diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,4764 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.95195530726257,
+  "eval_steps": 112,
+  "global_step": 669,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.004464285714285714,
+      "grad_norm": 5.096107617240597,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 0.81,
+      "step": 1
+    },
+    {
+      "epoch": 0.004464285714285714,
+      "eval_loss": 0.7919066548347473,
+      "eval_runtime": 32.2179,
+      "eval_samples_per_second": 82.873,
+      "eval_steps_per_second": 5.183,
+      "step": 1
+    },
+    {
+      "epoch": 0.008928571428571428,
+      "grad_norm": 4.9041704402978805,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.7839,
+      "step": 2
+    },
+    {
+      "epoch": 0.013392857142857142,
+      "grad_norm": 4.977917588021941,
+      "learning_rate": 1.5e-06,
+      "loss": 0.7745,
+      "step": 3
+    },
+    {
+      "epoch": 0.017857142857142856,
+      "grad_norm": 4.7373014121550705,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.7681,
+      "step": 4
+    },
+    {
+      "epoch": 0.022321428571428572,
+      "grad_norm": 4.563297129857777,
+      "learning_rate": 2.5e-06,
+      "loss": 0.7589,
+      "step": 5
+    },
+    {
+      "epoch": 0.026785714285714284,
+      "grad_norm": 3.5264108754930787,
+      "learning_rate": 3e-06,
+      "loss": 0.7629,
+      "step": 6
+    },
+    {
+      "epoch": 0.03125,
+      "grad_norm": 3.4126703272457166,
+      "learning_rate": 3.5e-06,
+      "loss": 0.7181,
+      "step": 7
+    },
+    {
+      "epoch": 0.03571428571428571,
+      "grad_norm": 3.6971429493758636,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.667,
+      "step": 8
+    },
+    {
+      "epoch": 0.04017857142857143,
+      "grad_norm": 1.940701767077778,
+      "learning_rate": 4.5e-06,
+      "loss": 0.6629,
+      "step": 9
+    },
+    {
+      "epoch": 0.044642857142857144,
+      "grad_norm": 12.40862535155878,
+      "learning_rate": 5e-06,
+      "loss": 0.6832,
+      "step": 10
+    },
+    {
+      "epoch": 0.049107142857142856,
+      "grad_norm": 2.8235547786373867,
+      "learning_rate": 4.992447129909366e-06,
+      "loss": 0.6213,
+      "step": 11
+    },
+    {
+      "epoch": 0.05357142857142857,
+      "grad_norm": 1.734602557718062,
+      "learning_rate": 4.984894259818732e-06,
+      "loss": 0.6279,
+      "step": 12
+    },
+    {
+      "epoch": 0.05803571428571429,
+      "grad_norm": 1.3840828717613172,
+      "learning_rate": 4.977341389728097e-06,
+      "loss": 0.6134,
+      "step": 13
+    },
+    {
+      "epoch": 0.0625,
+      "grad_norm": 1.1262172615918475,
+      "learning_rate": 4.969788519637463e-06,
+      "loss": 0.6029,
+      "step": 14
+    },
+    {
+      "epoch": 0.06696428571428571,
+      "grad_norm": 0.9263739512436016,
+      "learning_rate": 4.962235649546828e-06,
+      "loss": 0.6067,
+      "step": 15
+    },
+    {
+      "epoch": 0.07142857142857142,
+      "grad_norm": 0.8374584146721384,
+      "learning_rate": 4.954682779456194e-06,
+      "loss": 0.6014,
+      "step": 16
+    },
+    {
+      "epoch": 0.07589285714285714,
+      "grad_norm": 0.7757487058764371,
+      "learning_rate": 4.9471299093655595e-06,
+      "loss": 0.5906,
+      "step": 17
+    },
+    {
+      "epoch": 0.08035714285714286,
+      "grad_norm": 0.6531787929864147,
+      "learning_rate": 4.939577039274925e-06,
+      "loss": 0.5589,
+      "step": 18
+    },
+    {
+      "epoch": 0.08482142857142858,
+      "grad_norm": 0.7223853317191319,
+      "learning_rate": 4.93202416918429e-06,
+      "loss": 0.5877,
+      "step": 19
+    },
+    {
+      "epoch": 0.08928571428571429,
+      "grad_norm": 0.6793848952870568,
+      "learning_rate": 4.924471299093656e-06,
+      "loss": 0.6075,
+      "step": 20
+    },
+    {
+      "epoch": 0.09375,
+      "grad_norm": 0.65689704768199,
+      "learning_rate": 4.9169184290030215e-06,
+      "loss": 0.5659,
+      "step": 21
+    },
+    {
+      "epoch": 0.09821428571428571,
+      "grad_norm": 0.6553424447930819,
+      "learning_rate": 4.909365558912387e-06,
+      "loss": 0.5728,
+      "step": 22
+    },
+    {
+      "epoch": 0.10267857142857142,
+      "grad_norm": 0.6095768945247507,
+      "learning_rate": 4.901812688821753e-06,
+      "loss": 0.5716,
+      "step": 23
+    },
+    {
+      "epoch": 0.10714285714285714,
+      "grad_norm": 0.5874011280839848,
+      "learning_rate": 4.894259818731118e-06,
+      "loss": 0.5736,
+      "step": 24
+    },
+    {
+      "epoch": 0.11160714285714286,
+      "grad_norm": 0.642646780124128,
+      "learning_rate": 4.8867069486404835e-06,
+      "loss": 0.5889,
+      "step": 25
+    },
+    {
+      "epoch": 0.11607142857142858,
+      "grad_norm": 0.5585168252549827,
+      "learning_rate": 4.879154078549849e-06,
+      "loss": 0.5761,
+      "step": 26
+    },
+    {
+      "epoch": 0.12053571428571429,
+      "grad_norm": 0.4948199084353006,
+      "learning_rate": 4.871601208459215e-06,
+      "loss": 0.545,
+      "step": 27
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 0.580699538693282,
+      "learning_rate": 4.864048338368581e-06,
+      "loss": 0.5874,
+      "step": 28
+    },
+    {
+      "epoch": 0.12946428571428573,
+      "grad_norm": 0.5483036703790811,
+      "learning_rate": 4.8564954682779455e-06,
+      "loss": 0.553,
+      "step": 29
+    },
+    {
+      "epoch": 0.13392857142857142,
+      "grad_norm": 0.5207516065535861,
+      "learning_rate": 4.848942598187312e-06,
+      "loss": 0.54,
+      "step": 30
+    },
+    {
+      "epoch": 0.13839285714285715,
+      "grad_norm": 0.5453894595978196,
+      "learning_rate": 4.841389728096677e-06,
+      "loss": 0.5779,
+      "step": 31
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 0.5223897707340578,
+      "learning_rate": 4.833836858006043e-06,
+      "loss": 0.5518,
+      "step": 32
+    },
+    {
+      "epoch": 0.14732142857142858,
+      "grad_norm": 0.5473812522488812,
+      "learning_rate": 4.826283987915408e-06,
+      "loss": 0.5831,
+      "step": 33
+    },
+    {
+      "epoch": 0.15178571428571427,
+      "grad_norm": 0.5437546013329395,
+      "learning_rate": 4.818731117824774e-06,
+      "loss": 0.5884,
+      "step": 34
+    },
+    {
+      "epoch": 0.15625,
+      "grad_norm": 0.5305579139277316,
+      "learning_rate": 4.81117824773414e-06,
+      "loss": 0.5398,
+      "step": 35
+    },
+    {
+      "epoch": 0.16071428571428573,
+      "grad_norm": 0.5258419287632591,
+      "learning_rate": 4.803625377643505e-06,
+      "loss": 0.5593,
+      "step": 36
+    },
+    {
+      "epoch": 0.16517857142857142,
+      "grad_norm": 0.5158851354579528,
+      "learning_rate": 4.79607250755287e-06,
+      "loss": 0.5564,
+      "step": 37
+    },
+    {
+      "epoch": 0.16964285714285715,
+      "grad_norm": 0.4824844062486759,
+      "learning_rate": 4.788519637462236e-06,
+      "loss": 0.5616,
+      "step": 38
+    },
+    {
+      "epoch": 0.17410714285714285,
+      "grad_norm": 0.5173226727977146,
+      "learning_rate": 4.780966767371602e-06,
+      "loss": 0.5674,
+      "step": 39
+    },
+    {
+      "epoch": 0.17857142857142858,
+      "grad_norm": 0.4966629400190002,
+      "learning_rate": 4.773413897280967e-06,
+      "loss": 0.5833,
+      "step": 40
+    },
+    {
+      "epoch": 0.18303571428571427,
+      "grad_norm": 0.48186435126437244,
+      "learning_rate": 4.765861027190333e-06,
+      "loss": 0.5571,
+      "step": 41
+    },
+    {
+      "epoch": 0.1875,
+      "grad_norm": 0.5029558558326376,
+      "learning_rate": 4.758308157099698e-06,
+      "loss": 0.5358,
+      "step": 42
+    },
+    {
+      "epoch": 0.19196428571428573,
+      "grad_norm": 0.5249190794688006,
+      "learning_rate": 4.750755287009064e-06,
+      "loss": 0.5542,
+      "step": 43
+    },
+    {
+      "epoch": 0.19642857142857142,
+      "grad_norm": 0.4622817114582695,
+      "learning_rate": 4.743202416918429e-06,
+      "loss": 0.54,
+      "step": 44
+    },
+    {
+      "epoch": 0.20089285714285715,
+      "grad_norm": 0.49789824735563454,
+      "learning_rate": 4.735649546827795e-06,
+      "loss": 0.5308,
+      "step": 45
+    },
+    {
+      "epoch": 0.20535714285714285,
+      "grad_norm": 0.51199116739647,
+      "learning_rate": 4.728096676737161e-06,
+      "loss": 0.5784,
+      "step": 46
+    },
+    {
+      "epoch": 0.20982142857142858,
+      "grad_norm": 0.47152096107780506,
+      "learning_rate": 4.720543806646526e-06,
+      "loss": 0.5654,
+      "step": 47
+    },
+    {
+      "epoch": 0.21428571428571427,
+      "grad_norm": 0.5233581267403502,
+      "learning_rate": 4.712990936555891e-06,
+      "loss": 0.5727,
+      "step": 48
+    },
+    {
+      "epoch": 0.21875,
+      "grad_norm": 0.517908769266997,
+      "learning_rate": 4.705438066465257e-06,
+      "loss": 0.5258,
+      "step": 49
+    },
+    {
+      "epoch": 0.22321428571428573,
+      "grad_norm": 0.5103782222366013,
+      "learning_rate": 4.697885196374623e-06,
+      "loss": 0.564,
+      "step": 50
+    },
+    {
+      "epoch": 0.22767857142857142,
+      "grad_norm": 0.5894196264675328,
+      "learning_rate": 4.6903323262839885e-06,
+      "loss": 0.5355,
+      "step": 51
+    },
+    {
+      "epoch": 0.23214285714285715,
+      "grad_norm": 0.48136408513935,
+      "learning_rate": 4.682779456193353e-06,
+      "loss": 0.5714,
+      "step": 52
+    },
+    {
+      "epoch": 0.23660714285714285,
+      "grad_norm": 0.5996790928794941,
+      "learning_rate": 4.67522658610272e-06,
+      "loss": 0.5385,
+      "step": 53
+    },
+    {
+      "epoch": 0.24107142857142858,
+      "grad_norm": 0.5231061994660855,
+      "learning_rate": 4.667673716012085e-06,
+      "loss": 0.5429,
+      "step": 54
+    },
+    {
+      "epoch": 0.24553571428571427,
+      "grad_norm": 0.5416056944426403,
+      "learning_rate": 4.6601208459214505e-06,
+      "loss": 0.5423,
+      "step": 55
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.5734105721889117,
+      "learning_rate": 4.652567975830816e-06,
+      "loss": 0.5756,
+      "step": 56
+    },
+    {
+      "epoch": 0.2544642857142857,
+      "grad_norm": 0.5318541491738474,
+      "learning_rate": 4.645015105740182e-06,
+      "loss": 0.5486,
+      "step": 57
+    },
+    {
+      "epoch": 0.25892857142857145,
+      "grad_norm": 0.4839987879630258,
+      "learning_rate": 4.637462235649548e-06,
+      "loss": 0.5517,
+      "step": 58
+    },
+    {
+      "epoch": 0.26339285714285715,
+      "grad_norm": 0.5843387907103592,
+      "learning_rate": 4.6299093655589125e-06,
+      "loss": 0.5288,
+      "step": 59
+    },
+    {
+      "epoch": 0.26785714285714285,
+      "grad_norm": 0.5496996598474941,
+      "learning_rate": 4.622356495468278e-06,
+      "loss": 0.542,
+      "step": 60
+    },
+    {
+      "epoch": 0.27232142857142855,
+      "grad_norm": 0.5192508705264864,
+      "learning_rate": 4.614803625377644e-06,
+      "loss": 0.5291,
+      "step": 61
+    },
+    {
+      "epoch": 0.2767857142857143,
+      "grad_norm": 0.5918952697319948,
+      "learning_rate": 4.60725075528701e-06,
+      "loss": 0.5229,
+      "step": 62
+    },
+    {
+      "epoch": 0.28125,
+      "grad_norm": 0.5914173604947244,
+      "learning_rate": 4.5996978851963745e-06,
+      "loss": 0.5754,
+      "step": 63
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 0.5298896547584293,
+      "learning_rate": 4.592145015105741e-06,
+      "loss": 0.5687,
+      "step": 64
+    },
+    {
+      "epoch": 0.29017857142857145,
+      "grad_norm": 0.5671535875314645,
+      "learning_rate": 4.584592145015106e-06,
+      "loss": 0.5513,
+      "step": 65
+    },
+    {
+      "epoch": 0.29464285714285715,
+      "grad_norm": 0.5234306194196137,
+      "learning_rate": 4.577039274924472e-06,
+      "loss": 0.5226,
+      "step": 66
+    },
+    {
+      "epoch": 0.29910714285714285,
+      "grad_norm": 0.4959853502285333,
+      "learning_rate": 4.569486404833837e-06,
+      "loss": 0.5335,
+      "step": 67
+    },
+    {
+      "epoch": 0.30357142857142855,
+      "grad_norm": 0.47729026424742405,
+      "learning_rate": 4.561933534743202e-06,
+      "loss": 0.5248,
+      "step": 68
+    },
+    {
+      "epoch": 0.3080357142857143,
+      "grad_norm": 0.47294297292566195,
+      "learning_rate": 4.554380664652569e-06,
+      "loss": 0.5588,
+      "step": 69
+    },
+    {
+      "epoch": 0.3125,
+      "grad_norm": 0.5275786368006364,
+      "learning_rate": 4.5468277945619336e-06,
+      "loss": 0.5483,
+      "step": 70
+    },
+    {
+      "epoch": 0.3169642857142857,
+      "grad_norm": 0.5057705930103896,
+      "learning_rate": 4.539274924471299e-06,
+      "loss": 0.5236,
+      "step": 71
+    },
+    {
+      "epoch": 0.32142857142857145,
+      "grad_norm": 0.49625696780589473,
+      "learning_rate": 4.531722054380665e-06,
+      "loss": 0.5221,
+      "step": 72
+    },
+    {
+      "epoch": 0.32589285714285715,
+      "grad_norm": 0.49548627058675154,
+      "learning_rate": 4.524169184290031e-06,
+      "loss": 0.5176,
+      "step": 73
+    },
+    {
+      "epoch": 0.33035714285714285,
+      "grad_norm": 0.5011897814029462,
+      "learning_rate": 4.516616314199396e-06,
+      "loss": 0.5431,
+      "step": 74
+    },
+    {
+      "epoch": 0.33482142857142855,
+      "grad_norm": 0.5168007116628185,
+      "learning_rate": 4.509063444108761e-06,
+      "loss": 0.5289,
+      "step": 75
+    },
+    {
+      "epoch": 0.3392857142857143,
+      "grad_norm": 0.5199488476439844,
+      "learning_rate": 4.501510574018128e-06,
+      "loss": 0.5437,
+      "step": 76
+    },
+    {
+      "epoch": 0.34375,
+      "grad_norm": 0.5776348374651958,
+      "learning_rate": 4.493957703927493e-06,
+      "loss": 0.5424,
+      "step": 77
+    },
+    {
+      "epoch": 0.3482142857142857,
+      "grad_norm": 0.5785274375251012,
+      "learning_rate": 4.486404833836858e-06,
+      "loss": 0.5252,
+      "step": 78
+    },
+    {
+      "epoch": 0.35267857142857145,
+      "grad_norm": 0.5624252036929408,
+      "learning_rate": 4.478851963746224e-06,
+      "loss": 0.5502,
+      "step": 79
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 0.5630273292563954,
+      "learning_rate": 4.47129909365559e-06,
+      "loss": 0.5249,
+      "step": 80
+    },
+    {
+      "epoch": 0.36160714285714285,
+      "grad_norm": 0.5229643469397499,
+      "learning_rate": 4.463746223564955e-06,
+      "loss": 0.572,
+      "step": 81
+    },
+    {
+      "epoch": 0.36607142857142855,
+      "grad_norm": 0.5401109169653766,
+      "learning_rate": 4.45619335347432e-06,
+      "loss": 0.5174,
+      "step": 82
+    },
+    {
+      "epoch": 0.3705357142857143,
+      "grad_norm": 0.572853251261138,
+      "learning_rate": 4.448640483383686e-06,
+      "loss": 0.5336,
+      "step": 83
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 0.49833694539628726,
+      "learning_rate": 4.441087613293052e-06,
+      "loss": 0.5205,
+      "step": 84
+    },
+    {
+      "epoch": 0.3794642857142857,
