{
  "best_global_step": 25000,
  "best_metric": 0.10385506600141525,
  "best_model_checkpoint": "myanmar_text_segmentation_model/checkpoint-25000",
  "epoch": 2.297160709363227,
  "eval_steps": 1000,
  "global_step": 25000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04594321418726454,
      "grad_norm": 0.577109158039093,
      "learning_rate": 1.9816594688964442e-05,
      "loss": 0.1144,
      "step": 500
    },
    {
      "epoch": 0.09188642837452908,
      "grad_norm": 0.8819814324378967,
      "learning_rate": 1.9632821832215383e-05,
      "loss": 0.1133,
      "step": 1000
    },
    {
      "epoch": 0.09188642837452908,
      "eval_accuracy": 0.9564004775738034,
      "eval_f1": 0.7581789101913106,
      "eval_loss": 0.11448723077774048,
      "eval_precision": 0.7703745996379334,
      "eval_recall": 0.7463633395484631,
      "eval_runtime": 465.4607,
      "eval_samples_per_second": 61.528,
      "eval_steps_per_second": 1.231,
      "step": 1000
    },
    {
      "epoch": 0.1378296425617936,
      "grad_norm": 0.5137984156608582,
      "learning_rate": 1.9449048975466325e-05,
      "loss": 0.1119,
      "step": 1500
    },
    {
      "epoch": 0.18377285674905816,
      "grad_norm": 0.6753482222557068,
      "learning_rate": 1.9265276118717266e-05,
      "loss": 0.1115,
      "step": 2000
    },
    {
      "epoch": 0.18377285674905816,
      "eval_accuracy": 0.9565513036953667,
      "eval_f1": 0.7631813253052621,
      "eval_loss": 0.11509840190410614,
      "eval_precision": 0.7619568901058112,
      "eval_recall": 0.7644097020785157,
      "eval_runtime": 471.4797,
      "eval_samples_per_second": 60.743,
      "eval_steps_per_second": 1.215,
      "step": 2000
    },
    {
      "epoch": 0.2297160709363227,
      "grad_norm": 0.46862444281578064,
      "learning_rate": 1.9081503261968207e-05,
      "loss": 0.1083,
      "step": 2500
    },
    {
      "epoch": 0.2756592851235872,
      "grad_norm": 0.6514066457748413,
      "learning_rate": 1.889773040521915e-05,
      "loss": 0.1085,
      "step": 3000
    },
    {
      "epoch": 0.2756592851235872,
      "eval_accuracy": 0.955793370748351,
      "eval_f1": 0.7617613129080717,
      "eval_loss": 0.11627302318811417,
      "eval_precision": 0.7514122514122514,
      "eval_recall": 0.7723994268768838,
      "eval_runtime": 464.541,
      "eval_samples_per_second": 61.65,
      "eval_steps_per_second": 1.233,
      "step": 3000
    },
    {
      "epoch": 0.3216024993108518,
      "grad_norm": 0.5837205052375793,
      "learning_rate": 1.8713957548470093e-05,
      "loss": 0.1068,
      "step": 3500
    },
    {
      "epoch": 0.3675457134981163,
      "grad_norm": 0.590044379234314,
      "learning_rate": 1.8530184691721034e-05,
      "loss": 0.1076,
      "step": 4000
    },
    {
      "epoch": 0.3675457134981163,
      "eval_accuracy": 0.9567566300121167,
      "eval_f1": 0.7569129075628852,
      "eval_loss": 0.11681018769741058,
      "eval_precision": 0.7790560842692136,
      "eval_recall": 0.7359936967250493,
      "eval_runtime": 464.1614,
      "eval_samples_per_second": 61.701,
      "eval_steps_per_second": 1.234,
      "step": 4000
    },
    {
      "epoch": 0.41348892768538087,
      "grad_norm": 0.43107616901397705,
      "learning_rate": 1.8346411834971975e-05,
      "loss": 0.1071,
      "step": 4500
    },
    {
      "epoch": 0.4594321418726454,
      "grad_norm": 0.6132990121841431,
      "learning_rate": 1.8162638978222916e-05,
      "loss": 0.1056,
      "step": 5000
    },
    {
      "epoch": 0.4594321418726454,
      "eval_accuracy": 0.9576032842071108,
      "eval_f1": 0.7661190672004459,
      "eval_loss": 0.11183962970972061,
      "eval_precision": 0.7719609680194174,
