{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9517860860771542,
  "eval_steps": 1000,
  "global_step": 32000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy_step_1": 0.5574,
      "entropy_step_2": 0.5867,
      "entropy_step_3": 0.5973,
      "epoch": 0.0002974331518991107,
      "grad_norm": 162.92401123046875,
      "learning_rate": 5.350772889417361e-07,
      "loss": 0.3215684652328491,
      "loss_step_1": 0.3252,
      "loss_step_2": 0.3209,
      "loss_step_3": 0.3019,
      "step": 10
    },
    {
      "entropy_step_1": 0.6364,
      "entropy_step_2": 0.6498,
      "entropy_step_3": 0.6497,
      "epoch": 0.0005948663037982214,
      "grad_norm": 138.43966674804688,
      "learning_rate": 1.1296076099881094e-06,
      "loss": 0.2773189306259155,
      "loss_step_1": 0.2791,
      "loss_step_2": 0.2683,
      "loss_step_3": 0.2623,
      "step": 20
    },
    {
      "entropy_step_1": 0.6347,
      "entropy_step_2": 0.6099,
      "entropy_step_3": 0.6167,
      "epoch": 0.000892299455697332,
      "grad_norm": 99.22649383544922,
      "learning_rate": 1.724137931034483e-06,
      "loss": 0.24944260120391845,
      "loss_step_1": 0.2468,
      "loss_step_2": 0.2526,
      "loss_step_3": 0.2453,
      "step": 30
    },
    {
      "entropy_step_1": 0.6302,
      "entropy_step_2": 0.6364,
      "entropy_step_3": 0.6339,
      "epoch": 0.0011897326075964428,
      "grad_norm": 76.41724395751953,
      "learning_rate": 2.318668252080856e-06,
      "loss": 0.23605918884277344,
      "loss_step_1": 0.2367,
      "loss_step_2": 0.2374,
      "loss_step_3": 0.2243,
      "step": 40
    },
    {
      "entropy_step_1": 0.6061,
      "entropy_step_2": 0.6221,
      "entropy_step_3": 0.6245,
      "epoch": 0.0014871657594955534,
      "grad_norm": 64.6061782836914,
      "learning_rate": 2.9131985731272296e-06,
      "loss": 0.23342597484588623,
      "loss_step_1": 0.2432,
      "loss_step_2": 0.2355,
      "loss_step_3": 0.2516,
      "step": 50
    },
    {
      "entropy_step_1": 0.6041,
      "entropy_step_2": 0.6097,
      "entropy_step_3": 0.6189,
      "epoch": 0.001784598911394664,
      "grad_norm": 72.35734558105469,
      "learning_rate": 3.507728894173603e-06,
      "loss": 0.2176830530166626,
      "loss_step_1": 0.2357,
      "loss_step_2": 0.2281,
      "loss_step_3": 0.2322,
      "step": 60
    },
    {
      "entropy_step_1": 0.5959,
      "entropy_step_2": 0.6024,
      "entropy_step_3": 0.6097,
      "epoch": 0.002082032063293775,
      "grad_norm": 71.08319854736328,
      "learning_rate": 4.102259215219977e-06,
      "loss": 0.22477352619171143,
      "loss_step_1": 0.2167,
      "loss_step_2": 0.2139,
      "loss_step_3": 0.2198,
      "step": 70
    },
    {
      "entropy_step_1": 0.5917,
      "entropy_step_2": 0.6043,
      "entropy_step_3": 0.5917,
      "epoch": 0.0023794652151928855,
      "grad_norm": 70.38824462890625,
      "learning_rate": 4.69678953626635e-06,
      "loss": 0.22552130222320557,
      "loss_step_1": 0.2289,
      "loss_step_2": 0.2241,
      "loss_step_3": 0.2012,
      "step": 80
    },
    {
      "entropy_step_1": 0.5699,
      "entropy_step_2": 0.5789,
      "entropy_step_3": 0.5862,
      "epoch": 0.002676898367091996,
      "grad_norm": 78.26751708984375,
      "learning_rate": 5.291319857312723e-06,
      "loss": 0.21276428699493408,
      "loss_step_1": 0.2129,
      "loss_step_2": 0.21,
      "loss_step_3": 0.2113,
      "step": 90
    },
    {
      "entropy_step_1": 0.5897,
      "entropy_step_2": 0.5947,
      "entropy_step_3": 0.6,
      "epoch": 0.002974331518991107,
      "grad_norm": 69.95088958740234,
      "learning_rate": 5.885850178359096e-06,
      "loss": 0.21249134540557862,
      "loss_step_1": 0.2177,
      "loss_step_2": 0.217,
      "loss_step_3": 0.2222,
      "step": 100
    },
    {
      "entropy_step_1": 0.5877,
      "entropy_step_2": 0.5869,
      "entropy_step_3": 0.5894,
      "epoch": 0.0032717646708902174,
      "grad_norm": 52.85276412963867,
      "learning_rate": 6.48038049940547e-06,
      "loss": 0.20930936336517333,
      "loss_step_1": 0.2033,
      "loss_step_2": 0.2034,
      "loss_step_3": 0.2047,
      "step": 110
    },
    {
      "entropy_step_1": 0.5727,
      "entropy_step_2": 0.5771,
      "entropy_step_3": 0.5672,
      "epoch": 0.003569197822789328,
      "grad_norm": 52.416690826416016,
      "learning_rate": 7.0749108204518435e-06,
      "loss": 0.20670053958892823,
      "loss_step_1": 0.1978,
      "loss_step_2": 0.198,
      "loss_step_3": 0.1824,
      "step": 120
    },
    {
      "entropy_step_1": 0.5537,
      "entropy_step_2": 0.5498,
      "entropy_step_3": 0.5684,
      "epoch": 0.0038666309746884387,
      "grad_norm": 58.702884674072266,
      "learning_rate": 7.669441141498216e-06,
      "loss": 0.20105493068695068,
      "loss_step_1": 0.1938,
      "loss_step_2": 0.1937,
      "loss_step_3": 0.1967,
      "step": 130
    },
    {
      "entropy_step_1": 0.5618,
      "entropy_step_2": 0.5576,
      "entropy_step_3": 0.5678,
      "epoch": 0.00416406412658755,
      "grad_norm": 77.5804443359375,
      "learning_rate": 8.263971462544589e-06,
      "loss": 0.2052671194076538,
      "loss_step_1": 0.1971,
      "loss_step_2": 0.1957,
      "loss_step_3": 0.2072,
      "step": 140
    },
    {
      "entropy_step_1": 0.5501,
      "entropy_step_2": 0.5504,
      "entropy_step_3": 0.5201,
      "epoch": 0.00446149727848666,
      "grad_norm": 56.97880935668945,
      "learning_rate": 8.858501783590963e-06,
      "loss": 0.19344786405563355,
      "loss_step_1": 0.1905,
      "loss_step_2": 0.1857,
      "loss_step_3": 0.1724,
      "step": 150
    },
    {
      "entropy_step_1": 0.5799,
      "entropy_step_2": 0.5794,
      "entropy_step_3": 0.5708,
      "epoch": 0.004758930430385771,
      "grad_norm": 82.2699966430664,
      "learning_rate": 9.453032104637338e-06,
      "loss": 0.21766309738159179,
      "loss_step_1": 0.2142,
      "loss_step_2": 0.2108,
      "loss_step_3": 0.2131,
      "step": 160
    },
    {
      "entropy_step_1": 0.5607,
      "entropy_step_2": 0.5618,
      "entropy_step_3": 0.5508,
      "epoch": 0.005056363582284881,
      "grad_norm": 42.434654235839844,
      "learning_rate": 1.004756242568371e-05,
      "loss": 0.19557396173477173,
      "loss_step_1": 0.1937,
      "loss_step_2": 0.1933,
      "loss_step_3": 0.1833,
      "step": 170
    },
    {
      "entropy_step_1": 0.5562,
      "entropy_step_2": 0.559,
      "entropy_step_3": 0.5618,
      "epoch": 0.005353796734183992,
      "grad_norm": 55.32823944091797,
      "learning_rate": 1.0642092746730085e-05,
      "loss": 0.2103808879852295,
      "loss_step_1": 0.202,
      "loss_step_2": 0.2028,
      "loss_step_3": 0.1974,
      "step": 180
    },
    {
      "entropy_step_1": 0.59,
      "entropy_step_2": 0.5933,
      "entropy_step_3": 0.6043,
      "epoch": 0.0056512298860831025,
      "grad_norm": 48.03270721435547,
      "learning_rate": 1.1236623067776457e-05,
      "loss": 0.1974504590034485,
      "loss_step_1": 0.2117,
      "loss_step_2": 0.2083,
      "loss_step_3": 0.2147,
      "step": 190
    },
    {
      "entropy_step_1": 0.569,
      "entropy_step_2": 0.5735,
      "entropy_step_3": 0.5693,
      "epoch": 0.005948663037982214,
      "grad_norm": 38.456172943115234,
      "learning_rate": 1.183115338882283e-05,
      "loss": 0.20741612911224366,
      "loss_step_1": 0.206,
      "loss_step_2": 0.205,
      "loss_step_3": 0.1927,
      "step": 200
    },
    {
      "entropy_step_1": 0.5561,
      "entropy_step_2": 0.5584,
      "entropy_step_3": 0.559,
      "epoch": 0.006246096189881324,
      "grad_norm": 46.90557861328125,
      "learning_rate": 1.2425683709869204e-05,
      "loss": 0.18477752208709716,
      "loss_step_1": 0.183,
      "loss_step_2": 0.1849,
      "loss_step_3": 0.1847,
      "step": 210
    },
    {
      "entropy_step_1": 0.5877,
      "entropy_step_2": 0.5827,
      "entropy_step_3": 0.591,
      "epoch": 0.006543529341780435,
      "grad_norm": 44.64488220214844,
      "learning_rate": 1.3020214030915579e-05,
      "loss": 0.21059150695800782,
      "loss_step_1": 0.2116,
      "loss_step_2": 0.2111,
      "loss_step_3": 0.2005,
      "step": 220
    },
    {
      "entropy_step_1": 0.5666,
      "entropy_step_2": 0.5555,
      "entropy_step_3": 0.5535,
      "epoch": 0.006840962493679546,
      "grad_norm": 56.662330627441406,
      "learning_rate": 1.361474435196195e-05,
      "loss": 0.20618062019348143,
      "loss_step_1": 0.2025,
      "loss_step_2": 0.2042,
      "loss_step_3": 0.1987,
      "step": 230
    },
    {
      "entropy_step_1": 0.5708,
      "entropy_step_2": 0.5715,
      "entropy_step_3": 0.5785,
      "epoch": 0.007138395645578656,
      "grad_norm": 42.134891510009766,
      "learning_rate": 1.4209274673008324e-05,
      "loss": 0.1935202956199646,
      "loss_step_1": 0.1966,
      "loss_step_2": 0.1948,
      "loss_step_3": 0.1927,
      "step": 240
    },
    {
      "entropy_step_1": 0.5654,
      "entropy_step_2": 0.5638,
      "entropy_step_3": 0.5576,
      "epoch": 0.007435828797477767,
      "grad_norm": 81.76813507080078,
      "learning_rate": 1.4803804994054699e-05,
      "loss": 0.19635610580444335,
      "loss_step_1": 0.1933,
      "loss_step_2": 0.1925,
      "loss_step_3": 0.1918,
      "step": 250
    },
    {
      "entropy_step_1": 0.5573,
      "entropy_step_2": 0.567,
      "entropy_step_3": 0.5738,
      "epoch": 0.007733261949376877,
      "grad_norm": 28.948509216308594,
      "learning_rate": 1.539833531510107e-05,
      "loss": 0.19321547746658324,
      "loss_step_1": 0.1932,
      "loss_step_2": 0.1936,
      "loss_step_3": 0.1991,
      "step": 260
    },
    {
      "entropy_step_1": 0.5584,
      "entropy_step_2": 0.559,
      "entropy_step_3": 0.5613,
      "epoch": 0.008030695101275988,
      "grad_norm": 31.79094886779785,
      "learning_rate": 1.5992865636147444e-05,
      "loss": 0.19771844148635864,
      "loss_step_1": 0.1971,
      "loss_step_2": 0.1988,
      "loss_step_3": 0.1991,
      "step": 270
    },
    {
      "entropy_step_1": 0.5779,
      "entropy_step_2": 0.583,
      "entropy_step_3": 0.5728,
      "epoch": 0.0083281282531751,
      "grad_norm": 29.815677642822266,
      "learning_rate": 1.658739595719382e-05,
      "loss": 0.19258413314819336,
      "loss_step_1": 0.1946,
      "loss_step_2": 0.1931,
      "loss_step_3": 0.1849,
      "step": 280
    },
    {
      "entropy_step_1": 0.5501,
      "entropy_step_2": 0.5524,
      "entropy_step_3": 0.5647,
      "epoch": 0.008625561405074209,
      "grad_norm": 54.16142654418945,
      "learning_rate": 1.718192627824019e-05,
      "loss": 0.19478131532669068,
      "loss_step_1": 0.1948,
      "loss_step_2": 0.1946,
      "loss_step_3": 0.206,
      "step": 290
    },
    {
      "entropy_step_1": 0.5446,
      "entropy_step_2": 0.5507,
      "entropy_step_3": 0.5415,
      "epoch": 0.00892299455697332,
      "grad_norm": 36.75969314575195,
      "learning_rate": 1.7776456599286565e-05,
      "loss": 0.1888433575630188,
      "loss_step_1": 0.1771,
      "loss_step_2": 0.177,
      "loss_step_3": 0.1692,
      "step": 300
    },
    {
      "entropy_step_1": 0.5554,
      "entropy_step_2": 0.5559,
      "entropy_step_3": 0.5548,
      "epoch": 0.009220427708872431,
      "grad_norm": 57.10541915893555,
      "learning_rate": 1.8370986920332938e-05,
      "loss": 0.19630483388900757,
      "loss_step_1": 0.197,
      "loss_step_2": 0.1967,
      "loss_step_3": 0.196,
      "step": 310
    },
    {
      "entropy_step_1": 0.5826,
      "entropy_step_2": 0.5804,
      "entropy_step_3": 0.5792,
      "epoch": 0.009517860860771542,
      "grad_norm": 40.1431884765625,
      "learning_rate": 1.896551724137931e-05,
      "loss": 0.20554792881011963,
      "loss_step_1": 0.2015,
      "loss_step_2": 0.2016,
      "loss_step_3": 0.1977,
      "step": 320
    },
    {
      "entropy_step_1": 0.559,
      "entropy_step_2": 0.5525,
      "entropy_step_3": 0.5781,
      "epoch": 0.009815294012670651,
      "grad_norm": 29.274396896362305,
      "learning_rate": 1.9560047562425683e-05,
      "loss": 0.19082272052764893,
      "loss_step_1": 0.1892,
      "loss_step_2": 0.1893,
      "loss_step_3": 0.202,
      "step": 330
    },
    {
      "entropy_step_1": 0.5422,
      "entropy_step_2": 0.5361,
      "entropy_step_3": 0.5295,
      "epoch": 0.010112727164569763,
      "grad_norm": 37.429935455322266,
      "learning_rate": 2.015457788347206e-05,
      "loss": 0.18911318778991698,
      "loss_step_1": 0.1846,
      "loss_step_2": 0.1871,
      "loss_step_3": 0.1825,
      "step": 340
    },
    {
      "entropy_step_1": 0.5332,
      "entropy_step_2": 0.5442,
      "entropy_step_3": 0.5563,
      "epoch": 0.010410160316468874,
      "grad_norm": 47.78837203979492,
      "learning_rate": 2.074910820451843e-05,
      "loss": 0.19185755252838135,
      "loss_step_1": 0.1794,
      "loss_step_2": 0.1795,
      "loss_step_3": 0.1924,
      "step": 350
    },
    {
      "entropy_step_1": 0.5673,
      "entropy_step_2": 0.5728,
      "entropy_step_3": 0.5753,
      "epoch": 0.010707593468367985,
      "grad_norm": 28.353885650634766,
      "learning_rate": 2.1343638525564805e-05,
      "loss": 0.1985328435897827,
      "loss_step_1": 0.2041,
      "loss_step_2": 0.2049,
      "loss_step_3": 0.2027,
      "step": 360
    },
    {
      "entropy_step_1": 0.5535,
      "entropy_step_2": 0.5447,
      "entropy_step_3": 0.5335,
      "epoch": 0.011005026620267096,
      "grad_norm": 46.05038070678711,
      "learning_rate": 2.1938168846611178e-05,
      "loss": 0.19938653707504272,
      "loss_step_1": 0.1959,
      "loss_step_2": 0.1973,
      "loss_step_3": 0.1927,
      "step": 370
    },
    {
      "entropy_step_1": 0.5496,
      "entropy_step_2": 0.5464,
      "entropy_step_3": 0.546,
      "epoch": 0.011302459772166205,
      "grad_norm": 54.46418380737305,
      "learning_rate": 2.253269916765755e-05,
      "loss": 0.18759881258010863,
      "loss_step_1": 0.1861,
      "loss_step_2": 0.1883,
      "loss_step_3": 0.1956,
      "step": 380
    },
    {
      "entropy_step_1": 0.5379,
      "entropy_step_2": 0.5481,
      "entropy_step_3": 0.565,
      "epoch": 0.011599892924065316,
      "grad_norm": 42.184871673583984,
      "learning_rate": 2.3127229488703926e-05,
      "loss": 0.18169955015182496,
      "loss_step_1": 0.1803,
      "loss_step_2": 0.1792,
      "loss_step_3": 0.185,
      "step": 390
    },
    {
      "entropy_step_1": 0.528,
      "entropy_step_2": 0.5326,
      "entropy_step_3": 0.5504,
      "epoch": 0.011897326075964427,
      "grad_norm": 39.868492126464844,
      "learning_rate": 2.37217598097503e-05,
      "loss": 0.1739505410194397,
      "loss_step_1": 0.1702,
      "loss_step_2": 0.172,
      "loss_step_3": 0.1728,
      "step": 400
    },
    {
      "entropy_step_1": 0.5223,
      "entropy_step_2": 0.5218,
      "entropy_step_3": 0.5308,
      "epoch": 0.012194759227863538,
      "grad_norm": 48.86769485473633,
      "learning_rate": 2.4316290130796672e-05,
      "loss": 0.18009419441223146,
      "loss_step_1": 0.183,
      "loss_step_2": 0.1835,
      "loss_step_3": 0.1877,
      "step": 410
    },
    {
      "entropy_step_1": 0.5146,
      "entropy_step_2": 0.5166,
      "entropy_step_3": 0.5248,
      "epoch": 0.012492192379762648,
      "grad_norm": 59.004981994628906,
      "learning_rate": 2.4910820451843044e-05,
      "loss": 0.16364353895187378,
      "loss_step_1": 0.1589,
      "loss_step_2": 0.1587,
      "loss_step_3": 0.1617,
      "step": 420
    },
    {
      "entropy_step_1": 0.4767,
      "entropy_step_2": 0.4816,
      "entropy_step_3": 0.4928,
      "epoch": 0.012789625531661759,
      "grad_norm": 67.21807861328125,
      "learning_rate": 2.550535077288942e-05,
      "loss": 0.15321747064590455,
      "loss_step_1": 0.1493,
      "loss_step_2": 0.15,
      "loss_step_3": 0.1525,
      "step": 430
    },
    {
      "entropy_step_1": 0.4931,
      "entropy_step_2": 0.4884,
      "entropy_step_3": 0.5137,
      "epoch": 0.01308705868356087,
      "grad_norm": 63.38396453857422,
      "learning_rate": 2.609988109393579e-05,
      "loss": 0.15500762462615966,
      "loss_step_1": 0.1582,
      "loss_step_2": 0.1602,
      "loss_step_3": 0.1598,
      "step": 440
    },
    {
      "entropy_step_1": 0.5009,
      "entropy_step_2": 0.5029,
      "entropy_step_3": 0.5167,
      "epoch": 0.01338449183545998,
      "grad_norm": 110.44959259033203,
      "learning_rate": 2.669441141498217e-05,
      "loss": 0.1509174585342407,
      "loss_step_1": 0.1583,
      "loss_step_2": 0.1567,
      "loss_step_3": 0.1663,
      "step": 450
    },
    {
      "entropy_step_1": 0.4673,
      "entropy_step_2": 0.4673,
      "entropy_step_3": 0.4752,
      "epoch": 0.013681924987359092,
      "grad_norm": 78.8842544555664,
      "learning_rate": 2.728894173602854e-05,
      "loss": 0.15575270652770995,
      "loss_step_1": 0.1538,
      "loss_step_2": 0.1551,
      "loss_step_3": 0.1623,
      "step": 460
    },
    {
      "entropy_step_1": 0.4462,
      "entropy_step_2": 0.4515,
      "entropy_step_3": 0.4481,
      "epoch": 0.013979358139258201,
      "grad_norm": 72.851318359375,
      "learning_rate": 2.788347205707491e-05,
      "loss": 0.14301614761352538,
      "loss_step_1": 0.1389,
      "loss_step_2": 0.138,
      "loss_step_3": 0.1252,
      "step": 470
    },
    {
      "entropy_step_1": 0.4504,
      "entropy_step_2": 0.4502,
      "entropy_step_3": 0.4467,
      "epoch": 0.014276791291157312,
      "grad_norm": 114.49141693115234,
      "learning_rate": 2.8478002378121287e-05,
      "loss": 0.15186270475387573,
      "loss_step_1": 0.1524,
      "loss_step_2": 0.151,
      "loss_step_3": 0.1458,
      "step": 480
    },
    {
      "entropy_step_1": 0.3992,
      "entropy_step_2": 0.407,
      "entropy_step_3": 0.4194,
      "epoch": 0.014574224443056423,
      "grad_norm": 109.79158782958984,
      "learning_rate": 2.907253269916766e-05,
      "loss": 0.13704211711883546,
      "loss_step_1": 0.1289,
      "loss_step_2": 0.1286,
      "loss_step_3": 0.1288,
      "step": 490
    },
    {
      "entropy_step_1": 0.4395,
      "entropy_step_2": 0.4432,
      "entropy_step_3": 0.4492,
      "epoch": 0.014871657594955534,
      "grad_norm": 86.21638488769531,
      "learning_rate": 2.966706302021403e-05,
      "loss": 0.14735809564590455,
      "loss_step_1": 0.1358,
      "loss_step_2": 0.1349,
      "loss_step_3": 0.127,
      "step": 500
    },
    {
      "entropy_step_1": 0.4342,
      "entropy_step_2": 0.4302,
      "entropy_step_3": 0.4503,
      "epoch": 0.015169090746854644,
      "grad_norm": 53.078609466552734,
      "learning_rate": 3.026159334126041e-05,
      "loss": 0.12529877424240113,
      "loss_step_1": 0.1301,
      "loss_step_2": 0.1304,
      "loss_step_3": 0.1288,
      "step": 510
    },
    {
      "entropy_step_1": 0.4351,
      "entropy_step_2": 0.4383,
      "entropy_step_3": 0.435,
      "epoch": 0.015466523898753755,
      "grad_norm": 90.13272857666016,
      "learning_rate": 3.085612366230678e-05,
      "loss": 0.1448341965675354,
      "loss_step_1": 0.1409,
      "loss_step_2": 0.141,
      "loss_step_3": 0.1394,
      "step": 520
    },
    {
      "entropy_step_1": 0.4243,
      "entropy_step_2": 0.4273,
      "entropy_step_3": 0.4297,
      "epoch": 0.015763957050652866,
      "grad_norm": 160.3775634765625,
      "learning_rate": 3.1450653983353154e-05,
      "loss": 0.12917875051498412,
      "loss_step_1": 0.13,
      "loss_step_2": 0.1299,
      "loss_step_3": 0.1341,
      "step": 530
    },
    {
      "entropy_step_1": 0.4311,
      "entropy_step_2": 0.4398,
      "entropy_step_3": 0.4745,
      "epoch": 0.016061390202551977,
      "grad_norm": 145.440673828125,
      "learning_rate": 3.204518430439953e-05,
      "loss": 0.140031898021698,
      "loss_step_1": 0.1436,
      "loss_step_2": 0.1431,
      "loss_step_3": 0.1626,
      "step": 540
    },
    {
      "entropy_step_1": 0.4004,
      "entropy_step_2": 0.4065,
      "entropy_step_3": 0.4005,
      "epoch": 0.016358823354451088,
      "grad_norm": 68.19512939453125,
      "learning_rate": 3.26397146254459e-05,
      "loss": 0.12515480518341066,
      "loss_step_1": 0.1195,
      "loss_step_2": 0.1208,
      "loss_step_3": 0.1265,
      "step": 550
    },
    {
      "entropy_step_1": 0.4173,
      "entropy_step_2": 0.4177,
      "entropy_step_3": 0.4328,
      "epoch": 0.0166562565063502,
      "grad_norm": 90.94459533691406,
      "learning_rate": 3.323424494649227e-05,
      "loss": 0.1318908452987671,
      "loss_step_1": 0.1285,
      "loss_step_2": 0.1302,
      "loss_step_3": 0.1442,
      "step": 560
    },
    {
      "entropy_step_1": 0.4003,
      "entropy_step_2": 0.4086,
      "entropy_step_3": 0.4292,
      "epoch": 0.01695368965824931,
      "grad_norm": 88.39466857910156,
      "learning_rate": 3.3828775267538645e-05,
      "loss": 0.1357468843460083,
      "loss_step_1": 0.1359,
      "loss_step_2": 0.1337,
      "loss_step_3": 0.146,
      "step": 570
    },
    {
      "entropy_step_1": 0.4233,
      "entropy_step_2": 0.4313,
      "entropy_step_3": 0.4292,
      "epoch": 0.017251122810148418,
      "grad_norm": 98.74026489257812,
      "learning_rate": 3.442330558858502e-05,
      "loss": 0.12442314624786377,
      "loss_step_1": 0.1246,
      "loss_step_2": 0.1235,
      "loss_step_3": 0.1173,
      "step": 580
    },
    {
      "entropy_step_1": 0.4181,
      "entropy_step_2": 0.42,
      "entropy_step_3": 0.4232,
      "epoch": 0.01754855596204753,
      "grad_norm": 126.22911834716797,
      "learning_rate": 3.501783590963139e-05,
      "loss": 0.12905603647232056,
      "loss_step_1": 0.1267,
      "loss_step_2": 0.127,
      "loss_step_3": 0.1304,
      "step": 590
    },
    {
      "entropy_step_1": 0.386,
      "entropy_step_2": 0.385,
      "entropy_step_3": 0.3641,
      "epoch": 0.01784598911394664,
      "grad_norm": 94.7985610961914,
      "learning_rate": 3.5612366230677766e-05,
      "loss": 0.11679408550262452,
      "loss_step_1": 0.1191,
      "loss_step_2": 0.1176,
      "loss_step_3": 0.11,
      "step": 600
    },
    {
      "entropy_step_1": 0.3727,
      "entropy_step_2": 0.3709,
      "entropy_step_3": 0.3859,
      "epoch": 0.01814342226584575,
      "grad_norm": 83.83550262451172,
      "learning_rate": 3.620689655172414e-05,
      "loss": 0.11542613506317138,
      "loss_step_1": 0.1087,
      "loss_step_2": 0.109,
      "loss_step_3": 0.1154,
      "step": 610
    },
    {
      "entropy_step_1": 0.399,
      "entropy_step_2": 0.3999,
      "entropy_step_3": 0.3965,
      "epoch": 0.018440855417744862,
      "grad_norm": 129.48643493652344,
      "learning_rate": 3.680142687277051e-05,
      "loss": 0.1190953254699707,
      "loss_step_1": 0.1168,
      "loss_step_2": 0.116,
      "loss_step_3": 0.1226,
      "step": 620
    },
    {
      "entropy_step_1": 0.3631,
      "entropy_step_2": 0.3578,
      "entropy_step_3": 0.3419,
      "epoch": 0.018738288569643973,
      "grad_norm": 106.9564208984375,
      "learning_rate": 3.739595719381689e-05,
      "loss": 0.10964417457580566,
      "loss_step_1": 0.1011,
      "loss_step_2": 0.1015,
      "loss_step_3": 0.0901,
      "step": 630
    },
    {
      "entropy_step_1": 0.3591,
      "entropy_step_2": 0.3585,
      "entropy_step_3": 0.3671,
      "epoch": 0.019035721721543084,
      "grad_norm": 72.85526275634766,
      "learning_rate": 3.799048751486326e-05,
      "loss": 0.11281304359436035,
      "loss_step_1": 0.1128,
      "loss_step_2": 0.1122,
      "loss_step_3": 0.1283,
      "step": 640
    },
    {
      "entropy_step_1": 0.3837,
      "entropy_step_2": 0.3853,
      "entropy_step_3": 0.3958,
      "epoch": 0.019333154873442195,
      "grad_norm": 56.292606353759766,
      "learning_rate": 3.858501783590963e-05,
      "loss": 0.11508851051330567,
      "loss_step_1": 0.1223,
      "loss_step_2": 0.1229,
      "loss_step_3": 0.1231,
      "step": 650
    },
    {
      "entropy_step_1": 0.3848,
      "entropy_step_2": 0.3897,
      "entropy_step_3": 0.3771,
      "epoch": 0.019630588025341303,
      "grad_norm": 91.88412475585938,
      "learning_rate": 3.917954815695601e-05,
      "loss": 0.12149256467819214,
      "loss_step_1": 0.1265,
      "loss_step_2": 0.1265,
      "loss_step_3": 0.1181,
      "step": 660
    },
    {
      "entropy_step_1": 0.3791,
      "entropy_step_2": 0.3882,
      "entropy_step_3": 0.3907,
      "epoch": 0.019928021177240414,
      "grad_norm": 180.7445526123047,
      "learning_rate": 3.977407847800238e-05,
      "loss": 0.1166884183883667,
      "loss_step_1": 0.1271,
      "loss_step_2": 0.1286,
      "loss_step_3": 0.1147,
      "step": 670
    },
    {
      "entropy_step_1": 0.3575,
      "entropy_step_2": 0.3594,
      "entropy_step_3": 0.3374,
      "epoch": 0.020225454329139525,
      "grad_norm": 71.66834259033203,
      "learning_rate": 4.0368608799048755e-05,
      "loss": 0.11815319061279297,
      "loss_step_1": 0.1161,
      "loss_step_2": 0.1154,
      "loss_step_3": 0.1084,
      "step": 680
    },
    {
      "entropy_step_1": 0.3448,
      "entropy_step_2": 0.3502,
      "entropy_step_3": 0.3639,
      "epoch": 0.020522887481038636,
      "grad_norm": 64.67585754394531,
      "learning_rate": 4.096313912009513e-05,
      "loss": 0.10558825731277466,
      "loss_step_1": 0.1094,
      "loss_step_2": 0.1095,
      "loss_step_3": 0.1099,
      "step": 690
    },
    {
      "entropy_step_1": 0.3516,
      "entropy_step_2": 0.3555,
      "entropy_step_3": 0.3647,
      "epoch": 0.020820320632937747,
      "grad_norm": 96.5094223022461,
      "learning_rate": 4.15576694411415e-05,
      "loss": 0.11291162967681885,
      "loss_step_1": 0.1161,
      "loss_step_2": 0.1158,
      "loss_step_3": 0.1154,
      "step": 700
    },
    {
      "entropy_step_1": 0.3361,
      "entropy_step_2": 0.3397,
      "entropy_step_3": 0.3232,
      "epoch": 0.021117753784836858,
      "grad_norm": 108.14176177978516,
      "learning_rate": 4.215219976218787e-05,
      "loss": 0.10692033767700196,
      "loss_step_1": 0.0992,
      "loss_step_2": 0.1002,
      "loss_step_3": 0.0899,
      "step": 710
    },
    {
      "entropy_step_1": 0.3333,
      "entropy_step_2": 0.3314,
      "entropy_step_3": 0.3467,
      "epoch": 0.02141518693673597,
      "grad_norm": 82.54590606689453,
      "learning_rate": 4.2746730083234245e-05,
      "loss": 0.11296931505203248,
      "loss_step_1": 0.1085,
      "loss_step_2": 0.1083,
      "loss_step_3": 0.1022,
      "step": 720
    },
    {
      "entropy_step_1": 0.3424,
      "entropy_step_2": 0.34,
      "entropy_step_3": 0.3394,
      "epoch": 0.02171262008863508,
      "grad_norm": 65.26094055175781,
      "learning_rate": 4.334126040428062e-05,
      "loss": 0.10505881309509277,
      "loss_step_1": 0.1012,
      "loss_step_2": 0.1017,
      "loss_step_3": 0.0968,
      "step": 730
    },
    {
      "entropy_step_1": 0.3465,
      "entropy_step_2": 0.3467,
      "entropy_step_3": 0.339,
      "epoch": 0.02201005324053419,
      "grad_norm": 92.81388092041016,
      "learning_rate": 4.393579072532699e-05,
      "loss": 0.1079214334487915,
      "loss_step_1": 0.1027,
      "loss_step_2": 0.1021,
      "loss_step_3": 0.0918,
      "step": 740
    },
    {
      "entropy_step_1": 0.3362,
      "entropy_step_2": 0.3339,
      "entropy_step_3": 0.3333,
      "epoch": 0.0223074863924333,
      "grad_norm": 100.06945037841797,
      "learning_rate": 4.453032104637337e-05,
      "loss": 0.10010629892349243,
      "loss_step_1": 0.0955,
      "loss_step_2": 0.095,
      "loss_step_3": 0.088,
      "step": 750
    },
    {
      "entropy_step_1": 0.3518,
      "entropy_step_2": 0.3405,
      "entropy_step_3": 0.3375,
      "epoch": 0.02260491954433241,
      "grad_norm": 81.24453735351562,
      "learning_rate": 4.512485136741974e-05,
      "loss": 0.11146723031997681,
      "loss_step_1": 0.1029,
      "loss_step_2": 0.1032,
      "loss_step_3": 0.0996,
      "step": 760
    },
    {
      "entropy_step_1": 0.3444,
      "entropy_step_2": 0.3323,
      "entropy_step_3": 0.3336,
      "epoch": 0.02290235269623152,
      "grad_norm": 57.85712814331055,
      "learning_rate": 4.571938168846611e-05,
      "loss": 0.09790316820144654,
      "loss_step_1": 0.092,
      "loss_step_2": 0.0925,
      "loss_step_3": 0.0941,
      "step": 770
    },
    {
      "entropy_step_1": 0.3562,
      "entropy_step_2": 0.3498,
      "entropy_step_3": 0.3641,
      "epoch": 0.023199785848130632,
      "grad_norm": 65.9686050415039,
      "learning_rate": 4.631391200951249e-05,
      "loss": 0.10945267677307129,
      "loss_step_1": 0.108,
      "loss_step_2": 0.1075,
      "loss_step_3": 0.1085,
      "step": 780
    },
    {
      "entropy_step_1": 0.3487,
      "entropy_step_2": 0.3444,
      "entropy_step_3": 0.3233,
      "epoch": 0.023497219000029743,
      "grad_norm": 96.999755859375,
      "learning_rate": 4.690844233055886e-05,
      "loss": 0.0960654854774475,
      "loss_step_1": 0.1006,
      "loss_step_2": 0.1,
      "loss_step_3": 0.0915,
      "step": 790
    },
    {
      "entropy_step_1": 0.3159,
      "entropy_step_2": 0.3102,
      "entropy_step_3": 0.3147,
      "epoch": 0.023794652151928854,
      "grad_norm": 95.82238006591797,
      "learning_rate": 4.7502972651605234e-05,
      "loss": 0.09667249321937561,
      "loss_step_1": 0.1001,
      "loss_step_2": 0.1009,
      "loss_step_3": 0.098,
      "step": 800
    },
    {
      "entropy_step_1": 0.305,
      "entropy_step_2": 0.3036,
      "entropy_step_3": 0.317,
      "epoch": 0.024092085303827965,
      "grad_norm": 289.61407470703125,
      "learning_rate": 4.809750297265161e-05,
      "loss": 0.10215064287185668,
      "loss_step_1": 0.1043,
      "loss_step_2": 0.1035,
      "loss_step_3": 0.1228,
      "step": 810
    },
    {
      "entropy_step_1": 0.3302,
      "entropy_step_2": 0.336,
      "entropy_step_3": 0.3547,
      "epoch": 0.024389518455727077,
      "grad_norm": 74.55059051513672,
      "learning_rate": 4.869203329369798e-05,
      "loss": 0.10470545291900635,
      "loss_step_1": 0.1038,
      "loss_step_2": 0.1041,
      "loss_step_3": 0.1046,
      "step": 820
    },
    {
      "entropy_step_1": 0.3399,
      "entropy_step_2": 0.3423,
      "entropy_step_3": 0.3354,
      "epoch": 0.024686951607626188,
      "grad_norm": 79.88460540771484,
      "learning_rate": 4.9286563614744355e-05,
      "loss": 0.0983396589756012,
      "loss_step_1": 0.1004,
      "loss_step_2": 0.0999,
      "loss_step_3": 0.0998,
      "step": 830
    },
    {
      "entropy_step_1": 0.3104,
      "entropy_step_2": 0.3102,
      "entropy_step_3": 0.2999,
      "epoch": 0.024984384759525295,
      "grad_norm": 86.53710174560547,
      "learning_rate": 4.988109393579073e-05,
      "loss": 0.10172228813171387,
      "loss_step_1": 0.0997,
      "loss_step_2": 0.0985,
      "loss_step_3": 0.0933,
      "step": 840
    },
    {
      "entropy_step_1": 0.3131,
      "entropy_step_2": 0.3127,
      "entropy_step_3": 0.3295,
      "epoch": 0.025281817911424406,
      "grad_norm": 100.49166870117188,
      "learning_rate": 5.047562425683711e-05,
      "loss": 0.09528225660324097,
      "loss_step_1": 0.098,
      "loss_step_2": 0.0979,
      "loss_step_3": 0.1039,
      "step": 850
    },
    {
      "entropy_step_1": 0.3373,
      "entropy_step_2": 0.335,
      "entropy_step_3": 0.3383,
      "epoch": 0.025579251063323517,
      "grad_norm": 51.40016555786133,
      "learning_rate": 5.107015457788348e-05,
      "loss": 0.09736132025718688,
      "loss_step_1": 0.0997,
      "loss_step_2": 0.0996,
      "loss_step_3": 0.1014,
      "step": 860
    },
    {
      "entropy_step_1": 0.2754,
      "entropy_step_2": 0.2778,
      "entropy_step_3": 0.2968,
      "epoch": 0.02587668421522263,
      "grad_norm": 130.0742950439453,
      "learning_rate": 5.1664684898929846e-05,
      "loss": 0.09698017239570618,
      "loss_step_1": 0.0968,
      "loss_step_2": 0.096,
      "loss_step_3": 0.1089,
      "step": 870
    },
    {
      "entropy_step_1": 0.3528,
      "entropy_step_2": 0.3576,
      "entropy_step_3": 0.3527,
      "epoch": 0.02617411736712174,
      "grad_norm": 124.55671691894531,
      "learning_rate": 5.225921521997622e-05,
      "loss": 0.09832508563995361,
      "loss_step_1": 0.1076,
      "loss_step_2": 0.1078,
      "loss_step_3": 0.097,
      "step": 880
    },
    {
      "entropy_step_1": 0.3168,
      "entropy_step_2": 0.3133,
      "entropy_step_3": 0.3232,
      "epoch": 0.02647155051902085,
      "grad_norm": 53.34364318847656,
      "learning_rate": 5.285374554102259e-05,
      "loss": 0.10538835525512695,
      "loss_step_1": 0.0959,
      "loss_step_2": 0.0967,
      "loss_step_3": 0.1027,
      "step": 890
    },
    {
      "entropy_step_1": 0.2644,
      "entropy_step_2": 0.2555,
      "entropy_step_3": 0.265,
      "epoch": 0.02676898367091996,
      "grad_norm": 93.38742065429688,
      "learning_rate": 5.344827586206896e-05,
      "loss": 0.08695085644721985,
      "loss_step_1": 0.0796,
      "loss_step_2": 0.0793,
      "loss_step_3": 0.0808,
      "step": 900
    },
    {
      "entropy_step_1": 0.3324,
      "entropy_step_2": 0.3215,
      "entropy_step_3": 0.3343,
      "epoch": 0.027066416822819073,
      "grad_norm": 145.6822967529297,
      "learning_rate": 5.4042806183115343e-05,
      "loss": 0.11273114681243897,
      "loss_step_1": 0.1147,
      "loss_step_2": 0.1149,
      "loss_step_3": 0.1237,
      "step": 910
    },
    {
      "entropy_step_1": 0.3032,
      "entropy_step_2": 0.3016,
      "entropy_step_3": 0.3137,
      "epoch": 0.027363849974718184,
      "grad_norm": 161.3321533203125,
      "learning_rate": 5.463733650416172e-05,
      "loss": 0.09883949756622315,
      "loss_step_1": 0.0893,
      "loss_step_2": 0.0892,
      "loss_step_3": 0.0998,
      "step": 920
    },
    {
      "entropy_step_1": 0.2993,
      "entropy_step_2": 0.3085,
      "entropy_step_3": 0.3035,
      "epoch": 0.02766128312661729,
      "grad_norm": 79.04064178466797,
      "learning_rate": 5.523186682520809e-05,
      "loss": 0.08958419561386108,
      "loss_step_1": 0.0932,
      "loss_step_2": 0.0935,
      "loss_step_3": 0.0904,
      "step": 930
    },
    {
      "entropy_step_1": 0.3024,
      "entropy_step_2": 0.3084,
      "entropy_step_3": 0.3166,
      "epoch": 0.027958716278516402,
      "grad_norm": 103.05496978759766,
      "learning_rate": 5.582639714625446e-05,
      "loss": 0.09011476635932922,
      "loss_step_1": 0.0999,
      "loss_step_2": 0.0994,
      "loss_step_3": 0.0925,
      "step": 940
    },
    {
      "entropy_step_1": 0.3042,
      "entropy_step_2": 0.2992,
      "entropy_step_3": 0.3214,
      "epoch": 0.028256149430415513,
      "grad_norm": 70.90872192382812,
      "learning_rate": 5.6420927467300834e-05,
      "loss": 0.0933004379272461,
      "loss_step_1": 0.0915,
      "loss_step_2": 0.091,
      "loss_step_3": 0.0979,
      "step": 950
    },
    {
      "entropy_step_1": 0.3128,
      "entropy_step_2": 0.3084,
      "entropy_step_3": 0.3255,
      "epoch": 0.028553582582314625,
      "grad_norm": 89.31499481201172,
      "learning_rate": 5.7015457788347204e-05,
      "loss": 0.10039864778518677,
      "loss_step_1": 0.098,
      "loss_step_2": 0.0978,
      "loss_step_3": 0.1069,
      "step": 960
    },
    {
      "entropy_step_1": 0.2873,
      "entropy_step_2": 0.2956,
      "entropy_step_3": 0.3108,
      "epoch": 0.028851015734213736,
      "grad_norm": 76.65453338623047,
      "learning_rate": 5.7609988109393586e-05,
      "loss": 0.09685376286506653,
      "loss_step_1": 0.0917,
      "loss_step_2": 0.0928,
      "loss_step_3": 0.0941,
      "step": 970
    },
    {
      "entropy_step_1": 0.3094,
      "entropy_step_2": 0.3164,
      "entropy_step_3": 0.3222,
      "epoch": 0.029148448886112847,
      "grad_norm": 48.64705276489258,
      "learning_rate": 5.8204518430439956e-05,
      "loss": 0.09931063652038574,
      "loss_step_1": 0.0991,
      "loss_step_2": 0.0995,
      "loss_step_3": 0.1004,
      "step": 980
    },
    {
      "entropy_step_1": 0.3401,
      "entropy_step_2": 0.3399,
      "entropy_step_3": 0.339,
      "epoch": 0.029445882038011958,
      "grad_norm": 103.26080322265625,
      "learning_rate": 5.879904875148633e-05,
      "loss": 0.09599415063858033,
      "loss_step_1": 0.1022,
      "loss_step_2": 0.102,
      "loss_step_3": 0.1041,
      "step": 990
    },
    {
      "entropy_step_1": 0.2635,
      "entropy_step_2": 0.2621,
      "entropy_step_3": 0.2612,
      "epoch": 0.02974331518991107,
      "grad_norm": 39.30586624145508,
      "learning_rate": 5.93935790725327e-05,
      "loss": 0.07901580929756165,
      "loss_step_1": 0.0822,
      "loss_step_2": 0.082,
      "loss_step_3": 0.0676,
      "step": 1000
    },
    {
      "epoch": 0.02974331518991107,
      "eval_accuracy": 0.636,
      "eval_f1": 0.6289334078184954,
      "eval_loss": 0.30754733085632324,
      "eval_precision": 0.6472141879838584,
      "eval_recall": 0.636,
      "eval_runtime": 6.4094,
      "eval_samples_per_second": 78.011,
      "eval_steps_per_second": 9.829,
      "step": 1000
    },
    {
      "entropy_step_1": 0.3349,
      "entropy_step_2": 0.3375,
      "entropy_step_3": 0.3388,
      "epoch": 0.030040748341810176,
      "grad_norm": 69.72509765625,
      "learning_rate": 5.998810939357907e-05,
      "loss": 0.0995664119720459,
      "loss_step_1": 0.0999,
      "loss_step_2": 0.0991,
      "loss_step_3": 0.0977,
      "step": 1010
    },
    {
      "entropy_step_1": 0.2868,
      "entropy_step_2": 0.2838,
      "entropy_step_3": 0.2805,
      "epoch": 0.030338181493709288,
      "grad_norm": 56.9851188659668,
      "learning_rate": 6.0582639714625446e-05,
      "loss": 0.08874453902244568,
      "loss_step_1": 0.0882,
      "loss_step_2": 0.0878,
      "loss_step_3": 0.0861,
      "step": 1020
    },
    {
      "entropy_step_1": 0.2881,
      "entropy_step_2": 0.288,
      "entropy_step_3": 0.2904,
      "epoch": 0.0306356146456084,
      "grad_norm": 45.18030548095703,
      "learning_rate": 6.117717003567182e-05,
      "loss": 0.085654616355896,
      "loss_step_1": 0.0842,
      "loss_step_2": 0.0835,
      "loss_step_3": 0.0878,
      "step": 1030
    },
    {
      "entropy_step_1": 0.3026,
      "entropy_step_2": 0.3029,
      "entropy_step_3": 0.3162,
      "epoch": 0.03093304779750751,
      "grad_norm": 174.9792022705078,
      "learning_rate": 6.177170035671819e-05,
      "loss": 0.10366265773773194,
      "loss_step_1": 0.1055,
      "loss_step_2": 0.105,
      "loss_step_3": 0.0957,
      "step": 1040
    },
    {
      "entropy_step_1": 0.323,
      "entropy_step_2": 0.3272,
      "entropy_step_3": 0.3322,
      "epoch": 0.03123048094940662,
      "grad_norm": 77.2298355102539,
      "learning_rate": 6.236623067776457e-05,
      "loss": 0.10002387762069702,
      "loss_step_1": 0.0964,
      "loss_step_2": 0.0969,
      "loss_step_3": 0.1029,
      "step": 1050
    },
    {
      "entropy_step_1": 0.2904,
      "entropy_step_2": 0.2885,
      "entropy_step_3": 0.2749,
      "epoch": 0.03152791410130573,
      "grad_norm": 53.86473083496094,
      "learning_rate": 6.296076099881094e-05,
      "loss": 0.09458569884300232,
      "loss_step_1": 0.0883,
      "loss_step_2": 0.0884,
      "loss_step_3": 0.0819,
      "step": 1060
    },
    {
      "entropy_step_1": 0.3146,
      "entropy_step_2": 0.3122,
      "entropy_step_3": 0.3005,
      "epoch": 0.03182534725320484,
      "grad_norm": 75.0403060913086,
      "learning_rate": 6.355529131985731e-05,
      "loss": 0.08831526041030884,
      "loss_step_1": 0.0896,
      "loss_step_2": 0.0888,
      "loss_step_3": 0.0848,
      "step": 1070
    },
    {
      "entropy_step_1": 0.3123,
      "entropy_step_2": 0.3173,
      "entropy_step_3": 0.3327,
      "epoch": 0.032122780405103954,
      "grad_norm": 57.19150161743164,
      "learning_rate": 6.414982164090368e-05,
      "loss": 0.09590400457382202,
      "loss_step_1": 0.1021,
      "loss_step_2": 0.103,
      "loss_step_3": 0.1097,
      "step": 1080
    },
    {
      "entropy_step_1": 0.2957,
      "entropy_step_2": 0.2972,
      "entropy_step_3": 0.3041,
      "epoch": 0.032420213557003065,
      "grad_norm": 128.4562225341797,
      "learning_rate": 6.474435196195007e-05,
      "loss": 0.09407123923301697,
      "loss_step_1": 0.094,
      "loss_step_2": 0.0944,
      "loss_step_3": 0.0981,
      "step": 1090
    },
    {
      "entropy_step_1": 0.2887,
      "entropy_step_2": 0.2899,
      "entropy_step_3": 0.2941,
      "epoch": 0.032717646708902176,
      "grad_norm": 104.38414764404297,
      "learning_rate": 6.533888228299643e-05,
      "loss": 0.08449937701225281,
      "loss_step_1": 0.082,
      "loss_step_2": 0.0812,
      "loss_step_3": 0.0852,
      "step": 1100
    },
    {
      "entropy_step_1": 0.2944,
      "entropy_step_2": 0.2878,
      "entropy_step_3": 0.2813,
      "epoch": 0.03301507986080129,
      "grad_norm": 54.77143859863281,
      "learning_rate": 6.59334126040428e-05,
      "loss": 0.09155243635177612,
      "loss_step_1": 0.0977,
      "loss_step_2": 0.0986,
      "loss_step_3": 0.0916,
      "step": 1110
    },
    {
      "entropy_step_1": 0.3255,
      "entropy_step_2": 0.3232,
      "entropy_step_3": 0.3029,
      "epoch": 0.0333125130127004,
      "grad_norm": 64.50007629394531,
      "learning_rate": 6.652794292508919e-05,
      "loss": 0.09625117182731628,
      "loss_step_1": 0.099,
      "loss_step_2": 0.0988,
      "loss_step_3": 0.0869,
      "step": 1120
    },
    {
      "entropy_step_1": 0.2882,
      "entropy_step_2": 0.291,
      "entropy_step_3": 0.2965,
      "epoch": 0.03360994616459951,
      "grad_norm": 48.59226608276367,
      "learning_rate": 6.712247324613556e-05,
      "loss": 0.09146907925605774,
      "loss_step_1": 0.0932,
      "loss_step_2": 0.0932,
      "loss_step_3": 0.0829,
      "step": 1130
    },
    {
      "entropy_step_1": 0.2931,
      "entropy_step_2": 0.2961,
      "entropy_step_3": 0.2927,
      "epoch": 0.03390737931649862,
      "grad_norm": 61.889652252197266,
      "learning_rate": 6.771700356718193e-05,
      "loss": 0.09044798612594604,
      "loss_step_1": 0.0933,
      "loss_step_2": 0.0937,
      "loss_step_3": 0.0898,
      "step": 1140
    },
    {
      "entropy_step_1": 0.2862,
      "entropy_step_2": 0.2918,
      "entropy_step_3": 0.3053,
      "epoch": 0.034204812468397724,
      "grad_norm": 87.90200805664062,
      "learning_rate": 6.831153388822831e-05,
      "loss": 0.08901283144950867,
      "loss_step_1": 0.0818,
      "loss_step_2": 0.0814,
      "loss_step_3": 0.0848,
      "step": 1150
    },
    {
      "entropy_step_1": 0.2886,
      "entropy_step_2": 0.2912,
      "entropy_step_3": 0.2778,
      "epoch": 0.034502245620296836,
      "grad_norm": 72.8223648071289,
      "learning_rate": 6.890606420927468e-05,
      "loss": 0.0841073751449585,
      "loss_step_1": 0.0833,
      "loss_step_2": 0.0834,
      "loss_step_3": 0.0736,
      "step": 1160
    },
    {
      "entropy_step_1": 0.3121,
      "entropy_step_2": 0.3245,
      "entropy_step_3": 0.3076,
      "epoch": 0.03479967877219595,
      "grad_norm": 180.87510681152344,
      "learning_rate": 6.950059453032105e-05,
      "loss": 0.10555994510650635,
      "loss_step_1": 0.1108,
      "loss_step_2": 0.11,
      "loss_step_3": 0.115,
      "step": 1170
    },
    {
      "entropy_step_1": 0.2937,
      "entropy_step_2": 0.3054,
      "entropy_step_3": 0.3343,
      "epoch": 0.03509711192409506,
      "grad_norm": 210.67050170898438,
      "learning_rate": 7.009512485136742e-05,
      "loss": 0.09792216420173645,
      "loss_step_1": 0.1014,
      "loss_step_2": 0.1028,
      "loss_step_3": 0.1089,
      "step": 1180
    },
    {
      "entropy_step_1": 0.2805,
      "entropy_step_2": 0.2848,
      "entropy_step_3": 0.2946,
      "epoch": 0.03539454507599417,
      "grad_norm": 63.91289520263672,
      "learning_rate": 7.06896551724138e-05,
      "loss": 0.08954975008964539,
      "loss_step_1": 0.0894,
      "loss_step_2": 0.0893,
      "loss_step_3": 0.0895,
      "step": 1190
    },
    {
      "entropy_step_1": 0.2878,
      "entropy_step_2": 0.2906,
      "entropy_step_3": 0.2968,
      "epoch": 0.03569197822789328,
      "grad_norm": 63.75553894042969,
      "learning_rate": 7.128418549346017e-05,
      "loss": 0.0836095929145813,
      "loss_step_1": 0.0888,
      "loss_step_2": 0.0895,
      "loss_step_3": 0.0942,
      "step": 1200
    },
    {
      "entropy_step_1": 0.2913,
      "entropy_step_2": 0.3038,
      "entropy_step_3": 0.2949,
      "epoch": 0.03598941137979239,
      "grad_norm": 56.57352828979492,
      "learning_rate": 7.187871581450655e-05,
      "loss": 0.08846445679664612,
      "loss_step_1": 0.0874,
      "loss_step_2": 0.0872,
      "loss_step_3": 0.086,
      "step": 1210
    },
    {
      "entropy_step_1": 0.2832,
      "entropy_step_2": 0.2935,
      "entropy_step_3": 0.317,
      "epoch": 0.0362868445316915,
      "grad_norm": 75.64051818847656,
      "learning_rate": 7.247324613555292e-05,
      "loss": 0.08633489608764648,
      "loss_step_1": 0.0885,
      "loss_step_2": 0.0879,
      "loss_step_3": 0.092,
      "step": 1220
    },
    {
      "entropy_step_1": 0.268,
      "entropy_step_2": 0.2709,
      "entropy_step_3": 0.2739,
      "epoch": 0.03658427768359061,
      "grad_norm": 83.17045593261719,
      "learning_rate": 7.306777645659929e-05,
      "loss": 0.07847254872322082,
      "loss_step_1": 0.075,
      "loss_step_2": 0.0746,
      "loss_step_3": 0.0702,
      "step": 1230
    },
    {
      "entropy_step_1": 0.3157,
      "entropy_step_2": 0.3149,
      "entropy_step_3": 0.2903,
      "epoch": 0.036881710835489724,
      "grad_norm": 78.23483276367188,
      "learning_rate": 7.366230677764566e-05,
      "loss": 0.09289465546607971,
      "loss_step_1": 0.0902,
      "loss_step_2": 0.0899,
      "loss_step_3": 0.0858,
      "step": 1240
    },
    {
      "entropy_step_1": 0.2763,
      "entropy_step_2": 0.2729,
      "entropy_step_3": 0.298,
      "epoch": 0.037179143987388835,
      "grad_norm": 36.367088317871094,
      "learning_rate": 7.425683709869203e-05,
      "loss": 0.08035939931869507,
      "loss_step_1": 0.0756,
      "loss_step_2": 0.0757,
      "loss_step_3": 0.0925,
      "step": 1250
    },
    {
      "entropy_step_1": 0.2645,
      "entropy_step_2": 0.2664,
      "entropy_step_3": 0.2836,
      "epoch": 0.037476577139287946,
      "grad_norm": 44.45389938354492,
      "learning_rate": 7.485136741973841e-05,
      "loss": 0.10236362218856812,
      "loss_step_1": 0.1049,
      "loss_step_2": 0.104,
      "loss_step_3": 0.1077,
      "step": 1260
    },
    {
      "entropy_step_1": 0.2744,
      "entropy_step_2": 0.2785,
      "entropy_step_3": 0.2634,
      "epoch": 0.03777401029118706,
      "grad_norm": 134.84405517578125,
      "learning_rate": 7.54458977407848e-05,
      "loss": 0.07677897810935974,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.0738,
      "loss_step_3": 0.0767,
      "step": 1270
    },
    {
      "entropy_step_1": 0.282,
      "entropy_step_2": 0.284,
      "entropy_step_3": 0.2746,
      "epoch": 0.03807144344308617,
      "grad_norm": 131.8893585205078,
      "learning_rate": 7.604042806183116e-05,
      "loss": 0.09748839139938355,
      "loss_step_1": 0.0947,
      "loss_step_2": 0.0941,
      "loss_step_3": 0.079,
      "step": 1280
    },
    {
      "entropy_step_1": 0.3111,
      "entropy_step_2": 0.3036,
      "entropy_step_3": 0.3,
      "epoch": 0.03836887659498528,
      "grad_norm": 283.9534606933594,
      "learning_rate": 7.663495838287753e-05,
      "loss": 0.09222343564033508,
      "loss_step_1": 0.0928,
      "loss_step_2": 0.0921,
      "loss_step_3": 0.0912,
      "step": 1290
    },
    {
      "entropy_step_1": 0.3337,
      "entropy_step_2": 0.3254,
      "entropy_step_3": 0.3117,
      "epoch": 0.03866630974688439,
      "grad_norm": 88.40239715576172,
      "learning_rate": 7.72294887039239e-05,
      "loss": 0.09623315930366516,
      "loss_step_1": 0.0894,
      "loss_step_2": 0.0887,
      "loss_step_3": 0.084,
      "step": 1300
    },
    {
      "entropy_step_1": 0.2759,
      "entropy_step_2": 0.2702,
      "entropy_step_3": 0.2541,
      "epoch": 0.0389637428987835,
      "grad_norm": 90.4374008178711,
      "learning_rate": 7.782401902497027e-05,
      "loss": 0.08166399598121643,
      "loss_step_1": 0.0809,
      "loss_step_2": 0.0808,
      "loss_step_3": 0.082,
      "step": 1310
    },
    {
      "entropy_step_1": 0.2966,
      "entropy_step_2": 0.2888,
      "entropy_step_3": 0.2786,
      "epoch": 0.039261176050682606,
      "grad_norm": 94.4776611328125,
      "learning_rate": 7.841854934601664e-05,
      "loss": 0.08977149128913879,
      "loss_step_1": 0.089,
      "loss_step_2": 0.0884,
      "loss_step_3": 0.088,
      "step": 1320
    },
    {
      "entropy_step_1": 0.2909,
      "entropy_step_2": 0.2858,
      "entropy_step_3": 0.2843,
      "epoch": 0.03955860920258172,
      "grad_norm": 56.156558990478516,
      "learning_rate": 7.901307966706302e-05,
      "loss": 0.08929533958435058,
      "loss_step_1": 0.0962,
      "loss_step_2": 0.0955,
      "loss_step_3": 0.0822,
      "step": 1330
    },
    {
      "entropy_step_1": 0.3027,
      "entropy_step_2": 0.3048,
      "entropy_step_3": 0.3085,
      "epoch": 0.03985604235448083,
      "grad_norm": 60.938720703125,
      "learning_rate": 7.960760998810939e-05,
      "loss": 0.0854015290737152,
      "loss_step_1": 0.0865,
      "loss_step_2": 0.086,
      "loss_step_3": 0.0854,
      "step": 1340
    },
    {
      "entropy_step_1": 0.2823,
      "entropy_step_2": 0.2777,
      "entropy_step_3": 0.293,
      "epoch": 0.04015347550637994,
      "grad_norm": 41.0838737487793,
      "learning_rate": 8.020214030915578e-05,
      "loss": 0.09074344635009765,
      "loss_step_1": 0.085,
      "loss_step_2": 0.0842,
      "loss_step_3": 0.0875,
      "step": 1350
    },
    {
      "entropy_step_1": 0.2764,
      "entropy_step_2": 0.2636,
      "entropy_step_3": 0.2604,
      "epoch": 0.04045090865827905,
      "grad_norm": 112.51982116699219,
      "learning_rate": 8.079667063020214e-05,
      "loss": 0.08926092386245728,
      "loss_step_1": 0.0831,
      "loss_step_2": 0.0819,
      "loss_step_3": 0.0937,
      "step": 1360
    },
    {
      "entropy_step_1": 0.2717,
      "entropy_step_2": 0.2594,
      "entropy_step_3": 0.2531,
      "epoch": 0.04074834181017816,
      "grad_norm": 108.85630798339844,
      "learning_rate": 8.139120095124851e-05,
      "loss": 0.0829021692276001,
      "loss_step_1": 0.079,
      "loss_step_2": 0.0782,
      "loss_step_3": 0.0805,
      "step": 1370
    },
    {
      "entropy_step_1": 0.2818,
      "entropy_step_2": 0.2805,
      "entropy_step_3": 0.262,
      "epoch": 0.04104577496207727,
      "grad_norm": 132.3631134033203,
      "learning_rate": 8.198573127229488e-05,
      "loss": 0.08340991139411927,
      "loss_step_1": 0.0834,
      "loss_step_2": 0.0831,
      "loss_step_3": 0.0716,
      "step": 1380
    },
    {
      "entropy_step_1": 0.2956,
      "entropy_step_2": 0.2974,
      "entropy_step_3": 0.2737,
      "epoch": 0.04134320811397638,
      "grad_norm": 86.60938262939453,
      "learning_rate": 8.258026159334127e-05,
      "loss": 0.08839019536972045,
      "loss_step_1": 0.0951,
      "loss_step_2": 0.0944,
      "loss_step_3": 0.084,
      "step": 1390
    },
    {
      "entropy_step_1": 0.2234,
      "entropy_step_2": 0.2297,
      "entropy_step_3": 0.2451,
      "epoch": 0.041640641265875494,
      "grad_norm": 61.96017837524414,
      "learning_rate": 8.317479191438764e-05,
      "loss": 0.10049197673797608,
      "loss_step_1": 0.0959,
      "loss_step_2": 0.095,
      "loss_step_3": 0.1079,
      "step": 1400
    },
    {
      "entropy_step_1": 0.271,
      "entropy_step_2": 0.2736,
      "entropy_step_3": 0.2799,
      "epoch": 0.041938074417774605,
      "grad_norm": 163.3586883544922,
      "learning_rate": 8.3769322235434e-05,
      "loss": 0.08350586295127868,
      "loss_step_1": 0.0821,
      "loss_step_2": 0.0824,
      "loss_step_3": 0.0924,
      "step": 1410
    },
    {
      "entropy_step_1": 0.2934,
      "entropy_step_2": 0.282,
      "entropy_step_3": 0.2803,
      "epoch": 0.042235507569673716,
      "grad_norm": 54.51877212524414,
      "learning_rate": 8.436385255648039e-05,
      "loss": 0.08952122926712036,
      "loss_step_1": 0.0901,
      "loss_step_2": 0.0894,
      "loss_step_3": 0.0924,
      "step": 1420
    },
    {
      "entropy_step_1": 0.2523,
      "entropy_step_2": 0.2499,
      "entropy_step_3": 0.2536,
      "epoch": 0.04253294072157283,
      "grad_norm": 62.19689178466797,
      "learning_rate": 8.495838287752676e-05,
      "loss": 0.08641036152839661,
      "loss_step_1": 0.0952,
      "loss_step_2": 0.0952,
      "loss_step_3": 0.0983,
      "step": 1430
    },
    {
      "entropy_step_1": 0.2955,
      "entropy_step_2": 0.2996,
      "entropy_step_3": 0.2895,
      "epoch": 0.04283037387347194,
      "grad_norm": 100.86942291259766,
      "learning_rate": 8.555291319857313e-05,
      "loss": 0.08813844919204712,
      "loss_step_1": 0.0936,
      "loss_step_2": 0.0938,
      "loss_step_3": 0.0714,
      "step": 1440
    },
    {
      "entropy_step_1": 0.2943,
      "entropy_step_2": 0.3017,
      "entropy_step_3": 0.317,
      "epoch": 0.04312780702537105,
      "grad_norm": 51.20358657836914,
      "learning_rate": 8.614744351961951e-05,
      "loss": 0.08375130891799927,
      "loss_step_1": 0.088,
      "loss_step_2": 0.0874,
      "loss_step_3": 0.0843,
      "step": 1450
    },
    {
      "entropy_step_1": 0.2621,
      "entropy_step_2": 0.2667,
      "entropy_step_3": 0.2621,
      "epoch": 0.04342524017727016,
      "grad_norm": 74.44619750976562,
      "learning_rate": 8.674197384066588e-05,
      "loss": 0.08156928420066833,
      "loss_step_1": 0.0919,
      "loss_step_2": 0.0915,
      "loss_step_3": 0.0956,
      "step": 1460
    },
    {
      "entropy_step_1": 0.2781,
      "entropy_step_2": 0.2876,
      "entropy_step_3": 0.2646,
      "epoch": 0.04372267332916927,
      "grad_norm": 64.73831176757812,
      "learning_rate": 8.733650416171225e-05,
      "loss": 0.08286420702934265,
      "loss_step_1": 0.0814,
      "loss_step_2": 0.0808,
      "loss_step_3": 0.0747,
      "step": 1470
    },
    {
      "entropy_step_1": 0.257,
      "entropy_step_2": 0.2623,
      "entropy_step_3": 0.2612,
      "epoch": 0.04402010648106838,
      "grad_norm": 86.007568359375,
      "learning_rate": 8.793103448275862e-05,
      "loss": 0.07883127927780151,
      "loss_step_1": 0.0802,
      "loss_step_2": 0.0813,
      "loss_step_3": 0.0758,
      "step": 1480
    },
    {
      "entropy_step_1": 0.2747,
      "entropy_step_2": 0.2766,
      "entropy_step_3": 0.2689,
      "epoch": 0.044317539632967494,
      "grad_norm": 134.92286682128906,
      "learning_rate": 8.8525564803805e-05,
      "loss": 0.08255268335342407,
      "loss_step_1": 0.0801,
      "loss_step_2": 0.08,
      "loss_step_3": 0.0732,
      "step": 1490
    },
    {
      "entropy_step_1": 0.2993,
      "entropy_step_2": 0.2964,
      "entropy_step_3": 0.3015,
      "epoch": 0.0446149727848666,
      "grad_norm": 51.06706237792969,
      "learning_rate": 8.912009512485138e-05,
      "loss": 0.08809412121772767,
      "loss_step_1": 0.0926,
      "loss_step_2": 0.0913,
      "loss_step_3": 0.0909,
      "step": 1500
    },
    {
      "entropy_step_1": 0.2623,
      "entropy_step_2": 0.2639,
      "entropy_step_3": 0.2504,
      "epoch": 0.04491240593676571,
      "grad_norm": 138.16163635253906,
      "learning_rate": 8.971462544589775e-05,
      "loss": 0.09326639771461487,
      "loss_step_1": 0.0939,
      "loss_step_2": 0.0934,
      "loss_step_3": 0.0941,
      "step": 1510
    },
    {
      "entropy_step_1": 0.3025,
      "entropy_step_2": 0.3084,
      "entropy_step_3": 0.3124,
      "epoch": 0.04520983908866482,
      "grad_norm": 103.42186737060547,
      "learning_rate": 9.030915576694412e-05,
      "loss": 0.08835214972496033,
      "loss_step_1": 0.0885,
      "loss_step_2": 0.0883,
      "loss_step_3": 0.0772,
      "step": 1520
    },
    {
      "entropy_step_1": 0.2771,
      "entropy_step_2": 0.2699,
      "entropy_step_3": 0.2693,
      "epoch": 0.04550727224056393,
      "grad_norm": 77.18631744384766,
      "learning_rate": 9.090368608799049e-05,
      "loss": 0.07198024988174438,
      "loss_step_1": 0.0772,
      "loss_step_2": 0.0771,
      "loss_step_3": 0.0834,
      "step": 1530
    },
    {
      "entropy_step_1": 0.2364,
      "entropy_step_2": 0.2225,
      "entropy_step_3": 0.2258,
      "epoch": 0.04580470539246304,
      "grad_norm": 72.52236938476562,
      "learning_rate": 9.149821640903686e-05,
      "loss": 0.08588604927062989,
      "loss_step_1": 0.0862,
      "loss_step_2": 0.0866,
      "loss_step_3": 0.0933,
      "step": 1540
    },
    {
      "entropy_step_1": 0.3071,
      "entropy_step_2": 0.3048,
      "entropy_step_3": 0.3029,
      "epoch": 0.04610213854436215,
      "grad_norm": 94.15626525878906,
      "learning_rate": 9.209274673008323e-05,
      "loss": 0.09396561384201049,
      "loss_step_1": 0.094,
      "loss_step_2": 0.0937,
      "loss_step_3": 0.0925,
      "step": 1550
    },
    {
      "entropy_step_1": 0.2845,
      "entropy_step_2": 0.2992,
      "entropy_step_3": 0.3059,
      "epoch": 0.046399571696261264,
      "grad_norm": 95.55241394042969,
      "learning_rate": 9.268727705112961e-05,
      "loss": 0.08582545518875122,
      "loss_step_1": 0.0882,
      "loss_step_2": 0.0881,
      "loss_step_3": 0.0937,
      "step": 1560
    },
    {
      "entropy_step_1": 0.2773,
      "entropy_step_2": 0.2823,
      "entropy_step_3": 0.2864,
      "epoch": 0.046697004848160376,
      "grad_norm": 101.51142883300781,
      "learning_rate": 9.3281807372176e-05,
      "loss": 0.08120195865631104,
      "loss_step_1": 0.09,
      "loss_step_2": 0.0888,
      "loss_step_3": 0.0887,
      "step": 1570
    },
    {
      "entropy_step_1": 0.2198,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.2225,
      "epoch": 0.04699443800005949,
      "grad_norm": 135.77780151367188,
      "learning_rate": 9.387633769322236e-05,
      "loss": 0.07690795063972473,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0667,
      "step": 1580
    },
    {
      "entropy_step_1": 0.2952,
      "entropy_step_2": 0.277,
      "entropy_step_3": 0.2552,
      "epoch": 0.0472918711519586,
      "grad_norm": 97.83174896240234,
      "learning_rate": 9.447086801426873e-05,
      "loss": 0.08830531239509583,
      "loss_step_1": 0.0908,
      "loss_step_2": 0.0903,
      "loss_step_3": 0.0825,
      "step": 1590
    },
    {
      "entropy_step_1": 0.2771,
      "entropy_step_2": 0.2686,
      "entropy_step_3": 0.2531,
      "epoch": 0.04758930430385771,
      "grad_norm": 85.93737030029297,
      "learning_rate": 9.50653983353151e-05,
      "loss": 0.07649304866790771,
      "loss_step_1": 0.0798,
      "loss_step_2": 0.0802,
      "loss_step_3": 0.0752,
      "step": 1600
    },
    {
      "entropy_step_1": 0.2611,
      "entropy_step_2": 0.2623,
      "entropy_step_3": 0.2691,
      "epoch": 0.04788673745575682,
      "grad_norm": 47.756866455078125,
      "learning_rate": 9.565992865636147e-05,
      "loss": 0.08350440263748168,
      "loss_step_1": 0.0827,
      "loss_step_2": 0.0821,
      "loss_step_3": 0.0799,
      "step": 1610
    },
    {
      "entropy_step_1": 0.2548,
      "entropy_step_2": 0.257,
      "entropy_step_3": 0.2401,
      "epoch": 0.04818417060765593,
      "grad_norm": 66.98975372314453,
      "learning_rate": 9.625445897740786e-05,
      "loss": 0.08464881777763367,
      "loss_step_1": 0.0846,
      "loss_step_2": 0.0838,
      "loss_step_3": 0.0811,
      "step": 1620
    },
    {
      "entropy_step_1": 0.2604,
      "entropy_step_2": 0.2585,
      "entropy_step_3": 0.2369,
      "epoch": 0.04848160375955504,
      "grad_norm": 82.49462127685547,
      "learning_rate": 9.684898929845422e-05,
      "loss": 0.08336356282234192,
      "loss_step_1": 0.0822,
      "loss_step_2": 0.0818,
      "loss_step_3": 0.0706,
      "step": 1630
    },
    {
      "entropy_step_1": 0.2994,
      "entropy_step_2": 0.2956,
      "entropy_step_3": 0.2756,
      "epoch": 0.04877903691145415,
      "grad_norm": 55.67588424682617,
      "learning_rate": 9.744351961950061e-05,
      "loss": 0.08344126939773559,
      "loss_step_1": 0.0888,
      "loss_step_2": 0.088,
      "loss_step_3": 0.0928,
      "step": 1640
    },
    {
      "entropy_step_1": 0.2713,
      "entropy_step_2": 0.2752,
      "entropy_step_3": 0.277,
      "epoch": 0.049076470063353264,
      "grad_norm": 48.5106315612793,
      "learning_rate": 9.803804994054698e-05,
      "loss": 0.08011029958724976,
      "loss_step_1": 0.0838,
      "loss_step_2": 0.0839,
      "loss_step_3": 0.0871,
      "step": 1650
    },
    {
      "entropy_step_1": 0.2771,
      "entropy_step_2": 0.2862,
      "entropy_step_3": 0.2951,
      "epoch": 0.049373903215252375,
      "grad_norm": 109.29071044921875,
      "learning_rate": 9.863258026159335e-05,
      "loss": 0.09034280180931091,
      "loss_step_1": 0.0862,
      "loss_step_2": 0.086,
      "loss_step_3": 0.0818,
      "step": 1660
    },
    {
      "entropy_step_1": 0.2809,
      "entropy_step_2": 0.2779,
      "entropy_step_3": 0.281,
      "epoch": 0.04967133636715148,
      "grad_norm": 39.64223098754883,
      "learning_rate": 9.922711058263972e-05,
      "loss": 0.08212672472000122,
      "loss_step_1": 0.0848,
      "loss_step_2": 0.0848,
      "loss_step_3": 0.0851,
      "step": 1670
    },
    {
      "entropy_step_1": 0.2653,
      "entropy_step_2": 0.2587,
      "entropy_step_3": 0.2711,
      "epoch": 0.04996876951905059,
      "grad_norm": 82.58372497558594,
      "learning_rate": 9.98216409036861e-05,
      "loss": 0.07961670756340027,
      "loss_step_1": 0.0779,
      "loss_step_2": 0.0776,
      "loss_step_3": 0.0951,
      "step": 1680
    },
    {
      "entropy_step_1": 0.2948,
      "entropy_step_2": 0.2935,
      "entropy_step_3": 0.2921,
      "epoch": 0.0502662026709497,
      "grad_norm": 78.49678039550781,
      "learning_rate": 9.999998814795783e-05,
      "loss": 0.0779748022556305,
      "loss_step_1": 0.0838,
      "loss_step_2": 0.0836,
      "loss_step_3": 0.0778,
      "step": 1690
    },
    {
      "entropy_step_1": 0.2576,
      "entropy_step_2": 0.2695,
      "entropy_step_3": 0.2698,
      "epoch": 0.05056363582284881,
      "grad_norm": 116.42617797851562,
      "learning_rate": 9.999993009715256e-05,
      "loss": 0.07281153202056885,
      "loss_step_1": 0.0781,
      "loss_step_2": 0.077,
      "loss_step_3": 0.0814,
      "step": 1700
    },
    {
      "entropy_step_1": 0.2418,
      "entropy_step_2": 0.2548,
      "entropy_step_3": 0.2612,
      "epoch": 0.050861068974747924,
      "grad_norm": 60.97153854370117,
      "learning_rate": 9.999982367073457e-05,
      "loss": 0.07447486519813537,
      "loss_step_1": 0.0768,
      "loss_step_2": 0.0762,
      "loss_step_3": 0.075,
      "step": 1710
    },
    {
      "entropy_step_1": 0.2656,
      "entropy_step_2": 0.2702,
      "entropy_step_3": 0.2533,
      "epoch": 0.051158502126647035,
      "grad_norm": 53.170772552490234,
      "learning_rate": 9.999966886880682e-05,
      "loss": 0.0834235668182373,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.074,
      "loss_step_3": 0.0621,
      "step": 1720
    },
    {
      "entropy_step_1": 0.2316,
      "entropy_step_2": 0.2223,
      "entropy_step_3": 0.2268,
      "epoch": 0.051455935278546146,
      "grad_norm": 52.9130744934082,
      "learning_rate": 9.99994656915191e-05,
      "loss": 0.06649897694587707,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0606,
      "step": 1730
    },
    {
      "entropy_step_1": 0.259,
      "entropy_step_2": 0.2491,
      "entropy_step_3": 0.2379,
      "epoch": 0.05175336843044526,
      "grad_norm": 114.27075958251953,
      "learning_rate": 9.999921413906798e-05,
      "loss": 0.07424667477607727,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.0737,
      "loss_step_3": 0.066,
      "step": 1740
    },
    {
      "entropy_step_1": 0.2653,
      "entropy_step_2": 0.2623,
      "entropy_step_3": 0.2636,
      "epoch": 0.05205080158234437,
      "grad_norm": 71.30533599853516,
      "learning_rate": 9.999891421169685e-05,
      "loss": 0.077163565158844,
      "loss_step_1": 0.0812,
      "loss_step_2": 0.081,
      "loss_step_3": 0.0744,
      "step": 1750
    },
    {
      "entropy_step_1": 0.2498,
      "entropy_step_2": 0.2537,
      "entropy_step_3": 0.2602,
      "epoch": 0.05234823473424348,
      "grad_norm": 51.61823654174805,
      "learning_rate": 9.999856590969588e-05,
      "loss": 0.08578778505325317,
      "loss_step_1": 0.0859,
      "loss_step_2": 0.0858,
      "loss_step_3": 0.0838,
      "step": 1760
    },
    {
      "entropy_step_1": 0.2402,
      "entropy_step_2": 0.2441,
      "entropy_step_3": 0.2394,
      "epoch": 0.05264566788614259,
      "grad_norm": 77.38929748535156,
      "learning_rate": 9.999816923340207e-05,
      "loss": 0.06947680115699768,
      "loss_step_1": 0.0713,
      "loss_step_2": 0.0712,
      "loss_step_3": 0.0699,
      "step": 1770
    },
    {
      "entropy_step_1": 0.2209,
      "entropy_step_2": 0.2247,
      "entropy_step_3": 0.2235,
      "epoch": 0.0529431010380417,
      "grad_norm": 73.325439453125,
      "learning_rate": 9.999772418319919e-05,
      "loss": 0.06767261028289795,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.068,
      "loss_step_3": 0.073,
      "step": 1780
    },
    {
      "entropy_step_1": 0.2391,
      "entropy_step_2": 0.2384,
      "entropy_step_3": 0.2438,
      "epoch": 0.05324053418994081,
      "grad_norm": 101.42700958251953,
      "learning_rate": 9.999723075951787e-05,
      "loss": 0.07186325788497924,
      "loss_step_1": 0.0706,
      "loss_step_2": 0.0702,
      "loss_step_3": 0.0706,
      "step": 1790
    },
    {
      "entropy_step_1": 0.2575,
      "entropy_step_2": 0.2532,
      "entropy_step_3": 0.2449,
      "epoch": 0.05353796734183992,
      "grad_norm": 126.40409088134766,
      "learning_rate": 9.999668896283545e-05,
      "loss": 0.08052533268928527,
      "loss_step_1": 0.0807,
      "loss_step_2": 0.0799,
      "loss_step_3": 0.0707,
      "step": 1800
    },
    {
      "entropy_step_1": 0.2675,
      "entropy_step_2": 0.2597,
      "entropy_step_3": 0.2667,
      "epoch": 0.053835400493739034,
      "grad_norm": 42.55577850341797,
      "learning_rate": 9.999609879367617e-05,
      "loss": 0.08270376920700073,
      "loss_step_1": 0.084,
      "loss_step_2": 0.0833,
      "loss_step_3": 0.0883,
      "step": 1810
    },
    {
      "entropy_step_1": 0.2574,
      "entropy_step_2": 0.2554,
      "entropy_step_3": 0.2545,
      "epoch": 0.054132833645638145,
      "grad_norm": 124.00222778320312,
      "learning_rate": 9.999546025261101e-05,
      "loss": 0.08148663640022277,
      "loss_step_1": 0.0814,
      "loss_step_2": 0.0807,
      "loss_step_3": 0.0727,
      "step": 1820
    },
    {
      "entropy_step_1": 0.2825,
      "entropy_step_2": 0.2814,
      "entropy_step_3": 0.3107,
      "epoch": 0.054430266797537256,
      "grad_norm": 74.45962524414062,
      "learning_rate": 9.999477334025778e-05,
      "loss": 0.0874531626701355,
      "loss_step_1": 0.0863,
      "loss_step_2": 0.0865,
      "loss_step_3": 0.1011,
      "step": 1830
    },
    {
      "entropy_step_1": 0.2508,
      "entropy_step_2": 0.248,
      "entropy_step_3": 0.2618,
      "epoch": 0.05472769994943637,
      "grad_norm": 62.12924575805664,
      "learning_rate": 9.999403805728106e-05,
      "loss": 0.0758641541004181,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0732,
      "loss_step_3": 0.0793,
      "step": 1840
    },
    {
      "entropy_step_1": 0.2362,
      "entropy_step_2": 0.2322,
      "entropy_step_3": 0.2327,
      "epoch": 0.05502513310133547,
      "grad_norm": 66.6561050415039,
      "learning_rate": 9.999325440439225e-05,
      "loss": 0.0738156020641327,
      "loss_step_1": 0.0751,
      "loss_step_2": 0.0749,
      "loss_step_3": 0.065,
      "step": 1850
    },
    {
      "entropy_step_1": 0.2373,
      "entropy_step_2": 0.2332,
      "entropy_step_3": 0.2381,
      "epoch": 0.05532256625323458,
      "grad_norm": 267.1378173828125,
      "learning_rate": 9.999242238234954e-05,
      "loss": 0.07641763091087342,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.0735,
      "loss_step_3": 0.0724,
      "step": 1860
    },
    {
      "entropy_step_1": 0.2606,
      "entropy_step_2": 0.2595,
      "entropy_step_3": 0.2608,
      "epoch": 0.055619999405133694,
      "grad_norm": 38.48652648925781,
      "learning_rate": 9.999154199195793e-05,
      "loss": 0.07933675646781921,
      "loss_step_1": 0.0764,
      "loss_step_2": 0.077,
      "loss_step_3": 0.0807,
      "step": 1870
    },
    {
      "entropy_step_1": 0.2536,
      "entropy_step_2": 0.2574,
      "entropy_step_3": 0.2733,
      "epoch": 0.055917432557032805,
      "grad_norm": 60.074283599853516,
      "learning_rate": 9.999061323406922e-05,
      "loss": 0.07240758538246155,
      "loss_step_1": 0.0721,
      "loss_step_2": 0.0718,
      "loss_step_3": 0.0814,
      "step": 1880
    },
    {
      "entropy_step_1": 0.2652,
      "entropy_step_2": 0.2672,
      "entropy_step_3": 0.2626,
      "epoch": 0.056214865708931916,
      "grad_norm": 68.41935729980469,
      "learning_rate": 9.998963610958198e-05,
      "loss": 0.0843345582485199,
      "loss_step_1": 0.0807,
      "loss_step_2": 0.0803,
      "loss_step_3": 0.073,
      "step": 1890
    },
    {
      "entropy_step_1": 0.2608,
      "entropy_step_2": 0.2595,
      "entropy_step_3": 0.2548,
      "epoch": 0.05651229886083103,
      "grad_norm": 51.40510177612305,
      "learning_rate": 9.99886106194416e-05,
      "loss": 0.08313073515892029,
      "loss_step_1": 0.0774,
      "loss_step_2": 0.0775,
      "loss_step_3": 0.0711,
      "step": 1900
    },
    {
      "entropy_step_1": 0.2433,
      "entropy_step_2": 0.2414,
      "entropy_step_3": 0.259,
      "epoch": 0.05680973201273014,
      "grad_norm": 52.32907485961914,
      "learning_rate": 9.998753676464024e-05,
      "loss": 0.07018977999687195,
      "loss_step_1": 0.0693,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0772,
      "step": 1910
    },
    {
      "entropy_step_1": 0.2664,
      "entropy_step_2": 0.2599,
      "entropy_step_3": 0.266,
      "epoch": 0.05710716516462925,
      "grad_norm": 61.1766242980957,
      "learning_rate": 9.99864145462169e-05,
      "loss": 0.079660302400589,
      "loss_step_1": 0.0799,
      "loss_step_2": 0.0791,
      "loss_step_3": 0.078,
      "step": 1920
    },
    {
      "entropy_step_1": 0.2817,
      "entropy_step_2": 0.2733,
      "entropy_step_3": 0.238,
      "epoch": 0.05740459831652836,
      "grad_norm": 62.163238525390625,
      "learning_rate": 9.99852439652573e-05,
      "loss": 0.08566530346870423,
      "loss_step_1": 0.0891,
      "loss_step_2": 0.0889,
      "loss_step_3": 0.0866,
      "step": 1930
    },
    {
      "entropy_step_1": 0.224,
      "entropy_step_2": 0.2289,
      "entropy_step_3": 0.2473,
      "epoch": 0.05770203146842747,
      "grad_norm": 26.986696243286133,
      "learning_rate": 9.998402502289403e-05,
      "loss": 0.07763261795043945,
      "loss_step_1": 0.0889,
      "loss_step_2": 0.0882,
      "loss_step_3": 0.0919,
      "step": 1940
    },
    {
      "entropy_step_1": 0.2289,
      "entropy_step_2": 0.2393,
      "entropy_step_3": 0.2432,
      "epoch": 0.05799946462032658,
      "grad_norm": 32.55070495605469,
      "learning_rate": 9.998275772030643e-05,
      "loss": 0.08005954027175903,
      "loss_step_1": 0.0801,
      "loss_step_2": 0.0795,
      "loss_step_3": 0.0744,
      "step": 1950
    },
    {
      "entropy_step_1": 0.2813,
      "entropy_step_2": 0.2874,
      "entropy_step_3": 0.2953,
      "epoch": 0.05829689777222569,
      "grad_norm": 42.67896270751953,
      "learning_rate": 9.998144205872059e-05,
      "loss": 0.08408699035644532,
      "loss_step_1": 0.0888,
      "loss_step_2": 0.0882,
      "loss_step_3": 0.0875,
      "step": 1960
    },
    {
      "entropy_step_1": 0.2939,
      "entropy_step_2": 0.2997,
      "entropy_step_3": 0.2776,
      "epoch": 0.058594330924124804,
      "grad_norm": 95.8947982788086,
      "learning_rate": 9.99800780394095e-05,
      "loss": 0.08339161276817322,
      "loss_step_1": 0.0892,
      "loss_step_2": 0.0892,
      "loss_step_3": 0.0894,
      "step": 1970
    },
    {
      "entropy_step_1": 0.2171,
      "entropy_step_2": 0.2218,
      "entropy_step_3": 0.2274,
      "epoch": 0.058891764076023916,
      "grad_norm": 63.40790939331055,
      "learning_rate": 9.997866566369281e-05,
      "loss": 0.08021209239959717,
      "loss_step_1": 0.0862,
      "loss_step_2": 0.0848,
      "loss_step_3": 0.0901,
      "step": 1980
    },
    {
      "entropy_step_1": 0.2651,
      "entropy_step_2": 0.2746,
      "entropy_step_3": 0.3018,
      "epoch": 0.05918919722792303,
      "grad_norm": 79.5086898803711,
      "learning_rate": 9.997720493293703e-05,
      "loss": 0.0768567979335785,
      "loss_step_1": 0.0824,
      "loss_step_2": 0.0822,
      "loss_step_3": 0.0976,
      "step": 1990
    },
    {
      "entropy_step_1": 0.2357,
      "entropy_step_2": 0.2412,
      "entropy_step_3": 0.2454,
      "epoch": 0.05948663037982214,
      "grad_norm": 35.92058181762695,
      "learning_rate": 9.997569584855544e-05,
      "loss": 0.07928791642189026,
      "loss_step_1": 0.08,
      "loss_step_2": 0.079,
      "loss_step_3": 0.0831,
      "step": 2000
    },
    {
      "epoch": 0.05948663037982214,
      "eval_accuracy": 0.722,
      "eval_f1": 0.713320491191308,
      "eval_loss": 0.21759271621704102,
      "eval_precision": 0.7525896124228577,
      "eval_recall": 0.722,
      "eval_runtime": 6.2897,
      "eval_samples_per_second": 79.495,
      "eval_steps_per_second": 10.016,
      "step": 2000
    },
    {
      "entropy_step_1": 0.2334,
      "entropy_step_2": 0.2396,
      "entropy_step_3": 0.2163,
      "epoch": 0.05978406353172125,
      "grad_norm": 143.2496795654297,
      "learning_rate": 9.99741384120081e-05,
      "loss": 0.0787390947341919,
      "loss_step_1": 0.0748,
      "loss_step_2": 0.0743,
      "loss_step_3": 0.0621,
      "step": 2010
    },
    {
      "entropy_step_1": 0.2819,
      "entropy_step_2": 0.2818,
      "entropy_step_3": 0.2724,
      "epoch": 0.06008149668362035,
      "grad_norm": 79.62905883789062,
      "learning_rate": 9.997253262480185e-05,
      "loss": 0.07629975080490112,
      "loss_step_1": 0.084,
      "loss_step_2": 0.0836,
      "loss_step_3": 0.0813,
      "step": 2020
    },
    {
      "entropy_step_1": 0.2586,
      "entropy_step_2": 0.2605,
      "entropy_step_3": 0.2679,
      "epoch": 0.060378929835519464,
      "grad_norm": 53.24638366699219,
      "learning_rate": 9.997087848849032e-05,
      "loss": 0.07560193538665771,
      "loss_step_1": 0.0745,
      "loss_step_2": 0.0737,
      "loss_step_3": 0.0753,
      "step": 2030
    },
    {
      "entropy_step_1": 0.2609,
      "entropy_step_2": 0.2672,
      "entropy_step_3": 0.2648,
      "epoch": 0.060676362987418575,
      "grad_norm": 82.38983154296875,
      "learning_rate": 9.996917600467388e-05,
      "loss": 0.08616210222244262,
      "loss_step_1": 0.0879,
      "loss_step_2": 0.0864,
      "loss_step_3": 0.084,
      "step": 2040
    },
    {
      "entropy_step_1": 0.2752,
      "entropy_step_2": 0.2797,
      "entropy_step_3": 0.291,
      "epoch": 0.060973796139317686,
      "grad_norm": 29.80046844482422,
      "learning_rate": 9.996742517499974e-05,
      "loss": 0.07907070517539978,
      "loss_step_1": 0.0904,
      "loss_step_2": 0.0901,
      "loss_step_3": 0.1025,
      "step": 2050
    },
    {
      "entropy_step_1": 0.2472,
      "entropy_step_2": 0.2506,
      "entropy_step_3": 0.2445,
      "epoch": 0.0612712292912168,
      "grad_norm": 57.651405334472656,
      "learning_rate": 9.996562600116183e-05,
      "loss": 0.08514626622200012,
      "loss_step_1": 0.0873,
      "loss_step_2": 0.0866,
      "loss_step_3": 0.0886,
      "step": 2060
    },
    {
      "entropy_step_1": 0.3008,
      "entropy_step_2": 0.3029,
      "entropy_step_3": 0.3063,
      "epoch": 0.06156866244311591,
      "grad_norm": 90.48078155517578,
      "learning_rate": 9.99637784849009e-05,
      "loss": 0.08591211438179017,
      "loss_step_1": 0.0898,
      "loss_step_2": 0.0892,
      "loss_step_3": 0.0819,
      "step": 2070
    },
    {
      "entropy_step_1": 0.254,
      "entropy_step_2": 0.2553,
      "entropy_step_3": 0.2665,
      "epoch": 0.06186609559501502,
      "grad_norm": 103.04530334472656,
      "learning_rate": 9.996188262800442e-05,
      "loss": 0.07352520823478699,
      "loss_step_1": 0.0769,
      "loss_step_2": 0.0762,
      "loss_step_3": 0.0888,
      "step": 2080
    },
    {
      "entropy_step_1": 0.2275,
      "entropy_step_2": 0.2317,
      "entropy_step_3": 0.2298,
      "epoch": 0.06216352874691413,
      "grad_norm": 76.96748352050781,
      "learning_rate": 9.995993843230667e-05,
      "loss": 0.07201567888259888,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.0618,
      "step": 2090
    },
    {
      "entropy_step_1": 0.224,
      "entropy_step_2": 0.2235,
      "entropy_step_3": 0.2316,
      "epoch": 0.06246096189881324,
      "grad_norm": 73.55329132080078,
      "learning_rate": 9.995794589968867e-05,
      "loss": 0.07075282335281372,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0714,
      "step": 2100
    },
    {
      "entropy_step_1": 0.2886,
      "entropy_step_2": 0.2887,
      "entropy_step_3": 0.2983,
      "epoch": 0.06275839505071235,
      "grad_norm": 76.9638671875,
      "learning_rate": 9.995590503207824e-05,
      "loss": 0.08560768961906433,
      "loss_step_1": 0.0893,
      "loss_step_2": 0.0884,
      "loss_step_3": 0.0873,
      "step": 2110
    },
    {
      "entropy_step_1": 0.2421,
      "entropy_step_2": 0.2344,
      "entropy_step_3": 0.239,
      "epoch": 0.06305582820261146,
      "grad_norm": 61.485660552978516,
      "learning_rate": 9.995381583144996e-05,
      "loss": 0.07411361932754516,
      "loss_step_1": 0.0691,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0649,
      "step": 2120
    },
    {
      "entropy_step_1": 0.2358,
      "entropy_step_2": 0.2254,
      "entropy_step_3": 0.2126,
      "epoch": 0.06335326135451057,
      "grad_norm": 71.77091979980469,
      "learning_rate": 9.995167829982513e-05,
      "loss": 0.06646403670310974,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0635,
      "step": 2130
    },
    {
      "entropy_step_1": 0.2532,
      "entropy_step_2": 0.2471,
      "entropy_step_3": 0.2317,
      "epoch": 0.06365069450640969,
      "grad_norm": 46.42973709106445,
      "learning_rate": 9.994949243927186e-05,
      "loss": 0.08322973847389221,
      "loss_step_1": 0.077,
      "loss_step_2": 0.0775,
      "loss_step_3": 0.0646,
      "step": 2140
    },
    {
      "entropy_step_1": 0.2541,
      "entropy_step_2": 0.248,
      "entropy_step_3": 0.2277,
      "epoch": 0.06394812765830879,
      "grad_norm": 68.15711212158203,
      "learning_rate": 9.994725825190498e-05,
      "loss": 0.07281010746955871,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.0702,
      "loss_step_3": 0.0596,
      "step": 2150
    },
    {
      "entropy_step_1": 0.2259,
      "entropy_step_2": 0.2186,
      "entropy_step_3": 0.2118,
      "epoch": 0.06424556081020791,
      "grad_norm": 50.12639236450195,
      "learning_rate": 9.994497573988612e-05,
      "loss": 0.0690816581249237,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0536,
      "step": 2160
    },
    {
      "entropy_step_1": 0.2692,
      "entropy_step_2": 0.2597,
      "entropy_step_3": 0.2546,
      "epoch": 0.06454299396210701,
      "grad_norm": 114.05488586425781,
      "learning_rate": 9.994264490542363e-05,
      "loss": 0.08579813838005065,
      "loss_step_1": 0.0975,
      "loss_step_2": 0.0966,
      "loss_step_3": 0.0978,
      "step": 2170
    },
    {
      "entropy_step_1": 0.2525,
      "entropy_step_2": 0.2535,
      "entropy_step_3": 0.2467,
      "epoch": 0.06484042711400613,
      "grad_norm": 157.7943878173828,
      "learning_rate": 9.994026575077263e-05,
      "loss": 0.07390696406364441,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0701,
      "loss_step_3": 0.0695,
      "step": 2180
    },
    {
      "entropy_step_1": 0.2477,
      "entropy_step_2": 0.2519,
      "entropy_step_3": 0.2481,
      "epoch": 0.06513786026590523,
      "grad_norm": 98.37014770507812,
      "learning_rate": 9.993783827823497e-05,
      "loss": 0.07627924680709838,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0732,
      "loss_step_3": 0.0749,
      "step": 2190
    },
    {
      "entropy_step_1": 0.2641,
      "entropy_step_2": 0.2676,
      "entropy_step_3": 0.2699,
      "epoch": 0.06543529341780435,
      "grad_norm": 40.15657043457031,
      "learning_rate": 9.993536249015929e-05,
      "loss": 0.07660680413246154,
      "loss_step_1": 0.0803,
      "loss_step_2": 0.0804,
      "loss_step_3": 0.0777,
      "step": 2200
    },
    {
      "entropy_step_1": 0.2531,
      "entropy_step_2": 0.2635,
      "entropy_step_3": 0.252,
      "epoch": 0.06573272656970346,
      "grad_norm": 54.28097915649414,
      "learning_rate": 9.993283838894092e-05,
      "loss": 0.07664843201637268,
      "loss_step_1": 0.0783,
      "loss_step_2": 0.0783,
      "loss_step_3": 0.0679,
      "step": 2210
    },
    {
      "entropy_step_1": 0.2665,
      "entropy_step_2": 0.2793,
      "entropy_step_3": 0.297,
      "epoch": 0.06603015972160257,
      "grad_norm": 90.59759521484375,
      "learning_rate": 9.993026597702197e-05,
      "loss": 0.08822803497314453,
      "loss_step_1": 0.0947,
      "loss_step_2": 0.0937,
      "loss_step_3": 0.0966,
      "step": 2220
    },
    {
      "entropy_step_1": 0.2292,
      "entropy_step_2": 0.2388,
      "entropy_step_3": 0.2615,
      "epoch": 0.06632759287350168,
      "grad_norm": 71.55923461914062,
      "learning_rate": 9.992764525689132e-05,
      "loss": 0.08183076381683349,
      "loss_step_1": 0.0804,
      "loss_step_2": 0.0788,
      "loss_step_3": 0.0838,
      "step": 2230
    },
    {
      "entropy_step_1": 0.2634,
      "entropy_step_2": 0.2681,
      "entropy_step_3": 0.2606,
      "epoch": 0.0666250260254008,
      "grad_norm": 46.60572814941406,
      "learning_rate": 9.992497623108448e-05,
      "loss": 0.07265828847885132,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0698,
      "loss_step_3": 0.0702,
      "step": 2240
    },
    {
      "entropy_step_1": 0.2585,
      "entropy_step_2": 0.2645,
      "entropy_step_3": 0.2515,
      "epoch": 0.0669224591772999,
      "grad_norm": 51.877960205078125,
      "learning_rate": 9.992225890218381e-05,
      "loss": 0.08632049560546876,
      "loss_step_1": 0.0864,
      "loss_step_2": 0.0856,
      "loss_step_3": 0.0769,
      "step": 2250
    },
    {
      "entropy_step_1": 0.2669,
      "entropy_step_2": 0.2694,
      "entropy_step_3": 0.292,
      "epoch": 0.06721989232919902,
      "grad_norm": 40.40311813354492,
      "learning_rate": 9.991949327281839e-05,
      "loss": 0.07915863990783692,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.0729,
      "loss_step_3": 0.0886,
      "step": 2260
    },
    {
      "entropy_step_1": 0.266,
      "entropy_step_2": 0.2648,
      "entropy_step_3": 0.2443,
      "epoch": 0.06751732548109812,
      "grad_norm": 72.37147521972656,
      "learning_rate": 9.991667934566395e-05,
      "loss": 0.07806136608123779,
      "loss_step_1": 0.0788,
      "loss_step_2": 0.0789,
      "loss_step_3": 0.0676,
      "step": 2270
    },
    {
      "entropy_step_1": 0.227,
      "entropy_step_2": 0.2295,
      "entropy_step_3": 0.2436,
      "epoch": 0.06781475863299724,
      "grad_norm": 71.24931335449219,
      "learning_rate": 9.991381712344305e-05,
      "loss": 0.0648786962032318,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.065,
      "step": 2280
    },
    {
      "entropy_step_1": 0.2569,
      "entropy_step_2": 0.2549,
      "entropy_step_3": 0.2657,
      "epoch": 0.06811219178489634,
      "grad_norm": 24.83136749267578,
      "learning_rate": 9.991090660892488e-05,
      "loss": 0.07430261373519897,
      "loss_step_1": 0.078,
      "loss_step_2": 0.0783,
      "loss_step_3": 0.0923,
      "step": 2290
    },
    {
      "entropy_step_1": 0.2531,
      "entropy_step_2": 0.2546,
      "entropy_step_3": 0.2435,
      "epoch": 0.06840962493679545,
      "grad_norm": 62.52503204345703,
      "learning_rate": 9.990794780492545e-05,
      "loss": 0.07406471371650696,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0695,
      "loss_step_3": 0.064,
      "step": 2300
    },
    {
      "entropy_step_1": 0.2449,
      "entropy_step_2": 0.2436,
      "entropy_step_3": 0.2369,
      "epoch": 0.06870705808869457,
      "grad_norm": 71.5512924194336,
      "learning_rate": 9.990494071430742e-05,
      "loss": 0.0801632285118103,
      "loss_step_1": 0.076,
      "loss_step_2": 0.0758,
      "loss_step_3": 0.0685,
      "step": 2310
    },
    {
      "entropy_step_1": 0.2543,
      "entropy_step_2": 0.2504,
      "entropy_step_3": 0.2308,
      "epoch": 0.06900449124059367,
      "grad_norm": 70.04234313964844,
      "learning_rate": 9.990188533998019e-05,
      "loss": 0.07215031385421752,
      "loss_step_1": 0.0713,
      "loss_step_2": 0.0712,
      "loss_step_3": 0.0647,
      "step": 2320
    },
    {
      "entropy_step_1": 0.2482,
      "entropy_step_2": 0.2416,
      "entropy_step_3": 0.2293,
      "epoch": 0.06930192439249279,
      "grad_norm": 46.662723541259766,
      "learning_rate": 9.989878168489987e-05,
      "loss": 0.07144889831542969,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0696,
      "loss_step_3": 0.0695,
      "step": 2330
    },
    {
      "entropy_step_1": 0.2305,
      "entropy_step_2": 0.2266,
      "entropy_step_3": 0.239,
      "epoch": 0.0695993575443919,
      "grad_norm": 102.29576873779297,
      "learning_rate": 9.989562975206932e-05,
      "loss": 0.07663939595222473,
      "loss_step_1": 0.0748,
      "loss_step_2": 0.0738,
      "loss_step_3": 0.068,
      "step": 2340
    },
    {
      "entropy_step_1": 0.2479,
      "entropy_step_2": 0.2494,
      "entropy_step_3": 0.2499,
      "epoch": 0.06989679069629101,
      "grad_norm": 108.23655700683594,
      "learning_rate": 9.989242954453805e-05,
      "loss": 0.0775681495666504,
      "loss_step_1": 0.0812,
      "loss_step_2": 0.081,
      "loss_step_3": 0.0742,
      "step": 2350
    },
    {
      "entropy_step_1": 0.2251,
      "entropy_step_2": 0.228,
      "entropy_step_3": 0.2431,
      "epoch": 0.07019422384819012,
      "grad_norm": 93.65106201171875,
      "learning_rate": 9.98891810654023e-05,
      "loss": 0.0837700366973877,
      "loss_step_1": 0.0849,
      "loss_step_2": 0.0826,
      "loss_step_3": 0.0899,
      "step": 2360
    },
    {
      "entropy_step_1": 0.2602,
      "entropy_step_2": 0.2627,
      "entropy_step_3": 0.2612,
      "epoch": 0.07049165700008923,
      "grad_norm": 52.628787994384766,
      "learning_rate": 9.988588431780506e-05,
      "loss": 0.0748688042163849,
      "loss_step_1": 0.0785,
      "loss_step_2": 0.079,
      "loss_step_3": 0.0767,
      "step": 2370
    },
    {
      "entropy_step_1": 0.2552,
      "entropy_step_2": 0.2489,
      "entropy_step_3": 0.258,
      "epoch": 0.07078909015198834,
      "grad_norm": 69.4281234741211,
      "learning_rate": 9.988253930493592e-05,
      "loss": 0.08052289485931396,
      "loss_step_1": 0.0851,
      "loss_step_2": 0.0855,
      "loss_step_3": 0.0963,
      "step": 2380
    },
    {
      "entropy_step_1": 0.2651,
      "entropy_step_2": 0.2676,
      "entropy_step_3": 0.2625,
      "epoch": 0.07108652330388746,
      "grad_norm": 351.78759765625,
      "learning_rate": 9.987914603003124e-05,
      "loss": 0.07214398384094238,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0735,
      "loss_step_3": 0.075,
      "step": 2390
    },
    {
      "entropy_step_1": 0.2385,
      "entropy_step_2": 0.2441,
      "entropy_step_3": 0.2543,
      "epoch": 0.07138395645578656,
      "grad_norm": 54.00967025756836,
      "learning_rate": 9.98757044963741e-05,
      "loss": 0.07631233930587769,
      "loss_step_1": 0.078,
      "loss_step_2": 0.0773,
      "loss_step_3": 0.0807,
      "step": 2400
    },
    {
      "entropy_step_1": 0.2462,
      "entropy_step_2": 0.2494,
      "entropy_step_3": 0.2565,
      "epoch": 0.07168138960768568,
      "grad_norm": 51.84700012207031,
      "learning_rate": 9.987221470729418e-05,
      "loss": 0.0761151671409607,
      "loss_step_1": 0.0755,
      "loss_step_2": 0.0749,
      "loss_step_3": 0.0707,
      "step": 2410
    },
    {
      "entropy_step_1": 0.2506,
      "entropy_step_2": 0.2506,
      "entropy_step_3": 0.2362,
      "epoch": 0.07197882275958478,
      "grad_norm": 50.575218200683594,
      "learning_rate": 9.986867666616794e-05,
      "loss": 0.07270088791847229,
      "loss_step_1": 0.0723,
      "loss_step_2": 0.072,
      "loss_step_3": 0.0668,
      "step": 2420
    },
    {
      "entropy_step_1": 0.2176,
      "entropy_step_2": 0.2241,
      "entropy_step_3": 0.2207,
      "epoch": 0.0722762559114839,
      "grad_norm": 38.89208984375,
      "learning_rate": 9.986509037641845e-05,
      "loss": 0.07021186351776124,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.073,
      "loss_step_3": 0.0753,
      "step": 2430
    },
    {
      "entropy_step_1": 0.2128,
      "entropy_step_2": 0.2197,
      "entropy_step_3": 0.2222,
      "epoch": 0.072573689063383,
      "grad_norm": 93.26402282714844,
      "learning_rate": 9.986145584151552e-05,
      "loss": 0.06566426157951355,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0637,
      "step": 2440
    },
    {
      "entropy_step_1": 0.2284,
      "entropy_step_2": 0.2333,
      "entropy_step_3": 0.2158,
      "epoch": 0.07287112221528212,
      "grad_norm": 43.36237716674805,
      "learning_rate": 9.985777306497559e-05,
      "loss": 0.07505200505256653,
      "loss_step_1": 0.0768,
      "loss_step_2": 0.0755,
      "loss_step_3": 0.0687,
      "step": 2450
    },
    {
      "entropy_step_1": 0.2219,
      "entropy_step_2": 0.2267,
      "entropy_step_3": 0.2222,
      "epoch": 0.07316855536718123,
      "grad_norm": 79.95824432373047,
      "learning_rate": 9.98540420503618e-05,
      "loss": 0.07091587781906128,
      "loss_step_1": 0.0814,
      "loss_step_2": 0.0806,
      "loss_step_3": 0.0865,
      "step": 2460
    },
    {
      "entropy_step_1": 0.2471,
      "entropy_step_2": 0.2592,
      "entropy_step_3": 0.2622,
      "epoch": 0.07346598851908033,
      "grad_norm": 39.83952713012695,
      "learning_rate": 9.985026280128398e-05,
      "loss": 0.07930325269699097,
      "loss_step_1": 0.0844,
      "loss_step_2": 0.0841,
      "loss_step_3": 0.0808,
      "step": 2470
    },
    {
      "entropy_step_1": 0.2115,
      "entropy_step_2": 0.2283,
      "entropy_step_3": 0.2293,
      "epoch": 0.07376342167097945,
      "grad_norm": 43.369842529296875,
      "learning_rate": 9.984643532139858e-05,
      "loss": 0.07224557995796203,
      "loss_step_1": 0.0729,
      "loss_step_2": 0.0717,
      "loss_step_3": 0.0693,
      "step": 2480
    },
    {
      "entropy_step_1": 0.238,
      "entropy_step_2": 0.249,
      "entropy_step_3": 0.2364,
      "epoch": 0.07406085482287855,
      "grad_norm": 67.97593688964844,
      "learning_rate": 9.984255961440875e-05,
      "loss": 0.07395975589752198,
      "loss_step_1": 0.0725,
      "loss_step_2": 0.0732,
      "loss_step_3": 0.0716,
      "step": 2490
    },
    {
      "entropy_step_1": 0.2694,
      "entropy_step_2": 0.2718,
      "entropy_step_3": 0.2705,
      "epoch": 0.07435828797477767,
      "grad_norm": 76.93289184570312,
      "learning_rate": 9.983863568406428e-05,
      "loss": 0.09417676329612731,
      "loss_step_1": 0.0954,
      "loss_step_2": 0.094,
      "loss_step_3": 0.0852,
      "step": 2500
    },
    {
      "entropy_step_1": 0.2732,
      "entropy_step_2": 0.2754,
      "entropy_step_3": 0.3017,
      "epoch": 0.07465572112667677,
      "grad_norm": 109.51927947998047,
      "learning_rate": 9.983466353416164e-05,
      "loss": 0.07684734463691711,
      "loss_step_1": 0.0877,
      "loss_step_2": 0.0877,
      "loss_step_3": 0.0869,
      "step": 2510
    },
    {
      "entropy_step_1": 0.2629,
      "entropy_step_2": 0.2654,
      "entropy_step_3": 0.2609,
      "epoch": 0.07495315427857589,
      "grad_norm": 65.96898651123047,
      "learning_rate": 9.983064316854393e-05,
      "loss": 0.08178932070732117,
      "loss_step_1": 0.0911,
      "loss_step_2": 0.0911,
      "loss_step_3": 0.081,
      "step": 2520
    },
    {
      "entropy_step_1": 0.2441,
      "entropy_step_2": 0.2476,
      "entropy_step_3": 0.2227,
      "epoch": 0.075250587430475,
      "grad_norm": 69.8066635131836,
      "learning_rate": 9.982657459110093e-05,
      "loss": 0.06854516267776489,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0739,
      "loss_step_3": 0.0644,
      "step": 2530
    },
    {
      "entropy_step_1": 0.2211,
      "entropy_step_2": 0.226,
      "entropy_step_3": 0.2242,
      "epoch": 0.07554802058237411,
      "grad_norm": 58.97880172729492,
      "learning_rate": 9.9822457805769e-05,
      "loss": 0.06446611881256104,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0625,
      "step": 2540
    },
    {
      "entropy_step_1": 0.2425,
      "entropy_step_2": 0.2452,
      "entropy_step_3": 0.2647,
      "epoch": 0.07584545373427322,
      "grad_norm": 137.86660766601562,
      "learning_rate": 9.981829281653121e-05,
      "loss": 0.07635207772254944,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0695,
      "loss_step_3": 0.079,
      "step": 2550
    },
    {
      "entropy_step_1": 0.2599,
      "entropy_step_2": 0.2507,
      "entropy_step_3": 0.2339,
      "epoch": 0.07614288688617234,
      "grad_norm": 41.59444046020508,
      "learning_rate": 9.981407962741724e-05,
      "loss": 0.0763142466545105,
      "loss_step_1": 0.073,
      "loss_step_2": 0.0726,
      "loss_step_3": 0.0725,
      "step": 2560
    },
    {
      "entropy_step_1": 0.231,
      "entropy_step_2": 0.2221,
      "entropy_step_3": 0.2246,
      "epoch": 0.07644032003807144,
      "grad_norm": 61.2664909362793,
      "learning_rate": 9.980981824250342e-05,
      "loss": 0.07745084166526794,
      "loss_step_1": 0.0812,
      "loss_step_2": 0.0813,
      "loss_step_3": 0.0707,
      "step": 2570
    },
    {
      "entropy_step_1": 0.2346,
      "entropy_step_2": 0.2327,
      "entropy_step_3": 0.2585,
      "epoch": 0.07673775318997056,
      "grad_norm": 39.27815246582031,
      "learning_rate": 9.980550866591269e-05,
      "loss": 0.06693739295005799,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0746,
      "step": 2580
    },
    {
      "entropy_step_1": 0.2248,
      "entropy_step_2": 0.232,
      "entropy_step_3": 0.2139,
      "epoch": 0.07703518634186966,
      "grad_norm": 75.93663024902344,
      "learning_rate": 9.980115090181463e-05,
      "loss": 0.07848586440086365,
      "loss_step_1": 0.0777,
      "loss_step_2": 0.077,
      "loss_step_3": 0.0735,
      "step": 2590
    },
    {
      "entropy_step_1": 0.2648,
      "entropy_step_2": 0.2742,
      "entropy_step_3": 0.3047,
      "epoch": 0.07733261949376878,
      "grad_norm": 44.71162796020508,
      "learning_rate": 9.979674495442544e-05,
      "loss": 0.07544406652450561,
      "loss_step_1": 0.0766,
      "loss_step_2": 0.0767,
      "loss_step_3": 0.0847,
      "step": 2600
    },
    {
      "entropy_step_1": 0.2438,
      "entropy_step_2": 0.2518,
      "entropy_step_3": 0.2596,
      "epoch": 0.07763005264566789,
      "grad_norm": 61.615665435791016,
      "learning_rate": 9.97922908280079e-05,
      "loss": 0.07344512343406677,
      "loss_step_1": 0.0771,
      "loss_step_2": 0.0763,
      "loss_step_3": 0.0796,
      "step": 2610
    },
    {
      "entropy_step_1": 0.2311,
      "entropy_step_2": 0.2338,
      "entropy_step_3": 0.2459,
      "epoch": 0.077927485797567,
      "grad_norm": 51.64616394042969,
      "learning_rate": 9.978778852687147e-05,
      "loss": 0.06391108632087708,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0687,
      "step": 2620
    },
    {
      "entropy_step_1": 0.2558,
      "entropy_step_2": 0.2586,
      "entropy_step_3": 0.2476,
      "epoch": 0.07822491894946611,
      "grad_norm": 83.07377624511719,
      "learning_rate": 9.978323805537217e-05,
      "loss": 0.0635058343410492,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0617,
      "step": 2630
    },
    {
      "entropy_step_1": 0.2439,
      "entropy_step_2": 0.2507,
      "entropy_step_3": 0.2396,
      "epoch": 0.07852235210136521,
      "grad_norm": 60.09807586669922,
      "learning_rate": 9.977863941791267e-05,
      "loss": 0.07660959959030152,
      "loss_step_1": 0.0768,
      "loss_step_2": 0.0766,
      "loss_step_3": 0.0724,
      "step": 2640
    },
    {
      "entropy_step_1": 0.2488,
      "entropy_step_2": 0.2551,
      "entropy_step_3": 0.2515,
      "epoch": 0.07881978525326433,
      "grad_norm": 136.55624389648438,
      "learning_rate": 9.977399261894218e-05,
      "loss": 0.07617241740226746,
      "loss_step_1": 0.0796,
      "loss_step_2": 0.0793,
      "loss_step_3": 0.0679,
      "step": 2650
    },
    {
      "entropy_step_1": 0.2188,
      "entropy_step_2": 0.2175,
      "entropy_step_3": 0.2196,
      "epoch": 0.07911721840516343,
      "grad_norm": 108.3440170288086,
      "learning_rate": 9.976929766295658e-05,
      "loss": 0.06502090692520142,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0557,
      "step": 2660
    },
    {
      "entropy_step_1": 0.2439,
      "entropy_step_2": 0.2412,
      "entropy_step_3": 0.2368,
      "epoch": 0.07941465155706255,
      "grad_norm": 37.33821105957031,
      "learning_rate": 9.976455455449827e-05,
      "loss": 0.07513614296913147,
      "loss_step_1": 0.0819,
      "loss_step_2": 0.0813,
      "loss_step_3": 0.082,
      "step": 2670
    },
    {
      "entropy_step_1": 0.2292,
      "entropy_step_2": 0.2278,
      "entropy_step_3": 0.239,
      "epoch": 0.07971208470896166,
      "grad_norm": 61.43324661254883,
      "learning_rate": 9.975976329815627e-05,
      "loss": 0.07820314764976502,
      "loss_step_1": 0.077,
      "loss_step_2": 0.0763,
      "loss_step_3": 0.0826,
      "step": 2680
    },
    {
      "entropy_step_1": 0.274,
      "entropy_step_2": 0.2751,
      "entropy_step_3": 0.2722,
      "epoch": 0.08000951786086077,
      "grad_norm": 79.48824310302734,
      "learning_rate": 9.975492389856622e-05,
      "loss": 0.07714401483535767,
      "loss_step_1": 0.0831,
      "loss_step_2": 0.082,
      "loss_step_3": 0.0782,
      "step": 2690
    },
    {
      "entropy_step_1": 0.2487,
      "entropy_step_2": 0.2537,
      "entropy_step_3": 0.2606,
      "epoch": 0.08030695101275988,
      "grad_norm": 35.237030029296875,
      "learning_rate": 9.975003636041027e-05,
      "loss": 0.07440330386161804,
      "loss_step_1": 0.078,
      "loss_step_2": 0.0762,
      "loss_step_3": 0.0791,
      "step": 2700
    },
    {
      "entropy_step_1": 0.2472,
      "entropy_step_2": 0.2514,
      "entropy_step_3": 0.2597,
      "epoch": 0.080604384164659,
      "grad_norm": 64.64996337890625,
      "learning_rate": 9.97451006884172e-05,
      "loss": 0.07222386598587036,
      "loss_step_1": 0.0774,
      "loss_step_2": 0.0762,
      "loss_step_3": 0.0819,
      "step": 2710
    },
    {
      "entropy_step_1": 0.2687,
      "entropy_step_2": 0.2673,
      "entropy_step_3": 0.2525,
      "epoch": 0.0809018173165581,
      "grad_norm": 117.28914642333984,
      "learning_rate": 9.974011688736233e-05,
      "loss": 0.07415847182273864,
      "loss_step_1": 0.0731,
      "loss_step_2": 0.0729,
      "loss_step_3": 0.0692,
      "step": 2720
    },
    {
      "entropy_step_1": 0.2375,
      "entropy_step_2": 0.2357,
      "entropy_step_3": 0.2447,
      "epoch": 0.08119925046845722,
      "grad_norm": 64.34429168701172,
      "learning_rate": 9.973508496206757e-05,
      "loss": 0.07866135835647584,
      "loss_step_1": 0.0745,
      "loss_step_2": 0.0744,
      "loss_step_3": 0.0768,
      "step": 2730
    },
    {
      "entropy_step_1": 0.2416,
      "entropy_step_2": 0.2425,
      "entropy_step_3": 0.2265,
      "epoch": 0.08149668362035632,
      "grad_norm": 67.45742797851562,
      "learning_rate": 9.973000491740135e-05,
      "loss": 0.0745775043964386,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0687,
      "loss_step_3": 0.063,
      "step": 2740
    },
    {
      "entropy_step_1": 0.2646,
      "entropy_step_2": 0.2644,
      "entropy_step_3": 0.2596,
      "epoch": 0.08179411677225544,
      "grad_norm": 34.127098083496094,
      "learning_rate": 9.972487675827871e-05,
      "loss": 0.07211064100265503,
      "loss_step_1": 0.0828,
      "loss_step_2": 0.0827,
      "loss_step_3": 0.0761,
      "step": 2750
    },
    {
      "entropy_step_1": 0.2178,
      "entropy_step_2": 0.2227,
      "entropy_step_3": 0.2344,
      "epoch": 0.08209154992415454,
      "grad_norm": 64.89350128173828,
      "learning_rate": 9.971970048966119e-05,
      "loss": 0.08188138604164123,
      "loss_step_1": 0.0755,
      "loss_step_2": 0.0744,
      "loss_step_3": 0.075,
      "step": 2760
    },
    {
      "entropy_step_1": 0.2402,
      "entropy_step_2": 0.2487,
      "entropy_step_3": 0.2554,
      "epoch": 0.08238898307605366,
      "grad_norm": 87.28723907470703,
      "learning_rate": 9.971447611655693e-05,
      "loss": 0.0659691035747528,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0628,
      "step": 2770
    },
    {
      "entropy_step_1": 0.2489,
      "entropy_step_2": 0.2554,
      "entropy_step_3": 0.266,
      "epoch": 0.08268641622795277,
      "grad_norm": 89.0658950805664,
      "learning_rate": 9.970920364402054e-05,
      "loss": 0.08060733079910279,
      "loss_step_1": 0.0778,
      "loss_step_2": 0.0777,
      "loss_step_3": 0.0809,
      "step": 2780
    },
    {
      "entropy_step_1": 0.2392,
      "entropy_step_2": 0.2473,
      "entropy_step_3": 0.2547,
      "epoch": 0.08298384937985188,
      "grad_norm": 50.316375732421875,
      "learning_rate": 9.970388307715326e-05,
      "loss": 0.0801543116569519,
      "loss_step_1": 0.0743,
      "loss_step_2": 0.0739,
      "loss_step_3": 0.0728,
      "step": 2790
    },
    {
      "entropy_step_1": 0.2703,
      "entropy_step_2": 0.2805,
      "entropy_step_3": 0.2802,
      "epoch": 0.08328128253175099,
      "grad_norm": 62.11091613769531,
      "learning_rate": 9.969851442110277e-05,
      "loss": 0.08444404006004333,
      "loss_step_1": 0.0839,
      "loss_step_2": 0.0834,
      "loss_step_3": 0.0857,
      "step": 2800
    },
    {
      "entropy_step_1": 0.2201,
      "entropy_step_2": 0.2262,
      "entropy_step_3": 0.2461,
      "epoch": 0.08357871568365009,
      "grad_norm": 60.31355667114258,
      "learning_rate": 9.969309768106332e-05,
      "loss": 0.07059417366981506,
      "loss_step_1": 0.073,
      "loss_step_2": 0.0721,
      "loss_step_3": 0.0861,
      "step": 2810
    },
    {
      "entropy_step_1": 0.2479,
      "entropy_step_2": 0.258,
      "entropy_step_3": 0.2401,
      "epoch": 0.08387614883554921,
      "grad_norm": 70.577880859375,
      "learning_rate": 9.968763286227571e-05,
      "loss": 0.0755907118320465,
      "loss_step_1": 0.0769,
      "loss_step_2": 0.0758,
      "loss_step_3": 0.0662,
      "step": 2820
    },
    {
      "entropy_step_1": 0.2415,
      "entropy_step_2": 0.2497,
      "entropy_step_3": 0.243,
      "epoch": 0.08417358198744831,
      "grad_norm": 32.18790817260742,
      "learning_rate": 9.968211997002718e-05,
      "loss": 0.06934824585914612,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0699,
      "loss_step_3": 0.0659,
      "step": 2830
    },
    {
      "entropy_step_1": 0.1937,
      "entropy_step_2": 0.1953,
      "entropy_step_3": 0.1939,
      "epoch": 0.08447101513934743,
      "grad_norm": 328.107666015625,
      "learning_rate": 9.967655900965158e-05,
      "loss": 0.06334558129310608,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0498,
      "step": 2840
    },
    {
      "entropy_step_1": 0.2548,
      "entropy_step_2": 0.2551,
      "entropy_step_3": 0.2533,
      "epoch": 0.08476844829124654,
      "grad_norm": 45.207763671875,
      "learning_rate": 9.967094998652919e-05,
      "loss": 0.0788927435874939,
      "loss_step_1": 0.0781,
      "loss_step_2": 0.0778,
      "loss_step_3": 0.0789,
      "step": 2850
    },
    {
      "entropy_step_1": 0.2178,
      "entropy_step_2": 0.2211,
      "entropy_step_3": 0.2187,
      "epoch": 0.08506588144314566,
      "grad_norm": 80.93568420410156,
      "learning_rate": 9.966529290608681e-05,
      "loss": 0.0702759325504303,
      "loss_step_1": 0.0792,
      "loss_step_2": 0.0775,
      "loss_step_3": 0.0748,
      "step": 2860
    },
    {
      "entropy_step_1": 0.2323,
      "entropy_step_2": 0.2432,
      "entropy_step_3": 0.2485,
      "epoch": 0.08536331459504476,
      "grad_norm": 43.99787902832031,
      "learning_rate": 9.965958777379776e-05,
      "loss": 0.07670790553092957,
      "loss_step_1": 0.0827,
      "loss_step_2": 0.0823,
      "loss_step_3": 0.0755,
      "step": 2870
    },
    {
      "entropy_step_1": 0.2531,
      "entropy_step_2": 0.2701,
      "entropy_step_3": 0.2701,
      "epoch": 0.08566074774694388,
      "grad_norm": 37.72101593017578,
      "learning_rate": 9.96538345951818e-05,
      "loss": 0.07555044293403626,
      "loss_step_1": 0.0788,
      "loss_step_2": 0.0777,
      "loss_step_3": 0.0862,
      "step": 2880
    },
    {
      "entropy_step_1": 0.2411,
      "entropy_step_2": 0.2556,
      "entropy_step_3": 0.2528,
      "epoch": 0.08595818089884298,
      "grad_norm": 82.98299407958984,
      "learning_rate": 9.964803337580526e-05,
      "loss": 0.0729517936706543,
      "loss_step_1": 0.0821,
      "loss_step_2": 0.0815,
      "loss_step_3": 0.0797,
      "step": 2890
    },
    {
      "entropy_step_1": 0.2359,
      "entropy_step_2": 0.2416,
      "entropy_step_3": 0.2356,
      "epoch": 0.0862556140507421,
      "grad_norm": 109.83242797851562,
      "learning_rate": 9.964218412128086e-05,
      "loss": 0.0693914532661438,
      "loss_step_1": 0.0718,
      "loss_step_2": 0.0716,
      "loss_step_3": 0.0633,
      "step": 2900
    },
    {
      "entropy_step_1": 0.2768,
      "entropy_step_2": 0.2811,
      "entropy_step_3": 0.2826,
      "epoch": 0.0865530472026412,
      "grad_norm": 72.14924621582031,
      "learning_rate": 9.963628683726782e-05,
      "loss": 0.08535857200622558,
      "loss_step_1": 0.0897,
      "loss_step_2": 0.0892,
      "loss_step_3": 0.0885,
      "step": 2910
    },
    {
      "entropy_step_1": 0.2311,
      "entropy_step_2": 0.2361,
      "entropy_step_3": 0.2448,
      "epoch": 0.08685048035454032,
      "grad_norm": 68.47631072998047,
      "learning_rate": 9.963034152947189e-05,
      "loss": 0.0768608272075653,
      "loss_step_1": 0.0752,
      "loss_step_2": 0.0746,
      "loss_step_3": 0.0827,
      "step": 2920
    },
    {
      "entropy_step_1": 0.2368,
      "entropy_step_2": 0.2463,
      "entropy_step_3": 0.216,
      "epoch": 0.08714791350643943,
      "grad_norm": 69.8570556640625,
      "learning_rate": 9.962434820364522e-05,
      "loss": 0.0659399688243866,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0547,
      "step": 2930
    },
    {
      "entropy_step_1": 0.2257,
      "entropy_step_2": 0.2384,
      "entropy_step_3": 0.2336,
      "epoch": 0.08744534665833854,
      "grad_norm": 76.96154022216797,
      "learning_rate": 9.96183068655864e-05,
      "loss": 0.07622541189193725,
      "loss_step_1": 0.0771,
      "loss_step_2": 0.0756,
      "loss_step_3": 0.0862,
      "step": 2940
    },
    {
      "entropy_step_1": 0.2276,
      "entropy_step_2": 0.2345,
      "entropy_step_3": 0.2223,
      "epoch": 0.08774277981023765,
      "grad_norm": 36.92588806152344,
      "learning_rate": 9.961221752114054e-05,
      "loss": 0.0643959403038025,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.057,
      "step": 2950
    },
    {
      "entropy_step_1": 0.2317,
      "entropy_step_2": 0.2321,
      "entropy_step_3": 0.2357,
      "epoch": 0.08804021296213677,
      "grad_norm": 25.4630069732666,
      "learning_rate": 9.960608017619917e-05,
      "loss": 0.06928234696388244,
      "loss_step_1": 0.0726,
      "loss_step_2": 0.0717,
      "loss_step_3": 0.0701,
      "step": 2960
    },
    {
      "entropy_step_1": 0.2256,
      "entropy_step_2": 0.2265,
      "entropy_step_3": 0.2518,
      "epoch": 0.08833764611403587,
      "grad_norm": 117.22377014160156,
      "learning_rate": 9.959989483670023e-05,
      "loss": 0.07012505531311035,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.078,
      "step": 2970
    },
    {
      "entropy_step_1": 0.2338,
      "entropy_step_2": 0.2344,
      "entropy_step_3": 0.2379,
      "epoch": 0.08863507926593499,
      "grad_norm": 38.577205657958984,
      "learning_rate": 9.959366150862813e-05,
      "loss": 0.07681465148925781,
      "loss_step_1": 0.0774,
      "loss_step_2": 0.0769,
      "loss_step_3": 0.0792,
      "step": 2980
    },
    {
      "entropy_step_1": 0.2933,
      "entropy_step_2": 0.2965,
      "entropy_step_3": 0.2948,
      "epoch": 0.08893251241783409,
      "grad_norm": 37.55236053466797,
      "learning_rate": 9.95873801980137e-05,
      "loss": 0.08828416466712952,
      "loss_step_1": 0.0909,
      "loss_step_2": 0.0903,
      "loss_step_3": 0.0854,
      "step": 2990
    },
    {
      "entropy_step_1": 0.2607,
      "entropy_step_2": 0.2632,
      "entropy_step_3": 0.2797,
      "epoch": 0.0892299455697332,
      "grad_norm": 73.1404800415039,
      "learning_rate": 9.95810509109342e-05,
      "loss": 0.0813741385936737,
      "loss_step_1": 0.0759,
      "loss_step_2": 0.0759,
      "loss_step_3": 0.0821,
      "step": 3000
    },
    {
      "epoch": 0.0892299455697332,
      "eval_accuracy": 0.705,
      "eval_f1": 0.6928359983715102,
      "eval_loss": 0.22351275384426117,
      "eval_precision": 0.7435848079125851,
      "eval_recall": 0.705,
      "eval_runtime": 6.3407,
      "eval_samples_per_second": 78.855,
      "eval_steps_per_second": 9.936,
      "step": 3000
    },
    {
      "entropy_step_1": 0.2436,
      "entropy_step_2": 0.2366,
      "entropy_step_3": 0.2341,
      "epoch": 0.08952737872163231,
      "grad_norm": 75.14173126220703,
      "learning_rate": 9.957467365351328e-05,
      "loss": 0.06646546721458435,
      "loss_step_1": 0.0741,
      "loss_step_2": 0.0739,
      "loss_step_3": 0.0741,
      "step": 3010
    },
    {
      "entropy_step_1": 0.228,
      "entropy_step_2": 0.2233,
      "entropy_step_3": 0.2349,
      "epoch": 0.08982481187353142,
      "grad_norm": 41.23202133178711,
      "learning_rate": 9.956824843192105e-05,
      "loss": 0.07197875380516053,
      "loss_step_1": 0.0744,
      "loss_step_2": 0.0744,
      "loss_step_3": 0.0682,
      "step": 3020
    },
    {
      "entropy_step_1": 0.2544,
      "entropy_step_2": 0.2507,
      "entropy_step_3": 0.2416,
      "epoch": 0.09012224502543054,
      "grad_norm": 78.43931579589844,
      "learning_rate": 9.9561775252374e-05,
      "loss": 0.0799466073513031,
      "loss_step_1": 0.0709,
      "loss_step_2": 0.0708,
      "loss_step_3": 0.0654,
      "step": 3030
    },
    {
      "entropy_step_1": 0.2149,
      "entropy_step_2": 0.2125,
      "entropy_step_3": 0.2264,
      "epoch": 0.09041967817732964,
      "grad_norm": 36.23701477050781,
      "learning_rate": 9.9555254121135e-05,
      "loss": 0.06360053420066833,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0646,
      "step": 3040
    },
    {
      "entropy_step_1": 0.2378,
      "entropy_step_2": 0.2329,
      "entropy_step_3": 0.2296,
      "epoch": 0.09071711132922876,
      "grad_norm": 81.32477569580078,
      "learning_rate": 9.954868504451333e-05,
      "loss": 0.06387787461280822,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0605,
      "step": 3050
    },
    {
      "entropy_step_1": 0.2175,
      "entropy_step_2": 0.2139,
      "entropy_step_3": 0.2033,
      "epoch": 0.09101454448112786,
      "grad_norm": 84.28453063964844,
      "learning_rate": 9.954206802886469e-05,
      "loss": 0.0780889093875885,
      "loss_step_1": 0.0876,
      "loss_step_2": 0.0874,
      "loss_step_3": 0.0754,
      "step": 3060
    },
    {
      "entropy_step_1": 0.2326,
      "entropy_step_2": 0.2336,
      "entropy_step_3": 0.2396,
      "epoch": 0.09131197763302698,
      "grad_norm": 106.6060562133789,
      "learning_rate": 9.953540308059111e-05,
      "loss": 0.07259621620178222,
      "loss_step_1": 0.0742,
      "loss_step_2": 0.073,
      "loss_step_3": 0.0776,
      "step": 3070
    },
    {
      "entropy_step_1": 0.2352,
      "entropy_step_2": 0.2417,
      "entropy_step_3": 0.2441,
      "epoch": 0.09160941078492608,
      "grad_norm": 50.737876892089844,
      "learning_rate": 9.952869020614102e-05,
      "loss": 0.07153601050376893,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0739,
      "loss_step_3": 0.0755,
      "step": 3080
    },
    {
      "entropy_step_1": 0.1915,
      "entropy_step_2": 0.1976,
      "entropy_step_3": 0.2012,
      "epoch": 0.0919068439368252,
      "grad_norm": 51.61606216430664,
      "learning_rate": 9.952192941200921e-05,
      "loss": 0.07550274729728698,
      "loss_step_1": 0.0762,
      "loss_step_2": 0.0766,
      "loss_step_3": 0.0869,
      "step": 3090
    },
    {
      "entropy_step_1": 0.2386,
      "entropy_step_2": 0.2397,
      "entropy_step_3": 0.2251,
      "epoch": 0.0922042770887243,
      "grad_norm": 110.19373321533203,
      "learning_rate": 9.951512070473685e-05,
      "loss": 0.06965585350990296,
      "loss_step_1": 0.0767,
      "loss_step_2": 0.0765,
      "loss_step_3": 0.0661,
      "step": 3100
    },
    {
      "entropy_step_1": 0.2436,
      "entropy_step_2": 0.2445,
      "entropy_step_3": 0.244,
      "epoch": 0.09250171024062342,
      "grad_norm": 74.94062042236328,
      "learning_rate": 9.950826409091148e-05,
      "loss": 0.06187580823898316,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.067,
      "step": 3110
    },
    {
      "entropy_step_1": 0.2366,
      "entropy_step_2": 0.2377,
      "entropy_step_3": 0.2416,
      "epoch": 0.09279914339252253,
      "grad_norm": 72.52946472167969,
      "learning_rate": 9.950135957716693e-05,
      "loss": 0.07065063118934631,
      "loss_step_1": 0.0666,
      "loss_step_2": 0.0662,
      "loss_step_3": 0.0586,
      "step": 3120
    },
    {
      "entropy_step_1": 0.2059,
      "entropy_step_2": 0.2068,
      "entropy_step_3": 0.2266,
      "epoch": 0.09309657654442165,
      "grad_norm": 58.135738372802734,
      "learning_rate": 9.949440717018343e-05,
      "loss": 0.06438869833946229,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0739,
      "step": 3130
    },
    {
      "entropy_step_1": 0.2559,
      "entropy_step_2": 0.2622,
      "entropy_step_3": 0.2581,
      "epoch": 0.09339400969632075,
      "grad_norm": 58.41170120239258,
      "learning_rate": 9.948740687668751e-05,
      "loss": 0.08012637495994568,
      "loss_step_1": 0.082,
      "loss_step_2": 0.0816,
      "loss_step_3": 0.0775,
      "step": 3140
    },
    {
      "entropy_step_1": 0.2481,
      "entropy_step_2": 0.2598,
      "entropy_step_3": 0.2606,
      "epoch": 0.09369144284821987,
      "grad_norm": 92.40338897705078,
      "learning_rate": 9.948035870345208e-05,
      "loss": 0.07408499717712402,
      "loss_step_1": 0.0756,
      "loss_step_2": 0.0752,
      "loss_step_3": 0.0829,
      "step": 3150
    },
    {
      "entropy_step_1": 0.2393,
      "entropy_step_2": 0.2467,
      "entropy_step_3": 0.2514,
      "epoch": 0.09398887600011897,
      "grad_norm": 132.43251037597656,
      "learning_rate": 9.947326265729632e-05,
      "loss": 0.0683690845966339,
      "loss_step_1": 0.0725,
      "loss_step_2": 0.0714,
      "loss_step_3": 0.0721,
      "step": 3160
    },
    {
      "entropy_step_1": 0.232,
      "entropy_step_2": 0.2314,
      "entropy_step_3": 0.2388,
      "epoch": 0.09428630915201808,
      "grad_norm": 78.08355712890625,
      "learning_rate": 9.946611874508576e-05,
      "loss": 0.07405186295509339,
      "loss_step_1": 0.0738,
      "loss_step_2": 0.0735,
      "loss_step_3": 0.0784,
      "step": 3170
    },
    {
      "entropy_step_1": 0.2736,
      "entropy_step_2": 0.2731,
      "entropy_step_3": 0.2908,
      "epoch": 0.0945837423039172,
      "grad_norm": 175.3792266845703,
      "learning_rate": 9.945892697373223e-05,
      "loss": 0.07923247814178466,
      "loss_step_1": 0.076,
      "loss_step_2": 0.076,
      "loss_step_3": 0.0842,
      "step": 3180
    },
    {
      "entropy_step_1": 0.2379,
      "entropy_step_2": 0.2305,
      "entropy_step_3": 0.2283,
      "epoch": 0.0948811754558163,
      "grad_norm": 53.54227066040039,
      "learning_rate": 9.945168735019386e-05,
      "loss": 0.07043417096138001,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0703,
      "step": 3190
    },
    {
      "entropy_step_1": 0.2094,
      "entropy_step_2": 0.2042,
      "entropy_step_3": 0.2244,
      "epoch": 0.09517860860771542,
      "grad_norm": 33.83897018432617,
      "learning_rate": 9.944439988147509e-05,
      "loss": 0.07055318355560303,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0869,
      "step": 3200
    },
    {
      "entropy_step_1": 0.2687,
      "entropy_step_2": 0.2592,
      "entropy_step_3": 0.2739,
      "epoch": 0.09547604175961452,
      "grad_norm": 44.17039108276367,
      "learning_rate": 9.943706457462666e-05,
      "loss": 0.07731128931045532,
      "loss_step_1": 0.0764,
      "loss_step_2": 0.0756,
      "loss_step_3": 0.0688,
      "step": 3210
    },
    {
      "entropy_step_1": 0.2327,
      "entropy_step_2": 0.2232,
      "entropy_step_3": 0.2318,
      "epoch": 0.09577347491151364,
      "grad_norm": 75.48301696777344,
      "learning_rate": 9.942968143674556e-05,
      "loss": 0.07332245707511902,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.0701,
      "step": 3220
    },
    {
      "entropy_step_1": 0.2406,
      "entropy_step_2": 0.2285,
      "entropy_step_3": 0.2443,
      "epoch": 0.09607090806341274,
      "grad_norm": 42.74138259887695,
      "learning_rate": 9.942225047497508e-05,
      "loss": 0.07878889441490174,
      "loss_step_1": 0.0841,
      "loss_step_2": 0.0829,
      "loss_step_3": 0.0802,
      "step": 3230
    },
    {
      "entropy_step_1": 0.2313,
      "entropy_step_2": 0.2283,
      "entropy_step_3": 0.2375,
      "epoch": 0.09636834121531186,
      "grad_norm": 153.071533203125,
      "learning_rate": 9.941477169650477e-05,
      "loss": 0.07998066544532775,
      "loss_step_1": 0.0896,
      "loss_step_2": 0.0889,
      "loss_step_3": 0.0756,
      "step": 3240
    },
    {
      "entropy_step_1": 0.2558,
      "entropy_step_2": 0.2658,
      "entropy_step_3": 0.2573,
      "epoch": 0.09666577436721097,
      "grad_norm": 43.1906623840332,
      "learning_rate": 9.940724510857046e-05,
      "loss": 0.0700314462184906,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.0736,
      "loss_step_3": 0.0747,
      "step": 3250
    },
    {
      "entropy_step_1": 0.2434,
      "entropy_step_2": 0.2538,
      "entropy_step_3": 0.2535,
      "epoch": 0.09696320751911008,
      "grad_norm": 74.72119903564453,
      "learning_rate": 9.939967071845423e-05,
      "loss": 0.07329361438751221,
      "loss_step_1": 0.0665,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0602,
      "step": 3260
    },
    {
      "entropy_step_1": 0.22,
      "entropy_step_2": 0.2169,
      "entropy_step_3": 0.2132,
      "epoch": 0.09726064067100919,
      "grad_norm": 63.93959426879883,
      "learning_rate": 9.93920485334844e-05,
      "loss": 0.06559511423110961,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0473,
      "step": 3270
    },
    {
      "entropy_step_1": 0.2259,
      "entropy_step_2": 0.2127,
      "entropy_step_3": 0.2239,
      "epoch": 0.0975580738229083,
      "grad_norm": 76.99048614501953,
      "learning_rate": 9.938437856103556e-05,
      "loss": 0.07277519702911377,
      "loss_step_1": 0.0744,
      "loss_step_2": 0.0756,
      "loss_step_3": 0.0734,
      "step": 3280
    },
    {
      "entropy_step_1": 0.2241,
      "entropy_step_2": 0.2152,
      "entropy_step_3": 0.2277,
      "epoch": 0.09785550697480741,
      "grad_norm": 95.83087158203125,
      "learning_rate": 9.937666080852845e-05,
      "loss": 0.06448097825050354,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0611,
      "step": 3290
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.1918,
      "epoch": 0.09815294012670653,
      "grad_norm": 43.664886474609375,
      "learning_rate": 9.936889528343017e-05,
      "loss": 0.05906238555908203,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.057,
      "step": 3300
    },
    {
      "entropy_step_1": 0.2423,
      "entropy_step_2": 0.2356,
      "entropy_step_3": 0.2454,
      "epoch": 0.09845037327860563,
      "grad_norm": 86.1562271118164,
      "learning_rate": 9.936108199325393e-05,
      "loss": 0.06829200983047486,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0693,
      "step": 3310
    },
    {
      "entropy_step_1": 0.2294,
      "entropy_step_2": 0.2185,
      "entropy_step_3": 0.2241,
      "epoch": 0.09874780643050475,
      "grad_norm": 68.32661437988281,
      "learning_rate": 9.935322094555921e-05,
      "loss": 0.07014741897583007,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0655,
      "step": 3320
    },
    {
      "entropy_step_1": 0.2296,
      "entropy_step_2": 0.2232,
      "entropy_step_3": 0.2274,
      "epoch": 0.09904523958240385,
      "grad_norm": 110.57166290283203,
      "learning_rate": 9.934531214795167e-05,
      "loss": 0.06195187568664551,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0577,
      "step": 3330
    },
    {
      "entropy_step_1": 0.2432,
      "entropy_step_2": 0.2434,
      "entropy_step_3": 0.2562,
      "epoch": 0.09934267273430296,
      "grad_norm": 64.0654067993164,
      "learning_rate": 9.93373556080832e-05,
      "loss": 0.07558727860450745,
      "loss_step_1": 0.0749,
      "loss_step_2": 0.0745,
      "loss_step_3": 0.0844,
      "step": 3340
    },
    {
      "entropy_step_1": 0.2209,
      "entropy_step_2": 0.2238,
      "entropy_step_3": 0.2117,
      "epoch": 0.09964010588620208,
      "grad_norm": 54.84423828125,
      "learning_rate": 9.932935133365184e-05,
      "loss": 0.06997796893119812,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0687,
      "loss_step_3": 0.0584,
      "step": 3350
    },
    {
      "entropy_step_1": 0.2171,
      "entropy_step_2": 0.2168,
      "entropy_step_3": 0.2287,
      "epoch": 0.09993753903810118,
      "grad_norm": 49.83549499511719,
      "learning_rate": 9.932129933240184e-05,
      "loss": 0.06757780909538269,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0546,
      "step": 3360
    },
    {
      "entropy_step_1": 0.2293,
      "entropy_step_2": 0.2336,
      "entropy_step_3": 0.2491,
      "epoch": 0.1002349721900003,
      "grad_norm": 64.41304779052734,
      "learning_rate": 9.931319961212361e-05,
      "loss": 0.07221124172210694,
      "loss_step_1": 0.0718,
      "loss_step_2": 0.0716,
      "loss_step_3": 0.0752,
      "step": 3370
    },
    {
      "entropy_step_1": 0.2606,
      "entropy_step_2": 0.269,
      "entropy_step_3": 0.277,
      "epoch": 0.1005324053418994,
      "grad_norm": 73.49200439453125,
      "learning_rate": 9.930505218065379e-05,
      "loss": 0.07262219190597534,
      "loss_step_1": 0.0784,
      "loss_step_2": 0.0783,
      "loss_step_3": 0.0789,
      "step": 3380
    },
    {
      "entropy_step_1": 0.2192,
      "entropy_step_2": 0.2248,
      "entropy_step_3": 0.2144,
      "epoch": 0.10082983849379852,
      "grad_norm": 57.0296516418457,
      "learning_rate": 9.929685704587506e-05,
      "loss": 0.06897035241127014,
      "loss_step_1": 0.07,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0704,
      "step": 3390
    },
    {
      "entropy_step_1": 0.2236,
      "entropy_step_2": 0.2191,
      "entropy_step_3": 0.2344,
      "epoch": 0.10112727164569763,
      "grad_norm": 37.8204231262207,
      "learning_rate": 9.928861421571636e-05,
      "loss": 0.06921069622039795,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0625,
      "step": 3400
    },
    {
      "entropy_step_1": 0.2142,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.2248,
      "epoch": 0.10142470479759674,
      "grad_norm": 85.19928741455078,
      "learning_rate": 9.928032369815274e-05,
      "loss": 0.06535682678222657,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0692,
      "loss_step_3": 0.0835,
      "step": 3410
    },
    {
      "entropy_step_1": 0.2673,
      "entropy_step_2": 0.2631,
      "entropy_step_3": 0.2671,
      "epoch": 0.10172213794949585,
      "grad_norm": 68.49748992919922,
      "learning_rate": 9.92719855012054e-05,
      "loss": 0.07386579513549804,
      "loss_step_1": 0.0777,
      "loss_step_2": 0.0776,
      "loss_step_3": 0.0777,
      "step": 3420
    },
    {
      "entropy_step_1": 0.2159,
      "entropy_step_2": 0.2102,
      "entropy_step_3": 0.1935,
      "epoch": 0.10201957110139497,
      "grad_norm": 25.71999168395996,
      "learning_rate": 9.926359963294162e-05,
      "loss": 0.07511736750602722,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0549,
      "step": 3430
    },
    {
      "entropy_step_1": 0.2383,
      "entropy_step_2": 0.2302,
      "entropy_step_3": 0.2137,
      "epoch": 0.10231700425329407,
      "grad_norm": 94.20822143554688,
      "learning_rate": 9.92551661014749e-05,
      "loss": 0.07046036124229431,
      "loss_step_1": 0.072,
      "loss_step_2": 0.0714,
      "loss_step_3": 0.0797,
      "step": 3440
    },
    {
      "entropy_step_1": 0.2442,
      "entropy_step_2": 0.2375,
      "entropy_step_3": 0.227,
      "epoch": 0.10261443740519319,
      "grad_norm": 62.989013671875,
      "learning_rate": 9.924668491496474e-05,
      "loss": 0.0731191098690033,
      "loss_step_1": 0.079,
      "loss_step_2": 0.0793,
      "loss_step_3": 0.073,
      "step": 3450
    },
    {
      "entropy_step_1": 0.2389,
      "entropy_step_2": 0.2406,
      "entropy_step_3": 0.2371,
      "epoch": 0.10291187055709229,
      "grad_norm": 65.50613403320312,
      "learning_rate": 9.923815608161684e-05,
      "loss": 0.0716052770614624,
      "loss_step_1": 0.0765,
      "loss_step_2": 0.0763,
      "loss_step_3": 0.071,
      "step": 3460
    },
    {
      "entropy_step_1": 0.2248,
      "entropy_step_2": 0.2319,
      "entropy_step_3": 0.2202,
      "epoch": 0.10320930370899141,
      "grad_norm": 607.02099609375,
      "learning_rate": 9.922957960968293e-05,
      "loss": 0.07119444012641907,
      "loss_step_1": 0.0768,
      "loss_step_2": 0.0741,
      "loss_step_3": 0.0786,
      "step": 3470
    },
    {
      "entropy_step_1": 0.2312,
      "entropy_step_2": 0.2425,
      "entropy_step_3": 0.2337,
      "epoch": 0.10350673686089051,
      "grad_norm": 95.19959259033203,
      "learning_rate": 9.922095550746089e-05,
      "loss": 0.06904032826423645,
      "loss_step_1": 0.0691,
      "loss_step_2": 0.0692,
      "loss_step_3": 0.0635,
      "step": 3480
    },
    {
      "entropy_step_1": 0.2101,
      "entropy_step_2": 0.2238,
      "entropy_step_3": 0.2324,
      "epoch": 0.10380417001278963,
      "grad_norm": 101.0452880859375,
      "learning_rate": 9.921228378329466e-05,
      "loss": 0.06267485618591309,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0632,
      "step": 3490
    },
    {
      "entropy_step_1": 0.224,
      "entropy_step_2": 0.2338,
      "entropy_step_3": 0.2348,
      "epoch": 0.10410160316468874,
      "grad_norm": 66.97811126708984,
      "learning_rate": 9.920356444557423e-05,
      "loss": 0.07382463812828063,
      "loss_step_1": 0.0784,
      "loss_step_2": 0.0764,
      "loss_step_3": 0.0796,
      "step": 3500
    },
    {
      "entropy_step_1": 0.246,
      "entropy_step_2": 0.2504,
      "entropy_step_3": 0.2679,
      "epoch": 0.10439903631658785,
      "grad_norm": 76.41088104248047,
      "learning_rate": 9.919479750273568e-05,
      "loss": 0.06931242942810059,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0654,
      "loss_step_3": 0.071,
      "step": 3510
    },
    {
      "entropy_step_1": 0.2407,
      "entropy_step_2": 0.2349,
      "entropy_step_3": 0.2363,
      "epoch": 0.10469646946848696,
      "grad_norm": 67.52484130859375,
      "learning_rate": 9.918598296326115e-05,
      "loss": 0.07431666254997253,
      "loss_step_1": 0.0755,
      "loss_step_2": 0.0742,
      "loss_step_3": 0.0662,
      "step": 3520
    },
    {
      "entropy_step_1": 0.2274,
      "entropy_step_2": 0.2191,
      "entropy_step_3": 0.2176,
      "epoch": 0.10499390262038606,
      "grad_norm": 219.4819793701172,
      "learning_rate": 9.917712083567885e-05,
      "loss": 0.07182353734970093,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0693,
      "loss_step_3": 0.0665,
      "step": 3530
    },
    {
      "entropy_step_1": 0.2232,
      "entropy_step_2": 0.2132,
      "entropy_step_3": 0.2279,
      "epoch": 0.10529133577228518,
      "grad_norm": 37.821510314941406,
      "learning_rate": 9.916821112856299e-05,
      "loss": 0.06691217422485352,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0636,
      "step": 3540
    },
    {
      "entropy_step_1": 0.2522,
      "entropy_step_2": 0.2469,
      "entropy_step_3": 0.2486,
      "epoch": 0.10558876892418428,
      "grad_norm": 79.50679016113281,
      "learning_rate": 9.915925385053383e-05,
      "loss": 0.07219929695129394,
      "loss_step_1": 0.0746,
      "loss_step_2": 0.0749,
      "loss_step_3": 0.0754,
      "step": 3550
    },
    {
      "entropy_step_1": 0.2334,
      "entropy_step_2": 0.2338,
      "entropy_step_3": 0.2501,
      "epoch": 0.1058862020760834,
      "grad_norm": 66.44237518310547,
      "learning_rate": 9.915024901025766e-05,
      "loss": 0.07319210767745972,
      "loss_step_1": 0.0754,
      "loss_step_2": 0.075,
      "loss_step_3": 0.0813,
      "step": 3560
    },
    {
      "entropy_step_1": 0.2022,
      "entropy_step_2": 0.2039,
      "entropy_step_3": 0.2013,
      "epoch": 0.1061836352279825,
      "grad_norm": 33.503292083740234,
      "learning_rate": 9.914119661644679e-05,
      "loss": 0.06814413070678711,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.066,
      "loss_step_3": 0.0611,
      "step": 3570
    },
    {
      "entropy_step_1": 0.243,
      "entropy_step_2": 0.246,
      "entropy_step_3": 0.2323,
      "epoch": 0.10648106837988162,
      "grad_norm": 110.08049011230469,
      "learning_rate": 9.913209667785952e-05,
      "loss": 0.08108659386634827,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0715,
      "loss_step_3": 0.062,
      "step": 3580
    },
    {
      "entropy_step_1": 0.2176,
      "entropy_step_2": 0.2215,
      "entropy_step_3": 0.2175,
      "epoch": 0.10677850153178073,
      "grad_norm": 56.7987060546875,
      "learning_rate": 9.912294920330018e-05,
      "loss": 0.06568795442581177,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0654,
      "step": 3590
    },
    {
      "entropy_step_1": 0.2334,
      "entropy_step_2": 0.2341,
      "entropy_step_3": 0.2407,
      "epoch": 0.10707593468367985,
      "grad_norm": 58.14353942871094,
      "learning_rate": 9.911375420161906e-05,
      "loss": 0.06667491793632507,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0583,
      "step": 3600
    },
    {
      "entropy_step_1": 0.2174,
      "entropy_step_2": 0.2228,
      "entropy_step_3": 0.2349,
      "epoch": 0.10737336783557895,
      "grad_norm": 28.316028594970703,
      "learning_rate": 9.910451168171248e-05,
      "loss": 0.0731643259525299,
      "loss_step_1": 0.0756,
      "loss_step_2": 0.0746,
      "loss_step_3": 0.0711,
      "step": 3610
    },
    {
      "entropy_step_1": 0.2311,
      "entropy_step_2": 0.2356,
      "entropy_step_3": 0.222,
      "epoch": 0.10767080098747807,
      "grad_norm": 45.80824279785156,
      "learning_rate": 9.909522165252267e-05,
      "loss": 0.06745876669883728,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0687,
      "loss_step_3": 0.0613,
      "step": 3620
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.187,
      "entropy_step_3": 0.2157,
      "epoch": 0.10796823413937717,
      "grad_norm": 141.91517639160156,
      "learning_rate": 9.908588412303787e-05,
      "loss": 0.06014142632484436,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0634,
      "step": 3630
    },
    {
      "entropy_step_1": 0.2427,
      "entropy_step_2": 0.2453,
      "entropy_step_3": 0.2699,
      "epoch": 0.10826566729127629,
      "grad_norm": 33.068172454833984,
      "learning_rate": 9.907649910229229e-05,
      "loss": 0.07164363265037536,
      "loss_step_1": 0.0781,
      "loss_step_2": 0.0775,
      "loss_step_3": 0.0795,
      "step": 3640
    },
    {
      "entropy_step_1": 0.2188,
      "entropy_step_2": 0.2266,
      "entropy_step_3": 0.2238,
      "epoch": 0.1085631004431754,
      "grad_norm": 106.63243103027344,
      "learning_rate": 9.906706659936604e-05,
      "loss": 0.07051247954368592,
      "loss_step_1": 0.0764,
      "loss_step_2": 0.0748,
      "loss_step_3": 0.0725,
      "step": 3650
    },
    {
      "entropy_step_1": 0.2306,
      "entropy_step_2": 0.2423,
      "entropy_step_3": 0.2423,
      "epoch": 0.10886053359507451,
      "grad_norm": 32.413047790527344,
      "learning_rate": 9.90575866233852e-05,
      "loss": 0.08417662978172302,
      "loss_step_1": 0.0902,
      "loss_step_2": 0.0885,
      "loss_step_3": 0.0943,
      "step": 3660
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.2147,
      "entropy_step_3": 0.2043,
      "epoch": 0.10915796674697362,
      "grad_norm": 107.09159088134766,
      "learning_rate": 9.904805918352176e-05,
      "loss": 0.06852722764015198,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0529,
      "step": 3670
    },
    {
      "entropy_step_1": 0.2273,
      "entropy_step_2": 0.2375,
      "entropy_step_3": 0.227,
      "epoch": 0.10945539989887274,
      "grad_norm": 49.06271743774414,
      "learning_rate": 9.90384842889937e-05,
      "loss": 0.07227377891540528,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0611,
      "step": 3680
    },
    {
      "entropy_step_1": 0.2129,
      "entropy_step_2": 0.2145,
      "entropy_step_3": 0.2185,
      "epoch": 0.10975283305077184,
      "grad_norm": 500.5906677246094,
      "learning_rate": 9.902886194906479e-05,
      "loss": 0.07667388319969178,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0712,
      "loss_step_3": 0.0756,
      "step": 3690
    },
    {
      "entropy_step_1": 0.2592,
      "entropy_step_2": 0.255,
      "entropy_step_3": 0.2458,
      "epoch": 0.11005026620267094,
      "grad_norm": 72.07833862304688,
      "learning_rate": 9.901919217304483e-05,
      "loss": 0.07178915143013001,
      "loss_step_1": 0.0727,
      "loss_step_2": 0.071,
      "loss_step_3": 0.0645,
      "step": 3700
    },
    {
      "entropy_step_1": 0.2264,
      "entropy_step_2": 0.2256,
      "entropy_step_3": 0.2084,
      "epoch": 0.11034769935457006,
      "grad_norm": 36.60491180419922,
      "learning_rate": 9.900947497028943e-05,
      "loss": 0.07105568647384644,
      "loss_step_1": 0.0704,
      "loss_step_2": 0.0699,
      "loss_step_3": 0.0712,
      "step": 3710
    },
    {
      "entropy_step_1": 0.2091,
      "entropy_step_2": 0.2106,
      "entropy_step_3": 0.2033,
      "epoch": 0.11064513250646917,
      "grad_norm": 44.734588623046875,
      "learning_rate": 9.899971035020012e-05,
      "loss": 0.058530253171920774,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0566,
      "step": 3720
    },
    {
      "entropy_step_1": 0.2388,
      "entropy_step_2": 0.2431,
      "entropy_step_3": 0.2195,
      "epoch": 0.11094256565836828,
      "grad_norm": 80.32772064208984,
      "learning_rate": 9.898989832222432e-05,
      "loss": 0.07143727540969849,
      "loss_step_1": 0.0747,
      "loss_step_2": 0.0738,
      "loss_step_3": 0.0628,
      "step": 3730
    },
    {
      "entropy_step_1": 0.2203,
      "entropy_step_2": 0.2265,
      "entropy_step_3": 0.2483,
      "epoch": 0.11123999881026739,
      "grad_norm": 79.00653076171875,
      "learning_rate": 9.89800388958553e-05,
      "loss": 0.07231315970420837,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0722,
      "step": 3740
    },
    {
      "entropy_step_1": 0.1993,
      "entropy_step_2": 0.2022,
      "entropy_step_3": 0.1945,
      "epoch": 0.1115374319621665,
      "grad_norm": 88.86993408203125,
      "learning_rate": 9.897013208063215e-05,
      "loss": 0.0673844039440155,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0611,
      "step": 3750
    },
    {
      "entropy_step_1": 0.2357,
      "entropy_step_2": 0.2374,
      "entropy_step_3": 0.2354,
      "epoch": 0.11183486511406561,
      "grad_norm": 163.4485321044922,
      "learning_rate": 9.896017788613989e-05,
      "loss": 0.07180223464965821,
      "loss_step_1": 0.0745,
      "loss_step_2": 0.0743,
      "loss_step_3": 0.0726,
      "step": 3760
    },
    {
      "entropy_step_1": 0.21,
      "entropy_step_2": 0.2173,
      "entropy_step_3": 0.2343,
      "epoch": 0.11213229826596473,
      "grad_norm": 74.25777435302734,
      "learning_rate": 9.895017632200934e-05,
      "loss": 0.06327153444290161,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0662,
      "step": 3770
    },
    {
      "entropy_step_1": 0.2308,
      "entropy_step_2": 0.2375,
      "entropy_step_3": 0.24,
      "epoch": 0.11242973141786383,
      "grad_norm": 44.54108428955078,
      "learning_rate": 9.894012739791711e-05,
      "loss": 0.06294716596603393,
      "loss_step_1": 0.0715,
      "loss_step_2": 0.0714,
      "loss_step_3": 0.075,
      "step": 3780
    },
    {
      "entropy_step_1": 0.2644,
      "entropy_step_2": 0.2774,
      "entropy_step_3": 0.2744,
      "epoch": 0.11272716456976295,
      "grad_norm": 101.40164947509766,
      "learning_rate": 9.893003112358573e-05,
      "loss": 0.07872834801673889,
      "loss_step_1": 0.0848,
      "loss_step_2": 0.0851,
      "loss_step_3": 0.092,
      "step": 3790
    },
    {
      "entropy_step_1": 0.2287,
      "entropy_step_2": 0.2435,
      "entropy_step_3": 0.2529,
      "epoch": 0.11302459772166205,
      "grad_norm": 75.5790023803711,
      "learning_rate": 9.891988750878342e-05,
      "loss": 0.07934074997901916,
      "loss_step_1": 0.0849,
      "loss_step_2": 0.083,
      "loss_step_3": 0.078,
      "step": 3800
    },
    {
      "entropy_step_1": 0.2323,
      "entropy_step_2": 0.2492,
      "entropy_step_3": 0.2622,
      "epoch": 0.11332203087356117,
      "grad_norm": 121.96820068359375,
      "learning_rate": 9.89096965633243e-05,
      "loss": 0.06383268237113952,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0696,
      "step": 3810
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2281,
      "entropy_step_3": 0.2419,
      "epoch": 0.11361946402546028,
      "grad_norm": 52.91718673706055,
      "learning_rate": 9.889945829706824e-05,
      "loss": 0.06366865634918213,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0757,
      "step": 3820
    },
    {
      "entropy_step_1": 0.2196,
      "entropy_step_2": 0.231,
      "entropy_step_3": 0.2166,
      "epoch": 0.1139168971773594,
      "grad_norm": 65.52130889892578,
      "learning_rate": 9.888917271992091e-05,
      "loss": 0.07390747666358947,
      "loss_step_1": 0.0715,
      "loss_step_2": 0.0704,
      "loss_step_3": 0.0718,
      "step": 3830
    },
    {
      "entropy_step_1": 0.2156,
      "entropy_step_2": 0.2151,
      "entropy_step_3": 0.2201,
      "epoch": 0.1142143303292585,
      "grad_norm": 86.46481323242188,
      "learning_rate": 9.887883984183373e-05,
      "loss": 0.06662839651107788,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0652,
