{
  "best_global_step": 12000,
  "best_metric": 0.8851364411848566,
  "best_model_checkpoint": "/home/rupak/Desktop/Topic-Modeling /topic modeling 25k/checkpoint/hindibert-25k-2e5-256-16-10/checkpoint-12000",
  "epoch": 10.0,
  "eval_steps": 1000,
  "global_step": 12510,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007996801279488205,
      "grad_norm": 0.4256272315979004,
      "learning_rate": 0.0,
      "loss": 1.6057,
      "step": 1
    },
    {
      "epoch": 0.07996801279488205,
      "grad_norm": 0.2559647262096405,
      "learning_rate": 1.5840000000000002e-06,
      "loss": 1.6096,
      "step": 100
    },
    {
      "epoch": 0.1599360255897641,
      "grad_norm": 0.3533499538898468,
      "learning_rate": 3.1840000000000003e-06,
      "loss": 1.6082,
      "step": 200
    },
    {
      "epoch": 0.23990403838464613,
      "grad_norm": 0.8860995769500732,
      "learning_rate": 4.784e-06,
      "loss": 1.5804,
      "step": 300
    },
    {
      "epoch": 0.3198720511795282,
      "grad_norm": 2.5284152030944824,
      "learning_rate": 6.384e-06,
      "loss": 1.5089,
      "step": 400
    },
    {
      "epoch": 0.39984006397441024,
      "grad_norm": 3.2023630142211914,
      "learning_rate": 7.984e-06,
      "loss": 1.4447,
      "step": 500
    },
    {
      "epoch": 0.47980807676929227,
      "grad_norm": 1.5101996660232544,
      "learning_rate": 9.584000000000002e-06,
      "loss": 1.3792,
      "step": 600
    },
    {
      "epoch": 0.5597760895641744,
      "grad_norm": 5.609898567199707,
      "learning_rate": 1.1184000000000002e-05,
      "loss": 1.304,
      "step": 700
    },
    {
      "epoch": 0.6397441023590564,
      "grad_norm": 2.335813283920288,
      "learning_rate": 1.2784000000000002e-05,
      "loss": 1.2209,
      "step": 800
    },
    {
      "epoch": 0.7197121151539384,
      "grad_norm": 8.134674072265625,
      "learning_rate": 1.4384e-05,
      "loss": 1.1316,
      "step": 900
    },
    {
      "epoch": 0.7996801279488205,
      "grad_norm": 3.204949378967285,
      "learning_rate": 1.5984e-05,
      "loss": 1.0329,
      "step": 1000
    },
    {
      "epoch": 0.7996801279488205,
      "eval_accuracy": 0.8532586965213914,
      "eval_auroc_macro": 0.9627139686967083,
      "eval_auroc_weighted": 0.9621425517364535,
      "eval_f1_macro": 0.8528821004884175,
      "eval_f1_weighted": 0.8513342469566539,
      "eval_loss": 0.947596549987793,
      "eval_precision_macro": 0.8554881698862349,
      "eval_precision_weighted": 0.8550249260047849,
      "eval_recall_macro": 0.8556520862335584,
      "eval_recall_weighted": 0.8532586965213914,
      "eval_runtime": 10.6709,
      "eval_samples_per_second": 234.375,
      "eval_steps_per_second": 29.332,
      "step": 1000
    },
    {
      "epoch": 0.8796481407437026,
      "grad_norm": 11.433937072753906,
      "learning_rate": 1.7584e-05,
      "loss": 0.9068,
      "step": 1100
    },
    {
      "epoch": 0.9596161535385845,
      "grad_norm": 7.322030067443848,
      "learning_rate": 1.9184e-05,
      "loss": 0.7819,
      "step": 1200
    },
    {
      "epoch": 1.0391843262694922,
      "grad_norm": 13.326150894165039,
      "learning_rate": 1.991296625222025e-05,
      "loss": 0.671,
      "step": 1300
    },
    {
      "epoch": 1.1191523390643743,
      "grad_norm": 4.284726619720459,
      "learning_rate": 1.9735346358792185e-05,
      "loss": 0.577,
      "step": 1400
    },
    {
      "epoch": 1.1991203518592564,
      "grad_norm": 18.762231826782227,
      "learning_rate": 1.955772646536412e-05,
      "loss": 0.5453,
      "step": 1500
    },
    {
      "epoch": 1.2790883646541382,
      "grad_norm": 7.304553508758545,
      "learning_rate": 1.938010657193606e-05,
      "loss": 0.5108,
