second": 7.777,
      "step": 1360
    },
    {
      "epoch": 7.29,
      "learning_rate": 0.00037888198757763975,
      "loss": 0.8547,
      "step": 1370
    },
    {
      "epoch": 7.34,
      "learning_rate": 0.0003779946761313221,
      "loss": 0.8289,
      "step": 1380
    },
    {
      "epoch": 7.34,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7525202035903931,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0147,
      "eval_samples_per_second": 249.387,
      "eval_steps_per_second": 7.814,
      "step": 1380
    },
    {
      "epoch": 7.39,
      "learning_rate": 0.0003771073646850044,
      "loss": 0.7157,
      "step": 1390
    },
    {
      "epoch": 7.45,
      "learning_rate": 0.0003762200532386868,
      "loss": 0.7269,
      "step": 1400
    },
    {
      "epoch": 7.45,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7548522353172302,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1004,
      "eval_samples_per_second": 245.886,
      "eval_steps_per_second": 7.704,
      "step": 1400
    },
    {
      "epoch": 7.5,
      "learning_rate": 0.00037533274179236915,
      "loss": 0.8475,
      "step": 1410
    },
    {
      "epoch": 7.55,
      "learning_rate": 0.0003744454303460515,
      "loss": 0.7227,
      "step": 1420
    },
    {
      "epoch": 7.55,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7969293594360352,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0981,
      "eval_samples_per_second": 245.98,
      "eval_steps_per_second": 7.707,
      "step": 1420
    },
    {
      "epoch": 7.61,
      "learning_rate": 0.00037355811889973377,
      "loss": 0.9252,
      "step": 1430
    },
    {
      "epoch": 7.66,
      "learning_rate": 0.00037267080745341616,
      "loss": 0.71,
      "step": 1440
    },
    {
      "epoch": 7.66,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8190247416496277,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0637,
      "eval_samples_per_second": 247.376,
      "eval_steps_per_second": 7.751,
      "step": 1440
    },
    {
      "epoch": 7.71,
      "learning_rate": 0.0003717834960070985,
      "loss": 0.8722,
      "step": 1450
    },
    {
      "epoch": 7.77,
      "learning_rate": 0.00037089618456078084,
      "loss": 1.0148,
      "step": 1460
    },
    {
      "epoch": 7.77,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7832964062690735,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.091,
      "eval_samples_per_second": 246.267,
      "eval_steps_per_second": 7.716,
      "step": 1460
    },
    {
      "epoch": 7.82,
      "learning_rate": 0.0003700088731144632,
      "loss": 0.772,
      "step": 1470
    },
    {
      "epoch": 7.87,
      "learning_rate": 0.0003691215616681455,
      "loss": 0.6867,
      "step": 1480
    },
    {
      "epoch": 7.87,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8290088772773743,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1496,
      "eval_samples_per_second": 243.918,
      "eval_steps_per_second": 7.643,
      "step": 1480
    },
    {
      "epoch": 7.93,
      "learning_rate": 0.0003682342502218279,
      "loss": 0.8245,
      "step": 1490
    },
    {
      "epoch": 7.98,
      "learning_rate": 0.00036734693877551024,
      "loss": 0.8665,
      "step": 1500
    },
    {
      "epoch": 7.98,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7548300623893738,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2167,
      "eval_samples_per_second": 241.285,
      "eval_steps_per_second": 7.56,
      "step": 1500
    },
    {
      "epoch": 8.03,
      "learning_rate": 0.0003664596273291926,
      "loss": 0.6394,
      "step": 1510
    },
    {
      "epoch": 8.09,
      "learning_rate": 0.00036557231588287486,
      "loss": 0.731,
      "step": 1520
    },
    {
      "epoch": 8.09,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7873489856719971,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2267,
      "eval_samples_per_second": 240.898,
      "eval_steps_per_second": 7.548,
      "step": 1520
    },
    {
      "epoch": 8.14,
      "learning_rate": 0.0003646850044365572,
      "loss": 0.7414,
      "step": 1530
    },
    {
      "epoch": 8.19,
      "learning_rate": 0.0003637976929902396,
      "loss": 0.7668,
      "step": 1540
    },
    {
      "epoch": 8.19,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.809903085231781,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.3017,
      "eval_samples_per_second": 238.03,
      "eval_steps_per_second": 7.458,
      "step": 1540
    },
    {
      "epoch": 8.24,
      "learning_rate": 0.0003629103815439219,
      "loss": 0.8027,
      "step": 1550
    },
    {
      "epoch": 8.3,
      "learning_rate": 0.00036202307009760426,
      "loss": 0.5895,
      "step": 1560
    },
    {
      "epoch": 8.3,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7850357890129089,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1562,
      "eval_samples_per_second": 243.656,
      "eval_steps_per_second": 7.635,
      "step": 1560
    },
    {
      "epoch": 8.35,
      "learning_rate": 0.0003611357586512866,
      "loss": 0.8055,
      "step": 1570
    },
    {
      "epoch": 8.4,
      "learning_rate": 0.000360248447204969,
      "loss": 0.8404,
      "step": 1580
    },
    {
      "epoch": 8.4,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8316419124603271,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.001,
      "eval_samples_per_second": 249.958,
      "eval_steps_per_second": 7.832,
      "step": 1580
    },
    {
      "epoch": 8.46,
      "learning_rate": 0.0003593611357586513,
      "loss": 0.7572,
      "step": 1590
    },
    {
      "epoch": 8.51,
      "learning_rate": 0.00035847382431233366,
      "loss": 0.7666,
      "step": 1600
    },
    {
