{
  "best_global_step": 900,
  "best_metric": 0.8504398826979472,
  "best_model_checkpoint": "/content/drive/MyDrive/expC_large_realonly/checkpoint-900",
  "epoch": 13.373177842565598,
  "eval_steps": 50,
  "global_step": 1150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011661807580174927,
      "grad_norm": 53.1702880859375,
      "learning_rate": 0.0,
      "loss": 0.6632702350616455,
      "step": 1
    },
    {
      "epoch": 0.2915451895043732,
      "grad_norm": 3.24454927444458,
      "learning_rate": 6.282722513089005e-06,
      "loss": 0.6155035893122355,
      "step": 25
    },
    {
      "epoch": 0.5830903790087464,
      "grad_norm": 1.3036233186721802,
      "learning_rate": 1.2827225130890053e-05,
      "loss": 0.4128861999511719,
      "step": 50
    },
    {
      "epoch": 0.5830903790087464,
      "eval_accuracy": 0.8467655276165914,
      "eval_f1": 0.48104575163398694,
      "eval_loss": 0.08503369987010956,
      "eval_precision": 0.4309133489461358,
      "eval_recall": 0.5443786982248521,
      "eval_runtime": 5.5979,
      "eval_samples_per_second": 60.915,
      "eval_steps_per_second": 1.965,
      "step": 50
    },
    {
      "epoch": 0.8746355685131195,
      "grad_norm": 1.6594698429107666,
      "learning_rate": 1.93717277486911e-05,
      "loss": 0.2590180397033691,
      "step": 75
    },
    {
      "epoch": 1.163265306122449,
      "grad_norm": 1.5420184135437012,
      "learning_rate": 2.591623036649215e-05,
      "loss": 0.217969970703125,
      "step": 100
    },
    {
      "epoch": 1.163265306122449,
      "eval_accuracy": 0.8611648398882441,
      "eval_f1": 0.5584725536992841,
      "eval_loss": 0.059965744614601135,
      "eval_precision": 0.468,
      "eval_recall": 0.6923076923076923,
      "eval_runtime": 2.9899,
      "eval_samples_per_second": 114.05,
      "eval_steps_per_second": 3.679,
      "step": 100
    },
    {
      "epoch": 1.4548104956268222,
      "grad_norm": 3.1667258739471436,
      "learning_rate": 3.246073298429319e-05,
      "loss": 0.18862422943115234,
      "step": 125
    },
    {
      "epoch": 1.7463556851311952,
      "grad_norm": 7.493422985076904,
      "learning_rate": 3.900523560209424e-05,
      "loss": 0.1803609848022461,
      "step": 150
    },
    {
      "epoch": 1.7463556851311952,
      "eval_accuracy": 0.864173651407694,
      "eval_f1": 0.5561372891215822,
      "eval_loss": 0.04819481074810028,
      "eval_precision": 0.4582933844678811,
      "eval_recall": 0.7071005917159763,
      "eval_runtime": 2.9913,
      "eval_samples_per_second": 113.997,
      "eval_steps_per_second": 3.677,
      "step": 150
    },
    {
      "epoch": 2.0349854227405246,
      "grad_norm": 2.291372060775757,
      "learning_rate": 4.554973821989529e-05,
      "loss": 0.1149876594543457,
      "step": 175
    },
    {
      "epoch": 2.326530612244898,
      "grad_norm": 3.1508946418762207,
      "learning_rate": 4.999346304613061e-05,
      "loss": 0.0918524169921875,
      "step": 200
    },
    {
      "epoch": 2.326530612244898,
      "eval_accuracy": 0.9348807221147647,
      "eval_f1": 0.7352328005559416,
      "eval_loss": 0.04772070050239563,
      "eval_precision": 0.6933158584534731,
      "eval_recall": 0.7825443786982249,
      "eval_runtime": 2.979,
      "eval_samples_per_second": 114.468,
      "eval_steps_per_second": 3.693,
      "step": 200
    },
    {
      "epoch": 2.618075801749271,
      "grad_norm": 2.248906135559082,
      "learning_rate": 4.98888472605987e-05,
      "loss": 0.11147353172302246,
      "step": 225
    },
    {
