diff --git "a/checkpoint-718/trainer_state.json" "b/checkpoint-718/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-718/trainer_state.json"
@@ -0,0 +1,5079 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9996519317786287,
+  "eval_steps": 180,
+  "global_step": 718,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.08854816108942032,
+      "learning_rate": 2e-05,
+      "loss": 1.1956,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "eval_loss": 1.1464052200317383,
+      "eval_runtime": 96.6552,
+      "eval_samples_per_second": 51.73,
+      "eval_steps_per_second": 12.933,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.10053814947605133,
+      "learning_rate": 4e-05,
+      "loss": 1.1428,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.09509150683879852,
+      "learning_rate": 6e-05,
+      "loss": 1.1963,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.0889890268445015,
+      "learning_rate": 8e-05,
+      "loss": 1.2185,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.0988571047782898,
+      "learning_rate": 0.0001,
+      "loss": 1.1157,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.089688241481781,
+      "learning_rate": 0.00012,
+      "loss": 1.2244,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.08509726822376251,
+      "learning_rate": 0.00014,
+      "loss": 1.2232,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.07318675518035889,
+      "learning_rate": 0.00016,
+      "loss": 1.131,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.07840697467327118,
+      "learning_rate": 0.00018,
+      "loss": 1.2006,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.08998328447341919,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.110952228307724,
+      "learning_rate": 0.0001999998926455355,
+      "loss": 1.0462,
+      "step": 11
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.08632127940654755,
+      "learning_rate": 0.0001999995705823725,
+      "loss": 1.1094,
+      "step": 12
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.06929520517587662,
+      "learning_rate": 0.00019999903381120245,
+      "loss": 1.1388,
+      "step": 13
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.058088913559913635,
+      "learning_rate": 0.0001999982823331779,
+      "loss": 1.0518,
+      "step": 14
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.05951884761452675,
+      "learning_rate": 0.0001999973161499123,
+      "loss": 1.1296,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.06432141363620758,
+      "learning_rate": 0.00019999613526348019,
+      "loss": 1.2231,
+      "step": 16
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.05998155102133751,
+      "learning_rate": 0.00019999473967641696,
+      "loss": 1.1548,
+      "step": 17
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.06434385478496552,
+      "learning_rate": 0.00019999312939171914,
+      "loss": 1.2023,
+      "step": 18
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.059631019830703735,
+      "learning_rate": 0.00019999130441284408,
+      "loss": 1.176,
+      "step": 19
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.06641547381877899,
+      "learning_rate": 0.00019998926474371022,
+      "loss": 1.1081,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.05724109336733818,
+      "learning_rate": 0.0001999870103886969,
+      "loss": 1.1644,
+      "step": 21
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.05426528677344322,
+      "learning_rate": 0.00019998454135264444,
+      "loss": 1.0424,
+      "step": 22
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.06212550774216652,
+      "learning_rate": 0.00019998185764085404,
+      "loss": 1.1966,
+      "step": 23
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.05805288627743721,
+      "learning_rate": 0.0001999789592590879,
+      "loss": 1.0189,
+      "step": 24
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.06322558224201202,
+      "learning_rate": 0.00019997584621356907,
+      "loss": 1.1605,
+      "step": 25
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.05367085337638855,
+      "learning_rate": 0.0001999725185109816,
+      "loss": 1.1008,
+      "step": 26
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.06441543996334076,
+      "learning_rate": 0.00019996897615847032,
+      "loss": 1.1974,
+      "step": 27
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.06023159250617027,
+      "learning_rate": 0.00019996521916364096,
+      "loss": 1.0898,
+      "step": 28
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.05423330143094063,
+      "learning_rate": 0.00019996124753456017,
+      "loss": 1.1021,
+      "step": 29
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.05826232209801674,
+      "learning_rate": 0.00019995706127975537,
+      "loss": 1.1573,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.06136321276426315,
+      "learning_rate": 0.00019995266040821483,
+      "loss": 1.0987,
+      "step": 31
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.057079821825027466,
+      "learning_rate": 0.0001999480449293876,
+      "loss": 1.1706,
+      "step": 32
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.0606551319360733,
+      "learning_rate": 0.00019994321485318354,
+      "loss": 1.1549,
+      "step": 33
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.059561554342508316,
+      "learning_rate": 0.00019993817018997323,
+      "loss": 1.1983,
+      "step": 34
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.05861816182732582,
+      "learning_rate": 0.00019993291095058804,
+      "loss": 1.1644,
+      "step": 35
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.05660370737314224,
+      "learning_rate": 0.00019992743714632,
+      "loss": 1.1671,
+      "step": 36
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.05508609116077423,
+      "learning_rate": 0.00019992174878892186,
+      "loss": 1.2054,
+      "step": 37
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.058200910687446594,
+      "learning_rate": 0.0001999158458906071,
+      "loss": 1.1693,
+      "step": 38
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.05510352924466133,
+      "learning_rate": 0.00019990972846404967,
+      "loss": 1.0654,
+      "step": 39
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.05602727457880974,
+      "learning_rate": 0.0001999033965223843,
+      "loss": 1.0574,
+      "step": 40
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.06220879405736923,
+      "learning_rate": 0.0001998968500792062,
+      "loss": 1.1197,
+      "step": 41
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.0568472184240818,
+      "learning_rate": 0.00019989008914857116,
+      "loss": 1.0929,
+      "step": 42
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.06510251760482788,
+      "learning_rate": 0.0001998831137449955,
+      "loss": 1.1372,
+      "step": 43
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.05792449042201042,
+      "learning_rate": 0.00019987592388345611,
+      "loss": 1.1506,
+      "step": 44
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.05337774381041527,
+      "learning_rate": 0.00019986851957939016,
+      "loss": 1.1419,
+      "step": 45
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.05266134813427925,
+      "learning_rate": 0.00019986090084869545,
+      "loss": 1.1284,
+      "step": 46
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.05520001798868179,
+      "learning_rate": 0.00019985306770773,
+      "loss": 1.1934,
+      "step": 47
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.056133098900318146,
+      "learning_rate": 0.00019984502017331225,
+      "loss": 1.1348,
+      "step": 48
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.05897597223520279,
+      "learning_rate": 0.00019983675826272106,
+      "loss": 1.0986,
+      "step": 49
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.0529349111020565,
+      "learning_rate": 0.00019982828199369541,
+      "loss": 1.0987,
+      "step": 50
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.0615835078060627,
+      "learning_rate": 0.00019981959138443467,
+      "loss": 1.1104,
+      "step": 51
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.059136223047971725,
+      "learning_rate": 0.0001998106864535983,
+      "loss": 1.1611,
+      "step": 52
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.05952458456158638,
+      "learning_rate": 0.000199801567220306,
+      "loss": 1.0697,
+      "step": 53
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.059737298637628555,
+      "learning_rate": 0.00019979223370413763,
+      "loss": 1.1294,
+      "step": 54
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.05678507685661316,
+      "learning_rate": 0.00019978268592513296,
+      "loss": 1.1584,
+      "step": 55
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.05472412705421448,
+      "learning_rate": 0.00019977292390379207,
+      "loss": 1.1416,
+      "step": 56
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.05907364934682846,
+      "learning_rate": 0.00019976294766107478,
+      "loss": 1.141,
+      "step": 57
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.05549243837594986,
+      "learning_rate": 0.00019975275721840103,
+      "loss": 1.0759,
+      "step": 58
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.062153417617082596,
+      "learning_rate": 0.0001997423525976506,
+      "loss": 1.1715,
+      "step": 59
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.05978319048881531,
+      "learning_rate": 0.0001997317338211631,
+      "loss": 1.1452,
+      "step": 60
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.053922947496175766,
+      "learning_rate": 0.00019972090091173805,
+      "loss": 1.204,
+      "step": 61
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.055929865688085556,
+      "learning_rate": 0.00019970985389263467,
+      "loss": 1.1136,
+      "step": 62
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.05876065790653229,
+      "learning_rate": 0.00019969859278757185,
+      "loss": 1.1364,
+      "step": 63
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.06039357930421829,
+      "learning_rate": 0.0001996871176207282,
+      "loss": 1.0784,
+      "step": 64
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.055668413639068604,
+      "learning_rate": 0.00019967542841674196,
+      "loss": 1.0948,
+      "step": 65
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.054827943444252014,
+      "learning_rate": 0.0001996635252007109,
+      "loss": 1.1803,
+      "step": 66
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.05121610313653946,
+      "learning_rate": 0.00019965140799819224,
+      "loss": 1.0376,
+      "step": 67
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.05753542110323906,
+      "learning_rate": 0.00019963907683520274,
+      "loss": 1.1206,
+      "step": 68
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.055351294577121735,
+      "learning_rate": 0.00019962653173821843,
+      "loss": 1.1842,
+      "step": 69
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.061469532549381256,
+      "learning_rate": 0.00019961377273417487,
+      "loss": 1.0909,
+      "step": 70
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.054115746170282364,
+      "learning_rate": 0.0001996007998504667,
+      "loss": 1.0843,
+      "step": 71
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.05452101677656174,
+      "learning_rate": 0.0001995876131149479,
+      "loss": 1.1006,
+      "step": 72
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.055213477462530136,
+      "learning_rate": 0.00019957421255593154,
+      "loss": 1.1685,
+      "step": 73
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.055633626878261566,
+      "learning_rate": 0.00019956059820218982,
+      "loss": 1.1772,
+      "step": 74
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.05437927320599556,
+      "learning_rate": 0.000199546770082954,
+      "loss": 1.1043,
+      "step": 75
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.058048781007528305,
+      "learning_rate": 0.00019953272822791424,
+      "loss": 1.1045,
+      "step": 76
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.06041900813579559,
+      "learning_rate": 0.0001995184726672197,
+      "loss": 1.1975,
+      "step": 77
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.060865532606840134,
+      "learning_rate": 0.00019950400343147833,
+      "loss": 1.1731,
+      "step": 78
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.056755781173706055,
+      "learning_rate": 0.00019948932055175686,
+      "loss": 1.0745,
+      "step": 79
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.058669883757829666,
+      "learning_rate": 0.00019947442405958074,
+      "loss": 1.0734,
+      "step": 80
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.058248650282621384,
+      "learning_rate": 0.00019945931398693408,
+      "loss": 1.1543,
+      "step": 81
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.060298267751932144,
+      "learning_rate": 0.0001994439903662596,
+      "loss": 1.1762,
+      "step": 82
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.05813004449009895,
+      "learning_rate": 0.00019942845323045837,
+      "loss": 1.1652,
+      "step": 83
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.056509047746658325,
+      "learning_rate": 0.00019941270261289012,
+      "loss": 1.1391,
+      "step": 84
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.05784199759364128,
+      "learning_rate": 0.00019939673854737277,
+      "loss": 1.1592,
+      "step": 85
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.05908602103590965,
+      "learning_rate": 0.00019938056106818261,
+      "loss": 1.1867,
+      "step": 86
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.05213363468647003,
+      "learning_rate": 0.00019936417021005414,
+      "loss": 1.0747,
+      "step": 87
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.0627497211098671,
+      "learning_rate": 0.00019934756600817997,
+      "loss": 1.1086,
+      "step": 88
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.05405977740883827,
+      "learning_rate": 0.00019933074849821084,
+      "loss": 0.9786,
+      "step": 89
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.057349707931280136,
+      "learning_rate": 0.00019931371771625544,
+      "loss": 1.1537,
+      "step": 90
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.05519171431660652,
+      "learning_rate": 0.00019929647369888034,