      "epoch": 8.51,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8102498650550842,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0058,
      "eval_samples_per_second": 249.759,
      "eval_steps_per_second": 7.826,
      "step": 1600
    },
    {
      "epoch": 8.56,
      "learning_rate": 0.00035758651286601595,
      "loss": 0.8272,
      "step": 1610
    },
    {
      "epoch": 8.62,
      "learning_rate": 0.0003566992014196983,
      "loss": 0.8208,
      "step": 1620
    },
    {
      "epoch": 8.62,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7474777102470398,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0376,
      "eval_samples_per_second": 248.442,
      "eval_steps_per_second": 7.785,
      "step": 1620
    },
    {
      "epoch": 8.67,
      "learning_rate": 0.0003558118899733807,
      "loss": 0.8661,
      "step": 1630
    },
    {
      "epoch": 8.72,
      "learning_rate": 0.000354924578527063,
      "loss": 0.6448,
      "step": 1640
    },
    {
      "epoch": 8.72,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7620306015014648,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9919,
      "eval_samples_per_second": 250.338,
      "eval_steps_per_second": 7.844,
      "step": 1640
    },
    {
      "epoch": 8.78,
      "learning_rate": 0.00035403726708074535,
      "loss": 0.8122,
      "step": 1650
    },
    {
      "epoch": 8.83,
      "learning_rate": 0.0003531499556344277,
      "loss": 0.8219,
      "step": 1660
    },
    {
      "epoch": 8.83,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7586666941642761,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9752,
      "eval_samples_per_second": 251.038,
      "eval_steps_per_second": 7.866,
      "step": 1660
    },
    {
      "epoch": 8.88,
      "learning_rate": 0.00035226264418811,
      "loss": 0.9411,
      "step": 1670
    },
    {
      "epoch": 8.94,
      "learning_rate": 0.0003513753327417924,
      "loss": 0.8689,
      "step": 1680
    },
    {
      "epoch": 8.94,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7479954361915588,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9816,
      "eval_samples_per_second": 250.768,
      "eval_steps_per_second": 7.857,
      "step": 1680
    },
    {
      "epoch": 8.99,
      "learning_rate": 0.00035048802129547475,
      "loss": 0.87,
      "step": 1690
    },
    {
      "epoch": 9.04,
      "learning_rate": 0.00034960070984915703,
      "loss": 0.6917,
      "step": 1700
    },
    {
      "epoch": 9.04,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7611201405525208,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0339,
      "eval_samples_per_second": 248.597,
      "eval_steps_per_second": 7.789,
      "step": 1700
    },
    {
      "epoch": 9.1,
      "learning_rate": 0.00034871339840283937,
      "loss": 0.7082,
      "step": 1710
    },
    {
      "epoch": 9.15,
      "learning_rate": 0.00034782608695652176,
      "loss": 0.8686,
      "step": 1720
    },
    {
      "epoch": 9.15,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7557721138000488,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2391,
      "eval_samples_per_second": 240.42,
      "eval_steps_per_second": 7.533,
      "step": 1720
    },
    {
      "epoch": 9.2,
      "learning_rate": 0.0003469387755102041,
      "loss": 0.6791,
      "step": 1730
    },
    {
      "epoch": 9.26,
      "learning_rate": 0.00034605146406388644,
      "loss": 0.8685,
      "step": 1740
    },
    {
      "epoch": 9.26,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7630966901779175,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2692,
      "eval_samples_per_second": 239.264,
      "eval_steps_per_second": 7.497,
      "step": 1740
    },
    {
      "epoch": 9.31,
      "learning_rate": 0.0003451641526175688,
      "loss": 0.7924,
      "step": 1750
    },
    {
      "epoch": 9.36,
      "learning_rate": 0.0003442768411712511,
      "loss": 0.7709,
      "step": 1760
    },
    {
      "epoch": 9.36,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7447947859764099,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0155,
      "eval_samples_per_second": 249.356,
      "eval_steps_per_second": 7.813,
      "step": 1760
    },
    {
      "epoch": 9.41,
      "learning_rate": 0.0003433895297249335,
      "loss": 0.5728,
      "step": 1770
    },
    {
      "epoch": 9.47,
      "learning_rate": 0.0003425022182786158,
      "loss": 0.7227,
      "step": 1780
    },
    {
      "epoch": 9.47,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8022724390029907,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9833,
      "eval_samples_per_second": 250.699,
      "eval_steps_per_second": 7.855,
      "step": 1780
    },
    {
      "epoch": 9.52,
      "learning_rate": 0.0003416149068322981,
      "loss": 0.906,
      "step": 1790
    },
    {
      "epoch": 9.57,
      "learning_rate": 0.00034072759538598046,
      "loss": 0.9409,
      "step": 1800
    },
    {
      "epoch": 9.57,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7532926201820374,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9683,
      "eval_samples_per_second": 251.328,
      "eval_steps_per_second": 7.875,
      "step": 1800
    },
    {
      "epoch": 9.63,
      "learning_rate": 0.0003398402839396628,
      "loss": 0.7015,
      "step": 1810
    },
    {
      "epoch": 9.68,
      "learning_rate": 0.0003389529724933452,
      "loss": 1.0591,
      "step": 1820
    },
    {
      "epoch": 9.68,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7541568875312805,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9742,
      "eval_samples_per_second": 251.08,
      "eval_steps_per_second": 7.867,
      "step": 1820
    },
    {
      "epoch": 9.73,
      "learning_rate": 0.0003380656610470275,
      "loss": 0.6827,
