{
  "best_metric": 0.8400702987697716,
  "best_model_checkpoint": "results/qa-suffix-roberta-large-2e-5-32/checkpoint-330",
  "epoch": 3.0,
  "global_step": 363,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 2.702702702702703e-06,
      "loss": 0.656,
      "step": 5
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.6743119266055045,
      "eval_f1": 0.0,
      "eval_loss": 0.638568639755249,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 7.0752,
      "eval_samples_per_second": 123.248,
      "step": 5
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.405405405405406e-06,
      "loss": 0.6251,
      "step": 10
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.6743119266055045,
      "eval_f1": 0.0,
      "eval_loss": 0.6357823610305786,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 6.4162,
      "eval_samples_per_second": 135.907,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.108108108108109e-06,
      "loss": 0.6534,
      "step": 15
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.6743119266055045,
      "eval_f1": 0.0,
      "eval_loss": 0.6329530477523804,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 5.2772,
      "eval_samples_per_second": 165.241,
      "step": 15
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.0810810810810812e-05,
      "loss": 0.6362,
      "step": 20
    },
    {
      "epoch": 0.17,
      "eval_accuracy": 0.6743119266055045,
      "eval_f1": 0.0,
      "eval_loss": 0.62663733959198,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 7.1397,
      "eval_samples_per_second": 122.133,
      "step": 20
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.3513513513513515e-05,
      "loss": 0.6653,
      "step": 25
    },
    {
      "epoch": 0.21,
      "eval_accuracy": 0.6743119266055045,
      "eval_f1": 0.0,
      "eval_loss": 0.614637017250061,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 5.5244,
      "eval_samples_per_second": 157.845,
      "step": 25
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6216216216216218e-05,
      "loss": 0.5946,
      "step": 30
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.6926605504587156,
      "eval_f1": 0.1518987341772152,
      "eval_loss": 0.5719833374023438,
      "eval_precision": 0.75,
      "eval_recall": 0.08450704225352113,
      "eval_runtime": 6.0074,
      "eval_samples_per_second": 145.154,
      "step": 30
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.891891891891892e-05,
      "loss": 0.5597,
      "step": 35
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.783256880733945,
      "eval_f1": 0.5900216919739696,
      "eval_loss": 0.4864228665828705,
      "eval_precision": 0.768361581920904,
      "eval_recall": 0.4788732394366197,
      "eval_runtime": 7.545,
      "eval_samples_per_second": 115.574,
      "step": 35
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.98159509202454e-05,
      "loss": 0.5029,
      "step": 40
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.7259174311926605,
      "eval_f1": 0.31123919308357345,
      "eval_loss": 0.5254184603691101,
      "eval_precision": 0.8571428571428571,
      "eval_recall": 0.19014084507042253,
      "eval_runtime": 7.5551,
      "eval_samples_per_second": 115.418,
      "step": 40
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.9509202453987733e-05,
      "loss": 0.5271,
      "step": 45
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.8360091743119266,
      "eval_f1": 0.7265774378585086,
      "eval_loss": 0.41283223032951355,
      "eval_precision": 0.7949790794979079,
      "eval_recall": 0.6690140845070423,
      "eval_runtime": 6.2573,
      "eval_samples_per_second": 139.357,
      "step": 45
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.9202453987730062e-05,
      "loss": 0.4492,
      "step": 50
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.8532110091743119,
      "eval_f1": 0.7714285714285714,
      "eval_loss": 0.3917687237262726,
