diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,6123 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "global_step": 1015,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.225806451612903e-06,
+      "loss": 1.0096,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.451612903225806e-06,
+      "loss": 0.9944,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.67741935483871e-06,
+      "loss": 1.1679,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2903225806451613e-05,
+      "loss": 1.0488,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6129032258064517e-05,
+      "loss": 1.0446,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.935483870967742e-05,
+      "loss": 0.988,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.258064516129032e-05,
+      "loss": 0.9541,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5806451612903226e-05,
+      "loss": 0.9757,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9032258064516133e-05,
+      "loss": 0.9468,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2258064516129034e-05,
+      "loss": 0.8748,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.548387096774194e-05,
+      "loss": 0.9589,
+      "step": 11
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.870967741935484e-05,
+      "loss": 0.9508,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1935483870967746e-05,
+      "loss": 0.8802,
+      "step": 13
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.516129032258064e-05,
+      "loss": 0.8195,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.8387096774193554e-05,
+      "loss": 0.8317,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.161290322580645e-05,
+      "loss": 0.8283,
+      "step": 16
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.4838709677419355e-05,
+      "loss": 0.8281,
+      "step": 17
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.8064516129032266e-05,
+      "loss": 0.8478,
+      "step": 18
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.129032258064517e-05,
+      "loss": 0.771,
+      "step": 19
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.451612903225807e-05,
+      "loss": 0.7952,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.774193548387096e-05,
+      "loss": 0.8332,
+      "step": 21
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.096774193548388e-05,
+      "loss": 0.7989,
+      "step": 22
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.419354838709677e-05,
+      "loss": 0.7948,
+      "step": 23
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.741935483870968e-05,
+      "loss": 0.7996,
+      "step": 24
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.064516129032258e-05,
+      "loss": 0.7529,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 8.387096774193549e-05,
+      "loss": 0.7816,
+      "step": 26
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 8.709677419354839e-05,
+      "loss": 0.7356,
+      "step": 27
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.032258064516129e-05,
+      "loss": 0.7724,
+      "step": 28
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.35483870967742e-05,
+      "loss": 0.8279,
+      "step": 29
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.677419354838711e-05,
+      "loss": 0.7699,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0001,
+      "loss": 0.7442,
+      "step": 31
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999974517080149e-05,
+      "loss": 0.7235,
+      "step": 32
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999898068580345e-05,
+      "loss": 0.7769,
+      "step": 33
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999770655279843e-05,
+      "loss": 0.8342,
+      "step": 34
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999592278477388e-05,
+      "loss": 0.6822,
+      "step": 35
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999362939991202e-05,
+      "loss": 0.7573,
+      "step": 36
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999082642158973e-05,
+      "loss": 0.7347,
+      "step": 37
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.998751387837822e-05,
+      "loss": 0.6936,
+      "step": 38
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.998369180404283e-05,
+      "loss": 0.7514,
+      "step": 39
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.997936023754257e-05,
+      "loss": 0.7383,
+      "step": 40
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.997451922302986e-05,
+      "loss": 0.6676,
+      "step": 41
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.996916880984995e-05,
+      "loss": 0.6739,
+      "step": 42
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.99633090525405e-05,
+      "loss": 0.7704,
+      "step": 43
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.995694001083102e-05,
+      "loss": 0.7319,
+      "step": 44
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.99500617496422e-05,
+      "loss": 0.7985,
+      "step": 45
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.994267433908533e-05,
+      "loss": 0.6941,
+      "step": 46
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.99347778544615e-05,
+      "loss": 0.7071,
+      "step": 47
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.992637237626092e-05,
+      "loss": 0.7374,
+      "step": 48
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.991745799016206e-05,
+      "loss": 0.7035,
+      "step": 49
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.990803478703072e-05,
+      "loss": 0.7026,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.989810286291923e-05,
+      "loss": 0.7069,
+      "step": 51
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.988766231906533e-05,
+      "loss": 0.6708,
+      "step": 52
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.987671326189123e-05,
+      "loss": 0.692,
+      "step": 53
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.986525580300253e-05,
+      "loss": 0.6346,
+      "step": 54
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.985329005918702e-05,
+      "loss": 0.6542,
+      "step": 55
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.984081615241356e-05,
+      "loss": 0.6343,
+      "step": 56
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.982783420983075e-05,
+      "loss": 0.6672,
+      "step": 57
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.981434436376572e-05,
+      "loss": 0.693,
+      "step": 58
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.980034675172274e-05,
+      "loss": 0.7088,
+      "step": 59
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.978584151638181e-05,
+      "loss": 0.7524,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.977082880559725e-05,
+      "loss": 0.6479,
+      "step": 61
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.975530877239612e-05,
+      "loss": 0.6936,
+      "step": 62
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.973928157497674e-05,
+      "loss": 0.647,
+      "step": 63
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.972274737670701e-05,
+      "loss": 0.6543,
+      "step": 64
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.970570634612282e-05,
+      "loss": 0.7236,
+      "step": 65
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.968815865692622e-05,
+      "loss": 0.6759,
+      "step": 66
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.967010448798375e-05,
+      "loss": 0.6666,
+      "step": 67
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.965154402332464e-05,
+      "loss": 0.6738,
+      "step": 68
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.963247745213876e-05,
+      "loss": 0.6549,
+      "step": 69
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.96129049687749e-05,
+      "loss": 0.6958,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.95928267727387e-05,
+      "loss": 0.7584,
+      "step": 71
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.957224306869053e-05,
+      "loss": 0.6363,
+      "step": 72
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.955115406644356e-05,
+      "loss": 0.6442,
+      "step": 73
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.952955998096155e-05,
+      "loss": 0.6741,
+      "step": 74
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.950746103235663e-05,
+      "loss": 0.6581,
+      "step": 75
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.948485744588709e-05,
+      "loss": 0.5917,
+      "step": 76
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.946174945195507e-05,
+      "loss": 0.6559,
+      "step": 77
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.943813728610427e-05,
+      "loss": 0.7137,
+      "step": 78
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.941402118901744e-05,
+      "loss": 0.6469,
+      "step": 79
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.938940140651398e-05,
+      "loss": 0.6427,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.936427818954752e-05,
+      "loss": 0.6794,
+      "step": 81
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.933865179420321e-05,
+      "loss": 0.6676,
+      "step": 82
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.931252248169518e-05,
+      "loss": 0.5922,
+      "step": 83
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.928589051836392e-05,
+      "loss": 0.6565,
+      "step": 84
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.92587561756735e-05,
+      "loss": 0.6451,
+      "step": 85
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.923111973020885e-05,
+      "loss": 0.6265,
+      "step": 86
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.920298146367286e-05,
+      "loss": 0.6503,
+      "step": 87
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.917434166288364e-05,
+      "loss": 0.6391,
+      "step": 88
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.914520061977149e-05,
+      "loss": 0.6591,
+      "step": 89
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.911555863137594e-05,
+      "loss": 0.6477,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.908541599984276e-05,
+      "loss": 0.6695,
+      "step": 91
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.905477303242086e-05,
+      "loss": 0.6701,
+      "step": 92
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.902363004145914e-05,
+      "loss": 0.6776,
+      "step": 93
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.899198734440335e-05,
+      "loss": 0.6633,
+      "step": 94
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.895984526379281e-05,
+      "loss": 0.6146,
+      "step": 95
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.892720412725715e-05,
+      "loss": 0.6483,
+      "step": 96
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.889406426751296e-05,
+      "loss": 0.6251,
+      "step": 97
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.886042602236039e-05,
+      "loss": 0.6589,
+      "step": 98
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.882628973467972e-05,
+      "loss": 0.6782,
+      "step": 99
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.879165575242787e-05,
+      "loss": 0.7015,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.875652442863484e-05,
+      "loss": 0.6615,
+      "step": 101
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.872089612140008e-05,
+      "loss": 0.6355,
+      "step": 102
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.868477119388896e-05,
+      "loss": 0.6717,
+      "step": 103
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.86481500143289e-05,
+      "loss": 0.674,
+      "step": 104
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.861103295600573e-05,
+      "loss": 0.6255,
+      "step": 105
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.85734203972599e-05,
+      "loss": 0.6478,
+      "step": 106
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.853531272148249e-05,
+      "loss": 0.6413,
+      "step": 107
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.849671031711145e-05,
+      "loss": 0.6312,
+      "step": 108
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.84576135776276e-05,
+      "loss": 0.7004,
+      "step": 109
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.841802290155053e-05,
+      "loss": 0.6366,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.837793869243468e-05,
+      "loss": 0.6672,
+      "step": 111
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.833736135886512e-05,
+      "loss": 0.6206,
+      "step": 112
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.829629131445342e-05,
+      "loss": 0.6524,
+      "step": 113
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.825472897783343e-05,
+      "loss": 0.6081,
+      "step": 114
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.821267477265705e-05,
+      "loss": 0.6321,
+      "step": 115
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.817012912758985e-05,
+      "loss": 0.6108,
+      "step": 116
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.812709247630671e-05,
+      "loss": 0.6418,
+      "step": 117
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.808356525748748e-05,
+      "loss": 0.6569,
+      "step": 118
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.803954791481239e-05,
+      "loss": 0.6715,
+      "step": 119
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.799504089695761e-05,
+      "loss": 0.5868,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.795004465759065e-05,
+      "loss": 0.6411,
+      "step": 121
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.790455965536574e-05,
+      "loss": 0.6375,
+      "step": 122
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.785858635391914e-05,
+      "loss": 0.6132,
+      "step": 123
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.781212522186443e-05,
+      "loss": 0.6415,
+      "step": 124
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.776517673278772e-05,
+      "loss": 0.6415,
+      "step": 125
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.771774136524286e-05,
+      "loss": 0.6793,
+      "step": 126
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.766981960274653e-05,
+      "loss": 0.6348,
+      "step": 127
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.762141193377329e-05,
+      "loss": 0.6299,
+      "step": 128
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.757251885175063e-05,
+      "loss": 0.5762,
+      "step": 129
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.752314085505395e-05,
+      "loss": 0.6699,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.747327844700147e-05,
+      "loss": 0.6534,
