{
  "best_global_step": null,
  "best_metric": 0.9040361144490907,
  "best_model_checkpoint": null,
  "epoch": 0.9947643979057592,
  "eval_steps": 16,
  "global_step": 760,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005235602094240838,
      "grad_norm": 43.404293060302734,
      "learning_rate": 5.217391304347826e-07,
      "loss": 5.0279,
      "step": 4
    },
    {
      "epoch": 0.010471204188481676,
      "grad_norm": 49.279701232910156,
      "learning_rate": 1.2173913043478262e-06,
      "loss": 4.7533,
      "step": 8
    },
    {
      "epoch": 0.010471204188481676,
      "eval_F1_err_corr": 0.7856240609830496,
      "eval_accuracy": 0.8077030162412993,
      "eval_correct_accuracy": 0.9044467268195442,
      "eval_error_accuracy": 0.69439691596995,
      "eval_f1": 0.11022044088176353,
      "eval_loss": 0.6095216870307922,
      "eval_pr_auc": 0.14219526881257386,
      "eval_precision": 0.14986376021798364,
      "eval_recall": 0.08716323296354993,
      "eval_runtime": 73.3028,
      "eval_samples_per_second": 66.751,
      "eval_steps_per_second": 0.532,
      "step": 8
    },
    {
      "epoch": 0.015706806282722512,
      "grad_norm": 68.26057434082031,
      "learning_rate": 1.9130434782608697e-06,
      "loss": 4.998,
      "step": 12
    },
    {
      "epoch": 0.020942408376963352,
      "grad_norm": 36.13338088989258,
      "learning_rate": 2.6086956521739132e-06,
      "loss": 4.5877,
      "step": 16
    },
    {
      "epoch": 0.020942408376963352,
      "eval_F1_err_corr": 0.7534917056196923,
      "eval_accuracy": 0.8118174787316319,
      "eval_correct_accuracy": 0.7896249617333692,
      "eval_error_accuracy": 0.7205206637453856,
      "eval_f1": 0.19526392379944438,
      "eval_loss": 0.5342526435852051,
      "eval_pr_auc": 0.20049527075325158,
      "eval_precision": 0.23488224061107574,
      "eval_recall": 0.1670817296807788,
      "eval_runtime": 73.2246,
      "eval_samples_per_second": 66.822,
      "eval_steps_per_second": 0.533,
      "step": 16
    },
    {
      "epoch": 0.02617801047120419,
      "grad_norm": 47.67475128173828,
      "learning_rate": 3.3043478260869567e-06,
      "loss": 3.9689,
      "step": 20
    },
    {
      "epoch": 0.031413612565445025,
      "grad_norm": 37.08432388305664,
      "learning_rate": 4.000000000000001e-06,
      "loss": 3.4487,
      "step": 24
    },
    {
      "epoch": 0.031413612565445025,
      "eval_F1_err_corr": 0.8365777452640643,
      "eval_accuracy": 0.859860788863109,
      "eval_correct_accuracy": 0.9621554711899274,
      "eval_error_accuracy": 0.7399956688054159,
      "eval_f1": 0.1793478260869565,
      "eval_loss": 0.407941073179245,
      "eval_pr_auc": 0.3074633641681041,
      "eval_precision": 0.4487760652765186,
      "eval_recall": 0.11206701381027848,
      "eval_runtime": 73.2248,
      "eval_samples_per_second": 66.822,
      "eval_steps_per_second": 0.533,
      "step": 24
    },
    {
      "epoch": 0.03664921465968586,
      "grad_norm": 25.600910186767578,
      "learning_rate": 4.695652173913044e-06,
      "loss": 3.0438,
      "step": 28
    },
    {
      "epoch": 0.041884816753926704,
      "grad_norm": 34.34440612792969,
      "learning_rate": 5.391304347826088e-06,
      "loss": 2.7638,
      "step": 32
    },
    {
      "epoch": 0.041884816753926704,
      "eval_F1_err_corr": 0.8428950058877069,
      "eval_accuracy": 0.8726991492652746,
      "eval_correct_accuracy": 0.9294179033702546,
      "eval_error_accuracy": 0.7711096044925128,
      "eval_f1": 0.3288207470233241,
      "eval_loss": 0.3438774645328522,
      "eval_pr_auc": 0.4404164901818042,
      "eval_precision": 0.588098016336056,
      "eval_recall": 0.22820919175911253,
      "eval_runtime": 73.4011,
      "eval_samples_per_second": 66.661,
      "eval_steps_per_second": 0.531,
      "step": 32
    },
    {
      "epoch": 0.04712041884816754,
      "grad_norm": 41.228370666503906,
      "learning_rate": 6.086956521739132e-06,
      "loss": 2.7747,
      "step": 36
    },
    {
      "epoch": 0.05235602094240838,
      "grad_norm": 32.323543548583984,
      "learning_rate": 6.782608695652174e-06,
      "loss": 2.5846,
      "step": 40
    },
    {
      "epoch": 0.05235602094240838,
      "eval_F1_err_corr": 0.8543599553580162,
      "eval_accuracy": 0.8773395204949729,
      "eval_correct_accuracy": 0.9330410885418402,
      "eval_error_accuracy": 0.7879168072379804,
      "eval_f1": 0.45929360425473886,
      "eval_loss": 0.31489187479019165,
      "eval_pr_auc": 0.5089374415932274,
      "eval_precision": 0.5775034293552812,
      "eval_recall": 0.3812542449626443,
      "eval_runtime": 73.3517,
      "eval_samples_per_second": 66.706,
      "eval_steps_per_second": 0.532,
      "step": 40
    },
    {
      "epoch": 0.05759162303664921,
      "grad_norm": 30.525619506835938,
      "learning_rate": 7.478260869565218e-06,
      "loss": 2.4505,
      "step": 44
    },
    {
      "epoch": 0.06282722513089005,
      "grad_norm": 73.03593444824219,
      "learning_rate": 8.173913043478263e-06,
      "loss": 2.5538,
      "step": 48
    },
    {
      "epoch": 0.06282722513089005,
      "eval_F1_err_corr": 0.8642908344403837,
      "eval_accuracy": 0.8819180201082754,
      "eval_correct_accuracy": 0.9289934972548709,
      "eval_error_accuracy": 0.8080141568385051,
      "eval_f1": 0.5086883768824817,
      "eval_loss": 0.300601601600647,
      "eval_pr_auc": 0.5509099723979306,
      "eval_precision": 0.5894988066825776,
      "eval_recall": 0.44736246321032375,
      "eval_runtime": 73.2719,
      "eval_samples_per_second": 66.779,
      "eval_steps_per_second": 0.532,
      "step": 48
    },
    {
      "epoch": 0.06806282722513089,
      "grad_norm": 20.755630493164062,
      "learning_rate": 8.869565217391306e-06,
      "loss": 2.3355,
      "step": 52
    },
    {
      "epoch": 0.07329842931937172,
      "grad_norm": 21.251449584960938,
      "learning_rate": 9.565217391304349e-06,
      "loss": 2.3184,
      "step": 56
    },
    {