      "step": 1830
    },
    {
      "epoch": 9.79,
      "learning_rate": 0.00033717834960070986,
      "loss": 0.839,
      "step": 1840
    },
    {
      "epoch": 9.79,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7814498543739319,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9684,
      "eval_samples_per_second": 251.322,
      "eval_steps_per_second": 7.875,
      "step": 1840
    },
    {
      "epoch": 9.84,
      "learning_rate": 0.0003362910381543922,
      "loss": 0.8837,
      "step": 1850
    },
    {
      "epoch": 9.89,
      "learning_rate": 0.0003354037267080746,
      "loss": 0.8196,
      "step": 1860
    },
    {
      "epoch": 9.89,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7539615631103516,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9945,
      "eval_samples_per_second": 250.228,
      "eval_steps_per_second": 7.84,
      "step": 1860
    },
    {
      "epoch": 9.95,
      "learning_rate": 0.00033451641526175687,
      "loss": 0.6146,
      "step": 1870
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0003336291038154392,
      "loss": 0.7688,
      "step": 1880
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8234023451805115,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0496,
      "eval_samples_per_second": 247.95,
      "eval_steps_per_second": 7.769,
      "step": 1880
    },
    {
      "epoch": 10.05,
      "learning_rate": 0.00033274179236912155,
      "loss": 0.7267,
      "step": 1890
    },
    {
      "epoch": 10.11,
      "learning_rate": 0.0003318544809228039,
      "loss": 0.7265,
      "step": 1900
    },
    {
      "epoch": 10.11,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7481959462165833,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2575,
      "eval_samples_per_second": 239.713,
      "eval_steps_per_second": 7.511,
      "step": 1900
    },
    {
      "epoch": 10.16,
      "learning_rate": 0.0003309671694764863,
      "loss": 0.667,
      "step": 1910
    },
    {
      "epoch": 10.21,
      "learning_rate": 0.0003300798580301686,
      "loss": 0.9785,
      "step": 1920
    },
    {
      "epoch": 10.21,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7461732029914856,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2167,
      "eval_samples_per_second": 241.284,
      "eval_steps_per_second": 7.56,
      "step": 1920
    },
    {
      "epoch": 10.27,
      "learning_rate": 0.00032919254658385095,
      "loss": 0.8112,
      "step": 1930
    },
    {
      "epoch": 10.32,
      "learning_rate": 0.0003283052351375333,
      "loss": 0.8718,
      "step": 1940
    },
    {
      "epoch": 10.32,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7645600438117981,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2468,
      "eval_samples_per_second": 240.122,
      "eval_steps_per_second": 7.524,
      "step": 1940
    },
    {
      "epoch": 10.37,
      "learning_rate": 0.0003274179236912156,
      "loss": 0.8389,
      "step": 1950
    },
    {
      "epoch": 10.43,
      "learning_rate": 0.00032653061224489796,
      "loss": 0.8501,
      "step": 1960
    },
    {
      "epoch": 10.43,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7544277310371399,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1932,
      "eval_samples_per_second": 242.202,
      "eval_steps_per_second": 7.589,
      "step": 1960
    },
    {
      "epoch": 10.48,
      "learning_rate": 0.0003256433007985803,
      "loss": 0.7671,
      "step": 1970
    },
    {
      "epoch": 10.53,
      "learning_rate": 0.00032475598935226263,
      "loss": 0.7454,
      "step": 1980
    },
    {
      "epoch": 10.53,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.789313793182373,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9911,
      "eval_samples_per_second": 250.371,
      "eval_steps_per_second": 7.845,
      "step": 1980
    },
    {
      "epoch": 10.59,
      "learning_rate": 0.00032386867790594497,
      "loss": 0.5969,
      "step": 1990
    },
    {
      "epoch": 10.64,
      "learning_rate": 0.0003229813664596273,
      "loss": 0.7165,
      "step": 2000
    },
    {
      "epoch": 10.64,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7718906402587891,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1745,
      "eval_samples_per_second": 242.934,
      "eval_steps_per_second": 7.612,
      "step": 2000
    },
    {
      "epoch": 10.69,
      "learning_rate": 0.0003220940550133097,
      "loss": 0.8527,
      "step": 2010
    },
    {
      "epoch": 10.74,
      "learning_rate": 0.00032120674356699204,
      "loss": 0.7001,
      "step": 2020
    },
    {
      "epoch": 10.74,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7477550506591797,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1635,
      "eval_samples_per_second": 243.367,
      "eval_steps_per_second": 7.625,
      "step": 2020
    },
    {
      "epoch": 10.8,
      "learning_rate": 0.00032040816326530617,
      "loss": 0.7705,
      "step": 2030
    },
    {
      "epoch": 10.85,
      "learning_rate": 0.0003195208518189885,
      "loss": 0.5604,
      "step": 2040
    },
    {
      "epoch": 10.85,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7897444367408752,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9586,
      "eval_samples_per_second": 251.738,
      "eval_steps_per_second": 7.888,
      "step": 2040
    },
    {
      "epoch": 10.9,
      "learning_rate": 0.00031863354037267084,
      "loss": 0.952,
      "step": 2050
    },
    {
      "epoch": 10.96,
      "learning_rate": 0.0003177462289263531,
      "loss": 0.7915,
      "step": 2060
    },
    {
      "epoch": 10.96,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7844055891036987,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.966,