+      "step": 131
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.74229321358491e-05,
+      "loss": 0.6244,
+      "step": 132
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.737210243478521e-05,
+      "loss": 0.6335,
+      "step": 133
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.732078986192552e-05,
+      "loss": 0.6844,
+      "step": 134
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.726899494030768e-05,
+      "loss": 0.609,
+      "step": 135
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.721671819788602e-05,
+      "loss": 0.6791,
+      "step": 136
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.716396016752615e-05,
+      "loss": 0.5922,
+      "step": 137
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.711072138699957e-05,
+      "loss": 0.6683,
+      "step": 138
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.705700239897809e-05,
+      "loss": 0.6376,
+      "step": 139
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.700280375102834e-05,
+      "loss": 0.6022,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.694812599560632e-05,
+      "loss": 0.6476,
+      "step": 141
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.68929696900515e-05,
+      "loss": 0.6353,
+      "step": 142
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.683733539658139e-05,
+      "loss": 0.6032,
+      "step": 143
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.67812236822857e-05,
+      "loss": 0.6191,
+      "step": 144
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.672463511912055e-05,
+      "loss": 0.6368,
+      "step": 145
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.666757028390266e-05,
+      "loss": 0.6672,
+      "step": 146
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.661002975830349e-05,
+      "loss": 0.6114,
+      "step": 147
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.655201412884327e-05,
+      "loss": 0.6349,
+      "step": 148
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.649352398688506e-05,
+      "loss": 0.6465,
+      "step": 149
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.64345599286287e-05,
+      "loss": 0.6231,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.637512255510475e-05,
+      "loss": 0.5874,
+      "step": 151
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.63152124721683e-05,
+      "loss": 0.66,
+      "step": 152
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.625483029049294e-05,
+      "loss": 0.61,
+      "step": 153
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.619397662556435e-05,
+      "loss": 0.6165,
+      "step": 154
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.613265209767417e-05,
+      "loss": 0.6148,
+      "step": 155
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.607085733191361e-05,
+      "loss": 0.6341,
+      "step": 156
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.600859295816708e-05,
+      "loss": 0.5915,
+      "step": 157
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.594585961110585e-05,
+      "loss": 0.654,
+      "step": 158
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.58826579301814e-05,
+      "loss": 0.6411,
+      "step": 159
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.581898855961912e-05,
+      "loss": 0.5904,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.575485214841157e-05,
+      "loss": 0.595,
+      "step": 161
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.569024935031198e-05,
+      "loss": 0.6914,
+      "step": 162
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.56251808238275e-05,
+      "loss": 0.6053,
+      "step": 163
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.555964723221258e-05,
+      "loss": 0.6598,
+      "step": 164
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.54936492434621e-05,
+      "loss": 0.619,
+      "step": 165
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.542718753030463e-05,
+      "loss": 0.6094,
+      "step": 166
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.536026277019561e-05,
+      "loss": 0.6026,
+      "step": 167
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.529287564531032e-05,
+      "loss": 0.6323,
+      "step": 168
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.522502684253709e-05,
+      "loss": 0.6005,
+      "step": 169
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.515671705347013e-05,
+      "loss": 0.6686,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.508794697440257e-05,
+      "loss": 0.6271,
+      "step": 171
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.501871730631942e-05,
+      "loss": 0.6502,
+      "step": 172
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.49490287548903e-05,
+      "loss": 0.5812,
+      "step": 173
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.487888203046231e-05,
+      "loss": 0.676,
+      "step": 174
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.480827784805278e-05,
+      "loss": 0.6245,
+      "step": 175
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.473721692734203e-05,
+      "loss": 0.6114,
+      "step": 176
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.466569999266594e-05,
+      "loss": 0.636,
+      "step": 177
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.459372777300864e-05,
+      "loss": 0.6285,
+      "step": 178
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.452130100199503e-05,
+      "loss": 0.6319,
+      "step": 179
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.444842041788338e-05,
+      "loss": 0.6073,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.437508676355773e-05,
+      "loss": 0.6383,
+      "step": 181
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.43013007865203e-05,
+      "loss": 0.6264,
+      "step": 182
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.422706323888397e-05,
+      "loss": 0.6256,
+      "step": 183
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.415237487736452e-05,
+      "loss": 0.6279,
+      "step": 184
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.407723646327298e-05,
+      "loss": 0.5873,
+      "step": 185
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.400164876250781e-05,
+      "loss": 0.6322,
+      "step": 186
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.392561254554713e-05,
+      "loss": 0.6531,
+      "step": 187
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.384912858744087e-05,
+      "loss": 0.6355,
+      "step": 188
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.377219766780287e-05,
+      "loss": 0.6333,
+      "step": 189
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.369482057080292e-05,
+      "loss": 0.6779,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.361699808515876e-05,
+      "loss": 0.6131,
+      "step": 191
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.353873100412805e-05,
+      "loss": 0.6125,
+      "step": 192
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.346002012550027e-05,
+      "loss": 0.63,
+      "step": 193
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.338086625158867e-05,
+      "loss": 0.5982,
+      "step": 194
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.330127018922194e-05,
+      "loss": 0.5919,
+      "step": 195
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.322123274973613e-05,
+      "loss": 0.6162,
+      "step": 196
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.314075474896631e-05,
+      "loss": 0.618,
+      "step": 197
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.305983700723824e-05,
+      "loss": 0.6567,
+      "step": 198
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.297848034936006e-05,
+      "loss": 0.5975,
+      "step": 199
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.289668560461384e-05,
+      "loss": 0.624,
+      "step": 200
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.281445360674717e-05,
+      "loss": 0.6248,
+      "step": 201
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.273178519396459e-05,
+      "loss": 0.6512,
+      "step": 202
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.264868120891912e-05,
+      "loss": 0.6872,
+      "step": 203
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.256514249870365e-05,
+      "loss": 0.6364,
+      "step": 204
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.248116991484229e-05,
+      "loss": 0.6088,
+      "step": 205
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.239676431328164e-05,
+      "loss": 0.7192,
+      "step": 206
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.231192655438221e-05,
+      "loss": 0.5991,
+      "step": 207
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.222665750290953e-05,
+      "loss": 0.6879,
+      "step": 208
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.214095802802534e-05,
+      "loss": 0.5814,
+      "step": 209
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.20548290032788e-05,
+      "loss": 0.6807,
+      "step": 210
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.19682713065975e-05,
+      "loss": 0.6252,
+      "step": 211
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.18812858202786e-05,
+      "loss": 0.6161,
+      "step": 212
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.179387343097977e-05,
+      "loss": 0.6506,
+      "step": 213
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.170603502971016e-05,
+      "loss": 0.6053,
+      "step": 214
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.161777151182136e-05,
+      "loss": 0.5975,
+      "step": 215
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.152908377699822e-05,
+      "loss": 0.6571,
+      "step": 216
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.143997272924973e-05,
+      "loss": 0.5967,
+      "step": 217
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.135043927689975e-05,
+      "loss": 0.644,
+      "step": 218
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.126048433257779e-05,
+      "loss": 0.5963,
+      "step": 219
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.117010881320973e-05,
+      "loss": 0.6178,
+      "step": 220
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.107931364000839e-05,
+      "loss": 0.6071,
+      "step": 221
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.098809973846425e-05,
+      "loss": 0.5899,
+      "step": 222
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.089646803833589e-05,
+      "loss": 0.5723,
+      "step": 223
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.080441947364064e-05,
+      "loss": 0.6222,
+      "step": 224
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.071195498264497e-05,
+      "loss": 0.6093,
+      "step": 225
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.061907550785498e-05,
+      "loss": 0.5738,
+      "step": 226
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.052578199600675e-05,
+      "loss": 0.5853,
+      "step": 227
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.04320753980567e-05,
+      "loss": 0.6473,
+      "step": 228
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.033795666917191e-05,
+      "loss": 0.6065,
+      "step": 229
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.024342676872043e-05,
+      "loss": 0.6703,
+      "step": 230
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.014848666026138e-05,
+      "loss": 0.6864,
+      "step": 231
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.005313731153524e-05,
+      "loss": 0.5884,
+      "step": 232
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.995737969445394e-05,
+      "loss": 0.5931,
+      "step": 233
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.986121478509096e-05,
+      "loss": 0.6571,
+      "step": 234
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.976464356367134e-05,
+      "loss": 0.661,
+      "step": 235
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.966766701456177e-05,
+      "loss": 0.5835,
+      "step": 236
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.95702861262605e-05,
+      "loss": 0.6734,
+      "step": 237
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.947250189138731e-05,
+      "loss": 0.606,
+      "step": 238
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.937431530667328e-05,
+      "loss": 0.5856,
+      "step": 239
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.92757273729508e-05,
+      "loss": 0.6284,
+      "step": 240
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.917673909514322e-05,
+      "loss": 0.5526,
+      "step": 241
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.907735148225465e-05,
+      "loss": 0.6407,
+      "step": 242
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.897756554735977e-05,
+      "loss": 0.5986,
+      "step": 243
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.887738230759333e-05,
+      "loss": 0.5807,
+      "step": 244
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.877680278413995e-05,
+      "loss": 0.5821,
+      "step": 245
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.867582800222358e-05,
+      "loss": 0.6089,
+      "step": 246
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.857445899109715e-05,
+      "loss": 0.592,
+      "step": 247
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.8472696784032e-05,
+      "loss": 0.606,
+      "step": 248
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.83705424183074e-05,
+      "loss": 0.6172,
+      "step": 249
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.826799693519996e-05,
+      "loss": 0.5878,
+      "step": 250
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.8165061379973e-05,
+      "loss": 0.5713,
+      "step": 251
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.806173680186593e-05,
+      "loss": 0.6061,
+      "step": 252
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.795802425408352e-05,
+      "loss": 0.5744,
+      "step": 253
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.785392479378521e-05,
+      "loss": 0.6243,
+      "step": 254
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.774943948207426e-05,
+      "loss": 0.5758,
+      "step": 255
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.7644569383987e-05,
+      "loss": 0.6559,
+      "step": 256
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.753931556848195e-05,
+      "loss": 0.5668,
+      "step": 257
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.743367910842894e-05,
+      "loss": 0.6137,
+      "step": 258
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.732766108059813e-05,
+      "loss": 0.6202,
+      "step": 259