      "eval_precision": 0.782608695652174,
      "eval_recall": 0.7605633802816901,
      "eval_runtime": 5.3023,
      "eval_samples_per_second": 164.456,
      "step": 50
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.8895705521472395e-05,
      "loss": 0.4193,
      "step": 55
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.8394495412844036,
      "eval_f1": 0.7188755020080323,
      "eval_loss": 0.35428276658058167,
      "eval_precision": 0.8364485981308412,
      "eval_recall": 0.6302816901408451,
      "eval_runtime": 7.0485,
      "eval_samples_per_second": 123.714,
      "step": 55
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.8588957055214724e-05,
      "loss": 0.369,
      "step": 60
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.8428899082568807,
      "eval_f1": 0.7674023769100169,
      "eval_loss": 0.37413832545280457,
      "eval_precision": 0.740983606557377,
      "eval_recall": 0.795774647887324,
      "eval_runtime": 7.2795,
      "eval_samples_per_second": 119.788,
      "step": 60
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.828220858895706e-05,
      "loss": 0.4063,
      "step": 65
    },
    {
      "epoch": 0.54,
      "eval_accuracy": 0.8096330275229358,
      "eval_f1": 0.7558823529411764,
      "eval_loss": 0.4283739924430847,
      "eval_precision": 0.648989898989899,
      "eval_recall": 0.9049295774647887,
      "eval_runtime": 6.9732,
      "eval_samples_per_second": 125.05,
      "step": 65
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.797546012269939e-05,
      "loss": 0.4608,
      "step": 70
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.8348623853211009,
      "eval_f1": 0.6869565217391304,
      "eval_loss": 0.37825068831443787,
      "eval_precision": 0.8977272727272727,
      "eval_recall": 0.5563380281690141,
      "eval_runtime": 5.0,
      "eval_samples_per_second": 174.401,
      "step": 70
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.766871165644172e-05,
      "loss": 0.4506,
      "step": 75
    },
    {
      "epoch": 0.62,
      "eval_accuracy": 0.8681192660550459,
      "eval_f1": 0.8067226890756302,
      "eval_loss": 0.3225836753845215,
      "eval_precision": 0.7717041800643086,
      "eval_recall": 0.8450704225352113,
      "eval_runtime": 6.6965,
      "eval_samples_per_second": 130.217,
      "step": 75
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.736196319018405e-05,
      "loss": 0.3557,
      "step": 80
    },
    {
      "epoch": 0.66,
      "eval_accuracy": 0.8497706422018348,
      "eval_f1": 0.7883683360258481,
      "eval_loss": 0.34664297103881836,
      "eval_precision": 0.7283582089552239,
      "eval_recall": 0.8591549295774648,
      "eval_runtime": 7.1969,
      "eval_samples_per_second": 121.162,
      "step": 80
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.7055214723926382e-05,
      "loss": 0.4147,
      "step": 85
    },
    {
      "epoch": 0.7,
      "eval_accuracy": 0.8600917431192661,
      "eval_f1": 0.7680608365019012,
      "eval_loss": 0.3466174304485321,
      "eval_precision": 0.8347107438016529,
      "eval_recall": 0.7112676056338029,
      "eval_runtime": 7.3038,
      "eval_samples_per_second": 119.39,
      "step": 85
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.6748466257668714e-05,
      "loss": 0.4009,
      "step": 90
    },
    {
      "epoch": 0.74,
      "eval_accuracy": 0.8394495412844036,
      "eval_f1": 0.779874213836478,
      "eval_loss": 0.35522547364234924,
      "eval_precision": 0.7045454545454546,
      "eval_recall": 0.8732394366197183,
      "eval_runtime": 5.9612,
      "eval_samples_per_second": 146.278,
      "step": 90
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.6441717791411043e-05,
      "loss": 0.355,
      "step": 95
    },
    {
      "epoch": 0.79,
      "eval_accuracy": 0.856651376146789,
      "eval_f1": 0.7731397459165154,
      "eval_loss": 0.33357954025268555,
