diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,14811 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9114583333333334,
+  "eval_steps": 300,
+  "global_step": 2100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00043402777777777775,
+      "grad_norm": 22.039037704467773,
+      "learning_rate": 7.194244604316547e-08,
+      "loss": 0.7083,
+      "step": 1
+    },
+    {
+      "epoch": 0.0008680555555555555,
+      "grad_norm": 20.530189514160156,
+      "learning_rate": 1.4388489208633095e-07,
+      "loss": 0.7177,
+      "step": 2
+    },
+    {
+      "epoch": 0.0013020833333333333,
+      "grad_norm": 23.356374740600586,
+      "learning_rate": 2.1582733812949643e-07,
+      "loss": 0.7513,
+      "step": 3
+    },
+    {
+      "epoch": 0.001736111111111111,
+      "grad_norm": 20.72550392150879,
+      "learning_rate": 2.877697841726619e-07,
+      "loss": 0.7145,
+      "step": 4
+    },
+    {
+      "epoch": 0.002170138888888889,
+      "grad_norm": 22.307851791381836,
+      "learning_rate": 3.5971223021582736e-07,
+      "loss": 0.6842,
+      "step": 5
+    },
+    {
+      "epoch": 0.0026041666666666665,
+      "grad_norm": 26.96861457824707,
+      "learning_rate": 4.3165467625899287e-07,
+      "loss": 0.7214,
+      "step": 6
+    },
+    {
+      "epoch": 0.0030381944444444445,
+      "grad_norm": 22.72673988342285,
+      "learning_rate": 5.035971223021583e-07,
+      "loss": 0.6912,
+      "step": 7
+    },
+    {
+      "epoch": 0.003472222222222222,
+      "grad_norm": 20.26327896118164,
+      "learning_rate": 5.755395683453238e-07,
+      "loss": 0.6848,
+      "step": 8
+    },
+    {
+      "epoch": 0.00390625,
+      "grad_norm": 19.785690307617188,
+      "learning_rate": 6.474820143884893e-07,
+      "loss": 0.6813,
+      "step": 9
+    },
+    {
+      "epoch": 0.004340277777777778,
+      "grad_norm": 20.36345863342285,
+      "learning_rate": 7.194244604316547e-07,
+      "loss": 0.6662,
+      "step": 10
+    },
+    {
+      "epoch": 0.004774305555555556,
+      "grad_norm": 22.651525497436523,
+      "learning_rate": 7.913669064748202e-07,
+      "loss": 0.6969,
+      "step": 11
+    },
+    {
+      "epoch": 0.005208333333333333,
+      "grad_norm": 24.08732795715332,
+      "learning_rate": 8.633093525179857e-07,
+      "loss": 0.6428,
+      "step": 12
+    },
+    {
+      "epoch": 0.005642361111111111,
+      "grad_norm": 24.18958854675293,
+      "learning_rate": 9.352517985611512e-07,
+      "loss": 0.5645,
+      "step": 13
+    },
+    {
+      "epoch": 0.006076388888888889,
+      "grad_norm": 23.265363693237305,
+      "learning_rate": 1.0071942446043167e-06,
+      "loss": 0.5913,
+      "step": 14
+    },
+    {
+      "epoch": 0.006510416666666667,
+      "grad_norm": 21.222742080688477,
+      "learning_rate": 1.079136690647482e-06,
+      "loss": 0.5673,
+      "step": 15
+    },
+    {
+      "epoch": 0.006944444444444444,
+      "grad_norm": 19.037031173706055,
+      "learning_rate": 1.1510791366906476e-06,
+      "loss": 0.529,
+      "step": 16
+    },
+    {
+      "epoch": 0.007378472222222222,
+      "grad_norm": 25.137651443481445,
+      "learning_rate": 1.2230215827338131e-06,
+      "loss": 0.3661,
+      "step": 17
+    },
+    {
+      "epoch": 0.0078125,
+      "grad_norm": 27.268735885620117,
+      "learning_rate": 1.2949640287769785e-06,
+      "loss": 0.3817,
+      "step": 18
+    },
+    {
+      "epoch": 0.008246527777777778,
+      "grad_norm": 26.401870727539062,
+      "learning_rate": 1.366906474820144e-06,
+      "loss": 0.36,
+      "step": 19
+    },
+    {
+      "epoch": 0.008680555555555556,
+      "grad_norm": 19.024900436401367,
+      "learning_rate": 1.4388489208633094e-06,
+      "loss": 0.3818,
+      "step": 20
+    },
+    {
+      "epoch": 0.009114583333333334,
+      "grad_norm": 15.296829223632812,
+      "learning_rate": 1.510791366906475e-06,
+      "loss": 0.3154,
+      "step": 21
+    },
+    {
+      "epoch": 0.009548611111111112,
+      "grad_norm": 21.349294662475586,
+      "learning_rate": 1.5827338129496403e-06,
+      "loss": 0.2649,
+      "step": 22
+    },
+    {
+      "epoch": 0.009982638888888888,
+      "grad_norm": 42.53651428222656,
+      "learning_rate": 1.654676258992806e-06,
+      "loss": 0.2867,
+      "step": 23
+    },
+    {
+      "epoch": 0.010416666666666666,
+      "grad_norm": 13.64511775970459,
+      "learning_rate": 1.7266187050359715e-06,
+      "loss": 0.2534,
+      "step": 24
+    },
+    {
+      "epoch": 0.010850694444444444,
+      "grad_norm": 20.433380126953125,
+      "learning_rate": 1.7985611510791368e-06,
+      "loss": 0.2721,
+      "step": 25
+    },
+    {
+      "epoch": 0.011284722222222222,
+      "grad_norm": 12.288137435913086,
+      "learning_rate": 1.8705035971223024e-06,
+      "loss": 0.2361,
+      "step": 26
+    },
+    {
+      "epoch": 0.01171875,
+      "grad_norm": 14.382365226745605,
+      "learning_rate": 1.942446043165468e-06,
+      "loss": 0.2462,
+      "step": 27
+    },
+    {
+      "epoch": 0.012152777777777778,
+      "grad_norm": 9.596738815307617,
+      "learning_rate": 2.0143884892086333e-06,
+      "loss": 0.1907,
+      "step": 28
+    },
+    {
+      "epoch": 0.012586805555555556,
+      "grad_norm": 7.768435478210449,
+      "learning_rate": 2.0863309352517987e-06,
+      "loss": 0.1846,
+      "step": 29
+    },
+    {
+      "epoch": 0.013020833333333334,
+      "grad_norm": 10.87830924987793,
+      "learning_rate": 2.158273381294964e-06,
+      "loss": 0.2154,
+      "step": 30
+    },
+    {
+      "epoch": 0.013454861111111112,
+      "grad_norm": 11.848028182983398,
+      "learning_rate": 2.23021582733813e-06,
+      "loss": 0.2569,
+      "step": 31
+    },
+    {
+      "epoch": 0.013888888888888888,
+      "grad_norm": 9.48557186126709,
+      "learning_rate": 2.302158273381295e-06,
+      "loss": 0.2271,
+      "step": 32
+    },
+    {
+      "epoch": 0.014322916666666666,
+      "grad_norm": 13.119791984558105,
+      "learning_rate": 2.3741007194244605e-06,
+      "loss": 0.1902,
+      "step": 33
+    },
+    {
+      "epoch": 0.014756944444444444,
+      "grad_norm": 8.415437698364258,
+      "learning_rate": 2.4460431654676263e-06,
+      "loss": 0.2007,
+      "step": 34
+    },
+    {
+      "epoch": 0.015190972222222222,
+      "grad_norm": 10.289463996887207,
+      "learning_rate": 2.5179856115107916e-06,
+      "loss": 0.1879,
+      "step": 35
+    },
+    {
+      "epoch": 0.015625,
+      "grad_norm": 14.64406681060791,
+      "learning_rate": 2.589928057553957e-06,
+      "loss": 0.173,
+      "step": 36
+    },
+    {
+      "epoch": 0.016059027777777776,
+      "grad_norm": 12.37926959991455,
+      "learning_rate": 2.6618705035971228e-06,
+      "loss": 0.2205,
+      "step": 37
+    },
+    {
+      "epoch": 0.016493055555555556,
+      "grad_norm": 13.952500343322754,
+      "learning_rate": 2.733812949640288e-06,
+      "loss": 0.2058,
+      "step": 38
+    },
+    {
+      "epoch": 0.016927083333333332,
+      "grad_norm": 10.09529972076416,
+      "learning_rate": 2.805755395683453e-06,
+      "loss": 0.1783,
+      "step": 39
+    },
+    {
+      "epoch": 0.017361111111111112,
+      "grad_norm": 11.618066787719727,
+      "learning_rate": 2.877697841726619e-06,
+      "loss": 0.1718,
+      "step": 40
+    },
+    {
+      "epoch": 0.017795138888888888,
+      "grad_norm": 5.552605628967285,
+      "learning_rate": 2.949640287769784e-06,
+      "loss": 0.1269,
+      "step": 41
+    },
+    {
+      "epoch": 0.018229166666666668,
+      "grad_norm": 6.753513336181641,
+      "learning_rate": 3.02158273381295e-06,
+      "loss": 0.1408,
+      "step": 42
+    },
+    {
+      "epoch": 0.018663194444444444,
+      "grad_norm": 13.211586952209473,
+      "learning_rate": 3.0935251798561158e-06,
+      "loss": 0.1528,
+      "step": 43
+    },
+    {
+      "epoch": 0.019097222222222224,
+      "grad_norm": 7.235265731811523,
+      "learning_rate": 3.1654676258992807e-06,
+      "loss": 0.1649,
+      "step": 44
+    },
+    {
+      "epoch": 0.01953125,
+      "grad_norm": 7.747194766998291,
+      "learning_rate": 3.237410071942446e-06,
+      "loss": 0.1485,
+      "step": 45
+    },
+    {
+      "epoch": 0.019965277777777776,
+      "grad_norm": 7.433028697967529,
+      "learning_rate": 3.309352517985612e-06,
+      "loss": 0.1665,
+      "step": 46
+    },
+    {
+      "epoch": 0.020399305555555556,
+      "grad_norm": 10.274096488952637,
+      "learning_rate": 3.381294964028777e-06,
+      "loss": 0.1545,
+      "step": 47
+    },
+    {
+      "epoch": 0.020833333333333332,
+      "grad_norm": 13.211413383483887,
+      "learning_rate": 3.453237410071943e-06,
+      "loss": 0.1682,
+      "step": 48
+    },
+    {
+      "epoch": 0.021267361111111112,
+      "grad_norm": 6.6468706130981445,
+      "learning_rate": 3.525179856115108e-06,
+      "loss": 0.1291,
+      "step": 49
+    },
+    {
+      "epoch": 0.021701388888888888,
+      "grad_norm": 8.226445198059082,
+      "learning_rate": 3.5971223021582737e-06,
+      "loss": 0.2014,
+      "step": 50
+    },
+    {
+      "epoch": 0.022135416666666668,
+      "grad_norm": 6.410370826721191,
+      "learning_rate": 3.669064748201439e-06,
+      "loss": 0.1378,
+      "step": 51
+    },
+    {
+      "epoch": 0.022569444444444444,
+      "grad_norm": 4.667335033416748,
+      "learning_rate": 3.741007194244605e-06,
+      "loss": 0.1412,
+      "step": 52
+    },
+    {
+      "epoch": 0.023003472222222224,
+      "grad_norm": 4.757472515106201,
+      "learning_rate": 3.81294964028777e-06,
+      "loss": 0.1376,
+      "step": 53
+    },
+    {
+      "epoch": 0.0234375,
+      "grad_norm": 3.980797529220581,
+      "learning_rate": 3.884892086330936e-06,
+      "loss": 0.1506,
+      "step": 54
+    },
+    {
+      "epoch": 0.023871527777777776,
+      "grad_norm": 5.135439872741699,
+      "learning_rate": 3.956834532374101e-06,
+      "loss": 0.1208,
+      "step": 55
+    },
+    {
+      "epoch": 0.024305555555555556,
+      "grad_norm": 4.567761421203613,
+      "learning_rate": 4.028776978417267e-06,
+      "loss": 0.1458,
+      "step": 56
+    },
+    {
+      "epoch": 0.024739583333333332,
+      "grad_norm": 5.0912699699401855,
+      "learning_rate": 4.100719424460432e-06,
+      "loss": 0.1469,
+      "step": 57
+    },
+    {
+      "epoch": 0.025173611111111112,
+      "grad_norm": 3.5036520957946777,
+      "learning_rate": 4.172661870503597e-06,
+      "loss": 0.1111,
+      "step": 58
+    },
+    {
+      "epoch": 0.025607638888888888,
+      "grad_norm": 3.6658434867858887,
+      "learning_rate": 4.244604316546763e-06,
+      "loss": 0.1222,
+      "step": 59
+    },
+    {
+      "epoch": 0.026041666666666668,
+      "grad_norm": 3.44124436378479,
+      "learning_rate": 4.316546762589928e-06,
+      "loss": 0.1143,
+      "step": 60
+    },
+    {
+      "epoch": 0.026475694444444444,
+      "grad_norm": 4.265533447265625,
+      "learning_rate": 4.388489208633094e-06,
+      "loss": 0.1016,
+      "step": 61
+    },
+    {
+      "epoch": 0.026909722222222224,
+      "grad_norm": 3.329289674758911,
+      "learning_rate": 4.46043165467626e-06,
+      "loss": 0.1105,
+      "step": 62
+    },
+    {
+      "epoch": 0.02734375,
+      "grad_norm": 6.9250383377075195,
+      "learning_rate": 4.5323741007194245e-06,
+      "loss": 0.2056,
+      "step": 63
+    },
+    {
+      "epoch": 0.027777777777777776,
+      "grad_norm": 29.69718360900879,
+      "learning_rate": 4.60431654676259e-06,
+      "loss": 0.1643,
+      "step": 64
+    },
+    {
+      "epoch": 0.028211805555555556,
+      "grad_norm": 3.4973223209381104,
+      "learning_rate": 4.676258992805755e-06,
+      "loss": 0.1296,
+      "step": 65
+    },
+    {
+      "epoch": 0.028645833333333332,
+      "grad_norm": 3.7894699573516846,
+      "learning_rate": 4.748201438848921e-06,
+      "loss": 0.1521,
+      "step": 66
+    },
+    {
+      "epoch": 0.029079861111111112,
+      "grad_norm": 4.099551200866699,
+      "learning_rate": 4.820143884892087e-06,
+      "loss": 0.11,
+      "step": 67
+    },
+    {
+      "epoch": 0.029513888888888888,
+      "grad_norm": 4.133640766143799,
+      "learning_rate": 4.892086330935253e-06,
+      "loss": 0.1544,
+      "step": 68
+    },
+    {
+      "epoch": 0.029947916666666668,
+      "grad_norm": 3.045886754989624,
+      "learning_rate": 4.9640287769784175e-06,
+      "loss": 0.1247,
+      "step": 69
+    },
+    {
+      "epoch": 0.030381944444444444,
+      "grad_norm": 4.6925950050354,
+      "learning_rate": 5.035971223021583e-06,
+      "loss": 0.1302,
+      "step": 70
+    },
+    {
+      "epoch": 0.030815972222222224,
+      "grad_norm": 4.342012882232666,
+      "learning_rate": 5.107913669064749e-06,
+      "loss": 0.1378,
+      "step": 71
+    },
+    {
+      "epoch": 0.03125,
+      "grad_norm": 3.7377989292144775,
+      "learning_rate": 5.179856115107914e-06,
+      "loss": 0.1376,
+      "step": 72
+    },
+    {
+      "epoch": 0.031684027777777776,
+      "grad_norm": 2.9397761821746826,
+      "learning_rate": 5.251798561151079e-06,
+      "loss": 0.1279,
+      "step": 73
+    },
+    {
+      "epoch": 0.03211805555555555,
+      "grad_norm": 2.6648385524749756,
+      "learning_rate": 5.3237410071942456e-06,
+      "loss": 0.125,
+      "step": 74
+    },
+    {
+      "epoch": 0.032552083333333336,
+      "grad_norm": 4.317229747772217,
+      "learning_rate": 5.3956834532374105e-06,
+      "loss": 0.1038,
+      "step": 75
+    },
+    {
+      "epoch": 0.03298611111111111,
+      "grad_norm": 3.513620376586914,
+      "learning_rate": 5.467625899280576e-06,
+      "loss": 0.0993,
+      "step": 76
+    },
+    {
+      "epoch": 0.03342013888888889,
+      "grad_norm": 3.500981330871582,
+      "learning_rate": 5.539568345323741e-06,
+      "loss": 0.122,
+      "step": 77
+    },
+    {
+      "epoch": 0.033854166666666664,
+      "grad_norm": 2.3485305309295654,
+      "learning_rate": 5.611510791366906e-06,
+      "loss": 0.0902,
+      "step": 78
+    },
+    {
+      "epoch": 0.03428819444444445,
+      "grad_norm": 2.7842323780059814,
+      "learning_rate": 5.683453237410073e-06,
+      "loss": 0.1088,
+      "step": 79
+    },
+    {
+      "epoch": 0.034722222222222224,
+      "grad_norm": 3.6362524032592773,
+      "learning_rate": 5.755395683453238e-06,
+      "loss": 0.1331,
+      "step": 80
+    },
+    {
+      "epoch": 0.03515625,
+      "grad_norm": 2.647966146469116,
+      "learning_rate": 5.8273381294964035e-06,
+      "loss": 0.1081,
+      "step": 81
+    },
+    {
+      "epoch": 0.035590277777777776,
+      "grad_norm": 4.212615013122559,
+      "learning_rate": 5.899280575539568e-06,
+      "loss": 0.1277,
+      "step": 82
+    },
+    {
+      "epoch": 0.03602430555555555,
+      "grad_norm": 2.0377724170684814,
+      "learning_rate": 5.971223021582734e-06,
+      "loss": 0.1064,
+      "step": 83
+    },
+    {
+      "epoch": 0.036458333333333336,
+      "grad_norm": 2.544142484664917,
+      "learning_rate": 6.0431654676259e-06,
+      "loss": 0.102,
+      "step": 84
+    },
+    {
+      "epoch": 0.03689236111111111,
+      "grad_norm": 2.4017419815063477,
+      "learning_rate": 6.115107913669065e-06,
+      "loss": 0.0823,
+      "step": 85
+    },
+    {
+      "epoch": 0.03732638888888889,
+      "grad_norm": 3.2943034172058105,
+      "learning_rate": 6.1870503597122315e-06,
+      "loss": 0.1098,
+      "step": 86
+    },
+    {
+      "epoch": 0.037760416666666664,
+      "grad_norm": 3.2607340812683105,
+      "learning_rate": 6.2589928057553964e-06,
+      "loss": 0.1552,
+      "step": 87
+    },
+    {
+      "epoch": 0.03819444444444445,
+      "grad_norm": 2.7917239665985107,
+      "learning_rate": 6.330935251798561e-06,
+      "loss": 0.1047,
+      "step": 88
+    },
+    {
+      "epoch": 0.038628472222222224,
+      "grad_norm": 3.072331190109253,
+      "learning_rate": 6.402877697841727e-06,
+      "loss": 0.0831,
+      "step": 89
+    },
+    {
+      "epoch": 0.0390625,
+      "grad_norm": 1.7378712892532349,
+      "learning_rate": 6.474820143884892e-06,
+      "loss": 0.113,
+      "step": 90
+    },
+    {
+      "epoch": 0.039496527777777776,
+      "grad_norm": 2.934422492980957,
+      "learning_rate": 6.546762589928059e-06,
+      "loss": 0.1121,
+      "step": 91
+    },
+    {
+      "epoch": 0.03993055555555555,
+      "grad_norm": 2.3598244190216064,
+      "learning_rate": 6.618705035971224e-06,
+      "loss": 0.0975,
+      "step": 92
+    },
+    {
+      "epoch": 0.040364583333333336,
+      "grad_norm": 3.3509647846221924,
+      "learning_rate": 6.6906474820143886e-06,
+      "loss": 0.1271,
+      "step": 93
+    },
+    {
+      "epoch": 0.04079861111111111,
+      "grad_norm": 3.8913393020629883,
+      "learning_rate": 6.762589928057554e-06,
+      "loss": 0.1008,
+      "step": 94
+    },
+    {
+      "epoch": 0.04123263888888889,
+      "grad_norm": 2.031092882156372,
+      "learning_rate": 6.834532374100719e-06,
+      "loss": 0.0998,
+      "step": 95
+    },
+    {
+      "epoch": 0.041666666666666664,
+      "grad_norm": 1.9853342771530151,
+      "learning_rate": 6.906474820143886e-06,
+      "loss": 0.1243,
+      "step": 96
+    },
+    {
+      "epoch": 0.04210069444444445,
+      "grad_norm": 1.718274712562561,
+      "learning_rate": 6.978417266187051e-06,
+      "loss": 0.109,
+      "step": 97
+    },
+    {
+      "epoch": 0.042534722222222224,
+      "grad_norm": 1.5930973291397095,
+      "learning_rate": 7.050359712230216e-06,
+      "loss": 0.1049,
+      "step": 98
+    },
+    {
+      "epoch": 0.04296875,
+      "grad_norm": 2.425053119659424,
+      "learning_rate": 7.122302158273382e-06,
+      "loss": 0.0991,
+      "step": 99
+    },
+    {
+      "epoch": 0.043402777777777776,
+      "grad_norm": 2.481341600418091,
+      "learning_rate": 7.194244604316547e-06,
+      "loss": 0.0843,
+      "step": 100
+    },
+    {
+      "epoch": 0.04383680555555555,
+      "grad_norm": 2.6645309925079346,
+      "learning_rate": 7.266187050359713e-06,
+      "loss": 0.098,
+      "step": 101
+    },
+    {
+      "epoch": 0.044270833333333336,
+      "grad_norm": 4.1276960372924805,
+      "learning_rate": 7.338129496402878e-06,
+      "loss": 0.1159,
+      "step": 102
+    },
+    {
+      "epoch": 0.04470486111111111,
+      "grad_norm": 4.709148406982422,
+      "learning_rate": 7.410071942446043e-06,
+      "loss": 0.0773,
+      "step": 103
+    },
+    {
+      "epoch": 0.04513888888888889,
+      "grad_norm": 6.506033897399902,
+      "learning_rate": 7.48201438848921e-06,
+      "loss": 0.1495,
+      "step": 104
+    },
+    {
+      "epoch": 0.045572916666666664,
+      "grad_norm": 2.1495561599731445,
+      "learning_rate": 7.5539568345323745e-06,
+      "loss": 0.1304,
+      "step": 105
+    },
+    {
+      "epoch": 0.04600694444444445,
+      "grad_norm": 2.843578577041626,
+      "learning_rate": 7.62589928057554e-06,
+      "loss": 0.1134,
+      "step": 106
+    },
+    {
+      "epoch": 0.046440972222222224,
+      "grad_norm": 2.714841842651367,
+      "learning_rate": 7.697841726618706e-06,
+      "loss": 0.0984,
+      "step": 107
+    },
+    {
+      "epoch": 0.046875,
+      "grad_norm": 2.3955094814300537,
+      "learning_rate": 7.769784172661872e-06,
+      "loss": 0.0714,
+      "step": 108
+    },
+    {
+      "epoch": 0.047309027777777776,
+      "grad_norm": 2.931508779525757,
+      "learning_rate": 7.841726618705036e-06,
+      "loss": 0.0862,
+      "step": 109
+    },
+    {
+      "epoch": 0.04774305555555555,
+      "grad_norm": 2.2287561893463135,
+      "learning_rate": 7.913669064748202e-06,
+      "loss": 0.0688,
+      "step": 110
+    },
+    {
+      "epoch": 0.048177083333333336,
+      "grad_norm": 4.007569313049316,
+      "learning_rate": 7.985611510791367e-06,
+      "loss": 0.1335,
+      "step": 111
+    },
+    {
+      "epoch": 0.04861111111111111,
+      "grad_norm": 2.308885097503662,
+      "learning_rate": 8.057553956834533e-06,
+      "loss": 0.0981,
+      "step": 112
+    },
+    {
+      "epoch": 0.04904513888888889,
+      "grad_norm": 3.60953950881958,
+      "learning_rate": 8.129496402877699e-06,
+      "loss": 0.0906,
+      "step": 113
+    },
+    {
+      "epoch": 0.049479166666666664,
+      "grad_norm": 3.5016674995422363,
+      "learning_rate": 8.201438848920865e-06,
+      "loss": 0.0946,
+      "step": 114
+    },
+    {
+      "epoch": 0.04991319444444445,
+      "grad_norm": 2.3107264041900635,
+      "learning_rate": 8.273381294964029e-06,
+      "loss": 0.1004,
+      "step": 115
+    },
+    {
+      "epoch": 0.050347222222222224,
+      "grad_norm": 2.496023654937744,
+      "learning_rate": 8.345323741007195e-06,
+      "loss": 0.1074,
+      "step": 116
+    },
+    {
+      "epoch": 0.05078125,
+      "grad_norm": 2.5439746379852295,
+      "learning_rate": 8.41726618705036e-06,
+      "loss": 0.1125,
+      "step": 117
+    },
+    {
+      "epoch": 0.051215277777777776,
+      "grad_norm": 2.115340232849121,
+      "learning_rate": 8.489208633093526e-06,
+      "loss": 0.0943,
+      "step": 118
+    },
+    {
+      "epoch": 0.05164930555555555,
+      "grad_norm": 1.8092048168182373,
+      "learning_rate": 8.561151079136692e-06,
+      "loss": 0.1043,
+      "step": 119
+    },
+    {
+      "epoch": 0.052083333333333336,
+      "grad_norm": 2.728424072265625,
+      "learning_rate": 8.633093525179856e-06,
+      "loss": 0.1212,
+      "step": 120
+    },
+    {
+      "epoch": 0.05251736111111111,
+      "grad_norm": 1.6297051906585693,
+      "learning_rate": 8.705035971223022e-06,
+      "loss": 0.0968,
+      "step": 121
+    },
+    {
+      "epoch": 0.05295138888888889,
+      "grad_norm": 1.4963327646255493,
+      "learning_rate": 8.776978417266188e-06,
+      "loss": 0.0692,
+      "step": 122
+    },
+    {
+      "epoch": 0.053385416666666664,
+      "grad_norm": 1.9528976678848267,
+      "learning_rate": 8.848920863309353e-06,
+      "loss": 0.0871,
+      "step": 123
+    },
+    {
+      "epoch": 0.05381944444444445,
+      "grad_norm": 3.2168524265289307,
+      "learning_rate": 8.92086330935252e-06,
+      "loss": 0.1252,
+      "step": 124
+    },
+    {
+      "epoch": 0.054253472222222224,
+      "grad_norm": 3.5068187713623047,
+      "learning_rate": 8.992805755395683e-06,
+      "loss": 0.1394,
+      "step": 125
+    },
+    {
+      "epoch": 0.0546875,
+      "grad_norm": 2.7131741046905518,
+      "learning_rate": 9.064748201438849e-06,
+      "loss": 0.1179,
+      "step": 126
+    },
+    {
+      "epoch": 0.055121527777777776,
+      "grad_norm": 2.037222146987915,
+      "learning_rate": 9.136690647482015e-06,
+      "loss": 0.0866,
+      "step": 127
+    },
+    {
+      "epoch": 0.05555555555555555,
+      "grad_norm": 1.1361604928970337,
+      "learning_rate": 9.20863309352518e-06,
+      "loss": 0.0829,
+      "step": 128
+    },
+    {
+      "epoch": 0.055989583333333336,
+      "grad_norm": 1.3983852863311768,
+      "learning_rate": 9.280575539568346e-06,
+      "loss": 0.1227,
+      "step": 129
+    },
+    {
+      "epoch": 0.05642361111111111,
+      "grad_norm": 1.1151443719863892,
+      "learning_rate": 9.35251798561151e-06,
+      "loss": 0.0963,
+      "step": 130
+    },
+    {
+      "epoch": 0.05685763888888889,
+      "grad_norm": 1.3063839673995972,
+      "learning_rate": 9.424460431654678e-06,
+      "loss": 0.0969,
+      "step": 131
+    },
+    {
+      "epoch": 0.057291666666666664,
+      "grad_norm": 1.619152307510376,
+      "learning_rate": 9.496402877697842e-06,
+      "loss": 0.1083,
+      "step": 132
+    },
+    {
+      "epoch": 0.05772569444444445,
+      "grad_norm": 1.5226192474365234,
+      "learning_rate": 9.568345323741008e-06,
+      "loss": 0.0849,
+      "step": 133
+    },
+    {
+      "epoch": 0.058159722222222224,
+      "grad_norm": 1.5232475996017456,
+      "learning_rate": 9.640287769784174e-06,
+      "loss": 0.0649,
+      "step": 134
+    },
+    {
+      "epoch": 0.05859375,
+      "grad_norm": 2.6182069778442383,
+      "learning_rate": 9.712230215827338e-06,
+      "loss": 0.0981,
+      "step": 135
+    },
+    {
+      "epoch": 0.059027777777777776,
+      "grad_norm": 2.3720853328704834,
+      "learning_rate": 9.784172661870505e-06,
+      "loss": 0.1011,
+      "step": 136
+    },
+    {
+      "epoch": 0.05946180555555555,
+      "grad_norm": 1.811148762702942,
+      "learning_rate": 9.85611510791367e-06,
+      "loss": 0.0983,
+      "step": 137
+    },
+    {
+      "epoch": 0.059895833333333336,
+      "grad_norm": 1.452284574508667,
+      "learning_rate": 9.928057553956835e-06,
+      "loss": 0.0656,
+      "step": 138
+    },
+    {
+      "epoch": 0.06032986111111111,
+      "grad_norm": 1.6373428106307983,
+      "learning_rate": 1e-05,
+      "loss": 0.1291,
+      "step": 139
+    },
+    {
+      "epoch": 0.06076388888888889,
+      "grad_norm": 1.5119938850402832,
+      "learning_rate": 9.999998764567487e-06,
+      "loss": 0.106,
+      "step": 140
+    },
+    {
+      "epoch": 0.061197916666666664,
+      "grad_norm": 1.1721326112747192,
+      "learning_rate": 9.999995058270555e-06,
+      "loss": 0.1004,
+      "step": 141
+    },
+    {
+      "epoch": 0.06163194444444445,
+      "grad_norm": 1.0162428617477417,
+      "learning_rate": 9.999988881111037e-06,
+      "loss": 0.0773,
+      "step": 142
+    },
+    {
+      "epoch": 0.062065972222222224,
+      "grad_norm": 4.524620532989502,
+      "learning_rate": 9.999980233091988e-06,
+      "loss": 0.0876,
+      "step": 143
+    },
+    {
+      "epoch": 0.0625,
+      "grad_norm": 2.312443733215332,
+      "learning_rate": 9.999969114217678e-06,
+      "loss": 0.1297,
+      "step": 144
+    },
+    {
+      "epoch": 0.06293402777777778,
+      "grad_norm": 2.083632707595825,
+      "learning_rate": 9.999955524493603e-06,
+      "loss": 0.1167,
+      "step": 145
+    },
+    {
+      "epoch": 0.06336805555555555,
+      "grad_norm": 1.0585899353027344,
+      "learning_rate": 9.99993946392648e-06,
+      "loss": 0.0778,
+      "step": 146
+    },
+    {
+      "epoch": 0.06380208333333333,
+      "grad_norm": 1.1303775310516357,
+      "learning_rate": 9.999920932524242e-06,
+      "loss": 0.1124,
+      "step": 147
+    },
+    {
+      "epoch": 0.0642361111111111,
+      "grad_norm": 1.113276720046997,
+      "learning_rate": 9.999899930296052e-06,
+      "loss": 0.1029,
+      "step": 148
+    },
+    {
+      "epoch": 0.0646701388888889,
+      "grad_norm": 1.1997543573379517,
+      "learning_rate": 9.999876457252281e-06,
+      "loss": 0.0998,
+      "step": 149
+    },
+    {
+      "epoch": 0.06510416666666667,
+      "grad_norm": 1.2771512269973755,
+      "learning_rate": 9.999850513404539e-06,
+      "loss": 0.109,
+      "step": 150
+    },
+    {
+      "epoch": 0.06553819444444445,
+      "grad_norm": 1.8055669069290161,
+      "learning_rate": 9.99982209876564e-06,
+      "loss": 0.1004,
+      "step": 151
+    },
+    {
+      "epoch": 0.06597222222222222,
+      "grad_norm": 1.3983486890792847,
+      "learning_rate": 9.999791213349624e-06,
+      "loss": 0.0636,
+      "step": 152
+    },
+    {
+      "epoch": 0.06640625,
+      "grad_norm": 1.934438705444336,
+      "learning_rate": 9.999757857171758e-06,
+      "loss": 0.1031,
+      "step": 153
+    },
+    {
+      "epoch": 0.06684027777777778,
+      "grad_norm": 1.6015806198120117,
+      "learning_rate": 9.999722030248526e-06,
+      "loss": 0.0763,
+      "step": 154
+    },
+    {
+      "epoch": 0.06727430555555555,
+      "grad_norm": 1.175926923751831,
+      "learning_rate": 9.999683732597631e-06,
+      "loss": 0.0813,
+      "step": 155
+    },
+    {
+      "epoch": 0.06770833333333333,
+      "grad_norm": 1.0643317699432373,
+      "learning_rate": 9.999642964238e-06,
+      "loss": 0.081,
+      "step": 156
+    },
+    {
+      "epoch": 0.0681423611111111,
+      "grad_norm": 1.2169849872589111,
+      "learning_rate": 9.999599725189777e-06,
+      "loss": 0.0734,
+      "step": 157
+    },
+    {
+      "epoch": 0.0685763888888889,
+      "grad_norm": 1.030631422996521,
+      "learning_rate": 9.99955401547433e-06,
+      "loss": 0.066,
+      "step": 158
+    },
+    {
+      "epoch": 0.06901041666666667,
+      "grad_norm": 1.4374827146530151,
+      "learning_rate": 9.999505835114252e-06,
+      "loss": 0.0832,
+      "step": 159
+    },
+    {
+      "epoch": 0.06944444444444445,
+      "grad_norm": 1.1652729511260986,
+      "learning_rate": 9.999455184133346e-06,
+      "loss": 0.0626,
+      "step": 160
+    },
+    {
+      "epoch": 0.06987847222222222,
+      "grad_norm": 2.172323226928711,
+      "learning_rate": 9.999402062556648e-06,
+      "loss": 0.1019,
+      "step": 161
+    },
+    {
+      "epoch": 0.0703125,
+      "grad_norm": 1.4461246728897095,
+      "learning_rate": 9.999346470410405e-06,
+      "loss": 0.0822,
+      "step": 162
+    },
+    {
+      "epoch": 0.07074652777777778,
+      "grad_norm": 1.3454128503799438,
+      "learning_rate": 9.999288407722089e-06,
+      "loss": 0.0975,
+      "step": 163
+    },
+    {
+      "epoch": 0.07118055555555555,
+      "grad_norm": 1.2311937808990479,
+      "learning_rate": 9.9992278745204e-06,
+      "loss": 0.0967,
+      "step": 164
+    },
+    {
+      "epoch": 0.07161458333333333,
+      "grad_norm": 0.9970878958702087,
+      "learning_rate": 9.999164870835241e-06,
+      "loss": 0.0988,
+      "step": 165
+    },
+    {
+      "epoch": 0.0720486111111111,
+      "grad_norm": 1.034321904182434,
+      "learning_rate": 9.999099396697755e-06,
+      "loss": 0.1032,
+      "step": 166
+    },
+    {
+      "epoch": 0.0724826388888889,
+      "grad_norm": 1.2639926671981812,
+      "learning_rate": 9.999031452140295e-06,
+      "loss": 0.0922,
+      "step": 167
+    },
+    {
+      "epoch": 0.07291666666666667,
+      "grad_norm": 0.9578790068626404,
+      "learning_rate": 9.998961037196437e-06,
+      "loss": 0.0644,
+      "step": 168
+    },
+    {
+      "epoch": 0.07335069444444445,
+      "grad_norm": 1.2277770042419434,
+      "learning_rate": 9.998888151900979e-06,
+      "loss": 0.0536,
+      "step": 169
+    },
+    {
+      "epoch": 0.07378472222222222,
+      "grad_norm": 2.1698157787323,
+      "learning_rate": 9.998812796289939e-06,
+      "loss": 0.1146,
+      "step": 170
+    },
+    {
+      "epoch": 0.07421875,
+      "grad_norm": 1.36558198928833,
+      "learning_rate": 9.998734970400554e-06,
+      "loss": 0.0612,
+      "step": 171
+    },
+    {
+      "epoch": 0.07465277777777778,
+      "grad_norm": 1.664151906967163,
+      "learning_rate": 9.998654674271285e-06,
+      "loss": 0.0615,
+      "step": 172
+    },
+    {
+      "epoch": 0.07508680555555555,
+      "grad_norm": 1.5223487615585327,
+      "learning_rate": 9.998571907941811e-06,
+      "loss": 0.0874,
+      "step": 173
+    },
+    {
+      "epoch": 0.07552083333333333,
+      "grad_norm": 1.261687994003296,
+      "learning_rate": 9.998486671453034e-06,
+      "loss": 0.0705,
+      "step": 174
+    },
+    {
+      "epoch": 0.0759548611111111,
+      "grad_norm": 1.260436773300171,
+      "learning_rate": 9.998398964847075e-06,
+      "loss": 0.0715,
+      "step": 175
+    },
+    {
+      "epoch": 0.0763888888888889,
+      "grad_norm": 0.9473410248756409,
+      "learning_rate": 9.998308788167277e-06,
+      "loss": 0.0807,
+      "step": 176
+    },
+    {
+      "epoch": 0.07682291666666667,
+      "grad_norm": 1.0099543333053589,
+      "learning_rate": 9.998216141458203e-06,
+      "loss": 0.0871,
+      "step": 177
+    },
+    {
+      "epoch": 0.07725694444444445,
+      "grad_norm": 0.9015682935714722,
+      "learning_rate": 9.998121024765635e-06,
+      "loss": 0.0744,
+      "step": 178
+    },
+    {
+      "epoch": 0.07769097222222222,
+      "grad_norm": 1.6776366233825684,
+      "learning_rate": 9.998023438136578e-06,
+      "loss": 0.0897,
+      "step": 179
+    },
+    {
+      "epoch": 0.078125,
+      "grad_norm": 1.5866495370864868,
+      "learning_rate": 9.997923381619257e-06,
+      "loss": 0.0935,
+      "step": 180
+    },
+    {
+      "epoch": 0.07855902777777778,
+      "grad_norm": 1.2076671123504639,
+      "learning_rate": 9.997820855263117e-06,
+      "loss": 0.089,
+      "step": 181
+    },
+    {
+      "epoch": 0.07899305555555555,
+      "grad_norm": 1.3449783325195312,
+      "learning_rate": 9.997715859118822e-06,
+      "loss": 0.0826,
+      "step": 182
+    },
+    {
+      "epoch": 0.07942708333333333,
+      "grad_norm": 1.6158901453018188,
+      "learning_rate": 9.99760839323826e-06,
+      "loss": 0.1154,
+      "step": 183
+    },
+    {
+      "epoch": 0.0798611111111111,
+      "grad_norm": 1.1235649585723877,
+      "learning_rate": 9.997498457674539e-06,
+      "loss": 0.0851,
+      "step": 184
+    },
+    {
+      "epoch": 0.0802951388888889,
+      "grad_norm": 0.9119734168052673,
+      "learning_rate": 9.997386052481985e-06,
+      "loss": 0.1149,
+      "step": 185
+    },
+    {
+      "epoch": 0.08072916666666667,
+      "grad_norm": 1.1313296556472778,
+      "learning_rate": 9.997271177716146e-06,
+      "loss": 0.073,
+      "step": 186
+    },
+    {
+      "epoch": 0.08116319444444445,
+      "grad_norm": 0.9562469720840454,
+      "learning_rate": 9.997153833433789e-06,
+      "loss": 0.0891,
+      "step": 187
+    },
+    {
+      "epoch": 0.08159722222222222,
+      "grad_norm": 0.916240930557251,
+      "learning_rate": 9.9970340196929e-06,
+      "loss": 0.1016,
+      "step": 188
+    },
+    {
+      "epoch": 0.08203125,
+      "grad_norm": 1.28097665309906,
+      "learning_rate": 9.996911736552694e-06,
+      "loss": 0.0879,
+      "step": 189
+    },
+    {
+      "epoch": 0.08246527777777778,
+      "grad_norm": 1.1469060182571411,
+      "learning_rate": 9.996786984073594e-06,
+      "loss": 0.0563,
+      "step": 190
+    },
+    {
+      "epoch": 0.08289930555555555,
+      "grad_norm": 3.393587112426758,
+      "learning_rate": 9.996659762317254e-06,
+      "loss": 0.1082,
+      "step": 191
+    },
+    {
+      "epoch": 0.08333333333333333,
+      "grad_norm": 3.045605421066284,
+      "learning_rate": 9.99653007134654e-06,
+      "loss": 0.0754,
+      "step": 192
+    },
+    {
+      "epoch": 0.0837673611111111,
+      "grad_norm": 1.3723241090774536,
+      "learning_rate": 9.996397911225544e-06,
+      "loss": 0.067,
+      "step": 193
+    },
+    {
+      "epoch": 0.0842013888888889,
+      "grad_norm": 1.394571304321289,
+      "learning_rate": 9.996263282019574e-06,
+      "loss": 0.0915,
+      "step": 194
+    },
+    {
+      "epoch": 0.08463541666666667,
+      "grad_norm": 1.0956934690475464,
+      "learning_rate": 9.996126183795162e-06,
+      "loss": 0.0637,
+      "step": 195
+    },
+    {
+      "epoch": 0.08506944444444445,
+      "grad_norm": 0.8720288872718811,
+      "learning_rate": 9.995986616620057e-06,
+      "loss": 0.0809,
+      "step": 196
+    },
+    {
+      "epoch": 0.08550347222222222,
+      "grad_norm": 1.0597174167633057,
+      "learning_rate": 9.995844580563232e-06,
+      "loss": 0.0856,
+      "step": 197
+    },
+    {
+      "epoch": 0.0859375,
+      "grad_norm": 1.317156195640564,
+      "learning_rate": 9.995700075694872e-06,
+      "loss": 0.0994,
+      "step": 198
+    },
+    {
+      "epoch": 0.08637152777777778,
+      "grad_norm": 1.026953935623169,
+      "learning_rate": 9.995553102086392e-06,
+      "loss": 0.0891,
+      "step": 199
+    },
+    {
+      "epoch": 0.08680555555555555,
+      "grad_norm": 1.7246663570404053,
+      "learning_rate": 9.995403659810422e-06,
+      "loss": 0.1019,
+      "step": 200
+    },
+    {
+      "epoch": 0.08723958333333333,
+      "grad_norm": 0.7544384598731995,
+      "learning_rate": 9.995251748940811e-06,
+      "loss": 0.0577,
+      "step": 201
+    },
+    {
+      "epoch": 0.0876736111111111,
+      "grad_norm": 0.9751232266426086,
+      "learning_rate": 9.995097369552628e-06,