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.722126256564911e-05,
+      "loss": 0.6303,
+      "step": 260
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.711448464811979e-05,
+      "loss": 0.591,
+      "step": 261
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.700732841641542e-05,
+      "loss": 0.5558,
+      "step": 262
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.689979496279746e-05,
+      "loss": 0.6066,
+      "step": 263
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.679188538337248e-05,
+      "loss": 0.6323,
+      "step": 264
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.668360077808093e-05,
+      "loss": 0.5949,
+      "step": 265
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.657494225068599e-05,
+      "loss": 0.5949,
+      "step": 266
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.646591090876224e-05,
+      "loss": 0.5917,
+      "step": 267
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.635650786368452e-05,
+      "loss": 0.544,
+      "step": 268
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.62467342306164e-05,
+      "loss": 0.6081,
+      "step": 269
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.613659112849896e-05,
+      "loss": 0.6385,
+      "step": 270
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.602607968003935e-05,
+      "loss": 0.6132,
+      "step": 271
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.591520101169932e-05,
+      "loss": 0.6243,
+      "step": 272
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.580395625368376e-05,
+      "loss": 0.6408,
+      "step": 273
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.569234653992916e-05,
+      "loss": 0.6679,
+      "step": 274
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.558037300809208e-05,
+      "loss": 0.64,
+      "step": 275
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.546803679953754e-05,
+      "loss": 0.6144,
+      "step": 276
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.535533905932738e-05,
+      "loss": 0.5899,
+      "step": 277
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.524228093620859e-05,
+      "loss": 0.6388,
+      "step": 278
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.512886358260162e-05,
+      "loss": 0.65,
+      "step": 279
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.501508815458855e-05,
+      "loss": 0.6227,
+      "step": 280
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.49009558119015e-05,
+      "loss": 0.6303,
+      "step": 281
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.478646771791054e-05,
+      "loss": 0.6207,
+      "step": 282
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.467162503961208e-05,
+      "loss": 0.6316,
+      "step": 283
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.455642894761684e-05,
+      "loss": 0.5995,
+      "step": 284
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.444088061613788e-05,
+      "loss": 0.6321,
+      "step": 285
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.432498122297878e-05,
+      "loss": 0.5882,
+      "step": 286
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.420873194952152e-05,
+      "loss": 0.6007,
+      "step": 287
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.409213398071447e-05,
+      "loss": 0.5974,
+      "step": 288
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.397518850506028e-05,
+      "loss": 0.6378,
+      "step": 289
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.385789671460386e-05,
+      "loss": 0.6056,
+      "step": 290
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.374025980492011e-05,
+      "loss": 0.5953,
+      "step": 291
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.36222789751018e-05,
+      "loss": 0.574,
+      "step": 292
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.350395542774736e-05,
+      "loss": 0.5908,
+      "step": 293
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.338529036894856e-05,
+      "loss": 0.6195,
+      "step": 294
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.326628500827826e-05,
+      "loss": 0.6435,
+      "step": 295
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.314694055877814e-05,
+      "loss": 0.5895,
+      "step": 296
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.302725823694618e-05,
+      "loss": 0.5496,
+      "step": 297
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.290723926272439e-05,
+      "loss": 0.627,
+      "step": 298
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.278688485948633e-05,
+      "loss": 0.5821,
+      "step": 299
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.266619625402465e-05,
+      "loss": 0.5868,
+      "step": 300
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.254517467653858e-05,
+      "loss": 0.5762,
+      "step": 301
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.242382136062134e-05,
+      "loss": 0.6811,
+      "step": 302
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.230213754324773e-05,
+      "loss": 0.6546,
+      "step": 303
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.218012446476128e-05,
+      "loss": 0.622,
+      "step": 304
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.205778336886182e-05,
+      "loss": 0.5442,
+      "step": 305
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.193511550259267e-05,
+      "loss": 0.6063,
+      "step": 306
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.181212211632799e-05,
+      "loss": 0.567,
+      "step": 307
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.168880446376003e-05,
+      "loss": 0.5632,
+      "step": 308
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.156516380188636e-05,
+      "loss": 0.6764,
+      "step": 309
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.144120139099697e-05,
+      "loss": 0.5993,
+      "step": 310
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.131691849466153e-05,
+      "loss": 0.6162,
+      "step": 311
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.119231637971651e-05,
+      "loss": 0.579,
+      "step": 312
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.106739631625217e-05,
+      "loss": 0.6314,
+      "step": 313
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.09421595775997e-05,
+      "loss": 0.6043,
+      "step": 314
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.081660744031819e-05,
+      "loss": 0.611,
+      "step": 315
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.06907411841817e-05,
+      "loss": 0.5617,
+      "step": 316
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.056456209216609e-05,
+      "loss": 0.5396,
+      "step": 317
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.043807145043604e-05,
+      "loss": 0.5615,
+      "step": 318
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.03112705483319e-05,
+      "loss": 0.586,
+      "step": 319
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 8.01841606783566e-05,
+      "loss": 0.592,
+      "step": 320
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 8.005674313616231e-05,
+      "loss": 0.5773,
+      "step": 321
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.992901922053752e-05,
+      "loss": 0.5994,
+      "step": 322
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.98009902333935e-05,
+      "loss": 0.6009,
+      "step": 323
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.967265747975123e-05,
+      "loss": 0.6054,
+      "step": 324
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.954402226772804e-05,
+      "loss": 0.5879,
+      "step": 325
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.941508590852423e-05,
+      "loss": 0.5954,
+      "step": 326
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.928584971640974e-05,
+      "loss": 0.621,
+      "step": 327
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.915631500871083e-05,
+      "loss": 0.6265,
+      "step": 328
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.90264831057965e-05,
+      "loss": 0.6192,
+      "step": 329
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.889635533106515e-05,
+      "loss": 0.5802,
+      "step": 330
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.876593301093104e-05,
+      "loss": 0.6511,
+      "step": 331
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.863521747481078e-05,
+      "loss": 0.5955,
+      "step": 332
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.850421005510977e-05,
+      "loss": 0.5973,
+      "step": 333
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.837291208720866e-05,
+      "loss": 0.6042,
+      "step": 334
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.824132490944967e-05,
+      "loss": 0.5843,
+      "step": 335
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.810944986312302e-05,
+      "loss": 0.6296,
+      "step": 336
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.797728829245321e-05,
+      "loss": 0.5901,
+      "step": 337
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.784484154458529e-05,
+      "loss": 0.5552,
+      "step": 338
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.771211096957125e-05,
+      "loss": 0.6132,
+      "step": 339
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.757909792035608e-05,
+      "loss": 0.5825,
+      "step": 340
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.744580375276416e-05,
+      "loss": 0.5957,
+      "step": 341
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.731222982548534e-05,
+      "loss": 0.6702,
+      "step": 342
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.717837750006106e-05,
+      "loss": 0.595,
+      "step": 343
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.704424814087057e-05,
+      "loss": 0.6249,
+      "step": 344
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.690984311511695e-05,
+      "loss": 0.5861,
+      "step": 345
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.677516379281321e-05,
+      "loss": 0.5083,
+      "step": 346
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.664021154676829e-05,
+      "loss": 0.5969,
+      "step": 347
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.650498775257309e-05,
+      "loss": 0.5887,
+      "step": 348
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.636949378858646e-05,
+      "loss": 0.586,
+      "step": 349
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.623373103592117e-05,
+      "loss": 0.5672,
+      "step": 350
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.609770087842969e-05,
+      "loss": 0.6095,
+      "step": 351
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.596140470269029e-05,
+      "loss": 0.5614,
+      "step": 352
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.582484389799278e-05,
+      "loss": 0.6383,
+      "step": 353
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.568801985632439e-05,
+      "loss": 0.6188,
+      "step": 354
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.555093397235552e-05,
+      "loss": 0.6059,
+      "step": 355
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.541358764342565e-05,
+      "loss": 0.5827,
+      "step": 356
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.527598226952894e-05,
+      "loss": 0.6178,
+      "step": 357
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.51381192533001e-05,
+      "loss": 0.5847,
+      "step": 358
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.5933,
+      "step": 359
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.486162591750138e-05,
+      "loss": 0.5939,
+      "step": 360
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.472299841627451e-05,
+      "loss": 0.5541,
+      "step": 361
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.45841189093728e-05,
+      "loss": 0.6037,
+      "step": 362
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.444498881241835e-05,
+      "loss": 0.5862,
+      "step": 363
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.430560954358764e-05,
+      "loss": 0.55,
+      "step": 364
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.416598252359694e-05,
+      "loss": 0.6235,
+      "step": 365
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.402610917568794e-05,
+      "loss": 0.5755,
+      "step": 366
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.388599092561315e-05,
+      "loss": 0.5998,
+      "step": 367
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.374562920162142e-05,
+      "loss": 0.6086,
+      "step": 368
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.360502543444338e-05,
+      "loss": 0.5528,
+      "step": 369
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.346418105727686e-05,
+      "loss": 0.6233,
+      "step": 370
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.332309750577222e-05,
+      "loss": 0.5943,
+      "step": 371
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.31817762180178e-05,
+      "loss": 0.587,
+      "step": 372
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.304021863452524e-05,
+      "loss": 0.5771,
+      "step": 373
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.289842619821475e-05,
+      "loss": 0.546,
+      "step": 374
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.275640035440045e-05,
+      "loss": 0.6116,
+      "step": 375
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.26141425507756e-05,
+      "loss": 0.5706,
+      "step": 376
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.24716542373979e-05,
+      "loss": 0.5593,
+      "step": 377
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.232893686667466e-05,
+      "loss": 0.5684,
+      "step": 378
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.218599189334799e-05,
+      "loss": 0.6221,
+      "step": 379
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.204282077448002e-05,
+      "loss": 0.622,
+      "step": 380
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.189942496943803e-05,
+      "loss": 0.6153,
+      "step": 381
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.175580593987951e-05,
+      "loss": 0.6249,
+      "step": 382
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.161196514973734e-05,
+      "loss": 0.6031,
+      "step": 383
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.14679040652049e-05,
+      "loss": 0.5638,
+      "step": 384
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.1323624154721e-05,
+      "loss": 0.6023,
+      "step": 385
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.117912688895495e-05,
+      "loss": 0.5987,
+      "step": 386
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.10344137407917e-05,
