{
  "best_global_step": 4152,
  "best_metric": 0.8636363636363636,
  "best_model_checkpoint": "./bert-tiny-heading-classifier\\checkpoint-4152",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 4152,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.036127167630057806,
      "grad_norm": 3.7394230365753174,
      "learning_rate": 2.45e-05,
      "loss": 0.484,
      "step": 50
    },
    {
      "epoch": 0.07225433526011561,
      "grad_norm": 2.0182766914367676,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 0.3104,
      "step": 100
    },
    {
      "epoch": 0.10838150289017341,
      "grad_norm": 1.2229622602462769,
      "learning_rate": 4.9549300956585726e-05,
      "loss": 0.2133,
      "step": 150
    },
    {
      "epoch": 0.14450867052023122,
      "grad_norm": 0.9460026025772095,
      "learning_rate": 4.9089403973509935e-05,
      "loss": 0.2163,
      "step": 200
    },
    {
      "epoch": 0.18063583815028902,
      "grad_norm": 0.7926827073097229,
      "learning_rate": 4.8629506990434144e-05,
      "loss": 0.182,
      "step": 250
    },
    {
      "epoch": 0.21676300578034682,
      "grad_norm": 0.5686139464378357,
      "learning_rate": 4.816961000735836e-05,
      "loss": 0.1811,
      "step": 300
    },
    {
      "epoch": 0.25289017341040465,
      "grad_norm": 1.6547473669052124,
      "learning_rate": 4.770971302428257e-05,
      "loss": 0.121,
      "step": 350
    },
    {
      "epoch": 0.28901734104046245,
      "grad_norm": 0.47385355830192566,
      "learning_rate": 4.724981604120677e-05,
      "loss": 0.1834,
      "step": 400
    },
    {
      "epoch": 0.32514450867052025,
      "grad_norm": 0.9103878140449524,
      "learning_rate": 4.678991905813098e-05,
      "loss": 0.1099,
      "step": 450
    },
    {
      "epoch": 0.36127167630057805,
      "grad_norm": 0.30629491806030273,
      "learning_rate": 4.633002207505519e-05,
      "loss": 0.1169,
      "step": 500
    },
    {
      "epoch": 0.39739884393063585,
      "grad_norm": 3.804182291030884,
      "learning_rate": 4.58701250919794e-05,
      "loss": 0.151,
      "step": 550
    },
    {
      "epoch": 0.43352601156069365,
      "grad_norm": 0.23083843290805817,
      "learning_rate": 4.5410228108903606e-05,
      "loss": 0.0816,
      "step": 600
    },
    {
      "epoch": 0.46965317919075145,
      "grad_norm": 0.31194067001342773,
      "learning_rate": 4.4950331125827815e-05,
      "loss": 0.1238,
      "step": 650
    },
    {
      "epoch": 0.5057803468208093,
      "grad_norm": 0.18349987268447876,
      "learning_rate": 4.4490434142752024e-05,
      "loss": 0.0834,
      "step": 700
    },
    {
      "epoch": 0.541907514450867,
      "grad_norm": 2.0543556213378906,
      "learning_rate": 4.403053715967623e-05,
      "loss": 0.1476,
      "step": 750
    },
    {
      "epoch": 0.5780346820809249,
      "grad_norm": 3.1080751419067383,
      "learning_rate": 4.357064017660044e-05,
      "loss": 0.1217,
      "step": 800
    },
    {
      "epoch": 0.6141618497109826,
      "grad_norm": 13.235072135925293,
      "learning_rate": 4.311074319352466e-05,
      "loss": 0.1886,
      "step": 850
    },
    {
      "epoch": 0.6502890173410405,
      "grad_norm": 0.290387362241745,
      "learning_rate": 4.265084621044887e-05,
      "loss": 0.1246,
      "step": 900
    },
    {
      "epoch": 0.6864161849710982,
      "grad_norm": 3.8590610027313232,
      "learning_rate": 4.219094922737307e-05,
      "loss": 0.079,
      "step": 950
    },
    {
      "epoch": 0.7225433526011561,
      "grad_norm": 0.1421326994895935,
      "learning_rate": 4.173105224429728e-05,
      "loss": 0.0346,
      "step": 1000
    },
    {
      "epoch": 0.7586705202312138,
      "grad_norm": 0.1293569654226303,
      "learning_rate": 4.127115526122149e-05,
      "loss": 0.0552,
      "step": 1050
    },
    {
      "epoch": 0.7947976878612717,
