{
  "best_metric": 1.0474289655685425,
  "best_model_checkpoint": "squarerun_large_model/checkpoint-261",
  "epoch": 25.0,
  "eval_steps": 500,
  "global_step": 725,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 16.904699325561523,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 2.0161,
      "step": 2
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 7.744546413421631,
      "learning_rate": 5.479452054794521e-06,
      "loss": 1.9714,
      "step": 4
    },
    {
      "epoch": 0.20689655172413793,
      "grad_norm": 11.83096694946289,
      "learning_rate": 8.21917808219178e-06,
      "loss": 2.0864,
      "step": 6
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 8.980508804321289,
      "learning_rate": 1.0958904109589042e-05,
      "loss": 1.9544,
      "step": 8
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 11.756573677062988,
      "learning_rate": 1.3698630136986302e-05,
      "loss": 1.872,
      "step": 10
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 12.744009971618652,
      "learning_rate": 1.643835616438356e-05,
      "loss": 2.128,
      "step": 12
    },
    {
      "epoch": 0.4827586206896552,
      "grad_norm": 11.320892333984375,
      "learning_rate": 1.9178082191780822e-05,
      "loss": 1.9613,
      "step": 14
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 7.5296406745910645,
      "learning_rate": 2.1917808219178083e-05,
      "loss": 2.0163,
      "step": 16
    },
    {
      "epoch": 0.6206896551724138,
      "grad_norm": 10.615029335021973,
      "learning_rate": 2.4657534246575342e-05,
      "loss": 1.8878,
      "step": 18
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 9.413975715637207,
      "learning_rate": 2.7397260273972603e-05,
      "loss": 1.9238,
      "step": 20
    },
    {
      "epoch": 0.7586206896551724,
      "grad_norm": 7.4806437492370605,
      "learning_rate": 3.0136986301369862e-05,
      "loss": 1.9426,
      "step": 22
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 4.6118035316467285,
      "learning_rate": 3.287671232876712e-05,
      "loss": 1.8594,
      "step": 24
    },
    {
      "epoch": 0.896551724137931,
      "grad_norm": 9.567954063415527,
      "learning_rate": 3.561643835616438e-05,
      "loss": 1.762,
      "step": 26
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 7.831980228424072,
      "learning_rate": 3.8356164383561644e-05,
      "loss": 1.917,
      "step": 28
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2196969696969697,
      "eval_f1_macro": 0.10662030635038826,
      "eval_f1_micro": 0.2196969696969697,
      "eval_f1_weighted": 0.12729797039644397,
      "eval_loss": 1.9114803075790405,
      "eval_precision_macro": 0.07802017358667604,
      "eval_precision_micro": 0.2196969696969697,
      "eval_precision_weighted": 0.09233343285067422,
      "eval_recall_macro": 0.18317460317460316,
      "eval_recall_micro": 0.2196969696969697,
      "eval_recall_weighted": 0.2196969696969697,
      "eval_runtime": 2.9641,
      "eval_samples_per_second": 44.533,
      "eval_steps_per_second": 5.735,
      "step": 29
    },
    {
      "epoch": 1.0344827586206897,
      "grad_norm": 11.95688533782959,
      "learning_rate": 4.1095890410958905e-05,
      "loss": 1.9276,
      "step": 30
    },
    {
      "epoch": 1.103448275862069,
      "grad_norm": 7.446818828582764,
      "learning_rate": 4.383561643835617e-05,
      "loss": 1.7188,
      "step": 32
    },
    {
      "epoch": 1.1724137931034484,
      "grad_norm": 7.449117183685303,
      "learning_rate": 4.657534246575342e-05,
      "loss": 1.7229,
      "step": 34
    },
    {
      "epoch": 1.2413793103448276,
      "grad_norm": 9.84091567993164,
      "learning_rate": 4.9315068493150684e-05,
      "loss": 2.0518,
      "step": 36
    },
    {
      "epoch": 1.3103448275862069,
      "grad_norm": 9.748187065124512,
      "learning_rate": 5.2054794520547945e-05,
      "loss": 1.838,
      "step": 38
    },
    {
      "epoch": 1.3793103448275863,
      "grad_norm": 10.7653226852417,
      "learning_rate": 5.479452054794521e-05,
      "loss": 1.9796,
      "step": 40
    },
    {
      "epoch": 1.4482758620689655,
      "grad_norm": 8.703463554382324,
      "learning_rate": 5.753424657534247e-05,
      "loss": 1.8676,
      "step": 42
    },
    {
      "epoch": 1.5172413793103448,
      "grad_norm": 8.969353675842285,
      "learning_rate": 6.0273972602739724e-05,
      "loss": 1.6242,
      "step": 44
    },
    {
      "epoch": 1.5862068965517242,
      "grad_norm": 6.287548542022705,
      "learning_rate": 6.301369863013699e-05,
      "loss": 1.6454,
      "step": 46
    },
    {
      "epoch": 1.6551724137931034,
      "grad_norm": 6.384350776672363,
      "learning_rate": 6.575342465753424e-05,
      "loss": 1.7246,
      "step": 48
    },
    {
      "epoch": 1.7241379310344827,
      "grad_norm": 8.735137939453125,
      "learning_rate": 6.84931506849315e-05,
      "loss": 1.8569,
      "step": 50
    },
    {
      "epoch": 1.793103448275862,
      "grad_norm": 7.929757118225098,
      "learning_rate": 7.123287671232876e-05,
      "loss": 1.6627,
      "step": 52
    },
    {
      "epoch": 1.8620689655172413,
      "grad_norm": 9.604327201843262,
      "learning_rate": 7.397260273972603e-05,
      "loss": 1.7562,
      "step": 54
    },
    {
      "epoch": 1.9310344827586206,
      "grad_norm": 8.560978889465332,
      "learning_rate": 7.671232876712329e-05,
      "loss": 1.5444,
      "step": 56
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.442503929138184,
      "learning_rate": 7.945205479452055e-05,
      "loss": 1.6762,
      "step": 58
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3560606060606061,
      "eval_f1_macro": 0.27325003502368067,
      "eval_f1_micro": 0.3560606060606061,
      "eval_f1_weighted": 0.30047625697016983,
      "eval_loss": 1.6721538305282593,
      "eval_precision_macro": 0.31412770236299653,
      "eval_precision_micro": 0.3560606060606061,
      "eval_precision_weighted": 0.368357405122111,
      "eval_recall_macro": 0.33546485260770975,
      "eval_recall_micro": 0.3560606060606061,
      "eval_recall_weighted": 0.3560606060606061,
      "eval_runtime": 3.2765,
      "eval_samples_per_second": 40.287,
      "eval_steps_per_second": 5.189,
      "step": 58
    },
    {
      "epoch": 2.0689655172413794,
      "grad_norm": 10.002808570861816,
      "learning_rate": 8.219178082191781e-05,
      "loss": 1.6511,
      "step": 60
    },
    {
      "epoch": 2.1379310344827585,
      "grad_norm": 32.81317901611328,
      "learning_rate": 8.493150684931507e-05,
      "loss": 1.5311,
      "step": 62
    },
    {
      "epoch": 2.206896551724138,
      "grad_norm": 11.362375259399414,
      "learning_rate": 8.767123287671233e-05,
      "loss": 1.5404,
      "step": 64
    },
    {
      "epoch": 2.2758620689655173,
      "grad_norm": 15.569432258605957,
      "learning_rate": 9.041095890410958e-05,
      "loss": 1.7608,
      "step": 66
    },
    {
      "epoch": 2.344827586206897,
      "grad_norm": 8.356627464294434,
      "learning_rate": 9.315068493150684e-05,
      "loss": 1.6792,
      "step": 68
    },
    {
      "epoch": 2.413793103448276,
      "grad_norm": 9.728259086608887,
      "learning_rate": 9.58904109589041e-05,
      "loss": 1.7309,
      "step": 70
    },
    {
      "epoch": 2.4827586206896552,
      "grad_norm": 8.866983413696289,
      "learning_rate": 9.863013698630137e-05,
      "loss": 1.8618,
      "step": 72
    },
    {
      "epoch": 2.5517241379310347,
      "grad_norm": 10.20724105834961,
      "learning_rate": 9.984662576687117e-05,
      "loss": 1.4866,
      "step": 74
    },
    {
      "epoch": 2.6206896551724137,
      "grad_norm": 10.527627944946289,
      "learning_rate": 9.95398773006135e-05,
      "loss": 1.7936,
      "step": 76
    },
    {
      "epoch": 2.689655172413793,
      "grad_norm": 5.849502086639404,
      "learning_rate": 9.923312883435584e-05,
      "loss": 1.7227,
      "step": 78
    },
    {
      "epoch": 2.7586206896551726,
      "grad_norm": 9.242766380310059,
      "learning_rate": 9.892638036809816e-05,
      "loss": 1.7347,
      "step": 80
    },
    {
      "epoch": 2.8275862068965516,
      "grad_norm": 9.303982734680176,
      "learning_rate": 9.861963190184049e-05,
      "loss": 1.5123,
      "step": 82
    },
    {
      "epoch": 2.896551724137931,
      "grad_norm": 7.220952987670898,
      "learning_rate": 9.831288343558283e-05,
      "loss": 1.9009,
      "step": 84
    },
    {
      "epoch": 2.9655172413793105,
      "grad_norm": 8.945652961730957,
      "learning_rate": 9.800613496932515e-05,
      "loss": 1.9664,
      "step": 86
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.45454545454545453,
      "eval_f1_macro": 0.35540571944322835,
      "eval_f1_micro": 0.45454545454545453,
      "eval_f1_weighted": 0.4060242434737366,
      "eval_loss": 1.5057227611541748,
      "eval_precision_macro": 0.3733890330325679,
      "eval_precision_micro": 0.45454545454545453,
      "eval_precision_weighted": 0.41290912992763085,
      "eval_recall_macro": 0.3856538170823885,
      "eval_recall_micro": 0.45454545454545453,
      "eval_recall_weighted": 0.45454545454545453,
      "eval_runtime": 2.9982,
      "eval_samples_per_second": 44.026,
      "eval_steps_per_second": 5.67,
      "step": 87
    },
    {
      "epoch": 3.0344827586206895,
      "grad_norm": 7.55537223815918,
      "learning_rate": 9.76993865030675e-05,
      "loss": 1.3214,
      "step": 88
    },
    {
      "epoch": 3.103448275862069,
      "grad_norm": 6.456240177154541,
      "learning_rate": 9.739263803680982e-05,
      "loss": 1.2262,
      "step": 90
    },
    {
      "epoch": 3.1724137931034484,
      "grad_norm": 10.980310440063477,
      "learning_rate": 9.708588957055215e-05,
      "loss": 1.6514,
