diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,6526 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 280,
+  "global_step": 916,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.001091703056768559,
+      "grad_norm": 6.9168360306400665,
+      "learning_rate": 2.2857142857142855e-07,
+      "loss": 0.8069,
+      "step": 1
+    },
+    {
+      "epoch": 0.002183406113537118,
+      "grad_norm": 8.619261383294274,
+      "learning_rate": 4.571428571428571e-07,
+      "loss": 0.918,
+      "step": 2
+    },
+    {
+      "epoch": 0.0032751091703056767,
+      "grad_norm": 7.692605504695248,
+      "learning_rate": 6.857142857142857e-07,
+      "loss": 0.9234,
+      "step": 3
+    },
+    {
+      "epoch": 0.004366812227074236,
+      "grad_norm": 7.122858120605265,
+      "learning_rate": 9.142857142857142e-07,
+      "loss": 0.9265,
+      "step": 4
+    },
+    {
+      "epoch": 0.0054585152838427945,
+      "grad_norm": 7.537326535977078,
+      "learning_rate": 1.1428571428571428e-06,
+      "loss": 0.7695,
+      "step": 5
+    },
+    {
+      "epoch": 0.006550218340611353,
+      "grad_norm": 6.260583368018694,
+      "learning_rate": 1.3714285714285715e-06,
+      "loss": 0.9215,
+      "step": 6
+    },
+    {
+      "epoch": 0.007641921397379912,
+      "grad_norm": 5.203784202592674,
+      "learning_rate": 1.6e-06,
+      "loss": 0.7629,
+      "step": 7
+    },
+    {
+      "epoch": 0.008733624454148471,
+      "grad_norm": 4.233618238431581,
+      "learning_rate": 1.8285714285714284e-06,
+      "loss": 0.8116,
+      "step": 8
+    },
+    {
+      "epoch": 0.009825327510917031,
+      "grad_norm": 3.327308477876956,
+      "learning_rate": 2.057142857142857e-06,
+      "loss": 0.7596,
+      "step": 9
+    },
+    {
+      "epoch": 0.010917030567685589,
+      "grad_norm": 2.9654255350852723,
+      "learning_rate": 2.2857142857142856e-06,
+      "loss": 0.7323,
+      "step": 10
+    },
+    {
+      "epoch": 0.012008733624454149,
+      "grad_norm": 3.4039963472657875,
+      "learning_rate": 2.5142857142857142e-06,
+      "loss": 0.773,
+      "step": 11
+    },
+    {
+      "epoch": 0.013100436681222707,
+      "grad_norm": 2.764952769431326,
+      "learning_rate": 2.742857142857143e-06,
+      "loss": 0.7028,
+      "step": 12
+    },
+    {
+      "epoch": 0.014192139737991267,
+      "grad_norm": 2.645047626519114,
+      "learning_rate": 2.9714285714285716e-06,
+      "loss": 0.6944,
+      "step": 13
+    },
+    {
+      "epoch": 0.015283842794759825,
+      "grad_norm": 2.5003386957924816,
+      "learning_rate": 3.2e-06,
+      "loss": 0.6705,
+      "step": 14
+    },
+    {
+      "epoch": 0.016375545851528384,
+      "grad_norm": 2.2872854432818657,
+      "learning_rate": 3.428571428571428e-06,
+      "loss": 0.6705,
+      "step": 15
+    },
+    {
+      "epoch": 0.017467248908296942,
+      "grad_norm": 1.928277117833659,
+      "learning_rate": 3.657142857142857e-06,
+      "loss": 0.7259,
+      "step": 16
+    },
+    {
+      "epoch": 0.018558951965065504,
+      "grad_norm": 1.8574032707760006,
+      "learning_rate": 3.885714285714286e-06,
+      "loss": 0.6962,
+      "step": 17
+    },
+    {
+      "epoch": 0.019650655021834062,
+      "grad_norm": 1.9310574643116005,
+      "learning_rate": 4.114285714285714e-06,
+      "loss": 0.7249,
+      "step": 18
+    },
+    {
+      "epoch": 0.02074235807860262,
+      "grad_norm": 1.8624427266334949,
+      "learning_rate": 4.342857142857142e-06,
+      "loss": 0.715,
+      "step": 19
+    },
+    {
+      "epoch": 0.021834061135371178,
+      "grad_norm": 1.8660926001733764,
+      "learning_rate": 4.571428571428571e-06,
+      "loss": 0.7817,
+      "step": 20
+    },
+    {
+      "epoch": 0.02292576419213974,
+      "grad_norm": 1.9682589360779612,
+      "learning_rate": 4.8e-06,
+      "loss": 0.5957,
+      "step": 21
+    },
+    {
+      "epoch": 0.024017467248908297,
+      "grad_norm": 1.7481064474330001,
+      "learning_rate": 5.0285714285714285e-06,
+      "loss": 0.6397,
+      "step": 22
+    },
+    {
+      "epoch": 0.025109170305676855,
+      "grad_norm": 1.4998856719874816,
+      "learning_rate": 5.257142857142857e-06,
+      "loss": 0.5489,
+      "step": 23
+    },
+    {
+      "epoch": 0.026200873362445413,
+      "grad_norm": 1.669636786128378,
+      "learning_rate": 5.485714285714286e-06,
+      "loss": 0.6943,
+      "step": 24
+    },
+    {
+      "epoch": 0.027292576419213975,
+      "grad_norm": 1.5957918499269743,
+      "learning_rate": 5.7142857142857145e-06,
+      "loss": 0.6927,
+      "step": 25
+    },
+    {
+      "epoch": 0.028384279475982533,
+      "grad_norm": 1.6133288301808093,
+      "learning_rate": 5.942857142857143e-06,
+      "loss": 0.7117,
+      "step": 26
+    },
+    {
+      "epoch": 0.02947598253275109,
+      "grad_norm": 1.4864207057916419,
+      "learning_rate": 6.171428571428571e-06,
+      "loss": 0.5169,
+      "step": 27
+    },
+    {
+      "epoch": 0.03056768558951965,
+      "grad_norm": 1.5150567145731058,
+      "learning_rate": 6.4e-06,
+      "loss": 0.5613,
+      "step": 28
+    },
+    {
+      "epoch": 0.03165938864628821,
+      "grad_norm": 1.564571555194077,
+      "learning_rate": 6.628571428571428e-06,
+      "loss": 0.6743,
+      "step": 29
+    },
+    {
+      "epoch": 0.03275109170305677,
+      "grad_norm": 1.93767931683908,
+      "learning_rate": 6.857142857142856e-06,
+      "loss": 0.642,
+      "step": 30
+    },
+    {
+      "epoch": 0.03384279475982533,
+      "grad_norm": 1.5789902137704794,
+      "learning_rate": 7.085714285714285e-06,
+      "loss": 0.6701,
+      "step": 31
+    },
+    {
+      "epoch": 0.034934497816593885,
+      "grad_norm": 1.6974375239936534,
+      "learning_rate": 7.314285714285714e-06,
+      "loss": 0.5485,
+      "step": 32
+    },
+    {
+      "epoch": 0.036026200873362446,
+      "grad_norm": 1.3204997268284404,
+      "learning_rate": 7.542857142857142e-06,
+      "loss": 0.5434,
+      "step": 33
+    },
+    {
+      "epoch": 0.03711790393013101,
+      "grad_norm": 1.4597825234641069,
+      "learning_rate": 7.771428571428572e-06,
+      "loss": 0.6605,
+      "step": 34
+    },
+    {
+      "epoch": 0.03820960698689956,
+      "grad_norm": 1.6181689606941454,
+      "learning_rate": 8e-06,
+      "loss": 0.5726,
+      "step": 35
+    },
+    {
+      "epoch": 0.039301310043668124,
+      "grad_norm": 1.637262387439095,
+      "learning_rate": 7.999980926133991e-06,
+      "loss": 0.6037,
+      "step": 36
+    },
+    {
+      "epoch": 0.04039301310043668,
+      "grad_norm": 1.664670294379062,
+      "learning_rate": 7.999923704778513e-06,
+      "loss": 0.7399,
+      "step": 37
+    },
+    {
+      "epoch": 0.04148471615720524,
+      "grad_norm": 1.7493240614718077,
+      "learning_rate": 7.99982833666118e-06,
+      "loss": 0.6127,
+      "step": 38
+    },
+    {
+      "epoch": 0.0425764192139738,
+      "grad_norm": 2.092392677058526,
+      "learning_rate": 7.999694822994692e-06,
+      "loss": 0.668,
+      "step": 39
+    },
+    {
+      "epoch": 0.043668122270742356,
+      "grad_norm": 1.7013397213106045,
+      "learning_rate": 7.999523165476792e-06,
+      "loss": 0.6856,
+      "step": 40
+    },
+    {
+      "epoch": 0.04475982532751092,
+      "grad_norm": 1.474118425250071,
+      "learning_rate": 7.999313366290262e-06,
+      "loss": 0.5802,
+      "step": 41
+    },
+    {
+      "epoch": 0.04585152838427948,
+      "grad_norm": 1.660443265060205,
+      "learning_rate": 7.99906542810289e-06,
+      "loss": 0.6324,
+      "step": 42
+    },
+    {
+      "epoch": 0.04694323144104803,
+      "grad_norm": 1.6459243125827774,
+      "learning_rate": 7.998779354067437e-06,
+      "loss": 0.6614,
+      "step": 43
+    },
+    {
+      "epoch": 0.048034934497816595,
+      "grad_norm": 1.5196086504507762,
+      "learning_rate": 7.998455147821595e-06,
+      "loss": 0.6091,
+      "step": 44
+    },
+    {
+      "epoch": 0.04912663755458515,
+      "grad_norm": 1.429090294880644,
+      "learning_rate": 7.99809281348794e-06,
+      "loss": 0.6485,
+      "step": 45
+    },
+    {
+      "epoch": 0.05021834061135371,
+      "grad_norm": 1.3827763726996563,
+      "learning_rate": 7.997692355673884e-06,
+      "loss": 0.5785,
+      "step": 46
+    },
+    {
+      "epoch": 0.05131004366812227,
+      "grad_norm": 1.7588449103100179,
+      "learning_rate": 7.997253779471614e-06,
+      "loss": 0.6473,
+      "step": 47
+    },
+    {
+      "epoch": 0.05240174672489083,
+      "grad_norm": 1.5172171755749084,
+      "learning_rate": 7.996777090458023e-06,
+      "loss": 0.581,
+      "step": 48
+    },
+    {
+      "epoch": 0.05349344978165939,
+      "grad_norm": 1.6076648219406835,
+      "learning_rate": 7.996262294694649e-06,
+      "loss": 0.5681,
+      "step": 49
+    },
+    {
+      "epoch": 0.05458515283842795,
+      "grad_norm": 1.4244628514708157,
+      "learning_rate": 7.995709398727587e-06,
+      "loss": 0.4754,
+      "step": 50
+    },
+    {
+      "epoch": 0.055676855895196505,
+      "grad_norm": 1.2927489245734367,
+      "learning_rate": 7.995118409587412e-06,
+      "loss": 0.6537,
+      "step": 51
+    },
+    {
+      "epoch": 0.056768558951965066,
+      "grad_norm": 1.6748840284841289,
+      "learning_rate": 7.99448933478909e-06,
+      "loss": 0.5275,
+      "step": 52
+    },
+    {
+      "epoch": 0.05786026200873363,
+      "grad_norm": 1.5128577258259677,
+      "learning_rate": 7.993822182331882e-06,
+      "loss": 0.6424,
+      "step": 53
+    },
+    {
+      "epoch": 0.05895196506550218,
+      "grad_norm": 1.3310295240779713,
+      "learning_rate": 7.993116960699236e-06,
+      "loss": 0.6035,
+      "step": 54
+    },
+    {
+      "epoch": 0.060043668122270744,
+      "grad_norm": 1.6799361475060814,
+      "learning_rate": 7.99237367885869e-06,
+      "loss": 0.6795,
+      "step": 55
+    },
+    {
+      "epoch": 0.0611353711790393,
+      "grad_norm": 1.546417812619001,
+      "learning_rate": 7.991592346261747e-06,
+      "loss": 0.5519,
+      "step": 56
+    },
+    {
+      "epoch": 0.06222707423580786,
+      "grad_norm": 1.5066644174359114,
+      "learning_rate": 7.990772972843764e-06,
+      "loss": 0.6043,
+      "step": 57
+    },
+    {
+      "epoch": 0.06331877729257641,
+      "grad_norm": 1.489274779760962,
+      "learning_rate": 7.989915569023819e-06,
+      "loss": 0.6095,
+      "step": 58
+    },
+    {
+      "epoch": 0.06441048034934498,
+      "grad_norm": 1.7091568832403692,
+      "learning_rate": 7.989020145704584e-06,
+      "loss": 0.7029,
+      "step": 59
+    },
+    {
+      "epoch": 0.06550218340611354,
+      "grad_norm": 1.3568973285059558,
+      "learning_rate": 7.98808671427218e-06,
+      "loss": 0.6448,
+      "step": 60
+    },
+    {
+      "epoch": 0.0665938864628821,
+      "grad_norm": 1.4801827231358118,
+      "learning_rate": 7.987115286596038e-06,
+      "loss": 0.6638,
+      "step": 61
+    },
+    {
+      "epoch": 0.06768558951965066,
+      "grad_norm": 1.600407108115019,
+      "learning_rate": 7.986105875028748e-06,
+      "loss": 0.6321,
+      "step": 62
+    },
+    {
+      "epoch": 0.06877729257641921,
+      "grad_norm": 1.5576366146182474,
+      "learning_rate": 7.985058492405896e-06,
+      "loss": 0.7121,
+      "step": 63
+    },
+    {
+      "epoch": 0.06986899563318777,
+      "grad_norm": 1.6147036677777984,
+      "learning_rate": 7.983973152045903e-06,
+      "loss": 0.6388,
+      "step": 64
+    },
+    {
+      "epoch": 0.07096069868995633,
+      "grad_norm": 1.7851503357701477,
+      "learning_rate": 7.982849867749865e-06,
+      "loss": 0.6858,
+      "step": 65
+    },
+    {
+      "epoch": 0.07205240174672489,
+      "grad_norm": 1.2859661673668508,
+      "learning_rate": 7.981688653801361e-06,
+      "loss": 0.5384,
+      "step": 66
+    },
+    {
+      "epoch": 0.07314410480349345,
+      "grad_norm": 1.60696565837722,
+      "learning_rate": 7.980489524966288e-06,
+      "loss": 0.5506,
+      "step": 67
+    },
+    {
+      "epoch": 0.07423580786026202,
+      "grad_norm": 1.5405282936232578,
+      "learning_rate": 7.979252496492655e-06,
+      "loss": 0.4455,
+      "step": 68
+    },
+    {
+      "epoch": 0.07532751091703056,
+      "grad_norm": 1.7009157245583173,
+      "learning_rate": 7.977977584110411e-06,
+      "loss": 0.5581,
+      "step": 69
+    },
+    {
+      "epoch": 0.07641921397379912,
+      "grad_norm": 1.4022427304469514,
+      "learning_rate": 7.976664804031223e-06,
+      "loss": 0.6153,
+      "step": 70
+    },
+    {
+      "epoch": 0.07751091703056769,
+      "grad_norm": 1.5035374403149062,
+      "learning_rate": 7.975314172948291e-06,
+      "loss": 0.684,
+      "step": 71
+    },
+    {
+      "epoch": 0.07860262008733625,
+      "grad_norm": 1.4093854673180422,
+      "learning_rate": 7.973925708036115e-06,
+      "loss": 0.7426,
+      "step": 72
+    },
+    {
+      "epoch": 0.07969432314410481,
+      "grad_norm": 1.6851458145111584,
+      "learning_rate": 7.972499426950292e-06,
+      "loss": 0.5668,
+      "step": 73
+    },
+    {
+      "epoch": 0.08078602620087336,
+      "grad_norm": 1.529050078017108,
+      "learning_rate": 7.971035347827286e-06,
+      "loss": 0.5895,
+      "step": 74
+    },
+    {
+      "epoch": 0.08187772925764192,
+      "grad_norm": 1.6942196136539975,
+      "learning_rate": 7.969533489284195e-06,
+      "loss": 0.7048,
+      "step": 75
+    },
+    {
+      "epoch": 0.08296943231441048,
+      "grad_norm": 1.4777780150565798,
+      "learning_rate": 7.967993870418517e-06,
+      "loss": 0.7128,
+      "step": 76
+    },
+    {
+      "epoch": 0.08406113537117904,
+      "grad_norm": 1.680209877486954,
+      "learning_rate": 7.96641651080791e-06,
+      "loss": 0.4864,
+      "step": 77
+    },
+    {
+      "epoch": 0.0851528384279476,
+      "grad_norm": 1.42998101662559,
+      "learning_rate": 7.964801430509938e-06,
+      "loss": 0.5904,
+      "step": 78
+    },
+    {
+      "epoch": 0.08624454148471616,
+      "grad_norm": 1.3659967634997836,
+      "learning_rate": 7.963148650061817e-06,
+      "loss": 0.5854,
+      "step": 79
+    },
+    {
+      "epoch": 0.08733624454148471,
+      "grad_norm": 1.5702401689586527,
+      "learning_rate": 7.961458190480154e-06,
+      "loss": 0.6448,
+      "step": 80
+    },
+    {
+      "epoch": 0.08842794759825327,
+      "grad_norm": 1.6021520859056189,
+      "learning_rate": 7.959730073260684e-06,
+      "loss": 0.62,
+      "step": 81
+    },
+    {
+      "epoch": 0.08951965065502183,
+      "grad_norm": 1.3559678768559176,
+      "learning_rate": 7.957964320377992e-06,
+      "loss": 0.5855,
+      "step": 82
+    },
+    {
+      "epoch": 0.0906113537117904,
+      "grad_norm": 1.5008897257417335,
+      "learning_rate": 7.956160954285234e-06,
+      "loss": 0.5917,
+      "step": 83
+    },
+    {
+      "epoch": 0.09170305676855896,
+      "grad_norm": 1.3325553922283428,
+      "learning_rate": 7.954319997913848e-06,
+      "loss": 0.5747,
+      "step": 84
+    },
+    {
+      "epoch": 0.0927947598253275,
+      "grad_norm": 1.6334929695605136,
+      "learning_rate": 7.952441474673275e-06,
+      "loss": 0.6271,
+      "step": 85
+    },
+    {
+      "epoch": 0.09388646288209607,
+      "grad_norm": 1.5780752995999416,
+      "learning_rate": 7.950525408450647e-06,
+      "loss": 0.6098,
+      "step": 86
+    },
+    {
+      "epoch": 0.09497816593886463,
+      "grad_norm": 1.5582598019357035,
+      "learning_rate": 7.948571823610492e-06,
+      "loss": 0.5995,
+      "step": 87
+    },
+    {
+      "epoch": 0.09606986899563319,
+      "grad_norm": 1.349598296171142,
+      "learning_rate": 7.946580744994418e-06,
+      "loss": 0.6231,
+      "step": 88
+    },
+    {
+      "epoch": 0.09716157205240175,
+      "grad_norm": 1.4067679148676697,
+      "learning_rate": 7.944552197920806e-06,
+      "loss": 0.7023,
+      "step": 89
+    },
+    {
+      "epoch": 0.0982532751091703,
+      "grad_norm": 1.489512752201376,
+      "learning_rate": 7.942486208184478e-06,
+      "loss": 0.6443,
+      "step": 90
+    },
+    {
+      "epoch": 0.09934497816593886,
+      "grad_norm": 1.422726492645584,
+      "learning_rate": 7.940382802056374e-06,
+      "loss": 0.6031,
+      "step": 91
+    },
+    {
+      "epoch": 0.10043668122270742,
+      "grad_norm": 1.55466766394215,
+      "learning_rate": 7.938242006283219e-06,
+      "loss": 0.5998,
+      "step": 92
+    },
+    {
+      "epoch": 0.10152838427947598,
+      "grad_norm": 1.6313884014991245,
+      "learning_rate": 7.936063848087182e-06,
+      "loss": 0.5822,
+      "step": 93
+    },
+    {
+      "epoch": 0.10262008733624454,
+      "grad_norm": 1.3837342347132842,
+      "learning_rate": 7.933848355165526e-06,
+      "loss": 0.6168,
+      "step": 94
+    },
+    {
+      "epoch": 0.1037117903930131,
+      "grad_norm": 1.3175406249386525,
+      "learning_rate": 7.931595555690261e-06,
+      "loss": 0.5629,
+      "step": 95
+    },
+    {
+      "epoch": 0.10480349344978165,
+      "grad_norm": 1.4980147704457836,
+      "learning_rate": 7.929305478307787e-06,
+      "loss": 0.6575,
+      "step": 96
+    },
+    {
+      "epoch": 0.10589519650655022,
+      "grad_norm": 1.4934490360354107,
+      "learning_rate": 7.92697815213852e-06,
+      "loss": 0.6395,
+      "step": 97
+    },
+    {
+      "epoch": 0.10698689956331878,
+      "grad_norm": 1.6143096306904618,
+      "learning_rate": 7.924613606776535e-06,
+      "loss": 0.6362,
+      "step": 98
+    },
+    {
+      "epoch": 0.10807860262008734,
+      "grad_norm": 1.5507045899750025,
+      "learning_rate": 7.922211872289176e-06,
+      "loss": 0.6258,
+      "step": 99
+    },
+    {
+      "epoch": 0.1091703056768559,
+      "grad_norm": 1.7417116462175397,
+      "learning_rate": 7.919772979216687e-06,
+      "loss": 0.578,
+      "step": 100
+    },
+    {
+      "epoch": 0.11026200873362445,
+      "grad_norm": 1.5564506490699812,
+      "learning_rate": 7.917296958571813e-06,
+      "loss": 0.5899,
+      "step": 101
+    },
+    {
+      "epoch": 0.11135371179039301,
+      "grad_norm": 1.51376629823034,
+      "learning_rate": 7.914783841839412e-06,
+      "loss": 0.546,
+      "step": 102
+    },
+    {
+      "epoch": 0.11244541484716157,
+      "grad_norm": 1.5871820189984336,
+      "learning_rate": 7.91223366097605e-06,
+      "loss": 0.7067,
+      "step": 103
+    },
+    {
+      "epoch": 0.11353711790393013,
+      "grad_norm": 1.684703978601094,
+      "learning_rate": 7.909646448409604e-06,
+      "loss": 0.5795,
+      "step": 104
+    },
+    {
+      "epoch": 0.1146288209606987,
+      "grad_norm": 1.504074132778666,
+      "learning_rate": 7.90702223703883e-06,
+      "loss": 0.6801,
+      "step": 105
+    },
+    {
+      "epoch": 0.11572052401746726,
+      "grad_norm": 1.3449630071756726,
+      "learning_rate": 7.904361060232974e-06,
+      "loss": 0.5508,
+      "step": 106
+    },
+    {
+      "epoch": 0.1168122270742358,
+      "grad_norm": 1.397180371022227,
+      "learning_rate": 7.901662951831316e-06,
+      "loss": 0.6593,
+      "step": 107
+    },
+    {
+      "epoch": 0.11790393013100436,
+      "grad_norm": 1.3698872191395284,
+      "learning_rate": 7.898927946142763e-06,
+      "loss": 0.6885,
+      "step": 108
+    },
+    {
+      "epoch": 0.11899563318777293,
+      "grad_norm": 1.5829183415829775,
+      "learning_rate": 7.896156077945407e-06,
+      "loss": 0.5311,
+      "step": 109
+    },
+    {
+      "epoch": 0.12008733624454149,
+      "grad_norm": 1.451977799352687,
+      "learning_rate": 7.893347382486072e-06,
+      "loss": 0.6202,
+      "step": 110
+    },
+    {
+      "epoch": 0.12117903930131005,
+      "grad_norm": 1.3994680718242534,
+      "learning_rate": 7.890501895479879e-06,
+      "loss": 0.6208,
+      "step": 111
+    },
+    {
+      "epoch": 0.1222707423580786,
+      "grad_norm": 1.6636406803513806,
+      "learning_rate": 7.887619653109788e-06,
+      "loss": 0.6677,
+      "step": 112
+    },
+    {
+      "epoch": 0.12336244541484716,
+      "grad_norm": 1.4532843659222607,
+      "learning_rate": 7.884700692026134e-06,
+      "loss": 0.5871,
+      "step": 113
+    },
+    {
+      "epoch": 0.12445414847161572,
+      "grad_norm": 1.4662200934134768,
+      "learning_rate": 7.881745049346169e-06,
+      "loss": 0.5987,
+      "step": 114
+    },
+    {
+      "epoch": 0.12554585152838427,
+      "grad_norm": 1.475099944420446,
+      "learning_rate": 7.878752762653575e-06,
+      "loss": 0.647,
+      "step": 115
+    },
+    {
+      "epoch": 0.12663755458515283,
+      "grad_norm": 1.5380967311487281,
+      "learning_rate": 7.875723869998009e-06,
+      "loss": 0.7604,
+      "step": 116
+    },
+    {
+      "epoch": 0.1277292576419214,
+      "grad_norm": 1.571718225913752,
+      "learning_rate": 7.872658409894593e-06,
