diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,5173 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.9976717112922002,
+  "global_step": 858,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0,
+      "loss": 0.7422,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.254921071067263e-06,
+      "loss": 0.811,
+      "step": 2
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.743890341169911e-06,
+      "loss": 0.7936,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.509842142134527e-06,
+      "loss": 0.7831,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.879620776439306e-06,
+      "loss": 0.7914,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.0998811412237173e-05,
+      "loss": 0.7931,
+      "step": 6
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1945073611827293e-05,
+      "loss": 0.7507,
+      "step": 7
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.276476321320179e-05,
+      "loss": 0.7546,
+      "step": 8
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3487780682339822e-05,
+      "loss": 0.7963,
+      "step": 9
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4134541847506571e-05,
+      "loss": 0.8279,
+      "step": 10
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.4719608488056167e-05,
+      "loss": 0.8287,
+      "step": 11
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.5253732483304438e-05,
+      "loss": 0.7954,
+      "step": 12
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.574507892893274e-05,
+      "loss": 0.7881,
+      "step": 13
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.6199994682894558e-05,
+      "loss": 0.7122,
+      "step": 14
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.6623511117609218e-05,
+      "loss": 0.7996,
+      "step": 15
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.7019684284269053e-05,
+      "loss": 0.7628,
+      "step": 16
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.7391831770440535e-05,
+      "loss": 0.7771,
+      "step": 17
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.7742701753407083e-05,
+      "loss": 0.7875,
+      "step": 18
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.8074596285317475e-05,
+      "loss": 0.7663,
+      "step": 19
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.838946291857383e-05,
+      "loss": 0.7906,
+      "step": 20
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.8688963952997205e-05,
+      "loss": 0.7768,
+      "step": 21
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.8974529559123432e-05,
+      "loss": 0.7778,
+      "step": 22
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.924739908310523e-05,
+      "loss": 0.7449,
+      "step": 23
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.95086535543717e-05,
+      "loss": 0.7659,
+      "step": 24
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9759241552878612e-05,
+      "loss": 0.8142,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2e-05,
+      "loss": 0.7864,
+      "step": 26
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2e-05,
+      "loss": 0.7993,
+      "step": 27
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9975961538461538e-05,
+      "loss": 0.7083,
+      "step": 28
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9951923076923078e-05,
+      "loss": 0.7924,
+      "step": 29
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9927884615384617e-05,
+      "loss": 0.7368,
+      "step": 30
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9903846153846157e-05,
+      "loss": 0.7919,
+      "step": 31
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9879807692307693e-05,
+      "loss": 0.7637,
+      "step": 32
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9855769230769233e-05,
+      "loss": 0.7758,
+      "step": 33
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.983173076923077e-05,
+      "loss": 0.7711,
+      "step": 34
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.980769230769231e-05,
+      "loss": 0.7734,
+      "step": 35
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.978365384615385e-05,
+      "loss": 0.7318,
+      "step": 36
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.975961538461539e-05,
+      "loss": 0.7486,
+      "step": 37
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9735576923076925e-05,
+      "loss": 0.743,
+      "step": 38
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9711538461538465e-05,
+      "loss": 0.7031,
+      "step": 39
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.96875e-05,
+      "loss": 0.7297,
+      "step": 40
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.966346153846154e-05,
+      "loss": 0.7214,
+      "step": 41
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.963942307692308e-05,
+      "loss": 0.6865,
+      "step": 42
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9615384615384617e-05,
+      "loss": 0.7313,
+      "step": 43
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9591346153846156e-05,
+      "loss": 0.6994,
+      "step": 44
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9567307692307693e-05,
+      "loss": 0.6914,
+      "step": 45
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9543269230769232e-05,
+      "loss": 0.6956,
+      "step": 46
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9519230769230772e-05,
+      "loss": 0.6747,
+      "step": 47
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.949519230769231e-05,
+      "loss": 0.6916,
+      "step": 48
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9471153846153848e-05,
+      "loss": 0.7272,
+      "step": 49
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9447115384615388e-05,
+      "loss": 0.7077,
+      "step": 50
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9423076923076924e-05,
+      "loss": 0.679,
+      "step": 51
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9399038461538464e-05,
+      "loss": 0.6818,
+      "step": 52
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9375e-05,
+      "loss": 0.6959,
+      "step": 53
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.935096153846154e-05,
+      "loss": 0.7511,
+      "step": 54
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.932692307692308e-05,
+      "loss": 0.6726,
+      "step": 55
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9302884615384616e-05,
+      "loss": 0.6754,
+      "step": 56
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9278846153846155e-05,
+      "loss": 0.6715,
+      "step": 57
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9254807692307692e-05,
+      "loss": 0.6854,
+      "step": 58
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": 0.7328,
+      "step": 59
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.920673076923077e-05,
+      "loss": 0.6677,
+      "step": 60
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.918269230769231e-05,
+      "loss": 0.6678,
+      "step": 61
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9158653846153847e-05,
+      "loss": 0.6942,
+      "step": 62
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9134615384615387e-05,
+      "loss": 0.6735,
+      "step": 63
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9110576923076923e-05,
+      "loss": 0.6584,
+      "step": 64
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9086538461538463e-05,
+      "loss": 0.6717,
+      "step": 65
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9062500000000003e-05,
+      "loss": 0.6701,
+      "step": 66
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.903846153846154e-05,
+      "loss": 0.6356,
+      "step": 67
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.901442307692308e-05,
+      "loss": 0.66,
+      "step": 68
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8990384615384615e-05,
+      "loss": 0.6898,
+      "step": 69
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8966346153846155e-05,
+      "loss": 0.6713,
+      "step": 70
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.8942307692307694e-05,
+      "loss": 0.6409,
+      "step": 71
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.8918269230769234e-05,
+      "loss": 0.6343,
+      "step": 72
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.889423076923077e-05,
+      "loss": 0.6791,
+      "step": 73
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.887019230769231e-05,
+      "loss": 0.641,
+      "step": 74
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.8846153846153846e-05,
+      "loss": 0.6472,
+      "step": 75
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.8822115384615386e-05,
+      "loss": 0.6777,
+      "step": 76
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.8798076923076926e-05,
+      "loss": 0.6636,
+      "step": 77
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.8774038461538465e-05,
+      "loss": 0.6776,
+      "step": 78
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.6891,
+      "step": 79
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.8725961538461538e-05,
+      "loss": 0.6069,
+      "step": 80
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.8701923076923078e-05,
+      "loss": 0.6426,
+      "step": 81
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.8677884615384617e-05,
+      "loss": 0.6459,
+      "step": 82
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.8653846153846157e-05,
+      "loss": 0.626,
+      "step": 83
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8629807692307693e-05,
+      "loss": 0.6017,
+      "step": 84
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8605769230769233e-05,
+      "loss": 0.6666,
+      "step": 85
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.858173076923077e-05,
+      "loss": 0.6367,
+      "step": 86
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.855769230769231e-05,
+      "loss": 0.6391,
+      "step": 87
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.853365384615385e-05,
+      "loss": 0.6155,
+      "step": 88
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.850961538461539e-05,
+      "loss": 0.663,
+      "step": 89
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.8485576923076925e-05,
+      "loss": 0.59,
+      "step": 90
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.8461538461538465e-05,
+      "loss": 0.674,
+      "step": 91
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.84375e-05,
+      "loss": 0.6134,
+      "step": 92
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.841346153846154e-05,
+      "loss": 0.6526,
+      "step": 93
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.8389423076923077e-05,
+      "loss": 0.6172,
+      "step": 94
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.8365384615384617e-05,
+      "loss": 0.6317,
+      "step": 95
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.8341346153846156e-05,
+      "loss": 0.6154,
+      "step": 96
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.8317307692307693e-05,
+      "loss": 0.6432,
+      "step": 97
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.8293269230769232e-05,
+      "loss": 0.6154,
+      "step": 98
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.826923076923077e-05,
+      "loss": 0.6252,
+      "step": 99
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.824519230769231e-05,
+      "loss": 0.6276,
+      "step": 100
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.8221153846153848e-05,
+      "loss": 0.6388,
+      "step": 101
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.8197115384615388e-05,
+      "loss": 0.5981,
+      "step": 102
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.8173076923076924e-05,
+      "loss": 0.6396,
+      "step": 103
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.8149038461538464e-05,
+      "loss": 0.6027,
+      "step": 104
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.8125e-05,
+      "loss": 0.5946,
+      "step": 105
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.810096153846154e-05,
+      "loss": 0.5966,
+      "step": 106
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.807692307692308e-05,
+      "loss": 0.6364,
+      "step": 107
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.8052884615384616e-05,
+      "loss": 0.5778,
+      "step": 108
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.8028846153846156e-05,
+      "loss": 0.6138,
