{
  "best_metric": 0.5942014872446945,
  "best_model_checkpoint": "models/fallacy_detector/checkpoint-2480",
  "epoch": 13.0,
  "eval_steps": 500,
  "global_step": 3224,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08064516129032258,
      "grad_norm": 116749.5,
      "learning_rate": 3.5842293906810035e-06,
      "loss": 1.8041,
      "step": 20
    },
    {
      "epoch": 0.16129032258064516,
      "grad_norm": 129446.0,
      "learning_rate": 7.168458781362007e-06,
      "loss": 1.8001,
      "step": 40
    },
    {
      "epoch": 0.24193548387096775,
      "grad_norm": 124799.328125,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 1.7913,
      "step": 60
    },
    {
      "epoch": 0.3225806451612903,
      "grad_norm": 120556.65625,
      "learning_rate": 1.4336917562724014e-05,
      "loss": 1.7964,
      "step": 80
    },
    {
      "epoch": 0.4032258064516129,
      "grad_norm": 147319.21875,
      "learning_rate": 1.7921146953405018e-05,
      "loss": 1.7976,
      "step": 100
    },
    {
      "epoch": 0.4838709677419355,
      "grad_norm": 187166.375,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 1.8002,
      "step": 120
    },
    {
      "epoch": 0.5645161290322581,
      "grad_norm": 128844.7421875,
      "learning_rate": 2.5089605734767026e-05,
      "loss": 1.8047,
      "step": 140
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 129834.9765625,
      "learning_rate": 2.8673835125448028e-05,
      "loss": 1.7942,
      "step": 160
    },
    {
      "epoch": 0.7258064516129032,
      "grad_norm": 103151.375,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 1.79,
      "step": 180
    },
    {
      "epoch": 0.8064516129032258,
      "grad_norm": 159776.875,
      "learning_rate": 3.5842293906810036e-05,
      "loss": 1.7913,
      "step": 200
    },
    {
      "epoch": 0.8870967741935484,
      "grad_norm": 136072.203125,
      "learning_rate": 3.9426523297491045e-05,
      "loss": 1.7903,
      "step": 220
    },
    {
      "epoch": 0.967741935483871,
      "grad_norm": 153388.328125,
      "learning_rate": 4.301075268817205e-05,
      "loss": 1.7666,
      "step": 240
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3097165991902834,
      "eval_f1_macro": 0.2437130087374885,
      "eval_f1_weighted": 0.27846991634919227,
      "eval_loss": 1.7279411554336548,
      "eval_runtime": 6.7607,
      "eval_samples_per_second": 73.069,
      "eval_steps_per_second": 2.367,
      "step": 248
    },
    {
      "epoch": 1.0483870967741935,
      "grad_norm": 150245.78125,
      "learning_rate": 4.659498207885305e-05,
      "loss": 1.7333,
      "step": 260
    },
    {
      "epoch": 1.129032258064516,
      "grad_norm": 145167.953125,
      "learning_rate": 5.017921146953405e-05,
      "loss": 1.6906,
      "step": 280
    },
    {
      "epoch": 1.2096774193548387,
      "grad_norm": 189463.4375,
      "learning_rate": 5.3763440860215054e-05,
      "loss": 1.621,
      "step": 300
    },
    {
      "epoch": 1.2903225806451613,
      "grad_norm": 231174.21875,
      "learning_rate": 5.7347670250896056e-05,
      "loss": 1.5355,
      "step": 320
    },
    {
      "epoch": 1.370967741935484,
      "grad_norm": 216882.625,
      "learning_rate": 6.093189964157706e-05,
      "loss": 1.4977,
      "step": 340
    },
    {
      "epoch": 1.4516129032258065,
      "grad_norm": 284663.46875,
      "learning_rate": 6.451612903225807e-05,
      "loss": 1.5157,
      "step": 360
    },
    {
      "epoch": 1.532258064516129,
      "grad_norm": 373533.09375,
      "learning_rate": 6.810035842293908e-05,
      "loss": 1.4792,
      "step": 380
    },
    {
      "epoch": 1.6129032258064515,
      "grad_norm": 229778.140625,
      "learning_rate": 7.168458781362007e-05,
      "loss": 1.5003,
      "step": 400
    },
    {
      "epoch": 1.6935483870967742,
      "grad_norm": 206755.359375,