      "step": 92
    },
    {
      "epoch": 3.2413793103448274,
      "grad_norm": 10.584870338439941,
      "learning_rate": 9.677914110429448e-05,
      "loss": 1.464,
      "step": 94
    },
    {
      "epoch": 3.310344827586207,
      "grad_norm": 7.440125465393066,
      "learning_rate": 9.647239263803681e-05,
      "loss": 1.2235,
      "step": 96
    },
    {
      "epoch": 3.3793103448275863,
      "grad_norm": 9.703545570373535,
      "learning_rate": 9.616564417177915e-05,
      "loss": 1.7171,
      "step": 98
    },
    {
      "epoch": 3.4482758620689653,
      "grad_norm": 7.038817882537842,
      "learning_rate": 9.585889570552147e-05,
      "loss": 1.3095,
      "step": 100
    },
    {
      "epoch": 3.5172413793103448,
      "grad_norm": 11.278383255004883,
      "learning_rate": 9.555214723926381e-05,
      "loss": 1.6106,
      "step": 102
    },
    {
      "epoch": 3.586206896551724,
      "grad_norm": 7.141289710998535,
      "learning_rate": 9.524539877300614e-05,
      "loss": 1.0537,
      "step": 104
    },
    {
      "epoch": 3.655172413793103,
      "grad_norm": 6.102043151855469,
      "learning_rate": 9.493865030674846e-05,
      "loss": 1.2514,
      "step": 106
    },
    {
      "epoch": 3.7241379310344827,
      "grad_norm": 9.988530158996582,
      "learning_rate": 9.46319018404908e-05,
      "loss": 1.1408,
      "step": 108
    },
    {
      "epoch": 3.793103448275862,
      "grad_norm": 7.49151611328125,
      "learning_rate": 9.432515337423313e-05,
      "loss": 1.5343,
      "step": 110
    },
    {
      "epoch": 3.862068965517241,
      "grad_norm": 13.044901847839355,
      "learning_rate": 9.401840490797547e-05,
      "loss": 1.9732,
      "step": 112
    },
    {
      "epoch": 3.9310344827586206,
      "grad_norm": 10.967622756958008,
      "learning_rate": 9.37116564417178e-05,
      "loss": 1.4529,
      "step": 114
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.224091529846191,
      "learning_rate": 9.340490797546013e-05,
      "loss": 1.1934,
      "step": 116
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4090909090909091,
      "eval_f1_macro": 0.3129633381731382,
      "eval_f1_micro": 0.4090909090909091,
      "eval_f1_weighted": 0.3530157221067592,
      "eval_loss": 1.4216643571853638,
      "eval_precision_macro": 0.34140786749482405,
      "eval_precision_micro": 0.4090909090909091,
      "eval_precision_weighted": 0.3818291611769873,
      "eval_recall_macro": 0.3628798185941043,
      "eval_recall_micro": 0.4090909090909091,
      "eval_recall_weighted": 0.4090909090909091,
      "eval_runtime": 2.9654,
      "eval_samples_per_second": 44.513,
      "eval_steps_per_second": 5.733,
      "step": 116
    },
    {
      "epoch": 4.068965517241379,
      "grad_norm": 10.413983345031738,
      "learning_rate": 9.309815950920246e-05,
      "loss": 1.4143,
      "step": 118
    },
    {
      "epoch": 4.137931034482759,
      "grad_norm": 6.670328617095947,
      "learning_rate": 9.279141104294478e-05,
      "loss": 1.1382,
      "step": 120
    },
    {
      "epoch": 4.206896551724138,
      "grad_norm": 7.599217414855957,
      "learning_rate": 9.248466257668712e-05,
      "loss": 1.291,
      "step": 122
    },
    {
      "epoch": 4.275862068965517,
      "grad_norm": 9.762212753295898,
      "learning_rate": 9.217791411042945e-05,
      "loss": 1.2167,
      "step": 124
    },
    {
      "epoch": 4.344827586206897,
      "grad_norm": 7.603570938110352,
      "learning_rate": 9.187116564417179e-05,
      "loss": 1.1758,
      "step": 126
    },
    {
      "epoch": 4.413793103448276,
      "grad_norm": 6.104551315307617,
      "learning_rate": 9.156441717791411e-05,
      "loss": 1.1649,
      "step": 128
    },
    {
      "epoch": 4.482758620689655,
      "grad_norm": 10.374088287353516,
      "learning_rate": 9.125766871165644e-05,
      "loss": 1.3607,
      "step": 130
    },
    {
      "epoch": 4.551724137931035,
      "grad_norm": 9.6392822265625,
      "learning_rate": 9.095092024539878e-05,
      "loss": 1.6339,
      "step": 132
    },
    {
      "epoch": 4.620689655172414,
      "grad_norm": 7.82922887802124,
      "learning_rate": 9.06441717791411e-05,
      "loss": 1.8073,
      "step": 134
    },
    {
      "epoch": 4.689655172413794,
      "grad_norm": 9.731952667236328,
      "learning_rate": 9.033742331288344e-05,
      "loss": 1.3304,
      "step": 136
    },
    {
      "epoch": 4.758620689655173,
      "grad_norm": 6.175458908081055,
      "learning_rate": 9.003067484662577e-05,
      "loss": 1.1931,
      "step": 138
    },
    {
      "epoch": 4.827586206896552,
      "grad_norm": 4.451329231262207,
      "learning_rate": 8.972392638036811e-05,
      "loss": 1.0355,
      "step": 140
    },
    {
      "epoch": 4.896551724137931,
      "grad_norm": 6.266848087310791,
      "learning_rate": 8.941717791411043e-05,
      "loss": 1.2382,
      "step": 142
    },
    {
      "epoch": 4.9655172413793105,
      "grad_norm": 5.750278949737549,
      "learning_rate": 8.911042944785276e-05,
      "loss": 1.0968,
      "step": 144
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5757575757575758,
      "eval_f1_macro": 0.4607554427944924,
      "eval_f1_micro": 0.5757575757575758,
      "eval_f1_weighted": 0.525821869552332,
      "eval_loss": 1.1879011392593384,
      "eval_precision_macro": 0.4807155657962109,
      "eval_precision_micro": 0.5757575757575758,
      "eval_precision_weighted": 0.5437836021505377,
      "eval_recall_macro": 0.5045275888133031,
      "eval_recall_micro": 0.5757575757575758,
      "eval_recall_weighted": 0.5757575757575758,
      "eval_runtime": 3.2227,
      "eval_samples_per_second": 40.96,
      "eval_steps_per_second": 5.275,
      "step": 145
    },
    {
      "epoch": 5.0344827586206895,
      "grad_norm": 7.826033592224121,
      "learning_rate": 8.88036809815951e-05,
      "loss": 1.5289,
      "step": 146
    },
    {
      "epoch": 5.103448275862069,
      "grad_norm": 11.248700141906738,
      "learning_rate": 8.849693251533742e-05,
      "loss": 1.0666,
      "step": 148
    },
    {
      "epoch": 5.172413793103448,
      "grad_norm": 6.100868225097656,
      "learning_rate": 8.819018404907976e-05,
      "loss": 0.9321,
      "step": 150
    },
    {
      "epoch": 5.241379310344827,
      "grad_norm": 5.934208393096924,
      "learning_rate": 8.788343558282209e-05,
      "loss": 0.8063,
      "step": 152
    },
    {
      "epoch": 5.310344827586207,
      "grad_norm": 6.6341471672058105,
      "learning_rate": 8.757668711656443e-05,
      "loss": 1.1262,
      "step": 154
    },
    {
      "epoch": 5.379310344827586,
      "grad_norm": 6.946788311004639,
      "learning_rate": 8.726993865030675e-05,
      "loss": 0.957,
      "step": 156
    },
    {
      "epoch": 5.448275862068965,
      "grad_norm": 11.869194030761719,
      "learning_rate": 8.696319018404908e-05,
      "loss": 1.6966,
      "step": 158
    },
    {
      "epoch": 5.517241379310345,
      "grad_norm": 8.566763877868652,
      "learning_rate": 8.665644171779142e-05,
      "loss": 1.222,
      "step": 160
    },
    {
      "epoch": 5.586206896551724,
      "grad_norm": 8.25841999053955,
      "learning_rate": 8.634969325153374e-05,
      "loss": 1.4671,
      "step": 162
    },
    {
      "epoch": 5.655172413793103,
      "grad_norm": 4.913756370544434,
      "learning_rate": 8.604294478527608e-05,
      "loss": 1.07,
      "step": 164
    },
    {
      "epoch": 5.724137931034483,
      "grad_norm": 7.771355152130127,
      "learning_rate": 8.573619631901841e-05,
      "loss": 0.8391,
      "step": 166
    },
    {
      "epoch": 5.793103448275862,
      "grad_norm": 8.31480598449707,
      "learning_rate": 8.542944785276073e-05,
      "loss": 1.311,
      "step": 168
    },
    {
      "epoch": 5.862068965517241,
      "grad_norm": 14.048528671264648,
      "learning_rate": 8.512269938650307e-05,
      "loss": 1.4462,
      "step": 170
    },
    {
      "epoch": 5.931034482758621,
      "grad_norm": 7.521085739135742,
      "learning_rate": 8.48159509202454e-05,
      "loss": 1.2066,
      "step": 172
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.221696376800537,
      "learning_rate": 8.450920245398774e-05,
      "loss": 1.1313,
      "step": 174
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.553030303030303,
      "eval_f1_macro": 0.4964328132779621,
      "eval_f1_micro": 0.553030303030303,
      "eval_f1_weighted": 0.5242598987794572,
      "eval_loss": 1.230722427368164,
      "eval_precision_macro": 0.5849505386679483,
      "eval_precision_micro": 0.553030303030303,
      "eval_precision_weighted": 0.6113855756896126,
      "eval_recall_macro": 0.5196371882086168,
      "eval_recall_micro": 0.553030303030303,
      "eval_recall_weighted": 0.553030303030303,
      "eval_runtime": 2.9624,
      "eval_samples_per_second": 44.559,
      "eval_steps_per_second": 5.739,
      "step": 174
    },
    {
      "epoch": 6.068965517241379,
      "grad_norm": 8.70207691192627,
      "learning_rate": 8.420245398773006e-05,
      "loss": 1.0758,
      "step": 176
    },
    {
      "epoch": 6.137931034482759,
      "grad_norm": 5.113523960113525,
      "learning_rate": 8.38957055214724e-05,
      "loss": 0.835,
      "step": 178
    },
    {
      "epoch": 6.206896551724138,
      "grad_norm": 8.031415939331055,
      "learning_rate": 8.358895705521473e-05,
      "loss": 0.9388,
      "step": 180
    },
    {
      "epoch": 6.275862068965517,
      "grad_norm": 6.924206256866455,
      "learning_rate": 8.328220858895705e-05,
      "loss": 1.2986,
      "step": 182
    },
    {
      "epoch": 6.344827586206897,
      "grad_norm": 8.165400505065918,
      "learning_rate": 8.297546012269939e-05,
      "loss": 0.9004,
      "step": 184
    },
    {
      "epoch": 6.413793103448276,