      "epoch": 0.07329842931937172,
      "eval_F1_err_corr": 0.8654107678235833,
      "eval_accuracy": 0.888631090487239,
      "eval_correct_accuracy": 0.9337635011162132,
      "eval_error_accuracy": 0.8063824940859462,
      "eval_f1": 0.49622166246851385,
      "eval_loss": 0.28497353196144104,
      "eval_pr_auc": 0.575085189348315,
      "eval_precision": 0.6496885305972884,
      "eval_recall": 0.4014036676477247,
      "eval_runtime": 73.2577,
      "eval_samples_per_second": 66.792,
      "eval_steps_per_second": 0.532,
      "step": 56
    },
    {
      "epoch": 0.07853403141361257,
      "grad_norm": 13.29970645904541,
      "learning_rate": 1.0260869565217393e-05,
      "loss": 2.1286,
      "step": 60
    },
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 7.41117000579834,
      "learning_rate": 1.0956521739130435e-05,
      "loss": 2.145,
      "step": 64
    },
    {
      "epoch": 0.08376963350785341,
      "eval_F1_err_corr": 0.8744012492187797,
      "eval_accuracy": 0.8954988399071926,
      "eval_correct_accuracy": 0.9417514666141938,
      "eval_error_accuracy": 0.8160413087302195,
      "eval_f1": 0.5564600840336135,
      "eval_loss": 0.269569456577301,
      "eval_pr_auc": 0.616943123420949,
      "eval_precision": 0.6623944982807127,
      "eval_recall": 0.4797373783110709,
      "eval_runtime": 73.2419,
      "eval_samples_per_second": 66.806,
      "eval_steps_per_second": 0.532,
      "step": 64
    },
    {
      "epoch": 0.08900523560209424,
      "grad_norm": 15.101433753967285,
      "learning_rate": 1.1652173913043478e-05,
      "loss": 2.0576,
      "step": 68
    },
    {
      "epoch": 0.09424083769633508,
      "grad_norm": 23.230924606323242,
      "learning_rate": 1.2347826086956523e-05,
      "loss": 2.0721,
      "step": 72
    },
    {
      "epoch": 0.09424083769633508,
      "eval_F1_err_corr": 0.8819337284716499,
      "eval_accuracy": 0.902799690641918,
      "eval_correct_accuracy": 0.9505328366301462,
      "eval_error_accuracy": 0.8225696122378097,
      "eval_f1": 0.5853787278965427,
      "eval_loss": 0.2565557360649109,
      "eval_pr_auc": 0.6563233866294026,
      "eval_precision": 0.7016766845934831,
      "eval_recall": 0.5021507810731266,
      "eval_runtime": 73.2462,
      "eval_samples_per_second": 66.802,
      "eval_steps_per_second": 0.532,
      "step": 72
    },
    {
      "epoch": 0.09947643979057591,
      "grad_norm": 42.72426223754883,
      "learning_rate": 1.3043478260869566e-05,
      "loss": 2.0267,
      "step": 76
    },
    {
      "epoch": 0.10471204188481675,
      "grad_norm": 6.76558256149292,
      "learning_rate": 1.373913043478261e-05,
      "loss": 1.99,
      "step": 80
    },
    {
      "epoch": 0.10471204188481675,
      "eval_F1_err_corr": 0.8814308259221694,
      "eval_accuracy": 0.9006960556844548,
      "eval_correct_accuracy": 0.9394550028793558,
      "eval_error_accuracy": 0.8301572763056169,
      "eval_f1": 0.6051660516605166,
      "eval_loss": 0.25570690631866455,
      "eval_pr_auc": 0.6715258461977598,
      "eval_precision": 0.6625370320495556,
      "eval_recall": 0.5569390989359294,
      "eval_runtime": 73.2805,
      "eval_samples_per_second": 66.771,
      "eval_steps_per_second": 0.532,
      "step": 80
    },
    {
      "epoch": 0.1099476439790576,
      "grad_norm": 10.198799133300781,
      "learning_rate": 1.4434782608695654e-05,
      "loss": 1.9619,
      "step": 84
    },
    {
      "epoch": 0.11518324607329843,
      "grad_norm": 14.052003860473633,
      "learning_rate": 1.5130434782608697e-05,
      "loss": 1.9286,
      "step": 88
    },
    {
      "epoch": 0.12041884816753927,
      "grad_norm": 25.076927185058594,
      "learning_rate": 1.582608695652174e-05,
      "loss": 1.9326,
      "step": 92
    },
    {
      "epoch": 0.1256544502617801,
      "grad_norm": 13.464066505432129,
      "learning_rate": 1.6521739130434785e-05,
      "loss": 1.9319,
      "step": 96
    },
    {
      "epoch": 0.13089005235602094,
      "grad_norm": 8.98434829711914,
      "learning_rate": 1.721739130434783e-05,
      "loss": 1.8496,
      "step": 100
    },
    {
      "epoch": 0.13612565445026178,
      "grad_norm": 37.530517578125,
      "learning_rate": 1.791304347826087e-05,
      "loss": 1.852,
      "step": 104
    },
    {
      "epoch": 0.14136125654450263,
      "grad_norm": 59.9710578918457,
      "learning_rate": 1.8608695652173912e-05,
      "loss": 2.0372,
      "step": 108
    },
    {
      "epoch": 0.14659685863874344,
      "grad_norm": 111.6693115234375,
      "learning_rate": 1.9304347826086957e-05,
      "loss": 2.0572,
      "step": 112
    },
    {
      "epoch": 0.1518324607329843,
      "grad_norm": 29.88340950012207,
      "learning_rate": 2e-05,
      "loss": 2.0327,
      "step": 116
    },
    {
      "epoch": 0.15706806282722513,
      "grad_norm": 32.678375244140625,
      "learning_rate": 1.9999942480792804e-05,
      "loss": 1.9213,
      "step": 120
    },
    {
      "epoch": 0.15706806282722513,
      "eval_F1_err_corr": 0.8993955274771063,
      "eval_accuracy": 0.911152358855375,
      "eval_correct_accuracy": 0.9550211716256353,
      "eval_error_accuracy": 0.8498931149830868,
      "eval_f1": 0.6542258608235011,
      "eval_loss": 0.23431183397769928,
      "eval_pr_auc": 0.723673140949982,
      "eval_precision": 0.6986371817948058,
      "eval_recall": 0.6151233869141951,
      "eval_runtime": 73.2147,
      "eval_samples_per_second": 66.831,
      "eval_steps_per_second": 0.533,
      "step": 120
    },
    {
      "epoch": 0.16230366492146597,
      "grad_norm": 23.950916290283203,
      "learning_rate": 1.999976992383291e-05,
      "loss": 1.8462,
      "step": 124
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 19.700756072998047,
      "learning_rate": 1.9999482331105377e-05,
      "loss": 1.7507,
      "step": 128
    },
    {
      "epoch": 0.17277486910994763,
      "grad_norm": 40.214717864990234,
      "learning_rate": 1.9999079705918636e-05,
      "loss": 1.9274,
      "step": 132
    },
