{
  "best_metric": 0.15622803568840027,
  "best_model_checkpoint": "frostsolutions/frost-vision-v2-google_vit-base-patch16-224/checkpoint-500",
  "epoch": 30.0,
  "eval_steps": 100,
  "global_step": 2610,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11494252873563218,
      "grad_norm": 1.843792200088501,
      "learning_rate": 1.9157088122605362e-06,
      "loss": 0.7018,
      "step": 10
    },
    {
      "epoch": 0.22988505747126436,
      "grad_norm": 1.8868529796600342,
      "learning_rate": 3.8314176245210725e-06,
      "loss": 0.6818,
      "step": 20
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 1.7420289516448975,
      "learning_rate": 5.747126436781609e-06,
      "loss": 0.6445,
      "step": 30
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 1.434963345527649,
      "learning_rate": 7.662835249042145e-06,
      "loss": 0.5892,
      "step": 40
    },
    {
      "epoch": 0.5747126436781609,
      "grad_norm": 1.3582804203033447,
      "learning_rate": 9.578544061302683e-06,
      "loss": 0.5362,
      "step": 50
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 1.2787097692489624,
      "learning_rate": 1.1494252873563218e-05,
      "loss": 0.4723,
      "step": 60
    },
    {
      "epoch": 0.8045977011494253,
      "grad_norm": 1.0611391067504883,
      "learning_rate": 1.3409961685823755e-05,
      "loss": 0.4433,
      "step": 70
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 1.0236091613769531,
      "learning_rate": 1.532567049808429e-05,
      "loss": 0.415,
      "step": 80
    },
    {
      "epoch": 1.0344827586206897,
      "grad_norm": 1.0555670261383057,
      "learning_rate": 1.7241379310344828e-05,
      "loss": 0.3879,
      "step": 90
    },
    {
      "epoch": 1.1494252873563218,
      "grad_norm": 1.0582798719406128,
      "learning_rate": 1.9157088122605367e-05,
      "loss": 0.3416,
      "step": 100
    },
    {
      "epoch": 1.1494252873563218,
      "eval_accuracy": 0.8771014492753623,
      "eval_f1": 0.6124314442413162,
      "eval_loss": 0.32730263471603394,
      "eval_precision": 0.9005376344086021,
      "eval_recall": 0.46398891966759004,
      "eval_runtime": 2.5472,
      "eval_samples_per_second": 135.44,
      "eval_steps_per_second": 17.274,
      "step": 100
    },
    {
      "epoch": 1.264367816091954,
      "grad_norm": 1.0698238611221313,
      "learning_rate": 2.10727969348659e-05,
      "loss": 0.3371,
      "step": 110
    },
    {
      "epoch": 1.3793103448275863,
      "grad_norm": 1.0518220663070679,
      "learning_rate": 2.2988505747126437e-05,
      "loss": 0.3122,
      "step": 120
    },
    {
      "epoch": 1.4942528735632183,
      "grad_norm": 1.3425720930099487,
      "learning_rate": 2.4904214559386975e-05,
      "loss": 0.2861,
      "step": 130
    },
    {
      "epoch": 1.6091954022988506,
      "grad_norm": 0.789143979549408,
      "learning_rate": 2.681992337164751e-05,
      "loss": 0.2678,
      "step": 140
    },
    {
      "epoch": 1.7241379310344827,
      "grad_norm": 0.8800056576728821,
      "learning_rate": 2.8735632183908045e-05,
      "loss": 0.2811,
      "step": 150
    },
    {
      "epoch": 1.839080459770115,
      "grad_norm": 0.943029522895813,
      "learning_rate": 3.065134099616858e-05,
      "loss": 0.2767,
      "step": 160
    },
    {
      "epoch": 1.9540229885057472,
      "grad_norm": 1.1474237442016602,
      "learning_rate": 3.256704980842912e-05,
      "loss": 0.266,
      "step": 170
    },
    {
      "epoch": 2.0689655172413794,
      "grad_norm": 1.0653332471847534,
      "learning_rate": 3.4482758620689657e-05,
      "loss": 0.2381,
      "step": 180
    },
    {
      "epoch": 2.1839080459770113,
      "grad_norm": 1.0384609699249268,
      "learning_rate": 3.6398467432950195e-05,
      "loss": 0.2266,
      "step": 190
    },
    {
      "epoch": 2.2988505747126435,
      "grad_norm": 0.7939756512641907,
      "learning_rate": 3.831417624521073e-05,
      "loss": 0.2215,
      "step": 200
    },
    {
      "epoch": 2.2988505747126435,
      "eval_accuracy": 0.9182608695652174,
      "eval_f1": 0.7901785714285714,
      "eval_loss": 0.21871700882911682,
      "eval_precision": 0.8536977491961415,
      "eval_recall": 0.7354570637119113,
      "eval_runtime": 2.9768,
      "eval_samples_per_second": 115.897,
      "eval_steps_per_second": 14.781,
      "step": 200
    },
    {
      "epoch": 2.413793103448276,
      "grad_norm": 0.8638312816619873,
      "learning_rate": 4.0229885057471265e-05,
      "loss": 0.2274,
      "step": 210
    },
    {
      "epoch": 2.528735632183908,
      "grad_norm": 1.0766372680664062,
      "learning_rate": 4.21455938697318e-05,
      "loss": 0.1967,
      "step": 220
    },
    {
      "epoch": 2.6436781609195403,
      "grad_norm": 0.8049170970916748,
      "learning_rate": 4.406130268199234e-05,
      "loss": 0.2041,
      "step": 230
    },
    {
      "epoch": 2.7586206896551726,
      "grad_norm": 0.8865797519683838,
      "learning_rate": 4.597701149425287e-05,
      "loss": 0.2288,
      "step": 240
    },
    {
      "epoch": 2.873563218390805,
      "grad_norm": 0.906036376953125,
      "learning_rate": 4.789272030651341e-05,
      "loss": 0.2205,
      "step": 250
    },
    {
      "epoch": 2.9885057471264367,
      "grad_norm": 1.1091452836990356,
      "learning_rate": 4.980842911877395e-05,
      "loss": 0.2075,
      "step": 260
    },
    {
      "epoch": 3.103448275862069,
      "grad_norm": 1.1074674129486084,
      "learning_rate": 4.980842911877395e-05,
      "loss": 0.1914,
      "step": 270
    },
    {
      "epoch": 3.218390804597701,
      "grad_norm": 1.6170417070388794,
      "learning_rate": 4.959557258407833e-05,
      "loss": 0.1792,
      "step": 280
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 1.5800983905792236,
      "learning_rate": 4.938271604938271e-05,
      "loss": 0.1985,
      "step": 290
    },
    {
      "epoch": 3.4482758620689653,
      "grad_norm": 1.2079609632492065,
      "learning_rate": 4.916985951468711e-05,
      "loss": 0.1753,
      "step": 300
    },
    {
      "epoch": 3.4482758620689653,
      "eval_accuracy": 0.923768115942029,
      "eval_f1": 0.8098336948662328,
      "eval_loss": 0.1898600310087204,
      "eval_precision": 0.8472012102874432,
      "eval_recall": 0.775623268698061,
      "eval_runtime": 3.0971,
      "eval_samples_per_second": 111.395,
      "eval_steps_per_second": 14.207,
      "step": 300
    },
    {
      "epoch": 3.5632183908045976,
      "grad_norm": 0.70270836353302,
      "learning_rate": 4.895700297999149e-05,
      "loss": 0.2106,
      "step": 310
    },
    {
      "epoch": 3.67816091954023,
      "grad_norm": 0.7896942496299744,
      "learning_rate": 4.874414644529587e-05,
      "loss": 0.168,
      "step": 320
    },
    {
      "epoch": 3.793103448275862,
      "grad_norm": 1.1563091278076172,
      "learning_rate": 4.853128991060026e-05,