      "grad_norm": 0.11882981657981873,
      "learning_rate": 4.0811258278145696e-05,
      "loss": 0.1326,
      "step": 1100
    },
    {
      "epoch": 0.8309248554913294,
      "grad_norm": 0.12036094069480896,
      "learning_rate": 4.0351361295069905e-05,
      "loss": 0.0938,
      "step": 1150
    },
    {
      "epoch": 0.8670520231213873,
      "grad_norm": 3.4779465198516846,
      "learning_rate": 3.9891464311994114e-05,
      "loss": 0.088,
      "step": 1200
    },
    {
      "epoch": 0.903179190751445,
      "grad_norm": 0.17521940171718597,
      "learning_rate": 3.943156732891832e-05,
      "loss": 0.0692,
      "step": 1250
    },
    {
      "epoch": 0.9393063583815029,
      "grad_norm": 4.401475429534912,
      "learning_rate": 3.897167034584253e-05,
      "loss": 0.067,
      "step": 1300
    },
    {
      "epoch": 0.9754335260115607,
      "grad_norm": 18.098052978515625,
      "learning_rate": 3.851177336276674e-05,
      "loss": 0.0704,
      "step": 1350
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.981064483111566,
      "eval_f1": 0.7932960893854749,
      "eval_loss": 0.08416279405355453,
      "eval_runtime": 1.6656,
      "eval_samples_per_second": 1173.183,
      "eval_steps_per_second": 147.098,
      "step": 1384
    },
    {
      "epoch": 1.0115606936416186,
      "grad_norm": 0.13371741771697998,
      "learning_rate": 3.805187637969095e-05,
      "loss": 0.0487,
      "step": 1400
    },
    {
      "epoch": 1.0476878612716762,
      "grad_norm": 0.1262962520122528,
      "learning_rate": 3.759197939661516e-05,
      "loss": 0.1117,
      "step": 1450
    },
    {
      "epoch": 1.083815028901734,
      "grad_norm": 5.906493663787842,
      "learning_rate": 3.713208241353937e-05,
      "loss": 0.0851,
      "step": 1500
    },
    {
      "epoch": 1.119942196531792,
      "grad_norm": 0.12109671533107758,
      "learning_rate": 3.667218543046358e-05,
      "loss": 0.0705,
      "step": 1550
    },
    {
      "epoch": 1.1560693641618498,
      "grad_norm": 0.08801592141389847,
      "learning_rate": 3.6212288447387786e-05,
      "loss": 0.1045,
      "step": 1600
    },
    {
      "epoch": 1.1921965317919074,
      "grad_norm": 0.28873875737190247,
      "learning_rate": 3.5752391464311995e-05,
      "loss": 0.0827,
      "step": 1650
    },
    {
      "epoch": 1.2283236994219653,
      "grad_norm": 0.5105486512184143,
      "learning_rate": 3.5292494481236204e-05,
      "loss": 0.0836,
      "step": 1700
    },
    {
      "epoch": 1.2644508670520231,
      "grad_norm": 3.9093217849731445,
      "learning_rate": 3.483259749816041e-05,
      "loss": 0.1053,
      "step": 1750
    },
    {
      "epoch": 1.300578034682081,
      "grad_norm": 0.19484883546829224,
      "learning_rate": 3.437270051508462e-05,
      "loss": 0.0774,
      "step": 1800
    },
    {
      "epoch": 1.3367052023121389,
      "grad_norm": 0.12063230574131012,
      "learning_rate": 3.391280353200883e-05,
      "loss": 0.1184,
      "step": 1850
    },
    {
      "epoch": 1.3728323699421965,
      "grad_norm": 0.0778161883354187,
      "learning_rate": 3.345290654893304e-05,
      "loss": 0.084,
      "step": 1900
    },
    {
      "epoch": 1.4089595375722543,
      "grad_norm": 0.15158304572105408,
      "learning_rate": 3.299300956585725e-05,
      "loss": 0.0636,
      "step": 1950
    },
    {
      "epoch": 1.4450867052023122,
      "grad_norm": 0.09447193890810013,
      "learning_rate": 3.253311258278146e-05,
      "loss": 0.0957,
      "step": 2000
    },
    {
      "epoch": 1.4812138728323698,
      "grad_norm": 0.06642930209636688,
      "learning_rate": 3.207321559970567e-05,
      "loss": 0.078,
      "step": 2050
    },
    {
      "epoch": 1.5173410404624277,
      "grad_norm": 1.1793818473815918,
      "learning_rate": 3.1613318616629876e-05,
      "loss": 0.0632,
      "step": 2100
    },
    {
      "epoch": 1.5534682080924855,