      "epoch": 2.9096209912536444,
      "grad_norm": 1.1041216850280762,
      "learning_rate": 4.965717280180432e-05,
      "loss": 0.12180458068847656,
      "step": 250
    },
    {
      "epoch": 2.9096209912536444,
      "eval_accuracy": 0.8517085751128304,
      "eval_f1": 0.606029106029106,
      "eval_loss": 0.031385358422994614,
      "eval_precision": 0.4671474358974359,
      "eval_recall": 0.8624260355029586,
      "eval_runtime": 2.9853,
      "eval_samples_per_second": 114.227,
      "eval_steps_per_second": 3.685,
      "step": 250
    },
    {
      "epoch": 3.198250728862974,
      "grad_norm": 2.275531053543091,
      "learning_rate": 4.9299622378173246e-05,
      "loss": 0.0561268424987793,
      "step": 275
    },
    {
      "epoch": 3.489795918367347,
      "grad_norm": 15.159272193908691,
      "learning_rate": 4.881802130053548e-05,
      "loss": 0.05052190780639648,
      "step": 300
    },
    {
      "epoch": 3.489795918367347,
      "eval_accuracy": 0.9101654846335697,
      "eval_f1": 0.7108140947752126,
      "eval_loss": 0.03343665972352028,
      "eval_precision": 0.6030927835051546,
      "eval_recall": 0.8653846153846154,
      "eval_runtime": 2.9916,
      "eval_samples_per_second": 113.985,
      "eval_steps_per_second": 3.677,
      "step": 300
    },
    {
      "epoch": 3.78134110787172,
      "grad_norm": 2.0696234703063965,
      "learning_rate": 4.821482816383218e-05,
      "loss": 0.06179484367370605,
      "step": 325
    },
    {
      "epoch": 4.069970845481049,
      "grad_norm": 1.3048503398895264,
      "learning_rate": 4.749312229587908e-05,
      "loss": 0.04184870719909668,
      "step": 350
    },
    {
      "epoch": 4.069970845481049,
      "eval_accuracy": 0.9226305609284333,
      "eval_f1": 0.7249357326478149,
      "eval_loss": 0.03706977888941765,
      "eval_precision": 0.6409090909090909,
      "eval_recall": 0.834319526627219,
      "eval_runtime": 2.9936,
      "eval_samples_per_second": 113.91,
      "eval_steps_per_second": 3.675,
      "step": 350
    },
    {
      "epoch": 4.3615160349854225,
      "grad_norm": 1.1634117364883423,
      "learning_rate": 4.6656588037260834e-05,
      "loss": 0.02826032876968384,
      "step": 375
    },
    {
      "epoch": 4.653061224489796,
      "grad_norm": 0.3656497597694397,
      "learning_rate": 4.570949593260837e-05,
      "loss": 0.02992666721343994,
      "step": 400
    },
    {
      "epoch": 4.653061224489796,
      "eval_accuracy": 0.941972920696325,
      "eval_f1": 0.7889413351314902,
      "eval_loss": 0.03734752908349037,
      "eval_precision": 0.724907063197026,
      "eval_recall": 0.8653846153846154,
      "eval_runtime": 3.0067,
      "eval_samples_per_second": 113.414,
      "eval_steps_per_second": 3.659,
      "step": 400
    },
    {
      "epoch": 4.944606413994169,
      "grad_norm": 0.7156509757041931,
      "learning_rate": 4.465668092927841e-05,
      "loss": 0.02784186840057373,
      "step": 425
    },
    {
      "epoch": 5.233236151603498,
      "grad_norm": 0.2143929898738861,
      "learning_rate": 4.350351769473198e-05,
      "loss": 0.014064539670944214,
      "step": 450
    },
    {
      "epoch": 5.233236151603498,
      "eval_accuracy": 0.9396088545024716,
      "eval_f1": 0.7841823056300269,
      "eval_loss": 0.04331827163696289,
      "eval_precision": 0.7169117647058824,
      "eval_recall": 0.8653846153846154,
      "eval_runtime": 2.9881,
      "eval_samples_per_second": 114.12,
      "eval_steps_per_second": 3.681,
      "step": 450
    },
    {
      "epoch": 5.524781341107872,