    {
      "epoch": 0.17801047120418848,
      "grad_norm": 16.36521339416504,
      "learning_rate": 1.999856205290442e-05,
      "loss": 1.6216,
      "step": 136
    },
    {
      "epoch": 0.18324607329842932,
      "grad_norm": 14.577583312988281,
      "learning_rate": 1.9997929378017723e-05,
      "loss": 1.6965,
      "step": 140
    },
    {
      "epoch": 0.18848167539267016,
      "grad_norm": 32.502071380615234,
      "learning_rate": 1.9997181688536746e-05,
      "loss": 1.6885,
      "step": 144
    },
    {
      "epoch": 0.193717277486911,
      "grad_norm": 20.771347045898438,
      "learning_rate": 1.999631899306278e-05,
      "loss": 1.6577,
      "step": 148
    },
    {
      "epoch": 0.19895287958115182,
      "grad_norm": 9.587984085083008,
      "learning_rate": 1.999534130152014e-05,
      "loss": 1.6842,
      "step": 152
    },
    {
      "epoch": 0.20418848167539266,
      "grad_norm": 23.711835861206055,
      "learning_rate": 1.999424862515604e-05,
      "loss": 1.8278,
      "step": 156
    },
    {
      "epoch": 0.2094240837696335,
      "grad_norm": 22.35334014892578,
      "learning_rate": 1.999304097654045e-05,
      "loss": 1.743,
      "step": 160
    },
    {
      "epoch": 0.2094240837696335,
      "eval_F1_err_corr": 0.8993887871915233,
      "eval_accuracy": 0.9107192575406032,
      "eval_correct_accuracy": 0.959441733768809,
      "eval_error_accuracy": 0.8464106325913571,
      "eval_f1": 0.6483061174750183,
      "eval_loss": 0.23133407533168793,
      "eval_pr_auc": 0.7244312751634396,
      "eval_precision": 0.7020321984692531,
      "eval_recall": 0.6022187004754358,
      "eval_runtime": 73.2104,
      "eval_samples_per_second": 66.835,
      "eval_steps_per_second": 0.533,
      "step": 160
    },
    {
      "epoch": 0.21465968586387435,
      "grad_norm": 18.791339874267578,
      "learning_rate": 1.999171836956597e-05,
      "loss": 1.6844,
      "step": 164
    },
    {
      "epoch": 0.2198952879581152,
      "grad_norm": 6.037458896636963,
      "learning_rate": 1.9990280819447662e-05,
      "loss": 1.6566,
      "step": 168
    },
    {
      "epoch": 0.225130890052356,
      "grad_norm": 7.139956951141357,
      "learning_rate": 1.998872834272287e-05,
      "loss": 1.6759,
      "step": 172
    },
    {
      "epoch": 0.23036649214659685,
      "grad_norm": 11.674882888793945,
      "learning_rate": 1.9987060957251047e-05,
      "loss": 1.6746,
      "step": 176
    },
    {
      "epoch": 0.2356020942408377,
      "grad_norm": 6.298519134521484,
      "learning_rate": 1.9985278682213525e-05,
      "loss": 1.4914,
      "step": 180
    },
    {
      "epoch": 0.24083769633507854,
      "grad_norm": 14.267353057861328,
      "learning_rate": 1.9983381538113317e-05,
      "loss": 1.6223,
      "step": 184
    },
    {
      "epoch": 0.24607329842931938,
      "grad_norm": 7.899906158447266,
      "learning_rate": 1.998136954677487e-05,
      "loss": 1.6448,
      "step": 188
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 20.155229568481445,
      "learning_rate": 1.9979242731343803e-05,
      "loss": 1.6562,
      "step": 192
    },
    {
      "epoch": 0.25654450261780104,
      "grad_norm": 21.51776123046875,
      "learning_rate": 1.9977001116286675e-05,
      "loss": 1.6211,
      "step": 196
    },
    {
      "epoch": 0.2617801047120419,
      "grad_norm": 17.941186904907227,
      "learning_rate": 1.9974644727390665e-05,
      "loss": 1.6279,
      "step": 200
    },
    {
      "epoch": 0.2617801047120419,
      "eval_F1_err_corr": 0.8988893377228542,
      "eval_accuracy": 0.9054601701469451,
      "eval_correct_accuracy": 0.9581679176379972,
      "eval_error_accuracy": 0.8465181481131606,
      "eval_f1": 0.6507428571428572,
      "eval_loss": 0.23669207096099854,
      "eval_pr_auc": 0.7193585351305057,
      "eval_precision": 0.6570505423494115,
      "eval_recall": 0.6445551279148743,
      "eval_runtime": 73.2472,
      "eval_samples_per_second": 66.801,
      "eval_steps_per_second": 0.532,
      "step": 200
    },
    {
      "epoch": 0.2670157068062827,
      "grad_norm": 22.866968154907227,
      "learning_rate": 1.9972173591763297e-05,
      "loss": 1.7616,
      "step": 204
    },
    {
      "epoch": 0.27225130890052357,
      "grad_norm": 20.52739715576172,
      "learning_rate": 1.996958773783213e-05,
      "loss": 1.5429,
      "step": 208
    },
    {
      "epoch": 0.2774869109947644,
      "grad_norm": 8.633818626403809,
      "learning_rate": 1.9966887195344403e-05,
      "loss": 1.5773,
      "step": 212
    },
    {
      "epoch": 0.28272251308900526,
      "grad_norm": 23.69244956970215,
      "learning_rate": 1.9964071995366744e-05,
      "loss": 1.5115,
      "step": 216
    },
    {
      "epoch": 0.2879581151832461,
      "grad_norm": 19.777175903320312,
      "learning_rate": 1.9961142170284762e-05,
      "loss": 1.4079,
      "step": 220
    },
    {
      "epoch": 0.2931937172774869,
      "grad_norm": 8.477285385131836,
      "learning_rate": 1.9958097753802693e-05,
      "loss": 1.5007,
      "step": 224
    },
    {
      "epoch": 0.29842931937172773,
      "grad_norm": 10.63436508178711,
      "learning_rate": 1.9954938780943034e-05,
      "loss": 1.4683,
      "step": 228
    },
    {
      "epoch": 0.3036649214659686,
      "grad_norm": 7.749914646148682,
      "learning_rate": 1.9951665288046098e-05,
      "loss": 1.5781,
      "step": 232
    },
    {
      "epoch": 0.3089005235602094,
      "grad_norm": 8.496782302856445,
      "learning_rate": 1.994827731276963e-05,
      "loss": 1.5299,
      "step": 236
    },
    {
      "epoch": 0.31413612565445026,
      "grad_norm": 19.28988265991211,
      "learning_rate": 1.9944774894088367e-05,
      "loss": 1.5927,
      "step": 240
    },
    {
      "epoch": 0.31413612565445026,
      "eval_F1_err_corr": 0.8974736809922041,
      "eval_accuracy": 0.8993039443155453,
      "eval_correct_accuracy": 0.9425683761157468,
      "eval_error_accuracy": 0.8564968507306808,
      "eval_f1": 0.665845395749923,