      "grad_norm": 8.674925804138184,
      "learning_rate": 8.266871165644172e-05,
      "loss": 0.8029,
      "step": 186
    },
    {
      "epoch": 6.482758620689655,
      "grad_norm": 7.9463114738464355,
      "learning_rate": 8.236196319018406e-05,
      "loss": 0.7277,
      "step": 188
    },
    {
      "epoch": 6.551724137931035,
      "grad_norm": 7.694469451904297,
      "learning_rate": 8.205521472392638e-05,
      "loss": 1.1082,
      "step": 190
    },
    {
      "epoch": 6.620689655172414,
      "grad_norm": 7.801606178283691,
      "learning_rate": 8.174846625766872e-05,
      "loss": 1.0687,
      "step": 192
    },
    {
      "epoch": 6.689655172413794,
      "grad_norm": 7.689370155334473,
      "learning_rate": 8.144171779141105e-05,
      "loss": 1.0604,
      "step": 194
    },
    {
      "epoch": 6.758620689655173,
      "grad_norm": 7.995850086212158,
      "learning_rate": 8.113496932515337e-05,
      "loss": 1.3311,
      "step": 196
    },
    {
      "epoch": 6.827586206896552,
      "grad_norm": 4.453602313995361,
      "learning_rate": 8.082822085889571e-05,
      "loss": 0.8307,
      "step": 198
    },
    {
      "epoch": 6.896551724137931,
      "grad_norm": 5.754306793212891,
      "learning_rate": 8.052147239263804e-05,
      "loss": 0.7514,
      "step": 200
    },
    {
      "epoch": 6.9655172413793105,
      "grad_norm": 6.210538387298584,
      "learning_rate": 8.021472392638038e-05,
      "loss": 1.0807,
      "step": 202
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5303030303030303,
      "eval_f1_macro": 0.4087798393817858,
      "eval_f1_micro": 0.5303030303030303,
      "eval_f1_weighted": 0.47721174256077536,
      "eval_loss": 1.2771176099777222,
      "eval_precision_macro": 0.5392650999793857,
      "eval_precision_micro": 0.5303030303030303,
      "eval_precision_weighted": 0.581616489571035,
      "eval_recall_macro": 0.4303930461073318,
      "eval_recall_micro": 0.5303030303030303,
      "eval_recall_weighted": 0.5303030303030303,
      "eval_runtime": 2.9556,
      "eval_samples_per_second": 44.661,
      "eval_steps_per_second": 5.752,
      "step": 203
    },
    {
      "epoch": 7.0344827586206895,
      "grad_norm": 5.119338512420654,
      "learning_rate": 7.99079754601227e-05,
      "loss": 0.962,
      "step": 204
    },
    {
      "epoch": 7.103448275862069,
      "grad_norm": 9.368791580200195,
      "learning_rate": 7.960122699386503e-05,
      "loss": 0.7753,
      "step": 206
    },
    {
      "epoch": 7.172413793103448,
      "grad_norm": 7.991408348083496,
      "learning_rate": 7.929447852760737e-05,
      "loss": 1.0809,
      "step": 208
    },
    {
      "epoch": 7.241379310344827,
      "grad_norm": 8.43641185760498,
      "learning_rate": 7.898773006134969e-05,
      "loss": 0.9518,
      "step": 210
    },
    {
      "epoch": 7.310344827586207,
      "grad_norm": 11.549813270568848,
      "learning_rate": 7.868098159509203e-05,
      "loss": 1.0382,
      "step": 212
    },
    {
      "epoch": 7.379310344827586,
      "grad_norm": 7.266906261444092,
      "learning_rate": 7.837423312883436e-05,
      "loss": 0.9945,
      "step": 214
    },
    {
      "epoch": 7.448275862068965,
      "grad_norm": 6.440887928009033,
      "learning_rate": 7.80674846625767e-05,
      "loss": 0.8706,
      "step": 216
    },
    {
      "epoch": 7.517241379310345,
      "grad_norm": 7.407945156097412,
      "learning_rate": 7.776073619631902e-05,
      "loss": 0.7985,
      "step": 218
    },
    {
      "epoch": 7.586206896551724,
      "grad_norm": 6.593641757965088,
      "learning_rate": 7.745398773006135e-05,
      "loss": 0.5974,
      "step": 220
    },
    {
      "epoch": 7.655172413793103,
      "grad_norm": 7.8406829833984375,
      "learning_rate": 7.714723926380369e-05,
      "loss": 0.8405,
      "step": 222
    },
    {
      "epoch": 7.724137931034483,
      "grad_norm": 9.805188179016113,
      "learning_rate": 7.684049079754601e-05,
      "loss": 1.177,
      "step": 224
    },
    {
      "epoch": 7.793103448275862,
      "grad_norm": 10.728581428527832,
      "learning_rate": 7.653374233128835e-05,
      "loss": 1.5151,
      "step": 226
    },
    {
      "epoch": 7.862068965517241,
      "grad_norm": 5.296911239624023,
      "learning_rate": 7.622699386503068e-05,
      "loss": 0.7184,
      "step": 228
    },
    {
      "epoch": 7.931034482758621,
      "grad_norm": 7.458691596984863,
      "learning_rate": 7.5920245398773e-05,
      "loss": 1.1221,
      "step": 230
    },
    {
      "epoch": 8.0,
      "grad_norm": 7.329806804656982,
      "learning_rate": 7.561349693251534e-05,
      "loss": 1.1825,
      "step": 232
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5681818181818182,
      "eval_f1_macro": 0.45282668788005603,
      "eval_f1_micro": 0.5681818181818182,
      "eval_f1_weighted": 0.517546575055234,
      "eval_loss": 1.2339295148849487,
      "eval_precision_macro": 0.5544132386496917,
      "eval_precision_micro": 0.5681818181818182,
      "eval_precision_weighted": 0.6168848132014276,
      "eval_recall_macro": 0.4919803476946334,
      "eval_recall_micro": 0.5681818181818182,
      "eval_recall_weighted": 0.5681818181818182,
      "eval_runtime": 3.0256,
      "eval_samples_per_second": 43.628,
      "eval_steps_per_second": 5.619,
      "step": 232
    },
    {
      "epoch": 8.068965517241379,
      "grad_norm": 12.180721282958984,
      "learning_rate": 7.530674846625767e-05,
      "loss": 0.9113,
      "step": 234
    },
    {
      "epoch": 8.137931034482758,
      "grad_norm": 4.219178199768066,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.8664,
      "step": 236
    },
    {
      "epoch": 8.206896551724139,
      "grad_norm": 7.4406304359436035,
      "learning_rate": 7.469325153374233e-05,
      "loss": 0.9639,
      "step": 238
    },
    {
      "epoch": 8.275862068965518,
      "grad_norm": 7.360535621643066,
      "learning_rate": 7.438650306748467e-05,
      "loss": 0.9976,
      "step": 240
    },
    {
      "epoch": 8.344827586206897,
      "grad_norm": 5.644505023956299,
      "learning_rate": 7.4079754601227e-05,
      "loss": 0.6907,
      "step": 242
    },
    {
      "epoch": 8.413793103448276,
      "grad_norm": 7.283102512359619,
      "learning_rate": 7.377300613496932e-05,
      "loss": 1.0068,
      "step": 244
    },
    {
      "epoch": 8.482758620689655,
      "grad_norm": 4.419933319091797,
      "learning_rate": 7.346625766871166e-05,
      "loss": 0.5921,
      "step": 246
    },
    {
      "epoch": 8.551724137931034,
      "grad_norm": 8.00572395324707,
      "learning_rate": 7.315950920245399e-05,
      "loss": 0.6962,
      "step": 248
    },
    {
      "epoch": 8.620689655172415,
      "grad_norm": 7.26607608795166,
      "learning_rate": 7.285276073619633e-05,
      "loss": 0.8763,
      "step": 250
    },
    {
      "epoch": 8.689655172413794,
      "grad_norm": 4.030153751373291,
      "learning_rate": 7.254601226993865e-05,
      "loss": 0.4548,
      "step": 252
    },
    {
      "epoch": 8.758620689655173,
      "grad_norm": 9.402144432067871,
      "learning_rate": 7.223926380368099e-05,
      "loss": 0.9571,
      "step": 254
    },
    {
      "epoch": 8.827586206896552,
      "grad_norm": 6.616663932800293,
      "learning_rate": 7.193251533742332e-05,
      "loss": 0.4974,
      "step": 256
    },
    {
      "epoch": 8.89655172413793,
      "grad_norm": 6.191842079162598,
      "learning_rate": 7.162576687116564e-05,
      "loss": 0.7361,
      "step": 258
    },
    {
      "epoch": 8.96551724137931,
      "grad_norm": 7.1751837730407715,
      "learning_rate": 7.131901840490798e-05,
      "loss": 0.4454,
      "step": 260
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.696969696969697,
      "eval_f1_macro": 0.6064190911129687,
      "eval_f1_micro": 0.696969696969697,
      "eval_f1_weighted": 0.6763255683710229,
      "eval_loss": 1.0474289655685425,
      "eval_precision_macro": 0.6334407447189402,
      "eval_precision_micro": 0.696969696969697,
      "eval_precision_weighted": 0.686842105263158,
      "eval_recall_macro": 0.6099622071050643,
      "eval_recall_micro": 0.696969696969697,
      "eval_recall_weighted": 0.696969696969697,
      "eval_runtime": 2.9691,
      "eval_samples_per_second": 44.457,
      "eval_steps_per_second": 5.726,
      "step": 261
    },
    {
      "epoch": 9.03448275862069,
      "grad_norm": 4.017953872680664,
      "learning_rate": 7.101226993865031e-05,
      "loss": 0.3868,
      "step": 262
    },
    {
      "epoch": 9.10344827586207,
      "grad_norm": 8.303009986877441,
      "learning_rate": 7.070552147239265e-05,
      "loss": 0.5722,
      "step": 264
    },
    {
      "epoch": 9.172413793103448,
      "grad_norm": 16.19445037841797,
      "learning_rate": 7.039877300613497e-05,
      "loss": 0.8712,
      "step": 266
    },
    {
      "epoch": 9.241379310344827,
      "grad_norm": 6.667177677154541,
      "learning_rate": 7.00920245398773e-05,
      "loss": 0.4407,
      "step": 268
    },
    {
      "epoch": 9.310344827586206,
      "grad_norm": 5.363644599914551,
      "learning_rate": 6.978527607361964e-05,
      "loss": 0.3215,
      "step": 270
    },
    {
      "epoch": 9.379310344827585,
      "grad_norm": 12.17090892791748,
      "learning_rate": 6.947852760736196e-05,
      "loss": 0.489,
      "step": 272
    },
    {
      "epoch": 9.448275862068966,
      "grad_norm": 10.571720123291016,
      "learning_rate": 6.91717791411043e-05,
      "loss": 0.8153,
      "step": 274
    },
    {
      "epoch": 9.517241379310345,
      "grad_norm": 6.873518466949463,
      "learning_rate": 6.886503067484663e-05,
      "loss": 0.379,
      "step": 276
    },
    {
      "epoch": 9.586206896551724,
      "grad_norm": 3.1050946712493896,
      "learning_rate": 6.855828220858897e-05,