      "grad_norm": 0.2881239950656891,
      "learning_rate": 4.2255893178617745e-05,
      "loss": 0.01728477954864502,
      "step": 475
    },
    {
      "epoch": 5.816326530612245,
      "grad_norm": 0.9025610089302063,
      "learning_rate": 4.092017655963198e-05,
      "loss": 0.013348644971847535,
      "step": 500
    },
    {
      "epoch": 5.816326530612245,
      "eval_accuracy": 0.9368149580915538,
      "eval_f1": 0.7799074686054197,
      "eval_loss": 0.04094831272959709,
      "eval_precision": 0.7048984468339307,
      "eval_recall": 0.8727810650887574,
      "eval_runtime": 2.9827,
      "eval_samples_per_second": 114.324,
      "eval_steps_per_second": 3.688,
      "step": 500
    },
    {
      "epoch": 6.104956268221574,
      "grad_norm": 0.5944705009460449,
      "learning_rate": 3.9503186730577816e-05,
      "loss": 0.017484543323516847,
      "step": 525
    },
    {
      "epoch": 6.396501457725948,
      "grad_norm": 0.20067809522151947,
      "learning_rate": 3.801215748761385e-05,
      "loss": 0.005213648080825806,
      "step": 550
    },
    {
      "epoch": 6.396501457725948,
      "eval_accuracy": 0.9608854502471523,
      "eval_f1": 0.8493932905067809,
      "eval_loss": 0.048467967659235,
      "eval_precision": 0.8206896551724138,
      "eval_recall": 0.8801775147928994,
      "eval_runtime": 2.9914,
      "eval_samples_per_second": 113.994,
      "eval_steps_per_second": 3.677,
      "step": 550
    },
    {
      "epoch": 6.688046647230321,
      "grad_norm": 0.25010251998901367,
      "learning_rate": 3.645470060140278e-05,
      "loss": 0.007735109925270081,
      "step": 575
    },
    {
      "epoch": 6.979591836734694,
      "grad_norm": 0.5211274027824402,
      "learning_rate": 3.4838766958683304e-05,
      "loss": 0.010291404724121093,
      "step": 600
    },
    {
      "epoch": 6.979591836734694,
      "eval_accuracy": 0.9432624113475178,
      "eval_f1": 0.8013377926421403,
      "eval_loss": 0.042320284992456436,
      "eval_precision": 0.7313797313797313,
      "eval_recall": 0.886094674556213,
      "eval_runtime": 2.992,
      "eval_samples_per_second": 113.971,
      "eval_steps_per_second": 3.676,
      "step": 600
    },
    {
      "epoch": 7.2682215743440235,
      "grad_norm": 1.0244024991989136,
      "learning_rate": 3.317260597263932e-05,
      "loss": 0.004896260499954223,
      "step": 625
    },
    {
      "epoch": 7.559766763848397,
      "grad_norm": 0.4166155755519867,
      "learning_rate": 3.146472346927845e-05,
      "loss": 0.0053441798686981205,
      "step": 650
    },
    {
      "epoch": 7.559766763848397,
      "eval_accuracy": 0.9537932516655921,
      "eval_f1": 0.8184397163120567,
      "eval_loss": 0.06086782366037369,
      "eval_precision": 0.7861035422343324,
      "eval_recall": 0.8535502958579881,
      "eval_runtime": 3.0122,
      "eval_samples_per_second": 113.206,
      "eval_steps_per_second": 3.652,
      "step": 650
    },
    {
      "epoch": 7.85131195335277,
      "grad_norm": 0.010130354203283787,
      "learning_rate": 2.9723838264811545e-05,
      "loss": 0.004410363137722016,
      "step": 675
    },
    {
      "epoch": 8.139941690962099,
      "grad_norm": 0.6293845176696777,
      "learning_rate": 2.7958837655707814e-05,
      "loss": 0.004301608800888062,
      "step": 700
    },
    {
      "epoch": 8.139941690962099,
      "eval_accuracy": 0.9546529120997206,
      "eval_f1": 0.8310055865921788,
      "eval_loss": 0.05279209837317467,
      "eval_precision": 0.7870370370370371,
      "eval_recall": 0.8801775147928994,
      "eval_runtime": 3.0184,