+      "loss": 1.1859,
+      "step": 91
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.05836858972907066,
+      "learning_rate": 0.00019927901648311003,
+      "loss": 1.1702,
+      "step": 92
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.058726366609334946,
+      "learning_rate": 0.00019926134610642667,
+      "loss": 1.0712,
+      "step": 93
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.06261308491230011,
+      "learning_rate": 0.0001992434626067702,
+      "loss": 1.0476,
+      "step": 94
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.05999467894434929,
+      "learning_rate": 0.00019922536602253802,
+      "loss": 1.0721,
+      "step": 95
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.06016852334141731,
+      "learning_rate": 0.00019920705639258517,
+      "loss": 1.1634,
+      "step": 96
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.055001530796289444,
+      "learning_rate": 0.00019918853375622402,
+      "loss": 1.1175,
+      "step": 97
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.05680612102150917,
+      "learning_rate": 0.00019916979815322433,
+      "loss": 1.1508,
+      "step": 98
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.061409611254930496,
+      "learning_rate": 0.0001991508496238131,
+      "loss": 1.0693,
+      "step": 99
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.06438402086496353,
+      "learning_rate": 0.00019913168820867458,
+      "loss": 1.1216,
+      "step": 100
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.06238175556063652,
+      "learning_rate": 0.00019911231394894997,
+      "loss": 1.1668,
+      "step": 101
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.057234276086091995,
+      "learning_rate": 0.00019909272688623756,
+      "loss": 1.1387,
+      "step": 102
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.07677511870861053,
+      "learning_rate": 0.00019907292706259253,
+      "loss": 1.2115,
+      "step": 103
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.059833280742168427,
+      "learning_rate": 0.00019905291452052687,
+      "loss": 1.1444,
+      "step": 104
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.05804947763681412,
+      "learning_rate": 0.00019903268930300926,
+      "loss": 1.1475,
+      "step": 105
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.06504326313734055,
+      "learning_rate": 0.0001990122514534651,
+      "loss": 1.1432,
+      "step": 106
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.057522453367710114,
+      "learning_rate": 0.00019899160101577624,
+      "loss": 1.1549,
+      "step": 107
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.06902632117271423,
+      "learning_rate": 0.00019897073803428104,
+      "loss": 1.1719,
+      "step": 108
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.05860123783349991,
+      "learning_rate": 0.00019894966255377416,
+      "loss": 1.0437,
+      "step": 109
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.05878465995192528,
+      "learning_rate": 0.00019892837461950652,
+      "loss": 1.1757,
+      "step": 110
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.05310744047164917,
+      "learning_rate": 0.00019890687427718528,
+      "loss": 1.0382,
+      "step": 111
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.05617301166057587,
+      "learning_rate": 0.00019888516157297358,
+      "loss": 1.0004,
+      "step": 112
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.05957025662064552,
+      "learning_rate": 0.00019886323655349047,
+      "loss": 1.0967,
+      "step": 113
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.05660326033830643,
+      "learning_rate": 0.00019884109926581096,
+      "loss": 1.1104,
+      "step": 114
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.0532040037214756,
+      "learning_rate": 0.00019881874975746582,
+      "loss": 1.1462,
+      "step": 115
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.054993633180856705,
+      "learning_rate": 0.00019879618807644138,
+      "loss": 1.1964,
+      "step": 116
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.06195160746574402,
+      "learning_rate": 0.00019877341427117966,
+      "loss": 1.1926,
+      "step": 117
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.0648837760090828,
+      "learning_rate": 0.00019875042839057798,
+      "loss": 1.1438,
+      "step": 118
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.055000726133584976,
+      "learning_rate": 0.00019872723048398912,
+      "loss": 1.1134,
+      "step": 119
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.06383106857538223,
+      "learning_rate": 0.000198703820601221,
+      "loss": 1.1825,
+      "step": 120
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.059399884194135666,
+      "learning_rate": 0.00019868019879253684,
+      "loss": 1.1175,
+      "step": 121
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.061064597219228745,
+      "learning_rate": 0.00019865636510865467,
+      "loss": 1.1299,
+      "step": 122
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.05886513367295265,
+      "learning_rate": 0.00019863231960074758,
+      "loss": 1.1096,
+      "step": 123
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.05704960227012634,
+      "learning_rate": 0.00019860806232044337,
+      "loss": 1.1557,
+      "step": 124
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.05626309663057327,
+      "learning_rate": 0.00019858359331982467,
+      "loss": 1.1348,
+      "step": 125
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.05755053088068962,
+      "learning_rate": 0.0001985589126514286,
+      "loss": 1.2132,
+      "step": 126
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.05787818506360054,
+      "learning_rate": 0.0001985340203682467,
+      "loss": 1.146,
+      "step": 127
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.05755281075835228,
+      "learning_rate": 0.0001985089165237249,
+      "loss": 1.1248,
+      "step": 128
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.05296372249722481,
+      "learning_rate": 0.00019848360117176352,
+      "loss": 1.1296,
+      "step": 129
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.05723942816257477,
+      "learning_rate": 0.0001984580743667168,
+      "loss": 1.0936,
+      "step": 130
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.057743217796087265,
+      "learning_rate": 0.00019843233616339306,
+      "loss": 1.1184,
+      "step": 131
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.05883733928203583,
+      "learning_rate": 0.00019840638661705454,
+      "loss": 1.1413,
+      "step": 132
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.06587976962327957,
+      "learning_rate": 0.00019838022578341725,
+      "loss": 1.1704,
+      "step": 133
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.053919676691293716,
+      "learning_rate": 0.0001983538537186508,
+      "loss": 1.08,
+      "step": 134
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.05607760325074196,
+      "learning_rate": 0.0001983272704793784,
+      "loss": 1.095,
+      "step": 135
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.055869605392217636,
+      "learning_rate": 0.00019830047612267663,
+      "loss": 1.1441,
+      "step": 136
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.058360978960990906,
+      "learning_rate": 0.00019827347070607536,
+      "loss": 1.1664,
+      "step": 137
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.0630207508802414,
+      "learning_rate": 0.0001982462542875576,
+      "loss": 1.2234,
+      "step": 138
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.05601505562663078,
+      "learning_rate": 0.0001982188269255595,
+      "loss": 1.1665,
+      "step": 139
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.06406079232692719,
+      "learning_rate": 0.00019819118867897003,
+      "loss": 1.21,
+      "step": 140
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.05720619857311249,
+      "learning_rate": 0.00019816333960713094,
+      "loss": 1.0835,
+      "step": 141
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.05414717271924019,
+      "learning_rate": 0.0001981352797698367,
+      "loss": 1.0968,
+      "step": 142
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.060023292899131775,
+      "learning_rate": 0.00019810700922733428,
+      "loss": 1.12,
+      "step": 143
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.06230263411998749,
+      "learning_rate": 0.00019807852804032305,
+      "loss": 1.2331,
+      "step": 144
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.05766064673662186,
+      "learning_rate": 0.00019804983626995468,
+      "loss": 1.1895,
+      "step": 145
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.06487365067005157,
+      "learning_rate": 0.00019802093397783296,
+      "loss": 1.116,
+      "step": 146
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.05351267755031586,
+      "learning_rate": 0.0001979918212260137,
+      "loss": 1.1481,
+      "step": 147
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.06439248472452164,
+      "learning_rate": 0.00019796249807700457,
+      "loss": 1.2206,
+      "step": 148
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.06001568213105202,
+      "learning_rate": 0.00019793296459376494,
+      "loss": 1.1055,
+      "step": 149
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.055722612887620926,
+      "learning_rate": 0.0001979032208397059,
+      "loss": 1.0998,
+      "step": 150
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.06338882446289062,
+      "learning_rate": 0.00019787326687868993,
+      "loss": 1.0981,
+      "step": 151
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.053316015750169754,
+      "learning_rate": 0.00019784310277503085,
+      "loss": 1.0689,
+      "step": 152
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.055426474660634995,
+      "learning_rate": 0.00019781272859349368,
+      "loss": 1.1028,
+      "step": 153
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.08009687811136246,
+      "learning_rate": 0.00019778214439929452,
+      "loss": 1.1934,
+      "step": 154
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.06281601637601852,
+      "learning_rate": 0.00019775135025810035,
+      "loss": 1.1871,
+      "step": 155
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.05907437577843666,
+      "learning_rate": 0.00019772034623602894,
+      "loss": 1.1801,
+      "step": 156
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.05395223945379257,
+      "learning_rate": 0.00019768913239964872,
+      "loss": 1.1755,
+      "step": 157
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.05565226450562477,
+      "learning_rate": 0.00019765770881597855,
+      "loss": 1.1147,
+      "step": 158
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.06353943049907684,
+      "learning_rate": 0.0001976260755524877,
+      "loss": 1.148,
+      "step": 159
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.056039877235889435,
+      "learning_rate": 0.00019759423267709555,
+      "loss": 1.1285,
+      "step": 160
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.06256404519081116,
+      "learning_rate": 0.00019756218025817165,
+      "loss": 1.1148,
+      "step": 161
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.06320231407880783,
+      "learning_rate": 0.00019752991836453543,
+      "loss": 1.1897,
+      "step": 162
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.06373140960931778,
+      "learning_rate": 0.00019749744706545598,
+      "loss": 1.147,
+      "step": 163
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.061453696340322495,
+      "learning_rate": 0.00019746476643065216,
+      "loss": 1.1333,
+      "step": 164
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.06347192078828812,
+      "learning_rate": 0.00019743187653029214,
+      "loss": 1.1834,
+      "step": 165
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.058654770255088806,
+      "learning_rate": 0.00019739877743499352,
+      "loss": 1.1774,
+      "step": 166
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.05696241557598114,
+      "learning_rate": 0.000197365469215823,
+      "loss": 1.1321,
+      "step": 167
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.057179030030965805,
+      "learning_rate": 0.00019733195194429628,
+      "loss": 1.0855,
+      "step": 168
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.058308880776166916,
+      "learning_rate": 0.00019729822569237797,
+      "loss": 1.0362,
+      "step": 169
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.06666514277458191,
+      "learning_rate": 0.0001972642905324813,
+      "loss": 1.1118,
+      "step": 170
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.05894763022661209,
+      "learning_rate": 0.00019723014653746815,
+      "loss": 1.1454,
+      "step": 171
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.06241476908326149,
+      "learning_rate": 0.00019719579378064869,
+      "loss": 1.1657,
+      "step": 172
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.05953037366271019,
+      "learning_rate": 0.00019716123233578134,
+      "loss": 1.1378,
+      "step": 173
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.05746786668896675,
+      "learning_rate": 0.00019712646227707263,
+      "loss": 1.1115,
+      "step": 174
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.05923418328166008,
+      "learning_rate": 0.00019709148367917697,
+      "loss": 1.0896,
+      "step": 175
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.05635738745331764,
+      "learning_rate": 0.00019705629661719652,
+      "loss": 1.1049,
+      "step": 176
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.05885479226708412,
+      "learning_rate": 0.0001970209011666811,
+      "loss": 1.1252,
+      "step": 177
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.057710710912942886,
+      "learning_rate": 0.00019698529740362785,
+      "loss": 1.1097,
+      "step": 178
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.06445460021495819,
+      "learning_rate": 0.00019694948540448123,
+      "loss": 1.2045,
+      "step": 179
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.05737968161702156,
+      "learning_rate": 0.00019691346524613286,
+      "loss": 1.1544,
+      "step": 180
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 1.0905169248580933,
+      "eval_runtime": 97.1012,
