{
  "best_metric": 0.9783398772157638,
  "best_model_checkpoint": "models/pos_final_mono_nl/checkpoint-2760",
  "epoch": 39.99638989169675,
  "global_step": 2760,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2575113142718755,
      "eval_f1": 0.16891972475874908,
      "eval_loss": 3.770303726196289,
      "eval_precision": 0.25972286447785947,
      "eval_recall": 0.12516136964406466,
      "eval_runtime": 10.2346,
      "eval_samples_per_second": 768.865,
      "eval_steps_per_second": 3.029,
      "step": 69
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8066356880136831,
      "eval_f1": 0.8033051608452517,
      "eval_loss": 1.0147907733917236,
      "eval_precision": 0.8057864338897609,
      "eval_recall": 0.8008391221491363,
      "eval_runtime": 10.7861,
      "eval_samples_per_second": 729.548,
      "eval_steps_per_second": 2.874,
      "step": 138
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.929904490895606,
      "eval_f1": 0.9289770104330163,
      "eval_loss": 0.3402073085308075,
      "eval_precision": 0.9301506840872673,
      "eval_recall": 0.9278062949529723,
      "eval_runtime": 10.421,
      "eval_samples_per_second": 755.108,
      "eval_steps_per_second": 2.975,
      "step": 207
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9560597575188824,
      "eval_f1": 0.9555069516784352,
      "eval_loss": 0.20157238841056824,
      "eval_precision": 0.9559442226785728,
      "eval_recall": 0.9550700805311366,
      "eval_runtime": 10.4533,
      "eval_samples_per_second": 752.775,
      "eval_steps_per_second": 2.966,
      "step": 276
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.964831156250473,
      "eval_f1": 0.964063779010111,
      "eval_loss": 0.14858682453632355,
      "eval_precision": 0.9642898327887757,
      "eval_recall": 0.9638378311919837,
      "eval_runtime": 11.0648,
      "eval_samples_per_second": 711.175,
      "eval_steps_per_second": 2.802,
      "step": 345
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9701515128581591,
      "eval_f1": 0.9696606418295832,
      "eval_loss": 0.12055634707212448,
      "eval_precision": 0.9697202582231786,
      "eval_recall": 0.969601032765722,
      "eval_runtime": 10.8285,
      "eval_samples_per_second": 726.696,
      "eval_steps_per_second": 2.863,
      "step": 414
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.972701954076922,
      "eval_f1": 0.9719634068091613,
      "eval_loss": 0.10631231963634491,
      "eval_precision": 0.9719820795967141,
      "eval_recall": 0.9719447347390422,
      "eval_runtime": 10.4365,
      "eval_samples_per_second": 753.985,
      "eval_steps_per_second": 2.97,
      "step": 483
    },
    {
      "epoch": 7.25,
      "learning_rate": 5e-05,
      "loss": 1.2192,
      "step": 500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9742231371183798,
      "eval_f1": 0.9734790710183069,
      "eval_loss": 0.09831023961305618,
      "eval_precision": 0.9734304527887268,
      "eval_recall": 0.973527694104629,
      "eval_runtime": 10.9399,
      "eval_samples_per_second": 719.291,
      "eval_steps_per_second": 2.834,
      "step": 552
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.975388620642681,
      "eval_f1": 0.9746202254443189,
      "eval_loss": 0.09469176828861237,
      "eval_precision": 0.974567806377257,
      "eval_recall": 0.9746726501506117,
      "eval_runtime": 10.9888,
      "eval_samples_per_second": 716.095,
      "eval_steps_per_second": 2.821,
      "step": 621
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9761454281259934,
      "eval_f1": 0.9753816837883316,
      "eval_loss": 0.09128155559301376,
      "eval_precision": 0.9752992516787289,
      "eval_recall": 0.975464129833405,
      "eval_runtime": 10.3458,
      "eval_samples_per_second": 760.6,
      "eval_steps_per_second": 2.996,
      "step": 690
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9770081886569695,
