diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,4903 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.9840490797546013,
+  "eval_steps": 500,
+  "global_step": 814,
+  "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": 1.317,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.000000000000001e-08,
+      "loss": 1.1248,
+      "step": 2
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2000000000000002e-07,
+      "loss": 1.0845,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8e-07,
+      "loss": 1.0864,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4000000000000003e-07,
+      "loss": 1.0715,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0000000000000004e-07,
+      "loss": 1.0756,
+      "step": 6
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6e-07,
+      "loss": 0.9961,
+      "step": 7
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2000000000000006e-07,
+      "loss": 0.9972,
+      "step": 8
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.800000000000001e-07,
+      "loss": 1.0152,
+      "step": 9
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.4e-07,
+      "loss": 1.0149,
+      "step": 10
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.000000000000001e-07,
+      "loss": 1.0061,
+      "step": 11
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.6e-07,
+      "loss": 0.9948,
+      "step": 12
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.2e-07,
+      "loss": 0.9787,
+      "step": 13
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.8e-07,
+      "loss": 0.9936,
+      "step": 14
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.400000000000001e-07,
+      "loss": 0.9496,
+      "step": 15
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9e-07,
+      "loss": 0.9368,
+      "step": 16
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.600000000000001e-07,
+      "loss": 0.9234,
+      "step": 17
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.0200000000000002e-06,
+      "loss": 0.9324,
+      "step": 18
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.08e-06,
+      "loss": 0.9183,
+      "step": 19
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.14e-06,
+      "loss": 0.9273,
+      "step": 20
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 0.9165,
+      "step": 21
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.26e-06,
+      "loss": 0.9161,
+      "step": 22
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.32e-06,
+      "loss": 0.8887,
+      "step": 23
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.3800000000000001e-06,
+      "loss": 0.9093,
+      "step": 24
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.44e-06,
+      "loss": 0.8891,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.5e-06,
+      "loss": 0.8893,
+      "step": 26
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.56e-06,
+      "loss": 0.8838,
+      "step": 27
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.6200000000000002e-06,
+      "loss": 0.8701,
+      "step": 28
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.6800000000000002e-06,
+      "loss": 0.8748,
+      "step": 29
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.7399999999999999e-06,
+      "loss": 0.8831,
+      "step": 30
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.8e-06,
+      "loss": 0.8656,
+      "step": 31
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.86e-06,
+      "loss": 0.8655,
+      "step": 32
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9200000000000003e-06,
+      "loss": 0.8681,
+      "step": 33
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.98e-06,
+      "loss": 0.8682,
+      "step": 34
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 2.0400000000000004e-06,
+      "loss": 0.8618,
+      "step": 35
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 2.1e-06,
+      "loss": 0.874,
+      "step": 36
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 2.16e-06,
+      "loss": 0.8305,
+      "step": 37
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 2.22e-06,
+      "loss": 0.8392,
+      "step": 38
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.28e-06,
+      "loss": 0.8441,
+      "step": 39
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.34e-06,
+      "loss": 0.829,
+      "step": 40
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 0.8422,
+      "step": 41
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.4599999999999997e-06,
+      "loss": 0.8295,
+      "step": 42
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.52e-06,
+      "loss": 0.8349,
+      "step": 43
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.58e-06,
+      "loss": 0.8204,
+      "step": 44
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.64e-06,
+      "loss": 0.8344,
+      "step": 45
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.7e-06,
+      "loss": 0.8121,
+      "step": 46
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 2.7600000000000003e-06,
+      "loss": 0.817,
+      "step": 47
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 2.82e-06,
+      "loss": 0.8095,
+      "step": 48
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 2.88e-06,
+      "loss": 0.7887,
+      "step": 49
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 2.9400000000000002e-06,
+      "loss": 0.7916,
+      "step": 50
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 3e-06,
+      "loss": 0.8136,
+      "step": 51
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 3.06e-06,
+      "loss": 0.8269,
+      "step": 52
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 3.12e-06,
+      "loss": 0.788,
+      "step": 53
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 3.18e-06,
+      "loss": 0.8035,
+      "step": 54
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 3.2400000000000003e-06,
+      "loss": 0.8079,
+      "step": 55
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 3.3e-06,
+      "loss": 0.8097,
+      "step": 56
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 3.3600000000000004e-06,
+      "loss": 0.825,
+      "step": 57
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 3.42e-06,
+      "loss": 0.8157,
+      "step": 58
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 3.4799999999999997e-06,
+      "loss": 0.8158,
+      "step": 59
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 3.54e-06,
+      "loss": 0.8033,
+      "step": 60
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 3.6e-06,
+      "loss": 0.7999,
+      "step": 61
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 3.66e-06,
+      "loss": 0.7915,
+      "step": 62
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 3.72e-06,
+      "loss": 0.8036,
+      "step": 63
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 3.7800000000000002e-06,
+      "loss": 0.7854,
+      "step": 64
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 3.8400000000000005e-06,
+      "loss": 0.7872,
+      "step": 65
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 3.9e-06,
+      "loss": 0.8166,
+      "step": 66
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 3.96e-06,
+      "loss": 0.7972,
+      "step": 67
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.0200000000000005e-06,
+      "loss": 0.7959,
+      "step": 68
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.080000000000001e-06,
+      "loss": 0.7973,
+      "step": 69
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.14e-06,
+      "loss": 0.7606,
+      "step": 70
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.2e-06,
+      "loss": 0.7859,
+      "step": 71
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.26e-06,
+      "loss": 0.798,
+      "step": 72
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.32e-06,
+      "loss": 0.7739,
+      "step": 73
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.38e-06,
+      "loss": 0.7898,
+      "step": 74
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.44e-06,
+      "loss": 0.7995,
+      "step": 75
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5e-06,
+      "loss": 0.7688,
+      "step": 76
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.56e-06,
+      "loss": 0.7937,
+      "step": 77
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.62e-06,
+      "loss": 0.7708,
+      "step": 78
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.68e-06,
+      "loss": 0.7963,
+      "step": 79
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.74e-06,
+      "loss": 0.7817,
+      "step": 80
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 0.752,
+      "step": 81
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.86e-06,
+      "loss": 0.7891,
+      "step": 82
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.9199999999999995e-06,
+      "loss": 0.7742,
+      "step": 83
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.98e-06,
+      "loss": 0.7814,
+      "step": 84
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 5.04e-06,
+      "loss": 0.7851,
+      "step": 85
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 5.1e-06,
+      "loss": 0.7643,
+      "step": 86
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 5.16e-06,
+      "loss": 0.7853,
+      "step": 87
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 5.22e-06,
+      "loss": 0.7654,
+      "step": 88
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 5.28e-06,
+      "loss": 0.7698,
+      "step": 89
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 5.3400000000000005e-06,
+      "loss": 0.764,
+      "step": 90
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 5.4e-06,
+      "loss": 0.7493,
+      "step": 91
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 5.46e-06,
+      "loss": 0.7841,
+      "step": 92
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 5.5200000000000005e-06,
+      "loss": 0.7774,
+      "step": 93
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 5.580000000000001e-06,
+      "loss": 0.7793,
+      "step": 94
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 5.64e-06,
+      "loss": 0.7687,
+      "step": 95
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 5.7e-06,
+      "loss": 0.7697,
+      "step": 96
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 5.76e-06,
+      "loss": 0.7685,
+      "step": 97
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 5.82e-06,
+      "loss": 0.7714,
+      "step": 98
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 5.8800000000000005e-06,
+      "loss": 0.7744,
+      "step": 99
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 5.94e-06,
+      "loss": 0.7819,
+      "step": 100
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 6e-06,
+      "loss": 0.7874,
+      "step": 101
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 5.991596638655462e-06,
+      "loss": 0.7769,
+      "step": 102
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 5.9831932773109244e-06,
+      "loss": 0.7852,
+      "step": 103
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 5.9747899159663866e-06,
+      "loss": 0.7852,
+      "step": 104
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 5.966386554621849e-06,
+      "loss": 0.7518,
+      "step": 105
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 5.957983193277311e-06,
+      "loss": 0.7671,
+      "step": 106
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 5.949579831932774e-06,
+      "loss": 0.7748,
+      "step": 107
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 5.941176470588236e-06,
+      "loss": 0.7515,
+      "step": 108
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 5.932773109243698e-06,
+      "loss": 0.7456,
+      "step": 109
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 5.92436974789916e-06,
