{
  "best_metric": 0.9965715706847758,
  "best_model_checkpoint": "output_dev/v48-deberta-v3-large-f1-warmup-greater-patience-subword-len-500-maxlen-512-6epoch/checkpoint-1800",
  "epoch": 6.0,
  "eval_steps": 200,
  "global_step": 3156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 3.1645569620253167e-06,
      "loss": 1.8539,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 6.329113924050633e-06,
      "loss": 1.0295,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.49367088607595e-06,
      "loss": 0.4899,
      "step": 30
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.2658227848101267e-05,
      "loss": 0.2892,
      "step": 40
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.5822784810126583e-05,
      "loss": 0.1988,
      "step": 50
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.89873417721519e-05,
      "loss": 0.1417,
      "step": 60
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.2151898734177217e-05,
      "loss": 0.0904,
      "step": 70
    },
    {
      "epoch": 0.15,
      "learning_rate": 2.5316455696202533e-05,
      "loss": 0.0814,
      "step": 80
    },
    {
      "epoch": 0.17,
      "learning_rate": 2.848101265822785e-05,
      "loss": 0.0335,
      "step": 90
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.1645569620253167e-05,
      "loss": 0.0401,
      "step": 100
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.4810126582278487e-05,
      "loss": 0.0304,
      "step": 110
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.79746835443038e-05,
      "loss": 0.0223,
      "step": 120
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.113924050632912e-05,
      "loss": 0.029,
      "step": 130
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.430379746835443e-05,
      "loss": 0.0312,
      "step": 140
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.7468354430379746e-05,
      "loss": 0.0295,
      "step": 150
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.996664442961975e-05,
      "loss": 0.0315,
      "step": 160
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.979986657771848e-05,
      "loss": 0.027,
      "step": 170
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.963308872581721e-05,
      "loss": 0.021,
      "step": 180
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.946631087391594e-05,
      "loss": 0.0258,
      "step": 190
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.929953302201468e-05,
      "loss": 0.0232,
      "step": 200
    },
    {
      "epoch": 0.38,
      "eval_f1": 0.9829244574920268,
      "eval_f1_0": 0.9946946557701127,
      "eval_f1_1": 0.99571802949886,
      "eval_f1_2": 0.9426792739319484,
      "eval_f1_3": 0.9986058707671863,
      "eval_loss": 0.02559051290154457,
      "eval_runtime": 19.2055,
      "eval_samples_per_second": 389.055,
      "eval_steps_per_second": 0.417,
      "step": 200
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.913275517011341e-05,
      "loss": 0.023,
      "step": 210
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.896597731821214e-05,
      "loss": 0.0145,
      "step": 220
    },
    {
      "epoch": 0.44,
      "learning_rate": 4.879919946631087e-05,
      "loss": 0.0319,
      "step": 230
    },
    {
      "epoch": 0.46,
      "learning_rate": 4.863242161440961e-05,
      "loss": 0.02,
      "step": 240
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.846564376250834e-05,
      "loss": 0.0204,
      "step": 250
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.829886591060707e-05,
      "loss": 0.0217,
      "step": 260
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.813208805870581e-05,
      "loss": 0.0239,
      "step": 270
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.796531020680454e-05,
      "loss": 0.0289,
      "step": 280
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.779853235490327e-05,
      "loss": 0.0298,
      "step": 290
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.7631754503002e-05,
      "loss": 0.0198,
      "step": 300
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.7464976651100736e-05,
      "loss": 0.0313,
      "step": 310
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.729819879919947e-05,
      "loss": 0.0187,
      "step": 320
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.71314209472982e-05,
      "loss": 0.0264,
      "step": 330
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.6964643095396935e-05,
      "loss": 0.0327,
      "step": 340
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.6797865243495666e-05,
      "loss": 0.0181,
      "step": 350
    },
    {
      "epoch": 0.68,
      "learning_rate": 4.6631087391594396e-05,
      "loss": 0.0162,
      "step": 360
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.646430953969313e-05,
      "loss": 0.0178,
      "step": 370
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.6297531687791864e-05,
      "loss": 0.0068,
      "step": 380
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.6130753835890595e-05,
      "loss": 0.0202,
      "step": 390
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.5963975983989326e-05,
      "loss": 0.0261,
      "step": 400
    },
    {
      "epoch": 0.76,
      "eval_f1": 0.9902730835281612,
      "eval_f1_0": 0.9970295053857255,
      "eval_f1_1": 0.9966460102285332,
      "eval_f1_2": 0.968532366748815,
      "eval_f1_3": 0.9988844517495713,
      "eval_loss": 0.01530930120497942,
      "eval_runtime": 19.2644,