      "eval_f1": 0.9762231314470121,
      "eval_loss": 0.08845613151788712,
      "eval_precision": 0.9761368787406173,
      "eval_recall": 0.9763093993975533,
      "eval_runtime": 10.591,
      "eval_samples_per_second": 742.993,
      "eval_steps_per_second": 2.927,
      "step": 759
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9772200947522969,
      "eval_f1": 0.9764375477837924,
      "eval_loss": 0.08773986995220184,
      "eval_precision": 0.9763812802053,
      "eval_recall": 0.976493821847913,
      "eval_runtime": 10.1848,
      "eval_samples_per_second": 772.621,
      "eval_steps_per_second": 3.044,
      "step": 828
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.977530385820455,
      "eval_f1": 0.9767615183960107,
      "eval_loss": 0.0878407210111618,
      "eval_precision": 0.9766527100218952,
      "eval_recall": 0.9768703510173972,
      "eval_runtime": 10.414,
      "eval_samples_per_second": 755.619,
      "eval_steps_per_second": 2.977,
      "step": 897
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9775909304191199,
      "eval_f1": 0.9767761714683069,
      "eval_loss": 0.08732089400291443,
      "eval_precision": 0.9766973731723995,
      "eval_recall": 0.9768549824798672,
      "eval_runtime": 10.2517,
      "eval_samples_per_second": 767.579,
      "eval_steps_per_second": 3.024,
      "step": 966
    },
    {
      "epoch": 14.49,
      "learning_rate": 3.893805309734514e-05,
      "loss": 0.0688,
      "step": 1000
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9779466299362768,
      "eval_f1": 0.9771878601613523,
      "eval_loss": 0.08772371709346771,
      "eval_precision": 0.9770827571371501,
      "eval_recall": 0.9772929857994713,
      "eval_runtime": 10.4702,
      "eval_samples_per_second": 751.563,
      "eval_steps_per_second": 2.961,
      "step": 1035
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9781282637322718,
      "eval_f1": 0.9773326264186318,
      "eval_loss": 0.08782745897769928,
      "eval_precision": 0.9772800614675374,
      "eval_recall": 0.9773851970246511,
      "eval_runtime": 10.0217,
      "eval_samples_per_second": 785.196,
      "eval_steps_per_second": 3.093,
      "step": 1104
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.9780601510587736,
      "eval_f1": 0.9772529072559422,
      "eval_loss": 0.08969255536794662,
      "eval_precision": 0.9771590568603499,
      "eval_recall": 0.9773467756808262,
      "eval_runtime": 10.2237,
      "eval_samples_per_second": 769.681,
      "eval_steps_per_second": 3.032,
      "step": 1173
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.978325033677933,
      "eval_f1": 0.9775826100987749,
      "eval_loss": 0.09088694304227829,
      "eval_precision": 0.9775187663749587,
      "eval_recall": 0.9776464621626606,
      "eval_runtime": 10.2448,
      "eval_samples_per_second": 768.099,
      "eval_steps_per_second": 3.026,
      "step": 1242
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.9784536909500962,
      "eval_f1": 0.9776833564477773,
      "eval_loss": 0.09170977026224136,
      "eval_precision": 0.9775819549334296,
      "eval_recall": 0.9777847790004304,
      "eval_runtime": 10.5329,
      "eval_samples_per_second": 747.089,
      "eval_steps_per_second": 2.943,
      "step": 1311
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.978695869344756,
      "eval_f1": 0.977932645393286,
      "eval_loss": 0.09237655997276306,
      "eval_precision": 0.977849997695109,
      "eval_recall": 0.9780153070633798,
      "eval_runtime": 10.1755,
      "eval_samples_per_second": 773.331,
      "eval_steps_per_second": 3.047,
      "step": 1380
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.9785369397732605,
      "eval_f1": 0.9777796559381027,
      "eval_loss": 0.09489051252603531,
      "eval_precision": 0.9776669790882412,