+      "loss": 0.7801,
+      "step": 110
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 5.915966386554622e-06,
+      "loss": 0.755,
+      "step": 111
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 5.9075630252100845e-06,
+      "loss": 0.7659,
+      "step": 112
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 5.899159663865546e-06,
+      "loss": 0.7668,
+      "step": 113
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 5.890756302521008e-06,
+      "loss": 0.7369,
+      "step": 114
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 5.88235294117647e-06,
+      "loss": 0.7754,
+      "step": 115
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 5.873949579831933e-06,
+      "loss": 0.7463,
+      "step": 116
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 5.865546218487395e-06,
+      "loss": 0.7554,
+      "step": 117
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 5.857142857142857e-06,
+      "loss": 0.7728,
+      "step": 118
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 5.8487394957983195e-06,
+      "loss": 0.7504,
+      "step": 119
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 5.840336134453782e-06,
+      "loss": 0.7919,
+      "step": 120
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 5.831932773109244e-06,
+      "loss": 0.7601,
+      "step": 121
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 5.823529411764706e-06,
+      "loss": 0.7127,
+      "step": 122
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 5.815126050420168e-06,
+      "loss": 0.7609,
+      "step": 123
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 5.80672268907563e-06,
+      "loss": 0.7695,
+      "step": 124
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 5.798319327731092e-06,
+      "loss": 0.7576,
+      "step": 125
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 5.7899159663865544e-06,
+      "loss": 0.7576,
+      "step": 126
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 5.781512605042017e-06,
+      "loss": 0.7702,
+      "step": 127
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 5.7731092436974796e-06,
+      "loss": 0.7298,
+      "step": 128
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 5.764705882352942e-06,
+      "loss": 0.7614,
+      "step": 129
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 5.756302521008404e-06,
+      "loss": 0.7551,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 5.747899159663866e-06,
+      "loss": 0.7866,
+      "step": 131
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 5.739495798319328e-06,
+      "loss": 0.7705,
+      "step": 132
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 5.73109243697479e-06,
+      "loss": 0.765,
+      "step": 133
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 5.722689075630252e-06,
+      "loss": 0.765,
+      "step": 134
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 5.7142857142857145e-06,
+      "loss": 0.7346,
+      "step": 135
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 5.705882352941177e-06,
+      "loss": 0.771,
+      "step": 136
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 5.697478991596639e-06,
+      "loss": 0.7689,
+      "step": 137
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 5.689075630252101e-06,
+      "loss": 0.7632,
+      "step": 138
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 5.680672268907563e-06,
+      "loss": 0.7282,
+      "step": 139
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 5.672268907563025e-06,
+      "loss": 0.7497,
+      "step": 140
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 5.663865546218487e-06,
+      "loss": 0.7766,
+      "step": 141
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 5.6554621848739495e-06,
+      "loss": 0.7348,
+      "step": 142
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 5.647058823529412e-06,
+      "loss": 0.776,
+      "step": 143
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 5.638655462184874e-06,
+      "loss": 0.7627,
+      "step": 144
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 5.630252100840336e-06,
+      "loss": 0.7495,
+      "step": 145
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 5.621848739495799e-06,
+      "loss": 0.7636,
+      "step": 146
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 5.613445378151261e-06,
+      "loss": 0.778,
+      "step": 147
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 5.605042016806723e-06,
+      "loss": 0.7537,
+      "step": 148
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 5.596638655462185e-06,
+      "loss": 0.7634,
+      "step": 149
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 5.588235294117647e-06,
+      "loss": 0.7456,
+      "step": 150
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 5.5798319327731096e-06,
+      "loss": 0.7694,
+      "step": 151
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 5.571428571428572e-06,
+      "loss": 0.7776,
+      "step": 152
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 5.563025210084034e-06,
+      "loss": 0.7586,
+      "step": 153
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 5.554621848739496e-06,
+      "loss": 0.726,
+      "step": 154
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 5.546218487394958e-06,
+      "loss": 0.7561,
+      "step": 155
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 5.53781512605042e-06,
+      "loss": 0.7481,
+      "step": 156
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 5.529411764705882e-06,
+      "loss": 0.7465,
+      "step": 157
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 5.5210084033613445e-06,
+      "loss": 0.7681,
+      "step": 158
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 5.512605042016807e-06,
+      "loss": 0.7355,
+      "step": 159
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 5.504201680672269e-06,
+      "loss": 0.7753,
+      "step": 160
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 5.495798319327731e-06,
+      "loss": 0.7721,
+      "step": 161
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 5.487394957983193e-06,
+      "loss": 0.7413,
+      "step": 162
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 5.478991596638655e-06,
+      "loss": 0.7471,
+      "step": 163
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 5.470588235294117e-06,
+      "loss": 0.7569,
+      "step": 164
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 5.4621848739495795e-06,
+      "loss": 0.7632,
+      "step": 165
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 5.4537815126050425e-06,
+      "loss": 0.7278,
+      "step": 166
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 5.445378151260505e-06,
+      "loss": 0.7726,
+      "step": 167
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 5.436974789915967e-06,
+      "loss": 0.7256,
+      "step": 168
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 5.428571428571429e-06,
+      "loss": 0.7721,
+      "step": 169
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.420168067226891e-06,
+      "loss": 0.7643,
+      "step": 170
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.411764705882353e-06,
+      "loss": 0.7544,
+      "step": 171
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.403361344537815e-06,
+      "loss": 0.7246,
+      "step": 172
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.394957983193277e-06,
+      "loss": 0.7754,
+      "step": 173
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 5.3865546218487396e-06,
+      "loss": 0.7644,
+      "step": 174
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 5.378151260504202e-06,
+      "loss": 0.7507,
+      "step": 175
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 5.369747899159664e-06,
+      "loss": 0.741,
+      "step": 176
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 5.361344537815127e-06,
+      "loss": 0.72,
+      "step": 177
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 5.352941176470589e-06,
+      "loss": 0.75,
+      "step": 178
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 5.344537815126051e-06,
+      "loss": 0.7542,
+      "step": 179
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 5.336134453781512e-06,
+      "loss": 0.7606,
+      "step": 180
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 5.3277310924369745e-06,
+      "loss": 0.7525,
+      "step": 181
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.319327731092437e-06,
+      "loss": 0.7413,
+      "step": 182
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.310924369747899e-06,
+      "loss": 0.7422,
+      "step": 183
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.302521008403361e-06,
+      "loss": 0.7243,
+      "step": 184
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.294117647058824e-06,
+      "loss": 0.7377,
+      "step": 185
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.285714285714286e-06,
+      "loss": 0.7262,
+      "step": 186
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.277310924369748e-06,
+      "loss": 0.7349,
+      "step": 187
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.26890756302521e-06,
+      "loss": 0.7409,
+      "step": 188
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.2605042016806725e-06,
+      "loss": 0.7494,
+      "step": 189
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.252100840336135e-06,
+      "loss": 0.7457,
+      "step": 190
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.243697478991597e-06,
+      "loss": 0.7596,
+      "step": 191
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.235294117647059e-06,
+      "loss": 0.726,
+      "step": 192
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.226890756302521e-06,
+      "loss": 0.7229,
+      "step": 193
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.218487394957983e-06,
+      "loss": 0.7744,
+      "step": 194
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.210084033613445e-06,
+      "loss": 0.7112,
+      "step": 195
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.201680672268908e-06,
+      "loss": 0.7394,
+      "step": 196
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.19327731092437e-06,
+      "loss": 0.7422,
+      "step": 197
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.1848739495798325e-06,
+      "loss": 0.7396,
+      "step": 198
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.176470588235295e-06,
+      "loss": 0.7332,
+      "step": 199
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.168067226890757e-06,
+      "loss": 0.7368,
+      "step": 200
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.159663865546219e-06,
+      "loss": 0.7457,
+      "step": 201
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.15126050420168e-06,
+      "loss": 0.734,
+      "step": 202
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.142857142857142e-06,
+      "loss": 0.7245,
+      "step": 203
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.1344537815126045e-06,
+      "loss": 0.7477,
+      "step": 204
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.1260504201680675e-06,
+      "loss": 0.7553,
+      "step": 205
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.11764705882353e-06,
+      "loss": 0.7347,
+      "step": 206
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.109243697478992e-06,
+      "loss": 0.7378,
+      "step": 207
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.100840336134454e-06,
+      "loss": 0.7463,
+      "step": 208
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.092436974789916e-06,
+      "loss": 0.7439,
+      "step": 209
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.084033613445378e-06,
+      "loss": 0.7549,
+      "step": 210