+      "loss": 0.0864,
+      "step": 202
+    },
+    {
+      "epoch": 0.0881076388888889,
+      "grad_norm": 1.1838496923446655,
+      "learning_rate": 9.994940521722167e-06,
+      "loss": 0.0692,
+      "step": 203
+    },
+    {
+      "epoch": 0.08854166666666667,
+      "grad_norm": 1.2722941637039185,
+      "learning_rate": 9.994781205526935e-06,
+      "loss": 0.0854,
+      "step": 204
+    },
+    {
+      "epoch": 0.08897569444444445,
+      "grad_norm": 1.445184350013733,
+      "learning_rate": 9.994619421045661e-06,
+      "loss": 0.1057,
+      "step": 205
+    },
+    {
+      "epoch": 0.08940972222222222,
+      "grad_norm": 0.7705734968185425,
+      "learning_rate": 9.994455168358297e-06,
+      "loss": 0.0696,
+      "step": 206
+    },
+    {
+      "epoch": 0.08984375,
+      "grad_norm": 0.8373009562492371,
+      "learning_rate": 9.994288447546013e-06,
+      "loss": 0.0837,
+      "step": 207
+    },
+    {
+      "epoch": 0.09027777777777778,
+      "grad_norm": 0.6120887994766235,
+      "learning_rate": 9.994119258691192e-06,
+      "loss": 0.0622,
+      "step": 208
+    },
+    {
+      "epoch": 0.09071180555555555,
+      "grad_norm": 0.9396751523017883,
+      "learning_rate": 9.993947601877451e-06,
+      "loss": 0.078,
+      "step": 209
+    },
+    {
+      "epoch": 0.09114583333333333,
+      "grad_norm": 1.0563628673553467,
+      "learning_rate": 9.993773477189613e-06,
+      "loss": 0.0684,
+      "step": 210
+    },
+    {
+      "epoch": 0.0915798611111111,
+      "grad_norm": 1.0072035789489746,
+      "learning_rate": 9.993596884713724e-06,
+      "loss": 0.0708,
+      "step": 211
+    },
+    {
+      "epoch": 0.0920138888888889,
+      "grad_norm": 1.5425307750701904,
+      "learning_rate": 9.993417824537055e-06,
+      "loss": 0.0955,
+      "step": 212
+    },
+    {
+      "epoch": 0.09244791666666667,
+      "grad_norm": 1.181329369544983,
+      "learning_rate": 9.993236296748093e-06,
+      "loss": 0.0776,
+      "step": 213
+    },
+    {
+      "epoch": 0.09288194444444445,
+      "grad_norm": 0.8895711898803711,
+      "learning_rate": 9.99305230143654e-06,
+      "loss": 0.0695,
+      "step": 214
+    },
+    {
+      "epoch": 0.09331597222222222,
+      "grad_norm": 1.0264010429382324,
+      "learning_rate": 9.992865838693328e-06,
+      "loss": 0.091,
+      "step": 215
+    },
+    {
+      "epoch": 0.09375,
+      "grad_norm": 0.741313099861145,
+      "learning_rate": 9.992676908610595e-06,
+      "loss": 0.0735,
+      "step": 216
+    },
+    {
+      "epoch": 0.09418402777777778,
+      "grad_norm": 0.715937614440918,
+      "learning_rate": 9.99248551128171e-06,
+      "loss": 0.0908,
+      "step": 217
+    },
+    {
+      "epoch": 0.09461805555555555,
+      "grad_norm": 0.7246452569961548,
+      "learning_rate": 9.992291646801253e-06,
+      "loss": 0.0793,
+      "step": 218
+    },
+    {
+      "epoch": 0.09505208333333333,
+      "grad_norm": 0.8316558003425598,
+      "learning_rate": 9.99209531526503e-06,
+      "loss": 0.0854,
+      "step": 219
+    },
+    {
+      "epoch": 0.0954861111111111,
+      "grad_norm": 0.8730294704437256,
+      "learning_rate": 9.991896516770058e-06,
+      "loss": 0.0746,
+      "step": 220
+    },
+    {
+      "epoch": 0.0959201388888889,
+      "grad_norm": 0.8386092185974121,
+      "learning_rate": 9.991695251414584e-06,
+      "loss": 0.0619,
+      "step": 221
+    },
+    {
+      "epoch": 0.09635416666666667,
+      "grad_norm": 1.0405433177947998,
+      "learning_rate": 9.991491519298063e-06,
+      "loss": 0.0892,
+      "step": 222
+    },
+    {
+      "epoch": 0.09678819444444445,
+      "grad_norm": 1.1907081604003906,
+      "learning_rate": 9.991285320521176e-06,
+      "loss": 0.0581,
+      "step": 223
+    },
+    {
+      "epoch": 0.09722222222222222,
+      "grad_norm": 2.635026693344116,
+      "learning_rate": 9.99107665518582e-06,
+      "loss": 0.0631,
+      "step": 224
+    },
+    {
+      "epoch": 0.09765625,
+      "grad_norm": 1.6267889738082886,
+      "learning_rate": 9.990865523395113e-06,
+      "loss": 0.0853,
+      "step": 225
+    },
+    {
+      "epoch": 0.09809027777777778,
+      "grad_norm": 0.743120551109314,
+      "learning_rate": 9.990651925253389e-06,
+      "loss": 0.0472,
+      "step": 226
+    },
+    {
+      "epoch": 0.09852430555555555,
+      "grad_norm": 1.1194217205047607,
+      "learning_rate": 9.990435860866204e-06,
+      "loss": 0.102,
+      "step": 227
+    },
+    {
+      "epoch": 0.09895833333333333,
+      "grad_norm": 0.7726638913154602,
+      "learning_rate": 9.99021733034033e-06,
+      "loss": 0.0681,
+      "step": 228
+    },
+    {
+      "epoch": 0.0993923611111111,
+      "grad_norm": 1.217424750328064,
+      "learning_rate": 9.98999633378376e-06,
+      "loss": 0.0882,
+      "step": 229
+    },
+    {
+      "epoch": 0.0998263888888889,
+      "grad_norm": 0.8821778297424316,
+      "learning_rate": 9.989772871305704e-06,
+      "loss": 0.0991,
+      "step": 230
+    },
+    {
+      "epoch": 0.10026041666666667,
+      "grad_norm": 1.0021941661834717,
+      "learning_rate": 9.98954694301659e-06,
+      "loss": 0.0786,
+      "step": 231
+    },
+    {
+      "epoch": 0.10069444444444445,
+      "grad_norm": 0.8431205749511719,
+      "learning_rate": 9.989318549028068e-06,
+      "loss": 0.0872,
+      "step": 232
+    },
+    {
+      "epoch": 0.10112847222222222,
+      "grad_norm": 0.9266021847724915,
+      "learning_rate": 9.989087689453001e-06,
+      "loss": 0.0735,
+      "step": 233
+    },
+    {
+      "epoch": 0.1015625,
+      "grad_norm": 1.1363204717636108,
+      "learning_rate": 9.988854364405477e-06,
+      "loss": 0.0777,
+      "step": 234
+    },
+    {
+      "epoch": 0.10199652777777778,
+      "grad_norm": 0.9753544330596924,
+      "learning_rate": 9.988618574000796e-06,
+      "loss": 0.069,
+      "step": 235
+    },
+    {
+      "epoch": 0.10243055555555555,
+      "grad_norm": 1.2155951261520386,
+      "learning_rate": 9.988380318355483e-06,
+      "loss": 0.0821,
+      "step": 236
+    },
+    {
+      "epoch": 0.10286458333333333,
+      "grad_norm": 0.957694947719574,
+      "learning_rate": 9.988139597587274e-06,
+      "loss": 0.0602,
+      "step": 237
+    },
+    {
+      "epoch": 0.1032986111111111,
+      "grad_norm": 0.958967387676239,
+      "learning_rate": 9.987896411815127e-06,
+      "loss": 0.0708,
+      "step": 238
+    },
+    {
+      "epoch": 0.1037326388888889,
+      "grad_norm": 1.111096978187561,
+      "learning_rate": 9.987650761159218e-06,
+      "loss": 0.0829,
+      "step": 239
+    },
+    {
+      "epoch": 0.10416666666666667,
+      "grad_norm": 0.9012123942375183,
+      "learning_rate": 9.987402645740943e-06,
+      "loss": 0.0839,
+      "step": 240
+    },
+    {
+      "epoch": 0.10460069444444445,
+      "grad_norm": 0.6766245365142822,
+      "learning_rate": 9.987152065682913e-06,
+      "loss": 0.0814,
+      "step": 241
+    },
+    {
+      "epoch": 0.10503472222222222,
+      "grad_norm": 0.7312656044960022,
+      "learning_rate": 9.986899021108957e-06,
+      "loss": 0.0674,
+      "step": 242
+    },
+    {
+      "epoch": 0.10546875,
+      "grad_norm": 1.022851824760437,
+      "learning_rate": 9.986643512144122e-06,
+      "loss": 0.0824,
+      "step": 243
+    },
+    {
+      "epoch": 0.10590277777777778,
+      "grad_norm": 0.8258888125419617,
+      "learning_rate": 9.986385538914676e-06,
+      "loss": 0.0866,
+      "step": 244
+    },
+    {
+      "epoch": 0.10633680555555555,
+      "grad_norm": 0.804434597492218,
+      "learning_rate": 9.986125101548101e-06,
+      "loss": 0.0868,
+      "step": 245
+    },
+    {
+      "epoch": 0.10677083333333333,
+      "grad_norm": 0.7631927132606506,
+      "learning_rate": 9.9858622001731e-06,
+      "loss": 0.0798,
+      "step": 246
+    },
+    {
+      "epoch": 0.1072048611111111,
+      "grad_norm": 0.703942596912384,
+      "learning_rate": 9.985596834919587e-06,
+      "loss": 0.0598,
+      "step": 247
+    },
+    {
+      "epoch": 0.1076388888888889,
+      "grad_norm": 0.9682087898254395,
+      "learning_rate": 9.985329005918702e-06,
+      "loss": 0.0817,
+      "step": 248
+    },
+    {
+      "epoch": 0.10807291666666667,
+      "grad_norm": 1.005095362663269,
+      "learning_rate": 9.9850587133028e-06,
+      "loss": 0.0734,
+      "step": 249
+    },
+    {
+      "epoch": 0.10850694444444445,
+      "grad_norm": 1.0828065872192383,
+      "learning_rate": 9.98478595720545e-06,
+      "loss": 0.0788,
+      "step": 250
+    },
+    {
+      "epoch": 0.10894097222222222,
+      "grad_norm": 1.217239499092102,
+      "learning_rate": 9.984510737761441e-06,
+      "loss": 0.1023,
+      "step": 251
+    },
+    {
+      "epoch": 0.109375,
+      "grad_norm": 0.8663279414176941,
+      "learning_rate": 9.98423305510678e-06,
+      "loss": 0.0839,
+      "step": 252
+    },
+    {
+      "epoch": 0.10980902777777778,
+      "grad_norm": 0.8627954125404358,
+      "learning_rate": 9.98395290937869e-06,
+      "loss": 0.0661,
+      "step": 253
+    },
+    {
+      "epoch": 0.11024305555555555,
+      "grad_norm": 1.0426639318466187,
+      "learning_rate": 9.98367030071561e-06,
+      "loss": 0.0921,
+      "step": 254
+    },
+    {
+      "epoch": 0.11067708333333333,
+      "grad_norm": 0.7946957945823669,
+      "learning_rate": 9.983385229257201e-06,
+      "loss": 0.0455,
+      "step": 255
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.8853251338005066,
+      "learning_rate": 9.983097695144332e-06,
+      "loss": 0.0662,
+      "step": 256
+    },
+    {
+      "epoch": 0.1115451388888889,
+      "grad_norm": 0.9303094744682312,
+      "learning_rate": 9.982807698519101e-06,
+      "loss": 0.0709,
+      "step": 257
+    },
+    {
+      "epoch": 0.11197916666666667,
+      "grad_norm": 2.260749101638794,
+      "learning_rate": 9.982515239524811e-06,
+      "loss": 0.08,
+      "step": 258
+    },
+    {
+      "epoch": 0.11241319444444445,
+      "grad_norm": 1.252661108970642,
+      "learning_rate": 9.98222031830599e-06,
+      "loss": 0.0714,
+      "step": 259
+    },
+    {
+      "epoch": 0.11284722222222222,
+      "grad_norm": 0.9974795579910278,
+      "learning_rate": 9.981922935008381e-06,
+      "loss": 0.0602,
+      "step": 260
+    },
+    {
+      "epoch": 0.11328125,
+      "grad_norm": 1.5989861488342285,
+      "learning_rate": 9.98162308977894e-06,
+      "loss": 0.0771,
+      "step": 261
+    },
+    {
+      "epoch": 0.11371527777777778,
+      "grad_norm": 2.777109384536743,
+      "learning_rate": 9.981320782765847e-06,
+      "loss": 0.1554,
+      "step": 262
+    },
+    {
+      "epoch": 0.11414930555555555,
+      "grad_norm": 1.3210549354553223,
+      "learning_rate": 9.98101601411849e-06,
+      "loss": 0.0628,
+      "step": 263
+    },
+    {
+      "epoch": 0.11458333333333333,
+      "grad_norm": 1.3631166219711304,
+      "learning_rate": 9.980708783987476e-06,
+      "loss": 0.0743,
+      "step": 264
+    },
+    {
+      "epoch": 0.1150173611111111,
+      "grad_norm": 1.32863450050354,
+      "learning_rate": 9.980399092524635e-06,
+      "loss": 0.1134,
+      "step": 265
+    },
+    {
+      "epoch": 0.1154513888888889,
+      "grad_norm": 0.6396360993385315,
+      "learning_rate": 9.980086939883004e-06,
+      "loss": 0.0687,
+      "step": 266
+    },
+    {
+      "epoch": 0.11588541666666667,
+      "grad_norm": 0.9112362861633301,
+      "learning_rate": 9.979772326216842e-06,
+      "loss": 0.0576,
+      "step": 267
+    },
+    {
+      "epoch": 0.11631944444444445,
+      "grad_norm": 0.8859538435935974,
+      "learning_rate": 9.979455251681622e-06,
+      "loss": 0.0751,
+      "step": 268
+    },
+    {
+      "epoch": 0.11675347222222222,
+      "grad_norm": 1.0131158828735352,
+      "learning_rate": 9.979135716434036e-06,
+      "loss": 0.0616,
+      "step": 269
+    },
+    {
+      "epoch": 0.1171875,
+      "grad_norm": 0.7546622157096863,
+      "learning_rate": 9.978813720631986e-06,
+      "loss": 0.0537,
+      "step": 270
+    },
+    {
+      "epoch": 0.11762152777777778,
+      "grad_norm": 1.2106804847717285,
+      "learning_rate": 9.978489264434597e-06,
+      "loss": 0.076,
+      "step": 271
+    },
+    {
+      "epoch": 0.11805555555555555,
+      "grad_norm": 0.7226711511611938,
+      "learning_rate": 9.978162348002205e-06,
+      "loss": 0.0513,
+      "step": 272
+    },
+    {
+      "epoch": 0.11848958333333333,
+      "grad_norm": 1.3641725778579712,
+      "learning_rate": 9.977832971496362e-06,
+      "loss": 0.084,
+      "step": 273
+    },
+    {
+      "epoch": 0.1189236111111111,
+      "grad_norm": 1.5535808801651,
+      "learning_rate": 9.977501135079839e-06,
+      "loss": 0.097,
+      "step": 274
+    },
+    {
+      "epoch": 0.1193576388888889,
+      "grad_norm": 0.8176865577697754,
+      "learning_rate": 9.977166838916619e-06,
+      "loss": 0.0646,
+      "step": 275
+    },
+    {
+      "epoch": 0.11979166666666667,
+      "grad_norm": 1.3481011390686035,
+      "learning_rate": 9.976830083171903e-06,
+      "loss": 0.0823,
+      "step": 276
+    },
+    {
+      "epoch": 0.12022569444444445,
+      "grad_norm": 1.4268255233764648,
+      "learning_rate": 9.976490868012108e-06,
+      "loss": 0.101,
+      "step": 277
+    },
+    {
+      "epoch": 0.12065972222222222,
+      "grad_norm": 1.1602445840835571,
+      "learning_rate": 9.976149193604862e-06,
+      "loss": 0.0882,
+      "step": 278
+    },
+    {
+      "epoch": 0.12109375,
+      "grad_norm": 0.6257598996162415,
+      "learning_rate": 9.975805060119013e-06,
+      "loss": 0.0558,
+      "step": 279
+    },
+    {
+      "epoch": 0.12152777777777778,
+      "grad_norm": 0.7104304432868958,
+      "learning_rate": 9.975458467724624e-06,
+      "loss": 0.079,
+      "step": 280
+    },
+    {
+      "epoch": 0.12196180555555555,
+      "grad_norm": 0.691241443157196,
+      "learning_rate": 9.975109416592967e-06,
+      "loss": 0.0705,
+      "step": 281
+    },
+    {
+      "epoch": 0.12239583333333333,
+      "grad_norm": 0.8947821855545044,
+      "learning_rate": 9.974757906896539e-06,
+      "loss": 0.0751,
+      "step": 282
+    },
+    {
+      "epoch": 0.1228298611111111,
+      "grad_norm": 0.8744166493415833,
+      "learning_rate": 9.974403938809044e-06,
+      "loss": 0.1026,
+      "step": 283
+    },
+    {
+      "epoch": 0.1232638888888889,
+      "grad_norm": 1.5625134706497192,
+      "learning_rate": 9.974047512505401e-06,
+      "loss": 0.0952,
+      "step": 284
+    },
+    {
+      "epoch": 0.12369791666666667,
+      "grad_norm": 1.6969810724258423,
+      "learning_rate": 9.97368862816175e-06,
+      "loss": 0.1144,
+      "step": 285
+    },
+    {
+      "epoch": 0.12413194444444445,
+      "grad_norm": 1.002626895904541,
+      "learning_rate": 9.973327285955442e-06,
+      "loss": 0.068,
+      "step": 286
+    },
+    {
+      "epoch": 0.12456597222222222,
+      "grad_norm": 0.8302503228187561,
+      "learning_rate": 9.972963486065042e-06,
+      "loss": 0.0987,
+      "step": 287
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 0.6188612580299377,
+      "learning_rate": 9.972597228670327e-06,
+      "loss": 0.0643,
+      "step": 288
+    },
+    {
+      "epoch": 0.1254340277777778,
+      "grad_norm": 0.9741168022155762,
+      "learning_rate": 9.972228513952295e-06,
+      "loss": 0.0579,
+      "step": 289
+    },
+    {
+      "epoch": 0.12586805555555555,
+      "grad_norm": 1.0956751108169556,
+      "learning_rate": 9.971857342093154e-06,
+      "loss": 0.0529,
+      "step": 290
+    },
+    {
+      "epoch": 0.12630208333333334,
+      "grad_norm": 1.3885653018951416,
+      "learning_rate": 9.971483713276326e-06,
+      "loss": 0.0843,
+      "step": 291
+    },
+    {
+      "epoch": 0.1267361111111111,
+      "grad_norm": 0.7927285432815552,
+      "learning_rate": 9.971107627686452e-06,
+      "loss": 0.0587,
+      "step": 292
+    },
+    {
+      "epoch": 0.1271701388888889,
+      "grad_norm": 0.9241272807121277,
+      "learning_rate": 9.970729085509379e-06,
+      "loss": 0.0531,
+      "step": 293
+    },
+    {
+      "epoch": 0.12760416666666666,
+      "grad_norm": 1.4300391674041748,
+      "learning_rate": 9.970348086932174e-06,
+      "loss": 0.1115,
+      "step": 294
+    },
+    {
+      "epoch": 0.12803819444444445,
+      "grad_norm": 0.7712482213973999,
+      "learning_rate": 9.969964632143114e-06,
+      "loss": 0.0469,
+      "step": 295
+    },
+    {
+      "epoch": 0.1284722222222222,
+      "grad_norm": 0.7699704766273499,
+      "learning_rate": 9.969578721331696e-06,
+      "loss": 0.0685,
+      "step": 296
+    },
+    {
+      "epoch": 0.12890625,
+      "grad_norm": 0.6662436127662659,
+      "learning_rate": 9.969190354688625e-06,
+      "loss": 0.0694,
+      "step": 297
+    },
+    {
+      "epoch": 0.1293402777777778,
+      "grad_norm": 0.7108180522918701,
+      "learning_rate": 9.968799532405823e-06,
+      "loss": 0.0681,
+      "step": 298
+    },
+    {
+      "epoch": 0.12977430555555555,
+      "grad_norm": 1.0249931812286377,
+      "learning_rate": 9.968406254676418e-06,
+      "loss": 0.0977,
+      "step": 299
+    },
+    {
+      "epoch": 0.13020833333333334,
+      "grad_norm": 0.7156983017921448,
+      "learning_rate": 9.968010521694762e-06,
+      "loss": 0.0662,
+      "step": 300
+    },
+    {
+      "epoch": 0.13020833333333334,
+      "eval_layout_rm_rank_test_accuracy": 0.9643132637732938,
+      "eval_layout_rm_rank_test_chosen_scores": 0.6622011065483093,
+      "eval_layout_rm_rank_test_loss": 0.07006000727415085,
+      "eval_layout_rm_rank_test_rejected_scores": 0.12240368872880936,
+      "eval_layout_rm_rank_test_runtime": 1177.9284,
+      "eval_layout_rm_rank_test_samples_per_second": 20.91,
+      "eval_layout_rm_rank_test_steps_per_second": 0.654,
+      "step": 300
+    },
+    {
+      "epoch": 0.1306423611111111,
+      "grad_norm": 0.8029769659042358,
+      "learning_rate": 9.967612333656417e-06,
+      "loss": 0.0798,
+      "step": 301
+    },
+    {
+      "epoch": 0.1310763888888889,
+      "grad_norm": 0.6051777601242065,
+      "learning_rate": 9.967211690758154e-06,
+      "loss": 0.038,
+      "step": 302
+    },
+    {
+      "epoch": 0.13151041666666666,
+      "grad_norm": 0.7547597289085388,
+      "learning_rate": 9.966808593197959e-06,
+      "loss": 0.0705,
+      "step": 303
+    },
+    {
+      "epoch": 0.13194444444444445,
+      "grad_norm": 1.097530484199524,
+      "learning_rate": 9.966403041175034e-06,
+      "loss": 0.0771,
+      "step": 304
+    },
+    {
+      "epoch": 0.1323784722222222,
+      "grad_norm": 1.216100811958313,
+      "learning_rate": 9.965995034889792e-06,
+      "loss": 0.0835,
+      "step": 305
+    },
+    {
+      "epoch": 0.1328125,
+      "grad_norm": 1.5321711301803589,
+      "learning_rate": 9.965584574543857e-06,
+      "loss": 0.0677,
+      "step": 306
+    },
+    {
+      "epoch": 0.1332465277777778,
+      "grad_norm": 1.0033652782440186,
+      "learning_rate": 9.96517166034007e-06,
+      "loss": 0.0707,
+      "step": 307
+    },
+    {
+      "epoch": 0.13368055555555555,
+      "grad_norm": 0.8167241215705872,
+      "learning_rate": 9.964756292482479e-06,
+      "loss": 0.0805,
+      "step": 308
+    },
+    {
+      "epoch": 0.13411458333333334,
+      "grad_norm": 0.8531818389892578,
+      "learning_rate": 9.964338471176349e-06,
+      "loss": 0.0721,
+      "step": 309
+    },
+    {
+      "epoch": 0.1345486111111111,
+      "grad_norm": 0.6103355884552002,
+      "learning_rate": 9.963918196628155e-06,
+      "loss": 0.049,
+      "step": 310
+    },
+    {
+      "epoch": 0.1349826388888889,
+      "grad_norm": 1.079995036125183,
+      "learning_rate": 9.963495469045587e-06,
+      "loss": 0.0943,
+      "step": 311
+    },
+    {
+      "epoch": 0.13541666666666666,
+      "grad_norm": 0.9208278656005859,
+      "learning_rate": 9.963070288637547e-06,
+      "loss": 0.0703,
+      "step": 312
+    },
+    {
+      "epoch": 0.13585069444444445,
+      "grad_norm": 0.7477347254753113,
+      "learning_rate": 9.962642655614143e-06,
+      "loss": 0.0477,
+      "step": 313
+    },
+    {
+      "epoch": 0.1362847222222222,
+      "grad_norm": 1.0080677270889282,
+      "learning_rate": 9.962212570186702e-06,
+      "loss": 0.0738,
+      "step": 314
+    },
+    {
+      "epoch": 0.13671875,
+      "grad_norm": 1.0213791131973267,
+      "learning_rate": 9.961780032567763e-06,
+      "loss": 0.0788,
+      "step": 315
+    },
+    {
+      "epoch": 0.1371527777777778,
+      "grad_norm": 1.3381445407867432,
+      "learning_rate": 9.961345042971071e-06,
+      "loss": 0.0824,
+      "step": 316
+    },
+    {
+      "epoch": 0.13758680555555555,
+      "grad_norm": 1.2868865728378296,
+      "learning_rate": 9.960907601611587e-06,
+      "loss": 0.099,
+      "step": 317
+    },
+    {
+      "epoch": 0.13802083333333334,
+      "grad_norm": 1.0978431701660156,
+      "learning_rate": 9.960467708705484e-06,
+      "loss": 0.0673,
+      "step": 318
+    },
+    {
+      "epoch": 0.1384548611111111,
+      "grad_norm": 0.8579413890838623,
+      "learning_rate": 9.960025364470145e-06,
+      "loss": 0.0579,
+      "step": 319
+    },
+    {
+      "epoch": 0.1388888888888889,
+      "grad_norm": 1.3059312105178833,
+      "learning_rate": 9.959580569124164e-06,
+      "loss": 0.1266,
+      "step": 320
+    },
+    {
+      "epoch": 0.13932291666666666,
+      "grad_norm": 0.8588614463806152,
+      "learning_rate": 9.959133322887345e-06,
+      "loss": 0.0716,
+      "step": 321
+    },
+    {
+      "epoch": 0.13975694444444445,
+      "grad_norm": 0.669397234916687,
+      "learning_rate": 9.958683625980708e-06,
+      "loss": 0.0721,
+      "step": 322
+    },
+    {
+      "epoch": 0.1401909722222222,
+      "grad_norm": 0.7610234022140503,
+      "learning_rate": 9.958231478626477e-06,
+      "loss": 0.0773,
+      "step": 323
+    },
+    {
+      "epoch": 0.140625,
+      "grad_norm": 0.6288635730743408,
+      "learning_rate": 9.957776881048097e-06,
+      "loss": 0.0824,
+      "step": 324
+    },
+    {
+      "epoch": 0.1410590277777778,
+      "grad_norm": 0.8414584994316101,
+      "learning_rate": 9.957319833470214e-06,
+      "loss": 0.0883,
+      "step": 325
+    },
+    {
+      "epoch": 0.14149305555555555,
+      "grad_norm": 0.6574875712394714,
+      "learning_rate": 9.95686033611869e-06,
+      "loss": 0.0778,
+      "step": 326
+    },
+    {
+      "epoch": 0.14192708333333334,
+      "grad_norm": 0.8761506676673889,
+      "learning_rate": 9.956398389220594e-06,
+      "loss": 0.0641,
+      "step": 327
+    },
+    {
+      "epoch": 0.1423611111111111,
+      "grad_norm": 0.7599236369132996,
+      "learning_rate": 9.95593399300421e-06,
+      "loss": 0.0696,
+      "step": 328
+    },
+    {
+      "epoch": 0.1427951388888889,
+      "grad_norm": 0.8212679624557495,
+      "learning_rate": 9.95546714769903e-06,
+      "loss": 0.0717,
+      "step": 329
+    },
+    {
+      "epoch": 0.14322916666666666,
+      "grad_norm": 0.7221580147743225,
+      "learning_rate": 9.954997853535754e-06,
+      "loss": 0.0418,
+      "step": 330
+    },
+    {
+      "epoch": 0.14366319444444445,
+      "grad_norm": 0.7911623120307922,
+      "learning_rate": 9.954526110746297e-06,
+      "loss": 0.0755,
+      "step": 331
+    },
+    {
+      "epoch": 0.1440972222222222,
+      "grad_norm": 0.7278416156768799,
+      "learning_rate": 9.95405191956378e-06,
+      "loss": 0.0622,
+      "step": 332
+    },
+    {
+      "epoch": 0.14453125,
+      "grad_norm": 0.8105444312095642,
+      "learning_rate": 9.953575280222537e-06,
+      "loss": 0.0554,
+      "step": 333
+    },
+    {
+      "epoch": 0.1449652777777778,
+      "grad_norm": 1.0872361660003662,
+      "learning_rate": 9.953096192958109e-06,
+      "loss": 0.0672,
+      "step": 334
+    },
+    {
+      "epoch": 0.14539930555555555,
+      "grad_norm": 1.1132135391235352,
+      "learning_rate": 9.952614658007247e-06,
+      "loss": 0.0759,
+      "step": 335
+    },
+    {
+      "epoch": 0.14583333333333334,
+      "grad_norm": 0.6996872425079346,
+      "learning_rate": 9.952130675607917e-06,
+      "loss": 0.057,
+      "step": 336
+    },
+    {
+      "epoch": 0.1462673611111111,
+      "grad_norm": 0.7257365584373474,
+      "learning_rate": 9.951644245999285e-06,
+      "loss": 0.0601,
+      "step": 337
+    },
+    {
+      "epoch": 0.1467013888888889,
+      "grad_norm": 1.0367202758789062,
+      "learning_rate": 9.951155369421733e-06,
+      "loss": 0.0853,
+      "step": 338
+    },
+    {
+      "epoch": 0.14713541666666666,
+      "grad_norm": 1.2991832494735718,
+      "learning_rate": 9.950664046116852e-06,
+      "loss": 0.0716,
+      "step": 339
+    },
+    {
+      "epoch": 0.14756944444444445,
+      "grad_norm": 0.9172611832618713,
+      "learning_rate": 9.95017027632744e-06,
+      "loss": 0.0646,
+      "step": 340
+    },
+    {
+      "epoch": 0.1480034722222222,
+      "grad_norm": 0.8245013952255249,
+      "learning_rate": 9.949674060297504e-06,
+      "loss": 0.0854,
+      "step": 341
+    },
+    {
+      "epoch": 0.1484375,
+      "grad_norm": 0.7438561320304871,
+      "learning_rate": 9.949175398272259e-06,
+      "loss": 0.0536,
+      "step": 342
+    },
+    {
+      "epoch": 0.1488715277777778,
+      "grad_norm": 0.6299500465393066,
+      "learning_rate": 9.948674290498134e-06,
+      "loss": 0.0459,
+      "step": 343
+    },
+    {
+      "epoch": 0.14930555555555555,
+      "grad_norm": 0.7602421641349792,
+      "learning_rate": 9.948170737222763e-06,
+      "loss": 0.078,
+      "step": 344
+    },
+    {
+      "epoch": 0.14973958333333334,
+      "grad_norm": 0.6525736451148987,
+      "learning_rate": 9.947664738694985e-06,
+      "loss": 0.0601,
+      "step": 345
+    },
+    {
+      "epoch": 0.1501736111111111,
+      "grad_norm": 1.00484299659729,
+      "learning_rate": 9.947156295164853e-06,
+      "loss": 0.0675,
+      "step": 346
+    },
+    {
+      "epoch": 0.1506076388888889,
+      "grad_norm": 0.8091078996658325,
+      "learning_rate": 9.946645406883625e-06,
+      "loss": 0.051,
+      "step": 347
+    },
+    {
+      "epoch": 0.15104166666666666,
+      "grad_norm": 0.7195174098014832,
+      "learning_rate": 9.946132074103769e-06,
+      "loss": 0.0546,
+      "step": 348
+    },
+    {
+      "epoch": 0.15147569444444445,
+      "grad_norm": 2.212528944015503,
+      "learning_rate": 9.94561629707896e-06,
+      "loss": 0.105,
+      "step": 349
+    },
+    {
+      "epoch": 0.1519097222222222,
+      "grad_norm": 1.7493406534194946,
+      "learning_rate": 9.94509807606408e-06,
+      "loss": 0.1233,
+      "step": 350
+    },
+    {
+      "epoch": 0.15234375,
+      "grad_norm": 1.3774362802505493,
+      "learning_rate": 9.944577411315222e-06,
+      "loss": 0.0697,
+      "step": 351
+    },
+    {
+      "epoch": 0.1527777777777778,
+      "grad_norm": 0.7310093641281128,
+      "learning_rate": 9.944054303089682e-06,
+      "loss": 0.0711,
+      "step": 352
+    },
+    {
+      "epoch": 0.15321180555555555,
+      "grad_norm": 0.8248127698898315,
+      "learning_rate": 9.94352875164597e-06,
+      "loss": 0.0725,
+      "step": 353
+    },
+    {
+      "epoch": 0.15364583333333334,
+      "grad_norm": 0.822748601436615,
+      "learning_rate": 9.943000757243795e-06,
+      "loss": 0.0709,
+      "step": 354
+    },
+    {
+      "epoch": 0.1540798611111111,
+      "grad_norm": 0.6451482772827148,
+      "learning_rate": 9.94247032014408e-06,
+      "loss": 0.0572,
+      "step": 355
+    },
+    {
+      "epoch": 0.1545138888888889,
+      "grad_norm": 0.6533085703849792,
+      "learning_rate": 9.941937440608951e-06,
+      "loss": 0.0712,
+      "step": 356
+    },
+    {
+      "epoch": 0.15494791666666666,
+      "grad_norm": 0.6614604592323303,
+      "learning_rate": 9.941402118901743e-06,
+      "loss": 0.0603,
+      "step": 357
+    },
+    {
+      "epoch": 0.15538194444444445,
+      "grad_norm": 1.1059565544128418,
+      "learning_rate": 9.940864355286999e-06,
+      "loss": 0.0697,
+      "step": 358
+    },
+    {
+      "epoch": 0.1558159722222222,
+      "grad_norm": 0.7619580626487732,
+      "learning_rate": 9.940324150030468e-06,
+      "loss": 0.0615,
+      "step": 359
+    },
+    {
+      "epoch": 0.15625,
+      "grad_norm": 1.6438034772872925,
+      "learning_rate": 9.9397815033991e-06,
+      "loss": 0.0975,
+      "step": 360
+    },
+    {
+      "epoch": 0.1566840277777778,
+      "grad_norm": 0.9065143465995789,
+      "learning_rate": 9.939236415661063e-06,
+      "loss": 0.0597,
+      "step": 361
+    },
+    {
+      "epoch": 0.15711805555555555,
+      "grad_norm": 0.8576701283454895,
+      "learning_rate": 9.938688887085717e-06,
+      "loss": 0.0656,
+      "step": 362
+    },
+    {
+      "epoch": 0.15755208333333334,
+      "grad_norm": 0.8027536273002625,
+      "learning_rate": 9.938138917943643e-06,
+      "loss": 0.0557,
+      "step": 363
+    },
+    {
+      "epoch": 0.1579861111111111,
+      "grad_norm": 0.8124869465827942,
+      "learning_rate": 9.937586508506617e-06,
+      "loss": 0.0781,
+      "step": 364
+    },
+    {
+      "epoch": 0.1584201388888889,
+      "grad_norm": 0.7721472382545471,
+      "learning_rate": 9.937031659047627e-06,
+      "loss": 0.0674,
+      "step": 365
+    },
+    {
+      "epoch": 0.15885416666666666,
+      "grad_norm": 0.7477177381515503,
+      "learning_rate": 9.936474369840862e-06,
+      "loss": 0.0634,
+      "step": 366
+    },
+    {
+      "epoch": 0.15928819444444445,
+      "grad_norm": 1.119555115699768,
+      "learning_rate": 9.93591464116172e-06,
+      "loss": 0.0763,
+      "step": 367
+    },
+    {
+      "epoch": 0.1597222222222222,
+      "grad_norm": 1.1497366428375244,
+      "learning_rate": 9.935352473286808e-06,
+      "loss": 0.0625,
+      "step": 368
+    },
+    {
+      "epoch": 0.16015625,
+      "grad_norm": 1.4596697092056274,
+      "learning_rate": 9.934787866493929e-06,
+      "loss": 0.09,
+      "step": 369
+    },
+    {
+      "epoch": 0.1605902777777778,
+      "grad_norm": 0.7065611481666565,
+      "learning_rate": 9.934220821062097e-06,
+      "loss": 0.0558,
+      "step": 370
+    },
+    {
+      "epoch": 0.16102430555555555,
+      "grad_norm": 0.850710391998291,
+      "learning_rate": 9.933651337271532e-06,
+      "loss": 0.0661,
+      "step": 371
+    },
+    {
+      "epoch": 0.16145833333333334,
+      "grad_norm": 0.637713611125946,
+      "learning_rate": 9.933079415403657e-06,
+      "loss": 0.0591,
+      "step": 372
+    },
+    {
+      "epoch": 0.1618923611111111,
+      "grad_norm": 0.6183542609214783,
+      "learning_rate": 9.932505055741103e-06,
+      "loss": 0.0535,
+      "step": 373
+    },
+    {
+      "epoch": 0.1623263888888889,
+      "grad_norm": 0.5929973125457764,
+      "learning_rate": 9.931928258567697e-06,
+      "loss": 0.0536,
+      "step": 374
+    },
+    {
+      "epoch": 0.16276041666666666,
+      "grad_norm": 0.8810651302337646,
+      "learning_rate": 9.931349024168483e-06,
+      "loss": 0.0658,
+      "step": 375
+    },
+    {
+      "epoch": 0.16319444444444445,
+      "grad_norm": 1.0665572881698608,
+      "learning_rate": 9.930767352829698e-06,
+      "loss": 0.0669,
+      "step": 376
+    },
+    {
+      "epoch": 0.1636284722222222,
+      "grad_norm": 0.6635595560073853,
+      "learning_rate": 9.930183244838792e-06,
+      "loss": 0.0575,
+      "step": 377
+    },
+    {
+      "epoch": 0.1640625,
+      "grad_norm": 1.1162075996398926,
+      "learning_rate": 9.929596700484412e-06,
+      "loss": 0.0806,
+      "step": 378
+    },
+    {
+      "epoch": 0.1644965277777778,
+      "grad_norm": 0.7197816371917725,
+      "learning_rate": 9.929007720056417e-06,
+      "loss": 0.0418,
+      "step": 379
+    },
+    {
+      "epoch": 0.16493055555555555,
+      "grad_norm": 1.1181401014328003,
+      "learning_rate": 9.92841630384586e-06,
+      "loss": 0.0798,
+      "step": 380
+    },
+    {
+      "epoch": 0.16536458333333334,
+      "grad_norm": 0.9076238870620728,
+      "learning_rate": 9.927822452145007e-06,
+      "loss": 0.0785,
+      "step": 381
+    },
+    {
+      "epoch": 0.1657986111111111,
+      "grad_norm": 0.8867773413658142,
+      "learning_rate": 9.92722616524732e-06,
+      "loss": 0.0604,
+      "step": 382
+    },
+    {
+      "epoch": 0.1662326388888889,
+      "grad_norm": 0.8934979438781738,
+      "learning_rate": 9.926627443447472e-06,
+      "loss": 0.0479,
+      "step": 383
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 0.801541268825531,
+      "learning_rate": 9.926026287041331e-06,
+      "loss": 0.0488,
+      "step": 384
+    },
+    {
+      "epoch": 0.16710069444444445,
+      "grad_norm": 0.7244776487350464,
+      "learning_rate": 9.925422696325976e-06,
+      "loss": 0.0615,
+      "step": 385
+    },
+    {
+      "epoch": 0.1675347222222222,
+      "grad_norm": 1.1914883852005005,
+      "learning_rate": 9.924816671599682e-06,
+      "loss": 0.0755,
+      "step": 386
+    },
+    {
+      "epoch": 0.16796875,
+      "grad_norm": 1.536291480064392,
+      "learning_rate": 9.924208213161932e-06,
+      "loss": 0.0713,
+      "step": 387
+    },
+    {
+      "epoch": 0.1684027777777778,
+      "grad_norm": 1.358291506767273,
+      "learning_rate": 9.92359732131341e-06,
+      "loss": 0.0888,
+      "step": 388
+    },
+    {
+      "epoch": 0.16883680555555555,
+      "grad_norm": 0.885543942451477,
+      "learning_rate": 9.922983996355999e-06,
+      "loss": 0.0556,
+      "step": 389
+    },
+    {
+      "epoch": 0.16927083333333334,
+      "grad_norm": 0.765248715877533,
+      "learning_rate": 9.922368238592792e-06,
+      "loss": 0.0585,
+      "step": 390
+    },
+    {
+      "epoch": 0.1697048611111111,
+      "grad_norm": 0.8303397297859192,
+      "learning_rate": 9.921750048328078e-06,
+      "loss": 0.0751,
+      "step": 391
+    },
+    {
+      "epoch": 0.1701388888888889,
+      "grad_norm": 0.8853304386138916,
+      "learning_rate": 9.92112942586735e-06,
+      "loss": 0.0598,
+      "step": 392
+    },
+    {
+      "epoch": 0.17057291666666666,
+      "grad_norm": 0.7674742341041565,
+      "learning_rate": 9.920506371517303e-06,
+      "loss": 0.0663,
+      "step": 393
+    },
+    {
+      "epoch": 0.17100694444444445,
+      "grad_norm": 0.9201746582984924,
+      "learning_rate": 9.919880885585833e-06,
+      "loss": 0.0674,
+      "step": 394
+    },
+    {
+      "epoch": 0.1714409722222222,
+      "grad_norm": 0.9129164814949036,
+      "learning_rate": 9.919252968382039e-06,
+      "loss": 0.0605,
+      "step": 395
+    },
+    {
+      "epoch": 0.171875,
+      "grad_norm": 0.8458436727523804,
+      "learning_rate": 9.918622620216218e-06,
+      "loss": 0.0425,
+      "step": 396
+    },
+    {
+      "epoch": 0.1723090277777778,
+      "grad_norm": 0.9405080080032349,
+      "learning_rate": 9.917989841399875e-06,
+      "loss": 0.0839,
+      "step": 397
+    },
+    {
+      "epoch": 0.17274305555555555,
+      "grad_norm": 0.874566376209259,
+      "learning_rate": 9.917354632245712e-06,
+      "loss": 0.0624,
+      "step": 398
+    },
+    {
+      "epoch": 0.17317708333333334,
+      "grad_norm": 0.4679570198059082,
+      "learning_rate": 9.91671699306763e-06,
+      "loss": 0.035,
+      "step": 399
+    },
+    {
+      "epoch": 0.1736111111111111,
+      "grad_norm": 0.9329303503036499,
+      "learning_rate": 9.916076924180734e-06,
+      "loss": 0.0934,
+      "step": 400
+    },
+    {
+      "epoch": 0.1740451388888889,
+      "grad_norm": 0.6457216143608093,
+      "learning_rate": 9.915434425901328e-06,
+      "loss": 0.0524,
+      "step": 401
+    },
+    {
+      "epoch": 0.17447916666666666,
+      "grad_norm": 0.5407356023788452,
+      "learning_rate": 9.91478949854692e-06,
+      "loss": 0.053,
+      "step": 402
+    },
+    {
