diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,5373 @@
+{
+  "best_metric": 0.5186672210693359,
+  "best_model_checkpoint": "/workspace/dolphin-2.6-mistral-7b/checkpoint-730",
+  "epoch": 2.3532627263409633,
+  "eval_steps": 73,
+  "global_step": 876,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 1.1255,
+      "step": 1
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 1.1614,
+      "step": 2
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5e-06,
+      "loss": 1.1455,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.0631,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5e-06,
+      "loss": 0.9595,
+      "step": 5
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3e-06,
+      "loss": 0.9296,
+      "step": 6
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5e-06,
+      "loss": 0.8787,
+      "step": 7
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.8616,
+      "step": 8
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.5e-06,
+      "loss": 0.852,
+      "step": 9
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5e-06,
+      "loss": 0.8329,
+      "step": 10
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.999994132223223e-06,
+      "loss": 0.8109,
+      "step": 11
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.999976528920434e-06,
+      "loss": 0.7917,
+      "step": 12
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.999947190174268e-06,
+      "loss": 0.8021,
+      "step": 13
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.999906116122448e-06,
+      "loss": 0.7972,
+      "step": 14
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.999853306957783e-06,
+      "loss": 0.7847,
+      "step": 15
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9997887629281724e-06,
+      "loss": 0.7669,
+      "step": 16
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9997124843366004e-06,
+      "loss": 0.7892,
+      "step": 17
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.999624471541135e-06,
+      "loss": 0.7752,
+      "step": 18
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.999524724954928e-06,
+      "loss": 0.7702,
+      "step": 19
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.99941324504621e-06,
+      "loss": 0.7345,
+      "step": 20
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.999290032338296e-06,
+      "loss": 0.7482,
+      "step": 21
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9991550874095715e-06,
+      "loss": 0.745,
+      "step": 22
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9990084108934975e-06,
+      "loss": 0.7291,
+      "step": 23
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.998850003478607e-06,
+      "loss": 0.7275,
+      "step": 24
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.998679865908499e-06,
+      "loss": 0.7432,
+      "step": 25
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.998497998981838e-06,
+      "loss": 0.718,
+      "step": 26
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.998304403552346e-06,
+      "loss": 0.7137,
+      "step": 27
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.998099080528803e-06,
+      "loss": 0.7057,
+      "step": 28
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.997882030875043e-06,
+      "loss": 0.7137,
+      "step": 29
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9976532556099425e-06,
+      "loss": 0.6866,
+      "step": 30
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9974127558074236e-06,
+      "loss": 0.6794,
+      "step": 31
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.997160532596447e-06,
+      "loss": 0.6774,
+      "step": 32
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.996896587161004e-06,
+      "loss": 0.6922,
+      "step": 33
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.996620920740111e-06,
+      "loss": 0.6962,
+      "step": 34
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.99633353462781e-06,
+      "loss": 0.687,
+      "step": 35
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.996034430173154e-06,
+      "loss": 0.6773,
+      "step": 36
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.995723608780204e-06,
+      "loss": 0.6822,
+      "step": 37
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.995401071908026e-06,
+      "loss": 0.6876,
+      "step": 38
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9950668210706795e-06,
+      "loss": 0.6959,
+      "step": 39
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.994720857837211e-06,
+      "loss": 0.6893,
+      "step": 40
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.99436318383165e-06,
+      "loss": 0.6734,
+      "step": 41
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9939938007329955e-06,
+      "loss": 0.6591,
+      "step": 42
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.993612710275216e-06,
+      "loss": 0.6988,
+      "step": 43
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.993219914247233e-06,
+      "loss": 0.66,
+      "step": 44
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.992815414492917e-06,
+      "loss": 0.6642,
+      "step": 45
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.992399212911082e-06,
+      "loss": 0.6652,
+      "step": 46
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.991971311455468e-06,
+      "loss": 0.6777,
+      "step": 47
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.99153171213474e-06,
+      "loss": 0.6647,
+      "step": 48
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.991080417012476e-06,
+      "loss": 0.6737,
+      "step": 49
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.990617428207153e-06,
+      "loss": 0.6442,
+      "step": 50
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.990142747892145e-06,
+      "loss": 0.6534,
+      "step": 51
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.989656378295704e-06,
+      "loss": 0.6519,
+      "step": 52
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.989158321700959e-06,
+      "loss": 0.669,
+      "step": 53
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.988648580445896e-06,
+      "loss": 0.6409,
+      "step": 54
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.988127156923355e-06,
+      "loss": 0.6372,
+      "step": 55
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.987594053581013e-06,
+      "loss": 0.6687,
+      "step": 56
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.987049272921374e-06,
+      "loss": 0.6508,
+      "step": 57
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.98649281750176e-06,
+      "loss": 0.649,
+      "step": 58
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.985924689934297e-06,
+      "loss": 0.6501,
+      "step": 59
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.985344892885899e-06,
+      "loss": 0.6478,
+      "step": 60
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.984753429078264e-06,
+      "loss": 0.6582,
+      "step": 61
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.984150301287854e-06,
+      "loss": 0.6382,
+      "step": 62
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.9835355123458815e-06,
+      "loss": 0.6563,
+      "step": 63
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.9829090651383064e-06,
+      "loss": 0.6414,
+      "step": 64
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.9822709626058065e-06,
+      "loss": 0.6518,
+      "step": 65
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.981621207743779e-06,
+      "loss": 0.6379,
+      "step": 66
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.980959803602315e-06,
+      "loss": 0.6356,
+      "step": 67
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.980286753286196e-06,
+      "loss": 0.6546,
+      "step": 68
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.979602059954863e-06,
+      "loss": 0.6511,
+      "step": 69
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.978905726822424e-06,
+      "loss": 0.6521,
+      "step": 70
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.978197757157618e-06,
+      "loss": 0.6404,
+      "step": 71
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.9774781542838115e-06,
+      "loss": 0.6349,
+      "step": 72
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.97674692157898e-06,
+      "loss": 0.6397,
+      "step": 73
+    },
+    {
+      "epoch": 0.2,
+      "eval_loss": 0.6273066401481628,
+      "eval_runtime": 70.5832,
+      "eval_samples_per_second": 71.745,
+      "eval_steps_per_second": 5.979,
+      "step": 73
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.976004062475691e-06,
+      "loss": 0.616,
+      "step": 74
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.975249580461092e-06,
+      "loss": 0.6359,
+      "step": 75
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.9744834790768855e-06,
+      "loss": 0.6313,
+      "step": 76
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.973705761919324e-06,
+      "loss": 0.6318,
+      "step": 77
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.972916432639182e-06,
+      "loss": 0.621,
+      "step": 78
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.972115494941746e-06,
+      "loss": 0.6438,
+      "step": 79
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.971302952586796e-06,
+      "loss": 0.6325,
+      "step": 80
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.970478809388585e-06,
+      "loss": 0.6426,
+      "step": 81
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.969643069215824e-06,
+      "loss": 0.6391,
+      "step": 82
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.968795735991661e-06,
+      "loss": 0.6215,
+      "step": 83
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.967936813693668e-06,
+      "loss": 0.6167,
+      "step": 84
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.967066306353816e-06,
+      "loss": 0.6232,
+      "step": 85
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.966184218058457e-06,
+      "loss": 0.6141,
+      "step": 86
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.965290552948312e-06,
+      "loss": 0.6269,
+      "step": 87
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.964385315218441e-06,
+      "loss": 0.6144,
+      "step": 88
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.9634685091182306e-06,
+      "loss": 0.633,
+      "step": 89
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.962540138951371e-06,
+      "loss": 0.6173,
+      "step": 90
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.9616002090758385e-06,
+      "loss": 0.6209,
+      "step": 91
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.960648723903872e-06,
+      "loss": 0.6298,
+      "step": 92
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.9596856879019524e-06,
+      "loss": 0.6239,
+      "step": 93
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.958711105590786e-06,
+      "loss": 0.6347,
+      "step": 94
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.957724981545276e-06,
+      "loss": 0.6137,
+      "step": 95
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.956727320394507e-06,
+      "loss": 0.6205,
+      "step": 96
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.9557181268217225e-06,
+      "loss": 0.588,
+      "step": 97
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.954697405564299e-06,
+      "loss": 0.6211,
+      "step": 98
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.953665161413731e-06,
+      "loss": 0.6264,
+      "step": 99
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.952621399215598e-06,
+      "loss": 0.6031,
+      "step": 100
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.951566123869553e-06,
+      "loss": 0.6186,
+      "step": 101
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.950499340329291e-06,
+      "loss": 0.6177,
+      "step": 102
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.9494210536025306e-06,
+      "loss": 0.6364,
+      "step": 103
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.948331268750988e-06,
+      "loss": 0.6214,
+      "step": 104
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.947229990890356e-06,
+      "loss": 0.6159,
+      "step": 105
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.946117225190274e-06,
+      "loss": 0.5957,
+      "step": 106
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.944992976874313e-06,
+      "loss": 0.6088,
+      "step": 107
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.9438572512199425e-06,
+      "loss": 0.6163,
+      "step": 108
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.942710053558511e-06,
+      "loss": 0.5964,
+      "step": 109
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.941551389275217e-06,
+      "loss": 0.6132,
+      "step": 110
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.940381263809089e-06,
+      "loss": 0.605,
+      "step": 111
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.9391996826529535e-06,
+      "loss": 0.6063,
+      "step": 112
