{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.11897326075964428,
  "eval_steps": 1000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy_step_1": 0.1269,
      "entropy_step_2": 0.1259,
      "entropy_step_3": 0.1199,
      "epoch": 0.0002974331518991107,
      "grad_norm": 14.0,
      "learning_rate": 5.3507728894173604e-08,
      "loss": 0.05490208864212036,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0431,
      "step": 10
    },
    {
      "entropy_step_1": 0.1207,
      "entropy_step_2": 0.1189,
      "entropy_step_3": 0.115,
      "epoch": 0.0005948663037982214,
      "grad_norm": 15.875,
      "learning_rate": 1.1296076099881095e-07,
      "loss": 0.04469497203826904,
      "loss_step_1": 0.0403,
      "loss_step_2": 0.0402,
      "loss_step_3": 0.044,
      "step": 20
    },
    {
      "entropy_step_1": 0.1305,
      "entropy_step_2": 0.1294,
      "entropy_step_3": 0.145,
      "epoch": 0.000892299455697332,
      "grad_norm": 11.0625,
      "learning_rate": 1.7241379310344828e-07,
      "loss": 0.04919552206993103,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0498,
      "loss_step_3": 0.0551,
      "step": 30
    },
    {
      "entropy_step_1": 0.136,
      "entropy_step_2": 0.135,
      "entropy_step_3": 0.1313,
      "epoch": 0.0011897326075964428,
      "grad_norm": 11.0625,
      "learning_rate": 2.318668252080856e-07,
      "loss": 0.04959763884544373,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0431,
      "step": 40
    },
    {
      "entropy_step_1": 0.115,
      "entropy_step_2": 0.1137,
      "entropy_step_3": 0.094,
      "epoch": 0.0014871657594955534,
      "grad_norm": 14.375,
      "learning_rate": 2.91319857312723e-07,
      "loss": 0.04376085996627808,
      "loss_step_1": 0.0421,
      "loss_step_2": 0.0422,
      "loss_step_3": 0.0359,
      "step": 50
    },
    {
      "entropy_step_1": 0.1394,
      "entropy_step_2": 0.1388,
      "entropy_step_3": 0.1556,
      "epoch": 0.001784598911394664,
      "grad_norm": 6.15625,
      "learning_rate": 3.507728894173603e-07,
      "loss": 0.04998557865619659,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0598,
      "step": 60
    },
    {
      "entropy_step_1": 0.1365,
      "entropy_step_2": 0.1362,
      "entropy_step_3": 0.1303,
      "epoch": 0.002082032063293775,
      "grad_norm": 11.1875,
      "learning_rate": 4.102259215219977e-07,
      "loss": 0.06390773057937622,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.0701,
      "step": 70
    },
    {
      "entropy_step_1": 0.1101,
      "entropy_step_2": 0.1098,
      "entropy_step_3": 0.1155,
      "epoch": 0.0023794652151928855,
      "grad_norm": 16.5,
      "learning_rate": 4.6967895362663505e-07,
      "loss": 0.04625580906867981,
      "loss_step_1": 0.0438,
      "loss_step_2": 0.0427,
      "loss_step_3": 0.0444,
      "step": 80
    },
    {
      "entropy_step_1": 0.1028,
      "entropy_step_2": 0.1025,
      "entropy_step_3": 0.1149,
      "epoch": 0.002676898367091996,
      "grad_norm": 34.5,
      "learning_rate": 5.291319857312724e-07,
      "loss": 0.04806084930896759,
      "loss_step_1": 0.0494,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0628,
      "step": 90
    },
    {
      "entropy_step_1": 0.1245,
      "entropy_step_2": 0.1236,
      "entropy_step_3": 0.1064,
      "epoch": 0.002974331518991107,
      "grad_norm": 8.0,
      "learning_rate": 5.885850178359096e-07,
      "loss": 0.06233357787132263,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0557,
      "step": 100
    },
    {
      "entropy_step_1": 0.1217,
      "entropy_step_2": 0.1208,
      "entropy_step_3": 0.1387,
      "epoch": 0.0032717646708902174,
      "grad_norm": 21.375,
      "learning_rate": 6.48038049940547e-07,
      "loss": 0.054258376359939575,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.056,
      "step": 110
    },
    {
      "entropy_step_1": 0.1273,
      "entropy_step_2": 0.1261,
      "entropy_step_3": 0.1379,
      "epoch": 0.003569197822789328,
      "grad_norm": 20.25,
      "learning_rate": 7.074910820451844e-07,
      "loss": 0.057795822620391846,
      "loss_step_1": 0.0514,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0441,
      "step": 120
    },
    {
      "entropy_step_1": 0.1262,
      "entropy_step_2": 0.1257,
      "entropy_step_3": 0.1129,
      "epoch": 0.0038666309746884387,
      "grad_norm": 47.25,
      "learning_rate": 7.669441141498217e-07,
      "loss": 0.060700607299804685,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0637,
      "step": 130
    },
    {
      "entropy_step_1": 0.1284,
      "entropy_step_2": 0.127,
      "entropy_step_3": 0.1284,
      "epoch": 0.00416406412658755,
      "grad_norm": 21.375,
      "learning_rate": 8.26397146254459e-07,
      "loss": 0.05582804679870605,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0557,
      "step": 140
    },
    {
      "entropy_step_1": 0.1399,
      "entropy_step_2": 0.1381,
      "entropy_step_3": 0.1233,
      "epoch": 0.00446149727848666,
      "grad_norm": 39.0,
      "learning_rate": 8.858501783590964e-07,
      "loss": 0.05377758741378784,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.056,
      "step": 150
    },
    {
      "entropy_step_1": 0.1536,
      "entropy_step_2": 0.1513,
      "entropy_step_3": 0.1396,
      "epoch": 0.004758930430385771,
      "grad_norm": 14.125,
      "learning_rate": 9.453032104637337e-07,
      "loss": 0.05115737318992615,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0535,
      "step": 160
    },
    {
      "entropy_step_1": 0.1138,
      "entropy_step_2": 0.1137,
      "entropy_step_3": 0.1191,
      "epoch": 0.005056363582284881,
      "grad_norm": 7.3125,
      "learning_rate": 1.004756242568371e-06,
      "loss": 0.04697312712669373,
      "loss_step_1": 0.0412,
      "loss_step_2": 0.0406,
      "loss_step_3": 0.0308,
      "step": 170
    },
    {
      "entropy_step_1": 0.1464,
      "entropy_step_2": 0.1453,
      "entropy_step_3": 0.1618,
      "epoch": 0.005353796734183992,
      "grad_norm": 13.1875,
      "learning_rate": 1.0642092746730084e-06,
      "loss": 0.05087941884994507,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0546,
      "step": 180
    },
    {
      "entropy_step_1": 0.131,
      "entropy_step_2": 0.1296,
      "entropy_step_3": 0.1155,
      "epoch": 0.0056512298860831025,
      "grad_norm": 7.09375,
      "learning_rate": 1.1236623067776457e-06,
      "loss": 0.050765782594680786,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0449,
      "step": 190
    },
    {
      "entropy_step_1": 0.1336,
      "entropy_step_2": 0.1325,
      "entropy_step_3": 0.1349,
      "epoch": 0.005948663037982214,
      "grad_norm": 24.875,
      "learning_rate": 1.183115338882283e-06,
      "loss": 0.0555966317653656,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0629,
      "step": 200
    },
    {
      "entropy_step_1": 0.1451,
      "entropy_step_2": 0.1444,
      "entropy_step_3": 0.133,
      "epoch": 0.006246096189881324,
      "grad_norm": 24.125,
      "learning_rate": 1.2425683709869204e-06,
      "loss": 0.04687873125076294,
      "loss_step_1": 0.0428,
      "loss_step_2": 0.0424,
      "loss_step_3": 0.0486,
      "step": 210
    },
    {
      "entropy_step_1": 0.1562,
      "entropy_step_2": 0.1543,
      "entropy_step_3": 0.1447,
      "epoch": 0.006543529341780435,
      "grad_norm": 8.0625,
      "learning_rate": 1.302021403091558e-06,
      "loss": 0.046315240859985354,
      "loss_step_1": 0.0466,
      "loss_step_2": 0.0463,
      "loss_step_3": 0.0383,
      "step": 220
    },
    {
      "entropy_step_1": 0.1257,
      "entropy_step_2": 0.125,
      "entropy_step_3": 0.1262,
      "epoch": 0.006840962493679546,
      "grad_norm": 4.6875,
      "learning_rate": 1.361474435196195e-06,
      "loss": 0.0415429562330246,
      "loss_step_1": 0.0401,
      "loss_step_2": 0.0402,
      "loss_step_3": 0.0409,
      "step": 230
    },
    {
      "entropy_step_1": 0.172,
      "entropy_step_2": 0.1712,
      "entropy_step_3": 0.1789,
      "epoch": 0.007138395645578656,
      "grad_norm": 18.375,
      "learning_rate": 1.4209274673008325e-06,
      "loss": 0.050160109996795654,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0619,
      "step": 240
    },
    {
      "entropy_step_1": 0.1698,
      "entropy_step_2": 0.1682,
      "entropy_step_3": 0.1759,
      "epoch": 0.007435828797477767,
      "grad_norm": 10.0625,
      "learning_rate": 1.48038049940547e-06,
      "loss": 0.05399076938629151,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0658,
      "step": 250
    },
    {
      "entropy_step_1": 0.148,
      "entropy_step_2": 0.1467,
      "entropy_step_3": 0.1566,
      "epoch": 0.007733261949376877,
      "grad_norm": 12.1875,
      "learning_rate": 1.539833531510107e-06,
      "loss": 0.04920393526554108,
      "loss_step_1": 0.0444,
      "loss_step_2": 0.0442,
      "loss_step_3": 0.0443,
      "step": 260
    },
    {
      "entropy_step_1": 0.1589,
      "entropy_step_2": 0.1574,
      "entropy_step_3": 0.1636,
      "epoch": 0.008030695101275988,
      "grad_norm": 22.25,
      "learning_rate": 1.5992865636147445e-06,
      "loss": 0.048985743522644044,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.0474,
      "step": 270
    },
    {
      "entropy_step_1": 0.1612,
      "entropy_step_2": 0.1602,
      "entropy_step_3": 0.1658,
      "epoch": 0.0083281282531751,
      "grad_norm": 4.21875,
      "learning_rate": 1.658739595719382e-06,
      "loss": 0.05151403546333313,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0554,
      "step": 280
    },
    {
      "entropy_step_1": 0.1453,
      "entropy_step_2": 0.1448,
      "entropy_step_3": 0.1311,
      "epoch": 0.008625561405074209,
      "grad_norm": 15.5,
      "learning_rate": 1.718192627824019e-06,
      "loss": 0.04945358633995056,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0434,
      "step": 290
    },
    {
      "entropy_step_1": 0.146,
      "entropy_step_2": 0.1443,
      "entropy_step_3": 0.1471,
      "epoch": 0.00892299455697332,
      "grad_norm": 11.6875,
      "learning_rate": 1.7776456599286565e-06,
      "loss": 0.04783240556716919,
      "loss_step_1": 0.0476,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.0537,
      "step": 300
    },
    {
      "entropy_step_1": 0.1487,
      "entropy_step_2": 0.1471,
      "entropy_step_3": 0.1367,
      "epoch": 0.009220427708872431,
      "grad_norm": 3.03125,
      "learning_rate": 1.837098692033294e-06,
      "loss": 0.04544385075569153,
      "loss_step_1": 0.0434,
      "loss_step_2": 0.0436,
      "loss_step_3": 0.0478,
      "step": 310
    },
    {
      "entropy_step_1": 0.1569,
      "entropy_step_2": 0.1558,
      "entropy_step_3": 0.1429,
      "epoch": 0.009517860860771542,
      "grad_norm": 7.34375,
      "learning_rate": 1.896551724137931e-06,
      "loss": 0.0506971538066864,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0593,
      "step": 320
    },
    {
      "entropy_step_1": 0.1506,
      "entropy_step_2": 0.1482,
      "entropy_step_3": 0.1662,
      "epoch": 0.009815294012670651,
      "grad_norm": 13.0,
      "learning_rate": 1.9560047562425685e-06,
      "loss": 0.045241162180900574,
      "loss_step_1": 0.0449,
      "loss_step_2": 0.0446,
      "loss_step_3": 0.0506,
      "step": 330
    },
    {
      "entropy_step_1": 0.1557,
      "entropy_step_2": 0.1545,
      "entropy_step_3": 0.1431,
      "epoch": 0.010112727164569763,
      "grad_norm": 15.4375,
      "learning_rate": 2.0154577883472058e-06,
      "loss": 0.05822656750679016,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0541,
      "step": 340
    },
    {
      "entropy_step_1": 0.1679,
      "entropy_step_2": 0.1654,
      "entropy_step_3": 0.1668,
      "epoch": 0.010410160316468874,
      "grad_norm": 5.96875,
      "learning_rate": 2.074910820451843e-06,
      "loss": 0.05308838486671448,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0568,
      "step": 350
    },
    {
      "entropy_step_1": 0.1384,
      "entropy_step_2": 0.1358,
      "entropy_step_3": 0.1377,
      "epoch": 0.010707593468367985,
      "grad_norm": 12.5,
      "learning_rate": 2.1343638525564807e-06,
      "loss": 0.04592644572257996,
      "loss_step_1": 0.0446,
      "loss_step_2": 0.0446,
      "loss_step_3": 0.0459,
      "step": 360
    },
    {
      "entropy_step_1": 0.1684,
      "entropy_step_2": 0.1655,
      "entropy_step_3": 0.1695,
      "epoch": 0.011005026620267096,
      "grad_norm": 7.15625,
      "learning_rate": 2.193816884661118e-06,
      "loss": 0.05064881443977356,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0501,
      "step": 370
    },
    {
      "entropy_step_1": 0.1694,
      "entropy_step_2": 0.1668,
      "entropy_step_3": 0.1727,
      "epoch": 0.011302459772166205,
      "grad_norm": 5.96875,
      "learning_rate": 2.2532699167657553e-06,
      "loss": 0.04952639639377594,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0581,
      "step": 380
    },
    {
      "entropy_step_1": 0.1522,
      "entropy_step_2": 0.1496,
      "entropy_step_3": 0.1574,
      "epoch": 0.011599892924065316,
      "grad_norm": 15.5625,
      "learning_rate": 2.3127229488703926e-06,
      "loss": 0.05158161520957947,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0737,
      "step": 390
    },
    {
      "entropy_step_1": 0.1568,
      "entropy_step_2": 0.1558,
      "entropy_step_3": 0.1474,
      "epoch": 0.011897326075964427,
      "grad_norm": 11.4375,
      "learning_rate": 2.37217598097503e-06,
      "loss": 0.047658208012580874,
      "loss_step_1": 0.0477,
      "loss_step_2": 0.0477,
      "loss_step_3": 0.0487,
      "step": 400
    },
    {
      "entropy_step_1": 0.15,
      "entropy_step_2": 0.1484,
      "entropy_step_3": 0.1594,
      "epoch": 0.012194759227863538,
      "grad_norm": 9.0625,
      "learning_rate": 2.431629013079667e-06,
      "loss": 0.04629878401756286,
      "loss_step_1": 0.0455,
      "loss_step_2": 0.0455,
      "loss_step_3": 0.0519,
      "step": 410
    },
    {
      "entropy_step_1": 0.1484,
      "entropy_step_2": 0.1478,
      "entropy_step_3": 0.1504,
      "epoch": 0.012492192379762648,
      "grad_norm": 5.0,
      "learning_rate": 2.4910820451843048e-06,
      "loss": 0.04484026730060577,
      "loss_step_1": 0.0455,
      "loss_step_2": 0.0454,
      "loss_step_3": 0.0464,
      "step": 420
    },
    {
      "entropy_step_1": 0.1718,
      "entropy_step_2": 0.1709,
      "entropy_step_3": 0.1441,
      "epoch": 0.012789625531661759,
      "grad_norm": 23.25,
      "learning_rate": 2.550535077288942e-06,
      "loss": 0.05759080052375794,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0527,
      "step": 430
    },
    {
      "entropy_step_1": 0.1541,
      "entropy_step_2": 0.1533,
      "entropy_step_3": 0.1482,
      "epoch": 0.01308705868356087,
      "grad_norm": 7.90625,
      "learning_rate": 2.6099881093935793e-06,
      "loss": 0.05002807378768921,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0492,
      "step": 440
    },
    {
      "entropy_step_1": 0.1462,
      "entropy_step_2": 0.1452,
      "entropy_step_3": 0.1463,
      "epoch": 0.01338449183545998,
      "grad_norm": 5.59375,
      "learning_rate": 2.669441141498217e-06,
      "loss": 0.04358697235584259,
      "loss_step_1": 0.0404,
      "loss_step_2": 0.0404,
      "loss_step_3": 0.0381,
      "step": 450
    },
    {
      "entropy_step_1": 0.1742,
      "entropy_step_2": 0.1725,
      "entropy_step_3": 0.1815,
      "epoch": 0.013681924987359092,
      "grad_norm": 3.609375,
      "learning_rate": 2.728894173602854e-06,
      "loss": 0.05483717322349548,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0693,
      "step": 460
    },
    {
      "entropy_step_1": 0.1476,
      "entropy_step_2": 0.1455,
      "entropy_step_3": 0.1431,
      "epoch": 0.013979358139258201,
      "grad_norm": 4.90625,
      "learning_rate": 2.788347205707491e-06,
      "loss": 0.04310814142227173,
      "loss_step_1": 0.0393,
      "loss_step_2": 0.0389,
      "loss_step_3": 0.0353,
      "step": 470
    },
    {
      "entropy_step_1": 0.1487,
      "entropy_step_2": 0.1464,
      "entropy_step_3": 0.1519,
      "epoch": 0.014276791291157312,
      "grad_norm": 9.75,
      "learning_rate": 2.847800237812129e-06,
      "loss": 0.04548307061195374,
      "loss_step_1": 0.0407,
      "loss_step_2": 0.0405,
      "loss_step_3": 0.0465,
      "step": 480
    },
    {
      "entropy_step_1": 0.1405,
      "entropy_step_2": 0.1389,
      "entropy_step_3": 0.1252,
      "epoch": 0.014574224443056423,
      "grad_norm": 8.6875,
      "learning_rate": 2.907253269916766e-06,
      "loss": 0.04080027043819427,
      "loss_step_1": 0.0417,
      "loss_step_2": 0.0413,
      "loss_step_3": 0.039,
      "step": 490
    },
    {
      "entropy_step_1": 0.1499,
      "entropy_step_2": 0.1482,
      "entropy_step_3": 0.1548,
      "epoch": 0.014871657594955534,
      "grad_norm": 16.75,
      "learning_rate": 2.9667063020214034e-06,
      "loss": 0.050579404830932616,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0524,
      "step": 500
    },
    {
      "entropy_step_1": 0.1307,
      "entropy_step_2": 0.129,
      "entropy_step_3": 0.1236,
      "epoch": 0.015169090746854644,
      "grad_norm": 8.375,
      "learning_rate": 3.026159334126041e-06,
      "loss": 0.04416508078575134,
      "loss_step_1": 0.0438,
      "loss_step_2": 0.0439,
      "loss_step_3": 0.0452,
