{
  "best_metric": 0.8422989695943094,
  "best_model_checkpoint": "./output//roberta-large_ipc0_5__5e-5_0.01_0.06_07-20-22_16-12/checkpoint-78000",
  "epoch": 0.14747043196404466,
  "global_step": 86000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5.71588616241119e-07,
      "loss": 0.3883,
      "step": 2000
    },
    {
      "epoch": 0.0,
      "eval_accuracy": 0.5407561450712831,
      "eval_f1": 0.6846357008233311,
      "eval_loss": 0.2273142784833908,
      "eval_roc_auc": 0.7884579701932871,
      "eval_runtime": 4707.9267,
      "eval_samples_per_second": 62.918,
      "step": 2000
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.143177232482238e-06,
      "loss": 0.2054,
      "step": 4000
    },
    {
      "epoch": 0.01,
      "eval_accuracy": 0.6222245478760217,
      "eval_f1": 0.7599527299062029,
      "eval_loss": 0.18135647475719452,
      "eval_roc_auc": 0.8430138570362599,
      "eval_runtime": 4704.9968,
      "eval_samples_per_second": 62.957,
      "step": 4000
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7147658487233569e-06,
      "loss": 0.1754,
      "step": 6000
    },
    {
      "epoch": 0.01,
      "eval_accuracy": 0.6469635026146725,
      "eval_f1": 0.7830456416949749,
      "eval_loss": 0.16404622793197632,
      "eval_roc_auc": 0.858923618616464,
      "eval_runtime": 4707.9088,
      "eval_samples_per_second": 62.918,
      "step": 6000
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.286354464964476e-06,
      "loss": 0.1619,
      "step": 8000
    },
    {
      "epoch": 0.01,
      "eval_accuracy": 0.6548936069652581,
      "eval_f1": 0.7882942239180966,
      "eval_loss": 0.15830162167549133,
      "eval_roc_auc": 0.8607893437318509,
      "eval_runtime": 4704.8509,
      "eval_samples_per_second": 62.959,
      "step": 8000
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.857943081205595e-06,
      "loss": 0.1536,
      "step": 10000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.6733600483435906,
      "eval_f1": 0.8012577006532831,
      "eval_loss": 0.14840050041675568,
      "eval_roc_auc": 0.8665618320388239,
      "eval_runtime": 4704.9917,
      "eval_samples_per_second": 62.957,
      "step": 10000
    },
    {
      "epoch": 0.02,
      "learning_rate": 3.4295316974467138e-06,
      "loss": 0.1479,
      "step": 12000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.6778433087001583,
      "eval_f1": 0.8084837754895617,
      "eval_loss": 0.14521972835063934,
      "eval_roc_auc": 0.8744209349192423,
      "eval_runtime": 4700.2006,
      "eval_samples_per_second": 63.021,
      "step": 12000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.001120313687832e-06,
      "loss": 0.1433,
      "step": 14000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.6842576119211513,
      "eval_f1": 0.8146811536296887,
      "eval_loss": 0.13989576697349548,
      "eval_roc_auc": 0.8784560258941168,
      "eval_runtime": 4713.9153,
      "eval_samples_per_second": 62.838,
      "step": 14000
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.572708929928952e-06,
      "loss": 0.1398,
      "step": 16000
    },
    {
      "epoch": 0.03,
      "eval_accuracy": 0.6886058343151719,
      "eval_f1": 0.8160765004171728,
      "eval_loss": 0.13773857057094574,
      "eval_roc_auc": 0.877459307877016,
      "eval_runtime": 4705.5648,
      "eval_samples_per_second": 62.95,
      "step": 16000
    },
    {
      "epoch": 0.03,
      "learning_rate": 5.144297546170071e-06,
      "loss": 0.1371,
      "step": 18000
    },
    {
      "epoch": 0.03,
      "eval_accuracy": 0.6879238926043084,
      "eval_f1": 0.8203506019576413,
      "eval_loss": 0.13596272468566895,
      "eval_roc_auc": 0.8839289957852754,
      "eval_runtime": 4711.3568,
      "eval_samples_per_second": 62.872,
      "step": 18000
    },
    {
      "epoch": 0.03,
      "learning_rate": 5.71588616241119e-06,
      "loss": 0.1348,
      "step": 20000
    },
    {
      "epoch": 0.03,
      "eval_accuracy": 0.6958573729039577,
      "eval_f1": 0.824050131286724,
      "eval_loss": 0.13410066068172455,
      "eval_roc_auc": 0.8851249727718851,
      "eval_runtime": 4700.4765,
      "eval_samples_per_second": 63.018,
      "step": 20000
    },
    {
      "epoch": 0.04,
      "learning_rate": 6.287474778652308e-06,
      "loss": 0.1337,
      "step": 22000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.6954286273728701,