      "eval_samples_per_second": 387.867,
      "eval_steps_per_second": 0.415,
      "step": 400
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.5797198132088056e-05,
      "loss": 0.0196,
      "step": 410
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.5630420280186794e-05,
      "loss": 0.0197,
      "step": 420
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.5463642428285524e-05,
      "loss": 0.0213,
      "step": 430
    },
    {
      "epoch": 0.84,
      "learning_rate": 4.5296864576384255e-05,
      "loss": 0.0192,
      "step": 440
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.513008672448299e-05,
      "loss": 0.0078,
      "step": 450
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.496330887258172e-05,
      "loss": 0.0219,
      "step": 460
    },
    {
      "epoch": 0.89,
      "learning_rate": 4.4796531020680454e-05,
      "loss": 0.0076,
      "step": 470
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.4629753168779184e-05,
      "loss": 0.0145,
      "step": 480
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.446297531687792e-05,
      "loss": 0.0121,
      "step": 490
    },
    {
      "epoch": 0.95,
      "learning_rate": 4.429619746497665e-05,
      "loss": 0.0146,
      "step": 500
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.412941961307538e-05,
      "loss": 0.0126,
      "step": 510
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.396264176117412e-05,
      "loss": 0.0142,
      "step": 520
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.379586390927285e-05,
      "loss": 0.0112,
      "step": 530
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.362908605737158e-05,
      "loss": 0.0066,
      "step": 540
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.346230820547031e-05,
      "loss": 0.0116,
      "step": 550
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.329553035356905e-05,
      "loss": 0.0121,
      "step": 560
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.312875250166778e-05,
      "loss": 0.0083,
      "step": 570
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.296197464976651e-05,
      "loss": 0.0085,
      "step": 580
    },
    {
      "epoch": 1.12,
      "learning_rate": 4.279519679786524e-05,
      "loss": 0.0146,
      "step": 590
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.262841894596398e-05,
      "loss": 0.0105,
      "step": 600
    },
    {
      "epoch": 1.14,
      "eval_f1": 0.9896097338429783,
      "eval_f1_0": 0.9968138351713054,
      "eval_f1_1": 0.9974716910591519,
      "eval_f1_2": 0.9651378712455954,
      "eval_f1_3": 0.9990155378958605,
      "eval_loss": 0.015092647634446621,
      "eval_runtime": 19.3157,
      "eval_samples_per_second": 386.835,
      "eval_steps_per_second": 0.414,
      "step": 600
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.246164109406271e-05,
      "loss": 0.0126,
      "step": 610
    },
    {
      "epoch": 1.18,
      "learning_rate": 4.229486324216144e-05,
      "loss": 0.0119,
      "step": 620
    },
    {
      "epoch": 1.2,
      "learning_rate": 4.212808539026018e-05,
      "loss": 0.0211,
      "step": 630
    },
    {
      "epoch": 1.22,
      "learning_rate": 4.196130753835891e-05,
      "loss": 0.0142,
      "step": 640
    },
    {
      "epoch": 1.24,
      "learning_rate": 4.179452968645764e-05,
      "loss": 0.0096,
      "step": 650
    },
    {
      "epoch": 1.25,
      "learning_rate": 4.162775183455637e-05,
      "loss": 0.0084,
      "step": 660
    },
    {
      "epoch": 1.27,
      "learning_rate": 4.146097398265511e-05,
      "loss": 0.0044,
      "step": 670
    },
    {
      "epoch": 1.29,
      "learning_rate": 4.129419613075384e-05,
      "loss": 0.0059,
      "step": 680
    },
    {
      "epoch": 1.31,
      "learning_rate": 4.112741827885257e-05,
      "loss": 0.0042,
      "step": 690
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.09606404269513e-05,
      "loss": 0.0117,
      "step": 700
    },
    {
      "epoch": 1.35,
      "learning_rate": 4.0793862575050036e-05,
      "loss": 0.0074,
      "step": 710
    },
    {
      "epoch": 1.37,
      "learning_rate": 4.062708472314877e-05,
      "loss": 0.0087,
      "step": 720
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.04603068712475e-05,
      "loss": 0.0094,
      "step": 730
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.0293529019346235e-05,
      "loss": 0.0119,
      "step": 740
    },
    {
      "epoch": 1.43,
      "learning_rate": 4.0126751167444966e-05,
      "loss": 0.0138,
      "step": 750
    },
    {
      "epoch": 1.44,
      "learning_rate": 3.9959973315543696e-05,
      "loss": 0.0073,
      "step": 760
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.979319546364243e-05,
      "loss": 0.0096,
      "step": 770
    },
    {
      "epoch": 1.48,
      "learning_rate": 3.9626417611741164e-05,
      "loss": 0.0119,
      "step": 780
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.9459639759839895e-05,
      "loss": 0.0045,
      "step": 790
    },
    {
      "epoch": 1.52,
      "learning_rate": 3.9292861907938625e-05,
      "loss": 0.0102,
      "step": 800
    },
    {
      "epoch": 1.52,
      "eval_f1": 0.9950750308913552,
      "eval_f1_0": 0.9985038101075736,
      "eval_f1_1": 0.9979478535606442,
      "eval_f1_2": 0.9849011258106944,
      "eval_f1_3": 0.9989473340865084,
      "eval_loss": 0.009680411778390408,
      "eval_runtime": 19.3566,