      "eval_precision": 0.797752808988764,
      "eval_recall": 0.75,
      "eval_runtime": 5.1139,
      "eval_samples_per_second": 170.515,
      "step": 95
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.6134969325153376e-05,
      "loss": 0.3783,
      "step": 100
    },
    {
      "epoch": 0.83,
      "eval_accuracy": 0.8543577981651376,
      "eval_f1": 0.7814113597246126,
      "eval_loss": 0.3445069193840027,
      "eval_precision": 0.7643097643097643,
      "eval_recall": 0.7992957746478874,
      "eval_runtime": 7.3177,
      "eval_samples_per_second": 119.163,
      "step": 100
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5828220858895708e-05,
      "loss": 0.3407,
      "step": 105
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.8612385321100917,
      "eval_f1": 0.7858407079646017,
      "eval_loss": 0.3221452236175537,
      "eval_precision": 0.7900355871886121,
      "eval_recall": 0.7816901408450704,
      "eval_runtime": 7.2195,
      "eval_samples_per_second": 120.784,
      "step": 105
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.5521472392638037e-05,
      "loss": 0.3886,
      "step": 110
    },
    {
      "epoch": 0.91,
      "eval_accuracy": 0.8532110091743119,
      "eval_f1": 0.7942122186495175,
      "eval_loss": 0.3414439558982849,
      "eval_precision": 0.7307692307692307,
      "eval_recall": 0.8697183098591549,
      "eval_runtime": 6.7246,
      "eval_samples_per_second": 129.673,
      "step": 110
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.5214723926380371e-05,
      "loss": 0.3325,
      "step": 115
    },
    {
      "epoch": 0.95,
      "eval_accuracy": 0.8646788990825688,
      "eval_f1": 0.7999999999999999,
      "eval_loss": 0.3119550347328186,
      "eval_precision": 0.7712418300653595,
      "eval_recall": 0.8309859154929577,
      "eval_runtime": 5.1226,
      "eval_samples_per_second": 170.227,
      "step": 115
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4907975460122702e-05,
      "loss": 0.3489,
      "step": 120
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.8669724770642202,
      "eval_f1": 0.802721088435374,
      "eval_loss": 0.3048677444458008,
      "eval_precision": 0.7763157894736842,
      "eval_recall": 0.8309859154929577,
      "eval_runtime": 5.8542,
      "eval_samples_per_second": 148.952,
      "step": 120
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4601226993865032e-05,
      "loss": 0.308,
      "step": 125
    },
    {
      "epoch": 1.03,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.8112874779541447,
      "eval_loss": 0.2938537299633026,
      "eval_precision": 0.8127208480565371,
      "eval_recall": 0.8098591549295775,
      "eval_runtime": 7.2512,
      "eval_samples_per_second": 120.256,
      "step": 125
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4294478527607363e-05,
      "loss": 0.3356,
      "step": 130
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.8704128440366973,
      "eval_f1": 0.8027923211169284,
      "eval_loss": 0.3023342192173004,
      "eval_precision": 0.7958477508650519,
      "eval_recall": 0.8098591549295775,
      "eval_runtime": 7.3245,
      "eval_samples_per_second": 119.053,
      "step": 130
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.3987730061349694e-05,
      "loss": 0.3,
      "step": 135
    },
    {
      "epoch": 1.12,
      "eval_accuracy": 0.8727064220183486,
      "eval_f1": 0.8035398230088495,
      "eval_loss": 0.3078989088535309,
      "eval_precision": 0.8078291814946619,
      "eval_recall": 0.7992957746478874,
      "eval_runtime": 6.5413,
      "eval_samples_per_second": 133.306,
      "step": 135
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3680981595092026e-05,
      "loss": 0.3106,
      "step": 140
    },
    {
      "epoch": 1.16,
      "eval_accuracy": 0.8658256880733946,
      "eval_f1": 0.8040201005025125,
      "eval_loss": 0.3164271116256714,
      "eval_precision": 0.7667731629392971,
