diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,4809 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9992652461425422,
+  "eval_steps": 250,
+  "global_step": 680,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0014695077149155032,
+      "grad_norm": 11.337424189328551,
+      "learning_rate": 4.7619047619047613e-08,
+      "loss": 1.5474,
+      "step": 1
+    },
+    {
+      "epoch": 0.0029390154298310064,
+      "grad_norm": 11.09462136805898,
+      "learning_rate": 9.523809523809523e-08,
+      "loss": 1.5705,
+      "step": 2
+    },
+    {
+      "epoch": 0.00440852314474651,
+      "grad_norm": 11.058506611133362,
+      "learning_rate": 1.4285714285714285e-07,
+      "loss": 1.4847,
+      "step": 3
+    },
+    {
+      "epoch": 0.005878030859662013,
+      "grad_norm": 10.9831239658005,
+      "learning_rate": 1.9047619047619045e-07,
+      "loss": 1.5536,
+      "step": 4
+    },
+    {
+      "epoch": 0.0073475385745775165,
+      "grad_norm": 10.8860240443791,
+      "learning_rate": 2.3809523809523806e-07,
+      "loss": 1.5523,
+      "step": 5
+    },
+    {
+      "epoch": 0.00881704628949302,
+      "grad_norm": 12.310500944697788,
+      "learning_rate": 2.857142857142857e-07,
+      "loss": 1.5543,
+      "step": 6
+    },
+    {
+      "epoch": 0.010286554004408524,
+      "grad_norm": 10.339190259363582,
+      "learning_rate": 3.333333333333333e-07,
+      "loss": 1.5789,
+      "step": 7
+    },
+    {
+      "epoch": 0.011756061719324026,
+      "grad_norm": 11.75096794982965,
+      "learning_rate": 3.809523809523809e-07,
+      "loss": 1.4619,
+      "step": 8
+    },
+    {
+      "epoch": 0.01322556943423953,
+      "grad_norm": 10.983629821806812,
+      "learning_rate": 4.285714285714285e-07,
+      "loss": 1.5461,
+      "step": 9
+    },
+    {
+      "epoch": 0.014695077149155033,
+      "grad_norm": 10.026453006105893,
+      "learning_rate": 4.761904761904761e-07,
+      "loss": 1.594,
+      "step": 10
+    },
+    {
+      "epoch": 0.016164584864070537,
+      "grad_norm": 9.2515708413017,
+      "learning_rate": 5.238095238095238e-07,
+      "loss": 1.4904,
+      "step": 11
+    },
+    {
+      "epoch": 0.01763409257898604,
+      "grad_norm": 8.8513471084721,
+      "learning_rate": 5.714285714285714e-07,
+      "loss": 1.5626,
+      "step": 12
+    },
+    {
+      "epoch": 0.019103600293901544,
+      "grad_norm": 9.013166716897066,
+      "learning_rate": 6.19047619047619e-07,
+      "loss": 1.5522,
+      "step": 13
+    },
+    {
+      "epoch": 0.020573108008817047,
+      "grad_norm": 8.149687442441557,
+      "learning_rate": 6.666666666666666e-07,
+      "loss": 1.5075,
+      "step": 14
+    },
+    {
+      "epoch": 0.02204261572373255,
+      "grad_norm": 6.880717564693127,
+      "learning_rate": 7.142857142857143e-07,
+      "loss": 1.538,
+      "step": 15
+    },
+    {
+      "epoch": 0.02351212343864805,
+      "grad_norm": 6.545463429781402,
+      "learning_rate": 7.619047619047618e-07,
+      "loss": 1.5044,
+      "step": 16
+    },
+    {
+      "epoch": 0.024981631153563555,
+      "grad_norm": 6.093488294374897,
+      "learning_rate": 8.095238095238095e-07,
+      "loss": 1.4779,
+      "step": 17
+    },
+    {
+      "epoch": 0.02645113886847906,
+      "grad_norm": 6.4020806034329425,
+      "learning_rate": 8.57142857142857e-07,
+      "loss": 1.4431,
+      "step": 18
+    },
+    {
+      "epoch": 0.027920646583394562,
+      "grad_norm": 6.142908310782321,
+      "learning_rate": 9.047619047619047e-07,
+      "loss": 1.5965,
+      "step": 19
+    },
+    {
+      "epoch": 0.029390154298310066,
+      "grad_norm": 5.136583641076265,
+      "learning_rate": 9.523809523809522e-07,
+      "loss": 1.3968,
+      "step": 20
+    },
+    {
+      "epoch": 0.03085966201322557,
+      "grad_norm": 5.02265599787556,
+      "learning_rate": 1e-06,
+      "loss": 1.4007,
+      "step": 21
+    },
+    {
+      "epoch": 0.03232916972814107,
+      "grad_norm": 4.976860924734495,
+      "learning_rate": 9.999943184333936e-07,
+      "loss": 1.4378,
+      "step": 22
+    },
+    {
+      "epoch": 0.03379867744305658,
+      "grad_norm": 4.89918762254008,
+      "learning_rate": 9.999772738626954e-07,
+      "loss": 1.4341,
+      "step": 23
+    },
+    {
+      "epoch": 0.03526818515797208,
+      "grad_norm": 4.659631180053279,
+      "learning_rate": 9.999488666752648e-07,
+      "loss": 1.4178,
+      "step": 24
+    },
+    {
+      "epoch": 0.036737692872887584,
+      "grad_norm": 4.542979691946772,
+      "learning_rate": 9.99909097516691e-07,
+      "loss": 1.4933,
+      "step": 25
+    },
+    {
+      "epoch": 0.03820720058780309,
+      "grad_norm": 4.342835973040556,
+      "learning_rate": 9.998579672907788e-07,
+      "loss": 1.4417,
+      "step": 26
+    },
+    {
+      "epoch": 0.03967670830271859,
+      "grad_norm": 3.2871765127170853,
+      "learning_rate": 9.99795477159527e-07,
+      "loss": 1.423,
+      "step": 27
+    },
+    {
+      "epoch": 0.041146216017634095,
+      "grad_norm": 3.2008427389003766,
+      "learning_rate": 9.99721628543103e-07,
+      "loss": 1.3282,
+      "step": 28
+    },
+    {
+      "epoch": 0.0426157237325496,
+      "grad_norm": 3.1114560975594836,
+      "learning_rate": 9.996364231198103e-07,
+      "loss": 1.3891,
+      "step": 29
+    },
+    {
+      "epoch": 0.0440852314474651,
+      "grad_norm": 2.9307530666626884,
+      "learning_rate": 9.9953986282605e-07,
+      "loss": 1.3589,
+      "step": 30
+    },
+    {
+      "epoch": 0.045554739162380606,
+      "grad_norm": 2.9544258327932,
+      "learning_rate": 9.99431949856277e-07,
+      "loss": 1.3487,
+      "step": 31
+    },
+    {
+      "epoch": 0.0470242468772961,
+      "grad_norm": 2.9071124300801565,
+      "learning_rate": 9.9931268666295e-07,
+      "loss": 1.3462,
+      "step": 32
+    },
+    {
+      "epoch": 0.048493754592211606,
+      "grad_norm": 2.673673317642893,
+      "learning_rate": 9.991820759564766e-07,
+      "loss": 1.3972,
+      "step": 33
+    },
+    {
+      "epoch": 0.04996326230712711,
+      "grad_norm": 2.746698468930808,
+      "learning_rate": 9.990401207051504e-07,
+      "loss": 1.2998,
+      "step": 34
+    },
+    {
+      "epoch": 0.05143277002204261,
+      "grad_norm": 2.699798149588662,
+      "learning_rate": 9.98886824135084e-07,
+      "loss": 1.3451,
+      "step": 35
+    },
+    {
+      "epoch": 0.05290227773695812,
+      "grad_norm": 2.7302270842982987,
+      "learning_rate": 9.98722189730136e-07,
+      "loss": 1.392,
+      "step": 36
+    },
+    {
+      "epoch": 0.05437178545187362,
+      "grad_norm": 2.535544436400141,
+      "learning_rate": 9.985462212318322e-07,
+      "loss": 1.3673,
+      "step": 37
+    },
+    {
+      "epoch": 0.055841293166789124,
+      "grad_norm": 2.5536809032653145,
+      "learning_rate": 9.983589226392792e-07,
+      "loss": 1.2696,
+      "step": 38
+    },
+    {
+      "epoch": 0.05731080088170463,
+      "grad_norm": 2.438242871155347,
+      "learning_rate": 9.98160298209075e-07,
+      "loss": 1.3818,
+      "step": 39
+    },
+    {
+      "epoch": 0.05878030859662013,
+      "grad_norm": 2.2111464394984814,
+      "learning_rate": 9.97950352455211e-07,
+      "loss": 1.2618,
+      "step": 40
+    },
+    {
+      "epoch": 0.060249816311535635,
+      "grad_norm": 2.0951995475252856,
+      "learning_rate": 9.977290901489707e-07,
+      "loss": 1.2963,
+      "step": 41
+    },
+    {
+      "epoch": 0.06171932402645114,
+      "grad_norm": 2.1449537100989846,
+      "learning_rate": 9.9749651631882e-07,
+      "loss": 1.2525,
+      "step": 42
+    },
+    {
+      "epoch": 0.06318883174136664,
+      "grad_norm": 2.617528852412968,
+      "learning_rate": 9.972526362502937e-07,
+      "loss": 1.3144,
+      "step": 43
+    },
+    {
+      "epoch": 0.06465833945628215,
+      "grad_norm": 2.210234169368085,
+      "learning_rate": 9.969974554858754e-07,
+      "loss": 1.2379,
+      "step": 44
+    },
+    {
+      "epoch": 0.06612784717119764,
+      "grad_norm": 2.42457485573731,
+      "learning_rate": 9.967309798248707e-07,
+      "loss": 1.2573,
+      "step": 45
+    },
+    {
+      "epoch": 0.06759735488611315,
+      "grad_norm": 2.0254014234209405,
+      "learning_rate": 9.96453215323277e-07,
+      "loss": 1.2536,
+      "step": 46
+    },
+    {
+      "epoch": 0.06906686260102865,
+      "grad_norm": 2.0725386643920958,
+      "learning_rate": 9.961641682936442e-07,
+      "loss": 1.2687,
+      "step": 47
+    },
+    {
+      "epoch": 0.07053637031594416,
+      "grad_norm": 1.8563249903888095,
+      "learning_rate": 9.95863845304932e-07,
+      "loss": 1.2335,
+      "step": 48
+    },
+    {
+      "epoch": 0.07200587803085966,
+      "grad_norm": 1.9548577428039762,
+      "learning_rate": 9.955522531823606e-07,
+      "loss": 1.3056,
+      "step": 49
+    },
+    {
+      "epoch": 0.07347538574577517,
+      "grad_norm": 1.855842269546498,
+      "learning_rate": 9.952293990072557e-07,
+      "loss": 1.2671,
+      "step": 50
+    },
+    {
+      "epoch": 0.07494489346069066,
+      "grad_norm": 1.7972312223565392,
+      "learning_rate": 9.948952901168874e-07,
+      "loss": 1.2718,
+      "step": 51
+    },
+    {
+      "epoch": 0.07641440117560618,
+      "grad_norm": 1.6423712015458758,
+      "learning_rate": 9.945499341043033e-07,
+      "loss": 1.2043,
+      "step": 52
+    },
+    {
+      "epoch": 0.07788390889052167,
+      "grad_norm": 1.6028195930568658,
+      "learning_rate": 9.94193338818156e-07,
+      "loss": 1.2136,
+      "step": 53
+    },
+    {
+      "epoch": 0.07935341660543718,
+      "grad_norm": 1.5486505262354593,
+      "learning_rate": 9.938255123625251e-07,
+      "loss": 1.2556,
+      "step": 54
+    },
+    {
+      "epoch": 0.08082292432035268,
+      "grad_norm": 1.5343630961919048,
+      "learning_rate": 9.934464630967328e-07,
+      "loss": 1.2519,
+      "step": 55
+    },
+    {
+      "epoch": 0.08229243203526819,
+      "grad_norm": 1.662816938451208,
+      "learning_rate": 9.930561996351533e-07,
+      "loss": 1.2482,
+      "step": 56
+    },
+    {
+      "epoch": 0.08376193975018369,
+      "grad_norm": 1.5931246608007381,
+      "learning_rate": 9.926547308470183e-07,
+      "loss": 1.1625,
+      "step": 57
+    },
+    {
+      "epoch": 0.0852314474650992,
+      "grad_norm": 1.5447707955729764,
+      "learning_rate": 9.922420658562144e-07,
+      "loss": 1.1873,
+      "step": 58
+    },
+    {
+      "epoch": 0.0867009551800147,
+      "grad_norm": 1.644709907245824,
+      "learning_rate": 9.91818214041076e-07,
+      "loss": 1.1914,
+      "step": 59
+    },
+    {
+      "epoch": 0.0881704628949302,
+      "grad_norm": 1.6366535743931827,
+      "learning_rate": 9.913831850341725e-07,
+      "loss": 1.1871,
+      "step": 60
+    },
+    {
+      "epoch": 0.0896399706098457,
+      "grad_norm": 1.5571790849844254,
+      "learning_rate": 9.90936988722089e-07,
+      "loss": 1.2605,
+      "step": 61
+    },
+    {
+      "epoch": 0.09110947832476121,
+      "grad_norm": 3.7097540672301386,
+      "learning_rate": 9.904796352452019e-07,
+      "loss": 1.1212,
+      "step": 62
+    },
+    {
+      "epoch": 0.09257898603967671,
+      "grad_norm": 1.5517830620847208,
+      "learning_rate": 9.900111349974478e-07,
+      "loss": 1.1623,
+      "step": 63
+    },
+    {
+      "epoch": 0.0940484937545922,
+      "grad_norm": 1.5680093555121737,
+      "learning_rate": 9.895314986260886e-07,
+      "loss": 1.1552,
+      "step": 64
+    },
+    {
+      "epoch": 0.09551800146950772,
+      "grad_norm": 1.6209106318425315,
+      "learning_rate": 9.890407370314677e-07,
+      "loss": 1.1337,
+      "step": 65
+    },
+    {
+      "epoch": 0.09698750918442321,
+      "grad_norm": 1.4100673270240494,
+      "learning_rate": 9.885388613667644e-07,
+      "loss": 1.2156,
+      "step": 66
+    },
+    {
+      "epoch": 0.09845701689933872,
+      "grad_norm": 1.5012099159985126,
+      "learning_rate": 9.880258830377386e-07,
+      "loss": 1.2427,
+      "step": 67
+    },
+    {
+      "epoch": 0.09992652461425422,
+      "grad_norm": 1.5660607972935188,
+      "learning_rate": 9.875018137024721e-07,
+      "loss": 1.1729,
+      "step": 68
+    },
+    {
+      "epoch": 0.10139603232916973,
+      "grad_norm": 1.5666357672603266,
+      "learning_rate": 9.869666652711049e-07,
+      "loss": 1.2838,
+      "step": 69
+    },
+    {
+      "epoch": 0.10286554004408523,
+      "grad_norm": 1.5178831815539953,
+      "learning_rate": 9.864204499055622e-07,
+      "loss": 1.2145,
+      "step": 70
+    },
+    {
+      "epoch": 0.10433504775900074,
+      "grad_norm": 1.71647656199295,
+      "learning_rate": 9.858631800192804e-07,
+      "loss": 1.1724,
+      "step": 71
+    },
+    {
+      "epoch": 0.10580455547391623,
+      "grad_norm": 1.5138224523065718,
+      "learning_rate": 9.852948682769234e-07,
+      "loss": 1.2672,
+      "step": 72
+    },
+    {
+      "epoch": 0.10727406318883174,
+      "grad_norm": 1.5726438604764952,
+      "learning_rate": 9.84715527594095e-07,
+      "loss": 1.1505,
+      "step": 73
+    },
+    {
+      "epoch": 0.10874357090374724,
+      "grad_norm": 1.5196898825337892,
+      "learning_rate": 9.841251711370457e-07,
+      "loss": 1.1801,
+      "step": 74
+    },
+    {
+      "epoch": 0.11021307861866275,
+      "grad_norm": 1.5385492133663783,
+      "learning_rate": 9.83523812322374e-07,
+      "loss": 1.2523,
+      "step": 75
+    },
+    {
+      "epoch": 0.11168258633357825,
+      "grad_norm": 1.4506696813615474,
+      "learning_rate": 9.829114648167206e-07,
+      "loss": 1.1798,
+      "step": 76
+    },
+    {
+      "epoch": 0.11315209404849376,
+      "grad_norm": 1.5471986855965765,
+      "learning_rate": 9.822881425364578e-07,
+      "loss": 1.1227,
+      "step": 77
+    },
+    {
+      "epoch": 0.11462160176340926,
+      "grad_norm": 1.4810764356218553,
+      "learning_rate": 9.81653859647374e-07,
+      "loss": 1.1904,
+      "step": 78
+    },
+    {
+      "epoch": 0.11609110947832477,
+      "grad_norm": 1.458422270393885,
+      "learning_rate": 9.810086305643511e-07,
+      "loss": 1.1688,
+      "step": 79
+    },
+    {
+      "epoch": 0.11756061719324026,
+      "grad_norm": 1.4150019691387836,
+      "learning_rate": 9.803524699510372e-07,
+      "loss": 1.1679,
+      "step": 80
+    },
+    {
+      "epoch": 0.11903012490815577,
+      "grad_norm": 1.5594916312404665,
+      "learning_rate": 9.79685392719513e-07,
+      "loss": 1.1915,
+      "step": 81
+    },
+    {
+      "epoch": 0.12049963262307127,
+      "grad_norm": 1.5378772800136011,
+      "learning_rate": 9.790074140299535e-07,
+      "loss": 1.2446,
+      "step": 82
+    },
+    {
+      "epoch": 0.12196914033798678,
+      "grad_norm": 1.3745260542581244,
+      "learning_rate": 9.783185492902831e-07,
+      "loss": 1.1894,
+      "step": 83
+    },
+    {
+      "epoch": 0.12343864805290228,
+      "grad_norm": 1.4281455021507763,
+      "learning_rate": 9.776188141558253e-07,
+      "loss": 1.1535,
+      "step": 84
+    },
+    {
+      "epoch": 0.12490815576781777,
+      "grad_norm": 1.4827991467262824,
+      "learning_rate": 9.769082245289472e-07,
+      "loss": 1.1812,
+      "step": 85
+    },