      "eval_samples_per_second": 386.019,
      "eval_steps_per_second": 0.413,
      "step": 800
    },
    {
      "epoch": 1.54,
      "learning_rate": 3.912608405603736e-05,
      "loss": 0.0046,
      "step": 810
    },
    {
      "epoch": 1.56,
      "learning_rate": 3.8959306204136094e-05,
      "loss": 0.0095,
      "step": 820
    },
    {
      "epoch": 1.58,
      "learning_rate": 3.8792528352234824e-05,
      "loss": 0.0092,
      "step": 830
    },
    {
      "epoch": 1.6,
      "learning_rate": 3.8625750500333555e-05,
      "loss": 0.0125,
      "step": 840
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.845897264843229e-05,
      "loss": 0.0182,
      "step": 850
    },
    {
      "epoch": 1.63,
      "learning_rate": 3.829219479653102e-05,
      "loss": 0.0189,
      "step": 860
    },
    {
      "epoch": 1.65,
      "learning_rate": 3.8125416944629753e-05,
      "loss": 0.0151,
      "step": 870
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.7958639092728484e-05,
      "loss": 0.0097,
      "step": 880
    },
    {
      "epoch": 1.69,
      "learning_rate": 3.779186124082722e-05,
      "loss": 0.0047,
      "step": 890
    },
    {
      "epoch": 1.71,
      "learning_rate": 3.762508338892595e-05,
      "loss": 0.0068,
      "step": 900
    },
    {
      "epoch": 1.73,
      "learning_rate": 3.745830553702468e-05,
      "loss": 0.0165,
      "step": 910
    },
    {
      "epoch": 1.75,
      "learning_rate": 3.729152768512342e-05,
      "loss": 0.011,
      "step": 920
    },
    {
      "epoch": 1.77,
      "learning_rate": 3.712474983322215e-05,
      "loss": 0.0128,
      "step": 930
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.695797198132088e-05,
      "loss": 0.0131,
      "step": 940
    },
    {
      "epoch": 1.81,
      "learning_rate": 3.679119412941961e-05,
      "loss": 0.0061,
      "step": 950
    },
    {
      "epoch": 1.83,
      "learning_rate": 3.662441627751835e-05,
      "loss": 0.006,
      "step": 960
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.645763842561708e-05,
      "loss": 0.0053,
      "step": 970
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.629086057371581e-05,
      "loss": 0.0062,
      "step": 980
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.612408272181455e-05,
      "loss": 0.0039,
      "step": 990
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.595730486991328e-05,
      "loss": 0.0123,
      "step": 1000
    },
    {
      "epoch": 1.9,
      "eval_f1": 0.9952965639551941,
      "eval_f1_0": 0.9985619538388097,
      "eval_f1_1": 0.9977381384652626,
      "eval_f1_2": 0.9859661604309984,
      "eval_f1_3": 0.9989200030857055,
      "eval_loss": 0.008907313458621502,
      "eval_runtime": 19.3729,
      "eval_samples_per_second": 385.693,
      "eval_steps_per_second": 0.413,
      "step": 1000
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.579052701801201e-05,
      "loss": 0.0081,
      "step": 1010
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.562374916611074e-05,
      "loss": 0.0037,
      "step": 1020
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.545697131420948e-05,
      "loss": 0.007,
      "step": 1030
    },
    {
      "epoch": 1.98,
      "learning_rate": 3.529019346230821e-05,
      "loss": 0.0043,
      "step": 1040
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.512341561040694e-05,
      "loss": 0.0068,
      "step": 1050
    },
    {
      "epoch": 2.02,
      "learning_rate": 3.495663775850567e-05,
      "loss": 0.0084,
      "step": 1060
    },
    {
      "epoch": 2.03,
      "learning_rate": 3.478985990660441e-05,
      "loss": 0.0023,
      "step": 1070
    },
    {
      "epoch": 2.05,
      "learning_rate": 3.462308205470314e-05,
      "loss": 0.0052,
      "step": 1080
    },
    {
      "epoch": 2.07,
      "learning_rate": 3.445630420280187e-05,
      "loss": 0.0037,
      "step": 1090
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.4289526350900605e-05,
      "loss": 0.0052,
      "step": 1100
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.4122748498999336e-05,
      "loss": 0.0074,
      "step": 1110
    },
    {
      "epoch": 2.13,
      "learning_rate": 3.395597064709807e-05,
      "loss": 0.0027,
      "step": 1120
    },
    {
      "epoch": 2.15,
      "learning_rate": 3.37891927951968e-05,
      "loss": 0.0029,
      "step": 1130
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.3622414943295535e-05,
      "loss": 0.0035,
      "step": 1140
    },
    {
      "epoch": 2.19,
      "learning_rate": 3.3455637091394265e-05,
      "loss": 0.0037,
      "step": 1150
    },
    {
      "epoch": 2.21,
      "learning_rate": 3.3288859239492996e-05,
      "loss": 0.0031,
      "step": 1160
    },
    {
      "epoch": 2.22,
      "learning_rate": 3.3122081387591733e-05,
      "loss": 0.0056,
      "step": 1170
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.2955303535690464e-05,
      "loss": 0.0062,
      "step": 1180
    },
    {
      "epoch": 2.26,
      "learning_rate": 3.2788525683789195e-05,
      "loss": 0.0025,
      "step": 1190
    },
    {
      "epoch": 2.28,
      "learning_rate": 3.2621747831887925e-05,
      "loss": 0.0071,
      "step": 1200
    },
    {
      "epoch": 2.28,
      "eval_f1": 0.9963329419211635,
      "eval_f1_0": 0.9988693819051201,
      "eval_f1_1": 0.9980296050889892,
      "eval_f1_2": 0.9892069880179274,
      "eval_f1_3": 0.9992257926726171,
      "eval_loss": 0.007162621710449457,
