{
  "best_metric": 0.6590909090909091,
  "best_model_checkpoint": "squarerun2/checkpoint-522",
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 580,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 2.9724416732788086,
      "learning_rate": 3.448275862068966e-06,
      "loss": 1.9767,
      "step": 2
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 1.7837220430374146,
      "learning_rate": 6.896551724137932e-06,
      "loss": 1.9373,
      "step": 4
    },
    {
      "epoch": 0.20689655172413793,
      "grad_norm": 1.9200481176376343,
      "learning_rate": 1.0344827586206897e-05,
      "loss": 1.9422,
      "step": 6
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 1.8607347011566162,
      "learning_rate": 1.3793103448275863e-05,
      "loss": 1.9203,
      "step": 8
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 2.282787322998047,
      "learning_rate": 1.7241379310344828e-05,
      "loss": 1.8967,
      "step": 10
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 1.8788440227508545,
      "learning_rate": 2.0689655172413793e-05,
      "loss": 1.9576,
      "step": 12
    },
    {
      "epoch": 0.4827586206896552,
      "grad_norm": 1.689126968383789,
      "learning_rate": 2.413793103448276e-05,
      "loss": 1.9238,
      "step": 14
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 1.5710227489471436,
      "learning_rate": 2.7586206896551727e-05,
      "loss": 1.9468,
      "step": 16
    },
    {
      "epoch": 0.6206896551724138,
      "grad_norm": 1.6929808855056763,
      "learning_rate": 3.103448275862069e-05,
      "loss": 1.83,
      "step": 18
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 1.9147762060165405,
      "learning_rate": 3.4482758620689657e-05,
      "loss": 1.8844,
      "step": 20
    },
    {
      "epoch": 0.7586206896551724,
      "grad_norm": 1.0664433240890503,
      "learning_rate": 3.793103448275862e-05,
      "loss": 1.8893,
      "step": 22
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 1.2909166812896729,
      "learning_rate": 4.1379310344827587e-05,
      "loss": 1.8794,
      "step": 24
    },
    {
      "epoch": 0.896551724137931,
      "grad_norm": 2.562612771987915,
      "learning_rate": 4.482758620689655e-05,
      "loss": 1.8722,
      "step": 26
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 1.766555905342102,
      "learning_rate": 4.827586206896552e-05,
      "loss": 1.9016,
      "step": 28
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.24242424242424243,
      "eval_f1_macro": 0.1010989010989011,
      "eval_f1_micro": 0.24242424242424243,
      "eval_f1_weighted": 0.14009324009324012,
      "eval_loss": 1.8763961791992188,
      "eval_precision_macro": 0.07205977686035006,
      "eval_precision_micro": 0.24242424242424243,
      "eval_precision_weighted": 0.10010783738990092,
      "eval_recall_macro": 0.1760846560846561,
      "eval_recall_micro": 0.24242424242424243,
      "eval_recall_weighted": 0.24242424242424243,
      "eval_runtime": 2.3126,
      "eval_samples_per_second": 57.08,
      "eval_steps_per_second": 7.351,
      "step": 29
    },
    {
      "epoch": 1.0344827586206897,
      "grad_norm": 2.116236448287964,
      "learning_rate": 5.172413793103449e-05,
      "loss": 1.9127,
      "step": 30
    },
    {
      "epoch": 1.103448275862069,
      "grad_norm": 1.5244780778884888,
      "learning_rate": 5.517241379310345e-05,
      "loss": 1.8217,
      "step": 32
    },
    {
      "epoch": 1.1724137931034484,
      "grad_norm": 1.2253153324127197,
      "learning_rate": 5.862068965517241e-05,
      "loss": 1.8662,
      "step": 34
    },
    {
      "epoch": 1.2413793103448276,
      "grad_norm": 1.1567929983139038,
      "learning_rate": 6.206896551724138e-05,
      "loss": 1.9031,
      "step": 36
    },
    {
      "epoch": 1.3103448275862069,
      "grad_norm": 1.4619667530059814,
      "learning_rate": 6.551724137931034e-05,
      "loss": 1.8464,
      "step": 38
    },
    {
      "epoch": 1.3793103448275863,
      "grad_norm": 1.5786610841751099,
      "learning_rate": 6.896551724137931e-05,
      "loss": 1.9057,
      "step": 40
    },
    {
      "epoch": 1.4482758620689655,
      "grad_norm": 1.1554012298583984,
      "learning_rate": 7.241379310344828e-05,
      "loss": 1.8748,
      "step": 42
    },
    {
      "epoch": 1.5172413793103448,
      "grad_norm": 2.0982630252838135,
      "learning_rate": 7.586206896551724e-05,
      "loss": 1.9052,
      "step": 44
    },
    {
      "epoch": 1.5862068965517242,
      "grad_norm": 1.046858787536621,
      "learning_rate": 7.931034482758621e-05,
      "loss": 1.9363,
      "step": 46
    },
    {
      "epoch": 1.6551724137931034,
      "grad_norm": 1.05802583694458,
      "learning_rate": 8.275862068965517e-05,
      "loss": 1.8712,
      "step": 48
    },
    {
      "epoch": 1.7241379310344827,
      "grad_norm": 1.0946681499481201,
      "learning_rate": 8.620689655172413e-05,
      "loss": 1.9325,
      "step": 50
    },
    {
      "epoch": 1.793103448275862,
      "grad_norm": 1.3410812616348267,
      "learning_rate": 8.96551724137931e-05,
      "loss": 1.8671,
      "step": 52
    },
    {
      "epoch": 1.8620689655172413,
      "grad_norm": 1.4638521671295166,
      "learning_rate": 9.310344827586207e-05,
      "loss": 1.9462,
      "step": 54
    },
    {
      "epoch": 1.9310344827586206,
      "grad_norm": 1.6821434497833252,
      "learning_rate": 9.655172413793105e-05,
      "loss": 1.8823,
      "step": 56
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.771202564239502,
      "learning_rate": 0.0001,
      "loss": 1.8787,
      "step": 58
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.20454545454545456,
      "eval_f1_macro": 0.04851752021563342,
      "eval_f1_micro": 0.20454545454545456,
      "eval_f1_weighted": 0.06946826758147512,
      "eval_loss": 1.8749967813491821,
      "eval_precision_macro": 0.029220779220779224,
      "eval_precision_micro": 0.20454545454545456,
      "eval_precision_weighted": 0.04183884297520662,
      "eval_recall_macro": 0.14285714285714285,
      "eval_recall_micro": 0.20454545454545456,
      "eval_recall_weighted": 0.20454545454545456,
      "eval_runtime": 2.315,
      "eval_samples_per_second": 57.02,
      "eval_steps_per_second": 7.343,
      "step": 58
    },
    {
      "epoch": 2.0689655172413794,
      "grad_norm": 1.4094064235687256,
      "learning_rate": 9.96168582375479e-05,
      "loss": 1.9068,
      "step": 60
    },
    {
      "epoch": 2.1379310344827585,
      "grad_norm": 0.7938138246536255,
      "learning_rate": 9.92337164750958e-05,
      "loss": 1.847,
      "step": 62
    },
    {
      "epoch": 2.206896551724138,
      "grad_norm": 1.7750108242034912,
      "learning_rate": 9.885057471264369e-05,
      "loss": 1.8632,
      "step": 64
    },
    {
      "epoch": 2.2758620689655173,
      "grad_norm": 2.120215654373169,
      "learning_rate": 9.846743295019157e-05,
      "loss": 1.9374,
      "step": 66
    },
    {
      "epoch": 2.344827586206897,
      "grad_norm": 1.1951868534088135,
      "learning_rate": 9.808429118773947e-05,
      "loss": 1.789,
      "step": 68
    },
    {
      "epoch": 2.413793103448276,
      "grad_norm": 1.5802347660064697,
      "learning_rate": 9.770114942528736e-05,
      "loss": 1.8232,
      "step": 70
    },
    {
      "epoch": 2.4827586206896552,
      "grad_norm": 1.5989313125610352,
      "learning_rate": 9.731800766283526e-05,
      "loss": 1.9379,
      "step": 72
    },
    {
      "epoch": 2.5517241379310347,
      "grad_norm": 1.3577500581741333,
      "learning_rate": 9.693486590038314e-05,
      "loss": 1.905,
      "step": 74
    },
    {
      "epoch": 2.6206896551724137,
      "grad_norm": 1.2363930940628052,
      "learning_rate": 9.655172413793105e-05,
      "loss": 1.8514,
      "step": 76
    },
    {
      "epoch": 2.689655172413793,
