{
  "best_metric": 0.7672541719320296,
  "best_model_checkpoint": "./CARES/checkpoints/roberta-stratified/run-6/checkpoint-3550",
  "epoch": 50.0,
  "global_step": 3550,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0,
      "eval_loss": 0.24808131158351898,
      "eval_macro_f1": 0.05481647030175482,
      "eval_macro_precision": 0.17800453514739228,
      "eval_macro_recall": 0.03330847588023425,
      "eval_micro_f1": 0.13594611145131658,
      "eval_micro_precision": 0.9568965517241379,
      "eval_micro_recall": 0.07317073170731707,
      "eval_runtime": 2.8186,
      "eval_samples_per_second": 342.724,
      "eval_steps_per_second": 21.642,
      "step": 142
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.16814221441745758,
      "eval_macro_f1": 0.37346208328620767,
      "eval_macro_precision": 0.4376472360907343,
      "eval_macro_recall": 0.3342883491414155,
      "eval_micro_f1": 0.7248062015503876,
      "eval_micro_precision": 0.8795860771401693,
      "eval_micro_recall": 0.6163480553724456,
      "eval_runtime": 2.8192,
      "eval_samples_per_second": 342.653,
      "eval_steps_per_second": 21.638,
      "step": 284
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.13150478899478912,
      "eval_macro_f1": 0.4319986220599845,
      "eval_macro_precision": 0.6389953058630822,
      "eval_macro_recall": 0.37598442828818834,
      "eval_micro_f1": 0.7641723356009071,
      "eval_micro_precision": 0.895482728077945,
      "eval_micro_recall": 0.6664469347396177,
      "eval_runtime": 2.8191,
      "eval_samples_per_second": 342.658,
      "eval_steps_per_second": 21.638,
      "step": 426
    },
    {
      "epoch": 7.04,
      "learning_rate": 2.8662355855330125e-05,
      "loss": 0.2172,
      "step": 500
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.1121998205780983,
      "eval_macro_f1": 0.5386615940425785,
      "eval_macro_precision": 0.7631201629660762,
      "eval_macro_recall": 0.46714612883136086,
      "eval_micro_f1": 0.8077485380116959,
      "eval_micro_precision": 0.9064807219031994,
      "eval_micro_recall": 0.7284113381674358,
      "eval_runtime": 2.8196,
      "eval_samples_per_second": 342.598,
      "eval_steps_per_second": 21.634,
      "step": 568
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.10608664900064468,
      "eval_macro_f1": 0.5851323545796894,
      "eval_macro_precision": 0.7572045278273729,
      "eval_macro_recall": 0.5132727928856573,
      "eval_micro_f1": 0.8146802325581396,
      "eval_micro_precision": 0.9076923076923077,
      "eval_micro_recall": 0.7389584706657878,
      "eval_runtime": 2.8204,
      "eval_samples_per_second": 342.507,
      "eval_steps_per_second": 21.628,
      "step": 710
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.100788913667202,
      "eval_macro_f1": 0.628825318779795,
      "eval_macro_precision": 0.750817062947988,
      "eval_macro_recall": 0.5578116563013593,
      "eval_micro_f1": 0.8245363766048501,
      "eval_micro_precision": 0.8982128982128982,
      "eval_micro_recall": 0.7620303230059328,
      "eval_runtime": 2.8211,
      "eval_samples_per_second": 342.425,
      "eval_steps_per_second": 21.623,
      "step": 852
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.09952697157859802,
      "eval_macro_f1": 0.648367052483795,
      "eval_macro_precision": 0.7408904755242313,
      "eval_macro_recall": 0.5923039028415692,
      "eval_micro_f1": 0.8379418970948547,
      "eval_micro_precision": 0.8932835820895523,
      "eval_micro_recall": 0.7890573500329597,
      "eval_runtime": 2.8203,
      "eval_samples_per_second": 342.519,
      "eval_steps_per_second": 21.629,
      "step": 994
    },
    {
      "epoch": 14.08,
      "learning_rate": 2.6490965260229357e-05,
      "loss": 0.0388,
      "step": 1000
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.10003374516963959,
      "eval_macro_f1": 0.7100297932545989,
      "eval_macro_precision": 0.8731926397728601,
      "eval_macro_recall": 0.6387457691830838,
      "eval_micro_f1": 0.837847344354555,
      "eval_micro_precision": 0.8981900452488688,
      "eval_micro_recall": 0.7851021753460777,
      "eval_runtime": 2.8204,
      "eval_samples_per_second": 342.505,
      "eval_steps_per_second": 21.628,
      "step": 1136
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.10116878896951675,
      "eval_macro_f1": 0.7102542897490418,
      "eval_macro_precision": 0.8708979085725551,
      "eval_macro_recall": 0.6278706438272531,