+      "eval_samples_per_second": 51.493,
+      "eval_steps_per_second": 12.873,
+      "step": 180
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.06000959873199463,
+      "learning_rate": 0.00019687723700592116,
+      "loss": 1.14,
+      "step": 181
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.06186012923717499,
+      "learning_rate": 0.00019684080076163142,
+      "loss": 1.1284,
+      "step": 182
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.061752576380968094,
+      "learning_rate": 0.00019680415659149554,
+      "loss": 1.1719,
+      "step": 183
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.05509311705827713,
+      "learning_rate": 0.00019676730457419178,
+      "loss": 1.1155,
+      "step": 184
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.06091266870498657,
+      "learning_rate": 0.00019673024478884473,
+      "loss": 1.0099,
+      "step": 185
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.065452940762043,
+      "learning_rate": 0.00019669297731502507,
+      "loss": 1.1272,
+      "step": 186
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.05998782813549042,
+      "learning_rate": 0.00019665550223274937,
+      "loss": 1.0957,
+      "step": 187
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.06403583288192749,
+      "learning_rate": 0.00019661781962248003,
+      "loss": 1.137,
+      "step": 188
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.06362204998731613,
+      "learning_rate": 0.0001965799295651249,
+      "loss": 1.1025,
+      "step": 189
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.06712298840284348,
+      "learning_rate": 0.0001965418321420374,
+      "loss": 1.1069,
+      "step": 190
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.059146981686353683,
+      "learning_rate": 0.00019650352743501604,
+      "loss": 1.1502,
+      "step": 191
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.0600997731089592,
+      "learning_rate": 0.00019646501552630444,
+      "loss": 1.1693,
+      "step": 192
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.06209671124815941,
+      "learning_rate": 0.0001964262964985912,
+      "loss": 1.1384,
+      "step": 193
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.061546314507722855,
+      "learning_rate": 0.0001963873704350094,
+      "loss": 1.1485,
+      "step": 194
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.06550047546625137,
+      "learning_rate": 0.00019634823741913687,
+      "loss": 1.1136,
+      "step": 195
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.05800144001841545,
+      "learning_rate": 0.0001963088975349956,
+      "loss": 1.1783,
+      "step": 196
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.06252887099981308,
+      "learning_rate": 0.00019626935086705194,
+      "loss": 1.1337,
+      "step": 197
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.06455714255571365,
+      "learning_rate": 0.00019622959750021605,
+      "loss": 1.1107,
+      "step": 198
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.06395559012889862,
+      "learning_rate": 0.00019618963751984195,
+      "loss": 1.0746,
+      "step": 199
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.05962133780121803,
+      "learning_rate": 0.00019614947101172732,
+      "loss": 1.2018,
+      "step": 200
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.06970924139022827,
+      "learning_rate": 0.00019610909806211323,
+      "loss": 1.2098,
+      "step": 201
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.06366313248872757,
+      "learning_rate": 0.000196068518757684,
+      "loss": 1.0914,
+      "step": 202
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.05959942191839218,
+      "learning_rate": 0.00019602773318556707,
+      "loss": 1.1557,
+      "step": 203
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.06882765144109726,
+      "learning_rate": 0.00019598674143333263,
+      "loss": 1.2375,
+      "step": 204
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.06210785359144211,
+      "learning_rate": 0.00019594554358899366,
+      "loss": 1.1748,
+      "step": 205
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.06589924544095993,
+      "learning_rate": 0.0001959041397410056,
+      "loss": 1.1242,
+      "step": 206
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.05691949278116226,
+      "learning_rate": 0.00019586252997826626,
+      "loss": 1.1532,
+      "step": 207
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.061684492975473404,
+      "learning_rate": 0.00019582071439011546,
+      "loss": 1.1318,
+      "step": 208
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.06065653637051582,
+      "learning_rate": 0.00019577869306633503,
+      "loss": 1.1281,
+      "step": 209
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.057913534343242645,
+      "learning_rate": 0.0001957364660971485,
+      "loss": 1.1318,
+      "step": 210
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.06302294880151749,
+      "learning_rate": 0.0001956940335732209,
+      "loss": 1.1971,
+      "step": 211
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.0630715936422348,
+      "learning_rate": 0.0001956513955856587,
+      "loss": 1.215,
+      "step": 212
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.067079097032547,
+      "learning_rate": 0.00019560855222600947,
+      "loss": 1.1484,
+      "step": 213
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.05915733054280281,
+      "learning_rate": 0.0001955655035862617,
+      "loss": 1.0748,
+      "step": 214
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.07278767973184586,
+      "learning_rate": 0.00019552224975884466,
+      "loss": 1.0966,
+      "step": 215
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.05778686702251434,
+      "learning_rate": 0.00019547879083662819,
+      "loss": 1.0561,
+      "step": 216
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.06521858274936676,
+      "learning_rate": 0.00019543512691292251,
+      "loss": 1.1249,
+      "step": 217
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.058512818068265915,
+      "learning_rate": 0.0001953912580814779,
+      "loss": 1.0463,
+      "step": 218
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.06136133521795273,
+      "learning_rate": 0.00019534718443648473,
+      "loss": 1.1914,
+      "step": 219
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.05871862918138504,
+      "learning_rate": 0.000195302906072573,
+      "loss": 1.0887,
+      "step": 220
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.05922750383615494,
+      "learning_rate": 0.00019525842308481227,
+      "loss": 1.0888,
+      "step": 221
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.06273704022169113,
+      "learning_rate": 0.0001952137355687116,
+      "loss": 1.1218,
+      "step": 222
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.05759681016206741,
+      "learning_rate": 0.000195168843620219,
+      "loss": 1.1278,
+      "step": 223
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.060312505811452866,
+      "learning_rate": 0.00019512374733572153,
+      "loss": 1.1688,
+      "step": 224
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.06145263463258743,
+      "learning_rate": 0.0001950784468120449,
+      "loss": 1.1055,
+      "step": 225
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.06581958383321762,
+      "learning_rate": 0.00019503294214645337,
+      "loss": 1.2683,
+      "step": 226
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.06561966240406036,
+      "learning_rate": 0.00019498723343664955,
+      "loss": 1.2038,
+      "step": 227
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.06790629774332047,
+      "learning_rate": 0.00019494132078077414,
+      "loss": 1.1307,
+      "step": 228
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.06045442819595337,
+      "learning_rate": 0.0001948952042774057,
+      "loss": 1.1119,
+      "step": 229
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.06775711476802826,
+      "learning_rate": 0.00019484888402556045,
+      "loss": 1.1636,
+      "step": 230
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.06277008354663849,
+      "learning_rate": 0.00019480236012469213,
+      "loss": 1.1908,
+      "step": 231
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.0631483793258667,
+      "learning_rate": 0.00019475563267469173,
+      "loss": 1.1828,
+      "step": 232
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.06697548180818558,
+      "learning_rate": 0.0001947087017758872,
+      "loss": 1.1107,
+      "step": 233
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.06013527512550354,
+      "learning_rate": 0.00019466156752904343,
+      "loss": 1.1176,
+      "step": 234
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.06709557771682739,
+      "learning_rate": 0.00019461423003536182,
+      "loss": 1.1575,
+      "step": 235
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.057948894798755646,
+      "learning_rate": 0.0001945666893964802,
+      "loss": 1.0455,
+      "step": 236
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.057657647877931595,
+      "learning_rate": 0.00019451894571447258,
+      "loss": 1.1221,
+      "step": 237
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.06317263096570969,
+      "learning_rate": 0.0001944709990918489,
+      "loss": 1.1305,
+      "step": 238
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.06375168263912201,
+      "learning_rate": 0.00019442284963155487,
+      "loss": 1.206,
+      "step": 239
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.06475285440683365,
+      "learning_rate": 0.00019437449743697164,
+      "loss": 1.204,
+      "step": 240
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.057945527136325836,
+      "learning_rate": 0.00019432594261191568,
+      "loss": 0.9143,
+      "step": 241
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.05976686626672745,
+      "learning_rate": 0.00019427718526063856,
+      "loss": 1.1177,
+      "step": 242
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.059271346777677536,
+      "learning_rate": 0.00019422822548782666,
+      "loss": 1.1089,
+      "step": 243
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.05958615615963936,
+      "learning_rate": 0.00019417906339860098,
+      "loss": 1.1451,
+      "step": 244
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.06455380469560623,
+      "learning_rate": 0.00019412969909851695,
+      "loss": 1.2423,
+      "step": 245
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.0598178505897522,
+      "learning_rate": 0.00019408013269356408,
+      "loss": 1.081,
+      "step": 246
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.06671278178691864,
+      "learning_rate": 0.0001940303642901659,
+      "loss": 1.1094,
+      "step": 247
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.06148644909262657,
+      "learning_rate": 0.0001939803939951796,
+      "loss": 1.1045,
+      "step": 248
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.06747152656316757,
+      "learning_rate": 0.00019393022191589588,
+      "loss": 1.1443,
+      "step": 249
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.060876958072185516,
+      "learning_rate": 0.00019387984816003867,
+      "loss": 1.1341,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.05667860805988312,
+      "learning_rate": 0.00019382927283576488,
+      "loss": 1.2032,
+      "step": 251
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.06507498770952225,
+      "learning_rate": 0.0001937784960516643,
+      "loss": 1.1355,
+      "step": 252
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.05810081586241722,
+      "learning_rate": 0.0001937275179167592,
+      "loss": 1.0911,
+      "step": 253
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.06210170313715935,
+      "learning_rate": 0.00019367633854050422,
+      "loss": 1.2247,
+      "step": 254
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.060430753976106644,
+      "learning_rate": 0.00019362495803278598,
+      "loss": 1.1633,
+      "step": 255
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.0607365183532238,
+      "learning_rate": 0.0001935733765039231,
+      "loss": 1.1253,
+      "step": 256
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.062377460300922394,
+      "learning_rate": 0.00019352159406466564,
+      "loss": 1.1525,
+      "step": 257
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.06079326942563057,
+      "learning_rate": 0.00019346961082619522,
+      "loss": 1.067,
+      "step": 258
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.06413824111223221,
+      "learning_rate": 0.00019341742690012442,
+      "loss": 1.0536,
+      "step": 259
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.06453590095043182,
+      "learning_rate": 0.00019336504239849677,
+      "loss": 1.1118,
+      "step": 260
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.06608355045318604,
+      "learning_rate": 0.00019331245743378658,
+      "loss": 1.2306,
+      "step": 261
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.06034080311655998,
+      "learning_rate": 0.00019325967211889834,
+      "loss": 1.0643,
+      "step": 262
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.06985899806022644,
+      "learning_rate": 0.0001932066865671669,
+      "loss": 1.1907,
+      "step": 263
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.06518577039241791,
+      "learning_rate": 0.000193153500892357,
+      "loss": 1.0499,
+      "step": 264
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.05966721102595329,
+      "learning_rate": 0.00019310011520866298,
+      "loss": 1.2111,
+      "step": 265
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.0628088116645813,
+      "learning_rate": 0.0001930465296307087,
+      "loss": 1.2063,
+      "step": 266
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.06214950606226921,
+      "learning_rate": 0.00019299274427354713,
+      "loss": 1.1141,
+      "step": 267
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.05994017794728279,
+      "learning_rate": 0.00019293875925266028,
+      "loss": 1.0661,
+      "step": 268
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.0646560862660408,
+      "learning_rate": 0.0001928845746839588,
+      "loss": 1.1208,
+      "step": 269
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.0596354603767395,
+      "learning_rate": 0.00019283019068378182,
+      "loss": 1.0839,