      "eval_runtime": 19.3722,
      "eval_samples_per_second": 385.707,
      "eval_steps_per_second": 0.413,
      "step": 1200
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.245496997998666e-05,
      "loss": 0.0083,
      "step": 1210
    },
    {
      "epoch": 2.32,
      "learning_rate": 3.228819212808539e-05,
      "loss": 0.0042,
      "step": 1220
    },
    {
      "epoch": 2.34,
      "learning_rate": 3.2121414276184124e-05,
      "loss": 0.0073,
      "step": 1230
    },
    {
      "epoch": 2.36,
      "learning_rate": 3.1954636424282855e-05,
      "loss": 0.0037,
      "step": 1240
    },
    {
      "epoch": 2.38,
      "learning_rate": 3.178785857238159e-05,
      "loss": 0.0029,
      "step": 1250
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.162108072048032e-05,
      "loss": 0.0031,
      "step": 1260
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.145430286857905e-05,
      "loss": 0.0018,
      "step": 1270
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.128752501667779e-05,
      "loss": 0.0067,
      "step": 1280
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.112074716477652e-05,
      "loss": 0.0023,
      "step": 1290
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.095396931287525e-05,
      "loss": 0.0097,
      "step": 1300
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.078719146097398e-05,
      "loss": 0.0178,
      "step": 1310
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.062041360907272e-05,
      "loss": 0.0035,
      "step": 1320
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.045363575717145e-05,
      "loss": 0.0055,
      "step": 1330
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.028685790527018e-05,
      "loss": 0.0068,
      "step": 1340
    },
    {
      "epoch": 2.57,
      "learning_rate": 3.0120080053368915e-05,
      "loss": 0.0032,
      "step": 1350
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.995330220146765e-05,
      "loss": 0.0047,
      "step": 1360
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.978652434956638e-05,
      "loss": 0.0062,
      "step": 1370
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.9619746497665114e-05,
      "loss": 0.0033,
      "step": 1380
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.9452968645763845e-05,
      "loss": 0.0024,
      "step": 1390
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.928619079386258e-05,
      "loss": 0.0024,
      "step": 1400
    },
    {
      "epoch": 2.66,
      "eval_f1": 0.9959387557872723,
      "eval_f1_0": 0.998752586951556,
      "eval_f1_1": 0.9977694492109829,
      "eval_f1_2": 0.9880693801743948,
      "eval_f1_3": 0.9991636068121555,
      "eval_loss": 0.00828399695456028,
      "eval_runtime": 19.3722,
      "eval_samples_per_second": 385.707,
      "eval_steps_per_second": 0.413,
      "step": 1400
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.911941294196131e-05,
      "loss": 0.0035,
      "step": 1410
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.8952635090060043e-05,
      "loss": 0.0023,
      "step": 1420
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.8785857238158774e-05,
      "loss": 0.0047,
      "step": 1430
    },
    {
      "epoch": 2.74,
      "learning_rate": 2.8619079386257508e-05,
      "loss": 0.0049,
      "step": 1440
    },
    {
      "epoch": 2.76,
      "learning_rate": 2.8452301534356242e-05,
      "loss": 0.0071,
      "step": 1450
    },
    {
      "epoch": 2.78,
      "learning_rate": 2.8285523682454973e-05,
      "loss": 0.0025,
      "step": 1460
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.8118745830553707e-05,
      "loss": 0.0028,
      "step": 1470
    },
    {
      "epoch": 2.81,
      "learning_rate": 2.7951967978652437e-05,
      "loss": 0.0041,
      "step": 1480
    },
    {
      "epoch": 2.83,
      "learning_rate": 2.778519012675117e-05,
      "loss": 0.0054,
      "step": 1490
    },
    {
      "epoch": 2.85,
      "learning_rate": 2.7618412274849902e-05,
      "loss": 0.0077,
      "step": 1500
    },
    {
      "epoch": 2.87,
      "learning_rate": 2.7451634422948636e-05,
      "loss": 0.0046,
      "step": 1510
    },
    {
      "epoch": 2.89,
      "learning_rate": 2.7284856571047366e-05,
      "loss": 0.004,
      "step": 1520
    },
    {
      "epoch": 2.91,
      "learning_rate": 2.71180787191461e-05,
      "loss": 0.0053,
      "step": 1530
    },
    {
      "epoch": 2.93,
      "learning_rate": 2.6951300867244835e-05,
      "loss": 0.0035,
      "step": 1540
    },
    {
      "epoch": 2.95,
      "learning_rate": 2.6784523015343565e-05,
      "loss": 0.0042,
      "step": 1550
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.66177451634423e-05,
      "loss": 0.01,
      "step": 1560
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.645096731154103e-05,
      "loss": 0.0045,
      "step": 1570
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.6284189459639764e-05,
      "loss": 0.0028,
      "step": 1580
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.6117411607738494e-05,
      "loss": 0.0028,
      "step": 1590
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.595063375583723e-05,
      "loss": 0.002,
      "step": 1600
    },
    {
      "epoch": 3.04,
      "eval_f1": 0.9960098503562397,
      "eval_f1_0": 0.9987963282060577,
      "eval_f1_1": 0.9984863058871551,
      "eval_f1_2": 0.9876790221409639,
      "eval_f1_3": 0.9990777451907822,
      "eval_loss": 0.009044110774993896,