      "eval_recall": 0.8450704225352113,
      "eval_runtime": 5.1365,
      "eval_samples_per_second": 169.766,
      "step": 140
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3374233128834357e-05,
      "loss": 0.2401,
      "step": 145
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.8761467889908257,
      "eval_f1": 0.8131487889273357,
      "eval_loss": 0.3013850152492523,
      "eval_precision": 0.7993197278911565,
      "eval_recall": 0.8274647887323944,
      "eval_runtime": 5.3284,
      "eval_samples_per_second": 163.653,
      "step": 145
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3067484662576687e-05,
      "loss": 0.3167,
      "step": 150
    },
    {
      "epoch": 1.24,
      "eval_accuracy": 0.856651376146789,
      "eval_f1": 0.7980613893376413,
      "eval_loss": 0.32322707772254944,
      "eval_precision": 0.7373134328358208,
      "eval_recall": 0.8697183098591549,
      "eval_runtime": 7.1641,
      "eval_samples_per_second": 121.717,
      "step": 150
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.276073619631902e-05,
      "loss": 0.3227,
      "step": 155
    },
    {
      "epoch": 1.28,
      "eval_accuracy": 0.8761467889908257,
      "eval_f1": 0.7954545454545454,
      "eval_loss": 0.3056000769138336,
      "eval_precision": 0.860655737704918,
      "eval_recall": 0.7394366197183099,
      "eval_runtime": 7.0886,
      "eval_samples_per_second": 123.014,
      "step": 155
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.245398773006135e-05,
      "loss": 0.3401,
      "step": 160
    },
    {
      "epoch": 1.32,
      "eval_accuracy": 0.8646788990825688,
      "eval_f1": 0.805921052631579,
      "eval_loss": 0.30931735038757324,
      "eval_precision": 0.7561728395061729,
      "eval_recall": 0.8626760563380281,
      "eval_runtime": 7.2974,
      "eval_samples_per_second": 119.494,
      "step": 160
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2147239263803683e-05,
      "loss": 0.272,
      "step": 165
    },
    {
      "epoch": 1.36,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.8158347676419967,
      "eval_loss": 0.2903553247451782,
      "eval_precision": 0.797979797979798,
      "eval_recall": 0.8345070422535211,
      "eval_runtime": 6.7437,
      "eval_samples_per_second": 129.307,
      "step": 165
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.1840490797546013e-05,
      "loss": 0.2701,
      "step": 170
    },
    {
      "epoch": 1.4,
      "eval_accuracy": 0.8795871559633027,
      "eval_f1": 0.8205128205128205,
      "eval_loss": 0.29034143686294556,
      "eval_precision": 0.7973421926910299,
      "eval_recall": 0.8450704225352113,
      "eval_runtime": 5.3852,
      "eval_samples_per_second": 161.925,
      "step": 170
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.1533742331288344e-05,
      "loss": 0.2579,
      "step": 175
    },
    {
      "epoch": 1.45,
      "eval_accuracy": 0.8864678899082569,
      "eval_f1": 0.8241563055062167,
      "eval_loss": 0.28575021028518677,
      "eval_precision": 0.8315412186379928,
      "eval_recall": 0.8169014084507042,
      "eval_runtime": 5.1529,
      "eval_samples_per_second": 169.224,
      "step": 175
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1226993865030675e-05,
      "loss": 0.2523,
      "step": 180
    },
    {
      "epoch": 1.49,
      "eval_accuracy": 0.8864678899082569,
      "eval_f1": 0.8241563055062167,
      "eval_loss": 0.2819433808326721,
      "eval_precision": 0.8315412186379928,
      "eval_recall": 0.8169014084507042,
      "eval_runtime": 6.7646,
      "eval_samples_per_second": 128.907,
      "step": 180
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.0920245398773005e-05,
      "loss": 0.3262,
      "step": 185
    },
    {
      "epoch": 1.53,
      "eval_accuracy": 0.8841743119266054,
      "eval_f1": 0.8206039076376554,
      "eval_loss": 0.27894383668899536,
      "eval_precision": 0.8279569892473119,