      "eval_loss": 0.2493041455745697,
      "eval_pr_auc": 0.733870235874116,
      "eval_precision": 0.6091284748309542,
      "eval_recall": 0.7342087389630971,
      "eval_runtime": 73.2151,
      "eval_samples_per_second": 66.83,
      "eval_steps_per_second": 0.533,
      "step": 240
    },
    {
      "epoch": 0.3193717277486911,
      "grad_norm": 15.320127487182617,
      "learning_rate": 1.994115807229357e-05,
      "loss": 1.5878,
      "step": 244
    },
    {
      "epoch": 0.32460732984293195,
      "grad_norm": 11.574797630310059,
      "learning_rate": 1.993742688899259e-05,
      "loss": 1.5463,
      "step": 248
    },
    {
      "epoch": 0.3298429319371728,
      "grad_norm": 10.225842475891113,
      "learning_rate": 1.9933581387108358e-05,
      "loss": 1.38,
      "step": 252
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 7.9395623207092285,
      "learning_rate": 1.992962161087893e-05,
      "loss": 1.4878,
      "step": 256
    },
    {
      "epoch": 0.3403141361256545,
      "grad_norm": 6.2755937576293945,
      "learning_rate": 1.9925547605856937e-05,
      "loss": 1.467,
      "step": 260
    },
    {
      "epoch": 0.34554973821989526,
      "grad_norm": 12.917094230651855,
      "learning_rate": 1.992135941890909e-05,
      "loss": 1.437,
      "step": 264
    },
    {
      "epoch": 0.3507853403141361,
      "grad_norm": 8.741606712341309,
      "learning_rate": 1.9917057098215624e-05,
      "loss": 1.4992,
      "step": 268
    },
    {
      "epoch": 0.35602094240837695,
      "grad_norm": 22.467348098754883,
      "learning_rate": 1.9912640693269754e-05,
      "loss": 1.5051,
      "step": 272
    },
    {
      "epoch": 0.3612565445026178,
      "grad_norm": 22.836654663085938,
      "learning_rate": 1.9908110254877107e-05,
      "loss": 1.4494,
      "step": 276
    },
    {
      "epoch": 0.36649214659685864,
      "grad_norm": 7.070972919464111,
      "learning_rate": 1.9903465835155124e-05,
      "loss": 1.3466,
      "step": 280
    },
    {
      "epoch": 0.36649214659685864,
      "eval_F1_err_corr": 0.904888679332032,
      "eval_accuracy": 0.9101005413766434,
      "eval_correct_accuracy": 0.9618798467956057,
      "eval_error_accuracy": 0.8542731828285801,
      "eval_f1": 0.665053019824804,
      "eval_loss": 0.2285982072353363,
      "eval_pr_auc": 0.7431078398519564,
      "eval_precision": 0.6773890584644283,
      "eval_recall": 0.6531582522073806,
      "eval_runtime": 73.4017,
      "eval_samples_per_second": 66.661,
      "eval_steps_per_second": 0.531,
      "step": 280
    },
    {
      "epoch": 0.3717277486910995,
      "grad_norm": 6.082576274871826,
      "learning_rate": 1.9898707487532475e-05,
      "loss": 1.4552,
      "step": 284
    },
    {
      "epoch": 0.3769633507853403,
      "grad_norm": 7.502638339996338,
      "learning_rate": 1.9893835266748437e-05,
      "loss": 1.4758,
      "step": 288
    },
    {
      "epoch": 0.38219895287958117,
      "grad_norm": 15.583269119262695,
      "learning_rate": 1.9888849228852262e-05,
      "loss": 1.4416,
      "step": 292
    },
    {
      "epoch": 0.387434554973822,
      "grad_norm": 6.652861595153809,
      "learning_rate": 1.988374943120254e-05,
      "loss": 1.42,
      "step": 296
    },
    {
      "epoch": 0.39267015706806285,
      "grad_norm": 16.742830276489258,
      "learning_rate": 1.987853593246654e-05,
      "loss": 1.4353,
      "step": 300
    },
    {
      "epoch": 0.39790575916230364,
      "grad_norm": 9.651524543762207,
      "learning_rate": 1.9873208792619517e-05,
      "loss": 1.3519,
      "step": 304
    },
    {
      "epoch": 0.4031413612565445,
      "grad_norm": 7.194350719451904,
      "learning_rate": 1.9867768072944047e-05,
      "loss": 1.4409,
      "step": 308
    },
    {
      "epoch": 0.4083769633507853,
      "grad_norm": 9.677583694458008,
      "learning_rate": 1.9862213836029308e-05,
      "loss": 1.3468,
      "step": 312
    },
    {
      "epoch": 0.41361256544502617,
      "grad_norm": 10.901875495910645,
      "learning_rate": 1.985654614577036e-05,
      "loss": 1.3667,
      "step": 316
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 11.349899291992188,
      "learning_rate": 1.985076506736741e-05,
      "loss": 1.3641,
      "step": 320
    },
    {
      "epoch": 0.418848167539267,
      "eval_F1_err_corr": 0.9031122184334268,
      "eval_accuracy": 0.9109667440061872,
      "eval_correct_accuracy": 0.9675908606266521,
      "eval_error_accuracy": 0.8466901950778333,
      "eval_f1": 0.656645192078263,
      "eval_loss": 0.22545810043811798,
      "eval_pr_auc": 0.7398421046192083,
      "eval_precision": 0.6940731399747794,
      "eval_recall": 0.6230473171836088,
      "eval_runtime": 73.2834,
      "eval_samples_per_second": 66.768,
      "eval_steps_per_second": 0.532,
      "step": 320
    },
    {
      "epoch": 0.42408376963350786,
      "grad_norm": 9.947056770324707,
      "learning_rate": 1.9844870667325073e-05,
      "loss": 1.3222,
      "step": 324
    },
    {
      "epoch": 0.4293193717277487,
      "grad_norm": 9.555671691894531,
      "learning_rate": 1.9838863013451587e-05,
      "loss": 1.3875,
      "step": 328
    },
    {
      "epoch": 0.43455497382198954,
      "grad_norm": 19.172666549682617,
      "learning_rate": 1.9832742174858052e-05,
      "loss": 1.3089,
      "step": 332
    },
    {
      "epoch": 0.4397905759162304,
      "grad_norm": 14.9385404586792,
      "learning_rate": 1.9826508221957624e-05,
      "loss": 1.3496,
      "step": 336
    },
    {
      "epoch": 0.44502617801047123,
      "grad_norm": 9.018355369567871,
      "learning_rate": 1.9820161226464708e-05,
      "loss": 1.2145,
      "step": 340
    },
    {
      "epoch": 0.450261780104712,
      "grad_norm": 9.747211456298828,
      "learning_rate": 1.9813701261394136e-05,
      "loss": 1.3075,
      "step": 344
    },
    {
      "epoch": 0.45549738219895286,
      "grad_norm": 6.160175323486328,
      "learning_rate": 1.980712840106032e-05,