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.938006651353414e-06,
+      "loss": 0.5971,
+      "step": 113
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.936802175510824e-06,
+      "loss": 0.607,
+      "step": 114
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.935586260779261e-06,
+      "loss": 0.6128,
+      "step": 115
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.934358912866497e-06,
+      "loss": 0.5985,
+      "step": 116
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.933120137533975e-06,
+      "loss": 0.6132,
+      "step": 117
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.931869940596779e-06,
+      "loss": 0.611,
+      "step": 118
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.9306083279236124e-06,
+      "loss": 0.5972,
+      "step": 119
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.929335305436764e-06,
+      "loss": 0.6025,
+      "step": 120
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.928050879112083e-06,
+      "loss": 0.618,
+      "step": 121
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.926755054978951e-06,
+      "loss": 0.6072,
+      "step": 122
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.925447839120254e-06,
+      "loss": 0.6045,
+      "step": 123
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.924129237672351e-06,
+      "loss": 0.6133,
+      "step": 124
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.922799256825052e-06,
+      "loss": 0.6027,
+      "step": 125
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.921457902821578e-06,
+      "loss": 0.5949,
+      "step": 126
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.920105181958545e-06,
+      "loss": 0.5807,
+      "step": 127
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.918741100585921e-06,
+      "loss": 0.6019,
+      "step": 128
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.917365665107009e-06,
+      "loss": 0.5998,
+      "step": 129
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.915978881978407e-06,
+      "loss": 0.6053,
+      "step": 130
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.91458075770998e-06,
+      "loss": 0.6015,
+      "step": 131
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.913171298864836e-06,
+      "loss": 0.5926,
+      "step": 132
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.911750512059285e-06,
+      "loss": 0.5923,
+      "step": 133
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.910318403962814e-06,
+      "loss": 0.5809,
+      "step": 134
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.908874981298058e-06,
+      "loss": 0.6,
+      "step": 135
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.907420250840761e-06,
+      "loss": 0.5861,
+      "step": 136
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.90595421941975e-06,
+      "loss": 0.5969,
+      "step": 137
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.904476893916901e-06,
+      "loss": 0.5932,
+      "step": 138
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.902988281267108e-06,
+      "loss": 0.6009,
+      "step": 139
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.901488388458247e-06,
+      "loss": 0.5853,
+      "step": 140
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.899977222531148e-06,
+      "loss": 0.605,
+      "step": 141
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.898454790579558e-06,
+      "loss": 0.5774,
+      "step": 142
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.89692109975011e-06,
+      "loss": 0.5982,
+      "step": 143
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.895376157242288e-06,
+      "loss": 0.5874,
+      "step": 144
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.893819970308394e-06,
+      "loss": 0.5886,
+      "step": 145
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.8922525462535154e-06,
+      "loss": 0.5784,
+      "step": 146
+    },
+    {
+      "epoch": 0.4,
+      "eval_loss": 0.5756770372390747,
+      "eval_runtime": 70.7014,
+      "eval_samples_per_second": 71.625,
+      "eval_steps_per_second": 5.969,
+      "step": 146
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.890673892435487e-06,
+      "loss": 0.589,
+      "step": 147
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.889084016264858e-06,
+      "loss": 0.5799,
+      "step": 148
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.887482925204861e-06,
+      "loss": 0.5827,
+      "step": 149
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.885870626771371e-06,
+      "loss": 0.5763,
+      "step": 150
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.884247128532874e-06,
+      "loss": 0.584,
+      "step": 151
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.88261243811043e-06,
+      "loss": 0.5953,
+      "step": 152
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.880966563177638e-06,
+      "loss": 0.5745,
+      "step": 153
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.879309511460601e-06,
+      "loss": 0.5961,
+      "step": 154
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": 0.5848,
+      "step": 155
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.875961908840486e-06,
+      "loss": 0.5758,
+      "step": 156
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.874271373651798e-06,
+      "loss": 0.5779,
+      "step": 157
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.872569693107564e-06,
+      "loss": 0.5827,
+      "step": 158
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.8708568751958516e-06,
+      "loss": 0.5683,
+      "step": 159
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.869132927957007e-06,
+      "loss": 0.587,
+      "step": 160
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.867397859483619e-06,
+      "loss": 0.5712,
+      "step": 161
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.865651677920484e-06,
+      "loss": 0.5786,
+      "step": 162
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.863894391464566e-06,
+      "loss": 0.5795,
+      "step": 163
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.862126008364954e-06,
+      "loss": 0.5824,
+      "step": 164
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.860346536922834e-06,
+      "loss": 0.5698,
+      "step": 165
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.858555985491434e-06,
+      "loss": 0.5971,
+      "step": 166
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.856754362476003e-06,
+      "loss": 0.5527,
+      "step": 167
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.854941676333756e-06,
+      "loss": 0.573,
+      "step": 168
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.853117935573844e-06,
+      "loss": 0.5791,
+      "step": 169
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.85128314875731e-06,
+      "loss": 0.5793,
+      "step": 170
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.849437324497049e-06,
+      "loss": 0.5636,
+      "step": 171
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.8475804714577704e-06,
+      "loss": 0.5799,
+      "step": 172
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.8457125983559506e-06,
+      "loss": 0.5865,
+      "step": 173
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.843833713959803e-06,
+      "loss": 0.5781,
+      "step": 174
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.841943827089223e-06,
+      "loss": 0.5907,
+      "step": 175
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.840042946615761e-06,
+      "loss": 0.5689,
+      "step": 176
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.8381310814625705e-06,
+      "loss": 0.5756,
+      "step": 177
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.836208240604369e-06,
+      "loss": 0.5755,
+      "step": 178
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.834274433067398e-06,
+      "loss": 0.5661,
+      "step": 179
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.832329667929378e-06,
+      "loss": 0.584,
+      "step": 180
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.830373954319467e-06,
+      "loss": 0.5673,
+      "step": 181
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.828407301418218e-06,
+      "loss": 0.5537,
+      "step": 182
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.826429718457534e-06,
+      "loss": 0.5636,
+      "step": 183
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.824441214720629e-06,
+      "loss": 0.5688,
+      "step": 184
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.822441799541979e-06,
+      "loss": 0.573,
+      "step": 185
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.820431482307282e-06,
+      "loss": 0.5746,
+      "step": 186
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.818410272453411e-06,
+      "loss": 0.5639,
+      "step": 187
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.816378179468375e-06,
+      "loss": 0.5686,
+      "step": 188
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.814335212891266e-06,
+      "loss": 0.5671,
+      "step": 189
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.812281382312222e-06,
+      "loss": 0.5697,
+      "step": 190
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.81021669737238e-06,
+      "loss": 0.5853,
+      "step": 191
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.808141167763827e-06,
+      "loss": 0.5627,
+      "step": 192
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.806054803229559e-06,
+      "loss": 0.5818,
+      "step": 193
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.8039576135634314e-06,
+      "loss": 0.5935,
+      "step": 194
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.801849608610119e-06,
+      "loss": 0.5656,
+      "step": 195
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.799730798265064e-06,
+      "loss": 0.5716,
+      "step": 196
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.79760119247443e-06,
+      "loss": 0.5605,
+      "step": 197
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.795460801235058e-06,
+      "loss": 0.5776,
+      "step": 198
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.793309634594418e-06,
+      "loss": 0.555,
+      "step": 199
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.7911477026505656e-06,
+      "loss": 0.5692,
+      "step": 200
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.788975015552085e-06,
+      "loss": 0.5529,
+      "step": 201
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.786791583498051e-06,
+      "loss": 0.5721,
+      "step": 202
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.784597416737978e-06,
+      "loss": 0.5488,
+      "step": 203
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.782392525571771e-06,
+      "loss": 0.5598,
+      "step": 204
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.780176920349675e-06,
+      "loss": 0.5603,
+      "step": 205
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.777950611472234e-06,
+      "loss": 0.5513,
+      "step": 206
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.775713609390234e-06,
+      "loss": 0.5668,
+      "step": 207
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.773465924604657e-06,
+      "loss": 0.5585,
+      "step": 208
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.771207567666635e-06,
+      "loss": 0.5636,
+      "step": 209
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.7689385491773934e-06,
+      "loss": 0.5654,
+      "step": 210
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.766658879788208e-06,
+      "loss": 0.5672,
+      "step": 211
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.7643685702003536e-06,
+      "loss": 0.5657,
+      "step": 212
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.762067631165049e-06,
+      "loss": 0.5686,
+      "step": 213
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.759756073483412e-06,
+      "loss": 0.547,
+      "step": 214
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.7574339080064046e-06,
+      "loss": 0.5575,
+      "step": 215
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.755101145634788e-06,
+      "loss": 0.5702,
+      "step": 216
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.752757797319064e-06,
+      "loss": 0.5498,
+      "step": 217
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.750403874059428e-06,
+      "loss": 0.5484,
+      "step": 218
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.748039386905719e-06,
+      "loss": 0.576,
+      "step": 219
+    },
+    {
+      "epoch": 0.6,
+      "eval_loss": 0.5469194054603577,
+      "eval_runtime": 70.7492,
+      "eval_samples_per_second": 71.577,
+      "eval_steps_per_second": 5.965,
+      "step": 219
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.745664346957362e-06,
+      "loss": 0.5703,
+      "step": 220
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.743278765363319e-06,
+      "loss": 0.5642,
+      "step": 221
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.74088265332204e-06,
+      "loss": 0.5533,
+      "step": 222