      "eval_samples_per_second": 251.427,
      "eval_steps_per_second": 7.878,
      "step": 2060
    },
    {
      "epoch": 11.01,
      "learning_rate": 0.00031685891748003546,
      "loss": 0.7949,
      "step": 2070
    },
    {
      "epoch": 11.06,
      "learning_rate": 0.00031597160603371785,
      "loss": 0.8174,
      "step": 2080
    },
    {
      "epoch": 11.06,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7565051913261414,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9628,
      "eval_samples_per_second": 251.56,
      "eval_steps_per_second": 7.882,
      "step": 2080
    },
    {
      "epoch": 11.12,
      "learning_rate": 0.0003150842945874002,
      "loss": 0.7171,
      "step": 2090
    },
    {
      "epoch": 11.17,
      "learning_rate": 0.0003141969831410825,
      "loss": 0.8376,
      "step": 2100
    },
    {
      "epoch": 11.17,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.750295877456665,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0042,
      "eval_samples_per_second": 249.826,
      "eval_steps_per_second": 7.828,
      "step": 2100
    },
    {
      "epoch": 11.22,
      "learning_rate": 0.00031330967169476486,
      "loss": 0.7579,
      "step": 2110
    },
    {
      "epoch": 11.28,
      "learning_rate": 0.0003124223602484472,
      "loss": 0.6963,
      "step": 2120
    },
    {
      "epoch": 11.28,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.751220703125,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9834,
      "eval_samples_per_second": 250.693,
      "eval_steps_per_second": 7.855,
      "step": 2120
    },
    {
      "epoch": 11.33,
      "learning_rate": 0.0003115350488021296,
      "loss": 0.7289,
      "step": 2130
    },
    {
      "epoch": 11.38,
      "learning_rate": 0.00031064773735581193,
      "loss": 0.8378,
      "step": 2140
    },
    {
      "epoch": 11.38,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7463984489440918,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9745,
      "eval_samples_per_second": 251.067,
      "eval_steps_per_second": 7.867,
      "step": 2140
    },
    {
      "epoch": 11.44,
      "learning_rate": 0.0003097604259094942,
      "loss": 0.6986,
      "step": 2150
    },
    {
      "epoch": 11.49,
      "learning_rate": 0.00030887311446317655,
      "loss": 0.8198,
      "step": 2160
    },
    {
      "epoch": 11.49,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7484407424926758,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1214,
      "eval_samples_per_second": 245.041,
      "eval_steps_per_second": 7.678,
      "step": 2160
    },
    {
      "epoch": 11.54,
      "learning_rate": 0.0003079858030168589,
      "loss": 0.7262,
      "step": 2170
    },
    {
      "epoch": 11.6,
      "learning_rate": 0.0003070984915705413,
      "loss": 0.7191,
      "step": 2180
    },
    {
      "epoch": 11.6,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8106263279914856,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1373,
      "eval_samples_per_second": 244.409,
      "eval_steps_per_second": 7.658,
      "step": 2180
    },
    {
      "epoch": 11.65,
      "learning_rate": 0.0003062111801242236,
      "loss": 0.6937,
      "step": 2190
    },
    {
      "epoch": 11.7,
      "learning_rate": 0.00030532386867790595,
      "loss": 0.7963,
      "step": 2200
    },
    {
      "epoch": 11.7,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7488634586334229,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1573,
      "eval_samples_per_second": 243.613,
      "eval_steps_per_second": 7.633,
      "step": 2200
    },
    {
      "epoch": 11.76,
      "learning_rate": 0.0003044365572315883,
      "loss": 0.83,
      "step": 2210
    },
    {
      "epoch": 11.81,
      "learning_rate": 0.0003035492457852707,
      "loss": 0.9264,
      "step": 2220
    },
    {
      "epoch": 11.81,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7818997502326965,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2194,
      "eval_samples_per_second": 241.181,
      "eval_steps_per_second": 7.557,
      "step": 2220
    },
    {
      "epoch": 11.86,
      "learning_rate": 0.000302661934338953,
      "loss": 0.7467,
      "step": 2230
    },
    {
      "epoch": 11.91,
      "learning_rate": 0.0003017746228926353,
      "loss": 0.8076,
      "step": 2240
    },
    {
      "epoch": 11.91,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8005937337875366,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1952,
      "eval_samples_per_second": 242.124,
      "eval_steps_per_second": 7.587,
      "step": 2240
    },
    {
      "epoch": 11.97,
      "learning_rate": 0.00030088731144631764,
      "loss": 0.7122,
      "step": 2250
    },
    {
      "epoch": 12.02,
      "learning_rate": 0.0003,
      "loss": 0.738,
      "step": 2260
    },
    {
      "epoch": 12.02,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8040432333946228,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1051,
      "eval_samples_per_second": 245.696,
      "eval_steps_per_second": 7.698,
      "step": 2260
    },
    {
      "epoch": 12.07,
      "learning_rate": 0.00029911268855368236,
      "loss": 0.8552,
      "step": 2270
    },
    {
      "epoch": 12.13,
      "learning_rate": 0.0002982253771073647,
      "loss": 0.7486,
      "step": 2280
    },
    {
      "epoch": 12.13,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7550094127655029,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9627,
      "eval_samples_per_second": 251.566,
      "eval_steps_per_second": 7.882,
      "step": 2280
    },
    {
      "epoch": 12.18,
      "learning_rate": 0.00029733806566104704,
      "loss": 0.7618,
      "step": 2290
    },
    {
      "epoch": 12.23,
      "learning_rate": 0.0002964507542147294,
