{
  "best_global_step": 22629,
  "best_metric": 0.7126982699264178,
  "best_model_checkpoint": "../output/checkpoint-22629",
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 23820,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 0.028375638648867607,
      "learning_rate": 1.9581024349286314e-05,
      "loss": 0.1387,
      "step": 500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 0.016328655183315277,
      "learning_rate": 1.9161209068010078e-05,
      "loss": 0.0525,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_Hamming Loss": 0.01128586278670676,
      "eval_loss": 0.04800482094287872,
      "eval_macro-F1": 0.0,
      "eval_micro-F1": 0.0,
      "eval_micro-Precision": 0.0,
      "eval_micro-ROC-AUC": 0.894420189398385,
      "eval_micro-Recall": 0.0,
      "eval_runtime": 56.2294,
      "eval_samples-Precision": 0.0,
      "eval_samples-ROC-AUC": 0.9075031413018584,
      "eval_samples-Recall": 0.0,
      "eval_samples_F1": 0.0,
      "eval_samples_per_second": 677.457,
      "eval_steps_per_second": 2.65,
      "eval_weighted-F1": 0.0,
      "eval_weighted-ROC-AUC": 0.6750014123328221,
      "step": 1191
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 0.022175736725330353,
      "learning_rate": 1.8741393786733838e-05,
      "loss": 0.0478,
      "step": 1500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 0.019683972001075745,
      "learning_rate": 1.83215785054576e-05,
      "loss": 0.0413,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_Hamming Loss": 0.009107120094851965,
      "eval_loss": 0.0341496467590332,
      "eval_macro-F1": 0.03927605624721043,
      "eval_micro-F1": 0.3742492143481414,
      "eval_micro-Precision": 0.8333402343588258,
      "eval_micro-ROC-AUC": 0.95328941167146,
      "eval_micro-Recall": 0.2413102646187787,
      "eval_runtime": 56.4959,
      "eval_samples-Precision": 0.4733787485014745,
      "eval_samples-ROC-AUC": 0.9533603293317763,
      "eval_samples-Recall": 0.3310842278806518,
      "eval_samples_F1": 0.36697773688171714,
      "eval_samples_per_second": 674.262,
      "eval_steps_per_second": 2.637,
      "eval_weighted-F1": 0.28759108086329915,
      "eval_weighted-ROC-AUC": 0.9137363969306936,
      "step": 2382
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 0.021357489749789238,
      "learning_rate": 1.790176322418136e-05,
      "loss": 0.0359,
      "step": 2500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 0.02431364730000496,
      "learning_rate": 1.7481947942905125e-05,
      "loss": 0.0322,
      "step": 3000
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 0.026525776833295822,
      "learning_rate": 1.7062132661628885e-05,
      "loss": 0.0299,
      "step": 3500
    },
    {
      "epoch": 3.0,
      "eval_Hamming Loss": 0.008002254926291353,
      "eval_loss": 0.027838103473186493,
      "eval_macro-F1": 0.0850764115643415,
      "eval_micro-F1": 0.5156952508865176,
      "eval_micro-Precision": 0.8134818106655642,
      "eval_micro-ROC-AUC": 0.9701655203023096,
      "eval_micro-Recall": 0.37750440631631954,
      "eval_runtime": 56.3401,
      "eval_samples-Precision": 0.6690008663008952,
      "eval_samples-ROC-AUC": 0.9693675900344925,
      "eval_samples-Recall": 0.4936407116876433,
      "eval_samples_F1": 0.5398431979370828,
      "eval_samples_per_second": 676.126,
      "eval_steps_per_second": 2.645,
      "eval_weighted-F1": 0.4436933206273856,
      "eval_weighted-ROC-AUC": 0.9440048360290951,
      "step": 3573
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 0.029091596603393555,
      "learning_rate": 1.6642317380352645e-05,
      "loss": 0.0276,
      "step": 4000
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 0.031437281519174576,
      "learning_rate": 1.622250209907641e-05,
      "loss": 0.0263,
      "step": 4500
    },
    {
      "epoch": 4.0,
      "eval_Hamming Loss": 0.007530945020339533,