      "loss": 0.168,
      "step": 278
    },
    {
      "epoch": 9.655172413793103,
      "grad_norm": 3.549830675125122,
      "learning_rate": 6.825153374233129e-05,
      "loss": 0.1496,
      "step": 280
    },
    {
      "epoch": 9.724137931034482,
      "grad_norm": 7.51619815826416,
      "learning_rate": 6.794478527607362e-05,
      "loss": 0.7824,
      "step": 282
    },
    {
      "epoch": 9.793103448275861,
      "grad_norm": 9.865971565246582,
      "learning_rate": 6.763803680981596e-05,
      "loss": 0.9922,
      "step": 284
    },
    {
      "epoch": 9.862068965517242,
      "grad_norm": 11.25515079498291,
      "learning_rate": 6.733128834355828e-05,
      "loss": 1.0158,
      "step": 286
    },
    {
      "epoch": 9.931034482758621,
      "grad_norm": 6.325614929199219,
      "learning_rate": 6.702453987730062e-05,
      "loss": 0.4932,
      "step": 288
    },
    {
      "epoch": 10.0,
      "grad_norm": 6.320476531982422,
      "learning_rate": 6.671779141104295e-05,
      "loss": 0.5439,
      "step": 290
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5151515151515151,
      "eval_f1_macro": 0.4580164261227385,
      "eval_f1_micro": 0.5151515151515151,
      "eval_f1_weighted": 0.491986025706956,
      "eval_loss": 1.6814864873886108,
      "eval_precision_macro": 0.5393793994024408,
      "eval_precision_micro": 0.5151515151515151,
      "eval_precision_weighted": 0.595123648752681,
      "eval_recall_macro": 0.4903250188964474,
      "eval_recall_micro": 0.5151515151515151,
      "eval_recall_weighted": 0.5151515151515151,
      "eval_runtime": 2.9552,
      "eval_samples_per_second": 44.667,
      "eval_steps_per_second": 5.753,
      "step": 290
    },
    {
      "epoch": 10.068965517241379,
      "grad_norm": 19.37687110900879,
      "learning_rate": 6.641104294478529e-05,
      "loss": 1.1536,
      "step": 292
    },
    {
      "epoch": 10.137931034482758,
      "grad_norm": 5.3166093826293945,
      "learning_rate": 6.610429447852761e-05,
      "loss": 0.6399,
      "step": 294
    },
    {
      "epoch": 10.206896551724139,
      "grad_norm": 6.868032932281494,
      "learning_rate": 6.579754601226994e-05,
      "loss": 0.5945,
      "step": 296
    },
    {
      "epoch": 10.275862068965518,
      "grad_norm": 6.491347312927246,
      "learning_rate": 6.549079754601228e-05,
      "loss": 0.5558,
      "step": 298
    },
    {
      "epoch": 10.344827586206897,
      "grad_norm": 5.701662063598633,
      "learning_rate": 6.51840490797546e-05,
      "loss": 0.5752,
      "step": 300
    },
    {
      "epoch": 10.413793103448276,
      "grad_norm": 6.190051555633545,
      "learning_rate": 6.487730061349694e-05,
      "loss": 0.4731,
      "step": 302
    },
    {
      "epoch": 10.482758620689655,
      "grad_norm": 6.490513801574707,
      "learning_rate": 6.457055214723927e-05,
      "loss": 0.5012,
      "step": 304
    },
    {
      "epoch": 10.551724137931034,
      "grad_norm": 7.6167521476745605,
      "learning_rate": 6.426380368098159e-05,
      "loss": 0.553,
      "step": 306
    },
    {
      "epoch": 10.620689655172415,
      "grad_norm": 9.545734405517578,
      "learning_rate": 6.395705521472393e-05,
      "loss": 0.6079,
      "step": 308
    },
    {
      "epoch": 10.689655172413794,
      "grad_norm": 10.175183296203613,
      "learning_rate": 6.365030674846626e-05,
      "loss": 0.6955,
      "step": 310
    },
    {
      "epoch": 10.758620689655173,
      "grad_norm": 8.330876350402832,
      "learning_rate": 6.33435582822086e-05,
      "loss": 0.739,
      "step": 312
    },
    {
      "epoch": 10.827586206896552,
      "grad_norm": 9.028648376464844,
      "learning_rate": 6.303680981595092e-05,
      "loss": 0.4514,
      "step": 314
    },
    {
      "epoch": 10.89655172413793,
      "grad_norm": 6.639448642730713,
      "learning_rate": 6.273006134969326e-05,
      "loss": 0.2532,
      "step": 316
    },
    {
      "epoch": 10.96551724137931,
      "grad_norm": 6.314249515533447,
      "learning_rate": 6.242331288343559e-05,
      "loss": 0.4256,
      "step": 318
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1_macro": 0.5799728079445362,
      "eval_f1_micro": 0.6666666666666666,
      "eval_f1_weighted": 0.649504493507492,
      "eval_loss": 1.137831687927246,
      "eval_precision_macro": 0.5800879765395894,
      "eval_precision_micro": 0.6666666666666666,
      "eval_precision_weighted": 0.6434914393198851,
      "eval_recall_macro": 0.590650037792895,
      "eval_recall_micro": 0.6666666666666666,
      "eval_recall_weighted": 0.6666666666666666,
      "eval_runtime": 2.949,
      "eval_samples_per_second": 44.761,
      "eval_steps_per_second": 5.765,
      "step": 319
    },
    {
      "epoch": 11.03448275862069,
      "grad_norm": 5.468910217285156,
      "learning_rate": 6.211656441717791e-05,
      "loss": 0.4358,
      "step": 320
    },
    {
      "epoch": 11.10344827586207,
      "grad_norm": 4.487988471984863,
      "learning_rate": 6.180981595092025e-05,
      "loss": 0.349,
      "step": 322
    },
    {
      "epoch": 11.172413793103448,
      "grad_norm": 4.8094587326049805,
      "learning_rate": 6.150306748466258e-05,
      "loss": 0.4052,
      "step": 324
    },
    {
      "epoch": 11.241379310344827,
      "grad_norm": 2.387544870376587,
      "learning_rate": 6.119631901840492e-05,
      "loss": 0.2168,
      "step": 326
    },
    {
      "epoch": 11.310344827586206,
      "grad_norm": 7.197230815887451,
      "learning_rate": 6.088957055214725e-05,
      "loss": 0.2149,
      "step": 328
    },
    {
      "epoch": 11.379310344827585,
      "grad_norm": 4.119611740112305,
      "learning_rate": 6.058282208588958e-05,
      "loss": 0.3088,
      "step": 330
    },
    {
      "epoch": 11.448275862068966,
      "grad_norm": 8.908929824829102,
      "learning_rate": 6.02760736196319e-05,
      "loss": 0.4711,
      "step": 332
    },
    {
      "epoch": 11.517241379310345,
      "grad_norm": 7.788853168487549,
      "learning_rate": 5.996932515337423e-05,
      "loss": 0.2111,
      "step": 334
    },
    {
      "epoch": 11.586206896551724,
      "grad_norm": 4.6214118003845215,
      "learning_rate": 5.9662576687116564e-05,
      "loss": 0.4032,
      "step": 336
    },
    {
      "epoch": 11.655172413793103,
      "grad_norm": 12.485306739807129,
      "learning_rate": 5.93558282208589e-05,
      "loss": 0.6185,
      "step": 338
    },
    {
      "epoch": 11.724137931034482,
      "grad_norm": 8.764460563659668,
      "learning_rate": 5.9049079754601235e-05,
      "loss": 0.3864,
      "step": 340
    },
    {
      "epoch": 11.793103448275861,
      "grad_norm": 7.914090156555176,
      "learning_rate": 5.874233128834357e-05,
      "loss": 0.3361,
      "step": 342
    },
    {
      "epoch": 11.862068965517242,
      "grad_norm": 7.184711933135986,
      "learning_rate": 5.8435582822085886e-05,
      "loss": 0.3037,
      "step": 344
    },
    {
      "epoch": 11.931034482758621,
      "grad_norm": 5.275500297546387,
      "learning_rate": 5.812883435582822e-05,
      "loss": 0.2329,
      "step": 346
    },
    {
      "epoch": 12.0,
      "grad_norm": 9.601807594299316,
      "learning_rate": 5.782208588957055e-05,
      "loss": 0.4968,
      "step": 348
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6136363636363636,
      "eval_f1_macro": 0.5307113938692886,
      "eval_f1_micro": 0.6136363636363636,
      "eval_f1_weighted": 0.6012838636642465,
      "eval_loss": 1.4229304790496826,
      "eval_precision_macro": 0.5347963458887828,
      "eval_precision_micro": 0.6136363636363636,
      "eval_precision_weighted": 0.6095054001916747,
      "eval_recall_macro": 0.5485941043083901,
      "eval_recall_micro": 0.6136363636363636,
      "eval_recall_weighted": 0.6136363636363636,
      "eval_runtime": 2.9496,
      "eval_samples_per_second": 44.752,
      "eval_steps_per_second": 5.764,
      "step": 348
    },
    {
      "epoch": 12.068965517241379,
      "grad_norm": 3.669062614440918,
      "learning_rate": 5.751533742331289e-05,
      "loss": 0.2376,
      "step": 350
    },
    {
      "epoch": 12.137931034482758,
      "grad_norm": 7.027435302734375,
      "learning_rate": 5.720858895705522e-05,
      "loss": 0.3657,
      "step": 352
    },
    {
      "epoch": 12.206896551724139,
      "grad_norm": 5.393192768096924,
      "learning_rate": 5.6901840490797555e-05,
      "loss": 0.3012,
      "step": 354
    },
    {
      "epoch": 12.275862068965518,
      "grad_norm": 1.6986196041107178,
      "learning_rate": 5.6595092024539874e-05,
      "loss": 0.1062,
      "step": 356
    },
    {
      "epoch": 12.344827586206897,
      "grad_norm": 11.825511932373047,
      "learning_rate": 5.6288343558282206e-05,
      "loss": 0.5445,
      "step": 358
    },
    {
      "epoch": 12.413793103448276,
      "grad_norm": 11.926621437072754,
      "learning_rate": 5.598159509202454e-05,
      "loss": 0.8014,
      "step": 360
    },
    {
      "epoch": 12.482758620689655,
      "grad_norm": 7.5627665519714355,
      "learning_rate": 5.567484662576688e-05,
      "loss": 0.3761,
      "step": 362
    },
    {
      "epoch": 12.551724137931034,
      "grad_norm": 5.975190162658691,
      "learning_rate": 5.536809815950921e-05,
      "loss": 0.3764,
      "step": 364
    },
    {
      "epoch": 12.620689655172415,
      "grad_norm": 5.467086315155029,
      "learning_rate": 5.506134969325154e-05,
      "loss": 0.3437,
      "step": 366
    },
    {
      "epoch": 12.689655172413794,
      "grad_norm": 7.421685218811035,
      "learning_rate": 5.475460122699386e-05,
      "loss": 0.242,
      "step": 368
    },
    {
      "epoch": 12.758620689655173,
      "grad_norm": 6.59246301651001,
      "learning_rate": 5.4447852760736193e-05,
      "loss": 0.4546,