+    {
+      "epoch": 0.12637766348273327,
+      "grad_norm": 1.4091217681303907,
+      "learning_rate": 9.76186796558698e-07,
+      "loss": 1.1733,
+      "step": 86
+    },
+    {
+      "epoch": 0.1278471711976488,
+      "grad_norm": 1.504340633323682,
+      "learning_rate": 9.754545466404423e-07,
+      "loss": 1.1891,
+      "step": 87
+    },
+    {
+      "epoch": 0.1293166789125643,
+      "grad_norm": 1.5341255256297364,
+      "learning_rate": 9.747114914154862e-07,
+      "loss": 1.1941,
+      "step": 88
+    },
+    {
+      "epoch": 0.1307861866274798,
+      "grad_norm": 1.448718524801724,
+      "learning_rate": 9.73957647770701e-07,
+      "loss": 1.2118,
+      "step": 89
+    },
+    {
+      "epoch": 0.13225569434239529,
+      "grad_norm": 1.4171871543912984,
+      "learning_rate": 9.731930328381384e-07,
+      "loss": 1.1745,
+      "step": 90
+    },
+    {
+      "epoch": 0.1337252020573108,
+      "grad_norm": 1.487952395310904,
+      "learning_rate": 9.72417663994641e-07,
+      "loss": 1.1625,
+      "step": 91
+    },
+    {
+      "epoch": 0.1351947097722263,
+      "grad_norm": 1.5914017741584792,
+      "learning_rate": 9.716315588614472e-07,
+      "loss": 1.1051,
+      "step": 92
+    },
+    {
+      "epoch": 0.1366642174871418,
+      "grad_norm": 1.6391622049937462,
+      "learning_rate": 9.708347353037924e-07,
+      "loss": 1.143,
+      "step": 93
+    },
+    {
+      "epoch": 0.1381337252020573,
+      "grad_norm": 1.4437878891336424,
+      "learning_rate": 9.700272114305008e-07,
+      "loss": 1.1648,
+      "step": 94
+    },
+    {
+      "epoch": 0.13960323291697282,
+      "grad_norm": 1.4177630579747587,
+      "learning_rate": 9.69209005593575e-07,
+      "loss": 1.1999,
+      "step": 95
+    },
+    {
+      "epoch": 0.14107274063188832,
+      "grad_norm": 1.5120811492994586,
+      "learning_rate": 9.68380136387779e-07,
+      "loss": 1.2242,
+      "step": 96
+    },
+    {
+      "epoch": 0.14254224834680382,
+      "grad_norm": 1.4738631960216506,
+      "learning_rate": 9.67540622650215e-07,
+      "loss": 1.1548,
+      "step": 97
+    },
+    {
+      "epoch": 0.14401175606171931,
+      "grad_norm": 1.5620452244313792,
+      "learning_rate": 9.66690483459896e-07,
+      "loss": 1.1875,
+      "step": 98
+    },
+    {
+      "epoch": 0.14548126377663484,
+      "grad_norm": 1.5673649210031055,
+      "learning_rate": 9.658297381373117e-07,
+      "loss": 1.198,
+      "step": 99
+    },
+    {
+      "epoch": 0.14695077149155034,
+      "grad_norm": 1.4854078674557176,
+      "learning_rate": 9.649584062439897e-07,
+      "loss": 1.2242,
+      "step": 100
+    },
+    {
+      "epoch": 0.14842027920646583,
+      "grad_norm": 1.5887802038965162,
+      "learning_rate": 9.640765075820508e-07,
+      "loss": 1.23,
+      "step": 101
+    },
+    {
+      "epoch": 0.14988978692138133,
+      "grad_norm": 1.5956321149777724,
+      "learning_rate": 9.631840621937585e-07,
+      "loss": 1.226,
+      "step": 102
+    },
+    {
+      "epoch": 0.15135929463629685,
+      "grad_norm": 1.4446949899977264,
+      "learning_rate": 9.622810903610653e-07,
+      "loss": 1.1796,
+      "step": 103
+    },
+    {
+      "epoch": 0.15282880235121235,
+      "grad_norm": 1.4023788130375525,
+      "learning_rate": 9.613676126051488e-07,
+      "loss": 1.1589,
+      "step": 104
+    },
+    {
+      "epoch": 0.15429831006612785,
+      "grad_norm": 1.5491829383620404,
+      "learning_rate": 9.604436496859482e-07,
+      "loss": 1.237,
+      "step": 105
+    },
+    {
+      "epoch": 0.15576781778104334,
+      "grad_norm": 1.3527676295882165,
+      "learning_rate": 9.595092226016912e-07,
+      "loss": 1.1329,
+      "step": 106
+    },
+    {
+      "epoch": 0.15723732549595884,
+      "grad_norm": 1.5174098623642012,
+      "learning_rate": 9.585643525884163e-07,
+      "loss": 1.179,
+      "step": 107
+    },
+    {
+      "epoch": 0.15870683321087437,
+      "grad_norm": 1.3787786506819375,
+      "learning_rate": 9.576090611194915e-07,
+      "loss": 1.1199,
+      "step": 108
+    },
+    {
+      "epoch": 0.16017634092578986,
+      "grad_norm": 1.4679013393236473,
+      "learning_rate": 9.566433699051248e-07,
+      "loss": 1.2768,
+      "step": 109
+    },
+    {
+      "epoch": 0.16164584864070536,
+      "grad_norm": 1.4363459788248092,
+      "learning_rate": 9.556673008918725e-07,
+      "loss": 1.1434,
+      "step": 110
+    },
+    {
+      "epoch": 0.16311535635562086,
+      "grad_norm": 1.5373366464111682,
+      "learning_rate": 9.546808762621385e-07,
+      "loss": 1.1391,
+      "step": 111
+    },
+    {
+      "epoch": 0.16458486407053638,
+      "grad_norm": 1.397660653092303,
+      "learning_rate": 9.536841184336725e-07,
+      "loss": 1.1599,
+      "step": 112
+    },
+    {
+      "epoch": 0.16605437178545188,
+      "grad_norm": 1.5338825854004605,
+      "learning_rate": 9.526770500590576e-07,
+      "loss": 1.0599,
+      "step": 113
+    },
+    {
+      "epoch": 0.16752387950036737,
+      "grad_norm": 1.7726346336471452,
+      "learning_rate": 9.516596940251986e-07,
+      "loss": 1.1347,
+      "step": 114
+    },
+    {
+      "epoch": 0.16899338721528287,
+      "grad_norm": 1.401274657699429,
+      "learning_rate": 9.506320734527997e-07,
+      "loss": 1.1538,
+      "step": 115
+    },
+    {
+      "epoch": 0.1704628949301984,
+      "grad_norm": 1.4313194549405384,
+      "learning_rate": 9.495942116958395e-07,
+      "loss": 1.2409,
+      "step": 116
+    },
+    {
+      "epoch": 0.1719324026451139,
+      "grad_norm": 1.338422913627801,
+      "learning_rate": 9.485461323410411e-07,
+      "loss": 1.2274,
+      "step": 117
+    },
+    {
+      "epoch": 0.1734019103600294,
+      "grad_norm": 1.414029707516539,
+      "learning_rate": 9.474878592073352e-07,
+      "loss": 1.1813,
+      "step": 118
+    },
+    {
+      "epoch": 0.17487141807494488,
+      "grad_norm": 1.5090772581920886,
+      "learning_rate": 9.464194163453188e-07,
+      "loss": 1.1576,
+      "step": 119
+    },
+    {
+      "epoch": 0.1763409257898604,
+      "grad_norm": 1.397279646873424,
+      "learning_rate": 9.45340828036709e-07,
+      "loss": 1.1172,
+      "step": 120
+    },
+    {
+      "epoch": 0.1778104335047759,
+      "grad_norm": 1.4335773048462084,
+      "learning_rate": 9.442521187937911e-07,
+      "loss": 1.2413,
+      "step": 121
+    },
+    {
+      "epoch": 0.1792799412196914,
+      "grad_norm": 1.4697864318337606,
+      "learning_rate": 9.431533133588616e-07,
+      "loss": 1.1551,
+      "step": 122
+    },
+    {
+      "epoch": 0.1807494489346069,
+      "grad_norm": 1.418392328936289,
+      "learning_rate": 9.420444367036653e-07,
+      "loss": 1.1411,
+      "step": 123
+    },
+    {
+      "epoch": 0.18221895664952242,
+      "grad_norm": 1.3863098352226801,
+      "learning_rate": 9.409255140288288e-07,
+      "loss": 1.1388,
+      "step": 124
+    },
+    {
+      "epoch": 0.18368846436443792,
+      "grad_norm": 1.4189204374991593,
+      "learning_rate": 9.397965707632866e-07,
+      "loss": 1.1512,
+      "step": 125
+    },
+    {
+      "epoch": 0.18515797207935342,
+      "grad_norm": 1.524052045769854,
+      "learning_rate": 9.386576325637043e-07,
+      "loss": 1.1759,
+      "step": 126
+    },
+    {
+      "epoch": 0.1866274797942689,
+      "grad_norm": 1.405091479549588,
+      "learning_rate": 9.375087253138951e-07,
+      "loss": 1.166,
+      "step": 127
+    },
+    {
+      "epoch": 0.1880969875091844,
+      "grad_norm": 1.544876861673977,
+      "learning_rate": 9.363498751242307e-07,
+      "loss": 1.1309,
+      "step": 128
+    },
+    {
+      "epoch": 0.18956649522409993,
+      "grad_norm": 1.423691492612366,
+      "learning_rate": 9.351811083310497e-07,
+      "loss": 1.1394,
+      "step": 129
+    },
+    {
+      "epoch": 0.19103600293901543,
+      "grad_norm": 1.466545968863769,
+      "learning_rate": 9.340024514960574e-07,
+      "loss": 1.1772,
+      "step": 130
+    },
+    {
+      "epoch": 0.19250551065393093,
+      "grad_norm": 1.4099417566779877,
+      "learning_rate": 9.328139314057233e-07,
+      "loss": 1.116,
+      "step": 131
+    },
+    {
+      "epoch": 0.19397501836884642,
+      "grad_norm": 1.3751112771189664,
+      "learning_rate": 9.316155750706713e-07,
+      "loss": 1.1249,
+      "step": 132
+    },
+    {
+      "epoch": 0.19544452608376195,
+      "grad_norm": 1.440365523396416,
+      "learning_rate": 9.304074097250671e-07,
+      "loss": 1.1607,
+      "step": 133
+    },
+    {
+      "epoch": 0.19691403379867745,
+      "grad_norm": 1.3429807064760693,
+      "learning_rate": 9.291894628259979e-07,
+      "loss": 1.148,
+      "step": 134
+    },
+    {
+      "epoch": 0.19838354151359294,
+      "grad_norm": 1.6100321463659257,
+      "learning_rate": 9.279617620528496e-07,
+      "loss": 1.17,
+      "step": 135
+    },
+    {
+      "epoch": 0.19985304922850844,
+      "grad_norm": 1.5750772736599945,
+      "learning_rate": 9.26724335306677e-07,
+      "loss": 1.1625,
+      "step": 136
+    },
+    {
+      "epoch": 0.20132255694342396,
+      "grad_norm": 1.4373009264887187,
+      "learning_rate": 9.254772107095701e-07,
+      "loss": 1.1595,
+      "step": 137
+    },
+    {
+      "epoch": 0.20279206465833946,
+      "grad_norm": 1.3522430478527927,
+      "learning_rate": 9.242204166040148e-07,
+      "loss": 1.105,
+      "step": 138
+    },
+    {
+      "epoch": 0.20426157237325496,
+      "grad_norm": 1.496220468060292,
+      "learning_rate": 9.229539815522485e-07,
+      "loss": 1.1729,
+      "step": 139
+    },
+    {
+      "epoch": 0.20573108008817045,
+      "grad_norm": 1.3692383152400416,
+      "learning_rate": 9.216779343356119e-07,
+      "loss": 1.0892,
+      "step": 140
+    },
+    {
+      "epoch": 0.20720058780308598,
+      "grad_norm": 1.415058934382544,
+      "learning_rate": 9.203923039538939e-07,
+      "loss": 1.098,
+      "step": 141
+    },
+    {
+      "epoch": 0.20867009551800147,
+      "grad_norm": 1.3906133545085528,
+      "learning_rate": 9.190971196246731e-07,
+      "loss": 1.1862,
+      "step": 142
+    },
+    {
+      "epoch": 0.21013960323291697,
+      "grad_norm": 1.4853242615363982,
+      "learning_rate": 9.177924107826535e-07,
+      "loss": 1.125,
+      "step": 143
+    },
+    {
+      "epoch": 0.21160911094783247,
+      "grad_norm": 1.4673768423852198,
+      "learning_rate": 9.164782070789961e-07,
+      "loss": 1.2024,
+      "step": 144
+    },
+    {
+      "epoch": 0.213078618662748,
+      "grad_norm": 1.357361299564151,
+      "learning_rate": 9.151545383806441e-07,
+      "loss": 1.0744,
+      "step": 145
+    },
+    {
+      "epoch": 0.2145481263776635,
+      "grad_norm": 1.4090842422695524,
+      "learning_rate": 9.138214347696453e-07,
+      "loss": 1.2193,
+      "step": 146
+    },
+    {
+      "epoch": 0.216017634092579,
+      "grad_norm": 1.403649440383755,
+      "learning_rate": 9.124789265424674e-07,
+      "loss": 1.1808,
+      "step": 147
+    },
+    {
+      "epoch": 0.21748714180749448,
+      "grad_norm": 1.3566301062411306,
+      "learning_rate": 9.1112704420931e-07,
+      "loss": 1.1419,
+      "step": 148
+    },
+    {
+      "epoch": 0.21895664952240998,
+      "grad_norm": 1.3695324273141942,
+      "learning_rate": 9.097658184934114e-07,
+      "loss": 1.1495,
+      "step": 149
+    },
+    {
+      "epoch": 0.2204261572373255,
+      "grad_norm": 1.4138005238611118,
+      "learning_rate": 9.083952803303496e-07,
+      "loss": 1.2471,
+      "step": 150
+    },
+    {
+      "epoch": 0.221895664952241,
+      "grad_norm": 1.3414103194314155,
+      "learning_rate": 9.070154608673402e-07,
+      "loss": 1.1672,
+      "step": 151
+    },
+    {
+      "epoch": 0.2233651726671565,
+      "grad_norm": 1.438725688375383,
+      "learning_rate": 9.056263914625277e-07,
+      "loss": 1.2144,
+      "step": 152
+    },
+    {
+      "epoch": 0.224834680382072,
+      "grad_norm": 1.564966532077944,
+      "learning_rate": 9.042281036842739e-07,
+      "loss": 1.1981,
+      "step": 153
+    },
+    {
+      "epoch": 0.22630418809698752,
+      "grad_norm": 1.275239691393889,
+      "learning_rate": 9.028206293104391e-07,
+      "loss": 1.1098,
+      "step": 154
+    },
+    {
+      "epoch": 0.22777369581190302,
+      "grad_norm": 1.4419905057143438,
+      "learning_rate": 9.014040003276611e-07,
+      "loss": 1.1783,
+      "step": 155
+    },
+    {
+      "epoch": 0.2292432035268185,
+      "grad_norm": 1.4105436779658675,
+      "learning_rate": 8.999782489306271e-07,
+      "loss": 1.1513,
+      "step": 156
+    },
+    {
+      "epoch": 0.230712711241734,
+      "grad_norm": 1.4197672181122762,
+      "learning_rate": 8.985434075213439e-07,
+      "loss": 1.1433,
+      "step": 157
+    },
+    {
+      "epoch": 0.23218221895664953,
+      "grad_norm": 1.3670907633233818,
+      "learning_rate": 8.970995087083992e-07,
+      "loss": 1.03,
+      "step": 158
+    },
+    {
+      "epoch": 0.23365172667156503,
+      "grad_norm": 1.3603227901129349,
+      "learning_rate": 8.956465853062222e-07,
+      "loss": 1.0851,
+      "step": 159
+    },
+    {
+      "epoch": 0.23512123438648053,
+      "grad_norm": 1.3852573758765583,
+      "learning_rate": 8.941846703343372e-07,
+      "loss": 1.2005,
+      "step": 160
+    },
+    {
+      "epoch": 0.23659074210139602,
+      "grad_norm": 1.3638595388577004,
+      "learning_rate": 8.927137970166135e-07,
+      "loss": 1.1217,
+      "step": 161
+    },
+    {
+      "epoch": 0.23806024981631155,
+      "grad_norm": 1.4806310433707142,
+      "learning_rate": 8.912339987805099e-07,
+      "loss": 1.1339,
+      "step": 162
+    },
+    {
+      "epoch": 0.23952975753122704,
+      "grad_norm": 1.4400386045554132,
+      "learning_rate": 8.897453092563153e-07,
+      "loss": 1.1343,
+      "step": 163
+    },
+    {
+      "epoch": 0.24099926524614254,
+      "grad_norm": 1.3697227632539914,
+      "learning_rate": 8.882477622763846e-07,
+      "loss": 1.1529,
+      "step": 164
+    },
+    {
+      "epoch": 0.24246877296105804,
+      "grad_norm": 1.3503406924429706,
+      "learning_rate": 8.867413918743693e-07,
+      "loss": 1.164,
+      "step": 165
+    },
+    {
+      "epoch": 0.24393828067597356,
+      "grad_norm": 1.3850072048890962,
+      "learning_rate": 8.852262322844444e-07,
+      "loss": 1.153,
+      "step": 166
+    },
+    {
+      "epoch": 0.24540778839088906,
+      "grad_norm": 1.359563698104864,
+      "learning_rate": 8.837023179405308e-07,
+      "loss": 1.1724,
+      "step": 167
+    },
+    {
+      "epoch": 0.24687729610580456,
+      "grad_norm": 1.3978149085145892,
+      "learning_rate": 8.821696834755117e-07,
+      "loss": 1.0965,
+      "step": 168
+    },
+    {
+      "epoch": 0.24834680382072005,
+      "grad_norm": 1.404265065477685,
+      "learning_rate": 8.806283637204462e-07,
+      "loss": 1.1879,
+      "step": 169
+    },
+    {
+      "epoch": 0.24981631153563555,
+      "grad_norm": 1.403937125406496,
+      "learning_rate": 8.790783937037776e-07,
+      "loss": 1.1251,
+      "step": 170
+    },
+    {
+      "epoch": 0.25128581925055105,
+      "grad_norm": 1.4373795084844492,
+      "learning_rate": 8.775198086505375e-07,