+      "loss": 0.5726,
+      "step": 387
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.088948618531667e-05,
+      "loss": 0.5621,
+      "step": 388
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.074434569980075e-05,
+      "loss": 0.5767,
+      "step": 389
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.05989937636853e-05,
+      "loss": 0.6063,
+      "step": 390
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.045343185856701e-05,
+      "loss": 0.5476,
+      "step": 391
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.03076614681828e-05,
+      "loss": 0.6066,
+      "step": 392
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.016168407839478e-05,
+      "loss": 0.5837,
+      "step": 393
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.0015501177175e-05,
+      "loss": 0.5339,
+      "step": 394
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 6.986911425459028e-05,
+      "loss": 0.6202,
+      "step": 395
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 6.972252480278713e-05,
+      "loss": 0.5303,
+      "step": 396
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 6.957573431597646e-05,
+      "loss": 0.561,
+      "step": 397
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 6.942874429041833e-05,
+      "loss": 0.5862,
+      "step": 398
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 6.92815562244068e-05,
+      "loss": 0.5895,
+      "step": 399
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 6.91341716182545e-05,
+      "loss": 0.5908,
+      "step": 400
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.898659197427748e-05,
+      "loss": 0.6063,
+      "step": 401
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.883881879677987e-05,
+      "loss": 0.6266,
+      "step": 402
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.869085359203844e-05,
+      "loss": 0.6089,
+      "step": 403
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.85426978682874e-05,
+      "loss": 0.6032,
+      "step": 404
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.839435313570293e-05,
+      "loss": 0.5379,
+      "step": 405
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.824582090638777e-05,
+      "loss": 0.6127,
+      "step": 406
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.809710269435589e-05,
+      "loss": 0.5964,
+      "step": 407
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.794820001551703e-05,
+      "loss": 0.5558,
+      "step": 408
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.779911438766116e-05,
+      "loss": 0.5178,
+      "step": 409
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.764984733044316e-05,
+      "loss": 0.5372,
+      "step": 410
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.750040036536718e-05,
+      "loss": 0.5735,
+      "step": 411
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.735077501577126e-05,
+      "loss": 0.6026,
+      "step": 412
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.72009728068117e-05,
+      "loss": 0.6029,
+      "step": 413
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.705099526544757e-05,
+      "loss": 0.5953,
+      "step": 414
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.690084392042513e-05,
+      "loss": 0.5679,
+      "step": 415
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.67505203022623e-05,
+      "loss": 0.5808,
+      "step": 416
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.660002594323293e-05,
+      "loss": 0.5633,
+      "step": 417
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.644936237735128e-05,
+      "loss": 0.5811,
+      "step": 418
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.629853114035642e-05,
+      "loss": 0.5865,
+      "step": 419
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.614753376969647e-05,
+      "loss": 0.5376,
+      "step": 420
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.599637180451294e-05,
+      "loss": 0.5553,
+      "step": 421
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.584504678562521e-05,
+      "loss": 0.613,
+      "step": 422
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.569356025551454e-05,
+      "loss": 0.6257,
+      "step": 423
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.55419137583086e-05,
+      "loss": 0.6456,
+      "step": 424
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.539010883976562e-05,
+      "loss": 0.6071,
+      "step": 425
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.523814704725861e-05,
+      "loss": 0.6283,
+      "step": 426
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.508602992975963e-05,
+      "loss": 0.5936,
+      "step": 427
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.493375903782402e-05,
+      "loss": 0.5787,
+      "step": 428
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.478133592357455e-05,
+      "loss": 0.5047,
+      "step": 429
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.462876214068562e-05,
+      "loss": 0.5937,
+      "step": 430
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.447603924436744e-05,
+      "loss": 0.5724,
+      "step": 431
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.432316879135012e-05,
+      "loss": 0.5647,
+      "step": 432
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.417015233986786e-05,
+      "loss": 0.5805,
+      "step": 433
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.401699144964307e-05,
+      "loss": 0.595,
+      "step": 434
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.38636876818704e-05,
+      "loss": 0.5649,
+      "step": 435
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.371024259920091e-05,
+      "loss": 0.6284,
+      "step": 436
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.355665776572611e-05,
+      "loss": 0.5849,
+      "step": 437
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.340293474696199e-05,
+      "loss": 0.5683,
+      "step": 438
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.32490751098331e-05,
+      "loss": 0.5621,
+      "step": 439
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.309508042265654e-05,
+      "loss": 0.5535,
+      "step": 440
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.294095225512603e-05,
+      "loss": 0.6141,
+      "step": 441
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.27866921782959e-05,
+      "loss": 0.5449,
+      "step": 442
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.263230176456498e-05,
+      "loss": 0.5957,
+      "step": 443
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.247778258766068e-05,
+      "loss": 0.5236,
+      "step": 444
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.232313622262296e-05,
+      "loss": 0.5646,
+      "step": 445
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.216836424578816e-05,
+      "loss": 0.5769,
+      "step": 446
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.201346823477303e-05,
+      "loss": 0.5718,
+      "step": 447
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.185844976845866e-05,
+      "loss": 0.5625,
+      "step": 448
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.170331042697425e-05,
+      "loss": 0.55,
+      "step": 449
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.154805179168121e-05,
+      "loss": 0.5898,
+      "step": 450
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.139267544515689e-05,
+      "loss": 0.5985,
+      "step": 451
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.123718297117845e-05,
+      "loss": 0.5617,
+      "step": 452
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.10815759547068e-05,
+      "loss": 0.554,
+      "step": 453
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.09258559818704e-05,
+      "loss": 0.5673,
+      "step": 454
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.0770024639949074e-05,
+      "loss": 0.5716,
+      "step": 455
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.0614083517357864e-05,
+      "loss": 0.546,
+      "step": 456
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.045803420363084e-05,
+      "loss": 0.5989,
+      "step": 457
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.0301878289404845e-05,
+      "loss": 0.6068,
+      "step": 458
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.014561736640334e-05,
+      "loss": 0.5801,
+      "step": 459
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.998925302742017e-05,
+      "loss": 0.5655,
+      "step": 460
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.983278686630327e-05,
+      "loss": 0.6082,
+      "step": 461
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.967622047793853e-05,
+      "loss": 0.613,
+      "step": 462
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.951955545823342e-05,
+      "loss": 0.6079,
+      "step": 463
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.936279340410081e-05,
+      "loss": 0.5899,
+      "step": 464
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.920593591344263e-05,
+      "loss": 0.5661,
+      "step": 465
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.9048984585133646e-05,
+      "loss": 0.5721,
+      "step": 466
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.889194101900509e-05,
+      "loss": 0.6085,
+      "step": 467
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.8734806815828404e-05,
+      "loss": 0.5602,
+      "step": 468
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.8577583577298924e-05,
+      "loss": 0.6018,
+      "step": 469
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.84202729060195e-05,
+      "loss": 0.5759,
+      "step": 470
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.826287640548425e-05,
+      "loss": 0.5825,
+      "step": 471
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.810539568006213e-05,
+      "loss": 0.583,
+      "step": 472
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.7947832334980614e-05,
+      "loss": 0.584,
+      "step": 473
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.779018797630934e-05,
+      "loss": 0.5954,
+      "step": 474
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.7632464210943726e-05,
+      "loss": 0.6305,
+      "step": 475
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.747466264658863e-05,
+      "loss": 0.5756,
+      "step": 476
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.7316784891741856e-05,
+      "loss": 0.6044,
+      "step": 477
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.71588325556779e-05,
+      "loss": 0.5662,
+      "step": 478
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.700080724843147e-05,
+      "loss": 0.5665,
+      "step": 479
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.684271058078101e-05,
+      "loss": 0.6109,
+      "step": 480
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.668454416423242e-05,
+      "loss": 0.6113,
+      "step": 481
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.6526309611002594e-05,
+      "loss": 0.5755,
+      "step": 482
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.636800853400285e-05,
+      "loss": 0.592,
+      "step": 483
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.620964254682266e-05,
+      "loss": 0.5917,
+      "step": 484
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.6051213263713156e-05,
+      "loss": 0.5476,
+      "step": 485
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.589272229957061e-05,
+      "loss": 0.6072,
+      "step": 486
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.573417126992003e-05,
+      "loss": 0.5326,
+      "step": 487
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.55755617908987e-05,
+      "loss": 0.5432,
+      "step": 488
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.541689547923966e-05,
+      "loss": 0.6199,
+      "step": 489
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.52581739522553e-05,
+      "loss": 0.5852,
+      "step": 490
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.509939882782077e-05,
+      "loss": 0.5538,
+      "step": 491
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.49405717243576e-05,
+      "loss": 0.5625,
+      "step": 492
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.478169426081712e-05,
+      "loss": 0.5473,
+      "step": 493
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.4622768056664e-05,
+      "loss": 0.5753,
+      "step": 494
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.446379473185972e-05,
+      "loss": 0.5639,
+      "step": 495
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.43047759068461e-05,
+      "loss": 0.621,
+      "step": 496
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.414571320252869e-05,
+      "loss": 0.5926,
+      "step": 497
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.398660824026039e-05,
+      "loss": 0.5648,
+      "step": 498
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.38274626418248e-05,
+      "loss": 0.5538,
+      "step": 499
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.3668278029419686e-05,
+      "loss": 0.5772,
+      "step": 500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.350905602564057e-05,
+      "loss": 0.5619,
+      "step": 501
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.334979825346409e-05,
+      "loss": 0.5457,
+      "step": 502
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.319050633623142e-05,
+      "loss": 0.5543,
+      "step": 503
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.3031181897631866e-05,
+      "loss": 0.5897,
+      "step": 504
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.287182656168618e-05,
+      "loss": 0.6105,
+      "step": 505
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.271244195273002e-05,
+      "loss": 0.6149,
+      "step": 506
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.2553029695397525e-05,
+      "loss": 0.5676,
+      "step": 507
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.2393591414604604e-05,
+      "loss": 0.5944,
+      "step": 508
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.2234128735532395e-05,
+      "loss": 0.6445,
+      "step": 509
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.207464328361078e-05,
+      "loss": 0.5683,
+      "step": 510
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.191513668450178e-05,
+      "loss": 0.6096,
+      "step": 511
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.175561056408291e-05,
+      "loss": 0.5649,
+      "step": 512
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.159606654843072e-05,
+      "loss": 0.5808,
+      "step": 513