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.07563025210084e-06,
+      "loss": 0.7304,
+      "step": 211
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.0672268907563025e-06,
+      "loss": 0.7288,
+      "step": 212
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.058823529411765e-06,
+      "loss": 0.7473,
+      "step": 213
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 5.050420168067227e-06,
+      "loss": 0.7537,
+      "step": 214
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 5.042016806722689e-06,
+      "loss": 0.7504,
+      "step": 215
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 5.033613445378152e-06,
+      "loss": 0.7619,
+      "step": 216
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 5.025210084033614e-06,
+      "loss": 0.7302,
+      "step": 217
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 5.016806722689076e-06,
+      "loss": 0.7483,
+      "step": 218
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 5.008403361344538e-06,
+      "loss": 0.7385,
+      "step": 219
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 5e-06,
+      "loss": 0.7463,
+      "step": 220
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.9915966386554625e-06,
+      "loss": 0.7225,
+      "step": 221
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.983193277310925e-06,
+      "loss": 0.7325,
+      "step": 222
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.974789915966387e-06,
+      "loss": 0.7461,
+      "step": 223
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.966386554621849e-06,
+      "loss": 0.72,
+      "step": 224
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.957983193277311e-06,
+      "loss": 0.7513,
+      "step": 225
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.949579831932773e-06,
+      "loss": 0.7179,
+      "step": 226
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.941176470588235e-06,
+      "loss": 0.7451,
+      "step": 227
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.9327731092436975e-06,
+      "loss": 0.7402,
+      "step": 228
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.92436974789916e-06,
+      "loss": 0.758,
+      "step": 229
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.915966386554622e-06,
+      "loss": 0.7291,
+      "step": 230
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.907563025210084e-06,
+      "loss": 0.7653,
+      "step": 231
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.899159663865546e-06,
+      "loss": 0.7223,
+      "step": 232
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.890756302521008e-06,
+      "loss": 0.7531,
+      "step": 233
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.88235294117647e-06,
+      "loss": 0.7324,
+      "step": 234
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.873949579831933e-06,
+      "loss": 0.7454,
+      "step": 235
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.8655462184873955e-06,
+      "loss": 0.7254,
+      "step": 236
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.857142857142858e-06,
+      "loss": 0.7589,
+      "step": 237
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.84873949579832e-06,
+      "loss": 0.7408,
+      "step": 238
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.840336134453782e-06,
+      "loss": 0.7279,
+      "step": 239
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.831932773109244e-06,
+      "loss": 0.7654,
+      "step": 240
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.823529411764706e-06,
+      "loss": 0.732,
+      "step": 241
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.815126050420168e-06,
+      "loss": 0.7226,
+      "step": 242
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.80672268907563e-06,
+      "loss": 0.7326,
+      "step": 243
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.7983193277310925e-06,
+      "loss": 0.7423,
+      "step": 244
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.789915966386555e-06,
+      "loss": 0.7442,
+      "step": 245
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.781512605042018e-06,
+      "loss": 0.7259,
+      "step": 246
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.773109243697479e-06,
+      "loss": 0.7485,
+      "step": 247
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.764705882352941e-06,
+      "loss": 0.7196,
+      "step": 248
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.756302521008403e-06,
+      "loss": 0.7476,
+      "step": 249
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.747899159663865e-06,
+      "loss": 0.7238,
+      "step": 250
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.7394957983193275e-06,
+      "loss": 0.7322,
+      "step": 251
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.73109243697479e-06,
+      "loss": 0.7438,
+      "step": 252
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.722689075630252e-06,
+      "loss": 0.7517,
+      "step": 253
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.714285714285714e-06,
+      "loss": 0.7327,
+      "step": 254
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.705882352941177e-06,
+      "loss": 0.7248,
+      "step": 255
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.697478991596639e-06,
+      "loss": 0.7339,
+      "step": 256
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.689075630252101e-06,
+      "loss": 0.7338,
+      "step": 257
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.680672268907563e-06,
+      "loss": 0.7384,
+      "step": 258
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.6722689075630255e-06,
+      "loss": 0.7279,
+      "step": 259
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.663865546218488e-06,
+      "loss": 0.7456,
+      "step": 260
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.65546218487395e-06,
+      "loss": 0.7599,
+      "step": 261
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.647058823529412e-06,
+      "loss": 0.7286,
+      "step": 262
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.638655462184874e-06,
+      "loss": 0.7471,
+      "step": 263
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.630252100840336e-06,
+      "loss": 0.7443,
+      "step": 264
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.621848739495798e-06,
+      "loss": 0.7352,
+      "step": 265
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.613445378151261e-06,
+      "loss": 0.7536,
+      "step": 266
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.605042016806723e-06,
+      "loss": 0.7499,
+      "step": 267
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.5966386554621855e-06,
+      "loss": 0.7519,
+      "step": 268
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.588235294117647e-06,
+      "loss": 0.7293,
+      "step": 269
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.579831932773109e-06,
+      "loss": 0.7201,
+      "step": 270
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.571428571428571e-06,
+      "loss": 0.7494,
+      "step": 271
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.563025210084033e-06,
+      "loss": 0.7146,
+      "step": 272
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.554621848739495e-06,
+      "loss": 0.7421,
+      "step": 273
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.5462184873949575e-06,
+      "loss": 0.71,
+      "step": 274
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.5378151260504205e-06,
+      "loss": 0.7479,
+      "step": 275
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.529411764705883e-06,
+      "loss": 0.7583,
+      "step": 276
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.521008403361345e-06,
+      "loss": 0.7293,
+      "step": 277
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.512605042016807e-06,
+      "loss": 0.7295,
+      "step": 278
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.504201680672269e-06,
+      "loss": 0.7468,
+      "step": 279
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.495798319327731e-06,
+      "loss": 0.744,
+      "step": 280
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.487394957983193e-06,
+      "loss": 0.7154,
+      "step": 281
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.4789915966386555e-06,
+      "loss": 0.7305,
+      "step": 282
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.470588235294118e-06,
+      "loss": 0.736,
+      "step": 283
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.46218487394958e-06,
+      "loss": 0.7375,
+      "step": 284
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.453781512605043e-06,
+      "loss": 0.7513,
+      "step": 285
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.445378151260505e-06,
+      "loss": 0.7035,
+      "step": 286
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.436974789915967e-06,
+      "loss": 0.7273,
+      "step": 287
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.428571428571429e-06,
+      "loss": 0.7304,
+      "step": 288
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.420168067226891e-06,
+      "loss": 0.7278,
+      "step": 289
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.411764705882353e-06,
+      "loss": 0.7299,
+      "step": 290
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.403361344537815e-06,
+      "loss": 0.7275,
+      "step": 291
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.394957983193277e-06,
+      "loss": 0.7232,
+      "step": 292
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.386554621848739e-06,
+      "loss": 0.697,
+      "step": 293
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.378151260504202e-06,
+      "loss": 0.7324,
+      "step": 294
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.369747899159664e-06,
+      "loss": 0.733,
+      "step": 295
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.361344537815126e-06,
+      "loss": 0.7218,
+      "step": 296
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.352941176470588e-06,
+      "loss": 0.6961,
+      "step": 297
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.3445378151260505e-06,
+      "loss": 0.7239,
+      "step": 298
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.336134453781513e-06,
+      "loss": 0.7167,
+      "step": 299
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.327731092436975e-06,
+      "loss": 0.7439,
+      "step": 300
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.319327731092437e-06,
+      "loss": 0.7214,
+      "step": 301
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.310924369747899e-06,
+      "loss": 0.7495,
+      "step": 302
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.302521008403361e-06,
+      "loss": 0.728,
+      "step": 303
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.294117647058823e-06,
+      "loss": 0.7445,
+      "step": 304
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.285714285714286e-06,
+      "loss": 0.7366,
+      "step": 305
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.2773109243697484e-06,
+      "loss": 0.7275,
+      "step": 306
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.268907563025211e-06,
+      "loss": 0.7455,
+      "step": 307
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.260504201680673e-06,
+      "loss": 0.7443,
+      "step": 308
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.252100840336135e-06,
+      "loss": 0.7016,
+      "step": 309
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.243697478991597e-06,
+      "loss": 0.7468,
+      "step": 310
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.235294117647059e-06,
+      "loss": 0.7084,
+      "step": 311
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.226890756302521e-06,
