{
  "best_metric": 0.7580168991143236,
  "best_model_checkpoint": "/nfs/production/literature/amina-mardiyyah/new_data/OT-Entity-Extraction-Pipeline/model_outputs/hf/CeLLaTe-AL-Test_1.0/base/cellate_tapt_freeze_llrd_ww_mask_LR_2e/checkpoint-920",
  "epoch": 30.0,
  "eval_steps": 500,
  "global_step": 3450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.0460078716278076,
      "learning_rate": 6.434782608695652e-06,
      "loss": 1.3009,
      "step": 115
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8241595949177176,
      "eval_f1": 0.16903515332834707,
      "eval_loss": 0.6194407939910889,
      "eval_precision": 0.3309900410076157,
      "eval_recall": 0.11349939734833267,
      "eval_runtime": 4.4674,
      "eval_samples_per_second": 213.098,
      "eval_steps_per_second": 13.431,
      "step": 115
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.8860368728637695,
      "learning_rate": 1.3101449275362319e-05,
      "loss": 0.3335,
      "step": 230
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9082938721927891,
      "eval_f1": 0.6236261015942172,
      "eval_loss": 0.29858899116516113,
      "eval_precision": 0.6149189611404022,
      "eval_recall": 0.6325833668139815,
      "eval_runtime": 4.5486,
      "eval_samples_per_second": 209.295,
      "eval_steps_per_second": 13.191,
      "step": 230
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.3074285984039307,
      "learning_rate": 1.9768115942028986e-05,
      "loss": 0.1312,
      "step": 345
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9199915607857846,
      "eval_f1": 0.6880798004987531,
      "eval_loss": 0.26223188638687134,
      "eval_precision": 0.6833762631266098,
      "eval_recall": 0.6928485335476094,
      "eval_runtime": 4.4824,
      "eval_samples_per_second": 212.386,
      "eval_steps_per_second": 13.386,
      "step": 345
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.443023204803467,
      "learning_rate": 1.9285024154589372e-05,
      "loss": 0.0797,
      "step": 460
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9264850672792911,
      "eval_f1": 0.7306420915556447,
      "eval_loss": 0.27330681681632996,
      "eval_precision": 0.7286713286713287,
      "eval_recall": 0.7326235435918039,
      "eval_runtime": 4.5427,
      "eval_samples_per_second": 209.567,
      "eval_steps_per_second": 13.208,
      "step": 460
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.006119251251221,
      "learning_rate": 1.8544283413848634e-05,
      "loss": 0.0566,
      "step": 575
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9305874630784378,
      "eval_f1": 0.746298320580344,
      "eval_loss": 0.2804930508136749,
      "eval_precision": 0.7384464110127827,
      "eval_recall": 0.75431900361591,
      "eval_runtime": 4.4864,
      "eval_samples_per_second": 212.196,
      "eval_steps_per_second": 13.374,
      "step": 575
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.0166711807250977,
      "learning_rate": 1.7803542673107893e-05,
      "loss": 0.0404,
      "step": 690
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.927797833935018,
      "eval_f1": 0.7355380362444968,
      "eval_loss": 0.3372795581817627,
      "eval_precision": 0.7500522029651284,
      "eval_recall": 0.7215749296906389,
      "eval_runtime": 4.5271,
      "eval_samples_per_second": 210.29,
      "eval_steps_per_second": 13.254,
      "step": 690
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.738591194152832,
      "learning_rate": 1.706280193236715e-05,
      "loss": 0.0299,
      "step": 805
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9269304702517699,
      "eval_f1": 0.7306815827636214,
      "eval_loss": 0.3340073227882385,
      "eval_precision": 0.7542771599657827,
      "eval_recall": 0.7085174768983528,
      "eval_runtime": 4.5362,
      "eval_samples_per_second": 209.869,
      "eval_steps_per_second": 13.227,
      "step": 805
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.060254454612732,
      "learning_rate": 1.632206119162641e-05,
      "loss": 0.0221,
      "step": 920