      "learning_rate": 7.526881720430108e-05,
      "loss": 1.4892,
      "step": 420
    },
    {
      "epoch": 1.7741935483870968,
      "grad_norm": 389183.5,
      "learning_rate": 7.885304659498209e-05,
      "loss": 1.5018,
      "step": 440
    },
    {
      "epoch": 1.8548387096774195,
      "grad_norm": 211097.21875,
      "learning_rate": 8.243727598566309e-05,
      "loss": 1.4836,
      "step": 460
    },
    {
      "epoch": 1.935483870967742,
      "grad_norm": 185060.09375,
      "learning_rate": 8.60215053763441e-05,
      "loss": 1.4278,
      "step": 480
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.39473684210526316,
      "eval_f1_macro": 0.3803420101129123,
      "eval_f1_weighted": 0.3678672758556732,
      "eval_loss": 1.453121542930603,
      "eval_runtime": 6.4873,
      "eval_samples_per_second": 76.149,
      "eval_steps_per_second": 2.466,
      "step": 496
    },
    {
      "epoch": 2.0161290322580645,
      "grad_norm": 217972.03125,
      "learning_rate": 8.960573476702509e-05,
      "loss": 1.4423,
      "step": 500
    },
    {
      "epoch": 2.096774193548387,
      "grad_norm": 202365.875,
      "learning_rate": 9.31899641577061e-05,
      "loss": 1.3673,
      "step": 520
    },
    {
      "epoch": 2.1774193548387095,
      "grad_norm": 257111.984375,
      "learning_rate": 9.677419354838711e-05,
      "loss": 1.3972,
      "step": 540
    },
    {
      "epoch": 2.258064516129032,
      "grad_norm": 179046.5625,
      "learning_rate": 9.993674889310564e-05,
      "loss": 1.3573,
      "step": 560
    },
    {
      "epoch": 2.338709677419355,
      "grad_norm": 247477.953125,
      "learning_rate": 9.930423782416193e-05,
      "loss": 1.3868,
      "step": 580
    },
    {
      "epoch": 2.4193548387096775,
      "grad_norm": 364917.625,
      "learning_rate": 9.867172675521823e-05,
      "loss": 1.2169,
      "step": 600
    },
    {
      "epoch": 2.5,
      "grad_norm": 232415.90625,
      "learning_rate": 9.80392156862745e-05,
      "loss": 1.3737,
      "step": 620
    },
    {
      "epoch": 2.5806451612903225,
      "grad_norm": 302076.875,
      "learning_rate": 9.740670461733081e-05,
      "loss": 1.2454,
      "step": 640
    },
    {
      "epoch": 2.661290322580645,
      "grad_norm": 327026.0625,
      "learning_rate": 9.677419354838711e-05,
      "loss": 1.4311,
      "step": 660
    },
    {
      "epoch": 2.741935483870968,
      "grad_norm": 183140.640625,
      "learning_rate": 9.61416824794434e-05,
      "loss": 1.3207,
      "step": 680
    },
    {
      "epoch": 2.8225806451612905,
      "grad_norm": 238072.3125,
      "learning_rate": 9.550917141049968e-05,
      "loss": 1.3202,
      "step": 700
    },
    {
      "epoch": 2.903225806451613,
      "grad_norm": 215563.453125,
      "learning_rate": 9.487666034155598e-05,
      "loss": 1.3908,
      "step": 720
    },
    {
      "epoch": 2.9838709677419355,
      "grad_norm": 424141.4375,
      "learning_rate": 9.424414927261227e-05,
      "loss": 1.1402,
      "step": 740
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.46558704453441296,
      "eval_f1_macro": 0.44600226194430165,
      "eval_f1_weighted": 0.4682600612720702,
      "eval_loss": 1.2723522186279297,
      "eval_runtime": 6.4886,
      "eval_samples_per_second": 76.133,
      "eval_steps_per_second": 2.466,
      "step": 744
    },
    {
      "epoch": 3.064516129032258,
      "grad_norm": 238018.875,
      "learning_rate": 9.361163820366858e-05,
      "loss": 1.1561,
      "step": 760
    },
    {
      "epoch": 3.1451612903225805,
      "grad_norm": 282444.9375,
      "learning_rate": 9.297912713472486e-05,
      "loss": 1.2526,
      "step": 780
    },
    {
      "epoch": 3.225806451612903,
      "grad_norm": 132935.109375,
      "learning_rate": 9.234661606578115e-05,
      "loss": 1.0902,
      "step": 800
    },
    {
      "epoch": 3.306451612903226,