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.738476022081405e-06,
+      "loss": 0.5696,
+      "step": 223
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.736058882938674e-06,
+      "loss": 0.5581,
+      "step": 224
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.733631247240435e-06,
+      "loss": 0.5368,
+      "step": 225
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.731193126382544e-06,
+      "loss": 0.5589,
+      "step": 226
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.728744531810082e-06,
+      "loss": 0.5446,
+      "step": 227
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.726285475017295e-06,
+      "loss": 0.5661,
+      "step": 228
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.723815967547539e-06,
+      "loss": 0.5493,
+      "step": 229
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.721336020993228e-06,
+      "loss": 0.5648,
+      "step": 230
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.718845646995782e-06,
+      "loss": 0.5503,
+      "step": 231
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.716344857245568e-06,
+      "loss": 0.5604,
+      "step": 232
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.713833663481845e-06,
+      "loss": 0.5536,
+      "step": 233
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.711312077492714e-06,
+      "loss": 0.5531,
+      "step": 234
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.708780111115058e-06,
+      "loss": 0.5584,
+      "step": 235
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.706237776234487e-06,
+      "loss": 0.5694,
+      "step": 236
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.7036850847852835e-06,
+      "loss": 0.562,
+      "step": 237
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.7011220487503485e-06,
+      "loss": 0.5542,
+      "step": 238
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.698548680161138e-06,
+      "loss": 0.5655,
+      "step": 239
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.6959649910976165e-06,
+      "loss": 0.5529,
+      "step": 240
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.693370993688191e-06,
+      "loss": 0.5675,
+      "step": 241
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.690766700109659e-06,
+      "loss": 0.556,
+      "step": 242
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.688152122587153e-06,
+      "loss": 0.5566,
+      "step": 243
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.685527273394078e-06,
+      "loss": 0.5462,
+      "step": 244
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.682892164852057e-06,
+      "loss": 0.5596,
+      "step": 245
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.680246809330874e-06,
+      "loss": 0.5605,
+      "step": 246
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.677591219248413e-06,
+      "loss": 0.5508,
+      "step": 247
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.674925407070602e-06,
+      "loss": 0.5716,
+      "step": 248
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.672249385311353e-06,
+      "loss": 0.5368,
+      "step": 249
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.669563166532504e-06,
+      "loss": 0.5474,
+      "step": 250
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.666866763343762e-06,
+      "loss": 0.557,
+      "step": 251
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.664160188402641e-06,
+      "loss": 0.5523,
+      "step": 252
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.661443454414403e-06,
+      "loss": 0.5547,
+      "step": 253
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.6587165741319975e-06,
+      "loss": 0.5347,
+      "step": 254
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.655979560356006e-06,
+      "loss": 0.5491,
+      "step": 255
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.6532324259345755e-06,
+      "loss": 0.5724,
+      "step": 256
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.650475183763365e-06,
+      "loss": 0.5611,
+      "step": 257
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.647707846785478e-06,
+      "loss": 0.5471,
+      "step": 258
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.644930427991408e-06,
+      "loss": 0.5585,
+      "step": 259
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.642142940418973e-06,
+      "loss": 0.5425,
+      "step": 260
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.639345397153259e-06,
+      "loss": 0.5544,
+      "step": 261
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.636537811326551e-06,
+      "loss": 0.5396,
+      "step": 262
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.633720196118281e-06,
+      "loss": 0.5348,
+      "step": 263
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.630892564754957e-06,
+      "loss": 0.5505,
+      "step": 264
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.6280549305101065e-06,
+      "loss": 0.5358,
+      "step": 265
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.625207306704213e-06,
+      "loss": 0.556,
+      "step": 266
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.6223497067046555e-06,
+      "loss": 0.5546,
+      "step": 267
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.619482143925637e-06,
+      "loss": 0.535,
+      "step": 268
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.616604631828137e-06,
+      "loss": 0.5645,
+      "step": 269
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.61371718391983e-06,
+      "loss": 0.5615,
+      "step": 270
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.610819813755038e-06,
+      "loss": 0.546,
+      "step": 271
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.607912534934658e-06,
+      "loss": 0.5517,
+      "step": 272
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.604995361106101e-06,
+      "loss": 0.5346,
+      "step": 273
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.602068305963225e-06,
+      "loss": 0.5422,
+      "step": 274
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.599131383246277e-06,
+      "loss": 0.5463,
+      "step": 275
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.596184606741821e-06,
+      "loss": 0.5452,
+      "step": 276
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.593227990282679e-06,
+      "loss": 0.5416,
+      "step": 277
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.590261547747864e-06,
+      "loss": 0.5486,
+      "step": 278
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.587285293062514e-06,
+      "loss": 0.5444,
+      "step": 279
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.584299240197826e-06,
+      "loss": 0.5399,
+      "step": 280
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.581303403170995e-06,
+      "loss": 0.5495,
+      "step": 281
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.578297796045142e-06,
+      "loss": 0.5382,
+      "step": 282
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.5752824329292534e-06,
+      "loss": 0.5349,
+      "step": 283
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.572257327978111e-06,
+      "loss": 0.5627,
+      "step": 284
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.569222495392227e-06,
+      "loss": 0.5395,
+      "step": 285
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.566177949417778e-06,
+      "loss": 0.5602,
+      "step": 286
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.5631237043465356e-06,
+      "loss": 0.5432,
+      "step": 287
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.560059774515805e-06,
+      "loss": 0.5407,
+      "step": 288
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.556986174308349e-06,
+      "loss": 0.5434,
+      "step": 289
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.553902918152329e-06,
+      "loss": 0.5449,
+      "step": 290
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.550810020521231e-06,
+      "loss": 0.5487,
+      "step": 291
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.547707495933802e-06,
+      "loss": 0.5619,
+      "step": 292
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 0.536051869392395,
+      "eval_runtime": 70.7206,
+      "eval_samples_per_second": 71.606,
+      "eval_steps_per_second": 5.967,
+      "step": 292
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.5445953589539784e-06,
+      "loss": 0.5463,
+      "step": 293
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.541473624190822e-06,
+      "loss": 0.5608,
+      "step": 294
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.5383423062984455e-06,
+      "loss": 0.5606,
+      "step": 295
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.535201419975948e-06,
+      "loss": 0.5392,
+      "step": 296
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.5320509799673476e-06,
+      "loss": 0.5356,
+      "step": 297
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.528891001061506e-06,
+      "loss": 0.5771,
+      "step": 298
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.5257214980920636e-06,
+      "loss": 0.5488,
+      "step": 299
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.522542485937369e-06,
+      "loss": 0.5658,
+      "step": 300
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.51935397952041e-06,
+      "loss": 0.5461,
+      "step": 301
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.516155993808741e-06,
+      "loss": 0.5524,
+      "step": 302
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.512948543814415e-06,
+      "loss": 0.5534,
+      "step": 303
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.5097316445939135e-06,
+      "loss": 0.5446,
+      "step": 304
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.5065053112480725e-06,
+      "loss": 0.5407,
+      "step": 305
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.503269558922015e-06,
+      "loss": 0.544,
+      "step": 306
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.500024402805081e-06,
+      "loss": 0.5447,
+      "step": 307
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.496769858130749e-06,
+      "loss": 0.5609,
+      "step": 308
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.493505940176574e-06,
+      "loss": 0.535,
+      "step": 309
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.49023266426411e-06,
+      "loss": 0.5494,
+      "step": 310
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.486950045758838e-06,
+      "loss": 0.5495,
+      "step": 311
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.483658100070095e-06,
+      "loss": 0.5522,
+      "step": 312
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.480356842651006e-06,
+      "loss": 0.556,
+      "step": 313
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.477046288998401e-06,
+      "loss": 0.5581,
+      "step": 314
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.473726454652755e-06,
+      "loss": 0.545,
+      "step": 315
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.470397355198103e-06,
+      "loss": 0.5489,
+      "step": 316
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.467059006261976e-06,
+      "loss": 0.5504,
+      "step": 317
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.463711423515324e-06,
+      "loss": 0.5629,
+      "step": 318
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.46035462267244e-06,
+      "loss": 0.5424,
+      "step": 319
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.45698861949089e-06,
+      "loss": 0.5585,
+      "step": 320
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.45361342977144e-06,
+      "loss": 0.5507,
+      "step": 321
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.450229069357978e-06,
+      "loss": 0.5627,
+      "step": 322
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.446835554137438e-06,
+      "loss": 0.5503,
+      "step": 323
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.443432900039737e-06,
+      "loss": 0.538,
+      "step": 324
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.440021123037683e-06,
+      "loss": 0.5408,
+      "step": 325
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.436600239146913e-06,
+      "loss": 0.5494,
+      "step": 326
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.433170264425814e-06,
+      "loss": 0.5485,
+      "step": 327
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.429731214975448e-06,
+      "loss": 0.5497,
+      "step": 328
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.426283106939474e-06,
+      "loss": 0.5406,
+      "step": 329
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.422825956504073e-06,
+      "loss": 0.5421,
+      "step": 330
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.4193597798978756e-06,
+      "loss": 0.5504,
+      "step": 331
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.415884593391882e-06,
+      "loss": 0.5368,
+      "step": 332
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.412400413299389e-06,
+      "loss": 0.5456,
+      "step": 333