      "eval_recall": 0.9778923587631401,
      "eval_runtime": 10.3245,
      "eval_samples_per_second": 762.165,
      "eval_steps_per_second": 3.003,
      "step": 1449
    },
    {
      "epoch": 21.74,
      "learning_rate": 2.7876106194690264e-05,
      "loss": 0.0366,
      "step": 1500
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.9783931463514312,
      "eval_f1": 0.977655520039341,
      "eval_loss": 0.09559858590364456,
      "eval_precision": 0.9775954268854877,
      "eval_recall": 0.9777156205815455,
      "eval_runtime": 10.2656,
      "eval_samples_per_second": 766.54,
      "eval_steps_per_second": 3.02,
      "step": 1518
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.9785899162970924,
      "eval_f1": 0.9778600762968741,
      "eval_loss": 0.0962114930152893,
      "eval_precision": 0.977758656453831,
      "eval_recall": 0.977961517182025,
      "eval_runtime": 10.2847,
      "eval_samples_per_second": 765.116,
      "eval_steps_per_second": 3.014,
      "step": 1587
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.978627756671258,
      "eval_f1": 0.9778800497871751,
      "eval_loss": 0.09919747710227966,
      "eval_precision": 0.9777448299173401,
      "eval_recall": 0.9780153070633798,
      "eval_runtime": 10.2497,
      "eval_samples_per_second": 767.73,
      "eval_steps_per_second": 3.024,
      "step": 1656
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.9787488458685879,
      "eval_f1": 0.9779865998709163,
      "eval_loss": 0.09993624687194824,
      "eval_precision": 0.9778964351567302,
      "eval_recall": 0.9780767812134997,
      "eval_runtime": 10.4654,
      "eval_samples_per_second": 751.908,
      "eval_steps_per_second": 2.962,
      "step": 1725
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.9788699350659179,
      "eval_f1": 0.9781095368349878,
      "eval_loss": 0.10065959393978119,
      "eval_precision": 0.978019360786724,
      "eval_recall": 0.9781997295137395,
      "eval_runtime": 10.3069,
      "eval_samples_per_second": 763.466,
      "eval_steps_per_second": 3.008,
      "step": 1794
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.9789229115897498,
      "eval_f1": 0.9781592282543133,
      "eval_loss": 0.10217240452766418,
      "eval_precision": 0.97808031838472,
      "eval_recall": 0.9782381508575644,
      "eval_runtime": 10.3246,
      "eval_samples_per_second": 762.158,
      "eval_steps_per_second": 3.003,
      "step": 1863
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9790061604129142,
      "eval_f1": 0.9782287156594198,
      "eval_loss": 0.10301286727190018,
      "eval_precision": 0.9781347715521547,
      "eval_recall": 0.9783226778139792,
      "eval_runtime": 10.6685,
      "eval_samples_per_second": 737.591,
      "eval_steps_per_second": 2.906,
      "step": 1932
    },
    {
      "epoch": 28.98,
      "learning_rate": 1.6814159292035402e-05,
      "loss": 0.0226,
      "step": 2000
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.9789456158142492,
      "eval_f1": 0.9781276533619175,
      "eval_loss": 0.10546565800905228,
      "eval_precision": 0.9780863177791267,
      "eval_recall": 0.9781689924386795,
      "eval_runtime": 10.2106,
      "eval_samples_per_second": 770.668,
      "eval_steps_per_second": 3.036,
      "step": 2001
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.9788775031407511,
      "eval_f1": 0.9781016850177108,
      "eval_loss": 0.10569430887699127,
      "eval_precision": 0.9780190230335438,
      "eval_recall": 0.9781843609762095,
      "eval_runtime": 10.1623,
      "eval_samples_per_second": 774.335,
      "eval_steps_per_second": 3.051,
      "step": 2070
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.9788169585420861,
      "eval_f1": 0.9780400473314586,
      "eval_loss": 0.10669872909784317,
      "eval_precision": 0.9779649036540766,
      "eval_recall": 0.9781152025573246,