      "step": 3840
    },
    {
      "entropy_step_1": 0.2223,
      "entropy_step_2": 0.2244,
      "entropy_step_3": 0.2453,
      "epoch": 0.11451176348115762,
      "grad_norm": 39.85425567626953,
      "learning_rate": 9.886845967280392e-05,
      "loss": 0.07500792741775512,
      "loss_step_1": 0.0733,
      "loss_step_2": 0.0728,
      "loss_step_3": 0.0861,
      "step": 3850
    },
    {
      "entropy_step_1": 0.2172,
      "entropy_step_2": 0.2137,
      "entropy_step_3": 0.2295,
      "epoch": 0.11480919663305672,
      "grad_norm": 79.7365951538086,
      "learning_rate": 9.885803222287441e-05,
      "loss": 0.060401296615600585,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0551,
      "step": 3860
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.2083,
      "epoch": 0.11510662978495582,
      "grad_norm": 74.47300720214844,
      "learning_rate": 9.884755750213393e-05,
      "loss": 0.0710746169090271,
      "loss_step_1": 0.0683,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0728,
      "step": 3870
    },
    {
      "entropy_step_1": 0.2422,
      "entropy_step_2": 0.2315,
      "entropy_step_3": 0.2333,
      "epoch": 0.11540406293685494,
      "grad_norm": 73.12374877929688,
      "learning_rate": 9.88370355207169e-05,
      "loss": 0.06992482542991638,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0529,
      "step": 3880
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.1976,
      "entropy_step_3": 0.2188,
      "epoch": 0.11570149608875405,
      "grad_norm": 92.99838256835938,
      "learning_rate": 9.882646628880347e-05,
      "loss": 0.06128813624382019,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0632,
      "step": 3890
    },
    {
      "entropy_step_1": 0.2098,
      "entropy_step_2": 0.2008,
      "entropy_step_3": 0.187,
      "epoch": 0.11599892924065316,
      "grad_norm": 45.518062591552734,
      "learning_rate": 9.881584981661954e-05,
      "loss": 0.06482232213020325,
      "loss_step_1": 0.0704,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0623,
      "step": 3900
    },
    {
      "entropy_step_1": 0.2583,
      "entropy_step_2": 0.2529,
      "entropy_step_3": 0.2754,
      "epoch": 0.11629636239255227,
      "grad_norm": 48.72062301635742,
      "learning_rate": 9.880518611443666e-05,
      "loss": 0.06551112532615662,
      "loss_step_1": 0.0641,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0761,
      "step": 3910
    },
    {
      "entropy_step_1": 0.2186,
      "entropy_step_2": 0.2217,
      "entropy_step_3": 0.2224,
      "epoch": 0.11659379554445139,
      "grad_norm": 35.46641159057617,
      "learning_rate": 9.879447519257213e-05,
      "loss": 0.0689254105091095,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0707,
      "loss_step_3": 0.0697,
      "step": 3920
    },
    {
      "entropy_step_1": 0.2259,
      "entropy_step_2": 0.2269,
      "entropy_step_3": 0.2578,
      "epoch": 0.11689122869635049,
      "grad_norm": 91.2370834350586,
      "learning_rate": 9.878371706138891e-05,
      "loss": 0.06508617401123047,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0735,
      "step": 3930
    },
    {
      "entropy_step_1": 0.2213,
      "entropy_step_2": 0.2121,
      "entropy_step_3": 0.2085,
      "epoch": 0.11718866184824961,
      "grad_norm": 57.47941970825195,
      "learning_rate": 9.877291173129564e-05,
      "loss": 0.06318227052688599,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0593,
      "step": 3940
    },
    {
      "entropy_step_1": 0.2338,
      "entropy_step_2": 0.2286,
      "entropy_step_3": 0.2365,
      "epoch": 0.11748609500014871,
      "grad_norm": 130.83340454101562,
      "learning_rate": 9.87620592127466e-05,
      "loss": 0.06797334551811218,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0696,
      "loss_step_3": 0.0705,
      "step": 3950
    },
    {
      "entropy_step_1": 0.2466,
      "entropy_step_2": 0.2454,
      "entropy_step_3": 0.2337,
      "epoch": 0.11778352815204783,
      "grad_norm": 38.773284912109375,
      "learning_rate": 9.875115951624179e-05,
      "loss": 0.07013702988624573,
      "loss_step_1": 0.0791,
      "loss_step_2": 0.0792,
      "loss_step_3": 0.0669,
      "step": 3960
    },
    {
      "entropy_step_1": 0.2267,
      "entropy_step_2": 0.2325,
      "entropy_step_3": 0.2184,
      "epoch": 0.11808096130394694,
      "grad_norm": 109.46712493896484,
      "learning_rate": 9.874021265232678e-05,
      "loss": 0.06910912990570069,
      "loss_step_1": 0.0716,
      "loss_step_2": 0.0709,
      "loss_step_3": 0.0634,
      "step": 3970
    },
    {
      "entropy_step_1": 0.2344,
      "entropy_step_2": 0.2389,
      "entropy_step_3": 0.2456,
      "epoch": 0.11837839445584605,
      "grad_norm": 47.61233139038086,
      "learning_rate": 9.87292186315928e-05,
      "loss": 0.0656329333782196,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0626,
      "step": 3980
    },
    {
      "entropy_step_1": 0.2413,
      "entropy_step_2": 0.2385,
      "entropy_step_3": 0.1955,
      "epoch": 0.11867582760774516,
      "grad_norm": 52.513832092285156,
      "learning_rate": 9.871817746467677e-05,
      "loss": 0.06900444030761718,
      "loss_step_1": 0.0776,
      "loss_step_2": 0.0751,
      "loss_step_3": 0.0544,
      "step": 3990
    },
    {
      "entropy_step_1": 0.239,
      "entropy_step_2": 0.2298,
      "entropy_step_3": 0.2351,
      "epoch": 0.11897326075964428,
      "grad_norm": 112.15646362304688,
      "learning_rate": 9.870708916226113e-05,
      "loss": 0.07454299330711364,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.073,
      "loss_step_3": 0.0747,
      "step": 4000
    },
    {
      "epoch": 0.11897326075964428,
      "eval_accuracy": 0.797,
      "eval_f1": 0.7958516656191075,
      "eval_loss": 0.19304126501083374,
      "eval_precision": 0.8038363171355498,
      "eval_recall": 0.797,
      "eval_runtime": 6.6924,
      "eval_samples_per_second": 74.711,
      "eval_steps_per_second": 9.414,
      "step": 4000
    },
    {
      "entropy_step_1": 0.2064,
      "entropy_step_2": 0.2098,
      "entropy_step_3": 0.2108,
      "epoch": 0.11927069391154338,
      "grad_norm": 103.89142608642578,
      "learning_rate": 9.869595373507397e-05,
      "loss": 0.06330393552780152,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0582,
      "step": 4010
    },
    {
      "entropy_step_1": 0.2221,
      "entropy_step_2": 0.2269,
      "entropy_step_3": 0.2214,
      "epoch": 0.1195681270634425,
      "grad_norm": 235.4517822265625,
      "learning_rate": 9.868477119388896e-05,
      "loss": 0.06764224767684937,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0595,
      "step": 4020
    },
    {
      "entropy_step_1": 0.2257,
      "entropy_step_2": 0.232,
      "entropy_step_3": 0.2339,
      "epoch": 0.1198655602153416,
      "grad_norm": 135.73146057128906,
      "learning_rate": 9.867354154952537e-05,
      "loss": 0.06816955804824829,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0706,
      "step": 4030
    },
    {
      "entropy_step_1": 0.2407,
      "entropy_step_2": 0.2471,
      "entropy_step_3": 0.2351,
      "epoch": 0.1201629933672407,
      "grad_norm": 62.5189323425293,
      "learning_rate": 9.866226481284803e-05,
      "loss": 0.08343331813812256,
      "loss_step_1": 0.0798,
      "loss_step_2": 0.079,
      "loss_step_3": 0.0756,
      "step": 4040
    },
    {
      "entropy_step_1": 0.2078,
      "entropy_step_2": 0.2133,
      "entropy_step_3": 0.2242,
      "epoch": 0.12046042651913982,
      "grad_norm": 116.95427703857422,
      "learning_rate": 9.865094099476734e-05,
      "loss": 0.0719325840473175,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0675,
      "step": 4050
    },
    {
      "entropy_step_1": 0.2413,
      "entropy_step_2": 0.2479,
      "entropy_step_3": 0.2271,
      "epoch": 0.12075785967103893,
      "grad_norm": 26.419902801513672,
      "learning_rate": 9.863957010623924e-05,
      "loss": 0.06883882880210876,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0696,
      "loss_step_3": 0.0586,
      "step": 4060
    },
    {
      "entropy_step_1": 0.2736,
      "entropy_step_2": 0.2714,
      "entropy_step_3": 0.2823,
      "epoch": 0.12105529282293805,
      "grad_norm": 51.16787338256836,
      "learning_rate": 9.862815215826523e-05,
      "loss": 0.0799972414970398,
      "loss_step_1": 0.0847,
      "loss_step_2": 0.0831,
      "loss_step_3": 0.0888,
      "step": 4070
    },
    {
      "entropy_step_1": 0.2252,
      "entropy_step_2": 0.2119,
      "entropy_step_3": 0.1919,
      "epoch": 0.12135272597483715,
      "grad_norm": 99.3899154663086,
      "learning_rate": 9.861668716189232e-05,
      "loss": 0.0693248450756073,
      "loss_step_1": 0.0708,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0622,
      "step": 4080
    },
    {
      "entropy_step_1": 0.2214,
      "entropy_step_2": 0.2116,
      "entropy_step_3": 0.2141,
      "epoch": 0.12165015912673627,
      "grad_norm": 42.973655700683594,
      "learning_rate": 9.860517512821306e-05,
      "loss": 0.05616840124130249,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0574,
      "step": 4090
    },
    {
      "entropy_step_1": 0.2222,
      "entropy_step_2": 0.2231,
      "entropy_step_3": 0.2394,
      "epoch": 0.12194759227863537,
      "grad_norm": 64.55619812011719,
      "learning_rate": 9.859361606836551e-05,
      "loss": 0.06693246364593505,
      "loss_step_1": 0.0726,
      "loss_step_2": 0.0724,
      "loss_step_3": 0.0803,
      "step": 4100
    },
    {
      "entropy_step_1": 0.2336,
      "entropy_step_2": 0.2419,
      "entropy_step_3": 0.2456,
      "epoch": 0.12224502543053449,
      "grad_norm": 73.97051239013672,
      "learning_rate": 9.858200999353318e-05,
      "loss": 0.0732642114162445,
      "loss_step_1": 0.0767,
      "loss_step_2": 0.0753,
      "loss_step_3": 0.0784,
      "step": 4110
    },
    {
      "entropy_step_1": 0.2193,
      "entropy_step_2": 0.2213,
      "entropy_step_3": 0.223,
      "epoch": 0.1225424585824336,
      "grad_norm": 74.816162109375,
      "learning_rate": 9.857035691494512e-05,
      "loss": 0.0654531478881836,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.0688,
      "step": 4120
    },
    {
      "entropy_step_1": 0.2228,
      "entropy_step_2": 0.222,
      "entropy_step_3": 0.215,
      "epoch": 0.12283989173433271,
      "grad_norm": 61.85882568359375,
      "learning_rate": 9.855865684387587e-05,
      "loss": 0.07305105328559876,
      "loss_step_1": 0.0731,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.0724,
      "step": 4130
    },
    {
      "entropy_step_1": 0.1863,
      "entropy_step_2": 0.1869,
      "entropy_step_3": 0.1776,
      "epoch": 0.12313732488623182,
      "grad_norm": 65.90156555175781,
      "learning_rate": 9.854690979164538e-05,
      "loss": 0.05881397128105163,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0492,
      "step": 4140
    },
    {
      "entropy_step_1": 0.2304,
      "entropy_step_2": 0.2322,
      "entropy_step_3": 0.2253,
      "epoch": 0.12343475803813093,
      "grad_norm": 50.78444290161133,
      "learning_rate": 9.853511576961909e-05,
      "loss": 0.07094047665596008,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0612,
      "step": 4150
    },
    {
      "entropy_step_1": 0.2265,
      "entropy_step_2": 0.2298,
      "entropy_step_3": 0.2357,
      "epoch": 0.12373219119003004,
      "grad_norm": 76.24791717529297,
      "learning_rate": 9.852327478920787e-05,
      "loss": 0.06819141507148743,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0712,
      "loss_step_3": 0.081,
      "step": 4160
    },
    {
      "entropy_step_1": 0.222,
      "entropy_step_2": 0.2283,
      "entropy_step_3": 0.2217,
      "epoch": 0.12402962434192916,
      "grad_norm": 62.76770782470703,
      "learning_rate": 9.851138686186805e-05,
      "loss": 0.06418318152427674,
      "loss_step_1": 0.0666,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.0782,
      "step": 4170
    },
    {
      "entropy_step_1": 0.2263,
      "entropy_step_2": 0.2348,
      "entropy_step_3": 0.2369,
      "epoch": 0.12432705749382826,
      "grad_norm": 36.04010772705078,
      "learning_rate": 9.849945199910134e-05,
      "loss": 0.07472273111343383,
      "loss_step_1": 0.0724,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0756,
      "step": 4180
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2082,
      "entropy_step_3": 0.2054,
      "epoch": 0.12462449064572738,
      "grad_norm": 89.32247161865234,
      "learning_rate": 9.848747021245489e-05,
      "loss": 0.058599847555160525,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0529,
      "step": 4190
    },
    {
      "entropy_step_1": 0.2111,
      "entropy_step_2": 0.2066,
      "entropy_step_3": 0.1979,
      "epoch": 0.12492192379762648,
      "grad_norm": 88.71621704101562,
      "learning_rate": 9.847544151352124e-05,
      "loss": 0.06577463150024414,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.065,
      "loss_step_3": 0.0606,
      "step": 4200
    },
    {
      "entropy_step_1": 0.2551,
      "entropy_step_2": 0.2477,
      "entropy_step_3": 0.253,
      "epoch": 0.1252193569495256,
      "grad_norm": 60.11402893066406,
      "learning_rate": 9.846336591393833e-05,
      "loss": 0.0696653664112091,
      "loss_step_1": 0.0716,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0735,
      "step": 4210
    },
    {
      "entropy_step_1": 0.245,
      "entropy_step_2": 0.2506,
      "entropy_step_3": 0.2531,
      "epoch": 0.1255167901014247,
      "grad_norm": 94.83946228027344,
      "learning_rate": 9.845124342538945e-05,
      "loss": 0.07007895708084107,
      "loss_step_1": 0.0729,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.0703,
      "step": 4220
    },
    {
      "entropy_step_1": 0.2318,
      "entropy_step_2": 0.2392,
      "entropy_step_3": 0.2392,
      "epoch": 0.1258142232533238,
      "grad_norm": 68.43324279785156,
      "learning_rate": 9.843907405960328e-05,
      "loss": 0.06751342415809632,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0712,
      "step": 4230
    },
    {
      "entropy_step_1": 0.2542,
      "entropy_step_2": 0.2529,
      "entropy_step_3": 0.244,
      "epoch": 0.12611165640522293,
      "grad_norm": 26.686241149902344,
      "learning_rate": 9.842685782835387e-05,
      "loss": 0.07681447267532349,
      "loss_step_1": 0.0766,
      "loss_step_2": 0.0761,
      "loss_step_3": 0.0749,
      "step": 4240
    },
    {
      "entropy_step_1": 0.2506,
      "entropy_step_2": 0.2555,
      "entropy_step_3": 0.2531,
      "epoch": 0.12640908955712205,
      "grad_norm": 139.24598693847656,
      "learning_rate": 9.841459474346056e-05,
      "loss": 0.08207847476005554,
      "loss_step_1": 0.0855,
      "loss_step_2": 0.0845,
      "loss_step_3": 0.0822,
      "step": 4250
    },
    {
      "entropy_step_1": 0.2624,
      "entropy_step_2": 0.2658,
      "entropy_step_3": 0.2665,
      "epoch": 0.12670652270902114,
      "grad_norm": 62.7042236328125,
      "learning_rate": 9.840228481678803e-05,
      "loss": 0.06694030165672302,
      "loss_step_1": 0.0674,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.075,
      "step": 4260
    },
    {
      "entropy_step_1": 0.1813,
      "entropy_step_2": 0.19,
      "entropy_step_3": 0.1881,
      "epoch": 0.12700395586092025,
      "grad_norm": 87.36475372314453,
      "learning_rate": 9.838992806024636e-05,
      "loss": 0.06987811923027039,
      "loss_step_1": 0.0692,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.0658,
      "step": 4270
    },
    {
      "entropy_step_1": 0.2276,
      "entropy_step_2": 0.2381,
      "entropy_step_3": 0.2477,
      "epoch": 0.12730138901281937,
      "grad_norm": 46.27036666870117,
      "learning_rate": 9.837752448579085e-05,
      "loss": 0.06287777423858643,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0639,
      "step": 4280
    },
    {
      "entropy_step_1": 0.2159,
      "entropy_step_2": 0.2214,
      "entropy_step_3": 0.2104,
      "epoch": 0.1275988221647185,
      "grad_norm": 30.873044967651367,
      "learning_rate": 9.836507410542213e-05,
      "loss": 0.06193908452987671,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0687,
      "loss_step_3": 0.0686,
      "step": 4290
    },
    {
      "entropy_step_1": 0.2196,
      "entropy_step_2": 0.2354,
      "entropy_step_3": 0.2382,
      "epoch": 0.12789625531661758,
      "grad_norm": 75.31753540039062,
      "learning_rate": 9.835257693118609e-05,
      "loss": 0.0680661678314209,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0604,
      "step": 4300
    },
    {
      "entropy_step_1": 0.2649,
      "entropy_step_2": 0.2707,
      "entropy_step_3": 0.2737,
      "epoch": 0.1281936884685167,
      "grad_norm": 42.54191970825195,
      "learning_rate": 9.834003297517395e-05,
      "loss": 0.07847943305969238,
      "loss_step_1": 0.0772,
      "loss_step_2": 0.0769,
      "loss_step_3": 0.0869,
      "step": 4310
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.1994,
      "epoch": 0.12849112162041582,
      "grad_norm": 80.55716705322266,
      "learning_rate": 9.832744224952216e-05,
      "loss": 0.06282431483268738,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.061,
      "step": 4320
    },
    {
      "entropy_step_1": 0.2299,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.2206,
      "epoch": 0.12878855477231493,
      "grad_norm": 73.0713882446289,
      "learning_rate": 9.831480476641239e-05,
      "loss": 0.06798039674758911,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0663,
      "step": 4330
    },
    {
      "entropy_step_1": 0.24,
      "entropy_step_2": 0.2255,
      "entropy_step_3": 0.2389,
      "epoch": 0.12908598792421402,
      "grad_norm": 46.855220794677734,
      "learning_rate": 9.830212053807158e-05,
      "loss": 0.07009733319282532,
      "loss_step_1": 0.0703,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0772,
      "step": 4340
    },
    {
      "entropy_step_1": 0.2242,
      "entropy_step_2": 0.2184,
      "entropy_step_3": 0.2152,
      "epoch": 0.12938342107611314,
      "grad_norm": 51.01715850830078,
      "learning_rate": 9.828938957677192e-05,
      "loss": 0.06720777750015258,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0734,
      "step": 4350
    },
    {
      "entropy_step_1": 0.2392,
      "entropy_step_2": 0.2366,
      "entropy_step_3": 0.2512,
      "epoch": 0.12968085422801226,
      "grad_norm": 207.3904266357422,
      "learning_rate": 9.827661189483076e-05,
      "loss": 0.07133277654647827,
      "loss_step_1": 0.0705,
      "loss_step_2": 0.0695,
      "loss_step_3": 0.0658,
      "step": 4360
    },
    {
      "entropy_step_1": 0.1978,
      "entropy_step_2": 0.1959,
      "entropy_step_3": 0.1698,
      "epoch": 0.12997828737991138,
      "grad_norm": 95.59587097167969,
      "learning_rate": 9.82637875046107e-05,
      "loss": 0.06217855215072632,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.057,
      "step": 4370
    },
    {
      "entropy_step_1": 0.2383,
      "entropy_step_2": 0.2371,
      "entropy_step_3": 0.2274,
      "epoch": 0.13027572053181047,
      "grad_norm": 64.4358139038086,
      "learning_rate": 9.825091641851951e-05,
      "loss": 0.0616992712020874,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.0562,
      "step": 4380
    },
    {
      "entropy_step_1": 0.2511,
      "entropy_step_2": 0.2465,
      "entropy_step_3": 0.2465,
      "epoch": 0.1305731536837096,
      "grad_norm": 81.79906463623047,
      "learning_rate": 9.823799864901013e-05,
      "loss": 0.07177466154098511,
      "loss_step_1": 0.0803,
      "loss_step_2": 0.0809,
      "loss_step_3": 0.0914,
      "step": 4390
    },
    {
      "entropy_step_1": 0.2156,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.1931,
      "epoch": 0.1308705868356087,
      "grad_norm": 146.6709442138672,
      "learning_rate": 9.822503420858069e-05,
      "loss": 0.06001982092857361,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0561,
      "step": 4400
    },
    {
      "entropy_step_1": 0.187,
      "entropy_step_2": 0.1859,
      "entropy_step_3": 0.1758,
      "epoch": 0.1311680199875078,
      "grad_norm": 50.082847595214844,
      "learning_rate": 9.821202310977444e-05,
      "loss": 0.06350985765457154,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0511,
      "step": 4410
    },
    {
      "entropy_step_1": 0.1857,
      "entropy_step_2": 0.1846,
      "entropy_step_3": 0.175,
      "epoch": 0.1314654531394069,
      "grad_norm": 112.16313171386719,
      "learning_rate": 9.819896536517983e-05,
      "loss": 0.061355584859848024,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0604,
      "step": 4420
    },
    {
      "entropy_step_1": 0.2141,
      "entropy_step_2": 0.2169,
      "entropy_step_3": 0.2085,
      "epoch": 0.13176288629130603,
      "grad_norm": 47.31580352783203,
      "learning_rate": 9.818586098743038e-05,
      "loss": 0.06689470410346984,
      "loss_step_1": 0.0674,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0547,
      "step": 4430
    },
    {
      "entropy_step_1": 0.23,
      "entropy_step_2": 0.2317,
      "entropy_step_3": 0.2403,
      "epoch": 0.13206031944320515,
      "grad_norm": 72.6303482055664,
      "learning_rate": 9.817270998920477e-05,
      "loss": 0.06828827857971191,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0651,
      "step": 4440
    },
    {
      "entropy_step_1": 0.2287,
      "entropy_step_2": 0.2273,
      "entropy_step_3": 0.2318,
      "epoch": 0.13235775259510424,
      "grad_norm": 83.15607452392578,
      "learning_rate": 9.815951238322674e-05,
      "loss": 0.07202479243278503,
      "loss_step_1": 0.0693,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0672,
      "step": 4450
    },
    {
      "entropy_step_1": 0.2258,
      "entropy_step_2": 0.2257,
      "entropy_step_3": 0.244,
      "epoch": 0.13265518574700336,
      "grad_norm": 49.6726188659668,
      "learning_rate": 9.814626818226519e-05,
      "loss": 0.06340648531913758,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0631,
      "step": 4460
    },
    {
      "entropy_step_1": 0.251,
      "entropy_step_2": 0.2489,
      "entropy_step_3": 0.2504,
      "epoch": 0.13295261889890247,
      "grad_norm": 70.0263671875,
      "learning_rate": 9.813297739913406e-05,
      "loss": 0.07430911064147949,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0611,
      "step": 4470
    },
    {
      "entropy_step_1": 0.2374,
      "entropy_step_2": 0.2313,
      "entropy_step_3": 0.2183,
      "epoch": 0.1332500520508016,
      "grad_norm": 57.5525016784668,
      "learning_rate": 9.811964004669232e-05,
      "loss": 0.06812782287597656,
      "loss_step_1": 0.0703,
      "loss_step_2": 0.0698,
      "loss_step_3": 0.0535,
      "step": 4480
    },
    {
      "entropy_step_1": 0.2375,
      "entropy_step_2": 0.2304,
      "entropy_step_3": 0.1966,
      "epoch": 0.13354748520270068,
      "grad_norm": 122.96155548095703,
      "learning_rate": 9.81062561378441e-05,
      "loss": 0.06583046317100524,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0531,
      "step": 4490
    },
    {
      "entropy_step_1": 0.2516,
      "entropy_step_2": 0.2457,
      "entropy_step_3": 0.2379,
      "epoch": 0.1338449183545998,
      "grad_norm": 40.29863357543945,
      "learning_rate": 9.809282568553846e-05,
      "loss": 0.07283094525337219,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0634,
      "step": 4500
    },
    {
      "entropy_step_1": 0.2374,
      "entropy_step_2": 0.2282,
      "entropy_step_3": 0.2183,
      "epoch": 0.13414235150649892,
      "grad_norm": 53.9449348449707,
      "learning_rate": 9.80793487027696e-05,
      "loss": 0.0662133276462555,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0633,
      "step": 4510
    },
    {
      "entropy_step_1": 0.1776,
      "entropy_step_2": 0.1746,
      "entropy_step_3": 0.179,
      "epoch": 0.13443978465839804,
      "grad_norm": 123.32917022705078,
      "learning_rate": 9.806582520257664e-05,
      "loss": 0.06226044297218323,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.075,
      "step": 4520
    },
    {
      "entropy_step_1": 0.2454,
      "entropy_step_2": 0.2366,
      "entropy_step_3": 0.2597,
      "epoch": 0.13473721781029713,
      "grad_norm": 23.7492733001709,
      "learning_rate": 9.805225519804377e-05,
      "loss": 0.0664846956729889,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0621,
      "step": 4530
    },
    {
      "entropy_step_1": 0.2149,
      "entropy_step_2": 0.212,
      "entropy_step_3": 0.2133,
      "epoch": 0.13503465096219625,
      "grad_norm": 72.93165588378906,
      "learning_rate": 9.803863870230014e-05,
      "loss": 0.06368963122367859,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0636,
      "step": 4540
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.1941,
      "entropy_step_3": 0.1868,
      "epoch": 0.13533208411409536,
      "grad_norm": 24.38273811340332,
      "learning_rate": 9.802497572851994e-05,
      "loss": 0.0665030837059021,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0537,
      "step": 4550
    },
    {
      "entropy_step_1": 0.1905,
      "entropy_step_2": 0.1814,
      "entropy_step_3": 0.1785,
      "epoch": 0.13562951726599448,
      "grad_norm": 62.87890625,
      "learning_rate": 9.801126628992221e-05,
      "loss": 0.06229486465454102,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0703,
      "step": 4560
    },
    {
      "entropy_step_1": 0.2046,
      "entropy_step_2": 0.1966,
      "entropy_step_3": 0.2047,
      "epoch": 0.13592695041789357,
      "grad_norm": 40.95473098754883,
      "learning_rate": 9.799751039977108e-05,
      "loss": 0.06656493544578553,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0597,
      "step": 4570
    },
    {
      "entropy_step_1": 0.2137,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.2027,
      "epoch": 0.1362243835697927,
      "grad_norm": 93.2742919921875,
      "learning_rate": 9.798370807137551e-05,
      "loss": 0.06617287397384644,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0679,
      "loss_step_3": 0.0717,
      "step": 4580
    },
    {
      "entropy_step_1": 0.2099,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.2171,
      "epoch": 0.1365218167216918,
      "grad_norm": 66.53085327148438,
      "learning_rate": 9.79698593180895e-05,
      "loss": 0.06404640078544617,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0623,
      "step": 4590
    },
    {
      "entropy_step_1": 0.2109,
      "entropy_step_2": 0.2154,
      "entropy_step_3": 0.1913,
      "epoch": 0.1368192498735909,
      "grad_norm": 47.722251892089844,
      "learning_rate": 9.795596415331186e-05,
      "loss": 0.07197282314300538,
      "loss_step_1": 0.07,
      "loss_step_2": 0.07,
      "loss_step_3": 0.0647,
      "step": 4600
    },
    {
      "entropy_step_1": 0.2231,
      "entropy_step_2": 0.2292,
      "entropy_step_3": 0.2304,
      "epoch": 0.13711668302549002,
      "grad_norm": 241.02317810058594,
      "learning_rate": 9.794202259048637e-05,
      "loss": 0.06701764464378357,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0734,
      "loss_step_3": 0.0825,
      "step": 4610
    },
    {
      "entropy_step_1": 0.2301,
      "entropy_step_2": 0.2348,
      "entropy_step_3": 0.2406,
      "epoch": 0.13741411617738913,
      "grad_norm": 47.1419792175293,
      "learning_rate": 9.792803464310168e-05,
      "loss": 0.07523521184921264,
      "loss_step_1": 0.0731,
      "loss_step_2": 0.0721,
      "loss_step_3": 0.0727,
      "step": 4620
    },
    {
      "entropy_step_1": 0.2454,
      "entropy_step_2": 0.2497,
      "entropy_step_3": 0.2379,
      "epoch": 0.13771154932928825,
      "grad_norm": 34.6539192199707,
      "learning_rate": 9.791400032469131e-05,
      "loss": 0.072628653049469,
      "loss_step_1": 0.0744,
      "loss_step_2": 0.0738,
      "loss_step_3": 0.065,
      "step": 4630
    },
    {
      "entropy_step_1": 0.2086,
      "entropy_step_2": 0.2166,
      "entropy_step_3": 0.23,
      "epoch": 0.13800898248118734,
      "grad_norm": 176.85841369628906,
      "learning_rate": 9.789991964883365e-05,
      "loss": 0.06619141101837159,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.073,
      "step": 4640
    },
    {
      "entropy_step_1": 0.2341,
      "entropy_step_2": 0.2457,
      "entropy_step_3": 0.2469,
      "epoch": 0.13830641563308646,
      "grad_norm": 73.32119750976562,
      "learning_rate": 9.788579262915199e-05,
      "loss": 0.06620604991912842,
      "loss_step_1": 0.0723,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0755,
      "step": 4650
    },
    {
      "entropy_step_1": 0.2199,
      "entropy_step_2": 0.2274,
      "entropy_step_3": 0.2349,
      "epoch": 0.13860384878498558,
      "grad_norm": 244.2790069580078,
      "learning_rate": 9.787161927931435e-05,
      "loss": 0.06308255791664123,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.064,
      "step": 4660
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1973,
      "entropy_step_3": 0.204,
      "epoch": 0.1389012819368847,
      "grad_norm": 46.16183090209961,
      "learning_rate": 9.785739961303369e-05,
      "loss": 0.06191205382347107,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0591,
      "step": 4670
    },
    {
      "entropy_step_1": 0.1978,
      "entropy_step_2": 0.2054,
      "entropy_step_3": 0.213,
      "epoch": 0.1391987150887838,
      "grad_norm": 79.2952880859375,
      "learning_rate": 9.78431336440677e-05,
      "loss": 0.0673409104347229,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0709,
      "loss_step_3": 0.0799,
      "step": 4680
    },
    {
      "entropy_step_1": 0.2751,
      "entropy_step_2": 0.2791,
      "entropy_step_3": 0.3079,
      "epoch": 0.1394961482406829,
      "grad_norm": 39.12159729003906,
      "learning_rate": 9.782882138621891e-05,
      "loss": 0.0789314091205597,
      "loss_step_1": 0.0821,
      "loss_step_2": 0.082,
      "loss_step_3": 0.1058,
      "step": 4690
    },
    {
      "entropy_step_1": 0.1829,
      "entropy_step_2": 0.1882,
      "entropy_step_3": 0.1819,
      "epoch": 0.13979358139258202,
      "grad_norm": 48.13775634765625,
      "learning_rate": 9.781446285333462e-05,
      "loss": 0.07446236610412597,
      "loss_step_1": 0.0778,
      "loss_step_2": 0.0756,
      "loss_step_3": 0.0632,
      "step": 4700
    },
    {
      "entropy_step_1": 0.2381,
      "entropy_step_2": 0.2377,
      "entropy_step_3": 0.2178,
      "epoch": 0.14009101454448114,
      "grad_norm": 45.4048957824707,
      "learning_rate": 9.780005805930692e-05,
      "loss": 0.06748455166816711,
      "loss_step_1": 0.0708,
      "loss_step_2": 0.0704,
      "loss_step_3": 0.0668,
      "step": 4710
    },
    {
      "entropy_step_1": 0.1966,
      "entropy_step_2": 0.1973,
      "entropy_step_3": 0.17,
      "epoch": 0.14038844769638023,
      "grad_norm": 126.15430450439453,
      "learning_rate": 9.778560701807262e-05,
      "loss": 0.06515680551528931,
      "loss_step_1": 0.0688,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0676,
      "step": 4720
    },
    {
      "entropy_step_1": 0.2167,
      "entropy_step_2": 0.2252,
      "entropy_step_3": 0.2163,
      "epoch": 0.14068588084827935,
      "grad_norm": 28.225322723388672,
      "learning_rate": 9.777110974361332e-05,
      "loss": 0.06446076035499573,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.073,
      "step": 4730
    },
    {
      "entropy_step_1": 0.2349,
      "entropy_step_2": 0.2396,
      "entropy_step_3": 0.2453,
      "epoch": 0.14098331400017847,
      "grad_norm": 58.51274108886719,
      "learning_rate": 9.775656624995532e-05,
      "loss": 0.07343543767929077,
      "loss_step_1": 0.0742,
      "loss_step_2": 0.0736,
      "loss_step_3": 0.0821,
      "step": 4740
    },
    {
      "entropy_step_1": 0.2409,
      "entropy_step_2": 0.2371,
      "entropy_step_3": 0.2243,
      "epoch": 0.14128074715207756,
      "grad_norm": 86.27022552490234,
      "learning_rate": 9.774197655116965e-05,
      "loss": 0.07190963625907898,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0652,
      "loss_step_3": 0.0521,
      "step": 4750
    },
    {
      "entropy_step_1": 0.2185,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.2125,
      "epoch": 0.14157818030397668,
      "grad_norm": 45.927982330322266,
      "learning_rate": 9.772734066137206e-05,
      "loss": 0.07318059206008912,
      "loss_step_1": 0.0751,
      "loss_step_2": 0.075,
      "loss_step_3": 0.0785,
      "step": 4760
    },
    {
      "entropy_step_1": 0.2195,
      "entropy_step_2": 0.2143,
      "entropy_step_3": 0.2213,
      "epoch": 0.1418756134558758,
      "grad_norm": 49.24163818359375,
      "learning_rate": 9.771265859472294e-05,
      "loss": 0.07425122261047364,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0711,
      "step": 4770
    },
    {
      "entropy_step_1": 0.2321,
      "entropy_step_2": 0.2334,
      "entropy_step_3": 0.219,
      "epoch": 0.1421730466077749,
      "grad_norm": 109.13839721679688,
      "learning_rate": 9.769793036542741e-05,
      "loss": 0.0711780309677124,
      "loss_step_1": 0.0665,
      "loss_step_2": 0.0658,
      "loss_step_3": 0.0575,
      "step": 4780
    },
    {
      "entropy_step_1": 0.2151,
      "entropy_step_2": 0.2231,
      "entropy_step_3": 0.1912,
      "epoch": 0.142470479759674,
      "grad_norm": 55.31061553955078,
      "learning_rate": 9.768315598773523e-05,
      "loss": 0.061631041765213015,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0469,
      "step": 4790
    },
    {
      "entropy_step_1": 0.2258,
      "entropy_step_2": 0.2424,
      "entropy_step_3": 0.2393,
      "epoch": 0.14276791291157312,
      "grad_norm": 55.70045852661133,
      "learning_rate": 9.766833547594081e-05,
      "loss": 0.0707964539527893,
      "loss_step_1": 0.0787,
      "loss_step_2": 0.0785,
      "loss_step_3": 0.0787,
      "step": 4800
    },
    {
      "entropy_step_1": 0.2185,
      "entropy_step_2": 0.2309,
      "entropy_step_3": 0.2261,
      "epoch": 0.14306534606347224,
      "grad_norm": 59.741329193115234,
      "learning_rate": 9.765346884438318e-05,
      "loss": 0.06316241025924682,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0548,
      "step": 4810
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2077,
      "entropy_step_3": 0.2048,
      "epoch": 0.14336277921537136,
      "grad_norm": 35.24050521850586,
      "learning_rate": 9.763855610744604e-05,
      "loss": 0.06395804286003112,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0698,
      "step": 4820
    },
    {
      "entropy_step_1": 0.2093,
      "entropy_step_2": 0.2029,
      "entropy_step_3": 0.1866,
      "epoch": 0.14366021236727045,
      "grad_norm": 59.27975845336914,
      "learning_rate": 9.762359727955764e-05,
      "loss": 0.06380795240402222,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0576,
      "step": 4830
    },
    {
      "entropy_step_1": 0.2277,
      "entropy_step_2": 0.2269,
      "entropy_step_3": 0.2069,
      "epoch": 0.14395764551916956,
      "grad_norm": 61.623477935791016,
      "learning_rate": 9.760859237519087e-05,
      "loss": 0.0690631091594696,
      "loss_step_1": 0.072,
      "loss_step_2": 0.071,
      "loss_step_3": 0.0686,
      "step": 4840
    },
    {
      "entropy_step_1": 0.1944,
      "entropy_step_2": 0.1991,
      "entropy_step_3": 0.2076,
      "epoch": 0.14425507867106868,
      "grad_norm": 32.632076263427734,
      "learning_rate": 9.759354140886315e-05,
      "loss": 0.058362793922424314,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0629,
      "step": 4850
    },
    {
      "entropy_step_1": 0.2328,
      "entropy_step_2": 0.2447,
      "entropy_step_3": 0.2196,
      "epoch": 0.1445525118229678,
      "grad_norm": 23.125045776367188,
      "learning_rate": 9.757844439513653e-05,
      "loss": 0.06655386090278625,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0578,
      "step": 4860
    },
    {
      "entropy_step_1": 0.1781,
      "entropy_step_2": 0.1845,
      "entropy_step_3": 0.171,
      "epoch": 0.1448499449748669,
      "grad_norm": 54.943153381347656,
      "learning_rate": 9.756330134861754e-05,
      "loss": 0.0600730299949646,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0482,
      "step": 4870
    },
    {
      "entropy_step_1": 0.1963,
      "entropy_step_2": 0.1938,
      "entropy_step_3": 0.1854,
      "epoch": 0.145147378126766,
      "grad_norm": 168.3514862060547,
      "learning_rate": 9.754811228395733e-05,
      "loss": 0.05887918472290039,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0607,
      "step": 4880
    },
    {
      "entropy_step_1": 0.227,
      "entropy_step_2": 0.2254,
      "entropy_step_3": 0.2432,
      "epoch": 0.14544481127866513,
      "grad_norm": 176.82965087890625,
      "learning_rate": 9.753287721585147e-05,
      "loss": 0.06678463220596313,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0675,
      "step": 4890
    },
    {
      "entropy_step_1": 0.2121,
      "entropy_step_2": 0.2063,
      "entropy_step_3": 0.219,
      "epoch": 0.14574224443056424,
      "grad_norm": 90.75521087646484,
      "learning_rate": 9.751759615904015e-05,
      "loss": 0.0652107298374176,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0753,
      "step": 4900
    },
    {
      "entropy_step_1": 0.2137,
      "entropy_step_2": 0.2136,
      "entropy_step_3": 0.1991,
      "epoch": 0.14603967758246333,
      "grad_norm": 53.07145690917969,
      "learning_rate": 9.750226912830795e-05,
      "loss": 0.05920582413673401,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0582,
      "step": 4910
    },
    {
      "entropy_step_1": 0.2255,
      "entropy_step_2": 0.2287,
      "entropy_step_3": 0.2115,
      "epoch": 0.14633711073436245,
      "grad_norm": 173.41539001464844,
      "learning_rate": 9.748689613848402e-05,
      "loss": 0.07371435165405274,
      "loss_step_1": 0.0785,
      "loss_step_2": 0.0758,
      "loss_step_3": 0.0617,
      "step": 4920
    },
    {
      "entropy_step_1": 0.2375,
      "entropy_step_2": 0.2422,
      "entropy_step_3": 0.2137,
      "epoch": 0.14663454388626157,
      "grad_norm": 49.070411682128906,
      "learning_rate": 9.747147720444194e-05,
      "loss": 0.06798727512359619,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0522,
      "step": 4930
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2066,
      "entropy_step_3": 0.2168,
      "epoch": 0.14693197703816066,
      "grad_norm": 85.59335327148438,
      "learning_rate": 9.74560123410997e-05,
      "loss": 0.06840584874153137,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0666,
      "step": 4940
    },
    {
      "entropy_step_1": 0.2278,
      "entropy_step_2": 0.2286,
      "entropy_step_3": 0.2285,
      "epoch": 0.14722941019005978,
      "grad_norm": 39.872520446777344,
      "learning_rate": 9.74405015634198e-05,
      "loss": 0.06912697553634643,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.0537,
      "step": 4950
    },
    {
      "entropy_step_1": 0.1996,
      "entropy_step_2": 0.1999,
      "entropy_step_3": 0.1941,
      "epoch": 0.1475268433419589,
      "grad_norm": 115.77137756347656,
      "learning_rate": 9.74249448864091e-05,
      "loss": 0.06861889362335205,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.0624,
      "step": 4960
    },
    {
      "entropy_step_1": 0.2259,
      "entropy_step_2": 0.2259,
      "entropy_step_3": 0.239,
      "epoch": 0.14782427649385801,
      "grad_norm": 50.118743896484375,
      "learning_rate": 9.740934232511894e-05,
      "loss": 0.06581351161003113,
      "loss_step_1": 0.0714,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0681,
      "step": 4970
    },
    {
      "entropy_step_1": 0.2194,
      "entropy_step_2": 0.2378,
      "entropy_step_3": 0.2299,
      "epoch": 0.1481217096457571,
      "grad_norm": 71.42698669433594,
      "learning_rate": 9.739369389464496e-05,
      "loss": 0.06291902661323548,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0636,
      "step": 4980
    },
    {
      "entropy_step_1": 0.2188,
      "entropy_step_2": 0.2416,
      "entropy_step_3": 0.2299,
      "epoch": 0.14841914279765622,
      "grad_norm": 52.11591720581055,
      "learning_rate": 9.737799961012726e-05,
      "loss": 0.06569313406944274,
      "loss_step_1": 0.0718,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0661,
      "step": 4990
    },
    {
      "entropy_step_1": 0.2075,
      "entropy_step_2": 0.2188,
      "entropy_step_3": 0.2129,
      "epoch": 0.14871657594955534,
      "grad_norm": 54.23481750488281,
      "learning_rate": 9.736225948675026e-05,
      "loss": 0.06307684779167175,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0441,
      "step": 5000
    },
    {
      "epoch": 0.14871657594955534,
      "eval_accuracy": 0.794,
      "eval_f1": 0.791834243469052,
      "eval_loss": 0.1733316034078598,
      "eval_precision": 0.8067663900899847,
      "eval_recall": 0.794,
      "eval_runtime": 6.438,
      "eval_samples_per_second": 77.663,
      "eval_steps_per_second": 9.786,
      "step": 5000
    },
    {
      "entropy_step_1": 0.2227,
      "entropy_step_2": 0.2209,
      "entropy_step_3": 0.2262,
      "epoch": 0.14901400910145446,
      "grad_norm": 208.07557678222656,
      "learning_rate": 9.734647353974275e-05,
      "loss": 0.06312817335128784,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0592,
      "loss_step_3": 0.0598,
      "step": 5010
    },
    {
      "entropy_step_1": 0.2151,
      "entropy_step_2": 0.2087,
      "entropy_step_3": 0.2005,
      "epoch": 0.14931144225335355,
      "grad_norm": 85.23271179199219,
      "learning_rate": 9.733064178437786e-05,
      "loss": 0.06466197371482849,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0565,
      "step": 5020
    },
    {
      "entropy_step_1": 0.2204,
      "entropy_step_2": 0.2212,
      "entropy_step_3": 0.2212,
      "epoch": 0.14960887540525267,
      "grad_norm": 60.99701690673828,
      "learning_rate": 9.7314764235973e-05,
      "loss": 0.07290710806846619,
      "loss_step_1": 0.0738,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.07,
      "step": 5030
    },
    {
      "entropy_step_1": 0.2544,
      "entropy_step_2": 0.2539,
      "entropy_step_3": 0.2699,
      "epoch": 0.14990630855715179,
      "grad_norm": 118.81817626953125,
      "learning_rate": 9.729884090988994e-05,
      "loss": 0.07992799878120423,
      "loss_step_1": 0.0758,
      "loss_step_2": 0.0756,
      "loss_step_3": 0.0841,
      "step": 5040
    },
    {
      "entropy_step_1": 0.1954,
      "entropy_step_2": 0.2014,
      "entropy_step_3": 0.2124,
      "epoch": 0.1502037417090509,
      "grad_norm": 78.73497772216797,
      "learning_rate": 9.72828718215347e-05,
      "loss": 0.05840023756027222,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0545,
      "step": 5050
    },
    {
      "entropy_step_1": 0.2325,
      "entropy_step_2": 0.2438,
      "entropy_step_3": 0.2526,
      "epoch": 0.15050117486095,
      "grad_norm": 98.08887481689453,
      "learning_rate": 9.72668569863576e-05,
      "loss": 0.06885097026824952,
      "loss_step_1": 0.069,
      "loss_step_2": 0.068,
      "loss_step_3": 0.068,
      "step": 5060
    },
    {
      "entropy_step_1": 0.1859,
      "entropy_step_2": 0.1913,
      "entropy_step_3": 0.1838,
      "epoch": 0.1507986080128491,
      "grad_norm": 56.83107376098633,
      "learning_rate": 9.725079641985322e-05,
      "loss": 0.056546366214752196,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0494,
      "step": 5070
    },
    {
      "entropy_step_1": 0.2191,
      "entropy_step_2": 0.2163,
      "entropy_step_3": 0.2086,
      "epoch": 0.15109604116474823,
      "grad_norm": 124.43631744384766,
      "learning_rate": 9.723469013756035e-05,
      "loss": 0.0672774851322174,
      "loss_step_1": 0.0674,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0622,
      "step": 5080
    },
    {
      "entropy_step_1": 0.235,
      "entropy_step_2": 0.2267,
      "entropy_step_3": 0.2336,
      "epoch": 0.15139347431664735,
      "grad_norm": 47.798500061035156,
      "learning_rate": 9.721853815506206e-05,
      "loss": 0.06590728759765625,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0704,
      "step": 5090
    },
    {
      "entropy_step_1": 0.2445,
      "entropy_step_2": 0.2323,
      "entropy_step_3": 0.2279,
      "epoch": 0.15169090746854644,
      "grad_norm": 88.26029968261719,
      "learning_rate": 9.72023404879856e-05,
      "loss": 0.07301787734031677,
      "loss_step_1": 0.0713,
      "loss_step_2": 0.071,
      "loss_step_3": 0.0688,
      "step": 5100
    },
    {
      "entropy_step_1": 0.2053,
      "entropy_step_2": 0.1981,
      "entropy_step_3": 0.2098,
      "epoch": 0.15198834062044556,
      "grad_norm": 73.34039306640625,
      "learning_rate": 9.718609715200241e-05,
      "loss": 0.06856504678726197,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0668,
      "step": 5110
    },
    {
      "entropy_step_1": 0.2572,
      "entropy_step_2": 0.256,
      "entropy_step_3": 0.2643,
      "epoch": 0.15228577377234467,
      "grad_norm": 41.78804016113281,
      "learning_rate": 9.716980816282821e-05,
      "loss": 0.07772153615951538,
      "loss_step_1": 0.0755,
      "loss_step_2": 0.0747,
      "loss_step_3": 0.0715,
      "step": 5120
    },
    {
      "entropy_step_1": 0.2296,
      "entropy_step_2": 0.2256,
      "entropy_step_3": 0.2073,
      "epoch": 0.15258320692424376,
      "grad_norm": 57.9200325012207,
      "learning_rate": 9.715347353622275e-05,
      "loss": 0.0640830636024475,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0499,
      "step": 5130
    },
    {
      "entropy_step_1": 0.2287,
      "entropy_step_2": 0.2198,
      "entropy_step_3": 0.2152,
      "epoch": 0.15288064007614288,
      "grad_norm": 72.92306518554688,
      "learning_rate": 9.713709328799003e-05,
      "loss": 0.06801908612251281,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0694,
      "loss_step_3": 0.0533,
      "step": 5140
    },
    {
      "entropy_step_1": 0.2169,
      "entropy_step_2": 0.2158,
      "entropy_step_3": 0.1968,
      "epoch": 0.153178073228042,
      "grad_norm": 31.205690383911133,
      "learning_rate": 9.712066743397817e-05,
      "loss": 0.06194753646850586,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0691,
      "step": 5150
    },
    {
      "entropy_step_1": 0.222,
      "entropy_step_2": 0.2289,
      "entropy_step_3": 0.2219,
      "epoch": 0.15347550637994112,
      "grad_norm": 71.76648712158203,
      "learning_rate": 9.710419599007939e-05,
      "loss": 0.06722007393836975,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0631,
      "step": 5160
    },
    {
      "entropy_step_1": 0.2294,
      "entropy_step_2": 0.2323,
      "entropy_step_3": 0.2254,
      "epoch": 0.1537729395318402,
      "grad_norm": 72.33699798583984,
      "learning_rate": 9.708767897223003e-05,
      "loss": 0.07324386835098266,
      "loss_step_1": 0.0799,
      "loss_step_2": 0.0792,
      "loss_step_3": 0.0745,
      "step": 5170
    },
    {
      "entropy_step_1": 0.2486,
      "entropy_step_2": 0.2577,
      "entropy_step_3": 0.2731,
      "epoch": 0.15407037268373933,
      "grad_norm": 61.215667724609375,
      "learning_rate": 9.707111639641057e-05,
      "loss": 0.062196379899978636,
      "loss_step_1": 0.0683,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0754,
      "step": 5180
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2224,
      "entropy_step_3": 0.227,
      "epoch": 0.15436780583563844,
      "grad_norm": 92.80889129638672,
      "learning_rate": 9.705450827864549e-05,
      "loss": 0.06256375908851623,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0634,
      "step": 5190
    },
    {
      "entropy_step_1": 0.2364,
      "entropy_step_2": 0.245,
      "entropy_step_3": 0.2695,
      "epoch": 0.15466523898753756,
      "grad_norm": 28.8593692779541,
      "learning_rate": 9.703785463500338e-05,
      "loss": 0.06254502534866332,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0758,
      "step": 5200
    },
    {
      "entropy_step_1": 0.1884,
      "entropy_step_2": 0.1895,
      "entropy_step_3": 0.1926,
      "epoch": 0.15496267213943665,
      "grad_norm": 67.05595397949219,
      "learning_rate": 9.702115548159687e-05,
      "loss": 0.0560248076915741,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0547,
      "step": 5210
    },
    {
      "entropy_step_1": 0.2043,
      "entropy_step_2": 0.2091,
      "entropy_step_3": 0.2161,
      "epoch": 0.15526010529133577,
      "grad_norm": 120.8702392578125,
      "learning_rate": 9.70044108345826e-05,
      "loss": 0.06601104736328126,
      "loss_step_1": 0.0691,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.0674,
      "step": 5220
    },
    {
      "entropy_step_1": 0.2199,
      "entropy_step_2": 0.2288,
      "entropy_step_3": 0.2326,
      "epoch": 0.1555575384432349,
      "grad_norm": 48.09918212890625,
      "learning_rate": 9.698762071016129e-05,
      "loss": 0.0585346519947052,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0646,
      "step": 5230
    },
    {
      "entropy_step_1": 0.1949,
      "entropy_step_2": 0.2017,
      "entropy_step_3": 0.2138,
      "epoch": 0.155854971595134,
      "grad_norm": 98.90666198730469,
      "learning_rate": 9.697078512457756e-05,
      "loss": 0.061171102523803714,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0715,
      "step": 5240
    },
    {
      "entropy_step_1": 0.2134,
      "entropy_step_2": 0.2189,
      "entropy_step_3": 0.2255,
      "epoch": 0.1561524047470331,
      "grad_norm": 39.46408462524414,
      "learning_rate": 9.69539040941201e-05,
      "loss": 0.06631507277488709,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0761,
      "step": 5250
    },
    {
      "entropy_step_1": 0.2326,
      "entropy_step_2": 0.2314,
      "entropy_step_3": 0.2336,
      "epoch": 0.15644983789893221,
      "grad_norm": 262.15625,
      "learning_rate": 9.693697763512153e-05,
      "loss": 0.06891655325889587,
      "loss_step_1": 0.07,
      "loss_step_2": 0.0693,
      "loss_step_3": 0.0774,
      "step": 5260
    },
    {
      "entropy_step_1": 0.2179,
      "entropy_step_2": 0.2231,
      "entropy_step_3": 0.2296,
      "epoch": 0.15674727105083133,
      "grad_norm": 67.7482681274414,
      "learning_rate": 9.692000576395843e-05,
      "loss": 0.06400622129440307,
      "loss_step_1": 0.0743,
      "loss_step_2": 0.0735,
      "loss_step_3": 0.0793,
      "step": 5270
    },
    {
      "entropy_step_1": 0.2265,
      "entropy_step_2": 0.2336,
      "entropy_step_3": 0.2346,
      "epoch": 0.15704470420273042,
      "grad_norm": 32.60036849975586,
      "learning_rate": 9.69029884970513e-05,
      "loss": 0.06562511920928955,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0624,
      "step": 5280
    },
    {
      "entropy_step_1": 0.1826,
      "entropy_step_2": 0.1908,
      "entropy_step_3": 0.1906,
      "epoch": 0.15734213735462954,
      "grad_norm": 101.91472625732422,
      "learning_rate": 9.688592585086462e-05,
      "loss": 0.07095950245857238,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.0738,
      "step": 5290
    },
    {
      "entropy_step_1": 0.231,
      "entropy_step_2": 0.2405,
      "entropy_step_3": 0.2433,
      "epoch": 0.15763957050652866,
      "grad_norm": 74.16458892822266,
      "learning_rate": 9.686881784190669e-05,
      "loss": 0.07079254388809204,
      "loss_step_1": 0.0772,
      "loss_step_2": 0.0754,
      "loss_step_3": 0.0815,
      "step": 5300
    },
    {
      "entropy_step_1": 0.2164,
      "entropy_step_2": 0.2315,
      "entropy_step_3": 0.2263,
      "epoch": 0.15793700365842778,
      "grad_norm": 72.61209106445312,
      "learning_rate": 9.685166448672974e-05,
      "loss": 0.06924202442169189,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0695,
      "step": 5310
    },
    {
      "entropy_step_1": 0.2038,
      "entropy_step_2": 0.2142,
      "entropy_step_3": 0.2148,
      "epoch": 0.15823443681032687,
      "grad_norm": 43.101417541503906,
      "learning_rate": 9.68344658019299e-05,
      "loss": 0.06427812576293945,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0608,
      "step": 5320
    },
    {
      "entropy_step_1": 0.2089,
      "entropy_step_2": 0.2096,
      "entropy_step_3": 0.2238,
      "epoch": 0.15853186996222599,
      "grad_norm": 81.31289672851562,
      "learning_rate": 9.681722180414713e-05,
      "loss": 0.063042551279068,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0674,
      "step": 5330
    },
    {
      "entropy_step_1": 0.2231,
      "entropy_step_2": 0.2158,
      "entropy_step_3": 0.2081,
      "epoch": 0.1588293031141251,
      "grad_norm": 105.83767700195312,
      "learning_rate": 9.679993251006521e-05,
      "loss": 0.05837448239326477,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0581,
      "step": 5340
    },
    {
      "entropy_step_1": 0.2244,
      "entropy_step_2": 0.2144,
      "entropy_step_3": 0.2112,
      "epoch": 0.15912673626602422,
      "grad_norm": 87.61250305175781,
      "learning_rate": 9.67825979364118e-05,
      "loss": 0.07322333455085754,
      "loss_step_1": 0.0729,
      "loss_step_2": 0.0731,
      "loss_step_3": 0.0797,
      "step": 5350
    },
    {
      "entropy_step_1": 0.2232,
      "entropy_step_2": 0.2175,
      "entropy_step_3": 0.2136,
      "epoch": 0.1594241694179233,
      "grad_norm": 59.78135681152344,
      "learning_rate": 9.676521809995831e-05,
      "loss": 0.06567410826683044,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.06,
      "step": 5360
    },
    {
      "entropy_step_1": 0.2011,
      "entropy_step_2": 0.2029,
      "entropy_step_3": 0.2008,
      "epoch": 0.15972160256982243,
      "grad_norm": 87.15909576416016,
      "learning_rate": 9.674779301751996e-05,
      "loss": 0.07030232548713684,
      "loss_step_1": 0.0724,
      "loss_step_2": 0.072,
      "loss_step_3": 0.0615,
      "step": 5370
    },
    {
      "entropy_step_1": 0.2459,
      "entropy_step_2": 0.2521,
      "entropy_step_3": 0.2387,
      "epoch": 0.16001903572172155,
      "grad_norm": 77.11946868896484,
      "learning_rate": 9.673032270595579e-05,
      "loss": 0.07102763056755065,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0711,
      "loss_step_3": 0.0664,
      "step": 5380
    },
    {
      "entropy_step_1": 0.2186,
      "entropy_step_2": 0.2217,
      "entropy_step_3": 0.2107,
      "epoch": 0.16031646887362067,
      "grad_norm": 69.05628967285156,
      "learning_rate": 9.671280718216854e-05,
      "loss": 0.06694316864013672,
      "loss_step_1": 0.0658,
      "loss_step_2": 0.0659,
      "loss_step_3": 0.0558,
      "step": 5390
    },
    {
      "entropy_step_1": 0.249,
      "entropy_step_2": 0.2433,
      "entropy_step_3": 0.2403,
      "epoch": 0.16061390202551976,
      "grad_norm": 152.5409698486328,
      "learning_rate": 9.669524646310471e-05,
      "loss": 0.0653900921344757,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.0608,
      "step": 5400
    },
    {
      "entropy_step_1": 0.1704,
      "entropy_step_2": 0.1674,
      "entropy_step_3": 0.1665,
      "epoch": 0.16091133517741887,
      "grad_norm": 70.17958068847656,
      "learning_rate": 9.667764056575454e-05,
      "loss": 0.05781432390213013,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0539,
      "step": 5410
    },
    {
      "entropy_step_1": 0.2319,
      "entropy_step_2": 0.2247,
      "entropy_step_3": 0.2319,
      "epoch": 0.161208768329318,
      "grad_norm": 99.63750457763672,
      "learning_rate": 9.665998950715201e-05,
      "loss": 0.07167447805404663,
      "loss_step_1": 0.07,
      "loss_step_2": 0.07,
      "loss_step_3": 0.0712,
      "step": 5420
    },
    {
      "entropy_step_1": 0.1959,
      "entropy_step_2": 0.188,
      "entropy_step_3": 0.1877,
      "epoch": 0.1615062014812171,
      "grad_norm": 58.319053649902344,
      "learning_rate": 9.664229330437471e-05,
      "loss": 0.0660612165927887,
      "loss_step_1": 0.0622,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0624,
      "step": 5430
    },
    {
      "entropy_step_1": 0.2388,
      "entropy_step_2": 0.2225,
      "entropy_step_3": 0.2338,
      "epoch": 0.1618036346331162,
      "grad_norm": 93.92231750488281,
      "learning_rate": 9.662455197454398e-05,
      "loss": 0.07396497130393982,
      "loss_step_1": 0.071,
      "loss_step_2": 0.0709,
      "loss_step_3": 0.0727,
      "step": 5440
    },
    {
      "entropy_step_1": 0.232,
      "entropy_step_2": 0.2191,
      "entropy_step_3": 0.2185,
      "epoch": 0.16210106778501532,
      "grad_norm": 35.18596267700195,
      "learning_rate": 9.66067655348248e-05,
      "loss": 0.07328044176101685,
      "loss_step_1": 0.076,
      "loss_step_2": 0.0751,
      "loss_step_3": 0.0648,
      "step": 5450
    },
    {
      "entropy_step_1": 0.2079,
      "entropy_step_2": 0.2014,
      "entropy_step_3": 0.2111,
      "epoch": 0.16239850093691444,
      "grad_norm": 62.36537551879883,
      "learning_rate": 9.658893400242578e-05,
      "loss": 0.06415877938270569,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.07,
      "step": 5460
    },
    {
      "entropy_step_1": 0.2213,
      "entropy_step_2": 0.2201,
      "entropy_step_3": 0.2357,
      "epoch": 0.16269593408881353,
      "grad_norm": 65.66902160644531,
      "learning_rate": 9.657105739459919e-05,
      "loss": 0.07026920914649963,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0696,
      "loss_step_3": 0.0729,
      "step": 5470
    },
    {
      "entropy_step_1": 0.2668,
      "entropy_step_2": 0.2683,
      "entropy_step_3": 0.264,
      "epoch": 0.16299336724071264,
      "grad_norm": 92.89129638671875,
      "learning_rate": 9.655313572864086e-05,
      "loss": 0.07211393117904663,
      "loss_step_1": 0.0745,
      "loss_step_2": 0.0734,
      "loss_step_3": 0.0763,
      "step": 5480
    },
    {
      "entropy_step_1": 0.2062,
      "entropy_step_2": 0.209,
      "entropy_step_3": 0.197,
      "epoch": 0.16329080039261176,
      "grad_norm": 54.85028839111328,
      "learning_rate": 9.653516902189028e-05,
      "loss": 0.07586533427238465,
      "loss_step_1": 0.0707,
      "loss_step_2": 0.0699,
      "loss_step_3": 0.0658,
      "step": 5490
    },
    {
      "entropy_step_1": 0.2064,
      "entropy_step_2": 0.2048,
      "entropy_step_3": 0.1802,
      "epoch": 0.16358823354451088,
      "grad_norm": 75.50467681884766,
      "learning_rate": 9.651715729173047e-05,
      "loss": 0.05915030837059021,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0498,
      "step": 5500
    },
    {
      "entropy_step_1": 0.2306,
      "entropy_step_2": 0.232,
      "entropy_step_3": 0.234,
      "epoch": 0.16388566669640997,
      "grad_norm": 42.92985153198242,
      "learning_rate": 9.649910055558803e-05,
      "loss": 0.06447831988334655,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0719,
      "step": 5510
    },
    {
      "entropy_step_1": 0.2383,
      "entropy_step_2": 0.2427,
      "entropy_step_3": 0.2195,
      "epoch": 0.1641830998483091,
      "grad_norm": 39.57537841796875,
      "learning_rate": 9.648099883093308e-05,
      "loss": 0.06412280797958374,
      "loss_step_1": 0.0674,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0651,
      "step": 5520
    },
    {
      "entropy_step_1": 0.2284,
      "entropy_step_2": 0.24,
      "entropy_step_3": 0.2309,
      "epoch": 0.1644805330002082,
      "grad_norm": 98.51581573486328,
      "learning_rate": 9.64628521352793e-05,
      "loss": 0.06971430778503418,
      "loss_step_1": 0.0742,
      "loss_step_2": 0.074,
      "loss_step_3": 0.0746,
      "step": 5530
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.2159,
      "entropy_step_3": 0.2204,
      "epoch": 0.16477796615210732,
      "grad_norm": 53.95223617553711,
      "learning_rate": 9.644466048618385e-05,
      "loss": 0.0586531400680542,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0529,
      "step": 5540
    },
    {
      "entropy_step_1": 0.2114,
      "entropy_step_2": 0.2181,
      "entropy_step_3": 0.2016,
      "epoch": 0.16507539930400641,
      "grad_norm": 113.91924285888672,
      "learning_rate": 9.642642390124743e-05,
      "loss": 0.06544971466064453,
      "loss_step_1": 0.0679,
      "loss_step_2": 0.0675,
      "loss_step_3": 0.0634,
      "step": 5550
    },
    {
      "entropy_step_1": 0.2061,
      "entropy_step_2": 0.22,
      "entropy_step_3": 0.2177,
      "epoch": 0.16537283245590553,
      "grad_norm": 71.58355712890625,
      "learning_rate": 9.640814239811414e-05,
      "loss": 0.06307232975959778,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0618,
      "step": 5560
    },
    {
      "entropy_step_1": 0.2198,
      "entropy_step_2": 0.2269,
      "entropy_step_3": 0.2346,
      "epoch": 0.16567026560780465,
      "grad_norm": 49.17924118041992,
      "learning_rate": 9.63898159944716e-05,
      "loss": 0.06419755220413208,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.0691,
      "step": 5570
    },
    {
      "entropy_step_1": 0.2158,
      "entropy_step_2": 0.2145,
      "entropy_step_3": 0.2167,
      "epoch": 0.16596769875970377,
      "grad_norm": 34.202362060546875,
      "learning_rate": 9.637144470805089e-05,
      "loss": 0.06394060254096985,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.058,
      "step": 5580
    },
    {
      "entropy_step_1": 0.2472,
      "entropy_step_2": 0.238,
      "entropy_step_3": 0.2562,
      "epoch": 0.16626513191160286,
      "grad_norm": 95.59196472167969,
      "learning_rate": 9.635302855662643e-05,
      "loss": 0.06977679133415222,
      "loss_step_1": 0.0722,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0861,
      "step": 5590
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.1975,
      "entropy_step_3": 0.1946,
      "epoch": 0.16656256506350198,
      "grad_norm": 77.39028930664062,
      "learning_rate": 9.633456755801613e-05,
      "loss": 0.07080674767494202,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0692,
      "loss_step_3": 0.0639,
      "step": 5600
    },
    {
      "entropy_step_1": 0.2251,
      "entropy_step_2": 0.2147,
      "entropy_step_3": 0.2175,
      "epoch": 0.1668599982154011,
      "grad_norm": 42.75494384765625,
      "learning_rate": 9.631606173008123e-05,
      "loss": 0.06804988980293274,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0616,
      "step": 5610
    },
    {
      "entropy_step_1": 0.2486,
      "entropy_step_2": 0.244,
      "entropy_step_3": 0.2483,
      "epoch": 0.16715743136730019,
      "grad_norm": 274.96435546875,
      "learning_rate": 9.629751109072638e-05,
      "loss": 0.07673221826553345,
      "loss_step_1": 0.078,
      "loss_step_2": 0.0771,
      "loss_step_3": 0.0733,
      "step": 5620
    },
    {
      "entropy_step_1": 0.2045,
      "entropy_step_2": 0.2139,
      "entropy_step_3": 0.2136,
      "epoch": 0.1674548645191993,
      "grad_norm": 99.70991516113281,
      "learning_rate": 9.62789156578996e-05,
      "loss": 0.06613027453422546,
      "loss_step_1": 0.0683,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.066,
      "step": 5630
    },
    {
      "entropy_step_1": 0.2249,
      "entropy_step_2": 0.2444,
      "entropy_step_3": 0.2379,
      "epoch": 0.16775229767109842,
      "grad_norm": 43.55630874633789,
      "learning_rate": 9.62602754495922e-05,
      "loss": 0.07186936140060425,
      "loss_step_1": 0.0717,
      "loss_step_2": 0.0723,
      "loss_step_3": 0.0614,
      "step": 5640
    },
    {
      "entropy_step_1": 0.2469,
      "entropy_step_2": 0.251,
      "entropy_step_3": 0.2568,
      "epoch": 0.16804973082299754,
      "grad_norm": 106.93896484375,
      "learning_rate": 9.624159048383884e-05,
      "loss": 0.07658292651176453,
      "loss_step_1": 0.068,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0734,
      "step": 5650
    },
    {
      "entropy_step_1": 0.2,
      "entropy_step_2": 0.1968,
      "entropy_step_3": 0.2047,
      "epoch": 0.16834716397489663,
      "grad_norm": 76.6806869506836,
      "learning_rate": 9.622286077871747e-05,
      "loss": 0.06451476216316224,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0655,
      "step": 5660
    },
    {
      "entropy_step_1": 0.2356,
      "entropy_step_2": 0.2389,
      "entropy_step_3": 0.2372,
      "epoch": 0.16864459712679575,
      "grad_norm": 41.139041900634766,
      "learning_rate": 9.620408635234936e-05,
      "loss": 0.0656774401664734,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.066,
      "loss_step_3": 0.0715,
      "step": 5670
    },
    {
      "entropy_step_1": 0.2048,
      "entropy_step_2": 0.2126,
      "entropy_step_3": 0.2238,
      "epoch": 0.16894203027869487,
      "grad_norm": 90.29292297363281,
      "learning_rate": 9.618526722289899e-05,
      "loss": 0.06307196617126465,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0609,
      "step": 5680
    },
    {
      "entropy_step_1": 0.2091,
      "entropy_step_2": 0.2135,
      "entropy_step_3": 0.203,
      "epoch": 0.16923946343059398,
      "grad_norm": 45.693199157714844,
      "learning_rate": 9.616640340857415e-05,
      "loss": 0.0625493884086609,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0654,
      "step": 5690
    },
    {
      "entropy_step_1": 0.2182,
      "entropy_step_2": 0.2198,
      "entropy_step_3": 0.2057,
      "epoch": 0.16953689658249307,
      "grad_norm": 67.22010803222656,
      "learning_rate": 9.614749492762582e-05,
      "loss": 0.06379120349884033,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0582,
      "step": 5700
    },
    {
      "entropy_step_1": 0.2264,
      "entropy_step_2": 0.2313,
      "entropy_step_3": 0.2169,
      "epoch": 0.1698343297343922,
      "grad_norm": 66.9043960571289,
      "learning_rate": 9.612854179834822e-05,
      "loss": 0.06475005149841309,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0668,
      "step": 5710
    },
    {
      "entropy_step_1": 0.2108,
      "entropy_step_2": 0.2165,
      "entropy_step_3": 0.2303,
      "epoch": 0.1701317628862913,
      "grad_norm": 29.355607986450195,
      "learning_rate": 9.610954403907877e-05,
      "loss": 0.05776153206825256,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0616,
      "step": 5720
    },
    {
      "entropy_step_1": 0.2005,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2057,
      "epoch": 0.17042919603819043,
      "grad_norm": 82.31197357177734,
      "learning_rate": 9.609050166819804e-05,
      "loss": 0.06593592166900634,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0568,
      "step": 5730
    },
    {
      "entropy_step_1": 0.2265,
      "entropy_step_2": 0.2248,
      "entropy_step_3": 0.216,
      "epoch": 0.17072662919008952,
      "grad_norm": 40.49549102783203,
      "learning_rate": 9.60714147041298e-05,
      "loss": 0.06463494300842285,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0579,
      "step": 5740
    },
    {
      "entropy_step_1": 0.1956,
      "entropy_step_2": 0.189,
      "entropy_step_3": 0.1895,
      "epoch": 0.17102406234198864,
      "grad_norm": 75.59264373779297,
      "learning_rate": 9.605228316534094e-05,
      "loss": 0.06986281275749207,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0652,
      "step": 5750
    },
    {
      "entropy_step_1": 0.2212,
      "entropy_step_2": 0.2138,
      "entropy_step_3": 0.2149,
      "epoch": 0.17132149549388775,
      "grad_norm": 41.339996337890625,
      "learning_rate": 9.603310707034149e-05,
      "loss": 0.05836338400840759,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0582,
      "step": 5760
    },
    {
      "entropy_step_1": 0.2224,
      "entropy_step_2": 0.2206,
      "entropy_step_3": 0.2227,
      "epoch": 0.17161892864578687,
      "grad_norm": 53.874019622802734,
      "learning_rate": 9.601388643768457e-05,
      "loss": 0.07050216794013978,
      "loss_step_1": 0.0787,
      "loss_step_2": 0.0781,
      "loss_step_3": 0.0748,
      "step": 5770
    },
    {
      "entropy_step_1": 0.2556,
      "entropy_step_2": 0.2581,
      "entropy_step_3": 0.2711,
      "epoch": 0.17191636179768596,
      "grad_norm": 107.93016052246094,
      "learning_rate": 9.599462128596643e-05,
      "loss": 0.07296949625015259,
      "loss_step_1": 0.0786,
      "loss_step_2": 0.0773,
      "loss_step_3": 0.0883,
      "step": 5780
    },
    {
      "entropy_step_1": 0.2127,
      "entropy_step_2": 0.2167,
      "entropy_step_3": 0.219,
      "epoch": 0.17221379494958508,
      "grad_norm": 85.64257049560547,
      "learning_rate": 9.597531163382633e-05,
      "loss": 0.06436319351196289,
      "loss_step_1": 0.0622,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0657,
      "step": 5790
    },
    {
      "entropy_step_1": 0.1974,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.2149,
      "epoch": 0.1725112281014842,
      "grad_norm": 97.65151977539062,
      "learning_rate": 9.595595749994664e-05,
      "loss": 0.06568825244903564,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0663,
      "loss_step_3": 0.0622,
      "step": 5800
    },
    {
      "entropy_step_1": 0.2373,
      "entropy_step_2": 0.2405,
      "entropy_step_3": 0.2413,
      "epoch": 0.1728086612533833,
      "grad_norm": 100.63510131835938,
      "learning_rate": 9.593655890305276e-05,
      "loss": 0.06058458089828491,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0585,
      "step": 5810
    },
    {
      "entropy_step_1": 0.2208,
      "entropy_step_2": 0.221,
      "entropy_step_3": 0.2206,
      "epoch": 0.1731060944052824,
      "grad_norm": 41.83860778808594,
      "learning_rate": 9.591711586191308e-05,
      "loss": 0.06836225390434265,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0598,
      "step": 5820
    },
    {
      "entropy_step_1": 0.1993,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.2048,
      "epoch": 0.17340352755718152,
      "grad_norm": 87.67767333984375,
      "learning_rate": 9.5897628395339e-05,
      "loss": 0.06586849689483643,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0589,
      "step": 5830
    },
    {
      "entropy_step_1": 0.2214,
      "entropy_step_2": 0.2101,
      "entropy_step_3": 0.212,
      "epoch": 0.17370096070908064,
      "grad_norm": 60.706539154052734,
      "learning_rate": 9.587809652218492e-05,
      "loss": 0.06490331888198853,
      "loss_step_1": 0.066,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0698,
      "step": 5840
    },
    {
      "entropy_step_1": 0.2089,
      "entropy_step_2": 0.2071,
      "entropy_step_3": 0.1903,
      "epoch": 0.17399839386097973,
      "grad_norm": 73.46343231201172,
      "learning_rate": 9.58585202613482e-05,
      "loss": 0.06298965811729432,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0544,
      "step": 5850
    },
    {
      "entropy_step_1": 0.2273,
      "entropy_step_2": 0.2255,
      "entropy_step_3": 0.2331,
      "epoch": 0.17429582701287885,
      "grad_norm": 68.8997573852539,
      "learning_rate": 9.583889963176913e-05,
      "loss": 0.06447206139564514,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0659,
      "step": 5860
    },
    {
      "entropy_step_1": 0.2129,
      "entropy_step_2": 0.2048,
      "entropy_step_3": 0.1817,
      "epoch": 0.17459326016477797,
      "grad_norm": 59.81831359863281,
      "learning_rate": 9.581923465243095e-05,
      "loss": 0.06611148118972779,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0518,
      "step": 5870
    },
    {
      "entropy_step_1": 0.2132,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2282,
      "epoch": 0.1748906933166771,
      "grad_norm": 53.071102142333984,
      "learning_rate": 9.579952534235977e-05,
      "loss": 0.07487742900848389,
      "loss_step_1": 0.0725,
      "loss_step_2": 0.0718,
      "loss_step_3": 0.0734,
      "step": 5880
    },
    {
      "entropy_step_1": 0.2277,
      "entropy_step_2": 0.2215,
      "entropy_step_3": 0.2275,
      "epoch": 0.17518812646857618,
      "grad_norm": 80.01422119140625,
      "learning_rate": 9.577977172062462e-05,
      "loss": 0.07231720685958862,
      "loss_step_1": 0.0697,
      "loss_step_2": 0.07,
      "loss_step_3": 0.0718,
      "step": 5890
    },
    {
      "entropy_step_1": 0.2182,
      "entropy_step_2": 0.2123,
      "entropy_step_3": 0.2218,
      "epoch": 0.1754855596204753,
      "grad_norm": 78.9100112915039,
      "learning_rate": 9.575997380633741e-05,
      "loss": 0.05940516591072083,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0624,
      "step": 5900
    },
    {
      "entropy_step_1": 0.2274,
      "entropy_step_2": 0.2266,
      "entropy_step_3": 0.2431,
      "epoch": 0.1757829927723744,
      "grad_norm": 89.43810272216797,
      "learning_rate": 9.57401316186529e-05,
      "loss": 0.066690593957901,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0863,
      "step": 5910
    },
    {
      "entropy_step_1": 0.2205,
      "entropy_step_2": 0.2239,
      "entropy_step_3": 0.2289,
      "epoch": 0.17608042592427353,
      "grad_norm": 91.92916107177734,
      "learning_rate": 9.572024517676866e-05,
      "loss": 0.0671968162059784,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0554,
      "step": 5920
    },
    {
      "entropy_step_1": 0.1867,
      "entropy_step_2": 0.1964,
      "entropy_step_3": 0.1924,
      "epoch": 0.17637785907617262,
      "grad_norm": 123.91554260253906,
      "learning_rate": 9.57003144999251e-05,
      "loss": 0.06775065660476684,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0629,
      "step": 5930
    },
    {
      "entropy_step_1": 0.2163,
      "entropy_step_2": 0.2373,
      "entropy_step_3": 0.2223,
      "epoch": 0.17667529222807174,
      "grad_norm": 101.04044342041016,
      "learning_rate": 9.568033960740541e-05,
      "loss": 0.060746073722839355,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0598,
      "step": 5940
    },
    {
      "entropy_step_1": 0.2044,
      "entropy_step_2": 0.2139,
      "entropy_step_3": 0.1919,
      "epoch": 0.17697272537997086,
      "grad_norm": 65.864013671875,
      "learning_rate": 9.566032051853558e-05,
      "loss": 0.0756988286972046,
      "loss_step_1": 0.076,
      "loss_step_2": 0.0747,
      "loss_step_3": 0.0772,
      "step": 5950
    },
    {
      "entropy_step_1": 0.22,
      "entropy_step_2": 0.2238,
      "entropy_step_3": 0.227,
      "epoch": 0.17727015853186998,
      "grad_norm": 44.40154266357422,
      "learning_rate": 9.564025725268435e-05,
      "loss": 0.05954664945602417,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0576,
      "step": 5960
    },
    {
      "entropy_step_1": 0.2057,
      "entropy_step_2": 0.2092,
      "entropy_step_3": 0.2022,
      "epoch": 0.17756759168376907,
      "grad_norm": 303.46722412109375,
      "learning_rate": 9.562014982926322e-05,
      "loss": 0.0685522973537445,
      "loss_step_1": 0.0717,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0727,
      "step": 5970
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.2154,
      "entropy_step_3": 0.234,
      "epoch": 0.17786502483566818,
      "grad_norm": 167.23681640625,
      "learning_rate": 9.559999826772635e-05,
      "loss": 0.06607893705368043,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0664,
      "step": 5980
    },
    {
      "entropy_step_1": 0.2085,
      "entropy_step_2": 0.2155,
      "entropy_step_3": 0.2182,
      "epoch": 0.1781624579875673,
      "grad_norm": 138.01992797851562,
      "learning_rate": 9.557980258757069e-05,
      "loss": 0.06350319981575012,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0697,
      "step": 5990
    },
    {
      "entropy_step_1": 0.1991,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.209,
      "epoch": 0.1784598911394664,
      "grad_norm": 55.26619338989258,
      "learning_rate": 9.555956280833582e-05,
      "loss": 0.06480395793914795,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0632,
      "step": 6000
    },
    {
      "epoch": 0.1784598911394664,
      "eval_accuracy": 0.856,
      "eval_f1": 0.8555469953775039,
      "eval_loss": 0.14669294655323029,
      "eval_precision": 0.8605223928964936,
      "eval_recall": 0.856,
      "eval_runtime": 6.2729,
      "eval_samples_per_second": 79.707,
      "eval_steps_per_second": 10.043,
      "step": 6000
    },
    {
      "entropy_step_1": 0.219,
      "entropy_step_2": 0.2319,
      "entropy_step_3": 0.2312,
      "epoch": 0.1787573242913655,
      "grad_norm": 158.2210235595703,
      "learning_rate": 9.553927894960401e-05,
      "loss": 0.06262257695198059,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0562,
      "step": 6010
    },
    {
      "entropy_step_1": 0.1866,
      "entropy_step_2": 0.2043,
      "entropy_step_3": 0.1984,
      "epoch": 0.17905475744326463,
      "grad_norm": 95.11447143554688,
      "learning_rate": 9.551895103100017e-05,
      "loss": 0.06659491062164306,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0545,
      "step": 6020
    },
    {
      "entropy_step_1": 0.2031,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2021,
      "epoch": 0.17935219059516375,
      "grad_norm": 28.566024780273438,
      "learning_rate": 9.549857907219184e-05,
      "loss": 0.05665794610977173,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0575,
      "step": 6030
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1778,
      "entropy_step_3": 0.1628,
      "epoch": 0.17964962374706284,
      "grad_norm": 55.56345748901367,
      "learning_rate": 9.547816309288915e-05,
      "loss": 0.06560707688331605,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.0562,
      "step": 6040
    },
    {
      "entropy_step_1": 0.2678,
      "entropy_step_2": 0.2624,
      "entropy_step_3": 0.2592,
      "epoch": 0.17994705689896195,
      "grad_norm": 37.189781188964844,
      "learning_rate": 9.545770311284483e-05,
      "loss": 0.06918668150901794,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0701,
      "step": 6050
    },
    {
      "entropy_step_1": 0.1949,
      "entropy_step_2": 0.188,
      "entropy_step_3": 0.1834,
      "epoch": 0.18024449005086107,
      "grad_norm": 89.78742980957031,
      "learning_rate": 9.543719915185423e-05,
      "loss": 0.07874386310577393,
      "loss_step_1": 0.0814,
      "loss_step_2": 0.0824,
      "loss_step_3": 0.0738,
      "step": 6060
    },
    {
      "entropy_step_1": 0.2313,
      "entropy_step_2": 0.2274,
      "entropy_step_3": 0.2273,
      "epoch": 0.1805419232027602,
      "grad_norm": 135.56715393066406,
      "learning_rate": 9.541665122975518e-05,
      "loss": 0.06623731851577759,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0601,
      "step": 6070
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.2043,
      "entropy_step_3": 0.1979,
      "epoch": 0.18083935635465928,
      "grad_norm": 54.64655685424805,
      "learning_rate": 9.539605936642808e-05,
      "loss": 0.05790523886680603,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0604,
      "step": 6080
    },
    {
      "entropy_step_1": 0.213,
      "entropy_step_2": 0.225,
      "entropy_step_3": 0.2111,
      "epoch": 0.1811367895065584,
      "grad_norm": 60.73484802246094,
      "learning_rate": 9.537542358179585e-05,
      "loss": 0.0678011953830719,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.0646,
      "step": 6090
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.213,
      "entropy_step_3": 0.216,
      "epoch": 0.18143422265845752,
      "grad_norm": 120.36260986328125,
      "learning_rate": 9.535474389582386e-05,
      "loss": 0.06522369980812073,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0597,
      "step": 6100
    },
    {
      "entropy_step_1": 0.2172,
      "entropy_step_2": 0.2241,
      "entropy_step_3": 0.2307,
      "epoch": 0.18173165581035663,
      "grad_norm": 111.98409271240234,
      "learning_rate": 9.533402032852004e-05,
      "loss": 0.07750526070594788,
      "loss_step_1": 0.0854,
      "loss_step_2": 0.0845,
      "loss_step_3": 0.0951,
      "step": 6110
    },
    {
      "entropy_step_1": 0.2123,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.2363,
      "epoch": 0.18202908896225573,
      "grad_norm": 56.107139587402344,
      "learning_rate": 9.531325289993466e-05,
      "loss": 0.059782886505126955,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0617,
      "step": 6120
    },
    {
      "entropy_step_1": 0.1886,
      "entropy_step_2": 0.1919,
      "entropy_step_3": 0.1941,
      "epoch": 0.18232652211415484,
      "grad_norm": 45.4063835144043,
      "learning_rate": 9.529244163016055e-05,
      "loss": 0.06904675364494324,
      "loss_step_1": 0.0733,
      "loss_step_2": 0.0716,
      "loss_step_3": 0.0686,
      "step": 6130
    },
    {
      "entropy_step_1": 0.2024,
      "entropy_step_2": 0.2034,
      "entropy_step_3": 0.2036,
      "epoch": 0.18262395526605396,
      "grad_norm": 69.9322738647461,
      "learning_rate": 9.527158653933285e-05,
      "loss": 0.0574566662311554,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0573,
      "step": 6140
    },
    {
      "entropy_step_1": 0.2311,
      "entropy_step_2": 0.2328,
      "entropy_step_3": 0.2476,
      "epoch": 0.18292138841795305,
      "grad_norm": 175.54042053222656,
      "learning_rate": 9.525068764762918e-05,
      "loss": 0.07159168720245361,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0695,
      "step": 6150
    },
    {
      "entropy_step_1": 0.24,
      "entropy_step_2": 0.2376,
      "entropy_step_3": 0.2468,
      "epoch": 0.18321882156985217,
      "grad_norm": 74.64618682861328,
      "learning_rate": 9.522974497526948e-05,
      "loss": 0.07036988139152527,
      "loss_step_1": 0.0726,
      "loss_step_2": 0.072,
      "loss_step_3": 0.0903,
      "step": 6160
    },
    {
      "entropy_step_1": 0.2099,
      "entropy_step_2": 0.2035,
      "entropy_step_3": 0.2056,
      "epoch": 0.1835162547217513,
      "grad_norm": 57.494171142578125,
      "learning_rate": 9.520875854251609e-05,
      "loss": 0.05572736859321594,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.052,
      "step": 6170
    },
    {
      "entropy_step_1": 0.2424,
      "entropy_step_2": 0.2385,
      "entropy_step_3": 0.2461,
      "epoch": 0.1838136878736504,
      "grad_norm": 102.74459838867188,
      "learning_rate": 9.518772836967365e-05,
      "loss": 0.06973839998245239,
      "loss_step_1": 0.0779,
      "loss_step_2": 0.0783,
      "loss_step_3": 0.0757,
      "step": 6180
    },
    {
      "entropy_step_1": 0.234,
      "entropy_step_2": 0.2348,
      "entropy_step_3": 0.2339,
      "epoch": 0.1841111210255495,
      "grad_norm": 119.75819396972656,
      "learning_rate": 9.516665447708912e-05,
      "loss": 0.06379510760307312,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.071,
      "step": 6190
    },
    {
      "entropy_step_1": 0.1748,
      "entropy_step_2": 0.1762,
      "entropy_step_3": 0.1728,
      "epoch": 0.1844085541774486,
      "grad_norm": 46.26518249511719,
      "learning_rate": 9.514553688515183e-05,
      "loss": 0.061240381002426146,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.06,
      "step": 6200
    },
    {
      "entropy_step_1": 0.2124,
      "entropy_step_2": 0.2164,
      "entropy_step_3": 0.2079,
      "epoch": 0.18470598732934773,
      "grad_norm": 82.5500259399414,
      "learning_rate": 9.512437561429331e-05,
      "loss": 0.06480839252471923,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0615,
      "step": 6210
    },
    {
      "entropy_step_1": 0.2323,
      "entropy_step_2": 0.2341,
      "entropy_step_3": 0.246,
      "epoch": 0.18500342048124685,
      "grad_norm": 104.79817962646484,
      "learning_rate": 9.510317068498736e-05,
      "loss": 0.0673658788204193,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0627,
      "step": 6220
    },
    {
      "entropy_step_1": 0.2172,
      "entropy_step_2": 0.2182,
      "entropy_step_3": 0.2398,
      "epoch": 0.18530085363314594,
      "grad_norm": 27.31346893310547,
      "learning_rate": 9.508192211775006e-05,
      "loss": 0.06280574202537537,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.0674,
      "step": 6230
    },
    {
      "entropy_step_1": 0.2134,
      "entropy_step_2": 0.218,
      "entropy_step_3": 0.2295,
      "epoch": 0.18559828678504506,
      "grad_norm": 110.57732391357422,
      "learning_rate": 9.506062993313968e-05,
      "loss": 0.06789652705192566,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0795,
      "step": 6240
    },
    {
      "entropy_step_1": 0.1991,
      "entropy_step_2": 0.2025,
      "entropy_step_3": 0.2031,
      "epoch": 0.18589571993694418,
      "grad_norm": 75.50672149658203,
      "learning_rate": 9.503929415175671e-05,
      "loss": 0.05858739018440247,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.06,
      "step": 6250
    },
    {
      "entropy_step_1": 0.2138,
      "entropy_step_2": 0.2143,
      "entropy_step_3": 0.2167,
      "epoch": 0.1861931530888433,
      "grad_norm": 159.8157196044922,
      "learning_rate": 9.50179147942438e-05,
      "loss": 0.07272651195526122,
      "loss_step_1": 0.0782,
      "loss_step_2": 0.0783,
      "loss_step_3": 0.0816,
      "step": 6260
    },
    {
      "entropy_step_1": 0.2322,
      "entropy_step_2": 0.232,
      "entropy_step_3": 0.2374,
      "epoch": 0.18649058624074238,
      "grad_norm": 137.36668395996094,
      "learning_rate": 9.499649188128577e-05,
      "loss": 0.0699052631855011,
      "loss_step_1": 0.0714,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.072,
      "step": 6270
    },
    {
      "entropy_step_1": 0.237,
      "entropy_step_2": 0.2387,
      "entropy_step_3": 0.2671,
      "epoch": 0.1867880193926415,
      "grad_norm": 51.03142547607422,
      "learning_rate": 9.497502543360957e-05,
      "loss": 0.06557508707046508,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0751,
      "step": 6280
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.204,
      "entropy_step_3": 0.1877,
      "epoch": 0.18708545254454062,
      "grad_norm": 86.6373062133789,
      "learning_rate": 9.495351547198428e-05,
      "loss": 0.0712066650390625,
      "loss_step_1": 0.075,
      "loss_step_2": 0.0745,
      "loss_step_3": 0.0782,
      "step": 6290
    },
    {
      "entropy_step_1": 0.2523,
      "entropy_step_2": 0.2571,
      "entropy_step_3": 0.2717,
      "epoch": 0.18738288569643974,
      "grad_norm": 97.6312484741211,
      "learning_rate": 9.49319620172211e-05,
      "loss": 0.06999778747558594,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0748,
      "step": 6300
    },
    {
      "entropy_step_1": 0.208,
      "entropy_step_2": 0.2107,
      "entropy_step_3": 0.215,
      "epoch": 0.18768031884833883,
      "grad_norm": 89.54481506347656,
      "learning_rate": 9.491036509017329e-05,
      "loss": 0.06812779307365417,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0642,
      "step": 6310
    },
    {
      "entropy_step_1": 0.2255,
      "entropy_step_2": 0.2272,
      "entropy_step_3": 0.2193,
      "epoch": 0.18797775200023795,
      "grad_norm": 212.5825653076172,
      "learning_rate": 9.488872471173616e-05,
      "loss": 0.06366437077522277,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0624,
      "step": 6320
    },
    {
      "entropy_step_1": 0.2184,
      "entropy_step_2": 0.2197,
      "entropy_step_3": 0.2302,
      "epoch": 0.18827518515213706,
      "grad_norm": 90.11189270019531,
      "learning_rate": 9.486704090284707e-05,
      "loss": 0.06314131021499633,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.062,
      "step": 6330
    },
    {
      "entropy_step_1": 0.1967,
      "entropy_step_2": 0.1969,
      "entropy_step_3": 0.2027,
      "epoch": 0.18857261830403615,
      "grad_norm": 331.96026611328125,
      "learning_rate": 9.484531368448539e-05,
      "loss": 0.06640812158584594,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0634,
      "step": 6340
    },
    {
      "entropy_step_1": 0.2381,
      "entropy_step_2": 0.2421,
      "entropy_step_3": 0.2446,
      "epoch": 0.18887005145593527,
      "grad_norm": 141.05245971679688,
      "learning_rate": 9.482354307767251e-05,
      "loss": 0.07524219155311584,
      "loss_step_1": 0.0835,
      "loss_step_2": 0.0829,
      "loss_step_3": 0.0894,
      "step": 6350
    },
    {
      "entropy_step_1": 0.1958,
      "entropy_step_2": 0.1983,
      "entropy_step_3": 0.2082,
      "epoch": 0.1891674846078344,
      "grad_norm": 230.9766845703125,
      "learning_rate": 9.48017291034718e-05,
      "loss": 0.06122902035713196,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0585,
      "step": 6360
    },
    {
      "entropy_step_1": 0.2436,
      "entropy_step_2": 0.2451,
      "entropy_step_3": 0.2376,
      "epoch": 0.1894649177597335,
      "grad_norm": 75.83438110351562,
      "learning_rate": 9.477987178298857e-05,
      "loss": 0.06450847387313843,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0546,
      "step": 6370
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2076,
      "entropy_step_3": 0.2127,
      "epoch": 0.1897623509116326,
      "grad_norm": 124.6250228881836,
      "learning_rate": 9.475797113737007e-05,
      "loss": 0.06116616129875183,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0587,
      "step": 6380
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.1966,
      "entropy_step_3": 0.1955,
      "epoch": 0.19005978406353172,
      "grad_norm": 40.51316452026367,
      "learning_rate": 9.473602718780548e-05,
      "loss": 0.059025251865386964,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0622,
      "step": 6390
    },
    {
      "entropy_step_1": 0.2344,
      "entropy_step_2": 0.2285,
      "entropy_step_3": 0.2288,
      "epoch": 0.19035721721543084,
      "grad_norm": 37.58058547973633,
      "learning_rate": 9.471403995552584e-05,
      "loss": 0.07048228979110718,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0676,
      "loss_step_3": 0.0704,
      "step": 6400
    },
    {
      "entropy_step_1": 0.2162,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.2055,
      "epoch": 0.19065465036732995,
      "grad_norm": 99.24775695800781,
      "learning_rate": 9.469200946180414e-05,
      "loss": 0.06676225066184997,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0701,
      "step": 6410
    },
    {
      "entropy_step_1": 0.2309,
      "entropy_step_2": 0.2266,
      "entropy_step_3": 0.2258,
      "epoch": 0.19095208351922904,
      "grad_norm": 36.49147033691406,
      "learning_rate": 9.466993572795515e-05,
      "loss": 0.06927371025085449,
      "loss_step_1": 0.0697,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0556,
      "step": 6420
    },
    {
      "entropy_step_1": 0.1661,
      "entropy_step_2": 0.1603,
      "entropy_step_3": 0.1562,
      "epoch": 0.19124951667112816,
      "grad_norm": 125.87777709960938,
      "learning_rate": 9.464781877533554e-05,
      "loss": 0.06029604077339172,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.062,
      "step": 6430
    },
    {
      "entropy_step_1": 0.215,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.2177,
      "epoch": 0.19154694982302728,
      "grad_norm": 32.568870544433594,
      "learning_rate": 9.462565862534374e-05,
      "loss": 0.06547333598136902,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0559,
      "step": 6440
    },
    {
      "entropy_step_1": 0.2139,
      "entropy_step_2": 0.2115,
      "entropy_step_3": 0.2218,
      "epoch": 0.1918443829749264,
      "grad_norm": 104.29120635986328,
      "learning_rate": 9.460345529941998e-05,
      "loss": 0.07128658294677734,
      "loss_step_1": 0.0729,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.0767,
      "step": 6450
    },
    {
      "entropy_step_1": 0.2499,
      "entropy_step_2": 0.2511,
      "entropy_step_3": 0.2404,
      "epoch": 0.1921418161268255,
      "grad_norm": 43.17768478393555,
      "learning_rate": 9.45812088190463e-05,
      "loss": 0.0701907753944397,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0726,
      "step": 6460
    },
    {
      "entropy_step_1": 0.2021,
      "entropy_step_2": 0.2079,
      "entropy_step_3": 0.1982,
      "epoch": 0.1924392492787246,
      "grad_norm": 59.77560806274414,
      "learning_rate": 9.45589192057465e-05,
      "loss": 0.0644864022731781,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0539,
      "step": 6470
    },
    {
      "entropy_step_1": 0.2559,
      "entropy_step_2": 0.2656,
      "entropy_step_3": 0.2395,
      "epoch": 0.19273668243062372,
      "grad_norm": 63.22099304199219,
      "learning_rate": 9.453658648108604e-05,
      "loss": 0.07528776526451111,
      "loss_step_1": 0.0764,
      "loss_step_2": 0.0762,
      "loss_step_3": 0.0747,
      "step": 6480
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.2146,
      "entropy_step_3": 0.2188,
      "epoch": 0.19303411558252284,
      "grad_norm": 79.61250305175781,
      "learning_rate": 9.451421066667216e-05,
      "loss": 0.07296299934387207,
      "loss_step_1": 0.0737,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.0729,
      "step": 6490
    },
    {
      "entropy_step_1": 0.2281,
      "entropy_step_2": 0.2298,
      "entropy_step_3": 0.2466,
      "epoch": 0.19333154873442193,
      "grad_norm": 81.41826629638672,
      "learning_rate": 9.449179178415375e-05,
      "loss": 0.06797922253608704,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0757,
      "step": 6500
    },
    {
      "entropy_step_1": 0.2103,
      "entropy_step_2": 0.2092,
      "entropy_step_3": 0.2006,
      "epoch": 0.19362898188632105,
      "grad_norm": 62.352806091308594,
      "learning_rate": 9.446932985522139e-05,
      "loss": 0.06535433530807495,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.0592,
      "step": 6510
    },
    {
      "entropy_step_1": 0.2264,
      "entropy_step_2": 0.2258,
      "entropy_step_3": 0.2333,
      "epoch": 0.19392641503822017,
      "grad_norm": 89.46855926513672,
      "learning_rate": 9.44468249016073e-05,
      "loss": 0.07696073651313781,
      "loss_step_1": 0.084,
      "loss_step_2": 0.0846,
      "loss_step_3": 0.1012,
      "step": 6520
    },
    {
      "entropy_step_1": 0.2396,
      "entropy_step_2": 0.239,
      "entropy_step_3": 0.2568,
      "epoch": 0.19422384819011926,
      "grad_norm": 58.7336311340332,
      "learning_rate": 9.442427694508533e-05,
      "loss": 0.062482357025146484,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0658,
      "step": 6530
    },
    {
      "entropy_step_1": 0.2095,
      "entropy_step_2": 0.2088,
      "entropy_step_3": 0.2237,
      "epoch": 0.19452128134201838,
      "grad_norm": 69.32096099853516,
      "learning_rate": 9.440168600747094e-05,
      "loss": 0.06445941925048829,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0628,
      "step": 6540
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.185,
      "entropy_step_3": 0.1661,
      "epoch": 0.1948187144939175,
      "grad_norm": 47.2012825012207,
      "learning_rate": 9.437905211062117e-05,
      "loss": 0.053538888692855835,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.0465,
      "step": 6550
    },
    {
      "entropy_step_1": 0.1931,
      "entropy_step_2": 0.1937,
      "entropy_step_3": 0.172,
      "epoch": 0.1951161476458166,
      "grad_norm": 187.96534729003906,
      "learning_rate": 9.43563752764346e-05,
      "loss": 0.06188880205154419,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0554,
      "step": 6560
    },
    {
      "entropy_step_1": 0.2016,
      "entropy_step_2": 0.2021,
      "entropy_step_3": 0.2034,
      "epoch": 0.1954135807977157,
      "grad_norm": 94.80298614501953,
      "learning_rate": 9.433365552685141e-05,
      "loss": 0.05990006923675537,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.079,
      "step": 6570
    },
    {
      "entropy_step_1": 0.1949,
      "entropy_step_2": 0.1989,
      "entropy_step_3": 0.2308,
      "epoch": 0.19571101394961482,
      "grad_norm": 170.6454315185547,
      "learning_rate": 9.431089288385325e-05,
      "loss": 0.0599044144153595,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0749,
      "step": 6580
    },
    {
      "entropy_step_1": 0.2028,
      "entropy_step_2": 0.2106,
      "entropy_step_3": 0.1986,
      "epoch": 0.19600844710151394,
      "grad_norm": 33.36080551147461,
      "learning_rate": 9.428808736946328e-05,
      "loss": 0.05954192876815796,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0595,
      "step": 6590
    },
    {
      "entropy_step_1": 0.2155,
      "entropy_step_2": 0.2253,
      "entropy_step_3": 0.239,
      "epoch": 0.19630588025341306,
      "grad_norm": 48.81224060058594,
      "learning_rate": 9.426523900574616e-05,
      "loss": 0.05861917734146118,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0607,
      "step": 6600
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.2051,
      "epoch": 0.19660331340531215,
      "grad_norm": 111.66925811767578,
      "learning_rate": 9.424234781480798e-05,
      "loss": 0.06460691094398499,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0677,
      "step": 6610
    },
    {
      "entropy_step_1": 0.2139,
      "entropy_step_2": 0.2182,
      "entropy_step_3": 0.2178,
      "epoch": 0.19690074655721126,
      "grad_norm": 175.781005859375,
      "learning_rate": 9.421941381879631e-05,
      "loss": 0.06692473888397217,
      "loss_step_1": 0.0724,
      "loss_step_2": 0.0715,
      "loss_step_3": 0.0703,
      "step": 6620
    },
    {
      "entropy_step_1": 0.2164,
      "entropy_step_2": 0.2223,
      "entropy_step_3": 0.2108,
      "epoch": 0.19719817970911038,
      "grad_norm": 63.12721633911133,
      "learning_rate": 9.419643703990004e-05,
      "loss": 0.06578449606895446,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0557,
      "step": 6630
    },
    {
      "entropy_step_1": 0.2163,
      "entropy_step_2": 0.2264,
      "entropy_step_3": 0.2214,
      "epoch": 0.1974956128610095,
      "grad_norm": 72.86148071289062,
      "learning_rate": 9.417341750034958e-05,
      "loss": 0.07163288593292236,
      "loss_step_1": 0.0725,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0667,
      "step": 6640
    },
    {
      "entropy_step_1": 0.218,
      "entropy_step_2": 0.2228,
      "entropy_step_3": 0.1959,
      "epoch": 0.1977930460129086,
      "grad_norm": 62.15083694458008,
      "learning_rate": 9.415035522241662e-05,
      "loss": 0.06668326258659363,
      "loss_step_1": 0.0683,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0568,
      "step": 6650
    },
    {
      "entropy_step_1": 0.2191,
      "entropy_step_2": 0.2212,
      "entropy_step_3": 0.2304,
      "epoch": 0.1980904791648077,
      "grad_norm": 54.67432403564453,
      "learning_rate": 9.412725022841422e-05,
      "loss": 0.07250729203224182,
      "loss_step_1": 0.0742,
      "loss_step_2": 0.0735,
      "loss_step_3": 0.085,
      "step": 6660
    },
    {
      "entropy_step_1": 0.2167,
      "entropy_step_2": 0.2235,
      "entropy_step_3": 0.2204,
      "epoch": 0.19838791231670683,
      "grad_norm": 61.65943145751953,
      "learning_rate": 9.410410254069679e-05,
      "loss": 0.07098269462585449,
      "loss_step_1": 0.0767,
      "loss_step_2": 0.0754,
      "loss_step_3": 0.0728,
      "step": 6670
    },
    {
      "entropy_step_1": 0.2304,
      "entropy_step_2": 0.2377,
      "entropy_step_3": 0.2345,
      "epoch": 0.19868534546860592,
      "grad_norm": 96.64888000488281,
      "learning_rate": 9.408091218166003e-05,
      "loss": 0.061352938413619995,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.05,
      "step": 6680
    },
    {
      "entropy_step_1": 0.207,
      "entropy_step_2": 0.2094,
      "entropy_step_3": 0.213,
      "epoch": 0.19898277862050504,
      "grad_norm": 227.78634643554688,
      "learning_rate": 9.405767917374093e-05,
      "loss": 0.06562969088554382,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0717,
      "step": 6690
    },
    {
      "entropy_step_1": 0.2218,
      "entropy_step_2": 0.2323,
      "entropy_step_3": 0.2451,
      "epoch": 0.19928021177240415,
      "grad_norm": 53.975669860839844,
      "learning_rate": 9.403440353941773e-05,
      "loss": 0.06063014268875122,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0668,
      "step": 6700
    },
    {
      "entropy_step_1": 0.2202,
      "entropy_step_2": 0.2283,
      "entropy_step_3": 0.2231,
      "epoch": 0.19957764492430327,
      "grad_norm": 59.74839782714844,
      "learning_rate": 9.401108530120993e-05,
      "loss": 0.06930052042007447,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0709,
      "loss_step_3": 0.059,
      "step": 6710
    },
    {
      "entropy_step_1": 0.2127,
      "entropy_step_2": 0.215,
      "entropy_step_3": 0.2285,
      "epoch": 0.19987507807620236,
      "grad_norm": 36.18467330932617,
      "learning_rate": 9.398772448167826e-05,
      "loss": 0.061637842655181886,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0544,
      "step": 6720
    },
    {
      "entropy_step_1": 0.2205,
      "entropy_step_2": 0.2132,
      "entropy_step_3": 0.239,
      "epoch": 0.20017251122810148,
      "grad_norm": 38.493751525878906,
      "learning_rate": 9.39643211034246e-05,
      "loss": 0.06043809652328491,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0754,
      "step": 6730
    },
    {
      "entropy_step_1": 0.2053,
      "entropy_step_2": 0.1996,
      "entropy_step_3": 0.2025,
      "epoch": 0.2004699443800006,
      "grad_norm": 93.41810607910156,
      "learning_rate": 9.394087518909205e-05,
      "loss": 0.06566618084907531,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0689,
      "step": 6740
    },
    {
      "entropy_step_1": 0.2005,
      "entropy_step_2": 0.1986,
      "entropy_step_3": 0.1737,
      "epoch": 0.20076737753189972,
      "grad_norm": 72.37875366210938,
      "learning_rate": 9.391738676136485e-05,
      "loss": 0.06484355926513671,
      "loss_step_1": 0.0723,
      "loss_step_2": 0.0714,
      "loss_step_3": 0.0539,
      "step": 6750
    },
    {
      "entropy_step_1": 0.2151,
      "entropy_step_2": 0.2181,
      "entropy_step_3": 0.2042,
      "epoch": 0.2010648106837988,
      "grad_norm": 57.659175872802734,
      "learning_rate": 9.389385584296838e-05,
      "loss": 0.06904300451278686,
      "loss_step_1": 0.0733,
      "loss_step_2": 0.0722,
      "loss_step_3": 0.0685,
      "step": 6760
    },
    {
      "entropy_step_1": 0.2311,
      "entropy_step_2": 0.2366,
      "entropy_step_3": 0.2483,
      "epoch": 0.20136224383569792,
      "grad_norm": 87.28399658203125,
      "learning_rate": 9.387028245666912e-05,
      "loss": 0.06968106031417846,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0737,
      "step": 6770
    },
    {
      "entropy_step_1": 0.2237,
      "entropy_step_2": 0.2295,
      "entropy_step_3": 0.2214,
      "epoch": 0.20165967698759704,
      "grad_norm": 32.20038986206055,
      "learning_rate": 9.384666662527464e-05,
      "loss": 0.058118730783462524,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0575,
      "step": 6780
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1923,
      "entropy_step_3": 0.1901,
      "epoch": 0.20195711013949616,
      "grad_norm": 149.3977813720703,
      "learning_rate": 9.382300837163359e-05,
      "loss": 0.0611992359161377,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.068,
      "step": 6790
    },
    {
      "entropy_step_1": 0.2327,
      "entropy_step_2": 0.2332,
      "entropy_step_3": 0.2265,
      "epoch": 0.20225454329139525,
      "grad_norm": 75.05015563964844,
      "learning_rate": 9.379930771863563e-05,
      "loss": 0.07380914092063903,
      "loss_step_1": 0.0794,
      "loss_step_2": 0.0793,
      "loss_step_3": 0.0632,
      "step": 6800
    },
    {
      "entropy_step_1": 0.2026,
      "entropy_step_2": 0.2089,
      "entropy_step_3": 0.222,
      "epoch": 0.20255197644329437,
      "grad_norm": 593.26904296875,
      "learning_rate": 9.377556468921149e-05,
      "loss": 0.06946783065795899,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0759,
      "step": 6810
    },
    {
      "entropy_step_1": 0.1968,
      "entropy_step_2": 0.2001,
      "entropy_step_3": 0.2058,
      "epoch": 0.2028494095951935,
      "grad_norm": 95.67847442626953,
      "learning_rate": 9.375177930633285e-05,
      "loss": 0.057048720121383664,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.058,
      "step": 6820
    },
    {
      "entropy_step_1": 0.2139,
      "entropy_step_2": 0.2177,
      "entropy_step_3": 0.2026,
      "epoch": 0.2031468427470926,
      "grad_norm": 69.13335418701172,
      "learning_rate": 9.37279515930124e-05,
      "loss": 0.06601886749267578,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0487,
      "step": 6830
    },
    {
      "entropy_step_1": 0.2016,
      "entropy_step_2": 0.2034,
      "entropy_step_3": 0.2132,
      "epoch": 0.2034442758989917,
      "grad_norm": 95.47032928466797,
      "learning_rate": 9.370408157230377e-05,
      "loss": 0.06950117349624634,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0595,
      "step": 6840
    },
    {
      "entropy_step_1": 0.2097,
      "entropy_step_2": 0.2063,
      "entropy_step_3": 0.2044,
      "epoch": 0.2037417090508908,
      "grad_norm": 36.21644973754883,
      "learning_rate": 9.368016926730156e-05,
      "loss": 0.06520209312438965,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0559,
      "step": 6850
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2045,
      "entropy_step_3": 0.1718,
      "epoch": 0.20403914220278993,
      "grad_norm": 89.46717834472656,
      "learning_rate": 9.365621470114122e-05,
      "loss": 0.06899858713150024,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0511,
      "step": 6860
    },
    {
      "entropy_step_1": 0.2222,
      "entropy_step_2": 0.2233,
      "entropy_step_3": 0.23,
      "epoch": 0.20433657535468902,
      "grad_norm": 111.42205810546875,
      "learning_rate": 9.363221789699912e-05,
      "loss": 0.07029114365577697,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0676,
      "step": 6870
    },
    {
      "entropy_step_1": 0.204,
      "entropy_step_2": 0.2054,
      "entropy_step_3": 0.2027,
      "epoch": 0.20463400850658814,
      "grad_norm": 112.18702697753906,
      "learning_rate": 9.360817887809252e-05,
      "loss": 0.07266149520874024,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0755,
      "step": 6880
    },
    {
      "entropy_step_1": 0.2096,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2222,
      "epoch": 0.20493144165848726,
      "grad_norm": 124.71671295166016,
      "learning_rate": 9.358409766767946e-05,
      "loss": 0.06543689966201782,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0622,
      "step": 6890
    },
    {
      "entropy_step_1": 0.2393,
      "entropy_step_2": 0.2421,
      "entropy_step_3": 0.2373,
      "epoch": 0.20522887481038637,
      "grad_norm": 59.45389938354492,
      "learning_rate": 9.355997428905887e-05,
      "loss": 0.07933424711227417,
      "loss_step_1": 0.0884,
      "loss_step_2": 0.0881,
      "loss_step_3": 0.0764,
      "step": 6900
    },
    {
      "entropy_step_1": 0.2265,
      "entropy_step_2": 0.2305,
      "entropy_step_3": 0.2426,
      "epoch": 0.20552630796228546,
      "grad_norm": 127.56971740722656,
      "learning_rate": 9.353580876557044e-05,
      "loss": 0.06936927437782288,
      "loss_step_1": 0.0732,
      "loss_step_2": 0.0723,
      "loss_step_3": 0.0707,
      "step": 6910
    },
    {
      "entropy_step_1": 0.2147,
      "entropy_step_2": 0.2201,
      "entropy_step_3": 0.2142,
      "epoch": 0.20582374111418458,
      "grad_norm": 36.40214157104492,
      "learning_rate": 9.351160112059463e-05,
      "loss": 0.07062640190124511,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.0712,
      "loss_step_3": 0.0777,
      "step": 6920
    },
    {
      "entropy_step_1": 0.2097,
      "entropy_step_2": 0.2094,
      "entropy_step_3": 0.2124,
      "epoch": 0.2061211742660837,
      "grad_norm": 116.89170837402344,
      "learning_rate": 9.348735137755269e-05,
      "loss": 0.0679658830165863,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.0733,
      "loss_step_3": 0.077,
      "step": 6930
    },
    {
      "entropy_step_1": 0.2031,
      "entropy_step_2": 0.2059,
      "entropy_step_3": 0.2031,
      "epoch": 0.20641860741798282,
      "grad_norm": 63.93560028076172,
      "learning_rate": 9.346305955990655e-05,
      "loss": 0.06324752569198608,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0538,
      "step": 6940
    },
    {
      "entropy_step_1": 0.2134,
      "entropy_step_2": 0.2198,
      "entropy_step_3": 0.2176,
      "epoch": 0.2067160405698819,
      "grad_norm": 61.35556411743164,
      "learning_rate": 9.343872569115888e-05,
      "loss": 0.06201986670494079,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0576,
      "step": 6950
    },
    {
      "entropy_step_1": 0.227,
      "entropy_step_2": 0.2221,
      "entropy_step_3": 0.2188,
      "epoch": 0.20701347372178103,
      "grad_norm": 210.376708984375,
      "learning_rate": 9.341434979485305e-05,
      "loss": 0.07546594738960266,
      "loss_step_1": 0.0726,
      "loss_step_2": 0.0709,
      "loss_step_3": 0.0843,
      "step": 6960
    },
    {
      "entropy_step_1": 0.2144,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.2024,
      "epoch": 0.20731090687368015,
      "grad_norm": 145.81179809570312,
      "learning_rate": 9.338993189457303e-05,
      "loss": 0.06640991568565369,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0694,
      "step": 6970
    },
    {
      "entropy_step_1": 0.2501,
      "entropy_step_2": 0.2199,
      "entropy_step_3": 0.2397,
      "epoch": 0.20760834002557926,
      "grad_norm": 51.25756072998047,
      "learning_rate": 9.336547201394353e-05,
      "loss": 0.07389606237411499,
      "loss_step_1": 0.068,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.0756,
      "step": 6980
    },
    {
      "entropy_step_1": 0.2241,
      "entropy_step_2": 0.1971,
      "entropy_step_3": 0.1972,
      "epoch": 0.20790577317747835,
      "grad_norm": 117.6095962524414,
      "learning_rate": 9.334097017662975e-05,
      "loss": 0.06880548596382141,
      "loss_step_1": 0.0735,
      "loss_step_2": 0.075,
      "loss_step_3": 0.0587,
      "step": 6990
    },
    {
      "entropy_step_1": 0.2325,
      "entropy_step_2": 0.2184,
      "entropy_step_3": 0.2257,
      "epoch": 0.20820320632937747,
      "grad_norm": 86.47283935546875,
      "learning_rate": 9.331642640633759e-05,
      "loss": 0.07386354207992554,
      "loss_step_1": 0.0785,
      "loss_step_2": 0.0785,
      "loss_step_3": 0.0815,
      "step": 7000
    },
    {
      "epoch": 0.20820320632937747,
      "eval_accuracy": 0.787,
      "eval_f1": 0.7821638189443842,
      "eval_loss": 0.1712484061717987,
      "eval_precision": 0.8149706539537047,
      "eval_recall": 0.787,
      "eval_runtime": 7.9498,
      "eval_samples_per_second": 62.894,
      "eval_steps_per_second": 7.925,
      "step": 7000
    },
    {
      "entropy_step_1": 0.2203,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.2191,
      "epoch": 0.2085006394812766,
      "grad_norm": 100.62167358398438,
      "learning_rate": 9.32918407268135e-05,
      "loss": 0.06714650392532348,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0744,
      "step": 7010
    },
    {
      "entropy_step_1": 0.2308,
      "entropy_step_2": 0.2344,
      "entropy_step_3": 0.2261,
      "epoch": 0.2087980726331757,
      "grad_norm": 69.75515747070312,
      "learning_rate": 9.326721316184442e-05,
      "loss": 0.0669415533542633,
      "loss_step_1": 0.0658,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0611,
      "step": 7020
    },
    {
      "entropy_step_1": 0.2227,
      "entropy_step_2": 0.2226,
      "entropy_step_3": 0.2239,
      "epoch": 0.2090955057850748,
      "grad_norm": 80.95523834228516,
      "learning_rate": 9.324254373525787e-05,
      "loss": 0.07338034510612487,
      "loss_step_1": 0.0721,
      "loss_step_2": 0.0717,
      "loss_step_3": 0.0702,
      "step": 7030
    },
    {
      "entropy_step_1": 0.2157,
      "entropy_step_2": 0.2232,
      "entropy_step_3": 0.2281,
      "epoch": 0.20939293893697392,
      "grad_norm": 81.83662414550781,
      "learning_rate": 9.321783247092187e-05,
      "loss": 0.07333041429519653,
      "loss_step_1": 0.0741,
      "loss_step_2": 0.0726,
      "loss_step_3": 0.0828,
      "step": 7040
    },
    {
      "entropy_step_1": 0.2246,
      "entropy_step_2": 0.239,
      "entropy_step_3": 0.2183,
      "epoch": 0.20969037208887303,
      "grad_norm": 92.56788635253906,
      "learning_rate": 9.319307939274487e-05,
      "loss": 0.06062267422676086,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0593,
      "step": 7050
    },
    {
      "entropy_step_1": 0.1881,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.2307,
      "epoch": 0.20998780524077212,
      "grad_norm": 137.2906036376953,
      "learning_rate": 9.316828452467585e-05,
      "loss": 0.06607145667076111,
      "loss_step_1": 0.0708,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0757,
      "step": 7060
    },
    {
      "entropy_step_1": 0.211,
      "entropy_step_2": 0.2333,
      "entropy_step_3": 0.1982,
      "epoch": 0.21028523839267124,
      "grad_norm": 172.29554748535156,
      "learning_rate": 9.314344789070415e-05,
      "loss": 0.06760268211364746,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0654,
      "loss_step_3": 0.0532,
      "step": 7070
    },
    {
      "entropy_step_1": 0.2053,
      "entropy_step_2": 0.2209,
      "entropy_step_3": 0.2187,
      "epoch": 0.21058267154457036,
      "grad_norm": 176.43148803710938,
      "learning_rate": 9.311856951485956e-05,
      "loss": 0.07442830204963684,
      "loss_step_1": 0.076,
      "loss_step_2": 0.0743,
      "loss_step_3": 0.0653,
      "step": 7080
    },
    {
      "entropy_step_1": 0.2252,
      "entropy_step_2": 0.2393,
      "entropy_step_3": 0.2452,
      "epoch": 0.21088010469646948,
      "grad_norm": 220.45516967773438,
      "learning_rate": 9.309364942121225e-05,
      "loss": 0.06750796437263488,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.066,
      "step": 7090
    },
    {
      "entropy_step_1": 0.2034,
      "entropy_step_2": 0.2129,
      "entropy_step_3": 0.2131,
      "epoch": 0.21117753784836857,
      "grad_norm": 97.50086975097656,
      "learning_rate": 9.306868763387277e-05,
      "loss": 0.06544797420501709,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0597,
      "step": 7100
    },
    {
      "entropy_step_1": 0.2069,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2373,
      "epoch": 0.2114749710002677,
      "grad_norm": 54.679603576660156,
      "learning_rate": 9.304368417699197e-05,
      "loss": 0.05781269073486328,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0827,
      "step": 7110
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.2156,
      "entropy_step_3": 0.2304,
      "epoch": 0.2117724041521668,
      "grad_norm": 52.687496185302734,
      "learning_rate": 9.3018639074761e-05,
      "loss": 0.06664073467254639,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0728,
      "step": 7120
    },
    {
      "entropy_step_1": 0.2449,
      "entropy_step_2": 0.2456,
      "entropy_step_3": 0.2507,
      "epoch": 0.21206983730406592,
      "grad_norm": 227.64622497558594,
      "learning_rate": 9.29935523514114e-05,
      "loss": 0.07129181623458862,
      "loss_step_1": 0.0745,
      "loss_step_2": 0.0741,
      "loss_step_3": 0.0781,
      "step": 7130
    },
    {
      "entropy_step_1": 0.2262,
      "entropy_step_2": 0.2281,
      "entropy_step_3": 0.2125,
      "epoch": 0.212367270455965,
      "grad_norm": 127.80057525634766,
      "learning_rate": 9.29684240312149e-05,
      "loss": 0.06359598636627198,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.068,
      "loss_step_3": 0.054,
      "step": 7140
    },
    {
      "entropy_step_1": 0.2086,
      "entropy_step_2": 0.2289,
      "entropy_step_3": 0.214,
      "epoch": 0.21266470360786413,
      "grad_norm": 51.73659896850586,
      "learning_rate": 9.294325413848346e-05,
      "loss": 0.07548996210098266,
      "loss_step_1": 0.0731,
      "loss_step_2": 0.071,
      "loss_step_3": 0.0676,
      "step": 7150
    },
    {
      "entropy_step_1": 0.2302,
      "entropy_step_2": 0.2375,
      "entropy_step_3": 0.2569,
      "epoch": 0.21296213675976325,
      "grad_norm": 95.93714141845703,
      "learning_rate": 9.291804269756932e-05,
      "loss": 0.07564598917961121,
      "loss_step_1": 0.0807,
      "loss_step_2": 0.0785,
      "loss_step_3": 0.0725,
      "step": 7160
    },
    {
      "entropy_step_1": 0.2349,
      "entropy_step_2": 0.2437,
      "entropy_step_3": 0.2484,
      "epoch": 0.21325956991166237,
      "grad_norm": 35.10371780395508,
      "learning_rate": 9.289278973286488e-05,
      "loss": 0.06581624150276184,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.07,
      "loss_step_3": 0.0754,
      "step": 7170
    },
    {
      "entropy_step_1": 0.1936,
      "entropy_step_2": 0.2069,
      "entropy_step_3": 0.2123,
      "epoch": 0.21355700306356146,
      "grad_norm": 123.65418243408203,
      "learning_rate": 9.286749526880273e-05,
      "loss": 0.06669719219207763,
      "loss_step_1": 0.0746,
      "loss_step_2": 0.0726,
      "loss_step_3": 0.074,
      "step": 7180
    },
    {
      "entropy_step_1": 0.216,
      "entropy_step_2": 0.2195,
      "entropy_step_3": 0.2009,
      "epoch": 0.21385443621546057,
      "grad_norm": 53.42976379394531,
      "learning_rate": 9.284215932985561e-05,
      "loss": 0.0607768714427948,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0454,
      "step": 7190
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2132,
      "entropy_step_3": 0.2167,
      "epoch": 0.2141518693673597,
      "grad_norm": 59.38926315307617,
      "learning_rate": 9.28167819405364e-05,
      "loss": 0.0729419469833374,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0718,
      "step": 7200
    },
    {
      "entropy_step_1": 0.2397,
      "entropy_step_2": 0.2494,
      "entropy_step_3": 0.273,
      "epoch": 0.21444930251925878,
      "grad_norm": 180.83798217773438,
      "learning_rate": 9.279136312539805e-05,
      "loss": 0.06501609683036805,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0722,
      "step": 7210
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.1921,
      "entropy_step_3": 0.2091,
      "epoch": 0.2147467356711579,
      "grad_norm": 93.47257995605469,
      "learning_rate": 9.276590290903362e-05,
      "loss": 0.056511479616165164,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.061,
      "step": 7220
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.207,
      "entropy_step_3": 0.1914,
      "epoch": 0.21504416882305702,
      "grad_norm": 27.583520889282227,
      "learning_rate": 9.274040131607619e-05,
      "loss": 0.06329840421676636,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.0628,
      "step": 7230
    },
    {
      "entropy_step_1": 0.2019,
      "entropy_step_2": 0.2027,
      "entropy_step_3": 0.1945,
      "epoch": 0.21534160197495614,
      "grad_norm": 64.9977798461914,
      "learning_rate": 9.271485837119893e-05,
      "loss": 0.05808520913124084,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0612,
      "step": 7240
    },
    {
      "entropy_step_1": 0.217,
      "entropy_step_2": 0.2146,
      "entropy_step_3": 0.2327,
      "epoch": 0.21563903512685523,
      "grad_norm": 114.96700286865234,
      "learning_rate": 9.268927409911498e-05,
      "loss": 0.06005658507347107,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0638,
      "step": 7250
    },
    {
      "entropy_step_1": 0.1819,
      "entropy_step_2": 0.1768,
      "entropy_step_3": 0.1728,
      "epoch": 0.21593646827875435,
      "grad_norm": 41.08003234863281,
      "learning_rate": 9.266364852457747e-05,
      "loss": 0.0716456115245819,
      "loss_step_1": 0.0772,
      "loss_step_2": 0.0773,
      "loss_step_3": 0.0898,
      "step": 7260
    },
    {
      "entropy_step_1": 0.2416,
      "entropy_step_2": 0.2305,
      "entropy_step_3": 0.2342,
      "epoch": 0.21623390143065346,
      "grad_norm": 285.49383544921875,
      "learning_rate": 9.26379816723795e-05,
      "loss": 0.06904637813568115,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0631,
      "step": 7270
    },
    {
      "entropy_step_1": 0.1963,
      "entropy_step_2": 0.1947,
      "entropy_step_3": 0.1943,
      "epoch": 0.21653133458255258,
      "grad_norm": 99.65970611572266,
      "learning_rate": 9.261227356735407e-05,
      "loss": 0.07285265326499939,
      "loss_step_1": 0.0778,
      "loss_step_2": 0.0766,
      "loss_step_3": 0.091,
      "step": 7280
    },
    {
      "entropy_step_1": 0.2374,
      "entropy_step_2": 0.2462,
      "entropy_step_3": 0.2378,
      "epoch": 0.21682876773445167,
      "grad_norm": 139.28457641601562,
      "learning_rate": 9.258652423437416e-05,
      "loss": 0.06588862538337707,
      "loss_step_1": 0.0717,
      "loss_step_2": 0.0719,
      "loss_step_3": 0.0715,
      "step": 7290
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.2049,
      "entropy_step_3": 0.1879,
      "epoch": 0.2171262008863508,
      "grad_norm": 63.120296478271484,
      "learning_rate": 9.256073369835254e-05,
      "loss": 0.06546322107315064,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.051,
      "step": 7300
    },
    {
      "entropy_step_1": 0.2372,
      "entropy_step_2": 0.2497,
      "entropy_step_3": 0.2484,
      "epoch": 0.2174236340382499,
      "grad_norm": 78.01323699951172,
      "learning_rate": 9.253490198424197e-05,
      "loss": 0.07134145498275757,
      "loss_step_1": 0.072,
      "loss_step_2": 0.071,
      "loss_step_3": 0.0638,
      "step": 7310
    },
    {
      "entropy_step_1": 0.214,
      "entropy_step_2": 0.223,
      "entropy_step_3": 0.232,
      "epoch": 0.21772106719014903,
      "grad_norm": 125.93977355957031,
      "learning_rate": 9.250902911703496e-05,
      "loss": 0.07156926393508911,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0636,
      "step": 7320
    },
    {
      "entropy_step_1": 0.232,
      "entropy_step_2": 0.2337,
      "entropy_step_3": 0.236,
      "epoch": 0.21801850034204812,
      "grad_norm": 79.39398193359375,
      "learning_rate": 9.248311512176386e-05,
      "loss": 0.06957606673240661,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0751,
      "step": 7330
    },
    {
      "entropy_step_1": 0.2009,
      "entropy_step_2": 0.1988,
      "entropy_step_3": 0.2113,
      "epoch": 0.21831593349394723,
      "grad_norm": 103.87754821777344,
      "learning_rate": 9.245716002350081e-05,
      "loss": 0.06900240182876587,
      "loss_step_1": 0.0752,
      "loss_step_2": 0.0729,
      "loss_step_3": 0.0778,
      "step": 7340
    },
    {
      "entropy_step_1": 0.2121,
      "entropy_step_2": 0.2163,
      "entropy_step_3": 0.2271,
      "epoch": 0.21861336664584635,
      "grad_norm": 72.35258483886719,
      "learning_rate": 9.243116384735773e-05,
      "loss": 0.0636155664920807,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0674,
      "step": 7350
    },
    {
      "entropy_step_1": 0.23,
      "entropy_step_2": 0.2417,
      "entropy_step_3": 0.2409,
      "epoch": 0.21891079979774547,
      "grad_norm": 89.30628204345703,
      "learning_rate": 9.240512661848627e-05,
      "loss": 0.06876441836357117,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0576,
      "step": 7360
    },
    {
      "entropy_step_1": 0.2237,
      "entropy_step_2": 0.2368,
      "entropy_step_3": 0.2003,
      "epoch": 0.21920823294964456,
      "grad_norm": 128.29522705078125,
      "learning_rate": 9.237904836207777e-05,
      "loss": 0.07122105956077576,
      "loss_step_1": 0.0754,
      "loss_step_2": 0.0732,
      "loss_step_3": 0.058,
      "step": 7370
    },
    {
      "entropy_step_1": 0.2116,
      "entropy_step_2": 0.2221,
      "entropy_step_3": 0.2517,
      "epoch": 0.21950566610154368,
      "grad_norm": 256.2572937011719,
      "learning_rate": 9.235292910336338e-05,
      "loss": 0.06040353178977966,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0677,
      "step": 7380
    },
    {
      "entropy_step_1": 0.2102,
      "entropy_step_2": 0.2163,
      "entropy_step_3": 0.2253,
      "epoch": 0.2198030992534428,
      "grad_norm": 188.35015869140625,
      "learning_rate": 9.232676886761375e-05,
      "loss": 0.06370631456375123,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0578,
      "step": 7390
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2061,
      "entropy_step_3": 0.2014,
      "epoch": 0.2201005324053419,
      "grad_norm": 83.70665740966797,
      "learning_rate": 9.23005676801393e-05,
      "loss": 0.06542852520942688,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0606,
      "step": 7400
    },
    {
      "entropy_step_1": 0.2275,
      "entropy_step_2": 0.223,
      "entropy_step_3": 0.2208,
      "epoch": 0.220397965557241,
      "grad_norm": 79.28530883789062,
      "learning_rate": 9.227432556629005e-05,
      "loss": 0.06677904725074768,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0595,
      "step": 7410
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.203,
      "entropy_step_3": 0.235,
      "epoch": 0.22069539870914012,
      "grad_norm": 77.17556762695312,
      "learning_rate": 9.22480425514556e-05,
      "loss": 0.06326159834861755,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.078,
      "step": 7420
    },
    {
      "entropy_step_1": 0.2266,
      "entropy_step_2": 0.2223,
      "entropy_step_3": 0.2382,
      "epoch": 0.22099283186103924,
      "grad_norm": 156.35107421875,
      "learning_rate": 9.22217186610651e-05,
      "loss": 0.06606478691101074,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0645,
      "step": 7430
    },
    {
      "entropy_step_1": 0.2274,
      "entropy_step_2": 0.225,
      "entropy_step_3": 0.2187,
      "epoch": 0.22129026501293833,
      "grad_norm": 403.597412109375,
      "learning_rate": 9.219535392058729e-05,
      "loss": 0.06654639840126038,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0652,
      "loss_step_3": 0.0608,
      "step": 7440
    },
    {
      "entropy_step_1": 0.208,
      "entropy_step_2": 0.205,
      "entropy_step_3": 0.21,
      "epoch": 0.22158769816483745,
      "grad_norm": 177.63861083984375,
      "learning_rate": 9.21689483555304e-05,
      "loss": 0.06157998442649841,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0556,
      "step": 7450
    },
    {
      "entropy_step_1": 0.232,
      "entropy_step_2": 0.2259,
      "entropy_step_3": 0.2189,
      "epoch": 0.22188513131673657,
      "grad_norm": 67.70587921142578,
      "learning_rate": 9.214250199144222e-05,
      "loss": 0.06631827354431152,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0583,
      "step": 7460
    },
    {
      "entropy_step_1": 0.2122,
      "entropy_step_2": 0.2059,
      "entropy_step_3": 0.2045,
      "epoch": 0.22218256446863568,
      "grad_norm": 57.387996673583984,
      "learning_rate": 9.211601485390989e-05,
      "loss": 0.06629712581634521,
      "loss_step_1": 0.0723,
      "loss_step_2": 0.0719,
      "loss_step_3": 0.0692,
      "step": 7470
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.1787,
      "entropy_step_3": 0.1899,
      "epoch": 0.22247999762053478,
      "grad_norm": 143.47653198242188,
      "learning_rate": 9.208948696856013e-05,
      "loss": 0.0583669900894165,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0678,
      "step": 7480
    },
    {
      "entropy_step_1": 0.2032,
      "entropy_step_2": 0.2033,
      "entropy_step_3": 0.2106,
      "epoch": 0.2227774307724339,
      "grad_norm": 100.39441680908203,
      "learning_rate": 9.206291836105902e-05,
      "loss": 0.06438320875167847,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0639,
      "step": 7490
    },
    {
      "entropy_step_1": 0.2548,
      "entropy_step_2": 0.2548,
      "entropy_step_3": 0.2735,
      "epoch": 0.223074863924333,
      "grad_norm": 122.77210998535156,
      "learning_rate": 9.203630905711205e-05,
      "loss": 0.06622755527496338,
      "loss_step_1": 0.0706,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0718,
      "step": 7500
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.1748,
      "epoch": 0.22337229707623213,
      "grad_norm": 151.29237365722656,
      "learning_rate": 9.200965908246407e-05,
      "loss": 0.06349515318870544,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0445,
      "step": 7510
    },
    {
      "entropy_step_1": 0.2184,
      "entropy_step_2": 0.2073,
      "entropy_step_3": 0.2095,
      "epoch": 0.22366973022813122,
      "grad_norm": 85.17562103271484,
      "learning_rate": 9.198296846289931e-05,
      "loss": 0.06599915623664857,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0654,
      "loss_step_3": 0.0617,
      "step": 7520
    },
    {
      "entropy_step_1": 0.2328,
      "entropy_step_2": 0.22,
      "entropy_step_3": 0.2159,
      "epoch": 0.22396716338003034,
      "grad_norm": 75.65653991699219,
      "learning_rate": 9.195623722424131e-05,
      "loss": 0.06411052942276001,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0584,
      "step": 7530
    },
    {
      "entropy_step_1": 0.2032,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.1811,
      "epoch": 0.22426459653192946,
      "grad_norm": 332.46588134765625,
      "learning_rate": 9.192946539235289e-05,
      "loss": 0.07069405317306518,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0576,
      "step": 7540
    },
    {
      "entropy_step_1": 0.2221,
      "entropy_step_2": 0.215,
      "entropy_step_3": 0.2003,
      "epoch": 0.22456202968382855,
      "grad_norm": 65.75765991210938,
      "learning_rate": 9.190265299313617e-05,
      "loss": 0.06622949838638306,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.0726,
      "loss_step_3": 0.054,
      "step": 7550
    },
    {
      "entropy_step_1": 0.2228,
      "entropy_step_2": 0.2234,
      "entropy_step_3": 0.2253,
      "epoch": 0.22485946283572766,
      "grad_norm": 206.6014862060547,
      "learning_rate": 9.187580005253251e-05,
      "loss": 0.06396301984786987,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0661,
      "step": 7560
    },
    {
      "entropy_step_1": 0.2233,
      "entropy_step_2": 0.2273,
      "entropy_step_3": 0.231,
      "epoch": 0.22515689598762678,
      "grad_norm": 107.84211730957031,
      "learning_rate": 9.184890659652248e-05,
      "loss": 0.07425774335861206,
      "loss_step_1": 0.0747,
      "loss_step_2": 0.0728,
      "loss_step_3": 0.0808,
      "step": 7570
    },
    {
      "entropy_step_1": 0.2088,
      "entropy_step_2": 0.2133,
      "entropy_step_3": 0.2205,
      "epoch": 0.2254543291395259,
      "grad_norm": 91.46566009521484,
      "learning_rate": 9.182197265112592e-05,
      "loss": 0.06614927649497986,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0578,
      "step": 7580
    },
    {
      "entropy_step_1": 0.2162,
      "entropy_step_2": 0.2147,
      "entropy_step_3": 0.2382,
      "epoch": 0.225751762291425,
      "grad_norm": 71.56591796875,
      "learning_rate": 9.179499824240173e-05,
      "loss": 0.06848678588867188,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0768,
      "step": 7590
    },
    {
      "entropy_step_1": 0.2299,
      "entropy_step_2": 0.2184,
      "entropy_step_3": 0.2051,
      "epoch": 0.2260491954433241,
      "grad_norm": 35.85456848144531,
      "learning_rate": 9.176798339644803e-05,
      "loss": 0.057939308881759646,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0561,
      "step": 7600
    },
    {
      "entropy_step_1": 0.2396,
      "entropy_step_2": 0.2333,
      "entropy_step_3": 0.247,
      "epoch": 0.22634662859522323,
      "grad_norm": 153.58026123046875,
      "learning_rate": 9.174092813940206e-05,
      "loss": 0.0693858504295349,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0702,
      "loss_step_3": 0.0803,
      "step": 7610
    },
    {
      "entropy_step_1": 0.2028,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.2222,
      "epoch": 0.22664406174712234,
      "grad_norm": 93.82486724853516,
      "learning_rate": 9.171383249744015e-05,
      "loss": 0.06733027696609498,
      "loss_step_1": 0.0714,
      "loss_step_2": 0.0698,
      "loss_step_3": 0.0659,
      "step": 7620
    },
    {
      "entropy_step_1": 0.1928,
      "entropy_step_2": 0.1996,
      "entropy_step_3": 0.2153,
      "epoch": 0.22694149489902143,
      "grad_norm": 160.62417602539062,
      "learning_rate": 9.168669649677771e-05,
      "loss": 0.06407296061515808,
      "loss_step_1": 0.0619,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0618,
      "step": 7630
    },
    {
      "entropy_step_1": 0.203,
      "entropy_step_2": 0.2003,
      "entropy_step_3": 0.1859,
      "epoch": 0.22723892805092055,
      "grad_norm": 103.29144287109375,
      "learning_rate": 9.165952016366916e-05,
      "loss": 0.05719894766807556,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.043,
      "step": 7640
    },
    {
      "entropy_step_1": 0.2177,
      "entropy_step_2": 0.2146,
      "entropy_step_3": 0.2256,
      "epoch": 0.22753636120281967,
      "grad_norm": 41.830684661865234,
      "learning_rate": 9.163230352440799e-05,
      "loss": 0.06180111169815063,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.068,
      "step": 7650
    },
    {
      "entropy_step_1": 0.22,
      "entropy_step_2": 0.2225,
      "entropy_step_3": 0.2316,
      "epoch": 0.2278337943547188,
      "grad_norm": 83.35968017578125,
      "learning_rate": 9.16050466053267e-05,
      "loss": 0.07082626819610596,
      "loss_step_1": 0.0718,
      "loss_step_2": 0.0707,
      "loss_step_3": 0.0705,
      "step": 7660
    },
    {
      "entropy_step_1": 0.2177,
      "entropy_step_2": 0.2215,
      "entropy_step_3": 0.2058,
      "epoch": 0.22813122750661788,
      "grad_norm": 40.72505187988281,
      "learning_rate": 9.157774943279667e-05,
      "loss": 0.0622658908367157,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0566,
      "step": 7670
    },
    {
      "entropy_step_1": 0.2203,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2132,
      "epoch": 0.228428660658517,
      "grad_norm": 43.29966735839844,
      "learning_rate": 9.155041203322834e-05,
      "loss": 0.07011635899543762,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0675,
      "step": 7680
    },
    {
      "entropy_step_1": 0.2162,
      "entropy_step_2": 0.2081,
      "entropy_step_3": 0.2046,
      "epoch": 0.22872609381041611,
      "grad_norm": 136.21620178222656,
      "learning_rate": 9.152303443307098e-05,
      "loss": 0.06630193591117858,
      "loss_step_1": 0.0622,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0608,
      "step": 7690
    },
    {
      "entropy_step_1": 0.2211,
      "entropy_step_2": 0.2148,
      "entropy_step_3": 0.2048,
      "epoch": 0.22902352696231523,
      "grad_norm": 71.95145416259766,
      "learning_rate": 9.149561665881282e-05,
      "loss": 0.06022083163261414,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0571,
      "step": 7700
    },
    {
      "entropy_step_1": 0.1705,
      "entropy_step_2": 0.1716,
      "entropy_step_3": 0.1714,
      "epoch": 0.22932096011421432,
      "grad_norm": 60.18083572387695,
      "learning_rate": 9.146815873698089e-05,
      "loss": 0.05949353575706482,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0538,
      "step": 7710
    },
    {
      "entropy_step_1": 0.2177,
      "entropy_step_2": 0.2142,
      "entropy_step_3": 0.2187,
      "epoch": 0.22961839326611344,
      "grad_norm": 78.26529693603516,
      "learning_rate": 9.144066069414113e-05,
      "loss": 0.05942248106002808,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0569,
      "step": 7720
    },
    {
      "entropy_step_1": 0.2285,
      "entropy_step_2": 0.2191,
      "entropy_step_3": 0.2353,
      "epoch": 0.22991582641801256,
      "grad_norm": 54.488712310791016,
      "learning_rate": 9.141312255689826e-05,
      "loss": 0.06590477228164673,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0676,
      "step": 7730
    },
    {
      "entropy_step_1": 0.2085,
      "entropy_step_2": 0.209,
      "entropy_step_3": 0.2163,
      "epoch": 0.23021325956991165,
      "grad_norm": 72.27156066894531,
      "learning_rate": 9.138554435189584e-05,
      "loss": 0.0689136266708374,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0711,
      "loss_step_3": 0.0606,
      "step": 7740
    },
    {
      "entropy_step_1": 0.2202,
      "entropy_step_2": 0.2266,
      "entropy_step_3": 0.2296,
      "epoch": 0.23051069272181077,
      "grad_norm": 118.47209167480469,
      "learning_rate": 9.13579261058161e-05,
      "loss": 0.06057863831520081,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0643,
      "step": 7750
    },
    {
      "entropy_step_1": 0.196,
      "entropy_step_2": 0.2089,
      "entropy_step_3": 0.218,
      "epoch": 0.23080812587370989,
      "grad_norm": 218.04754638671875,
      "learning_rate": 9.13302678453801e-05,
      "loss": 0.06316519379615784,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0566,
      "step": 7760
    },
    {
      "entropy_step_1": 0.2195,
      "entropy_step_2": 0.2328,
      "entropy_step_3": 0.2225,
      "epoch": 0.231105559025609,
      "grad_norm": 100.92310333251953,
      "learning_rate": 9.13025695973476e-05,
      "loss": 0.06749430894851685,
      "loss_step_1": 0.0693,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.062,
      "step": 7770
    },
    {
      "entropy_step_1": 0.2379,
      "entropy_step_2": 0.2419,
      "entropy_step_3": 0.2364,
      "epoch": 0.2314029921775081,
      "grad_norm": 76.20889282226562,
      "learning_rate": 9.127483138851698e-05,
      "loss": 0.06811223030090333,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.0705,
      "step": 7780
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1928,
      "entropy_step_3": 0.1743,
      "epoch": 0.2317004253294072,
      "grad_norm": 92.41940307617188,
      "learning_rate": 9.124705324572539e-05,
      "loss": 0.08293372392654419,
      "loss_step_1": 0.0902,
      "loss_step_2": 0.0827,
      "loss_step_3": 0.0804,
      "step": 7790
    },
    {
      "entropy_step_1": 0.2412,
      "entropy_step_2": 0.2279,
      "entropy_step_3": 0.2148,
      "epoch": 0.23199785848130633,
      "grad_norm": 74.68539428710938,
      "learning_rate": 9.121923519584852e-05,
      "loss": 0.06965017318725586,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.0563,
      "step": 7800
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.1758,
      "epoch": 0.23229529163320545,
      "grad_norm": 587.6491088867188,
      "learning_rate": 9.119137726580075e-05,
      "loss": 0.056436294317245485,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0458,
      "step": 7810
    },
    {
      "entropy_step_1": 0.2039,
      "entropy_step_2": 0.208,
      "entropy_step_3": 0.2088,
      "epoch": 0.23259272478510454,
      "grad_norm": 288.078857421875,
      "learning_rate": 9.116347948253497e-05,
      "loss": 0.06815359592437745,
      "loss_step_1": 0.0753,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.0673,
      "step": 7820
    },
    {
      "entropy_step_1": 0.2224,
      "entropy_step_2": 0.2262,
      "entropy_step_3": 0.2114,
      "epoch": 0.23289015793700366,
      "grad_norm": 41.04722595214844,
      "learning_rate": 9.11355418730427e-05,
      "loss": 0.07055039405822754,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0665,
      "step": 7830
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.2063,
      "entropy_step_3": 0.2041,
      "epoch": 0.23318759108890277,
      "grad_norm": 38.02643966674805,
      "learning_rate": 9.110756446435391e-05,
      "loss": 0.05942562222480774,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0562,
      "step": 7840
    },