+      "epoch": 0.17491319444444445,
+      "grad_norm": 0.6888991594314575,
+      "learning_rate": 9.914142142436212e-06,
+      "loss": 0.0707,
+      "step": 403
+    },
+    {
+      "epoch": 0.1753472222222222,
+      "grad_norm": 0.6760328412055969,
+      "learning_rate": 9.913492357889112e-06,
+      "loss": 0.0853,
+      "step": 404
+    },
+    {
+      "epoch": 0.17578125,
+      "grad_norm": 0.6837098598480225,
+      "learning_rate": 9.912840145226725e-06,
+      "loss": 0.0515,
+      "step": 405
+    },
+    {
+      "epoch": 0.1762152777777778,
+      "grad_norm": 0.7048181295394897,
+      "learning_rate": 9.912185504771359e-06,
+      "loss": 0.0474,
+      "step": 406
+    },
+    {
+      "epoch": 0.17664930555555555,
+      "grad_norm": 1.214665412902832,
+      "learning_rate": 9.911528436846519e-06,
+      "loss": 0.0874,
+      "step": 407
+    },
+    {
+      "epoch": 0.17708333333333334,
+      "grad_norm": 0.8326371908187866,
+      "learning_rate": 9.910868941776906e-06,
+      "loss": 0.0717,
+      "step": 408
+    },
+    {
+      "epoch": 0.1775173611111111,
+      "grad_norm": 0.818688690662384,
+      "learning_rate": 9.91020701988843e-06,
+      "loss": 0.0906,
+      "step": 409
+    },
+    {
+      "epoch": 0.1779513888888889,
+      "grad_norm": 1.0963165760040283,
+      "learning_rate": 9.909542671508192e-06,
+      "loss": 0.0805,
+      "step": 410
+    },
+    {
+      "epoch": 0.17838541666666666,
+      "grad_norm": 0.6087559461593628,
+      "learning_rate": 9.908875896964497e-06,
+      "loss": 0.0529,
+      "step": 411
+    },
+    {
+      "epoch": 0.17881944444444445,
+      "grad_norm": 0.5906699895858765,
+      "learning_rate": 9.908206696586846e-06,
+      "loss": 0.0787,
+      "step": 412
+    },
+    {
+      "epoch": 0.1792534722222222,
+      "grad_norm": 0.5716590285301208,
+      "learning_rate": 9.90753507070594e-06,
+      "loss": 0.0671,
+      "step": 413
+    },
+    {
+      "epoch": 0.1796875,
+      "grad_norm": 0.7733680605888367,
+      "learning_rate": 9.906861019653677e-06,
+      "loss": 0.0495,
+      "step": 414
+    },
+    {
+      "epoch": 0.1801215277777778,
+      "grad_norm": 0.8408774733543396,
+      "learning_rate": 9.906184543763158e-06,
+      "loss": 0.0848,
+      "step": 415
+    },
+    {
+      "epoch": 0.18055555555555555,
+      "grad_norm": 1.0400339365005493,
+      "learning_rate": 9.905505643368675e-06,
+      "loss": 0.0825,
+      "step": 416
+    },
+    {
+      "epoch": 0.18098958333333334,
+      "grad_norm": 0.5606048107147217,
+      "learning_rate": 9.904824318805725e-06,
+      "loss": 0.0548,
+      "step": 417
+    },
+    {
+      "epoch": 0.1814236111111111,
+      "grad_norm": 0.7534106373786926,
+      "learning_rate": 9.904140570410997e-06,
+      "loss": 0.0719,
+      "step": 418
+    },
+    {
+      "epoch": 0.1818576388888889,
+      "grad_norm": 0.6680476069450378,
+      "learning_rate": 9.903454398522388e-06,
+      "loss": 0.0634,
+      "step": 419
+    },
+    {
+      "epoch": 0.18229166666666666,
+      "grad_norm": 0.5825947523117065,
+      "learning_rate": 9.90276580347898e-06,
+      "loss": 0.0447,
+      "step": 420
+    },
+    {
+      "epoch": 0.18272569444444445,
+      "grad_norm": 0.7220215201377869,
+      "learning_rate": 9.902074785621058e-06,
+      "loss": 0.0438,
+      "step": 421
+    },
+    {
+      "epoch": 0.1831597222222222,
+      "grad_norm": 1.1537821292877197,
+      "learning_rate": 9.901381345290107e-06,
+      "loss": 0.0707,
+      "step": 422
+    },
+    {
+      "epoch": 0.18359375,
+      "grad_norm": 0.9954478740692139,
+      "learning_rate": 9.900685482828803e-06,
+      "loss": 0.0788,
+      "step": 423
+    },
+    {
+      "epoch": 0.1840277777777778,
+      "grad_norm": 1.56073796749115,
+      "learning_rate": 9.899987198581028e-06,
+      "loss": 0.0803,
+      "step": 424
+    },
+    {
+      "epoch": 0.18446180555555555,
+      "grad_norm": 0.9214962720870972,
+      "learning_rate": 9.899286492891849e-06,
+      "loss": 0.0592,
+      "step": 425
+    },
+    {
+      "epoch": 0.18489583333333334,
+      "grad_norm": 0.5767590403556824,
+      "learning_rate": 9.898583366107539e-06,
+      "loss": 0.0487,
+      "step": 426
+    },
+    {
+      "epoch": 0.1853298611111111,
+      "grad_norm": 0.8960171937942505,
+      "learning_rate": 9.897877818575563e-06,
+      "loss": 0.0545,
+      "step": 427
+    },
+    {
+      "epoch": 0.1857638888888889,
+      "grad_norm": 0.9727705121040344,
+      "learning_rate": 9.897169850644585e-06,
+      "loss": 0.0686,
+      "step": 428
+    },
+    {
+      "epoch": 0.18619791666666666,
+      "grad_norm": 1.2093175649642944,
+      "learning_rate": 9.896459462664465e-06,
+      "loss": 0.0697,
+      "step": 429
+    },
+    {
+      "epoch": 0.18663194444444445,
+      "grad_norm": 0.9476456046104431,
+      "learning_rate": 9.895746654986253e-06,
+      "loss": 0.0638,
+      "step": 430
+    },
+    {
+      "epoch": 0.1870659722222222,
+      "grad_norm": 0.5582579970359802,
+      "learning_rate": 9.895031427962205e-06,
+      "loss": 0.0528,
+      "step": 431
+    },
+    {
+      "epoch": 0.1875,
+      "grad_norm": 0.9498739838600159,
+      "learning_rate": 9.894313781945762e-06,
+      "loss": 0.0452,
+      "step": 432
+    },
+    {
+      "epoch": 0.1879340277777778,
+      "grad_norm": 2.0486114025115967,
+      "learning_rate": 9.893593717291568e-06,
+      "loss": 0.0941,
+      "step": 433
+    },
+    {
+      "epoch": 0.18836805555555555,
+      "grad_norm": 1.371767520904541,
+      "learning_rate": 9.892871234355457e-06,
+      "loss": 0.0568,
+      "step": 434
+    },
+    {
+      "epoch": 0.18880208333333334,
+      "grad_norm": 1.096907615661621,
+      "learning_rate": 9.892146333494463e-06,
+      "loss": 0.0472,
+      "step": 435
+    },
+    {
+      "epoch": 0.1892361111111111,
+      "grad_norm": 1.0589361190795898,
+      "learning_rate": 9.891419015066812e-06,
+      "loss": 0.0644,
+      "step": 436
+    },
+    {
+      "epoch": 0.1896701388888889,
+      "grad_norm": 0.9310447573661804,
+      "learning_rate": 9.890689279431924e-06,
+      "loss": 0.0709,
+      "step": 437
+    },
+    {
+      "epoch": 0.19010416666666666,
+      "grad_norm": 0.8208966851234436,
+      "learning_rate": 9.889957126950416e-06,
+      "loss": 0.0561,
+      "step": 438
+    },
+    {
+      "epoch": 0.19053819444444445,
+      "grad_norm": 1.5516587495803833,
+      "learning_rate": 9.889222557984098e-06,
+      "loss": 0.0617,
+      "step": 439
+    },
+    {
+      "epoch": 0.1909722222222222,
+      "grad_norm": 1.1894527673721313,
+      "learning_rate": 9.888485572895971e-06,
+      "loss": 0.0771,
+      "step": 440
+    },
+    {
+      "epoch": 0.19140625,
+      "grad_norm": 1.1664812564849854,
+      "learning_rate": 9.88774617205024e-06,
+      "loss": 0.0504,
+      "step": 441
+    },
+    {
+      "epoch": 0.1918402777777778,
+      "grad_norm": 0.7609825730323792,
+      "learning_rate": 9.88700435581229e-06,
+      "loss": 0.0606,
+      "step": 442
+    },
+    {
+      "epoch": 0.19227430555555555,
+      "grad_norm": 0.94313645362854,
+      "learning_rate": 9.88626012454871e-06,
+      "loss": 0.0642,
+      "step": 443
+    },
+    {
+      "epoch": 0.19270833333333334,
+      "grad_norm": 0.6481614112854004,
+      "learning_rate": 9.885513478627277e-06,
+      "loss": 0.0571,
+      "step": 444
+    },
+    {
+      "epoch": 0.1931423611111111,
+      "grad_norm": 0.8185029029846191,
+      "learning_rate": 9.884764418416965e-06,
+      "loss": 0.0529,
+      "step": 445
+    },
+    {
+      "epoch": 0.1935763888888889,
+      "grad_norm": 0.844165563583374,
+      "learning_rate": 9.884012944287939e-06,
+      "loss": 0.0875,
+      "step": 446
+    },
+    {
+      "epoch": 0.19401041666666666,
+      "grad_norm": 0.7095887660980225,
+      "learning_rate": 9.883259056611559e-06,
+      "loss": 0.0592,
+      "step": 447
+    },
+    {
+      "epoch": 0.19444444444444445,
+      "grad_norm": 0.6745681762695312,
+      "learning_rate": 9.882502755760371e-06,
+      "loss": 0.0615,
+      "step": 448
+    },
+    {
+      "epoch": 0.1948784722222222,
+      "grad_norm": 0.7381845116615295,
+      "learning_rate": 9.881744042108123e-06,
+      "loss": 0.0607,
+      "step": 449
+    },
+    {
+      "epoch": 0.1953125,
+      "grad_norm": 0.8988702893257141,
+      "learning_rate": 9.880982916029749e-06,
+      "loss": 0.0835,
+      "step": 450
+    },
+    {
+      "epoch": 0.1957465277777778,
+      "grad_norm": 0.8817563056945801,
+      "learning_rate": 9.880219377901377e-06,
+      "loss": 0.1022,
+      "step": 451
+    },
+    {
+      "epoch": 0.19618055555555555,
+      "grad_norm": 0.8826650381088257,
+      "learning_rate": 9.879453428100327e-06,
+      "loss": 0.0735,
+      "step": 452
+    },
+    {
+      "epoch": 0.19661458333333334,
+      "grad_norm": 0.6274738907814026,
+      "learning_rate": 9.87868506700511e-06,
+      "loss": 0.0611,
+      "step": 453
+    },
+    {
+      "epoch": 0.1970486111111111,
+      "grad_norm": 0.752539336681366,
+      "learning_rate": 9.877914294995432e-06,
+      "loss": 0.0621,
+      "step": 454
+    },
+    {
+      "epoch": 0.1974826388888889,
+      "grad_norm": 0.9866121411323547,
+      "learning_rate": 9.877141112452185e-06,
+      "loss": 0.0611,
+      "step": 455
+    },
+    {
+      "epoch": 0.19791666666666666,
+      "grad_norm": 0.6701814532279968,
+      "learning_rate": 9.876365519757457e-06,
+      "loss": 0.0699,
+      "step": 456
+    },
+    {
+      "epoch": 0.19835069444444445,
+      "grad_norm": 1.2707716226577759,
+      "learning_rate": 9.875587517294522e-06,
+      "loss": 0.0788,
+      "step": 457
+    },
+    {
+      "epoch": 0.1987847222222222,
+      "grad_norm": 1.0906691551208496,
+      "learning_rate": 9.87480710544785e-06,
+      "loss": 0.0987,
+      "step": 458
+    },
+    {
+      "epoch": 0.19921875,
+      "grad_norm": 1.0719773769378662,
+      "learning_rate": 9.874024284603099e-06,
+      "loss": 0.0619,
+      "step": 459
+    },
+    {
+      "epoch": 0.1996527777777778,
+      "grad_norm": 0.8737601041793823,
+      "learning_rate": 9.873239055147119e-06,
+      "loss": 0.0734,
+      "step": 460
+    },
+    {
+      "epoch": 0.20008680555555555,
+      "grad_norm": 0.6847056150436401,
+      "learning_rate": 9.872451417467947e-06,
+      "loss": 0.0614,
+      "step": 461
+    },
+    {
+      "epoch": 0.20052083333333334,
+      "grad_norm": 0.7287288308143616,
+      "learning_rate": 9.871661371954814e-06,
+      "loss": 0.0705,
+      "step": 462
+    },
+    {
+      "epoch": 0.2009548611111111,
+      "grad_norm": 0.7544885873794556,
+      "learning_rate": 9.87086891899814e-06,
+      "loss": 0.0804,
+      "step": 463
+    },
+    {
+      "epoch": 0.2013888888888889,
+      "grad_norm": 0.6416462063789368,
+      "learning_rate": 9.870074058989532e-06,
+      "loss": 0.0696,
+      "step": 464
+    },
+    {
+      "epoch": 0.20182291666666666,
+      "grad_norm": 0.6606134176254272,
+      "learning_rate": 9.869276792321786e-06,
+      "loss": 0.0498,
+      "step": 465
+    },
+    {
+      "epoch": 0.20225694444444445,
+      "grad_norm": 0.7388238906860352,
+      "learning_rate": 9.868477119388897e-06,
+      "loss": 0.0707,
+      "step": 466
+    },
+    {
+      "epoch": 0.2026909722222222,
+      "grad_norm": 0.8424966335296631,
+      "learning_rate": 9.867675040586035e-06,
+      "loss": 0.0956,
+      "step": 467
+    },
+    {
+      "epoch": 0.203125,
+      "grad_norm": 0.7446620464324951,
+      "learning_rate": 9.866870556309568e-06,
+      "loss": 0.0682,
+      "step": 468
+    },
+    {
+      "epoch": 0.2035590277777778,
+      "grad_norm": 0.6781134009361267,
+      "learning_rate": 9.86606366695705e-06,
+      "loss": 0.0595,
+      "step": 469
+    },
+    {
+      "epoch": 0.20399305555555555,
+      "grad_norm": 0.5591951012611389,
+      "learning_rate": 9.865254372927224e-06,
+      "loss": 0.0516,
+      "step": 470
+    },
+    {
+      "epoch": 0.20442708333333334,
+      "grad_norm": 0.81165611743927,
+      "learning_rate": 9.864442674620023e-06,
+      "loss": 0.0763,
+      "step": 471
+    },
+    {
+      "epoch": 0.2048611111111111,
+      "grad_norm": 0.6503605246543884,
+      "learning_rate": 9.863628572436565e-06,
+      "loss": 0.0426,
+      "step": 472
+    },
+    {
+      "epoch": 0.2052951388888889,
+      "grad_norm": 0.7912452220916748,
+      "learning_rate": 9.862812066779157e-06,
+      "loss": 0.0745,
+      "step": 473
+    },
+    {
+      "epoch": 0.20572916666666666,
+      "grad_norm": 0.590604841709137,
+      "learning_rate": 9.861993158051295e-06,
+      "loss": 0.0609,
+      "step": 474
+    },
+    {
+      "epoch": 0.20616319444444445,
+      "grad_norm": 1.0321898460388184,
+      "learning_rate": 9.86117184665766e-06,
+      "loss": 0.0697,
+      "step": 475
+    },
+    {
+      "epoch": 0.2065972222222222,
+      "grad_norm": 1.022867202758789,
+      "learning_rate": 9.860348133004123e-06,
+      "loss": 0.0532,
+      "step": 476
+    },
+    {
+      "epoch": 0.20703125,
+      "grad_norm": 0.6199460029602051,
+      "learning_rate": 9.859522017497742e-06,
+      "loss": 0.0623,
+      "step": 477
+    },
+    {
+      "epoch": 0.2074652777777778,
+      "grad_norm": 1.0487356185913086,
+      "learning_rate": 9.858693500546758e-06,
+      "loss": 0.0826,
+      "step": 478
+    },
+    {
+      "epoch": 0.20789930555555555,
+      "grad_norm": 0.42486223578453064,
+      "learning_rate": 9.857862582560606e-06,
+      "loss": 0.0333,
+      "step": 479
+    },
+    {
+      "epoch": 0.20833333333333334,
+      "grad_norm": 1.3734253644943237,
+      "learning_rate": 9.857029263949898e-06,
+      "loss": 0.1035,
+      "step": 480
+    },
+    {
+      "epoch": 0.2087673611111111,
+      "grad_norm": 0.8876743912696838,
+      "learning_rate": 9.856193545126444e-06,
+      "loss": 0.0786,
+      "step": 481
+    },
+    {
+      "epoch": 0.2092013888888889,
+      "grad_norm": 0.6525356769561768,
+      "learning_rate": 9.855355426503228e-06,
+      "loss": 0.0602,
+      "step": 482
+    },
+    {
+      "epoch": 0.20963541666666666,
+      "grad_norm": 1.0160382986068726,
+      "learning_rate": 9.854514908494427e-06,
+      "loss": 0.0816,
+      "step": 483
+    },
+    {
+      "epoch": 0.21006944444444445,
+      "grad_norm": 1.0310873985290527,
+      "learning_rate": 9.853671991515405e-06,
+      "loss": 0.0613,
+      "step": 484
+    },
+    {
+      "epoch": 0.2105034722222222,
+      "grad_norm": 0.7252808809280396,
+      "learning_rate": 9.852826675982705e-06,
+      "loss": 0.0674,
+      "step": 485
+    },
+    {
+      "epoch": 0.2109375,
+      "grad_norm": 0.7665201425552368,
+      "learning_rate": 9.851978962314063e-06,
+      "loss": 0.0509,
+      "step": 486
+    },
+    {
+      "epoch": 0.2113715277777778,
+      "grad_norm": 0.6184194684028625,
+      "learning_rate": 9.851128850928394e-06,
+      "loss": 0.0782,
+      "step": 487
+    },
+    {
+      "epoch": 0.21180555555555555,
+      "grad_norm": 0.6223265528678894,
+      "learning_rate": 9.850276342245797e-06,
+      "loss": 0.0454,
+      "step": 488
+    },
+    {
+      "epoch": 0.21223958333333334,
+      "grad_norm": 0.7626925110816956,
+      "learning_rate": 9.849421436687565e-06,
+      "loss": 0.0725,
+      "step": 489
+    },
+    {
+      "epoch": 0.2126736111111111,
+      "grad_norm": 0.8450047373771667,
+      "learning_rate": 9.848564134676166e-06,
+      "loss": 0.0682,
+      "step": 490
+    },
+    {
+      "epoch": 0.2131076388888889,
+      "grad_norm": 0.7744449377059937,
+      "learning_rate": 9.847704436635255e-06,
+      "loss": 0.0645,
+      "step": 491
+    },
+    {
+      "epoch": 0.21354166666666666,
+      "grad_norm": 1.0327328443527222,
+      "learning_rate": 9.846842342989675e-06,
+      "loss": 0.0837,
+      "step": 492
+    },
+    {
+      "epoch": 0.21397569444444445,
+      "grad_norm": 0.8731493949890137,
+      "learning_rate": 9.845977854165443e-06,
+      "loss": 0.0634,
+      "step": 493
+    },
+    {
+      "epoch": 0.2144097222222222,
+      "grad_norm": 0.5188009738922119,
+      "learning_rate": 9.845110970589773e-06,
+      "loss": 0.0391,
+      "step": 494
+    },
+    {
+      "epoch": 0.21484375,
+      "grad_norm": 0.7408533692359924,
+      "learning_rate": 9.84424169269105e-06,
+      "loss": 0.0556,
+      "step": 495
+    },
+    {
+      "epoch": 0.2152777777777778,
+      "grad_norm": 0.6349946856498718,
+      "learning_rate": 9.843370020898852e-06,
+      "loss": 0.0576,
+      "step": 496
+    },
+    {
+      "epoch": 0.21571180555555555,
+      "grad_norm": 0.8657709956169128,
+      "learning_rate": 9.842495955643932e-06,
+      "loss": 0.0552,
+      "step": 497
+    },
+    {
+      "epoch": 0.21614583333333334,
+      "grad_norm": 0.7064958214759827,
+      "learning_rate": 9.841619497358232e-06,
+      "loss": 0.0662,
+      "step": 498
+    },
+    {
+      "epoch": 0.2165798611111111,
+      "grad_norm": 0.9118136763572693,
+      "learning_rate": 9.840740646474873e-06,
+      "loss": 0.054,
+      "step": 499
+    },
+    {
+      "epoch": 0.2170138888888889,
+      "grad_norm": 0.577598512172699,
+      "learning_rate": 9.839859403428159e-06,
+      "loss": 0.0557,
+      "step": 500
+    },
+    {
+      "epoch": 0.21744791666666666,
+      "grad_norm": 0.7629770040512085,
+      "learning_rate": 9.838975768653579e-06,
+      "loss": 0.0657,
+      "step": 501
+    },
+    {
+      "epoch": 0.21788194444444445,
+      "grad_norm": 0.7026235461235046,
+      "learning_rate": 9.838089742587796e-06,
+      "loss": 0.0614,
+      "step": 502
+    },
+    {
+      "epoch": 0.2183159722222222,
+      "grad_norm": 0.8833339214324951,
+      "learning_rate": 9.837201325668667e-06,
+      "loss": 0.0532,
+      "step": 503
+    },
+    {
+      "epoch": 0.21875,
+      "grad_norm": 0.8416234254837036,
+      "learning_rate": 9.836310518335219e-06,
+      "loss": 0.0568,
+      "step": 504
+    },
+    {
+      "epoch": 0.2191840277777778,
+      "grad_norm": 1.0332963466644287,
+      "learning_rate": 9.835417321027664e-06,
+      "loss": 0.0687,
+      "step": 505
+    },
+    {
+      "epoch": 0.21961805555555555,
+      "grad_norm": 0.796916127204895,
+      "learning_rate": 9.8345217341874e-06,
+      "loss": 0.0631,
+      "step": 506
+    },
+    {
+      "epoch": 0.22005208333333334,
+      "grad_norm": 0.6922689080238342,
+      "learning_rate": 9.833623758257003e-06,
+      "loss": 0.0611,
+      "step": 507
+    },
+    {
+      "epoch": 0.2204861111111111,
+      "grad_norm": 0.9117850661277771,
+      "learning_rate": 9.832723393680222e-06,
+      "loss": 0.0769,
+      "step": 508
+    },
+    {
+      "epoch": 0.2209201388888889,
+      "grad_norm": 1.138522982597351,
+      "learning_rate": 9.831820640901997e-06,
+      "loss": 0.0677,
+      "step": 509
+    },
+    {
+      "epoch": 0.22135416666666666,
+      "grad_norm": 0.6167424917221069,
+      "learning_rate": 9.830915500368444e-06,
+      "loss": 0.0613,
+      "step": 510
+    },
+    {
+      "epoch": 0.22178819444444445,
+      "grad_norm": 0.8744767904281616,
+      "learning_rate": 9.830007972526858e-06,
+      "loss": 0.0858,
+      "step": 511
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.7937278151512146,
+      "learning_rate": 9.829098057825716e-06,
+      "loss": 0.0742,
+      "step": 512
+    },
+    {
+      "epoch": 0.22265625,
+      "grad_norm": 0.8917018175125122,
+      "learning_rate": 9.828185756714672e-06,
+      "loss": 0.0886,
+      "step": 513
+    },
+    {
+      "epoch": 0.2230902777777778,
+      "grad_norm": 0.6010713577270508,
+      "learning_rate": 9.827271069644562e-06,
+      "loss": 0.046,
+      "step": 514
+    },
+    {
+      "epoch": 0.22352430555555555,
+      "grad_norm": 0.7526761293411255,
+      "learning_rate": 9.826353997067398e-06,
+      "loss": 0.0556,
+      "step": 515
+    },
+    {
+      "epoch": 0.22395833333333334,
+      "grad_norm": 0.6621930599212646,
+      "learning_rate": 9.825434539436374e-06,
+      "loss": 0.0491,
+      "step": 516
+    },
+    {
+      "epoch": 0.2243923611111111,
+      "grad_norm": 1.0901848077774048,
+      "learning_rate": 9.82451269720586e-06,
+      "loss": 0.0706,
+      "step": 517
+    },
+    {
+      "epoch": 0.2248263888888889,
+      "grad_norm": 1.0902998447418213,
+      "learning_rate": 9.823588470831405e-06,
+      "loss": 0.0616,
+      "step": 518
+    },
+    {
+      "epoch": 0.22526041666666666,
+      "grad_norm": 1.2565516233444214,
+      "learning_rate": 9.822661860769738e-06,
+      "loss": 0.0952,
+      "step": 519
+    },
+    {
+      "epoch": 0.22569444444444445,
+      "grad_norm": 1.2477786540985107,
+      "learning_rate": 9.821732867478764e-06,
+      "loss": 0.0915,
+      "step": 520
+    },
+    {
+      "epoch": 0.2261284722222222,
+      "grad_norm": 1.1893221139907837,
+      "learning_rate": 9.820801491417569e-06,
+      "loss": 0.0878,
+      "step": 521
+    },
+    {
+      "epoch": 0.2265625,
+      "grad_norm": 0.8113741874694824,
+      "learning_rate": 9.81986773304641e-06,
+      "loss": 0.0625,
+      "step": 522
+    },
+    {
+      "epoch": 0.2269965277777778,
+      "grad_norm": 0.84547358751297,
+      "learning_rate": 9.818931592826727e-06,
+      "loss": 0.0617,
+      "step": 523
+    },
+    {
+      "epoch": 0.22743055555555555,
+      "grad_norm": 0.6920998692512512,
+      "learning_rate": 9.817993071221135e-06,
+      "loss": 0.0717,
+      "step": 524
+    },
+    {
+      "epoch": 0.22786458333333334,
+      "grad_norm": 0.9595234990119934,
+      "learning_rate": 9.817052168693427e-06,
+      "loss": 0.0931,
+      "step": 525
+    },
+    {
+      "epoch": 0.2282986111111111,
+      "grad_norm": 0.5811662673950195,
+      "learning_rate": 9.81610888570857e-06,
+      "loss": 0.0681,
+      "step": 526
+    },
+    {
+      "epoch": 0.2287326388888889,
+      "grad_norm": 0.8212871551513672,
+      "learning_rate": 9.81516322273271e-06,
+      "loss": 0.08,
+      "step": 527
+    },
+    {
+      "epoch": 0.22916666666666666,
+      "grad_norm": 0.5696889758110046,
+      "learning_rate": 9.814215180233167e-06,
+      "loss": 0.0671,
+      "step": 528
+    },
+    {
+      "epoch": 0.22960069444444445,
+      "grad_norm": 0.5984359383583069,
+      "learning_rate": 9.81326475867844e-06,
+      "loss": 0.0694,
+      "step": 529
+    },
+    {
+      "epoch": 0.2300347222222222,
+      "grad_norm": 0.6086364388465881,
+      "learning_rate": 9.812311958538202e-06,
+      "loss": 0.0545,
+      "step": 530
+    },
+    {
+      "epoch": 0.23046875,
+      "grad_norm": 0.8695059418678284,
+      "learning_rate": 9.811356780283297e-06,
+      "loss": 0.085,
+      "step": 531
+    },
+    {
+      "epoch": 0.2309027777777778,
+      "grad_norm": 0.7423482537269592,
+      "learning_rate": 9.810399224385752e-06,
+      "loss": 0.0552,
+      "step": 532
+    },
+    {
+      "epoch": 0.23133680555555555,
+      "grad_norm": 1.1487404108047485,
+      "learning_rate": 9.809439291318764e-06,
+      "loss": 0.09,
+      "step": 533
+    },
+    {
+      "epoch": 0.23177083333333334,
+      "grad_norm": 0.7295951247215271,
+      "learning_rate": 9.808476981556706e-06,
+      "loss": 0.0361,
+      "step": 534
+    },
+    {
+      "epoch": 0.2322048611111111,
+      "grad_norm": 0.7190055251121521,
+      "learning_rate": 9.807512295575125e-06,
+      "loss": 0.0418,
+      "step": 535
+    },
+    {
+      "epoch": 0.2326388888888889,
+      "grad_norm": 1.2923922538757324,
+      "learning_rate": 9.806545233850746e-06,
+      "loss": 0.0802,
+      "step": 536
+    },
+    {
+      "epoch": 0.23307291666666666,
+      "grad_norm": 1.0366101264953613,
+      "learning_rate": 9.805575796861461e-06,
+      "loss": 0.0744,
+      "step": 537
+    },
+    {
+      "epoch": 0.23350694444444445,
+      "grad_norm": 0.6617826819419861,
+      "learning_rate": 9.80460398508634e-06,
+      "loss": 0.0483,
+      "step": 538
+    },
+    {
+      "epoch": 0.2339409722222222,
+      "grad_norm": 1.1316651105880737,
+      "learning_rate": 9.80362979900563e-06,
+      "loss": 0.0825,
+      "step": 539
+    },
+    {
+      "epoch": 0.234375,
+      "grad_norm": 0.773888349533081,
+      "learning_rate": 9.802653239100742e-06,
+      "loss": 0.0719,
+      "step": 540
+    },
+    {
+      "epoch": 0.2348090277777778,
+      "grad_norm": 0.8033248782157898,
+      "learning_rate": 9.801674305854268e-06,
+      "loss": 0.0775,
+      "step": 541
+    },
+    {
+      "epoch": 0.23524305555555555,
+      "grad_norm": 0.8055210113525391,
+      "learning_rate": 9.800692999749973e-06,
+      "loss": 0.0765,
+      "step": 542
+    },
+    {
+      "epoch": 0.23567708333333334,
+      "grad_norm": 0.6195559501647949,
+      "learning_rate": 9.799709321272787e-06,
+      "loss": 0.0603,
+      "step": 543
+    },
+    {
+      "epoch": 0.2361111111111111,
+      "grad_norm": 0.9171823263168335,
+      "learning_rate": 9.79872327090882e-06,
+      "loss": 0.0691,
+      "step": 544
+    },
+    {
+      "epoch": 0.2365451388888889,
+      "grad_norm": 0.5740596055984497,
+      "learning_rate": 9.797734849145352e-06,
+      "loss": 0.054,
+      "step": 545
+    },
+    {
+      "epoch": 0.23697916666666666,
+      "grad_norm": 0.632985532283783,
+      "learning_rate": 9.796744056470832e-06,
+      "loss": 0.0605,
+      "step": 546
+    },
+    {
+      "epoch": 0.23741319444444445,
+      "grad_norm": 0.6397607326507568,
+      "learning_rate": 9.795750893374887e-06,
+      "loss": 0.0451,
+      "step": 547
+    },
+    {
+      "epoch": 0.2378472222222222,
+      "grad_norm": 0.7479575276374817,
+      "learning_rate": 9.794755360348309e-06,
+      "loss": 0.0541,
+      "step": 548
+    },
+    {
+      "epoch": 0.23828125,
+      "grad_norm": 1.0319230556488037,
+      "learning_rate": 9.793757457883062e-06,
+      "loss": 0.0708,
+      "step": 549
+    },
+    {
+      "epoch": 0.2387152777777778,
+      "grad_norm": 1.3229392766952515,
+      "learning_rate": 9.792757186472285e-06,
+      "loss": 0.0931,
+      "step": 550
+    },
+    {
+      "epoch": 0.23914930555555555,
+      "grad_norm": 1.2721326351165771,
+      "learning_rate": 9.791754546610284e-06,
+      "loss": 0.0785,
+      "step": 551
+    },
+    {
+      "epoch": 0.23958333333333334,
+      "grad_norm": 0.6858370304107666,
+      "learning_rate": 9.790749538792537e-06,
+      "loss": 0.0555,
+      "step": 552
+    },
+    {
+      "epoch": 0.2400173611111111,
+      "grad_norm": 0.7393696904182434,
+      "learning_rate": 9.789742163515689e-06,
+      "loss": 0.0775,
+      "step": 553
+    },
+    {
+      "epoch": 0.2404513888888889,
+      "grad_norm": 0.5606722831726074,
+      "learning_rate": 9.788732421277563e-06,
+      "loss": 0.0438,
+      "step": 554
+    },
+    {
+      "epoch": 0.24088541666666666,
+      "grad_norm": 0.6640605330467224,
+      "learning_rate": 9.787720312577142e-06,
+      "loss": 0.0712,
+      "step": 555
+    },
+    {
+      "epoch": 0.24131944444444445,
+      "grad_norm": 0.8472843170166016,
+      "learning_rate": 9.786705837914584e-06,
+      "loss": 0.0681,
+      "step": 556
+    },
+    {
+      "epoch": 0.2417534722222222,
+      "grad_norm": 0.5820501446723938,
+      "learning_rate": 9.785688997791218e-06,
+      "loss": 0.045,
+      "step": 557
+    },
+    {
+      "epoch": 0.2421875,
+      "grad_norm": 0.7918657064437866,
+      "learning_rate": 9.784669792709534e-06,
+      "loss": 0.08,
+      "step": 558
+    },
+    {
+      "epoch": 0.2426215277777778,
+      "grad_norm": 0.6729964017868042,
+      "learning_rate": 9.783648223173197e-06,
+      "loss": 0.058,
+      "step": 559
+    },
+    {
+      "epoch": 0.24305555555555555,
+      "grad_norm": 0.6780051589012146,
+      "learning_rate": 9.78262428968704e-06,
+      "loss": 0.0592,
+      "step": 560
+    },
+    {
+      "epoch": 0.24348958333333334,
+      "grad_norm": 0.7747189402580261,
+      "learning_rate": 9.781597992757066e-06,
+      "loss": 0.0537,
+      "step": 561
+    },
+    {
+      "epoch": 0.2439236111111111,
+      "grad_norm": 1.1778326034545898,
+      "learning_rate": 9.78056933289044e-06,
+      "loss": 0.0714,
+      "step": 562
+    },
+    {
+      "epoch": 0.2443576388888889,
+      "grad_norm": 0.8010048270225525,
+      "learning_rate": 9.779538310595495e-06,
+      "loss": 0.0547,
+      "step": 563
+    },
+    {
+      "epoch": 0.24479166666666666,
+      "grad_norm": 1.0312997102737427,
+      "learning_rate": 9.77850492638174e-06,
+      "loss": 0.0834,
+      "step": 564
+    },
+    {
+      "epoch": 0.24522569444444445,
+      "grad_norm": 1.1206309795379639,
+      "learning_rate": 9.777469180759844e-06,
+      "loss": 0.0829,
+      "step": 565
+    },
+    {
+      "epoch": 0.2456597222222222,
+      "grad_norm": 0.7379752993583679,
+      "learning_rate": 9.776431074241643e-06,
+      "loss": 0.0557,
+      "step": 566
+    },
+    {
+      "epoch": 0.24609375,
+      "grad_norm": 0.6607487797737122,
+      "learning_rate": 9.775390607340142e-06,
+      "loss": 0.0535,
+      "step": 567
+    },
+    {
+      "epoch": 0.2465277777777778,
+      "grad_norm": 0.7935291528701782,
+      "learning_rate": 9.774347780569513e-06,
+      "loss": 0.0608,
+      "step": 568
+    },
+    {
+      "epoch": 0.24696180555555555,
+      "grad_norm": 0.7309617400169373,
+      "learning_rate": 9.77330259444509e-06,
+      "loss": 0.0704,
+      "step": 569
+    },
+    {
+      "epoch": 0.24739583333333334,
+      "grad_norm": 0.9455975294113159,
+      "learning_rate": 9.772255049483378e-06,
+      "loss": 0.0779,
+      "step": 570
+    },
+    {
+      "epoch": 0.2478298611111111,
+      "grad_norm": 0.716031551361084,
+      "learning_rate": 9.771205146202045e-06,
+      "loss": 0.0576,
+      "step": 571
+    },
+    {
+      "epoch": 0.2482638888888889,
+      "grad_norm": 0.8444411754608154,
+      "learning_rate": 9.770152885119925e-06,
+      "loss": 0.0709,
+      "step": 572
+    },
+    {
+      "epoch": 0.24869791666666666,
+      "grad_norm": 0.5682095289230347,
+      "learning_rate": 9.769098266757015e-06,
+      "loss": 0.0494,
+      "step": 573
+    },
+    {
+      "epoch": 0.24913194444444445,
+      "grad_norm": 0.7053081393241882,
+      "learning_rate": 9.768041291634483e-06,
+      "loss": 0.0473,
+      "step": 574
+    },
+    {
+      "epoch": 0.2495659722222222,
+      "grad_norm": 0.7969498038291931,
+      "learning_rate": 9.766981960274653e-06,
+      "loss": 0.061,
+      "step": 575
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.7599273324012756,
+      "learning_rate": 9.765920273201021e-06,
+      "loss": 0.0515,
+      "step": 576
+    },
+    {
+      "epoch": 0.2504340277777778,
+      "grad_norm": 0.46713560819625854,
+      "learning_rate": 9.764856230938244e-06,
+      "loss": 0.0313,
+      "step": 577
+    },
+    {
+      "epoch": 0.2508680555555556,
+      "grad_norm": 1.0053895711898804,
+      "learning_rate": 9.763789834012141e-06,
+      "loss": 0.0377,
+      "step": 578
+    },
+    {
+      "epoch": 0.2513020833333333,
+      "grad_norm": 1.5747969150543213,
+      "learning_rate": 9.7627210829497e-06,
+      "loss": 0.0576,
+      "step": 579
+    },
+    {
+      "epoch": 0.2517361111111111,
+      "grad_norm": 1.6479192972183228,
+      "learning_rate": 9.761649978279065e-06,
+      "loss": 0.0642,
+      "step": 580
+    },
+    {
+      "epoch": 0.2521701388888889,
+      "grad_norm": 1.494860053062439,
+      "learning_rate": 9.760576520529548e-06,
+      "loss": 0.0698,
+      "step": 581
+    },
+    {
+      "epoch": 0.2526041666666667,
+      "grad_norm": 1.0137051343917847,
+      "learning_rate": 9.759500710231624e-06,
+      "loss": 0.0597,
+      "step": 582
+    },
+    {
+      "epoch": 0.2530381944444444,
+      "grad_norm": 1.1509759426116943,
+      "learning_rate": 9.758422547916931e-06,
+      "loss": 0.0637,
+      "step": 583
+    },
+    {
+      "epoch": 0.2534722222222222,
+      "grad_norm": 1.0559190511703491,
+      "learning_rate": 9.757342034118262e-06,
+      "loss": 0.0696,
+      "step": 584
+    },
+    {
+      "epoch": 0.25390625,
+      "grad_norm": 0.6210541129112244,
+      "learning_rate": 9.756259169369584e-06,
+      "loss": 0.0615,
+      "step": 585
+    },
+    {
+      "epoch": 0.2543402777777778,
+      "grad_norm": 0.6341214776039124,
+      "learning_rate": 9.755173954206016e-06,
+      "loss": 0.0633,
+      "step": 586
+    },
+    {
+      "epoch": 0.2547743055555556,
+      "grad_norm": 0.6376259326934814,
+      "learning_rate": 9.754086389163843e-06,
+      "loss": 0.0604,
+      "step": 587
+    },
+    {
+      "epoch": 0.2552083333333333,
+      "grad_norm": 1.0872777700424194,
+      "learning_rate": 9.752996474780512e-06,
+      "loss": 0.0542,
+      "step": 588
+    },
+    {
+      "epoch": 0.2556423611111111,
+      "grad_norm": 0.6137789487838745,
+      "learning_rate": 9.751904211594627e-06,
+      "loss": 0.0724,
+      "step": 589
+    },
+    {
+      "epoch": 0.2560763888888889,
+      "grad_norm": 0.6856177449226379,
+      "learning_rate": 9.750809600145955e-06,
+      "loss": 0.0747,
+      "step": 590
+    },
+    {
+      "epoch": 0.2565104166666667,
+      "grad_norm": 0.47774896025657654,
+      "learning_rate": 9.749712640975424e-06,
+      "loss": 0.0477,
+      "step": 591
+    },
+    {
+      "epoch": 0.2569444444444444,
+      "grad_norm": 1.0445431470870972,
+      "learning_rate": 9.748613334625122e-06,
+      "loss": 0.0501,
+      "step": 592
+    },
+    {
+      "epoch": 0.2573784722222222,
+      "grad_norm": 1.031905174255371,
+      "learning_rate": 9.747511681638295e-06,
+      "loss": 0.086,
+      "step": 593
+    },
+    {
+      "epoch": 0.2578125,
+      "grad_norm": 0.8072367906570435,
+      "learning_rate": 9.746407682559352e-06,
+      "loss": 0.069,
+      "step": 594
+    },
+    {
+      "epoch": 0.2582465277777778,
+      "grad_norm": 0.7235659956932068,
+      "learning_rate": 9.745301337933856e-06,
+      "loss": 0.0611,
+      "step": 595
+    },
+    {
+      "epoch": 0.2586805555555556,
+      "grad_norm": 2.203366994857788,
+      "learning_rate": 9.744192648308539e-06,
+      "loss": 0.0646,
+      "step": 596
+    },
+    {
+      "epoch": 0.2591145833333333,
+      "grad_norm": 0.9374741315841675,
+      "learning_rate": 9.743081614231283e-06,
+      "loss": 0.0763,
+      "step": 597
+    },
+    {
+      "epoch": 0.2595486111111111,
+      "grad_norm": 1.1199864149093628,
+      "learning_rate": 9.741968236251126e-06,
+      "loss": 0.0582,
+      "step": 598
+    },
+    {
+      "epoch": 0.2599826388888889,
+      "grad_norm": 0.6270304918289185,
+      "learning_rate": 9.740852514918275e-06,
+      "loss": 0.0439,
+      "step": 599
+    },
+    {
+      "epoch": 0.2604166666666667,
+      "grad_norm": 2.2203094959259033,
+      "learning_rate": 9.739734450784088e-06,
+      "loss": 0.1011,
+      "step": 600
+    },
+    {
+      "epoch": 0.2604166666666667,
+      "eval_layout_rm_rank_test_accuracy": 0.9295603101782307,
+      "eval_layout_rm_rank_test_chosen_scores": 0.7769265174865723,
+      "eval_layout_rm_rank_test_loss": 0.06331586837768555,
+      "eval_layout_rm_rank_test_rejected_scores": 0.17147307097911835,
+      "eval_layout_rm_rank_test_runtime": 1181.1502,
+      "eval_layout_rm_rank_test_samples_per_second": 20.853,
+      "eval_layout_rm_rank_test_steps_per_second": 0.652,
+      "step": 600
+    },
+    {
+      "epoch": 0.2608506944444444,
+      "grad_norm": 2.3830530643463135,
+      "learning_rate": 9.738614044401083e-06,
+      "loss": 0.0879,
+      "step": 601
+    },
+    {
+      "epoch": 0.2612847222222222,
+      "grad_norm": 0.488438218832016,
+      "learning_rate": 9.737491296322931e-06,
+      "loss": 0.0515,
+      "step": 602
+    },
+    {
+      "epoch": 0.26171875,
+      "grad_norm": 0.635438859462738,