+      "step": 270
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.06729655712842941,
+      "learning_rate": 0.00019277560736889656,
+      "loss": 1.1983,
+      "step": 271
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.060334715992212296,
+      "learning_rate": 0.0001927208248564984,
+      "loss": 1.1733,
+      "step": 272
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.06451458483934402,
+      "learning_rate": 0.0001926658432642102,
+      "loss": 1.1861,
+      "step": 273
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.0642094537615776,
+      "learning_rate": 0.00019261066271008235,
+      "loss": 1.2082,
+      "step": 274
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.06414438039064407,
+      "learning_rate": 0.00019255528331259245,
+      "loss": 1.2229,
+      "step": 275
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.06006895750761032,
+      "learning_rate": 0.00019249970519064503,
+      "loss": 1.1456,
+      "step": 276
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.06294744461774826,
+      "learning_rate": 0.00019244392846357123,
+      "loss": 1.072,
+      "step": 277
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.06234186887741089,
+      "learning_rate": 0.0001923879532511287,
+      "loss": 1.0956,
+      "step": 278
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.06028866767883301,
+      "learning_rate": 0.00019233177967350117,
+      "loss": 1.1874,
+      "step": 279
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.06322412937879562,
+      "learning_rate": 0.0001922754078512984,
+      "loss": 1.0697,
+      "step": 280
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.059378113597631454,
+      "learning_rate": 0.00019221883790555567,
+      "loss": 1.1663,
+      "step": 281
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.062111396342515945,
+      "learning_rate": 0.00019216206995773373,
+      "loss": 1.1077,
+      "step": 282
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.06687445938587189,
+      "learning_rate": 0.00019210510412971844,
+      "loss": 1.193,
+      "step": 283
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.06198667362332344,
+      "learning_rate": 0.00019204794054382052,
+      "loss": 1.2258,
+      "step": 284
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.05913461372256279,
+      "learning_rate": 0.00019199057932277525,
+      "loss": 1.141,
+      "step": 285
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.06464271247386932,
+      "learning_rate": 0.00019193302058974232,
+      "loss": 1.0863,
+      "step": 286
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.05979037657380104,
+      "learning_rate": 0.0001918752644683055,
+      "loss": 1.1686,
+      "step": 287
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.06707989424467087,
+      "learning_rate": 0.00019181731108247228,
+      "loss": 1.2413,
+      "step": 288
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.06815943866968155,
+      "learning_rate": 0.0001917591605566738,
+      "loss": 1.2022,
+      "step": 289
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.06593851000070572,
+      "learning_rate": 0.00019170081301576444,
+      "loss": 1.0464,
+      "step": 290
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.06178038939833641,
+      "learning_rate": 0.00019164226858502153,
+      "loss": 1.1011,
+      "step": 291
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.06253184378147125,
+      "learning_rate": 0.00019158352739014523,
+      "loss": 1.0302,
+      "step": 292
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.06402178853750229,
+      "learning_rate": 0.0001915245895572581,
+      "loss": 1.1971,
+      "step": 293
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.0673704445362091,
+      "learning_rate": 0.00019146545521290495,
+      "loss": 1.1555,
+      "step": 294
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.06606754660606384,
+      "learning_rate": 0.00019140612448405252,
+      "loss": 1.1057,
+      "step": 295
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.060698218643665314,
+      "learning_rate": 0.00019134659749808913,
+      "loss": 1.1279,
+      "step": 296
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.06506410986185074,
+      "learning_rate": 0.00019128687438282457,
+      "loss": 1.1735,
+      "step": 297
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.061889030039310455,
+      "learning_rate": 0.00019122695526648968,
+      "loss": 1.1158,
+      "step": 298
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.06519250571727753,
+      "learning_rate": 0.0001911668402777362,
+      "loss": 1.1934,
+      "step": 299
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.06114347651600838,
+      "learning_rate": 0.00019110652954563631,
+      "loss": 1.1105,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.06125333905220032,
+      "learning_rate": 0.0001910460231996826,
+      "loss": 1.0982,
+      "step": 301
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.06343477964401245,
+      "learning_rate": 0.00019098532136978754,
+      "loss": 1.1342,
+      "step": 302
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.060326121747493744,
+      "learning_rate": 0.00019092442418628343,
+      "loss": 1.1897,
+      "step": 303
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.06917954236268997,
+      "learning_rate": 0.00019086333177992191,
+      "loss": 1.0707,
+      "step": 304
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.06384435296058655,
+      "learning_rate": 0.00019080204428187388,
+      "loss": 1.0565,
+      "step": 305
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.06388357281684875,
+      "learning_rate": 0.00019074056182372907,
+      "loss": 1.1176,
+      "step": 306
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.0598420649766922,
+      "learning_rate": 0.00019067888453749575,
+      "loss": 1.1335,
+      "step": 307
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.05912897735834122,
+      "learning_rate": 0.0001906170125556006,
+      "loss": 1.0945,
+      "step": 308
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.0659290999174118,
+      "learning_rate": 0.00019055494601088834,
+      "loss": 1.1361,
+      "step": 309
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.0634288564324379,
+      "learning_rate": 0.00019049268503662126,
+      "loss": 1.1754,
+      "step": 310
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.06105343624949455,
+      "learning_rate": 0.00019043022976647934,
+      "loss": 1.1168,
+      "step": 311
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.06428004801273346,
+      "learning_rate": 0.00019036758033455956,
+      "loss": 1.0972,
+      "step": 312
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.061880819499492645,
+      "learning_rate": 0.0001903047368753759,
+      "loss": 1.1545,
+      "step": 313
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.07274632155895233,
+      "learning_rate": 0.00019024169952385885,
+      "loss": 1.1257,
+      "step": 314
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.06646310538053513,
+      "learning_rate": 0.00019017846841535522,
+      "loss": 1.154,
+      "step": 315
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.06078438088297844,
+      "learning_rate": 0.00019011504368562782,
+      "loss": 1.1303,
+      "step": 316
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.07181931287050247,
+      "learning_rate": 0.00019005142547085527,
+      "loss": 1.1603,
+      "step": 317
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.06724654138088226,
+      "learning_rate": 0.00018998761390763154,
+      "loss": 1.1718,
+      "step": 318
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.05919702351093292,
+      "learning_rate": 0.00018992360913296574,
+      "loss": 1.142,
+      "step": 319
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.06586825847625732,
+      "learning_rate": 0.00018985941128428185,
+      "loss": 1.1859,
+      "step": 320
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.06922873854637146,
+      "learning_rate": 0.00018979502049941833,
+      "loss": 1.1,
+      "step": 321
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.06558050960302353,
+      "learning_rate": 0.00018973043691662803,
+      "loss": 1.1329,
+      "step": 322
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.06796015799045563,
+      "learning_rate": 0.0001896656606745776,
+      "loss": 1.1413,
+      "step": 323
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.06253591179847717,
+      "learning_rate": 0.00018960069191234746,
+      "loss": 1.1416,
+      "step": 324
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.07260395586490631,
+      "learning_rate": 0.00018953553076943134,
+      "loss": 1.2171,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.06394602358341217,
+      "learning_rate": 0.000189470177385736,
+      "loss": 1.1857,
+      "step": 326
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.06450013071298599,
+      "learning_rate": 0.000189404631901581,
+      "loss": 1.1278,
+      "step": 327
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.06406421959400177,
+      "learning_rate": 0.00018933889445769836,
+      "loss": 1.2068,
+      "step": 328
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.06315187364816666,
+      "learning_rate": 0.00018927296519523226,
+      "loss": 1.0789,
+      "step": 329
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.06743728369474411,
+      "learning_rate": 0.00018920684425573865,
+      "loss": 1.1594,
+      "step": 330
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.06548736989498138,
+      "learning_rate": 0.0001891405317811852,
+      "loss": 1.0963,
+      "step": 331
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.06221878528594971,
+      "learning_rate": 0.00018907402791395057,
+      "loss": 1.1802,
+      "step": 332
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.06024456024169922,
+      "learning_rate": 0.00018900733279682462,
+      "loss": 1.1534,
+      "step": 333
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.0619971826672554,
+      "learning_rate": 0.00018894044657300765,
+      "loss": 1.1287,
+      "step": 334
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.061270926147699356,
+      "learning_rate": 0.0001888733693861104,
+      "loss": 1.1306,
+      "step": 335
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.06278882175683975,
+      "learning_rate": 0.00018880610138015356,
+      "loss": 1.0748,
+      "step": 336
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.061774447560310364,
+      "learning_rate": 0.00018873864269956756,
+      "loss": 1.083,
+      "step": 337
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.06414585560560226,
+      "learning_rate": 0.00018867099348919217,
+      "loss": 1.1351,
+      "step": 338
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.06463301926851273,
+      "learning_rate": 0.00018860315389427633,
+      "loss": 1.1633,
+      "step": 339
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.06183488667011261,
+      "learning_rate": 0.00018853512406047772,
+      "loss": 1.0311,
+      "step": 340
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.06121757999062538,
+      "learning_rate": 0.00018846690413386238,
+      "loss": 1.1787,
+      "step": 341
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.06119801104068756,
+      "learning_rate": 0.0001883984942609047,
+      "loss": 1.1699,
+      "step": 342
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.06786572933197021,
+      "learning_rate": 0.0001883298945884867,
+      "loss": 1.1863,
+      "step": 343
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.06401295214891434,
+      "learning_rate": 0.00018826110526389803,
+      "loss": 1.1483,
+      "step": 344
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.06333799660205841,
+      "learning_rate": 0.0001881921264348355,
+      "loss": 1.0793,
+      "step": 345
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.06162017583847046,
+      "learning_rate": 0.00018812295824940285,
+      "loss": 1.1106,
+      "step": 346
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.05900746211409569,
+      "learning_rate": 0.00018805360085611032,
+      "loss": 1.1257,
+      "step": 347
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.06269056349992752,
+      "learning_rate": 0.00018798405440387445,
+      "loss": 1.1076,
+      "step": 348
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.06586477160453796,
+      "learning_rate": 0.00018791431904201765,
+      "loss": 1.0705,
+      "step": 349
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.07007557153701782,
+      "learning_rate": 0.00018784439492026798,
+      "loss": 1.2659,
+      "step": 350
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.06069363281130791,
+      "learning_rate": 0.00018777428218875878,
+      "loss": 1.0526,
+      "step": 351
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.06026845797896385,
+      "learning_rate": 0.00018770398099802836,
+      "loss": 1.1309,
+      "step": 352
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.061220645904541016,
+      "learning_rate": 0.0001876334914990196,
+      "loss": 1.0999,
+      "step": 353
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.06346994638442993,
+      "learning_rate": 0.00018756281384307982,
+      "loss": 1.0391,
+      "step": 354
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.06735444813966751,
+      "learning_rate": 0.00018749194818196022,
+      "loss": 1.1974,
+      "step": 355
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.06252043694257736,
+      "learning_rate": 0.0001874208946678157,
+      "loss": 1.1771,
+      "step": 356
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.06052526459097862,
+      "learning_rate": 0.0001873496534532045,
+      "loss": 1.1878,
+      "step": 357
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.06671708077192307,
+      "learning_rate": 0.0001872782246910879,
+      "loss": 1.1576,
+      "step": 358
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.06357080489397049,
+      "learning_rate": 0.00018720660853482977,
+      "loss": 1.164,
+      "step": 359
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.06754573434591293,
+      "learning_rate": 0.00018713480513819644,
+      "loss": 1.1932,
+      "step": 360
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 1.0889657735824585,