      "eval_loss": 0.024892881512641907,
      "eval_macro-F1": 0.12433102464637438,
      "eval_micro-F1": 0.5772899893665502,
      "eval_micro-Precision": 0.7875126924591251,
      "eval_micro-ROC-AUC": 0.9778129934458253,
      "eval_micro-Recall": 0.4556550723595075,
      "eval_runtime": 57.5998,
      "eval_samples-Precision": 0.7262085397517104,
      "eval_samples-ROC-AUC": 0.9769735188597375,
      "eval_samples-Recall": 0.5706880912223564,
      "eval_samples_F1": 0.6109341836523597,
      "eval_samples_per_second": 661.339,
      "eval_steps_per_second": 2.587,
      "eval_weighted-F1": 0.5154308844020796,
      "eval_weighted-ROC-AUC": 0.9549957691180601,
      "step": 4764
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 0.03555045649409294,
      "learning_rate": 1.580268681780017e-05,
      "loss": 0.025,
      "step": 5000
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 0.03600688651204109,
      "learning_rate": 1.5382871536523932e-05,
      "loss": 0.0238,
      "step": 5500
    },
    {
      "epoch": 5.0,
      "eval_Hamming Loss": 0.0072420427380520975,
      "eval_loss": 0.023193439468741417,
      "eval_macro-F1": 0.16691609522798895,
      "eval_micro-F1": 0.6110000654160095,
      "eval_micro-Precision": 0.7757946029753775,
      "eval_micro-ROC-AUC": 0.9815662800440879,
      "eval_micro-Recall": 0.5039506972171265,
      "eval_runtime": 57.0641,
      "eval_samples-Precision": 0.7495485359764886,
      "eval_samples-ROC-AUC": 0.9811746136558583,
      "eval_samples-Recall": 0.6128593646762802,
      "eval_samples_F1": 0.6466909749113342,
      "eval_samples_per_second": 667.548,
      "eval_steps_per_second": 2.611,
      "eval_weighted-F1": 0.5637611718614859,
      "eval_weighted-ROC-AUC": 0.9602035986169475,
      "step": 5955
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 0.030450329184532166,
      "learning_rate": 1.4963056255247692e-05,
      "loss": 0.0232,
      "step": 6000
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 0.03672802075743675,
      "learning_rate": 1.4543240973971454e-05,
      "loss": 0.022,
      "step": 6500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 0.038189828395843506,
      "learning_rate": 1.4123425692695216e-05,
      "loss": 0.0216,
      "step": 7000
    },
    {
      "epoch": 6.0,
      "eval_Hamming Loss": 0.007131894514266631,
      "eval_loss": 0.022080443799495697,
      "eval_macro-F1": 0.1847520859923191,
      "eval_micro-F1": 0.6218258267739135,
      "eval_micro-Precision": 0.7742655993138446,
      "eval_micro-ROC-AUC": 0.9836965904784507,
      "eval_micro-Recall": 0.519537666510797,
      "eval_runtime": 56.5357,
      "eval_samples-Precision": 0.7550899031984796,
      "eval_samples-ROC-AUC": 0.9835234555861406,
      "eval_samples-Recall": 0.6243584834313252,
      "eval_samples_F1": 0.6567380367195391,
      "eval_samples_per_second": 673.787,
      "eval_steps_per_second": 2.636,
      "eval_weighted-F1": 0.5796689384762133,
      "eval_weighted-ROC-AUC": 0.9631894848329642,
      "step": 7146
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 0.041865091770887375,
      "learning_rate": 1.3703610411418977e-05,
      "loss": 0.0208,
      "step": 7500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 0.036912206560373306,
      "learning_rate": 1.3283795130142737e-05,
      "loss": 0.0204,
      "step": 8000
    },
    {
      "epoch": 7.0,
      "eval_Hamming Loss": 0.006943126980875075,
      "eval_loss": 0.021343262866139412,
      "eval_macro-F1": 0.21623803173319953,
      "eval_micro-F1": 0.64556105109005,
      "eval_micro-Precision": 0.7615097537523834,
      "eval_micro-ROC-AUC": 0.9852894966170272,
      "eval_micro-Recall": 0.5602556262964162,
      "eval_runtime": 56.5298,
      "eval_samples-Precision": 0.7660912057750061,
      "eval_samples-ROC-AUC": 0.9854023086128085,