+      "step": 109
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.8004807692307692e-05,
+      "loss": 0.6226,
+      "step": 110
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.798076923076923e-05,
+      "loss": 0.5759,
+      "step": 111
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.795673076923077e-05,
+      "loss": 0.6231,
+      "step": 112
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.793269230769231e-05,
+      "loss": 0.5883,
+      "step": 113
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.7908653846153847e-05,
+      "loss": 0.6646,
+      "step": 114
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.7884615384615387e-05,
+      "loss": 0.6122,
+      "step": 115
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.7860576923076923e-05,
+      "loss": 0.6049,
+      "step": 116
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.7836538461538463e-05,
+      "loss": 0.5839,
+      "step": 117
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.7812500000000003e-05,
+      "loss": 0.6089,
+      "step": 118
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.778846153846154e-05,
+      "loss": 0.6326,
+      "step": 119
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.776442307692308e-05,
+      "loss": 0.6625,
+      "step": 120
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.7740384615384615e-05,
+      "loss": 0.585,
+      "step": 121
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.7716346153846155e-05,
+      "loss": 0.6062,
+      "step": 122
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.7692307692307694e-05,
+      "loss": 0.63,
+      "step": 123
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.7668269230769234e-05,
+      "loss": 0.6506,
+      "step": 124
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.764423076923077e-05,
+      "loss": 0.5931,
+      "step": 125
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.762019230769231e-05,
+      "loss": 0.6179,
+      "step": 126
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.7596153846153846e-05,
+      "loss": 0.6346,
+      "step": 127
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.7572115384615386e-05,
+      "loss": 0.6125,
+      "step": 128
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.7548076923076926e-05,
+      "loss": 0.5992,
+      "step": 129
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.7524038461538466e-05,
+      "loss": 0.5912,
+      "step": 130
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 0.6053,
+      "step": 131
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.7475961538461538e-05,
+      "loss": 0.6335,
+      "step": 132
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.7451923076923078e-05,
+      "loss": 0.6382,
+      "step": 133
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.7427884615384618e-05,
+      "loss": 0.6007,
+      "step": 134
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.7403846153846157e-05,
+      "loss": 0.6569,
+      "step": 135
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.7379807692307694e-05,
+      "loss": 0.6395,
+      "step": 136
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.7355769230769233e-05,
+      "loss": 0.594,
+      "step": 137
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.733173076923077e-05,
+      "loss": 0.6044,
+      "step": 138
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.730769230769231e-05,
+      "loss": 0.6195,
+      "step": 139
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7283653846153846e-05,
+      "loss": 0.6055,
+      "step": 140
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7259615384615385e-05,
+      "loss": 0.656,
+      "step": 141
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7235576923076925e-05,
+      "loss": 0.6211,
+      "step": 142
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7211538461538465e-05,
+      "loss": 0.6229,
+      "step": 143
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.71875e-05,
+      "loss": 0.6284,
+      "step": 144
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.7163461538461537e-05,
+      "loss": 0.6261,
+      "step": 145
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.7139423076923077e-05,
+      "loss": 0.5551,
+      "step": 146
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.7115384615384617e-05,
+      "loss": 0.679,
+      "step": 147
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.7091346153846156e-05,
+      "loss": 0.5737,
+      "step": 148
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.7067307692307693e-05,
+      "loss": 0.5966,
+      "step": 149
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.7043269230769232e-05,
+      "loss": 0.6058,
+      "step": 150
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.701923076923077e-05,
+      "loss": 0.5706,
+      "step": 151
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.699519230769231e-05,
+      "loss": 0.5843,
+      "step": 152
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.6971153846153848e-05,
+      "loss": 0.6428,
+      "step": 153
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.6947115384615388e-05,
+      "loss": 0.6675,
+      "step": 154
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.6923076923076924e-05,
+      "loss": 0.6131,
+      "step": 155
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.6899038461538464e-05,
+      "loss": 0.6633,
+      "step": 156
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.6875e-05,
+      "loss": 0.5869,
+      "step": 157
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.685096153846154e-05,
+      "loss": 0.5723,
+      "step": 158
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.682692307692308e-05,
+      "loss": 0.6094,
+      "step": 159
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.6802884615384616e-05,
+      "loss": 0.6415,
+      "step": 160
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.6778846153846156e-05,
+      "loss": 0.6097,
+      "step": 161
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.6754807692307692e-05,
+      "loss": 0.5758,
+      "step": 162
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.673076923076923e-05,
+      "loss": 0.6197,
+      "step": 163
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.670673076923077e-05,
+      "loss": 0.6409,
+      "step": 164
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.668269230769231e-05,
+      "loss": 0.595,
+      "step": 165
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.6658653846153847e-05,
+      "loss": 0.5801,
+      "step": 166
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.6634615384615387e-05,
+      "loss": 0.5958,
+      "step": 167
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.6610576923076923e-05,
+      "loss": 0.5762,
+      "step": 168
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.6586538461538463e-05,
+      "loss": 0.5679,
+      "step": 169
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.6562500000000003e-05,
+      "loss": 0.5873,
+      "step": 170
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.653846153846154e-05,
+      "loss": 0.6121,
+      "step": 171
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.651442307692308e-05,
+      "loss": 0.6226,
+      "step": 172
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.6490384615384615e-05,
+      "loss": 0.5936,
+      "step": 173
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.6466346153846155e-05,
+      "loss": 0.5506,
+      "step": 174
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.6442307692307695e-05,
+      "loss": 0.5895,
+      "step": 175
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.6418269230769234e-05,
+      "loss": 0.5636,
+      "step": 176
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.639423076923077e-05,
+      "loss": 0.644,
+      "step": 177
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.637019230769231e-05,
+      "loss": 0.5985,
+      "step": 178
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.6346153846153847e-05,
+      "loss": 0.6207,
+      "step": 179
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.6322115384615386e-05,
+      "loss": 0.6017,
+      "step": 180
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.6298076923076923e-05,
+      "loss": 0.5789,
+      "step": 181
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.6274038461538462e-05,
+      "loss": 0.6289,
+      "step": 182
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.6250000000000002e-05,
+      "loss": 0.5999,
+      "step": 183
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.6225961538461538e-05,
+      "loss": 0.5892,
+      "step": 184
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.6201923076923078e-05,
+      "loss": 0.5971,
+      "step": 185
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.6177884615384614e-05,
+      "loss": 0.6273,
+      "step": 186
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.6153846153846154e-05,
+      "loss": 0.6107,
+      "step": 187
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.6129807692307694e-05,
+      "loss": 0.5817,
+      "step": 188
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.6105769230769233e-05,
+      "loss": 0.5933,
+      "step": 189
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.608173076923077e-05,
+      "loss": 0.6149,
+      "step": 190
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.605769230769231e-05,
+      "loss": 0.6081,
+      "step": 191
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.6033653846153846e-05,
+      "loss": 0.5656,
+      "step": 192
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.6009615384615385e-05,
+      "loss": 0.6105,
+      "step": 193
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.5985576923076925e-05,
+      "loss": 0.5958,
+      "step": 194
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.5961538461538465e-05,
+      "loss": 0.593,
+      "step": 195
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.59375e-05,
+      "loss": 0.5812,
+      "step": 196
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.5913461538461537e-05,
+      "loss": 0.6058,
+      "step": 197
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.5889423076923077e-05,
+      "loss": 0.5699,
+      "step": 198
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.5865384615384617e-05,
+      "loss": 0.6198,
+      "step": 199
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.5841346153846157e-05,
+      "loss": 0.5797,
+      "step": 200
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.5817307692307693e-05,
+      "loss": 0.6161,
+      "step": 201
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.5793269230769233e-05,
+      "loss": 0.5843,
+      "step": 202
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.576923076923077e-05,
+      "loss": 0.6367,
+      "step": 203
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.574519230769231e-05,
+      "loss": 0.6324,
+      "step": 204
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.5721153846153848e-05,
+      "loss": 0.5936,
+      "step": 205
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.5697115384615388e-05,
+      "loss": 0.5901,
+      "step": 206
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.5673076923076924e-05,
+      "loss": 0.5988,
+      "step": 207
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.5649038461538464e-05,
+      "loss": 0.5844,
+      "step": 208
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.5625e-05,
+      "loss": 0.5556,
+      "step": 209
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.560096153846154e-05,
+      "loss": 0.5861,
+      "step": 210
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.557692307692308e-05,
+      "loss": 0.5732,
+      "step": 211
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.5552884615384616e-05,
+      "loss": 0.608,
+      "step": 212
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.5528846153846156e-05,
+      "loss": 0.5851,
+      "step": 213
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.5504807692307692e-05,
+      "loss": 0.6018,
+      "step": 214
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.5480769230769232e-05,
+      "loss": 0.6022,
+      "step": 215
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.545673076923077e-05,
+      "loss": 0.5558,
+      "step": 216
