diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,12904 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "global_step": 2120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.0833333333333333e-07,
+      "loss": 0.7326,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1666666666666667e-07,
+      "loss": 0.7787,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.25e-07,
+      "loss": 0.7075,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.333333333333333e-07,
+      "loss": 0.7133,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.0416666666666667e-06,
+      "loss": 0.6808,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.25e-06,
+      "loss": 0.8208,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4583333333333335e-06,
+      "loss": 0.7355,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6666666666666667e-06,
+      "loss": 0.6799,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8750000000000003e-06,
+      "loss": 0.7339,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0833333333333334e-06,
+      "loss": 0.7616,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2916666666666666e-06,
+      "loss": 0.745,
+      "step": 11
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5e-06,
+      "loss": 0.6903,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7083333333333334e-06,
+      "loss": 0.7008,
+      "step": 13
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.916666666666667e-06,
+      "loss": 0.7561,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.125e-06,
+      "loss": 0.7029,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 0.6738,
+      "step": 16
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5416666666666673e-06,
+      "loss": 0.6514,
+      "step": 17
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.6679,
+      "step": 18
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.958333333333333e-06,
+      "loss": 0.6322,
+      "step": 19
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.166666666666667e-06,
+      "loss": 0.6581,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3750000000000005e-06,
+      "loss": 0.6535,
+      "step": 21
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.583333333333333e-06,
+      "loss": 0.5464,
+      "step": 22
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.791666666666668e-06,
+      "loss": 0.634,
+      "step": 23
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5e-06,
+      "loss": 0.6029,
+      "step": 24
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.208333333333334e-06,
+      "loss": 0.5905,
+      "step": 25
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.416666666666667e-06,
+      "loss": 0.6053,
+      "step": 26
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.625e-06,
+      "loss": 0.5884,
+      "step": 27
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.833333333333334e-06,
+      "loss": 0.531,
+      "step": 28
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.041666666666667e-06,
+      "loss": 0.6065,
+      "step": 29
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.25e-06,
+      "loss": 0.6337,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.458333333333334e-06,
+      "loss": 0.5916,
+      "step": 31
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.6664,
+      "step": 32
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.875e-06,
+      "loss": 0.6685,
+      "step": 33
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.083333333333335e-06,
+      "loss": 0.5514,
+      "step": 34
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.291666666666667e-06,
+      "loss": 0.5857,
+      "step": 35
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.5973,
+      "step": 36
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.708333333333334e-06,
+      "loss": 0.6512,
+      "step": 37
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.916666666666667e-06,
+      "loss": 0.6075,
+      "step": 38
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.125000000000001e-06,
+      "loss": 0.5983,
+      "step": 39
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.6215,
+      "step": 40
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.541666666666666e-06,
+      "loss": 0.5994,
+      "step": 41
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.750000000000001e-06,
+      "loss": 0.5578,
+      "step": 42
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.958333333333334e-06,
+      "loss": 0.6068,
+      "step": 43
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.166666666666666e-06,
+      "loss": 0.6059,
+      "step": 44
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.375000000000001e-06,
+      "loss": 0.5691,
+      "step": 45
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.583333333333335e-06,
+      "loss": 0.5952,
+      "step": 46
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.791666666666666e-06,
+      "loss": 0.6053,
+      "step": 47
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 0.612,
+      "step": 48
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.0208333333333334e-05,
+      "loss": 0.5973,
+      "step": 49
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.0416666666666668e-05,
+      "loss": 0.5418,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.0625e-05,
+      "loss": 0.637,
+      "step": 51
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.0833333333333334e-05,
+      "loss": 0.6408,
+      "step": 52
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.1041666666666668e-05,
+      "loss": 0.5574,
+      "step": 53
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.125e-05,
+      "loss": 0.6231,
+      "step": 54
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.1458333333333333e-05,
+      "loss": 0.619,
+      "step": 55
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 0.5988,
+      "step": 56
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.1875e-05,
+      "loss": 0.609,
+      "step": 57
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.2083333333333333e-05,
+      "loss": 0.5784,
+      "step": 58
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2291666666666668e-05,
+      "loss": 0.6546,
+      "step": 59
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.25e-05,
+      "loss": 0.6141,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2708333333333333e-05,
+      "loss": 0.5892,
+      "step": 61
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2916666666666668e-05,
+      "loss": 0.66,
+      "step": 62
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.3125e-05,
+      "loss": 0.6163,
+      "step": 63
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.6129,
+      "step": 64
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.3541666666666668e-05,
+      "loss": 0.6026,
+      "step": 65
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.375e-05,
+      "loss": 0.6215,
+      "step": 66
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.3958333333333333e-05,
+      "loss": 0.5868,
+      "step": 67
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.416666666666667e-05,
+      "loss": 0.5981,
+      "step": 68
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.4375e-05,
+      "loss": 0.589,
+      "step": 69
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.4583333333333333e-05,
+      "loss": 0.655,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.479166666666667e-05,
+      "loss": 0.5749,
+      "step": 71
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.5959,
+      "step": 72
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.5208333333333333e-05,
+      "loss": 0.6165,
+      "step": 73
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.5416666666666668e-05,
+      "loss": 0.5976,
+      "step": 74
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.5625e-05,
+      "loss": 0.5764,
+      "step": 75
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.5833333333333333e-05,
+      "loss": 0.548,
+      "step": 76
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.604166666666667e-05,
+      "loss": 0.6566,
+      "step": 77
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.6250000000000002e-05,
+      "loss": 0.5837,
+      "step": 78
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.6458333333333335e-05,
+      "loss": 0.5748,
+      "step": 79
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.5978,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.6875e-05,
+      "loss": 0.6396,
+      "step": 81
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.7083333333333333e-05,
+      "loss": 0.5216,
+      "step": 82
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.729166666666667e-05,
+      "loss": 0.6118,
+      "step": 83
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 0.6012,
+      "step": 84
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.7708333333333335e-05,
+      "loss": 0.6056,
+      "step": 85
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.7916666666666667e-05,
+      "loss": 0.6004,
+      "step": 86
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.8125e-05,
+      "loss": 0.6406,
+      "step": 87
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 0.5906,
+      "step": 88
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.854166666666667e-05,
+      "loss": 0.5875,
+      "step": 89
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.5592,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.8958333333333334e-05,
+      "loss": 0.5899,
+      "step": 91
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.916666666666667e-05,
+      "loss": 0.5738,
+      "step": 92
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9375e-05,
+      "loss": 0.5446,
+      "step": 93
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9583333333333333e-05,
+      "loss": 0.6337,
+      "step": 94
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.979166666666667e-05,
+      "loss": 0.6061,
+      "step": 95
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 2e-05,
+      "loss": 0.6291,
+      "step": 96
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999999481151013e-05,
+      "loss": 0.6291,
+      "step": 97
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9999979246045894e-05,
+      "loss": 0.591,
+      "step": 98
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999995330362345e-05,
+      "loss": 0.6333,
+      "step": 99
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9999916984269715e-05,
+      "loss": 0.6573,
+      "step": 100
+    },
+    {
+      "epoch": 0.09,
+      "eval_loss": 0.6141161322593689,
+      "eval_runtime": 98.2454,
+      "eval_samples_per_second": 7.044,
+      "eval_steps_per_second": 0.448,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999987028802238e-05,
+      "loss": 0.6257,
+      "step": 101
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.99998132149299e-05,
+      "loss": 0.649,
+      "step": 102
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9999745765051502e-05,
+      "loss": 0.6166,
+      "step": 103
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9999667938457175e-05,
+      "loss": 0.5607,
+      "step": 104
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999957973522768e-05,
+      "loss": 0.5782,
+      "step": 105
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9999481155454553e-05,
+      "loss": 0.5954,
+      "step": 106
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999937219924008e-05,
+      "loss": 0.6324,
+      "step": 107
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9999252866697328e-05,
+      "loss": 0.5947,
+      "step": 108
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999912315795013e-05,
+      "loss": 0.604,
+      "step": 109
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9998983073133082e-05,
+      "loss": 0.6668,
+      "step": 110
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999883261239155e-05,
+      "loss": 0.63,
+      "step": 111
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9998671775881668e-05,
+      "loss": 0.6292,
+      "step": 112
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999850056377034e-05,
+      "loss": 0.6459,
+      "step": 113
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999831897623522e-05,
+      "loss": 0.6096,
+      "step": 114
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9998127013464752e-05,
+      "loss": 0.5628,
+      "step": 115
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9997924675658134e-05,
+      "loss": 0.6543,
+      "step": 116
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999771196302533e-05,
+      "loss": 0.6258,
+      "step": 117
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999748887578707e-05,
+      "loss": 0.5721,
+      "step": 118
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999725541417485e-05,
+      "loss": 0.5796,
+      "step": 119
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9997011578430938e-05,
+      "loss": 0.6138,
+      "step": 120
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999675736880836e-05,
+      "loss": 0.616,
+      "step": 121
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9996492785570904e-05,
+      "loss": 0.5827,
+      "step": 122
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9996217828993134e-05,
+      "loss": 0.6213,
+      "step": 123
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999593249936037e-05,
+      "loss": 0.5762,
+      "step": 124
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9995636796968693e-05,
+      "loss": 0.6499,
+      "step": 125
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999533072212496e-05,
+      "loss": 0.6158,
+      "step": 126
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999501427514678e-05,
+      "loss": 0.5746,
+      "step": 127
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9994687456362532e-05,
+      "loss": 0.6043,
+      "step": 128
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999435026611135e-05,
+      "loss": 0.5502,
+      "step": 129
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9994002704743143e-05,
+      "loss": 0.6352,
+      "step": 130
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9993644772618567e-05,
+      "loss": 0.6016,
+      "step": 131
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9993276470109054e-05,
+      "loss": 0.6073,
+      "step": 132
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999289779759679e-05,
+      "loss": 0.5726,
+      "step": 133
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999250875547472e-05,
+      "loss": 0.6235,
+      "step": 134
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999210934414655e-05,
+      "loss": 0.5962,
+      "step": 135
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9991699564026755e-05,
+      "loss": 0.5792,
+      "step": 136
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999127941554056e-05,
+      "loss": 0.6162,
+      "step": 137
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999084889912395e-05,
+      "loss": 0.5992,
+      "step": 138
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9990408015223672e-05,
+      "loss": 0.6222,
+      "step": 139
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9989956764297232e-05,
+      "loss": 0.5913,
+      "step": 140
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.998949514681289e-05,
+      "loss": 0.5724,
+      "step": 141
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.998902316324967e-05,
+      "loss": 0.6323,
+      "step": 142
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9988540814097345e-05,
+      "loss": 0.5976,
+      "step": 143
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9988048099856443e-05,
+      "loss": 0.6314,
+      "step": 144
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9987545021038263e-05,
+      "loss": 0.6173,
+      "step": 145
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998703157816484e-05,
+      "loss": 0.5601,
+      "step": 146
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9986507771768976e-05,
+      "loss": 0.6025,
+      "step": 147
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998597360239422e-05,
+      "loss": 0.5823,
+      "step": 148
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998542907059489e-05,
+      "loss": 0.6384,
+      "step": 149
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9984874176936033e-05,
+      "loss": 0.5913,
+      "step": 150
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998430892199347e-05,
+      "loss": 0.5983,
+      "step": 151
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9983733306353755e-05,
+      "loss": 0.5786,
+      "step": 152
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9983147330614215e-05,
+      "loss": 0.6165,
+      "step": 153
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9982550995382905e-05,
+      "loss": 0.6236,
+      "step": 154
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9981944301278648e-05,
+      "loss": 0.6565,
+      "step": 155
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9981327248931008e-05,
+      "loss": 0.5939,
+      "step": 156
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9980699838980294e-05,
+      "loss": 0.5393,
+      "step": 157
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9980062072077576e-05,
+      "loss": 0.5869,
+      "step": 158
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9979413948884657e-05,
+      "loss": 0.6094,
+      "step": 159
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9978755470074093e-05,
+      "loss": 0.607,
+      "step": 160
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9978086636329195e-05,
+      "loss": 0.5368,
+      "step": 161
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9977407448344e-05,
+      "loss": 0.6108,
+      "step": 162
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9976717906823303e-05,
+      "loss": 0.5986,
+      "step": 163
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9976018012482642e-05,
+      "loss": 0.5919,
+      "step": 164
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9975307766048294e-05,
+      "loss": 0.6005,
+      "step": 165
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9974587168257282e-05,
+      "loss": 0.6247,
+      "step": 166
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9973856219857368e-05,
+      "loss": 0.5739,
+      "step": 167
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9973114921607055e-05,
+      "loss": 0.5933,
+      "step": 168
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997236327427559e-05,
+      "loss": 0.5759,
+      "step": 169
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997160127864295e-05,
+      "loss": 0.6027,
+      "step": 170
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997082893549986e-05,
+      "loss": 0.6017,
+      "step": 171
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9970046245647777e-05,
+      "loss": 0.649,
+      "step": 172
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9969253209898905e-05,
+      "loss": 0.6216,
+      "step": 173
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.996844982907616e-05,
+      "loss": 0.547,
+      "step": 174
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.996763610401322e-05,
+      "loss": 0.6113,
+      "step": 175
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.996681203555449e-05,
+      "loss": 0.6467,
+      "step": 176
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9965977624555087e-05,
+      "loss": 0.5634,
+      "step": 177
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9965132871880893e-05,
+      "loss": 0.6044,
+      "step": 178
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.99642777784085e-05,
+      "loss": 0.5879,
+      "step": 179
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.996341234502524e-05,
+      "loss": 0.5724,
+      "step": 180
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.996253657262917e-05,
+      "loss": 0.5138,
+      "step": 181
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9961650462129072e-05,
+      "loss": 0.5927,
+      "step": 182
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.996075401444447e-05,
+      "loss": 0.5524,
+      "step": 183
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.99598472305056e-05,
+      "loss": 0.5618,
+      "step": 184
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9958930111253432e-05,
+      "loss": 0.6033,
+      "step": 185
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9958002657639656e-05,
+      "loss": 0.5777,
+      "step": 186
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.995706487062669e-05,
+      "loss": 0.5729,
+      "step": 187
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.995611675118768e-05,
+      "loss": 0.5779,
+      "step": 188
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.995515830030648e-05,
+      "loss": 0.5908,
+      "step": 189
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9954189518977672e-05,
+      "loss": 0.5889,
+      "step": 190
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9953210408206565e-05,
+      "loss": 0.5787,
+      "step": 191
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9952220969009175e-05,
+      "loss": 0.6369,
+      "step": 192
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9951221202412242e-05,
+      "loss": 0.6182,
+      "step": 193
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9950211109453222e-05,
+      "loss": 0.6142,
+      "step": 194
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9949190691180292e-05,
+      "loss": 0.6141,
+      "step": 195
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9948159948652324e-05,
+      "loss": 0.5863,
+      "step": 196
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9947118882938935e-05,
+      "loss": 0.5989,
+      "step": 197
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.994606749512042e-05,
+      "loss": 0.59,
+      "step": 198
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9945005786287813e-05,
+      "loss": 0.6165,
+      "step": 199
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9943933757542847e-05,
+      "loss": 0.5721,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "eval_loss": 0.6079282164573669,
+      "eval_runtime": 98.3977,
+      "eval_samples_per_second": 7.033,
+      "eval_steps_per_second": 0.447,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9942851409997958e-05,
+      "loss": 0.5548,
+      "step": 201
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9941758744776302e-05,
+      "loss": 0.6455,
+      "step": 202
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9940655763011728e-05,
+      "loss": 0.5759,
+      "step": 203
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9939542465848808e-05,
+      "loss": 0.6353,
+      "step": 204
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9938418854442796e-05,
+      "loss": 0.6205,
+      "step": 205
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9937284929959674e-05,
+      "loss": 0.5948,
+      "step": 206
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9936140693576103e-05,
+      "loss": 0.5701,
+      "step": 207
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9934986146479457e-05,
+      "loss": 0.6363,
+      "step": 208
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9933821289867813e-05,
+      "loss": 0.5744,
+      "step": 209
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.993264612494994e-05,
+      "loss": 0.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9931460652945293e-05,
+      "loss": 0.6293,
+      "step": 211
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9930264875084045e-05,
+      "loss": 0.5879,
+      "step": 212
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.992905879260705e-05,
+      "loss": 0.6129,
+      "step": 213
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9927842406765852e-05,
+      "loss": 0.5958,
+      "step": 214
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9926615718822697e-05,
+      "loss": 0.6046,
+      "step": 215
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9925378730050518e-05,
+      "loss": 0.5521,
+      "step": 216
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9924131441732933e-05,
+      "loss": 0.596,
+      "step": 217
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.992287385516425e-05,
+      "loss": 0.5675,
+      "step": 218
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9921605971649467e-05,
+      "loss": 0.5719,
+      "step": 219
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9920327792504263e-05,
+      "loss": 0.5757,
+      "step": 220
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9919039319055e-05,
+      "loss": 0.5792,
+      "step": 221
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.991774055263873e-05,
+      "loss": 0.6579,
+      "step": 222
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9916431494603164e-05,
+      "loss": 0.5762,
+      "step": 223
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.991511214630673e-05,
+      "loss": 0.5887,
+      "step": 224
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.99137825091185e-05,
+      "loss": 0.6405,
+      "step": 225
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9912442584418243e-05,
+      "loss": 0.5899,
+      "step": 226
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.991109237359639e-05,
+      "loss": 0.621,
+      "step": 227
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9909731878054057e-05,
+      "loss": 0.6153,
+      "step": 228
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9908361099203026e-05,
+      "loss": 0.5963,
+      "step": 229
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9906980038465748e-05,
+      "loss": 0.5676,
+      "step": 230
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.990558869727535e-05,
+      "loss": 0.5886,
+      "step": 231
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.990418707707562e-05,
+      "loss": 0.5511,
+      "step": 232
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9902775179321026e-05,
+      "loss": 0.5729,
+      "step": 233
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.990135300547668e-05,
+      "loss": 0.5858,
+      "step": 234
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.989992055701838e-05,
+      "loss": 0.6288,
+      "step": 235
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9898477835432566e-05,
+      "loss": 0.6943,
+      "step": 236
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.989702484221635e-05,
+      "loss": 0.5418,
+      "step": 237
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.98955615788775e-05,
+      "loss": 0.5746,
+      "step": 238
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9894088046934445e-05,
+      "loss": 0.5643,
+      "step": 239
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.989260424791626e-05,
+      "loss": 0.5986,
+      "step": 240
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9891110183362684e-05,
+      "loss": 0.6039,
+      "step": 241
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.988960585482411e-05,
+      "loss": 0.5578,
+      "step": 242
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9888091263861563e-05,
+      "loss": 0.6179,
+      "step": 243
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.988656641204674e-05,
+      "loss": 0.6135,
+      "step": 244
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9885031300961975e-05,
+      "loss": 0.625,
+      "step": 245
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9883485932200256e-05,
+      "loss": 0.5913,
+      "step": 246
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9881930307365196e-05,
+      "loss": 0.5854,
+      "step": 247
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9880364428071073e-05,
+      "loss": 0.6206,
+      "step": 248
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9878788295942796e-05,
+      "loss": 0.5828,
+      "step": 249
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9877201912615913e-05,
+      "loss": 0.6236,
+      "step": 250
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.987560527973661e-05,
+      "loss": 0.5627,
+      "step": 251
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9873998398961708e-05,
+      "loss": 0.5604,
+      "step": 252
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9872381271958665e-05,
+      "loss": 0.5494,
+      "step": 253
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9870753900405575e-05,
+      "loss": 0.5877,
+      "step": 254
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9869116285991154e-05,
+      "loss": 0.5588,
+      "step": 255
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9867468430414752e-05,
+      "loss": 0.5789,
+      "step": 256
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9865810335386344e-05,
+      "loss": 0.5936,
+      "step": 257
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.986414200262653e-05,
+      "loss": 0.5678,
+      "step": 258
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9862463433866542e-05,
+      "loss": 0.6115,
+      "step": 259
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9860774630848228e-05,
+      "loss": 0.5876,
+      "step": 260
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9859075595324045e-05,
+      "loss": 0.5799,
+      "step": 261
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.985736632905709e-05,
+      "loss": 0.6088,
+      "step": 262
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.985564683382106e-05,
+      "loss": 0.5941,
+      "step": 263
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.985391711140027e-05,
+      "loss": 0.6513,
+      "step": 264
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.985217716358965e-05,
+      "loss": 0.5672,
+      "step": 265
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9850426992194743e-05,
+      "loss": 0.596,
+      "step": 266
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9848666599031698e-05,
+      "loss": 0.5958,
+      "step": 267
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.984689598592727e-05,
+      "loss": 0.6019,
+      "step": 268
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.984511515471882e-05,
+      "loss": 0.6043,
+      "step": 269
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.984332410725432e-05,
+      "loss": 0.5951,
+      "step": 270
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9841522845392322e-05,
+      "loss": 0.5682,
+      "step": 271
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9839711371002006e-05,
+      "loss": 0.5719,
+      "step": 272
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.983788968596313e-05,
+      "loss": 0.5692,
+      "step": 273
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9836057792166046e-05,
+      "loss": 0.5636,
+      "step": 274
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.983421569151172e-05,
+      "loss": 0.5984,
+      "step": 275
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9832363385911682e-05,