      "loss": 0.8463,
      "step": 2300
    },
    {
      "epoch": 12.23,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7614899277687073,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9421,
      "eval_samples_per_second": 252.437,
      "eval_steps_per_second": 7.91,
      "step": 2300
    },
    {
      "epoch": 12.29,
      "learning_rate": 0.0002955634427684117,
      "loss": 0.8668,
      "step": 2310
    },
    {
      "epoch": 12.34,
      "learning_rate": 0.0002946761313220941,
      "loss": 0.8266,
      "step": 2320
    },
    {
      "epoch": 12.34,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7446946501731873,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9489,
      "eval_samples_per_second": 252.147,
      "eval_steps_per_second": 7.901,
      "step": 2320
    },
    {
      "epoch": 12.39,
      "learning_rate": 0.0002937888198757764,
      "loss": 0.5528,
      "step": 2330
    },
    {
      "epoch": 12.45,
      "learning_rate": 0.0002929015084294587,
      "loss": 0.8261,
      "step": 2340
    },
    {
      "epoch": 12.45,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8049609661102295,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9564,
      "eval_samples_per_second": 251.832,
      "eval_steps_per_second": 7.891,
      "step": 2340
    },
    {
      "epoch": 12.5,
      "learning_rate": 0.00029201419698314106,
      "loss": 0.7592,
      "step": 2350
    },
    {
      "epoch": 12.55,
      "learning_rate": 0.00029112688553682345,
      "loss": 0.7599,
      "step": 2360
    },
    {
      "epoch": 12.55,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.748624324798584,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.938,
      "eval_samples_per_second": 252.611,
      "eval_steps_per_second": 7.915,
      "step": 2360
    },
    {
      "epoch": 12.61,
      "learning_rate": 0.0002902395740905058,
      "loss": 0.6873,
      "step": 2370
    },
    {
      "epoch": 12.66,
      "learning_rate": 0.0002893522626441881,
      "loss": 0.8146,
      "step": 2380
    },
    {
      "epoch": 12.66,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7453404664993286,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.962,
      "eval_samples_per_second": 251.594,
      "eval_steps_per_second": 7.883,
      "step": 2380
    },
    {
      "epoch": 12.71,
      "learning_rate": 0.00028846495119787046,
      "loss": 1.0175,
      "step": 2390
    },
    {
      "epoch": 12.77,
      "learning_rate": 0.0002875776397515528,
      "loss": 0.7378,
      "step": 2400
    },
    {
      "epoch": 12.77,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7540510892868042,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9518,
      "eval_samples_per_second": 252.023,
      "eval_steps_per_second": 7.897,
      "step": 2400
    },
    {
      "epoch": 12.82,
      "learning_rate": 0.0002866903283052352,
      "loss": 0.7038,
      "step": 2410
    },
    {
      "epoch": 12.87,
      "learning_rate": 0.0002858030168589175,
      "loss": 0.7705,
      "step": 2420
    },
    {
      "epoch": 12.87,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7760998010635376,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9477,
      "eval_samples_per_second": 252.199,
      "eval_steps_per_second": 7.902,
      "step": 2420
    },
    {
      "epoch": 12.93,
      "learning_rate": 0.0002849157054125998,
      "loss": 0.7126,
      "step": 2430
    },
    {
      "epoch": 12.98,
      "learning_rate": 0.00028402839396628215,
      "loss": 0.7215,
      "step": 2440
    },
    {
      "epoch": 12.98,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.756435215473175,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9606,
      "eval_samples_per_second": 251.651,
      "eval_steps_per_second": 7.885,
      "step": 2440
    },
    {
      "epoch": 13.03,
      "learning_rate": 0.0002831410825199645,
      "loss": 0.6946,
      "step": 2450
    },
    {
      "epoch": 13.09,
      "learning_rate": 0.0002822537710736469,
      "loss": 0.7055,
      "step": 2460
    },
    {
      "epoch": 13.09,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7603118419647217,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0008,
      "eval_samples_per_second": 249.968,
      "eval_steps_per_second": 7.832,
      "step": 2460
    },
    {
      "epoch": 13.14,
      "learning_rate": 0.0002813664596273292,
      "loss": 0.7502,
      "step": 2470
    },
    {
      "epoch": 13.19,
      "learning_rate": 0.00028047914818101155,
      "loss": 0.8595,
      "step": 2480
    },
    {
      "epoch": 13.19,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7446478009223938,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.949,
      "eval_samples_per_second": 252.142,
      "eval_steps_per_second": 7.9,
      "step": 2480
    },
    {
      "epoch": 13.24,
      "learning_rate": 0.0002795918367346939,
      "loss": 0.763,
      "step": 2490
    },
    {
      "epoch": 13.3,
      "learning_rate": 0.00027870452528837617,
      "loss": 0.686,
      "step": 2500
    },
    {
      "epoch": 13.3,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7445690035820007,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9445,
      "eval_samples_per_second": 252.332,
      "eval_steps_per_second": 7.906,
      "step": 2500
    },
    {
      "epoch": 13.35,
      "learning_rate": 0.00027781721384205856,
      "loss": 0.838,
      "step": 2510
    },
    {
      "epoch": 13.4,
      "learning_rate": 0.0002769299023957409,
      "loss": 0.7186,
      "step": 2520
    },
    {
      "epoch": 13.4,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7475804090499878,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9738,
      "eval_samples_per_second": 251.097,
      "eval_steps_per_second": 7.868,