      "step": 510
    },
    {
      "entropy_step_1": 0.1425,
      "entropy_step_2": 0.1418,
      "entropy_step_3": 0.1373,
      "epoch": 0.015466523898753755,
      "grad_norm": 13.6875,
      "learning_rate": 3.085612366230678e-06,
      "loss": 0.04708780348300934,
      "loss_step_1": 0.0412,
      "loss_step_2": 0.0412,
      "loss_step_3": 0.0353,
      "step": 520
    },
    {
      "entropy_step_1": 0.1576,
      "entropy_step_2": 0.1552,
      "entropy_step_3": 0.1799,
      "epoch": 0.015763957050652866,
      "grad_norm": 9.4375,
      "learning_rate": 3.145065398335315e-06,
      "loss": 0.05313628911972046,
      "loss_step_1": 0.0504,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0569,
      "step": 530
    },
    {
      "entropy_step_1": 0.1531,
      "entropy_step_2": 0.1501,
      "entropy_step_3": 0.1634,
      "epoch": 0.016061390202551977,
      "grad_norm": 5.0625,
      "learning_rate": 3.204518430439953e-06,
      "loss": 0.051033306121826175,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0574,
      "step": 540
    },
    {
      "entropy_step_1": 0.1596,
      "entropy_step_2": 0.1569,
      "entropy_step_3": 0.1328,
      "epoch": 0.016358823354451088,
      "grad_norm": 16.125,
      "learning_rate": 3.26397146254459e-06,
      "loss": 0.049698644876480104,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0458,
      "step": 550
    },
    {
      "entropy_step_1": 0.1557,
      "entropy_step_2": 0.1536,
      "entropy_step_3": 0.1587,
      "epoch": 0.0166562565063502,
      "grad_norm": 10.875,
      "learning_rate": 3.3234244946492274e-06,
      "loss": 0.05324244499206543,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0633,
      "step": 560
    },
    {
      "entropy_step_1": 0.183,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1813,
      "epoch": 0.01695368965824931,
      "grad_norm": 19.625,
      "learning_rate": 3.382877526753865e-06,
      "loss": 0.06462842226028442,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0667,
      "step": 570
    },
    {
      "entropy_step_1": 0.1527,
      "entropy_step_2": 0.1503,
      "entropy_step_3": 0.1274,
      "epoch": 0.017251122810148418,
      "grad_norm": 8.625,
      "learning_rate": 3.442330558858502e-06,
      "loss": 0.05306703448295593,
      "loss_step_1": 0.0514,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0428,
      "step": 580
    },
    {
      "entropy_step_1": 0.1837,
      "entropy_step_2": 0.1814,
      "entropy_step_3": 0.1805,
      "epoch": 0.01754855596204753,
      "grad_norm": 7.40625,
      "learning_rate": 3.501783590963139e-06,
      "loss": 0.05398572087287903,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.056,
      "step": 590
    },
    {
      "entropy_step_1": 0.1595,
      "entropy_step_2": 0.157,
      "entropy_step_3": 0.1698,
      "epoch": 0.01784598911394664,
      "grad_norm": 11.5,
      "learning_rate": 3.561236623067777e-06,
      "loss": 0.045186972618103026,
      "loss_step_1": 0.0459,
      "loss_step_2": 0.0459,
      "loss_step_3": 0.048,
      "step": 600
    },
    {
      "entropy_step_1": 0.169,
      "entropy_step_2": 0.1673,
      "entropy_step_3": 0.1549,
      "epoch": 0.01814342226584575,
      "grad_norm": 3.171875,
      "learning_rate": 3.620689655172414e-06,
      "loss": 0.04781789183616638,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0519,
      "step": 610
    },
    {
      "entropy_step_1": 0.1434,
      "entropy_step_2": 0.1411,
      "entropy_step_3": 0.1405,
      "epoch": 0.018440855417744862,
      "grad_norm": 12.5625,
      "learning_rate": 3.6801426872770514e-06,
      "loss": 0.043872755765914914,
      "loss_step_1": 0.0424,
      "loss_step_2": 0.0422,
      "loss_step_3": 0.0426,
      "step": 620
    },
    {
      "entropy_step_1": 0.1565,
      "entropy_step_2": 0.1549,
      "entropy_step_3": 0.171,
      "epoch": 0.018738288569643973,
      "grad_norm": 13.3125,
      "learning_rate": 3.739595719381689e-06,
      "loss": 0.04813492298126221,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0602,
      "step": 630
    },
    {
      "entropy_step_1": 0.1551,
      "entropy_step_2": 0.1539,
      "entropy_step_3": 0.1307,
      "epoch": 0.019035721721543084,
      "grad_norm": 19.75,
      "learning_rate": 3.799048751486326e-06,
      "loss": 0.05256493091583252,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0423,
      "step": 640
    },
    {
      "entropy_step_1": 0.1443,
      "entropy_step_2": 0.1412,
      "entropy_step_3": 0.1274,
      "epoch": 0.019333154873442195,
      "grad_norm": 3.796875,
      "learning_rate": 3.858501783590964e-06,
      "loss": 0.05315169095993042,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0394,
      "step": 650
    },
    {
      "entropy_step_1": 0.1629,
      "entropy_step_2": 0.1575,
      "entropy_step_3": 0.1655,
      "epoch": 0.019630588025341303,
      "grad_norm": 6.53125,
      "learning_rate": 3.917954815695601e-06,
      "loss": 0.04939117431640625,
      "loss_step_1": 0.0476,
      "loss_step_2": 0.0473,
      "loss_step_3": 0.0567,
      "step": 660
    },
    {
      "entropy_step_1": 0.1797,
      "entropy_step_2": 0.1737,
      "entropy_step_3": 0.1862,
      "epoch": 0.019928021177240414,
      "grad_norm": 4.625,
      "learning_rate": 3.977407847800238e-06,
      "loss": 0.053183114528656004,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0529,
      "step": 670
    },
    {
      "entropy_step_1": 0.1681,
      "entropy_step_2": 0.1631,
      "entropy_step_3": 0.1674,
      "epoch": 0.020225454329139525,
      "grad_norm": 11.375,
      "learning_rate": 4.0368608799048755e-06,
      "loss": 0.047932100296020505,
      "loss_step_1": 0.0469,
      "loss_step_2": 0.0468,
      "loss_step_3": 0.0491,
      "step": 680
    },
    {
      "entropy_step_1": 0.1416,
      "entropy_step_2": 0.1371,
      "entropy_step_3": 0.1383,
      "epoch": 0.020522887481038636,
      "grad_norm": 8.0,
      "learning_rate": 4.096313912009513e-06,
      "loss": 0.04632853865623474,
      "loss_step_1": 0.0465,
      "loss_step_2": 0.0467,
      "loss_step_3": 0.0444,
      "step": 690
    },
    {
      "entropy_step_1": 0.1612,
      "entropy_step_2": 0.1556,
      "entropy_step_3": 0.168,
      "epoch": 0.020820320632937747,
      "grad_norm": 6.90625,
      "learning_rate": 4.15576694411415e-06,
      "loss": 0.0525726854801178,
      "loss_step_1": 0.0508,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.056,
      "step": 700
    },
    {
      "entropy_step_1": 0.1394,
      "entropy_step_2": 0.1371,
      "entropy_step_3": 0.1398,
      "epoch": 0.021117753784836858,
      "grad_norm": 7.25,
      "learning_rate": 4.215219976218787e-06,
      "loss": 0.049972376227378844,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0473,
      "loss_step_3": 0.0452,
      "step": 710
    },
    {
      "entropy_step_1": 0.1313,
      "entropy_step_2": 0.1291,
      "entropy_step_3": 0.1283,
      "epoch": 0.02141518693673597,
      "grad_norm": 12.6875,
      "learning_rate": 4.274673008323425e-06,
      "loss": 0.04236210286617279,
      "loss_step_1": 0.0415,
      "loss_step_2": 0.0418,
      "loss_step_3": 0.0345,
      "step": 720
    },
    {
      "entropy_step_1": 0.1335,
      "entropy_step_2": 0.1317,
      "entropy_step_3": 0.1401,
      "epoch": 0.02171262008863508,
      "grad_norm": 8.75,
      "learning_rate": 4.334126040428062e-06,
      "loss": 0.05107964277267456,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.045,
      "step": 730
    },
    {
      "entropy_step_1": 0.1678,
      "entropy_step_2": 0.1647,
      "entropy_step_3": 0.18,
      "epoch": 0.02201005324053419,
      "grad_norm": 11.9375,
      "learning_rate": 4.393579072532699e-06,
      "loss": 0.04507316946983338,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0554,
      "step": 740
    },
    {
      "entropy_step_1": 0.1718,
      "entropy_step_2": 0.1687,
      "entropy_step_3": 0.1601,
      "epoch": 0.0223074863924333,
      "grad_norm": 5.125,
      "learning_rate": 4.453032104637337e-06,
      "loss": 0.047673165798187256,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0514,
      "loss_step_3": 0.0477,
      "step": 750
    },
    {
      "entropy_step_1": 0.1489,
      "entropy_step_2": 0.1473,
      "entropy_step_3": 0.1569,
      "epoch": 0.02260491954433241,
      "grad_norm": 2.828125,
      "learning_rate": 4.5124851367419745e-06,
      "loss": 0.05465167760848999,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0497,
      "loss_step_3": 0.0484,
      "step": 760
    },
    {
      "entropy_step_1": 0.1623,
      "entropy_step_2": 0.1609,
      "entropy_step_3": 0.1647,
      "epoch": 0.02290235269623152,
      "grad_norm": 11.6875,
      "learning_rate": 4.571938168846612e-06,
      "loss": 0.04933509826660156,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0479,
      "loss_step_3": 0.0518,
      "step": 770
    },
    {
      "entropy_step_1": 0.1508,
      "entropy_step_2": 0.1497,
      "entropy_step_3": 0.1503,
      "epoch": 0.023199785848130632,
      "grad_norm": 17.0,
      "learning_rate": 4.631391200951249e-06,
      "loss": 0.051386290788650514,
      "loss_step_1": 0.0461,
      "loss_step_2": 0.0462,
      "loss_step_3": 0.0517,
      "step": 780
    },
    {
      "entropy_step_1": 0.152,
      "entropy_step_2": 0.1514,
      "entropy_step_3": 0.1418,
      "epoch": 0.023497219000029743,
      "grad_norm": 3.59375,
      "learning_rate": 4.690844233055886e-06,
      "loss": 0.0495593398809433,
      "loss_step_1": 0.0442,
      "loss_step_2": 0.0442,
      "loss_step_3": 0.0499,
      "step": 790
    },
    {
      "entropy_step_1": 0.1539,
      "entropy_step_2": 0.1527,
      "entropy_step_3": 0.1514,
      "epoch": 0.023794652151928854,
      "grad_norm": 5.78125,
      "learning_rate": 4.7502972651605235e-06,
      "loss": 0.04992133378982544,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0484,
      "step": 800
    },
    {
      "entropy_step_1": 0.1582,
      "entropy_step_2": 0.1572,
      "entropy_step_3": 0.155,
      "epoch": 0.024092085303827965,
      "grad_norm": 7.0,
      "learning_rate": 4.809750297265161e-06,
      "loss": 0.04976120293140411,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0482,
      "loss_step_3": 0.0472,
      "step": 810
    },
    {
      "entropy_step_1": 0.1352,
      "entropy_step_2": 0.1343,
      "entropy_step_3": 0.1312,
      "epoch": 0.024389518455727077,
      "grad_norm": 11.5625,
      "learning_rate": 4.869203329369798e-06,
      "loss": 0.04785799980163574,
      "loss_step_1": 0.0466,
      "loss_step_2": 0.0465,
      "loss_step_3": 0.0435,
      "step": 820
    },
    {
      "entropy_step_1": 0.167,
      "entropy_step_2": 0.1654,
      "entropy_step_3": 0.1591,
      "epoch": 0.024686951607626188,
      "grad_norm": 5.03125,
      "learning_rate": 4.928656361474435e-06,
      "loss": 0.05038613677024841,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0532,
      "step": 830
    },
    {
      "entropy_step_1": 0.1657,
      "entropy_step_2": 0.1628,
      "entropy_step_3": 0.1522,
      "epoch": 0.024984384759525295,
      "grad_norm": 7.125,
      "learning_rate": 4.9881093935790735e-06,
      "loss": 0.04596621990203857,
      "loss_step_1": 0.047,
      "loss_step_2": 0.0463,
      "loss_step_3": 0.0489,
      "step": 840
    },
    {
      "entropy_step_1": 0.166,
      "entropy_step_2": 0.1654,
      "entropy_step_3": 0.1771,
      "epoch": 0.025281817911424406,
      "grad_norm": 12.625,
      "learning_rate": 5.047562425683711e-06,
      "loss": 0.049768340587615964,
      "loss_step_1": 0.0487,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0566,
      "step": 850
    },
    {
      "entropy_step_1": 0.157,
      "entropy_step_2": 0.1556,
      "entropy_step_3": 0.1571,
      "epoch": 0.025579251063323517,
      "grad_norm": 6.5,
      "learning_rate": 5.107015457788348e-06,
      "loss": 0.053108209371566774,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.056,
      "step": 860
    },
    {
      "entropy_step_1": 0.1664,
      "entropy_step_2": 0.1655,
      "entropy_step_3": 0.1631,
      "epoch": 0.02587668421522263,
      "grad_norm": 7.03125,
      "learning_rate": 5.166468489892985e-06,
      "loss": 0.04641217887401581,
      "loss_step_1": 0.046,
      "loss_step_2": 0.0461,
      "loss_step_3": 0.0429,
      "step": 870
    },
    {
      "entropy_step_1": 0.1295,
      "entropy_step_2": 0.1282,
      "entropy_step_3": 0.1353,
      "epoch": 0.02617411736712174,
      "grad_norm": 3.890625,
      "learning_rate": 5.2259215219976225e-06,
      "loss": 0.03801336884498596,
      "loss_step_1": 0.0394,
      "loss_step_2": 0.0395,
      "loss_step_3": 0.0398,
      "step": 880
    },
    {
      "entropy_step_1": 0.1529,
      "entropy_step_2": 0.153,
      "entropy_step_3": 0.1692,
      "epoch": 0.02647155051902085,
      "grad_norm": 9.8125,
      "learning_rate": 5.28537455410226e-06,
      "loss": 0.04736810028553009,
      "loss_step_1": 0.0465,
      "loss_step_2": 0.0465,
      "loss_step_3": 0.0523,
      "step": 890
    },
    {
      "entropy_step_1": 0.1464,
      "entropy_step_2": 0.1461,
      "entropy_step_3": 0.1516,
      "epoch": 0.02676898367091996,
      "grad_norm": 9.5625,
      "learning_rate": 5.344827586206896e-06,
      "loss": 0.05125091671943664,
      "loss_step_1": 0.0451,
      "loss_step_2": 0.0453,
      "loss_step_3": 0.0442,
      "step": 900
    },
    {
      "entropy_step_1": 0.1618,
      "entropy_step_2": 0.161,
      "entropy_step_3": 0.1617,
      "epoch": 0.027066416822819073,
      "grad_norm": 7.3125,
      "learning_rate": 5.404280618311535e-06,
      "loss": 0.043788307905197145,
      "loss_step_1": 0.0437,
      "loss_step_2": 0.0435,
      "loss_step_3": 0.0414,
      "step": 910
    },
    {
      "entropy_step_1": 0.1502,
      "entropy_step_2": 0.1493,
      "entropy_step_3": 0.138,
      "epoch": 0.027363849974718184,
      "grad_norm": 10.75,
      "learning_rate": 5.463733650416172e-06,
      "loss": 0.05158089399337769,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0395,
      "step": 920
    },
    {
      "entropy_step_1": 0.1398,
      "entropy_step_2": 0.1401,
      "entropy_step_3": 0.1323,
      "epoch": 0.02766128312661729,
      "grad_norm": 6.6875,
      "learning_rate": 5.523186682520809e-06,
      "loss": 0.04863620400428772,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0424,
      "step": 930
    },
    {
      "entropy_step_1": 0.1474,
      "entropy_step_2": 0.1476,
      "entropy_step_3": 0.1619,
      "epoch": 0.027958716278516402,
      "grad_norm": 4.5625,
      "learning_rate": 5.582639714625446e-06,
      "loss": 0.04505519270896911,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0461,
      "loss_step_3": 0.0624,
      "step": 940
    },
    {
      "entropy_step_1": 0.1642,
      "entropy_step_2": 0.1648,
      "entropy_step_3": 0.1885,
      "epoch": 0.028256149430415513,
      "grad_norm": 4.40625,
      "learning_rate": 5.642092746730083e-06,
      "loss": 0.047565269470214847,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0559,
      "step": 950
    },
    {
      "entropy_step_1": 0.1542,
      "entropy_step_2": 0.154,
      "entropy_step_3": 0.1534,
      "epoch": 0.028553582582314625,
      "grad_norm": 5.59375,
      "learning_rate": 5.701545778834721e-06,
      "loss": 0.053848940134048465,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.05,
      "step": 960
    },
    {
      "entropy_step_1": 0.1613,
      "entropy_step_2": 0.1599,
      "entropy_step_3": 0.1649,
      "epoch": 0.028851015734213736,
      "grad_norm": 6.34375,
      "learning_rate": 5.760998810939359e-06,
      "loss": 0.047441419959068296,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0479,
      "loss_step_3": 0.0556,
      "step": 970
    },
    {
      "entropy_step_1": 0.1582,
      "entropy_step_2": 0.1575,
      "entropy_step_3": 0.185,
      "epoch": 0.029148448886112847,
      "grad_norm": 10.375,
      "learning_rate": 5.820451843043996e-06,
      "loss": 0.051739168167114255,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0702,
      "step": 980
    },
    {
      "entropy_step_1": 0.1545,
      "entropy_step_2": 0.1537,
      "entropy_step_3": 0.1488,
      "epoch": 0.029445882038011958,
      "grad_norm": 11.0625,
      "learning_rate": 5.879904875148633e-06,
      "loss": 0.04818410277366638,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0521,
      "step": 990
    },
    {
      "entropy_step_1": 0.1667,
      "entropy_step_2": 0.1667,
      "entropy_step_3": 0.1615,
      "epoch": 0.02974331518991107,
      "grad_norm": 6.46875,
      "learning_rate": 5.939357907253271e-06,
      "loss": 0.05047672390937805,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.051,
      "step": 1000
    },
    {
      "epoch": 0.02974331518991107,
      "eval_accuracy": 0.978,
      "eval_f1": 0.977997799779978,
      "eval_loss": 0.042056433856487274,
      "eval_precision": 0.9781912765106042,
      "eval_recall": 0.978,
      "eval_runtime": 19.9813,
      "eval_samples_per_second": 25.023,
      "eval_steps_per_second": 3.153,
      "step": 1000
    },
    {
      "entropy_step_1": 0.1546,
      "entropy_step_2": 0.1537,
      "entropy_step_3": 0.1651,
      "epoch": 0.030040748341810176,
      "grad_norm": 19.875,
      "learning_rate": 5.998810939357908e-06,
      "loss": 0.04759654998779297,