+      "loss": 0.6236,
+      "step": 117
+    },
+    {
+      "epoch": 0.12882096069868995,
+      "grad_norm": 1.4096634500954293,
+      "learning_rate": 7.86955642132345e-06,
+      "loss": 0.6519,
+      "step": 118
+    },
+    {
+      "epoch": 0.1299126637554585,
+      "grad_norm": 1.4944998170948933,
+      "learning_rate": 7.866417943729186e-06,
+      "loss": 0.619,
+      "step": 119
+    },
+    {
+      "epoch": 0.13100436681222707,
+      "grad_norm": 1.5310843619532999,
+      "learning_rate": 7.863243017020403e-06,
+      "loss": 0.625,
+      "step": 120
+    },
+    {
+      "epoch": 0.13209606986899564,
+      "grad_norm": 1.441967778632453,
+      "learning_rate": 7.860031681569183e-06,
+      "loss": 0.6129,
+      "step": 121
+    },
+    {
+      "epoch": 0.1331877729257642,
+      "grad_norm": 1.4984295970887447,
+      "learning_rate": 7.856783978210586e-06,
+      "loss": 0.6419,
+      "step": 122
+    },
+    {
+      "epoch": 0.13427947598253276,
+      "grad_norm": 1.3778311870438549,
+      "learning_rate": 7.853499948242111e-06,
+      "loss": 0.6067,
+      "step": 123
+    },
+    {
+      "epoch": 0.13537117903930132,
+      "grad_norm": 1.5135481075273571,
+      "learning_rate": 7.850179633423195e-06,
+      "loss": 0.6112,
+      "step": 124
+    },
+    {
+      "epoch": 0.13646288209606988,
+      "grad_norm": 1.3279833242854147,
+      "learning_rate": 7.846823075974663e-06,
+      "loss": 0.5415,
+      "step": 125
+    },
+    {
+      "epoch": 0.13755458515283842,
+      "grad_norm": 1.4776236100804023,
+      "learning_rate": 7.843430318578198e-06,
+      "loss": 0.5892,
+      "step": 126
+    },
+    {
+      "epoch": 0.13864628820960698,
+      "grad_norm": 1.3091913650405782,
+      "learning_rate": 7.840001404375803e-06,
+      "loss": 0.6429,
+      "step": 127
+    },
+    {
+      "epoch": 0.13973799126637554,
+      "grad_norm": 1.2931286140842404,
+      "learning_rate": 7.836536376969242e-06,
+      "loss": 0.4929,
+      "step": 128
+    },
+    {
+      "epoch": 0.1408296943231441,
+      "grad_norm": 1.3706911268954516,
+      "learning_rate": 7.833035280419495e-06,
+      "loss": 0.7081,
+      "step": 129
+    },
+    {
+      "epoch": 0.14192139737991266,
+      "grad_norm": 1.4028173055022224,
+      "learning_rate": 7.829498159246193e-06,
+      "loss": 0.5754,
+      "step": 130
+    },
+    {
+      "epoch": 0.14301310043668122,
+      "grad_norm": 1.5579836702697056,
+      "learning_rate": 7.825925058427054e-06,
+      "loss": 0.5518,
+      "step": 131
+    },
+    {
+      "epoch": 0.14410480349344978,
+      "grad_norm": 1.3789925803317493,
+      "learning_rate": 7.822316023397308e-06,
+      "loss": 0.67,
+      "step": 132
+    },
+    {
+      "epoch": 0.14519650655021835,
+      "grad_norm": 1.6438796790386405,
+      "learning_rate": 7.81867110004912e-06,
+      "loss": 0.6262,
+      "step": 133
+    },
+    {
+      "epoch": 0.1462882096069869,
+      "grad_norm": 1.5932532515551787,
+      "learning_rate": 7.814990334731015e-06,
+      "loss": 0.5472,
+      "step": 134
+    },
+    {
+      "epoch": 0.14737991266375547,
+      "grad_norm": 1.4218597455087487,
+      "learning_rate": 7.811273774247271e-06,
+      "loss": 0.6212,
+      "step": 135
+    },
+    {
+      "epoch": 0.14847161572052403,
+      "grad_norm": 1.4979752973234424,
+      "learning_rate": 7.80752146585734e-06,
+      "loss": 0.5449,
+      "step": 136
+    },
+    {
+      "epoch": 0.14956331877729256,
+      "grad_norm": 1.5503434977081532,
+      "learning_rate": 7.803733457275241e-06,
+      "loss": 0.5627,
+      "step": 137
+    },
+    {
+      "epoch": 0.15065502183406113,
+      "grad_norm": 1.3385399338417763,
+      "learning_rate": 7.799909796668952e-06,
+      "loss": 0.6285,
+      "step": 138
+    },
+    {
+      "epoch": 0.1517467248908297,
+      "grad_norm": 1.6086857843123978,
+      "learning_rate": 7.796050532659801e-06,
+      "loss": 0.6201,
+      "step": 139
+    },
+    {
+      "epoch": 0.15283842794759825,
+      "grad_norm": 1.5042189043053371,
+      "learning_rate": 7.792155714321844e-06,
+      "loss": 0.5448,
+      "step": 140
+    },
+    {
+      "epoch": 0.1539301310043668,
+      "grad_norm": 1.496614591995746,
+      "learning_rate": 7.788225391181244e-06,
+      "loss": 0.5677,
+      "step": 141
+    },
+    {
+      "epoch": 0.15502183406113537,
+      "grad_norm": 1.3673478795664078,
+      "learning_rate": 7.784259613215637e-06,
+      "loss": 0.4902,
+      "step": 142
+    },
+    {
+      "epoch": 0.15611353711790393,
+      "grad_norm": 1.5463971091149502,
+      "learning_rate": 7.780258430853501e-06,
+      "loss": 0.6009,
+      "step": 143
+    },
+    {
+      "epoch": 0.1572052401746725,
+      "grad_norm": 1.528859261250924,
+      "learning_rate": 7.776221894973517e-06,
+      "loss": 0.6209,
+      "step": 144
+    },
+    {
+      "epoch": 0.15829694323144106,
+      "grad_norm": 1.427566766037643,
+      "learning_rate": 7.77215005690391e-06,
+      "loss": 0.6433,
+      "step": 145
+    },
+    {
+      "epoch": 0.15938864628820962,
+      "grad_norm": 1.5336258102759508,
+      "learning_rate": 7.768042968421814e-06,
+      "loss": 0.6244,
+      "step": 146
+    },
+    {
+      "epoch": 0.16048034934497818,
+      "grad_norm": 1.4516187807428587,
+      "learning_rate": 7.763900681752594e-06,
+      "loss": 0.4526,
+      "step": 147
+    },
+    {
+      "epoch": 0.1615720524017467,
+      "grad_norm": 1.2426578447642056,
+      "learning_rate": 7.759723249569202e-06,
+      "loss": 0.5119,
+      "step": 148
+    },
+    {
+      "epoch": 0.16266375545851527,
+      "grad_norm": 1.3690342156175288,
+      "learning_rate": 7.755510724991491e-06,
+      "loss": 0.596,
+      "step": 149
+    },
+    {
+      "epoch": 0.16375545851528384,
+      "grad_norm": 1.4398744824849907,
+      "learning_rate": 7.751263161585547e-06,
+      "loss": 0.5291,
+      "step": 150
+    },
+    {
+      "epoch": 0.1648471615720524,
+      "grad_norm": 1.4736827868329099,
+      "learning_rate": 7.746980613363004e-06,
+      "loss": 0.5622,
+      "step": 151
+    },
+    {
+      "epoch": 0.16593886462882096,
+      "grad_norm": 1.1576108683772315,
+      "learning_rate": 7.742663134780369e-06,
+      "loss": 0.4897,
+      "step": 152
+    },
+    {
+      "epoch": 0.16703056768558952,
+      "grad_norm": 1.5090960302871992,
+      "learning_rate": 7.738310780738307e-06,
+      "loss": 0.712,
+      "step": 153
+    },
+    {
+      "epoch": 0.16812227074235808,
+      "grad_norm": 1.3964072666308516,
+      "learning_rate": 7.733923606580966e-06,
+      "loss": 0.6472,
+      "step": 154
+    },
+    {
+      "epoch": 0.16921397379912664,
+      "grad_norm": 1.6059506328673856,
+      "learning_rate": 7.729501668095264e-06,
+      "loss": 0.6786,
+      "step": 155
+    },
+    {
+      "epoch": 0.1703056768558952,
+      "grad_norm": 1.3441849797912846,
+      "learning_rate": 7.72504502151017e-06,
+      "loss": 0.5136,
+      "step": 156
+    },
+    {
+      "epoch": 0.17139737991266377,
+      "grad_norm": 1.636706525867172,
+      "learning_rate": 7.72055372349601e-06,
+      "loss": 0.6795,
+      "step": 157
+    },
+    {
+      "epoch": 0.17248908296943233,
+      "grad_norm": 1.4433457862208092,
+      "learning_rate": 7.716027831163723e-06,
+      "loss": 0.5496,
+      "step": 158
+    },
+    {
+      "epoch": 0.17358078602620086,
+      "grad_norm": 1.4594871883636047,
+      "learning_rate": 7.711467402064155e-06,
+      "loss": 0.5887,
+      "step": 159
+    },
+    {
+      "epoch": 0.17467248908296942,
+      "grad_norm": 1.3298172020543817,
+      "learning_rate": 7.706872494187313e-06,
+      "loss": 0.4975,
+      "step": 160
+    },
+    {
+      "epoch": 0.17576419213973798,
+      "grad_norm": 1.4792366232102332,
+      "learning_rate": 7.702243165961635e-06,
+      "loss": 0.5748,
+      "step": 161
+    },
+    {
+      "epoch": 0.17685589519650655,
+      "grad_norm": 1.1963230765110093,
+      "learning_rate": 7.697579476253248e-06,
+      "loss": 0.5333,
+      "step": 162
+    },
+    {
+      "epoch": 0.1779475982532751,
+      "grad_norm": 1.3100993334968998,
+      "learning_rate": 7.69288148436521e-06,
+      "loss": 0.529,
+      "step": 163
+    },
+    {
+      "epoch": 0.17903930131004367,
+      "grad_norm": 1.4299716683505044,
+      "learning_rate": 7.68814925003677e-06,
+      "loss": 0.4922,
+      "step": 164
+    },
+    {
+      "epoch": 0.18013100436681223,
+      "grad_norm": 1.2819297147739053,
+      "learning_rate": 7.683382833442594e-06,
+      "loss": 0.4941,
+      "step": 165
+    },
+    {
+      "epoch": 0.1812227074235808,
+      "grad_norm": 1.383396362069611,
+      "learning_rate": 7.678582295192009e-06,
+      "loss": 0.519,
+      "step": 166
+    },
+    {
+      "epoch": 0.18231441048034935,
+      "grad_norm": 1.5772925336837018,
+      "learning_rate": 7.673747696328235e-06,
+      "loss": 0.7051,
+      "step": 167
+    },
+    {
+      "epoch": 0.18340611353711792,
+      "grad_norm": 1.5305700874690167,
+      "learning_rate": 7.668879098327596e-06,
+      "loss": 0.5226,
+      "step": 168
+    },
+    {
+      "epoch": 0.18449781659388648,
+      "grad_norm": 1.7043647886602058,
+      "learning_rate": 7.663976563098748e-06,
+      "loss": 0.5749,
+      "step": 169
+    },
+    {
+      "epoch": 0.185589519650655,
+      "grad_norm": 1.4744519506752671,
+      "learning_rate": 7.659040152981894e-06,
+      "loss": 0.6675,
+      "step": 170
+    },
+    {
+      "epoch": 0.18668122270742357,
+      "grad_norm": 1.4080879804081063,
+      "learning_rate": 7.654069930747982e-06,
+      "loss": 0.5026,
+      "step": 171
+    },
+    {
+      "epoch": 0.18777292576419213,
+      "grad_norm": 1.3184115543301183,
+      "learning_rate": 7.649065959597916e-06,
+      "loss": 0.5183,
+      "step": 172
+    },
+    {
+      "epoch": 0.1888646288209607,
+      "grad_norm": 1.6589387141904088,
+      "learning_rate": 7.644028303161743e-06,
+      "loss": 0.5887,
+      "step": 173
+    },
+    {
+      "epoch": 0.18995633187772926,
+      "grad_norm": 1.5321842642074932,
+      "learning_rate": 7.638957025497857e-06,
+      "loss": 0.6049,
+      "step": 174
+    },
+    {
+      "epoch": 0.19104803493449782,
+      "grad_norm": 1.5678479812006205,
+      "learning_rate": 7.633852191092166e-06,
+      "loss": 0.5664,
+      "step": 175
+    },
+    {
+      "epoch": 0.19213973799126638,
+      "grad_norm": 1.416930808688244,
+      "learning_rate": 7.6287138648572946e-06,
+      "loss": 0.5525,
+      "step": 176
+    },
+    {
+      "epoch": 0.19323144104803494,
+      "grad_norm": 1.335762536841671,
+      "learning_rate": 7.623542112131734e-06,
+      "loss": 0.5139,
+      "step": 177
+    },
+    {
+      "epoch": 0.1943231441048035,
+      "grad_norm": 1.473220842403004,
+      "learning_rate": 7.618336998679034e-06,
+      "loss": 0.6093,
+      "step": 178
+    },
+    {
+      "epoch": 0.19541484716157206,
+      "grad_norm": 1.4576307884390227,
+      "learning_rate": 7.6130985906869515e-06,
+      "loss": 0.5685,
+      "step": 179
+    },
+    {
+      "epoch": 0.1965065502183406,
+      "grad_norm": 1.4673412599144768,
+      "learning_rate": 7.6078269547666135e-06,
+      "loss": 0.5365,
+      "step": 180
+    },
+    {
+      "epoch": 0.19759825327510916,
+      "grad_norm": 1.5183161060171537,
+      "learning_rate": 7.602522157951672e-06,
+      "loss": 0.521,
+      "step": 181
+    },
+    {
+      "epoch": 0.19868995633187772,
+      "grad_norm": 1.4768425270547154,
+      "learning_rate": 7.597184267697449e-06,
+      "loss": 0.5472,
+      "step": 182
+    },
+    {
+      "epoch": 0.19978165938864628,
+      "grad_norm": 1.5653932400158284,
+      "learning_rate": 7.591813351880081e-06,
+      "loss": 0.5521,
+      "step": 183
+    },
+    {
+      "epoch": 0.20087336244541484,
+      "grad_norm": 1.739179296646375,
+      "learning_rate": 7.586409478795654e-06,
+      "loss": 0.5815,
+      "step": 184
+    },
+    {
+      "epoch": 0.2019650655021834,
+      "grad_norm": 1.8305945169666482,
+      "learning_rate": 7.580972717159335e-06,
+      "loss": 0.5967,
+      "step": 185
+    },
+    {
+      "epoch": 0.20305676855895197,
+      "grad_norm": 1.5537911239908309,
+      "learning_rate": 7.5755031361044985e-06,
+      "loss": 0.5975,
+      "step": 186
+    },
+    {
+      "epoch": 0.20414847161572053,
+      "grad_norm": 1.4213609050860452,
+      "learning_rate": 7.57000080518185e-06,
+      "loss": 0.4776,
+      "step": 187
+    },
+    {
+      "epoch": 0.2052401746724891,
+      "grad_norm": 1.4789155471626043,
+      "learning_rate": 7.564465794358536e-06,
+      "loss": 0.6826,
+      "step": 188
+    },
+    {
+      "epoch": 0.20633187772925765,
+      "grad_norm": 1.566883946086422,
+      "learning_rate": 7.558898174017262e-06,
+      "loss": 0.5313,
+      "step": 189
+    },
+    {
+      "epoch": 0.2074235807860262,
+      "grad_norm": 1.7724230033266728,
+      "learning_rate": 7.55329801495539e-06,
+      "loss": 0.5543,
+      "step": 190
+    },
+    {
+      "epoch": 0.20851528384279475,
+      "grad_norm": 1.494585694856317,
+      "learning_rate": 7.547665388384041e-06,
+      "loss": 0.5714,
+      "step": 191
+    },
+    {
+      "epoch": 0.2096069868995633,
+      "grad_norm": 1.6571585826088926,
+      "learning_rate": 7.542000365927194e-06,
+      "loss": 0.5344,
+      "step": 192
+    },
+    {
+      "epoch": 0.21069868995633187,
+      "grad_norm": 1.37196223933786,
+      "learning_rate": 7.536303019620765e-06,
+      "loss": 0.6217,
+      "step": 193
+    },
+    {
+      "epoch": 0.21179039301310043,
+      "grad_norm": 1.4468057089753752,
+      "learning_rate": 7.530573421911705e-06,
+      "loss": 0.5434,
+      "step": 194
+    },
+    {
+      "epoch": 0.212882096069869,
+      "grad_norm": 1.5177202673287582,
+      "learning_rate": 7.524811645657063e-06,
+      "loss": 0.6779,
+      "step": 195
+    },
+    {
+      "epoch": 0.21397379912663755,
+      "grad_norm": 1.5725281443663255,
+      "learning_rate": 7.519017764123074e-06,
+      "loss": 0.6112,
+      "step": 196
+    },
+    {
+      "epoch": 0.21506550218340612,
+      "grad_norm": 1.575239994763357,
+      "learning_rate": 7.513191850984215e-06,
+      "loss": 0.5418,
+      "step": 197
+    },
+    {
+      "epoch": 0.21615720524017468,
+      "grad_norm": 1.3004073921667239,
+      "learning_rate": 7.507333980322279e-06,
+      "loss": 0.5046,
+      "step": 198
+    },
+    {
+      "epoch": 0.21724890829694324,
+      "grad_norm": 1.332980149587294,
+      "learning_rate": 7.501444226625426e-06,
+      "loss": 0.5145,
+      "step": 199
+    },
+    {
+      "epoch": 0.2183406113537118,
+      "grad_norm": 1.4395725224087588,
+      "learning_rate": 7.495522664787237e-06,
+      "loss": 0.5503,
+      "step": 200
+    },
+    {
+      "epoch": 0.21943231441048036,
+      "grad_norm": 1.5415342261120932,
+      "learning_rate": 7.4895693701057655e-06,
+      "loss": 0.5744,
+      "step": 201
+    },
+    {
+      "epoch": 0.2205240174672489,
+      "grad_norm": 1.4676635626586152,
+      "learning_rate": 7.4835844182825726e-06,
+      "loss": 0.5928,
+      "step": 202
+    },
+    {
+      "epoch": 0.22161572052401746,
+      "grad_norm": 1.5559087687931528,
+      "learning_rate": 7.477567885421771e-06,
+      "loss": 0.5411,
+      "step": 203
+    },
+    {
+      "epoch": 0.22270742358078602,
+      "grad_norm": 1.296850939133172,
+      "learning_rate": 7.471519848029057e-06,
+      "loss": 0.5444,
+      "step": 204
+    },
+    {
+      "epoch": 0.22379912663755458,
+      "grad_norm": 1.4777489551219587,
+      "learning_rate": 7.465440383010733e-06,
+      "loss": 0.565,
+      "step": 205
+    },
+    {
+      "epoch": 0.22489082969432314,
+      "grad_norm": 1.42155992331424,
+      "learning_rate": 7.459329567672732e-06,
+      "loss": 0.654,
+      "step": 206
+    },
+    {
+      "epoch": 0.2259825327510917,
+      "grad_norm": 1.4070072461083232,
+      "learning_rate": 7.453187479719637e-06,
+      "loss": 0.5257,
+      "step": 207
+    },
+    {
+      "epoch": 0.22707423580786026,
+      "grad_norm": 1.4076749777799764,
+      "learning_rate": 7.4470141972536895e-06,
+      "loss": 0.6039,
+      "step": 208
+    },
+    {
+      "epoch": 0.22816593886462883,
+      "grad_norm": 1.3129034828037942,
+      "learning_rate": 7.440809798773798e-06,
+      "loss": 0.5754,
+      "step": 209
+    },
+    {
+      "epoch": 0.2292576419213974,
+      "grad_norm": 1.4692678959747856,
+      "learning_rate": 7.43457436317454e-06,
+      "loss": 0.6623,
+      "step": 210
+    },
+    {
+      "epoch": 0.23034934497816595,
+      "grad_norm": 1.2858994143144755,
+      "learning_rate": 7.428307969745156e-06,
+      "loss": 0.5861,
+      "step": 211
+    },
+    {
+      "epoch": 0.2314410480349345,
+      "grad_norm": 1.604597407185956,
+      "learning_rate": 7.422010698168546e-06,
+      "loss": 0.5987,
+      "step": 212
+    },
+    {
+      "epoch": 0.23253275109170304,
+      "grad_norm": 1.3655281919289834,
+      "learning_rate": 7.415682628520253e-06,
+      "loss": 0.5937,
+      "step": 213
+    },
+    {
+      "epoch": 0.2336244541484716,
+      "grad_norm": 1.506921841428789,
+      "learning_rate": 7.409323841267445e-06,
+      "loss": 0.6669,
+      "step": 214
+    },
+    {
+      "epoch": 0.23471615720524017,
+      "grad_norm": 1.5101680215020887,
+      "learning_rate": 7.402934417267893e-06,
+      "loss": 0.6308,
+      "step": 215
+    },
+    {
+      "epoch": 0.23580786026200873,
+      "grad_norm": 1.489150088241989,
+      "learning_rate": 7.396514437768942e-06,
+      "loss": 0.6392,
+      "step": 216
+    },
+    {
+      "epoch": 0.2368995633187773,
+      "grad_norm": 1.4119779294110884,
+      "learning_rate": 7.3900639844064754e-06,
+      "loss": 0.5056,
+      "step": 217
+    },
+    {
+      "epoch": 0.23799126637554585,
+      "grad_norm": 1.445389704500089,
+      "learning_rate": 7.383583139203887e-06,
+      "loss": 0.5456,
+      "step": 218
+    },
+    {
+      "epoch": 0.2390829694323144,
+      "grad_norm": 1.4709438420906595,
+      "learning_rate": 7.3770719845710206e-06,
+      "loss": 0.6319,
+      "step": 219
+    },
+    {
+      "epoch": 0.24017467248908297,
+      "grad_norm": 1.3633406909433219,
+      "learning_rate": 7.370530603303139e-06,
+      "loss": 0.5647,
+      "step": 220
+    },
+    {
+      "epoch": 0.24126637554585154,
+      "grad_norm": 1.4987673997428108,
+      "learning_rate": 7.363959078579863e-06,
+      "loss": 0.5942,
+      "step": 221
+    },
+    {
+      "epoch": 0.2423580786026201,
+      "grad_norm": 1.4832422188486167,
+      "learning_rate": 7.3573574939641115e-06,
+      "loss": 0.5636,
+      "step": 222
+    },
+    {
+      "epoch": 0.24344978165938866,
+      "grad_norm": 1.4162439765343409,
+      "learning_rate": 7.350725933401046e-06,
+      "loss": 0.5328,
+      "step": 223
+    },
+    {
+      "epoch": 0.2445414847161572,
+      "grad_norm": 1.4080662897353993,
+      "learning_rate": 7.344064481217e-06,
+      "loss": 0.5343,
+      "step": 224
+    },
+    {
+      "epoch": 0.24563318777292575,
+      "grad_norm": 1.5492494250984743,
+      "learning_rate": 7.3373732221184015e-06,
+      "loss": 0.5613,
+      "step": 225
+    },
+    {
+      "epoch": 0.24672489082969432,
+      "grad_norm": 2.649801771711655,
+      "learning_rate": 7.330652241190705e-06,
+      "loss": 0.6204,
+      "step": 226
+    },
+    {
+      "epoch": 0.24781659388646288,
+      "grad_norm": 1.472867143202289,
+      "learning_rate": 7.3239016238973034e-06,
+      "loss": 0.4709,
+      "step": 227
+    },
+    {
+      "epoch": 0.24890829694323144,
+      "grad_norm": 1.3984113233132964,
+      "learning_rate": 7.3171214560784426e-06,
+      "loss": 0.4408,
+      "step": 228
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 1.3248399752948739,
+      "learning_rate": 7.310311823950132e-06,
+      "loss": 0.6315,
+      "step": 229
+    },
+    {
+      "epoch": 0.25109170305676853,
+      "grad_norm": 1.393688466478258,
+      "learning_rate": 7.3034728141030436e-06,
+      "loss": 0.6316,
+      "step": 230
+    },
+    {
+      "epoch": 0.2521834061135371,
+      "grad_norm": 1.5554375243032204,
+      "learning_rate": 7.2966045135014176e-06,
+      "loss": 0.552,
+      "step": 231
+    },
+    {
+      "epoch": 0.25327510917030566,
+      "grad_norm": 1.5113408431573132,
+      "learning_rate": 7.289707009481951e-06,
+      "loss": 0.5209,
+      "step": 232
+    },
+    {
+      "epoch": 0.25436681222707425,
+      "grad_norm": 1.8008902213945412,
+      "learning_rate": 7.282780389752688e-06,
+      "loss": 0.5615,
+      "step": 233
+    },
+    {
+      "epoch": 0.2554585152838428,
+      "grad_norm": 1.5242476878510804,
+      "learning_rate": 7.275824742391906e-06,