+      "loss": 0.5976,
+      "step": 276
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.983050087728808e-05,
+      "loss": 0.5512,
+      "step": 277
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9828628167573626e-05,
+      "loss": 0.59,
+      "step": 278
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.982674525871163e-05,
+      "loss": 0.6002,
+      "step": 279
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.982485215265598e-05,
+      "loss": 0.581,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.982294885137115e-05,
+      "loss": 0.6359,
+      "step": 281
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9821035356832193e-05,
+      "loss": 0.6287,
+      "step": 282
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9819111671024737e-05,
+      "loss": 0.5836,
+      "step": 283
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.981717779594499e-05,
+      "loss": 0.6338,
+      "step": 284
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9815233733599725e-05,
+      "loss": 0.6441,
+      "step": 285
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9813279486006297e-05,
+      "loss": 0.5829,
+      "step": 286
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9811315055192618e-05,
+      "loss": 0.5692,
+      "step": 287
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.980934044319718e-05,
+      "loss": 0.5854,
+      "step": 288
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9807355652069032e-05,
+      "loss": 0.5706,
+      "step": 289
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.980536068386779e-05,
+      "loss": 0.5953,
+      "step": 290
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.980335554066362e-05,
+      "loss": 0.5619,
+      "step": 291
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9801340224537266e-05,
+      "loss": 0.6009,
+      "step": 292
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.979931473758001e-05,
+      "loss": 0.6175,
+      "step": 293
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.97972790818937e-05,
+      "loss": 0.5911,
+      "step": 294
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9795233259590733e-05,
+      "loss": 0.5678,
+      "step": 295
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9793177272794044e-05,
+      "loss": 0.5768,
+      "step": 296
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.979111112363714e-05,
+      "loss": 0.591,
+      "step": 297
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.978903481426405e-05,
+      "loss": 0.5897,
+      "step": 298
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9786948346829363e-05,
+      "loss": 0.572,
+      "step": 299
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.97848517234982e-05,
+      "loss": 0.5882,
+      "step": 300
+    },
+    {
+      "epoch": 0.28,
+      "eval_loss": 0.5981120467185974,
+      "eval_runtime": 98.3541,
+      "eval_samples_per_second": 7.036,
+      "eval_steps_per_second": 0.447,
+      "step": 300
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9782744946446216e-05,
+      "loss": 0.5629,
+      "step": 301
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.978062801785962e-05,
+      "loss": 0.5501,
+      "step": 302
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.977850093993514e-05,
+      "loss": 0.6137,
+      "step": 303
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9776363714880037e-05,
+      "loss": 0.5776,
+      "step": 304
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9774216344912106e-05,
+      "loss": 0.5619,
+      "step": 305
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9772058832259675e-05,
+      "loss": 0.5891,
+      "step": 306
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9769891179161584e-05,
+      "loss": 0.6069,
+      "step": 307
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9767713387867204e-05,
+      "loss": 0.6002,
+      "step": 308
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9765525460636423e-05,
+      "loss": 0.6514,
+      "step": 309
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9763327399739654e-05,
+      "loss": 0.6539,
+      "step": 310
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9761119207457815e-05,
+      "loss": 0.5384,
+      "step": 311
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9758900886082343e-05,
+      "loss": 0.5719,
+      "step": 312
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9756672437915194e-05,
+      "loss": 0.6166,
+      "step": 313
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.975443386526881e-05,
+      "loss": 0.5764,
+      "step": 314
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9752185170466163e-05,
+      "loss": 0.5817,
+      "step": 315
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9749926355840716e-05,
+      "loss": 0.5919,
+      "step": 316
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9747657423736438e-05,
+      "loss": 0.5948,
+      "step": 317
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.974537837650779e-05,
+      "loss": 0.5908,
+      "step": 318
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9743089216519745e-05,
+      "loss": 0.5959,
+      "step": 319
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9740789946147748e-05,
+      "loss": 0.5513,
+      "step": 320
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9738480567777758e-05,
+      "loss": 0.6033,
+      "step": 321
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9736161083806205e-05,
+      "loss": 0.5437,
+      "step": 322
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9733831496640014e-05,
+      "loss": 0.5903,
+      "step": 323
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9731491808696593e-05,
+      "loss": 0.6248,
+      "step": 324
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9729142022403834e-05,
+      "loss": 0.5799,
+      "step": 325
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9726782140200104e-05,
+      "loss": 0.5878,
+      "step": 326
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9724412164534246e-05,
+      "loss": 0.5698,
+      "step": 327
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.972203209786558e-05,
+      "loss": 0.5499,
+      "step": 328
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9719641942663898e-05,
+      "loss": 0.5743,
+      "step": 329
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.971724170140946e-05,
+      "loss": 0.5709,
+      "step": 330
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9714831376592987e-05,
+      "loss": 0.5838,
+      "step": 331
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9712410970715674e-05,
+      "loss": 0.5646,
+      "step": 332
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9709980486289165e-05,
+      "loss": 0.5821,
+      "step": 333
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.970753992583557e-05,
+      "loss": 0.5893,
+      "step": 334
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9705089291887457e-05,
+      "loss": 0.5818,
+      "step": 335
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9702628586987846e-05,
+      "loss": 0.6159,
+      "step": 336
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.97001578136902e-05,
+      "loss": 0.597,
+      "step": 337
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9697676974558434e-05,
+      "loss": 0.5776,
+      "step": 338
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9695186072166915e-05,
+      "loss": 0.5701,
+      "step": 339
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.969268510910045e-05,
+      "loss": 0.6197,
+      "step": 340
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9690174087954274e-05,
+      "loss": 0.5247,
+      "step": 341
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9687653011334076e-05,
+      "loss": 0.5455,
+      "step": 342
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9685121881855966e-05,
+      "loss": 0.5599,
+      "step": 343
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9682580702146497e-05,
+      "loss": 0.5966,
+      "step": 344
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9680029474842642e-05,
+      "loss": 0.6238,
+      "step": 345
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.967746820259181e-05,
+      "loss": 0.5181,
+      "step": 346
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9674896888051825e-05,
+      "loss": 0.5982,
+      "step": 347
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9672315533890934e-05,
+      "loss": 0.5558,
+      "step": 348
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9669724142787805e-05,
+      "loss": 0.6287,
+      "step": 349
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9667122717431516e-05,
+      "loss": 0.5811,
+      "step": 350
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9664511260521562e-05,
+      "loss": 0.5789,
+      "step": 351
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.966188977476785e-05,
+      "loss": 0.5449,
+      "step": 352
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9659258262890683e-05,
+      "loss": 0.5197,
+      "step": 353
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9656616727620785e-05,
+      "loss": 0.5649,
+      "step": 354
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9653965171699267e-05,
+      "loss": 0.5597,
+      "step": 355
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.965130359787764e-05,
+      "loss": 0.5642,
+      "step": 356
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.964863200891782e-05,
+      "loss": 0.5938,
+      "step": 357
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9645950407592106e-05,
+      "loss": 0.5417,
+      "step": 358
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9643258796683192e-05,
+      "loss": 0.5909,
+      "step": 359
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9640557178984152e-05,
+      "loss": 0.5885,
+      "step": 360
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.963784555729846e-05,
+      "loss": 0.5942,
+      "step": 361
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9635123934439945e-05,
+      "loss": 0.5829,
+      "step": 362
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9632392313232845e-05,
+      "loss": 0.5843,
+      "step": 363
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.962965069651175e-05,
+      "loss": 0.6169,
+      "step": 364
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9626899087121625e-05,
+      "loss": 0.5482,
+      "step": 365
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9624137487917824e-05,
+      "loss": 0.6272,
+      "step": 366
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9621365901766037e-05,
+      "loss": 0.5752,
+      "step": 367
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9618584331542345e-05,
+      "loss": 0.6456,
+      "step": 368
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9615792780133175e-05,
+      "loss": 0.6172,
+      "step": 369
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.961299125043531e-05,
+      "loss": 0.6144,
+      "step": 370
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.961017974535589e-05,
+      "loss": 0.5861,
+      "step": 371
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.960735826781242e-05,
+      "loss": 0.619,
+      "step": 372
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9604526820732733e-05,
+      "loss": 0.5604,
+      "step": 373
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9601685407055016e-05,
+      "loss": 0.5582,
+      "step": 374
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.95988340297278e-05,
+      "loss": 0.5521,
+      "step": 375
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.959597269170995e-05,
+      "loss": 0.5315,
+      "step": 376
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9593101395970674e-05,
+      "loss": 0.5407,
+      "step": 377
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9590220145489512e-05,
+      "loss": 0.6201,
+      "step": 378
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.958732894325633e-05,
+      "loss": 0.5554,
+      "step": 379
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9584427792271317e-05,
+      "loss": 0.5616,
+      "step": 380
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9581516695544997e-05,
+      "loss": 0.6026,
+      "step": 381
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9578595656098208e-05,
+      "loss": 0.5618,
+      "step": 382
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.957566467696211e-05,
+      "loss": 0.5805,
+      "step": 383
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9572723761178168e-05,
+      "loss": 0.62,
+      "step": 384
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.956977291179817e-05,
+      "loss": 0.625,
+      "step": 385
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9566812131884202e-05,
+      "loss": 0.5864,
+      "step": 386
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9563841424508664e-05,
+      "loss": 0.5998,
+      "step": 387
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9560860792754248e-05,
+      "loss": 0.597,
+      "step": 388
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9557870239713952e-05,
+      "loss": 0.5672,
+      "step": 389
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9554869768491065e-05,
+      "loss": 0.5854,
+      "step": 390
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9551859382199175e-05,
+      "loss": 0.6149,
+      "step": 391
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.954883908396215e-05,
+      "loss": 0.6169,
+      "step": 392
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9545808876914142e-05,
+      "loss": 0.587,
+      "step": 393
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9542768764199604e-05,
+      "loss": 0.6126,
+      "step": 394
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9539718748973242e-05,
+      "loss": 0.5933,
+      "step": 395
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9536658834400058e-05,
+      "loss": 0.618,
+      "step": 396
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9533589023655318e-05,
+      "loss": 0.5953,
+      "step": 397
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9530509319924554e-05,
+      "loss": 0.5976,
+      "step": 398
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9527419726403578e-05,
+      "loss": 0.6189,
+      "step": 399
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.952432024629844e-05,
+      "loss": 0.54,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "eval_loss": 0.5913203954696655,
+      "eval_runtime": 98.4678,
+      "eval_samples_per_second": 7.028,
+      "eval_steps_per_second": 0.447,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9521210882825483e-05,
+      "loss": 0.5819,
+      "step": 401
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.951809163921127e-05,
+      "loss": 0.5649,
+      "step": 402
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.951496251869264e-05,
+      "loss": 0.6142,
+      "step": 403
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9511823524516677e-05,
+      "loss": 0.5937,
+      "step": 404
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.950867465994071e-05,
+      "loss": 0.5806,
+      "step": 405
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9505515928232306e-05,
+      "loss": 0.5451,
+      "step": 406
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9502347332669276e-05,
+      "loss": 0.5579,
+      "step": 407
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9499168876539666e-05,
+      "loss": 0.5738,
+      "step": 408
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.949598056314175e-05,
+      "loss": 0.5665,
+      "step": 409
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9492782395784037e-05,
+      "loss": 0.5751,
+      "step": 410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.948957437778526e-05,
+      "loss": 0.6239,
+      "step": 411
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.948635651247437e-05,
+      "loss": 0.5535,
+      "step": 412
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.948312880319054e-05,
+      "loss": 0.5865,
+      "step": 413
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9479891253283158e-05,
+      "loss": 0.6061,
+      "step": 414
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9476643866111822e-05,
+      "loss": 0.6486,
+      "step": 415
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9473386645046344e-05,
+      "loss": 0.5527,
+      "step": 416
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9470119593466727e-05,
+      "loss": 0.5629,
+      "step": 417
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9466842714763192e-05,
+      "loss": 0.5948,
+      "step": 418
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9463556012336148e-05,
+      "loss": 0.5656,
+      "step": 419
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9460259489596192e-05,
+      "loss": 0.5688,
+      "step": 420
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9456953149964126e-05,
+      "loss": 0.6054,
+      "step": 421
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9453636996870932e-05,
+      "loss": 0.5693,
+      "step": 422
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9450311033757772e-05,
+      "loss": 0.5942,
+      "step": 423
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9446975264075993e-05,
+      "loss": 0.566,
+      "step": 424
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9443629691287116e-05,
+      "loss": 0.5768,
+      "step": 425
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9440274318862835e-05,
+      "loss": 0.5814,
+      "step": 426
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.943690915028502e-05,
+      "loss": 0.6063,
+      "step": 427
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.943353418904568e-05,
+      "loss": 0.6333,
+      "step": 428
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.943014943864703e-05,
+      "loss": 0.5858,
+      "step": 429
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9426754902601405e-05,
+      "loss": 0.5447,
+      "step": 430
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9423350584431308e-05,
+      "loss": 0.5683,
+      "step": 431
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9419936487669396e-05,
+      "loss": 0.5727,
+      "step": 432
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9416512615858468e-05,
+      "loss": 0.5733,
+      "step": 433
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9413078972551472e-05,
+      "loss": 0.5554,
+      "step": 434
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9409635561311492e-05,
+      "loss": 0.6282,
+      "step": 435
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.940618238571175e-05,
+      "loss": 0.5854,
+      "step": 436
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9402719449335595e-05,
+      "loss": 0.5892,
+      "step": 437
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9399246755776512e-05,
+      "loss": 0.5563,
+      "step": 438
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.939576430863811e-05,
+      "loss": 0.6043,
+      "step": 439
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.939227211153411e-05,
+      "loss": 0.4823,
+      "step": 440
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.938877016808837e-05,
+      "loss": 0.5842,
+      "step": 441
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9385258481934836e-05,
+      "loss": 0.582,
+      "step": 442
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.938173705671759e-05,
+      "loss": 0.5391,
+      "step": 443
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9378205896090796e-05,
+      "loss": 0.5581,
+      "step": 444
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.937466500371874e-05,
+      "loss": 0.5722,
+      "step": 445
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9371114383275798e-05,
+      "loss": 0.605,
+      "step": 446
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9367554038446437e-05,
+      "loss": 0.5818,
+      "step": 447
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.936398397292523e-05,
+      "loss": 0.5634,
+      "step": 448
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9360404190416818e-05,
+      "loss": 0.5855,
+      "step": 449
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9356814694635934e-05,
+      "loss": 0.6026,
+      "step": 450
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.935321548930739e-05,
+      "loss": 0.5612,
+      "step": 451
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.934960657816608e-05,
+      "loss": 0.6043,
+      "step": 452
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.934598796495696e-05,
+      "loss": 0.6415,
+      "step": 453
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.934235965343506e-05,
+      "loss": 0.5513,
+      "step": 454
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9338721647365465e-05,
+      "loss": 0.5614,
+      "step": 455
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9335073950523335e-05,
+      "loss": 0.5844,
+      "step": 456
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.933141656669387e-05,
+      "loss": 0.5609,
+      "step": 457
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9327749499672337e-05,
+      "loss": 0.6231,
+      "step": 458
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.932407275326404e-05,
+      "loss": 0.5584,
+      "step": 459
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9320386331284334e-05,
+      "loss": 0.6077,
+      "step": 460
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9316690237558603e-05,
+      "loss": 0.5578,
+      "step": 461
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9312984475922286e-05,
+      "loss": 0.5803,
+      "step": 462
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.930926905022084e-05,
+      "loss": 0.5313,
+      "step": 463
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9305543964309756e-05,
+      "loss": 0.5843,
+      "step": 464
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.930180922205455e-05,
+      "loss": 0.5592,
+      "step": 465
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.929806482733075e-05,
+      "loss": 0.5918,
+      "step": 466
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9294310784023915e-05,
+      "loss": 0.5871,
+      "step": 467
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.92905470960296e-05,
+      "loss": 0.6113,
+      "step": 468
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9286773767253383e-05,
+      "loss": 0.572,
+      "step": 469
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9282990801610837e-05,
+      "loss": 0.5669,
+      "step": 470
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9279198203027537e-05,
+      "loss": 0.5614,
+      "step": 471
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9275395975439054e-05,
+      "loss": 0.5682,
+      "step": 472
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9271584122790958e-05,
+      "loss": 0.5994,
+      "step": 473
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9267762649038794e-05,
+      "loss": 0.5869,
+      "step": 474
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9263931558148097e-05,
+      "loss": 0.6193,
+      "step": 475
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9260090854094388e-05,
+      "loss": 0.5862,
+      "step": 476
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9256240540863158e-05,
+      "loss": 0.5877,
+      "step": 477
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.925238062244986e-05,
+      "loss": 0.5495,
+      "step": 478
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.924851110285993e-05,
+      "loss": 0.5445,
+      "step": 479
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9244631986108768e-05,
+      "loss": 0.5571,
+      "step": 480
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.924074327622171e-05,
+      "loss": 0.549,
+      "step": 481
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9236844977234074e-05,
+      "loss": 0.5768,
+      "step": 482
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9232937093191112e-05,
+      "loss": 0.5613,
+      "step": 483
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.922901962814803e-05,
+      "loss": 0.5232,
+      "step": 484
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9225092586169972e-05,
+      "loss": 0.5351,
+      "step": 485
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.922115597133202e-05,
+      "loss": 0.5689,
+      "step": 486
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.921720978771919e-05,
+      "loss": 0.6027,
+      "step": 487
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.921325403942644e-05,
+      "loss": 0.5553,
+      "step": 488
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9209288730558632e-05,
+      "loss": 0.5257,
+      "step": 489
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.920531386523056e-05,
+      "loss": 0.634,
+      "step": 490
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.920132944756693e-05,
+      "loss": 0.5452,
+      "step": 491
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9197335481702374e-05,
+      "loss": 0.5689,
+      "step": 492
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.919333197178141e-05,
+      "loss": 0.5362,
+      "step": 493
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9189318921958485e-05,
+      "loss": 0.5794,
+      "step": 494
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9185296336397922e-05,
+      "loss": 0.5753,
+      "step": 495
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9181264219273957e-05,
+      "loss": 0.5522,
+      "step": 496
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.917722257477071e-05,
+      "loss": 0.5864,
+      "step": 497
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9173171407082183e-05,
+      "loss": 0.5841,
+      "step": 498
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9169110720412262e-05,
+      "loss": 0.558,
+      "step": 499
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.916504051897472e-05,
+      "loss": 0.5482,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "eval_loss": 0.5859342217445374,
+      "eval_runtime": 98.5474,
+      "eval_samples_per_second": 7.022,
+      "eval_steps_per_second": 0.446,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9160960806993197e-05,
+      "loss": 0.5682,
+      "step": 501
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9156871588701197e-05,
+      "loss": 0.5551,
+      "step": 502
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9152772868342095e-05,
+      "loss": 0.5835,
+      "step": 503
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9148664650169128e-05,
+      "loss": 0.6197,
+      "step": 504
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9144546938445382e-05,
+      "loss": 0.6248,
+      "step": 505
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9140419737443797e-05,
+      "loss": 0.6037,
+      "step": 506
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9136283051447164e-05,
+      "loss": 0.5955,
+      "step": 507
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9132136884748114e-05,
+      "loss": 0.5565,
+      "step": 508
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9127981241649113e-05,
+      "loss": 0.6062,
+      "step": 509
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9123816126462468e-05,
+      "loss": 0.5215,
+      "step": 510
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9119641543510306e-05,
+      "loss": 0.5553,
+      "step": 511
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9115457497124588e-05,
+      "loss": 0.5493,
+      "step": 512
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9111263991647086e-05,
+      "loss": 0.6129,
+      "step": 513
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.910706103142939e-05,
+      "loss": 0.5313,
+      "step": 514
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.910284862083291e-05,
+      "loss": 0.5578,
+      "step": 515
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9098626764228854e-05,
+      "loss": 0.5609,
+      "step": 516
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9094395465998233e-05,
+      "loss": 0.5836,
+      "step": 517
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9090154730531856e-05,
+      "loss": 0.5255,
+      "step": 518
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9085904562230327e-05,
+      "loss": 0.6473,
+      "step": 519
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9081644965504035e-05,
+      "loss": 0.5163,
+      "step": 520
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9077375944773156e-05,
+      "loss": 0.6049,
+      "step": 521
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9073097504467643e-05,
+      "loss": 0.5387,
+      "step": 522
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.906880964902723e-05,
+      "loss": 0.5614,
+      "step": 523
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.906451238290141e-05,
+      "loss": 0.5294,
+      "step": 524
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9060205710549446e-05,
+      "loss": 0.5636,
+      "step": 525
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.905588963644037e-05,
+      "loss": 0.5605,
+      "step": 526
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.905156416505296e-05,
+      "loss": 0.5989,
+      "step": 527
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.904722930087575e-05,
+      "loss": 0.5652,
+      "step": 528
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9042885048407017e-05,
+      "loss": 0.6249,
+      "step": 529
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9038531412154786e-05,
+      "loss": 0.5731,
+      "step": 530
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9034168396636815e-05,
+      "loss": 0.5145,
+      "step": 531
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9029796006380595e-05,
+      "loss": 0.531,
+      "step": 532
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.902541424592335e-05,
+      "loss": 0.5742,
+      "step": 533
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.902102311981202e-05,
+      "loss": 0.5841,
+      "step": 534
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.901662263260327e-05,
+      "loss": 0.5879,
+      "step": 535
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9012212788863475e-05,
+      "loss": 0.5815,
+      "step": 536
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9007793593168725e-05,
+      "loss": 0.6399,
+      "step": 537
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9003365050104803e-05,
+      "loss": 0.5358,
+      "step": 538
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8998927164267213e-05,
+      "loss": 0.564,
+      "step": 539
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8994479940261122e-05,
+      "loss": 0.5736,
+      "step": 540