      "grad_norm": 173232.328125,
      "learning_rate": 9.171410499683745e-05,
      "loss": 1.2506,
      "step": 820
    },
    {
      "epoch": 3.3870967741935485,
      "grad_norm": 269722.375,
      "learning_rate": 9.108159392789374e-05,
      "loss": 1.185,
      "step": 840
    },
    {
      "epoch": 3.467741935483871,
      "grad_norm": 163013.53125,
      "learning_rate": 9.044908285895004e-05,
      "loss": 1.1408,
      "step": 860
    },
    {
      "epoch": 3.5483870967741935,
      "grad_norm": 172902.015625,
      "learning_rate": 8.981657179000633e-05,
      "loss": 1.2523,
      "step": 880
    },
    {
      "epoch": 3.629032258064516,
      "grad_norm": 315512.5625,
      "learning_rate": 8.918406072106262e-05,
      "loss": 1.1325,
      "step": 900
    },
    {
      "epoch": 3.709677419354839,
      "grad_norm": 287570.34375,
      "learning_rate": 8.855154965211892e-05,
      "loss": 1.1493,
      "step": 920
    },
    {
      "epoch": 3.790322580645161,
      "grad_norm": 317774.9375,
      "learning_rate": 8.791903858317521e-05,
      "loss": 1.133,
      "step": 940
    },
    {
      "epoch": 3.870967741935484,
      "grad_norm": 307236.34375,
      "learning_rate": 8.72865275142315e-05,
      "loss": 1.2473,
      "step": 960
    },
    {
      "epoch": 3.9516129032258065,
      "grad_norm": 280431.59375,
      "learning_rate": 8.66540164452878e-05,
      "loss": 1.0287,
      "step": 980
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5141700404858299,
      "eval_f1_macro": 0.4818941736895477,
      "eval_f1_weighted": 0.4929543703966077,
      "eval_loss": 1.2111815214157104,
      "eval_runtime": 6.4861,
      "eval_samples_per_second": 76.162,
      "eval_steps_per_second": 2.467,
      "step": 992
    },
    {
      "epoch": 4.032258064516129,
      "grad_norm": 416063.03125,
      "learning_rate": 8.60215053763441e-05,
      "loss": 1.0325,
      "step": 1000
    },
    {
      "epoch": 4.112903225806452,
      "grad_norm": 365579.875,
      "learning_rate": 8.538899430740039e-05,
      "loss": 1.0955,
      "step": 1020
    },
    {
      "epoch": 4.193548387096774,
      "grad_norm": 298404.71875,
      "learning_rate": 8.475648323845667e-05,
      "loss": 1.0768,
      "step": 1040
    },
    {
      "epoch": 4.274193548387097,
      "grad_norm": 259632.609375,
      "learning_rate": 8.412397216951296e-05,
      "loss": 1.0736,
      "step": 1060
    },
    {
      "epoch": 4.354838709677419,
      "grad_norm": 314956.3125,
      "learning_rate": 8.349146110056927e-05,
      "loss": 0.9955,
      "step": 1080
    },
    {
      "epoch": 4.435483870967742,
      "grad_norm": 755572.6875,
      "learning_rate": 8.285895003162556e-05,
      "loss": 0.9618,
      "step": 1100
    },
    {
      "epoch": 4.516129032258064,
      "grad_norm": 247183.71875,
      "learning_rate": 8.222643896268185e-05,
      "loss": 1.013,
      "step": 1120
    },
    {
      "epoch": 4.596774193548387,
      "grad_norm": 301904.8125,
      "learning_rate": 8.159392789373814e-05,
      "loss": 1.0046,
      "step": 1140
    },
    {
      "epoch": 4.67741935483871,
      "grad_norm": 224825.875,
      "learning_rate": 8.096141682479443e-05,
      "loss": 1.1598,
      "step": 1160
    },
    {
      "epoch": 4.758064516129032,
      "grad_norm": 280016.9375,
      "learning_rate": 8.032890575585074e-05,
      "loss": 0.9723,
      "step": 1180
    },
    {
      "epoch": 4.838709677419355,
      "grad_norm": 294691.1875,
      "learning_rate": 7.969639468690702e-05,
      "loss": 1.0387,
      "step": 1200
    },
    {
      "epoch": 4.919354838709677,
      "grad_norm": 157460.875,
      "learning_rate": 7.906388361796332e-05,
      "loss": 0.9852,
      "step": 1220
    },
    {
      "epoch": 5.0,
      "grad_norm": 562566.0625,
      "learning_rate": 7.843137254901961e-05,
      "loss": 0.993,