      "eval_samples_per_second": 112.974,
      "eval_steps_per_second": 3.644,
      "step": 700
    },
    {
      "epoch": 8.431486880466473,
      "grad_norm": 0.07044461369514465,
      "learning_rate": 2.6178732048650694e-05,
      "loss": 0.0016899970173835754,
      "step": 725
    },
    {
      "epoch": 8.723032069970845,
      "grad_norm": 0.05460745468735695,
      "learning_rate": 2.4392608962010652e-05,
      "loss": 0.0013941892981529237,
      "step": 750
    },
    {
      "epoch": 8.723032069970845,
      "eval_accuracy": 0.9615301955727488,
      "eval_f1": 0.8472121650977552,
      "eval_loss": 0.058827999979257584,
      "eval_precision": 0.8297872340425532,
      "eval_recall": 0.8653846153846154,
      "eval_runtime": 3.0051,
      "eval_samples_per_second": 113.474,
      "eval_steps_per_second": 3.66,
      "step": 750
    },
    {
      "epoch": 9.011661807580175,
      "grad_norm": 0.00831658486276865,
      "learning_rate": 2.2609586633659256e-05,
      "loss": 0.001328500509262085,
      "step": 775
    },
    {
      "epoch": 9.303206997084548,
      "grad_norm": 0.0017224822659045458,
      "learning_rate": 2.0838767471959014e-05,
      "loss": 0.0004855974018573761,
      "step": 800
    },
    {
      "epoch": 9.303206997084548,
      "eval_accuracy": 0.9608854502471523,
      "eval_f1": 0.8466522678185745,
      "eval_loss": 0.061686139553785324,
      "eval_precision": 0.8246844319775596,
      "eval_recall": 0.8698224852071006,
      "eval_runtime": 2.9993,
      "eval_samples_per_second": 113.695,
      "eval_steps_per_second": 3.668,
      "step": 800
    },
    {
      "epoch": 9.594752186588922,
      "grad_norm": 0.03047182224690914,
      "learning_rate": 1.9089191587563417e-05,
      "loss": 0.0007374519854784012,
      "step": 825
    },
    {
      "epoch": 9.886297376093294,
      "grad_norm": 0.040594808757305145,
      "learning_rate": 1.7369790643249573e-05,
      "loss": 0.00048769511282444,
      "step": 850
    },
    {
      "epoch": 9.886297376093294,
      "eval_accuracy": 0.961960025789813,
      "eval_f1": 0.8477785870356883,
      "eval_loss": 0.06452534347772598,
      "eval_precision": 0.8350071736011477,
      "eval_recall": 0.8609467455621301,
      "eval_runtime": 2.9903,
      "eval_samples_per_second": 114.035,
      "eval_steps_per_second": 3.679,
      "step": 850
    },
    {
      "epoch": 10.174927113702624,
      "grad_norm": 0.00843559205532074,
      "learning_rate": 1.5689342257382207e-05,
      "loss": 0.000465988963842392,
      "step": 875
    },
    {
      "epoch": 10.466472303206997,
      "grad_norm": 0.026429006829857826,
      "learning_rate": 1.4056425193781048e-05,
      "loss": 0.0011860659718513488,
      "step": 900
    },
    {
      "epoch": 10.466472303206997,
      "eval_accuracy": 0.9628196862239415,
      "eval_f1": 0.8504398826979472,
      "eval_loss": 0.06725193560123444,
      "eval_precision": 0.8430232558139535,
      "eval_recall": 0.8579881656804734,
      "eval_runtime": 2.9861,
      "eval_samples_per_second": 114.195,
      "eval_steps_per_second": 3.684,
      "step": 900
    },
    {
      "epoch": 10.758017492711371,
      "grad_norm": 0.04398288577795029,
      "learning_rate": 1.2479375566749694e-05,
      "loss": 0.00036286499351263047,
      "step": 925
    },
    {
      "epoch": 11.0466472303207,
      "grad_norm": 0.33958083391189575,
      "learning_rate": 1.0966244284840926e-05,
      "loss": 0.0004555666074156761,
      "step": 950
    },
    {
      "epoch": 11.0466472303207,
      "eval_accuracy": 0.9600257898130239,