      "loss": 0.2003,
      "step": 330
    },
    {
      "epoch": 3.9080459770114944,
      "grad_norm": 1.192246437072754,
      "learning_rate": 4.831843337590464e-05,
      "loss": 0.203,
      "step": 340
    },
    {
      "epoch": 4.022988505747127,
      "grad_norm": 1.551155686378479,
      "learning_rate": 4.810557684120903e-05,
      "loss": 0.1936,
      "step": 350
    },
    {
      "epoch": 4.137931034482759,
      "grad_norm": 1.4264142513275146,
      "learning_rate": 4.789272030651341e-05,
      "loss": 0.1783,
      "step": 360
    },
    {
      "epoch": 4.252873563218391,
      "grad_norm": 0.6229875087738037,
      "learning_rate": 4.767986377181779e-05,
      "loss": 0.1797,
      "step": 370
    },
    {
      "epoch": 4.3678160919540225,
      "grad_norm": 0.6375844478607178,
      "learning_rate": 4.746700723712218e-05,
      "loss": 0.1589,
      "step": 380
    },
    {
      "epoch": 4.482758620689655,
      "grad_norm": 1.297573447227478,
      "learning_rate": 4.725415070242657e-05,
      "loss": 0.142,
      "step": 390
    },
    {
      "epoch": 4.597701149425287,
      "grad_norm": 0.7838256359100342,
      "learning_rate": 4.704129416773095e-05,
      "loss": 0.1656,
      "step": 400
    },
    {
      "epoch": 4.597701149425287,
      "eval_accuracy": 0.9272463768115942,
      "eval_f1": 0.8174545454545454,
      "eval_loss": 0.17321595549583435,
      "eval_precision": 0.8606431852986217,
      "eval_recall": 0.778393351800554,
      "eval_runtime": 4.6227,
      "eval_samples_per_second": 74.631,
      "eval_steps_per_second": 9.518,
      "step": 400
    },
    {
      "epoch": 4.712643678160919,
      "grad_norm": 0.8572351932525635,
      "learning_rate": 4.682843763303534e-05,
      "loss": 0.1843,
      "step": 410
    },
    {
      "epoch": 4.827586206896552,
      "grad_norm": 0.9022108912467957,
      "learning_rate": 4.661558109833972e-05,
      "loss": 0.1638,
      "step": 420
    },
    {
      "epoch": 4.942528735632184,
      "grad_norm": 1.1695079803466797,
      "learning_rate": 4.640272456364411e-05,
      "loss": 0.1447,
      "step": 430
    },
    {
      "epoch": 5.057471264367816,
      "grad_norm": 0.9919353723526001,
      "learning_rate": 4.618986802894849e-05,
      "loss": 0.1796,
      "step": 440
    },
    {
      "epoch": 5.172413793103448,
      "grad_norm": 0.9828383326530457,
      "learning_rate": 4.597701149425287e-05,
      "loss": 0.1376,
      "step": 450
    },
    {
      "epoch": 5.287356321839081,
      "grad_norm": 0.7474183440208435,
      "learning_rate": 4.576415495955726e-05,
      "loss": 0.1245,
      "step": 460
    },
    {
      "epoch": 5.402298850574713,
      "grad_norm": 0.9904183149337769,
      "learning_rate": 4.555129842486164e-05,
      "loss": 0.1531,
      "step": 470
    },
    {
      "epoch": 5.517241379310345,
      "grad_norm": 0.7156611680984497,
      "learning_rate": 4.5338441890166025e-05,
      "loss": 0.1551,
      "step": 480
    },
    {
      "epoch": 5.6321839080459775,
      "grad_norm": 0.7088082432746887,
      "learning_rate": 4.512558535547042e-05,
      "loss": 0.1444,
      "step": 490
    },
    {
      "epoch": 5.747126436781609,
      "grad_norm": 0.8901511430740356,
      "learning_rate": 4.49127288207748e-05,
      "loss": 0.1288,
      "step": 500
    },
    {
      "epoch": 5.747126436781609,
      "eval_accuracy": 0.9359420289855073,
      "eval_f1": 0.8380952380952381,
      "eval_loss": 0.15622803568840027,
      "eval_precision": 0.8895800933125972,
      "eval_recall": 0.7922437673130194,
      "eval_runtime": 2.5149,
      "eval_samples_per_second": 137.184,
      "eval_steps_per_second": 17.496,
      "step": 500
    },
    {
      "epoch": 5.862068965517241,
      "grad_norm": 0.8678961992263794,
      "learning_rate": 4.469987228607918e-05,
      "loss": 0.144,
      "step": 510
    },
    {
      "epoch": 5.977011494252873,
      "grad_norm": 0.8695370554924011,
      "learning_rate": 4.448701575138357e-05,
      "loss": 0.1543,
      "step": 520
    },
    {
      "epoch": 6.091954022988506,
      "grad_norm": 0.6334937810897827,
      "learning_rate": 4.427415921668795e-05,
      "loss": 0.118,
      "step": 530
    },
    {
      "epoch": 6.206896551724138,
      "grad_norm": 0.8258758187294006,
      "learning_rate": 4.406130268199234e-05,
      "loss": 0.1127,
      "step": 540
    },
    {
      "epoch": 6.32183908045977,
      "grad_norm": 1.4297465085983276,
      "learning_rate": 4.384844614729672e-05,
      "loss": 0.124,
      "step": 550
    },
    {
      "epoch": 6.436781609195402,
      "grad_norm": 0.9284784197807312,
      "learning_rate": 4.3635589612601105e-05,
      "loss": 0.1153,
      "step": 560
    },
    {
      "epoch": 6.551724137931035,
      "grad_norm": 0.7226102352142334,
      "learning_rate": 4.342273307790549e-05,
      "loss": 0.1204,
      "step": 570
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 1.2235733270645142,
      "learning_rate": 4.3209876543209875e-05,
      "loss": 0.1393,
      "step": 580
    },
    {
      "epoch": 6.781609195402299,
      "grad_norm": 0.6442233324050903,
      "learning_rate": 4.299702000851426e-05,
      "loss": 0.1311,
      "step": 590
    },
    {
      "epoch": 6.896551724137931,
      "grad_norm": 0.8887196183204651,
      "learning_rate": 4.278416347381865e-05,
      "loss": 0.1323,
      "step": 600
    },
    {
      "epoch": 6.896551724137931,
      "eval_accuracy": 0.9321739130434783,
      "eval_f1": 0.8326180257510729,
      "eval_loss": 0.1597267985343933,
      "eval_precision": 0.8609467455621301,
      "eval_recall": 0.8060941828254847,
      "eval_runtime": 2.5018,
      "eval_samples_per_second": 137.898,
      "eval_steps_per_second": 17.587,
      "step": 600
    },
    {
      "epoch": 7.011494252873563,
      "grad_norm": 0.535347044467926,
      "learning_rate": 4.257130693912303e-05,
      "loss": 0.1151,
      "step": 610
    },
    {
      "epoch": 7.126436781609195,
      "grad_norm": 1.1041584014892578,
      "learning_rate": 4.235845040442742e-05,
      "loss": 0.1196,
      "step": 620
    },
    {
      "epoch": 7.241379310344827,
      "grad_norm": 1.098777413368225,
      "learning_rate": 4.21455938697318e-05,
      "loss": 0.1262,
      "step": 630
    },
    {
      "epoch": 7.35632183908046,
      "grad_norm": 1.0132532119750977,
      "learning_rate": 4.1932737335036185e-05,
      "loss": 0.115,
      "step": 640
    },
    {
      "epoch": 7.471264367816092,
      "grad_norm": 1.0600366592407227,
      "learning_rate": 4.171988080034057e-05,
      "loss": 0.1205,
      "step": 650
    },
    {
      "epoch": 7.586206896551724,
      "grad_norm": 0.8362026810646057,
      "learning_rate": 4.1507024265644955e-05,
      "loss": 0.1294,
      "step": 660
    },
    {
      "epoch": 7.7011494252873565,
      "grad_norm": 0.9791185855865479,
      "learning_rate": 4.129416773094934e-05,