+      "loss": 1.1347,
+      "step": 171
+    },
+    {
+      "epoch": 0.25275532696546654,
+      "grad_norm": 1.4706572304655665,
+      "learning_rate": 8.759526439815455e-07,
+      "loss": 1.0477,
+      "step": 172
+    },
+    {
+      "epoch": 0.2542248346803821,
+      "grad_norm": 1.4008092483314147,
+      "learning_rate": 8.743769353126029e-07,
+      "loss": 1.1761,
+      "step": 173
+    },
+    {
+      "epoch": 0.2556943423952976,
+      "grad_norm": 1.3764920528768414,
+      "learning_rate": 8.727927184536849e-07,
+      "loss": 1.1277,
+      "step": 174
+    },
+    {
+      "epoch": 0.2571638501102131,
+      "grad_norm": 1.4213614638860257,
+      "learning_rate": 8.712000294081259e-07,
+      "loss": 1.0911,
+      "step": 175
+    },
+    {
+      "epoch": 0.2586333578251286,
+      "grad_norm": 1.3408094274422444,
+      "learning_rate": 8.695989043718015e-07,
+      "loss": 1.1602,
+      "step": 176
+    },
+    {
+      "epoch": 0.2601028655400441,
+      "grad_norm": 1.4025428044724146,
+      "learning_rate": 8.679893797323058e-07,
+      "loss": 1.1325,
+      "step": 177
+    },
+    {
+      "epoch": 0.2615723732549596,
+      "grad_norm": 1.3560427968027609,
+      "learning_rate": 8.663714920681245e-07,
+      "loss": 1.1789,
+      "step": 178
+    },
+    {
+      "epoch": 0.2630418809698751,
+      "grad_norm": 1.4997523461667033,
+      "learning_rate": 8.64745278147804e-07,
+      "loss": 1.1578,
+      "step": 179
+    },
+    {
+      "epoch": 0.26451138868479057,
+      "grad_norm": 1.4091572819165725,
+      "learning_rate": 8.631107749291148e-07,
+      "loss": 1.1344,
+      "step": 180
+    },
+    {
+      "epoch": 0.2659808963997061,
+      "grad_norm": 1.5029746752542426,
+      "learning_rate": 8.614680195582127e-07,
+      "loss": 1.1829,
+      "step": 181
+    },
+    {
+      "epoch": 0.2674504041146216,
+      "grad_norm": 1.4231917336486979,
+      "learning_rate": 8.598170493687939e-07,
+      "loss": 1.097,
+      "step": 182
+    },
+    {
+      "epoch": 0.2689199118295371,
+      "grad_norm": 1.3738447004488614,
+      "learning_rate": 8.581579018812468e-07,
+      "loss": 1.1539,
+      "step": 183
+    },
+    {
+      "epoch": 0.2703894195444526,
+      "grad_norm": 1.4706711617380483,
+      "learning_rate": 8.564906148017992e-07,
+      "loss": 1.179,
+      "step": 184
+    },
+    {
+      "epoch": 0.2718589272593681,
+      "grad_norm": 1.3788046176437054,
+      "learning_rate": 8.548152260216613e-07,
+      "loss": 1.1003,
+      "step": 185
+    },
+    {
+      "epoch": 0.2733284349742836,
+      "grad_norm": 1.3707247735299555,
+      "learning_rate": 8.531317736161652e-07,
+      "loss": 1.0604,
+      "step": 186
+    },
+    {
+      "epoch": 0.2747979426891991,
+      "grad_norm": 1.3354127082342266,
+      "learning_rate": 8.514402958438987e-07,
+      "loss": 1.1121,
+      "step": 187
+    },
+    {
+      "epoch": 0.2762674504041146,
+      "grad_norm": 1.336615144205344,
+      "learning_rate": 8.497408311458362e-07,
+      "loss": 1.0885,
+      "step": 188
+    },
+    {
+      "epoch": 0.2777369581190301,
+      "grad_norm": 1.3645293415775273,
+      "learning_rate": 8.480334181444652e-07,
+      "loss": 1.1334,
+      "step": 189
+    },
+    {
+      "epoch": 0.27920646583394565,
+      "grad_norm": 1.405142088376881,
+      "learning_rate": 8.463180956429085e-07,
+      "loss": 1.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.28067597354886115,
+      "grad_norm": 1.3868866326557587,
+      "learning_rate": 8.445949026240424e-07,
+      "loss": 1.2174,
+      "step": 191
+    },
+    {
+      "epoch": 0.28214548126377664,
+      "grad_norm": 1.335338615058983,
+      "learning_rate": 8.428638782496105e-07,
+      "loss": 1.1235,
+      "step": 192
+    },
+    {
+      "epoch": 0.28361498897869214,
+      "grad_norm": 1.3713807109243261,
+      "learning_rate": 8.411250618593337e-07,
+      "loss": 1.2191,
+      "step": 193
+    },
+    {
+      "epoch": 0.28508449669360764,
+      "grad_norm": 1.41726090350252,
+      "learning_rate": 8.393784929700169e-07,
+      "loss": 1.1522,
+      "step": 194
+    },
+    {
+      "epoch": 0.28655400440852313,
+      "grad_norm": 1.3529238461119495,
+      "learning_rate": 8.376242112746499e-07,
+      "loss": 1.0376,
+      "step": 195
+    },
+    {
+      "epoch": 0.28802351212343863,
+      "grad_norm": 1.4100766083722753,
+      "learning_rate": 8.358622566415057e-07,
+      "loss": 1.0645,
+      "step": 196
+    },
+    {
+      "epoch": 0.2894930198383541,
+      "grad_norm": 1.3130064224637852,
+      "learning_rate": 8.340926691132348e-07,
+      "loss": 1.1029,
+      "step": 197
+    },
+    {
+      "epoch": 0.2909625275532697,
+      "grad_norm": 1.3572153763312709,
+      "learning_rate": 8.323154889059549e-07,
+      "loss": 1.1309,
+      "step": 198
+    },
+    {
+      "epoch": 0.2924320352681852,
+      "grad_norm": 1.542624539487457,
+      "learning_rate": 8.305307564083368e-07,
+      "loss": 1.2663,
+      "step": 199
+    },
+    {
+      "epoch": 0.29390154298310067,
+      "grad_norm": 1.3525660807398292,
+      "learning_rate": 8.287385121806869e-07,
+      "loss": 1.1509,
+      "step": 200
+    },
+    {
+      "epoch": 0.29537105069801617,
+      "grad_norm": 1.514167322393567,
+      "learning_rate": 8.26938796954025e-07,
+      "loss": 1.233,
+      "step": 201
+    },
+    {
+      "epoch": 0.29684055841293167,
+      "grad_norm": 1.3922047754873006,
+      "learning_rate": 8.251316516291586e-07,
+      "loss": 1.1035,
+      "step": 202
+    },
+    {
+      "epoch": 0.29831006612784716,
+      "grad_norm": 1.450031885451194,
+      "learning_rate": 8.233171172757539e-07,
+      "loss": 1.1351,
+      "step": 203
+    },
+    {
+      "epoch": 0.29977957384276266,
+      "grad_norm": 1.3679188570513114,
+      "learning_rate": 8.214952351314022e-07,
+      "loss": 1.105,
+      "step": 204
+    },
+    {
+      "epoch": 0.30124908155767816,
+      "grad_norm": 1.3734569453159238,
+      "learning_rate": 8.196660466006823e-07,
+      "loss": 1.1665,
+      "step": 205
+    },
+    {
+      "epoch": 0.3027185892725937,
+      "grad_norm": 1.4252740827310968,
+      "learning_rate": 8.178295932542205e-07,
+      "loss": 1.1422,
+      "step": 206
+    },
+    {
+      "epoch": 0.3041880969875092,
+      "grad_norm": 1.38166857353608,
+      "learning_rate": 8.159859168277444e-07,
+      "loss": 1.0333,
+      "step": 207
+    },
+    {
+      "epoch": 0.3056576047024247,
+      "grad_norm": 1.4722910903729798,
+      "learning_rate": 8.141350592211358e-07,
+      "loss": 1.134,
+      "step": 208
+    },
+    {
+      "epoch": 0.3071271124173402,
+      "grad_norm": 1.4074834904053308,
+      "learning_rate": 8.122770624974778e-07,
+      "loss": 1.0712,
+      "step": 209
+    },
+    {
+      "epoch": 0.3085966201322557,
+      "grad_norm": 1.4280613261841697,
+      "learning_rate": 8.10411968882099e-07,
+      "loss": 1.1277,
+      "step": 210
+    },
+    {
+      "epoch": 0.3100661278471712,
+      "grad_norm": 1.3743456352090058,
+      "learning_rate": 8.085398207616138e-07,
+      "loss": 1.2189,
+      "step": 211
+    },
+    {
+      "epoch": 0.3115356355620867,
+      "grad_norm": 1.4591835041502073,
+      "learning_rate": 8.06660660682959e-07,
+      "loss": 1.0779,
+      "step": 212
+    },
+    {
+      "epoch": 0.3130051432770022,
+      "grad_norm": 1.3896625615186315,
+      "learning_rate": 8.047745313524275e-07,
+      "loss": 1.118,
+      "step": 213
+    },
+    {
+      "epoch": 0.3144746509919177,
+      "grad_norm": 1.4095222295795578,
+      "learning_rate": 8.028814756346967e-07,
+      "loss": 1.1015,
+      "step": 214
+    },
+    {
+      "epoch": 0.31594415870683323,
+      "grad_norm": 1.3835385844364931,
+      "learning_rate": 8.009815365518554e-07,
+      "loss": 1.1964,
+      "step": 215
+    },
+    {
+      "epoch": 0.31741366642174873,
+      "grad_norm": 1.4442582574354577,
+      "learning_rate": 7.990747572824253e-07,
+      "loss": 1.0994,
+      "step": 216
+    },
+    {
+      "epoch": 0.3188831741366642,
+      "grad_norm": 1.540269470315885,
+      "learning_rate": 7.971611811603803e-07,
+      "loss": 1.1612,
+      "step": 217
+    },
+    {
+      "epoch": 0.3203526818515797,
+      "grad_norm": 1.5012905943987096,
+      "learning_rate": 7.952408516741607e-07,
+      "loss": 1.1212,
+      "step": 218
+    },
+    {
+      "epoch": 0.3218221895664952,
+      "grad_norm": 1.377730602489123,
+      "learning_rate": 7.933138124656864e-07,
+      "loss": 1.136,
+      "step": 219
+    },
+    {
+      "epoch": 0.3232916972814107,
+      "grad_norm": 1.4387714462293955,
+      "learning_rate": 7.913801073293638e-07,
+      "loss": 1.1662,
+      "step": 220
+    },
+    {
+      "epoch": 0.3247612049963262,
+      "grad_norm": 1.3416069488877052,
+      "learning_rate": 7.894397802110908e-07,
+      "loss": 1.1242,
+      "step": 221
+    },
+    {
+      "epoch": 0.3262307127112417,
+      "grad_norm": 1.3608424909383003,
+      "learning_rate": 7.87492875207259e-07,
+      "loss": 1.2513,
+      "step": 222
+    },
+    {
+      "epoch": 0.32770022042615726,
+      "grad_norm": 1.3266709078386894,
+      "learning_rate": 7.855394365637495e-07,
+      "loss": 1.1342,
+      "step": 223
+    },
+    {
+      "epoch": 0.32916972814107276,
+      "grad_norm": 1.3442846396883863,
+      "learning_rate": 7.835795086749299e-07,
+      "loss": 1.1801,
+      "step": 224
+    },
+    {
+      "epoch": 0.33063923585598826,
+      "grad_norm": 1.4096410255441036,
+      "learning_rate": 7.816131360826434e-07,
+      "loss": 1.1195,
+      "step": 225
+    },
+    {
+      "epoch": 0.33210874357090375,
+      "grad_norm": 1.3103219218263102,
+      "learning_rate": 7.796403634751973e-07,
+      "loss": 1.1207,
+      "step": 226
+    },
+    {
+      "epoch": 0.33357825128581925,
+      "grad_norm": 1.3677511713167119,
+      "learning_rate": 7.776612356863477e-07,
+      "loss": 1.0818,
+      "step": 227
+    },
+    {
+      "epoch": 0.33504775900073475,
+      "grad_norm": 1.4153862144691094,
+      "learning_rate": 7.756757976942798e-07,
+      "loss": 1.1155,
+      "step": 228
+    },
+    {
+      "epoch": 0.33651726671565024,
+      "grad_norm": 1.4438500129836211,
+      "learning_rate": 7.736840946205865e-07,
+      "loss": 1.193,
+      "step": 229
+    },
+    {
+      "epoch": 0.33798677443056574,
+      "grad_norm": 1.456870674708481,
+      "learning_rate": 7.716861717292424e-07,
+      "loss": 1.1199,
+      "step": 230
+    },
+    {
+      "epoch": 0.33945628214548124,
+      "grad_norm": 1.3923730055470556,
+      "learning_rate": 7.696820744255756e-07,
+      "loss": 1.2182,
+      "step": 231
+    },
+    {
+      "epoch": 0.3409257898603968,
+      "grad_norm": 1.370735300132156,
+      "learning_rate": 7.676718482552353e-07,
+      "loss": 1.134,
+      "step": 232
+    },
+    {
+      "epoch": 0.3423952975753123,
+      "grad_norm": 1.3480114555788523,
+      "learning_rate": 7.65655538903157e-07,
+      "loss": 1.0998,
+      "step": 233
+    },
+    {
+      "epoch": 0.3438648052902278,
+      "grad_norm": 1.4276815010976192,
+      "learning_rate": 7.636331921925241e-07,
+      "loss": 1.1696,
+      "step": 234
+    },
+    {
+      "epoch": 0.3453343130051433,
+      "grad_norm": 1.4170232593240883,
+      "learning_rate": 7.61604854083727e-07,
+      "loss": 1.116,
+      "step": 235
+    },
+    {
+      "epoch": 0.3468038207200588,
+      "grad_norm": 1.4134392902146415,
+      "learning_rate": 7.595705706733178e-07,
+      "loss": 1.1178,
+      "step": 236
+    },
+    {
+      "epoch": 0.34827332843497427,
+      "grad_norm": 1.3744159242762746,
+      "learning_rate": 7.575303881929632e-07,
+      "loss": 1.1824,
+      "step": 237
+    },
+    {
+      "epoch": 0.34974283614988977,
+      "grad_norm": 1.38324732435889,
+      "learning_rate": 7.55484353008394e-07,
+      "loss": 1.1422,
+      "step": 238
+    },
+    {
+      "epoch": 0.35121234386480527,
+      "grad_norm": 1.3881791765597504,
+      "learning_rate": 7.534325116183508e-07,
+      "loss": 1.1597,
+      "step": 239
+    },
+    {
+      "epoch": 0.3526818515797208,
+      "grad_norm": 1.396997073324397,
+      "learning_rate": 7.513749106535278e-07,
+      "loss": 1.1062,
+      "step": 240
+    },
+    {
+      "epoch": 0.3541513592946363,
+      "grad_norm": 1.443575506575812,
+      "learning_rate": 7.493115968755125e-07,
+      "loss": 1.1133,
+      "step": 241
+    },
+    {
+      "epoch": 0.3556208670095518,
+      "grad_norm": 1.4651722014658863,
+      "learning_rate": 7.472426171757238e-07,
+      "loss": 1.1609,
+      "step": 242
+    },
+    {
+      "epoch": 0.3570903747244673,
+      "grad_norm": 1.358202452421609,
+      "learning_rate": 7.451680185743454e-07,
+      "loss": 1.0965,
+      "step": 243
+    },
+    {
+      "epoch": 0.3585598824393828,
+      "grad_norm": 1.5175842365493015,
+      "learning_rate": 7.430878482192579e-07,
+      "loss": 1.0451,
+      "step": 244
+    },
+    {
+      "epoch": 0.3600293901542983,
+      "grad_norm": 1.4934774283507497,
+      "learning_rate": 7.41002153384967e-07,
+      "loss": 1.168,
+      "step": 245
+    },
+    {
+      "epoch": 0.3614988978692138,
+      "grad_norm": 1.4081881789418684,
+      "learning_rate": 7.389109814715292e-07,
+      "loss": 1.1344,
+      "step": 246
+    },
+    {
+      "epoch": 0.3629684055841293,
+      "grad_norm": 1.4765256904942174,
+      "learning_rate": 7.368143800034745e-07,
+      "loss": 1.1749,
+      "step": 247
+    },
+    {
+      "epoch": 0.36443791329904485,
+      "grad_norm": 1.3509670984266027,
+      "learning_rate": 7.347123966287265e-07,
+      "loss": 1.1334,
+      "step": 248
+    },
+    {
+      "epoch": 0.36590742101396034,
+      "grad_norm": 1.53782552923439,
+      "learning_rate": 7.326050791175196e-07,
+      "loss": 1.2062,
+      "step": 249
+    },
+    {
+      "epoch": 0.36737692872887584,
+      "grad_norm": 1.3428470092287972,
+      "learning_rate": 7.304924753613127e-07,
+      "loss": 1.0949,
+      "step": 250
+    },
+    {
+      "epoch": 0.36737692872887584,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_loss": 2.756098508834839,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_runtime": 5.9542,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_samples_per_second": 15.619,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_steps_per_second": 2.015,
+      "step": 250
+    },
+    {
+      "epoch": 0.36884643644379134,
+      "grad_norm": 1.3548965420628396,
+      "learning_rate": 7.283746333717014e-07,
+      "loss": 1.1102,
+      "step": 251
+    },
+    {
+      "epoch": 0.37031594415870683,
+      "grad_norm": 1.3760628358821343,
+      "learning_rate": 7.262516012793276e-07,
+      "loss": 1.0957,
+      "step": 252
+    },
+    {
+      "epoch": 0.37178545187362233,
+      "grad_norm": 1.4363319701743325,
+      "learning_rate": 7.241234273327838e-07,
+      "loss": 1.0782,
+      "step": 253
+    },
+    {
+      "epoch": 0.3732549595885378,
+      "grad_norm": 1.3292955709874987,
+      "learning_rate": 7.219901598975185e-07,
+      "loss": 1.0904,
+      "step": 254
+    },
+    {
+      "epoch": 0.3747244673034533,
+      "grad_norm": 1.3562568095097107,