+      "grad_norm": 0.5167418841184229,
+      "learning_rate": 4.4335347432024175e-06,
+      "loss": 0.5405,
+      "step": 85
+    },
+    {
+      "epoch": 0.38392857142857145,
+      "grad_norm": 0.5243612918158316,
+      "learning_rate": 4.425981873111782e-06,
+      "loss": 0.5204,
+      "step": 86
+    },
+    {
+      "epoch": 0.38839285714285715,
+      "grad_norm": 0.5159624173670619,
+      "learning_rate": 4.418429003021149e-06,
+      "loss": 0.5315,
+      "step": 87
+    },
+    {
+      "epoch": 0.39285714285714285,
+      "grad_norm": 0.5227918139330733,
+      "learning_rate": 4.410876132930514e-06,
+      "loss": 0.5418,
+      "step": 88
+    },
+    {
+      "epoch": 0.39732142857142855,
+      "grad_norm": 0.48045718180033986,
+      "learning_rate": 4.4033232628398795e-06,
+      "loss": 0.506,
+      "step": 89
+    },
+    {
+      "epoch": 0.4017857142857143,
+      "grad_norm": 0.5791314708394477,
+      "learning_rate": 4.395770392749245e-06,
+      "loss": 0.596,
+      "step": 90
+    },
+    {
+      "epoch": 0.40625,
+      "grad_norm": 0.6076262289549809,
+      "learning_rate": 4.38821752265861e-06,
+      "loss": 0.5312,
+      "step": 91
+    },
+    {
+      "epoch": 0.4107142857142857,
+      "grad_norm": 0.4780377894473343,
+      "learning_rate": 4.380664652567977e-06,
+      "loss": 0.549,
+      "step": 92
+    },
+    {
+      "epoch": 0.41517857142857145,
+      "grad_norm": 0.5549693296410254,
+      "learning_rate": 4.3731117824773415e-06,
+      "loss": 0.5159,
+      "step": 93
+    },
+    {
+      "epoch": 0.41964285714285715,
+      "grad_norm": 0.5097636365128929,
+      "learning_rate": 4.365558912386707e-06,
+      "loss": 0.5612,
+      "step": 94
+    },
+    {
+      "epoch": 0.42410714285714285,
+      "grad_norm": 0.5021825140627123,
+      "learning_rate": 4.358006042296073e-06,
+      "loss": 0.5653,
+      "step": 95
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 0.537945728158161,
+      "learning_rate": 4.350453172205439e-06,
+      "loss": 0.528,
+      "step": 96
+    },
+    {
+      "epoch": 0.4330357142857143,
+      "grad_norm": 0.4846132130879552,
+      "learning_rate": 4.342900302114804e-06,
+      "loss": 0.5395,
+      "step": 97
+    },
+    {
+      "epoch": 0.4375,
+      "grad_norm": 0.5206116049646974,
+      "learning_rate": 4.335347432024169e-06,
+      "loss": 0.5445,
+      "step": 98
+    },
+    {
+      "epoch": 0.4419642857142857,
+      "grad_norm": 0.551262765339688,
+      "learning_rate": 4.327794561933535e-06,
+      "loss": 0.5346,
+      "step": 99
+    },
+    {
+      "epoch": 0.44642857142857145,
+      "grad_norm": 0.4984265695742978,
+      "learning_rate": 4.3202416918429006e-06,
+      "loss": 0.5161,
+      "step": 100
+    },
+    {
+      "epoch": 0.45089285714285715,
+      "grad_norm": 0.49943465819192145,
+      "learning_rate": 4.312688821752266e-06,
+      "loss": 0.5164,
+      "step": 101
+    },
+    {
+      "epoch": 0.45535714285714285,
+      "grad_norm": 0.48826983665318036,
+      "learning_rate": 4.305135951661632e-06,
+      "loss": 0.552,
+      "step": 102
+    },
+    {
+      "epoch": 0.45982142857142855,
+      "grad_norm": 0.5320671379410685,
+      "learning_rate": 4.297583081570998e-06,
+      "loss": 0.5512,
+      "step": 103
+    },
+    {
+      "epoch": 0.4642857142857143,
+      "grad_norm": 0.5027439491810191,
+      "learning_rate": 4.2900302114803626e-06,
+      "loss": 0.5728,
+      "step": 104
+    },
+    {
+      "epoch": 0.46875,
+      "grad_norm": 0.4875858103308158,
+      "learning_rate": 4.282477341389728e-06,
+      "loss": 0.5306,
+      "step": 105
+    },
+    {
+      "epoch": 0.4732142857142857,
+      "grad_norm": 0.5205285237402679,
+      "learning_rate": 4.274924471299094e-06,
+      "loss": 0.5274,
+      "step": 106
+    },
+    {
+      "epoch": 0.47767857142857145,
+      "grad_norm": 0.47059444782816534,
+      "learning_rate": 4.26737160120846e-06,
+      "loss": 0.5092,
+      "step": 107
+    },
+    {
+      "epoch": 0.48214285714285715,
+      "grad_norm": 0.4998056822147382,
+      "learning_rate": 4.259818731117825e-06,
+      "loss": 0.5481,
+      "step": 108
+    },
+    {
+      "epoch": 0.48660714285714285,
+      "grad_norm": 0.5403989848602658,
+      "learning_rate": 4.25226586102719e-06,
+      "loss": 0.518,
+      "step": 109
+    },
+    {
+      "epoch": 0.49107142857142855,
+      "grad_norm": 0.48519680312136604,
+      "learning_rate": 4.244712990936557e-06,
+      "loss": 0.5522,
+      "step": 110
+    },
+    {
+      "epoch": 0.4955357142857143,
+      "grad_norm": 0.5655590239975465,
+      "learning_rate": 4.237160120845922e-06,
+      "loss": 0.526,
+      "step": 111
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.49558151987556875,
+      "learning_rate": 4.229607250755287e-06,
+      "loss": 0.5275,
+      "step": 112
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 0.5346882343292236,
+      "eval_runtime": 32.2399,
+      "eval_samples_per_second": 82.817,
+      "eval_steps_per_second": 5.18,
+      "step": 112
+    },
+    {
+      "epoch": 0.5050279329608939,
+      "grad_norm": 0.5662939933016762,
+      "learning_rate": 4.218512898330804e-06,
+      "loss": 0.515,
+      "step": 113
+    },
+    {
+      "epoch": 0.5094972067039106,
+      "grad_norm": 0.5225180760232837,
+      "learning_rate": 4.2109256449165405e-06,
+      "loss": 0.5038,
+      "step": 114
+    },
+    {
+      "epoch": 0.5139664804469274,
+      "grad_norm": 0.5377410752157915,
+      "learning_rate": 4.203338391502276e-06,
+      "loss": 0.5043,
+      "step": 115
+    },
+    {
+      "epoch": 0.5184357541899441,
+      "grad_norm": 0.5915094303426857,
+      "learning_rate": 4.195751138088012e-06,
+      "loss": 0.5115,
+      "step": 116
+    },
+    {
+      "epoch": 0.5229050279329609,
+      "grad_norm": 0.5079762717076601,
+      "learning_rate": 4.1881638846737485e-06,
+      "loss": 0.5491,
+      "step": 117
+    },
+    {
+      "epoch": 0.5273743016759777,
+      "grad_norm": 0.5933329970849119,
+      "learning_rate": 4.180576631259484e-06,
+      "loss": 0.5095,
+      "step": 118
+    },
+    {
+      "epoch": 0.5318435754189944,
+      "grad_norm": 0.5053074762991974,
+      "learning_rate": 4.17298937784522e-06,
+      "loss": 0.4881,
+      "step": 119
+    },
+    {
+      "epoch": 0.5363128491620112,
+      "grad_norm": 0.6515342691068152,
+      "learning_rate": 4.1654021244309564e-06,
+      "loss": 0.5069,
+      "step": 120
+    },
+    {
+      "epoch": 0.5407821229050279,
+      "grad_norm": 0.5774345075195761,
+      "learning_rate": 4.157814871016692e-06,
+      "loss": 0.5007,
+      "step": 121
+    },
+    {
+      "epoch": 0.5452513966480447,
+      "grad_norm": 0.6308991657157696,
+      "learning_rate": 4.150227617602428e-06,
+      "loss": 0.5083,
+      "step": 122
+    },
+    {
+      "epoch": 0.5497206703910614,
+      "grad_norm": 0.5258198371155504,
+      "learning_rate": 4.142640364188164e-06,
+      "loss": 0.5103,
+      "step": 123
+    },
+    {
+      "epoch": 0.5541899441340782,
+      "grad_norm": 0.5629243370057063,
+      "learning_rate": 4.1350531107739e-06,
+      "loss": 0.5016,
+      "step": 124
+    },
+    {
+      "epoch": 0.5586592178770949,
+      "grad_norm": 0.5440095922230602,
+      "learning_rate": 4.127465857359636e-06,
+      "loss": 0.491,
+      "step": 125
+    },
+    {
+      "epoch": 0.5631284916201117,
+      "grad_norm": 0.5126793634050918,
+      "learning_rate": 4.119878603945372e-06,
+      "loss": 0.5239,
+      "step": 126
+    },
+    {
+      "epoch": 0.5675977653631284,
+      "grad_norm": 0.5476299968831635,
+      "learning_rate": 4.112291350531108e-06,
+      "loss": 0.5176,
+      "step": 127
+    },
+    {
+      "epoch": 0.5720670391061452,
+      "grad_norm": 0.5570481457317527,
+      "learning_rate": 4.104704097116844e-06,
+      "loss": 0.5084,
+      "step": 128
+    },
+    {
+      "epoch": 0.576536312849162,
+      "grad_norm": 0.49480777896006917,
+      "learning_rate": 4.09711684370258e-06,
+      "loss": 0.4966,
+      "step": 129
+    },
+    {
+      "epoch": 0.5810055865921788,
+      "grad_norm": 0.5089889598386482,
+      "learning_rate": 4.089529590288316e-06,
+      "loss": 0.5059,
+      "step": 130
+    },
+    {
+      "epoch": 0.5854748603351956,
+      "grad_norm": 0.4954780811768282,
+      "learning_rate": 4.081942336874052e-06,
+      "loss": 0.5166,
+      "step": 131
+    },
+    {
+      "epoch": 0.5899441340782123,
+      "grad_norm": 0.4809684003541764,
+      "learning_rate": 4.074355083459787e-06,
+      "loss": 0.4857,
+      "step": 132
+    },
+    {
+      "epoch": 0.5944134078212291,
+      "grad_norm": 0.5009748980192239,
+      "learning_rate": 4.066767830045524e-06,
+      "loss": 0.523,
+      "step": 133
+    },
+    {
+      "epoch": 0.5988826815642458,
+      "grad_norm": 0.48320950023108783,
+      "learning_rate": 4.05918057663126e-06,
+      "loss": 0.4956,
+      "step": 134
+    },
+    {
+      "epoch": 0.6033519553072626,
+      "grad_norm": 0.49399181892388616,
+      "learning_rate": 4.051593323216995e-06,
+      "loss": 0.5163,
+      "step": 135
+    },
+    {
+      "epoch": 0.6078212290502794,
+      "grad_norm": 0.4821527638840584,
+      "learning_rate": 4.044006069802732e-06,
+      "loss": 0.4981,
+      "step": 136
+    },
+    {
+      "epoch": 0.6122905027932961,
+      "grad_norm": 0.5099314085164022,
+      "learning_rate": 4.036418816388468e-06,
+      "loss": 0.5121,
+      "step": 137
+    },
+    {
+      "epoch": 0.6167597765363129,
+      "grad_norm": 0.5323495276796993,
+      "learning_rate": 4.028831562974203e-06,
+      "loss": 0.5074,
+      "step": 138
+    },
+    {
+      "epoch": 0.6212290502793296,
+      "grad_norm": 0.48851533865778524,
+      "learning_rate": 4.02124430955994e-06,
+      "loss": 0.52,
+      "step": 139
+    },
+    {
+      "epoch": 0.6256983240223464,
+      "grad_norm": 0.5281525574471102,
+      "learning_rate": 4.0136570561456756e-06,
+      "loss": 0.5039,
+      "step": 140
+    },
+    {
+      "epoch": 0.6301675977653631,
+      "grad_norm": 0.5279406819094602,
+      "learning_rate": 4.006069802731411e-06,
+      "loss": 0.4933,
+      "step": 141
+    },
+    {
+      "epoch": 0.6346368715083799,
+      "grad_norm": 0.45822529010842367,
+      "learning_rate": 3.998482549317148e-06,
+      "loss": 0.5151,
+      "step": 142
+    },
+    {
+      "epoch": 0.6391061452513966,
+      "grad_norm": 0.5349296018392958,
+      "learning_rate": 3.9908952959028835e-06,
+      "loss": 0.4983,
+      "step": 143
+    },
+    {
+      "epoch": 0.6435754189944134,
+      "grad_norm": 0.5316574655817645,
+      "learning_rate": 3.983308042488619e-06,
+      "loss": 0.5344,
+      "step": 144
+    },
+    {
+      "epoch": 0.6480446927374302,
+      "grad_norm": 0.5107562456487705,
+      "learning_rate": 3.975720789074356e-06,
+      "loss": 0.52,
+      "step": 145
+    },
+    {
+      "epoch": 0.6525139664804469,
+      "grad_norm": 0.5072583855988344,
+      "learning_rate": 3.9681335356600915e-06,
+      "loss": 0.4923,
+      "step": 146
+    },
+    {
+      "epoch": 0.6569832402234637,
+      "grad_norm": 0.5836000010606872,
+      "learning_rate": 3.960546282245827e-06,
+      "loss": 0.5295,
+      "step": 147
+    },
+    {
+      "epoch": 0.6614525139664804,
+      "grad_norm": 0.49884721978842206,
+      "learning_rate": 3.952959028831564e-06,
+      "loss": 0.5216,
+      "step": 148
+    },
+    {
+      "epoch": 0.6659217877094972,
+      "grad_norm": 0.5120991222561042,
+      "learning_rate": 3.945371775417299e-06,
+      "loss": 0.5414,
+      "step": 149
+    },
+    {
+      "epoch": 0.6703910614525139,
+      "grad_norm": 0.49838481084559894,
+      "learning_rate": 3.937784522003035e-06,
+      "loss": 0.5381,
+      "step": 150
+    },
+    {
+      "epoch": 0.6748603351955307,
+      "grad_norm": 0.5398799415397277,
+      "learning_rate": 3.930197268588772e-06,
+      "loss": 0.4896,
+      "step": 151
+    },
+    {
+      "epoch": 0.6793296089385474,
+      "grad_norm": 0.565748454568293,
+      "learning_rate": 3.922610015174507e-06,
+      "loss": 0.4985,
+      "step": 152
+    },
+    {
+      "epoch": 0.6837988826815642,
+      "grad_norm": 0.5986782671726579,
+      "learning_rate": 3.915022761760243e-06,
+      "loss": 0.5125,
+      "step": 153
+    },
+    {
+      "epoch": 0.6882681564245811,
+      "grad_norm": 0.5290750194980306,
+      "learning_rate": 3.907435508345979e-06,
+      "loss": 0.5078,
+      "step": 154
+    },
+    {
+      "epoch": 0.6927374301675978,
+      "grad_norm": 0.5706479438251948,
+      "learning_rate": 3.899848254931715e-06,
+      "loss": 0.5205,
+      "step": 155
+    },
+    {
+      "epoch": 0.6972067039106146,
+      "grad_norm": 0.5173864500214489,
+      "learning_rate": 3.892261001517451e-06,
+      "loss": 0.4988,
+      "step": 156
+    },
+    {
+      "epoch": 0.7016759776536313,
+      "grad_norm": 0.4920045809108581,
+      "learning_rate": 3.884673748103187e-06,
+      "loss": 0.4954,
+      "step": 157
+    },
+    {
+      "epoch": 0.7061452513966481,
+      "grad_norm": 0.5237901072069291,
+      "learning_rate": 3.877086494688923e-06,
+      "loss": 0.5253,
+      "step": 158
+    },
+    {
+      "epoch": 0.7106145251396648,
+      "grad_norm": 0.5212795514388354,
+      "learning_rate": 3.869499241274659e-06,
+      "loss": 0.5029,
+      "step": 159
+    },
+    {
+      "epoch": 0.7150837988826816,
+      "grad_norm": 0.48658569533086143,
+      "learning_rate": 3.861911987860395e-06,
+      "loss": 0.5023,
+      "step": 160
+    },
+    {
+      "epoch": 0.7195530726256983,
+      "grad_norm": 0.5145837815914728,
+      "learning_rate": 3.854324734446131e-06,
+      "loss": 0.5436,
+      "step": 161
+    },
+    {
+      "epoch": 0.7240223463687151,
+      "grad_norm": 0.49182827281373437,
+      "learning_rate": 3.846737481031867e-06,
+      "loss": 0.4948,
+      "step": 162
+    },
+    {
+      "epoch": 0.7284916201117319,
+      "grad_norm": 0.5113640333759738,
+      "learning_rate": 3.839150227617603e-06,
+      "loss": 0.4778,
+      "step": 163
+    },
+    {
+      "epoch": 0.7329608938547486,
+      "grad_norm": 0.47675402744999507,
+      "learning_rate": 3.831562974203339e-06,
+      "loss": 0.5182,
+      "step": 164
+    },
+    {
+      "epoch": 0.7374301675977654,
+      "grad_norm": 0.5214515656344721,
+      "learning_rate": 3.823975720789075e-06,
+      "loss": 0.5204,
+      "step": 165
+    },
+    {
+      "epoch": 0.7418994413407821,
+      "grad_norm": 0.5130606167326404,
+      "learning_rate": 3.816388467374811e-06,
+      "loss": 0.492,
+      "step": 166
+    },
+    {
+      "epoch": 0.7463687150837989,
+      "grad_norm": 0.49678736596959705,
+      "learning_rate": 3.8088012139605467e-06,
+      "loss": 0.4997,
+      "step": 167
+    },
+    {
+      "epoch": 0.7508379888268156,
+      "grad_norm": 0.5275415964748987,
+      "learning_rate": 3.801213960546283e-06,
+      "loss": 0.5163,
+      "step": 168
+    },
+    {