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.143650626380416e-05,
+      "loss": 0.6313,
+      "step": 514
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.127693133662801e-05,
+      "loss": 0.5658,
+      "step": 515
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.111734339347629e-05,
+      "loss": 0.5753,
+      "step": 516
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.095774406105571e-05,
+      "loss": 0.6195,
+      "step": 517
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.079813496618908e-05,
+      "loss": 0.5849,
+      "step": 518
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.0638517735798696e-05,
+      "loss": 0.5831,
+      "step": 519
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.0478893996889796e-05,
+      "loss": 0.5582,
+      "step": 520
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.0319265376533964e-05,
+      "loss": 0.5185,
+      "step": 521
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.015963350185253e-05,
+      "loss": 0.5689,
+      "step": 522
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5e-05,
+      "loss": 0.6145,
+      "step": 523
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.984036649814749e-05,
+      "loss": 0.5704,
+      "step": 524
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.968073462346605e-05,
+      "loss": 0.5127,
+      "step": 525
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.9521106003110216e-05,
+      "loss": 0.5992,
+      "step": 526
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.936148226420132e-05,
+      "loss": 0.5668,
+      "step": 527
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.920186503381093e-05,
+      "loss": 0.6318,
+      "step": 528
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.9042255938944296e-05,
+      "loss": 0.5816,
+      "step": 529
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.8882656606523734e-05,
+      "loss": 0.5896,
+      "step": 530
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.8723068663372006e-05,
+      "loss": 0.5534,
+      "step": 531
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.856349373619585e-05,
+      "loss": 0.6282,
+      "step": 532
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.84039334515693e-05,
+      "loss": 0.6118,
+      "step": 533
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.82443894359171e-05,
+      "loss": 0.5798,
+      "step": 534
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.8084863315498234e-05,
+      "loss": 0.5431,
+      "step": 535
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.792535671638923e-05,
+      "loss": 0.6097,
+      "step": 536
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.776587126446761e-05,
+      "loss": 0.5484,
+      "step": 537
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.760640858539541e-05,
+      "loss": 0.5415,
+      "step": 538
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.744697030460248e-05,
+      "loss": 0.5669,
+      "step": 539
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.728755804726998e-05,
+      "loss": 0.5588,
+      "step": 540
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.712817343831384e-05,
+      "loss": 0.6145,
+      "step": 541
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.696881810236815e-05,
+      "loss": 0.5715,
+      "step": 542
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.680949366376858e-05,
+      "loss": 0.5883,
+      "step": 543
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.665020174653592e-05,
+      "loss": 0.5971,
+      "step": 544
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.649094397435944e-05,
+      "loss": 0.5581,
+      "step": 545
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.633172197058034e-05,
+      "loss": 0.5936,
+      "step": 546
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.6172537358175214e-05,
+      "loss": 0.5448,
+      "step": 547
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.601339175973961e-05,
+      "loss": 0.5817,
+      "step": 548
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.585428679747132e-05,
+      "loss": 0.5846,
+      "step": 549
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.569522409315392e-05,
+      "loss": 0.556,
+      "step": 550
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.5536205268140294e-05,
+      "loss": 0.5799,
+      "step": 551
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.537723194333602e-05,
+      "loss": 0.6071,
+      "step": 552
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.521830573918289e-05,
+      "loss": 0.5573,
+      "step": 553
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.505942827564241e-05,
+      "loss": 0.5169,
+      "step": 554
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.4900601172179244e-05,
+      "loss": 0.5962,
+      "step": 555
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.474182604774471e-05,
+      "loss": 0.5597,
+      "step": 556
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.458310452076034e-05,
+      "loss": 0.5526,
+      "step": 557
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.442443820910133e-05,
+      "loss": 0.53,
+      "step": 558
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.4265828730079987e-05,
+      "loss": 0.5916,
+      "step": 559
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.410727770042941e-05,
+      "loss": 0.5789,
+      "step": 560
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.394878673628686e-05,
+      "loss": 0.5934,
+      "step": 561
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.379035745317735e-05,
+      "loss": 0.5456,
+      "step": 562
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.363199146599717e-05,
+      "loss": 0.5904,
+      "step": 563
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.347369038899744e-05,
+      "loss": 0.6391,
+      "step": 564
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.331545583576758e-05,
+      "loss": 0.6026,
+      "step": 565
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.3157289419219005e-05,
+      "loss": 0.5438,
+      "step": 566
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.2999192751568564e-05,
+      "loss": 0.6065,
+      "step": 567
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.28411674443221e-05,
+      "loss": 0.5455,
+      "step": 568
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.268321510825815e-05,
+      "loss": 0.5866,
+      "step": 569
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.252533735341139e-05,
+      "loss": 0.5703,
+      "step": 570
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.236753578905627e-05,
+      "loss": 0.5135,
+      "step": 571
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.220981202369067e-05,
+      "loss": 0.5781,
+      "step": 572
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.2052167665019404e-05,
+      "loss": 0.6279,
+      "step": 573
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.189460431993788e-05,
+      "loss": 0.5634,
+      "step": 574
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.1737123594515756e-05,
+      "loss": 0.5826,
+      "step": 575
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.1579727093980506e-05,
+      "loss": 0.575,
+      "step": 576
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.142241642270108e-05,
+      "loss": 0.6119,
+      "step": 577
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.126519318417161e-05,
+      "loss": 0.5825,
+      "step": 578
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.110805898099493e-05,
+      "loss": 0.5714,
+      "step": 579
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.095101541486636e-05,
+      "loss": 0.6145,
+      "step": 580
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.079406408655737e-05,
+      "loss": 0.5506,
+      "step": 581
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.06372065958992e-05,
+      "loss": 0.6151,
+      "step": 582
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.0480444541766576e-05,
+      "loss": 0.5394,
+      "step": 583
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.032377952206148e-05,
+      "loss": 0.618,
+      "step": 584
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.016721313369674e-05,
+      "loss": 0.5777,
+      "step": 585
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.001074697257986e-05,
+      "loss": 0.5703,
+      "step": 586
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 3.9854382633596664e-05,
+      "loss": 0.5362,
+      "step": 587
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 3.969812171059516e-05,
+      "loss": 0.5628,
+      "step": 588
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 3.954196579636918e-05,
+      "loss": 0.5766,
+      "step": 589
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 3.938591648264214e-05,
+      "loss": 0.5586,
+      "step": 590
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 3.922997536005094e-05,
+      "loss": 0.5889,
+      "step": 591
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 3.907414401812963e-05,
+      "loss": 0.5604,
+      "step": 592
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 3.891842404529321e-05,
+      "loss": 0.5297,
+      "step": 593
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.8762817028821564e-05,
+      "loss": 0.6072,
+      "step": 594
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.8607324554843136e-05,
+      "loss": 0.562,
+      "step": 595
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.845194820831878e-05,
+      "loss": 0.6005,
+      "step": 596
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.8296689573025756e-05,
+      "loss": 0.5684,
+      "step": 597
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.814155023154136e-05,
+      "loss": 0.5631,
+      "step": 598
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.7986531765226964e-05,
+      "loss": 0.6063,
+      "step": 599
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.783163575421185e-05,
+      "loss": 0.6216,
+      "step": 600
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.7676863777377054e-05,
+      "loss": 0.5871,
+      "step": 601
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.752221741233931e-05,
+      "loss": 0.5922,
+      "step": 602
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.7367698235435036e-05,
+      "loss": 0.5973,
+      "step": 603
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.721330782170411e-05,
+      "loss": 0.544,
+      "step": 604
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.705904774487396e-05,
+      "loss": 0.5758,
+      "step": 605
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.6904919577343476e-05,
+      "loss": 0.5237,
+      "step": 606
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.675092489016693e-05,
+      "loss": 0.5977,
+      "step": 607
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.6597065253038015e-05,
+      "loss": 0.5959,
+      "step": 608
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.64433422342739e-05,
+      "loss": 0.5602,
+      "step": 609
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.62897574007991e-05,
+      "loss": 0.5238,
+      "step": 610
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.61363123181296e-05,
+      "loss": 0.5655,
+      "step": 611
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.5983008550356946e-05,
+      "loss": 0.5461,
+      "step": 612
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.582984766013215e-05,
+      "loss": 0.559,
+      "step": 613
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.5676831208649885e-05,
+      "loss": 0.5565,
+      "step": 614
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.5523960755632574e-05,
+      "loss": 0.5758,
+      "step": 615
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.537123785931439e-05,
+      "loss": 0.5833,
+      "step": 616
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.5218664076425455e-05,
+      "loss": 0.5956,
+      "step": 617
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.5066240962176e-05,
+      "loss": 0.5945,
+      "step": 618
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.4913970070240386e-05,
+      "loss": 0.6321,
+      "step": 619
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.4761852952741404e-05,
+      "loss": 0.6075,
+      "step": 620
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.4609891160234386e-05,
+      "loss": 0.5874,
+      "step": 621
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.4458086241691415e-05,
+      "loss": 0.5703,
+      "step": 622
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.4306439744485454e-05,
+      "loss": 0.5554,
+      "step": 623
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.4154953214374795e-05,
+      "loss": 0.6022,
+      "step": 624
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.4003628195487057e-05,
+      "loss": 0.5806,
+      "step": 625
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.385246623030356e-05,
+      "loss": 0.5397,
+      "step": 626
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.370146885964358e-05,
+      "loss": 0.5616,
+      "step": 627
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.355063762264873e-05,
+      "loss": 0.5967,
+      "step": 628
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.339997405676709e-05,
+      "loss": 0.5536,
+      "step": 629
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.324947969773771e-05,
+      "loss": 0.5829,
+      "step": 630
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.309915607957487e-05,
+      "loss": 0.6141,
+      "step": 631
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.2949004734552444e-05,
+      "loss": 0.5826,
+      "step": 632
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.2799027193188316e-05,
+      "loss": 0.5896,
+      "step": 633
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.2649224984228756e-05,
+      "loss": 0.6202,
+      "step": 634
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.249959963463283e-05,
+      "loss": 0.5548,
+      "step": 635
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.235015266955684e-05,
+      "loss": 0.5614,
+      "step": 636
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.2200885612338845e-05,
+      "loss": 0.5642,
+      "step": 637
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.205179998448298e-05,
+      "loss": 0.5908,
+      "step": 638
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.1902897305644095e-05,
+      "loss": 0.6035,
+      "step": 639
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.175417909361225e-05,
+      "loss": 0.5686,