      "step": 1240
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5344129554655871,
      "eval_f1_macro": 0.5210663276171126,
      "eval_f1_weighted": 0.5358881669989737,
      "eval_loss": 1.1454784870147705,
      "eval_runtime": 6.4764,
      "eval_samples_per_second": 76.277,
      "eval_steps_per_second": 2.471,
      "step": 1240
    },
    {
      "epoch": 5.080645161290323,
      "grad_norm": 257956.03125,
      "learning_rate": 7.77988614800759e-05,
      "loss": 0.9462,
      "step": 1260
    },
    {
      "epoch": 5.161290322580645,
      "grad_norm": 241475.75,
      "learning_rate": 7.71663504111322e-05,
      "loss": 0.936,
      "step": 1280
    },
    {
      "epoch": 5.241935483870968,
      "grad_norm": 222372.890625,
      "learning_rate": 7.653383934218849e-05,
      "loss": 0.9063,
      "step": 1300
    },
    {
      "epoch": 5.32258064516129,
      "grad_norm": 272774.5625,
      "learning_rate": 7.590132827324479e-05,
      "loss": 0.8876,
      "step": 1320
    },
    {
      "epoch": 5.403225806451613,
      "grad_norm": 366996.75,
      "learning_rate": 7.526881720430108e-05,
      "loss": 0.8755,
      "step": 1340
    },
    {
      "epoch": 5.483870967741936,
      "grad_norm": 366414.875,
      "learning_rate": 7.463630613535737e-05,
      "loss": 0.9071,
      "step": 1360
    },
    {
      "epoch": 5.564516129032258,
      "grad_norm": 202618.3125,
      "learning_rate": 7.400379506641366e-05,
      "loss": 0.8593,
      "step": 1380
    },
    {
      "epoch": 5.645161290322581,
      "grad_norm": 228361.6875,
      "learning_rate": 7.337128399746996e-05,
      "loss": 0.8858,
      "step": 1400
    },
    {
      "epoch": 5.725806451612903,
      "grad_norm": 228972.40625,
      "learning_rate": 7.273877292852626e-05,
      "loss": 0.9162,
      "step": 1420
    },
    {
      "epoch": 5.806451612903226,
      "grad_norm": 282326.96875,
      "learning_rate": 7.210626185958255e-05,
      "loss": 0.9823,
      "step": 1440
    },
    {
      "epoch": 5.887096774193548,
      "grad_norm": 349532.15625,
      "learning_rate": 7.147375079063883e-05,
      "loss": 0.9268,
      "step": 1460
    },
    {
      "epoch": 5.967741935483871,
      "grad_norm": 344699.09375,
      "learning_rate": 7.084123972169513e-05,
      "loss": 0.9212,
      "step": 1480
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5647773279352226,
      "eval_f1_macro": 0.5566153646272254,
      "eval_f1_weighted": 0.5635035195555513,
      "eval_loss": 1.1447995901107788,
      "eval_runtime": 6.7478,
      "eval_samples_per_second": 73.209,
      "eval_steps_per_second": 2.371,
      "step": 1488
    },
    {
      "epoch": 6.048387096774194,
      "grad_norm": 380314.4375,
      "learning_rate": 7.020872865275142e-05,
      "loss": 0.8163,
      "step": 1500
    },
    {
      "epoch": 6.129032258064516,
      "grad_norm": 297659.90625,
      "learning_rate": 6.957621758380773e-05,
      "loss": 0.9182,
      "step": 1520
    },
    {
      "epoch": 6.209677419354839,
      "grad_norm": 336392.09375,
      "learning_rate": 6.894370651486401e-05,
      "loss": 0.7823,
      "step": 1540
    },
    {
      "epoch": 6.290322580645161,
      "grad_norm": 244166.359375,
      "learning_rate": 6.83111954459203e-05,
      "loss": 0.8199,
      "step": 1560
    },
    {
      "epoch": 6.370967741935484,
      "grad_norm": 306139.09375,
      "learning_rate": 6.76786843769766e-05,
      "loss": 0.8065,
      "step": 1580
    },
    {
      "epoch": 6.451612903225806,
      "grad_norm": 207655.90625,
      "learning_rate": 6.704617330803289e-05,
      "loss": 0.7914,
      "step": 1600
    },
    {
      "epoch": 6.532258064516129,
      "grad_norm": 240435.453125,
      "learning_rate": 6.64136622390892e-05,
      "loss": 0.7684,
      "step": 1620
    },
    {
      "epoch": 6.612903225806452,
      "grad_norm": 421195.90625,