    {
      "entropy_step_1": 0.2305,
      "entropy_step_2": 0.2232,
      "entropy_step_3": 0.2313,
      "epoch": 0.2334850242408019,
      "grad_norm": 107.94577026367188,
      "learning_rate": 9.107954728353716e-05,
      "loss": 0.06891375184059143,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0676,
      "loss_step_3": 0.0769,
      "step": 7850
    },
    {
      "entropy_step_1": 0.1929,
      "entropy_step_2": 0.1826,
      "entropy_step_3": 0.1945,
      "epoch": 0.23378245739270098,
      "grad_norm": 138.48590087890625,
      "learning_rate": 9.105149035769945e-05,
      "loss": 0.06123780608177185,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0557,
      "step": 7860
    },
    {
      "entropy_step_1": 0.2,
      "entropy_step_2": 0.1844,
      "entropy_step_3": 0.1863,
      "epoch": 0.2340798905446001,
      "grad_norm": 192.7019805908203,
      "learning_rate": 9.102339371398624e-05,
      "loss": 0.06207252740859985,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0486,
      "step": 7870
    },
    {
      "entropy_step_1": 0.2339,
      "entropy_step_2": 0.2116,
      "entropy_step_3": 0.2191,
      "epoch": 0.23437732369649922,
      "grad_norm": 603.9229125976562,
      "learning_rate": 9.099525737958138e-05,
      "loss": 0.06946084499359131,
      "loss_step_1": 0.0692,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0625,
      "step": 7880
    },
    {
      "entropy_step_1": 0.2137,
      "entropy_step_2": 0.1991,
      "entropy_step_3": 0.2096,
      "epoch": 0.23467475684839834,
      "grad_norm": 59.1090202331543,
      "learning_rate": 9.096708138170723e-05,
      "loss": 0.06295551061630249,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0651,
      "loss_step_3": 0.0676,
      "step": 7890
    },
    {
      "entropy_step_1": 0.1997,
      "entropy_step_2": 0.2008,
      "entropy_step_3": 0.2049,
      "epoch": 0.23497219000029743,
      "grad_norm": 144.68275451660156,
      "learning_rate": 9.093886574762436e-05,
      "loss": 0.06783191561698913,
      "loss_step_1": 0.0666,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0614,
      "step": 7900
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2063,
      "entropy_step_3": 0.2059,
      "epoch": 0.23526962315219654,
      "grad_norm": 71.75930786132812,
      "learning_rate": 9.091061050463186e-05,
      "loss": 0.06423907279968262,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0504,
      "step": 7910
    },
    {
      "entropy_step_1": 0.2066,
      "entropy_step_2": 0.2012,
      "entropy_step_3": 0.2008,
      "epoch": 0.23556705630409566,
      "grad_norm": 68.66214752197266,
      "learning_rate": 9.0882315680067e-05,
      "loss": 0.058623498678207396,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0599,
      "step": 7920
    },
    {
      "entropy_step_1": 0.2061,
      "entropy_step_2": 0.1961,
      "entropy_step_3": 0.1861,
      "epoch": 0.23586448945599475,
      "grad_norm": 226.1669921875,
      "learning_rate": 9.085398130130545e-05,
      "loss": 0.06737584471702576,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.056,
      "step": 7930
    },
    {
      "entropy_step_1": 0.1769,
      "entropy_step_2": 0.1762,
      "entropy_step_3": 0.1756,
      "epoch": 0.23616192260789387,
      "grad_norm": 64.59709930419922,
      "learning_rate": 9.082560739576109e-05,
      "loss": 0.06169006824493408,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0689,
      "step": 7940
    },
    {
      "entropy_step_1": 0.2073,
      "entropy_step_2": 0.2045,
      "entropy_step_3": 0.2161,
      "epoch": 0.236459355759793,
      "grad_norm": 25.30452537536621,
      "learning_rate": 9.079719399088606e-05,
      "loss": 0.07014750242233277,
      "loss_step_1": 0.0711,
      "loss_step_2": 0.0708,
      "loss_step_3": 0.0883,
      "step": 7950
    },
    {
      "entropy_step_1": 0.1979,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2021,
      "epoch": 0.2367567889116921,
      "grad_norm": 78.2759017944336,
      "learning_rate": 9.07687411141707e-05,
      "loss": 0.0652387022972107,
      "loss_step_1": 0.061,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0617,
      "step": 7960
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.2033,
      "entropy_step_3": 0.2175,
      "epoch": 0.2370542220635912,
      "grad_norm": 46.279109954833984,
      "learning_rate": 9.07402487931436e-05,
      "loss": 0.06280825734138488,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0673,
      "step": 7970
    },
    {
      "entropy_step_1": 0.231,
      "entropy_step_2": 0.2346,
      "entropy_step_3": 0.2334,
      "epoch": 0.23735165521549031,
      "grad_norm": 143.51316833496094,
      "learning_rate": 9.071171705537144e-05,
      "loss": 0.05892901420593262,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0608,
      "step": 7980
    },
    {
      "entropy_step_1": 0.2178,
      "entropy_step_2": 0.2185,
      "entropy_step_3": 0.2122,
      "epoch": 0.23764908836738943,
      "grad_norm": 92.74910736083984,
      "learning_rate": 9.068314592845905e-05,
      "loss": 0.07369182109832764,
      "loss_step_1": 0.0693,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0578,
      "step": 7990
    },
    {
      "entropy_step_1": 0.2522,
      "entropy_step_2": 0.2523,
      "entropy_step_3": 0.2518,
      "epoch": 0.23794652151928855,
      "grad_norm": 85.5526351928711,
      "learning_rate": 9.06545354400494e-05,
      "loss": 0.07310523390769959,
      "loss_step_1": 0.0738,
      "loss_step_2": 0.0736,
      "loss_step_3": 0.0697,
      "step": 8000
    },
    {
      "epoch": 0.23794652151928855,
      "eval_accuracy": 0.851,
      "eval_f1": 0.850921137281622,
      "eval_loss": 0.15751592814922333,
      "eval_precision": 0.8517442909195858,
      "eval_recall": 0.851,
      "eval_runtime": 6.6785,
      "eval_samples_per_second": 74.867,
      "eval_steps_per_second": 9.433,
      "step": 8000
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.1963,
      "entropy_step_3": 0.1909,
      "epoch": 0.23824395467118764,
      "grad_norm": 129.5052490234375,
      "learning_rate": 9.062588561782355e-05,
      "loss": 0.06812021732330323,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.061,
      "step": 8010
    },
    {
      "entropy_step_1": 0.2314,
      "entropy_step_2": 0.2329,
      "entropy_step_3": 0.2269,
      "epoch": 0.23854138782308676,
      "grad_norm": 48.13874435424805,
      "learning_rate": 9.059719648950055e-05,
      "loss": 0.06874669790267944,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.067,
      "loss_step_3": 0.065,
      "step": 8020
    },
    {
      "entropy_step_1": 0.2068,
      "entropy_step_2": 0.2026,
      "entropy_step_3": 0.1872,
      "epoch": 0.23883882097498588,
      "grad_norm": 45.548797607421875,
      "learning_rate": 9.056846808283754e-05,
      "loss": 0.0606534481048584,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0566,
      "step": 8030
    },
    {
      "entropy_step_1": 0.2489,
      "entropy_step_2": 0.234,
      "entropy_step_3": 0.2309,
      "epoch": 0.239136254126885,
      "grad_norm": 104.3013687133789,
      "learning_rate": 9.053970042562965e-05,
      "loss": 0.06466519236564636,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0666,
      "step": 8040
    },
    {
      "entropy_step_1": 0.1883,
      "entropy_step_2": 0.1943,
      "entropy_step_3": 0.2084,
      "epoch": 0.23943368727878409,
      "grad_norm": 47.20214080810547,
      "learning_rate": 9.051089354570998e-05,
      "loss": 0.06098911762237549,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0609,
      "step": 8050
    },
    {
      "entropy_step_1": 0.2377,
      "entropy_step_2": 0.2499,
      "entropy_step_3": 0.261,
      "epoch": 0.2397311204306832,
      "grad_norm": 129.29075622558594,
      "learning_rate": 9.048204747094957e-05,
      "loss": 0.06667697429656982,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0573,
      "step": 8060
    },
    {
      "entropy_step_1": 0.2069,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.2221,
      "epoch": 0.24002855358258232,
      "grad_norm": 52.66033172607422,
      "learning_rate": 9.04531622292574e-05,
      "loss": 0.0681131899356842,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.072,
      "step": 8070
    },
    {
      "entropy_step_1": 0.2117,
      "entropy_step_2": 0.2151,
      "entropy_step_3": 0.2282,
      "epoch": 0.2403259867344814,
      "grad_norm": 120.26878356933594,
      "learning_rate": 9.042423784858032e-05,
      "loss": 0.06259663105010986,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0701,
      "step": 8080
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.2051,
      "entropy_step_3": 0.2033,
      "epoch": 0.24062341988638053,
      "grad_norm": 39.56063461303711,
      "learning_rate": 9.039527435690307e-05,
      "loss": 0.05541638731956482,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0503,
      "step": 8090
    },
    {
      "entropy_step_1": 0.2061,
      "entropy_step_2": 0.2209,
      "entropy_step_3": 0.2305,
      "epoch": 0.24092085303827965,
      "grad_norm": 44.84477615356445,
      "learning_rate": 9.036627178224823e-05,
      "loss": 0.06260948777198791,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0723,
      "step": 8100
    },
    {
      "entropy_step_1": 0.182,
      "entropy_step_2": 0.1904,
      "entropy_step_3": 0.2098,
      "epoch": 0.24121828619017877,
      "grad_norm": 70.12020111083984,
      "learning_rate": 9.033723015267617e-05,
      "loss": 0.055849528312683104,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0651,
      "step": 8110
    },
    {
      "entropy_step_1": 0.2147,
      "entropy_step_2": 0.2216,
      "entropy_step_3": 0.2347,
      "epoch": 0.24151571934207786,
      "grad_norm": 31.811729431152344,
      "learning_rate": 9.030814949628508e-05,
      "loss": 0.06464970111846924,
      "loss_step_1": 0.071,
      "loss_step_2": 0.0704,
      "loss_step_3": 0.0716,
      "step": 8120
    },
    {
      "entropy_step_1": 0.2088,
      "entropy_step_2": 0.219,
      "entropy_step_3": 0.2302,
      "epoch": 0.24181315249397697,
      "grad_norm": 46.766414642333984,
      "learning_rate": 9.027902984121088e-05,
      "loss": 0.06171149015426636,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0731,
      "step": 8130
    },
    {
      "entropy_step_1": 0.2106,
      "entropy_step_2": 0.2212,
      "entropy_step_3": 0.2335,
      "epoch": 0.2421105856458761,
      "grad_norm": 105.47088623046875,
      "learning_rate": 9.024987121562722e-05,
      "loss": 0.05835903286933899,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0618,
      "step": 8140
    },
    {
      "entropy_step_1": 0.1982,
      "entropy_step_2": 0.2167,
      "entropy_step_3": 0.2294,
      "epoch": 0.2424080187977752,
      "grad_norm": 52.09465789794922,
      "learning_rate": 9.02206736477455e-05,
      "loss": 0.06269367337226868,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0775,
      "step": 8150
    },
    {
      "entropy_step_1": 0.2163,
      "entropy_step_2": 0.2322,
      "entropy_step_3": 0.2425,
      "epoch": 0.2427054519496743,
      "grad_norm": 158.0280303955078,
      "learning_rate": 9.019143716581472e-05,
      "loss": 0.05949435830116272,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0615,
      "step": 8160
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2127,
      "entropy_step_3": 0.2064,
      "epoch": 0.24300288510157342,
      "grad_norm": 98.657958984375,
      "learning_rate": 9.016216179812161e-05,
      "loss": 0.06730206608772278,
      "loss_step_1": 0.072,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0701,
      "step": 8170
    },
    {
      "entropy_step_1": 0.2186,
      "entropy_step_2": 0.2247,
      "entropy_step_3": 0.2466,
      "epoch": 0.24330031825347254,
      "grad_norm": 110.16642761230469,
      "learning_rate": 9.013284757299048e-05,
      "loss": 0.06028639078140259,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0596,
      "step": 8180
    },
    {
      "entropy_step_1": 0.23,
      "entropy_step_2": 0.2245,
      "entropy_step_3": 0.2405,
      "epoch": 0.24359775140537165,
      "grad_norm": 95.65181732177734,
      "learning_rate": 9.010349451878321e-05,
      "loss": 0.07601836919784546,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0737,
      "step": 8190
    },
    {
      "entropy_step_1": 0.2308,
      "entropy_step_2": 0.2181,
      "entropy_step_3": 0.2329,
      "epoch": 0.24389518455727074,
      "grad_norm": 64.61136627197266,
      "learning_rate": 9.007410266389935e-05,
      "loss": 0.06985559463500976,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0668,
      "step": 8200
    },
    {
      "entropy_step_1": 0.2266,
      "entropy_step_2": 0.2192,
      "entropy_step_3": 0.2267,
      "epoch": 0.24419261770916986,
      "grad_norm": 67.94685363769531,
      "learning_rate": 9.004467203677585e-05,
      "loss": 0.07035278081893921,
      "loss_step_1": 0.0769,
      "loss_step_2": 0.0746,
      "loss_step_3": 0.0753,
      "step": 8210
    },
    {
      "entropy_step_1": 0.2152,
      "entropy_step_2": 0.2137,
      "entropy_step_3": 0.1975,
      "epoch": 0.24449005086106898,
      "grad_norm": 180.1104736328125,
      "learning_rate": 9.001520266588727e-05,
      "loss": 0.0670454442501068,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0581,
      "step": 8220
    },
    {
      "entropy_step_1": 0.2093,
      "entropy_step_2": 0.2073,
      "entropy_step_3": 0.2089,
      "epoch": 0.2447874840129681,
      "grad_norm": 90.12586975097656,
      "learning_rate": 8.998569457974562e-05,
      "loss": 0.061086589097976686,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0519,
      "step": 8230
    },
    {
      "entropy_step_1": 0.2652,
      "entropy_step_2": 0.2627,
      "entropy_step_3": 0.2604,
      "epoch": 0.2450849171648672,
      "grad_norm": 367.2773132324219,
      "learning_rate": 8.995614780690038e-05,
      "loss": 0.0703213095664978,
      "loss_step_1": 0.0708,
      "loss_step_2": 0.0702,
      "loss_step_3": 0.0689,
      "step": 8240
    },
    {
      "entropy_step_1": 0.2008,
      "entropy_step_2": 0.2046,
      "entropy_step_3": 0.2113,
      "epoch": 0.2453823503167663,
      "grad_norm": 105.96823120117188,
      "learning_rate": 8.992656237593847e-05,
      "loss": 0.07392090559005737,
      "loss_step_1": 0.0781,
      "loss_step_2": 0.0775,
      "loss_step_3": 0.0577,
      "step": 8250
    },
    {
      "entropy_step_1": 0.2744,
      "entropy_step_2": 0.2713,
      "entropy_step_3": 0.2679,
      "epoch": 0.24567978346866542,
      "grad_norm": 420.2648010253906,
      "learning_rate": 8.989693831548416e-05,
      "loss": 0.07351195812225342,
      "loss_step_1": 0.0791,
      "loss_step_2": 0.078,
      "loss_step_3": 0.0777,
      "step": 8260
    },
    {
      "entropy_step_1": 0.1703,
      "entropy_step_2": 0.1833,
      "entropy_step_3": 0.1748,
      "epoch": 0.24597721662056451,
      "grad_norm": 105.54190826416016,
      "learning_rate": 8.986727565419917e-05,
      "loss": 0.08096259832382202,
      "loss_step_1": 0.0905,
      "loss_step_2": 0.0859,
      "loss_step_3": 0.0901,
      "step": 8270
    },
    {
      "entropy_step_1": 0.2264,
      "entropy_step_2": 0.2429,
      "entropy_step_3": 0.2248,
      "epoch": 0.24627464977246363,
      "grad_norm": 65.82577514648438,
      "learning_rate": 8.98375744207825e-05,
      "loss": 0.07226883769035339,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0602,
      "step": 8280
    },
    {
      "entropy_step_1": 0.2256,
      "entropy_step_2": 0.2391,
      "entropy_step_3": 0.233,
      "epoch": 0.24657208292436275,
      "grad_norm": 139.8597412109375,
      "learning_rate": 8.980783464397052e-05,
      "loss": 0.06149206161499023,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0654,
      "loss_step_3": 0.0584,
      "step": 8290
    },
    {
      "entropy_step_1": 0.2092,
      "entropy_step_2": 0.2266,
      "entropy_step_3": 0.2671,
      "epoch": 0.24686951607626187,
      "grad_norm": 99.8554916381836,
      "learning_rate": 8.977805635253686e-05,
      "loss": 0.06609837412834167,
      "loss_step_1": 0.0721,
      "loss_step_2": 0.0704,
      "loss_step_3": 0.0789,
      "step": 8300
    },
    {
      "entropy_step_1": 0.2129,
      "entropy_step_2": 0.2322,
      "entropy_step_3": 0.242,
      "epoch": 0.24716694922816096,
      "grad_norm": 35.98211669921875,
      "learning_rate": 8.974823957529242e-05,
      "loss": 0.06904119849205018,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0695,
      "step": 8310
    },
    {
      "entropy_step_1": 0.2376,
      "entropy_step_2": 0.2459,
      "entropy_step_3": 0.2529,
      "epoch": 0.24746438238006008,
      "grad_norm": 93.8543930053711,
      "learning_rate": 8.971838434108532e-05,
      "loss": 0.07086508870124816,
      "loss_step_1": 0.0733,
      "loss_step_2": 0.0739,
      "loss_step_3": 0.0775,
      "step": 8320
    },
    {
      "entropy_step_1": 0.2124,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.209,
      "epoch": 0.2477618155319592,
      "grad_norm": 90.0091323852539,
      "learning_rate": 8.968849067880093e-05,
      "loss": 0.05302618145942688,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0534,
      "step": 8330
    },
    {
      "entropy_step_1": 0.2206,
      "entropy_step_2": 0.223,
      "entropy_step_3": 0.2283,
      "epoch": 0.2480592486838583,
      "grad_norm": 51.48200225830078,
      "learning_rate": 8.965855861736179e-05,
      "loss": 0.06258568763732911,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0548,
      "step": 8340
    },
    {
      "entropy_step_1": 0.215,
      "entropy_step_2": 0.2101,
      "entropy_step_3": 0.1917,
      "epoch": 0.2483566818357574,
      "grad_norm": 110.04141998291016,
      "learning_rate": 8.962858818572753e-05,
      "loss": 0.06456630229949951,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0548,
      "step": 8350
    },
    {
      "entropy_step_1": 0.1958,
      "entropy_step_2": 0.1866,
      "entropy_step_3": 0.1714,
      "epoch": 0.24865411498765652,
      "grad_norm": 185.99061584472656,
      "learning_rate": 8.959857941289497e-05,
      "loss": 0.06738654375076295,
      "loss_step_1": 0.0674,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0537,
      "step": 8360
    },
    {
      "entropy_step_1": 0.2235,
      "entropy_step_2": 0.2004,
      "entropy_step_3": 0.2078,
      "epoch": 0.24895154813955564,
      "grad_norm": 49.30251693725586,
      "learning_rate": 8.956853232789804e-05,
      "loss": 0.06362667083740234,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0567,
      "step": 8370
    },
    {
      "entropy_step_1": 0.1848,
      "entropy_step_2": 0.1667,
      "entropy_step_3": 0.1772,
      "epoch": 0.24924898129145476,
      "grad_norm": 77.04841613769531,
      "learning_rate": 8.953844695980766e-05,
      "loss": 0.062405210733413694,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0722,
      "step": 8380
    },
    {
      "entropy_step_1": 0.2123,
      "entropy_step_2": 0.2167,
      "entropy_step_3": 0.2357,
      "epoch": 0.24954641444335385,
      "grad_norm": 125.4833755493164,
      "learning_rate": 8.950832333773184e-05,
      "loss": 0.0643155336380005,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0741,
      "step": 8390
    },
    {
      "entropy_step_1": 0.2057,
      "entropy_step_2": 0.2093,
      "entropy_step_3": 0.2323,
      "epoch": 0.24984384759525297,
      "grad_norm": 47.86640548706055,
      "learning_rate": 8.947816149081562e-05,
      "loss": 0.06518245339393616,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0681,
      "step": 8400
    },
    {
      "entropy_step_1": 0.2314,
      "entropy_step_2": 0.2251,
      "entropy_step_3": 0.2363,
      "epoch": 0.2501412807471521,
      "grad_norm": 41.14552688598633,
      "learning_rate": 8.944796144824097e-05,
      "loss": 0.0803784430027008,
      "loss_step_1": 0.0885,
      "loss_step_2": 0.0871,
      "loss_step_3": 0.0864,
      "step": 8410
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.173,
      "entropy_step_3": 0.1911,
      "epoch": 0.2504387138990512,
      "grad_norm": 54.52341079711914,
      "learning_rate": 8.941772323922685e-05,
      "loss": 0.05782074928283691,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0597,
      "step": 8420
    },
    {
      "entropy_step_1": 0.1967,
      "entropy_step_2": 0.1945,
      "entropy_step_3": 0.1973,
      "epoch": 0.2507361470509503,
      "grad_norm": 57.889041900634766,
      "learning_rate": 8.938744689302914e-05,
      "loss": 0.06326506733894348,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0624,
      "step": 8430
    },
    {
      "entropy_step_1": 0.2094,
      "entropy_step_2": 0.2021,
      "entropy_step_3": 0.2123,
      "epoch": 0.2510335802028494,
      "grad_norm": 128.7344207763672,
      "learning_rate": 8.935713243894061e-05,
      "loss": 0.06679847836494446,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.075,
      "step": 8440
    },
    {
      "entropy_step_1": 0.2399,
      "entropy_step_2": 0.2219,
      "entropy_step_3": 0.2059,
      "epoch": 0.2513310133547485,
      "grad_norm": 82.20132446289062,
      "learning_rate": 8.932677990629092e-05,
      "loss": 0.05916407108306885,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.053,
      "step": 8450
    },
    {
      "entropy_step_1": 0.2118,
      "entropy_step_2": 0.1969,
      "entropy_step_3": 0.2004,
      "epoch": 0.2516284465066476,
      "grad_norm": 61.826446533203125,
      "learning_rate": 8.929638932444658e-05,
      "loss": 0.07170878648757935,
      "loss_step_1": 0.0668,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0593,
      "step": 8460
    },
    {
      "entropy_step_1": 0.2307,
      "entropy_step_2": 0.2213,
      "entropy_step_3": 0.2362,
      "epoch": 0.25192587965854674,
      "grad_norm": 1313.1256103515625,
      "learning_rate": 8.926596072281083e-05,
      "loss": 0.06746459007263184,
      "loss_step_1": 0.0666,
      "loss_step_2": 0.0658,
      "loss_step_3": 0.0654,
      "step": 8470
    },
    {
      "entropy_step_1": 0.2487,
      "entropy_step_2": 0.2382,
      "entropy_step_3": 0.2209,
      "epoch": 0.25222331281044585,
      "grad_norm": 53.57522964477539,
      "learning_rate": 8.923549413082379e-05,
      "loss": 0.06657592058181763,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0572,
      "step": 8480
    },
    {
      "entropy_step_1": 0.2209,
      "entropy_step_2": 0.2194,
      "entropy_step_3": 0.2064,
      "epoch": 0.252520745962345,
      "grad_norm": 76.23648834228516,
      "learning_rate": 8.920498957796232e-05,
      "loss": 0.06759809255599976,
      "loss_step_1": 0.0718,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0629,
      "step": 8490
    },
    {
      "entropy_step_1": 0.2027,
      "entropy_step_2": 0.2009,
      "entropy_step_3": 0.1947,
      "epoch": 0.2528181791142441,
      "grad_norm": 122.58628845214844,
      "learning_rate": 8.917444709373996e-05,
      "loss": 0.06583386659622192,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0612,
      "step": 8500
    },
    {
      "entropy_step_1": 0.2217,
      "entropy_step_2": 0.2221,
      "entropy_step_3": 0.2303,
      "epoch": 0.2531156122661432,
      "grad_norm": 143.95516967773438,
      "learning_rate": 8.9143866707707e-05,
      "loss": 0.06737189292907715,
      "loss_step_1": 0.0665,
      "loss_step_2": 0.0659,
      "loss_step_3": 0.0632,
      "step": 8510
    },
    {
      "entropy_step_1": 0.192,
      "entropy_step_2": 0.1946,
      "entropy_step_3": 0.1936,
      "epoch": 0.25341304541804227,
      "grad_norm": 165.01986694335938,
      "learning_rate": 8.911324844945034e-05,
      "loss": 0.06619893908500671,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0607,
      "step": 8520
    },
    {
      "entropy_step_1": 0.2207,
      "entropy_step_2": 0.2252,
      "entropy_step_3": 0.2239,
      "epoch": 0.2537104785699414,
      "grad_norm": 88.15606689453125,
      "learning_rate": 8.908259234859364e-05,
      "loss": 0.0636023998260498,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0719,
      "step": 8530
    },
    {
      "entropy_step_1": 0.2292,
      "entropy_step_2": 0.2299,
      "entropy_step_3": 0.2466,
      "epoch": 0.2540079117218405,
      "grad_norm": 41.03483200073242,
      "learning_rate": 8.905189843479702e-05,
      "loss": 0.06697492599487305,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0739,
      "step": 8540
    },
    {
      "entropy_step_1": 0.2194,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.2173,
      "epoch": 0.2543053448737396,
      "grad_norm": 48.48267364501953,
      "learning_rate": 8.90211667377573e-05,
      "loss": 0.0629124402999878,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0682,
      "step": 8550
    },
    {
      "entropy_step_1": 0.2433,
      "entropy_step_2": 0.231,
      "entropy_step_3": 0.2295,
      "epoch": 0.25460277802563874,
      "grad_norm": 211.6405029296875,
      "learning_rate": 8.899039728720778e-05,
      "loss": 0.06860451698303223,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0675,
      "loss_step_3": 0.0691,
      "step": 8560
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.2065,
      "entropy_step_3": 0.2233,
      "epoch": 0.25490021117753786,
      "grad_norm": 121.27808380126953,
      "learning_rate": 8.895959011291836e-05,
      "loss": 0.06682854294776916,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0715,
      "step": 8570
    },
    {
      "entropy_step_1": 0.2453,
      "entropy_step_2": 0.2485,
      "entropy_step_3": 0.2441,
      "epoch": 0.255197644329437,
      "grad_norm": 126.60099792480469,
      "learning_rate": 8.892874524469538e-05,
      "loss": 0.06774318218231201,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.0679,
      "step": 8580
    },
    {
      "entropy_step_1": 0.1982,
      "entropy_step_2": 0.2038,
      "entropy_step_3": 0.1892,
      "epoch": 0.2554950774813361,
      "grad_norm": 59.70271682739258,
      "learning_rate": 8.889786271238168e-05,
      "loss": 0.06523889303207397,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0625,
      "step": 8590
    },
    {
      "entropy_step_1": 0.2118,
      "entropy_step_2": 0.2166,
      "entropy_step_3": 0.2029,
      "epoch": 0.25579251063323516,
      "grad_norm": 82.8730697631836,
      "learning_rate": 8.886694254585651e-05,
      "loss": 0.06683255434036255,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0588,
      "step": 8600
    },
    {
      "entropy_step_1": 0.2222,
      "entropy_step_2": 0.2167,
      "entropy_step_3": 0.2308,
      "epoch": 0.2560899437851343,
      "grad_norm": 29.20696258544922,
      "learning_rate": 8.883598477503556e-05,
      "loss": 0.059498411417007444,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0662,
      "step": 8610
    },
    {
      "entropy_step_1": 0.2147,
      "entropy_step_2": 0.2101,
      "entropy_step_3": 0.1931,
      "epoch": 0.2563873769370334,
      "grad_norm": 42.75193786621094,
      "learning_rate": 8.880498942987092e-05,
      "loss": 0.06401170492172241,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.055,
      "step": 8620
    },
    {
      "entropy_step_1": 0.1915,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.2071,
      "epoch": 0.2566848100889325,
      "grad_norm": 89.3790512084961,
      "learning_rate": 8.877395654035098e-05,
      "loss": 0.055965638160705565,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0608,
      "step": 8630
    },
    {
      "entropy_step_1": 0.2282,
      "entropy_step_2": 0.2287,
      "entropy_step_3": 0.2241,
      "epoch": 0.25698224324083163,
      "grad_norm": 93.27536010742188,
      "learning_rate": 8.874288613650049e-05,
      "loss": 0.06846846342086792,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0676,
      "loss_step_3": 0.0644,
      "step": 8640
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1822,
      "entropy_step_3": 0.1794,
      "epoch": 0.25727967639273075,
      "grad_norm": 122.57518768310547,
      "learning_rate": 8.871177824838048e-05,
      "loss": 0.059996843338012695,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0565,
      "step": 8650
    },
    {
      "entropy_step_1": 0.2258,
      "entropy_step_2": 0.214,
      "entropy_step_3": 0.2133,
      "epoch": 0.25757710954462987,
      "grad_norm": 83.35755157470703,
      "learning_rate": 8.86806329060883e-05,
      "loss": 0.06884328126907349,
      "loss_step_1": 0.0619,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0582,
      "step": 8660
    },
    {
      "entropy_step_1": 0.1767,
      "entropy_step_2": 0.1696,
      "entropy_step_3": 0.1604,
      "epoch": 0.25787454269652893,
      "grad_norm": 66.80352783203125,
      "learning_rate": 8.864945013975743e-05,
      "loss": 0.059555810689926145,
      "loss_step_1": 0.061,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0583,
      "step": 8670
    },
    {
      "entropy_step_1": 0.2135,
      "entropy_step_2": 0.1966,
      "entropy_step_3": 0.2117,
      "epoch": 0.25817197584842805,
      "grad_norm": 156.66519165039062,
      "learning_rate": 8.861822997955764e-05,
      "loss": 0.06557616591453552,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0632,
      "step": 8680
    },
    {
      "entropy_step_1": 0.2351,
      "entropy_step_2": 0.2153,
      "entropy_step_3": 0.2312,
      "epoch": 0.25846940900032717,
      "grad_norm": 552.9268188476562,
      "learning_rate": 8.858697245569488e-05,
      "loss": 0.07031681537628173,
      "loss_step_1": 0.0717,
      "loss_step_2": 0.0715,
      "loss_step_3": 0.0777,
      "step": 8690
    },
    {
      "entropy_step_1": 0.2092,
      "entropy_step_2": 0.2105,
      "entropy_step_3": 0.2112,
      "epoch": 0.2587668421522263,
      "grad_norm": 66.77665710449219,
      "learning_rate": 8.855567759841121e-05,
      "loss": 0.06184561848640442,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0526,
      "step": 8700
    },
    {
      "entropy_step_1": 0.2139,
      "entropy_step_2": 0.2164,
      "entropy_step_3": 0.2251,
      "epoch": 0.2590642753041254,
      "grad_norm": 92.1580810546875,
      "learning_rate": 8.852434543798485e-05,
      "loss": 0.06318224668502807,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0687,
      "step": 8710
    },
    {
      "entropy_step_1": 0.2146,
      "entropy_step_2": 0.2091,
      "entropy_step_3": 0.2065,
      "epoch": 0.2593617084560245,
      "grad_norm": 69.77599334716797,
      "learning_rate": 8.849297600473007e-05,
      "loss": 0.06639938354492188,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.0652,
      "loss_step_3": 0.0651,
      "step": 8720
    },
    {
      "entropy_step_1": 0.2263,
      "entropy_step_2": 0.2155,
      "entropy_step_3": 0.2145,
      "epoch": 0.25965914160792364,
      "grad_norm": 65.51806640625,
      "learning_rate": 8.846156932899723e-05,
      "loss": 0.06644797921180726,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.068,
      "step": 8730
    },
    {
      "entropy_step_1": 0.1868,
      "entropy_step_2": 0.176,
      "entropy_step_3": 0.1808,
      "epoch": 0.25995657475982276,
      "grad_norm": 65.04584503173828,
      "learning_rate": 8.843012544117273e-05,
      "loss": 0.061761653423309325,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.058,
      "step": 8740
    },
    {
      "entropy_step_1": 0.2073,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.2194,
      "epoch": 0.2602540079117218,
      "grad_norm": 124.52558898925781,
      "learning_rate": 8.839864437167894e-05,
      "loss": 0.06658529043197632,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0739,
      "step": 8750
    },
    {
      "entropy_step_1": 0.2358,
      "entropy_step_2": 0.2323,
      "entropy_step_3": 0.2368,
      "epoch": 0.26055144106362094,
      "grad_norm": 85.69343566894531,
      "learning_rate": 8.836712615097425e-05,
      "loss": 0.06573988795280457,
      "loss_step_1": 0.0737,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0726,
      "step": 8760
    },
    {
      "entropy_step_1": 0.195,
      "entropy_step_2": 0.2022,
      "entropy_step_3": 0.2048,
      "epoch": 0.26084887421552005,
      "grad_norm": 120.33555603027344,
      "learning_rate": 8.833557080955293e-05,
      "loss": 0.061267465353012085,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0605,
      "step": 8770
    },
    {
      "entropy_step_1": 0.2282,
      "entropy_step_2": 0.2327,
      "entropy_step_3": 0.2292,
      "epoch": 0.2611463073674192,
      "grad_norm": 72.79170989990234,
      "learning_rate": 8.830397837794521e-05,
      "loss": 0.0721830427646637,
      "loss_step_1": 0.0762,
      "loss_step_2": 0.0745,
      "loss_step_3": 0.0702,
      "step": 8780
    },
    {
      "entropy_step_1": 0.1977,
      "entropy_step_2": 0.2112,
      "entropy_step_3": 0.1988,
      "epoch": 0.2614437405193183,
      "grad_norm": 398.3570251464844,
      "learning_rate": 8.827234888671723e-05,
      "loss": 0.05646679401397705,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0603,
      "step": 8790
    },
    {
      "entropy_step_1": 0.2031,
      "entropy_step_2": 0.2257,
      "entropy_step_3": 0.2208,
      "epoch": 0.2617411736712174,
      "grad_norm": 93.45822143554688,
      "learning_rate": 8.824068236647091e-05,
      "loss": 0.0595172643661499,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.057,
      "step": 8800
    },
    {
      "entropy_step_1": 0.1935,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2123,
      "epoch": 0.2620386068231165,
      "grad_norm": 648.7051391601562,
      "learning_rate": 8.820897884784407e-05,
      "loss": 0.0705479085445404,
      "loss_step_1": 0.0773,
      "loss_step_2": 0.0732,
      "loss_step_3": 0.07,
      "step": 8810
    },
    {
      "entropy_step_1": 0.2172,
      "entropy_step_2": 0.2179,
      "entropy_step_3": 0.2005,
      "epoch": 0.2623360399750156,
      "grad_norm": 177.1621551513672,
      "learning_rate": 8.817723836151028e-05,
      "loss": 0.055916118621826175,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0633,
      "step": 8820
    },
    {
      "entropy_step_1": 0.1806,
      "entropy_step_2": 0.1838,
      "entropy_step_3": 0.197,
      "epoch": 0.2626334731269147,
      "grad_norm": 204.78369140625,
      "learning_rate": 8.814546093817892e-05,
      "loss": 0.058534014225006106,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0656,
      "step": 8830
    },
    {
      "entropy_step_1": 0.2106,
      "entropy_step_2": 0.2165,
      "entropy_step_3": 0.2282,
      "epoch": 0.2629309062788138,
      "grad_norm": 123.72289276123047,
      "learning_rate": 8.811364660859505e-05,
      "loss": 0.06223071813583374,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0696,
      "step": 8840
    },
    {
      "entropy_step_1": 0.1883,
      "entropy_step_2": 0.2015,
      "entropy_step_3": 0.1782,
      "epoch": 0.26322833943071294,
      "grad_norm": 49.39513397216797,
      "learning_rate": 8.808179540353947e-05,
      "loss": 0.054026877880096434,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0457,
      "step": 8850
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2115,
      "entropy_step_3": 0.2041,
      "epoch": 0.26352577258261206,
      "grad_norm": 48.42926788330078,
      "learning_rate": 8.804990735382865e-05,
      "loss": 0.05954674482345581,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.058,
      "step": 8860
    },
    {
      "entropy_step_1": 0.2132,
      "entropy_step_2": 0.2111,
      "entropy_step_3": 0.1971,
      "epoch": 0.2638232057345112,
      "grad_norm": 598.6470947265625,
      "learning_rate": 8.801798249031474e-05,
      "loss": 0.06299484372138978,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0572,
      "step": 8870
    },
    {
      "entropy_step_1": 0.2116,
      "entropy_step_2": 0.2096,
      "entropy_step_3": 0.1984,
      "epoch": 0.2641206388864103,
      "grad_norm": 78.64302825927734,
      "learning_rate": 8.798602084388545e-05,
      "loss": 0.06382767558097839,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0709,
      "loss_step_3": 0.0549,
      "step": 8880
    },
    {
      "entropy_step_1": 0.2214,
      "entropy_step_2": 0.2224,
      "entropy_step_3": 0.2402,
      "epoch": 0.2644180720383094,
      "grad_norm": 110.22325897216797,
      "learning_rate": 8.795402244546413e-05,
      "loss": 0.06757518649101257,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0683,
      "step": 8890
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.2021,
      "entropy_step_3": 0.2101,
      "epoch": 0.2647155051902085,
      "grad_norm": 40.99565124511719,
      "learning_rate": 8.792198732600963e-05,
      "loss": 0.05700868964195251,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0464,
      "step": 8900
    },
    {
      "entropy_step_1": 0.2224,
      "entropy_step_2": 0.225,
      "entropy_step_3": 0.2112,
      "epoch": 0.2650129383421076,
      "grad_norm": 51.714969635009766,
      "learning_rate": 8.788991551651642e-05,
      "loss": 0.06563426256179809,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0631,
      "step": 8910
    },
    {
      "entropy_step_1": 0.2185,
      "entropy_step_2": 0.2143,
      "entropy_step_3": 0.2069,
      "epoch": 0.2653103714940067,
      "grad_norm": 277.11865234375,
      "learning_rate": 8.785780704801436e-05,
      "loss": 0.06697630286216735,
      "loss_step_1": 0.0717,
      "loss_step_2": 0.0701,
      "loss_step_3": 0.0631,
      "step": 8920
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1861,
      "entropy_step_3": 0.1722,
      "epoch": 0.26560780464590583,
      "grad_norm": 49.48381423950195,
      "learning_rate": 8.782566195156886e-05,
      "loss": 0.058554995059967044,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0604,
      "step": 8930
    },
    {
      "entropy_step_1": 0.2374,
      "entropy_step_2": 0.257,
      "entropy_step_3": 0.2481,
      "epoch": 0.26590523779780495,
      "grad_norm": 74.90861511230469,
      "learning_rate": 8.779348025828071e-05,
      "loss": 0.06596698760986328,
      "loss_step_1": 0.0762,
      "loss_step_2": 0.0746,
      "loss_step_3": 0.0636,
      "step": 8940
    },
    {
      "entropy_step_1": 0.2309,
      "entropy_step_2": 0.2474,
      "entropy_step_3": 0.244,
      "epoch": 0.26620267094970407,
      "grad_norm": 68.46836853027344,
      "learning_rate": 8.776126199928619e-05,
      "loss": 0.07089759707450867,
      "loss_step_1": 0.0693,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0739,
      "step": 8950
    },
    {
      "entropy_step_1": 0.2184,
      "entropy_step_2": 0.2174,
      "entropy_step_3": 0.2349,
      "epoch": 0.2665001041016032,
      "grad_norm": 128.63772583007812,
      "learning_rate": 8.772900720575684e-05,
      "loss": 0.06563959121704102,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0725,
      "step": 8960
    },
    {
      "entropy_step_1": 0.2173,
      "entropy_step_2": 0.2046,
      "entropy_step_3": 0.1958,
      "epoch": 0.26679753725350225,
      "grad_norm": 30.589555740356445,
      "learning_rate": 8.769671590889965e-05,
      "loss": 0.0647486388683319,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0585,
      "step": 8970
    },
    {
      "entropy_step_1": 0.2151,
      "entropy_step_2": 0.2002,
      "entropy_step_3": 0.1816,
      "epoch": 0.26709497040540137,
      "grad_norm": 51.550106048583984,
      "learning_rate": 8.766438813995688e-05,
      "loss": 0.06575678586959839,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0537,
      "step": 8980
    },
    {
      "entropy_step_1": 0.2043,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.1944,
      "epoch": 0.2673924035573005,
      "grad_norm": 85.85247039794922,
      "learning_rate": 8.763202393020609e-05,
      "loss": 0.06392547488212585,
      "loss_step_1": 0.0641,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.0624,
      "step": 8990
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.2052,
      "entropy_step_3": 0.1998,
      "epoch": 0.2676898367091996,
      "grad_norm": 86.36274719238281,
      "learning_rate": 8.759962331096008e-05,
      "loss": 0.06862123012542724,
      "loss_step_1": 0.0703,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0747,
      "step": 9000
    },
    {
      "epoch": 0.2676898367091996,
      "eval_accuracy": 0.836,
      "eval_f1": 0.8348345928874136,
      "eval_loss": 0.16392123699188232,
      "eval_precision": 0.8457586933614331,
      "eval_recall": 0.836,
      "eval_runtime": 8.2945,
      "eval_samples_per_second": 60.281,
      "eval_steps_per_second": 7.595,
      "step": 9000
    },
    {
      "entropy_step_1": 0.2166,
      "entropy_step_2": 0.2192,
      "entropy_step_3": 0.2045,
      "epoch": 0.2679872698610987,
      "grad_norm": 61.709327697753906,
      "learning_rate": 8.756718631356692e-05,
      "loss": 0.057328718900680545,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0643,
      "step": 9010
    },
    {
      "entropy_step_1": 0.1905,
      "entropy_step_2": 0.2033,
      "entropy_step_3": 0.1917,
      "epoch": 0.26828470301299784,
      "grad_norm": 104.87523651123047,
      "learning_rate": 8.753471296940981e-05,
      "loss": 0.06480485796928406,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.0566,
      "step": 9020
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2138,
      "entropy_step_3": 0.2175,
      "epoch": 0.26858213616489696,
      "grad_norm": 121.35801696777344,
      "learning_rate": 8.750220330990718e-05,
      "loss": 0.06394923329353333,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0684,
      "step": 9030
    },
    {
      "entropy_step_1": 0.1961,
      "entropy_step_2": 0.2016,
      "entropy_step_3": 0.2001,
      "epoch": 0.2688795693167961,
      "grad_norm": 51.438812255859375,
      "learning_rate": 8.746965736651254e-05,
      "loss": 0.05925137400627136,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0548,
      "step": 9040
    },
    {
      "entropy_step_1": 0.2187,
      "entropy_step_2": 0.2222,
      "entropy_step_3": 0.1989,
      "epoch": 0.26917700246869514,
      "grad_norm": 326.1811828613281,
      "learning_rate": 8.743707517071455e-05,
      "loss": 0.06455451250076294,
      "loss_step_1": 0.0691,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0536,
      "step": 9050
    },
    {
      "entropy_step_1": 0.2584,
      "entropy_step_2": 0.258,
      "entropy_step_3": 0.2648,
      "epoch": 0.26947443562059425,
      "grad_norm": 54.0351676940918,
      "learning_rate": 8.740445675403694e-05,
      "loss": 0.07489205598831176,
      "loss_step_1": 0.0792,
      "loss_step_2": 0.0778,
      "loss_step_3": 0.0808,
      "step": 9060
    },
    {
      "entropy_step_1": 0.1905,
      "entropy_step_2": 0.1902,
      "entropy_step_3": 0.1983,
      "epoch": 0.2697718687724934,
      "grad_norm": 216.1924591064453,
      "learning_rate": 8.737180214803846e-05,
      "loss": 0.06876054406166077,
      "loss_step_1": 0.0743,
      "loss_step_2": 0.0737,
      "loss_step_3": 0.0888,
      "step": 9070
    },
    {
      "entropy_step_1": 0.2285,
      "entropy_step_2": 0.2295,
      "entropy_step_3": 0.2303,
      "epoch": 0.2700693019243925,
      "grad_norm": 52.9063606262207,
      "learning_rate": 8.733911138431288e-05,
      "loss": 0.06892918348312378,
      "loss_step_1": 0.0697,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0733,
      "step": 9080
    },
    {
      "entropy_step_1": 0.1958,
      "entropy_step_2": 0.1958,
      "entropy_step_3": 0.199,
      "epoch": 0.2703667350762916,
      "grad_norm": 71.05464935302734,
      "learning_rate": 8.730638449448896e-05,
      "loss": 0.05961363911628723,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0532,
      "step": 9090
    },
    {
      "entropy_step_1": 0.1905,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.1945,
      "epoch": 0.2706641682281907,
      "grad_norm": 103.81517791748047,
      "learning_rate": 8.727362151023041e-05,
      "loss": 0.06616173386573791,
      "loss_step_1": 0.071,
      "loss_step_2": 0.0702,
      "loss_step_3": 0.0726,
      "step": 9100
    },
    {
      "entropy_step_1": 0.2013,
      "entropy_step_2": 0.1992,
      "entropy_step_3": 0.1855,
      "epoch": 0.27096160138008984,
      "grad_norm": 38.39331817626953,
      "learning_rate": 8.724082246323587e-05,
      "loss": 0.054530000686645506,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.043,
      "step": 9110
    },
    {
      "entropy_step_1": 0.2066,
      "entropy_step_2": 0.2009,
      "entropy_step_3": 0.199,
      "epoch": 0.27125903453198896,
      "grad_norm": 50.760250091552734,
      "learning_rate": 8.720798738523887e-05,
      "loss": 0.06737218499183655,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.064,
      "step": 9120
    },
    {
      "entropy_step_1": 0.2128,
      "entropy_step_2": 0.2106,
      "entropy_step_3": 0.2123,
      "epoch": 0.271556467683888,
      "grad_norm": 87.1677474975586,
      "learning_rate": 8.717511630800779e-05,
      "loss": 0.06624113321304322,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0694,
      "step": 9130
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1988,
      "entropy_step_3": 0.1978,
      "epoch": 0.27185390083578714,
      "grad_norm": 106.00553131103516,
      "learning_rate": 8.714220926334583e-05,
      "loss": 0.06577042937278747,
      "loss_step_1": 0.073,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0661,
      "step": 9140
    },
    {
      "entropy_step_1": 0.2087,
      "entropy_step_2": 0.2294,
      "entropy_step_3": 0.2185,
      "epoch": 0.27215133398768626,
      "grad_norm": 133.99234008789062,
      "learning_rate": 8.710926628309102e-05,
      "loss": 0.06273196339607238,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0624,
      "step": 9150
    },
    {
      "entropy_step_1": 0.1983,
      "entropy_step_2": 0.2302,
      "entropy_step_3": 0.2393,
      "epoch": 0.2724487671395854,
      "grad_norm": 62.31758499145508,
      "learning_rate": 8.707628739911613e-05,
      "loss": 0.06895508766174316,
      "loss_step_1": 0.0774,
      "loss_step_2": 0.0748,
      "loss_step_3": 0.0737,
      "step": 9160
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.2156,
      "entropy_step_3": 0.2114,
      "epoch": 0.2727462002914845,
      "grad_norm": 87.86634826660156,
      "learning_rate": 8.70432726433287e-05,
      "loss": 0.057576483488082884,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0484,
      "step": 9170
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.187,
      "entropy_step_3": 0.1884,
      "epoch": 0.2730436334433836,
      "grad_norm": 46.23208236694336,
      "learning_rate": 8.701022204767094e-05,
      "loss": 0.06030834317207336,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0564,
      "step": 9180
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.2045,
      "entropy_step_3": 0.2002,
      "epoch": 0.27334106659528273,
      "grad_norm": 67.57634735107422,
      "learning_rate": 8.697713564411973e-05,
      "loss": 0.06171878576278687,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0589,
      "step": 9190
    },
    {
      "entropy_step_1": 0.1876,
      "entropy_step_2": 0.1971,
      "entropy_step_3": 0.2062,
      "epoch": 0.2736384997471818,
      "grad_norm": 390.91314697265625,
      "learning_rate": 8.694401346468664e-05,
      "loss": 0.05606215000152588,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0597,
      "step": 9200
    },
    {
      "entropy_step_1": 0.2189,
      "entropy_step_2": 0.2219,
      "entropy_step_3": 0.2215,
      "epoch": 0.2739359328990809,
      "grad_norm": 59.93711853027344,
      "learning_rate": 8.691085554141784e-05,
      "loss": 0.06488144397735596,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.064,
      "step": 9210
    },
    {
      "entropy_step_1": 0.2523,
      "entropy_step_2": 0.2426,
      "entropy_step_3": 0.2577,
      "epoch": 0.27423336605098003,
      "grad_norm": 89.80084991455078,
      "learning_rate": 8.687766190639406e-05,
      "loss": 0.06393908858299255,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0686,
      "step": 9220
    },
    {
      "entropy_step_1": 0.2073,
      "entropy_step_2": 0.2009,
      "entropy_step_3": 0.2118,
      "epoch": 0.27453079920287915,
      "grad_norm": 84.71327209472656,
      "learning_rate": 8.68444325917306e-05,
      "loss": 0.0599575400352478,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0642,
      "step": 9230
    },
    {
      "entropy_step_1": 0.1879,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.1608,
      "epoch": 0.27482823235477827,
      "grad_norm": 148.561279296875,
      "learning_rate": 8.681116762957724e-05,
      "loss": 0.06122756600379944,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.054,
      "step": 9240
    },
    {
      "entropy_step_1": 0.2171,
      "entropy_step_2": 0.2078,
      "entropy_step_3": 0.2136,
      "epoch": 0.2751256655066774,
      "grad_norm": 78.77574157714844,
      "learning_rate": 8.677786705211835e-05,
      "loss": 0.059562063217163085,
      "loss_step_1": 0.0619,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0582,
      "step": 9250
    },
    {
      "entropy_step_1": 0.163,
      "entropy_step_2": 0.1603,
      "entropy_step_3": 0.1586,
      "epoch": 0.2754230986585765,
      "grad_norm": 515.2099609375,
      "learning_rate": 8.674453089157261e-05,
      "loss": 0.05764968991279602,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0584,
      "step": 9260
    },
    {
      "entropy_step_1": 0.2054,
      "entropy_step_2": 0.203,
      "entropy_step_3": 0.2292,
      "epoch": 0.2757205318104756,
      "grad_norm": 101.32806396484375,
      "learning_rate": 8.671115918019327e-05,
      "loss": 0.06692843437194824,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0793,
      "step": 9270
    },
    {
      "entropy_step_1": 0.2083,
      "entropy_step_2": 0.2277,
      "entropy_step_3": 0.2262,
      "epoch": 0.2760179649623747,
      "grad_norm": 44.953269958496094,
      "learning_rate": 8.667775195026789e-05,
      "loss": 0.06550166606903077,
      "loss_step_1": 0.072,
      "loss_step_2": 0.0704,
      "loss_step_3": 0.0641,
      "step": 9280
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.21,
      "epoch": 0.2763153981142738,
      "grad_norm": 150.85052490234375,
      "learning_rate": 8.664430923411843e-05,
      "loss": 0.07006179094314575,
      "loss_step_1": 0.0765,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.0726,
      "step": 9290
    },
    {
      "entropy_step_1": 0.2293,
      "entropy_step_2": 0.2477,
      "entropy_step_3": 0.238,
      "epoch": 0.2766128312661729,
      "grad_norm": 69.01634979248047,
      "learning_rate": 8.661083106410118e-05,
      "loss": 0.06161515116691589,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0488,
      "step": 9300
    },
    {
      "entropy_step_1": 0.2315,
      "entropy_step_2": 0.2203,
      "entropy_step_3": 0.2219,
      "epoch": 0.27691026441807204,
      "grad_norm": 177.30075073242188,
      "learning_rate": 8.657731747260668e-05,
      "loss": 0.06410667896270753,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0555,
      "step": 9310
    },
    {
      "entropy_step_1": 0.2271,
      "entropy_step_2": 0.2188,
      "entropy_step_3": 0.2006,
      "epoch": 0.27720769756997116,
      "grad_norm": 45.80161666870117,
      "learning_rate": 8.654376849205985e-05,
      "loss": 0.06522660851478576,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0627,
      "step": 9320
    },
    {
      "entropy_step_1": 0.2204,
      "entropy_step_2": 0.2089,
      "entropy_step_3": 0.2,
      "epoch": 0.2775051307218703,
      "grad_norm": 57.53732681274414,
      "learning_rate": 8.651018415491974e-05,
      "loss": 0.06318278908729554,
      "loss_step_1": 0.0703,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0597,
      "step": 9330
    },
    {
      "entropy_step_1": 0.2485,
      "entropy_step_2": 0.233,
      "entropy_step_3": 0.2215,
      "epoch": 0.2778025638737694,
      "grad_norm": 144.98593139648438,
      "learning_rate": 8.647656449367967e-05,
      "loss": 0.06783353090286255,
      "loss_step_1": 0.07,
      "loss_step_2": 0.0692,
      "loss_step_3": 0.0678,
      "step": 9340
    },
    {
      "entropy_step_1": 0.2303,
      "entropy_step_2": 0.2166,
      "entropy_step_3": 0.2202,
      "epoch": 0.27809999702566846,
      "grad_norm": 34.8060302734375,
      "learning_rate": 8.644290954086711e-05,
      "loss": 0.06276975274085998,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0618,
      "step": 9350
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.1899,
      "entropy_step_3": 0.1877,
      "epoch": 0.2783974301775676,
      "grad_norm": 70.68110656738281,
      "learning_rate": 8.64092193290437e-05,
      "loss": 0.07040436267852783,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0672,
      "step": 9360
    },
    {
      "entropy_step_1": 0.2294,
      "entropy_step_2": 0.2201,
      "entropy_step_3": 0.2112,
      "epoch": 0.2786948633294667,
      "grad_norm": 213.91232299804688,
      "learning_rate": 8.637549389080516e-05,
      "loss": 0.07872928977012635,
      "loss_step_1": 0.0805,
      "loss_step_2": 0.0819,
      "loss_step_3": 0.0804,
      "step": 9370
    },
    {
      "entropy_step_1": 0.2154,
      "entropy_step_2": 0.2137,
      "entropy_step_3": 0.2314,
      "epoch": 0.2789922964813658,
      "grad_norm": 141.99757385253906,
      "learning_rate": 8.634173325878135e-05,
      "loss": 0.06695796847343445,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0615,
      "step": 9380
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.1664,
      "epoch": 0.2792897296332649,
      "grad_norm": 120.63399505615234,
      "learning_rate": 8.630793746563608e-05,
      "loss": 0.06171674132347107,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.0606,
      "step": 9390
    },
    {
      "entropy_step_1": 0.2345,
      "entropy_step_2": 0.2333,
      "entropy_step_3": 0.2344,
      "epoch": 0.27958716278516405,
      "grad_norm": 76.21360778808594,
      "learning_rate": 8.627410654406728e-05,
      "loss": 0.06298595666885376,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0576,
      "step": 9400
    },
    {
      "entropy_step_1": 0.2036,
      "entropy_step_2": 0.2056,
      "entropy_step_3": 0.2039,
      "epoch": 0.27988459593706316,
      "grad_norm": 69.52822875976562,
      "learning_rate": 8.624024052680682e-05,
      "loss": 0.060113584995269774,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0479,
      "step": 9410
    },
    {
      "entropy_step_1": 0.2089,
      "entropy_step_2": 0.208,
      "entropy_step_3": 0.2074,
      "epoch": 0.2801820290889623,
      "grad_norm": 111.6306381225586,
      "learning_rate": 8.620633944662053e-05,
      "loss": 0.0705528736114502,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.064,
      "step": 9420
    },
    {
      "entropy_step_1": 0.2418,
      "entropy_step_2": 0.2326,
      "entropy_step_3": 0.2106,
      "epoch": 0.28047946224086134,
      "grad_norm": 110.08809661865234,
      "learning_rate": 8.617240333630816e-05,
      "loss": 0.06324725747108459,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0467,
      "step": 9430
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.1838,
      "entropy_step_3": 0.1847,
      "epoch": 0.28077689539276046,
      "grad_norm": 192.30723571777344,
      "learning_rate": 8.613843222870337e-05,
      "loss": 0.061316901445388795,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0541,
      "step": 9440
    },
    {
      "entropy_step_1": 0.2491,
      "entropy_step_2": 0.2316,
      "entropy_step_3": 0.2297,
      "epoch": 0.2810743285446596,
      "grad_norm": 123.26160430908203,
      "learning_rate": 8.610442615667367e-05,
      "loss": 0.06472057700157166,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0615,
      "step": 9450
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.2019,
      "entropy_step_3": 0.2213,
      "epoch": 0.2813717616965587,
      "grad_norm": 131.59823608398438,
      "learning_rate": 8.607038515312038e-05,
      "loss": 0.0533504843711853,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0558,
      "step": 9460
    },
    {
      "entropy_step_1": 0.2112,
      "entropy_step_2": 0.2198,
      "entropy_step_3": 0.2234,
      "epoch": 0.2816691948484578,
      "grad_norm": 81.24897003173828,
      "learning_rate": 8.603630925097865e-05,
      "loss": 0.0723314881324768,
      "loss_step_1": 0.0725,
      "loss_step_2": 0.0715,
      "loss_step_3": 0.0792,
      "step": 9470
    },
    {
      "entropy_step_1": 0.1928,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.2065,
      "epoch": 0.28196662800035693,
      "grad_norm": 75.47930145263672,
      "learning_rate": 8.600219848321738e-05,
      "loss": 0.06913756728172302,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0538,
      "step": 9480
    },
    {
      "entropy_step_1": 0.2124,
      "entropy_step_2": 0.2257,
      "entropy_step_3": 0.2221,
      "epoch": 0.28226406115225605,
      "grad_norm": 418.8944396972656,
      "learning_rate": 8.59680528828392e-05,
      "loss": 0.06869551539421082,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.0613,
      "step": 9490
    },
    {
      "entropy_step_1": 0.1984,
      "entropy_step_2": 0.2113,
      "entropy_step_3": 0.1946,
      "epoch": 0.2825614943041551,
      "grad_norm": 91.87527465820312,
      "learning_rate": 8.593387248288042e-05,
      "loss": 0.061914265155792236,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0475,
      "step": 9500
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.2065,
      "entropy_step_3": 0.2122,
      "epoch": 0.28285892745605423,
      "grad_norm": 92.36898803710938,
      "learning_rate": 8.589965731641107e-05,
      "loss": 0.05765443444252014,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0668,
      "step": 9510
    },
    {
      "entropy_step_1": 0.1766,
      "entropy_step_2": 0.1812,
      "entropy_step_3": 0.1793,
      "epoch": 0.28315636060795335,
      "grad_norm": 62.304893493652344,
      "learning_rate": 8.58654074165348e-05,
      "loss": 0.05534162521362305,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0551,
      "step": 9520
    },
    {
      "entropy_step_1": 0.2062,
      "entropy_step_2": 0.2092,
      "entropy_step_3": 0.2136,
      "epoch": 0.28345379375985247,
      "grad_norm": 1063.4512939453125,
      "learning_rate": 8.583112281638881e-05,
      "loss": 0.0649878740310669,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.057,
      "step": 9530
    },
    {
      "entropy_step_1": 0.222,
      "entropy_step_2": 0.223,
      "entropy_step_3": 0.226,
      "epoch": 0.2837512269117516,
      "grad_norm": 76.82576751708984,
      "learning_rate": 8.579680354914396e-05,
      "loss": 0.07163575291633606,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0692,
      "loss_step_3": 0.0734,
      "step": 9540
    },
    {
      "entropy_step_1": 0.2159,
      "entropy_step_2": 0.2129,
      "entropy_step_3": 0.22,
      "epoch": 0.2840486600636507,
      "grad_norm": 71.44548034667969,
      "learning_rate": 8.576244964800458e-05,
      "loss": 0.061902803182601926,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0588,
      "step": 9550
    },
    {
      "entropy_step_1": 0.1879,
      "entropy_step_2": 0.1771,
      "entropy_step_3": 0.1836,
      "epoch": 0.2843460932155498,
      "grad_norm": 35.435462951660156,
      "learning_rate": 8.572806114620856e-05,
      "loss": 0.06811543703079223,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0581,
      "step": 9560
    },
    {
      "entropy_step_1": 0.2274,
      "entropy_step_2": 0.2116,
      "entropy_step_3": 0.2205,
      "epoch": 0.28464352636744894,
      "grad_norm": 131.624755859375,
      "learning_rate": 8.569363807702722e-05,
      "loss": 0.06995614767074584,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0619,
      "step": 9570
    },
    {
      "entropy_step_1": 0.2407,
      "entropy_step_2": 0.225,
      "entropy_step_3": 0.236,
      "epoch": 0.284940959519348,
      "grad_norm": 46.07236099243164,
      "learning_rate": 8.565918047376535e-05,
      "loss": 0.0691650927066803,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.067,
      "step": 9580
    },
    {
      "entropy_step_1": 0.1968,
      "entropy_step_2": 0.1895,
      "entropy_step_3": 0.19,
      "epoch": 0.2852383926712471,
      "grad_norm": 102.51307678222656,
      "learning_rate": 8.562468836976117e-05,
      "loss": 0.06789202094078065,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.057,
      "step": 9590
    },
    {
      "entropy_step_1": 0.2583,
      "entropy_step_2": 0.2497,
      "entropy_step_3": 0.2403,
      "epoch": 0.28553582582314624,
      "grad_norm": 56.62163162231445,
      "learning_rate": 8.559016179838624e-05,
      "loss": 0.07618694901466369,
      "loss_step_1": 0.0767,
      "loss_step_2": 0.0756,
      "loss_step_3": 0.0684,
      "step": 9600
    },
    {
      "entropy_step_1": 0.1808,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.1941,
      "epoch": 0.28583325897504536,
      "grad_norm": 97.47997283935547,
      "learning_rate": 8.55556007930455e-05,
      "loss": 0.07115705013275146,
      "loss_step_1": 0.0775,
      "loss_step_2": 0.0762,
      "loss_step_3": 0.0856,
      "step": 9610
    },
    {
      "entropy_step_1": 0.2212,
      "entropy_step_2": 0.2255,
      "entropy_step_3": 0.222,
      "epoch": 0.2861306921269445,
      "grad_norm": 77.5483627319336,
      "learning_rate": 8.55210053871772e-05,
      "loss": 0.06238356828689575,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0508,
      "step": 9620
    },
    {
      "entropy_step_1": 0.1919,
      "entropy_step_2": 0.1953,
      "entropy_step_3": 0.1911,
      "epoch": 0.2864281252788436,
      "grad_norm": 49.57651138305664,
      "learning_rate": 8.548637561425284e-05,
      "loss": 0.07524430155754089,
      "loss_step_1": 0.0784,
      "loss_step_2": 0.078,
      "loss_step_3": 0.0733,
      "step": 9630
    },
    {
      "entropy_step_1": 0.2202,
      "entropy_step_2": 0.2293,
      "entropy_step_3": 0.2239,
      "epoch": 0.2867255584307427,
      "grad_norm": 191.51925659179688,
      "learning_rate": 8.545171150777721e-05,
      "loss": 0.062330234050750735,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.0642,
      "step": 9640
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.2225,
      "entropy_step_3": 0.2202,
      "epoch": 0.28702299158264183,
      "grad_norm": 124.46849060058594,
      "learning_rate": 8.541701310128831e-05,
      "loss": 0.06366927623748779,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0798,
      "step": 9650
    },
    {
      "entropy_step_1": 0.2402,
      "entropy_step_2": 0.2575,
      "entropy_step_3": 0.2532,
      "epoch": 0.2873204247345409,
      "grad_norm": 69.5603256225586,
      "learning_rate": 8.538228042835732e-05,
      "loss": 0.06487061977386474,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0646,
      "step": 9660
    },
    {
      "entropy_step_1": 0.2129,
      "entropy_step_2": 0.224,
      "entropy_step_3": 0.219,
      "epoch": 0.28761785788644,
      "grad_norm": 80.86583709716797,
      "learning_rate": 8.534751352258857e-05,
      "loss": 0.06842782497406005,
      "loss_step_1": 0.0742,
      "loss_step_2": 0.0729,
      "loss_step_3": 0.0638,
      "step": 9670
    },
    {
      "entropy_step_1": 0.2006,
      "entropy_step_2": 0.2097,
      "entropy_step_3": 0.2183,
      "epoch": 0.2879152910383391,
      "grad_norm": 85.11266326904297,
      "learning_rate": 8.531271241761951e-05,
      "loss": 0.060686296224594115,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0656,
      "step": 9680
    },
    {
      "entropy_step_1": 0.2148,
      "entropy_step_2": 0.2224,
      "entropy_step_3": 0.2033,
      "epoch": 0.28821272419023825,
      "grad_norm": 70.9598617553711,
      "learning_rate": 8.527787714712069e-05,
      "loss": 0.064888334274292,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0613,
      "step": 9690
    },
    {
      "entropy_step_1": 0.2144,
      "entropy_step_2": 0.2182,
      "entropy_step_3": 0.2272,
      "epoch": 0.28851015734213736,
      "grad_norm": 54.86967086791992,
      "learning_rate": 8.52430077447957e-05,
      "loss": 0.0632986605167389,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0554,
      "step": 9700
    },
    {
      "entropy_step_1": 0.209,
      "entropy_step_2": 0.2084,
      "entropy_step_3": 0.1944,
      "epoch": 0.2888075904940365,
      "grad_norm": 94.21330261230469,
      "learning_rate": 8.520810424438117e-05,
      "loss": 0.06449058651924133,
      "loss_step_1": 0.0666,
      "loss_step_2": 0.0662,
      "loss_step_3": 0.0645,
      "step": 9710
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.2052,
      "entropy_step_3": 0.2022,
      "epoch": 0.2891050236459356,
      "grad_norm": 47.05815124511719,
      "learning_rate": 8.517316667964671e-05,
      "loss": 0.06490320563316346,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0734,
      "step": 9720
    },
    {
      "entropy_step_1": 0.2227,
      "entropy_step_2": 0.2173,
      "entropy_step_3": 0.2188,
      "epoch": 0.28940245679783466,
      "grad_norm": 203.0907745361328,
      "learning_rate": 8.513819508439488e-05,
      "loss": 0.06319669485092164,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0753,
      "step": 9730
    },
    {
      "entropy_step_1": 0.2018,
      "entropy_step_2": 0.1987,
      "entropy_step_3": 0.2015,
      "epoch": 0.2896998899497338,
      "grad_norm": 71.5558090209961,
      "learning_rate": 8.51031894924612e-05,
      "loss": 0.06133904457092285,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.06,
      "step": 9740
    },
    {
      "entropy_step_1": 0.1961,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.2056,
      "epoch": 0.2899973231016329,
      "grad_norm": 155.1604461669922,
      "learning_rate": 8.506814993771401e-05,
      "loss": 0.06265974640846253,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0527,
      "step": 9750
    },
    {
      "entropy_step_1": 0.2042,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.1893,
      "epoch": 0.290294756253532,
      "grad_norm": 69.71263885498047,
      "learning_rate": 8.50330764540546e-05,
      "loss": 0.06291218400001526,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0608,
      "step": 9760
    },
    {
      "entropy_step_1": 0.229,
      "entropy_step_2": 0.2294,
      "entropy_step_3": 0.2284,
      "epoch": 0.29059218940543113,
      "grad_norm": 136.19561767578125,
      "learning_rate": 8.499796907541703e-05,
      "loss": 0.06467481255531311,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0627,
      "step": 9770
    },
    {
      "entropy_step_1": 0.209,
      "entropy_step_2": 0.2075,
      "entropy_step_3": 0.2121,
      "epoch": 0.29088962255733025,
      "grad_norm": 91.09506225585938,
      "learning_rate": 8.496282783576815e-05,
      "loss": 0.057963764667510985,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0588,
      "step": 9780
    },
    {
      "entropy_step_1": 0.2042,
      "entropy_step_2": 0.2006,
      "entropy_step_3": 0.2031,
      "epoch": 0.29118705570922937,
      "grad_norm": 61.61308288574219,
      "learning_rate": 8.492765276910762e-05,
      "loss": 0.06293242573738098,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0599,
      "step": 9790
    },
    {
      "entropy_step_1": 0.2361,
      "entropy_step_2": 0.2258,
      "entropy_step_3": 0.2463,
      "epoch": 0.2914844888611285,
      "grad_norm": 53.27970504760742,
      "learning_rate": 8.489244390946775e-05,
      "loss": 0.06272225975990295,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0723,
      "step": 9800
    },
    {
      "entropy_step_1": 0.2075,
      "entropy_step_2": 0.1953,
      "entropy_step_3": 0.1904,
      "epoch": 0.29178192201302755,
      "grad_norm": 607.2037353515625,
      "learning_rate": 8.485720129091363e-05,
      "loss": 0.06685674786567689,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0528,
      "step": 9810
    },
    {
      "entropy_step_1": 0.2161,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.2104,
      "epoch": 0.29207935516492667,
      "grad_norm": 186.9063262939453,
      "learning_rate": 8.482192494754298e-05,
      "loss": 0.07437694072723389,
      "loss_step_1": 0.0706,
      "loss_step_2": 0.0687,
      "loss_step_3": 0.0685,
      "step": 9820
    },
    {
      "entropy_step_1": 0.2405,
      "entropy_step_2": 0.2334,
      "entropy_step_3": 0.2387,
      "epoch": 0.2923767883168258,
      "grad_norm": 203.8229522705078,
      "learning_rate": 8.47866149134861e-05,
      "loss": 0.06675038933753967,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0651,
      "loss_step_3": 0.0754,
      "step": 9830
    },
    {
      "entropy_step_1": 0.2394,
      "entropy_step_2": 0.2328,
      "entropy_step_3": 0.2294,
      "epoch": 0.2926742214687249,
      "grad_norm": 49.342899322509766,
      "learning_rate": 8.475127122290595e-05,
      "loss": 0.06870197057723999,
      "loss_step_1": 0.0725,
      "loss_step_2": 0.0728,
      "loss_step_3": 0.0576,
      "step": 9840
    },
    {
      "entropy_step_1": 0.2321,
      "entropy_step_2": 0.2235,
      "entropy_step_3": 0.2148,
      "epoch": 0.292971654620624,
      "grad_norm": 370.9245300292969,
      "learning_rate": 8.471589390999804e-05,
      "loss": 0.06033860445022583,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0534,
      "step": 9850
    },
    {
      "entropy_step_1": 0.2205,
      "entropy_step_2": 0.2162,
      "entropy_step_3": 0.2254,
      "epoch": 0.29326908777252314,
      "grad_norm": 47.94429016113281,
      "learning_rate": 8.46804830089904e-05,
      "loss": 0.07355170249938965,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0636,
      "step": 9860
    },
    {
      "entropy_step_1": 0.2106,
      "entropy_step_2": 0.2001,
      "entropy_step_3": 0.2076,
      "epoch": 0.29356652092442226,
      "grad_norm": 71.81089782714844,
      "learning_rate": 8.464503855414355e-05,
      "loss": 0.060817641019821164,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0587,
      "step": 9870
    },
    {
      "entropy_step_1": 0.2263,
      "entropy_step_2": 0.2155,
      "entropy_step_3": 0.2176,
      "epoch": 0.2938639540763213,
      "grad_norm": 89.56838989257812,
      "learning_rate": 8.460956057975047e-05,
      "loss": 0.06752274632453918,
      "loss_step_1": 0.0728,
      "loss_step_2": 0.0714,
      "loss_step_3": 0.0734,
      "step": 9880
    },
    {
      "entropy_step_1": 0.2139,
      "entropy_step_2": 0.209,
      "entropy_step_3": 0.2343,
      "epoch": 0.29416138722822044,
      "grad_norm": 83.70606994628906,
      "learning_rate": 8.45740491201366e-05,
      "loss": 0.07128597497940063,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0695,
      "step": 9890
    },
    {
      "entropy_step_1": 0.1804,
      "entropy_step_2": 0.183,
      "entropy_step_3": 0.192,
      "epoch": 0.29445882038011956,
      "grad_norm": 105.40206146240234,
      "learning_rate": 8.453850420965978e-05,
      "loss": 0.060174405574798584,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.059,
      "step": 9900
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.1929,
      "entropy_step_3": 0.1872,
      "epoch": 0.2947562535320187,
      "grad_norm": 53.762779235839844,
      "learning_rate": 8.450292588271015e-05,
      "loss": 0.06946982145309448,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0567,
      "step": 9910
    },
    {
      "entropy_step_1": 0.2104,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.2222,
      "epoch": 0.2950536866839178,
      "grad_norm": 401.31219482421875,
      "learning_rate": 8.446731417371026e-05,
      "loss": 0.058839458227157596,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0734,
      "step": 9920
    },
    {
      "entropy_step_1": 0.1642,
      "entropy_step_2": 0.1579,
      "entropy_step_3": 0.1501,
      "epoch": 0.2953511198358169,
      "grad_norm": 124.99465942382812,
      "learning_rate": 8.443166911711491e-05,
      "loss": 0.06569697260856629,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0467,
      "step": 9930
    },
    {
      "entropy_step_1": 0.2222,
      "entropy_step_2": 0.2105,
      "entropy_step_3": 0.2193,
      "epoch": 0.29564855298771603,
      "grad_norm": 44.17770004272461,
      "learning_rate": 8.439599074741118e-05,
      "loss": 0.058301502466201784,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0611,
      "step": 9940
    },
    {
      "entropy_step_1": 0.1946,
      "entropy_step_2": 0.1935,
      "entropy_step_3": 0.2087,
      "epoch": 0.29594598613961515,
      "grad_norm": 147.59478759765625,
      "learning_rate": 8.436027909911837e-05,
      "loss": 0.06107291579246521,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0595,
      "step": 9950
    },
    {
      "entropy_step_1": 0.1926,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.1868,
      "epoch": 0.2962434192915142,
      "grad_norm": 175.72506713867188,
      "learning_rate": 8.432453420678799e-05,
      "loss": 0.056718140840530396,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0496,
      "step": 9960
    },
    {
      "entropy_step_1": 0.2263,
      "entropy_step_2": 0.2228,
      "entropy_step_3": 0.2181,
      "epoch": 0.2965408524434133,
      "grad_norm": 85.6613540649414,
      "learning_rate": 8.428875610500372e-05,
      "loss": 0.06343095898628234,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0546,
      "step": 9970
    },
    {
      "entropy_step_1": 0.1938,
      "entropy_step_2": 0.1979,
      "entropy_step_3": 0.1908,
      "epoch": 0.29683828559531245,
      "grad_norm": 411.4942932128906,
      "learning_rate": 8.425294482838134e-05,
      "loss": 0.06483501195907593,
      "loss_step_1": 0.062,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0595,
      "step": 9980
    },
    {
      "entropy_step_1": 0.2495,
      "entropy_step_2": 0.2558,
      "entropy_step_3": 0.2685,
      "epoch": 0.29713571874721156,
      "grad_norm": 76.49911499023438,
      "learning_rate": 8.421710041156879e-05,
      "loss": 0.06625749468803406,
      "loss_step_1": 0.0707,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0778,
      "step": 9990
    },
    {
      "entropy_step_1": 0.2039,
      "entropy_step_2": 0.2163,
      "entropy_step_3": 0.2348,
      "epoch": 0.2974331518991107,
      "grad_norm": 144.39495849609375,
      "learning_rate": 8.418122288924596e-05,
      "loss": 0.07472147345542908,
      "loss_step_1": 0.0861,
      "loss_step_2": 0.0837,
      "loss_step_3": 0.087,
      "step": 10000
    },
    {
      "epoch": 0.2974331518991107,
      "eval_accuracy": 0.836,
      "eval_f1": 0.8346607520919447,
      "eval_loss": 0.15777605772018433,
      "eval_precision": 0.84725093013642,
      "eval_recall": 0.836,
      "eval_runtime": 8.3167,
      "eval_samples_per_second": 60.12,
      "eval_steps_per_second": 7.575,
      "step": 10000
    },
    {
      "entropy_step_1": 0.2573,
      "entropy_step_2": 0.254,
      "entropy_step_3": 0.24,
      "epoch": 0.2977305850510098,
      "grad_norm": 130.3545684814453,
      "learning_rate": 8.41453122961249e-05,
      "loss": 0.06668028831481934,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0727,
      "loss_step_3": 0.0646,
      "step": 10010
    },
    {
      "entropy_step_1": 0.2448,
      "entropy_step_2": 0.244,
      "entropy_step_3": 0.2476,
      "epoch": 0.2980280182029089,
      "grad_norm": 82.08531188964844,
      "learning_rate": 8.410936866694958e-05,
      "loss": 0.07011550664901733,
      "loss_step_1": 0.0695,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0719,
      "step": 10020
    },
    {
      "entropy_step_1": 0.2373,
      "entropy_step_2": 0.2269,
      "entropy_step_3": 0.2522,
      "epoch": 0.298325451354808,
      "grad_norm": 69.09362030029297,
      "learning_rate": 8.407339203649595e-05,
      "loss": 0.06589874625205994,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0679,
      "loss_step_3": 0.0742,
      "step": 10030
    },
    {
      "entropy_step_1": 0.2141,
      "entropy_step_2": 0.2126,
      "entropy_step_3": 0.2078,
      "epoch": 0.2986228845067071,
      "grad_norm": 88.0932846069336,
      "learning_rate": 8.40373824395719e-05,
      "loss": 0.06231223344802857,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0625,
      "step": 10040
    },
    {
      "entropy_step_1": 0.221,
      "entropy_step_2": 0.2197,
      "entropy_step_3": 0.2181,
      "epoch": 0.2989203176586062,
      "grad_norm": 56.73396682739258,
      "learning_rate": 8.400133991101718e-05,
      "loss": 0.060166704654693606,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0603,
      "step": 10050
    },
    {
      "entropy_step_1": 0.2045,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.1862,
      "epoch": 0.29921775081050533,
      "grad_norm": 64.90804290771484,
      "learning_rate": 8.396526448570345e-05,
      "loss": 0.06572257280349732,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0502,
      "step": 10060
    },
    {
      "entropy_step_1": 0.2264,
      "entropy_step_2": 0.2168,
      "entropy_step_3": 0.1982,
      "epoch": 0.29951518396240445,
      "grad_norm": 73.26856231689453,
      "learning_rate": 8.392915619853416e-05,
      "loss": 0.06265195608139038,
      "loss_step_1": 0.0629,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0549,
      "step": 10070
    },
    {
      "entropy_step_1": 0.199,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1788,
      "epoch": 0.29981261711430357,
      "grad_norm": 100.97982788085938,
      "learning_rate": 8.389301508444458e-05,
      "loss": 0.06488853096961975,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0681,
      "step": 10080
    },
    {
      "entropy_step_1": 0.2388,
      "entropy_step_2": 0.2376,
      "entropy_step_3": 0.2117,
      "epoch": 0.3001100502662027,
      "grad_norm": 63.205848693847656,
      "learning_rate": 8.385684117840172e-05,
      "loss": 0.0659994900226593,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0645,
      "step": 10090
    },
    {
      "entropy_step_1": 0.2045,
      "entropy_step_2": 0.2141,
      "entropy_step_3": 0.2174,
      "epoch": 0.3004074834181018,
      "grad_norm": 94.69514465332031,
      "learning_rate": 8.382063451540432e-05,
      "loss": 0.06574773788452148,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0619,
      "step": 10100
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.1829,
      "epoch": 0.30070491657000087,
      "grad_norm": 28.456104278564453,
      "learning_rate": 8.378439513048284e-05,
      "loss": 0.04987600147724151,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0477,
      "loss_step_3": 0.0428,
      "step": 10110
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.1746,
      "epoch": 0.3010023497219,
      "grad_norm": 36.162635803222656,
      "learning_rate": 8.374812305869936e-05,
      "loss": 0.07124140858650208,
      "loss_step_1": 0.0697,
      "loss_step_2": 0.0694,
      "loss_step_3": 0.0486,
      "step": 10120
    },
    {
      "entropy_step_1": 0.2241,
      "entropy_step_2": 0.2134,
      "entropy_step_3": 0.1864,
      "epoch": 0.3012997828737991,
      "grad_norm": 33.48667526245117,
      "learning_rate": 8.371181833514762e-05,
      "loss": 0.05840499997138977,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0517,
      "step": 10130
    },
    {
      "entropy_step_1": 0.2003,
      "entropy_step_2": 0.191,
      "entropy_step_3": 0.1728,
      "epoch": 0.3015972160256982,
      "grad_norm": 40.948604583740234,
      "learning_rate": 8.367548099495292e-05,
      "loss": 0.05684501528739929,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0411,
      "step": 10140
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.1949,
      "entropy_step_3": 0.1861,
      "epoch": 0.30189464917759734,
      "grad_norm": 75.50666046142578,
      "learning_rate": 8.363911107327215e-05,
      "loss": 0.06403612494468688,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.051,
      "step": 10150
    },
    {
      "entropy_step_1": 0.2097,
      "entropy_step_2": 0.2008,
      "entropy_step_3": 0.2052,
      "epoch": 0.30219208232949646,
      "grad_norm": 473.61920166015625,
      "learning_rate": 8.36027086052937e-05,
      "loss": 0.06987258195877075,
      "loss_step_1": 0.0721,
      "loss_step_2": 0.0723,
      "loss_step_3": 0.0786,
      "step": 10160
    },
    {
      "entropy_step_1": 0.1944,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1952,
      "epoch": 0.3024895154813956,
      "grad_norm": 68.98081970214844,
      "learning_rate": 8.356627362623741e-05,
      "loss": 0.06108096241950989,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0442,
      "step": 10170
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.1929,
      "epoch": 0.3027869486332947,
      "grad_norm": 48.576507568359375,
      "learning_rate": 8.352980617135467e-05,
      "loss": 0.06008191704750061,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0538,
      "step": 10180
    },
    {
      "entropy_step_1": 0.2044,
      "entropy_step_2": 0.1985,
      "entropy_step_3": 0.2131,
      "epoch": 0.30308438178519376,
      "grad_norm": 39.27751159667969,
      "learning_rate": 8.34933062759282e-05,
      "loss": 0.05887711644172668,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0582,
      "step": 10190
    },
    {
      "entropy_step_1": 0.1866,
      "entropy_step_2": 0.1893,
      "entropy_step_3": 0.1875,
      "epoch": 0.3033818149370929,
      "grad_norm": 129.86392211914062,
      "learning_rate": 8.345677397527219e-05,
      "loss": 0.06849842071533203,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0616,
      "step": 10200
    },
    {
      "entropy_step_1": 0.2084,
      "entropy_step_2": 0.2099,
      "entropy_step_3": 0.1998,
      "epoch": 0.303679248088992,
      "grad_norm": 101.73873901367188,
      "learning_rate": 8.342020930473211e-05,
      "loss": 0.05897098779678345,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.069,
      "step": 10210
    },
    {
      "entropy_step_1": 0.2297,
      "entropy_step_2": 0.2302,
      "entropy_step_3": 0.226,
      "epoch": 0.3039766812408911,
      "grad_norm": 96.45166015625,
      "learning_rate": 8.338361229968476e-05,
      "loss": 0.06687284111976624,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.0615,
      "step": 10220
    },
    {
      "entropy_step_1": 0.193,
      "entropy_step_2": 0.1904,
      "entropy_step_3": 0.1964,
      "epoch": 0.30427411439279023,
      "grad_norm": 118.66142272949219,
      "learning_rate": 8.334698299553828e-05,
      "loss": 0.06463314294815063,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0648,
      "step": 10230
    },
    {
      "entropy_step_1": 0.2274,
      "entropy_step_2": 0.2226,
      "entropy_step_3": 0.2132,
      "epoch": 0.30457154754468935,
      "grad_norm": 57.94724655151367,
      "learning_rate": 8.3310321427732e-05,
      "loss": 0.06509120464324951,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0625,
      "step": 10240
    },
    {
      "entropy_step_1": 0.2171,
      "entropy_step_2": 0.2159,
      "entropy_step_3": 0.1874,
      "epoch": 0.30486898069658847,
      "grad_norm": 111.96568298339844,
      "learning_rate": 8.327362763173647e-05,
      "loss": 0.0679026186466217,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0599,
      "step": 10250
    },
    {
      "entropy_step_1": 0.2328,
      "entropy_step_2": 0.238,
      "entropy_step_3": 0.2339,
      "epoch": 0.3051664138484875,
      "grad_norm": 82.1718521118164,
      "learning_rate": 8.323690164305346e-05,
      "loss": 0.06727625131607055,
      "loss_step_1": 0.0718,
      "loss_step_2": 0.0714,
      "loss_step_3": 0.0694,
      "step": 10260
    },
    {
      "entropy_step_1": 0.2042,
      "entropy_step_2": 0.2169,
      "entropy_step_3": 0.2343,
      "epoch": 0.30546384700038665,
      "grad_norm": 303.3708190917969,
      "learning_rate": 8.320014349721588e-05,
      "loss": 0.06688046455383301,
      "loss_step_1": 0.0726,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0796,
      "step": 10270
    },
    {
      "entropy_step_1": 0.2258,
      "entropy_step_2": 0.2356,
      "entropy_step_3": 0.262,
      "epoch": 0.30576128015228576,
      "grad_norm": 84.61814880371094,
      "learning_rate": 8.31633532297877e-05,
      "loss": 0.06812853813171386,
      "loss_step_1": 0.07,
      "loss_step_2": 0.0693,
      "loss_step_3": 0.0816,
      "step": 10280
    },
    {
      "entropy_step_1": 0.2022,
      "entropy_step_2": 0.2118,
      "entropy_step_3": 0.1933,
      "epoch": 0.3060587133041849,
      "grad_norm": 95.28242492675781,
      "learning_rate": 8.312653087636399e-05,
      "loss": 0.05763425827026367,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0538,
      "step": 10290
    },
    {
      "entropy_step_1": 0.2384,
      "entropy_step_2": 0.2407,
      "entropy_step_3": 0.227,
      "epoch": 0.306356146456084,
      "grad_norm": 105.62196350097656,
      "learning_rate": 8.308967647257093e-05,
      "loss": 0.07593395709991455,
      "loss_step_1": 0.0768,
      "loss_step_2": 0.0756,
      "loss_step_3": 0.075,
      "step": 10300
    },
    {
      "entropy_step_1": 0.2162,
      "entropy_step_2": 0.2208,
      "entropy_step_3": 0.2046,
      "epoch": 0.3066535796079831,
      "grad_norm": 43.61630630493164,
      "learning_rate": 8.305279005406563e-05,
      "loss": 0.05479799509048462,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0561,
      "step": 10310
    },
    {
      "entropy_step_1": 0.1741,
      "entropy_step_2": 0.1816,
      "entropy_step_3": 0.1782,
      "epoch": 0.30695101275988224,
      "grad_norm": 62.40982437133789,
      "learning_rate": 8.301587165653622e-05,
      "loss": 0.06788924932479859,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0536,
      "step": 10320
    },
    {
      "entropy_step_1": 0.2152,
      "entropy_step_2": 0.22,
      "entropy_step_3": 0.2277,
      "epoch": 0.30724844591178135,
      "grad_norm": 369.60870361328125,
      "learning_rate": 8.297892131570171e-05,
      "loss": 0.05839192271232605,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0591,
      "step": 10330
    },
    {
      "entropy_step_1": 0.1754,
      "entropy_step_2": 0.1784,
      "entropy_step_3": 0.1748,
      "epoch": 0.3075458790636804,
      "grad_norm": 77.55708312988281,
      "learning_rate": 8.294193906731209e-05,
      "loss": 0.05884115099906921,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0608,
      "step": 10340
    },
    {
      "entropy_step_1": 0.2176,
      "entropy_step_2": 0.2136,
      "entropy_step_3": 0.2126,
      "epoch": 0.30784331221557953,
      "grad_norm": 60.7965087890625,
      "learning_rate": 8.290492494714819e-05,
      "loss": 0.06497877240180969,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0584,
      "step": 10350
    },
    {
      "entropy_step_1": 0.1927,
      "entropy_step_2": 0.1952,
      "entropy_step_3": 0.1801,
      "epoch": 0.30814074536747865,
      "grad_norm": 53.23594665527344,
      "learning_rate": 8.286787899102167e-05,
      "loss": 0.06649846434593201,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0651,
      "loss_step_3": 0.053,
      "step": 10360
    },
    {
      "entropy_step_1": 0.2249,
      "entropy_step_2": 0.2355,
      "entropy_step_3": 0.2316,
      "epoch": 0.30843817851937777,
      "grad_norm": 170.78245544433594,
      "learning_rate": 8.283080123477498e-05,
      "loss": 0.06211053729057312,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.0672,
      "step": 10370
    },
    {
      "entropy_step_1": 0.2089,
      "entropy_step_2": 0.2133,
      "entropy_step_3": 0.2357,
      "epoch": 0.3087356116712769,
      "grad_norm": 186.97499084472656,
      "learning_rate": 8.279369171428137e-05,
      "loss": 0.06169096231460571,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.068,
      "step": 10380
    },
    {
      "entropy_step_1": 0.2046,
      "entropy_step_2": 0.2104,
      "entropy_step_3": 0.2162,
      "epoch": 0.309033044823176,
      "grad_norm": 60.72852325439453,
      "learning_rate": 8.275655046544481e-05,
      "loss": 0.06450085043907165,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0717,
      "step": 10390
    },
    {
      "entropy_step_1": 0.19,
      "entropy_step_2": 0.1905,
      "entropy_step_3": 0.1813,
      "epoch": 0.3093304779750751,
      "grad_norm": 54.542972564697266,
      "learning_rate": 8.271937752419998e-05,
      "loss": 0.06472359895706177,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0549,
      "step": 10400
    },
    {
      "entropy_step_1": 0.2105,
      "entropy_step_2": 0.2139,
      "entropy_step_3": 0.2347,
      "epoch": 0.3096279111269742,
      "grad_norm": 80.20269775390625,
      "learning_rate": 8.268217292651215e-05,
      "loss": 0.06905775666236877,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0766,
      "step": 10410
    },
    {
      "entropy_step_1": 0.2348,
      "entropy_step_2": 0.2339,
      "entropy_step_3": 0.2445,
      "epoch": 0.3099253442788733,
      "grad_norm": 140.16928100585938,
      "learning_rate": 8.264493670837732e-05,
      "loss": 0.07143082022666931,
      "loss_step_1": 0.0737,
      "loss_step_2": 0.0723,
      "loss_step_3": 0.0794,
      "step": 10420
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1886,
      "entropy_step_3": 0.1911,
      "epoch": 0.3102227774307724,
      "grad_norm": 56.373985290527344,
      "learning_rate": 8.260766890582203e-05,
      "loss": 0.06733845472335816,
      "loss_step_1": 0.0629,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0765,
      "step": 10430
    },
    {
      "entropy_step_1": 0.2154,
      "entropy_step_2": 0.2135,
      "entropy_step_3": 0.2067,
      "epoch": 0.31052021058267154,
      "grad_norm": 60.85152816772461,
      "learning_rate": 8.257036955490337e-05,
      "loss": 0.06661420464515685,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.056,
      "step": 10440
    },
    {
      "entropy_step_1": 0.2085,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.1903,
      "epoch": 0.31081764373457066,
      "grad_norm": 50.67261505126953,
      "learning_rate": 8.253303869170898e-05,
      "loss": 0.06574174761772156,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0721,
      "step": 10450
    },
    {
      "entropy_step_1": 0.2222,
      "entropy_step_2": 0.2064,
      "entropy_step_3": 0.2019,
      "epoch": 0.3111150768864698,
      "grad_norm": 47.949344635009766,
      "learning_rate": 8.249567635235699e-05,
      "loss": 0.06870338916778565,
      "loss_step_1": 0.063,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0598,
      "step": 10460
    },
    {
      "entropy_step_1": 0.2309,
      "entropy_step_2": 0.2194,
      "entropy_step_3": 0.2235,
      "epoch": 0.3114125100383689,
      "grad_norm": 72.30645751953125,
      "learning_rate": 8.245828257299596e-05,
      "loss": 0.06749626994132996,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.0737,
      "step": 10470
    },
    {
      "entropy_step_1": 0.2241,
      "entropy_step_2": 0.2183,
      "entropy_step_3": 0.2346,
      "epoch": 0.311709943190268,
      "grad_norm": 343.047607421875,
      "learning_rate": 8.242085738980488e-05,
      "loss": 0.061993527412414554,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0684,
      "step": 10480
    },
    {
      "entropy_step_1": 0.1827,
      "entropy_step_2": 0.1853,
      "entropy_step_3": 0.2113,
      "epoch": 0.3120073763421671,
      "grad_norm": 86.44266510009766,
      "learning_rate": 8.238340083899312e-05,
      "loss": 0.05784238576889038,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0554,
      "step": 10490
    },
    {
      "entropy_step_1": 0.1982,
      "entropy_step_2": 0.201,
      "entropy_step_3": 0.2098,
      "epoch": 0.3123048094940662,
      "grad_norm": 117.95409393310547,
      "learning_rate": 8.234591295680043e-05,
      "loss": 0.06299328804016113,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0589,
      "step": 10500
    },
    {
      "entropy_step_1": 0.1958,
      "entropy_step_2": 0.1998,
      "entropy_step_3": 0.1623,
      "epoch": 0.3126022426459653,
      "grad_norm": 849.3372192382812,
      "learning_rate": 8.23083937794968e-05,
      "loss": 0.0613726794719696,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0573,
      "step": 10510
    },
    {
      "entropy_step_1": 0.2024,
      "entropy_step_2": 0.1969,
      "entropy_step_3": 0.2034,
      "epoch": 0.31289967579786443,
      "grad_norm": 85.51011657714844,
      "learning_rate": 8.227084334338258e-05,
      "loss": 0.06780948042869568,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.0738,
      "step": 10520
    },
    {
      "entropy_step_1": 0.2413,
      "entropy_step_2": 0.241,
      "entropy_step_3": 0.2487,
      "epoch": 0.31319710894976355,
      "grad_norm": 110.20674896240234,
      "learning_rate": 8.223326168478831e-05,
      "loss": 0.0668554961681366,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0674,
      "step": 10530
    },
    {
      "entropy_step_1": 0.1883,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.219,
      "epoch": 0.31349454210166267,
      "grad_norm": 122.68174743652344,
      "learning_rate": 8.219564884007478e-05,
      "loss": 0.07082007527351379,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0672,
      "step": 10540
    },
    {
      "entropy_step_1": 0.2666,
      "entropy_step_2": 0.2611,
      "entropy_step_3": 0.273,
      "epoch": 0.3137919752535618,
      "grad_norm": 115.93325805664062,
      "learning_rate": 8.21580048456329e-05,
      "loss": 0.0732377827167511,
      "loss_step_1": 0.0784,
      "loss_step_2": 0.0753,
      "loss_step_3": 0.0716,
      "step": 10550
    },
    {
      "entropy_step_1": 0.2308,
      "entropy_step_2": 0.2442,
      "entropy_step_3": 0.25,
      "epoch": 0.31408940840546085,
      "grad_norm": 104.4388427734375,
      "learning_rate": 8.212032973788375e-05,
      "loss": 0.06489856839179993,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0645,
      "step": 10560
    },
    {
      "entropy_step_1": 0.18,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.2017,
      "epoch": 0.31438684155735996,
      "grad_norm": 310.51507568359375,
      "learning_rate": 8.208262355327853e-05,
      "loss": 0.058665573596954346,
      "loss_step_1": 0.0641,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.058,
      "step": 10570
    },
    {
      "entropy_step_1": 0.2271,
      "entropy_step_2": 0.2391,
      "entropy_step_3": 0.2273,
      "epoch": 0.3146842747092591,
      "grad_norm": 35.2140998840332,
      "learning_rate": 8.204488632829848e-05,
      "loss": 0.061242449283599856,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0595,
      "step": 10580
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.1925,
      "entropy_step_3": 0.2002,
      "epoch": 0.3149817078611582,
      "grad_norm": 72.69994354248047,
      "learning_rate": 8.200711809945485e-05,
      "loss": 0.0625920832157135,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0549,
      "step": 10590
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1935,
      "entropy_step_3": 0.1749,
      "epoch": 0.3152791410130573,
      "grad_norm": 44.415435791015625,
      "learning_rate": 8.196931890328896e-05,
      "loss": 0.05836077332496643,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0503,
      "step": 10600
    },
    {
      "entropy_step_1": 0.2081,
      "entropy_step_2": 0.2052,
      "entropy_step_3": 0.1936,
      "epoch": 0.31557657416495644,
      "grad_norm": 47.238319396972656,
      "learning_rate": 8.193148877637202e-05,
      "loss": 0.06720547676086426,
      "loss_step_1": 0.0688,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0646,
      "step": 10610
    },
    {
      "entropy_step_1": 0.2237,
      "entropy_step_2": 0.2219,
      "entropy_step_3": 0.2099,
      "epoch": 0.31587400731685555,
      "grad_norm": 272.0768127441406,
      "learning_rate": 8.18936277553052e-05,
      "loss": 0.06425940990447998,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0548,
      "step": 10620
    },
    {
      "entropy_step_1": 0.2322,
      "entropy_step_2": 0.2286,
      "entropy_step_3": 0.2441,
      "epoch": 0.31617144046875467,
      "grad_norm": 150.25059509277344,
      "learning_rate": 8.185573587671955e-05,
      "loss": 0.06595197916030884,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.0757,
      "step": 10630
    },
    {
      "entropy_step_1": 0.205,
      "entropy_step_2": 0.2056,
      "entropy_step_3": 0.2426,
      "epoch": 0.31646887362065373,
      "grad_norm": 27.815929412841797,
      "learning_rate": 8.181781317727597e-05,
      "loss": 0.05849953293800354,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0762,
      "step": 10640
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1889,
      "entropy_step_3": 0.1902,
      "epoch": 0.31676630677255285,
      "grad_norm": 46.53925704956055,
      "learning_rate": 8.17798596936652e-05,
      "loss": 0.06920443177223205,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0633,
      "step": 10650
    },
    {
      "entropy_step_1": 0.2273,
      "entropy_step_2": 0.2359,
      "entropy_step_3": 0.2299,
      "epoch": 0.31706373992445197,
      "grad_norm": 167.0477294921875,
      "learning_rate": 8.174187546260776e-05,
      "loss": 0.07547817826271057,
      "loss_step_1": 0.0748,
      "loss_step_2": 0.0746,
      "loss_step_3": 0.0707,
      "step": 10660
    },
    {
      "entropy_step_1": 0.2624,
      "entropy_step_2": 0.2551,
      "entropy_step_3": 0.2392,
      "epoch": 0.3173611730763511,
      "grad_norm": 271.2654113769531,
      "learning_rate": 8.17038605208539e-05,
      "loss": 0.06799860000610351,
      "loss_step_1": 0.0708,
      "loss_step_2": 0.0698,
      "loss_step_3": 0.07,
      "step": 10670
    },
    {
      "entropy_step_1": 0.1842,
      "entropy_step_2": 0.188,
      "entropy_step_3": 0.2005,
      "epoch": 0.3176586062282502,
      "grad_norm": 266.01171875,
      "learning_rate": 8.166581490518359e-05,
      "loss": 0.06156433820724487,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0625,
      "loss_step_3": 0.0643,
      "step": 10680
    },
    {
      "entropy_step_1": 0.2081,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.2394,
      "epoch": 0.3179560393801493,
      "grad_norm": 57.6634635925293,
      "learning_rate": 8.162773865240648e-05,
      "loss": 0.05976551175117493,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.058,
      "loss_step_3": 0.056,
      "step": 10690
    },
    {
      "entropy_step_1": 0.2206,
      "entropy_step_2": 0.217,
      "entropy_step_3": 0.2247,
      "epoch": 0.31825347253204844,
      "grad_norm": 83.87332153320312,
      "learning_rate": 8.158963179936186e-05,
      "loss": 0.06252371072769165,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0678,
      "step": 10700
    },
    {
      "entropy_step_1": 0.2129,
      "entropy_step_2": 0.2044,
      "entropy_step_3": 0.2003,
      "epoch": 0.3185509056839475,
      "grad_norm": 56.13838577270508,
      "learning_rate": 8.155149438291864e-05,
      "loss": 0.06358460783958435,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0542,
      "step": 10710
    },
    {
      "entropy_step_1": 0.2182,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2226,
      "epoch": 0.3188483388358466,
      "grad_norm": 141.27499389648438,
      "learning_rate": 8.151332643997529e-05,
      "loss": 0.06995119452476502,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.079,
      "step": 10720
    },
    {
      "entropy_step_1": 0.2156,
      "entropy_step_2": 0.2033,
      "entropy_step_3": 0.2039,
      "epoch": 0.31914577198774574,
      "grad_norm": 67.13939666748047,
      "learning_rate": 8.14751280074598e-05,
      "loss": 0.06603848934173584,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0626,
      "step": 10730
    },
    {
      "entropy_step_1": 0.2082,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.1881,
      "epoch": 0.31944320513964486,
      "grad_norm": 48.34375,
      "learning_rate": 8.143689912232967e-05,
      "loss": 0.0655871331691742,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0651,
      "loss_step_3": 0.0593,
      "step": 10740
    },
    {
      "entropy_step_1": 0.2175,
      "entropy_step_2": 0.2003,
      "entropy_step_3": 0.2034,
      "epoch": 0.319740638291544,
      "grad_norm": 69.61613464355469,
      "learning_rate": 8.13986398215719e-05,
      "loss": 0.06022469997406006,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0518,
      "step": 10750
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.1876,
      "entropy_step_3": 0.1884,
      "epoch": 0.3200380714434431,
      "grad_norm": 55.05569839477539,
      "learning_rate": 8.136035014220286e-05,
      "loss": 0.057646816968917845,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.051,
      "step": 10760
    },
    {
      "entropy_step_1": 0.2198,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.2082,
      "epoch": 0.3203355045953422,
      "grad_norm": 86.15158081054688,
      "learning_rate": 8.132203012126833e-05,
      "loss": 0.06677941679954529,
      "loss_step_1": 0.0665,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0761,
      "step": 10770
    },
    {
      "entropy_step_1": 0.1987,
      "entropy_step_2": 0.1911,
      "entropy_step_3": 0.1824,
      "epoch": 0.32063293774724133,
      "grad_norm": 60.7623405456543,
      "learning_rate": 8.128367979584346e-05,
      "loss": 0.05286737680435181,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0509,
      "step": 10780
    },
    {
      "entropy_step_1": 0.2075,
      "entropy_step_2": 0.197,
      "entropy_step_3": 0.1967,
      "epoch": 0.3209303708991404,
      "grad_norm": 114.80874633789062,
      "learning_rate": 8.124529920303272e-05,
      "loss": 0.0686872959136963,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.0749,
      "loss_step_3": 0.0715,
      "step": 10790
    },
    {
      "entropy_step_1": 0.2478,
      "entropy_step_2": 0.2312,
      "entropy_step_3": 0.2547,
      "epoch": 0.3212278040510395,
      "grad_norm": 63.785728454589844,
      "learning_rate": 8.120688837996984e-05,
      "loss": 0.06656766533851624,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0814,
      "step": 10800
    },
    {
      "entropy_step_1": 0.1518,
      "entropy_step_2": 0.1526,
      "entropy_step_3": 0.1652,
      "epoch": 0.32152523720293863,
      "grad_norm": 97.82560729980469,
      "learning_rate": 8.116844736381782e-05,
      "loss": 0.06147181987762451,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0605,
      "step": 10810
    },
    {
      "entropy_step_1": 0.2336,
      "entropy_step_2": 0.2227,
      "entropy_step_3": 0.2218,
      "epoch": 0.32182267035483775,
      "grad_norm": 127.74058532714844,
      "learning_rate": 8.112997619176887e-05,
      "loss": 0.06869691014289855,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0617,
      "step": 10820
    },
    {
      "entropy_step_1": 0.2242,
      "entropy_step_2": 0.2118,
      "entropy_step_3": 0.2096,
      "epoch": 0.32212010350673687,
      "grad_norm": 202.24949645996094,
      "learning_rate": 8.109147490104437e-05,
      "loss": 0.05624974966049194,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0537,
      "step": 10830
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.1871,
      "entropy_step_3": 0.1966,
      "epoch": 0.322417536658636,
      "grad_norm": 203.7653350830078,
      "learning_rate": 8.105294352889485e-05,
      "loss": 0.06046856045722961,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0684,
      "step": 10840
    },
    {
      "entropy_step_1": 0.2217,
      "entropy_step_2": 0.2141,
      "entropy_step_3": 0.2105,
      "epoch": 0.3227149698105351,
      "grad_norm": 57.664703369140625,
      "learning_rate": 8.101438211259991e-05,
      "loss": 0.0641352117061615,
      "loss_step_1": 0.0665,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0664,
      "step": 10850
    },
    {
      "entropy_step_1": 0.2104,
      "entropy_step_2": 0.2079,
      "entropy_step_3": 0.2113,
      "epoch": 0.3230124029624342,
      "grad_norm": 38.76139450073242,
      "learning_rate": 8.097579068946828e-05,
      "loss": 0.05890857577323914,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0659,
      "step": 10860
    },
    {
      "entropy_step_1": 0.2082,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.2202,
      "epoch": 0.3233098361143333,
      "grad_norm": 153.8860626220703,
      "learning_rate": 8.093716929683767e-05,
      "loss": 0.06325961351394653,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0744,
      "step": 10870
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.2183,
      "entropy_step_3": 0.2377,
      "epoch": 0.3236072692662324,
      "grad_norm": 69.20873260498047,
      "learning_rate": 8.089851797207479e-05,
      "loss": 0.061358028650283815,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0748,
      "step": 10880
    },
    {
      "entropy_step_1": 0.2174,
      "entropy_step_2": 0.23,
      "entropy_step_3": 0.2224,
      "epoch": 0.3239047024181315,
      "grad_norm": 38.66058349609375,
      "learning_rate": 8.085983675257536e-05,
      "loss": 0.059830772876739505,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0529,
      "step": 10890
    },
    {
      "entropy_step_1": 0.2288,
      "entropy_step_2": 0.2268,
      "entropy_step_3": 0.199,
      "epoch": 0.32420213557003064,
      "grad_norm": 73.27120971679688,
      "learning_rate": 8.082112567576395e-05,
      "loss": 0.0636944055557251,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.0588,
      "step": 10900
    },
    {
      "entropy_step_1": 0.1892,
      "entropy_step_2": 0.1884,
      "entropy_step_3": 0.175,
      "epoch": 0.32449956872192975,
      "grad_norm": 60.72011184692383,
      "learning_rate": 8.078238477909406e-05,
      "loss": 0.05767539143562317,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0522,
      "step": 10910
    },
    {
      "entropy_step_1": 0.2014,
      "entropy_step_2": 0.204,
      "entropy_step_3": 0.2124,
      "epoch": 0.32479700187382887,
      "grad_norm": 111.36566925048828,
      "learning_rate": 8.074361410004804e-05,
      "loss": 0.06746325492858887,
      "loss_step_1": 0.0724,
      "loss_step_2": 0.0727,
      "loss_step_3": 0.0768,
      "step": 10920
    },
    {
      "entropy_step_1": 0.203,
      "entropy_step_2": 0.2055,
      "entropy_step_3": 0.2125,
      "epoch": 0.325094435025728,
      "grad_norm": 85.39408111572266,
      "learning_rate": 8.070481367613707e-05,
      "loss": 0.059377992153167726,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0627,
      "step": 10930
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.2042,
      "entropy_step_3": 0.2051,
      "epoch": 0.32539186817762705,
      "grad_norm": 74.50660705566406,
      "learning_rate": 8.066598354490106e-05,
      "loss": 0.06272127032279969,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0667,
      "step": 10940
    },
    {
      "entropy_step_1": 0.2193,
      "entropy_step_2": 0.2255,
      "entropy_step_3": 0.2212,
      "epoch": 0.32568930132952617,
      "grad_norm": 49.31907272338867,
      "learning_rate": 8.06271237439087e-05,
      "loss": 0.0680263102054596,
      "loss_step_1": 0.0701,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0584,
      "step": 10950
    },
    {
      "entropy_step_1": 0.1812,
      "entropy_step_2": 0.1857,
      "entropy_step_3": 0.1855,
      "epoch": 0.3259867344814253,
      "grad_norm": 65.03980255126953,
      "learning_rate": 8.058823431075737e-05,
      "loss": 0.055943351984024045,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.052,
      "step": 10960
    },
    {
      "entropy_step_1": 0.2465,
      "entropy_step_2": 0.2472,
      "entropy_step_3": 0.2428,
      "epoch": 0.3262841676333244,
      "grad_norm": 83.54413604736328,
      "learning_rate": 8.054931528307312e-05,
      "loss": 0.07291365265846253,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0624,
      "step": 10970
    },
    {
      "entropy_step_1": 0.2062,
      "entropy_step_2": 0.2074,
      "entropy_step_3": 0.2379,
      "epoch": 0.3265816007852235,
      "grad_norm": 37.91787338256836,
      "learning_rate": 8.051036669851068e-05,
      "loss": 0.05541043281555176,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0556,
      "step": 10980
    },
    {
      "entropy_step_1": 0.2111,
      "entropy_step_2": 0.2119,
      "entropy_step_3": 0.2232,
      "epoch": 0.32687903393712264,
      "grad_norm": 306.0603332519531,
      "learning_rate": 8.047138859475328e-05,
      "loss": 0.07608755230903626,
      "loss_step_1": 0.0775,
      "loss_step_2": 0.0767,
      "loss_step_3": 0.079,
      "step": 10990
    },
    {
      "entropy_step_1": 0.2079,
      "entropy_step_2": 0.2075,
      "entropy_step_3": 0.1838,
      "epoch": 0.32717646708902176,
      "grad_norm": 91.66179656982422,
      "learning_rate": 8.04323810095128e-05,
      "loss": 0.061724013090133666,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0472,
      "step": 11000
    },
    {
      "epoch": 0.32717646708902176,
      "eval_accuracy": 0.817,
      "eval_f1": 0.8151886640968128,
      "eval_loss": 0.15558840334415436,
      "eval_precision": 0.8299347624261549,
      "eval_recall": 0.817,
      "eval_runtime": 8.1585,
      "eval_samples_per_second": 61.286,
      "eval_steps_per_second": 7.722,
      "step": 11000
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.1915,
      "entropy_step_3": 0.196,
      "epoch": 0.3274739002409209,
      "grad_norm": 100.0116195678711,
      "learning_rate": 8.03933439805296e-05,
      "loss": 0.06278694868087768,
      "loss_step_1": 0.0706,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0664,
      "step": 11010
    },
    {
      "entropy_step_1": 0.2249,
      "entropy_step_2": 0.2302,
      "entropy_step_3": 0.2451,
      "epoch": 0.32777133339281994,
      "grad_norm": 104.51634979248047,
      "learning_rate": 8.035427754557255e-05,
      "loss": 0.05838364362716675,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0687,
      "step": 11020
    },
    {
      "entropy_step_1": 0.1811,
      "entropy_step_2": 0.1802,
      "entropy_step_3": 0.1809,
      "epoch": 0.32806876654471906,
      "grad_norm": 44.7346076965332,
      "learning_rate": 8.031518174243892e-05,
      "loss": 0.060868841409683225,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0574,
      "step": 11030
    },
    {
      "entropy_step_1": 0.1893,
      "entropy_step_2": 0.1887,
      "entropy_step_3": 0.1985,
      "epoch": 0.3283661996966182,
      "grad_norm": 70.50875854492188,
      "learning_rate": 8.027605660895448e-05,
      "loss": 0.058949905633926394,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0662,
      "step": 11040
    },
    {
      "entropy_step_1": 0.2442,
      "entropy_step_2": 0.2431,
      "entropy_step_3": 0.245,
      "epoch": 0.3286636328485173,
      "grad_norm": 83.5273208618164,
      "learning_rate": 8.02369021829733e-05,
      "loss": 0.06598452925682068,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0697,
      "step": 11050
    },
    {
      "entropy_step_1": 0.1927,
      "entropy_step_2": 0.1943,
      "entropy_step_3": 0.1967,
      "epoch": 0.3289610660004164,
      "grad_norm": 180.66159057617188,
      "learning_rate": 8.019771850237781e-05,
      "loss": 0.06671789288520813,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0543,
      "step": 11060
    },
    {
      "entropy_step_1": 0.2287,
      "entropy_step_2": 0.2277,
      "entropy_step_3": 0.2392,
      "epoch": 0.32925849915231553,
      "grad_norm": 51.05860900878906,
      "learning_rate": 8.015850560507879e-05,
      "loss": 0.07396566867828369,
      "loss_step_1": 0.0787,
      "loss_step_2": 0.0788,
      "loss_step_3": 0.076,
      "step": 11070
    },
    {
      "entropy_step_1": 0.2299,
      "entropy_step_2": 0.2302,
      "entropy_step_3": 0.2195,
      "epoch": 0.32955593230421465,
      "grad_norm": 87.48316192626953,
      "learning_rate": 8.011926352901522e-05,
      "loss": 0.06302722692489623,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0696,
      "loss_step_3": 0.0674,
      "step": 11080
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.2161,
      "entropy_step_3": 0.2363,
      "epoch": 0.3298533654561137,
      "grad_norm": 108.4588851928711,
      "learning_rate": 8.007999231215437e-05,
      "loss": 0.07866873145103455,
      "loss_step_1": 0.0897,
      "loss_step_2": 0.0877,
      "loss_step_3": 0.1015,
      "step": 11090
    },
    {
      "entropy_step_1": 0.2457,
      "entropy_step_2": 0.2533,
      "entropy_step_3": 0.2568,
      "epoch": 0.33015079860801283,
      "grad_norm": 53.73530578613281,
      "learning_rate": 8.004069199249163e-05,
      "loss": 0.06509339213371276,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0598,
      "step": 11100
    },
    {
      "entropy_step_1": 0.2068,
      "entropy_step_2": 0.207,
      "entropy_step_3": 0.1966,
      "epoch": 0.33044823175991195,
      "grad_norm": 181.19754028320312,
      "learning_rate": 8.000136260805065e-05,
      "loss": 0.05842333436012268,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0526,
      "step": 11110
    },
    {
      "entropy_step_1": 0.1976,
      "entropy_step_2": 0.1926,
      "entropy_step_3": 0.1885,
      "epoch": 0.33074566491181107,
      "grad_norm": 118.0971450805664,
      "learning_rate": 7.996200419688311e-05,
      "loss": 0.060911357402801514,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.047,
      "step": 11120
    },
    {
      "entropy_step_1": 0.2261,
      "entropy_step_2": 0.2207,
      "entropy_step_3": 0.2154,
      "epoch": 0.3310430980637102,
      "grad_norm": 45.44434356689453,
      "learning_rate": 7.992261679706881e-05,
      "loss": 0.0568999171257019,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0582,
      "step": 11130
    },
    {
      "entropy_step_1": 0.2342,
      "entropy_step_2": 0.2334,
      "entropy_step_3": 0.2074,
      "epoch": 0.3313405312156093,
      "grad_norm": 171.27149963378906,
      "learning_rate": 7.988320044671562e-05,
      "loss": 0.06763806939125061,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0555,
      "step": 11140
    },
    {
      "entropy_step_1": 0.2447,
      "entropy_step_2": 0.243,
      "entropy_step_3": 0.2541,
      "epoch": 0.3316379643675084,
      "grad_norm": 48.21563720703125,
      "learning_rate": 7.984375518395937e-05,
      "loss": 0.06935151219367981,
      "loss_step_1": 0.0709,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0765,
      "step": 11150
    },
    {
      "entropy_step_1": 0.1997,
      "entropy_step_2": 0.201,
      "entropy_step_3": 0.2166,
      "epoch": 0.33193539751940754,
      "grad_norm": 491.33880615234375,
      "learning_rate": 7.98042810469639e-05,
      "loss": 0.05593779683113098,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0605,
      "step": 11160
    },
    {
      "entropy_step_1": 0.2231,
      "entropy_step_2": 0.2252,
      "entropy_step_3": 0.1979,
      "epoch": 0.3322328306713066,
      "grad_norm": 46.36606979370117,
      "learning_rate": 7.976477807392096e-05,
      "loss": 0.06594794392585754,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0679,
      "loss_step_3": 0.0613,
      "step": 11170
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.2023,
      "entropy_step_3": 0.2247,
      "epoch": 0.3325302638232057,
      "grad_norm": 65.0324478149414,
      "learning_rate": 7.972524630305022e-05,
      "loss": 0.05867213010787964,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.064,
      "step": 11180
    },
    {
      "entropy_step_1": 0.2316,
      "entropy_step_2": 0.2288,
      "entropy_step_3": 0.2143,
      "epoch": 0.33282769697510484,
      "grad_norm": 28.958284378051758,
      "learning_rate": 7.968568577259922e-05,
      "loss": 0.06318922638893128,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0612,
      "step": 11190
    },
    {
      "entropy_step_1": 0.2096,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.2089,
      "epoch": 0.33312513012700395,
      "grad_norm": 164.27520751953125,
      "learning_rate": 7.964609652084331e-05,
      "loss": 0.06626777648925782,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0622,
      "step": 11200
    },
    {
      "entropy_step_1": 0.2105,
      "entropy_step_2": 0.2061,
      "entropy_step_3": 0.2105,
      "epoch": 0.3334225632789031,
      "grad_norm": 220.19888305664062,
      "learning_rate": 7.960647858608561e-05,
      "loss": 0.06933976411819458,
      "loss_step_1": 0.072,
      "loss_step_2": 0.0701,
      "loss_step_3": 0.0676,
      "step": 11210
    },
    {
      "entropy_step_1": 0.2195,
      "entropy_step_2": 0.2148,
      "entropy_step_3": 0.1902,
      "epoch": 0.3337199964308022,
      "grad_norm": 138.90872192382812,
      "learning_rate": 7.956683200665703e-05,
      "loss": 0.05956287980079651,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0535,
      "step": 11220
    },
    {
      "entropy_step_1": 0.2224,
      "entropy_step_2": 0.2152,
      "entropy_step_3": 0.2156,
      "epoch": 0.3340174295827013,
      "grad_norm": 203.13204956054688,
      "learning_rate": 7.952715682091618e-05,
      "loss": 0.07346096038818359,
      "loss_step_1": 0.0736,
      "loss_step_2": 0.0748,
      "loss_step_3": 0.0722,
      "step": 11230
    },
    {
      "entropy_step_1": 0.2346,
      "entropy_step_2": 0.2202,
      "entropy_step_3": 0.2119,
      "epoch": 0.33431486273460037,
      "grad_norm": 250.878173828125,
      "learning_rate": 7.948745306724932e-05,
      "loss": 0.06823365688323975,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0654,
      "loss_step_3": 0.0573,
      "step": 11240
    },
    {
      "entropy_step_1": 0.2263,
      "entropy_step_2": 0.2144,
      "entropy_step_3": 0.2086,
      "epoch": 0.3346122958864995,
      "grad_norm": 102.0660400390625,
      "learning_rate": 7.944772078407038e-05,
      "loss": 0.06276549696922303,
      "loss_step_1": 0.0629,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0498,
      "step": 11250
    },
    {
      "entropy_step_1": 0.213,
      "entropy_step_2": 0.2106,
      "entropy_step_3": 0.2249,
      "epoch": 0.3349097290383986,
      "grad_norm": 176.61000061035156,
      "learning_rate": 7.940796000982089e-05,
      "loss": 0.07033122777938842,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0709,
      "step": 11260
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.1999,
      "entropy_step_3": 0.2061,
      "epoch": 0.3352071621902977,
      "grad_norm": 43.74397659301758,
      "learning_rate": 7.936817078296994e-05,
      "loss": 0.06094197630882263,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0636,
      "step": 11270
    },
    {
      "entropy_step_1": 0.2281,
      "entropy_step_2": 0.222,
      "entropy_step_3": 0.2269,
      "epoch": 0.33550459534219684,
      "grad_norm": 75.70973205566406,
      "learning_rate": 7.932835314201414e-05,
      "loss": 0.06720806360244751,
      "loss_step_1": 0.0703,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0651,
      "step": 11280
    },
    {
      "entropy_step_1": 0.2264,
      "entropy_step_2": 0.2251,
      "entropy_step_3": 0.2412,
      "epoch": 0.33580202849409596,
      "grad_norm": 89.28396606445312,
      "learning_rate": 7.92885071254776e-05,
      "loss": 0.06352121829986572,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0629,
      "step": 11290
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.1976,
      "entropy_step_3": 0.2238,
      "epoch": 0.3360994616459951,
      "grad_norm": 256.72320556640625,
      "learning_rate": 7.92486327719119e-05,
      "loss": 0.06215413808822632,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0734,
      "step": 11300
    },
    {
      "entropy_step_1": 0.2228,
      "entropy_step_2": 0.2253,
      "entropy_step_3": 0.2127,
      "epoch": 0.3363968947978942,
      "grad_norm": 72.33975219726562,
      "learning_rate": 7.9208730119896e-05,
      "loss": 0.061087149381637576,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0579,
      "step": 11310
    },
    {
      "entropy_step_1": 0.2006,
      "entropy_step_2": 0.2184,
      "entropy_step_3": 0.2004,
      "epoch": 0.33669432794979326,
      "grad_norm": 215.68069458007812,
      "learning_rate": 7.916879920803628e-05,
      "loss": 0.07099650502204895,
      "loss_step_1": 0.0781,
      "loss_step_2": 0.0765,
      "loss_step_3": 0.0823,
      "step": 11320
    },
    {
      "entropy_step_1": 0.2095,
      "entropy_step_2": 0.2412,
      "entropy_step_3": 0.2219,
      "epoch": 0.3369917611016924,
      "grad_norm": 71.57310485839844,
      "learning_rate": 7.912884007496641e-05,
      "loss": 0.05579655170440674,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0441,
      "step": 11330
    },
    {
      "entropy_step_1": 0.1892,
      "entropy_step_2": 0.2161,
      "entropy_step_3": 0.2191,
      "epoch": 0.3372891942535915,
      "grad_norm": 72.81035614013672,
      "learning_rate": 7.908885275934742e-05,
      "loss": 0.06366470456123352,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0549,
      "step": 11340
    },
    {
      "entropy_step_1": 0.2284,
      "entropy_step_2": 0.239,
      "entropy_step_3": 0.2637,
      "epoch": 0.3375866274054906,
      "grad_norm": 48.11268615722656,
      "learning_rate": 7.904883729986756e-05,
      "loss": 0.06588039994239807,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0716,
      "step": 11350
    },
    {
      "entropy_step_1": 0.2357,
      "entropy_step_2": 0.2428,
      "entropy_step_3": 0.247,
      "epoch": 0.33788406055738973,
      "grad_norm": 90.3067626953125,
      "learning_rate": 7.900879373524237e-05,
      "loss": 0.07173647284507752,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.0722,
      "loss_step_3": 0.0666,
      "step": 11360
    },
    {
      "entropy_step_1": 0.2497,
      "entropy_step_2": 0.251,
      "entropy_step_3": 0.2484,
      "epoch": 0.33818149370928885,
      "grad_norm": 71.9343490600586,
      "learning_rate": 7.896872210421451e-05,
      "loss": 0.06660389304161071,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0635,
      "step": 11370
    },
    {
      "entropy_step_1": 0.2229,
      "entropy_step_2": 0.2258,
      "entropy_step_3": 0.2203,
      "epoch": 0.33847892686118797,
      "grad_norm": 186.3178253173828,
      "learning_rate": 7.892862244555384e-05,
      "loss": 0.07164438962936401,
      "loss_step_1": 0.0781,
      "loss_step_2": 0.077,
      "loss_step_3": 0.0704,
      "step": 11380
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2115,
      "entropy_step_3": 0.2257,
      "epoch": 0.3387763600130871,
      "grad_norm": 102.1776351928711,
      "learning_rate": 7.888849479805731e-05,
      "loss": 0.05562134385108948,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.063,
      "step": 11390
    },
    {
      "entropy_step_1": 0.2166,
      "entropy_step_2": 0.2313,
      "entropy_step_3": 0.2309,
      "epoch": 0.33907379316498615,
      "grad_norm": 107.02006530761719,
      "learning_rate": 7.8848339200549e-05,
      "loss": 0.06870006322860718,
      "loss_step_1": 0.0674,
      "loss_step_2": 0.0654,
      "loss_step_3": 0.078,
      "step": 11400
    },
    {
      "entropy_step_1": 0.2164,
      "entropy_step_2": 0.232,
      "entropy_step_3": 0.2376,
      "epoch": 0.33937122631688527,
      "grad_norm": 143.024658203125,
      "learning_rate": 7.880815569187996e-05,
      "loss": 0.06637955307960511,
      "loss_step_1": 0.0701,
      "loss_step_2": 0.0698,
      "loss_step_3": 0.0667,
      "step": 11410
    },
    {
      "entropy_step_1": 0.2281,
      "entropy_step_2": 0.2248,
      "entropy_step_3": 0.2389,
      "epoch": 0.3396686594687844,
      "grad_norm": 41.32217025756836,
      "learning_rate": 7.87679443109283e-05,
      "loss": 0.06849156618118286,
      "loss_step_1": 0.0723,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0633,
      "step": 11420
    },
    {
      "entropy_step_1": 0.2087,
      "entropy_step_2": 0.2044,
      "entropy_step_3": 0.2049,
      "epoch": 0.3399660926206835,
      "grad_norm": 115.17191314697266,
      "learning_rate": 7.872770509659906e-05,
      "loss": 0.06189779639244079,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0607,
      "step": 11430
    },
    {
      "entropy_step_1": 0.2054,
      "entropy_step_2": 0.2056,
      "entropy_step_3": 0.2143,
      "epoch": 0.3402635257725826,
      "grad_norm": 102.3184585571289,
      "learning_rate": 7.868743808782426e-05,
      "loss": 0.06800362467765808,
      "loss_step_1": 0.0745,
      "loss_step_2": 0.0721,
      "loss_step_3": 0.0669,
      "step": 11440
    },
    {
      "entropy_step_1": 0.2018,
      "entropy_step_2": 0.2002,
      "entropy_step_3": 0.1762,
      "epoch": 0.34056095892448174,
      "grad_norm": 63.575218200683594,
      "learning_rate": 7.864714332356276e-05,
      "loss": 0.060475319623947144,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0475,
      "step": 11450
    },
    {
      "entropy_step_1": 0.1996,
      "entropy_step_2": 0.1989,
      "entropy_step_3": 0.1885,
      "epoch": 0.34085839207638086,
      "grad_norm": 53.711769104003906,
      "learning_rate": 7.860682084280029e-05,
      "loss": 0.057867962121963504,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0547,
      "step": 11460
    },
    {
      "entropy_step_1": 0.2192,
      "entropy_step_2": 0.2225,
      "entropy_step_3": 0.233,
      "epoch": 0.3411558252282799,
      "grad_norm": 894.4238891601562,
      "learning_rate": 7.85664706845494e-05,
      "loss": 0.06622000932693481,
      "loss_step_1": 0.0683,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0802,
      "step": 11470
    },
    {
      "entropy_step_1": 0.2034,
      "entropy_step_2": 0.2146,
      "entropy_step_3": 0.2175,
      "epoch": 0.34145325838017904,
      "grad_norm": 251.9961700439453,
      "learning_rate": 7.852609288784942e-05,
      "loss": 0.06632732152938843,
      "loss_step_1": 0.0759,
      "loss_step_2": 0.0737,
      "loss_step_3": 0.0705,
      "step": 11480
    },
    {
      "entropy_step_1": 0.218,
      "entropy_step_2": 0.2233,
      "entropy_step_3": 0.2266,
      "epoch": 0.34175069153207815,
      "grad_norm": 145.34063720703125,
      "learning_rate": 7.848568749176643e-05,
      "loss": 0.05449431538581848,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.059,
      "step": 11490
    },
    {
      "entropy_step_1": 0.2012,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.2149,
      "epoch": 0.3420481246839773,
      "grad_norm": 64.36278533935547,
      "learning_rate": 7.844525453539318e-05,
      "loss": 0.06691339612007141,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.06,
      "step": 11500
    },
    {
      "entropy_step_1": 0.2309,
      "entropy_step_2": 0.2255,
      "entropy_step_3": 0.2215,
      "epoch": 0.3423455578358764,
      "grad_norm": 86.2324447631836,
      "learning_rate": 7.840479405784911e-05,
      "loss": 0.0651869535446167,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0619,
      "step": 11510
    },
    {
      "entropy_step_1": 0.2237,
      "entropy_step_2": 0.2167,
      "entropy_step_3": 0.2257,
      "epoch": 0.3426429909877755,
      "grad_norm": 119.4519271850586,
      "learning_rate": 7.836430609828026e-05,
      "loss": 0.06966202855110168,
      "loss_step_1": 0.0683,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0645,
      "step": 11520
    },
    {
      "entropy_step_1": 0.2293,
      "entropy_step_2": 0.2216,
      "entropy_step_3": 0.2238,
      "epoch": 0.3429404241396746,
      "grad_norm": 95.2335433959961,
      "learning_rate": 7.832379069585936e-05,
      "loss": 0.06442174315452576,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0622,
      "step": 11530
    },
    {
      "entropy_step_1": 0.2046,
      "entropy_step_2": 0.2019,
      "entropy_step_3": 0.2055,
      "epoch": 0.34323785729157374,
      "grad_norm": 54.017330169677734,
      "learning_rate": 7.828324788978552e-05,
      "loss": 0.06276901960372924,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.053,
      "step": 11540
    },
    {
      "entropy_step_1": 0.2114,
      "entropy_step_2": 0.1996,
      "entropy_step_3": 0.1976,
      "epoch": 0.3435352904434728,
      "grad_norm": 84.32186126708984,
      "learning_rate": 7.824267771928451e-05,
      "loss": 0.06178892850875854,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0536,
      "step": 11550
    },
    {
      "entropy_step_1": 0.2188,
      "entropy_step_2": 0.2022,
      "entropy_step_3": 0.2089,
      "epoch": 0.3438327235953719,
      "grad_norm": 46.63831329345703,
      "learning_rate": 7.820208022360851e-05,
      "loss": 0.05695030093193054,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0589,
      "step": 11560
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.1803,
      "epoch": 0.34413015674727104,
      "grad_norm": 55.31596374511719,
      "learning_rate": 7.816145544203616e-05,
      "loss": 0.0651107132434845,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0469,
      "step": 11570
    },
    {
      "entropy_step_1": 0.198,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.1859,
      "epoch": 0.34442758989917016,
      "grad_norm": 152.4857940673828,
      "learning_rate": 7.812080341387246e-05,
      "loss": 0.05988108515739441,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0507,
      "step": 11580
    },
    {
      "entropy_step_1": 0.1777,
      "entropy_step_2": 0.1836,
      "entropy_step_3": 0.1701,
      "epoch": 0.3447250230510693,
      "grad_norm": 179.49827575683594,
      "learning_rate": 7.808012417844883e-05,
      "loss": 0.05922544002532959,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0483,
      "step": 11590
    },
    {
      "entropy_step_1": 0.194,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.2005,
      "epoch": 0.3450224562029684,
      "grad_norm": 130.37608337402344,
      "learning_rate": 7.803941777512297e-05,
      "loss": 0.06236262321472168,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0489,
      "step": 11600
    },
    {
      "entropy_step_1": 0.2019,
      "entropy_step_2": 0.2039,
      "entropy_step_3": 0.2015,
      "epoch": 0.3453198893548675,
      "grad_norm": 41.52959442138672,
      "learning_rate": 7.799868424327888e-05,
      "loss": 0.06214199662208557,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0601,
      "step": 11610
    },
    {
      "entropy_step_1": 0.1835,
      "entropy_step_2": 0.1901,
      "entropy_step_3": 0.1942,
      "epoch": 0.3456173225067666,
      "grad_norm": 56.447540283203125,
      "learning_rate": 7.795792362232681e-05,
      "loss": 0.05488287210464478,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0592,
      "step": 11620
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.2255,
      "epoch": 0.3459147556586657,
      "grad_norm": 878.9700317382812,
      "learning_rate": 7.79171359517032e-05,
      "loss": 0.0677986741065979,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0674,
      "step": 11630
    },
    {
      "entropy_step_1": 0.215,
      "entropy_step_2": 0.2225,
      "entropy_step_3": 0.2085,
      "epoch": 0.3462121888105648,
      "grad_norm": 65.46953582763672,
      "learning_rate": 7.78763212708707e-05,
      "loss": 0.06206746101379394,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0603,
      "step": 11640
    },
    {
      "entropy_step_1": 0.2462,
      "entropy_step_2": 0.2437,
      "entropy_step_3": 0.2346,
      "epoch": 0.34650962196246393,
      "grad_norm": 43.83197021484375,
      "learning_rate": 7.783547961931805e-05,
      "loss": 0.06715182662010193,
      "loss_step_1": 0.072,
      "loss_step_2": 0.0711,
      "loss_step_3": 0.0755,
      "step": 11650
    },
    {
      "entropy_step_1": 0.2389,
      "entropy_step_2": 0.2369,
      "entropy_step_3": 0.2295,
      "epoch": 0.34680705511436305,
      "grad_norm": 90.4649658203125,
      "learning_rate": 7.779461103656013e-05,
      "loss": 0.0638842225074768,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0651,
      "loss_step_3": 0.0622,
      "step": 11660
    },
    {
      "entropy_step_1": 0.229,
      "entropy_step_2": 0.2333,
      "entropy_step_3": 0.2479,
      "epoch": 0.34710448826626217,
      "grad_norm": 111.4400863647461,
      "learning_rate": 7.775371556213782e-05,
      "loss": 0.06783182621002197,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0657,
      "step": 11670
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1935,
      "entropy_step_3": 0.1714,
      "epoch": 0.3474019214181613,
      "grad_norm": 71.27393341064453,
      "learning_rate": 7.771279323561804e-05,
      "loss": 0.058218193054199216,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0498,
      "step": 11680
    },
    {
      "entropy_step_1": 0.2197,
      "entropy_step_2": 0.2238,
      "entropy_step_3": 0.2419,
      "epoch": 0.3476993545700604,
      "grad_norm": 33.55841064453125,
      "learning_rate": 7.767184409659375e-05,
      "loss": 0.06762500405311585,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0603,
      "step": 11690
    },
    {
      "entropy_step_1": 0.2153,
      "entropy_step_2": 0.2167,
      "entropy_step_3": 0.1994,
      "epoch": 0.34799678772195947,
      "grad_norm": 34.928741455078125,
      "learning_rate": 7.763086818468374e-05,
      "loss": 0.06867336630821227,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0559,
      "step": 11700
    },
    {
      "entropy_step_1": 0.1903,
      "entropy_step_2": 0.1896,
      "entropy_step_3": 0.1837,
      "epoch": 0.3482942208738586,
      "grad_norm": 94.46157836914062,
      "learning_rate": 7.758986553953282e-05,
      "loss": 0.06166561245918274,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0568,
      "step": 11710
    },
    {
      "entropy_step_1": 0.219,
      "entropy_step_2": 0.2076,
      "entropy_step_3": 0.2159,
      "epoch": 0.3485916540257577,
      "grad_norm": 118.01801300048828,
      "learning_rate": 7.754883620081155e-05,
      "loss": 0.0598416805267334,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0683,
      "step": 11720
    },
    {
      "entropy_step_1": 0.2362,
      "entropy_step_2": 0.228,
      "entropy_step_3": 0.2254,
      "epoch": 0.3488890871776568,
      "grad_norm": 120.20565795898438,
      "learning_rate": 7.750778020821641e-05,
      "loss": 0.07047515511512756,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0638,
      "step": 11730
    },
    {
      "entropy_step_1": 0.2196,
      "entropy_step_2": 0.2132,
      "entropy_step_3": 0.1985,
      "epoch": 0.34918652032955594,
      "grad_norm": 41.92041778564453,
      "learning_rate": 7.746669760146961e-05,
      "loss": 0.06261159181594848,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0691,
      "step": 11740
    },
    {
      "entropy_step_1": 0.1966,
      "entropy_step_2": 0.1975,
      "entropy_step_3": 0.1902,
      "epoch": 0.34948395348145506,
      "grad_norm": 1128.0052490234375,
      "learning_rate": 7.742558842031915e-05,
      "loss": 0.06629315614700318,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0521,
      "step": 11750
    },
    {
      "entropy_step_1": 0.1979,
      "entropy_step_2": 0.1986,
      "entropy_step_3": 0.2109,
      "epoch": 0.3497813866333542,
      "grad_norm": 96.4623794555664,
      "learning_rate": 7.738445270453874e-05,
      "loss": 0.06728439331054688,
      "loss_step_1": 0.0691,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0773,
      "step": 11760
    },
    {
      "entropy_step_1": 0.2546,
      "entropy_step_2": 0.2551,
      "entropy_step_3": 0.244,
      "epoch": 0.35007881978525324,
      "grad_norm": 70.3455810546875,
      "learning_rate": 7.734329049392768e-05,
      "loss": 0.0712314248085022,
      "loss_step_1": 0.0735,
      "loss_step_2": 0.0728,
      "loss_step_3": 0.0627,
      "step": 11770
    },
    {
      "entropy_step_1": 0.1927,
      "entropy_step_2": 0.2039,
      "entropy_step_3": 0.2117,
      "epoch": 0.35037625293715235,
      "grad_norm": 402.1440124511719,
      "learning_rate": 7.730210182831103e-05,
      "loss": 0.059861356019973756,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0725,
      "step": 11780
    },
    {
      "entropy_step_1": 0.1777,
      "entropy_step_2": 0.1903,
      "entropy_step_3": 0.2073,
      "epoch": 0.3506736860890515,
      "grad_norm": 342.7561340332031,
      "learning_rate": 7.726088674753935e-05,
      "loss": 0.06413132548332215,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0662,
      "step": 11790
    },
    {
      "entropy_step_1": 0.2131,
      "entropy_step_2": 0.2151,
      "entropy_step_3": 0.2066,
      "epoch": 0.3509711192409506,
      "grad_norm": 45.42009735107422,
      "learning_rate": 7.72196452914888e-05,
      "loss": 0.05264003872871399,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0578,
      "step": 11800
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.2063,
      "entropy_step_3": 0.1908,
      "epoch": 0.3512685523928497,
      "grad_norm": 45.06092071533203,
      "learning_rate": 7.717837750006106e-05,
      "loss": 0.05871639847755432,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0496,
      "step": 11810
    },
    {
      "entropy_step_1": 0.1826,
      "entropy_step_2": 0.1791,
      "entropy_step_3": 0.1991,
      "epoch": 0.3515659855447488,
      "grad_norm": 80.90025329589844,
      "learning_rate": 7.713708341318329e-05,
      "loss": 0.05715729594230652,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0635,
      "step": 11820
    },
    {
      "entropy_step_1": 0.2128,
      "entropy_step_2": 0.2053,
      "entropy_step_3": 0.1992,
      "epoch": 0.35186341869664794,
      "grad_norm": 132.2048797607422,
      "learning_rate": 7.709576307080803e-05,
      "loss": 0.07063705921173095,
      "loss_step_1": 0.069,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0607,
      "step": 11830
    },
    {
      "entropy_step_1": 0.1938,
      "entropy_step_2": 0.1834,
      "entropy_step_3": 0.2087,
      "epoch": 0.35216085184854706,
      "grad_norm": 68.2482681274414,
      "learning_rate": 7.705441651291334e-05,
      "loss": 0.05583434104919434,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0707,
      "step": 11840
    },
    {
      "entropy_step_1": 0.1524,
      "entropy_step_2": 0.1742,
      "entropy_step_3": 0.1728,
      "epoch": 0.3524582850004461,
      "grad_norm": 50.27001190185547,
      "learning_rate": 7.701304377950253e-05,
      "loss": 0.060333603620529176,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0639,
      "step": 11850
    },
    {
      "entropy_step_1": 0.2114,
      "entropy_step_2": 0.2258,
      "entropy_step_3": 0.2237,
      "epoch": 0.35275571815234524,
      "grad_norm": 51.185791015625,
      "learning_rate": 7.69716449106043e-05,
      "loss": 0.05548994541168213,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0572,
      "step": 11860
    },
    {
      "entropy_step_1": 0.2319,
      "entropy_step_2": 0.2265,
      "entropy_step_3": 0.239,
      "epoch": 0.35305315130424436,
      "grad_norm": 97.81754302978516,
      "learning_rate": 7.69302199462726e-05,
      "loss": 0.06921311616897582,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0669,
      "step": 11870
    },
    {
      "entropy_step_1": 0.1743,
      "entropy_step_2": 0.1711,
      "entropy_step_3": 0.156,
      "epoch": 0.3533505844561435,
      "grad_norm": 107.2760238647461,
      "learning_rate": 7.688876892658668e-05,
      "loss": 0.07101151943206788,
      "loss_step_1": 0.0747,
      "loss_step_2": 0.0739,
      "loss_step_3": 0.0649,
      "step": 11880
    },
    {
      "entropy_step_1": 0.2319,
      "entropy_step_2": 0.2296,