      "loss": 0.1253,
      "step": 670
    },
    {
      "epoch": 7.816091954022989,
      "grad_norm": 0.6780907511711121,
      "learning_rate": 4.1081311196253725e-05,
      "loss": 0.1072,
      "step": 680
    },
    {
      "epoch": 7.931034482758621,
      "grad_norm": 0.9286981225013733,
      "learning_rate": 4.086845466155811e-05,
      "loss": 0.114,
      "step": 690
    },
    {
      "epoch": 8.045977011494253,
      "grad_norm": 0.48207980394363403,
      "learning_rate": 4.06555981268625e-05,
      "loss": 0.1004,
      "step": 700
    },
    {
      "epoch": 8.045977011494253,
      "eval_accuracy": 0.9315942028985508,
      "eval_f1": 0.8323863636363636,
      "eval_loss": 0.161317378282547,
      "eval_precision": 0.8542274052478134,
      "eval_recall": 0.8116343490304709,
      "eval_runtime": 2.5046,
      "eval_samples_per_second": 137.746,
      "eval_steps_per_second": 17.568,
      "step": 700
    },
    {
      "epoch": 8.160919540229886,
      "grad_norm": 1.1616382598876953,
      "learning_rate": 4.044274159216688e-05,
      "loss": 0.1209,
      "step": 710
    },
    {
      "epoch": 8.275862068965518,
      "grad_norm": 0.7316162586212158,
      "learning_rate": 4.0229885057471265e-05,
      "loss": 0.096,
      "step": 720
    },
    {
      "epoch": 8.39080459770115,
      "grad_norm": 0.8447398543357849,
      "learning_rate": 4.001702852277565e-05,
      "loss": 0.1163,
      "step": 730
    },
    {
      "epoch": 8.505747126436782,
      "grad_norm": 0.6557740569114685,
      "learning_rate": 3.9804171988080035e-05,
      "loss": 0.0951,
      "step": 740
    },
    {
      "epoch": 8.620689655172415,
      "grad_norm": 0.9132739901542664,
      "learning_rate": 3.959131545338442e-05,
      "loss": 0.1045,
      "step": 750
    },
    {
      "epoch": 8.735632183908045,
      "grad_norm": 1.8383800983428955,
      "learning_rate": 3.9378458918688805e-05,
      "loss": 0.1007,
      "step": 760
    },
    {
      "epoch": 8.850574712643677,
      "grad_norm": 1.6262847185134888,
      "learning_rate": 3.9165602383993187e-05,
      "loss": 0.0886,
      "step": 770
    },
    {
      "epoch": 8.96551724137931,
      "grad_norm": 1.1226928234100342,
      "learning_rate": 3.8952745849297575e-05,
      "loss": 0.1136,
      "step": 780
    },
    {
      "epoch": 9.080459770114942,
      "grad_norm": 0.8685470223426819,
      "learning_rate": 3.873988931460196e-05,
      "loss": 0.086,
      "step": 790
    },
    {
      "epoch": 9.195402298850574,
      "grad_norm": 0.662932276725769,
      "learning_rate": 3.8527032779906345e-05,
      "loss": 0.0956,
      "step": 800
    },
    {
      "epoch": 9.195402298850574,
      "eval_accuracy": 0.933623188405797,
      "eval_f1": 0.8367783321454028,
      "eval_loss": 0.16119635105133057,
      "eval_precision": 0.8619676945668135,
      "eval_recall": 0.8130193905817175,
      "eval_runtime": 3.6353,
      "eval_samples_per_second": 94.902,
      "eval_steps_per_second": 12.103,
      "step": 800
    },
    {
      "epoch": 9.310344827586206,
      "grad_norm": 0.9096585512161255,
      "learning_rate": 3.831417624521073e-05,
      "loss": 0.0939,
      "step": 810
    },
    {
      "epoch": 9.425287356321839,
      "grad_norm": 0.6489679217338562,
      "learning_rate": 3.8101319710515115e-05,
      "loss": 0.0894,
      "step": 820
    },
    {
      "epoch": 9.540229885057471,
      "grad_norm": 1.230733871459961,
      "learning_rate": 3.7888463175819497e-05,
      "loss": 0.1027,
      "step": 830
    },
    {
      "epoch": 9.655172413793103,
      "grad_norm": 0.5888535976409912,
      "learning_rate": 3.7675606641123885e-05,
      "loss": 0.1034,
      "step": 840
    },
    {
      "epoch": 9.770114942528735,
      "grad_norm": 1.329034447669983,
      "learning_rate": 3.7462750106428267e-05,
      "loss": 0.0785,
      "step": 850
    },
    {
      "epoch": 9.885057471264368,
      "grad_norm": 0.7995342016220093,
      "learning_rate": 3.7249893571732655e-05,
      "loss": 0.1072,
      "step": 860
    },
    {
      "epoch": 10.0,
      "grad_norm": 2.6494181156158447,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.1007,
      "step": 870
    },
    {
      "epoch": 10.114942528735632,
      "grad_norm": 0.9277469515800476,
      "learning_rate": 3.682418050234142e-05,
      "loss": 0.0835,
      "step": 880
    },
    {
      "epoch": 10.229885057471265,
      "grad_norm": 0.8300806879997253,
      "learning_rate": 3.661132396764581e-05,
      "loss": 0.0853,
      "step": 890
    },
    {
      "epoch": 10.344827586206897,
      "grad_norm": 0.49981820583343506,
      "learning_rate": 3.6398467432950195e-05,
      "loss": 0.0841,
      "step": 900
    },
    {
      "epoch": 10.344827586206897,
      "eval_accuracy": 0.9344927536231884,
      "eval_f1": 0.8383404864091559,
      "eval_loss": 0.16212987899780273,
      "eval_precision": 0.8668639053254438,
      "eval_recall": 0.8116343490304709,
      "eval_runtime": 2.4973,
      "eval_samples_per_second": 138.151,
      "eval_steps_per_second": 17.619,
      "step": 900
    },
    {
      "epoch": 10.459770114942529,
      "grad_norm": 1.2328234910964966,
      "learning_rate": 3.6185610898254577e-05,
      "loss": 0.0825,
      "step": 910
    },
    {
      "epoch": 10.574712643678161,
      "grad_norm": 0.48059162497520447,
      "learning_rate": 3.5972754363558965e-05,
      "loss": 0.0765,
      "step": 920
    },
    {
      "epoch": 10.689655172413794,
      "grad_norm": 1.0247180461883545,
      "learning_rate": 3.5759897828863347e-05,
      "loss": 0.0876,
      "step": 930
    },
    {
      "epoch": 10.804597701149426,
      "grad_norm": 0.6012043356895447,
      "learning_rate": 3.5547041294167735e-05,
      "loss": 0.08,
      "step": 940
    },
    {
      "epoch": 10.919540229885058,
      "grad_norm": 0.8350706696510315,
      "learning_rate": 3.5334184759472117e-05,
      "loss": 0.0889,
      "step": 950
    },
    {
      "epoch": 11.03448275862069,
      "grad_norm": 0.46032950282096863,
      "learning_rate": 3.51213282247765e-05,
      "loss": 0.0815,
      "step": 960
    },
    {
      "epoch": 11.149425287356323,
      "grad_norm": 0.8548628687858582,
      "learning_rate": 3.4908471690080887e-05,
      "loss": 0.09,
      "step": 970
    },
    {
      "epoch": 11.264367816091955,
      "grad_norm": 0.9966102242469788,
      "learning_rate": 3.469561515538527e-05,
      "loss": 0.0707,
      "step": 980
    },
    {
      "epoch": 11.379310344827585,
      "grad_norm": 1.062538981437683,
      "learning_rate": 3.4482758620689657e-05,
      "loss": 0.0652,
      "step": 990
    },
    {
      "epoch": 11.494252873563218,
      "grad_norm": 0.6467556953430176,
      "learning_rate": 3.4269902085994045e-05,
      "loss": 0.0764,
      "step": 1000
    },
    {
      "epoch": 11.494252873563218,
      "eval_accuracy": 0.9359420289855073,