+      "epoch": 0.7553072625698324,
+      "grad_norm": 0.5323621998896121,
+      "learning_rate": 3.7936267071320185e-06,
+      "loss": 0.4846,
+      "step": 169
+    },
+    {
+      "epoch": 0.7597765363128491,
+      "grad_norm": 0.5302675218909635,
+      "learning_rate": 3.7860394537177547e-06,
+      "loss": 0.5193,
+      "step": 170
+    },
+    {
+      "epoch": 0.7642458100558659,
+      "grad_norm": 0.5097255453283106,
+      "learning_rate": 3.778452200303491e-06,
+      "loss": 0.5248,
+      "step": 171
+    },
+    {
+      "epoch": 0.7687150837988826,
+      "grad_norm": 0.5221084637587756,
+      "learning_rate": 3.7708649468892265e-06,
+      "loss": 0.5217,
+      "step": 172
+    },
+    {
+      "epoch": 0.7731843575418994,
+      "grad_norm": 0.5823307367451829,
+      "learning_rate": 3.7632776934749626e-06,
+      "loss": 0.5057,
+      "step": 173
+    },
+    {
+      "epoch": 0.7776536312849162,
+      "grad_norm": 0.48586162675121547,
+      "learning_rate": 3.7556904400606987e-06,
+      "loss": 0.5372,
+      "step": 174
+    },
+    {
+      "epoch": 0.7821229050279329,
+      "grad_norm": 0.5099958749272927,
+      "learning_rate": 3.748103186646434e-06,
+      "loss": 0.4844,
+      "step": 175
+    },
+    {
+      "epoch": 0.7865921787709497,
+      "grad_norm": 0.5221594966770278,
+      "learning_rate": 3.74051593323217e-06,
+      "loss": 0.4968,
+      "step": 176
+    },
+    {
+      "epoch": 0.7910614525139665,
+      "grad_norm": 0.5392129841421105,
+      "learning_rate": 3.732928679817906e-06,
+      "loss": 0.4906,
+      "step": 177
+    },
+    {
+      "epoch": 0.7955307262569833,
+      "grad_norm": 0.5239629013619267,
+      "learning_rate": 3.725341426403642e-06,
+      "loss": 0.5172,
+      "step": 178
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.537861417175095,
+      "learning_rate": 3.717754172989378e-06,
+      "loss": 0.4914,
+      "step": 179
+    },
+    {
+      "epoch": 0.8044692737430168,
+      "grad_norm": 0.5001204419984062,
+      "learning_rate": 3.710166919575114e-06,
+      "loss": 0.5014,
+      "step": 180
+    },
+    {
+      "epoch": 0.8089385474860336,
+      "grad_norm": 0.5243758405421232,
+      "learning_rate": 3.70257966616085e-06,
+      "loss": 0.4979,
+      "step": 181
+    },
+    {
+      "epoch": 0.8134078212290503,
+      "grad_norm": 0.507560920806033,
+      "learning_rate": 3.6949924127465856e-06,
+      "loss": 0.5003,
+      "step": 182
+    },
+    {
+      "epoch": 0.8178770949720671,
+      "grad_norm": 0.5623627396713589,
+      "learning_rate": 3.6874051593323218e-06,
+      "loss": 0.5011,
+      "step": 183
+    },
+    {
+      "epoch": 0.8223463687150838,
+      "grad_norm": 0.5225320984203384,
+      "learning_rate": 3.679817905918058e-06,
+      "loss": 0.5117,
+      "step": 184
+    },
+    {
+      "epoch": 0.8268156424581006,
+      "grad_norm": 0.5371208831947621,
+      "learning_rate": 3.6722306525037936e-06,
+      "loss": 0.5028,
+      "step": 185
+    },
+    {
+      "epoch": 0.8312849162011173,
+      "grad_norm": 0.5412513689241548,
+      "learning_rate": 3.6646433990895297e-06,
+      "loss": 0.5111,
+      "step": 186
+    },
+    {
+      "epoch": 0.8357541899441341,
+      "grad_norm": 0.49083593539874787,
+      "learning_rate": 3.657056145675266e-06,
+      "loss": 0.5002,
+      "step": 187
+    },
+    {
+      "epoch": 0.8402234636871508,
+      "grad_norm": 0.5144646076400369,
+      "learning_rate": 3.6494688922610015e-06,
+      "loss": 0.4875,
+      "step": 188
+    },
+    {
+      "epoch": 0.8446927374301676,
+      "grad_norm": 0.5385458846653849,
+      "learning_rate": 3.6418816388467377e-06,
+      "loss": 0.5072,
+      "step": 189
+    },
+    {
+      "epoch": 0.8491620111731844,
+      "grad_norm": 0.48088002744673064,
+      "learning_rate": 3.6342943854324738e-06,
+      "loss": 0.549,
+      "step": 190
+    },
+    {
+      "epoch": 0.8536312849162011,
+      "grad_norm": 0.563960197772806,
+      "learning_rate": 3.6267071320182095e-06,
+      "loss": 0.5344,
+      "step": 191
+    },
+    {
+      "epoch": 0.8581005586592179,
+      "grad_norm": 0.5235289059912077,
+      "learning_rate": 3.6191198786039456e-06,
+      "loss": 0.5172,
+      "step": 192
+    },
+    {
+      "epoch": 0.8625698324022346,
+      "grad_norm": 0.5059904029741168,
+      "learning_rate": 3.6115326251896813e-06,
+      "loss": 0.4908,
+      "step": 193
+    },
+    {
+      "epoch": 0.8670391061452514,
+      "grad_norm": 0.507980780742306,
+      "learning_rate": 3.6039453717754174e-06,
+      "loss": 0.4951,
+      "step": 194
+    },
+    {
+      "epoch": 0.8715083798882681,
+      "grad_norm": 0.5552145890763519,
+      "learning_rate": 3.5963581183611536e-06,
+      "loss": 0.505,
+      "step": 195
+    },
+    {
+      "epoch": 0.8759776536312849,
+      "grad_norm": 0.5379367693029108,
+      "learning_rate": 3.5887708649468893e-06,
+      "loss": 0.52,
+      "step": 196
+    },
+    {
+      "epoch": 0.8804469273743016,
+      "grad_norm": 0.5813281708032492,
+      "learning_rate": 3.5811836115326254e-06,
+      "loss": 0.4966,
+      "step": 197
+    },
+    {
+      "epoch": 0.8849162011173184,
+      "grad_norm": 0.5138017935214574,
+      "learning_rate": 3.5735963581183615e-06,
+      "loss": 0.495,
+      "step": 198
+    },
+    {
+      "epoch": 0.8893854748603351,
+      "grad_norm": 0.5116405158423362,
+      "learning_rate": 3.5660091047040972e-06,
+      "loss": 0.519,
+      "step": 199
+    },
+    {
+      "epoch": 0.8938547486033519,
+      "grad_norm": 0.5696283061922035,
+      "learning_rate": 3.5584218512898333e-06,
+      "loss": 0.5002,
+      "step": 200
+    },
+    {
+      "epoch": 0.8983240223463688,
+      "grad_norm": 0.5541549121266046,
+      "learning_rate": 3.5508345978755695e-06,
+      "loss": 0.5125,
+      "step": 201
+    },
+    {
+      "epoch": 0.9027932960893855,
+      "grad_norm": 0.5017863683272789,
+      "learning_rate": 3.543247344461305e-06,
+      "loss": 0.499,
+      "step": 202
+    },
+    {
+      "epoch": 0.9072625698324023,
+      "grad_norm": 0.5776107100347877,
+      "learning_rate": 3.5356600910470413e-06,
+      "loss": 0.4883,
+      "step": 203
+    },
+    {
+      "epoch": 0.911731843575419,
+      "grad_norm": 0.5350847562000018,
+      "learning_rate": 3.528072837632777e-06,
+      "loss": 0.5052,
+      "step": 204
+    },
+    {
+      "epoch": 0.9162011173184358,
+      "grad_norm": 0.5012614076658582,
+      "learning_rate": 3.520485584218513e-06,
+      "loss": 0.5422,
+      "step": 205
+    },
+    {
+      "epoch": 0.9206703910614525,
+      "grad_norm": 0.6107847324258918,
+      "learning_rate": 3.5128983308042493e-06,
+      "loss": 0.5204,
+      "step": 206
+    },
+    {
+      "epoch": 0.9251396648044693,
+      "grad_norm": 0.49314942553653257,
+      "learning_rate": 3.505311077389985e-06,
+      "loss": 0.5074,
+      "step": 207
+    },
+    {
+      "epoch": 0.929608938547486,
+      "grad_norm": 0.5340833706807174,
+      "learning_rate": 3.497723823975721e-06,
+      "loss": 0.4936,
+      "step": 208
+    },
+    {
+      "epoch": 0.9340782122905028,
+      "grad_norm": 0.5342137283430334,
+      "learning_rate": 3.490136570561457e-06,
+      "loss": 0.4945,
+      "step": 209
+    },
+    {
+      "epoch": 0.9385474860335196,
+      "grad_norm": 0.5301769352005536,
+      "learning_rate": 3.482549317147193e-06,
+      "loss": 0.4963,
+      "step": 210
+    },
+    {
+      "epoch": 0.9430167597765363,
+      "grad_norm": 0.5455545680194068,
+      "learning_rate": 3.474962063732929e-06,
+      "loss": 0.4931,
+      "step": 211
+    },
+    {
+      "epoch": 0.9474860335195531,
+      "grad_norm": 0.5282021847913991,
+      "learning_rate": 3.467374810318665e-06,
+      "loss": 0.4894,
+      "step": 212
+    },
+    {
+      "epoch": 0.9519553072625698,
+      "grad_norm": 0.5341400870442735,
+      "learning_rate": 3.459787556904401e-06,
+      "loss": 0.5102,
+      "step": 213
+    },
+    {
+      "epoch": 0.9564245810055866,
+      "grad_norm": 0.5313633005763995,
+      "learning_rate": 3.452200303490137e-06,
+      "loss": 0.501,
+      "step": 214
+    },
+    {
+      "epoch": 0.9608938547486033,
+      "grad_norm": 0.5774555510069799,
+      "learning_rate": 3.4446130500758727e-06,
+      "loss": 0.5021,
+      "step": 215
+    },
+    {
+      "epoch": 0.9653631284916201,
+      "grad_norm": 0.521646351973494,
+      "learning_rate": 3.437025796661609e-06,
+      "loss": 0.5071,
+      "step": 216
+    },
+    {
+      "epoch": 0.9698324022346368,
+      "grad_norm": 0.5335773786656196,
+      "learning_rate": 3.429438543247345e-06,
+      "loss": 0.4993,
+      "step": 217
+    },
+    {
+      "epoch": 0.9743016759776536,
+      "grad_norm": 0.5774515212071393,
+      "learning_rate": 3.4218512898330806e-06,
+      "loss": 0.4802,
+      "step": 218
+    },
+    {
+      "epoch": 0.9787709497206704,
+      "grad_norm": 0.5159511449041201,
+      "learning_rate": 3.4142640364188168e-06,
+      "loss": 0.483,
+      "step": 219
+    },
+    {
+      "epoch": 0.9832402234636871,
+      "grad_norm": 0.509916149384416,
+      "learning_rate": 3.406676783004553e-06,
+      "loss": 0.4774,
+      "step": 220
+    },
+    {
+      "epoch": 0.9877094972067039,
+      "grad_norm": 0.5226659969898425,
+      "learning_rate": 3.3990895295902886e-06,
+      "loss": 0.516,
+      "step": 221
+    },
+    {
+      "epoch": 0.9921787709497206,
+      "grad_norm": 0.5632866199425641,
+      "learning_rate": 3.3915022761760247e-06,
+      "loss": 0.5346,
+      "step": 222
+    },
+    {
+      "epoch": 0.9966480446927374,
+      "grad_norm": 0.5224220899066443,
+      "learning_rate": 3.383915022761761e-06,
+      "loss": 0.5441,
+      "step": 223
+    },
+    {
+      "epoch": 1.0011173184357542,
+      "grad_norm": 0.5965341004746808,
+      "learning_rate": 3.3763277693474965e-06,
+      "loss": 0.512,
+      "step": 224
+    },
+    {
+      "epoch": 1.0011173184357542,
+      "eval_loss": 0.5086758732795715,
+      "eval_runtime": 32.322,
+      "eval_samples_per_second": 82.606,
+      "eval_steps_per_second": 5.167,
+      "step": 224
+    },
+    {
+      "epoch": 1.005586592178771,
+      "grad_norm": 0.5234315419190534,
+      "learning_rate": 3.3687405159332327e-06,
+      "loss": 0.5054,
+      "step": 225
+    },
+    {
+      "epoch": 1.0100558659217878,
+      "grad_norm": 0.5575750854403085,
+      "learning_rate": 3.3611532625189684e-06,
+      "loss": 0.4878,
+      "step": 226
+    },
+    {
+      "epoch": 1.0145251396648045,
+      "grad_norm": 0.5253448160628447,
+      "learning_rate": 3.3535660091047045e-06,
+      "loss": 0.5054,
+      "step": 227
+    },
+    {
+      "epoch": 1.0189944134078213,
+      "grad_norm": 0.49020025660118194,
+      "learning_rate": 3.3459787556904406e-06,
+      "loss": 0.4818,
+      "step": 228
+    },
+    {
+      "epoch": 1.0044692737430168,
+      "grad_norm": 0.6277138162928673,
+      "learning_rate": 3.3383915022761763e-06,
+      "loss": 0.5008,
+      "step": 229
+    },
+    {
+      "epoch": 1.0089385474860335,
+      "grad_norm": 0.5276575275655174,
+      "learning_rate": 3.3308042488619125e-06,
+      "loss": 0.4822,
+      "step": 230
+    },
+    {
+      "epoch": 1.0134078212290503,
+      "grad_norm": 0.516729892988387,
+      "learning_rate": 3.3232169954476486e-06,
+      "loss": 0.4783,
+      "step": 231
+    },
+    {
+      "epoch": 1.017877094972067,
+      "grad_norm": 0.5648907712973277,
+      "learning_rate": 3.3156297420333843e-06,
+      "loss": 0.4927,
+      "step": 232
+    },
+    {
+      "epoch": 1.0223463687150838,
+      "grad_norm": 0.5521221146190025,
+      "learning_rate": 3.3080424886191204e-06,
+      "loss": 0.4742,
+      "step": 233
+    },
+    {
+      "epoch": 1.0268156424581005,
+      "grad_norm": 0.5288379239842262,
+      "learning_rate": 3.3004552352048565e-06,
+      "loss": 0.4863,
+      "step": 234
+    },
+    {
+      "epoch": 1.0312849162011173,
+      "grad_norm": 0.4932863276651226,
+      "learning_rate": 3.2928679817905922e-06,
+      "loss": 0.4613,
+      "step": 235
+    },
+    {
+      "epoch": 1.035754189944134,
+      "grad_norm": 0.5069055178884726,
+      "learning_rate": 3.2852807283763284e-06,
+      "loss": 0.4969,
+      "step": 236
+    },
+    {
+      "epoch": 1.0402234636871508,
+      "grad_norm": 0.49009866664123986,
+      "learning_rate": 3.277693474962064e-06,
+      "loss": 0.4759,
+      "step": 237
+    },
+    {
+      "epoch": 1.0446927374301676,
+      "grad_norm": 0.5220649844244807,
+      "learning_rate": 3.2701062215478e-06,
+      "loss": 0.4654,
+      "step": 238
+    },
+    {
+      "epoch": 1.0491620111731843,
+      "grad_norm": 0.5354627141925222,
+      "learning_rate": 3.2625189681335363e-06,
+      "loss": 0.5009,
+      "step": 239
+    },
+    {
+      "epoch": 1.053631284916201,
+      "grad_norm": 0.5561818122893151,
+      "learning_rate": 3.254931714719272e-06,
+      "loss": 0.4564,
+      "step": 240
+    },
+    {
+      "epoch": 1.0581005586592178,
+      "grad_norm": 0.48719455414113416,
+      "learning_rate": 3.2473444613050077e-06,
+      "loss": 0.4924,
+      "step": 241
+    },
+    {
+      "epoch": 1.0625698324022346,
+      "grad_norm": 0.5527905454518657,
+      "learning_rate": 3.2397572078907434e-06,
+      "loss": 0.4856,
+      "step": 242
+    },
+    {
+      "epoch": 1.0670391061452513,
+      "grad_norm": 0.5381863443762583,
+      "learning_rate": 3.2321699544764795e-06,
+      "loss": 0.4935,
+      "step": 243
+    },
+    {
+      "epoch": 1.071508379888268,
+      "grad_norm": 0.5355360844432822,
+      "learning_rate": 3.2245827010622157e-06,
+      "loss": 0.4858,
+      "step": 244
+    },
+    {
+      "epoch": 1.0759776536312848,
+      "grad_norm": 0.5191892602172652,
+      "learning_rate": 3.2169954476479514e-06,
+      "loss": 0.4819,
+      "step": 245
+    },
+    {
+      "epoch": 1.0804469273743016,
+      "grad_norm": 0.4952672717234948,
+      "learning_rate": 3.2094081942336875e-06,
+      "loss": 0.5007,
+      "step": 246
+    },
+    {
+      "epoch": 1.0849162011173183,
+      "grad_norm": 0.5331734326280092,
+      "learning_rate": 3.201820940819423e-06,
+      "loss": 0.4874,
+      "step": 247
+    },
+    {
+      "epoch": 1.089385474860335,
+      "grad_norm": 0.49333273079420603,
+      "learning_rate": 3.1942336874051593e-06,
+      "loss": 0.5045,
+      "step": 248
+    },
+    {
+      "epoch": 1.0938547486033519,
+      "grad_norm": 0.5143326582021585,
+      "learning_rate": 3.1866464339908955e-06,
+      "loss": 0.5047,
+      "step": 249
+    },
+    {
+      "epoch": 1.0983240223463686,
+      "grad_norm": 0.5027227799706213,
+      "learning_rate": 3.179059180576631e-06,
+      "loss": 0.4803,
+      "step": 250
+    },
+    {
+      "epoch": 1.1027932960893856,
+      "grad_norm": 0.5257359502707071,
+      "learning_rate": 3.1714719271623673e-06,
+      "loss": 0.4654,
+      "step": 251
+    },
+    {
+      "epoch": 1.1072625698324021,