+      "learning_rate": 7.198518474547354e-07,
+      "loss": 1.0553,
+      "step": 255
+    },
+    {
+      "epoch": 0.3761939750183688,
+      "grad_norm": 1.3922411026579398,
+      "learning_rate": 7.17708538600293e-07,
+      "loss": 1.1265,
+      "step": 256
+    },
+    {
+      "epoch": 0.3776634827332844,
+      "grad_norm": 1.3917980021891811,
+      "learning_rate": 7.155602820435992e-07,
+      "loss": 1.153,
+      "step": 257
+    },
+    {
+      "epoch": 0.37913299044819987,
+      "grad_norm": 1.4243974023254953,
+      "learning_rate": 7.134071266065051e-07,
+      "loss": 1.1354,
+      "step": 258
+    },
+    {
+      "epoch": 0.38060249816311537,
+      "grad_norm": 1.4529979227761458,
+      "learning_rate": 7.112491212221946e-07,
+      "loss": 1.0746,
+      "step": 259
+    },
+    {
+      "epoch": 0.38207200587803086,
+      "grad_norm": 1.381064754681566,
+      "learning_rate": 7.09086314934073e-07,
+      "loss": 1.0771,
+      "step": 260
+    },
+    {
+      "epoch": 0.38354151359294636,
+      "grad_norm": 1.3494944459929241,
+      "learning_rate": 7.069187568946524e-07,
+      "loss": 1.1317,
+      "step": 261
+    },
+    {
+      "epoch": 0.38501102130786186,
+      "grad_norm": 1.367365128216842,
+      "learning_rate": 7.047464963644342e-07,
+      "loss": 1.0841,
+      "step": 262
+    },
+    {
+      "epoch": 0.38648052902277735,
+      "grad_norm": 1.4111229982907987,
+      "learning_rate": 7.025695827107901e-07,
+      "loss": 1.1456,
+      "step": 263
+    },
+    {
+      "epoch": 0.38795003673769285,
+      "grad_norm": 1.4469253951516285,
+      "learning_rate": 7.003880654068395e-07,
+      "loss": 1.1497,
+      "step": 264
+    },
+    {
+      "epoch": 0.3894195444526084,
+      "grad_norm": 1.438128214166176,
+      "learning_rate": 6.98201994030326e-07,
+      "loss": 1.1566,
+      "step": 265
+    },
+    {
+      "epoch": 0.3908890521675239,
+      "grad_norm": 1.392547852669767,
+      "learning_rate": 6.960114182624902e-07,
+      "loss": 1.1729,
+      "step": 266
+    },
+    {
+      "epoch": 0.3923585598824394,
+      "grad_norm": 1.401971190522276,
+      "learning_rate": 6.938163878869405e-07,
+      "loss": 1.0718,
+      "step": 267
+    },
+    {
+      "epoch": 0.3938280675973549,
+      "grad_norm": 1.396822074711564,
+      "learning_rate": 6.916169527885221e-07,
+      "loss": 1.1141,
+      "step": 268
+    },
+    {
+      "epoch": 0.3952975753122704,
+      "grad_norm": 1.3883010823518094,
+      "learning_rate": 6.894131629521829e-07,
+      "loss": 1.2072,
+      "step": 269
+    },
+    {
+      "epoch": 0.3967670830271859,
+      "grad_norm": 1.400367946344679,
+      "learning_rate": 6.872050684618381e-07,
+      "loss": 1.1512,
+      "step": 270
+    },
+    {
+      "epoch": 0.3982365907421014,
+      "grad_norm": 1.386986813686829,
+      "learning_rate": 6.849927194992312e-07,
+      "loss": 1.0693,
+      "step": 271
+    },
+    {
+      "epoch": 0.3997060984570169,
+      "grad_norm": 1.4393959811216241,
+      "learning_rate": 6.827761663427943e-07,
+      "loss": 1.1799,
+      "step": 272
+    },
+    {
+      "epoch": 0.4011756061719324,
+      "grad_norm": 1.3989135808754347,
+      "learning_rate": 6.805554593665049e-07,
+      "loss": 1.1858,
+      "step": 273
+    },
+    {
+      "epoch": 0.4026451138868479,
+      "grad_norm": 1.3294118191399562,
+      "learning_rate": 6.783306490387414e-07,
+      "loss": 1.051,
+      "step": 274
+    },
+    {
+      "epoch": 0.4041146216017634,
+      "grad_norm": 1.3425765430711993,
+      "learning_rate": 6.761017859211359e-07,
+      "loss": 1.0931,
+      "step": 275
+    },
+    {
+      "epoch": 0.4055841293166789,
+      "grad_norm": 1.3979046769171386,
+      "learning_rate": 6.738689206674257e-07,
+      "loss": 1.0426,
+      "step": 276
+    },
+    {
+      "epoch": 0.4070536370315944,
+      "grad_norm": 1.3494939507393635,
+      "learning_rate": 6.716321040223014e-07,
+      "loss": 1.0846,
+      "step": 277
+    },
+    {
+      "epoch": 0.4085231447465099,
+      "grad_norm": 1.38158252960459,
+      "learning_rate": 6.693913868202539e-07,
+      "loss": 1.0751,
+      "step": 278
+    },
+    {
+      "epoch": 0.4099926524614254,
+      "grad_norm": 1.3216203318442303,
+      "learning_rate": 6.671468199844192e-07,
+      "loss": 1.074,
+      "step": 279
+    },
+    {
+      "epoch": 0.4114621601763409,
+      "grad_norm": 1.3720532658754991,
+      "learning_rate": 6.648984545254216e-07,
+      "loss": 1.1034,
+      "step": 280
+    },
+    {
+      "epoch": 0.4129316678912564,
+      "grad_norm": 1.3638059704989733,
+      "learning_rate": 6.626463415402131e-07,
+      "loss": 1.1033,
+      "step": 281
+    },
+    {
+      "epoch": 0.41440117560617196,
+      "grad_norm": 1.3293227220141846,
+      "learning_rate": 6.603905322109138e-07,
+      "loss": 1.1054,
+      "step": 282
+    },
+    {
+      "epoch": 0.41587068332108745,
+      "grad_norm": 1.4610059929514447,
+      "learning_rate": 6.581310778036474e-07,
+      "loss": 0.9441,
+      "step": 283
+    },
+    {
+      "epoch": 0.41734019103600295,
+      "grad_norm": 1.412604693032899,
+      "learning_rate": 6.558680296673766e-07,
+      "loss": 1.1464,
+      "step": 284
+    },
+    {
+      "epoch": 0.41880969875091845,
+      "grad_norm": 1.3384277299187781,
+      "learning_rate": 6.536014392327365e-07,
+      "loss": 1.1396,
+      "step": 285
+    },
+    {
+      "epoch": 0.42027920646583394,
+      "grad_norm": 1.3987728346478638,
+      "learning_rate": 6.51331358010865e-07,
+      "loss": 1.0833,
+      "step": 286
+    },
+    {
+      "epoch": 0.42174871418074944,
+      "grad_norm": 1.577477591087323,
+      "learning_rate": 6.490578375922328e-07,
+      "loss": 1.1428,
+      "step": 287
+    },
+    {
+      "epoch": 0.42321822189566494,
+      "grad_norm": 1.5588639617673477,
+      "learning_rate": 6.467809296454708e-07,
+      "loss": 1.0857,
+      "step": 288
+    },
+    {
+      "epoch": 0.42468772961058043,
+      "grad_norm": 1.4118713026809537,
+      "learning_rate": 6.445006859161956e-07,
+      "loss": 1.1891,
+      "step": 289
+    },
+    {
+      "epoch": 0.426157237325496,
+      "grad_norm": 1.414060079207387,
+      "learning_rate": 6.422171582258334e-07,
+      "loss": 1.0867,
+      "step": 290
+    },
+    {
+      "epoch": 0.4276267450404115,
+      "grad_norm": 1.4123941837634646,
+      "learning_rate": 6.399303984704432e-07,
+      "loss": 1.1262,
+      "step": 291
+    },
+    {
+      "epoch": 0.429096252755327,
+      "grad_norm": 1.445359468963558,
+      "learning_rate": 6.376404586195364e-07,
+      "loss": 1.0714,
+      "step": 292
+    },
+    {
+      "epoch": 0.4305657604702425,
+      "grad_norm": 1.4230042553733626,
+      "learning_rate": 6.353473907148961e-07,
+      "loss": 1.1363,
+      "step": 293
+    },
+    {
+      "epoch": 0.432035268185158,
+      "grad_norm": 1.325942136523235,
+      "learning_rate": 6.330512468693944e-07,
+      "loss": 1.062,
+      "step": 294
+    },
+    {
+      "epoch": 0.43350477590007347,
+      "grad_norm": 1.4212600729175167,
+      "learning_rate": 6.307520792658081e-07,
+      "loss": 1.1282,
+      "step": 295
+    },
+    {
+      "epoch": 0.43497428361498897,
+      "grad_norm": 1.4254820055813413,
+      "learning_rate": 6.284499401556328e-07,
+      "loss": 1.1743,
+      "step": 296
+    },
+    {
+      "epoch": 0.43644379132990446,
+      "grad_norm": 1.4651887201716516,
+      "learning_rate": 6.261448818578952e-07,
+      "loss": 1.0821,
+      "step": 297
+    },
+    {
+      "epoch": 0.43791329904481996,
+      "grad_norm": 1.360035413094489,
+      "learning_rate": 6.238369567579642e-07,
+      "loss": 1.1277,
+      "step": 298
+    },
+    {
+      "epoch": 0.4393828067597355,
+      "grad_norm": 1.3842334621043852,
+      "learning_rate": 6.215262173063607e-07,
+      "loss": 1.134,
+      "step": 299
+    },
+    {
+      "epoch": 0.440852314474651,
+      "grad_norm": 1.5072722906394864,
+      "learning_rate": 6.192127160175649e-07,
+      "loss": 1.1385,
+      "step": 300
+    },
+    {
+      "epoch": 0.4423218221895665,
+      "grad_norm": 1.3613054784017538,
+      "learning_rate": 6.168965054688238e-07,
+      "loss": 1.1117,
+      "step": 301
+    },
+    {
+      "epoch": 0.443791329904482,
+      "grad_norm": 1.4231270036725268,
+      "learning_rate": 6.145776382989552e-07,
+      "loss": 1.0889,
+      "step": 302
+    },
+    {
+      "epoch": 0.4452608376193975,
+      "grad_norm": 1.4588262408030779,
+      "learning_rate": 6.122561672071521e-07,
+      "loss": 1.0359,
+      "step": 303
+    },
+    {
+      "epoch": 0.446730345334313,
+      "grad_norm": 1.411260640294453,
+      "learning_rate": 6.099321449517851e-07,
+      "loss": 1.1288,
+      "step": 304
+    },
+    {
+      "epoch": 0.4481998530492285,
+      "grad_norm": 1.3956027122525714,
+      "learning_rate": 6.076056243492035e-07,
+      "loss": 1.1173,
+      "step": 305
+    },
+    {
+      "epoch": 0.449669360764144,
+      "grad_norm": 1.4227235203095976,
+      "learning_rate": 6.052766582725339e-07,
+      "loss": 1.0394,
+      "step": 306
+    },
+    {
+      "epoch": 0.45113886847905954,
+      "grad_norm": 1.3911528073579336,
+      "learning_rate": 6.029452996504801e-07,
+      "loss": 1.1346,
+      "step": 307
+    },
+    {
+      "epoch": 0.45260837619397504,
+      "grad_norm": 1.3598137504924201,
+      "learning_rate": 6.006116014661191e-07,
+      "loss": 1.1715,
+      "step": 308
+    },
+    {
+      "epoch": 0.45407788390889053,
+      "grad_norm": 1.3341402291681796,
+      "learning_rate": 5.982756167556978e-07,
+      "loss": 1.0628,
+      "step": 309
+    },
+    {
+      "epoch": 0.45554739162380603,
+      "grad_norm": 1.4336307282930423,
+      "learning_rate": 5.959373986074269e-07,
+      "loss": 1.1031,
+      "step": 310
+    },
+    {
+      "epoch": 0.4570168993387215,
+      "grad_norm": 1.4435742655464976,
+      "learning_rate": 5.935970001602751e-07,
+      "loss": 1.0705,
+      "step": 311
+    },
+    {
+      "epoch": 0.458486407053637,
+      "grad_norm": 1.295586975475687,
+      "learning_rate": 5.912544746027612e-07,
+      "loss": 1.082,
+      "step": 312
+    },
+    {
+      "epoch": 0.4599559147685525,
+      "grad_norm": 1.419413405526378,
+      "learning_rate": 5.88909875171745e-07,
+      "loss": 1.1675,
+      "step": 313
+    },
+    {
+      "epoch": 0.461425422483468,
+      "grad_norm": 1.437607892446711,
+      "learning_rate": 5.865632551512175e-07,
+      "loss": 1.0499,
+      "step": 314
+    },
+    {
+      "epoch": 0.4628949301983835,
+      "grad_norm": 1.308082043000319,
+      "learning_rate": 5.842146678710911e-07,
+      "loss": 1.1021,
+      "step": 315
+    },
+    {
+      "epoch": 0.46436443791329907,
+      "grad_norm": 1.4411999480884474,
+      "learning_rate": 5.818641667059856e-07,
+      "loss": 1.115,
+      "step": 316
+    },
+    {
+      "epoch": 0.46583394562821456,
+      "grad_norm": 1.4248859054466834,
+      "learning_rate": 5.795118050740169e-07,
+      "loss": 1.0735,
+      "step": 317
+    },
+    {
+      "epoch": 0.46730345334313006,
+      "grad_norm": 1.3950677919239058,
+      "learning_rate": 5.771576364355819e-07,
+      "loss": 1.1289,
+      "step": 318
+    },
+    {
+      "epoch": 0.46877296105804556,
+      "grad_norm": 1.4356515357949984,
+      "learning_rate": 5.748017142921448e-07,
+      "loss": 1.1771,
+      "step": 319
+    },
+    {
+      "epoch": 0.47024246877296105,
+      "grad_norm": 1.3803272984145392,
+      "learning_rate": 5.724440921850195e-07,
+      "loss": 1.2042,
+      "step": 320
+    },
+    {
+      "epoch": 0.47171197648787655,
+      "grad_norm": 1.3828275996220702,
+      "learning_rate": 5.700848236941543e-07,
+      "loss": 1.1014,
+      "step": 321
+    },
+    {
+      "epoch": 0.47318148420279205,
+      "grad_norm": 1.4372467929841282,
+      "learning_rate": 5.677239624369134e-07,
+      "loss": 1.0635,
+      "step": 322
+    },
+    {
+      "epoch": 0.47465099191770754,
+      "grad_norm": 1.4412482068570687,
+      "learning_rate": 5.653615620668589e-07,
+      "loss": 1.0709,
+      "step": 323
+    },
+    {
+      "epoch": 0.4761204996326231,
+      "grad_norm": 1.5223869305783915,
+      "learning_rate": 5.629976762725307e-07,
+      "loss": 1.0613,
+      "step": 324
+    },
+    {
+      "epoch": 0.4775900073475386,
+      "grad_norm": 1.4106384412205724,
+      "learning_rate": 5.606323587762275e-07,
+      "loss": 1.3318,
+      "step": 325
+    },
+    {
+      "epoch": 0.4790595150624541,
+      "grad_norm": 1.4808028294260986,
+      "learning_rate": 5.582656633327848e-07,
+      "loss": 1.1348,
+      "step": 326
+    },
+    {
+      "epoch": 0.4805290227773696,
+      "grad_norm": 1.411232450754018,
+      "learning_rate": 5.558976437283535e-07,
+      "loss": 1.1455,
+      "step": 327
+    },
+    {
+      "epoch": 0.4819985304922851,
+      "grad_norm": 1.3874875642549753,
+      "learning_rate": 5.535283537791785e-07,
+      "loss": 1.1502,
+      "step": 328
+    },
+    {
+      "epoch": 0.4834680382072006,
+      "grad_norm": 1.529379416962907,
+      "learning_rate": 5.511578473303742e-07,
+      "loss": 1.1306,
+      "step": 329
+    },
+    {
+      "epoch": 0.4849375459221161,
+      "grad_norm": 1.4353633142872315,
+      "learning_rate": 5.487861782547017e-07,
+      "loss": 1.1453,
+      "step": 330
+    },
+    {
+      "epoch": 0.4864070536370316,
+      "grad_norm": 1.3293126841762646,
+      "learning_rate": 5.464134004513442e-07,
+      "loss": 1.1422,
+      "step": 331
+    },
+    {
+      "epoch": 0.4878765613519471,
+      "grad_norm": 1.3853962798430477,
+      "learning_rate": 5.440395678446825e-07,
+      "loss": 1.1431,
+      "step": 332
+    },
+    {
+      "epoch": 0.4893460690668626,
+      "grad_norm": 1.394633338720216,
+      "learning_rate": 5.416647343830687e-07,
+      "loss": 1.0997,
+      "step": 333
+    },
+    {
+      "epoch": 0.4908155767817781,
+      "grad_norm": 1.3557987209086912,
+      "learning_rate": 5.392889540376006e-07,
+      "loss": 1.0817,
+      "step": 334
+    },
+    {
+      "epoch": 0.4922850844966936,
+      "grad_norm": 2.8096656165818645,
+      "learning_rate": 5.369122808008955e-07,
+      "loss": 1.101,
+      "step": 335
+    },
+    {
+      "epoch": 0.4937545922116091,
+      "grad_norm": 1.463042361823667,
+      "learning_rate": 5.345347686858626e-07,
+      "loss": 1.1002,
+      "step": 336
+    },
+    {
+      "epoch": 0.4952240999265246,
+      "grad_norm": 1.513004558088391,
+      "learning_rate": 5.321564717244757e-07,
+      "loss": 1.0977,
+      "step": 337
+    },
+    {
+      "epoch": 0.4966936076414401,
+      "grad_norm": 1.513966316502475,
+      "learning_rate": 5.297774439665449e-07,
+      "loss": 1.1179,
+      "step": 338
+    },
+    {
+      "epoch": 0.4981631153563556,
+      "grad_norm": 1.3659383294811929,
+      "learning_rate": 5.273977394784892e-07,
+      "loss": 1.1009,
+      "step": 339
+    },
+    {
+      "epoch": 0.4996326230712711,
+      "grad_norm": 1.4406236511077748,
+      "learning_rate": 5.250174123421068e-07,
+      "loss": 1.2597,
+      "step": 340
+    },
+    {
+      "epoch": 0.5011021307861866,