      "grad_norm": 1.3295987844467163,
      "learning_rate": 3.1153421633554085e-05,
      "loss": 0.0604,
      "step": 2150
    },
    {
      "epoch": 1.5895953757225434,
      "grad_norm": 2.1613638401031494,
      "learning_rate": 3.0693524650478294e-05,
      "loss": 0.0941,
      "step": 2200
    },
    {
      "epoch": 1.6257225433526012,
      "grad_norm": 7.148230075836182,
      "learning_rate": 3.0233627667402503e-05,
      "loss": 0.1062,
      "step": 2250
    },
    {
      "epoch": 1.661849710982659,
      "grad_norm": 27.96155548095703,
      "learning_rate": 2.977373068432671e-05,
      "loss": 0.1002,
      "step": 2300
    },
    {
      "epoch": 1.6979768786127167,
      "grad_norm": 2.712245225906372,
      "learning_rate": 2.931383370125092e-05,
      "loss": 0.0833,
      "step": 2350
    },
    {
      "epoch": 1.7341040462427746,
      "grad_norm": 1.0877773761749268,
      "learning_rate": 2.885393671817513e-05,
      "loss": 0.0918,
      "step": 2400
    },
    {
      "epoch": 1.7702312138728322,
      "grad_norm": 2.0854358673095703,
      "learning_rate": 2.839403973509934e-05,
      "loss": 0.1156,
      "step": 2450
    },
    {
      "epoch": 1.80635838150289,
      "grad_norm": 0.0537271611392498,
      "learning_rate": 2.7934142752023544e-05,
      "loss": 0.0496,
      "step": 2500
    },
    {
      "epoch": 1.842485549132948,
      "grad_norm": 0.10613041371107101,
      "learning_rate": 2.747424576894776e-05,
      "loss": 0.0605,
      "step": 2550
    },
    {
      "epoch": 1.8786127167630058,
      "grad_norm": 0.047803498804569244,
      "learning_rate": 2.701434878587197e-05,
      "loss": 0.0441,
      "step": 2600
    },
    {
      "epoch": 1.9147398843930636,
      "grad_norm": 0.0479620136320591,
      "learning_rate": 2.6554451802796178e-05,
      "loss": 0.0476,
      "step": 2650
    },
    {
      "epoch": 1.9508670520231215,
      "grad_norm": 0.10258156806230545,
      "learning_rate": 2.6094554819720383e-05,
      "loss": 0.0364,
      "step": 2700
    },
    {
      "epoch": 1.9869942196531793,
      "grad_norm": 0.07609525322914124,
      "learning_rate": 2.5634657836644592e-05,
      "loss": 0.0405,
      "step": 2750
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9851586489252815,
      "eval_f1": 0.8397790055248618,
      "eval_loss": 0.06949524581432343,
      "eval_runtime": 1.6296,
      "eval_samples_per_second": 1199.037,
      "eval_steps_per_second": 150.34,
      "step": 2768
    },
    {
      "epoch": 2.023121387283237,
      "grad_norm": 0.03800142556428909,
      "learning_rate": 2.51747608535688e-05,
      "loss": 0.0533,
      "step": 2800
    },
    {
      "epoch": 2.0592485549132946,
      "grad_norm": 0.10002855956554413,
      "learning_rate": 2.471486387049301e-05,
      "loss": 0.1105,
      "step": 2850
    },
    {
      "epoch": 2.0953757225433525,
      "grad_norm": 0.16160914301872253,
      "learning_rate": 2.425496688741722e-05,
      "loss": 0.0651,
      "step": 2900
    },
    {
      "epoch": 2.1315028901734103,
      "grad_norm": 0.08646886795759201,
      "learning_rate": 2.379506990434143e-05,
      "loss": 0.0538,
      "step": 2950
    },
    {
      "epoch": 2.167630057803468,
      "grad_norm": 62.76107406616211,
      "learning_rate": 2.3335172921265637e-05,
      "loss": 0.0594,
      "step": 3000
    },
    {
      "epoch": 2.203757225433526,
      "grad_norm": 0.06207644194364548,
      "learning_rate": 2.2875275938189846e-05,
      "loss": 0.0624,
      "step": 3050
    },
    {
      "epoch": 2.239884393063584,
      "grad_norm": 0.038836702704429626,
      "learning_rate": 2.2415378955114055e-05,
      "loss": 0.072,
      "step": 3100
    },
    {
      "epoch": 2.2760115606936417,
      "grad_norm": 0.2888094484806061,
      "learning_rate": 2.1955481972038264e-05,
      "loss": 0.09,
      "step": 3150
    },
    {