      "eval_f1": 0.8438162544169612,
      "eval_loss": 0.15857626497745514,
      "eval_precision": 0.8614718614718615,
      "eval_recall": 0.8268698060941828,
      "eval_runtime": 2.4932,
      "eval_samples_per_second": 138.379,
      "eval_steps_per_second": 17.648,
      "step": 1000
    },
    {
      "epoch": 11.60919540229885,
      "grad_norm": 0.9217528700828552,
      "learning_rate": 3.4057045551298427e-05,
      "loss": 0.0884,
      "step": 1010
    },
    {
      "epoch": 11.724137931034482,
      "grad_norm": 0.9330977201461792,
      "learning_rate": 3.3844189016602815e-05,
      "loss": 0.0869,
      "step": 1020
    },
    {
      "epoch": 11.839080459770114,
      "grad_norm": 1.4273611307144165,
      "learning_rate": 3.3631332481907197e-05,
      "loss": 0.0846,
      "step": 1030
    },
    {
      "epoch": 11.954022988505747,
      "grad_norm": 1.0101591348648071,
      "learning_rate": 3.341847594721158e-05,
      "loss": 0.0773,
      "step": 1040
    },
    {
      "epoch": 12.068965517241379,
      "grad_norm": 1.1758484840393066,
      "learning_rate": 3.3205619412515967e-05,
      "loss": 0.0744,
      "step": 1050
    },
    {
      "epoch": 12.183908045977011,
      "grad_norm": 0.8224564790725708,
      "learning_rate": 3.299276287782035e-05,
      "loss": 0.0795,
      "step": 1060
    },
    {
      "epoch": 12.298850574712644,
      "grad_norm": 0.468171626329422,
      "learning_rate": 3.277990634312474e-05,
      "loss": 0.0637,
      "step": 1070
    },
    {
      "epoch": 12.413793103448276,
      "grad_norm": 1.1701269149780273,
      "learning_rate": 3.256704980842912e-05,
      "loss": 0.0689,
      "step": 1080
    },
    {
      "epoch": 12.528735632183908,
      "grad_norm": 1.3815059661865234,
      "learning_rate": 3.235419327373351e-05,
      "loss": 0.0861,
      "step": 1090
    },
    {
      "epoch": 12.64367816091954,
      "grad_norm": 1.2272127866744995,
      "learning_rate": 3.2141336739037895e-05,
      "loss": 0.0726,
      "step": 1100
    },
    {
      "epoch": 12.64367816091954,
      "eval_accuracy": 0.9420289855072463,
      "eval_f1": 0.8593530239099859,
      "eval_loss": 0.15456102788448334,
      "eval_precision": 0.8728571428571429,
      "eval_recall": 0.8462603878116344,
      "eval_runtime": 2.5301,
      "eval_samples_per_second": 136.358,
      "eval_steps_per_second": 17.391,
      "step": 1100
    },
    {
      "epoch": 12.758620689655173,
      "grad_norm": 0.9871719479560852,
      "learning_rate": 3.192848020434228e-05,
      "loss": 0.0683,
      "step": 1110
    },
    {
      "epoch": 12.873563218390805,
      "grad_norm": 0.49420595169067383,
      "learning_rate": 3.171562366964666e-05,
      "loss": 0.0721,
      "step": 1120
    },
    {
      "epoch": 12.988505747126437,
      "grad_norm": 0.25016915798187256,
      "learning_rate": 3.150276713495105e-05,
      "loss": 0.0785,
      "step": 1130
    },
    {
      "epoch": 13.10344827586207,
      "grad_norm": 1.1972845792770386,
      "learning_rate": 3.128991060025543e-05,
      "loss": 0.0732,
      "step": 1140
    },
    {
      "epoch": 13.218390804597702,
      "grad_norm": 0.6661795973777771,
      "learning_rate": 3.107705406555981e-05,
      "loss": 0.0709,
      "step": 1150
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 0.5791137218475342,
      "learning_rate": 3.08641975308642e-05,
      "loss": 0.0805,
      "step": 1160
    },
    {
      "epoch": 13.448275862068966,
      "grad_norm": 0.6590031385421753,
      "learning_rate": 3.065134099616858e-05,
      "loss": 0.0665,
      "step": 1170
    },
    {
      "epoch": 13.563218390804598,
      "grad_norm": 0.5040526986122131,
      "learning_rate": 3.043848446147297e-05,
      "loss": 0.0766,
      "step": 1180
    },
    {
      "epoch": 13.678160919540229,
      "grad_norm": 0.785590648651123,
      "learning_rate": 3.0225627926777357e-05,
      "loss": 0.0557,
      "step": 1190
    },
    {
      "epoch": 13.793103448275861,
      "grad_norm": 0.7830068469047546,
      "learning_rate": 3.0012771392081738e-05,
      "loss": 0.0732,
      "step": 1200
    },
    {
      "epoch": 13.793103448275861,
      "eval_accuracy": 0.9408695652173913,
      "eval_f1": 0.8565400843881856,
      "eval_loss": 0.15285652875900269,
      "eval_precision": 0.87,
      "eval_recall": 0.8434903047091413,
      "eval_runtime": 2.5569,
      "eval_samples_per_second": 134.928,
      "eval_steps_per_second": 17.208,
      "step": 1200
    },
    {
      "epoch": 13.908045977011493,
      "grad_norm": 0.5470532178878784,
      "learning_rate": 2.9799914857386123e-05,
      "loss": 0.0665,
      "step": 1210
    },
    {
      "epoch": 14.022988505747126,
      "grad_norm": 0.9854804873466492,
      "learning_rate": 2.9587058322690508e-05,
      "loss": 0.063,
      "step": 1220
    },
    {
      "epoch": 14.137931034482758,
      "grad_norm": 0.562602162361145,
      "learning_rate": 2.9374201787994893e-05,
      "loss": 0.0683,
      "step": 1230
    },
    {
      "epoch": 14.25287356321839,
      "grad_norm": 0.5807965993881226,
      "learning_rate": 2.9161345253299278e-05,
      "loss": 0.0695,
      "step": 1240
    },
    {
      "epoch": 14.367816091954023,
      "grad_norm": 0.659021258354187,
      "learning_rate": 2.894848871860366e-05,
      "loss": 0.069,
      "step": 1250
    },
    {
      "epoch": 14.482758620689655,
      "grad_norm": 0.42295771837234497,
      "learning_rate": 2.8735632183908045e-05,
      "loss": 0.0692,
      "step": 1260
    },
    {
      "epoch": 14.597701149425287,
      "grad_norm": 1.1786874532699585,
      "learning_rate": 2.852277564921243e-05,
      "loss": 0.0564,
      "step": 1270
    },
    {
      "epoch": 14.71264367816092,
      "grad_norm": 0.7675245404243469,
      "learning_rate": 2.8309919114516818e-05,
      "loss": 0.063,
      "step": 1280
    },
    {
      "epoch": 14.827586206896552,
      "grad_norm": 0.6836528182029724,
      "learning_rate": 2.8097062579821203e-05,
      "loss": 0.0711,
      "step": 1290
    },
    {
      "epoch": 14.942528735632184,
      "grad_norm": 0.6420086026191711,
      "learning_rate": 2.7884206045125588e-05,
      "loss": 0.0626,
      "step": 1300
    },
    {
      "epoch": 14.942528735632184,
      "eval_accuracy": 0.9376811594202898,
      "eval_f1": 0.8484848484848485,
      "eval_loss": 0.1589481383562088,
      "eval_precision": 0.8637015781922525,
      "eval_recall": 0.8337950138504155,
      "eval_runtime": 2.7864,
      "eval_samples_per_second": 123.815,
      "eval_steps_per_second": 15.791,
      "step": 1300
    },
    {
      "epoch": 15.057471264367816,
      "grad_norm": 0.5504136681556702,
      "learning_rate": 2.7671349510429973e-05,
      "loss": 0.0451,
      "step": 1310
    },
    {
      "epoch": 15.172413793103448,
      "grad_norm": 1.0383538007736206,
      "learning_rate": 2.745849297573436e-05,