      "step": 1600
    },
    {
      "epoch": 1.3590563774490203,
      "grad_norm": 17.503097534179688,
      "learning_rate": 1.9202486678507993e-05,
      "loss": 0.4353,
      "step": 1700
    },
    {
      "epoch": 1.4390243902439024,
      "grad_norm": 8.452184677124023,
      "learning_rate": 1.902486678507993e-05,
      "loss": 0.4634,
      "step": 1800
    },
    {
      "epoch": 1.5189924030387845,
      "grad_norm": 49.753658294677734,
      "learning_rate": 1.884724689165187e-05,
      "loss": 0.4383,
      "step": 1900
    },
    {
      "epoch": 1.5989604158336665,
      "grad_norm": 3.9803502559661865,
      "learning_rate": 1.86696269982238e-05,
      "loss": 0.4214,
      "step": 2000
    },
    {
      "epoch": 1.5989604158336665,
      "eval_accuracy": 0.8780487804878049,
      "eval_auroc_macro": 0.9738165911813731,
      "eval_auroc_weighted": 0.973573288992315,
      "eval_f1_macro": 0.8793645166505886,
      "eval_f1_weighted": 0.8783420244558047,
      "eval_loss": 0.4454324245452881,
      "eval_precision_macro": 0.8818083342062639,
      "eval_precision_weighted": 0.8809044196758252,
      "eval_recall_macro": 0.8792211492934016,
      "eval_recall_weighted": 0.8780487804878049,
      "eval_runtime": 11.1254,
      "eval_samples_per_second": 224.801,
      "eval_steps_per_second": 28.134,
      "step": 2000
    },
    {
      "epoch": 1.6789284286285486,
      "grad_norm": 6.893618583679199,
      "learning_rate": 1.849200710479574e-05,
      "loss": 0.4203,
      "step": 2100
    },
    {
      "epoch": 1.7588964414234307,
      "grad_norm": 5.774092674255371,
      "learning_rate": 1.8314387211367673e-05,
      "loss": 0.3834,
      "step": 2200
    },
    {
      "epoch": 1.8388644542183128,
      "grad_norm": 25.236385345458984,
      "learning_rate": 1.813676731793961e-05,
      "loss": 0.3887,
      "step": 2300
    },
    {
      "epoch": 1.9188324670131949,
      "grad_norm": 19.330398559570312,
      "learning_rate": 1.795914742451155e-05,
      "loss": 0.3582,
      "step": 2400
    },
    {
      "epoch": 1.9988004798080767,
      "grad_norm": 36.057804107666016,
      "learning_rate": 1.778152753108348e-05,
      "loss": 0.3953,
      "step": 2500
    },
    {
      "epoch": 2.0783686525389844,
      "grad_norm": 15.274521827697754,
      "learning_rate": 1.760390763765542e-05,
      "loss": 0.3088,
      "step": 2600
    },
    {
      "epoch": 2.1583366653338665,
      "grad_norm": 1.7254096269607544,
      "learning_rate": 1.7426287744227356e-05,
      "loss": 0.264,
      "step": 2700
    },
    {
      "epoch": 2.2383046781287486,
      "grad_norm": 26.985990524291992,
      "learning_rate": 1.724866785079929e-05,
      "loss": 0.2689,
      "step": 2800
    },
    {
      "epoch": 2.3182726909236306,
      "grad_norm": 5.12957239151001,
      "learning_rate": 1.7071047957371228e-05,
      "loss": 0.2925,
      "step": 2900
    },
    {
      "epoch": 2.3982407037185127,
      "grad_norm": 4.839026927947998,
      "learning_rate": 1.6893428063943164e-05,
      "loss": 0.3011,
      "step": 3000
    },
    {
      "epoch": 2.3982407037185127,
      "eval_accuracy": 0.8788484606157537,
      "eval_auroc_macro": 0.9773569107548961,
      "eval_auroc_weighted": 0.9770285530114762,
      "eval_f1_macro": 0.8791828285515656,
      "eval_f1_weighted": 0.8779722862623245,
      "eval_loss": 0.4504168629646301,
      "eval_precision_macro": 0.879582196101898,
      "eval_precision_weighted": 0.8788972346636038,
      "eval_recall_macro": 0.8805120955311116,
      "eval_recall_weighted": 0.8788484606157537,
      "eval_runtime": 10.8688,
      "eval_samples_per_second": 230.108,
      "eval_steps_per_second": 28.798,
      "step": 3000
    },