+      "step": 640
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.160564686429709e-05,
+      "loss": 0.558,
+      "step": 641
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.145730213171259e-05,
+      "loss": 0.6503,
+      "step": 642
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.130914640796157e-05,
+      "loss": 0.5802,
+      "step": 643
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.116118120322015e-05,
+      "loss": 0.6125,
+      "step": 644
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.101340802572251e-05,
+      "loss": 0.587,
+      "step": 645
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 0.5675,
+      "step": 646
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.071844377559323e-05,
+      "loss": 0.5932,
+      "step": 647
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.057125570958168e-05,
+      "loss": 0.5953,
+      "step": 648
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.0424265684023558e-05,
+      "loss": 0.551,
+      "step": 649
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.0277475197212897e-05,
+      "loss": 0.5983,
+      "step": 650
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.013088574540974e-05,
+      "loss": 0.6117,
+      "step": 651
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.9984498822825025e-05,
+      "loss": 0.5665,
+      "step": 652
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.9838315921605236e-05,
+      "loss": 0.5289,
+      "step": 653
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.9692338531817206e-05,
+      "loss": 0.5433,
+      "step": 654
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.9546568141433006e-05,
+      "loss": 0.5402,
+      "step": 655
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.940100623631472e-05,
+      "loss": 0.5632,
+      "step": 656
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.9255654300199253e-05,
+      "loss": 0.5811,
+      "step": 657
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.9110513814683333e-05,
+      "loss": 0.5704,
+      "step": 658
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.8965586259208295e-05,
+      "loss": 0.4795,
+      "step": 659
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.8820873111045044e-05,
+      "loss": 0.5581,
+      "step": 660
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.8676375845279013e-05,
+      "loss": 0.5313,
+      "step": 661
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.85320959347951e-05,
+      "loss": 0.5227,
+      "step": 662
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.8388034850262646e-05,
+      "loss": 0.5722,
+      "step": 663
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.8244194060120498e-05,
+      "loss": 0.6306,
+      "step": 664
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.810057503056198e-05,
+      "loss": 0.5629,
+      "step": 665
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.795717922551998e-05,
+      "loss": 0.5715,
+      "step": 666
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7814008106652012e-05,
+      "loss": 0.5915,
+      "step": 667
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7671063133325346e-05,
+      "loss": 0.5895,
+      "step": 668
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7528345762602125e-05,
+      "loss": 0.6113,
+      "step": 669
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7385857449224405e-05,
+      "loss": 0.5967,
+      "step": 670
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7243599645599576e-05,
+      "loss": 0.5754,
+      "step": 671
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7101573801785268e-05,
+      "loss": 0.532,
+      "step": 672
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.6959781365474758e-05,
+      "loss": 0.5875,
+      "step": 673
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.681822378198221e-05,
+      "loss": 0.5583,
+      "step": 674
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.6676902494227795e-05,
+      "loss": 0.5467,
+      "step": 675
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.6535818942723144e-05,
+      "loss": 0.5427,
+      "step": 676
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.639497456555663e-05,
+      "loss": 0.6062,
+      "step": 677
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.625437079837859e-05,
+      "loss": 0.5915,
+      "step": 678
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.6114009074386846e-05,
+      "loss": 0.5837,
+      "step": 679
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.5973890824312076e-05,
+      "loss": 0.5773,
+      "step": 680
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.5834017476403073e-05,
+      "loss": 0.5643,
+      "step": 681
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.569439045641236e-05,
+      "loss": 0.5587,
+      "step": 682
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.555501118758167e-05,
+      "loss": 0.5024,
+      "step": 683
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.5415881090627226e-05,
+      "loss": 0.5782,
+      "step": 684
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.527700158372548e-05,
+      "loss": 0.588,
+      "step": 685
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.513837408249863e-05,
+      "loss": 0.5588,
+      "step": 686
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 0.5063,
+      "step": 687
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.486188074669989e-05,
+      "loss": 0.5688,
+      "step": 688
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.4724017730471072e-05,
+      "loss": 0.5739,
+      "step": 689
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.4586412356574367e-05,
+      "loss": 0.618,
+      "step": 690
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.4449066027644475e-05,
+      "loss": 0.6024,
+      "step": 691
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.4311980143675635e-05,
+      "loss": 0.5666,
+      "step": 692
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.4175156102007234e-05,
+      "loss": 0.5851,
+      "step": 693
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.4038595297309713e-05,
+      "loss": 0.5945,
+      "step": 694
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.3902299121570333e-05,
+      "loss": 0.5698,
+      "step": 695
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.376626896407886e-05,
+      "loss": 0.5464,
+      "step": 696
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.363050621141354e-05,
+      "loss": 0.5838,
+      "step": 697
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.3495012247426917e-05,
+      "loss": 0.564,
+      "step": 698
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.3359788453231724e-05,
+      "loss": 0.5732,
+      "step": 699
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.32248362071868e-05,
+      "loss": 0.5514,
+      "step": 700
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.3090156884883052e-05,
+      "loss": 0.5391,
+      "step": 701
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.295575185912944e-05,
+      "loss": 0.5309,
+      "step": 702
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.282162249993895e-05,
+      "loss": 0.6087,
+      "step": 703
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.2687770174514674e-05,
+      "loss": 0.581,
+      "step": 704
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.2554196247235843e-05,
+      "loss": 0.5863,
+      "step": 705
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.242090207964393e-05,
+      "loss": 0.5242,
+      "step": 706
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.228788903042877e-05,
+      "loss": 0.6191,
+      "step": 707
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.2155158455414714e-05,
+      "loss": 0.5353,
+      "step": 708
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.20227117075468e-05,
+      "loss": 0.6004,
+      "step": 709
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.189055013687698e-05,
+      "loss": 0.5758,
+      "step": 710
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.175867509055033e-05,
+      "loss": 0.5331,
+      "step": 711
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.162708791279135e-05,
+      "loss": 0.5621,
+      "step": 712
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.149578994489024e-05,
+      "loss": 0.5717,
+      "step": 713
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.1364782525189236e-05,
+      "loss": 0.5785,
+      "step": 714
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.1234066989068972e-05,
+      "loss": 0.5828,
+      "step": 715
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.1103644668934857e-05,
+      "loss": 0.5829,
+      "step": 716
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.0973516894203503e-05,
+      "loss": 0.5997,
+      "step": 717
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.0843684991289175e-05,
+      "loss": 0.5881,
+      "step": 718
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.071415028359026e-05,
+      "loss": 0.5449,
+      "step": 719
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.058491409147579e-05,
+      "loss": 0.629,
+      "step": 720
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.0455977732271993e-05,
+      "loss": 0.5469,
+      "step": 721
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.032734252024877e-05,
+      "loss": 0.5841,
+      "step": 722
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.0199009766606507e-05,
+      "loss": 0.566,
+      "step": 723
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.007098077946251e-05,
+      "loss": 0.5883,
+      "step": 724
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.994325686383769e-05,
+      "loss": 0.5717,
+      "step": 725
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.981583932164342e-05,
+      "loss": 0.5361,
+      "step": 726
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9688729451668114e-05,
+      "loss": 0.5647,
+      "step": 727
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9561928549563968e-05,
+      "loss": 0.5326,
+      "step": 728
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.943543790783392e-05,
+      "loss": 0.5501,
+      "step": 729
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9309258815818326e-05,
+      "loss": 0.6012,
+      "step": 730
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9183392559681812e-05,
+      "loss": 0.5951,
+      "step": 731
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.905784042240032e-05,
+      "loss": 0.5709,
+      "step": 732
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.893260368374786e-05,
+      "loss": 0.5487,
+      "step": 733
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.8807683620283494e-05,
+      "loss": 0.5636,
+      "step": 734
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.868308150533847e-05,
+      "loss": 0.5533,
+      "step": 735
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.8558798609003052e-05,
+      "loss": 0.5622,
+      "step": 736
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.843483619811364e-05,
+      "loss": 0.5718,
+      "step": 737
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.8311195536239955e-05,
+      "loss": 0.5652,
+      "step": 738
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.818787788367202e-05,
+      "loss": 0.5529,
+      "step": 739
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.8064884497407332e-05,
+      "loss": 0.5623,
+      "step": 740
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7942216631138187e-05,
+      "loss": 0.597,
+      "step": 741
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.781987553523874e-05,
+      "loss": 0.5436,
+      "step": 742
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7697862456752273e-05,
+      "loss": 0.5815,
+      "step": 743
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.757617863937865e-05,
+      "loss": 0.5678,
+      "step": 744
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7454825323461448e-05,
+      "loss": 0.5144,
+      "step": 745
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7333803745975352e-05,
+      "loss": 0.5121,
+      "step": 746
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7213115140513686e-05,
+      "loss": 0.6342,
+      "step": 747
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7092760737275625e-05,
+      "loss": 0.5379,
+      "step": 748
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.6972741763053835e-05,
+      "loss": 0.506,
+      "step": 749
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.685305944122187e-05,
+      "loss": 0.5759,
+      "step": 750
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.673371499172174e-05,
+      "loss": 0.5398,
+      "step": 751
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.6614709631051462e-05,
+      "loss": 0.5486,
+      "step": 752
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.649604457225266e-05,
+      "loss": 0.518,
+      "step": 753
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.6377721024898213e-05,
+      "loss": 0.5199,
+      "step": 754
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.6259740195079903e-05,
+      "loss": 0.567,
+      "step": 755
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.6142103285396148e-05,
+      "loss": 0.5024,
+      "step": 756
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.6024811494939724e-05,
+      "loss": 0.5851,
+      "step": 757
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.5907866019285545e-05,
+      "loss": 0.5437,
+      "step": 758
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.5791268050478486e-05,
+      "loss": 0.5952,
+      "step": 759
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.5675018777021223e-05,
+      "loss": 0.5774,
+      "step": 760
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.555911938386213e-05,
+      "loss": 0.5463,
+      "step": 761
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.5443571052383177e-05,
+      "loss": 0.6335,
+      "step": 762
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.532837496038792e-05,
+      "loss": 0.5745,
+      "step": 763
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.5213532282089466e-05,
+      "loss": 0.5821,
+      "step": 764
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.509904418809852e-05,
+      "loss": 0.5844,
+      "step": 765
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.4984911845411453e-05,
+      "loss": 0.5667,
+      "step": 766
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4871136417398406e-05,
+      "loss": 0.5334,