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.543269230769231e-05,
+      "loss": 0.5761,
+      "step": 217
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.5408653846153847e-05,
+      "loss": 0.5825,
+      "step": 218
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.556,
+      "step": 219
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.5360576923076923e-05,
+      "loss": 0.5722,
+      "step": 220
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.5336538461538463e-05,
+      "loss": 0.6098,
+      "step": 221
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.5312500000000003e-05,
+      "loss": 0.6149,
+      "step": 222
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.528846153846154e-05,
+      "loss": 0.5854,
+      "step": 223
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.526442307692308e-05,
+      "loss": 0.5865,
+      "step": 224
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.5240384615384615e-05,
+      "loss": 0.5743,
+      "step": 225
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.5216346153846155e-05,
+      "loss": 0.5512,
+      "step": 226
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.5192307692307693e-05,
+      "loss": 0.6032,
+      "step": 227
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.5168269230769233e-05,
+      "loss": 0.5869,
+      "step": 228
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.514423076923077e-05,
+      "loss": 0.5995,
+      "step": 229
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.512019230769231e-05,
+      "loss": 0.5888,
+      "step": 230
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.5096153846153847e-05,
+      "loss": 0.5775,
+      "step": 231
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.5072115384615386e-05,
+      "loss": 0.5972,
+      "step": 232
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.5048076923076924e-05,
+      "loss": 0.5632,
+      "step": 233
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.5024038461538464e-05,
+      "loss": 0.5654,
+      "step": 234
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.6083,
+      "step": 235
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.4975961538461538e-05,
+      "loss": 0.5793,
+      "step": 236
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.4951923076923078e-05,
+      "loss": 0.6203,
+      "step": 237
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.4927884615384616e-05,
+      "loss": 0.6068,
+      "step": 238
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.4903846153846156e-05,
+      "loss": 0.5823,
+      "step": 239
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.4879807692307692e-05,
+      "loss": 0.58,
+      "step": 240
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.4855769230769232e-05,
+      "loss": 0.5865,
+      "step": 241
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.483173076923077e-05,
+      "loss": 0.58,
+      "step": 242
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.480769230769231e-05,
+      "loss": 0.646,
+      "step": 243
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.4783653846153848e-05,
+      "loss": 0.5876,
+      "step": 244
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.4759615384615387e-05,
+      "loss": 0.5934,
+      "step": 245
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.4735576923076924e-05,
+      "loss": 0.5466,
+      "step": 246
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.4711538461538463e-05,
+      "loss": 0.6036,
+      "step": 247
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.4687500000000001e-05,
+      "loss": 0.6108,
+      "step": 248
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.466346153846154e-05,
+      "loss": 0.5597,
+      "step": 249
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.4639423076923079e-05,
+      "loss": 0.5692,
+      "step": 250
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.4615384615384615e-05,
+      "loss": 0.5468,
+      "step": 251
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.4591346153846155e-05,
+      "loss": 0.5733,
+      "step": 252
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.4567307692307693e-05,
+      "loss": 0.6086,
+      "step": 253
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.4543269230769233e-05,
+      "loss": 0.5834,
+      "step": 254
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.451923076923077e-05,
+      "loss": 0.6216,
+      "step": 255
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.449519230769231e-05,
+      "loss": 0.5997,
+      "step": 256
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.4471153846153847e-05,
+      "loss": 0.5901,
+      "step": 257
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.4447115384615386e-05,
+      "loss": 0.6099,
+      "step": 258
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.4423076923076924e-05,
+      "loss": 0.5922,
+      "step": 259
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4399038461538464e-05,
+      "loss": 0.5947,
+      "step": 260
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4375e-05,
+      "loss": 0.5753,
+      "step": 261
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4350961538461538e-05,
+      "loss": 0.5513,
+      "step": 262
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4326923076923078e-05,
+      "loss": 0.564,
+      "step": 263
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4302884615384616e-05,
+      "loss": 0.5453,
+      "step": 264
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.4278846153846156e-05,
+      "loss": 0.5563,
+      "step": 265
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.4254807692307692e-05,
+      "loss": 0.5737,
+      "step": 266
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.4230769230769232e-05,
+      "loss": 0.5407,
+      "step": 267
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.420673076923077e-05,
+      "loss": 0.5654,
+      "step": 268
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.418269230769231e-05,
+      "loss": 0.6007,
+      "step": 269
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.4158653846153848e-05,
+      "loss": 0.6246,
+      "step": 270
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.4134615384615387e-05,
+      "loss": 0.5637,
+      "step": 271
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.4110576923076924e-05,
+      "loss": 0.5661,
+      "step": 272
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.4086538461538463e-05,
+      "loss": 0.6215,
+      "step": 273
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.4062500000000001e-05,
+      "loss": 0.6113,
+      "step": 274
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.403846153846154e-05,
+      "loss": 0.5856,
+      "step": 275
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.4014423076923079e-05,
+      "loss": 0.5466,
+      "step": 276
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.3990384615384615e-05,
+      "loss": 0.6127,
+      "step": 277
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.3966346153846155e-05,
+      "loss": 0.5922,
+      "step": 278
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.3942307692307693e-05,
+      "loss": 0.5887,
+      "step": 279
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.3918269230769233e-05,
+      "loss": 0.5584,
+      "step": 280
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.3894230769230769e-05,
+      "loss": 0.5844,
+      "step": 281
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.3870192307692309e-05,
+      "loss": 0.6082,
+      "step": 282
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.3846153846153847e-05,
+      "loss": 0.586,
+      "step": 283
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.3822115384615386e-05,
+      "loss": 0.5995,
+      "step": 284
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.3798076923076924e-05,
+      "loss": 0.5915,
+      "step": 285
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.3774038461538464e-05,
+      "loss": 0.623,
+      "step": 286
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.375e-05,
+      "loss": 0.6166,
+      "step": 287
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.3725961538461538e-05,
+      "loss": 0.6105,
+      "step": 288
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.3701923076923078e-05,
+      "loss": 0.6051,
+      "step": 289
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.3677884615384616e-05,
+      "loss": 0.5828,
+      "step": 290
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.3653846153846156e-05,
+      "loss": 0.5863,
+      "step": 291
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.3629807692307692e-05,
+      "loss": 0.5673,
+      "step": 292
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.3605769230769232e-05,
+      "loss": 0.5942,
+      "step": 293
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.358173076923077e-05,
+      "loss": 0.5794,
+      "step": 294
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.355769230769231e-05,
+      "loss": 0.5921,
+      "step": 295
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.3533653846153848e-05,
+      "loss": 0.5549,
+      "step": 296
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.3509615384615387e-05,
+      "loss": 0.539,
+      "step": 297
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.3485576923076924e-05,
+      "loss": 0.6069,
+      "step": 298
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.3461538461538463e-05,
+      "loss": 0.5782,
+      "step": 299
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.3437500000000001e-05,
+      "loss": 0.5575,
+      "step": 300
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.341346153846154e-05,
+      "loss": 0.5728,
+      "step": 301
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.3389423076923077e-05,
+      "loss": 0.5773,
+      "step": 302
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.3365384615384615e-05,
+      "loss": 0.5626,
+      "step": 303
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.3341346153846155e-05,
+      "loss": 0.6014,
+      "step": 304
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.3317307692307693e-05,
+      "loss": 0.5702,
+      "step": 305
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.3293269230769233e-05,
+      "loss": 0.5823,
+      "step": 306
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.3269230769230769e-05,
+      "loss": 0.5821,
+      "step": 307
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.3245192307692309e-05,
+      "loss": 0.5438,
+      "step": 308
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.3221153846153847e-05,
+      "loss": 0.6171,
+      "step": 309
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.3197115384615387e-05,
+      "loss": 0.592,
+      "step": 310
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.3173076923076925e-05,
+      "loss": 0.5793,
+      "step": 311
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.3149038461538464e-05,
+      "loss": 0.5872,
+      "step": 312
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.3125e-05,
+      "loss": 0.6339,
+      "step": 313
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.3100961538461539e-05,
+      "loss": 0.606,
+      "step": 314
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.3076923076923078e-05,
+      "loss": 0.582,
+      "step": 315
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3052884615384616e-05,
+      "loss": 0.5718,
+      "step": 316
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3028846153846156e-05,
+      "loss": 0.5941,
+      "step": 317
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3004807692307692e-05,
+      "loss": 0.6062,
+      "step": 318
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.2980769230769232e-05,
+      "loss": 0.5908,
+      "step": 319
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.295673076923077e-05,
+      "loss": 0.5649,
+      "step": 320
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.293269230769231e-05,
+      "loss": 0.5444,
+      "step": 321
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.2908653846153848e-05,
+      "loss": 0.5807,
+      "step": 322
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.2884615384615386e-05,
+      "loss": 0.585,
+      "step": 323
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.2860576923076924e-05,
+      "loss": 0.5582,