      "eval_f1": 0.8228504902698741,
      "eval_loss": 0.13327716290950775,
      "eval_roc_auc": 0.8827479207014749,
      "eval_runtime": 4730.8966,
      "eval_samples_per_second": 62.612,
      "step": 22000
    },
    {
      "epoch": 0.04,
      "learning_rate": 6.8590633948934276e-06,
      "loss": 0.1331,
      "step": 24000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.6971469854462836,
      "eval_f1": 0.8237129350972603,
      "eval_loss": 0.13190585374832153,
      "eval_roc_auc": 0.883230389109756,
      "eval_runtime": 4692.6605,
      "eval_samples_per_second": 63.123,
      "step": 24000
    },
    {
      "epoch": 0.04,
      "learning_rate": 7.4306520111345465e-06,
      "loss": 0.1309,
      "step": 26000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.7027004216560381,
      "eval_f1": 0.8271447851497717,
      "eval_loss": 0.13012564182281494,
      "eval_roc_auc": 0.8860201435878601,
      "eval_runtime": 4762.2583,
      "eval_samples_per_second": 62.2,
      "step": 26000
    },
    {
      "epoch": 0.05,
      "learning_rate": 8.002240627375665e-06,
      "loss": 0.1298,
      "step": 28000
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.7013432901324385,
      "eval_f1": 0.8245749680737388,
      "eval_loss": 0.12999016046524048,
      "eval_roc_auc": 0.8808593693278481,
      "eval_runtime": 4730.1652,
      "eval_samples_per_second": 62.622,
      "step": 28000
    },
    {
      "epoch": 0.05,
      "learning_rate": 8.573829243616784e-06,
      "loss": 0.1284,
      "step": 30000
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.7064072137279592,
      "eval_f1": 0.8300594309848033,
      "eval_loss": 0.12789227068424225,
      "eval_roc_auc": 0.887462059567475,
      "eval_runtime": 4709.67,
      "eval_samples_per_second": 62.895,
      "step": 30000
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.145417859857904e-06,
      "loss": 0.1294,
      "step": 32000
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.7096987640650478,
      "eval_f1": 0.8326160218501911,
      "eval_loss": 0.12640197575092316,
      "eval_roc_auc": 0.8885914760719662,
      "eval_runtime": 4696.3704,
      "eval_samples_per_second": 63.073,
      "step": 32000
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.717006476099022e-06,
      "loss": 0.1281,
      "step": 34000
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.7070351402538039,
      "eval_f1": 0.8319073213351634,
      "eval_loss": 0.12621602416038513,
      "eval_roc_auc": 0.8896376613488131,
      "eval_runtime": 4698.8191,
      "eval_samples_per_second": 63.04,
      "step": 34000
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.0288595092340142e-05,
      "loss": 0.1266,
      "step": 36000
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.7081255718013727,
      "eval_f1": 0.8335308031653823,
      "eval_loss": 0.12684929370880127,
      "eval_roc_auc": 0.8919053253578205,
      "eval_runtime": 4698.6474,
      "eval_samples_per_second": 63.042,
      "step": 36000
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.086018370858126e-05,
      "loss": 0.1264,
      "step": 38000
    },
    {
      "epoch": 0.07,
      "eval_accuracy": 0.7080074135841439,
      "eval_f1": 0.8339326805263629,
      "eval_loss": 0.12483926117420197,
      "eval_roc_auc": 0.8912626509356554,
      "eval_runtime": 4703.9737,
      "eval_samples_per_second": 62.971,
      "step": 38000
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.143177232482238e-05,
      "loss": 0.1252,
      "step": 40000
    },
    {
      "epoch": 0.07,
      "eval_accuracy": 0.7134258118313511,
      "eval_f1": 0.8343130408901507,
      "eval_loss": 0.12493231147527695,
      "eval_roc_auc": 0.8898269175762451,
      "eval_runtime": 4715.1654,
      "eval_samples_per_second": 62.821,
      "step": 40000
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.20033609410635e-05,
      "loss": 0.1253,
      "step": 42000
    },
    {
      "epoch": 0.07,
      "eval_accuracy": 0.7114036183422064,
      "eval_f1": 0.8363716860622692,
      "eval_loss": 0.12403804063796997,
      "eval_roc_auc": 0.8935152243226613,
      "eval_runtime": 4721.8512,
      "eval_samples_per_second": 62.732,
      "step": 42000
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.2574949557304616e-05,
      "loss": 0.1241,
      "step": 44000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.7098405539257223,