      "entropy_step_3": 0.2205,
      "epoch": 0.3536480176080426,
      "grad_norm": 26.41079330444336,
      "learning_rate": 7.684729189165096e-05,
      "loss": 0.07167640328407288,
      "loss_step_1": 0.0731,
      "loss_step_2": 0.0727,
      "loss_step_3": 0.074,
      "step": 11890
    },
    {
      "entropy_step_1": 0.2189,
      "entropy_step_2": 0.2166,
      "entropy_step_3": 0.2056,
      "epoch": 0.3539454507599417,
      "grad_norm": 133.80679321289062,
      "learning_rate": 7.680578888159503e-05,
      "loss": 0.06136994957923889,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0493,
      "step": 11900
    },
    {
      "entropy_step_1": 0.2029,
      "entropy_step_2": 0.2025,
      "entropy_step_3": 0.1985,
      "epoch": 0.35424288391184083,
      "grad_norm": 58.64091110229492,
      "learning_rate": 7.676425993657357e-05,
      "loss": 0.06447811126708984,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0568,
      "step": 11910
    },
    {
      "entropy_step_1": 0.225,
      "entropy_step_2": 0.2243,
      "entropy_step_3": 0.2326,
      "epoch": 0.35454031706373995,
      "grad_norm": 41.892696380615234,
      "learning_rate": 7.672270509676647e-05,
      "loss": 0.05798177719116211,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0641,
      "step": 11920
    },
    {
      "entropy_step_1": 0.1796,
      "entropy_step_2": 0.1811,
      "entropy_step_3": 0.1733,
      "epoch": 0.354837750215639,
      "grad_norm": 112.64521789550781,
      "learning_rate": 7.668112440237857e-05,
      "loss": 0.04858611524105072,
      "loss_step_1": 0.0477,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0442,
      "step": 11930
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.2057,
      "entropy_step_3": 0.1765,
      "epoch": 0.35513518336753813,
      "grad_norm": 79.32926177978516,
      "learning_rate": 7.663951789363977e-05,
      "loss": 0.06212002635002136,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0581,
      "step": 11940
    },
    {
      "entropy_step_1": 0.2173,
      "entropy_step_2": 0.227,
      "entropy_step_3": 0.2413,
      "epoch": 0.35543261651943725,
      "grad_norm": 54.73219680786133,
      "learning_rate": 7.65978856108049e-05,
      "loss": 0.07039072513580322,
      "loss_step_1": 0.0724,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0635,
      "step": 11950
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.2136,
      "entropy_step_3": 0.1887,
      "epoch": 0.35573004967133637,
      "grad_norm": 100.07430267333984,
      "learning_rate": 7.65562275941538e-05,
      "loss": 0.06334390044212342,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0503,
      "step": 11960
    },
    {
      "entropy_step_1": 0.1872,
      "entropy_step_2": 0.1931,
      "entropy_step_3": 0.1862,
      "epoch": 0.3560274828232355,
      "grad_norm": 61.89216613769531,
      "learning_rate": 7.651454388399115e-05,
      "loss": 0.06435129046440125,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.067,
      "step": 11970
    },
    {
      "entropy_step_1": 0.1997,
      "entropy_step_2": 0.2004,
      "entropy_step_3": 0.2009,
      "epoch": 0.3563249159751346,
      "grad_norm": 49.83101272583008,
      "learning_rate": 7.647283452064651e-05,
      "loss": 0.055914664268493654,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0616,
      "step": 11980
    },
    {
      "entropy_step_1": 0.1848,
      "entropy_step_2": 0.1907,
      "entropy_step_3": 0.2089,
      "epoch": 0.3566223491270337,
      "grad_norm": 90.46575164794922,
      "learning_rate": 7.643109954447427e-05,
      "loss": 0.06854914426803589,
      "loss_step_1": 0.071,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0796,
      "step": 11990
    },
    {
      "entropy_step_1": 0.1964,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.2351,
      "epoch": 0.3569197822789328,
      "grad_norm": 113.8453140258789,
      "learning_rate": 7.638933899585354e-05,
      "loss": 0.06579188704490661,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0756,
      "step": 12000
    },
    {
      "epoch": 0.3569197822789328,
      "eval_accuracy": 0.853,
      "eval_f1": 0.8527524769136919,
      "eval_loss": 0.14898554980754852,
      "eval_precision": 0.8553896399389496,
      "eval_recall": 0.853,
      "eval_runtime": 7.1192,
      "eval_samples_per_second": 70.233,
      "eval_steps_per_second": 8.849,
      "step": 12000
    },
    {
      "entropy_step_1": 0.2277,
      "entropy_step_2": 0.2353,
      "entropy_step_3": 0.2315,
      "epoch": 0.3572172154308319,
      "grad_norm": 159.19505310058594,
      "learning_rate": 7.634755291518829e-05,
      "loss": 0.06629485487937928,
      "loss_step_1": 0.0701,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0636,
      "step": 12010
    },
    {
      "entropy_step_1": 0.2337,
      "entropy_step_2": 0.2383,
      "entropy_step_3": 0.2304,
      "epoch": 0.357514648582731,
      "grad_norm": 35.03372573852539,
      "learning_rate": 7.630574134290706e-05,
      "loss": 0.06636688113212585,
      "loss_step_1": 0.0692,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0578,
      "step": 12020
    },
    {
      "entropy_step_1": 0.2212,
      "entropy_step_2": 0.2212,
      "entropy_step_3": 0.2289,
      "epoch": 0.35781208173463014,
      "grad_norm": 132.68006896972656,
      "learning_rate": 7.626390431946316e-05,
      "loss": 0.07077829241752624,
      "loss_step_1": 0.0753,
      "loss_step_2": 0.0743,
      "loss_step_3": 0.0799,
      "step": 12030
    },
    {
      "entropy_step_1": 0.2062,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.1861,
      "epoch": 0.35810951488652926,
      "grad_norm": 49.4527702331543,
      "learning_rate": 7.622204188533444e-05,
      "loss": 0.05423392057418823,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0498,
      "loss_step_3": 0.0419,
      "step": 12040
    },
    {
      "entropy_step_1": 0.186,
      "entropy_step_2": 0.1819,
      "entropy_step_3": 0.1858,
      "epoch": 0.3584069480384284,
      "grad_norm": 770.1776123046875,
      "learning_rate": 7.61801540810234e-05,
      "loss": 0.06426676511764526,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0599,
      "step": 12050
    },
    {
      "entropy_step_1": 0.2058,
      "entropy_step_2": 0.2011,
      "entropy_step_3": 0.2061,
      "epoch": 0.3587043811903275,
      "grad_norm": 75.9125747680664,
      "learning_rate": 7.613824094705706e-05,
      "loss": 0.06941338777542114,
      "loss_step_1": 0.0717,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0718,
      "step": 12060
    },
    {
      "entropy_step_1": 0.2082,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.2152,
      "epoch": 0.3590018143422266,
      "grad_norm": 648.013427734375,
      "learning_rate": 7.609630252398694e-05,
      "loss": 0.06603046655654907,
      "loss_step_1": 0.0697,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0764,
      "step": 12070
    },
    {
      "entropy_step_1": 0.2229,
      "entropy_step_2": 0.2298,
      "entropy_step_3": 0.204,
      "epoch": 0.3592992474941257,
      "grad_norm": 44.174781799316406,
      "learning_rate": 7.605433885238903e-05,
      "loss": 0.062376105785369874,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0455,
      "step": 12080
    },
    {
      "entropy_step_1": 0.2011,
      "entropy_step_2": 0.2032,
      "entropy_step_3": 0.2172,
      "epoch": 0.3595966806460248,
      "grad_norm": 59.45381546020508,
      "learning_rate": 7.601234997286376e-05,
      "loss": 0.06406792402267455,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0703,
      "step": 12090
    },
    {
      "entropy_step_1": 0.2401,
      "entropy_step_2": 0.2411,
      "entropy_step_3": 0.2526,
      "epoch": 0.3598941137979239,
      "grad_norm": 70.46629333496094,
      "learning_rate": 7.597033592603595e-05,
      "loss": 0.06630628705024719,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0757,
      "step": 12100
    },
    {
      "entropy_step_1": 0.223,
      "entropy_step_2": 0.2233,
      "entropy_step_3": 0.2172,
      "epoch": 0.360191546949823,
      "grad_norm": 74.76010131835938,
      "learning_rate": 7.592829675255473e-05,
      "loss": 0.06968769431114197,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0664,
      "step": 12110
    },
    {
      "entropy_step_1": 0.2281,
      "entropy_step_2": 0.2223,
      "entropy_step_3": 0.2192,
      "epoch": 0.36048898010172215,
      "grad_norm": 82.20162963867188,
      "learning_rate": 7.588623249309364e-05,
      "loss": 0.06678667068481445,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.0582,
      "step": 12120
    },
    {
      "entropy_step_1": 0.22,
      "entropy_step_2": 0.2156,
      "entropy_step_3": 0.2182,
      "epoch": 0.36078641325362126,
      "grad_norm": 142.74465942382812,
      "learning_rate": 7.584414318835034e-05,
      "loss": 0.06140570640563965,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0679,
      "loss_step_3": 0.0631,
      "step": 12130
    },
    {
      "entropy_step_1": 0.2041,
      "entropy_step_2": 0.2067,
      "entropy_step_3": 0.1989,
      "epoch": 0.3610838464055204,
      "grad_norm": 86.99258422851562,
      "learning_rate": 7.580202887904686e-05,
      "loss": 0.06728482842445374,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.0693,
      "loss_step_3": 0.0664,
      "step": 12140
    },
    {
      "entropy_step_1": 0.2079,
      "entropy_step_2": 0.2144,
      "entropy_step_3": 0.2106,
      "epoch": 0.36138127955741944,
      "grad_norm": 74.90220642089844,
      "learning_rate": 7.575988960592934e-05,
      "loss": 0.05950371623039245,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0628,
      "step": 12150
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.197,
      "entropy_step_3": 0.1806,
      "epoch": 0.36167871270931856,
      "grad_norm": 194.65444946289062,
      "learning_rate": 7.571772540976812e-05,
      "loss": 0.055969220399856565,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0509,
      "step": 12160
    },
    {
      "entropy_step_1": 0.2046,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2197,
      "epoch": 0.3619761458612177,
      "grad_norm": 128.99603271484375,
      "learning_rate": 7.567553633135764e-05,
      "loss": 0.0625927984714508,
      "loss_step_1": 0.056,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0674,
      "step": 12170
    },
    {
      "entropy_step_1": 0.2025,
      "entropy_step_2": 0.2138,
      "entropy_step_3": 0.2251,
      "epoch": 0.3622735790131168,
      "grad_norm": 79.20844268798828,
      "learning_rate": 7.563332241151637e-05,
      "loss": 0.07046326398849487,
      "loss_step_1": 0.0744,
      "loss_step_2": 0.0718,
      "loss_step_3": 0.0782,
      "step": 12180
    },
    {
      "entropy_step_1": 0.209,
      "entropy_step_2": 0.2225,
      "entropy_step_3": 0.2227,
      "epoch": 0.3625710121650159,
      "grad_norm": 143.52027893066406,
      "learning_rate": 7.55910836910869e-05,
      "loss": 0.06484509110450745,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0644,
      "step": 12190
    },
    {
      "entropy_step_1": 0.2236,
      "entropy_step_2": 0.2329,
      "entropy_step_3": 0.243,
      "epoch": 0.36286844531691503,
      "grad_norm": 71.56117248535156,
      "learning_rate": 7.554882021093573e-05,
      "loss": 0.07158970832824707,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0711,
      "loss_step_3": 0.0688,
      "step": 12200
    },
    {
      "entropy_step_1": 0.2445,
      "entropy_step_2": 0.2467,
      "entropy_step_3": 0.2496,
      "epoch": 0.36316587846881415,
      "grad_norm": 144.6235809326172,
      "learning_rate": 7.550653201195336e-05,
      "loss": 0.06924105882644653,
      "loss_step_1": 0.0728,
      "loss_step_2": 0.0727,
      "loss_step_3": 0.0789,
      "step": 12210
    },
    {
      "entropy_step_1": 0.1939,
      "entropy_step_2": 0.1949,
      "entropy_step_3": 0.1878,
      "epoch": 0.36346331162071327,
      "grad_norm": 64.65184020996094,
      "learning_rate": 7.546421913505419e-05,
      "loss": 0.06136136054992676,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0439,
      "step": 12220
    },
    {
      "entropy_step_1": 0.2093,
      "entropy_step_2": 0.2099,
      "entropy_step_3": 0.2115,
      "epoch": 0.36376074477261233,
      "grad_norm": 106.21207427978516,
      "learning_rate": 7.542188162117654e-05,
      "loss": 0.05937599539756775,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0661,
      "step": 12230
    },
    {
      "entropy_step_1": 0.2073,
      "entropy_step_2": 0.209,
      "entropy_step_3": 0.2118,
      "epoch": 0.36405817792451145,
      "grad_norm": 106.84627532958984,
      "learning_rate": 7.53795195112825e-05,
      "loss": 0.06160744428634644,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0557,
      "step": 12240
    },
    {
      "entropy_step_1": 0.2198,
      "entropy_step_2": 0.2184,
      "entropy_step_3": 0.2135,
      "epoch": 0.36435561107641057,
      "grad_norm": 144.59327697753906,
      "learning_rate": 7.533713284635799e-05,
      "loss": 0.0596254289150238,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0583,
      "step": 12250
    },
    {
      "entropy_step_1": 0.2053,
      "entropy_step_2": 0.2142,
      "entropy_step_3": 0.2084,
      "epoch": 0.3646530442283097,
      "grad_norm": 75.3650894165039,
      "learning_rate": 7.529472166741269e-05,
      "loss": 0.06002505421638489,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0593,
      "step": 12260
    },
    {
      "entropy_step_1": 0.2418,
      "entropy_step_2": 0.2563,
      "entropy_step_3": 0.243,
      "epoch": 0.3649504773802088,
      "grad_norm": 66.29467010498047,
      "learning_rate": 7.525228601548e-05,
      "loss": 0.0754019558429718,
      "loss_step_1": 0.076,
      "loss_step_2": 0.0763,
      "loss_step_3": 0.0772,
      "step": 12270
    },
    {
      "entropy_step_1": 0.216,
      "entropy_step_2": 0.2307,
      "entropy_step_3": 0.2197,
      "epoch": 0.3652479105321079,
      "grad_norm": 165.92459106445312,
      "learning_rate": 7.520982593161698e-05,
      "loss": 0.06317379474639892,
      "loss_step_1": 0.0629,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0635,
      "step": 12280
    },
    {
      "entropy_step_1": 0.2391,
      "entropy_step_2": 0.2525,
      "entropy_step_3": 0.2453,
      "epoch": 0.36554534368400704,
      "grad_norm": 111.61275482177734,
      "learning_rate": 7.516734145690433e-05,
      "loss": 0.07092677950859069,
      "loss_step_1": 0.079,
      "loss_step_2": 0.0777,
      "loss_step_3": 0.0757,
      "step": 12290
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.2094,
      "entropy_step_3": 0.212,
      "epoch": 0.3658427768359061,
      "grad_norm": 112.26809692382812,
      "learning_rate": 7.512483263244637e-05,
      "loss": 0.06604694724082946,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0722,
      "step": 12300
    },
    {
      "entropy_step_1": 0.2306,
      "entropy_step_2": 0.2278,
      "entropy_step_3": 0.2216,
      "epoch": 0.3661402099878052,
      "grad_norm": 47.871788024902344,
      "learning_rate": 7.508229949937098e-05,
      "loss": 0.06021730899810791,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0577,
      "step": 12310
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.2091,
      "entropy_step_3": 0.2013,
      "epoch": 0.36643764313970434,
      "grad_norm": 45.88487243652344,
      "learning_rate": 7.503974209882951e-05,
      "loss": 0.06820017695426941,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.0621,
      "step": 12320
    },
    {
      "entropy_step_1": 0.2071,
      "entropy_step_2": 0.2063,
      "entropy_step_3": 0.2312,
      "epoch": 0.36673507629160346,
      "grad_norm": 51.37034225463867,
      "learning_rate": 7.499716047199687e-05,
      "loss": 0.06491823792457581,
      "loss_step_1": 0.0629,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0709,
      "step": 12330
    },
    {
      "entropy_step_1": 0.1967,
      "entropy_step_2": 0.1959,
      "entropy_step_3": 0.2022,
      "epoch": 0.3670325094435026,
      "grad_norm": 480.9462585449219,
      "learning_rate": 7.495455466007132e-05,
      "loss": 0.06630082130432129,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0654,
      "step": 12340
    },
    {
      "entropy_step_1": 0.195,
      "entropy_step_2": 0.1948,
      "entropy_step_3": 0.2307,
      "epoch": 0.3673299425954017,
      "grad_norm": 103.69374084472656,
      "learning_rate": 7.491192470427459e-05,
      "loss": 0.05902543663978577,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.063,
      "step": 12350
    },
    {
      "entropy_step_1": 0.1982,
      "entropy_step_2": 0.1981,
      "entropy_step_3": 0.1787,
      "epoch": 0.3676273757473008,
      "grad_norm": 53.189292907714844,
      "learning_rate": 7.486927064585172e-05,
      "loss": 0.0654319703578949,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0524,
      "step": 12360
    },
    {
      "entropy_step_1": 0.2177,
      "entropy_step_2": 0.2189,
      "entropy_step_3": 0.2058,
      "epoch": 0.36792480889919993,
      "grad_norm": 89.4727783203125,
      "learning_rate": 7.482659252607111e-05,
      "loss": 0.06378021240234374,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0659,
      "loss_step_3": 0.053,
      "step": 12370
    },
    {
      "entropy_step_1": 0.223,
      "entropy_step_2": 0.2238,
      "entropy_step_3": 0.2223,
      "epoch": 0.368222242051099,
      "grad_norm": 77.23822021484375,
      "learning_rate": 7.478389038622442e-05,
      "loss": 0.06213900446891785,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0699,
      "step": 12380
    },
    {
      "entropy_step_1": 0.2089,
      "entropy_step_2": 0.2078,
      "entropy_step_3": 0.2019,
      "epoch": 0.3685196752029981,
      "grad_norm": 110.51181030273438,
      "learning_rate": 7.474116426762654e-05,
      "loss": 0.06674733161926269,
      "loss_step_1": 0.0665,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0695,
      "step": 12390
    },
    {
      "entropy_step_1": 0.24,
      "entropy_step_2": 0.2367,
      "entropy_step_3": 0.2392,
      "epoch": 0.3688171083548972,
      "grad_norm": 148.4378662109375,
      "learning_rate": 7.469841421161559e-05,
      "loss": 0.06390424966812133,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0613,
      "step": 12400
    },
    {
      "entropy_step_1": 0.2028,
      "entropy_step_2": 0.2032,
      "entropy_step_3": 0.213,
      "epoch": 0.36911454150679635,
      "grad_norm": 88.94471740722656,
      "learning_rate": 7.465564025955282e-05,
      "loss": 0.06724964380264283,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.061,
      "loss_step_3": 0.069,
      "step": 12410
    },
    {
      "entropy_step_1": 0.2245,
      "entropy_step_2": 0.23,
      "entropy_step_3": 0.2244,
      "epoch": 0.36941197465869546,
      "grad_norm": 126.42184448242188,
      "learning_rate": 7.46128424528226e-05,
      "loss": 0.06182199716567993,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0624,
      "step": 12420
    },
    {
      "entropy_step_1": 0.187,
      "entropy_step_2": 0.1961,
      "entropy_step_3": 0.199,
      "epoch": 0.3697094078105946,
      "grad_norm": 36.52716827392578,
      "learning_rate": 7.457002083283242e-05,
      "loss": 0.06571245193481445,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0718,
      "step": 12430
    },
    {
      "entropy_step_1": 0.19,
      "entropy_step_2": 0.1964,
      "entropy_step_3": 0.2007,
      "epoch": 0.3700068409624937,
      "grad_norm": 91.41426849365234,
      "learning_rate": 7.452717544101274e-05,
      "loss": 0.055656588077545165,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0638,
      "step": 12440
    },
    {
      "entropy_step_1": 0.1871,
      "entropy_step_2": 0.1941,
      "entropy_step_3": 0.2175,
      "epoch": 0.3703042741143928,
      "grad_norm": 110.75361633300781,
      "learning_rate": 7.448430631881709e-05,
      "loss": 0.057427966594696046,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0693,
      "step": 12450
    },
    {
      "entropy_step_1": 0.2122,
      "entropy_step_2": 0.2199,
      "entropy_step_3": 0.2263,
      "epoch": 0.3706017072662919,
      "grad_norm": 86.78398895263672,
      "learning_rate": 7.444141350772191e-05,
      "loss": 0.05401798486709595,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0612,
      "step": 12460
    },
    {
      "entropy_step_1": 0.1747,
      "entropy_step_2": 0.1829,
      "entropy_step_3": 0.1669,
      "epoch": 0.370899140418191,
      "grad_norm": 214.9566650390625,
      "learning_rate": 7.43984970492266e-05,
      "loss": 0.06672013998031616,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0705,
      "step": 12470
    },
    {
      "entropy_step_1": 0.2053,
      "entropy_step_2": 0.2107,
      "entropy_step_3": 0.2018,
      "epoch": 0.3711965735700901,
      "grad_norm": 214.8955841064453,
      "learning_rate": 7.43555569848534e-05,
      "loss": 0.06413938403129578,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0625,
      "loss_step_3": 0.0515,
      "step": 12480
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.2098,
      "entropy_step_3": 0.2184,
      "epoch": 0.37149400672198923,
      "grad_norm": 73.98046875,
      "learning_rate": 7.431259335614742e-05,
      "loss": 0.06900299191474915,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.07,
      "step": 12490
    },
    {
      "entropy_step_1": 0.1979,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.2199,
      "epoch": 0.37179143987388835,
      "grad_norm": 76.68579864501953,
      "learning_rate": 7.426960620467652e-05,
      "loss": 0.07192163467407227,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0678,
      "step": 12500
    },
    {
      "entropy_step_1": 0.1957,
      "entropy_step_2": 0.1963,
      "entropy_step_3": 0.187,
      "epoch": 0.37208887302578747,
      "grad_norm": 684.3937377929688,
      "learning_rate": 7.422659557203141e-05,
      "loss": 0.05805978775024414,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0507,
      "step": 12510
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1737,
      "entropy_step_3": 0.1505,
      "epoch": 0.3723863061776866,
      "grad_norm": 94.70181274414062,
      "learning_rate": 7.418356149982541e-05,
      "loss": 0.06397337913513183,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0516,
      "step": 12520
    },
    {
      "entropy_step_1": 0.1987,
      "entropy_step_2": 0.1991,
      "entropy_step_3": 0.2022,
      "epoch": 0.37268373932958565,
      "grad_norm": 615.0824584960938,
      "learning_rate": 7.414050402969461e-05,
      "loss": 0.05887655615806579,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0597,
      "step": 12530
    },
    {
      "entropy_step_1": 0.1769,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1627,
      "epoch": 0.37298117248148477,
      "grad_norm": 177.86070251464844,
      "learning_rate": 7.409742320329769e-05,
      "loss": 0.06348645687103271,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0541,
      "step": 12540
    },
    {
      "entropy_step_1": 0.1956,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.1809,
      "epoch": 0.3732786056333839,
      "grad_norm": 32.177284240722656,
      "learning_rate": 7.405431906231591e-05,
      "loss": 0.05370649695396423,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0502,
      "step": 12550
    },
    {
      "entropy_step_1": 0.202,
      "entropy_step_2": 0.2006,
      "entropy_step_3": 0.1928,
      "epoch": 0.373576038785283,
      "grad_norm": 279.7656555175781,
      "learning_rate": 7.401119164845314e-05,
      "loss": 0.0697041094303131,
      "loss_step_1": 0.0743,
      "loss_step_2": 0.073,
      "loss_step_3": 0.0773,
      "step": 12560
    },
    {
      "entropy_step_1": 0.2,
      "entropy_step_2": 0.2009,
      "entropy_step_3": 0.2145,
      "epoch": 0.3738734719371821,
      "grad_norm": 45.52028274536133,
      "learning_rate": 7.396804100343573e-05,
      "loss": 0.06199411153793335,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0534,
      "step": 12570
    },
    {
      "entropy_step_1": 0.1889,
      "entropy_step_2": 0.1917,
      "entropy_step_3": 0.1959,
      "epoch": 0.37417090508908124,
      "grad_norm": 63.376914978027344,
      "learning_rate": 7.39248671690125e-05,
      "loss": 0.06318487524986267,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0483,
      "step": 12580
    },
    {
      "entropy_step_1": 0.1937,
      "entropy_step_2": 0.1949,
      "entropy_step_3": 0.2004,
      "epoch": 0.37446833824098036,
      "grad_norm": 185.25933837890625,
      "learning_rate": 7.388167018695477e-05,
      "loss": 0.0634133517742157,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0601,
      "step": 12590
    },
    {
      "entropy_step_1": 0.2353,
      "entropy_step_2": 0.2382,
      "entropy_step_3": 0.2184,
      "epoch": 0.3747657713928795,
      "grad_norm": 85.39836883544922,
      "learning_rate": 7.383845009905615e-05,
      "loss": 0.05718278884887695,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0577,
      "step": 12600
    },
    {
      "entropy_step_1": 0.1602,
      "entropy_step_2": 0.1635,
      "entropy_step_3": 0.1771,
      "epoch": 0.37506320454477854,
      "grad_norm": 113.29109954833984,
      "learning_rate": 7.379520694713271e-05,
      "loss": 0.061281800270080566,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.066,
      "step": 12610
    },
    {
      "entropy_step_1": 0.1983,
      "entropy_step_2": 0.2021,
      "entropy_step_3": 0.2112,
      "epoch": 0.37536063769667766,
      "grad_norm": 199.37380981445312,
      "learning_rate": 7.375194077302276e-05,
      "loss": 0.058479595184326175,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0565,
      "step": 12620
    },
    {
      "entropy_step_1": 0.1894,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.2034,
      "epoch": 0.3756580708485768,
      "grad_norm": 401.7574462890625,
      "learning_rate": 7.370865161858692e-05,
      "loss": 0.0652044415473938,
      "loss_step_1": 0.0726,
      "loss_step_2": 0.0719,
      "loss_step_3": 0.0788,
      "step": 12630
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.1904,
      "epoch": 0.3759555040004759,
      "grad_norm": 136.4241943359375,
      "learning_rate": 7.366533952570804e-05,
      "loss": 0.060348278284072875,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0557,
      "step": 12640
    },
    {
      "entropy_step_1": 0.2378,
      "entropy_step_2": 0.2382,
      "entropy_step_3": 0.2391,
      "epoch": 0.376252937152375,
      "grad_norm": 68.58273315429688,
      "learning_rate": 7.362200453629115e-05,
      "loss": 0.06567624807357789,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0722,
      "step": 12650
    },
    {
      "entropy_step_1": 0.1912,
      "entropy_step_2": 0.1948,
      "entropy_step_3": 0.2005,
      "epoch": 0.37655037030427413,
      "grad_norm": 197.44924926757812,
      "learning_rate": 7.357864669226344e-05,
      "loss": 0.05782486200332641,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0541,
      "step": 12660
    },
    {
      "entropy_step_1": 0.2138,
      "entropy_step_2": 0.2181,
      "entropy_step_3": 0.2114,
      "epoch": 0.37684780345617325,
      "grad_norm": 172.15032958984375,
      "learning_rate": 7.353526603557423e-05,
      "loss": 0.061404180526733396,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0592,
      "step": 12670
    },
    {
      "entropy_step_1": 0.2166,
      "entropy_step_2": 0.2191,
      "entropy_step_3": 0.2246,
      "epoch": 0.3771452366080723,
      "grad_norm": 95.06825256347656,
      "learning_rate": 7.349186260819488e-05,
      "loss": 0.06519854068756104,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.065,
      "step": 12680
    },
    {
      "entropy_step_1": 0.2022,
      "entropy_step_2": 0.2029,
      "entropy_step_3": 0.2123,
      "epoch": 0.3774426697599714,
      "grad_norm": 82.9030990600586,
      "learning_rate": 7.344843645211881e-05,
      "loss": 0.05735411047935486,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0566,
      "step": 12690
    },
    {
      "entropy_step_1": 0.1945,
      "entropy_step_2": 0.198,
      "entropy_step_3": 0.1911,
      "epoch": 0.37774010291187055,
      "grad_norm": 329.54986572265625,
      "learning_rate": 7.340498760936142e-05,
      "loss": 0.07412059307098388,
      "loss_step_1": 0.0691,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0554,
      "step": 12700
    },
    {
      "entropy_step_1": 0.2034,
      "entropy_step_2": 0.2094,
      "entropy_step_3": 0.2074,
      "epoch": 0.37803753606376966,
      "grad_norm": 140.10354614257812,
      "learning_rate": 7.336151612196002e-05,
      "loss": 0.060083305835723876,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0552,
      "step": 12710
    },
    {
      "entropy_step_1": 0.1886,
      "entropy_step_2": 0.1937,
      "entropy_step_3": 0.212,
      "epoch": 0.3783349692156688,
      "grad_norm": 84.42156219482422,
      "learning_rate": 7.331802203197392e-05,
      "loss": 0.058882641792297366,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0617,
      "step": 12720
    },
    {
      "entropy_step_1": 0.2044,
      "entropy_step_2": 0.205,
      "entropy_step_3": 0.1911,
      "epoch": 0.3786324023675679,
      "grad_norm": 34.579490661621094,
      "learning_rate": 7.327450538148421e-05,
      "loss": 0.06235569715499878,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0524,
      "step": 12730
    },
    {
      "entropy_step_1": 0.2227,
      "entropy_step_2": 0.2234,
      "entropy_step_3": 0.2207,
      "epoch": 0.378929835519467,
      "grad_norm": 86.15076446533203,
      "learning_rate": 7.323096621259385e-05,
      "loss": 0.07191731333732605,
      "loss_step_1": 0.0718,
      "loss_step_2": 0.0704,
      "loss_step_3": 0.0615,
      "step": 12740
    },
    {
      "entropy_step_1": 0.2081,
      "entropy_step_2": 0.2112,
      "entropy_step_3": 0.2076,
      "epoch": 0.37922726867136614,
      "grad_norm": 62.790164947509766,
      "learning_rate": 7.318740456742756e-05,
      "loss": 0.061525607109069826,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.059,
      "step": 12750
    },
    {
      "entropy_step_1": 0.2181,
      "entropy_step_2": 0.2199,
      "entropy_step_3": 0.2483,
      "epoch": 0.3795247018232652,
      "grad_norm": 42.960330963134766,
      "learning_rate": 7.314382048813185e-05,
      "loss": 0.06485064029693603,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.0722,
      "loss_step_3": 0.0813,
      "step": 12760
    },
    {
      "entropy_step_1": 0.2276,
      "entropy_step_2": 0.2241,
      "entropy_step_3": 0.2248,
      "epoch": 0.3798221349751643,
      "grad_norm": 21.41311264038086,
      "learning_rate": 7.31002140168749e-05,
      "loss": 0.05862925052642822,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0711,
      "step": 12770
    },
    {
      "entropy_step_1": 0.2036,
      "entropy_step_2": 0.2009,
      "entropy_step_3": 0.2024,
      "epoch": 0.38011956812706343,
      "grad_norm": 88.58800506591797,
      "learning_rate": 7.305658519584657e-05,
      "loss": 0.06730713844299316,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0688,
      "loss_step_3": 0.0685,
      "step": 12780
    },
    {
      "entropy_step_1": 0.2679,
      "entropy_step_2": 0.2572,
      "entropy_step_3": 0.2778,
      "epoch": 0.38041700127896255,
      "grad_norm": 55.101078033447266,
      "learning_rate": 7.301293406725833e-05,
      "loss": 0.06809641122817993,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0679,
      "loss_step_3": 0.0692,
      "step": 12790
    },
    {
      "entropy_step_1": 0.2195,
      "entropy_step_2": 0.2196,
      "entropy_step_3": 0.2157,
      "epoch": 0.38071443443086167,
      "grad_norm": 104.04218292236328,
      "learning_rate": 7.296926067334325e-05,
      "loss": 0.0671287715435028,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.0781,
      "step": 12800
    },
    {
      "entropy_step_1": 0.2229,
      "entropy_step_2": 0.2258,
      "entropy_step_3": 0.239,
      "epoch": 0.3810118675827608,
      "grad_norm": 111.86653900146484,
      "learning_rate": 7.292556505635591e-05,
      "loss": 0.061053919792175296,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0645,
      "step": 12810
    },
    {
      "entropy_step_1": 0.1932,
      "entropy_step_2": 0.1983,
      "entropy_step_3": 0.2055,
      "epoch": 0.3813093007346599,
      "grad_norm": 125.02615356445312,
      "learning_rate": 7.288184725857243e-05,
      "loss": 0.06049726009368896,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0655,
      "step": 12820
    },
    {
      "entropy_step_1": 0.2036,
      "entropy_step_2": 0.2057,
      "entropy_step_3": 0.2055,
      "epoch": 0.38160673388655897,
      "grad_norm": 47.464317321777344,
      "learning_rate": 7.28381073222904e-05,
      "loss": 0.06546868085861206,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.057,
      "loss_step_3": 0.062,
      "step": 12830
    },
    {
      "entropy_step_1": 0.2075,
      "entropy_step_2": 0.2084,
      "entropy_step_3": 0.1957,
      "epoch": 0.3819041670384581,
      "grad_norm": 84.94876861572266,
      "learning_rate": 7.279434528982876e-05,
      "loss": 0.05533039569854736,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0533,
      "step": 12840
    },
    {
      "entropy_step_1": 0.1764,
      "entropy_step_2": 0.1826,
      "entropy_step_3": 0.1858,
      "epoch": 0.3822016001903572,
      "grad_norm": 331.85601806640625,
      "learning_rate": 7.275056120352788e-05,
      "loss": 0.07012980580329894,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.0746,
      "step": 12850
    },
    {
      "entropy_step_1": 0.2098,
      "entropy_step_2": 0.2195,
      "entropy_step_3": 0.209,
      "epoch": 0.3824990333422563,
      "grad_norm": 61.35006332397461,
      "learning_rate": 7.270675510574949e-05,
      "loss": 0.07596932053565979,
      "loss_step_1": 0.0803,
      "loss_step_2": 0.079,
      "loss_step_3": 0.0701,
      "step": 12860
    },
    {
      "entropy_step_1": 0.2311,
      "entropy_step_2": 0.2413,
      "entropy_step_3": 0.2518,
      "epoch": 0.38279646649415544,
      "grad_norm": 94.51679992675781,
      "learning_rate": 7.266292703887655e-05,
      "loss": 0.07507551908493042,
      "loss_step_1": 0.0765,
      "loss_step_2": 0.0758,
      "loss_step_3": 0.0797,
      "step": 12870
    },
    {
      "entropy_step_1": 0.2019,
      "entropy_step_2": 0.2159,
      "entropy_step_3": 0.2212,
      "epoch": 0.38309389964605456,
      "grad_norm": 76.62822723388672,
      "learning_rate": 7.261907704531334e-05,
      "loss": 0.058053618669509886,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0555,
      "step": 12880
    },
    {
      "entropy_step_1": 0.2027,
      "entropy_step_2": 0.2169,
      "entropy_step_3": 0.2283,
      "epoch": 0.3833913327979537,
      "grad_norm": 258.05615234375,
      "learning_rate": 7.257520516748531e-05,
      "loss": 0.06582584977149963,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0708,
      "step": 12890
    },
    {
      "entropy_step_1": 0.2028,
      "entropy_step_2": 0.216,
      "entropy_step_3": 0.2127,
      "epoch": 0.3836887659498528,
      "grad_norm": 70.6363754272461,
      "learning_rate": 7.253131144783909e-05,
      "loss": 0.06337296962738037,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.054,
      "step": 12900
    },
    {
      "entropy_step_1": 0.2236,
      "entropy_step_2": 0.233,
      "entropy_step_3": 0.2398,
      "epoch": 0.38398619910175186,
      "grad_norm": 108.35055541992188,
      "learning_rate": 7.248739592884246e-05,
      "loss": 0.062057167291641235,
      "loss_step_1": 0.0619,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0619,
      "step": 12910
    },
    {
      "entropy_step_1": 0.2184,
      "entropy_step_2": 0.2161,
      "entropy_step_3": 0.2002,
      "epoch": 0.384283632253651,
      "grad_norm": 80.0225830078125,
      "learning_rate": 7.24434586529843e-05,
      "loss": 0.0682828664779663,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0613,
      "step": 12920
    },
    {
      "entropy_step_1": 0.2211,
      "entropy_step_2": 0.2195,
      "entropy_step_3": 0.2144,
      "epoch": 0.3845810654055501,
      "grad_norm": 41.26567077636719,
      "learning_rate": 7.239949966277449e-05,
      "loss": 0.07085387110710144,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0604,
      "step": 12930
    },
    {
      "entropy_step_1": 0.1874,
      "entropy_step_2": 0.1903,
      "entropy_step_3": 0.1936,
      "epoch": 0.3848784985574492,
      "grad_norm": 169.661865234375,
      "learning_rate": 7.235551900074396e-05,
      "loss": 0.056627309322357176,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0505,
      "step": 12940
    },
    {
      "entropy_step_1": 0.2142,
      "entropy_step_2": 0.2098,
      "entropy_step_3": 0.2139,
      "epoch": 0.38517593170934833,
      "grad_norm": 161.57020568847656,
      "learning_rate": 7.231151670944463e-05,
      "loss": 0.06860252618789672,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0755,
      "step": 12950
    },
    {
      "entropy_step_1": 0.2474,
      "entropy_step_2": 0.2409,
      "entropy_step_3": 0.2259,
      "epoch": 0.38547336486124745,
      "grad_norm": 291.9860534667969,
      "learning_rate": 7.226749283144927e-05,
      "loss": 0.06443529725074768,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0545,
      "step": 12960
    },
    {
      "entropy_step_1": 0.2188,
      "entropy_step_2": 0.2099,
      "entropy_step_3": 0.1805,
      "epoch": 0.38577079801314657,
      "grad_norm": 81.41329956054688,
      "learning_rate": 7.222344740935162e-05,
      "loss": 0.06542131304740906,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0449,
      "step": 12970
    },
    {
      "entropy_step_1": 0.1806,
      "entropy_step_2": 0.1762,
      "entropy_step_3": 0.1683,
      "epoch": 0.3860682311650457,
      "grad_norm": 163.16940307617188,
      "learning_rate": 7.21793804857662e-05,
      "loss": 0.06589994430541993,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0492,
      "step": 12980
    },
    {
      "entropy_step_1": 0.2442,
      "entropy_step_2": 0.2372,
      "entropy_step_3": 0.2383,
      "epoch": 0.38636566431694475,
      "grad_norm": 67.3374252319336,
      "learning_rate": 7.213529210332837e-05,
      "loss": 0.07629554271697998,
      "loss_step_1": 0.0764,
      "loss_step_2": 0.075,
      "loss_step_3": 0.0652,
      "step": 12990
    },
    {
      "entropy_step_1": 0.2041,
      "entropy_step_2": 0.2015,
      "entropy_step_3": 0.1779,
      "epoch": 0.38666309746884386,
      "grad_norm": 105.33760070800781,
      "learning_rate": 7.209118230469423e-05,
      "loss": 0.0622505784034729,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0538,
      "step": 13000
    },
    {
      "epoch": 0.38666309746884386,
      "eval_accuracy": 0.794,
      "eval_f1": 0.7916590814390492,
      "eval_loss": 0.17627739906311035,
      "eval_precision": 0.8078353520631252,
      "eval_recall": 0.794,
      "eval_runtime": 6.8156,
      "eval_samples_per_second": 73.361,
      "eval_steps_per_second": 9.243,
      "step": 13000
    },
    {
      "entropy_step_1": 0.1954,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.1878,
      "epoch": 0.386960530620743,
      "grad_norm": 57.4433479309082,
      "learning_rate": 7.204705113254061e-05,
      "loss": 0.06783888936042785,
      "loss_step_1": 0.0727,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0718,
      "step": 13010
    },
    {
      "entropy_step_1": 0.2,
      "entropy_step_2": 0.1968,
      "entropy_step_3": 0.2117,
      "epoch": 0.3872579637726421,
      "grad_norm": 278.66668701171875,
      "learning_rate": 7.200289862956504e-05,
      "loss": 0.0600049614906311,
      "loss_step_1": 0.064,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0681,
      "step": 13020
    },
    {
      "entropy_step_1": 0.1949,
      "entropy_step_2": 0.1958,
      "entropy_step_3": 0.1983,
      "epoch": 0.3875553969245412,
      "grad_norm": 171.44386291503906,
      "learning_rate": 7.195872483848565e-05,
      "loss": 0.06795578598976135,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0566,
      "step": 13030
    },
    {
      "entropy_step_1": 0.2131,
      "entropy_step_2": 0.2174,
      "entropy_step_3": 0.2189,
      "epoch": 0.38785283007644034,
      "grad_norm": 38.26158905029297,
      "learning_rate": 7.191452980204119e-05,
      "loss": 0.06286602616310119,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.068,
      "step": 13040
    },
    {
      "entropy_step_1": 0.1971,
      "entropy_step_2": 0.1964,
      "entropy_step_3": 0.212,
      "epoch": 0.38815026322833945,
      "grad_norm": 34.82015609741211,
      "learning_rate": 7.187031356299094e-05,
      "loss": 0.06509606838226319,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0646,
      "step": 13050
    },
    {
      "entropy_step_1": 0.18,
      "entropy_step_2": 0.1607,
      "entropy_step_3": 0.1554,
      "epoch": 0.3884476963802385,
      "grad_norm": 74.34246826171875,
      "learning_rate": 7.182607616411474e-05,
      "loss": 0.05425759553909302,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0414,
      "step": 13060
    },
    {
      "entropy_step_1": 0.1904,
      "entropy_step_2": 0.1641,
      "entropy_step_3": 0.1566,
      "epoch": 0.38874512953213763,
      "grad_norm": 44.00335693359375,
      "learning_rate": 7.178181764821286e-05,
      "loss": 0.05658444166183472,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0505,
      "step": 13070
    },
    {
      "entropy_step_1": 0.2295,
      "entropy_step_2": 0.2078,
      "entropy_step_3": 0.2134,
      "epoch": 0.38904256268403675,
      "grad_norm": 110.22832489013672,
      "learning_rate": 7.173753805810601e-05,
      "loss": 0.06630017757415771,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0708,
      "step": 13080
    },
    {
      "entropy_step_1": 0.217,
      "entropy_step_2": 0.2007,
      "entropy_step_3": 0.21,
      "epoch": 0.38933999583593587,
      "grad_norm": 57.86138153076172,
      "learning_rate": 7.169323743663533e-05,
      "loss": 0.06726662516593933,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0788,
      "step": 13090
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1697,
      "epoch": 0.389637428987835,
      "grad_norm": 91.0728988647461,
      "learning_rate": 7.164891582666224e-05,
      "loss": 0.06098896861076355,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0571,
      "step": 13100
    },
    {
      "entropy_step_1": 0.2054,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.1915,
      "epoch": 0.3899348621397341,
      "grad_norm": 81.67881774902344,
      "learning_rate": 7.160457327106852e-05,
      "loss": 0.06114634871482849,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0553,
      "step": 13110
    },
    {
      "entropy_step_1": 0.2234,
      "entropy_step_2": 0.2179,
      "entropy_step_3": 0.222,
      "epoch": 0.3902322952916332,
      "grad_norm": 57.391056060791016,
      "learning_rate": 7.156020981275617e-05,
      "loss": 0.05953074097633362,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0602,
      "step": 13120
    },
    {
      "entropy_step_1": 0.2279,
      "entropy_step_2": 0.2226,
      "entropy_step_3": 0.2073,
      "epoch": 0.39052972844353234,
      "grad_norm": 55.379268646240234,
      "learning_rate": 7.151582549464747e-05,
      "loss": 0.06862240433692932,
      "loss_step_1": 0.0679,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0627,
      "step": 13130
    },
    {
      "entropy_step_1": 0.183,
      "entropy_step_2": 0.1777,
      "entropy_step_3": 0.1698,
      "epoch": 0.3908271615954314,
      "grad_norm": 116.97119140625,
      "learning_rate": 7.147142035968484e-05,
      "loss": 0.06714056134223938,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0624,
      "step": 13140
    },
    {
      "entropy_step_1": 0.2239,
      "entropy_step_2": 0.2181,
      "entropy_step_3": 0.2335,
      "epoch": 0.3911245947473305,
      "grad_norm": 45.42938995361328,
      "learning_rate": 7.142699445083085e-05,
      "loss": 0.05848774313926697,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.058,
      "step": 13150
    },
    {
      "entropy_step_1": 0.2182,
      "entropy_step_2": 0.2152,
      "entropy_step_3": 0.2276,
      "epoch": 0.39142202789922964,
      "grad_norm": 115.80217742919922,
      "learning_rate": 7.138254781106817e-05,
      "loss": 0.07876406908035279,
      "loss_step_1": 0.0844,
      "loss_step_2": 0.0841,
      "loss_step_3": 0.0987,
      "step": 13160
    },
    {
      "entropy_step_1": 0.2614,
      "entropy_step_2": 0.2597,
      "entropy_step_3": 0.2362,
      "epoch": 0.39171946105112876,
      "grad_norm": 57.44655227661133,
      "learning_rate": 7.133808048339953e-05,
      "loss": 0.06986763477325439,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0542,
      "step": 13170
    },
    {
      "entropy_step_1": 0.1888,
      "entropy_step_2": 0.1916,
      "entropy_step_3": 0.1876,
      "epoch": 0.3920168942030279,
      "grad_norm": 94.52568054199219,
      "learning_rate": 7.129359251084769e-05,
      "loss": 0.06475082635879517,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0588,
      "step": 13180
    },
    {
      "entropy_step_1": 0.2119,
      "entropy_step_2": 0.2089,
      "entropy_step_3": 0.1967,
      "epoch": 0.392314327354927,
      "grad_norm": 62.7019157409668,
      "learning_rate": 7.124908393645533e-05,
      "loss": 0.06729393601417541,
      "loss_step_1": 0.0697,
      "loss_step_2": 0.0695,
      "loss_step_3": 0.0627,
      "step": 13190
    },
    {
      "entropy_step_1": 0.2032,
      "entropy_step_2": 0.2034,
      "entropy_step_3": 0.1844,
      "epoch": 0.3926117605068261,
      "grad_norm": 73.9620590209961,
      "learning_rate": 7.120455480328515e-05,
      "loss": 0.06033282279968262,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0597,
      "step": 13200
    },
    {
      "entropy_step_1": 0.2288,
      "entropy_step_2": 0.2382,
      "entropy_step_3": 0.239,
      "epoch": 0.3929091936587252,
      "grad_norm": 81.50350952148438,
      "learning_rate": 7.116000515441965e-05,
      "loss": 0.07063723802566528,
      "loss_step_1": 0.0783,
      "loss_step_2": 0.0771,
      "loss_step_3": 0.0698,
      "step": 13210
    },
    {
      "entropy_step_1": 0.2234,
      "entropy_step_2": 0.2403,
      "entropy_step_3": 0.2422,
      "epoch": 0.3932066268106243,
      "grad_norm": 1252.5675048828125,
      "learning_rate": 7.111543503296123e-05,
      "loss": 0.07774989604949951,
      "loss_step_1": 0.0674,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0676,
      "step": 13220
    },
    {
      "entropy_step_1": 0.2227,
      "entropy_step_2": 0.2326,
      "entropy_step_3": 0.2278,
      "epoch": 0.3935040599625234,
      "grad_norm": 143.14759826660156,
      "learning_rate": 7.107084448203212e-05,
      "loss": 0.06815189123153687,
      "loss_step_1": 0.0719,
      "loss_step_2": 0.0689,
      "loss_step_3": 0.0677,
      "step": 13230
    },
    {
      "entropy_step_1": 0.2177,
      "entropy_step_2": 0.2234,
      "entropy_step_3": 0.2299,
      "epoch": 0.39380149311442253,
      "grad_norm": 45.676239013671875,
      "learning_rate": 7.102623354477424e-05,
      "loss": 0.06429983973503113,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.0677,
      "step": 13240
    },
    {
      "entropy_step_1": 0.2339,
      "entropy_step_2": 0.2377,
      "entropy_step_3": 0.22,
      "epoch": 0.39409892626632165,
      "grad_norm": 78.21421813964844,
      "learning_rate": 7.098160226434933e-05,
      "loss": 0.05982611775398254,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0562,
      "step": 13250
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1836,
      "entropy_step_3": 0.1839,
      "epoch": 0.39439635941822077,
      "grad_norm": 173.44166564941406,
      "learning_rate": 7.09369506839387e-05,
      "loss": 0.06966316103935241,
      "loss_step_1": 0.0734,
      "loss_step_2": 0.0696,
      "loss_step_3": 0.0765,
      "step": 13260
    },
    {
      "entropy_step_1": 0.1977,
      "entropy_step_2": 0.2096,
      "entropy_step_3": 0.2181,
      "epoch": 0.3946937925701199,
      "grad_norm": 235.30015563964844,
      "learning_rate": 7.089227884674344e-05,
      "loss": 0.06111922264099121,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.059,
      "step": 13270
    },
    {
      "entropy_step_1": 0.2287,
      "entropy_step_2": 0.2376,
      "entropy_step_3": 0.2495,
      "epoch": 0.394991225722019,
      "grad_norm": 189.4049072265625,
      "learning_rate": 7.084758679598409e-05,
      "loss": 0.06659809350967408,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0675,
      "step": 13280
    },
    {
      "entropy_step_1": 0.2027,
      "entropy_step_2": 0.2066,
      "entropy_step_3": 0.1977,
      "epoch": 0.39528865887391806,
      "grad_norm": 136.48431396484375,
      "learning_rate": 7.080287457490084e-05,
      "loss": 0.06140934228897095,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0593,
      "step": 13290
    },
    {
      "entropy_step_1": 0.2136,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.2238,
      "epoch": 0.3955860920258172,
      "grad_norm": 151.83152770996094,
      "learning_rate": 7.075814222675338e-05,
      "loss": 0.06526874899864196,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0687,
      "step": 13300
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.2016,
      "entropy_step_3": 0.1964,
      "epoch": 0.3958835251777163,
      "grad_norm": 147.7934112548828,
      "learning_rate": 7.071338979482086e-05,
      "loss": 0.06256843209266663,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0659,
      "loss_step_3": 0.0589,
      "step": 13310
    },
    {
      "entropy_step_1": 0.2206,
      "entropy_step_2": 0.2205,
      "entropy_step_3": 0.2234,
      "epoch": 0.3961809583296154,
      "grad_norm": 187.06747436523438,
      "learning_rate": 7.066861732240186e-05,
      "loss": 0.05940934419631958,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0597,
      "step": 13320
    },
    {
      "entropy_step_1": 0.2154,
      "entropy_step_2": 0.2113,
      "entropy_step_3": 0.2199,
      "epoch": 0.39647839148151454,
      "grad_norm": 75.31459045410156,
      "learning_rate": 7.062382485281437e-05,
      "loss": 0.06734951734542846,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0815,
      "step": 13330
    },
    {
      "entropy_step_1": 0.2198,
      "entropy_step_2": 0.2172,
      "entropy_step_3": 0.2121,
      "epoch": 0.39677582463341365,
      "grad_norm": 476.14056396484375,
      "learning_rate": 7.05790124293957e-05,
      "loss": 0.06878939270973206,
      "loss_step_1": 0.0764,
      "loss_step_2": 0.0735,
      "loss_step_3": 0.067,
      "step": 13340
    },
    {
      "entropy_step_1": 0.2148,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.2029,
      "epoch": 0.39707325778531277,
      "grad_norm": 76.62419128417969,
      "learning_rate": 7.05341800955025e-05,
      "loss": 0.06605101823806762,
      "loss_step_1": 0.074,
      "loss_step_2": 0.0734,
      "loss_step_3": 0.0668,
      "step": 13350
    },
    {
      "entropy_step_1": 0.2135,
      "entropy_step_2": 0.2141,
      "entropy_step_3": 0.2148,
      "epoch": 0.39737069093721183,
      "grad_norm": 61.788822174072266,
      "learning_rate": 7.048932789451065e-05,
      "loss": 0.061317676305770875,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0586,
      "step": 13360
    },
    {
      "entropy_step_1": 0.2034,
      "entropy_step_2": 0.2043,
      "entropy_step_3": 0.1947,
      "epoch": 0.39766812408911095,
      "grad_norm": 65.87030029296875,
      "learning_rate": 7.044445586981525e-05,
      "loss": 0.06208389401435852,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0571,
      "step": 13370
    },
    {
      "entropy_step_1": 0.2296,
      "entropy_step_2": 0.2274,
      "entropy_step_3": 0.2214,
      "epoch": 0.39796555724101007,
      "grad_norm": 106.31910705566406,
      "learning_rate": 7.039956406483061e-05,
      "loss": 0.06329457759857178,
      "loss_step_1": 0.061,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0589,
      "step": 13380
    },
    {
      "entropy_step_1": 0.2096,
      "entropy_step_2": 0.2048,
      "entropy_step_3": 0.2114,
      "epoch": 0.3982629903929092,
      "grad_norm": 92.46717834472656,
      "learning_rate": 7.035465252299017e-05,
      "loss": 0.06906673312187195,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0766,
      "step": 13390
    },
    {
      "entropy_step_1": 0.2099,
      "entropy_step_2": 0.2027,
      "entropy_step_3": 0.1814,
      "epoch": 0.3985604235448083,
      "grad_norm": 47.21907043457031,
      "learning_rate": 7.030972128774644e-05,
      "loss": 0.05981747508049011,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.052,
      "loss_step_3": 0.046,
      "step": 13400
    },
    {
      "entropy_step_1": 0.2108,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.1717,
      "epoch": 0.3988578566967074,
      "grad_norm": 209.5569610595703,
      "learning_rate": 7.026477040257102e-05,
      "loss": 0.06027495265007019,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0495,
      "step": 13410
    },
    {
      "entropy_step_1": 0.1771,
      "entropy_step_2": 0.1732,
      "entropy_step_3": 0.1552,
      "epoch": 0.39915528984860654,
      "grad_norm": 108.24126434326172,
      "learning_rate": 7.021979991095449e-05,
      "loss": 0.06078647375106812,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0443,
      "step": 13420
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.2011,
      "entropy_step_3": 0.1913,
      "epoch": 0.39945272300050566,
      "grad_norm": 495.3605041503906,
      "learning_rate": 7.017480985640642e-05,
      "loss": 0.061386597156524655,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0588,
      "step": 13430
    },
    {
      "entropy_step_1": 0.2181,
      "entropy_step_2": 0.2114,
      "entropy_step_3": 0.2214,
      "epoch": 0.3997501561524047,
      "grad_norm": 61.75539779663086,
      "learning_rate": 7.012980028245529e-05,
      "loss": 0.06217736005783081,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0552,
      "step": 13440
    },
    {
      "entropy_step_1": 0.2088,
      "entropy_step_2": 0.2011,
      "entropy_step_3": 0.1828,
      "epoch": 0.40004758930430384,
      "grad_norm": 59.300254821777344,
      "learning_rate": 7.008477123264848e-05,
      "loss": 0.068492192029953,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0561,
      "step": 13450
    },
    {
      "entropy_step_1": 0.2208,
      "entropy_step_2": 0.2109,
      "entropy_step_3": 0.2048,
      "epoch": 0.40034502245620296,
      "grad_norm": 564.8363037109375,
      "learning_rate": 7.003972275055221e-05,
      "loss": 0.05691617727279663,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0508,
      "step": 13460
    },
    {
      "entropy_step_1": 0.1872,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.1829,
      "epoch": 0.4006424556081021,
      "grad_norm": 62.216922760009766,
      "learning_rate": 6.99946548797515e-05,
      "loss": 0.060161709785461426,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0488,
      "step": 13470
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.1918,
      "entropy_step_3": 0.193,
      "epoch": 0.4009398887600012,
      "grad_norm": 81.8545150756836,
      "learning_rate": 6.994956766385012e-05,
      "loss": 0.06283673644065857,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0628,
      "step": 13480
    },
    {
      "entropy_step_1": 0.2242,
      "entropy_step_2": 0.216,
      "entropy_step_3": 0.2142,
      "epoch": 0.4012373219119003,
      "grad_norm": 37.30115509033203,
      "learning_rate": 6.990446114647058e-05,
      "loss": 0.06382733583450317,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0588,
      "step": 13490
    },
    {
      "entropy_step_1": 0.1713,
      "entropy_step_2": 0.1679,
      "entropy_step_3": 0.1774,
      "epoch": 0.40153475506379943,
      "grad_norm": 72.29856872558594,
      "learning_rate": 6.985933537125404e-05,
      "loss": 0.058999812602996825,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0556,
      "step": 13500
    },
    {
      "entropy_step_1": 0.2017,
      "entropy_step_2": 0.1977,
      "entropy_step_3": 0.1799,
      "epoch": 0.40183218821569855,
      "grad_norm": 194.7978515625,
      "learning_rate": 6.981419038186031e-05,
      "loss": 0.061129868030548096,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0629,
      "step": 13510
    },
    {
      "entropy_step_1": 0.1989,
      "entropy_step_2": 0.196,
      "entropy_step_3": 0.1922,
      "epoch": 0.4021296213675976,
      "grad_norm": 67.89088439941406,
      "learning_rate": 6.976902622196777e-05,
      "loss": 0.06050979495048523,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0509,
      "step": 13520
    },
    {
      "entropy_step_1": 0.2123,
      "entropy_step_2": 0.2105,
      "entropy_step_3": 0.2188,
      "epoch": 0.40242705451949673,
      "grad_norm": 103.91695404052734,
      "learning_rate": 6.972384293527336e-05,
      "loss": 0.06994785070419311,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0694,
      "step": 13530
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.2034,
      "entropy_step_3": 0.2006,
      "epoch": 0.40272448767139585,
      "grad_norm": 341.35272216796875,
      "learning_rate": 6.967864056549254e-05,
      "loss": 0.05535833835601807,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0486,
      "step": 13540
    },
    {
      "entropy_step_1": 0.1813,
      "entropy_step_2": 0.184,
      "entropy_step_3": 0.1865,
      "epoch": 0.40302192082329497,
      "grad_norm": 170.4493408203125,
      "learning_rate": 6.963341915635919e-05,
      "loss": 0.05934275388717651,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0625,
      "step": 13550
    },
    {
      "entropy_step_1": 0.2214,
      "entropy_step_2": 0.2298,
      "entropy_step_3": 0.2272,
      "epoch": 0.4033193539751941,
      "grad_norm": 234.85728454589844,
      "learning_rate": 6.958817875162567e-05,
      "loss": 0.06601885557174683,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0634,
      "step": 13560
    },
    {
      "entropy_step_1": 0.2358,
      "entropy_step_2": 0.2436,
      "entropy_step_3": 0.254,
      "epoch": 0.4036167871270932,
      "grad_norm": 138.14076232910156,
      "learning_rate": 6.954291939506265e-05,
      "loss": 0.06842800378799438,
      "loss_step_1": 0.0733,
      "loss_step_2": 0.0727,
      "loss_step_3": 0.0801,
      "step": 13570
    },
    {
      "entropy_step_1": 0.2166,
      "entropy_step_2": 0.2247,
      "entropy_step_3": 0.2067,
      "epoch": 0.4039142202789923,
      "grad_norm": 167.58425903320312,
      "learning_rate": 6.949764113045922e-05,
      "loss": 0.05906093120574951,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0594,
      "step": 13580
    },
    {
      "entropy_step_1": 0.203,
      "entropy_step_2": 0.209,
      "entropy_step_3": 0.2069,
      "epoch": 0.4042116534308914,
      "grad_norm": 159.55133056640625,
      "learning_rate": 6.945234400162268e-05,
      "loss": 0.06714918613433837,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0666,
      "step": 13590
    },
    {
      "entropy_step_1": 0.2129,
      "entropy_step_2": 0.2194,
      "entropy_step_3": 0.1973,
      "epoch": 0.4045090865827905,
      "grad_norm": 104.52310180664062,
      "learning_rate": 6.940702805237862e-05,
      "loss": 0.06843621730804443,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0551,
      "step": 13600
    },
    {
      "entropy_step_1": 0.2314,
      "entropy_step_2": 0.2383,
      "entropy_step_3": 0.2192,
      "epoch": 0.4048065197346896,
      "grad_norm": 149.6254425048828,
      "learning_rate": 6.936169332657088e-05,
      "loss": 0.06609995365142822,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.052,
      "step": 13610
    },
    {
      "entropy_step_1": 0.1769,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.1656,
      "epoch": 0.40510395288658874,
      "grad_norm": 124.77953338623047,
      "learning_rate": 6.931633986806136e-05,
      "loss": 0.05296396017074585,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0495,
      "loss_step_3": 0.0411,
      "step": 13620
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.2005,
      "entropy_step_3": 0.2098,
      "epoch": 0.40540138603848785,
      "grad_norm": 213.5821075439453,
      "learning_rate": 6.927096772073019e-05,
      "loss": 0.06320421695709229,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0611,
      "step": 13630
    },
    {
      "entropy_step_1": 0.2065,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.2217,
      "epoch": 0.405698819190387,
      "grad_norm": 168.63613891601562,
      "learning_rate": 6.922557692847555e-05,
      "loss": 0.060931462049484256,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.058,
      "step": 13640
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.2036,
      "entropy_step_3": 0.2118,
      "epoch": 0.4059962523422861,
      "grad_norm": 44.67960739135742,
      "learning_rate": 6.91801675352136e-05,
      "loss": 0.05933077931404114,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0606,
      "step": 13650
    },
    {
      "entropy_step_1": 0.2236,
      "entropy_step_2": 0.223,
      "entropy_step_3": 0.2389,
      "epoch": 0.4062936854941852,
      "grad_norm": 131.80589294433594,
      "learning_rate": 6.913473958487859e-05,
      "loss": 0.06278392672538757,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0739,
      "step": 13660
    },
    {
      "entropy_step_1": 0.23,
      "entropy_step_2": 0.2295,
      "entropy_step_3": 0.2262,
      "epoch": 0.40659111864608427,
      "grad_norm": 136.1300048828125,
      "learning_rate": 6.908929312142267e-05,
      "loss": 0.07773579359054565,
      "loss_step_1": 0.0856,
      "loss_step_2": 0.0841,
      "loss_step_3": 0.0646,
      "step": 13670
    },
    {
      "entropy_step_1": 0.2197,
      "entropy_step_2": 0.2195,
      "entropy_step_3": 0.2332,
      "epoch": 0.4068885517979834,
      "grad_norm": 87.03961944580078,
      "learning_rate": 6.90438281888159e-05,
      "loss": 0.05791701674461365,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0663,
      "step": 13680
    },
    {
      "entropy_step_1": 0.2113,
      "entropy_step_2": 0.2134,
      "entropy_step_3": 0.2038,
      "epoch": 0.4071859849498825,
      "grad_norm": 68.48536682128906,
      "learning_rate": 6.899834483104624e-05,
      "loss": 0.06493307948112488,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0652,
      "loss_step_3": 0.0637,
      "step": 13690
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2097,
      "entropy_step_3": 0.2146,
      "epoch": 0.4074834181017816,
      "grad_norm": 102.70381927490234,
      "learning_rate": 6.895284309211943e-05,
      "loss": 0.06534689664840698,
      "loss_step_1": 0.0679,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.0776,
      "step": 13700
    },
    {
      "entropy_step_1": 0.2123,
      "entropy_step_2": 0.2141,
      "entropy_step_3": 0.226,
      "epoch": 0.40778085125368074,
      "grad_norm": 86.76974487304688,
      "learning_rate": 6.890732301605906e-05,
      "loss": 0.07374396324157714,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0704,
      "step": 13710
    },
    {
      "entropy_step_1": 0.2268,
      "entropy_step_2": 0.2304,
      "entropy_step_3": 0.2373,
      "epoch": 0.40807828440557986,
      "grad_norm": 50.808021545410156,
      "learning_rate": 6.886178464690639e-05,
      "loss": 0.07382552027702331,
      "loss_step_1": 0.0752,
      "loss_step_2": 0.0744,
      "loss_step_3": 0.0718,
      "step": 13720
    },
    {
      "entropy_step_1": 0.2249,
      "entropy_step_2": 0.2288,
      "entropy_step_3": 0.2333,
      "epoch": 0.408375717557479,
      "grad_norm": 43.860591888427734,
      "learning_rate": 6.88162280287204e-05,
      "loss": 0.062037086486816405,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0468,
      "step": 13730
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.2113,
      "epoch": 0.40867315070937804,
      "grad_norm": 542.2110595703125,
      "learning_rate": 6.877065320557778e-05,
      "loss": 0.0620654821395874,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0612,
      "step": 13740
    },
    {
      "entropy_step_1": 0.2179,
      "entropy_step_2": 0.2162,
      "entropy_step_3": 0.22,
      "epoch": 0.40897058386127716,
      "grad_norm": 71.64596557617188,
      "learning_rate": 6.87250602215728e-05,
      "loss": 0.05754964351654053,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0502,
      "step": 13750
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.1959,
      "entropy_step_3": 0.1945,
      "epoch": 0.4092680170131763,
      "grad_norm": 558.8438110351562,
      "learning_rate": 6.867944912081722e-05,
      "loss": 0.06185073852539062,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0606,
      "step": 13760
    },
    {
      "entropy_step_1": 0.1895,
      "entropy_step_2": 0.1846,
      "entropy_step_3": 0.1855,
      "epoch": 0.4095654501650754,
      "grad_norm": 187.91061401367188,
      "learning_rate": 6.863381994744046e-05,
      "loss": 0.056387138366699216,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0584,
      "step": 13770
    },
    {
      "entropy_step_1": 0.2057,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.2123,
      "epoch": 0.4098628833169745,
      "grad_norm": 58.833065032958984,
      "learning_rate": 6.858817274558935e-05,
      "loss": 0.05805739760398865,
      "loss_step_1": 0.057,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0585,
      "step": 13780
    },
    {
      "entropy_step_1": 0.2081,
      "entropy_step_2": 0.2034,
      "entropy_step_3": 0.2071,
      "epoch": 0.41016031646887363,
      "grad_norm": 106.3961410522461,
      "learning_rate": 6.854250755942817e-05,
      "loss": 0.0644540548324585,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0623,
      "step": 13790
    },
    {
      "entropy_step_1": 0.2031,
      "entropy_step_2": 0.1979,
      "entropy_step_3": 0.2079,
      "epoch": 0.41045774962077275,
      "grad_norm": 98.36650848388672,
      "learning_rate": 6.849682443313864e-05,
      "loss": 0.05667756199836731,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0564,
      "step": 13800
    },
    {
      "entropy_step_1": 0.2009,
      "entropy_step_2": 0.196,
      "entropy_step_3": 0.1876,
      "epoch": 0.41075518277267187,
      "grad_norm": 124.55608367919922,
      "learning_rate": 6.845112341091977e-05,
      "loss": 0.06299437284469604,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0528,
      "step": 13810
    },
    {
      "entropy_step_1": 0.2111,
      "entropy_step_2": 0.2071,
      "entropy_step_3": 0.2069,
      "epoch": 0.41105261592457093,
      "grad_norm": 113.73513793945312,
      "learning_rate": 6.840540453698795e-05,
      "loss": 0.06545457243919373,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0554,
      "step": 13820
    },
    {
      "entropy_step_1": 0.2212,
      "entropy_step_2": 0.2143,
      "entropy_step_3": 0.2123,
      "epoch": 0.41135004907647005,
      "grad_norm": 63.887725830078125,
      "learning_rate": 6.835966785557678e-05,
      "loss": 0.059613800048828124,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0708,
      "step": 13830
    },
    {
      "entropy_step_1": 0.2302,
      "entropy_step_2": 0.2247,
      "entropy_step_3": 0.2296,
      "epoch": 0.41164748222836917,
      "grad_norm": 100.81594848632812,
      "learning_rate": 6.831391341093716e-05,
      "loss": 0.06928393840789795,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0766,
      "step": 13840
    },
    {
      "entropy_step_1": 0.219,
      "entropy_step_2": 0.2119,
      "entropy_step_3": 0.2256,
      "epoch": 0.4119449153802683,
      "grad_norm": 69.2034912109375,
      "learning_rate": 6.82681412473371e-05,
      "loss": 0.05796271562576294,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0643,
      "step": 13850
    },
    {
      "entropy_step_1": 0.2015,
      "entropy_step_2": 0.1944,
      "entropy_step_3": 0.1898,
      "epoch": 0.4122423485321674,
      "grad_norm": 86.31202697753906,
      "learning_rate": 6.822235140906183e-05,
      "loss": 0.057194101810455325,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0556,
      "step": 13860
    },
    {
      "entropy_step_1": 0.2035,
      "entropy_step_2": 0.1941,
      "entropy_step_3": 0.1899,
      "epoch": 0.4125397816840665,
      "grad_norm": 296.0848388671875,
      "learning_rate": 6.817654394041359e-05,
      "loss": 0.053837788105010984,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0508,
      "step": 13870
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1846,
      "entropy_step_3": 0.1769,
      "epoch": 0.41283721483596564,
      "grad_norm": 84.71009826660156,
      "learning_rate": 6.813071888571178e-05,
      "loss": 0.0613441526889801,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0595,
      "step": 13880
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.1998,
      "entropy_step_3": 0.2211,
      "epoch": 0.4131346479878647,
      "grad_norm": 329.0058288574219,
      "learning_rate": 6.808487628929274e-05,
      "loss": 0.057714450359344485,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0603,
      "step": 13890
    },
    {
      "entropy_step_1": 0.1866,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.1859,
      "epoch": 0.4134320811397638,
      "grad_norm": 480.9157409667969,
      "learning_rate": 6.803901619550982e-05,
      "loss": 0.06261398792266845,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0596,
      "step": 13900
    },
    {
      "entropy_step_1": 0.2186,
      "entropy_step_2": 0.2113,
      "entropy_step_3": 0.2093,
      "epoch": 0.41372951429166294,
      "grad_norm": 72.52754974365234,
      "learning_rate": 6.799313864873328e-05,
      "loss": 0.06496147513389587,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0583,
      "step": 13910
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.1903,
      "entropy_step_3": 0.1817,
      "epoch": 0.41402694744356205,
      "grad_norm": 428.93743896484375,
      "learning_rate": 6.794724369335024e-05,
      "loss": 0.06321152448654174,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0578,
      "step": 13920
    },
    {
      "entropy_step_1": 0.233,
      "entropy_step_2": 0.2227,
      "entropy_step_3": 0.2081,
      "epoch": 0.4143243805954612,
      "grad_norm": 70.6868667602539,
      "learning_rate": 6.790133137376473e-05,
      "loss": 0.0669189989566803,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.067,
      "loss_step_3": 0.065,
      "step": 13930
    },
    {
      "entropy_step_1": 0.2294,
      "entropy_step_2": 0.2224,
      "entropy_step_3": 0.2443,
      "epoch": 0.4146218137473603,
      "grad_norm": 104.84321594238281,
      "learning_rate": 6.785540173439755e-05,
      "loss": 0.06973609328269958,
      "loss_step_1": 0.0732,
      "loss_step_2": 0.073,
      "loss_step_3": 0.0871,
      "step": 13940
    },
    {
      "entropy_step_1": 0.2353,
      "entropy_step_2": 0.2278,
      "entropy_step_3": 0.2131,
      "epoch": 0.4149192468992594,
      "grad_norm": 119.68803405761719,
      "learning_rate": 6.78094548196862e-05,
      "loss": 0.0635843813419342,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0641,
      "step": 13950
    },
    {
      "entropy_step_1": 0.2169,
      "entropy_step_2": 0.2139,
      "entropy_step_3": 0.2137,
      "epoch": 0.4152166800511585,
      "grad_norm": 75.29126739501953,
      "learning_rate": 6.7763490674085e-05,
      "loss": 0.0727946937084198,
      "loss_step_1": 0.0744,
      "loss_step_2": 0.0738,
      "loss_step_3": 0.0749,
      "step": 13960
    },
    {
      "entropy_step_1": 0.253,
      "entropy_step_2": 0.2506,
      "entropy_step_3": 0.2455,
      "epoch": 0.4155141132030576,
      "grad_norm": 34.935237884521484,
      "learning_rate": 6.771750934206485e-05,
      "loss": 0.0666684627532959,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.0739,
      "loss_step_3": 0.0768,
      "step": 13970
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.2112,
      "entropy_step_3": 0.2274,
      "epoch": 0.4158115463549567,
      "grad_norm": 133.88917541503906,
      "learning_rate": 6.767151086811335e-05,
      "loss": 0.06412237882614136,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0781,
      "step": 13980
    },
    {
      "entropy_step_1": 0.1941,
      "entropy_step_2": 0.1924,
      "entropy_step_3": 0.1928,
      "epoch": 0.4161089795068558,
      "grad_norm": 278.6361999511719,
      "learning_rate": 6.762549529673462e-05,
      "loss": 0.06359769105911255,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0586,
      "step": 13990
    },
    {
      "entropy_step_1": 0.218,
      "entropy_step_2": 0.2157,
      "entropy_step_3": 0.2193,
      "epoch": 0.41640641265875494,
      "grad_norm": 109.06614685058594,
      "learning_rate": 6.757946267244937e-05,
      "loss": 0.06789698004722595,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0651,
      "loss_step_3": 0.0691,
      "step": 14000
    },
    {
      "epoch": 0.41640641265875494,
      "eval_accuracy": 0.853,
      "eval_f1": 0.852212440093257,
      "eval_loss": 0.1587878167629242,
      "eval_precision": 0.8606884346735003,
      "eval_recall": 0.853,
      "eval_runtime": 6.9108,
      "eval_samples_per_second": 72.35,
      "eval_steps_per_second": 9.116,
      "step": 14000
    },
    {
      "entropy_step_1": 0.2086,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.2089,
      "epoch": 0.41670384581065406,
      "grad_norm": 102.80815124511719,
      "learning_rate": 6.753341303979477e-05,
      "loss": 0.06142747402191162,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0597,
      "step": 14010
    },
    {
      "entropy_step_1": 0.181,
      "entropy_step_2": 0.1817,
      "entropy_step_3": 0.1911,
      "epoch": 0.4170012789625532,
      "grad_norm": 99.97615814208984,
      "learning_rate": 6.748734644332448e-05,
      "loss": 0.062079203128814694,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0741,
      "step": 14020
    },
    {
      "entropy_step_1": 0.2026,
      "entropy_step_2": 0.2028,
      "entropy_step_3": 0.2031,
      "epoch": 0.4172987121144523,
      "grad_norm": 54.62283706665039,
      "learning_rate": 6.744126292760857e-05,
      "loss": 0.05464267134666443,
      "loss_step_1": 0.0485,
      "loss_step_2": 0.0482,
      "loss_step_3": 0.0464,
      "step": 14030
    },
    {
      "entropy_step_1": 0.2083,
      "entropy_step_2": 0.2082,
      "entropy_step_3": 0.2048,
      "epoch": 0.4175961452663514,
      "grad_norm": 148.655517578125,
      "learning_rate": 6.739516253723343e-05,
      "loss": 0.06908016204833985,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0595,
      "step": 14040
    },
    {
      "entropy_step_1": 0.2255,
      "entropy_step_2": 0.224,
      "entropy_step_3": 0.2299,
      "epoch": 0.4178935784182505,
      "grad_norm": 91.3832015991211,
      "learning_rate": 6.734904531680185e-05,
      "loss": 0.06845394968986511,
      "loss_step_1": 0.0713,
      "loss_step_2": 0.0711,
      "loss_step_3": 0.0736,
      "step": 14050
    },
    {
      "entropy_step_1": 0.2321,
      "entropy_step_2": 0.2289,
      "entropy_step_3": 0.2141,
      "epoch": 0.4181910115701496,
      "grad_norm": 424.1656494140625,
      "learning_rate": 6.730291131093287e-05,
      "loss": 0.07530132532119752,
      "loss_step_1": 0.0774,
      "loss_step_2": 0.078,
      "loss_step_3": 0.0851,
      "step": 14060
    },
    {
      "entropy_step_1": 0.2323,
      "entropy_step_2": 0.2222,
      "entropy_step_3": 0.2307,
      "epoch": 0.4184884447220487,
      "grad_norm": 303.9932861328125,
      "learning_rate": 6.725676056426174e-05,
      "loss": 0.06674252152442932,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.054,
      "loss_step_3": 0.061,
      "step": 14070
    },
    {
      "entropy_step_1": 0.2218,
      "entropy_step_2": 0.2154,
      "entropy_step_3": 0.2208,
      "epoch": 0.41878587787394783,
      "grad_norm": 144.67982482910156,
      "learning_rate": 6.721059312143996e-05,
      "loss": 0.0633568823337555,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.063,
      "step": 14080
    },
    {
      "entropy_step_1": 0.2159,
      "entropy_step_2": 0.2088,
      "entropy_step_3": 0.1989,
      "epoch": 0.41908331102584695,
      "grad_norm": 114.9125747680664,
      "learning_rate": 6.716440902713516e-05,
      "loss": 0.05821634531021118,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0582,
      "step": 14090
    },
    {
      "entropy_step_1": 0.2258,
      "entropy_step_2": 0.2223,
      "entropy_step_3": 0.217,
      "epoch": 0.41938074417774607,
      "grad_norm": 42.951725006103516,
      "learning_rate": 6.711820832603106e-05,
      "loss": 0.06258311867713928,
      "loss_step_1": 0.0688,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.066,
      "step": 14100
    },
    {
      "entropy_step_1": 0.2195,
      "entropy_step_2": 0.22,
      "entropy_step_3": 0.2271,
      "epoch": 0.4196781773296452,
      "grad_norm": 206.63711547851562,
      "learning_rate": 6.707199106282746e-05,
      "loss": 0.06443716883659363,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0728,
      "step": 14110
    },
    {
      "entropy_step_1": 0.2289,
      "entropy_step_2": 0.2302,
      "entropy_step_3": 0.2435,
      "epoch": 0.41997561048154425,
      "grad_norm": 126.21395111083984,
      "learning_rate": 6.702575728224023e-05,
      "loss": 0.06861551403999329,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0699,
      "step": 14120
    },
    {
      "entropy_step_1": 0.2135,
      "entropy_step_2": 0.2175,
      "entropy_step_3": 0.2287,
      "epoch": 0.42027304363344337,
      "grad_norm": 261.281982421875,
      "learning_rate": 6.697950702900115e-05,
      "loss": 0.06741804480552674,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0698,
      "step": 14130
    },
    {
      "entropy_step_1": 0.2161,
      "entropy_step_2": 0.2196,
      "entropy_step_3": 0.2451,
      "epoch": 0.4205704767853425,
      "grad_norm": 187.21888732910156,
      "learning_rate": 6.6933240347858e-05,
      "loss": 0.06411729454994201,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0691,
      "step": 14140
    },
    {
      "entropy_step_1": 0.1934,
      "entropy_step_2": 0.1963,
      "entropy_step_3": 0.2111,
      "epoch": 0.4208679099372416,
      "grad_norm": 319.3817138671875,
      "learning_rate": 6.688695728357439e-05,
      "loss": 0.05248534083366394,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.0581,
      "step": 14150
    },
    {
      "entropy_step_1": 0.2012,
      "entropy_step_2": 0.2028,
      "entropy_step_3": 0.2048,
      "epoch": 0.4211653430891407,
      "grad_norm": 273.8393249511719,
      "learning_rate": 6.684065788092983e-05,
      "loss": 0.06265203356742859,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0521,
      "step": 14160
    },
    {
      "entropy_step_1": 0.1977,
      "entropy_step_2": 0.1952,
      "entropy_step_3": 0.1872,
      "epoch": 0.42146277624103984,
      "grad_norm": 406.0248107910156,
      "learning_rate": 6.67943421847196e-05,
      "loss": 0.06411808729171753,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0687,
      "step": 14170
    },
    {
      "entropy_step_1": 0.189,
      "entropy_step_2": 0.1877,
      "entropy_step_3": 0.1955,
      "epoch": 0.42176020939293896,
      "grad_norm": 88.05773162841797,
      "learning_rate": 6.674801023975481e-05,
      "loss": 0.062335819005966187,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0603,
      "step": 14180
    },
    {
      "entropy_step_1": 0.1733,
      "entropy_step_2": 0.1698,
      "entropy_step_3": 0.1677,
      "epoch": 0.4220576425448381,
      "grad_norm": 106.60838317871094,
      "learning_rate": 6.670166209086222e-05,
      "loss": 0.057234418392181394,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0618,
      "step": 14190
    },
    {
      "entropy_step_1": 0.2338,
      "entropy_step_2": 0.2305,
      "entropy_step_3": 0.2191,
      "epoch": 0.42235507569673714,
      "grad_norm": 264.998779296875,
      "learning_rate": 6.665529778288429e-05,
      "loss": 0.07496945261955261,
      "loss_step_1": 0.0774,
      "loss_step_2": 0.0771,
      "loss_step_3": 0.0829,
      "step": 14200
    },
    {
      "entropy_step_1": 0.2021,
      "entropy_step_2": 0.203,
      "entropy_step_3": 0.2061,
      "epoch": 0.42265250884863625,
      "grad_norm": 153.83544921875,
      "learning_rate": 6.660891736067913e-05,
      "loss": 0.06324229836463928,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0562,
      "step": 14210
    },
    {
      "entropy_step_1": 0.1993,
      "entropy_step_2": 0.2017,
      "entropy_step_3": 0.2099,
      "epoch": 0.4229499420005354,
      "grad_norm": 2626.5576171875,
      "learning_rate": 6.656252086912044e-05,
      "loss": 0.06450698375701905,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0613,
      "step": 14220
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1974,
      "entropy_step_3": 0.1899,
      "epoch": 0.4232473751524345,
      "grad_norm": 91.29451751708984,
      "learning_rate": 6.651610835309743e-05,
      "loss": 0.06231285929679871,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.0513,
      "step": 14230
    },
    {
      "entropy_step_1": 0.2098,
      "entropy_step_2": 0.2106,
      "entropy_step_3": 0.2249,
      "epoch": 0.4235448083043336,
      "grad_norm": 59.28451919555664,
      "learning_rate": 6.646967985751485e-05,
      "loss": 0.057741963863372804,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0662,
      "step": 14240
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2054,
      "entropy_step_3": 0.1999,
      "epoch": 0.4238422414562327,
      "grad_norm": 60.95350646972656,
      "learning_rate": 6.642323542729292e-05,
      "loss": 0.05975069403648377,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0552,
      "step": 14250
    },
    {
      "entropy_step_1": 0.233,
      "entropy_step_2": 0.2289,
      "entropy_step_3": 0.2214,
      "epoch": 0.42413967460813184,
      "grad_norm": 166.70779418945312,
      "learning_rate": 6.637677510736724e-05,
      "loss": 0.06215118169784546,
      "loss_step_1": 0.0666,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0608,
      "step": 14260
    },
    {
      "entropy_step_1": 0.2019,
      "entropy_step_2": 0.2067,
      "entropy_step_3": 0.2076,
      "epoch": 0.4244371077600309,
      "grad_norm": 107.69351959228516,
      "learning_rate": 6.63302989426888e-05,
      "loss": 0.06754953265190125,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0708,
      "loss_step_3": 0.067,
      "step": 14270
    },
    {
      "entropy_step_1": 0.1899,
      "entropy_step_2": 0.1946,
      "entropy_step_3": 0.1893,
      "epoch": 0.42473454091193,
      "grad_norm": 78.45877075195312,
      "learning_rate": 6.628380697822392e-05,
      "loss": 0.06558354496955872,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0595,
      "step": 14280
    },
    {
      "entropy_step_1": 0.2031,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.213,
      "epoch": 0.42503197406382914,
      "grad_norm": 57.473297119140625,
      "learning_rate": 6.623729925895422e-05,
      "loss": 0.05540151596069336,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0589,
      "step": 14290
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1818,
      "epoch": 0.42532940721572826,
      "grad_norm": 277.17498779296875,
      "learning_rate": 6.619077582987657e-05,
      "loss": 0.0570567786693573,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0558,
      "step": 14300
    },
    {
      "entropy_step_1": 0.1886,
      "entropy_step_2": 0.1922,
      "entropy_step_3": 0.2018,
      "epoch": 0.4256268403676274,
      "grad_norm": 340.4908752441406,
      "learning_rate": 6.614423673600296e-05,
      "loss": 0.06529116630554199,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0697,
      "step": 14310
    },
    {
      "entropy_step_1": 0.2317,
      "entropy_step_2": 0.2344,
      "entropy_step_3": 0.2364,
      "epoch": 0.4259242735195265,
      "grad_norm": 122.33257293701172,
      "learning_rate": 6.609768202236067e-05,
      "loss": 0.074676775932312,
      "loss_step_1": 0.0725,
      "loss_step_2": 0.0726,
      "loss_step_3": 0.0677,
      "step": 14320
    },
    {
      "entropy_step_1": 0.1894,
      "entropy_step_2": 0.1906,
      "entropy_step_3": 0.1943,
      "epoch": 0.4262217066714256,
      "grad_norm": 144.03128051757812,
      "learning_rate": 6.605111173399194e-05,
      "loss": 0.05604594349861145,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0509,
      "step": 14330
    },
    {
      "entropy_step_1": 0.2036,
      "entropy_step_2": 0.2028,
      "entropy_step_3": 0.1758,
      "epoch": 0.42651913982332473,
      "grad_norm": 229.76739501953125,
      "learning_rate": 6.600452591595423e-05,
      "loss": 0.060060149431228636,
      "loss_step_1": 0.0679,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0637,
      "step": 14340
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2064,
      "entropy_step_3": 0.2137,
      "epoch": 0.4268165729752238,
      "grad_norm": 157.06439208984375,
      "learning_rate": 6.595792461331991e-05,
      "loss": 0.06216703057289123,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0652,
      "step": 14350
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.2057,
      "entropy_step_3": 0.202,
      "epoch": 0.4271140061271229,
      "grad_norm": 97.271484375,
      "learning_rate": 6.59113078711764e-05,
      "loss": 0.06084859371185303,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0529,
      "step": 14360
    },
    {
      "entropy_step_1": 0.1783,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1663,
      "epoch": 0.42741143927902203,
      "grad_norm": 272.292236328125,
      "learning_rate": 6.586467573462601e-05,
      "loss": 0.059296411275863645,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0569,
      "step": 14370
    },
    {
      "entropy_step_1": 0.2114,
      "entropy_step_2": 0.212,
      "entropy_step_3": 0.2136,
      "epoch": 0.42770887243092115,
      "grad_norm": 42.62760925292969,
      "learning_rate": 6.581802824878597e-05,
      "loss": 0.06095443964004517,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.0652,
      "step": 14380
    },
    {
      "entropy_step_1": 0.1915,
      "entropy_step_2": 0.1947,
      "entropy_step_3": 0.1875,
      "epoch": 0.42800630558282027,
      "grad_norm": 142.2401885986328,
      "learning_rate": 6.57713654587884e-05,
      "loss": 0.061324840784072875,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.055,
      "step": 14390
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1904,
      "entropy_step_3": 0.1986,
      "epoch": 0.4283037387347194,
      "grad_norm": 94.83280181884766,
      "learning_rate": 6.572468740978011e-05,
      "loss": 0.05356038212776184,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.046,
      "step": 14400
    },
    {
      "entropy_step_1": 0.2107,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.2153,
      "epoch": 0.4286011718866185,
      "grad_norm": 87.605224609375,
      "learning_rate": 6.56779941469228e-05,
      "loss": 0.06128644347190857,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0713,
      "step": 14410
    },
    {
      "entropy_step_1": 0.2168,
      "entropy_step_2": 0.2187,
      "entropy_step_3": 0.2168,
      "epoch": 0.42889860503851757,
      "grad_norm": 233.27687072753906,
      "learning_rate": 6.563128571539281e-05,
      "loss": 0.06629726886749268,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0604,
      "step": 14420
    },
    {
      "entropy_step_1": 0.222,
      "entropy_step_2": 0.2221,
      "entropy_step_3": 0.2096,
      "epoch": 0.4291960381904167,
      "grad_norm": 933.4888305664062,
      "learning_rate": 6.558456216038122e-05,
      "loss": 0.06381189227104186,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0611,
      "step": 14430
    },
    {
      "entropy_step_1": 0.2131,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.1981,
      "epoch": 0.4294934713423158,
      "grad_norm": 77.81133270263672,
      "learning_rate": 6.553782352709368e-05,
      "loss": 0.06323696970939637,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0693,
      "loss_step_3": 0.0633,
      "step": 14440
    },
    {
      "entropy_step_1": 0.2116,
      "entropy_step_2": 0.2119,
      "entropy_step_3": 0.2178,
      "epoch": 0.4297909044942149,
      "grad_norm": 64.0295181274414,
      "learning_rate": 6.549106986075043e-05,
      "loss": 0.06913046836853028,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0678,
      "step": 14450
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.216,
      "entropy_step_3": 0.2439,
      "epoch": 0.43008833764611404,
      "grad_norm": 148.4069061279297,
      "learning_rate": 6.544430120658632e-05,
      "loss": 0.060201328992843625,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0796,
      "step": 14460
    },
    {
      "entropy_step_1": 0.1976,
      "entropy_step_2": 0.2006,
      "entropy_step_3": 0.2213,
      "epoch": 0.43038577079801316,
      "grad_norm": 119.14771270751953,
      "learning_rate": 6.539751760985064e-05,
      "loss": 0.06099168062210083,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0645,
      "step": 14470
    },
    {
      "entropy_step_1": 0.2183,
      "entropy_step_2": 0.2188,
      "entropy_step_3": 0.2123,
      "epoch": 0.4306832039499123,
      "grad_norm": 70.07284545898438,
      "learning_rate": 6.535071911580716e-05,
      "loss": 0.06476163864135742,
      "loss_step_1": 0.0701,
      "loss_step_2": 0.0699,
      "loss_step_3": 0.069,
      "step": 14480
    },
    {
      "entropy_step_1": 0.2232,
      "entropy_step_2": 0.2251,
      "entropy_step_3": 0.2384,
      "epoch": 0.4309806371018114,
      "grad_norm": 159.3399658203125,
      "learning_rate": 6.530390576973406e-05,
      "loss": 0.0681956648826599,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0573,
      "step": 14490
    },
    {
      "entropy_step_1": 0.2189,
      "entropy_step_2": 0.2204,
      "entropy_step_3": 0.2065,
      "epoch": 0.43127807025371045,
      "grad_norm": 59.25735092163086,
      "learning_rate": 6.525707761692386e-05,
      "loss": 0.06422998905181884,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.065,
      "loss_step_3": 0.0605,
      "step": 14500
    },
    {
      "entropy_step_1": 0.2167,
      "entropy_step_2": 0.2154,
      "entropy_step_3": 0.2247,
      "epoch": 0.4315755034056096,
      "grad_norm": 67.13794708251953,
      "learning_rate": 6.52102347026835e-05,
      "loss": 0.06677732467651368,
      "loss_step_1": 0.0714,
      "loss_step_2": 0.0716,
      "loss_step_3": 0.0672,
      "step": 14510
    },
    {
      "entropy_step_1": 0.2087,
      "entropy_step_2": 0.2118,
      "entropy_step_3": 0.2124,
      "epoch": 0.4318729365575087,
      "grad_norm": 106.92160034179688,
      "learning_rate": 6.516337707233409e-05,
      "loss": 0.058934581279754636,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0668,
      "step": 14520
    },
    {
      "entropy_step_1": 0.2381,
      "entropy_step_2": 0.2464,
      "entropy_step_3": 0.2531,
      "epoch": 0.4321703697094078,
      "grad_norm": 115.24420928955078,
      "learning_rate": 6.511650477121103e-05,
      "loss": 0.06763035655021668,
      "loss_step_1": 0.0772,
      "loss_step_2": 0.0763,
      "loss_step_3": 0.0733,
      "step": 14530
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.2014,
      "epoch": 0.4324678028613069,
      "grad_norm": 116.28227233886719,
      "learning_rate": 6.506961784466391e-05,
      "loss": 0.05803382396697998,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0723,
      "step": 14540
    },
    {
      "entropy_step_1": 0.1944,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.2044,
      "epoch": 0.43276523601320604,
      "grad_norm": 79.23333740234375,
      "learning_rate": 6.502271633805649e-05,
      "loss": 0.059811604022979734,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0581,
      "step": 14550
    },
    {
      "entropy_step_1": 0.2227,
      "entropy_step_2": 0.2289,
      "entropy_step_3": 0.242,
      "epoch": 0.43306266916510516,
      "grad_norm": 109.35433197021484,
      "learning_rate": 6.497580029676658e-05,
      "loss": 0.06160650253295898,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0646,
      "step": 14560
    },
    {
      "entropy_step_1": 0.1821,
      "entropy_step_2": 0.1817,
      "entropy_step_3": 0.1865,
      "epoch": 0.4333601023170043,
      "grad_norm": 152.8376922607422,
      "learning_rate": 6.492886976618613e-05,
      "loss": 0.07059626579284668,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0668,
      "step": 14570
    },
    {
      "entropy_step_1": 0.2341,
      "entropy_step_2": 0.2262,
      "entropy_step_3": 0.2256,
      "epoch": 0.43365753546890334,
      "grad_norm": 76.92559051513672,
      "learning_rate": 6.488192479172102e-05,
      "loss": 0.05490562915802002,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.052,
      "step": 14580
    },
    {
      "entropy_step_1": 0.1937,
      "entropy_step_2": 0.1865,
      "entropy_step_3": 0.1934,
      "epoch": 0.43395496862080246,
      "grad_norm": 196.48117065429688,
      "learning_rate": 6.483496541879118e-05,
      "loss": 0.062566477060318,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0658,
      "step": 14590
    },
    {
      "entropy_step_1": 0.2154,
      "entropy_step_2": 0.2066,
      "entropy_step_3": 0.2068,
      "epoch": 0.4342524017727016,
      "grad_norm": 54.14310073852539,
      "learning_rate": 6.478799169283045e-05,
      "loss": 0.06357493996620178,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.063,
      "loss_step_3": 0.066,
      "step": 14600
    },
    {
      "entropy_step_1": 0.2296,
      "entropy_step_2": 0.2239,
      "entropy_step_3": 0.2024,
      "epoch": 0.4345498349246007,
      "grad_norm": 75.56774139404297,
      "learning_rate": 6.474100365928655e-05,
      "loss": 0.06152291893959046,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.053,
      "step": 14610
    },
    {
      "entropy_step_1": 0.1813,
      "entropy_step_2": 0.1827,
      "entropy_step_3": 0.1799,
      "epoch": 0.4348472680764998,
      "grad_norm": 79.66403198242188,
      "learning_rate": 6.469400136362101e-05,
      "loss": 0.05586747527122497,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0487,
      "loss_step_3": 0.0396,
      "step": 14620
    },
    {
      "entropy_step_1": 0.2026,
      "entropy_step_2": 0.2036,
      "entropy_step_3": 0.2239,
      "epoch": 0.43514470122839893,
      "grad_norm": 56.53398132324219,
      "learning_rate": 6.46469848513092e-05,
      "loss": 0.06174602508544922,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0736,
      "step": 14630
    },
    {
      "entropy_step_1": 0.1897,
      "entropy_step_2": 0.1915,
      "entropy_step_3": 0.1835,
      "epoch": 0.43544213438029805,
      "grad_norm": 83.14220428466797,
      "learning_rate": 6.459995416784028e-05,
      "loss": 0.059312057495117185,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0528,
      "step": 14640
    },
    {
      "entropy_step_1": 0.1739,
      "entropy_step_2": 0.1788,
      "entropy_step_3": 0.1612,
      "epoch": 0.4357395675321971,
      "grad_norm": 40.01176834106445,
      "learning_rate": 6.4552909358717e-05,
      "loss": 0.05310713648796082,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.0394,
      "step": 14650
    },
    {
      "entropy_step_1": 0.1844,
      "entropy_step_2": 0.1891,
      "entropy_step_3": 0.1709,
      "epoch": 0.43603700068409623,
      "grad_norm": 265.8713073730469,
      "learning_rate": 6.450585046945591e-05,
      "loss": 0.05080358982086182,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0429,
      "step": 14660
    },
    {
      "entropy_step_1": 0.1821,
      "entropy_step_2": 0.1854,
      "entropy_step_3": 0.1937,
      "epoch": 0.43633443383599535,
      "grad_norm": 90.63864135742188,
      "learning_rate": 6.445877754558712e-05,
      "loss": 0.06954418420791626,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0701,
      "step": 14670
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.1949,
      "entropy_step_3": 0.1928,
      "epoch": 0.43663186698789447,
      "grad_norm": 99.60881805419922,
      "learning_rate": 6.44116906326543e-05,
      "loss": 0.05479980707168579,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0563,
      "step": 14680
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.182,
      "entropy_step_3": 0.1971,
      "epoch": 0.4369293001397936,
      "grad_norm": 154.5539093017578,
      "learning_rate": 6.436458977621472e-05,
      "loss": 0.06326367855072021,
      "loss_step_1": 0.0714,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0596,
      "step": 14690
    },
    {
      "entropy_step_1": 0.2106,
      "entropy_step_2": 0.2147,
      "entropy_step_3": 0.2061,
      "epoch": 0.4372267332916927,
      "grad_norm": 85.76837921142578,
      "learning_rate": 6.431747502183903e-05,
      "loss": 0.06458974480628968,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.061,
      "step": 14700
    },
    {
      "entropy_step_1": 0.2463,
      "entropy_step_2": 0.2472,
      "entropy_step_3": 0.2349,
      "epoch": 0.4375241664435918,
      "grad_norm": 2129.005859375,
      "learning_rate": 6.427034641511147e-05,
      "loss": 0.07359606623649598,
      "loss_step_1": 0.0707,
      "loss_step_2": 0.0708,
      "loss_step_3": 0.0671,
      "step": 14710
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.2045,
      "entropy_step_3": 0.2021,
      "epoch": 0.43782159959549094,
      "grad_norm": 101.39295196533203,
      "learning_rate": 6.422320400162958e-05,
      "loss": 0.06486339569091797,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0505,
      "step": 14720
    },
    {
      "entropy_step_1": 0.2099,
      "entropy_step_2": 0.2032,
      "entropy_step_3": 0.203,
      "epoch": 0.43811903274739,
      "grad_norm": 56.21049499511719,
      "learning_rate": 6.417604782700426e-05,
      "loss": 0.05810639262199402,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0567,
      "step": 14730
    },
    {
      "entropy_step_1": 0.2043,
      "entropy_step_2": 0.1982,
      "entropy_step_3": 0.1703,
      "epoch": 0.4384164658992891,
      "grad_norm": 181.27230834960938,
      "learning_rate": 6.41288779368598e-05,
      "loss": 0.057360851764678956,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0537,
      "step": 14740
    },
    {
      "entropy_step_1": 0.2005,
      "entropy_step_2": 0.1906,
      "entropy_step_3": 0.1942,
      "epoch": 0.43871389905118824,
      "grad_norm": 116.9753646850586,
      "learning_rate": 6.408169437683367e-05,
      "loss": 0.06506509780883789,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0652,
      "loss_step_3": 0.0612,
      "step": 14750
    },
    {
      "entropy_step_1": 0.2376,
      "entropy_step_2": 0.2269,
      "entropy_step_3": 0.232,
      "epoch": 0.43901133220308736,
      "grad_norm": 128.08473205566406,
      "learning_rate": 6.403449719257662e-05,
      "loss": 0.06704225540161132,
      "loss_step_1": 0.0727,
      "loss_step_2": 0.0707,
      "loss_step_3": 0.075,
      "step": 14760
    },
    {
      "entropy_step_1": 0.2229,
      "entropy_step_2": 0.2183,
      "entropy_step_3": 0.2088,
      "epoch": 0.4393087653549865,
      "grad_norm": 165.74391174316406,
      "learning_rate": 6.398728642975257e-05,
      "loss": 0.06546757221221924,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0557,
      "step": 14770
    },
    {
      "entropy_step_1": 0.205,
      "entropy_step_2": 0.2028,
      "entropy_step_3": 0.1889,
      "epoch": 0.4396061985068856,
      "grad_norm": 191.6743621826172,
      "learning_rate": 6.394006213403858e-05,
      "loss": 0.05886639952659607,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0472,
      "step": 14780
    },
    {
      "entropy_step_1": 0.1988,
      "entropy_step_2": 0.1972,
      "entropy_step_3": 0.1885,
      "epoch": 0.4399036316587847,
      "grad_norm": 138.897705078125,
      "learning_rate": 6.38928243511248e-05,
      "loss": 0.06315731406211852,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0556,
      "step": 14790
    },
    {
      "entropy_step_1": 0.1974,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.2068,
      "epoch": 0.4402010648106838,
      "grad_norm": 179.09222412109375,
      "learning_rate": 6.38455731267144e-05,
      "loss": 0.0612740159034729,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0503,
      "step": 14800
    },
    {
      "entropy_step_1": 0.2004,
      "entropy_step_2": 0.2056,
      "entropy_step_3": 0.1911,
      "epoch": 0.4404984979625829,
      "grad_norm": 219.40328979492188,
      "learning_rate": 6.379830850652363e-05,
      "loss": 0.06551380753517151,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0648,
      "step": 14810
    },
    {
      "entropy_step_1": 0.1999,
      "entropy_step_2": 0.206,
      "entropy_step_3": 0.1883,
      "epoch": 0.440795931114482,
      "grad_norm": 275.4924621582031,
      "learning_rate": 6.375103053628161e-05,
      "loss": 0.06198967099189758,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.051,
      "step": 14820
    },
    {
      "entropy_step_1": 0.205,
      "entropy_step_2": 0.2098,
      "entropy_step_3": 0.1913,
      "epoch": 0.4410933642663811,
      "grad_norm": 62.99348831176758,
      "learning_rate": 6.370373926173044e-05,
      "loss": 0.06341620683670043,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0598,
      "step": 14830
    },
    {
      "entropy_step_1": 0.1842,
      "entropy_step_2": 0.1891,
      "entropy_step_3": 0.1868,
      "epoch": 0.44139079741828025,
      "grad_norm": 86.83837127685547,
      "learning_rate": 6.365643472862506e-05,
      "loss": 0.061961591243743896,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0584,
      "step": 14840
    },
    {
      "entropy_step_1": 0.2396,
      "entropy_step_2": 0.2428,
      "entropy_step_3": 0.236,
      "epoch": 0.44168823057017936,
      "grad_norm": 286.32171630859375,
      "learning_rate": 6.360911698273328e-05,
      "loss": 0.06786321997642517,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.0505,
      "step": 14850
    },
    {
      "entropy_step_1": 0.2347,
      "entropy_step_2": 0.2374,
      "entropy_step_3": 0.2476,
      "epoch": 0.4419856637220785,
      "grad_norm": 33.68107986450195,
      "learning_rate": 6.356178606983562e-05,
      "loss": 0.06900238394737243,
      "loss_step_1": 0.0741,
      "loss_step_2": 0.0723,
      "loss_step_3": 0.0735,
      "step": 14860
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1807,
      "entropy_step_3": 0.18,
      "epoch": 0.4422830968739776,
      "grad_norm": 71.35676574707031,
      "learning_rate": 6.351444203572541e-05,
      "loss": 0.06278805136680603,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.0676,
      "loss_step_3": 0.0698,
      "step": 14870
    },
    {
      "entropy_step_1": 0.2083,
      "entropy_step_2": 0.2089,
      "entropy_step_3": 0.2293,
      "epoch": 0.44258053002587666,
      "grad_norm": 129.43234252929688,
      "learning_rate": 6.346708492620864e-05,
      "loss": 0.06108798384666443,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0756,
      "step": 14880
    },
    {
      "entropy_step_1": 0.21,
      "entropy_step_2": 0.2139,
      "entropy_step_3": 0.2058,
      "epoch": 0.4428779631777758,
      "grad_norm": 110.30997467041016,
      "learning_rate": 6.341971478710393e-05,
      "loss": 0.06444236040115356,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.058,
      "step": 14890
    },
    {
      "entropy_step_1": 0.1801,
      "entropy_step_2": 0.1827,
      "entropy_step_3": 0.1676,
      "epoch": 0.4431753963296749,
      "grad_norm": 67.6080551147461,
      "learning_rate": 6.33723316642426e-05,
      "loss": 0.05776381492614746,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0478,
      "step": 14900
    },
    {
      "entropy_step_1": 0.2141,
      "entropy_step_2": 0.2161,
      "entropy_step_3": 0.2221,
      "epoch": 0.443472829481574,
      "grad_norm": 231.29244995117188,
      "learning_rate": 6.332493560346842e-05,
      "loss": 0.06661579012870789,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.0695,
      "step": 14910
    },
    {
      "entropy_step_1": 0.2107,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.1915,
      "epoch": 0.44377026263347313,
      "grad_norm": 48.293418884277344,
      "learning_rate": 6.327752665063773e-05,
      "loss": 0.06155579686164856,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0491,
      "step": 14920
    },
    {
      "entropy_step_1": 0.2305,
      "entropy_step_2": 0.2321,
      "entropy_step_3": 0.2245,
      "epoch": 0.44406769578537225,
      "grad_norm": 404.4087829589844,
      "learning_rate": 6.323010485161932e-05,
      "loss": 0.0672533631324768,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0642,
      "step": 14930
    },
    {
      "entropy_step_1": 0.1769,
      "entropy_step_2": 0.1797,
      "entropy_step_3": 0.1839,
      "epoch": 0.44436512893727137,
      "grad_norm": 176.1958770751953,
      "learning_rate": 6.318267025229447e-05,
      "loss": 0.05437643527984619,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0673,
      "step": 14940
    },
    {
      "entropy_step_1": 0.2234,
      "entropy_step_2": 0.2218,
      "entropy_step_3": 0.2066,
      "epoch": 0.44466256208917043,
      "grad_norm": 136.20188903808594,
      "learning_rate": 6.313522289855676e-05,
      "loss": 0.06686396598815918,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0608,
      "step": 14950
    },
    {
      "entropy_step_1": 0.2162,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.1985,
      "epoch": 0.44495999524106955,
      "grad_norm": 80.1042709350586,
      "learning_rate": 6.308776283631217e-05,
      "loss": 0.06252679228782654,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.055,
      "step": 14960
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.2011,
      "epoch": 0.44525742839296867,
      "grad_norm": 25.603918075561523,
      "learning_rate": 6.304029011147895e-05,
      "loss": 0.06522538661956787,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0627,
      "step": 14970
    },
    {
      "entropy_step_1": 0.1906,
      "entropy_step_2": 0.1822,
      "entropy_step_3": 0.1739,
      "epoch": 0.4455548615448678,
      "grad_norm": 63.79719924926758,
      "learning_rate": 6.299280476998761e-05,
      "loss": 0.054908227920532224,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.051,
      "step": 14980
    },
    {
      "entropy_step_1": 0.1905,
      "entropy_step_2": 0.1819,
      "entropy_step_3": 0.1869,
      "epoch": 0.4458522946967669,
      "grad_norm": 89.00721740722656,
      "learning_rate": 6.294530685778086e-05,
      "loss": 0.06351138949394226,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0499,
      "step": 14990
    },
    {
      "entropy_step_1": 0.2021,
      "entropy_step_2": 0.1893,
      "entropy_step_3": 0.1953,
      "epoch": 0.446149727848666,
      "grad_norm": 102.64193725585938,
      "learning_rate": 6.289779642081358e-05,
      "loss": 0.0650702178478241,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0681,
      "step": 15000
    },
    {
      "epoch": 0.446149727848666,
      "eval_accuracy": 0.855,
      "eval_f1": 0.8540893717692116,
      "eval_loss": 0.14843830466270447,
      "eval_precision": 0.8640891208119494,
      "eval_recall": 0.855,
      "eval_runtime": 6.7079,
      "eval_samples_per_second": 74.539,
      "eval_steps_per_second": 9.392,
      "step": 15000
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.1886,
      "entropy_step_3": 0.1812,
      "epoch": 0.44644716100056514,
      "grad_norm": 137.91494750976562,
      "learning_rate": 6.285027350505278e-05,
      "loss": 0.05766648054122925,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0549,
      "step": 15010
    },
    {
      "entropy_step_1": 0.1936,
      "entropy_step_2": 0.1859,
      "entropy_step_3": 0.1918,
      "epoch": 0.44674459415246426,
      "grad_norm": 101.07596588134766,
      "learning_rate": 6.280273815647748e-05,
      "loss": 0.05801478624343872,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0612,
      "step": 15020
    },
    {
      "entropy_step_1": 0.1844,
      "entropy_step_2": 0.1786,
      "entropy_step_3": 0.1742,
      "epoch": 0.4470420273043633,
      "grad_norm": 1125.7486572265625,
      "learning_rate": 6.275519042107883e-05,
      "loss": 0.0545143723487854,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0454,
      "step": 15030
    },
    {
      "entropy_step_1": 0.1693,
      "entropy_step_2": 0.1653,
      "entropy_step_3": 0.1688,
      "epoch": 0.44733946045626244,
      "grad_norm": 282.87310791015625,
      "learning_rate": 6.270763034485987e-05,
      "loss": 0.0565409779548645,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0531,
      "step": 15040
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1843,
      "entropy_step_3": 0.1806,
      "epoch": 0.44763689360816156,
      "grad_norm": 73.58560943603516,
      "learning_rate": 6.266005797383564e-05,
      "loss": 0.05732893943786621,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.058,
      "loss_step_3": 0.058,
      "step": 15050
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1824,
      "entropy_step_3": 0.1834,
      "epoch": 0.4479343267600607,
      "grad_norm": 228.90414428710938,
      "learning_rate": 6.261247335403306e-05,
      "loss": 0.07118995785713196,
      "loss_step_1": 0.0701,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0805,
      "step": 15060
    },
    {
      "entropy_step_1": 0.2095,
      "entropy_step_2": 0.209,
      "entropy_step_3": 0.194,
      "epoch": 0.4482317599119598,
      "grad_norm": 73.57329559326172,
      "learning_rate": 6.25648765314909e-05,
      "loss": 0.06014705896377563,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0596,
      "step": 15070
    },
    {
      "entropy_step_1": 0.1923,
      "entropy_step_2": 0.1985,
      "entropy_step_3": 0.2128,
      "epoch": 0.4485291930638589,
      "grad_norm": 51.356353759765625,
      "learning_rate": 6.251726755225972e-05,
      "loss": 0.06257429122924804,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0649,
      "step": 15080
    },
    {
      "entropy_step_1": 0.2041,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.1931,
      "epoch": 0.44882662621575803,
      "grad_norm": 54.1783332824707,
      "learning_rate": 6.246964646240185e-05,
      "loss": 0.06326314806938171,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0662,
      "loss_step_3": 0.0571,
      "step": 15090
    },
    {
      "entropy_step_1": 0.2015,
      "entropy_step_2": 0.2045,
      "entropy_step_3": 0.2112,
      "epoch": 0.4491240593676571,
      "grad_norm": 105.95112609863281,
      "learning_rate": 6.242201330799138e-05,
      "loss": 0.0613541305065155,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0643,
      "step": 15100
    },
    {
      "entropy_step_1": 0.1971,
      "entropy_step_2": 0.197,
      "entropy_step_3": 0.2034,
      "epoch": 0.4494214925195562,
      "grad_norm": 184.0570526123047,
      "learning_rate": 6.2374368135114e-05,
      "loss": 0.06009278893470764,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0657,
      "step": 15110
    },
    {
      "entropy_step_1": 0.1769,
      "entropy_step_2": 0.1785,
      "entropy_step_3": 0.1838,
      "epoch": 0.4497189256714553,
      "grad_norm": 151.35565185546875,
      "learning_rate": 6.232671098986705e-05,
      "loss": 0.05080335140228272,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.0488,
      "step": 15120
    },
    {
      "entropy_step_1": 0.1846,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1864,
      "epoch": 0.45001635882335445,
      "grad_norm": 173.8345947265625,
      "learning_rate": 6.227904191835952e-05,
      "loss": 0.061785298585891726,
      "loss_step_1": 0.0692,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0661,
      "step": 15130
    },
    {
      "entropy_step_1": 0.2052,
      "entropy_step_2": 0.2068,
      "entropy_step_3": 0.2142,
      "epoch": 0.45031379197525356,
      "grad_norm": 70.1549072265625,
      "learning_rate": 6.223136096671182e-05,
      "loss": 0.06562177538871765,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0661,
      "step": 15140
    },
    {
      "entropy_step_1": 0.2596,
      "entropy_step_2": 0.2612,
      "entropy_step_3": 0.2796,
      "epoch": 0.4506112251271527,
      "grad_norm": 75.69319152832031,
      "learning_rate": 6.218366818105597e-05,
      "loss": 0.06722786426544189,
      "loss_step_1": 0.0696,
      "loss_step_2": 0.0687,
      "loss_step_3": 0.0758,
      "step": 15150
    },
    {
      "entropy_step_1": 0.1892,
      "entropy_step_2": 0.1889,
      "entropy_step_3": 0.1784,
      "epoch": 0.4509086582790518,
      "grad_norm": 110.43570709228516,
      "learning_rate": 6.213596360753539e-05,
      "loss": 0.0603520929813385,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0631,
      "step": 15160
    },
    {
      "entropy_step_1": 0.1932,
      "entropy_step_2": 0.1908,
      "entropy_step_3": 0.1914,
      "epoch": 0.4512060914309509,
      "grad_norm": 124.68074798583984,
      "learning_rate": 6.208824729230487e-05,
      "loss": 0.06071839928627014,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0625,
      "loss_step_3": 0.077,
      "step": 15170
    },
    {
      "entropy_step_1": 0.2281,
      "entropy_step_2": 0.2199,
      "entropy_step_3": 0.2176,
      "epoch": 0.45150352458285,
      "grad_norm": 39.42001724243164,
      "learning_rate": 6.204051928153061e-05,
      "loss": 0.059786629676818845,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0591,
      "step": 15180
    },
    {
      "entropy_step_1": 0.2208,
      "entropy_step_2": 0.2106,
      "entropy_step_3": 0.2087,
      "epoch": 0.4518009577347491,
      "grad_norm": 95.51270294189453,
      "learning_rate": 6.199277962139011e-05,
      "loss": 0.07008896470069885,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0574,
      "step": 15190
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1724,
      "entropy_step_3": 0.1486,
      "epoch": 0.4520983908866482,
      "grad_norm": 102.8482437133789,
      "learning_rate": 6.194502835807213e-05,
      "loss": 0.0625311255455017,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0671,
      "step": 15200
    },
    {
      "entropy_step_1": 0.2547,
      "entropy_step_2": 0.2414,
      "entropy_step_3": 0.2527,
      "epoch": 0.45239582403854733,
      "grad_norm": 71.62409210205078,
      "learning_rate": 6.189726553777668e-05,
      "loss": 0.07591454982757569,
      "loss_step_1": 0.069,
      "loss_step_2": 0.07,
      "loss_step_3": 0.0656,
      "step": 15210
    },
    {
      "entropy_step_1": 0.2086,
      "entropy_step_2": 0.2089,
      "entropy_step_3": 0.2098,
      "epoch": 0.45269325719044645,
      "grad_norm": 74.06119537353516,
      "learning_rate": 6.184949120671495e-05,
      "loss": 0.06666609644889832,
      "loss_step_1": 0.0711,
      "loss_step_2": 0.0712,
      "loss_step_3": 0.0715,
      "step": 15220
    },
    {
      "entropy_step_1": 0.1768,
      "entropy_step_2": 0.1835,
      "entropy_step_3": 0.1794,
      "epoch": 0.45299069034234557,
      "grad_norm": 54.184730529785156,
      "learning_rate": 6.180170541110923e-05,
      "loss": 0.06074885725975036,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.052,
      "step": 15230
    },
    {
      "entropy_step_1": 0.1852,
      "entropy_step_2": 0.2028,
      "entropy_step_3": 0.191,
      "epoch": 0.4532881234942447,
      "grad_norm": 203.12478637695312,
      "learning_rate": 6.175390819719295e-05,
      "loss": 0.05890145301818848,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.05,
      "step": 15240
    },
    {
      "entropy_step_1": 0.1884,
      "entropy_step_2": 0.2016,
      "entropy_step_3": 0.2018,
      "epoch": 0.4535855566461438,
      "grad_norm": 309.8692626953125,
      "learning_rate": 6.170609961121054e-05,
      "loss": 0.05931199789047241,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0541,
      "step": 15250
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2284,
      "epoch": 0.45388298979804287,
      "grad_norm": 88.83383178710938,
      "learning_rate": 6.165827969941749e-05,
      "loss": 0.055824381113052365,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0464,
      "step": 15260
    },
    {
      "entropy_step_1": 0.1739,
      "entropy_step_2": 0.1844,
      "entropy_step_3": 0.1834,
      "epoch": 0.454180422949942,
      "grad_norm": 106.31608581542969,
      "learning_rate": 6.161044850808019e-05,
      "loss": 0.0676031768321991,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0602,
      "step": 15270
    },
    {
      "entropy_step_1": 0.194,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.1888,
      "epoch": 0.4544778561018411,
      "grad_norm": 80.88507080078125,
      "learning_rate": 6.156260608347596e-05,
      "loss": 0.06506986618041992,
      "loss_step_1": 0.0751,
      "loss_step_2": 0.0742,
      "loss_step_3": 0.0795,
      "step": 15280
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.2227,
      "entropy_step_3": 0.2148,
      "epoch": 0.4547752892537402,
      "grad_norm": 133.88075256347656,
      "learning_rate": 6.151475247189303e-05,
      "loss": 0.05860433578491211,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0604,
      "step": 15290
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.2053,
      "entropy_step_3": 0.2022,
      "epoch": 0.45507272240563934,
      "grad_norm": 93.36676025390625,
      "learning_rate": 6.14668877196304e-05,
      "loss": 0.059510540962219236,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0638,
      "step": 15300
    },
    {
      "entropy_step_1": 0.1831,
      "entropy_step_2": 0.1884,
      "entropy_step_3": 0.1891,
      "epoch": 0.45537015555753846,
      "grad_norm": 60.173526763916016,
      "learning_rate": 6.141901187299787e-05,
      "loss": 0.06367123723030091,
      "loss_step_1": 0.0668,
      "loss_step_2": 0.0664,
      "loss_step_3": 0.0659,
      "step": 15310
    },
    {
      "entropy_step_1": 0.1855,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.189,
      "epoch": 0.4556675887094376,
      "grad_norm": 86.61966705322266,
      "learning_rate": 6.137112497831598e-05,
      "loss": 0.05873661041259766,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0519,
      "step": 15320
    },
    {
      "entropy_step_1": 0.1845,
      "entropy_step_2": 0.1857,
      "entropy_step_3": 0.1895,
      "epoch": 0.45596502186133664,
      "grad_norm": 63.9205207824707,
      "learning_rate": 6.132322708191593e-05,
      "loss": 0.051580005884170534,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.0486,
      "step": 15330
    },
    {
      "entropy_step_1": 0.2016,
      "entropy_step_2": 0.1977,
      "entropy_step_3": 0.2074,
      "epoch": 0.45626245501323576,
      "grad_norm": 76.74940490722656,
      "learning_rate": 6.127531823013964e-05,
      "loss": 0.07142043113708496,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.066,
      "loss_step_3": 0.0613,
      "step": 15340
    },
    {
      "entropy_step_1": 0.195,
      "entropy_step_2": 0.1911,
      "entropy_step_3": 0.2044,
      "epoch": 0.4565598881651349,
      "grad_norm": 47.7672119140625,
      "learning_rate": 6.122739846933949e-05,
      "loss": 0.06436742544174194,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0685,
      "step": 15350
    },
    {
      "entropy_step_1": 0.2146,
      "entropy_step_2": 0.2155,
      "entropy_step_3": 0.2061,
      "epoch": 0.456857321317034,
      "grad_norm": 89.78275299072266,
      "learning_rate": 6.117946784587858e-05,
      "loss": 0.05529999136924744,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0572,
      "step": 15360
    },
    {
      "entropy_step_1": 0.1763,
      "entropy_step_2": 0.1801,
      "entropy_step_3": 0.1912,
      "epoch": 0.4571547544689331,
      "grad_norm": 105.66117858886719,
      "learning_rate": 6.11315264061304e-05,
      "loss": 0.061842328310012816,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0618,
      "step": 15370
    },
    {
      "entropy_step_1": 0.1979,
      "entropy_step_2": 0.2035,
      "entropy_step_3": 0.2018,
      "epoch": 0.45745218762083223,
      "grad_norm": 138.17852783203125,
      "learning_rate": 6.108357419647896e-05,
      "loss": 0.0655186653137207,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0503,
      "step": 15380
    },
    {
      "entropy_step_1": 0.2266,
      "entropy_step_2": 0.229,
      "entropy_step_3": 0.2309,
      "epoch": 0.45774962077273135,
      "grad_norm": 127.70921325683594,
      "learning_rate": 6.103561126331868e-05,
      "loss": 0.06000100374221802,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0608,
      "step": 15390
    },
    {
      "entropy_step_1": 0.1763,
      "entropy_step_2": 0.1815,
      "entropy_step_3": 0.1705,
      "epoch": 0.45804705392463047,
      "grad_norm": 82.0830078125,
      "learning_rate": 6.098763765305433e-05,
      "loss": 0.056957566738128663,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0479,
      "step": 15400
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.2057,
      "entropy_step_3": 0.2046,
      "epoch": 0.4583444870765295,
      "grad_norm": 60.65046310424805,
      "learning_rate": 6.093965341210104e-05,
      "loss": 0.05687534809112549,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0556,
      "step": 15410
    },
    {
      "entropy_step_1": 0.1953,
      "entropy_step_2": 0.1958,
      "entropy_step_3": 0.1999,
      "epoch": 0.45864192022842865,
      "grad_norm": 73.78093719482422,
      "learning_rate": 6.089165858688424e-05,
      "loss": 0.05679064989089966,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0612,
      "step": 15420
    },
    {
      "entropy_step_1": 0.1674,
      "entropy_step_2": 0.1644,
      "entropy_step_3": 0.1489,
      "epoch": 0.45893935338032776,
      "grad_norm": 204.95423889160156,
      "learning_rate": 6.084365322383953e-05,
      "loss": 0.05500693321228027,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0451,
      "step": 15430
    },
    {
      "entropy_step_1": 0.1922,
      "entropy_step_2": 0.1895,
      "entropy_step_3": 0.1924,
      "epoch": 0.4592367865322269,
      "grad_norm": 102.36491394042969,
      "learning_rate": 6.079563736941277e-05,
      "loss": 0.05620257258415222,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.068,
      "step": 15440
    },
    {
      "entropy_step_1": 0.1857,
      "entropy_step_2": 0.1876,
      "entropy_step_3": 0.1788,
      "epoch": 0.459534219684126,
      "grad_norm": 68.38166809082031,
      "learning_rate": 6.074761107005998e-05,
      "loss": 0.060596692562103274,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0569,
      "step": 15450
    },
    {
      "entropy_step_1": 0.2106,
      "entropy_step_2": 0.2189,
      "entropy_step_3": 0.2168,
      "epoch": 0.4598316528360251,
      "grad_norm": 1203.0147705078125,
      "learning_rate": 6.069957437224722e-05,
      "loss": 0.06441051363945008,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.0666,
      "step": 15460
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1857,
      "entropy_step_3": 0.185,
      "epoch": 0.46012908598792424,
      "grad_norm": 207.55735778808594,
      "learning_rate": 6.06515273224507e-05,
      "loss": 0.05889838933944702,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.053,
      "loss_step_3": 0.047,
      "step": 15470
    },
    {
      "entropy_step_1": 0.2014,
      "entropy_step_2": 0.2082,
      "entropy_step_3": 0.2006,
      "epoch": 0.4604265191398233,
      "grad_norm": 224.6645050048828,
      "learning_rate": 6.060346996715652e-05,
      "loss": 0.06026415228843689,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0653,
      "step": 15480
    },
    {
      "entropy_step_1": 0.1991,
      "entropy_step_2": 0.2065,
      "entropy_step_3": 0.2195,
      "epoch": 0.4607239522917224,
      "grad_norm": 278.9127197265625,
      "learning_rate": 6.0555402352860893e-05,
      "loss": 0.05575082302093506,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0566,
      "step": 15490
    },
    {
      "entropy_step_1": 0.1744,
      "entropy_step_2": 0.1806,
      "entropy_step_3": 0.164,
      "epoch": 0.46102138544362153,
      "grad_norm": 109.96845245361328,
      "learning_rate": 6.050732452606985e-05,
      "loss": 0.06095147728919983,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0482,
      "step": 15500
    },
    {
      "entropy_step_1": 0.1944,
      "entropy_step_2": 0.2023,
      "entropy_step_3": 0.1927,
      "epoch": 0.46131881859552065,
      "grad_norm": 145.16934204101562,
      "learning_rate": 6.045923653329937e-05,
      "loss": 0.061266982555389406,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0579,
      "step": 15510
    },
    {
      "entropy_step_1": 0.215,
      "entropy_step_2": 0.2227,
      "entropy_step_3": 0.2408,
      "epoch": 0.46161625174741977,
      "grad_norm": 36.02292251586914,
      "learning_rate": 6.0411138421075195e-05,
      "loss": 0.06970303058624268,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0714,
      "step": 15520
    },
    {
      "entropy_step_1": 0.2026,
      "entropy_step_2": 0.2095,
      "entropy_step_3": 0.2166,
      "epoch": 0.4619136848993189,
      "grad_norm": 62.49000549316406,
      "learning_rate": 6.036303023593296e-05,
      "loss": 0.05807809829711914,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0522,
      "loss_step_3": 0.0556,
      "step": 15530
    },
    {
      "entropy_step_1": 0.1929,
      "entropy_step_2": 0.1982,
      "entropy_step_3": 0.2073,
      "epoch": 0.462211118051218,
      "grad_norm": 88.59056091308594,
      "learning_rate": 6.031491202441797e-05,
      "loss": 0.05784912109375,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0734,
      "step": 15540
    },
    {
      "entropy_step_1": 0.1931,
      "entropy_step_2": 0.2015,
      "entropy_step_3": 0.213,
      "epoch": 0.4625085512031171,
      "grad_norm": 163.4883270263672,
      "learning_rate": 6.026678383308522e-05,
      "loss": 0.060122203826904294,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0478,
      "step": 15550
    },
    {
      "entropy_step_1": 0.1692,
      "entropy_step_2": 0.1768,
      "entropy_step_3": 0.1923,
      "epoch": 0.4628059843550162,
      "grad_norm": 69.49646759033203,
      "learning_rate": 6.0218645708499424e-05,
      "loss": 0.055084329843521115,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0548,
      "step": 15560
    },
    {
      "entropy_step_1": 0.2049,
      "entropy_step_2": 0.208,
      "entropy_step_3": 0.2143,
      "epoch": 0.4631034175069153,
      "grad_norm": 118.72130584716797,
      "learning_rate": 6.017049769723484e-05,
      "loss": 0.071623694896698,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0722,
      "step": 15570
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2083,
      "entropy_step_3": 0.2103,
      "epoch": 0.4634008506588144,
      "grad_norm": 106.35348510742188,
      "learning_rate": 6.0122339845875364e-05,
      "loss": 0.05961090326309204,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0596,
      "step": 15580
    },
    {
      "entropy_step_1": 0.1825,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1727,
      "epoch": 0.46369828381071354,
      "grad_norm": 248.5636444091797,
      "learning_rate": 6.007417220101435e-05,
      "loss": 0.053947484493255614,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0527,
      "loss_step_3": 0.0451,
      "step": 15590
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.1793,
      "entropy_step_3": 0.1699,
      "epoch": 0.46399571696261266,
      "grad_norm": 69.8867416381836,
      "learning_rate": 6.0025994809254635e-05,
      "loss": 0.05659646987915039,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0592,
      "loss_step_3": 0.0629,
      "step": 15600
    },
    {
      "entropy_step_1": 0.2021,
      "entropy_step_2": 0.2016,
      "entropy_step_3": 0.1968,
      "epoch": 0.4642931501145118,
      "grad_norm": 185.11605834960938,
      "learning_rate": 5.9977807717208546e-05,
      "loss": 0.06905574798583984,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0663,
      "loss_step_3": 0.0684,
      "step": 15610
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.195,
      "entropy_step_3": 0.2187,
      "epoch": 0.4645905832664109,
      "grad_norm": 149.5734100341797,
      "learning_rate": 5.9929610971497693e-05,
      "loss": 0.05815905928611755,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0637,
      "step": 15620
    },
    {
      "entropy_step_1": 0.1864,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1822,
      "epoch": 0.46488801641830996,
      "grad_norm": 40.97689437866211,
      "learning_rate": 5.9881404618753146e-05,
      "loss": 0.05854865908622742,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0556,
      "step": 15630
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1888,
      "entropy_step_3": 0.2018,
      "epoch": 0.4651854495702091,
      "grad_norm": 276.6614074707031,
      "learning_rate": 5.983318870561515e-05,
      "loss": 0.0627337634563446,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.05,
      "step": 15640
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1883,
      "entropy_step_3": 0.178,
      "epoch": 0.4654828827221082,
      "grad_norm": 276.271728515625,
      "learning_rate": 5.97849632787333e-05,
      "loss": 0.05566244721412659,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.057,
      "step": 15650
    },
    {
      "entropy_step_1": 0.1957,
      "entropy_step_2": 0.1899,
      "entropy_step_3": 0.2053,
      "epoch": 0.4657803158740073,
      "grad_norm": 111.893310546875,
      "learning_rate": 5.973672838476632e-05,
      "loss": 0.06276767253875733,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0685,
      "loss_step_3": 0.0737,
      "step": 15660
    },
    {
      "entropy_step_1": 0.2137,
      "entropy_step_2": 0.2125,
      "entropy_step_3": 0.2025,
      "epoch": 0.46607774902590643,
      "grad_norm": 275.874267578125,
      "learning_rate": 5.9688484070382135e-05,
      "loss": 0.06904608011245728,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.0614,
      "step": 15670
    },
    {
      "entropy_step_1": 0.2043,
      "entropy_step_2": 0.2068,
      "entropy_step_3": 0.2144,
      "epoch": 0.46637518217780555,
      "grad_norm": 44.41139221191406,
      "learning_rate": 5.964023038225779e-05,
      "loss": 0.05913774967193604,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0691,
      "step": 15680
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1799,
      "entropy_step_3": 0.1707,
      "epoch": 0.46667261532970467,
      "grad_norm": 340.8150939941406,
      "learning_rate": 5.9591967367079384e-05,
      "loss": 0.060050463676452635,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0575,
      "step": 15690
    },
    {
      "entropy_step_1": 0.1711,
      "entropy_step_2": 0.1734,
      "entropy_step_3": 0.165,
      "epoch": 0.4669700484816038,
      "grad_norm": 105.15835571289062,
      "learning_rate": 5.9543695071542026e-05,
      "loss": 0.05805985331535339,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0456,
      "step": 15700
    },
    {
      "entropy_step_1": 0.2251,
      "entropy_step_2": 0.2271,
      "entropy_step_3": 0.234,
      "epoch": 0.46726748163350285,
      "grad_norm": 87.28971862792969,
      "learning_rate": 5.949541354234982e-05,
      "loss": 0.06342824697494506,
      "loss_step_1": 0.0709,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.0759,
      "step": 15710
    },
    {
      "entropy_step_1": 0.1832,
      "entropy_step_2": 0.187,
      "entropy_step_3": 0.1958,
      "epoch": 0.46756491478540196,
      "grad_norm": 65.91068267822266,
      "learning_rate": 5.944712282621583e-05,
      "loss": 0.057448720932006835,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.061,
      "step": 15720
    },
    {
      "entropy_step_1": 0.1786,
      "entropy_step_2": 0.1856,
      "entropy_step_3": 0.1897,
      "epoch": 0.4678623479373011,
      "grad_norm": 57.13606262207031,
      "learning_rate": 5.939882296986193e-05,
      "loss": 0.07018072605133056,
      "loss_step_1": 0.0778,
      "loss_step_2": 0.0757,
      "loss_step_3": 0.0718,
      "step": 15730
    },
    {
      "entropy_step_1": 0.2029,
      "entropy_step_2": 0.2085,
      "entropy_step_3": 0.206,
      "epoch": 0.4681597810892002,
      "grad_norm": 79.62398529052734,
      "learning_rate": 5.935051402001893e-05,
      "loss": 0.06389217376708985,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0596,
      "step": 15740
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.2042,
      "entropy_step_3": 0.2071,
      "epoch": 0.4684572142410993,
      "grad_norm": 39.47242736816406,
      "learning_rate": 5.930219602342638e-05,
      "loss": 0.061096596717834475,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0551,
      "step": 15750
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.2309,
      "epoch": 0.46875464739299844,
      "grad_norm": 110.09537506103516,
      "learning_rate": 5.9253869026832595e-05,
      "loss": 0.06040909290313721,
      "loss_step_1": 0.0682,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.0772,
      "step": 15760
    },
    {
      "entropy_step_1": 0.1766,
      "entropy_step_2": 0.1812,
      "entropy_step_3": 0.1792,
      "epoch": 0.46905208054489755,
      "grad_norm": 65.3311996459961,
      "learning_rate": 5.9205533076994635e-05,
      "loss": 0.06051374077796936,
      "loss_step_1": 0.0629,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.063,
      "step": 15770
    },
    {
      "entropy_step_1": 0.1806,
      "entropy_step_2": 0.185,
      "entropy_step_3": 0.2041,
      "epoch": 0.46934951369679667,
      "grad_norm": 81.7621078491211,
      "learning_rate": 5.915718822067814e-05,
      "loss": 0.05539113879203796,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0527,
      "step": 15780
    },
    {
      "entropy_step_1": 0.1833,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.1693,
      "epoch": 0.46964694684869573,
      "grad_norm": 123.95222473144531,
      "learning_rate": 5.910883450465745e-05,
      "loss": 0.0640458345413208,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0495,
      "step": 15790
    },
    {
      "entropy_step_1": 0.2217,
      "entropy_step_2": 0.2211,
      "entropy_step_3": 0.2158,
      "epoch": 0.46994438000059485,
      "grad_norm": 82.5048828125,
      "learning_rate": 5.906047197571543e-05,
      "loss": 0.06253222227096558,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0431,
      "step": 15800
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1699,
      "epoch": 0.47024181315249397,
      "grad_norm": 216.6564483642578,
      "learning_rate": 5.9012100680643486e-05,
      "loss": 0.0584510326385498,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0493,
      "step": 15810
    },
    {
      "entropy_step_1": 0.2068,
      "entropy_step_2": 0.2025,
      "entropy_step_3": 0.2077,
      "epoch": 0.4705392463043931,
      "grad_norm": 243.41390991210938,
      "learning_rate": 5.8963720666241504e-05,
      "loss": 0.06237548589706421,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0552,
      "step": 15820
    },
    {
      "entropy_step_1": 0.2232,
      "entropy_step_2": 0.2168,
      "entropy_step_3": 0.1931,
      "epoch": 0.4708366794562922,
      "grad_norm": 199.99920654296875,
      "learning_rate": 5.891533197931781e-05,
      "loss": 0.07157328724861145,
      "loss_step_1": 0.073,
      "loss_step_2": 0.0722,
      "loss_step_3": 0.0719,
      "step": 15830
    },
    {
      "entropy_step_1": 0.2273,
      "entropy_step_2": 0.2157,
      "entropy_step_3": 0.2242,
      "epoch": 0.4711341126081913,
      "grad_norm": 110.72489929199219,
      "learning_rate": 5.886693466668912e-05,
      "loss": 0.061580151319503784,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0568,
      "step": 15840
    },
    {
      "entropy_step_1": 0.2131,
      "entropy_step_2": 0.2074,
      "entropy_step_3": 0.206,
      "epoch": 0.47143154576009044,
      "grad_norm": 105.93952178955078,
      "learning_rate": 5.881852877518048e-05,
      "loss": 0.06166269183158875,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0634,
      "step": 15850
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1855,
      "entropy_step_3": 0.1961,
      "epoch": 0.4717289789119895,
      "grad_norm": 320.7995300292969,
      "learning_rate": 5.877011435162527e-05,
      "loss": 0.05568305253982544,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.056,
      "step": 15860
    },
    {
      "entropy_step_1": 0.1885,
      "entropy_step_2": 0.1848,
      "entropy_step_3": 0.1907,
      "epoch": 0.4720264120638886,
      "grad_norm": 108.15721893310547,
      "learning_rate": 5.872169144286508e-05,
      "loss": 0.054564809799194335,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0558,
      "step": 15870
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1924,
      "entropy_step_3": 0.1866,
      "epoch": 0.47232384521578774,
      "grad_norm": 56.75605010986328,
      "learning_rate": 5.8673260095749746e-05,
      "loss": 0.06490077376365662,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0719,
      "step": 15880
    },
    {
      "entropy_step_1": 0.2183,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.2205,
      "epoch": 0.47262127836768686,
      "grad_norm": 61.52631378173828,
      "learning_rate": 5.862482035713727e-05,
      "loss": 0.06296765804290771,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0594,
      "step": 15890
    },
    {
      "entropy_step_1": 0.2042,
      "entropy_step_2": 0.2016,
      "entropy_step_3": 0.2007,
      "epoch": 0.472918711519586,
      "grad_norm": 151.9186248779297,
      "learning_rate": 5.857637227389371e-05,
      "loss": 0.05345821976661682,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0549,
      "step": 15900
    },
    {
      "entropy_step_1": 0.1862,
      "entropy_step_2": 0.1808,
      "entropy_step_3": 0.1788,
      "epoch": 0.4732161446714851,
      "grad_norm": 121.28878021240234,
      "learning_rate": 5.8527915892893303e-05,
      "loss": 0.0703115463256836,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0674,
      "loss_step_3": 0.0638,
      "step": 15910
    },
    {
      "entropy_step_1": 0.1805,
      "entropy_step_2": 0.1755,
      "entropy_step_3": 0.1803,
      "epoch": 0.4735135778233842,
      "grad_norm": 77.3372573852539,
      "learning_rate": 5.847945126101821e-05,
      "loss": 0.054128462076187135,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0539,
      "step": 15920
    },
    {
      "entropy_step_1": 0.2245,
      "entropy_step_2": 0.2189,
      "entropy_step_3": 0.2198,
      "epoch": 0.47381101097528333,
      "grad_norm": 164.8209991455078,
      "learning_rate": 5.843097842515866e-05,
      "loss": 0.06932728290557862,
      "loss_step_1": 0.0715,
      "loss_step_2": 0.071,
      "loss_step_3": 0.0706,
      "step": 15930
    },
    {
      "entropy_step_1": 0.169,
      "entropy_step_2": 0.1718,
      "entropy_step_3": 0.1509,
      "epoch": 0.4741084441271824,