+      "learning_rate": 9.73636620710447e-06,
+      "loss": 0.0663,
+      "step": 603
+    },
+    {
+      "epoch": 0.2621527777777778,
+      "grad_norm": 0.7975000739097595,
+      "learning_rate": 9.735238777301682e-06,
+      "loss": 0.0548,
+      "step": 604
+    },
+    {
+      "epoch": 0.2625868055555556,
+      "grad_norm": 0.7051642537117004,
+      "learning_rate": 9.734109007471715e-06,
+      "loss": 0.0533,
+      "step": 605
+    },
+    {
+      "epoch": 0.2630208333333333,
+      "grad_norm": 1.0443001985549927,
+      "learning_rate": 9.732976898172872e-06,
+      "loss": 0.0629,
+      "step": 606
+    },
+    {
+      "epoch": 0.2634548611111111,
+      "grad_norm": 1.143349051475525,
+      "learning_rate": 9.731842449964609e-06,
+      "loss": 0.0601,
+      "step": 607
+    },
+    {
+      "epoch": 0.2638888888888889,
+      "grad_norm": 1.0804708003997803,
+      "learning_rate": 9.730705663407541e-06,
+      "loss": 0.0595,
+      "step": 608
+    },
+    {
+      "epoch": 0.2643229166666667,
+      "grad_norm": 0.9708920121192932,
+      "learning_rate": 9.729566539063438e-06,
+      "loss": 0.0532,
+      "step": 609
+    },
+    {
+      "epoch": 0.2647569444444444,
+      "grad_norm": 0.7470878958702087,
+      "learning_rate": 9.728425077495222e-06,
+      "loss": 0.0556,
+      "step": 610
+    },
+    {
+      "epoch": 0.2651909722222222,
+      "grad_norm": 1.107346773147583,
+      "learning_rate": 9.727281279266973e-06,
+      "loss": 0.0589,
+      "step": 611
+    },
+    {
+      "epoch": 0.265625,
+      "grad_norm": 0.8929963707923889,
+      "learning_rate": 9.726135144943928e-06,
+      "loss": 0.0675,
+      "step": 612
+    },
+    {
+      "epoch": 0.2660590277777778,
+      "grad_norm": 0.6872580647468567,
+      "learning_rate": 9.724986675092473e-06,
+      "loss": 0.0551,
+      "step": 613
+    },
+    {
+      "epoch": 0.2664930555555556,
+      "grad_norm": 0.7875890135765076,
+      "learning_rate": 9.723835870280152e-06,
+      "loss": 0.0664,
+      "step": 614
+    },
+    {
+      "epoch": 0.2669270833333333,
+      "grad_norm": 0.7004210948944092,
+      "learning_rate": 9.722682731075658e-06,
+      "loss": 0.0559,
+      "step": 615
+    },
+    {
+      "epoch": 0.2673611111111111,
+      "grad_norm": 1.1991928815841675,
+      "learning_rate": 9.721527258048849e-06,
+      "loss": 0.0877,
+      "step": 616
+    },
+    {
+      "epoch": 0.2677951388888889,
+      "grad_norm": 0.785410463809967,
+      "learning_rate": 9.72036945177072e-06,
+      "loss": 0.0748,
+      "step": 617
+    },
+    {
+      "epoch": 0.2682291666666667,
+      "grad_norm": 1.1013269424438477,
+      "learning_rate": 9.719209312813434e-06,
+      "loss": 0.0641,
+      "step": 618
+    },
+    {
+      "epoch": 0.2686631944444444,
+      "grad_norm": 0.5591453909873962,
+      "learning_rate": 9.718046841750297e-06,
+      "loss": 0.0509,
+      "step": 619
+    },
+    {
+      "epoch": 0.2690972222222222,
+      "grad_norm": 0.45849889516830444,
+      "learning_rate": 9.716882039155773e-06,
+      "loss": 0.0555,
+      "step": 620
+    },
+    {
+      "epoch": 0.26953125,
+      "grad_norm": 0.895866870880127,
+      "learning_rate": 9.715714905605473e-06,
+      "loss": 0.0783,
+      "step": 621
+    },
+    {
+      "epoch": 0.2699652777777778,
+      "grad_norm": 0.8574803471565247,
+      "learning_rate": 9.714545441676167e-06,
+      "loss": 0.0629,
+      "step": 622
+    },
+    {
+      "epoch": 0.2703993055555556,
+      "grad_norm": 0.5782579779624939,
+      "learning_rate": 9.713373647945768e-06,
+      "loss": 0.0701,
+      "step": 623
+    },
+    {
+      "epoch": 0.2708333333333333,
+      "grad_norm": 0.740162193775177,
+      "learning_rate": 9.712199524993347e-06,
+      "loss": 0.0684,
+      "step": 624
+    },
+    {
+      "epoch": 0.2712673611111111,
+      "grad_norm": 0.6386349201202393,
+      "learning_rate": 9.711023073399124e-06,
+      "loss": 0.052,
+      "step": 625
+    },
+    {
+      "epoch": 0.2717013888888889,
+      "grad_norm": 0.6506643295288086,
+      "learning_rate": 9.70984429374447e-06,
+      "loss": 0.0659,
+      "step": 626
+    },
+    {
+      "epoch": 0.2721354166666667,
+      "grad_norm": 1.0170015096664429,
+      "learning_rate": 9.708663186611907e-06,
+      "loss": 0.0864,
+      "step": 627
+    },
+    {
+      "epoch": 0.2725694444444444,
+      "grad_norm": 0.7113478183746338,
+      "learning_rate": 9.707479752585103e-06,
+      "loss": 0.0552,
+      "step": 628
+    },
+    {
+      "epoch": 0.2730034722222222,
+      "grad_norm": 0.9804077744483948,
+      "learning_rate": 9.70629399224888e-06,
+      "loss": 0.0902,
+      "step": 629
+    },
+    {
+      "epoch": 0.2734375,
+      "grad_norm": 0.5103983283042908,
+      "learning_rate": 9.705105906189212e-06,
+      "loss": 0.0439,
+      "step": 630
+    },
+    {
+      "epoch": 0.2738715277777778,
+      "grad_norm": 0.7018703818321228,
+      "learning_rate": 9.703915494993215e-06,
+      "loss": 0.0494,
+      "step": 631
+    },
+    {
+      "epoch": 0.2743055555555556,
+      "grad_norm": 0.6729298233985901,
+      "learning_rate": 9.702722759249161e-06,
+      "loss": 0.0637,
+      "step": 632
+    },
+    {
+      "epoch": 0.2747395833333333,
+      "grad_norm": 0.5301738977432251,
+      "learning_rate": 9.701527699546465e-06,
+      "loss": 0.0431,
+      "step": 633
+    },
+    {
+      "epoch": 0.2751736111111111,
+      "grad_norm": 0.7354344725608826,
+      "learning_rate": 9.700330316475697e-06,
+      "loss": 0.0742,
+      "step": 634
+    },
+    {
+      "epoch": 0.2756076388888889,
+      "grad_norm": 0.8999168276786804,
+      "learning_rate": 9.69913061062857e-06,
+      "loss": 0.0602,
+      "step": 635
+    },
+    {
+      "epoch": 0.2760416666666667,
+      "grad_norm": 0.6171438694000244,
+      "learning_rate": 9.697928582597942e-06,
+      "loss": 0.0565,
+      "step": 636
+    },
+    {
+      "epoch": 0.2764756944444444,
+      "grad_norm": 0.9136935472488403,
+      "learning_rate": 9.696724232977829e-06,
+      "loss": 0.0715,
+      "step": 637
+    },
+    {
+      "epoch": 0.2769097222222222,
+      "grad_norm": 0.827450692653656,
+      "learning_rate": 9.695517562363385e-06,
+      "loss": 0.0705,
+      "step": 638
+    },
+    {
+      "epoch": 0.27734375,
+      "grad_norm": 0.5673283934593201,
+      "learning_rate": 9.694308571350914e-06,
+      "loss": 0.0611,
+      "step": 639
+    },
+    {
+      "epoch": 0.2777777777777778,
+      "grad_norm": 0.8250726461410522,
+      "learning_rate": 9.693097260537867e-06,
+      "loss": 0.0567,
+      "step": 640
+    },
+    {
+      "epoch": 0.2782118055555556,
+      "grad_norm": 0.677777886390686,
+      "learning_rate": 9.691883630522842e-06,
+      "loss": 0.0698,
+      "step": 641
+    },
+    {
+      "epoch": 0.2786458333333333,
+      "grad_norm": 0.7001973390579224,
+      "learning_rate": 9.690667681905582e-06,
+      "loss": 0.0364,
+      "step": 642
+    },
+    {
+      "epoch": 0.2790798611111111,
+      "grad_norm": 0.5818181037902832,
+      "learning_rate": 9.689449415286974e-06,
+      "loss": 0.0449,
+      "step": 643
+    },
+    {
+      "epoch": 0.2795138888888889,
+      "grad_norm": 0.7768517732620239,
+      "learning_rate": 9.688228831269056e-06,
+      "loss": 0.0649,
+      "step": 644
+    },
+    {
+      "epoch": 0.2799479166666667,
+      "grad_norm": 0.8156334161758423,
+      "learning_rate": 9.687005930455002e-06,
+      "loss": 0.0785,
+      "step": 645
+    },
+    {
+      "epoch": 0.2803819444444444,
+      "grad_norm": 0.9127948880195618,
+      "learning_rate": 9.685780713449143e-06,
+      "loss": 0.0817,
+      "step": 646
+    },
+    {
+      "epoch": 0.2808159722222222,
+      "grad_norm": 0.822266161441803,
+      "learning_rate": 9.684553180856945e-06,
+      "loss": 0.0688,
+      "step": 647
+    },
+    {
+      "epoch": 0.28125,
+      "grad_norm": 0.5691503882408142,
+      "learning_rate": 9.683323333285022e-06,
+      "loss": 0.0477,
+      "step": 648
+    },
+    {
+      "epoch": 0.2816840277777778,
+      "grad_norm": 0.7925100922584534,
+      "learning_rate": 9.68209117134113e-06,
+      "loss": 0.0538,
+      "step": 649
+    },
+    {
+      "epoch": 0.2821180555555556,
+      "grad_norm": 0.7010947465896606,
+      "learning_rate": 9.680856695634174e-06,
+      "loss": 0.0671,
+      "step": 650
+    },
+    {
+      "epoch": 0.2825520833333333,
+      "grad_norm": 0.9301623702049255,
+      "learning_rate": 9.679619906774193e-06,
+      "loss": 0.0634,
+      "step": 651
+    },
+    {
+      "epoch": 0.2829861111111111,
+      "grad_norm": 0.7186834216117859,
+      "learning_rate": 9.678380805372381e-06,
+      "loss": 0.0542,
+      "step": 652
+    },
+    {
+      "epoch": 0.2834201388888889,
+      "grad_norm": 0.8715153336524963,
+      "learning_rate": 9.677139392041062e-06,
+      "loss": 0.0525,
+      "step": 653
+    },
+    {
+      "epoch": 0.2838541666666667,
+      "grad_norm": 1.021162748336792,
+      "learning_rate": 9.675895667393713e-06,
+      "loss": 0.0723,
+      "step": 654
+    },
+    {
+      "epoch": 0.2842881944444444,
+      "grad_norm": 0.7757608890533447,
+      "learning_rate": 9.67464963204495e-06,
+      "loss": 0.044,
+      "step": 655
+    },
+    {
+      "epoch": 0.2847222222222222,
+      "grad_norm": 1.0429720878601074,
+      "learning_rate": 9.673401286610527e-06,
+      "loss": 0.0768,
+      "step": 656
+    },
+    {
+      "epoch": 0.28515625,
+      "grad_norm": 0.6734326481819153,
+      "learning_rate": 9.672150631707343e-06,
+      "loss": 0.0602,
+      "step": 657
+    },
+    {
+      "epoch": 0.2855902777777778,
+      "grad_norm": 0.598980724811554,
+      "learning_rate": 9.67089766795344e-06,
+      "loss": 0.0576,
+      "step": 658
+    },
+    {
+      "epoch": 0.2860243055555556,
+      "grad_norm": 0.7078015208244324,
+      "learning_rate": 9.669642395967997e-06,
+      "loss": 0.0677,
+      "step": 659
+    },
+    {
+      "epoch": 0.2864583333333333,
+      "grad_norm": 0.7563789486885071,
+      "learning_rate": 9.668384816371337e-06,
+      "loss": 0.0676,
+      "step": 660
+    },
+    {
+      "epoch": 0.2868923611111111,
+      "grad_norm": 0.6212117671966553,
+      "learning_rate": 9.667124929784922e-06,
+      "loss": 0.0415,
+      "step": 661
+    },
+    {
+      "epoch": 0.2873263888888889,
+      "grad_norm": 0.9101629257202148,
+      "learning_rate": 9.665862736831351e-06,
+      "loss": 0.0587,
+      "step": 662
+    },
+    {
+      "epoch": 0.2877604166666667,
+      "grad_norm": 0.7955373525619507,
+      "learning_rate": 9.664598238134369e-06,
+      "loss": 0.0563,
+      "step": 663
+    },
+    {
+      "epoch": 0.2881944444444444,
+      "grad_norm": 0.933698832988739,
+      "learning_rate": 9.663331434318855e-06,
+      "loss": 0.0659,
+      "step": 664
+    },
+    {
+      "epoch": 0.2886284722222222,
+      "grad_norm": 0.6098387241363525,
+      "learning_rate": 9.662062326010831e-06,
+      "loss": 0.0519,
+      "step": 665
+    },
+    {
+      "epoch": 0.2890625,
+      "grad_norm": 0.6711975336074829,
+      "learning_rate": 9.660790913837455e-06,
+      "loss": 0.0694,
+      "step": 666
+    },
+    {
+      "epoch": 0.2894965277777778,
+      "grad_norm": 0.680465817451477,
+      "learning_rate": 9.659517198427024e-06,
+      "loss": 0.0552,
+      "step": 667
+    },
+    {
+      "epoch": 0.2899305555555556,
+      "grad_norm": 0.6448178887367249,
+      "learning_rate": 9.658241180408974e-06,
+      "loss": 0.0498,
+      "step": 668
+    },
+    {
+      "epoch": 0.2903645833333333,
+      "grad_norm": 0.9048629403114319,
+      "learning_rate": 9.65696286041388e-06,
+      "loss": 0.0576,
+      "step": 669
+    },
+    {
+      "epoch": 0.2907986111111111,
+      "grad_norm": 0.7213271856307983,
+      "learning_rate": 9.655682239073452e-06,
+      "loss": 0.0548,
+      "step": 670
+    },
+    {
+      "epoch": 0.2912326388888889,
+      "grad_norm": 1.198144793510437,
+      "learning_rate": 9.65439931702054e-06,
+      "loss": 0.0648,
+      "step": 671
+    },
+    {
+      "epoch": 0.2916666666666667,
+      "grad_norm": 0.9591615796089172,
+      "learning_rate": 9.653114094889128e-06,
+      "loss": 0.0649,
+      "step": 672
+    },
+    {
+      "epoch": 0.2921006944444444,
+      "grad_norm": 0.8612697124481201,
+      "learning_rate": 9.651826573314338e-06,
+      "loss": 0.0481,
+      "step": 673
+    },
+    {
+      "epoch": 0.2925347222222222,
+      "grad_norm": 0.5536429286003113,
+      "learning_rate": 9.650536752932429e-06,
+      "loss": 0.0356,
+      "step": 674
+    },
+    {
+      "epoch": 0.29296875,
+      "grad_norm": 0.7021438479423523,
+      "learning_rate": 9.649244634380794e-06,
+      "loss": 0.0543,
+      "step": 675
+    },
+    {
+      "epoch": 0.2934027777777778,
+      "grad_norm": 0.7661638259887695,
+      "learning_rate": 9.647950218297966e-06,
+      "loss": 0.0589,
+      "step": 676
+    },
+    {
+      "epoch": 0.2938368055555556,
+      "grad_norm": 0.5391556620597839,
+      "learning_rate": 9.646653505323609e-06,
+      "loss": 0.041,
+      "step": 677
+    },
+    {
+      "epoch": 0.2942708333333333,
+      "grad_norm": 0.7588377594947815,
+      "learning_rate": 9.645354496098522e-06,
+      "loss": 0.0576,
+      "step": 678
+    },
+    {
+      "epoch": 0.2947048611111111,
+      "grad_norm": 1.0408495664596558,
+      "learning_rate": 9.644053191264643e-06,
+      "loss": 0.0923,
+      "step": 679
+    },
+    {
+      "epoch": 0.2951388888888889,
+      "grad_norm": 0.9011924266815186,
+      "learning_rate": 9.64274959146504e-06,
+      "loss": 0.0629,
+      "step": 680
+    },
+    {
+      "epoch": 0.2955729166666667,
+      "grad_norm": 1.4155269861221313,
+      "learning_rate": 9.641443697343918e-06,
+      "loss": 0.0723,
+      "step": 681
+    },
+    {
+      "epoch": 0.2960069444444444,
+      "grad_norm": 0.5065480470657349,
+      "learning_rate": 9.640135509546613e-06,
+      "loss": 0.0445,
+      "step": 682
+    },
+    {
+      "epoch": 0.2964409722222222,
+      "grad_norm": 0.5630937218666077,
+      "learning_rate": 9.638825028719596e-06,
+      "loss": 0.0435,
+      "step": 683
+    },
+    {
+      "epoch": 0.296875,
+      "grad_norm": 0.796134889125824,
+      "learning_rate": 9.637512255510475e-06,
+      "loss": 0.0415,
+      "step": 684
+    },
+    {
+      "epoch": 0.2973090277777778,
+      "grad_norm": 0.8620226383209229,
+      "learning_rate": 9.636197190567983e-06,
+      "loss": 0.087,
+      "step": 685
+    },
+    {
+      "epoch": 0.2977430555555556,
+      "grad_norm": 0.8691263198852539,
+      "learning_rate": 9.634879834541989e-06,
+      "loss": 0.0651,
+      "step": 686
+    },
+    {
+      "epoch": 0.2981770833333333,
+      "grad_norm": 0.4917859733104706,
+      "learning_rate": 9.633560188083498e-06,
+      "loss": 0.0367,
+      "step": 687
+    },
+    {
+      "epoch": 0.2986111111111111,
+      "grad_norm": 0.7439955472946167,
+      "learning_rate": 9.632238251844643e-06,
+      "loss": 0.0367,
+      "step": 688
+    },
+    {
+      "epoch": 0.2990451388888889,
+      "grad_norm": 1.3402535915374756,
+      "learning_rate": 9.630914026478688e-06,
+      "loss": 0.097,
+      "step": 689
+    },
+    {
+      "epoch": 0.2994791666666667,
+      "grad_norm": 1.0623676776885986,
+      "learning_rate": 9.62958751264003e-06,
+      "loss": 0.0483,
+      "step": 690
+    },
+    {
+      "epoch": 0.2999131944444444,
+      "grad_norm": 0.9007038474082947,
+      "learning_rate": 9.628258710984195e-06,
+      "loss": 0.0773,
+      "step": 691
+    },
+    {
+      "epoch": 0.3003472222222222,
+      "grad_norm": 0.7735984921455383,
+      "learning_rate": 9.626927622167844e-06,
+      "loss": 0.0538,
+      "step": 692
+    },
+    {
+      "epoch": 0.30078125,
+      "grad_norm": 0.6532759666442871,
+      "learning_rate": 9.625594246848762e-06,
+      "loss": 0.068,
+      "step": 693
+    },
+    {
+      "epoch": 0.3012152777777778,
+      "grad_norm": 1.0619275569915771,
+      "learning_rate": 9.62425858568587e-06,
+      "loss": 0.065,
+      "step": 694
+    },
+    {
+      "epoch": 0.3016493055555556,
+      "grad_norm": 0.8146920800209045,
+      "learning_rate": 9.622920639339212e-06,
+      "loss": 0.0624,
+      "step": 695
+    },
+    {
+      "epoch": 0.3020833333333333,
+      "grad_norm": 0.6527515053749084,
+      "learning_rate": 9.62158040846997e-06,
+      "loss": 0.0526,
+      "step": 696
+    },
+    {
+      "epoch": 0.3025173611111111,
+      "grad_norm": 0.6193270087242126,
+      "learning_rate": 9.620237893740446e-06,
+      "loss": 0.0566,
+      "step": 697
+    },
+    {
+      "epoch": 0.3029513888888889,
+      "grad_norm": 0.8723997473716736,
+      "learning_rate": 9.618893095814074e-06,
+      "loss": 0.058,
+      "step": 698
+    },
+    {
+      "epoch": 0.3033854166666667,
+      "grad_norm": 0.761968195438385,
+      "learning_rate": 9.61754601535542e-06,
+      "loss": 0.0479,
+      "step": 699
+    },
+    {
+      "epoch": 0.3038194444444444,
+      "grad_norm": 1.001143217086792,
+      "learning_rate": 9.616196653030173e-06,
+      "loss": 0.0608,
+      "step": 700
+    },
+    {
+      "epoch": 0.3042534722222222,
+      "grad_norm": 0.9322062730789185,
+      "learning_rate": 9.614845009505151e-06,
+      "loss": 0.0626,
+      "step": 701
+    },
+    {
+      "epoch": 0.3046875,
+      "grad_norm": 1.3316640853881836,
+      "learning_rate": 9.613491085448302e-06,
+      "loss": 0.0631,
+      "step": 702
+    },
+    {
+      "epoch": 0.3051215277777778,
+      "grad_norm": 0.8206167221069336,
+      "learning_rate": 9.612134881528695e-06,
+      "loss": 0.0455,
+      "step": 703
+    },
+    {
+      "epoch": 0.3055555555555556,
+      "grad_norm": 0.9950183629989624,
+      "learning_rate": 9.610776398416533e-06,
+      "loss": 0.0833,
+      "step": 704
+    },
+    {
+      "epoch": 0.3059895833333333,
+      "grad_norm": 0.6813333630561829,
+      "learning_rate": 9.609415636783138e-06,
+      "loss": 0.0583,
+      "step": 705
+    },
+    {
+      "epoch": 0.3064236111111111,
+      "grad_norm": 0.817216694355011,
+      "learning_rate": 9.608052597300965e-06,
+      "loss": 0.0722,
+      "step": 706
+    },
+    {
+      "epoch": 0.3068576388888889,
+      "grad_norm": 0.5392987132072449,
+      "learning_rate": 9.606687280643591e-06,
+      "loss": 0.0425,
+      "step": 707
+    },
+    {
+      "epoch": 0.3072916666666667,
+      "grad_norm": 0.7456854581832886,
+      "learning_rate": 9.605319687485718e-06,
+      "loss": 0.0659,
+      "step": 708
+    },
+    {
+      "epoch": 0.3077256944444444,
+      "grad_norm": 0.5358279347419739,
+      "learning_rate": 9.60394981850317e-06,
+      "loss": 0.0631,
+      "step": 709
+    },
+    {
+      "epoch": 0.3081597222222222,
+      "grad_norm": 0.5878068208694458,
+      "learning_rate": 9.602577674372907e-06,
+      "loss": 0.0358,
+      "step": 710
+    },
+    {
+      "epoch": 0.30859375,
+      "grad_norm": 0.7481245994567871,
+      "learning_rate": 9.601203255773e-06,
+      "loss": 0.0706,
+      "step": 711
+    },
+    {
+      "epoch": 0.3090277777777778,
+      "grad_norm": 0.7426470518112183,
+      "learning_rate": 9.599826563382649e-06,
+      "loss": 0.0496,
+      "step": 712
+    },
+    {
+      "epoch": 0.3094618055555556,
+      "grad_norm": 0.6087659001350403,
+      "learning_rate": 9.598447597882181e-06,
+      "loss": 0.0415,
+      "step": 713
+    },
+    {
+      "epoch": 0.3098958333333333,
+      "grad_norm": 0.8318248987197876,
+      "learning_rate": 9.597066359953043e-06,
+      "loss": 0.0335,
+      "step": 714
+    },
+    {
+      "epoch": 0.3103298611111111,
+      "grad_norm": 1.3557584285736084,
+      "learning_rate": 9.595682850277806e-06,
+      "loss": 0.0925,
+      "step": 715
+    },
+    {
+      "epoch": 0.3107638888888889,
+      "grad_norm": 0.9173680543899536,
+      "learning_rate": 9.59429706954016e-06,
+      "loss": 0.0587,
+      "step": 716
+    },
+    {
+      "epoch": 0.3111979166666667,
+      "grad_norm": 0.8296968340873718,
+      "learning_rate": 9.592909018424924e-06,
+      "loss": 0.0489,
+      "step": 717
+    },
+    {
+      "epoch": 0.3116319444444444,
+      "grad_norm": 0.6870502233505249,
+      "learning_rate": 9.591518697618033e-06,
+      "loss": 0.05,
+      "step": 718
+    },
+    {
+      "epoch": 0.3120659722222222,
+      "grad_norm": 1.0346730947494507,
+      "learning_rate": 9.59012610780655e-06,
+      "loss": 0.0654,
+      "step": 719
+    },
+    {
+      "epoch": 0.3125,
+      "grad_norm": 1.2370779514312744,
+      "learning_rate": 9.588731249678648e-06,
+      "loss": 0.0848,
+      "step": 720
+    },
+    {
+      "epoch": 0.3129340277777778,
+      "grad_norm": 0.6258195638656616,
+      "learning_rate": 9.587334123923634e-06,
+      "loss": 0.0511,
+      "step": 721
+    },
+    {
+      "epoch": 0.3133680555555556,
+      "grad_norm": 0.6205534338951111,
+      "learning_rate": 9.585934731231931e-06,
+      "loss": 0.0646,
+      "step": 722
+    },
+    {
+      "epoch": 0.3138020833333333,
+      "grad_norm": 0.8664010167121887,
+      "learning_rate": 9.584533072295077e-06,
+      "loss": 0.0706,
+      "step": 723
+    },
+    {
+      "epoch": 0.3142361111111111,
+      "grad_norm": 0.5968353152275085,
+      "learning_rate": 9.583129147805734e-06,
+      "loss": 0.0627,
+      "step": 724
+    },
+    {
+      "epoch": 0.3146701388888889,
+      "grad_norm": 0.464321106672287,
+      "learning_rate": 9.581722958457687e-06,
+      "loss": 0.0566,
+      "step": 725
+    },
+    {
+      "epoch": 0.3151041666666667,
+      "grad_norm": 0.5013965964317322,
+      "learning_rate": 9.580314504945834e-06,
+      "loss": 0.0379,
+      "step": 726
+    },
+    {
+      "epoch": 0.3155381944444444,
+      "grad_norm": 0.46232596039772034,
+      "learning_rate": 9.578903787966195e-06,
+      "loss": 0.0416,
+      "step": 727
+    },
+    {
+      "epoch": 0.3159722222222222,
+      "grad_norm": 0.6010047197341919,
+      "learning_rate": 9.57749080821591e-06,
+      "loss": 0.0531,
+      "step": 728
+    },
+    {
+      "epoch": 0.31640625,
+      "grad_norm": 0.5406182408332825,
+      "learning_rate": 9.576075566393232e-06,
+      "loss": 0.0513,
+      "step": 729
+    },
+    {
+      "epoch": 0.3168402777777778,
+      "grad_norm": 0.82621830701828,
+      "learning_rate": 9.57465806319754e-06,
+      "loss": 0.061,
+      "step": 730
+    },
+    {
+      "epoch": 0.3172743055555556,
+      "grad_norm": 1.1684813499450684,
+      "learning_rate": 9.573238299329323e-06,
+      "loss": 0.0912,
+      "step": 731
+    },
+    {
+      "epoch": 0.3177083333333333,
+      "grad_norm": 0.9138361811637878,
+      "learning_rate": 9.571816275490191e-06,
+      "loss": 0.0876,
+      "step": 732
+    },
+    {
+      "epoch": 0.3181423611111111,
+      "grad_norm": 0.9768379926681519,
+      "learning_rate": 9.570391992382869e-06,
+      "loss": 0.0627,
+      "step": 733
+    },
+    {
+      "epoch": 0.3185763888888889,
+      "grad_norm": 0.8773151636123657,
+      "learning_rate": 9.568965450711199e-06,
+      "loss": 0.0771,
+      "step": 734
+    },
+    {
+      "epoch": 0.3190104166666667,
+      "grad_norm": 0.6589311361312866,
+      "learning_rate": 9.567536651180139e-06,
+      "loss": 0.0638,
+      "step": 735
+    },
+    {
+      "epoch": 0.3194444444444444,
+      "grad_norm": 1.1911635398864746,
+      "learning_rate": 9.566105594495765e-06,
+      "loss": 0.0747,
+      "step": 736
+    },
+    {
+      "epoch": 0.3198784722222222,
+      "grad_norm": 0.592398464679718,
+      "learning_rate": 9.564672281365263e-06,
+      "loss": 0.0522,
+      "step": 737
+    },
+    {
+      "epoch": 0.3203125,
+      "grad_norm": 0.5514065623283386,
+      "learning_rate": 9.563236712496944e-06,
+      "loss": 0.0714,
+      "step": 738
+    },
+    {
+      "epoch": 0.3207465277777778,
+      "grad_norm": 0.7221823930740356,
+      "learning_rate": 9.561798888600222e-06,
+      "loss": 0.0463,
+      "step": 739
+    },
+    {
+      "epoch": 0.3211805555555556,
+      "grad_norm": 0.5506830811500549,
+      "learning_rate": 9.560358810385632e-06,
+      "loss": 0.068,
+      "step": 740
+    },
+    {
+      "epoch": 0.3216145833333333,
+      "grad_norm": 0.6145944595336914,
+      "learning_rate": 9.558916478564824e-06,
+      "loss": 0.0485,
+      "step": 741
+    },
+    {
+      "epoch": 0.3220486111111111,
+      "grad_norm": 0.5052325129508972,
+      "learning_rate": 9.557471893850556e-06,
+      "loss": 0.0425,
+      "step": 742
+    },
+    {
+      "epoch": 0.3224826388888889,
+      "grad_norm": 0.7461121678352356,
+      "learning_rate": 9.556025056956704e-06,
+      "loss": 0.0755,
+      "step": 743
+    },
+    {
+      "epoch": 0.3229166666666667,
+      "grad_norm": 0.7425948977470398,
+      "learning_rate": 9.554575968598257e-06,
+      "loss": 0.0411,
+      "step": 744
+    },
+    {
+      "epoch": 0.3233506944444444,
+      "grad_norm": 0.7867175340652466,
+      "learning_rate": 9.553124629491314e-06,
+      "loss": 0.0774,
+      "step": 745
+    },
+    {
+      "epoch": 0.3237847222222222,
+      "grad_norm": 0.5744606852531433,
+      "learning_rate": 9.551671040353088e-06,
+      "loss": 0.049,
+      "step": 746
+    },
+    {
+      "epoch": 0.32421875,
+      "grad_norm": 0.69539475440979,
+      "learning_rate": 9.550215201901904e-06,
+      "loss": 0.0381,
+      "step": 747
+    },
+    {
+      "epoch": 0.3246527777777778,
+      "grad_norm": 0.952163815498352,
+      "learning_rate": 9.548757114857198e-06,
+      "loss": 0.0714,
+      "step": 748
+    },
+    {
+      "epoch": 0.3250868055555556,
+      "grad_norm": 1.1044210195541382,
+      "learning_rate": 9.547296779939517e-06,
+      "loss": 0.0658,
+      "step": 749
+    },
+    {
+      "epoch": 0.3255208333333333,
+      "grad_norm": 1.186783790588379,
+      "learning_rate": 9.545834197870519e-06,
+      "loss": 0.0736,
+      "step": 750
+    },
+    {
+      "epoch": 0.3259548611111111,
+      "grad_norm": 0.6580379605293274,
+      "learning_rate": 9.54436936937297e-06,
+      "loss": 0.06,
+      "step": 751
+    },
+    {
+      "epoch": 0.3263888888888889,
+      "grad_norm": 0.7105837464332581,
+      "learning_rate": 9.542902295170755e-06,
+      "loss": 0.0486,
+      "step": 752
+    },
+    {
+      "epoch": 0.3268229166666667,
+      "grad_norm": 1.1437960863113403,
+      "learning_rate": 9.541432975988856e-06,
+      "loss": 0.0747,
+      "step": 753
+    },
+    {
+      "epoch": 0.3272569444444444,
+      "grad_norm": 0.8399361968040466,
+      "learning_rate": 9.539961412553375e-06,
+      "loss": 0.0799,
+      "step": 754
+    },
+    {
+      "epoch": 0.3276909722222222,
+      "grad_norm": 0.6523122191429138,
+      "learning_rate": 9.538487605591518e-06,
+      "loss": 0.0495,
+      "step": 755
+    },
+    {
+      "epoch": 0.328125,
+      "grad_norm": 0.6750720739364624,
+      "learning_rate": 9.5370115558316e-06,
+      "loss": 0.0621,
+      "step": 756
+    },
+    {
+      "epoch": 0.3285590277777778,
+      "grad_norm": 0.5642672181129456,
+      "learning_rate": 9.535533264003043e-06,
+      "loss": 0.0557,
+      "step": 757
+    },
+    {
+      "epoch": 0.3289930555555556,
+      "grad_norm": 0.47357630729675293,
+      "learning_rate": 9.534052730836383e-06,
+      "loss": 0.0506,
+      "step": 758
+    },
+    {
+      "epoch": 0.3294270833333333,
+      "grad_norm": 0.704122006893158,
+      "learning_rate": 9.532569957063257e-06,
+      "loss": 0.0553,
+      "step": 759
+    },
+    {
+      "epoch": 0.3298611111111111,
+      "grad_norm": 0.5729414820671082,
+      "learning_rate": 9.531084943416413e-06,
+      "loss": 0.0509,
+      "step": 760
+    },
+    {
+      "epoch": 0.3302951388888889,
+      "grad_norm": 0.6760425567626953,
+      "learning_rate": 9.529597690629702e-06,
+      "loss": 0.0471,
+      "step": 761
+    },
+    {
+      "epoch": 0.3307291666666667,
+      "grad_norm": 1.0773210525512695,
+      "learning_rate": 9.528108199438088e-06,
+      "loss": 0.0687,
+      "step": 762
+    },
+    {
+      "epoch": 0.3311631944444444,
+      "grad_norm": 0.6594577431678772,
+      "learning_rate": 9.526616470577636e-06,
+      "loss": 0.045,
+      "step": 763
+    },
+    {
+      "epoch": 0.3315972222222222,
+      "grad_norm": 1.134124517440796,
+      "learning_rate": 9.525122504785515e-06,
+      "loss": 0.0778,
+      "step": 764
+    },
+    {
+      "epoch": 0.33203125,
+      "grad_norm": 0.9284763932228088,
+      "learning_rate": 9.523626302800009e-06,
+      "loss": 0.0531,
+      "step": 765
+    },
+    {
+      "epoch": 0.3324652777777778,
+      "grad_norm": 0.855278491973877,
+      "learning_rate": 9.522127865360494e-06,
+      "loss": 0.0503,
+      "step": 766
+    },
+    {
+      "epoch": 0.3328993055555556,
+      "grad_norm": 1.1203644275665283,
+      "learning_rate": 9.520627193207462e-06,
+      "loss": 0.054,
+      "step": 767
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.8028052449226379,
+      "learning_rate": 9.5191242870825e-06,
+      "loss": 0.0824,
+      "step": 768
+    },
+    {
+      "epoch": 0.3337673611111111,
+      "grad_norm": 0.6946297883987427,
+      "learning_rate": 9.51761914772831e-06,
+      "loss": 0.0484,
+      "step": 769
+    },
+    {
+      "epoch": 0.3342013888888889,
+      "grad_norm": 0.531436562538147,
+      "learning_rate": 9.516111775888685e-06,
+      "loss": 0.04,
+      "step": 770
+    },
+    {
+      "epoch": 0.3346354166666667,
+      "grad_norm": 0.6533716917037964,
+      "learning_rate": 9.514602172308532e-06,
+      "loss": 0.0648,
+      "step": 771
+    },
+    {
+      "epoch": 0.3350694444444444,
+      "grad_norm": 1.1470931768417358,
+      "learning_rate": 9.513090337733852e-06,
+      "loss": 0.0774,
+      "step": 772
+    },
+    {
+      "epoch": 0.3355034722222222,
+      "grad_norm": 0.771140992641449,
+      "learning_rate": 9.511576272911757e-06,
+      "loss": 0.0708,
+      "step": 773
+    },
+    {
+      "epoch": 0.3359375,
+      "grad_norm": 0.6818678975105286,
+      "learning_rate": 9.510059978590456e-06,
+      "loss": 0.0732,
+      "step": 774
+    },
+    {
+      "epoch": 0.3363715277777778,
+      "grad_norm": 0.9621103405952454,
+      "learning_rate": 9.50854145551926e-06,
+      "loss": 0.0487,
+      "step": 775
+    },
+    {
+      "epoch": 0.3368055555555556,
+      "grad_norm": 0.7351205348968506,
+      "learning_rate": 9.507020704448581e-06,
+      "loss": 0.0747,
+      "step": 776
+    },
+    {
+      "epoch": 0.3372395833333333,
+      "grad_norm": 0.738548219203949,
+      "learning_rate": 9.505497726129935e-06,
+      "loss": 0.062,
+      "step": 777
+    },
+    {
+      "epoch": 0.3376736111111111,
+      "grad_norm": 0.5943174958229065,
+      "learning_rate": 9.503972521315935e-06,
+      "loss": 0.0591,
+      "step": 778
+    },
+    {
+      "epoch": 0.3381076388888889,
+      "grad_norm": 0.5805765986442566,
+      "learning_rate": 9.502445090760299e-06,
+      "loss": 0.0586,
+      "step": 779
+    },
+    {
+      "epoch": 0.3385416666666667,
+      "grad_norm": 0.6666689515113831,
+      "learning_rate": 9.50091543521784e-06,
+      "loss": 0.0748,
+      "step": 780
+    },
+    {
+      "epoch": 0.3389756944444444,
+      "grad_norm": 0.5110723972320557,
+      "learning_rate": 9.49938355544447e-06,
+      "loss": 0.0527,
+      "step": 781
+    },
+    {
+      "epoch": 0.3394097222222222,
+      "grad_norm": 1.0024490356445312,
+      "learning_rate": 9.497849452197207e-06,
+      "loss": 0.0641,
+      "step": 782
+    },
+    {
+      "epoch": 0.33984375,
+      "grad_norm": 0.682685911655426,
+      "learning_rate": 9.496313126234162e-06,
+      "loss": 0.0726,
+      "step": 783
+    },
+    {
+      "epoch": 0.3402777777777778,
+      "grad_norm": 0.6020975112915039,
+      "learning_rate": 9.494774578314545e-06,
+      "loss": 0.0635,
+      "step": 784
+    },
+    {
+      "epoch": 0.3407118055555556,
+      "grad_norm": 0.6309037804603577,
+      "learning_rate": 9.493233809198665e-06,
+      "loss": 0.0507,
+      "step": 785
+    },
+    {
+      "epoch": 0.3411458333333333,
+      "grad_norm": 0.7836750745773315,
+      "learning_rate": 9.491690819647927e-06,
+      "loss": 0.0488,
+      "step": 786
+    },
+    {
+      "epoch": 0.3415798611111111,
+      "grad_norm": 1.142842411994934,
+      "learning_rate": 9.49014561042484e-06,
+      "loss": 0.0704,
+      "step": 787
+    },
+    {
+      "epoch": 0.3420138888888889,
+      "grad_norm": 1.004252314567566,
+      "learning_rate": 9.488598182292998e-06,
+      "loss": 0.0467,
+      "step": 788
+    },
+    {
+      "epoch": 0.3424479166666667,
+      "grad_norm": 1.094494104385376,
+      "learning_rate": 9.487048536017103e-06,
+      "loss": 0.0831,
+      "step": 789
+    },
+    {
+      "epoch": 0.3428819444444444,
+      "grad_norm": 1.0647095441818237,
+      "learning_rate": 9.485496672362948e-06,
+      "loss": 0.0691,
+      "step": 790
+    },
+    {
+      "epoch": 0.3433159722222222,
+      "grad_norm": 0.8126431703567505,
+      "learning_rate": 9.48394259209742e-06,
+      "loss": 0.0514,
+      "step": 791
+    },
+    {
+      "epoch": 0.34375,
+      "grad_norm": 0.9838793873786926,
+      "learning_rate": 9.482386295988504e-06,
+      "loss": 0.0843,
+      "step": 792
+    },
+    {
+      "epoch": 0.3441840277777778,
+      "grad_norm": 0.7736327648162842,
+      "learning_rate": 9.480827784805278e-06,
+      "loss": 0.0423,
+      "step": 793
+    },
+    {
+      "epoch": 0.3446180555555556,
+      "grad_norm": 0.7374549508094788,
+      "learning_rate": 9.479267059317921e-06,
+      "loss": 0.0551,
+      "step": 794
+    },
+    {
+      "epoch": 0.3450520833333333,
+      "grad_norm": 0.9624958634376526,
+      "learning_rate": 9.477704120297698e-06,
+      "loss": 0.0599,
+      "step": 795
+    },
+    {
+      "epoch": 0.3454861111111111,
+      "grad_norm": 0.9699764847755432,
+      "learning_rate": 9.47613896851697e-06,
+      "loss": 0.0463,
+      "step": 796
+    },
+    {
+      "epoch": 0.3459201388888889,
+      "grad_norm": 0.7315773963928223,
+      "learning_rate": 9.474571604749196e-06,
+      "loss": 0.0585,
+      "step": 797
+    },
+    {
+      "epoch": 0.3463541666666667,
+      "grad_norm": 1.3439610004425049,
+      "learning_rate": 9.473002029768923e-06,
+      "loss": 0.0794,
+      "step": 798
+    },
+    {
+      "epoch": 0.3467881944444444,
+      "grad_norm": 0.6556320190429688,
+      "learning_rate": 9.471430244351793e-06,
+      "loss": 0.0647,
+      "step": 799
+    },
+    {
+      "epoch": 0.3472222222222222,
+      "grad_norm": 0.7467793822288513,
+      "learning_rate": 9.469856249274541e-06,
+      "loss": 0.0612,
+      "step": 800
+    },
+    {
+      "epoch": 0.34765625,
+      "grad_norm": 0.9917733073234558,
+      "learning_rate": 9.468280045314991e-06,
+      "loss": 0.0691,
+      "step": 801
+    },
+    {
+      "epoch": 0.3480902777777778,
+      "grad_norm": 0.577621579170227,
+      "learning_rate": 9.466701633252062e-06,
+      "loss": 0.0518,
+      "step": 802
+    },
+    {
+      "epoch": 0.3485243055555556,
+      "grad_norm": 0.8279784917831421,
+      "learning_rate": 9.465121013865761e-06,
+      "loss": 0.067,
+      "step": 803
+    },
+    {
+      "epoch": 0.3489583333333333,
+      "grad_norm": 0.5595852732658386,
+      "learning_rate": 9.46353818793719e-06,
+      "loss": 0.0478,
+      "step": 804
+    },
+    {
+      "epoch": 0.3493923611111111,
+      "grad_norm": 0.8331674933433533,
+      "learning_rate": 9.461953156248534e-06,
+      "loss": 0.0605,
+      "step": 805
+    },
+    {
+      "epoch": 0.3498263888888889,
+      "grad_norm": 0.7563780546188354,
+      "learning_rate": 9.460365919583077e-06,
+      "loss": 0.0374,
+      "step": 806
+    },
+    {