    {
      "epoch": 2.478208716513395,
      "grad_norm": 10.992266654968262,
      "learning_rate": 1.67158081705151e-05,
      "loss": 0.2742,
      "step": 3100
    },
    {
      "epoch": 2.5581767293082764,
      "grad_norm": 4.304646968841553,
      "learning_rate": 1.6538188277087036e-05,
      "loss": 0.2733,
      "step": 3200
    },
    {
      "epoch": 2.638144742103159,
      "grad_norm": 60.62465286254883,
      "learning_rate": 1.636056838365897e-05,
      "loss": 0.3115,
      "step": 3300
    },
    {
      "epoch": 2.7181127548980406,
      "grad_norm": 1.9873929023742676,
      "learning_rate": 1.6182948490230908e-05,
      "loss": 0.3169,
      "step": 3400
    },
    {
      "epoch": 2.7980807676929227,
      "grad_norm": 8.354615211486816,
      "learning_rate": 1.6005328596802844e-05,
      "loss": 0.3139,
      "step": 3500
    },
    {
      "epoch": 2.8780487804878048,
      "grad_norm": 27.897375106811523,
      "learning_rate": 1.582770870337478e-05,
      "loss": 0.2708,
      "step": 3600
    },
    {
      "epoch": 2.958016793282687,
      "grad_norm": 57.23756790161133,
      "learning_rate": 1.5650088809946716e-05,
      "loss": 0.2551,
      "step": 3700
    },
    {
      "epoch": 3.0375849660135947,
      "grad_norm": 2.6712305545806885,
      "learning_rate": 1.5472468916518652e-05,
      "loss": 0.262,
      "step": 3800
    },
    {
      "epoch": 3.117552978808477,
      "grad_norm": 4.725401401519775,
      "learning_rate": 1.5294849023090588e-05,
      "loss": 0.1919,
      "step": 3900
    },
    {
      "epoch": 3.1975209916033585,
      "grad_norm": 28.024898529052734,
      "learning_rate": 1.5117229129662522e-05,
      "loss": 0.1611,
      "step": 4000
    },
    {
      "epoch": 3.1975209916033585,
      "eval_accuracy": 0.884046381447421,
      "eval_auroc_macro": 0.9757735583965023,
      "eval_auroc_weighted": 0.9753618417054956,
      "eval_f1_macro": 0.8845660655909923,
      "eval_f1_weighted": 0.8834468599472054,
      "eval_loss": 0.4886566996574402,
      "eval_precision_macro": 0.8844933323070212,
      "eval_precision_weighted": 0.8837534689063661,
      "eval_recall_macro": 0.8855092755592715,
      "eval_recall_weighted": 0.884046381447421,
      "eval_runtime": 11.0978,
      "eval_samples_per_second": 225.361,
      "eval_steps_per_second": 28.204,
      "step": 4000
    },
    {
      "epoch": 3.2774890043982405,
      "grad_norm": 0.42076241970062256,
      "learning_rate": 1.493960923623446e-05,
      "loss": 0.1631,
      "step": 4100
    },
    {
      "epoch": 3.3574570171931226,
      "grad_norm": 22.10235595703125,
      "learning_rate": 1.4761989342806396e-05,
      "loss": 0.1887,
      "step": 4200
    },
    {
      "epoch": 3.4374250299880047,
      "grad_norm": 20.63714027404785,
      "learning_rate": 1.458436944937833e-05,
      "loss": 0.2175,
      "step": 4300
    },
    {
      "epoch": 3.517393042782887,
      "grad_norm": 3.4005095958709717,
      "learning_rate": 1.4406749555950268e-05,
      "loss": 0.224,
      "step": 4400
    },
    {
      "epoch": 3.597361055577769,
      "grad_norm": 0.43836817145347595,
      "learning_rate": 1.4229129662522204e-05,
      "loss": 0.1809,
      "step": 4500
    },
    {
      "epoch": 3.677329068372651,
      "grad_norm": 37.63475799560547,
      "learning_rate": 1.4051509769094141e-05,
      "loss": 0.22,
      "step": 4600
    },
    {
      "epoch": 3.757297081167533,
      "grad_norm": 9.212823867797852,
      "learning_rate": 1.3873889875666075e-05,
      "loss": 0.1919,
      "step": 4700
    },
    {
      "epoch": 3.837265093962415,
      "grad_norm": 15.547619819641113,
      "learning_rate": 1.3696269982238011e-05,
      "loss": 0.1902,
      "step": 4800