+      "loss": 0.7252,
+      "step": 312
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.2184873949579826e-06,
+      "loss": 0.7321,
+      "step": 313
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.2100840336134455e-06,
+      "loss": 0.7194,
+      "step": 314
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.201680672268908e-06,
+      "loss": 0.7089,
+      "step": 315
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.19327731092437e-06,
+      "loss": 0.731,
+      "step": 316
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.184873949579832e-06,
+      "loss": 0.7354,
+      "step": 317
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.176470588235294e-06,
+      "loss": 0.7188,
+      "step": 318
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.168067226890756e-06,
+      "loss": 0.7285,
+      "step": 319
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.159663865546218e-06,
+      "loss": 0.7187,
+      "step": 320
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.1512605042016805e-06,
+      "loss": 0.7402,
+      "step": 321
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.142857142857143e-06,
+      "loss": 0.7194,
+      "step": 322
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.134453781512605e-06,
+      "loss": 0.7422,
+      "step": 323
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.126050420168067e-06,
+      "loss": 0.7457,
+      "step": 324
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.11764705882353e-06,
+      "loss": 0.7158,
+      "step": 325
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.109243697478992e-06,
+      "loss": 0.7252,
+      "step": 326
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.100840336134454e-06,
+      "loss": 0.7123,
+      "step": 327
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.092436974789916e-06,
+      "loss": 0.7281,
+      "step": 328
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.0840336134453784e-06,
+      "loss": 0.7347,
+      "step": 329
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.075630252100841e-06,
+      "loss": 0.7285,
+      "step": 330
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.067226890756303e-06,
+      "loss": 0.7389,
+      "step": 331
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.058823529411765e-06,
+      "loss": 0.7376,
+      "step": 332
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.050420168067227e-06,
+      "loss": 0.7392,
+      "step": 333
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.042016806722689e-06,
+      "loss": 0.7471,
+      "step": 334
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.033613445378151e-06,
+      "loss": 0.737,
+      "step": 335
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.025210084033613e-06,
+      "loss": 0.7264,
+      "step": 336
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.0168067226890755e-06,
+      "loss": 0.7135,
+      "step": 337
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.008403361344538e-06,
+      "loss": 0.7076,
+      "step": 338
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4e-06,
+      "loss": 0.7435,
+      "step": 339
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.991596638655462e-06,
+      "loss": 0.7291,
+      "step": 340
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.983193277310924e-06,
+      "loss": 0.7362,
+      "step": 341
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.974789915966386e-06,
+      "loss": 0.7216,
+      "step": 342
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.966386554621848e-06,
+      "loss": 0.7304,
+      "step": 343
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.957983193277311e-06,
+      "loss": 0.7347,
+      "step": 344
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.9495798319327735e-06,
+      "loss": 0.725,
+      "step": 345
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.941176470588236e-06,
+      "loss": 0.735,
+      "step": 346
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.932773109243698e-06,
+      "loss": 0.6993,
+      "step": 347
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.92436974789916e-06,
+      "loss": 0.7304,
+      "step": 348
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.915966386554622e-06,
+      "loss": 0.7319,
+      "step": 349
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.907563025210084e-06,
+      "loss": 0.7302,
+      "step": 350
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.899159663865546e-06,
+      "loss": 0.7174,
+      "step": 351
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.8907563025210084e-06,
+      "loss": 0.7344,
+      "step": 352
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 3.882352941176471e-06,
+      "loss": 0.7472,
+      "step": 353
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 3.873949579831933e-06,
+      "loss": 0.7265,
+      "step": 354
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 3.865546218487396e-06,
+      "loss": 0.7508,
+      "step": 355
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 3.857142857142858e-06,
+      "loss": 0.7163,
+      "step": 356
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.84873949579832e-06,
+      "loss": 0.7336,
+      "step": 357
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.840336134453781e-06,
+      "loss": 0.7464,
+      "step": 358
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.831932773109243e-06,
+      "loss": 0.7261,
+      "step": 359
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 3.8235294117647055e-06,
+      "loss": 0.7222,
+      "step": 360
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.815126050420168e-06,
+      "loss": 0.7522,
+      "step": 361
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.8067226890756302e-06,
+      "loss": 0.7152,
+      "step": 362
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.7983193277310924e-06,
+      "loss": 0.7134,
+      "step": 363
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.7899159663865545e-06,
+      "loss": 0.7028,
+      "step": 364
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.7815126050420167e-06,
+      "loss": 0.7166,
+      "step": 365
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.773109243697479e-06,
+      "loss": 0.7329,
+      "step": 366
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.7647058823529414e-06,
+      "loss": 0.7305,
+      "step": 367
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.7563025210084035e-06,
+      "loss": 0.7561,
+      "step": 368
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 3.7478991596638656e-06,
+      "loss": 0.7281,
+      "step": 369
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 3.7394957983193278e-06,
+      "loss": 0.7117,
+      "step": 370
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 3.73109243697479e-06,
+      "loss": 0.7031,
+      "step": 371
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 3.7226890756302525e-06,
+      "loss": 0.7133,
+      "step": 372
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 3.7142857142857146e-06,
+      "loss": 0.7399,
+      "step": 373
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 3.7058823529411767e-06,
+      "loss": 0.7102,
+      "step": 374
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 3.697478991596639e-06,
+      "loss": 0.716,
+      "step": 375
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 3.689075630252101e-06,
+      "loss": 0.7448,
+      "step": 376
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.6806722689075636e-06,
+      "loss": 0.7139,
+      "step": 377
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.6722689075630257e-06,
+      "loss": 0.7244,
+      "step": 378
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.663865546218488e-06,
+      "loss": 0.731,
+      "step": 379
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.655462184873949e-06,
+      "loss": 0.7344,
+      "step": 380
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.6470588235294117e-06,
+      "loss": 0.7297,
+      "step": 381
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 3.638655462184874e-06,
+      "loss": 0.718,
+      "step": 382
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 3.630252100840336e-06,
+      "loss": 0.7262,
+      "step": 383
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 3.621848739495798e-06,
+      "loss": 0.7185,
+      "step": 384
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 3.6134453781512602e-06,
+      "loss": 0.7084,
+      "step": 385
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 3.605042016806723e-06,
+      "loss": 0.7152,
+      "step": 386
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 3.596638655462185e-06,
+      "loss": 0.6996,
+      "step": 387
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 3.588235294117647e-06,
+      "loss": 0.7184,
+      "step": 388
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 3.5798319327731092e-06,
+      "loss": 0.7,
+      "step": 389
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.5714285714285714e-06,
+      "loss": 0.7155,
+      "step": 390
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.5630252100840335e-06,
+      "loss": 0.7204,
+      "step": 391
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.554621848739496e-06,
+      "loss": 0.7248,
+      "step": 392
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.546218487394958e-06,
+      "loss": 0.7232,
+      "step": 393
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.5378151260504203e-06,
+      "loss": 0.7163,
+      "step": 394
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.5294117647058825e-06,
+      "loss": 0.7219,
+      "step": 395
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.5210084033613446e-06,
+      "loss": 0.7422,
+      "step": 396
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.512605042016807e-06,
+      "loss": 0.719,
+      "step": 397
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 3.5042016806722693e-06,
+      "loss": 0.7058,
+      "step": 398
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 3.4957983193277314e-06,
+      "loss": 0.7321,
+      "step": 399
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 3.4873949579831936e-06,
+      "loss": 0.7247,
+      "step": 400
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 3.4789915966386557e-06,
+      "loss": 0.721,
+      "step": 401
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 3.470588235294118e-06,
+      "loss": 0.7138,
+      "step": 402
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 3.4621848739495796e-06,
+      "loss": 0.7322,
+      "step": 403
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 3.4537815126050417e-06,
+      "loss": 0.6883,
+      "step": 404
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 3.445378151260504e-06,
+      "loss": 0.7016,
+      "step": 405
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4369747899159664e-06,
+      "loss": 0.7124,
+      "step": 406
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4285714285714285e-06,
+      "loss": 0.7302,
+      "step": 407
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4201680672268907e-06,
+      "loss": 0.7583,
+      "step": 408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.411764705882353e-06,
+      "loss": 0.727,
+      "step": 409
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.403361344537815e-06,
+      "loss": 0.7506,
+      "step": 410
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3949579831932775e-06,
+      "loss": 0.7356,
+      "step": 411
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3865546218487396e-06,
+      "loss": 0.7154,
+      "step": 412
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3781512605042018e-06,