      "learning_rate": 6.578115117014548e-05,
      "loss": 0.7626,
      "step": 1640
    },
    {
      "epoch": 6.693548387096774,
      "grad_norm": 308932.90625,
      "learning_rate": 6.514864010120177e-05,
      "loss": 0.9373,
      "step": 1660
    },
    {
      "epoch": 6.774193548387097,
      "grad_norm": 295667.90625,
      "learning_rate": 6.451612903225807e-05,
      "loss": 0.7873,
      "step": 1680
    },
    {
      "epoch": 6.854838709677419,
      "grad_norm": 411704.5,
      "learning_rate": 6.388361796331436e-05,
      "loss": 0.7532,
      "step": 1700
    },
    {
      "epoch": 6.935483870967742,
      "grad_norm": 171699.0625,
      "learning_rate": 6.325110689437066e-05,
      "loss": 0.735,
      "step": 1720
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5951417004048583,
      "eval_f1_macro": 0.5908663884882793,
      "eval_f1_weighted": 0.6001450425865201,
      "eval_loss": 1.1485521793365479,
      "eval_runtime": 6.7324,
      "eval_samples_per_second": 73.376,
      "eval_steps_per_second": 2.377,
      "step": 1736
    },
    {
      "epoch": 7.016129032258065,
      "grad_norm": 320877.96875,
      "learning_rate": 6.261859582542695e-05,
      "loss": 0.7661,
      "step": 1740
    },
    {
      "epoch": 7.096774193548387,
      "grad_norm": 219053.296875,
      "learning_rate": 6.198608475648324e-05,
      "loss": 0.6837,
      "step": 1760
    },
    {
      "epoch": 7.17741935483871,
      "grad_norm": 265047.4375,
      "learning_rate": 6.135357368753954e-05,
      "loss": 0.8169,
      "step": 1780
    },
    {
      "epoch": 7.258064516129032,
      "grad_norm": 254569.140625,
      "learning_rate": 6.0721062618595825e-05,
      "loss": 0.7038,
      "step": 1800
    },
    {
      "epoch": 7.338709677419355,
      "grad_norm": 305000.25,
      "learning_rate": 6.008855154965212e-05,
      "loss": 0.6717,
      "step": 1820
    },
    {
      "epoch": 7.419354838709677,
      "grad_norm": 379865.53125,
      "learning_rate": 5.945604048070842e-05,
      "loss": 0.7303,
      "step": 1840
    },
    {
      "epoch": 7.5,
      "grad_norm": 178170.015625,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.6484,
      "step": 1860
    },
    {
      "epoch": 7.580645161290323,
      "grad_norm": 430648.0,
      "learning_rate": 5.8191018342821e-05,
      "loss": 0.8021,
      "step": 1880
    },
    {
      "epoch": 7.661290322580645,
      "grad_norm": 297593.0625,
      "learning_rate": 5.7558507273877296e-05,
      "loss": 0.7074,
      "step": 1900
    },
    {
      "epoch": 7.741935483870968,
      "grad_norm": 338920.84375,
      "learning_rate": 5.692599620493358e-05,
      "loss": 0.6387,
      "step": 1920
    },
    {
      "epoch": 7.82258064516129,
      "grad_norm": 352287.0,
      "learning_rate": 5.6293485135989884e-05,
      "loss": 0.7579,
      "step": 1940
    },
    {
      "epoch": 7.903225806451613,
      "grad_norm": 208302.796875,
      "learning_rate": 5.566097406704618e-05,
      "loss": 0.7028,
      "step": 1960
    },
    {
      "epoch": 7.983870967741936,
      "grad_norm": 198196.296875,
      "learning_rate": 5.502846299810247e-05,
      "loss": 0.7811,
      "step": 1980
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5668016194331984,
      "eval_f1_macro": 0.5733215653407325,
      "eval_f1_weighted": 0.5705200897887893,
      "eval_loss": 1.165218472480774,
      "eval_runtime": 6.4814,
      "eval_samples_per_second": 76.218,
      "eval_steps_per_second": 2.469,
      "step": 1984
    },
    {
      "epoch": 8.064516129032258,
      "grad_norm": 146523.9375,
      "learning_rate": 5.439595192915876e-05,
      "loss": 0.6839,
      "step": 2000
    },
    {
      "epoch": 8.14516129032258,
      "grad_norm": 248096.859375,
      "learning_rate": 5.3763440860215054e-05,
      "loss": 0.6205,
      "step": 2020
    },
    {