      "loss": 0.0569,
      "step": 1320
    },
    {
      "epoch": 15.28735632183908,
      "grad_norm": 0.42968350648880005,
      "learning_rate": 2.724563644103874e-05,
      "loss": 0.0551,
      "step": 1330
    },
    {
      "epoch": 15.402298850574713,
      "grad_norm": 0.8487522602081299,
      "learning_rate": 2.7032779906343125e-05,
      "loss": 0.0501,
      "step": 1340
    },
    {
      "epoch": 15.517241379310345,
      "grad_norm": 0.8882860541343689,
      "learning_rate": 2.681992337164751e-05,
      "loss": 0.0467,
      "step": 1350
    },
    {
      "epoch": 15.632183908045977,
      "grad_norm": 1.2020477056503296,
      "learning_rate": 2.6607066836951895e-05,
      "loss": 0.0649,
      "step": 1360
    },
    {
      "epoch": 15.74712643678161,
      "grad_norm": 0.6664167046546936,
      "learning_rate": 2.6394210302256277e-05,
      "loss": 0.0462,
      "step": 1370
    },
    {
      "epoch": 15.862068965517242,
      "grad_norm": 0.6712772250175476,
      "learning_rate": 2.618135376756067e-05,
      "loss": 0.0729,
      "step": 1380
    },
    {
      "epoch": 15.977011494252874,
      "grad_norm": 0.5807361602783203,
      "learning_rate": 2.5968497232865053e-05,
      "loss": 0.0532,
      "step": 1390
    },
    {
      "epoch": 16.091954022988507,
      "grad_norm": 0.4083256423473358,
      "learning_rate": 2.5755640698169435e-05,
      "loss": 0.0481,
      "step": 1400
    },
    {
      "epoch": 16.091954022988507,
      "eval_accuracy": 0.9394202898550724,
      "eval_f1": 0.8510334996436208,
      "eval_loss": 0.16117151081562042,
      "eval_precision": 0.8766519823788547,
      "eval_recall": 0.8268698060941828,
      "eval_runtime": 3.3334,
      "eval_samples_per_second": 103.498,
      "eval_steps_per_second": 13.2,
      "step": 1400
    },
    {
      "epoch": 16.20689655172414,
      "grad_norm": 0.8922817707061768,
      "learning_rate": 2.554278416347382e-05,
      "loss": 0.0674,
      "step": 1410
    },
    {
      "epoch": 16.32183908045977,
      "grad_norm": 0.9631970524787903,
      "learning_rate": 2.5329927628778205e-05,
      "loss": 0.0503,
      "step": 1420
    },
    {
      "epoch": 16.436781609195403,
      "grad_norm": 0.8879241943359375,
      "learning_rate": 2.511707109408259e-05,
      "loss": 0.0482,
      "step": 1430
    },
    {
      "epoch": 16.551724137931036,
      "grad_norm": 0.7775533199310303,
      "learning_rate": 2.4904214559386975e-05,
      "loss": 0.0636,
      "step": 1440
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 0.9835919737815857,
      "learning_rate": 2.4691358024691357e-05,
      "loss": 0.0568,
      "step": 1450
    },
    {
      "epoch": 16.7816091954023,
      "grad_norm": 0.7925294041633606,
      "learning_rate": 2.4478501489995745e-05,
      "loss": 0.0529,
      "step": 1460
    },
    {
      "epoch": 16.896551724137932,
      "grad_norm": 0.6245427131652832,
      "learning_rate": 2.426564495530013e-05,
      "loss": 0.0641,
      "step": 1470
    },
    {
      "epoch": 17.011494252873565,
      "grad_norm": 0.5181954503059387,
      "learning_rate": 2.4052788420604515e-05,
      "loss": 0.046,
      "step": 1480
    },
    {
      "epoch": 17.126436781609197,
      "grad_norm": 0.40600207448005676,
      "learning_rate": 2.3839931885908897e-05,
      "loss": 0.039,
      "step": 1490
    },
    {
      "epoch": 17.24137931034483,
      "grad_norm": 0.7081565260887146,
      "learning_rate": 2.3627075351213285e-05,
      "loss": 0.0507,
      "step": 1500
    },
    {
      "epoch": 17.24137931034483,
      "eval_accuracy": 0.9339130434782609,
      "eval_f1": 0.8394366197183099,
      "eval_loss": 0.1679152399301529,
      "eval_precision": 0.8538681948424068,
      "eval_recall": 0.8254847645429363,
      "eval_runtime": 2.512,
      "eval_samples_per_second": 137.339,
      "eval_steps_per_second": 17.516,
      "step": 1500
    },
    {
      "epoch": 17.35632183908046,
      "grad_norm": 0.9278731346130371,
      "learning_rate": 2.341421881651767e-05,
      "loss": 0.0546,
      "step": 1510
    },
    {
      "epoch": 17.47126436781609,
      "grad_norm": 1.362691044807434,
      "learning_rate": 2.3201362281822055e-05,
      "loss": 0.0729,
      "step": 1520
    },
    {
      "epoch": 17.586206896551722,
      "grad_norm": 0.6156861782073975,
      "learning_rate": 2.2988505747126437e-05,
      "loss": 0.0587,
      "step": 1530
    },
    {
      "epoch": 17.701149425287355,
      "grad_norm": 0.530103862285614,
      "learning_rate": 2.277564921243082e-05,
      "loss": 0.0488,
      "step": 1540
    },
    {
      "epoch": 17.816091954022987,
      "grad_norm": 0.6204285025596619,
      "learning_rate": 2.256279267773521e-05,
      "loss": 0.0499,
      "step": 1550
    },
    {
      "epoch": 17.93103448275862,
      "grad_norm": 0.4868471920490265,
      "learning_rate": 2.234993614303959e-05,
      "loss": 0.0536,
      "step": 1560
    },
    {
      "epoch": 18.04597701149425,
      "grad_norm": 0.5951109528541565,
      "learning_rate": 2.2137079608343977e-05,
      "loss": 0.0436,
      "step": 1570
    },
    {
      "epoch": 18.160919540229884,
      "grad_norm": 1.3129435777664185,
      "learning_rate": 2.192422307364836e-05,
      "loss": 0.0455,
      "step": 1580
    },
    {
      "epoch": 18.275862068965516,
      "grad_norm": 0.4130817651748657,
      "learning_rate": 2.1711366538952747e-05,
      "loss": 0.0539,
      "step": 1590
    },
    {
      "epoch": 18.39080459770115,
      "grad_norm": 0.6867007613182068,
      "learning_rate": 2.149851000425713e-05,
      "loss": 0.0446,
      "step": 1600
    },
    {
      "epoch": 18.39080459770115,
      "eval_accuracy": 0.9417391304347826,
      "eval_f1": 0.8597348220516399,
      "eval_loss": 0.16227415204048157,
      "eval_precision": 0.8663853727144867,
      "eval_recall": 0.853185595567867,
      "eval_runtime": 2.7697,
      "eval_samples_per_second": 124.564,
      "eval_steps_per_second": 15.886,
      "step": 1600
    },
    {
      "epoch": 18.50574712643678,
      "grad_norm": 0.44690120220184326,
      "learning_rate": 2.1285653469561517e-05,
      "loss": 0.0411,
      "step": 1610
    },
    {
      "epoch": 18.620689655172413,
      "grad_norm": 0.47931063175201416,
      "learning_rate": 2.10727969348659e-05,
      "loss": 0.0487,
      "step": 1620
    },
    {
      "epoch": 18.735632183908045,
      "grad_norm": 0.5977224707603455,
      "learning_rate": 2.0859940400170287e-05,
      "loss": 0.0501,
      "step": 1630
    },
    {
      "epoch": 18.850574712643677,
      "grad_norm": 0.5454962253570557,
      "learning_rate": 2.064708386547467e-05,
      "loss": 0.0516,
      "step": 1640
    },
    {
      "epoch": 18.96551724137931,
      "grad_norm": 0.5072124004364014,
      "learning_rate": 2.0434227330779057e-05,
      "loss": 0.0407,
      "step": 1650
    },
    {
      "epoch": 19.080459770114942,