      "loss": 1.2454,
      "step": 348
    },
    {
      "epoch": 0.4607329842931937,
      "grad_norm": 16.006214141845703,
      "learning_rate": 1.9800442721076406e-05,
      "loss": 1.2183,
      "step": 352
    },
    {
      "epoch": 0.46596858638743455,
      "grad_norm": 8.38248062133789,
      "learning_rate": 1.979364429835339e-05,
      "loss": 1.3076,
      "step": 356
    },
    {
      "epoch": 0.4712041884816754,
      "grad_norm": 12.295247077941895,
      "learning_rate": 1.9786733211099257e-05,
      "loss": 1.3584,
      "step": 360
    },
    {
      "epoch": 0.4712041884816754,
      "eval_F1_err_corr": 0.9053759601017026,
      "eval_accuracy": 0.9166589327146172,
      "eval_correct_accuracy": 0.9761573206307471,
      "eval_error_accuracy": 0.8441653625762615,
      "eval_f1": 0.6447784810126582,
      "eval_loss": 0.2235945612192154,
      "eval_pr_auc": 0.7409747509705941,
      "eval_precision": 0.7720239974739501,
      "eval_recall": 0.5535431288204664,
      "eval_runtime": 73.3925,
      "eval_samples_per_second": 66.669,
      "eval_steps_per_second": 0.531,
      "step": 360
    },
    {
      "epoch": 0.47643979057591623,
      "grad_norm": 16.797306060791016,
      "learning_rate": 1.9779709538818052e-05,
      "loss": 1.3383,
      "step": 364
    },
    {
      "epoch": 0.4816753926701571,
      "grad_norm": 8.570252418518066,
      "learning_rate": 1.9772573362308992e-05,
      "loss": 1.3023,
      "step": 368
    },
    {
      "epoch": 0.4869109947643979,
      "grad_norm": 12.603636741638184,
      "learning_rate": 1.9765324763665516e-05,
      "loss": 1.2977,
      "step": 372
    },
    {
      "epoch": 0.49214659685863876,
      "grad_norm": 7.162085056304932,
      "learning_rate": 1.9757963826274357e-05,
      "loss": 1.218,
      "step": 376
    },
    {
      "epoch": 0.4973821989528796,
      "grad_norm": 7.1061882972717285,
      "learning_rate": 1.975049063481457e-05,
      "loss": 1.2535,
      "step": 380
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 10.717151641845703,
      "learning_rate": 1.974290527525657e-05,
      "loss": 1.3631,
      "step": 384
    },
    {
      "epoch": 0.5078534031413613,
      "grad_norm": 14.366579055786133,
      "learning_rate": 1.9735207834861117e-05,
      "loss": 1.2677,
      "step": 388
    },
    {
      "epoch": 0.5130890052356021,
      "grad_norm": 16.858505249023438,
      "learning_rate": 1.972739840217836e-05,
      "loss": 1.2859,
      "step": 392
    },
    {
      "epoch": 0.518324607329843,
      "grad_norm": 9.74325180053711,
      "learning_rate": 1.9719477067046768e-05,
      "loss": 1.2454,
      "step": 396
    },
    {
      "epoch": 0.5235602094240838,
      "grad_norm": 8.15595531463623,
      "learning_rate": 1.971144392059212e-05,
      "loss": 1.1893,
      "step": 400
    },
    {
      "epoch": 0.5235602094240838,
      "eval_F1_err_corr": 0.9038263356777348,
      "eval_accuracy": 0.911307037896365,
      "eval_correct_accuracy": 0.9625925807635642,
      "eval_error_accuracy": 0.8518225945701047,
      "eval_f1": 0.6579983299534773,
      "eval_loss": 0.22870948910713196,
      "eval_pr_auc": 0.736284937342107,
      "eval_precision": 0.6954109934442764,
      "eval_recall": 0.624405705229794,
      "eval_runtime": 73.2606,
      "eval_samples_per_second": 66.789,
      "eval_steps_per_second": 0.532,
      "step": 400
    },
    {
      "epoch": 0.5287958115183246,
      "grad_norm": 14.265653610229492,
      "learning_rate": 1.970329905522647e-05,
      "loss": 1.2027,
      "step": 404
    },
    {
      "epoch": 0.5340314136125655,
      "grad_norm": 11.760555267333984,
      "learning_rate": 1.9695042564647045e-05,
      "loss": 1.1617,
      "step": 408
    },
    {
      "epoch": 0.5392670157068062,
      "grad_norm": 6.7435383796691895,
      "learning_rate": 1.9686674543835208e-05,
      "loss": 1.2307,
      "step": 412
    },
    {
      "epoch": 0.5445026178010471,
      "grad_norm": 10.695906639099121,
      "learning_rate": 1.9678195089055347e-05,
      "loss": 1.098,
      "step": 416
    },
    {
      "epoch": 0.5497382198952879,
      "grad_norm": 7.993613243103027,
      "learning_rate": 1.9669604297853766e-05,
      "loss": 1.285,
      "step": 420
    },
    {
      "epoch": 0.5549738219895288,
      "grad_norm": 8.1636962890625,
      "learning_rate": 1.9660902269057558e-05,
      "loss": 1.2282,
      "step": 424
    },
    {
      "epoch": 0.5602094240837696,
      "grad_norm": 13.683623313903809,
      "learning_rate": 1.9652089102773487e-05,
      "loss": 1.2109,
      "step": 428
    },
    {
      "epoch": 0.5654450261780105,
      "grad_norm": 7.191526889801025,
      "learning_rate": 1.9643164900386824e-05,
      "loss": 1.1563,
      "step": 432
    },
    {
      "epoch": 0.5706806282722513,
      "grad_norm": 14.868063926696777,
      "learning_rate": 1.963412976456017e-05,
      "loss": 1.2169,
      "step": 436
    },
    {
      "epoch": 0.5759162303664922,
      "grad_norm": 16.254392623901367,
      "learning_rate": 1.96249837992323e-05,
      "loss": 1.2315,
      "step": 440
    },
    {
      "epoch": 0.5759162303664922,
      "eval_F1_err_corr": 0.9032073036079588,
      "eval_accuracy": 0.9144934261407579,
      "eval_correct_accuracy": 0.9674043692973997,
      "eval_error_accuracy": 0.8470002679889722,
      "eval_f1": 0.6580900544285008,
      "eval_loss": 0.2237122654914856,
      "eval_pr_auc": 0.7399598147055758,
      "eval_precision": 0.7253886010362695,
      "eval_recall": 0.6022187004754358,
      "eval_runtime": 73.2075,
      "eval_samples_per_second": 66.837,
      "eval_steps_per_second": 0.533,
      "step": 440
    },
    {
      "epoch": 0.581151832460733,
      "grad_norm": 13.422707557678223,
      "learning_rate": 1.961572710961695e-05,
      "loss": 1.0837,
      "step": 444
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 8.574109077453613,
      "learning_rate": 1.9606359802201608e-05,
      "loss": 1.1012,
      "step": 448
    },
    {
      "epoch": 0.5916230366492147,