+      "grad_norm": 0.5244897759431714,
+      "learning_rate": 3.1638846737481034e-06,
+      "loss": 0.5329,
+      "step": 252
+    },
+    {
+      "epoch": 1.111731843575419,
+      "grad_norm": 0.5028531015372357,
+      "learning_rate": 3.156297420333839e-06,
+      "loss": 0.4721,
+      "step": 253
+    },
+    {
+      "epoch": 1.1162011173184359,
+      "grad_norm": 0.5342121642252061,
+      "learning_rate": 3.1487101669195752e-06,
+      "loss": 0.4962,
+      "step": 254
+    },
+    {
+      "epoch": 1.1206703910614526,
+      "grad_norm": 0.49550125400926287,
+      "learning_rate": 3.1411229135053114e-06,
+      "loss": 0.4884,
+      "step": 255
+    },
+    {
+      "epoch": 1.1251396648044694,
+      "grad_norm": 0.5724921757672813,
+      "learning_rate": 3.133535660091047e-06,
+      "loss": 0.4676,
+      "step": 256
+    },
+    {
+      "epoch": 1.1296089385474861,
+      "grad_norm": 0.5012184887171401,
+      "learning_rate": 3.125948406676783e-06,
+      "loss": 0.48,
+      "step": 257
+    },
+    {
+      "epoch": 1.1340782122905029,
+      "grad_norm": 0.5049054849116781,
+      "learning_rate": 3.118361153262519e-06,
+      "loss": 0.494,
+      "step": 258
+    },
+    {
+      "epoch": 1.1385474860335196,
+      "grad_norm": 0.5291980859522503,
+      "learning_rate": 3.110773899848255e-06,
+      "loss": 0.4787,
+      "step": 259
+    },
+    {
+      "epoch": 1.1430167597765364,
+      "grad_norm": 0.5235318263865767,
+      "learning_rate": 3.103186646433991e-06,
+      "loss": 0.4745,
+      "step": 260
+    },
+    {
+      "epoch": 1.1474860335195531,
+      "grad_norm": 0.5067236081675356,
+      "learning_rate": 3.095599393019727e-06,
+      "loss": 0.5863,
+      "step": 261
+    },
+    {
+      "epoch": 1.15195530726257,
+      "grad_norm": 0.48763909341918976,
+      "learning_rate": 3.088012139605463e-06,
+      "loss": 0.5264,
+      "step": 262
+    },
+    {
+      "epoch": 1.1564245810055866,
+      "grad_norm": 0.5198110315693464,
+      "learning_rate": 3.080424886191199e-06,
+      "loss": 0.4716,
+      "step": 263
+    },
+    {
+      "epoch": 1.1608938547486034,
+      "grad_norm": 0.5093655208180958,
+      "learning_rate": 3.072837632776935e-06,
+      "loss": 0.4768,
+      "step": 264
+    },
+    {
+      "epoch": 1.1653631284916202,
+      "grad_norm": 0.49489186143964614,
+      "learning_rate": 3.065250379362671e-06,
+      "loss": 0.4765,
+      "step": 265
+    },
+    {
+      "epoch": 1.169832402234637,
+      "grad_norm": 0.5209973994944476,
+      "learning_rate": 3.057663125948407e-06,
+      "loss": 0.4869,
+      "step": 266
+    },
+    {
+      "epoch": 1.1743016759776537,
+      "grad_norm": 0.4873878645874658,
+      "learning_rate": 3.0500758725341427e-06,
+      "loss": 0.4623,
+      "step": 267
+    },
+    {
+      "epoch": 1.1787709497206704,
+      "grad_norm": 0.5074910158328607,
+      "learning_rate": 3.042488619119879e-06,
+      "loss": 0.4765,
+      "step": 268
+    },
+    {
+      "epoch": 1.1832402234636872,
+      "grad_norm": 0.5195087180758534,
+      "learning_rate": 3.0349013657056146e-06,
+      "loss": 0.4838,
+      "step": 269
+    },
+    {
+      "epoch": 1.187709497206704,
+      "grad_norm": 0.5264228931092361,
+      "learning_rate": 3.0273141122913507e-06,
+      "loss": 0.4638,
+      "step": 270
+    },
+    {
+      "epoch": 1.1921787709497207,
+      "grad_norm": 0.4919380409463276,
+      "learning_rate": 3.019726858877087e-06,
+      "loss": 0.4981,
+      "step": 271
+    },
+    {
+      "epoch": 1.1966480446927374,
+      "grad_norm": 0.5060903656555795,
+      "learning_rate": 3.0121396054628225e-06,
+      "loss": 0.477,
+      "step": 272
+    },
+    {
+      "epoch": 1.2011173184357542,
+      "grad_norm": 0.4938697284936987,
+      "learning_rate": 3.0045523520485587e-06,
+      "loss": 0.5003,
+      "step": 273
+    },
+    {
+      "epoch": 1.205586592178771,
+      "grad_norm": 0.5599516302969035,
+      "learning_rate": 2.9969650986342948e-06,
+      "loss": 0.4673,
+      "step": 274
+    },
+    {
+      "epoch": 1.2100558659217877,
+      "grad_norm": 0.5126307823030745,
+      "learning_rate": 2.9893778452200305e-06,
+      "loss": 0.4971,
+      "step": 275
+    },
+    {
+      "epoch": 1.2145251396648045,
+      "grad_norm": 0.5927333276589611,
+      "learning_rate": 2.9817905918057666e-06,
+      "loss": 0.5006,
+      "step": 276
+    },
+    {
+      "epoch": 1.2189944134078212,
+      "grad_norm": 0.5367579038483986,
+      "learning_rate": 2.9742033383915027e-06,
+      "loss": 0.5143,
+      "step": 277
+    },
+    {
+      "epoch": 1.223463687150838,
+      "grad_norm": 0.5179699664039845,
+      "learning_rate": 2.9666160849772384e-06,
+      "loss": 0.4733,
+      "step": 278
+    },
+    {
+      "epoch": 1.2279329608938547,
+      "grad_norm": 0.5015385250607317,
+      "learning_rate": 2.9590288315629746e-06,
+      "loss": 0.4786,
+      "step": 279
+    },
+    {
+      "epoch": 1.2324022346368715,
+      "grad_norm": 0.4925764923779409,
+      "learning_rate": 2.9514415781487103e-06,
+      "loss": 0.472,
+      "step": 280
+    },
+    {
+      "epoch": 1.2368715083798882,
+      "grad_norm": 0.5522056674324646,
+      "learning_rate": 2.9438543247344464e-06,
+      "loss": 0.4675,
+      "step": 281
+    },
+    {
+      "epoch": 1.241340782122905,
+      "grad_norm": 0.5128909667557014,
+      "learning_rate": 2.9362670713201825e-06,
+      "loss": 0.4778,
+      "step": 282
+    },
+    {
+      "epoch": 1.2458100558659218,
+      "grad_norm": 0.49958051066954534,
+      "learning_rate": 2.9286798179059182e-06,
+      "loss": 0.4744,
+      "step": 283
+    },
+    {
+      "epoch": 1.2502793296089385,
+      "grad_norm": 0.5186801983796817,
+      "learning_rate": 2.9210925644916543e-06,
+      "loss": 0.47,
+      "step": 284
+    },
+    {
+      "epoch": 1.2547486033519553,
+      "grad_norm": 0.49479081933964797,
+      "learning_rate": 2.9135053110773905e-06,
+      "loss": 0.4693,
+      "step": 285
+    },
+    {
+      "epoch": 1.259217877094972,
+      "grad_norm": 0.5172924586559385,
+      "learning_rate": 2.905918057663126e-06,
+      "loss": 0.5034,
+      "step": 286
+    },
+    {
+      "epoch": 1.2636871508379888,
+      "grad_norm": 0.49588955761543196,
+      "learning_rate": 2.8983308042488623e-06,
+      "loss": 0.4857,
+      "step": 287
+    },
+    {
+      "epoch": 1.2681564245810055,
+      "grad_norm": 0.5651670438316321,
+      "learning_rate": 2.8907435508345984e-06,
+      "loss": 0.4845,
+      "step": 288
+    },
+    {
+      "epoch": 1.2726256983240223,
+      "grad_norm": 0.47720259500943096,
+      "learning_rate": 2.883156297420334e-06,
+      "loss": 0.4482,
+      "step": 289
+    },
+    {
+      "epoch": 1.277094972067039,
+      "grad_norm": 0.5080092332022862,
+      "learning_rate": 2.8755690440060702e-06,
+      "loss": 0.4826,
+      "step": 290
+    },
+    {
+      "epoch": 1.2815642458100558,
+      "grad_norm": 0.6016729996144168,
+      "learning_rate": 2.867981790591806e-06,
+      "loss": 0.5592,
+      "step": 291
+    },
+    {
+      "epoch": 1.2860335195530725,
+      "grad_norm": 0.5456040494522852,
+      "learning_rate": 2.860394537177542e-06,
+      "loss": 0.4821,
+      "step": 292
+    },
+    {
+      "epoch": 1.2905027932960893,
+      "grad_norm": 0.5858331665900632,
+      "learning_rate": 2.852807283763278e-06,
+      "loss": 0.5089,
+      "step": 293
+    },
+    {
+      "epoch": 1.2949720670391063,
+      "grad_norm": 0.5895630636847224,
+      "learning_rate": 2.845220030349014e-06,
+      "loss": 0.5101,
+      "step": 294
+    },
+    {
+      "epoch": 1.2994413407821228,
+      "grad_norm": 0.505172328049275,
+      "learning_rate": 2.83763277693475e-06,
+      "loss": 0.5008,
+      "step": 295
+    },
+    {
+      "epoch": 1.3039106145251398,
+      "grad_norm": 0.5035894899292893,
+      "learning_rate": 2.830045523520486e-06,
+      "loss": 0.4812,
+      "step": 296
+    },
+    {
+      "epoch": 1.3083798882681563,
+      "grad_norm": 0.5304264887862729,
+      "learning_rate": 2.822458270106222e-06,
+      "loss": 0.4579,
+      "step": 297
+    },
+    {
+      "epoch": 1.3128491620111733,
+      "grad_norm": 0.4742288658363735,
+      "learning_rate": 2.814871016691958e-06,
+      "loss": 0.4801,
+      "step": 298
+    },
+    {
+      "epoch": 1.3173184357541898,
+      "grad_norm": 0.5293412371049411,
+      "learning_rate": 2.807283763277694e-06,
+      "loss": 0.4602,
+      "step": 299
+    },
+    {
+      "epoch": 1.3217877094972068,
+      "grad_norm": 0.5333831439310788,
+      "learning_rate": 2.79969650986343e-06,
+      "loss": 0.4963,
+      "step": 300
+    },
+    {
+      "epoch": 1.3262569832402233,
+      "grad_norm": 0.4947844594727205,
+      "learning_rate": 2.792109256449166e-06,
+      "loss": 0.4781,
+      "step": 301
+    },
+    {
+      "epoch": 1.3307262569832403,
+      "grad_norm": 0.5323122787674339,
+      "learning_rate": 2.7845220030349016e-06,
+      "loss": 0.4998,
+      "step": 302
+    },
+    {
+      "epoch": 1.3351955307262569,
+      "grad_norm": 0.48615316005162806,
+      "learning_rate": 2.7769347496206378e-06,
+      "loss": 0.466,
+      "step": 303
+    },
+    {
+      "epoch": 1.3396648044692738,
+      "grad_norm": 0.5038897355673981,
+      "learning_rate": 2.769347496206374e-06,
+      "loss": 0.531,
+      "step": 304
+    },
+    {
+      "epoch": 1.3441340782122906,
+      "grad_norm": 0.47725378544672703,
+      "learning_rate": 2.7617602427921096e-06,
+      "loss": 0.4917,
+      "step": 305
+    },
+    {
+      "epoch": 1.3486033519553073,
+      "grad_norm": 0.5241579403675019,
+      "learning_rate": 2.7541729893778457e-06,
+      "loss": 0.4673,
+      "step": 306
+    },
+    {
+      "epoch": 1.353072625698324,
+      "grad_norm": 0.5252091486066182,
+      "learning_rate": 2.746585735963581e-06,
+      "loss": 0.467,
+      "step": 307
+    },
+    {
+      "epoch": 1.3575418994413408,
+      "grad_norm": 0.4803361684555166,
+      "learning_rate": 2.738998482549317e-06,
+      "loss": 0.5257,
+      "step": 308
+    },
+    {
+      "epoch": 1.3620111731843576,
+      "grad_norm": 0.5300934772683465,
+      "learning_rate": 2.7314112291350532e-06,
+      "loss": 0.4889,
+      "step": 309
+    },
+    {
+      "epoch": 1.3664804469273744,
+      "grad_norm": 0.5001374114396476,
+      "learning_rate": 2.723823975720789e-06,
+      "loss": 0.4847,
+      "step": 310
+    },
+    {
+      "epoch": 1.3709497206703911,
+      "grad_norm": 0.5240875065710657,
+      "learning_rate": 2.716236722306525e-06,
+      "loss": 0.4578,
+      "step": 311
+    },
+    {
+      "epoch": 1.3754189944134079,
+      "grad_norm": 0.4714761354385937,
+      "learning_rate": 2.708649468892261e-06,
+      "loss": 0.4947,
+      "step": 312
+    },
+    {
+      "epoch": 1.3798882681564246,
+      "grad_norm": 0.5205222669608968,
+      "learning_rate": 2.701062215477997e-06,
+      "loss": 0.5049,
+      "step": 313
+    },
+    {
+      "epoch": 1.3843575418994414,
+      "grad_norm": 0.5409150510364881,
+      "learning_rate": 2.693474962063733e-06,
+      "loss": 0.4946,
+      "step": 314
+    },
+    {
+      "epoch": 1.3888268156424581,
+      "grad_norm": 0.49473773833387136,
+      "learning_rate": 2.6858877086494687e-06,
+      "loss": 0.5024,
+      "step": 315
+    },
+    {
+      "epoch": 1.393296089385475,
+      "grad_norm": 0.5322303350412809,
+      "learning_rate": 2.678300455235205e-06,
+      "loss": 0.5129,
+      "step": 316
+    },
+    {
+      "epoch": 1.3977653631284916,
+      "grad_norm": 0.5249345159944246,
+      "learning_rate": 2.670713201820941e-06,
+      "loss": 0.4852,
+      "step": 317
+    },
+    {
+      "epoch": 1.4022346368715084,
+      "grad_norm": 0.526125738227021,
+      "learning_rate": 2.6631259484066767e-06,
+      "loss": 0.4755,
+      "step": 318
+    },
+    {
+      "epoch": 1.4067039106145252,
+      "grad_norm": 0.5077327993266544,
+      "learning_rate": 2.655538694992413e-06,
+      "loss": 0.4708,
+      "step": 319
+    },
+    {
+      "epoch": 1.411173184357542,
+      "grad_norm": 0.509006777595202,
+      "learning_rate": 2.647951441578149e-06,
+      "loss": 0.4574,
+      "step": 320
+    },
+    {
+      "epoch": 1.4156424581005587,
+      "grad_norm": 0.5397063984545032,
+      "learning_rate": 2.6403641881638846e-06,
+      "loss": 0.4983,
+      "step": 321
+    },
+    {
+      "epoch": 1.4201117318435754,
+      "grad_norm": 0.588039527577832,
+      "learning_rate": 2.6327769347496208e-06,
+      "loss": 0.5063,
+      "step": 322
+    },
+    {
+      "epoch": 1.4245810055865922,
+      "grad_norm": 0.5086942941008908,
+      "learning_rate": 2.625189681335357e-06,
+      "loss": 0.5439,
+      "step": 323
+    },
+    {
+      "epoch": 1.429050279329609,
+      "grad_norm": 0.5593378373560197,
+      "learning_rate": 2.6176024279210926e-06,
+      "loss": 0.4789,
+      "step": 324
+    },
+    {
+      "epoch": 1.4335195530726257,
+      "grad_norm": 0.5473461722884195,
+      "learning_rate": 2.6100151745068287e-06,
+      "loss": 0.4861,
+      "step": 325
+    },
+    {
+      "epoch": 1.4379888268156424,
+      "grad_norm": 0.5744717523131797,
+      "learning_rate": 2.6024279210925644e-06,
+      "loss": 0.4818,
+      "step": 326
+    },
+    {
+      "epoch": 1.4424581005586592,
+      "grad_norm": 0.5240335872273282,
+      "learning_rate": 2.5948406676783005e-06,
+      "loss": 0.4785,
+      "step": 327
+    },
+    {
+      "epoch": 1.446927374301676,
+      "grad_norm": 0.5608475565499077,
+      "learning_rate": 2.5872534142640367e-06,
+      "loss": 0.5042,
+      "step": 328
+    },
+    {
+      "epoch": 1.4513966480446927,
+      "grad_norm": 0.5363050759442354,
+      "learning_rate": 2.5796661608497724e-06,
+      "loss": 0.4995,
+      "step": 329
+    },
+    {
+      "epoch": 1.4558659217877095,
+      "grad_norm": 0.5027607995293213,
+      "learning_rate": 2.5720789074355085e-06,
+      "loss": 0.4954,
+      "step": 330
+    },
+    {
+      "epoch": 1.4603351955307262,
+      "grad_norm": 0.4969954043343912,
+      "learning_rate": 2.5644916540212446e-06,
+      "loss": 0.4882,
+      "step": 331
+    },
+    {
+      "epoch": 1.464804469273743,
+      "grad_norm": 0.5718337637259987,
+      "learning_rate": 2.5569044006069803e-06,
+      "loss": 0.4726,
+      "step": 332
+    },
+    {
+      "epoch": 1.4692737430167597,
+      "grad_norm": 0.5234285673769393,
+      "learning_rate": 2.5493171471927164e-06,
+      "loss": 0.4697,
+      "step": 333
+    },
+    {
+      "epoch": 1.4737430167597765,
+      "grad_norm": 0.5219608223935474,
+      "learning_rate": 2.5417298937784526e-06,
+      "loss": 0.4776,
+      "step": 334
+    },
+    {
+      "epoch": 1.4782122905027932,
+      "grad_norm": 0.606768683027719,
+      "learning_rate": 2.5341426403641883e-06,
+      "loss": 0.4756,
+      "step": 335
+    },
+    {
+      "epoch": 1.48268156424581,
+      "grad_norm": 0.5432176614167162,
+      "learning_rate": 2.5265553869499244e-06,