      "eval_recall": 0.7603649207638403,
      "eval_runtime": 473.1438,
      "eval_samples_per_second": 60.529,
      "eval_steps_per_second": 1.211,
      "step": 5000
    },
    {
      "epoch": 0.5053753560599099,
      "grad_norm": 0.4892466962337494,
      "learning_rate": 1.7978866121473858e-05,
      "loss": 0.1154,
      "step": 5500
    },
    {
      "epoch": 0.5513185702471745,
      "grad_norm": 0.8282744884490967,
      "learning_rate": 1.7795093264724802e-05,
      "loss": 0.1141,
      "step": 6000
    },
    {
      "epoch": 0.5513185702471745,
      "eval_accuracy": 0.9580804777765948,
      "eval_f1": 0.7681685393564336,
      "eval_loss": 0.11200755834579468,
      "eval_precision": 0.7751412801894542,
      "eval_recall": 0.7613201259575663,
      "eval_runtime": 463.6002,
      "eval_samples_per_second": 61.775,
      "eval_steps_per_second": 1.236,
      "step": 6000
    },
    {
      "epoch": 0.597261784434439,
      "grad_norm": 0.44782352447509766,
      "learning_rate": 1.7611320407975743e-05,
      "loss": 0.1135,
      "step": 6500
    },
    {
      "epoch": 0.6432049986217035,
      "grad_norm": 0.5544679760932922,
      "learning_rate": 1.7427547551226684e-05,
      "loss": 0.1154,
      "step": 7000
    },
    {
      "epoch": 0.6432049986217035,
      "eval_accuracy": 0.9576146912247081,
      "eval_f1": 0.7686338805157804,
      "eval_loss": 0.11075501888990402,
      "eval_precision": 0.7656366294168357,
      "eval_recall": 0.7716546906241484,
      "eval_runtime": 472.6577,
      "eval_samples_per_second": 60.591,
      "eval_steps_per_second": 1.212,
      "step": 7000
    },
    {
      "epoch": 0.6891482128089681,
      "grad_norm": 0.4101652503013611,
      "learning_rate": 1.7243774694477626e-05,
      "loss": 0.1123,
      "step": 7500
    },
    {
      "epoch": 0.7350914269962326,
      "grad_norm": 0.5032928586006165,
      "learning_rate": 1.7060001837728567e-05,
      "loss": 0.1128,
      "step": 8000
    },
    {
      "epoch": 0.7350914269962326,
      "eval_accuracy": 0.9583707230021242,
      "eval_f1": 0.7702898098589065,
      "eval_loss": 0.10922538489103317,
      "eval_precision": 0.7744496660220761,
      "eval_recall": 0.7661744031991279,
      "eval_runtime": 465.4833,
      "eval_samples_per_second": 61.525,
      "eval_steps_per_second": 1.231,
      "step": 8000
    },
    {
      "epoch": 0.7810346411834972,
      "grad_norm": 0.5559138059616089,
      "learning_rate": 1.687622898097951e-05,
      "loss": 0.112,
      "step": 8500
    },
    {
      "epoch": 0.8269778553707617,
      "grad_norm": 0.7113578915596008,
      "learning_rate": 1.6692456124230453e-05,
      "loss": 0.1082,
      "step": 9000
    },
    {
      "epoch": 0.8269778553707617,
      "eval_accuracy": 0.9588998818739956,
      "eval_f1": 0.7713698035177239,
      "eval_loss": 0.10946337133646011,
      "eval_precision": 0.78099547261026,
      "eval_recall": 0.7619785159781004,
      "eval_runtime": 477.2564,
      "eval_samples_per_second": 60.008,
      "eval_steps_per_second": 1.201,
      "step": 9000
    },
    {
      "epoch": 0.8729210695580263,
      "grad_norm": 0.4333978593349457,
      "learning_rate": 1.6508683267481394e-05,
      "loss": 0.1116,
      "step": 9500
    },
    {
      "epoch": 0.9188642837452908,
      "grad_norm": 0.3579271137714386,
      "learning_rate": 1.6324910410732335e-05,
      "loss": 0.1101,
      "step": 10000
    },
    {
      "epoch": 0.9188642837452908,
      "eval_accuracy": 0.9590893651107495,
      "eval_f1": 0.7721416651417115,
      "eval_loss": 0.10736154764890671,
      "eval_precision": 0.7816850586599052,
      "eval_recall": 0.7628284867013311,
      "eval_runtime": 478.6898,
      "eval_samples_per_second": 59.828,