    },
    {
      "epoch": 3.917233106757297,
      "grad_norm": 36.606658935546875,
      "learning_rate": 1.3518650088809947e-05,
      "loss": 0.2083,
      "step": 4900
    },
    {
      "epoch": 3.9972011195521793,
      "grad_norm": 27.680450439453125,
      "learning_rate": 1.3341030195381883e-05,
      "loss": 0.2711,
      "step": 5000
    },
    {
      "epoch": 3.9972011195521793,
      "eval_accuracy": 0.868452618952419,
      "eval_auroc_macro": 0.9740067888918509,
      "eval_auroc_weighted": 0.9734915246888479,
      "eval_f1_macro": 0.8672746760101144,
      "eval_f1_weighted": 0.8657240340371538,
      "eval_loss": 0.5831986665725708,
      "eval_precision_macro": 0.8713544170659105,
      "eval_precision_weighted": 0.8714247844372666,
      "eval_recall_macro": 0.8712665611651452,
      "eval_recall_weighted": 0.868452618952419,
      "eval_runtime": 10.6548,
      "eval_samples_per_second": 234.729,
      "eval_steps_per_second": 29.376,
      "step": 5000
    },
    {
      "epoch": 4.076769292283087,
      "grad_norm": 7.681755542755127,
      "learning_rate": 1.3163410301953821e-05,
      "loss": 0.1556,
      "step": 5100
    },
    {
      "epoch": 4.156737305077969,
      "grad_norm": 0.0798175036907196,
      "learning_rate": 1.2985790408525755e-05,
      "loss": 0.1634,
      "step": 5200
    },
    {
      "epoch": 4.23670531787285,
      "grad_norm": 16.147348403930664,
      "learning_rate": 1.2808170515097691e-05,
      "loss": 0.1468,
      "step": 5300
    },
    {
      "epoch": 4.316673330667733,
      "grad_norm": 56.157772064208984,
      "learning_rate": 1.2630550621669629e-05,
      "loss": 0.1396,
      "step": 5400
    },
    {
      "epoch": 4.396641343462615,
      "grad_norm": 71.01326751708984,
      "learning_rate": 1.2452930728241563e-05,
      "loss": 0.1403,
      "step": 5500
    },
    {
      "epoch": 4.476609356257497,
      "grad_norm": 72.29058837890625,
      "learning_rate": 1.22753108348135e-05,
      "loss": 0.1346,
      "step": 5600
    },
    {
      "epoch": 4.556577369052379,
      "grad_norm": 12.029815673828125,
      "learning_rate": 1.2097690941385437e-05,
      "loss": 0.1661,
      "step": 5700
    },
    {
      "epoch": 4.636545381847261,
      "grad_norm": 105.95983123779297,
      "learning_rate": 1.1920071047957371e-05,
      "loss": 0.16,
      "step": 5800
    },
    {
      "epoch": 4.716513394642143,
      "grad_norm": 47.3993034362793,
      "learning_rate": 1.1742451154529309e-05,
      "loss": 0.1765,
      "step": 5900
    },
    {
      "epoch": 4.7964814074370254,
      "grad_norm": 0.10814545303583145,
      "learning_rate": 1.1564831261101243e-05,
      "loss": 0.1354,
      "step": 6000
    },
    {
      "epoch": 4.7964814074370254,
      "eval_accuracy": 0.8848460615753698,
      "eval_auroc_macro": 0.9771739659875225,
      "eval_auroc_weighted": 0.9767904877872695,
      "eval_f1_macro": 0.885015833309326,
      "eval_f1_weighted": 0.8839241041685038,
      "eval_loss": 0.5577186346054077,
      "eval_precision_macro": 0.8853823194756998,
      "eval_precision_weighted": 0.8853725074070478,
      "eval_recall_macro": 0.8869834176751912,
      "eval_recall_weighted": 0.8848460615753698,
      "eval_runtime": 10.9779,
      "eval_samples_per_second": 227.821,
      "eval_steps_per_second": 28.512,
      "step": 6000
    },
    {
      "epoch": 4.876449420231907,
      "grad_norm": 45.430564880371094,
      "learning_rate": 1.138721136767318e-05,
      "loss": 0.1184,
      "step": 6100
    },
    {
      "epoch": 4.95641743302679,
      "grad_norm": 0.07600370794534683,
      "learning_rate": 1.1209591474245117e-05,
      "loss": 0.1323,
      "step": 6200
    },
    {
      "epoch": 5.035985605757697,