      "eval_f1": 0.8442211055276383,
      "eval_loss": 0.06250197440385818,
      "eval_precision": 0.8200836820083682,
      "eval_recall": 0.8698224852071006,
      "eval_runtime": 2.9966,
      "eval_samples_per_second": 113.797,
      "eval_steps_per_second": 3.671,
      "step": 950
    },
    {
      "epoch": 11.338192419825074,
      "grad_norm": 0.04950733855366707,
      "learning_rate": 9.524755950610204e-06,
      "loss": 0.0002466118521988392,
      "step": 975
    },
    {
      "epoch": 11.629737609329446,
      "grad_norm": 0.0015039406716823578,
      "learning_rate": 8.162269426175681e-06,
      "loss": 0.0001478069555014372,
      "step": 1000
    },
    {
      "epoch": 11.629737609329446,
      "eval_accuracy": 0.961960025789813,
      "eval_f1": 0.8481751824817517,
      "eval_loss": 0.06827583909034729,
      "eval_precision": 0.8371757925072046,
      "eval_recall": 0.8594674556213018,
      "eval_runtime": 2.9913,
      "eval_samples_per_second": 113.999,
      "eval_steps_per_second": 3.677,
      "step": 1000
    },
    {
      "epoch": 11.921282798833818,
      "grad_norm": 0.00010757955897133797,
      "learning_rate": 6.885740265899526e-06,
      "loss": 0.0003006239794194698,
      "step": 1025
    },
    {
      "epoch": 12.209912536443149,
      "grad_norm": 0.0013954541645944118,
      "learning_rate": 5.701685207973243e-06,
      "loss": 0.00012959538027644158,
      "step": 1050
    },
    {
      "epoch": 12.209912536443149,
      "eval_accuracy": 0.9617451106812809,
      "eval_f1": 0.8468864468864469,
      "eval_loss": 0.07185657322406769,
      "eval_precision": 0.8388969521044993,
      "eval_recall": 0.8550295857988166,
      "eval_runtime": 3.0121,
      "eval_samples_per_second": 113.211,
      "eval_steps_per_second": 3.652,
      "step": 1050
    },
    {
      "epoch": 12.501457725947521,
      "grad_norm": 0.003432412398979068,
      "learning_rate": 4.616148906179082e-06,
      "loss": 8.389898575842381e-05,
      "step": 1075
    },
    {
      "epoch": 12.793002915451895,
      "grad_norm": 0.0069350870326161385,
      "learning_rate": 3.6346730716634025e-06,
      "loss": 7.618521340191364e-05,
      "step": 1100
    },
    {
      "epoch": 12.793002915451895,
      "eval_accuracy": 0.9617451106812809,
      "eval_f1": 0.8471104608632042,
      "eval_loss": 0.07132749259471893,
      "eval_precision": 0.8379160636758322,
      "eval_recall": 0.856508875739645,
      "eval_runtime": 3.0032,
      "eval_samples_per_second": 113.547,
      "eval_steps_per_second": 3.663,
      "step": 1100
    },
    {
      "epoch": 13.081632653061224,
      "grad_norm": 0.0016698213294148445,
      "learning_rate": 2.7622681822545764e-06,
      "loss": 7.001091726124286e-05,
      "step": 1125
    },
    {
      "epoch": 13.373177842565598,
      "grad_norm": 0.004049114417284727,
      "learning_rate": 2.0033879037506004e-06,
      "loss": 4.9526491202414037e-05,
      "step": 1150
    },
    {
      "epoch": 13.373177842565598,
      "eval_accuracy": 0.9615301955727488,
      "eval_f1": 0.8462664714494876,
      "eval_loss": 0.0716230645775795,
      "eval_precision": 0.8376811594202899,
      "eval_recall": 0.8550295857988166,
      "eval_runtime": 2.9914,
      "eval_samples_per_second": 113.993,
      "eval_steps_per_second": 3.677,
      "step": 1150
    }
  ],
  "logging_steps": 25,
  "max_steps": 1290,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 10,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.935609178420992e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}