diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,6019 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.8416622830089426,
+  "eval_steps": 500,
+  "global_step": 1000,
+  "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": 2.8259,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7894294565112984e-06,
+      "loss": 2.5557,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.421141086977404e-06,
+      "loss": 2.7329,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.578858913022597e-06,
+      "loss": 2.7307,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.47685462377997e-06,
+      "loss": 2.7151,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.210570543488702e-06,
+      "loss": 2.5732,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.830918514469461e-06,
+      "loss": 2.6619,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.368288369533896e-06,
+      "loss": 2.6477,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.842282173954808e-06,
+      "loss": 2.6763,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.26628408029127e-06,
+      "loss": 2.7761,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.64984045981344e-06,
+      "loss": 2.7258,
+      "step": 11
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1e-05,
+      "loss": 2.6809,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1e-05,
+      "loss": 2.5706,
+      "step": 13
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1e-05,
+      "loss": 2.7214,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1e-05,
+      "loss": 2.5208,
+      "step": 15
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1e-05,
+      "loss": 2.4438,
+      "step": 16
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1e-05,
+      "loss": 2.6816,
+      "step": 17
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.6377,
+      "step": 18
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.7363,
+      "step": 19
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.707,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.5566,
+      "step": 21
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.6118,
+      "step": 22
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.554,
+      "step": 23
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.5769,
+      "step": 24
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.6658,
+      "step": 25
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.7537,
+      "step": 26
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.3424,
+      "step": 27
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.5845,
+      "step": 28
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 2.6904,
+      "step": 29
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.6335,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.5015,
+      "step": 31
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.6611,
+      "step": 32
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.4358,
+      "step": 33
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.5933,
+      "step": 34
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.5588,
+      "step": 35
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.6606,
+      "step": 36
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.4919,
+      "step": 37
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.5081,
+      "step": 38
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.5645,
+      "step": 39
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.5762,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1e-05,
+      "loss": 2.5271,
+      "step": 41
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.6013,
+      "step": 42
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.5552,
+      "step": 43
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.4739,
+      "step": 44
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.5588,
+      "step": 45
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.4092,
+      "step": 46
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.4539,
+      "step": 47
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.4312,
+      "step": 48
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.6229,
+      "step": 49
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.4873,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.3577,
+      "step": 51
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.4965,
+      "step": 52
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 2.5461,
+      "step": 53
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.449,
+      "step": 54
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.5027,
+      "step": 55
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.4758,
+      "step": 56
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.4668,
+      "step": 57
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.5155,
+      "step": 58
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.4065,
+      "step": 59
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.4031,
+      "step": 60
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.3389,
+      "step": 61
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.2373,
+      "step": 62
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.4731,
+      "step": 63
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.4128,
+      "step": 64
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.5498,
+      "step": 65
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.4177,
+      "step": 66
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.4893,
+      "step": 67
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.2933,
+      "step": 68
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.5251,
+      "step": 69
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.4397,
+      "step": 70
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.4528,
+      "step": 71
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.5054,
+      "step": 72
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.3779,
+      "step": 73
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.4624,
+      "step": 74
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.4575,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.3148,
+      "step": 76
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 2.4647,
+      "step": 77
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.4636,
+      "step": 78
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.322,
+      "step": 79
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.5166,
+      "step": 80
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.4895,
+      "step": 81
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.4092,
+      "step": 82
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.4055,
+      "step": 83
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.488,
+      "step": 84
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.5962,
+      "step": 85
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.4296,
+      "step": 86
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.303,
+      "step": 87
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.4531,
+      "step": 88
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 2.2974,
+      "step": 89
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.4592,
+      "step": 90
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.3069,
+      "step": 91
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.4932,
+      "step": 92
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.4792,
+      "step": 93
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.2467,
+      "step": 94
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.4437,
+      "step": 95
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.531,
+      "step": 96
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.454,
+      "step": 97
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.349,
+      "step": 98
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.3259,
+      "step": 99
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 2.4045,
+      "step": 100
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.476,
+      "step": 101
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.5391,
+      "step": 102
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.5425,
+      "step": 103
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.4038,
+      "step": 104
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.4192,
+      "step": 105
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.4674,
+      "step": 106
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.4766,
+      "step": 107
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.4172,
+      "step": 108
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.3806,
+      "step": 109
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.5781,
+      "step": 110
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.2716,
+      "step": 111
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 2.2889,
+      "step": 112
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.4734,
+      "step": 113
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.2109,
+      "step": 114
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.475,
+      "step": 115
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.3792,
+      "step": 116
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.2552,
+      "step": 117
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.3818,
+      "step": 118
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.3567,
+      "step": 119
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.3621,
+      "step": 120
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.459,
+      "step": 121
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.3079,
+      "step": 122