      "grad_norm": 1.6196924448013306,
      "learning_rate": 9.616858237547893e-05,
      "loss": 1.9273,
      "step": 78
    },
    {
      "epoch": 2.7586206896551726,
      "grad_norm": 1.13129460811615,
      "learning_rate": 9.578544061302682e-05,
      "loss": 1.9407,
      "step": 80
    },
    {
      "epoch": 2.8275862068965516,
      "grad_norm": 1.0212408304214478,
      "learning_rate": 9.540229885057472e-05,
      "loss": 1.8273,
      "step": 82
    },
    {
      "epoch": 2.896551724137931,
      "grad_norm": 0.8608859777450562,
      "learning_rate": 9.501915708812261e-05,
      "loss": 1.8769,
      "step": 84
    },
    {
      "epoch": 2.9655172413793105,
      "grad_norm": 1.2102094888687134,
      "learning_rate": 9.463601532567051e-05,
      "loss": 1.9345,
      "step": 86
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.20454545454545456,
      "eval_f1_macro": 0.04851752021563342,
      "eval_f1_micro": 0.20454545454545456,
      "eval_f1_weighted": 0.06946826758147512,
      "eval_loss": 1.8624200820922852,
      "eval_precision_macro": 0.029220779220779224,
      "eval_precision_micro": 0.20454545454545456,
      "eval_precision_weighted": 0.04183884297520662,
      "eval_recall_macro": 0.14285714285714285,
      "eval_recall_micro": 0.20454545454545456,
      "eval_recall_weighted": 0.20454545454545456,
      "eval_runtime": 2.3224,
      "eval_samples_per_second": 56.838,
      "eval_steps_per_second": 7.32,
      "step": 87
    },
    {
      "epoch": 3.0344827586206895,
      "grad_norm": 1.7206976413726807,
      "learning_rate": 9.425287356321839e-05,
      "loss": 1.9875,
      "step": 88
    },
    {
      "epoch": 3.103448275862069,
      "grad_norm": 1.0909031629562378,
      "learning_rate": 9.38697318007663e-05,
      "loss": 1.7979,
      "step": 90
    },
    {
      "epoch": 3.1724137931034484,
      "grad_norm": 1.2875232696533203,
      "learning_rate": 9.348659003831418e-05,
      "loss": 1.8964,
      "step": 92
    },
    {
      "epoch": 3.2413793103448274,
      "grad_norm": 1.6919517517089844,
      "learning_rate": 9.310344827586207e-05,
      "loss": 1.8316,
      "step": 94
    },
    {
      "epoch": 3.310344827586207,
      "grad_norm": 1.7749555110931396,
      "learning_rate": 9.272030651340997e-05,
      "loss": 1.8712,
      "step": 96
    },
    {
      "epoch": 3.3793103448275863,
      "grad_norm": 1.273366093635559,
      "learning_rate": 9.233716475095786e-05,
      "loss": 1.8105,
      "step": 98
    },
    {
      "epoch": 3.4482758620689653,
      "grad_norm": 1.3420616388320923,
      "learning_rate": 9.195402298850575e-05,
      "loss": 1.7223,
      "step": 100
    },
    {
      "epoch": 3.5172413793103448,
      "grad_norm": 2.060871124267578,
      "learning_rate": 9.157088122605364e-05,
      "loss": 1.9009,
      "step": 102
    },
    {
      "epoch": 3.586206896551724,
      "grad_norm": 1.3717443943023682,
      "learning_rate": 9.118773946360154e-05,
      "loss": 1.7423,
      "step": 104
    },
    {
      "epoch": 3.655172413793103,
      "grad_norm": 2.35315203666687,
      "learning_rate": 9.080459770114943e-05,
      "loss": 1.8381,
      "step": 106
    },
    {
      "epoch": 3.7241379310344827,
      "grad_norm": 1.7000161409378052,
      "learning_rate": 9.042145593869731e-05,
      "loss": 1.6663,
      "step": 108
    },
    {
      "epoch": 3.793103448275862,
      "grad_norm": 2.3289718627929688,
      "learning_rate": 9.003831417624522e-05,
      "loss": 1.7653,
      "step": 110
    },
    {
      "epoch": 3.862068965517241,
      "grad_norm": 2.581888437271118,
      "learning_rate": 8.96551724137931e-05,
      "loss": 1.8899,
      "step": 112
    },
    {
      "epoch": 3.9310344827586206,
      "grad_norm": 2.3343546390533447,
      "learning_rate": 8.9272030651341e-05,
      "loss": 1.7854,
      "step": 114
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.989835262298584,
      "learning_rate": 8.888888888888889e-05,
      "loss": 1.6663,
      "step": 116
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3560606060606061,
      "eval_f1_macro": 0.2230450801010059,
      "eval_f1_micro": 0.3560606060606061,
      "eval_f1_weighted": 0.27375266721108504,
      "eval_loss": 1.7238999605178833,
      "eval_precision_macro": 0.31727902915703876,
      "eval_precision_micro": 0.3560606060606061,
      "eval_precision_weighted": 0.3548852832560697,
      "eval_recall_macro": 0.27251700680272106,
      "eval_recall_micro": 0.3560606060606061,
      "eval_recall_weighted": 0.3560606060606061,
      "eval_runtime": 2.3156,
      "eval_samples_per_second": 57.004,
      "eval_steps_per_second": 7.341,
      "step": 116
    },
    {
      "epoch": 4.068965517241379,
      "grad_norm": 2.951305389404297,
      "learning_rate": 8.850574712643679e-05,
      "loss": 1.6573,
      "step": 118
    },
    {
      "epoch": 4.137931034482759,
      "grad_norm": 2.0035061836242676,
      "learning_rate": 8.812260536398468e-05,
      "loss": 1.5898,
      "step": 120
    },
    {
      "epoch": 4.206896551724138,
      "grad_norm": 1.7943332195281982,
      "learning_rate": 8.773946360153256e-05,
      "loss": 1.5183,
      "step": 122
    },
    {
      "epoch": 4.275862068965517,
      "grad_norm": 2.8160884380340576,
      "learning_rate": 8.735632183908047e-05,
      "loss": 1.3985,
      "step": 124
    },
    {
      "epoch": 4.344827586206897,
      "grad_norm": 2.5276668071746826,
      "learning_rate": 8.697318007662835e-05,
      "loss": 1.5588,
      "step": 126
    },
    {
      "epoch": 4.413793103448276,
      "grad_norm": 1.8822064399719238,
      "learning_rate": 8.659003831417625e-05,
      "loss": 1.6377,
      "step": 128
    },
    {
      "epoch": 4.482758620689655,
      "grad_norm": 3.416983127593994,
      "learning_rate": 8.620689655172413e-05,
      "loss": 1.8148,
      "step": 130
    },
    {
      "epoch": 4.551724137931035,
      "grad_norm": 3.465373992919922,
      "learning_rate": 8.582375478927204e-05,
      "loss": 1.7206,
      "step": 132
    },
    {
      "epoch": 4.620689655172414,
      "grad_norm": 2.2419955730438232,
      "learning_rate": 8.544061302681992e-05,
      "loss": 1.8184,
      "step": 134
    },
    {
      "epoch": 4.689655172413794,
      "grad_norm": 2.772228240966797,
      "learning_rate": 8.505747126436782e-05,
      "loss": 1.6022,
      "step": 136
    },
    {
      "epoch": 4.758620689655173,
      "grad_norm": 2.6388156414031982,
      "learning_rate": 8.467432950191571e-05,
      "loss": 1.5941,
      "step": 138
    },
    {
      "epoch": 4.827586206896552,
      "grad_norm": 2.489682674407959,
      "learning_rate": 8.42911877394636e-05,
      "loss": 1.476,
      "step": 140
    },
    {
      "epoch": 4.896551724137931,
      "grad_norm": 2.517737865447998,
      "learning_rate": 8.39080459770115e-05,
      "loss": 1.5529,
      "step": 142
    },
    {
      "epoch": 4.9655172413793105,
      "grad_norm": 1.9954488277435303,
      "learning_rate": 8.35249042145594e-05,
      "loss": 1.3847,
      "step": 144
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4696969696969697,
      "eval_f1_macro": 0.3420460214829701,
      "eval_f1_micro": 0.4696969696969697,
      "eval_f1_weighted": 0.4038214256266929,
      "eval_loss": 1.4879895448684692,
      "eval_precision_macro": 0.4520649449711235,
      "eval_precision_micro": 0.4696969696969697,
      "eval_precision_weighted": 0.4845688001294408,
      "eval_recall_macro": 0.38925170068027215,
      "eval_recall_micro": 0.4696969696969697,
      "eval_recall_weighted": 0.4696969696969697,
      "eval_runtime": 2.3198,
      "eval_samples_per_second": 56.9,
      "eval_steps_per_second": 7.328,
      "step": 145
    },
    {
      "epoch": 5.0344827586206895,
      "grad_norm": 2.7297556400299072,
      "learning_rate": 8.314176245210729e-05,
      "loss": 1.6994,