      "grad_norm": 0.07455410063266754,
      "learning_rate": 1.103197158081705e-05,
      "loss": 0.1142,
      "step": 6300
    },
    {
      "epoch": 5.115953618552579,
      "grad_norm": 7.246333122253418,
      "learning_rate": 1.0854351687388988e-05,
      "loss": 0.1095,
      "step": 6400
    },
    {
      "epoch": 5.195921631347461,
      "grad_norm": 0.050232477486133575,
      "learning_rate": 1.0676731793960924e-05,
      "loss": 0.097,
      "step": 6500
    },
    {
      "epoch": 5.275889644142343,
      "grad_norm": 2.7280073165893555,
      "learning_rate": 1.0499111900532862e-05,
      "loss": 0.0806,
      "step": 6600
    },
    {
      "epoch": 5.355857656937225,
      "grad_norm": 5.1381354331970215,
      "learning_rate": 1.0321492007104796e-05,
      "loss": 0.1225,
      "step": 6700
    },
    {
      "epoch": 5.4358256697321075,
      "grad_norm": 73.33670806884766,
      "learning_rate": 1.0143872113676732e-05,
      "loss": 0.1127,
      "step": 6800
    },
    {
      "epoch": 5.515793682526989,
      "grad_norm": 0.04268639162182808,
      "learning_rate": 9.966252220248668e-06,
      "loss": 0.0951,
      "step": 6900
    },
    {
      "epoch": 5.595761695321872,
      "grad_norm": 14.716099739074707,
      "learning_rate": 9.788632326820606e-06,
      "loss": 0.134,
      "step": 7000
    },
    {
      "epoch": 5.595761695321872,
      "eval_accuracy": 0.8796481407437026,
      "eval_auroc_macro": 0.9736991302553468,
      "eval_auroc_weighted": 0.9733130910363977,
      "eval_f1_macro": 0.8809953099105533,
      "eval_f1_weighted": 0.8797942315789191,
      "eval_loss": 0.6088235974311829,
      "eval_precision_macro": 0.8815953725252788,
      "eval_precision_weighted": 0.8805306902636757,
      "eval_recall_macro": 0.8809931182513335,
      "eval_recall_weighted": 0.8796481407437026,
      "eval_runtime": 10.9456,
      "eval_samples_per_second": 228.493,
      "eval_steps_per_second": 28.596,
      "step": 7000
    },
    {
      "epoch": 5.675729708116753,
      "grad_norm": 0.0554790124297142,
      "learning_rate": 9.61101243339254e-06,
      "loss": 0.0785,
      "step": 7100
    },
    {
      "epoch": 5.755697720911636,
      "grad_norm": 0.03109428659081459,
      "learning_rate": 9.433392539964476e-06,
      "loss": 0.111,
      "step": 7200
    },
    {
      "epoch": 5.835665733706517,
      "grad_norm": 4.9373459815979,
      "learning_rate": 9.255772646536412e-06,
      "loss": 0.0948,
      "step": 7300
    },
    {
      "epoch": 5.915633746501399,
      "grad_norm": 72.38359832763672,
      "learning_rate": 9.07815275310835e-06,
      "loss": 0.1146,
      "step": 7400
    },
    {
      "epoch": 5.995601759296282,
      "grad_norm": 3.889324188232422,
      "learning_rate": 8.900532859680286e-06,
      "loss": 0.0973,
      "step": 7500
    },
    {
      "epoch": 6.0751699320271895,
      "grad_norm": 0.04162607342004776,
      "learning_rate": 8.72291296625222e-06,
      "loss": 0.0644,
      "step": 7600
    },
    {
      "epoch": 6.155137944822071,
      "grad_norm": 0.038938213139772415,
      "learning_rate": 8.545293072824158e-06,
      "loss": 0.0499,
      "step": 7700
    },
    {
      "epoch": 6.235105957616954,
      "grad_norm": 0.22254200279712677,
      "learning_rate": 8.367673179396094e-06,
      "loss": 0.0596,
      "step": 7800
    },
    {
      "epoch": 6.315073970411835,
      "grad_norm": 0.10705538839101791,
      "learning_rate": 8.19005328596803e-06,
      "loss": 0.0569,
      "step": 7900
    },
    {
      "epoch": 6.395041983206717,
      "grad_norm": 0.022395452484488487,
      "learning_rate": 8.012433392539965e-06,
      "loss": 0.0877,
      "step": 8000
    },
    {
      "epoch": 6.395041983206717,