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.3638,
+      "step": 123
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 2.2089,
+      "step": 124
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.3926,
+      "step": 125
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.1709,
+      "step": 126
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.4202,
+      "step": 127
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.4788,
+      "step": 128
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.5142,
+      "step": 129
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.3896,
+      "step": 130
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.3799,
+      "step": 131
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.7068,
+      "step": 132
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.3345,
+      "step": 133
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.3035,
+      "step": 134
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.3997,
+      "step": 135
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 2.2101,
+      "step": 136
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.3604,
+      "step": 137
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.2592,
+      "step": 138
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.3757,
+      "step": 139
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.245,
+      "step": 140
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.406,
+      "step": 141
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.4346,
+      "step": 142
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.479,
+      "step": 143
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.2181,
+      "step": 144
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.2773,
+      "step": 145
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.3723,
+      "step": 146
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.2892,
+      "step": 147
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 2.4207,
+      "step": 148
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.3695,
+      "step": 149
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.415,
+      "step": 150
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.3501,
+      "step": 151
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.4258,
+      "step": 152
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.2216,
+      "step": 153
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.5314,
+      "step": 154
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.2883,
+      "step": 155
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.3345,
+      "step": 156
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.3342,
+      "step": 157
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.4011,
+      "step": 158
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.5132,
+      "step": 159
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 2.4756,
+      "step": 160
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.3672,
+      "step": 161
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.2925,
+      "step": 162
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.4006,
+      "step": 163
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.2201,
+      "step": 164
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.309,
+      "step": 165
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.3008,
+      "step": 166
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.4033,
+      "step": 167
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.3649,
+      "step": 168
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.3375,
+      "step": 169
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.2927,
+      "step": 170
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.2726,
+      "step": 171
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 2.3896,
+      "step": 172
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.4485,
+      "step": 173
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.3717,
+      "step": 174
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.4014,
+      "step": 175
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.436,
+      "step": 176
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.4148,
+      "step": 177
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.3759,
+      "step": 178
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.1646,
+      "step": 179
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.2839,
+      "step": 180
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.2582,
+      "step": 181
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.4131,
+      "step": 182
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.4531,
+      "step": 183
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 2.5225,
+      "step": 184
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.4196,
+      "step": 185
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.3701,
+      "step": 186
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.4148,
+      "step": 187
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.2227,
+      "step": 188
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.2859,
+      "step": 189
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.4951,
+      "step": 190
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.4355,
+      "step": 191
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.3933,
+      "step": 192
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.1965,
+      "step": 193
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.0837,
+      "step": 194
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.3225,
+      "step": 195
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 2.5396,
+      "step": 196
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.4783,
+      "step": 197
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.3979,
+      "step": 198
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.2863,
+      "step": 199
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.3859,
+      "step": 200
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.3307,
+      "step": 201
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.363,
+      "step": 202
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.2008,
+      "step": 203
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.4919,
+      "step": 204
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.3655,
+      "step": 205
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.2611,
+      "step": 206
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 2.3362,
+      "step": 207
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.3208,
+      "step": 208
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.4622,
+      "step": 209
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.4822,
+      "step": 210
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.314,
+      "step": 211
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.3552,
+      "step": 212
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.444,
+      "step": 213
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.5292,
+      "step": 214
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.2057,
+      "step": 215
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.1686,
+      "step": 216
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.3187,
+      "step": 217
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.282,
+      "step": 218
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 2.3346,
+      "step": 219
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.3881,
+      "step": 220
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.335,
+      "step": 221
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.3267,
+      "step": 222
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.3275,
+      "step": 223
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.3184,
+      "step": 224
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.3699,
+      "step": 225
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.3748,
+      "step": 226
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.189,
+      "step": 227
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.2434,
+      "step": 228
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.3972,
+      "step": 229
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.2656,
+      "step": 230
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 2.5007,
+      "step": 231
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.1849,
+      "step": 232
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.405,
+      "step": 233
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.4366,
+      "step": 234
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.2173,
+      "step": 235
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.3785,
+      "step": 236
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.3518,
+      "step": 237
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.1577,
+      "step": 238
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.176,
+      "step": 239
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.4232,
+      "step": 240
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.4047,
+      "step": 241
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.3771,
+      "step": 242
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 2.3892,
+      "step": 243
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.3459,
+      "step": 244
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.3494,
+      "step": 245
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.3613,
+      "step": 246
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.2981,
+      "step": 247
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.1968,