      "eval_steps_per_second": 1.197,
      "step": 10000
    },
    {
      "epoch": 0.9648074979325554,
      "grad_norm": 0.4602518677711487,
      "learning_rate": 1.6141137553983276e-05,
      "loss": 0.1101,
      "step": 10500
    },
    {
      "epoch": 1.0107507121198198,
      "grad_norm": 0.5322664976119995,
      "learning_rate": 1.5957364697234217e-05,
      "loss": 0.1101,
      "step": 11000
    },
    {
      "epoch": 1.0107507121198198,
      "eval_accuracy": 0.9591349931811384,
      "eval_f1": 0.7744744313952638,
      "eval_loss": 0.10792569071054459,
      "eval_precision": 0.7768799906601179,
      "eval_recall": 0.7720837234653981,
      "eval_runtime": 477.7344,
      "eval_samples_per_second": 59.948,
      "eval_steps_per_second": 1.199,
      "step": 11000
    },
    {
      "epoch": 1.0566939263070845,
      "grad_norm": 0.5673955678939819,
      "learning_rate": 1.5773591840485162e-05,
      "loss": 0.1071,
      "step": 11500
    },
    {
      "epoch": 1.102637140494349,
      "grad_norm": 0.42657607793807983,
      "learning_rate": 1.5589818983736103e-05,
      "loss": 0.1058,
      "step": 12000
    },
    {
      "epoch": 1.102637140494349,
      "eval_accuracy": 0.9584847931780965,
      "eval_f1": 0.7735292150015388,
      "eval_loss": 0.10933595150709152,
      "eval_precision": 0.7671773857517072,
      "eval_recall": 0.779987102031565,
      "eval_runtime": 475.9965,
      "eval_samples_per_second": 60.166,
      "eval_steps_per_second": 1.204,
      "step": 12000
    },
    {
      "epoch": 1.1485803546816136,
      "grad_norm": 0.4599572718143463,
      "learning_rate": 1.5406046126987044e-05,
      "loss": 0.107,
      "step": 12500
    },
    {
      "epoch": 1.194523568868878,
      "grad_norm": 0.4379059970378876,
      "learning_rate": 1.5222273270237986e-05,
      "loss": 0.107,
      "step": 13000
    },
    {
      "epoch": 1.194523568868878,
      "eval_accuracy": 0.959601413456225,
      "eval_f1": 0.7737560775183182,
      "eval_loss": 0.1075456365942955,
      "eval_precision": 0.7856481314090709,
      "eval_recall": 0.76221866643641,
      "eval_runtime": 478.1835,
      "eval_samples_per_second": 59.891,
      "eval_steps_per_second": 1.198,
      "step": 13000
    },
    {
      "epoch": 1.2404667830561427,
      "grad_norm": 0.43774718046188354,
      "learning_rate": 1.5038500413488928e-05,
      "loss": 0.1055,
      "step": 13500
    },
    {
      "epoch": 1.286409997243407,
      "grad_norm": 0.4196317195892334,
      "learning_rate": 1.485472755673987e-05,
      "loss": 0.1055,
      "step": 14000
    },
    {
      "epoch": 1.286409997243407,
      "eval_accuracy": 0.959860606244962,
      "eval_f1": 0.7768597287158228,
      "eval_loss": 0.10658087581396103,
      "eval_precision": 0.7824100936011823,
      "eval_recall": 0.7713875569682759,
      "eval_runtime": 464.4067,
      "eval_samples_per_second": 61.668,
      "eval_steps_per_second": 1.234,
      "step": 14000
    },
    {
      "epoch": 1.3323532114306718,
      "grad_norm": 0.36208856105804443,
      "learning_rate": 1.467095469999081e-05,
      "loss": 0.1035,
      "step": 14500
    },
    {
      "epoch": 1.3782964256179362,
      "grad_norm": 0.5692697763442993,
      "learning_rate": 1.4487181843241755e-05,
      "loss": 0.1054,
      "step": 15000
    },
    {
      "epoch": 1.3782964256179362,
      "eval_accuracy": 0.9593396857746885,
      "eval_f1": 0.7750922296908528,
      "eval_loss": 0.10651089251041412,
      "eval_precision": 0.7777847213920075,
      "eval_recall": 0.7724183151151778,
      "eval_runtime": 484.7629,
      "eval_samples_per_second": 59.078,
      "eval_steps_per_second": 1.182,
      "step": 15000
    },
    {