      "epoch": 2.3121387283236996,
      "grad_norm": 0.043586716055870056,
      "learning_rate": 2.1495584988962473e-05,
      "loss": 0.0575,
      "step": 3200
    },
    {
      "epoch": 2.348265895953757,
      "grad_norm": 0.03632841259241104,
      "learning_rate": 2.1035688005886682e-05,
      "loss": 0.0537,
      "step": 3250
    },
    {
      "epoch": 2.384393063583815,
      "grad_norm": 0.05004638433456421,
      "learning_rate": 2.057579102281089e-05,
      "loss": 0.065,
      "step": 3300
    },
    {
      "epoch": 2.4205202312138727,
      "grad_norm": 0.04363992065191269,
      "learning_rate": 2.01158940397351e-05,
      "loss": 0.0448,
      "step": 3350
    },
    {
      "epoch": 2.4566473988439306,
      "grad_norm": 1.8408621549606323,
      "learning_rate": 1.965599705665931e-05,
      "loss": 0.0666,
      "step": 3400
    },
    {
      "epoch": 2.4927745664739884,
      "grad_norm": 0.037667375057935715,
      "learning_rate": 1.9196100073583518e-05,
      "loss": 0.0744,
      "step": 3450
    },
    {
      "epoch": 2.5289017341040463,
      "grad_norm": 0.0906088650226593,
      "learning_rate": 1.8736203090507727e-05,
      "loss": 0.066,
      "step": 3500
    },
    {
      "epoch": 2.565028901734104,
      "grad_norm": 0.0400952510535717,
      "learning_rate": 1.8276306107431936e-05,
      "loss": 0.0376,
      "step": 3550
    },
    {
      "epoch": 2.601156069364162,
      "grad_norm": 33.840248107910156,
      "learning_rate": 1.7816409124356145e-05,
      "loss": 0.0512,
      "step": 3600
    },
    {
      "epoch": 2.63728323699422,
      "grad_norm": 0.17998424172401428,
      "learning_rate": 1.7356512141280354e-05,
      "loss": 0.0655,
      "step": 3650
    },
    {
      "epoch": 2.6734104046242777,
      "grad_norm": 0.04229271411895752,
      "learning_rate": 1.6896615158204563e-05,
      "loss": 0.0517,
      "step": 3700
    },
    {
      "epoch": 2.709537572254335,
      "grad_norm": 0.03790134936571121,
      "learning_rate": 1.6436718175128772e-05,
      "loss": 0.068,
      "step": 3750
    },
    {
      "epoch": 2.745664739884393,
      "grad_norm": 31.895214080810547,
      "learning_rate": 1.597682119205298e-05,
      "loss": 0.0729,
      "step": 3800
    },
    {
      "epoch": 2.781791907514451,
      "grad_norm": 2.5294675827026367,
      "learning_rate": 1.551692420897719e-05,
      "loss": 0.0534,
      "step": 3850
    },
    {
      "epoch": 2.8179190751445087,
      "grad_norm": 0.03958361968398094,
      "learning_rate": 1.5057027225901399e-05,
      "loss": 0.0396,
      "step": 3900
    },
    {
      "epoch": 2.8540462427745665,
      "grad_norm": 0.03872457146644592,
      "learning_rate": 1.4597130242825608e-05,
      "loss": 0.0874,
      "step": 3950
    },
    {
      "epoch": 2.8901734104046244,
      "grad_norm": 3.4865188598632812,
      "learning_rate": 1.4137233259749815e-05,
      "loss": 0.0831,
      "step": 4000
    },
    {
      "epoch": 2.9263005780346822,
      "grad_norm": 0.03453204780817032,
      "learning_rate": 1.3677336276674028e-05,
      "loss": 0.0527,
      "step": 4050
    },
    {
      "epoch": 2.9624277456647397,
      "grad_norm": 1.6099286079406738,
      "learning_rate": 1.3217439293598235e-05,
      "loss": 0.0305,
      "step": 4100
    },
    {
      "epoch": 2.9985549132947975,
      "grad_norm": 0.035434894263744354,
      "learning_rate": 1.2757542310522444e-05,
      "loss": 0.1007,
      "step": 4150
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9877175025588536,
      "eval_f1": 0.8636363636363636,
      "eval_loss": 0.06188501790165901,
      "eval_runtime": 1.631,
      "eval_samples_per_second": 1198.045,
      "eval_steps_per_second": 150.216,
      "step": 4152
    }
  ],
  "logging_steps": 50,
  "max_steps": 5536,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 21094531568640.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}