+      "step": 248
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.1989,
+      "step": 249
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.385,
+      "step": 250
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.2654,
+      "step": 251
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.2053,
+      "step": 252
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.3613,
+      "step": 253
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.3159,
+      "step": 254
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 2.2936,
+      "step": 255
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.3887,
+      "step": 256
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.3291,
+      "step": 257
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.396,
+      "step": 258
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.3242,
+      "step": 259
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.2458,
+      "step": 260
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.1858,
+      "step": 261
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.3352,
+      "step": 262
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.2957,
+      "step": 263
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.3259,
+      "step": 264
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.3422,
+      "step": 265
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.1168,
+      "step": 266
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 2.262,
+      "step": 267
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.3177,
+      "step": 268
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.2103,
+      "step": 269
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.3483,
+      "step": 270
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.4521,
+      "step": 271
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.2559,
+      "step": 272
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.2421,
+      "step": 273
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.1936,
+      "step": 274
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.1454,
+      "step": 275
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.192,
+      "step": 276
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.1375,
+      "step": 277
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.3392,
+      "step": 278
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 2.292,
+      "step": 279
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.3164,
+      "step": 280
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.4335,
+      "step": 281
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.3114,
+      "step": 282
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.3052,
+      "step": 283
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.3796,
+      "step": 284
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.4504,
+      "step": 285
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.3525,
+      "step": 286
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.2864,
+      "step": 287
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.3616,
+      "step": 288
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.3833,
+      "step": 289
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.2781,
+      "step": 290
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 2.2812,
+      "step": 291
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.5166,
+      "step": 292
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.2191,
+      "step": 293
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.2686,
+      "step": 294
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.3453,
+      "step": 295
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.4126,
+      "step": 296
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.3767,
+      "step": 297
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.3721,
+      "step": 298
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.2933,
+      "step": 299
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.0715,
+      "step": 300
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.2462,
+      "step": 301
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 2.2107,
+      "step": 302
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.2766,
+      "step": 303
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.322,
+      "step": 304
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.1987,
+      "step": 305
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.3499,
+      "step": 306
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.2788,
+      "step": 307
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.4497,
+      "step": 308
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.3535,
+      "step": 309
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.3267,
+      "step": 310
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.2974,
+      "step": 311
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.2649,
+      "step": 312
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.2965,
+      "step": 313
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 2.5212,
+      "step": 314
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.2142,
+      "step": 315
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.4559,
+      "step": 316
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.4224,
+      "step": 317
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.3911,
+      "step": 318
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.2738,
+      "step": 319
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.3777,
+      "step": 320
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.3427,
+      "step": 321
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.3491,
+      "step": 322
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.4419,
+      "step": 323
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.3616,
+      "step": 324
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.3165,
+      "step": 325
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 2.5095,
+      "step": 326
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.5312,
+      "step": 327
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.488,
+      "step": 328
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.3965,
+      "step": 329
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.2384,
+      "step": 330
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.4305,
+      "step": 331
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.4185,
+      "step": 332
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.236,
+      "step": 333
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.509,
+      "step": 334
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.4294,
+      "step": 335
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.5247,
+      "step": 336
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.2194,
+      "step": 337
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 2.3369,
+      "step": 338
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.3704,
+      "step": 339
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.1064,
+      "step": 340
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.4399,
+      "step": 341
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.4681,
+      "step": 342
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.3401,
+      "step": 343
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.1954,
+      "step": 344
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.2196,
+      "step": 345
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.3447,
+      "step": 346
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.3713,
+      "step": 347
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.2354,
+      "step": 348
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.521,
+      "step": 349
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 2.3572,
+      "step": 350
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.1414,
+      "step": 351
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.2871,
+      "step": 352
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.2443,
+      "step": 353
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.4058,
+      "step": 354
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.3345,
+      "step": 355
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.4351,
+      "step": 356
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.1862,
+      "step": 357
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.2479,
+      "step": 358
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.3582,
+      "step": 359
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.3162,
+      "step": 360
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.3345,
+      "step": 361
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 2.4045,
+      "step": 362
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.3116,
+      "step": 363
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.3158,
+      "step": 364
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.3013,
+      "step": 365
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.3345,
+      "step": 366
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.3745,
+      "step": 367
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.2908,
+      "step": 368
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.2833,
+      "step": 369
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.4097,
+      "step": 370
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.2383,
+      "step": 371
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.3862,
+      "step": 372
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.4517,
+      "step": 373