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8990023382701418e-05,
+      "loss": 0.5332,
+      "step": 541
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8985557496212655e-05,
+      "loss": 0.5321,
+      "step": 542
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8981082285429084e-05,
+      "loss": 0.5531,
+      "step": 543
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8976597754994614e-05,
+      "loss": 0.5611,
+      "step": 544
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.897210390956283e-05,
+      "loss": 0.6184,
+      "step": 545
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8967600753796993e-05,
+      "loss": 0.5792,
+      "step": 546
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.896308829237002e-05,
+      "loss": 0.6083,
+      "step": 547
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8958566529964475e-05,
+      "loss": 0.5768,
+      "step": 548
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.895403547127259e-05,
+      "loss": 0.5414,
+      "step": 549
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.894949512099623e-05,
+      "loss": 0.5255,
+      "step": 550
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8944945483846905e-05,
+      "loss": 0.6145,
+      "step": 551
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8940386564545773e-05,
+      "loss": 0.6404,
+      "step": 552
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8935818367823612e-05,
+      "loss": 0.6009,
+      "step": 553
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8931240898420825e-05,
+      "loss": 0.6181,
+      "step": 554
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8926654161087452e-05,
+      "loss": 0.5804,
+      "step": 555
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8922058160583133e-05,
+      "loss": 0.5564,
+      "step": 556
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8917452901677135e-05,
+      "loss": 0.6215,
+      "step": 557
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8912838389148315e-05,
+      "loss": 0.5612,
+      "step": 558
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8908214627785155e-05,
+      "loss": 0.6129,
+      "step": 559
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.890358162238572e-05,
+      "loss": 0.5796,
+      "step": 560
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8898939377757666e-05,
+      "loss": 0.5629,
+      "step": 561
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8894287898718245e-05,
+      "loss": 0.5834,
+      "step": 562
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8889627190094285e-05,
+      "loss": 0.6389,
+      "step": 563
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8884957256722197e-05,
+      "loss": 0.5797,
+      "step": 564
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8880278103447957e-05,
+      "loss": 0.5707,
+      "step": 565
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8875589735127114e-05,
+      "loss": 0.5236,
+      "step": 566
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8870892156624778e-05,
+      "loss": 0.597,
+      "step": 567
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.886618537281562e-05,
+      "loss": 0.5582,
+      "step": 568
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.886146938858386e-05,
+      "loss": 0.5816,
+      "step": 569
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8856744208823263e-05,
+      "loss": 0.5846,
+      "step": 570
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8852009838437138e-05,
+      "loss": 0.5707,
+      "step": 571
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8847266282338333e-05,
+      "loss": 0.5756,
+      "step": 572
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8842513545449225e-05,
+      "loss": 0.5548,
+      "step": 573
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.883775163270172e-05,
+      "loss": 0.5167,
+      "step": 574
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.883298054903725e-05,
+      "loss": 0.5867,
+      "step": 575
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8828200299406747e-05,
+      "loss": 0.5603,
+      "step": 576
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.882341088877068e-05,
+      "loss": 0.6117,
+      "step": 577
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8818612322099002e-05,
+      "loss": 0.5951,
+      "step": 578
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8813804604371174e-05,
+      "loss": 0.5836,
+      "step": 579
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8808987740576166e-05,
+      "loss": 0.5795,
+      "step": 580
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.880416173571242e-05,
+      "loss": 0.5583,
+      "step": 581
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8799326594787873e-05,
+      "loss": 0.5554,
+      "step": 582
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.879448232281994e-05,
+      "loss": 0.5935,
+      "step": 583
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8789628924835512e-05,
+      "loss": 0.5972,
+      "step": 584
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.878476640587095e-05,
+      "loss": 0.5644,
+      "step": 585
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8779894770972085e-05,
+      "loss": 0.5458,
+      "step": 586
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8775014025194196e-05,
+      "loss": 0.5892,
+      "step": 587
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8770124173602026e-05,
+      "loss": 0.5848,
+      "step": 588
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8765225221269766e-05,
+      "loss": 0.5943,
+      "step": 589
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8760317173281044e-05,
+      "loss": 0.5725,
+      "step": 590
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8755400034728937e-05,
+      "loss": 0.6062,
+      "step": 591
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8750473810715944e-05,
+      "loss": 0.5403,
+      "step": 592
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8745538506354e-05,
+      "loss": 0.5488,
+      "step": 593
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.874059412676447e-05,
+      "loss": 0.5587,
+      "step": 594
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8735640677078107e-05,
+      "loss": 0.5366,
+      "step": 595
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8730678162435112e-05,
+      "loss": 0.5971,
+      "step": 596
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.872570658798507e-05,
+      "loss": 0.5939,
+      "step": 597
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8720725958886975e-05,
+      "loss": 0.6264,
+      "step": 598
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8715736280309212e-05,
+      "loss": 0.5661,
+      "step": 599
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.871073755742957e-05,
+      "loss": 0.6252,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "eval_loss": 0.5812298655509949,
+      "eval_runtime": 98.3288,
+      "eval_samples_per_second": 7.038,
+      "eval_steps_per_second": 0.447,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8705729795435202e-05,
+      "loss": 0.5515,
+      "step": 601
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8700712999522657e-05,
+      "loss": 0.5785,
+      "step": 602
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8695687174897862e-05,
+      "loss": 0.5765,
+      "step": 603
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8690652326776088e-05,
+      "loss": 0.5426,
+      "step": 604
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8685608460382002e-05,
+      "loss": 0.5697,
+      "step": 605
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8680555580949608e-05,
+      "loss": 0.5697,
+      "step": 606
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.867549369372227e-05,
+      "loss": 0.5621,
+      "step": 607
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8670422803952696e-05,
+      "loss": 0.5661,
+      "step": 608
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.866534291690294e-05,
+      "loss": 0.5666,
+      "step": 609
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 0.6096,
+      "step": 610
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8655156172057762e-05,
+      "loss": 0.6008,
+      "step": 611
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8650049324833107e-05,
+      "loss": 0.5491,
+      "step": 612
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8644933501469785e-05,
+      "loss": 0.5467,
+      "step": 613
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.863980870727648e-05,
+      "loss": 0.5452,
+      "step": 614
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.863467494757118e-05,
+      "loss": 0.5977,
+      "step": 615
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8629532227681175e-05,
+      "loss": 0.5929,
+      "step": 616
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8624380552943053e-05,
+      "loss": 0.5645,
+      "step": 617
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.86192199287027e-05,
+      "loss": 0.5638,
+      "step": 618
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8614050360315285e-05,
+      "loss": 0.6028,
+      "step": 619
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8608871853145257e-05,
+      "loss": 0.6058,
+      "step": 620
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8603684412566346e-05,
+      "loss": 0.6176,
+      "step": 621
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.859848804396154e-05,
+      "loss": 0.5663,
+      "step": 622
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8593282752723114e-05,
+      "loss": 0.6204,
+      "step": 623
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8588068544252572e-05,
+      "loss": 0.5066,
+      "step": 624
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8582845423960703e-05,
+      "loss": 0.5631,
+      "step": 625
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8577613397267518e-05,
+      "loss": 0.5109,
+      "step": 626
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8572372469602285e-05,
+      "loss": 0.5734,
+      "step": 627
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.85671226464035e-05,
+      "loss": 0.5528,
+      "step": 628
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.85618639331189e-05,
+      "loss": 0.589,
+      "step": 629
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8556596335205432e-05,
+      "loss": 0.586,
+      "step": 630
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.855131985812928e-05,
+      "loss": 0.5878,
+      "step": 631
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8546034507365832e-05,
+      "loss": 0.5807,
+      "step": 632
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8540740288399685e-05,
+      "loss": 0.5277,
+      "step": 633
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8535437206724642e-05,
+      "loss": 0.5955,
+      "step": 634
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8530125267843692e-05,
+      "loss": 0.5919,
+      "step": 635
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.852480447726903e-05,
+      "loss": 0.6462,
+      "step": 636
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.851947484052203e-05,
+      "loss": 0.5508,
+      "step": 637
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.851413636313324e-05,
+      "loss": 0.5659,
+      "step": 638
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.850878905064239e-05,
+      "loss": 0.5621,
+      "step": 639
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.850343290859838e-05,
+      "loss": 0.5726,
+      "step": 640
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8498067942559258e-05,
+      "loss": 0.5861,
+      "step": 641
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.849269415809225e-05,
+      "loss": 0.5666,
+      "step": 642
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.848731156077371e-05,
+      "loss": 0.5669,
+      "step": 643
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8481920156189157e-05,
+      "loss": 0.6032,
+      "step": 644
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8476519949933234e-05,
+      "loss": 0.5514,
+      "step": 645
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8471110947609728e-05,
+      "loss": 0.5775,
+      "step": 646
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.846569315483155e-05,
+      "loss": 0.5876,
+      "step": 647
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8460266577220733e-05,
+      "loss": 0.5553,
+      "step": 648
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.845483122040842e-05,
+      "loss": 0.5381,
+      "step": 649
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.844938709003488e-05,
+      "loss": 0.5954,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8443934191749466e-05,
+      "loss": 0.5495,
+      "step": 651
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8438472531210645e-05,
+      "loss": 0.5497,
+      "step": 652
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8433002114085966e-05,
+      "loss": 0.5824,
+      "step": 653
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8427522946052077e-05,
+      "loss": 0.5554,
+      "step": 654
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8422035032794693e-05,
+      "loss": 0.5746,
+      "step": 655
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8416538380008616e-05,
+      "loss": 0.5362,
+      "step": 656
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8411032993397708e-05,
+      "loss": 0.5981,
+      "step": 657
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.84055188786749e-05,
+      "loss": 0.5636,
+      "step": 658
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8399996041562174e-05,
+      "loss": 0.5736,
+      "step": 659
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.839446448779057e-05,
+      "loss": 0.5547,
+      "step": 660
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8388924223100166e-05,
+      "loss": 0.5013,
+      "step": 661
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.838337525324009e-05,
+      "loss": 0.56,
+      "step": 662
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.837781758396849e-05,
+      "loss": 0.531,
+      "step": 663
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8372251221052556e-05,
+      "loss": 0.5816,
+      "step": 664
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8366676170268482e-05,
+      "loss": 0.5402,
+      "step": 665
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8361092437401496e-05,
+      "loss": 0.5886,
+      "step": 666
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8355500028245822e-05,
+      "loss": 0.5915,
+      "step": 667
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8349898948604686e-05,
+      "loss": 0.5771,
+      "step": 668
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.834428920429033e-05,
+      "loss": 0.5671,
+      "step": 669
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8338670801123964e-05,
+      "loss": 0.5582,
+      "step": 670
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8333043744935797e-05,
+      "loss": 0.5921,
+      "step": 671
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8327408041565013e-05,
+      "loss": 0.5953,
+      "step": 672
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8321763696859772e-05,
+      "loss": 0.5167,
+      "step": 673
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8316110716677195e-05,
+      "loss": 0.556,
+      "step": 674
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.831044910688337e-05,
+      "loss": 0.5609,
+      "step": 675
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8304778873353345e-05,
+      "loss": 0.5767,
+      "step": 676
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.82991000219711e-05,
+      "loss": 0.534,
+      "step": 677
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8293412558629572e-05,
+      "loss": 0.5632,
+      "step": 678
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8287716489230626e-05,
+      "loss": 0.5917,
+      "step": 679
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8282011819685064e-05,
+      "loss": 0.5807,
+      "step": 680
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8276298555912613e-05,
+      "loss": 0.6039,
+      "step": 681
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8270576703841918e-05,
+      "loss": 0.51,
+      "step": 682
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8264846269410522e-05,
+      "loss": 0.5641,
+      "step": 683
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8259107258564897e-05,
+      "loss": 0.5617,
+      "step": 684
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8253359677260394e-05,
+      "loss": 0.5738,
+      "step": 685
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.824760353146127e-05,
+      "loss": 0.599,
+      "step": 686
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.824183882714067e-05,
+      "loss": 0.5792,
+      "step": 687
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.823606557028061e-05,
+      "loss": 0.5716,
+      "step": 688
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8230283766871986e-05,
+      "loss": 0.6217,
+      "step": 689
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8224493422914568e-05,
+      "loss": 0.5533,
+      "step": 690
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8218694544416985e-05,
+      "loss": 0.5485,
+      "step": 691
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.821288713739672e-05,
+      "loss": 0.5642,
+      "step": 692
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.82070712078801e-05,
+      "loss": 0.558,
+      "step": 693
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8201246761902318e-05,
+      "loss": 0.5973,
+      "step": 694
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8195413805507378e-05,
+      "loss": 0.5894,
+      "step": 695
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.818957234474813e-05,
+      "loss": 0.557,
+      "step": 696
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.818372238568625e-05,
+      "loss": 0.5588,
+      "step": 697
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8177863934392225e-05,
+      "loss": 0.5887,
+      "step": 698
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.817199699694536e-05,
+      "loss": 0.5801,
+      "step": 699
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.816612157943376e-05,
+      "loss": 0.5516,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "eval_loss": 0.5751967430114746,
+      "eval_runtime": 98.4652,
+      "eval_samples_per_second": 7.028,
+      "eval_steps_per_second": 0.447,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8160237687954336e-05,
+      "loss": 0.5639,
+      "step": 701
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.815434532861279e-05,
+      "loss": 0.5219,
+      "step": 702
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8148444507523617e-05,
+      "loss": 0.572,
+      "step": 703
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8142535230810078e-05,
+      "loss": 0.5793,
+      "step": 704
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8136617504604227e-05,
+      "loss": 0.5591,
+      "step": 705
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8130691335046865e-05,
+      "loss": 0.5816,
+      "step": 706
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8124756728287574e-05,
+      "loss": 0.5319,
+      "step": 707
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8118813690484686e-05,
+      "loss": 0.5282,
+      "step": 708
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.811286222780527e-05,
+      "loss": 0.6021,
+      "step": 709
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8106902346425156e-05,
+      "loss": 0.5507,
+      "step": 710
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8100934052528895e-05,
+      "loss": 0.6193,
+      "step": 711
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8094957352309773e-05,
+      "loss": 0.6059,
+      "step": 712
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8088972251969804e-05,
+      "loss": 0.6059,
+      "step": 713
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8082978757719712e-05,
+      "loss": 0.5644,
+      "step": 714
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8076976875778936e-05,
+      "loss": 0.54,
+      "step": 715
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8070966612375612e-05,
+      "loss": 0.6267,
+      "step": 716
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.806494797374658e-05,
+      "loss": 0.5758,
+      "step": 717
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8058920966137376e-05,
+      "loss": 0.5909,
+      "step": 718
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8052885595802206e-05,
+      "loss": 0.5816,
+      "step": 719
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8046841869003962e-05,
+      "loss": 0.5841,
+      "step": 720
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.804078979201421e-05,
+      "loss": 0.5332,
+      "step": 721
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8034729371113172e-05,
+      "loss": 0.5583,
+      "step": 722
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8028660612589745e-05,
+      "loss": 0.5751,
+      "step": 723
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8022583522741456e-05,
+      "loss": 0.5597,
+      "step": 724
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.80164981078745e-05,
+      "loss": 0.5712,
+      "step": 725
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.801040437430369e-05,
+      "loss": 0.5534,
+      "step": 726
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8004302328352487e-05,
+      "loss": 0.5057,
+      "step": 727
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.799819197635297e-05,
+      "loss": 0.5597,
+      "step": 728
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.7992073324645834e-05,
+      "loss": 0.629,
+      "step": 729
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.7985946379580403e-05,
+      "loss": 0.5792,
+      "step": 730
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.7979811147514586e-05,
+      "loss": 0.5347,
+      "step": 731
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.79736676348149e-05,
+      "loss": 0.6113,
+      "step": 732
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.796751584785646e-05,
+      "loss": 0.5606,
+      "step": 733
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.7961355793022966e-05,
+      "loss": 0.5367,
+      "step": 734
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.7955187476706692e-05,
+      "loss": 0.5819,
+      "step": 735
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.794901090530848e-05,
+      "loss": 0.6074,
+      "step": 736
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.794282608523775e-05,
+      "loss": 0.5375,
+      "step": 737
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7936633022912482e-05,
+      "loss": 0.5317,
+      "step": 738
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7930431724759203e-05,
+      "loss": 0.5551,
+      "step": 739
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7924222197212984e-05,
+      "loss": 0.5839,
+      "step": 740
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.791800444671744e-05,
+      "loss": 0.5848,
+      "step": 741
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7911778479724716e-05,
+      "loss": 0.5511,
+      "step": 742
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.790554430269549e-05,
+      "loss": 0.5907,
+      "step": 743
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7899301922098958e-05,
+      "loss": 0.5771,
+      "step": 744
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7893051344412815e-05,
+      "loss": 0.5675,
+      "step": 745
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7886792576123277e-05,
+      "loss": 0.5859,
+      "step": 746
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.788052562372506e-05,
+      "loss": 0.5992,
+      "step": 747
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7874250493721362e-05,
+      "loss": 0.5783,
+      "step": 748
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.786796719262388e-05,
+      "loss": 0.5282,
+      "step": 749
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7861675726952773e-05,
+      "loss": 0.5974,
+      "step": 750
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7855376103236687e-05,
+      "loss": 0.5755,
+      "step": 751
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7849068328012734e-05,
+      "loss": 0.5513,
+      "step": 752
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.784275240782647e-05,
+      "loss": 0.618,
+      "step": 753
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7836428349231915e-05,
+      "loss": 0.5876,
+      "step": 754
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.783009615879154e-05,
+      "loss": 0.5945,
+      "step": 755
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7823755843076233e-05,
+      "loss": 0.5791,
+      "step": 756
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.781740740866534e-05,
+      "loss": 0.5523,
+      "step": 757
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.781105086214661e-05,
+      "loss": 0.561,
+      "step": 758
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7804686210116217e-05,
+      "loss": 0.5497,
+      "step": 759
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7798313459178754e-05,
+      "loss": 0.5568,
+      "step": 760
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7791932615947213e-05,
+      "loss": 0.6005,
+      "step": 761
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7785543687042974e-05,
+      "loss": 0.5212,
+      "step": 762
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.777914667909582e-05,
+      "loss": 0.5558,
+      "step": 763
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7772741598743913e-05,
+      "loss": 0.5773,
+      "step": 764
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7766328452633793e-05,
+      "loss": 0.5674,
+      "step": 765
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.775990724742037e-05,
+      "loss": 0.5559,
+      "step": 766
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.775347798976691e-05,
+      "loss": 0.5437,
+      "step": 767
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.774704068634504e-05,
+      "loss": 0.5938,
+      "step": 768
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7740595343834746e-05,
+      "loss": 0.6039,
+      "step": 769
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7734141968924338e-05,
+      "loss": 0.6154,
+      "step": 770
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7727680568310476e-05,
+      "loss": 0.5407,
+      "step": 771
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.772121114869814e-05,
+      "loss": 0.541,
+      "step": 772
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.771473371680063e-05,
+      "loss": 0.5306,
+      "step": 773
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7708248279339573e-05,
+      "loss": 0.5437,
+      "step": 774
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7701754843044884e-05,
+      "loss": 0.585,
+      "step": 775
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7695253414654796e-05,
+      "loss": 0.5802,
+      "step": 776
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7688744000915825e-05,
+      "loss": 0.5494,
+      "step": 777
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7682226608582777e-05,
+      "loss": 0.5799,
+      "step": 778
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7675701244418733e-05,
+      "loss": 0.5412,
+      "step": 779
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.766916791519506e-05,
+      "loss": 0.5561,
+      "step": 780
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7662626627691374e-05,
+      "loss": 0.554,
+      "step": 781
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.765607738869555e-05,
+      "loss": 0.5593,
+      "step": 782
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.764952020500373e-05,
+      "loss": 0.5583,
+      "step": 783
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7642955083420288e-05,
+      "loss": 0.5537,
+      "step": 784
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7636382030757833e-05,
+      "loss": 0.5322,
+      "step": 785
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7629801053837215e-05,
+      "loss": 0.5585,
+      "step": 786
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7623212159487492e-05,
+      "loss": 0.5761,
+      "step": 787
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7616615354545956e-05,
+      "loss": 0.5533,
+      "step": 788
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7610010645858092e-05,
+      "loss": 0.6236,
+      "step": 789
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7603398040277592e-05,
+      "loss": 0.5322,
+      "step": 790
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7596777544666347e-05,
+      "loss": 0.5869,
+      "step": 791
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.759014916589443e-05,
+      "loss": 0.5849,
+      "step": 792
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7583512910840096e-05,
+      "loss": 0.5746,
+      "step": 793
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7576868786389773e-05,
+      "loss": 0.5317,
+      "step": 794
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7570216799438058e-05,
+      "loss": 0.5084,
+      "step": 795
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7563556956887705e-05,
+      "loss": 0.5295,
+      "step": 796
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.755688926564961e-05,
+      "loss": 0.5491,
+      "step": 797
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7550213732642837e-05,
+      "loss": 0.5506,
+      "step": 798
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7543530364794564e-05,
+      "loss": 0.5624,
+      "step": 799
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7536839169040107e-05,
+      "loss": 0.5744,
+      "step": 800
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 0.5711074471473694,
+      "eval_runtime": 98.3186,
+      "eval_samples_per_second": 7.038,
+      "eval_steps_per_second": 0.448,
+      "step": 800
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7530140152322914e-05,
+      "loss": 0.537,
+      "step": 801
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7523433321594533e-05,
+      "loss": 0.5683,
+      "step": 802
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7516718683814632e-05,
+      "loss": 0.5938,
+      "step": 803
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7509996245950977e-05,
+      "loss": 0.5449,
+      "step": 804
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.750326601497943e-05,
+      "loss": 0.5212,
+      "step": 805