      "epoch": 8.225806451612904,
      "grad_norm": 355304.8125,
      "learning_rate": 5.313092979127134e-05,
      "loss": 0.6835,
      "step": 2040
    },
    {
      "epoch": 8.306451612903226,
      "grad_norm": 559134.6875,
      "learning_rate": 5.249841872232765e-05,
      "loss": 0.691,
      "step": 2060
    },
    {
      "epoch": 8.387096774193548,
      "grad_norm": 313194.6875,
      "learning_rate": 5.1865907653383936e-05,
      "loss": 0.575,
      "step": 2080
    },
    {
      "epoch": 8.46774193548387,
      "grad_norm": 356380.125,
      "learning_rate": 5.123339658444023e-05,
      "loss": 0.6626,
      "step": 2100
    },
    {
      "epoch": 8.548387096774194,
      "grad_norm": 362203.6875,
      "learning_rate": 5.0600885515496524e-05,
      "loss": 0.6883,
      "step": 2120
    },
    {
      "epoch": 8.629032258064516,
      "grad_norm": 178366.890625,
      "learning_rate": 4.996837444655282e-05,
      "loss": 0.752,
      "step": 2140
    },
    {
      "epoch": 8.709677419354838,
      "grad_norm": 170288.078125,
      "learning_rate": 4.933586337760911e-05,
      "loss": 0.7214,
      "step": 2160
    },
    {
      "epoch": 8.790322580645162,
      "grad_norm": 210902.640625,
      "learning_rate": 4.870335230866541e-05,
      "loss": 0.6366,
      "step": 2180
    },
    {
      "epoch": 8.870967741935484,
      "grad_norm": 260362.5625,
      "learning_rate": 4.80708412397217e-05,
      "loss": 0.6006,
      "step": 2200
    },
    {
      "epoch": 8.951612903225806,
      "grad_norm": 328288.71875,
      "learning_rate": 4.743833017077799e-05,
      "loss": 0.6694,
      "step": 2220
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5789473684210527,
      "eval_f1_macro": 0.5848147378665898,
      "eval_f1_weighted": 0.5817724273381693,
      "eval_loss": 1.1652216911315918,
      "eval_runtime": 6.7288,
      "eval_samples_per_second": 73.416,
      "eval_steps_per_second": 2.378,
      "step": 2232
    },
    {
      "epoch": 9.03225806451613,
      "grad_norm": 390067.4375,
      "learning_rate": 4.680581910183429e-05,
      "loss": 0.505,
      "step": 2240
    },
    {
      "epoch": 9.112903225806452,
      "grad_norm": 412543.65625,
      "learning_rate": 4.6173308032890576e-05,
      "loss": 0.5729,
      "step": 2260
    },
    {
      "epoch": 9.193548387096774,
      "grad_norm": 116351.8046875,
      "learning_rate": 4.554079696394687e-05,
      "loss": 0.6273,
      "step": 2280
    },
    {
      "epoch": 9.274193548387096,
      "grad_norm": 431701.5625,
      "learning_rate": 4.4908285895003165e-05,
      "loss": 0.5603,
      "step": 2300
    },
    {
      "epoch": 9.35483870967742,
      "grad_norm": 227461.75,
      "learning_rate": 4.427577482605946e-05,
      "loss": 0.5625,
      "step": 2320
    },
    {
      "epoch": 9.435483870967742,
      "grad_norm": 571204.125,
      "learning_rate": 4.364326375711575e-05,
      "loss": 0.5643,
      "step": 2340
    },
    {
      "epoch": 9.516129032258064,
      "grad_norm": 377639.90625,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.6546,
      "step": 2360
    },
    {
      "epoch": 9.596774193548388,
      "grad_norm": 482121.9375,
      "learning_rate": 4.2378241619228334e-05,
      "loss": 0.6187,
      "step": 2380
    },
    {
      "epoch": 9.67741935483871,
      "grad_norm": 246030.6875,
      "learning_rate": 4.1745730550284635e-05,
      "loss": 0.5258,
      "step": 2400
    },
    {
      "epoch": 9.758064516129032,
      "grad_norm": 345079.0,
      "learning_rate": 4.111321948134092e-05,
      "loss": 0.5514,
      "step": 2420
    },
    {
      "epoch": 9.838709677419354,
      "grad_norm": 883186.4375,
      "learning_rate": 4.048070841239722e-05,
      "loss": 0.7296,
      "step": 2440
    },
    {
      "epoch": 9.919354838709678,
      "grad_norm": 354193.5625,
      "learning_rate": 3.984819734345351e-05,