      "grad_norm": 0.8368410468101501,
      "learning_rate": 2.022137079608344e-05,
      "loss": 0.0481,
      "step": 1660
    },
    {
      "epoch": 19.195402298850574,
      "grad_norm": 1.143817663192749,
      "learning_rate": 2.0008514261387827e-05,
      "loss": 0.0406,
      "step": 1670
    },
    {
      "epoch": 19.310344827586206,
      "grad_norm": 0.8621155023574829,
      "learning_rate": 1.979565772669221e-05,
      "loss": 0.0472,
      "step": 1680
    },
    {
      "epoch": 19.42528735632184,
      "grad_norm": 0.8814654350280762,
      "learning_rate": 1.9582801191996593e-05,
      "loss": 0.0403,
      "step": 1690
    },
    {
      "epoch": 19.54022988505747,
      "grad_norm": 0.37576496601104736,
      "learning_rate": 1.936994465730098e-05,
      "loss": 0.0498,
      "step": 1700
    },
    {
      "epoch": 19.54022988505747,
      "eval_accuracy": 0.9417391304347826,
      "eval_f1": 0.860125260960334,
      "eval_loss": 0.16253642737865448,
      "eval_precision": 0.8643356643356643,
      "eval_recall": 0.8559556786703602,
      "eval_runtime": 3.3471,
      "eval_samples_per_second": 103.074,
      "eval_steps_per_second": 13.146,
      "step": 1700
    },
    {
      "epoch": 19.655172413793103,
      "grad_norm": 1.052512764930725,
      "learning_rate": 1.9157088122605367e-05,
      "loss": 0.0569,
      "step": 1710
    },
    {
      "epoch": 19.770114942528735,
      "grad_norm": 0.47423890233039856,
      "learning_rate": 1.8944231587909748e-05,
      "loss": 0.039,
      "step": 1720
    },
    {
      "epoch": 19.885057471264368,
      "grad_norm": 0.922591507434845,
      "learning_rate": 1.8731375053214133e-05,
      "loss": 0.0376,
      "step": 1730
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.3589046001434326,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.0496,
      "step": 1740
    },
    {
      "epoch": 20.114942528735632,
      "grad_norm": 0.8147189617156982,
      "learning_rate": 1.8305661983822907e-05,
      "loss": 0.0355,
      "step": 1750
    },
    {
      "epoch": 20.229885057471265,
      "grad_norm": 0.2483612447977066,
      "learning_rate": 1.8092805449127288e-05,
      "loss": 0.0432,
      "step": 1760
    },
    {
      "epoch": 20.344827586206897,
      "grad_norm": 0.29129087924957275,
      "learning_rate": 1.7879948914431673e-05,
      "loss": 0.0394,
      "step": 1770
    },
    {
      "epoch": 20.45977011494253,
      "grad_norm": 0.6358793377876282,
      "learning_rate": 1.7667092379736058e-05,
      "loss": 0.0453,
      "step": 1780
    },
    {
      "epoch": 20.57471264367816,
      "grad_norm": 0.3951164782047272,
      "learning_rate": 1.7454235845040443e-05,
      "loss": 0.0499,
      "step": 1790
    },
    {
      "epoch": 20.689655172413794,
      "grad_norm": 0.5246742367744446,
      "learning_rate": 1.7241379310344828e-05,
      "loss": 0.0458,
      "step": 1800
    },
    {
      "epoch": 20.689655172413794,
      "eval_accuracy": 0.9397101449275362,
      "eval_f1": 0.8533145275035261,
      "eval_loss": 0.16005827486515045,
      "eval_precision": 0.8692528735632183,
      "eval_recall": 0.8379501385041551,
      "eval_runtime": 2.5451,
      "eval_samples_per_second": 135.555,
      "eval_steps_per_second": 17.288,
      "step": 1800
    },
    {
      "epoch": 20.804597701149426,
      "grad_norm": 0.4585019648075104,
      "learning_rate": 1.7028522775649213e-05,
      "loss": 0.0466,
      "step": 1810
    },
    {
      "epoch": 20.919540229885058,
      "grad_norm": 0.5193343162536621,
      "learning_rate": 1.6815666240953598e-05,
      "loss": 0.0321,
      "step": 1820
    },
    {
      "epoch": 21.03448275862069,
      "grad_norm": 0.3044123649597168,
      "learning_rate": 1.6602809706257983e-05,
      "loss": 0.0391,
      "step": 1830
    },
    {
      "epoch": 21.149425287356323,
      "grad_norm": 0.3363288342952728,
      "learning_rate": 1.638995317156237e-05,
      "loss": 0.0344,
      "step": 1840
    },
    {
      "epoch": 21.264367816091955,
      "grad_norm": 0.38722899556159973,
      "learning_rate": 1.6177096636866753e-05,
      "loss": 0.0352,
      "step": 1850
    },
    {
      "epoch": 21.379310344827587,
      "grad_norm": 0.5276902914047241,
      "learning_rate": 1.596424010217114e-05,
      "loss": 0.0354,
      "step": 1860
    },
    {
      "epoch": 21.49425287356322,
      "grad_norm": 0.37170952558517456,
      "learning_rate": 1.5751383567475523e-05,
      "loss": 0.0369,
      "step": 1870
    },
    {
      "epoch": 21.60919540229885,
      "grad_norm": 1.0527151823043823,
      "learning_rate": 1.5538527032779905e-05,
      "loss": 0.032,
      "step": 1880
    },
    {
      "epoch": 21.724137931034484,
      "grad_norm": 0.42286577820777893,
      "learning_rate": 1.532567049808429e-05,
      "loss": 0.0417,
      "step": 1890
    },
    {
      "epoch": 21.839080459770116,
      "grad_norm": 0.4092840254306793,
      "learning_rate": 1.5112813963388678e-05,
      "loss": 0.0307,
      "step": 1900
    },
    {
      "epoch": 21.839080459770116,
      "eval_accuracy": 0.9431884057971015,
      "eval_f1": 0.8636995827538247,
      "eval_loss": 0.16264864802360535,
      "eval_precision": 0.86731843575419,
      "eval_recall": 0.8601108033240997,
      "eval_runtime": 2.5207,
      "eval_samples_per_second": 136.867,
      "eval_steps_per_second": 17.455,
      "step": 1900
    },
    {
      "epoch": 21.95402298850575,
      "grad_norm": 0.4764251708984375,
      "learning_rate": 1.4899957428693062e-05,
      "loss": 0.042,
      "step": 1910
    },
    {
      "epoch": 22.06896551724138,
      "grad_norm": 0.925857424736023,
      "learning_rate": 1.4687100893997447e-05,
      "loss": 0.0342,
      "step": 1920
    },
    {
      "epoch": 22.183908045977013,
      "grad_norm": 0.6081854701042175,
      "learning_rate": 1.447424435930183e-05,
      "loss": 0.0443,
      "step": 1930
    },
    {
      "epoch": 22.298850574712645,
      "grad_norm": 0.35534176230430603,
      "learning_rate": 1.4261387824606215e-05,
      "loss": 0.0296,
      "step": 1940
    },
    {
      "epoch": 22.413793103448278,
      "grad_norm": 0.5677205324172974,
      "learning_rate": 1.4048531289910602e-05,
      "loss": 0.0319,
      "step": 1950
    },
    {
      "epoch": 22.52873563218391,
      "grad_norm": 0.9339087009429932,
      "learning_rate": 1.3835674755214987e-05,
      "loss": 0.0367,
      "step": 1960
    },
    {
      "epoch": 22.64367816091954,
      "grad_norm": 0.30769288539886475,
      "learning_rate": 1.362281822051937e-05,
      "loss": 0.0372,
      "step": 1970
    },
    {
      "epoch": 22.75862068965517,
      "grad_norm": 0.4518299698829651,
      "learning_rate": 1.3409961685823755e-05,
      "loss": 0.0501,
      "step": 1980
    },
    {
      "epoch": 22.873563218390803,
      "grad_norm": 0.3183022737503052,
      "learning_rate": 1.3197105151128138e-05,