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 2.3655,
+      "step": 374
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3655,
+      "step": 375
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3616,
+      "step": 376
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3538,
+      "step": 377
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.2383,
+      "step": 378
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.4049,
+      "step": 379
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3469,
+      "step": 380
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3098,
+      "step": 381
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.6006,
+      "step": 382
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3943,
+      "step": 383
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3174,
+      "step": 384
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3457,
+      "step": 385
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.5051,
+      "step": 386
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.3582,
+      "step": 387
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.4658,
+      "step": 388
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.2195,
+      "step": 389
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.3911,
+      "step": 390
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.269,
+      "step": 391
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.3667,
+      "step": 392
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.3599,
+      "step": 393
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.3384,
+      "step": 394
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.3212,
+      "step": 395
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.4569,
+      "step": 396
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.4934,
+      "step": 397
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 2.3193,
+      "step": 398
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.3307,
+      "step": 399
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.3055,
+      "step": 400
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.2667,
+      "step": 401
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.3383,
+      "step": 402
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.0842,
+      "step": 403
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.3457,
+      "step": 404
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.5002,
+      "step": 405
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.3591,
+      "step": 406
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.3533,
+      "step": 407
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.422,
+      "step": 408
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 2.3676,
+      "step": 409
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.1697,
+      "step": 410
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.3918,
+      "step": 411
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.3489,
+      "step": 412
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.2001,
+      "step": 413
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.4946,
+      "step": 414
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.5054,
+      "step": 415
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.4373,
+      "step": 416
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.3499,
+      "step": 417
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.2653,
+      "step": 418
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.1583,
+      "step": 419
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.3044,
+      "step": 420
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 2.1425,
+      "step": 421
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.295,
+      "step": 422
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.2476,
+      "step": 423
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.2086,
+      "step": 424
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.3085,
+      "step": 425
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.2019,
+      "step": 426
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.3374,
+      "step": 427
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.3936,
+      "step": 428
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.3114,
+      "step": 429
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.3997,
+      "step": 430
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.4136,
+      "step": 431
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.2758,
+      "step": 432
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 2.1997,
+      "step": 433
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.2594,
+      "step": 434
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.3005,
+      "step": 435
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.3423,
+      "step": 436
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.4279,
+      "step": 437
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.3577,
+      "step": 438
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.2664,
+      "step": 439
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.1888,
+      "step": 440
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.286,
+      "step": 441
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.3948,
+      "step": 442
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.2546,
+      "step": 443
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.3893,
+      "step": 444
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 2.3262,
+      "step": 445
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.2455,
+      "step": 446
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.2612,
+      "step": 447
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.3744,
+      "step": 448
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.4705,
+      "step": 449
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.2452,
+      "step": 450
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.2024,
+      "step": 451
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.1718,
+      "step": 452
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 1.8889,
+      "step": 453
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.2754,
+      "step": 454
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.3568,
+      "step": 455
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.2584,
+      "step": 456
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 2.2926,
+      "step": 457
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.2955,
+      "step": 458
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.2965,
+      "step": 459
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.3854,
+      "step": 460
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.364,
+      "step": 461
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.1924,
+      "step": 462
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.2109,
+      "step": 463
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.2834,
+      "step": 464
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.4985,
+      "step": 465
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.3274,
+      "step": 466
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.2981,
+      "step": 467
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.201,
+      "step": 468
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 2.2792,
+      "step": 469
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.3503,
+      "step": 470
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.2524,
+      "step": 471
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.3373,
+      "step": 472
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.2361,
+      "step": 473
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.2913,
+      "step": 474
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.097,
+      "step": 475
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.3242,
+      "step": 476
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.2937,
+      "step": 477
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.3176,
+      "step": 478
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.2639,
+      "step": 479
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.3733,
+      "step": 480
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 2.2101,
+      "step": 481
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.3608,
+      "step": 482
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.2839,
+      "step": 483
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.2975,
+      "step": 484
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.3962,
+      "step": 485
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.26,
+      "step": 486
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.4678,
+      "step": 487
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.4189,
+      "step": 488
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.2723,
+      "step": 489
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.3385,
+      "step": 490
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.2101,
+      "step": 491
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.3672,
+      "step": 492
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 2.3762,
+      "step": 493
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.3467,
+      "step": 494
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.4207,
+      "step": 495
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.406,
+      "step": 496
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.3048,
+      "step": 497
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.1755,
+      "step": 498
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.4397,
+      "step": 499