      "eval_recall": 0.8133802816901409,
      "eval_runtime": 7.2116,
      "eval_samples_per_second": 120.915,
      "step": 185
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.061349693251534e-05,
      "loss": 0.2973,
      "step": 190
    },
    {
      "epoch": 1.57,
      "eval_accuracy": 0.8795871559633027,
      "eval_f1": 0.8154657293497363,
      "eval_loss": 0.2783905565738678,
      "eval_precision": 0.8140350877192982,
      "eval_recall": 0.8169014084507042,
      "eval_runtime": 7.1139,
      "eval_samples_per_second": 122.577,
      "step": 190
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.030674846625767e-05,
      "loss": 0.3122,
      "step": 195
    },
    {
      "epoch": 1.61,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.8112874779541447,
      "eval_loss": 0.28692659735679626,
      "eval_precision": 0.8127208480565371,
      "eval_recall": 0.8098591549295775,
      "eval_runtime": 7.2602,
      "eval_samples_per_second": 120.107,
      "step": 195
    },
    {
      "epoch": 1.65,
      "learning_rate": 1e-05,
      "loss": 0.3195,
      "step": 200
    },
    {
      "epoch": 1.65,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.8078994614003591,
      "eval_loss": 0.2904263436794281,
      "eval_precision": 0.8241758241758241,
      "eval_recall": 0.7922535211267606,
      "eval_runtime": 6.5577,
      "eval_samples_per_second": 132.972,
      "step": 200
    },
    {
      "epoch": 1.69,
      "learning_rate": 9.693251533742331e-06,
      "loss": 0.2864,
      "step": 205
    },
    {
      "epoch": 1.69,
      "eval_accuracy": 0.8669724770642202,
      "eval_f1": 0.8020477815699658,
      "eval_loss": 0.29111164808273315,
      "eval_precision": 0.7781456953642384,
      "eval_recall": 0.8274647887323944,
      "eval_runtime": 5.8177,
      "eval_samples_per_second": 149.888,
      "step": 205
    },
    {
      "epoch": 1.74,
      "learning_rate": 9.386503067484664e-06,
      "loss": 0.3185,
      "step": 210
    },
    {
      "epoch": 1.74,
      "eval_accuracy": 0.8635321100917431,
      "eval_f1": 0.8006700167504189,
      "eval_loss": 0.29158464074134827,
      "eval_precision": 0.7635782747603834,
      "eval_recall": 0.8415492957746479,
      "eval_runtime": 5.1013,
      "eval_samples_per_second": 170.936,
      "step": 210
    },
    {
      "epoch": 1.78,
      "learning_rate": 9.079754601226994e-06,
      "loss": 0.2743,
      "step": 215
    },
    {
      "epoch": 1.78,
      "eval_accuracy": 0.8704128440366973,
      "eval_f1": 0.804159445407279,
      "eval_loss": 0.2849805951118469,
      "eval_precision": 0.7918088737201365,
      "eval_recall": 0.8169014084507042,
      "eval_runtime": 6.2932,
      "eval_samples_per_second": 138.562,
      "step": 215
    },
    {
      "epoch": 1.82,
      "learning_rate": 8.773006134969327e-06,
      "loss": 0.3158,
      "step": 220
    },
    {
      "epoch": 1.82,
      "eval_accuracy": 0.8669724770642202,
      "eval_f1": 0.8110749185667752,
      "eval_loss": 0.301179438829422,
      "eval_precision": 0.7545454545454545,
      "eval_recall": 0.8767605633802817,
      "eval_runtime": 7.2876,
      "eval_samples_per_second": 119.656,
      "step": 220
    },
    {
      "epoch": 1.86,
      "learning_rate": 8.466257668711658e-06,
      "loss": 0.2714,
      "step": 225
    },
    {
      "epoch": 1.86,
      "eval_accuracy": 0.8853211009174312,
      "eval_f1": 0.8161764705882353,
      "eval_loss": 0.28371983766555786,
      "eval_precision": 0.8538461538461538,
      "eval_recall": 0.7816901408450704,
      "eval_runtime": 7.0861,
      "eval_samples_per_second": 123.058,
      "step": 225
    },
    {
      "epoch": 1.9,
      "learning_rate": 8.159509202453988e-06,
      "loss": 0.2978,
      "step": 230
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.8738532110091743,
      "eval_f1": 0.8141891891891891,
      "eval_loss": 0.2848930060863495,
      "eval_precision": 0.7824675324675324,