+      "loss": 0.7204,
+      "step": 413
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.369747899159664e-06,
+      "loss": 0.6537,
+      "step": 414
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.361344537815126e-06,
+      "loss": 0.6609,
+      "step": 415
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.352941176470588e-06,
+      "loss": 0.662,
+      "step": 416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3445378151260508e-06,
+      "loss": 0.6924,
+      "step": 417
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.336134453781513e-06,
+      "loss": 0.6623,
+      "step": 418
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.327731092436975e-06,
+      "loss": 0.6464,
+      "step": 419
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.319327731092437e-06,
+      "loss": 0.6102,
+      "step": 420
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3109243697478993e-06,
+      "loss": 0.6213,
+      "step": 421
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.302521008403362e-06,
+      "loss": 0.6724,
+      "step": 422
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.294117647058824e-06,
+      "loss": 0.6709,
+      "step": 423
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.285714285714286e-06,
+      "loss": 0.654,
+      "step": 424
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.277310924369748e-06,
+      "loss": 0.6542,
+      "step": 425
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.26890756302521e-06,
+      "loss": 0.6449,
+      "step": 426
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.260504201680672e-06,
+      "loss": 0.6568,
+      "step": 427
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 3.2521008403361343e-06,
+      "loss": 0.6495,
+      "step": 428
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 3.2436974789915964e-06,
+      "loss": 0.6644,
+      "step": 429
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 3.2352941176470585e-06,
+      "loss": 0.6548,
+      "step": 430
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 3.226890756302521e-06,
+      "loss": 0.6472,
+      "step": 431
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.2184873949579832e-06,
+      "loss": 0.6447,
+      "step": 432
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.2100840336134454e-06,
+      "loss": 0.6323,
+      "step": 433
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.2016806722689075e-06,
+      "loss": 0.6511,
+      "step": 434
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.1932773109243696e-06,
+      "loss": 0.6498,
+      "step": 435
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 3.184873949579832e-06,
+      "loss": 0.6358,
+      "step": 436
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 3.1764705882352943e-06,
+      "loss": 0.6599,
+      "step": 437
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 3.1680672268907565e-06,
+      "loss": 0.6596,
+      "step": 438
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 3.1596638655462186e-06,
+      "loss": 0.6301,
+      "step": 439
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 3.1512605042016808e-06,
+      "loss": 0.6251,
+      "step": 440
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 3.142857142857143e-06,
+      "loss": 0.6552,
+      "step": 441
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 3.1344537815126055e-06,
+      "loss": 0.6479,
+      "step": 442
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 3.1260504201680676e-06,
+      "loss": 0.6602,
+      "step": 443
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.1176470588235297e-06,
+      "loss": 0.6412,
+      "step": 444
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.109243697478992e-06,
+      "loss": 0.6397,
+      "step": 445
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.100840336134454e-06,
+      "loss": 0.6677,
+      "step": 446
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.0924369747899157e-06,
+      "loss": 0.6144,
+      "step": 447
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 3.084033613445378e-06,
+      "loss": 0.6347,
+      "step": 448
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 3.07563025210084e-06,
+      "loss": 0.6562,
+      "step": 449
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 3.0672268907563026e-06,
+      "loss": 0.6583,
+      "step": 450
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 3.0588235294117647e-06,
+      "loss": 0.6759,
+      "step": 451
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.050420168067227e-06,
+      "loss": 0.6663,
+      "step": 452
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.042016806722689e-06,
+      "loss": 0.6551,
+      "step": 453
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.033613445378151e-06,
+      "loss": 0.6738,
+      "step": 454
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.0252100840336132e-06,
+      "loss": 0.6469,
+      "step": 455
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3.016806722689076e-06,
+      "loss": 0.6549,
+      "step": 456
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3.008403361344538e-06,
+      "loss": 0.6404,
+      "step": 457
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3e-06,
+      "loss": 0.6372,
+      "step": 458
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.9915966386554622e-06,
+      "loss": 0.6541,
+      "step": 459
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.9831932773109244e-06,
+      "loss": 0.6356,
+      "step": 460
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.974789915966387e-06,
+      "loss": 0.6241,
+      "step": 461
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.966386554621849e-06,
+      "loss": 0.6622,
+      "step": 462
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.957983193277311e-06,
+      "loss": 0.6708,
+      "step": 463
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.949579831932773e-06,
+      "loss": 0.6605,
+      "step": 464
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.941176470588235e-06,
+      "loss": 0.6521,
+      "step": 465
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.9327731092436976e-06,
+      "loss": 0.6373,
+      "step": 466
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.9243697478991597e-06,
+      "loss": 0.6655,
+      "step": 467
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.915966386554622e-06,
+      "loss": 0.6313,
+      "step": 468
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.907563025210084e-06,
+      "loss": 0.6155,
+      "step": 469
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.899159663865546e-06,
+      "loss": 0.6526,
+      "step": 470
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.8907563025210087e-06,
+      "loss": 0.6653,
+      "step": 471
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.882352941176471e-06,
+      "loss": 0.6351,
+      "step": 472
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 2.873949579831933e-06,
+      "loss": 0.6724,
+      "step": 473
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 2.865546218487395e-06,
+      "loss": 0.6386,
+      "step": 474
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 2.8571428571428573e-06,
+      "loss": 0.6352,
+      "step": 475
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 2.8487394957983194e-06,
+      "loss": 0.6566,
+      "step": 476
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 2.8403361344537815e-06,
+      "loss": 0.6931,
+      "step": 477
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 2.8319327731092437e-06,
+      "loss": 0.6714,
+      "step": 478
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 2.823529411764706e-06,
+      "loss": 0.6605,
+      "step": 479
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 2.815126050420168e-06,
+      "loss": 0.6342,
+      "step": 480
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 2.8067226890756305e-06,
+      "loss": 0.6425,
+      "step": 481
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 2.7983193277310926e-06,
+      "loss": 0.6374,
+      "step": 482
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 2.7899159663865548e-06,
+      "loss": 0.6417,
+      "step": 483
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 2.781512605042017e-06,
+      "loss": 0.6414,
+      "step": 484
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 2.773109243697479e-06,
+      "loss": 0.6314,
+      "step": 485
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 2.764705882352941e-06,
+      "loss": 0.6606,
+      "step": 486
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 2.7563025210084033e-06,
+      "loss": 0.6748,
+      "step": 487
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 2.7478991596638655e-06,
+      "loss": 0.6099,
+      "step": 488
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.7394957983193276e-06,
+      "loss": 0.6424,
+      "step": 489
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.7310924369747897e-06,
+      "loss": 0.6311,
+      "step": 490
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.7226890756302523e-06,
+      "loss": 0.6463,
+      "step": 491
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.7142857142857144e-06,
+      "loss": 0.6723,
+      "step": 492
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 2.7058823529411766e-06,
+      "loss": 0.6463,
+      "step": 493
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 2.6974789915966387e-06,
+      "loss": 0.6764,
+      "step": 494
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 2.689075630252101e-06,
+      "loss": 0.6475,
+      "step": 495
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 2.6806722689075634e-06,
+      "loss": 0.6526,
+      "step": 496
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 2.6722689075630255e-06,
+      "loss": 0.6402,
+      "step": 497
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 2.6638655462184873e-06,
+      "loss": 0.6331,
+      "step": 498
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 2.6554621848739494e-06,
+      "loss": 0.6346,
+      "step": 499
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 2.647058823529412e-06,
+      "loss": 0.6306,
+      "step": 500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 2.638655462184874e-06,
+      "loss": 0.6578,
+      "step": 501
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 2.6302521008403362e-06,
+      "loss": 0.6477,
+      "step": 502
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 2.6218487394957984e-06,
+      "loss": 0.6377,
+      "step": 503
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 2.6134453781512605e-06,
+      "loss": 0.6585,
+      "step": 504
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.6050420168067226e-06,
+      "loss": 0.6341,
+      "step": 505
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.596638655462185e-06,
+      "loss": 0.6534,
+      "step": 506
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.5882352941176473e-06,
+      "loss": 0.6521,
+      "step": 507
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.5798319327731095e-06,
+      "loss": 0.6457,
+      "step": 508
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.571428571428571e-06,
+      "loss": 0.6533,
+      "step": 509
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.5630252100840338e-06,
+      "loss": 0.655,
+      "step": 510
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.554621848739496e-06,
+      "loss": 0.6516,
+      "step": 511
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.546218487394958e-06,
+      "loss": 0.6452,
+      "step": 512
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 2.53781512605042e-06,
+      "loss": 0.6336,