      "step": 2520
    },
    {
      "epoch": 13.46,
      "learning_rate": 0.00027604259094942324,
      "loss": 0.9647,
      "step": 2530
    },
    {
      "epoch": 13.51,
      "learning_rate": 0.0002751552795031056,
      "loss": 0.7071,
      "step": 2540
    },
    {
      "epoch": 13.51,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7455338835716248,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9446,
      "eval_samples_per_second": 252.33,
      "eval_steps_per_second": 7.906,
      "step": 2540
    },
    {
      "epoch": 13.56,
      "learning_rate": 0.00027426796805678796,
      "loss": 0.6762,
      "step": 2550
    },
    {
      "epoch": 13.62,
      "learning_rate": 0.0002733806566104703,
      "loss": 0.9072,
      "step": 2560
    },
    {
      "epoch": 13.62,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7543990612030029,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9512,
      "eval_samples_per_second": 252.051,
      "eval_steps_per_second": 7.898,
      "step": 2560
    },
    {
      "epoch": 13.67,
      "learning_rate": 0.00027249334516415264,
      "loss": 0.8409,
      "step": 2570
    },
    {
      "epoch": 13.72,
      "learning_rate": 0.000271606033717835,
      "loss": 0.6846,
      "step": 2580
    },
    {
      "epoch": 13.72,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7723027467727661,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0229,
      "eval_samples_per_second": 249.05,
      "eval_steps_per_second": 7.804,
      "step": 2580
    },
    {
      "epoch": 13.78,
      "learning_rate": 0.00027071872227151726,
      "loss": 0.869,
      "step": 2590
    },
    {
      "epoch": 13.83,
      "learning_rate": 0.00026983141082519965,
      "loss": 0.9324,
      "step": 2600
    },
    {
      "epoch": 13.83,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7549225091934204,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9967,
      "eval_samples_per_second": 250.14,
      "eval_steps_per_second": 7.838,
      "step": 2600
    },
    {
      "epoch": 13.88,
      "learning_rate": 0.000268944099378882,
      "loss": 0.7796,
      "step": 2610
    },
    {
      "epoch": 13.94,
      "learning_rate": 0.0002680567879325643,
      "loss": 0.7263,
      "step": 2620
    },
    {
      "epoch": 13.94,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7473235726356506,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0285,
      "eval_samples_per_second": 248.818,
      "eval_steps_per_second": 7.796,
      "step": 2620
    },
    {
      "epoch": 13.99,
      "learning_rate": 0.00026716947648624666,
      "loss": 0.6593,
      "step": 2630
    },
    {
      "epoch": 14.04,
      "learning_rate": 0.000266282165039929,
      "loss": 0.8613,
      "step": 2640
    },
    {
      "epoch": 14.04,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7636823058128357,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.119,
      "eval_samples_per_second": 245.137,
      "eval_steps_per_second": 7.681,
      "step": 2640
    },
    {
      "epoch": 14.1,
      "learning_rate": 0.0002653948535936114,
      "loss": 0.7282,
      "step": 2650
    },
    {
      "epoch": 14.15,
      "learning_rate": 0.0002645075421472937,
      "loss": 0.7996,
      "step": 2660
    },
    {
      "epoch": 14.15,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7555644512176514,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1633,
      "eval_samples_per_second": 243.375,
      "eval_steps_per_second": 7.626,
      "step": 2660
    },
    {
      "epoch": 14.2,
      "learning_rate": 0.00026362023070097606,
      "loss": 0.8257,
      "step": 2670
    },
    {
      "epoch": 14.26,
      "learning_rate": 0.00026273291925465835,
      "loss": 0.775,
      "step": 2680
    },
    {
      "epoch": 14.26,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7547734379768372,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.967,
      "eval_samples_per_second": 251.383,
      "eval_steps_per_second": 7.877,
      "step": 2680
    },
    {
      "epoch": 14.31,
      "learning_rate": 0.00026184560780834074,
      "loss": 0.7809,
      "step": 2690
    },
    {
      "epoch": 14.36,
      "learning_rate": 0.0002609582963620231,
      "loss": 0.6581,
      "step": 2700
    },
    {
      "epoch": 14.36,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7868841290473938,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0114,
      "eval_samples_per_second": 249.528,
      "eval_steps_per_second": 7.819,
      "step": 2700
    },
    {
      "epoch": 14.41,
      "learning_rate": 0.0002600709849157054,
      "loss": 0.9386,
      "step": 2710
    },
    {
      "epoch": 14.47,
      "learning_rate": 0.00025918367346938775,
      "loss": 0.7711,
      "step": 2720
    },
    {
      "epoch": 14.47,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.757413923740387,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2156,
      "eval_samples_per_second": 241.328,
      "eval_steps_per_second": 7.562,
      "step": 2720
    },
    {
      "epoch": 14.52,
      "learning_rate": 0.0002582963620230701,
      "loss": 0.8136,
      "step": 2730
    },
    {
      "epoch": 14.57,
      "learning_rate": 0.0002574090505767525,
      "loss": 0.8535,
      "step": 2740
    },
    {
      "epoch": 14.57,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7570397257804871,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9416,
      "eval_samples_per_second": 252.457,
      "eval_steps_per_second": 7.91,
      "step": 2740
    },
    {
      "epoch": 14.63,
      "learning_rate": 0.0002565217391304348,
      "loss": 0.7214,
      "step": 2750
    },
    {
      "epoch": 14.68,
      "learning_rate": 0.00025563442768411715,
      "loss": 0.8601,
      "step": 2760
    },
    {