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9331426696047634,
      "eval_f1": 0.7580168991143236,
      "eval_loss": 0.31868091225624084,
      "eval_precision": 0.7684210526315789,
      "eval_recall": 0.747890719164323,
      "eval_runtime": 4.4814,
      "eval_samples_per_second": 212.432,
      "eval_steps_per_second": 13.389,
      "step": 920
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.8265031576156616,
      "learning_rate": 1.5581320450885667e-05,
      "loss": 0.0172,
      "step": 1035
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9277275071498898,
      "eval_f1": 0.7327145687767235,
      "eval_loss": 0.39539921283721924,
      "eval_precision": 0.7618737800910865,
      "eval_recall": 0.7057051024507834,
      "eval_runtime": 4.5117,
      "eval_samples_per_second": 211.008,
      "eval_steps_per_second": 13.299,
      "step": 1035
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.03647367283701897,
      "learning_rate": 1.484057971014493e-05,
      "loss": 0.0142,
      "step": 1150
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9295560035632238,
      "eval_f1": 0.7459492509935799,
      "eval_loss": 0.38732069730758667,
      "eval_precision": 0.7569803516028956,
      "eval_recall": 0.7352350341502611,
      "eval_runtime": 4.5775,
      "eval_samples_per_second": 207.974,
      "eval_steps_per_second": 13.108,
      "step": 1150
    },
    {
      "epoch": 11.0,
      "grad_norm": 2.521304130554199,
      "learning_rate": 1.4099838969404187e-05,
      "loss": 0.0116,
      "step": 1265
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9310328660509166,
      "eval_f1": 0.7478805006055712,
      "eval_loss": 0.3827316462993622,
      "eval_precision": 0.7515212981744422,
      "eval_recall": 0.7442748091603053,
      "eval_runtime": 4.4769,
      "eval_samples_per_second": 212.648,
      "eval_steps_per_second": 13.402,
      "step": 1265
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.6821355819702148,
      "learning_rate": 1.3359098228663448e-05,
      "loss": 0.0101,
      "step": 1380
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9300951755825402,
      "eval_f1": 0.7481072232453447,
      "eval_loss": 0.403070330619812,
      "eval_precision": 0.762301918265221,
      "eval_recall": 0.7344314985938127,
      "eval_runtime": 4.4976,
      "eval_samples_per_second": 211.666,
      "eval_steps_per_second": 13.34,
      "step": 1380
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.06722108274698257,
      "learning_rate": 1.2618357487922706e-05,
      "loss": 0.0074,
      "step": 1495
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.9326269398471565,
      "eval_f1": 0.7498711738637535,
      "eval_loss": 0.3884039521217346,
      "eval_precision": 0.7699470899470899,
      "eval_recall": 0.7308155885897951,
      "eval_runtime": 4.5432,
      "eval_samples_per_second": 209.544,
      "eval_steps_per_second": 13.207,
      "step": 1495
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.2433609515428543,
      "learning_rate": 1.1877616747181966e-05,
      "loss": 0.0061,
      "step": 1610
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9247972244362136,
      "eval_f1": 0.7140626626418237,
      "eval_loss": 0.4754845201969147,
      "eval_precision": 0.7409807733851804,
      "eval_recall": 0.6890317396544797,
      "eval_runtime": 4.5526,
      "eval_samples_per_second": 209.113,
      "eval_steps_per_second": 13.179,
      "step": 1610
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.013076603412628174,
      "learning_rate": 1.1136876006441224e-05,
      "loss": 0.0056,
      "step": 1725
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9288761779736509,
      "eval_f1": 0.7404103479036573,
      "eval_loss": 0.4164673686027527,
      "eval_precision": 0.7307767560164351,
      "eval_recall": 0.7503013258336682,
      "eval_runtime": 4.5295,
      "eval_samples_per_second": 210.177,
      "eval_steps_per_second": 13.246,
      "step": 1725
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.03739052265882492,