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.4163,
+      "step": 500
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.3713,
+      "step": 501
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.2992,
+      "step": 502
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.4268,
+      "step": 503
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 2.2449,
+      "step": 504
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.1991,
+      "step": 505
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.276,
+      "step": 506
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.4667,
+      "step": 507
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.3689,
+      "step": 508
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.3092,
+      "step": 509
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.3237,
+      "step": 510
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.5165,
+      "step": 511
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.3433,
+      "step": 512
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.2793,
+      "step": 513
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.3188,
+      "step": 514
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.2644,
+      "step": 515
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 2.3296,
+      "step": 516
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.2194,
+      "step": 517
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.3608,
+      "step": 518
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.4829,
+      "step": 519
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.3435,
+      "step": 520
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.33,
+      "step": 521
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.3451,
+      "step": 522
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.2209,
+      "step": 523
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.1531,
+      "step": 524
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.4272,
+      "step": 525
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.2444,
+      "step": 526
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.446,
+      "step": 527
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 2.0909,
+      "step": 528
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.1919,
+      "step": 529
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.2784,
+      "step": 530
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.3943,
+      "step": 531
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.3364,
+      "step": 532
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.3333,
+      "step": 533
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.478,
+      "step": 534
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.3312,
+      "step": 535
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.2998,
+      "step": 536
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.377,
+      "step": 537
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.1466,
+      "step": 538
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.3286,
+      "step": 539
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 2.2478,
+      "step": 540
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.3604,
+      "step": 541
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.4155,
+      "step": 542
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.2784,
+      "step": 543
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.3737,
+      "step": 544
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.0891,
+      "step": 545
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.3732,
+      "step": 546
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.0795,
+      "step": 547
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.2083,
+      "step": 548
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.2194,
+      "step": 549
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.3892,
+      "step": 550
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.4294,
+      "step": 551
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 2.1777,
+      "step": 552
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.4517,
+      "step": 553
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.3463,
+      "step": 554
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.2588,
+      "step": 555
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.1049,
+      "step": 556
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.3221,
+      "step": 557
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.2915,
+      "step": 558
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.1635,
+      "step": 559
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.2549,
+      "step": 560
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.3236,
+      "step": 561
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.1913,
+      "step": 562
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.194,
+      "step": 563
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 2.4624,
+      "step": 564
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.3615,
+      "step": 565
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.2692,
+      "step": 566
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.302,
+      "step": 567
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.1394,
+      "step": 568
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.2216,
+      "step": 569
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.3376,
+      "step": 570
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.2515,
+      "step": 571
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.3853,
+      "step": 572
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.3322,
+      "step": 573
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.2834,
+      "step": 574
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.171,
+      "step": 575
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 2.415,
+      "step": 576
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.2532,
+      "step": 577
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.1749,
+      "step": 578
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.2074,
+      "step": 579
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.2367,
+      "step": 580
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.5156,
+      "step": 581
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.501,
+      "step": 582
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.1338,
+      "step": 583
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.2324,
+      "step": 584
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.449,
+      "step": 585
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.2458,
+      "step": 586
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.2714,
+      "step": 587
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 2.2799,
+      "step": 588
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2622,
+      "step": 589
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.1775,
+      "step": 590
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2734,
+      "step": 591
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2943,
+      "step": 592
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.0663,
+      "step": 593
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2756,
+      "step": 594
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2773,
+      "step": 595
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.3901,
+      "step": 596
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2953,
+      "step": 597
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.3966,
+      "step": 598
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.1602,
+      "step": 599
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2133,
+      "step": 600
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.3267,
+      "step": 601
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2681,
+      "step": 602
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.3875,
+      "step": 603
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2686,
+      "step": 604
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2135,
+      "step": 605
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2495,
+      "step": 606
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2191,
+      "step": 607
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2448,
+      "step": 608
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.4113,
+      "step": 609
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2657,
+      "step": 610
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 2.2522,
+      "step": 611
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.2059,
+      "step": 612
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.218,
+      "step": 613
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.3721,
+      "step": 614
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.2502,
+      "step": 615
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.4062,
+      "step": 616
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.2446,
+      "step": 617
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.1578,
+      "step": 618
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.3721,
+      "step": 619
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.3496,
+      "step": 620
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.3186,
+      "step": 621
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.3555,
+      "step": 622
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 2.5234,
+      "step": 623
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.2859,
+      "step": 624
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.1616,
+      "step": 625