      "eval_runtime": 10.228,
      "eval_samples_per_second": 769.357,
      "eval_steps_per_second": 3.031,
      "step": 2139
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.9788850712155842,
      "eval_f1": 0.9781134626983792,
      "eval_loss": 0.10771454125642776,
      "eval_precision": 0.9780195296594217,
      "eval_recall": 0.9782074137825044,
      "eval_runtime": 10.2465,
      "eval_samples_per_second": 767.969,
      "eval_steps_per_second": 3.025,
      "step": 2208
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.9788547989162517,
      "eval_f1": 0.9780702765419577,
      "eval_loss": 0.10846679657697678,
      "eval_precision": 0.9780176719170188,
      "eval_recall": 0.9781228868260896,
      "eval_runtime": 10.262,
      "eval_samples_per_second": 766.809,
      "eval_steps_per_second": 3.021,
      "step": 2277
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.9789153435149167,
      "eval_f1": 0.9781402710760058,
      "eval_loss": 0.10942833125591278,
      "eval_precision": 0.9780500921942225,
      "eval_recall": 0.9782304665887994,
      "eval_runtime": 10.2289,
      "eval_samples_per_second": 769.29,
      "eval_steps_per_second": 3.031,
      "step": 2346
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.9791272496102441,
      "eval_f1": 0.9783318606170041,
      "eval_loss": 0.10954407602548599,
      "eval_precision": 0.9782642100895862,
      "eval_recall": 0.978399520501629,
      "eval_runtime": 10.2671,
      "eval_samples_per_second": 766.432,
      "eval_steps_per_second": 3.019,
      "step": 2415
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.9788775031407511,
      "eval_f1": 0.9780747081173908,
      "eval_loss": 0.11010610312223434,
      "eval_precision": 0.9779958050661893,
      "eval_recall": 0.9781536239011496,
      "eval_runtime": 10.1692,
      "eval_samples_per_second": 773.807,
      "eval_steps_per_second": 3.048,
      "step": 2484
    },
    {
      "epoch": 36.23,
      "learning_rate": 5.752212389380531e-06,
      "loss": 0.0159,
      "step": 2500
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.9791045453857448,
      "eval_f1": 0.9783088094048946,
      "eval_loss": 0.11143232136964798,
      "eval_precision": 0.9782411604714415,
      "eval_recall": 0.9783764676953342,
      "eval_runtime": 10.184,
      "eval_samples_per_second": 772.684,
      "eval_steps_per_second": 3.044,
      "step": 2553
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.9791272496102441,
      "eval_f1": 0.9783286015589884,
      "eval_loss": 0.11111290007829666,
      "eval_precision": 0.9782346478591898,
      "eval_recall": 0.978422573307924,
      "eval_runtime": 10.011,
      "eval_samples_per_second": 786.035,
      "eval_steps_per_second": 3.097,
      "step": 2622
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.9790894092360786,
      "eval_f1": 0.9782975339329141,
      "eval_loss": 0.11137838661670685,
      "eval_precision": 0.978218612905952,
      "eval_recall": 0.9783764676953342,
      "eval_runtime": 10.1814,
      "eval_samples_per_second": 772.882,
      "eval_steps_per_second": 3.045,
      "step": 2691
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.9791272496102441,
      "eval_f1": 0.9783398772157638,
      "eval_loss": 0.1115424633026123,
      "eval_precision": 0.9782571951013384,
      "eval_recall": 0.978422573307924,
      "eval_runtime": 10.2449,
      "eval_samples_per_second": 768.087,
      "eval_steps_per_second": 3.026,
      "step": 2760
    },
    {
      "epoch": 40.0,
      "step": 2760,
      "total_flos": 1.3845487354146643e+17,
      "train_loss": 0.24823836001796998,
      "train_runtime": 2048.5615,
      "train_samples_per_second": 1382.668,
      "train_steps_per_second": 1.347
    }
  ],
  "max_steps": 2760,
  "num_train_epochs": 40,
  "total_flos": 1.3845487354146643e+17,
  "trial_name": null,
  "trial_params": null
}