      "eval_micro_f1": 0.8369795342272407,
      "eval_micro_precision": 0.9005315110098709,
      "eval_micro_recall": 0.7818061964403428,
      "eval_runtime": 2.821,
      "eval_samples_per_second": 342.431,
      "eval_steps_per_second": 21.624,
      "step": 1278
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.10277832299470901,
      "eval_macro_f1": 0.7155479835258449,
      "eval_macro_precision": 0.8815899776220457,
      "eval_macro_recall": 0.635461059766739,
      "eval_micro_f1": 0.8361702127659574,
      "eval_micro_precision": 0.9048349961627015,
      "eval_micro_recall": 0.7771918259723137,
      "eval_runtime": 2.8201,
      "eval_samples_per_second": 342.543,
      "eval_steps_per_second": 21.631,
      "step": 1420
    },
    {
      "epoch": 21.13,
      "learning_rate": 2.431957466512859e-05,
      "loss": 0.0147,
      "step": 1500
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.10351855307817459,
      "eval_macro_f1": 0.7203873361325647,
      "eval_macro_precision": 0.87817690487536,
      "eval_macro_recall": 0.6443209986334517,
      "eval_micro_f1": 0.837389770723104,
      "eval_micro_precision": 0.9006069802731411,
      "eval_micro_recall": 0.7824653922214898,
      "eval_runtime": 2.8203,
      "eval_samples_per_second": 342.52,
      "eval_steps_per_second": 21.629,
      "step": 1562
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.10557578504085541,
      "eval_macro_f1": 0.7298035823714943,
      "eval_macro_precision": 0.8804692132123697,
      "eval_macro_recall": 0.6544575680820387,
      "eval_micro_f1": 0.8399153737658673,
      "eval_micro_precision": 0.9029567854435178,
      "eval_micro_recall": 0.7851021753460777,
      "eval_runtime": 2.8192,
      "eval_samples_per_second": 342.647,
      "eval_steps_per_second": 21.637,
      "step": 1704
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.10698197782039642,
      "eval_macro_f1": 0.719671126038702,
      "eval_macro_precision": 0.8799359996888543,
      "eval_macro_recall": 0.6422398711062759,
      "eval_micro_f1": 0.8375706214689265,
      "eval_micro_precision": 0.9019011406844106,
      "eval_micro_recall": 0.7818061964403428,
      "eval_runtime": 2.8198,
      "eval_samples_per_second": 342.58,
      "eval_steps_per_second": 21.633,
      "step": 1846
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.10863872617483139,
      "eval_macro_f1": 0.7318024399797272,
      "eval_macro_precision": 0.8781101314061108,
      "eval_macro_recall": 0.6548756753680312,
      "eval_micro_f1": 0.8389143461402891,
      "eval_micro_precision": 0.9015151515151515,
      "eval_micro_recall": 0.7844429795649308,
      "eval_runtime": 2.8198,
      "eval_samples_per_second": 342.583,
      "eval_steps_per_second": 21.633,
      "step": 1988
    },
    {
      "epoch": 28.17,
      "learning_rate": 2.214818407002782e-05,
      "loss": 0.0082,
      "step": 2000
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.1103406548500061,
      "eval_macro_f1": 0.728656455732704,
      "eval_macro_precision": 0.8784457053881349,
      "eval_macro_recall": 0.6552278977528909,
      "eval_micro_f1": 0.8400702987697715,
      "eval_micro_precision": 0.8998493975903614,
      "eval_micro_recall": 0.7877389584706658,
      "eval_runtime": 2.8227,
      "eval_samples_per_second": 342.231,
      "eval_steps_per_second": 21.611,
      "step": 2130
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.11085448414087296,
      "eval_macro_f1": 0.7344651790603305,
      "eval_macro_precision": 0.8647882525531505,
      "eval_macro_recall": 0.6579821769650391,
      "eval_micro_f1": 0.8402116402116402,
      "eval_micro_precision": 0.9036418816388467,
      "eval_micro_recall": 0.7851021753460777,
      "eval_runtime": 2.8204,
      "eval_samples_per_second": 342.508,
      "eval_steps_per_second": 21.628,
      "step": 2272
    },
    {
      "epoch": 34.0,
      "eval_loss": 0.11268670856952667,
      "eval_macro_f1": 0.7371526225457369,
      "eval_macro_precision": 0.8672638823695995,
      "eval_macro_recall": 0.6591834104386446,
      "eval_micro_f1": 0.8397323001056711,
      "eval_micro_precision": 0.9016641452344932,
      "eval_micro_recall": 0.7857613711272248,
      "eval_runtime": 2.8203,
      "eval_samples_per_second": 342.515,
      "eval_steps_per_second": 21.629,
      "step": 2414
    },
    {
      "epoch": 35.21,
      "learning_rate": 1.9976793474927056e-05,
      "loss": 0.0056,
      "step": 2500