+      "loss": 0.5996,
+      "step": 234
+    },
+    {
+      "epoch": 0.25655021834061137,
+      "grad_norm": 1.6201534221012874,
+      "learning_rate": 7.2688401558469955e-06,
+      "loss": 0.6023,
+      "step": 235
+    },
+    {
+      "epoch": 0.2576419213973799,
+      "grad_norm": 1.65373334840132,
+      "learning_rate": 7.261826718933335e-06,
+      "loss": 0.6784,
+      "step": 236
+    },
+    {
+      "epoch": 0.2587336244541485,
+      "grad_norm": 1.357403882003097,
+      "learning_rate": 7.254784520833163e-06,
+      "loss": 0.537,
+      "step": 237
+    },
+    {
+      "epoch": 0.259825327510917,
+      "grad_norm": 1.5116722340731783,
+      "learning_rate": 7.247713651094441e-06,
+      "loss": 0.5698,
+      "step": 238
+    },
+    {
+      "epoch": 0.2609170305676856,
+      "grad_norm": 1.4288603307083088,
+      "learning_rate": 7.240614199629716e-06,
+      "loss": 0.6177,
+      "step": 239
+    },
+    {
+      "epoch": 0.26200873362445415,
+      "grad_norm": 1.4338899312326936,
+      "learning_rate": 7.23348625671498e-06,
+      "loss": 0.6443,
+      "step": 240
+    },
+    {
+      "epoch": 0.2631004366812227,
+      "grad_norm": 1.4962646720928754,
+      "learning_rate": 7.226329912988517e-06,
+      "loss": 0.6777,
+      "step": 241
+    },
+    {
+      "epoch": 0.26419213973799127,
+      "grad_norm": 1.2699992707782755,
+      "learning_rate": 7.219145259449756e-06,
+      "loss": 0.4732,
+      "step": 242
+    },
+    {
+      "epoch": 0.2652838427947598,
+      "grad_norm": 1.4131898857557792,
+      "learning_rate": 7.211932387458108e-06,
+      "loss": 0.6738,
+      "step": 243
+    },
+    {
+      "epoch": 0.2663755458515284,
+      "grad_norm": 1.5277531162623728,
+      "learning_rate": 7.2046913887318105e-06,
+      "loss": 0.612,
+      "step": 244
+    },
+    {
+      "epoch": 0.26746724890829693,
+      "grad_norm": 1.7351026071707434,
+      "learning_rate": 7.197422355346756e-06,
+      "loss": 0.6016,
+      "step": 245
+    },
+    {
+      "epoch": 0.2685589519650655,
+      "grad_norm": 1.613296891262151,
+      "learning_rate": 7.190125379735322e-06,
+      "loss": 0.476,
+      "step": 246
+    },
+    {
+      "epoch": 0.26965065502183405,
+      "grad_norm": 1.2399403442028993,
+      "learning_rate": 7.182800554685201e-06,
+      "loss": 0.568,
+      "step": 247
+    },
+    {
+      "epoch": 0.27074235807860264,
+      "grad_norm": 1.602746922015574,
+      "learning_rate": 7.175447973338213e-06,
+      "loss": 0.6195,
+      "step": 248
+    },
+    {
+      "epoch": 0.2718340611353712,
+      "grad_norm": 1.3651825200088947,
+      "learning_rate": 7.168067729189126e-06,
+      "loss": 0.5219,
+      "step": 249
+    },
+    {
+      "epoch": 0.27292576419213976,
+      "grad_norm": 1.4769248274303346,
+      "learning_rate": 7.160659916084464e-06,
+      "loss": 0.5377,
+      "step": 250
+    },
+    {
+      "epoch": 0.2740174672489083,
+      "grad_norm": 1.625431355976466,
+      "learning_rate": 7.153224628221318e-06,
+      "loss": 0.5888,
+      "step": 251
+    },
+    {
+      "epoch": 0.27510917030567683,
+      "grad_norm": 1.363907269155982,
+      "learning_rate": 7.145761960146144e-06,
+      "loss": 0.6062,
+      "step": 252
+    },
+    {
+      "epoch": 0.2762008733624454,
+      "grad_norm": 1.85268561061027,
+      "learning_rate": 7.138272006753563e-06,
+      "loss": 0.4969,
+      "step": 253
+    },
+    {
+      "epoch": 0.27729257641921395,
+      "grad_norm": 1.5428882847739833,
+      "learning_rate": 7.130754863285153e-06,
+      "loss": 0.4758,
+      "step": 254
+    },
+    {
+      "epoch": 0.27838427947598254,
+      "grad_norm": 1.6972894826270524,
+      "learning_rate": 7.123210625328238e-06,
+      "loss": 0.6367,
+      "step": 255
+    },
+    {
+      "epoch": 0.2794759825327511,
+      "grad_norm": 1.56171538368843,
+      "learning_rate": 7.115639388814675e-06,
+      "loss": 0.5483,
+      "step": 256
+    },
+    {
+      "epoch": 0.28056768558951967,
+      "grad_norm": 1.616159463723734,
+      "learning_rate": 7.1080412500196316e-06,
+      "loss": 0.6188,
+      "step": 257
+    },
+    {
+      "epoch": 0.2816593886462882,
+      "grad_norm": 1.664727956285899,
+      "learning_rate": 7.10041630556036e-06,
+      "loss": 0.5104,
+      "step": 258
+    },
+    {
+      "epoch": 0.2827510917030568,
+      "grad_norm": 1.472585818104954,
+      "learning_rate": 7.092764652394974e-06,
+      "loss": 0.5146,
+      "step": 259
+    },
+    {
+      "epoch": 0.2838427947598253,
+      "grad_norm": 1.2854288778227903,
+      "learning_rate": 7.085086387821213e-06,
+      "loss": 0.6012,
+      "step": 260
+    },
+    {
+      "epoch": 0.2849344978165939,
+      "grad_norm": 1.2852267764551335,
+      "learning_rate": 7.0773816094752015e-06,
+      "loss": 0.6049,
+      "step": 261
+    },
+    {
+      "epoch": 0.28602620087336245,
+      "grad_norm": 1.3567820396359396,
+      "learning_rate": 7.069650415330213e-06,
+      "loss": 0.531,
+      "step": 262
+    },
+    {
+      "epoch": 0.287117903930131,
+      "grad_norm": 1.460404193275241,
+      "learning_rate": 7.061892903695423e-06,
+      "loss": 0.5811,
+      "step": 263
+    },
+    {
+      "epoch": 0.28820960698689957,
+      "grad_norm": 1.4744335745677344,
+      "learning_rate": 7.0541091732146535e-06,
+      "loss": 0.6736,
+      "step": 264
+    },
+    {
+      "epoch": 0.2893013100436681,
+      "grad_norm": 1.4944561183840575,
+      "learning_rate": 7.046299322865131e-06,
+      "loss": 0.62,
+      "step": 265
+    },
+    {
+      "epoch": 0.2903930131004367,
+      "grad_norm": 1.5993459040844344,
+      "learning_rate": 7.038463451956209e-06,
+      "loss": 0.6203,
+      "step": 266
+    },
+    {
+      "epoch": 0.2914847161572052,
+      "grad_norm": 1.398558452093382,
+      "learning_rate": 7.030601660128127e-06,
+      "loss": 0.6007,
+      "step": 267
+    },
+    {
+      "epoch": 0.2925764192139738,
+      "grad_norm": 1.490504616058761,
+      "learning_rate": 7.0227140473507245e-06,
+      "loss": 0.5718,
+      "step": 268
+    },
+    {
+      "epoch": 0.29366812227074235,
+      "grad_norm": 1.3890197651526783,
+      "learning_rate": 7.014800713922181e-06,
+      "loss": 0.6712,
+      "step": 269
+    },
+    {
+      "epoch": 0.29475982532751094,
+      "grad_norm": 1.475360288498539,
+      "learning_rate": 7.006861760467741e-06,
+      "loss": 0.6257,
+      "step": 270
+    },
+    {
+      "epoch": 0.29585152838427947,
+      "grad_norm": 1.4498851473925982,
+      "learning_rate": 6.998897287938424e-06,
+      "loss": 0.5378,
+      "step": 271
+    },
+    {
+      "epoch": 0.29694323144104806,
+      "grad_norm": 1.3144961611880193,
+      "learning_rate": 6.990907397609751e-06,
+      "loss": 0.5108,
+      "step": 272
+    },
+    {
+      "epoch": 0.2980349344978166,
+      "grad_norm": 1.3859005230286172,
+      "learning_rate": 6.982892191080456e-06,
+      "loss": 0.4885,
+      "step": 273
+    },
+    {
+      "epoch": 0.29912663755458513,
+      "grad_norm": 1.221223897043554,
+      "learning_rate": 6.974851770271186e-06,
+      "loss": 0.5224,
+      "step": 274
+    },
+    {
+      "epoch": 0.3002183406113537,
+      "grad_norm": 1.38427922807046,
+      "learning_rate": 6.966786237423218e-06,
+      "loss": 0.5956,
+      "step": 275
+    },
+    {
+      "epoch": 0.30131004366812225,
+      "grad_norm": 1.2894973461951942,
+      "learning_rate": 6.958695695097145e-06,
+      "loss": 0.4793,
+      "step": 276
+    },
+    {
+      "epoch": 0.30240174672489084,
+      "grad_norm": 1.4482446520872734,
+      "learning_rate": 6.950580246171581e-06,
+      "loss": 0.6768,
+      "step": 277
+    },
+    {
+      "epoch": 0.3034934497816594,
+      "grad_norm": 1.5030922903307724,
+      "learning_rate": 6.942439993841848e-06,
+      "loss": 0.4834,
+      "step": 278
+    },
+    {
+      "epoch": 0.30458515283842796,
+      "grad_norm": 1.4439095189849094,
+      "learning_rate": 6.93427504161867e-06,
+      "loss": 0.4781,
+      "step": 279
+    },
+    {
+      "epoch": 0.3056768558951965,
+      "grad_norm": 1.4054270284734758,
+      "learning_rate": 6.9260854933268466e-06,
+      "loss": 0.5164,
+      "step": 280
+    },
+    {
+      "epoch": 0.3056768558951965,
+      "eval_accuracy": 0.8122814246384281,
+      "eval_accuracy_first_token": 0.7722520229111738,
+      "eval_accuracy_first_token_<": 0.9699042407660738,
+      "eval_accuracy_first_token_<_total": 2924,
+      "eval_accuracy_first_token_<|python_tag|>": 0.8661971830985915,
+      "eval_accuracy_first_token_<|python_tag|>_total": 994,
+      "eval_accuracy_first_token_Here": 0.6010362694300518,
+      "eval_accuracy_first_token_Here_total": 386,
+      "eval_accuracy_first_token_The": 0.9305943358595932,
+      "eval_accuracy_first_token_The_total": 2507,
+      "eval_accuracy_first_token_To": 0.7403846153846154,
+      "eval_accuracy_first_token_To_total": 936,
+      "eval_first_token_param_values": 0.9361062892241955,
+      "eval_first_token_param_values_total": 14978,
+      "eval_loss": 0.5944764614105225,
+      "eval_perplexity": 1.1145611460638327,
+      "eval_runtime": 404.5683,
+      "eval_samples_per_second": 1.916,
+      "eval_steps_per_second": 0.24,
+      "eval_total_number_first_token": 10999,
+      "step": 280
+    },
+    {
+      "epoch": 0.3067685589519651,
+      "grad_norm": 1.2731720997992346,
+      "learning_rate": 6.917871453103946e-06,
+      "loss": 0.5342,
+      "step": 281
+    },
+    {
+      "epoch": 0.3078602620087336,
+      "grad_norm": 1.4627726722347458,
+      "learning_rate": 6.909633025398968e-06,
+      "loss": 0.4906,
+      "step": 282
+    },
+    {
+      "epoch": 0.3089519650655022,
+      "grad_norm": 1.391286615450489,
+      "learning_rate": 6.901370314971025e-06,
+      "loss": 0.5165,
+      "step": 283
+    },
+    {
+      "epoch": 0.31004366812227074,
+      "grad_norm": 1.461047803005325,
+      "learning_rate": 6.893083426888001e-06,
+      "loss": 0.6427,
+      "step": 284
+    },
+    {
+      "epoch": 0.3111353711790393,
+      "grad_norm": 1.4333799644406657,
+      "learning_rate": 6.884772466525229e-06,
+      "loss": 0.5478,
+      "step": 285
+    },
+    {
+      "epoch": 0.31222707423580787,
+      "grad_norm": 1.4838513414542995,
+      "learning_rate": 6.8764375395641365e-06,
+      "loss": 0.6074,
+      "step": 286
+    },
+    {
+      "epoch": 0.3133187772925764,
+      "grad_norm": 1.2646171911891406,
+      "learning_rate": 6.86807875199091e-06,
+      "loss": 0.5755,
+      "step": 287
+    },
+    {
+      "epoch": 0.314410480349345,
+      "grad_norm": 1.2542325136137744,
+      "learning_rate": 6.859696210095146e-06,
+      "loss": 0.5502,
+      "step": 288
+    },
+    {
+      "epoch": 0.3155021834061135,
+      "grad_norm": 1.2321799849318462,
+      "learning_rate": 6.851290020468498e-06,
+      "loss": 0.6221,
+      "step": 289
+    },
+    {
+      "epoch": 0.3165938864628821,
+      "grad_norm": 1.2953758966810032,
+      "learning_rate": 6.842860290003322e-06,
+      "loss": 0.5903,
+      "step": 290
+    },
+    {
+      "epoch": 0.31768558951965065,
+      "grad_norm": 1.7395783863975123,
+      "learning_rate": 6.8344071258913206e-06,
+      "loss": 0.5963,
+      "step": 291
+    },
+    {
+      "epoch": 0.31877729257641924,
+      "grad_norm": 1.4642573503581784,
+      "learning_rate": 6.825930635622169e-06,
+      "loss": 0.5011,
+      "step": 292
+    },
+    {
+      "epoch": 0.31986899563318777,
+      "grad_norm": 1.7487735502303772,
+      "learning_rate": 6.817430926982163e-06,
+      "loss": 0.5699,
+      "step": 293
+    },
+    {
+      "epoch": 0.32096069868995636,
+      "grad_norm": 1.4436349581997563,
+      "learning_rate": 6.80890810805284e-06,
+      "loss": 0.559,
+      "step": 294
+    },
+    {
+      "epoch": 0.3220524017467249,
+      "grad_norm": 1.4902082225296027,
+      "learning_rate": 6.8003622872096e-06,
+      "loss": 0.6149,
+      "step": 295
+    },
+    {
+      "epoch": 0.3231441048034934,
+      "grad_norm": 1.5021927515338005,
+      "learning_rate": 6.791793573120342e-06,
+      "loss": 0.5098,
+      "step": 296
+    },
+    {
+      "epoch": 0.324235807860262,
+      "grad_norm": 1.2568489718668712,
+      "learning_rate": 6.7832020747440645e-06,
+      "loss": 0.4616,
+      "step": 297
+    },
+    {
+      "epoch": 0.32532751091703055,
+      "grad_norm": 1.85090367162997,
+      "learning_rate": 6.7745879013294945e-06,
+      "loss": 0.6342,
+      "step": 298
+    },
+    {
+      "epoch": 0.32641921397379914,
+      "grad_norm": 1.3622630820375037,
+      "learning_rate": 6.765951162413695e-06,
+      "loss": 0.574,
+      "step": 299
+    },
+    {
+      "epoch": 0.32751091703056767,
+      "grad_norm": 1.3544768555004285,
+      "learning_rate": 6.757291967820663e-06,
+      "loss": 0.4971,
+      "step": 300
+    },
+    {
+      "epoch": 0.32860262008733626,
+      "grad_norm": 1.2812885948189523,
+      "learning_rate": 6.748610427659944e-06,
+      "loss": 0.4419,
+      "step": 301
+    },
+    {
+      "epoch": 0.3296943231441048,
+      "grad_norm": 1.3044190835387586,
+      "learning_rate": 6.739906652325228e-06,
+      "loss": 0.5298,
+      "step": 302
+    },
+    {
+      "epoch": 0.3307860262008734,
+      "grad_norm": 1.6049637774921335,
+      "learning_rate": 6.731180752492944e-06,
+      "loss": 0.5731,
+      "step": 303
+    },
+    {
+      "epoch": 0.3318777292576419,
+      "grad_norm": 1.5188076633711975,
+      "learning_rate": 6.722432839120857e-06,
+      "loss": 0.5466,
+      "step": 304
+    },
+    {
+      "epoch": 0.3329694323144105,
+      "grad_norm": 1.2200311097592393,
+      "learning_rate": 6.713663023446648e-06,
+      "loss": 0.5754,
+      "step": 305
+    },
+    {
+      "epoch": 0.33406113537117904,
+      "grad_norm": 1.1943617882666278,
+      "learning_rate": 6.704871416986513e-06,
+      "loss": 0.6153,
+      "step": 306
+    },
+    {
+      "epoch": 0.3351528384279476,
+      "grad_norm": 1.2566424888713876,
+      "learning_rate": 6.696058131533732e-06,
+      "loss": 0.5391,
+      "step": 307
+    },
+    {
+      "epoch": 0.33624454148471616,
+      "grad_norm": 1.5354142684922412,
+      "learning_rate": 6.687223279157258e-06,
+      "loss": 0.512,
+      "step": 308
+    },
+    {
+      "epoch": 0.3373362445414847,
+      "grad_norm": 1.7632124793181871,
+      "learning_rate": 6.678366972200284e-06,
+      "loss": 0.5118,
+      "step": 309
+    },
+    {
+      "epoch": 0.3384279475982533,
+      "grad_norm": 1.4051894291399412,
+      "learning_rate": 6.669489323278818e-06,
+      "loss": 0.5959,
+      "step": 310
+    },
+    {
+      "epoch": 0.3395196506550218,
+      "grad_norm": 1.5575194677153044,
+      "learning_rate": 6.660590445280248e-06,
+      "loss": 0.5217,
+      "step": 311
+    },
+    {
+      "epoch": 0.3406113537117904,
+      "grad_norm": 1.445909145564962,
+      "learning_rate": 6.651670451361916e-06,
+      "loss": 0.5329,
+      "step": 312
+    },
+    {
+      "epoch": 0.34170305676855894,
+      "grad_norm": 1.6148708469036988,
+      "learning_rate": 6.642729454949665e-06,
+      "loss": 0.5441,
+      "step": 313
+    },
+    {
+      "epoch": 0.34279475982532753,
+      "grad_norm": 1.3877288714622236,
+      "learning_rate": 6.633767569736406e-06,
+      "loss": 0.4973,
+      "step": 314
+    },
+    {
+      "epoch": 0.34388646288209607,
+      "grad_norm": 1.5705634975133718,
+      "learning_rate": 6.624784909680674e-06,
+      "loss": 0.5379,
+      "step": 315
+    },
+    {
+      "epoch": 0.34497816593886466,
+      "grad_norm": 1.4683909348634254,
+      "learning_rate": 6.61578158900517e-06,
+      "loss": 0.5646,
+      "step": 316
+    },
+    {
+      "epoch": 0.3460698689956332,
+      "grad_norm": 1.444455571760505,
+      "learning_rate": 6.606757722195315e-06,
+      "loss": 0.6329,
+      "step": 317
+    },
+    {
+      "epoch": 0.3471615720524017,
+      "grad_norm": 1.4016572432346093,
+      "learning_rate": 6.597713423997795e-06,
+      "loss": 0.5736,
+      "step": 318
+    },
+    {
+      "epoch": 0.3482532751091703,
+      "grad_norm": 1.8469454503294669,
+      "learning_rate": 6.588648809419096e-06,
+      "loss": 0.5856,
+      "step": 319
+    },
+    {
+      "epoch": 0.34934497816593885,
+      "grad_norm": 1.3217897187946839,
+      "learning_rate": 6.579563993724048e-06,
+      "loss": 0.5124,
+      "step": 320
+    },
+    {
+      "epoch": 0.35043668122270744,
+      "grad_norm": 1.5194866076068712,
+      "learning_rate": 6.570459092434356e-06,
+      "loss": 0.4867,
+      "step": 321
+    },
+    {
+      "epoch": 0.35152838427947597,
+      "grad_norm": 1.408417064891368,
+      "learning_rate": 6.561334221327133e-06,
+      "loss": 0.5776,
+      "step": 322
+    },
+    {
+      "epoch": 0.35262008733624456,
+      "grad_norm": 1.4452329373302548,
+      "learning_rate": 6.552189496433421e-06,
+      "loss": 0.5251,
+      "step": 323
+    },
+    {
+      "epoch": 0.3537117903930131,
+      "grad_norm": 1.4435343192846628,
+      "learning_rate": 6.543025034036729e-06,
+      "loss": 0.5047,
+      "step": 324
+    },
+    {
+      "epoch": 0.3548034934497817,
+      "grad_norm": 1.3953834298183048,
+      "learning_rate": 6.533840950671539e-06,
+      "loss": 0.5376,
+      "step": 325
+    },
+    {
+      "epoch": 0.3558951965065502,
+      "grad_norm": 1.5746860190061827,
+      "learning_rate": 6.524637363121837e-06,
+      "loss": 0.5363,
+      "step": 326
+    },
+    {
+      "epoch": 0.3569868995633188,
+      "grad_norm": 1.430598020020894,
+      "learning_rate": 6.515414388419619e-06,
+      "loss": 0.6226,
+      "step": 327
+    },
+    {
+      "epoch": 0.35807860262008734,
+      "grad_norm": 1.2707811617436353,
+      "learning_rate": 6.5061721438434076e-06,
+      "loss": 0.4761,
+      "step": 328
+    },
+    {
+      "epoch": 0.35917030567685587,
+      "grad_norm": 1.5783432705751568,
+      "learning_rate": 6.496910746916759e-06,
+      "loss": 0.5214,
+      "step": 329
+    },
+    {
+      "epoch": 0.36026200873362446,
+      "grad_norm": 1.83574416889433,
+      "learning_rate": 6.487630315406768e-06,
+      "loss": 0.5599,
+      "step": 330
+    },
+    {
+      "epoch": 0.361353711790393,
+      "grad_norm": 1.322420845333774,
+      "learning_rate": 6.478330967322577e-06,
+      "loss": 0.5336,
+      "step": 331
+    },
+    {
+      "epoch": 0.3624454148471616,
+      "grad_norm": 1.3612789015918931,
+      "learning_rate": 6.469012820913861e-06,
+      "loss": 0.5508,
+      "step": 332
+    },
+    {
+      "epoch": 0.3635371179039301,
+      "grad_norm": 1.3891730681529915,
+      "learning_rate": 6.45967599466934e-06,
+      "loss": 0.4888,
+      "step": 333
+    },
+    {
+      "epoch": 0.3646288209606987,
+      "grad_norm": 1.4417746815764274,
+      "learning_rate": 6.45032060731526e-06,
+      "loss": 0.5416,
+      "step": 334
+    },
+    {
+      "epoch": 0.36572052401746724,
+      "grad_norm": 1.4895891533428605,
+      "learning_rate": 6.440946777813895e-06,
+      "loss": 0.5417,
+      "step": 335
+    },
+    {
+      "epoch": 0.36681222707423583,
+      "grad_norm": 1.4141847222045185,
+      "learning_rate": 6.431554625362018e-06,
+      "loss": 0.4896,
+      "step": 336
+    },
+    {
+      "epoch": 0.36790393013100436,
+      "grad_norm": 1.5117794325198755,
+      "learning_rate": 6.422144269389406e-06,
+      "loss": 0.5532,
+      "step": 337
+    },
+    {
+      "epoch": 0.36899563318777295,
+      "grad_norm": 1.58498155036596,
+      "learning_rate": 6.412715829557301e-06,
+      "loss": 0.4635,
+      "step": 338
+    },
+    {
+      "epoch": 0.3700873362445415,
+      "grad_norm": 1.3847721442020027,
+      "learning_rate": 6.403269425756905e-06,
+      "loss": 0.4702,
+      "step": 339
+    },
+    {
+      "epoch": 0.37117903930131,
+      "grad_norm": 1.3681322527990918,
+      "learning_rate": 6.3938051781078405e-06,
+      "loss": 0.5593,
+      "step": 340
+    },
+    {
+      "epoch": 0.3722707423580786,
+      "grad_norm": 1.503661732921771,
+      "learning_rate": 6.3843232069566396e-06,
+      "loss": 0.5693,
+      "step": 341
+    },
+    {
+      "epoch": 0.37336244541484714,
+      "grad_norm": 1.5045172592179,
+      "learning_rate": 6.374823632875197e-06,
+      "loss": 0.5619,
+      "step": 342
+    },
+    {
+      "epoch": 0.37445414847161573,
+      "grad_norm": 1.4568114586028496,
+      "learning_rate": 6.365306576659252e-06,
+      "loss": 0.5992,
+      "step": 343
+    },
+    {
+      "epoch": 0.37554585152838427,
+      "grad_norm": 1.2603187279915327,
+      "learning_rate": 6.3557721593268375e-06,
+      "loss": 0.5808,
+      "step": 344
+    },
+    {
+      "epoch": 0.37663755458515286,
+      "grad_norm": 1.5105002968929564,
+      "learning_rate": 6.3462205021167545e-06,