      "eval_runtime": 19.4406,
      "eval_samples_per_second": 384.35,
      "eval_steps_per_second": 0.412,
      "step": 1600
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.578385590393596e-05,
      "loss": 0.0038,
      "step": 1610
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.5617078052034693e-05,
      "loss": 0.0027,
      "step": 1620
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.5450300200133427e-05,
      "loss": 0.0021,
      "step": 1630
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.5283522348232158e-05,
      "loss": 0.0025,
      "step": 1640
    },
    {
      "epoch": 3.14,
      "learning_rate": 2.5116744496330892e-05,
      "loss": 0.0038,
      "step": 1650
    },
    {
      "epoch": 3.16,
      "learning_rate": 2.494996664442962e-05,
      "loss": 0.0024,
      "step": 1660
    },
    {
      "epoch": 3.17,
      "learning_rate": 2.4783188792528353e-05,
      "loss": 0.002,
      "step": 1670
    },
    {
      "epoch": 3.19,
      "learning_rate": 2.4616410940627087e-05,
      "loss": 0.0034,
      "step": 1680
    },
    {
      "epoch": 3.21,
      "learning_rate": 2.4449633088725818e-05,
      "loss": 0.002,
      "step": 1690
    },
    {
      "epoch": 3.23,
      "learning_rate": 2.4282855236824552e-05,
      "loss": 0.0018,
      "step": 1700
    },
    {
      "epoch": 3.25,
      "learning_rate": 2.4116077384923282e-05,
      "loss": 0.0026,
      "step": 1710
    },
    {
      "epoch": 3.27,
      "learning_rate": 2.3949299533022016e-05,
      "loss": 0.0019,
      "step": 1720
    },
    {
      "epoch": 3.29,
      "learning_rate": 2.3782521681120747e-05,
      "loss": 0.0076,
      "step": 1730
    },
    {
      "epoch": 3.31,
      "learning_rate": 2.361574382921948e-05,
      "loss": 0.003,
      "step": 1740
    },
    {
      "epoch": 3.33,
      "learning_rate": 2.344896597731821e-05,
      "loss": 0.0018,
      "step": 1750
    },
    {
      "epoch": 3.35,
      "learning_rate": 2.3282188125416946e-05,
      "loss": 0.0025,
      "step": 1760
    },
    {
      "epoch": 3.37,
      "learning_rate": 2.311541027351568e-05,
      "loss": 0.0025,
      "step": 1770
    },
    {
      "epoch": 3.38,
      "learning_rate": 2.294863242161441e-05,
      "loss": 0.0025,
      "step": 1780
    },
    {
      "epoch": 3.4,
      "learning_rate": 2.2781854569713144e-05,
      "loss": 0.0018,
      "step": 1790
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.2615076717811875e-05,
      "loss": 0.0017,
      "step": 1800
    },
    {
      "epoch": 3.42,
      "eval_f1": 0.9965715706847758,
      "eval_f1_0": 0.9989613839913065,
      "eval_f1_1": 0.9985897813972129,
      "eval_f1_2": 0.9894973463811291,
      "eval_f1_3": 0.9992377709694545,
      "eval_loss": 0.0082913339138031,
      "eval_runtime": 19.4143,
      "eval_samples_per_second": 384.87,
      "eval_steps_per_second": 0.412,
      "step": 1800
    },
    {
      "epoch": 3.44,
      "learning_rate": 2.244829886591061e-05,
      "loss": 0.0019,
      "step": 1810
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.228152101400934e-05,
      "loss": 0.0012,
      "step": 1820
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.2114743162108074e-05,
      "loss": 0.0014,
      "step": 1830
    },
    {
      "epoch": 3.5,
      "learning_rate": 2.1947965310206804e-05,
      "loss": 0.0018,
      "step": 1840
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.1781187458305538e-05,
      "loss": 0.0017,
      "step": 1850
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.161440960640427e-05,
      "loss": 0.0017,
      "step": 1860
    },
    {
      "epoch": 3.56,
      "learning_rate": 2.1447631754503003e-05,
      "loss": 0.0047,
      "step": 1870
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.1280853902601737e-05,
      "loss": 0.0024,
      "step": 1880
    },
    {
      "epoch": 3.59,
      "learning_rate": 2.1114076050700468e-05,
      "loss": 0.0021,
      "step": 1890
    },
    {
      "epoch": 3.61,
      "learning_rate": 2.09472981987992e-05,
      "loss": 0.0019,
      "step": 1900
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.0780520346897932e-05,
      "loss": 0.0016,
      "step": 1910
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.0613742494996666e-05,
      "loss": 0.0017,
      "step": 1920
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.0446964643095397e-05,
      "loss": 0.0014,
      "step": 1930
    },
    {
      "epoch": 3.69,
      "learning_rate": 2.028018679119413e-05,
      "loss": 0.0015,
      "step": 1940
    },
    {
      "epoch": 3.71,
      "learning_rate": 2.011340893929286e-05,
      "loss": 0.0022,
      "step": 1950
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.9946631087391596e-05,
      "loss": 0.0026,
      "step": 1960
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.977985323549033e-05,
      "loss": 0.0017,
      "step": 1970
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.961307538358906e-05,
      "loss": 0.0017,
      "step": 1980
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.9446297531687794e-05,
      "loss": 0.0038,
      "step": 1990
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.9279519679786525e-05,
      "loss": 0.0019,
      "step": 2000
    },
    {
      "epoch": 3.8,
      "eval_f1": 0.9954966488952333,
      "eval_f1_0": 0.9986378180905452,
      "eval_f1_1": 0.998616706855238,
      "eval_f1_2": 0.9854914081132,
      "eval_f1_3": 0.9992406625219497,