      "loss_step_1": 0.0494,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0491,
      "step": 1010
    },
    {
      "entropy_step_1": 0.1457,
      "entropy_step_2": 0.1457,
      "entropy_step_3": 0.15,
      "epoch": 0.030338181493709288,
      "grad_norm": 5.90625,
      "learning_rate": 6.058263971462544e-06,
      "loss": 0.04542114436626434,
      "loss_step_1": 0.0461,
      "loss_step_2": 0.0457,
      "loss_step_3": 0.0432,
      "step": 1020
    },
    {
      "entropy_step_1": 0.1377,
      "entropy_step_2": 0.1372,
      "entropy_step_3": 0.1368,
      "epoch": 0.0306356146456084,
      "grad_norm": 11.75,
      "learning_rate": 6.117717003567183e-06,
      "loss": 0.05444192886352539,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0581,
      "step": 1030
    },
    {
      "entropy_step_1": 0.1448,
      "entropy_step_2": 0.1449,
      "entropy_step_3": 0.1537,
      "epoch": 0.03093304779750751,
      "grad_norm": 13.1875,
      "learning_rate": 6.17717003567182e-06,
      "loss": 0.05507417917251587,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0605,
      "step": 1040
    },
    {
      "entropy_step_1": 0.1604,
      "entropy_step_2": 0.16,
      "entropy_step_3": 0.1395,
      "epoch": 0.03123048094940662,
      "grad_norm": 7.625,
      "learning_rate": 6.236623067776457e-06,
      "loss": 0.05474355220794678,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0439,
      "step": 1050
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1749,
      "entropy_step_3": 0.1811,
      "epoch": 0.03152791410130573,
      "grad_norm": 9.8125,
      "learning_rate": 6.296076099881094e-06,
      "loss": 0.052248412370681764,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0601,
      "step": 1060
    },
    {
      "entropy_step_1": 0.1735,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1631,
      "epoch": 0.03182534725320484,
      "grad_norm": 7.5,
      "learning_rate": 6.3555291319857315e-06,
      "loss": 0.05373838543891907,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0429,
      "step": 1070
    },
    {
      "entropy_step_1": 0.1384,
      "entropy_step_2": 0.1376,
      "entropy_step_3": 0.1361,
      "epoch": 0.032122780405103954,
      "grad_norm": 13.3125,
      "learning_rate": 6.414982164090369e-06,
      "loss": 0.04695771038532257,
      "loss_step_1": 0.0414,
      "loss_step_2": 0.0417,
      "loss_step_3": 0.0379,
      "step": 1080
    },
    {
      "entropy_step_1": 0.1573,
      "entropy_step_2": 0.1575,
      "entropy_step_3": 0.132,
      "epoch": 0.032420213557003065,
      "grad_norm": 8.25,
      "learning_rate": 6.474435196195007e-06,
      "loss": 0.04581833779811859,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.04,
      "step": 1090
    },
    {
      "entropy_step_1": 0.14,
      "entropy_step_2": 0.1401,
      "entropy_step_3": 0.138,
      "epoch": 0.032717646708902176,
      "grad_norm": 7.59375,
      "learning_rate": 6.533888228299644e-06,
      "loss": 0.05054736137390137,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0579,
      "step": 1100
    },
    {
      "entropy_step_1": 0.1602,
      "entropy_step_2": 0.1592,
      "entropy_step_3": 0.1409,
      "epoch": 0.03301507986080129,
      "grad_norm": 4.125,
      "learning_rate": 6.593341260404281e-06,
      "loss": 0.061014586687088014,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0403,
      "step": 1110
    },
    {
      "entropy_step_1": 0.1541,
      "entropy_step_2": 0.1541,
      "entropy_step_3": 0.1625,
      "epoch": 0.0333125130127004,
      "grad_norm": 9.5,
      "learning_rate": 6.652794292508919e-06,
      "loss": 0.046003499627113344,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.0485,
      "loss_step_3": 0.0476,
      "step": 1120
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.1843,
      "epoch": 0.03360994616459951,
      "grad_norm": 3.609375,
      "learning_rate": 6.712247324613556e-06,
      "loss": 0.045214283466339114,
      "loss_step_1": 0.0452,
      "loss_step_2": 0.0448,
      "loss_step_3": 0.0434,
      "step": 1130
    },
    {
      "entropy_step_1": 0.1633,
      "entropy_step_2": 0.1627,
      "entropy_step_3": 0.1681,
      "epoch": 0.03390737931649862,
      "grad_norm": 7.625,
      "learning_rate": 6.771700356718192e-06,
      "loss": 0.048197215795516966,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0459,
      "loss_step_3": 0.051,
      "step": 1140
    },
    {
      "entropy_step_1": 0.1351,
      "entropy_step_2": 0.1352,
      "entropy_step_3": 0.1384,
      "epoch": 0.034204812468397724,
      "grad_norm": 8.8125,
      "learning_rate": 6.831153388822831e-06,
      "loss": 0.0490300714969635,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0587,
      "step": 1150
    },
    {
      "entropy_step_1": 0.15,
      "entropy_step_2": 0.1502,
      "entropy_step_3": 0.1546,
      "epoch": 0.034502245620296836,
      "grad_norm": 9.0,
      "learning_rate": 6.890606420927468e-06,
      "loss": 0.04294655025005341,
      "loss_step_1": 0.0428,
      "loss_step_2": 0.0429,
      "loss_step_3": 0.0487,
      "step": 1160
    },
    {
      "entropy_step_1": 0.1552,
      "entropy_step_2": 0.1552,
      "entropy_step_3": 0.1413,
      "epoch": 0.03479967877219595,
      "grad_norm": 11.875,
      "learning_rate": 6.950059453032105e-06,
      "loss": 0.05439215898513794,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0419,
      "step": 1170
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1742,
      "entropy_step_3": 0.1728,
      "epoch": 0.03509711192409506,
      "grad_norm": 10.8125,
      "learning_rate": 7.009512485136742e-06,
      "loss": 0.056864798069000244,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0611,
      "step": 1180
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.1935,
      "epoch": 0.03539454507599417,
      "grad_norm": 5.3125,
      "learning_rate": 7.0689655172413796e-06,
      "loss": 0.053636187314987184,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0547,
      "step": 1190
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1722,
      "entropy_step_3": 0.1577,
      "epoch": 0.03569197822789328,
      "grad_norm": 8.0,
      "learning_rate": 7.128418549346017e-06,
      "loss": 0.050388699769973753,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.045,
      "step": 1200
    },
    {
      "entropy_step_1": 0.1382,
      "entropy_step_2": 0.1384,
      "entropy_step_3": 0.1302,
      "epoch": 0.03598941137979239,
      "grad_norm": 12.0,
      "learning_rate": 7.187871581450655e-06,
      "loss": 0.047644391655921936,
      "loss_step_1": 0.0442,
      "loss_step_2": 0.044,
      "loss_step_3": 0.0375,
      "step": 1210
    },
    {
      "entropy_step_1": 0.1523,
      "entropy_step_2": 0.1528,
      "entropy_step_3": 0.1388,
      "epoch": 0.0362868445316915,
      "grad_norm": 9.25,
      "learning_rate": 7.247324613555292e-06,
      "loss": 0.049605792760849,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0432,
      "step": 1220
    },
    {
      "entropy_step_1": 0.1211,
      "entropy_step_2": 0.1219,
      "entropy_step_3": 0.1116,
      "epoch": 0.03658427768359061,
      "grad_norm": 6.8125,
      "learning_rate": 7.3067776456599295e-06,
      "loss": 0.045650610327720643,
      "loss_step_1": 0.0462,
      "loss_step_2": 0.0459,
      "loss_step_3": 0.0355,
      "step": 1230
    },
    {
      "entropy_step_1": 0.1516,
      "entropy_step_2": 0.1517,
      "entropy_step_3": 0.1638,
      "epoch": 0.036881710835489724,
      "grad_norm": 6.96875,
      "learning_rate": 7.366230677764567e-06,
      "loss": 0.0467339426279068,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0497,
      "loss_step_3": 0.057,
      "step": 1240
    },
    {
      "entropy_step_1": 0.1505,
      "entropy_step_2": 0.1516,
      "entropy_step_3": 0.1588,
      "epoch": 0.037179143987388835,
      "grad_norm": 4.40625,
      "learning_rate": 7.425683709869204e-06,
      "loss": 0.04607460498809814,
      "loss_step_1": 0.0455,
      "loss_step_2": 0.0459,
      "loss_step_3": 0.0507,
      "step": 1250
    },
    {
      "entropy_step_1": 0.1461,
      "entropy_step_2": 0.1451,
      "entropy_step_3": 0.1543,
      "epoch": 0.037476577139287946,
      "grad_norm": 9.6875,
      "learning_rate": 7.4851367419738404e-06,
      "loss": 0.04839504063129425,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0589,
      "step": 1260
    },
    {
      "entropy_step_1": 0.174,
      "entropy_step_2": 0.1744,
      "entropy_step_3": 0.1693,
      "epoch": 0.03777401029118706,
      "grad_norm": 8.5,
      "learning_rate": 7.544589774078479e-06,
      "loss": 0.05641516447067261,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0531,
      "step": 1270
    },
    {
      "entropy_step_1": 0.1375,
      "entropy_step_2": 0.138,
      "entropy_step_3": 0.1551,
      "epoch": 0.03807144344308617,
      "grad_norm": 5.84375,
      "learning_rate": 7.604042806183116e-06,
      "loss": 0.0454856276512146,
      "loss_step_1": 0.0458,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.0527,
      "step": 1280
    },
    {
      "entropy_step_1": 0.1581,
      "entropy_step_2": 0.1585,
      "entropy_step_3": 0.1633,
      "epoch": 0.03836887659498528,
      "grad_norm": 11.9375,
      "learning_rate": 7.663495838287754e-06,
      "loss": 0.052429449558258054,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0465,
      "step": 1290
    },
    {
      "entropy_step_1": 0.1682,
      "entropy_step_2": 0.1682,
      "entropy_step_3": 0.1719,
      "epoch": 0.03866630974688439,
      "grad_norm": 4.25,
      "learning_rate": 7.722948870392391e-06,
      "loss": 0.05473124980926514,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0581,
      "step": 1300
    },
    {
      "entropy_step_1": 0.1579,
      "entropy_step_2": 0.1585,
      "entropy_step_3": 0.1689,
      "epoch": 0.0389637428987835,
      "grad_norm": 8.875,
      "learning_rate": 7.782401902497028e-06,
      "loss": 0.053840750455856325,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.054,
      "loss_step_3": 0.052,
      "step": 1310
    },
    {
      "entropy_step_1": 0.1494,
      "entropy_step_2": 0.1497,
      "entropy_step_3": 0.1502,
      "epoch": 0.039261176050682606,
      "grad_norm": 4.28125,
      "learning_rate": 7.841854934601664e-06,
      "loss": 0.04606353044509888,
      "loss_step_1": 0.0444,
      "loss_step_2": 0.0447,
      "loss_step_3": 0.0471,
      "step": 1320
    },
    {
      "entropy_step_1": 0.1315,
      "entropy_step_2": 0.1315,
      "entropy_step_3": 0.1423,
      "epoch": 0.03955860920258172,
      "grad_norm": 5.21875,
      "learning_rate": 7.901307966706303e-06,
      "loss": 0.041921356320381166,
      "loss_step_1": 0.0413,
      "loss_step_2": 0.041,
      "loss_step_3": 0.0404,
      "step": 1330
    },
    {
      "entropy_step_1": 0.1166,
      "entropy_step_2": 0.1157,
      "entropy_step_3": 0.1218,
      "epoch": 0.03985604235448083,
      "grad_norm": 27.875,
      "learning_rate": 7.96076099881094e-06,
      "loss": 0.04634135365486145,
      "loss_step_1": 0.0451,
      "loss_step_2": 0.045,
      "loss_step_3": 0.0481,
      "step": 1340
    },
    {
      "entropy_step_1": 0.1414,
      "entropy_step_2": 0.1415,
      "entropy_step_3": 0.1114,
      "epoch": 0.04015347550637994,
      "grad_norm": 6.0,
      "learning_rate": 8.020214030915578e-06,
      "loss": 0.04665260910987854,
      "loss_step_1": 0.0383,
      "loss_step_2": 0.0382,
      "loss_step_3": 0.0298,
      "step": 1350
    },
    {
      "entropy_step_1": 0.1489,
      "entropy_step_2": 0.1477,
      "entropy_step_3": 0.1583,
      "epoch": 0.04045090865827905,
      "grad_norm": 4.875,
      "learning_rate": 8.079667063020215e-06,
      "loss": 0.05266072154045105,
      "loss_step_1": 0.0483,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0574,
      "step": 1360
    },
    {
      "entropy_step_1": 0.1646,
      "entropy_step_2": 0.1648,
      "entropy_step_3": 0.1738,
      "epoch": 0.04074834181017816,
      "grad_norm": 10.0,
      "learning_rate": 8.139120095124852e-06,
      "loss": 0.04945930540561676,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0606,
      "step": 1370
    },
    {
      "entropy_step_1": 0.1587,
      "entropy_step_2": 0.1592,
      "entropy_step_3": 0.1547,
      "epoch": 0.04104577496207727,
      "grad_norm": 8.8125,
      "learning_rate": 8.19857312722949e-06,
      "loss": 0.04905755519866943,
      "loss_step_1": 0.0483,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0499,
      "step": 1380
    },
    {
      "entropy_step_1": 0.1453,
      "entropy_step_2": 0.1461,
      "entropy_step_3": 0.1711,
      "epoch": 0.04134320811397638,
      "grad_norm": 9.0,
      "learning_rate": 8.258026159334127e-06,
      "loss": 0.04061069786548614,
      "loss_step_1": 0.0419,
      "loss_step_2": 0.042,
      "loss_step_3": 0.055,
      "step": 1390
    },
    {
      "entropy_step_1": 0.1441,
      "entropy_step_2": 0.1439,
      "entropy_step_3": 0.1345,
      "epoch": 0.041640641265875494,
      "grad_norm": 25.75,
      "learning_rate": 8.317479191438764e-06,
      "loss": 0.05381606817245484,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0504,
      "step": 1400
    },
    {
      "entropy_step_1": 0.1265,
      "entropy_step_2": 0.1271,
      "entropy_step_3": 0.1228,
      "epoch": 0.041938074417774605,
      "grad_norm": 10.125,
      "learning_rate": 8.376932223543401e-06,
      "loss": 0.04885972142219543,
      "loss_step_1": 0.0456,
      "loss_step_2": 0.045,
      "loss_step_3": 0.0506,
      "step": 1410
    },
    {
      "entropy_step_1": 0.1541,
      "entropy_step_2": 0.1541,
      "entropy_step_3": 0.1599,
      "epoch": 0.042235507569673716,
      "grad_norm": 5.75,
      "learning_rate": 8.436385255648038e-06,
      "loss": 0.05052608251571655,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.047,
      "loss_step_3": 0.0435,
      "step": 1420
    },
    {
      "entropy_step_1": 0.1492,
      "entropy_step_2": 0.1471,
      "entropy_step_3": 0.1502,
      "epoch": 0.04253294072157283,
      "grad_norm": 9.0625,
      "learning_rate": 8.495838287752676e-06,
      "loss": 0.05112156271934509,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0392,
      "step": 1430
    },
    {
      "entropy_step_1": 0.1398,
      "entropy_step_2": 0.1397,
      "entropy_step_3": 0.1268,
      "epoch": 0.04283037387347194,
      "grad_norm": 22.625,
      "learning_rate": 8.555291319857313e-06,
      "loss": 0.047648513317108156,
      "loss_step_1": 0.0418,
      "loss_step_2": 0.0418,
      "loss_step_3": 0.0354,
      "step": 1440
    },
    {
      "entropy_step_1": 0.1361,
      "entropy_step_2": 0.1348,
      "entropy_step_3": 0.1454,
      "epoch": 0.04312780702537105,
      "grad_norm": 7.0625,
      "learning_rate": 8.614744351961952e-06,
      "loss": 0.0374517410993576,
      "loss_step_1": 0.0378,
      "loss_step_2": 0.0375,
      "loss_step_3": 0.0373,
      "step": 1450
    },
    {
      "entropy_step_1": 0.1605,
      "entropy_step_2": 0.1604,
      "entropy_step_3": 0.1639,
      "epoch": 0.04342524017727016,
      "grad_norm": 3.875,
      "learning_rate": 8.674197384066588e-06,
      "loss": 0.05011860728263855,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0542,
      "step": 1460
    },
    {
      "entropy_step_1": 0.1416,
      "entropy_step_2": 0.142,
      "entropy_step_3": 0.1389,
      "epoch": 0.04372267332916927,
      "grad_norm": 13.75,
      "learning_rate": 8.733650416171225e-06,
      "loss": 0.039521968364715575,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.0439,
      "loss_step_3": 0.0481,
      "step": 1470
    },
    {
      "entropy_step_1": 0.1185,
      "entropy_step_2": 0.1183,
      "entropy_step_3": 0.1209,
      "epoch": 0.04402010648106838,
      "grad_norm": 4.75,
      "learning_rate": 8.793103448275862e-06,
      "loss": 0.041888114809989926,
      "loss_step_1": 0.042,
      "loss_step_2": 0.0418,
      "loss_step_3": 0.044,
      "step": 1480
    },
    {
      "entropy_step_1": 0.1283,
      "entropy_step_2": 0.1277,
      "entropy_step_3": 0.1299,
      "epoch": 0.044317539632967494,
      "grad_norm": 11.5625,
      "learning_rate": 8.8525564803805e-06,
      "loss": 0.04854902625083923,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0479,
      "step": 1490
    },
    {
      "entropy_step_1": 0.1828,
      "entropy_step_2": 0.1823,
      "entropy_step_3": 0.1614,
      "epoch": 0.0446149727848666,
      "grad_norm": 8.3125,
      "learning_rate": 8.912009512485138e-06,
      "loss": 0.05644582509994507,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0587,
      "step": 1500
    },
    {
      "entropy_step_1": 0.1541,
      "entropy_step_2": 0.1533,
      "entropy_step_3": 0.1482,
      "epoch": 0.04491240593676571,
      "grad_norm": 6.6875,
      "learning_rate": 8.971462544589776e-06,
      "loss": 0.042718881368637086,
      "loss_step_1": 0.042,
      "loss_step_2": 0.0417,
      "loss_step_3": 0.0404,
      "step": 1510
    },
    {
      "entropy_step_1": 0.1639,
      "entropy_step_2": 0.1632,
      "entropy_step_3": 0.1521,
      "epoch": 0.04520983908866482,
      "grad_norm": 8.875,
      "learning_rate": 9.030915576694413e-06,
      "loss": 0.05260757207870483,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0559,
      "step": 1520
    },
    {
      "entropy_step_1": 0.1345,
      "entropy_step_2": 0.1352,