+      "step": 324
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.2836538461538463e-05,
+      "loss": 0.5879,
+      "step": 325
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.2812500000000001e-05,
+      "loss": 0.5916,
+      "step": 326
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.2788461538461538e-05,
+      "loss": 0.5654,
+      "step": 327
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.2764423076923077e-05,
+      "loss": 0.5762,
+      "step": 328
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.2740384615384615e-05,
+      "loss": 0.5632,
+      "step": 329
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.2716346153846155e-05,
+      "loss": 0.5815,
+      "step": 330
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.2692307692307693e-05,
+      "loss": 0.5544,
+      "step": 331
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.2668269230769233e-05,
+      "loss": 0.5836,
+      "step": 332
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.264423076923077e-05,
+      "loss": 0.5917,
+      "step": 333
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2620192307692309e-05,
+      "loss": 0.5436,
+      "step": 334
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2596153846153847e-05,
+      "loss": 0.5763,
+      "step": 335
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2572115384615387e-05,
+      "loss": 0.5854,
+      "step": 336
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2548076923076925e-05,
+      "loss": 0.5848,
+      "step": 337
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.2524038461538464e-05,
+      "loss": 0.5593,
+      "step": 338
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.25e-05,
+      "loss": 0.5729,
+      "step": 339
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.2475961538461539e-05,
+      "loss": 0.5872,
+      "step": 340
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.2451923076923078e-05,
+      "loss": 0.6385,
+      "step": 341
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.2427884615384616e-05,
+      "loss": 0.5848,
+      "step": 342
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.2403846153846156e-05,
+      "loss": 0.6006,
+      "step": 343
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.2379807692307692e-05,
+      "loss": 0.5952,
+      "step": 344
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.2355769230769232e-05,
+      "loss": 0.5433,
+      "step": 345
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.233173076923077e-05,
+      "loss": 0.6088,
+      "step": 346
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.230769230769231e-05,
+      "loss": 0.5818,
+      "step": 347
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.2283653846153846e-05,
+      "loss": 0.5653,
+      "step": 348
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.2259615384615386e-05,
+      "loss": 0.5521,
+      "step": 349
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.2235576923076924e-05,
+      "loss": 0.5397,
+      "step": 350
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.2211538461538463e-05,
+      "loss": 0.5787,
+      "step": 351
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.2187500000000001e-05,
+      "loss": 0.5623,
+      "step": 352
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.2163461538461538e-05,
+      "loss": 0.5457,
+      "step": 353
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.2139423076923077e-05,
+      "loss": 0.5912,
+      "step": 354
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.2115384615384615e-05,
+      "loss": 0.5615,
+      "step": 355
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.2091346153846155e-05,
+      "loss": 0.5859,
+      "step": 356
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.2067307692307693e-05,
+      "loss": 0.5529,
+      "step": 357
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.2043269230769233e-05,
+      "loss": 0.6052,
+      "step": 358
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.201923076923077e-05,
+      "loss": 0.5309,
+      "step": 359
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.1995192307692309e-05,
+      "loss": 0.5909,
+      "step": 360
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.1971153846153847e-05,
+      "loss": 0.6188,
+      "step": 361
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.1947115384615387e-05,
+      "loss": 0.6019,
+      "step": 362
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.1923076923076925e-05,
+      "loss": 0.5941,
+      "step": 363
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.1899038461538464e-05,
+      "loss": 0.5953,
+      "step": 364
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.1875e-05,
+      "loss": 0.57,
+      "step": 365
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.1850961538461539e-05,
+      "loss": 0.5834,
+      "step": 366
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.1826923076923078e-05,
+      "loss": 0.5331,
+      "step": 367
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.1802884615384615e-05,
+      "loss": 0.6333,
+      "step": 368
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.1778846153846154e-05,
+      "loss": 0.5826,
+      "step": 369
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.1754807692307692e-05,
+      "loss": 0.5818,
+      "step": 370
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.1730769230769232e-05,
+      "loss": 0.5869,
+      "step": 371
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.170673076923077e-05,
+      "loss": 0.5817,
+      "step": 372
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.168269230769231e-05,
+      "loss": 0.5742,
+      "step": 373
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.1658653846153846e-05,
+      "loss": 0.5462,
+      "step": 374
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.1634615384615386e-05,
+      "loss": 0.5851,
+      "step": 375
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.1610576923076924e-05,
+      "loss": 0.5891,
+      "step": 376
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.1586538461538464e-05,
+      "loss": 0.5687,
+      "step": 377
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.1562500000000002e-05,
+      "loss": 0.5648,
+      "step": 378
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.1538461538461538e-05,
+      "loss": 0.5909,
+      "step": 379
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.1514423076923078e-05,
+      "loss": 0.5784,
+      "step": 380
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.1490384615384616e-05,
+      "loss": 0.5534,
+      "step": 381
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.1466346153846155e-05,
+      "loss": 0.5947,
+      "step": 382
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.1442307692307693e-05,
+      "loss": 0.5397,
+      "step": 383
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.1418269230769233e-05,
+      "loss": 0.568,
+      "step": 384
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.139423076923077e-05,
+      "loss": 0.5931,
+      "step": 385
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.1370192307692309e-05,
+      "loss": 0.5522,
+      "step": 386
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.1346153846153847e-05,
+      "loss": 0.6481,
+      "step": 387
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.1322115384615387e-05,
+      "loss": 0.5383,
+      "step": 388
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.1298076923076923e-05,
+      "loss": 0.574,
+      "step": 389
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.1274038461538463e-05,
+      "loss": 0.5602,
+      "step": 390
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.125e-05,
+      "loss": 0.5638,
+      "step": 391
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.1225961538461539e-05,
+      "loss": 0.5927,
+      "step": 392
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.1201923076923078e-05,
+      "loss": 0.5886,
+      "step": 393
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.1177884615384615e-05,
+      "loss": 0.5739,
+      "step": 394
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.1153846153846154e-05,
+      "loss": 0.5886,
+      "step": 395
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.1129807692307692e-05,
+      "loss": 0.5989,
+      "step": 396
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.1105769230769232e-05,
+      "loss": 0.5391,
+      "step": 397
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.108173076923077e-05,
+      "loss": 0.6239,
+      "step": 398
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.105769230769231e-05,
+      "loss": 0.6043,
+      "step": 399
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.1033653846153846e-05,
+      "loss": 0.5794,
+      "step": 400
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.1009615384615386e-05,
+      "loss": 0.5703,
+      "step": 401
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.0985576923076924e-05,
+      "loss": 0.5712,
+      "step": 402
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.0961538461538464e-05,
+      "loss": 0.5726,
+      "step": 403
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.0937500000000002e-05,
+      "loss": 0.5776,
+      "step": 404
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.0913461538461538e-05,
+      "loss": 0.5926,
+      "step": 405
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.0889423076923078e-05,
+      "loss": 0.5388,
+      "step": 406
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.0865384615384616e-05,
+      "loss": 0.5392,
+      "step": 407
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.0841346153846155e-05,
+      "loss": 0.5408,
+      "step": 408
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.0817307692307693e-05,
+      "loss": 0.5864,
+      "step": 409
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.0793269230769231e-05,
+      "loss": 0.5912,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.076923076923077e-05,
+      "loss": 0.5715,
+      "step": 411
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.0745192307692309e-05,
+      "loss": 0.584,
+      "step": 412
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.0721153846153847e-05,
+      "loss": 0.5664,
+      "step": 413
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.0697115384615387e-05,
+      "loss": 0.5957,
+      "step": 414
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.0673076923076923e-05,
+      "loss": 0.598,
+      "step": 415
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.0649038461538463e-05,
+      "loss": 0.5529,
+      "step": 416
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.0625e-05,
+      "loss": 0.5624,
+      "step": 417
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.0600961538461539e-05,
+      "loss": 0.5889,
+      "step": 418
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.0576923076923078e-05,
+      "loss": 0.6202,
+      "step": 419
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.0552884615384615e-05,
+      "loss": 0.5405,
+      "step": 420
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.0528846153846154e-05,
+      "loss": 0.565,
+      "step": 421
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.0504807692307692e-05,
+      "loss": 0.5762,
+      "step": 422
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.0480769230769232e-05,
+      "loss": 0.5851,
+      "step": 423
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.045673076923077e-05,
+      "loss": 0.5716,
+      "step": 424
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.043269230769231e-05,
+      "loss": 0.5518,
+      "step": 425
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.0408653846153846e-05,
+      "loss": 0.603,
+      "step": 426
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.0384615384615386e-05,
+      "loss": 0.5858,
+      "step": 427
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0360576923076924e-05,
+      "loss": 0.5506,
+      "step": 428
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0336538461538464e-05,
+      "loss": 0.6052,
+      "step": 429
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0312500000000002e-05,
+      "loss": 0.5869,
+      "step": 430
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0288461538461538e-05,
+      "loss": 0.6164,
+      "step": 431