      "step": 370
    },
    {
      "epoch": 12.827586206896552,
      "grad_norm": 7.327258586883545,
      "learning_rate": 5.4141104294478526e-05,
      "loss": 0.4652,
      "step": 372
    },
    {
      "epoch": 12.89655172413793,
      "grad_norm": 8.353236198425293,
      "learning_rate": 5.3834355828220865e-05,
      "loss": 0.2649,
      "step": 374
    },
    {
      "epoch": 12.96551724137931,
      "grad_norm": 5.369735240936279,
      "learning_rate": 5.35276073619632e-05,
      "loss": 0.3408,
      "step": 376
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6287878787878788,
      "eval_f1_macro": 0.5426263900444636,
      "eval_f1_micro": 0.6287878787878788,
      "eval_f1_weighted": 0.6095195120641993,
      "eval_loss": 1.4445136785507202,
      "eval_precision_macro": 0.5558757032441243,
      "eval_precision_micro": 0.6287878787878788,
      "eval_precision_weighted": 0.6306541677355074,
      "eval_recall_macro": 0.5621390778533636,
      "eval_recall_micro": 0.6287878787878788,
      "eval_recall_weighted": 0.6287878787878788,
      "eval_runtime": 2.9534,
      "eval_samples_per_second": 44.694,
      "eval_steps_per_second": 5.756,
      "step": 377
    },
    {
      "epoch": 13.03448275862069,
      "grad_norm": 3.363588333129883,
      "learning_rate": 5.322085889570553e-05,
      "loss": 0.2143,
      "step": 378
    },
    {
      "epoch": 13.10344827586207,
      "grad_norm": 5.9551682472229,
      "learning_rate": 5.291411042944786e-05,
      "loss": 0.3482,
      "step": 380
    },
    {
      "epoch": 13.172413793103448,
      "grad_norm": 3.5881054401397705,
      "learning_rate": 5.260736196319018e-05,
      "loss": 0.3513,
      "step": 382
    },
    {
      "epoch": 13.241379310344827,
      "grad_norm": 3.4177823066711426,
      "learning_rate": 5.230061349693251e-05,
      "loss": 0.141,
      "step": 384
    },
    {
      "epoch": 13.310344827586206,
      "grad_norm": 1.6237398386001587,
      "learning_rate": 5.1993865030674845e-05,
      "loss": 0.1823,
      "step": 386
    },
    {
      "epoch": 13.379310344827585,
      "grad_norm": 4.370217323303223,
      "learning_rate": 5.1687116564417185e-05,
      "loss": 0.1641,
      "step": 388
    },
    {
      "epoch": 13.448275862068966,
      "grad_norm": 7.227107048034668,
      "learning_rate": 5.138036809815952e-05,
      "loss": 0.3134,
      "step": 390
    },
    {
      "epoch": 13.517241379310345,
      "grad_norm": 8.068262100219727,
      "learning_rate": 5.107361963190185e-05,
      "loss": 0.2103,
      "step": 392
    },
    {
      "epoch": 13.586206896551724,
      "grad_norm": 0.982545793056488,
      "learning_rate": 5.076687116564417e-05,
      "loss": 0.0825,
      "step": 394
    },
    {
      "epoch": 13.655172413793103,
      "grad_norm": 5.129421234130859,
      "learning_rate": 5.04601226993865e-05,
      "loss": 0.3593,
      "step": 396
    },
    {
      "epoch": 13.724137931034482,
      "grad_norm": 7.760966777801514,
      "learning_rate": 5.015337423312883e-05,
      "loss": 0.4634,
      "step": 398
    },
    {
      "epoch": 13.793103448275861,
      "grad_norm": 5.949847221374512,
      "learning_rate": 4.984662576687117e-05,
      "loss": 0.4661,
      "step": 400
    },
    {
      "epoch": 13.862068965517242,
      "grad_norm": 6.836796760559082,
      "learning_rate": 4.9539877300613504e-05,
      "loss": 0.4485,
      "step": 402
    },
    {
      "epoch": 13.931034482758621,
      "grad_norm": 8.495439529418945,
      "learning_rate": 4.923312883435583e-05,
      "loss": 0.6775,
      "step": 404
    },
    {
      "epoch": 14.0,
      "grad_norm": 6.360949993133545,
      "learning_rate": 4.892638036809816e-05,
      "loss": 0.2914,
      "step": 406
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6515151515151515,
      "eval_f1_macro": 0.6009180806560772,
      "eval_f1_micro": 0.6515151515151515,
      "eval_f1_weighted": 0.6470399238998288,
      "eval_loss": 1.4276713132858276,
      "eval_precision_macro": 0.7068301168482659,
      "eval_precision_micro": 0.6515151515151515,
      "eval_precision_weighted": 0.6867810122256585,
      "eval_recall_macro": 0.5958201058201057,
      "eval_recall_micro": 0.6515151515151515,
      "eval_recall_weighted": 0.6515151515151515,
      "eval_runtime": 2.9541,
      "eval_samples_per_second": 44.683,
      "eval_steps_per_second": 5.755,
      "step": 406
    },
    {
      "epoch": 14.068965517241379,
      "grad_norm": 5.016369342803955,
      "learning_rate": 4.8619631901840495e-05,
      "loss": 0.1568,
      "step": 408
    },
    {
      "epoch": 14.137931034482758,
      "grad_norm": 1.709494948387146,
      "learning_rate": 4.831288343558282e-05,
      "loss": 0.1725,
      "step": 410
    },
    {
      "epoch": 14.206896551724139,
      "grad_norm": 11.258108139038086,
      "learning_rate": 4.800613496932516e-05,
      "loss": 0.4265,
      "step": 412
    },
    {
      "epoch": 14.275862068965518,
      "grad_norm": 7.010964870452881,
      "learning_rate": 4.769938650306749e-05,
      "loss": 0.2611,
      "step": 414
    },
    {
      "epoch": 14.344827586206897,
      "grad_norm": 2.1494526863098145,
      "learning_rate": 4.739263803680982e-05,
      "loss": 0.2004,
      "step": 416
    },
    {
      "epoch": 14.413793103448276,
      "grad_norm": 1.0723114013671875,
      "learning_rate": 4.708588957055215e-05,
      "loss": 0.1463,
      "step": 418
    },
    {
      "epoch": 14.482758620689655,
      "grad_norm": 4.130954742431641,
      "learning_rate": 4.677914110429448e-05,
      "loss": 0.0826,
      "step": 420
    },
    {
      "epoch": 14.551724137931034,
      "grad_norm": 2.124276876449585,
      "learning_rate": 4.647239263803681e-05,
      "loss": 0.2997,
      "step": 422
    },
    {
      "epoch": 14.620689655172415,
      "grad_norm": 6.786820888519287,
      "learning_rate": 4.616564417177914e-05,
      "loss": 0.186,
      "step": 424
    },
    {
      "epoch": 14.689655172413794,
      "grad_norm": 1.8289989233016968,
      "learning_rate": 4.585889570552148e-05,
      "loss": 0.1018,
      "step": 426
    },
    {
      "epoch": 14.758620689655173,
      "grad_norm": 2.392906904220581,
      "learning_rate": 4.5552147239263805e-05,
      "loss": 0.1582,
      "step": 428
    },
    {
      "epoch": 14.827586206896552,
      "grad_norm": 0.6232361197471619,
      "learning_rate": 4.524539877300614e-05,
      "loss": 0.0487,
      "step": 430
    },
    {
      "epoch": 14.89655172413793,
      "grad_norm": 10.35379695892334,
      "learning_rate": 4.493865030674847e-05,
      "loss": 0.3485,
      "step": 432
    },
    {
      "epoch": 14.96551724137931,
      "grad_norm": 3.712934970855713,
      "learning_rate": 4.4631901840490795e-05,
      "loss": 0.2003,
      "step": 434
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6287878787878788,
      "eval_f1_macro": 0.5769950459342476,
      "eval_f1_micro": 0.6287878787878788,
      "eval_f1_weighted": 0.6296265645546432,
      "eval_loss": 1.5517460107803345,
      "eval_precision_macro": 0.5889880952380953,
      "eval_precision_micro": 0.6287878787878788,
      "eval_precision_weighted": 0.6474589646464647,
      "eval_recall_macro": 0.579191232048375,
      "eval_recall_micro": 0.6287878787878788,
      "eval_recall_weighted": 0.6287878787878788,
      "eval_runtime": 2.9553,
      "eval_samples_per_second": 44.666,
      "eval_steps_per_second": 5.752,
      "step": 435
    },
    {
      "epoch": 15.03448275862069,
      "grad_norm": 6.708681106567383,
      "learning_rate": 4.432515337423313e-05,
      "loss": 0.0948,
      "step": 436
    },
    {
      "epoch": 15.10344827586207,
      "grad_norm": 2.6313998699188232,
      "learning_rate": 4.4018404907975466e-05,
      "loss": 0.1151,
      "step": 438
    },
    {
      "epoch": 15.172413793103448,
      "grad_norm": 1.044400930404663,
      "learning_rate": 4.371165644171779e-05,
      "loss": 0.0373,
      "step": 440
    },
    {
      "epoch": 15.241379310344827,
      "grad_norm": 4.7035064697265625,
      "learning_rate": 4.3404907975460124e-05,
      "loss": 0.1399,
      "step": 442
    },
    {
      "epoch": 15.310344827586206,
      "grad_norm": 10.513556480407715,
      "learning_rate": 4.309815950920246e-05,
      "loss": 0.5143,
      "step": 444
    },
    {
      "epoch": 15.379310344827585,
      "grad_norm": 2.8816163539886475,
      "learning_rate": 4.279141104294479e-05,
      "loss": 0.0437,
      "step": 446
    },
    {
      "epoch": 15.448275862068966,
      "grad_norm": 0.7600950598716736,
      "learning_rate": 4.2484662576687115e-05,
      "loss": 0.112,
      "step": 448
    },
    {
      "epoch": 15.517241379310345,
      "grad_norm": 4.796483516693115,
      "learning_rate": 4.2177914110429454e-05,
      "loss": 0.0414,
      "step": 450
    },
    {
      "epoch": 15.586206896551724,
      "grad_norm": 9.010687828063965,
      "learning_rate": 4.1871165644171786e-05,
      "loss": 0.1845,
      "step": 452
    },
    {
      "epoch": 15.655172413793103,
      "grad_norm": 5.453256607055664,
      "learning_rate": 4.156441717791411e-05,
      "loss": 0.2598,
      "step": 454
    },
    {
      "epoch": 15.724137931034482,
      "grad_norm": 0.6193873286247253,
      "learning_rate": 4.1257668711656444e-05,
      "loss": 0.1249,
      "step": 456
    },
    {
      "epoch": 15.793103448275861,
      "grad_norm": 1.3485605716705322,
      "learning_rate": 4.0950920245398776e-05,
      "loss": 0.0884,
      "step": 458
    },
    {
      "epoch": 15.862068965517242,
      "grad_norm": 5.937335014343262,
      "learning_rate": 4.06441717791411e-05,
      "loss": 0.2239,
      "step": 460
    },
    {
      "epoch": 15.931034482758621,
      "grad_norm": 1.5148670673370361,
      "learning_rate": 4.033742331288344e-05,