      "entropy_step_3": 0.1096,
      "epoch": 0.04550727224056393,
      "grad_norm": 11.75,
      "learning_rate": 9.09036860879905e-06,
      "loss": 0.046257045865058896,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0494,
      "step": 1530
    },
    {
      "entropy_step_1": 0.1373,
      "entropy_step_2": 0.1375,
      "entropy_step_3": 0.1398,
      "epoch": 0.04580470539246304,
      "grad_norm": 32.0,
      "learning_rate": 9.149821640903687e-06,
      "loss": 0.052933961153030396,
      "loss_step_1": 0.0446,
      "loss_step_2": 0.0447,
      "loss_step_3": 0.0504,
      "step": 1540
    },
    {
      "entropy_step_1": 0.1479,
      "entropy_step_2": 0.149,
      "entropy_step_3": 0.1588,
      "epoch": 0.04610213854436215,
      "grad_norm": 10.0625,
      "learning_rate": 9.209274673008323e-06,
      "loss": 0.052151286602020265,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0487,
      "loss_step_3": 0.0567,
      "step": 1550
    },
    {
      "entropy_step_1": 0.1376,
      "entropy_step_2": 0.1384,
      "entropy_step_3": 0.1427,
      "epoch": 0.046399571696261264,
      "grad_norm": 23.625,
      "learning_rate": 9.268727705112962e-06,
      "loss": 0.04249287545681,
      "loss_step_1": 0.0407,
      "loss_step_2": 0.0404,
      "loss_step_3": 0.0413,
      "step": 1560
    },
    {
      "entropy_step_1": 0.1459,
      "entropy_step_2": 0.1462,
      "entropy_step_3": 0.1541,
      "epoch": 0.046697004848160376,
      "grad_norm": 12.0625,
      "learning_rate": 9.3281807372176e-06,
      "loss": 0.04203498959541321,
      "loss_step_1": 0.0447,
      "loss_step_2": 0.0447,
      "loss_step_3": 0.0513,
      "step": 1570
    },
    {
      "entropy_step_1": 0.1407,
      "entropy_step_2": 0.1418,
      "entropy_step_3": 0.1341,
      "epoch": 0.04699443800005949,
      "grad_norm": 7.15625,
      "learning_rate": 9.387633769322236e-06,
      "loss": 0.043540766835212706,
      "loss_step_1": 0.0415,
      "loss_step_2": 0.0416,
      "loss_step_3": 0.0384,
      "step": 1580
    },
    {
      "entropy_step_1": 0.1248,
      "entropy_step_2": 0.1247,
      "entropy_step_3": 0.1256,
      "epoch": 0.0472918711519586,
      "grad_norm": 4.40625,
      "learning_rate": 9.447086801426874e-06,
      "loss": 0.05017787218093872,
      "loss_step_1": 0.0365,
      "loss_step_2": 0.0365,
      "loss_step_3": 0.031,
      "step": 1590
    },
    {
      "entropy_step_1": 0.1409,
      "entropy_step_2": 0.1405,
      "entropy_step_3": 0.1482,
      "epoch": 0.04758930430385771,
      "grad_norm": 5.96875,
      "learning_rate": 9.506539833531511e-06,
      "loss": 0.04474652707576752,
      "loss_step_1": 0.0453,
      "loss_step_2": 0.0451,
      "loss_step_3": 0.0462,
      "step": 1600
    },
    {
      "entropy_step_1": 0.1578,
      "entropy_step_2": 0.1578,
      "entropy_step_3": 0.153,
      "epoch": 0.04788673745575682,
      "grad_norm": 6.28125,
      "learning_rate": 9.565992865636148e-06,
      "loss": 0.04405922293663025,
      "loss_step_1": 0.0404,
      "loss_step_2": 0.0406,
      "loss_step_3": 0.0406,
      "step": 1610
    },
    {
      "entropy_step_1": 0.1539,
      "entropy_step_2": 0.1542,
      "entropy_step_3": 0.142,
      "epoch": 0.04818417060765593,
      "grad_norm": 6.96875,
      "learning_rate": 9.625445897740786e-06,
      "loss": 0.05125764012336731,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.065,
      "step": 1620
    },
    {
      "entropy_step_1": 0.1664,
      "entropy_step_2": 0.1667,
      "entropy_step_3": 0.1553,
      "epoch": 0.04848160375955504,
      "grad_norm": 4.03125,
      "learning_rate": 9.684898929845423e-06,
      "loss": 0.051525342464447024,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0509,
      "step": 1630
    },
    {
      "entropy_step_1": 0.1458,
      "entropy_step_2": 0.1454,
      "entropy_step_3": 0.1441,
      "epoch": 0.04877903691145415,
      "grad_norm": 6.09375,
      "learning_rate": 9.74435196195006e-06,
      "loss": 0.04568221271038055,
      "loss_step_1": 0.0447,
      "loss_step_2": 0.0443,
      "loss_step_3": 0.0507,
      "step": 1640
    },
    {
      "entropy_step_1": 0.1716,
      "entropy_step_2": 0.1712,
      "entropy_step_3": 0.1914,
      "epoch": 0.049076470063353264,
      "grad_norm": 11.75,
      "learning_rate": 9.803804994054697e-06,
      "loss": 0.04871885776519776,
      "loss_step_1": 0.0497,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0546,
      "step": 1650
    },
    {
      "entropy_step_1": 0.1423,
      "entropy_step_2": 0.1421,
      "entropy_step_3": 0.1587,
      "epoch": 0.049373903215252375,
      "grad_norm": 3.421875,
      "learning_rate": 9.863258026159335e-06,
      "loss": 0.04711183607578277,
      "loss_step_1": 0.0429,
      "loss_step_2": 0.0432,
      "loss_step_3": 0.0449,
      "step": 1660
    },
    {
      "entropy_step_1": 0.1397,
      "entropy_step_2": 0.1385,
      "entropy_step_3": 0.1276,
      "epoch": 0.04967133636715148,
      "grad_norm": 4.03125,
      "learning_rate": 9.922711058263972e-06,
      "loss": 0.047143077850341795,
      "loss_step_1": 0.0448,
      "loss_step_2": 0.045,
      "loss_step_3": 0.0426,
      "step": 1670
    },
    {
      "entropy_step_1": 0.1341,
      "entropy_step_2": 0.1341,
      "entropy_step_3": 0.1338,
      "epoch": 0.04996876951905059,
      "grad_norm": 7.0625,
      "learning_rate": 9.98216409036861e-06,
      "loss": 0.044937309622764585,
      "loss_step_1": 0.0431,
      "loss_step_2": 0.0429,
      "loss_step_3": 0.0465,
      "step": 1680
    },
    {
      "entropy_step_1": 0.1584,
      "entropy_step_2": 0.1562,
      "entropy_step_3": 0.1543,
      "epoch": 0.0502662026709497,
      "grad_norm": 6.53125,
      "learning_rate": 9.999998814795783e-06,
      "loss": 0.044633609056472776,
      "loss_step_1": 0.0445,
      "loss_step_2": 0.0446,
      "loss_step_3": 0.0411,
      "step": 1690
    },
    {
      "entropy_step_1": 0.1366,
      "entropy_step_2": 0.1352,
      "entropy_step_3": 0.1343,
      "epoch": 0.05056363582284881,
      "grad_norm": 15.0,
      "learning_rate": 9.999993009715256e-06,
      "loss": 0.04364912509918213,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0497,
      "loss_step_3": 0.0396,
      "step": 1700
    },
    {
      "entropy_step_1": 0.1451,
      "entropy_step_2": 0.1449,
      "entropy_step_3": 0.1467,
      "epoch": 0.050861068974747924,
      "grad_norm": 12.375,
      "learning_rate": 9.999982367073456e-06,
      "loss": 0.04366686344146729,
      "loss_step_1": 0.0457,
      "loss_step_2": 0.0455,
      "loss_step_3": 0.0477,
      "step": 1710
    },
    {
      "entropy_step_1": 0.1574,
      "entropy_step_2": 0.1563,
      "entropy_step_3": 0.1592,
      "epoch": 0.051158502126647035,
      "grad_norm": 3.6875,
      "learning_rate": 9.999966886880682e-06,
      "loss": 0.045107656717300416,
      "loss_step_1": 0.0472,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0452,
      "step": 1720
    },
    {
      "entropy_step_1": 0.1418,
      "entropy_step_2": 0.1409,
      "entropy_step_3": 0.1379,
      "epoch": 0.051455935278546146,
      "grad_norm": 21.125,
      "learning_rate": 9.99994656915191e-06,
      "loss": 0.047960993647575376,
      "loss_step_1": 0.044,
      "loss_step_2": 0.044,
      "loss_step_3": 0.0413,
      "step": 1730
    },
    {
      "entropy_step_1": 0.1399,
      "entropy_step_2": 0.1392,
      "entropy_step_3": 0.1474,
      "epoch": 0.05175336843044526,
      "grad_norm": 6.28125,
      "learning_rate": 9.9999214139068e-06,
      "loss": 0.052089530229568484,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0522,
      "loss_step_3": 0.0592,
      "step": 1740
    },
    {
      "entropy_step_1": 0.1655,
      "entropy_step_2": 0.165,
      "entropy_step_3": 0.1658,
      "epoch": 0.05205080158234437,
      "grad_norm": 10.75,
      "learning_rate": 9.999891421169685e-06,
      "loss": 0.05429313778877258,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0596,
      "step": 1750
    },
    {
      "entropy_step_1": 0.1488,
      "entropy_step_2": 0.148,
      "entropy_step_3": 0.1526,
      "epoch": 0.05234823473424348,
      "grad_norm": 9.375,
      "learning_rate": 9.999856590969588e-06,
      "loss": 0.03836148977279663,
      "loss_step_1": 0.0384,
      "loss_step_2": 0.038,
      "loss_step_3": 0.0359,
      "step": 1760
    },
    {
      "entropy_step_1": 0.1389,
      "entropy_step_2": 0.137,
      "entropy_step_3": 0.1597,
      "epoch": 0.05264566788614259,
      "grad_norm": 25.625,
      "learning_rate": 9.999816923340208e-06,
      "loss": 0.04749746024608612,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0654,
      "step": 1770
    },
    {
      "entropy_step_1": 0.1696,
      "entropy_step_2": 0.1687,
      "entropy_step_3": 0.1698,
      "epoch": 0.0529431010380417,
      "grad_norm": 2.59375,
      "learning_rate": 9.99977241831992e-06,
      "loss": 0.04730409681797028,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0489,
      "step": 1780
    },
    {
      "entropy_step_1": 0.1529,
      "entropy_step_2": 0.1515,
      "entropy_step_3": 0.1431,
      "epoch": 0.05324053418994081,
      "grad_norm": 12.0,
      "learning_rate": 9.999723075951787e-06,
      "loss": 0.049559152126312254,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0481,
      "step": 1790
    },
    {
      "entropy_step_1": 0.1368,
      "entropy_step_2": 0.1361,
      "entropy_step_3": 0.1482,
      "epoch": 0.05353796734183992,
      "grad_norm": 4.28125,
      "learning_rate": 9.999668896283547e-06,
      "loss": 0.044942346215248105,
      "loss_step_1": 0.0437,
      "loss_step_2": 0.0435,
      "loss_step_3": 0.0468,
      "step": 1800
    },
    {
      "entropy_step_1": 0.1297,
      "entropy_step_2": 0.1292,
      "entropy_step_3": 0.121,
      "epoch": 0.053835400493739034,
      "grad_norm": 26.625,
      "learning_rate": 9.999609879367618e-06,
      "loss": 0.043008828163146974,
      "loss_step_1": 0.0451,
      "loss_step_2": 0.0449,
      "loss_step_3": 0.0403,
      "step": 1810
    },
    {
      "entropy_step_1": 0.154,
      "entropy_step_2": 0.1532,
      "entropy_step_3": 0.1662,
      "epoch": 0.054132833645638145,
      "grad_norm": 7.6875,
      "learning_rate": 9.999546025261101e-06,
      "loss": 0.04807855486869812,
      "loss_step_1": 0.0434,
      "loss_step_2": 0.0437,
      "loss_step_3": 0.0448,
      "step": 1820
    },
    {
      "entropy_step_1": 0.1507,
      "entropy_step_2": 0.1494,
      "entropy_step_3": 0.1471,
      "epoch": 0.054430266797537256,
      "grad_norm": 6.46875,
      "learning_rate": 9.999477334025778e-06,
      "loss": 0.05159052610397339,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0526,
      "step": 1830
    },
    {
      "entropy_step_1": 0.1433,
      "entropy_step_2": 0.1418,
      "entropy_step_3": 0.1339,
      "epoch": 0.05472769994943637,
      "grad_norm": 4.125,
      "learning_rate": 9.999403805728107e-06,
      "loss": 0.04667559266090393,
      "loss_step_1": 0.0451,
      "loss_step_2": 0.0448,
      "loss_step_3": 0.0385,
      "step": 1840
    },
    {
      "entropy_step_1": 0.1409,
      "entropy_step_2": 0.1392,
      "entropy_step_3": 0.145,
      "epoch": 0.05502513310133547,
      "grad_norm": 5.59375,
      "learning_rate": 9.999325440439226e-06,
      "loss": 0.0442452609539032,
      "loss_step_1": 0.0428,
      "loss_step_2": 0.0424,
      "loss_step_3": 0.0471,
      "step": 1850
    },
    {
      "entropy_step_1": 0.1575,
      "entropy_step_2": 0.1554,
      "entropy_step_3": 0.1493,
      "epoch": 0.05532256625323458,
      "grad_norm": 10.75,
      "learning_rate": 9.999242238234954e-06,
      "loss": 0.04568090736865997,
      "loss_step_1": 0.047,
      "loss_step_2": 0.047,
      "loss_step_3": 0.0399,
      "step": 1860
    },
    {
      "entropy_step_1": 0.1537,
      "entropy_step_2": 0.1529,
      "entropy_step_3": 0.1484,
      "epoch": 0.055619999405133694,
      "grad_norm": 7.59375,
      "learning_rate": 9.999154199195795e-06,
      "loss": 0.05228644609451294,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0406,
      "step": 1870
    },
    {
      "entropy_step_1": 0.1462,
      "entropy_step_2": 0.1469,
      "entropy_step_3": 0.1566,
      "epoch": 0.055917432557032805,
      "grad_norm": 14.0625,
      "learning_rate": 9.999061323406922e-06,
      "loss": 0.05599297881126404,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.056,
      "step": 1880
    },
    {
      "entropy_step_1": 0.1599,
      "entropy_step_2": 0.159,
      "entropy_step_3": 0.156,
      "epoch": 0.056214865708931916,
      "grad_norm": 6.8125,
      "learning_rate": 9.9989636109582e-06,
      "loss": 0.05439034104347229,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0637,
      "step": 1890
    },
    {
      "entropy_step_1": 0.1346,
      "entropy_step_2": 0.1338,
      "entropy_step_3": 0.1393,
      "epoch": 0.05651229886083103,
      "grad_norm": 5.40625,
      "learning_rate": 9.99886106194416e-06,
      "loss": 0.04190440475940704,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.0439,
      "loss_step_3": 0.0532,
      "step": 1900
    },
    {
      "entropy_step_1": 0.1623,
      "entropy_step_2": 0.1625,
      "entropy_step_3": 0.1476,
      "epoch": 0.05680973201273014,
      "grad_norm": 6.84375,
      "learning_rate": 9.998753676464025e-06,
      "loss": 0.058756327629089354,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0639,
      "step": 1910
    },
    {
      "entropy_step_1": 0.1552,
      "entropy_step_2": 0.1548,
      "entropy_step_3": 0.1665,
      "epoch": 0.05710716516462925,
      "grad_norm": 10.375,
      "learning_rate": 9.998641454621688e-06,
      "loss": 0.04401671886444092,
      "loss_step_1": 0.046,
      "loss_step_2": 0.0464,
      "loss_step_3": 0.0469,
      "step": 1920
    },
    {
      "entropy_step_1": 0.1465,
      "entropy_step_2": 0.1463,
      "entropy_step_3": 0.1548,
      "epoch": 0.05740459831652836,
      "grad_norm": 8.375,
      "learning_rate": 9.998524396525731e-06,
      "loss": 0.03999495208263397,
      "loss_step_1": 0.0404,
      "loss_step_2": 0.0401,
      "loss_step_3": 0.0458,
      "step": 1930
    },
    {
      "entropy_step_1": 0.1245,
      "entropy_step_2": 0.1238,
      "entropy_step_3": 0.1187,
      "epoch": 0.05770203146842747,
      "grad_norm": 8.8125,
      "learning_rate": 9.998402502289404e-06,
      "loss": 0.040645363926887515,
      "loss_step_1": 0.0392,
      "loss_step_2": 0.0393,
      "loss_step_3": 0.0463,
      "step": 1940
    },
    {
      "entropy_step_1": 0.1165,
      "entropy_step_2": 0.1161,
      "entropy_step_3": 0.109,
      "epoch": 0.05799946462032658,
      "grad_norm": 8.0625,
      "learning_rate": 9.998275772030643e-06,
      "loss": 0.0474224328994751,
      "loss_step_1": 0.0421,
      "loss_step_2": 0.0417,
      "loss_step_3": 0.0272,
      "step": 1950
    },
    {
      "entropy_step_1": 0.125,
      "entropy_step_2": 0.1246,
      "entropy_step_3": 0.1306,
      "epoch": 0.05829689777222569,
      "grad_norm": 6.0,
      "learning_rate": 9.99814420587206e-06,
      "loss": 0.045671552419662476,
      "loss_step_1": 0.044,
      "loss_step_2": 0.0434,
      "loss_step_3": 0.0494,
      "step": 1960
    },
    {
      "entropy_step_1": 0.1621,
      "entropy_step_2": 0.1603,
      "entropy_step_3": 0.1698,
      "epoch": 0.058594330924124804,
      "grad_norm": 12.125,
      "learning_rate": 9.99800780394095e-06,
      "loss": 0.052239859104156496,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0516,
      "step": 1970
    },
    {
      "entropy_step_1": 0.1849,
      "entropy_step_2": 0.184,
      "entropy_step_3": 0.2014,
      "epoch": 0.058891764076023916,
      "grad_norm": 11.9375,
      "learning_rate": 9.997866566369281e-06,
      "loss": 0.04442194700241089,
      "loss_step_1": 0.0461,
      "loss_step_2": 0.0462,
      "loss_step_3": 0.052,
      "step": 1980
    },
    {
      "entropy_step_1": 0.1798,
      "entropy_step_2": 0.1789,
      "entropy_step_3": 0.1963,
      "epoch": 0.05918919722792303,
      "grad_norm": 9.5625,
      "learning_rate": 9.997720493293703e-06,
      "loss": 0.05281427502632141,
      "loss_step_1": 0.0457,
      "loss_step_2": 0.0455,
      "loss_step_3": 0.0527,
      "step": 1990
    },
    {
      "entropy_step_1": 0.1658,
      "entropy_step_2": 0.1648,
      "entropy_step_3": 0.1573,
      "epoch": 0.05948663037982214,
      "grad_norm": 8.375,
      "learning_rate": 9.997569584855544e-06,
      "loss": 0.04867420494556427,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0514,
      "step": 2000
    },
    {
      "epoch": 0.05948663037982214,
      "eval_accuracy": 0.974,
      "eval_f1": 0.9739915732697393,
      "eval_loss": 0.03760460391640663,
      "eval_precision": 0.9746151011711176,
      "eval_recall": 0.974,
      "eval_runtime": 19.7471,
      "eval_samples_per_second": 25.32,
      "eval_steps_per_second": 3.19,
      "step": 2000
    },
    {
      "entropy_step_1": 0.1573,
      "entropy_step_2": 0.1571,
      "entropy_step_3": 0.1541,
      "epoch": 0.05978406353172125,
      "grad_norm": 11.875,
      "learning_rate": 9.997413841200811e-06,
      "loss": 0.04664312899112701,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0478,
      "step": 2010
    },
    {
      "entropy_step_1": 0.1557,
      "entropy_step_2": 0.1544,
      "entropy_step_3": 0.1683,
      "epoch": 0.06008149668362035,
      "grad_norm": 6.5,