      "epoch": 1.4242396398052009,
      "grad_norm": 0.4805659055709839,
      "learning_rate": 1.4303408986492697e-05,
      "loss": 0.1043,
      "step": 15500
    },
    {
      "epoch": 1.4701828539924653,
      "grad_norm": 0.5373257994651794,
      "learning_rate": 1.411963612974364e-05,
      "loss": 0.1038,
      "step": 16000
    },
    {
      "epoch": 1.4701828539924653,
      "eval_accuracy": 0.9594822734946539,
      "eval_f1": 0.77620383584155,
      "eval_loss": 0.10620830953121185,
      "eval_precision": 0.7775004264051742,
      "eval_recall": 0.7749115625699877,
      "eval_runtime": 484.6548,
      "eval_samples_per_second": 59.092,
      "eval_steps_per_second": 1.182,
      "step": 16000
    },
    {
      "epoch": 1.5161260681797297,
      "grad_norm": 0.5227932929992676,
      "learning_rate": 1.393586327299458e-05,
      "loss": 0.1057,
      "step": 16500
    },
    {
      "epoch": 1.5620692823669944,
      "grad_norm": 0.2855619192123413,
      "learning_rate": 1.3752090416245523e-05,
      "loss": 0.1052,
      "step": 17000
    },
    {
      "epoch": 1.5620692823669944,
      "eval_accuracy": 0.9600310777857204,
      "eval_f1": 0.7757748564065305,
      "eval_loss": 0.10526877641677856,
      "eval_precision": 0.7884331924355441,
      "eval_recall": 0.7635165582391844,
      "eval_runtime": 483.926,
      "eval_samples_per_second": 59.181,
      "eval_steps_per_second": 1.184,
      "step": 17000
    },
    {
      "epoch": 1.608012496554259,
      "grad_norm": 0.36337897181510925,
      "learning_rate": 1.3568317559496465e-05,
      "loss": 0.1058,
      "step": 17500
    },
    {
      "epoch": 1.6539557107415235,
      "grad_norm": 0.7537007927894592,
      "learning_rate": 1.3384544702747406e-05,
      "loss": 0.1061,
      "step": 18000
    },
    {
      "epoch": 1.6539557107415235,
      "eval_accuracy": 0.9598663097537605,
      "eval_f1": 0.7781213928303998,
      "eval_loss": 0.1061909943819046,
      "eval_precision": 0.7795501631707945,
      "eval_recall": 0.7766978502486501,
      "eval_runtime": 483.1917,
      "eval_samples_per_second": 59.27,
      "eval_steps_per_second": 1.186,
      "step": 18000
    },
    {
      "epoch": 1.699898924928788,
      "grad_norm": 0.7511648535728455,
      "learning_rate": 1.3200771845998349e-05,
      "loss": 0.1035,
      "step": 18500
    },
    {
      "epoch": 1.7458421391160526,
      "grad_norm": 0.5765513181686401,
      "learning_rate": 1.301699898924929e-05,
      "loss": 0.1038,
      "step": 19000
    },
    {
      "epoch": 1.7458421391160526,
      "eval_accuracy": 0.9596578148210112,
      "eval_f1": 0.7782083489589002,
      "eval_loss": 0.10631386935710907,
      "eval_precision": 0.7753205917958497,
      "eval_recall": 0.7811176980094495,
      "eval_runtime": 481.5937,
      "eval_samples_per_second": 59.467,
      "eval_steps_per_second": 1.19,
      "step": 19000
    },
    {
      "epoch": 1.7917853533033172,
      "grad_norm": 1.3671506643295288,
      "learning_rate": 1.2833226132500231e-05,
      "loss": 0.1037,
      "step": 19500
    },
    {
      "epoch": 1.8377285674905817,
      "grad_norm": 0.6475801467895508,
      "learning_rate": 1.2649453275751174e-05,
      "loss": 0.1017,
      "step": 20000
    },
    {
      "epoch": 1.8377285674905817,
      "eval_accuracy": 0.9603023113152545,
      "eval_f1": 0.7790918283030356,
      "eval_loss": 0.10478387773036957,
      "eval_precision": 0.784845306908137,
      "eval_recall": 0.7734220900645168,
      "eval_runtime": 483.6282,
      "eval_samples_per_second": 59.217,
      "eval_steps_per_second": 1.185,
      "step": 20000
    },
    {
      "epoch": 1.883671781677846,
      "grad_norm": 0.898071825504303,