+      "loss": 0.4838,
+      "step": 345
+    },
+    {
+      "epoch": 0.3777292576419214,
+      "grad_norm": 1.7236853852664615,
+      "learning_rate": 6.336651726487023e-06,
+      "loss": 0.519,
+      "step": 346
+    },
+    {
+      "epoch": 0.37882096069869,
+      "grad_norm": 1.5823366228954794,
+      "learning_rate": 6.327065954113339e-06,
+      "loss": 0.5357,
+      "step": 347
+    },
+    {
+      "epoch": 0.3799126637554585,
+      "grad_norm": 1.6501056758652983,
+      "learning_rate": 6.317463306887525e-06,
+      "loss": 0.7162,
+      "step": 348
+    },
+    {
+      "epoch": 0.38100436681222705,
+      "grad_norm": 1.612984303794767,
+      "learning_rate": 6.307843906915991e-06,
+      "loss": 0.5861,
+      "step": 349
+    },
+    {
+      "epoch": 0.38209606986899564,
+      "grad_norm": 2.0839316924516287,
+      "learning_rate": 6.298207876518162e-06,
+      "loss": 0.5356,
+      "step": 350
+    },
+    {
+      "epoch": 0.38318777292576417,
+      "grad_norm": 1.6442415918972877,
+      "learning_rate": 6.288555338224944e-06,
+      "loss": 0.4907,
+      "step": 351
+    },
+    {
+      "epoch": 0.38427947598253276,
+      "grad_norm": 1.4680570384981029,
+      "learning_rate": 6.278886414777148e-06,
+      "loss": 0.5877,
+      "step": 352
+    },
+    {
+      "epoch": 0.3853711790393013,
+      "grad_norm": 1.5004977869240974,
+      "learning_rate": 6.269201229123943e-06,
+      "loss": 0.5518,
+      "step": 353
+    },
+    {
+      "epoch": 0.3864628820960699,
+      "grad_norm": 1.5820396873030045,
+      "learning_rate": 6.259499904421285e-06,
+      "loss": 0.566,
+      "step": 354
+    },
+    {
+      "epoch": 0.3875545851528384,
+      "grad_norm": 1.2944420053683188,
+      "learning_rate": 6.249782564030352e-06,
+      "loss": 0.41,
+      "step": 355
+    },
+    {
+      "epoch": 0.388646288209607,
+      "grad_norm": 1.2182802013913994,
+      "learning_rate": 6.240049331515976e-06,
+      "loss": 0.4647,
+      "step": 356
+    },
+    {
+      "epoch": 0.38973799126637554,
+      "grad_norm": 1.5151046008059181,
+      "learning_rate": 6.230300330645072e-06,
+      "loss": 0.591,
+      "step": 357
+    },
+    {
+      "epoch": 0.39082969432314413,
+      "grad_norm": 1.4656723716881035,
+      "learning_rate": 6.220535685385064e-06,
+      "loss": 0.5486,
+      "step": 358
+    },
+    {
+      "epoch": 0.39192139737991266,
+      "grad_norm": 1.256305336512033,
+      "learning_rate": 6.21075551990231e-06,
+      "loss": 0.558,
+      "step": 359
+    },
+    {
+      "epoch": 0.3930131004366812,
+      "grad_norm": 1.3298929325184357,
+      "learning_rate": 6.20095995856052e-06,
+      "loss": 0.5805,
+      "step": 360
+    },
+    {
+      "epoch": 0.3941048034934498,
+      "grad_norm": 1.4212920029170337,
+      "learning_rate": 6.1911491259191766e-06,
+      "loss": 0.4801,
+      "step": 361
+    },
+    {
+      "epoch": 0.3951965065502183,
+      "grad_norm": 1.366679203076844,
+      "learning_rate": 6.181323146731951e-06,
+      "loss": 0.5107,
+      "step": 362
+    },
+    {
+      "epoch": 0.3962882096069869,
+      "grad_norm": 1.38703232756071,
+      "learning_rate": 6.1714821459451186e-06,
+      "loss": 0.5678,
+      "step": 363
+    },
+    {
+      "epoch": 0.39737991266375544,
+      "grad_norm": 1.4858572589764882,
+      "learning_rate": 6.1616262486959646e-06,
+      "loss": 0.4529,
+      "step": 364
+    },
+    {
+      "epoch": 0.39847161572052403,
+      "grad_norm": 1.5348563857001403,
+      "learning_rate": 6.151755580311199e-06,
+      "loss": 0.5933,
+      "step": 365
+    },
+    {
+      "epoch": 0.39956331877729256,
+      "grad_norm": 1.4176620057000997,
+      "learning_rate": 6.141870266305358e-06,
+      "loss": 0.4918,
+      "step": 366
+    },
+    {
+      "epoch": 0.40065502183406115,
+      "grad_norm": 1.4951222520742187,
+      "learning_rate": 6.131970432379213e-06,
+      "loss": 0.5709,
+      "step": 367
+    },
+    {
+      "epoch": 0.4017467248908297,
+      "grad_norm": 1.6210277269111266,
+      "learning_rate": 6.122056204418166e-06,
+      "loss": 0.5339,
+      "step": 368
+    },
+    {
+      "epoch": 0.4028384279475983,
+      "grad_norm": 1.8538643894848557,
+      "learning_rate": 6.112127708490657e-06,
+      "loss": 0.5994,
+      "step": 369
+    },
+    {
+      "epoch": 0.4039301310043668,
+      "grad_norm": 1.3107005664617082,
+      "learning_rate": 6.102185070846551e-06,
+      "loss": 0.5299,
+      "step": 370
+    },
+    {
+      "epoch": 0.40502183406113534,
+      "grad_norm": 1.52599657247444,
+      "learning_rate": 6.092228417915541e-06,
+      "loss": 0.6073,
+      "step": 371
+    },
+    {
+      "epoch": 0.40611353711790393,
+      "grad_norm": 1.4601043256949482,
+      "learning_rate": 6.082257876305534e-06,
+      "loss": 0.5638,
+      "step": 372
+    },
+    {
+      "epoch": 0.40720524017467247,
+      "grad_norm": 1.4005447907301034,
+      "learning_rate": 6.07227357280105e-06,
+      "loss": 0.5575,
+      "step": 373
+    },
+    {
+      "epoch": 0.40829694323144106,
+      "grad_norm": 1.5877468909166246,
+      "learning_rate": 6.062275634361599e-06,
+      "loss": 0.6142,
+      "step": 374
+    },
+    {
+      "epoch": 0.4093886462882096,
+      "grad_norm": 1.8682639604384457,
+      "learning_rate": 6.0522641881200734e-06,
+      "loss": 0.6107,
+      "step": 375
+    },
+    {
+      "epoch": 0.4104803493449782,
+      "grad_norm": 1.3289776248271887,
+      "learning_rate": 6.042239361381128e-06,
+      "loss": 0.5571,
+      "step": 376
+    },
+    {
+      "epoch": 0.4115720524017467,
+      "grad_norm": 1.2005221136761943,
+      "learning_rate": 6.032201281619565e-06,
+      "loss": 0.6254,
+      "step": 377
+    },
+    {
+      "epoch": 0.4126637554585153,
+      "grad_norm": 1.3166126771313549,
+      "learning_rate": 6.02215007647871e-06,
+      "loss": 0.4617,
+      "step": 378
+    },
+    {
+      "epoch": 0.41375545851528384,
+      "grad_norm": 1.5210638954921578,
+      "learning_rate": 6.012085873768791e-06,
+      "loss": 0.5461,
+      "step": 379
+    },
+    {
+      "epoch": 0.4148471615720524,
+      "grad_norm": 1.5029108053726574,
+      "learning_rate": 6.002008801465307e-06,
+      "loss": 0.5954,
+      "step": 380
+    },
+    {
+      "epoch": 0.41593886462882096,
+      "grad_norm": 1.4322948512523916,
+      "learning_rate": 5.991918987707414e-06,
+      "loss": 0.5222,
+      "step": 381
+    },
+    {
+      "epoch": 0.4170305676855895,
+      "grad_norm": 1.5951914575628332,
+      "learning_rate": 5.981816560796278e-06,
+      "loss": 0.5121,
+      "step": 382
+    },
+    {
+      "epoch": 0.4181222707423581,
+      "grad_norm": 1.4733981716395201,
+      "learning_rate": 5.971701649193461e-06,
+      "loss": 0.5417,
+      "step": 383
+    },
+    {
+      "epoch": 0.4192139737991266,
+      "grad_norm": 1.4093851168934195,
+      "learning_rate": 5.961574381519272e-06,
+      "loss": 0.532,
+      "step": 384
+    },
+    {
+      "epoch": 0.4203056768558952,
+      "grad_norm": 1.8731153406678291,
+      "learning_rate": 5.951434886551144e-06,
+      "loss": 0.596,
+      "step": 385
+    },
+    {
+      "epoch": 0.42139737991266374,
+      "grad_norm": 1.3983487740826432,
+      "learning_rate": 5.941283293221989e-06,
+      "loss": 0.5818,
+      "step": 386
+    },
+    {
+      "epoch": 0.42248908296943233,
+      "grad_norm": 1.438105551234948,
+      "learning_rate": 5.931119730618561e-06,
+      "loss": 0.6056,
+      "step": 387
+    },
+    {
+      "epoch": 0.42358078602620086,
+      "grad_norm": 1.5050579045957733,
+      "learning_rate": 5.920944327979814e-06,
+      "loss": 0.5443,
+      "step": 388
+    },
+    {
+      "epoch": 0.42467248908296945,
+      "grad_norm": 1.7282154077005178,
+      "learning_rate": 5.910757214695259e-06,
+      "loss": 0.5875,
+      "step": 389
+    },
+    {
+      "epoch": 0.425764192139738,
+      "grad_norm": 1.280595735861228,
+      "learning_rate": 5.900558520303318e-06,
+      "loss": 0.5573,
+      "step": 390
+    },
+    {
+      "epoch": 0.4268558951965066,
+      "grad_norm": 1.2728154446007562,
+      "learning_rate": 5.89034837448968e-06,
+      "loss": 0.523,
+      "step": 391
+    },
+    {
+      "epoch": 0.4279475982532751,
+      "grad_norm": 1.5426782593172892,
+      "learning_rate": 5.880126907085643e-06,
+      "loss": 0.5358,
+      "step": 392
+    },
+    {
+      "epoch": 0.42903930131004364,
+      "grad_norm": 1.5200061103097575,
+      "learning_rate": 5.8698942480664775e-06,
+      "loss": 0.5418,
+      "step": 393
+    },
+    {
+      "epoch": 0.43013100436681223,
+      "grad_norm": 1.7356764870493646,
+      "learning_rate": 5.859650527549761e-06,
+      "loss": 0.5654,
+      "step": 394
+    },
+    {
+      "epoch": 0.43122270742358076,
+      "grad_norm": 1.3654496081014231,
+      "learning_rate": 5.849395875793726e-06,
+      "loss": 0.5915,
+      "step": 395
+    },
+    {
+      "epoch": 0.43231441048034935,
+      "grad_norm": 2.808147594668986,
+      "learning_rate": 5.839130423195609e-06,
+      "loss": 0.5665,
+      "step": 396
+    },
+    {
+      "epoch": 0.4334061135371179,
+      "grad_norm": 1.4833752079280529,
+      "learning_rate": 5.82885430028999e-06,
+      "loss": 0.5413,
+      "step": 397
+    },
+    {
+      "epoch": 0.4344978165938865,
+      "grad_norm": 1.6449672988864015,
+      "learning_rate": 5.8185676377471295e-06,
+      "loss": 0.5401,
+      "step": 398
+    },
+    {
+      "epoch": 0.435589519650655,
+      "grad_norm": 1.1977061243594427,
+      "learning_rate": 5.808270566371307e-06,
+      "loss": 0.4718,
+      "step": 399
+    },
+    {
+      "epoch": 0.4366812227074236,
+      "grad_norm": 1.3405831214725057,
+      "learning_rate": 5.7979632170991665e-06,
+      "loss": 0.568,
+      "step": 400
+    },
+    {
+      "epoch": 0.43777292576419213,
+      "grad_norm": 2.094672745602528,
+      "learning_rate": 5.787645720998037e-06,
+      "loss": 0.6035,
+      "step": 401
+    },
+    {
+      "epoch": 0.4388646288209607,
+      "grad_norm": 1.2660085035873359,
+      "learning_rate": 5.7773182092642785e-06,
+      "loss": 0.5269,
+      "step": 402
+    },
+    {
+      "epoch": 0.43995633187772926,
+      "grad_norm": 1.4836506744346154,
+      "learning_rate": 5.766980813221607e-06,
+      "loss": 0.5382,
+      "step": 403
+    },
+    {
+      "epoch": 0.4410480349344978,
+      "grad_norm": 1.4272157208447975,
+      "learning_rate": 5.75663366431943e-06,
+      "loss": 0.5292,
+      "step": 404
+    },
+    {
+      "epoch": 0.4421397379912664,
+      "grad_norm": 1.712850218127739,
+      "learning_rate": 5.746276894131165e-06,
+      "loss": 0.5852,
+      "step": 405
+    },
+    {
+      "epoch": 0.4432314410480349,
+      "grad_norm": 1.2286185333492807,
+      "learning_rate": 5.735910634352579e-06,
+      "loss": 0.5017,
+      "step": 406
+    },
+    {
+      "epoch": 0.4443231441048035,
+      "grad_norm": 1.238407584969687,
+      "learning_rate": 5.725535016800103e-06,
+      "loss": 0.4684,
+      "step": 407
+    },
+    {
+      "epoch": 0.44541484716157204,
+      "grad_norm": 1.5146880310733182,
+      "learning_rate": 5.7151501734091655e-06,
+      "loss": 0.6328,
+      "step": 408
+    },
+    {
+      "epoch": 0.4465065502183406,
+      "grad_norm": 1.2536815602440197,
+      "learning_rate": 5.704756236232505e-06,
+      "loss": 0.4686,
+      "step": 409
+    },
+    {
+      "epoch": 0.44759825327510916,
+      "grad_norm": 1.3163930911885626,
+      "learning_rate": 5.6943533374385e-06,
+      "loss": 0.5248,
+      "step": 410
+    },
+    {
+      "epoch": 0.44868995633187775,
+      "grad_norm": 1.4295888211351506,
+      "learning_rate": 5.683941609309483e-06,
+      "loss": 0.5344,
+      "step": 411
+    },
+    {
+      "epoch": 0.4497816593886463,
+      "grad_norm": 1.3389542681794442,
+      "learning_rate": 5.6735211842400565e-06,
+      "loss": 0.5823,
+      "step": 412
+    },
+    {
+      "epoch": 0.45087336244541487,
+      "grad_norm": 1.411103751375743,
+      "learning_rate": 5.6630921947354155e-06,
+      "loss": 0.5351,
+      "step": 413
+    },
+    {
+      "epoch": 0.4519650655021834,
+      "grad_norm": 1.3449450319960814,
+      "learning_rate": 5.652654773409659e-06,
+      "loss": 0.5219,
+      "step": 414
+    },
+    {
+      "epoch": 0.45305676855895194,
+      "grad_norm": 1.2260548801396869,
+      "learning_rate": 5.642209052984106e-06,
+      "loss": 0.5583,
+      "step": 415
+    },
+    {
+      "epoch": 0.45414847161572053,
+      "grad_norm": 1.2016754154637224,
+      "learning_rate": 5.631755166285602e-06,
+      "loss": 0.4704,
+      "step": 416
+    },
+    {
+      "epoch": 0.45524017467248906,
+      "grad_norm": 1.6555020620471408,
+      "learning_rate": 5.6212932462448376e-06,
+      "loss": 0.484,
+      "step": 417
+    },
+    {
+      "epoch": 0.45633187772925765,
+      "grad_norm": 1.5311624701116364,
+      "learning_rate": 5.610823425894653e-06,
+      "loss": 0.5493,
+      "step": 418
+    },
+    {
+      "epoch": 0.4574235807860262,
+      "grad_norm": 1.7090470948405863,
+      "learning_rate": 5.600345838368348e-06,
+      "loss": 0.5095,
+      "step": 419
+    },
+    {
+      "epoch": 0.4585152838427948,
+      "grad_norm": 1.512713841712156,
+      "learning_rate": 5.589860616897991e-06,
+      "loss": 0.5147,
+      "step": 420
+    },
+    {
+      "epoch": 0.4596069868995633,
+      "grad_norm": 1.7872665411814004,
+      "learning_rate": 5.5793678948127214e-06,
+      "loss": 0.5598,
+      "step": 421
+    },
+    {
+      "epoch": 0.4606986899563319,
+      "grad_norm": 1.6670125232959876,
+      "learning_rate": 5.568867805537055e-06,
+      "loss": 0.6108,
+      "step": 422
+    },
+    {
+      "epoch": 0.46179039301310043,
+      "grad_norm": 1.3800090550625805,
+      "learning_rate": 5.558360482589189e-06,
+      "loss": 0.4621,
+      "step": 423
+    },
+    {
+      "epoch": 0.462882096069869,
+      "grad_norm": 1.4361795723883242,
+      "learning_rate": 5.547846059579304e-06,
+      "loss": 0.6029,
+      "step": 424
+    },
+    {
+      "epoch": 0.46397379912663755,
+      "grad_norm": 1.2533304542951986,
+      "learning_rate": 5.537324670207865e-06,
+      "loss": 0.5281,
+      "step": 425
+    },
+    {
+      "epoch": 0.4650655021834061,
+      "grad_norm": 1.5542519820585374,
+      "learning_rate": 5.526796448263917e-06,
+      "loss": 0.4954,
+      "step": 426
+    },
+    {
+      "epoch": 0.4661572052401747,
+      "grad_norm": 1.6743052467665012,
+      "learning_rate": 5.516261527623391e-06,
+      "loss": 0.5058,
+      "step": 427
+    },
+    {
+      "epoch": 0.4672489082969432,
+      "grad_norm": 1.4149693200756928,
+      "learning_rate": 5.5057200422473964e-06,
+      "loss": 0.5571,
+      "step": 428
+    },
+    {
+      "epoch": 0.4683406113537118,
+      "grad_norm": 1.377375482313666,
+      "learning_rate": 5.49517212618052e-06,
+      "loss": 0.5241,
+      "step": 429
+    },
+    {
+      "epoch": 0.46943231441048033,
+      "grad_norm": 1.7214155025614608,
+      "learning_rate": 5.48461791354912e-06,
+      "loss": 0.5233,
+      "step": 430
+    },
+    {
+      "epoch": 0.4705240174672489,
+      "grad_norm": 1.4624248908282391,
+      "learning_rate": 5.474057538559622e-06,
+      "loss": 0.7009,
+      "step": 431
+    },
+    {
+      "epoch": 0.47161572052401746,
+      "grad_norm": 1.3755085340809183,
+      "learning_rate": 5.463491135496809e-06,
+      "loss": 0.559,
+      "step": 432
+    },
+    {
+      "epoch": 0.47270742358078605,
+      "grad_norm": 1.8480554892919654,
+      "learning_rate": 5.452918838722122e-06,
+      "loss": 0.5358,
+      "step": 433
+    },
+    {
+      "epoch": 0.4737991266375546,
+      "grad_norm": 1.370591853187486,
+      "learning_rate": 5.44234078267194e-06,
+      "loss": 0.4598,
+      "step": 434
+    },
+    {
+      "epoch": 0.47489082969432317,
+      "grad_norm": 1.4652044355564495,
+      "learning_rate": 5.431757101855878e-06,
+      "loss": 0.5229,
+      "step": 435
+    },
+    {
+      "epoch": 0.4759825327510917,
+      "grad_norm": 1.6442082838896726,
+      "learning_rate": 5.421167930855078e-06,
+      "loss": 0.4233,
+      "step": 436
+    },
+    {
+      "epoch": 0.47707423580786024,
+      "grad_norm": 1.4443466098404825,
+      "learning_rate": 5.410573404320491e-06,
+      "loss": 0.5414,
+      "step": 437
+    },
+    {
+      "epoch": 0.4781659388646288,
+      "grad_norm": 1.3720527302642247,
+      "learning_rate": 5.399973656971172e-06,
+      "loss": 0.5629,
+      "step": 438
+    },
+    {
+      "epoch": 0.47925764192139736,
+      "grad_norm": 1.420619258102723,
+      "learning_rate": 5.3893688235925585e-06,
+      "loss": 0.5166,
+      "step": 439
+    },
+    {
+      "epoch": 0.48034934497816595,
+      "grad_norm": 1.5277571440758837,
+      "learning_rate": 5.378759039034766e-06,
+      "loss": 0.554,
+      "step": 440
+    },
+    {
+      "epoch": 0.4814410480349345,
+      "grad_norm": 1.4638317307089392,
+      "learning_rate": 5.368144438210867e-06,
+      "loss": 0.4772,
+      "step": 441
+    },
+    {
+      "epoch": 0.48253275109170307,
+      "grad_norm": 1.5797672430691585,
+      "learning_rate": 5.357525156095178e-06,
+      "loss": 0.5166,
+      "step": 442
+    },
+    {
+      "epoch": 0.4836244541484716,
+      "grad_norm": 1.8314646046493335,
+      "learning_rate": 5.346901327721542e-06,
+      "loss": 0.5416,
+      "step": 443
+    },
+    {
+      "epoch": 0.4847161572052402,
+      "grad_norm": 1.6655392592879021,
+      "learning_rate": 5.3362730881816105e-06,
+      "loss": 0.5114,
+      "step": 444
+    },
+    {
+      "epoch": 0.48580786026200873,
+      "grad_norm": 1.717857275474268,
+      "learning_rate": 5.325640572623128e-06,
+      "loss": 0.5098,
+      "step": 445
+    },
+    {
+      "epoch": 0.4868995633187773,
+      "grad_norm": 1.3491745179920687,
+      "learning_rate": 5.315003916248214e-06,
+      "loss": 0.4975,
+      "step": 446
+    },
+    {
+      "epoch": 0.48799126637554585,
+      "grad_norm": 1.2024838454094096,
+      "learning_rate": 5.304363254311639e-06,
+      "loss": 0.4725,
+      "step": 447
+    },
+    {
+      "epoch": 0.4890829694323144,
+      "grad_norm": 1.6943727192213853,
+      "learning_rate": 5.2937187221191095e-06,
+      "loss": 0.6957,
+      "step": 448
+    },
+    {
+      "epoch": 0.490174672489083,
+      "grad_norm": 1.4214839188213413,
+      "learning_rate": 5.283070455025549e-06,
+      "loss": 0.5138,
+      "step": 449
+    },
+    {
+      "epoch": 0.4912663755458515,
+      "grad_norm": 1.7874547320870642,
+      "learning_rate": 5.272418588433366e-06,
+      "loss": 0.4976,
+      "step": 450
+    },
+    {
+      "epoch": 0.4923580786026201,
+      "grad_norm": 1.6025810609170663,
+      "learning_rate": 5.26176325779075e-06,
+      "loss": 0.4854,
+      "step": 451
+    },
+    {
+      "epoch": 0.49344978165938863,
+      "grad_norm": 1.3537064325091892,
+      "learning_rate": 5.25110459858993e-06,
+      "loss": 0.5205,
+      "step": 452
+    },
+    {
+      "epoch": 0.4945414847161572,
+      "grad_norm": 1.4501493604145486,
+      "learning_rate": 5.240442746365465e-06,
+      "loss": 0.7607,
+      "step": 453
+    },
+    {
+      "epoch": 0.49563318777292575,
+      "grad_norm": 1.4215275546800406,
+      "learning_rate": 5.229777836692516e-06,
+      "loss": 0.4881,
+      "step": 454
+    },
+    {
+      "epoch": 0.49672489082969434,
+      "grad_norm": 1.6725560271729976,
+      "learning_rate": 5.219110005185121e-06,
+      "loss": 0.7687,
+      "step": 455
+    },
+    {
+      "epoch": 0.4978165938864629,
+      "grad_norm": 1.5074766611660562,
+      "learning_rate": 5.208439387494473e-06,
+      "loss": 0.6112,
+      "step": 456
+    },
+    {
+      "epoch": 0.49890829694323147,
+      "grad_norm": 1.3597839340846551,
+      "learning_rate": 5.197766119307193e-06,
+      "loss": 0.4853,
+      "step": 457
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 1.6691192782977187,
+      "learning_rate": 5.187090336343605e-06,
+      "loss": 0.5366,
+      "step": 458
+    },
+    {
+      "epoch": 0.5010917030567685,
+      "grad_norm": 1.2963947269544678,
+      "learning_rate": 5.176412174356015e-06,
+      "loss": 0.4896,
+      "step": 459
+    },
+    {
+      "epoch": 0.5021834061135371,
+      "grad_norm": 1.4474979354084274,
+      "learning_rate": 5.165731769126973e-06,
+      "loss": 0.4894,
+      "step": 460
+    },
+    {
+      "epoch": 0.5032751091703057,
+      "grad_norm": 1.4441187219316847,
+      "learning_rate": 5.15504925646756e-06,
+      "loss": 0.4403,
+      "step": 461
+    },
+    {
+      "epoch": 0.5043668122270742,
+      "grad_norm": 1.4514534998495667,
+      "learning_rate": 5.14436477221565e-06,