      "step": 146
    },
    {
      "epoch": 5.103448275862069,
      "grad_norm": 2.7157135009765625,
      "learning_rate": 8.275862068965517e-05,
      "loss": 1.409,
      "step": 148
    },
    {
      "epoch": 5.172413793103448,
      "grad_norm": 3.6584227085113525,
      "learning_rate": 8.237547892720307e-05,
      "loss": 1.3627,
      "step": 150
    },
    {
      "epoch": 5.241379310344827,
      "grad_norm": 2.6041762828826904,
      "learning_rate": 8.199233716475096e-05,
      "loss": 1.3385,
      "step": 152
    },
    {
      "epoch": 5.310344827586207,
      "grad_norm": 2.7281856536865234,
      "learning_rate": 8.160919540229886e-05,
      "loss": 1.4468,
      "step": 154
    },
    {
      "epoch": 5.379310344827586,
      "grad_norm": 2.5631356239318848,
      "learning_rate": 8.122605363984674e-05,
      "loss": 1.3639,
      "step": 156
    },
    {
      "epoch": 5.448275862068965,
      "grad_norm": 3.118602991104126,
      "learning_rate": 8.084291187739465e-05,
      "loss": 1.5097,
      "step": 158
    },
    {
      "epoch": 5.517241379310345,
      "grad_norm": 2.653867483139038,
      "learning_rate": 8.045977011494253e-05,
      "loss": 1.3556,
      "step": 160
    },
    {
      "epoch": 5.586206896551724,
      "grad_norm": 2.3653130531311035,
      "learning_rate": 8.007662835249042e-05,
      "loss": 1.504,
      "step": 162
    },
    {
      "epoch": 5.655172413793103,
      "grad_norm": 2.039597749710083,
      "learning_rate": 7.969348659003832e-05,
      "loss": 1.3222,
      "step": 164
    },
    {
      "epoch": 5.724137931034483,
      "grad_norm": 2.91522216796875,
      "learning_rate": 7.931034482758621e-05,
      "loss": 1.2761,
      "step": 166
    },
    {
      "epoch": 5.793103448275862,
      "grad_norm": 2.8554606437683105,
      "learning_rate": 7.892720306513411e-05,
      "loss": 1.3019,
      "step": 168
    },
    {
      "epoch": 5.862068965517241,
      "grad_norm": 3.730526924133301,
      "learning_rate": 7.854406130268199e-05,
      "loss": 1.2626,
      "step": 170
    },
    {
      "epoch": 5.931034482758621,
      "grad_norm": 5.193080425262451,
      "learning_rate": 7.81609195402299e-05,
      "loss": 1.5394,
      "step": 172
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.901132583618164,
      "learning_rate": 7.777777777777778e-05,
      "loss": 1.6559,
      "step": 174
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4772727272727273,
      "eval_f1_macro": 0.3479108279730525,
      "eval_f1_micro": 0.4772727272727273,
      "eval_f1_weighted": 0.4108418905152118,
      "eval_loss": 1.4056475162506104,
      "eval_precision_macro": 0.3864548957927566,
      "eval_precision_micro": 0.4772727272727273,
      "eval_precision_weighted": 0.42757837676034927,
      "eval_recall_macro": 0.38701436130007555,
      "eval_recall_micro": 0.4772727272727273,
      "eval_recall_weighted": 0.4772727272727273,
      "eval_runtime": 2.3209,
      "eval_samples_per_second": 56.874,
      "eval_steps_per_second": 7.325,
      "step": 174
    },
    {
      "epoch": 6.068965517241379,
      "grad_norm": 3.3514347076416016,
      "learning_rate": 7.739463601532568e-05,
      "loss": 1.2022,
      "step": 176
    },
    {
      "epoch": 6.137931034482759,
      "grad_norm": 2.6016125679016113,
      "learning_rate": 7.701149425287356e-05,
      "loss": 1.1793,
      "step": 178
    },
    {
      "epoch": 6.206896551724138,
      "grad_norm": 3.9388816356658936,
      "learning_rate": 7.662835249042147e-05,
      "loss": 1.2841,
      "step": 180
    },
    {
      "epoch": 6.275862068965517,
      "grad_norm": 2.01733660697937,
      "learning_rate": 7.624521072796935e-05,
      "loss": 1.3875,
      "step": 182
    },
    {
      "epoch": 6.344827586206897,
      "grad_norm": 2.873060703277588,
      "learning_rate": 7.586206896551724e-05,
      "loss": 1.3029,
      "step": 184
    },
    {
      "epoch": 6.413793103448276,
      "grad_norm": 3.8860301971435547,
      "learning_rate": 7.547892720306514e-05,
      "loss": 1.0694,
      "step": 186
    },
    {
      "epoch": 6.482758620689655,
      "grad_norm": 4.136936664581299,
      "learning_rate": 7.509578544061303e-05,
      "loss": 1.0778,
      "step": 188
    },
    {
      "epoch": 6.551724137931035,
      "grad_norm": 2.326791763305664,
      "learning_rate": 7.471264367816091e-05,
      "loss": 1.2761,
      "step": 190
    },
    {
      "epoch": 6.620689655172414,
      "grad_norm": 3.5550026893615723,
      "learning_rate": 7.432950191570882e-05,
      "loss": 1.3036,
      "step": 192
    },
    {
      "epoch": 6.689655172413794,
      "grad_norm": 2.3610475063323975,
      "learning_rate": 7.39463601532567e-05,
      "loss": 1.2765,
      "step": 194
    },
    {
      "epoch": 6.758620689655173,
      "grad_norm": 3.0064663887023926,
      "learning_rate": 7.35632183908046e-05,
      "loss": 1.5579,
      "step": 196
    },
    {
      "epoch": 6.827586206896552,
      "grad_norm": 2.8558247089385986,
      "learning_rate": 7.31800766283525e-05,
      "loss": 1.1987,
      "step": 198
    },
    {
      "epoch": 6.896551724137931,
      "grad_norm": 3.3801021575927734,
      "learning_rate": 7.279693486590039e-05,
      "loss": 1.1622,
      "step": 200
    },
    {
      "epoch": 6.9655172413793105,
      "grad_norm": 3.4124093055725098,
      "learning_rate": 7.241379310344828e-05,
      "loss": 1.335,
      "step": 202
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5151515151515151,
      "eval_f1_macro": 0.38747536212974004,
      "eval_f1_micro": 0.5151515151515151,
      "eval_f1_weighted": 0.4526784325171422,
      "eval_loss": 1.3767547607421875,
      "eval_precision_macro": 0.3933148791431466,
      "eval_precision_micro": 0.5151515151515151,
      "eval_precision_weighted": 0.44469956019158147,
      "eval_recall_macro": 0.4265457294028723,
      "eval_recall_micro": 0.5151515151515151,
      "eval_recall_weighted": 0.5151515151515151,
      "eval_runtime": 2.312,
      "eval_samples_per_second": 57.094,
      "eval_steps_per_second": 7.353,
      "step": 203
    },
    {
      "epoch": 7.0344827586206895,
      "grad_norm": 3.0827553272247314,
      "learning_rate": 7.203065134099617e-05,
      "loss": 1.0667,
      "step": 204
    },
    {
      "epoch": 7.103448275862069,
      "grad_norm": 1.9460029602050781,
      "learning_rate": 7.164750957854408e-05,
      "loss": 1.0915,
      "step": 206
    },
    {
      "epoch": 7.172413793103448,
      "grad_norm": 3.7591891288757324,
      "learning_rate": 7.126436781609196e-05,
      "loss": 1.2813,
      "step": 208
    },
    {
      "epoch": 7.241379310344827,
      "grad_norm": 2.6194112300872803,
      "learning_rate": 7.088122605363985e-05,
      "loss": 1.1527,
      "step": 210
    },
    {
      "epoch": 7.310344827586207,
      "grad_norm": 3.020498514175415,
      "learning_rate": 7.049808429118773e-05,
      "loss": 1.2731,
      "step": 212
    },
    {
      "epoch": 7.379310344827586,
      "grad_norm": 3.303375005722046,
      "learning_rate": 7.011494252873564e-05,
      "loss": 1.0312,
      "step": 214
    },
    {
      "epoch": 7.448275862068965,
      "grad_norm": 2.705366849899292,
      "learning_rate": 6.973180076628352e-05,
      "loss": 1.2032,
      "step": 216
    },
    {
      "epoch": 7.517241379310345,
      "grad_norm": 3.3683817386627197,
      "learning_rate": 6.934865900383142e-05,
      "loss": 1.166,
      "step": 218
    },
    {
      "epoch": 7.586206896551724,
      "grad_norm": 3.079350471496582,
      "learning_rate": 6.896551724137931e-05,
      "loss": 0.9884,
      "step": 220
    },
    {
      "epoch": 7.655172413793103,
      "grad_norm": 2.3934454917907715,
      "learning_rate": 6.858237547892721e-05,
      "loss": 1.1715,
      "step": 222
    },
    {
      "epoch": 7.724137931034483,
      "grad_norm": 3.6560773849487305,