      "loss": 0.1397,
      "step": 462
    },
    {
      "epoch": 16.0,
      "grad_norm": 6.244295120239258,
      "learning_rate": 4.0030674846625773e-05,
      "loss": 0.0871,
      "step": 464
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6515151515151515,
      "eval_f1_macro": 0.5701671714225517,
      "eval_f1_micro": 0.6515151515151515,
      "eval_f1_weighted": 0.6407448459978784,
      "eval_loss": 1.481210708618164,
      "eval_precision_macro": 0.5776942355889724,
      "eval_precision_micro": 0.6515151515151515,
      "eval_precision_weighted": 0.6491323004480899,
      "eval_recall_macro": 0.5785411942554799,
      "eval_recall_micro": 0.6515151515151515,
      "eval_recall_weighted": 0.6515151515151515,
      "eval_runtime": 2.9649,
      "eval_samples_per_second": 44.52,
      "eval_steps_per_second": 5.734,
      "step": 464
    },
    {
      "epoch": 16.06896551724138,
      "grad_norm": 11.335488319396973,
      "learning_rate": 3.97239263803681e-05,
      "loss": 0.3416,
      "step": 466
    },
    {
      "epoch": 16.137931034482758,
      "grad_norm": 3.1509413719177246,
      "learning_rate": 3.941717791411043e-05,
      "loss": 0.0441,
      "step": 468
    },
    {
      "epoch": 16.20689655172414,
      "grad_norm": 11.243932723999023,
      "learning_rate": 3.9110429447852764e-05,
      "loss": 0.297,
      "step": 470
    },
    {
      "epoch": 16.275862068965516,
      "grad_norm": 4.241555690765381,
      "learning_rate": 3.880368098159509e-05,
      "loss": 0.0561,
      "step": 472
    },
    {
      "epoch": 16.344827586206897,
      "grad_norm": 2.7286536693573,
      "learning_rate": 3.849693251533742e-05,
      "loss": 0.1139,
      "step": 474
    },
    {
      "epoch": 16.413793103448278,
      "grad_norm": 0.8547440767288208,
      "learning_rate": 3.819018404907976e-05,
      "loss": 0.0259,
      "step": 476
    },
    {
      "epoch": 16.482758620689655,
      "grad_norm": 0.9898260235786438,
      "learning_rate": 3.7883435582822086e-05,
      "loss": 0.1832,
      "step": 478
    },
    {
      "epoch": 16.551724137931036,
      "grad_norm": 1.3115100860595703,
      "learning_rate": 3.757668711656442e-05,
      "loss": 0.098,
      "step": 480
    },
    {
      "epoch": 16.620689655172413,
      "grad_norm": 7.043308734893799,
      "learning_rate": 3.726993865030675e-05,
      "loss": 0.0928,
      "step": 482
    },
    {
      "epoch": 16.689655172413794,
      "grad_norm": 5.262262344360352,
      "learning_rate": 3.696319018404908e-05,
      "loss": 0.2486,
      "step": 484
    },
    {
      "epoch": 16.75862068965517,
      "grad_norm": 4.680488586425781,
      "learning_rate": 3.665644171779141e-05,
      "loss": 0.0745,
      "step": 486
    },
    {
      "epoch": 16.82758620689655,
      "grad_norm": 1.229391098022461,
      "learning_rate": 3.634969325153375e-05,
      "loss": 0.0213,
      "step": 488
    },
    {
      "epoch": 16.896551724137932,
      "grad_norm": 3.3319144248962402,
      "learning_rate": 3.6042944785276074e-05,
      "loss": 0.0361,
      "step": 490
    },
    {
      "epoch": 16.96551724137931,
      "grad_norm": 2.358989953994751,
      "learning_rate": 3.5736196319018406e-05,
      "loss": 0.0352,
      "step": 492
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5984848484848485,
      "eval_f1_macro": 0.5007356485225759,
      "eval_f1_micro": 0.5984848484848485,
      "eval_f1_weighted": 0.5743528123674804,
      "eval_loss": 2.10516357421875,
      "eval_precision_macro": 0.5466466490902581,
      "eval_precision_micro": 0.5984848484848485,
      "eval_precision_weighted": 0.6130060430000621,
      "eval_recall_macro": 0.5126681783824641,
      "eval_recall_micro": 0.5984848484848485,
      "eval_recall_weighted": 0.5984848484848485,
      "eval_runtime": 2.9571,
      "eval_samples_per_second": 44.638,
      "eval_steps_per_second": 5.749,
      "step": 493
    },
    {
      "epoch": 17.03448275862069,
      "grad_norm": 7.632280349731445,
      "learning_rate": 3.542944785276074e-05,
      "loss": 0.0805,
      "step": 494
    },
    {
      "epoch": 17.103448275862068,
      "grad_norm": 2.5213019847869873,
      "learning_rate": 3.512269938650307e-05,
      "loss": 0.0341,
      "step": 496
    },
    {
      "epoch": 17.17241379310345,
      "grad_norm": 0.9358583688735962,
      "learning_rate": 3.4815950920245396e-05,
      "loss": 0.0441,
      "step": 498
    },
    {
      "epoch": 17.24137931034483,
      "grad_norm": 1.5655701160430908,
      "learning_rate": 3.4509202453987735e-05,
      "loss": 0.0189,
      "step": 500
    },
    {
      "epoch": 17.310344827586206,
      "grad_norm": 0.4072040319442749,
      "learning_rate": 3.420245398773007e-05,
      "loss": 0.0209,
      "step": 502
    },
    {
      "epoch": 17.379310344827587,
      "grad_norm": 7.48984956741333,
      "learning_rate": 3.3895705521472393e-05,
      "loss": 0.1448,
      "step": 504
    },
    {
      "epoch": 17.448275862068964,
      "grad_norm": 3.6645641326904297,
      "learning_rate": 3.3588957055214726e-05,
      "loss": 0.1325,
      "step": 506
    },
    {
      "epoch": 17.517241379310345,
      "grad_norm": 1.4333422183990479,
      "learning_rate": 3.328220858895706e-05,
      "loss": 0.0464,
      "step": 508
    },
    {
      "epoch": 17.586206896551722,
      "grad_norm": 2.419375419616699,
      "learning_rate": 3.2975460122699384e-05,
      "loss": 0.018,
      "step": 510
    },
    {
      "epoch": 17.655172413793103,
      "grad_norm": 9.471452713012695,
      "learning_rate": 3.266871165644172e-05,
      "loss": 0.1255,
      "step": 512
    },
    {
      "epoch": 17.724137931034484,
      "grad_norm": 5.6544575691223145,
      "learning_rate": 3.2361963190184055e-05,
      "loss": 0.0936,
      "step": 514
    },
    {
      "epoch": 17.79310344827586,
      "grad_norm": 3.275522470474243,
      "learning_rate": 3.205521472392638e-05,
      "loss": 0.1898,
      "step": 516
    },
    {
      "epoch": 17.862068965517242,
      "grad_norm": 0.8146153092384338,
      "learning_rate": 3.174846625766871e-05,
      "loss": 0.0209,
      "step": 518
    },
    {
      "epoch": 17.93103448275862,
      "grad_norm": 9.057777404785156,
      "learning_rate": 3.1441717791411045e-05,
      "loss": 0.2627,
      "step": 520
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.2130121886730194,
      "learning_rate": 3.113496932515337e-05,
      "loss": 0.0101,
      "step": 522
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6212121212121212,
      "eval_f1_macro": 0.5724632695552766,
      "eval_f1_micro": 0.6212121212121212,
      "eval_f1_weighted": 0.6223060135184718,
      "eval_loss": 1.9977855682373047,
      "eval_precision_macro": 0.615182436611008,
      "eval_precision_micro": 0.6212121212121212,
      "eval_precision_weighted": 0.6558523547159911,
      "eval_recall_macro": 0.5672486772486772,
      "eval_recall_micro": 0.6212121212121212,
      "eval_recall_weighted": 0.6212121212121212,
      "eval_runtime": 2.9607,
      "eval_samples_per_second": 44.583,
      "eval_steps_per_second": 5.742,
      "step": 522
    },
    {
      "epoch": 18.06896551724138,
      "grad_norm": 5.074588775634766,
      "learning_rate": 3.0828220858895703e-05,
      "loss": 0.0407,
      "step": 524
    },
    {
      "epoch": 18.137931034482758,
      "grad_norm": 4.95729923248291,
      "learning_rate": 3.052147239263804e-05,
      "loss": 0.0346,
      "step": 526
    },
    {
      "epoch": 18.20689655172414,
      "grad_norm": 0.48612403869628906,
      "learning_rate": 3.0214723926380368e-05,
      "loss": 0.006,
      "step": 528
    },
    {
      "epoch": 18.275862068965516,
      "grad_norm": 1.2487165927886963,
      "learning_rate": 2.99079754601227e-05,
      "loss": 0.0738,
      "step": 530
    },
    {
      "epoch": 18.344827586206897,
      "grad_norm": 6.368688106536865,
      "learning_rate": 2.9601226993865033e-05,
      "loss": 0.1009,
      "step": 532
    },
    {
      "epoch": 18.413793103448278,
      "grad_norm": 0.3834826648235321,
      "learning_rate": 2.9294478527607362e-05,
      "loss": 0.0337,
      "step": 534
    },
    {
      "epoch": 18.482758620689655,
      "grad_norm": 7.203604221343994,
      "learning_rate": 2.8987730061349694e-05,
      "loss": 0.1217,
      "step": 536
    },
    {
      "epoch": 18.551724137931036,
      "grad_norm": 0.1774144023656845,
      "learning_rate": 2.8680981595092026e-05,
      "loss": 0.0448,
      "step": 538
    },
    {
      "epoch": 18.620689655172413,
      "grad_norm": 1.4739829301834106,
      "learning_rate": 2.837423312883436e-05,
      "loss": 0.042,
      "step": 540
    },
    {
      "epoch": 18.689655172413794,
      "grad_norm": 7.63886022567749,
      "learning_rate": 2.8067484662576688e-05,
      "loss": 0.1012,
      "step": 542
    },
    {
      "epoch": 18.75862068965517,
      "grad_norm": 0.5444397926330566,
      "learning_rate": 2.776073619631902e-05,
      "loss": 0.0614,
      "step": 544
    },
    {
      "epoch": 18.82758620689655,
      "grad_norm": 0.015687117353081703,
      "learning_rate": 2.7453987730061353e-05,
      "loss": 0.1418,
      "step": 546
    },
    {
      "epoch": 18.896551724137932,
      "grad_norm": 1.4215542078018188,
      "learning_rate": 2.714723926380368e-05,
      "loss": 0.0078,
      "step": 548
    },
    {
      "epoch": 18.96551724137931,
      "grad_norm": 0.5485539436340332,
      "learning_rate": 2.6840490797546014e-05,
      "loss": 0.0035,
      "step": 550
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6439393939393939,
      "eval_f1_macro": 0.587968716690295,
      "eval_f1_micro": 0.6439393939393939,
      "eval_f1_weighted": 0.638753653210352,
      "eval_loss": 2.030425786972046,
      "eval_precision_macro": 0.6697785789090137,