      "grad_norm": 574.3944702148438,
      "learning_rate": 5.838249743221275e-05,
      "loss": 0.06466525197029113,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0529,
      "step": 15940
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1864,
      "entropy_step_3": 0.1945,
      "epoch": 0.4744058772790815,
      "grad_norm": 187.51980590820312,
      "learning_rate": 5.8334008329086506e-05,
      "loss": 0.060955810546875,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0601,
      "step": 15950
    },
    {
      "entropy_step_1": 0.2269,
      "entropy_step_2": 0.2334,
      "entropy_step_3": 0.2305,
      "epoch": 0.47470331043098063,
      "grad_norm": 125.92294311523438,
      "learning_rate": 5.828551116269382e-05,
      "loss": 0.06288990378379822,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0651,
      "loss_step_3": 0.0577,
      "step": 15960
    },
    {
      "entropy_step_1": 0.1661,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1806,
      "epoch": 0.47500074358287975,
      "grad_norm": 93.29901123046875,
      "learning_rate": 5.8237005979956335e-05,
      "loss": 0.04753435254096985,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0568,
      "step": 15970
    },
    {
      "entropy_step_1": 0.1579,
      "entropy_step_2": 0.1668,
      "entropy_step_3": 0.1756,
      "epoch": 0.47529817673477887,
      "grad_norm": 140.8844451904297,
      "learning_rate": 5.818849282780349e-05,
      "loss": 0.06339052319526672,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0512,
      "step": 15980
    },
    {
      "entropy_step_1": 0.2025,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.2015,
      "epoch": 0.475595609886678,
      "grad_norm": 76.2470474243164,
      "learning_rate": 5.813997175317241e-05,
      "loss": 0.06016140580177307,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0676,
      "step": 15990
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.2167,
      "entropy_step_3": 0.2195,
      "epoch": 0.4758930430385771,
      "grad_norm": 64.30619812011719,
      "learning_rate": 5.80914428030079e-05,
      "loss": 0.06477230787277222,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.0661,
      "step": 16000
    },
    {
      "epoch": 0.4758930430385771,
      "eval_accuracy": 0.873,
      "eval_f1": 0.8728065387454318,
      "eval_loss": 0.13431505858898163,
      "eval_precision": 0.8752832231295199,
      "eval_recall": 0.873,
      "eval_runtime": 6.7906,
      "eval_samples_per_second": 73.631,
      "eval_steps_per_second": 9.277,
      "step": 16000
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.2059,
      "entropy_step_3": 0.2158,
      "epoch": 0.47619047619047616,
      "grad_norm": 121.04645538330078,
      "learning_rate": 5.804290602426237e-05,
      "loss": 0.060852831602096556,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0628,
      "step": 16010
    },
    {
      "entropy_step_1": 0.2025,
      "entropy_step_2": 0.2158,
      "entropy_step_3": 0.216,
      "epoch": 0.4764879093423753,
      "grad_norm": 88.24828338623047,
      "learning_rate": 5.799436146389583e-05,
      "loss": 0.06766157746315002,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0563,
      "step": 16020
    },
    {
      "entropy_step_1": 0.1987,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.1775,
      "epoch": 0.4767853424942744,
      "grad_norm": 1501.5135498046875,
      "learning_rate": 5.7945809168875786e-05,
      "loss": 0.06890509724617004,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0491,
      "step": 16030
    },
    {
      "entropy_step_1": 0.2374,
      "entropy_step_2": 0.2412,
      "entropy_step_3": 0.2608,
      "epoch": 0.4770827756461735,
      "grad_norm": 73.32955169677734,
      "learning_rate": 5.789724918617727e-05,
      "loss": 0.07427366375923157,
      "loss_step_1": 0.0767,
      "loss_step_2": 0.0764,
      "loss_step_3": 0.0864,
      "step": 16040
    },
    {
      "entropy_step_1": 0.2058,
      "entropy_step_2": 0.212,
      "entropy_step_3": 0.2225,
      "epoch": 0.47738020879807264,
      "grad_norm": 93.24539184570312,
      "learning_rate": 5.784868156278269e-05,
      "loss": 0.05737714767456055,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0627,
      "step": 16050
    },
    {
      "entropy_step_1": 0.1983,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.2055,
      "epoch": 0.47767764194997175,
      "grad_norm": 57.837364196777344,
      "learning_rate": 5.780010634568192e-05,
      "loss": 0.0692915678024292,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0655,
      "step": 16060
    },
    {
      "entropy_step_1": 0.235,
      "entropy_step_2": 0.2312,
      "entropy_step_3": 0.2157,
      "epoch": 0.47797507510187087,
      "grad_norm": 50.927520751953125,
      "learning_rate": 5.7751523581872116e-05,
      "loss": 0.06294230222702027,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0612,
      "step": 16070
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.1988,
      "epoch": 0.47827250825377,
      "grad_norm": 95.93053436279297,
      "learning_rate": 5.770293331835776e-05,
      "loss": 0.0574152946472168,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0569,
      "step": 16080
    },
    {
      "entropy_step_1": 0.202,
      "entropy_step_2": 0.2,
      "entropy_step_3": 0.212,
      "epoch": 0.47856994140566905,
      "grad_norm": 60.6068229675293,
      "learning_rate": 5.765433560215063e-05,
      "loss": 0.06266436576843262,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0605,
      "step": 16090
    },
    {
      "entropy_step_1": 0.1715,
      "entropy_step_2": 0.1707,
      "entropy_step_3": 0.1759,
      "epoch": 0.47886737455756817,
      "grad_norm": 141.27455139160156,
      "learning_rate": 5.7605730480269624e-05,
      "loss": 0.059669983386993405,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0634,
      "step": 16100
    },
    {
      "entropy_step_1": 0.1752,
      "entropy_step_2": 0.1739,
      "entropy_step_3": 0.1885,
      "epoch": 0.4791648077094673,
      "grad_norm": 72.92115020751953,
      "learning_rate": 5.755711799974092e-05,
      "loss": 0.06597381830215454,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0644,
      "step": 16110
    },
    {
      "entropy_step_1": 0.2169,
      "entropy_step_2": 0.2134,
      "entropy_step_3": 0.209,
      "epoch": 0.4794622408613664,
      "grad_norm": 164.23574829101562,
      "learning_rate": 5.7508498207597694e-05,
      "loss": 0.06582216620445251,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0615,
      "step": 16120
    },
    {
      "entropy_step_1": 0.2268,
      "entropy_step_2": 0.2275,
      "entropy_step_3": 0.2211,
      "epoch": 0.4797596740132655,
      "grad_norm": 164.72525024414062,
      "learning_rate": 5.7459871150880294e-05,
      "loss": 0.07368611097335816,
      "loss_step_1": 0.0756,
      "loss_step_2": 0.0753,
      "loss_step_3": 0.0593,
      "step": 16130
    },
    {
      "entropy_step_1": 0.2021,
      "entropy_step_2": 0.2012,
      "entropy_step_3": 0.204,
      "epoch": 0.48005710716516464,
      "grad_norm": 208.92327880859375,
      "learning_rate": 5.7411236876636046e-05,
      "loss": 0.05588871836662292,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0576,
      "step": 16140
    },
    {
      "entropy_step_1": 0.1825,
      "entropy_step_2": 0.1836,
      "entropy_step_3": 0.1914,
      "epoch": 0.48035454031706376,
      "grad_norm": 89.89017486572266,
      "learning_rate": 5.7362595431919276e-05,
      "loss": 0.05889769196510315,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.0569,
      "step": 16150
    },
    {
      "entropy_step_1": 0.1978,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.1587,
      "epoch": 0.4806519734689628,
      "grad_norm": 52.505950927734375,
      "learning_rate": 5.731394686379124e-05,
      "loss": 0.05647816061973572,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0402,
      "step": 16160
    },
    {
      "entropy_step_1": 0.2091,
      "entropy_step_2": 0.2076,
      "entropy_step_3": 0.1986,
      "epoch": 0.48094940662086194,
      "grad_norm": 56.58296585083008,
      "learning_rate": 5.72652912193201e-05,
      "loss": 0.06250560283660889,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0588,
      "step": 16170
    },
    {
      "entropy_step_1": 0.1747,
      "entropy_step_2": 0.1755,
      "entropy_step_3": 0.1753,
      "epoch": 0.48124683977276106,
      "grad_norm": 50.326904296875,
      "learning_rate": 5.7216628545580854e-05,
      "loss": 0.05682839155197143,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0568,
      "step": 16180
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1886,
      "entropy_step_3": 0.1888,
      "epoch": 0.4815442729246602,
      "grad_norm": 241.25344848632812,
      "learning_rate": 5.716795888965527e-05,
      "loss": 0.05219350457191467,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0564,
      "step": 16190
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.1941,
      "entropy_step_3": 0.1876,
      "epoch": 0.4818417060765593,
      "grad_norm": 105.20832061767578,
      "learning_rate": 5.7119282298631946e-05,
      "loss": 0.0587559700012207,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0511,
      "step": 16200
    },
    {
      "entropy_step_1": 0.2203,
      "entropy_step_2": 0.2235,
      "entropy_step_3": 0.2503,
      "epoch": 0.4821391392284584,
      "grad_norm": 190.45291137695312,
      "learning_rate": 5.707059881960611e-05,
      "loss": 0.0665766417980194,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0652,
      "loss_step_3": 0.0692,
      "step": 16210
    },
    {
      "entropy_step_1": 0.193,
      "entropy_step_2": 0.1987,
      "entropy_step_3": 0.1971,
      "epoch": 0.48243657238035753,
      "grad_norm": 38.48942565917969,
      "learning_rate": 5.7021908499679724e-05,
      "loss": 0.05784050226211548,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0651,
      "step": 16220
    },
    {
      "entropy_step_1": 0.1849,
      "entropy_step_2": 0.1923,
      "entropy_step_3": 0.1968,
      "epoch": 0.48273400553225665,
      "grad_norm": 169.78277587890625,
      "learning_rate": 5.697321138596131e-05,
      "loss": 0.05982022285461426,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0668,
      "step": 16230
    },
    {
      "entropy_step_1": 0.2122,
      "entropy_step_2": 0.2212,
      "entropy_step_3": 0.2279,
      "epoch": 0.4830314386841557,
      "grad_norm": 222.0338134765625,
      "learning_rate": 5.6924507525566006e-05,
      "loss": 0.06449351906776428,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0625,
      "loss_step_3": 0.0624,
      "step": 16240
    },
    {
      "entropy_step_1": 0.1785,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.2062,
      "epoch": 0.48332887183605483,
      "grad_norm": 91.4129867553711,
      "learning_rate": 5.6875796965615436e-05,
      "loss": 0.061080753803253174,
      "loss_step_1": 0.0622,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0717,
      "step": 16250
    },
    {
      "entropy_step_1": 0.2214,
      "entropy_step_2": 0.2338,
      "entropy_step_3": 0.2302,
      "epoch": 0.48362630498795395,
      "grad_norm": 189.5098876953125,
      "learning_rate": 5.682707975323774e-05,
      "loss": 0.0632901668548584,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0642,
      "step": 16260
    },
    {
      "entropy_step_1": 0.1912,
      "entropy_step_2": 0.2004,
      "entropy_step_3": 0.1928,
      "epoch": 0.48392373813985307,
      "grad_norm": 94.67030334472656,
      "learning_rate": 5.677835593556751e-05,
      "loss": 0.053927117586135866,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0504,
      "step": 16270
    },
    {
      "entropy_step_1": 0.1699,
      "entropy_step_2": 0.174,
      "entropy_step_3": 0.1749,
      "epoch": 0.4842211712917522,
      "grad_norm": 89.16207122802734,
      "learning_rate": 5.672962555974566e-05,
      "loss": 0.058881980180740354,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0531,
      "step": 16280
    },
    {
      "entropy_step_1": 0.2412,
      "entropy_step_2": 0.238,
      "entropy_step_3": 0.2471,
      "epoch": 0.4845186044436513,
      "grad_norm": 81.19690704345703,
      "learning_rate": 5.668088867291952e-05,
      "loss": 0.0697544276714325,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0707,
      "step": 16290
    },
    {
      "entropy_step_1": 0.1727,
      "entropy_step_2": 0.1694,
      "entropy_step_3": 0.168,
      "epoch": 0.4848160375955504,
      "grad_norm": 77.62663269042969,
      "learning_rate": 5.6632145322242666e-05,
      "loss": 0.05972687005996704,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0495,
      "step": 16300
    },
    {
      "entropy_step_1": 0.1628,
      "entropy_step_2": 0.1606,
      "entropy_step_3": 0.1695,
      "epoch": 0.48511347074744954,
      "grad_norm": 168.35516357421875,
      "learning_rate": 5.658339555487497e-05,
      "loss": 0.06270343065261841,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0719,
      "step": 16310
    },
    {
      "entropy_step_1": 0.2198,
      "entropy_step_2": 0.2171,
      "entropy_step_3": 0.2289,
      "epoch": 0.4854109038993486,
      "grad_norm": 220.9829559326172,
      "learning_rate": 5.6534639417982515e-05,
      "loss": 0.06413742303848266,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0597,
      "step": 16320
    },
    {
      "entropy_step_1": 0.2189,
      "entropy_step_2": 0.2197,
      "entropy_step_3": 0.2274,
      "epoch": 0.4857083370512477,
      "grad_norm": 47.358558654785156,
      "learning_rate": 5.648587695873749e-05,
      "loss": 0.0625583827495575,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0566,
      "step": 16330
    },
    {
      "entropy_step_1": 0.223,
      "entropy_step_2": 0.2262,
      "entropy_step_3": 0.2209,
      "epoch": 0.48600577020314684,
      "grad_norm": 162.04080200195312,
      "learning_rate": 5.643710822431827e-05,
      "loss": 0.06240063309669495,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0621,
      "step": 16340
    },
    {
      "entropy_step_1": 0.1986,
      "entropy_step_2": 0.1999,
      "entropy_step_3": 0.2075,
      "epoch": 0.48630320335504595,
      "grad_norm": 440.0152282714844,
      "learning_rate": 5.6388333261909235e-05,
      "loss": 0.05932015776634216,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0645,
      "step": 16350
    },
    {
      "entropy_step_1": 0.193,
      "entropy_step_2": 0.187,
      "entropy_step_3": 0.1791,
      "epoch": 0.4866006365069451,
      "grad_norm": 127.05791473388672,
      "learning_rate": 5.633955211870085e-05,
      "loss": 0.05856330990791321,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.049,
      "step": 16360
    },
    {
      "entropy_step_1": 0.2122,
      "entropy_step_2": 0.2052,
      "entropy_step_3": 0.2122,
      "epoch": 0.4868980696588442,
      "grad_norm": 360.4495544433594,
      "learning_rate": 5.6290764841889534e-05,
      "loss": 0.06233299374580383,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0608,
      "step": 16370
    },
    {
      "entropy_step_1": 0.204,
      "entropy_step_2": 0.1943,
      "entropy_step_3": 0.1965,
      "epoch": 0.4871955028107433,
      "grad_norm": 61.26724624633789,
      "learning_rate": 5.6241971478677626e-05,
      "loss": 0.05486531257629394,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0507,
      "step": 16380
    },
    {
      "entropy_step_1": 0.1896,
      "entropy_step_2": 0.1775,
      "entropy_step_3": 0.1956,
      "epoch": 0.48749293596264237,
      "grad_norm": 52.55599594116211,
      "learning_rate": 5.619317207627338e-05,
      "loss": 0.05019201040267944,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0498,
      "loss_step_3": 0.0565,
      "step": 16390
    },
    {
      "entropy_step_1": 0.213,
      "entropy_step_2": 0.2028,
      "entropy_step_3": 0.1987,
      "epoch": 0.4877903691145415,
      "grad_norm": 75.17625427246094,
      "learning_rate": 5.614436668189088e-05,
      "loss": 0.05928993225097656,
      "loss_step_1": 0.065,
      "loss_step_2": 0.065,
      "loss_step_3": 0.0604,
      "step": 16400
    },
    {
      "entropy_step_1": 0.203,
      "entropy_step_2": 0.1935,
      "entropy_step_3": 0.2062,
      "epoch": 0.4880878022664406,
      "grad_norm": 116.53997039794922,
      "learning_rate": 5.6095555342750014e-05,
      "loss": 0.06476052403450012,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0657,
      "loss_step_3": 0.0717,
      "step": 16410
    },
    {
      "entropy_step_1": 0.1989,
      "entropy_step_2": 0.1897,
      "entropy_step_3": 0.192,
      "epoch": 0.4883852354183397,
      "grad_norm": 128.47789001464844,
      "learning_rate": 5.604673810607643e-05,
      "loss": 0.06148002743721008,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0603,
      "step": 16420
    },
    {
      "entropy_step_1": 0.2136,
      "entropy_step_2": 0.2024,
      "entropy_step_3": 0.1973,
      "epoch": 0.48868266857023884,
      "grad_norm": 129.4574432373047,
      "learning_rate": 5.599791501910142e-05,
      "loss": 0.0643936574459076,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0633,
      "step": 16430
    },
    {
      "entropy_step_1": 0.2046,
      "entropy_step_2": 0.1956,
      "entropy_step_3": 0.2001,
      "epoch": 0.48898010172213796,
      "grad_norm": 95.34495544433594,
      "learning_rate": 5.594908612906202e-05,
      "loss": 0.05683154463768005,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0556,
      "step": 16440
    },
    {
      "entropy_step_1": 0.2006,
      "entropy_step_2": 0.1939,
      "entropy_step_3": 0.189,
      "epoch": 0.4892775348740371,
      "grad_norm": 142.17662048339844,
      "learning_rate": 5.590025148320086e-05,
      "loss": 0.05966925621032715,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0698,
      "step": 16450
    },
    {
      "entropy_step_1": 0.2131,
      "entropy_step_2": 0.2074,
      "entropy_step_3": 0.2073,
      "epoch": 0.4895749680259362,
      "grad_norm": 149.56362915039062,
      "learning_rate": 5.5851411128766086e-05,
      "loss": 0.07218506336212158,
      "loss_step_1": 0.0715,
      "loss_step_2": 0.0702,
      "loss_step_3": 0.0636,
      "step": 16460
    },
    {
      "entropy_step_1": 0.2008,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.2095,
      "epoch": 0.48987240117783526,
      "grad_norm": 220.00962829589844,
      "learning_rate": 5.580256511301143e-05,
      "loss": 0.052476251125335695,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0541,
      "step": 16470
    },
    {
      "entropy_step_1": 0.1872,
      "entropy_step_2": 0.1834,
      "entropy_step_3": 0.1664,
      "epoch": 0.4901698343297344,
      "grad_norm": 49.501949310302734,
      "learning_rate": 5.575371348319608e-05,
      "loss": 0.06015034914016724,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0593,
      "step": 16480
    },
    {
      "entropy_step_1": 0.2049,
      "entropy_step_2": 0.201,
      "entropy_step_3": 0.1972,
      "epoch": 0.4904672674816335,
      "grad_norm": 332.6080017089844,
      "learning_rate": 5.570485628658463e-05,
      "loss": 0.06115477085113526,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0525,
      "step": 16490
    },
    {
      "entropy_step_1": 0.1921,
      "entropy_step_2": 0.1901,
      "entropy_step_3": 0.1783,
      "epoch": 0.4907647006335326,
      "grad_norm": 291.2117004394531,
      "learning_rate": 5.565599357044711e-05,
      "loss": 0.05709276795387268,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0549,
      "step": 16500
    },
    {
      "entropy_step_1": 0.2095,
      "entropy_step_2": 0.2074,
      "entropy_step_3": 0.216,
      "epoch": 0.49106213378543173,
      "grad_norm": 163.1171112060547,
      "learning_rate": 5.5607125382058846e-05,
      "loss": 0.06114852428436279,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0578,
      "step": 16510
    },
    {
      "entropy_step_1": 0.2153,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.2405,
      "epoch": 0.49135956693733085,
      "grad_norm": 190.5419158935547,
      "learning_rate": 5.555825176870048e-05,
      "loss": 0.06348549127578736,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.0741,
      "step": 16520
    },
    {
      "entropy_step_1": 0.1939,
      "entropy_step_2": 0.1928,
      "entropy_step_3": 0.175,
      "epoch": 0.49165700008922997,
      "grad_norm": 38.26572799682617,
      "learning_rate": 5.5509372777657884e-05,
      "loss": 0.06030902862548828,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0526,
      "step": 16530
    },
    {
      "entropy_step_1": 0.2096,
      "entropy_step_2": 0.2118,
      "entropy_step_3": 0.214,
      "epoch": 0.49195443324112903,
      "grad_norm": 45.632041931152344,
      "learning_rate": 5.546048845622217e-05,
      "loss": 0.05596621632575989,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0622,
      "step": 16540
    },
    {
      "entropy_step_1": 0.1885,
      "entropy_step_2": 0.191,
      "entropy_step_3": 0.1786,
      "epoch": 0.49225186639302815,
      "grad_norm": 64.57176208496094,
      "learning_rate": 5.541159885168957e-05,
      "loss": 0.05742693543434143,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.053,
      "step": 16550
    },
    {
      "entropy_step_1": 0.2045,
      "entropy_step_2": 0.205,
      "entropy_step_3": 0.2029,
      "epoch": 0.49254929954492727,
      "grad_norm": 126.47038269042969,
      "learning_rate": 5.536270401136145e-05,
      "loss": 0.06009189486503601,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0601,
      "step": 16560
    },
    {
      "entropy_step_1": 0.2182,
      "entropy_step_2": 0.2143,
      "entropy_step_3": 0.216,
      "epoch": 0.4928467326968264,
      "grad_norm": 38.32261657714844,
      "learning_rate": 5.531380398254423e-05,
      "loss": 0.06938607096672059,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0754,
      "step": 16570
    },
    {
      "entropy_step_1": 0.2179,
      "entropy_step_2": 0.2119,
      "entropy_step_3": 0.2153,
      "epoch": 0.4931441658487255,
      "grad_norm": 67.19957733154297,
      "learning_rate": 5.526489881254935e-05,
      "loss": 0.06266205310821533,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0582,
      "step": 16580
    },
    {
      "entropy_step_1": 0.2166,
      "entropy_step_2": 0.2084,
      "entropy_step_3": 0.2248,
      "epoch": 0.4934415990006246,
      "grad_norm": 51.745277404785156,
      "learning_rate": 5.5215988548693234e-05,
      "loss": 0.06660573482513428,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0607,
      "step": 16590
    },
    {
      "entropy_step_1": 0.2109,
      "entropy_step_2": 0.1995,
      "entropy_step_3": 0.1972,
      "epoch": 0.49373903215252374,
      "grad_norm": 100.64446258544922,
      "learning_rate": 5.516707323829722e-05,
      "loss": 0.06287384629249573,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.064,
      "step": 16600
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1771,
      "entropy_step_3": 0.1816,
      "epoch": 0.49403646530442286,
      "grad_norm": 72.1899185180664,
      "learning_rate": 5.511815292868754e-05,
      "loss": 0.06603637337684631,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0515,
      "step": 16610
    },
    {
      "entropy_step_1": 0.2288,
      "entropy_step_2": 0.2169,
      "entropy_step_3": 0.2142,
      "epoch": 0.4943338984563219,
      "grad_norm": 73.42185974121094,
      "learning_rate": 5.5069227667195254e-05,
      "loss": 0.06504569053649903,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0514,
      "loss_step_3": 0.0463,
      "step": 16620
    },
    {
      "entropy_step_1": 0.2282,
      "entropy_step_2": 0.2137,
      "entropy_step_3": 0.2008,
      "epoch": 0.49463133160822104,
      "grad_norm": 74.30448913574219,
      "learning_rate": 5.502029750115623e-05,
      "loss": 0.058086717128753663,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0513,
      "step": 16630
    },
    {
      "entropy_step_1": 0.2162,
      "entropy_step_2": 0.207,
      "entropy_step_3": 0.2053,
      "epoch": 0.49492876476012015,
      "grad_norm": 448.57025146484375,
      "learning_rate": 5.497136247791106e-05,
      "loss": 0.060901814699172975,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.071,
      "step": 16640
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.1996,
      "entropy_step_3": 0.2099,
      "epoch": 0.4952261979120193,
      "grad_norm": 1539.8446044921875,
      "learning_rate": 5.492242264480505e-05,
      "loss": 0.05914214849472046,
      "loss_step_1": 0.0619,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0691,
      "step": 16650
    },
    {
      "entropy_step_1": 0.2175,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.2032,
      "epoch": 0.4955236310639184,
      "grad_norm": 293.3633728027344,
      "learning_rate": 5.4873478049188154e-05,
      "loss": 0.0653329312801361,
      "loss_step_1": 0.0704,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0669,
      "step": 16660
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1818,
      "entropy_step_3": 0.1846,
      "epoch": 0.4958210642158175,
      "grad_norm": 1070.7099609375,
      "learning_rate": 5.482452873841492e-05,
      "loss": 0.057849884033203125,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0544,
      "step": 16670
    },
    {
      "entropy_step_1": 0.2151,
      "entropy_step_2": 0.2165,
      "entropy_step_3": 0.2101,
      "epoch": 0.4961184973677166,
      "grad_norm": 103.82299041748047,
      "learning_rate": 5.477557475984449e-05,
      "loss": 0.06582703590393066,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.059,
      "step": 16680
    },
    {
      "entropy_step_1": 0.2249,
      "entropy_step_2": 0.2253,
      "entropy_step_3": 0.2189,
      "epoch": 0.4964159305196157,
      "grad_norm": 61.01148986816406,
      "learning_rate": 5.4726616160840494e-05,
      "loss": 0.06860421895980835,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0615,
      "step": 16690
    },
    {
      "entropy_step_1": 0.2005,
      "entropy_step_2": 0.2066,
      "entropy_step_3": 0.2137,
      "epoch": 0.4967133636715148,
      "grad_norm": 91.19450378417969,
      "learning_rate": 5.467765298877106e-05,
      "loss": 0.060850077867507936,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0581,
      "step": 16700
    },
    {
      "entropy_step_1": 0.1953,
      "entropy_step_2": 0.2012,
      "entropy_step_3": 0.2037,
      "epoch": 0.4970107968234139,
      "grad_norm": 104.37213134765625,
      "learning_rate": 5.4628685291008706e-05,
      "loss": 0.055830967426300046,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0602,
      "step": 16710
    },
    {
      "entropy_step_1": 0.2039,
      "entropy_step_2": 0.2081,
      "entropy_step_3": 0.2038,
      "epoch": 0.49730822997531304,
      "grad_norm": 143.06492614746094,
      "learning_rate": 5.457971311493034e-05,
      "loss": 0.060984230041503905,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.056,
      "step": 16720
    },
    {
      "entropy_step_1": 0.1985,
      "entropy_step_2": 0.2031,
      "entropy_step_3": 0.1989,
      "epoch": 0.49760566312721216,
      "grad_norm": 121.39657592773438,
      "learning_rate": 5.453073650791724e-05,
      "loss": 0.06090055108070373,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0581,
      "step": 16730
    },
    {
      "entropy_step_1": 0.2066,
      "entropy_step_2": 0.2114,
      "entropy_step_3": 0.204,
      "epoch": 0.4979030962791113,
      "grad_norm": 492.5432434082031,
      "learning_rate": 5.448175551735489e-05,
      "loss": 0.06307801604270935,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.0663,
      "loss_step_3": 0.0556,
      "step": 16740
    },
    {
      "entropy_step_1": 0.1948,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.2037,
      "epoch": 0.4982005294310104,
      "grad_norm": 125.80400848388672,
      "learning_rate": 5.443277019063312e-05,
      "loss": 0.057364809513092044,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0581,
      "step": 16750
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1922,
      "entropy_step_3": 0.1839,
      "epoch": 0.4984979625829095,
      "grad_norm": 59.43564987182617,
      "learning_rate": 5.438378057514586e-05,
      "loss": 0.06607611775398255,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0668,
      "step": 16760
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.1925,
      "epoch": 0.4987953957348086,
      "grad_norm": 215.8719024658203,
      "learning_rate": 5.433478671829124e-05,
      "loss": 0.06124025583267212,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0746,
      "step": 16770
    },
    {
      "entropy_step_1": 0.2024,
      "entropy_step_2": 0.2102,
      "entropy_step_3": 0.2065,
      "epoch": 0.4990928288867077,
      "grad_norm": 50.86967468261719,
      "learning_rate": 5.4285788667471505e-05,
      "loss": 0.055994123220443726,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0573,
      "step": 16780
    },
    {
      "entropy_step_1": 0.1762,
      "entropy_step_2": 0.1826,
      "entropy_step_3": 0.1687,
      "epoch": 0.4993902620386068,
      "grad_norm": 55.10786819458008,
      "learning_rate": 5.4236786470092915e-05,
      "loss": 0.06097780466079712,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0494,
      "step": 16790
    },
    {
      "entropy_step_1": 0.2319,
      "entropy_step_2": 0.2343,
      "entropy_step_3": 0.2463,
      "epoch": 0.49968769519050593,
      "grad_norm": 178.34368896484375,
      "learning_rate": 5.418778017356576e-05,
      "loss": 0.06839253902435302,
      "loss_step_1": 0.077,
      "loss_step_2": 0.0763,
      "loss_step_3": 0.0861,
      "step": 16800
    },
    {
      "entropy_step_1": 0.2163,
      "entropy_step_2": 0.2219,
      "entropy_step_3": 0.213,
      "epoch": 0.49998512834240505,
      "grad_norm": 51.213706970214844,
      "learning_rate": 5.41387698253043e-05,
      "loss": 0.06144183874130249,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0573,
      "step": 16810
    },
    {
      "entropy_step_1": 0.1736,
      "entropy_step_2": 0.1835,
      "entropy_step_3": 0.1752,
      "epoch": 0.5002825614943042,
      "grad_norm": 178.15931701660156,
      "learning_rate": 5.4089755472726725e-05,
      "loss": 0.06222821474075317,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0584,
      "step": 16820
    },
    {
      "entropy_step_1": 0.2102,
      "entropy_step_2": 0.2268,
      "entropy_step_3": 0.2247,
      "epoch": 0.5005799946462033,
      "grad_norm": 110.48286437988281,
      "learning_rate": 5.404073716325507e-05,
      "loss": 0.06755802035331726,
      "loss_step_1": 0.0723,
      "loss_step_2": 0.0708,
      "loss_step_3": 0.0712,
      "step": 16830
    },
    {
      "entropy_step_1": 0.2204,
      "entropy_step_2": 0.2377,
      "entropy_step_3": 0.2298,
      "epoch": 0.5008774277981024,
      "grad_norm": 230.42733764648438,
      "learning_rate": 5.399171494431524e-05,
      "loss": 0.05822170376777649,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0507,
      "step": 16840
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.2157,
      "entropy_step_3": 0.2457,
      "epoch": 0.5011748609500015,
      "grad_norm": 301.1647033691406,
      "learning_rate": 5.3942688863336876e-05,
      "loss": 0.06304863691329957,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0774,
      "step": 16850
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.19,
      "entropy_step_3": 0.1908,
      "epoch": 0.5014722941019006,
      "grad_norm": 56.89492416381836,
      "learning_rate": 5.389365896775339e-05,
      "loss": 0.0560316264629364,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0467,
      "step": 16860
    },
    {
      "entropy_step_1": 0.2071,
      "entropy_step_2": 0.2161,
      "entropy_step_3": 0.2033,
      "epoch": 0.5017697272537998,
      "grad_norm": 141.61094665527344,
      "learning_rate": 5.384462530500187e-05,
      "loss": 0.06653942465782166,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0505,
      "step": 16870
    },
    {
      "entropy_step_1": 0.1767,
      "entropy_step_2": 0.1875,
      "entropy_step_3": 0.1866,
      "epoch": 0.5020671604056988,
      "grad_norm": 59.391990661621094,
      "learning_rate": 5.379558792252306e-05,
      "loss": 0.06145574450492859,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0642,
      "step": 16880
    },
    {
      "entropy_step_1": 0.1895,
      "entropy_step_2": 0.1945,
      "entropy_step_3": 0.2031,
      "epoch": 0.5023645935575979,
      "grad_norm": 479.56427001953125,
      "learning_rate": 5.374654686776129e-05,
      "loss": 0.055600911378860474,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0607,
      "step": 16890
    },
    {
      "entropy_step_1": 0.2413,
      "entropy_step_2": 0.2446,
      "entropy_step_3": 0.2406,
      "epoch": 0.502662026709497,
      "grad_norm": 283.7814636230469,
      "learning_rate": 5.3697502188164437e-05,
      "loss": 0.07119568586349487,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0726,
      "step": 16900
    },
    {
      "entropy_step_1": 0.1985,
      "entropy_step_2": 0.1984,
      "entropy_step_3": 0.197,
      "epoch": 0.5029594598613961,
      "grad_norm": 158.61431884765625,
      "learning_rate": 5.364845393118392e-05,
      "loss": 0.054815882444381715,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0514,
      "step": 16910
    },
    {
      "entropy_step_1": 0.2454,
      "entropy_step_2": 0.2388,
      "entropy_step_3": 0.2317,
      "epoch": 0.5032568930132952,
      "grad_norm": 116.61573791503906,
      "learning_rate": 5.359940214427456e-05,
      "loss": 0.0651834785938263,
      "loss_step_1": 0.0688,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0674,
      "step": 16920
    },
    {
      "entropy_step_1": 0.2115,
      "entropy_step_2": 0.2081,
      "entropy_step_3": 0.1964,
      "epoch": 0.5035543261651944,
      "grad_norm": 189.38211059570312,
      "learning_rate": 5.355034687489465e-05,
      "loss": 0.06675602793693543,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0512,
      "step": 16930
    },
    {
      "entropy_step_1": 0.1956,
      "entropy_step_2": 0.191,
      "entropy_step_3": 0.1844,
      "epoch": 0.5038517593170935,
      "grad_norm": 170.8079833984375,
      "learning_rate": 5.3501288170505844e-05,
      "loss": 0.05485016107559204,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0562,
      "step": 16940
    },
    {
      "entropy_step_1": 0.2218,
      "entropy_step_2": 0.212,
      "entropy_step_3": 0.1989,
      "epoch": 0.5041491924689926,
      "grad_norm": 108.20915222167969,
      "learning_rate": 5.345222607857309e-05,
      "loss": 0.0742453694343567,
      "loss_step_1": 0.0813,
      "loss_step_2": 0.0826,
      "loss_step_3": 0.0698,
      "step": 16950
    },
    {
      "entropy_step_1": 0.208,
      "entropy_step_2": 0.2,
      "entropy_step_3": 0.1947,
      "epoch": 0.5044466256208917,
      "grad_norm": 53.89240646362305,
      "learning_rate": 5.340316064656463e-05,
      "loss": 0.06287675499916076,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0635,
      "step": 16960
    },
    {
      "entropy_step_1": 0.2102,
      "entropy_step_2": 0.2042,
      "entropy_step_3": 0.221,
      "epoch": 0.5047440587727908,
      "grad_norm": 155.30885314941406,
      "learning_rate": 5.335409192195194e-05,
      "loss": 0.06084800362586975,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0599,
      "step": 16970
    },
    {
      "entropy_step_1": 0.1944,
      "entropy_step_2": 0.1933,
      "entropy_step_3": 0.1946,
      "epoch": 0.50504149192469,
      "grad_norm": 229.28363037109375,
      "learning_rate": 5.330501995220967e-05,
      "loss": 0.058504623174667356,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0469,
      "step": 16980
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2031,
      "epoch": 0.5053389250765891,
      "grad_norm": 168.28659057617188,
      "learning_rate": 5.3255944784815626e-05,
      "loss": 0.07272011637687684,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0559,
      "step": 16990
    },
    {
      "entropy_step_1": 0.2218,
      "entropy_step_2": 0.2196,
      "entropy_step_3": 0.2315,
      "epoch": 0.5056363582284882,
      "grad_norm": 164.99539184570312,
      "learning_rate": 5.32068664672507e-05,
      "loss": 0.06205034255981445,
      "loss_step_1": 0.068,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0724,
      "step": 17000
    },
    {
      "epoch": 0.5056363582284882,
      "eval_accuracy": 0.849,
      "eval_f1": 0.8476248139458614,
      "eval_loss": 0.13509759306907654,
      "eval_precision": 0.8620707542276169,
      "eval_recall": 0.849,
      "eval_runtime": 6.6961,
      "eval_samples_per_second": 74.671,
      "eval_steps_per_second": 9.408,
      "step": 17000
    },
    {
      "entropy_step_1": 0.1831,
      "entropy_step_2": 0.1831,
      "entropy_step_3": 0.1966,
      "epoch": 0.5059337913803873,
      "grad_norm": 119.83193969726562,
      "learning_rate": 5.315778504699883e-05,
      "loss": 0.058578068017959596,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0717,
      "step": 17010
    },
    {
      "entropy_step_1": 0.2171,
      "entropy_step_2": 0.216,
      "entropy_step_3": 0.2058,
      "epoch": 0.5062312245322864,
      "grad_norm": 346.76043701171875,
      "learning_rate": 5.3108700571546955e-05,
      "loss": 0.06399516463279724,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0628,
      "step": 17020
    },
    {
      "entropy_step_1": 0.2015,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.199,
      "epoch": 0.5065286576841854,
      "grad_norm": 69.52174377441406,
      "learning_rate": 5.3059613088384986e-05,
      "loss": 0.048333349823951724,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0504,
      "step": 17030
    },
    {
      "entropy_step_1": 0.1735,
      "entropy_step_2": 0.1796,
      "entropy_step_3": 0.186,
      "epoch": 0.5068260908360845,
      "grad_norm": 405.0927734375,
      "learning_rate": 5.301052264500571e-05,
      "loss": 0.060031354427337646,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0598,
      "step": 17040
    },
    {
      "entropy_step_1": 0.192,
      "entropy_step_2": 0.1957,
      "entropy_step_3": 0.2035,
      "epoch": 0.5071235239879837,
      "grad_norm": 182.29661560058594,
      "learning_rate": 5.2961429288904815e-05,
      "loss": 0.056834614276885985,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0569,
      "step": 17050
    },
    {
      "entropy_step_1": 0.1911,
      "entropy_step_2": 0.1932,
      "entropy_step_3": 0.166,
      "epoch": 0.5074209571398828,
      "grad_norm": 66.84671783447266,
      "learning_rate": 5.291233306758079e-05,
      "loss": 0.06158579587936401,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0659,
      "step": 17060
    },
    {
      "entropy_step_1": 0.2122,
      "entropy_step_2": 0.2122,
      "entropy_step_3": 0.2093,
      "epoch": 0.5077183902917819,
      "grad_norm": 145.2693634033203,
      "learning_rate": 5.2863234028534906e-05,
      "loss": 0.05768169760704041,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.057,
      "loss_step_3": 0.057,
      "step": 17070
    },
    {
      "entropy_step_1": 0.205,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.2078,
      "epoch": 0.508015823443681,
      "grad_norm": 215.14801025390625,
      "learning_rate": 5.2814132219271115e-05,
      "loss": 0.07049989104270935,
      "loss_step_1": 0.0728,
      "loss_step_2": 0.071,
      "loss_step_3": 0.0718,
      "step": 17080
    },
    {
      "entropy_step_1": 0.2205,
      "entropy_step_2": 0.2293,
      "entropy_step_3": 0.2259,
      "epoch": 0.5083132565955801,
      "grad_norm": 121.04875946044922,
      "learning_rate": 5.276502768729612e-05,
      "loss": 0.06322267651557922,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0662,
      "loss_step_3": 0.0722,
      "step": 17090
    },
    {
      "entropy_step_1": 0.2277,
      "entropy_step_2": 0.2408,
      "entropy_step_3": 0.2382,
      "epoch": 0.5086106897474792,
      "grad_norm": 108.33673858642578,
      "learning_rate": 5.2715920480119216e-05,
      "loss": 0.06569361686706543,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.0675,
      "loss_step_3": 0.0705,
      "step": 17100
    },
    {
      "entropy_step_1": 0.1984,
      "entropy_step_2": 0.2114,
      "entropy_step_3": 0.1914,
      "epoch": 0.5089081228993784,
      "grad_norm": 82.5300064086914,
      "learning_rate": 5.2666810645252294e-05,
      "loss": 0.05975935459136963,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0482,
      "step": 17110
    },
    {
      "entropy_step_1": 0.2199,
      "entropy_step_2": 0.231,
      "entropy_step_3": 0.2233,
      "epoch": 0.5092055560512775,
      "grad_norm": 161.33074951171875,
      "learning_rate": 5.261769823020979e-05,
      "loss": 0.06112962365150452,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0621,
      "step": 17120
    },
    {
      "entropy_step_1": 0.1819,
      "entropy_step_2": 0.1928,
      "entropy_step_3": 0.1938,
      "epoch": 0.5095029892031766,
      "grad_norm": 215.83241271972656,
      "learning_rate": 5.256858328250861e-05,
      "loss": 0.057076412439346316,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0576,
      "step": 17130
    },
    {
      "entropy_step_1": 0.2063,
      "entropy_step_2": 0.2141,
      "entropy_step_3": 0.2361,
      "epoch": 0.5098004223550757,
      "grad_norm": 74.51325225830078,
      "learning_rate": 5.2519465849668184e-05,
      "loss": 0.06619070768356324,
      "loss_step_1": 0.069,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0761,
      "step": 17140
    },
    {
      "entropy_step_1": 0.22,
      "entropy_step_2": 0.2286,
      "entropy_step_3": 0.2134,
      "epoch": 0.5100978555069748,
      "grad_norm": 120.63079071044922,
      "learning_rate": 5.247034597921023e-05,
      "loss": 0.059654563665390015,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0589,
      "step": 17150
    },
    {
      "entropy_step_1": 0.1993,
      "entropy_step_2": 0.2077,
      "entropy_step_3": 0.2161,
      "epoch": 0.510395288658874,
      "grad_norm": 57.4451904296875,
      "learning_rate": 5.242122371865895e-05,
      "loss": 0.06490705013275147,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0708,
      "step": 17160
    },
    {
      "entropy_step_1": 0.219,
      "entropy_step_2": 0.2265,
      "entropy_step_3": 0.2128,
      "epoch": 0.5106927218107731,
      "grad_norm": 65.5269775390625,
      "learning_rate": 5.2372099115540786e-05,
      "loss": 0.05931301116943359,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0625,
      "step": 17170
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.2129,
      "entropy_step_3": 0.2192,
      "epoch": 0.5109901549626722,
      "grad_norm": 201.71896362304688,
      "learning_rate": 5.232297221738444e-05,
      "loss": 0.06355693340301513,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0629,
      "step": 17180
    },
    {
      "entropy_step_1": 0.1982,
      "entropy_step_2": 0.2056,
      "entropy_step_3": 0.2026,
      "epoch": 0.5112875881145712,
      "grad_norm": 97.78790283203125,
      "learning_rate": 5.227384307172086e-05,
      "loss": 0.0690767526626587,
      "loss_step_1": 0.0685,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.0611,
      "step": 17190
    },
    {
      "entropy_step_1": 0.2375,
      "entropy_step_2": 0.2462,
      "entropy_step_3": 0.2332,
      "epoch": 0.5115850212664703,
      "grad_norm": 88.45771026611328,
      "learning_rate": 5.2224711726083166e-05,
      "loss": 0.07225090265274048,
      "loss_step_1": 0.0705,
      "loss_step_2": 0.0706,
      "loss_step_3": 0.0712,
      "step": 17200
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2182,
      "entropy_step_3": 0.2248,
      "epoch": 0.5118824544183694,
      "grad_norm": 35.83905029296875,
      "learning_rate": 5.2175578228006615e-05,
      "loss": 0.0640450119972229,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.066,
      "loss_step_3": 0.0789,
      "step": 17210
    },
    {
      "entropy_step_1": 0.2104,
      "entropy_step_2": 0.2264,
      "entropy_step_3": 0.2036,
      "epoch": 0.5121798875702686,
      "grad_norm": 80.35198974609375,
      "learning_rate": 5.2126442625028517e-05,
      "loss": 0.06861768364906311,
      "loss_step_1": 0.0694,
      "loss_step_2": 0.069,
      "loss_step_3": 0.0622,
      "step": 17220
    },
    {
      "entropy_step_1": 0.1879,
      "entropy_step_2": 0.2017,
      "entropy_step_3": 0.2029,
      "epoch": 0.5124773207221677,
      "grad_norm": 218.68492126464844,
      "learning_rate": 5.2077304964688244e-05,
      "loss": 0.05905605554580688,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0569,
      "step": 17230
    },
    {
      "entropy_step_1": 0.1765,
      "entropy_step_2": 0.1898,
      "entropy_step_3": 0.1808,
      "epoch": 0.5127747538740668,
      "grad_norm": 91.24811553955078,
      "learning_rate": 5.2028165294527154e-05,
      "loss": 0.0573864996433258,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0522,
      "step": 17240
    },
    {
      "entropy_step_1": 0.2229,
      "entropy_step_2": 0.2354,
      "entropy_step_3": 0.2262,
      "epoch": 0.5130721870259659,
      "grad_norm": 135.62039184570312,
      "learning_rate": 5.197902366208854e-05,
      "loss": 0.06319940686225892,
      "loss_step_1": 0.0681,
      "loss_step_2": 0.0678,
      "loss_step_3": 0.0636,
      "step": 17250
    },
    {
      "entropy_step_1": 0.2104,
      "entropy_step_2": 0.2233,
      "entropy_step_3": 0.2189,
      "epoch": 0.513369620177865,
      "grad_norm": 210.33787536621094,
      "learning_rate": 5.1929880114917615e-05,
      "loss": 0.06193084120750427,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0635,
      "step": 17260
    },
    {
      "entropy_step_1": 0.1645,
      "entropy_step_2": 0.177,
      "entropy_step_3": 0.1809,
      "epoch": 0.5136670533297641,
      "grad_norm": 68.73419952392578,
      "learning_rate": 5.188073470056142e-05,
      "loss": 0.06353251934051514,
      "loss_step_1": 0.065,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0664,
      "step": 17270
    },
    {
      "entropy_step_1": 0.2252,
      "entropy_step_2": 0.2359,
      "entropy_step_3": 0.2404,
      "epoch": 0.5139644864816633,
      "grad_norm": 86.25569152832031,
      "learning_rate": 5.18315874665688e-05,
      "loss": 0.06499286890029907,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0527,
      "step": 17280
    },
    {
      "entropy_step_1": 0.1913,
      "entropy_step_2": 0.201,
      "entropy_step_3": 0.233,
      "epoch": 0.5142619196335624,
      "grad_norm": 47.10398864746094,
      "learning_rate": 5.178243846049039e-05,
      "loss": 0.05948798060417175,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0704,
      "step": 17290
    },
    {
      "entropy_step_1": 0.2012,
      "entropy_step_2": 0.2103,
      "entropy_step_3": 0.2172,
      "epoch": 0.5145593527854615,
      "grad_norm": 71.74885559082031,
      "learning_rate": 5.173328772987852e-05,
      "loss": 0.06144368648529053,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0657,
      "step": 17300
    },
    {
      "entropy_step_1": 0.1981,
      "entropy_step_2": 0.2066,
      "entropy_step_3": 0.2125,
      "epoch": 0.5148567859373606,
      "grad_norm": 92.94609069824219,
      "learning_rate": 5.168413532228718e-05,
      "loss": 0.05913251042366028,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0745,
      "step": 17310
    },
    {
      "entropy_step_1": 0.1668,
      "entropy_step_2": 0.1744,
      "entropy_step_3": 0.1648,
      "epoch": 0.5151542190892597,
      "grad_norm": 53.621734619140625,
      "learning_rate": 5.1634981285272e-05,
      "loss": 0.05610477924346924,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0494,
      "step": 17320
    },
    {
      "entropy_step_1": 0.1915,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.1921,
      "epoch": 0.5154516522411589,
      "grad_norm": 60.77692413330078,
      "learning_rate": 5.158582566639017e-05,
      "loss": 0.06714847683906555,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.0725,
      "loss_step_3": 0.0897,
      "step": 17330
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2134,
      "entropy_step_3": 0.2226,
      "epoch": 0.5157490853930579,
      "grad_norm": 60.81392288208008,
      "learning_rate": 5.153666851320046e-05,
      "loss": 0.05782371759414673,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0674,
      "step": 17340
    },
    {
      "entropy_step_1": 0.1725,
      "entropy_step_2": 0.1809,
      "entropy_step_3": 0.1858,
      "epoch": 0.516046518544957,
      "grad_norm": 145.66954040527344,
      "learning_rate": 5.1487509873263026e-05,
      "loss": 0.05974566340446472,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0552,
      "step": 17350
    },
    {
      "entropy_step_1": 0.2049,
      "entropy_step_2": 0.2105,
      "entropy_step_3": 0.1924,
      "epoch": 0.5163439516968561,
      "grad_norm": 111.03235626220703,
      "learning_rate": 5.143834979413956e-05,
      "loss": 0.05707759857177734,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0552,
      "step": 17360
    },
    {
      "entropy_step_1": 0.2,
      "entropy_step_2": 0.2041,
      "entropy_step_3": 0.2115,
      "epoch": 0.5166413848487552,
      "grad_norm": 238.2884521484375,
      "learning_rate": 5.13891883233931e-05,
      "loss": 0.05679718255996704,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0608,
      "step": 17370
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.2021,
      "entropy_step_3": 0.218,
      "epoch": 0.5169388180006543,
      "grad_norm": 134.72975158691406,
      "learning_rate": 5.134002550858802e-05,
      "loss": 0.06501379609107971,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0756,
      "step": 17380
    },
    {
      "entropy_step_1": 0.205,
      "entropy_step_2": 0.2119,
      "entropy_step_3": 0.2238,
      "epoch": 0.5172362511525535,
      "grad_norm": 143.84249877929688,
      "learning_rate": 5.1290861397290044e-05,
      "loss": 0.05846807360649109,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0626,
      "step": 17390
    },
    {
      "entropy_step_1": 0.1811,
      "entropy_step_2": 0.1895,
      "entropy_step_3": 0.1851,
      "epoch": 0.5175336843044526,
      "grad_norm": 238.79293823242188,
      "learning_rate": 5.124169603706609e-05,
      "loss": 0.054303741455078124,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0459,
      "step": 17400
    },
    {
      "entropy_step_1": 0.1842,
      "entropy_step_2": 0.1924,
      "entropy_step_3": 0.2114,
      "epoch": 0.5178311174563517,
      "grad_norm": 109.94941711425781,
      "learning_rate": 5.119252947548432e-05,
      "loss": 0.05870357155799866,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.069,
      "step": 17410
    },
    {
      "entropy_step_1": 0.1982,
      "entropy_step_2": 0.2085,
      "entropy_step_3": 0.1986,
      "epoch": 0.5181285506082508,
      "grad_norm": 64.07581329345703,
      "learning_rate": 5.1143361760114074e-05,
      "loss": 0.06556310057640076,
      "loss_step_1": 0.0724,
      "loss_step_2": 0.0707,
      "loss_step_3": 0.0626,
      "step": 17420
    },
    {
      "entropy_step_1": 0.1804,
      "entropy_step_2": 0.192,
      "entropy_step_3": 0.1919,
      "epoch": 0.5184259837601499,
      "grad_norm": 75.25715637207031,
      "learning_rate": 5.109419293852574e-05,
      "loss": 0.05925664901733398,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0577,
      "step": 17430
    },
    {
      "entropy_step_1": 0.1915,
      "entropy_step_2": 0.1989,
      "entropy_step_3": 0.192,
      "epoch": 0.518723416912049,
      "grad_norm": 82.6480484008789,
      "learning_rate": 5.1045023058290876e-05,
      "loss": 0.05459856390953064,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.049,
      "step": 17440
    },
    {
      "entropy_step_1": 0.1805,
      "entropy_step_2": 0.1837,
      "entropy_step_3": 0.1857,
      "epoch": 0.5190208500639482,
      "grad_norm": 100.27513885498047,
      "learning_rate": 5.0995852166981974e-05,
      "loss": 0.06454513669013977,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0641,
      "step": 17450
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.1918,
      "entropy_step_3": 0.1812,
      "epoch": 0.5193182832158473,
      "grad_norm": 84.63385772705078,
      "learning_rate": 5.0946680312172546e-05,
      "loss": 0.058523702621459964,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0499,
      "step": 17460
    },
    {
      "entropy_step_1": 0.1893,
      "entropy_step_2": 0.1884,
      "entropy_step_3": 0.1969,
      "epoch": 0.5196157163677464,
      "grad_norm": 84.65625,
      "learning_rate": 5.089750754143704e-05,
      "loss": 0.053081804513931276,
      "loss_step_1": 0.0508,
      "loss_step_2": 0.0498,
      "loss_step_3": 0.0523,
      "step": 17470
    },
    {
      "entropy_step_1": 0.2049,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.2078,
      "epoch": 0.5199131495196455,
      "grad_norm": 56.23965072631836,
      "learning_rate": 5.084833390235077e-05,
      "loss": 0.05756900310516357,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0593,
      "step": 17480
    },
    {
      "entropy_step_1": 0.1958,
      "entropy_step_2": 0.1896,
      "entropy_step_3": 0.1866,
      "epoch": 0.5202105826715445,
      "grad_norm": 135.04827880859375,
      "learning_rate": 5.0799159442489904e-05,
      "loss": 0.06396843791007996,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0576,
      "step": 17490
    },
    {
      "entropy_step_1": 0.2008,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.1889,
      "epoch": 0.5205080158234436,
      "grad_norm": 47.038673400878906,
      "learning_rate": 5.07499842094314e-05,
      "loss": 0.06319708824157715,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0584,
      "step": 17500
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.2043,
      "epoch": 0.5208054489753428,
      "grad_norm": 77.67926025390625,
      "learning_rate": 5.070080825075299e-05,
      "loss": 0.06429147720336914,
      "loss_step_1": 0.0667,
      "loss_step_2": 0.0658,
      "loss_step_3": 0.0591,
      "step": 17510
    },
    {
      "entropy_step_1": 0.2284,
      "entropy_step_2": 0.2208,
      "entropy_step_3": 0.2142,
      "epoch": 0.5211028821272419,
      "grad_norm": 70.49113464355469,
      "learning_rate": 5.0651631614033055e-05,
      "loss": 0.0591708242893219,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0514,
      "step": 17520
    },
    {
      "entropy_step_1": 0.2393,
      "entropy_step_2": 0.232,
      "entropy_step_3": 0.2276,
      "epoch": 0.521400315279141,
      "grad_norm": 69.46260833740234,
      "learning_rate": 5.060245434685066e-05,
      "loss": 0.06489192843437194,
      "loss_step_1": 0.0727,
      "loss_step_2": 0.0734,
      "loss_step_3": 0.0708,
      "step": 17530
    },
    {
      "entropy_step_1": 0.2038,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.1932,
      "epoch": 0.5216977484310401,
      "grad_norm": 53.41841506958008,
      "learning_rate": 5.0553276496785485e-05,
      "loss": 0.05710709095001221,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0572,
      "step": 17540
    },
    {
      "entropy_step_1": 0.176,
      "entropy_step_2": 0.1764,
      "entropy_step_3": 0.1714,
      "epoch": 0.5219951815829392,
      "grad_norm": 137.4767303466797,
      "learning_rate": 5.0504098111417776e-05,
      "loss": 0.05359223484992981,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0509,
      "step": 17550
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.2027,
      "entropy_step_3": 0.2087,
      "epoch": 0.5222926147348383,
      "grad_norm": 92.85308074951172,
      "learning_rate": 5.045491923832828e-05,
      "loss": 0.06314199566841125,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0662,
      "step": 17560
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.2096,
      "entropy_step_3": 0.2101,
      "epoch": 0.5225900478867375,
      "grad_norm": 66.9843521118164,
      "learning_rate": 5.0405739925098225e-05,
      "loss": 0.06048135161399841,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0585,
      "step": 17570
    },
    {
      "entropy_step_1": 0.1855,
      "entropy_step_2": 0.1891,
      "entropy_step_3": 0.2201,
      "epoch": 0.5228874810386366,
      "grad_norm": 181.85093688964844,
      "learning_rate": 5.035656021930929e-05,
      "loss": 0.05590811371803284,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0646,
      "step": 17580
    },
    {
      "entropy_step_1": 0.2027,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.1966,
      "epoch": 0.5231849141905357,
      "grad_norm": 176.5374755859375,
      "learning_rate": 5.0307380168543474e-05,
      "loss": 0.06335694789886474,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0693,
      "step": 17590
    },
    {
      "entropy_step_1": 0.1955,
      "entropy_step_2": 0.1933,
      "entropy_step_3": 0.2184,
      "epoch": 0.5234823473424348,
      "grad_norm": 196.97625732421875,
      "learning_rate": 5.0258199820383166e-05,
      "loss": 0.06186491250991821,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0658,
      "step": 17600
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.1888,
      "entropy_step_3": 0.1605,
      "epoch": 0.5237797804943339,
      "grad_norm": 59.8787727355957,
      "learning_rate": 5.020901922241102e-05,
      "loss": 0.054576045274734496,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0461,
      "step": 17610
    },
    {
      "entropy_step_1": 0.1861,
      "entropy_step_2": 0.1921,
      "entropy_step_3": 0.2018,
      "epoch": 0.524077213646233,
      "grad_norm": 215.4490203857422,
      "learning_rate": 5.015983842220994e-05,
      "loss": 0.057634323835372925,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0646,
      "step": 17620
    },
    {
      "entropy_step_1": 0.21,
      "entropy_step_2": 0.2142,
      "entropy_step_3": 0.2067,
      "epoch": 0.5243746467981322,
      "grad_norm": 126.3728256225586,
      "learning_rate": 5.0110657467362995e-05,
      "loss": 0.061766022443771364,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0554,
      "step": 17630
    },
    {
      "entropy_step_1": 0.2093,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.2065,
      "epoch": 0.5246720799500312,
      "grad_norm": 209.0712127685547,
      "learning_rate": 5.006147640545346e-05,
      "loss": 0.0616299569606781,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0651,
      "step": 17640
    },
    {
      "entropy_step_1": 0.1727,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.1676,
      "epoch": 0.5249695131019303,
      "grad_norm": 79.71540069580078,
      "learning_rate": 5.001229528406466e-05,
      "loss": 0.05129517316818237,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.048,
      "loss_step_3": 0.0394,
      "step": 17650
    },
    {
      "entropy_step_1": 0.1826,
      "entropy_step_2": 0.1894,
      "entropy_step_3": 0.2,
      "epoch": 0.5252669462538294,
      "grad_norm": 111.7955093383789,
      "learning_rate": 4.996311415078e-05,
      "loss": 0.062320882081985475,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0671,
      "step": 17660
    },
    {
      "entropy_step_1": 0.2029,
      "entropy_step_2": 0.2094,
      "entropy_step_3": 0.218,
      "epoch": 0.5255643794057285,
      "grad_norm": 55.53071212768555,
      "learning_rate": 4.991393305318291e-05,
      "loss": 0.06126497387886047,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0575,
      "step": 17670
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.2034,
      "epoch": 0.5258618125576277,
      "grad_norm": 230.04689025878906,
      "learning_rate": 4.986475203885677e-05,
      "loss": 0.05412328839302063,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0573,
      "step": 17680
    },
    {
      "entropy_step_1": 0.1803,
      "entropy_step_2": 0.1854,
      "entropy_step_3": 0.1941,
      "epoch": 0.5261592457095268,
      "grad_norm": 90.98799896240234,
      "learning_rate": 4.981557115538485e-05,
      "loss": 0.061507958173751834,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0608,
      "step": 17690
    },
    {
      "entropy_step_1": 0.1955,
      "entropy_step_2": 0.1975,
      "entropy_step_3": 0.2043,
      "epoch": 0.5264566788614259,
      "grad_norm": 77.3071060180664,
      "learning_rate": 4.976639045035037e-05,
      "loss": 0.05400455594062805,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0569,
      "step": 17700
    },
    {
      "entropy_step_1": 0.1969,
      "entropy_step_2": 0.1946,
      "entropy_step_3": 0.2251,
      "epoch": 0.526754112013325,
      "grad_norm": 173.75978088378906,
      "learning_rate": 4.971720997133629e-05,
      "loss": 0.0642869532108307,
      "loss_step_1": 0.0703,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0704,
      "step": 17710
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.2104,
      "entropy_step_3": 0.2234,
      "epoch": 0.5270515451652241,
      "grad_norm": 75.75122833251953,
      "learning_rate": 4.9668029765925415e-05,
      "loss": 0.06398981809616089,
      "loss_step_1": 0.0688,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0722,
      "step": 17720
    },
    {
      "entropy_step_1": 0.2175,
      "entropy_step_2": 0.2146,
      "entropy_step_3": 0.2407,
      "epoch": 0.5273489783171232,
      "grad_norm": 72.70016479492188,
      "learning_rate": 4.961884988170027e-05,
      "loss": 0.06136447191238403,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0608,
      "step": 17730
    },
    {
      "entropy_step_1": 0.216,
      "entropy_step_2": 0.2165,
      "entropy_step_3": 0.2432,
      "epoch": 0.5276464114690224,
      "grad_norm": 246.0258331298828,
      "learning_rate": 4.956967036624304e-05,
      "loss": 0.07247918844223022,
      "loss_step_1": 0.0722,
      "loss_step_2": 0.0719,
      "loss_step_3": 0.0739,
      "step": 17740
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2047,
      "entropy_step_3": 0.215,
      "epoch": 0.5279438446209215,
      "grad_norm": 96.25411224365234,
      "learning_rate": 4.952049126713558e-05,
      "loss": 0.0655779480934143,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0632,
      "step": 17750
    },
    {
      "entropy_step_1": 0.2282,
      "entropy_step_2": 0.2239,
      "entropy_step_3": 0.2236,
      "epoch": 0.5282412777728206,
      "grad_norm": 53.93830871582031,
      "learning_rate": 4.947131263195935e-05,
      "loss": 0.06043356657028198,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0644,
      "step": 17760
    },
    {
      "entropy_step_1": 0.1985,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.1841,
      "epoch": 0.5285387109247197,
      "grad_norm": 192.39993286132812,
      "learning_rate": 4.942213450829536e-05,
      "loss": 0.058592194318771364,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0529,
      "step": 17770
    },
    {
      "entropy_step_1": 0.1977,
      "entropy_step_2": 0.2009,
      "entropy_step_3": 0.1877,
      "epoch": 0.5288361440766188,
      "grad_norm": 102.90760040283203,
      "learning_rate": 4.937295694372407e-05,
      "loss": 0.05557789206504822,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0498,
      "step": 17780
    },
    {
      "entropy_step_1": 0.1935,
      "entropy_step_2": 0.1935,
      "entropy_step_3": 0.1893,
      "epoch": 0.5291335772285178,
      "grad_norm": 186.2827606201172,
      "learning_rate": 4.932377998582548e-05,
      "loss": 0.06067904233932495,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0592,
      "step": 17790
    },
    {
      "entropy_step_1": 0.2082,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.2373,
      "epoch": 0.529431010380417,
      "grad_norm": 94.62532043457031,
      "learning_rate": 4.927460368217895e-05,
      "loss": 0.059291130304336546,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0863,
      "step": 17800
    },
    {
      "entropy_step_1": 0.2159,
      "entropy_step_2": 0.2194,
      "entropy_step_3": 0.2254,
      "epoch": 0.5297284435323161,
      "grad_norm": 631.2059936523438,
      "learning_rate": 4.922542808036322e-05,
      "loss": 0.06676125526428223,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.0723,
      "step": 17810
    },
    {
      "entropy_step_1": 0.1953,
      "entropy_step_2": 0.2005,
      "entropy_step_3": 0.2179,
      "epoch": 0.5300258766842152,
      "grad_norm": 84.25624084472656,
      "learning_rate": 4.917625322795639e-05,
      "loss": 0.062269604206085204,
      "loss_step_1": 0.0647,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.0676,
      "step": 17820
    },
    {
      "entropy_step_1": 0.2076,
      "entropy_step_2": 0.2136,
      "entropy_step_3": 0.2256,
      "epoch": 0.5303233098361143,
      "grad_norm": 52.00728988647461,
      "learning_rate": 4.9127079172535755e-05,
      "loss": 0.05747621059417725,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0531,
      "step": 17830
    },
    {
      "entropy_step_1": 0.1813,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.1963,
      "epoch": 0.5306207429880134,
      "grad_norm": 129.48751831054688,
      "learning_rate": 4.907790596167789e-05,
      "loss": 0.0587393581867218,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0609,
      "step": 17840
    },
    {
      "entropy_step_1": 0.1828,
      "entropy_step_2": 0.1926,
      "entropy_step_3": 0.1794,
      "epoch": 0.5309181761399125,
      "grad_norm": 203.2945556640625,
      "learning_rate": 4.902873364295856e-05,
      "loss": 0.061969465017318724,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0584,
      "step": 17850
    },
    {
      "entropy_step_1": 0.1955,
      "entropy_step_2": 0.2035,
      "entropy_step_3": 0.2108,
      "epoch": 0.5312156092918117,
      "grad_norm": 108.30489349365234,
      "learning_rate": 4.897956226395266e-05,
      "loss": 0.06415273547172547,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0578,
      "step": 17860
    },
    {
      "entropy_step_1": 0.2327,
      "entropy_step_2": 0.2373,
      "entropy_step_3": 0.234,
      "epoch": 0.5315130424437108,
      "grad_norm": 66.34414672851562,
      "learning_rate": 4.893039187223414e-05,
      "loss": 0.06744056344032287,
      "loss_step_1": 0.068,
      "loss_step_2": 0.0675,
      "loss_step_3": 0.0643,
      "step": 17870
    },
    {
      "entropy_step_1": 0.2295,
      "entropy_step_2": 0.2385,
      "entropy_step_3": 0.2299,
      "epoch": 0.5318104755956099,
      "grad_norm": 113.97661590576172,
      "learning_rate": 4.8881222515376034e-05,
      "loss": 0.06651098728179931,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0655,
      "step": 17880
    },
    {
      "entropy_step_1": 0.2224,
      "entropy_step_2": 0.2263,
      "entropy_step_3": 0.2236,
      "epoch": 0.532107908747509,
      "grad_norm": 126.71465301513672,
      "learning_rate": 4.8832054240950384e-05,
      "loss": 0.07017752528190613,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0627,
      "step": 17890
    },
    {
      "entropy_step_1": 0.2106,
      "entropy_step_2": 0.2092,
      "entropy_step_3": 0.2215,
      "epoch": 0.5324053418994081,
      "grad_norm": 192.26312255859375,
      "learning_rate": 4.878288709652813e-05,
      "loss": 0.06304375529289245,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0708,
      "step": 17900
    },
    {
      "entropy_step_1": 0.2114,
      "entropy_step_2": 0.2078,
      "entropy_step_3": 0.2157,
      "epoch": 0.5327027750513073,
      "grad_norm": 103.72122955322266,
      "learning_rate": 4.87337211296792e-05,
      "loss": 0.06304352283477783,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0549,
      "step": 17910
    },
    {
      "entropy_step_1": 0.2126,
      "entropy_step_2": 0.2039,
      "entropy_step_3": 0.2178,
      "epoch": 0.5330002082032064,
      "grad_norm": 872.387451171875,
      "learning_rate": 4.868455638797228e-05,
      "loss": 0.06207572221755982,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.07,
      "step": 17920
    },
    {
      "entropy_step_1": 0.1987,
      "entropy_step_2": 0.1946,
      "entropy_step_3": 0.2031,
      "epoch": 0.5332976413551055,
      "grad_norm": 115.23902130126953,
      "learning_rate": 4.8635392918974955e-05,
      "loss": 0.052202391624450686,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0581,
      "step": 17930
    },
    {
      "entropy_step_1": 0.2113,
      "entropy_step_2": 0.209,
      "entropy_step_3": 0.2248,
      "epoch": 0.5335950745070045,
      "grad_norm": 77.26659393310547,
      "learning_rate": 4.858623077025355e-05,
      "loss": 0.06088613867759705,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.0716,
      "step": 17940
    },
    {
      "entropy_step_1": 0.1925,
      "entropy_step_2": 0.1926,
      "entropy_step_3": 0.1738,
      "epoch": 0.5338925076589036,
      "grad_norm": 575.1509399414062,
      "learning_rate": 4.8537069989373116e-05,
      "loss": 0.06057254076004028,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0466,
      "step": 17950
    },
    {
      "entropy_step_1": 0.2139,
      "entropy_step_2": 0.2125,
      "entropy_step_3": 0.21,
      "epoch": 0.5341899408108027,
      "grad_norm": 68.07270050048828,
      "learning_rate": 4.8487910623897385e-05,
      "loss": 0.06492554545402526,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0556,
      "step": 17960
    },
    {
      "entropy_step_1": 0.218,
      "entropy_step_2": 0.2132,
      "entropy_step_3": 0.2177,
      "epoch": 0.5344873739627019,
      "grad_norm": 211.41354370117188,
      "learning_rate": 4.8438752721388684e-05,
      "loss": 0.06363405585289002,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.0654,
      "step": 17970
    },
    {
      "entropy_step_1": 0.1981,
      "entropy_step_2": 0.1969,
      "entropy_step_3": 0.2126,
      "epoch": 0.534784807114601,
      "grad_norm": 87.13505554199219,
      "learning_rate": 4.838959632940797e-05,
      "loss": 0.06280394792556762,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0799,
      "step": 17980
    },
    {
      "entropy_step_1": 0.1941,
      "entropy_step_2": 0.1915,
      "entropy_step_3": 0.1923,
      "epoch": 0.5350822402665001,
      "grad_norm": 60.013668060302734,
      "learning_rate": 4.834044149551472e-05,
      "loss": 0.058250820636749266,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0607,
      "step": 17990
    },
    {
      "entropy_step_1": 0.1628,
      "entropy_step_2": 0.1628,
      "entropy_step_3": 0.182,
      "epoch": 0.5353796734183992,
      "grad_norm": 412.2667236328125,
      "learning_rate": 4.8291288267266925e-05,
      "loss": 0.06088484525680542,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0579,
      "step": 18000
    },
    {
      "epoch": 0.5353796734183992,
      "eval_accuracy": 0.865,
      "eval_f1": 0.8644272049408752,
      "eval_loss": 0.1279831975698471,
      "eval_precision": 0.8712745397212898,
      "eval_recall": 0.865,
      "eval_runtime": 6.5539,
      "eval_samples_per_second": 76.29,
      "eval_steps_per_second": 9.613,
      "step": 18000
    },
    {
      "entropy_step_1": 0.1713,
      "entropy_step_2": 0.1727,
      "entropy_step_3": 0.2032,
      "epoch": 0.5356771065702983,
      "grad_norm": 108.24931335449219,
      "learning_rate": 4.824213669222096e-05,
      "loss": 0.055678766965866086,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0722,
      "step": 18010
    },
    {
      "entropy_step_1": 0.2092,
      "entropy_step_2": 0.2079,
      "entropy_step_3": 0.1951,
      "epoch": 0.5359745397221974,
      "grad_norm": 85.41950988769531,
      "learning_rate": 4.8192986817931666e-05,
      "loss": 0.060350388288497925,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.055,
      "step": 18020
    },
    {
      "entropy_step_1": 0.1871,
      "entropy_step_2": 0.1889,
      "entropy_step_3": 0.1906,
      "epoch": 0.5362719728740966,
      "grad_norm": 41.7197380065918,
      "learning_rate": 4.8143838691952215e-05,
      "loss": 0.05576645135879517,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0556,
      "step": 18030
    },
    {
      "entropy_step_1": 0.1928,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.1977,
      "epoch": 0.5365694060259957,
      "grad_norm": 181.98269653320312,
      "learning_rate": 4.8094692361834095e-05,
      "loss": 0.05729478597640991,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0557,
      "step": 18040
    },
    {
      "entropy_step_1": 0.1958,
      "entropy_step_2": 0.2034,
      "entropy_step_3": 0.2035,
      "epoch": 0.5368668391778948,
      "grad_norm": 38.982948303222656,
      "learning_rate": 4.804554787512704e-05,
      "loss": 0.06007210612297058,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0697,
      "step": 18050
    },
    {
      "entropy_step_1": 0.2095,
      "entropy_step_2": 0.2165,
      "entropy_step_3": 0.2315,
      "epoch": 0.5371642723297939,
      "grad_norm": 126.90457916259766,
      "learning_rate": 4.7996405279379e-05,
      "loss": 0.058781874179840085,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0649,
      "step": 18060
    },
    {
      "entropy_step_1": 0.1955,
      "entropy_step_2": 0.1937,
      "entropy_step_3": 0.1722,
      "epoch": 0.537461705481693,
      "grad_norm": 99.76029205322266,
      "learning_rate": 4.794726462213612e-05,
      "loss": 0.05340844392776489,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0395,
      "step": 18070
    },
    {
      "entropy_step_1": 0.1711,
      "entropy_step_2": 0.1718,
      "entropy_step_3": 0.1625,
      "epoch": 0.5377591386335921,
      "grad_norm": 51.21047592163086,
      "learning_rate": 4.789812595094265e-05,
      "loss": 0.05429608821868896,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0502,
      "step": 18080
    },
    {
      "entropy_step_1": 0.2041,
      "entropy_step_2": 0.2061,
      "entropy_step_3": 0.2068,
      "epoch": 0.5380565717854912,
      "grad_norm": 139.80613708496094,
      "learning_rate": 4.7848989313340925e-05,
      "loss": 0.06473402976989746,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0602,
      "step": 18090
    },
    {
      "entropy_step_1": 0.2106,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.2033,
      "epoch": 0.5383540049373903,
      "grad_norm": 144.36183166503906,
      "learning_rate": 4.779985475687131e-05,
      "loss": 0.06557906866073608,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0685,
      "step": 18100
    },
    {
      "entropy_step_1": 0.1935,
      "entropy_step_2": 0.1925,
      "entropy_step_3": 0.1981,
      "epoch": 0.5386514380892894,
      "grad_norm": 162.57876586914062,
      "learning_rate": 4.775072232907215e-05,
      "loss": 0.05357608199119568,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.052,
      "loss_step_3": 0.0553,
      "step": 18110
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1869,
      "entropy_step_3": 0.1867,
      "epoch": 0.5389488712411885,
      "grad_norm": 118.41423034667969,
      "learning_rate": 4.770159207747973e-05,
      "loss": 0.05308179259300232,
      "loss_step_1": 0.0514,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0443,
      "step": 18120
    },
    {
      "entropy_step_1": 0.1954,
      "entropy_step_2": 0.1941,
      "entropy_step_3": 0.1935,
      "epoch": 0.5392463043930876,
      "grad_norm": 85.88375091552734,
      "learning_rate": 4.7652464049628256e-05,
      "loss": 0.05897650718688965,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.06,
      "step": 18130
    },
    {
      "entropy_step_1": 0.1725,
      "entropy_step_2": 0.173,
      "entropy_step_3": 0.1841,
      "epoch": 0.5395437375449867,
      "grad_norm": 190.71632385253906,
      "learning_rate": 4.760333829304978e-05,
      "loss": 0.05242436528205872,
      "loss_step_1": 0.0504,
      "loss_step_2": 0.0494,
      "loss_step_3": 0.0545,
      "step": 18140
    },
    {
      "entropy_step_1": 0.1981,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.2043,
      "epoch": 0.5398411706968859,
      "grad_norm": 98.22464752197266,
      "learning_rate": 4.75542148552741e-05,
      "loss": 0.0666219413280487,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0625,
      "step": 18150
    },
    {
      "entropy_step_1": 0.2063,
      "entropy_step_2": 0.2,
      "entropy_step_3": 0.2167,
      "epoch": 0.540138603848785,
      "grad_norm": 131.6064910888672,
      "learning_rate": 4.750509378382882e-05,
      "loss": 0.06591436862945557,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0689,
      "loss_step_3": 0.0911,
      "step": 18160
    },
    {
      "entropy_step_1": 0.213,
      "entropy_step_2": 0.2043,
      "entropy_step_3": 0.218,
      "epoch": 0.5404360370006841,
      "grad_norm": 133.586181640625,
      "learning_rate": 4.745597512623925e-05,
      "loss": 0.056950193643569944,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0657,
      "step": 18170
    },
    {
      "entropy_step_1": 0.1958,
      "entropy_step_2": 0.1901,
      "entropy_step_3": 0.1981,
      "epoch": 0.5407334701525832,
      "grad_norm": 84.86332702636719,
      "learning_rate": 4.7406858930028387e-05,
      "loss": 0.06042500734329224,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0607,
      "step": 18180
    },
    {
      "entropy_step_1": 0.2068,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.1838,
      "epoch": 0.5410309033044823,
      "grad_norm": 171.87120056152344,
      "learning_rate": 4.7357745242716775e-05,
      "loss": 0.0630229115486145,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0597,
      "step": 18190
    },
    {
      "entropy_step_1": 0.2149,
      "entropy_step_2": 0.2069,
      "entropy_step_3": 0.1997,
      "epoch": 0.5413283364563815,
      "grad_norm": 148.844482421875,
      "learning_rate": 4.730863411182259e-05,
      "loss": 0.05876861214637756,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0631,
      "step": 18200
    },
    {
      "entropy_step_1": 0.2058,
      "entropy_step_2": 0.1991,
      "entropy_step_3": 0.2039,
      "epoch": 0.5416257696082806,
      "grad_norm": 90.87458801269531,
      "learning_rate": 4.7259525584861524e-05,
      "loss": 0.0556038498878479,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0577,
      "step": 18210
    },
    {
      "entropy_step_1": 0.1683,
      "entropy_step_2": 0.165,
      "entropy_step_3": 0.1778,
      "epoch": 0.5419232027601797,
      "grad_norm": 337.1847229003906,
      "learning_rate": 4.7210419709346756e-05,
      "loss": 0.060315167903900145,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0639,
      "step": 18220
    },
    {
      "entropy_step_1": 0.2245,
      "entropy_step_2": 0.2192,
      "entropy_step_3": 0.2097,
      "epoch": 0.5422206359120788,
      "grad_norm": 424.48944091796875,
      "learning_rate": 4.7161316532788876e-05,
      "loss": 0.06965391635894776,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0597,
      "step": 18230
    },
    {
      "entropy_step_1": 0.213,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.1946,
      "epoch": 0.5425180690639779,
      "grad_norm": 75.38604736328125,
      "learning_rate": 4.711221610269586e-05,
      "loss": 0.06265370845794678,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0524,
      "step": 18240
    },
    {
      "entropy_step_1": 0.2083,
      "entropy_step_2": 0.2045,
      "entropy_step_3": 0.218,
      "epoch": 0.5428155022158769,
      "grad_norm": 108.89220428466797,
      "learning_rate": 4.706311846657308e-05,
      "loss": 0.05510169267654419,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.066,
      "step": 18250
    },
    {
      "entropy_step_1": 0.1732,
      "entropy_step_2": 0.171,
      "entropy_step_3": 0.1887,
      "epoch": 0.543112935367776,
      "grad_norm": 154.59237670898438,
      "learning_rate": 4.701402367192313e-05,
      "loss": 0.06363030672073364,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0639,
      "step": 18260
    },
    {
      "entropy_step_1": 0.1918,
      "entropy_step_2": 0.1883,
      "entropy_step_3": 0.2038,
      "epoch": 0.5434103685196752,
      "grad_norm": 117.29946899414062,
      "learning_rate": 4.696493176624591e-05,
      "loss": 0.06123839616775513,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0652,
      "step": 18270
    },
    {
      "entropy_step_1": 0.2234,
      "entropy_step_2": 0.2196,
      "entropy_step_3": 0.2136,
      "epoch": 0.5437078016715743,
      "grad_norm": 74.16645812988281,
      "learning_rate": 4.6915842797038526e-05,
      "loss": 0.06360086798667908,
      "loss_step_1": 0.068,
      "loss_step_2": 0.0675,
      "loss_step_3": 0.0664,
      "step": 18280
    },
    {
      "entropy_step_1": 0.2187,
      "entropy_step_2": 0.2151,
      "entropy_step_3": 0.2188,
      "epoch": 0.5440052348234734,
      "grad_norm": 67.2227554321289,
      "learning_rate": 4.6866756811795173e-05,
      "loss": 0.06500118970870972,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0542,
      "step": 18290
    },
    {
      "entropy_step_1": 0.2131,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2071,
      "epoch": 0.5443026679753725,
      "grad_norm": 48.815223693847656,
      "learning_rate": 4.681767385800725e-05,
      "loss": 0.058232074975967406,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0629,
      "step": 18300
    },
    {
      "entropy_step_1": 0.1897,
      "entropy_step_2": 0.1824,
      "entropy_step_3": 0.2,
      "epoch": 0.5446001011272716,
      "grad_norm": 277.75177001953125,
      "learning_rate": 4.676859398316316e-05,
      "loss": 0.05558261275291443,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0607,
      "step": 18310
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.174,
      "entropy_step_3": 0.1818,
      "epoch": 0.5448975342791708,
      "grad_norm": 59.570091247558594,
      "learning_rate": 4.671951723474839e-05,
      "loss": 0.057631248235702516,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0603,
      "step": 18320
    },
    {
      "entropy_step_1": 0.2032,
      "entropy_step_2": 0.1973,
      "entropy_step_3": 0.1998,
      "epoch": 0.5451949674310699,
      "grad_norm": 122.17579650878906,
      "learning_rate": 4.667044366024531e-05,
      "loss": 0.05772364139556885,
      "loss_step_1": 0.058,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0594,
      "step": 18330
    },
    {
      "entropy_step_1": 0.191,
      "entropy_step_2": 0.1831,
      "entropy_step_3": 0.1866,
      "epoch": 0.545492400582969,
      "grad_norm": 81.11031341552734,
      "learning_rate": 4.6621373307133286e-05,
      "loss": 0.057651680707931516,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0651,
      "step": 18340
    },
    {
      "entropy_step_1": 0.198,
      "entropy_step_2": 0.1883,
      "entropy_step_3": 0.2055,
      "epoch": 0.5457898337348681,
      "grad_norm": 82.10770416259766,
      "learning_rate": 4.657230622288857e-05,
      "loss": 0.0613916277885437,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.0695,
      "step": 18350
    },
    {
      "entropy_step_1": 0.2119,
      "entropy_step_2": 0.2042,
      "entropy_step_3": 0.2187,
      "epoch": 0.5460872668867672,
      "grad_norm": 118.46488952636719,
      "learning_rate": 4.6523242454984226e-05,
      "loss": 0.05936986804008484,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0605,
      "step": 18360
    },
    {
      "entropy_step_1": 0.185,
      "entropy_step_2": 0.1813,
      "entropy_step_3": 0.1769,
      "epoch": 0.5463847000386663,
      "grad_norm": 106.26722717285156,
      "learning_rate": 4.6474182050890126e-05,
      "loss": 0.05461503863334656,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.052,
      "step": 18370
    },
    {
      "entropy_step_1": 0.1976,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.2019,
      "epoch": 0.5466821331905655,
      "grad_norm": 99.22588348388672,
      "learning_rate": 4.642512505807285e-05,
      "loss": 0.06393234729766846,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0556,
      "step": 18380
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.213,
      "entropy_step_3": 0.2276,
      "epoch": 0.5469795663424646,
      "grad_norm": 42.79029846191406,
      "learning_rate": 4.637607152399572e-05,
      "loss": 0.0639247715473175,
      "loss_step_1": 0.0713,
      "loss_step_2": 0.0707,
      "loss_step_3": 0.0704,
      "step": 18390
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1889,
      "entropy_step_3": 0.1954,
      "epoch": 0.5472769994943636,
      "grad_norm": 106.71253967285156,
      "learning_rate": 4.632702149611871e-05,
      "loss": 0.05947005748748779,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0619,
      "step": 18400
    },
    {
      "entropy_step_1": 0.2181,
      "entropy_step_2": 0.2187,
      "entropy_step_3": 0.2163,
      "epoch": 0.5475744326462627,
      "grad_norm": 172.10865783691406,
      "learning_rate": 4.62779750218984e-05,
      "loss": 0.059919679164886476,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0673,
      "step": 18410
    },
    {
      "entropy_step_1": 0.1752,
      "entropy_step_2": 0.1728,
      "entropy_step_3": 0.1697,
      "epoch": 0.5478718657981618,
      "grad_norm": 98.50541687011719,
      "learning_rate": 4.622893214878788e-05,
      "loss": 0.0514198899269104,
      "loss_step_1": 0.0506,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.049,
      "step": 18420
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1748,
      "entropy_step_3": 0.1617,
      "epoch": 0.548169298950061,
      "grad_norm": 80.95499420166016,
      "learning_rate": 4.617989292423682e-05,
      "loss": 0.06207348108291626,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0558,
      "step": 18430
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1793,
      "entropy_step_3": 0.1933,
      "epoch": 0.5484667321019601,
      "grad_norm": 58.255958557128906,
      "learning_rate": 4.613085739569134e-05,
      "loss": 0.0544292151927948,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.06,
      "step": 18440
    },
    {
      "entropy_step_1": 0.2165,
      "entropy_step_2": 0.2112,
      "entropy_step_3": 0.211,
      "epoch": 0.5487641652538592,
      "grad_norm": 88.45796966552734,
      "learning_rate": 4.6081825610593964e-05,
      "loss": 0.05549845099449158,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0436,
      "step": 18450
    },
    {
      "entropy_step_1": 0.1817,
      "entropy_step_2": 0.1778,
      "entropy_step_3": 0.1836,
      "epoch": 0.5490615984057583,
      "grad_norm": 345.8533630371094,
      "learning_rate": 4.6032797616383656e-05,
      "loss": 0.058418309688568114,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0495,
      "loss_step_3": 0.0538,
      "step": 18460
    },
    {
      "entropy_step_1": 0.1997,
      "entropy_step_2": 0.1928,
      "entropy_step_3": 0.2008,
      "epoch": 0.5493590315576574,
      "grad_norm": 159.07745361328125,
      "learning_rate": 4.598377346049561e-05,
      "loss": 0.0608882486820221,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0658,
      "step": 18470
    },
    {
      "entropy_step_1": 0.2093,
      "entropy_step_2": 0.2046,
      "entropy_step_3": 0.2206,
      "epoch": 0.5496564647095565,
      "grad_norm": 112.54340362548828,
      "learning_rate": 4.593475319036139e-05,
      "loss": 0.061955219507217406,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0723,
      "step": 18480
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.1872,
      "entropy_step_3": 0.1938,
      "epoch": 0.5499538978614557,
      "grad_norm": 61.368568420410156,
      "learning_rate": 4.588573685340877e-05,
      "loss": 0.051070022583007815,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0595,
      "step": 18490
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.176,
      "entropy_step_3": 0.1649,
      "epoch": 0.5502513310133548,
      "grad_norm": 112.87258911132812,
      "learning_rate": 4.5836724497061736e-05,
      "loss": 0.053687053918838504,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0561,
      "step": 18500
    },
    {
      "entropy_step_1": 0.1949,
      "entropy_step_2": 0.1947,
      "entropy_step_3": 0.2041,
      "epoch": 0.5505487641652539,
      "grad_norm": 97.57627868652344,
      "learning_rate": 4.5787716168740414e-05,
      "loss": 0.0549771785736084,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0693,
      "step": 18510
    },
    {
      "entropy_step_1": 0.1924,
      "entropy_step_2": 0.1951,
      "entropy_step_3": 0.1887,
      "epoch": 0.550846197317153,
      "grad_norm": 232.63841247558594,
      "learning_rate": 4.5738711915860994e-05,
      "loss": 0.058793914318084714,
      "loss_step_1": 0.0656,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0654,
      "step": 18520
    },
    {
      "entropy_step_1": 0.1819,
      "entropy_step_2": 0.1875,
      "entropy_step_3": 0.1895,
      "epoch": 0.5511436304690521,
      "grad_norm": 734.416259765625,
      "learning_rate": 4.568971178583578e-05,
      "loss": 0.05338549017906189,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0546,
      "step": 18530
    },
    {
      "entropy_step_1": 0.186,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.2016,
      "epoch": 0.5514410636209512,
      "grad_norm": 59.0011100769043,
      "learning_rate": 4.564071582607307e-05,
      "loss": 0.0552869975566864,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0531,
      "step": 18540
    },
    {
      "entropy_step_1": 0.1863,
      "entropy_step_2": 0.1991,
      "entropy_step_3": 0.2085,
      "epoch": 0.5517384967728503,
      "grad_norm": 57.891292572021484,
      "learning_rate": 4.559172408397713e-05,
      "loss": 0.061862516403198245,
      "loss_step_1": 0.0668,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0684,
      "step": 18550
    },
    {
      "entropy_step_1": 0.1837,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.2129,
      "epoch": 0.5520359299247494,
      "grad_norm": 243.36483764648438,
      "learning_rate": 4.55427366069481e-05,
      "loss": 0.055832237005233765,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0747,
      "step": 18560
    },
    {
      "entropy_step_1": 0.1864,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.1988,
      "epoch": 0.5523333630766485,
      "grad_norm": 52.17774200439453,
      "learning_rate": 4.5493753442382045e-05,
      "loss": 0.05906707644462585,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.063,
      "step": 18570
    },
    {
      "entropy_step_1": 0.1881,
      "entropy_step_2": 0.1998,
      "entropy_step_3": 0.2015,
      "epoch": 0.5526307962285476,
      "grad_norm": 156.29603576660156,
      "learning_rate": 4.544477463767086e-05,
      "loss": 0.062281334400177,
      "loss_step_1": 0.0646,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0572,
      "step": 18580
    },
    {
      "entropy_step_1": 0.185,
      "entropy_step_2": 0.1957,
      "entropy_step_3": 0.2102,
      "epoch": 0.5529282293804467,
      "grad_norm": 502.4995422363281,
      "learning_rate": 4.5395800240202216e-05,
      "loss": 0.05238840579986572,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.063,
      "step": 18590
    },
    {
      "entropy_step_1": 0.1978,
      "entropy_step_2": 0.2055,
      "entropy_step_3": 0.2221,
      "epoch": 0.5532256625323458,
      "grad_norm": 203.61489868164062,
      "learning_rate": 4.5346830297359474e-05,
      "loss": 0.05950433015823364,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0781,
      "step": 18600
    },
    {
      "entropy_step_1": 0.2008,
      "entropy_step_2": 0.2057,
      "entropy_step_3": 0.2035,
      "epoch": 0.553523095684245,
      "grad_norm": 80.4872817993164,
      "learning_rate": 4.529786485652173e-05,
      "loss": 0.05878233313560486,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0567,
      "step": 18610
    },
    {
      "entropy_step_1": 0.1848,
      "entropy_step_2": 0.1873,
      "entropy_step_3": 0.1971,
      "epoch": 0.5538205288361441,
      "grad_norm": 106.3681411743164,
      "learning_rate": 4.524890396506373e-05,
      "loss": 0.06237127184867859,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.0555,
      "step": 18620
    },
    {
      "entropy_step_1": 0.2022,
      "entropy_step_2": 0.2041,
      "entropy_step_3": 0.1997,
      "epoch": 0.5541179619880432,
      "grad_norm": 99.4492416381836,
      "learning_rate": 4.519994767035578e-05,
      "loss": 0.06671778559684753,
      "loss_step_1": 0.071,
      "loss_step_2": 0.0694,
      "loss_step_3": 0.0658,
      "step": 18630
    },
    {
      "entropy_step_1": 0.1846,
      "entropy_step_2": 0.1874,
      "entropy_step_3": 0.1761,
      "epoch": 0.5544153951399423,
      "grad_norm": 185.8136444091797,
      "learning_rate": 4.5150996019763783e-05,
      "loss": 0.05498636960983276,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0555,
      "step": 18640
    },
    {
      "entropy_step_1": 0.214,
      "entropy_step_2": 0.2148,
      "entropy_step_3": 0.2275,
      "epoch": 0.5547128282918414,
      "grad_norm": 76.56024932861328,
      "learning_rate": 4.5102049060649114e-05,
      "loss": 0.07494753003120422,
      "loss_step_1": 0.0794,
      "loss_step_2": 0.0804,
      "loss_step_3": 0.0839,
      "step": 18650
    },
    {
      "entropy_step_1": 0.2409,
      "entropy_step_2": 0.2427,
      "entropy_step_3": 0.2544,
      "epoch": 0.5550102614437405,
      "grad_norm": 67.06379699707031,
      "learning_rate": 4.505310684036862e-05,
      "loss": 0.06527275443077088,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0645,
      "loss_step_3": 0.0697,
      "step": 18660
    },
    {
      "entropy_step_1": 0.2053,
      "entropy_step_2": 0.2121,
      "entropy_step_3": 0.194,
      "epoch": 0.5553076945956397,
      "grad_norm": 179.6946258544922,
      "learning_rate": 4.500416940627457e-05,
      "loss": 0.06697585582733154,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0598,
      "step": 18670
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.1926,
      "entropy_step_3": 0.208,
      "epoch": 0.5556051277475388,
      "grad_norm": 67.18636322021484,
      "learning_rate": 4.49552368057146e-05,
      "loss": 0.05604323744773865,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.053,
      "step": 18680
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1959,
      "entropy_step_3": 0.1973,
      "epoch": 0.5559025608994379,
      "grad_norm": 167.35101318359375,
      "learning_rate": 4.490630908603168e-05,
      "loss": 0.0507239043712616,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0498,
      "loss_step_3": 0.0514,
      "step": 18690
    },
    {
      "entropy_step_1": 0.1947,
      "entropy_step_2": 0.2048,
      "entropy_step_3": 0.1987,
      "epoch": 0.5561999940513369,
      "grad_norm": 136.00181579589844,
      "learning_rate": 4.4857386294564005e-05,
      "loss": 0.058824574947357176,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0666,
      "step": 18700
    },
    {
      "entropy_step_1": 0.1758,
      "entropy_step_2": 0.1859,
      "entropy_step_3": 0.1757,
      "epoch": 0.556497427203236,
      "grad_norm": 78.69524383544922,
      "learning_rate": 4.480846847864507e-05,
      "loss": 0.05532684922218323,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0474,
      "step": 18710
    },
    {
      "entropy_step_1": 0.2012,
      "entropy_step_2": 0.2084,
      "entropy_step_3": 0.1984,
      "epoch": 0.5567948603551351,
      "grad_norm": 178.29197692871094,
      "learning_rate": 4.475955568560352e-05,
      "loss": 0.06188907623291016,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0562,
      "step": 18720
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.2053,
      "epoch": 0.5570922935070343,
      "grad_norm": 91.81689453125,
      "learning_rate": 4.471064796276318e-05,
      "loss": 0.07128803133964538,
      "loss_step_1": 0.0737,
      "loss_step_2": 0.072,
      "loss_step_3": 0.059,
      "step": 18730
    },
    {
      "entropy_step_1": 0.1879,
      "entropy_step_2": 0.1972,
      "entropy_step_3": 0.1827,
      "epoch": 0.5573897266589334,
      "grad_norm": 252.3079376220703,
      "learning_rate": 4.466174535744288e-05,
      "loss": 0.056901609897613524,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0521,
      "step": 18740
    },
    {
      "entropy_step_1": 0.1831,
      "entropy_step_2": 0.1922,
      "entropy_step_3": 0.1822,
      "epoch": 0.5576871598108325,
      "grad_norm": 115.51028442382812,
      "learning_rate": 4.4612847916956584e-05,
      "loss": 0.06041080355644226,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0529,
      "step": 18750
    },
    {
      "entropy_step_1": 0.1938,
      "entropy_step_2": 0.2008,
      "entropy_step_3": 0.2166,
      "epoch": 0.5579845929627316,
      "grad_norm": 111.15742492675781,
      "learning_rate": 4.4563955688613235e-05,
      "loss": 0.0618876039981842,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0677,
      "step": 18760
    },
    {
      "entropy_step_1": 0.1991,
      "entropy_step_2": 0.2051,
      "entropy_step_3": 0.2203,
      "epoch": 0.5582820261146307,
      "grad_norm": 78.23863983154297,
      "learning_rate": 4.4515068719716735e-05,
      "loss": 0.052621346712112424,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0494,
      "step": 18770
    },
    {
      "entropy_step_1": 0.1512,
      "entropy_step_2": 0.1564,
      "entropy_step_3": 0.1632,
      "epoch": 0.5585794592665299,
      "grad_norm": 321.9844055175781,
      "learning_rate": 4.446618705756588e-05,
      "loss": 0.05491922497749328,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0578,
      "step": 18780
    },
    {
      "entropy_step_1": 0.175,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1666,
      "epoch": 0.558876892418429,
      "grad_norm": 89.81301879882812,
      "learning_rate": 4.441731074945433e-05,
      "loss": 0.06422932147979736,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0496,
      "step": 18790
    },
    {
      "entropy_step_1": 0.2256,
      "entropy_step_2": 0.2216,
      "entropy_step_3": 0.1961,
      "epoch": 0.5591743255703281,
      "grad_norm": 72.99198913574219,
      "learning_rate": 4.436843984267062e-05,
      "loss": 0.05880879163742066,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0451,
      "step": 18800
    },
    {
      "entropy_step_1": 0.2014,
      "entropy_step_2": 0.2,
      "entropy_step_3": 0.1579,
      "epoch": 0.5594717587222272,
      "grad_norm": 70.32460021972656,
      "learning_rate": 4.4319574384497966e-05,
      "loss": 0.06556882858276367,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0563,
      "step": 18810
    },
    {
      "entropy_step_1": 0.1953,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.1797,
      "epoch": 0.5597691918741263,
      "grad_norm": 90.84107971191406,
      "learning_rate": 4.427071442221439e-05,
      "loss": 0.05758882164955139,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0545,
      "step": 18820
    },
    {
      "entropy_step_1": 0.1748,
      "entropy_step_2": 0.1806,
      "entropy_step_3": 0.1812,
      "epoch": 0.5600666250260254,
      "grad_norm": 89.17076110839844,
      "learning_rate": 4.422186000309259e-05,
      "loss": 0.054364120960235594,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.052,
      "loss_step_3": 0.0545,
      "step": 18830
    },
    {
      "entropy_step_1": 0.1895,
      "entropy_step_2": 0.1942,
      "entropy_step_3": 0.1895,
      "epoch": 0.5603640581779246,
      "grad_norm": 143.40902709960938,
      "learning_rate": 4.4173011174399844e-05,
      "loss": 0.06057219505310059,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0524,
      "step": 18840
    },
    {
      "entropy_step_1": 0.2092,
      "entropy_step_2": 0.2131,
      "entropy_step_3": 0.2201,
      "epoch": 0.5606614913298236,
      "grad_norm": 141.65338134765625,
      "learning_rate": 4.4124167983398065e-05,
      "loss": 0.06276803016662598,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.068,
      "step": 18850
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.2003,
      "entropy_step_3": 0.2061,
      "epoch": 0.5609589244817227,
      "grad_norm": 182.9359130859375,
      "learning_rate": 4.407533047734373e-05,
      "loss": 0.0568616509437561,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0496,
      "step": 18860
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.188,
      "epoch": 0.5612563576336218,
      "grad_norm": 416.01861572265625,
      "learning_rate": 4.402649870348778e-05,
      "loss": 0.05776512622833252,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0658,
      "loss_step_3": 0.0547,
      "step": 18870
    },
    {
      "entropy_step_1": 0.2103,
      "entropy_step_2": 0.2117,
      "entropy_step_3": 0.2195,
      "epoch": 0.5615537907855209,
      "grad_norm": 110.08731842041016,
      "learning_rate": 4.39776727090756e-05,
      "loss": 0.07165269255638122,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0595,
      "step": 18880
    },
    {
      "entropy_step_1": 0.1931,
      "entropy_step_2": 0.1935,
      "entropy_step_3": 0.2062,
      "epoch": 0.56185122393742,
      "grad_norm": 143.75819396972656,
      "learning_rate": 4.392885254134702e-05,
      "loss": 0.0590234637260437,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0497,
      "step": 18890
    },
    {
      "entropy_step_1": 0.2317,
      "entropy_step_2": 0.2233,
      "entropy_step_3": 0.2141,
      "epoch": 0.5621486570893192,
      "grad_norm": 156.7171630859375,
      "learning_rate": 4.388003824753622e-05,
      "loss": 0.0762662947177887,
      "loss_step_1": 0.0847,
      "loss_step_2": 0.084,
      "loss_step_3": 0.0951,
      "step": 18900
    },
    {
      "entropy_step_1": 0.2043,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.2001,
      "epoch": 0.5624460902412183,
      "grad_norm": 42.96922302246094,
      "learning_rate": 4.383122987487169e-05,
      "loss": 0.056509703397750854,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0619,
      "step": 18910
    },
    {
      "entropy_step_1": 0.2009,
      "entropy_step_2": 0.2008,
      "entropy_step_3": 0.2063,
      "epoch": 0.5627435233931174,
      "grad_norm": 63.981292724609375,
      "learning_rate": 4.378242747057621e-05,
      "loss": 0.05610054135322571,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0632,
      "step": 18920
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1841,
      "entropy_step_3": 0.1616,
      "epoch": 0.5630409565450165,
      "grad_norm": 79.21601104736328,
      "learning_rate": 4.373363108186673e-05,
      "loss": 0.0596732497215271,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0482,
      "step": 18930
    },
    {
      "entropy_step_1": 0.2085,
      "entropy_step_2": 0.2111,
      "entropy_step_3": 0.2094,
      "epoch": 0.5633383896969156,
      "grad_norm": 75.13182830810547,
      "learning_rate": 4.368484075595445e-05,
      "loss": 0.05639088749885559,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0504,
      "step": 18940
    },
    {
      "entropy_step_1": 0.2028,
      "entropy_step_2": 0.2008,
      "entropy_step_3": 0.1941,
      "epoch": 0.5636358228488147,
      "grad_norm": 34.80706787109375,
      "learning_rate": 4.363605654004469e-05,
      "loss": 0.06254071593284607,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0624,
      "step": 18950
    },
    {
      "entropy_step_1": 0.1759,
      "entropy_step_2": 0.1749,
      "entropy_step_3": 0.1637,
      "epoch": 0.5639332560007139,
      "grad_norm": 85.22718048095703,
      "learning_rate": 4.35872784813368e-05,
      "loss": 0.05269927978515625,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0536,
      "step": 18960
    },
    {
      "entropy_step_1": 0.1831,
      "entropy_step_2": 0.1833,
      "entropy_step_3": 0.1872,
      "epoch": 0.564230689152613,
      "grad_norm": 135.88778686523438,
      "learning_rate": 4.353850662702424e-05,
      "loss": 0.06220476627349854,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0534,
      "step": 18970
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.188,
      "entropy_step_3": 0.1905,
      "epoch": 0.5645281223045121,
      "grad_norm": 323.3546142578125,
      "learning_rate": 4.3489741024294453e-05,
      "loss": 0.056964796781539914,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0533,
      "step": 18980
    },
    {
      "entropy_step_1": 0.1887,
      "entropy_step_2": 0.1872,
      "entropy_step_3": 0.1955,
      "epoch": 0.5648255554564112,
      "grad_norm": 79.35916137695312,
      "learning_rate": 4.3440981720328786e-05,
      "loss": 0.06929381489753723,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0697,
      "loss_step_3": 0.0723,
      "step": 18990
    },
    {
      "entropy_step_1": 0.1897,
      "entropy_step_2": 0.1865,
      "entropy_step_3": 0.1713,
      "epoch": 0.5651229886083102,
      "grad_norm": 62.08083724975586,
      "learning_rate": 4.3392228762302565e-05,
      "loss": 0.061182659864425656,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0563,
      "step": 19000
    },
    {
      "epoch": 0.5651229886083102,
      "eval_accuracy": 0.859,
      "eval_f1": 0.8578226291923418,
      "eval_loss": 0.12557853758335114,
      "eval_precision": 0.8712989049267951,
      "eval_recall": 0.859,
      "eval_runtime": 6.7832,
      "eval_samples_per_second": 73.712,
      "eval_steps_per_second": 9.288,
      "step": 19000
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1671,
      "epoch": 0.5654204217602093,
      "grad_norm": 84.8452377319336,
      "learning_rate": 4.334348219738494e-05,
      "loss": 0.05591617226600647,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0475,
      "step": 19010
    },
    {
      "entropy_step_1": 0.1815,
      "entropy_step_2": 0.1749,
      "entropy_step_3": 0.1913,
      "epoch": 0.5657178549121085,
      "grad_norm": 58.497188568115234,
      "learning_rate": 4.329474207273885e-05,
      "loss": 0.06225336194038391,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.071,
      "step": 19020
    },
    {
      "entropy_step_1": 0.2211,
      "entropy_step_2": 0.2108,
      "entropy_step_3": 0.2076,
      "epoch": 0.5660152880640076,
      "grad_norm": 54.36385726928711,
      "learning_rate": 4.3246008435521046e-05,
      "loss": 0.07124340534210205,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0694,
      "loss_step_3": 0.0598,
      "step": 19030
    },
    {
      "entropy_step_1": 0.2114,
      "entropy_step_2": 0.2025,
      "entropy_step_3": 0.2099,
      "epoch": 0.5663127212159067,
      "grad_norm": 96.48873901367188,
      "learning_rate": 4.319728133288197e-05,
      "loss": 0.05550875067710877,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0493,
      "step": 19040
    },
    {
      "entropy_step_1": 0.2118,
      "entropy_step_2": 0.2006,
      "entropy_step_3": 0.1912,
      "epoch": 0.5666101543678058,
      "grad_norm": 216.1522216796875,
      "learning_rate": 4.3148560811965776e-05,
      "loss": 0.06287323832511901,
      "loss_step_1": 0.0664,
      "loss_step_2": 0.0662,
      "loss_step_3": 0.0566,
      "step": 19050
    },
    {
      "entropy_step_1": 0.2292,
      "entropy_step_2": 0.2147,
      "entropy_step_3": 0.2156,
      "epoch": 0.5669075875197049,
      "grad_norm": 97.05988311767578,
      "learning_rate": 4.309984691991025e-05,
      "loss": 0.07463021874427796,
      "loss_step_1": 0.0699,
      "loss_step_2": 0.0698,
      "loss_step_3": 0.0716,
      "step": 19060
    },
    {
      "entropy_step_1": 0.2475,
      "entropy_step_2": 0.2361,
      "entropy_step_3": 0.2281,
      "epoch": 0.567205020671604,
      "grad_norm": 124.04602813720703,
      "learning_rate": 4.3051139703846695e-05,
      "loss": 0.06626745462417602,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.0596,
      "step": 19070
    },
    {
      "entropy_step_1": 0.2118,
      "entropy_step_2": 0.2056,
      "entropy_step_3": 0.1993,
      "epoch": 0.5675024538235032,
      "grad_norm": 171.5989227294922,
      "learning_rate": 4.300243921090002e-05,
      "loss": 0.05242066383361817,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0553,
      "step": 19080
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.1752,
      "entropy_step_3": 0.1509,
      "epoch": 0.5677998869754023,
      "grad_norm": 53.76844024658203,
      "learning_rate": 4.295374548818864e-05,
      "loss": 0.05757691860198975,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0515,
      "step": 19090
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.1975,
      "entropy_step_3": 0.1876,
      "epoch": 0.5680973201273014,
      "grad_norm": 47.436893463134766,
      "learning_rate": 4.2905058582824386e-05,
      "loss": 0.06317124962806701,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0597,
      "step": 19100
    },
    {
      "entropy_step_1": 0.2042,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.206,
      "epoch": 0.5683947532792005,
      "grad_norm": 55.209007263183594,
      "learning_rate": 4.285637854191248e-05,
      "loss": 0.06010483503341675,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0739,
      "step": 19110
    },
    {
      "entropy_step_1": 0.2042,
      "entropy_step_2": 0.2021,
      "entropy_step_3": 0.2079,
      "epoch": 0.5686921864310996,
      "grad_norm": 100.56433868408203,
      "learning_rate": 4.280770541255154e-05,
      "loss": 0.055810433626174924,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0688,
      "step": 19120
    },
    {
      "entropy_step_1": 0.1829,
      "entropy_step_2": 0.1844,
      "entropy_step_3": 0.1935,
      "epoch": 0.5689896195829988,
      "grad_norm": 138.644775390625,
      "learning_rate": 4.2759039241833465e-05,
      "loss": 0.059523981809616086,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.05,
      "step": 19130
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1878,
      "entropy_step_3": 0.1855,
      "epoch": 0.5692870527348979,
      "grad_norm": 98.15403747558594,
      "learning_rate": 4.271038007684347e-05,
      "loss": 0.05154913067817688,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0515,
      "step": 19140
    },
    {
      "entropy_step_1": 0.195,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.1947,
      "epoch": 0.5695844858867969,
      "grad_norm": 73.54793548583984,
      "learning_rate": 4.2661727964659944e-05,
      "loss": 0.0616570770740509,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0625,
      "step": 19150
    },
    {
      "entropy_step_1": 0.2163,
      "entropy_step_2": 0.2199,
      "entropy_step_3": 0.2253,
      "epoch": 0.569881919038696,
      "grad_norm": 215.12086486816406,
      "learning_rate": 4.261308295235446e-05,
      "loss": 0.06026779413223267,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0746,
      "step": 19160
    },
    {
      "entropy_step_1": 0.1827,
      "entropy_step_2": 0.1901,
      "entropy_step_3": 0.1804,
      "epoch": 0.5701793521905951,
      "grad_norm": 151.50341796875,
      "learning_rate": 4.2564445086991725e-05,
      "loss": 0.053606486320495604,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0477,
      "step": 19170
    },
    {
      "entropy_step_1": 0.1711,
      "entropy_step_2": 0.1813,
      "entropy_step_3": 0.1802,
      "epoch": 0.5704767853424942,
      "grad_norm": 77.94953155517578,
      "learning_rate": 4.251581441562954e-05,
      "loss": 0.05215710997581482,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0585,
      "step": 19180
    },
    {
      "entropy_step_1": 0.1781,
      "entropy_step_2": 0.1875,
      "entropy_step_3": 0.1707,
      "epoch": 0.5707742184943934,
      "grad_norm": 77.09612274169922,
      "learning_rate": 4.2467190985318765e-05,
      "loss": 0.05858508944511413,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0556,
      "step": 19190
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.205,
      "entropy_step_3": 0.1958,
      "epoch": 0.5710716516462925,
      "grad_norm": 290.561767578125,
      "learning_rate": 4.241857484310319e-05,
      "loss": 0.06027058362960815,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0518,
      "step": 19200
    },
    {
      "entropy_step_1": 0.2002,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.2153,
      "epoch": 0.5713690847981916,
      "grad_norm": 131.71226501464844,
      "learning_rate": 4.236996603601962e-05,
      "loss": 0.06626551151275635,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0646,
      "step": 19210
    },
    {
      "entropy_step_1": 0.215,
      "entropy_step_2": 0.2307,
      "entropy_step_3": 0.2379,
      "epoch": 0.5716665179500907,
      "grad_norm": 286.98095703125,
      "learning_rate": 4.232136461109773e-05,
      "loss": 0.06282846331596374,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.0672,
      "loss_step_3": 0.0716,
      "step": 19220
    },
    {
      "entropy_step_1": 0.2006,
      "entropy_step_2": 0.2113,
      "entropy_step_3": 0.2096,
      "epoch": 0.5719639511019898,
      "grad_norm": 87.31452941894531,
      "learning_rate": 4.2272770615360055e-05,
      "loss": 0.06071043014526367,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.061,
      "step": 19230
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.1914,
      "entropy_step_3": 0.1726,
      "epoch": 0.572261384253889,
      "grad_norm": 165.22023010253906,
      "learning_rate": 4.222418409582198e-05,
      "loss": 0.06202999353408813,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0625,
      "step": 19240
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.2041,
      "epoch": 0.5725588174057881,
      "grad_norm": 126.24760437011719,
      "learning_rate": 4.2175605099491586e-05,
      "loss": 0.06675829887390136,
      "loss_step_1": 0.0737,
      "loss_step_2": 0.0722,
      "loss_step_3": 0.0695,
      "step": 19250
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.2119,
      "entropy_step_3": 0.2216,
      "epoch": 0.5728562505576872,
      "grad_norm": 250.09970092773438,
      "learning_rate": 4.212703367336971e-05,
      "loss": 0.05599821805953979,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0583,
      "step": 19260
    },
    {
      "entropy_step_1": 0.196,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.2033,
      "epoch": 0.5731536837095863,
      "grad_norm": 133.1664276123047,
      "learning_rate": 4.2078469864449885e-05,
      "loss": 0.06178041696548462,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0598,
      "step": 19270
    },
    {
      "entropy_step_1": 0.1754,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.19,
      "epoch": 0.5734511168614854,
      "grad_norm": 171.18353271484375,
      "learning_rate": 4.202991371971828e-05,
      "loss": 0.05868121385574341,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0571,
      "step": 19280
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.2082,
      "entropy_step_3": 0.2328,
      "epoch": 0.5737485500133845,
      "grad_norm": 83.9222183227539,
      "learning_rate": 4.1981365286153576e-05,
      "loss": 0.05646793842315674,
      "loss_step_1": 0.0663,
      "loss_step_2": 0.066,
      "loss_step_3": 0.0747,
      "step": 19290
    },
    {
      "entropy_step_1": 0.1991,
      "entropy_step_2": 0.1988,
      "entropy_step_3": 0.1992,
      "epoch": 0.5740459831652837,
      "grad_norm": 258.2501220703125,
      "learning_rate": 4.193282461072707e-05,
      "loss": 0.060895538330078124,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0655,
      "loss_step_3": 0.0681,
      "step": 19300
    },
    {
      "entropy_step_1": 0.2145,
      "entropy_step_2": 0.2135,
      "entropy_step_3": 0.206,
      "epoch": 0.5743434163171827,
      "grad_norm": 40.908756256103516,
      "learning_rate": 4.188429174040252e-05,
      "loss": 0.057654070854187014,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0674,
      "step": 19310
    },
    {
      "entropy_step_1": 0.1762,
      "entropy_step_2": 0.1801,
      "entropy_step_3": 0.1752,
      "epoch": 0.5746408494690818,
      "grad_norm": 52.660892486572266,
      "learning_rate": 4.183576672213615e-05,
      "loss": 0.055414503812789916,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0533,
      "step": 19320
    },
    {
      "entropy_step_1": 0.2038,
      "entropy_step_2": 0.2084,
      "entropy_step_3": 0.2,
      "epoch": 0.5749382826209809,
      "grad_norm": 91.29291534423828,
      "learning_rate": 4.1787249602876574e-05,
      "loss": 0.06500672698020935,
      "loss_step_1": 0.0713,
      "loss_step_2": 0.0702,
      "loss_step_3": 0.0633,
      "step": 19330
    },
    {
      "entropy_step_1": 0.1953,
      "entropy_step_2": 0.1956,
      "entropy_step_3": 0.2027,
      "epoch": 0.57523571577288,
      "grad_norm": 422.7021179199219,
      "learning_rate": 4.173874042956474e-05,
      "loss": 0.06309418082237243,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0585,
      "step": 19340
    },
    {
      "entropy_step_1": 0.2029,
      "entropy_step_2": 0.2047,
      "entropy_step_3": 0.2179,
      "epoch": 0.5755331489247791,
      "grad_norm": 54.573360443115234,
      "learning_rate": 4.169023924913397e-05,
      "loss": 0.05805949568748474,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0636,
      "step": 19350
    },
    {
      "entropy_step_1": 0.2069,
      "entropy_step_2": 0.2082,
      "entropy_step_3": 0.1941,
      "epoch": 0.5758305820766783,
      "grad_norm": 159.25155639648438,
      "learning_rate": 4.164174610850977e-05,
      "loss": 0.07164167761802673,
      "loss_step_1": 0.0715,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0526,
      "step": 19360
    },
    {
      "entropy_step_1": 0.2082,
      "entropy_step_2": 0.204,
      "entropy_step_3": 0.2029,
      "epoch": 0.5761280152285774,
      "grad_norm": 76.85353088378906,
      "learning_rate": 4.1593261054609956e-05,
      "loss": 0.050719863176345824,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.0577,
      "step": 19370
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.2031,
      "entropy_step_3": 0.1903,
      "epoch": 0.5764254483804765,
      "grad_norm": 174.00567626953125,
      "learning_rate": 4.154478413434448e-05,
      "loss": 0.0632291555404663,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0561,
      "step": 19380
    },
    {
      "entropy_step_1": 0.2025,
      "entropy_step_2": 0.2027,
      "entropy_step_3": 0.2078,
      "epoch": 0.5767228815323756,
      "grad_norm": 88.86378479003906,
      "learning_rate": 4.149631539461538e-05,
      "loss": 0.06285106539726257,
      "loss_step_1": 0.0666,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0765,
      "step": 19390
    },
    {
      "entropy_step_1": 0.2006,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.21,
      "epoch": 0.5770203146842747,
      "grad_norm": 84.48973083496094,
      "learning_rate": 4.1447854882316844e-05,
      "loss": 0.058304572105407716,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0661,
      "step": 19400
    },
    {
      "entropy_step_1": 0.1926,
      "entropy_step_2": 0.1929,
      "entropy_step_3": 0.192,
      "epoch": 0.5773177478361738,
      "grad_norm": 68.50330352783203,
      "learning_rate": 4.1399402644335084e-05,
      "loss": 0.0532630443572998,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0509,
      "step": 19410
    },
    {
      "entropy_step_1": 0.1867,
      "entropy_step_2": 0.1886,
      "entropy_step_3": 0.1744,
      "epoch": 0.577615180988073,
      "grad_norm": 207.48118591308594,
      "learning_rate": 4.1350958727548314e-05,
      "loss": 0.06025764346122742,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0548,
      "step": 19420
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1921,
      "entropy_step_3": 0.2074,
      "epoch": 0.5779126141399721,
      "grad_norm": 105.81290435791016,
      "learning_rate": 4.1302523178826635e-05,
      "loss": 0.0665053129196167,
      "loss_step_1": 0.0672,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0685,
      "step": 19430
    },
    {
      "entropy_step_1": 0.2423,
      "entropy_step_2": 0.2434,
      "entropy_step_3": 0.2396,
      "epoch": 0.5782100472918712,
      "grad_norm": 145.30247497558594,
      "learning_rate": 4.125409604503213e-05,
      "loss": 0.06735702753067016,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0681,
      "loss_step_3": 0.0725,
      "step": 19440
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1837,
      "entropy_step_3": 0.1806,
      "epoch": 0.5785074804437703,
      "grad_norm": 69.32466125488281,
      "learning_rate": 4.12056773730187e-05,
      "loss": 0.05405907034873962,
      "loss_step_1": 0.0487,
      "loss_step_2": 0.0481,
      "loss_step_3": 0.048,
      "step": 19450
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1824,
      "entropy_step_3": 0.1943,
      "epoch": 0.5788049135956693,
      "grad_norm": 195.7584228515625,
      "learning_rate": 4.115726720963209e-05,
      "loss": 0.06122941970825195,
      "loss_step_1": 0.0608,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0661,
      "step": 19460
    },
    {
      "entropy_step_1": 0.1889,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.1913,
      "epoch": 0.5791023467475684,
      "grad_norm": 70.49547576904297,
      "learning_rate": 4.110886560170979e-05,
      "loss": 0.06371496915817261,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0558,
      "step": 19470
    },
    {
      "entropy_step_1": 0.1885,
      "entropy_step_2": 0.1896,
      "entropy_step_3": 0.1926,
      "epoch": 0.5793997798994676,
      "grad_norm": 443.20086669921875,
      "learning_rate": 4.106047259608098e-05,
      "loss": 0.055362391471862796,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0529,
      "step": 19480
    },
    {
      "entropy_step_1": 0.1796,
      "entropy_step_2": 0.1813,
      "entropy_step_3": 0.1691,
      "epoch": 0.5796972130513667,
      "grad_norm": 93.3935546875,
      "learning_rate": 4.101208823956657e-05,
      "loss": 0.050969105958938596,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0455,
      "step": 19490
    },
    {
      "entropy_step_1": 0.1671,
      "entropy_step_2": 0.1682,
      "entropy_step_3": 0.1679,
      "epoch": 0.5799946462032658,
      "grad_norm": 86.32384490966797,
      "learning_rate": 4.096371257897911e-05,
      "loss": 0.05702660083770752,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0562,
      "step": 19500
    },
    {
      "entropy_step_1": 0.1751,
      "entropy_step_2": 0.1762,
      "entropy_step_3": 0.1735,
      "epoch": 0.5802920793551649,
      "grad_norm": 610.5455322265625,
      "learning_rate": 4.0915345661122674e-05,
      "loss": 0.05549699068069458,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0494,
      "step": 19510
    },
    {
      "entropy_step_1": 0.1976,
      "entropy_step_2": 0.1984,
      "entropy_step_3": 0.1827,
      "epoch": 0.580589512507064,
      "grad_norm": 35.902530670166016,
      "learning_rate": 4.0866987532792924e-05,
      "loss": 0.06157469153404236,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0528,
      "step": 19520
    },
    {
      "entropy_step_1": 0.2082,
      "entropy_step_2": 0.2106,
      "entropy_step_3": 0.2091,
      "epoch": 0.5808869456589631,
      "grad_norm": 98.23756408691406,
      "learning_rate": 4.0818638240777016e-05,
      "loss": 0.06535804867744446,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0611,
      "step": 19530
    },
    {
      "entropy_step_1": 0.2125,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.217,
      "epoch": 0.5811843788108623,
      "grad_norm": 69.98674774169922,
      "learning_rate": 4.0770297831853524e-05,
      "loss": 0.06847931742668152,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0665,
      "loss_step_3": 0.0683,
      "step": 19540
    },
    {
      "entropy_step_1": 0.1926,
      "entropy_step_2": 0.1896,
      "entropy_step_3": 0.1896,
      "epoch": 0.5814818119627614,
      "grad_norm": 172.89317321777344,
      "learning_rate": 4.0721966352792476e-05,
      "loss": 0.05691440105438232,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0553,
      "step": 19550
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.1929,
      "entropy_step_3": 0.2031,
      "epoch": 0.5817792451146605,
      "grad_norm": 400.0040283203125,
      "learning_rate": 4.067364385035524e-05,
      "loss": 0.06905391216278076,
      "loss_step_1": 0.0684,
      "loss_step_2": 0.0686,
      "loss_step_3": 0.0792,
      "step": 19560
    },
    {
      "entropy_step_1": 0.2254,
      "entropy_step_2": 0.2221,
      "entropy_step_3": 0.2409,
      "epoch": 0.5820766782665596,
      "grad_norm": 236.5614013671875,
      "learning_rate": 4.062533037129447e-05,
      "loss": 0.0630277156829834,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0573,
      "step": 19570
    },
    {
      "entropy_step_1": 0.2298,
      "entropy_step_2": 0.2316,
      "entropy_step_3": 0.2283,
      "epoch": 0.5823741114184587,
      "grad_norm": 236.5664825439453,
      "learning_rate": 4.057702596235412e-05,
      "loss": 0.06915771961212158,
      "loss_step_1": 0.0688,
      "loss_step_2": 0.0684,
      "loss_step_3": 0.0702,
      "step": 19580
    },
    {
      "entropy_step_1": 0.2132,
      "entropy_step_2": 0.2146,
      "entropy_step_3": 0.2168,
      "epoch": 0.5826715445703579,
      "grad_norm": 142.52059936523438,
      "learning_rate": 4.0528730670269374e-05,
      "loss": 0.0648705542087555,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0711,
      "step": 19590
    },
    {
      "entropy_step_1": 0.1816,
      "entropy_step_2": 0.1786,
      "entropy_step_3": 0.1846,
      "epoch": 0.582968977722257,
      "grad_norm": 44.47079849243164,
      "learning_rate": 4.0480444541766576e-05,
      "loss": 0.04832501709461212,
      "loss_step_1": 0.0459,
      "loss_step_2": 0.0451,
      "loss_step_3": 0.046,
      "step": 19600
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1829,
      "epoch": 0.583266410874156,
      "grad_norm": 201.18341064453125,
      "learning_rate": 4.043216762356324e-05,
      "loss": 0.05495594739913941,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0543,
      "step": 19610
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.1788,
      "entropy_step_3": 0.1818,
      "epoch": 0.5835638440260551,
      "grad_norm": 127.7668228149414,
      "learning_rate": 4.0383899962367896e-05,
      "loss": 0.05935842394828796,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0583,
      "step": 19620
    },
    {
      "entropy_step_1": 0.2039,
      "entropy_step_2": 0.2017,
      "entropy_step_3": 0.1871,
      "epoch": 0.5838612771779542,
      "grad_norm": 264.21636962890625,
      "learning_rate": 4.033564160488018e-05,
      "loss": 0.06832671761512757,
      "loss_step_1": 0.0662,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0643,
      "step": 19630
    },
    {
      "entropy_step_1": 0.2041,
      "entropy_step_2": 0.2002,
      "entropy_step_3": 0.1969,
      "epoch": 0.5841587103298533,
      "grad_norm": 323.26495361328125,
      "learning_rate": 4.028739259779072e-05,
      "loss": 0.05784640312194824,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.051,
      "step": 19640
    },
    {
      "entropy_step_1": 0.1799,
      "entropy_step_2": 0.1783,
      "entropy_step_3": 0.1933,
      "epoch": 0.5844561434817525,
      "grad_norm": 69.7060775756836,
      "learning_rate": 4.0239152987781107e-05,
      "loss": 0.057776767015457156,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0522,
      "step": 19650
    },
    {
      "entropy_step_1": 0.1903,
      "entropy_step_2": 0.1923,
      "entropy_step_3": 0.1859,
      "epoch": 0.5847535766336516,
      "grad_norm": 73.80256652832031,
      "learning_rate": 4.019092282152376e-05,
      "loss": 0.057708388566970824,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0512,
      "step": 19660
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1879,
      "entropy_step_3": 0.1959,
      "epoch": 0.5850510097855507,
      "grad_norm": 84.64006042480469,
      "learning_rate": 4.014270214568206e-05,
      "loss": 0.05754738450050354,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0625,
      "step": 19670
    },
    {
      "entropy_step_1": 0.2051,
      "entropy_step_2": 0.2039,
      "entropy_step_3": 0.1875,
      "epoch": 0.5853484429374498,
      "grad_norm": 70.5657730102539,
      "learning_rate": 4.009449100691017e-05,
      "loss": 0.06405075192451477,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0426,
      "step": 19680
    },
    {
      "entropy_step_1": 0.2048,
      "entropy_step_2": 0.2031,
      "entropy_step_3": 0.1982,
      "epoch": 0.5856458760893489,
      "grad_norm": 41.73463821411133,
      "learning_rate": 4.004628945185301e-05,
      "loss": 0.05987480878829956,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0513,
      "step": 19690
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.2035,
      "entropy_step_3": 0.187,
      "epoch": 0.585943309241248,
      "grad_norm": 78.6194839477539,
      "learning_rate": 3.9998097527146256e-05,
      "loss": 0.056178057193756105,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0493,
      "step": 19700
    },
    {
      "entropy_step_1": 0.1768,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.2064,
      "epoch": 0.5862407423931472,
      "grad_norm": 54.50684356689453,
      "learning_rate": 3.9949915279416236e-05,
      "loss": 0.0587524950504303,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0696,
      "step": 19710
    },
    {
      "entropy_step_1": 0.1687,
      "entropy_step_2": 0.1676,
      "entropy_step_3": 0.1558,
      "epoch": 0.5865381755450463,
      "grad_norm": 65.79401397705078,
      "learning_rate": 3.9901742755279925e-05,
      "loss": 0.05644544959068298,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0472,
      "step": 19720
    },
    {
      "entropy_step_1": 0.1912,
      "entropy_step_2": 0.1918,
      "entropy_step_3": 0.173,
      "epoch": 0.5868356086969454,
      "grad_norm": 190.9129638671875,
      "learning_rate": 3.9853580001344905e-05,
      "loss": 0.05609474182128906,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0581,
      "step": 19730
    },
    {
      "entropy_step_1": 0.2066,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.1977,
      "epoch": 0.5871330418488445,
      "grad_norm": 282.2143249511719,
      "learning_rate": 3.980542706420932e-05,
      "loss": 0.051657915115356445,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0478,
      "step": 19740
    },
    {
      "entropy_step_1": 0.196,
      "entropy_step_2": 0.196,
      "entropy_step_3": 0.1794,
      "epoch": 0.5874304750007436,
      "grad_norm": 59.66783142089844,
      "learning_rate": 3.975728399046174e-05,
      "loss": 0.0552801251411438,
      "loss_step_1": 0.0487,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0443,
      "step": 19750
    },
    {
      "entropy_step_1": 0.2072,
      "entropy_step_2": 0.2061,
      "entropy_step_3": 0.2166,
      "epoch": 0.5877279081526426,
      "grad_norm": 534.12060546875,
      "learning_rate": 3.970915082668127e-05,
      "loss": 0.061721622943878174,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.058,
      "step": 19760
    },
    {
      "entropy_step_1": 0.1598,
      "entropy_step_2": 0.1612,
      "entropy_step_3": 0.1767,
      "epoch": 0.5880253413045418,
      "grad_norm": 475.2884521484375,
      "learning_rate": 3.9661027619437416e-05,
      "loss": 0.060315316915512084,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0652,
      "step": 19770
    },
    {
      "entropy_step_1": 0.1929,
      "entropy_step_2": 0.1901,
      "entropy_step_3": 0.1846,
      "epoch": 0.5883227744564409,
      "grad_norm": 213.11660766601562,
      "learning_rate": 3.9612914415290015e-05,
      "loss": 0.07077226042747498,
      "loss_step_1": 0.07,
      "loss_step_2": 0.0699,
      "loss_step_3": 0.0567,
      "step": 19780
    },
    {
      "entropy_step_1": 0.2185,
      "entropy_step_2": 0.2126,
      "entropy_step_3": 0.2055,
      "epoch": 0.58862020760834,
      "grad_norm": 691.7734985351562,
      "learning_rate": 3.956481126078927e-05,
      "loss": 0.05642499327659607,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0556,
      "step": 19790
    },
    {
      "entropy_step_1": 0.2057,
      "entropy_step_2": 0.1999,
      "entropy_step_3": 0.2001,
      "epoch": 0.5889176407602391,
      "grad_norm": 231.9148406982422,
      "learning_rate": 3.9516718202475606e-05,
      "loss": 0.05941671133041382,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0553,
      "step": 19800
    },
    {
      "entropy_step_1": 0.1883,
      "entropy_step_2": 0.1862,
      "entropy_step_3": 0.1806,
      "epoch": 0.5892150739121382,
      "grad_norm": 84.64604949951172,
      "learning_rate": 3.9468635286879725e-05,
      "loss": 0.053077811002731325,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0499,
      "step": 19810
    },
    {
      "entropy_step_1": 0.1885,
      "entropy_step_2": 0.1832,
      "entropy_step_3": 0.1889,
      "epoch": 0.5895125070640373,
      "grad_norm": 88.53345489501953,
      "learning_rate": 3.94205625605225e-05,
      "loss": 0.06461418867111206,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0652,
      "loss_step_3": 0.0679,
      "step": 19820
    },
    {
      "entropy_step_1": 0.2177,
      "entropy_step_2": 0.2129,
      "entropy_step_3": 0.1893,
      "epoch": 0.5898099402159365,
      "grad_norm": 74.88169860839844,
      "learning_rate": 3.937250006991498e-05,
      "loss": 0.057462912797927854,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0511,
      "step": 19830
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1885,
      "entropy_step_3": 0.1981,
      "epoch": 0.5901073733678356,
      "grad_norm": 138.2755126953125,
      "learning_rate": 3.932444786155823e-05,
      "loss": 0.05926623344421387,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0498,
      "step": 19840
    },
    {
      "entropy_step_1": 0.1923,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.1773,
      "epoch": 0.5904048065197347,
      "grad_norm": 141.722412109375,
      "learning_rate": 3.927640598194343e-05,
      "loss": 0.060703355073928836,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.054,
      "step": 19850
    },
    {
      "entropy_step_1": 0.2004,
      "entropy_step_2": 0.1971,
      "entropy_step_3": 0.1901,
      "epoch": 0.5907022396716338,
      "grad_norm": 47.97555923461914,
      "learning_rate": 3.922837447755176e-05,
      "loss": 0.06165195107460022,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0586,
      "step": 19860
    },
    {
      "entropy_step_1": 0.1978,
      "entropy_step_2": 0.1951,
      "entropy_step_3": 0.1833,
      "epoch": 0.5909996728235329,
      "grad_norm": 116.88706970214844,
      "learning_rate": 3.918035339485436e-05,
      "loss": 0.05634922385215759,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0471,
      "step": 19870
    },
    {
      "entropy_step_1": 0.1741,
      "entropy_step_2": 0.173,
      "entropy_step_3": 0.1557,
      "epoch": 0.5912971059754321,
      "grad_norm": 95.19502258300781,
      "learning_rate": 3.913234278031228e-05,
      "loss": 0.06430634260177612,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0601,
      "step": 19880
    },
    {
      "entropy_step_1": 0.2003,
      "entropy_step_2": 0.1986,
      "entropy_step_3": 0.1925,
      "epoch": 0.5915945391273312,
      "grad_norm": 101.61078643798828,
      "learning_rate": 3.908434268037642e-05,
      "loss": 0.05256163477897644,
      "loss_step_1": 0.0508,
      "loss_step_2": 0.0505,
      "loss_step_3": 0.0435,
      "step": 19890
    },
    {
      "entropy_step_1": 0.2045,
      "entropy_step_2": 0.2036,
      "entropy_step_3": 0.1993,
      "epoch": 0.5918919722792303,
      "grad_norm": 314.7499694824219,
      "learning_rate": 3.903635314148756e-05,
      "loss": 0.06258108019828797,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0571,
      "step": 19900
    },
    {
      "entropy_step_1": 0.1715,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1769,
      "epoch": 0.5921894054311293,
      "grad_norm": 40.80882263183594,
      "learning_rate": 3.89883742100762e-05,
      "loss": 0.05359824895858765,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0536,
      "step": 19910
    },
    {
      "entropy_step_1": 0.2244,
      "entropy_step_2": 0.2236,
      "entropy_step_3": 0.2263,
      "epoch": 0.5924868385830284,
      "grad_norm": 67.14085388183594,
      "learning_rate": 3.894040593256263e-05,
      "loss": 0.06736806631088257,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0711,
      "step": 19920
    },
    {
      "entropy_step_1": 0.2095,
      "entropy_step_2": 0.21,
      "entropy_step_3": 0.2048,
      "epoch": 0.5927842717349275,
      "grad_norm": 120.39933776855469,
      "learning_rate": 3.889244835535683e-05,
      "loss": 0.06006978154182434,
      "loss_step_1": 0.0614,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0531,
      "step": 19930
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1905,
      "entropy_step_3": 0.1877,
      "epoch": 0.5930817048868267,
      "grad_norm": 92.28241729736328,
      "learning_rate": 3.884450152485837e-05,
      "loss": 0.05731523633003235,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.0525,
      "step": 19940
    },
    {
      "entropy_step_1": 0.2031,
      "entropy_step_2": 0.2048,
      "entropy_step_3": 0.2166,
      "epoch": 0.5933791380387258,
      "grad_norm": 323.92431640625,
      "learning_rate": 3.879656548745647e-05,
      "loss": 0.054384803771972655,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0562,
      "step": 19950
    },
    {
      "entropy_step_1": 0.2125,
      "entropy_step_2": 0.2163,
      "entropy_step_3": 0.2265,
      "epoch": 0.5936765711906249,
      "grad_norm": 53.86293029785156,
      "learning_rate": 3.874864028952991e-05,
      "loss": 0.060738992691040036,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0561,
      "step": 19960
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1963,
      "entropy_step_3": 0.2016,
      "epoch": 0.593974004342524,
      "grad_norm": 60.39168930053711,
      "learning_rate": 3.870072597744698e-05,
      "loss": 0.055077725648880006,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0523,
      "step": 19970
    },
    {
      "entropy_step_1": 0.2215,
      "entropy_step_2": 0.2249,
      "entropy_step_3": 0.2199,
      "epoch": 0.5942714374944231,
      "grad_norm": 288.5180358886719,
      "learning_rate": 3.865282259756541e-05,
      "loss": 0.05956679582595825,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.065,
      "loss_step_3": 0.0745,
      "step": 19980
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1871,
      "entropy_step_3": 0.1885,
      "epoch": 0.5945688706463222,
      "grad_norm": 199.3173828125,
      "learning_rate": 3.860493019623236e-05,
      "loss": 0.055958354473114015,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0576,
      "step": 19990
    },
    {
      "entropy_step_1": 0.1952,
      "entropy_step_2": 0.1984,
      "entropy_step_3": 0.2008,
      "epoch": 0.5948663037982214,
      "grad_norm": 238.18702697753906,
      "learning_rate": 3.8557048819784414e-05,
      "loss": 0.058176225423812865,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0538,
      "step": 20000
    },
    {
      "epoch": 0.5948663037982214,
      "eval_accuracy": 0.89,
      "eval_f1": 0.88965395480226,
      "eval_loss": 0.10713086277246475,
      "eval_precision": 0.8949543068248104,
      "eval_recall": 0.89,
      "eval_runtime": 6.6649,
      "eval_samples_per_second": 75.02,
      "eval_steps_per_second": 9.453,
      "step": 20000
    },
    {
      "entropy_step_1": 0.1897,