      "learning_rate": 1.0396135265700484e-05,
      "loss": 0.0052,
      "step": 1840
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9296966571334803,
      "eval_f1": 0.744733952281122,
      "eval_loss": 0.43249383568763733,
      "eval_precision": 0.7402262353641595,
      "eval_recall": 0.7492969063881076,
      "eval_runtime": 4.5747,
      "eval_samples_per_second": 208.102,
      "eval_steps_per_second": 13.116,
      "step": 1840
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.07633493095636368,
      "learning_rate": 9.655394524959744e-06,
      "loss": 0.0059,
      "step": 1955
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.9313141731914295,
      "eval_f1": 0.7502017756255044,
      "eval_loss": 0.4149955213069916,
      "eval_precision": 0.7535468179975678,
      "eval_recall": 0.7468862997187625,
      "eval_runtime": 4.5473,
      "eval_samples_per_second": 209.353,
      "eval_steps_per_second": 13.195,
      "step": 1955
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.006521929986774921,
      "learning_rate": 8.914653784219003e-06,
      "loss": 0.0034,
      "step": 2070
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.9325097285386094,
      "eval_f1": 0.750355474304286,
      "eval_loss": 0.4321502447128296,
      "eval_precision": 0.7588331963845522,
      "eval_recall": 0.7420650863800723,
      "eval_runtime": 4.5005,
      "eval_samples_per_second": 211.53,
      "eval_steps_per_second": 13.332,
      "step": 2070
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.045015785843133926,
      "learning_rate": 8.173913043478263e-06,
      "loss": 0.0037,
      "step": 2185
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.9316892493787801,
      "eval_f1": 0.7498972461981094,
      "eval_loss": 0.4423995316028595,
      "eval_precision": 0.7675641564997896,
      "eval_recall": 0.7330253113700281,
      "eval_runtime": 4.5385,
      "eval_samples_per_second": 209.761,
      "eval_steps_per_second": 13.22,
      "step": 2185
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.007699380628764629,
      "learning_rate": 7.43317230273752e-06,
      "loss": 0.0034,
      "step": 2300
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9285011017863003,
      "eval_f1": 0.7358452138492871,
      "eval_loss": 0.4641232490539551,
      "eval_precision": 0.7461792647666253,
      "eval_recall": 0.7257934913619928,
      "eval_runtime": 4.5632,
      "eval_samples_per_second": 208.626,
      "eval_steps_per_second": 13.149,
      "step": 2300
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.019550923258066177,
      "learning_rate": 6.692431561996779e-06,
      "loss": 0.0028,
      "step": 2415
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.9311031928360448,
      "eval_f1": 0.7502527805864508,
      "eval_loss": 0.4524008631706238,
      "eval_precision": 0.7552931596091205,
      "eval_recall": 0.7452792286058658,
      "eval_runtime": 4.5032,
      "eval_samples_per_second": 211.407,
      "eval_steps_per_second": 13.324,
      "step": 2415
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.006682181265205145,
      "learning_rate": 5.951690821256038e-06,
      "loss": 0.0023,
      "step": 2530
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.9294153499929674,
      "eval_f1": 0.745173745173745,
      "eval_loss": 0.46753570437431335,
      "eval_precision": 0.75390625,
      "eval_recall": 0.7366412213740458,
      "eval_runtime": 4.4674,
      "eval_samples_per_second": 213.099,
      "eval_steps_per_second": 13.431,
      "step": 2530
    },
    {
      "epoch": 23.0,
      "grad_norm": 0.005257639102637768,
      "learning_rate": 5.210950080515298e-06,
      "loss": 0.0022,
      "step": 2645
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.9308453279572413,
      "eval_f1": 0.7522655534059668,
      "eval_loss": 0.4650319218635559,
      "eval_precision": 0.7627503613462729,
      "eval_recall": 0.7420650863800723,
      "eval_runtime": 4.5526,
      "eval_samples_per_second": 209.113,
      "eval_steps_per_second": 13.179,
      "step": 2645