      "grad_norm": 8.540183067321777,
      "learning_rate": 1.9596881984746288e-05,
      "loss": 1.1766,
      "step": 452
    },
    {
      "epoch": 0.5968586387434555,
      "grad_norm": 10.323470115661621,
      "learning_rate": 1.958729376628231e-05,
      "loss": 1.0225,
      "step": 456
    },
    {
      "epoch": 0.6020942408376964,
      "grad_norm": 7.158539772033691,
      "learning_rate": 1.957759525711101e-05,
      "loss": 1.1934,
      "step": 460
    },
    {
      "epoch": 0.6073298429319371,
      "grad_norm": 11.638604164123535,
      "learning_rate": 1.9567786568802503e-05,
      "loss": 1.1326,
      "step": 464
    },
    {
      "epoch": 0.612565445026178,
      "grad_norm": 8.771400451660156,
      "learning_rate": 1.9557867814194385e-05,
      "loss": 1.1736,
      "step": 468
    },
    {
      "epoch": 0.6178010471204188,
      "grad_norm": 7.182300567626953,
      "learning_rate": 1.9547839107390435e-05,
      "loss": 1.0966,
      "step": 472
    },
    {
      "epoch": 0.6230366492146597,
      "grad_norm": 7.561171054840088,
      "learning_rate": 1.9537700563759303e-05,
      "loss": 1.1993,
      "step": 476
    },
    {
      "epoch": 0.6282722513089005,
      "grad_norm": 11.45133113861084,
      "learning_rate": 1.9527452299933192e-05,
      "loss": 1.1176,
      "step": 480
    },
    {
      "epoch": 0.6282722513089005,
      "eval_F1_err_corr": 0.8976839881538714,
      "eval_accuracy": 0.9037896365042537,
      "eval_correct_accuracy": 0.9433196060921855,
      "eval_error_accuracy": 0.8562601071557276,
      "eval_f1": 0.6728382074479277,
      "eval_loss": 0.24335815012454987,
      "eval_pr_auc": 0.7403424614633654,
      "eval_precision": 0.6284142267636078,
      "eval_recall": 0.7240208286167081,
      "eval_runtime": 73.2495,
      "eval_samples_per_second": 66.799,
      "eval_steps_per_second": 0.532,
      "step": 480
    },
    {
      "epoch": 0.6335078534031413,
      "grad_norm": 6.921932697296143,
      "learning_rate": 1.95170944338065e-05,
      "loss": 1.2263,
      "step": 484
    },
    {
      "epoch": 0.6387434554973822,
      "grad_norm": 8.900130271911621,
      "learning_rate": 1.9506627084534486e-05,
      "loss": 1.0606,
      "step": 488
    },
    {
      "epoch": 0.643979057591623,
      "grad_norm": 7.923438549041748,
      "learning_rate": 1.9496050372531864e-05,
      "loss": 1.0479,
      "step": 492
    },
    {
      "epoch": 0.6492146596858639,
      "grad_norm": 7.973376750946045,
      "learning_rate": 1.9485364419471454e-05,
      "loss": 1.0766,
      "step": 496
    },
    {
      "epoch": 0.6544502617801047,
      "grad_norm": 8.22862434387207,
      "learning_rate": 1.9474569348282774e-05,
      "loss": 1.0183,
      "step": 500
    },
    {
      "epoch": 0.6596858638743456,
      "grad_norm": 7.483217716217041,
      "learning_rate": 1.9463665283150604e-05,
      "loss": 1.1458,
      "step": 504
    },
    {
      "epoch": 0.6649214659685864,
      "grad_norm": 9.2059907913208,
      "learning_rate": 1.9452652349513587e-05,
      "loss": 1.0281,
      "step": 508
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 9.361949920654297,
      "learning_rate": 1.9441530674062754e-05,
      "loss": 1.1281,
      "step": 512
    },
    {
      "epoch": 0.675392670157068,
      "grad_norm": 8.502827644348145,
      "learning_rate": 1.9430300384740108e-05,
      "loss": 1.0423,
      "step": 516
    },
    {
      "epoch": 0.680628272251309,
      "grad_norm": 8.099173545837402,
      "learning_rate": 1.941896161073711e-05,
      "loss": 1.0373,
      "step": 520
    },
    {
      "epoch": 0.680628272251309,
      "eval_F1_err_corr": 0.8887221516150057,
      "eval_accuracy": 0.8935189481825213,
      "eval_correct_accuracy": 0.9295264018044566,
      "eval_error_accuracy": 0.8513496941450817,
      "eval_f1": 0.6533037872683319,
      "eval_loss": 0.2687327563762665,
      "eval_pr_auc": 0.7259388976075478,
      "eval_precision": 0.5884594447468698,
      "eval_recall": 0.7342087389630971,
      "eval_runtime": 73.2743,
      "eval_samples_per_second": 66.776,
      "eval_steps_per_second": 0.532,
      "step": 520
    },
    {
      "epoch": 0.6858638743455497,
      "grad_norm": 10.470741271972656,
      "learning_rate": 1.9407514482493214e-05,
      "loss": 1.1026,
      "step": 524
    },
    {
      "epoch": 0.6910994764397905,
      "grad_norm": 8.012195587158203,
      "learning_rate": 1.939595913169438e-05,
      "loss": 1.0198,
      "step": 528
    },
    {
      "epoch": 0.6963350785340314,
      "grad_norm": 9.073850631713867,
      "learning_rate": 1.9384295691271523e-05,
      "loss": 1.0025,
      "step": 532
    },
    {
      "epoch": 0.7015706806282722,
      "grad_norm": 7.318668365478516,
      "learning_rate": 1.9372524295399014e-05,
      "loss": 1.0328,
      "step": 536
    },
    {
      "epoch": 0.7068062827225131,
      "grad_norm": 8.4859037399292,
      "learning_rate": 1.9360645079493126e-05,
      "loss": 1.017,
      "step": 540
    },
    {
      "epoch": 0.7120418848167539,
      "grad_norm": 8.274374008178711,
      "learning_rate": 1.9348658180210473e-05,
      "loss": 0.9398,
      "step": 544
    },
    {
      "epoch": 0.7172774869109948,
      "grad_norm": 9.273284912109375,
      "learning_rate": 1.933656373544645e-05,
      "loss": 0.9819,
      "step": 548
    },
    {
      "epoch": 0.7225130890052356,
      "grad_norm": 7.139795780181885,
      "learning_rate": 1.932436188433362e-05,
      "loss": 0.9578,
      "step": 552
    },
    {
      "epoch": 0.7277486910994765,
      "grad_norm": 11.788651466369629,
      "learning_rate": 1.9312052767240153e-05,
      "loss": 0.965,
      "step": 556
    },
    {
      "epoch": 0.7329842931937173,
      "grad_norm": 8.076215744018555,
      "learning_rate": 1.9299636525768176e-05,
      "loss": 0.9347,
      "step": 560
    },
    {
      "epoch": 0.7329842931937173,
      "eval_F1_err_corr": 0.9048476894056305,
      "eval_accuracy": 0.9066666666666666,