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.408907255975907e-06,
+      "loss": 0.5644,
+      "step": 334
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.4054051378190915e-06,
+      "loss": 0.5358,
+      "step": 335
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.40189407526866e-06,
+      "loss": 0.5566,
+      "step": 336
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.3983740848063175e-06,
+      "loss": 0.5332,
+      "step": 337
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.394845182955678e-06,
+      "loss": 0.5389,
+      "step": 338
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.39130738628219e-06,
+      "loss": 0.5406,
+      "step": 339
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.387760711393052e-06,
+      "loss": 0.537,
+      "step": 340
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.3842051749371435e-06,
+      "loss": 0.5301,
+      "step": 341
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.380640793604938e-06,
+      "loss": 0.547,
+      "step": 342
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.377067584128432e-06,
+      "loss": 0.555,
+      "step": 343
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.3734855632810624e-06,
+      "loss": 0.5568,
+      "step": 344
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.369894747877627e-06,
+      "loss": 0.5529,
+      "step": 345
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.3662951547742085e-06,
+      "loss": 0.5668,
+      "step": 346
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.3626868008680946e-06,
+      "loss": 0.5351,
+      "step": 347
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.359069703097697e-06,
+      "loss": 0.5553,
+      "step": 348
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.355443878442474e-06,
+      "loss": 0.5262,
+      "step": 349
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.3518093439228484e-06,
+      "loss": 0.5428,
+      "step": 350
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.348166116600131e-06,
+      "loss": 0.5474,
+      "step": 351
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.3445142135764375e-06,
+      "loss": 0.5279,
+      "step": 352
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.340853651994608e-06,
+      "loss": 0.5614,
+      "step": 353
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.337184449038131e-06,
+      "loss": 0.5533,
+      "step": 354
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.333506621931056e-06,
+      "loss": 0.539,
+      "step": 355
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.32982018793792e-06,
+      "loss": 0.5574,
+      "step": 356
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.326125164363658e-06,
+      "loss": 0.5527,
+      "step": 357
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.322421568553529e-06,
+      "loss": 0.5425,
+      "step": 358
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.318709417893033e-06,
+      "loss": 0.551,
+      "step": 359
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.3149887298078275e-06,
+      "loss": 0.5664,
+      "step": 360
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.311259521763645e-06,
+      "loss": 0.5376,
+      "step": 361
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.307521811266214e-06,
+      "loss": 0.5583,
+      "step": 362
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.303775615861175e-06,
+      "loss": 0.5444,
+      "step": 363
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.300020953134e-06,
+      "loss": 0.5359,
+      "step": 364
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.296257840709906e-06,
+      "loss": 0.5338,
+      "step": 365
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.5293887853622437,
+      "eval_runtime": 71.2151,
+      "eval_samples_per_second": 71.108,
+      "eval_steps_per_second": 5.926,
+      "step": 365
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.292486296253776e-06,
+      "loss": 0.541,
+      "step": 366
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.288706337470076e-06,
+      "loss": 0.5501,
+      "step": 367
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2849179821027684e-06,
+      "loss": 0.5359,
+      "step": 368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.281121247935231e-06,
+      "loss": 0.5362,
+      "step": 369
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.277316152790177e-06,
+      "loss": 0.5382,
+      "step": 370
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.273502714529564e-06,
+      "loss": 0.5298,
+      "step": 371
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.269680951054517e-06,
+      "loss": 0.5776,
+      "step": 372
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.265850880305238e-06,
+      "loss": 0.5497,
+      "step": 373
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.262012520260928e-06,
+      "loss": 0.5263,
+      "step": 374
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2581658889397e-06,
+      "loss": 0.528,
+      "step": 375
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.254311004398492e-06,
+      "loss": 0.5341,
+      "step": 376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.250447884732986e-06,
+      "loss": 0.5362,
+      "step": 377
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2465765480775215e-06,
+      "loss": 0.5306,
+      "step": 378
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.242697012605008e-06,
+      "loss": 0.5191,
+      "step": 379
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.238809296526847e-06,
+      "loss": 0.5287,
+      "step": 380
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.234913418092837e-06,
+      "loss": 0.5421,
+      "step": 381
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.231009395591093e-06,
+      "loss": 0.5201,
+      "step": 382
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.227097247347962e-06,
+      "loss": 0.5283,
+      "step": 383
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.223176991727935e-06,
+      "loss": 0.5288,
+      "step": 384
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.219248647133559e-06,
+      "loss": 0.5204,
+      "step": 385
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.215312232005353e-06,
+      "loss": 0.5182,
+      "step": 386
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.211367764821722e-06,
+      "loss": 0.5431,
+      "step": 387
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.207415264098868e-06,
+      "loss": 0.5424,
+      "step": 388
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.203454748390705e-06,
+      "loss": 0.5213,
+      "step": 389
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.19948623628877e-06,
+      "loss": 0.5278,
+      "step": 390
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.195509746422138e-06,
+      "loss": 0.5443,
+      "step": 391
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.191525297457332e-06,
+      "loss": 0.5378,
+      "step": 392
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.187532908098239e-06,
+      "loss": 0.5287,
+      "step": 393
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.183532597086018e-06,
+      "loss": 0.5301,
+      "step": 394
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.179524383199016e-06,
+      "loss": 0.5317,
+      "step": 395
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.175508285252674e-06,
+      "loss": 0.534,
+      "step": 396
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.171484322099446e-06,
+      "loss": 0.5177,
+      "step": 397
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.167452512628707e-06,
+      "loss": 0.5344,
+      "step": 398
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.163412875766661e-06,
+      "loss": 0.5329,
+      "step": 399
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.159365430476262e-06,
+      "loss": 0.5297,
+      "step": 400
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.15531019575711e-06,
+      "loss": 0.5154,
+      "step": 401
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.151247190645378e-06,
+      "loss": 0.5344,
+      "step": 402
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.1471764342137095e-06,
+      "loss": 0.5193,
+      "step": 403
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.143097945571137e-06,
+      "loss": 0.5023,
+      "step": 404
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.139011743862991e-06,
+      "loss": 0.5235,
+      "step": 405
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.1349178482708044e-06,
+      "loss": 0.5086,
+      "step": 406
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.130816278012231e-06,
+      "loss": 0.5083,
+      "step": 407
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.126707052340949e-06,
+      "loss": 0.5234,
+      "step": 408
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.122590190546576e-06,
+      "loss": 0.5257,
+      "step": 409
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.11846571195457e-06,
+      "loss": 0.5202,
+      "step": 410
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.114333635926147e-06,
+      "loss": 0.5215,
+      "step": 411
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.1101939818581885e-06,
+      "loss": 0.5431,
+      "step": 412
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.106046769183146e-06,
+      "loss": 0.5299,
+      "step": 413
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.1018920173689555e-06,
+      "loss": 0.5259,
+      "step": 414
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.0977297459189405e-06,
+      "loss": 0.5023,
+      "step": 415
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.093559974371725e-06,
+      "loss": 0.5258,
+      "step": 416
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.08938272230114e-06,
+      "loss": 0.5287,
+      "step": 417
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.085198009316132e-06,
+      "loss": 0.5117,
+      "step": 418
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.0810058550606706e-06,
+      "loss": 0.5191,
+      "step": 419
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.076806279213656e-06,
+      "loss": 0.5234,
+      "step": 420
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.072599301488826e-06,
+      "loss": 0.5298,
+      "step": 421
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.068384941634665e-06,
+      "loss": 0.5183,
+      "step": 422
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.064163219434315e-06,
+      "loss": 0.5244,
+      "step": 423
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.059934154705472e-06,
+      "loss": 0.5057,
+      "step": 424
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.055697767300302e-06,
+      "loss": 0.5116,
+      "step": 425
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.051454077105347e-06,
+      "loss": 0.5208,
+      "step": 426
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.047203104041427e-06,
+      "loss": 0.5038,
+      "step": 427
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.042944868063552e-06,
+      "loss": 0.4986,
+      "step": 428
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.038679389160823e-06,
+      "loss": 0.5153,
+      "step": 429
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.034406687356344e-06,
+      "loss": 0.5256,
+      "step": 430
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.030126782707123e-06,
+      "loss": 0.5164,
+      "step": 431
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.02583969530398e-06,
+      "loss": 0.5315,
+      "step": 432
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.021545445271453e-06,
+      "loss": 0.5049,
+      "step": 433
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.0172440527677e-06,
+      "loss": 0.5222,
+      "step": 434
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.012935537984414e-06,
+      "loss": 0.5012,
+      "step": 435
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.0086199211467135e-06,
+      "loss": 0.5325,
+      "step": 436
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.004297222513062e-06,
+      "loss": 0.5121,
+      "step": 437
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 3.9999674623751625e-06,
+      "loss": 0.5227,
+      "step": 438
+    },
+    {
+      "epoch": 1.18,
+      "eval_loss": 0.526366651058197,
+      "eval_runtime": 70.7579,
+      "eval_samples_per_second": 71.568,
+      "eval_steps_per_second": 5.964,
+      "step": 438
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 3.995630661057869e-06,
+      "loss": 0.5069,
+      "step": 439
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 3.991286838919086e-06,
+      "loss": 0.5114,
+      "step": 440
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 3.986936016349677e-06,
+      "loss": 0.5242,
+      "step": 441
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 3.982578213773368e-06,
+      "loss": 0.5121,
+      "step": 442
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 3.978213451646646e-06,