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0264423076923078e-05,
+      "loss": 0.6347,
+      "step": 432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0240384615384616e-05,
+      "loss": 0.6035,
+      "step": 433
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0216346153846155e-05,
+      "loss": 0.5835,
+      "step": 434
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0192307692307692e-05,
+      "loss": 0.6033,
+      "step": 435
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0168269230769231e-05,
+      "loss": 0.5874,
+      "step": 436
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.014423076923077e-05,
+      "loss": 0.5692,
+      "step": 437
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0120192307692309e-05,
+      "loss": 0.5365,
+      "step": 438
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0096153846153847e-05,
+      "loss": 0.5518,
+      "step": 439
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0072115384615387e-05,
+      "loss": 0.5838,
+      "step": 440
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.0048076923076923e-05,
+      "loss": 0.5861,
+      "step": 441
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.0024038461538463e-05,
+      "loss": 0.5746,
+      "step": 442
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1e-05,
+      "loss": 0.5803,
+      "step": 443
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.975961538461539e-06,
+      "loss": 0.6219,
+      "step": 444
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.951923076923079e-06,
+      "loss": 0.5853,
+      "step": 445
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.927884615384617e-06,
+      "loss": 0.5894,
+      "step": 446
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.903846153846155e-06,
+      "loss": 0.5536,
+      "step": 447
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.879807692307694e-06,
+      "loss": 0.5812,
+      "step": 448
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.855769230769232e-06,
+      "loss": 0.5742,
+      "step": 449
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.83173076923077e-06,
+      "loss": 0.575,
+      "step": 450
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.807692307692308e-06,
+      "loss": 0.5495,
+      "step": 451
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.783653846153846e-06,
+      "loss": 0.5423,
+      "step": 452
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.759615384615386e-06,
+      "loss": 0.548,
+      "step": 453
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.735576923076924e-06,
+      "loss": 0.5906,
+      "step": 454
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.711538461538462e-06,
+      "loss": 0.59,
+      "step": 455
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.6875e-06,
+      "loss": 0.6059,
+      "step": 456
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.66346153846154e-06,
+      "loss": 0.5551,
+      "step": 457
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.639423076923078e-06,
+      "loss": 0.5867,
+      "step": 458
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.615384615384616e-06,
+      "loss": 0.5696,
+      "step": 459
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.591346153846155e-06,
+      "loss": 0.5521,
+      "step": 460
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.567307692307693e-06,
+      "loss": 0.5807,
+      "step": 461
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.543269230769231e-06,
+      "loss": 0.5821,
+      "step": 462
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.51923076923077e-06,
+      "loss": 0.5961,
+      "step": 463
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.495192307692307e-06,
+      "loss": 0.5657,
+      "step": 464
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.471153846153847e-06,
+      "loss": 0.5487,
+      "step": 465
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.447115384615385e-06,
+      "loss": 0.6098,
+      "step": 466
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.423076923076923e-06,
+      "loss": 0.556,
+      "step": 467
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.399038461538463e-06,
+      "loss": 0.5521,
+      "step": 468
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.375000000000001e-06,
+      "loss": 0.5531,
+      "step": 469
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.350961538461539e-06,
+      "loss": 0.5908,
+      "step": 470
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.326923076923079e-06,
+      "loss": 0.586,
+      "step": 471
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.302884615384617e-06,
+      "loss": 0.5721,
+      "step": 472
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.278846153846155e-06,
+      "loss": 0.5881,
+      "step": 473
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.254807692307694e-06,
+      "loss": 0.554,
+      "step": 474
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.230769230769232e-06,
+      "loss": 0.558,
+      "step": 475
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.20673076923077e-06,
+      "loss": 0.5883,
+      "step": 476
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.182692307692308e-06,
+      "loss": 0.5684,
+      "step": 477
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.158653846153846e-06,
+      "loss": 0.5792,
+      "step": 478
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.134615384615384e-06,
+      "loss": 0.5644,
+      "step": 479
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.110576923076924e-06,
+      "loss": 0.5571,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.086538461538462e-06,
+      "loss": 0.5823,
+      "step": 481
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.0625e-06,
+      "loss": 0.583,
+      "step": 482
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.03846153846154e-06,
+      "loss": 0.6022,
+      "step": 483
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.014423076923078e-06,
+      "loss": 0.5704,
+      "step": 484
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 8.990384615384616e-06,
+      "loss": 0.5538,
+      "step": 485
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 8.966346153846155e-06,
+      "loss": 0.5836,
+      "step": 486
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 8.942307692307693e-06,
+      "loss": 0.6102,
+      "step": 487
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 8.918269230769231e-06,
+      "loss": 0.5811,
+      "step": 488
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 8.89423076923077e-06,
+      "loss": 0.5609,
+      "step": 489
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 8.870192307692308e-06,
+      "loss": 0.5853,
+      "step": 490
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.5538,
+      "step": 491
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 8.822115384615385e-06,
+      "loss": 0.5734,
+      "step": 492
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 8.798076923076923e-06,
+      "loss": 0.5479,
+      "step": 493
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 8.774038461538463e-06,
+      "loss": 0.6357,
+      "step": 494
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 8.750000000000001e-06,
+      "loss": 0.5924,
+      "step": 495
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 8.725961538461539e-06,
+      "loss": 0.5247,
+      "step": 496
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 8.701923076923079e-06,
+      "loss": 0.5947,
+      "step": 497
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 8.677884615384617e-06,
+      "loss": 0.6095,
+      "step": 498
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 8.653846153846155e-06,
+      "loss": 0.6045,
+      "step": 499
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 8.629807692307693e-06,
+      "loss": 0.554,
+      "step": 500
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 8.605769230769232e-06,
+      "loss": 0.5972,
+      "step": 501
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 8.581730769230769e-06,
+      "loss": 0.5682,
+      "step": 502
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 8.557692307692308e-06,
+      "loss": 0.5703,
+      "step": 503
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 8.533653846153846e-06,
+      "loss": 0.5617,
+      "step": 504
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 8.509615384615384e-06,
+      "loss": 0.576,
+      "step": 505
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 8.485576923076924e-06,
+      "loss": 0.5925,
+      "step": 506
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 8.461538461538462e-06,
+      "loss": 0.588,
+      "step": 507
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 8.4375e-06,
+      "loss": 0.588,
+      "step": 508
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 8.41346153846154e-06,
+      "loss": 0.562,
+      "step": 509
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 8.389423076923078e-06,
+      "loss": 0.5432,
+      "step": 510
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 8.365384615384616e-06,
+      "loss": 0.5578,
+      "step": 511
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 8.341346153846156e-06,
+      "loss": 0.543,
+      "step": 512
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 8.317307692307694e-06,
+      "loss": 0.5429,
+      "step": 513
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 8.293269230769232e-06,
+      "loss": 0.5457,
+      "step": 514
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 8.26923076923077e-06,
+      "loss": 0.6041,
+      "step": 515
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 8.245192307692308e-06,
+      "loss": 0.5931,
+      "step": 516
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 8.221153846153847e-06,
+      "loss": 0.5622,
+      "step": 517
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 8.197115384615385e-06,
+      "loss": 0.5353,
+      "step": 518
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 8.173076923076923e-06,
+      "loss": 0.5195,
+      "step": 519
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 8.149038461538461e-06,
+      "loss": 0.5721,
+      "step": 520
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 8.125000000000001e-06,
+      "loss": 0.6149,
+      "step": 521
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 8.100961538461539e-06,
+      "loss": 0.5822,
+      "step": 522
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 8.076923076923077e-06,
+      "loss": 0.5433,
+      "step": 523
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 8.052884615384617e-06,
+      "loss": 0.5626,
+      "step": 524
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 8.028846153846155e-06,
+      "loss": 0.5816,
+      "step": 525
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 8.004807692307693e-06,
+      "loss": 0.5583,
+      "step": 526
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 7.980769230769232e-06,
+      "loss": 0.5719,
+      "step": 527
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 7.956730769230769e-06,
+      "loss": 0.591,
+      "step": 528
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 7.932692307692308e-06,
+      "loss": 0.5675,
+      "step": 529
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 7.908653846153846e-06,
+      "loss": 0.5986,
+      "step": 530
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 7.884615384615384e-06,
+      "loss": 0.5621,
+      "step": 531
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 7.860576923076924e-06,
+      "loss": 0.6024,
+      "step": 532
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 7.836538461538462e-06,
+      "loss": 0.5634,
+      "step": 533
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 7.8125e-06,
+      "loss": 0.5562,
+      "step": 534
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 7.78846153846154e-06,
+      "loss": 0.5527,
+      "step": 535
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 7.764423076923078e-06,
+      "loss": 0.5609,
+      "step": 536
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 7.740384615384616e-06,
+      "loss": 0.5874,
+      "step": 537
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 7.716346153846156e-06,
+      "loss": 0.5712,
+      "step": 538
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": 0.5751,