+      "loss": 0.4619,
+      "step": 336
+    },
+    {
+      "epoch": 1.48268156424581,
+      "eval_loss": 0.5032872557640076,
+      "eval_runtime": 32.2653,
+      "eval_samples_per_second": 82.751,
+      "eval_steps_per_second": 5.176,
+      "step": 336
+    },
+    {
+      "epoch": 1.4871508379888267,
+      "grad_norm": 0.5377224565720307,
+      "learning_rate": 2.51896813353566e-06,
+      "loss": 0.4667,
+      "step": 337
+    },
+    {
+      "epoch": 1.4916201117318435,
+      "grad_norm": 0.5875842821019925,
+      "learning_rate": 2.5113808801213962e-06,
+      "loss": 0.4873,
+      "step": 338
+    },
+    {
+      "epoch": 1.4960893854748603,
+      "grad_norm": 0.5624618850955563,
+      "learning_rate": 2.5037936267071324e-06,
+      "loss": 0.4626,
+      "step": 339
+    },
+    {
+      "epoch": 1.500558659217877,
+      "grad_norm": 0.5380665704623208,
+      "learning_rate": 2.496206373292868e-06,
+      "loss": 0.4749,
+      "step": 340
+    },
+    {
+      "epoch": 1.505027932960894,
+      "grad_norm": 0.5515048280645092,
+      "learning_rate": 2.488619119878604e-06,
+      "loss": 0.4862,
+      "step": 341
+    },
+    {
+      "epoch": 1.5094972067039105,
+      "grad_norm": 0.5541495309875792,
+      "learning_rate": 2.4810318664643403e-06,
+      "loss": 0.4533,
+      "step": 342
+    },
+    {
+      "epoch": 1.5139664804469275,
+      "grad_norm": 0.514569610923219,
+      "learning_rate": 2.473444613050076e-06,
+      "loss": 0.4585,
+      "step": 343
+    },
+    {
+      "epoch": 1.518435754189944,
+      "grad_norm": 0.5862261680615611,
+      "learning_rate": 2.465857359635812e-06,
+      "loss": 0.4835,
+      "step": 344
+    },
+    {
+      "epoch": 1.522905027932961,
+      "grad_norm": 0.536975433650959,
+      "learning_rate": 2.458270106221548e-06,
+      "loss": 0.4638,
+      "step": 345
+    },
+    {
+      "epoch": 1.5273743016759775,
+      "grad_norm": 0.5142279881627296,
+      "learning_rate": 2.450682852807284e-06,
+      "loss": 0.4832,
+      "step": 346
+    },
+    {
+      "epoch": 1.5318435754189945,
+      "grad_norm": 0.5124062345457664,
+      "learning_rate": 2.44309559939302e-06,
+      "loss": 0.4857,
+      "step": 347
+    },
+    {
+      "epoch": 1.536312849162011,
+      "grad_norm": 0.509905421315166,
+      "learning_rate": 2.435508345978756e-06,
+      "loss": 0.4807,
+      "step": 348
+    },
+    {
+      "epoch": 1.540782122905028,
+      "grad_norm": 0.5275507665643101,
+      "learning_rate": 2.427921092564492e-06,
+      "loss": 0.4722,
+      "step": 349
+    },
+    {
+      "epoch": 1.5452513966480446,
+      "grad_norm": 0.5003951978639241,
+      "learning_rate": 2.420333839150228e-06,
+      "loss": 0.4941,
+      "step": 350
+    },
+    {
+      "epoch": 1.5497206703910615,
+      "grad_norm": 0.5251614039487292,
+      "learning_rate": 2.4127465857359637e-06,
+      "loss": 0.5376,
+      "step": 351
+    },
+    {
+      "epoch": 1.554189944134078,
+      "grad_norm": 0.5128097582851606,
+      "learning_rate": 2.4051593323217e-06,
+      "loss": 0.5001,
+      "step": 352
+    },
+    {
+      "epoch": 1.558659217877095,
+      "grad_norm": 0.4971288144993544,
+      "learning_rate": 2.397572078907436e-06,
+      "loss": 0.4721,
+      "step": 353
+    },
+    {
+      "epoch": 1.5631284916201116,
+      "grad_norm": 0.5111491573396423,
+      "learning_rate": 2.3899848254931717e-06,
+      "loss": 0.5192,
+      "step": 354
+    },
+    {
+      "epoch": 1.5675977653631286,
+      "grad_norm": 0.5025703950184777,
+      "learning_rate": 2.382397572078908e-06,
+      "loss": 0.4541,
+      "step": 355
+    },
+    {
+      "epoch": 1.572067039106145,
+      "grad_norm": 0.5011840454618501,
+      "learning_rate": 2.3748103186646435e-06,
+      "loss": 0.4789,
+      "step": 356
+    },
+    {
+      "epoch": 1.576536312849162,
+      "grad_norm": 0.4981690031939279,
+      "learning_rate": 2.3672230652503792e-06,
+      "loss": 0.497,
+      "step": 357
+    },
+    {
+      "epoch": 1.5810055865921788,
+      "grad_norm": 0.4994506284814992,
+      "learning_rate": 2.3596358118361154e-06,
+      "loss": 0.4655,
+      "step": 358
+    },
+    {
+      "epoch": 1.5854748603351956,
+      "grad_norm": 0.49694019288140645,
+      "learning_rate": 2.3520485584218515e-06,
+      "loss": 0.4883,
+      "step": 359
+    },
+    {
+      "epoch": 1.5899441340782123,
+      "grad_norm": 0.5103222760000443,
+      "learning_rate": 2.344461305007587e-06,
+      "loss": 0.4615,
+      "step": 360
+    },
+    {
+      "epoch": 1.594413407821229,
+      "grad_norm": 0.4960772434204748,
+      "learning_rate": 2.3368740515933233e-06,
+      "loss": 0.4748,
+      "step": 361
+    },
+    {
+      "epoch": 1.5988826815642458,
+      "grad_norm": 0.5372239368130248,
+      "learning_rate": 2.3292867981790594e-06,
+      "loss": 0.461,
+      "step": 362
+    },
+    {
+      "epoch": 1.6033519553072626,
+      "grad_norm": 0.5002183877541576,
+      "learning_rate": 2.321699544764795e-06,
+      "loss": 0.4985,
+      "step": 363
+    },
+    {
+      "epoch": 1.6078212290502794,
+      "grad_norm": 0.5065612982842012,
+      "learning_rate": 2.3141122913505313e-06,
+      "loss": 0.4846,
+      "step": 364
+    },
+    {
+      "epoch": 1.612290502793296,
+      "grad_norm": 0.5294145045424744,
+      "learning_rate": 2.3065250379362674e-06,
+      "loss": 0.4769,
+      "step": 365
+    },
+    {
+      "epoch": 1.6167597765363129,
+      "grad_norm": 0.5019842435412675,
+      "learning_rate": 2.298937784522003e-06,
+      "loss": 0.4738,
+      "step": 366
+    },
+    {
+      "epoch": 1.6212290502793296,
+      "grad_norm": 0.4973490272845337,
+      "learning_rate": 2.291350531107739e-06,
+      "loss": 0.4773,
+      "step": 367
+    },
+    {
+      "epoch": 1.6256983240223464,
+      "grad_norm": 0.4917027064554639,
+      "learning_rate": 2.283763277693475e-06,
+      "loss": 0.4921,
+      "step": 368
+    },
+    {
+      "epoch": 1.6301675977653631,
+      "grad_norm": 0.5191932977137537,
+      "learning_rate": 2.276176024279211e-06,
+      "loss": 0.4823,
+      "step": 369
+    },
+    {
+      "epoch": 1.6346368715083799,
+      "grad_norm": 0.5130853619632637,
+      "learning_rate": 2.268588770864947e-06,
+      "loss": 0.4631,
+      "step": 370
+    },
+    {
+      "epoch": 1.6391061452513966,
+      "grad_norm": 0.5402133951776769,
+      "learning_rate": 2.261001517450683e-06,
+      "loss": 0.456,
+      "step": 371
+    },
+    {
+      "epoch": 1.6435754189944134,
+      "grad_norm": 0.5136201214507364,
+      "learning_rate": 2.253414264036419e-06,
+      "loss": 0.4715,
+      "step": 372
+    },
+    {
+      "epoch": 1.6480446927374302,
+      "grad_norm": 0.5397223114168315,
+      "learning_rate": 2.245827010622155e-06,
+      "loss": 0.465,
+      "step": 373
+    },
+    {
+      "epoch": 1.652513966480447,
+      "grad_norm": 0.5178224930093817,
+      "learning_rate": 2.238239757207891e-06,
+      "loss": 0.4526,
+      "step": 374
+    },
+    {
+      "epoch": 1.6569832402234637,
+      "grad_norm": 0.5399618966600667,
+      "learning_rate": 2.230652503793627e-06,
+      "loss": 0.4873,
+      "step": 375
+    },
+    {
+      "epoch": 1.6614525139664804,
+      "grad_norm": 0.5106529140250409,
+      "learning_rate": 2.223065250379363e-06,
+      "loss": 0.4771,
+      "step": 376
+    },
+    {
+      "epoch": 1.6659217877094972,
+      "grad_norm": 0.5460841160518792,
+      "learning_rate": 2.2154779969650988e-06,
+      "loss": 0.4743,
+      "step": 377
+    },
+    {
+      "epoch": 1.670391061452514,
+      "grad_norm": 0.5846706720135237,
+      "learning_rate": 2.207890743550835e-06,
+      "loss": 0.4552,
+      "step": 378
+    },
+    {
+      "epoch": 1.6748603351955307,
+      "grad_norm": 0.534699591229056,
+      "learning_rate": 2.2003034901365706e-06,
+      "loss": 0.495,
+      "step": 379
+    },
+    {
+      "epoch": 1.6793296089385474,
+      "grad_norm": 0.5377842578832556,
+      "learning_rate": 2.1927162367223067e-06,
+      "loss": 0.4824,
+      "step": 380
+    },
+    {
+      "epoch": 1.6837988826815642,
+      "grad_norm": 0.5799148072307362,
+      "learning_rate": 2.185128983308043e-06,
+      "loss": 0.4708,
+      "step": 381
+    },
+    {
+      "epoch": 1.6882681564245812,
+      "grad_norm": 0.5076181067753225,
+      "learning_rate": 2.1775417298937786e-06,
+      "loss": 0.4631,
+      "step": 382
+    },
+    {
+      "epoch": 1.6927374301675977,
+      "grad_norm": 0.5025103762077108,
+      "learning_rate": 2.1699544764795147e-06,
+      "loss": 0.4676,
+      "step": 383
+    },
+    {
+      "epoch": 1.6972067039106147,
+      "grad_norm": 0.48231447315943815,
+      "learning_rate": 2.162367223065251e-06,
+      "loss": 0.4694,
+      "step": 384
+    },
+    {
+      "epoch": 1.7016759776536312,
+      "grad_norm": 0.5003437712619672,
+      "learning_rate": 2.1547799696509865e-06,
+      "loss": 0.4607,
+      "step": 385
+    },
+    {
+      "epoch": 1.7061452513966482,
+      "grad_norm": 0.5149834201342615,
+      "learning_rate": 2.1471927162367226e-06,
+      "loss": 0.5253,
+      "step": 386
+    },
+    {
+      "epoch": 1.7106145251396647,
+      "grad_norm": 0.5406857228194807,
+      "learning_rate": 2.1396054628224588e-06,
+      "loss": 0.5352,
+      "step": 387
+    },
+    {
+      "epoch": 1.7150837988826817,
+      "grad_norm": 0.5204717082880796,
+      "learning_rate": 2.1320182094081945e-06,
+      "loss": 0.4936,
+      "step": 388
+    },
+    {
+      "epoch": 1.7195530726256982,
+      "grad_norm": 0.5202749448515015,
+      "learning_rate": 2.12443095599393e-06,
+      "loss": 0.4594,
+      "step": 389
+    },
+    {
+      "epoch": 1.7240223463687152,
+      "grad_norm": 0.4952910050954741,
+      "learning_rate": 2.1168437025796663e-06,
+      "loss": 0.4793,
+      "step": 390
+    },
+    {
+      "epoch": 1.7284916201117317,
+      "grad_norm": 0.48637344511275765,
+      "learning_rate": 2.109256449165402e-06,
+      "loss": 0.4745,
+      "step": 391
+    },
+    {
+      "epoch": 1.7329608938547487,
+      "grad_norm": 0.4906271558417189,
+      "learning_rate": 2.101669195751138e-06,
+      "loss": 0.4652,
+      "step": 392
+    },
+    {
+      "epoch": 1.7374301675977653,
+      "grad_norm": 0.5090727761430808,
+      "learning_rate": 2.0940819423368742e-06,
+      "loss": 0.5298,
+      "step": 393
+    },
+    {
+      "epoch": 1.7418994413407822,
+      "grad_norm": 0.4866596430096002,
+      "learning_rate": 2.08649468892261e-06,
+      "loss": 0.4838,
+      "step": 394
+    },
+    {
+      "epoch": 1.7463687150837988,
+      "grad_norm": 0.49259128094078225,
+      "learning_rate": 2.078907435508346e-06,
+      "loss": 0.4737,
+      "step": 395
+    },
+    {
+      "epoch": 1.7508379888268157,
+      "grad_norm": 0.5031034927365164,
+      "learning_rate": 2.071320182094082e-06,
+      "loss": 0.4777,
+      "step": 396
+    },
+    {
+      "epoch": 1.7553072625698323,
+      "grad_norm": 0.48520021809068,
+      "learning_rate": 2.063732928679818e-06,
+      "loss": 0.454,
+      "step": 397
+    },
+    {
+      "epoch": 1.7597765363128492,
+      "grad_norm": 0.5217902917498033,
+      "learning_rate": 2.056145675265554e-06,
+      "loss": 0.5199,
+      "step": 398
+    },
+    {
+      "epoch": 1.7642458100558658,
+      "grad_norm": 0.4963706354662438,
+      "learning_rate": 2.04855842185129e-06,
+      "loss": 0.4649,
+      "step": 399
+    },
+    {
+      "epoch": 1.7687150837988828,
+      "grad_norm": 0.49552580249683814,
+      "learning_rate": 2.040971168437026e-06,
+      "loss": 0.4887,
+      "step": 400
+    },
+    {
+      "epoch": 1.7731843575418993,
+      "grad_norm": 0.49468231483436587,
+      "learning_rate": 2.033383915022762e-06,
+      "loss": 0.4757,
+      "step": 401
+    },
+    {
+      "epoch": 1.7776536312849163,
+      "grad_norm": 0.47502769186897137,
+      "learning_rate": 2.0257966616084977e-06,
+      "loss": 0.4846,
+      "step": 402
+    },
+    {
+      "epoch": 1.7821229050279328,
+      "grad_norm": 0.5089134108324753,
+      "learning_rate": 2.018209408194234e-06,
+      "loss": 0.4692,
+      "step": 403
+    },
+    {
+      "epoch": 1.7865921787709498,
+      "grad_norm": 0.4981731250179096,
+      "learning_rate": 2.01062215477997e-06,
+      "loss": 0.4665,
+      "step": 404
+    },
+    {
+      "epoch": 1.7910614525139665,
+      "grad_norm": 0.4690174096990647,
+      "learning_rate": 2.0030349013657056e-06,
+      "loss": 0.4606,
+      "step": 405
+    },
+    {
+      "epoch": 1.7955307262569833,
+      "grad_norm": 0.4914411018326139,
+      "learning_rate": 1.9954476479514418e-06,
+      "loss": 0.4847,
+      "step": 406
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 0.5513335287615446,
+      "learning_rate": 1.987860394537178e-06,
+      "loss": 0.488,
+      "step": 407
+    },
+    {
+      "epoch": 1.8044692737430168,
+      "grad_norm": 0.48992847389625854,
+      "learning_rate": 1.9802731411229136e-06,
+      "loss": 0.465,
+      "step": 408
+    },
+    {
+      "epoch": 1.8089385474860336,
+      "grad_norm": 0.5185912675971102,
+      "learning_rate": 1.9726858877086497e-06,
+      "loss": 0.4872,
+      "step": 409
+    },
+    {
+      "epoch": 1.8134078212290503,
+      "grad_norm": 0.47300994432774746,
+      "learning_rate": 1.965098634294386e-06,
+      "loss": 0.4534,
+      "step": 410
+    },
+    {
+      "epoch": 1.817877094972067,
+      "grad_norm": 0.5034275716173061,
+      "learning_rate": 1.9575113808801215e-06,
+      "loss": 0.4668,
+      "step": 411
+    },
+    {
+      "epoch": 1.8223463687150838,
+      "grad_norm": 0.46435516708821234,
+      "learning_rate": 1.9499241274658577e-06,
+      "loss": 0.491,
+      "step": 412
+    },
+    {
+      "epoch": 1.8268156424581006,
+      "grad_norm": 0.4915994950607221,
+      "learning_rate": 1.9423368740515934e-06,
+      "loss": 0.4793,
+      "step": 413
+    },
+    {
+      "epoch": 1.8312849162011173,
+      "grad_norm": 0.49834550842622505,
+      "learning_rate": 1.9347496206373295e-06,
+      "loss": 0.4897,
+      "step": 414
+    },
+    {
+      "epoch": 1.835754189944134,
+      "grad_norm": 0.5275123854480922,
+      "learning_rate": 1.9271623672230656e-06,
+      "loss": 0.5167,
+      "step": 415
+    },
+    {
+      "epoch": 1.8402234636871508,
+      "grad_norm": 0.501447167095016,
+      "learning_rate": 1.9195751138088013e-06,
+      "loss": 0.5522,
+      "step": 416
+    },
+    {
+      "epoch": 1.8446927374301676,
+      "grad_norm": 0.4929358533949542,
+      "learning_rate": 1.9119878603945374e-06,
+      "loss": 0.464,
+      "step": 417
+    },
+    {
+      "epoch": 1.8491620111731844,
+      "grad_norm": 0.4753893398726578,
+      "learning_rate": 1.9044006069802734e-06,
+      "loss": 0.4707,
+      "step": 418
+    },
+    {
+      "epoch": 1.853631284916201,
+      "grad_norm": 0.49249135995729143,
+      "learning_rate": 1.8968133535660093e-06,
+      "loss": 0.4674,
+      "step": 419
+    },
+    {