      "eval_accuracy": 0.8740503798480608,
      "eval_auroc_macro": 0.974347054440335,
      "eval_auroc_weighted": 0.9739933549338261,
      "eval_f1_macro": 0.8735827403639422,
      "eval_f1_weighted": 0.8723502643051826,
      "eval_loss": 0.7089990377426147,
      "eval_precision_macro": 0.8751255771125654,
      "eval_precision_weighted": 0.8753003975964507,
      "eval_recall_macro": 0.8765722187069322,
      "eval_recall_weighted": 0.8740503798480608,
      "eval_runtime": 5.2646,
      "eval_samples_per_second": 475.056,
      "eval_steps_per_second": 59.453,
      "step": 8000
    },
    {
      "epoch": 6.475009996001599,
      "grad_norm": 0.02474510669708252,
      "learning_rate": 7.834813499111901e-06,
      "loss": 0.0933,
      "step": 8100
    },
    {
      "epoch": 6.554978008796481,
      "grad_norm": 0.051300596445798874,
      "learning_rate": 7.657193605683837e-06,
      "loss": 0.0686,
      "step": 8200
    },
    {
      "epoch": 6.634946021591364,
      "grad_norm": 0.041189149022102356,
      "learning_rate": 7.479573712255773e-06,
      "loss": 0.0829,
      "step": 8300
    },
    {
      "epoch": 6.714914034386245,
      "grad_norm": 5.698017120361328,
      "learning_rate": 7.301953818827709e-06,
      "loss": 0.0625,
      "step": 8400
    },
    {
      "epoch": 6.794882047181128,
      "grad_norm": 0.09351879358291626,
      "learning_rate": 7.124333925399646e-06,
      "loss": 0.0704,
      "step": 8500
    },
    {
      "epoch": 6.874850059976009,
      "grad_norm": 0.3524141311645508,
      "learning_rate": 6.946714031971581e-06,
      "loss": 0.0769,
      "step": 8600
    },
    {
      "epoch": 6.954818072770892,
      "grad_norm": 0.03331638127565384,
      "learning_rate": 6.769094138543517e-06,
      "loss": 0.0717,
      "step": 8700
    },
    {
      "epoch": 7.034386245501799,
      "grad_norm": 0.6107684969902039,
      "learning_rate": 6.591474245115453e-06,
      "loss": 0.049,
      "step": 8800
    },
    {
      "epoch": 7.114354258296681,
      "grad_norm": 0.027097152546048164,
      "learning_rate": 6.41385435168739e-06,
      "loss": 0.0437,
      "step": 8900
    },
    {
      "epoch": 7.194322271091563,
      "grad_norm": 0.019937481731176376,
      "learning_rate": 6.236234458259326e-06,
      "loss": 0.0741,
      "step": 9000
    },
    {
      "epoch": 7.194322271091563,
      "eval_accuracy": 0.8780487804878049,
      "eval_auroc_macro": 0.9715743344544261,
      "eval_auroc_weighted": 0.9710196639814077,
      "eval_f1_macro": 0.8781859377674458,
      "eval_f1_weighted": 0.8769698095840991,
      "eval_loss": 0.7172689437866211,
      "eval_precision_macro": 0.8782446963481615,
      "eval_precision_weighted": 0.8779260460077373,
      "eval_recall_macro": 0.8800909624847286,
      "eval_recall_weighted": 0.8780487804878049,
      "eval_runtime": 5.2461,
      "eval_samples_per_second": 476.737,
      "eval_steps_per_second": 59.664,
      "step": 9000
    },
    {
      "epoch": 7.274290283886446,
      "grad_norm": 0.1962634027004242,
      "learning_rate": 6.058614564831261e-06,
      "loss": 0.0461,
      "step": 9100
    },
    {
      "epoch": 7.354258296681327,
      "grad_norm": 0.3250579237937927,
      "learning_rate": 5.880994671403197e-06,
      "loss": 0.0465,
      "step": 9200
    },
    {
      "epoch": 7.43422630947621,
      "grad_norm": 0.034404072910547256,
      "learning_rate": 5.703374777975134e-06,
      "loss": 0.0429,
      "step": 9300
    },
    {
      "epoch": 7.514194322271091,
      "grad_norm": 0.021159108728170395,
      "learning_rate": 5.52575488454707e-06,
      "loss": 0.0352,
      "step": 9400
    },
    {
      "epoch": 7.594162335065974,
      "grad_norm": 0.08214077353477478,