      "learning_rate": 6.81992337164751e-05,
      "loss": 1.0007,
      "step": 224
    },
    {
      "epoch": 7.793103448275862,
      "grad_norm": 4.3467912673950195,
      "learning_rate": 6.781609195402298e-05,
      "loss": 1.4675,
      "step": 226
    },
    {
      "epoch": 7.862068965517241,
      "grad_norm": 2.5141751766204834,
      "learning_rate": 6.74329501915709e-05,
      "loss": 0.9672,
      "step": 228
    },
    {
      "epoch": 7.931034482758621,
      "grad_norm": 4.107297897338867,
      "learning_rate": 6.704980842911877e-05,
      "loss": 1.4593,
      "step": 230
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.193042278289795,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.2514,
      "step": 232
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5606060606060606,
      "eval_f1_macro": 0.4536205964777393,
      "eval_f1_micro": 0.5606060606060606,
      "eval_f1_weighted": 0.5206725615816525,
      "eval_loss": 1.2344759702682495,
      "eval_precision_macro": 0.47009218749545173,
      "eval_precision_micro": 0.5606060606060606,
      "eval_precision_weighted": 0.5257434050758826,
      "eval_recall_macro": 0.4766061980347694,
      "eval_recall_micro": 0.5606060606060606,
      "eval_recall_weighted": 0.5606060606060606,
      "eval_runtime": 2.3242,
      "eval_samples_per_second": 56.794,
      "eval_steps_per_second": 7.314,
      "step": 232
    },
    {
      "epoch": 8.068965517241379,
      "grad_norm": 3.166414260864258,
      "learning_rate": 6.628352490421456e-05,
      "loss": 0.8487,
      "step": 234
    },
    {
      "epoch": 8.137931034482758,
      "grad_norm": 3.4050002098083496,
      "learning_rate": 6.590038314176246e-05,
      "loss": 1.2154,
      "step": 236
    },
    {
      "epoch": 8.206896551724139,
      "grad_norm": 2.827500104904175,
      "learning_rate": 6.551724137931034e-05,
      "loss": 1.1884,
      "step": 238
    },
    {
      "epoch": 8.275862068965518,
      "grad_norm": 3.644810199737549,
      "learning_rate": 6.513409961685824e-05,
      "loss": 1.1444,
      "step": 240
    },
    {
      "epoch": 8.344827586206897,
      "grad_norm": 2.5931830406188965,
      "learning_rate": 6.475095785440613e-05,
      "loss": 0.9335,
      "step": 242
    },
    {
      "epoch": 8.413793103448276,
      "grad_norm": 4.586955547332764,
      "learning_rate": 6.436781609195403e-05,
      "loss": 1.174,
      "step": 244
    },
    {
      "epoch": 8.482758620689655,
      "grad_norm": 4.06891393661499,
      "learning_rate": 6.398467432950191e-05,
      "loss": 1.1579,
      "step": 246
    },
    {
      "epoch": 8.551724137931034,
      "grad_norm": 3.6428885459899902,
      "learning_rate": 6.360153256704982e-05,
      "loss": 0.8764,
      "step": 248
    },
    {
      "epoch": 8.620689655172415,
      "grad_norm": 6.163786888122559,
      "learning_rate": 6.32183908045977e-05,
      "loss": 1.0204,
      "step": 250
    },
    {
      "epoch": 8.689655172413794,
      "grad_norm": 2.775923252105713,
      "learning_rate": 6.283524904214559e-05,
      "loss": 0.8183,
      "step": 252
    },
    {
      "epoch": 8.758620689655173,
      "grad_norm": 3.2083027362823486,
      "learning_rate": 6.245210727969349e-05,
      "loss": 1.3757,
      "step": 254
    },
    {
      "epoch": 8.827586206896552,
      "grad_norm": 5.378493309020996,
      "learning_rate": 6.206896551724138e-05,
      "loss": 1.0219,
      "step": 256
    },
    {
      "epoch": 8.89655172413793,
      "grad_norm": 3.8027050495147705,
      "learning_rate": 6.168582375478928e-05,
      "loss": 1.0405,
      "step": 258
    },
    {
      "epoch": 8.96551724137931,
      "grad_norm": 3.604165554046631,
      "learning_rate": 6.130268199233716e-05,
      "loss": 0.6979,
      "step": 260
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1_macro": 0.530518587020135,
      "eval_f1_micro": 0.6363636363636364,
      "eval_f1_weighted": 0.6097181236324214,
      "eval_loss": 1.1500554084777832,
      "eval_precision_macro": 0.5490794091920358,
      "eval_precision_micro": 0.6363636363636364,
      "eval_precision_weighted": 0.6127338831200314,
      "eval_recall_macro": 0.5391232048374907,
      "eval_recall_micro": 0.6363636363636364,
      "eval_recall_weighted": 0.6363636363636364,
      "eval_runtime": 2.3286,
      "eval_samples_per_second": 56.687,
      "eval_steps_per_second": 7.301,
      "step": 261
    },
    {
      "epoch": 9.03448275862069,
      "grad_norm": 2.9546728134155273,
      "learning_rate": 6.091954022988506e-05,
      "loss": 0.9733,
      "step": 262
    },
    {
      "epoch": 9.10344827586207,
      "grad_norm": 3.742382049560547,
      "learning_rate": 6.053639846743295e-05,
      "loss": 0.8539,
      "step": 264
    },
    {
      "epoch": 9.172413793103448,
      "grad_norm": 4.167463779449463,
      "learning_rate": 6.0153256704980845e-05,
      "loss": 0.9605,
      "step": 266
    },
    {
      "epoch": 9.241379310344827,
      "grad_norm": 2.0738613605499268,
      "learning_rate": 5.977011494252874e-05,
      "loss": 0.7348,
      "step": 268
    },
    {
      "epoch": 9.310344827586206,
      "grad_norm": 3.2274346351623535,
      "learning_rate": 5.9386973180076635e-05,
      "loss": 0.9138,
      "step": 270
    },
    {
      "epoch": 9.379310344827585,
      "grad_norm": 3.5369961261749268,
      "learning_rate": 5.900383141762452e-05,
      "loss": 0.7026,
      "step": 272
    },
    {
      "epoch": 9.448275862068966,
      "grad_norm": 3.977078914642334,
      "learning_rate": 5.862068965517241e-05,
      "loss": 1.0063,
      "step": 274
    },
    {
      "epoch": 9.517241379310345,
      "grad_norm": 3.6526713371276855,
      "learning_rate": 5.823754789272031e-05,
      "loss": 0.7266,
      "step": 276
    },
    {
      "epoch": 9.586206896551724,
      "grad_norm": 1.9557095766067505,
      "learning_rate": 5.78544061302682e-05,
      "loss": 0.7732,
      "step": 278
    },
    {
      "epoch": 9.655172413793103,
      "grad_norm": 1.2573161125183105,
      "learning_rate": 5.747126436781609e-05,
      "loss": 0.5661,
      "step": 280
    },
    {
      "epoch": 9.724137931034482,
      "grad_norm": 2.927410364151001,
      "learning_rate": 5.708812260536399e-05,
      "loss": 1.0745,
      "step": 282
    },
    {
      "epoch": 9.793103448275861,
      "grad_norm": 4.574418544769287,
      "learning_rate": 5.670498084291188e-05,
      "loss": 0.9002,
      "step": 284
    },
    {
      "epoch": 9.862068965517242,
      "grad_norm": 3.5535728931427,
      "learning_rate": 5.632183908045977e-05,
      "loss": 0.9265,
      "step": 286
    },
    {
      "epoch": 9.931034482758621,
      "grad_norm": 4.7521162033081055,
      "learning_rate": 5.593869731800766e-05,
      "loss": 0.8453,
      "step": 288
    },
    {
      "epoch": 10.0,
      "grad_norm": 4.0816216468811035,
      "learning_rate": 5.555555555555556e-05,
      "loss": 1.0417,
      "step": 290
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6136363636363636,
      "eval_f1_macro": 0.5205916396815399,
      "eval_f1_micro": 0.6136363636363636,
      "eval_f1_weighted": 0.5899988122518645,
      "eval_loss": 1.1654376983642578,
      "eval_precision_macro": 0.5214674214674214,
      "eval_precision_micro": 0.6136363636363636,
      "eval_precision_weighted": 0.5934943844034754,
      "eval_recall_macro": 0.5463643235071807,
      "eval_recall_micro": 0.6136363636363636,
      "eval_recall_weighted": 0.6136363636363636,
      "eval_runtime": 2.3259,
      "eval_samples_per_second": 56.753,
      "eval_steps_per_second": 7.309,
      "step": 290
    },
    {
      "epoch": 10.068965517241379,
      "grad_norm": 4.20100736618042,
      "learning_rate": 5.517241379310345e-05,
      "loss": 0.9384,
      "step": 292
    },
    {
      "epoch": 10.137931034482758,
      "grad_norm": 3.5482981204986572,