+      "epoch": 1.8581005586592179,
+      "grad_norm": 0.4959065755228421,
+      "learning_rate": 1.8892261001517454e-06,
+      "loss": 0.4731,
+      "step": 420
+    },
+    {
+      "epoch": 1.8625698324022346,
+      "grad_norm": 0.4830053823448673,
+      "learning_rate": 1.8816388467374813e-06,
+      "loss": 0.4688,
+      "step": 421
+    },
+    {
+      "epoch": 1.8670391061452514,
+      "grad_norm": 0.5026774544021604,
+      "learning_rate": 1.874051593323217e-06,
+      "loss": 0.4777,
+      "step": 422
+    },
+    {
+      "epoch": 1.8715083798882681,
+      "grad_norm": 0.48376669794821286,
+      "learning_rate": 1.866464339908953e-06,
+      "loss": 0.4532,
+      "step": 423
+    },
+    {
+      "epoch": 1.8759776536312849,
+      "grad_norm": 0.5036910075540445,
+      "learning_rate": 1.858877086494689e-06,
+      "loss": 0.4749,
+      "step": 424
+    },
+    {
+      "epoch": 1.8804469273743016,
+      "grad_norm": 0.464687327497754,
+      "learning_rate": 1.851289833080425e-06,
+      "loss": 0.4582,
+      "step": 425
+    },
+    {
+      "epoch": 1.8849162011173184,
+      "grad_norm": 0.4713537790626329,
+      "learning_rate": 1.8437025796661609e-06,
+      "loss": 0.4954,
+      "step": 426
+    },
+    {
+      "epoch": 1.8893854748603351,
+      "grad_norm": 0.4964027557908742,
+      "learning_rate": 1.8361153262518968e-06,
+      "loss": 0.4777,
+      "step": 427
+    },
+    {
+      "epoch": 1.893854748603352,
+      "grad_norm": 0.5022160749373462,
+      "learning_rate": 1.828528072837633e-06,
+      "loss": 0.4723,
+      "step": 428
+    },
+    {
+      "epoch": 1.8983240223463689,
+      "grad_norm": 0.4971783807561644,
+      "learning_rate": 1.8209408194233688e-06,
+      "loss": 0.4619,
+      "step": 429
+    },
+    {
+      "epoch": 1.9027932960893854,
+      "grad_norm": 0.5020482529233811,
+      "learning_rate": 1.8133535660091047e-06,
+      "loss": 0.4923,
+      "step": 430
+    },
+    {
+      "epoch": 1.9072625698324024,
+      "grad_norm": 0.5027930539648167,
+      "learning_rate": 1.8057663125948407e-06,
+      "loss": 0.4936,
+      "step": 431
+    },
+    {
+      "epoch": 1.911731843575419,
+      "grad_norm": 0.5048109425957825,
+      "learning_rate": 1.7981790591805768e-06,
+      "loss": 0.4647,
+      "step": 432
+    },
+    {
+      "epoch": 1.916201117318436,
+      "grad_norm": 0.5152625504278344,
+      "learning_rate": 1.7905918057663127e-06,
+      "loss": 0.4962,
+      "step": 433
+    },
+    {
+      "epoch": 1.9206703910614524,
+      "grad_norm": 0.5486472960937676,
+      "learning_rate": 1.7830045523520486e-06,
+      "loss": 0.4643,
+      "step": 434
+    },
+    {
+      "epoch": 1.9251396648044694,
+      "grad_norm": 0.5074159325604022,
+      "learning_rate": 1.7754172989377847e-06,
+      "loss": 0.4838,
+      "step": 435
+    },
+    {
+      "epoch": 1.929608938547486,
+      "grad_norm": 0.487237277239409,
+      "learning_rate": 1.7678300455235207e-06,
+      "loss": 0.4953,
+      "step": 436
+    },
+    {
+      "epoch": 1.934078212290503,
+      "grad_norm": 0.5030243156211434,
+      "learning_rate": 1.7602427921092566e-06,
+      "loss": 0.4669,
+      "step": 437
+    },
+    {
+      "epoch": 1.9385474860335195,
+      "grad_norm": 0.54024162061195,
+      "learning_rate": 1.7526555386949925e-06,
+      "loss": 0.4645,
+      "step": 438
+    },
+    {
+      "epoch": 1.9430167597765364,
+      "grad_norm": 0.4924548318787792,
+      "learning_rate": 1.7450682852807286e-06,
+      "loss": 0.487,
+      "step": 439
+    },
+    {
+      "epoch": 1.947486033519553,
+      "grad_norm": 0.5002700607677886,
+      "learning_rate": 1.7374810318664645e-06,
+      "loss": 0.4904,
+      "step": 440
+    },
+    {
+      "epoch": 1.95195530726257,
+      "grad_norm": 0.49052978017153487,
+      "learning_rate": 1.7298937784522004e-06,
+      "loss": 0.4721,
+      "step": 441
+    },
+    {
+      "epoch": 1.9564245810055865,
+      "grad_norm": 0.5330277427973065,
+      "learning_rate": 1.7223065250379363e-06,
+      "loss": 0.4674,
+      "step": 442
+    },
+    {
+      "epoch": 1.9608938547486034,
+      "grad_norm": 0.49644600432443986,
+      "learning_rate": 1.7147192716236725e-06,
+      "loss": 0.4735,
+      "step": 443
+    },
+    {
+      "epoch": 1.96536312849162,
+      "grad_norm": 0.4763428511189237,
+      "learning_rate": 1.7071320182094084e-06,
+      "loss": 0.4758,
+      "step": 444
+    },
+    {
+      "epoch": 1.969832402234637,
+      "grad_norm": 0.49662051116381534,
+      "learning_rate": 1.6995447647951443e-06,
+      "loss": 0.5025,
+      "step": 445
+    },
+    {
+      "epoch": 1.9743016759776535,
+      "grad_norm": 0.5368027446646121,
+      "learning_rate": 1.6919575113808804e-06,
+      "loss": 0.4708,
+      "step": 446
+    },
+    {
+      "epoch": 1.9787709497206705,
+      "grad_norm": 0.49270124177132796,
+      "learning_rate": 1.6843702579666163e-06,
+      "loss": 0.4859,
+      "step": 447
+    },
+    {
+      "epoch": 1.983240223463687,
+      "grad_norm": 0.4886244120662818,
+      "learning_rate": 1.6767830045523523e-06,
+      "loss": 0.4542,
+      "step": 448
+    },
+    {
+      "epoch": 1.983240223463687,
+      "eval_loss": 0.49883514642715454,
+      "eval_runtime": 32.2692,
+      "eval_samples_per_second": 82.741,
+      "eval_steps_per_second": 5.175,
+      "step": 448
+    },
+    {
+      "epoch": 1.987709497206704,
+      "grad_norm": 0.48474622370426634,
+      "learning_rate": 1.6691957511380882e-06,
+      "loss": 0.4752,
+      "step": 449
+    },
+    {
+      "epoch": 1.9921787709497205,
+      "grad_norm": 0.5089879136002737,
+      "learning_rate": 1.6616084977238243e-06,
+      "loss": 0.4835,
+      "step": 450
+    },
+    {
+      "epoch": 1.9966480446927375,
+      "grad_norm": 0.540987114653738,
+      "learning_rate": 1.6540212443095602e-06,
+      "loss": 0.4685,
+      "step": 451
+    },
+    {
+      "epoch": 2.001117318435754,
+      "grad_norm": 0.5330263791441335,
+      "learning_rate": 1.6464339908952961e-06,
+      "loss": 0.4729,
+      "step": 452
+    },
+    {
+      "epoch": 2.005586592178771,
+      "grad_norm": 0.5025591514510047,
+      "learning_rate": 1.638846737481032e-06,
+      "loss": 0.4739,
+      "step": 453
+    },
+    {
+      "epoch": 2.010055865921788,
+      "grad_norm": 0.4938210284492649,
+      "learning_rate": 1.6312594840667682e-06,
+      "loss": 0.4772,
+      "step": 454
+    },
+    {
+      "epoch": 2.0145251396648045,
+      "grad_norm": 0.5590423643222178,
+      "learning_rate": 1.6236722306525039e-06,
+      "loss": 0.4651,
+      "step": 455
+    },
+    {
+      "epoch": 2.018994413407821,
+      "grad_norm": 0.4994221649154916,
+      "learning_rate": 1.6160849772382398e-06,
+      "loss": 0.4989,
+      "step": 456
+    },
+    {
+      "epoch": 2.004469273743017,
+      "grad_norm": 0.5578750520607872,
+      "learning_rate": 1.6084977238239757e-06,
+      "loss": 0.4779,
+      "step": 457
+    },
+    {
+      "epoch": 2.0089385474860335,
+      "grad_norm": 0.5277075050354076,
+      "learning_rate": 1.6009104704097116e-06,
+      "loss": 0.462,
+      "step": 458
+    },
+    {
+      "epoch": 2.0134078212290505,
+      "grad_norm": 0.5125996961203592,
+      "learning_rate": 1.5933232169954477e-06,
+      "loss": 0.4892,
+      "step": 459
+    },
+    {
+      "epoch": 2.017877094972067,
+      "grad_norm": 0.5150882727539035,
+      "learning_rate": 1.5857359635811836e-06,
+      "loss": 0.4913,
+      "step": 460
+    },
+    {
+      "epoch": 2.022346368715084,
+      "grad_norm": 0.5395569256138755,
+      "learning_rate": 1.5781487101669196e-06,
+      "loss": 0.4789,
+      "step": 461
+    },
+    {
+      "epoch": 2.0268156424581005,
+      "grad_norm": 0.4907655840755226,
+      "learning_rate": 1.5705614567526557e-06,
+      "loss": 0.4699,
+      "step": 462
+    },
+    {
+      "epoch": 2.0312849162011175,
+      "grad_norm": 0.508487377683692,
+      "learning_rate": 1.5629742033383916e-06,
+      "loss": 0.4671,
+      "step": 463
+    },
+    {
+      "epoch": 2.035754189944134,
+      "grad_norm": 0.5211842893319257,
+      "learning_rate": 1.5553869499241275e-06,
+      "loss": 0.48,
+      "step": 464
+    },
+    {
+      "epoch": 2.040223463687151,
+      "grad_norm": 0.5540445804429339,
+      "learning_rate": 1.5477996965098634e-06,
+      "loss": 0.4839,
+      "step": 465
+    },
+    {
+      "epoch": 2.0446927374301676,
+      "grad_norm": 0.502144462775385,
+      "learning_rate": 1.5402124430955995e-06,
+      "loss": 0.4991,
+      "step": 466
+    },
+    {
+      "epoch": 2.0491620111731845,
+      "grad_norm": 0.4963826920124019,
+      "learning_rate": 1.5326251896813355e-06,
+      "loss": 0.5268,
+      "step": 467
+    },
+    {
+      "epoch": 2.053631284916201,
+      "grad_norm": 0.48862357759049696,
+      "learning_rate": 1.5250379362670714e-06,
+      "loss": 0.4572,
+      "step": 468
+    },
+    {
+      "epoch": 2.058100558659218,
+      "grad_norm": 0.48705814461293884,
+      "learning_rate": 1.5174506828528073e-06,
+      "loss": 0.4761,
+      "step": 469
+    },
+    {
+      "epoch": 2.0625698324022346,
+      "grad_norm": 0.50151248823518,
+      "learning_rate": 1.5098634294385434e-06,
+      "loss": 0.468,
+      "step": 470
+    },
+    {
+      "epoch": 2.0670391061452515,
+      "grad_norm": 0.47075309247993236,
+      "learning_rate": 1.5022761760242793e-06,
+      "loss": 0.4496,
+      "step": 471
+    },
+    {
+      "epoch": 2.071508379888268,
+      "grad_norm": 0.5055609634042088,
+      "learning_rate": 1.4946889226100152e-06,
+      "loss": 0.4626,
+      "step": 472
+    },
+    {
+      "epoch": 2.075977653631285,
+      "grad_norm": 0.4931560723293831,
+      "learning_rate": 1.4871016691957514e-06,
+      "loss": 0.4743,
+      "step": 473
+    },
+    {
+      "epoch": 2.0804469273743016,
+      "grad_norm": 0.4970179492434698,
+      "learning_rate": 1.4795144157814873e-06,
+      "loss": 0.4849,
+      "step": 474
+    },
+    {
+      "epoch": 2.0849162011173186,
+      "grad_norm": 0.5205249715468657,
+      "learning_rate": 1.4719271623672232e-06,
+      "loss": 0.4647,
+      "step": 475
+    },
+    {
+      "epoch": 2.089385474860335,
+      "grad_norm": 0.5106891771853166,
+      "learning_rate": 1.4643399089529591e-06,
+      "loss": 0.4698,
+      "step": 476
+    },
+    {
+      "epoch": 2.093854748603352,
+      "grad_norm": 0.5030533511191912,
+      "learning_rate": 1.4567526555386952e-06,
+      "loss": 0.4435,
+      "step": 477
+    },
+    {
+      "epoch": 2.0983240223463686,
+      "grad_norm": 0.5267045945145785,
+      "learning_rate": 1.4491654021244311e-06,
+      "loss": 0.4551,
+      "step": 478
+    },
+    {
+      "epoch": 2.1027932960893856,
+      "grad_norm": 0.4910735265010368,
+      "learning_rate": 1.441578148710167e-06,
+      "loss": 0.4893,
+      "step": 479
+    },
+    {
+      "epoch": 2.107262569832402,
+      "grad_norm": 0.5098674286346987,
+      "learning_rate": 1.433990895295903e-06,
+      "loss": 0.4631,
+      "step": 480
+    },
+    {
+      "epoch": 2.111731843575419,
+      "grad_norm": 0.49833407519162715,
+      "learning_rate": 1.426403641881639e-06,
+      "loss": 0.4478,
+      "step": 481
+    },
+    {
+      "epoch": 2.1162011173184356,
+      "grad_norm": 0.4968638787102979,
+      "learning_rate": 1.418816388467375e-06,
+      "loss": 0.4661,
+      "step": 482
+    },
+    {
+      "epoch": 2.1206703910614526,
+      "grad_norm": 0.5287581344581138,
+      "learning_rate": 1.411229135053111e-06,
+      "loss": 0.487,
+      "step": 483
+    },
+    {
+      "epoch": 2.125139664804469,
+      "grad_norm": 0.5169520017940434,
+      "learning_rate": 1.403641881638847e-06,
+      "loss": 0.4641,
+      "step": 484
+    },
+    {
+      "epoch": 2.129608938547486,
+      "grad_norm": 0.4889710176907046,
+      "learning_rate": 1.396054628224583e-06,
+      "loss": 0.4657,
+      "step": 485
+    },
+    {
+      "epoch": 2.1340782122905027,
+      "grad_norm": 0.5068054201127302,
+      "learning_rate": 1.3884673748103189e-06,
+      "loss": 0.4655,
+      "step": 486
+    },
+    {
+      "epoch": 2.1385474860335196,
+      "grad_norm": 0.5227391076748372,
+      "learning_rate": 1.3808801213960548e-06,
+      "loss": 0.4489,
+      "step": 487
+    },
+    {
+      "epoch": 2.143016759776536,
+      "grad_norm": 0.4685531353247096,
+      "learning_rate": 1.3732928679817905e-06,
+      "loss": 0.4518,
+      "step": 488
+    },
+    {
+      "epoch": 2.147486033519553,
+      "grad_norm": 0.48729624697805496,
+      "learning_rate": 1.3657056145675266e-06,
+      "loss": 0.4539,
+      "step": 489
+    },
+    {
+      "epoch": 2.1519553072625697,
+      "grad_norm": 0.49686659002832273,
+      "learning_rate": 1.3581183611532625e-06,
+      "loss": 0.4698,
+      "step": 490
+    },
+    {
+      "epoch": 2.1564245810055866,
+      "grad_norm": 0.5165370496352449,
+      "learning_rate": 1.3505311077389985e-06,
+      "loss": 0.5282,
+      "step": 491
+    },
+    {
+      "epoch": 2.160893854748603,
+      "grad_norm": 0.49390583090447004,
+      "learning_rate": 1.3429438543247344e-06,
+      "loss": 0.5011,
+      "step": 492
+    },
+    {
+      "epoch": 2.16536312849162,
+      "grad_norm": 0.4966083299512843,
+      "learning_rate": 1.3353566009104705e-06,
+      "loss": 0.4888,
+      "step": 493
+    },
+    {
+      "epoch": 2.1698324022346367,
+      "grad_norm": 0.47651097727052893,
+      "learning_rate": 1.3277693474962064e-06,
+      "loss": 0.4629,
+      "step": 494
+    },
+    {
+      "epoch": 2.1743016759776537,
+      "grad_norm": 0.5171229401967155,
+      "learning_rate": 1.3201820940819423e-06,
+      "loss": 0.4538,
+      "step": 495
+    },
+    {
+      "epoch": 2.17877094972067,
+      "grad_norm": 0.5177579051508318,
+      "learning_rate": 1.3125948406676784e-06,
+      "loss": 0.4822,
+      "step": 496
+    },
+    {
+      "epoch": 2.183240223463687,
+      "grad_norm": 0.5363730349952008,
+      "learning_rate": 1.3050075872534144e-06,
+      "loss": 0.4879,
+      "step": 497
+    },
+    {
+      "epoch": 2.1877094972067037,
+      "grad_norm": 0.49547274684149156,
+      "learning_rate": 1.2974203338391503e-06,
+      "loss": 0.4732,
+      "step": 498
+    },
+    {
+      "epoch": 2.1921787709497207,
+      "grad_norm": 0.48560343902497966,
+      "learning_rate": 1.2898330804248862e-06,
+      "loss": 0.4918,
+      "step": 499
+    },
+    {
+      "epoch": 2.1966480446927372,
+      "grad_norm": 0.49966297308311575,
+      "learning_rate": 1.2822458270106223e-06,
+      "loss": 0.4618,
+      "step": 500
+    },
+    {
+      "epoch": 2.201117318435754,
+      "grad_norm": 0.472067295226221,
+      "learning_rate": 1.2746585735963582e-06,
+      "loss": 0.4953,
+      "step": 501
+    },
+    {
+      "epoch": 2.205586592178771,
+      "grad_norm": 0.5241464474412257,
+      "learning_rate": 1.2670713201820941e-06,
+      "loss": 0.4797,
+      "step": 502
+    },
+    {
+      "epoch": 2.2100558659217877,