+      "eval_runtime": 97.0368,
+      "eval_samples_per_second": 51.527,
+      "eval_steps_per_second": 12.882,
+      "step": 360
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.06117720529437065,
+      "learning_rate": 0.0001870628146553562,
+      "loss": 1.0225,
+      "step": 361
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.06431601196527481,
+      "learning_rate": 0.00018699063724087904,
+      "loss": 1.1954,
+      "step": 362
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.06532489508390427,
+      "learning_rate": 0.0001869182730497363,
+      "loss": 1.1475,
+      "step": 363
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.06369902938604355,
+      "learning_rate": 0.00018684572223730045,
+      "loss": 1.0935,
+      "step": 364
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.06582259386777878,
+      "learning_rate": 0.0001867729849593444,
+      "loss": 1.163,
+      "step": 365
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.06448414921760559,
+      "learning_rate": 0.0001867000613720417,
+      "loss": 1.1137,
+      "step": 366
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.0670081228017807,
+      "learning_rate": 0.00018662695163196579,
+      "loss": 1.1921,
+      "step": 367
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.06068355217576027,
+      "learning_rate": 0.0001865536558960898,
+      "loss": 0.9705,
+      "step": 368
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.06467951089143753,
+      "learning_rate": 0.0001864801743217862,
+      "loss": 1.0868,
+      "step": 369
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.06502547860145569,
+      "learning_rate": 0.0001864065070668265,
+      "loss": 1.0937,
+      "step": 370
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.07280650734901428,
+      "learning_rate": 0.00018633265428938088,
+      "loss": 1.1701,
+      "step": 371
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.05973858758807182,
+      "learning_rate": 0.00018625861614801785,
+      "loss": 1.0324,
+      "step": 372
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.06514280289411545,
+      "learning_rate": 0.00018618439280170392,
+      "loss": 1.1338,
+      "step": 373
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.06471429020166397,
+      "learning_rate": 0.00018610998440980324,
+      "loss": 1.1876,
+      "step": 374
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.0652419924736023,
+      "learning_rate": 0.00018603539113207722,
+      "loss": 1.2204,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.06470503658056259,
+      "learning_rate": 0.0001859606131286843,
+      "loss": 1.2483,
+      "step": 376
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.06816300749778748,
+      "learning_rate": 0.00018588565056017965,
+      "loss": 1.1036,
+      "step": 377
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.062083158642053604,
+      "learning_rate": 0.00018581050358751445,
+      "loss": 1.1232,
+      "step": 378
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.06199049949645996,
+      "learning_rate": 0.00018573517237203602,
+      "loss": 1.0623,
+      "step": 379
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.06976150721311569,
+      "learning_rate": 0.0001856596570754872,
+      "loss": 1.0361,
+      "step": 380
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.06474665552377701,
+      "learning_rate": 0.0001855839578600061,
+      "loss": 1.1786,
+      "step": 381
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.06963939964771271,
+      "learning_rate": 0.00018550807488812562,
+      "loss": 1.1381,
+      "step": 382
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.06306161731481552,
+      "learning_rate": 0.00018543200832277337,
+      "loss": 1.1783,
+      "step": 383
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.06826300919055939,
+      "learning_rate": 0.00018535575832727102,
+      "loss": 1.1401,
+      "step": 384
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.061530325561761856,
+      "learning_rate": 0.0001852793250653341,
+      "loss": 1.0992,
+      "step": 385
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.06267255544662476,
+      "learning_rate": 0.00018520270870107166,
+      "loss": 1.0652,
+      "step": 386
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.059738241136074066,
+      "learning_rate": 0.0001851259093989859,
+      "loss": 1.1092,
+      "step": 387
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.062138039618730545,
+      "learning_rate": 0.00018504892732397173,
+      "loss": 1.0607,
+      "step": 388
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.06500416994094849,
+      "learning_rate": 0.00018497176264131656,
+      "loss": 1.0877,
+      "step": 389
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.06768455356359482,
+      "learning_rate": 0.00018489441551669986,
+      "loss": 1.2099,
+      "step": 390
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.06814403086900711,
+      "learning_rate": 0.00018481688611619285,
+      "loss": 1.1743,
+      "step": 391
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.06558533012866974,
+      "learning_rate": 0.00018473917460625798,
+      "loss": 1.0566,
+      "step": 392
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.06515612453222275,
+      "learning_rate": 0.00018466128115374888,
+      "loss": 1.1267,
+      "step": 393
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.06616493314504623,
+      "learning_rate": 0.00018458320592590975,
+      "loss": 1.1703,
+      "step": 394
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.06906623393297195,
+      "learning_rate": 0.00018450494909037502,
+      "loss": 1.2327,
+      "step": 395
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.06478845328092575,
+      "learning_rate": 0.00018442651081516917,
+      "loss": 1.1189,
+      "step": 396
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.0638127252459526,
+      "learning_rate": 0.00018434789126870612,
+      "loss": 1.1117,
+      "step": 397
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.06424181908369064,
+      "learning_rate": 0.00018426909061978908,
+      "loss": 1.1469,
+      "step": 398
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.06596982479095459,
+      "learning_rate": 0.0001841901090376101,
+      "loss": 1.1003,
+      "step": 399
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.06202305108308792,
+      "learning_rate": 0.00018411094669174965,
+      "loss": 1.0425,
+      "step": 400
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.0672823116183281,
+      "learning_rate": 0.00018403160375217637,
+      "loss": 1.1015,
+      "step": 401
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.06283591687679291,
+      "learning_rate": 0.00018395208038924667,
+      "loss": 1.0863,
+      "step": 402
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.06623289734125137,
+      "learning_rate": 0.00018387237677370427,
+      "loss": 1.0828,
+      "step": 403
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.06633896380662918,
+      "learning_rate": 0.00018379249307667994,
+      "loss": 1.145,
+      "step": 404
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.06487153470516205,
+      "learning_rate": 0.00018371242946969117,
+      "loss": 1.1447,
+      "step": 405
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.06051245704293251,
+      "learning_rate": 0.00018363218612464158,
+      "loss": 1.0637,
+      "step": 406
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.05917385593056679,
+      "learning_rate": 0.00018355176321382087,
+      "loss": 1.1459,
+      "step": 407
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.061461642384529114,
+      "learning_rate": 0.00018347116090990424,
+      "loss": 1.0601,
+      "step": 408
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.06090409681200981,
+      "learning_rate": 0.00018339037938595193,
+      "loss": 1.0531,
+      "step": 409
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.0663984939455986,
+      "learning_rate": 0.00018330941881540915,
+      "loss": 1.1315,
+      "step": 410
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.06271735578775406,
+      "learning_rate": 0.0001832282793721055,
+      "loss": 1.0869,
+      "step": 411
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.06272199004888535,
+      "learning_rate": 0.00018314696123025454,
+      "loss": 1.1414,
+      "step": 412
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.06375264376401901,
+      "learning_rate": 0.0001830654645644536,
+      "loss": 1.0946,
+      "step": 413
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.06219152733683586,
+      "learning_rate": 0.00018298378954968337,
+      "loss": 1.0448,
+      "step": 414
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.06392863392829895,
+      "learning_rate": 0.00018290193636130727,
+      "loss": 1.1606,
+      "step": 415
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.06729146838188171,
+      "learning_rate": 0.00018281990517507156,
+      "loss": 1.1532,
+      "step": 416
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.0799422636628151,
+      "learning_rate": 0.00018273769616710436,
+      "loss": 1.1133,
+      "step": 417
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.06902763992547989,
+      "learning_rate": 0.0001826553095139159,
+      "loss": 1.1404,
+      "step": 418
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.06674634665250778,
+      "learning_rate": 0.00018257274539239756,
+      "loss": 1.2019,
+      "step": 419
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.06378789991140366,
+      "learning_rate": 0.00018249000397982195,
+      "loss": 1.1069,
+      "step": 420
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.06818845123052597,
+      "learning_rate": 0.0001824070854538422,
+      "loss": 1.1844,
+      "step": 421
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.07562380284070969,
+      "learning_rate": 0.00018232398999249192,
+      "loss": 1.0978,
+      "step": 422
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.07126368582248688,
+      "learning_rate": 0.00018224071777418437,
+      "loss": 1.2169,
+      "step": 423
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.06599973887205124,
+      "learning_rate": 0.00018215726897771248,
+      "loss": 1.1339,
+      "step": 424
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.06484373658895493,
+      "learning_rate": 0.00018207364378224826,
+      "loss": 1.1609,
+      "step": 425
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.062379587441682816,
+      "learning_rate": 0.00018198984236734246,
+      "loss": 1.1661,
+      "step": 426
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.0662432610988617,
+      "learning_rate": 0.00018190586491292424,
+      "loss": 1.1372,
+      "step": 427
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.06816010177135468,
+      "learning_rate": 0.00018182171159930065,
+      "loss": 1.1678,
+      "step": 428
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.0649806335568428,
+      "learning_rate": 0.0001817373826071564,
+      "loss": 1.1688,
+      "step": 429
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.06410276144742966,
+      "learning_rate": 0.0001816528781175533,
+      "loss": 1.1723,
+      "step": 430
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.06425531208515167,
+      "learning_rate": 0.00018156819831193012,
+      "loss": 1.148,
+      "step": 431
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.06638933718204498,
+      "learning_rate": 0.00018148334337210193,
+      "loss": 1.0427,
+      "step": 432
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.06576802581548691,
+      "learning_rate": 0.00018139831348025988,
+      "loss": 1.0814,
+      "step": 433
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.06729135662317276,
+      "learning_rate": 0.0001813131088189707,
+      "loss": 1.2238,
+      "step": 434
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.066316619515419,
+      "learning_rate": 0.00018122772957117645,
+      "loss": 1.124,
+      "step": 435
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.0657489225268364,
+      "learning_rate": 0.00018114217592019393,
+      "loss": 1.1616,
+      "step": 436
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.06329985707998276,
+      "learning_rate": 0.00018105644804971454,
+      "loss": 1.1662,
+      "step": 437
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.06630551815032959,
+      "learning_rate": 0.00018097054614380365,
+      "loss": 1.0863,
+      "step": 438
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.0653151199221611,
+      "learning_rate": 0.00018088447038690033,
+      "loss": 1.0568,
+      "step": 439
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.06651636958122253,
+      "learning_rate": 0.00018079822096381688,
+      "loss": 1.1219,
+      "step": 440
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.06870090961456299,
+      "learning_rate": 0.00018071179805973856,
+      "loss": 1.1327,
+      "step": 441
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.06400321424007416,
+      "learning_rate": 0.000180625201860223,
+      "loss": 1.0736,
+      "step": 442
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.06521004438400269,
+      "learning_rate": 0.00018053843255120002,
+      "loss": 1.0366,
+      "step": 443
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.0785718560218811,
+      "learning_rate": 0.0001804514903189711,
+      "loss": 1.0535,
+      "step": 444
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.0732305571436882,
+      "learning_rate": 0.0001803643753502089,
+      "loss": 1.0828,
+      "step": 445
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.06389783322811127,
+      "learning_rate": 0.0001802770878319571,
+      "loss": 1.0966,
+      "step": 446
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.06395277380943298,
+      "learning_rate": 0.00018018962795162977,
+      "loss": 1.1414,
+      "step": 447
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.06490381062030792,
+      "learning_rate": 0.00018010199589701107,
+      "loss": 1.1405,
+      "step": 448
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.06930892169475555,
+      "learning_rate": 0.00018001419185625488,
+      "loss": 1.0145,
+      "step": 449
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.07346969097852707,