      "learning_rate": 9.997253262480186e-06,
      "loss": 0.05134890079498291,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.059,
      "step": 2020
    },
    {
      "entropy_step_1": 0.1256,
      "entropy_step_2": 0.1257,
      "entropy_step_3": 0.1278,
      "epoch": 0.060378929835519464,
      "grad_norm": 4.65625,
      "learning_rate": 9.997087848849032e-06,
      "loss": 0.04458970129489899,
      "loss_step_1": 0.0444,
      "loss_step_2": 0.0439,
      "loss_step_3": 0.0374,
      "step": 2030
    },
    {
      "entropy_step_1": 0.1524,
      "entropy_step_2": 0.152,
      "entropy_step_3": 0.1541,
      "epoch": 0.060676362987418575,
      "grad_norm": 3.203125,
      "learning_rate": 9.996917600467388e-06,
      "loss": 0.04961158335208893,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0439,
      "step": 2040
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.1668,
      "epoch": 0.060973796139317686,
      "grad_norm": 5.4375,
      "learning_rate": 9.996742517499975e-06,
      "loss": 0.05399472713470459,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0482,
      "step": 2050
    },
    {
      "entropy_step_1": 0.1573,
      "entropy_step_2": 0.157,
      "entropy_step_3": 0.152,
      "epoch": 0.0612712292912168,
      "grad_norm": 10.375,
      "learning_rate": 9.996562600116185e-06,
      "loss": 0.04791352152824402,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0556,
      "step": 2060
    },
    {
      "entropy_step_1": 0.1446,
      "entropy_step_2": 0.1442,
      "entropy_step_3": 0.1381,
      "epoch": 0.06156866244311591,
      "grad_norm": 3.671875,
      "learning_rate": 9.99637784849009e-06,
      "loss": 0.045578169822692874,
      "loss_step_1": 0.0453,
      "loss_step_2": 0.0454,
      "loss_step_3": 0.0439,
      "step": 2070
    },
    {
      "entropy_step_1": 0.1684,
      "entropy_step_2": 0.1679,
      "entropy_step_3": 0.159,
      "epoch": 0.06186609559501502,
      "grad_norm": 9.125,
      "learning_rate": 9.996188262800442e-06,
      "loss": 0.05216138362884522,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0554,
      "step": 2080
    },
    {
      "entropy_step_1": 0.1487,
      "entropy_step_2": 0.1474,
      "entropy_step_3": 0.1578,
      "epoch": 0.06216352874691413,
      "grad_norm": 11.375,
      "learning_rate": 9.995993843230667e-06,
      "loss": 0.04256968796253204,
      "loss_step_1": 0.0417,
      "loss_step_2": 0.0412,
      "loss_step_3": 0.0452,
      "step": 2090
    },
    {
      "entropy_step_1": 0.1516,
      "entropy_step_2": 0.1518,
      "entropy_step_3": 0.1509,
      "epoch": 0.06246096189881324,
      "grad_norm": 11.5625,
      "learning_rate": 9.995794589968868e-06,
      "loss": 0.04561046659946442,
      "loss_step_1": 0.0454,
      "loss_step_2": 0.046,
      "loss_step_3": 0.0517,
      "step": 2100
    },
    {
      "entropy_step_1": 0.1238,
      "entropy_step_2": 0.1237,
      "entropy_step_3": 0.1276,
      "epoch": 0.06275839505071235,
      "grad_norm": 7.0,
      "learning_rate": 9.995590503207824e-06,
      "loss": 0.046977975964546205,
      "loss_step_1": 0.0379,
      "loss_step_2": 0.0376,
      "loss_step_3": 0.033,
      "step": 2110
    },
    {
      "entropy_step_1": 0.1445,
      "entropy_step_2": 0.1448,
      "entropy_step_3": 0.1263,
      "epoch": 0.06305582820261146,
      "grad_norm": 16.875,
      "learning_rate": 9.995381583144997e-06,
      "loss": 0.054321813583374026,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0391,
      "step": 2120
    },
    {
      "entropy_step_1": 0.1489,
      "entropy_step_2": 0.1502,
      "entropy_step_3": 0.1287,
      "epoch": 0.06335326135451057,
      "grad_norm": 15.3125,
      "learning_rate": 9.995167829982513e-06,
      "loss": 0.05158072113990784,
      "loss_step_1": 0.0477,
      "loss_step_2": 0.0477,
      "loss_step_3": 0.037,
      "step": 2130
    },
    {
      "entropy_step_1": 0.1346,
      "entropy_step_2": 0.1337,
      "entropy_step_3": 0.1439,
      "epoch": 0.06365069450640969,
      "grad_norm": 3.390625,
      "learning_rate": 9.994949243927185e-06,
      "loss": 0.045175564289093015,
      "loss_step_1": 0.0383,
      "loss_step_2": 0.0385,
      "loss_step_3": 0.0398,
      "step": 2140
    },
    {
      "entropy_step_1": 0.1424,
      "entropy_step_2": 0.1428,
      "entropy_step_3": 0.1536,
      "epoch": 0.06394812765830879,
      "grad_norm": 5.5,
      "learning_rate": 9.9947258251905e-06,
      "loss": 0.04198208451271057,
      "loss_step_1": 0.0445,
      "loss_step_2": 0.0445,
      "loss_step_3": 0.0469,
      "step": 2150
    },
    {
      "entropy_step_1": 0.1583,
      "entropy_step_2": 0.1572,
      "entropy_step_3": 0.1546,
      "epoch": 0.06424556081020791,
      "grad_norm": 7.875,
      "learning_rate": 9.994497573988613e-06,
      "loss": 0.04684409499168396,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0495,
      "loss_step_3": 0.0501,
      "step": 2160
    },
    {
      "entropy_step_1": 0.1508,
      "entropy_step_2": 0.1506,
      "entropy_step_3": 0.1283,
      "epoch": 0.06454299396210701,
      "grad_norm": 3.0,
      "learning_rate": 9.994264490542364e-06,
      "loss": 0.04069793522357941,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0462,
      "loss_step_3": 0.0421,
      "step": 2170
    },
    {
      "entropy_step_1": 0.1251,
      "entropy_step_2": 0.1257,
      "entropy_step_3": 0.1357,
      "epoch": 0.06484042711400613,
      "grad_norm": 5.15625,
      "learning_rate": 9.994026575077264e-06,
      "loss": 0.05518852472305298,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0666,
      "step": 2180
    },
    {
      "entropy_step_1": 0.1418,
      "entropy_step_2": 0.141,
      "entropy_step_3": 0.1354,
      "epoch": 0.06513786026590523,
      "grad_norm": 7.9375,
      "learning_rate": 9.993783827823498e-06,
      "loss": 0.045247966051101686,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0434,
      "step": 2190
    },
    {
      "entropy_step_1": 0.1322,
      "entropy_step_2": 0.1325,
      "entropy_step_3": 0.1514,
      "epoch": 0.06543529341780435,
      "grad_norm": 23.875,
      "learning_rate": 9.99353624901593e-06,
      "loss": 0.04181052446365356,
      "loss_step_1": 0.0419,
      "loss_step_2": 0.0416,
      "loss_step_3": 0.0506,
      "step": 2200
    },
    {
      "entropy_step_1": 0.1395,
      "entropy_step_2": 0.14,
      "entropy_step_3": 0.1413,
      "epoch": 0.06573272656970346,
      "grad_norm": 16.625,
      "learning_rate": 9.993283838894092e-06,
      "loss": 0.042797237634658813,
      "loss_step_1": 0.0457,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.0559,
      "step": 2210
    },
    {
      "entropy_step_1": 0.1425,
      "entropy_step_2": 0.1421,
      "entropy_step_3": 0.1289,
      "epoch": 0.06603015972160257,
      "grad_norm": 9.0625,
      "learning_rate": 9.993026597702198e-06,
      "loss": 0.04693845808506012,
      "loss_step_1": 0.0425,
      "loss_step_2": 0.0429,
      "loss_step_3": 0.0364,
      "step": 2220
    },
    {
      "entropy_step_1": 0.1437,
      "entropy_step_2": 0.1438,
      "entropy_step_3": 0.1374,
      "epoch": 0.06632759287350168,
      "grad_norm": 10.6875,
      "learning_rate": 9.992764525689132e-06,
      "loss": 0.04637958407402039,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0468,
      "loss_step_3": 0.0415,
      "step": 2230
    },
    {
      "entropy_step_1": 0.1443,
      "entropy_step_2": 0.1447,
      "entropy_step_3": 0.1288,
      "epoch": 0.0666250260254008,
      "grad_norm": 12.1875,
      "learning_rate": 9.992497623108447e-06,
      "loss": 0.04864843785762787,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0523,
      "step": 2240
    },
    {
      "entropy_step_1": 0.1559,
      "entropy_step_2": 0.1573,
      "entropy_step_3": 0.16,
      "epoch": 0.0669224591772999,
      "grad_norm": 6.46875,
      "learning_rate": 9.992225890218382e-06,
      "loss": 0.04461889863014221,
      "loss_step_1": 0.0413,
      "loss_step_2": 0.0414,
      "loss_step_3": 0.0397,
      "step": 2250
    },
    {
      "entropy_step_1": 0.1504,
      "entropy_step_2": 0.1506,
      "entropy_step_3": 0.1536,
      "epoch": 0.06721989232919902,
      "grad_norm": 9.3125,
      "learning_rate": 9.991949327281839e-06,
      "loss": 0.04224473237991333,
      "loss_step_1": 0.0439,
      "loss_step_2": 0.0437,
      "loss_step_3": 0.0376,
      "step": 2260
    },
    {
      "entropy_step_1": 0.1355,
      "entropy_step_2": 0.1368,
      "entropy_step_3": 0.1418,
      "epoch": 0.06751732548109812,
      "grad_norm": 15.3125,
      "learning_rate": 9.991667934566395e-06,
      "loss": 0.04837535619735718,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0519,
      "step": 2270
    },
    {
      "entropy_step_1": 0.1351,
      "entropy_step_2": 0.1375,
      "entropy_step_3": 0.1417,
      "epoch": 0.06781475863299724,
      "grad_norm": 3.96875,
      "learning_rate": 9.991381712344304e-06,
      "loss": 0.04701527953147888,
      "loss_step_1": 0.0466,
      "loss_step_2": 0.0461,
      "loss_step_3": 0.0422,
      "step": 2280
    },
    {
      "entropy_step_1": 0.1347,
      "entropy_step_2": 0.1345,
      "entropy_step_3": 0.1196,
      "epoch": 0.06811219178489634,
      "grad_norm": 7.75,
      "learning_rate": 9.991090660892488e-06,
      "loss": 0.04557845890522003,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.044,
      "loss_step_3": 0.0363,
      "step": 2290
    },
    {
      "entropy_step_1": 0.1343,
      "entropy_step_2": 0.1348,
      "entropy_step_3": 0.1271,
      "epoch": 0.06840962493679545,
      "grad_norm": 16.5,
      "learning_rate": 9.990794780492546e-06,
      "loss": 0.05021997690200806,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0491,
      "step": 2300
    },
    {
      "entropy_step_1": 0.1611,
      "entropy_step_2": 0.1617,
      "entropy_step_3": 0.1696,
      "epoch": 0.06870705808869457,
      "grad_norm": 7.0625,
      "learning_rate": 9.990494071430743e-06,
      "loss": 0.0506308913230896,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0602,
      "step": 2310
    },
    {
      "entropy_step_1": 0.1441,
      "entropy_step_2": 0.1449,
      "entropy_step_3": 0.1307,
      "epoch": 0.06900449124059367,
      "grad_norm": 8.5625,
      "learning_rate": 9.99018853399802e-06,
      "loss": 0.042967775464057924,
      "loss_step_1": 0.0402,
      "loss_step_2": 0.0404,
      "loss_step_3": 0.0374,
      "step": 2320
    },
    {
      "entropy_step_1": 0.1553,
      "entropy_step_2": 0.156,
      "entropy_step_3": 0.1647,
      "epoch": 0.06930192439249279,
      "grad_norm": 7.5,
      "learning_rate": 9.989878168489989e-06,
      "loss": 0.04547125399112702,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0561,
      "step": 2330
    },
    {
      "entropy_step_1": 0.1508,
      "entropy_step_2": 0.1526,
      "entropy_step_3": 0.1581,
      "epoch": 0.0695993575443919,
      "grad_norm": 11.3125,
      "learning_rate": 9.989562975206933e-06,
      "loss": 0.04681617319583893,
      "loss_step_1": 0.0458,
      "loss_step_2": 0.0459,
      "loss_step_3": 0.0434,
      "step": 2340
    },
    {
      "entropy_step_1": 0.1595,
      "entropy_step_2": 0.1615,
      "entropy_step_3": 0.158,
      "epoch": 0.06989679069629101,
      "grad_norm": 21.25,
      "learning_rate": 9.989242954453805e-06,
      "loss": 0.04921803176403046,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0461,
      "step": 2350
    },
    {
      "entropy_step_1": 0.1237,
      "entropy_step_2": 0.1254,
      "entropy_step_3": 0.1302,
      "epoch": 0.07019422384819012,
      "grad_norm": 4.5,
      "learning_rate": 9.98891810654023e-06,
      "loss": 0.03930695950984955,
      "loss_step_1": 0.0398,
      "loss_step_2": 0.0397,
      "loss_step_3": 0.0423,
      "step": 2360
    },
    {
      "entropy_step_1": 0.136,
      "entropy_step_2": 0.1372,
      "entropy_step_3": 0.1431,
      "epoch": 0.07049165700008923,
      "grad_norm": 4.90625,
      "learning_rate": 9.988588431780507e-06,
      "loss": 0.052748292684555054,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0594,
      "step": 2370
    },
    {
      "entropy_step_1": 0.1389,
      "entropy_step_2": 0.1402,
      "entropy_step_3": 0.1482,
      "epoch": 0.07078909015198834,
      "grad_norm": 8.3125,
      "learning_rate": 9.988253930493592e-06,
      "loss": 0.050638127326965335,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0602,
      "step": 2380
    },
    {
      "entropy_step_1": 0.1588,
      "entropy_step_2": 0.1603,
      "entropy_step_3": 0.1498,
      "epoch": 0.07108652330388746,
      "grad_norm": 9.9375,
      "learning_rate": 9.987914603003124e-06,
      "loss": 0.05277670621871948,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0485,
      "loss_step_3": 0.0439,
      "step": 2390
    },
    {
      "entropy_step_1": 0.1527,
      "entropy_step_2": 0.156,
      "entropy_step_3": 0.1564,
      "epoch": 0.07138395645578656,
      "grad_norm": 9.0,
      "learning_rate": 9.98757044963741e-06,
      "loss": 0.04667108654975891,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0445,
      "step": 2400
    },
    {
      "entropy_step_1": 0.1327,
      "entropy_step_2": 0.1353,
      "entropy_step_3": 0.13,
      "epoch": 0.07168138960768568,
      "grad_norm": 19.375,
      "learning_rate": 9.987221470729418e-06,
      "loss": 0.04717521071434021,
      "loss_step_1": 0.0416,
      "loss_step_2": 0.0417,
      "loss_step_3": 0.0404,
      "step": 2410
    },
    {
      "entropy_step_1": 0.144,
      "entropy_step_2": 0.1464,
      "entropy_step_3": 0.1472,
      "epoch": 0.07197882275958478,
      "grad_norm": 12.9375,
      "learning_rate": 9.986867666616794e-06,
      "loss": 0.04947497546672821,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0472,
      "step": 2420
    },
    {
      "entropy_step_1": 0.1715,
      "entropy_step_2": 0.1744,
      "entropy_step_3": 0.169,
      "epoch": 0.0722762559114839,
      "grad_norm": 8.3125,
      "learning_rate": 9.986509037641846e-06,
      "loss": 0.05462733507156372,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0527,
      "step": 2430
    },
    {
      "entropy_step_1": 0.1676,
      "entropy_step_2": 0.1697,
      "entropy_step_3": 0.1729,
      "epoch": 0.072573689063383,
      "grad_norm": 7.75,
      "learning_rate": 9.986145584151551e-06,
      "loss": 0.04701300859451294,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.0474,
      "loss_step_3": 0.0426,
      "step": 2440
    },
    {
      "entropy_step_1": 0.1604,
      "entropy_step_2": 0.1625,
      "entropy_step_3": 0.1601,
      "epoch": 0.07287112221528212,
      "grad_norm": 4.53125,
      "learning_rate": 9.98577730649756e-06,
      "loss": 0.05368818044662475,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0457,
      "step": 2450
    },
    {
      "entropy_step_1": 0.1458,
      "entropy_step_2": 0.1474,
      "entropy_step_3": 0.1451,
      "epoch": 0.07316855536718123,
      "grad_norm": 8.3125,
      "learning_rate": 9.98540420503618e-06,
      "loss": 0.03969095051288605,
      "loss_step_1": 0.0398,
      "loss_step_2": 0.0399,
      "loss_step_3": 0.0366,
      "step": 2460
    },
    {
      "entropy_step_1": 0.1477,
      "entropy_step_2": 0.1489,
      "entropy_step_3": 0.1525,
      "epoch": 0.07346598851908033,
      "grad_norm": 7.84375,
      "learning_rate": 9.985026280128398e-06,
      "loss": 0.04592232704162598,
      "loss_step_1": 0.0474,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0547,
      "step": 2470
    },
    {
      "entropy_step_1": 0.1406,
      "entropy_step_2": 0.1417,
      "entropy_step_3": 0.1295,
      "epoch": 0.07376342167097945,
      "grad_norm": 8.8125,
      "learning_rate": 9.984643532139859e-06,
      "loss": 0.0388120025396347,
      "loss_step_1": 0.0438,
      "loss_step_2": 0.0436,
      "loss_step_3": 0.0414,
      "step": 2480
    },
    {
      "entropy_step_1": 0.1441,
      "entropy_step_2": 0.1462,
      "entropy_step_3": 0.1345,
      "epoch": 0.07406085482287855,
      "grad_norm": 5.875,
      "learning_rate": 9.984255961440875e-06,
      "loss": 0.05029494762420654,
      "loss_step_1": 0.0508,
      "loss_step_2": 0.051,
      "loss_step_3": 0.051,
      "step": 2490
    },
    {
      "entropy_step_1": 0.1391,
      "entropy_step_2": 0.1416,
      "entropy_step_3": 0.1193,
      "epoch": 0.07435828797477767,
      "grad_norm": 11.5,
      "learning_rate": 9.98386356840643e-06,
      "loss": 0.051236027479171754,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0396,
      "step": 2500
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1752,
      "entropy_step_3": 0.1867,
      "epoch": 0.07465572112667677,
      "grad_norm": 10.0,
      "learning_rate": 9.983466353416165e-06,
      "loss": 0.056961292028427125,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0592,
      "loss_step_3": 0.0686,
      "step": 2510
    },
    {
      "entropy_step_1": 0.1581,
      "entropy_step_2": 0.1597,
      "entropy_step_3": 0.1578,
      "epoch": 0.07495315427857589,
      "grad_norm": 12.0,
      "learning_rate": 9.983064316854394e-06,
      "loss": 0.04939534366130829,
      "loss_step_1": 0.0485,
      "loss_step_2": 0.0479,
      "loss_step_3": 0.0491,
      "step": 2520
    },
    {
      "entropy_step_1": 0.1507,
      "entropy_step_2": 0.1534,
      "entropy_step_3": 0.1567,
      "epoch": 0.075250587430475,