      "eval_loss": 0.010157858952879906,
      "eval_runtime": 19.3129,
      "eval_samples_per_second": 386.893,
      "eval_steps_per_second": 0.414,
      "step": 2000
    },
    {
      "epoch": 3.82,
      "learning_rate": 1.911274182788526e-05,
      "loss": 0.0038,
      "step": 2010
    },
    {
      "epoch": 3.84,
      "learning_rate": 1.894596397598399e-05,
      "loss": 0.0018,
      "step": 2020
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.8779186124082724e-05,
      "loss": 0.0014,
      "step": 2030
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.8612408272181454e-05,
      "loss": 0.0027,
      "step": 2040
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.8445630420280188e-05,
      "loss": 0.0022,
      "step": 2050
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.8278852568378922e-05,
      "loss": 0.0034,
      "step": 2060
    },
    {
      "epoch": 3.94,
      "learning_rate": 1.8112074716477653e-05,
      "loss": 0.0026,
      "step": 2070
    },
    {
      "epoch": 3.95,
      "learning_rate": 1.7945296864576387e-05,
      "loss": 0.0013,
      "step": 2080
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.7778519012675117e-05,
      "loss": 0.0028,
      "step": 2090
    },
    {
      "epoch": 3.99,
      "learning_rate": 1.761174116077385e-05,
      "loss": 0.0024,
      "step": 2100
    },
    {
      "epoch": 4.01,
      "learning_rate": 1.7444963308872582e-05,
      "loss": 0.0024,
      "step": 2110
    },
    {
      "epoch": 4.03,
      "learning_rate": 1.7278185456971316e-05,
      "loss": 0.0012,
      "step": 2120
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.7111407605070047e-05,
      "loss": 0.0013,
      "step": 2130
    },
    {
      "epoch": 4.07,
      "learning_rate": 1.694462975316878e-05,
      "loss": 0.0013,
      "step": 2140
    },
    {
      "epoch": 4.09,
      "learning_rate": 1.6777851901267515e-05,
      "loss": 0.002,
      "step": 2150
    },
    {
      "epoch": 4.11,
      "learning_rate": 1.6611074049366245e-05,
      "loss": 0.0017,
      "step": 2160
    },
    {
      "epoch": 4.13,
      "learning_rate": 1.644429619746498e-05,
      "loss": 0.0013,
      "step": 2170
    },
    {
      "epoch": 4.14,
      "learning_rate": 1.627751834556371e-05,
      "loss": 0.0015,
      "step": 2180
    },
    {
      "epoch": 4.16,
      "learning_rate": 1.6110740493662444e-05,
      "loss": 0.0012,
      "step": 2190
    },
    {
      "epoch": 4.18,
      "learning_rate": 1.5943962641761175e-05,
      "loss": 0.0016,
      "step": 2200
    },
    {
      "epoch": 4.18,
      "eval_f1": 0.9964539401625592,
      "eval_f1_0": 0.9989316835915937,
      "eval_f1_1": 0.9986976730119667,
      "eval_f1_2": 0.9889220122285377,
      "eval_f1_3": 0.9992643918181386,
      "eval_loss": 0.008549452759325504,
      "eval_runtime": 19.2872,
      "eval_samples_per_second": 387.407,
      "eval_steps_per_second": 0.415,
      "step": 2200
    },
    {
      "epoch": 4.2,
      "learning_rate": 1.577718478985991e-05,
      "loss": 0.0013,
      "step": 2210
    },
    {
      "epoch": 4.22,
      "learning_rate": 1.561040693795864e-05,
      "loss": 0.0018,
      "step": 2220
    },
    {
      "epoch": 4.24,
      "learning_rate": 1.5443629086057373e-05,
      "loss": 0.0012,
      "step": 2230
    },
    {
      "epoch": 4.26,
      "learning_rate": 1.5276851234156107e-05,
      "loss": 0.0014,
      "step": 2240
    },
    {
      "epoch": 4.28,
      "learning_rate": 1.5110073382254838e-05,
      "loss": 0.0011,
      "step": 2250
    },
    {
      "epoch": 4.3,
      "learning_rate": 1.494329553035357e-05,
      "loss": 0.0044,
      "step": 2260
    },
    {
      "epoch": 4.32,
      "learning_rate": 1.4776517678452303e-05,
      "loss": 0.0028,
      "step": 2270
    },
    {
      "epoch": 4.33,
      "learning_rate": 1.4609739826551037e-05,
      "loss": 0.0014,
      "step": 2280
    },
    {
      "epoch": 4.35,
      "learning_rate": 1.4442961974649769e-05,
      "loss": 0.001,
      "step": 2290
    },
    {
      "epoch": 4.37,
      "learning_rate": 1.4276184122748501e-05,
      "loss": 0.0018,
      "step": 2300
    },
    {
      "epoch": 4.39,
      "learning_rate": 1.4109406270847234e-05,
      "loss": 0.0012,
      "step": 2310
    },
    {
      "epoch": 4.41,
      "learning_rate": 1.3942628418945963e-05,
      "loss": 0.0011,
      "step": 2320
    },
    {
      "epoch": 4.43,
      "learning_rate": 1.3775850567044695e-05,
      "loss": 0.0011,
      "step": 2330
    },
    {
      "epoch": 4.45,
      "learning_rate": 1.3609072715143429e-05,
      "loss": 0.0013,
      "step": 2340
    },
    {
      "epoch": 4.47,
      "learning_rate": 1.3442294863242161e-05,
      "loss": 0.0022,
      "step": 2350
    },
    {
      "epoch": 4.49,
      "learning_rate": 1.3275517011340894e-05,
      "loss": 0.0012,
      "step": 2360
    },
    {
      "epoch": 4.51,
      "learning_rate": 1.3108739159439626e-05,
      "loss": 0.0013,
      "step": 2370
    },
    {
      "epoch": 4.52,
      "learning_rate": 1.2941961307538358e-05,
      "loss": 0.0014,
      "step": 2380
    },
    {
      "epoch": 4.54,
      "learning_rate": 1.277518345563709e-05,
      "loss": 0.0011,
      "step": 2390
    },
    {
      "epoch": 4.56,
      "learning_rate": 1.2608405603735823e-05,
      "loss": 0.0013,
      "step": 2400
    },
    {
      "epoch": 4.56,
      "eval_f1": 0.9965431300136034,