+      "grad_norm": 0.5063960714766558,
+      "learning_rate": 1.25948406676783e-06,
+      "loss": 0.4493,
+      "step": 503
+    },
+    {
+      "epoch": 2.2145251396648042,
+      "grad_norm": 0.48407847297667117,
+      "learning_rate": 1.2518968133535662e-06,
+      "loss": 0.4736,
+      "step": 504
+    },
+    {
+      "epoch": 2.218994413407821,
+      "grad_norm": 0.4859389089228068,
+      "learning_rate": 1.244309559939302e-06,
+      "loss": 0.4476,
+      "step": 505
+    },
+    {
+      "epoch": 2.223463687150838,
+      "grad_norm": 0.49468343574845597,
+      "learning_rate": 1.236722306525038e-06,
+      "loss": 0.4447,
+      "step": 506
+    },
+    {
+      "epoch": 2.2279329608938547,
+      "grad_norm": 0.45980997966401754,
+      "learning_rate": 1.229135053110774e-06,
+      "loss": 0.4599,
+      "step": 507
+    },
+    {
+      "epoch": 2.2324022346368717,
+      "grad_norm": 0.5135776498149942,
+      "learning_rate": 1.22154779969651e-06,
+      "loss": 0.4852,
+      "step": 508
+    },
+    {
+      "epoch": 2.2368715083798882,
+      "grad_norm": 0.48749385873656575,
+      "learning_rate": 1.213960546282246e-06,
+      "loss": 0.5157,
+      "step": 509
+    },
+    {
+      "epoch": 2.241340782122905,
+      "grad_norm": 0.48899276420726745,
+      "learning_rate": 1.2063732928679819e-06,
+      "loss": 0.4437,
+      "step": 510
+    },
+    {
+      "epoch": 2.2458100558659218,
+      "grad_norm": 0.4918342780724834,
+      "learning_rate": 1.198786039453718e-06,
+      "loss": 0.4974,
+      "step": 511
+    },
+    {
+      "epoch": 2.2502793296089387,
+      "grad_norm": 0.4774969997416472,
+      "learning_rate": 1.191198786039454e-06,
+      "loss": 0.4509,
+      "step": 512
+    },
+    {
+      "epoch": 2.2547486033519553,
+      "grad_norm": 0.4791749757272505,
+      "learning_rate": 1.1836115326251896e-06,
+      "loss": 0.4599,
+      "step": 513
+    },
+    {
+      "epoch": 2.2592178770949722,
+      "grad_norm": 0.5072314853369309,
+      "learning_rate": 1.1760242792109257e-06,
+      "loss": 0.4643,
+      "step": 514
+    },
+    {
+      "epoch": 2.2636871508379888,
+      "grad_norm": 0.5004472392134528,
+      "learning_rate": 1.1684370257966617e-06,
+      "loss": 0.4688,
+      "step": 515
+    },
+    {
+      "epoch": 2.2681564245810057,
+      "grad_norm": 0.48150392498478517,
+      "learning_rate": 1.1608497723823976e-06,
+      "loss": 0.5069,
+      "step": 516
+    },
+    {
+      "epoch": 2.2726256983240223,
+      "grad_norm": 0.4828855445414688,
+      "learning_rate": 1.1532625189681337e-06,
+      "loss": 0.4571,
+      "step": 517
+    },
+    {
+      "epoch": 2.2770949720670393,
+      "grad_norm": 0.49789128855450915,
+      "learning_rate": 1.1456752655538696e-06,
+      "loss": 0.4651,
+      "step": 518
+    },
+    {
+      "epoch": 2.281564245810056,
+      "grad_norm": 0.49890243849633514,
+      "learning_rate": 1.1380880121396055e-06,
+      "loss": 0.4757,
+      "step": 519
+    },
+    {
+      "epoch": 2.2860335195530728,
+      "grad_norm": 0.48825603773254755,
+      "learning_rate": 1.1305007587253414e-06,
+      "loss": 0.4668,
+      "step": 520
+    },
+    {
+      "epoch": 2.2905027932960893,
+      "grad_norm": 0.48797823917514366,
+      "learning_rate": 1.1229135053110776e-06,
+      "loss": 0.4576,
+      "step": 521
+    },
+    {
+      "epoch": 2.2949720670391063,
+      "grad_norm": 0.4713265522467315,
+      "learning_rate": 1.1153262518968135e-06,
+      "loss": 0.4998,
+      "step": 522
+    },
+    {
+      "epoch": 2.299441340782123,
+      "grad_norm": 0.48731093463299924,
+      "learning_rate": 1.1077389984825494e-06,
+      "loss": 0.4583,
+      "step": 523
+    },
+    {
+      "epoch": 2.30391061452514,
+      "grad_norm": 0.5356992050391173,
+      "learning_rate": 1.1001517450682853e-06,
+      "loss": 0.4964,
+      "step": 524
+    },
+    {
+      "epoch": 2.3083798882681563,
+      "grad_norm": 0.5185855845215128,
+      "learning_rate": 1.0925644916540214e-06,
+      "loss": 0.4456,
+      "step": 525
+    },
+    {
+      "epoch": 2.3128491620111733,
+      "grad_norm": 0.5340341729511069,
+      "learning_rate": 1.0849772382397573e-06,
+      "loss": 0.4667,
+      "step": 526
+    },
+    {
+      "epoch": 2.31731843575419,
+      "grad_norm": 0.46791114784776194,
+      "learning_rate": 1.0773899848254933e-06,
+      "loss": 0.4907,
+      "step": 527
+    },
+    {
+      "epoch": 2.321787709497207,
+      "grad_norm": 0.5224682012273069,
+      "learning_rate": 1.0698027314112294e-06,
+      "loss": 0.4411,
+      "step": 528
+    },
+    {
+      "epoch": 2.3262569832402233,
+      "grad_norm": 0.5013175127206561,
+      "learning_rate": 1.062215477996965e-06,
+      "loss": 0.4872,
+      "step": 529
+    },
+    {
+      "epoch": 2.3307262569832403,
+      "grad_norm": 0.49567401298066655,
+      "learning_rate": 1.054628224582701e-06,
+      "loss": 0.4547,
+      "step": 530
+    },
+    {
+      "epoch": 2.335195530726257,
+      "grad_norm": 0.5092075971225333,
+      "learning_rate": 1.0470409711684371e-06,
+      "loss": 0.4774,
+      "step": 531
+    },
+    {
+      "epoch": 2.339664804469274,
+      "grad_norm": 0.5206537811339336,
+      "learning_rate": 1.039453717754173e-06,
+      "loss": 0.4593,
+      "step": 532
+    },
+    {
+      "epoch": 2.3441340782122904,
+      "grad_norm": 0.5247188237207532,
+      "learning_rate": 1.031866464339909e-06,
+      "loss": 0.4947,
+      "step": 533
+    },
+    {
+      "epoch": 2.3486033519553073,
+      "grad_norm": 0.48511496092549006,
+      "learning_rate": 1.024279210925645e-06,
+      "loss": 0.4675,
+      "step": 534
+    },
+    {
+      "epoch": 2.353072625698324,
+      "grad_norm": 0.49031229609343957,
+      "learning_rate": 1.016691957511381e-06,
+      "loss": 0.4654,
+      "step": 535
+    },
+    {
+      "epoch": 2.357541899441341,
+      "grad_norm": 0.5022437562865831,
+      "learning_rate": 1.009104704097117e-06,
+      "loss": 0.4881,
+      "step": 536
+    },
+    {
+      "epoch": 2.3620111731843574,
+      "grad_norm": 0.48427536854604036,
+      "learning_rate": 1.0015174506828528e-06,
+      "loss": 0.4722,
+      "step": 537
+    },
+    {
+      "epoch": 2.3664804469273744,
+      "grad_norm": 0.4951077677256604,
+      "learning_rate": 9.93930197268589e-07,
+      "loss": 0.4816,
+      "step": 538
+    },
+    {
+      "epoch": 2.370949720670391,
+      "grad_norm": 0.506971274674449,
+      "learning_rate": 9.863429438543249e-07,
+      "loss": 0.448,
+      "step": 539
+    },
+    {
+      "epoch": 2.375418994413408,
+      "grad_norm": 0.4843767211745505,
+      "learning_rate": 9.787556904400608e-07,
+      "loss": 0.4644,
+      "step": 540
+    },
+    {
+      "epoch": 2.3798882681564244,
+      "grad_norm": 0.4787312911085378,
+      "learning_rate": 9.711684370257967e-07,
+      "loss": 0.4681,
+      "step": 541
+    },
+    {
+      "epoch": 2.3843575418994414,
+      "grad_norm": 0.5040950848521704,
+      "learning_rate": 9.635811836115328e-07,
+      "loss": 0.4595,
+      "step": 542
+    },
+    {
+      "epoch": 2.388826815642458,
+      "grad_norm": 0.4742009596031962,
+      "learning_rate": 9.559939301972687e-07,
+      "loss": 0.4542,
+      "step": 543
+    },
+    {
+      "epoch": 2.393296089385475,
+      "grad_norm": 0.4819914757091554,
+      "learning_rate": 9.484066767830046e-07,
+      "loss": 0.4818,
+      "step": 544
+    },
+    {
+      "epoch": 2.3977653631284914,
+      "grad_norm": 0.4794763925814031,
+      "learning_rate": 9.408194233687407e-07,
+      "loss": 0.4459,
+      "step": 545
+    },
+    {
+      "epoch": 2.4022346368715084,
+      "grad_norm": 0.5065706041345039,
+      "learning_rate": 9.332321699544765e-07,
+      "loss": 0.4986,
+      "step": 546
+    },
+    {
+      "epoch": 2.4067039106145254,
+      "grad_norm": 0.4918492784353524,
+      "learning_rate": 9.256449165402125e-07,
+      "loss": 0.4618,
+      "step": 547
+    },
+    {
+      "epoch": 2.411173184357542,
+      "grad_norm": 0.48410824249150014,
+      "learning_rate": 9.180576631259484e-07,
+      "loss": 0.4358,
+      "step": 548
+    },
+    {
+      "epoch": 2.4156424581005584,
+      "grad_norm": 0.5071386174484,
+      "learning_rate": 9.104704097116844e-07,
+      "loss": 0.4522,
+      "step": 549
+    },
+    {
+      "epoch": 2.4201117318435754,
+      "grad_norm": 0.5179035085963222,
+      "learning_rate": 9.028831562974203e-07,
+      "loss": 0.4718,
+      "step": 550
+    },
+    {
+      "epoch": 2.4245810055865924,
+      "grad_norm": 0.48695002255677194,
+      "learning_rate": 8.952959028831563e-07,
+      "loss": 0.4946,
+      "step": 551
+    },
+    {
+      "epoch": 2.429050279329609,
+      "grad_norm": 0.4697467884861594,
+      "learning_rate": 8.877086494688924e-07,
+      "loss": 0.482,
+      "step": 552
+    },
+    {
+      "epoch": 2.4335195530726255,
+      "grad_norm": 0.5039869709064255,
+      "learning_rate": 8.801213960546283e-07,
+      "loss": 0.4528,
+      "step": 553
+    },
+    {
+      "epoch": 2.4379888268156424,
+      "grad_norm": 0.4956541767737936,
+      "learning_rate": 8.725341426403643e-07,
+      "loss": 0.4416,
+      "step": 554
+    },
+    {
+      "epoch": 2.4424581005586594,
+      "grad_norm": 0.5220748437476452,
+      "learning_rate": 8.649468892261002e-07,
+      "loss": 0.4534,
+      "step": 555
+    },
+    {
+      "epoch": 2.446927374301676,
+      "grad_norm": 0.5004114462806725,
+      "learning_rate": 8.573596358118362e-07,
+      "loss": 0.4523,
+      "step": 556
+    },
+    {
+      "epoch": 2.451396648044693,
+      "grad_norm": 0.5002883358165275,
+      "learning_rate": 8.497723823975721e-07,
+      "loss": 0.4483,
+      "step": 557
+    },
+    {
+      "epoch": 2.4558659217877095,
+      "grad_norm": 0.5001315929982186,
+      "learning_rate": 8.421851289833082e-07,
+      "loss": 0.4547,
+      "step": 558
+    },
+    {
+      "epoch": 2.4603351955307264,
+      "grad_norm": 0.5141518132214271,
+      "learning_rate": 8.345978755690441e-07,
+      "loss": 0.5064,
+      "step": 559
+    },
+    {
+      "epoch": 2.464804469273743,
+      "grad_norm": 0.4960566337702956,
+      "learning_rate": 8.270106221547801e-07,
+      "loss": 0.4473,
+      "step": 560
+    },
+    {
+      "epoch": 2.464804469273743,
+      "eval_loss": 0.4975128769874573,
+      "eval_runtime": 32.1892,
+      "eval_samples_per_second": 82.947,
+      "eval_steps_per_second": 5.188,
+      "step": 560
+    },
+    {
+      "epoch": 2.46927374301676,
+      "grad_norm": 0.4897198333575608,
+      "learning_rate": 8.19423368740516e-07,
+      "loss": 0.4559,
+      "step": 561
+    },
+    {
+      "epoch": 2.4737430167597765,
+      "grad_norm": 0.49985877359817477,
+      "learning_rate": 8.118361153262519e-07,
+      "loss": 0.4651,
+      "step": 562
+    },
+    {
+      "epoch": 2.4782122905027935,
+      "grad_norm": 0.49485306049230043,
+      "learning_rate": 8.042488619119878e-07,
+      "loss": 0.4803,
+      "step": 563
+    },
+    {
+      "epoch": 2.48268156424581,
+      "grad_norm": 0.5208900276481403,
+      "learning_rate": 7.966616084977239e-07,
+      "loss": 0.4553,
+      "step": 564
+    },
+    {
+      "epoch": 2.487150837988827,
+      "grad_norm": 0.48837650054879367,
+      "learning_rate": 7.890743550834598e-07,
+      "loss": 0.4681,
+      "step": 565
+    },
+    {
+      "epoch": 2.4916201117318435,
+      "grad_norm": 0.47305516115923485,
+      "learning_rate": 7.814871016691958e-07,
+      "loss": 0.4564,
+      "step": 566
+    },
+    {
+      "epoch": 2.4960893854748605,
+      "grad_norm": 0.5049602967209187,
+      "learning_rate": 7.738998482549317e-07,
+      "loss": 0.4632,
+      "step": 567
+    },
+    {
+      "epoch": 2.500558659217877,
+      "grad_norm": 0.48547661489581506,
+      "learning_rate": 7.663125948406677e-07,
+      "loss": 0.448,
+      "step": 568
+    },
+    {
+      "epoch": 2.505027932960894,
+      "grad_norm": 0.5747686136950829,
+      "learning_rate": 7.587253414264036e-07,
+      "loss": 0.4432,
+      "step": 569
+    },
+    {
+      "epoch": 2.5094972067039105,
+      "grad_norm": 0.5499322063415041,
+      "learning_rate": 7.511380880121397e-07,
+      "loss": 0.432,
+      "step": 570
+    },
+    {
+      "epoch": 2.5139664804469275,
+      "grad_norm": 0.49909881450641125,
+      "learning_rate": 7.435508345978757e-07,
+      "loss": 0.4325,
+      "step": 571
+    },
+    {
+      "epoch": 2.518435754189944,
+      "grad_norm": 0.5137001617086465,
+      "learning_rate": 7.359635811836116e-07,
+      "loss": 0.4374,
+      "step": 572
+    },
+    {
+      "epoch": 2.522905027932961,
+      "grad_norm": 0.5043757440405017,
+      "learning_rate": 7.283763277693476e-07,
+      "loss": 0.4751,
+      "step": 573
+    },
+    {
+      "epoch": 2.5273743016759775,
+      "grad_norm": 0.4882894387336232,
+      "learning_rate": 7.207890743550835e-07,
+      "loss": 0.4368,
+      "step": 574
+    },
+    {
+      "epoch": 2.5318435754189945,
+      "grad_norm": 0.5302357654250439,
+      "learning_rate": 7.132018209408196e-07,
+      "loss": 0.4174,
+      "step": 575
+    },
+    {
+      "epoch": 2.536312849162011,
+      "grad_norm": 0.5229623842607946,
+      "learning_rate": 7.056145675265555e-07,
+      "loss": 0.4353,
+      "step": 576
+    },
+    {
+      "epoch": 2.540782122905028,
+      "grad_norm": 0.5145290981136341,
+      "learning_rate": 6.980273141122915e-07,
+      "loss": 0.4278,
+      "step": 577
+    },
+    {
+      "epoch": 2.5452513966480446,
+      "grad_norm": 0.49830066459201744,
+      "learning_rate": 6.904400606980274e-07,
+      "loss": 0.434,
+      "step": 578
+    },
+    {
+      "epoch": 2.5497206703910615,
+      "grad_norm": 0.5220459047427843,
+      "learning_rate": 6.828528072837633e-07,
+      "loss": 0.4375,
+      "step": 579
+    },
+    {
+      "epoch": 2.554189944134078,
+      "grad_norm": 0.5111391623585094,
+      "learning_rate": 6.752655538694992e-07,
+      "loss": 0.4295,
+      "step": 580
+    },
+    {
+      "epoch": 2.558659217877095,
+      "grad_norm": 0.4944043554942434,
+      "learning_rate": 6.676783004552352e-07,
+      "loss": 0.4205,
+      "step": 581
+    },
+    {
+      "epoch": 2.5631284916201116,
+      "grad_norm": 0.49154601956517674,
+      "learning_rate": 6.600910470409712e-07,
+      "loss": 0.4502,
+      "step": 582
+    },
+    {
+      "epoch": 2.5675977653631286,
+      "grad_norm": 0.5008348019898189,
+      "learning_rate": 6.525037936267072e-07,
+      "loss": 0.4453,
+      "step": 583
+    },
+    {
+      "epoch": 2.572067039106145,
+      "grad_norm": 0.5005572885290496,
+      "learning_rate": 6.449165402124431e-07,
+      "loss": 0.4353,
+      "step": 584
+    },
+    {
+      "epoch": 2.576536312849162,
+      "grad_norm": 0.5302967367534124,
+      "learning_rate": 6.373292867981791e-07,
+      "loss": 0.4269,
+      "step": 585
+    },
+    {
+      "epoch": 2.5810055865921786,
+      "grad_norm": 0.5050943269077487,