+      "step": 539
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.668269230769232e-06,
+      "loss": 0.5978,
+      "step": 540
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.64423076923077e-06,
+      "loss": 0.5703,
+      "step": 541
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.620192307692308e-06,
+      "loss": 0.5851,
+      "step": 542
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.5961538461538465e-06,
+      "loss": 0.5512,
+      "step": 543
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 7.572115384615385e-06,
+      "loss": 0.5604,
+      "step": 544
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 7.548076923076923e-06,
+      "loss": 0.5328,
+      "step": 545
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 7.524038461538462e-06,
+      "loss": 0.5616,
+      "step": 546
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.5295,
+      "step": 547
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 7.475961538461539e-06,
+      "loss": 0.5721,
+      "step": 548
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 7.451923076923078e-06,
+      "loss": 0.5834,
+      "step": 549
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 7.427884615384616e-06,
+      "loss": 0.5781,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 7.403846153846155e-06,
+      "loss": 0.546,
+      "step": 551
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.379807692307694e-06,
+      "loss": 0.6229,
+      "step": 552
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.355769230769232e-06,
+      "loss": 0.5565,
+      "step": 553
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.33173076923077e-06,
+      "loss": 0.5745,
+      "step": 554
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.307692307692308e-06,
+      "loss": 0.5603,
+      "step": 555
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.2836538461538465e-06,
+      "loss": 0.5776,
+      "step": 556
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 7.259615384615385e-06,
+      "loss": 0.608,
+      "step": 557
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 7.235576923076923e-06,
+      "loss": 0.5664,
+      "step": 558
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 7.211538461538462e-06,
+      "loss": 0.5754,
+      "step": 559
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 7.1875e-06,
+      "loss": 0.591,
+      "step": 560
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 7.163461538461539e-06,
+      "loss": 0.6003,
+      "step": 561
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 7.139423076923078e-06,
+      "loss": 0.5684,
+      "step": 562
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 7.115384615384616e-06,
+      "loss": 0.5704,
+      "step": 563
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 7.091346153846155e-06,
+      "loss": 0.6031,
+      "step": 564
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 7.067307692307694e-06,
+      "loss": 0.569,
+      "step": 565
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 7.043269230769232e-06,
+      "loss": 0.5806,
+      "step": 566
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 7.01923076923077e-06,
+      "loss": 0.6005,
+      "step": 567
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 6.995192307692308e-06,
+      "loss": 0.6056,
+      "step": 568
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 6.9711538461538465e-06,
+      "loss": 0.5342,
+      "step": 569
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 6.9471153846153845e-06,
+      "loss": 0.6388,
+      "step": 570
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 6.923076923076923e-06,
+      "loss": 0.54,
+      "step": 571
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 6.899038461538462e-06,
+      "loss": 0.5943,
+      "step": 572
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 6.875e-06,
+      "loss": 0.5397,
+      "step": 573
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 6.850961538461539e-06,
+      "loss": 0.5663,
+      "step": 574
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 6.826923076923078e-06,
+      "loss": 0.5677,
+      "step": 575
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 6.802884615384616e-06,
+      "loss": 0.5883,
+      "step": 576
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 6.778846153846155e-06,
+      "loss": 0.6236,
+      "step": 577
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 6.754807692307694e-06,
+      "loss": 0.5652,
+      "step": 578
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 6.730769230769232e-06,
+      "loss": 0.5629,
+      "step": 579
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 6.70673076923077e-06,
+      "loss": 0.5783,
+      "step": 580
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 6.682692307692308e-06,
+      "loss": 0.538,
+      "step": 581
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 6.6586538461538465e-06,
+      "loss": 0.5702,
+      "step": 582
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 6.6346153846153846e-06,
+      "loss": 0.5765,
+      "step": 583
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 6.610576923076923e-06,
+      "loss": 0.5654,
+      "step": 584
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 6.586538461538462e-06,
+      "loss": 0.564,
+      "step": 585
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 6.5625e-06,
+      "loss": 0.5911,
+      "step": 586
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 6.538461538461539e-06,
+      "loss": 0.541,
+      "step": 587
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 6.514423076923078e-06,
+      "loss": 0.6175,
+      "step": 588
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 6.490384615384616e-06,
+      "loss": 0.5613,
+      "step": 589
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 6.466346153846155e-06,
+      "loss": 0.5631,
+      "step": 590
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 6.442307692307693e-06,
+      "loss": 0.5273,
+      "step": 591
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 6.418269230769232e-06,
+      "loss": 0.5717,
+      "step": 592
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 6.394230769230769e-06,
+      "loss": 0.6063,
+      "step": 593
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 6.370192307692308e-06,
+      "loss": 0.5787,
+      "step": 594
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 6.3461538461538466e-06,
+      "loss": 0.579,
+      "step": 595
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 6.322115384615385e-06,
+      "loss": 0.5468,
+      "step": 596
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 6.2980769230769234e-06,
+      "loss": 0.5497,
+      "step": 597
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 6.274038461538462e-06,
+      "loss": 0.6032,
+      "step": 598
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 6.25e-06,
+      "loss": 0.5299,
+      "step": 599
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 6.225961538461539e-06,
+      "loss": 0.6111,
+      "step": 600
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 6.201923076923078e-06,
+      "loss": 0.5409,
+      "step": 601
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 6.177884615384616e-06,
+      "loss": 0.5344,
+      "step": 602
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 6.153846153846155e-06,
+      "loss": 0.579,
+      "step": 603
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 6.129807692307693e-06,
+      "loss": 0.5859,
+      "step": 604
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 6.105769230769232e-06,
+      "loss": 0.5696,
+      "step": 605
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 6.081730769230769e-06,
+      "loss": 0.5406,
+      "step": 606
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 6.057692307692308e-06,
+      "loss": 0.6291,
+      "step": 607
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 6.033653846153847e-06,
+      "loss": 0.583,
+      "step": 608
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 6.009615384615385e-06,
+      "loss": 0.6049,
+      "step": 609
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.9855769230769235e-06,
+      "loss": 0.5971,
+      "step": 610
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.961538461538462e-06,
+      "loss": 0.5846,
+      "step": 611
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.9375e-06,
+      "loss": 0.5445,
+      "step": 612
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 5.913461538461539e-06,
+      "loss": 0.5817,
+      "step": 613
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 5.889423076923077e-06,
+      "loss": 0.5707,
+      "step": 614
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 5.865384615384616e-06,
+      "loss": 0.6075,
+      "step": 615
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 5.841346153846155e-06,
+      "loss": 0.5947,
+      "step": 616
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 5.817307692307693e-06,
+      "loss": 0.5655,
+      "step": 617
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 5.793269230769232e-06,
+      "loss": 0.5497,
+      "step": 618
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 5.769230769230769e-06,
+      "loss": 0.6269,
+      "step": 619
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 5.745192307692308e-06,
+      "loss": 0.5865,
+      "step": 620
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.721153846153847e-06,
+      "loss": 0.5353,
+      "step": 621
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.697115384615385e-06,
+      "loss": 0.5736,
+      "step": 622
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.6730769230769235e-06,
+      "loss": 0.6105,
+      "step": 623
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.6490384615384615e-06,
+      "loss": 0.5777,
+      "step": 624
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 5.625e-06,
+      "loss": 0.5785,
+      "step": 625
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 5.600961538461539e-06,
+      "loss": 0.5775,
+      "step": 626
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 5.576923076923077e-06,
+      "loss": 0.5749,
+      "step": 627
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 5.552884615384616e-06,
+      "loss": 0.5714,
+      "step": 628
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 5.528846153846155e-06,
+      "loss": 0.5971,
+      "step": 629
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 5.504807692307693e-06,
+      "loss": 0.5645,
+      "step": 630
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 5.480769230769232e-06,
+      "loss": 0.5997,
+      "step": 631
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 5.456730769230769e-06,
+      "loss": 0.5377,
+      "step": 632
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 5.432692307692308e-06,
+      "loss": 0.5543,
+      "step": 633
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.408653846153847e-06,
+      "loss": 0.5677,
+      "step": 634
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.384615384615385e-06,
+      "loss": 0.624,
+      "step": 635
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.3605769230769235e-06,
+      "loss": 0.5744,
+      "step": 636
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.3365384615384615e-06,
+      "loss": 0.5352,
+      "step": 637
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 5.3125e-06,
+      "loss": 0.5955,
+      "step": 638
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 5.288461538461539e-06,
+      "loss": 0.5307,
+      "step": 639
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 5.264423076923077e-06,
+      "loss": 0.5699,
+      "step": 640
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 5.240384615384616e-06,
+      "loss": 0.5429,
+      "step": 641
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 5.216346153846155e-06,
+      "loss": 0.5446,
+      "step": 642
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 5.192307692307693e-06,
+      "loss": 0.5272,
+      "step": 643
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 5.168269230769232e-06,
+      "loss": 0.5578,
+      "step": 644
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 5.144230769230769e-06,
+      "loss": 0.5155,
+      "step": 645
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 5.120192307692308e-06,
+      "loss": 0.5559,