+      "loss": 0.5265,
+      "step": 443
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 3.9738417504586734e-06,
+      "loss": 0.5144,
+      "step": 444
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3.969463130731183e-06,
+      "loss": 0.5083,
+      "step": 445
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3.965077613018385e-06,
+      "loss": 0.5113,
+      "step": 446
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3.960685217906871e-06,
+      "loss": 0.4961,
+      "step": 447
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3.956285966015516e-06,
+      "loss": 0.5018,
+      "step": 448
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 3.951879877995381e-06,
+      "loss": 0.5033,
+      "step": 449
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 3.947466974529622e-06,
+      "loss": 0.5151,
+      "step": 450
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 3.943047276333383e-06,
+      "loss": 0.5079,
+      "step": 451
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 3.9386208041537055e-06,
+      "loss": 0.4961,
+      "step": 452
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 3.93418757876943e-06,
+      "loss": 0.5098,
+      "step": 453
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 3.9297476209911e-06,
+      "loss": 0.5238,
+      "step": 454
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 3.925300951660859e-06,
+      "loss": 0.5197,
+      "step": 455
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 3.920847591652357e-06,
+      "loss": 0.5014,
+      "step": 456
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 3.916387561870653e-06,
+      "loss": 0.4977,
+      "step": 457
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 3.911920883252114e-06,
+      "loss": 0.5065,
+      "step": 458
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 3.9074475767643175e-06,
+      "loss": 0.4941,
+      "step": 459
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 3.9029676634059565e-06,
+      "loss": 0.5123,
+      "step": 460
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 3.898481164206734e-06,
+      "loss": 0.4956,
+      "step": 461
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 3.893988100227273e-06,
+      "loss": 0.5138,
+      "step": 462
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 3.8894884925590095e-06,
+      "loss": 0.5094,
+      "step": 463
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 3.884982362324098e-06,
+      "loss": 0.4993,
+      "step": 464
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 3.880469730675311e-06,
+      "loss": 0.5128,
+      "step": 465
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 3.875950618795943e-06,
+      "loss": 0.5091,
+      "step": 466
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.871425047899704e-06,
+      "loss": 0.5187,
+      "step": 467
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.866893039230626e-06,
+      "loss": 0.4973,
+      "step": 468
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.862354614062961e-06,
+      "loss": 0.5104,
+      "step": 469
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.857809793701082e-06,
+      "loss": 0.494,
+      "step": 470
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 3.853258599479383e-06,
+      "loss": 0.4944,
+      "step": 471
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 3.848701052762176e-06,
+      "loss": 0.5071,
+      "step": 472
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 3.844137174943594e-06,
+      "loss": 0.5059,
+      "step": 473
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 3.839566987447492e-06,
+      "loss": 0.4971,
+      "step": 474
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 3.834990511727341e-06,
+      "loss": 0.5109,
+      "step": 475
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 3.8304077692661305e-06,
+      "loss": 0.5275,
+      "step": 476
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 3.82581878157627e-06,
+      "loss": 0.5171,
+      "step": 477
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 3.821223570199483e-06,
+      "loss": 0.5085,
+      "step": 478
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 3.816622156706709e-06,
+      "loss": 0.4948,
+      "step": 479
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 3.812014562698002e-06,
+      "loss": 0.4959,
+      "step": 480
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 3.8074008098024282e-06,
+      "loss": 0.5087,
+      "step": 481
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 3.802780919677966e-06,
+      "loss": 0.4993,
+      "step": 482
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 3.7981549140114015e-06,
+      "loss": 0.5091,
+      "step": 483
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 3.7935228145182314e-06,
+      "loss": 0.4971,
+      "step": 484
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 3.788884642942555e-06,
+      "loss": 0.4976,
+      "step": 485
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 3.784240421056976e-06,
+      "loss": 0.4976,
+      "step": 486
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 3.7795901706625025e-06,
+      "loss": 0.496,
+      "step": 487
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 3.7749339135884373e-06,
+      "loss": 0.5072,
+      "step": 488
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 3.7702716716922825e-06,
+      "loss": 0.5074,
+      "step": 489
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 3.765603466859635e-06,
+      "loss": 0.5106,
+      "step": 490
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 3.760929321004082e-06,
+      "loss": 0.5052,
+      "step": 491
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 3.756249256067097e-06,
+      "loss": 0.4925,
+      "step": 492
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.751563294017944e-06,
+      "loss": 0.4922,
+      "step": 493
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.7468714568535643e-06,
+      "loss": 0.5246,
+      "step": 494
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.7421737665984807e-06,
+      "loss": 0.5016,
+      "step": 495
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 3.737470245304692e-06,
+      "loss": 0.5108,
+      "step": 496
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 3.7327609150515685e-06,
+      "loss": 0.5088,
+      "step": 497
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 3.728045797945749e-06,
+      "loss": 0.502,
+      "step": 498
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 3.723324916121038e-06,
+      "loss": 0.5099,
+      "step": 499
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 3.7185982917382986e-06,
+      "loss": 0.4917,
+      "step": 500
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 3.7138659469853535e-06,
+      "loss": 0.5076,
+      "step": 501
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 3.7091279040768752e-06,
+      "loss": 0.4993,
+      "step": 502
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 3.7043841852542884e-06,
+      "loss": 0.5042,
+      "step": 503
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 3.6996348127856584e-06,
+      "loss": 0.508,
+      "step": 504
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 3.6948798089655913e-06,
+      "loss": 0.5046,
+      "step": 505
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 3.6901191961151285e-06,
+      "loss": 0.5037,
+      "step": 506
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 3.6853529965816394e-06,
+      "loss": 0.488,
+      "step": 507
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 3.6805812327387212e-06,
+      "loss": 0.5089,
+      "step": 508
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 3.67580392698609e-06,
+      "loss": 0.4839,
+      "step": 509
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 3.671021101749476e-06,
+      "loss": 0.5099,
+      "step": 510
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 3.6662327794805203e-06,
+      "loss": 0.507,
+      "step": 511
+    },
+    {
+      "epoch": 1.38,
+      "eval_loss": 0.5247712135314941,
+      "eval_runtime": 70.9408,
+      "eval_samples_per_second": 71.383,
+      "eval_steps_per_second": 5.949,
+      "step": 511
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 3.6614389826566678e-06,
+      "loss": 0.4947,
+      "step": 512
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 3.656639733781062e-06,
+      "loss": 0.5022,
+      "step": 513
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 3.6518350553824406e-06,
+      "loss": 0.5206,
+      "step": 514
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 3.6470249700150273e-06,
+      "loss": 0.5084,
+      "step": 515
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 3.642209500258428e-06,
+      "loss": 0.4943,
+      "step": 516
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 3.637388668717523e-06,
+      "loss": 0.5118,
+      "step": 517
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 3.632562498022365e-06,
+      "loss": 0.5063,
+      "step": 518
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 3.6277310108280662e-06,
+      "loss": 0.4828,
+      "step": 519
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 3.6228942298146985e-06,
+      "loss": 0.498,
+      "step": 520
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 3.6180521776871815e-06,
+      "loss": 0.5064,
+      "step": 521
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 3.613204877175181e-06,
+      "loss": 0.5094,
+      "step": 522
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 3.608352351032999e-06,
+      "loss": 0.4858,
+      "step": 523
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 3.6034946220394667e-06,
+      "loss": 0.5,
+      "step": 524
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 3.598631712997841e-06,
+      "loss": 0.5078,
+      "step": 525
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 3.593763646735693e-06,
+      "loss": 0.5025,
+      "step": 526
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 3.5888904461048037e-06,
+      "loss": 0.5095,
+      "step": 527
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 3.5840121339810558e-06,
+      "loss": 0.5085,
+      "step": 528
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 3.579128733264326e-06,
+      "loss": 0.5069,
+      "step": 529
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 3.5742402668783797e-06,
+      "loss": 0.4872,
+      "step": 530
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 3.5693467577707597e-06,
+      "loss": 0.5029,
+      "step": 531
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 3.564448228912682e-06,
+      "loss": 0.4977,
+      "step": 532
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.559544703298925e-06,
+      "loss": 0.5014,
+      "step": 533
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.5546362039477244e-06,
+      "loss": 0.4964,
+      "step": 534
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.549722753900662e-06,
+      "loss": 0.5073,
+      "step": 535
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.54480437622256e-06,
+      "loss": 0.4972,
+      "step": 536
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 3.539881094001375e-06,
+      "loss": 0.508,
+      "step": 537
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 3.534952930348081e-06,
+      "loss": 0.5,
+      "step": 538
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 3.530019908396571e-06,
+      "loss": 0.5208,
+      "step": 539
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 3.5250820513035403e-06,
+      "loss": 0.4958,
+      "step": 540
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 3.520139382248386e-06,
+      "loss": 0.4894,
+      "step": 541
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 3.515191924433089e-06,
+      "loss": 0.4965,
+      "step": 542
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 3.5102397010821133e-06,
+      "loss": 0.5107,
+      "step": 543
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 3.5052827354422913e-06,
+      "loss": 0.4939,
+      "step": 544
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 3.500321050782717e-06,
+      "loss": 0.4948,
+      "step": 545
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 3.4953546703946366e-06,
+      "loss": 0.5096,
+      "step": 546
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 3.4903836175913402e-06,
+      "loss": 0.5124,
+      "step": 547
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 3.4854079157080502e-06,
+      "loss": 0.5176,
+      "step": 548
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 3.480427588101812e-06,
+      "loss": 0.5004,
+      "step": 549
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 3.4754426581513866e-06,
+      "loss": 0.5058,
+      "step": 550
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 3.470453149257139e-06,
+      "loss": 0.4973,
+      "step": 551
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 3.4654590848409296e-06,