      "eval_samples-Recall": 0.6604805442421895,
      "eval_samples_F1": 0.6840440071687744,
      "eval_samples_per_second": 673.857,
      "eval_steps_per_second": 2.636,
      "eval_weighted-F1": 0.6097152841377028,
      "eval_weighted-ROC-AUC": 0.9655545290298909,
      "step": 8337
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 0.0445188544690609,
      "learning_rate": 1.28639798488665e-05,
      "loss": 0.02,
      "step": 8500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 0.0462997630238533,
      "learning_rate": 1.2444164567590263e-05,
      "loss": 0.0192,
      "step": 9000
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 0.04490142688155174,
      "learning_rate": 1.2024349286314023e-05,
      "loss": 0.0193,
      "step": 9500
    },
    {
      "epoch": 8.0,
      "eval_Hamming Loss": 0.006875333049528,
      "eval_loss": 0.020876074209809303,
      "eval_macro-F1": 0.2250253559054493,
      "eval_micro-F1": 0.6473776624169784,
      "eval_micro-Precision": 0.7685462874468576,
      "eval_micro-ROC-AUC": 0.986105483089175,
      "eval_micro-Recall": 0.5592124983513782,
      "eval_runtime": 56.4221,
      "eval_samples-Precision": 0.7713289197407014,
      "eval_samples-ROC-AUC": 0.9862885403293905,
      "eval_samples-Recall": 0.6614123512440093,
      "eval_samples_F1": 0.6862704472864863,
      "eval_samples_per_second": 675.143,
      "eval_steps_per_second": 2.641,
      "eval_weighted-F1": 0.614858261261732,
      "eval_weighted-ROC-AUC": 0.9667338002054853,
      "step": 9528
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 0.0457521490752697,
      "learning_rate": 1.1604534005037784e-05,
      "loss": 0.0184,
      "step": 10000
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 0.044266458600759506,
      "learning_rate": 1.1184718723761545e-05,
      "loss": 0.0183,
      "step": 10500
    },
    {
      "epoch": 9.0,
      "eval_Hamming Loss": 0.006861936643932471,
      "eval_loss": 0.020720839500427246,
      "eval_macro-F1": 0.2427473530774573,
      "eval_micro-F1": 0.6579472789574509,
      "eval_micro-Precision": 0.7520701938349088,
      "eval_micro-ROC-AUC": 0.9866980845480343,
      "eval_micro-Recall": 0.5847631380166181,
      "eval_runtime": 56.8787,
      "eval_samples-Precision": 0.767462494671562,
      "eval_samples-ROC-AUC": 0.9869244929437454,
      "eval_samples-Recall": 0.6802752681053619,
      "eval_samples_F1": 0.6963376282148493,
      "eval_samples_per_second": 669.723,
      "eval_steps_per_second": 2.62,
      "eval_weighted-F1": 0.6282531162024942,
      "eval_weighted-ROC-AUC": 0.967585319293872,
      "step": 10719
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 0.045110948383808136,
      "learning_rate": 1.0764903442485308e-05,
      "loss": 0.0179,
      "step": 11000
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 0.047784559428691864,
      "learning_rate": 1.034508816120907e-05,
      "loss": 0.0176,
      "step": 11500
    },
    {
      "epoch": 10.0,
      "eval_Hamming Loss": 0.006811598635028056,
      "eval_loss": 0.020576098933815956,
      "eval_macro-F1": 0.2500687036894995,
      "eval_micro-F1": 0.658813322669414,
      "eval_micro-Precision": 0.7577765650580807,
      "eval_micro-ROC-AUC": 0.9868861197392078,
      "eval_micro-Recall": 0.5827128520556815,
      "eval_runtime": 56.8276,
      "eval_samples-Precision": 0.7700463756412356,
      "eval_samples-ROC-AUC": 0.9873274265352923,
      "eval_samples-Recall": 0.677322176316279,
      "eval_samples_F1": 0.696440018353974,
      "eval_samples_per_second": 670.325,
      "eval_steps_per_second": 2.622,
      "eval_weighted-F1": 0.6286963446756773,
      "eval_weighted-ROC-AUC": 0.9675386582556953,
      "step": 11910
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 0.054966043680906296,
      "learning_rate": 9.92527287993283e-06,
      "loss": 0.0174,
      "step": 12000