      "grad_norm": 5.8125,
      "learning_rate": 9.982657459110092e-06,
      "loss": 0.04895128309726715,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0474,
      "step": 2530
    },
    {
      "entropy_step_1": 0.1476,
      "entropy_step_2": 0.1508,
      "entropy_step_3": 0.153,
      "epoch": 0.07554802058237411,
      "grad_norm": 3.34375,
      "learning_rate": 9.982245780576901e-06,
      "loss": 0.044051170349121094,
      "loss_step_1": 0.047,
      "loss_step_2": 0.047,
      "loss_step_3": 0.0471,
      "step": 2540
    },
    {
      "entropy_step_1": 0.142,
      "entropy_step_2": 0.1436,
      "entropy_step_3": 0.1381,
      "epoch": 0.07584545373427322,
      "grad_norm": 9.625,
      "learning_rate": 9.981829281653121e-06,
      "loss": 0.047561845183372496,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0444,
      "step": 2550
    },
    {
      "entropy_step_1": 0.1376,
      "entropy_step_2": 0.1386,
      "entropy_step_3": 0.1341,
      "epoch": 0.07614288688617234,
      "grad_norm": 13.0,
      "learning_rate": 9.981407962741725e-06,
      "loss": 0.04671943187713623,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.0443,
      "loss_step_3": 0.0366,
      "step": 2560
    },
    {
      "entropy_step_1": 0.1218,
      "entropy_step_2": 0.1226,
      "entropy_step_3": 0.1189,
      "epoch": 0.07644032003807144,
      "grad_norm": 7.0625,
      "learning_rate": 9.980981824250344e-06,
      "loss": 0.04137278497219086,
      "loss_step_1": 0.0378,
      "loss_step_2": 0.0376,
      "loss_step_3": 0.0308,
      "step": 2570
    },
    {
      "entropy_step_1": 0.1434,
      "entropy_step_2": 0.1444,
      "entropy_step_3": 0.1504,
      "epoch": 0.07673775318997056,
      "grad_norm": 4.375,
      "learning_rate": 9.98055086659127e-06,
      "loss": 0.03801659643650055,
      "loss_step_1": 0.0401,
      "loss_step_2": 0.0398,
      "loss_step_3": 0.0436,
      "step": 2580
    },
    {
      "entropy_step_1": 0.135,
      "entropy_step_2": 0.1359,
      "entropy_step_3": 0.1343,
      "epoch": 0.07703518634186966,
      "grad_norm": 13.0625,
      "learning_rate": 9.980115090181464e-06,
      "loss": 0.043946340680122375,
      "loss_step_1": 0.045,
      "loss_step_2": 0.0448,
      "loss_step_3": 0.0442,
      "step": 2590
    },
    {
      "entropy_step_1": 0.1325,
      "entropy_step_2": 0.1336,
      "entropy_step_3": 0.1476,
      "epoch": 0.07733261949376878,
      "grad_norm": 5.125,
      "learning_rate": 9.979674495442543e-06,
      "loss": 0.048748940229415894,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.054,
      "step": 2600
    },
    {
      "entropy_step_1": 0.1457,
      "entropy_step_2": 0.1464,
      "entropy_step_3": 0.1521,
      "epoch": 0.07763005264566789,
      "grad_norm": 5.375,
      "learning_rate": 9.97922908280079e-06,
      "loss": 0.041534245014190674,
      "loss_step_1": 0.0379,
      "loss_step_2": 0.0376,
      "loss_step_3": 0.0363,
      "step": 2610
    },
    {
      "entropy_step_1": 0.1498,
      "entropy_step_2": 0.1497,
      "entropy_step_3": 0.1668,
      "epoch": 0.077927485797567,
      "grad_norm": 6.71875,
      "learning_rate": 9.978778852687148e-06,
      "loss": 0.05402216911315918,
      "loss_step_1": 0.0454,
      "loss_step_2": 0.045,
      "loss_step_3": 0.0518,
      "step": 2620
    },
    {
      "entropy_step_1": 0.1546,
      "entropy_step_2": 0.1554,
      "entropy_step_3": 0.1673,
      "epoch": 0.07822491894946611,
      "grad_norm": 7.375,
      "learning_rate": 9.978323805537218e-06,
      "loss": 0.05215198993682861,
      "loss_step_1": 0.0503,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0512,
      "step": 2630
    },
    {
      "entropy_step_1": 0.1492,
      "entropy_step_2": 0.1488,
      "entropy_step_3": 0.1486,
      "epoch": 0.07852235210136521,
      "grad_norm": 12.1875,
      "learning_rate": 9.977863941791267e-06,
      "loss": 0.037174805998802185,
      "loss_step_1": 0.0378,
      "loss_step_2": 0.038,
      "loss_step_3": 0.0378,
      "step": 2640
    },
    {
      "entropy_step_1": 0.1467,
      "entropy_step_2": 0.1477,
      "entropy_step_3": 0.1441,
      "epoch": 0.07881978525326433,
      "grad_norm": 3.703125,
      "learning_rate": 9.977399261894218e-06,
      "loss": 0.04143659770488739,
      "loss_step_1": 0.041,
      "loss_step_2": 0.0407,
      "loss_step_3": 0.0437,
      "step": 2650
    },
    {
      "entropy_step_1": 0.146,
      "entropy_step_2": 0.1461,
      "entropy_step_3": 0.141,
      "epoch": 0.07911721840516343,
      "grad_norm": 4.46875,
      "learning_rate": 9.976929766295658e-06,
      "loss": 0.04193979501724243,
      "loss_step_1": 0.0404,
      "loss_step_2": 0.0401,
      "loss_step_3": 0.0344,
      "step": 2660
    },
    {
      "entropy_step_1": 0.1483,
      "entropy_step_2": 0.1492,
      "entropy_step_3": 0.1528,
      "epoch": 0.07941465155706255,
      "grad_norm": 5.40625,
      "learning_rate": 9.976455455449828e-06,
      "loss": 0.05000919699668884,
      "loss_step_1": 0.0483,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0479,
      "step": 2670
    },
    {
      "entropy_step_1": 0.1554,
      "entropy_step_2": 0.1568,
      "entropy_step_3": 0.1612,
      "epoch": 0.07971208470896166,
      "grad_norm": 9.5,
      "learning_rate": 9.975976329815627e-06,
      "loss": 0.04767659306526184,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0541,
      "step": 2680
    },
    {
      "entropy_step_1": 0.1558,
      "entropy_step_2": 0.1563,
      "entropy_step_3": 0.1647,
      "epoch": 0.08000951786086077,
      "grad_norm": 3.15625,
      "learning_rate": 9.975492389856622e-06,
      "loss": 0.0565362811088562,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0511,
      "step": 2690
    },
    {
      "entropy_step_1": 0.1548,
      "entropy_step_2": 0.1554,
      "entropy_step_3": 0.1596,
      "epoch": 0.08030695101275988,
      "grad_norm": 5.15625,
      "learning_rate": 9.975003636041028e-06,
      "loss": 0.04910741746425629,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0482,
      "loss_step_3": 0.054,
      "step": 2700
    },
    {
      "entropy_step_1": 0.1541,
      "entropy_step_2": 0.1537,
      "entropy_step_3": 0.1647,
      "epoch": 0.080604384164659,
      "grad_norm": 4.59375,
      "learning_rate": 9.97451006884172e-06,
      "loss": 0.044136989116668704,
      "loss_step_1": 0.0469,
      "loss_step_2": 0.047,
      "loss_step_3": 0.0489,
      "step": 2710
    },
    {
      "entropy_step_1": 0.1403,
      "entropy_step_2": 0.1409,
      "entropy_step_3": 0.1379,
      "epoch": 0.0809018173165581,
      "grad_norm": 13.125,
      "learning_rate": 9.974011688736233e-06,
      "loss": 0.04663305282592774,
      "loss_step_1": 0.0434,
      "loss_step_2": 0.0434,
      "loss_step_3": 0.0399,
      "step": 2720
    },
    {
      "entropy_step_1": 0.1475,
      "entropy_step_2": 0.1475,
      "entropy_step_3": 0.1442,
      "epoch": 0.08119925046845722,
      "grad_norm": 11.3125,
      "learning_rate": 9.973508496206756e-06,
      "loss": 0.04529033601284027,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0461,
      "loss_step_3": 0.0368,
      "step": 2730
    },
    {
      "entropy_step_1": 0.1591,
      "entropy_step_2": 0.159,
      "entropy_step_3": 0.1551,
      "epoch": 0.08149668362035632,
      "grad_norm": 4.84375,
      "learning_rate": 9.973000491740135e-06,
      "loss": 0.050670129060745236,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0461,
      "step": 2740
    },
    {
      "entropy_step_1": 0.136,
      "entropy_step_2": 0.1348,
      "entropy_step_3": 0.1422,
      "epoch": 0.08179411677225544,
      "grad_norm": 5.5625,
      "learning_rate": 9.97248767582787e-06,
      "loss": 0.043995720148086545,
      "loss_step_1": 0.0466,
      "loss_step_2": 0.0463,
      "loss_step_3": 0.0461,
      "step": 2750
    },
    {
      "entropy_step_1": 0.1319,
      "entropy_step_2": 0.1309,
      "entropy_step_3": 0.1257,
      "epoch": 0.08209154992415454,
      "grad_norm": 6.34375,
      "learning_rate": 9.97197004896612e-06,
      "loss": 0.043430691957473753,
      "loss_step_1": 0.0452,
      "loss_step_2": 0.0453,
      "loss_step_3": 0.0391,
      "step": 2760
    },
    {
      "entropy_step_1": 0.1126,
      "entropy_step_2": 0.1114,
      "entropy_step_3": 0.1137,
      "epoch": 0.08238898307605366,
      "grad_norm": 7.5,
      "learning_rate": 9.971447611655694e-06,
      "loss": 0.03872762322425842,
      "loss_step_1": 0.0339,
      "loss_step_2": 0.0339,
      "loss_step_3": 0.0321,
      "step": 2770
    },
    {
      "entropy_step_1": 0.1405,
      "entropy_step_2": 0.1387,
      "entropy_step_3": 0.1508,
      "epoch": 0.08268641622795277,
      "grad_norm": 10.75,
      "learning_rate": 9.970920364402055e-06,
      "loss": 0.0470515638589859,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0487,
      "loss_step_3": 0.0622,
      "step": 2780
    },
    {
      "entropy_step_1": 0.1492,
      "entropy_step_2": 0.1469,
      "entropy_step_3": 0.1349,
      "epoch": 0.08298384937985188,
      "grad_norm": 13.25,
      "learning_rate": 9.970388307715326e-06,
      "loss": 0.04258919656276703,
      "loss_step_1": 0.0455,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.039,
      "step": 2790
    },
    {
      "entropy_step_1": 0.146,
      "entropy_step_2": 0.1432,
      "entropy_step_3": 0.1294,
      "epoch": 0.08328128253175099,
      "grad_norm": 31.125,
      "learning_rate": 9.969851442110277e-06,
      "loss": 0.05227503180503845,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0429,
      "step": 2800
    },
    {
      "entropy_step_1": 0.1395,
      "entropy_step_2": 0.1383,
      "entropy_step_3": 0.1578,
      "epoch": 0.08357871568365009,
      "grad_norm": 14.1875,
      "learning_rate": 9.969309768106332e-06,
      "loss": 0.04820596873760223,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0569,
      "step": 2810
    },
    {
      "entropy_step_1": 0.1418,
      "entropy_step_2": 0.1418,
      "entropy_step_3": 0.1504,
      "epoch": 0.08387614883554921,
      "grad_norm": 11.25,
      "learning_rate": 9.96876328622757e-06,
      "loss": 0.043928271532058714,
      "loss_step_1": 0.0423,
      "loss_step_2": 0.0421,
      "loss_step_3": 0.0553,
      "step": 2820
    },
    {
      "entropy_step_1": 0.1556,
      "entropy_step_2": 0.1559,
      "entropy_step_3": 0.1429,
      "epoch": 0.08417358198744831,
      "grad_norm": 6.71875,
      "learning_rate": 9.96821199700272e-06,
      "loss": 0.053042399883270266,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0486,
      "step": 2830
    },
    {
      "entropy_step_1": 0.1459,
      "entropy_step_2": 0.1461,
      "entropy_step_3": 0.1438,
      "epoch": 0.08447101513934743,
      "grad_norm": 16.625,
      "learning_rate": 9.96765590096516e-06,
      "loss": 0.043458014726638794,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0585,
      "step": 2840
    },
    {
      "entropy_step_1": 0.1589,
      "entropy_step_2": 0.1587,
      "entropy_step_3": 0.1594,
      "epoch": 0.08476844829124654,
      "grad_norm": 6.875,
      "learning_rate": 9.967094998652919e-06,
      "loss": 0.042645373940467836,
      "loss_step_1": 0.0462,
      "loss_step_2": 0.046,
      "loss_step_3": 0.0491,
      "step": 2850
    },
    {
      "entropy_step_1": 0.1884,
      "entropy_step_2": 0.1871,
      "entropy_step_3": 0.1837,
      "epoch": 0.08506588144314566,
      "grad_norm": 5.90625,
      "learning_rate": 9.966529290608682e-06,
      "loss": 0.05687616467475891,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0535,
      "step": 2860
    },
    {
      "entropy_step_1": 0.1535,
      "entropy_step_2": 0.1538,
      "entropy_step_3": 0.1603,
      "epoch": 0.08536331459504476,
      "grad_norm": 10.5,
      "learning_rate": 9.965958777379776e-06,
      "loss": 0.046384000778198244,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0523,
      "step": 2870
    },
    {
      "entropy_step_1": 0.1458,
      "entropy_step_2": 0.1456,
      "entropy_step_3": 0.137,
      "epoch": 0.08566074774694388,
      "grad_norm": 32.5,
      "learning_rate": 9.96538345951818e-06,
      "loss": 0.04779348075389862,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0475,
      "step": 2880
    },
    {
      "entropy_step_1": 0.1431,
      "entropy_step_2": 0.1429,
      "entropy_step_3": 0.1523,
      "epoch": 0.08595818089884298,
      "grad_norm": 15.0,
      "learning_rate": 9.964803337580527e-06,
      "loss": 0.0437695324420929,
      "loss_step_1": 0.0433,
      "loss_step_2": 0.0441,
      "loss_step_3": 0.0458,
      "step": 2890
    },
    {
      "entropy_step_1": 0.164,
      "entropy_step_2": 0.1632,
      "entropy_step_3": 0.1801,
      "epoch": 0.0862556140507421,
      "grad_norm": 7.59375,
      "learning_rate": 9.964218412128086e-06,
      "loss": 0.04932941496372223,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0555,
      "step": 2900
    },
    {
      "entropy_step_1": 0.1564,
      "entropy_step_2": 0.1562,
      "entropy_step_3": 0.1466,
      "epoch": 0.0865530472026412,
      "grad_norm": 6.0625,
      "learning_rate": 9.963628683726783e-06,
      "loss": 0.04746485352516174,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.051,
      "step": 2910
    },
    {
      "entropy_step_1": 0.1597,
      "entropy_step_2": 0.1584,
      "entropy_step_3": 0.1351,
      "epoch": 0.08685048035454032,
      "grad_norm": 5.90625,
      "learning_rate": 9.96303415294719e-06,
      "loss": 0.03794457614421844,
      "loss_step_1": 0.0385,
      "loss_step_2": 0.0386,
      "loss_step_3": 0.044,
      "step": 2920
    },
    {
      "entropy_step_1": 0.1416,
      "entropy_step_2": 0.1412,
      "entropy_step_3": 0.1572,
      "epoch": 0.08714791350643943,
      "grad_norm": 3.1875,
      "learning_rate": 9.96243482036452e-06,
      "loss": 0.04426727294921875,
      "loss_step_1": 0.0444,
      "loss_step_2": 0.0443,
      "loss_step_3": 0.0418,
      "step": 2930
    },
    {
      "entropy_step_1": 0.1597,
      "entropy_step_2": 0.1603,
      "entropy_step_3": 0.1648,
      "epoch": 0.08744534665833854,
      "grad_norm": 20.25,
      "learning_rate": 9.961830686558641e-06,
      "loss": 0.0491367906332016,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0628,
      "step": 2940
    },
    {
      "entropy_step_1": 0.1365,
      "entropy_step_2": 0.1371,
      "entropy_step_3": 0.1456,
      "epoch": 0.08774277981023765,
      "grad_norm": 3.953125,
      "learning_rate": 9.961221752114054e-06,
      "loss": 0.04207809865474701,
      "loss_step_1": 0.0406,
      "loss_step_2": 0.0405,
      "loss_step_3": 0.0447,
      "step": 2950
    },
    {
      "entropy_step_1": 0.1457,
      "entropy_step_2": 0.1457,
      "entropy_step_3": 0.1461,
      "epoch": 0.08804021296213677,
      "grad_norm": 18.5,
      "learning_rate": 9.960608017619918e-06,
      "loss": 0.04562165141105652,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0575,
      "step": 2960
    },
    {
      "entropy_step_1": 0.1493,
      "entropy_step_2": 0.1489,
      "entropy_step_3": 0.1562,
      "epoch": 0.08833764611403587,
      "grad_norm": 12.1875,
      "learning_rate": 9.959989483670022e-06,
      "loss": 0.04898250102996826,
      "loss_step_1": 0.0468,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0499,
      "step": 2970
    },
    {
      "entropy_step_1": 0.1689,
      "entropy_step_2": 0.1687,
      "entropy_step_3": 0.159,
      "epoch": 0.08863507926593499,
      "grad_norm": 4.9375,
      "learning_rate": 9.959366150862812e-06,
      "loss": 0.04850467443466187,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0533,
      "step": 2980
    },
    {
      "entropy_step_1": 0.1595,
      "entropy_step_2": 0.1587,
      "entropy_step_3": 0.147,
      "epoch": 0.08893251241783409,
      "grad_norm": 9.0625,
      "learning_rate": 9.958738019801371e-06,
      "loss": 0.0444347470998764,
      "loss_step_1": 0.0471,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0411,
      "step": 2990
    },
    {
      "entropy_step_1": 0.1688,
      "entropy_step_2": 0.1686,
      "entropy_step_3": 0.1757,
      "epoch": 0.0892299455697332,
      "grad_norm": 4.9375,
      "learning_rate": 9.958105091093419e-06,
      "loss": 0.048799967765808104,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0565,
      "step": 3000
    },
    {
      "epoch": 0.0892299455697332,
      "eval_accuracy": 0.967,
      "eval_f1": 0.9669825337603593,
      "eval_loss": 0.04609069973230362,
      "eval_precision": 0.9679902674058307,
      "eval_recall": 0.967,
      "eval_runtime": 20.3518,
      "eval_samples_per_second": 24.568,
      "eval_steps_per_second": 3.096,
      "step": 3000
    },
    {
      "entropy_step_1": 0.1458,
      "entropy_step_2": 0.1451,
      "entropy_step_3": 0.1305,
      "epoch": 0.08952737872163231,
      "grad_norm": 4.46875,
      "learning_rate": 9.957467365351328e-06,
      "loss": 0.04540419280529022,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.0444,
      "loss_step_3": 0.0376,
      "step": 3010
    },
    {
      "entropy_step_1": 0.1449,
      "entropy_step_2": 0.1446,
      "entropy_step_3": 0.1465,
      "epoch": 0.08982481187353142,
      "grad_norm": 7.75,
      "learning_rate": 9.956824843192105e-06,
      "loss": 0.046947437524795535,
      "loss_step_1": 0.0452,
      "loss_step_2": 0.0445,
      "loss_step_3": 0.0449,
      "step": 3020