+      "grad_norm": 1.4120228303295752,
+      "learning_rate": 5.226365166533458e-07,
+      "loss": 1.1148,
+      "step": 341
+    },
+    {
+      "epoch": 0.5025716385011021,
+      "grad_norm": 1.3150193319220376,
+      "learning_rate": 5.202551065210768e-07,
+      "loss": 1.0589,
+      "step": 342
+    },
+    {
+      "epoch": 0.5040411462160176,
+      "grad_norm": 1.3552885803220251,
+      "learning_rate": 5.178732360658605e-07,
+      "loss": 1.1281,
+      "step": 343
+    },
+    {
+      "epoch": 0.5055106539309331,
+      "grad_norm": 1.3782165714758237,
+      "learning_rate": 5.154909594187192e-07,
+      "loss": 1.115,
+      "step": 344
+    },
+    {
+      "epoch": 0.5069801616458487,
+      "grad_norm": 1.3839651456991835,
+      "learning_rate": 5.131083307199071e-07,
+      "loss": 1.1355,
+      "step": 345
+    },
+    {
+      "epoch": 0.5084496693607642,
+      "grad_norm": 1.3239848936899146,
+      "learning_rate": 5.107254041176788e-07,
+      "loss": 1.144,
+      "step": 346
+    },
+    {
+      "epoch": 0.5099191770756797,
+      "grad_norm": 1.4174744669080177,
+      "learning_rate": 5.08342233767059e-07,
+      "loss": 1.0891,
+      "step": 347
+    },
+    {
+      "epoch": 0.5113886847905952,
+      "grad_norm": 1.4197584070932192,
+      "learning_rate": 5.059588738286118e-07,
+      "loss": 1.1711,
+      "step": 348
+    },
+    {
+      "epoch": 0.5128581925055107,
+      "grad_norm": 1.460332635848889,
+      "learning_rate": 5.035753784672105e-07,
+      "loss": 1.1173,
+      "step": 349
+    },
+    {
+      "epoch": 0.5143277002204262,
+      "grad_norm": 1.4181458807955096,
+      "learning_rate": 5.011918018508057e-07,
+      "loss": 1.1409,
+      "step": 350
+    },
+    {
+      "epoch": 0.5157972079353417,
+      "grad_norm": 1.401868758466356,
+      "learning_rate": 4.988081981491944e-07,
+      "loss": 1.08,
+      "step": 351
+    },
+    {
+      "epoch": 0.5172667156502572,
+      "grad_norm": 1.4708608243218393,
+      "learning_rate": 4.964246215327894e-07,
+      "loss": 1.2088,
+      "step": 352
+    },
+    {
+      "epoch": 0.5187362233651727,
+      "grad_norm": 1.3087914829912093,
+      "learning_rate": 4.940411261713882e-07,
+      "loss": 1.088,
+      "step": 353
+    },
+    {
+      "epoch": 0.5202057310800882,
+      "grad_norm": 1.3639757292403556,
+      "learning_rate": 4.91657766232941e-07,
+      "loss": 1.1503,
+      "step": 354
+    },
+    {
+      "epoch": 0.5216752387950037,
+      "grad_norm": 1.3230263031119276,
+      "learning_rate": 4.892745958823213e-07,
+      "loss": 1.096,
+      "step": 355
+    },
+    {
+      "epoch": 0.5231447465099192,
+      "grad_norm": 1.3801068609972764,
+      "learning_rate": 4.868916692800928e-07,
+      "loss": 1.1064,
+      "step": 356
+    },
+    {
+      "epoch": 0.5246142542248347,
+      "grad_norm": 1.4818081892670187,
+      "learning_rate": 4.845090405812809e-07,
+      "loss": 1.0788,
+      "step": 357
+    },
+    {
+      "epoch": 0.5260837619397501,
+      "grad_norm": 1.4257930922965831,
+      "learning_rate": 4.821267639341397e-07,
+      "loss": 1.0475,
+      "step": 358
+    },
+    {
+      "epoch": 0.5275532696546656,
+      "grad_norm": 1.3664827162861548,
+      "learning_rate": 4.797448934789232e-07,
+      "loss": 1.0764,
+      "step": 359
+    },
+    {
+      "epoch": 0.5290227773695811,
+      "grad_norm": 1.384036889671994,
+      "learning_rate": 4.773634833466541e-07,
+      "loss": 1.0491,
+      "step": 360
+    },
+    {
+      "epoch": 0.5304922850844966,
+      "grad_norm": 1.3641782448500568,
+      "learning_rate": 4.7498258765789335e-07,
+      "loss": 1.0597,
+      "step": 361
+    },
+    {
+      "epoch": 0.5319617927994122,
+      "grad_norm": 1.423492847149906,
+      "learning_rate": 4.726022605215108e-07,
+      "loss": 1.2022,
+      "step": 362
+    },
+    {
+      "epoch": 0.5334313005143277,
+      "grad_norm": 1.4584247852241217,
+      "learning_rate": 4.7022255603345504e-07,
+      "loss": 1.0439,
+      "step": 363
+    },
+    {
+      "epoch": 0.5349008082292432,
+      "grad_norm": 1.4065237109476447,
+      "learning_rate": 4.6784352827552433e-07,
+      "loss": 1.0945,
+      "step": 364
+    },
+    {
+      "epoch": 0.5363703159441587,
+      "grad_norm": 1.359008408282674,
+      "learning_rate": 4.6546523131413737e-07,
+      "loss": 1.0858,
+      "step": 365
+    },
+    {
+      "epoch": 0.5378398236590742,
+      "grad_norm": 1.395920829743356,
+      "learning_rate": 4.6308771919910455e-07,
+      "loss": 1.0821,
+      "step": 366
+    },
+    {
+      "epoch": 0.5393093313739897,
+      "grad_norm": 1.4101595258715152,
+      "learning_rate": 4.607110459623994e-07,
+      "loss": 1.0566,
+      "step": 367
+    },
+    {
+      "epoch": 0.5407788390889052,
+      "grad_norm": 1.4400551979551548,
+      "learning_rate": 4.5833526561693146e-07,
+      "loss": 1.0875,
+      "step": 368
+    },
+    {
+      "epoch": 0.5422483468038207,
+      "grad_norm": 1.4569650644657774,
+      "learning_rate": 4.559604321553176e-07,
+      "loss": 1.0945,
+      "step": 369
+    },
+    {
+      "epoch": 0.5437178545187362,
+      "grad_norm": 1.4073481731963642,
+      "learning_rate": 4.535865995486559e-07,
+      "loss": 1.1326,
+      "step": 370
+    },
+    {
+      "epoch": 0.5451873622336517,
+      "grad_norm": 1.3843052803402691,
+      "learning_rate": 4.512138217452984e-07,
+      "loss": 1.0445,
+      "step": 371
+    },
+    {
+      "epoch": 0.5466568699485672,
+      "grad_norm": 1.482116494998853,
+      "learning_rate": 4.488421526696259e-07,
+      "loss": 1.0871,
+      "step": 372
+    },
+    {
+      "epoch": 0.5481263776634827,
+      "grad_norm": 1.3248580273316146,
+      "learning_rate": 4.464716462208216e-07,
+      "loss": 1.0515,
+      "step": 373
+    },
+    {
+      "epoch": 0.5495958853783982,
+      "grad_norm": 1.412909824990006,
+      "learning_rate": 4.441023562716464e-07,
+      "loss": 1.1317,
+      "step": 374
+    },
+    {
+      "epoch": 0.5510653930933137,
+      "grad_norm": 1.3923300607754752,
+      "learning_rate": 4.417343366672154e-07,
+      "loss": 1.0727,
+      "step": 375
+    },
+    {
+      "epoch": 0.5525349008082292,
+      "grad_norm": 1.3989792278937028,
+      "learning_rate": 4.393676412237726e-07,
+      "loss": 1.1355,
+      "step": 376
+    },
+    {
+      "epoch": 0.5540044085231447,
+      "grad_norm": 1.3946392251166513,
+      "learning_rate": 4.370023237274693e-07,
+      "loss": 1.0268,
+      "step": 377
+    },
+    {
+      "epoch": 0.5554739162380602,
+      "grad_norm": 1.41416247304468,
+      "learning_rate": 4.3463843793314123e-07,
+      "loss": 1.0584,
+      "step": 378
+    },
+    {
+      "epoch": 0.5569434239529758,
+      "grad_norm": 1.413788607440302,
+      "learning_rate": 4.322760375630867e-07,
+      "loss": 1.1846,
+      "step": 379
+    },
+    {
+      "epoch": 0.5584129316678913,
+      "grad_norm": 1.3360695239659472,
+      "learning_rate": 4.299151763058457e-07,
+      "loss": 1.1525,
+      "step": 380
+    },
+    {
+      "epoch": 0.5598824393828068,
+      "grad_norm": 1.3928904548700607,
+      "learning_rate": 4.2755590781498056e-07,
+      "loss": 1.0295,
+      "step": 381
+    },
+    {
+      "epoch": 0.5613519470977223,
+      "grad_norm": 1.4059368688788658,
+      "learning_rate": 4.251982857078553e-07,
+      "loss": 1.0474,
+      "step": 382
+    },
+    {
+      "epoch": 0.5628214548126378,
+      "grad_norm": 1.5189950174163334,
+      "learning_rate": 4.2284236356441817e-07,
+      "loss": 1.1166,
+      "step": 383
+    },
+    {
+      "epoch": 0.5642909625275533,
+      "grad_norm": 1.5773519459794447,
+      "learning_rate": 4.204881949259832e-07,
+      "loss": 1.216,
+      "step": 384
+    },
+    {
+      "epoch": 0.5657604702424688,
+      "grad_norm": 1.36093051783186,
+      "learning_rate": 4.181358332940144e-07,
+      "loss": 1.1028,
+      "step": 385
+    },
+    {
+      "epoch": 0.5672299779573843,
+      "grad_norm": 1.371801222763194,
+      "learning_rate": 4.157853321289089e-07,
+      "loss": 1.1299,
+      "step": 386
+    },
+    {
+      "epoch": 0.5686994856722998,
+      "grad_norm": 1.4309586056121164,
+      "learning_rate": 4.1343674484878236e-07,
+      "loss": 1.0961,
+      "step": 387
+    },
+    {
+      "epoch": 0.5701689933872153,
+      "grad_norm": 1.3916946581373462,
+      "learning_rate": 4.11090124828255e-07,
+      "loss": 1.084,
+      "step": 388
+    },
+    {
+      "epoch": 0.5716385011021308,
+      "grad_norm": 1.3379667194162195,
+      "learning_rate": 4.0874552539723873e-07,
+      "loss": 1.0954,
+      "step": 389
+    },
+    {
+      "epoch": 0.5731080088170463,
+      "grad_norm": 1.4682746859969864,
+      "learning_rate": 4.064029998397247e-07,
+      "loss": 1.123,
+      "step": 390
+    },
+    {
+      "epoch": 0.5745775165319618,
+      "grad_norm": 1.3797386330911343,
+      "learning_rate": 4.04062601392573e-07,
+      "loss": 1.1146,
+      "step": 391
+    },
+    {
+      "epoch": 0.5760470242468773,
+      "grad_norm": 1.4080533490533538,
+      "learning_rate": 4.017243832443021e-07,
+      "loss": 1.1117,
+      "step": 392
+    },
+    {
+      "epoch": 0.5775165319617928,
+      "grad_norm": 1.4279995751396037,
+      "learning_rate": 3.993883985338808e-07,
+      "loss": 1.0809,
+      "step": 393
+    },
+    {
+      "epoch": 0.5789860396767083,
+      "grad_norm": 1.3175097784254468,
+      "learning_rate": 3.9705470034951986e-07,
+      "loss": 1.0567,
+      "step": 394
+    },
+    {
+      "epoch": 0.5804555473916239,
+      "grad_norm": 1.3671088379708558,
+      "learning_rate": 3.9472334172746596e-07,
+      "loss": 1.1201,
+      "step": 395
+    },
+    {
+      "epoch": 0.5819250551065394,
+      "grad_norm": 1.4728248748732817,
+      "learning_rate": 3.9239437565079645e-07,
+      "loss": 1.1274,
+      "step": 396
+    },
+    {
+      "epoch": 0.5833945628214549,
+      "grad_norm": 1.4007045355088419,
+      "learning_rate": 3.900678550482147e-07,
+      "loss": 1.1204,
+      "step": 397
+    },
+    {
+      "epoch": 0.5848640705363704,
+      "grad_norm": 1.3734381784823617,
+      "learning_rate": 3.877438327928478e-07,
+      "loss": 1.1203,
+      "step": 398
+    },
+    {
+      "epoch": 0.5863335782512858,
+      "grad_norm": 1.4011368340836778,
+      "learning_rate": 3.854223617010448e-07,
+      "loss": 1.0533,
+      "step": 399
+    },
+    {
+      "epoch": 0.5878030859662013,
+      "grad_norm": 1.419012334062334,
+      "learning_rate": 3.8310349453117617e-07,
+      "loss": 1.0898,
+      "step": 400
+    },
+    {
+      "epoch": 0.5892725936811168,
+      "grad_norm": 1.3485874322249378,
+      "learning_rate": 3.8078728398243503e-07,
+      "loss": 1.1228,
+      "step": 401
+    },
+    {
+      "epoch": 0.5907421013960323,
+      "grad_norm": 1.4516071343212644,
+      "learning_rate": 3.784737826936393e-07,
+      "loss": 1.0773,
+      "step": 402
+    },
+    {
+      "epoch": 0.5922116091109478,
+      "grad_norm": 1.2870748285607192,
+      "learning_rate": 3.761630432420358e-07,
+      "loss": 1.0489,
+      "step": 403
+    },
+    {
+      "epoch": 0.5936811168258633,
+      "grad_norm": 1.4253057600978802,
+      "learning_rate": 3.7385511814210493e-07,
+      "loss": 1.1059,
+      "step": 404
+    },
+    {
+      "epoch": 0.5951506245407788,
+      "grad_norm": 1.4362421490175297,
+      "learning_rate": 3.715500598443672e-07,
+      "loss": 1.0798,
+      "step": 405
+    },
+    {
+      "epoch": 0.5966201322556943,
+      "grad_norm": 1.3142558631846668,
+      "learning_rate": 3.6924792073419193e-07,
+      "loss": 1.095,
+      "step": 406
+    },
+    {
+      "epoch": 0.5980896399706098,
+      "grad_norm": 1.4180620072245396,
+      "learning_rate": 3.6694875313060567e-07,
+      "loss": 1.1618,
+      "step": 407
+    },
+    {
+      "epoch": 0.5995591476855253,
+      "grad_norm": 1.4440333294474847,
+      "learning_rate": 3.646526092851039e-07,
+      "loss": 1.1742,
+      "step": 408
+    },
+    {
+      "epoch": 0.6010286554004408,
+      "grad_norm": 1.365950277580121,
+      "learning_rate": 3.623595413804636e-07,
+      "loss": 1.1049,
+      "step": 409
+    },
+    {
+      "epoch": 0.6024981631153563,
+      "grad_norm": 1.3800394626203474,
+      "learning_rate": 3.600696015295568e-07,
+      "loss": 1.0812,
+      "step": 410
+    },
+    {
+      "epoch": 0.6039676708302718,
+      "grad_norm": 1.4309055297758246,
+      "learning_rate": 3.577828417741665e-07,
+      "loss": 1.1591,
+      "step": 411
+    },
+    {
+      "epoch": 0.6054371785451874,
+      "grad_norm": 1.4376053158203683,
+      "learning_rate": 3.5549931408380446e-07,
+      "loss": 1.1083,
+      "step": 412
+    },
+    {
+      "epoch": 0.6069066862601029,
+      "grad_norm": 1.378409944475253,
+      "learning_rate": 3.5321907035452913e-07,
+      "loss": 1.0625,
+      "step": 413
+    },
+    {
+      "epoch": 0.6083761939750184,
+      "grad_norm": 1.4032013297402026,
+      "learning_rate": 3.509421624077672e-07,
+      "loss": 1.099,
+      "step": 414
+    },
+    {
+      "epoch": 0.6098457016899339,
+      "grad_norm": 1.403659794537397,
+      "learning_rate": 3.486686419891349e-07,
+      "loss": 1.0488,
+      "step": 415
+    },
+    {
+      "epoch": 0.6113152094048494,
+      "grad_norm": 1.3788151443608003,
+      "learning_rate": 3.4639856076726346e-07,
+      "loss": 1.113,
+      "step": 416
+    },
+    {
+      "epoch": 0.6127847171197649,
+      "grad_norm": 1.4287954215942635,
+      "learning_rate": 3.4413197033262343e-07,
+      "loss": 1.0992,
+      "step": 417
+    },
+    {
+      "epoch": 0.6142542248346804,
+      "grad_norm": 1.439755106322846,
+      "learning_rate": 3.4186892219635254e-07,
+      "loss": 1.0662,
+      "step": 418
+    },
+    {
+      "epoch": 0.6157237325495959,
+      "grad_norm": 1.4201132330693378,
+      "learning_rate": 3.396094677890862e-07,
+      "loss": 1.1108,
+      "step": 419
+    },
+    {
+      "epoch": 0.6171932402645114,
+      "grad_norm": 1.4151859235424602,
+      "learning_rate": 3.373536584597869e-07,
+      "loss": 1.043,
+      "step": 420
+    },
+    {
+      "epoch": 0.6186627479794269,
+      "grad_norm": 1.3436922877791724,
+      "learning_rate": 3.3510154547457845e-07,
+      "loss": 1.1002,
+      "step": 421
+    },
+    {
+      "epoch": 0.6201322556943424,
+      "grad_norm": 1.3768814743005016,
+      "learning_rate": 3.3285318001558076e-07,
+      "loss": 1.099,
+      "step": 422
+    },
+    {
+      "epoch": 0.6216017634092579,
+      "grad_norm": 1.9818793407102133,
+      "learning_rate": 3.306086131797462e-07,
+      "loss": 1.101,
+      "step": 423
+    },
+    {
+      "epoch": 0.6230712711241734,
+      "grad_norm": 1.5501432398127912,
+      "learning_rate": 3.283678959776986e-07,
+      "loss": 1.1401,
+      "step": 424
+    },
+    {
+      "epoch": 0.6245407788390889,
+      "grad_norm": 1.3451070295410414,
+      "learning_rate": 3.261310793325742e-07,
+      "loss": 1.0917,
+      "step": 425
+    },
+    {
+      "epoch": 0.6260102865540044,
+      "grad_norm": 1.3998218088329646,
+      "learning_rate": 3.23898214078864e-07,
+      "loss": 1.1472,
+      "step": 426