    },
    {
      "epoch": 36.0,
      "eval_loss": 0.11449939757585526,
      "eval_macro_f1": 0.7435877211427984,
      "eval_macro_precision": 0.8803719443456752,
      "eval_macro_recall": 0.6671207777218493,
      "eval_micro_f1": 0.8435852372583479,
      "eval_micro_precision": 0.9036144578313253,
      "eval_micro_recall": 0.7910349373764007,
      "eval_runtime": 2.8219,
      "eval_samples_per_second": 342.324,
      "eval_steps_per_second": 21.617,
      "step": 2556
    },
    {
      "epoch": 38.0,
      "eval_loss": 0.1153542622923851,
      "eval_macro_f1": 0.7450728657264144,
      "eval_macro_precision": 0.879277762192774,
      "eval_macro_recall": 0.6681699293595025,
      "eval_micro_f1": 0.8419570573741639,
      "eval_micro_precision": 0.9033232628398792,
      "eval_micro_recall": 0.7883981542518128,
      "eval_runtime": 2.8191,
      "eval_samples_per_second": 342.661,
      "eval_steps_per_second": 21.638,
      "step": 2698
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.11756419390439987,
      "eval_macro_f1": 0.744466331732378,
      "eval_macro_precision": 0.8796020530253235,
      "eval_macro_recall": 0.6677481100215579,
      "eval_micro_f1": 0.8419570573741639,
      "eval_micro_precision": 0.9033232628398792,
      "eval_micro_recall": 0.7883981542518128,
      "eval_runtime": 2.8181,
      "eval_samples_per_second": 342.785,
      "eval_steps_per_second": 21.646,
      "step": 2840
    },
    {
      "epoch": 42.0,
      "eval_loss": 0.11846820265054703,
      "eval_macro_f1": 0.7434279482403103,
      "eval_macro_precision": 0.8813168271357439,
      "eval_macro_recall": 0.6651456053129234,
      "eval_micro_f1": 0.8427717200140696,
      "eval_micro_precision": 0.9034690799396682,
      "eval_micro_recall": 0.7897165458141068,
      "eval_runtime": 2.8203,
      "eval_samples_per_second": 342.52,
      "eval_steps_per_second": 21.629,
      "step": 2982
    },
    {
      "epoch": 42.25,
      "learning_rate": 1.7805402879826288e-05,
      "loss": 0.004,
      "step": 3000
    },
    {
      "epoch": 44.0,
      "eval_loss": 0.11996057629585266,
      "eval_macro_f1": 0.7394753977857482,
      "eval_macro_precision": 0.8783969916238008,
      "eval_macro_recall": 0.6650662737686901,
      "eval_micro_f1": 0.8427717200140696,
      "eval_micro_precision": 0.9034690799396682,
      "eval_micro_recall": 0.7897165458141068,
      "eval_runtime": 2.82,
      "eval_samples_per_second": 342.552,
      "eval_steps_per_second": 21.631,
      "step": 3124
    },
    {
      "epoch": 46.0,
      "eval_loss": 0.12046220153570175,
      "eval_macro_f1": 0.7423287819151947,
      "eval_macro_precision": 0.8810998402784964,
      "eval_macro_recall": 0.6639465302563645,
      "eval_micro_f1": 0.8432546671363156,
      "eval_micro_precision": 0.905446293494705,
      "eval_micro_recall": 0.7890573500329597,
      "eval_runtime": 2.819,
      "eval_samples_per_second": 342.677,
      "eval_steps_per_second": 21.639,
      "step": 3266
    },
    {
      "epoch": 48.0,
      "eval_loss": 0.12299305200576782,
      "eval_macro_f1": 0.7584256567760465,
      "eval_macro_precision": 0.9421857420436691,
      "eval_macro_recall": 0.6745195368122752,
      "eval_micro_f1": 0.8419570573741639,
      "eval_micro_precision": 0.9033232628398792,
      "eval_micro_recall": 0.7883981542518128,
      "eval_runtime": 2.8184,
      "eval_samples_per_second": 342.744,
      "eval_steps_per_second": 21.643,
      "step": 3408
    },
    {
      "epoch": 49.3,
      "learning_rate": 1.563401228472552e-05,
      "loss": 0.0031,
      "step": 3500
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.12287621200084686,
      "eval_macro_f1": 0.7672541719320296,
      "eval_macro_precision": 0.9303028445349943,
      "eval_macro_recall": 0.6844658294965831,
      "eval_micro_f1": 0.8449122807017543,
      "eval_micro_precision": 0.9032258064516129,
      "eval_micro_recall": 0.7936717205009888,
      "eval_runtime": 2.8187,
      "eval_samples_per_second": 342.711,
      "eval_steps_per_second": 21.641,
      "step": 3550
    }
  ],
  "max_steps": 7100,
  "num_train_epochs": 100,
  "total_flos": 3.007255197834797e+16,
  "trial_name": null,
  "trial_params": {
    "adam_epsilon": 6.447418463180699e-08,
    "learning_rate": 3.0290898801655698e-05,
    "per_device_eval_batch_size": 16,
    "per_device_train_batch_size": 32,
    "seed": 320,
    "warmup_steps": 125,
    "weight_decay": 4.5126980713116176e-08
  }
}