      "eval_recall": 0.8485915492957746,
      "eval_runtime": 7.1621,
      "eval_samples_per_second": 121.752,
      "step": 230
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.85276073619632e-06,
      "loss": 0.2857,
      "step": 235
    },
    {
      "epoch": 1.94,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.8189509306260575,
      "eval_loss": 0.28239545226097107,
      "eval_precision": 0.7882736156351792,
      "eval_recall": 0.852112676056338,
      "eval_runtime": 6.4537,
      "eval_samples_per_second": 135.115,
      "step": 235
    },
    {
      "epoch": 1.98,
      "learning_rate": 7.54601226993865e-06,
      "loss": 0.2669,
      "step": 240
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.8876146788990825,
      "eval_f1": 0.8150943396226414,
      "eval_loss": 0.29343676567077637,
      "eval_precision": 0.8780487804878049,
      "eval_recall": 0.7605633802816901,
      "eval_runtime": 5.6057,
      "eval_samples_per_second": 155.557,
      "step": 240
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.239263803680983e-06,
      "loss": 0.2786,
      "step": 245
    },
    {
      "epoch": 2.02,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.8189509306260575,
      "eval_loss": 0.28877732157707214,
      "eval_precision": 0.7882736156351792,
      "eval_recall": 0.852112676056338,
      "eval_runtime": 5.106,
      "eval_samples_per_second": 170.781,
      "step": 245
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.932515337423313e-06,
      "loss": 0.2484,
      "step": 250
    },
    {
      "epoch": 2.07,
      "eval_accuracy": 0.8807339449541285,
      "eval_f1": 0.8237288135593219,
      "eval_loss": 0.285278856754303,
      "eval_precision": 0.7941176470588235,
      "eval_recall": 0.8556338028169014,
      "eval_runtime": 5.1204,
      "eval_samples_per_second": 170.3,
      "step": 250
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.625766871165644e-06,
      "loss": 0.2006,
      "step": 255
    },
    {
      "epoch": 2.11,
      "eval_accuracy": 0.8899082568807339,
      "eval_f1": 0.8241758241758241,
      "eval_loss": 0.2904672622680664,
      "eval_precision": 0.8587786259541985,
      "eval_recall": 0.7922535211267606,
      "eval_runtime": 7.1075,
      "eval_samples_per_second": 122.688,
      "step": 255
    },
    {
      "epoch": 2.15,
      "learning_rate": 6.319018404907976e-06,
      "loss": 0.2564,
      "step": 260
    },
    {
      "epoch": 2.15,
      "eval_accuracy": 0.8807339449541285,
      "eval_f1": 0.8272425249169436,
      "eval_loss": 0.31366848945617676,
      "eval_precision": 0.7830188679245284,
      "eval_recall": 0.8767605633802817,
      "eval_runtime": 7.0515,
      "eval_samples_per_second": 123.661,
      "step": 260
    },
    {
      "epoch": 2.19,
      "learning_rate": 6.012269938650307e-06,
      "loss": 0.2652,
      "step": 265
    },
    {
      "epoch": 2.19,
      "eval_accuracy": 0.8795871559633027,
      "eval_f1": 0.8247078464106845,
      "eval_loss": 0.29686439037323,
      "eval_precision": 0.7841269841269841,
      "eval_recall": 0.8697183098591549,
      "eval_runtime": 7.3289,
      "eval_samples_per_second": 118.981,
      "step": 265
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.7055214723926385e-06,
      "loss": 0.2267,
      "step": 270
    },
    {
      "epoch": 2.23,
      "eval_accuracy": 0.8876146788990825,
      "eval_f1": 0.8256227758007119,
      "eval_loss": 0.2745262682437897,
      "eval_precision": 0.8345323741007195,
      "eval_recall": 0.8169014084507042,
      "eval_runtime": 6.8361,
      "eval_samples_per_second": 127.558,
      "step": 270
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.39877300613497e-06,
      "loss": 0.2089,
      "step": 275
    },
    {
      "epoch": 2.27,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.81195079086116,
      "eval_loss": 0.28398409485816956,
      "eval_precision": 0.8105263157894737,