      "learning_rate": 5.478927203065134e-05,
      "loss": 0.7163,
      "step": 294
    },
    {
      "epoch": 10.206896551724139,
      "grad_norm": 5.511715412139893,
      "learning_rate": 5.440613026819924e-05,
      "loss": 1.0096,
      "step": 296
    },
    {
      "epoch": 10.275862068965518,
      "grad_norm": 4.4264984130859375,
      "learning_rate": 5.402298850574713e-05,
      "loss": 0.775,
      "step": 298
    },
    {
      "epoch": 10.344827586206897,
      "grad_norm": 2.2772464752197266,
      "learning_rate": 5.363984674329502e-05,
      "loss": 1.0071,
      "step": 300
    },
    {
      "epoch": 10.413793103448276,
      "grad_norm": 2.6942243576049805,
      "learning_rate": 5.325670498084292e-05,
      "loss": 0.6,
      "step": 302
    },
    {
      "epoch": 10.482758620689655,
      "grad_norm": 3.3119990825653076,
      "learning_rate": 5.287356321839081e-05,
      "loss": 0.7233,
      "step": 304
    },
    {
      "epoch": 10.551724137931034,
      "grad_norm": 3.8363144397735596,
      "learning_rate": 5.24904214559387e-05,
      "loss": 1.084,
      "step": 306
    },
    {
      "epoch": 10.620689655172415,
      "grad_norm": 3.457904815673828,
      "learning_rate": 5.2107279693486586e-05,
      "loss": 0.7188,
      "step": 308
    },
    {
      "epoch": 10.689655172413794,
      "grad_norm": 4.752615451812744,
      "learning_rate": 5.172413793103449e-05,
      "loss": 1.0076,
      "step": 310
    },
    {
      "epoch": 10.758620689655173,
      "grad_norm": 2.8279869556427,
      "learning_rate": 5.1340996168582377e-05,
      "loss": 1.0505,
      "step": 312
    },
    {
      "epoch": 10.827586206896552,
      "grad_norm": 5.538084506988525,
      "learning_rate": 5.0957854406130265e-05,
      "loss": 0.6474,
      "step": 314
    },
    {
      "epoch": 10.89655172413793,
      "grad_norm": 3.197183847427368,
      "learning_rate": 5.057471264367817e-05,
      "loss": 0.6146,
      "step": 316
    },
    {
      "epoch": 10.96551724137931,
      "grad_norm": 3.427384614944458,
      "learning_rate": 5.0191570881226055e-05,
      "loss": 0.7314,
      "step": 318
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6212121212121212,
      "eval_f1_macro": 0.537594465565654,
      "eval_f1_micro": 0.6212121212121212,
      "eval_f1_weighted": 0.6109434365736887,
      "eval_loss": 1.1566157341003418,
      "eval_precision_macro": 0.5386796536796536,
      "eval_precision_micro": 0.6212121212121212,
      "eval_precision_weighted": 0.6153512396694215,
      "eval_recall_macro": 0.5513832199546485,
      "eval_recall_micro": 0.6212121212121212,
      "eval_recall_weighted": 0.6212121212121212,
      "eval_runtime": 2.3247,
      "eval_samples_per_second": 56.78,
      "eval_steps_per_second": 7.313,
      "step": 319
    },
    {
      "epoch": 11.03448275862069,
      "grad_norm": 3.1624715328216553,
      "learning_rate": 4.980842911877395e-05,
      "loss": 0.5936,
      "step": 320
    },
    {
      "epoch": 11.10344827586207,
      "grad_norm": 3.2586793899536133,
      "learning_rate": 4.9425287356321845e-05,
      "loss": 0.7724,
      "step": 322
    },
    {
      "epoch": 11.172413793103448,
      "grad_norm": 4.481448173522949,
      "learning_rate": 4.904214559386973e-05,
      "loss": 0.6748,
      "step": 324
    },
    {
      "epoch": 11.241379310344827,
      "grad_norm": 2.7534396648406982,
      "learning_rate": 4.865900383141763e-05,
      "loss": 0.5199,
      "step": 326
    },
    {
      "epoch": 11.310344827586206,
      "grad_norm": 2.4173717498779297,
      "learning_rate": 4.827586206896552e-05,
      "loss": 0.8016,
      "step": 328
    },
    {
      "epoch": 11.379310344827585,
      "grad_norm": 2.85591459274292,
      "learning_rate": 4.789272030651341e-05,
      "loss": 0.8319,
      "step": 330
    },
    {
      "epoch": 11.448275862068966,
      "grad_norm": 5.3054585456848145,
      "learning_rate": 4.7509578544061307e-05,
      "loss": 1.0745,
      "step": 332
    },
    {
      "epoch": 11.517241379310345,
      "grad_norm": 3.0006418228149414,
      "learning_rate": 4.7126436781609195e-05,
      "loss": 0.5067,
      "step": 334
    },
    {
      "epoch": 11.586206896551724,
      "grad_norm": 3.4974889755249023,
      "learning_rate": 4.674329501915709e-05,
      "loss": 0.7971,
      "step": 336
    },
    {
      "epoch": 11.655172413793103,
      "grad_norm": 3.117962598800659,
      "learning_rate": 4.6360153256704985e-05,
      "loss": 0.6665,
      "step": 338
    },
    {
      "epoch": 11.724137931034482,
      "grad_norm": 4.9846343994140625,
      "learning_rate": 4.597701149425287e-05,
      "loss": 0.7402,
      "step": 340
    },
    {
      "epoch": 11.793103448275861,
      "grad_norm": 6.1209001541137695,
      "learning_rate": 4.559386973180077e-05,
      "loss": 0.9042,
      "step": 342
    },
    {
      "epoch": 11.862068965517242,
      "grad_norm": 2.3766987323760986,
      "learning_rate": 4.5210727969348656e-05,
      "loss": 0.6191,
      "step": 344
    },
    {
      "epoch": 11.931034482758621,
      "grad_norm": 5.1854143142700195,
      "learning_rate": 4.482758620689655e-05,
      "loss": 0.6116,
      "step": 346
    },
    {
      "epoch": 12.0,
      "grad_norm": 5.495875835418701,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.7902,
      "step": 348
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6212121212121212,
      "eval_f1_macro": 0.5396961800902688,
      "eval_f1_micro": 0.6212121212121212,
      "eval_f1_weighted": 0.6140481105214648,
      "eval_loss": 1.1624207496643066,
      "eval_precision_macro": 0.5422025161353321,
      "eval_precision_micro": 0.6212121212121212,
      "eval_precision_weighted": 0.6208899459169562,
      "eval_recall_macro": 0.5504686318972033,
      "eval_recall_micro": 0.6212121212121212,
      "eval_recall_weighted": 0.6212121212121212,
      "eval_runtime": 2.3261,
      "eval_samples_per_second": 56.747,
      "eval_steps_per_second": 7.308,
      "step": 348
    },
    {
      "epoch": 12.068965517241379,
      "grad_norm": 4.319167613983154,
      "learning_rate": 4.406130268199234e-05,
      "loss": 0.7979,
      "step": 350
    },
    {
      "epoch": 12.137931034482758,
      "grad_norm": 3.6471261978149414,
      "learning_rate": 4.367816091954024e-05,
      "loss": 0.7733,
      "step": 352
    },
    {
      "epoch": 12.206896551724139,
      "grad_norm": 3.5435140132904053,
      "learning_rate": 4.3295019157088125e-05,
      "loss": 0.8753,
      "step": 354
    },
    {
      "epoch": 12.275862068965518,
      "grad_norm": 3.007453441619873,
      "learning_rate": 4.291187739463602e-05,
      "loss": 0.5425,
      "step": 356
    },
    {
      "epoch": 12.344827586206897,
      "grad_norm": 4.487671852111816,
      "learning_rate": 4.252873563218391e-05,
      "loss": 0.5562,
      "step": 358
    },
    {
      "epoch": 12.413793103448276,
      "grad_norm": 3.367605447769165,
      "learning_rate": 4.21455938697318e-05,
      "loss": 0.625,
      "step": 360
    },
    {
      "epoch": 12.482758620689655,
      "grad_norm": 5.727989673614502,
      "learning_rate": 4.17624521072797e-05,
      "loss": 0.6552,
      "step": 362
    },
    {
      "epoch": 12.551724137931034,
      "grad_norm": 2.801624298095703,
      "learning_rate": 4.1379310344827587e-05,
      "loss": 0.5104,
      "step": 364
    },
    {
      "epoch": 12.620689655172415,
      "grad_norm": 4.306697368621826,
      "learning_rate": 4.099616858237548e-05,
      "loss": 0.6764,
      "step": 366
    },
    {
      "epoch": 12.689655172413794,
      "grad_norm": 7.026360988616943,
      "learning_rate": 4.061302681992337e-05,
      "loss": 0.7224,
      "step": 368
    },
    {
      "epoch": 12.758620689655173,
      "grad_norm": 5.398407936096191,
      "learning_rate": 4.0229885057471265e-05,
      "loss": 0.8992,