+      "epoch": 0.3502604166666667,
+      "grad_norm": 0.6543465852737427,
+      "learning_rate": 9.45877647872519e-06,
+      "loss": 0.0458,
+      "step": 807
+    },
+    {
+      "epoch": 0.3506944444444444,
+      "grad_norm": 0.9966139793395996,
+      "learning_rate": 9.457184834460327e-06,
+      "loss": 0.0615,
+      "step": 808
+    },
+    {
+      "epoch": 0.3511284722222222,
+      "grad_norm": 0.8658338189125061,
+      "learning_rate": 9.455590987575038e-06,
+      "loss": 0.0586,
+      "step": 809
+    },
+    {
+      "epoch": 0.3515625,
+      "grad_norm": 0.5670788884162903,
+      "learning_rate": 9.453994938856958e-06,
+      "loss": 0.0378,
+      "step": 810
+    },
+    {
+      "epoch": 0.3519965277777778,
+      "grad_norm": 0.9414218664169312,
+      "learning_rate": 9.452396689094815e-06,
+      "loss": 0.0694,
+      "step": 811
+    },
+    {
+      "epoch": 0.3524305555555556,
+      "grad_norm": 0.8863024115562439,
+      "learning_rate": 9.450796239078417e-06,
+      "loss": 0.0554,
+      "step": 812
+    },
+    {
+      "epoch": 0.3528645833333333,
+      "grad_norm": 0.7250595092773438,
+      "learning_rate": 9.449193589598664e-06,
+      "loss": 0.0565,
+      "step": 813
+    },
+    {
+      "epoch": 0.3532986111111111,
+      "grad_norm": 0.845879316329956,
+      "learning_rate": 9.447588741447542e-06,
+      "loss": 0.0733,
+      "step": 814
+    },
+    {
+      "epoch": 0.3537326388888889,
+      "grad_norm": 1.1272249221801758,
+      "learning_rate": 9.445981695418124e-06,
+      "loss": 0.0833,
+      "step": 815
+    },
+    {
+      "epoch": 0.3541666666666667,
+      "grad_norm": 0.516075611114502,
+      "learning_rate": 9.444372452304571e-06,
+      "loss": 0.0444,
+      "step": 816
+    },
+    {
+      "epoch": 0.3546006944444444,
+      "grad_norm": 0.5312237739562988,
+      "learning_rate": 9.442761012902125e-06,
+      "loss": 0.0417,
+      "step": 817
+    },
+    {
+      "epoch": 0.3550347222222222,
+      "grad_norm": 0.6717514991760254,
+      "learning_rate": 9.441147378007113e-06,
+      "loss": 0.0562,
+      "step": 818
+    },
+    {
+      "epoch": 0.35546875,
+      "grad_norm": 0.6433478593826294,
+      "learning_rate": 9.439531548416955e-06,
+      "loss": 0.0521,
+      "step": 819
+    },
+    {
+      "epoch": 0.3559027777777778,
+      "grad_norm": 0.7206553220748901,
+      "learning_rate": 9.437913524930148e-06,
+      "loss": 0.0727,
+      "step": 820
+    },
+    {
+      "epoch": 0.3563368055555556,
+      "grad_norm": 1.5485148429870605,
+      "learning_rate": 9.436293308346275e-06,
+      "loss": 0.0875,
+      "step": 821
+    },
+    {
+      "epoch": 0.3567708333333333,
+      "grad_norm": 1.1846742630004883,
+      "learning_rate": 9.434670899466007e-06,
+      "loss": 0.0647,
+      "step": 822
+    },
+    {
+      "epoch": 0.3572048611111111,
+      "grad_norm": 0.6657606959342957,
+      "learning_rate": 9.433046299091091e-06,
+      "loss": 0.0464,
+      "step": 823
+    },
+    {
+      "epoch": 0.3576388888888889,
+      "grad_norm": 0.6496115326881409,
+      "learning_rate": 9.431419508024359e-06,
+      "loss": 0.0716,
+      "step": 824
+    },
+    {
+      "epoch": 0.3580729166666667,
+      "grad_norm": 0.9403788447380066,
+      "learning_rate": 9.429790527069733e-06,
+      "loss": 0.0627,
+      "step": 825
+    },
+    {
+      "epoch": 0.3585069444444444,
+      "grad_norm": 0.5675524473190308,
+      "learning_rate": 9.428159357032207e-06,
+      "loss": 0.0527,
+      "step": 826
+    },
+    {
+      "epoch": 0.3589409722222222,
+      "grad_norm": 0.7261064648628235,
+      "learning_rate": 9.426525998717861e-06,
+      "loss": 0.0603,
+      "step": 827
+    },
+    {
+      "epoch": 0.359375,
+      "grad_norm": 0.8921183943748474,
+      "learning_rate": 9.42489045293386e-06,
+      "loss": 0.077,
+      "step": 828
+    },
+    {
+      "epoch": 0.3598090277777778,
+      "grad_norm": 0.4754103124141693,
+      "learning_rate": 9.423252720488445e-06,
+      "loss": 0.042,
+      "step": 829
+    },
+    {
+      "epoch": 0.3602430555555556,
+      "grad_norm": 0.5494545698165894,
+      "learning_rate": 9.421612802190936e-06,
+      "loss": 0.0545,
+      "step": 830
+    },
+    {
+      "epoch": 0.3606770833333333,
+      "grad_norm": 0.9630483984947205,
+      "learning_rate": 9.419970698851739e-06,
+      "loss": 0.0702,
+      "step": 831
+    },
+    {
+      "epoch": 0.3611111111111111,
+      "grad_norm": 0.7188891172409058,
+      "learning_rate": 9.418326411282338e-06,
+      "loss": 0.0555,
+      "step": 832
+    },
+    {
+      "epoch": 0.3615451388888889,
+      "grad_norm": 0.5888816714286804,
+      "learning_rate": 9.416679940295296e-06,
+      "loss": 0.0425,
+      "step": 833
+    },
+    {
+      "epoch": 0.3619791666666667,
+      "grad_norm": 0.7920419573783875,
+      "learning_rate": 9.415031286704253e-06,
+      "loss": 0.0584,
+      "step": 834
+    },
+    {
+      "epoch": 0.3624131944444444,
+      "grad_norm": 0.6306520700454712,
+      "learning_rate": 9.413380451323929e-06,
+      "loss": 0.0616,
+      "step": 835
+    },
+    {
+      "epoch": 0.3628472222222222,
+      "grad_norm": 0.8524011969566345,
+      "learning_rate": 9.411727434970121e-06,
+      "loss": 0.0443,
+      "step": 836
+    },
+    {
+      "epoch": 0.36328125,
+      "grad_norm": 0.9743005633354187,
+      "learning_rate": 9.410072238459709e-06,
+      "loss": 0.0522,
+      "step": 837
+    },
+    {
+      "epoch": 0.3637152777777778,
+      "grad_norm": 0.791600227355957,
+      "learning_rate": 9.408414862610645e-06,
+      "loss": 0.0479,
+      "step": 838
+    },
+    {
+      "epoch": 0.3641493055555556,
+      "grad_norm": 0.7155187726020813,
+      "learning_rate": 9.406755308241955e-06,
+      "loss": 0.0482,
+      "step": 839
+    },
+    {
+      "epoch": 0.3645833333333333,
+      "grad_norm": 0.7736154794692993,
+      "learning_rate": 9.405093576173752e-06,
+      "loss": 0.0328,
+      "step": 840
+    },
+    {
+      "epoch": 0.3650173611111111,
+      "grad_norm": 0.6123819351196289,
+      "learning_rate": 9.403429667227216e-06,
+      "loss": 0.0467,
+      "step": 841
+    },
+    {
+      "epoch": 0.3654513888888889,
+      "grad_norm": 0.43281838297843933,
+      "learning_rate": 9.401763582224606e-06,
+      "loss": 0.0335,
+      "step": 842
+    },
+    {
+      "epoch": 0.3658854166666667,
+      "grad_norm": 1.0776960849761963,
+      "learning_rate": 9.400095321989256e-06,
+      "loss": 0.0619,
+      "step": 843
+    },
+    {
+      "epoch": 0.3663194444444444,
+      "grad_norm": 1.0118237733840942,
+      "learning_rate": 9.398424887345577e-06,
+      "loss": 0.0668,
+      "step": 844
+    },
+    {
+      "epoch": 0.3667534722222222,
+      "grad_norm": 1.4794554710388184,
+      "learning_rate": 9.396752279119052e-06,
+      "loss": 0.0898,
+      "step": 845
+    },
+    {
+      "epoch": 0.3671875,
+      "grad_norm": 0.6297768354415894,
+      "learning_rate": 9.395077498136239e-06,
+      "loss": 0.0445,
+      "step": 846
+    },
+    {
+      "epoch": 0.3676215277777778,
+      "grad_norm": 0.8366924524307251,
+      "learning_rate": 9.393400545224766e-06,
+      "loss": 0.0539,
+      "step": 847
+    },
+    {
+      "epoch": 0.3680555555555556,
+      "grad_norm": 0.6405556797981262,
+      "learning_rate": 9.391721421213343e-06,
+      "loss": 0.0534,
+      "step": 848
+    },
+    {
+      "epoch": 0.3684895833333333,
+      "grad_norm": 0.711905300617218,
+      "learning_rate": 9.390040126931746e-06,
+      "loss": 0.0543,
+      "step": 849
+    },
+    {
+      "epoch": 0.3689236111111111,
+      "grad_norm": 0.6284782290458679,
+      "learning_rate": 9.388356663210824e-06,
+      "loss": 0.0685,
+      "step": 850
+    },
+    {
+      "epoch": 0.3693576388888889,
+      "grad_norm": 0.8485714793205261,
+      "learning_rate": 9.3866710308825e-06,
+      "loss": 0.0701,
+      "step": 851
+    },
+    {
+      "epoch": 0.3697916666666667,
+      "grad_norm": 0.847075879573822,
+      "learning_rate": 9.384983230779769e-06,
+      "loss": 0.0737,
+      "step": 852
+    },
+    {
+      "epoch": 0.3702256944444444,
+      "grad_norm": 0.5557371377944946,
+      "learning_rate": 9.383293263736694e-06,
+      "loss": 0.0513,
+      "step": 853
+    },
+    {
+      "epoch": 0.3706597222222222,
+      "grad_norm": 0.8466718196868896,
+      "learning_rate": 9.381601130588414e-06,
+      "loss": 0.0659,
+      "step": 854
+    },
+    {
+      "epoch": 0.37109375,
+      "grad_norm": 0.6541239023208618,
+      "learning_rate": 9.379906832171135e-06,
+      "loss": 0.0709,
+      "step": 855
+    },
+    {
+      "epoch": 0.3715277777777778,
+      "grad_norm": 0.604433536529541,
+      "learning_rate": 9.37821036932213e-06,
+      "loss": 0.0592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3719618055555556,
+      "grad_norm": 0.701170027256012,
+      "learning_rate": 9.376511742879749e-06,
+      "loss": 0.074,
+      "step": 857
+    },
+    {
+      "epoch": 0.3723958333333333,
+      "grad_norm": 0.7357682585716248,
+      "learning_rate": 9.374810953683407e-06,
+      "loss": 0.0583,
+      "step": 858
+    },
+    {
+      "epoch": 0.3728298611111111,
+      "grad_norm": 0.7110288143157959,
+      "learning_rate": 9.373108002573586e-06,
+      "loss": 0.0659,
+      "step": 859
+    },
+    {
+      "epoch": 0.3732638888888889,
+      "grad_norm": 0.5729966759681702,
+      "learning_rate": 9.371402890391839e-06,
+      "loss": 0.0611,
+      "step": 860
+    },
+    {
+      "epoch": 0.3736979166666667,
+      "grad_norm": 0.46971800923347473,
+      "learning_rate": 9.369695617980789e-06,
+      "loss": 0.0402,
+      "step": 861
+    },
+    {
+      "epoch": 0.3741319444444444,
+      "grad_norm": 0.7789297103881836,
+      "learning_rate": 9.36798618618412e-06,
+      "loss": 0.0589,
+      "step": 862
+    },
+    {
+      "epoch": 0.3745659722222222,
+      "grad_norm": 0.6817132234573364,
+      "learning_rate": 9.366274595846592e-06,
+      "loss": 0.0618,
+      "step": 863
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 0.9710595607757568,
+      "learning_rate": 9.36456084781402e-06,
+      "loss": 0.0818,
+      "step": 864
+    },
+    {
+      "epoch": 0.3754340277777778,
+      "grad_norm": 0.7942180037498474,
+      "learning_rate": 9.362844942933297e-06,
+      "loss": 0.0528,
+      "step": 865
+    },
+    {
+      "epoch": 0.3758680555555556,
+      "grad_norm": 0.5004309415817261,
+      "learning_rate": 9.361126882052378e-06,
+      "loss": 0.0429,
+      "step": 866
+    },
+    {
+      "epoch": 0.3763020833333333,
+      "grad_norm": 0.6871036291122437,
+      "learning_rate": 9.359406666020279e-06,
+      "loss": 0.0609,
+      "step": 867
+    },
+    {
+      "epoch": 0.3767361111111111,
+      "grad_norm": 0.5601425170898438,
+      "learning_rate": 9.357684295687084e-06,
+      "loss": 0.0543,
+      "step": 868
+    },
+    {
+      "epoch": 0.3771701388888889,
+      "grad_norm": 0.5933509469032288,
+      "learning_rate": 9.355959771903945e-06,
+      "loss": 0.0577,
+      "step": 869
+    },
+    {
+      "epoch": 0.3776041666666667,
+      "grad_norm": 0.7128217816352844,
+      "learning_rate": 9.354233095523071e-06,
+      "loss": 0.0635,
+      "step": 870
+    },
+    {
+      "epoch": 0.3780381944444444,
+      "grad_norm": 0.6243704557418823,
+      "learning_rate": 9.352504267397742e-06,
+      "loss": 0.0712,
+      "step": 871
+    },
+    {
+      "epoch": 0.3784722222222222,
+      "grad_norm": 0.5012828707695007,
+      "learning_rate": 9.350773288382297e-06,
+      "loss": 0.0405,
+      "step": 872
+    },
+    {
+      "epoch": 0.37890625,
+      "grad_norm": 0.5739850997924805,
+      "learning_rate": 9.349040159332141e-06,
+      "loss": 0.0549,
+      "step": 873
+    },
+    {
+      "epoch": 0.3793402777777778,
+      "grad_norm": 0.6758148074150085,
+      "learning_rate": 9.347304881103738e-06,
+      "loss": 0.0522,
+      "step": 874
+    },
+    {
+      "epoch": 0.3797743055555556,
+      "grad_norm": 0.5921206474304199,
+      "learning_rate": 9.345567454554614e-06,
+      "loss": 0.0523,
+      "step": 875
+    },
+    {
+      "epoch": 0.3802083333333333,
+      "grad_norm": 0.523183286190033,
+      "learning_rate": 9.343827880543362e-06,
+      "loss": 0.0388,
+      "step": 876
+    },
+    {
+      "epoch": 0.3806423611111111,
+      "grad_norm": 0.65218186378479,
+      "learning_rate": 9.342086159929629e-06,
+      "loss": 0.0501,
+      "step": 877
+    },
+    {
+      "epoch": 0.3810763888888889,
+      "grad_norm": 0.7650300860404968,
+      "learning_rate": 9.34034229357413e-06,
+      "loss": 0.0551,
+      "step": 878
+    },
+    {
+      "epoch": 0.3815104166666667,
+      "grad_norm": 0.9504824876785278,
+      "learning_rate": 9.338596282338632e-06,
+      "loss": 0.0323,
+      "step": 879
+    },
+    {
+      "epoch": 0.3819444444444444,
+      "grad_norm": 0.980655312538147,
+      "learning_rate": 9.336848127085972e-06,
+      "loss": 0.0538,
+      "step": 880
+    },
+    {
+      "epoch": 0.3823784722222222,
+      "grad_norm": 1.006734848022461,
+      "learning_rate": 9.335097828680035e-06,
+      "loss": 0.0497,
+      "step": 881
+    },
+    {
+      "epoch": 0.3828125,
+      "grad_norm": 1.2040218114852905,
+      "learning_rate": 9.333345387985777e-06,
+      "loss": 0.0711,
+      "step": 882
+    },
+    {
+      "epoch": 0.3832465277777778,
+      "grad_norm": 1.9897956848144531,
+      "learning_rate": 9.331590805869204e-06,
+      "loss": 0.0861,
+      "step": 883
+    },
+    {
+      "epoch": 0.3836805555555556,
+      "grad_norm": 1.4332646131515503,
+      "learning_rate": 9.329834083197382e-06,
+      "loss": 0.0663,
+      "step": 884
+    },
+    {
+      "epoch": 0.3841145833333333,
+      "grad_norm": 1.001236915588379,
+      "learning_rate": 9.32807522083844e-06,
+      "loss": 0.0723,
+      "step": 885
+    },
+    {
+      "epoch": 0.3845486111111111,
+      "grad_norm": 1.7063037157058716,
+      "learning_rate": 9.326314219661556e-06,
+      "loss": 0.0749,
+      "step": 886
+    },
+    {
+      "epoch": 0.3849826388888889,
+      "grad_norm": 0.6326516270637512,
+      "learning_rate": 9.324551080536972e-06,
+      "loss": 0.0539,
+      "step": 887
+    },
+    {
+      "epoch": 0.3854166666666667,
+      "grad_norm": 0.8508548736572266,
+      "learning_rate": 9.32278580433598e-06,
+      "loss": 0.046,
+      "step": 888
+    },
+    {
+      "epoch": 0.3858506944444444,
+      "grad_norm": 0.5909950137138367,
+      "learning_rate": 9.321018391930939e-06,
+      "loss": 0.0591,
+      "step": 889
+    },
+    {
+      "epoch": 0.3862847222222222,
+      "grad_norm": 0.4594472050666809,
+      "learning_rate": 9.319248844195248e-06,
+      "loss": 0.0428,
+      "step": 890
+    },
+    {
+      "epoch": 0.38671875,
+      "grad_norm": 0.42079484462738037,
+      "learning_rate": 9.317477162003376e-06,
+      "loss": 0.0352,
+      "step": 891
+    },
+    {
+      "epoch": 0.3871527777777778,
+      "grad_norm": 0.9136016964912415,
+      "learning_rate": 9.315703346230837e-06,
+      "loss": 0.0628,
+      "step": 892
+    },
+    {
+      "epoch": 0.3875868055555556,
+      "grad_norm": 0.7341428995132446,
+      "learning_rate": 9.313927397754205e-06,
+      "loss": 0.0535,
+      "step": 893
+    },
+    {
+      "epoch": 0.3880208333333333,
+      "grad_norm": 0.6203869581222534,
+      "learning_rate": 9.312149317451107e-06,
+      "loss": 0.0409,
+      "step": 894
+    },
+    {
+      "epoch": 0.3884548611111111,
+      "grad_norm": 0.4617060422897339,
+      "learning_rate": 9.310369106200218e-06,
+      "loss": 0.0214,
+      "step": 895
+    },
+    {
+      "epoch": 0.3888888888888889,
+      "grad_norm": 0.8191637992858887,
+      "learning_rate": 9.308586764881273e-06,
+      "loss": 0.0526,
+      "step": 896
+    },
+    {
+      "epoch": 0.3893229166666667,
+      "grad_norm": 1.3729549646377563,
+      "learning_rate": 9.306802294375056e-06,
+      "loss": 0.0779,
+      "step": 897
+    },
+    {
+      "epoch": 0.3897569444444444,
+      "grad_norm": 1.1696858406066895,
+      "learning_rate": 9.305015695563406e-06,
+      "loss": 0.0521,
+      "step": 898
+    },
+    {
+      "epoch": 0.3901909722222222,
+      "grad_norm": 1.3810319900512695,
+      "learning_rate": 9.30322696932921e-06,
+      "loss": 0.0716,
+      "step": 899
+    },
+    {
+      "epoch": 0.390625,
+      "grad_norm": 1.531559705734253,
+      "learning_rate": 9.30143611655641e-06,
+      "loss": 0.0768,
+      "step": 900
+    },
+    {
+      "epoch": 0.390625,
+      "eval_layout_rm_rank_test_accuracy": 0.9460030043441192,
+      "eval_layout_rm_rank_test_chosen_scores": 0.6126484870910645,
+      "eval_layout_rm_rank_test_loss": 0.06283911317586899,
+      "eval_layout_rm_rank_test_rejected_scores": 0.09528347104787827,
+      "eval_layout_rm_rank_test_runtime": 1176.1022,
+      "eval_layout_rm_rank_test_samples_per_second": 20.943,
+      "eval_layout_rm_rank_test_steps_per_second": 0.655,
+      "step": 900
+    },
+    {
+      "epoch": 0.3910590277777778,
+      "grad_norm": 1.5760308504104614,
+      "learning_rate": 9.299643138129995e-06,
+      "loss": 0.0661,
+      "step": 901
+    },
+    {
+      "epoch": 0.3914930555555556,
+      "grad_norm": 0.9271313548088074,
+      "learning_rate": 9.297848034936007e-06,
+      "loss": 0.0591,
+      "step": 902
+    },
+    {
+      "epoch": 0.3919270833333333,
+      "grad_norm": 1.0862689018249512,
+      "learning_rate": 9.296050807861538e-06,
+      "loss": 0.0826,
+      "step": 903
+    },
+    {
+      "epoch": 0.3923611111111111,
+      "grad_norm": 0.6337171196937561,
+      "learning_rate": 9.29425145779473e-06,
+      "loss": 0.0509,
+      "step": 904
+    },
+    {
+      "epoch": 0.3927951388888889,
+      "grad_norm": 0.7217728495597839,
+      "learning_rate": 9.292449985624771e-06,
+      "loss": 0.0407,
+      "step": 905
+    },
+    {
+      "epoch": 0.3932291666666667,
+      "grad_norm": 0.6203964948654175,
+      "learning_rate": 9.290646392241904e-06,
+      "loss": 0.0633,
+      "step": 906
+    },
+    {
+      "epoch": 0.3936631944444444,
+      "grad_norm": 0.4933103322982788,
+      "learning_rate": 9.288840678537412e-06,
+      "loss": 0.0455,
+      "step": 907
+    },
+    {
+      "epoch": 0.3940972222222222,
+      "grad_norm": 0.571537435054779,
+      "learning_rate": 9.28703284540363e-06,
+      "loss": 0.0584,
+      "step": 908
+    },
+    {
+      "epoch": 0.39453125,
+      "grad_norm": 0.4912552237510681,
+      "learning_rate": 9.285222893733943e-06,
+      "loss": 0.0415,
+      "step": 909
+    },
+    {
+      "epoch": 0.3949652777777778,
+      "grad_norm": 0.5689923167228699,
+      "learning_rate": 9.28341082442278e-06,
+      "loss": 0.0708,
+      "step": 910
+    },
+    {
+      "epoch": 0.3953993055555556,
+      "grad_norm": 0.49310195446014404,
+      "learning_rate": 9.281596638365615e-06,
+      "loss": 0.0424,
+      "step": 911
+    },
+    {
+      "epoch": 0.3958333333333333,
+      "grad_norm": 0.538809061050415,
+      "learning_rate": 9.27978033645897e-06,
+      "loss": 0.0513,
+      "step": 912
+    },
+    {
+      "epoch": 0.3962673611111111,
+      "grad_norm": 0.7360814213752747,
+      "learning_rate": 9.277961919600413e-06,
+      "loss": 0.0598,
+      "step": 913
+    },
+    {
+      "epoch": 0.3967013888888889,
+      "grad_norm": 0.6100806593894958,
+      "learning_rate": 9.276141388688557e-06,
+      "loss": 0.0347,
+      "step": 914
+    },
+    {
+      "epoch": 0.3971354166666667,
+      "grad_norm": 0.7792655229568481,
+      "learning_rate": 9.27431874462306e-06,
+      "loss": 0.0456,
+      "step": 915
+    },
+    {
+      "epoch": 0.3975694444444444,
+      "grad_norm": 1.1377533674240112,
+      "learning_rate": 9.272493988304621e-06,
+      "loss": 0.0721,
+      "step": 916
+    },
+    {
+      "epoch": 0.3980034722222222,
+      "grad_norm": 1.3492984771728516,
+      "learning_rate": 9.270667120634985e-06,
+      "loss": 0.0954,
+      "step": 917
+    },
+    {
+      "epoch": 0.3984375,
+      "grad_norm": 0.84878009557724,
+      "learning_rate": 9.268838142516943e-06,
+      "loss": 0.0449,
+      "step": 918
+    },
+    {
+      "epoch": 0.3988715277777778,
+      "grad_norm": 0.7643359899520874,
+      "learning_rate": 9.267007054854327e-06,
+      "loss": 0.0472,
+      "step": 919
+    },
+    {
+      "epoch": 0.3993055555555556,
+      "grad_norm": 0.8544363975524902,
+      "learning_rate": 9.26517385855201e-06,
+      "loss": 0.0617,
+      "step": 920
+    },
+    {
+      "epoch": 0.3997395833333333,
+      "grad_norm": 0.6072924733161926,
+      "learning_rate": 9.263338554515906e-06,
+      "loss": 0.0382,
+      "step": 921
+    },
+    {
+      "epoch": 0.4001736111111111,
+      "grad_norm": 0.9464804530143738,
+      "learning_rate": 9.261501143652975e-06,
+      "loss": 0.041,
+      "step": 922
+    },
+    {
+      "epoch": 0.4006076388888889,
+      "grad_norm": 0.6826421022415161,
+      "learning_rate": 9.259661626871217e-06,
+      "loss": 0.0559,
+      "step": 923
+    },
+    {
+      "epoch": 0.4010416666666667,
+      "grad_norm": 0.871494710445404,
+      "learning_rate": 9.25782000507967e-06,
+      "loss": 0.0757,
+      "step": 924
+    },
+    {
+      "epoch": 0.4014756944444444,
+      "grad_norm": 0.7773535251617432,
+      "learning_rate": 9.255976279188412e-06,
+      "loss": 0.0597,
+      "step": 925
+    },
+    {
+      "epoch": 0.4019097222222222,
+      "grad_norm": 0.6741392612457275,
+      "learning_rate": 9.254130450108567e-06,
+      "loss": 0.0409,
+      "step": 926
+    },
+    {
+      "epoch": 0.40234375,
+      "grad_norm": 0.9787711501121521,
+      "learning_rate": 9.25228251875229e-06,
+      "loss": 0.0507,
+      "step": 927
+    },
+    {
+      "epoch": 0.4027777777777778,
+      "grad_norm": 0.7004270553588867,
+      "learning_rate": 9.25043248603278e-06,
+      "loss": 0.0441,
+      "step": 928
+    },
+    {
+      "epoch": 0.4032118055555556,
+      "grad_norm": 1.1758992671966553,
+      "learning_rate": 9.248580352864273e-06,
+      "loss": 0.0761,
+      "step": 929
+    },
+    {
+      "epoch": 0.4036458333333333,
+      "grad_norm": 1.0420787334442139,
+      "learning_rate": 9.246726120162046e-06,
+      "loss": 0.0704,
+      "step": 930
+    },
+    {
+      "epoch": 0.4040798611111111,
+      "grad_norm": 0.7530882954597473,
+      "learning_rate": 9.244869788842406e-06,
+      "loss": 0.051,
+      "step": 931
+    },
+    {
+      "epoch": 0.4045138888888889,
+      "grad_norm": 0.8708016872406006,
+      "learning_rate": 9.243011359822704e-06,
+      "loss": 0.0427,
+      "step": 932
+    },
+    {
+      "epoch": 0.4049479166666667,
+      "grad_norm": 0.7921614646911621,
+      "learning_rate": 9.241150834021326e-06,
+      "loss": 0.0502,
+      "step": 933
+    },
+    {
+      "epoch": 0.4053819444444444,
+      "grad_norm": 0.6828484535217285,
+      "learning_rate": 9.239288212357694e-06,
+      "loss": 0.0558,
+      "step": 934
+    },
+    {
+      "epoch": 0.4058159722222222,
+      "grad_norm": 1.3188660144805908,
+      "learning_rate": 9.237423495752265e-06,
+      "loss": 0.0686,
+      "step": 935
+    },
+    {
+      "epoch": 0.40625,
+      "grad_norm": 1.3304487466812134,
+      "learning_rate": 9.23555668512653e-06,
+      "loss": 0.0532,
+      "step": 936
+    },
+    {
+      "epoch": 0.4066840277777778,
+      "grad_norm": 3.606907606124878,
+      "learning_rate": 9.233687781403017e-06,
+      "loss": 0.0708,
+      "step": 937
+    },
+    {
+      "epoch": 0.4071180555555556,
+      "grad_norm": 1.2716693878173828,
+      "learning_rate": 9.23181678550529e-06,
+      "loss": 0.101,
+      "step": 938
+    },
+    {
+      "epoch": 0.4075520833333333,
+      "grad_norm": 92.44752502441406,
+      "learning_rate": 9.229943698357942e-06,
+      "loss": 0.1493,
+      "step": 939
+    },
+    {
+      "epoch": 0.4079861111111111,
+      "grad_norm": 7.013112545013428,
+      "learning_rate": 9.228068520886604e-06,
+      "loss": 0.1524,
+      "step": 940
+    },
+    {
+      "epoch": 0.4084201388888889,
+      "grad_norm": 1.4616622924804688,
+      "learning_rate": 9.226191254017937e-06,
+      "loss": 0.1457,
+      "step": 941
+    },
+    {
+      "epoch": 0.4088541666666667,
+      "grad_norm": 78.1362533569336,
+      "learning_rate": 9.224311898679637e-06,
+      "loss": 0.1682,
+      "step": 942
+    },
+    {
+      "epoch": 0.4092881944444444,
+      "grad_norm": 34.54515075683594,
+      "learning_rate": 9.222430455800427e-06,
+      "loss": 0.1422,
+      "step": 943
+    },
+    {
+      "epoch": 0.4097222222222222,
+      "grad_norm": 39.45749282836914,
+      "learning_rate": 9.22054692631007e-06,
+      "loss": 0.1298,
+      "step": 944
+    },
+    {
+      "epoch": 0.41015625,
+      "grad_norm": 10.574701309204102,
+      "learning_rate": 9.218661311139352e-06,
+      "loss": 0.1894,
+      "step": 945
+    },
+    {
+      "epoch": 0.4105902777777778,
+      "grad_norm": 68.54234313964844,
+      "learning_rate": 9.216773611220094e-06,
+      "loss": 0.234,
+      "step": 946
+    },
+    {
+      "epoch": 0.4110243055555556,
+      "grad_norm": 48.71664810180664,
+      "learning_rate": 9.214883827485149e-06,
+      "loss": 0.256,
+      "step": 947
+    },
+    {
+      "epoch": 0.4114583333333333,
+      "grad_norm": 43.572425842285156,
+      "learning_rate": 9.212991960868392e-06,
+      "loss": 0.2555,
+      "step": 948
+    },
+    {
+      "epoch": 0.4118923611111111,
+      "grad_norm": 39.20548629760742,
+      "learning_rate": 9.211098012304736e-06,
+      "loss": 0.238,
+      "step": 949
+    },
+    {
+      "epoch": 0.4123263888888889,
+      "grad_norm": 67.0279769897461,
+      "learning_rate": 9.209201982730118e-06,
+      "loss": 0.3597,
+      "step": 950
+    },
+    {
+      "epoch": 0.4127604166666667,
+      "grad_norm": 43.72623062133789,
+      "learning_rate": 9.207303873081506e-06,
+      "loss": 0.3006,
+      "step": 951
+    },
+    {
+      "epoch": 0.4131944444444444,
+      "grad_norm": 11.67404556274414,
+      "learning_rate": 9.205403684296895e-06,
+      "loss": 0.2382,
+      "step": 952
+    },
+    {
+      "epoch": 0.4136284722222222,
+      "grad_norm": 11.699862480163574,
+      "learning_rate": 9.203501417315303e-06,
+      "loss": 0.2221,
+      "step": 953
+    },
+    {
+      "epoch": 0.4140625,
+      "grad_norm": 13.271985054016113,
+      "learning_rate": 9.201597073076784e-06,
+      "loss": 0.2537,
+      "step": 954
+    },
+    {
+      "epoch": 0.4144965277777778,
+      "grad_norm": 62.22166061401367,
+      "learning_rate": 9.199690652522409e-06,
+      "loss": 0.2715,
+      "step": 955
+    },
+    {
+      "epoch": 0.4149305555555556,
+      "grad_norm": 23.315994262695312,
+      "learning_rate": 9.197782156594284e-06,
+      "loss": 0.3128,
+      "step": 956
+    },
+    {
+      "epoch": 0.4153645833333333,
+      "grad_norm": 49.76742935180664,
+      "learning_rate": 9.195871586235531e-06,
+      "loss": 0.2271,
+      "step": 957
+    },
+    {
+      "epoch": 0.4157986111111111,
+      "grad_norm": 4.723781585693359,
+      "learning_rate": 9.19395894239031e-06,
+      "loss": 0.1952,
+      "step": 958
+    },
+    {
+      "epoch": 0.4162326388888889,
+      "grad_norm": 43.25361251831055,
+      "learning_rate": 9.19204422600379e-06,
+      "loss": 0.2309,
+      "step": 959
+    },
+    {
+      "epoch": 0.4166666666666667,
+      "grad_norm": 6.243020057678223,
+      "learning_rate": 9.190127438022175e-06,
+      "loss": 0.2332,
+      "step": 960
+    },
+    {
+      "epoch": 0.4171006944444444,
+      "grad_norm": 8.359853744506836,
+      "learning_rate": 9.18820857939269e-06,
+      "loss": 0.2122,
+      "step": 961
+    },
+    {
+      "epoch": 0.4175347222222222,
+      "grad_norm": 1.9266326427459717,
+      "learning_rate": 9.186287651063584e-06,
+      "loss": 0.1968,
+      "step": 962
+    },
+    {
+      "epoch": 0.41796875,
+      "grad_norm": 3.176651954650879,
+      "learning_rate": 9.184364653984126e-06,
+      "loss": 0.1657,
+      "step": 963
+    },
+    {
+      "epoch": 0.4184027777777778,
+      "grad_norm": 2.063450574874878,
+      "learning_rate": 9.182439589104612e-06,
+      "loss": 0.151,
+      "step": 964
+    },
+    {
+      "epoch": 0.4188368055555556,
+      "grad_norm": 27.922056198120117,
+      "learning_rate": 9.180512457376356e-06,
+      "loss": 0.1588,
+      "step": 965
+    },
+    {
+      "epoch": 0.4192708333333333,
+      "grad_norm": 4.232086181640625,
+      "learning_rate": 9.178583259751695e-06,
+      "loss": 0.1896,
+      "step": 966
+    },
+    {
+      "epoch": 0.4197048611111111,
+      "grad_norm": 4.373309135437012,
+      "learning_rate": 9.176651997183983e-06,
+      "loss": 0.1747,
+      "step": 967
+    },
+    {
+      "epoch": 0.4201388888888889,
+      "grad_norm": 3.191253900527954,
+      "learning_rate": 9.1747186706276e-06,
+      "loss": 0.1562,
+      "step": 968
+    },
+    {
+      "epoch": 0.4205729166666667,
+      "grad_norm": 8.11868667602539,
+      "learning_rate": 9.172783281037947e-06,
+      "loss": 0.1517,
+      "step": 969
+    },
+    {
+      "epoch": 0.4210069444444444,
+      "grad_norm": 28.279821395874023,
+      "learning_rate": 9.170845829371435e-06,
+      "loss": 0.1372,
+      "step": 970
+    },
+    {
+      "epoch": 0.4214409722222222,
+      "grad_norm": 23.073993682861328,
+      "learning_rate": 9.168906316585506e-06,
+      "loss": 0.1703,
+      "step": 971
+    },
+    {
+      "epoch": 0.421875,
+      "grad_norm": 6.043205261230469,
+      "learning_rate": 9.166964743638612e-06,
+      "loss": 0.15,
+      "step": 972
+    },
+    {
+      "epoch": 0.4223090277777778,
+      "grad_norm": 20.86329460144043,
+      "learning_rate": 9.165021111490226e-06,
+      "loss": 0.145,
+      "step": 973
+    },
+    {
+      "epoch": 0.4227430555555556,
+      "grad_norm": 7.663023471832275,
+      "learning_rate": 9.16307542110084e-06,
+      "loss": 0.123,
+      "step": 974
+    },
+    {
+      "epoch": 0.4231770833333333,
+      "grad_norm": 4.735728740692139,
+      "learning_rate": 9.16112767343196e-06,
+      "loss": 0.1135,
+      "step": 975
+    },
+    {
+      "epoch": 0.4236111111111111,
+      "grad_norm": 9.393878936767578,
+      "learning_rate": 9.159177869446112e-06,
+      "loss": 0.1371,
+      "step": 976
+    },
+    {
+      "epoch": 0.4240451388888889,
+      "grad_norm": 33.43635940551758,
+      "learning_rate": 9.157226010106835e-06,
+      "loss": 0.2043,
+      "step": 977
+    },
+    {
+      "epoch": 0.4244791666666667,
+      "grad_norm": 13.90460205078125,
+      "learning_rate": 9.155272096378685e-06,
+      "loss": 0.1522,
+      "step": 978
+    },
+    {
+      "epoch": 0.4249131944444444,
+      "grad_norm": 4.844347953796387,
+      "learning_rate": 9.153316129227234e-06,
+      "loss": 0.1117,
+      "step": 979
+    },
+    {
+      "epoch": 0.4253472222222222,
+      "grad_norm": 2.9909286499023438,
+      "learning_rate": 9.15135810961907e-06,
+      "loss": 0.0999,
+      "step": 980
+    },
+    {
+      "epoch": 0.42578125,
+      "grad_norm": 9.572826385498047,
+      "learning_rate": 9.14939803852179e-06,
+      "loss": 0.1549,
+      "step": 981
+    },
+    {
+      "epoch": 0.4262152777777778,
+      "grad_norm": 2.522773027420044,
+      "learning_rate": 9.147435916904011e-06,
+      "loss": 0.1092,
+      "step": 982
+    },
+    {
+      "epoch": 0.4266493055555556,
+      "grad_norm": 16.659290313720703,
+      "learning_rate": 9.145471745735359e-06,
+      "loss": 0.121,
+      "step": 983
+    },
+    {
+      "epoch": 0.4270833333333333,
+      "grad_norm": 4.65269136428833,
+      "learning_rate": 9.143505525986476e-06,
+      "loss": 0.1374,
+      "step": 984
+    },
+    {
+      "epoch": 0.4275173611111111,
+      "grad_norm": 1.8901381492614746,
+      "learning_rate": 9.141537258629012e-06,
+      "loss": 0.1186,
+      "step": 985
+    },
+    {
+      "epoch": 0.4279513888888889,
+      "grad_norm": 2.9920859336853027,
+      "learning_rate": 9.139566944635634e-06,
+      "loss": 0.1187,
+      "step": 986
+    },
+    {
+      "epoch": 0.4283854166666667,
+      "grad_norm": 1.7451281547546387,
+      "learning_rate": 9.137594584980017e-06,
+      "loss": 0.0959,
+      "step": 987
+    },
+    {
+      "epoch": 0.4288194444444444,
+      "grad_norm": 3.1266098022460938,
+      "learning_rate": 9.135620180636848e-06,
+      "loss": 0.1327,
+      "step": 988
+    },
+    {
+      "epoch": 0.4292534722222222,
+      "grad_norm": 1.3737506866455078,
+      "learning_rate": 9.133643732581824e-06,
+      "loss": 0.0942,
+      "step": 989
+    },
+    {
+      "epoch": 0.4296875,
+      "grad_norm": 1.38458251953125,
+      "learning_rate": 9.131665241791654e-06,
+      "loss": 0.0729,
+      "step": 990
+    },
+    {
+      "epoch": 0.4301215277777778,
+      "grad_norm": 1.934052586555481,
+      "learning_rate": 9.129684709244053e-06,
+      "loss": 0.0995,
+      "step": 991
+    },
+    {
+      "epoch": 0.4305555555555556,
+      "grad_norm": 21.256853103637695,
+      "learning_rate": 9.127702135917747e-06,
+      "loss": 0.0876,
+      "step": 992
+    },
+    {
+      "epoch": 0.4309895833333333,
+      "grad_norm": 7.968606472015381,
+      "learning_rate": 9.12571752279247e-06,
+      "loss": 0.0875,
+      "step": 993
+    },
+    {
+      "epoch": 0.4314236111111111,
+      "grad_norm": 6.5588250160217285,
+      "learning_rate": 9.123730870848963e-06,
+      "loss": 0.1262,
+      "step": 994
+    },
+    {
+      "epoch": 0.4318576388888889,
+      "grad_norm": 28.4091739654541,
+      "learning_rate": 9.12174218106898e-06,
+      "loss": 0.1219,
+      "step": 995
+    },
+    {
+      "epoch": 0.4322916666666667,
+      "grad_norm": 4.547117233276367,
+      "learning_rate": 9.119751454435275e-06,
+      "loss": 0.1213,
+      "step": 996
+    },
+    {
+      "epoch": 0.4327256944444444,
+      "grad_norm": 4.822508335113525,
+      "learning_rate": 9.11775869193161e-06,
+      "loss": 0.1223,
+      "step": 997
+    },
+    {
+      "epoch": 0.4331597222222222,
+      "grad_norm": 1.3423714637756348,
+      "learning_rate": 9.115763894542758e-06,
+      "loss": 0.0783,
+      "step": 998
+    },
+    {
+      "epoch": 0.43359375,
+      "grad_norm": 1.7178046703338623,
+      "learning_rate": 9.11376706325449e-06,
+      "loss": 0.0802,
+      "step": 999
+    },
+    {
+      "epoch": 0.4340277777777778,
+      "grad_norm": 5.980998992919922,
+      "learning_rate": 9.111768199053588e-06,
+      "loss": 0.0992,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4344618055555556,
+      "grad_norm": 1.5980966091156006,
+      "learning_rate": 9.109767302927839e-06,
+      "loss": 0.0947,
+      "step": 1001
+    },
+    {
+      "epoch": 0.4348958333333333,
+      "grad_norm": 1.964292049407959,
+      "learning_rate": 9.107764375866028e-06,
+      "loss": 0.1008,
+      "step": 1002
+    },
+    {
+      "epoch": 0.4353298611111111,
+      "grad_norm": 1.4555721282958984,
+      "learning_rate": 9.10575941885795e-06,
+      "loss": 0.084,
+      "step": 1003
+    },
+    {
+      "epoch": 0.4357638888888889,
+      "grad_norm": 1.8440698385238647,
+      "learning_rate": 9.103752432894398e-06,
+      "loss": 0.1092,
+      "step": 1004
+    },
+    {
+      "epoch": 0.4361979166666667,
+      "grad_norm": 10.594161033630371,
+      "learning_rate": 9.101743418967174e-06,
+      "loss": 0.0622,
+      "step": 1005
+    },
+    {
+      "epoch": 0.4366319444444444,
+      "grad_norm": 1.5675795078277588,
+      "learning_rate": 9.099732378069076e-06,
+      "loss": 0.1005,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4370659722222222,
+      "grad_norm": 1.1693589687347412,
+      "learning_rate": 9.097719311193907e-06,
+      "loss": 0.1029,
+      "step": 1007