      "eval_precision_micro": 0.6439393939393939,
      "eval_precision_weighted": 0.6935502946767769,
      "eval_recall_macro": 0.5804761904761905,
      "eval_recall_micro": 0.6439393939393939,
      "eval_recall_weighted": 0.6439393939393939,
      "eval_runtime": 2.9436,
      "eval_samples_per_second": 44.843,
      "eval_steps_per_second": 5.775,
      "step": 551
    },
    {
      "epoch": 19.03448275862069,
      "grad_norm": 0.7515120506286621,
      "learning_rate": 2.6533742331288346e-05,
      "loss": 0.0053,
      "step": 552
    },
    {
      "epoch": 19.103448275862068,
      "grad_norm": 6.6086835861206055,
      "learning_rate": 2.6226993865030675e-05,
      "loss": 0.0638,
      "step": 554
    },
    {
      "epoch": 19.17241379310345,
      "grad_norm": 1.8158479928970337,
      "learning_rate": 2.5920245398773008e-05,
      "loss": 0.0173,
      "step": 556
    },
    {
      "epoch": 19.24137931034483,
      "grad_norm": 1.9973891973495483,
      "learning_rate": 2.561349693251534e-05,
      "loss": 0.0127,
      "step": 558
    },
    {
      "epoch": 19.310344827586206,
      "grad_norm": 0.03836736083030701,
      "learning_rate": 2.530674846625767e-05,
      "loss": 0.0009,
      "step": 560
    },
    {
      "epoch": 19.379310344827587,
      "grad_norm": 2.1399827003479004,
      "learning_rate": 2.5e-05,
      "loss": 0.0149,
      "step": 562
    },
    {
      "epoch": 19.448275862068964,
      "grad_norm": 1.5504521131515503,
      "learning_rate": 2.469325153374233e-05,
      "loss": 0.0125,
      "step": 564
    },
    {
      "epoch": 19.517241379310345,
      "grad_norm": 2.622783660888672,
      "learning_rate": 2.4386503067484666e-05,
      "loss": 0.0509,
      "step": 566
    },
    {
      "epoch": 19.586206896551722,
      "grad_norm": 6.634346008300781,
      "learning_rate": 2.4079754601226995e-05,
      "loss": 0.0351,
      "step": 568
    },
    {
      "epoch": 19.655172413793103,
      "grad_norm": 0.067986860871315,
      "learning_rate": 2.3773006134969324e-05,
      "loss": 0.1117,
      "step": 570
    },
    {
      "epoch": 19.724137931034484,
      "grad_norm": 0.664193332195282,
      "learning_rate": 2.346625766871166e-05,
      "loss": 0.0052,
      "step": 572
    },
    {
      "epoch": 19.79310344827586,
      "grad_norm": 11.805370330810547,
      "learning_rate": 2.315950920245399e-05,
      "loss": 0.2013,
      "step": 574
    },
    {
      "epoch": 19.862068965517242,
      "grad_norm": 0.12794336676597595,
      "learning_rate": 2.285276073619632e-05,
      "loss": 0.0086,
      "step": 576
    },
    {
      "epoch": 19.93103448275862,
      "grad_norm": 0.3751371204853058,
      "learning_rate": 2.2546012269938653e-05,
      "loss": 0.0046,
      "step": 578
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.025759520009160042,
      "learning_rate": 2.2239263803680982e-05,
      "loss": 0.0013,
      "step": 580
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1_macro": 0.5513619025246932,
      "eval_f1_micro": 0.6363636363636364,
      "eval_f1_weighted": 0.6223818820101413,
      "eval_loss": 2.1374073028564453,
      "eval_precision_macro": 0.6025479764157659,
      "eval_precision_micro": 0.6363636363636364,
      "eval_precision_weighted": 0.6765262979680514,
      "eval_recall_macro": 0.5684882842025699,
      "eval_recall_micro": 0.6363636363636364,
      "eval_recall_weighted": 0.6363636363636364,
      "eval_runtime": 3.0501,
      "eval_samples_per_second": 43.277,
      "eval_steps_per_second": 5.574,
      "step": 580
    },
    {
      "epoch": 20.06896551724138,
      "grad_norm": 5.584213733673096,
      "learning_rate": 2.1932515337423315e-05,
      "loss": 0.0257,
      "step": 582
    },
    {
      "epoch": 20.137931034482758,
      "grad_norm": 6.075766086578369,
      "learning_rate": 2.1625766871165647e-05,
      "loss": 0.0319,
      "step": 584
    },
    {
      "epoch": 20.20689655172414,
      "grad_norm": 0.27452629804611206,
      "learning_rate": 2.1319018404907976e-05,
      "loss": 0.0029,
      "step": 586
    },
    {
      "epoch": 20.275862068965516,
      "grad_norm": 0.04866321012377739,
      "learning_rate": 2.1012269938650308e-05,
      "loss": 0.0021,
      "step": 588
    },
    {
      "epoch": 20.344827586206897,
      "grad_norm": 1.3495759963989258,
      "learning_rate": 2.0705521472392637e-05,
      "loss": 0.0088,
      "step": 590
    },
    {
      "epoch": 20.413793103448278,
      "grad_norm": 2.283107042312622,
      "learning_rate": 2.039877300613497e-05,
      "loss": 0.0166,
      "step": 592
    },
    {
      "epoch": 20.482758620689655,
      "grad_norm": 1.5386358499526978,
      "learning_rate": 2.0092024539877302e-05,
      "loss": 0.0544,
      "step": 594
    },
    {
      "epoch": 20.551724137931036,
      "grad_norm": 5.752704620361328,
      "learning_rate": 1.978527607361963e-05,
      "loss": 0.0246,
      "step": 596
    },
    {
      "epoch": 20.620689655172413,
      "grad_norm": 3.968679904937744,
      "learning_rate": 1.9478527607361967e-05,
      "loss": 0.0161,
      "step": 598
    },
    {
      "epoch": 20.689655172413794,
      "grad_norm": 0.8225330710411072,
      "learning_rate": 1.9171779141104296e-05,
      "loss": 0.0056,
      "step": 600
    },
    {
      "epoch": 20.75862068965517,
      "grad_norm": 6.011229991912842,
      "learning_rate": 1.8865030674846625e-05,
      "loss": 0.1674,
      "step": 602
    },
    {
      "epoch": 20.82758620689655,
      "grad_norm": 0.19111225008964539,
      "learning_rate": 1.855828220858896e-05,
      "loss": 0.0069,
      "step": 604
    },
    {
      "epoch": 20.896551724137932,
      "grad_norm": 14.381773948669434,
      "learning_rate": 1.825153374233129e-05,
      "loss": 0.1019,
      "step": 606
    },
    {
      "epoch": 20.96551724137931,
      "grad_norm": 7.869429111480713,
      "learning_rate": 1.7944785276073618e-05,
      "loss": 0.0589,
      "step": 608
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.6439393939393939,
      "eval_f1_macro": 0.5879137999438752,
      "eval_f1_micro": 0.6439393939393939,
      "eval_f1_weighted": 0.6396446806663525,
      "eval_loss": 1.767621636390686,
      "eval_precision_macro": 0.5940388301069416,
      "eval_precision_micro": 0.6439393939393939,
      "eval_precision_weighted": 0.640713517648502,
      "eval_recall_macro": 0.5888813303099018,
      "eval_recall_micro": 0.6439393939393939,
      "eval_recall_weighted": 0.6439393939393939,
      "eval_runtime": 2.9525,
      "eval_samples_per_second": 44.709,
      "eval_steps_per_second": 5.758,
      "step": 609
    },
    {
      "epoch": 21.03448275862069,
      "grad_norm": 0.8013659119606018,
      "learning_rate": 1.7638036809815954e-05,
      "loss": 0.0109,
      "step": 610
    },
    {
      "epoch": 21.103448275862068,
      "grad_norm": 0.09149477630853653,
      "learning_rate": 1.7331288343558283e-05,
      "loss": 0.0015,
      "step": 612
    },
    {
      "epoch": 21.17241379310345,
      "grad_norm": 0.035085856914520264,
      "learning_rate": 1.7024539877300612e-05,
      "loss": 0.0278,
      "step": 614
    },
    {
      "epoch": 21.24137931034483,
      "grad_norm": 3.7100307941436768,
      "learning_rate": 1.6717791411042948e-05,
      "loss": 0.0217,
      "step": 616
    },
    {
      "epoch": 21.310344827586206,
      "grad_norm": 0.08583565056324005,
      "learning_rate": 1.6411042944785277e-05,
      "loss": 0.0083,
      "step": 618
    },
    {
      "epoch": 21.379310344827587,
      "grad_norm": 3.9796557426452637,
      "learning_rate": 1.6104294478527606e-05,
      "loss": 0.1204,
      "step": 620
    },
    {
      "epoch": 21.448275862068964,
      "grad_norm": 3.607825994491577,
      "learning_rate": 1.579754601226994e-05,
      "loss": 0.0264,
      "step": 622
    },
    {
      "epoch": 21.517241379310345,
      "grad_norm": 0.23625929653644562,
      "learning_rate": 1.549079754601227e-05,
      "loss": 0.0061,
      "step": 624
    },
    {
      "epoch": 21.586206896551722,
      "grad_norm": 0.34624549746513367,
      "learning_rate": 1.5184049079754603e-05,
      "loss": 0.0655,
      "step": 626
    },
    {
      "epoch": 21.655172413793103,
      "grad_norm": 4.58054256439209,
      "learning_rate": 1.4877300613496933e-05,
      "loss": 0.0465,
      "step": 628
    },
    {
      "epoch": 21.724137931034484,
      "grad_norm": 0.03262769430875778,
      "learning_rate": 1.4570552147239264e-05,
      "loss": 0.0025,
      "step": 630
    },
    {
      "epoch": 21.79310344827586,
      "grad_norm": 0.6401829123497009,
      "learning_rate": 1.4263803680981596e-05,
      "loss": 0.0849,
      "step": 632
    },
    {
      "epoch": 21.862068965517242,
      "grad_norm": 4.184028148651123,
      "learning_rate": 1.3957055214723927e-05,
      "loss": 0.0624,
      "step": 634
    },
    {
      "epoch": 21.93103448275862,
      "grad_norm": 0.050627466291189194,
      "learning_rate": 1.3650306748466258e-05,
      "loss": 0.0023,
      "step": 636
    },
    {
      "epoch": 22.0,
      "grad_norm": 3.8194804191589355,
      "learning_rate": 1.334355828220859e-05,
      "loss": 0.0263,
      "step": 638
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6439393939393939,
      "eval_f1_macro": 0.5785271807221967,
      "eval_f1_micro": 0.6439393939393939,
      "eval_f1_weighted": 0.6327110651723412,
      "eval_loss": 1.841572880744934,
      "eval_precision_macro": 0.6016241640968781,
      "eval_precision_micro": 0.6439393939393939,
      "eval_precision_weighted": 0.6454148108541149,
      "eval_recall_macro": 0.5757520786092215,
      "eval_recall_micro": 0.6439393939393939,
      "eval_recall_weighted": 0.6439393939393939,
      "eval_runtime": 2.9422,
      "eval_samples_per_second": 44.864,
      "eval_steps_per_second": 5.778,