      "loss": 0.6133,
      "step": 2460
    },
    {
      "epoch": 10.0,
      "grad_norm": 91423.2890625,
      "learning_rate": 3.9215686274509805e-05,
      "loss": 0.5622,
      "step": 2480
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.597165991902834,
      "eval_f1_macro": 0.5942014872446945,
      "eval_f1_weighted": 0.5978411174633815,
      "eval_loss": 1.294508934020996,
      "eval_runtime": 6.4737,
      "eval_samples_per_second": 76.309,
      "eval_steps_per_second": 2.472,
      "step": 2480
    },
    {
      "epoch": 10.080645161290322,
      "grad_norm": 335653.0,
      "learning_rate": 3.85831752055661e-05,
      "loss": 0.5815,
      "step": 2500
    },
    {
      "epoch": 10.161290322580646,
      "grad_norm": 485680.0625,
      "learning_rate": 3.795066413662239e-05,
      "loss": 0.5126,
      "step": 2520
    },
    {
      "epoch": 10.241935483870968,
      "grad_norm": 425775.0,
      "learning_rate": 3.731815306767869e-05,
      "loss": 0.5556,
      "step": 2540
    },
    {
      "epoch": 10.32258064516129,
      "grad_norm": 484495.375,
      "learning_rate": 3.668564199873498e-05,
      "loss": 0.5939,
      "step": 2560
    },
    {
      "epoch": 10.403225806451612,
      "grad_norm": 234732.140625,
      "learning_rate": 3.6053130929791276e-05,
      "loss": 0.4965,
      "step": 2580
    },
    {
      "epoch": 10.483870967741936,
      "grad_norm": 222726.84375,
      "learning_rate": 3.542061986084756e-05,
      "loss": 0.4604,
      "step": 2600
    },
    {
      "epoch": 10.564516129032258,
      "grad_norm": 430689.0625,
      "learning_rate": 3.4788108791903864e-05,
      "loss": 0.5109,
      "step": 2620
    },
    {
      "epoch": 10.64516129032258,
      "grad_norm": 174353.125,
      "learning_rate": 3.415559772296015e-05,
      "loss": 0.5173,
      "step": 2640
    },
    {
      "epoch": 10.725806451612904,
      "grad_norm": 350186.15625,
      "learning_rate": 3.3523086654016445e-05,
      "loss": 0.5976,
      "step": 2660
    },
    {
      "epoch": 10.806451612903226,
      "grad_norm": 342012.46875,
      "learning_rate": 3.289057558507274e-05,
      "loss": 0.5656,
      "step": 2680
    },
    {
      "epoch": 10.887096774193548,
      "grad_norm": 364261.15625,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.5657,
      "step": 2700
    },
    {
      "epoch": 10.967741935483872,
      "grad_norm": 565346.0,
      "learning_rate": 3.162555344718533e-05,
      "loss": 0.5226,
      "step": 2720
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5809716599190283,
      "eval_f1_macro": 0.5665436314892837,
      "eval_f1_weighted": 0.5852790985052031,
      "eval_loss": 1.233909249305725,
      "eval_runtime": 6.4774,
      "eval_samples_per_second": 76.265,
      "eval_steps_per_second": 2.47,
      "step": 2728
    },
    {
      "epoch": 11.048387096774194,
      "grad_norm": 787262.375,
      "learning_rate": 3.099304237824162e-05,
      "loss": 0.5361,
      "step": 2740
    },
    {
      "epoch": 11.129032258064516,
      "grad_norm": 254161.28125,
      "learning_rate": 3.0360531309297913e-05,
      "loss": 0.4992,
      "step": 2760
    },
    {
      "epoch": 11.209677419354838,
      "grad_norm": 154221.5625,
      "learning_rate": 2.972802024035421e-05,
      "loss": 0.4863,
      "step": 2780
    },
    {
      "epoch": 11.290322580645162,
      "grad_norm": 172757.25,
      "learning_rate": 2.90955091714105e-05,
      "loss": 0.4555,
      "step": 2800
    },
    {
      "epoch": 11.370967741935484,
      "grad_norm": 600217.1875,
      "learning_rate": 2.846299810246679e-05,
      "loss": 0.5628,
      "step": 2820
    },
    {
      "epoch": 11.451612903225806,
      "grad_norm": 292712.21875,
      "learning_rate": 2.783048703352309e-05,
      "loss": 0.4477,
      "step": 2840
    },
    {
      "epoch": 11.532258064516128,
      "grad_norm": 347355.25,