+      "loss": 0.5085,
+      "step": 552
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 3.460460488346002e-06,
+      "loss": 0.5075,
+      "step": 553
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 3.4554573832368743e-06,
+      "loss": 0.5046,
+      "step": 554
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 3.45044979299923e-06,
+      "loss": 0.4852,
+      "step": 555
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 3.445437741139807e-06,
+      "loss": 0.4921,
+      "step": 556
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 3.440421251186286e-06,
+      "loss": 0.4943,
+      "step": 557
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 3.435400346687182e-06,
+      "loss": 0.5066,
+      "step": 558
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 3.4303750512117324e-06,
+      "loss": 0.5039,
+      "step": 559
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 3.425345388349787e-06,
+      "loss": 0.4971,
+      "step": 560
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 3.4203113817116955e-06,
+      "loss": 0.5009,
+      "step": 561
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 3.4152730549282007e-06,
+      "loss": 0.4939,
+      "step": 562
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 3.4102304316503256e-06,
+      "loss": 0.504,
+      "step": 563
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 3.4051835355492578e-06,
+      "loss": 0.5192,
+      "step": 564
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 3.4001323903162476e-06,
+      "loss": 0.5001,
+      "step": 565
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 3.395077019662487e-06,
+      "loss": 0.5132,
+      "step": 566
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 3.3900174473190066e-06,
+      "loss": 0.5215,
+      "step": 567
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 3.3849536970365586e-06,
+      "loss": 0.5056,
+      "step": 568
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 3.3798857925855083e-06,
+      "loss": 0.5085,
+      "step": 569
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 3.3748137577557216e-06,
+      "loss": 0.4918,
+      "step": 570
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 3.369737616356452e-06,
+      "loss": 0.5194,
+      "step": 571
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 3.364657392216233e-06,
+      "loss": 0.4927,
+      "step": 572
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 3.35957310918276e-06,
+      "loss": 0.5003,
+      "step": 573
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 3.354484791122784e-06,
+      "loss": 0.51,
+      "step": 574
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 3.3493924619219964e-06,
+      "loss": 0.4934,
+      "step": 575
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 3.344296145484918e-06,
+      "loss": 0.4863,
+      "step": 576
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.339195865734788e-06,
+      "loss": 0.4977,
+      "step": 577
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.3340916466134458e-06,
+      "loss": 0.5064,
+      "step": 578
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.3289835120812297e-06,
+      "loss": 0.4869,
+      "step": 579
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.3238714861168513e-06,
+      "loss": 0.4952,
+      "step": 580
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 3.318755592717294e-06,
+      "loss": 0.4991,
+      "step": 581
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 3.313635855897693e-06,
+      "loss": 0.4963,
+      "step": 582
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 3.3085122996912284e-06,
+      "loss": 0.5106,
+      "step": 583
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 3.3033849481490067e-06,
+      "loss": 0.4995,
+      "step": 584
+    },
+    {
+      "epoch": 1.58,
+      "eval_loss": 0.5225731134414673,
+      "eval_runtime": 70.7593,
+      "eval_samples_per_second": 71.567,
+      "eval_steps_per_second": 5.964,
+      "step": 584
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 3.29825382533995e-06,
+      "loss": 0.5121,
+      "step": 585
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 3.293118955350687e-06,
+      "loss": 0.4968,
+      "step": 586
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 3.287980362285433e-06,
+      "loss": 0.4936,
+      "step": 587
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 3.2828380702658823e-06,
+      "loss": 0.4992,
+      "step": 588
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 3.2776921034310917e-06,
+      "loss": 0.5006,
+      "step": 589
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 3.272542485937369e-06,
+      "loss": 0.4895,
+      "step": 590
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 3.2673892419581595e-06,
+      "loss": 0.4973,
+      "step": 591
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 3.262232395683931e-06,
+      "loss": 0.4921,
+      "step": 592
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 3.257071971322062e-06,
+      "loss": 0.5214,
+      "step": 593
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 3.251907993096728e-06,
+      "loss": 0.5045,
+      "step": 594
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 3.2467404852487846e-06,
+      "loss": 0.5033,
+      "step": 595
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 3.241569472035658e-06,
+      "loss": 0.494,
+      "step": 596
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 3.23639497773123e-06,
+      "loss": 0.5056,
+      "step": 597
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 3.2312170266257225e-06,
+      "loss": 0.4868,
+      "step": 598
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.2260356430255845e-06,
+      "loss": 0.5029,
+      "step": 599
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.2208508512533777e-06,
+      "loss": 0.4825,
+      "step": 600
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.2156626756476624e-06,
+      "loss": 0.4996,
+      "step": 601
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.210471140562883e-06,
+      "loss": 0.4989,
+      "step": 602
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.2052762703692574e-06,
+      "loss": 0.4891,
+      "step": 603
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.200078089452654e-06,
+      "loss": 0.507,
+      "step": 604
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.1948766222144863e-06,
+      "loss": 0.4998,
+      "step": 605
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.189671893071592e-06,
+      "loss": 0.4925,
+      "step": 606
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.184463926456124e-06,
+      "loss": 0.4935,
+      "step": 607
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.179252746815429e-06,
+      "loss": 0.4964,
+      "step": 608
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.174038378611939e-06,
+      "loss": 0.504,
+      "step": 609
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.168820846323053e-06,
+      "loss": 0.5163,
+      "step": 610
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.1636001744410218e-06,
+      "loss": 0.4929,
+      "step": 611
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.1583763874728364e-06,
+      "loss": 0.5051,
+      "step": 612
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.153149509940108e-06,
+      "loss": 0.4999,
+      "step": 613
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.1479195663789596e-06,
+      "loss": 0.5059,
+      "step": 614
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.142686581339902e-06,
+      "loss": 0.4932,
+      "step": 615
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.137450579387727e-06,
+      "loss": 0.5076,
+      "step": 616
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.1322115851013867e-06,
+      "loss": 0.4981,
+      "step": 617
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.12696962307388e-06,
+      "loss": 0.4902,
+      "step": 618
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.121724717912138e-06,
+      "loss": 0.5056,
+      "step": 619
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.1164768942369058e-06,
+      "loss": 0.4962,
+      "step": 620
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.1112261766826303e-06,
+      "loss": 0.5111,
+      "step": 621
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.105972589897342e-06,
+      "loss": 0.492,
+      "step": 622
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.1007161585425417e-06,
+      "loss": 0.4836,
+      "step": 623
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.0954569072930813e-06,
+      "loss": 0.4998,
+      "step": 624
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.0901948608370503e-06,
+      "loss": 0.5057,
+      "step": 625
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.0849300438756607e-06,
+      "loss": 0.4881,
+      "step": 626
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.0796624811231286e-06,
+      "loss": 0.4896,
+      "step": 627
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.07439219730656e-06,
+      "loss": 0.493,
+      "step": 628
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.0691192171658335e-06,
+      "loss": 0.5068,
+      "step": 629
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.063843565453486e-06,
+      "loss": 0.5064,
+      "step": 630
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.0585652669345944e-06,
+      "loss": 0.4997,
+      "step": 631
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 3.053284346386662e-06,
+      "loss": 0.486,
+      "step": 632
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 3.0480008285994976e-06,
+      "loss": 0.4983,
+      "step": 633
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 3.0427147383751043e-06,
+      "loss": 0.5115,
+      "step": 634
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 3.0374261005275606e-06,
+      "loss": 0.4859,
+      "step": 635
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 3.0321349398829026e-06,
+      "loss": 0.4691,
+      "step": 636
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 3.02684128127901e-06,
+      "loss": 0.4983,
+      "step": 637
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 3.021545149565489e-06,
+      "loss": 0.4794,
+      "step": 638
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 3.0162465696035536e-06,
+      "loss": 0.5014,
+      "step": 639
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 3.0109455662659126e-06,
+      "loss": 0.5093,
+      "step": 640
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 3.0056421644366486e-06,
+      "loss": 0.4815,
+      "step": 641
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 3.0003363890111042e-06,
+      "loss": 0.4993,
+      "step": 642
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.9950282648957646e-06,
+      "loss": 0.5145,
+      "step": 643
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.989717817008139e-06,
+      "loss": 0.4909,
+      "step": 644
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.984405070276646e-06,
+      "loss": 0.4992,
+      "step": 645
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.979090049640495e-06,
+      "loss": 0.4954,
+      "step": 646
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.9737727800495706e-06,
+      "loss": 0.4714,
+      "step": 647
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.9684532864643123e-06,
+      "loss": 0.4903,
+      "step": 648
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2.963131593855601e-06,
+      "loss": 0.4932,
+      "step": 649
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.9578077272046407e-06,
+      "loss": 0.4887,
+      "step": 650
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.95248171150284e-06,
+      "loss": 0.5008,
+      "step": 651
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.947153571751696e-06,
+      "loss": 0.4912,
+      "step": 652
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.9418233329626765e-06,
+      "loss": 0.4917,
+      "step": 653
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.936491020157103e-06,
+      "loss": 0.4844,
+      "step": 654
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.931156658366032e-06,
+      "loss": 0.4927,
+      "step": 655
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.925820272630138e-06,
+      "loss": 0.4881,
+      "step": 656
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.9204818879995995e-06,
+      "loss": 0.4988,
+      "step": 657
+    },
+    {
+      "epoch": 1.78,
+      "eval_loss": 0.521266520023346,
+      "eval_runtime": 70.745,
+      "eval_samples_per_second": 71.581,
+      "eval_steps_per_second": 5.965,
+      "step": 657
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.9151415295339734e-06,
+      "loss": 0.4906,
+      "step": 658
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.909799222302087e-06,
+      "loss": 0.5025,
+      "step": 659
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.9044549913819125e-06,
+      "loss": 0.4962,