+      "step": 767
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.475771906379142e-05,
+      "loss": 0.545,
+      "step": 768
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 0.5777,
+      "step": 769
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4531963200462468e-05,
+      "loss": 0.5528,
+      "step": 770
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4419626991907925e-05,
+      "loss": 0.624,
+      "step": 771
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4307653460070846e-05,
+      "loss": 0.6083,
+      "step": 772
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4196043746316251e-05,
+      "loss": 0.6201,
+      "step": 773
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4084798988300684e-05,
+      "loss": 0.5476,
+      "step": 774
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.3973920319960655e-05,
+      "loss": 0.5868,
+      "step": 775
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.3863408871501044e-05,
+      "loss": 0.5568,
+      "step": 776
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.3753265769383606e-05,
+      "loss": 0.5279,
+      "step": 777
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.3643492136315483e-05,
+      "loss": 0.5726,
+      "step": 778
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.3534089091237755e-05,
+      "loss": 0.5671,
+      "step": 779
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.3425057749314024e-05,
+      "loss": 0.5591,
+      "step": 780
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.3316399221919074e-05,
+      "loss": 0.5713,
+      "step": 781
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.3208114616627526e-05,
+      "loss": 0.554,
+      "step": 782
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.310020503720254e-05,
+      "loss": 0.5506,
+      "step": 783
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.2992671583584587e-05,
+      "loss": 0.5398,
+      "step": 784
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.2885515351880217e-05,
+      "loss": 0.574,
+      "step": 785
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.2778737434350895e-05,
+      "loss": 0.5774,
+      "step": 786
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2672338919401866e-05,
+      "loss": 0.5914,
+      "step": 787
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2566320891571064e-05,
+      "loss": 0.6046,
+      "step": 788
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2460684431518055e-05,
+      "loss": 0.5605,
+      "step": 789
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.235543061601301e-05,
+      "loss": 0.5634,
+      "step": 790
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2250560517925746e-05,
+      "loss": 0.5992,
+      "step": 791
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2146075206214808e-05,
+      "loss": 0.5418,
+      "step": 792
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2041975745916472e-05,
+      "loss": 0.5505,
+      "step": 793
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.1938263198134087e-05,
+      "loss": 0.5557,
+      "step": 794
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.183493862002702e-05,
+      "loss": 0.5453,
+      "step": 795
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.1732003064800045e-05,
+      "loss": 0.5731,
+      "step": 796
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1629457581692616e-05,
+      "loss": 0.5961,
+      "step": 797
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1527303215968016e-05,
+      "loss": 0.5717,
+      "step": 798
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1425541008902851e-05,
+      "loss": 0.5704,
+      "step": 799
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1324171997776428e-05,
+      "loss": 0.6271,
+      "step": 800
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1223197215860065e-05,
+      "loss": 0.5258,
+      "step": 801
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1122617692406667e-05,
+      "loss": 0.6027,
+      "step": 802
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.102243445264025e-05,
+      "loss": 0.5784,
+      "step": 803
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.092264851774536e-05,
+      "loss": 0.5464,
+      "step": 804
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.082326090485679e-05,
+      "loss": 0.5768,
+      "step": 805
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.0724272627049214e-05,
+      "loss": 0.5478,
+      "step": 806
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.0625684693326727e-05,
+      "loss": 0.5595,
+      "step": 807
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.0527498108612694e-05,
+      "loss": 0.5885,
+      "step": 808
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.0429713873739505e-05,
+      "loss": 0.5256,
+      "step": 809
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.0332332985438248e-05,
+      "loss": 0.5896,
+      "step": 810
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.0235356436328675e-05,
+      "loss": 0.584,
+      "step": 811
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.0138785214909063e-05,
+      "loss": 0.5153,
+      "step": 812
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.0042620305546069e-05,
+      "loss": 0.5431,
+      "step": 813
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.946862688464753e-06,
+      "loss": 0.5634,
+      "step": 814
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.851513339738628e-06,
+      "loss": 0.5113,
+      "step": 815
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.756573231279582e-06,
+      "loss": 0.5775,
+      "step": 816
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.662043330828085e-06,
+      "loss": 0.4822,
+      "step": 817
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.567924601943323e-06,
+      "loss": 0.587,
+      "step": 818
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.474218003993273e-06,
+      "loss": 0.5371,
+      "step": 819
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.380924492145032e-06,
+      "loss": 0.5646,
+      "step": 820
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.288045017355036e-06,
+      "loss": 0.5559,
+      "step": 821
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.195580526359371e-06,
+      "loss": 0.5451,
+      "step": 822
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.103531961664118e-06,
+      "loss": 0.5505,
+      "step": 823
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.011900261535765e-06,
+      "loss": 0.54,
+      "step": 824
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 8.920686359991614e-06,
+      "loss": 0.5686,
+      "step": 825
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 8.829891186790279e-06,
+      "loss": 0.5763,
+      "step": 826
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 8.73951566742221e-06,
+      "loss": 0.5461,
+      "step": 827
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.649560723100258e-06,
+      "loss": 0.5585,
+      "step": 828
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.560027270750277e-06,
+      "loss": 0.5984,
+      "step": 829
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.470916223001779e-06,
+      "loss": 0.5397,
+      "step": 830
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.38222848817864e-06,
+      "loss": 0.5468,
+      "step": 831
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.293964970289841e-06,
+      "loss": 0.5758,
+      "step": 832
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.206126569020234e-06,
+      "loss": 0.5751,
+      "step": 833
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.118714179721403e-06,
+      "loss": 0.5921,
+      "step": 834
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 8.031728693402502e-06,
+      "loss": 0.5655,
+      "step": 835
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 7.945170996721212e-06,
+      "loss": 0.6005,
+      "step": 836
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 7.859041971974668e-06,
+      "loss": 0.585,
+      "step": 837
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.773342497090486e-06,
+      "loss": 0.5836,
+      "step": 838
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.688073445617799e-06,
+      "loss": 0.5005,
+      "step": 839
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.603235686718374e-06,
+      "loss": 0.5537,
+      "step": 840
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.518830085157735e-06,
+      "loss": 0.571,
+      "step": 841
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.434857501296355e-06,
+      "loss": 0.5404,
+      "step": 842
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.35131879108088e-06,
+      "loss": 0.5323,
+      "step": 843
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.268214806035423e-06,
+      "loss": 0.6176,
+      "step": 844
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.185546393252834e-06,
+      "loss": 0.556,
+      "step": 845
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.103314395386157e-06,
+      "loss": 0.5161,
+      "step": 846
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.0215196506399515e-06,
+      "loss": 0.514,
+      "step": 847
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.940162992761762e-06,
+      "loss": 0.5945,
+      "step": 848
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.859245251033697e-06,
+      "loss": 0.5571,
+      "step": 849
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.7787672502638844e-06,
+      "loss": 0.5089,
+      "step": 850
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 0.5204,
+      "step": 851
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.619133748411344e-06,
+      "loss": 0.5581,
+      "step": 852
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.539979874499747e-06,
+      "loss": 0.5571,
+      "step": 853
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.461268995871966e-06,
+      "loss": 0.5825,
+      "step": 854
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.3830019148412525e-06,
+      "loss": 0.5894,
+      "step": 855
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.3051794291970944e-06,
+      "loss": 0.577,
+      "step": 856
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 6.227802332197124e-06,
+      "loss": 0.545,
+      "step": 857
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 6.150871412559134e-06,
+      "loss": 0.5687,
+      "step": 858
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 6.07438745445289e-06,
+      "loss": 0.5773,
+      "step": 859
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.9983512374922e-06,
+      "loss": 0.5853,
+      "step": 860
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.922763536727022e-06,
+      "loss": 0.5539,
+      "step": 861
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.8476251226354805e-06,
+      "loss": 0.5794,
+      "step": 862
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.772936761116027e-06,
+      "loss": 0.6266,
+      "step": 863
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.6986992134796965e-06,
+      "loss": 0.5659,
+      "step": 864
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.624913236442286e-06,
+      "loss": 0.5533,
+      "step": 865
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.551579582116612e-06,
+      "loss": 0.5576,
+      "step": 866
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 5.478698998004967e-06,
+      "loss": 0.5997,
+      "step": 867
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 5.406272226991383e-06,
+      "loss": 0.5555,
+      "step": 868
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 5.334300007334064e-06,
+      "loss": 0.5989,
+      "step": 869
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 5.262783072657984e-06,
+      "loss": 0.5888,
+      "step": 870
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 5.191722151947226e-06,
+      "loss": 0.5595,
+      "step": 871
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 5.1211179695377035e-06,
+      "loss": 0.5513,
+      "step": 872
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 5.050971245109709e-06,
+      "loss": 0.6141,
+      "step": 873
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.981282693680583e-06,
+      "loss": 0.5219,
+      "step": 874
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.912053025597429e-06,
+      "loss": 0.506,
+      "step": 875
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.8432829465298865e-06,
+      "loss": 0.5505,
+      "step": 876
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.7749731574629196e-06,
+      "loss": 0.5144,
+      "step": 877
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.707124354689679e-06,
+      "loss": 0.5986,
+      "step": 878
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.639737229804403e-06,
+      "loss": 0.5857,
+      "step": 879
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.572812469695381e-06,
+      "loss": 0.5612,
+      "step": 880
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.5063507565379195e-06,
+      "loss": 0.5714,
+      "step": 881
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.440352767787437e-06,
+      "loss": 0.6141,
+      "step": 882
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.374819176172501e-06,
+      "loss": 0.5798,
+      "step": 883
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.309750649688033e-06,
+      "loss": 0.5606,
+      "step": 884
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.245147851588438e-06,
+      "loss": 0.6166,
+      "step": 885
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.181011440380889e-06,
+      "loss": 0.6443,
+      "step": 886
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.117342069818603e-06,
+      "loss": 0.5473,
+      "step": 887
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.054140388894162e-06,
+      "loss": 0.5993,
+      "step": 888
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.991407041832912e-06,
+      "loss": 0.5317,
+      "step": 889
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.9291426680863995e-06,
+      "loss": 0.5084,
+      "step": 890
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.8673479023258465e-06,
+      "loss": 0.5799,
+      "step": 891
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 0.6053,
+      "step": 892
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.7451697095070736e-06,
+      "loss": 0.5232,
+      "step": 893
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.6847875278317068e-06,
+      "loss": 0.578,
+      "step": 894