+      "loss": 0.637,
+      "step": 462
+    },
+    {
+      "epoch": 0.5054585152838428,
+      "grad_norm": 1.5553610264211595,
+      "learning_rate": 5.133678452234194e-06,
+      "loss": 0.4584,
+      "step": 463
+    },
+    {
+      "epoch": 0.5065502183406113,
+      "grad_norm": 1.6280569904422444,
+      "learning_rate": 5.122990432409481e-06,
+      "loss": 0.5366,
+      "step": 464
+    },
+    {
+      "epoch": 0.50764192139738,
+      "grad_norm": 1.6171411741302026,
+      "learning_rate": 5.112300848649417e-06,
+      "loss": 0.5351,
+      "step": 465
+    },
+    {
+      "epoch": 0.5087336244541485,
+      "grad_norm": 1.5642871863260377,
+      "learning_rate": 5.101609836881791e-06,
+      "loss": 0.5112,
+      "step": 466
+    },
+    {
+      "epoch": 0.509825327510917,
+      "grad_norm": 1.3618839192737355,
+      "learning_rate": 5.0909175330525565e-06,
+      "loss": 0.5649,
+      "step": 467
+    },
+    {
+      "epoch": 0.5109170305676856,
+      "grad_norm": 1.4461003640687564,
+      "learning_rate": 5.080224073124093e-06,
+      "loss": 0.4662,
+      "step": 468
+    },
+    {
+      "epoch": 0.5120087336244541,
+      "grad_norm": 1.448092639750284,
+      "learning_rate": 5.069529593073484e-06,
+      "loss": 0.5392,
+      "step": 469
+    },
+    {
+      "epoch": 0.5131004366812227,
+      "grad_norm": 1.4766768695379655,
+      "learning_rate": 5.058834228890778e-06,
+      "loss": 0.5216,
+      "step": 470
+    },
+    {
+      "epoch": 0.5141921397379913,
+      "grad_norm": 1.388192539641165,
+      "learning_rate": 5.048138116577276e-06,
+      "loss": 0.5121,
+      "step": 471
+    },
+    {
+      "epoch": 0.5152838427947598,
+      "grad_norm": 1.442377666445445,
+      "learning_rate": 5.03744139214378e-06,
+      "loss": 0.4964,
+      "step": 472
+    },
+    {
+      "epoch": 0.5163755458515283,
+      "grad_norm": 1.4380858560991479,
+      "learning_rate": 5.026744191608888e-06,
+      "loss": 0.6296,
+      "step": 473
+    },
+    {
+      "epoch": 0.517467248908297,
+      "grad_norm": 1.4066071827611244,
+      "learning_rate": 5.0160466509972444e-06,
+      "loss": 0.5366,
+      "step": 474
+    },
+    {
+      "epoch": 0.5185589519650655,
+      "grad_norm": 1.4771695892234902,
+      "learning_rate": 5.005348906337819e-06,
+      "loss": 0.4818,
+      "step": 475
+    },
+    {
+      "epoch": 0.519650655021834,
+      "grad_norm": 1.3482597250692996,
+      "learning_rate": 4.994651093662181e-06,
+      "loss": 0.621,
+      "step": 476
+    },
+    {
+      "epoch": 0.5207423580786026,
+      "grad_norm": 1.4656112769819456,
+      "learning_rate": 4.9839533490027555e-06,
+      "loss": 0.5968,
+      "step": 477
+    },
+    {
+      "epoch": 0.5218340611353712,
+      "grad_norm": 1.4194678586304768,
+      "learning_rate": 4.973255808391112e-06,
+      "loss": 0.4855,
+      "step": 478
+    },
+    {
+      "epoch": 0.5229257641921398,
+      "grad_norm": 1.3385237861754027,
+      "learning_rate": 4.96255860785622e-06,
+      "loss": 0.446,
+      "step": 479
+    },
+    {
+      "epoch": 0.5240174672489083,
+      "grad_norm": 1.4199997853851958,
+      "learning_rate": 4.951861883422725e-06,
+      "loss": 0.5232,
+      "step": 480
+    },
+    {
+      "epoch": 0.5251091703056768,
+      "grad_norm": 1.4504354793413279,
+      "learning_rate": 4.941165771109221e-06,
+      "loss": 0.5941,
+      "step": 481
+    },
+    {
+      "epoch": 0.5262008733624454,
+      "grad_norm": 1.2469715040512819,
+      "learning_rate": 4.930470406926517e-06,
+      "loss": 0.5222,
+      "step": 482
+    },
+    {
+      "epoch": 0.527292576419214,
+      "grad_norm": 1.2822739519744986,
+      "learning_rate": 4.919775926875906e-06,
+      "loss": 0.4774,
+      "step": 483
+    },
+    {
+      "epoch": 0.5283842794759825,
+      "grad_norm": 1.5801715611186389,
+      "learning_rate": 4.909082466947443e-06,
+      "loss": 0.4725,
+      "step": 484
+    },
+    {
+      "epoch": 0.5294759825327511,
+      "grad_norm": 1.474196427294419,
+      "learning_rate": 4.898390163118207e-06,
+      "loss": 0.5055,
+      "step": 485
+    },
+    {
+      "epoch": 0.5305676855895196,
+      "grad_norm": 1.2140076028979254,
+      "learning_rate": 4.887699151350585e-06,
+      "loss": 0.5698,
+      "step": 486
+    },
+    {
+      "epoch": 0.5316593886462883,
+      "grad_norm": 1.55426631015283,
+      "learning_rate": 4.877009567590519e-06,
+      "loss": 0.5059,
+      "step": 487
+    },
+    {
+      "epoch": 0.5327510917030568,
+      "grad_norm": 1.0298037690986848,
+      "learning_rate": 4.866321547765805e-06,
+      "loss": 0.5384,
+      "step": 488
+    },
+    {
+      "epoch": 0.5338427947598253,
+      "grad_norm": 1.4188304779909608,
+      "learning_rate": 4.855635227784349e-06,
+      "loss": 0.509,
+      "step": 489
+    },
+    {
+      "epoch": 0.5349344978165939,
+      "grad_norm": 1.3254069119369634,
+      "learning_rate": 4.8449507435324414e-06,
+      "loss": 0.4632,
+      "step": 490
+    },
+    {
+      "epoch": 0.5360262008733624,
+      "grad_norm": 1.5099230232121612,
+      "learning_rate": 4.834268230873028e-06,
+      "loss": 0.4655,
+      "step": 491
+    },
+    {
+      "epoch": 0.537117903930131,
+      "grad_norm": 1.6476817572317615,
+      "learning_rate": 4.823587825643985e-06,
+      "loss": 0.4971,
+      "step": 492
+    },
+    {
+      "epoch": 0.5382096069868996,
+      "grad_norm": 1.3898484269183884,
+      "learning_rate": 4.812909663656393e-06,
+      "loss": 0.5795,
+      "step": 493
+    },
+    {
+      "epoch": 0.5393013100436681,
+      "grad_norm": 1.3882570065353528,
+      "learning_rate": 4.802233880692807e-06,
+      "loss": 0.4848,
+      "step": 494
+    },
+    {
+      "epoch": 0.5403930131004366,
+      "grad_norm": 1.28772744355436,
+      "learning_rate": 4.791560612505528e-06,
+      "loss": 0.5398,
+      "step": 495
+    },
+    {
+      "epoch": 0.5414847161572053,
+      "grad_norm": 1.368216878184212,
+      "learning_rate": 4.780889994814879e-06,
+      "loss": 0.5512,
+      "step": 496
+    },
+    {
+      "epoch": 0.5425764192139738,
+      "grad_norm": 1.4144230061626577,
+      "learning_rate": 4.770222163307484e-06,
+      "loss": 0.4963,
+      "step": 497
+    },
+    {
+      "epoch": 0.5436681222707423,
+      "grad_norm": 1.3794044239685685,
+      "learning_rate": 4.759557253634534e-06,
+      "loss": 0.4659,
+      "step": 498
+    },
+    {
+      "epoch": 0.5447598253275109,
+      "grad_norm": 1.524418837426529,
+      "learning_rate": 4.748895401410071e-06,
+      "loss": 0.5042,
+      "step": 499
+    },
+    {
+      "epoch": 0.5458515283842795,
+      "grad_norm": 1.6044354987036333,
+      "learning_rate": 4.738236742209249e-06,
+      "loss": 0.4827,
+      "step": 500
+    },
+    {
+      "epoch": 0.5469432314410481,
+      "grad_norm": 1.5500030376180924,
+      "learning_rate": 4.727581411566632e-06,
+      "loss": 0.5257,
+      "step": 501
+    },
+    {
+      "epoch": 0.5480349344978166,
+      "grad_norm": 1.4421277932778935,
+      "learning_rate": 4.716929544974451e-06,
+      "loss": 0.5487,
+      "step": 502
+    },
+    {
+      "epoch": 0.5491266375545851,
+      "grad_norm": 1.5990643037111425,
+      "learning_rate": 4.7062812778808904e-06,
+      "loss": 0.5438,
+      "step": 503
+    },
+    {
+      "epoch": 0.5502183406113537,
+      "grad_norm": 1.4502389607629171,
+      "learning_rate": 4.695636745688361e-06,
+      "loss": 0.4929,
+      "step": 504
+    },
+    {
+      "epoch": 0.5513100436681223,
+      "grad_norm": 1.3204441944279035,
+      "learning_rate": 4.684996083751785e-06,
+      "loss": 0.5448,
+      "step": 505
+    },
+    {
+      "epoch": 0.5524017467248908,
+      "grad_norm": 1.4355806504382989,
+      "learning_rate": 4.674359427376872e-06,
+      "loss": 0.5551,
+      "step": 506
+    },
+    {
+      "epoch": 0.5534934497816594,
+      "grad_norm": 1.3692565246056831,
+      "learning_rate": 4.66372691181839e-06,
+      "loss": 0.5209,
+      "step": 507
+    },
+    {
+      "epoch": 0.5545851528384279,
+      "grad_norm": 1.3879798213129202,
+      "learning_rate": 4.653098672278458e-06,
+      "loss": 0.5219,
+      "step": 508
+    },
+    {
+      "epoch": 0.5556768558951966,
+      "grad_norm": 1.481778083844796,
+      "learning_rate": 4.6424748439048204e-06,
+      "loss": 0.5543,
+      "step": 509
+    },
+    {
+      "epoch": 0.5567685589519651,
+      "grad_norm": 1.7087241965837792,
+      "learning_rate": 4.631855561789132e-06,
+      "loss": 0.5794,
+      "step": 510
+    },
+    {
+      "epoch": 0.5578602620087336,
+      "grad_norm": 1.477176476343145,
+      "learning_rate": 4.621240960965233e-06,
+      "loss": 0.5764,
+      "step": 511
+    },
+    {
+      "epoch": 0.5589519650655022,
+      "grad_norm": 1.3365328113829895,
+      "learning_rate": 4.610631176407442e-06,
+      "loss": 0.4716,
+      "step": 512
+    },
+    {
+      "epoch": 0.5600436681222707,
+      "grad_norm": 1.3617920239093109,
+      "learning_rate": 4.600026343028828e-06,
+      "loss": 0.4821,
+      "step": 513
+    },
+    {
+      "epoch": 0.5611353711790393,
+      "grad_norm": 1.4742907544937358,
+      "learning_rate": 4.589426595679509e-06,
+      "loss": 0.5276,
+      "step": 514
+    },
+    {
+      "epoch": 0.5622270742358079,
+      "grad_norm": 1.2133762965906438,
+      "learning_rate": 4.578832069144922e-06,
+      "loss": 0.554,
+      "step": 515
+    },
+    {
+      "epoch": 0.5633187772925764,
+      "grad_norm": 1.3499071499642659,
+      "learning_rate": 4.568242898144121e-06,
+      "loss": 0.5079,
+      "step": 516
+    },
+    {
+      "epoch": 0.5644104803493449,
+      "grad_norm": 1.716937961960104,
+      "learning_rate": 4.55765921732806e-06,
+      "loss": 0.5128,
+      "step": 517
+    },
+    {
+      "epoch": 0.5655021834061136,
+      "grad_norm": 1.4301074339999107,
+      "learning_rate": 4.547081161277878e-06,
+      "loss": 0.4896,
+      "step": 518
+    },
+    {
+      "epoch": 0.5665938864628821,
+      "grad_norm": 1.470837341341121,
+      "learning_rate": 4.53650886450319e-06,
+      "loss": 0.5082,
+      "step": 519
+    },
+    {
+      "epoch": 0.5676855895196506,
+      "grad_norm": 1.2685205731280647,
+      "learning_rate": 4.5259424614403774e-06,
+      "loss": 0.5761,
+      "step": 520
+    },
+    {
+      "epoch": 0.5687772925764192,
+      "grad_norm": 1.5260496629984845,
+      "learning_rate": 4.515382086450879e-06,
+      "loss": 0.5009,
+      "step": 521
+    },
+    {
+      "epoch": 0.5698689956331878,
+      "grad_norm": 1.3553156778244764,
+      "learning_rate": 4.50482787381948e-06,
+      "loss": 0.4602,
+      "step": 522
+    },
+    {
+      "epoch": 0.5709606986899564,
+      "grad_norm": 1.5048924602650442,
+      "learning_rate": 4.4942799577526035e-06,
+      "loss": 0.6232,
+      "step": 523
+    },
+    {
+      "epoch": 0.5720524017467249,
+      "grad_norm": 1.4163718659252338,
+      "learning_rate": 4.483738472376609e-06,
+      "loss": 0.595,
+      "step": 524
+    },
+    {
+      "epoch": 0.5731441048034934,
+      "grad_norm": 1.256504078133773,
+      "learning_rate": 4.473203551736082e-06,
+      "loss": 0.4841,
+      "step": 525
+    },
+    {
+      "epoch": 0.574235807860262,
+      "grad_norm": 1.8309113360149538,
+      "learning_rate": 4.462675329792135e-06,
+      "loss": 0.5268,
+      "step": 526
+    },
+    {
+      "epoch": 0.5753275109170306,
+      "grad_norm": 1.4616749715688826,
+      "learning_rate": 4.4521539404206955e-06,
+      "loss": 0.5201,
+      "step": 527
+    },
+    {
+      "epoch": 0.5764192139737991,
+      "grad_norm": 1.4866810995067377,
+      "learning_rate": 4.44163951741081e-06,
+      "loss": 0.4819,
+      "step": 528
+    },
+    {
+      "epoch": 0.5775109170305677,
+      "grad_norm": 1.2959474071911306,
+      "learning_rate": 4.431132194462945e-06,
+      "loss": 0.4599,
+      "step": 529
+    },
+    {
+      "epoch": 0.5786026200873362,
+      "grad_norm": 1.503165783836535,
+      "learning_rate": 4.4206321051872785e-06,
+      "loss": 0.5553,
+      "step": 530
+    },
+    {
+      "epoch": 0.5796943231441049,
+      "grad_norm": 1.4738805985403882,
+      "learning_rate": 4.410139383102009e-06,
+      "loss": 0.4724,
+      "step": 531
+    },
+    {
+      "epoch": 0.5807860262008734,
+      "grad_norm": 1.331771857702602,
+      "learning_rate": 4.3996541616316516e-06,
+      "loss": 0.4447,
+      "step": 532
+    },
+    {
+      "epoch": 0.5818777292576419,
+      "grad_norm": 1.4673541177247749,
+      "learning_rate": 4.3891765741053465e-06,
+      "loss": 0.6408,
+      "step": 533
+    },
+    {
+      "epoch": 0.5829694323144105,
+      "grad_norm": 1.4020555533377803,
+      "learning_rate": 4.378706753755162e-06,
+      "loss": 0.5344,
+      "step": 534
+    },
+    {
+      "epoch": 0.584061135371179,
+      "grad_norm": 1.6218461769836316,
+      "learning_rate": 4.368244833714397e-06,
+      "loss": 0.5655,
+      "step": 535
+    },
+    {
+      "epoch": 0.5851528384279476,
+      "grad_norm": 1.4187950178971758,
+      "learning_rate": 4.357790947015893e-06,
+      "loss": 0.5704,
+      "step": 536
+    },
+    {
+      "epoch": 0.5862445414847162,
+      "grad_norm": 1.7413424456473428,
+      "learning_rate": 4.34734522659034e-06,
+      "loss": 0.4735,
+      "step": 537
+    },
+    {
+      "epoch": 0.5873362445414847,
+      "grad_norm": 1.4967102117444473,
+      "learning_rate": 4.3369078052645845e-06,
+      "loss": 0.4856,
+      "step": 538
+    },
+    {
+      "epoch": 0.5884279475982532,
+      "grad_norm": 1.3591454057028716,
+      "learning_rate": 4.3264788157599435e-06,
+      "loss": 0.6019,
+      "step": 539
+    },
+    {
+      "epoch": 0.5895196506550219,
+      "grad_norm": 1.4646999752701761,
+      "learning_rate": 4.316058390690517e-06,
+      "loss": 0.6017,
+      "step": 540
+    },
+    {
+      "epoch": 0.5906113537117904,
+      "grad_norm": 1.1539983694011575,
+      "learning_rate": 4.305646662561498e-06,
+      "loss": 0.5945,
+      "step": 541
+    },
+    {
+      "epoch": 0.5917030567685589,
+      "grad_norm": 1.499171453680021,
+      "learning_rate": 4.2952437637674944e-06,
+      "loss": 0.6161,
+      "step": 542
+    },
+    {
+      "epoch": 0.5927947598253275,
+      "grad_norm": 1.2754209219978767,
+      "learning_rate": 4.284849826590835e-06,
+      "loss": 0.4973,
+      "step": 543
+    },
+    {
+      "epoch": 0.5938864628820961,
+      "grad_norm": 1.6027113994625042,
+      "learning_rate": 4.2744649831998964e-06,
+      "loss": 0.5622,
+      "step": 544
+    },
+    {
+      "epoch": 0.5949781659388647,
+      "grad_norm": 1.5148520907544383,
+      "learning_rate": 4.264089365647421e-06,
+      "loss": 0.5654,
+      "step": 545
+    },
+    {
+      "epoch": 0.5960698689956332,
+      "grad_norm": 1.409327342683608,
+      "learning_rate": 4.253723105868835e-06,
+      "loss": 0.5958,
+      "step": 546
+    },
+    {
+      "epoch": 0.5971615720524017,
+      "grad_norm": 1.432267443294078,
+      "learning_rate": 4.24336633568057e-06,
+      "loss": 0.4894,
+      "step": 547
+    },
+    {
+      "epoch": 0.5982532751091703,
+      "grad_norm": 1.4322901416716356,
+      "learning_rate": 4.233019186778392e-06,
+      "loss": 0.4671,
+      "step": 548
+    },
+    {
+      "epoch": 0.5993449781659389,
+      "grad_norm": 1.4857292999991856,
+      "learning_rate": 4.222681790735721e-06,
+      "loss": 0.4431,
+      "step": 549
+    },
+    {
+      "epoch": 0.6004366812227074,
+      "grad_norm": 1.5103278913268807,
+      "learning_rate": 4.212354279001963e-06,
+      "loss": 0.5286,
+      "step": 550
+    },
+    {
+      "epoch": 0.601528384279476,
+      "grad_norm": 1.3661166974305412,
+      "learning_rate": 4.2020367829008335e-06,
+      "loss": 0.495,
+      "step": 551
+    },
+    {
+      "epoch": 0.6026200873362445,
+      "grad_norm": 1.4657473729106474,
+      "learning_rate": 4.191729433628692e-06,
+      "loss": 0.4859,
+      "step": 552
+    },
+    {
+      "epoch": 0.6037117903930131,
+      "grad_norm": 1.37621037582073,
+      "learning_rate": 4.1814323622528704e-06,
+      "loss": 0.5261,
+      "step": 553
+    },
+    {
+      "epoch": 0.6048034934497817,
+      "grad_norm": 1.8525067725209052,
+      "learning_rate": 4.17114569971001e-06,
+      "loss": 0.6356,
+      "step": 554
+    },
+    {
+      "epoch": 0.6058951965065502,
+      "grad_norm": 1.3676971383382535,
+      "learning_rate": 4.160869576804391e-06,
+      "loss": 0.4843,
+      "step": 555
+    },
+    {
+      "epoch": 0.6069868995633187,
+      "grad_norm": 1.454872378059662,
+      "learning_rate": 4.150604124206275e-06,
+      "loss": 0.5104,
+      "step": 556
+    },
+    {
+      "epoch": 0.6080786026200873,
+      "grad_norm": 1.6090732359643547,
+      "learning_rate": 4.140349472450239e-06,
+      "loss": 0.4617,
+      "step": 557
+    },
+    {
+      "epoch": 0.6091703056768559,
+      "grad_norm": 1.3513538232854985,
+      "learning_rate": 4.1301057519335225e-06,
+      "loss": 0.4536,
+      "step": 558
+    },
+    {
+      "epoch": 0.6102620087336245,
+      "grad_norm": 1.4626389154409178,
+      "learning_rate": 4.119873092914357e-06,
+      "loss": 0.4666,
+      "step": 559
+    },
+    {
+      "epoch": 0.611353711790393,
+      "grad_norm": 1.5581131305282525,
+      "learning_rate": 4.10965162551032e-06,
+      "loss": 0.4777,
+      "step": 560
+    },
+    {
+      "epoch": 0.611353711790393,
+      "eval_accuracy": 0.8165034143027314,
+      "eval_accuracy_first_token": 0.7887080643694881,
+      "eval_accuracy_first_token_<": 0.9859781121751026,
+      "eval_accuracy_first_token_<_total": 2924,
+      "eval_accuracy_first_token_<|python_tag|>": 0.9255533199195171,
+      "eval_accuracy_first_token_<|python_tag|>_total": 994,
+      "eval_accuracy_first_token_Here": 0.572538860103627,
+      "eval_accuracy_first_token_Here_total": 386,
+      "eval_accuracy_first_token_The": 0.8994814519345832,
+      "eval_accuracy_first_token_The_total": 2507,
+      "eval_accuracy_first_token_To": 0.7873931623931624,
+      "eval_accuracy_first_token_To_total": 936,
+      "eval_first_token_param_values": 0.9387768727466952,
+      "eval_first_token_param_values_total": 14978,
+      "eval_loss": 0.5781970620155334,
+      "eval_perplexity": 1.1114759783723378,
+      "eval_runtime": 391.1056,
+      "eval_samples_per_second": 1.982,
+      "eval_steps_per_second": 0.248,
+      "eval_total_number_first_token": 10999,
+      "step": 560
+    },
+    {
+      "epoch": 0.6124454148471615,
+      "grad_norm": 1.582252798674935,
+      "learning_rate": 4.099441479696681e-06,
+      "loss": 0.4602,
+      "step": 561
+    },
+    {
+      "epoch": 0.6135371179039302,
+      "grad_norm": 1.608295893756117,
+      "learning_rate": 4.089242785304742e-06,
+      "loss": 0.5076,
+      "step": 562
+    },
+    {
+      "epoch": 0.6146288209606987,
+      "grad_norm": 1.3485004651056143,
+      "learning_rate": 4.079055672020186e-06,
+      "loss": 0.4661,
+      "step": 563
+    },
+    {
+      "epoch": 0.6157205240174672,
+      "grad_norm": 1.291513507710692,
+      "learning_rate": 4.06888026938144e-06,
+      "loss": 0.5642,
+      "step": 564
+    },
+    {
+      "epoch": 0.6168122270742358,
+      "grad_norm": 1.3170533583485662,
+      "learning_rate": 4.058716706778011e-06,
+      "loss": 0.5438,
+      "step": 565
+    },
+    {
+      "epoch": 0.6179039301310044,
+      "grad_norm": 1.4441714827082741,
+      "learning_rate": 4.048565113448857e-06,
+      "loss": 0.5078,
+      "step": 566
+    },
+    {
+      "epoch": 0.618995633187773,
+      "grad_norm": 1.6783163773061296,
+      "learning_rate": 4.038425618480729e-06,
+      "loss": 0.5576,
+      "step": 567
+    },
+    {
+      "epoch": 0.6200873362445415,
+      "grad_norm": 1.4188507978385323,
+      "learning_rate": 4.02829835080654e-06,
+      "loss": 0.5537,
+      "step": 568
+    },
+    {
+      "epoch": 0.62117903930131,
+      "grad_norm": 1.3779540707380475,
+      "learning_rate": 4.01818343920372e-06,
+      "loss": 0.4405,
+      "step": 569
+    },
+    {
+      "epoch": 0.6222707423580786,
+      "grad_norm": 1.4499236635379809,
+      "learning_rate": 4.008081012292587e-06,
+      "loss": 0.4844,
+      "step": 570
+    },
+    {
+      "epoch": 0.6233624454148472,
+      "grad_norm": 1.523491954701273,
+      "learning_rate": 3.997991198534692e-06,
+      "loss": 0.5044,
+      "step": 571
+    },
+    {
+      "epoch": 0.6244541484716157,
+      "grad_norm": 1.4598644387268775,
+      "learning_rate": 3.98791412623121e-06,
+      "loss": 0.5019,
+      "step": 572
+    },
+    {
+      "epoch": 0.6255458515283843,
+      "grad_norm": 1.4990035332267648,
+      "learning_rate": 3.977849923521289e-06,
+      "loss": 0.5348,
+      "step": 573