+      "step": 660
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.8991088618604522e-06,
+      "loss": 0.4934,
+      "step": 661
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.8937608588336234e-06,
+      "loss": 0.4848,
+      "step": 662
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.8884110074061357e-06,
+      "loss": 0.4976,
+      "step": 663
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.883059332691377e-06,
+      "loss": 0.491,
+      "step": 664
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.877705859811292e-06,
+      "loss": 0.5074,
+      "step": 665
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.8723506138962694e-06,
+      "loss": 0.4974,
+      "step": 666
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.866993620085018e-06,
+      "loss": 0.5048,
+      "step": 667
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 2.8616349035244535e-06,
+      "loss": 0.5067,
+      "step": 668
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 2.856274489369577e-06,
+      "loss": 0.4891,
+      "step": 669
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 2.850912402783361e-06,
+      "loss": 0.4988,
+      "step": 670
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 2.8455486689366262e-06,
+      "loss": 0.4999,
+      "step": 671
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 2.8401833130079277e-06,
+      "loss": 0.5119,
+      "step": 672
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 2.834816360183434e-06,
+      "loss": 0.504,
+      "step": 673
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 2.8294478356568096e-06,
+      "loss": 0.4941,
+      "step": 674
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 2.8240777646290973e-06,
+      "loss": 0.4963,
+      "step": 675
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 2.8187061723086e-06,
+      "loss": 0.5062,
+      "step": 676
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 2.813333083910761e-06,
+      "loss": 0.506,
+      "step": 677
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 2.8079585246580476e-06,
+      "loss": 0.4828,
+      "step": 678
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 2.80258251977983e-06,
+      "loss": 0.4919,
+      "step": 679
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 2.7972050945122666e-06,
+      "loss": 0.5074,
+      "step": 680
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 2.791826274098181e-06,
+      "loss": 0.4908,
+      "step": 681
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 2.7864460837869483e-06,
+      "loss": 0.4934,
+      "step": 682
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 2.7810645488343737e-06,
+      "loss": 0.499,
+      "step": 683
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 2.775681694502572e-06,
+      "loss": 0.4893,
+      "step": 684
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 2.7702975460598545e-06,
+      "loss": 0.5136,
+      "step": 685
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 2.7649121287806064e-06,
+      "loss": 0.5038,
+      "step": 686
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 2.759525467945168e-06,
+      "loss": 0.4965,
+      "step": 687
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 2.754137588839718e-06,
+      "loss": 0.4993,
+      "step": 688
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 2.7487485167561546e-06,
+      "loss": 0.4945,
+      "step": 689
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 2.7433582769919752e-06,
+      "loss": 0.5069,
+      "step": 690
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 2.7379668948501575e-06,
+      "loss": 0.4958,
+      "step": 691
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 2.7325743956390456e-06,
+      "loss": 0.5077,
+      "step": 692
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 2.7271808046722224e-06,
+      "loss": 0.4892,
+      "step": 693
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 2.7217861472684004e-06,
+      "loss": 0.5081,
+      "step": 694
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 2.716390448751294e-06,
+      "loss": 0.497,
+      "step": 695
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 2.7109937344495076e-06,
+      "loss": 0.5157,
+      "step": 696
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 2.7055960296964134e-06,
+      "loss": 0.4954,
+      "step": 697
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 2.700197359830032e-06,
+      "loss": 0.4886,
+      "step": 698
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 2.6947977501929153e-06,
+      "loss": 0.4987,
+      "step": 699
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 2.6893972261320265e-06,
+      "loss": 0.4989,
+      "step": 700
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.6839958129986215e-06,
+      "loss": 0.4992,
+      "step": 701
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.678593536148129e-06,
+      "loss": 0.4941,
+      "step": 702
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.673190420940034e-06,
+      "loss": 0.4832,
+      "step": 703
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.667786492737755e-06,
+      "loss": 0.5095,
+      "step": 704
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.6623817769085268e-06,
+      "loss": 0.4958,
+      "step": 705
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.6569762988232838e-06,
+      "loss": 0.481,
+      "step": 706
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.6515700838565355e-06,
+      "loss": 0.5198,
+      "step": 707
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.6461631573862528e-06,
+      "loss": 0.4827,
+      "step": 708
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.6407555447937455e-06,
+      "loss": 0.5017,
+      "step": 709
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.6353472714635443e-06,
+      "loss": 0.5014,
+      "step": 710
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.629938362783282e-06,
+      "loss": 0.4826,
+      "step": 711
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.6245288441435734e-06,
+      "loss": 0.4939,
+      "step": 712
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.619118740937898e-06,
+      "loss": 0.489,
+      "step": 713
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.613708078562476e-06,
+      "loss": 0.4769,
+      "step": 714
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.6082968824161558e-06,
+      "loss": 0.4963,
+      "step": 715
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 2.60288517790029e-06,
+      "loss": 0.499,
+      "step": 716
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 2.5974729904186187e-06,
+      "loss": 0.505,
+      "step": 717
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 2.5920603453771475e-06,
+      "loss": 0.5134,
+      "step": 718
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 2.586647268184031e-06,
+      "loss": 0.5112,
+      "step": 719
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 2.5812337842494517e-06,
+      "loss": 0.496,
+      "step": 720
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 2.575819918985502e-06,
+      "loss": 0.5056,
+      "step": 721
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 2.5704056978060643e-06,
+      "loss": 0.4804,
+      "step": 722
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 2.5649911461266923e-06,
+      "loss": 0.4924,
+      "step": 723
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 2.55957628936449e-06,
+      "loss": 0.4994,
+      "step": 724
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 2.554161152937994e-06,
+      "loss": 0.4814,
+      "step": 725
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 2.548745762267053e-06,
+      "loss": 0.498,
+      "step": 726
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.543330142772711e-06,
+      "loss": 0.5082,
+      "step": 727
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.537914319877084e-06,
+      "loss": 0.5103,
+      "step": 728
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.532498319003245e-06,
+      "loss": 0.4912,
+      "step": 729
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.5270821655750997e-06,
+      "loss": 0.4983,
+      "step": 730
+    },
+    {
+      "epoch": 1.97,
+      "eval_loss": 0.5186672210693359,
+      "eval_runtime": 70.712,
+      "eval_samples_per_second": 71.614,
+      "eval_steps_per_second": 5.968,
+      "step": 730
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.5216658850172737e-06,
+      "loss": 0.5086,
+      "step": 731
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.5162495027549862e-06,
+      "loss": 0.5001,
+      "step": 732
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.510833044213934e-06,
+      "loss": 0.516,
+      "step": 733
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 2.505416534820174e-06,
+      "loss": 0.4995,
+      "step": 734
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 2.5e-06,
+      "loss": 0.4981,
+      "step": 735
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 2.4945834651798266e-06,
+      "loss": 0.5008,
+      "step": 736
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 2.4891669557860664e-06,
+      "loss": 0.4864,
+      "step": 737
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.483750497245015e-06,
+      "loss": 0.5013,
+      "step": 738
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.478334114982727e-06,
+      "loss": 0.49,
+      "step": 739
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4729178344249007e-06,
+      "loss": 0.4878,
+      "step": 740
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.467501680996756e-06,
+      "loss": 0.4954,
+      "step": 741
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4620856801229166e-06,
+      "loss": 0.4922,
+      "step": 742
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4566698572272902e-06,
+      "loss": 0.4923,
+      "step": 743
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.451254237732948e-06,
+      "loss": 0.4844,
+      "step": 744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4458388470620066e-06,
+      "loss": 0.5174,
+      "step": 745
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4404237106355104e-06,
+      "loss": 0.5085,
+      "step": 746
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4350088538733086e-06,
+      "loss": 0.4873,
+      "step": 747
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.429594302193936e-06,
+      "loss": 0.4799,
+      "step": 748
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4241800810144987e-06,
+      "loss": 0.493,
+      "step": 749
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.418766215750549e-06,
+      "loss": 0.4899,
+      "step": 750
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.41335273181597e-06,
+      "loss": 0.4776,
+      "step": 751
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4079396546228537e-06,
+      "loss": 0.4784,
+      "step": 752
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.402527009581382e-06,
+      "loss": 0.4842,
+      "step": 753
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3971148220997103e-06,
+      "loss": 0.4924,
+      "step": 754
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3917031175838447e-06,
+      "loss": 0.4846,
+      "step": 755
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 2.3862919214375244e-06,
+      "loss": 0.4832,
+      "step": 756
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 2.380881259062104e-06,
+      "loss": 0.478,
+      "step": 757
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 2.3754711558564274e-06,
+      "loss": 0.4786,
+      "step": 758
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 2.370061637216719e-06,
+      "loss": 0.4766,
+      "step": 759
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 2.3646527285364565e-06,
+      "loss": 0.4826,
+      "step": 760
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 2.3592444552062557e-06,
+      "loss": 0.4917,
+      "step": 761
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 2.353836842613748e-06,
+      "loss": 0.494,
+      "step": 762
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 2.3484299161434653e-06,
+      "loss": 0.4787,
+      "step": 763
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 2.3430237011767166e-06,
+      "loss": 0.4948,
+      "step": 764
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 2.3376182230914728e-06,
+      "loss": 0.4891,
+      "step": 765
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 2.3322135072622463e-06,
+      "loss": 0.4881,
+      "step": 766
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 2.326809579059967e-06,
+      "loss": 0.479,
+      "step": 767
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 2.3214064638518714e-06,
+      "loss": 0.4843,
+      "step": 768
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 2.3160041870013794e-06,
+      "loss": 0.5038,
+      "step": 769
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 2.3106027738679743e-06,