      "eval_recall": 0.8133802816901409,
      "eval_runtime": 5.1533,
      "eval_samples_per_second": 169.213,
      "step": 275
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.092024539877301e-06,
      "loss": 0.2417,
      "step": 280
    },
    {
      "epoch": 2.31,
      "eval_accuracy": 0.8784403669724771,
      "eval_f1": 0.8239202657807309,
      "eval_loss": 0.30549222230911255,
      "eval_precision": 0.779874213836478,
      "eval_recall": 0.8732394366197183,
      "eval_runtime": 5.0137,
      "eval_samples_per_second": 173.924,
      "step": 280
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.785276073619632e-06,
      "loss": 0.1859,
      "step": 285
    },
    {
      "epoch": 2.36,
      "eval_accuracy": 0.875,
      "eval_f1": 0.8174204355108877,
      "eval_loss": 0.2901022434234619,
      "eval_precision": 0.7795527156549521,
      "eval_recall": 0.8591549295774648,
      "eval_runtime": 5.0546,
      "eval_samples_per_second": 172.515,
      "step": 285
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.478527607361964e-06,
      "loss": 0.162,
      "step": 290
    },
    {
      "epoch": 2.4,
      "eval_accuracy": 0.8841743119266054,
      "eval_f1": 0.8186714542190306,
      "eval_loss": 0.2908870279788971,
      "eval_precision": 0.8351648351648352,
      "eval_recall": 0.8028169014084507,
      "eval_runtime": 5.0477,
      "eval_samples_per_second": 172.752,
      "step": 290
    },
    {
      "epoch": 2.44,
      "learning_rate": 4.171779141104294e-06,
      "loss": 0.2655,
      "step": 295
    },
    {
      "epoch": 2.44,
      "eval_accuracy": 0.8853211009174312,
      "eval_f1": 0.8220640569395018,
      "eval_loss": 0.2885466516017914,
      "eval_precision": 0.8309352517985612,
      "eval_recall": 0.8133802816901409,
      "eval_runtime": 5.0462,
      "eval_samples_per_second": 172.803,
      "step": 295
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.865030674846626e-06,
      "loss": 0.2198,
      "step": 300
    },
    {
      "epoch": 2.48,
      "eval_accuracy": 0.8772935779816514,
      "eval_f1": 0.8183361629881155,
      "eval_loss": 0.2900920510292053,
      "eval_precision": 0.7901639344262295,
      "eval_recall": 0.8485915492957746,
      "eval_runtime": 5.0183,
      "eval_samples_per_second": 173.764,
      "step": 300
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.5582822085889574e-06,
      "loss": 0.1997,
      "step": 305
    },
    {
      "epoch": 2.52,
      "eval_accuracy": 0.875,
      "eval_f1": 0.8174204355108877,
      "eval_loss": 0.29444408416748047,
      "eval_precision": 0.7795527156549521,
      "eval_recall": 0.8591549295774648,
      "eval_runtime": 4.9821,
      "eval_samples_per_second": 175.028,
      "step": 305
    },
    {
      "epoch": 2.56,
      "learning_rate": 3.251533742331289e-06,
      "loss": 0.2164,
      "step": 310
    },
    {
      "epoch": 2.56,
      "eval_accuracy": 0.8876146788990825,
      "eval_f1": 0.8286713286713286,
      "eval_loss": 0.2741115391254425,
      "eval_precision": 0.8229166666666666,
      "eval_recall": 0.8345070422535211,
      "eval_runtime": 5.0084,
      "eval_samples_per_second": 174.109,
      "step": 310
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.94478527607362e-06,
      "loss": 0.2352,
      "step": 315
    },
    {
      "epoch": 2.6,
      "eval_accuracy": 0.8922018348623854,
      "eval_f1": 0.8362369337979093,
      "eval_loss": 0.2667850852012634,
      "eval_precision": 0.8275862068965517,
      "eval_recall": 0.8450704225352113,
      "eval_runtime": 5.0123,
      "eval_samples_per_second": 173.972,
      "step": 315
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.638036809815951e-06,
      "loss": 0.1848,
      "step": 320
    },
    {
      "epoch": 2.64,
      "eval_accuracy": 0.8853211009174312,
      "eval_f1": 0.8310810810810811,
      "eval_loss": 0.2768068015575409,
      "eval_precision": 0.7987012987012987,