+    },
+    {
+      "epoch": 0.6274797942689199,
+      "grad_norm": 1.4699494469190635,
+      "learning_rate": 3.216693509612587e-07,
+      "loss": 1.0937,
+      "step": 427
+    },
+    {
+      "epoch": 0.6289493019838354,
+      "grad_norm": 1.463414683070946,
+      "learning_rate": 3.19444540633495e-07,
+      "loss": 1.1389,
+      "step": 428
+    },
+    {
+      "epoch": 0.630418809698751,
+      "grad_norm": 1.4765841861514846,
+      "learning_rate": 3.172238336572056e-07,
+      "loss": 1.2141,
+      "step": 429
+    },
+    {
+      "epoch": 0.6318883174136665,
+      "grad_norm": 1.4207175135849464,
+      "learning_rate": 3.1500728050076873e-07,
+      "loss": 1.0713,
+      "step": 430
+    },
+    {
+      "epoch": 0.633357825128582,
+      "grad_norm": 1.3488069296637555,
+      "learning_rate": 3.1279493153816183e-07,
+      "loss": 1.0483,
+      "step": 431
+    },
+    {
+      "epoch": 0.6348273328434975,
+      "grad_norm": 1.4026690915785558,
+      "learning_rate": 3.1058683704781707e-07,
+      "loss": 1.0378,
+      "step": 432
+    },
+    {
+      "epoch": 0.636296840558413,
+      "grad_norm": 1.368530405030037,
+      "learning_rate": 3.0838304721147803e-07,
+      "loss": 1.0595,
+      "step": 433
+    },
+    {
+      "epoch": 0.6377663482733285,
+      "grad_norm": 1.3798655085044127,
+      "learning_rate": 3.0618361211305956e-07,
+      "loss": 1.1855,
+      "step": 434
+    },
+    {
+      "epoch": 0.639235855988244,
+      "grad_norm": 1.4358356459308954,
+      "learning_rate": 3.0398858173750994e-07,
+      "loss": 1.1314,
+      "step": 435
+    },
+    {
+      "epoch": 0.6407053637031594,
+      "grad_norm": 1.3911243386477645,
+      "learning_rate": 3.0179800596967414e-07,
+      "loss": 1.1479,
+      "step": 436
+    },
+    {
+      "epoch": 0.6421748714180749,
+      "grad_norm": 1.3787211953067466,
+      "learning_rate": 2.996119345931607e-07,
+      "loss": 1.0852,
+      "step": 437
+    },
+    {
+      "epoch": 0.6436443791329904,
+      "grad_norm": 1.3784361879287548,
+      "learning_rate": 2.9743041728921004e-07,
+      "loss": 1.0617,
+      "step": 438
+    },
+    {
+      "epoch": 0.6451138868479059,
+      "grad_norm": 1.4563498056244404,
+      "learning_rate": 2.952535036355659e-07,
+      "loss": 1.0172,
+      "step": 439
+    },
+    {
+      "epoch": 0.6465833945628214,
+      "grad_norm": 1.3911833970178598,
+      "learning_rate": 2.930812431053477e-07,
+      "loss": 1.1772,
+      "step": 440
+    },
+    {
+      "epoch": 0.6480529022777369,
+      "grad_norm": 1.456141852044334,
+      "learning_rate": 2.9091368506592704e-07,
+      "loss": 1.1163,
+      "step": 441
+    },
+    {
+      "epoch": 0.6495224099926524,
+      "grad_norm": 1.4409576849555352,
+      "learning_rate": 2.8875087877780547e-07,
+      "loss": 1.0099,
+      "step": 442
+    },
+    {
+      "epoch": 0.6509919177075679,
+      "grad_norm": 1.3739666211555663,
+      "learning_rate": 2.865928733934951e-07,
+      "loss": 1.074,
+      "step": 443
+    },
+    {
+      "epoch": 0.6524614254224834,
+      "grad_norm": 1.4333872958890426,
+      "learning_rate": 2.844397179564009e-07,
+      "loss": 1.119,
+      "step": 444
+    },
+    {
+      "epoch": 0.6539309331373989,
+      "grad_norm": 1.307442428138931,
+      "learning_rate": 2.8229146139970725e-07,
+      "loss": 1.0908,
+      "step": 445
+    },
+    {
+      "epoch": 0.6554004408523145,
+      "grad_norm": 1.364990935232205,
+      "learning_rate": 2.8014815254526475e-07,
+      "loss": 1.0506,
+      "step": 446
+    },
+    {
+      "epoch": 0.65686994856723,
+      "grad_norm": 1.4165052577271036,
+      "learning_rate": 2.780098401024816e-07,
+      "loss": 1.1602,
+      "step": 447
+    },
+    {
+      "epoch": 0.6583394562821455,
+      "grad_norm": 1.4602397372513622,
+      "learning_rate": 2.7587657266721633e-07,
+      "loss": 1.0126,
+      "step": 448
+    },
+    {
+      "epoch": 0.659808963997061,
+      "grad_norm": 1.3583523079909423,
+      "learning_rate": 2.737483987206725e-07,
+      "loss": 1.0467,
+      "step": 449
+    },
+    {
+      "epoch": 0.6612784717119765,
+      "grad_norm": 1.376858530036489,
+      "learning_rate": 2.7162536662829836e-07,
+      "loss": 1.0401,
+      "step": 450
+    },
+    {
+      "epoch": 0.662747979426892,
+      "grad_norm": 1.403999668213476,
+      "learning_rate": 2.695075246386874e-07,
+      "loss": 1.0438,
+      "step": 451
+    },
+    {
+      "epoch": 0.6642174871418075,
+      "grad_norm": 1.3971075292397948,
+      "learning_rate": 2.673949208824804e-07,
+      "loss": 1.0229,
+      "step": 452
+    },
+    {
+      "epoch": 0.665686994856723,
+      "grad_norm": 1.4601504097247877,
+      "learning_rate": 2.6528760337127344e-07,
+      "loss": 1.0741,
+      "step": 453
+    },
+    {
+      "epoch": 0.6671565025716385,
+      "grad_norm": 1.3882632206062708,
+      "learning_rate": 2.6318561999652543e-07,
+      "loss": 1.1148,
+      "step": 454
+    },
+    {
+      "epoch": 0.668626010286554,
+      "grad_norm": 1.4454566469056964,
+      "learning_rate": 2.610890185284707e-07,
+      "loss": 1.155,
+      "step": 455
+    },
+    {
+      "epoch": 0.6700955180014695,
+      "grad_norm": 1.4144159704417698,
+      "learning_rate": 2.5899784661503306e-07,
+      "loss": 1.1574,
+      "step": 456
+    },
+    {
+      "epoch": 0.671565025716385,
+      "grad_norm": 1.3227253044283795,
+      "learning_rate": 2.569121517807421e-07,
+      "loss": 1.0814,
+      "step": 457
+    },
+    {
+      "epoch": 0.6730345334313005,
+      "grad_norm": 1.4016559184728539,
+      "learning_rate": 2.5483198142565454e-07,
+      "loss": 1.1234,
+      "step": 458
+    },
+    {
+      "epoch": 0.674504041146216,
+      "grad_norm": 1.407280713108974,
+      "learning_rate": 2.5275738282427627e-07,
+      "loss": 1.1664,
+      "step": 459
+    },
+    {
+      "epoch": 0.6759735488611315,
+      "grad_norm": 1.4047211167155722,
+      "learning_rate": 2.506884031244875e-07,
+      "loss": 1.0663,
+      "step": 460
+    },
+    {
+      "epoch": 0.677443056576047,
+      "grad_norm": 1.4210107328567902,
+      "learning_rate": 2.4862508934647215e-07,
+      "loss": 1.0806,
+      "step": 461
+    },
+    {
+      "epoch": 0.6789125642909625,
+      "grad_norm": 1.3638119047997457,
+      "learning_rate": 2.465674883816492e-07,
+      "loss": 1.1283,
+      "step": 462
+    },
+    {
+      "epoch": 0.6803820720058781,
+      "grad_norm": 1.3846202967932413,
+      "learning_rate": 2.445156469916059e-07,
+      "loss": 1.0391,
+      "step": 463
+    },
+    {
+      "epoch": 0.6818515797207936,
+      "grad_norm": 1.4492638193493528,
+      "learning_rate": 2.424696118070367e-07,
+      "loss": 1.1258,
+      "step": 464
+    },
+    {
+      "epoch": 0.6833210874357091,
+      "grad_norm": 1.3695917043777497,
+      "learning_rate": 2.404294293266823e-07,
+      "loss": 1.0027,
+      "step": 465
+    },
+    {
+      "epoch": 0.6847905951506246,
+      "grad_norm": 1.3993284836421556,
+      "learning_rate": 2.3839514591627298e-07,
+      "loss": 1.1171,
+      "step": 466
+    },
+    {
+      "epoch": 0.6862601028655401,
+      "grad_norm": 1.3323985034089447,
+      "learning_rate": 2.3636680780747574e-07,
+      "loss": 1.1226,
+      "step": 467
+    },
+    {
+      "epoch": 0.6877296105804556,
+      "grad_norm": 1.3810241812868833,
+      "learning_rate": 2.3434446109684303e-07,
+      "loss": 1.1231,
+      "step": 468
+    },
+    {
+      "epoch": 0.6891991182953711,
+      "grad_norm": 1.3868467700476377,
+      "learning_rate": 2.323281517447646e-07,
+      "loss": 1.1609,
+      "step": 469
+    },
+    {
+      "epoch": 0.6906686260102866,
+      "grad_norm": 1.4013984837064835,
+      "learning_rate": 2.3031792557442426e-07,
+      "loss": 1.0457,
+      "step": 470
+    },
+    {
+      "epoch": 0.692138133725202,
+      "grad_norm": 1.4712416790362257,
+      "learning_rate": 2.2831382827075758e-07,
+      "loss": 1.0643,
+      "step": 471
+    },
+    {
+      "epoch": 0.6936076414401176,
+      "grad_norm": 1.3781792405401019,
+      "learning_rate": 2.2631590537941348e-07,
+      "loss": 1.0947,
+      "step": 472
+    },
+    {
+      "epoch": 0.695077149155033,
+      "grad_norm": 1.4117456798087704,
+      "learning_rate": 2.2432420230572014e-07,
+      "loss": 1.1242,
+      "step": 473
+    },
+    {
+      "epoch": 0.6965466568699485,
+      "grad_norm": 1.3486912287876163,
+      "learning_rate": 2.223387643136524e-07,
+      "loss": 1.1288,
+      "step": 474
+    },
+    {
+      "epoch": 0.698016164584864,
+      "grad_norm": 1.4003361923857591,
+      "learning_rate": 2.2035963652480266e-07,
+      "loss": 1.0847,
+      "step": 475
+    },
+    {
+      "epoch": 0.6994856722997795,
+      "grad_norm": 1.4405126411409146,
+      "learning_rate": 2.183868639173568e-07,
+      "loss": 1.0318,
+      "step": 476
+    },
+    {
+      "epoch": 0.700955180014695,
+      "grad_norm": 1.3517140108244523,
+      "learning_rate": 2.1642049132507013e-07,
+      "loss": 1.1525,
+      "step": 477
+    },
+    {
+      "epoch": 0.7024246877296105,
+      "grad_norm": 1.401420474546432,
+      "learning_rate": 2.144605634362504e-07,
+      "loss": 1.0637,
+      "step": 478
+    },
+    {
+      "epoch": 0.7038941954445261,
+      "grad_norm": 1.4272891896192437,
+      "learning_rate": 2.125071247927412e-07,
+      "loss": 1.0649,
+      "step": 479
+    },
+    {
+      "epoch": 0.7053637031594416,
+      "grad_norm": 1.4147072802744984,
+      "learning_rate": 2.1056021978890915e-07,
+      "loss": 1.0819,
+      "step": 480
+    },
+    {
+      "epoch": 0.7068332108743571,
+      "grad_norm": 1.489071300786935,
+      "learning_rate": 2.0861989267063622e-07,
+      "loss": 1.1086,
+      "step": 481
+    },
+    {
+      "epoch": 0.7083027185892726,
+      "grad_norm": 1.3674178536485293,
+      "learning_rate": 2.0668618753431372e-07,
+      "loss": 1.126,
+      "step": 482
+    },
+    {
+      "epoch": 0.7097722263041881,
+      "grad_norm": 1.4610080335049196,
+      "learning_rate": 2.0475914832583936e-07,
+      "loss": 1.1212,
+      "step": 483
+    },
+    {
+      "epoch": 0.7112417340191036,
+      "grad_norm": 1.3760740804936542,
+      "learning_rate": 2.0283881883961978e-07,
+      "loss": 1.1276,
+      "step": 484
+    },
+    {
+      "epoch": 0.7127112417340191,
+      "grad_norm": 1.3670564421892966,
+      "learning_rate": 2.0092524271757472e-07,
+      "loss": 1.0993,
+      "step": 485
+    },
+    {
+      "epoch": 0.7141807494489346,
+      "grad_norm": 1.394984491530565,
+      "learning_rate": 1.990184634481446e-07,
+      "loss": 1.089,
+      "step": 486
+    },
+    {
+      "epoch": 0.7156502571638501,
+      "grad_norm": 1.4571577516475638,
+      "learning_rate": 1.9711852436530318e-07,
+      "loss": 1.125,
+      "step": 487
+    },
+    {
+      "epoch": 0.7171197648787656,
+      "grad_norm": 1.5499839330419916,
+      "learning_rate": 1.952254686475726e-07,
+      "loss": 1.1264,
+      "step": 488
+    },
+    {
+      "epoch": 0.7185892725936811,
+      "grad_norm": 1.3781629330474126,
+      "learning_rate": 1.9333933931704098e-07,
+      "loss": 1.0566,
+      "step": 489
+    },
+    {
+      "epoch": 0.7200587803085966,
+      "grad_norm": 1.4387738954738303,
+      "learning_rate": 1.914601792383862e-07,
+      "loss": 1.0665,
+      "step": 490
+    },
+    {
+      "epoch": 0.7215282880235121,
+      "grad_norm": 1.3566125913075402,
+      "learning_rate": 1.8958803111790105e-07,
+      "loss": 1.0703,
+      "step": 491
+    },
+    {
+      "epoch": 0.7229977957384276,
+      "grad_norm": 1.3556380519533342,
+      "learning_rate": 1.877229375025222e-07,
+      "loss": 1.1557,
+      "step": 492
+    },
+    {
+      "epoch": 0.7244673034533431,
+      "grad_norm": 1.3370842913006449,
+      "learning_rate": 1.8586494077886416e-07,
+      "loss": 1.098,
+      "step": 493
+    },
+    {
+      "epoch": 0.7259368111682586,
+      "grad_norm": 1.5097558318383415,
+      "learning_rate": 1.840140831722557e-07,
+      "loss": 1.1044,
+      "step": 494
+    },
+    {
+      "epoch": 0.7274063188831741,
+      "grad_norm": 1.4260075475587533,
+      "learning_rate": 1.821704067457795e-07,
+      "loss": 1.0891,
+      "step": 495
+    },
+    {
+      "epoch": 0.7288758265980897,
+      "grad_norm": 1.4163474844245585,
+      "learning_rate": 1.803339533993175e-07,
+      "loss": 1.1143,
+      "step": 496
+    },
+    {
+      "epoch": 0.7303453343130052,
+      "grad_norm": 1.3471773518411894,
+      "learning_rate": 1.7850476486859784e-07,
+      "loss": 1.1001,
+      "step": 497
+    },
+    {
+      "epoch": 0.7318148420279207,
+      "grad_norm": 1.349662778135853,
+      "learning_rate": 1.766828827242461e-07,
+      "loss": 1.0691,
+      "step": 498
+    },
+    {
+      "epoch": 0.7332843497428362,
+      "grad_norm": 1.413952156125001,
+      "learning_rate": 1.7486834837084147e-07,
+      "loss": 1.1791,
+      "step": 499
+    },
+    {
+      "epoch": 0.7347538574577517,
+      "grad_norm": 1.4517736063725735,
+      "learning_rate": 1.7306120304597516e-07,
+      "loss": 1.1076,
+      "step": 500
+    },
+    {
+      "epoch": 0.7347538574577517,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_loss": 2.7672901153564453,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_runtime": 5.874,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_samples_per_second": 15.832,
+      "eval_ical_mcts_chains_sft_val_MORECHAINS_steps_per_second": 2.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.7362233651726672,
+      "grad_norm": 1.4017708695893076,
+      "learning_rate": 1.7126148781931309e-07,
+      "loss": 1.2015,
+      "step": 501
+    },
+    {
+      "epoch": 0.7376928728875827,
+      "grad_norm": 1.373359024365262,
+      "learning_rate": 1.6946924359166332e-07,
+      "loss": 1.1271,
+      "step": 502
+    },
+    {
+      "epoch": 0.7391623806024982,
+      "grad_norm": 1.3527598369707952,
+      "learning_rate": 1.6768451109404518e-07,
+      "loss": 1.1452,
+      "step": 503
+    },
+    {
+      "epoch": 0.7406318883174137,
+      "grad_norm": 1.4342813387133044,
+      "learning_rate": 1.659073308867653e-07,
+      "loss": 1.0873,
+      "step": 504
+    },
+    {
+      "epoch": 0.7421013960323292,
+      "grad_norm": 1.3934397068509279,
+      "learning_rate": 1.641377433584945e-07,
+      "loss": 1.1442,
+      "step": 505
+    },
+    {
+      "epoch": 0.7435709037472447,
+      "grad_norm": 1.3468387357646852,
+      "learning_rate": 1.6237578872535023e-07,
+      "loss": 1.0956,
+      "step": 506
+    },
+    {
+      "epoch": 0.7450404114621602,
+      "grad_norm": 1.3379393102784192,
+      "learning_rate": 1.6062150702998307e-07,
+      "loss": 1.1059,
+      "step": 507
+    },
+    {
+      "epoch": 0.7465099191770757,
+      "grad_norm": 1.4208106676166712,
+      "learning_rate": 1.5887493814066632e-07,
+      "loss": 1.0594,
+      "step": 508
+    },
+    {
+      "epoch": 0.7479794268919912,
+      "grad_norm": 1.3284093386527018,
+      "learning_rate": 1.5713612175038953e-07,
+      "loss": 1.0709,
+      "step": 509
+    },
+    {
+      "epoch": 0.7494489346069066,
+      "grad_norm": 1.3401638532562588,