      "eval_f1_0": 0.9989591777046161,
      "eval_f1_1": 0.9986785664998649,
      "eval_f1_2": 0.9892763163558415,
      "eval_f1_3": 0.9992584594940914,
      "eval_loss": 0.008574241772294044,
      "eval_runtime": 19.3355,
      "eval_samples_per_second": 386.439,
      "eval_steps_per_second": 0.414,
      "step": 2400
    },
    {
      "epoch": 4.58,
      "learning_rate": 1.2441627751834557e-05,
      "loss": 0.0018,
      "step": 2410
    },
    {
      "epoch": 4.6,
      "learning_rate": 1.227484989993329e-05,
      "loss": 0.0011,
      "step": 2420
    },
    {
      "epoch": 4.62,
      "learning_rate": 1.2108072048032022e-05,
      "loss": 0.0012,
      "step": 2430
    },
    {
      "epoch": 4.64,
      "learning_rate": 1.1941294196130754e-05,
      "loss": 0.0018,
      "step": 2440
    },
    {
      "epoch": 4.66,
      "learning_rate": 1.1774516344229488e-05,
      "loss": 0.0012,
      "step": 2450
    },
    {
      "epoch": 4.68,
      "learning_rate": 1.160773849232822e-05,
      "loss": 0.0019,
      "step": 2460
    },
    {
      "epoch": 4.7,
      "learning_rate": 1.1440960640426953e-05,
      "loss": 0.0013,
      "step": 2470
    },
    {
      "epoch": 4.71,
      "learning_rate": 1.1274182788525685e-05,
      "loss": 0.0014,
      "step": 2480
    },
    {
      "epoch": 4.73,
      "learning_rate": 1.1107404936624417e-05,
      "loss": 0.0012,
      "step": 2490
    },
    {
      "epoch": 4.75,
      "learning_rate": 1.094062708472315e-05,
      "loss": 0.0015,
      "step": 2500
    },
    {
      "epoch": 4.77,
      "learning_rate": 1.0773849232821882e-05,
      "loss": 0.0012,
      "step": 2510
    },
    {
      "epoch": 4.79,
      "learning_rate": 1.0607071380920614e-05,
      "loss": 0.0011,
      "step": 2520
    },
    {
      "epoch": 4.81,
      "learning_rate": 1.0440293529019347e-05,
      "loss": 0.0016,
      "step": 2530
    },
    {
      "epoch": 4.83,
      "learning_rate": 1.027351567711808e-05,
      "loss": 0.0025,
      "step": 2540
    },
    {
      "epoch": 4.85,
      "learning_rate": 1.0106737825216811e-05,
      "loss": 0.0019,
      "step": 2550
    },
    {
      "epoch": 4.87,
      "learning_rate": 9.939959973315544e-06,
      "loss": 0.0012,
      "step": 2560
    },
    {
      "epoch": 4.89,
      "learning_rate": 9.773182121414276e-06,
      "loss": 0.0011,
      "step": 2570
    },
    {
      "epoch": 4.9,
      "learning_rate": 9.606404269513008e-06,
      "loss": 0.0018,
      "step": 2580
    },
    {
      "epoch": 4.92,
      "learning_rate": 9.43962641761174e-06,
      "loss": 0.001,
      "step": 2590
    },
    {
      "epoch": 4.94,
      "learning_rate": 9.272848565710473e-06,
      "loss": 0.0016,
      "step": 2600
    },
    {
      "epoch": 4.94,
      "eval_f1": 0.9963671849973992,
      "eval_f1_0": 0.9988809930374426,
      "eval_f1_1": 0.9980475327397732,
      "eval_f1_2": 0.9892787776027725,
      "eval_f1_3": 0.9992614366096084,
      "eval_loss": 0.009876401163637638,
      "eval_runtime": 19.3714,
      "eval_samples_per_second": 385.724,
      "eval_steps_per_second": 0.413,
      "step": 2600
    },
    {
      "epoch": 4.96,
      "learning_rate": 9.106070713809207e-06,
      "loss": 0.001,
      "step": 2610
    },
    {
      "epoch": 4.98,
      "learning_rate": 8.93929286190794e-06,
      "loss": 0.0023,
      "step": 2620
    },
    {
      "epoch": 5.0,
      "learning_rate": 8.772515010006672e-06,
      "loss": 0.0024,
      "step": 2630
    },
    {
      "epoch": 5.02,
      "learning_rate": 8.605737158105404e-06,
      "loss": 0.0026,
      "step": 2640
    },
    {
      "epoch": 5.04,
      "learning_rate": 8.438959306204136e-06,
      "loss": 0.0011,
      "step": 2650
    },
    {
      "epoch": 5.06,
      "learning_rate": 8.272181454302869e-06,
      "loss": 0.0012,
      "step": 2660
    },
    {
      "epoch": 5.08,
      "learning_rate": 8.105403602401601e-06,
      "loss": 0.001,
      "step": 2670
    },
    {
      "epoch": 5.1,
      "learning_rate": 7.938625750500333e-06,
      "loss": 0.001,
      "step": 2680
    },
    {
      "epoch": 5.11,
      "learning_rate": 7.771847898599065e-06,
      "loss": 0.001,
      "step": 2690
    },
    {
      "epoch": 5.13,
      "learning_rate": 7.605070046697799e-06,
      "loss": 0.002,
      "step": 2700
    },
    {
      "epoch": 5.15,
      "learning_rate": 7.438292194796531e-06,
      "loss": 0.0013,
      "step": 2710
    },
    {
      "epoch": 5.17,
      "learning_rate": 7.271514342895263e-06,
      "loss": 0.0012,
      "step": 2720
    },
    {
      "epoch": 5.19,
      "learning_rate": 7.1047364909939965e-06,
      "loss": 0.0011,
      "step": 2730
    },
    {
      "epoch": 5.21,
      "learning_rate": 6.937958639092729e-06,
      "loss": 0.0009,
      "step": 2740
    },
    {
      "epoch": 5.23,
      "learning_rate": 6.771180787191461e-06,
      "loss": 0.0011,
      "step": 2750
    },
    {
      "epoch": 5.25,
      "learning_rate": 6.6044029352901935e-06,
      "loss": 0.0012,
      "step": 2760
    },
    {
      "epoch": 5.27,
      "learning_rate": 6.437625083388927e-06,
      "loss": 0.0011,
      "step": 2770
    },
    {
      "epoch": 5.29,
      "learning_rate": 6.270847231487659e-06,
      "loss": 0.001,
      "step": 2780
    },
    {
      "epoch": 5.3,
      "learning_rate": 6.104069379586391e-06,
      "loss": 0.0008,
      "step": 2790
    },
    {
      "epoch": 5.32,
      "learning_rate": 5.937291527685124e-06,