+      "step": 646
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 5.096153846153846e-06,
+      "loss": 0.5923,
+      "step": 647
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 5.072115384615385e-06,
+      "loss": 0.588,
+      "step": 648
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 5.0480769230769235e-06,
+      "loss": 0.6105,
+      "step": 649
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 5.0240384615384616e-06,
+      "loss": 0.5372,
+      "step": 650
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 5e-06,
+      "loss": 0.5566,
+      "step": 651
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.975961538461539e-06,
+      "loss": 0.5709,
+      "step": 652
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.951923076923077e-06,
+      "loss": 0.5494,
+      "step": 653
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.927884615384616e-06,
+      "loss": 0.5717,
+      "step": 654
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.903846153846154e-06,
+      "loss": 0.567,
+      "step": 655
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.879807692307693e-06,
+      "loss": 0.5234,
+      "step": 656
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.855769230769231e-06,
+      "loss": 0.5668,
+      "step": 657
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.83173076923077e-06,
+      "loss": 0.5875,
+      "step": 658
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.807692307692308e-06,
+      "loss": 0.6101,
+      "step": 659
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.783653846153847e-06,
+      "loss": 0.5547,
+      "step": 660
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.759615384615385e-06,
+      "loss": 0.5435,
+      "step": 661
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.7355769230769236e-06,
+      "loss": 0.5875,
+      "step": 662
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.711538461538462e-06,
+      "loss": 0.589,
+      "step": 663
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.6875000000000004e-06,
+      "loss": 0.5746,
+      "step": 664
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.663461538461539e-06,
+      "loss": 0.5612,
+      "step": 665
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.639423076923077e-06,
+      "loss": 0.5919,
+      "step": 666
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.615384615384616e-06,
+      "loss": 0.5341,
+      "step": 667
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.591346153846154e-06,
+      "loss": 0.5659,
+      "step": 668
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.567307692307692e-06,
+      "loss": 0.5679,
+      "step": 669
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.543269230769231e-06,
+      "loss": 0.5583,
+      "step": 670
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.51923076923077e-06,
+      "loss": 0.5677,
+      "step": 671
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.495192307692308e-06,
+      "loss": 0.5428,
+      "step": 672
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.471153846153847e-06,
+      "loss": 0.5833,
+      "step": 673
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.447115384615385e-06,
+      "loss": 0.5408,
+      "step": 674
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.423076923076924e-06,
+      "loss": 0.5661,
+      "step": 675
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.399038461538462e-06,
+      "loss": 0.5667,
+      "step": 676
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.3750000000000005e-06,
+      "loss": 0.577,
+      "step": 677
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.350961538461539e-06,
+      "loss": 0.6151,
+      "step": 678
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.326923076923077e-06,
+      "loss": 0.5657,
+      "step": 679
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.302884615384616e-06,
+      "loss": 0.5956,
+      "step": 680
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.278846153846154e-06,
+      "loss": 0.5724,
+      "step": 681
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.254807692307692e-06,
+      "loss": 0.5389,
+      "step": 682
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.230769230769231e-06,
+      "loss": 0.5206,
+      "step": 683
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.20673076923077e-06,
+      "loss": 0.5973,
+      "step": 684
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.182692307692308e-06,
+      "loss": 0.5916,
+      "step": 685
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.158653846153847e-06,
+      "loss": 0.5646,
+      "step": 686
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.134615384615385e-06,
+      "loss": 0.5906,
+      "step": 687
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.110576923076924e-06,
+      "loss": 0.5751,
+      "step": 688
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.086538461538462e-06,
+      "loss": 0.5749,
+      "step": 689
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.0625000000000005e-06,
+      "loss": 0.5598,
+      "step": 690
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.0384615384615385e-06,
+      "loss": 0.5733,
+      "step": 691
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.014423076923077e-06,
+      "loss": 0.5831,
+      "step": 692
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 3.990384615384616e-06,
+      "loss": 0.6017,
+      "step": 693
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.966346153846154e-06,
+      "loss": 0.59,
+      "step": 694
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.942307692307692e-06,
+      "loss": 0.5706,
+      "step": 695
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.918269230769231e-06,
+      "loss": 0.5545,
+      "step": 696
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.89423076923077e-06,
+      "loss": 0.5924,
+      "step": 697
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.870192307692308e-06,
+      "loss": 0.5701,
+      "step": 698
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": 0.6022,
+      "step": 699
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.822115384615385e-06,
+      "loss": 0.584,
+      "step": 700
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.7980769230769232e-06,
+      "loss": 0.5518,
+      "step": 701
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.7740384615384617e-06,
+      "loss": 0.6042,
+      "step": 702
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.5284,
+      "step": 703
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.725961538461539e-06,
+      "loss": 0.5829,
+      "step": 704
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.7019230769230774e-06,
+      "loss": 0.5931,
+      "step": 705
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.677884615384616e-06,
+      "loss": 0.5654,
+      "step": 706
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.653846153846154e-06,
+      "loss": 0.5387,
+      "step": 707
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.6298076923076927e-06,
+      "loss": 0.5472,
+      "step": 708
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.605769230769231e-06,
+      "loss": 0.5762,
+      "step": 709
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.5817307692307695e-06,
+      "loss": 0.5704,
+      "step": 710
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.557692307692308e-06,
+      "loss": 0.5401,
+      "step": 711
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.533653846153847e-06,
+      "loss": 0.5818,
+      "step": 712
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.509615384615385e-06,
+      "loss": 0.5797,
+      "step": 713
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.4855769230769233e-06,
+      "loss": 0.5397,
+      "step": 714
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.4615384615384617e-06,
+      "loss": 0.5687,
+      "step": 715
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.4375e-06,
+      "loss": 0.5544,
+      "step": 716
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.413461538461539e-06,
+      "loss": 0.5822,
+      "step": 717
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.3894230769230774e-06,
+      "loss": 0.6182,
+      "step": 718
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.365384615384616e-06,
+      "loss": 0.5567,
+      "step": 719
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.341346153846154e-06,
+      "loss": 0.5692,
+      "step": 720
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.3173076923076923e-06,
+      "loss": 0.5689,
+      "step": 721
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.293269230769231e-06,
+      "loss": 0.574,
+      "step": 722
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.2692307692307696e-06,
+      "loss": 0.5841,
+      "step": 723
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.245192307692308e-06,
+      "loss": 0.5365,
+      "step": 724
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.2211538461538464e-06,
+      "loss": 0.5425,
+      "step": 725
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.1971153846153844e-06,
+      "loss": 0.5562,
+      "step": 726
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.1730769230769233e-06,
+      "loss": 0.5227,
+      "step": 727
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.1490384615384617e-06,
+      "loss": 0.5789,
+      "step": 728
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.125e-06,
+      "loss": 0.5599,
+      "step": 729
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.100961538461539e-06,
+      "loss": 0.5673,
+      "step": 730
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.0769230769230774e-06,
+      "loss": 0.6091,
+      "step": 731
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.052884615384616e-06,
+      "loss": 0.6308,
+      "step": 732
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 3.028846153846154e-06,
+      "loss": 0.569,
+      "step": 733
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 3.0048076923076923e-06,
+      "loss": 0.5713,
+      "step": 734
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.980769230769231e-06,
+      "loss": 0.5671,
+      "step": 735
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.9567307692307696e-06,
+      "loss": 0.5172,
+      "step": 736
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.932692307692308e-06,
+      "loss": 0.5975,
+      "step": 737
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.9086538461538465e-06,
+      "loss": 0.5636,
+      "step": 738
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.8846153846153845e-06,
+      "loss": 0.5864,
+      "step": 739
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.8605769230769233e-06,
+      "loss": 0.5733,
+      "step": 740
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.8365384615384617e-06,
+      "loss": 0.6079,
+      "step": 741
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.8125e-06,
+      "loss": 0.5758,
+      "step": 742
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.7884615384615386e-06,
+      "loss": 0.5284,
+      "step": 743
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.7644230769230775e-06,
+      "loss": 0.5422,
+      "step": 744
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.740384615384616e-06,
+      "loss": 0.5345,
+      "step": 745
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.716346153846154e-06,
+      "loss": 0.5493,
+      "step": 746
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.6923076923076923e-06,
+      "loss": 0.5594,
+      "step": 747
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.6682692307692308e-06,
+      "loss": 0.569,
+      "step": 748
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.6442307692307696e-06,
+      "loss": 0.5296,
+      "step": 749
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.620192307692308e-06,
+      "loss": 0.5775,
+      "step": 750
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.5961538461538465e-06,
+      "loss": 0.5868,
+      "step": 751
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.5721153846153845e-06,
+      "loss": 0.5633,
+      "step": 752
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.548076923076923e-06,
+      "loss": 0.5348,
+      "step": 753
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.5240384615384618e-06,
+      "loss": 0.5995,