      "step": 370
    },
    {
      "epoch": 12.827586206896552,
      "grad_norm": 2.986201524734497,
      "learning_rate": 3.984674329501916e-05,
      "loss": 0.5226,
      "step": 372
    },
    {
      "epoch": 12.89655172413793,
      "grad_norm": 5.425571918487549,
      "learning_rate": 3.9463601532567055e-05,
      "loss": 0.6649,
      "step": 374
    },
    {
      "epoch": 12.96551724137931,
      "grad_norm": 4.2015461921691895,
      "learning_rate": 3.908045977011495e-05,
      "loss": 0.7503,
      "step": 376
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6287878787878788,
      "eval_f1_macro": 0.5376898842937335,
      "eval_f1_micro": 0.6287878787878788,
      "eval_f1_weighted": 0.6125773067483171,
      "eval_loss": 1.135877013206482,
      "eval_precision_macro": 0.5471859541424758,
      "eval_precision_micro": 0.6287878787878788,
      "eval_precision_weighted": 0.6142989641013356,
      "eval_recall_macro": 0.5455328798185941,
      "eval_recall_micro": 0.6287878787878788,
      "eval_recall_weighted": 0.6287878787878788,
      "eval_runtime": 2.3097,
      "eval_samples_per_second": 57.149,
      "eval_steps_per_second": 7.36,
      "step": 377
    },
    {
      "epoch": 13.03448275862069,
      "grad_norm": 3.527031660079956,
      "learning_rate": 3.869731800766284e-05,
      "loss": 0.6196,
      "step": 378
    },
    {
      "epoch": 13.10344827586207,
      "grad_norm": 4.090692520141602,
      "learning_rate": 3.831417624521073e-05,
      "loss": 0.4906,
      "step": 380
    },
    {
      "epoch": 13.172413793103448,
      "grad_norm": 3.5445234775543213,
      "learning_rate": 3.793103448275862e-05,
      "loss": 0.6832,
      "step": 382
    },
    {
      "epoch": 13.241379310344827,
      "grad_norm": 2.648242473602295,
      "learning_rate": 3.7547892720306517e-05,
      "loss": 0.7041,
      "step": 384
    },
    {
      "epoch": 13.310344827586206,
      "grad_norm": 2.421206474304199,
      "learning_rate": 3.716475095785441e-05,
      "loss": 0.3506,
      "step": 386
    },
    {
      "epoch": 13.379310344827585,
      "grad_norm": 4.197497367858887,
      "learning_rate": 3.67816091954023e-05,
      "loss": 0.7564,
      "step": 388
    },
    {
      "epoch": 13.448275862068966,
      "grad_norm": 4.1512131690979,
      "learning_rate": 3.6398467432950195e-05,
      "loss": 0.7134,
      "step": 390
    },
    {
      "epoch": 13.517241379310345,
      "grad_norm": 2.6926424503326416,
      "learning_rate": 3.601532567049808e-05,
      "loss": 0.5195,
      "step": 392
    },
    {
      "epoch": 13.586206896551724,
      "grad_norm": 5.174130916595459,
      "learning_rate": 3.563218390804598e-05,
      "loss": 0.485,
      "step": 394
    },
    {
      "epoch": 13.655172413793103,
      "grad_norm": 2.1289005279541016,
      "learning_rate": 3.5249042145593867e-05,
      "loss": 0.6963,
      "step": 396
    },
    {
      "epoch": 13.724137931034482,
      "grad_norm": 5.4137654304504395,
      "learning_rate": 3.486590038314176e-05,
      "loss": 0.6374,
      "step": 398
    },
    {
      "epoch": 13.793103448275861,
      "grad_norm": 3.6843011379241943,
      "learning_rate": 3.4482758620689657e-05,
      "loss": 0.9498,
      "step": 400
    },
    {
      "epoch": 13.862068965517242,
      "grad_norm": 4.690808296203613,
      "learning_rate": 3.409961685823755e-05,
      "loss": 0.7182,
      "step": 402
    },
    {
      "epoch": 13.931034482758621,
      "grad_norm": 4.772737979888916,
      "learning_rate": 3.371647509578545e-05,
      "loss": 0.7981,
      "step": 404
    },
    {
      "epoch": 14.0,
      "grad_norm": 6.463994979858398,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.586,
      "step": 406
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6287878787878788,
      "eval_f1_macro": 0.5440812816073651,
      "eval_f1_micro": 0.6287878787878788,
      "eval_f1_weighted": 0.614080662953901,
      "eval_loss": 1.151163101196289,
      "eval_precision_macro": 0.5361375661375661,
      "eval_precision_micro": 0.6287878787878788,
      "eval_precision_weighted": 0.6032631874298542,
      "eval_recall_macro": 0.5557218442932729,
      "eval_recall_micro": 0.6287878787878788,
      "eval_recall_weighted": 0.6287878787878788,
      "eval_runtime": 2.3161,
      "eval_samples_per_second": 56.992,
      "eval_steps_per_second": 7.34,
      "step": 406
    },
    {
      "epoch": 14.068965517241379,
      "grad_norm": 4.363000392913818,
      "learning_rate": 3.295019157088123e-05,
      "loss": 0.5204,
      "step": 408
    },
    {
      "epoch": 14.137931034482758,
      "grad_norm": 2.4504146575927734,
      "learning_rate": 3.256704980842912e-05,
      "loss": 0.3989,
      "step": 410
    },
    {
      "epoch": 14.206896551724139,
      "grad_norm": 2.4213919639587402,
      "learning_rate": 3.218390804597701e-05,
      "loss": 0.4246,
      "step": 412
    },
    {
      "epoch": 14.275862068965518,
      "grad_norm": 4.8299031257629395,
      "learning_rate": 3.180076628352491e-05,
      "loss": 0.6999,
      "step": 414
    },
    {
      "epoch": 14.344827586206897,
      "grad_norm": 2.637255907058716,
      "learning_rate": 3.1417624521072797e-05,
      "loss": 0.5376,
      "step": 416
    },
    {
      "epoch": 14.413793103448276,
      "grad_norm": 3.4667327404022217,
      "learning_rate": 3.103448275862069e-05,
      "loss": 0.5109,
      "step": 418
    },
    {
      "epoch": 14.482758620689655,
      "grad_norm": 6.673337459564209,
      "learning_rate": 3.065134099616858e-05,
      "loss": 0.4014,
      "step": 420
    },
    {
      "epoch": 14.551724137931034,
      "grad_norm": 2.5311481952667236,
      "learning_rate": 3.0268199233716475e-05,
      "loss": 0.6741,
      "step": 422
    },
    {
      "epoch": 14.620689655172415,
      "grad_norm": 3.339940071105957,
      "learning_rate": 2.988505747126437e-05,
      "loss": 0.4081,
      "step": 424
    },
    {
      "epoch": 14.689655172413794,
      "grad_norm": 2.3433589935302734,
      "learning_rate": 2.950191570881226e-05,
      "loss": 0.5574,
      "step": 426
    },
    {
      "epoch": 14.758620689655173,
      "grad_norm": 3.5291073322296143,
      "learning_rate": 2.9118773946360157e-05,
      "loss": 0.6136,
      "step": 428
    },
    {
      "epoch": 14.827586206896552,
      "grad_norm": 3.0121891498565674,
      "learning_rate": 2.8735632183908045e-05,
      "loss": 0.3841,
      "step": 430
    },
    {
      "epoch": 14.89655172413793,
      "grad_norm": 3.474233388900757,
      "learning_rate": 2.835249042145594e-05,
      "loss": 0.4371,
      "step": 432
    },
    {
      "epoch": 14.96551724137931,
      "grad_norm": 3.466447591781616,
      "learning_rate": 2.796934865900383e-05,
      "loss": 0.6869,
      "step": 434
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6287878787878788,
      "eval_f1_macro": 0.5323304385684055,
      "eval_f1_micro": 0.6287878787878788,
      "eval_f1_weighted": 0.6116664108577855,
      "eval_loss": 1.13055419921875,
      "eval_precision_macro": 0.5269524536765916,
      "eval_precision_micro": 0.6287878787878788,
      "eval_precision_weighted": 0.6043089290542268,
      "eval_recall_macro": 0.547490551776266,
      "eval_recall_micro": 0.6287878787878788,
      "eval_recall_weighted": 0.6287878787878788,
      "eval_runtime": 2.3194,
      "eval_samples_per_second": 56.911,
      "eval_steps_per_second": 7.329,
      "step": 435
    },
    {
      "epoch": 15.03448275862069,
      "grad_norm": 2.4587948322296143,
      "learning_rate": 2.7586206896551727e-05,
      "loss": 0.6758,
      "step": 436
    },
    {
      "epoch": 15.10344827586207,
      "grad_norm": 3.373725414276123,
      "learning_rate": 2.720306513409962e-05,
      "loss": 0.516,
      "step": 438
    },
    {
      "epoch": 15.172413793103448,
      "grad_norm": 4.587845802307129,