      "step": 638
    },
    {
      "epoch": 22.06896551724138,
      "grad_norm": 0.16209831833839417,
      "learning_rate": 1.303680981595092e-05,
      "loss": 0.0465,
      "step": 640
    },
    {
      "epoch": 22.137931034482758,
      "grad_norm": 0.1036888137459755,
      "learning_rate": 1.2730061349693251e-05,
      "loss": 0.0657,
      "step": 642
    },
    {
      "epoch": 22.20689655172414,
      "grad_norm": 1.0510048866271973,
      "learning_rate": 1.2423312883435584e-05,
      "loss": 0.0062,
      "step": 644
    },
    {
      "epoch": 22.275862068965516,
      "grad_norm": 3.261247158050537,
      "learning_rate": 1.2116564417177914e-05,
      "loss": 0.0175,
      "step": 646
    },
    {
      "epoch": 22.344827586206897,
      "grad_norm": 0.41699743270874023,
      "learning_rate": 1.1809815950920245e-05,
      "loss": 0.0136,
      "step": 648
    },
    {
      "epoch": 22.413793103448278,
      "grad_norm": 1.2335162162780762,
      "learning_rate": 1.1503067484662577e-05,
      "loss": 0.005,
      "step": 650
    },
    {
      "epoch": 22.482758620689655,
      "grad_norm": 0.03782954812049866,
      "learning_rate": 1.119631901840491e-05,
      "loss": 0.2332,
      "step": 652
    },
    {
      "epoch": 22.551724137931036,
      "grad_norm": 4.644011497497559,
      "learning_rate": 1.0889570552147239e-05,
      "loss": 0.0251,
      "step": 654
    },
    {
      "epoch": 22.620689655172413,
      "grad_norm": 2.634734630584717,
      "learning_rate": 1.0582822085889571e-05,
      "loss": 0.0357,
      "step": 656
    },
    {
      "epoch": 22.689655172413794,
      "grad_norm": 0.04239609092473984,
      "learning_rate": 1.0276073619631903e-05,
      "loss": 0.0025,
      "step": 658
    },
    {
      "epoch": 22.75862068965517,
      "grad_norm": 1.9931837320327759,
      "learning_rate": 9.969325153374232e-06,
      "loss": 0.0139,
      "step": 660
    },
    {
      "epoch": 22.82758620689655,
      "grad_norm": 0.26363053917884827,
      "learning_rate": 9.662576687116565e-06,
      "loss": 0.0047,
      "step": 662
    },
    {
      "epoch": 22.896551724137932,
      "grad_norm": 0.2392028421163559,
      "learning_rate": 9.355828220858897e-06,
      "loss": 0.003,
      "step": 664
    },
    {
      "epoch": 22.96551724137931,
      "grad_norm": 0.5267300605773926,
      "learning_rate": 9.049079754601228e-06,
      "loss": 0.0028,
      "step": 666
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1_macro": 0.6068035568035569,
      "eval_f1_micro": 0.6666666666666666,
      "eval_f1_weighted": 0.6569024750842932,
      "eval_loss": 1.984299659729004,
      "eval_precision_macro": 0.663051653896554,
      "eval_precision_micro": 0.6666666666666666,
      "eval_precision_weighted": 0.6882214684446634,
      "eval_recall_macro": 0.6069312169312168,
      "eval_recall_micro": 0.6666666666666666,
      "eval_recall_weighted": 0.6666666666666666,
      "eval_runtime": 2.9646,
      "eval_samples_per_second": 44.526,
      "eval_steps_per_second": 5.734,
      "step": 667
    },
    {
      "epoch": 23.03448275862069,
      "grad_norm": 0.5740432739257812,
      "learning_rate": 8.742331288343558e-06,
      "loss": 0.003,
      "step": 668
    },
    {
      "epoch": 23.103448275862068,
      "grad_norm": 0.03469856455922127,
      "learning_rate": 8.435582822085889e-06,
      "loss": 0.0006,
      "step": 670
    },
    {
      "epoch": 23.17241379310345,
      "grad_norm": 0.05739348381757736,
      "learning_rate": 8.128834355828221e-06,
      "loss": 0.0562,
      "step": 672
    },
    {
      "epoch": 23.24137931034483,
      "grad_norm": 0.6837252974510193,
      "learning_rate": 7.822085889570554e-06,
      "loss": 0.0441,
      "step": 674
    },
    {
      "epoch": 23.310344827586206,
      "grad_norm": 10.602856636047363,
      "learning_rate": 7.5153374233128836e-06,
      "loss": 0.0493,
      "step": 676
    },
    {
      "epoch": 23.379310344827587,
      "grad_norm": 0.026621289551258087,
      "learning_rate": 7.208588957055215e-06,
      "loss": 0.0009,
      "step": 678
    },
    {
      "epoch": 23.448275862068964,
      "grad_norm": 0.05075424537062645,
      "learning_rate": 6.901840490797547e-06,
      "loss": 0.001,
      "step": 680
    },
    {
      "epoch": 23.517241379310345,
      "grad_norm": 3.2929599285125732,
      "learning_rate": 6.595092024539877e-06,
      "loss": 0.0166,
      "step": 682
    },
    {
      "epoch": 23.586206896551722,
      "grad_norm": 1.3883030414581299,
      "learning_rate": 6.288343558282209e-06,
      "loss": 0.0044,
      "step": 684
    },
    {
      "epoch": 23.655172413793103,
      "grad_norm": 3.663137197494507,
      "learning_rate": 5.98159509202454e-06,
      "loss": 0.0135,
      "step": 686
    },
    {
      "epoch": 23.724137931034484,
      "grad_norm": 0.037959225475788116,
      "learning_rate": 5.674846625766871e-06,
      "loss": 0.0009,
      "step": 688
    },
    {
      "epoch": 23.79310344827586,
      "grad_norm": 0.024055300280451775,
      "learning_rate": 5.368098159509203e-06,
      "loss": 0.001,
      "step": 690
    },
    {
      "epoch": 23.862068965517242,
      "grad_norm": 0.03909283131361008,
      "learning_rate": 5.061349693251534e-06,
      "loss": 0.0008,
      "step": 692
    },
    {
      "epoch": 23.93103448275862,
      "grad_norm": 0.017916660755872726,
      "learning_rate": 4.7546012269938654e-06,
      "loss": 0.0007,
      "step": 694
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.009021880105137825,
      "learning_rate": 4.447852760736196e-06,
      "loss": 0.0006,
      "step": 696
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6742424242424242,
      "eval_f1_macro": 0.615661662832857,
      "eval_f1_micro": 0.6742424242424242,
      "eval_f1_weighted": 0.6655260925563127,
      "eval_loss": 1.9431589841842651,
      "eval_precision_macro": 0.6602878688132144,
      "eval_precision_micro": 0.6742424242424242,
      "eval_precision_weighted": 0.6853071985036794,
      "eval_recall_macro": 0.6152078609221466,
      "eval_recall_micro": 0.6742424242424242,
      "eval_recall_weighted": 0.6742424242424242,
      "eval_runtime": 2.9569,
      "eval_samples_per_second": 44.642,
      "eval_steps_per_second": 5.749,
      "step": 696
    },
    {
      "epoch": 24.06896551724138,
      "grad_norm": 0.08362489938735962,
      "learning_rate": 4.141104294478528e-06,
      "loss": 0.001,
      "step": 698
    },
    {
      "epoch": 24.137931034482758,
      "grad_norm": 2.7230169773101807,
      "learning_rate": 3.834355828220859e-06,
      "loss": 0.0105,
      "step": 700
    },
    {
      "epoch": 24.20689655172414,
      "grad_norm": 0.5051957368850708,
      "learning_rate": 3.52760736196319e-06,
      "loss": 0.0038,
      "step": 702
    },
    {
      "epoch": 24.275862068965516,
      "grad_norm": 0.6274848580360413,
      "learning_rate": 3.2208588957055217e-06,
      "loss": 0.0022,
      "step": 704
    },
    {
      "epoch": 24.344827586206897,
      "grad_norm": 1.267530083656311,
      "learning_rate": 2.914110429447853e-06,
      "loss": 0.0208,
      "step": 706
    },
    {
      "epoch": 24.413793103448278,
      "grad_norm": 0.0061959754675626755,
      "learning_rate": 2.607361963190184e-06,
      "loss": 0.0003,
      "step": 708
    },
    {
      "epoch": 24.482758620689655,
      "grad_norm": 0.01164495013654232,
      "learning_rate": 2.3006134969325154e-06,
      "loss": 0.0006,
      "step": 710
    },
    {
      "epoch": 24.551724137931036,
      "grad_norm": 0.051692791283130646,
      "learning_rate": 1.9938650306748465e-06,
      "loss": 0.0007,
      "step": 712
    },
    {
      "epoch": 24.620689655172413,
      "grad_norm": 1.7959578037261963,
      "learning_rate": 1.687116564417178e-06,
      "loss": 0.0073,
      "step": 714
    },
    {
      "epoch": 24.689655172413794,
      "grad_norm": 0.06827358901500702,
      "learning_rate": 1.3803680981595093e-06,
      "loss": 0.001,
      "step": 716
    },
    {
      "epoch": 24.75862068965517,
      "grad_norm": 0.014322124421596527,
      "learning_rate": 1.0736196319018406e-06,
      "loss": 0.0006,
      "step": 718
    },
    {
      "epoch": 24.82758620689655,
      "grad_norm": 0.24370881915092468,
      "learning_rate": 7.668711656441718e-07,
      "loss": 0.0045,
      "step": 720
    },
    {
      "epoch": 24.896551724137932,
      "grad_norm": 0.017789943143725395,
      "learning_rate": 4.601226993865031e-07,
      "loss": 0.0005,
      "step": 722
    },
    {
      "epoch": 24.96551724137931,
      "grad_norm": 0.03513872250914574,
      "learning_rate": 1.5337423312883438e-07,
      "loss": 0.0004,
      "step": 724
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1_macro": 0.6089082756725702,
      "eval_f1_micro": 0.6666666666666666,
      "eval_f1_weighted": 0.6569429652354989,
      "eval_loss": 1.9346412420272827,
      "eval_precision_macro": 0.654845691942466,
      "eval_precision_micro": 0.6666666666666666,
      "eval_precision_weighted": 0.6762704019668829,
      "eval_recall_macro": 0.6072713529856387,
      "eval_recall_micro": 0.6666666666666666,
      "eval_recall_weighted": 0.6666666666666666,
      "eval_runtime": 2.9677,
      "eval_samples_per_second": 44.478,
      "eval_steps_per_second": 5.728,
      "step": 725
    },
    {
      "epoch": 25.0,
      "step": 725,
      "total_flos": 3.163993239336653e+18,
      "train_loss": 0.5988656284373478,
      "train_runtime": 971.0069,
      "train_samples_per_second": 11.895,
      "train_steps_per_second": 0.747
    }
  ],
  "logging_steps": 2,
  "max_steps": 725,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "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": 3.163993239336653e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}