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7496527997883936e-05,
+      "loss": 0.5568,
+      "step": 806
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7489782201656522e-05,
+      "loss": 0.5302,
+      "step": 807
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7483028633297287e-05,
+      "loss": 0.5373,
+      "step": 808
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7476267299814393e-05,
+      "loss": 0.5704,
+      "step": 809
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7469498208224064e-05,
+      "loss": 0.5744,
+      "step": 810
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7462721365550576e-05,
+      "loss": 0.5463,
+      "step": 811
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.745593677882624e-05,
+      "loss": 0.5985,
+      "step": 812
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7449144455091415e-05,
+      "loss": 0.5272,
+      "step": 813
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7442344401394468e-05,
+      "loss": 0.58,
+      "step": 814
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7435536624791815e-05,
+      "loss": 0.5413,
+      "step": 815
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7428721132347863e-05,
+      "loss": 0.6074,
+      "step": 816
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.742189793113504e-05,
+      "loss": 0.5446,
+      "step": 817
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7415067028233768e-05,
+      "loss": 0.5378,
+      "step": 818
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7408228430732456e-05,
+      "loss": 0.5408,
+      "step": 819
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.740138214572751e-05,
+      "loss": 0.5625,
+      "step": 820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7394528180323297e-05,
+      "loss": 0.5833,
+      "step": 821
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7387666541632165e-05,
+      "loss": 0.5358,
+      "step": 822
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7380797236774433e-05,
+      "loss": 0.6098,
+      "step": 823
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7373920272878357e-05,
+      "loss": 0.5453,
+      "step": 824
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7367035657080148e-05,
+      "loss": 0.5288,
+      "step": 825
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7360143396523957e-05,
+      "loss": 0.5975,
+      "step": 826
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7353243498361873e-05,
+      "loss": 0.5055,
+      "step": 827
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.73463359697539e-05,
+      "loss": 0.5545,
+      "step": 828
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.733942081786797e-05,
+      "loss": 0.549,
+      "step": 829
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7332498049879925e-05,
+      "loss": 0.5206,
+      "step": 830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7325567672973503e-05,
+      "loss": 0.5628,
+      "step": 831
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.731862969434035e-05,
+      "loss": 0.5937,
+      "step": 832
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7311684121179978e-05,
+      "loss": 0.5614,
+      "step": 833
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7304730960699808e-05,
+      "loss": 0.6126,
+      "step": 834
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.729777022011511e-05,
+      "loss": 0.5883,
+      "step": 835
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.729080190664904e-05,
+      "loss": 0.5279,
+      "step": 836
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7283826027532603e-05,
+      "loss": 0.4975,
+      "step": 837
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.727684259000464e-05,
+      "loss": 0.5782,
+      "step": 838
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7269851601311865e-05,
+      "loss": 0.6091,
+      "step": 839
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7262853068708807e-05,
+      "loss": 0.5793,
+      "step": 840
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7255846999457828e-05,
+      "loss": 0.6473,
+      "step": 841
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7248833400829113e-05,
+      "loss": 0.5733,
+      "step": 842
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7241812280100664e-05,
+      "loss": 0.5713,
+      "step": 843
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7234783644558276e-05,
+      "loss": 0.5942,
+      "step": 844
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7227747501495553e-05,
+      "loss": 0.5634,
+      "step": 845
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7220703858213888e-05,
+      "loss": 0.5657,
+      "step": 846
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7213652722022454e-05,
+      "loss": 0.5706,
+      "step": 847
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.72065941002382e-05,
+      "loss": 0.578,
+      "step": 848
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7199528000185843e-05,
+      "loss": 0.6053,
+      "step": 849
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7192454429197865e-05,
+      "loss": 0.5268,
+      "step": 850
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7185373394614492e-05,
+      "loss": 0.6419,
+      "step": 851
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7178284903783698e-05,
+      "loss": 0.5771,
+      "step": 852
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7171188964061204e-05,
+      "loss": 0.538,
+      "step": 853
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.716408558281044e-05,
+      "loss": 0.5697,
+      "step": 854
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.715697476740258e-05,
+      "loss": 0.5882,
+      "step": 855
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.71498565252165e-05,
+      "loss": 0.5613,
+      "step": 856
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7142730863638784e-05,
+      "loss": 0.5116,
+      "step": 857
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.713559779006372e-05,
+      "loss": 0.582,
+      "step": 858
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7128457311893282e-05,
+      "loss": 0.5637,
+      "step": 859
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7121309436537125e-05,
+      "loss": 0.5551,
+      "step": 860
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.711415417141259e-05,
+      "loss": 0.6389,
+      "step": 861
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7106991523944684e-05,
+      "loss": 0.5288,
+      "step": 862
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7099821501566066e-05,
+      "loss": 0.5572,
+      "step": 863
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7092644111717052e-05,
+      "loss": 0.5429,
+      "step": 864
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7085459361845613e-05,
+      "loss": 0.5585,
+      "step": 865
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7078267259407345e-05,
+      "loss": 0.5374,
+      "step": 866
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7071067811865477e-05,
+      "loss": 0.5516,
+      "step": 867
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7063861026690863e-05,
+      "loss": 0.5464,
+      "step": 868
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.705664691136197e-05,
+      "loss": 0.5738,
+      "step": 869
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7049425473364866e-05,
+      "loss": 0.5332,
+      "step": 870
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7042196720193226e-05,
+      "loss": 0.5698,
+      "step": 871
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7034960659348314e-05,
+      "loss": 0.5944,
+      "step": 872
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7027717298338976e-05,
+      "loss": 0.591,
+      "step": 873
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.702046664468163e-05,
+      "loss": 0.5404,
+      "step": 874
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.7013208705900264e-05,
+      "loss": 0.5736,
+      "step": 875
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.7005943489526432e-05,
+      "loss": 0.5596,
+      "step": 876
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.699867100309923e-05,
+      "loss": 0.5405,
+      "step": 877
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.69913912541653e-05,
+      "loss": 0.6028,
+      "step": 878
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.698410425027883e-05,
+      "loss": 0.5792,
+      "step": 879
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.697680999900152e-05,
+      "loss": 0.5822,
+      "step": 880
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.696950850790261e-05,
+      "loss": 0.5607,
+      "step": 881
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.6962199784558835e-05,
+      "loss": 0.5861,
+      "step": 882
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.6954883836554444e-05,
+      "loss": 0.5676,
+      "step": 883
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.6947560671481182e-05,
+      "loss": 0.593,
+      "step": 884
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.6940230296938285e-05,
+      "loss": 0.6377,
+      "step": 885
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6932892720532463e-05,
+      "loss": 0.5205,
+      "step": 886
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.692554794987791e-05,
+      "loss": 0.5067,
+      "step": 887
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6918195992596274e-05,
+      "loss": 0.6044,
+      "step": 888
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6910836856316668e-05,
+      "loss": 0.5691,
+      "step": 889
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6903470548675654e-05,
+      "loss": 0.565,
+      "step": 890
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6896097077317236e-05,
+      "loss": 0.5827,
+      "step": 891
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6888716449892843e-05,
+      "loss": 0.52,
+      "step": 892
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6881328674061346e-05,
+      "loss": 0.5876,
+      "step": 893
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.687393375748902e-05,
+      "loss": 0.5625,
+      "step": 894
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6866531707849557e-05,
+      "loss": 0.5758,
+      "step": 895
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.685912253282405e-05,
+      "loss": 0.6238,
+      "step": 896
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.685170624010098e-05,
+      "loss": 0.5413,
+      "step": 897
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.684428283737622e-05,
+      "loss": 0.5402,
+      "step": 898
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6836852332353028e-05,
+      "loss": 0.5596,
+      "step": 899
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6829414732742013e-05,
+      "loss": 0.566,
+      "step": 900
+    },
+    {
+      "epoch": 0.85,
+      "eval_loss": 0.5677770376205444,
+      "eval_runtime": 98.4786,
+      "eval_samples_per_second": 7.027,
+      "eval_steps_per_second": 0.447,
+      "step": 900
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6821970046261166e-05,
+      "loss": 0.539,
+      "step": 901
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6814518280635818e-05,
+      "loss": 0.5555,
+      "step": 902
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.680705944359865e-05,
+      "loss": 0.6241,
+      "step": 903
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6799593542889686e-05,
+      "loss": 0.5491,
+      "step": 904
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6792120586256276e-05,
+      "loss": 0.5899,
+      "step": 905
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6784640581453086e-05,
+      "loss": 0.567,
+      "step": 906
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.677715353624211e-05,
+      "loss": 0.5559,
+      "step": 907
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6769659458392636e-05,
+      "loss": 0.5617,
+      "step": 908
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.676215835568125e-05,
+      "loss": 0.5907,
+      "step": 909
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6754650235891836e-05,
+      "loss": 0.5894,
+      "step": 910
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6747135106815554e-05,
+      "loss": 0.5952,
+      "step": 911
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6739612976250836e-05,
+      "loss": 0.5373,
+      "step": 912
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6732083852003383e-05,
+      "loss": 0.5474,
+      "step": 913
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6724547741886154e-05,
+      "loss": 0.5781,
+      "step": 914
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6717004653719348e-05,
+      "loss": 0.5935,
+      "step": 915
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.670945459533042e-05,
+      "loss": 0.5254,
+      "step": 916
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6701897574554048e-05,
+      "loss": 0.5605,
+      "step": 917
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6694333599232138e-05,
+      "loss": 0.5427,
+      "step": 918
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.668676267721381e-05,
+      "loss": 0.6043,
+      "step": 919
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6679184816355393e-05,
+      "loss": 0.5631,
+      "step": 920
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6671600024520423e-05,
+      "loss": 0.5819,
+      "step": 921
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.666400830957962e-05,
+      "loss": 0.5539,
+      "step": 922
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.665640967941089e-05,
+      "loss": 0.535,
+      "step": 923
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6648804141899316e-05,
+      "loss": 0.5455,
+      "step": 924
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6641191704937154e-05,
+      "loss": 0.5763,
+      "step": 925
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.663357237642381e-05,
+      "loss": 0.5259,
+      "step": 926
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6625946164265844e-05,
+      "loss": 0.5334,
+      "step": 927
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6618313076376966e-05,
+      "loss": 0.5205,
+      "step": 928
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.661067312067801e-05,
+      "loss": 0.5583,
+      "step": 929
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.660302630509695e-05,
+      "loss": 0.5419,
+      "step": 930
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6595372637568866e-05,
+      "loss": 0.5859,
+      "step": 931
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.658771212603595e-05,
+      "loss": 0.5572,
+      "step": 932
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6580044778447508e-05,
+      "loss": 0.5436,
+      "step": 933
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6572370602759926e-05,
+      "loss": 0.6003,
+      "step": 934
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.656468960693668e-05,
+      "loss": 0.5119,
+      "step": 935
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6557001798948324e-05,
+      "loss": 0.5342,
+      "step": 936
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6549307186772483e-05,
+      "loss": 0.5493,
+      "step": 937
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.654160577839384e-05,
+      "loss": 0.5049,
+      "step": 938
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.653389758180413e-05,
+      "loss": 0.5507,
+      "step": 939
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.652618260500213e-05,
+      "loss": 0.5508,
+      "step": 940
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.651846085599366e-05,
+      "loss": 0.5589,
+      "step": 941
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6510732342791566e-05,
+      "loss": 0.5043,
+      "step": 942
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6502997073415703e-05,
+      "loss": 0.5735,
+      "step": 943
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6495255055892948e-05,
+      "loss": 0.5402,
+      "step": 944
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6487506298257178e-05,
+      "loss": 0.5327,
+      "step": 945
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6479750808549263e-05,
+      "loss": 0.5555,
+      "step": 946
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6471988594817054e-05,
+      "loss": 0.5492,
+      "step": 947
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6464219665115392e-05,
+      "loss": 0.562,
+      "step": 948
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6456444027506074e-05,
+      "loss": 0.5365,
+      "step": 949
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6448661690057868e-05,
+      "loss": 0.5159,
+      "step": 950
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6440872660846487e-05,
+      "loss": 0.5558,
+      "step": 951
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6433076947954593e-05,
+      "loss": 0.591,
+      "step": 952
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6425274559471773e-05,
+      "loss": 0.5571,
+      "step": 953
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6417465503494562e-05,
+      "loss": 0.5504,
+      "step": 954
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6409649788126394e-05,
+      "loss": 0.5667,
+      "step": 955
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6401827421477625e-05,
+      "loss": 0.5456,
+      "step": 956
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6393998411665504e-05,
+      "loss": 0.5573,
+      "step": 957
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6386162766814182e-05,
+      "loss": 0.518,
+      "step": 958
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6378320495054688e-05,
+      "loss": 0.5527,
+      "step": 959
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.637047160452494e-05,
+      "loss": 0.5357,
+      "step": 960
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.636261610336971e-05,
+      "loss": 0.5911,
+      "step": 961
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6354753999740637e-05,
+      "loss": 0.5745,
+      "step": 962
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6346885301796206e-05,
+      "loss": 0.5909,
+      "step": 963
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6339010017701757e-05,
+      "loss": 0.5341,
+      "step": 964
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6331128155629446e-05,
+      "loss": 0.5578,
+      "step": 965
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6323239723758272e-05,
+      "loss": 0.5771,
+      "step": 966
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6315344730274046e-05,
+      "loss": 0.539,
+      "step": 967
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6307443183369382e-05,
+      "loss": 0.5707,
+      "step": 968
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6299535091243705e-05,
+      "loss": 0.5479,
+      "step": 969
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6291620462103222e-05,
+      "loss": 0.5931,
+      "step": 970
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6283699304160924e-05,
+      "loss": 0.5331,
+      "step": 971
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6275771625636592e-05,
+      "loss": 0.5524,
+      "step": 972
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.626783743475675e-05,
+      "loss": 0.5412,
+      "step": 973
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.62598967397547e-05,
+      "loss": 0.5271,
+      "step": 974
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.625194954887048e-05,
+      "loss": 0.4928,
+      "step": 975
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6243995870350875e-05,
+      "loss": 0.5172,
+      "step": 976
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6236035712449404e-05,
+      "loss": 0.5434,
+      "step": 977
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6228069083426304e-05,
+      "loss": 0.552,
+      "step": 978
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.622009599154853e-05,
+      "loss": 0.51,
+      "step": 979
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.621211644508974e-05,
+      "loss": 0.5655,
+      "step": 980
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.62041304523303e-05,
+      "loss": 0.5305,
+      "step": 981
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6196138021557256e-05,
+      "loss": 0.5741,
+      "step": 982
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6188139161064337e-05,
+      "loss": 0.5802,
+      "step": 983
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6180133879151943e-05,
+      "loss": 0.5169,
+      "step": 984
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6172122184127142e-05,
+      "loss": 0.5267,
+      "step": 985
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.616410408430365e-05,
+      "loss": 0.5721,
+      "step": 986
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.615607958800183e-05,
+      "loss": 0.514,
+      "step": 987
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6148048703548696e-05,
+      "loss": 0.5524,
+      "step": 988
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.614001143927787e-05,
+      "loss": 0.6135,
+      "step": 989
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6131967803529613e-05,
+      "loss": 0.5576,
+      "step": 990
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6123917804650783e-05,
+      "loss": 0.5319,
+      "step": 991
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6115861450994855e-05,
+      "loss": 0.5944,
+      "step": 992
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.610779875092188e-05,
+      "loss": 0.5411,
+      "step": 993
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6099729712798517e-05,
+      "loss": 0.556,
+      "step": 994
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6091654344997984e-05,
+      "loss": 0.566,
+      "step": 995
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6083572655900074e-05,
+      "loss": 0.5432,
+      "step": 996
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.607548465389114e-05,
+      "loss": 0.5501,
+      "step": 997
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.606739034736409e-05,
+      "loss": 0.585,
+      "step": 998
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6059289744718364e-05,
+      "loss": 0.5551,
+      "step": 999
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6051182854359935e-05,
+      "loss": 0.4931,
+      "step": 1000
+    },
+    {
+      "epoch": 0.94,
+      "eval_loss": 0.5632990598678589,
+      "eval_runtime": 98.3793,
+      "eval_samples_per_second": 7.034,
+      "eval_steps_per_second": 0.447,
+      "step": 1000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6043069684701323e-05,
+      "loss": 0.5575,
+      "step": 1001
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6034950244161534e-05,
+      "loss": 0.5676,
+      "step": 1002
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6026824541166104e-05,
+      "loss": 0.5723,
+      "step": 1003
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.601869258414705e-05,
+      "loss": 0.585,
+      "step": 1004
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6010554381542894e-05,
+      "loss": 0.5463,
+      "step": 1005
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.600240994179863e-05,
+      "loss": 0.5417,
+      "step": 1006
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.5994259273365726e-05,
+      "loss": 0.5163,
+      "step": 1007
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.5986102384702112e-05,
+      "loss": 0.5514,
+      "step": 1008
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.597793928427218e-05,
+      "loss": 0.5568,
+      "step": 1009
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.5969769980546762e-05,
+      "loss": 0.5915,
+      "step": 1010
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.5961594482003126e-05,
+      "loss": 0.566,
+      "step": 1011
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.5953412797124968e-05,
+      "loss": 0.5827,
+      "step": 1012
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.594522493440241e-05,
+      "loss": 0.5052,
+      "step": 1013
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.593703090233198e-05,
+      "loss": 0.6149,
+      "step": 1014
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5928830709416607e-05,
+      "loss": 0.5692,
+      "step": 1015
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.592062436416561e-05,
+      "loss": 0.519,
+      "step": 1016
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.591241187509471e-05,
+      "loss": 0.5332,
+      "step": 1017
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5904193250725978e-05,
+      "loss": 0.5427,
+      "step": 1018
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.589596849958787e-05,
+      "loss": 0.5564,
+      "step": 1019
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5887737630215188e-05,
+      "loss": 0.5379,
+      "step": 1020
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5879500651149093e-05,
+      "loss": 0.525,
+      "step": 1021
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5871257570937084e-05,
+      "loss": 0.5347,
+      "step": 1022
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5863008398132982e-05,
+      "loss": 0.5236,
+      "step": 1023
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5854753141296938e-05,
+      "loss": 0.5352,
+      "step": 1024
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5846491808995422e-05,
+      "loss": 0.5599,
+      "step": 1025
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5838224409801193e-05,
+      "loss": 0.5835,
+      "step": 1026
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5829950952293324e-05,
+      "loss": 0.5847,
+      "step": 1027
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5821671445057157e-05,
+      "loss": 0.549,
+      "step": 1028
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.581338589668432e-05,
+      "loss": 0.5461,
+      "step": 1029
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5805094315772716e-05,
+      "loss": 0.5552,
+      "step": 1030
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5796796710926496e-05,
+      "loss": 0.5382,
+      "step": 1031
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5788493090756074e-05,
+      "loss": 0.56,
+      "step": 1032
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.578018346387809e-05,
+      "loss": 0.5851,
+      "step": 1033
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5771867838915434e-05,
+      "loss": 0.5412,
+      "step": 1034
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5763546224497214e-05,
+      "loss": 0.5705,
+      "step": 1035
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5755218629258747e-05,
+      "loss": 0.5509,
+      "step": 1036
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.574688506184157e-05,
+      "loss": 0.536,
+      "step": 1037
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5738545530893402e-05,
+      "loss": 0.5262,
+      "step": 1038
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5730200045068157e-05,
+      "loss": 0.5661,
+      "step": 1039
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.572184861302593e-05,
+      "loss": 0.5689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.571349124343299e-05,
+      "loss": 0.518,
+      "step": 1041
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.570512794496176e-05,
+      "loss": 0.5438,
+      "step": 1042
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5696758726290813e-05,
+      "loss": 0.5671,
+      "step": 1043
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5688383596104875e-05,
+      "loss": 0.505,
+      "step": 1044
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.56800025630948e-05,
+      "loss": 0.5393,
+      "step": 1045
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.567161563595757e-05,
+      "loss": 0.562,
+      "step": 1046
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5663222823396282e-05,
+      "loss": 0.5085,
+      "step": 1047
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.565482413412014e-05,
+      "loss": 0.559,
+      "step": 1048
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.564641957684445e-05,
+      "loss": 0.5815,
+      "step": 1049
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5638009160290597e-05,
+      "loss": 0.5505,
+      "step": 1050
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.562959289318606e-05,
+      "loss": 0.5386,
+      "step": 1051
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.562117078426438e-05,
+      "loss": 0.5308,
+      "step": 1052
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.561274284226516e-05,
+      "loss": 0.5453,
+      "step": 1053
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.560430907593406e-05,
+      "loss": 0.526,
+      "step": 1054
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5595869494022787e-05,
+      "loss": 0.629,
+      "step": 1055
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.558742410528907e-05,
+      "loss": 0.572,
+      "step": 1056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5578972918496678e-05,
+      "loss": 0.5566,
+      "step": 1057
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5570515942415387e-05,
+      "loss": 0.511,
+      "step": 1058
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5562053185820986e-05,
+      "loss": 0.6134,
+      "step": 1059
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5553584657495257e-05,
+      "loss": 0.5505,
+      "step": 1060
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5545110366225983e-05,
+      "loss": 0.3655,
+      "step": 1061
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5536630320806907e-05,
+      "loss": 0.3365,
+      "step": 1062
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5528144530037763e-05,
+      "loss": 0.3175,
+      "step": 1063
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5519653002724235e-05,
+      "loss": 0.344,
+      "step": 1064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.551115574767797e-05,
+      "loss": 0.347,
+      "step": 1065
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5502652773716542e-05,
+      "loss": 0.3681,
+      "step": 1066
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.549414408966348e-05,
+      "loss": 0.3384,
+      "step": 1067
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5485629704348223e-05,
+      "loss": 0.3399,
+      "step": 1068
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.547710962660613e-05,
+      "loss": 0.3414,
+      "step": 1069
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5468583865278474e-05,
+      "loss": 0.3383,
+      "step": 1070
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5460052429212414e-05,