      "epoch": 14.68,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7461080551147461,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1807,
      "eval_samples_per_second": 242.692,
      "eval_steps_per_second": 7.604,
      "step": 2760
    },
    {
      "epoch": 14.73,
      "learning_rate": 0.00025474711623779943,
      "loss": 0.756,
      "step": 2770
    },
    {
      "epoch": 14.79,
      "learning_rate": 0.00025385980479148177,
      "loss": 0.6971,
      "step": 2780
    },
    {
      "epoch": 14.79,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7542539238929749,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0212,
      "eval_samples_per_second": 249.121,
      "eval_steps_per_second": 7.806,
      "step": 2780
    },
    {
      "epoch": 14.84,
      "learning_rate": 0.00025297249334516416,
      "loss": 0.8085,
      "step": 2790
    },
    {
      "epoch": 14.89,
      "learning_rate": 0.0002520851818988465,
      "loss": 0.6066,
      "step": 2800
    },
    {
      "epoch": 14.89,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7800114154815674,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1051,
      "eval_samples_per_second": 245.697,
      "eval_steps_per_second": 7.698,
      "step": 2800
    },
    {
      "epoch": 14.95,
      "learning_rate": 0.00025119787045252884,
      "loss": 0.7488,
      "step": 2810
    },
    {
      "epoch": 15.0,
      "learning_rate": 0.0002503105590062112,
      "loss": 0.7273,
      "step": 2820
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7626035213470459,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9645,
      "eval_samples_per_second": 251.489,
      "eval_steps_per_second": 7.88,
      "step": 2820
    },
    {
      "epoch": 15.05,
      "learning_rate": 0.00024942324755989356,
      "loss": 0.7263,
      "step": 2830
    },
    {
      "epoch": 15.11,
      "learning_rate": 0.00024853593611357585,
      "loss": 0.6408,
      "step": 2840
    },
    {
      "epoch": 15.11,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.757431149482727,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9894,
      "eval_samples_per_second": 250.444,
      "eval_steps_per_second": 7.847,
      "step": 2840
    },
    {
      "epoch": 15.16,
      "learning_rate": 0.0002476486246672582,
      "loss": 0.7053,
      "step": 2850
    },
    {
      "epoch": 15.21,
      "learning_rate": 0.0002467613132209406,
      "loss": 0.8312,
      "step": 2860
    },
    {
      "epoch": 15.21,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.755646824836731,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9635,
      "eval_samples_per_second": 251.531,
      "eval_steps_per_second": 7.881,
      "step": 2860
    },
    {
      "epoch": 15.27,
      "learning_rate": 0.0002458740017746229,
      "loss": 0.7453,
      "step": 2870
    },
    {
      "epoch": 15.32,
      "learning_rate": 0.00024498669032830525,
      "loss": 0.8348,
      "step": 2880
    },
    {
      "epoch": 15.32,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7451900839805603,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9416,
      "eval_samples_per_second": 252.456,
      "eval_steps_per_second": 7.91,
      "step": 2880
    },
    {
      "epoch": 15.37,
      "learning_rate": 0.0002440993788819876,
      "loss": 1.005,
      "step": 2890
    },
    {
      "epoch": 15.43,
      "learning_rate": 0.00024321206743566992,
      "loss": 0.9043,
      "step": 2900
    },
    {
      "epoch": 15.43,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.757602870464325,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.958,
      "eval_samples_per_second": 251.764,
      "eval_steps_per_second": 7.889,
      "step": 2900
    },
    {
      "epoch": 15.48,
      "learning_rate": 0.00024232475598935226,
      "loss": 0.7133,
      "step": 2910
    },
    {
      "epoch": 15.53,
      "learning_rate": 0.00024143744454303463,
      "loss": 0.7276,
      "step": 2920
    },
    {
      "epoch": 15.53,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7593382000923157,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.2356,
      "eval_samples_per_second": 240.553,
      "eval_steps_per_second": 7.537,
      "step": 2920
    },
    {
      "epoch": 15.59,
      "learning_rate": 0.00024055013309671694,
      "loss": 0.7108,
      "step": 2930
    },
    {
      "epoch": 15.64,
      "learning_rate": 0.0002396628216503993,
      "loss": 0.8336,
      "step": 2940
    },
    {
      "epoch": 15.64,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.745688796043396,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9489,
      "eval_samples_per_second": 252.148,
      "eval_steps_per_second": 7.901,
      "step": 2940
    },
    {
      "epoch": 15.69,
      "learning_rate": 0.00023877551020408164,
      "loss": 0.7997,
      "step": 2950
    },
    {
      "epoch": 15.74,
      "learning_rate": 0.000237888198757764,
      "loss": 0.775,
      "step": 2960
    },
    {
      "epoch": 15.74,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7659602761268616,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9877,
      "eval_samples_per_second": 250.514,
      "eval_steps_per_second": 7.849,
      "step": 2960
    },
    {
      "epoch": 15.8,
      "learning_rate": 0.0002370008873114463,
      "loss": 0.7057,
      "step": 2970
    },
    {
      "epoch": 15.85,
      "learning_rate": 0.00023611357586512865,
      "loss": 0.7903,
      "step": 2980
    },
    {
      "epoch": 15.85,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7482553720474243,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9849,
      "eval_samples_per_second": 250.632,
      "eval_steps_per_second": 7.853,
      "step": 2980
    },
    {
      "epoch": 15.9,
      "learning_rate": 0.000235226264418811,