+      "step": 513
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 2.5294117647058823e-06,
+      "loss": 0.6476,
+      "step": 514
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 2.5210084033613444e-06,
+      "loss": 0.6659,
+      "step": 515
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 2.512605042016807e-06,
+      "loss": 0.6405,
+      "step": 516
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.504201680672269e-06,
+      "loss": 0.6357,
+      "step": 517
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.4957983193277313e-06,
+      "loss": 0.6498,
+      "step": 518
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.4873949579831934e-06,
+      "loss": 0.6374,
+      "step": 519
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.4789915966386555e-06,
+      "loss": 0.6538,
+      "step": 520
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.4705882352941177e-06,
+      "loss": 0.6679,
+      "step": 521
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.46218487394958e-06,
+      "loss": 0.6247,
+      "step": 522
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.453781512605042e-06,
+      "loss": 0.6425,
+      "step": 523
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.445378151260504e-06,
+      "loss": 0.6356,
+      "step": 524
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.4369747899159667e-06,
+      "loss": 0.6557,
+      "step": 525
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.428571428571429e-06,
+      "loss": 0.6327,
+      "step": 526
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.420168067226891e-06,
+      "loss": 0.6267,
+      "step": 527
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.411764705882353e-06,
+      "loss": 0.6438,
+      "step": 528
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.403361344537815e-06,
+      "loss": 0.6407,
+      "step": 529
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.3949579831932773e-06,
+      "loss": 0.6389,
+      "step": 530
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.3865546218487395e-06,
+      "loss": 0.6279,
+      "step": 531
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.3781512605042016e-06,
+      "loss": 0.6104,
+      "step": 532
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.3697478991596638e-06,
+      "loss": 0.6451,
+      "step": 533
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 2.361344537815126e-06,
+      "loss": 0.6474,
+      "step": 534
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 2.3529411764705885e-06,
+      "loss": 0.6498,
+      "step": 535
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 2.3445378151260506e-06,
+      "loss": 0.6576,
+      "step": 536
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 2.3361344537815127e-06,
+      "loss": 0.6659,
+      "step": 537
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.327731092436975e-06,
+      "loss": 0.6441,
+      "step": 538
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.319327731092437e-06,
+      "loss": 0.645,
+      "step": 539
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.310924369747899e-06,
+      "loss": 0.6449,
+      "step": 540
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.3025210084033617e-06,
+      "loss": 0.6666,
+      "step": 541
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 2.2941176470588234e-06,
+      "loss": 0.6338,
+      "step": 542
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 2.2857142857142856e-06,
+      "loss": 0.6317,
+      "step": 543
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 2.2773109243697477e-06,
+      "loss": 0.6484,
+      "step": 544
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 2.2689075630252102e-06,
+      "loss": 0.6445,
+      "step": 545
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 2.2605042016806724e-06,
+      "loss": 0.6414,
+      "step": 546
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 2.2521008403361345e-06,
+      "loss": 0.6443,
+      "step": 547
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 2.2436974789915967e-06,
+      "loss": 0.6488,
+      "step": 548
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 2.235294117647059e-06,
+      "loss": 0.6479,
+      "step": 549
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 2.2268907563025214e-06,
+      "loss": 0.6078,
+      "step": 550
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 2.2184873949579835e-06,
+      "loss": 0.6432,
+      "step": 551
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 2.2100840336134456e-06,
+      "loss": 0.6512,
+      "step": 552
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 2.2016806722689073e-06,
+      "loss": 0.6522,
+      "step": 553
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.1932773109243695e-06,
+      "loss": 0.6406,
+      "step": 554
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.184873949579832e-06,
+      "loss": 0.6389,
+      "step": 555
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.176470588235294e-06,
+      "loss": 0.6268,
+      "step": 556
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.1680672268907563e-06,
+      "loss": 0.6258,
+      "step": 557
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 2.1596638655462185e-06,
+      "loss": 0.6555,
+      "step": 558
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 2.1512605042016806e-06,
+      "loss": 0.6401,
+      "step": 559
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 2.142857142857143e-06,
+      "loss": 0.6552,
+      "step": 560
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 2.1344537815126053e-06,
+      "loss": 0.6703,
+      "step": 561
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.1260504201680674e-06,
+      "loss": 0.6512,
+      "step": 562
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.1176470588235296e-06,
+      "loss": 0.6415,
+      "step": 563
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.1092436974789913e-06,
+      "loss": 0.6572,
+      "step": 564
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.100840336134454e-06,
+      "loss": 0.6622,
+      "step": 565
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 2.092436974789916e-06,
+      "loss": 0.6691,
+      "step": 566
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 2.084033613445378e-06,
+      "loss": 0.6422,
+      "step": 567
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 2.0756302521008403e-06,
+      "loss": 0.6557,
+      "step": 568
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 2.0672268907563024e-06,
+      "loss": 0.6606,
+      "step": 569
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.058823529411765e-06,
+      "loss": 0.6533,
+      "step": 570
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.050420168067227e-06,
+      "loss": 0.6407,
+      "step": 571
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.0420168067226892e-06,
+      "loss": 0.6524,
+      "step": 572
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.0336134453781514e-06,
+      "loss": 0.645,
+      "step": 573
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.0252100840336135e-06,
+      "loss": 0.6445,
+      "step": 574
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.0168067226890756e-06,
+      "loss": 0.6602,
+      "step": 575
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.0084033613445378e-06,
+      "loss": 0.6412,
+      "step": 576
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2e-06,
+      "loss": 0.6525,
+      "step": 577
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.991596638655462e-06,
+      "loss": 0.6369,
+      "step": 578
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.983193277310924e-06,
+      "loss": 0.6349,
+      "step": 579
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.9747899159663867e-06,
+      "loss": 0.6658,
+      "step": 580
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.966386554621849e-06,
+      "loss": 0.6326,
+      "step": 581
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.957983193277311e-06,
+      "loss": 0.6361,
+      "step": 582
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.949579831932773e-06,
+      "loss": 0.6565,
+      "step": 583
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.9411764705882353e-06,
+      "loss": 0.6351,
+      "step": 584
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.932773109243698e-06,
+      "loss": 0.6596,
+      "step": 585
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.92436974789916e-06,
+      "loss": 0.6425,
+      "step": 586
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.9159663865546217e-06,
+      "loss": 0.647,
+      "step": 587
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.907563025210084e-06,
+      "loss": 0.653,
+      "step": 588
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.8991596638655462e-06,
+      "loss": 0.663,
+      "step": 589
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.8907563025210083e-06,
+      "loss": 0.6504,
+      "step": 590
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.8823529411764707e-06,
+      "loss": 0.6495,
+      "step": 591
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.8739495798319328e-06,
+      "loss": 0.6636,
+      "step": 592
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.865546218487395e-06,
+      "loss": 0.6387,
+      "step": 593
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.8571428571428573e-06,
+      "loss": 0.6671,
+      "step": 594
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.8487394957983194e-06,
+      "loss": 0.6744,
+      "step": 595
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.8403361344537818e-06,
+      "loss": 0.6377,
+      "step": 596
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.831932773109244e-06,
+      "loss": 0.66,
+      "step": 597
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.8235294117647058e-06,
+      "loss": 0.6387,
+      "step": 598
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.815126050420168e-06,
+      "loss": 0.6523,
+      "step": 599
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.8067226890756301e-06,
+      "loss": 0.6298,
+      "step": 600
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.7983193277310925e-06,
+      "loss": 0.635,
+      "step": 601
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.7899159663865546e-06,
+      "loss": 0.6395,
+      "step": 602
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.7815126050420167e-06,
+      "loss": 0.6644,
+      "step": 603
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.773109243697479e-06,
+      "loss": 0.6434,
+      "step": 604
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.7647058823529412e-06,
+      "loss": 0.6566,
+      "step": 605
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.7563025210084036e-06,
+      "loss": 0.6506,
+      "step": 606
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.7478991596638657e-06,
+      "loss": 0.6525,
+      "step": 607
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.7394957983193279e-06,
+      "loss": 0.6327,
+      "step": 608
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.7310924369747898e-06,
+      "loss": 0.6519,
+      "step": 609
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.722689075630252e-06,
+      "loss": 0.6306,
+      "step": 610
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.7142857142857143e-06,
+      "loss": 0.639,
+      "step": 611
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.7058823529411764e-06,
+      "loss": 0.6392,
+      "step": 612
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6974789915966388e-06,
+      "loss": 0.6482,
+      "step": 613
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6890756302521009e-06,