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.318,
+      "step": 626
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.4158,
+      "step": 627
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.2285,
+      "step": 628
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.183,
+      "step": 629
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.5471,
+      "step": 630
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.2767,
+      "step": 631
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.3423,
+      "step": 632
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.1931,
+      "step": 633
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.3174,
+      "step": 634
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 2.3783,
+      "step": 635
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.0183,
+      "step": 636
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3057,
+      "step": 637
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3179,
+      "step": 638
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3076,
+      "step": 639
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3433,
+      "step": 640
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3546,
+      "step": 641
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3936,
+      "step": 642
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.2456,
+      "step": 643
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3695,
+      "step": 644
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.2272,
+      "step": 645
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.3754,
+      "step": 646
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 2.4569,
+      "step": 647
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.5757,
+      "step": 648
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.3358,
+      "step": 649
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.1724,
+      "step": 650
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.3162,
+      "step": 651
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.0896,
+      "step": 652
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.2563,
+      "step": 653
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.2859,
+      "step": 654
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.407,
+      "step": 655
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.3098,
+      "step": 656
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.3785,
+      "step": 657
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.2231,
+      "step": 658
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 2.2656,
+      "step": 659
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.4192,
+      "step": 660
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.4021,
+      "step": 661
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.1509,
+      "step": 662
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.0681,
+      "step": 663
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.3452,
+      "step": 664
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.2079,
+      "step": 665
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.3108,
+      "step": 666
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.5054,
+      "step": 667
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.1946,
+      "step": 668
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.1594,
+      "step": 669
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.1561,
+      "step": 670
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 2.2847,
+      "step": 671
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.272,
+      "step": 672
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.3601,
+      "step": 673
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.149,
+      "step": 674
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.3762,
+      "step": 675
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.2578,
+      "step": 676
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.3818,
+      "step": 677
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.2131,
+      "step": 678
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.2686,
+      "step": 679
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.4736,
+      "step": 680
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.2466,
+      "step": 681
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.2479,
+      "step": 682
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 2.3712,
+      "step": 683
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.2623,
+      "step": 684
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.2332,
+      "step": 685
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.3524,
+      "step": 686
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.3604,
+      "step": 687
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.3427,
+      "step": 688
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.1012,
+      "step": 689
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.3156,
+      "step": 690
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.2996,
+      "step": 691
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.2499,
+      "step": 692
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.2836,
+      "step": 693
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.2133,
+      "step": 694
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 2.4011,
+      "step": 695
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.0957,
+      "step": 696
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.2141,
+      "step": 697
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.4292,
+      "step": 698
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.1443,
+      "step": 699
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.319,
+      "step": 700
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.1442,
+      "step": 701
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.3586,
+      "step": 702
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.196,
+      "step": 703
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.2236,
+      "step": 704
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.2445,
+      "step": 705
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 2.3231,
+      "step": 706
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.2097,
+      "step": 707
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.2391,
+      "step": 708
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.257,
+      "step": 709
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.2209,
+      "step": 710
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.2224,
+      "step": 711
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.1322,
+      "step": 712
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.1532,
+      "step": 713
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.2419,
+      "step": 714
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.215,
+      "step": 715
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.3076,
+      "step": 716
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.3958,
+      "step": 717
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 2.3296,
+      "step": 718
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.4495,
+      "step": 719
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.332,
+      "step": 720
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.3201,
+      "step": 721
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.2604,
+      "step": 722
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.323,
+      "step": 723
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.0908,
+      "step": 724
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.3885,
+      "step": 725
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.3027,
+      "step": 726
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.4523,
+      "step": 727
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.2412,
+      "step": 728
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.3002,
+      "step": 729
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 2.3855,
+      "step": 730
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.4489,
+      "step": 731
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.3313,
+      "step": 732
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.2338,
+      "step": 733
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.2643,
+      "step": 734
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.2771,
+      "step": 735
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.1143,
+      "step": 736
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.3489,
+      "step": 737
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.1638,
+      "step": 738
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.4102,
+      "step": 739
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.3492,
+      "step": 740
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.3265,
+      "step": 741
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 2.2836,
+      "step": 742
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.4073,
+      "step": 743
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.2771,
+      "step": 744
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.3496,
+      "step": 745
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.3823,
+      "step": 746
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.3186,
+      "step": 747
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.2688,
+      "step": 748
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.2229,
+      "step": 749
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.298,
+      "step": 750
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.3521,