      "loss": 0.0011,
      "step": 2800
    },
    {
      "epoch": 5.32,
      "eval_f1": 0.9965125366318218,
      "eval_f1_0": 0.9989578910626513,
      "eval_f1_1": 0.9988072938423473,
      "eval_f1_2": 0.9889968193931058,
      "eval_f1_3": 0.9992881422291826,
      "eval_loss": 0.009526900947093964,
      "eval_runtime": 19.4195,
      "eval_samples_per_second": 384.768,
      "eval_steps_per_second": 0.412,
      "step": 2800
    },
    {
      "epoch": 5.34,
      "learning_rate": 5.770513675783856e-06,
      "loss": 0.0018,
      "step": 2810
    },
    {
      "epoch": 5.36,
      "learning_rate": 5.603735823882589e-06,
      "loss": 0.001,
      "step": 2820
    },
    {
      "epoch": 5.38,
      "learning_rate": 5.4369579719813214e-06,
      "loss": 0.0012,
      "step": 2830
    },
    {
      "epoch": 5.4,
      "learning_rate": 5.270180120080054e-06,
      "loss": 0.0011,
      "step": 2840
    },
    {
      "epoch": 5.42,
      "learning_rate": 5.103402268178786e-06,
      "loss": 0.0008,
      "step": 2850
    },
    {
      "epoch": 5.44,
      "learning_rate": 4.936624416277518e-06,
      "loss": 0.0013,
      "step": 2860
    },
    {
      "epoch": 5.46,
      "learning_rate": 4.769846564376252e-06,
      "loss": 0.0011,
      "step": 2870
    },
    {
      "epoch": 5.48,
      "learning_rate": 4.603068712474984e-06,
      "loss": 0.0013,
      "step": 2880
    },
    {
      "epoch": 5.49,
      "learning_rate": 4.436290860573716e-06,
      "loss": 0.0009,
      "step": 2890
    },
    {
      "epoch": 5.51,
      "learning_rate": 4.2695130086724486e-06,
      "loss": 0.001,
      "step": 2900
    },
    {
      "epoch": 5.53,
      "learning_rate": 4.102735156771181e-06,
      "loss": 0.001,
      "step": 2910
    },
    {
      "epoch": 5.55,
      "learning_rate": 3.935957304869913e-06,
      "loss": 0.0022,
      "step": 2920
    },
    {
      "epoch": 5.57,
      "learning_rate": 3.7691794529686455e-06,
      "loss": 0.001,
      "step": 2930
    },
    {
      "epoch": 5.59,
      "learning_rate": 3.6024016010673783e-06,
      "loss": 0.0013,
      "step": 2940
    },
    {
      "epoch": 5.61,
      "learning_rate": 3.4356237491661106e-06,
      "loss": 0.0009,
      "step": 2950
    },
    {
      "epoch": 5.63,
      "learning_rate": 3.2688458972648434e-06,
      "loss": 0.0011,
      "step": 2960
    },
    {
      "epoch": 5.65,
      "learning_rate": 3.1020680453635757e-06,
      "loss": 0.0013,
      "step": 2970
    },
    {
      "epoch": 5.67,
      "learning_rate": 2.9352901934623084e-06,
      "loss": 0.001,
      "step": 2980
    },
    {
      "epoch": 5.68,
      "learning_rate": 2.7685123415610408e-06,
      "loss": 0.0018,
      "step": 2990
    },
    {
      "epoch": 5.7,
      "learning_rate": 2.601734489659773e-06,
      "loss": 0.0024,
      "step": 3000
    },
    {
      "epoch": 5.7,
      "eval_f1": 0.9965564466847002,
      "eval_f1_0": 0.9989728439908399,
      "eval_f1_1": 0.998860218996728,
      "eval_f1_2": 0.9891016301666877,
      "eval_f1_3": 0.9992910935845453,
      "eval_loss": 0.009422173723578453,
      "eval_runtime": 19.3758,
      "eval_samples_per_second": 385.635,
      "eval_steps_per_second": 0.413,
      "step": 3000
    },
    {
      "epoch": 5.72,
      "learning_rate": 2.434956637758506e-06,
      "loss": 0.0013,
      "step": 3010
    },
    {
      "epoch": 5.74,
      "learning_rate": 2.268178785857238e-06,
      "loss": 0.0014,
      "step": 3020
    },
    {
      "epoch": 5.76,
      "learning_rate": 2.101400933955971e-06,
      "loss": 0.001,
      "step": 3030
    },
    {
      "epoch": 5.78,
      "learning_rate": 1.9346230820547032e-06,
      "loss": 0.0013,
      "step": 3040
    },
    {
      "epoch": 5.8,
      "learning_rate": 1.7678452301534358e-06,
      "loss": 0.0015,
      "step": 3050
    },
    {
      "epoch": 5.82,
      "learning_rate": 1.6010673782521683e-06,
      "loss": 0.0012,
      "step": 3060
    },
    {
      "epoch": 5.84,
      "learning_rate": 1.4342895263509006e-06,
      "loss": 0.0014,
      "step": 3070
    },
    {
      "epoch": 5.86,
      "learning_rate": 1.2675116744496332e-06,
      "loss": 0.0012,
      "step": 3080
    },
    {
      "epoch": 5.87,
      "learning_rate": 1.1007338225483657e-06,
      "loss": 0.0011,
      "step": 3090
    },
    {
      "epoch": 5.89,
      "learning_rate": 9.339559706470982e-07,
      "loss": 0.0008,
      "step": 3100
    },
    {
      "epoch": 5.91,
      "learning_rate": 7.671781187458306e-07,
      "loss": 0.0013,
      "step": 3110
    },
    {
      "epoch": 5.93,
      "learning_rate": 6.00400266844563e-07,
      "loss": 0.0011,
      "step": 3120
    },
    {
      "epoch": 5.95,
      "learning_rate": 4.3362241494329555e-07,
      "loss": 0.001,
      "step": 3130
    },
    {
      "epoch": 5.97,
      "learning_rate": 2.6684456304202804e-07,
      "loss": 0.0012,
      "step": 3140
    },
    {
      "epoch": 5.99,
      "learning_rate": 1.0006671114076052e-07,
      "loss": 0.0011,
      "step": 3150
    },
    {
      "epoch": 6.0,
      "step": 3156,
      "total_flos": 3.671049324168479e+17,
      "train_loss": 0.019633541173813333,
      "train_runtime": 2021.1766,
      "train_samples_per_second": 199.604,
      "train_steps_per_second": 1.561
    }
  ],
  "logging_steps": 10,
  "max_steps": 3156,
  "num_train_epochs": 6,
  "save_steps": 200,
  "total_flos": 3.671049324168479e+17,
  "trial_name": null,
  "trial_params": null
}