+      "loss": 0.3254,
+      "step": 1071
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5451515327261012e-05,
+      "loss": 0.3237,
+      "step": 1072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.544297256828319e-05,
+      "loss": 0.3459,
+      "step": 1073
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.543442416114376e-05,
+      "loss": 0.3413,
+      "step": 1074
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.542587011471339e-05,
+      "loss": 0.3238,
+      "step": 1075
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5417310437868586e-05,
+      "loss": 0.3296,
+      "step": 1076
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5408745139491714e-05,
+      "loss": 0.3596,
+      "step": 1077
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5400174228470967e-05,
+      "loss": 0.3333,
+      "step": 1078
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5391597713700358e-05,
+      "loss": 0.3321,
+      "step": 1079
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5383015604079723e-05,
+      "loss": 0.3476,
+      "step": 1080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.53744279085147e-05,
+      "loss": 0.3851,
+      "step": 1081
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5365834635916718e-05,
+      "loss": 0.3393,
+      "step": 1082
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5357235795203003e-05,
+      "loss": 0.3622,
+      "step": 1083
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5348631395296553e-05,
+      "loss": 0.3355,
+      "step": 1084
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.534002144512614e-05,
+      "loss": 0.3355,
+      "step": 1085
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5331405953626284e-05,
+      "loss": 0.3487,
+      "step": 1086
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5322784929737276e-05,
+      "loss": 0.354,
+      "step": 1087
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.531415838240512e-05,
+      "loss": 0.3015,
+      "step": 1088
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.530552632058158e-05,
+      "loss": 0.3634,
+      "step": 1089
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5296888753224116e-05,
+      "loss": 0.382,
+      "step": 1090
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5288245689295928e-05,
+      "loss": 0.3368,
+      "step": 1091
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5279597137765894e-05,
+      "loss": 0.3213,
+      "step": 1092
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5270943107608608e-05,
+      "loss": 0.3663,
+      "step": 1093
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.526228360780433e-05,
+      "loss": 0.3456,
+      "step": 1094
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5253618647339013e-05,
+      "loss": 0.3662,
+      "step": 1095
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5244948235204265e-05,
+      "loss": 0.3623,
+      "step": 1096
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5236272380397359e-05,
+      "loss": 0.334,
+      "step": 1097
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5227591091921205e-05,
+      "loss": 0.3288,
+      "step": 1098
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5218904378784367e-05,
+      "loss": 0.2924,
+      "step": 1099
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5210212250001024e-05,
+      "loss": 0.3037,
+      "step": 1100
+    },
+    {
+      "epoch": 1.04,
+      "eval_loss": 0.5881170630455017,
+      "eval_runtime": 98.672,
+      "eval_samples_per_second": 7.013,
+      "eval_steps_per_second": 0.446,
+      "step": 1100
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.520151471459098e-05,
+      "loss": 0.3418,
+      "step": 1101
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5192811781579655e-05,
+      "loss": 0.3401,
+      "step": 1102
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5184103459998058e-05,
+      "loss": 0.3127,
+      "step": 1103
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5175389758882803e-05,
+      "loss": 0.3371,
+      "step": 1104
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5166670687276077e-05,
+      "loss": 0.3286,
+      "step": 1105
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5157946254225643e-05,
+      "loss": 0.3329,
+      "step": 1106
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5149216468784829e-05,
+      "loss": 0.3337,
+      "step": 1107
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5140481340012515e-05,
+      "loss": 0.2903,
+      "step": 1108
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5131740876973125e-05,
+      "loss": 0.3251,
+      "step": 1109
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5122995088736619e-05,
+      "loss": 0.3164,
+      "step": 1110
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5114243984378485e-05,
+      "loss": 0.3287,
+      "step": 1111
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5105487572979729e-05,
+      "loss": 0.3703,
+      "step": 1112
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5096725863626857e-05,
+      "loss": 0.3278,
+      "step": 1113
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5087958865411882e-05,
+      "loss": 0.3245,
+      "step": 1114
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5079186587432295e-05,
+      "loss": 0.3051,
+      "step": 1115
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5070409038791077e-05,
+      "loss": 0.3617,
+      "step": 1116
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5061626228596667e-05,
+      "loss": 0.339,
+      "step": 1117
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.505283816596297e-05,
+      "loss": 0.3695,
+      "step": 1118
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5044044860009344e-05,
+      "loss": 0.3262,
+      "step": 1119
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5035246319860578e-05,
+      "loss": 0.3282,
+      "step": 1120
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.502644255464691e-05,
+      "loss": 0.345,
+      "step": 1121
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5017633573503983e-05,
+      "loss": 0.3135,
+      "step": 1122
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.500881938557286e-05,
+      "loss": 0.3439,
+      "step": 1123
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.3545,
+      "step": 1124
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.4991175425937275e-05,
+      "loss": 0.3432,
+      "step": 1125
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.4982345672541914e-05,
+      "loss": 0.3532,
+      "step": 1126
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.4973510748976541e-05,
+      "loss": 0.3268,
+      "step": 1127
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.4964670664409136e-05,
+      "loss": 0.3502,
+      "step": 1128
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4955825428013038e-05,
+      "loss": 0.3375,
+      "step": 1129
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4946975048966932e-05,
+      "loss": 0.3423,
+      "step": 1130
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4938119536454838e-05,
+      "loss": 0.3167,
+      "step": 1131
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4929258899666102e-05,
+      "loss": 0.2661,
+      "step": 1132
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4920393147795387e-05,
+      "loss": 0.3372,
+      "step": 1133
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.491152229004267e-05,
+      "loss": 0.3389,
+      "step": 1134
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.490264633561322e-05,
+      "loss": 0.3693,
+      "step": 1135
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.48937652937176e-05,
+      "loss": 0.3383,
+      "step": 1136
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4884879173571645e-05,
+      "loss": 0.3414,
+      "step": 1137
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4875987984396466e-05,
+      "loss": 0.3479,
+      "step": 1138
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4867091735418426e-05,
+      "loss": 0.3511,
+      "step": 1139
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4858190435869155e-05,
+      "loss": 0.3227,
+      "step": 1140
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4849284094985507e-05,
+      "loss": 0.3268,
+      "step": 1141
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4840372722009578e-05,
+      "loss": 0.3145,
+      "step": 1142
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4831456326188676e-05,
+      "loss": 0.4911,
+      "step": 1143
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.482253491677533e-05,
+      "loss": 0.3222,
+      "step": 1144
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.481360850302727e-05,
+      "loss": 0.3364,
+      "step": 1145
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4804677094207416e-05,
+      "loss": 0.3446,
+      "step": 1146
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4795740699583872e-05,
+      "loss": 0.3145,
+      "step": 1147
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4786799328429915e-05,
+      "loss": 0.3038,
+      "step": 1148
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4777852990023993e-05,
+      "loss": 0.3269,
+      "step": 1149
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.47689016936497e-05,
+      "loss": 0.3216,
+      "step": 1150
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4759945448595776e-05,
+      "loss": 0.293,
+      "step": 1151
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4750984264156103e-05,
+      "loss": 0.3495,
+      "step": 1152
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4742018149629683e-05,
+      "loss": 0.3248,
+      "step": 1153
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4733047114320631e-05,
+      "loss": 0.3684,
+      "step": 1154
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4724071167538175e-05,
+      "loss": 0.371,
+      "step": 1155
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4715090318596637e-05,
+      "loss": 0.3339,
+      "step": 1156
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4706104576815423e-05,
+      "loss": 0.3416,
+      "step": 1157
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4697113951519024e-05,
+      "loss": 0.3113,
+      "step": 1158
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.468811845203699e-05,
+      "loss": 0.3357,
+      "step": 1159
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4679118087703933e-05,
+      "loss": 0.3768,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4670112867859515e-05,
+      "loss": 0.3498,
+      "step": 1161
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4661102801848435e-05,
+      "loss": 0.3002,
+      "step": 1162
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4652087899020415e-05,
+      "loss": 0.3634,
+      "step": 1163
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4643068168730203e-05,
+      "loss": 0.3245,
+      "step": 1164
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4634043620337557e-05,
+      "loss": 0.2912,
+      "step": 1165
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4625014263207233e-05,
+      "loss": 0.3251,
+      "step": 1166
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4615980106708978e-05,
+      "loss": 0.3365,
+      "step": 1167
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4606941160217511e-05,
+      "loss": 0.3191,
+      "step": 1168
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4597897433112538e-05,
+      "loss": 0.3369,
+      "step": 1169
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4588848934778708e-05,
+      "loss": 0.3128,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4579795674605632e-05,
+      "loss": 0.3343,
+      "step": 1171
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4570737661987862e-05,
+      "loss": 0.3275,
+      "step": 1172
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4561674906324874e-05,
+      "loss": 0.2924,
+      "step": 1173
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4552607417021079e-05,
+      "loss": 0.3452,
+      "step": 1174
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4543535203485786e-05,
+      "loss": 0.3886,
+      "step": 1175
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4534458275133214e-05,
+      "loss": 0.3366,
+      "step": 1176
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4525376641382476e-05,
+      "loss": 0.2977,
+      "step": 1177
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4516290311657563e-05,
+      "loss": 0.3316,
+      "step": 1178
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4507199295387339e-05,
+      "loss": 0.3363,
+      "step": 1179
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4498103602005537e-05,
+      "loss": 0.3153,
+      "step": 1180
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4489003240950736e-05,
+      "loss": 0.3874,
+      "step": 1181
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4479898221666364e-05,
+      "loss": 0.339,
+      "step": 1182
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4470788553600681e-05,
+      "loss": 0.3449,
+      "step": 1183
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.446167424620677e-05,
+      "loss": 0.3049,
+      "step": 1184
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4452555308942533e-05,
+      "loss": 0.308,
+      "step": 1185
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.444343175127067e-05,
+      "loss": 0.3006,
+      "step": 1186
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4434303582658675e-05,
+      "loss": 0.3171,
+      "step": 1187
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4425170812578837e-05,
+      "loss": 0.3636,
+      "step": 1188
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.441603345050821e-05,
+      "loss": 0.3517,
+      "step": 1189
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4406891505928613e-05,
+      "loss": 0.3628,
+      "step": 1190
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4397744988326625e-05,
+      "loss": 0.3813,
+      "step": 1191
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4388593907193571e-05,
+      "loss": 0.3602,
+      "step": 1192
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4379438272025507e-05,
+      "loss": 0.3197,
+      "step": 1193
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4370278092323222e-05,
+      "loss": 0.3113,
+      "step": 1194
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.436111337759221e-05,
+      "loss": 0.3269,
+      "step": 1195
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.435194413734268e-05,
+      "loss": 0.3117,
+      "step": 1196
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4342770381089532e-05,
+      "loss": 0.3396,
+      "step": 1197
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4333592118352355e-05,
+      "loss": 0.2882,
+      "step": 1198
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4324409358655413e-05,
+      "loss": 0.3342,
+      "step": 1199
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.431522211152764e-05,
+      "loss": 0.4182,
+      "step": 1200
+    },
+    {
+      "epoch": 1.13,
+      "eval_loss": 0.5983697772026062,
+      "eval_runtime": 98.4449,
+      "eval_samples_per_second": 7.029,
+      "eval_steps_per_second": 0.447,
+      "step": 1200
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.430603038650262e-05,
+      "loss": 0.3331,
+      "step": 1201
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4296834193118592e-05,
+      "loss": 0.3151,
+      "step": 1202
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4287633540918422e-05,
+      "loss": 0.3507,
+      "step": 1203
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4278428439449613e-05,
+      "loss": 0.3913,
+      "step": 1204
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4269218898264276e-05,
+      "loss": 0.3041,
+      "step": 1205
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4260004926919138e-05,
+      "loss": 0.3967,
+      "step": 1206
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4250786534975514e-05,
+      "loss": 0.3515,
+      "step": 1207
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.424156373199931e-05,
+      "loss": 0.3178,
+      "step": 1208
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4232336527561013e-05,
+      "loss": 0.3427,
+      "step": 1209
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4223104931235675e-05,
+      "loss": 0.3304,
+      "step": 1210
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.42138689526029e-05,
+      "loss": 0.3646,
+      "step": 1211
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.420462860124685e-05,
+      "loss": 0.346,
+      "step": 1212
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4195383886756217e-05,
+      "loss": 0.3654,
+      "step": 1213
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4186134818724223e-05,
+      "loss": 0.3286,
+      "step": 1214
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4176881406748604e-05,
+      "loss": 0.3258,
+      "step": 1215
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.416762366043161e-05,
+      "loss": 0.3822,
+      "step": 1216
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4158361589379983e-05,
+      "loss": 0.3493,
+      "step": 1217
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4149095203204959e-05,
+      "loss": 0.3332,
+      "step": 1218
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4139824511522245e-05,
+      "loss": 0.3566,
+      "step": 1219
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4130549523952022e-05,
+      "loss": 0.3538,
+      "step": 1220
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4121270250118922e-05,
+      "loss": 0.3511,
+      "step": 1221
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4111986699652031e-05,
+      "loss": 0.329,
+      "step": 1222
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4102698882184871e-05,
+      "loss": 0.3353,
+      "step": 1223
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4093406807355389e-05,
+      "loss": 0.3448,
+      "step": 1224
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4084110484805955e-05,
+      "loss": 0.3145,
+      "step": 1225
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4074809924183341e-05,
+      "loss": 0.3293,
+      "step": 1226
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4065505135138724e-05,
+      "loss": 0.3519,
+      "step": 1227
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4056196127327661e-05,
+      "loss": 0.3057,
+      "step": 1228
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4046882910410092e-05,
+      "loss": 0.3556,
+      "step": 1229
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4037565494050322e-05,
+      "loss": 0.3424,
+      "step": 1230
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4028243887917022e-05,
+      "loss": 0.3103,
+      "step": 1231
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4018918101683191e-05,
+      "loss": 0.3389,
+      "step": 1232
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4009588145026192e-05,
+      "loss": 0.318,
+      "step": 1233
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4000254027627694e-05,
+      "loss": 0.3225,
+      "step": 1234
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3990915759173694e-05,
+      "loss": 0.3103,
+      "step": 1235
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3981573349354492e-05,
+      "loss": 0.3123,
+      "step": 1236
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3972226807864688e-05,
+      "loss": 0.3696,
+      "step": 1237
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3962876144403175e-05,
+      "loss": 0.3529,
+      "step": 1238
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3953521368673112e-05,
+      "loss": 0.3275,
+      "step": 1239
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3944162490381931e-05,
+      "loss": 0.3504,
+      "step": 1240
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3934799519241322e-05,
+      "loss": 0.3312,
+      "step": 1241
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3925432464967223e-05,
+      "loss": 0.6195,
+      "step": 1242
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3916061337279807e-05,
+      "loss": 0.3368,
+      "step": 1243
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3906686145903467e-05,
+      "loss": 0.3307,
+      "step": 1244
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3897306900566832e-05,
+      "loss": 0.3799,
+      "step": 1245
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3887923611002716e-05,
+      "loss": 0.3528,
+      "step": 1246
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3878536286948145e-05,
+      "loss": 0.3292,
+      "step": 1247
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3869144938144325e-05,
+      "loss": 0.3341,
+      "step": 1248
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.385974957433664e-05,
+      "loss": 0.3477,
+      "step": 1249
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3850350205274639e-05,
+      "loss": 0.3437,
+      "step": 1250
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3840946840712026e-05,
+      "loss": 0.3408,
+      "step": 1251
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3831539490406658e-05,
+      "loss": 0.332,
+      "step": 1252
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3822128164120519e-05,
+      "loss": 0.3637,
+      "step": 1253
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3812712871619727e-05,
+      "loss": 0.3737,
+      "step": 1254
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3803293622674509e-05,
+      "loss": 0.3294,
+      "step": 1255
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3793870427059204e-05,
+      "loss": 0.2935,
+      "step": 1256
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.378444329455224e-05,
+      "loss": 0.3493,
+      "step": 1257
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3775012234936135e-05,
+      "loss": 0.3421,
+      "step": 1258
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.376557725799748e-05,
+      "loss": 0.3403,
+      "step": 1259
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3756138373526931e-05,
+      "loss": 0.3231,
+      "step": 1260
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3746695591319197e-05,
+      "loss": 0.2763,
+      "step": 1261
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3737248921173037e-05,
+      "loss": 0.3717,
+      "step": 1262
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3727798372891241e-05,
+      "loss": 0.321,
+      "step": 1263
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3718343956280624e-05,
+      "loss": 0.3068,
+      "step": 1264
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3708885681152015e-05,
+      "loss": 0.3545,
+      "step": 1265
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3699423557320248e-05,
+      "loss": 0.2884,
+      "step": 1266
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3689957594604146e-05,
+      "loss": 0.3314,
+      "step": 1267
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3680487802826523e-05,
+      "loss": 0.3455,
+      "step": 1268
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3671014191814163e-05,
+      "loss": 0.3364,
+      "step": 1269
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3661536771397806e-05,
+      "loss": 0.312,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3652055551412161e-05,
+      "loss": 0.3206,
+      "step": 1271
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3642570541695867e-05,
+      "loss": 0.3609,
+      "step": 1272
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3633081752091499e-05,
+      "loss": 0.3252,
+      "step": 1273
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3623589192445556e-05,
+      "loss": 0.3354,
+      "step": 1274
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3614092872608449e-05,
+      "loss": 0.3355,
+      "step": 1275
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3604592802434486e-05,
+      "loss": 0.3277,
+      "step": 1276
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.359508899178187e-05,
+      "loss": 0.3388,
+      "step": 1277
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3585581450512693e-05,
+      "loss": 0.3346,
+      "step": 1278
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3576070188492905e-05,
+      "loss": 0.38,
+      "step": 1279
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3566555215592323e-05,
+      "loss": 0.31,
+      "step": 1280
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.355703654168462e-05,
+      "loss": 0.3434,
+      "step": 1281
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3547514176647303e-05,
+      "loss": 0.333,
+      "step": 1282
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3537988130361708e-05,
+      "loss": 0.33,
+      "step": 1283
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3528458412712996e-05,
+      "loss": 0.3182,
+      "step": 1284
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3518925033590137e-05,
+      "loss": 0.3153,
+      "step": 1285
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3509388002885899e-05,
+      "loss": 0.3291,
+      "step": 1286
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3499847330496838e-05,
+      "loss": 0.3102,
+      "step": 1287
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3490303026323289e-05,
+      "loss": 0.3435,
+      "step": 1288
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3480755100269356e-05,
+      "loss": 0.3328,
+      "step": 1289
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3471203562242909e-05,
+      "loss": 0.3329,
+      "step": 1290
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3461648422155556e-05,
+      "loss": 0.3385,
+      "step": 1291
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3452089689922646e-05,
+      "loss": 0.321,
+      "step": 1292
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3442527375463257e-05,
+      "loss": 0.3469,
+      "step": 1293
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3432961488700182e-05,
+      "loss": 0.2984,
+      "step": 1294
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3423392039559926e-05,
+      "loss": 0.3799,
+      "step": 1295
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3413819037972682e-05,
+      "loss": 0.2902,
+      "step": 1296
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3404242493872334e-05,
+      "loss": 0.3138,
+      "step": 1297
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.339466241719645e-05,
+      "loss": 0.3411,
+      "step": 1298
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3385078817886249e-05,
+      "loss": 0.3576,
+      "step": 1299
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3375491705886614e-05,
+      "loss": 0.3376,
+      "step": 1300
+    },
+    {
+      "epoch": 1.23,
+      "eval_loss": 0.595194935798645,
+      "eval_runtime": 98.5396,
+      "eval_samples_per_second": 7.023,
+      "eval_steps_per_second": 0.447,
+      "step": 1300
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3365901091146073e-05,
+      "loss": 0.3101,
+      "step": 1301
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3356306983616787e-05,
+      "loss": 0.3151,
+      "step": 1302
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3346709393254543e-05,
+      "loss": 0.3997,
+      "step": 1303
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3337108330018737e-05,
+      "loss": 0.3259,
+      "step": 1304
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3327503803872379e-05,
+      "loss": 0.3335,
+      "step": 1305
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.331789582478206e-05,
+      "loss": 0.3377,
+      "step": 1306
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3308284402717968e-05,
+      "loss": 0.345,
+      "step": 1307
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3298669547653848e-05,
+      "loss": 0.3052,
+      "step": 1308
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3289051269567021e-05,
+      "loss": 0.3403,
+      "step": 1309
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3279429578438354e-05,
+      "loss": 0.3266,
+      "step": 1310
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3269804484252255e-05,
+      "loss": 0.3031,
+      "step": 1311
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3260175996996667e-05,
+      "loss": 0.3641,
+      "step": 1312
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3250544126663047e-05,
+      "loss": 0.3145,
+      "step": 1313
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3240908883246373e-05,
+      "loss": 0.3467,
+      "step": 1314
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3231270276745111e-05,
+      "loss": 0.3566,
+      "step": 1315
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.322162831716123e-05,
+      "loss": 0.3044,
+      "step": 1316
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3211983014500169e-05,
+      "loss": 0.37,
+      "step": 1317
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.320233437877084e-05,
+      "loss": 0.3223,
+      "step": 1318
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3192682419985614e-05,
+      "loss": 0.3647,
+      "step": 1319
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3183027148160304e-05,
+      "loss": 0.3007,
+      "step": 1320
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3173368573314171e-05,
+      "loss": 0.3466,
+      "step": 1321
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3163706705469897e-05,
+      "loss": 0.3462,
+      "step": 1322
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3154041554653578e-05,
+      "loss": 0.3565,
+      "step": 1323
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3144373130894729e-05,
+      "loss": 0.3317,
+      "step": 1324
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.313470144422625e-05,
+      "loss": 0.3139,
+      "step": 1325
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3125026504684432e-05,
+      "loss": 0.3401,
+      "step": 1326
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3115348322308941e-05,
+      "loss": 0.3597,
+      "step": 1327
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3105666907142804e-05,
+      "loss": 0.3317,
+      "step": 1328
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3095982269232408e-05,
+      "loss": 0.3562,
+      "step": 1329
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3086294418627483e-05,
+      "loss": 0.3103,
+      "step": 1330
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3076603365381088e-05,
+      "loss": 0.3081,
+      "step": 1331
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3066909119549615e-05,
+      "loss": 0.3399,
+      "step": 1332
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3057211691192757e-05,
+      "loss": 0.3582,