+      "learning_rate": 0.00017992621601788428,
+      "loss": 1.1745,
+      "step": 450
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.07306042313575745,
+      "learning_rate": 0.0001798380685707913,
+      "loss": 1.1333,
+      "step": 451
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.06085991859436035,
+      "learning_rate": 0.00017974974970423632,
+      "loss": 1.0801,
+      "step": 452
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.0631103590130806,
+      "learning_rate": 0.00017966125960784787,
+      "loss": 1.1209,
+      "step": 453
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.06399189680814743,
+      "learning_rate": 0.00017957259847162205,
+      "loss": 1.1683,
+      "step": 454
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.06143182888627052,
+      "learning_rate": 0.00017948376648592232,
+      "loss": 1.0855,
+      "step": 455
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.06295006722211838,
+      "learning_rate": 0.00017939476384147877,
+      "loss": 1.1398,
+      "step": 456
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.06850682944059372,
+      "learning_rate": 0.0001793055907293881,
+      "loss": 1.1245,
+      "step": 457
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.061626896262168884,
+      "learning_rate": 0.00017921624734111292,
+      "loss": 1.0763,
+      "step": 458
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.06570126861333847,
+      "learning_rate": 0.00017912673386848152,
+      "loss": 1.1547,
+      "step": 459
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.06566940248012543,
+      "learning_rate": 0.00017903705050368722,
+      "loss": 1.1218,
+      "step": 460
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.06877337396144867,
+      "learning_rate": 0.00017894719743928827,
+      "loss": 1.2173,
+      "step": 461
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.06467350572347641,
+      "learning_rate": 0.00017885717486820722,
+      "loss": 1.1347,
+      "step": 462
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.08330734819173813,
+      "learning_rate": 0.00017876698298373053,
+      "loss": 1.1373,
+      "step": 463
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.06419449299573898,
+      "learning_rate": 0.0001786766219795083,
+      "loss": 1.1285,
+      "step": 464
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.07984116673469543,
+      "learning_rate": 0.0001785860920495536,
+      "loss": 1.1309,
+      "step": 465
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.07000021636486053,
+      "learning_rate": 0.00017849539338824231,
+      "loss": 1.1193,
+      "step": 466
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.07003198564052582,
+      "learning_rate": 0.00017840452619031258,
+      "loss": 1.1596,
+      "step": 467
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.07209260761737823,
+      "learning_rate": 0.00017831349065086435,
+      "loss": 1.2199,
+      "step": 468
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.06799673289060593,
+      "learning_rate": 0.00017822228696535907,
+      "loss": 1.1247,
+      "step": 469
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.06913288682699203,
+      "learning_rate": 0.0001781309153296192,
+      "loss": 1.1035,
+      "step": 470
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.06872755289077759,
+      "learning_rate": 0.0001780393759398278,
+      "loss": 1.1613,
+      "step": 471
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.06622165441513062,
+      "learning_rate": 0.00017794766899252812,
+      "loss": 1.0869,
+      "step": 472
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.06728823482990265,
+      "learning_rate": 0.00017785579468462316,
+      "loss": 1.1408,
+      "step": 473
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.06555955857038498,
+      "learning_rate": 0.00017776375321337521,
+      "loss": 1.1584,
+      "step": 474
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.0661938264966011,
+      "learning_rate": 0.0001776715447764056,
+      "loss": 1.1542,
+      "step": 475
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.07007063180208206,
+      "learning_rate": 0.00017757916957169404,
+      "loss": 1.1424,
+      "step": 476
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.06441737711429596,
+      "learning_rate": 0.0001774866277975784,
+      "loss": 1.1361,
+      "step": 477
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.06655770540237427,
+      "learning_rate": 0.00017739391965275404,
+      "loss": 1.0968,
+      "step": 478
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.07319090515375137,
+      "learning_rate": 0.0001773010453362737,
+      "loss": 1.107,
+      "step": 479
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.06766431778669357,
+      "learning_rate": 0.0001772080050475468,
+      "loss": 1.1571,
+      "step": 480
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.06634720414876938,
+      "learning_rate": 0.00017711479898633914,
+      "loss": 1.175,
+      "step": 481
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.06589143723249435,
+      "learning_rate": 0.00017702142735277247,
+      "loss": 1.1475,
+      "step": 482
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.06698860973119736,
+      "learning_rate": 0.00017692789034732403,
+      "loss": 1.1042,
+      "step": 483
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.06417112797498703,
+      "learning_rate": 0.0001768341881708261,
+      "loss": 1.0588,
+      "step": 484
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06298335641622543,
+      "learning_rate": 0.00017674032102446563,
+      "loss": 0.9964,
+      "step": 485
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06435716152191162,
+      "learning_rate": 0.00017664628910978375,
+      "loss": 1.1126,
+      "step": 486
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06721868366003036,
+      "learning_rate": 0.00017655209262867541,
+      "loss": 1.1541,
+      "step": 487
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06489331275224686,
+      "learning_rate": 0.00017645773178338886,
+      "loss": 1.0539,
+      "step": 488
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06617634743452072,
+      "learning_rate": 0.00017636320677652526,
+      "loss": 1.094,
+      "step": 489
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06776611506938934,
+      "learning_rate": 0.0001762685178110382,
+      "loss": 1.0155,
+      "step": 490
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06376908719539642,
+      "learning_rate": 0.00017617366509023338,
+      "loss": 1.2292,
+      "step": 491
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.06204572319984436,
+      "learning_rate": 0.00017607864881776807,
+      "loss": 1.0968,
+      "step": 492
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.07098660618066788,
+      "learning_rate": 0.00017598346919765067,
+      "loss": 1.1067,
+      "step": 493
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.06708233058452606,
+      "learning_rate": 0.00017588812643424032,
+      "loss": 1.126,
+      "step": 494
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.0708741545677185,
+      "learning_rate": 0.00017579262073224647,
+      "loss": 1.0234,
+      "step": 495
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.06812895089387894,
+      "learning_rate": 0.00017569695229672835,
+      "loss": 1.14,
+      "step": 496
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.06947667896747589,
+      "learning_rate": 0.00017560112133309467,
+      "loss": 1.1888,
+      "step": 497
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.06499818712472916,
+      "learning_rate": 0.0001755051280471031,
+      "loss": 1.0889,
+      "step": 498
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.063695028424263,
+      "learning_rate": 0.00017540897264485967,
+      "loss": 1.0678,
+      "step": 499
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.06530368328094482,
+      "learning_rate": 0.00017531265533281872,
+      "loss": 1.0348,
+      "step": 500
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.07548119127750397,
+      "learning_rate": 0.0001752161763177821,
+      "loss": 1.1479,
+      "step": 501
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.06523353606462479,
+      "learning_rate": 0.00017511953580689888,
+      "loss": 1.1491,
+      "step": 502
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.06849244982004166,
+      "learning_rate": 0.00017502273400766484,
+      "loss": 1.143,
+      "step": 503
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.06976161897182465,
+      "learning_rate": 0.00017492577112792208,
+      "loss": 1.1597,
+      "step": 504
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.06484940648078918,
+      "learning_rate": 0.0001748286473758586,
+      "loss": 1.1013,
+      "step": 505
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.06747712939977646,
+      "learning_rate": 0.00017473136296000772,
+      "loss": 1.1129,
+      "step": 506
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.06816747039556503,
+      "learning_rate": 0.0001746339180892478,
+      "loss": 1.2297,
+      "step": 507
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.06335021555423737,
+      "learning_rate": 0.00017453631297280166,
+      "loss": 1.0683,
+      "step": 508
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.06458504498004913,
+      "learning_rate": 0.00017443854782023624,
+      "loss": 1.0101,
+      "step": 509
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.0690096914768219,
+      "learning_rate": 0.000174340622841462,
+      "loss": 1.1363,
+      "step": 510
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.06535279005765915,
+      "learning_rate": 0.00017424253824673263,
+      "loss": 1.0364,
+      "step": 511
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.0651826336979866,
+      "learning_rate": 0.00017414429424664454,
+      "loss": 1.1102,
+      "step": 512
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.06632277369499207,
+      "learning_rate": 0.00017404589105213632,
+      "loss": 1.1461,
+      "step": 513
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.06560949981212616,
+      "learning_rate": 0.00017394732887448847,
+      "loss": 1.1939,
+      "step": 514
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.06644757837057114,
+      "learning_rate": 0.00017384860792532276,
+      "loss": 1.1603,
+      "step": 515
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.06334047764539719,
+      "learning_rate": 0.00017374972841660186,
+      "loss": 1.1102,
+      "step": 516
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.061144351959228516,
+      "learning_rate": 0.00017365069056062894,
+      "loss": 1.0735,
+      "step": 517
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.0644732415676117,
+      "learning_rate": 0.00017355149457004709,
+      "loss": 1.1449,
+      "step": 518
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.06779829412698746,
+      "learning_rate": 0.00017345214065783897,
+      "loss": 1.0644,
+      "step": 519
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.06311120837926865,
+      "learning_rate": 0.00017335262903732634,
+      "loss": 1.0737,
+      "step": 520
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.070918507874012,
+      "learning_rate": 0.0001732529599221695,
+      "loss": 1.002,
+      "step": 521
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.06275039911270142,
+      "learning_rate": 0.0001731531335263669,
+      "loss": 1.0902,
+      "step": 522
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.06746424734592438,
+      "learning_rate": 0.0001730531500642548,
+      "loss": 1.1495,
+      "step": 523
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.07816082239151001,
+      "learning_rate": 0.00017295300975050658,
+      "loss": 1.0652,
+      "step": 524
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.06773856282234192,
+      "learning_rate": 0.00017285271280013245,
+      "loss": 1.1856,
+      "step": 525
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.0698360875248909,
+      "learning_rate": 0.0001727522594284789,
+      "loss": 1.201,
+      "step": 526
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.07427448779344559,
+      "learning_rate": 0.0001726516498512283,
+      "loss": 1.116,
+      "step": 527
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.06394834816455841,
+      "learning_rate": 0.00017255088428439836,
+      "loss": 1.1401,
+      "step": 528
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.06733064353466034,
+      "learning_rate": 0.0001724499629443418,
+      "loss": 1.1239,
+      "step": 529
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.06488535553216934,
+      "learning_rate": 0.00017234888604774574,
+      "loss": 1.1434,
+      "step": 530
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.0653986781835556,
+      "learning_rate": 0.00017224765381163132,
+      "loss": 1.1392,
+      "step": 531
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.07161843031644821,
+      "learning_rate": 0.00017214626645335314,
+      "loss": 1.0808,
+      "step": 532
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.06368163228034973,
+      "learning_rate": 0.00017204472419059895,
+      "loss": 1.1395,
+      "step": 533
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.07309909909963608,
+      "learning_rate": 0.00017194302724138903,
+      "loss": 1.1411,
+      "step": 534
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.0663491040468216,
+      "learning_rate": 0.00017184117582407578,
+      "loss": 1.1509,
+      "step": 535
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.06098710745573044,
+      "learning_rate": 0.00017173917015734336,
+      "loss": 1.0276,
+      "step": 536
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.0714559257030487,
+      "learning_rate": 0.00017163701046020707,
+      "loss": 1.2508,
+      "step": 537
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.06662353128194809,
+      "learning_rate": 0.00017153469695201277,
+      "loss": 1.0658,
+      "step": 538
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.0661638155579567,
+      "learning_rate": 0.0001714322298524368,
+      "loss": 1.1066,
+      "step": 539
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.07075709104537964,
+      "learning_rate": 0.00017132960938148512,
+      "loss": 1.1827,