      "eval_recall": 0.8661971830985915,
      "eval_runtime": 4.9976,
      "eval_samples_per_second": 174.484,
      "step": 320
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.331288343558282e-06,
      "loss": 0.2405,
      "step": 325
    },
    {
      "epoch": 2.69,
      "eval_accuracy": 0.8887614678899083,
      "eval_f1": 0.8324697754749567,
      "eval_loss": 0.2706994116306305,
      "eval_precision": 0.8169491525423729,
      "eval_recall": 0.8485915492957746,
      "eval_runtime": 4.9789,
      "eval_samples_per_second": 175.138,
      "step": 325
    },
    {
      "epoch": 2.73,
      "learning_rate": 2.0245398773006137e-06,
      "loss": 0.2175,
      "step": 330
    },
    {
      "epoch": 2.73,
      "eval_accuracy": 0.8956422018348624,
      "eval_f1": 0.8400702987697716,
      "eval_loss": 0.2676824629306793,
      "eval_precision": 0.8385964912280702,
      "eval_recall": 0.8415492957746479,
      "eval_runtime": 4.9748,
      "eval_samples_per_second": 175.283,
      "step": 330
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.717791411042945e-06,
      "loss": 0.2392,
      "step": 335
    },
    {
      "epoch": 2.77,
      "eval_accuracy": 0.8864678899082569,
      "eval_f1": 0.82960413080895,
      "eval_loss": 0.2721306383609772,
      "eval_precision": 0.8114478114478114,
      "eval_recall": 0.8485915492957746,
      "eval_runtime": 4.9711,
      "eval_samples_per_second": 175.412,
      "step": 335
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.411042944785276e-06,
      "loss": 0.2155,
      "step": 340
    },
    {
      "epoch": 2.81,
      "eval_accuracy": 0.8818807339449541,
      "eval_f1": 0.8263069139966273,
      "eval_loss": 0.2795032560825348,
      "eval_precision": 0.7928802588996764,
      "eval_recall": 0.8626760563380281,
      "eval_runtime": 4.9888,
      "eval_samples_per_second": 174.791,
      "step": 340
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.1042944785276075e-06,
      "loss": 0.2833,
      "step": 345
    },
    {
      "epoch": 2.85,
      "eval_accuracy": 0.8818807339449541,
      "eval_f1": 0.826890756302521,
      "eval_loss": 0.28040066361427307,
      "eval_precision": 0.7909967845659164,
      "eval_recall": 0.8661971830985915,
      "eval_runtime": 4.9996,
      "eval_samples_per_second": 174.413,
      "step": 345
    },
    {
      "epoch": 2.89,
      "learning_rate": 7.975460122699387e-07,
      "loss": 0.2018,
      "step": 350
    },
    {
      "epoch": 2.89,
      "eval_accuracy": 0.8864678899082569,
      "eval_f1": 0.82960413080895,
      "eval_loss": 0.27195218205451965,
      "eval_precision": 0.8114478114478114,
      "eval_recall": 0.8485915492957746,
      "eval_runtime": 4.9745,
      "eval_samples_per_second": 175.295,
      "step": 350
    },
    {
      "epoch": 2.93,
      "learning_rate": 4.9079754601227e-07,
      "loss": 0.2184,
      "step": 355
    },
    {
      "epoch": 2.93,
      "eval_accuracy": 0.8899082568807339,
      "eval_f1": 0.832752613240418,
      "eval_loss": 0.26814883947372437,
      "eval_precision": 0.8241379310344827,
      "eval_recall": 0.8415492957746479,
      "eval_runtime": 5.1102,
      "eval_samples_per_second": 170.638,
      "step": 355
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.8404907975460125e-07,
      "loss": 0.2232,
      "step": 360
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.8910550458715596,
      "eval_f1": 0.8342059336823734,
      "eval_loss": 0.26773521304130554,
      "eval_precision": 0.8269896193771626,
      "eval_recall": 0.8415492957746479,
      "eval_runtime": 4.9742,
      "eval_samples_per_second": 175.303,
      "step": 360
    },
    {
      "epoch": 3.0,
      "step": 363,
      "total_flos": 1.259652372638976e+16,
      "train_runtime": 1963.3463,
      "train_samples_per_second": 0.185
    }
  ],
  "max_steps": 363,
  "num_train_epochs": 3,
  "total_flos": 1.259652372638976e+16,
  "trial_name": null,
  "trial_params": null
}