    },
    {
      "entropy_step_1": 0.1667,
      "entropy_step_2": 0.1672,
      "entropy_step_3": 0.1544,
      "epoch": 0.09012224502543054,
      "grad_norm": 10.8125,
      "learning_rate": 9.9561775252374e-06,
      "loss": 0.044453924894332884,
      "loss_step_1": 0.0437,
      "loss_step_2": 0.0434,
      "loss_step_3": 0.0442,
      "step": 3030
    },
    {
      "entropy_step_1": 0.1309,
      "entropy_step_2": 0.1308,
      "entropy_step_3": 0.1235,
      "epoch": 0.09041967817732964,
      "grad_norm": 10.9375,
      "learning_rate": 9.9555254121135e-06,
      "loss": 0.03822645246982574,
      "loss_step_1": 0.0391,
      "loss_step_2": 0.0387,
      "loss_step_3": 0.0405,
      "step": 3040
    },
    {
      "entropy_step_1": 0.1389,
      "entropy_step_2": 0.1393,
      "entropy_step_3": 0.1436,
      "epoch": 0.09071711132922876,
      "grad_norm": 10.5,
      "learning_rate": 9.954868504451333e-06,
      "loss": 0.046877297759056094,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0495,
      "loss_step_3": 0.0503,
      "step": 3050
    },
    {
      "entropy_step_1": 0.1552,
      "entropy_step_2": 0.1546,
      "entropy_step_3": 0.1594,
      "epoch": 0.09101454448112786,
      "grad_norm": 10.375,
      "learning_rate": 9.95420680288647e-06,
      "loss": 0.046022814512252805,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0526,
      "step": 3060
    },
    {
      "entropy_step_1": 0.147,
      "entropy_step_2": 0.147,
      "entropy_step_3": 0.1344,
      "epoch": 0.09131197763302698,
      "grad_norm": 3.09375,
      "learning_rate": 9.953540308059112e-06,
      "loss": 0.04152839481830597,
      "loss_step_1": 0.0416,
      "loss_step_2": 0.0414,
      "loss_step_3": 0.0411,
      "step": 3070
    },
    {
      "entropy_step_1": 0.1409,
      "entropy_step_2": 0.1404,
      "entropy_step_3": 0.1461,
      "epoch": 0.09160941078492608,
      "grad_norm": 6.375,
      "learning_rate": 9.952869020614101e-06,
      "loss": 0.038666000962257384,
      "loss_step_1": 0.041,
      "loss_step_2": 0.0405,
      "loss_step_3": 0.0356,
      "step": 3080
    },
    {
      "entropy_step_1": 0.1324,
      "entropy_step_2": 0.1325,
      "entropy_step_3": 0.1467,
      "epoch": 0.0919068439368252,
      "grad_norm": 2.59375,
      "learning_rate": 9.952192941200923e-06,
      "loss": 0.05622309446334839,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.059,
      "step": 3090
    },
    {
      "entropy_step_1": 0.1267,
      "entropy_step_2": 0.1258,
      "entropy_step_3": 0.1301,
      "epoch": 0.0922042770887243,
      "grad_norm": 2.953125,
      "learning_rate": 9.951512070473686e-06,
      "loss": 0.04418583512306214,
      "loss_step_1": 0.0432,
      "loss_step_2": 0.0429,
      "loss_step_3": 0.0492,
      "step": 3100
    },
    {
      "entropy_step_1": 0.1361,
      "entropy_step_2": 0.1354,
      "entropy_step_3": 0.1481,
      "epoch": 0.09250171024062342,
      "grad_norm": 8.875,
      "learning_rate": 9.950826409091148e-06,
      "loss": 0.04911676049232483,
      "loss_step_1": 0.0451,
      "loss_step_2": 0.0453,
      "loss_step_3": 0.0426,
      "step": 3110
    },
    {
      "entropy_step_1": 0.1382,
      "entropy_step_2": 0.1376,
      "entropy_step_3": 0.134,
      "epoch": 0.09279914339252253,
      "grad_norm": 6.125,
      "learning_rate": 9.950135957716693e-06,
      "loss": 0.04658166170120239,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0421,
      "step": 3120
    },
    {
      "entropy_step_1": 0.155,
      "entropy_step_2": 0.1546,
      "entropy_step_3": 0.1459,
      "epoch": 0.09309657654442165,
      "grad_norm": 30.0,
      "learning_rate": 9.949440717018343e-06,
      "loss": 0.04999850988388062,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0536,
      "step": 3130
    },
    {
      "entropy_step_1": 0.1518,
      "entropy_step_2": 0.1525,
      "entropy_step_3": 0.1588,
      "epoch": 0.09339400969632075,
      "grad_norm": 4.03125,
      "learning_rate": 9.948740687668751e-06,
      "loss": 0.04651845991611481,
      "loss_step_1": 0.0438,
      "loss_step_2": 0.0435,
      "loss_step_3": 0.04,
      "step": 3140
    },
    {
      "entropy_step_1": 0.1442,
      "entropy_step_2": 0.1448,
      "entropy_step_3": 0.1569,
      "epoch": 0.09369144284821987,
      "grad_norm": 4.03125,
      "learning_rate": 9.948035870345208e-06,
      "loss": 0.04850611984729767,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0568,
      "step": 3150
    },
    {
      "entropy_step_1": 0.1622,
      "entropy_step_2": 0.162,
      "entropy_step_3": 0.1707,
      "epoch": 0.09398887600011897,
      "grad_norm": 6.28125,
      "learning_rate": 9.947326265729633e-06,
      "loss": 0.047752085328102115,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0464,
      "loss_step_3": 0.0508,
      "step": 3160
    },
    {
      "entropy_step_1": 0.1455,
      "entropy_step_2": 0.1464,
      "entropy_step_3": 0.1388,
      "epoch": 0.09428630915201808,
      "grad_norm": 6.6875,
      "learning_rate": 9.946611874508576e-06,
      "loss": 0.047635114192962645,
      "loss_step_1": 0.0455,
      "loss_step_2": 0.0455,
      "loss_step_3": 0.0452,
      "step": 3170
    },
    {
      "entropy_step_1": 0.1208,
      "entropy_step_2": 0.122,
      "entropy_step_3": 0.1314,
      "epoch": 0.0945837423039172,
      "grad_norm": 19.25,
      "learning_rate": 9.945892697373222e-06,
      "loss": 0.03833120763301849,
      "loss_step_1": 0.0398,
      "loss_step_2": 0.04,
      "loss_step_3": 0.0408,
      "step": 3180
    },
    {
      "entropy_step_1": 0.1495,
      "entropy_step_2": 0.1504,
      "entropy_step_3": 0.1503,
      "epoch": 0.0948811754558163,
      "grad_norm": 4.5,
      "learning_rate": 9.945168735019386e-06,
      "loss": 0.04421415328979492,
      "loss_step_1": 0.043,
      "loss_step_2": 0.0429,
      "loss_step_3": 0.0454,
      "step": 3190
    },
    {
      "entropy_step_1": 0.1476,
      "entropy_step_2": 0.1479,
      "entropy_step_3": 0.1484,
      "epoch": 0.09517860860771542,
      "grad_norm": 4.0,
      "learning_rate": 9.944439988147509e-06,
      "loss": 0.045053505897521974,
      "loss_step_1": 0.0455,
      "loss_step_2": 0.0451,
      "loss_step_3": 0.0411,
      "step": 3200
    },
    {
      "entropy_step_1": 0.1534,
      "entropy_step_2": 0.1542,
      "entropy_step_3": 0.1565,
      "epoch": 0.09547604175961452,
      "grad_norm": 9.625,
      "learning_rate": 9.943706457462667e-06,
      "loss": 0.04477883279323578,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.0442,
      "loss_step_3": 0.0398,
      "step": 3210
    },
    {
      "entropy_step_1": 0.1467,
      "entropy_step_2": 0.1475,
      "entropy_step_3": 0.1475,
      "epoch": 0.09577347491151364,
      "grad_norm": 2.78125,
      "learning_rate": 9.942968143674556e-06,
      "loss": 0.042651569843292235,
      "loss_step_1": 0.0438,
      "loss_step_2": 0.0437,
      "loss_step_3": 0.0509,
      "step": 3220
    },
    {
      "entropy_step_1": 0.1375,
      "entropy_step_2": 0.1385,
      "entropy_step_3": 0.1337,
      "epoch": 0.09607090806341274,
      "grad_norm": 5.40625,
      "learning_rate": 9.942225047497508e-06,
      "loss": 0.05120767951011658,
      "loss_step_1": 0.0458,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.0449,
      "step": 3230
    },
    {
      "entropy_step_1": 0.1362,
      "entropy_step_2": 0.1364,
      "entropy_step_3": 0.1133,
      "epoch": 0.09636834121531186,
      "grad_norm": 12.25,
      "learning_rate": 9.941477169650477e-06,
      "loss": 0.04052704572677612,
      "loss_step_1": 0.0417,
      "loss_step_2": 0.0414,
      "loss_step_3": 0.0264,
      "step": 3240
    },
    {
      "entropy_step_1": 0.1541,
      "entropy_step_2": 0.1543,
      "entropy_step_3": 0.1537,
      "epoch": 0.09666577436721097,
      "grad_norm": 6.375,
      "learning_rate": 9.940724510857048e-06,
      "loss": 0.04626941382884979,
      "loss_step_1": 0.0487,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0386,
      "step": 3250
    },
    {
      "entropy_step_1": 0.1301,
      "entropy_step_2": 0.1304,
      "entropy_step_3": 0.1155,
      "epoch": 0.09696320751911008,
      "grad_norm": 8.875,
      "learning_rate": 9.939967071845424e-06,
      "loss": 0.04930768609046936,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.044,
      "loss_step_3": 0.0319,
      "step": 3260
    },
    {
      "entropy_step_1": 0.1397,
      "entropy_step_2": 0.1405,
      "entropy_step_3": 0.137,
      "epoch": 0.09726064067100919,
      "grad_norm": 4.21875,
      "learning_rate": 9.93920485334844e-06,
      "loss": 0.05649701952934265,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0601,
      "step": 3270
    },
    {
      "entropy_step_1": 0.1471,
      "entropy_step_2": 0.148,
      "entropy_step_3": 0.1377,
      "epoch": 0.0975580738229083,
      "grad_norm": 19.75,
      "learning_rate": 9.938437856103556e-06,
      "loss": 0.05016440749168396,
      "loss_step_1": 0.0471,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0441,
      "step": 3280
    },
    {
      "entropy_step_1": 0.1573,
      "entropy_step_2": 0.1578,
      "entropy_step_3": 0.1723,
      "epoch": 0.09785550697480741,
      "grad_norm": 7.15625,
      "learning_rate": 9.937666080852846e-06,
      "loss": 0.048886454105377196,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0576,
      "step": 3290
    },
    {
      "entropy_step_1": 0.1492,
      "entropy_step_2": 0.1501,
      "entropy_step_3": 0.141,
      "epoch": 0.09815294012670653,
      "grad_norm": 5.65625,
      "learning_rate": 9.936889528343016e-06,
      "loss": 0.04790023565292358,
      "loss_step_1": 0.0412,
      "loss_step_2": 0.0407,
      "loss_step_3": 0.0385,
      "step": 3300
    },
    {
      "entropy_step_1": 0.1612,
      "entropy_step_2": 0.1621,
      "entropy_step_3": 0.1502,
      "epoch": 0.09845037327860563,
      "grad_norm": 18.375,
      "learning_rate": 9.936108199325394e-06,
      "loss": 0.0499936580657959,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0427,
      "step": 3310
    },
    {
      "entropy_step_1": 0.1544,
      "entropy_step_2": 0.1553,
      "entropy_step_3": 0.1586,
      "epoch": 0.09874780643050475,
      "grad_norm": 7.21875,
      "learning_rate": 9.93532209455592e-06,
      "loss": 0.04682372212409973,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0514,
      "step": 3320
    },
    {
      "entropy_step_1": 0.1451,
      "entropy_step_2": 0.1454,
      "entropy_step_3": 0.1577,
      "epoch": 0.09904523958240385,
      "grad_norm": 16.5,
      "learning_rate": 9.934531214795168e-06,
      "loss": 0.045251893997192386,
      "loss_step_1": 0.0456,
      "loss_step_2": 0.0459,
      "loss_step_3": 0.0553,
      "step": 3330
    },
    {
      "entropy_step_1": 0.1416,
      "entropy_step_2": 0.1421,
      "entropy_step_3": 0.1259,
      "epoch": 0.09934267273430296,
      "grad_norm": 2.75,
      "learning_rate": 9.933735560808319e-06,
      "loss": 0.042500489950180055,
      "loss_step_1": 0.0471,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0476,
      "step": 3340
    },
    {
      "entropy_step_1": 0.142,
      "entropy_step_2": 0.1438,
      "entropy_step_3": 0.1339,
      "epoch": 0.09964010588620208,
      "grad_norm": 9.0,
      "learning_rate": 9.932935133365185e-06,
      "loss": 0.03854290246963501,
      "loss_step_1": 0.0363,
      "loss_step_2": 0.0367,
      "loss_step_3": 0.0316,
      "step": 3350
    },
    {
      "entropy_step_1": 0.1317,
      "entropy_step_2": 0.1326,
      "entropy_step_3": 0.1249,
      "epoch": 0.09993753903810118,
      "grad_norm": 5.15625,
      "learning_rate": 9.932129933240185e-06,
      "loss": 0.044186833500862124,
      "loss_step_1": 0.0457,
      "loss_step_2": 0.0462,
      "loss_step_3": 0.0521,
      "step": 3360
    },
    {
      "entropy_step_1": 0.1442,
      "entropy_step_2": 0.1457,
      "entropy_step_3": 0.1409,
      "epoch": 0.1002349721900003,
      "grad_norm": 12.4375,
      "learning_rate": 9.931319961212363e-06,
      "loss": 0.04510790705680847,
      "loss_step_1": 0.0448,
      "loss_step_2": 0.0452,
      "loss_step_3": 0.0469,
      "step": 3370
    },
    {
      "entropy_step_1": 0.1494,
      "entropy_step_2": 0.1504,
      "entropy_step_3": 0.1414,
      "epoch": 0.1005324053418994,
      "grad_norm": 4.28125,
      "learning_rate": 9.930505218065378e-06,
      "loss": 0.043421033024787906,
      "loss_step_1": 0.0434,
      "loss_step_2": 0.0433,
      "loss_step_3": 0.0399,
      "step": 3380
    },
    {
      "entropy_step_1": 0.1523,
      "entropy_step_2": 0.1524,
      "entropy_step_3": 0.1476,
      "epoch": 0.10082983849379852,
      "grad_norm": 6.6875,
      "learning_rate": 9.929685704587506e-06,
      "loss": 0.042116639018058774,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.0482,
      "loss_step_3": 0.0391,
      "step": 3390
    },
    {
      "entropy_step_1": 0.135,
      "entropy_step_2": 0.1364,
      "entropy_step_3": 0.1403,
      "epoch": 0.10112727164569763,
      "grad_norm": 7.71875,
      "learning_rate": 9.928861421571636e-06,
      "loss": 0.04866665899753571,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0517,
      "step": 3400
    },
    {
      "entropy_step_1": 0.1248,
      "entropy_step_2": 0.1272,
      "entropy_step_3": 0.1237,
      "epoch": 0.10142470479759674,
      "grad_norm": 6.34375,
      "learning_rate": 9.928032369815274e-06,
      "loss": 0.03891539871692658,
      "loss_step_1": 0.0395,
      "loss_step_2": 0.0391,
      "loss_step_3": 0.0332,
      "step": 3410
    },
    {
      "entropy_step_1": 0.1279,
      "entropy_step_2": 0.1298,
      "entropy_step_3": 0.1241,
      "epoch": 0.10172213794949585,
      "grad_norm": 11.25,
      "learning_rate": 9.92719855012054e-06,
      "loss": 0.045831847190856936,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.043,
      "step": 3420
    },
    {
      "entropy_step_1": 0.1296,
      "entropy_step_2": 0.1315,
      "entropy_step_3": 0.1311,
      "epoch": 0.10201957110139497,
      "grad_norm": 4.53125,
      "learning_rate": 9.926359963294164e-06,
      "loss": 0.04285036325454712,
      "loss_step_1": 0.0434,
      "loss_step_2": 0.0432,
      "loss_step_3": 0.0454,
      "step": 3430
    },
    {
      "entropy_step_1": 0.1447,
      "entropy_step_2": 0.1459,
      "entropy_step_3": 0.1456,
      "epoch": 0.10231700425329407,
      "grad_norm": 9.625,
      "learning_rate": 9.92551661014749e-06,
      "loss": 0.04622603058815002,
      "loss_step_1": 0.0405,
      "loss_step_2": 0.0404,
      "loss_step_3": 0.0382,
      "step": 3440
    },
    {
      "entropy_step_1": 0.1227,
      "entropy_step_2": 0.1239,
      "entropy_step_3": 0.1229,
      "epoch": 0.10261443740519319,
      "grad_norm": 3.59375,
      "learning_rate": 9.924668491496474e-06,
      "loss": 0.03945842981338501,
      "loss_step_1": 0.0366,
      "loss_step_2": 0.0364,
      "loss_step_3": 0.0387,
      "step": 3450
    },
    {
      "entropy_step_1": 0.1307,
      "entropy_step_2": 0.1307,
      "entropy_step_3": 0.1169,
      "epoch": 0.10291187055709229,
      "grad_norm": 9.3125,
      "learning_rate": 9.923815608161685e-06,
      "loss": 0.039426401257514954,
      "loss_step_1": 0.0345,
      "loss_step_2": 0.0337,
      "loss_step_3": 0.0311,
      "step": 3460
    },
    {
      "entropy_step_1": 0.1357,
      "entropy_step_2": 0.1356,
      "entropy_step_3": 0.1298,
      "epoch": 0.10320930370899141,
      "grad_norm": 6.15625,
      "learning_rate": 9.922957960968293e-06,
      "loss": 0.04038623869419098,
      "loss_step_1": 0.041,
      "loss_step_2": 0.0413,
      "loss_step_3": 0.042,
      "step": 3470
    },
    {
      "entropy_step_1": 0.1397,
      "entropy_step_2": 0.1402,
      "entropy_step_3": 0.1386,
      "epoch": 0.10350673686089051,
      "grad_norm": 2.359375,
      "learning_rate": 9.92209555074609e-06,
      "loss": 0.040235075354576114,
      "loss_step_1": 0.0417,
      "loss_step_2": 0.0417,
      "loss_step_3": 0.0464,
      "step": 3480
    },
    {
      "entropy_step_1": 0.1406,
      "entropy_step_2": 0.1413,
      "entropy_step_3": 0.1464,
      "epoch": 0.10380417001278963,
      "grad_norm": 21.625,
      "learning_rate": 9.921228378329466e-06,
      "loss": 0.04837905466556549,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0494,
      "loss_step_3": 0.0549,
      "step": 3490
    },
    {
      "entropy_step_1": 0.1398,
      "entropy_step_2": 0.1416,
      "entropy_step_3": 0.1332,
      "epoch": 0.10410160316468874,
      "grad_norm": 14.875,
      "learning_rate": 9.920356444557423e-06,
      "loss": 0.060268628597259524,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0664,
      "step": 3500
    },
    {
      "entropy_step_1": 0.1698,
      "entropy_step_2": 0.1706,
      "entropy_step_3": 0.1637,
      "epoch": 0.10439903631658785,
      "grad_norm": 6.65625,
      "learning_rate": 9.919479750273568e-06,
      "loss": 0.052022480964660646,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0609,
      "step": 3510
    },
    {
      "entropy_step_1": 0.1485,
      "entropy_step_2": 0.1497,
      "entropy_step_3": 0.1657,
      "epoch": 0.10469646946848696,
      "grad_norm": 9.6875,
      "learning_rate": 9.918598296326116e-06,
      "loss": 0.04009193480014801,
      "loss_step_1": 0.0401,
      "loss_step_2": 0.0401,
      "loss_step_3": 0.0467,
      "step": 3520
    },
    {