      "entropy_step_2": 0.1926,
      "entropy_step_3": 0.1836,
      "epoch": 0.5951637369501205,
      "grad_norm": 90.50101470947266,
      "learning_rate": 3.850917851454743e-05,
      "loss": 0.06028162240982056,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0611,
      "step": 20010
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.1849,
      "epoch": 0.5954611701020196,
      "grad_norm": 209.38514709472656,
      "learning_rate": 3.846131932683661e-05,
      "loss": 0.05738074779510498,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0508,
      "step": 20020
    },
    {
      "entropy_step_1": 0.1842,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.2047,
      "epoch": 0.5957586032539187,
      "grad_norm": 55.43560028076172,
      "learning_rate": 3.841347130295633e-05,
      "loss": 0.05602952837944031,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.06,
      "step": 20030
    },
    {
      "entropy_step_1": 0.2174,
      "entropy_step_2": 0.216,
      "entropy_step_3": 0.2232,
      "epoch": 0.5960560364058178,
      "grad_norm": 144.7454833984375,
      "learning_rate": 3.8365634489200217e-05,
      "loss": 0.06809983849525451,
      "loss_step_1": 0.0705,
      "loss_step_2": 0.0694,
      "loss_step_3": 0.0711,
      "step": 20040
    },
    {
      "entropy_step_1": 0.1985,
      "entropy_step_2": 0.1977,
      "entropy_step_3": 0.216,
      "epoch": 0.596353469557717,
      "grad_norm": 64.54296875,
      "learning_rate": 3.831780893185105e-05,
      "loss": 0.05924052596092224,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.066,
      "step": 20050
    },
    {
      "entropy_step_1": 0.2083,
      "entropy_step_2": 0.2094,
      "entropy_step_3": 0.1853,
      "epoch": 0.596650902709616,
      "grad_norm": 51.23402404785156,
      "learning_rate": 3.826999467718071e-05,
      "loss": 0.05770724415779114,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.059,
      "step": 20060
    },
    {
      "entropy_step_1": 0.1622,
      "entropy_step_2": 0.1662,
      "entropy_step_3": 0.1613,
      "epoch": 0.5969483358615151,
      "grad_norm": 102.76099395751953,
      "learning_rate": 3.8222191771450114e-05,
      "loss": 0.051210612058639526,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0522,
      "loss_step_3": 0.0443,
      "step": 20070
    },
    {
      "entropy_step_1": 0.2092,
      "entropy_step_2": 0.2139,
      "entropy_step_3": 0.231,
      "epoch": 0.5972457690134142,
      "grad_norm": 65.28624725341797,
      "learning_rate": 3.817440026090925e-05,
      "loss": 0.06437631249427796,
      "loss_step_1": 0.0658,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0672,
      "step": 20080
    },
    {
      "entropy_step_1": 0.2207,
      "entropy_step_2": 0.2193,
      "entropy_step_3": 0.2228,
      "epoch": 0.5975432021653133,
      "grad_norm": 108.28068542480469,
      "learning_rate": 3.812662019179704e-05,
      "loss": 0.059273332357406616,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0637,
      "step": 20090
    },
    {
      "entropy_step_1": 0.2144,
      "entropy_step_2": 0.2109,
      "entropy_step_3": 0.2056,
      "epoch": 0.5978406353172124,
      "grad_norm": 262.4874267578125,
      "learning_rate": 3.807885161034135e-05,
      "loss": 0.06277346611022949,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0605,
      "step": 20100
    },
    {
      "entropy_step_1": 0.1841,
      "entropy_step_2": 0.1816,
      "entropy_step_3": 0.1803,
      "epoch": 0.5981380684691116,
      "grad_norm": 144.32501220703125,
      "learning_rate": 3.803109456275896e-05,
      "loss": 0.06755815744400025,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0678,
      "step": 20110
    },
    {
      "entropy_step_1": 0.2114,
      "entropy_step_2": 0.2027,
      "entropy_step_3": 0.2409,
      "epoch": 0.5984355016210107,
      "grad_norm": 98.80068969726562,
      "learning_rate": 3.798334909525542e-05,
      "loss": 0.06192567348480225,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0705,
      "step": 20120
    },
    {
      "entropy_step_1": 0.2297,
      "entropy_step_2": 0.2156,
      "entropy_step_3": 0.2201,
      "epoch": 0.5987329347729098,
      "grad_norm": 104.43498229980469,
      "learning_rate": 3.7935615254025144e-05,
      "loss": 0.057291990518569945,
      "loss_step_1": 0.0622,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0649,
      "step": 20130
    },
    {
      "entropy_step_1": 0.1971,
      "entropy_step_2": 0.1852,
      "entropy_step_3": 0.1642,
      "epoch": 0.5990303679248089,
      "grad_norm": 214.98098754882812,
      "learning_rate": 3.7887893085251265e-05,
      "loss": 0.0597685694694519,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0427,
      "step": 20140
    },
    {
      "entropy_step_1": 0.1974,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1911,
      "epoch": 0.599327801076708,
      "grad_norm": 102.25267791748047,
      "learning_rate": 3.784018263510564e-05,
      "loss": 0.056016397476196286,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.058,
      "step": 20150
    },
    {
      "entropy_step_1": 0.2001,
      "entropy_step_2": 0.1948,
      "entropy_step_3": 0.1834,
      "epoch": 0.5996252342286071,
      "grad_norm": 110.49813842773438,
      "learning_rate": 3.7792483949748785e-05,
      "loss": 0.05750487446784973,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0507,
      "step": 20160
    },
    {
      "entropy_step_1": 0.1935,
      "entropy_step_2": 0.1881,
      "entropy_step_3": 0.1903,
      "epoch": 0.5999226673805063,
      "grad_norm": 51.393157958984375,
      "learning_rate": 3.77447970753298e-05,
      "loss": 0.052628624439239505,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0544,
      "step": 20170
    },
    {
      "entropy_step_1": 0.187,
      "entropy_step_2": 0.1845,
      "entropy_step_3": 0.1988,
      "epoch": 0.6002201005324054,
      "grad_norm": 64.56586456298828,
      "learning_rate": 3.76971220579864e-05,
      "loss": 0.06282070279121399,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0592,
      "loss_step_3": 0.063,
      "step": 20180
    },
    {
      "entropy_step_1": 0.1895,
      "entropy_step_2": 0.1874,
      "entropy_step_3": 0.1867,
      "epoch": 0.6005175336843045,
      "grad_norm": 34.86843490600586,
      "learning_rate": 3.764945894384482e-05,
      "loss": 0.05437226295471191,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0501,
      "step": 20190
    },
    {
      "entropy_step_1": 0.2013,
      "entropy_step_2": 0.2014,
      "entropy_step_3": 0.2181,
      "epoch": 0.6008149668362036,
      "grad_norm": 157.62254333496094,
      "learning_rate": 3.760180777901978e-05,
      "loss": 0.05670430660247803,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.066,
      "step": 20200
    },
    {
      "entropy_step_1": 0.1913,
      "entropy_step_2": 0.1947,
      "entropy_step_3": 0.1924,
      "epoch": 0.6011123999881026,
      "grad_norm": 134.87643432617188,
      "learning_rate": 3.7554168609614396e-05,
      "loss": 0.060536956787109374,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0586,
      "step": 20210
    },
    {
      "entropy_step_1": 0.2124,
      "entropy_step_2": 0.2173,
      "entropy_step_3": 0.2328,
      "epoch": 0.6014098331400017,
      "grad_norm": 167.46331787109375,
      "learning_rate": 3.750654148172024e-05,
      "loss": 0.059992605447769166,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0591,
      "step": 20220
    },
    {
      "entropy_step_1": 0.1899,
      "entropy_step_2": 0.197,
      "entropy_step_3": 0.2068,
      "epoch": 0.6017072662919009,
      "grad_norm": 96.23123931884766,
      "learning_rate": 3.74589264414172e-05,
      "loss": 0.055201733112335206,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0655,
      "step": 20230
    },
    {
      "entropy_step_1": 0.1746,
      "entropy_step_2": 0.183,
      "entropy_step_3": 0.1923,
      "epoch": 0.6020046994438,
      "grad_norm": 90.05731964111328,
      "learning_rate": 3.741132353477346e-05,
      "loss": 0.05750715732574463,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0523,
      "step": 20240
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.214,
      "entropy_step_3": 0.2203,
      "epoch": 0.6023021325956991,
      "grad_norm": 127.72913360595703,
      "learning_rate": 3.736373280784551e-05,
      "loss": 0.06389034390449524,
      "loss_step_1": 0.0715,
      "loss_step_2": 0.0718,
      "loss_step_3": 0.0842,
      "step": 20250
    },
    {
      "entropy_step_1": 0.2078,
      "entropy_step_2": 0.2168,
      "entropy_step_3": 0.236,
      "epoch": 0.6025995657475982,
      "grad_norm": 39.351261138916016,
      "learning_rate": 3.731615430667801e-05,
      "loss": 0.06076964735984802,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0801,
      "step": 20260
    },
    {
      "entropy_step_1": 0.211,
      "entropy_step_2": 0.2208,
      "entropy_step_3": 0.2511,
      "epoch": 0.6028969988994973,
      "grad_norm": 184.3130340576172,
      "learning_rate": 3.726858807730379e-05,
      "loss": 0.06318485736846924,
      "loss_step_1": 0.0665,
      "loss_step_2": 0.0653,
      "loss_step_3": 0.0885,
      "step": 20270
    },
    {
      "entropy_step_1": 0.1978,
      "entropy_step_2": 0.2083,
      "entropy_step_3": 0.2103,
      "epoch": 0.6031944320513964,
      "grad_norm": 87.05314636230469,
      "learning_rate": 3.722103416574385e-05,
      "loss": 0.06244376301765442,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0577,
      "step": 20280
    },
    {
      "entropy_step_1": 0.2004,
      "entropy_step_2": 0.2146,
      "entropy_step_3": 0.2126,
      "epoch": 0.6034918652032956,
      "grad_norm": 484.98388671875,
      "learning_rate": 3.7173492618007257e-05,
      "loss": 0.062928307056427,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0556,
      "step": 20290
    },
    {
      "entropy_step_1": 0.2116,
      "entropy_step_2": 0.2233,
      "entropy_step_3": 0.2363,
      "epoch": 0.6037892983551947,
      "grad_norm": 155.38076782226562,
      "learning_rate": 3.712596348009107e-05,
      "loss": 0.06012246012687683,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.064,
      "step": 20300
    },
    {
      "entropy_step_1": 0.2116,
      "entropy_step_2": 0.2217,
      "entropy_step_3": 0.2182,
      "epoch": 0.6040867315070938,
      "grad_norm": 109.75410461425781,
      "learning_rate": 3.707844679798038e-05,
      "loss": 0.06164546608924866,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.058,
      "step": 20310
    },
    {
      "entropy_step_1": 0.1634,
      "entropy_step_2": 0.1734,
      "entropy_step_3": 0.1874,
      "epoch": 0.6043841646589929,
      "grad_norm": 174.4270477294922,
      "learning_rate": 3.703094261764824e-05,
      "loss": 0.055745869874954224,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0638,
      "step": 20320
    },
    {
      "entropy_step_1": 0.1823,
      "entropy_step_2": 0.1894,
      "entropy_step_3": 0.1793,
      "epoch": 0.604681597810892,
      "grad_norm": 68.77952575683594,
      "learning_rate": 3.698345098505559e-05,
      "loss": 0.05733895897865295,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0458,
      "step": 20330
    },
    {
      "entropy_step_1": 0.1699,
      "entropy_step_2": 0.1757,
      "entropy_step_3": 0.1603,
      "epoch": 0.6049790309627912,
      "grad_norm": 55.1526985168457,
      "learning_rate": 3.6935971946151246e-05,
      "loss": 0.05401483178138733,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0514,
      "step": 20340
    },
    {
      "entropy_step_1": 0.1934,
      "entropy_step_2": 0.1972,
      "entropy_step_3": 0.1985,
      "epoch": 0.6052764641146903,
      "grad_norm": 67.17137908935547,
      "learning_rate": 3.688850554687179e-05,
      "loss": 0.051606875658035276,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0542,
      "step": 20350
    },
    {
      "entropy_step_1": 0.179,
      "entropy_step_2": 0.1852,
      "entropy_step_3": 0.2117,
      "epoch": 0.6055738972665894,
      "grad_norm": 79.57300567626953,
      "learning_rate": 3.684105183314162e-05,
      "loss": 0.06681877374649048,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0599,
      "step": 20360
    },
    {
      "entropy_step_1": 0.1813,
      "entropy_step_2": 0.1869,
      "entropy_step_3": 0.1758,
      "epoch": 0.6058713304184884,
      "grad_norm": 171.08255004882812,
      "learning_rate": 3.679361085087285e-05,
      "loss": 0.06387911438941955,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0551,
      "step": 20370
    },
    {
      "entropy_step_1": 0.2206,
      "entropy_step_2": 0.2229,
      "entropy_step_3": 0.2229,
      "epoch": 0.6061687635703875,
      "grad_norm": 48.2006950378418,
      "learning_rate": 3.674618264596532e-05,
      "loss": 0.059476602077484134,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0661,
      "step": 20380
    },
    {
      "entropy_step_1": 0.2136,
      "entropy_step_2": 0.2152,
      "entropy_step_3": 0.2299,
      "epoch": 0.6064661967222866,
      "grad_norm": 83.37467956542969,
      "learning_rate": 3.669876726430639e-05,
      "loss": 0.05905461311340332,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0645,
      "step": 20390
    },
    {
      "entropy_step_1": 0.1796,
      "entropy_step_2": 0.1842,
      "entropy_step_3": 0.1937,
      "epoch": 0.6067636298741858,
      "grad_norm": 214.24049377441406,
      "learning_rate": 3.6651364751771124e-05,
      "loss": 0.05578776001930237,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0547,
      "step": 20400
    },
    {
      "entropy_step_1": 0.207,
      "entropy_step_2": 0.2107,
      "entropy_step_3": 0.2047,
      "epoch": 0.6070610630260849,
      "grad_norm": 151.21981811523438,
      "learning_rate": 3.66039751542221e-05,
      "loss": 0.058873385190963745,
      "loss_step_1": 0.0619,
      "loss_step_2": 0.0618,
      "loss_step_3": 0.0592,
      "step": 20410
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1973,
      "entropy_step_3": 0.1982,
      "epoch": 0.607358496177984,
      "grad_norm": 139.91375732421875,
      "learning_rate": 3.655659851750939e-05,
      "loss": 0.056081336736679074,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0521,
      "step": 20420
    },
    {
      "entropy_step_1": 0.1847,
      "entropy_step_2": 0.1914,
      "entropy_step_3": 0.1905,
      "epoch": 0.6076559293298831,
      "grad_norm": 57.60966110229492,
      "learning_rate": 3.650923488747054e-05,
      "loss": 0.06101163625717163,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0518,
      "step": 20430
    },
    {
      "entropy_step_1": 0.1713,
      "entropy_step_2": 0.1784,
      "entropy_step_3": 0.1794,
      "epoch": 0.6079533624817822,
      "grad_norm": 105.50906372070312,
      "learning_rate": 3.6461884309930514e-05,
      "loss": 0.04992647469043732,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0505,
      "loss_step_3": 0.0498,
      "step": 20440
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1953,
      "entropy_step_3": 0.2072,
      "epoch": 0.6082507956336813,
      "grad_norm": 89.37676239013672,
      "learning_rate": 3.6414546830701624e-05,
      "loss": 0.07068484425544738,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0632,
      "step": 20450
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.1951,
      "entropy_step_3": 0.1854,
      "epoch": 0.6085482287855805,
      "grad_norm": 44.6318244934082,
      "learning_rate": 3.636722249558353e-05,
      "loss": 0.05641216039657593,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0517,
      "step": 20460
    },
    {
      "entropy_step_1": 0.1876,
      "entropy_step_2": 0.192,
      "entropy_step_3": 0.1907,
      "epoch": 0.6088456619374796,
      "grad_norm": 170.93826293945312,
      "learning_rate": 3.631991135036316e-05,
      "loss": 0.06240773797035217,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0558,
      "step": 20470
    },
    {
      "entropy_step_1": 0.183,
      "entropy_step_2": 0.1834,
      "entropy_step_3": 0.178,
      "epoch": 0.6091430950893787,
      "grad_norm": 369.0099182128906,
      "learning_rate": 3.6272613440814734e-05,
      "loss": 0.05333451628684997,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0608,
      "step": 20480
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2025,
      "entropy_step_3": 0.1945,
      "epoch": 0.6094405282412778,
      "grad_norm": 95.69799041748047,
      "learning_rate": 3.622532881269957e-05,
      "loss": 0.05996165871620178,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0488,
      "step": 20490
    },
    {
      "entropy_step_1": 0.1739,
      "entropy_step_2": 0.1695,
      "entropy_step_3": 0.1558,
      "epoch": 0.6097379613931769,
      "grad_norm": 147.04437255859375,
      "learning_rate": 3.617805751176621e-05,
      "loss": 0.052134865522384645,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.049,
      "step": 20500
    },
    {
      "entropy_step_1": 0.1684,
      "entropy_step_2": 0.1679,
      "entropy_step_3": 0.1686,
      "epoch": 0.610035394545076,
      "grad_norm": 43.61548614501953,
      "learning_rate": 3.6130799583750286e-05,
      "loss": 0.05639688372611999,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0444,
      "step": 20510
    },
    {
      "entropy_step_1": 0.1828,
      "entropy_step_2": 0.1827,
      "entropy_step_3": 0.186,
      "epoch": 0.610332827696975,
      "grad_norm": 80.57930755615234,
      "learning_rate": 3.60835550743745e-05,
      "loss": 0.05439964532852173,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.052,
      "loss_step_3": 0.052,
      "step": 20520
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.1945,
      "entropy_step_3": 0.2082,
      "epoch": 0.6106302608488742,
      "grad_norm": 143.45814514160156,
      "learning_rate": 3.603632402934852e-05,
      "loss": 0.05909920334815979,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0563,
      "step": 20530
    },
    {
      "entropy_step_1": 0.2258,
      "entropy_step_2": 0.2236,
      "entropy_step_3": 0.2186,
      "epoch": 0.6109276940007733,
      "grad_norm": 108.9649658203125,
      "learning_rate": 3.598910649436904e-05,
      "loss": 0.05979558825492859,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.0662,
      "step": 20540
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.2165,
      "epoch": 0.6112251271526724,
      "grad_norm": 64.23970031738281,
      "learning_rate": 3.594190251511968e-05,
      "loss": 0.05972906351089478,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0641,
      "step": 20550
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.1999,
      "epoch": 0.6115225603045715,
      "grad_norm": 118.84539794921875,
      "learning_rate": 3.5894712137270925e-05,
      "loss": 0.05408384203910828,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0535,
      "step": 20560
    },
    {
      "entropy_step_1": 0.1882,
      "entropy_step_2": 0.189,
      "entropy_step_3": 0.1954,
      "epoch": 0.6118199934564706,
      "grad_norm": 104.80683898925781,
      "learning_rate": 3.584753540648012e-05,
      "loss": 0.05828939080238342,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.065,
      "loss_step_3": 0.0748,
      "step": 20570
    },
    {
      "entropy_step_1": 0.2131,
      "entropy_step_2": 0.2108,
      "entropy_step_3": 0.2139,
      "epoch": 0.6121174266083698,
      "grad_norm": 137.74700927734375,
      "learning_rate": 3.5800372368391367e-05,
      "loss": 0.06259790062904358,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.061,
      "step": 20580
    },
    {
      "entropy_step_1": 0.209,
      "entropy_step_2": 0.2023,
      "entropy_step_3": 0.2052,
      "epoch": 0.6124148597602689,
      "grad_norm": 75.40719604492188,
      "learning_rate": 3.575322306863556e-05,
      "loss": 0.055827587842941284,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0581,
      "step": 20590
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1713,
      "entropy_step_3": 0.1703,
      "epoch": 0.612712292912168,
      "grad_norm": 71.12861633300781,
      "learning_rate": 3.57060875528303e-05,
      "loss": 0.05749703049659729,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0476,
      "step": 20600
    },
    {
      "entropy_step_1": 0.2041,
      "entropy_step_2": 0.2007,
      "entropy_step_3": 0.194,
      "epoch": 0.6130097260640671,
      "grad_norm": 70.67737579345703,
      "learning_rate": 3.565896586657983e-05,
      "loss": 0.0620015025138855,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0711,
      "step": 20610
    },
    {
      "entropy_step_1": 0.1887,
      "entropy_step_2": 0.185,
      "entropy_step_3": 0.1888,
      "epoch": 0.6133071592159662,
      "grad_norm": 101.38948822021484,
      "learning_rate": 3.561185805547501e-05,
      "loss": 0.055582886934280394,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0594,
      "step": 20620
    },
    {
      "entropy_step_1": 0.1983,
      "entropy_step_2": 0.1964,
      "entropy_step_3": 0.1993,
      "epoch": 0.6136045923678654,
      "grad_norm": 106.0345230102539,
      "learning_rate": 3.556476416509331e-05,
      "loss": 0.06918581128120423,
      "loss_step_1": 0.0692,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0577,
      "step": 20630
    },
    {
      "entropy_step_1": 0.1981,
      "entropy_step_2": 0.1952,
      "entropy_step_3": 0.1946,
      "epoch": 0.6139020255197645,
      "grad_norm": 91.95156860351562,
      "learning_rate": 3.551768424099869e-05,
      "loss": 0.055512112379074094,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.05,
      "step": 20640
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.1951,
      "entropy_step_3": 0.1911,
      "epoch": 0.6141994586716636,
      "grad_norm": 178.4049835205078,
      "learning_rate": 3.547061832874163e-05,
      "loss": 0.06782884001731873,
      "loss_step_1": 0.0737,
      "loss_step_2": 0.0734,
      "loss_step_3": 0.0652,
      "step": 20650
    },
    {
      "entropy_step_1": 0.1911,
      "entropy_step_2": 0.1929,
      "entropy_step_3": 0.2055,
      "epoch": 0.6144968918235627,
      "grad_norm": 105.36692810058594,
      "learning_rate": 3.542356647385906e-05,
      "loss": 0.06021566390991211,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0569,
      "step": 20660
    },
    {
      "entropy_step_1": 0.2283,
      "entropy_step_2": 0.2279,
      "entropy_step_3": 0.2356,
      "epoch": 0.6147943249754617,
      "grad_norm": 65.3862533569336,
      "learning_rate": 3.537652872187425e-05,
      "loss": 0.06829466223716736,
      "loss_step_1": 0.0724,
      "loss_step_2": 0.0715,
      "loss_step_3": 0.0666,
      "step": 20670
    },
    {
      "entropy_step_1": 0.1932,
      "entropy_step_2": 0.1974,
      "entropy_step_3": 0.1877,
      "epoch": 0.6150917581273608,
      "grad_norm": 56.22062301635742,
      "learning_rate": 3.5329505118296874e-05,
      "loss": 0.05650286078453064,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0572,
      "step": 20680
    },
    {
      "entropy_step_1": 0.1789,
      "entropy_step_2": 0.1821,
      "entropy_step_3": 0.1893,
      "epoch": 0.61538919127926,
      "grad_norm": 72.19248962402344,
      "learning_rate": 3.528249570862292e-05,
      "loss": 0.06252461075782775,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0586,
      "step": 20690
    },
    {
      "entropy_step_1": 0.1757,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.1856,
      "epoch": 0.6156866244311591,
      "grad_norm": 47.099586486816406,
      "learning_rate": 3.5235500538334634e-05,
      "loss": 0.05268696546554565,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.048,
      "loss_step_3": 0.055,
      "step": 20700
    },
    {
      "entropy_step_1": 0.1859,
      "entropy_step_2": 0.19,
      "entropy_step_3": 0.1939,
      "epoch": 0.6159840575830582,
      "grad_norm": 68.8216323852539,
      "learning_rate": 3.5188519652900486e-05,
      "loss": 0.05650825500488281,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0601,
      "step": 20710
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1914,
      "entropy_step_3": 0.1875,
      "epoch": 0.6162814907349573,
      "grad_norm": 193.31497192382812,
      "learning_rate": 3.51415530977751e-05,
      "loss": 0.06436445116996765,
      "loss_step_1": 0.0723,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0787,
      "step": 20720
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.1998,
      "entropy_step_3": 0.1997,
      "epoch": 0.6165789238868564,
      "grad_norm": 315.0943603515625,
      "learning_rate": 3.509460091839927e-05,
      "loss": 0.05260817408561706,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0534,
      "step": 20730
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.1913,
      "epoch": 0.6168763570387555,
      "grad_norm": 147.66253662109375,
      "learning_rate": 3.5047663160199875e-05,
      "loss": 0.062235665321350095,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0629,
      "step": 20740
    },
    {
      "entropy_step_1": 0.1814,
      "entropy_step_2": 0.1873,
      "entropy_step_3": 0.1762,
      "epoch": 0.6171737901906547,
      "grad_norm": 611.5122680664062,
      "learning_rate": 3.500073986858984e-05,
      "loss": 0.05608916878700256,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0406,
      "step": 20750
    },
    {
      "entropy_step_1": 0.2059,
      "entropy_step_2": 0.2111,
      "entropy_step_3": 0.2202,
      "epoch": 0.6174712233425538,
      "grad_norm": 95.51826477050781,
      "learning_rate": 3.495383108896806e-05,
      "loss": 0.05522394776344299,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0603,
      "step": 20760
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1949,
      "entropy_step_3": 0.1821,
      "epoch": 0.6177686564944529,
      "grad_norm": 61.84101486206055,
      "learning_rate": 3.490693686671943e-05,
      "loss": 0.05204648375511169,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0459,
      "step": 20770
    },
    {
      "entropy_step_1": 0.17,
      "entropy_step_2": 0.1748,
      "entropy_step_3": 0.1955,
      "epoch": 0.618066089646352,
      "grad_norm": 77.71707153320312,
      "learning_rate": 3.486005724721476e-05,
      "loss": 0.06565064191818237,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0538,
      "step": 20780
    },
    {
      "entropy_step_1": 0.1829,
      "entropy_step_2": 0.1842,
      "entropy_step_3": 0.2017,
      "epoch": 0.6183635227982511,
      "grad_norm": 72.3709945678711,
      "learning_rate": 3.481319227581071e-05,
      "loss": 0.05226080417633057,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.048,
      "loss_step_3": 0.0506,
      "step": 20790
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.2038,
      "entropy_step_3": 0.2263,
      "epoch": 0.6186609559501502,
      "grad_norm": 144.65931701660156,
      "learning_rate": 3.476634199784979e-05,
      "loss": 0.06323503851890563,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0685,
      "step": 20800
    },
    {
      "entropy_step_1": 0.1896,
      "entropy_step_2": 0.1983,
      "entropy_step_3": 0.2033,
      "epoch": 0.6189583891020494,
      "grad_norm": 45.03359603881836,
      "learning_rate": 3.4719506458660265e-05,
      "loss": 0.05984938144683838,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0573,
      "step": 20810
    },
    {
      "entropy_step_1": 0.18,
      "entropy_step_2": 0.1872,
      "entropy_step_3": 0.1719,
      "epoch": 0.6192558222539484,
      "grad_norm": 44.51306915283203,
      "learning_rate": 3.4672685703556154e-05,
      "loss": 0.05343493223190308,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0453,
      "step": 20820
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.2152,
      "entropy_step_3": 0.2221,
      "epoch": 0.6195532554058475,
      "grad_norm": 70.38258361816406,
      "learning_rate": 3.462587977783719e-05,
      "loss": 0.06062894463539124,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0578,
      "step": 20830
    },
    {
      "entropy_step_1": 0.2091,
      "entropy_step_2": 0.2151,
      "entropy_step_3": 0.2075,
      "epoch": 0.6198506885577466,
      "grad_norm": 50.32840347290039,
      "learning_rate": 3.457908872678875e-05,
      "loss": 0.05413135290145874,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0566,
      "step": 20840
    },
    {
      "entropy_step_1": 0.1926,
      "entropy_step_2": 0.1995,
      "entropy_step_3": 0.2009,
      "epoch": 0.6201481217096457,
      "grad_norm": 82.97097778320312,
      "learning_rate": 3.453231259568178e-05,
      "loss": 0.0632794976234436,
      "loss_step_1": 0.0658,
      "loss_step_2": 0.0647,
      "loss_step_3": 0.0692,
      "step": 20850
    },
    {
      "entropy_step_1": 0.1623,
      "entropy_step_2": 0.1697,
      "entropy_step_3": 0.1707,
      "epoch": 0.6204455548615448,
      "grad_norm": 798.8577880859375,
      "learning_rate": 3.448555142977284e-05,
      "loss": 0.04957353472709656,
      "loss_step_1": 0.0423,
      "loss_step_2": 0.0419,
      "loss_step_3": 0.0392,
      "step": 20860
    },
    {
      "entropy_step_1": 0.2144,
      "entropy_step_2": 0.2179,
      "entropy_step_3": 0.1983,
      "epoch": 0.620742988013444,
      "grad_norm": 129.68800354003906,
      "learning_rate": 3.4438805274304e-05,
      "loss": 0.0667620837688446,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0567,
      "step": 20870
    },
    {
      "entropy_step_1": 0.1835,
      "entropy_step_2": 0.1865,
      "entropy_step_3": 0.194,
      "epoch": 0.6210404211653431,
      "grad_norm": 68.73981475830078,
      "learning_rate": 3.4392074174502795e-05,
      "loss": 0.056096047163009644,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0649,
      "step": 20880
    },
    {
      "entropy_step_1": 0.1762,
      "entropy_step_2": 0.179,
      "entropy_step_3": 0.1849,
      "epoch": 0.6213378543172422,
      "grad_norm": 145.4446563720703,
      "learning_rate": 3.434535817558222e-05,
      "loss": 0.05253089666366577,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0481,
      "step": 20890
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1952,
      "entropy_step_3": 0.183,
      "epoch": 0.6216352874691413,
      "grad_norm": 246.09840393066406,
      "learning_rate": 3.4298657322740607e-05,
      "loss": 0.05754222869873047,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0571,
      "step": 20900
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.1837,
      "entropy_step_3": 0.1752,
      "epoch": 0.6219327206210404,
      "grad_norm": 83.83330535888672,
      "learning_rate": 3.425197166116168e-05,
      "loss": 0.05169695019721985,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0498,
      "step": 20910
    },
    {
      "entropy_step_1": 0.2128,
      "entropy_step_2": 0.2124,
      "entropy_step_3": 0.205,
      "epoch": 0.6222301537729396,
      "grad_norm": 189.83856201171875,
      "learning_rate": 3.4205301236014454e-05,
      "loss": 0.06587401032447815,
      "loss_step_1": 0.0678,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.0678,
      "step": 20920
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2089,
      "epoch": 0.6225275869248387,
      "grad_norm": 45.293190002441406,
      "learning_rate": 3.415864609245322e-05,
      "loss": 0.051925623416900636,
      "loss_step_1": 0.0476,
      "loss_step_2": 0.0469,
      "loss_step_3": 0.048,
      "step": 20930
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.198,
      "entropy_step_3": 0.1757,
      "epoch": 0.6228250200767378,
      "grad_norm": 132.44593811035156,
      "learning_rate": 3.4112006275617424e-05,
      "loss": 0.06491210460662841,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0592,
      "loss_step_3": 0.0464,
      "step": 20940
    },
    {
      "entropy_step_1": 0.2043,
      "entropy_step_2": 0.2049,
      "entropy_step_3": 0.2174,
      "epoch": 0.6231224532286369,
      "grad_norm": 259.00775146484375,
      "learning_rate": 3.406538183063175e-05,
      "loss": 0.0627710461616516,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0685,
      "step": 20950
    },
    {
      "entropy_step_1": 0.2153,
      "entropy_step_2": 0.2159,
      "entropy_step_3": 0.2236,
      "epoch": 0.623419886380536,
      "grad_norm": 46.011451721191406,
      "learning_rate": 3.401877280260598e-05,
      "loss": 0.06145049929618836,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0575,
      "step": 20960
    },
    {
      "entropy_step_1": 0.2005,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.217,
      "epoch": 0.623717319532435,
      "grad_norm": 122.15967559814453,
      "learning_rate": 3.3972179236634973e-05,
      "loss": 0.05488742589950561,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0604,
      "step": 20970
    },
    {
      "entropy_step_1": 0.1629,
      "entropy_step_2": 0.1675,
      "entropy_step_3": 0.1641,
      "epoch": 0.6240147526843342,
      "grad_norm": 69.38267517089844,
      "learning_rate": 3.392560117779865e-05,
      "loss": 0.05760624408721924,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0476,
      "step": 20980
    },
    {
      "entropy_step_1": 0.2045,
      "entropy_step_2": 0.2063,
      "entropy_step_3": 0.214,
      "epoch": 0.6243121858362333,
      "grad_norm": 265.7038269042969,
      "learning_rate": 3.3879038671161913e-05,
      "loss": 0.0591814398765564,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.048,
      "step": 20990
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1802,
      "entropy_step_3": 0.1835,
      "epoch": 0.6246096189881324,
      "grad_norm": 56.58501434326172,
      "learning_rate": 3.383249176177461e-05,
      "loss": 0.0633453905582428,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0588,
      "step": 21000
    },
    {
      "epoch": 0.6246096189881324,
      "eval_accuracy": 0.915,
      "eval_f1": 0.9149284548305124,
      "eval_loss": 0.09670388698577881,
      "eval_precision": 0.9164007721976729,
      "eval_recall": 0.915,
      "eval_runtime": 8.6805,
      "eval_samples_per_second": 57.6,
      "eval_steps_per_second": 7.258,
      "step": 21000
    },
    {
      "entropy_step_1": 0.1947,
      "entropy_step_2": 0.1915,
      "entropy_step_3": 0.1895,
      "epoch": 0.6249070521400315,
      "grad_norm": 75.06735229492188,
      "learning_rate": 3.378596049467152e-05,
      "loss": 0.058677035570144656,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0516,
      "step": 21010
    },
    {
      "entropy_step_1": 0.1656,
      "entropy_step_2": 0.1673,
      "entropy_step_3": 0.1594,
      "epoch": 0.6252044852919306,
      "grad_norm": 90.23967742919922,
      "learning_rate": 3.3739444914872275e-05,
      "loss": 0.05751475691795349,
      "loss_step_1": 0.046,
      "loss_step_2": 0.0464,
      "loss_step_3": 0.0438,
      "step": 21020
    },
    {
      "entropy_step_1": 0.1761,
      "entropy_step_2": 0.176,
      "entropy_step_3": 0.1678,
      "epoch": 0.6255019184438297,
      "grad_norm": 190.0738525390625,
      "learning_rate": 3.369294506738135e-05,
      "loss": 0.05387151837348938,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0543,
      "step": 21030
    },
    {
      "entropy_step_1": 0.1927,
      "entropy_step_2": 0.1908,
      "entropy_step_3": 0.1836,
      "epoch": 0.6257993515957289,
      "grad_norm": 32.72391128540039,
      "learning_rate": 3.364646099718794e-05,
      "loss": 0.05712136626243591,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0505,
      "step": 21040
    },
    {
      "entropy_step_1": 0.2149,
      "entropy_step_2": 0.2115,
      "entropy_step_3": 0.2222,
      "epoch": 0.626096784747628,
      "grad_norm": 89.91844940185547,
      "learning_rate": 3.359999274926604e-05,
      "loss": 0.05921645164489746,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0592,
      "loss_step_3": 0.0626,
      "step": 21050
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1815,
      "entropy_step_3": 0.1832,
      "epoch": 0.6263942178995271,
      "grad_norm": 56.89751434326172,
      "learning_rate": 3.355354036857431e-05,
      "loss": 0.05089718699455261,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0467,
      "step": 21060
    },
    {
      "entropy_step_1": 0.1739,
      "entropy_step_2": 0.1783,
      "entropy_step_3": 0.1685,
      "epoch": 0.6266916510514262,
      "grad_norm": 252.4061279296875,
      "learning_rate": 3.350710390005608e-05,
      "loss": 0.05698692202568054,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0535,
      "step": 21070
    },
    {
      "entropy_step_1": 0.2022,
      "entropy_step_2": 0.2087,
      "entropy_step_3": 0.2042,
      "epoch": 0.6269890842033253,
      "grad_norm": 72.46099090576172,
      "learning_rate": 3.3460683388639224e-05,
      "loss": 0.05885496735572815,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0589,
      "step": 21080
    },
    {
      "entropy_step_1": 0.2188,
      "entropy_step_2": 0.2267,
      "entropy_step_3": 0.2254,
      "epoch": 0.6272865173552244,
      "grad_norm": 396.13885498046875,
      "learning_rate": 3.341427887923624e-05,
      "loss": 0.06652520895004273,
      "loss_step_1": 0.0717,
      "loss_step_2": 0.0718,
      "loss_step_3": 0.0676,
      "step": 21090
    },
    {
      "entropy_step_1": 0.2024,
      "entropy_step_2": 0.2093,
      "entropy_step_3": 0.2089,
      "epoch": 0.6275839505071236,
      "grad_norm": 86.88043212890625,
      "learning_rate": 3.3367890416744115e-05,
      "loss": 0.06598552465438842,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.067,
      "loss_step_3": 0.0813,
      "step": 21100
    },
    {
      "entropy_step_1": 0.1946,
      "entropy_step_2": 0.1999,
      "entropy_step_3": 0.1885,
      "epoch": 0.6278813836590227,
      "grad_norm": 59.34747314453125,
      "learning_rate": 3.332151804604434e-05,
      "loss": 0.05634210705757141,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0505,
      "step": 21110
    },
    {
      "entropy_step_1": 0.1823,
      "entropy_step_2": 0.1861,
      "entropy_step_3": 0.1761,
      "epoch": 0.6281788168109217,
      "grad_norm": 119.59941101074219,
      "learning_rate": 3.327516181200281e-05,
      "loss": 0.055208373069763186,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0546,
      "step": 21120
    },
    {
      "entropy_step_1": 0.2172,
      "entropy_step_2": 0.219,
      "entropy_step_3": 0.1988,
      "epoch": 0.6284762499628208,
      "grad_norm": 46.19968032836914,
      "learning_rate": 3.322882175946978e-05,
      "loss": 0.06302379965782165,
      "loss_step_1": 0.065,
      "loss_step_2": 0.0642,
      "loss_step_3": 0.0616,
      "step": 21130
    },
    {
      "entropy_step_1": 0.1829,
      "entropy_step_2": 0.185,
      "entropy_step_3": 0.1583,
      "epoch": 0.6287736831147199,
      "grad_norm": 55.10160827636719,
      "learning_rate": 3.3182497933279896e-05,
      "loss": 0.051950889825820926,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0347,
      "step": 21140
    },
    {
      "entropy_step_1": 0.162,
      "entropy_step_2": 0.1679,
      "entropy_step_3": 0.1772,
      "epoch": 0.629071116266619,
      "grad_norm": 119.44499969482422,
      "learning_rate": 3.313619037825212e-05,
      "loss": 0.05540719032287598,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0494,
      "loss_step_3": 0.0525,
      "step": 21150
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.1883,
      "entropy_step_3": 0.176,
      "epoch": 0.6293685494185182,
      "grad_norm": 1011.2365112304688,
      "learning_rate": 3.308989913918962e-05,
      "loss": 0.061947232484817503,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0559,
      "step": 21160
    },
    {
      "entropy_step_1": 0.2032,
      "entropy_step_2": 0.2032,
      "entropy_step_3": 0.1933,
      "epoch": 0.6296659825704173,
      "grad_norm": 72.60134887695312,
      "learning_rate": 3.304362426087979e-05,
      "loss": 0.05800338387489319,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0596,
      "step": 21170
    },
    {
      "entropy_step_1": 0.1841,
      "entropy_step_2": 0.1851,
      "entropy_step_3": 0.1906,
      "epoch": 0.6299634157223164,
      "grad_norm": 94.76763916015625,
      "learning_rate": 3.29973657880942e-05,
      "loss": 0.058477896451950076,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0556,
      "step": 21180
    },
    {
      "entropy_step_1": 0.2025,
      "entropy_step_2": 0.2014,
      "entropy_step_3": 0.2241,
      "epoch": 0.6302608488742155,
      "grad_norm": 92.11673736572266,
      "learning_rate": 3.295112376558857e-05,
      "loss": 0.060472095012664796,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0688,
      "step": 21190
    },
    {
      "entropy_step_1": 0.1995,
      "entropy_step_2": 0.1954,
      "entropy_step_3": 0.2051,
      "epoch": 0.6305582820261146,
      "grad_norm": 132.22039794921875,
      "learning_rate": 3.2904898238102674e-05,
      "loss": 0.059990537166595456,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0528,
      "step": 21200
    },
    {
      "entropy_step_1": 0.1924,
      "entropy_step_2": 0.1881,
      "entropy_step_3": 0.1785,
      "epoch": 0.6308557151780138,
      "grad_norm": 96.62421417236328,
      "learning_rate": 3.2858689250360375e-05,
      "loss": 0.06070765256881714,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0511,
      "step": 21210
    },
    {
      "entropy_step_1": 0.2045,
      "entropy_step_2": 0.1998,
      "entropy_step_3": 0.1878,
      "epoch": 0.6311531483299129,
      "grad_norm": 52.40727996826172,
      "learning_rate": 3.281249684706944e-05,
      "loss": 0.0628304898738861,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0613,
      "step": 21220
    },
    {
      "entropy_step_1": 0.2135,
      "entropy_step_2": 0.2072,
      "entropy_step_3": 0.2088,
      "epoch": 0.631450581481812,
      "grad_norm": 138.2692108154297,
      "learning_rate": 3.276632107292167e-05,
      "loss": 0.05702245831489563,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0598,
      "step": 21230
    },
    {
      "entropy_step_1": 0.1957,
      "entropy_step_2": 0.1907,
      "entropy_step_3": 0.1787,
      "epoch": 0.6317480146337111,
      "grad_norm": 123.81265258789062,
      "learning_rate": 3.2720161972592773e-05,
      "loss": 0.062141692638397215,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0548,
      "step": 21240
    },
    {
      "entropy_step_1": 0.1919,
      "entropy_step_2": 0.1874,
      "entropy_step_3": 0.1628,
      "epoch": 0.6320454477856102,
      "grad_norm": 259.5120544433594,
      "learning_rate": 3.267401959074229e-05,
      "loss": 0.06261420249938965,
      "loss_step_1": 0.0632,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0574,
      "step": 21250
    },
    {
      "entropy_step_1": 0.1997,
      "entropy_step_2": 0.1948,
      "entropy_step_3": 0.2184,
      "epoch": 0.6323428809375093,
      "grad_norm": 282.00103759765625,
      "learning_rate": 3.262789397201362e-05,
      "loss": 0.0592460036277771,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0638,
      "step": 21260
    },
    {
      "entropy_step_1": 0.2159,
      "entropy_step_2": 0.2134,
      "entropy_step_3": 0.2145,
      "epoch": 0.6326403140894084,
      "grad_norm": 79.18666076660156,
      "learning_rate": 3.2581785161033905e-05,
      "loss": 0.05553638935089111,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0683,
      "step": 21270
    },
    {
      "entropy_step_1": 0.1596,
      "entropy_step_2": 0.1655,
      "entropy_step_3": 0.1719,
      "epoch": 0.6329377472413075,
      "grad_norm": 84.02420043945312,
      "learning_rate": 3.253569320241405e-05,
      "loss": 0.052918583154678345,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0505,
      "step": 21280
    },
    {
      "entropy_step_1": 0.1589,
      "entropy_step_2": 0.1649,
      "entropy_step_3": 0.164,
      "epoch": 0.6332351803932066,
      "grad_norm": 78.43517303466797,
      "learning_rate": 3.248961814074866e-05,
      "loss": 0.05787983536720276,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.055,
      "step": 21290
    },
    {
      "entropy_step_1": 0.232,
      "entropy_step_2": 0.2337,
      "entropy_step_3": 0.2283,
      "epoch": 0.6335326135451057,
      "grad_norm": 77.66947174072266,
      "learning_rate": 3.2443560020615995e-05,
      "loss": 0.062475192546844485,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0575,
      "step": 21300
    },
    {
      "entropy_step_1": 0.2195,
      "entropy_step_2": 0.2189,
      "entropy_step_3": 0.2212,
      "epoch": 0.6338300466970048,
      "grad_norm": 104.14019012451172,
      "learning_rate": 3.2397518886577885e-05,
      "loss": 0.057330656051635745,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0565,
      "step": 21310
    },
    {
      "entropy_step_1": 0.1983,
      "entropy_step_2": 0.1962,
      "entropy_step_3": 0.2019,
      "epoch": 0.6341274798489039,
      "grad_norm": 115.67605590820312,
      "learning_rate": 3.235149478317977e-05,
      "loss": 0.0614640474319458,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0615,
      "step": 21320
    },
    {
      "entropy_step_1": 0.1996,
      "entropy_step_2": 0.1955,
      "entropy_step_3": 0.1891,
      "epoch": 0.6344249130008031,
      "grad_norm": 126.28974914550781,
      "learning_rate": 3.23054877549506e-05,
      "loss": 0.06539899110794067,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0626,
      "step": 21330
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.2087,
      "entropy_step_3": 0.2123,
      "epoch": 0.6347223461527022,
      "grad_norm": 239.8892364501953,
      "learning_rate": 3.225949784640279e-05,
      "loss": 0.06431586146354676,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0658,
      "step": 21340
    },
    {
      "entropy_step_1": 0.2037,
      "entropy_step_2": 0.1984,
      "entropy_step_3": 0.2214,
      "epoch": 0.6350197793046013,
      "grad_norm": 217.26841735839844,
      "learning_rate": 3.2213525102032214e-05,
      "loss": 0.06481112241744995,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0768,
      "step": 21350
    },
    {
      "entropy_step_1": 0.2024,
      "entropy_step_2": 0.197,
      "entropy_step_3": 0.1843,
      "epoch": 0.6353172124565004,
      "grad_norm": 126.13460540771484,
      "learning_rate": 3.216756956631813e-05,
      "loss": 0.060137540102005005,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0553,
      "step": 21360
    },
    {
      "entropy_step_1": 0.2067,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.197,
      "epoch": 0.6356146456083995,
      "grad_norm": 56.90803527832031,
      "learning_rate": 3.212163128372312e-05,
      "loss": 0.05624587535858154,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0527,
      "loss_step_3": 0.0526,
      "step": 21370
    },
    {
      "entropy_step_1": 0.1861,
      "entropy_step_2": 0.1816,
      "entropy_step_3": 0.1803,
      "epoch": 0.6359120787602986,
      "grad_norm": 59.995914459228516,
      "learning_rate": 3.207571029869312e-05,
      "loss": 0.06555274724960328,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.057,
      "step": 21380
    },
    {
      "entropy_step_1": 0.1733,
      "entropy_step_2": 0.1684,
      "entropy_step_3": 0.1556,
      "epoch": 0.6362095119121978,
      "grad_norm": 53.239906311035156,
      "learning_rate": 3.202980665565731e-05,
      "loss": 0.0544681191444397,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0516,
      "step": 21390
    },
    {
      "entropy_step_1": 0.1857,
      "entropy_step_2": 0.1808,
      "entropy_step_3": 0.1708,
      "epoch": 0.6365069450640969,
      "grad_norm": 213.80894470214844,
      "learning_rate": 3.1983920399028065e-05,
      "loss": 0.05750279426574707,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0463,
      "step": 21400
    },
    {
      "entropy_step_1": 0.207,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.211,
      "epoch": 0.636804378215996,
      "grad_norm": 134.47213745117188,
      "learning_rate": 3.193805157320098e-05,
      "loss": 0.05550087690353393,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.062,
      "step": 21410
    },
    {
      "entropy_step_1": 0.1746,
      "entropy_step_2": 0.1677,
      "entropy_step_3": 0.1647,
      "epoch": 0.637101811367895,
      "grad_norm": 114.53059387207031,
      "learning_rate": 3.189220022255477e-05,
      "loss": 0.050685518980026247,
      "loss_step_1": 0.0458,
      "loss_step_2": 0.0453,
      "loss_step_3": 0.0479,
      "step": 21420
    },
    {
      "entropy_step_1": 0.1803,
      "entropy_step_2": 0.1756,
      "entropy_step_3": 0.1687,
      "epoch": 0.6373992445197941,
      "grad_norm": 79.10222625732422,
      "learning_rate": 3.184636639145123e-05,
      "loss": 0.050681459903717044,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0438,
      "step": 21430
    },
    {
      "entropy_step_1": 0.2046,
      "entropy_step_2": 0.1985,
      "entropy_step_3": 0.213,
      "epoch": 0.6376966776716932,
      "grad_norm": 66.77716064453125,
      "learning_rate": 3.180055012423525e-05,
      "loss": 0.05505739450454712,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0691,
      "step": 21440
    },
    {
      "entropy_step_1": 0.1779,
      "entropy_step_2": 0.1728,
      "entropy_step_3": 0.1693,
      "epoch": 0.6379941108235924,
      "grad_norm": 147.69769287109375,
      "learning_rate": 3.175475146523464e-05,
      "loss": 0.05604998469352722,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0514,
      "step": 21450
    },
    {
      "entropy_step_1": 0.2166,
      "entropy_step_2": 0.2096,
      "entropy_step_3": 0.2087,
      "epoch": 0.6382915439754915,
      "grad_norm": 124.56412506103516,
      "learning_rate": 3.170897045876024e-05,
      "loss": 0.06500259637832642,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0687,
      "loss_step_3": 0.0739,
      "step": 21460
    },
    {
      "entropy_step_1": 0.2256,
      "entropy_step_2": 0.2183,
      "entropy_step_3": 0.2185,
      "epoch": 0.6385889771273906,
      "grad_norm": 54.35158920288086,
      "learning_rate": 3.166320714910582e-05,
      "loss": 0.06325310468673706,
      "loss_step_1": 0.0648,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0593,
      "step": 21470
    },
    {
      "entropy_step_1": 0.2065,
      "entropy_step_2": 0.206,
      "entropy_step_3": 0.1993,
      "epoch": 0.6388864102792897,
      "grad_norm": 85.4570541381836,
      "learning_rate": 3.1617461580548013e-05,
      "loss": 0.06485379338264466,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0577,
      "step": 21480
    },
    {
      "entropy_step_1": 0.1806,
      "entropy_step_2": 0.1801,
      "entropy_step_3": 0.1704,
      "epoch": 0.6391838434311888,
      "grad_norm": 99.39859008789062,
      "learning_rate": 3.157173379734623e-05,
      "loss": 0.04976917803287506,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0483,
      "step": 21490
    },
    {
      "entropy_step_1": 0.1647,
      "entropy_step_2": 0.167,
      "entropy_step_3": 0.1547,
      "epoch": 0.639481276583088,
      "grad_norm": 422.6686706542969,
      "learning_rate": 3.152602384374275e-05,
      "loss": 0.05370214581489563,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0399,
      "step": 21500
    },
    {
      "entropy_step_1": 0.171,
      "entropy_step_2": 0.174,
      "entropy_step_3": 0.1614,
      "epoch": 0.6397787097349871,
      "grad_norm": 192.48678588867188,
      "learning_rate": 3.148033176396259e-05,
      "loss": 0.05400800108909607,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0545,
      "step": 21510
    },
    {
      "entropy_step_1": 0.1878,
      "entropy_step_2": 0.1923,
      "entropy_step_3": 0.1928,
      "epoch": 0.6400761428868862,
      "grad_norm": 88.21436309814453,
      "learning_rate": 3.143465760221344e-05,
      "loss": 0.05717363953590393,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0621,
      "step": 21520
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1869,
      "epoch": 0.6403735760387853,
      "grad_norm": 82.5219497680664,
      "learning_rate": 3.138900140268568e-05,
      "loss": 0.06092990636825561,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0644,
      "loss_step_3": 0.0745,
      "step": 21530
    },
    {
      "entropy_step_1": 0.2158,
      "entropy_step_2": 0.2252,
      "entropy_step_3": 0.2292,
      "epoch": 0.6406710091906844,
      "grad_norm": 102.17415618896484,
      "learning_rate": 3.13433632095523e-05,
      "loss": 0.05491744875907898,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0653,
      "step": 21540
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1853,
      "entropy_step_3": 0.1872,
      "epoch": 0.6409684423425835,
      "grad_norm": 212.56907653808594,
      "learning_rate": 3.129774306696887e-05,
      "loss": 0.05946866273880005,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0587,
      "step": 21550
    },
    {
      "entropy_step_1": 0.19,
      "entropy_step_2": 0.1917,
      "entropy_step_3": 0.196,
      "epoch": 0.6412658754944827,
      "grad_norm": 85.45793151855469,
      "learning_rate": 3.12521410190735e-05,
      "loss": 0.05263943076133728,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0428,
      "step": 21560
    },
    {
      "entropy_step_1": 0.2017,
      "entropy_step_2": 0.2031,
      "entropy_step_3": 0.2257,
      "epoch": 0.6415633086463818,
      "grad_norm": 41.855079650878906,
      "learning_rate": 3.12065571099868e-05,
      "loss": 0.05566168427467346,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0673,
      "step": 21570
    },
    {
      "entropy_step_1": 0.2094,
      "entropy_step_2": 0.2112,
      "entropy_step_3": 0.2129,
      "epoch": 0.6418607417982808,
      "grad_norm": 92.29579162597656,
      "learning_rate": 3.1160991383811835e-05,
      "loss": 0.05982135534286499,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0609,
      "step": 21580
    },
    {
      "entropy_step_1": 0.1758,
      "entropy_step_2": 0.1805,
      "entropy_step_3": 0.1812,
      "epoch": 0.6421581749501799,
      "grad_norm": 198.42469787597656,
      "learning_rate": 3.1115443884634035e-05,
      "loss": 0.0616804301738739,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0628,
      "step": 21590
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2051,
      "entropy_step_3": 0.236,
      "epoch": 0.642455608102079,
      "grad_norm": 137.30938720703125,
      "learning_rate": 3.1069914656521235e-05,
      "loss": 0.06285324096679687,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0625,
      "step": 21600
    },
    {
      "entropy_step_1": 0.1951,
      "entropy_step_2": 0.1887,
      "entropy_step_3": 0.1705,
      "epoch": 0.6427530412539781,
      "grad_norm": 77.50193786621094,
      "learning_rate": 3.102440374352358e-05,
      "loss": 0.05756099224090576,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0549,
      "step": 21610
    },
    {
      "entropy_step_1": 0.1953,
      "entropy_step_2": 0.1911,
      "entropy_step_3": 0.1749,
      "epoch": 0.6430504744058773,
      "grad_norm": 137.0287322998047,
      "learning_rate": 3.0978911189673526e-05,
      "loss": 0.053158634901046754,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0437,
      "step": 21620
    },
    {
      "entropy_step_1": 0.2052,
      "entropy_step_2": 0.2035,
      "entropy_step_3": 0.2031,
      "epoch": 0.6433479075577764,
      "grad_norm": 63.999595642089844,
      "learning_rate": 3.093343703898569e-05,
      "loss": 0.057539844512939455,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0696,
      "step": 21630
    },
    {
      "entropy_step_1": 0.2015,
      "entropy_step_2": 0.2031,
      "entropy_step_3": 0.1952,
      "epoch": 0.6436453407096755,
      "grad_norm": 84.42562103271484,
      "learning_rate": 3.088798133545697e-05,
      "loss": 0.06176241636276245,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0524,
      "step": 21640
    },
    {
      "entropy_step_1": 0.1788,
      "entropy_step_2": 0.182,
      "entropy_step_3": 0.2047,
      "epoch": 0.6439427738615746,
      "grad_norm": 142.6101531982422,
      "learning_rate": 3.084254412306636e-05,
      "loss": 0.055750131607055664,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0584,
      "step": 21650
    },
    {
      "entropy_step_1": 0.1888,
      "entropy_step_2": 0.1903,
      "entropy_step_3": 0.1906,
      "epoch": 0.6442402070134737,
      "grad_norm": 57.946998596191406,
      "learning_rate": 3.079712544577499e-05,
      "loss": 0.053691315650939944,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0452,
      "step": 21660
    },
    {
      "entropy_step_1": 0.1744,
      "entropy_step_2": 0.1778,
      "entropy_step_3": 0.1671,
      "epoch": 0.6445376401653728,
      "grad_norm": 180.7014617919922,
      "learning_rate": 3.075172534752605e-05,
      "loss": 0.0587919294834137,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.049,
      "step": 21670
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1804,
      "entropy_step_3": 0.1822,
      "epoch": 0.644835073317272,
      "grad_norm": 99.82331848144531,
      "learning_rate": 3.070634387224474e-05,
      "loss": 0.0556318998336792,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.0461,
      "step": 21680
    },
    {
      "entropy_step_1": 0.2124,
      "entropy_step_2": 0.2078,
      "entropy_step_3": 0.2024,
      "epoch": 0.6451325064691711,
      "grad_norm": 80.5784683227539,
      "learning_rate": 3.066098106383826e-05,
      "loss": 0.06164829134941101,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0666,
      "loss_step_3": 0.0553,
      "step": 21690
    },
    {
      "entropy_step_1": 0.1927,
      "entropy_step_2": 0.1889,
      "entropy_step_3": 0.1813,
      "epoch": 0.6454299396210702,
      "grad_norm": 65.15019226074219,
      "learning_rate": 3.061563696619574e-05,
      "loss": 0.054516375064849854,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0578,
      "step": 21700
    },
    {
      "entropy_step_1": 0.1695,
      "entropy_step_2": 0.1697,
      "entropy_step_3": 0.1575,
      "epoch": 0.6457273727729693,
      "grad_norm": 89.43103790283203,
      "learning_rate": 3.057031162318824e-05,
      "loss": 0.06090713143348694,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0481,
      "step": 21710
    },
    {
      "entropy_step_1": 0.1731,
      "entropy_step_2": 0.1734,
      "entropy_step_3": 0.1738,
      "epoch": 0.6460248059248684,
      "grad_norm": 130.64990234375,
      "learning_rate": 3.052500507866861e-05,
      "loss": 0.060098910331726076,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0568,
      "step": 21720
    },
    {
      "entropy_step_1": 0.1753,
      "entropy_step_2": 0.1739,
      "entropy_step_3": 0.173,
      "epoch": 0.6463222390767674,
      "grad_norm": 272.2684326171875,
      "learning_rate": 3.0479717376471554e-05,
      "loss": 0.0551250159740448,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0554,
      "step": 21730
    },
    {
      "entropy_step_1": 0.2009,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.1891,
      "epoch": 0.6466196722286666,
      "grad_norm": 82.57572937011719,
      "learning_rate": 3.0434448560413555e-05,
      "loss": 0.05998104214668274,
      "loss_step_1": 0.0629,
      "loss_step_2": 0.0619,
      "loss_step_3": 0.0665,
      "step": 21740
    },
    {
      "entropy_step_1": 0.1718,
      "entropy_step_2": 0.1723,
      "entropy_step_3": 0.1735,
      "epoch": 0.6469171053805657,
      "grad_norm": 1447.8170166015625,
      "learning_rate": 3.0389198674292797e-05,
      "loss": 0.04908641278743744,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.0438,
      "step": 21750
    },
    {
      "entropy_step_1": 0.1705,
      "entropy_step_2": 0.1723,
      "entropy_step_3": 0.1628,
      "epoch": 0.6472145385324648,
      "grad_norm": 93.20700073242188,
      "learning_rate": 3.034396776188918e-05,
      "loss": 0.06527349948883057,
      "loss_step_1": 0.0692,
      "loss_step_2": 0.0671,
      "loss_step_3": 0.0594,
      "step": 21760
    },
    {
      "entropy_step_1": 0.1804,
      "entropy_step_2": 0.1809,
      "entropy_step_3": 0.189,
      "epoch": 0.6475119716843639,
      "grad_norm": 73.15001678466797,
      "learning_rate": 3.02987558669642e-05,
      "loss": 0.05868271589279175,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0627,
      "step": 21770
    },
    {
      "entropy_step_1": 0.1877,
      "entropy_step_2": 0.1861,
      "entropy_step_3": 0.1753,
      "epoch": 0.647809404836263,
      "grad_norm": 60.100624084472656,
      "learning_rate": 3.0253563033260988e-05,
      "loss": 0.056305873394012454,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0521,
      "step": 21780
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.1987,
      "entropy_step_3": 0.2063,
      "epoch": 0.6481068379881622,
      "grad_norm": 204.2509307861328,
      "learning_rate": 3.0208389304504235e-05,
      "loss": 0.06061859130859375,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0645,
      "step": 21790
    },
    {
      "entropy_step_1": 0.1919,
      "entropy_step_2": 0.1915,
      "entropy_step_3": 0.1801,
      "epoch": 0.6484042711400613,
      "grad_norm": 89.58988952636719,
      "learning_rate": 3.0163234724400136e-05,
      "loss": 0.0608328104019165,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0612,
      "step": 21800
    },
    {
      "entropy_step_1": 0.1951,
      "entropy_step_2": 0.1943,
      "entropy_step_3": 0.1942,
      "epoch": 0.6487017042919604,
      "grad_norm": 55.485862731933594,
      "learning_rate": 3.0118099336636375e-05,
      "loss": 0.05399664640426636,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.059,
      "step": 21810
    },
    {
      "entropy_step_1": 0.1777,
      "entropy_step_2": 0.1771,
      "entropy_step_3": 0.1904,
      "epoch": 0.6489991374438595,
      "grad_norm": 146.5654754638672,
      "learning_rate": 3.007298318488202e-05,
      "loss": 0.04763535261154175,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0509,
      "step": 21820
    },
    {
      "entropy_step_1": 0.1794,
      "entropy_step_2": 0.1821,
      "entropy_step_3": 0.1767,
      "epoch": 0.6492965705957586,
      "grad_norm": 105.45609283447266,
      "learning_rate": 3.002788631278758e-05,
      "loss": 0.06267492771148682,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0476,
      "step": 21830
    },
    {
      "entropy_step_1": 0.1996,
      "entropy_step_2": 0.1985,
      "entropy_step_3": 0.1971,
      "epoch": 0.6495940037476577,
      "grad_norm": 496.58990478515625,
      "learning_rate": 2.998280876398489e-05,
      "loss": 0.07261739373207092,
      "loss_step_1": 0.0767,
      "loss_step_2": 0.0757,
      "loss_step_3": 0.0902,
      "step": 21840
    },
    {
      "entropy_step_1": 0.2183,
      "entropy_step_2": 0.2165,
      "entropy_step_3": 0.2233,
      "epoch": 0.6498914368995569,
      "grad_norm": 147.41429138183594,
      "learning_rate": 2.993775058208711e-05,
      "loss": 0.05972027778625488,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0633,
      "loss_step_3": 0.0675,
      "step": 21850
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2013,
      "entropy_step_3": 0.2108,
      "epoch": 0.650188870051456,
      "grad_norm": 104.0100326538086,
      "learning_rate": 2.9892711810688613e-05,
      "loss": 0.06370716094970703,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0602,
      "step": 21860
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.1806,
      "entropy_step_3": 0.1863,
      "epoch": 0.6504863032033551,
      "grad_norm": 169.1422119140625,
      "learning_rate": 2.9847692493365044e-05,
      "loss": 0.05584896206855774,
      "loss_step_1": 0.0497,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0509,
      "step": 21870
    },
    {
      "entropy_step_1": 0.1773,
      "entropy_step_2": 0.1763,
      "entropy_step_3": 0.1757,
      "epoch": 0.6507837363552541,
      "grad_norm": 111.89019012451172,
      "learning_rate": 2.9802692673673204e-05,
      "loss": 0.05348804593086243,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0541,
      "step": 21880
    },
    {
      "entropy_step_1": 0.182,
      "entropy_step_2": 0.1839,
      "entropy_step_3": 0.1834,
      "epoch": 0.6510811695071532,
      "grad_norm": 102.23084259033203,
      "learning_rate": 2.9757712395151015e-05,
      "loss": 0.062223637104034425,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0678,
      "step": 21890
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.1895,
      "epoch": 0.6513786026590523,
      "grad_norm": 255.9478759765625,
      "learning_rate": 2.971275170131753e-05,
      "loss": 0.05866650938987732,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0554,
      "step": 21900
    },
    {
      "entropy_step_1": 0.2009,
      "entropy_step_2": 0.1973,
      "entropy_step_3": 0.1891,
      "epoch": 0.6516760358109515,
      "grad_norm": 225.91073608398438,
      "learning_rate": 2.966781063567282e-05,
      "loss": 0.05638231635093689,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0543,
      "step": 21910
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.1764,
      "entropy_step_3": 0.1715,
      "epoch": 0.6519734689628506,
      "grad_norm": 117.77657318115234,
      "learning_rate": 2.9622889241697982e-05,
      "loss": 0.05620511770248413,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0516,
      "step": 21920
    },
    {
      "entropy_step_1": 0.1821,
      "entropy_step_2": 0.1784,
      "entropy_step_3": 0.1797,
      "epoch": 0.6522709021147497,
      "grad_norm": 89.43612670898438,
      "learning_rate": 2.957798756285506e-05,
      "loss": 0.05473183393478394,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0529,
      "step": 21930
    },
    {
      "entropy_step_1": 0.1819,
      "entropy_step_2": 0.1798,
      "entropy_step_3": 0.1772,
      "epoch": 0.6525683352666488,
      "grad_norm": 87.60095977783203,
      "learning_rate": 2.953310564258708e-05,
      "loss": 0.05692439675331116,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0489,
      "step": 21940
    },
    {
      "entropy_step_1": 0.2003,
      "entropy_step_2": 0.2008,
      "entropy_step_3": 0.1914,
      "epoch": 0.6528657684185479,
      "grad_norm": 107.45437622070312,
      "learning_rate": 2.948824352431786e-05,
      "loss": 0.0616338312625885,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0621,
      "step": 21950
    },
    {
      "entropy_step_1": 0.1685,
      "entropy_step_2": 0.167,
      "entropy_step_3": 0.1712,
      "epoch": 0.653163201570447,
      "grad_norm": 91.54277038574219,
      "learning_rate": 2.9443401251452128e-05,
      "loss": 0.05026301145553589,
      "loss_step_1": 0.0462,
      "loss_step_2": 0.046,
      "loss_step_3": 0.0459,
      "step": 21960
    },
    {
      "entropy_step_1": 0.1728,
      "entropy_step_2": 0.1705,
      "entropy_step_3": 0.1758,
      "epoch": 0.6534606347223462,
      "grad_norm": 267.67559814453125,
      "learning_rate": 2.9398578867375402e-05,
      "loss": 0.05660305619239807,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0613,
      "step": 21970
    },
    {
      "entropy_step_1": 0.161,
      "entropy_step_2": 0.1604,
      "entropy_step_3": 0.1568,
      "epoch": 0.6537580678742453,
      "grad_norm": 47.02507781982422,
      "learning_rate": 2.9353776415453948e-05,
      "loss": 0.0533237099647522,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0562,
      "step": 21980
    },
    {
      "entropy_step_1": 0.1874,
      "entropy_step_2": 0.1866,
      "entropy_step_3": 0.2012,
      "epoch": 0.6540555010261444,
      "grad_norm": 399.64862060546875,
      "learning_rate": 2.9308993939034767e-05,
      "loss": 0.05744631290435791,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0517,
      "step": 21990
    },
    {
      "entropy_step_1": 0.1896,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1932,
      "epoch": 0.6543529341780435,
      "grad_norm": 290.31890869140625,
      "learning_rate": 2.926423148144547e-05,
      "loss": 0.05946905016899109,
      "loss_step_1": 0.0506,
      "loss_step_2": 0.0497,
      "loss_step_3": 0.0493,
      "step": 22000
    },
    {
      "epoch": 0.6543529341780435,
      "eval_accuracy": 0.907,
      "eval_f1": 0.9069418386491557,
      "eval_loss": 0.10049314796924591,
      "eval_precision": 0.9080200501253132,
      "eval_recall": 0.907,
      "eval_runtime": 6.517,
      "eval_samples_per_second": 76.723,
      "eval_steps_per_second": 9.667,
      "step": 22000
    },
    {
      "entropy_step_1": 0.2002,
      "entropy_step_2": 0.2002,
      "entropy_step_3": 0.2053,
      "epoch": 0.6546503673299426,
      "grad_norm": 114.84507751464844,
      "learning_rate": 2.9219489085994363e-05,
      "loss": 0.05779011249542236,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.045,
      "step": 22010
    },
    {
      "entropy_step_1": 0.1762,
      "entropy_step_2": 0.1755,
      "entropy_step_3": 0.1795,
      "epoch": 0.6549478004818418,
      "grad_norm": 97.8281021118164,
      "learning_rate": 2.917476679597032e-05,
      "loss": 0.05471767783164978,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0606,
      "step": 22020
    },
    {
      "entropy_step_1": 0.2104,
      "entropy_step_2": 0.2082,
      "entropy_step_3": 0.1839,
      "epoch": 0.6552452336337408,
      "grad_norm": 76.71304321289062,
      "learning_rate": 2.913006465464281e-05,
      "loss": 0.06356891989707947,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0655,
      "step": 22030
    },
    {
      "entropy_step_1": 0.2059,
      "entropy_step_2": 0.2048,
      "entropy_step_3": 0.1992,
      "epoch": 0.6555426667856399,
      "grad_norm": 136.49249267578125,
      "learning_rate": 2.9085382705261677e-05,
      "loss": 0.05477166771888733,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0493,
      "step": 22040
    },
    {
      "entropy_step_1": 0.1906,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.1878,
      "epoch": 0.655840099937539,
      "grad_norm": 68.46590423583984,
      "learning_rate": 2.904072099105739e-05,
      "loss": 0.06369816064834595,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0526,
      "step": 22050
    },
    {
      "entropy_step_1": 0.1864,
      "entropy_step_2": 0.1844,
      "entropy_step_3": 0.1874,
      "epoch": 0.6561375330894381,
      "grad_norm": 165.23524475097656,
      "learning_rate": 2.8996079555240706e-05,
      "loss": 0.06668729782104492,
      "loss_step_1": 0.0675,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0603,
      "step": 22060
    },
    {
      "entropy_step_1": 0.2104,
      "entropy_step_2": 0.2079,
      "entropy_step_3": 0.2195,
      "epoch": 0.6564349662413372,
      "grad_norm": 132.77101135253906,
      "learning_rate": 2.8951458441002877e-05,
      "loss": 0.05318529605865478,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0515,
      "step": 22070
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.1913,
      "entropy_step_3": 0.184,
      "epoch": 0.6567323993932364,
      "grad_norm": 344.63079833984375,
      "learning_rate": 2.890685769151541e-05,
      "loss": 0.06022869944572449,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0694,
      "step": 22080
    },
    {
      "entropy_step_1": 0.1929,
      "entropy_step_2": 0.1945,
      "entropy_step_3": 0.1967,
      "epoch": 0.6570298325451355,
      "grad_norm": 264.8663635253906,
      "learning_rate": 2.8862277349930144e-05,
      "loss": 0.06034838557243347,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0709,
      "step": 22090
    },
    {
      "entropy_step_1": 0.1927,
      "entropy_step_2": 0.1972,
      "entropy_step_3": 0.2156,
      "epoch": 0.6573272656970346,
      "grad_norm": 142.92198181152344,
      "learning_rate": 2.881771745937914e-05,
      "loss": 0.058598142862319944,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0625,
      "step": 22100
    },
    {
      "entropy_step_1": 0.1972,
      "entropy_step_2": 0.2025,
      "entropy_step_3": 0.21,
      "epoch": 0.6576246988489337,
      "grad_norm": 130.2975311279297,
      "learning_rate": 2.8773178062974742e-05,
      "loss": 0.06101212501525879,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0591,
      "step": 22110
    },
    {
      "entropy_step_1": 0.1912,
      "entropy_step_2": 0.1974,
      "entropy_step_3": 0.1957,
      "epoch": 0.6579221320008328,
      "grad_norm": 550.7367553710938,
      "learning_rate": 2.8728659203809398e-05,
      "loss": 0.05780099034309387,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0536,
      "step": 22120
    },
    {
      "entropy_step_1": 0.2112,
      "entropy_step_2": 0.2177,
      "entropy_step_3": 0.2221,
      "epoch": 0.658219565152732,
      "grad_norm": 144.1768035888672,
      "learning_rate": 2.8684160924955716e-05,
      "loss": 0.05928115844726563,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0568,
      "step": 22130
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.1958,
      "entropy_step_3": 0.1934,
      "epoch": 0.6585169983046311,
      "grad_norm": 93.71833038330078,
      "learning_rate": 2.8639683269466368e-05,
      "loss": 0.050447916984558104,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.053,
      "step": 22140
    },
    {
      "entropy_step_1": 0.2148,
      "entropy_step_2": 0.2149,
      "entropy_step_3": 0.2177,
      "epoch": 0.6588144314565302,
      "grad_norm": 105.52024841308594,
      "learning_rate": 2.8595226280374114e-05,
      "loss": 0.061935806274414064,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.07,
      "step": 22150
    },
    {
      "entropy_step_1": 0.2061,
      "entropy_step_2": 0.2038,
      "entropy_step_3": 0.2106,
      "epoch": 0.6591118646084293,
      "grad_norm": 81.22308349609375,
      "learning_rate": 2.8550790000691698e-05,
      "loss": 0.05713624358177185,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0568,
      "step": 22160
    },
    {
      "entropy_step_1": 0.1894,
      "entropy_step_2": 0.1936,
      "entropy_step_3": 0.1848,
      "epoch": 0.6594092977603284,
      "grad_norm": 97.1379623413086,
      "learning_rate": 2.850637447341181e-05,
      "loss": 0.06148044466972351,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0578,
      "step": 22170
    },
    {
      "entropy_step_1": 0.1862,
      "entropy_step_2": 0.1903,
      "entropy_step_3": 0.1896,
      "epoch": 0.6597067309122274,
      "grad_norm": 127.93582153320312,
      "learning_rate": 2.8461979741507084e-05,
      "loss": 0.05444079637527466,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0477,
      "loss_step_3": 0.0463,
      "step": 22180
    },
    {
      "entropy_step_1": 0.1707,
      "entropy_step_2": 0.1732,
      "entropy_step_3": 0.1684,
      "epoch": 0.6600041640641265,
      "grad_norm": 43.308013916015625,
      "learning_rate": 2.841760584793e-05,
      "loss": 0.051870471239089964,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0461,
      "step": 22190
    },
    {
      "entropy_step_1": 0.1846,
      "entropy_step_2": 0.1844,
      "entropy_step_3": 0.1777,
      "epoch": 0.6603015972160257,
      "grad_norm": 60.78437042236328,
      "learning_rate": 2.8373252835612958e-05,
      "loss": 0.059866982698440555,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0678,
      "step": 22200
    },
    {
      "entropy_step_1": 0.1904,
      "entropy_step_2": 0.1879,
      "entropy_step_3": 0.1867,
      "epoch": 0.6605990303679248,
      "grad_norm": 43.86735916137695,
      "learning_rate": 2.8328920747468047e-05,
      "loss": 0.05781788825988769,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0602,
      "step": 22210
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.2044,
      "entropy_step_3": 0.1699,
      "epoch": 0.6608964635198239,
      "grad_norm": 249.3356475830078,
      "learning_rate": 2.8284609626387226e-05,
      "loss": 0.06534186005592346,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0552,
      "step": 22220
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.1942,
      "epoch": 0.661193896671723,
      "grad_norm": 127.52993774414062,
      "learning_rate": 2.824031951524204e-05,
      "loss": 0.057655084133148196,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0576,
      "step": 22230
    },
    {
      "entropy_step_1": 0.1892,
      "entropy_step_2": 0.1874,
      "entropy_step_3": 0.1713,
      "epoch": 0.6614913298236221,
      "grad_norm": 116.5815658569336,
      "learning_rate": 2.8196050456883828e-05,
      "loss": 0.05693889856338501,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0501,
      "step": 22240
    },
    {
      "entropy_step_1": 0.2002,
      "entropy_step_2": 0.1971,
      "entropy_step_3": 0.1936,
      "epoch": 0.6617887629755212,
      "grad_norm": 68.06834411621094,
      "learning_rate": 2.8151802494143452e-05,
      "loss": 0.05689901113510132,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0496,
      "step": 22250
    },
    {
      "entropy_step_1": 0.2011,
      "entropy_step_2": 0.1999,
      "entropy_step_3": 0.2126,
      "epoch": 0.6620861961274204,
      "grad_norm": 104.0520248413086,
      "learning_rate": 2.8107575669831503e-05,
      "loss": 0.05852494239807129,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0642,
      "step": 22260
    },
    {
      "entropy_step_1": 0.2014,
      "entropy_step_2": 0.2025,
      "entropy_step_3": 0.1931,
      "epoch": 0.6623836292793195,
      "grad_norm": 97.80985260009766,
      "learning_rate": 2.806337002673794e-05,
      "loss": 0.059280604124069214,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0529,
      "step": 22270
    },
    {
      "entropy_step_1": 0.1884,
      "entropy_step_2": 0.1862,
      "entropy_step_3": 0.1798,
      "epoch": 0.6626810624312186,
      "grad_norm": 90.6190185546875,
      "learning_rate": 2.8019185607632386e-05,
      "loss": 0.05645957589149475,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0483,
      "step": 22280
    },
    {
      "entropy_step_1": 0.1936,
      "entropy_step_2": 0.19,
      "entropy_step_3": 0.1924,
      "epoch": 0.6629784955831177,
      "grad_norm": 95.50624084472656,
      "learning_rate": 2.7975022455263833e-05,
      "loss": 0.061503255367279054,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0564,
      "step": 22290
    },
    {
      "entropy_step_1": 0.2019,
      "entropy_step_2": 0.1994,
      "entropy_step_3": 0.1834,
      "epoch": 0.6632759287350168,
      "grad_norm": 65.94950103759766,
      "learning_rate": 2.793088061236076e-05,
      "loss": 0.05521695613861084,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0602,
      "step": 22300
    },
    {
      "entropy_step_1": 0.2042,
      "entropy_step_2": 0.2054,
      "entropy_step_3": 0.1977,
      "epoch": 0.663573361886916,
      "grad_norm": 166.6420440673828,
      "learning_rate": 2.7886760121630996e-05,
      "loss": 0.05489189028739929,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0528,
      "step": 22310
    },
    {
      "entropy_step_1": 0.1741,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1827,
      "epoch": 0.6638707950388151,
      "grad_norm": 621.2066650390625,
      "learning_rate": 2.78426610257617e-05,
      "loss": 0.06490917801856995,
      "loss_step_1": 0.0649,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0596,
      "step": 22320
    },
    {
      "entropy_step_1": 0.1926,
      "entropy_step_2": 0.1955,
      "entropy_step_3": 0.1961,
      "epoch": 0.6641682281907141,
      "grad_norm": 83.42292022705078,
      "learning_rate": 2.779858336741936e-05,
      "loss": 0.05017282366752625,
      "loss_step_1": 0.0506,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0455,
      "step": 22330
    },
    {
      "entropy_step_1": 0.1677,
      "entropy_step_2": 0.1717,
      "entropy_step_3": 0.172,
      "epoch": 0.6644656613426132,
      "grad_norm": 177.7521514892578,
      "learning_rate": 2.7754527189249692e-05,
      "loss": 0.05467460751533508,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.059,
      "step": 22340
    },
    {
      "entropy_step_1": 0.2039,
      "entropy_step_2": 0.2053,
      "entropy_step_3": 0.2114,
      "epoch": 0.6647630944945123,
      "grad_norm": 134.91998291015625,
      "learning_rate": 2.7710492533877673e-05,
      "loss": 0.05859636664390564,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0479,
      "step": 22350
    },
    {
      "entropy_step_1": 0.2028,
      "entropy_step_2": 0.2007,
      "entropy_step_3": 0.2133,
      "epoch": 0.6650605276464114,
      "grad_norm": 43.78353500366211,
      "learning_rate": 2.7666479443907422e-05,
      "loss": 0.06716082692146301,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0528,
      "step": 22360
    },
    {
      "entropy_step_1": 0.1947,
      "entropy_step_2": 0.191,
      "entropy_step_3": 0.2031,
      "epoch": 0.6653579607983106,
      "grad_norm": 101.8807601928711,
      "learning_rate": 2.7622487961922206e-05,
      "loss": 0.05885818600654602,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0706,
      "step": 22370
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.1772,
      "entropy_step_3": 0.1671,
      "epoch": 0.6656553939502097,
      "grad_norm": 35.383724212646484,
      "learning_rate": 2.7578518130484354e-05,
      "loss": 0.05279458165168762,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0493,
      "step": 22380
    },
    {
      "entropy_step_1": 0.191,
      "entropy_step_2": 0.1927,
      "entropy_step_3": 0.1851,
      "epoch": 0.6659528271021088,
      "grad_norm": 77.91779327392578,
      "learning_rate": 2.7534569992135316e-05,
      "loss": 0.0633106291294098,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0616,
      "loss_step_3": 0.0574,
      "step": 22390
    },
    {
      "entropy_step_1": 0.1572,
      "entropy_step_2": 0.1608,
      "entropy_step_3": 0.1578,
      "epoch": 0.6662502602540079,
      "grad_norm": 63.783485412597656,
      "learning_rate": 2.7490643589395498e-05,
      "loss": 0.05340284705162048,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0495,
      "step": 22400
    },
    {
      "entropy_step_1": 0.1805,
      "entropy_step_2": 0.1816,
      "entropy_step_3": 0.1778,
      "epoch": 0.666547693405907,
      "grad_norm": 106.46318054199219,
      "learning_rate": 2.7446738964764296e-05,
      "loss": 0.05920029878616333,
      "loss_step_1": 0.0612,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0596,
      "step": 22410
    },
    {
      "entropy_step_1": 0.195,
      "entropy_step_2": 0.1957,
      "entropy_step_3": 0.1886,
      "epoch": 0.6668451265578061,
      "grad_norm": 35.68791198730469,
      "learning_rate": 2.7402856160720003e-05,
      "loss": 0.05634062886238098,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0569,
      "step": 22420
    },
    {
      "entropy_step_1": 0.1672,
      "entropy_step_2": 0.1705,
      "entropy_step_3": 0.1738,
      "epoch": 0.6671425597097053,
      "grad_norm": 100.4879150390625,
      "learning_rate": 2.7358995219719875e-05,
      "loss": 0.04828576743602753,
      "loss_step_1": 0.043,
      "loss_step_2": 0.0432,
      "loss_step_3": 0.0501,
      "step": 22430
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.1871,
      "entropy_step_3": 0.1877,
      "epoch": 0.6674399928616044,
      "grad_norm": 186.290283203125,
      "learning_rate": 2.7315156184199924e-05,
      "loss": 0.055532991886138916,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0435,
      "step": 22440
    },
    {
      "entropy_step_1": 0.177,
      "entropy_step_2": 0.182,
      "entropy_step_3": 0.1959,
      "epoch": 0.6677374260135035,
      "grad_norm": 95.88404846191406,
      "learning_rate": 2.7271339096575076e-05,
      "loss": 0.06663373708724976,
      "loss_step_1": 0.0659,
      "loss_step_2": 0.0649,
      "loss_step_3": 0.0655,
      "step": 22450
    },
    {
      "entropy_step_1": 0.2109,
      "entropy_step_2": 0.2145,
      "entropy_step_3": 0.2149,
      "epoch": 0.6680348591654026,
      "grad_norm": 71.90017700195312,
      "learning_rate": 2.722754399923889e-05,
      "loss": 0.05544441342353821,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0561,
      "step": 22460
    },
    {
      "entropy_step_1": 0.1741,
      "entropy_step_2": 0.1787,
      "entropy_step_3": 0.184,
      "epoch": 0.6683322923173017,
      "grad_norm": 117.04560089111328,
      "learning_rate": 2.7183770934563757e-05,
      "loss": 0.047841128706932065,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0511,
      "step": 22470
    },
    {
      "entropy_step_1": 0.1612,
      "entropy_step_2": 0.1665,
      "entropy_step_3": 0.1579,
      "epoch": 0.6686297254692007,
      "grad_norm": 199.7707061767578,
      "learning_rate": 2.7140019944900696e-05,
      "loss": 0.05698502659797668,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0469,
      "step": 22480
    },
    {
      "entropy_step_1": 0.1975,
      "entropy_step_2": 0.201,
      "entropy_step_3": 0.1925,
      "epoch": 0.6689271586210999,
      "grad_norm": 55.23896026611328,
      "learning_rate": 2.7096291072579426e-05,
      "loss": 0.06414267420768738,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0533,
      "step": 22490
    },
    {
      "entropy_step_1": 0.1815,
      "entropy_step_2": 0.182,
      "entropy_step_3": 0.1964,
      "epoch": 0.669224591772999,
      "grad_norm": 110.75791931152344,
      "learning_rate": 2.705258435990815e-05,
      "loss": 0.05485146641731262,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0562,
      "step": 22500
    },
    {
      "entropy_step_1": 0.2,
      "entropy_step_2": 0.2005,
      "entropy_step_3": 0.1953,
      "epoch": 0.6695220249248981,
      "grad_norm": 107.32481384277344,
      "learning_rate": 2.700889984917378e-05,
      "loss": 0.0530930757522583,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0516,
      "step": 22510
    },
    {
      "entropy_step_1": 0.1647,
      "entropy_step_2": 0.1629,
      "entropy_step_3": 0.167,
      "epoch": 0.6698194580767972,
      "grad_norm": 122.10095977783203,
      "learning_rate": 2.696523758264164e-05,
      "loss": 0.05286581516265869,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0525,
      "step": 22520
    },
    {
      "entropy_step_1": 0.2103,
      "entropy_step_2": 0.2069,
      "entropy_step_3": 0.1995,
      "epoch": 0.6701168912286963,
      "grad_norm": 432.71612548828125,
      "learning_rate": 2.6921597602555554e-05,
      "loss": 0.0648653507232666,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.0747,
      "step": 22530
    },
    {
      "entropy_step_1": 0.2005,
      "entropy_step_2": 0.1996,
      "entropy_step_3": 0.2199,
      "epoch": 0.6704143243805954,
      "grad_norm": 120.34432220458984,
      "learning_rate": 2.6877979951137855e-05,
      "loss": 0.06300555467605591,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0656,
      "loss_step_3": 0.0643,
      "step": 22540
    },
    {
      "entropy_step_1": 0.1727,
      "entropy_step_2": 0.174,
      "entropy_step_3": 0.18,
      "epoch": 0.6707117575324946,
      "grad_norm": 80.04916381835938,
      "learning_rate": 2.683438467058913e-05,
      "loss": 0.05149502158164978,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0575,
      "step": 22550
    },
    {
      "entropy_step_1": 0.1622,
      "entropy_step_2": 0.1647,
      "entropy_step_3": 0.1865,
      "epoch": 0.6710091906843937,
      "grad_norm": 203.5390625,
      "learning_rate": 2.6790811803088473e-05,
      "loss": 0.07380011677742004,
      "loss_step_1": 0.0689,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0813,
      "step": 22560
    },
    {
      "entropy_step_1": 0.1945,
      "entropy_step_2": 0.1956,
      "entropy_step_3": 0.1735,
      "epoch": 0.6713066238362928,
      "grad_norm": 49.321292877197266,
      "learning_rate": 2.6747261390793182e-05,
      "loss": 0.0533049464225769,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0474,
      "loss_step_3": 0.0356,
      "step": 22570
    },
    {
      "entropy_step_1": 0.1882,
      "entropy_step_2": 0.189,
      "entropy_step_3": 0.2022,
      "epoch": 0.6716040569881919,
      "grad_norm": 91.83394622802734,
      "learning_rate": 2.6703733475838932e-05,
      "loss": 0.05472918748855591,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0531,
      "step": 22580
    },
    {
      "entropy_step_1": 0.1918,
      "entropy_step_2": 0.1911,
      "entropy_step_3": 0.1957,
      "epoch": 0.671901490140091,
      "grad_norm": 109.96269226074219,
      "learning_rate": 2.6660228100339502e-05,
      "loss": 0.05885461568832397,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0574,
      "step": 22590
    },
    {
      "entropy_step_1": 0.1928,
      "entropy_step_2": 0.1916,
      "entropy_step_3": 0.2099,
      "epoch": 0.6721989232919902,
      "grad_norm": 123.61125183105469,
      "learning_rate": 2.661674530638698e-05,
      "loss": 0.05198657512664795,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0555,
      "step": 22600
    },
    {
      "entropy_step_1": 0.1866,
      "entropy_step_2": 0.1855,
      "entropy_step_3": 0.1855,
      "epoch": 0.6724963564438893,
      "grad_norm": 1978.1302490234375,
      "learning_rate": 2.657328513605154e-05,
      "loss": 0.054232990741729735,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0534,
      "step": 22610
    },
    {
      "entropy_step_1": 0.168,
      "entropy_step_2": 0.1667,
      "entropy_step_3": 0.1766,
      "epoch": 0.6727937895957884,
      "grad_norm": 100.66785430908203,
      "learning_rate": 2.652984763138151e-05,
      "loss": 0.05336818695068359,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0549,
      "step": 22620
    },
    {
      "entropy_step_1": 0.1841,
      "entropy_step_2": 0.1809,
      "entropy_step_3": 0.2049,
      "epoch": 0.6730912227476875,
      "grad_norm": 143.3184356689453,
      "learning_rate": 2.6486432834403264e-05,
      "loss": 0.05743290781974793,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0632,
      "step": 22630
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.1798,
      "entropy_step_3": 0.1493,
      "epoch": 0.6733886558995865,
      "grad_norm": 53.48335647583008,
      "learning_rate": 2.6443040787121186e-05,
      "loss": 0.051775646209716794,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.041,
      "step": 22640
    },
    {
      "entropy_step_1": 0.1829,
      "entropy_step_2": 0.1819,
      "entropy_step_3": 0.1739,
      "epoch": 0.6736860890514856,
      "grad_norm": 99.33316802978516,
      "learning_rate": 2.6399671531517672e-05,
      "loss": 0.05531848073005676,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0489,
      "step": 22650
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.172,
      "epoch": 0.6739835222033848,
      "grad_norm": 70.8333969116211,
      "learning_rate": 2.63563251095531e-05,
      "loss": 0.05247322916984558,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0533,
      "step": 22660
    },
    {
      "entropy_step_1": 0.1966,
      "entropy_step_2": 0.1936,
      "entropy_step_3": 0.2039,
      "epoch": 0.6742809553552839,
      "grad_norm": 89.21510314941406,
      "learning_rate": 2.63130015631657e-05,
      "loss": 0.05503894686698914,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0631,
      "step": 22670
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.184,
      "entropy_step_3": 0.1886,
      "epoch": 0.674578388507183,
      "grad_norm": 61.67454528808594,
      "learning_rate": 2.62697009342716e-05,
      "loss": 0.05619686245918274,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.0523,
      "step": 22680
    },
    {
      "entropy_step_1": 0.1866,
      "entropy_step_2": 0.1824,
      "entropy_step_3": 0.1687,
      "epoch": 0.6748758216590821,
      "grad_norm": 138.63217163085938,
      "learning_rate": 2.622642326476472e-05,
      "loss": 0.05903367400169372,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.054,
      "step": 22690
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2019,
      "entropy_step_3": 0.1964,
      "epoch": 0.6751732548109812,
      "grad_norm": 244.9290771484375,
      "learning_rate": 2.6183168596516837e-05,
      "loss": 0.051948726177215576,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0494,
      "step": 22700
    },
    {
      "entropy_step_1": 0.2151,
      "entropy_step_2": 0.2115,
      "entropy_step_3": 0.2137,
      "epoch": 0.6754706879628803,
      "grad_norm": 33.29358673095703,
      "learning_rate": 2.613993697137742e-05,
      "loss": 0.05839239358901978,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0625,
      "step": 22710
    },
    {
      "entropy_step_1": 0.1681,
      "entropy_step_2": 0.1708,
      "entropy_step_3": 0.1674,
      "epoch": 0.6757681211147795,
      "grad_norm": 36.56572723388672,
      "learning_rate": 2.6096728431173638e-05,
      "loss": 0.05091709494590759,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0509,
      "step": 22720
    },
    {
      "entropy_step_1": 0.1627,
      "entropy_step_2": 0.1699,
      "entropy_step_3": 0.1727,
      "epoch": 0.6760655542666786,
      "grad_norm": 91.581298828125,
      "learning_rate": 2.605354301771036e-05,
      "loss": 0.057533055543899536,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0583,
      "step": 22730
    },
    {
      "entropy_step_1": 0.1604,
      "entropy_step_2": 0.1672,
      "entropy_step_3": 0.1729,
      "epoch": 0.6763629874185777,
      "grad_norm": 99.4872055053711,
      "learning_rate": 2.6010380772770014e-05,
      "loss": 0.05387184619903564,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0495,
      "step": 22740
    },
    {
      "entropy_step_1": 0.1821,
      "entropy_step_2": 0.1882,
      "entropy_step_3": 0.1951,
      "epoch": 0.6766604205704768,
      "grad_norm": 232.2489776611328,
      "learning_rate": 2.5967241738112736e-05,
      "loss": 0.05691277384757996,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0702,
      "step": 22750
    },
    {
      "entropy_step_1": 0.1642,
      "entropy_step_2": 0.1711,
      "entropy_step_3": 0.1817,
      "epoch": 0.6769578537223759,
      "grad_norm": 123.0716552734375,
      "learning_rate": 2.5924125955476054e-05,
      "loss": 0.055909329652786256,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0551,
      "step": 22760
    },
    {
      "entropy_step_1": 0.1598,
      "entropy_step_2": 0.1662,
      "entropy_step_3": 0.1427,
      "epoch": 0.677255286874275,
      "grad_norm": 436.1871337890625,
      "learning_rate": 2.5881033466575155e-05,
      "loss": 0.04857642948627472,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0427,
      "step": 22770
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1925,
      "entropy_step_3": 0.198,
      "epoch": 0.6775527200261742,
      "grad_norm": 91.71216583251953,
      "learning_rate": 2.5837964313102524e-05,
      "loss": 0.06141836047172546,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0636,
      "step": 22780
    },
    {
      "entropy_step_1": 0.1859,
      "entropy_step_2": 0.1886,
      "entropy_step_3": 0.1944,
      "epoch": 0.6778501531780732,
      "grad_norm": 164.308349609375,
      "learning_rate": 2.5794918536728212e-05,
      "loss": 0.057374817132949826,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0594,
      "step": 22790
    },
    {
      "entropy_step_1": 0.1849,
      "entropy_step_2": 0.1852,
      "entropy_step_3": 0.1671,
      "epoch": 0.6781475863299723,
      "grad_norm": 109.0682373046875,
      "learning_rate": 2.5751896179099556e-05,
      "loss": 0.060530835390090944,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0438,
      "step": 22800
    },
    {
      "entropy_step_1": 0.1994,
      "entropy_step_2": 0.1984,
      "entropy_step_3": 0.2118,
      "epoch": 0.6784450194818714,
      "grad_norm": 151.0442657470703,
      "learning_rate": 2.5708897281841325e-05,
      "loss": 0.05961846709251404,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0583,
      "step": 22810
    },
    {
      "entropy_step_1": 0.1938,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.1954,
      "epoch": 0.6787424526337705,
      "grad_norm": 121.77062225341797,
      "learning_rate": 2.5665921886555466e-05,
      "loss": 0.057525748014450075,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0685,
      "step": 22820
    },
    {
      "entropy_step_1": 0.1956,
      "entropy_step_2": 0.1915,
      "entropy_step_3": 0.1884,
      "epoch": 0.6790398857856697,
      "grad_norm": 115.10416412353516,
      "learning_rate": 2.5622970034821314e-05,
      "loss": 0.052308082580566406,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0546,
      "step": 22830
    },
    {
      "entropy_step_1": 0.1751,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1691,
      "epoch": 0.6793373189375688,
      "grad_norm": 157.5750732421875,
      "learning_rate": 2.558004176819533e-05,
      "loss": 0.05298585295677185,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0619,
      "step": 22840
    },
    {
      "entropy_step_1": 0.1688,
      "entropy_step_2": 0.1687,
      "entropy_step_3": 0.1603,
      "epoch": 0.6796347520894679,
      "grad_norm": 243.57437133789062,
      "learning_rate": 2.5537137128211235e-05,
      "loss": 0.05260777473449707,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.0483,
      "step": 22850
    },
    {
      "entropy_step_1": 0.1819,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1815,
      "epoch": 0.679932185241367,
      "grad_norm": 144.0232696533203,
      "learning_rate": 2.5494256156379836e-05,
      "loss": 0.05297564268112183,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0533,
      "step": 22860
    },
    {
      "entropy_step_1": 0.1768,
      "entropy_step_2": 0.1763,
      "entropy_step_3": 0.1726,
      "epoch": 0.6802296183932661,
      "grad_norm": 85.81971740722656,
      "learning_rate": 2.545139889418906e-05,
      "loss": 0.05363874435424805,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.0488,
      "step": 22870
    },
    {
      "entropy_step_1": 0.182,
      "entropy_step_2": 0.1836,
      "entropy_step_3": 0.1676,
      "epoch": 0.6805270515451652,
      "grad_norm": 48.46589279174805,
      "learning_rate": 2.5408565383103878e-05,
      "loss": 0.05305519700050354,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0478,
      "step": 22880
    },
    {
      "entropy_step_1": 0.1872,
      "entropy_step_2": 0.1878,
      "entropy_step_3": 0.1875,
      "epoch": 0.6808244846970644,
      "grad_norm": 174.94656372070312,
      "learning_rate": 2.5365755664566283e-05,
      "loss": 0.05757282972335816,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0508,
      "step": 22890
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.193,
      "entropy_step_3": 0.1828,
      "epoch": 0.6811219178489635,
      "grad_norm": 53.143707275390625,
      "learning_rate": 2.5322969779995304e-05,
      "loss": 0.06093593239784241,
      "loss_step_1": 0.0676,
      "loss_step_2": 0.066,
      "loss_step_3": 0.0647,
      "step": 22900
    },
    {
      "entropy_step_1": 0.1934,
      "entropy_step_2": 0.1975,
      "entropy_step_3": 0.1645,
      "epoch": 0.6814193510008626,
      "grad_norm": 119.52581787109375,
      "learning_rate": 2.528020777078684e-05,
      "loss": 0.05420032739639282,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0423,
      "step": 22910
    },
    {
      "entropy_step_1": 0.1667,
      "entropy_step_2": 0.1722,
      "entropy_step_3": 0.1589,
      "epoch": 0.6817167841527617,
      "grad_norm": 104.17737579345703,
      "learning_rate": 2.523746967831372e-05,
      "loss": 0.052574056386947635,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0485,
      "loss_step_3": 0.0396,
      "step": 22920
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1968,
      "epoch": 0.6820142173046608,
      "grad_norm": 301.20037841796875,
      "learning_rate": 2.5194755543925607e-05,
      "loss": 0.05702744126319885,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0513,
      "step": 22930
    },
    {
      "entropy_step_1": 0.2143,
      "entropy_step_2": 0.2195,
      "entropy_step_3": 0.2242,
      "epoch": 0.6823116504565598,
      "grad_norm": 120.72415161132812,
      "learning_rate": 2.5152065408949056e-05,
      "loss": 0.06614369750022889,
      "loss_step_1": 0.0639,
      "loss_step_2": 0.0625,
      "loss_step_3": 0.0669,
      "step": 22940
    },
    {
      "entropy_step_1": 0.2055,
      "entropy_step_2": 0.2141,
      "entropy_step_3": 0.2117,
      "epoch": 0.682609083608459,
      "grad_norm": 50.92434310913086,
      "learning_rate": 2.510939931468732e-05,
      "loss": 0.05455536842346191,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0527,
      "step": 22950
    },
    {
      "entropy_step_1": 0.1889,
      "entropy_step_2": 0.1981,
      "entropy_step_3": 0.2015,
      "epoch": 0.6829065167603581,
      "grad_norm": 109.88640594482422,
      "learning_rate": 2.5066757302420436e-05,
      "loss": 0.05677238702774048,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0541,
      "step": 22960
    },
    {
      "entropy_step_1": 0.1696,
      "entropy_step_2": 0.1761,
      "entropy_step_3": 0.1764,
      "epoch": 0.6832039499122572,
      "grad_norm": 39.6287956237793,
      "learning_rate": 2.5024139413405106e-05,
      "loss": 0.05024879574775696,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0468,
      "step": 22970
    },
    {
      "entropy_step_1": 0.1785,
      "entropy_step_2": 0.1837,
      "entropy_step_3": 0.2072,
      "epoch": 0.6835013830641563,
      "grad_norm": 122.56188201904297,
      "learning_rate": 2.4981545688874763e-05,
      "loss": 0.06471734642982482,
      "loss_step_1": 0.0673,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0931,
      "step": 22980
    },
    {
      "entropy_step_1": 0.1884,
      "entropy_step_2": 0.1923,
      "entropy_step_3": 0.1942,
      "epoch": 0.6837988162160554,
      "grad_norm": 40.306602478027344,
      "learning_rate": 2.4938976170039368e-05,
      "loss": 0.06646927595138549,
      "loss_step_1": 0.0688,
      "loss_step_2": 0.0679,
      "loss_step_3": 0.0737,
      "step": 22990
    },
    {
      "entropy_step_1": 0.2119,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.2038,
      "epoch": 0.6840962493679545,
      "grad_norm": 84.67130279541016,
      "learning_rate": 2.489643089808557e-05,
      "loss": 0.06316642761230469,
      "loss_step_1": 0.0657,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.0645,
      "step": 23000
    },
    {
      "epoch": 0.6840962493679545,
      "eval_accuracy": 0.908,
      "eval_f1": 0.9078806132748041,
      "eval_loss": 0.10222207754850388,
      "eval_precision": 0.9101260936695832,
      "eval_recall": 0.908,
      "eval_runtime": 6.5334,
      "eval_samples_per_second": 76.53,
      "eval_steps_per_second": 9.643,
      "step": 23000
    },
    {
      "entropy_step_1": 0.1944,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.1975,
      "epoch": 0.6843936825198537,
      "grad_norm": 176.15130615234375,
      "learning_rate": 2.4853909914176433e-05,
      "loss": 0.053844916820526126,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0486,
      "step": 23010
    },
    {
      "entropy_step_1": 0.1762,
      "entropy_step_2": 0.1824,
      "entropy_step_3": 0.1906,
      "epoch": 0.6846911156717528,
      "grad_norm": 72.4700698852539,
      "learning_rate": 2.4811413259451626e-05,
      "loss": 0.06228358149528503,
      "loss_step_1": 0.0654,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0646,
      "step": 23020
    },
    {
      "entropy_step_1": 0.1799,
      "entropy_step_2": 0.1827,
      "entropy_step_3": 0.172,
      "epoch": 0.6849885488236519,
      "grad_norm": 62.63392639160156,
      "learning_rate": 2.4768940975027234e-05,
      "loss": 0.05291963219642639,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0494,
      "step": 23030
    },
    {
      "entropy_step_1": 0.1979,
      "entropy_step_2": 0.1979,
      "entropy_step_3": 0.1895,
      "epoch": 0.685285981975551,
      "grad_norm": 172.0845489501953,
      "learning_rate": 2.4726493101995762e-05,
      "loss": 0.05816420316696167,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0564,
      "step": 23040
    },
    {
      "entropy_step_1": 0.178,
      "entropy_step_2": 0.1807,
      "entropy_step_3": 0.1864,
      "epoch": 0.6855834151274501,
      "grad_norm": 70.68472290039062,
      "learning_rate": 2.46840696814261e-05,
      "loss": 0.05529614090919495,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0534,
      "step": 23050
    },
    {
      "entropy_step_1": 0.1862,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1774,
      "epoch": 0.6858808482793493,
      "grad_norm": 61.77821731567383,
      "learning_rate": 2.4641670754363504e-05,
      "loss": 0.049790683388710025,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0503,
      "step": 23060
    },
    {
      "entropy_step_1": 0.1839,
      "entropy_step_2": 0.1804,
      "entropy_step_3": 0.1826,
      "epoch": 0.6861782814312484,
      "grad_norm": 81.7649154663086,
      "learning_rate": 2.459929636182951e-05,
      "loss": 0.055150282382965085,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0512,
      "step": 23070
    },
    {
      "entropy_step_1": 0.1709,
      "entropy_step_2": 0.1647,
      "entropy_step_3": 0.1689,
      "epoch": 0.6864757145831475,
      "grad_norm": 61.57117462158203,
      "learning_rate": 2.455694654482189e-05,
      "loss": 0.06451029777526855,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0623,
      "loss_step_3": 0.0623,
      "step": 23080
    },
    {
      "entropy_step_1": 0.1928,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1821,
      "epoch": 0.6867731477350465,
      "grad_norm": 74.61197662353516,
      "learning_rate": 2.451462134431473e-05,
      "loss": 0.05788557529449463,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0595,
      "step": 23090
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.1948,
      "entropy_step_3": 0.2035,
      "epoch": 0.6870705808869456,
      "grad_norm": 103.54224395751953,
      "learning_rate": 2.447232080125816e-05,
      "loss": 0.05656808614730835,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0551,
      "step": 23100
    },
    {
      "entropy_step_1": 0.1825,
      "entropy_step_2": 0.1768,
      "entropy_step_3": 0.1728,
      "epoch": 0.6873680140388447,
      "grad_norm": 65.13648986816406,
      "learning_rate": 2.4430044956578592e-05,
      "loss": 0.050528496503829956,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0468,
      "step": 23110
    },
    {
      "entropy_step_1": 0.1695,
      "entropy_step_2": 0.1639,
      "entropy_step_3": 0.1587,
      "epoch": 0.6876654471907439,
      "grad_norm": 100.65654754638672,
      "learning_rate": 2.438779385117843e-05,
      "loss": 0.05103880167007446,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.048,
      "loss_step_3": 0.046,
      "step": 23120
    },
    {
      "entropy_step_1": 0.2077,
      "entropy_step_2": 0.1996,
      "entropy_step_3": 0.2107,
      "epoch": 0.687962880342643,
      "grad_norm": 83.45304870605469,
      "learning_rate": 2.4345567525936265e-05,
      "loss": 0.059189784526824954,
      "loss_step_1": 0.0721,
      "loss_step_2": 0.0728,
      "loss_step_3": 0.0811,
      "step": 23130
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1843,
      "entropy_step_3": 0.1823,
      "epoch": 0.6882603134945421,
      "grad_norm": 101.66228485107422,
      "learning_rate": 2.430336602170656e-05,
      "loss": 0.05642288327217102,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0512,
      "step": 23140
    },
    {
      "entropy_step_1": 0.2085,
      "entropy_step_2": 0.2057,
      "entropy_step_3": 0.2116,
      "epoch": 0.6885577466464412,
      "grad_norm": 155.68637084960938,
      "learning_rate": 2.4261189379319904e-05,
      "loss": 0.07084101438522339,
      "loss_step_1": 0.083,
      "loss_step_2": 0.0821,
      "loss_step_3": 0.0917,
      "step": 23150
    },
    {
      "entropy_step_1": 0.2021,
      "entropy_step_2": 0.2016,
      "entropy_step_3": 0.1915,
      "epoch": 0.6888551797983403,
      "grad_norm": 86.80657196044922,
      "learning_rate": 2.4219037639582724e-05,
      "loss": 0.05853807330131531,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0474,
      "loss_step_3": 0.0474,
      "step": 23160
    },
    {
      "entropy_step_1": 0.1999,
      "entropy_step_2": 0.1966,
      "entropy_step_3": 0.1976,
      "epoch": 0.6891526129502394,
      "grad_norm": 230.0216827392578,
      "learning_rate": 2.4176910843277457e-05,
      "loss": 0.05795689225196839,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0566,
      "step": 23170
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.1968,
      "entropy_step_3": 0.2052,
      "epoch": 0.6894500461021386,
      "grad_norm": 123.20011138916016,
      "learning_rate": 2.4134809031162314e-05,
      "loss": 0.05506593585014343,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0598,
      "step": 23180
    },
    {
      "entropy_step_1": 0.2017,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.1918,
      "epoch": 0.6897474792540377,
      "grad_norm": 73.11959075927734,
      "learning_rate": 2.409273224397139e-05,
      "loss": 0.06522631645202637,
      "loss_step_1": 0.072,
      "loss_step_2": 0.0714,
      "loss_step_3": 0.0634,
      "step": 23190
    },
    {
      "entropy_step_1": 0.198,
      "entropy_step_2": 0.1933,
      "entropy_step_3": 0.178,
      "epoch": 0.6900449124059368,
      "grad_norm": 80.12405395507812,
      "learning_rate": 2.4050680522414508e-05,
      "loss": 0.06092005968093872,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0464,
      "step": 23200
    },
    {
      "entropy_step_1": 0.2133,
      "entropy_step_2": 0.2092,
      "entropy_step_3": 0.2078,
      "epoch": 0.6903423455578359,
      "grad_norm": 51.933876037597656,
      "learning_rate": 2.4008653907177342e-05,
      "loss": 0.056951552629470825,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0569,
      "step": 23210
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1745,
      "entropy_step_3": 0.184,
      "epoch": 0.690639778709735,
      "grad_norm": 29.662595748901367,
      "learning_rate": 2.3966652438921178e-05,
      "loss": 0.05464853048324585,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0615,
      "step": 23220
    },
    {
      "entropy_step_1": 0.195,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.1966,
      "epoch": 0.6909372118616341,
      "grad_norm": 205.88941955566406,
      "learning_rate": 2.3924676158283017e-05,
      "loss": 0.056998223066329956,
      "loss_step_1": 0.0643,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.062,
      "step": 23230
    },
    {
      "entropy_step_1": 0.1812,
      "entropy_step_2": 0.1787,
      "entropy_step_3": 0.1823,
      "epoch": 0.6912346450135332,
      "grad_norm": 79.86702728271484,
      "learning_rate": 2.388272510587546e-05,
      "loss": 0.05565258860588074,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.047,
      "step": 23240
    },
    {
      "entropy_step_1": 0.2088,
      "entropy_step_2": 0.2048,
      "entropy_step_3": 0.1932,
      "epoch": 0.6915320781654323,
      "grad_norm": 65.60102844238281,
      "learning_rate": 2.384079932228676e-05,
      "loss": 0.05636966824531555,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0456,
      "step": 23250
    },
    {
      "entropy_step_1": 0.1968,
      "entropy_step_2": 0.1947,
      "entropy_step_3": 0.1938,
      "epoch": 0.6918295113173314,
      "grad_norm": 224.942138671875,
      "learning_rate": 2.3798898848080657e-05,
      "loss": 0.0581218421459198,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0613,
      "step": 23260
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1879,
      "entropy_step_3": 0.1762,
      "epoch": 0.6921269444692305,
      "grad_norm": 108.0385513305664,
      "learning_rate": 2.3757023723796446e-05,
      "loss": 0.06420650482177734,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0508,
      "step": 23270
    },
    {
      "entropy_step_1": 0.2079,
      "entropy_step_2": 0.2006,
      "entropy_step_3": 0.1873,
      "epoch": 0.6924243776211296,
      "grad_norm": 244.72256469726562,
      "learning_rate": 2.3715173989948865e-05,
      "loss": 0.05236122012138367,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0481,
      "loss_step_3": 0.0424,
      "step": 23280
    },
    {
      "entropy_step_1": 0.2054,
      "entropy_step_2": 0.1985,
      "entropy_step_3": 0.2113,
      "epoch": 0.6927218107730287,
      "grad_norm": 60.721492767333984,
      "learning_rate": 2.3673349687028085e-05,
      "loss": 0.050927764177322386,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.056,
      "step": 23290
    },
    {
      "entropy_step_1": 0.1586,
      "entropy_step_2": 0.1564,
      "entropy_step_3": 0.1624,
      "epoch": 0.6930192439249279,
      "grad_norm": 891.4352416992188,
      "learning_rate": 2.363155085549973e-05,
      "loss": 0.055005669593811035,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0573,
      "step": 23300
    },
    {
      "entropy_step_1": 0.1855,
      "entropy_step_2": 0.1845,
      "entropy_step_3": 0.1705,
      "epoch": 0.693316677076827,
      "grad_norm": 46.19875717163086,
      "learning_rate": 2.3589777535804697e-05,
      "loss": 0.05574337840080261,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0468,
      "step": 23310
    },
    {
      "entropy_step_1": 0.1915,
      "entropy_step_2": 0.1899,
      "entropy_step_3": 0.1826,
      "epoch": 0.6936141102287261,
      "grad_norm": 94.03810119628906,
      "learning_rate": 2.35480297683593e-05,
      "loss": 0.0558876097202301,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0533,
      "step": 23320
    },
    {
      "entropy_step_1": 0.1954,
      "entropy_step_2": 0.1918,
      "entropy_step_3": 0.1715,
      "epoch": 0.6939115433806252,
      "grad_norm": 725.734130859375,
      "learning_rate": 2.3506307593554995e-05,
      "loss": 0.0621854305267334,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0529,
      "step": 23330
    },
    {
      "entropy_step_1": 0.1903,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1983,
      "epoch": 0.6942089765325243,
      "grad_norm": 185.3500518798828,
      "learning_rate": 2.3464611051758605e-05,
      "loss": 0.056862431764602664,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0585,
      "step": 23340
    },
    {
      "entropy_step_1": 0.2149,
      "entropy_step_2": 0.2113,
      "entropy_step_3": 0.2096,
      "epoch": 0.6945064096844235,
      "grad_norm": 97.02572631835938,
      "learning_rate": 2.3422940183312077e-05,
      "loss": 0.060358017683029175,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0511,
      "step": 23350
    },
    {
      "entropy_step_1": 0.2033,
      "entropy_step_2": 0.1999,
      "entropy_step_3": 0.198,
      "epoch": 0.6948038428363226,
      "grad_norm": 111.40656280517578,
      "learning_rate": 2.3381295028532592e-05,
      "loss": 0.05669351816177368,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0582,
      "step": 23360
    },
    {
      "entropy_step_1": 0.212,
      "entropy_step_2": 0.2088,
      "entropy_step_3": 0.1972,
      "epoch": 0.6951012759882217,
      "grad_norm": 3073.374755859375,
      "learning_rate": 2.3339675627712327e-05,
      "loss": 0.05588808059692383,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0506,
      "step": 23370
    },
    {
      "entropy_step_1": 0.1783,
      "entropy_step_2": 0.1765,
      "entropy_step_3": 0.1818,
      "epoch": 0.6953987091401208,
      "grad_norm": 92.12902069091797,
      "learning_rate": 2.3298082021118677e-05,
      "loss": 0.05662175416946411,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0517,
      "step": 23380
    },
    {
      "entropy_step_1": 0.194,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.1961,
      "epoch": 0.6956961422920198,
      "grad_norm": 246.09169006347656,
      "learning_rate": 2.3256514248993986e-05,
      "loss": 0.06098656058311462,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.065,
      "step": 23390
    },
    {
      "entropy_step_1": 0.217,
      "entropy_step_2": 0.2107,
      "entropy_step_3": 0.2112,
      "epoch": 0.6959935754439189,
      "grad_norm": 107.54694366455078,
      "learning_rate": 2.321497235155568e-05,
      "loss": 0.06477683782577515,
      "loss_step_1": 0.0644,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.063,
      "step": 23400
    },
    {
      "entropy_step_1": 0.1881,
      "entropy_step_2": 0.1839,
      "entropy_step_3": 0.2004,
      "epoch": 0.696291008595818,
      "grad_norm": 51.19167709350586,
      "learning_rate": 2.3173456368996084e-05,
      "loss": 0.05451714396476746,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0653,
      "step": 23410
    },
    {
      "entropy_step_1": 0.176,
      "entropy_step_2": 0.1725,
      "entropy_step_3": 0.1686,
      "epoch": 0.6965884417477172,
      "grad_norm": 94.12152099609375,
      "learning_rate": 2.3131966341482476e-05,
      "loss": 0.053905165195465087,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.052,
      "loss_step_3": 0.0506,
      "step": 23420
    },
    {
      "entropy_step_1": 0.1614,
      "entropy_step_2": 0.1615,
      "entropy_step_3": 0.1574,
      "epoch": 0.6968858748996163,
      "grad_norm": 208.7970428466797,
      "learning_rate": 2.3090502309157025e-05,
      "loss": 0.04607577919960022,
      "loss_step_1": 0.0474,
      "loss_step_2": 0.0467,
      "loss_step_3": 0.0483,
      "step": 23430
    },
    {
      "entropy_step_1": 0.1916,
      "entropy_step_2": 0.1926,
      "entropy_step_3": 0.1886,
      "epoch": 0.6971833080515154,
      "grad_norm": 98.48442077636719,
      "learning_rate": 2.304906431213673e-05,
      "loss": 0.06005974411964417,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.053,
      "step": 23440
    },
    {
      "entropy_step_1": 0.1777,
      "entropy_step_2": 0.1783,
      "entropy_step_3": 0.1713,
      "epoch": 0.6974807412034145,
      "grad_norm": 53.30849838256836,
      "learning_rate": 2.300765239051345e-05,
      "loss": 0.05700350999832153,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0555,
      "step": 23450
    },
    {
      "entropy_step_1": 0.1882,
      "entropy_step_2": 0.1873,
      "entropy_step_3": 0.1937,
      "epoch": 0.6977781743553136,
      "grad_norm": 69.30826568603516,
      "learning_rate": 2.2966266584353774e-05,
      "loss": 0.05629355907440185,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.066,
      "step": 23460
    },
    {
      "entropy_step_1": 0.1922,
      "entropy_step_2": 0.1902,
      "entropy_step_3": 0.1902,
      "epoch": 0.6980756075072128,
      "grad_norm": 76.15665435791016,
      "learning_rate": 2.292490693369902e-05,
      "loss": 0.0549294114112854,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0595,
      "step": 23470
    },
    {
      "entropy_step_1": 0.1768,
      "entropy_step_2": 0.1764,
      "entropy_step_3": 0.1808,
      "epoch": 0.6983730406591119,
      "grad_norm": 64.6866455078125,
      "learning_rate": 2.2883573478565194e-05,
      "loss": 0.05042150616645813,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0609,
      "step": 23480
    },
    {
      "entropy_step_1": 0.1795,
      "entropy_step_2": 0.1794,
      "entropy_step_3": 0.1751,
      "epoch": 0.698670473811011,
      "grad_norm": 69.34090423583984,
      "learning_rate": 2.2842266258943024e-05,
      "loss": 0.05556519031524658,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0545,
      "step": 23490
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1873,
      "entropy_step_3": 0.1824,
      "epoch": 0.6989679069629101,
      "grad_norm": 127.0266342163086,
      "learning_rate": 2.280098531479778e-05,
      "loss": 0.05644726753234863,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0466,
      "step": 23500
    },
    {
      "entropy_step_1": 0.1848,
      "entropy_step_2": 0.1845,
      "entropy_step_3": 0.1879,
      "epoch": 0.6992653401148092,
      "grad_norm": 273.0392761230469,
      "learning_rate": 2.2759730686069342e-05,
      "loss": 0.05399341583251953,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0553,
      "step": 23510
    },
    {
      "entropy_step_1": 0.1996,
      "entropy_step_2": 0.1987,
      "entropy_step_3": 0.1787,
      "epoch": 0.6995627732667083,
      "grad_norm": 105.09366607666016,
      "learning_rate": 2.2718502412672106e-05,
      "loss": 0.05433200597763062,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0454,
      "step": 23520
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.194,
      "entropy_step_3": 0.1989,
      "epoch": 0.6998602064186075,
      "grad_norm": 73.98827362060547,
      "learning_rate": 2.2677300534495022e-05,
      "loss": 0.05384817123413086,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0582,
      "step": 23530
    },
    {
      "entropy_step_1": 0.1763,
      "entropy_step_2": 0.173,
      "entropy_step_3": 0.1895,
      "epoch": 0.7001576395705065,
      "grad_norm": 51.58806228637695,
      "learning_rate": 2.2636125091401428e-05,
      "loss": 0.054679489135742186,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0634,
      "step": 23540
    },
    {
      "entropy_step_1": 0.203,
      "entropy_step_2": 0.1991,
      "entropy_step_3": 0.1868,
      "epoch": 0.7004550727224056,
      "grad_norm": 103.58052062988281,
      "learning_rate": 2.259497612322919e-05,
      "loss": 0.054370170831680296,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0453,
      "step": 23550
    },
    {
      "entropy_step_1": 0.185,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1851,
      "epoch": 0.7007525058743047,
      "grad_norm": 118.88732147216797,
      "learning_rate": 2.2553853669790408e-05,
      "loss": 0.05710299015045166,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0536,
      "step": 23560
    },
    {
      "entropy_step_1": 0.2034,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.1811,
      "epoch": 0.7010499390262038,
      "grad_norm": 52.81699752807617,
      "learning_rate": 2.2512757770871683e-05,
      "loss": 0.062327778339385985,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0568,
      "step": 23570
    },
    {
      "entropy_step_1": 0.1677,
      "entropy_step_2": 0.158,
      "entropy_step_3": 0.1594,
      "epoch": 0.701347372178103,
      "grad_norm": 176.03053283691406,
      "learning_rate": 2.247168846623383e-05,
      "loss": 0.045443907380104065,
      "loss_step_1": 0.0409,
      "loss_step_2": 0.0402,
      "loss_step_3": 0.0393,
      "step": 23580
    },
    {
      "entropy_step_1": 0.1753,
      "entropy_step_2": 0.1653,
      "entropy_step_3": 0.1668,
      "epoch": 0.7016448053300021,
      "grad_norm": 124.26058959960938,
      "learning_rate": 2.2430645795611964e-05,
      "loss": 0.05486435294151306,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0533,
      "step": 23590
    },
    {
      "entropy_step_1": 0.1881,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.195,
      "epoch": 0.7019422384819012,
      "grad_norm": 135.30287170410156,
      "learning_rate": 2.2389629798715412e-05,
      "loss": 0.05321797728538513,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0606,
      "step": 23600
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.176,
      "entropy_step_3": 0.1786,
      "epoch": 0.7022396716338003,
      "grad_norm": 61.164886474609375,
      "learning_rate": 2.2348640515227737e-05,
      "loss": 0.05104342699050903,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0627,
      "step": 23610
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1735,
      "entropy_step_3": 0.1748,
      "epoch": 0.7025371047856994,
      "grad_norm": 197.11439514160156,
      "learning_rate": 2.2307677984806617e-05,
      "loss": 0.052069085836410525,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.048,
      "step": 23620
    },
    {
      "entropy_step_1": 0.1868,
      "entropy_step_2": 0.1834,
      "entropy_step_3": 0.1908,
      "epoch": 0.7028345379375985,
      "grad_norm": 154.45977783203125,
      "learning_rate": 2.2266742247083834e-05,
      "loss": 0.0631567358970642,
      "loss_step_1": 0.0698,
      "loss_step_2": 0.0692,
      "loss_step_3": 0.0755,
      "step": 23630
    },
    {
      "entropy_step_1": 0.2136,
      "entropy_step_2": 0.2069,
      "entropy_step_3": 0.1873,
      "epoch": 0.7031319710894977,
      "grad_norm": 64.42179870605469,
      "learning_rate": 2.2225833341665332e-05,
      "loss": 0.061726218461990355,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0449,
      "step": 23640
    },
    {
      "entropy_step_1": 0.1999,
      "entropy_step_2": 0.1958,
      "entropy_step_3": 0.202,
      "epoch": 0.7034294042413968,
      "grad_norm": 92.22738647460938,
      "learning_rate": 2.2184951308130957e-05,
      "loss": 0.05380958914756775,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0624,
      "step": 23650
    },
    {
      "entropy_step_1": 0.1916,
      "entropy_step_2": 0.1897,
      "entropy_step_3": 0.1926,
      "epoch": 0.7037268373932959,
      "grad_norm": 162.51646423339844,
      "learning_rate": 2.2144096186034686e-05,
      "loss": 0.05568719506263733,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0629,
      "step": 23660
    },
    {
      "entropy_step_1": 0.1705,
      "entropy_step_2": 0.1673,
      "entropy_step_3": 0.18,
      "epoch": 0.704024270545195,
      "grad_norm": 41.96104431152344,
      "learning_rate": 2.2103268014904373e-05,
      "loss": 0.04896848201751709,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0516,
      "step": 23670
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1882,
      "entropy_step_3": 0.196,
      "epoch": 0.7043217036970941,
      "grad_norm": 66.20469665527344,
      "learning_rate": 2.2062466834241863e-05,
      "loss": 0.055636155605316165,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.061,
      "step": 23680
    },
    {
      "entropy_step_1": 0.2102,
      "entropy_step_2": 0.2049,
      "entropy_step_3": 0.2217,
      "epoch": 0.7046191368489932,
      "grad_norm": 92.62857055664062,
      "learning_rate": 2.2021692683522836e-05,
      "loss": 0.06416250467300415,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.06,
      "loss_step_3": 0.0655,
      "step": 23690
    },
    {
      "entropy_step_1": 0.1945,
      "entropy_step_2": 0.1879,
      "entropy_step_3": 0.1927,
      "epoch": 0.7049165700008923,
      "grad_norm": 168.03298950195312,
      "learning_rate": 2.198094560219684e-05,
      "loss": 0.054973697662353514,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0537,
      "step": 23700
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.1768,
      "entropy_step_3": 0.1734,
      "epoch": 0.7052140031527914,
      "grad_norm": 73.98169708251953,
      "learning_rate": 2.1940225629687212e-05,
      "loss": 0.05474941730499268,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0497,
      "loss_step_3": 0.0502,
      "step": 23710
    },
    {
      "entropy_step_1": 0.1975,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.1969,
      "epoch": 0.7055114363046905,
      "grad_norm": 63.94557189941406,
      "learning_rate": 2.189953280539111e-05,
      "loss": 0.06320036053657532,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0615,
      "step": 23720
    },
    {
      "entropy_step_1": 0.1983,
      "entropy_step_2": 0.1909,
      "entropy_step_3": 0.2122,
      "epoch": 0.7058088694565896,
      "grad_norm": 124.68968200683594,
      "learning_rate": 2.1858867168679383e-05,
      "loss": 0.05711124539375305,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0664,
      "step": 23730
    },
    {
      "entropy_step_1": 0.2066,
      "entropy_step_2": 0.1998,
      "entropy_step_3": 0.1833,
      "epoch": 0.7061063026084887,
      "grad_norm": 276.95965576171875,
      "learning_rate": 2.181822875889658e-05,
      "loss": 0.05831743478775024,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0537,
      "step": 23740
    },
    {
      "entropy_step_1": 0.1944,
      "entropy_step_2": 0.1888,
      "entropy_step_3": 0.1863,
      "epoch": 0.7064037357603878,
      "grad_norm": 87.88914489746094,
      "learning_rate": 2.1777617615360908e-05,
      "loss": 0.05614701509475708,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0581,
      "step": 23750
    },
    {
      "entropy_step_1": 0.1795,
      "entropy_step_2": 0.1766,
      "entropy_step_3": 0.1651,
      "epoch": 0.706701168912287,
      "grad_norm": 97.35838317871094,
      "learning_rate": 2.173703377736423e-05,
      "loss": 0.05545757412910461,
      "loss_step_1": 0.0461,
      "loss_step_2": 0.0454,
      "loss_step_3": 0.0398,
      "step": 23760
    },
    {
      "entropy_step_1": 0.1826,
      "entropy_step_2": 0.1785,
      "entropy_step_3": 0.185,
      "epoch": 0.7069986020641861,
      "grad_norm": 41.44435501098633,
      "learning_rate": 2.1696477284171953e-05,
      "loss": 0.05790496468544006,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0592,
      "step": 23770
    },
    {
      "entropy_step_1": 0.1753,
      "entropy_step_2": 0.1711,
      "entropy_step_3": 0.1809,
      "epoch": 0.7072960352160852,
      "grad_norm": 76.01248168945312,
      "learning_rate": 2.1655948175023023e-05,
      "loss": 0.050975024700164795,
      "loss_step_1": 0.0503,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0451,
      "step": 23780
    },
    {
      "entropy_step_1": 0.1898,
      "entropy_step_2": 0.1864,
      "entropy_step_3": 0.1979,
      "epoch": 0.7075934683679843,
      "grad_norm": 65.09829711914062,
      "learning_rate": 2.1615446489129908e-05,
      "loss": 0.054419207572937014,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0611,
      "step": 23790
    },
    {
      "entropy_step_1": 0.1957,
      "entropy_step_2": 0.1925,
      "entropy_step_3": 0.1907,
      "epoch": 0.7078909015198834,
      "grad_norm": 69.6996841430664,
      "learning_rate": 2.1574972265678527e-05,
      "loss": 0.06036187410354614,
      "loss_step_1": 0.0623,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0642,
      "step": 23800
    },
    {
      "entropy_step_1": 0.2079,
      "entropy_step_2": 0.2046,
      "entropy_step_3": 0.216,
      "epoch": 0.7081883346717825,
      "grad_norm": 43.417022705078125,
      "learning_rate": 2.1534525543828278e-05,
      "loss": 0.0544048547744751,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0574,
      "step": 23810
    },
    {
      "entropy_step_1": 0.1649,
      "entropy_step_2": 0.1653,
      "entropy_step_3": 0.1611,
      "epoch": 0.7084857678236817,
      "grad_norm": 64.17594909667969,
      "learning_rate": 2.149410636271189e-05,
      "loss": 0.05304552316665649,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.056,
      "step": 23820
    },
    {
      "entropy_step_1": 0.1856,
      "entropy_step_2": 0.1853,
      "entropy_step_3": 0.2073,
      "epoch": 0.7087832009755808,
      "grad_norm": 138.0419464111328,
      "learning_rate": 2.145371476143548e-05,
      "loss": 0.05891075134277344,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0559,
      "step": 23830
    },
    {
      "entropy_step_1": 0.1643,
      "entropy_step_2": 0.1617,
      "entropy_step_3": 0.1759,
      "epoch": 0.7090806341274799,
      "grad_norm": 87.70024871826172,
      "learning_rate": 2.141335077907844e-05,
      "loss": 0.05101536512374878,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0633,
      "step": 23840
    },
    {
      "entropy_step_1": 0.177,
      "entropy_step_2": 0.1706,
      "entropy_step_3": 0.1638,
      "epoch": 0.7093780672793789,
      "grad_norm": 81.43804931640625,
      "learning_rate": 2.1373014454693525e-05,
      "loss": 0.0527552604675293,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0439,
      "step": 23850
    },
    {
      "entropy_step_1": 0.1965,
      "entropy_step_2": 0.1893,
      "entropy_step_3": 0.2029,
      "epoch": 0.709675500431278,
      "grad_norm": 196.4976806640625,
      "learning_rate": 2.133270582730663e-05,
      "loss": 0.0537581741809845,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0586,
      "step": 23860
    },
    {
      "entropy_step_1": 0.1847,
      "entropy_step_2": 0.179,
      "entropy_step_3": 0.161,
      "epoch": 0.7099729335831771,
      "grad_norm": 120.3901596069336,
      "learning_rate": 2.129242493591696e-05,
      "loss": 0.05556259751319885,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0492,
      "step": 23870
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.2025,
      "epoch": 0.7102703667350763,
      "grad_norm": 44.76449966430664,
      "learning_rate": 2.1252171819496753e-05,
      "loss": 0.05805350542068481,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0722,
      "step": 23880
    },
    {
      "entropy_step_1": 0.1747,
      "entropy_step_2": 0.1704,
      "entropy_step_3": 0.1693,
      "epoch": 0.7105677998869754,
      "grad_norm": 127.85621643066406,
      "learning_rate": 2.1211946516991494e-05,
      "loss": 0.0541381299495697,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0434,
      "step": 23890
    },
    {
      "entropy_step_1": 0.1777,
      "entropy_step_2": 0.1746,
      "entropy_step_3": 0.1951,
      "epoch": 0.7108652330388745,
      "grad_norm": 76.5628662109375,
      "learning_rate": 2.1171749067319685e-05,
      "loss": 0.055712330341339114,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0634,
      "step": 23900
    },
    {
      "entropy_step_1": 0.1911,
      "entropy_step_2": 0.1875,
      "entropy_step_3": 0.1945,
      "epoch": 0.7111626661907736,
      "grad_norm": 66.13215637207031,
      "learning_rate": 2.113157950937295e-05,
      "loss": 0.057327330112457275,
      "loss_step_1": 0.047,
      "loss_step_2": 0.0474,
      "loss_step_3": 0.0463,
      "step": 23910
    },
    {
      "entropy_step_1": 0.214,
      "entropy_step_2": 0.2062,
      "entropy_step_3": 0.2007,
      "epoch": 0.7114600993426727,
      "grad_norm": 43.10114669799805,
      "learning_rate": 2.109143788201582e-05,
      "loss": 0.05616176128387451,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0441,
      "step": 23920
    },
    {
      "entropy_step_1": 0.2044,
      "entropy_step_2": 0.1955,
      "entropy_step_3": 0.2024,
      "epoch": 0.7117575324945719,
      "grad_norm": 215.2279510498047,
      "learning_rate": 2.1051324224085916e-05,
      "loss": 0.0600796103477478,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0628,
      "step": 23930
    },
    {
      "entropy_step_1": 0.1919,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1923,
      "epoch": 0.712054965646471,
      "grad_norm": 62.033058166503906,
      "learning_rate": 2.1011238574393706e-05,
      "loss": 0.05077669620513916,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0472,
      "step": 23940
    },
    {
      "entropy_step_1": 0.1712,
      "entropy_step_2": 0.1652,
      "entropy_step_3": 0.1725,
      "epoch": 0.7123523987983701,
      "grad_norm": 117.54737091064453,
      "learning_rate": 2.0971180971722647e-05,
      "loss": 0.04749472141265869,
      "loss_step_1": 0.0436,
      "loss_step_2": 0.0434,
      "loss_step_3": 0.0442,
      "step": 23950
    },
    {
      "entropy_step_1": 0.1814,
      "entropy_step_2": 0.175,
      "entropy_step_3": 0.1892,
      "epoch": 0.7126498319502692,
      "grad_norm": 155.38702392578125,
      "learning_rate": 2.093115145482899e-05,
      "loss": 0.05654044151306152,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0591,
      "step": 23960
    },
    {
      "entropy_step_1": 0.1993,
      "entropy_step_2": 0.1946,
      "entropy_step_3": 0.1999,
      "epoch": 0.7129472651021683,
      "grad_norm": 102.32942962646484,
      "learning_rate": 2.089115006244184e-05,
      "loss": 0.056354820728302,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.054,
      "step": 23970
    },
    {
      "entropy_step_1": 0.1844,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.1752,
      "epoch": 0.7132446982540674,
      "grad_norm": 261.42071533203125,
      "learning_rate": 2.0851176833263087e-05,
      "loss": 0.0534764289855957,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0519,
      "step": 23980
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.1791,
      "entropy_step_3": 0.1871,
      "epoch": 0.7135421314059666,
      "grad_norm": 74.93724060058594,
      "learning_rate": 2.0811231805967356e-05,
      "loss": 0.05304608941078186,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0495,
      "step": 23990
    },
    {
      "entropy_step_1": 0.1828,
      "entropy_step_2": 0.1809,
      "entropy_step_3": 0.1746,
      "epoch": 0.7138395645578656,
      "grad_norm": 369.4526672363281,
      "learning_rate": 2.0771315019202044e-05,
      "loss": 0.05821476578712463,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0495,
      "step": 24000
    },
    {
      "epoch": 0.7138395645578656,
      "eval_accuracy": 0.909,
      "eval_f1": 0.9088883882756377,
      "eval_loss": 0.09460483491420746,
      "eval_precision": 0.9110139684453825,
      "eval_recall": 0.909,
      "eval_runtime": 6.6738,
      "eval_samples_per_second": 74.92,
      "eval_steps_per_second": 9.44,
      "step": 24000
    },
    {
      "entropy_step_1": 0.1717,
      "entropy_step_2": 0.1715,
      "entropy_step_3": 0.1686,
      "epoch": 0.7141369977097647,
      "grad_norm": 94.78185272216797,
      "learning_rate": 2.073142651158716e-05,
      "loss": 0.05256630182266235,
      "loss_step_1": 0.0476,
      "loss_step_2": 0.0469,
      "loss_step_3": 0.0511,
      "step": 24010
    },
    {
      "entropy_step_1": 0.1821,
      "entropy_step_2": 0.1827,
      "entropy_step_3": 0.176,
      "epoch": 0.7144344308616638,
      "grad_norm": 40.54111099243164,
      "learning_rate": 2.069156632171539e-05,
      "loss": 0.05368496179580688,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0641,
      "step": 24020
    },
    {
      "entropy_step_1": 0.1978,
      "entropy_step_2": 0.1969,
      "entropy_step_3": 0.23,
      "epoch": 0.7147318640135629,
      "grad_norm": 47.91811752319336,
      "learning_rate": 2.065173448815198e-05,
      "loss": 0.056087535619735715,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0689,
      "step": 24030
    },
    {
      "entropy_step_1": 0.1875,
      "entropy_step_2": 0.189,
      "entropy_step_3": 0.186,
      "epoch": 0.715029297165462,
      "grad_norm": 71.36123657226562,
      "learning_rate": 2.0611931049434814e-05,
      "loss": 0.0544248104095459,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0631,
      "step": 24040
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1767,
      "entropy_step_3": 0.1882,
      "epoch": 0.7153267303173612,
      "grad_norm": 232.13389587402344,
      "learning_rate": 2.0572156044074237e-05,
      "loss": 0.061617374420166016,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0675,
      "step": 24050
    },
    {
      "entropy_step_1": 0.1771,
      "entropy_step_2": 0.1798,
      "entropy_step_3": 0.1959,
      "epoch": 0.7156241634692603,
      "grad_norm": 66.84609985351562,
      "learning_rate": 2.0532409510553112e-05,
      "loss": 0.05289684534072876,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0549,
      "step": 24060
    },
    {
      "entropy_step_1": 0.1623,
      "entropy_step_2": 0.1649,
      "entropy_step_3": 0.1569,
      "epoch": 0.7159215966211594,
      "grad_norm": 110.82044219970703,
      "learning_rate": 2.049269148732673e-05,
      "loss": 0.04570325613021851,
      "loss_step_1": 0.0441,
      "loss_step_2": 0.0436,
      "loss_step_3": 0.0366,
      "step": 24070
    },
    {
      "entropy_step_1": 0.1788,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.1737,
      "epoch": 0.7162190297730585,
      "grad_norm": 95.44224548339844,
      "learning_rate": 2.0453002012822863e-05,
      "loss": 0.053427159786224365,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0492,