+      "loss": 0.6739,
+      "step": 614
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.680672268907563e-06,
+      "loss": 0.6375,
+      "step": 615
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6722689075630254e-06,
+      "loss": 0.6422,
+      "step": 616
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6638655462184875e-06,
+      "loss": 0.6439,
+      "step": 617
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6554621848739497e-06,
+      "loss": 0.6437,
+      "step": 618
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.647058823529412e-06,
+      "loss": 0.6452,
+      "step": 619
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.638655462184874e-06,
+      "loss": 0.632,
+      "step": 620
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.630252100840336e-06,
+      "loss": 0.6265,
+      "step": 621
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6218487394957982e-06,
+      "loss": 0.6403,
+      "step": 622
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.6134453781512606e-06,
+      "loss": 0.6337,
+      "step": 623
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.6050420168067227e-06,
+      "loss": 0.6491,
+      "step": 624
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5966386554621848e-06,
+      "loss": 0.6457,
+      "step": 625
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5882352941176472e-06,
+      "loss": 0.6451,
+      "step": 626
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5798319327731093e-06,
+      "loss": 0.6409,
+      "step": 627
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5714285714285714e-06,
+      "loss": 0.6372,
+      "step": 628
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5630252100840338e-06,
+      "loss": 0.6497,
+      "step": 629
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.554621848739496e-06,
+      "loss": 0.652,
+      "step": 630
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5462184873949579e-06,
+      "loss": 0.6405,
+      "step": 631
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.53781512605042e-06,
+      "loss": 0.6295,
+      "step": 632
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5294117647058823e-06,
+      "loss": 0.673,
+      "step": 633
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5210084033613445e-06,
+      "loss": 0.6458,
+      "step": 634
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5126050420168066e-06,
+      "loss": 0.6345,
+      "step": 635
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.504201680672269e-06,
+      "loss": 0.6352,
+      "step": 636
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.4957983193277311e-06,
+      "loss": 0.6429,
+      "step": 637
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.4873949579831935e-06,
+      "loss": 0.6468,
+      "step": 638
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.4789915966386556e-06,
+      "loss": 0.6425,
+      "step": 639
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.4705882352941175e-06,
+      "loss": 0.6462,
+      "step": 640
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.4621848739495799e-06,
+      "loss": 0.6341,
+      "step": 641
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.453781512605042e-06,
+      "loss": 0.63,
+      "step": 642
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.4453781512605044e-06,
+      "loss": 0.6397,
+      "step": 643
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.4369747899159665e-06,
+      "loss": 0.655,
+      "step": 644
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.4285714285714286e-06,
+      "loss": 0.6586,
+      "step": 645
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.4201680672268908e-06,
+      "loss": 0.6419,
+      "step": 646
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.411764705882353e-06,
+      "loss": 0.6472,
+      "step": 647
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.4033613445378153e-06,
+      "loss": 0.6337,
+      "step": 648
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.3949579831932774e-06,
+      "loss": 0.648,
+      "step": 649
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.3865546218487395e-06,
+      "loss": 0.6655,
+      "step": 650
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.3781512605042017e-06,
+      "loss": 0.6463,
+      "step": 651
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.3697478991596638e-06,
+      "loss": 0.656,
+      "step": 652
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.3613445378151261e-06,
+      "loss": 0.6236,
+      "step": 653
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.3529411764705883e-06,
+      "loss": 0.6536,
+      "step": 654
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.3445378151260504e-06,
+      "loss": 0.6288,
+      "step": 655
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.3361344537815128e-06,
+      "loss": 0.636,
+      "step": 656
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.3277310924369747e-06,
+      "loss": 0.6292,
+      "step": 657
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.319327731092437e-06,
+      "loss": 0.6474,
+      "step": 658
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.3109243697478992e-06,
+      "loss": 0.6468,
+      "step": 659
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.3025210084033613e-06,
+      "loss": 0.6522,
+      "step": 660
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.2941176470588237e-06,
+      "loss": 0.6227,
+      "step": 661
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.2857142857142856e-06,
+      "loss": 0.6327,
+      "step": 662
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.277310924369748e-06,
+      "loss": 0.6501,
+      "step": 663
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.26890756302521e-06,
+      "loss": 0.6375,
+      "step": 664
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.2605042016806722e-06,
+      "loss": 0.6385,
+      "step": 665
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.2521008403361346e-06,
+      "loss": 0.6374,
+      "step": 666
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.2436974789915967e-06,
+      "loss": 0.6332,
+      "step": 667
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.2352941176470588e-06,
+      "loss": 0.635,
+      "step": 668
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.226890756302521e-06,
+      "loss": 0.6404,
+      "step": 669
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.2184873949579833e-06,
+      "loss": 0.6256,
+      "step": 670
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.2100840336134455e-06,
+      "loss": 0.6076,
+      "step": 671
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.2016806722689076e-06,
+      "loss": 0.6469,
+      "step": 672
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.1932773109243697e-06,
+      "loss": 0.6477,
+      "step": 673
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.1848739495798319e-06,
+      "loss": 0.6408,
+      "step": 674
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.1764705882352942e-06,
+      "loss": 0.6448,
+      "step": 675
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.1680672268907564e-06,
+      "loss": 0.6288,
+      "step": 676
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.1596638655462185e-06,
+      "loss": 0.6398,
+      "step": 677
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.1512605042016808e-06,
+      "loss": 0.6497,
+      "step": 678
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.1428571428571428e-06,
+      "loss": 0.6583,
+      "step": 679
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.1344537815126051e-06,
+      "loss": 0.6346,
+      "step": 680
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.1260504201680673e-06,
+      "loss": 0.6394,
+      "step": 681
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.1176470588235294e-06,
+      "loss": 0.6409,
+      "step": 682
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.1092436974789917e-06,
+      "loss": 0.647,
+      "step": 683
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.1008403361344537e-06,
+      "loss": 0.6454,
+      "step": 684
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.092436974789916e-06,
+      "loss": 0.6292,
+      "step": 685
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.0840336134453782e-06,
+      "loss": 0.6492,
+      "step": 686
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.0756302521008403e-06,
+      "loss": 0.6405,
+      "step": 687
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.0672268907563026e-06,
+      "loss": 0.655,
+      "step": 688
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.0588235294117648e-06,
+      "loss": 0.6375,
+      "step": 689
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.050420168067227e-06,
+      "loss": 0.6613,
+      "step": 690
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.042016806722689e-06,
+      "loss": 0.6513,
+      "step": 691
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.0336134453781512e-06,
+      "loss": 0.6345,
+      "step": 692
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.0252100840336135e-06,
+      "loss": 0.6273,
+      "step": 693
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.0168067226890757e-06,
+      "loss": 0.6186,
+      "step": 694
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.0084033613445378e-06,
+      "loss": 0.6356,
+      "step": 695
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1e-06,
+      "loss": 0.6361,
+      "step": 696
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.91596638655462e-07,
+      "loss": 0.6363,
+      "step": 697
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.831932773109244e-07,
+      "loss": 0.6368,
+      "step": 698
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.747899159663866e-07,
+      "loss": 0.6479,
+      "step": 699
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.66386554621849e-07,
+      "loss": 0.6361,
+      "step": 700
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.579831932773109e-07,
+      "loss": 0.6435,
+      "step": 701
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.495798319327731e-07,
+      "loss": 0.621,
+      "step": 702
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.411764705882353e-07,
+      "loss": 0.6486,
+      "step": 703
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.327731092436975e-07,
+      "loss": 0.6738,
+      "step": 704
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.243697478991597e-07,
+      "loss": 0.6344,
+      "step": 705
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.15966386554622e-07,
+      "loss": 0.6547,
+      "step": 706
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.07563025210084e-07,
+      "loss": 0.616,
+      "step": 707
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.991596638655462e-07,
+      "loss": 0.6652,
+      "step": 708
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.907563025210084e-07,
+      "loss": 0.6593,
+      "step": 709
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.823529411764706e-07,
+      "loss": 0.6381,
+      "step": 710
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.739495798319329e-07,
+      "loss": 0.6556,
+      "step": 711
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.655462184873949e-07,
+      "loss": 0.6465,
+      "step": 712
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 8.571428571428571e-07,
+      "loss": 0.6563,
+      "step": 713
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 8.487394957983194e-07,
+      "loss": 0.6384,
+      "step": 714
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 8.403361344537815e-07,
+      "loss": 0.629,