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 0.05237603560090065,
      "learning_rate": 9.505457598656592e-06,
      "loss": 0.0168,
      "step": 12500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 0.05216987058520317,
      "learning_rate": 9.085642317380353e-06,
      "loss": 0.0169,
      "step": 13000
    },
    {
      "epoch": 11.0,
      "eval_Hamming Loss": 0.006823777185569446,
      "eval_loss": 0.020509161055088043,
      "eval_macro-F1": 0.26200424662299254,
      "eval_micro-F1": 0.6628468275723741,
      "eval_micro-Precision": 0.7491801048861215,
      "eval_micro-ROC-AUC": 0.9871619338418166,
      "eval_micro-Recall": 0.5943551191204154,
      "eval_runtime": 56.922,
      "eval_samples-Precision": 0.7695165679998852,
      "eval_samples-ROC-AUC": 0.9875443461940913,
      "eval_samples-Recall": 0.6863968374135058,
      "eval_samples_F1": 0.70167593015519,
      "eval_samples_per_second": 669.214,
      "eval_steps_per_second": 2.618,
      "eval_weighted-F1": 0.6349789993502296,
      "eval_weighted-ROC-AUC": 0.9679648550158526,
      "step": 13101
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 0.05352295935153961,
      "learning_rate": 8.665827036104115e-06,
      "loss": 0.0164,
      "step": 13500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 0.0528065450489521,
      "learning_rate": 8.246011754827877e-06,
      "loss": 0.0162,
      "step": 14000
    },
    {
      "epoch": 12.0,
      "eval_Hamming Loss": 0.006750841199549339,
      "eval_loss": 0.02036031149327755,
      "eval_macro-F1": 0.265561098619209,
      "eval_micro-F1": 0.6638230200605117,
      "eval_micro-Precision": 0.7578079325517708,
      "eval_micro-ROC-AUC": 0.9873305369247557,
      "eval_micro-Recall": 0.5905782765607952,
      "eval_runtime": 56.6104,
      "eval_samples-Precision": 0.7715894350459755,
      "eval_samples-ROC-AUC": 0.9878245117062334,
      "eval_samples-Recall": 0.6835733060735801,
      "eval_samples_F1": 0.7007100494384872,
      "eval_samples_per_second": 672.898,
      "eval_steps_per_second": 2.632,
      "eval_weighted-F1": 0.6369107935631289,
      "eval_weighted-ROC-AUC": 0.9682565361809532,
      "step": 14292
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 0.05650531128048897,
      "learning_rate": 7.826196473551639e-06,
      "loss": 0.016,
      "step": 14500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 0.0515579879283905,
      "learning_rate": 7.407220822837951e-06,
      "loss": 0.0157,
      "step": 15000
    },
    {
      "epoch": 13.0,
      "eval_Hamming Loss": 0.006759230867700076,
      "eval_loss": 0.020438507199287415,
      "eval_macro-F1": 0.27797168493479635,
      "eval_micro-F1": 0.6675363572831042,
      "eval_micro-Precision": 0.7502318901289609,
      "eval_micro-ROC-AUC": 0.9872694884494536,
      "eval_micro-Recall": 0.6012613455151493,
      "eval_runtime": 57.4109,
      "eval_samples-Precision": 0.7702766543216233,
      "eval_samples-ROC-AUC": 0.9878260312181315,
      "eval_samples-Recall": 0.6914046486678713,
      "eval_samples_F1": 0.7051631413757338,
      "eval_samples_per_second": 663.515,
      "eval_steps_per_second": 2.595,
      "eval_weighted-F1": 0.6422683921775024,
      "eval_weighted-ROC-AUC": 0.9680052742102324,
      "step": 15483
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 0.053546659648418427,
      "learning_rate": 6.987405541561714e-06,
      "loss": 0.0158,
      "step": 15500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 0.07182997465133667,
      "learning_rate": 6.567590260285475e-06,
      "loss": 0.0152,
      "step": 16000
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 0.06407944113016129,
      "learning_rate": 6.147774979009237e-06,
      "loss": 0.0153,
      "step": 16500
    },
    {
      "epoch": 14.0,
      "eval_Hamming Loss": 0.00676301975009073,
      "eval_loss": 0.020446496084332466,