      "step": 24080
    },
    {
      "entropy_step_1": 0.1786,
      "entropy_step_2": 0.1775,
      "entropy_step_3": 0.1719,
      "epoch": 0.7165164629249576,
      "grad_norm": 105.59371948242188,
      "learning_rate": 2.041334112544157e-05,
      "loss": 0.04885632991790771,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0513,
      "step": 24090
    },
    {
      "entropy_step_1": 0.1837,
      "entropy_step_2": 0.1836,
      "entropy_step_3": 0.189,
      "epoch": 0.7168138960768567,
      "grad_norm": 180.6376953125,
      "learning_rate": 2.0373708863555365e-05,
      "loss": 0.06099672913551331,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0649,
      "step": 24100
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.212,
      "epoch": 0.7171113292287559,
      "grad_norm": 79.78173828125,
      "learning_rate": 2.0334105265508918e-05,
      "loss": 0.06359982490539551,
      "loss_step_1": 0.0705,
      "loss_step_2": 0.07,
      "loss_step_3": 0.0784,
      "step": 24110
    },
    {
      "entropy_step_1": 0.2168,
      "entropy_step_2": 0.2159,
      "entropy_step_3": 0.2111,
      "epoch": 0.717408762380655,
      "grad_norm": 146.3166046142578,
      "learning_rate": 2.0294530369619312e-05,
      "loss": 0.05487605929374695,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.054,
      "step": 24120
    },
    {
      "entropy_step_1": 0.2026,
      "entropy_step_2": 0.2036,
      "entropy_step_3": 0.1919,
      "epoch": 0.7177061955325541,
      "grad_norm": 235.3155517578125,
      "learning_rate": 2.0254984214175772e-05,
      "loss": 0.05210350751876831,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0537,
      "step": 24130
    },
    {
      "entropy_step_1": 0.1823,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1929,
      "epoch": 0.7180036286844532,
      "grad_norm": 136.30233764648438,
      "learning_rate": 2.0215466837439744e-05,
      "loss": 0.06333529353141784,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0639,
      "step": 24140
    },
    {
      "entropy_step_1": 0.2001,
      "entropy_step_2": 0.2021,
      "entropy_step_3": 0.2095,
      "epoch": 0.7183010618363522,
      "grad_norm": 69.13755798339844,
      "learning_rate": 2.0175978277644808e-05,
      "loss": 0.05876409411430359,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0639,
      "step": 24150
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1913,
      "entropy_step_3": 0.1999,
      "epoch": 0.7185984949882513,
      "grad_norm": 197.47764587402344,
      "learning_rate": 2.0136518572996727e-05,
      "loss": 0.05979647040367127,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0636,
      "step": 24160
    },
    {
      "entropy_step_1": 0.1921,
      "entropy_step_2": 0.1944,
      "entropy_step_3": 0.1877,
      "epoch": 0.7188959281401505,
      "grad_norm": 140.27528381347656,
      "learning_rate": 2.0097087761673272e-05,
      "loss": 0.053825384378433226,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0478,
      "step": 24170
    },
    {
      "entropy_step_1": 0.1771,
      "entropy_step_2": 0.1788,
      "entropy_step_3": 0.1924,
      "epoch": 0.7191933612920496,
      "grad_norm": 113.43978881835938,
      "learning_rate": 2.005768588182428e-05,
      "loss": 0.055615711212158206,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0642,
      "step": 24180
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.1839,
      "entropy_step_3": 0.1828,
      "epoch": 0.7194907944439487,
      "grad_norm": 73.97344970703125,
      "learning_rate": 2.0018312971571657e-05,
      "loss": 0.054335886240005495,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0532,
      "step": 24190
    },
    {
      "entropy_step_1": 0.181,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.2031,
      "epoch": 0.7197882275958478,
      "grad_norm": 205.99325561523438,
      "learning_rate": 1.9978969069009158e-05,
      "loss": 0.05460028648376465,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0727,
      "step": 24200
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1865,
      "entropy_step_3": 0.1989,
      "epoch": 0.7200856607477469,
      "grad_norm": 115.69854736328125,
      "learning_rate": 1.9939654212202596e-05,
      "loss": 0.04976097643375397,
      "loss_step_1": 0.0474,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0429,
      "step": 24210
    },
    {
      "entropy_step_1": 0.2011,
      "entropy_step_2": 0.2047,
      "entropy_step_3": 0.2003,
      "epoch": 0.720383093899646,
      "grad_norm": 140.18838500976562,
      "learning_rate": 1.9900368439189598e-05,
      "loss": 0.06185164451599121,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0622,
      "loss_step_3": 0.0573,
      "step": 24220
    },
    {
      "entropy_step_1": 0.1843,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1777,
      "epoch": 0.7206805270515452,
      "grad_norm": 92.15230560302734,
      "learning_rate": 1.986111178797972e-05,
      "loss": 0.05877179503440857,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0477,
      "step": 24230
    },
    {
      "entropy_step_1": 0.1817,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.1883,
      "epoch": 0.7209779602034443,
      "grad_norm": 105.06607818603516,
      "learning_rate": 1.9821884296554282e-05,
      "loss": 0.06063199043273926,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0561,
      "step": 24240
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1785,
      "entropy_step_3": 0.1809,
      "epoch": 0.7212753933553434,
      "grad_norm": 86.77295684814453,
      "learning_rate": 1.9782686002866425e-05,
      "loss": 0.049558022618293764,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0535,
      "step": 24250
    },
    {
      "entropy_step_1": 0.1752,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1844,
      "epoch": 0.7215728265072425,
      "grad_norm": 36.26230239868164,
      "learning_rate": 1.9743516944841006e-05,
      "loss": 0.05250606536865234,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0624,
      "step": 24260
    },
    {
      "entropy_step_1": 0.1877,
      "entropy_step_2": 0.1906,
      "entropy_step_3": 0.1923,
      "epoch": 0.7218702596591416,
      "grad_norm": 132.33863830566406,
      "learning_rate": 1.970437716037466e-05,
      "loss": 0.0620689332485199,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0548,
      "step": 24270
    },
    {
      "entropy_step_1": 0.1876,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.1941,
      "epoch": 0.7221676928110408,
      "grad_norm": 90.27806091308594,
      "learning_rate": 1.9665266687335642e-05,
      "loss": 0.05860249996185303,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0556,
      "step": 24280
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1811,
      "entropy_step_3": 0.1781,
      "epoch": 0.7224651259629399,
      "grad_norm": 33.95071029663086,
      "learning_rate": 1.9626185563563865e-05,
      "loss": 0.04854942262172699,
      "loss_step_1": 0.0477,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0414,
      "step": 24290
    },
    {
      "entropy_step_1": 0.1647,
      "entropy_step_2": 0.1659,
      "entropy_step_3": 0.1558,
      "epoch": 0.7227625591148389,
      "grad_norm": 117.21508026123047,
      "learning_rate": 1.958713382687084e-05,
      "loss": 0.0634238600730896,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.062,
      "step": 24300
    },
    {
      "entropy_step_1": 0.1965,
      "entropy_step_2": 0.1975,
      "entropy_step_3": 0.1945,
      "epoch": 0.723059992266738,
      "grad_norm": 115.01028442382812,
      "learning_rate": 1.9548111515039675e-05,
      "loss": 0.046435761451721194,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0542,
      "step": 24310
    },
    {
      "entropy_step_1": 0.1732,
      "entropy_step_2": 0.175,
      "entropy_step_3": 0.1834,
      "epoch": 0.7233574254186371,
      "grad_norm": 108.55249786376953,
      "learning_rate": 1.950911866582498e-05,
      "loss": 0.05553098320960999,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0692,
      "step": 24320
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2057,
      "entropy_step_3": 0.2341,
      "epoch": 0.7236548585705362,
      "grad_norm": 120.23101043701172,
      "learning_rate": 1.9470155316952863e-05,
      "loss": 0.057425510883331296,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0725,
      "step": 24330
    },
    {
      "entropy_step_1": 0.2047,
      "entropy_step_2": 0.2074,
      "entropy_step_3": 0.2211,
      "epoch": 0.7239522917224354,
      "grad_norm": 147.35003662109375,
      "learning_rate": 1.94312215061209e-05,
      "loss": 0.07296893000602722,
      "loss_step_1": 0.0802,
      "loss_step_2": 0.0795,
      "loss_step_3": 0.0923,
      "step": 24340
    },
    {
      "entropy_step_1": 0.2035,
      "entropy_step_2": 0.2091,
      "entropy_step_3": 0.2016,
      "epoch": 0.7242497248743345,
      "grad_norm": 81.24919128417969,
      "learning_rate": 1.939231727099806e-05,
      "loss": 0.050601357221603395,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0527,
      "loss_step_3": 0.0464,
      "step": 24350
    },
    {
      "entropy_step_1": 0.186,
      "entropy_step_2": 0.1914,
      "entropy_step_3": 0.1843,
      "epoch": 0.7245471580262336,
      "grad_norm": 67.76310729980469,
      "learning_rate": 1.9353442649224752e-05,
      "loss": 0.059068959951400754,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0521,
      "step": 24360
    },
    {
      "entropy_step_1": 0.1665,
      "entropy_step_2": 0.1708,
      "entropy_step_3": 0.1762,
      "epoch": 0.7248445911781327,
      "grad_norm": 114.29119873046875,
      "learning_rate": 1.9314597678412687e-05,
      "loss": 0.05655400156974792,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.0578,
      "step": 24370
    },
    {
      "entropy_step_1": 0.1923,
      "entropy_step_2": 0.1924,
      "entropy_step_3": 0.1938,
      "epoch": 0.7251420243300318,
      "grad_norm": 67.81736755371094,
      "learning_rate": 1.927578239614491e-05,
      "loss": 0.06119224429130554,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0601,
      "step": 24380
    },
    {
      "entropy_step_1": 0.2141,
      "entropy_step_2": 0.2115,
      "entropy_step_3": 0.2209,
      "epoch": 0.725439457481931,
      "grad_norm": 45.466800689697266,
      "learning_rate": 1.9236996839975707e-05,
      "loss": 0.05099594593048096,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0666,
      "step": 24390
    },
    {
      "entropy_step_1": 0.2135,
      "entropy_step_2": 0.2164,
      "entropy_step_3": 0.2098,
      "epoch": 0.7257368906338301,
      "grad_norm": 102.03018951416016,
      "learning_rate": 1.919824104743067e-05,
      "loss": 0.05949515104293823,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0624,
      "step": 24400
    },
    {
      "entropy_step_1": 0.1744,
      "entropy_step_2": 0.1842,
      "entropy_step_3": 0.1923,
      "epoch": 0.7260343237857292,
      "grad_norm": 51.292884826660156,
      "learning_rate": 1.9159515056006523e-05,
      "loss": 0.05496042370796204,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0593,
      "step": 24410
    },
    {
      "entropy_step_1": 0.1815,
      "entropy_step_2": 0.1939,
      "entropy_step_3": 0.1773,
      "epoch": 0.7263317569376283,
      "grad_norm": 169.91603088378906,
      "learning_rate": 1.9120818903171232e-05,
      "loss": 0.05268785357475281,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.0505,
      "loss_step_3": 0.0452,
      "step": 24420
    },
    {
      "entropy_step_1": 0.1494,
      "entropy_step_2": 0.1659,
      "entropy_step_3": 0.1699,
      "epoch": 0.7266291900895274,
      "grad_norm": 337.00372314453125,
      "learning_rate": 1.90821526263638e-05,
      "loss": 0.05095857381820679,
      "loss_step_1": 0.0412,
      "loss_step_2": 0.0401,
      "loss_step_3": 0.0368,
      "step": 24430
    },
    {
      "entropy_step_1": 0.1788,
      "entropy_step_2": 0.1919,
      "entropy_step_3": 0.1801,
      "epoch": 0.7269266232414265,
      "grad_norm": 88.95618438720703,
      "learning_rate": 1.904351626299441e-05,
      "loss": 0.05209163427352905,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0486,
      "step": 24440
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1916,
      "entropy_step_3": 0.1974,
      "epoch": 0.7272240563933255,
      "grad_norm": 40.38401794433594,
      "learning_rate": 1.900490985044424e-05,
      "loss": 0.059921085834503174,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0539,
      "step": 24450
    },
    {
      "entropy_step_1": 0.197,
      "entropy_step_2": 0.2043,
      "entropy_step_3": 0.2209,
      "epoch": 0.7275214895452247,
      "grad_norm": 84.50074005126953,
      "learning_rate": 1.896633342606558e-05,
      "loss": 0.05696706771850586,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0601,
      "step": 24460
    },
    {
      "entropy_step_1": 0.1705,
      "entropy_step_2": 0.1757,
      "entropy_step_3": 0.1689,
      "epoch": 0.7278189226971238,
      "grad_norm": 59.01376724243164,
      "learning_rate": 1.892778702718157e-05,
      "loss": 0.050771623849868774,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0488,
      "step": 24470
    },
    {
      "entropy_step_1": 0.2011,
      "entropy_step_2": 0.2053,
      "entropy_step_3": 0.1956,
      "epoch": 0.7281163558490229,
      "grad_norm": 69.38468170166016,
      "learning_rate": 1.8889270691086426e-05,
      "loss": 0.06743090152740479,
      "loss_step_1": 0.0735,
      "loss_step_2": 0.0726,
      "loss_step_3": 0.066,
      "step": 24480
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.2016,
      "entropy_step_3": 0.197,
      "epoch": 0.728413789000922,
      "grad_norm": 174.54901123046875,
      "learning_rate": 1.885078445504519e-05,
      "loss": 0.06049860119819641,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0584,
      "step": 24490
    },
    {
      "entropy_step_1": 0.167,
      "entropy_step_2": 0.1701,
      "entropy_step_3": 0.1762,
      "epoch": 0.7287112221528211,
      "grad_norm": 67.3869857788086,
      "learning_rate": 1.881232835629386e-05,
      "loss": 0.05054461359977722,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0475,
      "step": 24500
    },
    {
      "entropy_step_1": 0.1605,
      "entropy_step_2": 0.1667,
      "entropy_step_3": 0.1767,
      "epoch": 0.7290086553047203,
      "grad_norm": 71.41658020019531,
      "learning_rate": 1.877390243203922e-05,
      "loss": 0.046273154020309445,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0476,
      "step": 24510
    },
    {
      "entropy_step_1": 0.2025,
      "entropy_step_2": 0.2086,
      "entropy_step_3": 0.2196,
      "epoch": 0.7293060884566194,
      "grad_norm": 279.87542724609375,
      "learning_rate": 1.8735506719458862e-05,
      "loss": 0.06502461433410645,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0648,
      "loss_step_3": 0.068,
      "step": 24520
    },
    {
      "entropy_step_1": 0.1815,
      "entropy_step_2": 0.1863,
      "entropy_step_3": 0.1849,
      "epoch": 0.7296035216085185,
      "grad_norm": 39.09892272949219,
      "learning_rate": 1.8697141255701174e-05,
      "loss": 0.05046471953392029,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.0544,
      "step": 24530
    },
    {
      "entropy_step_1": 0.1853,
      "entropy_step_2": 0.1899,
      "entropy_step_3": 0.2011,
      "epoch": 0.7299009547604176,
      "grad_norm": 376.38824462890625,
      "learning_rate": 1.8658806077885233e-05,
      "loss": 0.05413447022438049,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0529,
      "step": 24540
    },
    {
      "entropy_step_1": 0.1771,
      "entropy_step_2": 0.1805,
      "entropy_step_3": 0.1617,
      "epoch": 0.7301983879123167,
      "grad_norm": 69.27806854248047,
      "learning_rate": 1.8620501223100894e-05,
      "loss": 0.050594758987426755,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0502,
      "step": 24550
    },
    {
      "entropy_step_1": 0.1557,
      "entropy_step_2": 0.1584,
      "entropy_step_3": 0.1472,
      "epoch": 0.7304958210642158,
      "grad_norm": 27.971057891845703,
      "learning_rate": 1.8582226728408603e-05,
      "loss": 0.05262129306793213,
      "loss_step_1": 0.047,
      "loss_step_2": 0.0464,
      "loss_step_3": 0.0379,
      "step": 24560
    },
    {
      "entropy_step_1": 0.1951,
      "entropy_step_2": 0.1961,
      "entropy_step_3": 0.1943,
      "epoch": 0.730793254216115,
      "grad_norm": 76.12445831298828,
      "learning_rate": 1.8543982630839447e-05,
      "loss": 0.06050131320953369,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0662,
      "step": 24570
    },
    {
      "entropy_step_1": 0.1942,
      "entropy_step_2": 0.192,
      "entropy_step_3": 0.1943,
      "epoch": 0.7310906873680141,
      "grad_norm": 175.85641479492188,
      "learning_rate": 1.8505768967395103e-05,
      "loss": 0.056738466024398804,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0509,
      "step": 24580
    },
    {
      "entropy_step_1": 0.1913,
      "entropy_step_2": 0.1899,
      "entropy_step_3": 0.1954,
      "epoch": 0.7313881205199132,
      "grad_norm": 155.88523864746094,
      "learning_rate": 1.846758577504784e-05,
      "loss": 0.05508959889411926,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0632,
      "step": 24590
    },
    {
      "entropy_step_1": 0.1748,
      "entropy_step_2": 0.1721,
      "entropy_step_3": 0.1685,
      "epoch": 0.7316855536718122,
      "grad_norm": 90.42666625976562,
      "learning_rate": 1.84294330907404e-05,
      "loss": 0.052173280715942384,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0522,
      "loss_step_3": 0.0497,
      "step": 24600
    },
    {
      "entropy_step_1": 0.1874,
      "entropy_step_2": 0.187,
      "entropy_step_3": 0.182,
      "epoch": 0.7319829868237113,
      "grad_norm": 64.50325012207031,
      "learning_rate": 1.8391310951386026e-05,
      "loss": 0.05891335606575012,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.0545,
      "step": 24610
    },
    {
      "entropy_step_1": 0.1733,
      "entropy_step_2": 0.1739,
      "entropy_step_3": 0.1699,
      "epoch": 0.7322804199756104,
      "grad_norm": 133.53048706054688,
      "learning_rate": 1.8353219393868387e-05,
      "loss": 0.054579544067382815,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.062,
      "step": 24620
    },
    {
      "entropy_step_1": 0.1722,
      "entropy_step_2": 0.1703,
      "entropy_step_3": 0.1505,
      "epoch": 0.7325778531275096,
      "grad_norm": 86.90140533447266,
      "learning_rate": 1.831515845504162e-05,
      "loss": 0.054340654611587526,
      "loss_step_1": 0.0469,
      "loss_step_2": 0.0469,
      "loss_step_3": 0.0403,
      "step": 24630
    },
    {
      "entropy_step_1": 0.1895,
      "entropy_step_2": 0.1875,
      "entropy_step_3": 0.1947,
      "epoch": 0.7328752862794087,
      "grad_norm": 208.96644592285156,
      "learning_rate": 1.827712817173017e-05,
      "loss": 0.05735622644424439,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0625,
      "loss_step_3": 0.0645,
      "step": 24640
    },
    {
      "entropy_step_1": 0.1939,
      "entropy_step_2": 0.1922,
      "entropy_step_3": 0.2022,
      "epoch": 0.7331727194313078,
      "grad_norm": 89.31363677978516,
      "learning_rate": 1.823912858072889e-05,
      "loss": 0.06604982614517212,
      "loss_step_1": 0.066,
      "loss_step_2": 0.0659,
      "loss_step_3": 0.0744,
      "step": 24650
    },
    {
      "entropy_step_1": 0.2117,
      "entropy_step_2": 0.2089,
      "entropy_step_3": 0.1972,
      "epoch": 0.7334701525832069,
      "grad_norm": 145.69313049316406,
      "learning_rate": 1.8201159718802848e-05,
      "loss": 0.05913482308387756,
      "loss_step_1": 0.0627,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0633,
      "step": 24660
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1781,
      "entropy_step_3": 0.1656,
      "epoch": 0.733767585735106,
      "grad_norm": 149.7771453857422,
      "learning_rate": 1.8163221622687482e-05,
      "loss": 0.050341558456420896,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0438,
      "step": 24670
    },
    {
      "entropy_step_1": 0.1752,
      "entropy_step_2": 0.1781,
      "entropy_step_3": 0.1714,
      "epoch": 0.7340650188870051,
      "grad_norm": 56.01728439331055,
      "learning_rate": 1.8125314329088393e-05,
      "loss": 0.052354103326797484,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.048,
      "loss_step_3": 0.0508,
      "step": 24680
    },
    {
      "entropy_step_1": 0.1769,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1779,
      "epoch": 0.7343624520389043,
      "grad_norm": 85.85904693603516,
      "learning_rate": 1.808743787468141e-05,
      "loss": 0.05273523330688477,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0632,
      "step": 24690
    },
    {
      "entropy_step_1": 0.1991,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.1934,
      "epoch": 0.7346598851908034,
      "grad_norm": 129.34393310546875,
      "learning_rate": 1.8049592296112505e-05,
      "loss": 0.0522189199924469,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0481,
      "loss_step_3": 0.0462,
      "step": 24700
    },
    {
      "entropy_step_1": 0.1905,
      "entropy_step_2": 0.1907,
      "entropy_step_3": 0.1925,
      "epoch": 0.7349573183427025,
      "grad_norm": 60.07848358154297,
      "learning_rate": 1.8011777629997817e-05,
      "loss": 0.0558661937713623,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0606,
      "step": 24710
    },
    {
      "entropy_step_1": 0.1646,
      "entropy_step_2": 0.1689,
      "entropy_step_3": 0.1808,
      "epoch": 0.7352547514946016,
      "grad_norm": 208.58021545410156,
      "learning_rate": 1.797399391292353e-05,
      "loss": 0.05151734352111816,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0524,
      "step": 24720
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1836,
      "entropy_step_3": 0.178,
      "epoch": 0.7355521846465007,
      "grad_norm": 139.96109008789062,
      "learning_rate": 1.79362411814459e-05,
      "loss": 0.05402460694313049,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0476,
      "step": 24730
    },
    {
      "entropy_step_1": 0.1869,
      "entropy_step_2": 0.19,
      "entropy_step_3": 0.1923,
      "epoch": 0.7358496177983999,
      "grad_norm": 131.4214324951172,
      "learning_rate": 1.7898519472091245e-05,
      "loss": 0.053771352767944335,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0484,
      "step": 24740
    },
    {
      "entropy_step_1": 0.168,
      "entropy_step_2": 0.1722,
      "entropy_step_3": 0.1762,
      "epoch": 0.736147050950299,
      "grad_norm": 101.95176696777344,
      "learning_rate": 1.7860828821355774e-05,
      "loss": 0.04624178111553192,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0493,
      "step": 24750
    },
    {
      "entropy_step_1": 0.1913,
      "entropy_step_2": 0.1916,
      "entropy_step_3": 0.1679,
      "epoch": 0.736444484102198,
      "grad_norm": 73.66506958007812,
      "learning_rate": 1.7823169265705753e-05,
      "loss": 0.06114517450332642,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.044,
      "step": 24760
    },
    {
      "entropy_step_1": 0.1864,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1714,
      "epoch": 0.7367419172540971,
      "grad_norm": 144.23934936523438,
      "learning_rate": 1.7785540841577276e-05,
      "loss": 0.05553857684135437,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.056,
      "step": 24770
    },
    {
      "entropy_step_1": 0.1887,
      "entropy_step_2": 0.1871,
      "entropy_step_3": 0.1854,
      "epoch": 0.7370393504059962,
      "grad_norm": 89.50142669677734,
      "learning_rate": 1.7747943585376393e-05,
      "loss": 0.05700067281723022,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0576,
      "step": 24780
    },
    {
      "entropy_step_1": 0.1745,
      "entropy_step_2": 0.1724,
      "entropy_step_3": 0.1816,
      "epoch": 0.7373367835578953,
      "grad_norm": 104.70270538330078,
      "learning_rate": 1.7710377533478944e-05,
      "loss": 0.05085597038269043,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0427,
      "step": 24790
    },
    {
      "entropy_step_1": 0.1895,
      "entropy_step_2": 0.1878,
      "entropy_step_3": 0.193,
      "epoch": 0.7376342167097945,
      "grad_norm": 149.8698272705078,
      "learning_rate": 1.7672842722230597e-05,
      "loss": 0.05726882815361023,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0666,
      "step": 24800
    },
    {
      "entropy_step_1": 0.1789,
      "entropy_step_2": 0.1784,
      "entropy_step_3": 0.1701,
      "epoch": 0.7379316498616936,
      "grad_norm": 139.1858367919922,
      "learning_rate": 1.7635339187946776e-05,
      "loss": 0.055407369136810304,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0388,
      "step": 24810
    },
    {
      "entropy_step_1": 0.1988,
      "entropy_step_2": 0.1957,
      "entropy_step_3": 0.1973,
      "epoch": 0.7382290830135927,
      "grad_norm": 530.8410034179688,
      "learning_rate": 1.759786696691269e-05,
      "loss": 0.05732191801071167,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0488,
      "step": 24820
    },
    {
      "entropy_step_1": 0.2093,
      "entropy_step_2": 0.2035,
      "entropy_step_3": 0.2002,
      "epoch": 0.7385265161654918,
      "grad_norm": 108.04729461669922,
      "learning_rate": 1.7560426095383208e-05,
      "loss": 0.06427470445632935,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0569,
      "step": 24830
    },
    {
      "entropy_step_1": 0.2268,
      "entropy_step_2": 0.2183,
      "entropy_step_3": 0.2123,
      "epoch": 0.7388239493173909,
      "grad_norm": 81.08525085449219,
      "learning_rate": 1.752301660958288e-05,
      "loss": 0.06844622492790223,
      "loss_step_1": 0.0692,
      "loss_step_2": 0.068,
      "loss_step_3": 0.0569,
      "step": 24840
    },
    {
      "entropy_step_1": 0.2019,
      "entropy_step_2": 0.1964,
      "entropy_step_3": 0.2009,
      "epoch": 0.73912138246929,
      "grad_norm": 90.81783294677734,
      "learning_rate": 1.7485638545705884e-05,
      "loss": 0.047540795803070066,
      "loss_step_1": 0.0469,
      "loss_step_2": 0.0461,
      "loss_step_3": 0.0467,
      "step": 24850
    },
    {
      "entropy_step_1": 0.1689,
      "entropy_step_2": 0.1672,
      "entropy_step_3": 0.1586,
      "epoch": 0.7394188156211892,
      "grad_norm": 166.8651885986328,
      "learning_rate": 1.7448291939916035e-05,
      "loss": 0.055978381633758546,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0552,
      "step": 24860
    },
    {
      "entropy_step_1": 0.182,
      "entropy_step_2": 0.1806,
      "entropy_step_3": 0.1817,
      "epoch": 0.7397162487730883,
      "grad_norm": 164.9336395263672,
      "learning_rate": 1.741097682834666e-05,
      "loss": 0.05108861923217774,
      "loss_step_1": 0.0497,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0509,
      "step": 24870
    },
    {
      "entropy_step_1": 0.2065,
      "entropy_step_2": 0.2034,
      "entropy_step_3": 0.1939,
      "epoch": 0.7400136819249874,
      "grad_norm": 1127.873779296875,
      "learning_rate": 1.7373693247100643e-05,
      "loss": 0.061770886182785034,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0589,
      "step": 24880
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1755,
      "entropy_step_3": 0.1957,
      "epoch": 0.7403111150768865,
      "grad_norm": 90.69304656982422,
      "learning_rate": 1.7336441232250357e-05,
      "loss": 0.05048884749412537,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0528,
      "step": 24890
    },
    {
      "entropy_step_1": 0.1827,
      "entropy_step_2": 0.1776,
      "entropy_step_3": 0.1751,
      "epoch": 0.7406085482287856,
      "grad_norm": 64.91251373291016,
      "learning_rate": 1.729922081983761e-05,
      "loss": 0.057067716121673585,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0636,
      "step": 24900
    },
    {
      "entropy_step_1": 0.1979,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.1931,
      "epoch": 0.7409059813806846,
      "grad_norm": 55.579368591308594,
      "learning_rate": 1.7262032045873694e-05,
      "loss": 0.055897456407546994,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0442,
      "step": 24910
    },
    {
      "entropy_step_1": 0.2068,
      "entropy_step_2": 0.2035,
      "entropy_step_3": 0.1905,
      "epoch": 0.7412034145325838,
      "grad_norm": 69.16661071777344,
      "learning_rate": 1.7224874946339243e-05,
      "loss": 0.05747237801551819,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0498,
      "step": 24920
    },
    {
      "entropy_step_1": 0.1661,
      "entropy_step_2": 0.1643,
      "entropy_step_3": 0.1593,
      "epoch": 0.7415008476844829,
      "grad_norm": 105.56983184814453,
      "learning_rate": 1.7187749557184245e-05,
      "loss": 0.05292752981185913,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.05,
      "step": 24930
    },
    {
      "entropy_step_1": 0.1903,
      "entropy_step_2": 0.1883,
      "entropy_step_3": 0.1706,
      "epoch": 0.741798280836382,
      "grad_norm": 60.89784240722656,
      "learning_rate": 1.7150655914328016e-05,
      "loss": 0.05875328779220581,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0509,
      "step": 24940
    },
    {
      "entropy_step_1": 0.187,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1766,
      "epoch": 0.7420957139882811,
      "grad_norm": 92.1558609008789,
      "learning_rate": 1.7113594053659183e-05,
      "loss": 0.05746340751647949,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0491,
      "step": 24950
    },
    {
      "entropy_step_1": 0.1836,
      "entropy_step_2": 0.184,
      "entropy_step_3": 0.1902,
      "epoch": 0.7423931471401802,
      "grad_norm": 162.9598846435547,
      "learning_rate": 1.7076564011035572e-05,
      "loss": 0.05604220628738403,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0593,
      "step": 24960
    },
    {
      "entropy_step_1": 0.2014,
      "entropy_step_2": 0.1996,
      "entropy_step_3": 0.1919,
      "epoch": 0.7426905802920793,
      "grad_norm": 72.9642333984375,
      "learning_rate": 1.7039565822284304e-05,
      "loss": 0.049677681922912595,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0397,
      "step": 24970
    },
    {
      "entropy_step_1": 0.1688,
      "entropy_step_2": 0.1703,
      "entropy_step_3": 0.1875,
      "epoch": 0.7429880134439785,
      "grad_norm": 84.5833511352539,
      "learning_rate": 1.7002599523201558e-05,
      "loss": 0.05345211029052734,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0704,
      "step": 24980
    },
    {
      "entropy_step_1": 0.1583,
      "entropy_step_2": 0.1601,
      "entropy_step_3": 0.1471,
      "epoch": 0.7432854465958776,
      "grad_norm": 176.9222412109375,
      "learning_rate": 1.6965665149552784e-05,
      "loss": 0.04762327373027801,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0479,
      "loss_step_3": 0.0503,
      "step": 24990
    },
    {
      "entropy_step_1": 0.1668,
      "entropy_step_2": 0.1665,
      "entropy_step_3": 0.1899,
      "epoch": 0.7435828797477767,
      "grad_norm": 166.47010803222656,
      "learning_rate": 1.692876273707246e-05,
      "loss": 0.052280157804489136,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0605,
      "step": 25000
    },
    {
      "epoch": 0.7435828797477767,
      "eval_accuracy": 0.91,
      "eval_f1": 0.9099189270343309,
      "eval_loss": 0.09640539437532425,
      "eval_precision": 0.911481332798073,
      "eval_recall": 0.91,
      "eval_runtime": 6.5108,
      "eval_samples_per_second": 76.796,
      "eval_steps_per_second": 9.676,
      "step": 25000
    },
    {
      "entropy_step_1": 0.1911,
      "entropy_step_2": 0.1901,
      "entropy_step_3": 0.1922,
      "epoch": 0.7438803128996758,
      "grad_norm": 130.90133666992188,
      "learning_rate": 1.689189232146422e-05,
      "loss": 0.05287748575210571,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0518,
      "step": 25010
    },
    {
      "entropy_step_1": 0.1998,
      "entropy_step_2": 0.2004,
      "entropy_step_3": 0.2109,
      "epoch": 0.7441777460515749,
      "grad_norm": 73.1501693725586,
      "learning_rate": 1.6855053938400628e-05,
      "loss": 0.057415890693664554,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0627,
      "loss_step_3": 0.0637,
      "step": 25020
    },
    {
      "entropy_step_1": 0.1751,
      "entropy_step_2": 0.1775,
      "entropy_step_3": 0.1946,
      "epoch": 0.7444751792034741,
      "grad_norm": 70.74671936035156,
      "learning_rate": 1.681824762352337e-05,
      "loss": 0.04672623574733734,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0558,
      "step": 25030
    },
    {
      "entropy_step_1": 0.1801,
      "entropy_step_2": 0.1833,
      "entropy_step_3": 0.1804,
      "epoch": 0.7447726123553732,
      "grad_norm": 178.75286865234375,
      "learning_rate": 1.6781473412443037e-05,
      "loss": 0.05544346570968628,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0519,
      "step": 25040
    },
    {
      "entropy_step_1": 0.182,
      "entropy_step_2": 0.1826,
      "entropy_step_3": 0.1977,
      "epoch": 0.7450700455072723,
      "grad_norm": 101.14956665039062,
      "learning_rate": 1.6744731340739166e-05,
      "loss": 0.05185205340385437,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0559,
      "step": 25050
    },
    {
      "entropy_step_1": 0.2108,
      "entropy_step_2": 0.2096,
      "entropy_step_3": 0.2219,
      "epoch": 0.7453674786591713,
      "grad_norm": 120.61801147460938,
      "learning_rate": 1.670802144396024e-05,
      "loss": 0.051784825325012204,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0586,
      "step": 25060
    },
    {
      "entropy_step_1": 0.1772,
      "entropy_step_2": 0.1781,
      "entropy_step_3": 0.1904,
      "epoch": 0.7456649118110704,
      "grad_norm": 231.65130615234375,
      "learning_rate": 1.6671343757623577e-05,
      "loss": 0.05337649583816528,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0561,
      "step": 25070
    },
    {
      "entropy_step_1": 0.1787,
      "entropy_step_2": 0.1776,
      "entropy_step_3": 0.1455,
      "epoch": 0.7459623449629695,
      "grad_norm": 268.3359069824219,
      "learning_rate": 1.6634698317215326e-05,
      "loss": 0.057470285892486574,
      "loss_step_1": 0.0611,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0493,
      "step": 25080
    },
    {
      "entropy_step_1": 0.1775,
      "entropy_step_2": 0.1772,
      "entropy_step_3": 0.1791,
      "epoch": 0.7462597781148687,
      "grad_norm": 64.0761489868164,
      "learning_rate": 1.6598085158190435e-05,
      "loss": 0.05193066000938416,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0525,
      "step": 25090
    },
    {
      "entropy_step_1": 0.165,
      "entropy_step_2": 0.1647,
      "entropy_step_3": 0.1463,
      "epoch": 0.7465572112667678,
      "grad_norm": 39.28268051147461,
      "learning_rate": 1.656150431597267e-05,
      "loss": 0.04764939546585083,
      "loss_step_1": 0.0458,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.0361,
      "step": 25100
    },
    {
      "entropy_step_1": 0.1591,
      "entropy_step_2": 0.1617,
      "entropy_step_3": 0.1511,
      "epoch": 0.7468546444186669,
      "grad_norm": 94.41314697265625,
      "learning_rate": 1.6524955825954473e-05,
      "loss": 0.06118384003639221,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0621,
      "step": 25110
    },
    {
      "entropy_step_1": 0.203,
      "entropy_step_2": 0.2041,
      "entropy_step_3": 0.194,
      "epoch": 0.747152077570566,
      "grad_norm": 280.5613098144531,
      "learning_rate": 1.6488439723497017e-05,
      "loss": 0.055451196432113645,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0534,
      "step": 25120
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.1925,
      "entropy_step_3": 0.1763,
      "epoch": 0.7474495107224651,
      "grad_norm": 109.90045928955078,
      "learning_rate": 1.6451956043930094e-05,
      "loss": 0.05871983766555786,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0589,
      "step": 25130
    },
    {
      "entropy_step_1": 0.1599,
      "entropy_step_2": 0.1613,
      "entropy_step_3": 0.1571,
      "epoch": 0.7477469438743642,
      "grad_norm": 117.2219009399414,
      "learning_rate": 1.6415504822552208e-05,
      "loss": 0.047693383693695066,
      "loss_step_1": 0.044,
      "loss_step_2": 0.044,
      "loss_step_3": 0.0491,
      "step": 25140
    },
    {
      "entropy_step_1": 0.1586,
      "entropy_step_2": 0.1633,
      "entropy_step_3": 0.1737,
      "epoch": 0.7480443770262634,
      "grad_norm": 77.00016784667969,
      "learning_rate": 1.6379086094630388e-05,
      "loss": 0.048672768473625186,
      "loss_step_1": 0.0456,
      "loss_step_2": 0.045,
      "loss_step_3": 0.0461,
      "step": 25150
    },
    {
      "entropy_step_1": 0.1711,
      "entropy_step_2": 0.1736,
      "entropy_step_3": 0.1789,
      "epoch": 0.7483418101781625,
      "grad_norm": 64.68899536132812,
      "learning_rate": 1.634269989540025e-05,
      "loss": 0.05386747717857361,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0495,
      "loss_step_3": 0.048,
      "step": 25160
    },
    {
      "entropy_step_1": 0.2079,
      "entropy_step_2": 0.2081,
      "entropy_step_3": 0.1981,
      "epoch": 0.7486392433300616,
      "grad_norm": 99.88109588623047,
      "learning_rate": 1.6306346260065918e-05,
      "loss": 0.05322621464729309,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0682,
      "step": 25170
    },
    {
      "entropy_step_1": 0.1726,
      "entropy_step_2": 0.1728,
      "entropy_step_3": 0.1827,
      "epoch": 0.7489366764819607,
      "grad_norm": 88.71745300292969,
      "learning_rate": 1.6270025223800067e-05,
      "loss": 0.055701655149459836,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0584,
      "step": 25180
    },
    {
      "entropy_step_1": 0.1685,
      "entropy_step_2": 0.1711,
      "entropy_step_3": 0.1735,
      "epoch": 0.7492341096338598,
      "grad_norm": 170.2450408935547,
      "learning_rate": 1.623373682174376e-05,
      "loss": 0.05961583256721496,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0595,
      "step": 25190
    },
    {
      "entropy_step_1": 0.1809,
      "entropy_step_2": 0.1825,
      "entropy_step_3": 0.1923,
      "epoch": 0.749531542785759,
      "grad_norm": 40.588844299316406,
      "learning_rate": 1.6197481089006568e-05,
      "loss": 0.04636775553226471,
      "loss_step_1": 0.0464,
      "loss_step_2": 0.0459,
      "loss_step_3": 0.0527,
      "step": 25200
    },
    {
      "entropy_step_1": 0.1728,
      "entropy_step_2": 0.1769,
      "entropy_step_3": 0.1767,
      "epoch": 0.749828975937658,
      "grad_norm": 74.61652374267578,
      "learning_rate": 1.6161258060666335e-05,
      "loss": 0.055732548236846924,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0554,
      "step": 25210
    },
    {
      "entropy_step_1": 0.1901,
      "entropy_step_2": 0.1946,
      "entropy_step_3": 0.2004,
      "epoch": 0.7501264090895571,
      "grad_norm": 85.9818344116211,
      "learning_rate": 1.6125067771769386e-05,
      "loss": 0.06015433073043823,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0642,
      "step": 25220
    },
    {
      "entropy_step_1": 0.1948,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.2133,
      "epoch": 0.7504238422414562,
      "grad_norm": 263.78765869140625,
      "learning_rate": 1.608891025733031e-05,
      "loss": 0.052209514379501346,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0594,
      "step": 25230
    },
    {
      "entropy_step_1": 0.1863,
      "entropy_step_2": 0.1874,
      "entropy_step_3": 0.1793,
      "epoch": 0.7507212753933553,
      "grad_norm": 117.12543487548828,
      "learning_rate": 1.6052785552331983e-05,
      "loss": 0.05207518935203552,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.04,
      "step": 25240
    },
    {
      "entropy_step_1": 0.1925,
      "entropy_step_2": 0.196,
      "entropy_step_3": 0.1928,
      "epoch": 0.7510187085452544,
      "grad_norm": 156.75083923339844,
      "learning_rate": 1.6016693691725536e-05,
      "loss": 0.059199005365371704,
      "loss_step_1": 0.0653,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0576,
      "step": 25250
    },
    {
      "entropy_step_1": 0.1744,
      "entropy_step_2": 0.1786,
      "entropy_step_3": 0.1959,
      "epoch": 0.7513161416971536,
      "grad_norm": 34.46798324584961,
      "learning_rate": 1.598063471043037e-05,
      "loss": 0.04759067595005036,
      "loss_step_1": 0.0506,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.059,
      "step": 25260
    },
    {
      "entropy_step_1": 0.1836,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1931,
      "epoch": 0.7516135748490527,
      "grad_norm": 87.98756408691406,
      "learning_rate": 1.5944608643334024e-05,
      "loss": 0.055248755216598514,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0625,
      "step": 25270
    },
    {
      "entropy_step_1": 0.1585,
      "entropy_step_2": 0.1609,
      "entropy_step_3": 0.1654,
      "epoch": 0.7519110080009518,
      "grad_norm": 66.5262680053711,
      "learning_rate": 1.5908615525292185e-05,
      "loss": 0.05288137793540955,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0595,
      "step": 25280
    },
    {
      "entropy_step_1": 0.1597,
      "entropy_step_2": 0.162,
      "entropy_step_3": 0.1727,
      "epoch": 0.7522084411528509,
      "grad_norm": 59.119388580322266,
      "learning_rate": 1.5872655391128744e-05,
      "loss": 0.06001521944999695,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0567,
      "step": 25290
    },
    {
      "entropy_step_1": 0.1779,
      "entropy_step_2": 0.1776,
      "entropy_step_3": 0.1824,
      "epoch": 0.75250587430475,
      "grad_norm": 118.1175537109375,
      "learning_rate": 1.5836728275635543e-05,
      "loss": 0.04982998967170715,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0498,
      "step": 25300
    },
    {
      "entropy_step_1": 0.2108,
      "entropy_step_2": 0.2128,
      "entropy_step_3": 0.1935,
      "epoch": 0.7528033074566491,
      "grad_norm": 154.07754516601562,
      "learning_rate": 1.5800834213572614e-05,
      "loss": 0.06281092762947083,
      "loss_step_1": 0.0631,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0526,
      "step": 25310
    },
    {
      "entropy_step_1": 0.1739,
      "entropy_step_2": 0.177,
      "entropy_step_3": 0.1917,
      "epoch": 0.7531007406085483,
      "grad_norm": 31.14584732055664,
      "learning_rate": 1.5764973239667913e-05,
      "loss": 0.050091058015823364,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0543,
      "step": 25320
    },
    {
      "entropy_step_1": 0.1829,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1712,
      "epoch": 0.7533981737604474,
      "grad_norm": 230.37591552734375,
      "learning_rate": 1.5729145388617457e-05,
      "loss": 0.06469632983207703,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.065,
      "loss_step_3": 0.0642,
      "step": 25330
    },
    {
      "entropy_step_1": 0.1977,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.2157,
      "epoch": 0.7536956069123465,
      "grad_norm": 68.21881866455078,
      "learning_rate": 1.5693350695085158e-05,
      "loss": 0.057040446996688844,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0669,
      "step": 25340
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1749,
      "entropy_step_3": 0.1789,
      "epoch": 0.7539930400642456,
      "grad_norm": 41.00544357299805,
      "learning_rate": 1.565758919370287e-05,
      "loss": 0.04866970479488373,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0484,
      "step": 25350
    },
    {
      "entropy_step_1": 0.1846,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1947,
      "epoch": 0.7542904732161446,
      "grad_norm": 59.21843719482422,
      "learning_rate": 1.5621860919070312e-05,
      "loss": 0.05331472754478454,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.0543,
      "step": 25360
    },
    {
      "entropy_step_1": 0.1967,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.1901,
      "epoch": 0.7545879063680437,
      "grad_norm": 211.2704620361328,
      "learning_rate": 1.5586165905755123e-05,
      "loss": 0.0771350920200348,
      "loss_step_1": 0.0715,
      "loss_step_2": 0.0703,
      "loss_step_3": 0.061,
      "step": 25370
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1739,
      "entropy_step_3": 0.185,
      "epoch": 0.7548853395199429,
      "grad_norm": 89.18038177490234,
      "learning_rate": 1.5550504188292687e-05,
      "loss": 0.050066685676574706,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0415,
      "step": 25380
    },
    {
      "entropy_step_1": 0.1774,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.2018,
      "epoch": 0.755182772671842,
      "grad_norm": 61.07392883300781,
      "learning_rate": 1.5514875801186196e-05,
      "loss": 0.05667838454246521,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0524,
      "step": 25390
    },
    {
      "entropy_step_1": 0.1916,
      "entropy_step_2": 0.1918,
      "entropy_step_3": 0.1902,
      "epoch": 0.7554802058237411,
      "grad_norm": 125.19800567626953,
      "learning_rate": 1.547928077890659e-05,
      "loss": 0.04871191084384918,
      "loss_step_1": 0.0472,
      "loss_step_2": 0.0469,
      "loss_step_3": 0.0465,
      "step": 25400
    },
    {
      "entropy_step_1": 0.1671,
      "entropy_step_2": 0.1674,
      "entropy_step_3": 0.17,
      "epoch": 0.7557776389756402,
      "grad_norm": 62.40972900390625,
      "learning_rate": 1.5443719155892573e-05,
      "loss": 0.049714303016662596,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.0479,
      "loss_step_3": 0.052,
      "step": 25410
    },
    {
      "entropy_step_1": 0.1807,
      "entropy_step_2": 0.1789,
      "entropy_step_3": 0.1724,
      "epoch": 0.7560750721275393,
      "grad_norm": 120.8071517944336,
      "learning_rate": 1.5408190966550477e-05,
      "loss": 0.053949993848800656,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0517,
      "step": 25420
    },
    {
      "entropy_step_1": 0.2151,
      "entropy_step_2": 0.2129,
      "entropy_step_3": 0.2185,
      "epoch": 0.7563725052794384,
      "grad_norm": 135.96041870117188,
      "learning_rate": 1.537269624525431e-05,
      "loss": 0.061277967691421506,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0672,
      "step": 25430
    },
    {
      "entropy_step_1": 0.1876,
      "entropy_step_2": 0.1863,
      "entropy_step_3": 0.1882,
      "epoch": 0.7566699384313376,
      "grad_norm": 103.6975326538086,
      "learning_rate": 1.5337235026345705e-05,
      "loss": 0.0583420991897583,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0504,
      "step": 25440
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1854,
      "entropy_step_3": 0.1866,
      "epoch": 0.7569673715832367,
      "grad_norm": 112.99028015136719,
      "learning_rate": 1.530180734413384e-05,
      "loss": 0.05712806582450867,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.0636,
      "step": 25450
    },
    {
      "entropy_step_1": 0.1718,
      "entropy_step_2": 0.1709,
      "entropy_step_3": 0.1598,
      "epoch": 0.7572648047351358,
      "grad_norm": 66.77775573730469,
      "learning_rate": 1.5266413232895533e-05,
      "loss": 0.04804437756538391,
      "loss_step_1": 0.0457,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.0394,
      "step": 25460
    },
    {
      "entropy_step_1": 0.1702,
      "entropy_step_2": 0.1707,
      "entropy_step_3": 0.1561,
      "epoch": 0.7575622378870349,
      "grad_norm": 83.65337371826172,
      "learning_rate": 1.5231052726875033e-05,
      "loss": 0.052529752254486084,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0464,
      "step": 25470
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1797,
      "entropy_step_3": 0.1857,
      "epoch": 0.757859671038934,
      "grad_norm": 67.42005157470703,
      "learning_rate": 1.5195725860284126e-05,
      "loss": 0.049905669689178464,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.0594,
      "step": 25480
    },
    {
      "entropy_step_1": 0.1911,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.2028,
      "epoch": 0.7581571041908332,
      "grad_norm": 82.40177154541016,
      "learning_rate": 1.5160432667302011e-05,
      "loss": 0.050308609008789064,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0487,
      "step": 25490
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1711,
      "epoch": 0.7584545373427323,
      "grad_norm": 72.703125,
      "learning_rate": 1.512517318207537e-05,
      "loss": 0.05747021436691284,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0549,
      "step": 25500
    },
    {
      "entropy_step_1": 0.1691,
      "entropy_step_2": 0.1745,
      "entropy_step_3": 0.1671,
      "epoch": 0.7587519704946313,
      "grad_norm": 68.52166748046875,
      "learning_rate": 1.5089947438718199e-05,
      "loss": 0.046883097290992735,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0535,
      "step": 25510
    },
    {
      "entropy_step_1": 0.1772,
      "entropy_step_2": 0.1823,
      "entropy_step_3": 0.1719,
      "epoch": 0.7590494036465304,
      "grad_norm": 63.5421028137207,
      "learning_rate": 1.5054755471311933e-05,
      "loss": 0.05414860248565674,
      "loss_step_1": 0.0582,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0485,
      "step": 25520
    },
    {
      "entropy_step_1": 0.1747,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.1686,
      "epoch": 0.7593468367984295,
      "grad_norm": 235.40797424316406,
      "learning_rate": 1.5019597313905216e-05,
      "loss": 0.052218401432037355,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0425,
      "step": 25530
    },
    {
      "entropy_step_1": 0.1847,
      "entropy_step_2": 0.1874,
      "entropy_step_3": 0.1959,
      "epoch": 0.7596442699503286,
      "grad_norm": 148.48240661621094,
      "learning_rate": 1.4984473000514088e-05,
      "loss": 0.056113576889038085,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0552,
      "step": 25540
    },
    {
      "entropy_step_1": 0.1801,
      "entropy_step_2": 0.1812,
      "entropy_step_3": 0.1639,
      "epoch": 0.7599417031022278,
      "grad_norm": 102.62010192871094,
      "learning_rate": 1.4949382565121772e-05,
      "loss": 0.0530855655670166,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.052,
      "step": 25550
    },
    {
      "entropy_step_1": 0.1656,
      "entropy_step_2": 0.1679,
      "entropy_step_3": 0.158,
      "epoch": 0.7602391362541269,
      "grad_norm": 115.52043151855469,
      "learning_rate": 1.4914326041678788e-05,
      "loss": 0.05446317195892334,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0573,
      "loss_step_3": 0.0471,
      "step": 25560
    },
    {
      "entropy_step_1": 0.162,
      "entropy_step_2": 0.1656,
      "entropy_step_3": 0.1557,
      "epoch": 0.760536569406026,
      "grad_norm": 44.25603103637695,
      "learning_rate": 1.487930346410274e-05,
      "loss": 0.05654385685920715,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0438,
      "step": 25570
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1811,
      "entropy_step_3": 0.183,
      "epoch": 0.7608340025579251,
      "grad_norm": 103.84410095214844,
      "learning_rate": 1.484431486627849e-05,
      "loss": 0.05836474895477295,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0567,
      "step": 25580
    },
    {
      "entropy_step_1": 0.1951,
      "entropy_step_2": 0.1957,
      "entropy_step_3": 0.1854,
      "epoch": 0.7611314357098242,
      "grad_norm": 109.20671081542969,
      "learning_rate": 1.4809360282057972e-05,
      "loss": 0.06007876992225647,
      "loss_step_1": 0.0633,
      "loss_step_2": 0.0626,
      "loss_step_3": 0.055,
      "step": 25590
    },
    {
      "entropy_step_1": 0.1976,
      "entropy_step_2": 0.1979,
      "entropy_step_3": 0.1914,
      "epoch": 0.7614288688617233,
      "grad_norm": 114.22127532958984,
      "learning_rate": 1.4774439745260204e-05,
      "loss": 0.057676959037780764,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0603,
      "step": 25600
    },
    {
      "entropy_step_1": 0.1849,
      "entropy_step_2": 0.1873,
      "entropy_step_3": 0.1799,
      "epoch": 0.7617263020136225,
      "grad_norm": 52.996402740478516,
      "learning_rate": 1.4739553289671309e-05,
      "loss": 0.05096781849861145,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0553,
      "step": 25610
    },
    {
      "entropy_step_1": 0.1658,
      "entropy_step_2": 0.1681,
      "entropy_step_3": 0.159,
      "epoch": 0.7620237351655216,
      "grad_norm": 150.27398681640625,
      "learning_rate": 1.4704700949044393e-05,
      "loss": 0.055685579776763916,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0513,
      "step": 25620
    },
    {
      "entropy_step_1": 0.192,
      "entropy_step_2": 0.1943,
      "entropy_step_3": 0.2114,
      "epoch": 0.7623211683174207,
      "grad_norm": 66.00611114501953,
      "learning_rate": 1.4669882757099574e-05,
      "loss": 0.05290108919143677,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0627,
      "step": 25630
    },
    {
      "entropy_step_1": 0.1479,
      "entropy_step_2": 0.1519,
      "entropy_step_3": 0.143,
      "epoch": 0.7626186014693198,
      "grad_norm": 64.00811767578125,
      "learning_rate": 1.4635098747523907e-05,
      "loss": 0.056760191917419434,
      "loss_step_1": 0.0475,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.039,
      "step": 25640
    },
    {
      "entropy_step_1": 0.2002,
      "entropy_step_2": 0.2033,
      "entropy_step_3": 0.2008,
      "epoch": 0.7629160346212189,
      "grad_norm": 114.6508560180664,
      "learning_rate": 1.460034895397142e-05,
      "loss": 0.055380433797836304,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0563,
      "step": 25650
    },
    {
      "entropy_step_1": 0.1714,
      "entropy_step_2": 0.1729,
      "entropy_step_3": 0.1823,
      "epoch": 0.7632134677731179,
      "grad_norm": 84.80828857421875,
      "learning_rate": 1.4565633410063012e-05,
      "loss": 0.05558071136474609,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.0494,
      "loss_step_3": 0.0512,
      "step": 25660
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.1837,
      "entropy_step_3": 0.204,
      "epoch": 0.763510900925017,
      "grad_norm": 85.955078125,
      "learning_rate": 1.453095214938643e-05,
      "loss": 0.05341622829437256,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0546,
      "step": 25670
    },
    {
      "entropy_step_1": 0.1885,
      "entropy_step_2": 0.1881,
      "entropy_step_3": 0.1868,
      "epoch": 0.7638083340769162,
      "grad_norm": 66.93406677246094,
      "learning_rate": 1.4496305205496252e-05,
      "loss": 0.05519431233406067,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0548,
      "step": 25680
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1729,
      "entropy_step_3": 0.1857,
      "epoch": 0.7641057672288153,
      "grad_norm": 37.242610931396484,
      "learning_rate": 1.4461692611913901e-05,
      "loss": 0.061084359884262085,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.065,
      "step": 25690
    },
    {
      "entropy_step_1": 0.1698,
      "entropy_step_2": 0.1685,
      "entropy_step_3": 0.1664,
      "epoch": 0.7644032003807144,
      "grad_norm": 396.5051574707031,
      "learning_rate": 1.4427114402127522e-05,
      "loss": 0.054235410690307614,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0548,
      "step": 25700
    },
    {
      "entropy_step_1": 0.1932,
      "entropy_step_2": 0.1919,
      "entropy_step_3": 0.1896,
      "epoch": 0.7647006335326135,
      "grad_norm": 161.75723266601562,
      "learning_rate": 1.4392570609591994e-05,
      "loss": 0.055458837747573854,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0533,
      "step": 25710
    },
    {
      "entropy_step_1": 0.1963,
      "entropy_step_2": 0.1968,
      "entropy_step_3": 0.2149,
      "epoch": 0.7649980666845126,
      "grad_norm": 242.17530822753906,
      "learning_rate": 1.4358061267728896e-05,
      "loss": 0.0583021342754364,
      "loss_step_1": 0.0671,
      "loss_step_2": 0.0658,
      "loss_step_3": 0.0695,
      "step": 25720
    },
    {
      "entropy_step_1": 0.1632,
      "entropy_step_2": 0.1649,
      "entropy_step_3": 0.1454,
      "epoch": 0.7652954998364118,
      "grad_norm": 320.912841796875,
      "learning_rate": 1.4323586409926516e-05,
      "loss": 0.04860883057117462,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0461,
      "step": 25730
    },
    {
      "entropy_step_1": 0.1886,
      "entropy_step_2": 0.1898,
      "entropy_step_3": 0.1977,
      "epoch": 0.7655929329883109,
      "grad_norm": 50.32980728149414,
      "learning_rate": 1.4289146069539717e-05,
      "loss": 0.047960165143013,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0585,
      "step": 25740
    },
    {
      "entropy_step_1": 0.1846,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1975,
      "epoch": 0.76589036614021,
      "grad_norm": 120.4278793334961,
      "learning_rate": 1.4254740279890039e-05,
      "loss": 0.05114186406135559,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0531,
      "step": 25750
    },
    {
      "entropy_step_1": 0.1572,
      "entropy_step_2": 0.1606,
      "entropy_step_3": 0.1744,
      "epoch": 0.7661877992921091,
      "grad_norm": 253.22340393066406,
      "learning_rate": 1.42203690742655e-05,
      "loss": 0.0527573823928833,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0574,
      "step": 25760
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1866,
      "entropy_step_3": 0.1824,
      "epoch": 0.7664852324440082,
      "grad_norm": 228.39468383789062,
      "learning_rate": 1.4186032485920747e-05,
      "loss": 0.06032891869544983,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.048,
      "step": 25770
    },
    {
      "entropy_step_1": 0.1832,
      "entropy_step_2": 0.1832,
      "entropy_step_3": 0.1734,
      "epoch": 0.7667826655959074,
      "grad_norm": 83.16265869140625,
      "learning_rate": 1.415173054807688e-05,
      "loss": 0.059750872850418094,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0531,
      "step": 25780
    },
    {
      "entropy_step_1": 0.1923,
      "entropy_step_2": 0.1939,
      "entropy_step_3": 0.1635,
      "epoch": 0.7670800987478065,
      "grad_norm": 128.85821533203125,
      "learning_rate": 1.4117463293921502e-05,
      "loss": 0.055103343725204465,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0471,
      "step": 25790
    },
    {
      "entropy_step_1": 0.1875,
      "entropy_step_2": 0.1889,
      "entropy_step_3": 0.1774,
      "epoch": 0.7673775318997056,
      "grad_norm": 80.15033721923828,
      "learning_rate": 1.4083230756608634e-05,
      "loss": 0.05629996657371521,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.047,
      "step": 25800
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1787,
      "entropy_step_3": 0.1874,
      "epoch": 0.7676749650516046,
      "grad_norm": 251.9918670654297,
      "learning_rate": 1.4049032969258714e-05,
      "loss": 0.05552170872688293,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0602,
      "step": 25810
    },
    {
      "entropy_step_1": 0.1977,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.1881,
      "epoch": 0.7679723982035037,
      "grad_norm": 70.31330108642578,
      "learning_rate": 1.4014869964958594e-05,
      "loss": 0.056977540254592896,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0508,
      "step": 25820
    },
    {
      "entropy_step_1": 0.1785,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.1792,
      "epoch": 0.7682698313554028,
      "grad_norm": 89.01847076416016,
      "learning_rate": 1.3980741776761414e-05,
      "loss": 0.050771260261535646,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0561,
      "step": 25830
    },
    {
      "entropy_step_1": 0.1784,
      "entropy_step_2": 0.1732,
      "entropy_step_3": 0.1696,
      "epoch": 0.768567264507302,
      "grad_norm": 125.99321746826172,
      "learning_rate": 1.394664843768671e-05,
      "loss": 0.05460711121559143,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.0485,
      "step": 25840
    },
    {
      "entropy_step_1": 0.2142,
      "entropy_step_2": 0.2085,
      "entropy_step_3": 0.2231,
      "epoch": 0.7688646976592011,
      "grad_norm": 74.66107940673828,
      "learning_rate": 1.3912589980720192e-05,
      "loss": 0.06959118247032166,
      "loss_step_1": 0.0702,
      "loss_step_2": 0.0712,
      "loss_step_3": 0.072,
      "step": 25850
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1846,
      "entropy_step_3": 0.1727,
      "epoch": 0.7691621308111002,
      "grad_norm": 71.22103118896484,
      "learning_rate": 1.3878566438813922e-05,
      "loss": 0.05087482929229736,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0436,
      "step": 25860
    },
    {
      "entropy_step_1": 0.1841,
      "entropy_step_2": 0.1785,
      "entropy_step_3": 0.1816,
      "epoch": 0.7694595639629993,
      "grad_norm": 49.52678298950195,
      "learning_rate": 1.3844577844886108e-05,
      "loss": 0.0474699467420578,
      "loss_step_1": 0.0472,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0509,
      "step": 25870
    },
    {
      "entropy_step_1": 0.1658,
      "entropy_step_2": 0.16,
      "entropy_step_3": 0.1798,
      "epoch": 0.7697569971148984,
      "grad_norm": 59.93171691894531,
      "learning_rate": 1.3810624231821212e-05,
      "loss": 0.05253161191940307,
      "loss_step_1": 0.0483,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0547,
      "step": 25880
    },
    {
      "entropy_step_1": 0.2027,
      "entropy_step_2": 0.196,
      "entropy_step_3": 0.1946,
      "epoch": 0.7700544302667975,
      "grad_norm": 267.4481201171875,
      "learning_rate": 1.37767056324698e-05,
      "loss": 0.06336039304733276,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0522,
      "step": 25890
    },
    {
      "entropy_step_1": 0.1786,
      "entropy_step_2": 0.1735,
      "entropy_step_3": 0.1425,
      "epoch": 0.7703518634186967,
      "grad_norm": 51.2086296081543,
      "learning_rate": 1.3742822079648577e-05,
      "loss": 0.05151424407958984,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.045,
      "step": 25900
    },
    {
      "entropy_step_1": 0.1837,
      "entropy_step_2": 0.1803,
      "entropy_step_3": 0.1718,
      "epoch": 0.7706492965705958,
      "grad_norm": 67.04029083251953,
      "learning_rate": 1.370897360614033e-05,
      "loss": 0.051437145471572875,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0498,
      "loss_step_3": 0.0469,
      "step": 25910
    },
    {
      "entropy_step_1": 0.1796,
      "entropy_step_2": 0.1764,
      "entropy_step_3": 0.1764,
      "epoch": 0.7709467297224949,
      "grad_norm": 75.02259063720703,
      "learning_rate": 1.367516024469394e-05,
      "loss": 0.04980939030647278,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0546,
      "step": 25920
    },
    {
      "entropy_step_1": 0.1736,
      "entropy_step_2": 0.1724,
      "entropy_step_3": 0.1811,
      "epoch": 0.771244162874394,
      "grad_norm": 80.56145477294922,
      "learning_rate": 1.3641382028024285e-05,
      "loss": 0.05279460549354553,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0553,
      "step": 25930
    },
    {
      "entropy_step_1": 0.1714,
      "entropy_step_2": 0.1702,
      "entropy_step_3": 0.1723,
      "epoch": 0.7715415960262931,
      "grad_norm": 120.51045227050781,
      "learning_rate": 1.3607638988812255e-05,
      "loss": 0.05315805673599243,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0479,
      "step": 25940
    },
    {
      "entropy_step_1": 0.1586,
      "entropy_step_2": 0.1568,
      "entropy_step_3": 0.1458,
      "epoch": 0.7718390291781922,
      "grad_norm": 130.65814208984375,
      "learning_rate": 1.3573931159704673e-05,
      "loss": 0.04890690445899963,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0489,
      "step": 25950
    },
    {
      "entropy_step_1": 0.1588,
      "entropy_step_2": 0.158,
      "entropy_step_3": 0.1612,
      "epoch": 0.7721364623300914,
      "grad_norm": 103.95976257324219,
      "learning_rate": 1.3540258573314357e-05,
      "loss": 0.05478057265281677,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0506,
      "step": 25960
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.1981,
      "entropy_step_3": 0.2026,
      "epoch": 0.7724338954819904,
      "grad_norm": 188.96047973632812,
      "learning_rate": 1.3506621262219977e-05,
      "loss": 0.06020932793617249,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0552,
      "step": 25970
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1833,
      "entropy_step_3": 0.1803,
      "epoch": 0.7727313286338895,
      "grad_norm": 2136.552490234375,
      "learning_rate": 1.3473019258966092e-05,
      "loss": 0.06269159317016601,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0473,
      "loss_step_3": 0.0471,
      "step": 25980
    },
    {
      "entropy_step_1": 0.2038,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.178,
      "epoch": 0.7730287617857886,
      "grad_norm": 172.69117736816406,
      "learning_rate": 1.3439452596063095e-05,
      "loss": 0.05932940244674682,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0578,
      "step": 25990
    },
    {
      "entropy_step_1": 0.1986,
      "entropy_step_2": 0.1957,
      "entropy_step_3": 0.1806,
      "epoch": 0.7733261949376877,
      "grad_norm": 221.3213653564453,
      "learning_rate": 1.3405921305987173e-05,
      "loss": 0.056193023920059204,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0566,
      "step": 26000
    },
    {
      "epoch": 0.7733261949376877,
      "eval_accuracy": 0.915,
      "eval_f1": 0.9149624984618218,
      "eval_loss": 0.09449252486228943,
      "eval_precision": 0.9157333536358135,
      "eval_recall": 0.915,
      "eval_runtime": 6.4844,
      "eval_samples_per_second": 77.108,
      "eval_steps_per_second": 9.716,
      "step": 26000
    },
    {
      "entropy_step_1": 0.1659,
      "entropy_step_2": 0.1662,
      "entropy_step_3": 0.1712,
      "epoch": 0.7736236280895868,
      "grad_norm": 96.95243072509766,
      "learning_rate": 1.337242542118034e-05,
      "loss": 0.05315739512443542,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0493,
      "step": 26010
    },
    {
      "entropy_step_1": 0.1767,
      "entropy_step_2": 0.1757,
      "entropy_step_3": 0.1761,
      "epoch": 0.773921061241486,
      "grad_norm": 40.08684158325195,
      "learning_rate": 1.3338964974050305e-05,
      "loss": 0.05247824788093567,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0512,
      "step": 26020
    },
    {
      "entropy_step_1": 0.1665,
      "entropy_step_2": 0.1677,
      "entropy_step_3": 0.198,
      "epoch": 0.7742184943933851,
      "grad_norm": 133.98779296875,
      "learning_rate": 1.3305539996970495e-05,
      "loss": 0.05883731245994568,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0672,
      "step": 26030
    },
    {
      "entropy_step_1": 0.1935,
      "entropy_step_2": 0.1934,
      "entropy_step_3": 0.1825,
      "epoch": 0.7745159275452842,
      "grad_norm": 191.15139770507812,
      "learning_rate": 1.327215052228003e-05,
      "loss": 0.056642967462539676,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.0527,
      "loss_step_3": 0.0478,
      "step": 26040
    },
    {
      "entropy_step_1": 0.1842,
      "entropy_step_2": 0.1824,
      "entropy_step_3": 0.1919,
      "epoch": 0.7748133606971833,
      "grad_norm": 54.469791412353516,
      "learning_rate": 1.3238796582283703e-05,
      "loss": 0.059070253372192384,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0666,
      "step": 26050
    },
    {
      "entropy_step_1": 0.1747,
      "entropy_step_2": 0.1722,
      "entropy_step_3": 0.1811,
      "epoch": 0.7751107938490824,
      "grad_norm": 113.58892822265625,
      "learning_rate": 1.3205478209251875e-05,
      "loss": 0.05072833895683289,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0609,
      "step": 26060
    },
    {
      "entropy_step_1": 0.1808,
      "entropy_step_2": 0.1792,
      "entropy_step_3": 0.1797,
      "epoch": 0.7754082270009816,
      "grad_norm": 97.1191177368164,
      "learning_rate": 1.317219543542057e-05,
      "loss": 0.060140413045883176,
      "loss_step_1": 0.0626,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0677,
      "step": 26070
    },
    {
      "entropy_step_1": 0.1636,
      "entropy_step_2": 0.1658,
      "entropy_step_3": 0.1621,
      "epoch": 0.7757056601528807,
      "grad_norm": 200.96974182128906,
      "learning_rate": 1.3138948292991272e-05,
      "loss": 0.05267404913902283,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0497,
      "loss_step_3": 0.0509,
      "step": 26080
    },
    {
      "entropy_step_1": 0.1757,
      "entropy_step_2": 0.1744,
      "entropy_step_3": 0.1688,
      "epoch": 0.7760030933047798,
      "grad_norm": 126.23387908935547,
      "learning_rate": 1.3105736814131075e-05,
      "loss": 0.04891044795513153,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.05,
      "step": 26090
    },
    {
      "entropy_step_1": 0.1913,
      "entropy_step_2": 0.1872,
      "entropy_step_3": 0.1904,
      "epoch": 0.7763005264566789,
      "grad_norm": 51.45682907104492,
      "learning_rate": 1.307256103097252e-05,
      "loss": 0.0556588351726532,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0548,
      "step": 26100
    },
    {
      "entropy_step_1": 0.1926,
      "entropy_step_2": 0.1882,
      "entropy_step_3": 0.1852,
      "epoch": 0.776597959608578,
      "grad_norm": 148.6895294189453,
      "learning_rate": 1.3039420975613664e-05,
      "loss": 0.053625750541687014,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0399,
      "step": 26110
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.1976,
      "entropy_step_3": 0.2105,
      "epoch": 0.776895392760477,
      "grad_norm": 131.82447814941406,
      "learning_rate": 1.3006316680117909e-05,
      "loss": 0.05499733090400696,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0513,
      "step": 26120
    },
    {
      "entropy_step_1": 0.2064,
      "entropy_step_2": 0.2041,
      "entropy_step_3": 0.2033,
      "epoch": 0.7771928259123762,
      "grad_norm": 73.72447204589844,
      "learning_rate": 1.2973248176514152e-05,
      "loss": 0.061909687519073484,
      "loss_step_1": 0.0693,
      "loss_step_2": 0.0682,
      "loss_step_3": 0.075,
      "step": 26130
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.182,
      "entropy_step_3": 0.1606,
      "epoch": 0.7774902590642753,
      "grad_norm": 165.0065155029297,
      "learning_rate": 1.2940215496796609e-05,
      "loss": 0.05366050004959107,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0445,
      "step": 26140
    },
    {
      "entropy_step_1": 0.1835,
      "entropy_step_2": 0.1809,
      "entropy_step_3": 0.1825,
      "epoch": 0.7777876922161744,
      "grad_norm": 76.33038330078125,
      "learning_rate": 1.2907218672924825e-05,
      "loss": 0.05057266354560852,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.051,
      "loss_step_3": 0.057,
      "step": 26150
    },
    {
      "entropy_step_1": 0.1901,
      "entropy_step_2": 0.1872,
      "entropy_step_3": 0.1895,
      "epoch": 0.7780851253680735,
      "grad_norm": 126.38690185546875,
      "learning_rate": 1.2874257736823714e-05,
      "loss": 0.053757643699645995,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0547,
      "step": 26160
    },
    {
      "entropy_step_1": 0.1722,
      "entropy_step_2": 0.1721,
      "entropy_step_3": 0.1855,
      "epoch": 0.7783825585199726,
      "grad_norm": 81.77179718017578,
      "learning_rate": 1.2841332720383408e-05,
      "loss": 0.05764763951301575,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0589,
      "step": 26170
    },
    {
      "entropy_step_1": 0.1858,
      "entropy_step_2": 0.1843,
      "entropy_step_3": 0.1911,
      "epoch": 0.7786799916718717,
      "grad_norm": 115.05111694335938,
      "learning_rate": 1.2808443655459313e-05,
      "loss": 0.057241004705429074,
      "loss_step_1": 0.0592,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.069,
      "step": 26180
    },
    {
      "entropy_step_1": 0.2002,
      "entropy_step_2": 0.1967,
      "entropy_step_3": 0.1989,
      "epoch": 0.7789774248237709,
      "grad_norm": 72.68643951416016,
      "learning_rate": 1.2775590573872031e-05,
      "loss": 0.05364304780960083,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0589,
      "step": 26190
    },
    {
      "entropy_step_1": 0.2149,
      "entropy_step_2": 0.2143,
      "entropy_step_3": 0.1926,
      "epoch": 0.77927485797567,
      "grad_norm": 190.98190307617188,
      "learning_rate": 1.2742773507407397e-05,
      "loss": 0.05747292637825012,
      "loss_step_1": 0.0635,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0512,
      "step": 26200
    },
    {
      "entropy_step_1": 0.1795,
      "entropy_step_2": 0.1779,
      "entropy_step_3": 0.1789,
      "epoch": 0.7795722911275691,
      "grad_norm": 91.10305786132812,
      "learning_rate": 1.2709992487816347e-05,
      "loss": 0.05376814603805542,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0539,
      "step": 26210
    },
    {
      "entropy_step_1": 0.1925,
      "entropy_step_2": 0.1926,
      "entropy_step_3": 0.1839,
      "epoch": 0.7798697242794682,
      "grad_norm": 78.90355682373047,
      "learning_rate": 1.2677247546814975e-05,
      "loss": 0.05321159362792969,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0516,
      "step": 26220
    },
    {
      "entropy_step_1": 0.1863,
      "entropy_step_2": 0.1887,
      "entropy_step_3": 0.1983,
      "epoch": 0.7801671574313673,
      "grad_norm": 106.8844223022461,
      "learning_rate": 1.2644538716084431e-05,
      "loss": 0.06108141541481018,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0586,
      "step": 26230
    },
    {
      "entropy_step_1": 0.1827,
      "entropy_step_2": 0.1834,
      "entropy_step_3": 0.1877,
      "epoch": 0.7804645905832664,
      "grad_norm": 81.53189086914062,
      "learning_rate": 1.261186602727099e-05,
      "loss": 0.055882114171981814,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0493,
      "step": 26240
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.19,
      "entropy_step_3": 0.1732,
      "epoch": 0.7807620237351656,
      "grad_norm": 67.39456176757812,
      "learning_rate": 1.2579229511985913e-05,
      "loss": 0.05737369060516358,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.0453,
      "step": 26250
    },
    {
      "entropy_step_1": 0.1903,
      "entropy_step_2": 0.1885,
      "entropy_step_3": 0.1949,
      "epoch": 0.7810594568870647,
      "grad_norm": 194.6658935546875,
      "learning_rate": 1.254662920180546e-05,
      "loss": 0.05318688154220581,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0506,
      "step": 26260
    },
    {
      "entropy_step_1": 0.1846,
      "entropy_step_2": 0.1821,
      "entropy_step_3": 0.1826,
      "epoch": 0.7813568900389637,
      "grad_norm": 44.714595794677734,
      "learning_rate": 1.2514065128270868e-05,
      "loss": 0.05647706985473633,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0634,
      "step": 26270
    },
    {
      "entropy_step_1": 0.1796,
      "entropy_step_2": 0.1766,
      "entropy_step_3": 0.1661,
      "epoch": 0.7816543231908628,
      "grad_norm": 51.677467346191406,
      "learning_rate": 1.2481537322888348e-05,
      "loss": 0.05159374475479126,
      "loss_step_1": 0.0461,
      "loss_step_2": 0.0457,
      "loss_step_3": 0.0435,
      "step": 26280
    },
    {
      "entropy_step_1": 0.1796,
      "entropy_step_2": 0.1752,
      "entropy_step_3": 0.1727,
      "epoch": 0.7819517563427619,
      "grad_norm": 55.45363235473633,
      "learning_rate": 1.2449045817128985e-05,
      "loss": 0.05497862696647644,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0449,
      "step": 26290
    },
    {
      "entropy_step_1": 0.1962,
      "entropy_step_2": 0.1911,
      "entropy_step_3": 0.1872,
      "epoch": 0.782249189494661,
      "grad_norm": 48.97903823852539,
      "learning_rate": 1.2416590642428744e-05,
      "loss": 0.05891569256782532,
      "loss_step_1": 0.0686,
      "loss_step_2": 0.0677,
      "loss_step_3": 0.067,
      "step": 26300
    },
    {
      "entropy_step_1": 0.1959,
      "entropy_step_2": 0.1904,
      "entropy_step_3": 0.1837,
      "epoch": 0.7825466226465602,
      "grad_norm": 57.201358795166016,
      "learning_rate": 1.238417183018844e-05,
      "loss": 0.05769276022911072,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0568,
      "step": 26310
    },
    {
      "entropy_step_1": 0.2074,
      "entropy_step_2": 0.2012,
      "entropy_step_3": 0.2104,
      "epoch": 0.7828440557984593,
      "grad_norm": 116.12928009033203,
      "learning_rate": 1.235178941177374e-05,
      "loss": 0.057035863399505615,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0601,
      "step": 26320
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.1794,
      "entropy_step_3": 0.1837,
      "epoch": 0.7831414889503584,
      "grad_norm": 226.7863006591797,
      "learning_rate": 1.2319443418515065e-05,
      "loss": 0.05371706485748291,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0464,
      "step": 26330
    },
    {
      "entropy_step_1": 0.1689,
      "entropy_step_2": 0.1658,
      "entropy_step_3": 0.1513,
      "epoch": 0.7834389221022575,
      "grad_norm": 95.53885650634766,
      "learning_rate": 1.2287133881707601e-05,
      "loss": 0.05590540766716003,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0508,
      "step": 26340
    },
    {
      "entropy_step_1": 0.1731,
      "entropy_step_2": 0.1709,
      "entropy_step_3": 0.1747,
      "epoch": 0.7837363552541566,
      "grad_norm": 90.3128433227539,
      "learning_rate": 1.2254860832611265e-05,
      "loss": 0.05452269911766052,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0549,
      "step": 26350
    },
    {
      "entropy_step_1": 0.175,
      "entropy_step_2": 0.1721,
      "entropy_step_3": 0.169,
      "epoch": 0.7840337884060558,
      "grad_norm": 49.407814025878906,
      "learning_rate": 1.2222624302450652e-05,
      "loss": 0.05745288133621216,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0578,
      "step": 26360
    },
    {
      "entropy_step_1": 0.1639,
      "entropy_step_2": 0.1595,
      "entropy_step_3": 0.1541,
      "epoch": 0.7843312215579549,
      "grad_norm": 66.21541595458984,
      "learning_rate": 1.2190424322415078e-05,
      "loss": 0.049406629800796506,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0442,
      "step": 26370
    },
    {
      "entropy_step_1": 0.179,
      "entropy_step_2": 0.1737,
      "entropy_step_3": 0.1692,
      "epoch": 0.784628654709854,
      "grad_norm": 80.61353302001953,
      "learning_rate": 1.2158260923658432e-05,
      "loss": 0.05544132590293884,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.052,
      "loss_step_3": 0.0541,
      "step": 26380
    },
    {
      "entropy_step_1": 0.184,
      "entropy_step_2": 0.1797,
      "entropy_step_3": 0.1739,
      "epoch": 0.7849260878617531,
      "grad_norm": 223.25450134277344,
      "learning_rate": 1.2126134137299282e-05,
      "loss": 0.05569378137588501,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0508,
      "step": 26390
    },
    {
      "entropy_step_1": 0.1825,
      "entropy_step_2": 0.1772,
      "entropy_step_3": 0.1748,
      "epoch": 0.7852235210136522,
      "grad_norm": 127.22700500488281,
      "learning_rate": 1.2094043994420679e-05,
      "loss": 0.05367483496665955,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0495,
      "step": 26400
    },
    {
      "entropy_step_1": 0.1948,
      "entropy_step_2": 0.1875,
      "entropy_step_3": 0.1987,
      "epoch": 0.7855209541655513,
      "grad_norm": 42.15559768676758,
      "learning_rate": 1.2061990526070305e-05,
      "loss": 0.05852107405662536,
      "loss_step_1": 0.0677,
      "loss_step_2": 0.0668,
      "loss_step_3": 0.0851,
      "step": 26410
    },
    {
      "entropy_step_1": 0.1828,
      "entropy_step_2": 0.1782,
      "entropy_step_3": 0.1786,
      "epoch": 0.7858183873174504,
      "grad_norm": 237.2361602783203,
      "learning_rate": 1.2029973763260304e-05,
      "loss": 0.05757169127464294,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0528,
      "step": 26420
    },
    {
      "entropy_step_1": 0.191,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1865,
      "epoch": 0.7861158204693495,
      "grad_norm": 94.63184356689453,
      "learning_rate": 1.1997993736967362e-05,
      "loss": 0.05533782839775085,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0591,
      "step": 26430
    },
    {
      "entropy_step_1": 0.1596,
      "entropy_step_2": 0.1561,
      "entropy_step_3": 0.1493,
      "epoch": 0.7864132536212486,
      "grad_norm": 69.08625030517578,
      "learning_rate": 1.1966050478132574e-05,
      "loss": 0.052882665395736696,
      "loss_step_1": 0.0442,
      "loss_step_2": 0.0442,
      "loss_step_3": 0.0373,
      "step": 26440
    },
    {
      "entropy_step_1": 0.1779,
      "entropy_step_2": 0.1767,
      "entropy_step_3": 0.1756,
      "epoch": 0.7867106867731477,
      "grad_norm": 88.79198455810547,
      "learning_rate": 1.1934144017661469e-05,
      "loss": 0.05124973058700562,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0514,
      "step": 26450
    },
    {
      "entropy_step_1": 0.1867,
      "entropy_step_2": 0.1847,
      "entropy_step_3": 0.2048,
      "epoch": 0.7870081199250468,
      "grad_norm": 257.6824035644531,
      "learning_rate": 1.1902274386423973e-05,
      "loss": 0.05491839647293091,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0633,
      "step": 26460
    },
    {
      "entropy_step_1": 0.1781,
      "entropy_step_2": 0.1752,
      "entropy_step_3": 0.21,
      "epoch": 0.7873055530769459,
      "grad_norm": 150.7900390625,
      "learning_rate": 1.1870441615254424e-05,
      "loss": 0.054723703861236574,
      "loss_step_1": 0.0483,
      "loss_step_2": 0.0478,
      "loss_step_3": 0.0596,
      "step": 26470
    },
    {
      "entropy_step_1": 0.1966,
      "entropy_step_2": 0.1944,
      "entropy_step_3": 0.2029,
      "epoch": 0.7876029862288451,
      "grad_norm": 188.9451904296875,
      "learning_rate": 1.1838645734951432e-05,
      "loss": 0.05419694781303406,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0619,
      "step": 26480
    },
    {
      "entropy_step_1": 0.1762,
      "entropy_step_2": 0.1753,
      "entropy_step_3": 0.1644,
      "epoch": 0.7879004193807442,
      "grad_norm": 64.26065826416016,
      "learning_rate": 1.180688677627796e-05,
      "loss": 0.056912779808044434,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0588,
      "step": 26490
    },
    {
      "entropy_step_1": 0.1995,
      "entropy_step_2": 0.1987,
      "entropy_step_3": 0.2099,
      "epoch": 0.7881978525326433,
      "grad_norm": 418.37542724609375,
      "learning_rate": 1.17751647699612e-05,
      "loss": 0.05405132174491882,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0697,
      "step": 26500
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.1768,
      "entropy_step_3": 0.1789,
      "epoch": 0.7884952856845424,
      "grad_norm": 47.56760025024414,
      "learning_rate": 1.174347974669267e-05,
      "loss": 0.0493237167596817,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0473,
      "step": 26510
    },
    {
      "entropy_step_1": 0.1638,
      "entropy_step_2": 0.1626,
      "entropy_step_3": 0.1636,
      "epoch": 0.7887927188364415,
      "grad_norm": 83.68789672851562,
      "learning_rate": 1.1711831737128042e-05,
      "loss": 0.05316762924194336,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0536,
      "step": 26520
    },
    {
      "entropy_step_1": 0.1712,
      "entropy_step_2": 0.1684,
      "entropy_step_3": 0.1538,
      "epoch": 0.7890901519883406,
      "grad_norm": 52.74006271362305,
      "learning_rate": 1.168022077188719e-05,
      "loss": 0.05802299976348877,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0402,
      "step": 26530
    },
    {
      "entropy_step_1": 0.1798,
      "entropy_step_2": 0.1757,
      "entropy_step_3": 0.167,
      "epoch": 0.7893875851402398,
      "grad_norm": 70.50919342041016,
      "learning_rate": 1.164864688155416e-05,
      "loss": 0.058165884017944335,
      "loss_step_1": 0.0622,
      "loss_step_2": 0.0612,
      "loss_step_3": 0.0592,
      "step": 26540
    },
    {
      "entropy_step_1": 0.1954,
      "entropy_step_2": 0.1918,
      "entropy_step_3": 0.1976,
      "epoch": 0.7896850182921389,
      "grad_norm": 39.860660552978516,
      "learning_rate": 1.1617110096677103e-05,
      "loss": 0.05697503685951233,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.0631,
      "step": 26550
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1849,
      "entropy_step_3": 0.1722,
      "epoch": 0.789982451444038,
      "grad_norm": 76.54179382324219,
      "learning_rate": 1.158561044776832e-05,
      "loss": 0.05569036602973938,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0491,
      "step": 26560
    },
    {
      "entropy_step_1": 0.1621,
      "entropy_step_2": 0.1604,
      "entropy_step_3": 0.1796,
      "epoch": 0.790279884595937,
      "grad_norm": 35.75904846191406,
      "learning_rate": 1.155414796530413e-05,
      "loss": 0.04476657509803772,
      "loss_step_1": 0.0436,
      "loss_step_2": 0.0433,
      "loss_step_3": 0.0548,
      "step": 26570
    },
    {
      "entropy_step_1": 0.1664,
      "entropy_step_2": 0.1654,
      "entropy_step_3": 0.1604,
      "epoch": 0.7905773177478361,
      "grad_norm": 52.88188934326172,
      "learning_rate": 1.1522722679724912e-05,
      "loss": 0.05461059212684631,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0475,
      "step": 26580
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.1775,
      "entropy_step_3": 0.1731,
      "epoch": 0.7908747508997352,
      "grad_norm": 78.0169448852539,
      "learning_rate": 1.1491334621435045e-05,
      "loss": 0.059604990482330325,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0457,
      "step": 26590
    },
    {
      "entropy_step_1": 0.1841,
      "entropy_step_2": 0.1827,
      "entropy_step_3": 0.1732,
      "epoch": 0.7911721840516344,
      "grad_norm": 100.04763793945312,
      "learning_rate": 1.1459983820802938e-05,
      "loss": 0.055077898502349856,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0426,
      "step": 26600
    },
    {
      "entropy_step_1": 0.1817,
      "entropy_step_2": 0.1793,
      "entropy_step_3": 0.1754,
      "epoch": 0.7914696172035335,
      "grad_norm": 104.91999053955078,
      "learning_rate": 1.1428670308160882e-05,
      "loss": 0.05926501750946045,
      "loss_step_1": 0.0575,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0538,
      "step": 26610
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.1741,
      "epoch": 0.7917670503554326,
      "grad_norm": 70.43406677246094,
      "learning_rate": 1.1397394113805165e-05,
      "loss": 0.053535908460617065,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0571,
      "step": 26620
    },
    {
      "entropy_step_1": 0.2015,
      "entropy_step_2": 0.1971,
      "entropy_step_3": 0.2132,
      "epoch": 0.7920644835073317,
      "grad_norm": 73.22088623046875,
      "learning_rate": 1.1366155267995886e-05,
      "loss": 0.06023057103157044,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0577,
      "step": 26630
    },
    {
      "entropy_step_1": 0.1881,
      "entropy_step_2": 0.1833,
      "entropy_step_3": 0.1835,
      "epoch": 0.7923619166592308,
      "grad_norm": 214.16969299316406,
      "learning_rate": 1.1334953800957083e-05,
      "loss": 0.05552459359169006,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0576,
      "step": 26640
    },
    {
      "entropy_step_1": 0.1718,
      "entropy_step_2": 0.1673,
      "entropy_step_3": 0.1914,
      "epoch": 0.79265934981113,
      "grad_norm": 181.71435546875,
      "learning_rate": 1.130378974287658e-05,
      "loss": 0.0513361930847168,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0579,
      "step": 26650
    },
    {
      "entropy_step_1": 0.1738,
      "entropy_step_2": 0.1699,
      "entropy_step_3": 0.1514,
      "epoch": 0.7929567829630291,
      "grad_norm": 90.35481262207031,
      "learning_rate": 1.1272663123906068e-05,
      "loss": 0.05236164331436157,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0511,
      "step": 26660
    },
    {
      "entropy_step_1": 0.198,
      "entropy_step_2": 0.1955,
      "entropy_step_3": 0.1976,
      "epoch": 0.7932542161149282,
      "grad_norm": 100.69215393066406,
      "learning_rate": 1.1241573974160918e-05,
      "loss": 0.05885744094848633,
      "loss_step_1": 0.063,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0592,
      "step": 26670
    },
    {
      "entropy_step_1": 0.1951,
      "entropy_step_2": 0.1917,
      "entropy_step_3": 0.1899,
      "epoch": 0.7935516492668273,
      "grad_norm": 62.661773681640625,
      "learning_rate": 1.1210522323720347e-05,
      "loss": 0.050590747594833375,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0529,
      "step": 26680
    },
    {
      "entropy_step_1": 0.1912,
      "entropy_step_2": 0.1876,
      "entropy_step_3": 0.1937,
      "epoch": 0.7938490824187264,
      "grad_norm": 69.6530990600586,
      "learning_rate": 1.117950820262723e-05,
      "loss": 0.053054678440094,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0533,
      "step": 26690
    },
    {
      "entropy_step_1": 0.1682,
      "entropy_step_2": 0.1655,
      "entropy_step_3": 0.1444,
      "epoch": 0.7941465155706255,
      "grad_norm": 196.85411071777344,
      "learning_rate": 1.1148531640888148e-05,
      "loss": 0.050985592603683474,
      "loss_step_1": 0.0467,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0364,
      "step": 26700
    },
    {
      "entropy_step_1": 0.1709,
      "entropy_step_2": 0.1685,
      "entropy_step_3": 0.1828,
      "epoch": 0.7944439487225247,
      "grad_norm": 134.5421142578125,
      "learning_rate": 1.1117592668473364e-05,
      "loss": 0.06094163656234741,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.063,
      "step": 26710
    },
    {
      "entropy_step_1": 0.1954,
      "entropy_step_2": 0.1916,
      "entropy_step_3": 0.1921,
      "epoch": 0.7947413818744237,
      "grad_norm": 124.11528778076172,
      "learning_rate": 1.108669131531675e-05,
      "loss": 0.06736127138137818,
      "loss_step_1": 0.0743,
      "loss_step_2": 0.0745,
      "loss_step_3": 0.0606,
      "step": 26720
    },
    {
      "entropy_step_1": 0.1991,
      "entropy_step_2": 0.194,
      "entropy_step_3": 0.1856,
      "epoch": 0.7950388150263228,
      "grad_norm": 57.26310729980469,
      "learning_rate": 1.1055827611315778e-05,
      "loss": 0.05457510352134705,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0525,
      "step": 26730
    },
    {
      "entropy_step_1": 0.1847,
      "entropy_step_2": 0.1805,
      "entropy_step_3": 0.2032,
      "epoch": 0.7953362481782219,
      "grad_norm": 278.6754150390625,
      "learning_rate": 1.1025001586331485e-05,
      "loss": 0.05381411910057068,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0622,
      "step": 26740
    },
    {
      "entropy_step_1": 0.1759,
      "entropy_step_2": 0.1777,
      "entropy_step_3": 0.1607,
      "epoch": 0.795633681330121,
      "grad_norm": 63.6336669921875,
      "learning_rate": 1.0994213270188509e-05,
      "loss": 0.05396866798400879,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0495,
      "step": 26750
    },
    {
      "entropy_step_1": 0.1781,
      "entropy_step_2": 0.1775,
      "entropy_step_3": 0.1727,
      "epoch": 0.7959311144820201,
      "grad_norm": 50.446617126464844,
      "learning_rate": 1.0963462692674947e-05,
      "loss": 0.05251196026802063,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0524,
      "step": 26760
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1885,
      "entropy_step_3": 0.1784,
      "epoch": 0.7962285476339193,
      "grad_norm": 894.7803344726562,
      "learning_rate": 1.0932749883542392e-05,
      "loss": 0.05597807168960571,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0596,
      "loss_step_3": 0.062,
      "step": 26770
    },
    {
      "entropy_step_1": 0.1821,
      "entropy_step_2": 0.1809,
      "entropy_step_3": 0.1758,
      "epoch": 0.7965259807858184,
      "grad_norm": 74.04255676269531,
      "learning_rate": 1.0902074872505902e-05,
      "loss": 0.054308468103408815,
      "loss_step_1": 0.0549,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0574,
      "step": 26780
    },
    {
      "entropy_step_1": 0.1803,
      "entropy_step_2": 0.1803,
      "entropy_step_3": 0.1581,
      "epoch": 0.7968234139377175,
      "grad_norm": 173.05035400390625,
      "learning_rate": 1.087143768924399e-05,
      "loss": 0.05063085556030274,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0534,
      "step": 26790
    },
    {
      "entropy_step_1": 0.1722,
      "entropy_step_2": 0.1724,
      "entropy_step_3": 0.1962,
      "epoch": 0.7971208470896166,
      "grad_norm": 55.09196090698242,
      "learning_rate": 1.0840838363398537e-05,
      "loss": 0.062434858083724974,
      "loss_step_1": 0.0704,
      "loss_step_2": 0.0691,
      "loss_step_3": 0.0804,
      "step": 26800
    },
    {
      "entropy_step_1": 0.1831,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.1694,
      "epoch": 0.7974182802415157,
      "grad_norm": 66.28376007080078,
      "learning_rate": 1.0810276924574802e-05,
      "loss": 0.04791578650474548,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0489,
      "step": 26810
    },
    {
      "entropy_step_1": 0.1863,
      "entropy_step_2": 0.1856,
      "entropy_step_3": 0.1758,
      "epoch": 0.7977157133934148,
      "grad_norm": 310.7424011230469,
      "learning_rate": 1.0779753402341381e-05,
      "loss": 0.05864846706390381,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0469,
      "step": 26820
    },
    {
      "entropy_step_1": 0.1957,
      "entropy_step_2": 0.1954,
      "entropy_step_3": 0.198,
      "epoch": 0.798013146545314,
      "grad_norm": 59.35341262817383,
      "learning_rate": 1.074926782623023e-05,
      "loss": 0.05615267753601074,
      "loss_step_1": 0.06,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0699,
      "step": 26830
    },
    {
      "entropy_step_1": 0.1869,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.1721,
      "epoch": 0.7983105796972131,
      "grad_norm": 67.27317810058594,
      "learning_rate": 1.0718820225736536e-05,
      "loss": 0.055638629198074344,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.044,
      "step": 26840
    },
    {
      "entropy_step_1": 0.1932,
      "entropy_step_2": 0.1941,
      "entropy_step_3": 0.1734,
      "epoch": 0.7986080128491122,
      "grad_norm": 132.8995819091797,
      "learning_rate": 1.0688410630318773e-05,
      "loss": 0.06475812792778016,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0652,
      "step": 26850
    },
    {
      "entropy_step_1": 0.1849,
      "entropy_step_2": 0.1848,
      "entropy_step_3": 0.205,
      "epoch": 0.7989054460010113,
      "grad_norm": 207.82211303710938,
      "learning_rate": 1.0658039069398623e-05,
      "loss": 0.062323647737503055,
      "loss_step_1": 0.0615,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0703,
      "step": 26860
    },
    {
      "entropy_step_1": 0.182,
      "entropy_step_2": 0.1813,
      "entropy_step_3": 0.1764,
      "epoch": 0.7992028791529103,
      "grad_norm": 123.16419219970703,
      "learning_rate": 1.0627705572361013e-05,
      "loss": 0.04787949621677399,
      "loss_step_1": 0.0465,
      "loss_step_2": 0.0455,
      "loss_step_3": 0.038,
      "step": 26870
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1785,
      "entropy_step_3": 0.1762,
      "epoch": 0.7995003123048094,
      "grad_norm": 61.95934295654297,
      "learning_rate": 1.0597410168554e-05,
      "loss": 0.054610294103622434,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0559,
      "step": 26880
    },
    {
      "entropy_step_1": 0.1754,
      "entropy_step_2": 0.1751,
      "entropy_step_3": 0.1843,
      "epoch": 0.7997977454567086,
      "grad_norm": 56.32499313354492,
      "learning_rate": 1.0567152887288801e-05,
      "loss": 0.056771624088287356,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0531,
      "step": 26890
    },
    {
      "entropy_step_1": 0.1708,
      "entropy_step_2": 0.1708,
      "entropy_step_3": 0.1546,
      "epoch": 0.8000951786086077,
      "grad_norm": 220.79403686523438,
      "learning_rate": 1.053693375783974e-05,
      "loss": 0.052260851860046385,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0425,
      "step": 26900
    },
    {
      "entropy_step_1": 0.1816,
      "entropy_step_2": 0.1819,
      "entropy_step_3": 0.1704,
      "epoch": 0.8003926117605068,
      "grad_norm": 89.9606704711914,
      "learning_rate": 1.0506752809444232e-05,
      "loss": 0.05848245620727539,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.063,
      "loss_step_3": 0.0562,
      "step": 26910
    },
    {
      "entropy_step_1": 0.1756,
      "entropy_step_2": 0.177,
      "entropy_step_3": 0.1663,
      "epoch": 0.8006900449124059,
      "grad_norm": 82.49109649658203,
      "learning_rate": 1.0476610071302772e-05,
      "loss": 0.052726221084594724,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0446,
      "step": 26920
    },
    {
      "entropy_step_1": 0.1759,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1713,
      "epoch": 0.800987478064305,
      "grad_norm": 145.38658142089844,
      "learning_rate": 1.0446505572578852e-05,
      "loss": 0.04911128580570221,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0528,
      "step": 26930
    },
    {
      "entropy_step_1": 0.1564,
      "entropy_step_2": 0.1561,
      "entropy_step_3": 0.1631,
      "epoch": 0.8012849112162042,
      "grad_norm": 108.37654876708984,
      "learning_rate": 1.041643934239902e-05,
      "loss": 0.05474559664726257,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.052,
      "loss_step_3": 0.051,
      "step": 26940
    },
    {
      "entropy_step_1": 0.1777,
      "entropy_step_2": 0.1783,
      "entropy_step_3": 0.1872,
      "epoch": 0.8015823443681033,
      "grad_norm": 122.16942596435547,
      "learning_rate": 1.0386411409852708e-05,
      "loss": 0.05806172490119934,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0661,
      "step": 26950
    },
    {
      "entropy_step_1": 0.1745,
      "entropy_step_2": 0.1739,
      "entropy_step_3": 0.1706,
      "epoch": 0.8018797775200024,
      "grad_norm": 65.57919311523438,
      "learning_rate": 1.0356421803992395e-05,
      "loss": 0.05420033931732178,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0534,
      "step": 26960
    },
    {
      "entropy_step_1": 0.1774,
      "entropy_step_2": 0.1776,
      "entropy_step_3": 0.1456,
      "epoch": 0.8021772106719015,
      "grad_norm": 67.17960357666016,
      "learning_rate": 1.03264705538334e-05,
      "loss": 0.050645488500595096,
      "loss_step_1": 0.0514,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0352,
      "step": 26970
    },
    {
      "entropy_step_1": 0.1783,
      "entropy_step_2": 0.1786,
      "entropy_step_3": 0.1733,
      "epoch": 0.8024746438238006,
      "grad_norm": 109.60295867919922,
      "learning_rate": 1.0296557688353998e-05,
      "loss": 0.05037055611610412,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0424,
      "step": 26980
    },
    {
      "entropy_step_1": 0.1774,
      "entropy_step_2": 0.1768,
      "entropy_step_3": 0.1775,
      "epoch": 0.8027720769756997,
      "grad_norm": 86.39027404785156,
      "learning_rate": 1.0266683236495266e-05,
      "loss": 0.05181623101234436,
      "loss_step_1": 0.048,
      "loss_step_2": 0.048,
      "loss_step_3": 0.0417,
      "step": 26990
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1761,
      "entropy_step_3": 0.16,
      "epoch": 0.8030695101275989,
      "grad_norm": 75.12550354003906,
      "learning_rate": 1.0236847227161162e-05,
      "loss": 0.05632302165031433,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0588,
      "step": 27000
    },
    {
      "epoch": 0.8030695101275989,
      "eval_accuracy": 0.92,
      "eval_f1": 0.9199372307889385,
      "eval_loss": 0.08683381974697113,
      "eval_precision": 0.9213212634822805,
      "eval_recall": 0.92,
      "eval_runtime": 6.5711,
      "eval_samples_per_second": 76.091,
      "eval_steps_per_second": 9.587,
      "step": 27000
    },
    {
      "entropy_step_1": 0.1685,
      "entropy_step_2": 0.1697,
      "entropy_step_3": 0.1619,
      "epoch": 0.803366943279498,
      "grad_norm": 32.99424743652344,
      "learning_rate": 1.0207049689218407e-05,
      "loss": 0.054788851737976076,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0477,
      "step": 27010
    },
    {
      "entropy_step_1": 0.1609,
      "entropy_step_2": 0.1616,
      "entropy_step_3": 0.1676,
      "epoch": 0.8036643764313971,
      "grad_norm": 85.08507537841797,
      "learning_rate": 1.0177290651496551e-05,
      "loss": 0.054577356576919554,
      "loss_step_1": 0.0541,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0498,
      "step": 27020
    },
    {
      "entropy_step_1": 0.1855,
      "entropy_step_2": 0.1851,
      "entropy_step_3": 0.1781,
      "epoch": 0.8039618095832961,
      "grad_norm": 69.87444305419922,
      "learning_rate": 1.0147570142787855e-05,
      "loss": 0.0558816134929657,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.0553,
      "step": 27030
    },
    {
      "entropy_step_1": 0.1732,
      "entropy_step_2": 0.1712,
      "entropy_step_3": 0.1702,
      "epoch": 0.8042592427351952,
      "grad_norm": 179.3473663330078,
      "learning_rate": 1.0117888191847319e-05,
      "loss": 0.04713957309722901,
      "loss_step_1": 0.046,
      "loss_step_2": 0.0456,
      "loss_step_3": 0.0482,
      "step": 27040
    },
    {
      "entropy_step_1": 0.1864,
      "entropy_step_2": 0.187,
      "entropy_step_3": 0.1762,
      "epoch": 0.8045566758870943,
      "grad_norm": 136.78208923339844,
      "learning_rate": 1.0088244827392628e-05,
      "loss": 0.058124327659606935,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0581,
      "step": 27050
    },
    {
      "entropy_step_1": 0.1894,
      "entropy_step_2": 0.19,
      "entropy_step_3": 0.1884,
      "epoch": 0.8048541090389935,
      "grad_norm": 97.91500091552734,
      "learning_rate": 1.0058640078104137e-05,
      "loss": 0.05365650057792663,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.061,
      "step": 27060
    },
    {
      "entropy_step_1": 0.1916,
      "entropy_step_2": 0.1928,
      "entropy_step_3": 0.1949,
      "epoch": 0.8051515421908926,
      "grad_norm": 99.73834991455078,
      "learning_rate": 1.0029073972624863e-05,
      "loss": 0.054393672943115236,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0656,
      "step": 27070
    },
    {
      "entropy_step_1": 0.1757,
      "entropy_step_2": 0.1762,
      "entropy_step_3": 0.1744,
      "epoch": 0.8054489753427917,
      "grad_norm": 82.2139663696289,
      "learning_rate": 9.9995465395604e-06,
      "loss": 0.05685808062553406,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.055,
      "loss_step_3": 0.054,
      "step": 27080
    },
    {
      "entropy_step_1": 0.1629,
      "entropy_step_2": 0.1632,
      "entropy_step_3": 0.153,
      "epoch": 0.8057464084946908,
      "grad_norm": 89.71196746826172,
      "learning_rate": 9.97005780747895e-06,
      "loss": 0.047339871525764465,
      "loss_step_1": 0.0429,
      "loss_step_2": 0.0426,
      "loss_step_3": 0.0383,
      "step": 27090
    },
    {
      "entropy_step_1": 0.1901,
      "entropy_step_2": 0.1892,
      "entropy_step_3": 0.2011,
      "epoch": 0.8060438416465899,
      "grad_norm": 109.8268814086914,
      "learning_rate": 9.940607804911239e-06,
      "loss": 0.05130714178085327,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0496,
      "step": 27100
    },
    {
      "entropy_step_1": 0.1789,
      "entropy_step_2": 0.1801,
      "entropy_step_3": 0.1907,
      "epoch": 0.806341274798489,
      "grad_norm": 514.252197265625,
      "learning_rate": 9.911196560350582e-06,
      "loss": 0.050984245538711545,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0564,
      "step": 27110
    },
    {
      "entropy_step_1": 0.1596,
      "entropy_step_2": 0.1604,
      "entropy_step_3": 0.1636,
      "epoch": 0.8066387079503882,
      "grad_norm": 85.6751480102539,
      "learning_rate": 9.88182410225274e-06,
      "loss": 0.05099818706512451,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0416,
      "step": 27120
    },
    {
      "entropy_step_1": 0.1657,
      "entropy_step_2": 0.1667,
      "entropy_step_3": 0.1653,
      "epoch": 0.8069361411022873,
      "grad_norm": 60.578189849853516,
      "learning_rate": 9.852490459035962e-06,
      "loss": 0.05707752108573914,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0428,
      "step": 27130
    },
    {
      "entropy_step_1": 0.1671,
      "entropy_step_2": 0.1672,
      "entropy_step_3": 0.165,
      "epoch": 0.8072335742541864,
      "grad_norm": 81.00643157958984,
      "learning_rate": 9.823195659080941e-06,
      "loss": 0.05935595631599426,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0618,
      "step": 27140
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.1811,
      "entropy_step_3": 0.1985,
      "epoch": 0.8075310074060855,
      "grad_norm": 91.62298583984375,
      "learning_rate": 9.793939730730817e-06,
      "loss": 0.05405814051628113,
      "loss_step_1": 0.0585,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0646,
      "step": 27150
    },
    {
      "entropy_step_1": 0.1868,
      "entropy_step_2": 0.1859,
      "entropy_step_3": 0.1871,
      "epoch": 0.8078284405579846,
      "grad_norm": 36.166744232177734,
      "learning_rate": 9.764722702291068e-06,
      "loss": 0.053636443614959714,
      "loss_step_1": 0.0483,
      "loss_step_2": 0.0481,
      "loss_step_3": 0.0464,
      "step": 27160
    },
    {
      "entropy_step_1": 0.196,
      "entropy_step_2": 0.1956,
      "entropy_step_3": 0.1672,
      "epoch": 0.8081258737098838,
      "grad_norm": 262.3193054199219,
      "learning_rate": 9.735544602029617e-06,
      "loss": 0.05486850738525391,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0525,
      "step": 27170
    },
    {
      "entropy_step_1": 0.1739,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1646,
      "epoch": 0.8084233068617828,
      "grad_norm": 143.3649139404297,
      "learning_rate": 9.706405458176616e-06,
      "loss": 0.05433112382888794,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0534,
      "step": 27180
    },
    {
      "entropy_step_1": 0.1828,
      "entropy_step_2": 0.1801,
      "entropy_step_3": 0.168,
      "epoch": 0.8087207400136819,
      "grad_norm": 154.9728240966797,
      "learning_rate": 9.677305298924615e-06,
      "loss": 0.04819539785385132,
      "loss_step_1": 0.0467,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0473,
      "step": 27190
    },
    {
      "entropy_step_1": 0.1766,
      "entropy_step_2": 0.1766,
      "entropy_step_3": 0.1802,
      "epoch": 0.809018173165581,
      "grad_norm": 59.89656066894531,
      "learning_rate": 9.648244152428392e-06,
      "loss": 0.05499519109725952,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0571,
      "step": 27200
    },
    {
      "entropy_step_1": 0.2032,
      "entropy_step_2": 0.2018,
      "entropy_step_3": 0.1925,
      "epoch": 0.8093156063174801,
      "grad_norm": 351.8331604003906,
      "learning_rate": 9.619222046805048e-06,
      "loss": 0.04974181056022644,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0583,
      "step": 27210
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1859,
      "entropy_step_3": 0.1846,
      "epoch": 0.8096130394693792,
      "grad_norm": 98.73587036132812,
      "learning_rate": 9.59023901013381e-06,
      "loss": 0.05675364732742309,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0513,
      "step": 27220
    },
    {
      "entropy_step_1": 0.1934,
      "entropy_step_2": 0.1936,
      "entropy_step_3": 0.1823,
      "epoch": 0.8099104726212784,
      "grad_norm": 68.17870330810547,
      "learning_rate": 9.561295070456195e-06,
      "loss": 0.06328614354133606,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0677,
      "step": 27230
    },
    {
      "entropy_step_1": 0.203,
      "entropy_step_2": 0.2002,
      "entropy_step_3": 0.2036,
      "epoch": 0.8102079057731775,
      "grad_norm": 97.75910949707031,
      "learning_rate": 9.532390255775847e-06,
      "loss": 0.06193810701370239,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0644,
      "step": 27240
    },
    {
      "entropy_step_1": 0.2161,
      "entropy_step_2": 0.2127,
      "entropy_step_3": 0.1973,
      "epoch": 0.8105053389250766,
      "grad_norm": 183.05491638183594,
      "learning_rate": 9.503524594058555e-06,
      "loss": 0.05883685350418091,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0564,
      "step": 27250
    },
    {
      "entropy_step_1": 0.1867,
      "entropy_step_2": 0.1853,
      "entropy_step_3": 0.2078,
      "epoch": 0.8108027720769757,
      "grad_norm": 83.41854858398438,
      "learning_rate": 9.474698113232266e-06,
      "loss": 0.057850229740142825,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0662,
      "step": 27260
    },
    {
      "entropy_step_1": 0.1764,
      "entropy_step_2": 0.1767,
      "entropy_step_3": 0.1802,
      "epoch": 0.8111002052288748,
      "grad_norm": 64.42989349365234,
      "learning_rate": 9.445910841186978e-06,
      "loss": 0.06057213544845581,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0587,
      "step": 27270
    },
    {
      "entropy_step_1": 0.1835,
      "entropy_step_2": 0.1829,
      "entropy_step_3": 0.1886,
      "epoch": 0.811397638380774,
      "grad_norm": 55.75263214111328,
      "learning_rate": 9.41716280577477e-06,
      "loss": 0.05400930643081665,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0436,
      "step": 27280
    },
    {
      "entropy_step_1": 0.1917,
      "entropy_step_2": 0.1916,
      "entropy_step_3": 0.1761,
      "epoch": 0.8116950715326731,
      "grad_norm": 87.61705780029297,
      "learning_rate": 9.38845403480974e-06,
      "loss": 0.05290176868438721,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.058,
      "step": 27290
    },
    {
      "entropy_step_1": 0.1758,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.1544,
      "epoch": 0.8119925046845722,
      "grad_norm": 76.18836975097656,
      "learning_rate": 9.359784556068052e-06,
      "loss": 0.05116009712219238,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.0421,
      "step": 27300
    },
    {
      "entropy_step_1": 0.16,
      "entropy_step_2": 0.1638,
      "entropy_step_3": 0.1655,
      "epoch": 0.8122899378364713,
      "grad_norm": 45.2452392578125,
      "learning_rate": 9.331154397287789e-06,
      "loss": 0.05672905445098877,
      "loss_step_1": 0.0601,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0685,
      "step": 27310
    },
    {
      "entropy_step_1": 0.1703,
      "entropy_step_2": 0.1742,
      "entropy_step_3": 0.1723,
      "epoch": 0.8125873709883704,
      "grad_norm": 172.66879272460938,
      "learning_rate": 9.302563586169033e-06,
      "loss": 0.0473465621471405,
      "loss_step_1": 0.0488,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0449,
      "step": 27320
    },
    {
      "entropy_step_1": 0.1839,
      "entropy_step_2": 0.1884,
      "entropy_step_3": 0.2025,
      "epoch": 0.8128848041402694,
      "grad_norm": 78.42886352539062,
      "learning_rate": 9.274012150373762e-06,
      "loss": 0.058009034395217894,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.061,
      "step": 27330
    },
    {
      "entropy_step_1": 0.2007,
      "entropy_step_2": 0.205,
      "entropy_step_3": 0.205,
      "epoch": 0.8131822372921685,
      "grad_norm": 318.9114074707031,
      "learning_rate": 9.245500117525901e-06,
      "loss": 0.06493971347808838,
      "loss_step_1": 0.0668,
      "loss_step_2": 0.0663,
      "loss_step_3": 0.0673,
      "step": 27340
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.1867,
      "epoch": 0.8134796704440677,
      "grad_norm": 64.96910095214844,
      "learning_rate": 9.217027515211225e-06,
      "loss": 0.048818585276603696,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0545,
      "step": 27350
    },
    {
      "entropy_step_1": 0.172,
      "entropy_step_2": 0.1742,
      "entropy_step_3": 0.166,
      "epoch": 0.8137771035959668,
      "grad_norm": 87.05496978759766,
      "learning_rate": 9.18859437097736e-06,
      "loss": 0.052887117862701415,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0528,
      "step": 27360
    },
    {
      "entropy_step_1": 0.2029,
      "entropy_step_2": 0.2037,
      "entropy_step_3": 0.1976,
      "epoch": 0.8140745367478659,
      "grad_norm": 109.76853942871094,
      "learning_rate": 9.16020071233375e-06,
      "loss": 0.05393452644348144,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0546,
      "step": 27370
    },
    {
      "entropy_step_1": 0.1788,
      "entropy_step_2": 0.1808,
      "entropy_step_3": 0.1892,
      "epoch": 0.814371969899765,
      "grad_norm": 54.39186096191406,
      "learning_rate": 9.131846566751666e-06,
      "loss": 0.051887303590774536,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0595,
      "step": 27380
    },
    {
      "entropy_step_1": 0.1875,
      "entropy_step_2": 0.1908,
      "entropy_step_3": 0.1742,
      "epoch": 0.8146694030516641,
      "grad_norm": 59.49507522583008,
      "learning_rate": 9.103531961664118e-06,
      "loss": 0.0624355673789978,
      "loss_step_1": 0.0739,
      "loss_step_2": 0.073,
      "loss_step_3": 0.0936,
      "step": 27390
    },
    {
      "entropy_step_1": 0.1815,
      "entropy_step_2": 0.1843,
      "entropy_step_3": 0.2018,
      "epoch": 0.8149668362035632,
      "grad_norm": 94.45585632324219,
      "learning_rate": 9.075256924465881e-06,
      "loss": 0.05929840803146362,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0619,
      "step": 27400
    },
    {
      "entropy_step_1": 0.1783,
      "entropy_step_2": 0.1791,
      "entropy_step_3": 0.189,
      "epoch": 0.8152642693554624,
      "grad_norm": 30.53302001953125,
      "learning_rate": 9.047021482513419e-06,
      "loss": 0.05189482569694519,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.06,
      "step": 27410
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.2011,
      "entropy_step_3": 0.2081,
      "epoch": 0.8155617025073615,
      "grad_norm": 106.93376922607422,
      "learning_rate": 9.018825663124924e-06,
      "loss": 0.05367009043693542,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0571,
      "step": 27420
    },
    {
      "entropy_step_1": 0.1843,
      "entropy_step_2": 0.186,
      "entropy_step_3": 0.1824,
      "epoch": 0.8158591356592606,
      "grad_norm": 43.296329498291016,
      "learning_rate": 8.990669493580234e-06,
      "loss": 0.04611226618289947,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.0479,
      "loss_step_3": 0.0421,
      "step": 27430
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.1846,
      "entropy_step_3": 0.1828,
      "epoch": 0.8161565688111597,
      "grad_norm": 95.0047836303711,
      "learning_rate": 8.962553001120822e-06,
      "loss": 0.05171793103218079,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.051,
      "step": 27440
    },
    {
      "entropy_step_1": 0.1636,
      "entropy_step_2": 0.1658,
      "entropy_step_3": 0.1629,
      "epoch": 0.8164540019630588,
      "grad_norm": 110.71160888671875,
      "learning_rate": 8.934476212949777e-06,
      "loss": 0.04443917870521545,
      "loss_step_1": 0.0395,
      "loss_step_2": 0.0391,
      "loss_step_3": 0.04,
      "step": 27450
    },
    {
      "entropy_step_1": 0.1648,
      "entropy_step_2": 0.166,
      "entropy_step_3": 0.1413,
      "epoch": 0.816751435114958,
      "grad_norm": 109.58948516845703,
      "learning_rate": 8.906439156231761e-06,
      "loss": 0.0531241774559021,
      "loss_step_1": 0.0574,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0479,
      "step": 27460
    },
    {
      "entropy_step_1": 0.1688,
      "entropy_step_2": 0.1693,
      "entropy_step_3": 0.1681,
      "epoch": 0.8170488682668571,
      "grad_norm": 99.45525360107422,
      "learning_rate": 8.87844185809303e-06,
      "loss": 0.04668900966644287,
      "loss_step_1": 0.0451,
      "loss_step_2": 0.0445,
      "loss_step_3": 0.0411,
      "step": 27470
    },
    {
      "entropy_step_1": 0.1836,
      "entropy_step_2": 0.1829,
      "entropy_step_3": 0.183,
      "epoch": 0.8173463014187561,
      "grad_norm": 125.13640594482422,
      "learning_rate": 8.850484345621329e-06,
      "loss": 0.050099807977676394,
      "loss_step_1": 0.0474,
      "loss_step_2": 0.0469,
      "loss_step_3": 0.0481,
      "step": 27480
    },
    {
      "entropy_step_1": 0.16,
      "entropy_step_2": 0.1595,
      "entropy_step_3": 0.1401,
      "epoch": 0.8176437345706552,
      "grad_norm": 40.811790466308594,
      "learning_rate": 8.822566645865971e-06,
      "loss": 0.051526957750320436,
      "loss_step_1": 0.0494,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0407,
      "step": 27490
    },
    {
      "entropy_step_1": 0.1782,
      "entropy_step_2": 0.1775,
      "entropy_step_3": 0.1625,
      "epoch": 0.8179411677225543,
      "grad_norm": 264.822509765625,
      "learning_rate": 8.794688785837657e-06,
      "loss": 0.05491752028465271,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0558,
      "step": 27500
    },
    {
      "entropy_step_1": 0.177,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.1439,
      "epoch": 0.8182386008744534,
      "grad_norm": 54.462440490722656,
      "learning_rate": 8.766850792508636e-06,
      "loss": 0.05253893733024597,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0454,
      "step": 27510
    },
    {
      "entropy_step_1": 0.1575,
      "entropy_step_2": 0.1591,
      "entropy_step_3": 0.1443,
      "epoch": 0.8185360340263526,
      "grad_norm": 82.06842803955078,
      "learning_rate": 8.739052692812516e-06,
      "loss": 0.04806165099143982,
      "loss_step_1": 0.0444,
      "loss_step_2": 0.044,
      "loss_step_3": 0.042,
      "step": 27520
    },
    {
      "entropy_step_1": 0.1748,
      "entropy_step_2": 0.1765,
      "entropy_step_3": 0.1825,
      "epoch": 0.8188334671782517,
      "grad_norm": 104.8587875366211,
      "learning_rate": 8.711294513644368e-06,
      "loss": 0.06528111696243286,
      "loss_step_1": 0.067,
      "loss_step_2": 0.0661,
      "loss_step_3": 0.087,
      "step": 27530
    },
    {
      "entropy_step_1": 0.1774,
      "entropy_step_2": 0.1799,
      "entropy_step_3": 0.1938,
      "epoch": 0.8191309003301508,
      "grad_norm": 63.87154769897461,
      "learning_rate": 8.68357628186059e-06,
      "loss": 0.06358966827392579,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0658,
      "step": 27540
    },
    {
      "entropy_step_1": 0.1923,
      "entropy_step_2": 0.1965,
      "entropy_step_3": 0.1959,
      "epoch": 0.8194283334820499,
      "grad_norm": 82.88481903076172,
      "learning_rate": 8.655898024278952e-06,
      "loss": 0.051618915796279904,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.055,
      "step": 27550
    },
    {
      "entropy_step_1": 0.1777,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1548,
      "epoch": 0.819725766633949,
      "grad_norm": 51.33026123046875,
      "learning_rate": 8.628259767678531e-06,
      "loss": 0.055460613965988156,
      "loss_step_1": 0.0598,
      "loss_step_2": 0.0601,
      "loss_step_3": 0.0525,
      "step": 27560
    },
    {
      "entropy_step_1": 0.1931,
      "entropy_step_2": 0.197,
      "entropy_step_3": 0.2087,
      "epoch": 0.8200231997858481,
      "grad_norm": 114.8902587890625,
      "learning_rate": 8.600661538799737e-06,
      "loss": 0.05684747695922852,
      "loss_step_1": 0.0655,
      "loss_step_2": 0.0646,
      "loss_step_3": 0.0723,
      "step": 27570
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.1952,
      "entropy_step_3": 0.199,
      "epoch": 0.8203206329377473,
      "grad_norm": 44.86023712158203,
      "learning_rate": 8.573103364344232e-06,
      "loss": 0.05834214091300964,
      "loss_step_1": 0.0604,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0673,
      "step": 27580
    },
    {
      "entropy_step_1": 0.1784,
      "entropy_step_2": 0.1815,
      "entropy_step_3": 0.1841,
      "epoch": 0.8206180660896464,
      "grad_norm": 229.4903106689453,
      "learning_rate": 8.545585270974909e-06,
      "loss": 0.05217735767364502,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.052,
      "loss_step_3": 0.0461,
      "step": 27590
    },
    {
      "entropy_step_1": 0.172,
      "entropy_step_2": 0.1744,
      "entropy_step_3": 0.1732,
      "epoch": 0.8209154992415455,
      "grad_norm": 130.78492736816406,
      "learning_rate": 8.518107285315908e-06,
      "loss": 0.0544883131980896,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.057,
      "step": 27600
    },
    {
      "entropy_step_1": 0.1984,
      "entropy_step_2": 0.2002,
      "entropy_step_3": 0.2139,
      "epoch": 0.8212129323934446,
      "grad_norm": 128.0166778564453,
      "learning_rate": 8.490669433952541e-06,
      "loss": 0.05711832642555237,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0626,
      "step": 27610
    },
    {
      "entropy_step_1": 0.1758,
      "entropy_step_2": 0.1757,
      "entropy_step_3": 0.1941,
      "epoch": 0.8215103655453437,
      "grad_norm": 152.59400939941406,
      "learning_rate": 8.463271743431322e-06,
      "loss": 0.053520554304122926,
      "loss_step_1": 0.0418,
      "loss_step_2": 0.0415,
      "loss_step_3": 0.0471,
      "step": 27620
    },
    {
      "entropy_step_1": 0.1773,
      "entropy_step_2": 0.1775,
      "entropy_step_3": 0.1726,
      "epoch": 0.8218077986972427,
      "grad_norm": 55.47144317626953,
      "learning_rate": 8.435914240259884e-06,
      "loss": 0.04828590750694275,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0473,
      "step": 27630
    },
    {
      "entropy_step_1": 0.1856,
      "entropy_step_2": 0.1831,
      "entropy_step_3": 0.1938,
      "epoch": 0.8221052318491419,
      "grad_norm": 111.4009017944336,
      "learning_rate": 8.408596950906978e-06,
      "loss": 0.062238121032714845,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0627,
      "step": 27640
    },
    {
      "entropy_step_1": 0.1931,
      "entropy_step_2": 0.1919,
      "entropy_step_3": 0.1946,
      "epoch": 0.822402665001041,
      "grad_norm": 60.30396270751953,
      "learning_rate": 8.381319901802442e-06,
      "loss": 0.050401753187179564,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0458,
      "step": 27650
    },
    {
      "entropy_step_1": 0.1695,
      "entropy_step_2": 0.1695,
      "entropy_step_3": 0.1711,
      "epoch": 0.8227000981529401,
      "grad_norm": 69.82632446289062,
      "learning_rate": 8.354083119337219e-06,
      "loss": 0.04929390251636505,
      "loss_step_1": 0.0462,
      "loss_step_2": 0.0453,
      "loss_step_3": 0.0499,
      "step": 27660
    },
    {
      "entropy_step_1": 0.1686,
      "entropy_step_2": 0.1684,
      "entropy_step_3": 0.1584,
      "epoch": 0.8229975313048392,
      "grad_norm": 98.13349914550781,
      "learning_rate": 8.32688662986325e-06,
      "loss": 0.05403222441673279,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0501,
      "step": 27670
    },
    {
      "entropy_step_1": 0.1685,
      "entropy_step_2": 0.1705,
      "entropy_step_3": 0.1657,
      "epoch": 0.8232949644567383,
      "grad_norm": 93.57759094238281,
      "learning_rate": 8.299730459693512e-06,
      "loss": 0.056455254554748535,
      "loss_step_1": 0.055,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0556,
      "step": 27680
    },
    {
      "entropy_step_1": 0.1801,
      "entropy_step_2": 0.1823,
      "entropy_step_3": 0.1648,
      "epoch": 0.8235923976086374,
      "grad_norm": 58.68349838256836,
      "learning_rate": 8.272614635101956e-06,
      "loss": 0.06247115135192871,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0608,
      "loss_step_3": 0.0606,
      "step": 27690
    },
    {
      "entropy_step_1": 0.1789,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.188,
      "epoch": 0.8238898307605366,
      "grad_norm": 1340.1500244140625,
      "learning_rate": 8.245539182323536e-06,
      "loss": 0.054061079025268556,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0528,
      "step": 27700
    },
    {
      "entropy_step_1": 0.1848,
      "entropy_step_2": 0.1837,
      "entropy_step_3": 0.1829,
      "epoch": 0.8241872639124357,
      "grad_norm": 94.93693542480469,
      "learning_rate": 8.218504127554099e-06,
      "loss": 0.06258640885353088,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.059,
      "loss_step_3": 0.0572,
      "step": 27710
    },
    {
      "entropy_step_1": 0.1992,
      "entropy_step_2": 0.1978,
      "entropy_step_3": 0.2126,
      "epoch": 0.8244846970643348,
      "grad_norm": 137.86415100097656,
      "learning_rate": 8.191509496950462e-06,
      "loss": 0.0544150173664093,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0568,
      "step": 27720
    },
    {
      "entropy_step_1": 0.2004,
      "entropy_step_2": 0.1979,
      "entropy_step_3": 0.1841,
      "epoch": 0.8247821302162339,
      "grad_norm": 62.04273986816406,
      "learning_rate": 8.164555316630252e-06,
      "loss": 0.051478123664855956,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0534,
      "loss_step_3": 0.0538,
      "step": 27730
    },
    {
      "entropy_step_1": 0.1685,
      "entropy_step_2": 0.1669,
      "entropy_step_3": 0.1751,
      "epoch": 0.825079563368133,
      "grad_norm": 280.9319152832031,
      "learning_rate": 8.137641612672036e-06,
      "loss": 0.05021074414253235,
      "loss_step_1": 0.0463,
      "loss_step_2": 0.0462,
      "loss_step_3": 0.0462,
      "step": 27740
    },
    {
      "entropy_step_1": 0.1574,
      "entropy_step_2": 0.1565,
      "entropy_step_3": 0.1643,
      "epoch": 0.8253769965200322,
      "grad_norm": 171.60641479492188,
      "learning_rate": 8.110768411115165e-06,
      "loss": 0.04987340271472931,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0494,
      "loss_step_3": 0.0538,
      "step": 27750
    },
    {
      "entropy_step_1": 0.1649,
      "entropy_step_2": 0.1647,
      "entropy_step_3": 0.1644,
      "epoch": 0.8256744296719313,
      "grad_norm": 219.2447052001953,
      "learning_rate": 8.083935737959869e-06,
      "loss": 0.05068824291229248,
      "loss_step_1": 0.0535,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0468,
      "step": 27760
    },
    {
      "entropy_step_1": 0.1715,
      "entropy_step_2": 0.1729,
      "entropy_step_3": 0.1664,
      "epoch": 0.8259718628238304,
      "grad_norm": 139.71578979492188,
      "learning_rate": 8.057143619167073e-06,
      "loss": 0.05142989754676819,
      "loss_step_1": 0.0494,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.053,
      "step": 27770
    },
    {
      "entropy_step_1": 0.1708,
      "entropy_step_2": 0.1703,
      "entropy_step_3": 0.1769,
      "epoch": 0.8262692959757294,
      "grad_norm": 116.83667755126953,
      "learning_rate": 8.03039208065855e-06,
      "loss": 0.054502183198928834,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.051,
      "step": 27780
    },
    {
      "entropy_step_1": 0.1643,
      "entropy_step_2": 0.1646,
      "entropy_step_3": 0.178,
      "epoch": 0.8265667291276285,
      "grad_norm": 91.07199096679688,
      "learning_rate": 8.003681148316772e-06,
      "loss": 0.05868169069290161,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0527,
      "loss_step_3": 0.0524,
      "step": 27790
    },
    {
      "entropy_step_1": 0.1886,
      "entropy_step_2": 0.188,
      "entropy_step_3": 0.2114,
      "epoch": 0.8268641622795276,
      "grad_norm": 141.42404174804688,
      "learning_rate": 7.977010847984911e-06,
      "loss": 0.0526261568069458,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0647,
      "step": 27800
    },
    {
      "entropy_step_1": 0.1764,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.1918,
      "epoch": 0.8271615954314268,
      "grad_norm": 109.57869720458984,
      "learning_rate": 7.950381205466883e-06,
      "loss": 0.056158751249313354,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.058,
      "step": 27810
    },
    {
      "entropy_step_1": 0.1879,
      "entropy_step_2": 0.1874,
      "entropy_step_3": 0.1919,
      "epoch": 0.8274590285833259,
      "grad_norm": 112.70845794677734,
      "learning_rate": 7.923792246527184e-06,
      "loss": 0.05024382472038269,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0544,
      "step": 27820
    },
    {
      "entropy_step_1": 0.1795,
      "entropy_step_2": 0.1793,
      "entropy_step_3": 0.1807,
      "epoch": 0.827756461735225,
      "grad_norm": 50.53764724731445,
      "learning_rate": 7.89724399689103e-06,
      "loss": 0.05424159169197083,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0621,
      "step": 27830
    },
    {
      "entropy_step_1": 0.1753,
      "entropy_step_2": 0.1747,
      "entropy_step_3": 0.1883,
      "epoch": 0.8280538948871241,
      "grad_norm": 49.3902473449707,
      "learning_rate": 7.87073648224419e-06,
      "loss": 0.055009812116622925,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0553,
      "step": 27840
    },
    {
      "entropy_step_1": 0.1941,
      "entropy_step_2": 0.1937,
      "entropy_step_3": 0.2017,
      "epoch": 0.8283513280390232,
      "grad_norm": 48.57270431518555,
      "learning_rate": 7.844269728233062e-06,
      "loss": 0.055241113901138304,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.0629,
      "loss_step_3": 0.0584,
      "step": 27850
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.182,
      "entropy_step_3": 0.1788,
      "epoch": 0.8286487611909223,
      "grad_norm": 114.18180847167969,
      "learning_rate": 7.817843760464583e-06,
      "loss": 0.05425373911857605,
      "loss_step_1": 0.0482,
      "loss_step_2": 0.048,
      "loss_step_3": 0.0467,
      "step": 27860
    },
    {
      "entropy_step_1": 0.1918,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.185,
      "epoch": 0.8289461943428215,
      "grad_norm": 51.330909729003906,
      "learning_rate": 7.79145860450624e-06,
      "loss": 0.05773600935935974,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0617,
      "loss_step_3": 0.0589,
      "step": 27870
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.1806,
      "entropy_step_3": 0.2119,
      "epoch": 0.8292436274947206,
      "grad_norm": 125.89280700683594,
      "learning_rate": 7.765114285886016e-06,
      "loss": 0.05907989144325256,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.059,
      "step": 27880
    },
    {
      "entropy_step_1": 0.1993,
      "entropy_step_2": 0.1966,
      "entropy_step_3": 0.2016,
      "epoch": 0.8295410606466197,
      "grad_norm": 147.3596649169922,
      "learning_rate": 7.738810830092419e-06,
      "loss": 0.05794189572334289,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.057,
      "step": 27890
    },
    {
      "entropy_step_1": 0.1771,
      "entropy_step_2": 0.1735,
      "entropy_step_3": 0.1583,
      "epoch": 0.8298384937985188,
      "grad_norm": 53.64116668701172,
      "learning_rate": 7.712548262574394e-06,
      "loss": 0.05351541638374328,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0606,
      "step": 27900
    },
    {
      "entropy_step_1": 0.1966,
      "entropy_step_2": 0.1945,
      "entropy_step_3": 0.1908,
      "epoch": 0.8301359269504179,
      "grad_norm": 67.95960998535156,
      "learning_rate": 7.68632660874133e-06,
      "loss": 0.056626802682876586,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0516,
      "step": 27910
    },
    {
      "entropy_step_1": 0.1892,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1867,
      "epoch": 0.830433360102317,
      "grad_norm": 263.52142333984375,
      "learning_rate": 7.660145893963017e-06,
      "loss": 0.050809407234191896,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0574,
      "step": 27920
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1798,
      "entropy_step_3": 0.1844,
      "epoch": 0.8307307932542161,
      "grad_norm": 117.22303009033203,
      "learning_rate": 7.634006143569677e-06,
      "loss": 0.0539015531539917,
      "loss_step_1": 0.0595,
      "loss_step_2": 0.0589,
      "loss_step_3": 0.0694,
      "step": 27930
    },
    {
      "entropy_step_1": 0.1693,
      "entropy_step_2": 0.1681,
      "entropy_step_3": 0.1669,
      "epoch": 0.8310282264061152,
      "grad_norm": 189.10060119628906,
      "learning_rate": 7.60790738285187e-06,
      "loss": 0.053892111778259276,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.05,
      "step": 27940
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.1813,
      "entropy_step_3": 0.1832,
      "epoch": 0.8313256595580143,
      "grad_norm": 44.97697067260742,
      "learning_rate": 7.5818496370605005e-06,
      "loss": 0.04874639511108399,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.0492,
      "step": 27950
    },
    {
      "entropy_step_1": 0.176,
      "entropy_step_2": 0.1741,
      "entropy_step_3": 0.188,
      "epoch": 0.8316230927099134,
      "grad_norm": 120.40001678466797,
      "learning_rate": 7.5558329314067756e-06,
      "loss": 0.057329535484313965,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.05,
      "step": 27960
    },
    {
      "entropy_step_1": 0.1755,
      "entropy_step_2": 0.1746,
      "entropy_step_3": 0.1667,
      "epoch": 0.8319205258618125,
      "grad_norm": 44.7884635925293,
      "learning_rate": 7.5298572910622365e-06,
      "loss": 0.05254605412483215,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0455,
      "step": 27970
    },
    {
      "entropy_step_1": 0.1864,
      "entropy_step_2": 0.1848,
      "entropy_step_3": 0.176,
      "epoch": 0.8322179590137117,
      "grad_norm": 53.56235885620117,
      "learning_rate": 7.503922741158664e-06,
      "loss": 0.05909929275512695,
      "loss_step_1": 0.064,
      "loss_step_2": 0.0639,
      "loss_step_3": 0.0572,
      "step": 27980
    },
    {
      "entropy_step_1": 0.1899,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1649,
      "epoch": 0.8325153921656108,
      "grad_norm": 72.99501037597656,
      "learning_rate": 7.478029306788076e-06,
      "loss": 0.051134896278381345,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0539,
      "step": 27990
    },
    {
      "entropy_step_1": 0.1725,
      "entropy_step_2": 0.1689,
      "entropy_step_3": 0.1772,
      "epoch": 0.8328128253175099,
      "grad_norm": 74.78450775146484,
      "learning_rate": 7.452177013002732e-06,
      "loss": 0.052736663818359376,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.0607,
      "step": 28000
    },
    {
      "epoch": 0.8328128253175099,
      "eval_accuracy": 0.927,
      "eval_f1": 0.9269543464665416,
      "eval_loss": 0.08618003875017166,
      "eval_precision": 0.9280701754385965,
      "eval_recall": 0.927,
      "eval_runtime": 6.5662,
      "eval_samples_per_second": 76.148,
      "eval_steps_per_second": 9.595,
      "step": 28000
    },
    {
      "entropy_step_1": 0.1771,
      "entropy_step_2": 0.174,
      "entropy_step_3": 0.1831,
      "epoch": 0.833110258469409,
      "grad_norm": 48.39197540283203,
      "learning_rate": 7.426365884815062e-06,
      "loss": 0.050180470943450926,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0455,
      "step": 28010
    },
    {
      "entropy_step_1": 0.1789,
      "entropy_step_2": 0.1758,
      "entropy_step_3": 0.1801,
      "epoch": 0.8334076916213081,
      "grad_norm": 187.7471160888672,
      "learning_rate": 7.400595947197708e-06,
      "loss": 0.05753237009048462,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0503,
      "step": 28020
    },
    {
      "entropy_step_1": 0.2022,
      "entropy_step_2": 0.1988,
      "entropy_step_3": 0.1972,
      "epoch": 0.8337051247732072,
      "grad_norm": 372.3309631347656,
      "learning_rate": 7.374867225083415e-06,
      "loss": 0.06144528388977051,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.059,
      "step": 28030
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.1857,
      "entropy_step_3": 0.1765,
      "epoch": 0.8340025579251064,
      "grad_norm": 157.75352478027344,
      "learning_rate": 7.349179743365098e-06,
      "loss": 0.06445008516311646,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.051,
      "step": 28040
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1812,
      "entropy_step_3": 0.1775,
      "epoch": 0.8342999910770055,
      "grad_norm": 86.67451477050781,
      "learning_rate": 7.3235335268957095e-06,
      "loss": 0.05479282736778259,
      "loss_step_1": 0.0492,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0476,
      "step": 28050
    },
    {
      "entropy_step_1": 0.1888,
      "entropy_step_2": 0.1855,
      "entropy_step_3": 0.1805,
      "epoch": 0.8345974242289046,
      "grad_norm": 34.344749450683594,
      "learning_rate": 7.29792860048834e-06,
      "loss": 0.05976927876472473,
      "loss_step_1": 0.058,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0663,
      "step": 28060
    },
    {
      "entropy_step_1": 0.1681,
      "entropy_step_2": 0.165,
      "entropy_step_3": 0.162,
      "epoch": 0.8348948573808037,
      "grad_norm": 108.65086364746094,
      "learning_rate": 7.272364988916086e-06,
      "loss": 0.05304458737373352,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.0432,
      "step": 28070
    },
    {
      "entropy_step_1": 0.1843,
      "entropy_step_2": 0.1829,
      "entropy_step_3": 0.1969,
      "epoch": 0.8351922905327028,
      "grad_norm": 33.815818786621094,
      "learning_rate": 7.2468427169121115e-06,
      "loss": 0.05131205916404724,
      "loss_step_1": 0.052,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0513,
      "step": 28080
    },
    {
      "entropy_step_1": 0.176,
      "entropy_step_2": 0.1733,
      "entropy_step_3": 0.1845,
      "epoch": 0.8354897236846018,
      "grad_norm": 32.32892608642578,
      "learning_rate": 7.221361809169547e-06,
      "loss": 0.0501206636428833,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0466,
      "step": 28090
    },
    {
      "entropy_step_1": 0.1776,
      "entropy_step_2": 0.1747,
      "entropy_step_3": 0.1722,
      "epoch": 0.835787156836501,
      "grad_norm": 101.07667541503906,
      "learning_rate": 7.195922290341522e-06,
      "loss": 0.05470102429389954,
      "loss_step_1": 0.0554,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0545,
      "step": 28100
    },
    {
      "entropy_step_1": 0.1678,
      "entropy_step_2": 0.1666,
      "entropy_step_3": 0.1758,
      "epoch": 0.8360845899884001,
      "grad_norm": 59.509742736816406,
      "learning_rate": 7.1705241850411e-06,
      "loss": 0.04988945126533508,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0429,
      "step": 28110
    },
    {