+    },
+    {
+      "epoch": 0.4375,
+      "grad_norm": 0.9129657745361328,
+      "learning_rate": 9.09570421933647e-06,
+      "loss": 0.0799,
+      "step": 1008
+    },
+    {
+      "epoch": 0.4379340277777778,
+      "grad_norm": 0.9033944010734558,
+      "learning_rate": 9.093687103492568e-06,
+      "loss": 0.0671,
+      "step": 1009
+    },
+    {
+      "epoch": 0.4383680555555556,
+      "grad_norm": 1.1482619047164917,
+      "learning_rate": 9.091667964659006e-06,
+      "loss": 0.0788,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4388020833333333,
+      "grad_norm": 1.9808341264724731,
+      "learning_rate": 9.089646803833589e-06,
+      "loss": 0.0724,
+      "step": 1011
+    },
+    {
+      "epoch": 0.4392361111111111,
+      "grad_norm": 1.0761967897415161,
+      "learning_rate": 9.08762362201512e-06,
+      "loss": 0.0662,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4396701388888889,
+      "grad_norm": 1.2144149541854858,
+      "learning_rate": 9.085598420203399e-06,
+      "loss": 0.0727,
+      "step": 1013
+    },
+    {
+      "epoch": 0.4401041666666667,
+      "grad_norm": 0.7854107618331909,
+      "learning_rate": 9.083571199399226e-06,
+      "loss": 0.0791,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4405381944444444,
+      "grad_norm": 1.2169189453125,
+      "learning_rate": 9.0815419606044e-06,
+      "loss": 0.0784,
+      "step": 1015
+    },
+    {
+      "epoch": 0.4409722222222222,
+      "grad_norm": 0.8333733081817627,
+      "learning_rate": 9.079510704821718e-06,
+      "loss": 0.0771,
+      "step": 1016
+    },
+    {
+      "epoch": 0.44140625,
+      "grad_norm": 2.004037857055664,
+      "learning_rate": 9.077477433054969e-06,
+      "loss": 0.0582,
+      "step": 1017
+    },
+    {
+      "epoch": 0.4418402777777778,
+      "grad_norm": 1.4720275402069092,
+      "learning_rate": 9.075442146308942e-06,
+      "loss": 0.073,
+      "step": 1018
+    },
+    {
+      "epoch": 0.4422743055555556,
+      "grad_norm": 0.634091854095459,
+      "learning_rate": 9.07340484558942e-06,
+      "loss": 0.0397,
+      "step": 1019
+    },
+    {
+      "epoch": 0.4427083333333333,
+      "grad_norm": 1.0841357707977295,
+      "learning_rate": 9.071365531903182e-06,
+      "loss": 0.0938,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4431423611111111,
+      "grad_norm": 2.343041181564331,
+      "learning_rate": 9.069324206258003e-06,
+      "loss": 0.0891,
+      "step": 1021
+    },
+    {
+      "epoch": 0.4435763888888889,
+      "grad_norm": 1.1680972576141357,
+      "learning_rate": 9.06728086966265e-06,
+      "loss": 0.0546,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4440104166666667,
+      "grad_norm": 0.9939153790473938,
+      "learning_rate": 9.065235523126886e-06,
+      "loss": 0.0732,
+      "step": 1023
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 1.5492585897445679,
+      "learning_rate": 9.063188167661463e-06,
+      "loss": 0.0707,
+      "step": 1024
+    },
+    {
+      "epoch": 0.4448784722222222,
+      "grad_norm": 1.1362593173980713,
+      "learning_rate": 9.061138804278134e-06,
+      "loss": 0.0978,
+      "step": 1025
+    },
+    {
+      "epoch": 0.4453125,
+      "grad_norm": 1.016692876815796,
+      "learning_rate": 9.059087433989633e-06,
+      "loss": 0.0663,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4457465277777778,
+      "grad_norm": 1.037646770477295,
+      "learning_rate": 9.057034057809697e-06,
+      "loss": 0.0792,
+      "step": 1027
+    },
+    {
+      "epoch": 0.4461805555555556,
+      "grad_norm": 2.41023850440979,
+      "learning_rate": 9.054978676753045e-06,
+      "loss": 0.0951,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4466145833333333,
+      "grad_norm": 1.0070242881774902,
+      "learning_rate": 9.052921291835395e-06,
+      "loss": 0.0646,
+      "step": 1029
+    },
+    {
+      "epoch": 0.4470486111111111,
+      "grad_norm": 2.8583602905273438,
+      "learning_rate": 9.050861904073447e-06,
+      "loss": 0.082,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4474826388888889,
+      "grad_norm": 1.827767014503479,
+      "learning_rate": 9.048800514484897e-06,
+      "loss": 0.0818,
+      "step": 1031
+    },
+    {
+      "epoch": 0.4479166666666667,
+      "grad_norm": 2.483755350112915,
+      "learning_rate": 9.046737124088428e-06,
+      "loss": 0.0591,
+      "step": 1032
+    },
+    {
+      "epoch": 0.4483506944444444,
+      "grad_norm": 1.2786849737167358,
+      "learning_rate": 9.044671733903712e-06,
+      "loss": 0.0886,
+      "step": 1033
+    },
+    {
+      "epoch": 0.4487847222222222,
+      "grad_norm": 0.9632377624511719,
+      "learning_rate": 9.04260434495141e-06,
+      "loss": 0.0591,
+      "step": 1034
+    },
+    {
+      "epoch": 0.44921875,
+      "grad_norm": 1.3248785734176636,
+      "learning_rate": 9.040534958253166e-06,
+      "loss": 0.0667,
+      "step": 1035
+    },
+    {
+      "epoch": 0.4496527777777778,
+      "grad_norm": 0.7189981937408447,
+      "learning_rate": 9.038463574831617e-06,
+      "loss": 0.0641,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4500868055555556,
+      "grad_norm": 0.7724950313568115,
+      "learning_rate": 9.036390195710388e-06,
+      "loss": 0.0513,
+      "step": 1037
+    },
+    {
+      "epoch": 0.4505208333333333,
+      "grad_norm": 1.7117736339569092,
+      "learning_rate": 9.034314821914083e-06,
+      "loss": 0.0645,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4509548611111111,
+      "grad_norm": 32.858551025390625,
+      "learning_rate": 9.032237454468298e-06,
+      "loss": 0.0486,
+      "step": 1039
+    },
+    {
+      "epoch": 0.4513888888888889,
+      "grad_norm": 0.7045724391937256,
+      "learning_rate": 9.03015809439961e-06,
+      "loss": 0.0514,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4518229166666667,
+      "grad_norm": 0.6073102355003357,
+      "learning_rate": 9.028076742735583e-06,
+      "loss": 0.0487,
+      "step": 1041
+    },
+    {
+      "epoch": 0.4522569444444444,
+      "grad_norm": 1.0917977094650269,
+      "learning_rate": 9.025993400504766e-06,
+      "loss": 0.0608,
+      "step": 1042
+    },
+    {
+      "epoch": 0.4526909722222222,
+      "grad_norm": 4.855196475982666,
+      "learning_rate": 9.023908068736688e-06,
+      "loss": 0.0763,
+      "step": 1043
+    },
+    {
+      "epoch": 0.453125,
+      "grad_norm": 2.3733937740325928,
+      "learning_rate": 9.021820748461868e-06,
+      "loss": 0.077,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4535590277777778,
+      "grad_norm": 11.598405838012695,
+      "learning_rate": 9.0197314407118e-06,
+      "loss": 0.0901,
+      "step": 1045
+    },
+    {
+      "epoch": 0.4539930555555556,
+      "grad_norm": 0.9859849810600281,
+      "learning_rate": 9.017640146518964e-06,
+      "loss": 0.029,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4544270833333333,
+      "grad_norm": 0.6398284435272217,
+      "learning_rate": 9.015546866916823e-06,
+      "loss": 0.0509,
+      "step": 1047
+    },
+    {
+      "epoch": 0.4548611111111111,
+      "grad_norm": 0.8597634434700012,
+      "learning_rate": 9.013451602939818e-06,
+      "loss": 0.0555,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4552951388888889,
+      "grad_norm": 0.7557821869850159,
+      "learning_rate": 9.01135435562337e-06,
+      "loss": 0.0634,
+      "step": 1049
+    },
+    {
+      "epoch": 0.4557291666666667,
+      "grad_norm": 1.052837610244751,
+      "learning_rate": 9.009255126003883e-06,
+      "loss": 0.0639,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4561631944444444,
+      "grad_norm": 0.8531177043914795,
+      "learning_rate": 9.00715391511874e-06,
+      "loss": 0.0651,
+      "step": 1051
+    },
+    {
+      "epoch": 0.4565972222222222,
+      "grad_norm": 0.8589138984680176,
+      "learning_rate": 9.005050724006305e-06,
+      "loss": 0.0677,
+      "step": 1052
+    },
+    {
+      "epoch": 0.45703125,
+      "grad_norm": 0.5279788970947266,
+      "learning_rate": 9.002945553705915e-06,
+      "loss": 0.0518,
+      "step": 1053
+    },
+    {
+      "epoch": 0.4574652777777778,
+      "grad_norm": 0.5493040680885315,
+      "learning_rate": 9.00083840525789e-06,
+      "loss": 0.0452,
+      "step": 1054
+    },
+    {
+      "epoch": 0.4578993055555556,
+      "grad_norm": 0.4292406737804413,
+      "learning_rate": 8.998729279703527e-06,
+      "loss": 0.0248,
+      "step": 1055
+    },
+    {
+      "epoch": 0.4583333333333333,
+      "grad_norm": 0.5679145455360413,
+      "learning_rate": 8.996618178085095e-06,
+      "loss": 0.0414,
+      "step": 1056
+    },
+    {
+      "epoch": 0.4587673611111111,
+      "grad_norm": 0.8989630937576294,
+      "learning_rate": 8.994505101445845e-06,
+      "loss": 0.0517,
+      "step": 1057
+    },
+    {
+      "epoch": 0.4592013888888889,
+      "grad_norm": 0.522352933883667,
+      "learning_rate": 8.992390050830005e-06,
+      "loss": 0.0354,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4596354166666667,
+      "grad_norm": 0.7745173573493958,
+      "learning_rate": 8.990273027282774e-06,
+      "loss": 0.0589,
+      "step": 1059
+    },
+    {
+      "epoch": 0.4600694444444444,
+      "grad_norm": 1.2535181045532227,
+      "learning_rate": 8.988154031850328e-06,
+      "loss": 0.0539,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4605034722222222,
+      "grad_norm": 0.5835824012756348,
+      "learning_rate": 8.986033065579819e-06,
+      "loss": 0.035,
+      "step": 1061
+    },
+    {
+      "epoch": 0.4609375,
+      "grad_norm": 1.5330524444580078,
+      "learning_rate": 8.983910129519367e-06,
+      "loss": 0.0911,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4613715277777778,
+      "grad_norm": 0.8115177750587463,
+      "learning_rate": 8.981785224718073e-06,
+      "loss": 0.065,
+      "step": 1063
+    },
+    {
+      "epoch": 0.4618055555555556,
+      "grad_norm": 0.8075844049453735,
+      "learning_rate": 8.979658352226006e-06,
+      "loss": 0.0637,
+      "step": 1064
+    },
+    {
+      "epoch": 0.4622395833333333,
+      "grad_norm": 0.8678948879241943,
+      "learning_rate": 8.977529513094212e-06,
+      "loss": 0.0705,
+      "step": 1065
+    },
+    {
+      "epoch": 0.4626736111111111,
+      "grad_norm": 1.0560071468353271,
+      "learning_rate": 8.975398708374702e-06,
+      "loss": 0.072,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4631076388888889,
+      "grad_norm": 0.5890426635742188,
+      "learning_rate": 8.973265939120465e-06,
+      "loss": 0.0577,
+      "step": 1067
+    },
+    {
+      "epoch": 0.4635416666666667,
+      "grad_norm": 0.9163851737976074,
+      "learning_rate": 8.971131206385457e-06,
+      "loss": 0.0633,
+      "step": 1068
+    },
+    {
+      "epoch": 0.4639756944444444,
+      "grad_norm": 0.6653317809104919,
+      "learning_rate": 8.968994511224606e-06,
+      "loss": 0.0658,
+      "step": 1069
+    },
+    {
+      "epoch": 0.4644097222222222,
+      "grad_norm": 0.8337730169296265,
+      "learning_rate": 8.966855854693805e-06,
+      "loss": 0.0492,
+      "step": 1070
+    },
+    {
+      "epoch": 0.46484375,
+      "grad_norm": 0.7300952672958374,
+      "learning_rate": 8.964715237849928e-06,
+      "loss": 0.0733,
+      "step": 1071
+    },
+    {
+      "epoch": 0.4652777777777778,
+      "grad_norm": 1.1915796995162964,
+      "learning_rate": 8.962572661750802e-06,
+      "loss": 0.0896,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4657118055555556,
+      "grad_norm": 0.898739218711853,
+      "learning_rate": 8.960428127455236e-06,
+      "loss": 0.0527,
+      "step": 1073
+    },
+    {
+      "epoch": 0.4661458333333333,
+      "grad_norm": 0.6602903604507446,
+      "learning_rate": 8.958281636022998e-06,
+      "loss": 0.0511,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4665798611111111,
+      "grad_norm": 0.735146164894104,
+      "learning_rate": 8.956133188514827e-06,
+      "loss": 0.0372,
+      "step": 1075
+    },
+    {
+      "epoch": 0.4670138888888889,
+      "grad_norm": 0.6647898554801941,
+      "learning_rate": 8.953982785992428e-06,
+      "loss": 0.031,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4674479166666667,
+      "grad_norm": 1.3238353729248047,
+      "learning_rate": 8.951830429518471e-06,
+      "loss": 0.0795,
+      "step": 1077
+    },
+    {
+      "epoch": 0.4678819444444444,
+      "grad_norm": 0.9088767766952515,
+      "learning_rate": 8.949676120156593e-06,
+      "loss": 0.0487,
+      "step": 1078
+    },
+    {
+      "epoch": 0.4683159722222222,
+      "grad_norm": 0.882414698600769,
+      "learning_rate": 8.947519858971398e-06,
+      "loss": 0.0464,
+      "step": 1079
+    },
+    {
+      "epoch": 0.46875,
+      "grad_norm": 1.1095579862594604,
+      "learning_rate": 8.945361647028447e-06,
+      "loss": 0.0553,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4691840277777778,
+      "grad_norm": 1.1162978410720825,
+      "learning_rate": 8.943201485394276e-06,
+      "loss": 0.0579,
+      "step": 1081
+    },
+    {
+      "epoch": 0.4696180555555556,
+      "grad_norm": 0.7105571627616882,
+      "learning_rate": 8.94103937513637e-06,
+      "loss": 0.0349,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4700520833333333,
+      "grad_norm": 0.9360809326171875,
+      "learning_rate": 8.938875317323198e-06,
+      "loss": 0.058,
+      "step": 1083
+    },
+    {
+      "epoch": 0.4704861111111111,
+      "grad_norm": 0.7727855443954468,
+      "learning_rate": 8.936709313024165e-06,
+      "loss": 0.043,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4709201388888889,
+      "grad_norm": 0.773892879486084,
+      "learning_rate": 8.934541363309664e-06,
+      "loss": 0.0568,
+      "step": 1085
+    },
+    {
+      "epoch": 0.4713541666666667,
+      "grad_norm": 1.0830212831497192,
+      "learning_rate": 8.932371469251028e-06,
+      "loss": 0.0757,
+      "step": 1086
+    },
+    {
+      "epoch": 0.4717881944444444,
+      "grad_norm": 0.6006203889846802,
+      "learning_rate": 8.930199631920567e-06,
+      "loss": 0.044,
+      "step": 1087
+    },
+    {
+      "epoch": 0.4722222222222222,
+      "grad_norm": 0.7148776650428772,
+      "learning_rate": 8.928025852391539e-06,
+      "loss": 0.0493,
+      "step": 1088
+    },
+    {
+      "epoch": 0.47265625,
+      "grad_norm": 0.6793000102043152,
+      "learning_rate": 8.925850131738169e-06,
+      "loss": 0.0616,
+      "step": 1089
+    },
+    {
+      "epoch": 0.4730902777777778,
+      "grad_norm": 0.6601008772850037,
+      "learning_rate": 8.923672471035643e-06,
+      "loss": 0.0675,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4735243055555556,
+      "grad_norm": 0.603920578956604,
+      "learning_rate": 8.921492871360098e-06,
+      "loss": 0.0612,
+      "step": 1091
+    },
+    {
+      "epoch": 0.4739583333333333,
+      "grad_norm": 0.7035510540008545,
+      "learning_rate": 8.919311333788633e-06,
+      "loss": 0.0479,
+      "step": 1092
+    },
+    {
+      "epoch": 0.4743923611111111,
+      "grad_norm": 0.6087512373924255,
+      "learning_rate": 8.917127859399308e-06,
+      "loss": 0.0476,
+      "step": 1093
+    },
+    {
+      "epoch": 0.4748263888888889,
+      "grad_norm": 0.6310524940490723,
+      "learning_rate": 8.914942449271134e-06,
+      "loss": 0.0549,
+      "step": 1094
+    },
+    {
+      "epoch": 0.4752604166666667,
+      "grad_norm": 0.7847822904586792,
+      "learning_rate": 8.912755104484085e-06,
+      "loss": 0.0612,
+      "step": 1095
+    },
+    {
+      "epoch": 0.4756944444444444,
+      "grad_norm": 0.5045846700668335,
+      "learning_rate": 8.910565826119084e-06,
+      "loss": 0.0364,
+      "step": 1096
+    },
+    {
+      "epoch": 0.4761284722222222,
+      "grad_norm": 0.7675495743751526,
+      "learning_rate": 8.908374615258016e-06,
+      "loss": 0.062,
+      "step": 1097
+    },
+    {
+      "epoch": 0.4765625,
+      "grad_norm": 0.784545361995697,
+      "learning_rate": 8.906181472983717e-06,
+      "loss": 0.0458,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4769965277777778,
+      "grad_norm": 1.0218698978424072,
+      "learning_rate": 8.90398640037998e-06,
+      "loss": 0.0656,
+      "step": 1099
+    },
+    {
+      "epoch": 0.4774305555555556,
+      "grad_norm": 0.5361638069152832,
+      "learning_rate": 8.901789398531551e-06,
+      "loss": 0.0456,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4778645833333333,
+      "grad_norm": 0.6312048435211182,
+      "learning_rate": 8.899590468524125e-06,
+      "loss": 0.0485,
+      "step": 1101
+    },
+    {
+      "epoch": 0.4782986111111111,
+      "grad_norm": 0.8763497471809387,
+      "learning_rate": 8.897389611444359e-06,
+      "loss": 0.0427,
+      "step": 1102
+    },
+    {
+      "epoch": 0.4787326388888889,
+      "grad_norm": 0.9114484786987305,
+      "learning_rate": 8.895186828379855e-06,
+      "loss": 0.0651,
+      "step": 1103
+    },
+    {
+      "epoch": 0.4791666666666667,
+      "grad_norm": 1.0543360710144043,
+      "learning_rate": 8.892982120419166e-06,
+      "loss": 0.0765,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4796006944444444,
+      "grad_norm": 1.0119502544403076,
+      "learning_rate": 8.890775488651804e-06,
+      "loss": 0.0747,
+      "step": 1105
+    },
+    {
+      "epoch": 0.4800347222222222,
+      "grad_norm": 0.5783020257949829,
+      "learning_rate": 8.888566934168223e-06,
+      "loss": 0.0371,
+      "step": 1106
+    },
+    {
+      "epoch": 0.48046875,
+      "grad_norm": 0.5611697435379028,
+      "learning_rate": 8.886356458059836e-06,
+      "loss": 0.0469,
+      "step": 1107
+    },
+    {
+      "epoch": 0.4809027777777778,
+      "grad_norm": 1.545137882232666,
+      "learning_rate": 8.884144061418993e-06,
+      "loss": 0.0574,
+      "step": 1108
+    },
+    {
+      "epoch": 0.4813368055555556,
+      "grad_norm": 0.4729805290699005,
+      "learning_rate": 8.881929745339008e-06,
+      "loss": 0.041,
+      "step": 1109
+    },
+    {
+      "epoch": 0.4817708333333333,
+      "grad_norm": 0.6251962780952454,
+      "learning_rate": 8.879713510914132e-06,
+      "loss": 0.0535,
+      "step": 1110
+    },
+    {
+      "epoch": 0.4822048611111111,
+      "grad_norm": 0.5359340906143188,
+      "learning_rate": 8.87749535923957e-06,
+      "loss": 0.0436,
+      "step": 1111
+    },
+    {
+      "epoch": 0.4826388888888889,
+      "grad_norm": 0.6215574741363525,
+      "learning_rate": 8.875275291411473e-06,
+      "loss": 0.0552,
+      "step": 1112
+    },
+    {
+      "epoch": 0.4830729166666667,
+      "grad_norm": 0.5203942656517029,
+      "learning_rate": 8.873053308526938e-06,
+      "loss": 0.0617,
+      "step": 1113
+    },
+    {
+      "epoch": 0.4835069444444444,
+      "grad_norm": 0.4644133150577545,
+      "learning_rate": 8.870829411684007e-06,
+      "loss": 0.031,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4839409722222222,
+      "grad_norm": 0.5880697965621948,
+      "learning_rate": 8.868603601981672e-06,
+      "loss": 0.0415,
+      "step": 1115
+    },
+    {
+      "epoch": 0.484375,
+      "grad_norm": 1.0086796283721924,
+      "learning_rate": 8.866375880519868e-06,
+      "loss": 0.0526,
+      "step": 1116
+    },
+    {
+      "epoch": 0.4848090277777778,
+      "grad_norm": 1.1692084074020386,
+      "learning_rate": 8.864146248399476e-06,
+      "loss": 0.0713,
+      "step": 1117
+    },
+    {
+      "epoch": 0.4852430555555556,
+      "grad_norm": 0.7209148406982422,
+      "learning_rate": 8.861914706722316e-06,
+      "loss": 0.0501,
+      "step": 1118
+    },
+    {
+      "epoch": 0.4856770833333333,
+      "grad_norm": 0.5479106903076172,
+      "learning_rate": 8.85968125659116e-06,
+      "loss": 0.0426,
+      "step": 1119
+    },
+    {
+      "epoch": 0.4861111111111111,
+      "grad_norm": 0.8221974968910217,
+      "learning_rate": 8.857445899109716e-06,
+      "loss": 0.0591,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4865451388888889,
+      "grad_norm": 0.5368654131889343,
+      "learning_rate": 8.855208635382637e-06,
+      "loss": 0.0327,
+      "step": 1121
+    },
+    {
+      "epoch": 0.4869791666666667,
+      "grad_norm": 0.794711172580719,
+      "learning_rate": 8.852969466515522e-06,
+      "loss": 0.0695,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4874131944444444,
+      "grad_norm": 0.8020004034042358,
+      "learning_rate": 8.850728393614903e-06,
+      "loss": 0.0524,
+      "step": 1123
+    },
+    {
+      "epoch": 0.4878472222222222,
+      "grad_norm": 0.6102003455162048,
+      "learning_rate": 8.848485417788262e-06,
+      "loss": 0.0399,
+      "step": 1124
+    },
+    {
+      "epoch": 0.48828125,
+      "grad_norm": 0.7068143486976624,
+      "learning_rate": 8.846240540144013e-06,
+      "loss": 0.0561,
+      "step": 1125
+    },
+    {
+      "epoch": 0.4887152777777778,
+      "grad_norm": 0.9423394203186035,
+      "learning_rate": 8.843993761791518e-06,
+      "loss": 0.0814,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4891493055555556,
+      "grad_norm": 0.5660754442214966,
+      "learning_rate": 8.841745083841071e-06,
+      "loss": 0.047,
+      "step": 1127
+    },
+    {
+      "epoch": 0.4895833333333333,
+      "grad_norm": 0.5480889678001404,
+      "learning_rate": 8.839494507403909e-06,
+      "loss": 0.0481,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4900173611111111,
+      "grad_norm": 0.6959866881370544,
+      "learning_rate": 8.837242033592208e-06,
+      "loss": 0.0436,
+      "step": 1129
+    },
+    {
+      "epoch": 0.4904513888888889,
+      "grad_norm": 0.867567777633667,
+      "learning_rate": 8.834987663519079e-06,
+      "loss": 0.0702,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4908854166666667,
+      "grad_norm": 0.6856876015663147,
+      "learning_rate": 8.832731398298567e-06,
+      "loss": 0.0743,
+      "step": 1131
+    },
+    {
+      "epoch": 0.4913194444444444,
+      "grad_norm": 0.5049101114273071,
+      "learning_rate": 8.830473239045663e-06,
+      "loss": 0.0554,
+      "step": 1132
+    },
+    {
+      "epoch": 0.4917534722222222,
+      "grad_norm": 0.5671724081039429,
+      "learning_rate": 8.828213186876283e-06,
+      "loss": 0.0438,
+      "step": 1133
+    },
+    {
+      "epoch": 0.4921875,
+      "grad_norm": 0.6206570863723755,
+      "learning_rate": 8.825951242907288e-06,
+      "loss": 0.0399,
+      "step": 1134
+    },
+    {
+      "epoch": 0.4926215277777778,
+      "grad_norm": 0.5550037622451782,
+      "learning_rate": 8.823687408256468e-06,
+      "loss": 0.0433,
+      "step": 1135
+    },
+    {
+      "epoch": 0.4930555555555556,
+      "grad_norm": 0.8700177669525146,
+      "learning_rate": 8.82142168404255e-06,
+      "loss": 0.0727,
+      "step": 1136
+    },
+    {
+      "epoch": 0.4934895833333333,
+      "grad_norm": 0.9669756293296814,
+      "learning_rate": 8.81915407138519e-06,
+      "loss": 0.0738,
+      "step": 1137
+    },
+    {
+      "epoch": 0.4939236111111111,
+      "grad_norm": 0.7427315711975098,
+      "learning_rate": 8.816884571404987e-06,
+      "loss": 0.048,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4943576388888889,
+      "grad_norm": 1.1001490354537964,
+      "learning_rate": 8.814613185223462e-06,
+      "loss": 0.0752,
+      "step": 1139
+    },
+    {
+      "epoch": 0.4947916666666667,
+      "grad_norm": 0.7257412075996399,
+      "learning_rate": 8.812339913963074e-06,
+      "loss": 0.0563,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4952256944444444,
+      "grad_norm": 1.2158013582229614,
+      "learning_rate": 8.810064758747212e-06,
+      "loss": 0.0592,
+      "step": 1141
+    },
+    {
+      "epoch": 0.4956597222222222,
+      "grad_norm": 0.4755469262599945,
+      "learning_rate": 8.807787720700199e-06,
+      "loss": 0.0584,
+      "step": 1142
+    },
+    {
+      "epoch": 0.49609375,
+      "grad_norm": 0.4486525058746338,
+      "learning_rate": 8.80550880094728e-06,
+      "loss": 0.0314,
+      "step": 1143
+    },
+    {
+      "epoch": 0.4965277777777778,
+      "grad_norm": 0.6543397307395935,
+      "learning_rate": 8.803228000614642e-06,
+      "loss": 0.0534,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4969618055555556,
+      "grad_norm": 0.8404059410095215,
+      "learning_rate": 8.800945320829391e-06,
+      "loss": 0.0746,
+      "step": 1145
+    },
+    {
+      "epoch": 0.4973958333333333,
+      "grad_norm": 0.7537004947662354,
+      "learning_rate": 8.798660762719567e-06,
+      "loss": 0.0408,
+      "step": 1146
+    },
+    {
+      "epoch": 0.4978298611111111,
+      "grad_norm": 0.8327997922897339,
+      "learning_rate": 8.796374327414135e-06,
+      "loss": 0.0513,
+      "step": 1147
+    },
+    {
+      "epoch": 0.4982638888888889,
+      "grad_norm": 0.8455732464790344,
+      "learning_rate": 8.794086016042993e-06,
+      "loss": 0.0599,
+      "step": 1148
+    },
+    {
+      "epoch": 0.4986979166666667,
+      "grad_norm": 0.5204771757125854,
+      "learning_rate": 8.79179582973696e-06,
+      "loss": 0.0413,
+      "step": 1149
+    },
+    {
+      "epoch": 0.4991319444444444,
+      "grad_norm": 0.8038812875747681,
+      "learning_rate": 8.789503769627786e-06,
+      "loss": 0.0552,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4995659722222222,
+      "grad_norm": 0.781603217124939,
+      "learning_rate": 8.787209836848144e-06,
+      "loss": 0.0594,
+      "step": 1151
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.508056640625,
+      "learning_rate": 8.784914032531636e-06,
+      "loss": 0.0434,
+      "step": 1152
+    },
+    {
+      "epoch": 0.5004340277777778,
+      "grad_norm": 0.5347216725349426,
+      "learning_rate": 8.782616357812784e-06,
+      "loss": 0.0614,
+      "step": 1153
+    },
+    {
+      "epoch": 0.5008680555555556,
+      "grad_norm": 0.7684710621833801,
+      "learning_rate": 8.780316813827037e-06,
+      "loss": 0.0601,
+      "step": 1154
+    },
+    {
+      "epoch": 0.5013020833333334,
+      "grad_norm": 0.625655472278595,
+      "learning_rate": 8.778015401710766e-06,
+      "loss": 0.0524,
+      "step": 1155
+    },
+    {
+      "epoch": 0.5017361111111112,
+      "grad_norm": 1.0762808322906494,
+      "learning_rate": 8.775712122601273e-06,
+      "loss": 0.0763,
+      "step": 1156
+    },
+    {
+      "epoch": 0.5021701388888888,
+      "grad_norm": 0.592757523059845,
+      "learning_rate": 8.773406977636769e-06,
+      "loss": 0.0445,
+      "step": 1157
+    },
+    {
+      "epoch": 0.5026041666666666,
+      "grad_norm": 0.8303428292274475,
+      "learning_rate": 8.771099967956398e-06,
+      "loss": 0.0608,
+      "step": 1158
+    },
+    {
+      "epoch": 0.5030381944444444,
+      "grad_norm": 1.07639479637146,
+      "learning_rate": 8.768791094700222e-06,
+      "loss": 0.0608,
+      "step": 1159
+    },
+    {
+      "epoch": 0.5034722222222222,
+      "grad_norm": 4.158738613128662,
+      "learning_rate": 8.766480359009224e-06,
+      "loss": 0.065,
+      "step": 1160
+    },
+    {
+      "epoch": 0.50390625,
+      "grad_norm": 0.5527358055114746,
+      "learning_rate": 8.764167762025306e-06,
+      "loss": 0.0453,
+      "step": 1161
+    },
+    {
+      "epoch": 0.5043402777777778,
+      "grad_norm": 0.7811183929443359,
+      "learning_rate": 8.761853304891293e-06,
+      "loss": 0.0733,
+      "step": 1162
+    },
+    {
+      "epoch": 0.5047743055555556,
+      "grad_norm": 0.8103577494621277,
+      "learning_rate": 8.759536988750922e-06,
+      "loss": 0.0657,
+      "step": 1163
+    },
+    {
+      "epoch": 0.5052083333333334,
+      "grad_norm": 0.6777807474136353,
+      "learning_rate": 8.75721881474886e-06,
+      "loss": 0.0614,
+      "step": 1164
+    },
+    {
+      "epoch": 0.5056423611111112,
+      "grad_norm": 0.7982746362686157,
+      "learning_rate": 8.754898784030685e-06,
+      "loss": 0.0604,
+      "step": 1165
+    },
+    {
+      "epoch": 0.5060763888888888,
+      "grad_norm": 0.5058833360671997,
+      "learning_rate": 8.75257689774289e-06,
+      "loss": 0.0461,
+      "step": 1166
+    },
+    {
+      "epoch": 0.5065104166666666,
+      "grad_norm": 0.5670070648193359,
+      "learning_rate": 8.750253157032893e-06,
+      "loss": 0.0578,
+      "step": 1167
+    },
+    {
+      "epoch": 0.5069444444444444,
+      "grad_norm": 0.5332913398742676,
+      "learning_rate": 8.74792756304902e-06,
+      "loss": 0.0598,
+      "step": 1168
+    },
+    {
+      "epoch": 0.5073784722222222,
+      "grad_norm": 0.6165770888328552,
+      "learning_rate": 8.745600116940518e-06,
+      "loss": 0.0678,
+      "step": 1169
+    },
+    {
+      "epoch": 0.5078125,
+      "grad_norm": 0.5424283146858215,
+      "learning_rate": 8.74327081985755e-06,
+      "loss": 0.0496,
+      "step": 1170
+    },
+    {
+      "epoch": 0.5082465277777778,
+      "grad_norm": 0.7060388326644897,
+      "learning_rate": 8.74093967295119e-06,
+      "loss": 0.0462,
+      "step": 1171
+    },
+    {
+      "epoch": 0.5086805555555556,
+      "grad_norm": 0.7423508167266846,
+      "learning_rate": 8.738606677373426e-06,
+      "loss": 0.0543,
+      "step": 1172
+    },
+    {
+      "epoch": 0.5091145833333334,
+      "grad_norm": 0.5764522552490234,
+      "learning_rate": 8.736271834277167e-06,
+      "loss": 0.0493,
+      "step": 1173
+    },
+    {
+      "epoch": 0.5095486111111112,
+      "grad_norm": 0.8961094617843628,
+      "learning_rate": 8.733935144816224e-06,
+      "loss": 0.0708,
+      "step": 1174
+    },
+    {
+      "epoch": 0.5099826388888888,
+      "grad_norm": 1.347813367843628,
+      "learning_rate": 8.731596610145327e-06,
+      "loss": 0.082,
+      "step": 1175
+    },
+    {
+      "epoch": 0.5104166666666666,
+      "grad_norm": 0.664887011051178,
+      "learning_rate": 8.72925623142012e-06,
+      "loss": 0.0464,
+      "step": 1176
+    },
+    {
+      "epoch": 0.5108506944444444,
+      "grad_norm": 0.4864520728588104,
+      "learning_rate": 8.726914009797152e-06,
+      "loss": 0.0359,
+      "step": 1177
+    },
+    {
+      "epoch": 0.5112847222222222,
+      "grad_norm": 0.5455866456031799,
+      "learning_rate": 8.724569946433884e-06,
+      "loss": 0.0514,
+      "step": 1178
+    },
+    {
+      "epoch": 0.51171875,
+      "grad_norm": 0.5329997539520264,
+      "learning_rate": 8.722224042488694e-06,
+      "loss": 0.0435,
+      "step": 1179
+    },
+    {
+      "epoch": 0.5121527777777778,
+      "grad_norm": 0.6375029683113098,
+      "learning_rate": 8.71987629912086e-06,
+      "loss": 0.0666,
+      "step": 1180
+    },
+    {
+      "epoch": 0.5125868055555556,
+      "grad_norm": 0.6707751750946045,
+      "learning_rate": 8.717526717490575e-06,
+      "loss": 0.0605,
+      "step": 1181
+    },
+    {
+      "epoch": 0.5130208333333334,
+      "grad_norm": 0.7720732688903809,
+      "learning_rate": 8.715175298758938e-06,
+      "loss": 0.0455,
+      "step": 1182
+    },
+    {
+      "epoch": 0.5134548611111112,
+      "grad_norm": 0.6852914094924927,
+      "learning_rate": 8.712822044087959e-06,
+      "loss": 0.0661,
+      "step": 1183
+    },
+    {
+      "epoch": 0.5138888888888888,
+      "grad_norm": 0.6983592510223389,
+      "learning_rate": 8.710466954640551e-06,
+      "loss": 0.0494,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5143229166666666,
+      "grad_norm": 1.2967052459716797,
+      "learning_rate": 8.708110031580534e-06,
+      "loss": 0.0837,
+      "step": 1185
+    },
+    {
+      "epoch": 0.5147569444444444,
+      "grad_norm": 0.42146065831184387,
+      "learning_rate": 8.705751276072639e-06,
+      "loss": 0.0308,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5151909722222222,
+      "grad_norm": 1.0599000453948975,
+      "learning_rate": 8.703390689282498e-06,
+      "loss": 0.0834,
+      "step": 1187
+    },
+    {
+      "epoch": 0.515625,
+      "grad_norm": 0.8723039031028748,
+      "learning_rate": 8.70102827237665e-06,
+      "loss": 0.0641,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5160590277777778,
+      "grad_norm": 0.8208059668540955,
+      "learning_rate": 8.698664026522536e-06,
+      "loss": 0.07,
+      "step": 1189
+    },
+    {
+      "epoch": 0.5164930555555556,
+      "grad_norm": 0.5073002576828003,
+      "learning_rate": 8.696297952888503e-06,
+      "loss": 0.0469,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5169270833333334,
+      "grad_norm": 0.7411544322967529,
+      "learning_rate": 8.693930052643803e-06,
+      "loss": 0.0574,
+      "step": 1191
+    },
+    {
+      "epoch": 0.5173611111111112,
+      "grad_norm": 0.5245022773742676,
+      "learning_rate": 8.691560326958584e-06,
+      "loss": 0.0458,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5177951388888888,
+      "grad_norm": 0.6204422116279602,
+      "learning_rate": 8.689188777003906e-06,
+      "loss": 0.0533,
+      "step": 1193
+    },
+    {
+      "epoch": 0.5182291666666666,
+      "grad_norm": 0.9399925470352173,
+      "learning_rate": 8.686815403951719e-06,
+      "loss": 0.0675,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5186631944444444,
+      "grad_norm": 0.6003794074058533,
+      "learning_rate": 8.684440208974884e-06,
+      "loss": 0.0674,
+      "step": 1195
+    },
+    {
+      "epoch": 0.5190972222222222,
+      "grad_norm": 0.580943763256073,
+      "learning_rate": 8.682063193247157e-06,
+      "loss": 0.045,
+      "step": 1196
+    },
+    {
+      "epoch": 0.51953125,
+      "grad_norm": 0.5490971803665161,
+      "learning_rate": 8.679684357943195e-06,
+      "loss": 0.0505,
+      "step": 1197
+    },
+    {
+      "epoch": 0.5199652777777778,
+      "grad_norm": 0.6057333946228027,
+      "learning_rate": 8.677303704238553e-06,
+      "loss": 0.0453,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5203993055555556,
+      "grad_norm": 0.5196250081062317,
+      "learning_rate": 8.674921233309689e-06,
+      "loss": 0.0402,
+      "step": 1199
+    },
+    {
+      "epoch": 0.5208333333333334,
+      "grad_norm": 0.6459466814994812,
+      "learning_rate": 8.672536946333954e-06,
+      "loss": 0.0427,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5208333333333334,
+      "eval_layout_rm_rank_test_accuracy": 0.9691039746660712,
+      "eval_layout_rm_rank_test_chosen_scores": 0.5440658926963806,
+      "eval_layout_rm_rank_test_loss": 0.052996955811977386,
+      "eval_layout_rm_rank_test_rejected_scores": 0.07446130365133286,
+      "eval_layout_rm_rank_test_runtime": 1175.8343,
+      "eval_layout_rm_rank_test_samples_per_second": 20.948,
+      "eval_layout_rm_rank_test_steps_per_second": 0.655,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5212673611111112,
+      "grad_norm": 1.0727156400680542,
+      "learning_rate": 8.670150844489597e-06,
+      "loss": 0.0927,
+      "step": 1201
+    },
+    {
+      "epoch": 0.5217013888888888,
+      "grad_norm": 0.6895102262496948,
+      "learning_rate": 8.667762928955766e-06,
+      "loss": 0.0402,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5221354166666666,
+      "grad_norm": 1.0803653001785278,
+      "learning_rate": 8.665373200912506e-06,
+      "loss": 0.0585,
+      "step": 1203
+    },
+    {
+      "epoch": 0.5225694444444444,
+      "grad_norm": 1.0241613388061523,
+      "learning_rate": 8.662981661540753e-06,
+      "loss": 0.0748,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5230034722222222,
+      "grad_norm": 1.1917078495025635,
+      "learning_rate": 8.660588312022345e-06,
+      "loss": 0.0787,
+      "step": 1205
+    },
+    {
+      "epoch": 0.5234375,
+      "grad_norm": 0.8959947824478149,
+      "learning_rate": 8.658193153540006e-06,
+      "loss": 0.0604,
+      "step": 1206
+    },
+    {
+      "epoch": 0.5238715277777778,
+      "grad_norm": 0.5525012612342834,
+      "learning_rate": 8.655796187277364e-06,
+      "loss": 0.0398,
+      "step": 1207