      "eval_macro-F1": 0.28564157254020134,
      "eval_micro-F1": 0.6716854213060587,
      "eval_micro-Precision": 0.7428152151803098,
      "eval_micro-ROC-AUC": 0.9873961174551684,
      "eval_micro-Recall": 0.6129875424145415,
      "eval_runtime": 56.9561,
      "eval_samples-Precision": 0.769045098066178,
      "eval_samples-ROC-AUC": 0.9878886845834632,
      "eval_samples-Recall": 0.7013103269260028,
      "eval_samples_F1": 0.7106548871148202,
      "eval_samples_per_second": 668.814,
      "eval_steps_per_second": 2.616,
      "eval_weighted-F1": 0.6488405029590205,
      "eval_weighted-ROC-AUC": 0.9681342377692144,
      "step": 16674
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 0.06374882906675339,
      "learning_rate": 5.727959697732998e-06,
      "loss": 0.015,
      "step": 17000
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 0.05428186058998108,
      "learning_rate": 5.308144416456759e-06,
      "loss": 0.0148,
      "step": 17500
    },
    {
      "epoch": 15.0,
      "eval_Hamming Loss": 0.006737038842269096,
      "eval_loss": 0.020485879853367805,
      "eval_macro-F1": 0.2903380825378071,
      "eval_micro-F1": 0.6732943546534901,
      "eval_micro-Precision": 0.7436365744767206,
      "eval_micro-ROC-AUC": 0.9873218978561575,
      "eval_micro-Recall": 0.6151097682337566,
      "eval_runtime": 57.4469,
      "eval_samples-Precision": 0.7693417878451875,
      "eval_samples-ROC-AUC": 0.9879143861488975,
      "eval_samples-Recall": 0.7036283765729578,
      "eval_samples_F1": 0.7118673049549253,
      "eval_samples_per_second": 663.099,
      "eval_steps_per_second": 2.594,
      "eval_weighted-F1": 0.650088931259366,
      "eval_weighted-ROC-AUC": 0.968105545390811,
      "step": 17865
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 0.062481027096509933,
      "learning_rate": 4.888329135180521e-06,
      "loss": 0.0147,
      "step": 18000
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 0.06039604917168617,
      "learning_rate": 4.468513853904283e-06,
      "loss": 0.0145,
      "step": 18500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 0.07048299908638,
      "learning_rate": 4.0495382031905965e-06,
      "loss": 0.0145,
      "step": 19000
    },
    {
      "epoch": 16.0,
      "eval_Hamming Loss": 0.006746240413789259,
      "eval_loss": 0.020490169525146484,
      "eval_macro-F1": 0.2914569617906651,
      "eval_micro-F1": 0.672792307944738,
      "eval_micro-Precision": 0.7432354050056553,
      "eval_micro-ROC-AUC": 0.9873475904073974,
      "eval_micro-Recall": 0.6145462393439085,
      "eval_runtime": 56.6829,
      "eval_samples-Precision": 0.7698330401911111,
      "eval_samples-ROC-AUC": 0.9879202729867449,
      "eval_samples-Recall": 0.7029155752261821,
      "eval_samples_F1": 0.7116816735252254,
      "eval_samples_per_second": 672.037,
      "eval_steps_per_second": 2.629,
      "eval_weighted-F1": 0.65027977195525,
      "eval_weighted-ROC-AUC": 0.9680130603724392,
      "step": 19056
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 0.05856049060821533,
      "learning_rate": 3.6297229219143583e-06,
      "loss": 0.0142,
      "step": 19500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 0.059780750423669815,
      "learning_rate": 3.2099076406381196e-06,
      "loss": 0.0142,
      "step": 20000
    },
    {
      "epoch": 17.0,
      "eval_Hamming Loss": 0.006711193251675701,
      "eval_loss": 0.020483586937189102,
      "eval_macro-F1": 0.28941686346175016,
      "eval_micro-F1": 0.6717279358229306,
      "eval_micro-Precision": 0.7497598959795505,
      "eval_micro-ROC-AUC": 0.9873218576917725,
      "eval_micro-Recall": 0.6084073714374783,
      "eval_runtime": 57.2121,
      "eval_samples-Precision": 0.7701005038212662,
      "eval_samples-ROC-AUC": 0.9878985418085805,
      "eval_samples-Recall": 0.6967815659603637,