+      "step": 1333
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3047511090373521e-05,
+      "loss": 0.291,
+      "step": 1334
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.30378073271582e-05,
+      "loss": 0.3775,
+      "step": 1335
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3028100411616369e-05,
+      "loss": 0.3298,
+      "step": 1336
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3018390353820871e-05,
+      "loss": 0.3206,
+      "step": 1337
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3008677163847819e-05,
+      "loss": 0.3238,
+      "step": 1338
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.2998960851776565e-05,
+      "loss": 0.3681,
+      "step": 1339
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.2989241427689712e-05,
+      "loss": 0.3616,
+      "step": 1340
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2979518901673082e-05,
+      "loss": 0.3355,
+      "step": 1341
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2969793283815721e-05,
+      "loss": 0.3493,
+      "step": 1342
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2960064584209889e-05,
+      "loss": 0.3298,
+      "step": 1343
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.295033281295103e-05,
+      "loss": 0.3141,
+      "step": 1344
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2940597980137786e-05,
+      "loss": 0.3191,
+      "step": 1345
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2930860095871975e-05,
+      "loss": 0.3335,
+      "step": 1346
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2921119170258579e-05,
+      "loss": 0.3235,
+      "step": 1347
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2911375213405738e-05,
+      "loss": 0.3313,
+      "step": 1348
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2901628235424732e-05,
+      "loss": 0.3257,
+      "step": 1349
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2891878246429981e-05,
+      "loss": 0.3474,
+      "step": 1350
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2882125256539032e-05,
+      "loss": 0.3357,
+      "step": 1351
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2872369275872547e-05,
+      "loss": 0.3356,
+      "step": 1352
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2862610314554277e-05,
+      "loss": 0.3015,
+      "step": 1353
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2852848382711083e-05,
+      "loss": 0.3249,
+      "step": 1354
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2843083490472901e-05,
+      "loss": 0.3129,
+      "step": 1355
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.283331564797274e-05,
+      "loss": 0.3403,
+      "step": 1356
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2823544865346666e-05,
+      "loss": 0.3485,
+      "step": 1357
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2813771152733808e-05,
+      "loss": 0.3436,
+      "step": 1358
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.280399452027632e-05,
+      "loss": 0.3302,
+      "step": 1359
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2794214978119399e-05,
+      "loss": 0.3164,
+      "step": 1360
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2784432536411256e-05,
+      "loss": 0.377,
+      "step": 1361
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2774647205303109e-05,
+      "loss": 0.3053,
+      "step": 1362
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2764858994949176e-05,
+      "loss": 0.3225,
+      "step": 1363
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2755067915506664e-05,
+      "loss": 0.3351,
+      "step": 1364
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2745273977135757e-05,
+      "loss": 0.3588,
+      "step": 1365
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2735477189999603e-05,
+      "loss": 0.3756,
+      "step": 1366
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.272567756426431e-05,
+      "loss": 0.3565,
+      "step": 1367
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.271587511009893e-05,
+      "loss": 0.37,
+      "step": 1368
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2706069837675448e-05,
+      "loss": 0.3374,
+      "step": 1369
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2696261757168775e-05,
+      "loss": 0.373,
+      "step": 1370
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.268645087875674e-05,
+      "loss": 0.2926,
+      "step": 1371
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2676637212620069e-05,
+      "loss": 0.3165,
+      "step": 1372
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2666820768942381e-05,
+      "loss": 0.377,
+      "step": 1373
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2657001557910182e-05,
+      "loss": 0.3297,
+      "step": 1374
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2647179589712853e-05,
+      "loss": 0.2901,
+      "step": 1375
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.263735487454262e-05,
+      "loss": 0.3604,
+      "step": 1376
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.262752742259458e-05,
+      "loss": 0.3415,
+      "step": 1377
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2617697244066651e-05,
+      "loss": 0.345,
+      "step": 1378
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2607864349159595e-05,
+      "loss": 0.3544,
+      "step": 1379
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2598028748076987e-05,
+      "loss": 0.3664,
+      "step": 1380
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2588190451025209e-05,
+      "loss": 0.3462,
+      "step": 1381
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2578349468213442e-05,
+      "loss": 0.3356,
+      "step": 1382
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2568505809853653e-05,
+      "loss": 0.335,
+      "step": 1383
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.255865948616059e-05,
+      "loss": 0.3573,
+      "step": 1384
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.254881050735176e-05,
+      "loss": 0.333,
+      "step": 1385
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2538958883647423e-05,
+      "loss": 0.3515,
+      "step": 1386
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2529104625270598e-05,
+      "loss": 0.364,
+      "step": 1387
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2519247742447029e-05,
+      "loss": 0.3316,
+      "step": 1388
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2509388245405179e-05,
+      "loss": 0.3637,
+      "step": 1389
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2499526144376228e-05,
+      "loss": 0.3881,
+      "step": 1390
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.248966144959406e-05,
+      "loss": 0.3583,
+      "step": 1391
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2479794171295248e-05,
+      "loss": 0.3406,
+      "step": 1392
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2469924319719047e-05,
+      "loss": 0.3371,
+      "step": 1393
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2460051905107382e-05,
+      "loss": 0.3241,
+      "step": 1394
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2450176937704837e-05,
+      "loss": 0.3144,
+      "step": 1395
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2440299427758646e-05,
+      "loss": 0.3163,
+      "step": 1396
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2430419385518683e-05,
+      "loss": 0.3293,
+      "step": 1397
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2420536821237446e-05,
+      "loss": 0.333,
+      "step": 1398
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2410651745170053e-05,
+      "loss": 0.3596,
+      "step": 1399
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2400764167574225e-05,
+      "loss": 0.3193,
+      "step": 1400
+    },
+    {
+      "epoch": 1.32,
+      "eval_loss": 0.5944209694862366,
+      "eval_runtime": 98.4799,
+      "eval_samples_per_second": 7.027,
+      "eval_steps_per_second": 0.447,
+      "step": 1400
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2390874098710283e-05,
+      "loss": 0.3949,
+      "step": 1401
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2380981548841131e-05,
+      "loss": 0.3728,
+      "step": 1402
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2371086528232246e-05,
+      "loss": 0.3524,
+      "step": 1403
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2361189047151676e-05,
+      "loss": 0.3413,
+      "step": 1404
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2351289115870014e-05,
+      "loss": 0.3241,
+      "step": 1405
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2341386744660398e-05,
+      "loss": 0.3273,
+      "step": 1406
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.23314819437985e-05,
+      "loss": 0.321,
+      "step": 1407
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2321574723562511e-05,
+      "loss": 0.3183,
+      "step": 1408
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2311665094233134e-05,
+      "loss": 0.377,
+      "step": 1409
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2301753066093569e-05,
+      "loss": 0.2972,
+      "step": 1410
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.229183864942951e-05,
+      "loss": 0.3415,
+      "step": 1411
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2281921854529127e-05,
+      "loss": 0.3528,
+      "step": 1412
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2272002691683055e-05,
+      "loss": 0.3476,
+      "step": 1413
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2262081171184395e-05,
+      "loss": 0.3253,
+      "step": 1414
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2252157303328682e-05,
+      "loss": 0.342,
+      "step": 1415
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.22422310984139e-05,
+      "loss": 0.3108,
+      "step": 1416
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2232302566740447e-05,
+      "loss": 0.3779,
+      "step": 1417
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2222371718611142e-05,
+      "loss": 0.3341,
+      "step": 1418
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2212438564331206e-05,
+      "loss": 0.3361,
+      "step": 1419
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2202503114208251e-05,
+      "loss": 0.3545,
+      "step": 1420
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2192565378552277e-05,
+      "loss": 0.3494,
+      "step": 1421
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2182625367675646e-05,
+      "loss": 0.3209,
+      "step": 1422
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2172683091893096e-05,
+      "loss": 0.3221,
+      "step": 1423
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2162738561521704e-05,
+      "loss": 0.3475,
+      "step": 1424
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2152791786880884e-05,
+      "loss": 0.3892,
+      "step": 1425
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2142842778292388e-05,
+      "loss": 0.3523,
+      "step": 1426
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2132891546080282e-05,
+      "loss": 0.2987,
+      "step": 1427
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2122938100570938e-05,
+      "loss": 0.3095,
+      "step": 1428
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2112982452093026e-05,
+      "loss": 0.3627,
+      "step": 1429
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2103024610977502e-05,
+      "loss": 0.3476,
+      "step": 1430
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.20930645875576e-05,
+      "loss": 0.3766,
+      "step": 1431
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2083102392168814e-05,
+      "loss": 0.3337,
+      "step": 1432
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2073138035148896e-05,
+      "loss": 0.3556,
+      "step": 1433
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2063171526837838e-05,
+      "loss": 0.3461,
+      "step": 1434
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2053202877577866e-05,
+      "loss": 0.3565,
+      "step": 1435
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2043232097713424e-05,
+      "loss": 0.3372,
+      "step": 1436
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2033259197591174e-05,
+      "loss": 0.363,
+      "step": 1437
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2023284187559974e-05,
+      "loss": 0.3415,
+      "step": 1438
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2013307077970867e-05,
+      "loss": 0.3099,
+      "step": 1439
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2003327879177085e-05,
+      "loss": 0.3326,
+      "step": 1440
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.1993346601534019e-05,
+      "loss": 0.3136,
+      "step": 1441
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.1983363255399226e-05,
+      "loss": 0.3315,
+      "step": 1442
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.1973377851132395e-05,
+      "loss": 0.3462,
+      "step": 1443
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.1963390399095368e-05,
+      "loss": 0.3213,
+      "step": 1444
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.19534009096521e-05,
+      "loss": 0.3205,
+      "step": 1445
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.1943409393168663e-05,
+      "loss": 0.3535,
+      "step": 1446
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1933415860013235e-05,
+      "loss": 0.3419,
+      "step": 1447
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1923420320556085e-05,
+      "loss": 0.3147,
+      "step": 1448
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1913422785169564e-05,
+      "loss": 0.3218,
+      "step": 1449
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1903423264228096e-05,
+      "loss": 0.3157,
+      "step": 1450
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.189342176810816e-05,
+      "loss": 0.2924,
+      "step": 1451
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1883418307188292e-05,
+      "loss": 0.314,
+      "step": 1452
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.187341289184906e-05,
+      "loss": 0.3675,
+      "step": 1453
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1863405532473065e-05,
+      "loss": 0.347,
+      "step": 1454
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1853396239444923e-05,
+      "loss": 0.341,
+      "step": 1455
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1843385023151255e-05,
+      "loss": 0.3096,
+      "step": 1456
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1833371893980685e-05,
+      "loss": 0.3427,
+      "step": 1457
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1823356862323814e-05,
+      "loss": 0.3251,
+      "step": 1458
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1813339938573218e-05,
+      "loss": 0.3277,
+      "step": 1459
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1803321133123442e-05,
+      "loss": 0.3474,
+      "step": 1460
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1793300456370977e-05,
+      "loss": 0.3161,
+      "step": 1461
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1783277918714262e-05,
+      "loss": 0.3099,
+      "step": 1462
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1773253530553663e-05,
+      "loss": 0.3378,
+      "step": 1463
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1763227302291464e-05,
+      "loss": 0.3404,
+      "step": 1464
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1753199244331865e-05,
+      "loss": 0.3446,
+      "step": 1465
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1743169367080964e-05,
+      "loss": 0.3253,
+      "step": 1466
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.173313768094674e-05,
+      "loss": 0.35,
+      "step": 1467
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1723104196339055e-05,
+      "loss": 0.2982,
+      "step": 1468
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1713068923669634e-05,
+      "loss": 0.3354,
+      "step": 1469
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1703031873352063e-05,
+      "loss": 0.3208,
+      "step": 1470
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1692993055801765e-05,
+      "loss": 0.3123,
+      "step": 1471
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1682952481436e-05,
+      "loss": 0.3421,
+      "step": 1472
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1672910160673858e-05,
+      "loss": 0.3329,
+      "step": 1473
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1662866103936228e-05,
+      "loss": 0.3077,
+      "step": 1474
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1652820321645808e-05,
+      "loss": 0.3169,
+      "step": 1475
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1642772824227091e-05,
+      "loss": 0.3413,
+      "step": 1476
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1632723622106339e-05,
+      "loss": 0.3234,
+      "step": 1477
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1622672725711595e-05,
+      "loss": 0.3326,
+      "step": 1478
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1612620145472646e-05,
+      "loss": 0.3228,
+      "step": 1479
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.160256589182104e-05,
+      "loss": 0.3238,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1592509975190055e-05,
+      "loss": 0.3268,
+      "step": 1481
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1582452406014693e-05,
+      "loss": 0.3359,
+      "step": 1482
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1572393194731676e-05,
+      "loss": 0.3663,
+      "step": 1483
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1562332351779422e-05,
+      "loss": 0.3349,
+      "step": 1484
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1552269887598051e-05,
+      "loss": 0.3323,
+      "step": 1485
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1542205812629366e-05,
+      "loss": 0.327,
+      "step": 1486
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.153214013731683e-05,
+      "loss": 0.315,
+      "step": 1487
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1522072872105576e-05,
+      "loss": 0.3236,
+      "step": 1488
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1512004027442387e-05,
+      "loss": 0.3437,
+      "step": 1489
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1501933613775682e-05,
+      "loss": 0.3049,
+      "step": 1490
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1491861641555506e-05,
+      "loss": 0.3121,
+      "step": 1491
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1481788121233523e-05,
+      "loss": 0.3438,
+      "step": 1492
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1471713063263013e-05,
+      "loss": 0.3549,
+      "step": 1493
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1461636478098837e-05,
+      "loss": 0.3109,
+      "step": 1494
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1451558376197446e-05,
+      "loss": 0.3818,
+      "step": 1495
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1441478768016871e-05,
+      "loss": 0.299,
+      "step": 1496
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1431397664016697e-05,
+      "loss": 0.3182,
+      "step": 1497
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1421315074658066e-05,
+      "loss": 0.3506,
+      "step": 1498
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.141123101040366e-05,
+      "loss": 0.3094,
+      "step": 1499
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1401145481717695e-05,
+      "loss": 0.3182,
+      "step": 1500
+    },
+    {
+      "epoch": 1.42,
+      "eval_loss": 0.5894672274589539,
+      "eval_runtime": 98.6495,
+      "eval_samples_per_second": 7.015,
+      "eval_steps_per_second": 0.446,
+      "step": 1500
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1391058499065897e-05,
+      "loss": 0.3466,
+      "step": 1501
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1380970072915516e-05,
+      "loss": 0.3069,
+      "step": 1502
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1370880213735285e-05,
+      "loss": 0.3419,
+      "step": 1503
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1360788931995434e-05,
+      "loss": 0.3165,
+      "step": 1504
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1350696238167664e-05,
+      "loss": 0.303,
+      "step": 1505
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1340602142725143e-05,
+      "loss": 0.3346,
+      "step": 1506
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1330506656142494e-05,
+      "loss": 0.3288,
+      "step": 1507
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1320409788895786e-05,
+      "loss": 0.3382,
+      "step": 1508
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1310311551462507e-05,
+      "loss": 0.3304,
+      "step": 1509
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1300211954321592e-05,
+      "loss": 0.3314,
+      "step": 1510
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1290111007953362e-05,
+      "loss": 0.3405,
+      "step": 1511
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1280008722839552e-05,
+      "loss": 0.3287,
+      "step": 1512
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1269905109463283e-05,
+      "loss": 0.3319,
+      "step": 1513
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1259800178309057e-05,
+      "loss": 0.3323,
+      "step": 1514
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1249693939862733e-05,
+      "loss": 0.3644,
+      "step": 1515
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1239586404611542e-05,
+      "loss": 0.3634,
+      "step": 1516
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1229477583044049e-05,
+      "loss": 0.3503,
+      "step": 1517
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1219367485650156e-05,
+      "loss": 0.368,
+      "step": 1518
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1209256122921095e-05,
+      "loss": 0.3342,
+      "step": 1519
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1199143505349401e-05,
+      "loss": 0.3469,
+      "step": 1520
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1189029643428924e-05,
+      "loss": 0.339,
+      "step": 1521
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1178914547654794e-05,
+      "loss": 0.3442,
+      "step": 1522
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1168798228523426e-05,
+      "loss": 0.3329,
+      "step": 1523
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1158680696532502e-05,
+      "loss": 0.3405,
+      "step": 1524
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1148561962180966e-05,
+      "loss": 0.3406,
+      "step": 1525
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.113844203596901e-05,
+      "loss": 0.3566,
+      "step": 1526
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1128320928398058e-05,
+      "loss": 0.3215,
+      "step": 1527
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1118198649970762e-05,
+      "loss": 0.3185,
+      "step": 1528
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1108075211190994e-05,
+      "loss": 0.3302,
+      "step": 1529
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1097950622563822e-05,
+      "loss": 0.3243,
+      "step": 1530
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1087824894595514e-05,
+      "loss": 0.3379,
+      "step": 1531
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1077698037793516e-05,
+      "loss": 0.3392,
+      "step": 1532
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1067570062666449e-05,
+      "loss": 0.3264,
+      "step": 1533
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1057440979724085e-05,
+      "loss": 0.3334,
+      "step": 1534
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1047310799477362e-05,
+      "loss": 0.3746,
+      "step": 1535
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1037179532438345e-05,
+      "loss": 0.3294,
+      "step": 1536
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1027047189120224e-05,
+      "loss": 0.3681,
+      "step": 1537
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1016913780037315e-05,
+      "loss": 0.3285,
+      "step": 1538
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1006779315705038e-05,
+      "loss": 0.3371,
+      "step": 1539
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.0996643806639904e-05,
+      "loss": 0.3148,
+      "step": 1540
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.098650726335951e-05,
+      "loss": 0.3389,
+      "step": 1541
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.097636969638253e-05,
+      "loss": 0.3567,
+      "step": 1542
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0966231116228689e-05,
+      "loss": 0.3352,
+      "step": 1543
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0956091533418777e-05,
+      "loss": 0.3568,
+      "step": 1544
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0945950958474619e-05,
+      "loss": 0.3188,
+      "step": 1545
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0935809401919064e-05,
+      "loss": 0.3531,
+      "step": 1546
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0925666874275991e-05,
+      "loss": 0.3129,
+      "step": 1547
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0915523386070277e-05,
+      "loss": 0.3526,
+      "step": 1548
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0905378947827798e-05,
+      "loss": 0.3484,
+      "step": 1549
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.089523357007542e-05,
+      "loss": 0.3469,
+      "step": 1550
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0885087263340981e-05,
+      "loss": 0.3455,
+      "step": 1551
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0874940038153279e-05,
+      "loss": 0.3621,
+      "step": 1552
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0864791905042072e-05,
+      "loss": 0.3376,
+      "step": 1553
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0854642874538054e-05,
+      "loss": 0.3063,
+      "step": 1554
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.084449295717286e-05,
+      "loss": 0.3679,
+      "step": 1555
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0834342163479035e-05,
+      "loss": 0.3234,
+      "step": 1556
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0824190503990035e-05,
+      "loss": 0.3443,
+      "step": 1557
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0814037989240217e-05,
+      "loss": 0.3304,
+      "step": 1558
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0803884629764828e-05,
+      "loss": 0.342,
+      "step": 1559
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.079373043609999e-05,
+      "loss": 0.3105,
+      "step": 1560
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0783575418782684e-05,
+      "loss": 0.3391,
+      "step": 1561
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0773419588350748e-05,
+      "loss": 0.2811,
+      "step": 1562
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0763262955342878e-05,
+      "loss": 0.3592,
+      "step": 1563
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0753105530298582e-05,
+      "loss": 0.3271,
+      "step": 1564
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0742947323758204e-05,
+      "loss": 0.3459,
+      "step": 1565
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.073278834626289e-05,
+      "loss": 0.3489,
+      "step": 1566
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0722628608354595e-05,
+      "loss": 0.3343,
+      "step": 1567
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0712468120576059e-05,
+      "loss": 0.3181,
+      "step": 1568
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0702306893470796e-05,
+      "loss": 0.3209,
+      "step": 1569
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0692144937583091e-05,
+      "loss": 0.3262,
+      "step": 1570
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0681982263457987e-05,
+      "loss": 0.3268,
+      "step": 1571
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.067181888164127e-05,
+      "loss": 0.3305,
+      "step": 1572
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0661654802679455e-05,
+      "loss": 0.3481,
+      "step": 1573
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0651490037119796e-05,
+      "loss": 0.2735,
+      "step": 1574
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0641324595510245e-05,
+      "loss": 0.3193,
+      "step": 1575
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0631158488399463e-05,
+      "loss": 0.3504,
+      "step": 1576
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.062099172633679e-05,
+      "loss": 0.3556,
+      "step": 1577
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0610824319872266e-05,
+      "loss": 0.3321,
+      "step": 1578
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0600656279556578e-05,
+      "loss": 0.3353,
+      "step": 1579
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0590487615941088e-05,
+      "loss": 0.3326,
+      "step": 1580
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.058031833957779e-05,
+      "loss": 0.3742,
+      "step": 1581
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0570148461019332e-05,
+      "loss": 0.3302,
+      "step": 1582
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.055997799081897e-05,
+      "loss": 0.3151,
+      "step": 1583
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.054980693953058e-05,
+      "loss": 0.3065,
+      "step": 1584
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0539635317708644e-05,
+      "loss": 0.3316,
+      "step": 1585
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0529463135908229e-05,
+      "loss": 0.3381,
+      "step": 1586
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0519290404684992e-05,
+      "loss": 0.3199,
+      "step": 1587
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0509117134595152e-05,
+      "loss": 0.3331,
+      "step": 1588
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0498943336195494e-05,
+      "loss": 0.3372,
+      "step": 1589
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0488769020043346e-05,
+      "loss": 0.3091,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0478594196696577e-05,
+      "loss": 0.345,
+      "step": 1591
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.046841887671358e-05,
+      "loss": 0.3654,
+      "step": 1592
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0458243070653263e-05,
+      "loss": 0.334,
+      "step": 1593
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0448066789075037e-05,
+      "loss": 0.3196,
+      "step": 1594
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0437890042538814e-05,
+      "loss": 0.3404,
+      "step": 1595
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.042771284160498e-05,
+      "loss": 0.3297,
+      "step": 1596
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0417535196834396e-05,