+    },
+    {
+      "epoch": 0.5243055555555556,
+      "grad_norm": 0.6186550855636597,
+      "learning_rate": 8.65339741441893e-06,
+      "loss": 0.0442,
+      "step": 1208
+    },
+    {
+      "epoch": 0.5247395833333334,
+      "grad_norm": 0.5443890690803528,
+      "learning_rate": 8.650996836150117e-06,
+      "loss": 0.0361,
+      "step": 1209
+    },
+    {
+      "epoch": 0.5251736111111112,
+      "grad_norm": 0.6587841510772705,
+      "learning_rate": 8.648594453657224e-06,
+      "loss": 0.0485,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5256076388888888,
+      "grad_norm": 0.6534464359283447,
+      "learning_rate": 8.646190268127444e-06,
+      "loss": 0.0595,
+      "step": 1211
+    },
+    {
+      "epoch": 0.5260416666666666,
+      "grad_norm": 0.5813679695129395,
+      "learning_rate": 8.64378428074886e-06,
+      "loss": 0.0537,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5264756944444444,
+      "grad_norm": 0.6579888463020325,
+      "learning_rate": 8.641376492710445e-06,
+      "loss": 0.0393,
+      "step": 1213
+    },
+    {
+      "epoch": 0.5269097222222222,
+      "grad_norm": 0.6142035126686096,
+      "learning_rate": 8.638966905202064e-06,
+      "loss": 0.0438,
+      "step": 1214
+    },
+    {
+      "epoch": 0.52734375,
+      "grad_norm": 0.6465262770652771,
+      "learning_rate": 8.63655551941447e-06,
+      "loss": 0.0417,
+      "step": 1215
+    },
+    {
+      "epoch": 0.5277777777777778,
+      "grad_norm": 0.8512789607048035,
+      "learning_rate": 8.634142336539308e-06,
+      "loss": 0.0578,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5282118055555556,
+      "grad_norm": 0.868228554725647,
+      "learning_rate": 8.631727357769103e-06,
+      "loss": 0.0677,
+      "step": 1217
+    },
+    {
+      "epoch": 0.5286458333333334,
+      "grad_norm": 1.382534384727478,
+      "learning_rate": 8.629310584297274e-06,
+      "loss": 0.072,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5290798611111112,
+      "grad_norm": 0.895889937877655,
+      "learning_rate": 8.626892017318127e-06,
+      "loss": 0.0639,
+      "step": 1219
+    },
+    {
+      "epoch": 0.5295138888888888,
+      "grad_norm": 1.282185435295105,
+      "learning_rate": 8.624471658026848e-06,
+      "loss": 0.0581,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5299479166666666,
+      "grad_norm": 0.7685604095458984,
+      "learning_rate": 8.622049507619519e-06,
+      "loss": 0.0589,
+      "step": 1221
+    },
+    {
+      "epoch": 0.5303819444444444,
+      "grad_norm": 0.9641804099082947,
+      "learning_rate": 8.619625567293096e-06,
+      "loss": 0.0525,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5308159722222222,
+      "grad_norm": 0.6586122512817383,
+      "learning_rate": 8.61719983824543e-06,
+      "loss": 0.0557,
+      "step": 1223
+    },
+    {
+      "epoch": 0.53125,
+      "grad_norm": 0.625971257686615,
+      "learning_rate": 8.614772321675245e-06,
+      "loss": 0.0574,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5316840277777778,
+      "grad_norm": 0.6470603346824646,
+      "learning_rate": 8.61234301878216e-06,
+      "loss": 0.0594,
+      "step": 1225
+    },
+    {
+      "epoch": 0.5321180555555556,
+      "grad_norm": 0.600440502166748,
+      "learning_rate": 8.609911930766668e-06,
+      "loss": 0.0591,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5325520833333334,
+      "grad_norm": 0.5906641483306885,
+      "learning_rate": 8.607479058830144e-06,
+      "loss": 0.0453,
+      "step": 1227
+    },
+    {
+      "epoch": 0.5329861111111112,
+      "grad_norm": 0.800161600112915,
+      "learning_rate": 8.605044404174854e-06,
+      "loss": 0.0521,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5334201388888888,
+      "grad_norm": 0.7596476078033447,
+      "learning_rate": 8.602607968003935e-06,
+      "loss": 0.0514,
+      "step": 1229
+    },
+    {
+      "epoch": 0.5338541666666666,
+      "grad_norm": 0.729378342628479,
+      "learning_rate": 8.60016975152141e-06,
+      "loss": 0.0489,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5342881944444444,
+      "grad_norm": 0.6288821697235107,
+      "learning_rate": 8.597729755932175e-06,
+      "loss": 0.0494,
+      "step": 1231
+    },
+    {
+      "epoch": 0.5347222222222222,
+      "grad_norm": 0.841924250125885,
+      "learning_rate": 8.595287982442016e-06,
+      "loss": 0.0512,
+      "step": 1232
+    },
+    {
+      "epoch": 0.53515625,
+      "grad_norm": 0.7113053798675537,
+      "learning_rate": 8.592844432257588e-06,
+      "loss": 0.0411,
+      "step": 1233
+    },
+    {
+      "epoch": 0.5355902777777778,
+      "grad_norm": 0.6260560154914856,
+      "learning_rate": 8.590399106586429e-06,
+      "loss": 0.0517,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5360243055555556,
+      "grad_norm": 0.754421055316925,
+      "learning_rate": 8.58795200663695e-06,
+      "loss": 0.0551,
+      "step": 1235
+    },
+    {
+      "epoch": 0.5364583333333334,
+      "grad_norm": 0.7027459144592285,
+      "learning_rate": 8.585503133618447e-06,
+      "loss": 0.058,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5368923611111112,
+      "grad_norm": 0.7734227180480957,
+      "learning_rate": 8.583052488741081e-06,
+      "loss": 0.0474,
+      "step": 1237
+    },
+    {
+      "epoch": 0.5373263888888888,
+      "grad_norm": 0.6809510588645935,
+      "learning_rate": 8.5806000732159e-06,
+      "loss": 0.0562,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5377604166666666,
+      "grad_norm": 0.7577083110809326,
+      "learning_rate": 8.578145888254819e-06,
+      "loss": 0.0789,
+      "step": 1239
+    },
+    {
+      "epoch": 0.5381944444444444,
+      "grad_norm": 0.45938336849212646,
+      "learning_rate": 8.575689935070628e-06,
+      "loss": 0.0381,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5386284722222222,
+      "grad_norm": 0.43321073055267334,
+      "learning_rate": 8.573232214876996e-06,
+      "loss": 0.0473,
+      "step": 1241
+    },
+    {
+      "epoch": 0.5390625,
+      "grad_norm": 0.5972332954406738,
+      "learning_rate": 8.57077272888846e-06,
+      "loss": 0.0629,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5394965277777778,
+      "grad_norm": 0.5967984199523926,
+      "learning_rate": 8.568311478320434e-06,
+      "loss": 0.0446,
+      "step": 1243
+    },
+    {
+      "epoch": 0.5399305555555556,
+      "grad_norm": 0.5215777158737183,
+      "learning_rate": 8.565848464389198e-06,
+      "loss": 0.0404,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5403645833333334,
+      "grad_norm": 0.5675219297409058,
+      "learning_rate": 8.56338368831191e-06,
+      "loss": 0.0415,
+      "step": 1245
+    },
+    {
+      "epoch": 0.5407986111111112,
+      "grad_norm": 0.8522801399230957,
+      "learning_rate": 8.560917151306594e-06,
+      "loss": 0.0583,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5412326388888888,
+      "grad_norm": 0.4838396906852722,
+      "learning_rate": 8.558448854592148e-06,
+      "loss": 0.0371,
+      "step": 1247
+    },
+    {
+      "epoch": 0.5416666666666666,
+      "grad_norm": 0.5317425727844238,
+      "learning_rate": 8.555978799388336e-06,
+      "loss": 0.0357,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5421006944444444,
+      "grad_norm": 0.6244840025901794,
+      "learning_rate": 8.553506986915792e-06,
+      "loss": 0.0507,
+      "step": 1249
+    },
+    {
+      "epoch": 0.5425347222222222,
+      "grad_norm": 1.0391993522644043,
+      "learning_rate": 8.551033418396019e-06,
+      "loss": 0.0585,
+      "step": 1250
+    },
+    {
+      "epoch": 0.54296875,
+      "grad_norm": 0.9313353300094604,
+      "learning_rate": 8.548558095051391e-06,
+      "loss": 0.0814,
+      "step": 1251
+    },
+    {
+      "epoch": 0.5434027777777778,
+      "grad_norm": 0.835850715637207,
+      "learning_rate": 8.546081018105142e-06,
+      "loss": 0.0507,
+      "step": 1252
+    },
+    {
+      "epoch": 0.5438368055555556,
+      "grad_norm": 0.7506844401359558,
+      "learning_rate": 8.54360218878138e-06,
+      "loss": 0.0575,
+      "step": 1253
+    },
+    {
+      "epoch": 0.5442708333333334,
+      "grad_norm": 1.0784194469451904,
+      "learning_rate": 8.541121608305072e-06,
+      "loss": 0.0724,
+      "step": 1254
+    },
+    {
+      "epoch": 0.5447048611111112,
+      "grad_norm": 0.6933436393737793,
+      "learning_rate": 8.538639277902057e-06,
+      "loss": 0.0648,
+      "step": 1255
+    },
+    {
+      "epoch": 0.5451388888888888,
+      "grad_norm": 0.6776074171066284,
+      "learning_rate": 8.536155198799034e-06,
+      "loss": 0.0606,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5455729166666666,
+      "grad_norm": 0.5607663989067078,
+      "learning_rate": 8.533669372223568e-06,
+      "loss": 0.0589,
+      "step": 1257
+    },
+    {
+      "epoch": 0.5460069444444444,
+      "grad_norm": 0.4452672600746155,
+      "learning_rate": 8.531181799404088e-06,
+      "loss": 0.0436,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5464409722222222,
+      "grad_norm": 0.43998244404792786,
+      "learning_rate": 8.528692481569884e-06,
+      "loss": 0.0317,
+      "step": 1259
+    },
+    {
+      "epoch": 0.546875,
+      "grad_norm": 0.5600937604904175,
+      "learning_rate": 8.526201419951113e-06,
+      "loss": 0.0451,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5473090277777778,
+      "grad_norm": 0.5619065165519714,
+      "learning_rate": 8.523708615778785e-06,
+      "loss": 0.0556,
+      "step": 1261
+    },
+    {
+      "epoch": 0.5477430555555556,
+      "grad_norm": 0.48861682415008545,
+      "learning_rate": 8.521214070284782e-06,
+      "loss": 0.0364,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5481770833333334,
+      "grad_norm": 0.4935634732246399,
+      "learning_rate": 8.518717784701838e-06,
+      "loss": 0.0421,
+      "step": 1263
+    },
+    {
+      "epoch": 0.5486111111111112,
+      "grad_norm": 1.013365387916565,
+      "learning_rate": 8.51621976026355e-06,
+      "loss": 0.0546,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5490451388888888,
+      "grad_norm": 0.9382243156433105,
+      "learning_rate": 8.513719998204374e-06,
+      "loss": 0.0687,
+      "step": 1265
+    },
+    {
+      "epoch": 0.5494791666666666,
+      "grad_norm": 0.8279402852058411,
+      "learning_rate": 8.511218499759625e-06,
+      "loss": 0.0526,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5499131944444444,
+      "grad_norm": 0.8852784633636475,
+      "learning_rate": 8.508715266165479e-06,
+      "loss": 0.0669,
+      "step": 1267
+    },
+    {
+      "epoch": 0.5503472222222222,
+      "grad_norm": 0.8998196125030518,
+      "learning_rate": 8.506210298658964e-06,
+      "loss": 0.0599,
+      "step": 1268
+    },
+    {
+      "epoch": 0.55078125,
+      "grad_norm": 1.1187841892242432,
+      "learning_rate": 8.503703598477967e-06,
+      "loss": 0.0596,
+      "step": 1269
+    },
+    {
+      "epoch": 0.5512152777777778,
+      "grad_norm": 0.6818974018096924,
+      "learning_rate": 8.501195166861234e-06,
+      "loss": 0.0552,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5516493055555556,
+      "grad_norm": 0.8382002711296082,
+      "learning_rate": 8.49868500504836e-06,
+      "loss": 0.0564,
+      "step": 1271
+    },
+    {
+      "epoch": 0.5520833333333334,
+      "grad_norm": 0.8586729764938354,
+      "learning_rate": 8.496173114279803e-06,
+      "loss": 0.0607,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5525173611111112,
+      "grad_norm": 0.6577945947647095,
+      "learning_rate": 8.493659495796867e-06,
+      "loss": 0.0464,
+      "step": 1273
+    },
+    {
+      "epoch": 0.5529513888888888,
+      "grad_norm": 0.869432270526886,
+      "learning_rate": 8.49114415084172e-06,
+      "loss": 0.0598,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5533854166666666,
+      "grad_norm": 0.4027788043022156,
+      "learning_rate": 8.488627080657375e-06,
+      "loss": 0.0307,
+      "step": 1275
+    },
+    {
+      "epoch": 0.5538194444444444,
+      "grad_norm": 0.7500314116477966,
+      "learning_rate": 8.486108286487698e-06,
+      "loss": 0.0472,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5542534722222222,
+      "grad_norm": 0.5049936175346375,
+      "learning_rate": 8.483587769577414e-06,
+      "loss": 0.0402,
+      "step": 1277
+    },
+    {
+      "epoch": 0.5546875,
+      "grad_norm": 0.7624374628067017,
+      "learning_rate": 8.481065531172091e-06,
+      "loss": 0.0467,
+      "step": 1278
+    },
+    {
+      "epoch": 0.5551215277777778,
+      "grad_norm": 0.7847470045089722,
+      "learning_rate": 8.478541572518152e-06,
+      "loss": 0.0489,
+      "step": 1279
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.5217390656471252,
+      "learning_rate": 8.476015894862867e-06,
+      "loss": 0.0383,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5559895833333334,
+      "grad_norm": 0.8662663102149963,
+      "learning_rate": 8.473488499454363e-06,
+      "loss": 0.0702,
+      "step": 1281
+    },
+    {
+      "epoch": 0.5564236111111112,
+      "grad_norm": 0.8891013860702515,
+      "learning_rate": 8.470959387541606e-06,
+      "loss": 0.0722,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5568576388888888,
+      "grad_norm": 0.6959691047668457,
+      "learning_rate": 8.468428560374416e-06,
+      "loss": 0.0471,
+      "step": 1283
+    },
+    {
+      "epoch": 0.5572916666666666,
+      "grad_norm": 0.9065499901771545,
+      "learning_rate": 8.46589601920346e-06,
+      "loss": 0.0418,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5577256944444444,
+      "grad_norm": 0.694658637046814,
+      "learning_rate": 8.463361765280252e-06,
+      "loss": 0.0469,
+      "step": 1285
+    },
+    {
+      "epoch": 0.5581597222222222,
+      "grad_norm": 0.7287715673446655,
+      "learning_rate": 8.460825799857148e-06,
+      "loss": 0.0484,
+      "step": 1286
+    },
+    {
+      "epoch": 0.55859375,
+      "grad_norm": 0.4626723825931549,
+      "learning_rate": 8.45828812418736e-06,
+      "loss": 0.0405,
+      "step": 1287
+    },
+    {
+      "epoch": 0.5590277777777778,
+      "grad_norm": 0.7418968677520752,
+      "learning_rate": 8.455748739524934e-06,
+      "loss": 0.0672,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5594618055555556,
+      "grad_norm": 0.8423126935958862,
+      "learning_rate": 8.453207647124769e-06,
+      "loss": 0.0591,
+      "step": 1289
+    },
+    {
+      "epoch": 0.5598958333333334,
+      "grad_norm": 0.7815362215042114,
+      "learning_rate": 8.450664848242599e-06,
+      "loss": 0.0431,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5603298611111112,
+      "grad_norm": 0.6513561010360718,
+      "learning_rate": 8.448120344135012e-06,
+      "loss": 0.0601,
+      "step": 1291
+    },
+    {
+      "epoch": 0.5607638888888888,
+      "grad_norm": 0.5690886378288269,
+      "learning_rate": 8.445574136059429e-06,
+      "loss": 0.0555,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5611979166666666,
+      "grad_norm": 0.7079150676727295,
+      "learning_rate": 8.44302622527412e-06,
+      "loss": 0.0528,
+      "step": 1293
+    },
+    {
+      "epoch": 0.5616319444444444,
+      "grad_norm": 0.5692477822303772,
+      "learning_rate": 8.440476613038192e-06,
+      "loss": 0.0412,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5620659722222222,
+      "grad_norm": 0.8501294851303101,
+      "learning_rate": 8.437925300611598e-06,
+      "loss": 0.052,
+      "step": 1295
+    },
+    {
+      "epoch": 0.5625,
+      "grad_norm": 0.8660826086997986,
+      "learning_rate": 8.435372289255123e-06,
+      "loss": 0.0473,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5629340277777778,
+      "grad_norm": 0.6774151921272278,
+      "learning_rate": 8.4328175802304e-06,
+      "loss": 0.0582,
+      "step": 1297
+    },
+    {
+      "epoch": 0.5633680555555556,
+      "grad_norm": 0.6546648740768433,
+      "learning_rate": 8.430261174799897e-06,
+      "loss": 0.0432,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5638020833333334,
+      "grad_norm": 0.7087470889091492,
+      "learning_rate": 8.427703074226916e-06,
+      "loss": 0.0367,
+      "step": 1299
+    },
+    {
+      "epoch": 0.5642361111111112,
+      "grad_norm": 0.6567984223365784,
+      "learning_rate": 8.425143279775607e-06,
+      "loss": 0.0455,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5646701388888888,
+      "grad_norm": 1.0748047828674316,
+      "learning_rate": 8.422581792710948e-06,
+      "loss": 0.0557,
+      "step": 1301
+    },
+    {
+      "epoch": 0.5651041666666666,
+      "grad_norm": 0.6882471442222595,
+      "learning_rate": 8.420018614298758e-06,
+      "loss": 0.0524,
+      "step": 1302
+    },
+    {
+      "epoch": 0.5655381944444444,
+      "grad_norm": 0.7942001819610596,
+      "learning_rate": 8.41745374580569e-06,
+      "loss": 0.0609,
+      "step": 1303
+    },
+    {
+      "epoch": 0.5659722222222222,
+      "grad_norm": 1.2504606246948242,
+      "learning_rate": 8.414887188499233e-06,
+      "loss": 0.0386,
+      "step": 1304
+    },
+    {
+      "epoch": 0.56640625,
+      "grad_norm": 0.587224006652832,
+      "learning_rate": 8.41231894364771e-06,
+      "loss": 0.0586,
+      "step": 1305
+    },
+    {
+      "epoch": 0.5668402777777778,
+      "grad_norm": 0.8295255899429321,
+      "learning_rate": 8.409749012520278e-06,
+      "loss": 0.048,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5672743055555556,
+      "grad_norm": 0.7238820791244507,
+      "learning_rate": 8.40717739638693e-06,
+      "loss": 0.0789,
+      "step": 1307
+    },
+    {
+      "epoch": 0.5677083333333334,
+      "grad_norm": 0.5116128325462341,
+      "learning_rate": 8.404604096518486e-06,
+      "loss": 0.0454,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5681423611111112,
+      "grad_norm": 0.4796935021877289,
+      "learning_rate": 8.402029114186604e-06,
+      "loss": 0.0498,
+      "step": 1309
+    },
+    {
+      "epoch": 0.5685763888888888,
+      "grad_norm": 0.638282299041748,
+      "learning_rate": 8.39945245066377e-06,
+      "loss": 0.0459,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5690104166666666,
+      "grad_norm": 0.4643474817276001,
+      "learning_rate": 8.396874107223298e-06,
+      "loss": 0.0386,
+      "step": 1311
+    },
+    {
+      "epoch": 0.5694444444444444,
+      "grad_norm": 0.6372327208518982,
+      "learning_rate": 8.394294085139343e-06,
+      "loss": 0.0641,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5698784722222222,
+      "grad_norm": 0.7297875881195068,
+      "learning_rate": 8.391712385686877e-06,
+      "loss": 0.0651,
+      "step": 1313
+    },
+    {
+      "epoch": 0.5703125,
+      "grad_norm": 0.5666899681091309,
+      "learning_rate": 8.389129010141705e-06,
+      "loss": 0.0525,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5707465277777778,
+      "grad_norm": 0.749977171421051,
+      "learning_rate": 8.386543959780465e-06,
+      "loss": 0.0591,
+      "step": 1315
+    },
+    {
+      "epoch": 0.5711805555555556,
+      "grad_norm": 0.6804627776145935,
+      "learning_rate": 8.383957235880617e-06,
+      "loss": 0.0495,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5716145833333334,
+      "grad_norm": 0.5683985948562622,
+      "learning_rate": 8.381368839720454e-06,
+      "loss": 0.047,
+      "step": 1317
+    },
+    {
+      "epoch": 0.5720486111111112,
+      "grad_norm": 0.7459953427314758,
+      "learning_rate": 8.378778772579087e-06,
+      "loss": 0.0377,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5724826388888888,
+      "grad_norm": 1.290941834449768,
+      "learning_rate": 8.376187035736458e-06,
+      "loss": 0.0675,
+      "step": 1319
+    },
+    {
+      "epoch": 0.5729166666666666,
+      "grad_norm": 0.6719200015068054,
+      "learning_rate": 8.373593630473334e-06,
+      "loss": 0.0664,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5733506944444444,
+      "grad_norm": 0.9134562015533447,
+      "learning_rate": 8.370998558071308e-06,
+      "loss": 0.0481,
+      "step": 1321
+    },
+    {
+      "epoch": 0.5737847222222222,
+      "grad_norm": 0.6994345188140869,
+      "learning_rate": 8.36840181981279e-06,
+      "loss": 0.047,
+      "step": 1322
+    },
+    {
+      "epoch": 0.57421875,
+      "grad_norm": 0.7651628255844116,
+      "learning_rate": 8.365803416981022e-06,
+      "loss": 0.0486,
+      "step": 1323
+    },
+    {
+      "epoch": 0.5746527777777778,
+      "grad_norm": 0.8562982678413391,
+      "learning_rate": 8.363203350860064e-06,
+      "loss": 0.0383,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5750868055555556,
+      "grad_norm": 0.8496001958847046,
+      "learning_rate": 8.360601622734796e-06,
+      "loss": 0.0565,
+      "step": 1325
+    },
+    {
+      "epoch": 0.5755208333333334,
+      "grad_norm": 0.7761862277984619,
+      "learning_rate": 8.357998233890925e-06,
+      "loss": 0.075,
+      "step": 1326
+    },
+    {
+      "epoch": 0.5759548611111112,
+      "grad_norm": 0.5182710289955139,
+      "learning_rate": 8.355393185614974e-06,
+      "loss": 0.0417,
+      "step": 1327
+    },
+    {
+      "epoch": 0.5763888888888888,
+      "grad_norm": 0.45891663432121277,
+      "learning_rate": 8.352786479194288e-06,
+      "loss": 0.028,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5768229166666666,
+      "grad_norm": 0.6198312044143677,
+      "learning_rate": 8.35017811591703e-06,
+      "loss": 0.053,
+      "step": 1329
+    },
+    {
+      "epoch": 0.5772569444444444,
+      "grad_norm": 0.620874285697937,
+      "learning_rate": 8.347568097072183e-06,
+      "loss": 0.0599,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5776909722222222,
+      "grad_norm": 0.6952489018440247,
+      "learning_rate": 8.344956423949547e-06,
+      "loss": 0.0645,
+      "step": 1331
+    },
+    {
+      "epoch": 0.578125,
+      "grad_norm": 0.9362801313400269,
+      "learning_rate": 8.342343097839743e-06,
+      "loss": 0.0738,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5785590277777778,
+      "grad_norm": 0.7076898813247681,
+      "learning_rate": 8.339728120034204e-06,
+      "loss": 0.0539,
+      "step": 1333
+    },
+    {
+      "epoch": 0.5789930555555556,
+      "grad_norm": 0.4984985888004303,
+      "learning_rate": 8.337111491825183e-06,
+      "loss": 0.0365,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5794270833333334,
+      "grad_norm": 0.4663236439228058,
+      "learning_rate": 8.334493214505746e-06,
+      "loss": 0.0405,
+      "step": 1335
+    },
+    {
+      "epoch": 0.5798611111111112,
+      "grad_norm": 0.7300651669502258,
+      "learning_rate": 8.331873289369774e-06,
+      "loss": 0.0512,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5802951388888888,
+      "grad_norm": 0.8619731664657593,
+      "learning_rate": 8.329251717711964e-06,
+      "loss": 0.0483,
+      "step": 1337
+    },
+    {
+      "epoch": 0.5807291666666666,
+      "grad_norm": 1.2407528162002563,
+      "learning_rate": 8.326628500827826e-06,
+      "loss": 0.0408,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5811631944444444,
+      "grad_norm": 0.7284979820251465,
+      "learning_rate": 8.324003640013685e-06,
+      "loss": 0.0612,
+      "step": 1339
+    },
+    {
+      "epoch": 0.5815972222222222,
+      "grad_norm": 0.565274715423584,
+      "learning_rate": 8.321377136566675e-06,
+      "loss": 0.0563,
+      "step": 1340
+    },
+    {
+      "epoch": 0.58203125,
+      "grad_norm": 0.6407968401908875,
+      "learning_rate": 8.318748991784742e-06,
+      "loss": 0.048,
+      "step": 1341
+    },
+    {
+      "epoch": 0.5824652777777778,
+      "grad_norm": 1.3209383487701416,
+      "learning_rate": 8.316119206966644e-06,
+      "loss": 0.0977,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5828993055555556,
+      "grad_norm": 1.2708077430725098,
+      "learning_rate": 8.313487783411954e-06,
+      "loss": 0.082,
+      "step": 1343
+    },
+    {
+      "epoch": 0.5833333333333334,
+      "grad_norm": 0.7060262560844421,
+      "learning_rate": 8.310854722421044e-06,
+      "loss": 0.0539,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5837673611111112,
+      "grad_norm": 0.7067934274673462,
+      "learning_rate": 8.308220025295107e-06,
+      "loss": 0.0442,
+      "step": 1345
+    },
+    {
+      "epoch": 0.5842013888888888,
+      "grad_norm": 0.5657163858413696,
+      "learning_rate": 8.305583693336135e-06,
+      "loss": 0.0424,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5846354166666666,
+      "grad_norm": 0.8705339431762695,
+      "learning_rate": 8.302945727846934e-06,
+      "loss": 0.0774,
+      "step": 1347
+    },
+    {
+      "epoch": 0.5850694444444444,
+      "grad_norm": 0.651258647441864,
+      "learning_rate": 8.300306130131116e-06,
+      "loss": 0.048,
+      "step": 1348
+    },
+    {
+      "epoch": 0.5855034722222222,
+      "grad_norm": 0.7655849456787109,
+      "learning_rate": 8.297664901493098e-06,
+      "loss": 0.0577,
+      "step": 1349
+    },
+    {
+      "epoch": 0.5859375,
+      "grad_norm": 0.5116808414459229,
+      "learning_rate": 8.295022043238104e-06,
+      "loss": 0.0353,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5863715277777778,
+      "grad_norm": 0.5585806965827942,
+      "learning_rate": 8.292377556672165e-06,
+      "loss": 0.0502,
+      "step": 1351
+    },
+    {
+      "epoch": 0.5868055555555556,
+      "grad_norm": 0.6864737272262573,
+      "learning_rate": 8.289731443102111e-06,
+      "loss": 0.0511,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5872395833333334,
+      "grad_norm": 0.9937721490859985,
+      "learning_rate": 8.287083703835584e-06,
+      "loss": 0.0469,
+      "step": 1353
+    },
+    {
+      "epoch": 0.5876736111111112,
+      "grad_norm": 1.1562503576278687,
+      "learning_rate": 8.284434340181025e-06,
+      "loss": 0.0751,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5881076388888888,
+      "grad_norm": 0.8913924098014832,
+      "learning_rate": 8.281783353447675e-06,
+      "loss": 0.0599,
+      "step": 1355
+    },
+    {
+      "epoch": 0.5885416666666666,
+      "grad_norm": 0.8412242531776428,
+      "learning_rate": 8.27913074494558e-06,
+      "loss": 0.04,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5889756944444444,
+      "grad_norm": 0.9167409539222717,
+      "learning_rate": 8.27647651598559e-06,
+      "loss": 0.0371,
+      "step": 1357
+    },
+    {
+      "epoch": 0.5894097222222222,
+      "grad_norm": 1.0268688201904297,
+      "learning_rate": 8.273820667879351e-06,
+      "loss": 0.0747,
+      "step": 1358
+    },
+    {
+      "epoch": 0.58984375,
+      "grad_norm": 0.6086006164550781,
+      "learning_rate": 8.271163201939316e-06,
+      "loss": 0.0311,
+      "step": 1359
+    },
+    {
+      "epoch": 0.5902777777777778,
+      "grad_norm": 0.5621817708015442,
+      "learning_rate": 8.268504119478729e-06,
+      "loss": 0.047,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5907118055555556,
+      "grad_norm": 0.8023841381072998,
+      "learning_rate": 8.265843421811636e-06,
+      "loss": 0.0553,
+      "step": 1361
+    },
+    {
+      "epoch": 0.5911458333333334,
+      "grad_norm": 0.9353199601173401,
+      "learning_rate": 8.263181110252885e-06,
+      "loss": 0.0641,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5915798611111112,
+      "grad_norm": 0.6181360483169556,
+      "learning_rate": 8.260517186118116e-06,
+      "loss": 0.0639,
+      "step": 1363
+    },
+    {
+      "epoch": 0.5920138888888888,
+      "grad_norm": 0.5545939803123474,
+      "learning_rate": 8.25785165072377e-06,
+      "loss": 0.0585,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5924479166666666,
+      "grad_norm": 0.6505069136619568,
+      "learning_rate": 8.255184505387082e-06,
+      "loss": 0.0544,
+      "step": 1365
+    },
+    {
+      "epoch": 0.5928819444444444,
+      "grad_norm": 0.4361746907234192,
+      "learning_rate": 8.252515751426082e-06,
+      "loss": 0.03,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5933159722222222,
+      "grad_norm": 0.5293030142784119,
+      "learning_rate": 8.249845390159598e-06,
+      "loss": 0.0559,
+      "step": 1367
+    },
+    {
+      "epoch": 0.59375,
+      "grad_norm": 0.5465980172157288,
+      "learning_rate": 8.24717342290725e-06,
+      "loss": 0.0444,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5941840277777778,
+      "grad_norm": 0.6178024411201477,
+      "learning_rate": 8.244499850989453e-06,
+      "loss": 0.0571,
+      "step": 1369
+    },
+    {
+      "epoch": 0.5946180555555556,
+      "grad_norm": 0.5191216468811035,
+      "learning_rate": 8.241824675727412e-06,
+      "loss": 0.0347,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5950520833333334,
+      "grad_norm": 0.5588525533676147,
+      "learning_rate": 8.239147898443126e-06,
+      "loss": 0.0491,
+      "step": 1371
+    },
+    {
+      "epoch": 0.5954861111111112,
+      "grad_norm": 0.9492883682250977,
+      "learning_rate": 8.236469520459391e-06,
+      "loss": 0.0513,
+      "step": 1372
+    },
+    {
+      "epoch": 0.5959201388888888,
+      "grad_norm": 1.0945905447006226,
+      "learning_rate": 8.233789543099784e-06,
+      "loss": 0.0533,
+      "step": 1373
+    },
+    {
+      "epoch": 0.5963541666666666,
+      "grad_norm": 0.7427625060081482,
+      "learning_rate": 8.231107967688678e-06,
+      "loss": 0.0579,
+      "step": 1374
+    },
+    {
+      "epoch": 0.5967881944444444,
+      "grad_norm": 0.7160750031471252,
+      "learning_rate": 8.228424795551238e-06,
+      "loss": 0.045,
+      "step": 1375
+    },
+    {
+      "epoch": 0.5972222222222222,
+      "grad_norm": 0.7438015937805176,
+      "learning_rate": 8.225740028013413e-06,
+      "loss": 0.0457,
+      "step": 1376
+    },
+    {
+      "epoch": 0.59765625,
+      "grad_norm": 1.2287431955337524,
+      "learning_rate": 8.223053666401942e-06,
+      "loss": 0.0553,
+      "step": 1377
+    },
+    {
+      "epoch": 0.5980902777777778,
+      "grad_norm": 0.8054426908493042,
+      "learning_rate": 8.220365712044353e-06,
+      "loss": 0.0453,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5985243055555556,
+      "grad_norm": 0.6114641427993774,
+      "learning_rate": 8.217676166268965e-06,
+      "loss": 0.0504,
+      "step": 1379
+    },
+    {
+      "epoch": 0.5989583333333334,
+      "grad_norm": 0.6696768403053284,
+      "learning_rate": 8.214985030404872e-06,
+      "loss": 0.0514,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5993923611111112,
+      "grad_norm": 0.7519408464431763,
+      "learning_rate": 8.212292305781964e-06,
+      "loss": 0.0686,
+      "step": 1381
+    },
+    {
+      "epoch": 0.5998263888888888,
+      "grad_norm": 0.48269274830818176,
+      "learning_rate": 8.20959799373091e-06,
+      "loss": 0.0331,
+      "step": 1382
+    },
+    {
+      "epoch": 0.6002604166666666,
+      "grad_norm": 0.7272772192955017,
+      "learning_rate": 8.206902095583174e-06,
+      "loss": 0.0518,
+      "step": 1383
+    },
+    {
+      "epoch": 0.6006944444444444,
+      "grad_norm": 0.5394783020019531,
+      "learning_rate": 8.204204612670987e-06,
+      "loss": 0.0481,
+      "step": 1384
+    },
+    {
+      "epoch": 0.6011284722222222,
+      "grad_norm": 0.5102927088737488,
+      "learning_rate": 8.201505546327377e-06,
+      "loss": 0.0378,
+      "step": 1385
+    },
+    {
+      "epoch": 0.6015625,
+      "grad_norm": 0.6936540007591248,
+      "learning_rate": 8.19880489788615e-06,
+      "loss": 0.0534,
+      "step": 1386
+    },
+    {
+      "epoch": 0.6019965277777778,
+      "grad_norm": 0.6128141283988953,
+      "learning_rate": 8.196102668681891e-06,
+      "loss": 0.0477,
+      "step": 1387
+    },
+    {
+      "epoch": 0.6024305555555556,
+      "grad_norm": 0.7972360849380493,
+      "learning_rate": 8.19339886004997e-06,
+      "loss": 0.0593,
+      "step": 1388
+    },
+    {
+      "epoch": 0.6028645833333334,
+      "grad_norm": 0.8966538906097412,
+      "learning_rate": 8.190693473326536e-06,
+      "loss": 0.0715,
+      "step": 1389
+    },
+    {
+      "epoch": 0.6032986111111112,
+      "grad_norm": 0.6679428815841675,
+      "learning_rate": 8.18798650984852e-06,
+      "loss": 0.0416,
+      "step": 1390
+    },
+    {
+      "epoch": 0.6037326388888888,
+      "grad_norm": 0.45375049114227295,
+      "learning_rate": 8.185277970953627e-06,
+      "loss": 0.0371,
+      "step": 1391
+    },
+    {
+      "epoch": 0.6041666666666666,
+      "grad_norm": 0.8478909134864807,
+      "learning_rate": 8.182567857980347e-06,
+      "loss": 0.0746,
+      "step": 1392
+    },
+    {
+      "epoch": 0.6046006944444444,
+      "grad_norm": 1.3403058052062988,
+      "learning_rate": 8.179856172267942e-06,
+      "loss": 0.064,
+      "step": 1393
+    },
+    {
+      "epoch": 0.6050347222222222,
+      "grad_norm": 0.5338839292526245,
+      "learning_rate": 8.177142915156455e-06,
+      "loss": 0.0311,
+      "step": 1394
+    },
+    {
+      "epoch": 0.60546875,
+      "grad_norm": 0.8129806518554688,
+      "learning_rate": 8.174428087986705e-06,
+      "loss": 0.0701,
+      "step": 1395
+    },
+    {
+      "epoch": 0.6059027777777778,
+      "grad_norm": 0.76915442943573,
+      "learning_rate": 8.171711692100286e-06,
+      "loss": 0.0345,
+      "step": 1396
+    },
+    {
+      "epoch": 0.6063368055555556,
+      "grad_norm": 0.6448995471000671,
+      "learning_rate": 8.168993728839567e-06,
+      "loss": 0.0431,
+      "step": 1397
+    },
+    {
+      "epoch": 0.6067708333333334,
+      "grad_norm": 0.5470688343048096,
+      "learning_rate": 8.166274199547692e-06,
+      "loss": 0.034,
+      "step": 1398
+    },
+    {
+      "epoch": 0.6072048611111112,
+      "grad_norm": 1.3783670663833618,
+      "learning_rate": 8.16355310556858e-06,
+      "loss": 0.0559,
+      "step": 1399
+    },
+    {
+      "epoch": 0.6076388888888888,
+      "grad_norm": 0.6607531905174255,
+      "learning_rate": 8.160830448246922e-06,
+      "loss": 0.0434,
+      "step": 1400
+    },
+    {
+      "epoch": 0.6080729166666666,
+      "grad_norm": 1.23941969871521,
+      "learning_rate": 8.15810622892818e-06,
+      "loss": 0.0739,
+      "step": 1401
+    },
+    {
+      "epoch": 0.6085069444444444,
+      "grad_norm": 0.6775326728820801,
+      "learning_rate": 8.155380448958589e-06,
+      "loss": 0.0522,
+      "step": 1402
+    },
+    {
+      "epoch": 0.6089409722222222,
+      "grad_norm": 0.5448042750358582,
+      "learning_rate": 8.152653109685163e-06,
+      "loss": 0.0347,
+      "step": 1403
+    },
+    {
+      "epoch": 0.609375,
+      "grad_norm": 0.6932007670402527,
+      "learning_rate": 8.14992421245567e-06,
+      "loss": 0.0492,
+      "step": 1404
+    },
+    {
+      "epoch": 0.6098090277777778,
+      "grad_norm": 0.7181015014648438,
+      "learning_rate": 8.147193758618664e-06,
+      "loss": 0.0485,
+      "step": 1405
+    },
+    {
+      "epoch": 0.6102430555555556,
+      "grad_norm": 0.5694456696510315,
+      "learning_rate": 8.144461749523458e-06,
+      "loss": 0.0484,
+      "step": 1406
+    },
+    {
+      "epoch": 0.6106770833333334,
+      "grad_norm": 0.649516224861145,
+      "learning_rate": 8.141728186520138e-06,
+      "loss": 0.048,
+      "step": 1407
+    },
+    {
+      "epoch": 0.6111111111111112,
+      "grad_norm": 0.6933144330978394,
+      "learning_rate": 8.138993070959557e-06,
+      "loss": 0.0624,
+      "step": 1408
+    },
+    {
+      "epoch": 0.6115451388888888,
+      "grad_norm": 1.0121896266937256,
+      "learning_rate": 8.136256404193336e-06,
+      "loss": 0.0527,
+      "step": 1409
+    },
+    {
+      "epoch": 0.6119791666666666,
+      "grad_norm": 0.766247034072876,
+      "learning_rate": 8.133518187573864e-06,