      "loss": 0.0338,
      "step": 1990
    },
    {
      "epoch": 22.988505747126435,
      "grad_norm": 0.7140583395957947,
      "learning_rate": 1.2984248616432527e-05,
      "loss": 0.0334,
      "step": 2000
    },
    {
      "epoch": 22.988505747126435,
      "eval_accuracy": 0.9443478260869566,
      "eval_f1": 0.8642149929278642,
      "eval_loss": 0.16209261119365692,
      "eval_precision": 0.8829479768786127,
      "eval_recall": 0.8462603878116344,
      "eval_runtime": 3.2664,
      "eval_samples_per_second": 105.621,
      "eval_steps_per_second": 13.471,
      "step": 2000
    },
    {
      "epoch": 23.103448275862068,
      "grad_norm": 0.647758424282074,
      "learning_rate": 1.277139208173691e-05,
      "loss": 0.036,
      "step": 2010
    },
    {
      "epoch": 23.2183908045977,
      "grad_norm": 0.7994110584259033,
      "learning_rate": 1.2558535547041295e-05,
      "loss": 0.0313,
      "step": 2020
    },
    {
      "epoch": 23.333333333333332,
      "grad_norm": 0.33477237820625305,
      "learning_rate": 1.2345679012345678e-05,
      "loss": 0.034,
      "step": 2030
    },
    {
      "epoch": 23.448275862068964,
      "grad_norm": 0.17108239233493805,
      "learning_rate": 1.2132822477650065e-05,
      "loss": 0.0262,
      "step": 2040
    },
    {
      "epoch": 23.563218390804597,
      "grad_norm": 0.40771496295928955,
      "learning_rate": 1.1919965942954448e-05,
      "loss": 0.0375,
      "step": 2050
    },
    {
      "epoch": 23.67816091954023,
      "grad_norm": 1.7193657159805298,
      "learning_rate": 1.1707109408258835e-05,
      "loss": 0.0348,
      "step": 2060
    },
    {
      "epoch": 23.79310344827586,
      "grad_norm": 0.2730618715286255,
      "learning_rate": 1.1494252873563218e-05,
      "loss": 0.0335,
      "step": 2070
    },
    {
      "epoch": 23.908045977011493,
      "grad_norm": 0.12540611624717712,
      "learning_rate": 1.1281396338867605e-05,
      "loss": 0.0338,
      "step": 2080
    },
    {
      "epoch": 24.022988505747126,
      "grad_norm": 0.6330766081809998,
      "learning_rate": 1.1068539804171988e-05,
      "loss": 0.0409,
      "step": 2090
    },
    {
      "epoch": 24.137931034482758,
      "grad_norm": 0.5552668571472168,
      "learning_rate": 1.0855683269476373e-05,
      "loss": 0.0339,
      "step": 2100
    },
    {
      "epoch": 24.137931034482758,
      "eval_accuracy": 0.9434782608695652,
      "eval_f1": 0.8644892286309938,
      "eval_loss": 0.1680324375629425,
      "eval_precision": 0.8675034867503487,
      "eval_recall": 0.8614958448753463,
      "eval_runtime": 2.5297,
      "eval_samples_per_second": 136.381,
      "eval_steps_per_second": 17.394,
      "step": 2100
    },
    {
      "epoch": 24.25287356321839,
      "grad_norm": 0.8327674269676208,
      "learning_rate": 1.0642826734780758e-05,
      "loss": 0.0307,
      "step": 2110
    },
    {
      "epoch": 24.367816091954023,
      "grad_norm": 0.33239325881004333,
      "learning_rate": 1.0429970200085143e-05,
      "loss": 0.0333,
      "step": 2120
    },
    {
      "epoch": 24.482758620689655,
      "grad_norm": 0.44145530462265015,
      "learning_rate": 1.0217113665389528e-05,
      "loss": 0.037,
      "step": 2130
    },
    {
      "epoch": 24.597701149425287,
      "grad_norm": 0.7869518995285034,
      "learning_rate": 1.0004257130693913e-05,
      "loss": 0.0364,
      "step": 2140
    },
    {
      "epoch": 24.71264367816092,
      "grad_norm": 0.31890323758125305,
      "learning_rate": 9.791400595998297e-06,
      "loss": 0.0329,
      "step": 2150
    },
    {
      "epoch": 24.82758620689655,
      "grad_norm": 0.53341144323349,
      "learning_rate": 9.578544061302683e-06,
      "loss": 0.0481,
      "step": 2160
    },
    {
      "epoch": 24.942528735632184,
      "grad_norm": 0.4981224834918976,
      "learning_rate": 9.365687526607067e-06,
      "loss": 0.0368,
      "step": 2170
    },
    {
      "epoch": 25.057471264367816,
      "grad_norm": 1.0361753702163696,
      "learning_rate": 9.152830991911453e-06,
      "loss": 0.0464,
      "step": 2180
    },
    {
      "epoch": 25.17241379310345,
      "grad_norm": 0.1470736712217331,
      "learning_rate": 8.939974457215837e-06,
      "loss": 0.0372,
      "step": 2190
    },
    {
      "epoch": 25.28735632183908,
      "grad_norm": 0.5003727078437805,
      "learning_rate": 8.727117922520222e-06,
      "loss": 0.0222,
      "step": 2200
    },
    {
      "epoch": 25.28735632183908,
      "eval_accuracy": 0.9394202898550724,
      "eval_f1": 0.853743876836949,
      "eval_loss": 0.16558243334293365,
      "eval_precision": 0.8628005657708628,
      "eval_recall": 0.8448753462603878,
      "eval_runtime": 2.7223,
      "eval_samples_per_second": 126.732,
      "eval_steps_per_second": 16.163,
      "step": 2200
    },
    {
      "epoch": 25.402298850574713,
      "grad_norm": 1.032658576965332,
      "learning_rate": 8.514261387824607e-06,
      "loss": 0.036,
      "step": 2210
    },
    {
      "epoch": 25.517241379310345,
      "grad_norm": 0.3298107087612152,
      "learning_rate": 8.301404853128992e-06,
      "loss": 0.0311,
      "step": 2220
    },
    {
      "epoch": 25.632183908045977,
      "grad_norm": 1.8201491832733154,
      "learning_rate": 8.088548318433377e-06,
      "loss": 0.0373,
      "step": 2230
    },
    {
      "epoch": 25.74712643678161,
      "grad_norm": 0.5777905583381653,
      "learning_rate": 7.875691783737762e-06,
      "loss": 0.0323,
      "step": 2240
    },
    {
      "epoch": 25.862068965517242,
      "grad_norm": 0.3212037980556488,
      "learning_rate": 7.662835249042145e-06,
      "loss": 0.0337,
      "step": 2250
    },
    {
      "epoch": 25.977011494252874,
      "grad_norm": 1.0763506889343262,
      "learning_rate": 7.449978714346531e-06,
      "loss": 0.0455,
      "step": 2260
    },
    {
      "epoch": 26.091954022988507,
      "grad_norm": 0.7833639979362488,
      "learning_rate": 7.237122179650915e-06,
      "loss": 0.0289,
      "step": 2270
    },
    {
      "epoch": 26.20689655172414,
      "grad_norm": 1.0893276929855347,
      "learning_rate": 7.024265644955301e-06,
      "loss": 0.037,
      "step": 2280
    },
    {
      "epoch": 26.32183908045977,
      "grad_norm": 0.534814715385437,
      "learning_rate": 6.811409110259685e-06,
      "loss": 0.0309,
      "step": 2290
    },
    {
      "epoch": 26.436781609195403,
      "grad_norm": 1.0022532939910889,
      "learning_rate": 6.598552575564069e-06,
      "loss": 0.026,
      "step": 2300
    },
    {
      "epoch": 26.436781609195403,
      "eval_accuracy": 0.9385507246376812,
      "eval_f1": 0.8515406162464986,
      "eval_loss": 0.16867290437221527,
      "eval_precision": 0.8611898016997167,
      "eval_recall": 0.8421052631578947,
      "eval_runtime": 3.2071,
      "eval_samples_per_second": 107.575,
      "eval_steps_per_second": 13.72,