+      "learning_rate": 1.5540509737595752e-07,
+      "loss": 1.0478,
+      "step": 510
+    },
+    {
+      "epoch": 0.7509184423218221,
+      "grad_norm": 1.387779204411568,
+      "learning_rate": 1.536819043570915e-07,
+      "loss": 1.0749,
+      "step": 511
+    },
+    {
+      "epoch": 0.7523879500367376,
+      "grad_norm": 1.3361938042809636,
+      "learning_rate": 1.5196658185553484e-07,
+      "loss": 1.0633,
+      "step": 512
+    },
+    {
+      "epoch": 0.7538574577516532,
+      "grad_norm": 1.4464634414727173,
+      "learning_rate": 1.5025916885416385e-07,
+      "loss": 1.0915,
+      "step": 513
+    },
+    {
+      "epoch": 0.7553269654665687,
+      "grad_norm": 1.4564936678530274,
+      "learning_rate": 1.485597041561014e-07,
+      "loss": 1.181,
+      "step": 514
+    },
+    {
+      "epoch": 0.7567964731814842,
+      "grad_norm": 1.3813741239097441,
+      "learning_rate": 1.4686822638383485e-07,
+      "loss": 1.1353,
+      "step": 515
+    },
+    {
+      "epoch": 0.7582659808963997,
+      "grad_norm": 1.3284963146667195,
+      "learning_rate": 1.4518477397833868e-07,
+      "loss": 1.0876,
+      "step": 516
+    },
+    {
+      "epoch": 0.7597354886113152,
+      "grad_norm": 1.3825280093987735,
+      "learning_rate": 1.4350938519820082e-07,
+      "loss": 1.1206,
+      "step": 517
+    },
+    {
+      "epoch": 0.7612049963262307,
+      "grad_norm": 1.4124836120841386,
+      "learning_rate": 1.4184209811875314e-07,
+      "loss": 1.0521,
+      "step": 518
+    },
+    {
+      "epoch": 0.7626745040411462,
+      "grad_norm": 1.3957995246414707,
+      "learning_rate": 1.401829506312061e-07,
+      "loss": 1.1441,
+      "step": 519
+    },
+    {
+      "epoch": 0.7641440117560617,
+      "grad_norm": 2.064618166415374,
+      "learning_rate": 1.385319804417872e-07,
+      "loss": 1.0951,
+      "step": 520
+    },
+    {
+      "epoch": 0.7656135194709772,
+      "grad_norm": 1.3751431109514682,
+      "learning_rate": 1.3688922507088506e-07,
+      "loss": 1.0622,
+      "step": 521
+    },
+    {
+      "epoch": 0.7670830271858927,
+      "grad_norm": 1.4467861329866494,
+      "learning_rate": 1.35254721852196e-07,
+      "loss": 1.1695,
+      "step": 522
+    },
+    {
+      "epoch": 0.7685525349008082,
+      "grad_norm": 1.37927918872005,
+      "learning_rate": 1.3362850793187536e-07,
+      "loss": 1.1666,
+      "step": 523
+    },
+    {
+      "epoch": 0.7700220426157237,
+      "grad_norm": 1.402761123927447,
+      "learning_rate": 1.3201062026769415e-07,
+      "loss": 1.1472,
+      "step": 524
+    },
+    {
+      "epoch": 0.7714915503306392,
+      "grad_norm": 1.4020269134593764,
+      "learning_rate": 1.3040109562819852e-07,
+      "loss": 1.1132,
+      "step": 525
+    },
+    {
+      "epoch": 0.7729610580455547,
+      "grad_norm": 1.3179769606979768,
+      "learning_rate": 1.2879997059187402e-07,
+      "loss": 1.1577,
+      "step": 526
+    },
+    {
+      "epoch": 0.7744305657604702,
+      "grad_norm": 1.3608872827916707,
+      "learning_rate": 1.27207281546315e-07,
+      "loss": 1.105,
+      "step": 527
+    },
+    {
+      "epoch": 0.7759000734753857,
+      "grad_norm": 1.4191003452870128,
+      "learning_rate": 1.2562306468739707e-07,
+      "loss": 1.061,
+      "step": 528
+    },
+    {
+      "epoch": 0.7773695811903012,
+      "grad_norm": 1.42228631098522,
+      "learning_rate": 1.2404735601845446e-07,
+      "loss": 1.0901,
+      "step": 529
+    },
+    {
+      "epoch": 0.7788390889052168,
+      "grad_norm": 1.374432034447507,
+      "learning_rate": 1.2248019134946224e-07,
+      "loss": 1.1478,
+      "step": 530
+    },
+    {
+      "epoch": 0.7803085966201323,
+      "grad_norm": 1.4436649330903892,
+      "learning_rate": 1.2092160629622243e-07,
+      "loss": 1.2448,
+      "step": 531
+    },
+    {
+      "epoch": 0.7817781043350478,
+      "grad_norm": 1.3822744122903736,
+      "learning_rate": 1.1937163627955388e-07,
+      "loss": 1.0715,
+      "step": 532
+    },
+    {
+      "epoch": 0.7832476120499633,
+      "grad_norm": 1.4003007054983763,
+      "learning_rate": 1.1783031652448844e-07,
+      "loss": 1.0834,
+      "step": 533
+    },
+    {
+      "epoch": 0.7847171197648788,
+      "grad_norm": 1.3009002478832499,
+      "learning_rate": 1.1629768205946916e-07,
+      "loss": 1.0867,
+      "step": 534
+    },
+    {
+      "epoch": 0.7861866274797943,
+      "grad_norm": 1.384225049312056,
+      "learning_rate": 1.1477376771555547e-07,
+      "loss": 1.111,
+      "step": 535
+    },
+    {
+      "epoch": 0.7876561351947098,
+      "grad_norm": 1.4951881255718384,
+      "learning_rate": 1.1325860812563082e-07,
+      "loss": 1.0905,
+      "step": 536
+    },
+    {
+      "epoch": 0.7891256429096253,
+      "grad_norm": 1.3855016278493693,
+      "learning_rate": 1.1175223772361548e-07,
+      "loss": 1.1674,
+      "step": 537
+    },
+    {
+      "epoch": 0.7905951506245408,
+      "grad_norm": 1.402883620503283,
+      "learning_rate": 1.1025469074368465e-07,
+      "loss": 1.0934,
+      "step": 538
+    },
+    {
+      "epoch": 0.7920646583394563,
+      "grad_norm": 1.3721282887774249,
+      "learning_rate": 1.0876600121949014e-07,
+      "loss": 1.053,
+      "step": 539
+    },
+    {
+      "epoch": 0.7935341660543718,
+      "grad_norm": 1.3027250754946427,
+      "learning_rate": 1.0728620298338647e-07,
+      "loss": 0.9932,
+      "step": 540
+    },
+    {
+      "epoch": 0.7950036737692873,
+      "grad_norm": 1.3472091716626784,
+      "learning_rate": 1.058153296656627e-07,
+      "loss": 1.0411,
+      "step": 541
+    },
+    {
+      "epoch": 0.7964731814842028,
+      "grad_norm": 1.2515159561759799,
+      "learning_rate": 1.0435341469377785e-07,
+      "loss": 1.0409,
+      "step": 542
+    },
+    {
+      "epoch": 0.7979426891991183,
+      "grad_norm": 1.3955411540097424,
+      "learning_rate": 1.0290049129160083e-07,
+      "loss": 1.0847,
+      "step": 543
+    },
+    {
+      "epoch": 0.7994121969140338,
+      "grad_norm": 1.378388349777933,
+      "learning_rate": 1.0145659247865606e-07,
+      "loss": 1.0637,
+      "step": 544
+    },
+    {
+      "epoch": 0.8008817046289493,
+      "grad_norm": 1.3665203225260263,
+      "learning_rate": 1.0002175106937282e-07,
+      "loss": 1.083,
+      "step": 545
+    },
+    {
+      "epoch": 0.8023512123438648,
+      "grad_norm": 1.3325384707568575,
+      "learning_rate": 9.859599967233901e-08,
+      "loss": 1.0619,
+      "step": 546
+    },
+    {
+      "epoch": 0.8038207200587804,
+      "grad_norm": 1.3079374620261692,
+      "learning_rate": 9.717937068956083e-08,
+      "loss": 1.0893,
+      "step": 547
+    },
+    {
+      "epoch": 0.8052902277736959,
+      "grad_norm": 1.386818107212991,
+      "learning_rate": 9.577189631572613e-08,
+      "loss": 1.1115,
+      "step": 548
+    },
+    {
+      "epoch": 0.8067597354886114,
+      "grad_norm": 1.374316392770813,
+      "learning_rate": 9.437360853747223e-08,
+      "loss": 1.1449,
+      "step": 549
+    },
+    {
+      "epoch": 0.8082292432035268,
+      "grad_norm": 1.345287353676708,
+      "learning_rate": 9.29845391326598e-08,
+      "loss": 1.0576,
+      "step": 550
+    },
+    {
+      "epoch": 0.8096987509184423,
+      "grad_norm": 1.4813724241887494,
+      "learning_rate": 9.16047196696505e-08,
+      "loss": 1.092,
+      "step": 551
+    },
+    {
+      "epoch": 0.8111682586333578,
+      "grad_norm": 1.3806739470510292,
+      "learning_rate": 9.023418150658863e-08,
+      "loss": 1.0925,
+      "step": 552
+    },
+    {
+      "epoch": 0.8126377663482733,
+      "grad_norm": 1.3661938482257079,
+      "learning_rate": 8.887295579068988e-08,
+      "loss": 1.0278,
+      "step": 553
+    },
+    {
+      "epoch": 0.8141072740631888,
+      "grad_norm": 1.376286461998827,
+      "learning_rate": 8.752107345753262e-08,
+      "loss": 1.0203,
+      "step": 554
+    },
+    {
+      "epoch": 0.8155767817781043,
+      "grad_norm": 1.3714175983707415,
+      "learning_rate": 8.617856523035466e-08,
+      "loss": 1.0947,
+      "step": 555
+    },
+    {
+      "epoch": 0.8170462894930198,
+      "grad_norm": 1.3804861761749325,
+      "learning_rate": 8.484546161935596e-08,
+      "loss": 1.0772,
+      "step": 556
+    },
+    {
+      "epoch": 0.8185157972079353,
+      "grad_norm": 1.374313789743847,
+      "learning_rate": 8.352179292100403e-08,
+      "loss": 1.0653,
+      "step": 557
+    },
+    {
+      "epoch": 0.8199853049228508,
+      "grad_norm": 1.3618284844063149,
+      "learning_rate": 8.220758921734649e-08,
+      "loss": 1.0513,
+      "step": 558
+    },
+    {
+      "epoch": 0.8214548126377663,
+      "grad_norm": 1.3803405029205986,
+      "learning_rate": 8.090288037532706e-08,
+      "loss": 1.1005,
+      "step": 559
+    },
+    {
+      "epoch": 0.8229243203526818,
+      "grad_norm": 1.3988974278658783,
+      "learning_rate": 7.960769604610618e-08,
+      "loss": 1.1205,
+      "step": 560
+    },
+    {
+      "epoch": 0.8243938280675973,
+      "grad_norm": 1.3526087062032914,
+      "learning_rate": 7.83220656643881e-08,
+      "loss": 1.0322,
+      "step": 561
+    },
+    {
+      "epoch": 0.8258633357825128,
+      "grad_norm": 1.3213029228844033,
+      "learning_rate": 7.704601844775155e-08,
+      "loss": 1.071,
+      "step": 562
+    },
+    {
+      "epoch": 0.8273328434974284,
+      "grad_norm": 1.4988675658645414,
+      "learning_rate": 7.577958339598529e-08,
+      "loss": 1.0068,
+      "step": 563
+    },
+    {
+      "epoch": 0.8288023512123439,
+      "grad_norm": 1.4472214140414408,
+      "learning_rate": 7.452278929042982e-08,
+      "loss": 1.1286,
+      "step": 564
+    },
+    {
+      "epoch": 0.8302718589272594,
+      "grad_norm": 1.3652634375890536,
+      "learning_rate": 7.327566469332303e-08,
+      "loss": 1.1308,
+      "step": 565
+    },
+    {
+      "epoch": 0.8317413666421749,
+      "grad_norm": 1.3655886385805245,
+      "learning_rate": 7.203823794715041e-08,
+      "loss": 1.1015,
+      "step": 566
+    },
+    {
+      "epoch": 0.8332108743570904,
+      "grad_norm": 1.2502888959822873,
+      "learning_rate": 7.08105371740021e-08,
+      "loss": 1.0829,
+      "step": 567
+    },
+    {
+      "epoch": 0.8346803820720059,
+      "grad_norm": 1.366330743048358,
+      "learning_rate": 6.959259027493303e-08,
+      "loss": 1.1187,
+      "step": 568
+    },
+    {
+      "epoch": 0.8361498897869214,
+      "grad_norm": 1.3645999375102753,
+      "learning_rate": 6.838442492932867e-08,
+      "loss": 1.0662,
+      "step": 569
+    },
+    {
+      "epoch": 0.8376193975018369,
+      "grad_norm": 1.4139397317247722,
+      "learning_rate": 6.718606859427673e-08,
+      "loss": 1.1166,
+      "step": 570
+    },
+    {
+      "epoch": 0.8390889052167524,
+      "grad_norm": 1.4003308125547838,
+      "learning_rate": 6.599754850394263e-08,
+      "loss": 1.1006,
+      "step": 571
+    },
+    {
+      "epoch": 0.8405584129316679,
+      "grad_norm": 1.3719288368595044,
+      "learning_rate": 6.481889166895033e-08,
+      "loss": 1.0538,
+      "step": 572
+    },
+    {
+      "epoch": 0.8420279206465834,
+      "grad_norm": 1.3259310033546516,
+      "learning_rate": 6.365012487576926e-08,
+      "loss": 1.0573,
+      "step": 573
+    },
+    {
+      "epoch": 0.8434974283614989,
+      "grad_norm": 1.4183409718497981,
+      "learning_rate": 6.249127468610504e-08,
+      "loss": 1.1412,
+      "step": 574
+    },
+    {
+      "epoch": 0.8449669360764144,
+      "grad_norm": 1.3764865756818885,
+      "learning_rate": 6.134236743629562e-08,
+      "loss": 1.1218,
+      "step": 575
+    },
+    {
+      "epoch": 0.8464364437913299,
+      "grad_norm": 1.4471292335999226,
+      "learning_rate": 6.020342923671334e-08,
+      "loss": 1.0495,
+      "step": 576
+    },
+    {
+      "epoch": 0.8479059515062454,
+      "grad_norm": 1.4152415184940785,
+      "learning_rate": 5.907448597117126e-08,
+      "loss": 1.1413,
+      "step": 577
+    },
+    {
+      "epoch": 0.8493754592211609,
+      "grad_norm": 1.3749110659106119,
+      "learning_rate": 5.7955563296334664e-08,
+      "loss": 1.1659,
+      "step": 578
+    },
+    {
+      "epoch": 0.8508449669360764,
+      "grad_norm": 1.4403389990835915,
+      "learning_rate": 5.6846686641138394e-08,
+      "loss": 1.0724,
+      "step": 579
+    },
+    {
+      "epoch": 0.852314474650992,
+      "grad_norm": 1.3113544123287635,
+      "learning_rate": 5.5747881206208936e-08,
+      "loss": 1.1002,
+      "step": 580
+    },
+    {
+      "epoch": 0.8537839823659075,
+      "grad_norm": 1.4504739763414178,
+      "learning_rate": 5.465917196329106e-08,
+      "loss": 1.0829,
+      "step": 581
+    },
+    {
+      "epoch": 0.855253490080823,
+      "grad_norm": 1.461654698052296,
+      "learning_rate": 5.3580583654681266e-08,
+      "loss": 1.0559,
+      "step": 582
+    },
+    {
+      "epoch": 0.8567229977957385,
+      "grad_norm": 1.4101135770661162,
+      "learning_rate": 5.251214079266475e-08,
+      "loss": 1.069,
+      "step": 583
+    },
+    {
+      "epoch": 0.858192505510654,
+      "grad_norm": 1.3634335354582763,
+      "learning_rate": 5.1453867658958704e-08,
+      "loss": 1.1287,
+      "step": 584
+    },
+    {
+      "epoch": 0.8596620132255695,
+      "grad_norm": 1.4283589117852606,
+      "learning_rate": 5.0405788304160426e-08,
+      "loss": 1.0935,
+      "step": 585
+    },
+    {
+      "epoch": 0.861131520940485,
+      "grad_norm": 1.434406958645411,
+      "learning_rate": 4.936792654720029e-08,
+      "loss": 1.0494,
+      "step": 586
+    },
+    {
+      "epoch": 0.8626010286554004,
+      "grad_norm": 1.4389756558713183,
+      "learning_rate": 4.8340305974801266e-08,
+      "loss": 1.0656,
+      "step": 587
+    },
+    {
+      "epoch": 0.864070536370316,
+      "grad_norm": 1.3842315564330259,
+      "learning_rate": 4.7322949940942325e-08,
+      "loss": 1.0512,
+      "step": 588
+    },
+    {
+      "epoch": 0.8655400440852314,
+      "grad_norm": 1.4148624042139615,
+      "learning_rate": 4.63158815663276e-08,
+      "loss": 1.0613,
+      "step": 589
+    },
+    {
+      "epoch": 0.8670095518001469,
+      "grad_norm": 1.3366794867264846,
+      "learning_rate": 4.53191237378614e-08,
+      "loss": 1.0486,
+      "step": 590
+    },
+    {
+      "epoch": 0.8684790595150624,
+      "grad_norm": 1.3575430817405785,
+      "learning_rate": 4.433269910812759e-08,
+      "loss": 1.0735,
+      "step": 591
+    },
+    {
+      "epoch": 0.8699485672299779,
+      "grad_norm": 1.3289314595584882,
+      "learning_rate": 4.335663009487511e-08,
+      "loss": 1.1759,
+      "step": 592
+    },
+    {
+      "epoch": 0.8714180749448934,
+      "grad_norm": 1.4000173838833194,
+      "learning_rate": 4.2390938880508595e-08,
+      "loss": 1.0408,
+      "step": 593
+    },
+    {
+      "epoch": 0.8728875826598089,
+      "grad_norm": 1.372416148633152,
+      "learning_rate": 4.143564741158362e-08,
+      "loss": 1.0626,
+      "step": 594
+    },
+    {
+      "epoch": 0.8743570903747244,
+      "grad_norm": 1.4265940660232175,
+      "learning_rate": 4.0490777398308753e-08,
+      "loss": 1.2163,
+      "step": 595
+    },
+    {
+      "epoch": 0.8758265980896399,