      "entropy_step_1": 0.1286,
      "entropy_step_2": 0.1297,
      "entropy_step_3": 0.1231,
      "epoch": 0.10499390262038606,
      "grad_norm": 2.921875,
      "learning_rate": 9.917712083567887e-06,
      "loss": 0.04177891910076141,
      "loss_step_1": 0.0432,
      "loss_step_2": 0.0433,
      "loss_step_3": 0.0418,
      "step": 3530
    },
    {
      "entropy_step_1": 0.1242,
      "entropy_step_2": 0.1253,
      "entropy_step_3": 0.1262,
      "epoch": 0.10529133577228518,
      "grad_norm": 4.78125,
      "learning_rate": 9.9168211128563e-06,
      "loss": 0.042792341113090514,
      "loss_step_1": 0.0429,
      "loss_step_2": 0.0428,
      "loss_step_3": 0.0422,
      "step": 3540
    },
    {
      "entropy_step_1": 0.1367,
      "entropy_step_2": 0.1377,
      "entropy_step_3": 0.135,
      "epoch": 0.10558876892418428,
      "grad_norm": 9.375,
      "learning_rate": 9.915925385053383e-06,
      "loss": 0.04003942608833313,
      "loss_step_1": 0.0423,
      "loss_step_2": 0.0421,
      "loss_step_3": 0.0399,
      "step": 3550
    },
    {
      "entropy_step_1": 0.1372,
      "entropy_step_2": 0.1385,
      "entropy_step_3": 0.1417,
      "epoch": 0.1058862020760834,
      "grad_norm": 8.375,
      "learning_rate": 9.915024901025766e-06,
      "loss": 0.040530309081077576,
      "loss_step_1": 0.0421,
      "loss_step_2": 0.0419,
      "loss_step_3": 0.0452,
      "step": 3560
    },
    {
      "entropy_step_1": 0.1291,
      "entropy_step_2": 0.1298,
      "entropy_step_3": 0.1263,
      "epoch": 0.1061836352279825,
      "grad_norm": 6.8125,
      "learning_rate": 9.914119661644678e-06,
      "loss": 0.043109795451164244,
      "loss_step_1": 0.0399,
      "loss_step_2": 0.0397,
      "loss_step_3": 0.0398,
      "step": 3570
    },
    {
      "entropy_step_1": 0.1497,
      "entropy_step_2": 0.1504,
      "entropy_step_3": 0.152,
      "epoch": 0.10648106837988162,
      "grad_norm": 8.6875,
      "learning_rate": 9.913209667785952e-06,
      "loss": 0.04984990656375885,
      "loss_step_1": 0.0503,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.051,
      "step": 3580
    },
    {
      "entropy_step_1": 0.1404,
      "entropy_step_2": 0.1415,
      "entropy_step_3": 0.1404,
      "epoch": 0.10677850153178073,
      "grad_norm": 6.84375,
      "learning_rate": 9.912294920330019e-06,
      "loss": 0.036192023754119874,
      "loss_step_1": 0.0359,
      "loss_step_2": 0.0359,
      "loss_step_3": 0.0375,
      "step": 3590
    },
    {
      "entropy_step_1": 0.1371,
      "entropy_step_2": 0.1393,
      "entropy_step_3": 0.1485,
      "epoch": 0.10707593468367985,
      "grad_norm": 26.5,
      "learning_rate": 9.911375420161906e-06,
      "loss": 0.04664945900440216,
      "loss_step_1": 0.0417,
      "loss_step_2": 0.042,
      "loss_step_3": 0.0489,
      "step": 3600
    },
    {
      "entropy_step_1": 0.1408,
      "entropy_step_2": 0.142,
      "entropy_step_3": 0.1251,
      "epoch": 0.10737336783557895,
      "grad_norm": 36.5,
      "learning_rate": 9.910451168171248e-06,
      "loss": 0.05754275918006897,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.035,
      "step": 3610
    },
    {
      "entropy_step_1": 0.1354,
      "entropy_step_2": 0.1365,
      "entropy_step_3": 0.1444,
      "epoch": 0.10767080098747807,
      "grad_norm": 47.75,
      "learning_rate": 9.909522165252268e-06,
      "loss": 0.049727970361709596,
      "loss_step_1": 0.0504,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.044,
      "step": 3620
    },
    {
      "entropy_step_1": 0.1317,
      "entropy_step_2": 0.1332,
      "entropy_step_3": 0.1371,
      "epoch": 0.10796823413937717,
      "grad_norm": 3.609375,
      "learning_rate": 9.908588412303788e-06,
      "loss": 0.04896300435066223,
      "loss_step_1": 0.0476,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0495,
      "step": 3630
    },
    {
      "entropy_step_1": 0.1371,
      "entropy_step_2": 0.1385,
      "entropy_step_3": 0.1105,
      "epoch": 0.10826566729127629,
      "grad_norm": 3.078125,
      "learning_rate": 9.907649910229229e-06,
      "loss": 0.04440605044364929,
      "loss_step_1": 0.0453,
      "loss_step_2": 0.0454,
      "loss_step_3": 0.0388,
      "step": 3640
    },
    {
      "entropy_step_1": 0.1054,
      "entropy_step_2": 0.1063,
      "entropy_step_3": 0.1099,
      "epoch": 0.1085631004431754,
      "grad_norm": 3.328125,
      "learning_rate": 9.906706659936604e-06,
      "loss": 0.03872308433055878,
      "loss_step_1": 0.0368,
      "loss_step_2": 0.0368,
      "loss_step_3": 0.0426,
      "step": 3650
    },
    {
      "entropy_step_1": 0.1307,
      "entropy_step_2": 0.1308,
      "entropy_step_3": 0.1384,
      "epoch": 0.10886053359507451,
      "grad_norm": 9.875,
      "learning_rate": 9.90575866233852e-06,
      "loss": 0.04512480199337006,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0546,
      "step": 3660
    },
    {
      "entropy_step_1": 0.1236,
      "entropy_step_2": 0.1246,
      "entropy_step_3": 0.1082,
      "epoch": 0.10915796674697362,
      "grad_norm": 7.03125,
      "learning_rate": 9.904805918352177e-06,
      "loss": 0.03932113647460937,
      "loss_step_1": 0.0364,
      "loss_step_2": 0.035,
      "loss_step_3": 0.0293,
      "step": 3670
    },
    {
      "entropy_step_1": 0.1413,
      "entropy_step_2": 0.1414,
      "entropy_step_3": 0.1359,
      "epoch": 0.10945539989887274,
      "grad_norm": 8.4375,
      "learning_rate": 9.90384842889937e-06,
      "loss": 0.047062253952026366,
      "loss_step_1": 0.0461,
      "loss_step_2": 0.0461,
      "loss_step_3": 0.0419,
      "step": 3680
    },
    {
      "entropy_step_1": 0.1384,
      "entropy_step_2": 0.1399,
      "entropy_step_3": 0.1323,
      "epoch": 0.10975283305077184,
      "grad_norm": 5.125,
      "learning_rate": 9.902886194906479e-06,
      "loss": 0.0502930998802185,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0405,
      "step": 3690
    },
    {
      "entropy_step_1": 0.1098,
      "entropy_step_2": 0.1108,
      "entropy_step_3": 0.1071,
      "epoch": 0.11005026620267094,
      "grad_norm": 4.375,
      "learning_rate": 9.901919217304483e-06,
      "loss": 0.04219180047512054,
      "loss_step_1": 0.0365,
      "loss_step_2": 0.0365,
      "loss_step_3": 0.0315,
      "step": 3700
    },
    {
      "entropy_step_1": 0.1318,
      "entropy_step_2": 0.1318,
      "entropy_step_3": 0.1452,
      "epoch": 0.11034769935457006,
      "grad_norm": 11.4375,
      "learning_rate": 9.900947497028942e-06,
      "loss": 0.0463946133852005,
      "loss_step_1": 0.0465,
      "loss_step_2": 0.0463,
      "loss_step_3": 0.0529,
      "step": 3710
    },
    {
      "entropy_step_1": 0.1374,
      "entropy_step_2": 0.1371,
      "entropy_step_3": 0.1399,
      "epoch": 0.11064513250646917,
      "grad_norm": 5.46875,
      "learning_rate": 9.899971035020012e-06,
      "loss": 0.045424675941467284,
      "loss_step_1": 0.046,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.0505,
      "step": 3720
    },
    {
      "entropy_step_1": 0.1495,
      "entropy_step_2": 0.1497,
      "entropy_step_3": 0.158,
      "epoch": 0.11094256565836828,
      "grad_norm": 3.6875,
      "learning_rate": 9.898989832222432e-06,
      "loss": 0.04216669499874115,
      "loss_step_1": 0.0447,
      "loss_step_2": 0.0447,
      "loss_step_3": 0.0385,
      "step": 3730
    },
    {
      "entropy_step_1": 0.1328,
      "entropy_step_2": 0.1325,
      "entropy_step_3": 0.1078,
      "epoch": 0.11123999881026739,
      "grad_norm": 6.71875,
      "learning_rate": 9.89800388958553e-06,
      "loss": 0.044822913408279416,
      "loss_step_1": 0.0457,
      "loss_step_2": 0.0458,
      "loss_step_3": 0.0451,
      "step": 3740
    },
    {
      "entropy_step_1": 0.1457,
      "entropy_step_2": 0.1454,
      "entropy_step_3": 0.1452,
      "epoch": 0.1115374319621665,
      "grad_norm": 15.1875,
      "learning_rate": 9.897013208063216e-06,
      "loss": 0.053514760732650754,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0432,
      "step": 3750
    },
    {
      "entropy_step_1": 0.1258,
      "entropy_step_2": 0.1273,
      "entropy_step_3": 0.1216,
      "epoch": 0.11183486511406561,
      "grad_norm": 3.984375,
      "learning_rate": 9.896017788613989e-06,
      "loss": 0.045339322090148924,
      "loss_step_1": 0.0449,
      "loss_step_2": 0.0441,
      "loss_step_3": 0.0411,
      "step": 3760
    },
    {
      "entropy_step_1": 0.1266,
      "entropy_step_2": 0.1276,
      "entropy_step_3": 0.1393,
      "epoch": 0.11213229826596473,
      "grad_norm": 9.5625,
      "learning_rate": 9.895017632200934e-06,
      "loss": 0.041536948084831236,
      "loss_step_1": 0.0388,
      "loss_step_2": 0.0387,
      "loss_step_3": 0.0375,
      "step": 3770
    },
    {
      "entropy_step_1": 0.1511,
      "entropy_step_2": 0.1501,
      "entropy_step_3": 0.1575,
      "epoch": 0.11242973141786383,
      "grad_norm": 7.6875,
      "learning_rate": 9.894012739791712e-06,
      "loss": 0.05298884510993958,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0505,
      "loss_step_3": 0.0594,
      "step": 3780
    },
    {
      "entropy_step_1": 0.1728,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1703,
      "epoch": 0.11272716456976295,
      "grad_norm": 13.4375,
      "learning_rate": 9.893003112358573e-06,
      "loss": 0.05069786310195923,
      "loss_step_1": 0.0494,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.057,
      "step": 3790
    },
    {
      "entropy_step_1": 0.1405,
      "entropy_step_2": 0.1405,
      "entropy_step_3": 0.1335,
      "epoch": 0.11302459772166205,
      "grad_norm": 3.875,
      "learning_rate": 9.891988750878341e-06,
      "loss": 0.037297758460044864,
      "loss_step_1": 0.0371,
      "loss_step_2": 0.037,
      "loss_step_3": 0.0363,
      "step": 3800
    },
    {
      "entropy_step_1": 0.1371,
      "entropy_step_2": 0.1381,
      "entropy_step_3": 0.1442,
      "epoch": 0.11332203087356117,
      "grad_norm": 7.53125,
      "learning_rate": 9.89096965633243e-06,
      "loss": 0.044370564818382266,
      "loss_step_1": 0.0397,
      "loss_step_2": 0.0392,
      "loss_step_3": 0.0397,
      "step": 3810
    },
    {
      "entropy_step_1": 0.1157,
      "entropy_step_2": 0.1153,
      "entropy_step_3": 0.1098,
      "epoch": 0.11361946402546028,
      "grad_norm": 3.984375,
      "learning_rate": 9.889945829706824e-06,
      "loss": 0.043782484531402585,
      "loss_step_1": 0.0431,
      "loss_step_2": 0.0428,
      "loss_step_3": 0.0391,
      "step": 3820
    },
    {
      "entropy_step_1": 0.1453,
      "entropy_step_2": 0.1462,
      "entropy_step_3": 0.169,
      "epoch": 0.1139168971773594,
      "grad_norm": 15.5,
      "learning_rate": 9.88891727199209e-06,
      "loss": 0.0477424293756485,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.0485,
      "loss_step_3": 0.055,
      "step": 3830
    },
    {
      "entropy_step_1": 0.135,
      "entropy_step_2": 0.1361,
      "entropy_step_3": 0.1374,
      "epoch": 0.1142143303292585,
      "grad_norm": 9.125,
      "learning_rate": 9.887883984183373e-06,
      "loss": 0.03830094635486603,
      "loss_step_1": 0.0387,
      "loss_step_2": 0.039,
      "loss_step_3": 0.0397,
      "step": 3840
    },
    {
      "entropy_step_1": 0.1254,
      "entropy_step_2": 0.1265,
      "entropy_step_3": 0.12,
      "epoch": 0.11451176348115762,
      "grad_norm": 3.234375,
      "learning_rate": 9.886845967280392e-06,
      "loss": 0.04889472424983978,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.0485,
      "loss_step_3": 0.0375,
      "step": 3850
    },
    {
      "entropy_step_1": 0.1361,
      "entropy_step_2": 0.1363,
      "entropy_step_3": 0.1291,
      "epoch": 0.11480919663305672,
      "grad_norm": 8.25,
      "learning_rate": 9.885803222287442e-06,
      "loss": 0.04206637442111969,
      "loss_step_1": 0.0402,
      "loss_step_2": 0.0402,
      "loss_step_3": 0.0365,
      "step": 3860
    },
    {
      "entropy_step_1": 0.1311,
      "entropy_step_2": 0.132,
      "entropy_step_3": 0.1504,
      "epoch": 0.11510662978495582,
      "grad_norm": 6.125,
      "learning_rate": 9.884755750213394e-06,
      "loss": 0.04366500675678253,
      "loss_step_1": 0.0396,
      "loss_step_2": 0.0396,
      "loss_step_3": 0.0435,
      "step": 3870
    },
    {
      "entropy_step_1": 0.1199,
      "entropy_step_2": 0.1208,
      "entropy_step_3": 0.1239,
      "epoch": 0.11540406293685494,
      "grad_norm": 3.0,
      "learning_rate": 9.88370355207169e-06,
      "loss": 0.0393339604139328,
      "loss_step_1": 0.0364,
      "loss_step_2": 0.0365,
      "loss_step_3": 0.036,
      "step": 3880
    },
    {
      "entropy_step_1": 0.1239,
      "entropy_step_2": 0.1248,
      "entropy_step_3": 0.1084,
      "epoch": 0.11570149608875405,
      "grad_norm": 6.6875,
      "learning_rate": 9.882646628880349e-06,
      "loss": 0.04032142758369446,
      "loss_step_1": 0.0345,
      "loss_step_2": 0.0343,
      "loss_step_3": 0.0292,
      "step": 3890
    },
    {
      "entropy_step_1": 0.1276,
      "entropy_step_2": 0.128,
      "entropy_step_3": 0.122,
      "epoch": 0.11599892924065316,
      "grad_norm": 22.5,
      "learning_rate": 9.881584981661953e-06,
      "loss": 0.042543929815292356,
      "loss_step_1": 0.0421,
      "loss_step_2": 0.0415,
      "loss_step_3": 0.0468,
      "step": 3900
    },
    {
      "entropy_step_1": 0.1466,
      "entropy_step_2": 0.1476,
      "entropy_step_3": 0.1568,
      "epoch": 0.11629636239255227,
      "grad_norm": 6.3125,
      "learning_rate": 9.880518611443666e-06,
      "loss": 0.04578683972358703,
      "loss_step_1": 0.0418,
      "loss_step_2": 0.0413,
      "loss_step_3": 0.05,
      "step": 3910
    },
    {
      "entropy_step_1": 0.1343,
      "entropy_step_2": 0.1351,
      "entropy_step_3": 0.1398,
      "epoch": 0.11659379554445139,
      "grad_norm": 5.59375,
      "learning_rate": 9.879447519257213e-06,
      "loss": 0.04463169574737549,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.0435,
      "loss_step_3": 0.0489,
      "step": 3920
    },
    {
      "entropy_step_1": 0.1596,
      "entropy_step_2": 0.1606,
      "entropy_step_3": 0.1572,
      "epoch": 0.11689122869635049,
      "grad_norm": 4.03125,
      "learning_rate": 9.878371706138892e-06,
      "loss": 0.05086718797683716,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0476,
      "step": 3930
    },
    {
      "entropy_step_1": 0.1301,
      "entropy_step_2": 0.13,
      "entropy_step_3": 0.1258,
      "epoch": 0.11718866184824961,
      "grad_norm": 8.875,
      "learning_rate": 9.877291173129563e-06,
      "loss": 0.039041122794151305,
      "loss_step_1": 0.0332,
      "loss_step_2": 0.0331,
      "loss_step_3": 0.03,
      "step": 3940
    },
    {
      "entropy_step_1": 0.1218,
      "entropy_step_2": 0.1218,
      "entropy_step_3": 0.1258,
      "epoch": 0.11748609500014871,
      "grad_norm": 4.53125,
      "learning_rate": 9.87620592127466e-06,
      "loss": 0.04348518252372742,
      "loss_step_1": 0.0435,
      "loss_step_2": 0.0432,
      "loss_step_3": 0.0405,
      "step": 3950
    },
    {
      "entropy_step_1": 0.1269,
      "entropy_step_2": 0.1271,
      "entropy_step_3": 0.1251,
      "epoch": 0.11778352815204783,
      "grad_norm": 3.828125,
      "learning_rate": 9.875115951624178e-06,
      "loss": 0.0423738420009613,
      "loss_step_1": 0.0404,
      "loss_step_2": 0.0401,
      "loss_step_3": 0.0435,
      "step": 3960
    },
    {
      "entropy_step_1": 0.1458,
      "entropy_step_2": 0.1456,
      "entropy_step_3": 0.1403,
      "epoch": 0.11808096130394694,
      "grad_norm": 5.1875,
      "learning_rate": 9.874021265232678e-06,
      "loss": 0.052425879240036014,
      "loss_step_1": 0.0506,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.0437,
      "step": 3970
    },
    {
      "entropy_step_1": 0.1531,
      "entropy_step_2": 0.1532,
      "entropy_step_3": 0.1689,
      "epoch": 0.11837839445584605,
      "grad_norm": 9.0625,
      "learning_rate": 9.872921863159281e-06,
      "loss": 0.04407007396221161,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0462,
      "loss_step_3": 0.0493,
      "step": 3980
    },
    {
      "entropy_step_1": 0.1491,
      "entropy_step_2": 0.1487,
      "entropy_step_3": 0.1352,
      "epoch": 0.11867582760774516,
      "grad_norm": 15.8125,
      "learning_rate": 9.871817746467677e-06,
      "loss": 0.04245059788227081,
      "loss_step_1": 0.044,
      "loss_step_2": 0.0442,
      "loss_step_3": 0.0435,
      "step": 3990
    },
    {
      "entropy_step_1": 0.1284,
      "entropy_step_2": 0.1279,
      "entropy_step_3": 0.1293,
      "epoch": 0.11897326075964428,
      "grad_norm": 6.0625,
      "learning_rate": 9.870708916226112e-06,
      "loss": 0.04059685170650482,
      "loss_step_1": 0.0429,
      "loss_step_2": 0.0425,
      "loss_step_3": 0.0452,
      "step": 4000
    },
    {
      "epoch": 0.11897326075964428,
      "eval_accuracy": 0.959,
      "eval_f1": 0.9589782995204464,
      "eval_loss": 0.042586516588926315,
      "eval_precision": 0.9599733035102277,
      "eval_recall": 0.959,
      "eval_runtime": 20.0482,
      "eval_samples_per_second": 24.94,
      "eval_steps_per_second": 3.142,
      "step": 4000
    }
  ],
  "logging_steps": 10,
  "max_steps": 33621,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.752320583377261e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}