      "eval_correct_accuracy": 0.9648984138016987,
      "eval_error_accuracy": 0.851833586371737,
      "eval_f1": 0.663544106167057,
      "eval_loss": 0.2468423694372177,
      "eval_pr_auc": 0.7270364099943972,
      "eval_precision": 0.6538461538461539,
      "eval_recall": 0.6735340729001584,
      "eval_runtime": 73.26,
      "eval_samples_per_second": 66.79,
      "eval_steps_per_second": 0.532,
      "step": 560
    },
    {
      "epoch": 0.7382198952879581,
      "grad_norm": 7.399974346160889,
      "learning_rate": 1.9287113302752167e-05,
      "loss": 0.9323,
      "step": 564
    },
    {
      "epoch": 0.743455497382199,
      "grad_norm": 8.363114356994629,
      "learning_rate": 1.927448324225729e-05,
      "loss": 0.9909,
      "step": 568
    },
    {
      "epoch": 0.7486910994764397,
      "grad_norm": 8.044791221618652,
      "learning_rate": 1.9261746489577767e-05,
      "loss": 0.9245,
      "step": 572
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 10.069940567016602,
      "learning_rate": 1.9248903191235177e-05,
      "loss": 0.9564,
      "step": 576
    },
    {
      "epoch": 0.7591623036649214,
      "grad_norm": 15.672539710998535,
      "learning_rate": 1.9235953494976786e-05,
      "loss": 1.0271,
      "step": 580
    },
    {
      "epoch": 0.7643979057591623,
      "grad_norm": 9.93994140625,
      "learning_rate": 1.922289754977385e-05,
      "loss": 1.0102,
      "step": 584
    },
    {
      "epoch": 0.7696335078534031,
      "grad_norm": 11.041925430297852,
      "learning_rate": 1.920973550581989e-05,
      "loss": 0.9714,
      "step": 588
    },
    {
      "epoch": 0.774869109947644,
      "grad_norm": 7.117118835449219,
      "learning_rate": 1.9196467514528973e-05,
      "loss": 0.9632,
      "step": 592
    },
    {
      "epoch": 0.7801047120418848,
      "grad_norm": 8.919998168945312,
      "learning_rate": 1.9183093728533966e-05,
      "loss": 0.9669,
      "step": 596
    },
    {
      "epoch": 0.7853403141361257,
      "grad_norm": 10.06571102142334,
      "learning_rate": 1.9169614301684786e-05,
      "loss": 1.0159,
      "step": 600
    },
    {
      "epoch": 0.7853403141361257,
      "eval_F1_err_corr": 0.8977009019045425,
      "eval_accuracy": 0.8963959783449342,
      "eval_correct_accuracy": 0.9496939887354193,
      "eval_error_accuracy": 0.8511052713135295,
      "eval_f1": 0.6557006271203866,
      "eval_loss": 0.2661343812942505,
      "eval_pr_auc": 0.7220293951604384,
      "eval_precision": 0.6005649717514124,
      "eval_recall": 0.7219832465474304,
      "eval_runtime": 73.2795,
      "eval_samples_per_second": 66.772,
      "eval_steps_per_second": 0.532,
      "step": 600
    },
    {
      "epoch": 0.7905759162303665,
      "grad_norm": 8.285188674926758,
      "learning_rate": 1.915602938904662e-05,
      "loss": 1.0173,
      "step": 604
    },
    {
      "epoch": 0.7958115183246073,
      "grad_norm": 7.242547035217285,
      "learning_rate": 1.914233914689815e-05,
      "loss": 0.9253,
      "step": 608
    },
    {
      "epoch": 0.8010471204188482,
      "grad_norm": 8.031044960021973,
      "learning_rate": 1.912854373272975e-05,
      "loss": 0.9148,
      "step": 612
    },
    {
      "epoch": 0.806282722513089,
      "grad_norm": 11.873749732971191,
      "learning_rate": 1.9114643305241678e-05,
      "loss": 0.9732,
      "step": 616
    },
    {
      "epoch": 0.8115183246073299,
      "grad_norm": 9.597413063049316,
      "learning_rate": 1.9100638024342245e-05,
      "loss": 0.9687,
      "step": 620
    },
    {
      "epoch": 0.8167539267015707,
      "grad_norm": 8.924697875976562,
      "learning_rate": 1.908652805114598e-05,
      "loss": 0.9346,
      "step": 624
    },
    {
      "epoch": 0.8219895287958116,
      "grad_norm": 7.8740081787109375,
      "learning_rate": 1.907231354797179e-05,
      "loss": 0.916,
      "step": 628
    },
    {
      "epoch": 0.8272251308900523,
      "grad_norm": 8.93918514251709,
      "learning_rate": 1.9057994678341053e-05,
      "loss": 0.9274,
      "step": 632
    },
    {
      "epoch": 0.8324607329842932,
      "grad_norm": 8.441919326782227,
      "learning_rate": 1.9043571606975776e-05,
      "loss": 0.9868,
      "step": 636
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 7.841855049133301,
      "learning_rate": 1.902904449979669e-05,
      "loss": 0.8763,
      "step": 640
    },
    {
      "epoch": 0.837696335078534,
      "eval_F1_err_corr": 0.9048139002664162,
      "eval_accuracy": 0.9069450889404486,
      "eval_correct_accuracy": 0.9629414818645937,
      "eval_error_accuracy": 0.8533045110380948,
      "eval_f1": 0.6676977463543968,
      "eval_loss": 0.24821153283119202,
      "eval_pr_auc": 0.7226788592887777,
      "eval_precision": 0.6519956850053937,
      "eval_recall": 0.6841747792619425,
      "eval_runtime": 73.2534,
      "eval_samples_per_second": 66.796,
      "eval_steps_per_second": 0.532,
      "step": 640
    },
    {
      "epoch": 0.8429319371727748,
      "grad_norm": 8.774613380432129,
      "learning_rate": 1.901441352392133e-05,
      "loss": 0.9297,
      "step": 644
    },
    {
      "epoch": 0.8481675392670157,
      "grad_norm": 7.270528316497803,
      "learning_rate": 1.8999678847662124e-05,
      "loss": 0.961,
      "step": 648
    },
    {
      "epoch": 0.8534031413612565,
      "grad_norm": 9.619499206542969,
      "learning_rate": 1.8984840640524445e-05,
      "loss": 0.8385,
      "step": 652
    },
    {
      "epoch": 0.8586387434554974,
      "grad_norm": 9.912280082702637,
      "learning_rate": 1.8969899073204687e-05,
      "loss": 0.8858,
      "step": 656
    },
    {
      "epoch": 0.8638743455497382,
      "grad_norm": 7.621968746185303,
      "learning_rate": 1.8954854317588262e-05,
      "loss": 0.9501,
      "step": 660
    },
    {
      "epoch": 0.8691099476439791,
      "grad_norm": 6.877166271209717,
      "learning_rate": 1.8939706546747656e-05,
      "loss": 0.8896,
      "step": 664
    },
    {
      "epoch": 0.8743455497382199,