+      "loss": 0.3055,
+      "step": 1597
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0407357118788384e-05,
+      "loss": 0.3294,
+      "step": 1598
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0397178618028716e-05,
+      "loss": 0.3575,
+      "step": 1599
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.03869997051176e-05,
+      "loss": 0.2995,
+      "step": 1600
+    },
+    {
+      "epoch": 1.51,
+      "eval_loss": 0.5877659320831299,
+      "eval_runtime": 98.5523,
+      "eval_samples_per_second": 7.022,
+      "eval_steps_per_second": 0.446,
+      "step": 1600
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0376820390617675e-05,
+      "loss": 0.3224,
+      "step": 1601
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.036664068509199e-05,
+      "loss": 0.3273,
+      "step": 1602
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0356460599104012e-05,
+      "loss": 0.3125,
+      "step": 1603
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0346280143217593e-05,
+      "loss": 0.323,
+      "step": 1604
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.033609932799697e-05,
+      "loss": 0.3263,
+      "step": 1605
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0325918164006754e-05,
+      "loss": 0.338,
+      "step": 1606
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.031573666181192e-05,
+      "loss": 0.3471,
+      "step": 1607
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0305554831977788e-05,
+      "loss": 0.3336,
+      "step": 1608
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.029537268507003e-05,
+      "loss": 0.3166,
+      "step": 1609
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0285190231654636e-05,
+      "loss": 0.3127,
+      "step": 1610
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0275007482297914e-05,
+      "loss": 0.3169,
+      "step": 1611
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0264824447566484e-05,
+      "loss": 0.3849,
+      "step": 1612
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0254641138027263e-05,
+      "loss": 0.3246,
+      "step": 1613
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0244457564247446e-05,
+      "loss": 0.362,
+      "step": 1614
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0234273736794512e-05,
+      "loss": 0.3419,
+      "step": 1615
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0224089666236195e-05,
+      "loss": 0.3274,
+      "step": 1616
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0213905363140484e-05,
+      "loss": 0.3156,
+      "step": 1617
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0203720838075613e-05,
+      "loss": 0.3594,
+      "step": 1618
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.019353610161004e-05,
+      "loss": 0.3497,
+      "step": 1619
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0183351164312448e-05,
+      "loss": 0.3771,
+      "step": 1620
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0173166036751723e-05,
+      "loss": 0.3583,
+      "step": 1621
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0162980729496952e-05,
+      "loss": 0.3451,
+      "step": 1622
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0152795253117407e-05,
+      "loss": 0.3088,
+      "step": 1623
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0142609618182539e-05,
+      "loss": 0.3373,
+      "step": 1624
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.013242383526196e-05,
+      "loss": 0.3092,
+      "step": 1625
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0122237914925432e-05,
+      "loss": 0.336,
+      "step": 1626
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.011205186774287e-05,
+      "loss": 0.3302,
+      "step": 1627
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.010186570428431e-05,
+      "loss": 0.3237,
+      "step": 1628
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0091679435119916e-05,
+      "loss": 0.3087,
+      "step": 1629
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0081493070819958e-05,
+      "loss": 0.3633,
+      "step": 1630
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0071306621954807e-05,
+      "loss": 0.3203,
+      "step": 1631
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0061120099094917e-05,
+      "loss": 0.3476,
+      "step": 1632
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0050933512810825e-05,
+      "loss": 0.3246,
+      "step": 1633
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0040746873673126e-05,
+      "loss": 0.3428,
+      "step": 1634
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0030560192252483e-05,
+      "loss": 0.3195,
+      "step": 1635
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.002037347911959e-05,
+      "loss": 0.3241,
+      "step": 1636
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0010186744845178e-05,
+      "loss": 0.3404,
+      "step": 1637
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1e-05,
+      "loss": 0.3601,
+      "step": 1638
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.989813255154826e-06,
+      "loss": 0.3645,
+      "step": 1639
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.979626520880413e-06,
+      "loss": 0.3495,
+      "step": 1640
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.96943980774752e-06,
+      "loss": 0.3091,
+      "step": 1641
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.959253126326876e-06,
+      "loss": 0.329,
+      "step": 1642
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.949066487189178e-06,
+      "loss": 0.3246,
+      "step": 1643
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.938879900905086e-06,
+      "loss": 0.3204,
+      "step": 1644
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.928693378045197e-06,
+      "loss": 0.3148,
+      "step": 1645
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.918506929180044e-06,
+      "loss": 0.3125,
+      "step": 1646
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.908320564880087e-06,
+      "loss": 0.3501,
+      "step": 1647
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.898134295715693e-06,
+      "loss": 0.33,
+      "step": 1648
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.887948132257134e-06,
+      "loss": 0.327,
+      "step": 1649
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.877762085074571e-06,
+      "loss": 0.313,
+      "step": 1650
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.867576164738043e-06,
+      "loss": 0.346,
+      "step": 1651
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.857390381817463e-06,
+      "loss": 0.2988,
+      "step": 1652
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.847204746882594e-06,
+      "loss": 0.3451,
+      "step": 1653
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.837019270503053e-06,
+      "loss": 0.346,
+      "step": 1654
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.82683396324828e-06,
+      "loss": 0.33,
+      "step": 1655
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.816648835687557e-06,
+      "loss": 0.3472,
+      "step": 1656
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.806463898389963e-06,
+      "loss": 0.361,
+      "step": 1657
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.79627916192439e-06,
+      "loss": 0.3392,
+      "step": 1658
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.78609463685952e-06,
+      "loss": 0.3648,
+      "step": 1659
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.775910333763809e-06,
+      "loss": 0.3457,
+      "step": 1660
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.765726263205491e-06,
+      "loss": 0.3352,
+      "step": 1661
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.755542435752557e-06,
+      "loss": 0.3422,
+      "step": 1662
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.745358861972742e-06,
+      "loss": 0.3406,
+      "step": 1663
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.73517555243352e-06,
+      "loss": 0.323,
+      "step": 1664
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.724992517702091e-06,
+      "loss": 0.3219,
+      "step": 1665
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.71480976834537e-06,
+      "loss": 0.3465,
+      "step": 1666
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.704627314929974e-06,
+      "loss": 0.3106,
+      "step": 1667
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.694445168022214e-06,
+      "loss": 0.3566,
+      "step": 1668
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.684263338188084e-06,
+      "loss": 0.3459,
+      "step": 1669
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.674081835993251e-06,
+      "loss": 0.3218,
+      "step": 1670
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.663900672003036e-06,
+      "loss": 0.3461,
+      "step": 1671
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.653719856782412e-06,
+      "loss": 0.3243,
+      "step": 1672
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.643539400895991e-06,
+      "loss": 0.3139,
+      "step": 1673
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.633359314908013e-06,
+      "loss": 0.3312,
+      "step": 1674
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.623179609382332e-06,
+      "loss": 0.3456,
+      "step": 1675
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.613000294882406e-06,
+      "loss": 0.3586,
+      "step": 1676
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.602821381971286e-06,
+      "loss": 0.3399,
+      "step": 1677
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.59264288121162e-06,
+      "loss": 0.3328,
+      "step": 1678
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.582464803165607e-06,
+      "loss": 0.3201,
+      "step": 1679
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.572287158395025e-06,
+      "loss": 0.3686,
+      "step": 1680
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.56210995746119e-06,
+      "loss": 0.3327,
+      "step": 1681
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.551933210924968e-06,
+      "loss": 0.3186,
+      "step": 1682
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.541756929346744e-06,
+      "loss": 0.3029,
+      "step": 1683
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.531581123286426e-06,
+      "loss": 0.3064,
+      "step": 1684
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.521405803303427e-06,
+      "loss": 0.3204,
+      "step": 1685
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.511230979956655e-06,
+      "loss": 0.3323,
+      "step": 1686
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.501056663804507e-06,
+      "loss": 0.3106,
+      "step": 1687
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.49088286540485e-06,
+      "loss": 0.3342,
+      "step": 1688
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.480709595315008e-06,
+      "loss": 0.3437,
+      "step": 1689
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.470536864091771e-06,
+      "loss": 0.3299,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.460364682291358e-06,
+      "loss": 0.3163,
+      "step": 1691
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.45019306046942e-06,
+      "loss": 0.311,
+      "step": 1692
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.440022009181032e-06,
+      "loss": 0.3208,
+      "step": 1693
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.429851538980668e-06,
+      "loss": 0.323,
+      "step": 1694
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.41968166042221e-06,
+      "loss": 0.3461,
+      "step": 1695
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.409512384058916e-06,
+      "loss": 0.3373,
+      "step": 1696
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.399343720443424e-06,
+      "loss": 0.3092,
+      "step": 1697
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.389175680127736e-06,
+      "loss": 0.3027,
+      "step": 1698
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.37900827366321e-06,
+      "loss": 0.3456,
+      "step": 1699
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.368841511600539e-06,
+      "loss": 0.3245,
+      "step": 1700
+    },
+    {
+      "epoch": 1.6,
+      "eval_loss": 0.5855045318603516,
+      "eval_runtime": 98.2402,
+      "eval_samples_per_second": 7.044,
+      "eval_steps_per_second": 0.448,
+      "step": 1700
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.358675404489754e-06,
+      "loss": 0.3381,
+      "step": 1701
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.348509962880203e-06,
+      "loss": 0.3359,
+      "step": 1702
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.338345197320545e-06,
+      "loss": 0.3509,
+      "step": 1703
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.328181118358734e-06,
+      "loss": 0.3378,
+      "step": 1704
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.318017736542017e-06,
+      "loss": 0.3563,
+      "step": 1705
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.30785506241691e-06,
+      "loss": 0.3598,
+      "step": 1706
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.297693106529206e-06,
+      "loss": 0.3126,
+      "step": 1707
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.287531879423943e-06,
+      "loss": 0.3409,
+      "step": 1708
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.277371391645405e-06,
+      "loss": 0.3553,
+      "step": 1709
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.267211653737112e-06,
+      "loss": 0.32,
+      "step": 1710
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.2570526762418e-06,
+      "loss": 0.331,
+      "step": 1711
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.246894469701423e-06,
+      "loss": 0.3223,
+      "step": 1712
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.236737044657126e-06,
+      "loss": 0.3283,
+      "step": 1713
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.226580411649253e-06,
+      "loss": 0.3226,
+      "step": 1714
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.21642458121732e-06,
+      "loss": 0.3208,
+      "step": 1715
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.206269563900013e-06,
+      "loss": 0.3545,
+      "step": 1716
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.196115370235173e-06,
+      "loss": 0.3443,
+      "step": 1717
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.185962010759784e-06,
+      "loss": 0.345,
+      "step": 1718
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.175809496009969e-06,
+      "loss": 0.3033,
+      "step": 1719
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.165657836520969e-06,
+      "loss": 0.3861,
+      "step": 1720
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.155507042827143e-06,
+      "loss": 0.366,
+      "step": 1721
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.145357125461947e-06,
+      "loss": 0.3246,
+      "step": 1722
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.13520809495793e-06,
+      "loss": 0.3108,
+      "step": 1723
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.125059961846725e-06,
+      "loss": 0.336,
+      "step": 1724
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.114912736659022e-06,
+      "loss": 0.3058,
+      "step": 1725
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.104766429924583e-06,
+      "loss": 0.2907,
+      "step": 1726
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.094621052172203e-06,
+      "loss": 0.2952,
+      "step": 1727
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.084476613929726e-06,
+      "loss": 0.2939,
+      "step": 1728
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.074333125724012e-06,
+      "loss": 0.3154,
+      "step": 1729
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.064190598080938e-06,
+      "loss": 0.3314,
+      "step": 1730
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.054049041525386e-06,
+      "loss": 0.3357,
+      "step": 1731
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.043908466581225e-06,
+      "loss": 0.3078,
+      "step": 1732
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.033768883771313e-06,
+      "loss": 0.2827,
+      "step": 1733
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.023630303617474e-06,
+      "loss": 0.3476,
+      "step": 1734
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.01349273664049e-06,
+      "loss": 0.3421,
+      "step": 1735
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.003356193360097e-06,
+      "loss": 0.3368,
+      "step": 1736
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.993220684294964e-06,
+      "loss": 0.3415,
+      "step": 1737
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.983086219962688e-06,
+      "loss": 0.3297,
+      "step": 1738
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.972952810879781e-06,
+      "loss": 0.3187,
+      "step": 1739
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.96282046756166e-06,
+      "loss": 0.3383,
+      "step": 1740
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.95268920052264e-06,
+      "loss": 0.3484,
+      "step": 1741
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.942559020275916e-06,
+      "loss": 0.309,
+      "step": 1742
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.932429937333555e-06,
+      "loss": 0.3403,
+      "step": 1743
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.922301962206487e-06,
+      "loss": 0.3353,
+      "step": 1744
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.91217510540449e-06,
+      "loss": 0.3144,
+      "step": 1745
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.90204937743618e-06,
+      "loss": 0.3433,
+      "step": 1746
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.89192478880901e-06,
+      "loss": 0.3746,
+      "step": 1747
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.881801350029242e-06,
+      "loss": 0.3186,
+      "step": 1748
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.871679071601946e-06,
+      "loss": 0.3403,
+      "step": 1749
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.861557964030993e-06,
+      "loss": 0.3259,
+      "step": 1750
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.851438037819035e-06,
+      "loss": 0.3158,
+      "step": 1751
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.841319303467502e-06,
+      "loss": 0.3296,
+      "step": 1752
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.831201771476579e-06,
+      "loss": 0.3373,
+      "step": 1753
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.82108545234521e-06,
+      "loss": 0.3331,
+      "step": 1754
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.81097035657108e-06,
+      "loss": 0.333,
+      "step": 1755
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.8008564946506e-06,
+      "loss": 0.305,
+      "step": 1756
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.790743877078908e-06,
+      "loss": 0.3512,
+      "step": 1757
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.780632514349847e-06,
+      "loss": 0.3637,
+      "step": 1758
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.770522416955955e-06,
+      "loss": 0.3467,
+      "step": 1759
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.760413595388463e-06,
+      "loss": 0.3336,
+      "step": 1760
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.75030606013727e-06,
+      "loss": 0.3181,
+      "step": 1761
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.740199821690948e-06,
+      "loss": 0.3386,
+      "step": 1762
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.73009489053672e-06,
+      "loss": 0.3205,
+      "step": 1763
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.719991277160453e-06,
+      "loss": 0.3168,
+      "step": 1764
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.709888992046643e-06,
+      "loss": 0.3486,
+      "step": 1765
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.699788045678413e-06,
+      "loss": 0.3108,
+      "step": 1766
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.689688448537495e-06,
+      "loss": 0.3513,
+      "step": 1767
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.67959021110422e-06,
+      "loss": 0.3233,
+      "step": 1768
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.669493343857509e-06,
+      "loss": 0.357,
+      "step": 1769
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.65939785727486e-06,
+      "loss": 0.3115,
+      "step": 1770
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.649303761832341e-06,
+      "loss": 0.3159,
+      "step": 1771
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.63921106800457e-06,
+      "loss": 0.338,
+      "step": 1772
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.62911978626472e-06,
+      "loss": 0.3617,
+      "step": 1773
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.619029927084487e-06,
+      "loss": 0.3281,
+      "step": 1774
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.608941500934106e-06,
+      "loss": 0.3282,
+      "step": 1775
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.59885451828231e-06,
+      "loss": 0.3252,
+      "step": 1776
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.588768989596345e-06,
+      "loss": 0.3527,
+      "step": 1777
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.578684925341939e-06,
+      "loss": 0.318,
+      "step": 1778
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.568602335983308e-06,
+      "loss": 0.3571,
+      "step": 1779
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.558521231983134e-06,
+      "loss": 0.3327,
+      "step": 1780
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.548441623802557e-06,
+      "loss": 0.3407,
+      "step": 1781
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.538363521901168e-06,
+      "loss": 0.3615,
+      "step": 1782
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.528286936736987e-06,
+      "loss": 0.3377,
+      "step": 1783
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.518211878766475e-06,
+      "loss": 0.3415,
+      "step": 1784
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.508138358444497e-06,
+      "loss": 0.3048,
+      "step": 1785
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.49806638622432e-06,
+      "loss": 0.3157,
+      "step": 1786
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.487995972557613e-06,
+      "loss": 0.3536,
+      "step": 1787
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.477927127894424e-06,
+      "loss": 0.2963,
+      "step": 1788
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.46785986268317e-06,
+      "loss": 0.2898,
+      "step": 1789
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.457794187370636e-06,
+      "loss": 0.268,
+      "step": 1790
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.447730112401949e-06,
+      "loss": 0.3104,
+      "step": 1791
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.43766764822058e-06,
+      "loss": 0.3839,
+      "step": 1792
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.427606805268328e-06,
+      "loss": 0.3266,
+      "step": 1793
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.417547593985309e-06,
+      "loss": 0.3183,
+      "step": 1794
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.407490024809945e-06,
+      "loss": 0.3617,
+      "step": 1795
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.39743410817896e-06,
+      "loss": 0.3178,
+      "step": 1796
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.387379854527354e-06,
+      "loss": 0.3213,
+      "step": 1797
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.377327274288407e-06,
+      "loss": 0.2741,
+      "step": 1798
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.367276377893661e-06,
+      "loss": 0.3618,
+      "step": 1799
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.35722717577291e-06,
+      "loss": 0.302,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7,
+      "eval_loss": 0.5799716114997864,
+      "eval_runtime": 98.5652,
+      "eval_samples_per_second": 7.021,
+      "eval_steps_per_second": 0.446,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.347179678354194e-06,
+      "loss": 0.3222,
+      "step": 1801
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.337133896063777e-06,
+      "loss": 0.3258,
+      "step": 1802
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.327089839326145e-06,
+      "loss": 0.3463,
+      "step": 1803
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.317047518564e-06,
+      "loss": 0.3393,
+      "step": 1804
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.307006944198236e-06,
+      "loss": 0.3295,
+      "step": 1805
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.296968126647939e-06,
+      "loss": 0.3162,
+      "step": 1806
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.286931076330367e-06,
+      "loss": 0.3457,
+      "step": 1807
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.276895803660947e-06,
+      "loss": 0.3333,
+      "step": 1808
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.266862319053262e-06,
+      "loss": 0.336,
+      "step": 1809
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.256830632919038e-06,
+      "loss": 0.3355,
+      "step": 1810
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.246800755668137e-06,
+      "loss": 0.3212,
+      "step": 1811
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.236772697708537e-06,
+      "loss": 0.3424,
+      "step": 1812
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.22674646944634e-06,
+      "loss": 0.3544,
+      "step": 1813
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.216722081285741e-06,
+      "loss": 0.3179,
+      "step": 1814
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.206699543629025e-06,
+      "loss": 0.3146,
+      "step": 1815
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.196678866876561e-06,
+      "loss": 0.3291,
+      "step": 1816
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.186660061426785e-06,
+      "loss": 0.2784,
+      "step": 1817
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.17664313767619e-06,
+      "loss": 0.3135,
+      "step": 1818
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.166628106019318e-06,
+      "loss": 0.3738,
+      "step": 1819
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.156614976848748e-06,
+      "loss": 0.3302,
+      "step": 1820
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.14660376055508e-06,
+      "loss": 0.3198,
+      "step": 1821
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.136594467526938e-06,
+      "loss": 0.3454,
+      "step": 1822
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.126587108150943e-06,
+      "loss": 0.3437,
+      "step": 1823
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.116581692811711e-06,
+      "loss": 0.3218,
+      "step": 1824
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.106578231891842e-06,
+      "loss": 0.324,
+      "step": 1825
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.096576735771908e-06,
+      "loss": 0.3202,
+      "step": 1826
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.086577214830437e-06,
+      "loss": 0.39,
+      "step": 1827
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.076579679443919e-06,
+      "loss": 0.3357,
+      "step": 1828
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.066584139986766e-06,
+      "loss": 0.3235,
+      "step": 1829
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.05659060683134e-06,
+      "loss": 0.3343,
+      "step": 1830
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.046599090347904e-06,
+      "loss": 0.3126,
+      "step": 1831
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.036609600904636e-06,
+      "loss": 0.3806,
+      "step": 1832
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.026622148867608e-06,
+      "loss": 0.3055,
+      "step": 1833
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.01663674460078e-06,
+      "loss": 0.335,
+      "step": 1834
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.006653398465983e-06,
+      "loss": 0.3195,
+      "step": 1835
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.996672120822918e-06,
+      "loss": 0.3641,
+      "step": 1836
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.986692922029137e-06,
+      "loss": 0.3239,
+      "step": 1837
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.97671581244003e-06,
+      "loss": 0.3149,
+      "step": 1838
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.96674080240883e-06,
+      "loss": 0.3276,
+      "step": 1839
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.956767902286578e-06,
+      "loss": 0.3199,
+      "step": 1840
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.946797122422138e-06,
+      "loss": 0.3206,
+      "step": 1841
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.936828473162166e-06,
+      "loss": 0.3394,
+      "step": 1842
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.926861964851107e-06,
+      "loss": 0.3357,
+      "step": 1843
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.916897607831189e-06,
+      "loss": 0.3261,
+      "step": 1844
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.906935412442403e-06,
+      "loss": 0.3362,
+      "step": 1845
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.8969753890225e-06,
+      "loss": 0.3298,
+      "step": 1846
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.887017547906976e-06,
+      "loss": 0.3266,
+      "step": 1847
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.877061899429067e-06,
+      "loss": 0.295,
+      "step": 1848
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.86710845391972e-06,
+      "loss": 0.3311,
+      "step": 1849
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.857157221707614e-06,
+      "loss": 0.3283,
+      "step": 1850
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.847208213119119e-06,
+      "loss": 0.3041,
+      "step": 1851
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.8372614384783e-06,
+      "loss": 0.316,
+      "step": 1852
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.827316908106905e-06,
+      "loss": 0.3297,
+      "step": 1853
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.817374632324355e-06,
+      "loss": 0.313,
+      "step": 1854
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.807434621447726e-06,
+      "loss": 0.3198,
+      "step": 1855
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.797496885791754e-06,
+      "loss": 0.3314,
+      "step": 1856
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.787561435668801e-06,
+      "loss": 0.3496,
+      "step": 1857
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.777628281388863e-06,
+      "loss": 0.3167,
+      "step": 1858
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.767697433259559e-06,
+      "loss": 0.3119,
+      "step": 1859