+      "step": 751
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.5012,
+      "step": 752
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.1053,
+      "step": 753
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 2.3867,
+      "step": 754
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.3063,
+      "step": 755
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.282,
+      "step": 756
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.2903,
+      "step": 757
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.3392,
+      "step": 758
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.2308,
+      "step": 759
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.3164,
+      "step": 760
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.3813,
+      "step": 761
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.4695,
+      "step": 762
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.2874,
+      "step": 763
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.2224,
+      "step": 764
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.2678,
+      "step": 765
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 2.35,
+      "step": 766
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.3552,
+      "step": 767
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.283,
+      "step": 768
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.3071,
+      "step": 769
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.3927,
+      "step": 770
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.1398,
+      "step": 771
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.4207,
+      "step": 772
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.2115,
+      "step": 773
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.3557,
+      "step": 774
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.3668,
+      "step": 775
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.2931,
+      "step": 776
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.189,
+      "step": 777
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 2.3682,
+      "step": 778
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.3931,
+      "step": 779
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.3817,
+      "step": 780
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.3748,
+      "step": 781
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.2474,
+      "step": 782
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.3446,
+      "step": 783
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.2495,
+      "step": 784
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.3745,
+      "step": 785
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.2609,
+      "step": 786
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.3351,
+      "step": 787
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.3806,
+      "step": 788
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.1624,
+      "step": 789
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 2.1378,
+      "step": 790
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.1924,
+      "step": 791
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.1378,
+      "step": 792
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.176,
+      "step": 793
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.2009,
+      "step": 794
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.2955,
+      "step": 795
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 1.9546,
+      "step": 796
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.3124,
+      "step": 797
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.364,
+      "step": 798
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.2609,
+      "step": 799
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.3723,
+      "step": 800
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 2.4556,
+      "step": 801
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.224,
+      "step": 802
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.1948,
+      "step": 803
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.1469,
+      "step": 804
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.3168,
+      "step": 805
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.4487,
+      "step": 806
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.1589,
+      "step": 807
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.3718,
+      "step": 808
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.4338,
+      "step": 809
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.1456,
+      "step": 810
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.2983,
+      "step": 811
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.2098,
+      "step": 812
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 2.3372,
+      "step": 813
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.2562,
+      "step": 814
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.1631,
+      "step": 815
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.2562,
+      "step": 816
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.1465,
+      "step": 817
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.3113,
+      "step": 818
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.3125,
+      "step": 819
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.2483,
+      "step": 820
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.2314,
+      "step": 821
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.4141,
+      "step": 822
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.3204,
+      "step": 823
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.3163,
+      "step": 824
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 2.1493,
+      "step": 825
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.2576,
+      "step": 826
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.239,
+      "step": 827
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.353,
+      "step": 828
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.1693,
+      "step": 829
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.087,
+      "step": 830
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.3022,
+      "step": 831
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.302,
+      "step": 832
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.2426,
+      "step": 833
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.3514,
+      "step": 834
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.2394,
+      "step": 835
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.4016,
+      "step": 836
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 2.2247,
+      "step": 837
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.3806,
+      "step": 838
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.1702,
+      "step": 839
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.2688,
+      "step": 840
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.2903,
+      "step": 841
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.3085,
+      "step": 842
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.3633,
+      "step": 843
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.1893,
+      "step": 844
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.2922,
+      "step": 845
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.2444,
+      "step": 846
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.3173,
+      "step": 847
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.2429,
+      "step": 848
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 2.3679,
+      "step": 849
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.2257,
+      "step": 850
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.3265,
+      "step": 851
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.1279,
+      "step": 852
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.3496,
+      "step": 853
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.2643,
+      "step": 854
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.3385,
+      "step": 855
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.264,
+      "step": 856
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.0507,
+      "step": 857
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.1364,
+      "step": 858
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.3486,
+      "step": 859
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.3557,
+      "step": 860
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 2.3137,
+      "step": 861
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.3318,
+      "step": 862
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.4092,
+      "step": 863
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.2642,
+      "step": 864
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.3483,
+      "step": 865
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.2908,
+      "step": 866
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.4294,
+      "step": 867
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.4043,
+      "step": 868
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.4182,
+      "step": 869
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.3607,
+      "step": 870
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.444,
+      "step": 871
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.286,
+      "step": 872
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 2.2229,
+      "step": 873
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.1622,
+      "step": 874
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.3403,
+      "step": 875
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.3408,
+      "step": 876
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.1488,