      "entropy_step_1": 0.1584,
      "entropy_step_2": 0.1556,
      "entropy_step_3": 0.1585,
      "epoch": 0.8363820231402992,
      "grad_norm": 142.23660278320312,
      "learning_rate": 7.145167517841323e-06,
      "loss": 0.05051820278167725,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0507,
      "step": 28120
    },
    {
      "entropy_step_1": 0.1965,
      "entropy_step_2": 0.1925,
      "entropy_step_3": 0.1839,
      "epoch": 0.8366794562921983,
      "grad_norm": 107.88270568847656,
      "learning_rate": 7.119852313275094e-06,
      "loss": 0.05567841529846192,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0514,
      "step": 28130
    },
    {
      "entropy_step_1": 0.1662,
      "entropy_step_2": 0.1629,
      "entropy_step_3": 0.1754,
      "epoch": 0.8369768894440974,
      "grad_norm": 103.18074035644531,
      "learning_rate": 7.094578595835222e-06,
      "loss": 0.04797137379646301,
      "loss_step_1": 0.0503,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0554,
      "step": 28140
    },
    {
      "entropy_step_1": 0.1835,
      "entropy_step_2": 0.1792,
      "entropy_step_3": 0.1663,
      "epoch": 0.8372743225959965,
      "grad_norm": 91.1995849609375,
      "learning_rate": 7.069346389974374e-06,
      "loss": 0.05692607760429382,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0436,
      "step": 28150
    },
    {
      "entropy_step_1": 0.1855,
      "entropy_step_2": 0.1805,
      "entropy_step_3": 0.195,
      "epoch": 0.8375717557478957,
      "grad_norm": 66.25833892822266,
      "learning_rate": 7.0441557201050335e-06,
      "loss": 0.053322404623031616,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0694,
      "step": 28160
    },
    {
      "entropy_step_1": 0.1959,
      "entropy_step_2": 0.1917,
      "entropy_step_3": 0.1865,
      "epoch": 0.8378691888997948,
      "grad_norm": 109.1150894165039,
      "learning_rate": 7.019006610599549e-06,
      "loss": 0.054472440481185914,
      "loss_step_1": 0.0594,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.069,
      "step": 28170
    },
    {
      "entropy_step_1": 0.2068,
      "entropy_step_2": 0.2023,
      "entropy_step_3": 0.1993,
      "epoch": 0.8381666220516939,
      "grad_norm": 115.07359313964844,
      "learning_rate": 6.9938990857900065e-06,
      "loss": 0.05443033576011658,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0591,
      "loss_step_3": 0.058,
      "step": 28180
    },
    {
      "entropy_step_1": 0.1857,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.1892,
      "epoch": 0.838464055203593,
      "grad_norm": 77.84455108642578,
      "learning_rate": 6.968833169968281e-06,
      "loss": 0.051639598608016965,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0503,
      "step": 28190
    },
    {
      "entropy_step_1": 0.1784,
      "entropy_step_2": 0.1763,
      "entropy_step_3": 0.1484,
      "epoch": 0.8387614883554921,
      "grad_norm": 110.66744232177734,
      "learning_rate": 6.943808887385978e-06,
      "loss": 0.05864618420600891,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0532,
      "step": 28200
    },
    {
      "entropy_step_1": 0.1673,
      "entropy_step_2": 0.164,
      "entropy_step_3": 0.1449,
      "epoch": 0.8390589215073913,
      "grad_norm": 84.84320068359375,
      "learning_rate": 6.9188262622544516e-06,
      "loss": 0.05590488910675049,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0374,
      "step": 28210
    },
    {
      "entropy_step_1": 0.1887,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.1968,
      "epoch": 0.8393563546592904,
      "grad_norm": 146.42616271972656,
      "learning_rate": 6.8938853187447195e-06,
      "loss": 0.0533530056476593,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0631,
      "step": 28220
    },
    {
      "entropy_step_1": 0.1995,
      "entropy_step_2": 0.1976,
      "entropy_step_3": 0.1953,
      "epoch": 0.8396537878111895,
      "grad_norm": 62.617794036865234,
      "learning_rate": 6.868986080987488e-06,
      "loss": 0.05462751984596252,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0548,
      "step": 28230
    },
    {
      "entropy_step_1": 0.1663,
      "entropy_step_2": 0.1661,
      "entropy_step_3": 0.1618,
      "epoch": 0.8399512209630885,
      "grad_norm": 162.5062713623047,
      "learning_rate": 6.844128573073094e-06,
      "loss": 0.04966341257095337,
      "loss_step_1": 0.043,
      "loss_step_2": 0.0426,
      "loss_step_3": 0.041,
      "step": 28240
    },
    {
      "entropy_step_1": 0.179,
      "entropy_step_2": 0.1777,
      "entropy_step_3": 0.1909,
      "epoch": 0.8402486541149876,
      "grad_norm": 102.89517211914062,
      "learning_rate": 6.819312819051532e-06,
      "loss": 0.05681213140487671,
      "loss_step_1": 0.056,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0613,
      "step": 28250
    },
    {
      "entropy_step_1": 0.1699,
      "entropy_step_2": 0.1687,
      "entropy_step_3": 0.1599,
      "epoch": 0.8405460872668867,
      "grad_norm": 135.3939666748047,
      "learning_rate": 6.794538842932369e-06,
      "loss": 0.04946093559265137,
      "loss_step_1": 0.0468,
      "loss_step_2": 0.0468,
      "loss_step_3": 0.0441,
      "step": 28260
    },
    {
      "entropy_step_1": 0.1779,
      "entropy_step_2": 0.1772,
      "entropy_step_3": 0.1541,
      "epoch": 0.8408435204187859,
      "grad_norm": 53.57524871826172,
      "learning_rate": 6.769806668684797e-06,
      "loss": 0.055438727140426636,
      "loss_step_1": 0.0616,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0546,
      "step": 28270
    },
    {
      "entropy_step_1": 0.175,
      "entropy_step_2": 0.1765,
      "entropy_step_3": 0.1758,
      "epoch": 0.841140953570685,
      "grad_norm": 108.79643249511719,
      "learning_rate": 6.745116320237477e-06,
      "loss": 0.05382736921310425,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0478,
      "step": 28280
    },
    {
      "entropy_step_1": 0.1765,
      "entropy_step_2": 0.1767,
      "entropy_step_3": 0.1902,
      "epoch": 0.8414383867225841,
      "grad_norm": 77.95794677734375,
      "learning_rate": 6.7204678214786985e-06,
      "loss": 0.05814581513404846,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0575,
      "step": 28290
    },
    {
      "entropy_step_1": 0.1644,
      "entropy_step_2": 0.1662,
      "entropy_step_3": 0.131,
      "epoch": 0.8417358198744832,
      "grad_norm": 63.66585159301758,
      "learning_rate": 6.695861196256204e-06,
      "loss": 0.051068997383117674,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0382,
      "step": 28300
    },
    {
      "entropy_step_1": 0.1996,
      "entropy_step_2": 0.1998,
      "entropy_step_3": 0.2121,
      "epoch": 0.8420332530263823,
      "grad_norm": 112.58494567871094,
      "learning_rate": 6.671296468377242e-06,
      "loss": 0.06184842586517334,
      "loss_step_1": 0.0641,
      "loss_step_2": 0.0636,
      "loss_step_3": 0.0653,
      "step": 28310
    },
    {
      "entropy_step_1": 0.1852,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1768,
      "epoch": 0.8423306861782814,
      "grad_norm": 83.76473236083984,
      "learning_rate": 6.646773661608507e-06,
      "loss": 0.05657117366790772,
      "loss_step_1": 0.0576,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0501,
      "step": 28320
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1833,
      "entropy_step_3": 0.1763,
      "epoch": 0.8426281193301806,
      "grad_norm": 163.41859436035156,
      "learning_rate": 6.622292799676172e-06,
      "loss": 0.05519229173660278,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0546,
      "loss_step_3": 0.0573,
      "step": 28330
    },
    {
      "entropy_step_1": 0.1693,
      "entropy_step_2": 0.1732,
      "entropy_step_3": 0.165,
      "epoch": 0.8429255524820797,
      "grad_norm": 51.6901969909668,
      "learning_rate": 6.597853906265794e-06,
      "loss": 0.05652382969856262,
      "loss_step_1": 0.0593,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0561,
      "step": 28340
    },
    {
      "entropy_step_1": 0.1826,
      "entropy_step_2": 0.1838,
      "entropy_step_3": 0.1888,
      "epoch": 0.8432229856339788,
      "grad_norm": 300.09033203125,
      "learning_rate": 6.573457005022321e-06,
      "loss": 0.04887286424636841,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.046,
      "step": 28350
    },
    {
      "entropy_step_1": 0.1981,
      "entropy_step_2": 0.2018,
      "entropy_step_3": 0.2047,
      "epoch": 0.8435204187858779,
      "grad_norm": 150.10003662109375,
      "learning_rate": 6.549102119550121e-06,
      "loss": 0.058596014976501465,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0612,
      "step": 28360
    },
    {
      "entropy_step_1": 0.1775,
      "entropy_step_2": 0.1803,
      "entropy_step_3": 0.1709,
      "epoch": 0.843817851937777,
      "grad_norm": 346.2752685546875,
      "learning_rate": 6.524789273412835e-06,
      "loss": 0.05626042485237122,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0547,
      "loss_step_3": 0.0533,
      "step": 28370
    },
    {
      "entropy_step_1": 0.1696,
      "entropy_step_2": 0.1721,
      "entropy_step_3": 0.1616,
      "epoch": 0.8441152850896761,
      "grad_norm": 52.822566986083984,
      "learning_rate": 6.500518490133501e-06,
      "loss": 0.051169782876968384,
      "loss_step_1": 0.0556,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0536,
      "step": 28380
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.1827,
      "entropy_step_3": 0.1657,
      "epoch": 0.8444127182415752,
      "grad_norm": 221.0810089111328,
      "learning_rate": 6.4762897931944126e-06,
      "loss": 0.05529147982597351,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0472,
      "step": 28390
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1921,
      "entropy_step_3": 0.1844,
      "epoch": 0.8447101513934743,
      "grad_norm": 86.01910400390625,
      "learning_rate": 6.452103206037191e-06,
      "loss": 0.054635119438171384,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0502,
      "step": 28400
    },
    {
      "entropy_step_1": 0.1773,
      "entropy_step_2": 0.1803,
      "entropy_step_3": 0.1759,
      "epoch": 0.8450075845453734,
      "grad_norm": 210.90296936035156,
      "learning_rate": 6.427958752062668e-06,
      "loss": 0.052049100399017334,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0517,
      "step": 28410
    },
    {
      "entropy_step_1": 0.166,
      "entropy_step_2": 0.1698,
      "entropy_step_3": 0.1721,
      "epoch": 0.8453050176972725,
      "grad_norm": 69.55463409423828,
      "learning_rate": 6.403856454630941e-06,
      "loss": 0.04890067279338837,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0482,
      "step": 28420
    },
    {
      "entropy_step_1": 0.1774,
      "entropy_step_2": 0.1795,
      "entropy_step_3": 0.1852,
      "epoch": 0.8456024508491716,
      "grad_norm": 83.47843933105469,
      "learning_rate": 6.379796337061289e-06,
      "loss": 0.05461939573287964,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0484,
      "step": 28430
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.1878,
      "entropy_step_3": 0.1985,
      "epoch": 0.8458998840010707,
      "grad_norm": 161.80030822753906,
      "learning_rate": 6.3557784226322305e-06,
      "loss": 0.048410159349441526,
      "loss_step_1": 0.054,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0675,
      "step": 28440
    },
    {
      "entropy_step_1": 0.172,
      "entropy_step_2": 0.1749,
      "entropy_step_3": 0.1636,
      "epoch": 0.8461973171529699,
      "grad_norm": 231.64285278320312,
      "learning_rate": 6.331802734581415e-06,
      "loss": 0.059021925926208495,
      "loss_step_1": 0.0605,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0591,
      "step": 28450
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1899,
      "entropy_step_3": 0.1876,
      "epoch": 0.846494750304869,
      "grad_norm": 45.38887405395508,
      "learning_rate": 6.307869296105651e-06,
      "loss": 0.054048961400985716,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0594,
      "loss_step_3": 0.0748,
      "step": 28460
    },
    {
      "entropy_step_1": 0.2001,
      "entropy_step_2": 0.2038,
      "entropy_step_3": 0.2109,
      "epoch": 0.8467921834567681,
      "grad_norm": 158.38858032226562,
      "learning_rate": 6.283978130360846e-06,
      "loss": 0.05702344179153442,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0653,
      "step": 28470
    },
    {
      "entropy_step_1": 0.1749,
      "entropy_step_2": 0.1777,
      "entropy_step_3": 0.1658,
      "epoch": 0.8470896166086672,
      "grad_norm": 317.7591552734375,
      "learning_rate": 6.26012926046205e-06,
      "loss": 0.05422494411468506,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0483,
      "step": 28480
    },
    {
      "entropy_step_1": 0.1747,
      "entropy_step_2": 0.1779,
      "entropy_step_3": 0.1739,
      "epoch": 0.8473870497605663,
      "grad_norm": 71.20256805419922,
      "learning_rate": 6.236322709483366e-06,
      "loss": 0.055418151617050174,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.055,
      "step": 28490
    },
    {
      "entropy_step_1": 0.1714,
      "entropy_step_2": 0.1748,
      "entropy_step_3": 0.1824,
      "epoch": 0.8476844829124655,
      "grad_norm": 67.6072006225586,
      "learning_rate": 6.212558500457949e-06,
      "loss": 0.051512610912322995,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0522,
      "step": 28500
    },
    {
      "entropy_step_1": 0.1934,
      "entropy_step_2": 0.1966,
      "entropy_step_3": 0.191,
      "epoch": 0.8479819160643646,
      "grad_norm": 64.74650573730469,
      "learning_rate": 6.188836656377988e-06,
      "loss": 0.05199328064918518,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0517,
      "step": 28510
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.2045,
      "entropy_step_3": 0.1897,
      "epoch": 0.8482793492162637,
      "grad_norm": 71.58460998535156,
      "learning_rate": 6.16515720019471e-06,
      "loss": 0.05984630584716797,
      "loss_step_1": 0.0621,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0593,
      "step": 28520
    },
    {
      "entropy_step_1": 0.1871,
      "entropy_step_2": 0.1915,
      "entropy_step_3": 0.1869,
      "epoch": 0.8485767823681628,
      "grad_norm": 56.07693099975586,
      "learning_rate": 6.141520154818298e-06,
      "loss": 0.053925716876983644,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0533,
      "step": 28530
    },
    {
      "entropy_step_1": 0.163,
      "entropy_step_2": 0.167,
      "entropy_step_3": 0.1588,
      "epoch": 0.8488742155200618,
      "grad_norm": 210.94271850585938,
      "learning_rate": 6.117925543117919e-06,
      "loss": 0.05762397646903992,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0629,
      "step": 28540
    },
    {
      "entropy_step_1": 0.1977,
      "entropy_step_2": 0.2027,
      "entropy_step_3": 0.1918,
      "epoch": 0.8491716486719609,
      "grad_norm": 178.23304748535156,
      "learning_rate": 6.094373387921682e-06,
      "loss": 0.0603740394115448,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.057,
      "loss_step_3": 0.0469,
      "step": 28550
    },
    {
      "entropy_step_1": 0.1619,
      "entropy_step_2": 0.165,
      "entropy_step_3": 0.1589,
      "epoch": 0.84946908182386,
      "grad_norm": 76.00088500976562,
      "learning_rate": 6.0708637120166e-06,
      "loss": 0.04566846489906311,
      "loss_step_1": 0.0429,
      "loss_step_2": 0.0431,
      "loss_step_3": 0.0397,
      "step": 28560
    },
    {
      "entropy_step_1": 0.1799,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.1794,
      "epoch": 0.8497665149757592,
      "grad_norm": 68.27888488769531,
      "learning_rate": 6.047396538148631e-06,
      "loss": 0.05415703654289246,
      "loss_step_1": 0.0536,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.05,
      "step": 28570
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1756,
      "entropy_step_3": 0.1616,
      "epoch": 0.8500639481276583,
      "grad_norm": 80.37216186523438,
      "learning_rate": 6.023971889022562e-06,
      "loss": 0.049941590428352355,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0492,
      "step": 28580
    },
    {
      "entropy_step_1": 0.1731,
      "entropy_step_2": 0.1756,
      "entropy_step_3": 0.1812,
      "epoch": 0.8503613812795574,
      "grad_norm": 100.55516052246094,
      "learning_rate": 6.000589787302091e-06,
      "loss": 0.050256204605102536,
      "loss_step_1": 0.05,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.051,
      "step": 28590
    },
    {
      "entropy_step_1": 0.175,
      "entropy_step_2": 0.1777,
      "entropy_step_3": 0.1592,
      "epoch": 0.8506588144314565,
      "grad_norm": 48.06235122680664,
      "learning_rate": 5.977250255609679e-06,
      "loss": 0.050561392307281496,
      "loss_step_1": 0.0475,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0407,
      "step": 28600
    },
    {
      "entropy_step_1": 0.1875,
      "entropy_step_2": 0.1892,
      "entropy_step_3": 0.1892,
      "epoch": 0.8509562475833556,
      "grad_norm": 121.0284652709961,
      "learning_rate": 5.953953316526673e-06,
      "loss": 0.055054080486297605,
      "loss_step_1": 0.0607,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0593,
      "step": 28610
    },
    {
      "entropy_step_1": 0.1611,
      "entropy_step_2": 0.1629,
      "entropy_step_3": 0.1792,
      "epoch": 0.8512536807352548,
      "grad_norm": 202.42312622070312,
      "learning_rate": 5.930698992593153e-06,
      "loss": 0.04472721219062805,
      "loss_step_1": 0.0419,
      "loss_step_2": 0.0417,
      "loss_step_3": 0.0427,
      "step": 28620
    },
    {
      "entropy_step_1": 0.169,
      "entropy_step_2": 0.1706,
      "entropy_step_3": 0.1696,
      "epoch": 0.8515511138871539,
      "grad_norm": 94.49061584472656,
      "learning_rate": 5.9074873063080185e-06,
      "loss": 0.047876057028770444,
      "loss_step_1": 0.0453,
      "loss_step_2": 0.0451,
      "loss_step_3": 0.0466,
      "step": 28630
    },
    {
      "entropy_step_1": 0.1761,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.1636,
      "epoch": 0.851848547039053,
      "grad_norm": 60.455039978027344,
      "learning_rate": 5.884318280128886e-06,
      "loss": 0.05131875276565552,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0555,
      "step": 28640
    },
    {
      "entropy_step_1": 0.159,
      "entropy_step_2": 0.1603,
      "entropy_step_3": 0.152,
      "epoch": 0.8521459801909521,
      "grad_norm": 212.8666534423828,
      "learning_rate": 5.861191936472093e-06,
      "loss": 0.05080507397651672,
      "loss_step_1": 0.046,
      "loss_step_2": 0.046,
      "loss_step_3": 0.0449,
      "step": 28650
    },
    {
      "entropy_step_1": 0.1665,
      "entropy_step_2": 0.1678,
      "entropy_step_3": 0.1638,
      "epoch": 0.8524434133428512,
      "grad_norm": 211.43826293945312,
      "learning_rate": 5.838108297712686e-06,
      "loss": 0.053921055793762204,
      "loss_step_1": 0.0497,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0397,
      "step": 28660
    },
    {
      "entropy_step_1": 0.1776,
      "entropy_step_2": 0.1785,
      "entropy_step_3": 0.1646,
      "epoch": 0.8527408464947503,
      "grad_norm": 66.73004150390625,
      "learning_rate": 5.815067386184425e-06,
      "loss": 0.06465761065483093,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0547,
      "step": 28670
    },
    {
      "entropy_step_1": 0.1645,
      "entropy_step_2": 0.1641,
      "entropy_step_3": 0.1705,
      "epoch": 0.8530382796466495,
      "grad_norm": 92.6981201171875,
      "learning_rate": 5.7920692241796855e-06,
      "loss": 0.052901816368103025,
      "loss_step_1": 0.0465,
      "loss_step_2": 0.0464,
      "loss_step_3": 0.0486,
      "step": 28680
    },
    {
      "entropy_step_1": 0.1813,
      "entropy_step_2": 0.1803,
      "entropy_step_3": 0.1909,
      "epoch": 0.8533357127985485,
      "grad_norm": 87.46652221679688,
      "learning_rate": 5.769113833949513e-06,
      "loss": 0.05190612673759461,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.058,
      "step": 28690
    },
    {
      "entropy_step_1": 0.1749,
      "entropy_step_2": 0.1737,
      "entropy_step_3": 0.1615,
      "epoch": 0.8536331459504476,
      "grad_norm": 95.99945831298828,
      "learning_rate": 5.746201237703553e-06,
      "loss": 0.05110121965408325,
      "loss_step_1": 0.0518,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0461,
      "step": 28700
    },
    {
      "entropy_step_1": 0.1661,
      "entropy_step_2": 0.1662,
      "entropy_step_3": 0.1602,
      "epoch": 0.8539305791023467,
      "grad_norm": 79.45622253417969,
      "learning_rate": 5.7233314576100424e-06,
      "loss": 0.052876758575439456,
      "loss_step_1": 0.0526,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0558,
      "step": 28710
    },
    {
      "entropy_step_1": 0.1813,
      "entropy_step_2": 0.1823,
      "entropy_step_3": 0.1918,
      "epoch": 0.8542280122542458,
      "grad_norm": 66.01034545898438,
      "learning_rate": 5.700504515795829e-06,
      "loss": 0.05325074195861816,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.0583,
      "step": 28720
    },
    {
      "entropy_step_1": 0.1846,
      "entropy_step_2": 0.1851,
      "entropy_step_3": 0.1853,
      "epoch": 0.854525445406145,
      "grad_norm": 326.3972473144531,
      "learning_rate": 5.677720434346284e-06,
      "loss": 0.059257930517196654,
      "loss_step_1": 0.0661,
      "loss_step_2": 0.0673,
      "loss_step_3": 0.0537,
      "step": 28730
    },
    {
      "entropy_step_1": 0.1825,
      "entropy_step_2": 0.1843,
      "entropy_step_3": 0.1901,
      "epoch": 0.8548228785580441,
      "grad_norm": 96.45205688476562,
      "learning_rate": 5.654979235305302e-06,
      "loss": 0.059499037265777585,
      "loss_step_1": 0.06,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0593,
      "step": 28740
    },
    {
      "entropy_step_1": 0.1633,
      "entropy_step_2": 0.164,
      "entropy_step_3": 0.1619,
      "epoch": 0.8551203117099432,
      "grad_norm": 67.29574584960938,
      "learning_rate": 5.632280940675311e-06,
      "loss": 0.0596635639667511,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0496,
      "step": 28750
    },
    {
      "entropy_step_1": 0.1914,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.2029,
      "epoch": 0.8554177448618423,
      "grad_norm": 133.03721618652344,
      "learning_rate": 5.609625572417221e-06,
      "loss": 0.054972892999649046,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0571,
      "step": 28760
    },
    {
      "entropy_step_1": 0.1896,
      "entropy_step_2": 0.1888,
      "entropy_step_3": 0.1813,
      "epoch": 0.8557151780137414,
      "grad_norm": 50.46190643310547,
      "learning_rate": 5.587013152450415e-06,
      "loss": 0.05538906455039978,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0579,
      "loss_step_3": 0.0615,
      "step": 28770
    },
    {
      "entropy_step_1": 0.1981,
      "entropy_step_2": 0.1968,
      "entropy_step_3": 0.197,
      "epoch": 0.8560126111656405,
      "grad_norm": 106.23004150390625,
      "learning_rate": 5.564443702652705e-06,
      "loss": 0.046374392509460446,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.0559,
      "step": 28780
    },
    {
      "entropy_step_1": 0.1577,
      "entropy_step_2": 0.1584,
      "entropy_step_3": 0.1526,
      "epoch": 0.8563100443175397,
      "grad_norm": 34.857452392578125,
      "learning_rate": 5.5419172448603395e-06,
      "loss": 0.05062415599822998,
      "loss_step_1": 0.044,
      "loss_step_2": 0.0436,
      "loss_step_3": 0.0431,
      "step": 28790
    },
    {
      "entropy_step_1": 0.1997,
      "entropy_step_2": 0.1993,
      "entropy_step_3": 0.1817,
      "epoch": 0.8566074774694388,
      "grad_norm": 244.31105041503906,
      "learning_rate": 5.519433800867985e-06,
      "loss": 0.07305803894996643,
      "loss_step_1": 0.0712,
      "loss_step_2": 0.0713,
      "loss_step_3": 0.0663,
      "step": 28800
    },
    {
      "entropy_step_1": 0.1733,
      "entropy_step_2": 0.1731,
      "entropy_step_3": 0.1667,
      "epoch": 0.8569049106213379,
      "grad_norm": 62.454219818115234,
      "learning_rate": 5.496993392428668e-06,
      "loss": 0.0544049084186554,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.051,
      "step": 28810
    },
    {
      "entropy_step_1": 0.1927,
      "entropy_step_2": 0.1922,
      "entropy_step_3": 0.1937,
      "epoch": 0.857202343773237,
      "grad_norm": 90.4268798828125,
      "learning_rate": 5.4745960412538155e-06,
      "loss": 0.05630825757980347,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.055,
      "step": 28820
    },
    {
      "entropy_step_1": 0.1769,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1856,
      "epoch": 0.8574997769251361,
      "grad_norm": 102.70836639404297,
      "learning_rate": 5.452241769013128e-06,
      "loss": 0.04631878733634949,
      "loss_step_1": 0.0423,
      "loss_step_2": 0.0422,
      "loss_step_3": 0.0429,
      "step": 28830
    },
    {
      "entropy_step_1": 0.1723,
      "entropy_step_2": 0.1713,
      "entropy_step_3": 0.1759,
      "epoch": 0.8577972100770351,
      "grad_norm": 91.1958236694336,
      "learning_rate": 5.4299305973346985e-06,
      "loss": 0.050998526811599734,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.0583,
      "step": 28840
    },
    {
      "entropy_step_1": 0.185,
      "entropy_step_2": 0.1847,
      "entropy_step_3": 0.1968,
      "epoch": 0.8580946432289343,
      "grad_norm": 167.24526977539062,
      "learning_rate": 5.407662547804876e-06,
      "loss": 0.051507973670959474,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0531,
      "loss_step_3": 0.0619,
      "step": 28850
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1865,
      "epoch": 0.8583920763808334,
      "grad_norm": 58.9383659362793,
      "learning_rate": 5.385437641968311e-06,
      "loss": 0.05194282531738281,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.048,
      "step": 28860
    },
    {
      "entropy_step_1": 0.1703,
      "entropy_step_2": 0.1683,
      "entropy_step_3": 0.1661,
      "epoch": 0.8586895095327325,
      "grad_norm": 145.28065490722656,
      "learning_rate": 5.363255901327885e-06,
      "loss": 0.05235227346420288,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0578,
      "step": 28870
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.1779,
      "entropy_step_3": 0.1596,
      "epoch": 0.8589869426846316,
      "grad_norm": 120.98243713378906,
      "learning_rate": 5.341117347344754e-06,
      "loss": 0.05004074573516846,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.05,
      "loss_step_3": 0.0391,
      "step": 28880
    },
    {
      "entropy_step_1": 0.1731,
      "entropy_step_2": 0.1729,
      "entropy_step_3": 0.1691,
      "epoch": 0.8592843758365307,
      "grad_norm": 65.94908142089844,
      "learning_rate": 5.3190220014382615e-06,
      "loss": 0.05878115892410278,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0481,
      "step": 28890
    },
    {
      "entropy_step_1": 0.1876,
      "entropy_step_2": 0.1869,
      "entropy_step_3": 0.1977,
      "epoch": 0.8595818089884298,
      "grad_norm": 91.8592529296875,
      "learning_rate": 5.296969884985947e-06,
      "loss": 0.06020437479019165,
      "loss_step_1": 0.0454,
      "loss_step_2": 0.0451,
      "loss_step_3": 0.0485,
      "step": 28900
    },
    {
      "entropy_step_1": 0.2029,
      "entropy_step_2": 0.2014,
      "entropy_step_3": 0.2174,
      "epoch": 0.859879242140329,
      "grad_norm": 93.7215805053711,
      "learning_rate": 5.27496101932356e-06,
      "loss": 0.055393409729003903,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.065,
      "step": 28910
    },
    {
      "entropy_step_1": 0.1963,
      "entropy_step_2": 0.1937,
      "entropy_step_3": 0.1981,
      "epoch": 0.8601766752922281,
      "grad_norm": 136.00181579589844,
      "learning_rate": 5.2529954257449385e-06,
      "loss": 0.05582925081253052,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0516,
      "step": 28920
    },
    {
      "entropy_step_1": 0.1792,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.1857,
      "epoch": 0.8604741084441272,
      "grad_norm": 75.43812561035156,
      "learning_rate": 5.231073125502122e-06,
      "loss": 0.05500972867012024,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0522,
      "step": 28930
    },
    {
      "entropy_step_1": 0.198,
      "entropy_step_2": 0.1949,
      "entropy_step_3": 0.1904,
      "epoch": 0.8607715415960263,
      "grad_norm": 90.2042465209961,
      "learning_rate": 5.2091941398052235e-06,
      "loss": 0.05409574508666992,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0593,
      "loss_step_3": 0.0613,
      "step": 28940
    },
    {
      "entropy_step_1": 0.2008,
      "entropy_step_2": 0.199,
      "entropy_step_3": 0.2136,
      "epoch": 0.8610689747479254,
      "grad_norm": 232.81248474121094,
      "learning_rate": 5.187358489822463e-06,
      "loss": 0.06080581545829773,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0611,
      "loss_step_3": 0.0537,
      "step": 28950
    },
    {
      "entropy_step_1": 0.1856,
      "entropy_step_2": 0.1842,
      "entropy_step_3": 0.1898,
      "epoch": 0.8613664078998245,
      "grad_norm": 69.67938232421875,
      "learning_rate": 5.165566196680133e-06,
      "loss": 0.04992128908634186,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.0502,
      "loss_step_3": 0.0573,
      "step": 28960
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1782,
      "entropy_step_3": 0.1803,
      "epoch": 0.8616638410517237,
      "grad_norm": 31.142148971557617,
      "learning_rate": 5.143817281462571e-06,
      "loss": 0.05145231485366821,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0578,
      "step": 28970
    },
    {
      "entropy_step_1": 0.1787,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.1675,
      "epoch": 0.8619612742036228,
      "grad_norm": 68.3557357788086,
      "learning_rate": 5.122111765212128e-06,
      "loss": 0.05149380564689636,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0484,
      "loss_step_3": 0.0464,
      "step": 28980
    },
    {
      "entropy_step_1": 0.1803,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.1794,
      "epoch": 0.8622587073555218,
      "grad_norm": 85.00010681152344,
      "learning_rate": 5.100449668929219e-06,
      "loss": 0.0625561535358429,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0538,
      "step": 28990
    },
    {
      "entropy_step_1": 0.1872,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.2111,
      "epoch": 0.8625561405074209,
      "grad_norm": 107.26691436767578,
      "learning_rate": 5.078831013572205e-06,
      "loss": 0.05425798296928406,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0619,
      "step": 29000
    },
    {
      "epoch": 0.8625561405074209,
      "eval_accuracy": 0.921,
      "eval_f1": 0.9209651456292224,
      "eval_loss": 0.08691337704658508,
      "eval_precision": 0.921743956338982,
      "eval_recall": 0.921,
      "eval_runtime": 6.4683,
      "eval_samples_per_second": 77.3,
      "eval_steps_per_second": 9.74,
      "step": 29000
    },
    {
      "entropy_step_1": 0.1924,
      "entropy_step_2": 0.1902,
      "entropy_step_3": 0.1963,
      "epoch": 0.86285357365932,
      "grad_norm": 81.69336700439453,
      "learning_rate": 5.057255820057421e-06,
      "loss": 0.06486771106719971,
      "loss_step_1": 0.0703,
      "loss_step_2": 0.0705,
      "loss_step_3": 0.0758,
      "step": 29010
    },
    {
      "entropy_step_1": 0.1803,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.194,
      "epoch": 0.8631510068112191,
      "grad_norm": 147.53390502929688,
      "learning_rate": 5.03572410925916e-06,
      "loss": 0.051316410303115845,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0516,
      "loss_step_3": 0.0568,
      "step": 29020
    },
    {
      "entropy_step_1": 0.1794,
      "entropy_step_2": 0.1767,
      "entropy_step_3": 0.1881,
      "epoch": 0.8634484399631183,
      "grad_norm": 64.37777709960938,
      "learning_rate": 5.014235902009656e-06,
      "loss": 0.05637383460998535,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0568,
      "loss_step_3": 0.0626,
      "step": 29030
    },
    {
      "entropy_step_1": 0.201,
      "entropy_step_2": 0.1986,
      "entropy_step_3": 0.2007,
      "epoch": 0.8637458731150174,
      "grad_norm": 1032.992431640625,
      "learning_rate": 4.992791219099041e-06,
      "loss": 0.05181877017021179,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0588,
      "step": 29040
    },
    {
      "entropy_step_1": 0.199,
      "entropy_step_2": 0.198,
      "entropy_step_3": 0.1781,
      "epoch": 0.8640433062669165,
      "grad_norm": 176.36279296875,
      "learning_rate": 4.971390081275334e-06,
      "loss": 0.0588593065738678,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.049,
      "step": 29050
    },
    {
      "entropy_step_1": 0.1819,
      "entropy_step_2": 0.182,
      "entropy_step_3": 0.1819,
      "epoch": 0.8643407394188156,
      "grad_norm": 255.520751953125,
      "learning_rate": 4.950032509244423e-06,
      "loss": 0.046191900968551636,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0473,
      "loss_step_3": 0.0478,
      "step": 29060
    },
    {
      "entropy_step_1": 0.1866,
      "entropy_step_2": 0.1857,
      "entropy_step_3": 0.186,
      "epoch": 0.8646381725707147,
      "grad_norm": 257.456787109375,
      "learning_rate": 4.928718523670045e-06,
      "loss": 0.053509187698364255,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0485,
      "step": 29070
    },
    {
      "entropy_step_1": 0.1712,
      "entropy_step_2": 0.1707,
      "entropy_step_3": 0.1674,
      "epoch": 0.8649356057226139,
      "grad_norm": 101.75361633300781,
      "learning_rate": 4.907448145173788e-06,
      "loss": 0.05482549667358398,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0555,
      "loss_step_3": 0.0574,
      "step": 29080
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.187,
      "entropy_step_3": 0.1908,
      "epoch": 0.865233038874513,
      "grad_norm": 53.77777099609375,
      "learning_rate": 4.886221394335022e-06,
      "loss": 0.048434489965438844,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0667,
      "step": 29090
    },
    {
      "entropy_step_1": 0.1884,
      "entropy_step_2": 0.1883,
      "entropy_step_3": 0.1965,
      "epoch": 0.8655304720264121,
      "grad_norm": 108.08840942382812,
      "learning_rate": 4.865038291690921e-06,
      "loss": 0.05524916052818298,
      "loss_step_1": 0.0617,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0553,
      "step": 29100
    },
    {
      "entropy_step_1": 0.1826,
      "entropy_step_2": 0.1814,
      "entropy_step_3": 0.1668,
      "epoch": 0.8658279051783112,
      "grad_norm": 110.67262268066406,
      "learning_rate": 4.843898857736412e-06,
      "loss": 0.05320161581039429,
      "loss_step_1": 0.0524,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.05,
      "step": 29110
    },
    {
      "entropy_step_1": 0.175,
      "entropy_step_2": 0.1748,
      "entropy_step_3": 0.1918,
      "epoch": 0.8661253383302103,
      "grad_norm": 110.04277801513672,
      "learning_rate": 4.822803112924207e-06,
      "loss": 0.04970325231552124,
      "loss_step_1": 0.0467,
      "loss_step_2": 0.0465,
      "loss_step_3": 0.0516,
      "step": 29120
    },
    {
      "entropy_step_1": 0.1779,
      "entropy_step_2": 0.1782,
      "entropy_step_3": 0.1667,
      "epoch": 0.8664227714821094,
      "grad_norm": 99.20436096191406,
      "learning_rate": 4.801751077664701e-06,
      "loss": 0.04767073392868042,
      "loss_step_1": 0.0447,
      "loss_step_2": 0.0443,
      "loss_step_3": 0.0333,
      "step": 29130
    },
    {
      "entropy_step_1": 0.1653,
      "entropy_step_2": 0.1667,
      "entropy_step_3": 0.1523,
      "epoch": 0.8667202046340086,
      "grad_norm": 295.0946350097656,
      "learning_rate": 4.780742772326063e-06,
      "loss": 0.06090835332870483,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.061,
      "loss_step_3": 0.0582,
      "step": 29140
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1794,
      "entropy_step_3": 0.1811,
      "epoch": 0.8670176377859076,
      "grad_norm": 52.59861373901367,
      "learning_rate": 4.759778217234068e-06,
      "loss": 0.05409176349639892,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0566,
      "loss_step_3": 0.0534,
      "step": 29150
    },
    {
      "entropy_step_1": 0.1748,
      "entropy_step_2": 0.1748,
      "entropy_step_3": 0.1872,
      "epoch": 0.8673150709378067,
      "grad_norm": 162.92263793945312,
      "learning_rate": 4.738857432672239e-06,
      "loss": 0.045197388529777525,
      "loss_step_1": 0.0435,
      "loss_step_2": 0.0433,
      "loss_step_3": 0.0479,
      "step": 29160
    },
    {
      "entropy_step_1": 0.17,
      "entropy_step_2": 0.1693,
      "entropy_step_3": 0.1925,
      "epoch": 0.8676125040897058,
      "grad_norm": 195.6207275390625,
      "learning_rate": 4.717980438881709e-06,
      "loss": 0.0536444365978241,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0603,
      "step": 29170
    },
    {
      "entropy_step_1": 0.1798,
      "entropy_step_2": 0.1787,
      "entropy_step_3": 0.1878,
      "epoch": 0.8679099372416049,
      "grad_norm": 626.7207641601562,
      "learning_rate": 4.697147256061257e-06,
      "loss": 0.05682559609413147,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0626,
      "step": 29180
    },
    {
      "entropy_step_1": 0.1847,
      "entropy_step_2": 0.1843,
      "entropy_step_3": 0.1934,
      "epoch": 0.868207370393504,
      "grad_norm": 26.836177825927734,
      "learning_rate": 4.676357904367268e-06,
      "loss": 0.0529729962348938,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.054,
      "loss_step_3": 0.05,
      "step": 29190
    },
    {
      "entropy_step_1": 0.1775,
      "entropy_step_2": 0.1762,
      "entropy_step_3": 0.2096,
      "epoch": 0.8685048035454032,
      "grad_norm": 148.40252685546875,
      "learning_rate": 4.6556124039137285e-06,
      "loss": 0.057889968156814575,
      "loss_step_1": 0.0565,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0765,
      "step": 29200
    },
    {
      "entropy_step_1": 0.1667,
      "entropy_step_2": 0.1664,
      "entropy_step_3": 0.178,
      "epoch": 0.8688022366973023,
      "grad_norm": 53.529598236083984,
      "learning_rate": 4.634910774772178e-06,
      "loss": 0.04984340369701386,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.0555,
      "step": 29210
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1741,
      "entropy_step_3": 0.1678,
      "epoch": 0.8690996698492014,
      "grad_norm": 88.69329071044922,
      "learning_rate": 4.614253036971744e-06,
      "loss": 0.04858801960945129,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.042,
      "step": 29220
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1686,
      "epoch": 0.8693971030011005,
      "grad_norm": 61.63928985595703,
      "learning_rate": 4.593639210499057e-06,
      "loss": 0.051460057497024536,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0512,
      "loss_step_3": 0.0495,
      "step": 29230
    },
    {
      "entropy_step_1": 0.1629,
      "entropy_step_2": 0.1617,
      "entropy_step_3": 0.1805,
      "epoch": 0.8696945361529996,
      "grad_norm": 96.73866271972656,
      "learning_rate": 4.5730693152982785e-06,
      "loss": 0.05918816328048706,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0548,
      "loss_step_3": 0.0469,
      "step": 29240
    },
    {
      "entropy_step_1": 0.1825,
      "entropy_step_2": 0.1822,
      "entropy_step_3": 0.1842,
      "epoch": 0.8699919693048987,
      "grad_norm": 440.6553955078125,
      "learning_rate": 4.5525433712710615e-06,
      "loss": 0.058233118057250975,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.064,
      "loss_step_3": 0.0651,
      "step": 29250
    },
    {
      "entropy_step_1": 0.1759,
      "entropy_step_2": 0.1761,
      "entropy_step_3": 0.1651,
      "epoch": 0.8702894024567979,
      "grad_norm": 45.77598190307617,
      "learning_rate": 4.532061398276522e-06,
      "loss": 0.05204371809959411,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0562,
      "step": 29260
    },
    {
      "entropy_step_1": 0.1797,
      "entropy_step_2": 0.1793,
      "entropy_step_3": 0.1871,
      "epoch": 0.870586835608697,
      "grad_norm": 129.64804077148438,
      "learning_rate": 4.511623416131272e-06,
      "loss": 0.05165377855300903,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0594,
      "step": 29270
    },
    {
      "entropy_step_1": 0.1584,
      "entropy_step_2": 0.1592,
      "entropy_step_3": 0.1545,
      "epoch": 0.8708842687605961,
      "grad_norm": 40.52796173095703,
      "learning_rate": 4.491229444609329e-06,
      "loss": 0.056244033575057986,
      "loss_step_1": 0.0505,
      "loss_step_2": 0.0505,
      "loss_step_3": 0.0505,
      "step": 29280
    },
    {
      "entropy_step_1": 0.1861,
      "entropy_step_2": 0.1857,
      "entropy_step_3": 0.1994,
      "epoch": 0.8711817019124952,
      "grad_norm": 147.51800537109375,
      "learning_rate": 4.470879503442132e-06,
      "loss": 0.053068381547927854,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.061,
      "step": 29290
    },
    {
      "entropy_step_1": 0.1869,
      "entropy_step_2": 0.188,
      "entropy_step_3": 0.1942,
      "epoch": 0.8714791350643942,
      "grad_norm": 112.16609191894531,
      "learning_rate": 4.450573612318515e-06,
      "loss": 0.06042516231536865,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0578,
      "loss_step_3": 0.0628,
      "step": 29300
    },
    {
      "entropy_step_1": 0.1937,
      "entropy_step_2": 0.1936,
      "entropy_step_3": 0.2067,
      "epoch": 0.8717765682162933,
      "grad_norm": 150.12728881835938,
      "learning_rate": 4.430311790884734e-06,
      "loss": 0.06042754054069519,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0612,
      "step": 29310
    },
    {
      "entropy_step_1": 0.1971,
      "entropy_step_2": 0.1969,
      "entropy_step_3": 0.1962,
      "epoch": 0.8720740013681925,
      "grad_norm": 66.69120025634766,
      "learning_rate": 4.410094058744357e-06,
      "loss": 0.05515967011451721,
      "loss_step_1": 0.0487,
      "loss_step_2": 0.0487,
      "loss_step_3": 0.0447,
      "step": 29320
    },
    {
      "entropy_step_1": 0.1733,
      "entropy_step_2": 0.1738,
      "entropy_step_3": 0.1662,
      "epoch": 0.8723714345200916,
      "grad_norm": 144.09136962890625,
      "learning_rate": 4.3899204354583225e-06,
      "loss": 0.04886681139469147,
      "loss_step_1": 0.0486,
      "loss_step_2": 0.048,
      "loss_step_3": 0.0456,
      "step": 29330
    },
    {
      "entropy_step_1": 0.1592,
      "entropy_step_2": 0.1591,
      "entropy_step_3": 0.1479,
      "epoch": 0.8726688676719907,
      "grad_norm": 78.03034210205078,
      "learning_rate": 4.369790940544866e-06,
      "loss": 0.052037286758422854,
      "loss_step_1": 0.0503,
      "loss_step_2": 0.0496,
      "loss_step_3": 0.0469,
      "step": 29340
    },
    {
      "entropy_step_1": 0.1748,
      "entropy_step_2": 0.1757,
      "entropy_step_3": 0.1582,
      "epoch": 0.8729663008238898,
      "grad_norm": 56.31568908691406,
      "learning_rate": 4.34970559347958e-06,
      "loss": 0.06296008229255676,
      "loss_step_1": 0.0652,
      "loss_step_2": 0.0641,
      "loss_step_3": 0.0528,
      "step": 29350
    },
    {
      "entropy_step_1": 0.1876,
      "entropy_step_2": 0.1886,
      "entropy_step_3": 0.2049,
      "epoch": 0.8732637339757889,
      "grad_norm": 113.16852569580078,
      "learning_rate": 4.329664413695278e-06,
      "loss": 0.05010383725166321,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.061,
      "step": 29360
    },
    {
      "entropy_step_1": 0.1742,
      "entropy_step_2": 0.1742,
      "entropy_step_3": 0.1593,
      "epoch": 0.873561167127688,
      "grad_norm": 93.95831298828125,
      "learning_rate": 4.3096674205821145e-06,
      "loss": 0.05623990297317505,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0613,
      "loss_step_3": 0.0688,
      "step": 29370
    },
    {
      "entropy_step_1": 0.1942,
      "entropy_step_2": 0.1946,
      "entropy_step_3": 0.1811,
      "epoch": 0.8738586002795872,
      "grad_norm": 74.82335662841797,
      "learning_rate": 4.289714633487407e-06,
      "loss": 0.05728927254676819,
      "loss_step_1": 0.0609,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0544,
      "step": 29380
    },
    {
      "entropy_step_1": 0.176,
      "entropy_step_2": 0.1764,
      "entropy_step_3": 0.1778,
      "epoch": 0.8741560334314863,
      "grad_norm": 74.3353042602539,
      "learning_rate": 4.269806071715782e-06,
      "loss": 0.05506452918052673,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0516,
      "step": 29390
    },
    {
      "entropy_step_1": 0.1746,
      "entropy_step_2": 0.1755,
      "entropy_step_3": 0.1732,
      "epoch": 0.8744534665833854,
      "grad_norm": 64.14389038085938,
      "learning_rate": 4.24994175452903e-06,
      "loss": 0.054993027448654176,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0517,
      "loss_step_3": 0.0616,
      "step": 29400
    },
    {
      "entropy_step_1": 0.18,
      "entropy_step_2": 0.1811,
      "entropy_step_3": 0.1719,
      "epoch": 0.8747508997352845,
      "grad_norm": 136.8904571533203,
      "learning_rate": 4.230121701146151e-06,
      "loss": 0.053672367334365846,
      "loss_step_1": 0.0523,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0458,
      "step": 29410
    },
    {
      "entropy_step_1": 0.1843,
      "entropy_step_2": 0.1839,
      "entropy_step_3": 0.2152,
      "epoch": 0.8750483328871836,
      "grad_norm": 67.4659194946289,
      "learning_rate": 4.210345930743309e-06,
      "loss": 0.062191784381866455,
      "loss_step_1": 0.0628,
      "loss_step_2": 0.062,
      "loss_step_3": 0.0724,
      "step": 29420
    },
    {
      "entropy_step_1": 0.1821,
      "entropy_step_2": 0.1817,
      "entropy_step_3": 0.1829,
      "epoch": 0.8753457660390828,
      "grad_norm": 130.91082763671875,
      "learning_rate": 4.1906144624538406e-06,
      "loss": 0.05964751243591308,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0497,
      "step": 29430
    },
    {
      "entropy_step_1": 0.1854,
      "entropy_step_2": 0.1848,
      "entropy_step_3": 0.204,
      "epoch": 0.8756431991909819,
      "grad_norm": 225.62586975097656,
      "learning_rate": 4.170927315368211e-06,
      "loss": 0.05283790230751038,
      "loss_step_1": 0.0586,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0704,
      "step": 29440
    },
    {
      "entropy_step_1": 0.1759,
      "entropy_step_2": 0.1758,
      "entropy_step_3": 0.1741,
      "epoch": 0.8759406323428809,
      "grad_norm": 45.41159439086914,
      "learning_rate": 4.151284508533987e-06,
      "loss": 0.040751099586486816,
      "loss_step_1": 0.0437,
      "loss_step_2": 0.0436,
      "loss_step_3": 0.0465,
      "step": 29450
    },
    {
      "entropy_step_1": 0.1656,
      "entropy_step_2": 0.1654,
      "entropy_step_3": 0.1747,
      "epoch": 0.87623806549478,
      "grad_norm": 83.40444946289062,
      "learning_rate": 4.1316860609558864e-06,
      "loss": 0.042640841007232665,
      "loss_step_1": 0.0398,
      "loss_step_2": 0.0396,
      "loss_step_3": 0.0475,
      "step": 29460
    },
    {
      "entropy_step_1": 0.1683,
      "entropy_step_2": 0.168,
      "entropy_step_3": 0.1714,
      "epoch": 0.8765354986466791,
      "grad_norm": 386.4276123046875,
      "learning_rate": 4.112131991595625e-06,
      "loss": 0.05496073961257934,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0562,
      "step": 29470
    },
    {
      "entropy_step_1": 0.161,
      "entropy_step_2": 0.1611,
      "entropy_step_3": 0.1708,
      "epoch": 0.8768329317985782,
      "grad_norm": 40.68601608276367,
      "learning_rate": 4.092622319372069e-06,
      "loss": 0.0481176882982254,
      "loss_step_1": 0.0468,
      "loss_step_2": 0.0465,
      "loss_step_3": 0.0488,
      "step": 29480
    },
    {
      "entropy_step_1": 0.1591,
      "entropy_step_2": 0.1596,
      "entropy_step_3": 0.1669,
      "epoch": 0.8771303649504774,
      "grad_norm": 35.49180603027344,
      "learning_rate": 4.07315706316107e-06,
      "loss": 0.048048976063728335,
      "loss_step_1": 0.047,
      "loss_step_2": 0.0467,
      "loss_step_3": 0.0444,
      "step": 29490
    },
    {
      "entropy_step_1": 0.1957,
      "entropy_step_2": 0.1956,
      "entropy_step_3": 0.1974,
      "epoch": 0.8774277981023765,
      "grad_norm": 99.44112396240234,
      "learning_rate": 4.053736241795552e-06,
      "loss": 0.05229554176330566,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0571,
      "step": 29500
    },
    {
      "entropy_step_1": 0.1715,
      "entropy_step_2": 0.1721,
      "entropy_step_3": 0.1744,
      "epoch": 0.8777252312542756,
      "grad_norm": 74.9602279663086,
      "learning_rate": 4.034359874065402e-06,
      "loss": 0.05135861039161682,
      "loss_step_1": 0.0499,
      "loss_step_2": 0.0497,
      "loss_step_3": 0.0552,
      "step": 29510
    },
    {
      "entropy_step_1": 0.1607,
      "entropy_step_2": 0.1617,
      "entropy_step_3": 0.1554,
      "epoch": 0.8780226644061747,
      "grad_norm": 78.5976791381836,
      "learning_rate": 4.0150279787175285e-06,
      "loss": 0.04991500675678253,
      "loss_step_1": 0.0489,
      "loss_step_2": 0.0481,
      "loss_step_3": 0.0428,
      "step": 29520
    },
    {
      "entropy_step_1": 0.1719,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1946,
      "epoch": 0.8783200975580738,
      "grad_norm": 83.67645263671875,
      "learning_rate": 3.9957405744557975e-06,
      "loss": 0.05055598020553589,
      "loss_step_1": 0.0521,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0645,
      "step": 29530
    },
    {
      "entropy_step_1": 0.1677,
      "entropy_step_2": 0.1683,
      "entropy_step_3": 0.1597,
      "epoch": 0.878617530709973,
      "grad_norm": 167.5458984375,
      "learning_rate": 3.976497679941044e-06,
      "loss": 0.04938175678253174,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0493,
      "step": 29540
    },
    {
      "entropy_step_1": 0.1739,
      "entropy_step_2": 0.1745,
      "entropy_step_3": 0.1599,
      "epoch": 0.8789149638618721,
      "grad_norm": 45.78474807739258,
      "learning_rate": 3.957299313791024e-06,
      "loss": 0.0532406747341156,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0439,
      "step": 29550
    },
    {
      "entropy_step_1": 0.1676,
      "entropy_step_2": 0.1681,
      "entropy_step_3": 0.1675,
      "epoch": 0.8792123970137712,
      "grad_norm": 78.68744659423828,
      "learning_rate": 3.938145494580426e-06,
      "loss": 0.051917660236358645,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0506,
      "step": 29560
    },
    {
      "entropy_step_1": 0.1808,
      "entropy_step_2": 0.1802,
      "entropy_step_3": 0.1859,
      "epoch": 0.8795098301656703,
      "grad_norm": 89.27519226074219,
      "learning_rate": 3.919036240840807e-06,
      "loss": 0.05229764580726624,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0581,
      "step": 29570
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1732,
      "entropy_step_3": 0.1846,
      "epoch": 0.8798072633175694,
      "grad_norm": 72.31246185302734,
      "learning_rate": 3.899971571060662e-06,
      "loss": 0.05603710412979126,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0522,
      "loss_step_3": 0.0538,
      "step": 29580
    },
    {
      "entropy_step_1": 0.1685,
      "entropy_step_2": 0.1681,
      "entropy_step_3": 0.1642,
      "epoch": 0.8801046964694685,
      "grad_norm": 53.79189682006836,
      "learning_rate": 3.880951503685304e-06,
      "loss": 0.05338633060455322,
      "loss_step_1": 0.0501,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.0442,
      "step": 29590
    },
    {
      "entropy_step_1": 0.2057,
      "entropy_step_2": 0.2065,
      "entropy_step_3": 0.2064,
      "epoch": 0.8804021296213675,
      "grad_norm": 61.93375778198242,
      "learning_rate": 3.861976057116906e-06,
      "loss": 0.0645368218421936,
      "loss_step_1": 0.0669,
      "loss_step_2": 0.0667,
      "loss_step_3": 0.0601,
      "step": 29600
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.1916,
      "entropy_step_3": 0.2051,
      "epoch": 0.8806995627732667,
      "grad_norm": 79.4150390625,
      "learning_rate": 3.843045249714467e-06,
      "loss": 0.0513114333152771,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0527,
      "step": 29610
    },
    {
      "entropy_step_1": 0.1754,
      "entropy_step_2": 0.1763,
      "entropy_step_3": 0.1786,
      "epoch": 0.8809969959251658,
      "grad_norm": 66.29681396484375,
      "learning_rate": 3.8241590997938e-06,
      "loss": 0.04935804903507233,
      "loss_step_1": 0.0502,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.047,
      "step": 29620
    },
    {
      "entropy_step_1": 0.1687,
      "entropy_step_2": 0.1694,
      "entropy_step_3": 0.1974,
      "epoch": 0.8812944290770649,
      "grad_norm": 152.65089416503906,
      "learning_rate": 3.805317625627525e-06,
      "loss": 0.05216379165649414,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0563,
      "step": 29630
    },
    {
      "entropy_step_1": 0.178,
      "entropy_step_2": 0.1804,
      "entropy_step_3": 0.1889,
      "epoch": 0.881591862228964,
      "grad_norm": 106.0730209350586,
      "learning_rate": 3.7865208454450186e-06,
      "loss": 0.052115947008132935,
      "loss_step_1": 0.0483,
      "loss_step_2": 0.0481,
      "loss_step_3": 0.0477,
      "step": 29640
    },
    {
      "entropy_step_1": 0.1759,
      "entropy_step_2": 0.1774,
      "entropy_step_3": 0.1787,
      "epoch": 0.8818892953808631,
      "grad_norm": 111.82254028320312,
      "learning_rate": 3.767768777432418e-06,
      "loss": 0.053159463405609134,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0543,
      "step": 29650
    },
    {
      "entropy_step_1": 0.1798,
      "entropy_step_2": 0.1821,
      "entropy_step_3": 0.1744,
      "epoch": 0.8821867285327623,
      "grad_norm": 68.80625915527344,
      "learning_rate": 3.7490614397325986e-06,
      "loss": 0.05305522680282593,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0474,
      "step": 29660
    },
    {
      "entropy_step_1": 0.1701,
      "entropy_step_2": 0.1715,
      "entropy_step_3": 0.1697,
      "epoch": 0.8824841616846614,
      "grad_norm": 139.09263610839844,
      "learning_rate": 3.7303988504451825e-06,
      "loss": 0.06141452789306641,
      "loss_step_1": 0.0613,
      "loss_step_2": 0.0614,
      "loss_step_3": 0.0638,
      "step": 29670
    },
    {
      "entropy_step_1": 0.1885,
      "entropy_step_2": 0.1901,
      "entropy_step_3": 0.1911,
      "epoch": 0.8827815948365605,
      "grad_norm": 126.0421371459961,
      "learning_rate": 3.7117810276264542e-06,
      "loss": 0.06346554160118104,
      "loss_step_1": 0.061,
      "loss_step_2": 0.0603,
      "loss_step_3": 0.0539,
      "step": 29680
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1735,
      "entropy_step_3": 0.159,
      "epoch": 0.8830790279884596,
      "grad_norm": 66.74425506591797,
      "learning_rate": 3.693207989289443e-06,
      "loss": 0.05707541704177856,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0517,
      "step": 29690
    },
    {
      "entropy_step_1": 0.1771,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.1628,
      "epoch": 0.8833764611403587,
      "grad_norm": 103.9315414428711,
      "learning_rate": 3.6746797534037813e-06,
      "loss": 0.04800902009010315,
      "loss_step_1": 0.0472,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0441,
      "step": 29700
    },
    {
      "entropy_step_1": 0.1934,
      "entropy_step_2": 0.1941,
      "entropy_step_3": 0.1877,
      "epoch": 0.8836738942922578,
      "grad_norm": 87.68122100830078,
      "learning_rate": 3.656196337895812e-06,
      "loss": 0.06230604648590088,
      "loss_step_1": 0.0638,
      "loss_step_2": 0.0637,
      "loss_step_3": 0.0651,
      "step": 29710
    },
    {
      "entropy_step_1": 0.2056,
      "entropy_step_2": 0.2058,
      "entropy_step_3": 0.2086,
      "epoch": 0.883971327444157,
      "grad_norm": 122.61603546142578,
      "learning_rate": 3.6377577606484747e-06,
      "loss": 0.05864364504814148,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0527,
      "loss_step_3": 0.0542,
      "step": 29720
    },
    {
      "entropy_step_1": 0.1697,
      "entropy_step_2": 0.1703,
      "entropy_step_3": 0.1889,
      "epoch": 0.8842687605960561,
      "grad_norm": 130.04798889160156,
      "learning_rate": 3.619364039501366e-06,
      "loss": 0.05529029965400696,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0674,
      "step": 29730
    },
    {
      "entropy_step_1": 0.1695,
      "entropy_step_2": 0.1692,
      "entropy_step_3": 0.1853,
      "epoch": 0.8845661937479552,
      "grad_norm": 50.70771408081055,
      "learning_rate": 3.6010151922506506e-06,
      "loss": 0.058148699998855594,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0648,
      "step": 29740
    },
    {
      "entropy_step_1": 0.1826,
      "entropy_step_2": 0.1822,
      "entropy_step_3": 0.1984,
      "epoch": 0.8848636268998542,
      "grad_norm": 124.90799713134766,
      "learning_rate": 3.5827112366490777e-06,
      "loss": 0.051013171672821045,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0646,
      "step": 29750
    },
    {
      "entropy_step_1": 0.1847,
      "entropy_step_2": 0.1855,
      "entropy_step_3": 0.1766,
      "epoch": 0.8851610600517533,
      "grad_norm": 53.40135955810547,
      "learning_rate": 3.564452190405981e-06,
      "loss": 0.05320115685462952,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0555,
      "step": 29760
    },
    {
      "entropy_step_1": 0.1801,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1943,
      "epoch": 0.8854584932036524,
      "grad_norm": 167.0932159423828,
      "learning_rate": 3.5462380711872465e-06,
      "loss": 0.05163782238960266,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0522,
      "loss_step_3": 0.0567,
      "step": 29770
    },
    {
      "entropy_step_1": 0.1588,
      "entropy_step_2": 0.1609,
      "entropy_step_3": 0.145,
      "epoch": 0.8857559263555516,
      "grad_norm": 57.75163650512695,
      "learning_rate": 3.5280688966152665e-06,
      "loss": 0.045277222990989685,
      "loss_step_1": 0.045,
      "loss_step_2": 0.0446,
      "loss_step_3": 0.0382,
      "step": 29780
    },
    {
      "entropy_step_1": 0.1638,
      "entropy_step_2": 0.165,
      "entropy_step_3": 0.1904,
      "epoch": 0.8860533595074507,
      "grad_norm": 155.95382690429688,
      "learning_rate": 3.5099446842689786e-06,
      "loss": 0.05111013054847717,
      "loss_step_1": 0.0456,
      "loss_step_2": 0.0454,
      "loss_step_3": 0.055,
      "step": 29790
    },
    {
      "entropy_step_1": 0.185,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1844,
      "epoch": 0.8863507926593498,
      "grad_norm": 134.78677368164062,
      "learning_rate": 3.49186545168379e-06,
      "loss": 0.053440093994140625,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0526,
      "loss_step_3": 0.06,
      "step": 29800
    },
    {
      "entropy_step_1": 0.1683,
      "entropy_step_2": 0.1699,
      "entropy_step_3": 0.1781,
      "epoch": 0.8866482258112489,
      "grad_norm": 77.98626708984375,
      "learning_rate": 3.4738312163516083e-06,
      "loss": 0.05191800594329834,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0542,
      "step": 29810
    },
    {
      "entropy_step_1": 0.1698,
      "entropy_step_2": 0.1707,
      "entropy_step_3": 0.1668,
      "epoch": 0.886945658963148,
      "grad_norm": 45.037811279296875,
      "learning_rate": 3.4558419957208098e-06,
      "loss": 0.055291324853897095,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0484,
      "step": 29820
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.183,
      "entropy_step_3": 0.19,
      "epoch": 0.8872430921150471,
      "grad_norm": 124.22222900390625,
      "learning_rate": 3.437897807196211e-06,
      "loss": 0.05356694459915161,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.053,
      "step": 29830
    },
    {
      "entropy_step_1": 0.1788,
      "entropy_step_2": 0.1799,
      "entropy_step_3": 0.1701,
      "epoch": 0.8875405252669463,
      "grad_norm": 68.21562194824219,
      "learning_rate": 3.4199986681390516e-06,
      "loss": 0.05687881708145141,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0514,
      "step": 29840
    },
    {
      "entropy_step_1": 0.164,
      "entropy_step_2": 0.1641,
      "entropy_step_3": 0.1542,
      "epoch": 0.8878379584188454,
      "grad_norm": 64.16886901855469,
      "learning_rate": 3.4021445958669794e-06,
      "loss": 0.048159024119377135,
      "loss_step_1": 0.0429,
      "loss_step_2": 0.0427,
      "loss_step_3": 0.0424,
      "step": 29850
    },
    {
      "entropy_step_1": 0.1968,
      "entropy_step_2": 0.1981,
      "entropy_step_3": 0.1745,
      "epoch": 0.8881353915707445,
      "grad_norm": 34.01306915283203,
      "learning_rate": 3.3843356076540823e-06,
      "loss": 0.06177148222923279,
      "loss_step_1": 0.0693,
      "loss_step_2": 0.0692,
      "loss_step_3": 0.0537,
      "step": 29860
    },
    {
      "entropy_step_1": 0.1798,
      "entropy_step_2": 0.1805,
      "entropy_step_3": 0.1821,
      "epoch": 0.8884328247226436,
      "grad_norm": 278.8060302734375,
      "learning_rate": 3.366571720730777e-06,
      "loss": 0.05826011300086975,
      "loss_step_1": 0.0514,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0482,
      "step": 29870
    },
    {
      "entropy_step_1": 0.1896,
      "entropy_step_2": 0.189,
      "entropy_step_3": 0.1824,
      "epoch": 0.8887302578745427,
      "grad_norm": 43.35847854614258,
      "learning_rate": 3.3488529522838706e-06,
      "loss": 0.054384684562683104,
      "loss_step_1": 0.0557,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0596,
      "step": 29880
    },
    {
      "entropy_step_1": 0.17,
      "entropy_step_2": 0.17,
      "entropy_step_3": 0.1696,
      "epoch": 0.8890276910264419,
      "grad_norm": 162.24720764160156,
      "learning_rate": 3.3311793194565122e-06,
      "loss": 0.057324498891830444,
      "loss_step_1": 0.0602,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0548,
      "step": 29890
    },
    {
      "entropy_step_1": 0.1678,
      "entropy_step_2": 0.1661,
      "entropy_step_3": 0.1696,
      "epoch": 0.8893251241783409,
      "grad_norm": 41.623085021972656,
      "learning_rate": 3.313550839348184e-06,
      "loss": 0.04868876934051514,
      "loss_step_1": 0.048,
      "loss_step_2": 0.048,
      "loss_step_3": 0.0533,
      "step": 29900
    },
    {
      "entropy_step_1": 0.1784,
      "entropy_step_2": 0.1781,
      "entropy_step_3": 0.1665,
      "epoch": 0.88962255733024,
      "grad_norm": 34.44908905029297,
      "learning_rate": 3.2959675290146763e-06,
      "loss": 0.0477314293384552,
      "loss_step_1": 0.0459,
      "loss_step_2": 0.0458,
      "loss_step_3": 0.0433,
      "step": 29910
    },
    {
      "entropy_step_1": 0.1845,
      "entropy_step_2": 0.1846,
      "entropy_step_3": 0.1894,
      "epoch": 0.8899199904821391,
      "grad_norm": 98.04450988769531,
      "learning_rate": 3.278429405468103e-06,
      "loss": 0.05328910350799561,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0509,
      "loss_step_3": 0.051,
      "step": 29920
    },
    {
      "entropy_step_1": 0.1904,
      "entropy_step_2": 0.192,
      "entropy_step_3": 0.2048,
      "epoch": 0.8902174236340382,
      "grad_norm": 77.34526062011719,
      "learning_rate": 3.2609364856767964e-06,
      "loss": 0.0547396183013916,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0575,
      "loss_step_3": 0.0608,
      "step": 29930
    },
    {
      "entropy_step_1": 0.164,
      "entropy_step_2": 0.1641,
      "entropy_step_3": 0.1713,
      "epoch": 0.8905148567859373,
      "grad_norm": 190.0548095703125,
      "learning_rate": 3.2434887865654294e-06,
      "loss": 0.053805035352706906,
      "loss_step_1": 0.0491,
      "loss_step_2": 0.0492,
      "loss_step_3": 0.0418,
      "step": 29940
    },
    {
      "entropy_step_1": 0.1976,
      "entropy_step_2": 0.1972,
      "entropy_step_3": 0.2145,
      "epoch": 0.8908122899378365,
      "grad_norm": 823.0490112304688,
      "learning_rate": 3.2260863250148766e-06,
      "loss": 0.05352504849433899,
      "loss_step_1": 0.056,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0625,
      "step": 29950
    },
    {
      "entropy_step_1": 0.1579,
      "entropy_step_2": 0.1576,
      "entropy_step_3": 0.1561,
      "epoch": 0.8911097230897356,
      "grad_norm": 156.8701934814453,
      "learning_rate": 3.2087291178622524e-06,
      "loss": 0.04776493608951569,
      "loss_step_1": 0.0454,
      "loss_step_2": 0.045,
      "loss_step_3": 0.0489,
      "step": 29960
    },
    {
      "entropy_step_1": 0.171,
      "entropy_step_2": 0.1708,
      "entropy_step_3": 0.1634,
      "epoch": 0.8914071562416347,
      "grad_norm": 51.22069549560547,
      "learning_rate": 3.191417181900885e-06,
      "loss": 0.049851030111312866,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.0483,
      "loss_step_3": 0.0526,
      "step": 29970
    },
    {
      "entropy_step_1": 0.1652,
      "entropy_step_2": 0.1658,
      "entropy_step_3": 0.1735,
      "epoch": 0.8917045893935338,
      "grad_norm": 87.46971130371094,
      "learning_rate": 3.1741505338803324e-06,
      "loss": 0.05383874177932739,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0561,
      "step": 29980
    },
    {
      "entropy_step_1": 0.1775,
      "entropy_step_2": 0.1791,
      "entropy_step_3": 0.1466,
      "epoch": 0.8920020225454329,
      "grad_norm": 160.4195556640625,
      "learning_rate": 3.1569291905062915e-06,
      "loss": 0.05347929000854492,
      "loss_step_1": 0.0561,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0392,
      "step": 29990
    },
    {
      "entropy_step_1": 0.1626,
      "entropy_step_2": 0.1632,
      "entropy_step_3": 0.1634,
      "epoch": 0.892299455697332,
      "grad_norm": 95.40682983398438,
      "learning_rate": 3.13975316844064e-06,
      "loss": 0.04756808280944824,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0454,
      "step": 30000
    },
    {
      "epoch": 0.892299455697332,
      "eval_accuracy": 0.927,
      "eval_f1": 0.9269385553250283,
      "eval_loss": 0.0817301794886589,
      "eval_precision": 0.9284412764539912,
      "eval_recall": 0.927,
      "eval_runtime": 6.7011,
      "eval_samples_per_second": 74.615,
      "eval_steps_per_second": 9.401,
      "step": 30000
    },
    {
      "entropy_step_1": 0.1513,
      "entropy_step_2": 0.1522,
      "entropy_step_3": 0.1608,
      "epoch": 0.8925968888492312,
      "grad_norm": 49.18912124633789,
      "learning_rate": 3.122622484301446e-06,
      "loss": 0.05130372643470764,
      "loss_step_1": 0.0468,
      "loss_step_2": 0.046,
      "loss_step_3": 0.0518,
      "step": 30010
    },
    {
      "entropy_step_1": 0.1902,
      "entropy_step_2": 0.1904,
      "entropy_step_3": 0.2135,
      "epoch": 0.8928943220011303,
      "grad_norm": 59.909339904785156,
      "learning_rate": 3.1055371546628355e-06,
      "loss": 0.05321921706199646,
      "loss_step_1": 0.0559,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0707,
      "step": 30020
    },
    {
      "entropy_step_1": 0.1765,
      "entropy_step_2": 0.178,
      "entropy_step_3": 0.1505,
      "epoch": 0.8931917551530294,
      "grad_norm": 160.75,
      "learning_rate": 3.0884971960551344e-06,
      "loss": 0.05512760877609253,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0439,
      "step": 30030
    },
    {
      "entropy_step_1": 0.1618,
      "entropy_step_2": 0.1631,
      "entropy_step_3": 0.1869,
      "epoch": 0.8934891883049285,
      "grad_norm": 52.168052673339844,
      "learning_rate": 3.071502624964706e-06,
      "loss": 0.04942240118980408,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0631,
      "step": 30040
    },
    {
      "entropy_step_1": 0.1602,
      "entropy_step_2": 0.1608,
      "entropy_step_3": 0.1559,
      "epoch": 0.8937866214568275,
      "grad_norm": 100.23635864257812,
      "learning_rate": 3.054553457834053e-06,
      "loss": 0.05064432024955749,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0483,
      "step": 30050
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1738,
      "entropy_step_3": 0.1734,
      "epoch": 0.8940840546087266,
      "grad_norm": 29.44612693786621,
      "learning_rate": 3.037649711061713e-06,
      "loss": 0.05075105428695679,
      "loss_step_1": 0.053,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.0548,
      "step": 30060
    },
    {
      "entropy_step_1": 0.1819,
      "entropy_step_2": 0.1825,
      "entropy_step_3": 0.1818,
      "epoch": 0.8943814877606258,
      "grad_norm": 87.33452606201172,
      "learning_rate": 3.020791401002293e-06,
      "loss": 0.05256506204605103,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0504,
      "loss_step_3": 0.0529,
      "step": 30070
    },
    {
      "entropy_step_1": 0.1687,
      "entropy_step_2": 0.1702,
      "entropy_step_3": 0.1668,
      "epoch": 0.8946789209125249,
      "grad_norm": 78.53443908691406,
      "learning_rate": 3.003978543966429e-06,
      "loss": 0.04626070261001587,
      "loss_step_1": 0.0472,
      "loss_step_2": 0.047,
      "loss_step_3": 0.0502,
      "step": 30080
    },
    {
      "entropy_step_1": 0.1696,
      "entropy_step_2": 0.1703,
      "entropy_step_3": 0.1623,
      "epoch": 0.894976354064424,
      "grad_norm": 145.2967529296875,
      "learning_rate": 2.9872111562208094e-06,
      "loss": 0.05648094415664673,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0561,
      "loss_step_3": 0.0607,
      "step": 30090
    },
    {
      "entropy_step_1": 0.1776,
      "entropy_step_2": 0.1797,
      "entropy_step_3": 0.1716,
      "epoch": 0.8952737872163231,
      "grad_norm": 60.97075271606445,
      "learning_rate": 2.9704892539880958e-06,
      "loss": 0.054554450511932376,
      "loss_step_1": 0.0577,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0457,
      "step": 30100
    },
    {
      "entropy_step_1": 0.1597,
      "entropy_step_2": 0.1603,
      "entropy_step_3": 0.1682,
      "epoch": 0.8955712203682222,
      "grad_norm": 94.43553161621094,
      "learning_rate": 2.953812853446952e-06,
      "loss": 0.05008809566497803,
      "loss_step_1": 0.0468,
      "loss_step_2": 0.0466,
      "loss_step_3": 0.0485,
      "step": 30110
    },
    {
      "entropy_step_1": 0.1512,
      "entropy_step_2": 0.1513,
      "entropy_step_3": 0.1442,
      "epoch": 0.8958686535201213,
      "grad_norm": 29.928607940673828,
      "learning_rate": 2.937181970732028e-06,
      "loss": 0.04527064263820648,
      "loss_step_1": 0.0444,
      "loss_step_2": 0.0441,
      "loss_step_3": 0.0377,
      "step": 30120
    },
    {
      "entropy_step_1": 0.1721,
      "entropy_step_2": 0.173,
      "entropy_step_3": 0.1607,
      "epoch": 0.8961660866720205,
      "grad_norm": 87.33631896972656,
      "learning_rate": 2.9205966219339355e-06,
      "loss": 0.04563286900520325,
      "loss_step_1": 0.0424,
      "loss_step_2": 0.0421,
      "loss_step_3": 0.0343,
      "step": 30130
    },
    {
      "entropy_step_1": 0.1752,
      "entropy_step_2": 0.176,
      "entropy_step_3": 0.1731,
      "epoch": 0.8964635198239196,
      "grad_norm": 94.10301208496094,
      "learning_rate": 2.9040568230992215e-06,
      "loss": 0.05163455009460449,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0488,
      "step": 30140
    },
    {
      "entropy_step_1": 0.1686,
      "entropy_step_2": 0.1704,
      "entropy_step_3": 0.1757,
      "epoch": 0.8967609529758187,
      "grad_norm": 97.24726104736328,
      "learning_rate": 2.8875625902303584e-06,
      "loss": 0.0548955500125885,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0526,
      "step": 30150
    },
    {
      "entropy_step_1": 0.1785,
      "entropy_step_2": 0.1778,
      "entropy_step_3": 0.1826,
      "epoch": 0.8970583861277178,
      "grad_norm": 41.16071701049805,
      "learning_rate": 2.8711139392857523e-06,
      "loss": 0.05522969961166382,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0633,
      "step": 30160
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1886,
      "entropy_step_3": 0.1761,
      "epoch": 0.8973558192796169,
      "grad_norm": 179.08152770996094,
      "learning_rate": 2.854710886179679e-06,
      "loss": 0.057697457075119016,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0455,
      "step": 30170
    },
    {
      "entropy_step_1": 0.1726,
      "entropy_step_2": 0.1729,
      "entropy_step_3": 0.1564,
      "epoch": 0.8976532524315161,
      "grad_norm": 873.7451782226562,
      "learning_rate": 2.838353446782338e-06,
      "loss": 0.052847301959991454,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0402,
      "step": 30180
    },
    {
      "entropy_step_1": 0.194,
      "entropy_step_2": 0.1937,
      "entropy_step_3": 0.2035,
      "epoch": 0.8979506855834152,
      "grad_norm": 170.68409729003906,
      "learning_rate": 2.8220416369197644e-06,
      "loss": 0.05654796361923218,
      "loss_step_1": 0.0567,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0549,
      "step": 30190
    },
    {
      "entropy_step_1": 0.1643,
      "entropy_step_2": 0.1641,
      "entropy_step_3": 0.1492,
      "epoch": 0.8982481187353142,
      "grad_norm": 35.591400146484375,
      "learning_rate": 2.805775472373856e-06,
      "loss": 0.051840537786483766,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0475,
      "loss_step_3": 0.0382,
      "step": 30200
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.1839,
      "entropy_step_3": 0.171,
      "epoch": 0.8985455518872133,
      "grad_norm": 213.47537231445312,
      "learning_rate": 2.7895549688823407e-06,
      "loss": 0.062960684299469,
      "loss_step_1": 0.0645,
      "loss_step_2": 0.0635,
      "loss_step_3": 0.0626,
      "step": 30210
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1729,
      "entropy_step_3": 0.2108,
      "epoch": 0.8988429850391124,
      "grad_norm": 113.97025299072266,
      "learning_rate": 2.773380142138793e-06,
      "loss": 0.053213053941726686,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0698,
      "step": 30220
    },
    {
      "entropy_step_1": 0.1908,
      "entropy_step_2": 0.1902,
      "entropy_step_3": 0.2293,
      "epoch": 0.8991404181910115,
      "grad_norm": 97.82579803466797,
      "learning_rate": 2.7572510077925608e-06,
      "loss": 0.05471912622451782,
      "loss_step_1": 0.0543,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0545,
      "step": 30230
    },
    {
      "entropy_step_1": 0.17,
      "entropy_step_2": 0.1694,
      "entropy_step_3": 0.1698,
      "epoch": 0.8994378513429107,
      "grad_norm": 92.31330108642578,
      "learning_rate": 2.741167581448828e-06,
      "loss": 0.048064425587654114,
      "loss_step_1": 0.0464,
      "loss_step_2": 0.0464,
      "loss_step_3": 0.045,
      "step": 30240
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.1821,
      "entropy_step_3": 0.1614,
      "epoch": 0.8997352844948098,
      "grad_norm": 100.8052749633789,
      "learning_rate": 2.7251298786684866e-06,
      "loss": 0.04817017614841461,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.0376,
      "step": 30250
    },
    {
      "entropy_step_1": 0.1811,
      "entropy_step_2": 0.1822,
      "entropy_step_3": 0.167,
      "epoch": 0.9000327176467089,
      "grad_norm": 163.78907775878906,
      "learning_rate": 2.7091379149682685e-06,
      "loss": 0.052247810363769534,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0554,
      "loss_step_3": 0.0509,
      "step": 30260
    },
    {
      "entropy_step_1": 0.1574,
      "entropy_step_2": 0.158,
      "entropy_step_3": 0.1449,
      "epoch": 0.900330150798608,
      "grad_norm": 63.65280532836914,
      "learning_rate": 2.693191705820597e-06,
      "loss": 0.04783724844455719,
      "loss_step_1": 0.0452,
      "loss_step_2": 0.0446,
      "loss_step_3": 0.0437,
      "step": 30270
    },
    {
      "entropy_step_1": 0.1816,
      "entropy_step_2": 0.1816,
      "entropy_step_3": 0.1682,
      "epoch": 0.9006275839505071,
      "grad_norm": 103.95982360839844,
      "learning_rate": 2.6772912666536533e-06,
      "loss": 0.05596853494644165,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0473,
      "step": 30280
    },
    {
      "entropy_step_1": 0.1687,
      "entropy_step_2": 0.1686,
      "entropy_step_3": 0.1526,
      "epoch": 0.9009250171024062,
      "grad_norm": 48.50819396972656,
      "learning_rate": 2.661436612851331e-06,
      "loss": 0.04335886240005493,
      "loss_step_1": 0.0424,
      "loss_step_2": 0.0417,
      "loss_step_3": 0.0406,
      "step": 30290
    },
    {
      "entropy_step_1": 0.169,
      "entropy_step_2": 0.1701,
      "entropy_step_3": 0.1607,
      "epoch": 0.9012224502543054,
      "grad_norm": 133.37274169921875,
      "learning_rate": 2.645627759753222e-06,
      "loss": 0.04932448863983154,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0479,
      "loss_step_3": 0.0442,
      "step": 30300
    },
    {
      "entropy_step_1": 0.1709,
      "entropy_step_2": 0.1718,
      "entropy_step_3": 0.1821,
      "epoch": 0.9015198834062045,
      "grad_norm": 332.2200927734375,
      "learning_rate": 2.629864722654601e-06,
      "loss": 0.05049888491630554,
      "loss_step_1": 0.0537,
      "loss_step_2": 0.0535,
      "loss_step_3": 0.06,
      "step": 30310
    },
    {
      "entropy_step_1": 0.1572,
      "entropy_step_2": 0.1579,
      "entropy_step_3": 0.1509,
      "epoch": 0.9018173165581036,
      "grad_norm": 263.7810363769531,
      "learning_rate": 2.61414751680642e-06,
      "loss": 0.051727902889251706,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.0413,
      "step": 30320
    },
    {
      "entropy_step_1": 0.1838,
      "entropy_step_2": 0.1838,
      "entropy_step_3": 0.1837,
      "epoch": 0.9021147497100027,
      "grad_norm": 120.26066589355469,
      "learning_rate": 2.598476157415303e-06,
      "loss": 0.05231645703315735,
      "loss_step_1": 0.0496,
      "loss_step_2": 0.0494,
      "loss_step_3": 0.0585,
      "step": 30330
    },
    {
      "entropy_step_1": 0.1848,
      "entropy_step_2": 0.1842,
      "entropy_step_3": 0.204,
      "epoch": 0.9024121828619018,
      "grad_norm": 155.13699340820312,
      "learning_rate": 2.582850659643493e-06,
      "loss": 0.05972120761871338,
      "loss_step_1": 0.0619,
      "loss_step_2": 0.0621,
      "loss_step_3": 0.0719,
      "step": 30340
    },
    {
      "entropy_step_1": 0.1721,
      "entropy_step_2": 0.1718,
      "entropy_step_3": 0.147,
      "epoch": 0.902709616013801,
      "grad_norm": 156.8197021484375,
      "learning_rate": 2.5672710386088762e-06,
      "loss": 0.04998199343681335,
      "loss_step_1": 0.047,
      "loss_step_2": 0.0465,
      "loss_step_3": 0.0416,
      "step": 30350
    },
    {
      "entropy_step_1": 0.19,
      "entropy_step_2": 0.1895,
      "entropy_step_3": 0.1619,
      "epoch": 0.9030070491657,
      "grad_norm": 249.9617919921875,
      "learning_rate": 2.5517373093849386e-06,
      "loss": 0.056445884704589847,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0647,
      "step": 30360
    },
    {
      "entropy_step_1": 0.1713,
      "entropy_step_2": 0.1693,
      "entropy_step_3": 0.1793,
      "epoch": 0.9033044823175991,
      "grad_norm": 74.69705200195312,
      "learning_rate": 2.5362494870007856e-06,
      "loss": 0.05078247785568237,
      "loss_step_1": 0.0493,
      "loss_step_2": 0.0487,
      "loss_step_3": 0.049,
      "step": 30370
    },
    {
      "entropy_step_1": 0.1922,
      "entropy_step_2": 0.1912,
      "entropy_step_3": 0.2044,
      "epoch": 0.9036019154694982,
      "grad_norm": 127.87740325927734,
      "learning_rate": 2.520807586441087e-06,
      "loss": 0.05709457993507385,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0598,
      "loss_step_3": 0.0638,
      "step": 30380
    },
    {
      "entropy_step_1": 0.2059,
      "entropy_step_2": 0.2053,
      "entropy_step_3": 0.2,
      "epoch": 0.9038993486213973,
      "grad_norm": 60.05362319946289,
      "learning_rate": 2.5054116226461034e-06,
      "loss": 0.06371331810951233,
      "loss_step_1": 0.068,
      "loss_step_2": 0.0669,
      "loss_step_3": 0.0856,
      "step": 30390
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1848,
      "entropy_step_3": 0.1801,
      "epoch": 0.9041967817732964,
      "grad_norm": 293.8211364746094,
      "learning_rate": 2.4900616105116235e-06,
      "loss": 0.05846682786941528,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0556,
      "loss_step_3": 0.0483,
      "step": 30400
    },
    {
      "entropy_step_1": 0.1988,
      "entropy_step_2": 0.1983,
      "entropy_step_3": 0.2091,
      "epoch": 0.9044942149251955,
      "grad_norm": 73.35356140136719,
      "learning_rate": 2.4747575648890075e-06,
      "loss": 0.058540767431259154,
      "loss_step_1": 0.0532,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0621,
      "step": 30410
    },
    {
      "entropy_step_1": 0.1875,
      "entropy_step_2": 0.1876,
      "entropy_step_3": 0.1962,
      "epoch": 0.9047916480770947,
      "grad_norm": 57.823551177978516,
      "learning_rate": 2.459499500585127e-06,
      "loss": 0.054387104511260984,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0626,
      "step": 30420
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.188,
      "entropy_step_3": 0.2155,
      "epoch": 0.9050890812289938,
      "grad_norm": 108.52796936035156,
      "learning_rate": 2.4442874323623575e-06,
      "loss": 0.05319525003433227,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0587,
      "loss_step_3": 0.0715,
      "step": 30430
    },
    {
      "entropy_step_1": 0.1728,
      "entropy_step_2": 0.1722,
      "entropy_step_3": 0.1681,
      "epoch": 0.9053865143808929,
      "grad_norm": 214.35867309570312,
      "learning_rate": 2.429121374938581e-06,
      "loss": 0.0557421088218689,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0586,
      "loss_step_3": 0.0544,
      "step": 30440
    },
    {
      "entropy_step_1": 0.169,
      "entropy_step_2": 0.1687,
      "entropy_step_3": 0.1549,
      "epoch": 0.905683947532792,
      "grad_norm": 326.89495849609375,
      "learning_rate": 2.4140013429871713e-06,
      "loss": 0.05103065371513367,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0525,
      "loss_step_3": 0.0455,
      "step": 30450
    },
    {
      "entropy_step_1": 0.1779,
      "entropy_step_2": 0.1782,
      "entropy_step_3": 0.1814,
      "epoch": 0.9059813806846911,
      "grad_norm": 121.2320785522461,
      "learning_rate": 2.3989273511369646e-06,
      "loss": 0.05255172848701477,
      "loss_step_1": 0.0514,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0492,
      "step": 30460
    },
    {
      "entropy_step_1": 0.1789,
      "entropy_step_2": 0.1781,
      "entropy_step_3": 0.1756,
      "epoch": 0.9062788138365903,
      "grad_norm": 82.54296875,
      "learning_rate": 2.3838994139722626e-06,
      "loss": 0.0599978506565094,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0638,
      "loss_step_3": 0.0609,
      "step": 30470
    },
    {
      "entropy_step_1": 0.2039,
      "entropy_step_2": 0.2042,
      "entropy_step_3": 0.1908,
      "epoch": 0.9065762469884894,
      "grad_norm": 64.42805480957031,
      "learning_rate": 2.3689175460327728e-06,
      "loss": 0.05272164344787598,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0605,
      "loss_step_3": 0.0603,
      "step": 30480
    },
    {
      "entropy_step_1": 0.174,
      "entropy_step_2": 0.1743,
      "entropy_step_3": 0.15,
      "epoch": 0.9068736801403885,
      "grad_norm": 68.26420593261719,
      "learning_rate": 2.3539817618136785e-06,
      "loss": 0.04722130298614502,
      "loss_step_1": 0.0484,
      "loss_step_2": 0.0489,
      "loss_step_3": 0.0377,
      "step": 30490
    },
    {
      "entropy_step_1": 0.153,
      "entropy_step_2": 0.1534,
      "entropy_step_3": 0.1666,
      "epoch": 0.9071711132922876,
      "grad_norm": 684.6937866210938,
      "learning_rate": 2.3390920757655534e-06,
      "loss": 0.050508898496627805,
      "loss_step_1": 0.0431,
      "loss_step_2": 0.0425,
      "loss_step_3": 0.0456,
      "step": 30500
    },
    {
      "entropy_step_1": 0.1928,
      "entropy_step_2": 0.1923,
      "entropy_step_3": 0.1948,
      "epoch": 0.9074685464441866,
      "grad_norm": 63.823326110839844,
      "learning_rate": 2.324248502294357e-06,
      "loss": 0.05778360962867737,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0495,
      "step": 30510
    },
    {
      "entropy_step_1": 0.1772,
      "entropy_step_2": 0.1761,
      "entropy_step_3": 0.1727,
      "epoch": 0.9077659795960857,
      "grad_norm": 267.81927490234375,
      "learning_rate": 2.3094510557614577e-06,
      "loss": 0.05154444575309754,
      "loss_step_1": 0.0494,
      "loss_step_2": 0.0494,
      "loss_step_3": 0.0487,
      "step": 30520
    },
    {
      "entropy_step_1": 0.191,
      "entropy_step_2": 0.1903,
      "entropy_step_3": 0.1997,
      "epoch": 0.9080634127479849,
      "grad_norm": 623.5189208984375,
      "learning_rate": 2.2946997504835923e-06,
      "loss": 0.0601675808429718,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0634,
      "loss_step_3": 0.0755,
      "step": 30530
    },
    {
      "entropy_step_1": 0.1702,
      "entropy_step_2": 0.1697,
      "entropy_step_3": 0.1849,
      "epoch": 0.908360845899884,
      "grad_norm": 46.82630157470703,
      "learning_rate": 2.2799946007328377e-06,
      "loss": 0.05026869177818298,
      "loss_step_1": 0.0455,
      "loss_step_2": 0.046,
      "loss_step_3": 0.0507,
      "step": 30540
    },
    {
      "entropy_step_1": 0.1654,
      "entropy_step_2": 0.1654,
      "entropy_step_3": 0.1735,
      "epoch": 0.9086582790517831,
      "grad_norm": 529.7313842773438,
      "learning_rate": 2.265335620736625e-06,
      "loss": 0.04405585527420044,
      "loss_step_1": 0.0424,
      "loss_step_2": 0.0423,
      "loss_step_3": 0.0446,
      "step": 30550
    },
    {
      "entropy_step_1": 0.1929,
      "entropy_step_2": 0.1932,
      "entropy_step_3": 0.1681,
      "epoch": 0.9089557122036822,
      "grad_norm": 334.9998474121094,
      "learning_rate": 2.2507228246777425e-06,
      "loss": 0.06179994344711304,
      "loss_step_1": 0.0637,
      "loss_step_2": 0.0632,
      "loss_step_3": 0.0493,
      "step": 30560
    },
    {
      "entropy_step_1": 0.1843,
      "entropy_step_2": 0.1837,
      "entropy_step_3": 0.1584,
      "epoch": 0.9092531453555813,
      "grad_norm": 44.32616424560547,
      "learning_rate": 2.236156226694236e-06,
      "loss": 0.04898936152458191,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0505,
      "step": 30570
    },
    {
      "entropy_step_1": 0.1788,
      "entropy_step_2": 0.1788,
      "entropy_step_3": 0.1977,
      "epoch": 0.9095505785074804,
      "grad_norm": 94.98783874511719,
      "learning_rate": 2.2216358408795044e-06,
      "loss": 0.055912482738494876,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0553,
      "loss_step_3": 0.0647,
      "step": 30580
    },
    {
      "entropy_step_1": 0.1783,
      "entropy_step_2": 0.1779,
      "entropy_step_3": 0.1597,
      "epoch": 0.9098480116593796,
      "grad_norm": 55.621952056884766,
      "learning_rate": 2.2071616812822107e-06,
      "loss": 0.055477875471115115,
      "loss_step_1": 0.0512,
      "loss_step_2": 0.0511,
      "loss_step_3": 0.0495,
      "step": 30590
    },
    {
      "entropy_step_1": 0.1823,
      "entropy_step_2": 0.1825,
      "entropy_step_3": 0.1745,
      "epoch": 0.9101454448112787,
      "grad_norm": 93.67191314697266,
      "learning_rate": 2.1927337619063127e-06,
      "loss": 0.05264759063720703,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0491,
      "step": 30600
    },
    {
      "entropy_step_1": 0.1669,
      "entropy_step_2": 0.1673,
      "entropy_step_3": 0.1605,
      "epoch": 0.9104428779631778,
      "grad_norm": 40.306976318359375,
      "learning_rate": 2.178352096711012e-06,
      "loss": 0.049588122963905336,
      "loss_step_1": 0.0431,
      "loss_step_2": 0.0421,
      "loss_step_3": 0.0397,
      "step": 30610
    },
    {
      "entropy_step_1": 0.1453,
      "entropy_step_2": 0.1455,
      "entropy_step_3": 0.1452,
      "epoch": 0.9107403111150769,
      "grad_norm": 112.67816925048828,
      "learning_rate": 2.1640166996107604e-06,
      "loss": 0.051680988073349,
      "loss_step_1": 0.046,
      "loss_step_2": 0.0458,
      "loss_step_3": 0.0462,
      "step": 30620
    },
    {
      "entropy_step_1": 0.1662,
      "entropy_step_2": 0.1681,
      "entropy_step_3": 0.1588,
      "epoch": 0.911037744266976,
      "grad_norm": 91.10005950927734,
      "learning_rate": 2.149727584475253e-06,
      "loss": 0.049520176649093625,
      "loss_step_1": 0.0471,
      "loss_step_2": 0.0463,
      "loss_step_3": 0.0451,
      "step": 30630
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1829,
      "entropy_step_3": 0.1699,
      "epoch": 0.9113351774188752,
      "grad_norm": 90.78986358642578,
      "learning_rate": 2.135484765129414e-06,
      "loss": 0.05361611247062683,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0405,
      "step": 30640
    },
    {
      "entropy_step_1": 0.1746,
      "entropy_step_2": 0.1762,
      "entropy_step_3": 0.1595,
      "epoch": 0.9116326105707743,
      "grad_norm": 248.92745971679688,
      "learning_rate": 2.1212882553533544e-06,
      "loss": 0.052522724866867064,
      "loss_step_1": 0.0476,
      "loss_step_2": 0.0473,
      "loss_step_3": 0.0459,
      "step": 30650
    },
    {
      "entropy_step_1": 0.1694,
      "entropy_step_2": 0.1701,
      "entropy_step_3": 0.1732,
      "epoch": 0.9119300437226733,
      "grad_norm": 41.811553955078125,
      "learning_rate": 2.107138068882386e-06,
      "loss": 0.0616966724395752,
      "loss_step_1": 0.0624,
      "loss_step_2": 0.0607,
      "loss_step_3": 0.059,
      "step": 30660
    },
    {
      "entropy_step_1": 0.1676,
      "entropy_step_2": 0.1683,
      "entropy_step_3": 0.1715,
      "epoch": 0.9122274768745724,
      "grad_norm": 311.60601806640625,
      "learning_rate": 2.093034219407014e-06,
      "loss": 0.05861341953277588,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0615,
      "loss_step_3": 0.0656,
      "step": 30670
    },
    {
      "entropy_step_1": 0.1783,
      "entropy_step_2": 0.1792,
      "entropy_step_3": 0.1933,
      "epoch": 0.9125249100264715,
      "grad_norm": 39.349205017089844,
      "learning_rate": 2.078976720572906e-06,
      "loss": 0.055619442462921144,
      "loss_step_1": 0.062,
      "loss_step_2": 0.0606,
      "loss_step_3": 0.0565,
      "step": 30680
    },
    {
      "entropy_step_1": 0.1587,
      "entropy_step_2": 0.1592,
      "entropy_step_3": 0.1719,
      "epoch": 0.9128223431783706,
      "grad_norm": 165.7517547607422,
      "learning_rate": 2.0649655859808825e-06,
      "loss": 0.05361272692680359,
      "loss_step_1": 0.0434,
      "loss_step_2": 0.0433,
      "loss_step_3": 0.0472,
      "step": 30690
    },
    {
      "entropy_step_1": 0.1676,
      "entropy_step_2": 0.1674,
      "entropy_step_3": 0.1834,
      "epoch": 0.9131197763302698,
      "grad_norm": 209.58961486816406,
      "learning_rate": 2.0510008291869053e-06,
      "loss": 0.054637229442596434,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0549,
      "loss_step_3": 0.0616,
      "step": 30700
    },
    {
      "entropy_step_1": 0.188,
      "entropy_step_2": 0.1872,
      "entropy_step_3": 0.1788,
      "epoch": 0.9134172094821689,
      "grad_norm": 117.81775665283203,
      "learning_rate": 2.0370824637020615e-06,
      "loss": 0.057844418287277224,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0477,
      "step": 30710
    },
    {
      "entropy_step_1": 0.1843,
      "entropy_step_2": 0.1835,
      "entropy_step_3": 0.1899,
      "epoch": 0.913714642634068,
      "grad_norm": 100.90587615966797,
      "learning_rate": 2.023210502992562e-06,
      "loss": 0.05908823013305664,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0583,
      "loss_step_3": 0.0594,
      "step": 30720
    },
    {
      "entropy_step_1": 0.166,
      "entropy_step_2": 0.1649,
      "entropy_step_3": 0.1708,
      "epoch": 0.9140120757859671,
      "grad_norm": 487.1527404785156,
      "learning_rate": 2.0093849604797267e-06,
      "loss": 0.051097053289413455,
      "loss_step_1": 0.0471,
      "loss_step_2": 0.0468,
      "loss_step_3": 0.0553,
      "step": 30730
    },
    {
      "entropy_step_1": 0.1809,
      "entropy_step_2": 0.1797,
      "entropy_step_3": 0.1681,
      "epoch": 0.9143095089378662,
      "grad_norm": 70.69013977050781,
      "learning_rate": 1.99560584953995e-06,
      "loss": 0.055016893148422244,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0545,
      "loss_step_3": 0.0494,
      "step": 30740
    },
    {
      "entropy_step_1": 0.1703,
      "entropy_step_2": 0.1693,
      "entropy_step_3": 0.1593,
      "epoch": 0.9146069420897653,
      "grad_norm": 51.7568473815918,
      "learning_rate": 1.9818731835047065e-06,
      "loss": 0.04645791351795196,
      "loss_step_1": 0.0478,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0502,
      "step": 30750
    },
    {
      "entropy_step_1": 0.1785,
      "entropy_step_2": 0.1769,
      "entropy_step_3": 0.1702,
      "epoch": 0.9149043752416645,
      "grad_norm": 91.16558837890625,
      "learning_rate": 1.9681869756605298e-06,
      "loss": 0.053270000219345096,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0466,
      "step": 30760
    },
    {
      "entropy_step_1": 0.178,
      "entropy_step_2": 0.1769,
      "entropy_step_3": 0.1704,
      "epoch": 0.9152018083935636,
      "grad_norm": 146.61294555664062,
      "learning_rate": 1.954547239249033e-06,
      "loss": 0.061105310916900635,
      "loss_step_1": 0.058,
      "loss_step_2": 0.058,
      "loss_step_3": 0.0613,
      "step": 30770
    },
    {
      "entropy_step_1": 0.1655,
      "entropy_step_2": 0.1658,
      "entropy_step_3": 0.1561,
      "epoch": 0.9154992415454627,
      "grad_norm": 153.50721740722656,
      "learning_rate": 1.940953987466826e-06,
      "loss": 0.04920172095298767,
      "loss_step_1": 0.0456,
      "loss_step_2": 0.046,
      "loss_step_3": 0.0443,
      "step": 30780
    },
    {
      "entropy_step_1": 0.1851,
      "entropy_step_2": 0.1838,
      "entropy_step_3": 0.187,
      "epoch": 0.9157966746973618,
      "grad_norm": 50.910850524902344,
      "learning_rate": 1.9274072334655834e-06,
      "loss": 0.05389063954353333,
      "loss_step_1": 0.0538,
      "loss_step_2": 0.0541,
      "loss_step_3": 0.0559,
      "step": 30790
    },
    {
      "entropy_step_1": 0.174,
      "entropy_step_2": 0.1736,
      "entropy_step_3": 0.1644,
      "epoch": 0.9160941078492609,
      "grad_norm": 67.56734466552734,
      "learning_rate": 1.91390699035196e-06,
      "loss": 0.05515690445899964,
      "loss_step_1": 0.0564,
      "loss_step_2": 0.056,
      "loss_step_3": 0.0594,
      "step": 30800
    },
    {
      "entropy_step_1": 0.1767,
      "entropy_step_2": 0.1764,
      "entropy_step_3": 0.1652,
      "epoch": 0.9163915410011599,
      "grad_norm": 125.31333923339844,
      "learning_rate": 1.90045327118763e-06,
      "loss": 0.05226263403892517,
      "loss_step_1": 0.0507,
      "loss_step_2": 0.0503,
      "loss_step_3": 0.0534,
      "step": 30810
    },
    {
      "entropy_step_1": 0.1842,
      "entropy_step_2": 0.1841,
      "entropy_step_3": 0.186,
      "epoch": 0.916688974153059,
      "grad_norm": 67.70930480957031,
      "learning_rate": 1.8870460889892472e-06,
      "loss": 0.06129276156425476,
      "loss_step_1": 0.0569,
      "loss_step_2": 0.0567,
      "loss_step_3": 0.0586,
      "step": 30820
    },
    {
      "entropy_step_1": 0.1815,
      "entropy_step_2": 0.1812,
      "entropy_step_3": 0.1881,
      "epoch": 0.9169864073049582,
      "grad_norm": 71.42411804199219,
      "learning_rate": 1.8736854567284468e-06,
      "loss": 0.04754634201526642,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0542,
      "step": 30830
    },
    {
      "entropy_step_1": 0.1805,
      "entropy_step_2": 0.1799,
      "entropy_step_3": 0.1813,
      "epoch": 0.9172838404568573,
      "grad_norm": 95.0378646850586,
      "learning_rate": 1.8603713873318219e-06,
      "loss": 0.05417336225509643,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0598,
      "step": 30840
    },
    {
      "entropy_step_1": 0.1689,
      "entropy_step_2": 0.1682,
      "entropy_step_3": 0.1732,
      "epoch": 0.9175812736087564,
      "grad_norm": 175.0673828125,
      "learning_rate": 1.8471038936809125e-06,
      "loss": 0.04972945749759674,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0512,
      "step": 30850
    },
    {
      "entropy_step_1": 0.1899,
      "entropy_step_2": 0.191,
      "entropy_step_3": 0.1851,
      "epoch": 0.9178787067606555,
      "grad_norm": 87.21027374267578,
      "learning_rate": 1.8338829886122067e-06,
      "loss": 0.05779378414154053,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.059,
      "loss_step_3": 0.061,
      "step": 30860
    },
    {
      "entropy_step_1": 0.176,
      "entropy_step_2": 0.1773,
      "entropy_step_3": 0.1787,
      "epoch": 0.9181761399125546,
      "grad_norm": 113.27627563476562,
      "learning_rate": 1.820708684917094e-06,
      "loss": 0.05019668936729431,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0488,
      "loss_step_3": 0.0508,
      "step": 30870
    },
    {
      "entropy_step_1": 0.1713,
      "entropy_step_2": 0.1715,
      "entropy_step_3": 0.1667,
      "epoch": 0.9184735730644538,
      "grad_norm": 141.28921508789062,
      "learning_rate": 1.8075809953419066e-06,
      "loss": 0.05109061598777771,
      "loss_step_1": 0.0475,
      "loss_step_2": 0.0472,
      "loss_step_3": 0.0457,
      "step": 30880
    },
    {
      "entropy_step_1": 0.1788,
      "entropy_step_2": 0.1785,
      "entropy_step_3": 0.1898,
      "epoch": 0.9187710062163529,
      "grad_norm": 175.65579223632812,
      "learning_rate": 1.794499932587862e-06,
      "loss": 0.0577808141708374,
      "loss_step_1": 0.0651,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0602,
      "step": 30890
    },
    {
      "entropy_step_1": 0.1822,
      "entropy_step_2": 0.1821,
      "entropy_step_3": 0.1691,
      "epoch": 0.919068439368252,
      "grad_norm": 122.98747253417969,
      "learning_rate": 1.7814655093110643e-06,
      "loss": 0.056814706325531004,
      "loss_step_1": 0.0597,
      "loss_step_2": 0.0597,
      "loss_step_3": 0.0584,
      "step": 30900
    },
    {
      "entropy_step_1": 0.1585,
      "entropy_step_2": 0.1598,
      "entropy_step_3": 0.1626,
      "epoch": 0.9193658725201511,
      "grad_norm": 112.6215591430664,
      "learning_rate": 1.7684777381224871e-06,
      "loss": 0.04860070943832397,
      "loss_step_1": 0.0458,
      "loss_step_2": 0.0458,
      "loss_step_3": 0.0446,
      "step": 30910
    },
    {
      "entropy_step_1": 0.1832,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.1745,
      "epoch": 0.9196633056720502,
      "grad_norm": 56.80144500732422,
      "learning_rate": 1.7555366315879895e-06,
      "loss": 0.05193840265274048,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0506,
      "step": 30920
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.1812,
      "entropy_step_3": 0.1864,
      "epoch": 0.9199607388239494,
      "grad_norm": 251.60479736328125,
      "learning_rate": 1.7426422022282619e-06,
      "loss": 0.04833770990371704,
      "loss_step_1": 0.0481,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0476,
      "step": 30930
    },
    {
      "entropy_step_1": 0.175,
      "entropy_step_2": 0.1761,
      "entropy_step_3": 0.1918,
      "epoch": 0.9202581719758485,
      "grad_norm": 39.679080963134766,
      "learning_rate": 1.729794462518841e-06,
      "loss": 0.05486779808998108,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0539,
      "loss_step_3": 0.0655,
      "step": 30940
    },
    {
      "entropy_step_1": 0.1604,
      "entropy_step_2": 0.1618,
      "entropy_step_3": 0.155,
      "epoch": 0.9205556051277476,
      "grad_norm": 48.148887634277344,
      "learning_rate": 1.7169934248900844e-06,
      "loss": 0.04492431581020355,
      "loss_step_1": 0.04,
      "loss_step_2": 0.0393,
      "loss_step_3": 0.0345,
      "step": 30950
    },
    {
      "entropy_step_1": 0.1525,
      "entropy_step_2": 0.1521,
      "entropy_step_3": 0.1412,
      "epoch": 0.9208530382796466,
      "grad_norm": 48.73377227783203,
      "learning_rate": 1.7042391017271731e-06,
      "loss": 0.048019447922706605,
      "loss_step_1": 0.0369,
      "loss_step_2": 0.0367,
      "loss_step_3": 0.0359,
      "step": 30960
    },
    {
      "entropy_step_1": 0.1715,
      "entropy_step_2": 0.1733,
      "entropy_step_3": 0.1577,
      "epoch": 0.9211504714315457,
      "grad_norm": 91.20909118652344,
      "learning_rate": 1.6915315053700976e-06,
      "loss": 0.05571114420890808,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0565,
      "loss_step_3": 0.0501,
      "step": 30970
    },
    {
      "entropy_step_1": 0.1753,
      "entropy_step_2": 0.1767,
      "entropy_step_3": 0.174,
      "epoch": 0.9214479045834448,
      "grad_norm": 70.29375457763672,
      "learning_rate": 1.6788706481136174e-06,
      "loss": 0.04837614297866821,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.0532,
      "loss_step_3": 0.0574,
      "step": 30980
    },
    {
      "entropy_step_1": 0.1809,
      "entropy_step_2": 0.1799,
      "entropy_step_3": 0.1595,
      "epoch": 0.921745337735344,
      "grad_norm": 82.87957763671875,
      "learning_rate": 1.6662565422072894e-06,
      "loss": 0.06352748870849609,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0619,
      "step": 30990
    },
    {
      "entropy_step_1": 0.1728,
      "entropy_step_2": 0.1731,
      "entropy_step_3": 0.1647,
      "epoch": 0.9220427708872431,
      "grad_norm": 283.85699462890625,
      "learning_rate": 1.653689199855435e-06,
      "loss": 0.057045894861221316,
      "loss_step_1": 0.057,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0479,
      "step": 31000
    },
    {
      "epoch": 0.9220427708872431,
      "eval_accuracy": 0.926,
      "eval_f1": 0.9259499421609008,
      "eval_loss": 0.08331938087940216,
      "eval_precision": 0.9271550271935312,
      "eval_recall": 0.926,
      "eval_runtime": 6.7161,
      "eval_samples_per_second": 74.448,
      "eval_steps_per_second": 9.38,
      "step": 31000
    },
    {
      "entropy_step_1": 0.1787,
      "entropy_step_2": 0.1796,
      "entropy_step_3": 0.1728,
      "epoch": 0.9223402040391422,
      "grad_norm": 263.3998107910156,
      "learning_rate": 1.6411686332171217e-06,
      "loss": 0.055403506755828856,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0424,
      "step": 31010
    },
    {
      "entropy_step_1": 0.1745,
      "entropy_step_2": 0.175,
      "entropy_step_3": 0.1806,
      "epoch": 0.9226376371910413,
      "grad_norm": 63.050418853759766,
      "learning_rate": 1.6286948544061819e-06,
      "loss": 0.05074782371520996,
      "loss_step_1": 0.054,
      "loss_step_2": 0.0533,
      "loss_step_3": 0.0577,
      "step": 31020
    },
    {
      "entropy_step_1": 0.1767,
      "entropy_step_2": 0.1788,
      "entropy_step_3": 0.1892,
      "epoch": 0.9229350703429404,
      "grad_norm": 91.11695098876953,
      "learning_rate": 1.6162678754911454e-06,
      "loss": 0.049527278542518614,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0597,
      "step": 31030
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.18,
      "entropy_step_3": 0.173,
      "epoch": 0.9232325034948395,
      "grad_norm": 105.78659057617188,
      "learning_rate": 1.6038877084953052e-06,
      "loss": 0.05458589792251587,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0599,
      "loss_step_3": 0.0649,
      "step": 31040
    },
    {
      "entropy_step_1": 0.1422,
      "entropy_step_2": 0.1432,
      "entropy_step_3": 0.1538,
      "epoch": 0.9235299366467387,
      "grad_norm": 92.70982360839844,
      "learning_rate": 1.5915543653966303e-06,
      "loss": 0.052969717979431154,
      "loss_step_1": 0.0517,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0557,
      "step": 31050
    },
    {
      "entropy_step_1": 0.1726,
      "entropy_step_2": 0.1733,
      "entropy_step_3": 0.1847,
      "epoch": 0.9238273697986378,
      "grad_norm": 99.90088653564453,
      "learning_rate": 1.5792678581277975e-06,
      "loss": 0.057742524147033694,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.0518,
      "loss_step_3": 0.0623,
      "step": 31060
    },
    {
      "entropy_step_1": 0.1855,
      "entropy_step_2": 0.1862,
      "entropy_step_3": 0.1839,
      "epoch": 0.9241248029505369,
      "grad_norm": 122.88029479980469,
      "learning_rate": 1.5670281985761702e-06,
      "loss": 0.05505260229110718,
      "loss_step_1": 0.0584,
      "loss_step_2": 0.0576,
      "loss_step_3": 0.0508,
      "step": 31070
    },
    {
      "entropy_step_1": 0.1722,
      "entropy_step_2": 0.1725,
      "entropy_step_3": 0.157,
      "epoch": 0.924422236102436,
      "grad_norm": 40.67207336425781,
      "learning_rate": 1.5548353985837872e-06,
      "loss": 0.0517300546169281,
      "loss_step_1": 0.0588,
      "loss_step_2": 0.0584,
      "loss_step_3": 0.0445,
      "step": 31080
    },
    {
      "entropy_step_1": 0.1814,
      "entropy_step_2": 0.1817,
      "entropy_step_3": 0.1924,
      "epoch": 0.9247196692543351,
      "grad_norm": 873.2354125976562,
      "learning_rate": 1.5426894699473448e-06,
      "loss": 0.051506704092025755,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0529,
      "loss_step_3": 0.0531,
      "step": 31090
    },
    {
      "entropy_step_1": 0.1903,
      "entropy_step_2": 0.1918,
      "entropy_step_3": 0.1967,
      "epoch": 0.9250171024062342,
      "grad_norm": 52.41923904418945,
      "learning_rate": 1.530590424418199e-06,
      "loss": 0.056190776824951175,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0558,
      "loss_step_3": 0.0579,
      "step": 31100
    },
    {
      "entropy_step_1": 0.1845,
      "entropy_step_2": 0.1845,
      "entropy_step_3": 0.1742,
      "epoch": 0.9253145355581333,
      "grad_norm": 59.67142868041992,
      "learning_rate": 1.5185382737023468e-06,
      "loss": 0.0534189522266388,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0543,
      "loss_step_3": 0.0456,
      "step": 31110
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1734,
      "entropy_step_3": 0.1693,
      "epoch": 0.9256119687100324,
      "grad_norm": 211.4268341064453,
      "learning_rate": 1.5065330294603941e-06,
      "loss": 0.06265792846679688,
      "loss_step_1": 0.0562,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0543,
      "step": 31120
    },
    {
      "entropy_step_1": 0.1759,
      "entropy_step_2": 0.1758,
      "entropy_step_3": 0.1631,
      "epoch": 0.9259094018619315,
      "grad_norm": 679.1532592773438,
      "learning_rate": 1.494574703307583e-06,
      "loss": 0.04945343434810638,
      "loss_step_1": 0.0528,
      "loss_step_2": 0.0524,
      "loss_step_3": 0.0454,
      "step": 31130
    },
    {
      "entropy_step_1": 0.1709,
      "entropy_step_2": 0.172,
      "entropy_step_3": 0.1841,
      "epoch": 0.9262068350138306,
      "grad_norm": 188.3584747314453,
      "learning_rate": 1.4826633068137586e-06,
      "loss": 0.05300154685974121,
      "loss_step_1": 0.0465,
      "loss_step_2": 0.0465,
      "loss_step_3": 0.0549,
      "step": 31140
    },
    {
      "entropy_step_1": 0.1724,
      "entropy_step_2": 0.1731,
      "entropy_step_3": 0.1603,
      "epoch": 0.9265042681657297,
      "grad_norm": 44.43907928466797,
      "learning_rate": 1.4707988515033632e-06,
      "loss": 0.05593100190162659,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0497,
      "step": 31150
    },
    {
      "entropy_step_1": 0.1756,
      "entropy_step_2": 0.1761,
      "entropy_step_3": 0.1583,
      "epoch": 0.9268017013176288,
      "grad_norm": 52.27767562866211,
      "learning_rate": 1.458981348855415e-06,
      "loss": 0.05449582934379578,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0552,
      "loss_step_3": 0.0494,
      "step": 31160
    },
    {
      "entropy_step_1": 0.1947,
      "entropy_step_2": 0.1958,
      "entropy_step_3": 0.2114,
      "epoch": 0.927099134469528,
      "grad_norm": 162.9822540283203,
      "learning_rate": 1.4472108103035064e-06,
      "loss": 0.057255369424819944,
      "loss_step_1": 0.0591,
      "loss_step_2": 0.0582,
      "loss_step_3": 0.0606,
      "step": 31170
    },
    {
      "entropy_step_1": 0.1727,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1723,
      "epoch": 0.9273965676214271,
      "grad_norm": 65.709716796875,
      "learning_rate": 1.4354872472357895e-06,
      "loss": 0.06220555305480957,
      "loss_step_1": 0.0599,
      "loss_step_2": 0.0592,
      "loss_step_3": 0.0552,
      "step": 31180
    },
    {
      "entropy_step_1": 0.1744,
      "entropy_step_2": 0.1757,
      "entropy_step_3": 0.1658,
      "epoch": 0.9276940007733262,
      "grad_norm": 54.35511779785156,
      "learning_rate": 1.4238106709949795e-06,
      "loss": 0.05794340968132019,
      "loss_step_1": 0.0606,
      "loss_step_2": 0.0604,
      "loss_step_3": 0.0578,
      "step": 31190
    },
    {
      "entropy_step_1": 0.1753,
      "entropy_step_2": 0.1759,
      "entropy_step_3": 0.1537,
      "epoch": 0.9279914339252253,
      "grad_norm": 124.92884826660156,
      "learning_rate": 1.412181092878323e-06,
      "loss": 0.053211653232574464,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0559,
      "loss_step_3": 0.0594,
      "step": 31200
    },
    {
      "entropy_step_1": 0.1729,
      "entropy_step_2": 0.1718,
      "entropy_step_3": 0.1922,
      "epoch": 0.9282888670771244,
      "grad_norm": 54.00627899169922,
      "learning_rate": 1.400598524137592e-06,
      "loss": 0.04842055141925812,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0491,
      "loss_step_3": 0.0567,
      "step": 31210
    },
    {
      "entropy_step_1": 0.1742,
      "entropy_step_2": 0.1745,
      "entropy_step_3": 0.1626,
      "epoch": 0.9285863002290236,
      "grad_norm": 67.40380859375,
      "learning_rate": 1.3890629759790774e-06,
      "loss": 0.05150827169418335,
      "loss_step_1": 0.049,
      "loss_step_2": 0.049,
      "loss_step_3": 0.0425,
      "step": 31220
    },
    {
      "entropy_step_1": 0.1858,
      "entropy_step_2": 0.1864,
      "entropy_step_3": 0.1985,
      "epoch": 0.9288837333809227,
      "grad_norm": 179.95675659179688,
      "learning_rate": 1.3775744595635854e-06,
      "loss": 0.05992352366447449,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0623,
      "step": 31230
    },
    {
      "entropy_step_1": 0.1705,
      "entropy_step_2": 0.1709,
      "entropy_step_3": 0.1789,
      "epoch": 0.9291811665328218,
      "grad_norm": 48.50631332397461,
      "learning_rate": 1.366132986006402e-06,
      "loss": 0.049374693632125856,
      "loss_step_1": 0.0495,
      "loss_step_2": 0.0493,
      "loss_step_3": 0.0539,
      "step": 31240
    },
    {
      "entropy_step_1": 0.1698,
      "entropy_step_2": 0.1702,
      "entropy_step_3": 0.172,
      "epoch": 0.9294785996847209,
      "grad_norm": 46.372982025146484,
      "learning_rate": 1.354738566377317e-06,
      "loss": 0.04998637139797211,
      "loss_step_1": 0.0485,
      "loss_step_2": 0.0485,
      "loss_step_3": 0.0444,
      "step": 31250
    },
    {
      "entropy_step_1": 0.1986,
      "entropy_step_2": 0.1982,
      "entropy_step_3": 0.208,
      "epoch": 0.9297760328366199,
      "grad_norm": 40.30546569824219,
      "learning_rate": 1.3433912117005898e-06,
      "loss": 0.05575247406959534,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.0538,
      "loss_step_3": 0.0608,
      "step": 31260
    },
    {
      "entropy_step_1": 0.1802,
      "entropy_step_2": 0.1794,
      "entropy_step_3": 0.1858,
      "epoch": 0.930073465988519,
      "grad_norm": 259.21697998046875,
      "learning_rate": 1.3320909329549213e-06,
      "loss": 0.05675622820854187,
      "loss_step_1": 0.0583,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0653,
      "step": 31270
    },
    {
      "entropy_step_1": 0.1702,
      "entropy_step_2": 0.1702,
      "entropy_step_3": 0.1917,
      "epoch": 0.9303708991404182,
      "grad_norm": 116.03636932373047,
      "learning_rate": 1.320837741073505e-06,
      "loss": 0.057075494527816774,
      "loss_step_1": 0.0551,
      "loss_step_2": 0.0557,
      "loss_step_3": 0.0563,
      "step": 31280
    },
    {
      "entropy_step_1": 0.1754,
      "entropy_step_2": 0.1747,
      "entropy_step_3": 0.1693,
      "epoch": 0.9306683322923173,
      "grad_norm": 52.762901306152344,
      "learning_rate": 1.309631646943954e-06,
      "loss": 0.05104058384895325,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0551,
      "step": 31290
    },
    {
      "entropy_step_1": 0.1746,
      "entropy_step_2": 0.1746,
      "entropy_step_3": 0.1751,
      "epoch": 0.9309657654442164,
      "grad_norm": 198.08740234375,
      "learning_rate": 1.2984726614083065e-06,
      "loss": 0.061705803871154784,
      "loss_step_1": 0.0642,
      "loss_step_2": 0.0643,
      "loss_step_3": 0.0727,
      "step": 31300
    },
    {
      "entropy_step_1": 0.1907,
      "entropy_step_2": 0.1908,
      "entropy_step_3": 0.1896,
      "epoch": 0.9312631985961155,
      "grad_norm": 309.1460266113281,
      "learning_rate": 1.2873607952630373e-06,
      "loss": 0.05308011174201965,
      "loss_step_1": 0.0581,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0601,
      "step": 31310
    },
    {
      "entropy_step_1": 0.1679,
      "entropy_step_2": 0.1673,
      "entropy_step_3": 0.1561,
      "epoch": 0.9315606317480146,
      "grad_norm": 61.187294006347656,
      "learning_rate": 1.2762960592590356e-06,
      "loss": 0.05559890866279602,
      "loss_step_1": 0.0497,
      "loss_step_2": 0.0495,
      "loss_step_3": 0.0464,
      "step": 31320
    },
    {
      "entropy_step_1": 0.1778,
      "entropy_step_2": 0.1777,
      "entropy_step_3": 0.1647,
      "epoch": 0.9318580648999137,
      "grad_norm": 158.34820556640625,
      "learning_rate": 1.2652784641015768e-06,
      "loss": 0.05207775235176086,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0523,
      "loss_step_3": 0.0446,
      "step": 31330
    },
    {
      "entropy_step_1": 0.1763,
      "entropy_step_2": 0.1758,
      "entropy_step_3": 0.1941,
      "epoch": 0.9321554980518129,
      "grad_norm": 129.64186096191406,
      "learning_rate": 1.25430802045034e-06,
      "loss": 0.05028018355369568,
      "loss_step_1": 0.0539,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0623,
      "step": 31340
    },
    {
      "entropy_step_1": 0.1909,
      "entropy_step_2": 0.1908,
      "entropy_step_3": 0.1837,
      "epoch": 0.932452931203712,
      "grad_norm": 135.13182067871094,
      "learning_rate": 1.2433847389193676e-06,
      "loss": 0.053122717142105105,
      "loss_step_1": 0.0469,
      "loss_step_2": 0.0463,
      "loss_step_3": 0.0473,
      "step": 31350
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1794,
      "entropy_step_3": 0.1818,
      "epoch": 0.9327503643556111,
      "grad_norm": 67.86357879638672,
      "learning_rate": 1.2325086300770948e-06,
      "loss": 0.054536813497543336,
      "loss_step_1": 0.0534,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0587,
      "step": 31360
    },
    {
      "entropy_step_1": 0.1849,
      "entropy_step_2": 0.1858,
      "entropy_step_3": 0.1988,
      "epoch": 0.9330477975075102,
      "grad_norm": 59.658172607421875,
      "learning_rate": 1.2216797044462936e-06,
      "loss": 0.05831027030944824,
      "loss_step_1": 0.0571,
      "loss_step_2": 0.0569,
      "loss_step_3": 0.0588,
      "step": 31370
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.1828,
      "entropy_step_3": 0.2046,
      "epoch": 0.9333452306594093,
      "grad_norm": 110.52294921875,
      "learning_rate": 1.2108979725041104e-06,
      "loss": 0.05040190815925598,
      "loss_step_1": 0.0563,
      "loss_step_2": 0.0562,
      "loss_step_3": 0.0695,
      "step": 31380
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1793,
      "entropy_step_3": 0.2096,
      "epoch": 0.9336426638113084,
      "grad_norm": 426.3851013183594,
      "learning_rate": 1.200163444682012e-06,
      "loss": 0.05031658411026001,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0683,
      "step": 31390
    },
    {
      "entropy_step_1": 0.174,
      "entropy_step_2": 0.1736,
      "entropy_step_3": 0.1778,
      "epoch": 0.9339400969632076,
      "grad_norm": 43.07750701904297,
      "learning_rate": 1.1894761313657964e-06,
      "loss": 0.05125383734703064,
      "loss_step_1": 0.0552,
      "loss_step_2": 0.055,
      "loss_step_3": 0.0584,
      "step": 31400
    },
    {
      "entropy_step_1": 0.172,
      "entropy_step_2": 0.1718,
      "entropy_step_3": 0.178,
      "epoch": 0.9342375301151067,
      "grad_norm": 70.70314025878906,
      "learning_rate": 1.178836042895587e-06,
      "loss": 0.052258217334747316,
      "loss_step_1": 0.0504,
      "loss_step_2": 0.0499,
      "loss_step_3": 0.0519,
      "step": 31410
    },
    {
      "entropy_step_1": 0.1736,
      "entropy_step_2": 0.1732,
      "entropy_step_3": 0.1617,
      "epoch": 0.9345349632670057,
      "grad_norm": 183.1492156982422,
      "learning_rate": 1.1682431895658097e-06,
      "loss": 0.0512657105922699,
      "loss_step_1": 0.0479,
      "loss_step_2": 0.0476,
      "loss_step_3": 0.0398,
      "step": 31420
    },
    {
      "entropy_step_1": 0.1616,
      "entropy_step_2": 0.1622,
      "entropy_step_3": 0.1806,
      "epoch": 0.9348323964189048,
      "grad_norm": 219.20802307128906,
      "learning_rate": 1.1576975816252056e-06,
      "loss": 0.05024052858352661,
      "loss_step_1": 0.0466,
      "loss_step_2": 0.0461,
      "loss_step_3": 0.0491,
      "step": 31430
    },
    {
      "entropy_step_1": 0.1702,
      "entropy_step_2": 0.1709,
      "entropy_step_3": 0.1639,
      "epoch": 0.9351298295708039,
      "grad_norm": 259.4607238769531,
      "learning_rate": 1.1471992292767852e-06,
      "loss": 0.05908637046813965,
      "loss_step_1": 0.0587,
      "loss_step_2": 0.0588,
      "loss_step_3": 0.0531,
      "step": 31440
    },
    {
      "entropy_step_1": 0.1871,
      "entropy_step_2": 0.1867,
      "entropy_step_3": 0.1846,
      "epoch": 0.935427262722703,
      "grad_norm": 227.81492614746094,
      "learning_rate": 1.1367481426778515e-06,
      "loss": 0.055381500720977785,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.0508,
      "step": 31450
    },
    {
      "entropy_step_1": 0.1853,
      "entropy_step_2": 0.1846,
      "entropy_step_3": 0.1776,
      "epoch": 0.9357246958746022,
      "grad_norm": 105.46141052246094,
      "learning_rate": 1.1263443319399602e-06,
      "loss": 0.05870810747146606,
      "loss_step_1": 0.0625,
      "loss_step_2": 0.0628,
      "loss_step_3": 0.0617,
      "step": 31460
    },
    {
      "entropy_step_1": 0.1791,
      "entropy_step_2": 0.1781,
      "entropy_step_3": 0.1823,
      "epoch": 0.9360221290265013,
      "grad_norm": 39.75111770629883,
      "learning_rate": 1.1159878071289598e-06,
      "loss": 0.05113060474395752,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0546,
      "step": 31470
    },
    {
      "entropy_step_1": 0.1834,
      "entropy_step_2": 0.1829,
      "entropy_step_3": 0.166,
      "epoch": 0.9363195621784004,
      "grad_norm": 475.1911926269531,
      "learning_rate": 1.105678578264918e-06,
      "loss": 0.053466242551803586,
      "loss_step_1": 0.0513,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0485,
      "step": 31480
    },
    {
      "entropy_step_1": 0.1942,
      "entropy_step_2": 0.193,
      "entropy_step_3": 0.184,
      "epoch": 0.9366169953302995,
      "grad_norm": 63.501468658447266,
      "learning_rate": 1.0954166553221568e-06,
      "loss": 0.06263448596000672,
      "loss_step_1": 0.0687,
      "loss_step_2": 0.0683,
      "loss_step_3": 0.067,
      "step": 31490
    },
    {
      "entropy_step_1": 0.1632,
      "entropy_step_2": 0.1619,
      "entropy_step_3": 0.1569,
      "epoch": 0.9369144284821986,
      "grad_norm": 74.32173919677734,
      "learning_rate": 1.0852020482292169e-06,
      "loss": 0.05692804455757141,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0542,
      "loss_step_3": 0.0471,
      "step": 31500
    },
    {
      "entropy_step_1": 0.1815,
      "entropy_step_2": 0.1809,
      "entropy_step_3": 0.1575,
      "epoch": 0.9372118616340978,
      "grad_norm": 70.23541259765625,
      "learning_rate": 1.0750347668688822e-06,
      "loss": 0.05353488326072693,
      "loss_step_1": 0.0553,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0402,
      "step": 31510
    },
    {
      "entropy_step_1": 0.1824,
      "entropy_step_2": 0.1818,
      "entropy_step_3": 0.1822,
      "epoch": 0.9375092947859969,
      "grad_norm": 51.616825103759766,
      "learning_rate": 1.0649148210781335e-06,
      "loss": 0.053082597255706784,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0566,
      "step": 31520
    },
    {
      "entropy_step_1": 0.187,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1649,
      "epoch": 0.937806727937896,
      "grad_norm": 148.09303283691406,
      "learning_rate": 1.0548422206481556e-06,
      "loss": 0.05524142980575562,
      "loss_step_1": 0.0555,
      "loss_step_2": 0.0551,
      "loss_step_3": 0.0496,
      "step": 31530
    },
    {
      "entropy_step_1": 0.1818,
      "entropy_step_2": 0.1814,
      "entropy_step_3": 0.1776,
      "epoch": 0.9381041610897951,
      "grad_norm": 247.95187377929688,
      "learning_rate": 1.0448169753243143e-06,
      "loss": 0.04824863374233246,
      "loss_step_1": 0.048,
      "loss_step_2": 0.0477,
      "loss_step_3": 0.0442,
      "step": 31540
    },
    {
      "entropy_step_1": 0.1662,
      "entropy_step_2": 0.1674,
      "entropy_step_3": 0.19,
      "epoch": 0.9384015942416942,
      "grad_norm": 68.01253509521484,
      "learning_rate": 1.0348390948061892e-06,
      "loss": 0.049005982279777524,
      "loss_step_1": 0.049,
      "loss_step_2": 0.0487,
      "loss_step_3": 0.0526,
      "step": 31550
    },
    {
      "entropy_step_1": 0.1651,
      "entropy_step_2": 0.165,
      "entropy_step_3": 0.1481,
      "epoch": 0.9386990273935933,
      "grad_norm": 81.65979766845703,
      "learning_rate": 1.0249085887475084e-06,
      "loss": 0.05900182127952576,
      "loss_step_1": 0.0545,
      "loss_step_2": 0.0536,
      "loss_step_3": 0.0486,
      "step": 31560
    },
    {
      "entropy_step_1": 0.1933,
      "entropy_step_2": 0.1929,
      "entropy_step_3": 0.1884,
      "epoch": 0.9389964605454924,
      "grad_norm": 146.90403747558594,
      "learning_rate": 1.0150254667561642e-06,
      "loss": 0.050295460224151614,
      "loss_step_1": 0.0566,
      "loss_step_2": 0.0563,
      "loss_step_3": 0.0537,
      "step": 31570
    },
    {
      "entropy_step_1": 0.16,
      "entropy_step_2": 0.1598,
      "entropy_step_3": 0.161,
      "epoch": 0.9392938936973915,
      "grad_norm": 57.02243423461914,
      "learning_rate": 1.0051897383942188e-06,
      "loss": 0.053691303730010985,
      "loss_step_1": 0.0472,
      "loss_step_2": 0.0468,
      "loss_step_3": 0.0423,
      "step": 31580
    },
    {
      "entropy_step_1": 0.1588,
      "entropy_step_2": 0.1595,
      "entropy_step_3": 0.1653,
      "epoch": 0.9395913268492906,
      "grad_norm": 106.61156463623047,
      "learning_rate": 9.954014131778767e-07,
      "loss": 0.05217209458351135,
      "loss_step_1": 0.0434,
      "loss_step_2": 0.0431,
      "loss_step_3": 0.0418,
      "step": 31590
    },
    {
      "entropy_step_1": 0.1732,
      "entropy_step_2": 0.1731,
      "entropy_step_3": 0.1616,
      "epoch": 0.9398887600011897,
      "grad_norm": 36.022056579589844,
      "learning_rate": 9.856605005774732e-07,
      "loss": 0.05503876209259033,
      "loss_step_1": 0.0546,
      "loss_step_2": 0.0544,
      "loss_step_3": 0.0569,
      "step": 31600
    },
    {
      "entropy_step_1": 0.1999,
      "entropy_step_2": 0.2006,
      "entropy_step_3": 0.2129,
      "epoch": 0.9401861931530888,
      "grad_norm": 198.62820434570312,
      "learning_rate": 9.7596701001747e-07,
      "loss": 0.05581230521202087,
      "loss_step_1": 0.0634,
      "loss_step_2": 0.0631,
      "loss_step_3": 0.0691,
      "step": 31610
    },
    {
      "entropy_step_1": 0.1716,
      "entropy_step_2": 0.1711,
      "entropy_step_3": 0.1768,
      "epoch": 0.9404836263049879,
      "grad_norm": 65.87617492675781,
      "learning_rate": 9.663209508764537e-07,
      "loss": 0.04808252453804016,
      "loss_step_1": 0.0498,
      "loss_step_2": 0.0498,
      "loss_step_3": 0.0515,
      "step": 31620
    },
    {
      "entropy_step_1": 0.174,
      "entropy_step_2": 0.1743,
      "entropy_step_3": 0.18,
      "epoch": 0.9407810594568871,
      "grad_norm": 706.6361083984375,
      "learning_rate": 9.56722332487109e-07,
      "loss": 0.04839664995670319,
      "loss_step_1": 0.0511,
      "loss_step_2": 0.0508,
      "loss_step_3": 0.0542,
      "step": 31630
    },
    {
      "entropy_step_1": 0.1855,
      "entropy_step_2": 0.1853,
      "entropy_step_3": 0.1604,
      "epoch": 0.9410784926087862,
      "grad_norm": 72.54371643066406,
      "learning_rate": 9.471711641362457e-07,
      "loss": 0.05307072401046753,
      "loss_step_1": 0.0572,
      "loss_step_2": 0.0572,
      "loss_step_3": 0.0407,
      "step": 31640
    },
    {
      "entropy_step_1": 0.185,
      "entropy_step_2": 0.185,
      "entropy_step_3": 0.1871,
      "epoch": 0.9413759257606853,
      "grad_norm": 191.2977294921875,
      "learning_rate": 9.376674550647391e-07,
      "loss": 0.053853672742843625,
      "loss_step_1": 0.0573,
      "loss_step_2": 0.0571,
      "loss_step_3": 0.0553,
      "step": 31650
    },
    {
      "entropy_step_1": 0.1806,
      "entropy_step_2": 0.1807,
      "entropy_step_3": 0.1782,
      "epoch": 0.9416733589125844,
      "grad_norm": 147.48089599609375,
      "learning_rate": 9.282112144675669e-07,
      "loss": 0.05418532490730286,
      "loss_step_1": 0.059,
      "loss_step_2": 0.0585,
      "loss_step_3": 0.0605,
      "step": 31660
    },
    {
      "entropy_step_1": 0.1873,
      "entropy_step_2": 0.1872,
      "entropy_step_3": 0.1977,
      "epoch": 0.9419707920644835,
      "grad_norm": 140.83290100097656,
      "learning_rate": 9.188024514937554e-07,
      "loss": 0.047946909070014955,
      "loss_step_1": 0.0527,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0589,
      "step": 31670
    },
    {
      "entropy_step_1": 0.169,
      "entropy_step_2": 0.17,
      "entropy_step_3": 0.1745,
      "epoch": 0.9422682252163826,
      "grad_norm": 40.04920196533203,
      "learning_rate": 9.094411752464171e-07,
      "loss": 0.05609031915664673,
      "loss_step_1": 0.0548,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0474,
      "step": 31680
    },
    {
      "entropy_step_1": 0.1877,
      "entropy_step_2": 0.1871,
      "entropy_step_3": 0.1964,
      "epoch": 0.9425656583682818,
      "grad_norm": 62.46466827392578,
      "learning_rate": 9.001273947827183e-07,
      "loss": 0.054705333709716794,
      "loss_step_1": 0.0558,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.058,
      "step": 31690
    },
    {
      "entropy_step_1": 0.183,
      "entropy_step_2": 0.1831,
      "entropy_step_3": 0.1634,
      "epoch": 0.9428630915201809,
      "grad_norm": 115.6763687133789,
      "learning_rate": 8.908611191138672e-07,
      "loss": 0.05008330345153809,
      "loss_step_1": 0.0515,
      "loss_step_2": 0.0513,
      "loss_step_3": 0.0401,
      "step": 31700
    },
    {
      "entropy_step_1": 0.1737,
      "entropy_step_2": 0.1743,
      "entropy_step_3": 0.172,
      "epoch": 0.94316052467208,
      "grad_norm": 70.97384643554688,
      "learning_rate": 8.816423572051091e-07,
      "loss": 0.044180291891098025,
      "loss_step_1": 0.0418,
      "loss_step_2": 0.0416,
      "loss_step_3": 0.037,
      "step": 31710
    },
    {
      "entropy_step_1": 0.1974,
      "entropy_step_2": 0.197,
      "entropy_step_3": 0.2099,
      "epoch": 0.943457957823979,
      "grad_norm": 84.55717468261719,
      "learning_rate": 8.724711179757206e-07,
      "loss": 0.05709373354911804,
      "loss_step_1": 0.0596,
      "loss_step_2": 0.0595,
      "loss_step_3": 0.0639,
      "step": 31720
    },
    {
      "entropy_step_1": 0.1779,
      "entropy_step_2": 0.1789,
      "entropy_step_3": 0.1923,
      "epoch": 0.9437553909758781,
      "grad_norm": 158.9545135498047,
      "learning_rate": 8.633474102989981e-07,
      "loss": 0.054889333248138425,
      "loss_step_1": 0.0542,
      "loss_step_2": 0.0537,
      "loss_step_3": 0.0559,
      "step": 31730
    },
    {
      "entropy_step_1": 0.1847,
      "entropy_step_2": 0.1851,
      "entropy_step_3": 0.1723,
      "epoch": 0.9440528241277772,
      "grad_norm": 322.800048828125,
      "learning_rate": 8.542712430022637e-07,
      "loss": 0.0544084370136261,
      "loss_step_1": 0.0579,
      "loss_step_2": 0.0574,
      "loss_step_3": 0.0499,
      "step": 31740
    },
    {
      "entropy_step_1": 0.1668,
      "entropy_step_2": 0.1686,
      "entropy_step_3": 0.1487,
      "epoch": 0.9443502572796764,
      "grad_norm": 98.31466674804688,
      "learning_rate": 8.452426248668266e-07,
      "loss": 0.050858336687088015,
      "loss_step_1": 0.0531,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0405,
      "step": 31750
    },
    {
      "entropy_step_1": 0.1841,
      "entropy_step_2": 0.184,
      "entropy_step_3": 0.1795,
      "epoch": 0.9446476904315755,
      "grad_norm": 368.7783508300781,
      "learning_rate": 8.362615646279992e-07,
      "loss": 0.06090267896652222,
      "loss_step_1": 0.0568,
      "loss_step_2": 0.0564,
      "loss_step_3": 0.0572,
      "step": 31760
    },
    {
      "entropy_step_1": 0.1712,
      "entropy_step_2": 0.1709,
      "entropy_step_3": 0.1774,
      "epoch": 0.9449451235834746,
      "grad_norm": 828.5620727539062,
      "learning_rate": 8.273280709750697e-07,
      "loss": 0.05506492257118225,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.051,
      "loss_step_3": 0.0533,
      "step": 31770
    },
    {
      "entropy_step_1": 0.1671,
      "entropy_step_2": 0.1673,
      "entropy_step_3": 0.1496,
      "epoch": 0.9452425567353737,
      "grad_norm": 52.42852783203125,
      "learning_rate": 8.184421525513297e-07,
      "loss": 0.050803637504577635,
      "loss_step_1": 0.0433,
      "loss_step_2": 0.0427,
      "loss_step_3": 0.041,
      "step": 31780
    },
    {
      "entropy_step_1": 0.1636,
      "entropy_step_2": 0.1651,
      "entropy_step_3": 0.1687,
      "epoch": 0.9455399898872728,
      "grad_norm": 69.06958770751953,
      "learning_rate": 8.096038179540133e-07,
      "loss": 0.05123254656791687,
      "loss_step_1": 0.0544,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0611,
      "step": 31790
    },
    {
      "entropy_step_1": 0.1833,
      "entropy_step_2": 0.1833,
      "entropy_step_3": 0.1795,
      "epoch": 0.945837423039172,
      "grad_norm": 54.71040725708008,
      "learning_rate": 8.008130757343413e-07,
      "loss": 0.05252204537391662,
      "loss_step_1": 0.0516,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0525,
      "step": 31800
    },
    {
      "entropy_step_1": 0.1914,
      "entropy_step_2": 0.1922,
      "entropy_step_3": 0.1967,
      "epoch": 0.9461348561910711,
      "grad_norm": 88.88165283203125,
      "learning_rate": 7.920699343974658e-07,
      "loss": 0.06048430204391479,
      "loss_step_1": 0.0618,
      "loss_step_2": 0.0609,
      "loss_step_3": 0.0591,
      "step": 31810
    },
    {
      "entropy_step_1": 0.1793,
      "entropy_step_2": 0.1789,
      "entropy_step_3": 0.1863,
      "epoch": 0.9464322893429702,
      "grad_norm": 154.00071716308594,
      "learning_rate": 7.833744024024981e-07,
      "loss": 0.05219094753265381,
      "loss_step_1": 0.0522,
      "loss_step_2": 0.052,
      "loss_step_3": 0.0531,
      "step": 31820
    },
    {
      "entropy_step_1": 0.1803,
      "entropy_step_2": 0.1814,
      "entropy_step_3": 0.1552,
      "epoch": 0.9467297224948693,
      "grad_norm": 258.5691223144531,
      "learning_rate": 7.747264881624916e-07,
      "loss": 0.05755976438522339,
      "loss_step_1": 0.0529,
      "loss_step_2": 0.0528,
      "loss_step_3": 0.0448,
      "step": 31830
    },
    {
      "entropy_step_1": 0.1795,
      "entropy_step_2": 0.1791,
      "entropy_step_3": 0.193,
      "epoch": 0.9470271556467684,
      "grad_norm": 121.60330963134766,
      "learning_rate": 7.661262000444091e-07,
      "loss": 0.05454160571098328,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0602,
      "step": 31840
    },
    {
      "entropy_step_1": 0.1721,
      "entropy_step_2": 0.1726,
      "entropy_step_3": 0.1646,
      "epoch": 0.9473245887986675,
      "grad_norm": 57.596500396728516,
      "learning_rate": 7.575735463691558e-07,
      "loss": 0.055898630619049074,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0519,
      "loss_step_3": 0.0503,
      "step": 31850
    },
    {
      "entropy_step_1": 0.1814,
      "entropy_step_2": 0.1803,
      "entropy_step_3": 0.1943,
      "epoch": 0.9476220219505667,
      "grad_norm": 63.627777099609375,
      "learning_rate": 7.490685354115345e-07,
      "loss": 0.05422821044921875,
      "loss_step_1": 0.0473,
      "loss_step_2": 0.0471,
      "loss_step_3": 0.053,
      "step": 31860
    },
    {
      "entropy_step_1": 0.1809,
      "entropy_step_2": 0.181,
      "entropy_step_3": 0.1814,
      "epoch": 0.9479194551024657,
      "grad_norm": 66.42160034179688,
      "learning_rate": 7.406111754002631e-07,
      "loss": 0.05751404762268066,
      "loss_step_1": 0.0636,
      "loss_step_2": 0.0624,
      "loss_step_3": 0.0694,
      "step": 31870
    },
    {
      "entropy_step_1": 0.1734,
      "entropy_step_2": 0.1737,
      "entropy_step_3": 0.1852,
      "epoch": 0.9482168882543648,
      "grad_norm": 115.54087829589844,
      "learning_rate": 7.322014745179461e-07,
      "loss": 0.054847830533981325,
      "loss_step_1": 0.0603,
      "loss_step_2": 0.0602,
      "loss_step_3": 0.0605,
      "step": 31880
    },
    {
      "entropy_step_1": 0.174,
      "entropy_step_2": 0.1751,
      "entropy_step_3": 0.163,
      "epoch": 0.9485143214062639,
      "grad_norm": 134.42105102539062,
      "learning_rate": 7.238394409010973e-07,
      "loss": 0.052763766050338744,
      "loss_step_1": 0.0547,
      "loss_step_2": 0.054,
      "loss_step_3": 0.0605,
      "step": 31890
    },
    {
      "entropy_step_1": 0.1865,
      "entropy_step_2": 0.1868,
      "entropy_step_3": 0.1932,
      "epoch": 0.948811754558163,
      "grad_norm": 108.19246673583984,
      "learning_rate": 7.155250826400783e-07,
      "loss": 0.05342689752578735,
      "loss_step_1": 0.0525,
      "loss_step_2": 0.0521,
      "loss_step_3": 0.0537,
      "step": 31900
    },
    {
      "entropy_step_1": 0.176,
      "entropy_step_2": 0.176,
      "entropy_step_3": 0.1755,
      "epoch": 0.9491091877100621,
      "grad_norm": 212.2714385986328,
      "learning_rate": 7.0725840777916e-07,
      "loss": 0.059696877002716066,
      "loss_step_1": 0.0519,
      "loss_step_2": 0.0515,
      "loss_step_3": 0.0511,
      "step": 31910
    },
    {
      "entropy_step_1": 0.1577,
      "entropy_step_2": 0.1577,
      "entropy_step_3": 0.1632,
      "epoch": 0.9494066208619613,
      "grad_norm": 93.82364654541016,
      "learning_rate": 6.9903942431645e-07,
      "loss": 0.048799258470535276,
      "loss_step_1": 0.0446,
      "loss_step_2": 0.0442,
      "loss_step_3": 0.0455,
      "step": 31920
    },
    {
      "entropy_step_1": 0.1869,
      "entropy_step_2": 0.1881,
      "entropy_step_3": 0.1996,
      "epoch": 0.9497040540138604,
      "grad_norm": 100.1299819946289,
      "learning_rate": 6.908681402039374e-07,
      "loss": 0.052562862634658813,
      "loss_step_1": 0.0533,
      "loss_step_2": 0.053,
      "loss_step_3": 0.0578,
      "step": 31930
    },
    {
      "entropy_step_1": 0.1691,
      "entropy_step_2": 0.1695,
      "entropy_step_3": 0.1815,
      "epoch": 0.9500014871657595,
      "grad_norm": 221.89898681640625,
      "learning_rate": 6.827445633474539e-07,
      "loss": 0.05067140460014343,
      "loss_step_1": 0.0514,
      "loss_step_2": 0.0507,
      "loss_step_3": 0.0553,
      "step": 31940
    },
    {
      "entropy_step_1": 0.1758,
      "entropy_step_2": 0.176,
      "entropy_step_3": 0.1763,
      "epoch": 0.9502989203176586,
      "grad_norm": 77.32289123535156,
      "learning_rate": 6.746687016066566e-07,
      "loss": 0.04989548623561859,
      "loss_step_1": 0.0509,
      "loss_step_2": 0.0506,
      "loss_step_3": 0.0548,
      "step": 31950
    },
    {
      "entropy_step_1": 0.1768,
      "entropy_step_2": 0.1772,
      "entropy_step_3": 0.1866,
      "epoch": 0.9505963534695577,
      "grad_norm": 250.47525024414062,
      "learning_rate": 6.666405627950623e-07,
      "loss": 0.05419592261314392,
      "loss_step_1": 0.0485,
      "loss_step_2": 0.0486,
      "loss_step_3": 0.0492,
      "step": 31960
    },
    {
      "entropy_step_1": 0.1753,
      "entropy_step_2": 0.1751,
      "entropy_step_3": 0.1719,
      "epoch": 0.9508937866214568,
      "grad_norm": 322.4310302734375,
      "learning_rate": 6.586601546799964e-07,
      "loss": 0.05585911273956299,
      "loss_step_1": 0.0578,
      "loss_step_2": 0.0577,
      "loss_step_3": 0.0538,
      "step": 31970
    },
    {
      "entropy_step_1": 0.1762,
      "entropy_step_2": 0.177,
      "entropy_step_3": 0.1582,
      "epoch": 0.951191219773356,
      "grad_norm": 122.58663940429688,
      "learning_rate": 6.507274849826217e-07,
      "loss": 0.06044396162033081,
      "loss_step_1": 0.0589,
      "loss_step_2": 0.0581,
      "loss_step_3": 0.0514,
      "step": 31980
    },
    {
      "entropy_step_1": 0.1626,
      "entropy_step_2": 0.1621,
      "entropy_step_3": 0.1576,
      "epoch": 0.9514886529252551,
      "grad_norm": 55.507347106933594,
      "learning_rate": 6.428425613778988e-07,
      "loss": 0.04683448076248169,
      "loss_step_1": 0.0471,
      "loss_step_2": 0.0468,
      "loss_step_3": 0.045,
      "step": 31990
    },
    {
      "entropy_step_1": 0.1699,
      "entropy_step_2": 0.1698,
      "entropy_step_3": 0.178,
      "epoch": 0.9517860860771542,
      "grad_norm": 55.912254333496094,
      "learning_rate": 6.350053914945975e-07,
      "loss": 0.05246758460998535,
      "loss_step_1": 0.051,
      "loss_step_2": 0.0501,
      "loss_step_3": 0.054,
      "step": 32000
    },
    {
      "epoch": 0.9517860860771542,
      "eval_accuracy": 0.925,
      "eval_f1": 0.9249603040008164,
      "eval_loss": 0.08453197032213211,
      "eval_precision": 0.9259012069539145,
      "eval_recall": 0.925,
      "eval_runtime": 6.4912,
      "eval_samples_per_second": 77.027,
      "eval_steps_per_second": 9.705,
      "step": 32000
    }
  ],
  "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": 2.1467860731849994e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}