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.757768901586106e-06,
+      "loss": 0.306,
+      "step": 1860
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.747842696671322e-06,
+      "loss": 0.3258,
+      "step": 1861
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.73791882881561e-06,
+      "loss": 0.337,
+      "step": 1862
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.727997308316947e-06,
+      "loss": 0.3322,
+      "step": 1863
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.718078145470878e-06,
+      "loss": 0.3189,
+      "step": 1864
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.708161350570493e-06,
+      "loss": 0.3471,
+      "step": 1865
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.698246933906435e-06,
+      "loss": 0.3191,
+      "step": 1866
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.688334905766871e-06,
+      "loss": 0.3334,
+      "step": 1867
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.678425276437494e-06,
+      "loss": 0.3138,
+      "step": 1868
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.668518056201506e-06,
+      "loss": 0.3378,
+      "step": 1869
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.658613255339607e-06,
+      "loss": 0.3379,
+      "step": 1870
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.648710884129991e-06,
+      "loss": 0.3302,
+      "step": 1871
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.638810952848328e-06,
+      "loss": 0.3116,
+      "step": 1872
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.628913471767756e-06,
+      "loss": 0.3427,
+      "step": 1873
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.619018451158874e-06,
+      "loss": 0.3454,
+      "step": 1874
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.609125901289722e-06,
+      "loss": 0.3173,
+      "step": 1875
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.599235832425781e-06,
+      "loss": 0.3636,
+      "step": 1876
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.589348254829954e-06,
+      "loss": 0.3589,
+      "step": 1877
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.579463178762559e-06,
+      "loss": 0.2967,
+      "step": 1878
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.569580614481318e-06,
+      "loss": 0.3481,
+      "step": 1879
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.559700572241354e-06,
+      "loss": 0.306,
+      "step": 1880
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.549823062295164e-06,
+      "loss": 0.2933,
+      "step": 1881
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.539948094892621e-06,
+      "loss": 0.3309,
+      "step": 1882
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.530075680280956e-06,
+      "loss": 0.322,
+      "step": 1883
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.520205828704753e-06,
+      "loss": 0.3323,
+      "step": 1884
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.510338550405942e-06,
+      "loss": 0.3084,
+      "step": 1885
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.500473855623773e-06,
+      "loss": 0.3173,
+      "step": 1886
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.490611754594823e-06,
+      "loss": 0.3407,
+      "step": 1887
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.480752257552972e-06,
+      "loss": 0.3233,
+      "step": 1888
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.470895374729401e-06,
+      "loss": 0.3234,
+      "step": 1889
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.461041116352579e-06,
+      "loss": 0.3467,
+      "step": 1890
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.451189492648246e-06,
+      "loss": 0.3357,
+      "step": 1891
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.441340513839412e-06,
+      "loss": 0.3223,
+      "step": 1892
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.431494190146346e-06,
+      "loss": 0.3227,
+      "step": 1893
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.42165053178656e-06,
+      "loss": 0.3235,
+      "step": 1894
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.411809548974792e-06,
+      "loss": 0.3447,
+      "step": 1895
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.401971251923015e-06,
+      "loss": 0.364,
+      "step": 1896
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.392135650840406e-06,
+      "loss": 0.3442,
+      "step": 1897
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.38230275593335e-06,
+      "loss": 0.295,
+      "step": 1898
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.372472577405424e-06,
+      "loss": 0.3458,
+      "step": 1899
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.362645125457381e-06,
+      "loss": 0.3128,
+      "step": 1900
+    },
+    {
+      "epoch": 1.79,
+      "eval_loss": 0.5777141451835632,
+      "eval_runtime": 98.3078,
+      "eval_samples_per_second": 7.039,
+      "eval_steps_per_second": 0.448,
+      "step": 1900
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.3528204102871495e-06,
+      "loss": 0.3343,
+      "step": 1901
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.342998442089817e-06,
+      "loss": 0.357,
+      "step": 1902
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.333179231057621e-06,
+      "loss": 0.321,
+      "step": 1903
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.323362787379935e-06,
+      "loss": 0.3088,
+      "step": 1904
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.313549121243264e-06,
+      "loss": 0.2855,
+      "step": 1905
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.303738242831228e-06,
+      "loss": 0.3278,
+      "step": 1906
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.2939301623245564e-06,
+      "loss": 0.3378,
+      "step": 1907
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.284124889901074e-06,
+      "loss": 0.3064,
+      "step": 1908
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.274322435735693e-06,
+      "loss": 0.3085,
+      "step": 1909
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.264522810000398e-06,
+      "loss": 0.2902,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.254726022864245e-06,
+      "loss": 0.3448,
+      "step": 1911
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.244932084493338e-06,
+      "loss": 0.3327,
+      "step": 1912
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.235141005050826e-06,
+      "loss": 0.3106,
+      "step": 1913
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.225352794696894e-06,
+      "loss": 0.3249,
+      "step": 1914
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.215567463588747e-06,
+      "loss": 0.3335,
+      "step": 1915
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.205785021880603e-06,
+      "loss": 0.2965,
+      "step": 1916
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.196005479723682e-06,
+      "loss": 0.3067,
+      "step": 1917
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.1862288472661946e-06,
+      "loss": 0.3559,
+      "step": 1918
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.176455134653335e-06,
+      "loss": 0.3741,
+      "step": 1919
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.166684352027265e-06,
+      "loss": 0.3233,
+      "step": 1920
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.156916509527101e-06,
+      "loss": 0.3282,
+      "step": 1921
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.147151617288919e-06,
+      "loss": 0.3059,
+      "step": 1922
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.137389685445726e-06,
+      "loss": 0.3059,
+      "step": 1923
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.127630724127457e-06,
+      "loss": 0.3115,
+      "step": 1924
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.1178747434609685e-06,
+      "loss": 0.2881,
+      "step": 1925
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.108121753570021e-06,
+      "loss": 0.3323,
+      "step": 1926
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.098371764575271e-06,
+      "loss": 0.3429,
+      "step": 1927
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.088624786594266e-06,
+      "loss": 0.3176,
+      "step": 1928
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.078880829741424e-06,
+      "loss": 0.3078,
+      "step": 1929
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.069139904128027e-06,
+      "loss": 0.346,
+      "step": 1930
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.059402019862216e-06,
+      "loss": 0.3319,
+      "step": 1931
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.049667187048974e-06,
+      "loss": 0.3107,
+      "step": 1932
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.039935415790115e-06,
+      "loss": 0.3191,
+      "step": 1933
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.03020671618428e-06,
+      "loss": 0.314,
+      "step": 1934
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.02048109832692e-06,
+      "loss": 0.3426,
+      "step": 1935
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.01075857231029e-06,
+      "loss": 0.3317,
+      "step": 1936
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.001039148223436e-06,
+      "loss": 0.3155,
+      "step": 1937
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.9913228361521836e-06,
+      "loss": 0.3449,
+      "step": 1938
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.98160964617913e-06,
+      "loss": 0.3184,
+      "step": 1939
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.971899588383635e-06,
+      "loss": 0.3206,
+      "step": 1940
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.962192672841802e-06,
+      "loss": 0.3228,
+      "step": 1941
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.95248890962648e-06,
+      "loss": 0.3077,
+      "step": 1942
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.942788308807245e-06,
+      "loss": 0.3176,
+      "step": 1943
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.9330908804503874e-06,
+      "loss": 0.3398,
+      "step": 1944
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.923396634618913e-06,
+      "loss": 0.3321,
+      "step": 1945
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.913705581372521e-06,
+      "loss": 0.344,
+      "step": 1946
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.904017730767594e-06,
+      "loss": 0.302,
+      "step": 1947
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.894333092857199e-06,
+      "loss": 0.3549,
+      "step": 1948
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.884651677691062e-06,
+      "loss": 0.2956,
+      "step": 1949
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.87497349531557e-06,
+      "loss": 0.3045,
+      "step": 1950
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.865298555773753e-06,
+      "loss": 0.318,
+      "step": 1951
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.855626869105273e-06,
+      "loss": 0.3209,
+      "step": 1952
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.845958445346425e-06,
+      "loss": 0.3451,
+      "step": 1953
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.836293294530109e-06,
+      "loss": 0.3368,
+      "step": 1954
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.826631426685834e-06,
+      "loss": 0.3587,
+      "step": 1955
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.8169728518397e-06,
+      "loss": 0.2992,
+      "step": 1956
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.807317580014392e-06,
+      "loss": 0.3227,
+      "step": 1957
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.797665621229164e-06,
+      "loss": 0.3106,
+      "step": 1958
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.788016985499834e-06,
+      "loss": 0.3347,
+      "step": 1959
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.778371682838775e-06,
+      "loss": 0.3581,
+      "step": 1960
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.768729723254892e-06,
+      "loss": 0.3118,
+      "step": 1961
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.759091116753632e-06,
+      "loss": 0.3052,
+      "step": 1962
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.749455873336956e-06,
+      "loss": 0.3185,
+      "step": 1963
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.739824003003339e-06,
+      "loss": 0.3143,
+      "step": 1964
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.730195515747748e-06,
+      "loss": 0.3511,
+      "step": 1965
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.72057042156165e-06,
+      "loss": 0.3057,
+      "step": 1966
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.710948730432982e-06,
+      "loss": 0.3166,
+      "step": 1967
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.701330452346156e-06,
+      "loss": 0.3475,
+      "step": 1968
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.691715597282036e-06,
+      "loss": 0.3048,
+      "step": 1969
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.682104175217942e-06,
+      "loss": 0.3093,
+      "step": 1970
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.672496196127624e-06,
+      "loss": 0.341,
+      "step": 1971
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.662891669981266e-06,
+      "loss": 0.3126,
+      "step": 1972
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.653290606745463e-06,
+      "loss": 0.3279,
+      "step": 1973
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.643693016383218e-06,
+      "loss": 0.3116,
+      "step": 1974
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.6340989088539325e-06,
+      "loss": 0.3208,
+      "step": 1975
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.624508294113388e-06,
+      "loss": 0.3312,
+      "step": 1976
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.614921182113754e-06,
+      "loss": 0.3323,
+      "step": 1977
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.6053375828035525e-06,
+      "loss": 0.3417,
+      "step": 1978
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.595757506127667e-06,
+      "loss": 0.3165,
+      "step": 1979
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.586180962027322e-06,
+      "loss": 0.3364,
+      "step": 1980
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.576607960440076e-06,
+      "loss": 0.3235,
+      "step": 1981
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.567038511299817e-06,
+      "loss": 0.2917,
+      "step": 1982
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.557472624536743e-06,
+      "loss": 0.3307,
+      "step": 1983
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.547910310077354e-06,
+      "loss": 0.3085,
+      "step": 1984
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.538351577844444e-06,
+      "loss": 0.3023,
+      "step": 1985
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.528796437757091e-06,
+      "loss": 0.3324,
+      "step": 1986
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.5192448997306435e-06,
+      "loss": 0.2859,
+      "step": 1987
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.509696973676714e-06,
+      "loss": 0.3115,
+      "step": 1988
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.500152669503166e-06,
+      "loss": 0.2929,
+      "step": 1989
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.4906119971141025e-06,
+      "loss": 0.338,
+      "step": 1990
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.481074966409861e-06,
+      "loss": 0.2805,
+      "step": 1991
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.471541587287003e-06,
+      "loss": 0.3133,
+      "step": 1992
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.462011869638292e-06,
+      "loss": 0.317,
+      "step": 1993
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.452485823352699e-06,
+      "loss": 0.302,
+      "step": 1994
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.442963458315381e-06,
+      "loss": 0.3386,
+      "step": 1995
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.433444784407679e-06,
+      "loss": 0.3089,
+      "step": 1996
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.423929811507099e-06,
+      "loss": 0.3437,
+      "step": 1997
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.414418549487309e-06,
+      "loss": 0.3271,
+      "step": 1998
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.404911008218131e-06,
+      "loss": 0.3221,
+      "step": 1999
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.395407197565518e-06,
+      "loss": 0.3125,
+      "step": 2000
+    },
+    {
+      "epoch": 1.89,
+      "eval_loss": 0.5761173963546753,
+      "eval_runtime": 98.4004,
+      "eval_samples_per_second": 7.032,
+      "eval_steps_per_second": 0.447,
+      "step": 2000
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.385907127391555e-06,
+      "loss": 0.338,
+      "step": 2001
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.376410807554445e-06,
+      "loss": 0.3654,
+      "step": 2002
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.366918247908502e-06,
+      "loss": 0.3491,
+      "step": 2003
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.357429458304135e-06,
+      "loss": 0.3128,
+      "step": 2004
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.347944448587842e-06,
+      "loss": 0.3164,
+      "step": 2005
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.338463228602196e-06,
+      "loss": 0.3253,
+      "step": 2006
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.328985808185842e-06,
+      "loss": 0.3394,
+      "step": 2007
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.319512197173479e-06,
+      "loss": 0.3314,
+      "step": 2008
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.310042405395855e-06,
+      "loss": 0.3357,
+      "step": 2009
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.300576442679755e-06,
+      "loss": 0.2974,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.291114318847987e-06,
+      "loss": 0.3112,
+      "step": 2011
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.2816560437193786e-06,
+      "loss": 0.3086,
+      "step": 2012
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.272201627108762e-06,
+      "loss": 0.2942,
+      "step": 2013
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.262751078826968e-06,
+      "loss": 0.3442,
+      "step": 2014
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.253304408680806e-06,
+      "loss": 0.3358,
+      "step": 2015
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.243861626473073e-06,
+      "loss": 0.3316,
+      "step": 2016
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.234422742002524e-06,
+      "loss": 0.3128,
+      "step": 2017
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.224987765063868e-06,
+      "loss": 0.3597,
+      "step": 2018
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.215556705447761e-06,
+      "loss": 0.3377,
+      "step": 2019
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.206129572940798e-06,
+      "loss": 0.3578,
+      "step": 2020
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.196706377325492e-06,
+      "loss": 0.3202,
+      "step": 2021
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.187287128380276e-06,
+      "loss": 0.3071,
+      "step": 2022
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.177871835879483e-06,
+      "loss": 0.313,
+      "step": 2023
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.168460509593343e-06,
+      "loss": 0.3151,
+      "step": 2024
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.159053159287975e-06,
+      "loss": 0.3018,
+      "step": 2025
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.1496497947253655e-06,
+      "loss": 0.3016,
+      "step": 2026
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.1402504256633635e-06,
+      "loss": 0.3314,
+      "step": 2027
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.1308550618556765e-06,
+      "loss": 0.3333,
+      "step": 2028
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.121463713051856e-06,
+      "loss": 0.3149,
+      "step": 2029
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.112076388997285e-06,
+      "loss": 0.3301,
+      "step": 2030
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.102693099433172e-06,
+      "loss": 0.2957,
+      "step": 2031
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.093313854096534e-06,
+      "loss": 0.3059,
+      "step": 2032
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.083938662720197e-06,
+      "loss": 0.3255,
+      "step": 2033
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.074567535032778e-06,
+      "loss": 0.3188,
+      "step": 2034
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.06520048075868e-06,
+      "loss": 0.3138,
+      "step": 2035
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.0558375096180724e-06,
+      "loss": 0.3286,
+      "step": 2036
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.046478631326891e-06,
+      "loss": 0.3198,
+      "step": 2037
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.037123855596828e-06,
+      "loss": 0.3426,
+      "step": 2038
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.027773192135313e-06,
+      "loss": 0.2955,
+      "step": 2039
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.0184266506455125e-06,
+      "loss": 0.3348,
+      "step": 2040
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.00908424082631e-06,
+      "loss": 0.3563,
+      "step": 2041
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.999745972372308e-06,
+      "loss": 0.3336,
+      "step": 2042
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.99041185497381e-06,
+      "loss": 0.3118,
+      "step": 2043
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.981081898316809e-06,
+      "loss": 0.3152,
+      "step": 2044
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.971756112082982e-06,
+      "loss": 0.2987,
+      "step": 2045
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.962434505949679e-06,
+      "loss": 0.3158,
+      "step": 2046
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.953117089589911e-06,
+      "loss": 0.3142,
+      "step": 2047
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.943803872672344e-06,
+      "loss": 0.2926,
+      "step": 2048
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.934494864861281e-06,
+      "loss": 0.3038,
+      "step": 2049
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.925190075816661e-06,
+      "loss": 0.309,
+      "step": 2050
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.915889515194049e-06,
+      "loss": 0.3301,
+      "step": 2051
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.906593192644615e-06,
+      "loss": 0.3095,
+      "step": 2052
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.897301117815134e-06,
+      "loss": 0.3184,
+      "step": 2053
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.888013300347974e-06,
+      "loss": 0.3321,
+      "step": 2054
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.878729749881082e-06,
+      "loss": 0.3135,
+      "step": 2055
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.869450476047984e-06,
+      "loss": 0.3392,
+      "step": 2056
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.86017548847776e-06,
+      "loss": 0.3078,
+      "step": 2057
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.850904796795045e-06,
+      "loss": 0.285,
+      "step": 2058
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.84163841062002e-06,
+      "loss": 0.3522,
+      "step": 2059
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.832376339568394e-06,
+      "loss": 0.3761,
+      "step": 2060
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.823118593251402e-06,
+      "loss": 0.3437,
+      "step": 2061
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.813865181275782e-06,
+      "loss": 0.3208,
+      "step": 2062
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.804616113243784e-06,
+      "loss": 0.3271,
+      "step": 2063
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.795371398753153e-06,
+      "loss": 0.3567,
+      "step": 2064
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.786131047397101e-06,
+      "loss": 0.3133,
+      "step": 2065
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.776895068764331e-06,
+      "loss": 0.314,
+      "step": 2066
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.76766347243899e-06,
+      "loss": 0.3101,
+      "step": 2067
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.758436268000696e-06,
+      "loss": 0.3422,
+      "step": 2068
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.749213465024493e-06,
+      "loss": 0.2809,
+      "step": 2069
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.73999507308087e-06,
+      "loss": 0.3145,
+      "step": 2070
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.730781101735727e-06,
+      "loss": 0.2908,
+      "step": 2071
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.72157156055039e-06,
+      "loss": 0.3375,
+      "step": 2072
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.712366459081578e-06,
+      "loss": 0.3193,
+      "step": 2073
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.703165806881411e-06,
+      "loss": 0.2946,
+      "step": 2074
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.69396961349738e-06,
+      "loss": 0.3123,
+      "step": 2075
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.684777888472359e-06,
+      "loss": 0.3241,
+      "step": 2076
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.675590641344588e-06,
+      "loss": 0.3021,
+      "step": 2077
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.666407881647646e-06,
+      "loss": 0.3379,
+      "step": 2078
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.657229618910472e-06,
+      "loss": 0.3162,
+      "step": 2079
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.648055862657321e-06,
+      "loss": 0.2978,
+      "step": 2080
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.638886622407793e-06,
+      "loss": 0.3317,
+      "step": 2081
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.629721907676778e-06,
+      "loss": 0.3348,
+      "step": 2082
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.620561727974489e-06,
+      "loss": 0.3164,
+      "step": 2083
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.611406092806429e-06,
+      "loss": 0.3581,
+      "step": 2084
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.602255011673374e-06,
+      "loss": 0.3254,
+      "step": 2085
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.593108494071391e-06,
+      "loss": 0.3292,
+      "step": 2086
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.583966549491792e-06,
+      "loss": 0.3273,
+      "step": 2087
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.574829187421166e-06,
+      "loss": 0.3118,
+      "step": 2088
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.565696417341324e-06,
+      "loss": 0.3303,
+      "step": 2089
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.556568248729334e-06,
+      "loss": 0.2907,
+      "step": 2090
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.547444691057469e-06,
+      "loss": 0.295,
+      "step": 2091
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.538325753793227e-06,
+      "loss": 0.3245,
+      "step": 2092
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.529211446399321e-06,
+      "loss": 0.3041,
+      "step": 2093
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.520101778333636e-06,
+      "loss": 0.2954,
+      "step": 2094
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.510996759049268e-06,
+      "loss": 0.3276,
+      "step": 2095
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.501896397994465e-06,
+      "loss": 0.3224,
+      "step": 2096
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.492800704612664e-06,
+      "loss": 0.3231,
+      "step": 2097
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.48370968834244e-06,
+      "loss": 0.3064,
+      "step": 2098
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.474623358617528e-06,
+      "loss": 0.3022,
+      "step": 2099
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.465541724866787e-06,
+      "loss": 0.3169,
+      "step": 2100
+    },
+    {
+      "epoch": 1.98,
+      "eval_loss": 0.5736315250396729,
+      "eval_runtime": 98.7761,
+      "eval_samples_per_second": 7.006,
+      "eval_steps_per_second": 0.445,
+      "step": 2100
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.456464796514216e-06,
+      "loss": 0.3218,
+      "step": 2101
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.447392582978925e-06,
+      "loss": 0.3305,
+      "step": 2102
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.438325093675127e-06,
+      "loss": 0.3109,
+      "step": 2103
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.429262338012144e-06,
+      "loss": 0.3316,
+      "step": 2104
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.4202043253943715e-06,
+      "loss": 0.2759,
+      "step": 2105
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.411151065221298e-06,
+      "loss": 0.3586,
+      "step": 2106
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.402102566887466e-06,
+      "loss": 0.3236,
+      "step": 2107
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.393058839782492e-06,
+      "loss": 0.332,
+      "step": 2108
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.384019893291026e-06,
+      "loss": 0.326,
+      "step": 2109
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.374985736792766e-06,
+      "loss": 0.2691,
+      "step": 2110
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.365956379662444e-06,
+      "loss": 0.3365,
+      "step": 2111
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.356931831269798e-06,
+      "loss": 0.3473,
+      "step": 2112
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.34791210097959e-06,
+      "loss": 0.3284,
+      "step": 2113
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.338897198151569e-06,
+      "loss": 0.3151,
+      "step": 2114
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.329887132140488e-06,
+      "loss": 0.3085,
+      "step": 2115
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.3208819122960674e-06,
+      "loss": 0.279,
+      "step": 2116
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.3118815479630095e-06,
+      "loss": 0.3412,
+      "step": 2117
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.302886048480978e-06,
+      "loss": 0.324,
+      "step": 2118
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.293895423184577e-06,
+      "loss": 0.3282,
+      "step": 2119
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.284909681403367e-06,
+      "loss": 0.2847,
+      "step": 2120
+    }
+  ],
+  "max_steps": 3180,
+  "num_train_epochs": 3,
+  "total_flos": 1.3493648433453466e+18,
+  "trial_name": null,
+  "trial_params": null
+}