      "grad_norm": 12.053464889526367,
      "learning_rate": 1.8924455934940424e-05,
      "loss": 0.8812,
      "step": 668
    },
    {
      "epoch": 0.8795811518324608,
      "grad_norm": 8.33543872833252,
      "learning_rate": 1.8909102657607182e-05,
      "loss": 0.8902,
      "step": 672
    },
    {
      "epoch": 0.8848167539267016,
      "grad_norm": 10.474023818969727,
      "learning_rate": 1.88936468913696e-05,
      "loss": 0.8269,
      "step": 676
    },
    {
      "epoch": 0.8900523560209425,
      "grad_norm": 6.880638122558594,
      "learning_rate": 1.8878088814028365e-05,
      "loss": 0.8485,
      "step": 680
    },
    {
      "epoch": 0.8900523560209425,
      "eval_F1_err_corr": 0.9059432251011823,
      "eval_accuracy": 0.9063882443928848,
      "eval_correct_accuracy": 0.9703779177060071,
      "eval_error_accuracy": 0.8495328444726701,
      "eval_f1": 0.6530612244897959,
      "eval_loss": 0.24516551196575165,
      "eval_pr_auc": 0.7134608408684431,
      "eval_precision": 0.6615563298490128,
      "eval_recall": 0.6447815259225719,
      "eval_runtime": 73.2495,
      "eval_samples_per_second": 66.799,
      "eval_steps_per_second": 0.532,
      "step": 680
    },
    {
      "epoch": 0.8952879581151832,
      "grad_norm": 6.686065673828125,
      "learning_rate": 1.886242860456113e-05,
      "loss": 0.8777,
      "step": 684
    },
    {
      "epoch": 0.900523560209424,
      "grad_norm": 7.866757869720459,
      "learning_rate": 1.884666644312046e-05,
      "loss": 0.8438,
      "step": 688
    },
    {
      "epoch": 0.9057591623036649,
      "grad_norm": 8.994976043701172,
      "learning_rate": 1.8830802511031763e-05,
      "loss": 0.8494,
      "step": 692
    },
    {
      "epoch": 0.9109947643979057,
      "grad_norm": 11.257708549499512,
      "learning_rate": 1.88148369907912e-05,
      "loss": 0.8245,
      "step": 696
    },
    {
      "epoch": 0.9162303664921466,
      "grad_norm": 7.707327365875244,
      "learning_rate": 1.8798770066063577e-05,
      "loss": 0.8057,
      "step": 700
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 11.17904281616211,
      "learning_rate": 1.8782601921680258e-05,
      "loss": 0.89,
      "step": 704
    },
    {
      "epoch": 0.9267015706806283,
      "grad_norm": 8.393304824829102,
      "learning_rate": 1.8766332743637002e-05,
      "loss": 0.776,
      "step": 708
    },
    {
      "epoch": 0.9319371727748691,
      "grad_norm": 10.385251998901367,
      "learning_rate": 1.8749962719091864e-05,
      "loss": 0.8372,
      "step": 712
    },
    {
      "epoch": 0.93717277486911,
      "grad_norm": 7.805019855499268,
      "learning_rate": 1.8733492036363007e-05,
      "loss": 0.826,
      "step": 716
    },
    {
      "epoch": 0.9424083769633508,
      "grad_norm": 7.322630405426025,
      "learning_rate": 1.871692088492655e-05,
      "loss": 0.8387,
      "step": 720
    },
    {
      "epoch": 0.9424083769633508,
      "eval_F1_err_corr": 0.9048057208808482,
      "eval_accuracy": 0.9011291569992266,
      "eval_correct_accuracy": 0.9635766222061855,
      "eval_error_accuracy": 0.8527918542488826,
      "eval_f1": 0.6579623287671232,
      "eval_loss": 0.25547730922698975,
      "eval_pr_auc": 0.7115432250363661,
      "eval_precision": 0.6239090724578851,
      "eval_recall": 0.6959474756622142,
      "eval_runtime": 73.2576,
      "eval_samples_per_second": 66.792,
      "eval_steps_per_second": 0.532,
      "step": 720
    },
    {
      "epoch": 0.9476439790575916,
      "grad_norm": 8.483357429504395,
      "learning_rate": 1.8700249455414394e-05,
      "loss": 0.796,
      "step": 724
    },
    {
      "epoch": 0.9528795811518325,
      "grad_norm": 7.0828423500061035,
      "learning_rate": 1.8683477939612024e-05,
      "loss": 0.7919,
      "step": 728
    },
    {
      "epoch": 0.9581151832460733,
      "grad_norm": 7.401891231536865,
      "learning_rate": 1.866660653045629e-05,
      "loss": 0.7811,
      "step": 732
    },
    {
      "epoch": 0.9633507853403142,
      "grad_norm": 7.2336273193359375,
      "learning_rate": 1.8649635422033218e-05,
      "loss": 0.7897,
      "step": 736
    },
    {
      "epoch": 0.9685863874345549,
      "grad_norm": 8.54906940460205,
      "learning_rate": 1.863256480957574e-05,
      "loss": 0.7877,
      "step": 740
    },
    {
      "epoch": 0.9738219895287958,
      "grad_norm": 6.458954334259033,
      "learning_rate": 1.861539488946148e-05,
      "loss": 0.8188,
      "step": 744
    },
    {
      "epoch": 0.9790575916230366,
      "grad_norm": 6.701814651489258,
      "learning_rate": 1.8598125859210475e-05,
      "loss": 0.8441,
      "step": 748
    },
    {
      "epoch": 0.9842931937172775,
      "grad_norm": 9.568708419799805,
      "learning_rate": 1.858075791748291e-05,
      "loss": 0.8249,
      "step": 752
    },
    {
      "epoch": 0.9895287958115183,
      "grad_norm": 7.8331170082092285,
      "learning_rate": 1.8563291264076834e-05,
      "loss": 0.7904,
      "step": 756
    },
    {
      "epoch": 0.9947643979057592,
      "grad_norm": 8.621946334838867,
      "learning_rate": 1.854572609992586e-05,
      "loss": 0.7826,
      "step": 760
    },
    {
      "epoch": 0.9947643979057592,
      "eval_F1_err_corr": 0.9040361144490907,
      "eval_accuracy": 0.9081515854601702,
      "eval_correct_accuracy": 0.9695169328815136,
      "eval_error_accuracy": 0.8468407972694707,
      "eval_f1": 0.6513211978860833,
      "eval_loss": 0.24031765758991241,
      "eval_pr_auc": 0.7155756128338253,
      "eval_precision": 0.6766715470961444,
      "eval_recall": 0.627801675345257,
      "eval_runtime": 73.2204,
      "eval_samples_per_second": 66.826,
      "eval_steps_per_second": 0.533,
      "step": 760
    }
  ],
  "logging_steps": 4,
  "max_steps": 3820,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 16,
  "stateful_callbacks": {
    "MinEpochEarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5127896664393646e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}