      "learning_rate": 2.681992337164751e-05,
      "loss": 0.5093,
      "step": 440
    },
    {
      "epoch": 15.241379310344827,
      "grad_norm": 6.124247074127197,
      "learning_rate": 2.6436781609195405e-05,
      "loss": 0.5237,
      "step": 442
    },
    {
      "epoch": 15.310344827586206,
      "grad_norm": 3.777029514312744,
      "learning_rate": 2.6053639846743293e-05,
      "loss": 0.5128,
      "step": 444
    },
    {
      "epoch": 15.379310344827585,
      "grad_norm": 2.960836887359619,
      "learning_rate": 2.5670498084291188e-05,
      "loss": 0.6366,
      "step": 446
    },
    {
      "epoch": 15.448275862068966,
      "grad_norm": 3.1089887619018555,
      "learning_rate": 2.5287356321839083e-05,
      "loss": 0.5369,
      "step": 448
    },
    {
      "epoch": 15.517241379310345,
      "grad_norm": 2.6847426891326904,
      "learning_rate": 2.4904214559386975e-05,
      "loss": 0.3325,
      "step": 450
    },
    {
      "epoch": 15.586206896551724,
      "grad_norm": 6.938262462615967,
      "learning_rate": 2.4521072796934867e-05,
      "loss": 0.6207,
      "step": 452
    },
    {
      "epoch": 15.655172413793103,
      "grad_norm": 1.7349634170532227,
      "learning_rate": 2.413793103448276e-05,
      "loss": 0.3663,
      "step": 454
    },
    {
      "epoch": 15.724137931034482,
      "grad_norm": 1.5232642889022827,
      "learning_rate": 2.3754789272030653e-05,
      "loss": 0.5473,
      "step": 456
    },
    {
      "epoch": 15.793103448275861,
      "grad_norm": 4.805527210235596,
      "learning_rate": 2.3371647509578545e-05,
      "loss": 0.3652,
      "step": 458
    },
    {
      "epoch": 15.862068965517242,
      "grad_norm": 4.195865631103516,
      "learning_rate": 2.2988505747126437e-05,
      "loss": 0.5624,
      "step": 460
    },
    {
      "epoch": 15.931034482758621,
      "grad_norm": 1.5092368125915527,
      "learning_rate": 2.2605363984674328e-05,
      "loss": 0.4081,
      "step": 462
    },
    {
      "epoch": 16.0,
      "grad_norm": 3.367891311645508,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.5498,
      "step": 464
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6287878787878788,
      "eval_f1_macro": 0.5372783597671154,
      "eval_f1_micro": 0.6287878787878788,
      "eval_f1_weighted": 0.6117070792278643,
      "eval_loss": 1.1293154954910278,
      "eval_precision_macro": 0.5353323147440795,
      "eval_precision_micro": 0.6287878787878788,
      "eval_precision_weighted": 0.603853373305245,
      "eval_recall_macro": 0.5471201814058957,
      "eval_recall_micro": 0.6287878787878788,
      "eval_recall_weighted": 0.6287878787878788,
      "eval_runtime": 2.3165,
      "eval_samples_per_second": 56.983,
      "eval_steps_per_second": 7.339,
      "step": 464
    },
    {
      "epoch": 16.06896551724138,
      "grad_norm": 3.580066442489624,
      "learning_rate": 2.183908045977012e-05,
      "loss": 0.4746,
      "step": 466
    },
    {
      "epoch": 16.137931034482758,
      "grad_norm": 3.583418130874634,
      "learning_rate": 2.145593869731801e-05,
      "loss": 0.4795,
      "step": 468
    },
    {
      "epoch": 16.20689655172414,
      "grad_norm": 2.073596239089966,
      "learning_rate": 2.10727969348659e-05,
      "loss": 0.3328,
      "step": 470
    },
    {
      "epoch": 16.275862068965516,
      "grad_norm": 3.2745907306671143,
      "learning_rate": 2.0689655172413793e-05,
      "loss": 0.4395,
      "step": 472
    },
    {
      "epoch": 16.344827586206897,
      "grad_norm": 1.9064489603042603,
      "learning_rate": 2.0306513409961685e-05,
      "loss": 0.3172,
      "step": 474
    },
    {
      "epoch": 16.413793103448278,
      "grad_norm": 4.838860988616943,
      "learning_rate": 1.992337164750958e-05,
      "loss": 0.6758,
      "step": 476
    },
    {
      "epoch": 16.482758620689655,
      "grad_norm": 1.598775029182434,
      "learning_rate": 1.9540229885057475e-05,
      "loss": 0.3767,
      "step": 478
    },
    {
      "epoch": 16.551724137931036,
      "grad_norm": 4.009886264801025,
      "learning_rate": 1.9157088122605367e-05,
      "loss": 0.4511,
      "step": 480
    },
    {
      "epoch": 16.620689655172413,
      "grad_norm": 3.5439836978912354,
      "learning_rate": 1.8773946360153258e-05,
      "loss": 0.4542,
      "step": 482
    },
    {
      "epoch": 16.689655172413794,
      "grad_norm": 3.5532007217407227,
      "learning_rate": 1.839080459770115e-05,
      "loss": 0.5007,
      "step": 484
    },
    {
      "epoch": 16.75862068965517,
      "grad_norm": 2.5344693660736084,
      "learning_rate": 1.800766283524904e-05,
      "loss": 0.3181,
      "step": 486
    },
    {
      "epoch": 16.82758620689655,
      "grad_norm": 1.9428218603134155,
      "learning_rate": 1.7624521072796933e-05,
      "loss": 0.4047,
      "step": 488
    },
    {
      "epoch": 16.896551724137932,
      "grad_norm": 2.117885112762451,
      "learning_rate": 1.7241379310344828e-05,
      "loss": 0.3487,
      "step": 490
    },
    {
      "epoch": 16.96551724137931,
      "grad_norm": 3.377067804336548,
      "learning_rate": 1.6858237547892723e-05,
      "loss": 0.5037,
      "step": 492
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.6212121212121212,
      "eval_f1_macro": 0.5290491049649699,
      "eval_f1_micro": 0.6212121212121212,
      "eval_f1_weighted": 0.6005376615693997,
      "eval_loss": 1.163466453552246,
      "eval_precision_macro": 0.5374433106575963,
      "eval_precision_micro": 0.6212121212121212,
      "eval_precision_weighted": 0.6021569865319866,
      "eval_recall_macro": 0.5397581254724112,
      "eval_recall_micro": 0.6212121212121212,
      "eval_recall_weighted": 0.6212121212121212,
      "eval_runtime": 2.3304,
      "eval_samples_per_second": 56.642,
      "eval_steps_per_second": 7.295,
      "step": 493
    },
    {
      "epoch": 17.03448275862069,
      "grad_norm": 3.5657002925872803,
      "learning_rate": 1.6475095785440615e-05,
      "loss": 0.4446,
      "step": 494
    },
    {
      "epoch": 17.103448275862068,
      "grad_norm": 4.189113616943359,
      "learning_rate": 1.6091954022988507e-05,
      "loss": 0.3923,
      "step": 496
    },
    {
      "epoch": 17.17241379310345,
      "grad_norm": 2.554723024368286,
      "learning_rate": 1.5708812260536398e-05,
      "loss": 0.358,
      "step": 498
    },
    {
      "epoch": 17.24137931034483,
      "grad_norm": 3.3316357135772705,
      "learning_rate": 1.532567049808429e-05,
      "loss": 0.2683,
      "step": 500
    },
    {
      "epoch": 17.310344827586206,
      "grad_norm": 3.2318475246429443,
      "learning_rate": 1.4942528735632185e-05,
      "loss": 0.3611,
      "step": 502
    },
    {
      "epoch": 17.379310344827587,
      "grad_norm": 3.2837979793548584,
      "learning_rate": 1.4559386973180078e-05,
      "loss": 0.435,
      "step": 504
    },
    {
      "epoch": 17.448275862068964,
      "grad_norm": 3.8966495990753174,
      "learning_rate": 1.417624521072797e-05,
      "loss": 0.2982,
      "step": 506
    },
    {
      "epoch": 17.517241379310345,
      "grad_norm": 6.666224479675293,
      "learning_rate": 1.3793103448275863e-05,
      "loss": 0.4624,
      "step": 508
    },
    {
      "epoch": 17.586206896551722,
      "grad_norm": 4.43505334854126,
      "learning_rate": 1.3409961685823755e-05,
      "loss": 0.6202,
      "step": 510
    },
    {
      "epoch": 17.655172413793103,
      "grad_norm": 3.7163617610931396,
      "learning_rate": 1.3026819923371647e-05,
      "loss": 0.3422,
      "step": 512
    },
    {
      "epoch": 17.724137931034484,
      "grad_norm": 2.855036735534668,
      "learning_rate": 1.2643678160919542e-05,
      "loss": 0.5171,
      "step": 514
    },
    {
      "epoch": 17.79310344827586,
      "grad_norm": 3.121839761734009,
      "learning_rate": 1.2260536398467433e-05,