+      "loss": 0.4713,
+      "step": 770
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 2.305202249807086e-06,
+      "loss": 0.4877,
+      "step": 771
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 2.2998026401699693e-06,
+      "loss": 0.4998,
+      "step": 772
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 2.2944039703035874e-06,
+      "loss": 0.4752,
+      "step": 773
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 2.289006265550493e-06,
+      "loss": 0.4736,
+      "step": 774
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 2.2836095512487063e-06,
+      "loss": 0.4884,
+      "step": 775
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 2.278213852731601e-06,
+      "loss": 0.4754,
+      "step": 776
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 2.272819195327778e-06,
+      "loss": 0.4737,
+      "step": 777
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 2.2674256043609553e-06,
+      "loss": 0.4739,
+      "step": 778
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 2.2620331051498425e-06,
+      "loss": 0.4585,
+      "step": 779
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 2.256641723008026e-06,
+      "loss": 0.468,
+      "step": 780
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 2.2512514832438466e-06,
+      "loss": 0.4789,
+      "step": 781
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 2.245862411160283e-06,
+      "loss": 0.4844,
+      "step": 782
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 2.2404745320548326e-06,
+      "loss": 0.4731,
+      "step": 783
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 2.235087871219394e-06,
+      "loss": 0.474,
+      "step": 784
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 2.2297024539401463e-06,
+      "loss": 0.4877,
+      "step": 785
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 2.224318305497429e-06,
+      "loss": 0.4919,
+      "step": 786
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 2.2189354511656267e-06,
+      "loss": 0.4884,
+      "step": 787
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 2.2135539162130516e-06,
+      "loss": 0.4605,
+      "step": 788
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 2.208173725901819e-06,
+      "loss": 0.4737,
+      "step": 789
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 2.2027949054877342e-06,
+      "loss": 0.4881,
+      "step": 790
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 2.1974174802201708e-06,
+      "loss": 0.4724,
+      "step": 791
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 2.192041475341953e-06,
+      "loss": 0.4779,
+      "step": 792
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 2.186666916089239e-06,
+      "loss": 0.471,
+      "step": 793
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 2.1812938276914016e-06,
+      "loss": 0.4874,
+      "step": 794
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 2.175922235370904e-06,
+      "loss": 0.4786,
+      "step": 795
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 2.1705521643431916e-06,
+      "loss": 0.485,
+      "step": 796
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 2.165183639816567e-06,
+      "loss": 0.4651,
+      "step": 797
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 2.1598166869920723e-06,
+      "loss": 0.4594,
+      "step": 798
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 2.154451331063374e-06,
+      "loss": 0.4779,
+      "step": 799
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 2.1490875972166394e-06,
+      "loss": 0.4697,
+      "step": 800
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 2.1437255106304232e-06,
+      "loss": 0.4598,
+      "step": 801
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 2.1383650964755473e-06,
+      "loss": 0.4685,
+      "step": 802
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 2.1330063799149825e-06,
+      "loss": 0.4808,
+      "step": 803
+    },
+    {
+      "epoch": 2.15,
+      "eval_loss": 0.520827054977417,
+      "eval_runtime": 70.6736,
+      "eval_samples_per_second": 71.653,
+      "eval_steps_per_second": 5.971,
+      "step": 803
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 2.127649386103732e-06,
+      "loss": 0.4769,
+      "step": 804
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 2.1222941401887087e-06,
+      "loss": 0.4773,
+      "step": 805
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 2.116940667308624e-06,
+      "loss": 0.4658,
+      "step": 806
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 2.1115889925938642e-06,
+      "loss": 0.4772,
+      "step": 807
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 2.106239141166377e-06,
+      "loss": 0.4744,
+      "step": 808
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 2.100891138139548e-06,
+      "loss": 0.4728,
+      "step": 809
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 2.0955450086180883e-06,
+      "loss": 0.4777,
+      "step": 810
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 2.0902007776979133e-06,
+      "loss": 0.4752,
+      "step": 811
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 2.0848584704660266e-06,
+      "loss": 0.4649,
+      "step": 812
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 2.079518112000402e-06,
+      "loss": 0.4732,
+      "step": 813
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 2.0741797273698627e-06,
+      "loss": 0.4756,
+      "step": 814
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.0688433416339694e-06,
+      "loss": 0.4684,
+      "step": 815
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.0635089798428976e-06,
+      "loss": 0.4799,
+      "step": 816
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.058176667037324e-06,
+      "loss": 0.483,
+      "step": 817
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.0528464282483047e-06,
+      "loss": 0.4611,
+      "step": 818
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 2.0475182884971606e-06,
+      "loss": 0.4668,
+      "step": 819
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 2.0421922727953597e-06,
+      "loss": 0.4638,
+      "step": 820
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 2.0368684061443995e-06,
+      "loss": 0.4475,
+      "step": 821
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 2.031546713535688e-06,
+      "loss": 0.4765,
+      "step": 822
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 2.0262272199504306e-06,
+      "loss": 0.4625,
+      "step": 823
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 2.0209099503595052e-06,
+      "loss": 0.4708,
+      "step": 824
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 2.0155949297233542e-06,
+      "loss": 0.4522,
+      "step": 825
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 2.010282182991861e-06,
+      "loss": 0.4681,
+      "step": 826
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 2.0049717351042366e-06,
+      "loss": 0.4742,
+      "step": 827
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.9996636109888966e-06,
+      "loss": 0.4777,
+      "step": 828
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.9943578355633523e-06,
+      "loss": 0.4621,
+      "step": 829
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 1.9890544337340882e-06,
+      "loss": 0.4574,
+      "step": 830
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 1.983753430396447e-06,
+      "loss": 0.4681,
+      "step": 831
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 1.9784548504345123e-06,
+      "loss": 0.4549,
+      "step": 832
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.973158718720991e-06,
+      "loss": 0.4592,
+      "step": 833
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.9678650601170986e-06,
+      "loss": 0.4699,
+      "step": 834
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.96257389947244e-06,
+      "loss": 0.4619,
+      "step": 835
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.9572852616248957e-06,
+      "loss": 0.4725,
+      "step": 836
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 1.951999171400503e-06,
+      "loss": 0.4565,
+      "step": 837
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 1.946715653613339e-06,
+      "loss": 0.4708,
+      "step": 838
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 1.941434733065406e-06,
+      "loss": 0.4544,
+      "step": 839
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 1.936156434546515e-06,
+      "loss": 0.4817,
+      "step": 840
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.930880782834168e-06,
+      "loss": 0.4692,
+      "step": 841
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.9256078026934417e-06,
+      "loss": 0.4655,
+      "step": 842
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.9203375188768726e-06,
+      "loss": 0.4584,
+      "step": 843
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 1.9150699561243397e-06,
+      "loss": 0.4458,
+      "step": 844
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 1.90980513916295e-06,
+      "loss": 0.468,
+      "step": 845
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 1.90454309270692e-06,
+      "loss": 0.4652,
+      "step": 846
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 1.8992838414574593e-06,
+      "loss": 0.4526,
+      "step": 847
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 1.8940274101026584e-06,
+      "loss": 0.4693,
+      "step": 848
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 1.8887738233173708e-06,
+      "loss": 0.4787,
+      "step": 849
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 1.8835231057630955e-06,
+      "loss": 0.4896,
+      "step": 850
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 1.8782752820878636e-06,
+      "loss": 0.4637,
+      "step": 851
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.873030376926121e-06,
+      "loss": 0.4592,
+      "step": 852
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.867788414898614e-06,
+      "loss": 0.4505,
+      "step": 853
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.8625494206122732e-06,
+      "loss": 0.4714,
+      "step": 854
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 1.8573134186600978e-06,
+      "loss": 0.4629,
+      "step": 855
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 1.8520804336210413e-06,
+      "loss": 0.4598,
+      "step": 856
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 1.8468504900598922e-06,
+      "loss": 0.4632,
+      "step": 857
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 1.8416236125271647e-06,
+      "loss": 0.454,
+      "step": 858
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.836399825558979e-06,
+      "loss": 0.4589,
+      "step": 859
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.8311791536769485e-06,
+      "loss": 0.4514,
+      "step": 860
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.8259616213880622e-06,
+      "loss": 0.4682,
+      "step": 861
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.8207472531845716e-06,
+      "loss": 0.4674,
+      "step": 862
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.8155360735438766e-06,
+      "loss": 0.4664,
+      "step": 863
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.810328106928408e-06,
+      "loss": 0.4622,
+      "step": 864
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.805123377785515e-06,
+      "loss": 0.4618,
+      "step": 865
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 1.799921910547347e-06,
+      "loss": 0.4491,
+      "step": 866
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 1.7947237296307434e-06,
+      "loss": 0.4786,
+      "step": 867
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 1.7895288594371168e-06,
+      "loss": 0.4658,
+      "step": 868
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 1.7843373243523385e-06,
+      "loss": 0.4683,
+      "step": 869
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 1.7791491487466234e-06,
+      "loss": 0.4687,
+      "step": 870
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 1.7739643569744163e-06,
+      "loss": 0.4692,
+      "step": 871
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 1.7687829733742779e-06,
+      "loss": 0.47,
+      "step": 872
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.76360502226877e-06,
+      "loss": 0.4536,
+      "step": 873
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.7584305279643433e-06,
+      "loss": 0.4499,
+      "step": 874
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.7532595147512167e-06,
+      "loss": 0.4632,
+      "step": 875
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.748092006903273e-06,
+      "loss": 0.4647,
+      "step": 876
+    },
+    {
+      "epoch": 2.35,
+      "eval_loss": 0.52239590883255,
+      "eval_runtime": 71.0114,
+      "eval_samples_per_second": 71.313,
+      "eval_steps_per_second": 5.943,
+      "step": 876
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 73,
+  "total_flos": 2200684871024640.0,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}