      "step": 2300
    },
    {
      "epoch": 26.551724137931036,
      "grad_norm": 0.396712988615036,
      "learning_rate": 6.385696040868455e-06,
      "loss": 0.0422,
      "step": 2310
    },
    {
      "epoch": 26.666666666666668,
      "grad_norm": 0.6816790103912354,
      "learning_rate": 6.172839506172839e-06,
      "loss": 0.0259,
      "step": 2320
    },
    {
      "epoch": 26.7816091954023,
      "grad_norm": 0.44069159030914307,
      "learning_rate": 5.959982971477224e-06,
      "loss": 0.0314,
      "step": 2330
    },
    {
      "epoch": 26.896551724137932,
      "grad_norm": 0.8928768634796143,
      "learning_rate": 5.747126436781609e-06,
      "loss": 0.0451,
      "step": 2340
    },
    {
      "epoch": 27.011494252873565,
      "grad_norm": 0.15559493005275726,
      "learning_rate": 5.534269902085994e-06,
      "loss": 0.0292,
      "step": 2350
    },
    {
      "epoch": 27.126436781609197,
      "grad_norm": 0.57392817735672,
      "learning_rate": 5.321413367390379e-06,
      "loss": 0.0339,
      "step": 2360
    },
    {
      "epoch": 27.24137931034483,
      "grad_norm": 0.4282006621360779,
      "learning_rate": 5.108556832694764e-06,
      "loss": 0.0272,
      "step": 2370
    },
    {
      "epoch": 27.35632183908046,
      "grad_norm": 0.3765332102775574,
      "learning_rate": 4.895700297999148e-06,
      "loss": 0.0224,
      "step": 2380
    },
    {
      "epoch": 27.47126436781609,
      "grad_norm": 0.26194748282432556,
      "learning_rate": 4.682843763303533e-06,
      "loss": 0.025,
      "step": 2390
    },
    {
      "epoch": 27.586206896551722,
      "grad_norm": 0.6783250570297241,
      "learning_rate": 4.469987228607918e-06,
      "loss": 0.0353,
      "step": 2400
    },
    {
      "epoch": 27.586206896551722,
      "eval_accuracy": 0.9402898550724638,
      "eval_f1": 0.8555399719495091,
      "eval_loss": 0.16655248403549194,
      "eval_precision": 0.8664772727272727,
      "eval_recall": 0.8448753462603878,
      "eval_runtime": 2.5159,
      "eval_samples_per_second": 137.126,
      "eval_steps_per_second": 17.489,
      "step": 2400
    },
    {
      "epoch": 27.701149425287355,
      "grad_norm": 0.7945173382759094,
      "learning_rate": 4.257130693912303e-06,
      "loss": 0.0346,
      "step": 2410
    },
    {
      "epoch": 27.816091954022987,
      "grad_norm": 0.34426242113113403,
      "learning_rate": 4.044274159216688e-06,
      "loss": 0.0354,
      "step": 2420
    },
    {
      "epoch": 27.93103448275862,
      "grad_norm": 0.2353845238685608,
      "learning_rate": 3.8314176245210725e-06,
      "loss": 0.0306,
      "step": 2430
    },
    {
      "epoch": 28.04597701149425,
      "grad_norm": 0.7292235493659973,
      "learning_rate": 3.6185610898254575e-06,
      "loss": 0.0268,
      "step": 2440
    },
    {
      "epoch": 28.160919540229884,
      "grad_norm": 0.65985107421875,
      "learning_rate": 3.4057045551298425e-06,
      "loss": 0.0235,
      "step": 2450
    },
    {
      "epoch": 28.275862068965516,
      "grad_norm": 0.36814385652542114,
      "learning_rate": 3.1928480204342275e-06,
      "loss": 0.0234,
      "step": 2460
    },
    {
      "epoch": 28.39080459770115,
      "grad_norm": 0.2845553159713745,
      "learning_rate": 2.979991485738612e-06,
      "loss": 0.0249,
      "step": 2470
    },
    {
      "epoch": 28.50574712643678,
      "grad_norm": 0.36483725905418396,
      "learning_rate": 2.767134951042997e-06,
      "loss": 0.0267,
      "step": 2480
    },
    {
      "epoch": 28.620689655172413,
      "grad_norm": 0.5152461528778076,
      "learning_rate": 2.554278416347382e-06,
      "loss": 0.0325,
      "step": 2490
    },
    {
      "epoch": 28.735632183908045,
      "grad_norm": 0.775412380695343,
      "learning_rate": 2.3414218816517667e-06,
      "loss": 0.0294,
      "step": 2500
    },
    {
      "epoch": 28.735632183908045,
      "eval_accuracy": 0.9428985507246377,
      "eval_f1": 0.8613652357494722,
      "eval_loss": 0.1659679114818573,
      "eval_precision": 0.8755364806866953,
      "eval_recall": 0.8476454293628809,
      "eval_runtime": 2.5486,
      "eval_samples_per_second": 135.371,
      "eval_steps_per_second": 17.265,
      "step": 2500
    },
    {
      "epoch": 28.850574712643677,
      "grad_norm": 0.25289997458457947,
      "learning_rate": 2.1285653469561517e-06,
      "loss": 0.0285,
      "step": 2510
    },
    {
      "epoch": 28.96551724137931,
      "grad_norm": 0.11768297851085663,
      "learning_rate": 1.9157088122605362e-06,
      "loss": 0.0233,
      "step": 2520
    },
    {
      "epoch": 29.080459770114942,
      "grad_norm": 0.4232161045074463,
      "learning_rate": 1.7028522775649212e-06,
      "loss": 0.021,
      "step": 2530
    },
    {
      "epoch": 29.195402298850574,
      "grad_norm": 0.6776716709136963,
      "learning_rate": 1.489995742869306e-06,
      "loss": 0.0282,
      "step": 2540
    },
    {
      "epoch": 29.310344827586206,
      "grad_norm": 0.2837054431438446,
      "learning_rate": 1.277139208173691e-06,
      "loss": 0.034,
      "step": 2550
    },
    {
      "epoch": 29.42528735632184,
      "grad_norm": 0.13374456763267517,
      "learning_rate": 1.0642826734780758e-06,
      "loss": 0.0261,
      "step": 2560
    },
    {
      "epoch": 29.54022988505747,
      "grad_norm": 0.32526296377182007,
      "learning_rate": 8.514261387824606e-07,
      "loss": 0.0318,
      "step": 2570
    },
    {
      "epoch": 29.655172413793103,
      "grad_norm": 0.7919621467590332,
      "learning_rate": 6.385696040868455e-07,
      "loss": 0.0305,
      "step": 2580
    },
    {
      "epoch": 29.770114942528735,
      "grad_norm": 0.1575266569852829,
      "learning_rate": 4.257130693912303e-07,
      "loss": 0.0298,
      "step": 2590
    },
    {
      "epoch": 29.885057471264368,
      "grad_norm": 0.35859954357147217,
      "learning_rate": 2.1285653469561516e-07,
      "loss": 0.0243,
      "step": 2600
    },
    {
      "epoch": 29.885057471264368,
      "eval_accuracy": 0.9423188405797102,
      "eval_f1": 0.8589652728561304,
      "eval_loss": 0.16644859313964844,
      "eval_precision": 0.8795355587808418,
      "eval_recall": 0.8393351800554016,
      "eval_runtime": 2.5418,
      "eval_samples_per_second": 135.729,
      "eval_steps_per_second": 17.31,
      "step": 2600
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.5356839299201965,
      "learning_rate": 0.0,
      "loss": 0.0375,
      "step": 2610
    },
    {
      "epoch": 30.0,
      "step": 2610,
      "total_flos": 3.2060734740537754e+18,
      "train_loss": 0.09901393407606074,
      "train_runtime": 730.7896,
      "train_samples_per_second": 56.61,
      "train_steps_per_second": 3.571
    }
  ],
  "logging_steps": 10,
  "max_steps": 2610,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.2060734740537754e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}