      "eval_f1": 0.8373545897529787,
      "eval_loss": 0.12371223419904709,
      "eval_roc_auc": 0.8963795632569544,
      "eval_runtime": 4728.9925,
      "eval_samples_per_second": 62.638,
      "step": 44000
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.3146538173545735e-05,
      "loss": 0.1238,
      "step": 46000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.7124602903991385,
      "eval_f1": 0.8370424197096473,
      "eval_loss": 0.12373730540275574,
      "eval_roc_auc": 0.8944393753665816,
      "eval_runtime": 4701.3381,
      "eval_samples_per_second": 63.006,
      "step": 46000
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.3718126789786855e-05,
      "loss": 0.1238,
      "step": 48000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.7092024995526868,
      "eval_f1": 0.8358656658561511,
      "eval_loss": 0.125064879655838,
      "eval_roc_auc": 0.895248662380464,
      "eval_runtime": 4691.6911,
      "eval_samples_per_second": 63.136,
      "step": 48000
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.4289715406027973e-05,
      "loss": 0.1228,
      "step": 50000
    },
    {
      "epoch": 0.09,
      "eval_accuracy": 0.7139119484965211,
      "eval_f1": 0.8380575940983299,
      "eval_loss": 0.12305936962366104,
      "eval_roc_auc": 0.8943202372027479,
      "eval_runtime": 4708.5133,
      "eval_samples_per_second": 62.91,
      "step": 50000
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.4861304022269093e-05,
      "loss": 0.123,
      "step": 52000
    },
    {
      "epoch": 0.09,
      "eval_accuracy": 0.7159408938837931,
      "eval_f1": 0.8405024216747294,
      "eval_loss": 0.12146713584661484,
      "eval_roc_auc": 0.8976514718844044,
      "eval_runtime": 4705.5754,
      "eval_samples_per_second": 62.949,
      "step": 52000
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.5432892638510214e-05,
      "loss": 0.123,
      "step": 54000
    },
    {
      "epoch": 0.09,
      "eval_accuracy": 0.7114069942912701,
      "eval_f1": 0.8382327921361191,
      "eval_loss": 0.12516580522060394,
      "eval_roc_auc": 0.8973778420243191,
      "eval_runtime": 4690.4939,
      "eval_samples_per_second": 63.152,
      "step": 54000
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.600448125475133e-05,
      "loss": 0.1222,
      "step": 56000
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.7157248331437175,
      "eval_f1": 0.8389102808480002,
      "eval_loss": 0.1224837601184845,
      "eval_roc_auc": 0.8953561781764309,
      "eval_runtime": 4699.8836,
      "eval_samples_per_second": 63.026,
      "step": 56000
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.657606987099245e-05,
      "loss": 0.1222,
      "step": 58000
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.7158564951572011,
      "eval_f1": 0.8399927583173271,
      "eval_loss": 0.12127628922462463,
      "eval_roc_auc": 0.8965841568131228,
      "eval_runtime": 4698.6002,
      "eval_samples_per_second": 63.043,
      "step": 58000
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.714765848723357e-05,
      "loss": 0.1219,
      "step": 60000
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.7170279494822982,
      "eval_f1": 0.8398695690490878,
      "eval_loss": 0.12054365128278732,
      "eval_roc_auc": 0.8949977373242076,
      "eval_runtime": 4766.0245,
      "eval_samples_per_second": 62.151,
      "step": 60000
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.7719247103474687e-05,
      "loss": 0.1217,
      "step": 62000
    },
    {
      "epoch": 0.11,
      "eval_accuracy": 0.7173587924905389,
      "eval_f1": 0.8384044315001998,
      "eval_loss": 0.12145062536001205,
      "eval_roc_auc": 0.8928348976264423,
      "eval_runtime": 4726.8677,
      "eval_samples_per_second": 62.666,
      "step": 62000
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.8290835719715808e-05,
      "loss": 0.1212,
      "step": 64000
    },
    {
      "epoch": 0.11,
      "eval_accuracy": 0.7183074341774331,
      "eval_f1": 0.8391874180865005,
      "eval_loss": 0.12094008922576904,
      "eval_roc_auc": 0.8939367915265275,
      "eval_runtime": 4721.7671,
      "eval_samples_per_second": 62.734,
      "step": 64000
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.8862424335956926e-05,
      "loss": 0.1217,
      "step": 66000
    },
    {
      "epoch": 0.11,