      "loss": 0.8058,
      "step": 2990
    },
    {
      "epoch": 15.96,
      "learning_rate": 0.00023433895297249335,
      "loss": 0.8326,
      "step": 3000
    },
    {
      "epoch": 15.96,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7606054544448853,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9684,
      "eval_samples_per_second": 251.325,
      "eval_steps_per_second": 7.875,
      "step": 3000
    },
    {
      "epoch": 16.01,
      "learning_rate": 0.00023345164152617569,
      "loss": 0.4943,
      "step": 3010
    },
    {
      "epoch": 16.06,
      "learning_rate": 0.00023256433007985802,
      "loss": 1.0116,
      "step": 3020
    },
    {
      "epoch": 16.06,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8067740797996521,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1344,
      "eval_samples_per_second": 244.522,
      "eval_steps_per_second": 7.662,
      "step": 3020
    },
    {
      "epoch": 16.12,
      "learning_rate": 0.0002316770186335404,
      "loss": 0.8801,
      "step": 3030
    },
    {
      "epoch": 16.17,
      "learning_rate": 0.00023078970718722272,
      "loss": 0.8899,
      "step": 3040
    },
    {
      "epoch": 16.17,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7769070863723755,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.163,
      "eval_samples_per_second": 243.388,
      "eval_steps_per_second": 7.626,
      "step": 3040
    },
    {
      "epoch": 16.22,
      "learning_rate": 0.00022990239574090506,
      "loss": 0.5783,
      "step": 3050
    },
    {
      "epoch": 16.28,
      "learning_rate": 0.0002290150842945874,
      "loss": 0.8161,
      "step": 3060
    },
    {
      "epoch": 16.28,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.8031461834907532,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1927,
      "eval_samples_per_second": 242.22,
      "eval_steps_per_second": 7.59,
      "step": 3060
    },
    {
      "epoch": 16.33,
      "learning_rate": 0.00022812777284826974,
      "loss": 0.8135,
      "step": 3070
    },
    {
      "epoch": 16.38,
      "learning_rate": 0.0002272404614019521,
      "loss": 0.627,
      "step": 3080
    },
    {
      "epoch": 16.38,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7487474083900452,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.1077,
      "eval_samples_per_second": 245.593,
      "eval_steps_per_second": 7.695,
      "step": 3080
    },
    {
      "epoch": 16.44,
      "learning_rate": 0.00022635314995563444,
      "loss": 0.8599,
      "step": 3090
    },
    {
      "epoch": 16.49,
      "learning_rate": 0.00022546583850931677,
      "loss": 0.7,
      "step": 3100
    },
    {
      "epoch": 16.49,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7545273303985596,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.107,
      "eval_samples_per_second": 245.618,
      "eval_steps_per_second": 7.696,
      "step": 3100
    },
    {
      "epoch": 16.54,
      "learning_rate": 0.0002245785270629991,
      "loss": 0.6884,
      "step": 3110
    },
    {
      "epoch": 16.6,
      "learning_rate": 0.00022369121561668145,
      "loss": 0.808,
      "step": 3120
    },
    {
      "epoch": 16.6,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7591673135757446,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.122,
      "eval_samples_per_second": 245.016,
      "eval_steps_per_second": 7.677,
      "step": 3120
    },
    {
      "epoch": 16.65,
      "learning_rate": 0.0002228039041703638,
      "loss": 0.8319,
      "step": 3130
    },
    {
      "epoch": 16.7,
      "learning_rate": 0.00022191659272404615,
      "loss": 0.6893,
      "step": 3140
    },
    {
      "epoch": 16.7,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7453346252441406,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9958,
      "eval_samples_per_second": 250.175,
      "eval_steps_per_second": 7.839,
      "step": 3140
    },
    {
      "epoch": 16.76,
      "learning_rate": 0.00022102928127772849,
      "loss": 0.8172,
      "step": 3150
    },
    {
      "epoch": 16.81,
      "learning_rate": 0.00022014196983141082,
      "loss": 0.7099,
      "step": 3160
    },
    {
      "epoch": 16.81,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7497317790985107,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 5.9652,
      "eval_samples_per_second": 251.458,
      "eval_steps_per_second": 7.879,
      "step": 3160
    },
    {
      "epoch": 16.86,
      "learning_rate": 0.0002192546583850932,
      "loss": 0.7704,
      "step": 3170
    },
    {
      "epoch": 16.91,
      "learning_rate": 0.00021836734693877552,
      "loss": 0.6546,
      "step": 3180
    },
    {
      "epoch": 16.91,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7506041526794434,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0263,
      "eval_samples_per_second": 248.911,
      "eval_steps_per_second": 7.799,
      "step": 3180
    },
    {
      "epoch": 16.97,
      "learning_rate": 0.00021748003549245783,
      "loss": 0.8562,
      "step": 3190
    },
    {
      "epoch": 17.02,
      "learning_rate": 0.0002165927240461402,
      "loss": 0.7748,
      "step": 3200
    },
    {
      "epoch": 17.02,
      "eval_accuracy": 0.7386666666666667,
      "eval_combined_score": 0.6626504648943422,
      "eval_f1": 0.6276400817995911,
      "eval_loss": 0.7657597661018372,
      "eval_precision": 0.5456284444444445,
      "eval_recall": 0.7386666666666667,
      "eval_runtime": 6.0294,
      "eval_samples_per_second": 248.783,
      "eval_steps_per_second": 7.795,
      "step": 3200
    }
  ],
  "logging_steps": 10,
  "max_steps": 5640,
  "num_train_epochs": 30,
  "save_steps": 100,
  "total_flos": 6717811546128384.0,
  "trial_name": null,
  "trial_params": null
}