      "learning_rate": 2.719797596457938e-05,
      "loss": 0.4457,
      "step": 2860
    },
    {
      "epoch": 11.612903225806452,
      "grad_norm": 448712.625,
      "learning_rate": 2.656546489563567e-05,
      "loss": 0.4794,
      "step": 2880
    },
    {
      "epoch": 11.693548387096774,
      "grad_norm": 236689.59375,
      "learning_rate": 2.5932953826691968e-05,
      "loss": 0.4448,
      "step": 2900
    },
    {
      "epoch": 11.774193548387096,
      "grad_norm": 182560.375,
      "learning_rate": 2.5300442757748262e-05,
      "loss": 0.4482,
      "step": 2920
    },
    {
      "epoch": 11.85483870967742,
      "grad_norm": 478094.78125,
      "learning_rate": 2.4667931688804556e-05,
      "loss": 0.6564,
      "step": 2940
    },
    {
      "epoch": 11.935483870967742,
      "grad_norm": 132165.9375,
      "learning_rate": 2.403542061986085e-05,
      "loss": 0.5092,
      "step": 2960
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5991902834008097,
      "eval_f1_macro": 0.5808594109741428,
      "eval_f1_weighted": 0.601863769019088,
      "eval_loss": 1.3011294603347778,
      "eval_runtime": 6.7266,
      "eval_samples_per_second": 73.44,
      "eval_steps_per_second": 2.379,
      "step": 2976
    },
    {
      "epoch": 12.016129032258064,
      "grad_norm": 360605.40625,
      "learning_rate": 2.3402909550917145e-05,
      "loss": 0.5019,
      "step": 2980
    },
    {
      "epoch": 12.096774193548388,
      "grad_norm": 383628.65625,
      "learning_rate": 2.2770398481973435e-05,
      "loss": 0.4511,
      "step": 3000
    },
    {
      "epoch": 12.17741935483871,
      "grad_norm": 293539.46875,
      "learning_rate": 2.213788741302973e-05,
      "loss": 0.4609,
      "step": 3020
    },
    {
      "epoch": 12.258064516129032,
      "grad_norm": 255752.75,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.4805,
      "step": 3040
    },
    {
      "epoch": 12.338709677419354,
      "grad_norm": 358986.28125,
      "learning_rate": 2.0872865275142318e-05,
      "loss": 0.4505,
      "step": 3060
    },
    {
      "epoch": 12.419354838709678,
      "grad_norm": 296015.03125,
      "learning_rate": 2.024035420619861e-05,
      "loss": 0.4456,
      "step": 3080
    },
    {
      "epoch": 12.5,
      "grad_norm": 709465.75,
      "learning_rate": 1.9607843137254903e-05,
      "loss": 0.4525,
      "step": 3100
    },
    {
      "epoch": 12.580645161290322,
      "grad_norm": 718661.8125,
      "learning_rate": 1.8975332068311197e-05,
      "loss": 0.4375,
      "step": 3120
    },
    {
      "epoch": 12.661290322580646,
      "grad_norm": 665258.75,
      "learning_rate": 1.834282099936749e-05,
      "loss": 0.575,
      "step": 3140
    },
    {
      "epoch": 12.741935483870968,
      "grad_norm": 338319.0,
      "learning_rate": 1.771030993042378e-05,
      "loss": 0.4769,
      "step": 3160
    },
    {
      "epoch": 12.82258064516129,
      "grad_norm": 389402.6875,
      "learning_rate": 1.7077798861480076e-05,
      "loss": 0.4512,
      "step": 3180
    },
    {
      "epoch": 12.903225806451612,
      "grad_norm": 488243.21875,
      "learning_rate": 1.644528779253637e-05,
      "loss": 0.5537,
      "step": 3200
    },
    {
      "epoch": 12.983870967741936,
      "grad_norm": 515383.25,
      "learning_rate": 1.5812776723592664e-05,
      "loss": 0.4968,
      "step": 3220
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5910931174089069,
      "eval_f1_macro": 0.5810430688559796,
      "eval_f1_weighted": 0.5958028312757723,
      "eval_loss": 1.3416694402694702,
      "eval_runtime": 6.488,
      "eval_samples_per_second": 76.141,
      "eval_steps_per_second": 2.466,
      "step": 3224
    }
  ],
  "logging_steps": 20,
  "max_steps": 3720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7069215821930496.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}