      "learning_rate": 5.3481349911190065e-06,
      "loss": 0.0597,
      "step": 9500
    },
    {
      "epoch": 7.674130347860856,
      "grad_norm": 0.026783457025885582,
      "learning_rate": 5.170515097690942e-06,
      "loss": 0.0488,
      "step": 9600
    },
    {
      "epoch": 7.754098360655737,
      "grad_norm": 30.816205978393555,
      "learning_rate": 4.992895204262878e-06,
      "loss": 0.0636,
      "step": 9700
    },
    {
      "epoch": 7.83406637345062,
      "grad_norm": 0.01820790022611618,
      "learning_rate": 4.8152753108348135e-06,
      "loss": 0.0432,
      "step": 9800
    },
    {
      "epoch": 7.914034386245502,
      "grad_norm": 0.13603974878787994,
      "learning_rate": 4.63765541740675e-06,
      "loss": 0.0376,
      "step": 9900
    },
    {
      "epoch": 7.994002399040384,
      "grad_norm": 0.05620984360575676,
      "learning_rate": 4.4600355239786855e-06,
      "loss": 0.0532,
      "step": 10000
    },
    {
      "epoch": 7.994002399040384,
      "eval_accuracy": 0.8808476609356257,
      "eval_auroc_macro": 0.9712888548840507,
      "eval_auroc_weighted": 0.9708107694628548,
      "eval_f1_macro": 0.8808544746623334,
      "eval_f1_weighted": 0.8795725129000943,
      "eval_loss": 0.7412012815475464,
      "eval_precision_macro": 0.8813573909705014,
      "eval_precision_weighted": 0.8808757699391696,
      "eval_recall_macro": 0.882832478485879,
      "eval_recall_weighted": 0.8808476609356257,
      "eval_runtime": 5.0716,
      "eval_samples_per_second": 493.136,
      "eval_steps_per_second": 61.716,
      "step": 10000
    },
    {
      "epoch": 8.073570571771292,
      "grad_norm": 0.29488369822502136,
      "learning_rate": 4.282415630550622e-06,
      "loss": 0.0316,
      "step": 10100
    },
    {
      "epoch": 8.153538584566174,
      "grad_norm": 0.01641344465315342,
      "learning_rate": 4.104795737122558e-06,
      "loss": 0.0302,
      "step": 10200
    },
    {
      "epoch": 8.233506597361055,
      "grad_norm": 0.016257140785455704,
      "learning_rate": 3.927175843694494e-06,
      "loss": 0.0377,
      "step": 10300
    },
    {
      "epoch": 8.313474610155938,
      "grad_norm": 0.016158757731318474,
      "learning_rate": 3.74955595026643e-06,
      "loss": 0.0158,
      "step": 10400
    },
    {
      "epoch": 8.39344262295082,
      "grad_norm": 63.62160873413086,
      "learning_rate": 3.571936056838366e-06,
      "loss": 0.032,
      "step": 10500
    },
    {
      "epoch": 8.4734106357457,
      "grad_norm": 0.01895180717110634,
      "learning_rate": 3.394316163410302e-06,
      "loss": 0.031,
      "step": 10600
    },
    {
      "epoch": 8.553378648540583,
      "grad_norm": 0.013899303041398525,
      "learning_rate": 3.2166962699822384e-06,
      "loss": 0.0338,
      "step": 10700
    },
    {
      "epoch": 8.633346661335466,
      "grad_norm": 0.01123654842376709,
      "learning_rate": 3.0390763765541744e-06,
      "loss": 0.0288,
      "step": 10800
    },
    {
      "epoch": 8.713314674130348,
      "grad_norm": 113.46405792236328,
      "learning_rate": 2.8614564831261103e-06,
      "loss": 0.035,
      "step": 10900
    },
    {
      "epoch": 8.79328268692523,
      "grad_norm": 0.3093399703502655,
      "learning_rate": 2.6838365896980463e-06,
      "loss": 0.0133,
      "step": 11000
    },
    {
      "epoch": 8.79328268692523,
      "eval_accuracy": 0.8852459016393442,
      "eval_auroc_macro": 0.9700113689142762,
      "eval_auroc_weighted": 0.9694695236422716,
      "eval_f1_macro": 0.8852194619626491,
      "eval_f1_weighted": 0.8840464836518896,
      "eval_loss": 0.7584248185157776,
      "eval_precision_macro": 0.8853058007130382,
      "eval_precision_weighted": 0.8849852171837517,