+      "step": 540
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 1.0872660875320435,
+      "eval_runtime": 96.8426,
+      "eval_samples_per_second": 51.63,
+      "eval_steps_per_second": 12.908,
+      "step": 540
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.07708127051591873,
+      "learning_rate": 0.00017122683575949307,
+      "loss": 1.2037,
+      "step": 541
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.06289788335561752,
+      "learning_rate": 0.0001711239092071248,
+      "loss": 1.0755,
+      "step": 542
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.0677611231803894,
+      "learning_rate": 0.00017102082994537282,
+      "loss": 1.1438,
+      "step": 543
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.06381526589393616,
+      "learning_rate": 0.00017091759819555744,
+      "loss": 1.1003,
+      "step": 544
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.06795594096183777,
+      "learning_rate": 0.00017081421417932652,
+      "loss": 1.1063,
+      "step": 545
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.06619235128164291,
+      "learning_rate": 0.00017071067811865476,
+      "loss": 1.1499,
+      "step": 546
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.06958078593015671,
+      "learning_rate": 0.0001706069902358433,
+      "loss": 1.0918,
+      "step": 547
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.06839771568775177,
+      "learning_rate": 0.0001705031507535193,
+      "loss": 1.0789,
+      "step": 548
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.0677015408873558,
+      "learning_rate": 0.00017039915989463538,
+      "loss": 1.1419,
+      "step": 549
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.0648641586303711,
+      "learning_rate": 0.00017029501788246924,
+      "loss": 1.1461,
+      "step": 550
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.06705692410469055,
+      "learning_rate": 0.00017019072494062304,
+      "loss": 1.1288,
+      "step": 551
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.06862998753786087,
+      "learning_rate": 0.00017008628129302307,
+      "loss": 1.107,
+      "step": 552
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.07801419496536255,
+      "learning_rate": 0.00016998168716391915,
+      "loss": 1.097,
+      "step": 553
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.07198420912027359,
+      "learning_rate": 0.00016987694277788417,
+      "loss": 1.1328,
+      "step": 554
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.06974530965089798,
+      "learning_rate": 0.00016977204835981374,
+      "loss": 1.233,
+      "step": 555
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.06750082969665527,
+      "learning_rate": 0.00016966700413492556,
+      "loss": 1.1373,
+      "step": 556
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.06531986594200134,
+      "learning_rate": 0.00016956181032875894,
+      "loss": 1.0542,
+      "step": 557
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.06485234200954437,
+      "learning_rate": 0.0001694564671671743,
+      "loss": 1.0888,
+      "step": 558
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.06844603270292282,
+      "learning_rate": 0.0001693509748763529,
+      "loss": 1.1358,
+      "step": 559
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.06846935302019119,
+      "learning_rate": 0.00016924533368279607,
+      "loss": 1.186,
+      "step": 560
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.07385078817605972,
+      "learning_rate": 0.00016913954381332492,
+      "loss": 1.137,
+      "step": 561
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.06992754340171814,
+      "learning_rate": 0.0001690336054950797,
+      "loss": 1.1749,
+      "step": 562
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.06728782504796982,
+      "learning_rate": 0.0001689275189555195,
+      "loss": 1.041,
+      "step": 563
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.0680394098162651,
+      "learning_rate": 0.00016882128442242156,
+      "loss": 1.171,
+      "step": 564
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.06516265124082565,
+      "learning_rate": 0.00016871490212388087,
+      "loss": 1.126,
+      "step": 565
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.06674471497535706,
+      "learning_rate": 0.00016860837228830974,
+      "loss": 1.1474,
+      "step": 566
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.06331098824739456,
+      "learning_rate": 0.00016850169514443728,
+      "loss": 1.1349,
+      "step": 567
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.06607010960578918,
+      "learning_rate": 0.00016839487092130883,
+      "loss": 1.1848,
+      "step": 568
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.06833096593618393,
+      "learning_rate": 0.0001682878998482855,
+      "loss": 1.1646,
+      "step": 569
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.06246986240148544,
+      "learning_rate": 0.0001681807821550438,
+      "loss": 1.0467,
+      "step": 570
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.06146036833524704,
+      "learning_rate": 0.0001680735180715749,
+      "loss": 0.9574,
+      "step": 571
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.07327894866466522,
+      "learning_rate": 0.00016796610782818442,
+      "loss": 1.1735,
+      "step": 572
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.06582464277744293,
+      "learning_rate": 0.00016785855165549175,
+      "loss": 1.0437,
+      "step": 573
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.06511417776346207,
+      "learning_rate": 0.00016775084978442955,
+      "loss": 1.0813,
+      "step": 574
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.06808367371559143,
+      "learning_rate": 0.00016764300244624338,
+      "loss": 1.0771,
+      "step": 575
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.06845704466104507,
+      "learning_rate": 0.0001675350098724911,
+      "loss": 1.1645,
+      "step": 576
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.0673985630273819,
+      "learning_rate": 0.00016742687229504246,
+      "loss": 1.0358,
+      "step": 577
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.0681719034910202,
+      "learning_rate": 0.00016731858994607838,
+      "loss": 1.0971,
+      "step": 578
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.06640625,
+      "learning_rate": 0.00016721016305809084,
+      "loss": 1.242,
+      "step": 579
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.07134019583463669,
+      "learning_rate": 0.00016710159186388203,
+      "loss": 1.0486,
+      "step": 580
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.07230845093727112,
+      "learning_rate": 0.00016699287659656395,
+      "loss": 1.1678,
+      "step": 581
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.06907068192958832,
+      "learning_rate": 0.00016688401748955802,
+      "loss": 1.1183,
+      "step": 582
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.06549103558063507,
+      "learning_rate": 0.00016677501477659445,
+      "loss": 1.1073,
+      "step": 583
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.06864264607429504,
+      "learning_rate": 0.0001666658686917118,
+      "loss": 1.1528,
+      "step": 584
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.06810259073972702,
+      "learning_rate": 0.00016655657946925646,
+      "loss": 1.1308,
+      "step": 585
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.06780505925416946,
+      "learning_rate": 0.00016644714734388217,
+      "loss": 1.16,
+      "step": 586
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.06738951057195663,
+      "learning_rate": 0.00016633757255054947,
+      "loss": 1.1706,
+      "step": 587
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.06679335236549377,
+      "learning_rate": 0.0001662278553245252,
+      "loss": 1.1704,
+      "step": 588
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.06979308277368546,
+      "learning_rate": 0.00016611799590138203,
+      "loss": 1.1482,
+      "step": 589
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.0626835972070694,
+      "learning_rate": 0.00016600799451699802,
+      "loss": 1.1115,
+      "step": 590
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.06404287368059158,
+      "learning_rate": 0.0001658978514075559,
+      "loss": 1.175,
+      "step": 591
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.07459414005279541,
+      "learning_rate": 0.00016578756680954277,
+      "loss": 1.1302,
+      "step": 592
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.0666908472776413,
+      "learning_rate": 0.00016567714095974952,
+      "loss": 1.1636,
+      "step": 593
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.06724884361028671,
+      "learning_rate": 0.0001655665740952703,
+      "loss": 1.1355,
+      "step": 594
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.0689745619893074,
+      "learning_rate": 0.00016545586645350205,
+      "loss": 1.1162,
+      "step": 595
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.06672396510839462,
+      "learning_rate": 0.000165345018272144,
+      "loss": 1.0862,
+      "step": 596
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.06393066048622131,
+      "learning_rate": 0.00016523402978919704,
+      "loss": 1.1315,
+      "step": 597
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.07766008377075195,
+      "learning_rate": 0.00016512290124296336,
+      "loss": 1.1024,
+      "step": 598
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.07702287286520004,
+      "learning_rate": 0.00016501163287204588,
+      "loss": 1.1509,
+      "step": 599
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.06490621715784073,
+      "learning_rate": 0.00016490022491534768,
+      "loss": 1.0167,
+      "step": 600
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.06604059785604477,
+      "learning_rate": 0.0001647886776120717,
+      "loss": 1.106,
+      "step": 601
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.06731244921684265,
+      "learning_rate": 0.00016467699120171987,
+      "loss": 1.0898,
+      "step": 602
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.07821813970804214,
+      "learning_rate": 0.00016456516592409294,
+      "loss": 1.1176,
+      "step": 603
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.06816477328538895,
+      "learning_rate": 0.0001644532020192897,
+      "loss": 1.1786,
+      "step": 604
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.06874791532754898,
+      "learning_rate": 0.0001643410997277067,
+      "loss": 1.0926,
+      "step": 605
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.06932558119297028,
+      "learning_rate": 0.00016422885929003758,
+      "loss": 1.1679,
+      "step": 606
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.06875893473625183,
+      "learning_rate": 0.00016411648094727253,
+      "loss": 1.1935,
+      "step": 607
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.06305741518735886,
+      "learning_rate": 0.00016400396494069792,
+      "loss": 1.0654,
+      "step": 608
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.06925616413354874,
+      "learning_rate": 0.00016389131151189567,
+      "loss": 1.1218,
+      "step": 609
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.07191671431064606,
+      "learning_rate": 0.00016377852090274276,
+      "loss": 1.1875,
+      "step": 610
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.06721782684326172,
+      "learning_rate": 0.00016366559335541066,
+      "loss": 1.2297,
+      "step": 611
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.07050056755542755,
+      "learning_rate": 0.00016355252911236492,
+      "loss": 1.2071,
+      "step": 612
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.0658205896615982,
+      "learning_rate": 0.00016343932841636456,
+      "loss": 1.1652,
+      "step": 613
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.07456327974796295,
+      "learning_rate": 0.0001633259915104616,
+      "loss": 1.0894,
+      "step": 614
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.06406401097774506,
+      "learning_rate": 0.00016321251863800045,
+      "loss": 1.0864,
+      "step": 615
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.06928929686546326,
+      "learning_rate": 0.00016309891004261755,
+      "loss": 1.0215,
+      "step": 616
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.06436607241630554,
+      "learning_rate": 0.00016298516596824068,
+      "loss": 1.1855,
+      "step": 617
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.06957562267780304,
+      "learning_rate": 0.0001628712866590885,
+      "loss": 1.2085,
+      "step": 618
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.07085030525922775,
+      "learning_rate": 0.0001627572723596701,
+      "loss": 1.1079,
+      "step": 619
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.06608451902866364,
+      "learning_rate": 0.0001626431233147843,
+      "loss": 1.1456,
+      "step": 620
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.0686989277601242,
+      "learning_rate": 0.00016252883976951938,
+      "loss": 1.1112,
+      "step": 621
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.06871567666530609,
+      "learning_rate": 0.00016241442196925223,
+      "loss": 1.1612,
+      "step": 622
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.06559213250875473,
+      "learning_rate": 0.00016229987015964814,
+      "loss": 1.1588,
+      "step": 623
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.06943194568157196,
+      "learning_rate": 0.00016218518458666008,
+      "loss": 1.1108,
+      "step": 624
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.06725773215293884,
+      "learning_rate": 0.00016207036549652814,
+      "loss": 1.1942,
+      "step": 625
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.06677956134080887,
+      "learning_rate": 0.00016195541313577923,
+      "loss": 1.129,
+      "step": 626
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.06949102878570557,
+      "learning_rate": 0.00016184032775122631,
+      "loss": 1.0889,
+      "step": 627
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.06555166095495224,
+      "learning_rate": 0.00016172510958996795,
+      "loss": 1.1484,
+      "step": 628
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.07066816091537476,
+      "learning_rate": 0.0001616097588993879,
+      "loss": 1.0807,
+      "step": 629