+      "step": 715
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 8.319327731092438e-07,
+      "loss": 0.6367,
+      "step": 716
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.23529411764706e-07,
+      "loss": 0.6457,
+      "step": 717
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.15126050420168e-07,
+      "loss": 0.6242,
+      "step": 718
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.067226890756303e-07,
+      "loss": 0.6495,
+      "step": 719
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.983193277310924e-07,
+      "loss": 0.6441,
+      "step": 720
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.899159663865547e-07,
+      "loss": 0.6172,
+      "step": 721
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.815126050420169e-07,
+      "loss": 0.6385,
+      "step": 722
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.731092436974789e-07,
+      "loss": 0.6298,
+      "step": 723
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.647058823529412e-07,
+      "loss": 0.6435,
+      "step": 724
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.563025210084033e-07,
+      "loss": 0.6407,
+      "step": 725
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.478991596638656e-07,
+      "loss": 0.6434,
+      "step": 726
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.394957983193278e-07,
+      "loss": 0.639,
+      "step": 727
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.310924369747899e-07,
+      "loss": 0.6611,
+      "step": 728
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.226890756302522e-07,
+      "loss": 0.6266,
+      "step": 729
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.142857142857143e-07,
+      "loss": 0.6462,
+      "step": 730
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.058823529411765e-07,
+      "loss": 0.6476,
+      "step": 731
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 6.974789915966387e-07,
+      "loss": 0.654,
+      "step": 732
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 6.890756302521008e-07,
+      "loss": 0.6634,
+      "step": 733
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 6.806722689075631e-07,
+      "loss": 0.651,
+      "step": 734
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 6.722689075630252e-07,
+      "loss": 0.6557,
+      "step": 735
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 6.638655462184873e-07,
+      "loss": 0.64,
+      "step": 736
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.554621848739496e-07,
+      "loss": 0.6361,
+      "step": 737
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.470588235294118e-07,
+      "loss": 0.6434,
+      "step": 738
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.38655462184874e-07,
+      "loss": 0.6376,
+      "step": 739
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.302521008403361e-07,
+      "loss": 0.6415,
+      "step": 740
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.218487394957984e-07,
+      "loss": 0.6419,
+      "step": 741
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.134453781512605e-07,
+      "loss": 0.654,
+      "step": 742
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.050420168067227e-07,
+      "loss": 0.6393,
+      "step": 743
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 5.966386554621849e-07,
+      "loss": 0.6239,
+      "step": 744
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 5.882352941176471e-07,
+      "loss": 0.6343,
+      "step": 745
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 5.798319327731093e-07,
+      "loss": 0.6712,
+      "step": 746
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 5.714285714285714e-07,
+      "loss": 0.636,
+      "step": 747
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 5.630252100840336e-07,
+      "loss": 0.6427,
+      "step": 748
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 5.546218487394959e-07,
+      "loss": 0.6402,
+      "step": 749
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 5.46218487394958e-07,
+      "loss": 0.642,
+      "step": 750
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 5.378151260504201e-07,
+      "loss": 0.6279,
+      "step": 751
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 5.294117647058824e-07,
+      "loss": 0.6566,
+      "step": 752
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 5.210084033613445e-07,
+      "loss": 0.6504,
+      "step": 753
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 5.126050420168068e-07,
+      "loss": 0.6647,
+      "step": 754
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 5.042016806722689e-07,
+      "loss": 0.6461,
+      "step": 755
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 4.95798319327731e-07,
+      "loss": 0.6468,
+      "step": 756
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 4.873949579831933e-07,
+      "loss": 0.6488,
+      "step": 757
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.789915966386554e-07,
+      "loss": 0.6778,
+      "step": 758
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.7058823529411767e-07,
+      "loss": 0.6471,
+      "step": 759
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.6218487394957986e-07,
+      "loss": 0.6335,
+      "step": 760
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.53781512605042e-07,
+      "loss": 0.6581,
+      "step": 761
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.453781512605042e-07,
+      "loss": 0.6384,
+      "step": 762
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.3697478991596643e-07,
+      "loss": 0.6279,
+      "step": 763
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.2857142857142857e-07,
+      "loss": 0.6318,
+      "step": 764
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.2016806722689076e-07,
+      "loss": 0.6521,
+      "step": 765
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.11764705882353e-07,
+      "loss": 0.64,
+      "step": 766
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.0336134453781514e-07,
+      "loss": 0.6286,
+      "step": 767
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 3.9495798319327733e-07,
+      "loss": 0.6287,
+      "step": 768
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 3.8655462184873946e-07,
+      "loss": 0.6286,
+      "step": 769
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 3.7815126050420166e-07,
+      "loss": 0.6454,
+      "step": 770
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 3.697478991596639e-07,
+      "loss": 0.6305,
+      "step": 771
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 3.613445378151261e-07,
+      "loss": 0.6136,
+      "step": 772
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 3.529411764705882e-07,
+      "loss": 0.6458,
+      "step": 773
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.445378151260504e-07,
+      "loss": 0.6486,
+      "step": 774
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.361344537815126e-07,
+      "loss": 0.6313,
+      "step": 775
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.277310924369748e-07,
+      "loss": 0.6246,
+      "step": 776
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.19327731092437e-07,
+      "loss": 0.6345,
+      "step": 777
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 3.109243697478992e-07,
+      "loss": 0.6417,
+      "step": 778
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 3.0252100840336137e-07,
+      "loss": 0.635,
+      "step": 779
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.9411764705882356e-07,
+      "loss": 0.6473,
+      "step": 780
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.857142857142857e-07,
+      "loss": 0.6351,
+      "step": 781
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.7731092436974794e-07,
+      "loss": 0.6268,
+      "step": 782
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.689075630252101e-07,
+      "loss": 0.6251,
+      "step": 783
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.6050420168067226e-07,
+      "loss": 0.6354,
+      "step": 784
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.5210084033613445e-07,
+      "loss": 0.6434,
+      "step": 785
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.4369747899159664e-07,
+      "loss": 0.6279,
+      "step": 786
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.3529411764705883e-07,
+      "loss": 0.627,
+      "step": 787
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.26890756302521e-07,
+      "loss": 0.6324,
+      "step": 788
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.1848739495798321e-07,
+      "loss": 0.6511,
+      "step": 789
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.1008403361344538e-07,
+      "loss": 0.6578,
+      "step": 790
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.0168067226890757e-07,
+      "loss": 0.65,
+      "step": 791
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.9327731092436973e-07,
+      "loss": 0.6523,
+      "step": 792
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.8487394957983195e-07,
+      "loss": 0.6384,
+      "step": 793
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.764705882352941e-07,
+      "loss": 0.636,
+      "step": 794
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.680672268907563e-07,
+      "loss": 0.6417,
+      "step": 795
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.596638655462185e-07,
+      "loss": 0.6311,
+      "step": 796
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.5126050420168068e-07,
+      "loss": 0.6418,
+      "step": 797
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.4285714285714285e-07,
+      "loss": 0.6569,
+      "step": 798
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.3445378151260504e-07,
+      "loss": 0.6397,
+      "step": 799
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.2605042016806723e-07,
+      "loss": 0.6338,
+      "step": 800
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.1764705882352942e-07,
+      "loss": 0.6209,
+      "step": 801
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.0924369747899161e-07,
+      "loss": 0.6602,
+      "step": 802
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.0084033613445378e-07,
+      "loss": 0.6634,
+      "step": 803
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.243697478991597e-08,
+      "loss": 0.6433,
+      "step": 804
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 8.403361344537815e-08,
+      "loss": 0.6666,
+      "step": 805
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 7.563025210084034e-08,
+      "loss": 0.6515,
+      "step": 806
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 6.722689075630252e-08,
+      "loss": 0.6698,
+      "step": 807
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.882352941176471e-08,
+      "loss": 0.6463,
+      "step": 808
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.042016806722689e-08,
+      "loss": 0.6522,
+      "step": 809
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.2016806722689076e-08,
+      "loss": 0.6149,
+      "step": 810
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 3.361344537815126e-08,
+      "loss": 0.6234,
+      "step": 811
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.5210084033613446e-08,
+      "loss": 0.642,
+      "step": 812
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.680672268907563e-08,
+      "loss": 0.6412,
+      "step": 813
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 8.403361344537815e-09,
+      "loss": 0.6388,
+      "step": 814
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 814,
+  "num_train_epochs": 2,
+  "save_steps": 407,
+  "total_flos": 2.275967145406261e+19,
+  "trial_name": null,
+  "trial_params": null
+}