+    },
+    {
+      "epoch": 0.6266375545851528,
+      "grad_norm": 1.5319645001618205,
+      "learning_rate": 3.967798718380435e-06,
+      "loss": 0.4354,
+      "step": 574
+    },
+    {
+      "epoch": 0.6277292576419214,
+      "grad_norm": 1.3922243251918538,
+      "learning_rate": 3.957760638618872e-06,
+      "loss": 0.4926,
+      "step": 575
+    },
+    {
+      "epoch": 0.62882096069869,
+      "grad_norm": 1.533896226358977,
+      "learning_rate": 3.9477358118799265e-06,
+      "loss": 0.4322,
+      "step": 576
+    },
+    {
+      "epoch": 0.6299126637554585,
+      "grad_norm": 1.422842914473825,
+      "learning_rate": 3.9377243656384e-06,
+      "loss": 0.4712,
+      "step": 577
+    },
+    {
+      "epoch": 0.631004366812227,
+      "grad_norm": 1.2246188538633846,
+      "learning_rate": 3.927726427198949e-06,
+      "loss": 0.4364,
+      "step": 578
+    },
+    {
+      "epoch": 0.6320960698689956,
+      "grad_norm": 1.4664888990426166,
+      "learning_rate": 3.917742123694466e-06,
+      "loss": 0.4323,
+      "step": 579
+    },
+    {
+      "epoch": 0.6331877729257642,
+      "grad_norm": 1.310290618069973,
+      "learning_rate": 3.907771582084459e-06,
+      "loss": 0.5143,
+      "step": 580
+    },
+    {
+      "epoch": 0.6342794759825328,
+      "grad_norm": 1.372424891579117,
+      "learning_rate": 3.897814929153449e-06,
+      "loss": 0.4977,
+      "step": 581
+    },
+    {
+      "epoch": 0.6353711790393013,
+      "grad_norm": 1.5281095237236761,
+      "learning_rate": 3.887872291509343e-06,
+      "loss": 0.5587,
+      "step": 582
+    },
+    {
+      "epoch": 0.6364628820960698,
+      "grad_norm": 1.3721835230254955,
+      "learning_rate": 3.877943795581833e-06,
+      "loss": 0.4939,
+      "step": 583
+    },
+    {
+      "epoch": 0.6375545851528385,
+      "grad_norm": 1.388427881647579,
+      "learning_rate": 3.868029567620787e-06,
+      "loss": 0.4247,
+      "step": 584
+    },
+    {
+      "epoch": 0.638646288209607,
+      "grad_norm": 1.497847291110233,
+      "learning_rate": 3.858129733694641e-06,
+      "loss": 0.5298,
+      "step": 585
+    },
+    {
+      "epoch": 0.6397379912663755,
+      "grad_norm": 1.4255242399699755,
+      "learning_rate": 3.848244419688801e-06,
+      "loss": 0.5293,
+      "step": 586
+    },
+    {
+      "epoch": 0.6408296943231441,
+      "grad_norm": 1.4552294636702017,
+      "learning_rate": 3.838373751304035e-06,
+      "loss": 0.4229,
+      "step": 587
+    },
+    {
+      "epoch": 0.6419213973799127,
+      "grad_norm": 1.626066546172521,
+      "learning_rate": 3.828517854054881e-06,
+      "loss": 0.5642,
+      "step": 588
+    },
+    {
+      "epoch": 0.6430131004366813,
+      "grad_norm": 1.456810943997373,
+      "learning_rate": 3.818676853268048e-06,
+      "loss": 0.5599,
+      "step": 589
+    },
+    {
+      "epoch": 0.6441048034934498,
+      "grad_norm": 1.4719729624676086,
+      "learning_rate": 3.8088508740808243e-06,
+      "loss": 0.5027,
+      "step": 590
+    },
+    {
+      "epoch": 0.6451965065502183,
+      "grad_norm": 1.5160593159695923,
+      "learning_rate": 3.79904004143948e-06,
+      "loss": 0.5762,
+      "step": 591
+    },
+    {
+      "epoch": 0.6462882096069869,
+      "grad_norm": 1.5202694953006204,
+      "learning_rate": 3.7892444800976896e-06,
+      "loss": 0.4658,
+      "step": 592
+    },
+    {
+      "epoch": 0.6473799126637555,
+      "grad_norm": 1.665985658035627,
+      "learning_rate": 3.7794643146149357e-06,
+      "loss": 0.5121,
+      "step": 593
+    },
+    {
+      "epoch": 0.648471615720524,
+      "grad_norm": 1.4305093245291967,
+      "learning_rate": 3.769699669354928e-06,
+      "loss": 0.5636,
+      "step": 594
+    },
+    {
+      "epoch": 0.6495633187772926,
+      "grad_norm": 1.3175981932047456,
+      "learning_rate": 3.759950668484024e-06,
+      "loss": 0.5314,
+      "step": 595
+    },
+    {
+      "epoch": 0.6506550218340611,
+      "grad_norm": 1.4531299264572408,
+      "learning_rate": 3.750217435969648e-06,
+      "loss": 0.4879,
+      "step": 596
+    },
+    {
+      "epoch": 0.6517467248908297,
+      "grad_norm": 1.4355775633746033,
+      "learning_rate": 3.7405000955787146e-06,
+      "loss": 0.4265,
+      "step": 597
+    },
+    {
+      "epoch": 0.6528384279475983,
+      "grad_norm": 1.3377384511360848,
+      "learning_rate": 3.730798770876057e-06,
+      "loss": 0.5584,
+      "step": 598
+    },
+    {
+      "epoch": 0.6539301310043668,
+      "grad_norm": 1.3618246171688768,
+      "learning_rate": 3.721113585222853e-06,
+      "loss": 0.5308,
+      "step": 599
+    },
+    {
+      "epoch": 0.6550218340611353,
+      "grad_norm": 1.5237374024272177,
+      "learning_rate": 3.711444661775057e-06,
+      "loss": 0.5637,
+      "step": 600
+    },
+    {
+      "epoch": 0.6561135371179039,
+      "grad_norm": 1.4476718647379752,
+      "learning_rate": 3.7017921234818383e-06,
+      "loss": 0.5441,
+      "step": 601
+    },
+    {
+      "epoch": 0.6572052401746725,
+      "grad_norm": 1.5976684200512656,
+      "learning_rate": 3.6921560930840103e-06,
+      "loss": 0.5968,
+      "step": 602
+    },
+    {
+      "epoch": 0.6582969432314411,
+      "grad_norm": 1.4849123279512841,
+      "learning_rate": 3.6825366931124744e-06,
+      "loss": 0.4681,
+      "step": 603
+    },
+    {
+      "epoch": 0.6593886462882096,
+      "grad_norm": 1.4283815786057126,
+      "learning_rate": 3.672934045886661e-06,
+      "loss": 0.5145,
+      "step": 604
+    },
+    {
+      "epoch": 0.6604803493449781,
+      "grad_norm": 1.4467852106957109,
+      "learning_rate": 3.6633482735129765e-06,
+      "loss": 0.5078,
+      "step": 605
+    },
+    {
+      "epoch": 0.6615720524017468,
+      "grad_norm": 1.3301330189214893,
+      "learning_rate": 3.6537794978832454e-06,
+      "loss": 0.4711,
+      "step": 606
+    },
+    {
+      "epoch": 0.6626637554585153,
+      "grad_norm": 1.2939570426381055,
+      "learning_rate": 3.6442278406731616e-06,
+      "loss": 0.5188,
+      "step": 607
+    },
+    {
+      "epoch": 0.6637554585152838,
+      "grad_norm": 1.4335427746020373,
+      "learning_rate": 3.6346934233407475e-06,
+      "loss": 0.4148,
+      "step": 608
+    },
+    {
+      "epoch": 0.6648471615720524,
+      "grad_norm": 1.312602515327031,
+      "learning_rate": 3.6251763671248022e-06,
+      "loss": 0.4515,
+      "step": 609
+    },
+    {
+      "epoch": 0.665938864628821,
+      "grad_norm": 1.4935409138139255,
+      "learning_rate": 3.615676793043361e-06,
+      "loss": 0.4728,
+      "step": 610
+    },
+    {
+      "epoch": 0.6670305676855895,
+      "grad_norm": 1.564394383484283,
+      "learning_rate": 3.6061948218921587e-06,
+      "loss": 0.5202,
+      "step": 611
+    },
+    {
+      "epoch": 0.6681222707423581,
+      "grad_norm": 1.420319000780053,
+      "learning_rate": 3.5967305742430954e-06,
+      "loss": 0.4943,
+      "step": 612
+    },
+    {
+      "epoch": 0.6692139737991266,
+      "grad_norm": 1.3312001561036553,
+      "learning_rate": 3.5872841704426983e-06,
+      "loss": 0.4934,
+      "step": 613
+    },
+    {
+      "epoch": 0.6703056768558951,
+      "grad_norm": 1.3965985189004924,
+      "learning_rate": 3.5778557306105945e-06,
+      "loss": 0.5391,
+      "step": 614
+    },
+    {
+      "epoch": 0.6713973799126638,
+      "grad_norm": 1.3680149448615597,
+      "learning_rate": 3.5684453746379806e-06,
+      "loss": 0.538,
+      "step": 615
+    },
+    {
+      "epoch": 0.6724890829694323,
+      "grad_norm": 1.542142361004548,
+      "learning_rate": 3.5590532221861053e-06,
+      "loss": 0.4204,
+      "step": 616
+    },
+    {
+      "epoch": 0.6735807860262009,
+      "grad_norm": 1.34324454974894,
+      "learning_rate": 3.5496793926847383e-06,
+      "loss": 0.5017,
+      "step": 617
+    },
+    {
+      "epoch": 0.6746724890829694,
+      "grad_norm": 1.3232757727120024,
+      "learning_rate": 3.540324005330661e-06,
+      "loss": 0.4585,
+      "step": 618
+    },
+    {
+      "epoch": 0.675764192139738,
+      "grad_norm": 1.416803667589595,
+      "learning_rate": 3.5309871790861385e-06,
+      "loss": 0.4689,
+      "step": 619
+    },
+    {
+      "epoch": 0.6768558951965066,
+      "grad_norm": 1.5089099592021908,
+      "learning_rate": 3.521669032677423e-06,
+      "loss": 0.4154,
+      "step": 620
+    },
+    {
+      "epoch": 0.6779475982532751,
+      "grad_norm": 1.85540232423345,
+      "learning_rate": 3.5123696845932294e-06,
+      "loss": 0.445,
+      "step": 621
+    },
+    {
+      "epoch": 0.6790393013100436,
+      "grad_norm": 1.4563985724787059,
+      "learning_rate": 3.5030892530832416e-06,
+      "loss": 0.5235,
+      "step": 622
+    },
+    {
+      "epoch": 0.6801310043668122,
+      "grad_norm": 1.3838318346145242,
+      "learning_rate": 3.4938278561565924e-06,
+      "loss": 0.6341,
+      "step": 623
+    },
+    {
+      "epoch": 0.6812227074235808,
+      "grad_norm": 1.31089716942976,
+      "learning_rate": 3.4845856115803815e-06,
+      "loss": 0.4908,
+      "step": 624
+    },
+    {
+      "epoch": 0.6823144104803494,
+      "grad_norm": 1.2353223624644207,
+      "learning_rate": 3.4753626368781625e-06,
+      "loss": 0.4287,
+      "step": 625
+    },
+    {
+      "epoch": 0.6834061135371179,
+      "grad_norm": 1.6062380899237767,
+      "learning_rate": 3.4661590493284615e-06,
+      "loss": 0.4917,
+      "step": 626
+    },
+    {
+      "epoch": 0.6844978165938864,
+      "grad_norm": 1.5759254996812047,
+      "learning_rate": 3.456974965963271e-06,
+      "loss": 0.4592,
+      "step": 627
+    },
+    {
+      "epoch": 0.6855895196506551,
+      "grad_norm": 1.5669304480812016,
+      "learning_rate": 3.447810503566579e-06,
+      "loss": 0.484,
+      "step": 628
+    },
+    {
+      "epoch": 0.6866812227074236,
+      "grad_norm": 1.5033628816852151,
+      "learning_rate": 3.438665778672867e-06,
+      "loss": 0.5692,
+      "step": 629
+    },
+    {
+      "epoch": 0.6877729257641921,
+      "grad_norm": 1.4892610870512741,
+      "learning_rate": 3.429540907565644e-06,
+      "loss": 0.5574,
+      "step": 630
+    },
+    {
+      "epoch": 0.6888646288209607,
+      "grad_norm": 1.4346767102540499,
+      "learning_rate": 3.4204360062759518e-06,
+      "loss": 0.4922,
+      "step": 631
+    },
+    {
+      "epoch": 0.6899563318777293,
+      "grad_norm": 1.6385962403810959,
+      "learning_rate": 3.4113511905809043e-06,
+      "loss": 0.5122,
+      "step": 632
+    },
+    {
+      "epoch": 0.6910480349344978,
+      "grad_norm": 1.5484369704845515,
+      "learning_rate": 3.4022865760022044e-06,
+      "loss": 0.5603,
+      "step": 633
+    },
+    {
+      "epoch": 0.6921397379912664,
+      "grad_norm": 1.3328218691814475,
+      "learning_rate": 3.3932422778046854e-06,
+      "loss": 0.4696,
+      "step": 634
+    },
+    {
+      "epoch": 0.6932314410480349,
+      "grad_norm": 1.3201936226659459,
+      "learning_rate": 3.3842184109948303e-06,
+      "loss": 0.6217,
+      "step": 635
+    },
+    {
+      "epoch": 0.6943231441048034,
+      "grad_norm": 1.463942512713303,
+      "learning_rate": 3.3752150903193265e-06,
+      "loss": 0.622,
+      "step": 636
+    },
+    {
+      "epoch": 0.6954148471615721,
+      "grad_norm": 1.6139395561000496,
+      "learning_rate": 3.366232430263593e-06,
+      "loss": 0.4245,
+      "step": 637
+    },
+    {
+      "epoch": 0.6965065502183406,
+      "grad_norm": 1.3285149685117137,
+      "learning_rate": 3.357270545050337e-06,
+      "loss": 0.4324,
+      "step": 638
+    },
+    {
+      "epoch": 0.6975982532751092,
+      "grad_norm": 1.5088166603297204,
+      "learning_rate": 3.3483295486380846e-06,
+      "loss": 0.4563,
+      "step": 639
+    },
+    {
+      "epoch": 0.6986899563318777,
+      "grad_norm": 1.3095999354264778,
+      "learning_rate": 3.3394095547197526e-06,
+      "loss": 0.4683,
+      "step": 640
+    },
+    {
+      "epoch": 0.6997816593886463,
+      "grad_norm": 1.58544588878384,
+      "learning_rate": 3.330510676721183e-06,
+      "loss": 0.5485,
+      "step": 641
+    },
+    {
+      "epoch": 0.7008733624454149,
+      "grad_norm": 1.495660640862744,
+      "learning_rate": 3.3216330277997157e-06,
+      "loss": 0.5082,
+      "step": 642
+    },
+    {
+      "epoch": 0.7019650655021834,
+      "grad_norm": 1.3658052652721495,
+      "learning_rate": 3.312776720842741e-06,
+      "loss": 0.4509,
+      "step": 643
+    },
+    {
+      "epoch": 0.7030567685589519,
+      "grad_norm": 1.4640768663316701,
+      "learning_rate": 3.3039418684662676e-06,
+      "loss": 0.5417,
+      "step": 644
+    },
+    {
+      "epoch": 0.7041484716157205,
+      "grad_norm": 1.4247855209179192,
+      "learning_rate": 3.295128583013486e-06,
+      "loss": 0.4318,
+      "step": 645
+    },
+    {
+      "epoch": 0.7052401746724891,
+      "grad_norm": 1.3800459453105371,
+      "learning_rate": 3.286336976553352e-06,
+      "loss": 0.5208,
+      "step": 646
+    },
+    {
+      "epoch": 0.7063318777292577,
+      "grad_norm": 1.2554511851900485,
+      "learning_rate": 3.277567160879144e-06,
+      "loss": 0.4666,
+      "step": 647
+    },
+    {
+      "epoch": 0.7074235807860262,
+      "grad_norm": 1.629496057033825,
+      "learning_rate": 3.268819247507056e-06,
+      "loss": 0.5119,
+      "step": 648
+    },
+    {
+      "epoch": 0.7085152838427947,
+      "grad_norm": 1.679530653212758,
+      "learning_rate": 3.260093347674772e-06,
+      "loss": 0.5256,
+      "step": 649
+    },
+    {
+      "epoch": 0.7096069868995634,
+      "grad_norm": 1.5573971849859753,
+      "learning_rate": 3.251389572340056e-06,
+      "loss": 0.5222,
+      "step": 650
+    },
+    {
+      "epoch": 0.7106986899563319,
+      "grad_norm": 1.3889141401111582,
+      "learning_rate": 3.2427080321793374e-06,
+      "loss": 0.4432,
+      "step": 651
+    },
+    {
+      "epoch": 0.7117903930131004,
+      "grad_norm": 1.3127355149090743,
+      "learning_rate": 3.2340488375863047e-06,
+      "loss": 0.5384,
+      "step": 652
+    },
+    {
+      "epoch": 0.712882096069869,
+      "grad_norm": 1.3983921070638459,
+      "learning_rate": 3.225412098670504e-06,
+      "loss": 0.4823,
+      "step": 653
+    },
+    {
+      "epoch": 0.7139737991266376,
+      "grad_norm": 1.4752891184006638,
+      "learning_rate": 3.216797925255936e-06,
+      "loss": 0.5408,
+      "step": 654
+    },
+    {
+      "epoch": 0.7150655021834061,
+      "grad_norm": 1.4616253015066785,
+      "learning_rate": 3.2082064268796592e-06,
+      "loss": 0.5643,
+      "step": 655
+    },
+    {
+      "epoch": 0.7161572052401747,
+      "grad_norm": 1.6766027674126829,
+      "learning_rate": 3.1996377127903985e-06,
+      "loss": 0.5628,
+      "step": 656
+    },
+    {
+      "epoch": 0.7172489082969432,
+      "grad_norm": 1.5205329442665274,
+      "learning_rate": 3.1910918919471596e-06,
+      "loss": 0.514,
+      "step": 657
+    },
+    {
+      "epoch": 0.7183406113537117,
+      "grad_norm": 1.6037099096063592,
+      "learning_rate": 3.1825690730178357e-06,
+      "loss": 0.399,
+      "step": 658
+    },
+    {
+      "epoch": 0.7194323144104804,
+      "grad_norm": 1.7203937525493616,
+      "learning_rate": 3.17406936437783e-06,
+      "loss": 0.4892,
+      "step": 659
+    },
+    {
+      "epoch": 0.7205240174672489,
+      "grad_norm": 1.6170397886834587,
+      "learning_rate": 3.165592874108679e-06,
+      "loss": 0.4833,
+      "step": 660
+    },
+    {
+      "epoch": 0.7216157205240175,
+      "grad_norm": 1.4171579690456109,
+      "learning_rate": 3.1571397099966767e-06,
+      "loss": 0.4428,
+      "step": 661
+    },
+    {
+      "epoch": 0.722707423580786,
+      "grad_norm": 1.7081299315746081,
+      "learning_rate": 3.148709979531502e-06,
+      "loss": 0.4435,
+      "step": 662
+    },
+    {
+      "epoch": 0.7237991266375546,
+      "grad_norm": 1.4489036867577625,
+      "learning_rate": 3.1403037899048547e-06,
+      "loss": 0.4866,
+      "step": 663
+    },
+    {
+      "epoch": 0.7248908296943232,
+      "grad_norm": 1.6131445114377558,
+      "learning_rate": 3.1319212480090895e-06,
+      "loss": 0.4946,
+      "step": 664
+    },
+    {
+      "epoch": 0.7259825327510917,
+      "grad_norm": 1.8063942650562066,
+      "learning_rate": 3.123562460435863e-06,
+      "loss": 0.5204,
+      "step": 665
+    },
+    {
+      "epoch": 0.7270742358078602,
+      "grad_norm": 1.7518173609793894,
+      "learning_rate": 3.115227533474771e-06,
+      "loss": 0.5224,
+      "step": 666
+    },
+    {
+      "epoch": 0.7281659388646288,
+      "grad_norm": 1.6254743756924752,
+      "learning_rate": 3.1069165731119987e-06,
+      "loss": 0.5313,
+      "step": 667
+    },
+    {
+      "epoch": 0.7292576419213974,
+      "grad_norm": 1.3009078308401743,
+      "learning_rate": 3.0986296850289756e-06,
+      "loss": 0.5709,
+      "step": 668
+    },
+    {
+      "epoch": 0.730349344978166,
+      "grad_norm": 1.448104409567261,
+      "learning_rate": 3.0903669746010312e-06,
+      "loss": 0.4633,
+      "step": 669
+    },
+    {
+      "epoch": 0.7314410480349345,
+      "grad_norm": 1.4606508569088716,
+      "learning_rate": 3.0821285468960542e-06,
+      "loss": 0.5297,
+      "step": 670
+    },
+    {
+      "epoch": 0.732532751091703,
+      "grad_norm": 1.4136587749469944,
+      "learning_rate": 3.0739145066731534e-06,
+      "loss": 0.5312,
+      "step": 671
+    },
+    {
+      "epoch": 0.7336244541484717,
+      "grad_norm": 1.5905997409531767,
+      "learning_rate": 3.0657249583813307e-06,
+      "loss": 0.5227,
+      "step": 672
+    },
+    {
+      "epoch": 0.7347161572052402,
+      "grad_norm": 1.5538511136048576,
+      "learning_rate": 3.0575600061581518e-06,
+      "loss": 0.5521,
+      "step": 673
+    },
+    {
+      "epoch": 0.7358078602620087,
+      "grad_norm": 1.486422100568419,
+      "learning_rate": 3.0494197538284193e-06,
+      "loss": 0.5114,
+      "step": 674
+    },
+    {
+      "epoch": 0.7368995633187773,
+      "grad_norm": 1.3670813674791609,
+      "learning_rate": 3.041304304902855e-06,
+      "loss": 0.4841,
+      "step": 675
+    },
+    {
+      "epoch": 0.7379912663755459,
+      "grad_norm": 1.334243639446627,
+      "learning_rate": 3.033213762576781e-06,
+      "loss": 0.4348,
+      "step": 676
+    },
+    {
+      "epoch": 0.7390829694323144,
+      "grad_norm": 1.3652145593061606,
+      "learning_rate": 3.0251482297288125e-06,
+      "loss": 0.5032,
+      "step": 677
+    },
+    {
+      "epoch": 0.740174672489083,
+      "grad_norm": 1.5217375260885644,
+      "learning_rate": 3.0171078089195447e-06,
+      "loss": 0.5636,
+      "step": 678
+    },
+    {
+      "epoch": 0.7412663755458515,
+      "grad_norm": 1.382594820324151,
+      "learning_rate": 3.0090926023902494e-06,
+      "loss": 0.4378,
+      "step": 679
+    },
+    {
+      "epoch": 0.74235807860262,
+      "grad_norm": 1.45824384976189,
+      "learning_rate": 3.001102712061577e-06,
+      "loss": 0.4924,
+      "step": 680
+    },
+    {
+      "epoch": 0.7434497816593887,
+      "grad_norm": 1.3772141004468592,
+      "learning_rate": 2.9931382395322595e-06,
+      "loss": 0.5824,
+      "step": 681
+    },
+    {
+      "epoch": 0.7445414847161572,
+      "grad_norm": 1.5056226458754594,
+      "learning_rate": 2.985199286077818e-06,
+      "loss": 0.5283,
+      "step": 682
+    },
+    {
+      "epoch": 0.7456331877729258,
+      "grad_norm": 1.2862985634824098,
+      "learning_rate": 2.977285952649276e-06,
+      "loss": 0.4862,
+      "step": 683
+    },
+    {
+      "epoch": 0.7467248908296943,
+      "grad_norm": 1.358956864438878,
+      "learning_rate": 2.9693983398718738e-06,
+      "loss": 0.4513,
+      "step": 684
+    },
+    {
+      "epoch": 0.7478165938864629,
+      "grad_norm": 1.681388781002674,
+      "learning_rate": 2.9615365480437903e-06,
+      "loss": 0.4229,
+      "step": 685
+    },
+    {
+      "epoch": 0.7489082969432315,
+      "grad_norm": 1.7706615580410843,
+      "learning_rate": 2.95370067713487e-06,
+      "loss": 0.4856,
+      "step": 686
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 1.5889451935714294,
+      "learning_rate": 2.9458908267853452e-06,
+      "loss": 0.4419,
+      "step": 687
+    },
+    {
+      "epoch": 0.7510917030567685,
+      "grad_norm": 1.544459411681686,
+      "learning_rate": 2.9381070963045776e-06,
+      "loss": 0.5503,
+      "step": 688
+    },
+    {
+      "epoch": 0.7521834061135371,
+      "grad_norm": 1.4057907314213063,
+      "learning_rate": 2.930349584669787e-06,
+      "loss": 0.5223,
+      "step": 689
+    },
+    {
+      "epoch": 0.7532751091703057,
+      "grad_norm": 1.5378927118318613,
+      "learning_rate": 2.922618390524799e-06,
+      "loss": 0.5557,
+      "step": 690
+    },
+    {
+      "epoch": 0.7543668122270742,