+      "learning_rate": 6.29742033383915e-07,
+      "loss": 0.4372,
+      "step": 586
+    },
+    {
+      "epoch": 2.5854748603351956,
+      "grad_norm": 0.5223556534474804,
+      "learning_rate": 6.22154779969651e-07,
+      "loss": 0.445,
+      "step": 587
+    },
+    {
+      "epoch": 2.5899441340782126,
+      "grad_norm": 0.4891137574673572,
+      "learning_rate": 6.14567526555387e-07,
+      "loss": 0.4171,
+      "step": 588
+    },
+    {
+      "epoch": 2.594413407821229,
+      "grad_norm": 0.5172276124686148,
+      "learning_rate": 6.06980273141123e-07,
+      "loss": 0.4522,
+      "step": 589
+    },
+    {
+      "epoch": 2.5988826815642456,
+      "grad_norm": 0.48873823307103154,
+      "learning_rate": 5.99393019726859e-07,
+      "loss": 0.4286,
+      "step": 590
+    },
+    {
+      "epoch": 2.6033519553072626,
+      "grad_norm": 0.502230133682332,
+      "learning_rate": 5.918057663125948e-07,
+      "loss": 0.4461,
+      "step": 591
+    },
+    {
+      "epoch": 2.6078212290502796,
+      "grad_norm": 0.4898396410998909,
+      "learning_rate": 5.842185128983308e-07,
+      "loss": 0.4289,
+      "step": 592
+    },
+    {
+      "epoch": 2.612290502793296,
+      "grad_norm": 0.48422568768513635,
+      "learning_rate": 5.766312594840668e-07,
+      "loss": 0.4423,
+      "step": 593
+    },
+    {
+      "epoch": 2.6167597765363126,
+      "grad_norm": 0.5003994120834477,
+      "learning_rate": 5.690440060698028e-07,
+      "loss": 0.437,
+      "step": 594
+    },
+    {
+      "epoch": 2.6212290502793296,
+      "grad_norm": 0.5079705228889307,
+      "learning_rate": 5.614567526555388e-07,
+      "loss": 0.4516,
+      "step": 595
+    },
+    {
+      "epoch": 2.6256983240223466,
+      "grad_norm": 0.49246631419316356,
+      "learning_rate": 5.538694992412747e-07,
+      "loss": 0.4368,
+      "step": 596
+    },
+    {
+      "epoch": 2.630167597765363,
+      "grad_norm": 0.5086922889782935,
+      "learning_rate": 5.462822458270107e-07,
+      "loss": 0.4227,
+      "step": 597
+    },
+    {
+      "epoch": 2.6346368715083797,
+      "grad_norm": 0.4780759168718152,
+      "learning_rate": 5.386949924127466e-07,
+      "loss": 0.4462,
+      "step": 598
+    },
+    {
+      "epoch": 2.6391061452513966,
+      "grad_norm": 0.464930849547806,
+      "learning_rate": 5.311077389984825e-07,
+      "loss": 0.4334,
+      "step": 599
+    },
+    {
+      "epoch": 2.6435754189944136,
+      "grad_norm": 0.5002300455424397,
+      "learning_rate": 5.235204855842186e-07,
+      "loss": 0.4641,
+      "step": 600
+    },
+    {
+      "epoch": 2.64804469273743,
+      "grad_norm": 0.4901819784256759,
+      "learning_rate": 5.159332321699545e-07,
+      "loss": 0.4534,
+      "step": 601
+    },
+    {
+      "epoch": 2.6525139664804467,
+      "grad_norm": 0.48520187595124403,
+      "learning_rate": 5.083459787556905e-07,
+      "loss": 0.4246,
+      "step": 602
+    },
+    {
+      "epoch": 2.6569832402234637,
+      "grad_norm": 0.5093634792454178,
+      "learning_rate": 5.007587253414264e-07,
+      "loss": 0.458,
+      "step": 603
+    },
+    {
+      "epoch": 2.6614525139664806,
+      "grad_norm": 0.47795678253705354,
+      "learning_rate": 4.931714719271624e-07,
+      "loss": 0.4535,
+      "step": 604
+    },
+    {
+      "epoch": 2.665921787709497,
+      "grad_norm": 0.48340944166993405,
+      "learning_rate": 4.855842185128983e-07,
+      "loss": 0.4721,
+      "step": 605
+    },
+    {
+      "epoch": 2.6703910614525137,
+      "grad_norm": 0.47996848455446794,
+      "learning_rate": 4.779969650986344e-07,
+      "loss": 0.4718,
+      "step": 606
+    },
+    {
+      "epoch": 2.6748603351955307,
+      "grad_norm": 0.5064110255390989,
+      "learning_rate": 4.7040971168437033e-07,
+      "loss": 0.4232,
+      "step": 607
+    },
+    {
+      "epoch": 2.6793296089385477,
+      "grad_norm": 0.4824240302682651,
+      "learning_rate": 4.6282245827010624e-07,
+      "loss": 0.4307,
+      "step": 608
+    },
+    {
+      "epoch": 2.683798882681564,
+      "grad_norm": 0.4920800315955025,
+      "learning_rate": 4.552352048558422e-07,
+      "loss": 0.4435,
+      "step": 609
+    },
+    {
+      "epoch": 2.688268156424581,
+      "grad_norm": 0.48921504704563673,
+      "learning_rate": 4.476479514415782e-07,
+      "loss": 0.4409,
+      "step": 610
+    },
+    {
+      "epoch": 2.6927374301675977,
+      "grad_norm": 0.500357475284234,
+      "learning_rate": 4.4006069802731414e-07,
+      "loss": 0.4512,
+      "step": 611
+    },
+    {
+      "epoch": 2.6972067039106147,
+      "grad_norm": 0.4772853981005586,
+      "learning_rate": 4.324734446130501e-07,
+      "loss": 0.4305,
+      "step": 612
+    },
+    {
+      "epoch": 2.701675977653631,
+      "grad_norm": 0.5075964338612743,
+      "learning_rate": 4.248861911987861e-07,
+      "loss": 0.4285,
+      "step": 613
+    },
+    {
+      "epoch": 2.706145251396648,
+      "grad_norm": 0.4916721748305679,
+      "learning_rate": 4.1729893778452204e-07,
+      "loss": 0.4586,
+      "step": 614
+    },
+    {
+      "epoch": 2.7106145251396647,
+      "grad_norm": 0.4746003660659448,
+      "learning_rate": 4.09711684370258e-07,
+      "loss": 0.4385,
+      "step": 615
+    },
+    {
+      "epoch": 2.7150837988826817,
+      "grad_norm": 0.49004649736827927,
+      "learning_rate": 4.021244309559939e-07,
+      "loss": 0.4349,
+      "step": 616
+    },
+    {
+      "epoch": 2.7195530726256982,
+      "grad_norm": 0.5055169203153188,
+      "learning_rate": 3.945371775417299e-07,
+      "loss": 0.4741,
+      "step": 617
+    },
+    {
+      "epoch": 2.724022346368715,
+      "grad_norm": 0.5491416139361283,
+      "learning_rate": 3.8694992412746586e-07,
+      "loss": 0.4285,
+      "step": 618
+    },
+    {
+      "epoch": 2.7284916201117317,
+      "grad_norm": 0.4860669923591883,
+      "learning_rate": 3.793626707132018e-07,
+      "loss": 0.4147,
+      "step": 619
+    },
+    {
+      "epoch": 2.7329608938547487,
+      "grad_norm": 0.48537127261627433,
+      "learning_rate": 3.7177541729893784e-07,
+      "loss": 0.4538,
+      "step": 620
+    },
+    {
+      "epoch": 2.7374301675977653,
+      "grad_norm": 0.47557197343237195,
+      "learning_rate": 3.641881638846738e-07,
+      "loss": 0.454,
+      "step": 621
+    },
+    {
+      "epoch": 2.7418994413407822,
+      "grad_norm": 0.4844044051975442,
+      "learning_rate": 3.566009104704098e-07,
+      "loss": 0.4273,
+      "step": 622
+    },
+    {
+      "epoch": 2.7463687150837988,
+      "grad_norm": 0.5000172198487197,
+      "learning_rate": 3.4901365705614574e-07,
+      "loss": 0.4315,
+      "step": 623
+    },
+    {
+      "epoch": 2.7508379888268157,
+      "grad_norm": 0.4848986778566373,
+      "learning_rate": 3.4142640364188166e-07,
+      "loss": 0.4495,
+      "step": 624
+    },
+    {
+      "epoch": 2.7553072625698323,
+      "grad_norm": 0.48082262468066483,
+      "learning_rate": 3.338391502276176e-07,
+      "loss": 0.4185,
+      "step": 625
+    },
+    {
+      "epoch": 2.7597765363128492,
+      "grad_norm": 0.4865964770369407,
+      "learning_rate": 3.262518968133536e-07,
+      "loss": 0.4538,
+      "step": 626
+    },
+    {
+      "epoch": 2.764245810055866,
+      "grad_norm": 0.4815694373638968,
+      "learning_rate": 3.1866464339908956e-07,
+      "loss": 0.459,
+      "step": 627
+    },
+    {
+      "epoch": 2.7687150837988828,
+      "grad_norm": 0.49485301744734556,
+      "learning_rate": 3.110773899848255e-07,
+      "loss": 0.4527,
+      "step": 628
+    },
+    {
+      "epoch": 2.7731843575418993,
+      "grad_norm": 0.5214886405175438,
+      "learning_rate": 3.034901365705615e-07,
+      "loss": 0.4393,
+      "step": 629
+    },
+    {
+      "epoch": 2.7776536312849163,
+      "grad_norm": 0.4645817613806632,
+      "learning_rate": 2.959028831562974e-07,
+      "loss": 0.4723,
+      "step": 630
+    },
+    {
+      "epoch": 2.782122905027933,
+      "grad_norm": 0.4887881586065441,
+      "learning_rate": 2.883156297420334e-07,
+      "loss": 0.418,
+      "step": 631
+    },
+    {
+      "epoch": 2.78659217877095,
+      "grad_norm": 0.47562569136752975,
+      "learning_rate": 2.807283763277694e-07,
+      "loss": 0.4347,
+      "step": 632
+    },
+    {
+      "epoch": 2.7910614525139668,
+      "grad_norm": 0.4848933584055101,
+      "learning_rate": 2.7314112291350536e-07,
+      "loss": 0.4266,
+      "step": 633
+    },
+    {
+      "epoch": 2.7955307262569833,
+      "grad_norm": 0.48829866920710585,
+      "learning_rate": 2.6555386949924127e-07,
+      "loss": 0.4522,
+      "step": 634
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 0.4776396492507318,
+      "learning_rate": 2.5796661608497724e-07,
+      "loss": 0.4293,
+      "step": 635
+    },
+    {
+      "epoch": 2.804469273743017,
+      "grad_norm": 0.48342626581685083,
+      "learning_rate": 2.503793626707132e-07,
+      "loss": 0.437,
+      "step": 636
+    },
+    {
+      "epoch": 2.8089385474860338,
+      "grad_norm": 0.47481545268800873,
+      "learning_rate": 2.4279210925644917e-07,
+      "loss": 0.4333,
+      "step": 637
+    },
+    {
+      "epoch": 2.8134078212290503,
+      "grad_norm": 0.5105857690012979,
+      "learning_rate": 2.3520485584218516e-07,
+      "loss": 0.4367,
+      "step": 638
+    },
+    {
+      "epoch": 2.817877094972067,
+      "grad_norm": 0.49397311695562557,
+      "learning_rate": 2.276176024279211e-07,
+      "loss": 0.4355,
+      "step": 639
+    },
+    {
+      "epoch": 2.822346368715084,
+      "grad_norm": 0.4766966322472669,
+      "learning_rate": 2.2003034901365707e-07,
+      "loss": 0.4478,
+      "step": 640
+    },
+    {
+      "epoch": 2.826815642458101,
+      "grad_norm": 0.47484165703817144,
+      "learning_rate": 2.1244309559939304e-07,
+      "loss": 0.4366,
+      "step": 641
+    },
+    {
+      "epoch": 2.8312849162011173,
+      "grad_norm": 0.47271019258961905,
+      "learning_rate": 2.04855842185129e-07,
+      "loss": 0.4473,
+      "step": 642
+    },
+    {
+      "epoch": 2.835754189944134,
+      "grad_norm": 0.4668544421054751,
+      "learning_rate": 1.9726858877086494e-07,
+      "loss": 0.4341,
+      "step": 643
+    },
+    {
+      "epoch": 2.840223463687151,
+      "grad_norm": 0.47716986376667936,
+      "learning_rate": 1.896813353566009e-07,
+      "loss": 0.4223,
+      "step": 644
+    },
+    {
+      "epoch": 2.844692737430168,
+      "grad_norm": 0.4939991882199546,
+      "learning_rate": 1.820940819423369e-07,
+      "loss": 0.4441,
+      "step": 645
+    },
+    {
+      "epoch": 2.8491620111731844,
+      "grad_norm": 0.47705877463455093,
+      "learning_rate": 1.7450682852807287e-07,
+      "loss": 0.485,
+      "step": 646
+    },
+    {
+      "epoch": 2.853631284916201,
+      "grad_norm": 0.49941990019846055,
+      "learning_rate": 1.669195751138088e-07,
+      "loss": 0.4701,
+      "step": 647
+    },
+    {
+      "epoch": 2.858100558659218,
+      "grad_norm": 0.47625945090740057,
+      "learning_rate": 1.5933232169954478e-07,
+      "loss": 0.454,
+      "step": 648
+    },
+    {
+      "epoch": 2.862569832402235,
+      "grad_norm": 0.5028271948555065,
+      "learning_rate": 1.5174506828528074e-07,
+      "loss": 0.4296,
+      "step": 649
+    },
+    {
+      "epoch": 2.8670391061452514,
+      "grad_norm": 0.49358352504671604,
+      "learning_rate": 1.441578148710167e-07,
+      "loss": 0.4324,
+      "step": 650
+    },
+    {
+      "epoch": 2.871508379888268,
+      "grad_norm": 0.48576223310259714,
+      "learning_rate": 1.3657056145675268e-07,
+      "loss": 0.4424,
+      "step": 651
+    },
+    {
+      "epoch": 2.875977653631285,
+      "grad_norm": 0.47971171549572994,
+      "learning_rate": 1.2898330804248862e-07,
+      "loss": 0.4555,
+      "step": 652
+    },
+    {
+      "epoch": 2.880446927374302,
+      "grad_norm": 0.49955392754214234,
+      "learning_rate": 1.2139605462822459e-07,
+      "loss": 0.4338,
+      "step": 653
+    },
+    {
+      "epoch": 2.8849162011173184,
+      "grad_norm": 0.47232821100099637,
+      "learning_rate": 1.1380880121396055e-07,
+      "loss": 0.4349,
+      "step": 654
+    },
+    {
+      "epoch": 2.889385474860335,
+      "grad_norm": 0.4834170945858496,
+      "learning_rate": 1.0622154779969652e-07,
+      "loss": 0.4564,
+      "step": 655
+    },
+    {
+      "epoch": 2.893854748603352,
+      "grad_norm": 0.48493984226629705,
+      "learning_rate": 9.863429438543247e-08,
+      "loss": 0.4384,
+      "step": 656
+    },
+    {
+      "epoch": 2.898324022346369,
+      "grad_norm": 0.4601302405573049,
+      "learning_rate": 9.104704097116845e-08,
+      "loss": 0.4512,
+      "step": 657
+    },
+    {
+      "epoch": 2.9027932960893854,
+      "grad_norm": 0.4768143158802154,
+      "learning_rate": 8.34597875569044e-08,
+      "loss": 0.4354,
+      "step": 658
+    },
+    {
+      "epoch": 2.9072625698324024,
+      "grad_norm": 0.49427262358096746,
+      "learning_rate": 7.587253414264037e-08,
+      "loss": 0.4259,
+      "step": 659
+    },
+    {
+      "epoch": 2.911731843575419,
+      "grad_norm": 0.4749814243858473,
+      "learning_rate": 6.828528072837634e-08,
+      "loss": 0.4445,
+      "step": 660
+    },
+    {
+      "epoch": 2.916201117318436,
+      "grad_norm": 0.48533516583361297,
+      "learning_rate": 6.069802731411229e-08,
+      "loss": 0.4799,
+      "step": 661
+    },
+    {
+      "epoch": 2.9206703910614524,
+      "grad_norm": 0.4839142899304728,
+      "learning_rate": 5.311077389984826e-08,
+      "loss": 0.4574,
+      "step": 662
+    },
+    {
+      "epoch": 2.9251396648044694,
+      "grad_norm": 0.486543073762813,
+      "learning_rate": 4.5523520485584226e-08,
+      "loss": 0.4446,
+      "step": 663
+    },
+    {
+      "epoch": 2.929608938547486,
+      "grad_norm": 0.4980109939724879,
+      "learning_rate": 3.7936267071320186e-08,
+      "loss": 0.4315,
+      "step": 664
+    },
+    {
+      "epoch": 2.934078212290503,
+      "grad_norm": 0.48840296818117557,
+      "learning_rate": 3.0349013657056146e-08,
+      "loss": 0.4318,
+      "step": 665
+    },
+    {
+      "epoch": 2.9385474860335195,
+      "grad_norm": 0.4903615270099699,
+      "learning_rate": 2.2761760242792113e-08,
+      "loss": 0.4357,
+      "step": 666
+    },
+    {
+      "epoch": 2.9430167597765364,
+      "grad_norm": 0.48850430880690693,
+      "learning_rate": 1.5174506828528073e-08,
+      "loss": 0.4334,
+      "step": 667
+    },
+    {
+      "epoch": 2.947486033519553,
+      "grad_norm": 0.4773488572878224,
+      "learning_rate": 7.587253414264037e-09,
+      "loss": 0.4281,
+      "step": 668
+    },
+    {
+      "epoch": 2.95195530726257,
+      "grad_norm": 0.4760606483764159,
+      "learning_rate": 0.0,
+      "loss": 0.4504,
+      "step": 669
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 669,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 56,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 559881199288320.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}