+      "grad_norm": 1.3925507856636796,
+      "learning_rate": 3.955635031405169e-08,
+      "loss": 1.1028,
+      "step": 596
+    },
+    {
+      "epoch": 0.8772961058045555,
+      "grad_norm": 1.324576005377705,
+      "learning_rate": 3.86323873948512e-08,
+      "loss": 1.182,
+      "step": 597
+    },
+    {
+      "epoch": 0.878765613519471,
+      "grad_norm": 1.351029057416015,
+      "learning_rate": 3.771890963893476e-08,
+      "loss": 1.1285,
+      "step": 598
+    },
+    {
+      "epoch": 0.8802351212343865,
+      "grad_norm": 1.3994327447522252,
+      "learning_rate": 3.681593780624137e-08,
+      "loss": 1.0868,
+      "step": 599
+    },
+    {
+      "epoch": 0.881704628949302,
+      "grad_norm": 1.3561535752329212,
+      "learning_rate": 3.5923492417949285e-08,
+      "loss": 1.1401,
+      "step": 600
+    },
+    {
+      "epoch": 0.8831741366642175,
+      "grad_norm": 1.3988534133786137,
+      "learning_rate": 3.5041593756010234e-08,
+      "loss": 1.0106,
+      "step": 601
+    },
+    {
+      "epoch": 0.884643644379133,
+      "grad_norm": 1.3859821796457574,
+      "learning_rate": 3.417026186268829e-08,
+      "loss": 1.0873,
+      "step": 602
+    },
+    {
+      "epoch": 0.8861131520940485,
+      "grad_norm": 1.3465915197780411,
+      "learning_rate": 3.3309516540104e-08,
+      "loss": 1.1312,
+      "step": 603
+    },
+    {
+      "epoch": 0.887582659808964,
+      "grad_norm": 1.3934103882388418,
+      "learning_rate": 3.2459377349784986e-08,
+      "loss": 1.0802,
+      "step": 604
+    },
+    {
+      "epoch": 0.8890521675238795,
+      "grad_norm": 1.437718999786439,
+      "learning_rate": 3.1619863612221075e-08,
+      "loss": 1.182,
+      "step": 605
+    },
+    {
+      "epoch": 0.890521675238795,
+      "grad_norm": 1.4477961897782676,
+      "learning_rate": 3.079099440642496e-08,
+      "loss": 1.1751,
+      "step": 606
+    },
+    {
+      "epoch": 0.8919911829537105,
+      "grad_norm": 1.3924612008055068,
+      "learning_rate": 2.997278856949914e-08,
+      "loss": 1.0903,
+      "step": 607
+    },
+    {
+      "epoch": 0.893460690668626,
+      "grad_norm": 1.3907249036661788,
+      "learning_rate": 2.916526469620756e-08,
+      "loss": 1.0676,
+      "step": 608
+    },
+    {
+      "epoch": 0.8949301983835415,
+      "grad_norm": 1.4577046884258984,
+      "learning_rate": 2.836844113855269e-08,
+      "loss": 1.2377,
+      "step": 609
+    },
+    {
+      "epoch": 0.896399706098457,
+      "grad_norm": 1.4063512104136562,
+      "learning_rate": 2.758233600535914e-08,
+      "loss": 1.0655,
+      "step": 610
+    },
+    {
+      "epoch": 0.8978692138133725,
+      "grad_norm": 1.3811245692933716,
+      "learning_rate": 2.6806967161861593e-08,
+      "loss": 1.0515,
+      "step": 611
+    },
+    {
+      "epoch": 0.899338721528288,
+      "grad_norm": 1.4687882128856604,
+      "learning_rate": 2.6042352229298902e-08,
+      "loss": 1.0785,
+      "step": 612
+    },
+    {
+      "epoch": 0.9008082292432035,
+      "grad_norm": 1.4655898408700718,
+      "learning_rate": 2.5288508584513814e-08,
+      "loss": 1.0829,
+      "step": 613
+    },
+    {
+      "epoch": 0.9022777369581191,
+      "grad_norm": 1.3913574357745526,
+      "learning_rate": 2.4545453359557765e-08,
+      "loss": 1.0768,
+      "step": 614
+    },
+    {
+      "epoch": 0.9037472446730346,
+      "grad_norm": 1.3607919197214273,
+      "learning_rate": 2.3813203441301778e-08,
+      "loss": 1.1146,
+      "step": 615
+    },
+    {
+      "epoch": 0.9052167523879501,
+      "grad_norm": 1.4196701845145032,
+      "learning_rate": 2.3091775471052734e-08,
+      "loss": 1.1263,
+      "step": 616
+    },
+    {
+      "epoch": 0.9066862601028656,
+      "grad_norm": 1.3646134363098568,
+      "learning_rate": 2.2381185844174644e-08,
+      "loss": 1.027,
+      "step": 617
+    },
+    {
+      "epoch": 0.9081557678177811,
+      "grad_norm": 1.4522612796957168,
+      "learning_rate": 2.168145070971683e-08,
+      "loss": 1.0925,
+      "step": 618
+    },
+    {
+      "epoch": 0.9096252755326966,
+      "grad_norm": 1.332216974123385,
+      "learning_rate": 2.099258597004644e-08,
+      "loss": 1.057,
+      "step": 619
+    },
+    {
+      "epoch": 0.9110947832476121,
+      "grad_norm": 1.4818724681035569,
+      "learning_rate": 2.031460728048695e-08,
+      "loss": 0.9901,
+      "step": 620
+    },
+    {
+      "epoch": 0.9125642909625276,
+      "grad_norm": 1.3874842041637756,
+      "learning_rate": 1.9647530048962747e-08,
+      "loss": 1.0946,
+      "step": 621
+    },
+    {
+      "epoch": 0.914033798677443,
+      "grad_norm": 1.3490442752252867,
+      "learning_rate": 1.8991369435648774e-08,
+      "loss": 1.1295,
+      "step": 622
+    },
+    {
+      "epoch": 0.9155033063923586,
+      "grad_norm": 1.3339563383295754,
+      "learning_rate": 1.8346140352625883e-08,
+      "loss": 1.0182,
+      "step": 623
+    },
+    {
+      "epoch": 0.916972814107274,
+      "grad_norm": 1.4464185646745273,
+      "learning_rate": 1.771185746354209e-08,
+      "loss": 1.0926,
+      "step": 624
+    },
+    {
+      "epoch": 0.9184423218221895,
+      "grad_norm": 1.3165206597074977,
+      "learning_rate": 1.7088535183279407e-08,
+      "loss": 1.1249,
+      "step": 625
+    },
+    {
+      "epoch": 0.919911829537105,
+      "grad_norm": 1.3819515848899369,
+      "learning_rate": 1.647618767762593e-08,
+      "loss": 1.1573,
+      "step": 626
+    },
+    {
+      "epoch": 0.9213813372520205,
+      "grad_norm": 1.3001816716305787,
+      "learning_rate": 1.5874828862954327e-08,
+      "loss": 1.0266,
+      "step": 627
+    },
+    {
+      "epoch": 0.922850844966936,
+      "grad_norm": 1.319276221215603,
+      "learning_rate": 1.5284472405905247e-08,
+      "loss": 1.0694,
+      "step": 628
+    },
+    {
+      "epoch": 0.9243203526818515,
+      "grad_norm": 1.4228373976898447,
+      "learning_rate": 1.4705131723076692e-08,
+      "loss": 1.1516,
+      "step": 629
+    },
+    {
+      "epoch": 0.925789860396767,
+      "grad_norm": 1.3595993483558204,
+      "learning_rate": 1.4136819980719472e-08,
+      "loss": 1.0336,
+      "step": 630
+    },
+    {
+      "epoch": 0.9272593681116826,
+      "grad_norm": 1.3326370942028576,
+      "learning_rate": 1.3579550094437676e-08,
+      "loss": 1.0843,
+      "step": 631
+    },
+    {
+      "epoch": 0.9287288758265981,
+      "grad_norm": 1.3933226824191673,
+      "learning_rate": 1.3033334728895119e-08,
+      "loss": 1.1692,
+      "step": 632
+    },
+    {
+      "epoch": 0.9301983835415136,
+      "grad_norm": 1.3772541537537315,
+      "learning_rate": 1.2498186297527802e-08,
+      "loss": 1.1253,
+      "step": 633
+    },
+    {
+      "epoch": 0.9316678912564291,
+      "grad_norm": 1.4025797539924674,
+      "learning_rate": 1.1974116962261527e-08,
+      "loss": 1.1508,
+      "step": 634
+    },
+    {
+      "epoch": 0.9331373989713446,
+      "grad_norm": 1.3051053877123164,
+      "learning_rate": 1.1461138633235611e-08,
+      "loss": 1.0791,
+      "step": 635
+    },
+    {
+      "epoch": 0.9346069066862601,
+      "grad_norm": 1.3333191944243534,
+      "learning_rate": 1.095926296853228e-08,
+      "loss": 1.1968,
+      "step": 636
+    },
+    {
+      "epoch": 0.9360764144011756,
+      "grad_norm": 1.3418559426331282,
+      "learning_rate": 1.0468501373911532e-08,
+      "loss": 1.0896,
+      "step": 637
+    },
+    {
+      "epoch": 0.9375459221160911,
+      "grad_norm": 1.34359709316688,
+      "learning_rate": 9.988865002552138e-09,
+      "loss": 1.1065,
+      "step": 638
+    },
+    {
+      "epoch": 0.9390154298310066,
+      "grad_norm": 1.3620227287509608,
+      "learning_rate": 9.520364754798116e-09,
+      "loss": 1.0744,
+      "step": 639
+    },
+    {
+      "epoch": 0.9404849375459221,
+      "grad_norm": 1.3087398924414495,
+      "learning_rate": 9.06301127791087e-09,
+      "loss": 1.072,
+      "step": 640
+    },
+    {
+      "epoch": 0.9419544452608376,
+      "grad_norm": 1.4212527547101943,
+      "learning_rate": 8.61681496582739e-09,
+      "loss": 1.0904,
+      "step": 641
+    },
+    {
+      "epoch": 0.9434239529757531,
+      "grad_norm": 1.4044007797229823,
+      "learning_rate": 8.181785958923938e-09,
+      "loss": 1.1127,
+      "step": 642
+    },
+    {
+      "epoch": 0.9448934606906686,
+      "grad_norm": 1.3492187382144503,
+      "learning_rate": 7.757934143785561e-09,
+      "loss": 1.1309,
+      "step": 643
+    },
+    {
+      "epoch": 0.9463629684055841,
+      "grad_norm": 1.3882974174543632,
+      "learning_rate": 7.345269152981614e-09,
+      "loss": 1.0746,
+      "step": 644
+    },
+    {
+      "epoch": 0.9478324761204996,
+      "grad_norm": 1.3919108675813483,
+      "learning_rate": 6.943800364846653e-09,
+      "loss": 1.1149,
+      "step": 645
+    },
+    {
+      "epoch": 0.9493019838354151,
+      "grad_norm": 1.4427576763993042,
+      "learning_rate": 6.5535369032672095e-09,
+      "loss": 1.1289,
+      "step": 646
+    },
+    {
+      "epoch": 0.9507714915503307,
+      "grad_norm": 1.4253308850896702,
+      "learning_rate": 6.174487637474801e-09,
+      "loss": 1.1042,
+      "step": 647
+    },
+    {
+      "epoch": 0.9522409992652462,
+      "grad_norm": 1.400034959006393,
+      "learning_rate": 5.806661181843919e-09,
+      "loss": 1.0803,
+      "step": 648
+    },
+    {
+      "epoch": 0.9537105069801617,
+      "grad_norm": 1.435130822553809,
+      "learning_rate": 5.450065895696632e-09,
+      "loss": 1.0796,
+      "step": 649
+    },
+    {
+      "epoch": 0.9551800146950772,
+      "grad_norm": 1.4035193875819156,
+      "learning_rate": 5.1047098831125124e-09,
+      "loss": 1.1129,
+      "step": 650
+    },
+    {
+      "epoch": 0.9566495224099927,
+      "grad_norm": 1.5238894343006477,
+      "learning_rate": 4.770600992744178e-09,
+      "loss": 1.1102,
+      "step": 651
+    },
+    {
+      "epoch": 0.9581190301249082,
+      "grad_norm": 1.3438191535373563,
+      "learning_rate": 4.4477468176393196e-09,
+      "loss": 1.0581,
+      "step": 652
+    },
+    {
+      "epoch": 0.9595885378398237,
+      "grad_norm": 1.3010475336208236,
+      "learning_rate": 4.136154695068006e-09,
+      "loss": 1.0478,
+      "step": 653
+    },
+    {
+      "epoch": 0.9610580455547392,
+      "grad_norm": 1.3915831877261908,
+      "learning_rate": 3.8358317063557635e-09,
+      "loss": 1.0616,
+      "step": 654
+    },
+    {
+      "epoch": 0.9625275532696547,
+      "grad_norm": 1.3754622477764384,
+      "learning_rate": 3.546784676722925e-09,
+      "loss": 1.0684,
+      "step": 655
+    },
+    {
+      "epoch": 0.9639970609845702,
+      "grad_norm": 1.3958030478092502,
+      "learning_rate": 3.2690201751292002e-09,
+      "loss": 1.1125,
+      "step": 656
+    },
+    {
+      "epoch": 0.9654665686994857,
+      "grad_norm": 1.4118012322738474,
+      "learning_rate": 3.002544514124683e-09,
+      "loss": 1.0673,
+      "step": 657
+    },
+    {
+      "epoch": 0.9669360764144012,
+      "grad_norm": 1.3931161551594535,
+      "learning_rate": 2.747363749706244e-09,
+      "loss": 1.1434,
+      "step": 658
+    },
+    {
+      "epoch": 0.9684055841293167,
+      "grad_norm": 1.320233907465206,
+      "learning_rate": 2.5034836811799744e-09,
+      "loss": 1.0626,
+      "step": 659
+    },
+    {
+      "epoch": 0.9698750918442322,
+      "grad_norm": 1.40459539845844,
+      "learning_rate": 2.2709098510292347e-09,
+      "loss": 1.1047,
+      "step": 660
+    },
+    {
+      "epoch": 0.9713445995591476,
+      "grad_norm": 1.4426402012148907,
+      "learning_rate": 2.049647544788813e-09,
+      "loss": 1.0249,
+      "step": 661
+    },
+    {
+      "epoch": 0.9728141072740631,
+      "grad_norm": 1.37434918164306,
+      "learning_rate": 1.8397017909249634e-09,
+      "loss": 1.0911,
+      "step": 662
+    },
+    {
+      "epoch": 0.9742836149889786,
+      "grad_norm": 1.4108384424402636,
+      "learning_rate": 1.6410773607206663e-09,
+      "loss": 1.1213,
+      "step": 663
+    },
+    {
+      "epoch": 0.9757531227038942,
+      "grad_norm": 1.4926584455705627,
+      "learning_rate": 1.4537787681677683e-09,
+      "loss": 1.1082,
+      "step": 664
+    },
+    {
+      "epoch": 0.9772226304188097,
+      "grad_norm": 1.3595606693253082,
+      "learning_rate": 1.2778102698638993e-09,
+      "loss": 1.0751,
+      "step": 665
+    },
+    {
+      "epoch": 0.9786921381337252,
+      "grad_norm": 1.4509988964421765,
+      "learning_rate": 1.1131758649160494e-09,
+      "loss": 1.0865,
+      "step": 666
+    },
+    {
+      "epoch": 0.9801616458486407,
+      "grad_norm": 1.3679120087919863,
+      "learning_rate": 9.598792948496414e-10,
+      "loss": 1.0855,
+      "step": 667
+    },
+    {
+      "epoch": 0.9816311535635562,
+      "grad_norm": 1.346685189196511,
+      "learning_rate": 8.179240435232659e-10,
+      "loss": 1.0846,
+      "step": 668
+    },
+    {
+      "epoch": 0.9831006612784717,
+      "grad_norm": 1.3196217753280912,
+      "learning_rate": 6.873133370498551e-10,
+      "loss": 1.0291,
+      "step": 669
+    },
+    {
+      "epoch": 0.9845701689933872,
+      "grad_norm": 1.3795048616719339,
+      "learning_rate": 5.680501437230755e-10,
+      "loss": 1.0951,
+      "step": 670
+    },
+    {
+      "epoch": 0.9860396767083027,
+      "grad_norm": 1.3485249223159093,
+      "learning_rate": 4.6013717395010365e-10,
+      "loss": 1.0971,
+      "step": 671
+    },
+    {
+      "epoch": 0.9875091844232182,
+      "grad_norm": 1.507892762769766,
+      "learning_rate": 3.63576880189731e-10,
+      "loss": 1.1302,
+      "step": 672
+    },
+    {
+      "epoch": 0.9889786921381337,
+      "grad_norm": 1.362116408246938,
+      "learning_rate": 2.783714568970197e-10,
+      "loss": 1.1335,
+      "step": 673
+    },
+    {
+      "epoch": 0.9904481998530492,
+      "grad_norm": 1.4376872456307725,
+      "learning_rate": 2.045228404731203e-10,
+      "loss": 1.0601,
+      "step": 674
+    },
+    {
+      "epoch": 0.9919177075679647,
+      "grad_norm": 1.3558644824462847,
+      "learning_rate": 1.4203270922125143e-10,
+      "loss": 1.0517,
+      "step": 675
+    },
+    {
+      "epoch": 0.9933872152828802,
+      "grad_norm": 1.4975627645392091,
+      "learning_rate": 9.090248330889671e-11,
+      "loss": 1.0872,
+      "step": 676
+    },
+    {
+      "epoch": 0.9948567229977957,
+      "grad_norm": 1.5061190709997394,
+      "learning_rate": 5.1133324735164183e-11,
+      "loss": 1.0224,
+      "step": 677
+    },
+    {
+      "epoch": 0.9963262307127112,
+      "grad_norm": 1.4530193394151236,
+      "learning_rate": 2.2726137304529546e-11,
+      "loss": 1.0538,
+      "step": 678
+    },
+    {
+      "epoch": 0.9977957384276267,
+      "grad_norm": 1.3983394789120556,
+      "learning_rate": 5.6815666063525505e-12,
+      "loss": 1.0876,
+      "step": 679
+    },
+    {
+      "epoch": 0.9992652461425422,
+      "grad_norm": 1.4207479289834348,
+      "learning_rate": 0.0,
+      "loss": 1.1348,
+      "step": 680
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 680,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 711821505200128.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}