+      "grad_norm": 1.4035201644525876,
+      "learning_rate": 2.914913612178787e-06,
+      "loss": 0.4913,
+      "step": 691
+    },
+    {
+      "epoch": 0.7554585152838428,
+      "grad_norm": 1.561354983159288,
+      "learning_rate": 2.9072353476050254e-06,
+      "loss": 0.4545,
+      "step": 692
+    },
+    {
+      "epoch": 0.7565502183406113,
+      "grad_norm": 1.2907793257634048,
+      "learning_rate": 2.8995836944396394e-06,
+      "loss": 0.545,
+      "step": 693
+    },
+    {
+      "epoch": 0.75764192139738,
+      "grad_norm": 1.429751557254624,
+      "learning_rate": 2.891958749980369e-06,
+      "loss": 0.5632,
+      "step": 694
+    },
+    {
+      "epoch": 0.7587336244541485,
+      "grad_norm": 1.3060316717397793,
+      "learning_rate": 2.8843606111853247e-06,
+      "loss": 0.4993,
+      "step": 695
+    },
+    {
+      "epoch": 0.759825327510917,
+      "grad_norm": 1.3881628331724296,
+      "learning_rate": 2.8767893746717614e-06,
+      "loss": 0.4583,
+      "step": 696
+    },
+    {
+      "epoch": 0.7609170305676856,
+      "grad_norm": 1.2188750020693087,
+      "learning_rate": 2.869245136714846e-06,
+      "loss": 0.4878,
+      "step": 697
+    },
+    {
+      "epoch": 0.7620087336244541,
+      "grad_norm": 1.3688514324159553,
+      "learning_rate": 2.861727993246437e-06,
+      "loss": 0.5454,
+      "step": 698
+    },
+    {
+      "epoch": 0.7631004366812227,
+      "grad_norm": 1.4693627823600672,
+      "learning_rate": 2.854238039853855e-06,
+      "loss": 0.4887,
+      "step": 699
+    },
+    {
+      "epoch": 0.7641921397379913,
+      "grad_norm": 1.4063749888320647,
+      "learning_rate": 2.8467753717786818e-06,
+      "loss": 0.514,
+      "step": 700
+    },
+    {
+      "epoch": 0.7652838427947598,
+      "grad_norm": 1.7262735326314635,
+      "learning_rate": 2.8393400839155354e-06,
+      "loss": 0.5922,
+      "step": 701
+    },
+    {
+      "epoch": 0.7663755458515283,
+      "grad_norm": 1.2160175934089779,
+      "learning_rate": 2.831932270810875e-06,
+      "loss": 0.4369,
+      "step": 702
+    },
+    {
+      "epoch": 0.767467248908297,
+      "grad_norm": 1.5736504616224942,
+      "learning_rate": 2.824552026661787e-06,
+      "loss": 0.3896,
+      "step": 703
+    },
+    {
+      "epoch": 0.7685589519650655,
+      "grad_norm": 1.334413826428926,
+      "learning_rate": 2.8171994453147993e-06,
+      "loss": 0.4353,
+      "step": 704
+    },
+    {
+      "epoch": 0.769650655021834,
+      "grad_norm": 1.2822977853771844,
+      "learning_rate": 2.8098746202646773e-06,
+      "loss": 0.4925,
+      "step": 705
+    },
+    {
+      "epoch": 0.7707423580786026,
+      "grad_norm": 1.5498532031156085,
+      "learning_rate": 2.8025776446532452e-06,
+      "loss": 0.5633,
+      "step": 706
+    },
+    {
+      "epoch": 0.7718340611353712,
+      "grad_norm": 1.2449777128846868,
+      "learning_rate": 2.7953086112681894e-06,
+      "loss": 0.4179,
+      "step": 707
+    },
+    {
+      "epoch": 0.7729257641921398,
+      "grad_norm": 1.5646874311946461,
+      "learning_rate": 2.7880676125418926e-06,
+      "loss": 0.4719,
+      "step": 708
+    },
+    {
+      "epoch": 0.7740174672489083,
+      "grad_norm": 1.3665728473586682,
+      "learning_rate": 2.7808547405502437e-06,
+      "loss": 0.3991,
+      "step": 709
+    },
+    {
+      "epoch": 0.7751091703056768,
+      "grad_norm": 1.501171484718762,
+      "learning_rate": 2.773670087011484e-06,
+      "loss": 0.5744,
+      "step": 710
+    },
+    {
+      "epoch": 0.7762008733624454,
+      "grad_norm": 1.6011435161099792,
+      "learning_rate": 2.7665137432850203e-06,
+      "loss": 0.5695,
+      "step": 711
+    },
+    {
+      "epoch": 0.777292576419214,
+      "grad_norm": 1.5658472473873142,
+      "learning_rate": 2.759385800370284e-06,
+      "loss": 0.5355,
+      "step": 712
+    },
+    {
+      "epoch": 0.7783842794759825,
+      "grad_norm": 1.4901124794820078,
+      "learning_rate": 2.7522863489055586e-06,
+      "loss": 0.462,
+      "step": 713
+    },
+    {
+      "epoch": 0.7794759825327511,
+      "grad_norm": 1.5298330174312822,
+      "learning_rate": 2.7452154791668375e-06,
+      "loss": 0.5594,
+      "step": 714
+    },
+    {
+      "epoch": 0.7805676855895196,
+      "grad_norm": 1.4147385808285522,
+      "learning_rate": 2.7381732810666647e-06,
+      "loss": 0.4539,
+      "step": 715
+    },
+    {
+      "epoch": 0.7816593886462883,
+      "grad_norm": 1.4608859654086022,
+      "learning_rate": 2.731159844153005e-06,
+      "loss": 0.4534,
+      "step": 716
+    },
+    {
+      "epoch": 0.7827510917030568,
+      "grad_norm": 1.5117169302712787,
+      "learning_rate": 2.724175257608094e-06,
+      "loss": 0.5258,
+      "step": 717
+    },
+    {
+      "epoch": 0.7838427947598253,
+      "grad_norm": 1.2658257724311885,
+      "learning_rate": 2.7172196102473122e-06,
+      "loss": 0.4096,
+      "step": 718
+    },
+    {
+      "epoch": 0.7849344978165939,
+      "grad_norm": 1.551758238229153,
+      "learning_rate": 2.710292990518049e-06,
+      "loss": 0.6434,
+      "step": 719
+    },
+    {
+      "epoch": 0.7860262008733624,
+      "grad_norm": 1.4076094951453688,
+      "learning_rate": 2.703395486498582e-06,
+      "loss": 0.5038,
+      "step": 720
+    },
+    {
+      "epoch": 0.787117903930131,
+      "grad_norm": 1.5353091781216452,
+      "learning_rate": 2.6965271858969556e-06,
+      "loss": 0.5531,
+      "step": 721
+    },
+    {
+      "epoch": 0.7882096069868996,
+      "grad_norm": 1.443747275507301,
+      "learning_rate": 2.6896881760498684e-06,
+      "loss": 0.6324,
+      "step": 722
+    },
+    {
+      "epoch": 0.7893013100436681,
+      "grad_norm": 2.2066790531801828,
+      "learning_rate": 2.6828785439215574e-06,
+      "loss": 0.4867,
+      "step": 723
+    },
+    {
+      "epoch": 0.7903930131004366,
+      "grad_norm": 1.3039506134634729,
+      "learning_rate": 2.676098376102697e-06,
+      "loss": 0.3967,
+      "step": 724
+    },
+    {
+      "epoch": 0.7914847161572053,
+      "grad_norm": 1.4973744024054314,
+      "learning_rate": 2.6693477588092945e-06,
+      "loss": 0.4741,
+      "step": 725
+    },
+    {
+      "epoch": 0.7925764192139738,
+      "grad_norm": 1.6625078614001911,
+      "learning_rate": 2.6626267778815984e-06,
+      "loss": 0.4454,
+      "step": 726
+    },
+    {
+      "epoch": 0.7936681222707423,
+      "grad_norm": 1.5392825520879216,
+      "learning_rate": 2.655935518783e-06,
+      "loss": 0.4482,
+      "step": 727
+    },
+    {
+      "epoch": 0.7947598253275109,
+      "grad_norm": 1.2279287786501736,
+      "learning_rate": 2.649274066598952e-06,
+      "loss": 0.4494,
+      "step": 728
+    },
+    {
+      "epoch": 0.7958515283842795,
+      "grad_norm": 1.644472266754525,
+      "learning_rate": 2.6426425060358877e-06,
+      "loss": 0.4583,
+      "step": 729
+    },
+    {
+      "epoch": 0.7969432314410481,
+      "grad_norm": 1.3349263839032603,
+      "learning_rate": 2.6360409214201368e-06,
+      "loss": 0.5458,
+      "step": 730
+    },
+    {
+      "epoch": 0.7980349344978166,
+      "grad_norm": 1.353325917442229,
+      "learning_rate": 2.6294693966968603e-06,
+      "loss": 0.4996,
+      "step": 731
+    },
+    {
+      "epoch": 0.7991266375545851,
+      "grad_norm": 1.75020327032623,
+      "learning_rate": 2.622928015428979e-06,
+      "loss": 0.5484,
+      "step": 732
+    },
+    {
+      "epoch": 0.8002183406113537,
+      "grad_norm": 1.7507615696882926,
+      "learning_rate": 2.616416860796113e-06,
+      "loss": 0.5119,
+      "step": 733
+    },
+    {
+      "epoch": 0.8013100436681223,
+      "grad_norm": 1.5319391439632237,
+      "learning_rate": 2.6099360155935237e-06,
+      "loss": 0.5243,
+      "step": 734
+    },
+    {
+      "epoch": 0.8024017467248908,
+      "grad_norm": 1.271596095829789,
+      "learning_rate": 2.603485562231059e-06,
+      "loss": 0.4004,
+      "step": 735
+    },
+    {
+      "epoch": 0.8034934497816594,
+      "grad_norm": 1.5168565334204749,
+      "learning_rate": 2.5970655827321066e-06,
+      "loss": 0.5833,
+      "step": 736
+    },
+    {
+      "epoch": 0.8045851528384279,
+      "grad_norm": 1.425305835480303,
+      "learning_rate": 2.590676158732555e-06,
+      "loss": 0.6107,
+      "step": 737
+    },
+    {
+      "epoch": 0.8056768558951966,
+      "grad_norm": 1.433983687848821,
+      "learning_rate": 2.584317371479747e-06,
+      "loss": 0.5336,
+      "step": 738
+    },
+    {
+      "epoch": 0.8067685589519651,
+      "grad_norm": 1.5171771559883467,
+      "learning_rate": 2.5779893018314545e-06,
+      "loss": 0.4953,
+      "step": 739
+    },
+    {
+      "epoch": 0.8078602620087336,
+      "grad_norm": 1.2466215612737617,
+      "learning_rate": 2.571692030254844e-06,
+      "loss": 0.4111,
+      "step": 740
+    },
+    {
+      "epoch": 0.8089519650655022,
+      "grad_norm": 1.2958293442264301,
+      "learning_rate": 2.5654256368254606e-06,
+      "loss": 0.4796,
+      "step": 741
+    },
+    {
+      "epoch": 0.8100436681222707,
+      "grad_norm": 1.366221486976765,
+      "learning_rate": 2.5591902012262018e-06,
+      "loss": 0.479,
+      "step": 742
+    },
+    {
+      "epoch": 0.8111353711790393,
+      "grad_norm": 1.4757039124958742,
+      "learning_rate": 2.552985802746311e-06,
+      "loss": 0.4448,
+      "step": 743
+    },
+    {
+      "epoch": 0.8122270742358079,
+      "grad_norm": 1.3937974822769155,
+      "learning_rate": 2.5468125202803624e-06,
+      "loss": 0.4178,
+      "step": 744
+    },
+    {
+      "epoch": 0.8133187772925764,
+      "grad_norm": 1.595101474275964,
+      "learning_rate": 2.5406704323272677e-06,
+      "loss": 0.5009,
+      "step": 745
+    },
+    {
+      "epoch": 0.8144104803493449,
+      "grad_norm": 1.4109055917732352,
+      "learning_rate": 2.5345596169892675e-06,
+      "loss": 0.3471,
+      "step": 746
+    },
+    {
+      "epoch": 0.8155021834061136,
+      "grad_norm": 1.434951280260996,
+      "learning_rate": 2.5284801519709426e-06,
+      "loss": 0.5631,
+      "step": 747
+    },
+    {
+      "epoch": 0.8165938864628821,
+      "grad_norm": 1.3792152619204336,
+      "learning_rate": 2.522432114578228e-06,
+      "loss": 0.5299,
+      "step": 748
+    },
+    {
+      "epoch": 0.8176855895196506,
+      "grad_norm": 1.2965847499667549,
+      "learning_rate": 2.5164155817174274e-06,
+      "loss": 0.5049,
+      "step": 749
+    },
+    {
+      "epoch": 0.8187772925764192,
+      "grad_norm": 1.7699360693412276,
+      "learning_rate": 2.5104306298942345e-06,
+      "loss": 0.5608,
+      "step": 750
+    },
+    {
+      "epoch": 0.8198689956331878,
+      "grad_norm": 1.4105028322760174,
+      "learning_rate": 2.5044773352127624e-06,
+      "loss": 0.5555,
+      "step": 751
+    },
+    {
+      "epoch": 0.8209606986899564,
+      "grad_norm": 1.7025442354294646,
+      "learning_rate": 2.4985557733745733e-06,
+      "loss": 0.4905,
+      "step": 752
+    },
+    {
+      "epoch": 0.8220524017467249,
+      "grad_norm": 1.5676630984513402,
+      "learning_rate": 2.4926660196777203e-06,
+      "loss": 0.4812,
+      "step": 753
+    },
+    {
+      "epoch": 0.8231441048034934,
+      "grad_norm": 1.5833360948505186,
+      "learning_rate": 2.486808149015785e-06,
+      "loss": 0.5356,
+      "step": 754
+    },
+    {
+      "epoch": 0.824235807860262,
+      "grad_norm": 1.4178730350285935,
+      "learning_rate": 2.480982235876927e-06,
+      "loss": 0.5062,
+      "step": 755
+    },
+    {
+      "epoch": 0.8253275109170306,
+      "grad_norm": 1.5145732571443271,
+      "learning_rate": 2.4751883543429365e-06,
+      "loss": 0.502,
+      "step": 756
+    },
+    {
+      "epoch": 0.8264192139737991,
+      "grad_norm": 1.507197459921713,
+      "learning_rate": 2.4694265780882952e-06,
+      "loss": 0.4461,
+      "step": 757
+    },
+    {
+      "epoch": 0.8275109170305677,
+      "grad_norm": 1.6314783612840427,
+      "learning_rate": 2.4636969803792347e-06,
+      "loss": 0.5186,
+      "step": 758
+    },
+    {
+      "epoch": 0.8286026200873362,
+      "grad_norm": 1.2676790104347715,
+      "learning_rate": 2.457999634072807e-06,
+      "loss": 0.4751,
+      "step": 759
+    },
+    {
+      "epoch": 0.8296943231441049,
+      "grad_norm": 1.48067735218539,
+      "learning_rate": 2.4523346116159586e-06,
+      "loss": 0.4172,
+      "step": 760
+    },
+    {
+      "epoch": 0.8307860262008734,
+      "grad_norm": 1.5204394932078549,
+      "learning_rate": 2.446701985044611e-06,
+      "loss": 0.5344,
+      "step": 761
+    },
+    {
+      "epoch": 0.8318777292576419,
+      "grad_norm": 1.5467915314176128,
+      "learning_rate": 2.441101825982738e-06,
+      "loss": 0.4792,
+      "step": 762
+    },
+    {
+      "epoch": 0.8329694323144105,
+      "grad_norm": 1.5788360356770055,
+      "learning_rate": 2.4355342056414632e-06,
+      "loss": 0.555,
+      "step": 763
+    },
+    {
+      "epoch": 0.834061135371179,
+      "grad_norm": 1.3178495446349063,
+      "learning_rate": 2.4299991948181503e-06,
+      "loss": 0.4323,
+      "step": 764
+    },
+    {
+      "epoch": 0.8351528384279476,
+      "grad_norm": 1.6402225350694717,
+      "learning_rate": 2.4244968638955014e-06,
+      "loss": 0.5475,
+      "step": 765
+    },
+    {
+      "epoch": 0.8362445414847162,
+      "grad_norm": 1.3512463029540156,
+      "learning_rate": 2.419027282840665e-06,
+      "loss": 0.5547,
+      "step": 766
+    },
+    {
+      "epoch": 0.8373362445414847,
+      "grad_norm": 1.4426233346314499,
+      "learning_rate": 2.4135905212043453e-06,
+      "loss": 0.5333,
+      "step": 767
+    },
+    {
+      "epoch": 0.8384279475982532,
+      "grad_norm": 1.5182561345065841,
+      "learning_rate": 2.408186648119918e-06,
+      "loss": 0.5274,
+      "step": 768
+    },
+    {
+      "epoch": 0.8395196506550219,
+      "grad_norm": 1.5815269832741783,
+      "learning_rate": 2.40281573230255e-06,
+      "loss": 0.5156,
+      "step": 769
+    },
+    {
+      "epoch": 0.8406113537117904,
+      "grad_norm": 1.3659003932004359,
+      "learning_rate": 2.397477842048328e-06,
+      "loss": 0.522,
+      "step": 770
+    },
+    {
+      "epoch": 0.8417030567685589,
+      "grad_norm": 1.4916431814675448,
+      "learning_rate": 2.3921730452333865e-06,
+      "loss": 0.4973,
+      "step": 771
+    },
+    {
+      "epoch": 0.8427947598253275,
+      "grad_norm": 1.5921458181749883,
+      "learning_rate": 2.3869014093130484e-06,
+      "loss": 0.5033,
+      "step": 772
+    },
+    {
+      "epoch": 0.8438864628820961,
+      "grad_norm": 1.3488255455196068,
+      "learning_rate": 2.3816630013209653e-06,
+      "loss": 0.4981,
+      "step": 773
+    },
+    {
+      "epoch": 0.8449781659388647,
+      "grad_norm": 1.4884625204401833,
+      "learning_rate": 2.3764578878682658e-06,
+      "loss": 0.4492,
+      "step": 774
+    },
+    {
+      "epoch": 0.8460698689956332,
+      "grad_norm": 1.403679565738453,
+      "learning_rate": 2.371286135142706e-06,
+      "loss": 0.4386,
+      "step": 775
+    },
+    {
+      "epoch": 0.8471615720524017,
+      "grad_norm": 1.5140972915615711,
+      "learning_rate": 2.366147808907833e-06,
+      "loss": 0.3872,
+      "step": 776
+    },
+    {
+      "epoch": 0.8482532751091703,
+      "grad_norm": 1.2931794718268395,
+      "learning_rate": 2.3610429745021433e-06,
+      "loss": 0.5277,
+      "step": 777
+    },
+    {
+      "epoch": 0.8493449781659389,
+      "grad_norm": 1.65094958261332,
+      "learning_rate": 2.3559716968382566e-06,
+      "loss": 0.5036,
+      "step": 778
+    },
+    {
+      "epoch": 0.8504366812227074,
+      "grad_norm": 1.3484114437655963,
+      "learning_rate": 2.3509340404020845e-06,
+      "loss": 0.4085,
+      "step": 779
+    },
+    {
+      "epoch": 0.851528384279476,
+      "grad_norm": 1.5024637038030444,
+      "learning_rate": 2.3459300692520177e-06,
+      "loss": 0.5749,
+      "step": 780
+    },
+    {
+      "epoch": 0.8526200873362445,
+      "grad_norm": 1.7268258750716359,
+      "learning_rate": 2.340959847018106e-06,
+      "loss": 0.5552,
+      "step": 781
+    },
+    {
+      "epoch": 0.8537117903930131,
+      "grad_norm": 1.4471671123643692,
+      "learning_rate": 2.336023436901253e-06,
+      "loss": 0.5025,
+      "step": 782
+    },
+    {
+      "epoch": 0.8548034934497817,
+      "grad_norm": 1.3544346171446477,
+      "learning_rate": 2.3311209016724047e-06,
+      "loss": 0.4235,
+      "step": 783
+    },
+    {
+      "epoch": 0.8558951965065502,
+      "grad_norm": 1.668582349166636,
+      "learning_rate": 2.3262523036717657e-06,
+      "loss": 0.5257,
+      "step": 784
+    },
+    {
+      "epoch": 0.8569868995633187,
+      "grad_norm": 1.849430371551395,
+      "learning_rate": 2.32141770480799e-06,
+      "loss": 0.4966,
+      "step": 785
+    },
+    {
+      "epoch": 0.8580786026200873,
+      "grad_norm": 1.4257795015278287,
+      "learning_rate": 2.3166171665574068e-06,
+      "loss": 0.3834,
+      "step": 786
+    },
+    {
+      "epoch": 0.8591703056768559,
+      "grad_norm": 1.5867478801563297,
+      "learning_rate": 2.3118507499632304e-06,
+      "loss": 0.512,
+      "step": 787
+    },
+    {
+      "epoch": 0.8602620087336245,
+      "grad_norm": 1.6022282216088801,
+      "learning_rate": 2.3071185156347896e-06,
+      "loss": 0.5756,
+      "step": 788
+    },
+    {
+      "epoch": 0.861353711790393,
+      "grad_norm": 1.653250014006023,
+      "learning_rate": 2.302420523746752e-06,
+      "loss": 0.4727,
+      "step": 789
+    },
+    {
+      "epoch": 0.8624454148471615,
+      "grad_norm": 1.5698471805784935,
+      "learning_rate": 2.297756834038365e-06,
+      "loss": 0.5493,
+      "step": 790
+    },
+    {
+      "epoch": 0.8635371179039302,
+      "grad_norm": 1.3890720239074847,
+      "learning_rate": 2.293127505812687e-06,
+      "loss": 0.442,
+      "step": 791
+    },
+    {
+      "epoch": 0.8646288209606987,
+      "grad_norm": 1.3473874382458382,
+      "learning_rate": 2.288532597935846e-06,
+      "loss": 0.4535,
+      "step": 792
+    },
+    {
+      "epoch": 0.8657205240174672,
+      "grad_norm": 1.5754647979260004,
+      "learning_rate": 2.2839721688362764e-06,
+      "loss": 0.4703,
+      "step": 793
+    },
+    {
+      "epoch": 0.8668122270742358,
+      "grad_norm": 1.3874265635155072,
+      "learning_rate": 2.27944627650399e-06,
+      "loss": 0.4903,
+      "step": 794
+    },
+    {
+      "epoch": 0.8679039301310044,
+      "grad_norm": 1.5255706234718944,
+      "learning_rate": 2.2749549784898288e-06,
+      "loss": 0.4296,
+      "step": 795
+    },
+    {
+      "epoch": 0.868995633187773,
+      "grad_norm": 1.3874604308608358,
+      "learning_rate": 2.2704983319047363e-06,
+      "loss": 0.436,
+      "step": 796
+    },
+    {
+      "epoch": 0.8700873362445415,
+      "grad_norm": 1.4466236624321545,
+      "learning_rate": 2.2660763934190323e-06,
+      "loss": 0.5945,
+      "step": 797
+    },
+    {
+      "epoch": 0.87117903930131,
+      "grad_norm": 1.6017282651545426,
+      "learning_rate": 2.261689219261693e-06,
+      "loss": 0.5227,
+      "step": 798
+    },
+    {
+      "epoch": 0.8722707423580786,
+      "grad_norm": 1.63954651756713,
+      "learning_rate": 2.2573368652196324e-06,
+      "loss": 0.5144,
+      "step": 799
+    },
+    {
+      "epoch": 0.8733624454148472,
+      "grad_norm": 1.9052064611930404,
+      "learning_rate": 2.253019386636995e-06,
+      "loss": 0.4255,
+      "step": 800
+    },
+    {
+      "epoch": 0.8744541484716157,
+      "grad_norm": 1.44956679873312,
+      "learning_rate": 2.248736838414454e-06,
+      "loss": 0.4413,
+      "step": 801
+    },
+    {
+      "epoch": 0.8755458515283843,
+      "grad_norm": 1.5373891062534013,
+      "learning_rate": 2.244489275008509e-06,
+      "loss": 0.477,
+      "step": 802
+    },
+    {
+      "epoch": 0.8766375545851528,
+      "grad_norm": 1.337436818012808,
+      "learning_rate": 2.2402767504307973e-06,
+      "loss": 0.4437,
+      "step": 803
+    },
+    {
+      "epoch": 0.8777292576419214,
+      "grad_norm": 1.4288499680577358,
+      "learning_rate": 2.2360993182474043e-06,
+      "loss": 0.4113,
+      "step": 804
+    },
+    {
+      "epoch": 0.87882096069869,
+      "grad_norm": 1.4761288915146535,
+      "learning_rate": 2.231957031578186e-06,
+      "loss": 0.53,
+      "step": 805
+    },
+    {
+      "epoch": 0.8799126637554585,
+      "grad_norm": 1.4559595441230249,
+      "learning_rate": 2.227849943096089e-06,
+      "loss": 0.5839,
+      "step": 806
+    },
+    {
+      "epoch": 0.881004366812227,
+      "grad_norm": 1.356122629089398,
+      "learning_rate": 2.223778105026483e-06,
+      "loss": 0.5175,
+      "step": 807
+    },
+    {
+      "epoch": 0.8820960698689956,