+      "step": 877
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.2775,
+      "step": 878
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.2701,
+      "step": 879
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.374,
+      "step": 880
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.4299,
+      "step": 881
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.3677,
+      "step": 882
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.3252,
+      "step": 883
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.2539,
+      "step": 884
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 2.0759,
+      "step": 885
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.4619,
+      "step": 886
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.3337,
+      "step": 887
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.4041,
+      "step": 888
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.3657,
+      "step": 889
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.1604,
+      "step": 890
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.2185,
+      "step": 891
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.3638,
+      "step": 892
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.2451,
+      "step": 893
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.2507,
+      "step": 894
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.252,
+      "step": 895
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.2349,
+      "step": 896
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 2.3,
+      "step": 897
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.1538,
+      "step": 898
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.3563,
+      "step": 899
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.4106,
+      "step": 900
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.3315,
+      "step": 901
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.1498,
+      "step": 902
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.3896,
+      "step": 903
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.1454,
+      "step": 904
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.2047,
+      "step": 905
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.4373,
+      "step": 906
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.2754,
+      "step": 907
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 2.2401,
+      "step": 908
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.2321,
+      "step": 909
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.295,
+      "step": 910
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.3613,
+      "step": 911
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.3862,
+      "step": 912
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.2732,
+      "step": 913
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.1107,
+      "step": 914
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.3704,
+      "step": 915
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.1769,
+      "step": 916
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.2924,
+      "step": 917
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.2174,
+      "step": 918
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.2549,
+      "step": 919
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 2.2839,
+      "step": 920
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.182,
+      "step": 921
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.3416,
+      "step": 922
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.2506,
+      "step": 923
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.1416,
+      "step": 924
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.2079,
+      "step": 925
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.3428,
+      "step": 926
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.3228,
+      "step": 927
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.3235,
+      "step": 928
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.2609,
+      "step": 929
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.3467,
+      "step": 930
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.3949,
+      "step": 931
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 2.3855,
+      "step": 932
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.313,
+      "step": 933
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.3362,
+      "step": 934
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.2506,
+      "step": 935
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.3218,
+      "step": 936
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.224,
+      "step": 937
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.2747,
+      "step": 938
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.4739,
+      "step": 939
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.3003,
+      "step": 940
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.2889,
+      "step": 941
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.3801,
+      "step": 942
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.2177,
+      "step": 943
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 2.2465,
+      "step": 944
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.2833,
+      "step": 945
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.4143,
+      "step": 946
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.2596,
+      "step": 947
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.2372,
+      "step": 948
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.5713,
+      "step": 949
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.283,
+      "step": 950
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.2523,
+      "step": 951
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.3534,
+      "step": 952
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.1779,
+      "step": 953
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.1659,
+      "step": 954
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.4424,
+      "step": 955
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 2.2162,
+      "step": 956
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.1028,
+      "step": 957
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.2529,
+      "step": 958
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.2043,
+      "step": 959
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.2157,
+      "step": 960
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.1705,
+      "step": 961
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.1287,
+      "step": 962
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.3091,
+      "step": 963
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.1337,
+      "step": 964
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.2189,
+      "step": 965
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.2202,
+      "step": 966
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.2535,
+      "step": 967
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 2.3262,
+      "step": 968
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.3403,
+      "step": 969
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.2163,
+      "step": 970
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.1024,
+      "step": 971
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.3712,
+      "step": 972
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.4084,
+      "step": 973
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.2809,
+      "step": 974
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.3275,
+      "step": 975
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.2328,
+      "step": 976
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.3545,
+      "step": 977
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.2781,
+      "step": 978
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.1577,
+      "step": 979
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 2.2452,
+      "step": 980
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.2096,
+      "step": 981
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.3462,
+      "step": 982
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.3002,
+      "step": 983
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.3538,
+      "step": 984
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.2612,
+      "step": 985
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.2605,
+      "step": 986
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.1879,
+      "step": 987
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.2686,
+      "step": 988
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.3109,
+      "step": 989
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.3866,
+      "step": 990
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.2834,
+      "step": 991
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 2.3059,
+      "step": 992
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.1727,
+      "step": 993
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.3635,
+      "step": 994
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.2708,
+      "step": 995
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.2556,
+      "step": 996
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.2397,
+      "step": 997
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.2693,
+      "step": 998
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.2034,
+      "step": 999
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 2.222,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 1188,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "total_flos": 1.0500456841216e+18,
+  "trial_name": null,
+  "trial_params": null
+}