+      "step": 754
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.5e-06,
+      "loss": 0.5748,
+      "step": 755
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.4759615384615386e-06,
+      "loss": 0.5933,
+      "step": 756
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.451923076923077e-06,
+      "loss": 0.5773,
+      "step": 757
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.4278846153846155e-06,
+      "loss": 0.5765,
+      "step": 758
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.403846153846154e-06,
+      "loss": 0.5733,
+      "step": 759
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.3798076923076924e-06,
+      "loss": 0.5916,
+      "step": 760
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.355769230769231e-06,
+      "loss": 0.559,
+      "step": 761
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.3317307692307696e-06,
+      "loss": 0.5641,
+      "step": 762
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.307692307692308e-06,
+      "loss": 0.5606,
+      "step": 763
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.283653846153846e-06,
+      "loss": 0.5945,
+      "step": 764
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.259615384615385e-06,
+      "loss": 0.5992,
+      "step": 765
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.2355769230769234e-06,
+      "loss": 0.6104,
+      "step": 766
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.211538461538462e-06,
+      "loss": 0.5675,
+      "step": 767
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.1875000000000002e-06,
+      "loss": 0.5704,
+      "step": 768
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.1634615384615387e-06,
+      "loss": 0.5529,
+      "step": 769
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.139423076923077e-06,
+      "loss": 0.5756,
+      "step": 770
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.1153846153846155e-06,
+      "loss": 0.5558,
+      "step": 771
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.091346153846154e-06,
+      "loss": 0.5777,
+      "step": 772
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.0673076923076924e-06,
+      "loss": 0.5465,
+      "step": 773
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.043269230769231e-06,
+      "loss": 0.5587,
+      "step": 774
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.0192307692307692e-06,
+      "loss": 0.5646,
+      "step": 775
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.995192307692308e-06,
+      "loss": 0.5456,
+      "step": 776
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.971153846153846e-06,
+      "loss": 0.5291,
+      "step": 777
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.947115384615385e-06,
+      "loss": 0.5699,
+      "step": 778
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.9230769230769234e-06,
+      "loss": 0.5853,
+      "step": 779
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.8990384615384616e-06,
+      "loss": 0.5972,
+      "step": 780
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.8750000000000003e-06,
+      "loss": 0.583,
+      "step": 781
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.8509615384615387e-06,
+      "loss": 0.5487,
+      "step": 782
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.826923076923077e-06,
+      "loss": 0.6015,
+      "step": 783
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.8028846153846156e-06,
+      "loss": 0.5643,
+      "step": 784
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.778846153846154e-06,
+      "loss": 0.598,
+      "step": 785
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.7548076923076924e-06,
+      "loss": 0.5805,
+      "step": 786
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.7307692307692308e-06,
+      "loss": 0.5818,
+      "step": 787
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.7067307692307695e-06,
+      "loss": 0.5413,
+      "step": 788
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.682692307692308e-06,
+      "loss": 0.5826,
+      "step": 789
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.6586538461538461e-06,
+      "loss": 0.5505,
+      "step": 790
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.6346153846153848e-06,
+      "loss": 0.5916,
+      "step": 791
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.6105769230769232e-06,
+      "loss": 0.6,
+      "step": 792
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.5865384615384616e-06,
+      "loss": 0.5938,
+      "step": 793
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.5625e-06,
+      "loss": 0.5692,
+      "step": 794
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.5384615384615387e-06,
+      "loss": 0.5505,
+      "step": 795
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.514423076923077e-06,
+      "loss": 0.524,
+      "step": 796
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4903846153846156e-06,
+      "loss": 0.536,
+      "step": 797
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.466346153846154e-06,
+      "loss": 0.6124,
+      "step": 798
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4423076923076922e-06,
+      "loss": 0.5699,
+      "step": 799
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4182692307692309e-06,
+      "loss": 0.5652,
+      "step": 800
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.3942307692307693e-06,
+      "loss": 0.5985,
+      "step": 801
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.370192307692308e-06,
+      "loss": 0.6199,
+      "step": 802
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.3461538461538462e-06,
+      "loss": 0.6243,
+      "step": 803
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.3221153846153848e-06,
+      "loss": 0.5363,
+      "step": 804
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.2980769230769232e-06,
+      "loss": 0.5646,
+      "step": 805
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.2740384615384615e-06,
+      "loss": 0.5824,
+      "step": 806
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.25e-06,
+      "loss": 0.5426,
+      "step": 807
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.2259615384615385e-06,
+      "loss": 0.5691,
+      "step": 808
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.201923076923077e-06,
+      "loss": 0.5524,
+      "step": 809
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.1778846153846154e-06,
+      "loss": 0.5539,
+      "step": 810
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 0.5635,
+      "step": 811
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.1298076923076925e-06,
+      "loss": 0.5777,
+      "step": 812
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.105769230769231e-06,
+      "loss": 0.5822,
+      "step": 813
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.0817307692307693e-06,
+      "loss": 0.5582,
+      "step": 814
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.0576923076923078e-06,
+      "loss": 0.5512,
+      "step": 815
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.0336538461538462e-06,
+      "loss": 0.5706,
+      "step": 816
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.0096153846153846e-06,
+      "loss": 0.5636,
+      "step": 817
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.85576923076923e-07,
+      "loss": 0.566,
+      "step": 818
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.615384615384617e-07,
+      "loss": 0.5789,
+      "step": 819
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.375000000000001e-07,
+      "loss": 0.553,
+      "step": 820
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.134615384615385e-07,
+      "loss": 0.6072,
+      "step": 821
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 8.89423076923077e-07,
+      "loss": 0.5767,
+      "step": 822
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 8.653846153846154e-07,
+      "loss": 0.572,
+      "step": 823
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 8.41346153846154e-07,
+      "loss": 0.5745,
+      "step": 824
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 8.173076923076924e-07,
+      "loss": 0.5441,
+      "step": 825
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 7.932692307692308e-07,
+      "loss": 0.553,
+      "step": 826
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.692307692307694e-07,
+      "loss": 0.5425,
+      "step": 827
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.451923076923078e-07,
+      "loss": 0.5328,
+      "step": 828
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.211538461538461e-07,
+      "loss": 0.5143,
+      "step": 829
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.971153846153847e-07,
+      "loss": 0.5445,
+      "step": 830
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.730769230769231e-07,
+      "loss": 0.5471,
+      "step": 831
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 6.490384615384616e-07,
+      "loss": 0.6031,
+      "step": 832
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 6.25e-07,
+      "loss": 0.5807,
+      "step": 833
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 6.009615384615385e-07,
+      "loss": 0.5375,
+      "step": 834
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.76923076923077e-07,
+      "loss": 0.5817,
+      "step": 835
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.528846153846155e-07,
+      "loss": 0.547,
+      "step": 836
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.288461538461539e-07,
+      "loss": 0.6014,
+      "step": 837
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.048076923076923e-07,
+      "loss": 0.6045,
+      "step": 838
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.807692307692308e-07,
+      "loss": 0.5987,
+      "step": 839
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 4.5673076923076923e-07,
+      "loss": 0.5557,
+      "step": 840
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 4.326923076923077e-07,
+      "loss": 0.5507,
+      "step": 841
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 4.086538461538462e-07,
+      "loss": 0.6052,
+      "step": 842
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 3.846153846153847e-07,
+      "loss": 0.5356,
+      "step": 843
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.6057692307692306e-07,
+      "loss": 0.5851,
+      "step": 844
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.3653846153846154e-07,
+      "loss": 0.5717,
+      "step": 845
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.125e-07,
+      "loss": 0.6172,
+      "step": 846
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.884615384615385e-07,
+      "loss": 0.5966,
+      "step": 847
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.6442307692307694e-07,
+      "loss": 0.5934,
+      "step": 848
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.403846153846154e-07,
+      "loss": 0.5319,
+      "step": 849
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.1634615384615386e-07,
+      "loss": 0.5696,
+      "step": 850
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.9230769230769234e-07,
+      "loss": 0.5562,
+      "step": 851
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.6826923076923077e-07,
+      "loss": 0.5972,
+      "step": 852
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.4423076923076925e-07,
+      "loss": 0.5502,
+      "step": 853
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.201923076923077e-07,
+      "loss": 0.598,
+      "step": 854
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.615384615384617e-08,
+      "loss": 0.5461,
+      "step": 855
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 7.211538461538463e-08,
+      "loss": 0.5471,
+      "step": 856
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8076923076923085e-08,
+      "loss": 0.5612,
+      "step": 857
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4038461538461542e-08,
+      "loss": 0.5822,
+      "step": 858
+    },
+    {
+      "epoch": 2.0,
+      "step": 858,
+      "total_flos": 3.272941187438215e+18,
+      "train_loss": 0.5945545899006592,
+      "train_runtime": 12838.6545,
+      "train_samples_per_second": 8.564,
+      "train_steps_per_second": 0.067
+    }
+  ],
+  "max_steps": 858,
+  "num_train_epochs": 2,
+  "total_flos": 3.272941187438215e+18,
+  "trial_name": null,
+  "trial_params": null
+}