      "learning_rate": 1.2465680419002115e-05,
      "loss": 0.1027,
      "step": 20500
    },
    {
      "epoch": 1.9296149958651108,
      "grad_norm": 0.6122450232505798,
      "learning_rate": 1.2281907562253056e-05,
      "loss": 0.1029,
      "step": 21000
    },
    {
      "epoch": 1.9296149958651108,
      "eval_accuracy": 0.96032068928805,
      "eval_f1": 0.7800494094554808,
      "eval_loss": 0.10455518215894699,
      "eval_precision": 0.7826737729678044,
      "eval_recall": 0.7774425865013855,
      "eval_runtime": 483.8492,
      "eval_samples_per_second": 59.19,
      "eval_steps_per_second": 1.184,
      "step": 21000
    },
    {
      "epoch": 1.9755582100523754,
      "grad_norm": 0.6479480266571045,
      "learning_rate": 1.2098134705504e-05,
      "loss": 0.1032,
      "step": 21500
    },
    {
      "epoch": 2.0215014242396396,
      "grad_norm": 0.9193481206893921,
      "learning_rate": 1.191436184875494e-05,
      "loss": 0.1018,
      "step": 22000
    },
    {
      "epoch": 2.0215014242396396,
      "eval_accuracy": 0.9604075093664289,
      "eval_f1": 0.7798349386244161,
      "eval_loss": 0.10434387624263763,
      "eval_precision": 0.7843734048157843,
      "eval_recall": 0.7753486903705062,
      "eval_runtime": 482.4895,
      "eval_samples_per_second": 59.357,
      "eval_steps_per_second": 1.188,
      "step": 22000
    },
    {
      "epoch": 2.0674446384269043,
      "grad_norm": 0.41459354758262634,
      "learning_rate": 1.1730588992005882e-05,
      "loss": 0.1015,
      "step": 22500
    },
    {
      "epoch": 2.113387852614169,
      "grad_norm": 0.5848496556282043,
      "learning_rate": 1.1546816135256825e-05,
      "loss": 0.1007,
      "step": 23000
    },
    {
      "epoch": 2.113387852614169,
      "eval_accuracy": 0.960530451667199,
      "eval_f1": 0.7787409881226326,
      "eval_loss": 0.10433882474899292,
      "eval_precision": 0.7888630191217876,
      "eval_recall": 0.7688754212751719,
      "eval_runtime": 481.4982,
      "eval_samples_per_second": 59.479,
      "eval_steps_per_second": 1.19,
      "step": 23000
    },
    {
      "epoch": 2.1593310668014336,
      "grad_norm": 0.5300314426422119,
      "learning_rate": 1.1363043278507766e-05,
      "loss": 0.1008,
      "step": 23500
    },
    {
      "epoch": 2.205274280988698,
      "grad_norm": 0.34308552742004395,
      "learning_rate": 1.1179270421758709e-05,
      "loss": 0.1008,
      "step": 24000
    },
    {
      "epoch": 2.205274280988698,
      "eval_accuracy": 0.9605203120960014,
      "eval_f1": 0.7802636835797044,
      "eval_loss": 0.10480789840221405,
      "eval_precision": 0.7859456144721461,
      "eval_recall": 0.7746633171524092,
      "eval_runtime": 482.2132,
      "eval_samples_per_second": 59.391,
      "eval_steps_per_second": 1.188,
      "step": 24000
    },
    {
      "epoch": 2.2512174951759625,
      "grad_norm": 0.668408215045929,
      "learning_rate": 1.099549756500965e-05,
      "loss": 0.101,
      "step": 24500
    },
    {
      "epoch": 2.297160709363227,
      "grad_norm": 0.6432192921638489,
      "learning_rate": 1.0811724708260591e-05,
      "loss": 0.1015,
      "step": 25000
    },
    {
      "epoch": 2.297160709363227,
      "eval_accuracy": 0.9606768417263634,
      "eval_f1": 0.7810441317665336,
      "eval_loss": 0.10385506600141525,
      "eval_precision": 0.78621070532204,
      "eval_recall": 0.7759450190366459,
      "eval_runtime": 480.9669,
      "eval_samples_per_second": 59.545,
      "eval_steps_per_second": 1.191,
      "step": 25000
    }
  ],
  "logging_steps": 500,
  "max_steps": 54415,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.905786123724367e+17,
  "train_batch_size": 50,
  "trial_name": null,
  "trial_params": null
}