      "eval_samples_F1": 0.7080840509321433,
      "eval_samples_per_second": 665.821,
      "eval_steps_per_second": 2.604,
      "eval_weighted-F1": 0.6485769518381072,
      "eval_weighted-ROC-AUC": 0.9679402457458093,
      "step": 20247
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 0.055906716734170914,
      "learning_rate": 2.790092359361881e-06,
      "loss": 0.014,
      "step": 20500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 0.0654979944229126,
      "learning_rate": 2.3702770780856423e-06,
      "loss": 0.014,
      "step": 21000
    },
    {
      "epoch": 18.0,
      "eval_Hamming Loss": 0.006715658720207544,
      "eval_loss": 0.020548846572637558,
      "eval_macro-F1": 0.29578964881588043,
      "eval_micro-F1": 0.6735771742776524,
      "eval_micro-Precision": 0.7460370650970336,
      "eval_micro-ROC-AUC": 0.9872232935911024,
      "eval_micro-Recall": 0.6139467405249212,
      "eval_runtime": 56.7384,
      "eval_samples-Precision": 0.7704018084547316,
      "eval_samples-ROC-AUC": 0.9878310051497012,
      "eval_samples-Recall": 0.7018452283252362,
      "eval_samples_F1": 0.7113833671548305,
      "eval_samples_per_second": 671.38,
      "eval_steps_per_second": 2.626,
      "eval_weighted-F1": 0.6511503106110951,
      "eval_weighted-ROC-AUC": 0.9678236309782818,
      "step": 21438
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 0.06082635000348091,
      "learning_rate": 1.950461796809404e-06,
      "loss": 0.0138,
      "step": 21500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 0.0555616095662117,
      "learning_rate": 1.5306465155331654e-06,
      "loss": 0.0137,
      "step": 22000
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 0.05737029016017914,
      "learning_rate": 1.1116708648194795e-06,
      "loss": 0.0137,
      "step": 22500
    },
    {
      "epoch": 19.0,
      "eval_Hamming Loss": 0.006720665457652338,
      "eval_loss": 0.02056426741182804,
      "eval_macro-F1": 0.2955852120530338,
      "eval_micro-F1": 0.6742999540953505,
      "eval_micro-Precision": 0.7441631565996497,
      "eval_micro-ROC-AUC": 0.9872403324248038,
      "eval_micro-Recall": 0.6164286656355287,
      "eval_runtime": 56.7074,
      "eval_samples-Precision": 0.7709621898619877,
      "eval_samples-ROC-AUC": 0.9878666504914567,
      "eval_samples-Recall": 0.7039134950136432,
      "eval_samples_F1": 0.7126982699264178,
      "eval_samples_per_second": 671.747,
      "eval_steps_per_second": 2.628,
      "eval_weighted-F1": 0.6516401231059971,
      "eval_weighted-ROC-AUC": 0.9678204645643085,
      "step": 22629
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 0.06911616027355194,
      "learning_rate": 6.91855583543241e-07,
      "loss": 0.0136,
      "step": 23000
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 0.0590774230659008,
      "learning_rate": 2.7204030226700255e-07,
      "loss": 0.0137,
      "step": 23500
    },
    {
      "epoch": 20.0,
      "eval_Hamming Loss": 0.006729596394716025,
      "eval_loss": 0.02057667076587677,
      "eval_macro-F1": 0.29651580072093586,
      "eval_micro-F1": 0.6740253270758502,
      "eval_micro-Precision": 0.7434247625106635,
      "eval_micro-ROC-AUC": 0.9872140373684055,
      "eval_micro-Recall": 0.6164766255410477,
      "eval_runtime": 56.6316,
      "eval_samples-Precision": 0.7701395534639176,
      "eval_samples-ROC-AUC": 0.9878538452217698,
      "eval_samples-Recall": 0.70390428453663,
      "eval_samples_F1": 0.7124879028701201,
      "eval_samples_per_second": 672.646,
      "eval_steps_per_second": 2.631,
      "eval_weighted-F1": 0.6519540409689584,
      "eval_weighted-ROC-AUC": 0.9677342040265031,
      "step": 23820
    }
  ],
  "logging_steps": 500,
  "max_steps": 23820,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 1.6063684357289165e+18,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}