+      "grad_norm": 1.4285104513981248,
+      "learning_rate": 2.2197415691464978e-06,
+      "loss": 0.5656,
+      "step": 808
+    },
+    {
+      "epoch": 0.8831877729257642,
+      "grad_norm": 1.6446151125998247,
+      "learning_rate": 2.215740386784363e-06,
+      "loss": 0.4949,
+      "step": 809
+    },
+    {
+      "epoch": 0.8842794759825328,
+      "grad_norm": 1.5697329931755901,
+      "learning_rate": 2.211774608818756e-06,
+      "loss": 0.4485,
+      "step": 810
+    },
+    {
+      "epoch": 0.8853711790393013,
+      "grad_norm": 1.4212775687957804,
+      "learning_rate": 2.207844285678155e-06,
+      "loss": 0.4552,
+      "step": 811
+    },
+    {
+      "epoch": 0.8864628820960698,
+      "grad_norm": 1.5563125426433158,
+      "learning_rate": 2.203949467340198e-06,
+      "loss": 0.4584,
+      "step": 812
+    },
+    {
+      "epoch": 0.8875545851528385,
+      "grad_norm": 1.4100313864446468,
+      "learning_rate": 2.200090203331047e-06,
+      "loss": 0.4462,
+      "step": 813
+    },
+    {
+      "epoch": 0.888646288209607,
+      "grad_norm": 1.3400049168659227,
+      "learning_rate": 2.196266542724759e-06,
+      "loss": 0.5074,
+      "step": 814
+    },
+    {
+      "epoch": 0.8897379912663755,
+      "grad_norm": 1.2777197081112,
+      "learning_rate": 2.1924785341426607e-06,
+      "loss": 0.4549,
+      "step": 815
+    },
+    {
+      "epoch": 0.8908296943231441,
+      "grad_norm": 1.6729756071220223,
+      "learning_rate": 2.18872622575273e-06,
+      "loss": 0.5515,
+      "step": 816
+    },
+    {
+      "epoch": 0.8919213973799127,
+      "grad_norm": 1.5391238422917355,
+      "learning_rate": 2.185009665268985e-06,
+      "loss": 0.4776,
+      "step": 817
+    },
+    {
+      "epoch": 0.8930131004366813,
+      "grad_norm": 1.3890773151501015,
+      "learning_rate": 2.181328899950878e-06,
+      "loss": 0.4669,
+      "step": 818
+    },
+    {
+      "epoch": 0.8941048034934498,
+      "grad_norm": 1.441746914318561,
+      "learning_rate": 2.1776839766026917e-06,
+      "loss": 0.5172,
+      "step": 819
+    },
+    {
+      "epoch": 0.8951965065502183,
+      "grad_norm": 1.4698990931490696,
+      "learning_rate": 2.174074941572945e-06,
+      "loss": 0.4015,
+      "step": 820
+    },
+    {
+      "epoch": 0.8962882096069869,
+      "grad_norm": 1.5602622399868518,
+      "learning_rate": 2.1705018407538052e-06,
+      "loss": 0.4702,
+      "step": 821
+    },
+    {
+      "epoch": 0.8973799126637555,
+      "grad_norm": 1.6630230037043825,
+      "learning_rate": 2.1669647195805045e-06,
+      "loss": 0.4247,
+      "step": 822
+    },
+    {
+      "epoch": 0.898471615720524,
+      "grad_norm": 1.23838619288341,
+      "learning_rate": 2.1634636230307583e-06,
+      "loss": 0.5549,
+      "step": 823
+    },
+    {
+      "epoch": 0.8995633187772926,
+      "grad_norm": 1.303752704544832,
+      "learning_rate": 2.1599985956241974e-06,
+      "loss": 0.5294,
+      "step": 824
+    },
+    {
+      "epoch": 0.9006550218340611,
+      "grad_norm": 1.5965392918352368,
+      "learning_rate": 2.156569681421802e-06,
+      "loss": 0.5207,
+      "step": 825
+    },
+    {
+      "epoch": 0.9017467248908297,
+      "grad_norm": 1.5658093803338047,
+      "learning_rate": 2.153176924025338e-06,
+      "loss": 0.5036,
+      "step": 826
+    },
+    {
+      "epoch": 0.9028384279475983,
+      "grad_norm": 1.4362641435970003,
+      "learning_rate": 2.1498203665768054e-06,
+      "loss": 0.5309,
+      "step": 827
+    },
+    {
+      "epoch": 0.9039301310043668,
+      "grad_norm": 1.3768885226349667,
+      "learning_rate": 2.1465000517578882e-06,
+      "loss": 0.4467,
+      "step": 828
+    },
+    {
+      "epoch": 0.9050218340611353,
+      "grad_norm": 1.560858950459365,
+      "learning_rate": 2.1432160217894144e-06,
+      "loss": 0.5254,
+      "step": 829
+    },
+    {
+      "epoch": 0.9061135371179039,
+      "grad_norm": 1.4857990016793732,
+      "learning_rate": 2.1399683184308157e-06,
+      "loss": 0.4494,
+      "step": 830
+    },
+    {
+      "epoch": 0.9072052401746725,
+      "grad_norm": 1.3924147851916997,
+      "learning_rate": 2.1367569829795975e-06,
+      "loss": 0.4939,
+      "step": 831
+    },
+    {
+      "epoch": 0.9082969432314411,
+      "grad_norm": 1.467257632878726,
+      "learning_rate": 2.1335820562708133e-06,
+      "loss": 0.3817,
+      "step": 832
+    },
+    {
+      "epoch": 0.9093886462882096,
+      "grad_norm": 1.4338221579858819,
+      "learning_rate": 2.1304435786765496e-06,
+      "loss": 0.4756,
+      "step": 833
+    },
+    {
+      "epoch": 0.9104803493449781,
+      "grad_norm": 1.4476290815470778,
+      "learning_rate": 2.127341590105406e-06,
+      "loss": 0.4872,
+      "step": 834
+    },
+    {
+      "epoch": 0.9115720524017468,
+      "grad_norm": 1.4128458095745486,
+      "learning_rate": 2.124276130001992e-06,
+      "loss": 0.6456,
+      "step": 835
+    },
+    {
+      "epoch": 0.9126637554585153,
+      "grad_norm": 1.6789922152845889,
+      "learning_rate": 2.1212472373464245e-06,
+      "loss": 0.4783,
+      "step": 836
+    },
+    {
+      "epoch": 0.9137554585152838,
+      "grad_norm": 1.3775065717539268,
+      "learning_rate": 2.1182549506538323e-06,
+      "loss": 0.46,
+      "step": 837
+    },
+    {
+      "epoch": 0.9148471615720524,
+      "grad_norm": 1.5218752861219855,
+      "learning_rate": 2.115299307973865e-06,
+      "loss": 0.4672,
+      "step": 838
+    },
+    {
+      "epoch": 0.915938864628821,
+      "grad_norm": 1.6179423760169276,
+      "learning_rate": 2.1123803468902117e-06,
+      "loss": 0.459,
+      "step": 839
+    },
+    {
+      "epoch": 0.9170305676855895,
+      "grad_norm": 1.6296532181092986,
+      "learning_rate": 2.1094981045201208e-06,
+      "loss": 0.4474,
+      "step": 840
+    },
+    {
+      "epoch": 0.9170305676855895,
+      "eval_accuracy": 0.8187617923415552,
+      "eval_accuracy_first_token": 0.7932539321756523,
+      "eval_accuracy_first_token_<": 0.9678522571819426,
+      "eval_accuracy_first_token_<_total": 2924,
+      "eval_accuracy_first_token_<|python_tag|>": 0.8853118712273642,
+      "eval_accuracy_first_token_<|python_tag|>_total": 994,
+      "eval_accuracy_first_token_Here": 0.7616580310880829,
+      "eval_accuracy_first_token_Here_total": 386,
+      "eval_accuracy_first_token_The": 0.9054646988432389,
+      "eval_accuracy_first_token_The_total": 2507,
+      "eval_accuracy_first_token_To": 0.8397435897435898,
+      "eval_accuracy_first_token_To_total": 936,
+      "eval_first_token_param_values": 0.9391774602750701,
+      "eval_first_token_param_values_total": 14978,
+      "eval_loss": 0.5705814957618713,
+      "eval_perplexity": 1.1099794468961024,
+      "eval_runtime": 393.4493,
+      "eval_samples_per_second": 1.97,
+      "eval_steps_per_second": 0.247,
+      "eval_total_number_first_token": 10999,
+      "step": 840
+    },
+    {
+      "epoch": 0.9181222707423581,
+      "grad_norm": 1.205254153426699,
+      "learning_rate": 2.1066526175139284e-06,
+      "loss": 0.5144,
+      "step": 841
+    },
+    {
+      "epoch": 0.9192139737991266,
+      "grad_norm": 1.6709674318452725,
+      "learning_rate": 2.103843922054593e-06,
+      "loss": 0.5163,
+      "step": 842
+    },
+    {
+      "epoch": 0.9203056768558951,
+      "grad_norm": 1.437027858243901,
+      "learning_rate": 2.1010720538572353e-06,
+      "loss": 0.4309,
+      "step": 843
+    },
+    {
+      "epoch": 0.9213973799126638,
+      "grad_norm": 1.4494707759764556,
+      "learning_rate": 2.098337048168684e-06,
+      "loss": 0.5106,
+      "step": 844
+    },
+    {
+      "epoch": 0.9224890829694323,
+      "grad_norm": 1.1335356667481837,
+      "learning_rate": 2.095638939767026e-06,
+      "loss": 0.5262,
+      "step": 845
+    },
+    {
+      "epoch": 0.9235807860262009,
+      "grad_norm": 1.4089437898384287,
+      "learning_rate": 2.092977762961169e-06,
+      "loss": 0.456,
+      "step": 846
+    },
+    {
+      "epoch": 0.9246724890829694,
+      "grad_norm": 1.5580485353987292,
+      "learning_rate": 2.0903535515903973e-06,
+      "loss": 0.4332,
+      "step": 847
+    },
+    {
+      "epoch": 0.925764192139738,
+      "grad_norm": 1.4205570846295756,
+      "learning_rate": 2.087766339023949e-06,
+      "loss": 0.3294,
+      "step": 848
+    },
+    {
+      "epoch": 0.9268558951965066,
+      "grad_norm": 1.5282550968605033,
+      "learning_rate": 2.0852161581605876e-06,
+      "loss": 0.4687,
+      "step": 849
+    },
+    {
+      "epoch": 0.9279475982532751,
+      "grad_norm": 1.7856045037336745,
+      "learning_rate": 2.0827030414281865e-06,
+      "loss": 0.4517,
+      "step": 850
+    },
+    {
+      "epoch": 0.9290393013100436,
+      "grad_norm": 1.5465010658627751,
+      "learning_rate": 2.0802270207833125e-06,
+      "loss": 0.4412,
+      "step": 851
+    },
+    {
+      "epoch": 0.9301310043668122,
+      "grad_norm": 1.3738941993193405,
+      "learning_rate": 2.0777881277108233e-06,
+      "loss": 0.542,
+      "step": 852
+    },
+    {
+      "epoch": 0.9312227074235808,
+      "grad_norm": 2.321698082996745,
+      "learning_rate": 2.075386393223465e-06,
+      "loss": 0.4124,
+      "step": 853
+    },
+    {
+      "epoch": 0.9323144104803494,
+      "grad_norm": 1.324438187196309,
+      "learning_rate": 2.073021847861479e-06,
+      "loss": 0.4605,
+      "step": 854
+    },
+    {
+      "epoch": 0.9334061135371179,
+      "grad_norm": 1.463444924977783,
+      "learning_rate": 2.0706945216922125e-06,
+      "loss": 0.5823,
+      "step": 855
+    },
+    {
+      "epoch": 0.9344978165938864,
+      "grad_norm": 1.4922667981099709,
+      "learning_rate": 2.068404444309738e-06,
+      "loss": 0.4646,
+      "step": 856
+    },
+    {
+      "epoch": 0.9355895196506551,
+      "grad_norm": 1.3847315177416661,
+      "learning_rate": 2.066151644834474e-06,
+      "loss": 0.4343,
+      "step": 857
+    },
+    {
+      "epoch": 0.9366812227074236,
+      "grad_norm": 1.5940379300724075,
+      "learning_rate": 2.0639361519128192e-06,
+      "loss": 0.5482,
+      "step": 858
+    },
+    {
+      "epoch": 0.9377729257641921,
+      "grad_norm": 1.4677814031748417,
+      "learning_rate": 2.0617579937167808e-06,
+      "loss": 0.517,
+      "step": 859
+    },
+    {
+      "epoch": 0.9388646288209607,
+      "grad_norm": 1.7341602702329544,
+      "learning_rate": 2.0596171979436262e-06,
+      "loss": 0.4918,
+      "step": 860
+    },
+    {
+      "epoch": 0.9399563318777293,
+      "grad_norm": 1.634439110526545,
+      "learning_rate": 2.0575137918155216e-06,
+      "loss": 0.4462,
+      "step": 861
+    },
+    {
+      "epoch": 0.9410480349344978,
+      "grad_norm": 1.5065600663982117,
+      "learning_rate": 2.055447802079193e-06,
+      "loss": 0.6019,
+      "step": 862
+    },
+    {
+      "epoch": 0.9421397379912664,
+      "grad_norm": 2.0495851259000837,
+      "learning_rate": 2.053419255005581e-06,
+      "loss": 0.5173,
+      "step": 863
+    },
+    {
+      "epoch": 0.9432314410480349,
+      "grad_norm": 1.5961930641084219,
+      "learning_rate": 2.0514281763895078e-06,
+      "loss": 0.4916,
+      "step": 864
+    },
+    {
+      "epoch": 0.9443231441048034,
+      "grad_norm": 1.4756850948436897,
+      "learning_rate": 2.0494745915493524e-06,
+      "loss": 0.4054,
+      "step": 865
+    },
+    {
+      "epoch": 0.9454148471615721,
+      "grad_norm": 1.4560227055974817,
+      "learning_rate": 2.0475585253267246e-06,
+      "loss": 0.4926,
+      "step": 866
+    },
+    {
+      "epoch": 0.9465065502183406,
+      "grad_norm": 1.4957883384826696,
+      "learning_rate": 2.0456800020861516e-06,
+      "loss": 0.4415,
+      "step": 867
+    },
+    {
+      "epoch": 0.9475982532751092,
+      "grad_norm": 1.5238486125969655,
+      "learning_rate": 2.043839045714766e-06,
+      "loss": 0.4261,
+      "step": 868
+    },
+    {
+      "epoch": 0.9486899563318777,
+      "grad_norm": 1.434004624625696,
+      "learning_rate": 2.0420356796220064e-06,
+      "loss": 0.422,
+      "step": 869
+    },
+    {
+      "epoch": 0.9497816593886463,
+      "grad_norm": 1.406938061379921,
+      "learning_rate": 2.040269926739314e-06,
+      "loss": 0.4445,
+      "step": 870
+    },
+    {
+      "epoch": 0.9508733624454149,
+      "grad_norm": 1.4076211144815238,
+      "learning_rate": 2.038541809519845e-06,
+      "loss": 0.4958,
+      "step": 871
+    },
+    {
+      "epoch": 0.9519650655021834,
+      "grad_norm": 1.4839385418962134,
+      "learning_rate": 2.0368513499381834e-06,
+      "loss": 0.4866,
+      "step": 872
+    },
+    {
+      "epoch": 0.9530567685589519,
+      "grad_norm": 1.3849492456103984,
+      "learning_rate": 2.0351985694900617e-06,
+      "loss": 0.509,
+      "step": 873
+    },
+    {
+      "epoch": 0.9541484716157205,
+      "grad_norm": 1.4308287691303287,
+      "learning_rate": 2.0335834891920897e-06,
+      "loss": 0.4827,
+      "step": 874
+    },
+    {
+      "epoch": 0.9552401746724891,
+      "grad_norm": 1.2106458283071144,
+      "learning_rate": 2.0320061295814825e-06,
+      "loss": 0.4897,
+      "step": 875
+    },
+    {
+      "epoch": 0.9563318777292577,
+      "grad_norm": 1.5606989919569907,
+      "learning_rate": 2.0304665107158052e-06,
+      "loss": 0.4196,
+      "step": 876
+    },
+    {
+      "epoch": 0.9574235807860262,
+      "grad_norm": 1.3347004227714003,
+      "learning_rate": 2.028964652172714e-06,
+      "loss": 0.3604,
+      "step": 877
+    },
+    {
+      "epoch": 0.9585152838427947,
+      "grad_norm": 1.4116522769368487,
+      "learning_rate": 2.027500573049707e-06,
+      "loss": 0.5023,
+      "step": 878
+    },
+    {
+      "epoch": 0.9596069868995634,
+      "grad_norm": 1.5195001851239565,
+      "learning_rate": 2.0260742919638845e-06,
+      "loss": 0.4883,
+      "step": 879
+    },
+    {
+      "epoch": 0.9606986899563319,
+      "grad_norm": 1.5253017255371202,
+      "learning_rate": 2.0246858270517086e-06,
+      "loss": 0.4582,
+      "step": 880
+    },
+    {
+      "epoch": 0.9617903930131004,
+      "grad_norm": 1.5989482719027508,
+      "learning_rate": 2.0233351959687758e-06,
+      "loss": 0.4625,
+      "step": 881
+    },
+    {
+      "epoch": 0.962882096069869,
+      "grad_norm": 1.4605962460768873,
+      "learning_rate": 2.0220224158895895e-06,
+      "loss": 0.4964,
+      "step": 882
+    },
+    {
+      "epoch": 0.9639737991266376,
+      "grad_norm": 1.4426450316388095,
+      "learning_rate": 2.020747503507345e-06,
+      "loss": 0.4593,
+      "step": 883
+    },
+    {
+      "epoch": 0.9650655021834061,
+      "grad_norm": 1.469436868223501,
+      "learning_rate": 2.0195104750337126e-06,
+      "loss": 0.6242,
+      "step": 884
+    },
+    {
+      "epoch": 0.9661572052401747,
+      "grad_norm": 1.4929428338509867,
+      "learning_rate": 2.0183113461986374e-06,
+      "loss": 0.5312,
+      "step": 885
+    },
+    {
+      "epoch": 0.9672489082969432,
+      "grad_norm": 1.49713177404691,
+      "learning_rate": 2.017150132250134e-06,
+      "loss": 0.5664,
+      "step": 886
+    },
+    {
+      "epoch": 0.9683406113537117,
+      "grad_norm": 1.592472152020729,
+      "learning_rate": 2.0160268479540953e-06,
+      "loss": 0.4745,
+      "step": 887
+    },
+    {
+      "epoch": 0.9694323144104804,
+      "grad_norm": 1.4810365652583912,
+      "learning_rate": 2.0149415075941037e-06,
+      "loss": 0.4927,
+      "step": 888
+    },
+    {
+      "epoch": 0.9705240174672489,
+      "grad_norm": 1.4236888171085202,
+      "learning_rate": 2.013894124971251e-06,
+      "loss": 0.6274,
+      "step": 889
+    },
+    {
+      "epoch": 0.9716157205240175,
+      "grad_norm": 1.4546508808937473,
+      "learning_rate": 2.0128847134039603e-06,
+      "loss": 0.4766,
+      "step": 890
+    },
+    {
+      "epoch": 0.972707423580786,
+      "grad_norm": 1.5088320910965887,
+      "learning_rate": 2.01191328572782e-06,
+      "loss": 0.5803,
+      "step": 891
+    },
+    {
+      "epoch": 0.9737991266375546,
+      "grad_norm": 1.4031003174167702,
+      "learning_rate": 2.0109798542954166e-06,
+      "loss": 0.5278,
+      "step": 892
+    },
+    {
+      "epoch": 0.9748908296943232,
+      "grad_norm": 1.370703125544038,
+      "learning_rate": 2.010084430976181e-06,
+      "loss": 0.527,
+      "step": 893
+    },
+    {
+      "epoch": 0.9759825327510917,
+      "grad_norm": 1.349030326926829,
+      "learning_rate": 2.009227027156236e-06,
+      "loss": 0.539,
+      "step": 894
+    },
+    {
+      "epoch": 0.9770742358078602,
+      "grad_norm": 1.43566220138461,
+      "learning_rate": 2.008407653738253e-06,
+      "loss": 0.5392,
+      "step": 895
+    },
+    {
+      "epoch": 0.9781659388646288,
+      "grad_norm": 1.5178256542279296,
+      "learning_rate": 2.0076263211413094e-06,
+      "loss": 0.4182,
+      "step": 896
+    },
+    {
+      "epoch": 0.9792576419213974,
+      "grad_norm": 1.5979341650799257,
+      "learning_rate": 2.0068830393007625e-06,
+      "loss": 0.5014,
+      "step": 897
+    },
+    {
+      "epoch": 0.980349344978166,
+      "grad_norm": 1.4678783759443443,
+      "learning_rate": 2.0061778176681174e-06,
+      "loss": 0.4365,
+      "step": 898
+    },
+    {
+      "epoch": 0.9814410480349345,
+      "grad_norm": 1.4727108227205978,
+      "learning_rate": 2.005510665210909e-06,
+      "loss": 0.5425,
+      "step": 899
+    },
+    {
+      "epoch": 0.982532751091703,
+      "grad_norm": 1.4590600569771008,
+      "learning_rate": 2.004881590412587e-06,
+      "loss": 0.4367,
+      "step": 900
+    },
+    {
+      "epoch": 0.9836244541484717,
+      "grad_norm": 1.4380866049192125,
+      "learning_rate": 2.004290601272413e-06,
+      "loss": 0.4171,
+      "step": 901
+    },
+    {
+      "epoch": 0.9847161572052402,
+      "grad_norm": 1.370884653997314,
+      "learning_rate": 2.003737705305351e-06,
+      "loss": 0.4152,
+      "step": 902
+    },
+    {
+      "epoch": 0.9858078602620087,
+      "grad_norm": 1.5721117133850033,
+      "learning_rate": 2.003222909541976e-06,
+      "loss": 0.4556,
+      "step": 903
+    },
+    {
+      "epoch": 0.9868995633187773,
+      "grad_norm": 1.6954405110580915,
+      "learning_rate": 2.0027462205283858e-06,
+      "loss": 0.4836,
+      "step": 904
+    },
+    {
+      "epoch": 0.9879912663755459,
+      "grad_norm": 1.5357666181603002,
+      "learning_rate": 2.002307644326115e-06,
+      "loss": 0.4659,
+      "step": 905
+    },
+    {
+      "epoch": 0.9890829694323144,
+      "grad_norm": 1.3036572912689448,
+      "learning_rate": 2.0019071865120597e-06,
+      "loss": 0.4509,
+      "step": 906
+    },
+    {
+      "epoch": 0.990174672489083,
+      "grad_norm": 1.4565706250847452,
+      "learning_rate": 2.0015448521784045e-06,
+      "loss": 0.5369,
+      "step": 907
+    },
+    {
+      "epoch": 0.9912663755458515,
+      "grad_norm": 1.6290260214981989,
+      "learning_rate": 2.0012206459325623e-06,
+      "loss": 0.53,
+      "step": 908
+    },
+    {
+      "epoch": 0.99235807860262,
+      "grad_norm": 1.475184053581015,
+      "learning_rate": 2.0009345718971095e-06,
+      "loss": 0.6147,
+      "step": 909
+    },
+    {
+      "epoch": 0.9934497816593887,
+      "grad_norm": 1.3972057243681786,
+      "learning_rate": 2.0006866337097385e-06,
+      "loss": 0.3797,
+      "step": 910
+    },
+    {
+      "epoch": 0.9945414847161572,
+      "grad_norm": 1.539646657802833,
+      "learning_rate": 2.0004768345232086e-06,
+      "loss": 0.4444,
+      "step": 911
+    },
+    {
+      "epoch": 0.9956331877729258,
+      "grad_norm": 1.448519957622226,
+      "learning_rate": 2.000305177005308e-06,
+      "loss": 0.496,
+      "step": 912
+    },
+    {
+      "epoch": 0.9967248908296943,
+      "grad_norm": 1.5861995286170092,
+      "learning_rate": 2.000171663338818e-06,
+      "loss": 0.5035,
+      "step": 913
+    },
+    {
+      "epoch": 0.9978165938864629,
+      "grad_norm": 1.5202816160223769,
+      "learning_rate": 2.000076295221487e-06,
+      "loss": 0.558,
+      "step": 914
+    },
+    {
+      "epoch": 0.9989082969432315,
+      "grad_norm": 1.4109310416901248,
+      "learning_rate": 2.0000190738660073e-06,
+      "loss": 0.5627,
+      "step": 915
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.541149773382253,
+      "learning_rate": 2e-06,
+      "loss": 0.4722,
+      "step": 916
+    },
+    {
+      "epoch": 1.0,
+      "step": 916,
+      "total_flos": 287687646904320.0,
+      "train_loss": 0.5413380475088498,
+      "train_runtime": 14548.2375,
+      "train_samples_per_second": 0.755,
+      "train_steps_per_second": 0.063
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 916,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 287687646904320.0,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}