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.6248774448952695e-06,
+      "loss": 0.5703,
+      "step": 895
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.5654400713713066e-06,
+      "loss": 0.5265,
+      "step": 896
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.5064760131149453e-06,
+      "loss": 0.5291,
+      "step": 897
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.4479858711567337e-06,
+      "loss": 0.5323,
+      "step": 898
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.389970241696516e-06,
+      "loss": 0.5664,
+      "step": 899
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.3324297160973405e-06,
+      "loss": 0.6099,
+      "step": 900
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.2753648808794503e-06,
+      "loss": 0.5493,
+      "step": 901
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.2187763177143003e-06,
+      "loss": 0.5663,
+      "step": 902
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.162664603418608e-06,
+      "loss": 0.5925,
+      "step": 903
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.1070303099485056e-06,
+      "loss": 0.5373,
+      "step": 904
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.05187400439369e-06,
+      "loss": 0.5661,
+      "step": 905
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9971962489716508e-06,
+      "loss": 0.525,
+      "step": 906
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.942997601021924e-06,
+      "loss": 0.5767,
+      "step": 907
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.889278613000429e-06,
+      "loss": 0.5366,
+      "step": 908
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.8360398324738414e-06,
+      "loss": 0.5325,
+      "step": 909
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.783281802113985e-06,
+      "loss": 0.5241,
+      "step": 910
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.731005059692332e-06,
+      "loss": 0.5382,
+      "step": 911
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.6792101380744837e-06,
+      "loss": 0.5729,
+      "step": 912
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.6278975652147875e-06,
+      "loss": 0.5881,
+      "step": 913
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.577067864150906e-06,
+      "loss": 0.5548,
+      "step": 914
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.5267215529985342e-06,
+      "loss": 0.5335,
+      "step": 915
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.476859144946053e-06,
+      "loss": 0.5598,
+      "step": 916
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.4274811482493833e-06,
+      "loss": 0.5954,
+      "step": 917
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.3785880662267224e-06,
+      "loss": 0.5871,
+      "step": 918
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.330180397253473e-06,
+      "loss": 0.5587,
+      "step": 919
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.282258634757145e-06,
+      "loss": 0.6146,
+      "step": 920
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.2348232672122936e-06,
+      "loss": 0.5853,
+      "step": 921
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.1878747781355846e-06,
+      "loss": 0.5564,
+      "step": 922
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.1414136460808807e-06,
+      "loss": 0.5798,
+      "step": 923
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.0954403446342753e-06,
+      "loss": 0.5305,
+      "step": 924
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.049955342409349e-06,
+      "loss": 0.5847,
+      "step": 925
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.0049591030423977e-06,
+      "loss": 0.6053,
+      "step": 926
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.9604520851876198e-06,
+      "loss": 0.5832,
+      "step": 927
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.916434742512524e-06,
+      "loss": 0.5101,
+      "step": 928
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.8729075236932903e-06,
+      "loss": 0.5337,
+      "step": 929
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.829870872410161e-06,
+      "loss": 0.5416,
+      "step": 930
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.7873252273429509e-06,
+      "loss": 0.5341,
+      "step": 931
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.7452710221665724e-06,
+      "loss": 0.5215,
+      "step": 932
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.70370868554659e-06,
+      "loss": 0.5805,
+      "step": 933
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6626386411348782e-06,
+      "loss": 0.6582,
+      "step": 934
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6220613075653202e-06,
+      "loss": 0.5899,
+      "step": 935
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.5819770984494719e-06,
+      "loss": 0.5234,
+      "step": 936
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.542386422372405e-06,
+      "loss": 0.582,
+      "step": 937
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.5032896828885479e-06,
+      "loss": 0.5365,
+      "step": 938
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.464687278517518e-06,
+      "loss": 0.5538,
+      "step": 939
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.426579602740108e-06,
+      "loss": 0.6153,
+      "step": 940
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.3889670439942659e-06,
+      "loss": 0.5332,
+      "step": 941
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.3518499856711108e-06,
+      "loss": 0.5557,
+      "step": 942
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.3152288061110518e-06,
+      "loss": 0.5163,
+      "step": 943
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.2791038785999243e-06,
+      "loss": 0.5944,
+      "step": 944
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.243475571365177e-06,
+      "loss": 0.5656,
+      "step": 945
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.2083442475721352e-06,
+      "loss": 0.5602,
+      "step": 946
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.1737102653202826e-06,
+      "loss": 0.5633,
+      "step": 947
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.139573977639613e-06,
+      "loss": 0.5478,
+      "step": 948
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.1059357324870455e-06,
+      "loss": 0.5181,
+      "step": 949
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.072795872742849e-06,
+      "loss": 0.5564,
+      "step": 950
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.040154736207194e-06,
+      "loss": 0.5676,
+      "step": 951
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.0080126555966574e-06,
+      "loss": 0.5634,
+      "step": 952
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.763699585408737e-07,
+      "loss": 0.5172,
+      "step": 953
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.452269675791603e-07,
+      "loss": 0.5594,
+      "step": 954
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.145840001572537e-07,
+      "loss": 0.527,
+      "step": 955
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 8.844413686240727e-07,
+      "loss": 0.5014,
+      "step": 956
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 8.547993802285215e-07,
+      "loss": 0.5435,
+      "step": 957
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 8.256583371163585e-07,
+      "loss": 0.5452,
+      "step": 958
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 7.970185363271431e-07,
+      "loss": 0.6109,
+      "step": 959
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 7.688802697911667e-07,
+      "loss": 0.5099,
+      "step": 960
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 7.41243824326504e-07,
+      "loss": 0.5195,
+      "step": 961
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 7.141094816360882e-07,
+      "loss": 0.5538,
+      "step": 962
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 6.874775183048299e-07,
+      "loss": 0.5455,
+      "step": 963
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 6.613482057968024e-07,
+      "loss": 0.5602,
+      "step": 964
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 6.357218104524831e-07,
+      "loss": 0.5121,
+      "step": 965
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 6.10598593486017e-07,
+      "loss": 0.4987,
+      "step": 966
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 5.859788109825793e-07,
+      "loss": 0.5466,
+      "step": 967
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 5.618627138957333e-07,
+      "loss": 0.5478,
+      "step": 968
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 5.382505480449273e-07,
+      "loss": 0.5153,
+      "step": 969
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 5.151425541129184e-07,
+      "loss": 0.5882,
+      "step": 970
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.925389676433745e-07,
+      "loss": 0.5359,
+      "step": 971
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.7044001903844835e-07,
+      "loss": 0.5362,
+      "step": 972
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.48845933556441e-07,
+      "loss": 0.5693,
+      "step": 973
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.277569313094809e-07,
+      "loss": 0.5227,
+      "step": 974
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.071732272613149e-07,
+      "loss": 0.5506,
+      "step": 975
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.8709503122509873e-07,
+      "loss": 0.5338,
+      "step": 976
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.6752254786124317e-07,
+      "loss": 0.5324,
+      "step": 977
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.4845597667537145e-07,
+      "loss": 0.5885,
+      "step": 978
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.2989551201624835e-07,
+      "loss": 0.5455,
+      "step": 979
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.118413430737932e-07,
+      "loss": 0.6255,
+      "step": 980
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.9429365387719234e-07,
+      "loss": 0.5458,
+      "step": 981
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.772526232929895e-07,
+      "loss": 0.5417,
+      "step": 982
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.6071842502326527e-07,
+      "loss": 0.573,
+      "step": 983
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.446912276038826e-07,
+      "loss": 0.524,
+      "step": 984
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.2917119440275524e-07,
+      "loss": 0.5645,
+      "step": 985
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.1415848361818758e-07,
+      "loss": 0.5242,
+      "step": 986
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.996532482772595e-07,
+      "loss": 0.5226,
+      "step": 987
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8565563623428318e-07,
+      "loss": 0.6118,
+      "step": 988
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.7216579016925416e-07,
+      "loss": 0.5773,
+      "step": 989
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5918384758644155e-07,
+      "loss": 0.5582,
+      "step": 990
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.4670994081297795e-07,
+      "loss": 0.5882,
+      "step": 991
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.347441969974772e-07,
+      "loss": 0.5924,
+      "step": 992
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.2328673810877988e-07,
+      "loss": 0.5463,
+      "step": 993
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.1233768093468766e-07,
+      "loss": 0.6127,
+      "step": 994
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.0189713708078085e-07,
+      "loss": 0.5762,
+      "step": 995
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.196521296927496e-08,
+      "loss": 0.5832,
+      "step": 996
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 8.254200983794369e-08,
+      "loss": 0.584,
+      "step": 997
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 7.36276237390754e-08,
+      "loss": 0.5341,
+      "step": 998
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 6.522214553850159e-08,
+      "loss": 0.5559,
+      "step": 999
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 5.732566091468106e-08,
+      "loss": 0.4961,
+      "step": 1000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.993825035780608e-08,
+      "loss": 0.5108,
+      "step": 1001
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.305998916898646e-08,
+      "loss": 0.5416,
+      "step": 1002
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 3.669094745950008e-08,
+      "loss": 0.5816,
+      "step": 1003
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 3.083119015005465e-08,
+      "loss": 0.5774,
+      "step": 1004
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.5480776970143726e-08,
+      "loss": 0.6052,
+      "step": 1005
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.06397624574306e-08,
+      "loss": 0.5694,
+      "step": 1006
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.6308195957182027e-08,
+      "loss": 0.5383,
+      "step": 1007
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.2486121621785308e-08,
+      "loss": 0.569,
+      "step": 1008
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.17357841028199e-09,
+      "loss": 0.5661,
+      "step": 1009
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.3706000879903975e-09,
+      "loss": 0.5828,
+      "step": 1010
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.07721522613369e-09,
+      "loss": 0.5406,
+      "step": 1011
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2934472015734287e-09,
+      "loss": 0.538,
+      "step": 1012
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0193141965486596e-09,
+      "loss": 0.5358,
+      "step": 1013
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.548291985149387e-10,
+      "loss": 0.57,
+      "step": 1014
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0,
+      "loss": 0.5424,
+      "step": 1015
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.5613698959350586,
+      "eval_runtime": 4279.4254,
+      "eval_samples_per_second": 2.846,
+      "eval_steps_per_second": 0.119,
+      "step": 1015
+    },
+    {
+      "epoch": 1.0,
+      "step": 1015,
+      "total_flos": 3.6073842611805225e+18,
+      "train_loss": 0.5994285663360446,
+      "train_runtime": 58656.2372,
+      "train_samples_per_second": 0.83,
+      "train_steps_per_second": 0.017
+    }
+  ],
+  "max_steps": 1015,
+  "num_train_epochs": 1,
+  "total_flos": 3.6073842611805225e+18,
+  "trial_name": null,
+  "trial_params": null
+}