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.06895067542791367,
+      "learning_rate": 0.00016149427592715432,
+      "loss": 1.1586,
+      "step": 630
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.07003293186426163,
+      "learning_rate": 0.00016137866092121953,
+      "loss": 1.1059,
+      "step": 631
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.06466325372457504,
+      "learning_rate": 0.0001612629141298192,
+      "loss": 1.1759,
+      "step": 632
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.06705661863088608,
+      "learning_rate": 0.00016114703580147202,
+      "loss": 1.1766,
+      "step": 633
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.07274851948022842,
+      "learning_rate": 0.00016103102618497922,
+      "loss": 1.211,
+      "step": 634
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.06761591881513596,
+      "learning_rate": 0.0001609148855294237,
+      "loss": 1.1049,
+      "step": 635
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.06395487487316132,
+      "learning_rate": 0.00016079861408416985,
+      "loss": 1.0404,
+      "step": 636
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.07156820595264435,
+      "learning_rate": 0.00016068221209886288,
+      "loss": 1.1569,
+      "step": 637
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.06414105743169785,
+      "learning_rate": 0.00016056567982342817,
+      "loss": 1.1568,
+      "step": 638
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.06608510762453079,
+      "learning_rate": 0.00016044901750807098,
+      "loss": 1.1529,
+      "step": 639
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.06756897270679474,
+      "learning_rate": 0.00016033222540327567,
+      "loss": 1.0845,
+      "step": 640
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.06797973066568375,
+      "learning_rate": 0.00016021530375980535,
+      "loss": 1.0768,
+      "step": 641
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.06994897872209549,
+      "learning_rate": 0.00016009825282870126,
+      "loss": 1.1013,
+      "step": 642
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.06616318970918655,
+      "learning_rate": 0.00015998107286128215,
+      "loss": 1.0895,
+      "step": 643
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.07222306728363037,
+      "learning_rate": 0.00015986376410914388,
+      "loss": 1.1811,
+      "step": 644
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.06701509654521942,
+      "learning_rate": 0.00015974632682415885,
+      "loss": 1.1747,
+      "step": 645
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.0668727234005928,
+      "learning_rate": 0.00015962876125847535,
+      "loss": 1.1241,
+      "step": 646
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.06632225960493088,
+      "learning_rate": 0.00015951106766451715,
+      "loss": 1.1301,
+      "step": 647
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.06712299585342407,
+      "learning_rate": 0.00015939324629498294,
+      "loss": 1.1646,
+      "step": 648
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.06921354681253433,
+      "learning_rate": 0.0001592752974028457,
+      "loss": 1.1075,
+      "step": 649
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.06251943111419678,
+      "learning_rate": 0.00015915722124135227,
+      "loss": 1.139,
+      "step": 650
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.06602328270673752,
+      "learning_rate": 0.00015903901806402264,
+      "loss": 1.1455,
+      "step": 651
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.06831764429807663,
+      "learning_rate": 0.00015892068812464963,
+      "loss": 1.1503,
+      "step": 652
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.0706949532032013,
+      "learning_rate": 0.00015880223167729818,
+      "loss": 1.1477,
+      "step": 653
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.07188212126493454,
+      "learning_rate": 0.0001586836489763049,
+      "loss": 1.0946,
+      "step": 654
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.0642421543598175,
+      "learning_rate": 0.00015856494027627734,
+      "loss": 1.1365,
+      "step": 655
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.07110840082168579,
+      "learning_rate": 0.00015844610583209373,
+      "loss": 1.1358,
+      "step": 656
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.06591419875621796,
+      "learning_rate": 0.0001583271458989023,
+      "loss": 1.1176,
+      "step": 657
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.0683559700846672,
+      "learning_rate": 0.00015820806073212055,
+      "loss": 1.0383,
+      "step": 658
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.06749717891216278,
+      "learning_rate": 0.00015808885058743498,
+      "loss": 1.1389,
+      "step": 659
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.07007017731666565,
+      "learning_rate": 0.00015796951572080047,
+      "loss": 1.1519,
+      "step": 660
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.06702763587236404,
+      "learning_rate": 0.00015785005638843956,
+      "loss": 1.1068,
+      "step": 661
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.07349605113267899,
+      "learning_rate": 0.0001577304728468422,
+      "loss": 1.2148,
+      "step": 662
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.06870365887880325,
+      "learning_rate": 0.00015761076535276486,
+      "loss": 1.1951,
+      "step": 663
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.07068511843681335,
+      "learning_rate": 0.00015749093416323024,
+      "loss": 1.1536,
+      "step": 664
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.0644054263830185,
+      "learning_rate": 0.00015737097953552658,
+      "loss": 1.1183,
+      "step": 665
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.06498972326517105,
+      "learning_rate": 0.0001572509017272072,
+      "loss": 1.0997,
+      "step": 666
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.06673012673854828,
+      "learning_rate": 0.00015713070099608988,
+      "loss": 1.1471,
+      "step": 667
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.06598661094903946,
+      "learning_rate": 0.0001570103776002563,
+      "loss": 1.0986,
+      "step": 668
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.06478138267993927,
+      "learning_rate": 0.00015688993179805154,
+      "loss": 1.0817,
+      "step": 669
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.06719911098480225,
+      "learning_rate": 0.00015676936384808354,
+      "loss": 1.1287,
+      "step": 670
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.06300705671310425,
+      "learning_rate": 0.00015664867400922239,
+      "loss": 1.0937,
+      "step": 671
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.06910447776317596,
+      "learning_rate": 0.00015652786254059998,
+      "loss": 1.0682,
+      "step": 672
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.06596586108207703,
+      "learning_rate": 0.0001564069297016093,
+      "loss": 1.0551,
+      "step": 673
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.06589323282241821,
+      "learning_rate": 0.00015628587575190395,
+      "loss": 1.1609,
+      "step": 674
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.07057927548885345,
+      "learning_rate": 0.00015616470095139762,
+      "loss": 1.1117,
+      "step": 675
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.07004386186599731,
+      "learning_rate": 0.0001560434055602634,
+      "loss": 1.1485,
+      "step": 676
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.06499215215444565,
+      "learning_rate": 0.00015592198983893329,
+      "loss": 1.0118,
+      "step": 677
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.06642549484968185,
+      "learning_rate": 0.00015580045404809772,
+      "loss": 1.1556,
+      "step": 678
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.06729692965745926,
+      "learning_rate": 0.0001556787984487049,
+      "loss": 1.1552,
+      "step": 679
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.069004587829113,
+      "learning_rate": 0.00015555702330196023,
+      "loss": 1.1373,
+      "step": 680
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.07328642904758453,
+      "learning_rate": 0.00015543512886932582,
+      "loss": 1.1266,
+      "step": 681
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.06573385000228882,
+      "learning_rate": 0.00015531311541251995,
+      "loss": 1.1094,
+      "step": 682
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.0658872127532959,
+      "learning_rate": 0.00015519098319351636,
+      "loss": 1.087,
+      "step": 683
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.0687699168920517,
+      "learning_rate": 0.00015506873247454384,
+      "loss": 1.0914,
+      "step": 684
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.06617342680692673,
+      "learning_rate": 0.00015494636351808563,
+      "loss": 1.1377,
+      "step": 685
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.06719239801168442,
+      "learning_rate": 0.00015482387658687875,
+      "loss": 1.2187,
+      "step": 686
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.0662359967827797,
+      "learning_rate": 0.0001547012719439136,
+      "loss": 1.0325,
+      "step": 687
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.0648040920495987,
+      "learning_rate": 0.0001545785498524333,
+      "loss": 1.0903,
+      "step": 688
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.06930810958147049,
+      "learning_rate": 0.00015445571057593315,
+      "loss": 1.128,
+      "step": 689
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.06874281167984009,
+      "learning_rate": 0.00015433275437816004,
+      "loss": 1.2245,
+      "step": 690
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.06701396405696869,
+      "learning_rate": 0.0001542096815231119,
+      "loss": 1.1042,
+      "step": 691
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.06753288954496384,
+      "learning_rate": 0.00015408649227503714,
+      "loss": 1.1593,
+      "step": 692
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.06974251568317413,
+      "learning_rate": 0.00015396318689843408,
+      "loss": 1.1662,
+      "step": 693
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.07186184823513031,
+      "learning_rate": 0.00015383976565805035,
+      "loss": 1.2292,
+      "step": 694
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.06782180815935135,
+      "learning_rate": 0.0001537162288188824,
+      "loss": 1.1006,
+      "step": 695
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.06984351575374603,
+      "learning_rate": 0.00015359257664617485,
+      "loss": 1.0839,
+      "step": 696
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.0696905329823494,
+      "learning_rate": 0.00015346880940541997,
+      "loss": 1.0847,
+      "step": 697
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.06678467988967896,
+      "learning_rate": 0.00015334492736235705,
+      "loss": 1.145,
+      "step": 698
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.06899551302194595,
+      "learning_rate": 0.0001532209307829719,
+      "loss": 1.0262,
+      "step": 699
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.06538818776607513,
+      "learning_rate": 0.00015309681993349626,
+      "loss": 1.1179,
+      "step": 700
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.0666823759675026,
+      "learning_rate": 0.00015297259508040718,
+      "loss": 1.1654,
+      "step": 701
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.0722886174917221,
+      "learning_rate": 0.00015284825649042655,
+      "loss": 1.1446,
+      "step": 702
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.0693565309047699,
+      "learning_rate": 0.0001527238044305204,
+      "loss": 1.2072,
+      "step": 703
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.07026789337396622,
+      "learning_rate": 0.00015259923916789844,
+      "loss": 1.1308,
+      "step": 704
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.07177083194255829,
+      "learning_rate": 0.00015247456097001338,
+      "loss": 1.1002,
+      "step": 705
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.0645269900560379,
+      "learning_rate": 0.00015234977010456047,
+      "loss": 1.1355,
+      "step": 706
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.0640435442328453,
+      "learning_rate": 0.00015222486683947678,
+      "loss": 1.1648,
+      "step": 707
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.06458823382854462,
+      "learning_rate": 0.0001520998514429409,
+      "loss": 1.0977,
+      "step": 708
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.06381873786449432,
+      "learning_rate": 0.0001519747241833719,
+      "loss": 1.0494,
+      "step": 709
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.07047054916620255,
+      "learning_rate": 0.00015184948532942928,
+      "loss": 1.0674,
+      "step": 710
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.07138516753911972,
+      "learning_rate": 0.000151724135150012,
+      "loss": 1.2869,
+      "step": 711
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.06848011165857315,
+      "learning_rate": 0.0001515986739142581,
+      "loss": 1.1813,
+      "step": 712
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.0663144439458847,
+      "learning_rate": 0.00015147310189154406,
+      "loss": 1.2019,
+      "step": 713
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.07065600156784058,
+      "learning_rate": 0.0001513474193514842,
+      "loss": 1.1387,
+      "step": 714
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.06766260415315628,
+      "learning_rate": 0.0001512216265639302,
+      "loss": 1.1266,
+      "step": 715
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.06984108686447144,
+      "learning_rate": 0.00015109572379897035,
+      "loss": 1.2299,
+      "step": 716
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.06617981195449829,
+      "learning_rate": 0.00015096971132692917,
+      "loss": 1.1855,
+      "step": 717
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.07162600755691528,
+      "learning_rate": 0.0001508435894183667,
+      "loss": 1.0786,
+      "step": 718
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2154,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 718,
+  "total_flos": 2.1425420186167542e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}