      "eval_recall_macro": 0.8871884880754444,
      "eval_recall_weighted": 0.8852459016393442,
      "eval_runtime": 2.9336,
      "eval_samples_per_second": 852.549,
      "eval_steps_per_second": 106.696,
      "step": 11000
    },
    {
      "epoch": 8.873250699720112,
      "grad_norm": 1.7755924463272095,
      "learning_rate": 2.5062166962699823e-06,
      "loss": 0.041,
      "step": 11100
    },
    {
      "epoch": 8.953218712514994,
      "grad_norm": 0.01337814424186945,
      "learning_rate": 2.3285968028419182e-06,
      "loss": 0.0393,
      "step": 11200
    },
    {
      "epoch": 9.032786885245901,
      "grad_norm": 0.07428575307130814,
      "learning_rate": 2.1509769094138546e-06,
      "loss": 0.0284,
      "step": 11300
    },
    {
      "epoch": 9.112754898040784,
      "grad_norm": 0.02315426617860794,
      "learning_rate": 1.9733570159857906e-06,
      "loss": 0.0143,
      "step": 11400
    },
    {
      "epoch": 9.192722910835666,
      "grad_norm": 0.0839715525507927,
      "learning_rate": 1.7957371225577267e-06,
      "loss": 0.0245,
      "step": 11500
    },
    {
      "epoch": 9.272690923630547,
      "grad_norm": 0.13397835195064545,
      "learning_rate": 1.6181172291296627e-06,
      "loss": 0.0356,
      "step": 11600
    },
    {
      "epoch": 9.35265893642543,
      "grad_norm": 0.06111783906817436,
      "learning_rate": 1.4404973357015986e-06,
      "loss": 0.0241,
      "step": 11700
    },
    {
      "epoch": 9.432626949220312,
      "grad_norm": 0.7590734362602234,
      "learning_rate": 1.2628774422735348e-06,
      "loss": 0.0208,
      "step": 11800
    },
    {
      "epoch": 9.512594962015195,
      "grad_norm": 0.01169233862310648,
      "learning_rate": 1.0852575488454708e-06,
      "loss": 0.0475,
      "step": 11900
    },
    {
      "epoch": 9.592562974810075,
      "grad_norm": 0.015741437673568726,
      "learning_rate": 9.076376554174068e-07,
      "loss": 0.0357,
      "step": 12000
    },
    {
      "epoch": 9.592562974810075,
      "eval_accuracy": 0.8860455817672931,
      "eval_auroc_macro": 0.9713983635818311,
      "eval_auroc_weighted": 0.9709143753060755,
      "eval_f1_macro": 0.8862374765222798,
      "eval_f1_weighted": 0.8851364411848566,
      "eval_loss": 0.7631213068962097,
      "eval_precision_macro": 0.8860695429970787,
      "eval_precision_weighted": 0.8856898975950318,
      "eval_recall_macro": 0.8878156849040522,
      "eval_recall_weighted": 0.8860455817672931,
      "eval_runtime": 2.9757,
      "eval_samples_per_second": 840.488,
      "eval_steps_per_second": 105.187,
      "step": 12000
    },
    {
      "epoch": 9.672530987604958,
      "grad_norm": 0.07069654017686844,
      "learning_rate": 7.300177619893428e-07,
      "loss": 0.0268,
      "step": 12100
    },
    {
      "epoch": 9.75249900039984,
      "grad_norm": 0.06307296454906464,
      "learning_rate": 5.523978685612789e-07,
      "loss": 0.0137,
      "step": 12200
    },
    {
      "epoch": 9.832467013194723,
      "grad_norm": 0.10373717546463013,
      "learning_rate": 3.747779751332149e-07,
      "loss": 0.0392,
      "step": 12300
    },
    {
      "epoch": 9.912435025989604,
      "grad_norm": 0.06551024317741394,
      "learning_rate": 1.9715808170515098e-07,
      "loss": 0.0207,
      "step": 12400
    },
    {
      "epoch": 9.992403038784486,
      "grad_norm": 0.2663117051124573,
      "learning_rate": 1.9538188277087032e-08,
      "loss": 0.0306,
      "step": 12500
    }
  ],
  "logging_steps": 100,
  "max_steps": 12510,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 15,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3366600744047856.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}