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.04593547806143761,
      "learning_rate": 4.4702093397745575e-06,
      "loss": 0.0021,
      "step": 2760
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9316658071170707,
      "eval_f1": 0.7469412724306688,
      "eval_loss": 0.47501814365386963,
      "eval_precision": 0.7583850931677019,
      "eval_recall": 0.7358376858175975,
      "eval_runtime": 26.4469,
      "eval_samples_per_second": 35.997,
      "eval_steps_per_second": 2.269,
      "step": 2760
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.1780153512954712,
      "learning_rate": 3.729468599033817e-06,
      "loss": 0.0022,
      "step": 2875
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.9307750011721131,
      "eval_f1": 0.7475728155339806,
      "eval_loss": 0.4700013995170593,
      "eval_precision": 0.7527494908350305,
      "eval_recall": 0.7424668541582965,
      "eval_runtime": 4.5324,
      "eval_samples_per_second": 210.043,
      "eval_steps_per_second": 13.238,
      "step": 2875
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.2522285580635071,
      "learning_rate": 2.988727858293076e-06,
      "loss": 0.0015,
      "step": 2990
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.9324628440151906,
      "eval_f1": 0.7531491263713938,
      "eval_loss": 0.47662001848220825,
      "eval_precision": 0.761816687217427,
      "eval_recall": 0.7446765769385295,
      "eval_runtime": 4.5408,
      "eval_samples_per_second": 209.653,
      "eval_steps_per_second": 13.213,
      "step": 2990
    },
    {
      "epoch": 27.0,
      "grad_norm": 0.290810763835907,
      "learning_rate": 2.247987117552335e-06,
      "loss": 0.0018,
      "step": 3105
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.9317361339021989,
      "eval_f1": 0.7472753444375899,
      "eval_loss": 0.48660585284233093,
      "eval_precision": 0.7653748946925021,
      "eval_recall": 0.7300120530333467,
      "eval_runtime": 4.5274,
      "eval_samples_per_second": 210.274,
      "eval_steps_per_second": 13.253,
      "step": 3105
    },
    {
      "epoch": 28.0,
      "grad_norm": 1.0475250482559204,
      "learning_rate": 1.5072463768115944e-06,
      "loss": 0.0016,
      "step": 3220
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9320408833044213,
      "eval_f1": 0.752308472856418,
      "eval_loss": 0.4792047142982483,
      "eval_precision": 0.7600984211605495,
      "eval_recall": 0.7446765769385295,
      "eval_runtime": 4.5334,
      "eval_samples_per_second": 209.997,
      "eval_steps_per_second": 13.235,
      "step": 3220
    },
    {
      "epoch": 29.0,
      "grad_norm": 0.029815517365932465,
      "learning_rate": 7.665056360708536e-07,
      "loss": 0.0014,
      "step": 3335
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.9320174410427118,
      "eval_f1": 0.7504327461561959,
      "eval_loss": 0.4807119071483612,
      "eval_precision": 0.7608920090852778,
      "eval_recall": 0.7402571313780635,
      "eval_runtime": 4.4183,
      "eval_samples_per_second": 215.467,
      "eval_steps_per_second": 13.58,
      "step": 3335
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.06022266671061516,
      "learning_rate": 2.5764895330112724e-08,
      "loss": 0.0014,
      "step": 3450
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.9322049791363871,
      "eval_f1": 0.7508629441624366,
      "eval_loss": 0.4801517724990845,
      "eval_precision": 0.7590311986863711,
      "eval_recall": 0.7428686219365207,
      "eval_runtime": 4.5052,
      "eval_samples_per_second": 211.31,
      "eval_steps_per_second": 13.318,
      "step": 3450
    },
    {
      "epoch": 30.0,
      "step": 3450,
      "total_flos": 1629028238600664.0,
      "train_loss": 0.07025153513403906,
      "train_runtime": 736.8598,
      "train_samples_per_second": 74.79,
      "train_steps_per_second": 4.682
    }
  ],
  "logging_steps": 500,
  "max_steps": 3450,
  "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": 1629028238600664.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}