      "eval_accuracy": 0.7173655443886663,
      "eval_f1": 0.8390122732329008,
      "eval_loss": 0.12199072539806366,
      "eval_roc_auc": 0.8938961477593685,
      "eval_runtime": 4689.9026,
      "eval_samples_per_second": 63.16,
      "step": 66000
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9434012952198044e-05,
      "loss": 0.1211,
      "step": 68000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.7192932113040278,
      "eval_f1": 0.8406356467928897,
      "eval_loss": 0.12074108421802521,
      "eval_roc_auc": 0.8959624591328685,
      "eval_runtime": 4689.6506,
      "eval_samples_per_second": 63.163,
      "step": 68000
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.0005601568439166e-05,
      "loss": 0.1218,
      "step": 70000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.718013726608893,
      "eval_f1": 0.8394379496338936,
      "eval_loss": 0.1212056577205658,
      "eval_roc_auc": 0.8946202640469076,
      "eval_runtime": 4729.1038,
      "eval_samples_per_second": 62.636,
      "step": 70000
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.0577190184680284e-05,
      "loss": 0.1214,
      "step": 72000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.7165721963587013,
      "eval_f1": 0.8417356069559346,
      "eval_loss": 0.12099753320217133,
      "eval_roc_auc": 0.8992088434208416,
      "eval_runtime": 4735.2667,
      "eval_samples_per_second": 62.555,
      "step": 72000
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.11487788009214e-05,
      "loss": 0.1208,
      "step": 74000
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.7096379969819016,
      "eval_f1": 0.8359202990403864,
      "eval_loss": 0.12574784457683563,
      "eval_roc_auc": 0.8958735028365457,
      "eval_runtime": 4708.3575,
      "eval_samples_per_second": 62.912,
      "step": 74000
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.172036741716252e-05,
      "loss": 0.1202,
      "step": 76000
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.7170211975841708,
      "eval_f1": 0.8380538350579405,
      "eval_loss": 0.12220340967178345,
      "eval_roc_auc": 0.8937227496461092,
      "eval_runtime": 4705.5508,
      "eval_samples_per_second": 62.95,
      "step": 76000
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.229195603340364e-05,
      "loss": 0.1211,
      "step": 78000
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.7159375179347294,
      "eval_f1": 0.8422989695943094,
      "eval_loss": 0.12021646648645401,
      "eval_roc_auc": 0.8995460792466595,
      "eval_runtime": 4706.4093,
      "eval_samples_per_second": 62.938,
      "step": 78000
    },
    {
      "epoch": 0.14,
      "learning_rate": 2.286354464964476e-05,
      "loss": 0.1212,
      "step": 80000
    },
    {
      "epoch": 0.14,
      "eval_accuracy": 0.7161333229804229,
      "eval_f1": 0.8413767292128633,
      "eval_loss": 0.12168838083744049,
      "eval_roc_auc": 0.8989878842731752,
      "eval_runtime": 4758.053,
      "eval_samples_per_second": 62.255,
      "step": 80000
    },
    {
      "epoch": 0.14,
      "learning_rate": 2.3435133265885877e-05,
      "loss": 0.1211,
      "step": 82000
    },
    {
      "epoch": 0.14,
      "eval_accuracy": 0.7183985848021525,
      "eval_f1": 0.8372888052842701,
      "eval_loss": 0.12127470970153809,
      "eval_roc_auc": 0.891376589798138,
      "eval_runtime": 4718.8962,
      "eval_samples_per_second": 62.772,
      "step": 82000
    },
    {
      "epoch": 0.14,
      "learning_rate": 2.4006721882127e-05,
      "loss": 0.1208,
      "step": 84000
    },
    {
      "epoch": 0.14,
      "eval_accuracy": 0.7164675419377272,
      "eval_f1": 0.8410196283399609,
      "eval_loss": 0.12072896212339401,
      "eval_roc_auc": 0.8975580711258158,
      "eval_runtime": 4700.4599,
      "eval_samples_per_second": 63.018,
      "step": 84000
    },
    {
      "epoch": 0.15,
      "learning_rate": 2.4578310498368117e-05,
      "loss": 0.1207,
      "step": 86000
    },
    {
      "epoch": 0.15,
      "eval_accuracy": 0.7182635468396052,
      "eval_f1": 0.8404677971481037,
      "eval_loss": 0.12209167331457138,
      "eval_roc_auc": 0.8963189679938305,
      "eval_runtime": 4690.2466,
      "eval_samples_per_second": 63.155,
      "step": 86000
    }
  ],
  "max_steps": 2915835,
  "num_train_epochs": 5,
  "total_flos": 1.39736361467904e+17,
  "trial_name": null,
  "trial_params": null
}