      "loss": 0.3575,
      "step": 516
    },
    {
      "epoch": 17.862068965517242,
      "grad_norm": 2.77312970161438,
      "learning_rate": 1.1877394636015327e-05,
      "loss": 0.3051,
      "step": 518
    },
    {
      "epoch": 17.93103448275862,
      "grad_norm": 2.2301366329193115,
      "learning_rate": 1.1494252873563218e-05,
      "loss": 0.4635,
      "step": 520
    },
    {
      "epoch": 18.0,
      "grad_norm": 4.1137518882751465,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.3624,
      "step": 522
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6590909090909091,
      "eval_f1_macro": 0.5699944413104846,
      "eval_f1_micro": 0.6590909090909091,
      "eval_f1_weighted": 0.6414064024247821,
      "eval_loss": 1.09944748878479,
      "eval_precision_macro": 0.5815085815085814,
      "eval_precision_micro": 0.6590909090909091,
      "eval_precision_weighted": 0.640854299945209,
      "eval_recall_macro": 0.5742857142857142,
      "eval_recall_micro": 0.6590909090909091,
      "eval_recall_weighted": 0.6590909090909091,
      "eval_runtime": 2.3246,
      "eval_samples_per_second": 56.785,
      "eval_steps_per_second": 7.313,
      "step": 522
    },
    {
      "epoch": 18.06896551724138,
      "grad_norm": 5.545546054840088,
      "learning_rate": 1.0727969348659005e-05,
      "loss": 0.4351,
      "step": 524
    },
    {
      "epoch": 18.137931034482758,
      "grad_norm": 2.5802762508392334,
      "learning_rate": 1.0344827586206897e-05,
      "loss": 0.3958,
      "step": 526
    },
    {
      "epoch": 18.20689655172414,
      "grad_norm": 3.738755702972412,
      "learning_rate": 9.96168582375479e-06,
      "loss": 0.3933,
      "step": 528
    },
    {
      "epoch": 18.275862068965516,
      "grad_norm": 2.6017744541168213,
      "learning_rate": 9.578544061302683e-06,
      "loss": 0.4367,
      "step": 530
    },
    {
      "epoch": 18.344827586206897,
      "grad_norm": 4.551992416381836,
      "learning_rate": 9.195402298850575e-06,
      "loss": 0.3844,
      "step": 532
    },
    {
      "epoch": 18.413793103448278,
      "grad_norm": 3.3013875484466553,
      "learning_rate": 8.812260536398467e-06,
      "loss": 0.3592,
      "step": 534
    },
    {
      "epoch": 18.482758620689655,
      "grad_norm": 8.250885009765625,
      "learning_rate": 8.429118773946362e-06,
      "loss": 0.5747,
      "step": 536
    },
    {
      "epoch": 18.551724137931036,
      "grad_norm": 2.1234164237976074,
      "learning_rate": 8.045977011494253e-06,
      "loss": 0.3561,
      "step": 538
    },
    {
      "epoch": 18.620689655172413,
      "grad_norm": 2.730742931365967,
      "learning_rate": 7.662835249042145e-06,
      "loss": 0.3182,
      "step": 540
    },
    {
      "epoch": 18.689655172413794,
      "grad_norm": 1.1862895488739014,
      "learning_rate": 7.279693486590039e-06,
      "loss": 0.2739,
      "step": 542
    },
    {
      "epoch": 18.75862068965517,
      "grad_norm": 3.530163049697876,
      "learning_rate": 6.896551724137932e-06,
      "loss": 0.4006,
      "step": 544
    },
    {
      "epoch": 18.82758620689655,
      "grad_norm": 3.7491860389709473,
      "learning_rate": 6.513409961685823e-06,
      "loss": 0.2657,
      "step": 546
    },
    {
      "epoch": 18.896551724137932,
      "grad_norm": 6.904880523681641,
      "learning_rate": 6.130268199233717e-06,
      "loss": 0.4191,
      "step": 548
    },
    {
      "epoch": 18.96551724137931,
      "grad_norm": 1.2763559818267822,
      "learning_rate": 5.747126436781609e-06,
      "loss": 0.3387,
      "step": 550
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6515151515151515,
      "eval_f1_macro": 0.5642594468235355,
      "eval_f1_micro": 0.6515151515151515,
      "eval_f1_weighted": 0.6367337845877334,
      "eval_loss": 1.094382643699646,
      "eval_precision_macro": 0.5555528431468281,
      "eval_precision_micro": 0.6515151515151515,
      "eval_precision_weighted": 0.6268442283394436,
      "eval_recall_macro": 0.5781405895691609,
      "eval_recall_micro": 0.6515151515151515,
      "eval_recall_weighted": 0.6515151515151515,
      "eval_runtime": 2.3159,
      "eval_samples_per_second": 56.998,
      "eval_steps_per_second": 7.341,
      "step": 551
    },
    {
      "epoch": 19.03448275862069,
      "grad_norm": 3.0644407272338867,
      "learning_rate": 5.3639846743295025e-06,
      "loss": 0.3165,
      "step": 552
    },
    {
      "epoch": 19.103448275862068,
      "grad_norm": 3.8234410285949707,
      "learning_rate": 4.980842911877395e-06,
      "loss": 0.4086,
      "step": 554
    },
    {
      "epoch": 19.17241379310345,
      "grad_norm": 1.5022215843200684,
      "learning_rate": 4.5977011494252875e-06,
      "loss": 0.2806,
      "step": 556
    },
    {
      "epoch": 19.24137931034483,
      "grad_norm": 4.573976039886475,
      "learning_rate": 4.214559386973181e-06,
      "loss": 0.4335,
      "step": 558
    },
    {
      "epoch": 19.310344827586206,
      "grad_norm": 3.7575619220733643,
      "learning_rate": 3.8314176245210725e-06,
      "loss": 0.2748,
      "step": 560
    },
    {
      "epoch": 19.379310344827587,
      "grad_norm": 1.8237051963806152,
      "learning_rate": 3.448275862068966e-06,
      "loss": 0.3879,
      "step": 562
    },
    {
      "epoch": 19.448275862068964,
      "grad_norm": 5.58415412902832,
      "learning_rate": 3.0651340996168583e-06,
      "loss": 0.2736,
      "step": 564
    },
    {
      "epoch": 19.517241379310345,
      "grad_norm": 1.1507081985473633,
      "learning_rate": 2.6819923371647512e-06,
      "loss": 0.2612,
      "step": 566
    },
    {
      "epoch": 19.586206896551722,
      "grad_norm": 1.060652494430542,
      "learning_rate": 2.2988505747126437e-06,
      "loss": 0.2706,
      "step": 568
    },
    {
      "epoch": 19.655172413793103,
      "grad_norm": 1.4286646842956543,
      "learning_rate": 1.9157088122605362e-06,
      "loss": 0.2616,
      "step": 570
    },
    {
      "epoch": 19.724137931034484,
      "grad_norm": 5.675417900085449,
      "learning_rate": 1.5325670498084292e-06,
      "loss": 0.4929,
      "step": 572
    },
    {
      "epoch": 19.79310344827586,
      "grad_norm": 1.839812994003296,
      "learning_rate": 1.1494252873563219e-06,
      "loss": 0.2418,
      "step": 574
    },
    {
      "epoch": 19.862068965517242,
      "grad_norm": 3.5628745555877686,
      "learning_rate": 7.662835249042146e-07,
      "loss": 0.4503,
      "step": 576
    },
    {
      "epoch": 19.93103448275862,
      "grad_norm": 2.046889543533325,
      "learning_rate": 3.831417624521073e-07,
      "loss": 0.2689,
      "step": 578
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.0142828226089478,
      "learning_rate": 0.0,
      "loss": 0.4052,
      "step": 580
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6590909090909091,
      "eval_f1_macro": 0.5683267149224223,
      "eval_f1_micro": 0.6590909090909091,
      "eval_f1_weighted": 0.6432058978241244,
      "eval_loss": 1.0934416055679321,
      "eval_precision_macro": 0.5681032097512798,
      "eval_precision_micro": 0.6590909090909091,
      "eval_precision_weighted": 0.6393054520079716,
      "eval_recall_macro": 0.5797581254724111,
      "eval_recall_micro": 0.6590909090909091,
      "eval_recall_weighted": 0.6590909090909091,
      "eval_runtime": 2.3382,
      "eval_samples_per_second": 56.454,
      "eval_steps_per_second": 7.271,
      "step": 580
    },
    {
      "epoch": 20.0,
      "step": 580,
      "total_flos": 7.160580716109005e+17,
      "train_loss": 1.0144407548781098,
      "train_runtime": 564.1231,
      "train_samples_per_second": 16.379,
      "train_steps_per_second": 1.028
    }
  ],
  "logging_steps": 2,
  "max_steps": 580,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.160580716109005e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}