+      "loss": 0.0443,
+      "step": 1410
+    },
+    {
+      "epoch": 0.6124131944444444,
+      "grad_norm": 0.8101610541343689,
+      "learning_rate": 8.13077842245429e-06,
+      "loss": 0.0502,
+      "step": 1411
+    },
+    {
+      "epoch": 0.6128472222222222,
+      "grad_norm": 0.47420060634613037,
+      "learning_rate": 8.128037110188531e-06,
+      "loss": 0.0233,
+      "step": 1412
+    },
+    {
+      "epoch": 0.61328125,
+      "grad_norm": 1.5402026176452637,
+      "learning_rate": 8.125294252131273e-06,
+      "loss": 0.0785,
+      "step": 1413
+    },
+    {
+      "epoch": 0.6137152777777778,
+      "grad_norm": 0.6392644047737122,
+      "learning_rate": 8.122549849637961e-06,
+      "loss": 0.0357,
+      "step": 1414
+    },
+    {
+      "epoch": 0.6141493055555556,
+      "grad_norm": 0.7949815392494202,
+      "learning_rate": 8.119803904064805e-06,
+      "loss": 0.0415,
+      "step": 1415
+    },
+    {
+      "epoch": 0.6145833333333334,
+      "grad_norm": 1.336789608001709,
+      "learning_rate": 8.117056416768776e-06,
+      "loss": 0.0581,
+      "step": 1416
+    },
+    {
+      "epoch": 0.6150173611111112,
+      "grad_norm": 0.9395679831504822,
+      "learning_rate": 8.11430738910761e-06,
+      "loss": 0.0379,
+      "step": 1417
+    },
+    {
+      "epoch": 0.6154513888888888,
+      "grad_norm": 1.2471702098846436,
+      "learning_rate": 8.111556822439801e-06,
+      "loss": 0.0507,
+      "step": 1418
+    },
+    {
+      "epoch": 0.6158854166666666,
+      "grad_norm": 0.8362313508987427,
+      "learning_rate": 8.108804718124606e-06,
+      "loss": 0.0548,
+      "step": 1419
+    },
+    {
+      "epoch": 0.6163194444444444,
+      "grad_norm": 0.6505581736564636,
+      "learning_rate": 8.106051077522037e-06,
+      "loss": 0.0469,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6167534722222222,
+      "grad_norm": 0.6922495365142822,
+      "learning_rate": 8.103295901992874e-06,
+      "loss": 0.0505,
+      "step": 1421
+    },
+    {
+      "epoch": 0.6171875,
+      "grad_norm": 0.8996455669403076,
+      "learning_rate": 8.100539192898648e-06,
+      "loss": 0.05,
+      "step": 1422
+    },
+    {
+      "epoch": 0.6176215277777778,
+      "grad_norm": 0.46958956122398376,
+      "learning_rate": 8.097780951601647e-06,
+      "loss": 0.0282,
+      "step": 1423
+    },
+    {
+      "epoch": 0.6180555555555556,
+      "grad_norm": 0.9579825401306152,
+      "learning_rate": 8.095021179464926e-06,
+      "loss": 0.0497,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6184895833333334,
+      "grad_norm": 0.49152427911758423,
+      "learning_rate": 8.092259877852287e-06,
+      "loss": 0.0315,
+      "step": 1425
+    },
+    {
+      "epoch": 0.6189236111111112,
+      "grad_norm": 1.2000850439071655,
+      "learning_rate": 8.089497048128288e-06,
+      "loss": 0.0611,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6193576388888888,
+      "grad_norm": 1.0212475061416626,
+      "learning_rate": 8.086732691658247e-06,
+      "loss": 0.0651,
+      "step": 1427
+    },
+    {
+      "epoch": 0.6197916666666666,
+      "grad_norm": 0.9792412519454956,
+      "learning_rate": 8.083966809808234e-06,
+      "loss": 0.0486,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6202256944444444,
+      "grad_norm": 0.9700614213943481,
+      "learning_rate": 8.081199403945075e-06,
+      "loss": 0.0617,
+      "step": 1429
+    },
+    {
+      "epoch": 0.6206597222222222,
+      "grad_norm": 0.6700954437255859,
+      "learning_rate": 8.078430475436343e-06,
+      "loss": 0.0431,
+      "step": 1430
+    },
+    {
+      "epoch": 0.62109375,
+      "grad_norm": 0.9820812344551086,
+      "learning_rate": 8.075660025650374e-06,
+      "loss": 0.0478,
+      "step": 1431
+    },
+    {
+      "epoch": 0.6215277777777778,
+      "grad_norm": 0.6720173954963684,
+      "learning_rate": 8.072888055956246e-06,
+      "loss": 0.0436,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6219618055555556,
+      "grad_norm": 0.5771016478538513,
+      "learning_rate": 8.070114567723788e-06,
+      "loss": 0.0377,
+      "step": 1433
+    },
+    {
+      "epoch": 0.6223958333333334,
+      "grad_norm": 0.7792287468910217,
+      "learning_rate": 8.06733956232359e-06,
+      "loss": 0.0638,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6228298611111112,
+      "grad_norm": 0.9756418466567993,
+      "learning_rate": 8.064563041126981e-06,
+      "loss": 0.0557,
+      "step": 1435
+    },
+    {
+      "epoch": 0.6232638888888888,
+      "grad_norm": 0.5423485040664673,
+      "learning_rate": 8.06178500550604e-06,
+      "loss": 0.0351,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6236979166666666,
+      "grad_norm": 0.6620336771011353,
+      "learning_rate": 8.059005456833604e-06,
+      "loss": 0.0478,
+      "step": 1437
+    },
+    {
+      "epoch": 0.6241319444444444,
+      "grad_norm": 0.6537438631057739,
+      "learning_rate": 8.056224396483246e-06,
+      "loss": 0.0382,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6245659722222222,
+      "grad_norm": 0.6597551107406616,
+      "learning_rate": 8.05344182582929e-06,
+      "loss": 0.0465,
+      "step": 1439
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 1.3213461637496948,
+      "learning_rate": 8.050657746246812e-06,
+      "loss": 0.0539,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6254340277777778,
+      "grad_norm": 0.6487838625907898,
+      "learning_rate": 8.047872159111627e-06,
+      "loss": 0.0393,
+      "step": 1441
+    },
+    {
+      "epoch": 0.6258680555555556,
+      "grad_norm": 1.4973907470703125,
+      "learning_rate": 8.045085065800297e-06,
+      "loss": 0.0764,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6263020833333334,
+      "grad_norm": 0.8121576905250549,
+      "learning_rate": 8.042296467690126e-06,
+      "loss": 0.0571,
+      "step": 1443
+    },
+    {
+      "epoch": 0.6267361111111112,
+      "grad_norm": 0.5702161192893982,
+      "learning_rate": 8.039506366159166e-06,
+      "loss": 0.0279,
+      "step": 1444
+    },
+    {
+      "epoch": 0.6271701388888888,
+      "grad_norm": 0.5276521444320679,
+      "learning_rate": 8.03671476258621e-06,
+      "loss": 0.0328,
+      "step": 1445
+    },
+    {
+      "epoch": 0.6276041666666666,
+      "grad_norm": 0.8054888844490051,
+      "learning_rate": 8.033921658350793e-06,
+      "loss": 0.0596,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6280381944444444,
+      "grad_norm": 0.9613228440284729,
+      "learning_rate": 8.031127054833192e-06,
+      "loss": 0.0617,
+      "step": 1447
+    },
+    {
+      "epoch": 0.6284722222222222,
+      "grad_norm": 1.0140703916549683,
+      "learning_rate": 8.028330953414423e-06,
+      "loss": 0.0744,
+      "step": 1448
+    },
+    {
+      "epoch": 0.62890625,
+      "grad_norm": 0.652394711971283,
+      "learning_rate": 8.025533355476243e-06,
+      "loss": 0.037,
+      "step": 1449
+    },
+    {
+      "epoch": 0.6293402777777778,
+      "grad_norm": 0.636411726474762,
+      "learning_rate": 8.022734262401154e-06,
+      "loss": 0.0488,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6297743055555556,
+      "grad_norm": 0.6581993103027344,
+      "learning_rate": 8.019933675572389e-06,
+      "loss": 0.0488,
+      "step": 1451
+    },
+    {
+      "epoch": 0.6302083333333334,
+      "grad_norm": 0.6974286437034607,
+      "learning_rate": 8.017131596373922e-06,
+      "loss": 0.0555,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6306423611111112,
+      "grad_norm": 0.5334892868995667,
+      "learning_rate": 8.014328026190466e-06,
+      "loss": 0.0583,
+      "step": 1453
+    },
+    {
+      "epoch": 0.6310763888888888,
+      "grad_norm": 0.6498401761054993,
+      "learning_rate": 8.01152296640747e-06,
+      "loss": 0.0625,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6315104166666666,
+      "grad_norm": 0.6348752379417419,
+      "learning_rate": 8.008716418411117e-06,
+      "loss": 0.0551,
+      "step": 1455
+    },
+    {
+      "epoch": 0.6319444444444444,
+      "grad_norm": 0.5549450516700745,
+      "learning_rate": 8.005908383588332e-06,
+      "loss": 0.048,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6323784722222222,
+      "grad_norm": 0.5023066997528076,
+      "learning_rate": 8.003098863326764e-06,
+      "loss": 0.0485,
+      "step": 1457
+    },
+    {
+      "epoch": 0.6328125,
+      "grad_norm": 0.5993810892105103,
+      "learning_rate": 8.000287859014804e-06,
+      "loss": 0.0727,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6332465277777778,
+      "grad_norm": 0.6572069525718689,
+      "learning_rate": 7.997475372041576e-06,
+      "loss": 0.0471,
+      "step": 1459
+    },
+    {
+      "epoch": 0.6336805555555556,
+      "grad_norm": 0.5218420028686523,
+      "learning_rate": 7.994661403796935e-06,
+      "loss": 0.0416,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6341145833333334,
+      "grad_norm": 0.7828699946403503,
+      "learning_rate": 7.991845955671467e-06,
+      "loss": 0.0725,
+      "step": 1461
+    },
+    {
+      "epoch": 0.6345486111111112,
+      "grad_norm": 0.5772749781608582,
+      "learning_rate": 7.98902902905649e-06,
+      "loss": 0.0349,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6349826388888888,
+      "grad_norm": 0.480615496635437,
+      "learning_rate": 7.986210625344054e-06,
+      "loss": 0.0505,
+      "step": 1463
+    },
+    {
+      "epoch": 0.6354166666666666,
+      "grad_norm": 0.6188299059867859,
+      "learning_rate": 7.98339074592694e-06,
+      "loss": 0.0374,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6358506944444444,
+      "grad_norm": 0.5962648391723633,
+      "learning_rate": 7.980569392198651e-06,
+      "loss": 0.0526,
+      "step": 1465
+    },
+    {
+      "epoch": 0.6362847222222222,
+      "grad_norm": 0.9205628037452698,
+      "learning_rate": 7.97774656555343e-06,
+      "loss": 0.078,
+      "step": 1466
+    },
+    {
+      "epoch": 0.63671875,
+      "grad_norm": 0.6367231011390686,
+      "learning_rate": 7.974922267386239e-06,
+      "loss": 0.0384,
+      "step": 1467
+    },
+    {
+      "epoch": 0.6371527777777778,
+      "grad_norm": 0.9891589879989624,
+      "learning_rate": 7.972096499092767e-06,
+      "loss": 0.0694,
+      "step": 1468
+    },
+    {
+      "epoch": 0.6375868055555556,
+      "grad_norm": 0.7711750268936157,
+      "learning_rate": 7.969269262069438e-06,
+      "loss": 0.0552,
+      "step": 1469
+    },
+    {
+      "epoch": 0.6380208333333334,
+      "grad_norm": 0.623945951461792,
+      "learning_rate": 7.96644055771339e-06,
+      "loss": 0.0542,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6384548611111112,
+      "grad_norm": 0.6759617328643799,
+      "learning_rate": 7.963610387422499e-06,
+      "loss": 0.0433,
+      "step": 1471
+    },
+    {
+      "epoch": 0.6388888888888888,
+      "grad_norm": 0.6105266809463501,
+      "learning_rate": 7.960778752595355e-06,
+      "loss": 0.0407,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6393229166666666,
+      "grad_norm": 0.7171297669410706,
+      "learning_rate": 7.957945654631275e-06,
+      "loss": 0.0447,
+      "step": 1473
+    },
+    {
+      "epoch": 0.6397569444444444,
+      "grad_norm": 0.821792483329773,
+      "learning_rate": 7.9551110949303e-06,
+      "loss": 0.0581,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6401909722222222,
+      "grad_norm": 0.6385071277618408,
+      "learning_rate": 7.952275074893195e-06,
+      "loss": 0.0569,
+      "step": 1475
+    },
+    {
+      "epoch": 0.640625,
+      "grad_norm": 0.8872828483581543,
+      "learning_rate": 7.949437595921443e-06,
+      "loss": 0.0658,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6410590277777778,
+      "grad_norm": 0.6728609204292297,
+      "learning_rate": 7.946598659417249e-06,
+      "loss": 0.0396,
+      "step": 1477
+    },
+    {
+      "epoch": 0.6414930555555556,
+      "grad_norm": 0.6835218071937561,
+      "learning_rate": 7.943758266783539e-06,
+      "loss": 0.0656,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6419270833333334,
+      "grad_norm": 0.6220277547836304,
+      "learning_rate": 7.940916419423959e-06,
+      "loss": 0.048,
+      "step": 1479
+    },
+    {
+      "epoch": 0.6423611111111112,
+      "grad_norm": 0.46455779671669006,
+      "learning_rate": 7.938073118742871e-06,
+      "loss": 0.0382,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6427951388888888,
+      "grad_norm": 0.6037314534187317,
+      "learning_rate": 7.93522836614536e-06,
+      "loss": 0.0402,
+      "step": 1481
+    },
+    {
+      "epoch": 0.6432291666666666,
+      "grad_norm": 0.8157238364219666,
+      "learning_rate": 7.932382163037227e-06,
+      "loss": 0.0733,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6436631944444444,
+      "grad_norm": 1.062686800956726,
+      "learning_rate": 7.929534510824985e-06,
+      "loss": 0.067,
+      "step": 1483
+    },
+    {
+      "epoch": 0.6440972222222222,
+      "grad_norm": 0.7600874900817871,
+      "learning_rate": 7.926685410915871e-06,
+      "loss": 0.0513,
+      "step": 1484
+    },
+    {
+      "epoch": 0.64453125,
+      "grad_norm": 0.9816238880157471,
+      "learning_rate": 7.923834864717828e-06,
+      "loss": 0.0582,
+      "step": 1485
+    },
+    {
+      "epoch": 0.6449652777777778,
+      "grad_norm": 0.4806270897388458,
+      "learning_rate": 7.920982873639524e-06,
+      "loss": 0.0378,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6453993055555556,
+      "grad_norm": 0.7819468975067139,
+      "learning_rate": 7.918129439090335e-06,
+      "loss": 0.0562,
+      "step": 1487
+    },
+    {
+      "epoch": 0.6458333333333334,
+      "grad_norm": 0.6223804354667664,
+      "learning_rate": 7.915274562480349e-06,
+      "loss": 0.0492,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6462673611111112,
+      "grad_norm": 0.6152021884918213,
+      "learning_rate": 7.912418245220372e-06,
+      "loss": 0.0537,
+      "step": 1489
+    },
+    {
+      "epoch": 0.6467013888888888,
+      "grad_norm": 0.695647656917572,
+      "learning_rate": 7.909560488721915e-06,
+      "loss": 0.0484,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6471354166666666,
+      "grad_norm": 0.6031627655029297,
+      "learning_rate": 7.906701294397208e-06,
+      "loss": 0.0543,
+      "step": 1491
+    },
+    {
+      "epoch": 0.6475694444444444,
+      "grad_norm": 0.5799106955528259,
+      "learning_rate": 7.903840663659186e-06,
+      "loss": 0.0505,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6480034722222222,
+      "grad_norm": 1.1803741455078125,
+      "learning_rate": 7.900978597921494e-06,
+      "loss": 0.0398,
+      "step": 1493
+    },
+    {
+      "epoch": 0.6484375,
+      "grad_norm": 0.9330025315284729,
+      "learning_rate": 7.898115098598492e-06,
+      "loss": 0.0815,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6488715277777778,
+      "grad_norm": 0.6871426701545715,
+      "learning_rate": 7.895250167105239e-06,
+      "loss": 0.0501,
+      "step": 1495
+    },
+    {
+      "epoch": 0.6493055555555556,
+      "grad_norm": 0.7065111994743347,
+      "learning_rate": 7.89238380485751e-06,
+      "loss": 0.0519,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6497395833333334,
+      "grad_norm": 0.7787265777587891,
+      "learning_rate": 7.889516013271782e-06,
+      "loss": 0.0652,
+      "step": 1497
+    },
+    {
+      "epoch": 0.6501736111111112,
+      "grad_norm": 0.5526673197746277,
+      "learning_rate": 7.88664679376524e-06,
+      "loss": 0.0435,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6506076388888888,
+      "grad_norm": 0.7988463044166565,
+      "learning_rate": 7.883776147755777e-06,
+      "loss": 0.0602,
+      "step": 1499
+    },
+    {
+      "epoch": 0.6510416666666666,
+      "grad_norm": 0.6189616918563843,
+      "learning_rate": 7.880904076661988e-06,
+      "loss": 0.048,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6510416666666666,
+      "eval_layout_rm_rank_test_accuracy": 0.9717835248264383,
+      "eval_layout_rm_rank_test_chosen_scores": 0.5814792513847351,
+      "eval_layout_rm_rank_test_loss": 0.05069770663976669,
+      "eval_layout_rm_rank_test_rejected_scores": 0.08338902145624161,
+      "eval_layout_rm_rank_test_runtime": 1178.5822,
+      "eval_layout_rm_rank_test_samples_per_second": 20.899,
+      "eval_layout_rm_rank_test_steps_per_second": 0.653,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6514756944444444,
+      "grad_norm": 0.562228798866272,
+      "learning_rate": 7.878030581903171e-06,
+      "loss": 0.0367,
+      "step": 1501
+    },
+    {
+      "epoch": 0.6519097222222222,
+      "grad_norm": 0.6598196625709534,
+      "learning_rate": 7.875155664899334e-06,
+      "loss": 0.0547,
+      "step": 1502
+    },
+    {
+      "epoch": 0.65234375,
+      "grad_norm": 0.7484297752380371,
+      "learning_rate": 7.87227932707118e-06,
+      "loss": 0.0445,
+      "step": 1503
+    },
+    {
+      "epoch": 0.6527777777777778,
+      "grad_norm": 0.820593535900116,
+      "learning_rate": 7.869401569840117e-06,
+      "loss": 0.0469,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6532118055555556,
+      "grad_norm": 1.0721542835235596,
+      "learning_rate": 7.866522394628256e-06,
+      "loss": 0.0588,
+      "step": 1505
+    },
+    {
+      "epoch": 0.6536458333333334,
+      "grad_norm": 0.735040009021759,
+      "learning_rate": 7.863641802858408e-06,
+      "loss": 0.0462,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6540798611111112,
+      "grad_norm": 0.8309510946273804,
+      "learning_rate": 7.860759795954085e-06,
+      "loss": 0.0512,
+      "step": 1507
+    },
+    {
+      "epoch": 0.6545138888888888,
+      "grad_norm": 0.6727327704429626,
+      "learning_rate": 7.857876375339493e-06,
+      "loss": 0.051,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6549479166666666,
+      "grad_norm": 0.9708459377288818,
+      "learning_rate": 7.854991542439544e-06,
+      "loss": 0.0706,
+      "step": 1509
+    },
+    {
+      "epoch": 0.6553819444444444,
+      "grad_norm": 0.7068577408790588,
+      "learning_rate": 7.852105298679844e-06,
+      "loss": 0.0523,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6558159722222222,
+      "grad_norm": 0.5113874077796936,
+      "learning_rate": 7.849217645486695e-06,
+      "loss": 0.0292,
+      "step": 1511
+    },
+    {
+      "epoch": 0.65625,
+      "grad_norm": 0.5951700806617737,
+      "learning_rate": 7.8463285842871e-06,
+      "loss": 0.0374,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6566840277777778,
+      "grad_norm": 1.22222101688385,
+      "learning_rate": 7.843438116508752e-06,
+      "loss": 0.0734,
+      "step": 1513
+    },
+    {
+      "epoch": 0.6571180555555556,
+      "grad_norm": 0.6699300408363342,
+      "learning_rate": 7.840546243580045e-06,
+      "loss": 0.0442,
+      "step": 1514
+    },
+    {
+      "epoch": 0.6575520833333334,
+      "grad_norm": 0.5567951202392578,
+      "learning_rate": 7.837652966930064e-06,
+      "loss": 0.0396,
+      "step": 1515
+    },
+    {
+      "epoch": 0.6579861111111112,
+      "grad_norm": 0.7641263008117676,
+      "learning_rate": 7.834758287988586e-06,
+      "loss": 0.0441,
+      "step": 1516
+    },
+    {
+      "epoch": 0.6584201388888888,
+      "grad_norm": 0.7480666041374207,
+      "learning_rate": 7.831862208186085e-06,
+      "loss": 0.0463,
+      "step": 1517
+    },
+    {
+      "epoch": 0.6588541666666666,
+      "grad_norm": 0.8478124141693115,
+      "learning_rate": 7.828964728953728e-06,
+      "loss": 0.051,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6592881944444444,
+      "grad_norm": 0.7010301947593689,
+      "learning_rate": 7.826065851723366e-06,
+      "loss": 0.0378,
+      "step": 1519
+    },
+    {
+      "epoch": 0.6597222222222222,
+      "grad_norm": 0.7114124894142151,
+      "learning_rate": 7.823165577927547e-06,
+      "loss": 0.0587,
+      "step": 1520
+    },
+    {
+      "epoch": 0.66015625,
+      "grad_norm": 0.6724095940589905,
+      "learning_rate": 7.820263908999508e-06,
+      "loss": 0.0541,
+      "step": 1521
+    },
+    {
+      "epoch": 0.6605902777777778,
+      "grad_norm": 0.6992164254188538,
+      "learning_rate": 7.81736084637318e-06,
+      "loss": 0.0546,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6610243055555556,
+      "grad_norm": 0.5857810974121094,
+      "learning_rate": 7.814456391483174e-06,
+      "loss": 0.0522,
+      "step": 1523
+    },
+    {
+      "epoch": 0.6614583333333334,
+      "grad_norm": 0.6716225147247314,
+      "learning_rate": 7.811550545764792e-06,
+      "loss": 0.0477,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6618923611111112,
+      "grad_norm": 0.6821210980415344,
+      "learning_rate": 7.808643310654029e-06,
+      "loss": 0.0423,
+      "step": 1525
+    },
+    {
+      "epoch": 0.6623263888888888,
+      "grad_norm": 0.597215473651886,
+      "learning_rate": 7.805734687587559e-06,
+      "loss": 0.0376,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6627604166666666,
+      "grad_norm": 0.8344597816467285,
+      "learning_rate": 7.802824678002744e-06,
+      "loss": 0.0553,
+      "step": 1527
+    },
+    {
+      "epoch": 0.6631944444444444,
+      "grad_norm": 0.7426319718360901,
+      "learning_rate": 7.799913283337634e-06,
+      "loss": 0.0537,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6636284722222222,
+      "grad_norm": 0.7442464828491211,
+      "learning_rate": 7.797000505030962e-06,
+      "loss": 0.0667,
+      "step": 1529
+    },
+    {
+      "epoch": 0.6640625,
+      "grad_norm": 0.6925808191299438,
+      "learning_rate": 7.794086344522144e-06,
+      "loss": 0.039,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6644965277777778,
+      "grad_norm": 0.8789548277854919,
+      "learning_rate": 7.79117080325128e-06,
+      "loss": 0.0631,
+      "step": 1531
+    },
+    {
+      "epoch": 0.6649305555555556,
+      "grad_norm": 0.6922028660774231,
+      "learning_rate": 7.78825388265915e-06,
+      "loss": 0.0556,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6653645833333334,
+      "grad_norm": 0.6471952795982361,
+      "learning_rate": 7.78533558418722e-06,
+      "loss": 0.0449,
+      "step": 1533
+    },
+    {
+      "epoch": 0.6657986111111112,
+      "grad_norm": 0.6730069518089294,
+      "learning_rate": 7.782415909277632e-06,
+      "loss": 0.0557,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6662326388888888,
+      "grad_norm": 0.6181557774543762,
+      "learning_rate": 7.77949485937321e-06,
+      "loss": 0.0572,
+      "step": 1535
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.48969897627830505,
+      "learning_rate": 7.77657243591746e-06,
+      "loss": 0.0443,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6671006944444444,
+      "grad_norm": 0.6379843354225159,
+      "learning_rate": 7.773648640354564e-06,
+      "loss": 0.0525,
+      "step": 1537
+    },
+    {
+      "epoch": 0.6675347222222222,
+      "grad_norm": 0.6808762550354004,
+      "learning_rate": 7.770723474129385e-06,
+      "loss": 0.0578,
+      "step": 1538
+    },
+    {
+      "epoch": 0.66796875,
+      "grad_norm": 0.7974750995635986,
+      "learning_rate": 7.767796938687458e-06,
+      "loss": 0.0574,
+      "step": 1539
+    },
+    {
+      "epoch": 0.6684027777777778,
+      "grad_norm": 0.7461864352226257,
+      "learning_rate": 7.764869035474998e-06,
+      "loss": 0.0487,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6688368055555556,
+      "grad_norm": 0.7737073302268982,
+      "learning_rate": 7.761939765938897e-06,
+      "loss": 0.0547,
+      "step": 1541
+    },
+    {
+      "epoch": 0.6692708333333334,
+      "grad_norm": 0.9133140444755554,
+      "learning_rate": 7.75900913152672e-06,
+      "loss": 0.0438,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6697048611111112,
+      "grad_norm": 0.5385271906852722,
+      "learning_rate": 7.756077133686711e-06,
+      "loss": 0.0438,
+      "step": 1543
+    },
+    {
+      "epoch": 0.6701388888888888,
+      "grad_norm": 0.7025004029273987,
+      "learning_rate": 7.753143773867779e-06,
+      "loss": 0.0495,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6705729166666666,
+      "grad_norm": 0.6356387138366699,
+      "learning_rate": 7.750209053519513e-06,
+      "loss": 0.0504,
+      "step": 1545
+    },
+    {
+      "epoch": 0.6710069444444444,
+      "grad_norm": 0.6328694224357605,
+      "learning_rate": 7.747272974092173e-06,
+      "loss": 0.0452,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6714409722222222,
+      "grad_norm": 0.8696878552436829,
+      "learning_rate": 7.744335537036692e-06,
+      "loss": 0.0492,
+      "step": 1547
+    },
+    {
+      "epoch": 0.671875,
+      "grad_norm": 0.5818842649459839,
+      "learning_rate": 7.74139674380467e-06,
+      "loss": 0.0412,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6723090277777778,
+      "grad_norm": 1.057603359222412,
+      "learning_rate": 7.738456595848378e-06,
+      "loss": 0.042,
+      "step": 1549
+    },
+    {
+      "epoch": 0.6727430555555556,
+      "grad_norm": 0.4587070643901825,
+      "learning_rate": 7.735515094620761e-06,
+      "loss": 0.0432,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6731770833333334,
+      "grad_norm": 1.1908316612243652,
+      "learning_rate": 7.732572241575427e-06,
+      "loss": 0.059,
+      "step": 1551
+    },
+    {
+      "epoch": 0.6736111111111112,
+      "grad_norm": 0.5407928824424744,
+      "learning_rate": 7.729628038166655e-06,
+      "loss": 0.0471,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6740451388888888,
+      "grad_norm": 1.1224613189697266,
+      "learning_rate": 7.726682485849392e-06,
+      "loss": 0.062,
+      "step": 1553
+    },
+    {
+      "epoch": 0.6744791666666666,
+      "grad_norm": 1.0806435346603394,
+      "learning_rate": 7.723735586079251e-06,
+      "loss": 0.0655,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6749131944444444,
+      "grad_norm": 0.5161687731742859,
+      "learning_rate": 7.720787340312507e-06,
+      "loss": 0.0483,
+      "step": 1555
+    },
+    {
+      "epoch": 0.6753472222222222,
+      "grad_norm": 0.6101528406143188,
+      "learning_rate": 7.717837750006106e-06,
+      "loss": 0.0304,
+      "step": 1556
+    },
+    {
+      "epoch": 0.67578125,
+      "grad_norm": 0.6650863885879517,
+      "learning_rate": 7.714886816617656e-06,
+      "loss": 0.0492,
+      "step": 1557
+    },
+    {
+      "epoch": 0.6762152777777778,
+      "grad_norm": 0.9439712762832642,
+      "learning_rate": 7.711934541605427e-06,
+      "loss": 0.0711,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6766493055555556,
+      "grad_norm": 1.069583773612976,
+      "learning_rate": 7.708980926428356e-06,
+      "loss": 0.0608,
+      "step": 1559
+    },
+    {
+      "epoch": 0.6770833333333334,
+      "grad_norm": 0.8776530623435974,
+      "learning_rate": 7.706025972546039e-06,
+      "loss": 0.0635,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6775173611111112,
+      "grad_norm": 0.6253235936164856,
+      "learning_rate": 7.703069681418732e-06,
+      "loss": 0.0499,
+      "step": 1561
+    },
+    {
+      "epoch": 0.6779513888888888,
+      "grad_norm": 0.7380187511444092,
+      "learning_rate": 7.700112054507357e-06,
+      "loss": 0.0504,
+      "step": 1562
+    },
+    {
+      "epoch": 0.6783854166666666,
+      "grad_norm": 0.5171223282814026,
+      "learning_rate": 7.697153093273492e-06,
+      "loss": 0.0515,
+      "step": 1563
+    },
+    {
+      "epoch": 0.6788194444444444,
+      "grad_norm": 0.6803078055381775,
+      "learning_rate": 7.694192799179377e-06,
+      "loss": 0.0477,
+      "step": 1564
+    },
+    {
+      "epoch": 0.6792534722222222,
+      "grad_norm": 0.5703323483467102,
+      "learning_rate": 7.691231173687909e-06,
+      "loss": 0.0378,
+      "step": 1565
+    },
+    {
+      "epoch": 0.6796875,
+      "grad_norm": 0.6021401286125183,
+      "learning_rate": 7.688268218262645e-06,
+      "loss": 0.059,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6801215277777778,
+      "grad_norm": 0.5687470436096191,
+      "learning_rate": 7.685303934367793e-06,
+      "loss": 0.046,
+      "step": 1567
+    },
+    {
+      "epoch": 0.6805555555555556,
+      "grad_norm": 0.9414650201797485,
+      "learning_rate": 7.682338323468227e-06,
+      "loss": 0.0621,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6809895833333334,
+      "grad_norm": 0.8877261281013489,
+      "learning_rate": 7.679371387029468e-06,
+      "loss": 0.0636,
+      "step": 1569
+    },
+    {
+      "epoch": 0.6814236111111112,
+      "grad_norm": 0.7038198709487915,
+      "learning_rate": 7.676403126517698e-06,
+      "loss": 0.0404,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6818576388888888,
+      "grad_norm": 1.1424853801727295,
+      "learning_rate": 7.67343354339975e-06,
+      "loss": 0.0591,
+      "step": 1571
+    },
+    {
+      "epoch": 0.6822916666666666,
+      "grad_norm": 1.11978280544281,
+      "learning_rate": 7.670462639143115e-06,
+      "loss": 0.0397,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6827256944444444,
+      "grad_norm": 0.6603884696960449,
+      "learning_rate": 7.667490415215929e-06,
+      "loss": 0.0518,
+      "step": 1573
+    },
+    {
+      "epoch": 0.6831597222222222,
+      "grad_norm": 1.0959213972091675,
+      "learning_rate": 7.664516873086987e-06,
+      "loss": 0.0497,
+      "step": 1574
+    },
+    {
+      "epoch": 0.68359375,
+      "grad_norm": 0.6479515433311462,
+      "learning_rate": 7.661542014225735e-06,
+      "loss": 0.0527,
+      "step": 1575
+    },
+    {
+      "epoch": 0.6840277777777778,
+      "grad_norm": 0.9234191179275513,
+      "learning_rate": 7.658565840102265e-06,
+      "loss": 0.0648,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6844618055555556,
+      "grad_norm": 0.9269822835922241,
+      "learning_rate": 7.655588352187325e-06,
+      "loss": 0.0615,
+      "step": 1577
+    },
+    {
+      "epoch": 0.6848958333333334,
+      "grad_norm": 0.6565536260604858,
+      "learning_rate": 7.652609551952306e-06,
+      "loss": 0.0344,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6853298611111112,
+      "grad_norm": 0.5279817581176758,
+      "learning_rate": 7.649629440869254e-06,
+      "loss": 0.0355,
+      "step": 1579
+    },
+    {
+      "epoch": 0.6857638888888888,
+      "grad_norm": 0.5780903697013855,
+      "learning_rate": 7.646648020410857e-06,
+      "loss": 0.0485,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6861979166666666,
+      "grad_norm": 0.5379993319511414,
+      "learning_rate": 7.64366529205045e-06,
+      "loss": 0.0319,
+      "step": 1581
+    },
+    {
+      "epoch": 0.6866319444444444,
+      "grad_norm": 0.559840738773346,
+      "learning_rate": 7.640681257262023e-06,
+      "loss": 0.0424,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6870659722222222,
+      "grad_norm": 0.9723001718521118,
+      "learning_rate": 7.637695917520201e-06,
+      "loss": 0.0453,
+      "step": 1583
+    },
+    {
+      "epoch": 0.6875,
+      "grad_norm": 0.8327869772911072,
+      "learning_rate": 7.63470927430026e-06,
+      "loss": 0.0497,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6879340277777778,
+      "grad_norm": 0.8263087868690491,
+      "learning_rate": 7.631721329078118e-06,
+      "loss": 0.0587,
+      "step": 1585
+    },
+    {
+      "epoch": 0.6883680555555556,
+      "grad_norm": 1.0317976474761963,
+      "learning_rate": 7.628732083330338e-06,
+      "loss": 0.0699,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6888020833333334,
+      "grad_norm": 0.6567073464393616,
+      "learning_rate": 7.625741538534123e-06,
+      "loss": 0.052,
+      "step": 1587
+    },
+    {
+      "epoch": 0.6892361111111112,
+      "grad_norm": 0.8320260643959045,
+      "learning_rate": 7.622749696167318e-06,
+      "loss": 0.0462,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6896701388888888,
+      "grad_norm": 1.1415224075317383,
+      "learning_rate": 7.619756557708415e-06,
+      "loss": 0.0721,
+      "step": 1589
+    },
+    {
+      "epoch": 0.6901041666666666,
+      "grad_norm": 0.49437031149864197,
+      "learning_rate": 7.61676212463654e-06,
+      "loss": 0.0414,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6905381944444444,
+      "grad_norm": 0.6207118630409241,
+      "learning_rate": 7.613766398431462e-06,
+      "loss": 0.0658,
+      "step": 1591
+    },
+    {
+      "epoch": 0.6909722222222222,
+      "grad_norm": 0.9149729013442993,
+      "learning_rate": 7.610769380573584e-06,
+      "loss": 0.0702,
+      "step": 1592
+    },
+    {
+      "epoch": 0.69140625,
+      "grad_norm": 0.5002617239952087,
+      "learning_rate": 7.607771072543955e-06,
+      "loss": 0.041,
+      "step": 1593
+    },
+    {
+      "epoch": 0.6918402777777778,
+      "grad_norm": 0.49440690875053406,
+      "learning_rate": 7.604771475824258e-06,
+      "loss": 0.0492,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6922743055555556,
+      "grad_norm": 0.47741520404815674,
+      "learning_rate": 7.6017705918968134e-06,
+      "loss": 0.0399,
+      "step": 1595
+    },
+    {
+      "epoch": 0.6927083333333334,
+      "grad_norm": 0.4327121376991272,
+      "learning_rate": 7.598768422244573e-06,
+      "loss": 0.0406,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6931423611111112,
+      "grad_norm": 0.633120059967041,
+      "learning_rate": 7.595764968351132e-06,
+      "loss": 0.0511,
+      "step": 1597
+    },
+    {
+      "epoch": 0.6935763888888888,
+      "grad_norm": 0.8574790954589844,
+      "learning_rate": 7.592760231700715e-06,
+      "loss": 0.0784,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6940104166666666,
+      "grad_norm": 0.6364694237709045,
+      "learning_rate": 7.589754213778181e-06,
+      "loss": 0.0475,
+      "step": 1599
+    },
+    {
+      "epoch": 0.6944444444444444,
+      "grad_norm": 1.3502572774887085,
+      "learning_rate": 7.586746916069023e-06,
+      "loss": 0.0619,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6948784722222222,
+      "grad_norm": 0.5626755952835083,
+      "learning_rate": 7.583738340059368e-06,
+      "loss": 0.048,
+      "step": 1601
+    },
+    {
+      "epoch": 0.6953125,
+      "grad_norm": 0.6648493409156799,
+      "learning_rate": 7.580728487235971e-06,
+      "loss": 0.0498,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6957465277777778,
+      "grad_norm": 0.8648276925086975,
+      "learning_rate": 7.577717359086221e-06,
+      "loss": 0.0558,
+      "step": 1603
+    },
+    {
+      "epoch": 0.6961805555555556,
+      "grad_norm": 0.9568964242935181,
+      "learning_rate": 7.574704957098135e-06,
+      "loss": 0.0615,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6966145833333334,
+      "grad_norm": 0.7265316247940063,
+      "learning_rate": 7.571691282760364e-06,
+      "loss": 0.0465,
+      "step": 1605
+    },
+    {
+      "epoch": 0.6970486111111112,
+      "grad_norm": 0.5609695315361023,
+      "learning_rate": 7.568676337562182e-06,
+      "loss": 0.0579,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6974826388888888,
+      "grad_norm": 0.4945490062236786,
+      "learning_rate": 7.565660122993494e-06,
+      "loss": 0.0338,
+      "step": 1607
+    },
+    {
+      "epoch": 0.6979166666666666,
+      "grad_norm": 0.691581130027771,
+      "learning_rate": 7.56264264054483e-06,
+      "loss": 0.0491,
+      "step": 1608
+    },
+    {
+      "epoch": 0.6983506944444444,
+      "grad_norm": 0.4934881329536438,
+      "learning_rate": 7.55962389170735e-06,
+      "loss": 0.0305,
+      "step": 1609
+    },
+    {
+      "epoch": 0.6987847222222222,
+      "grad_norm": 1.052710771560669,
+      "learning_rate": 7.556603877972841e-06,
+      "loss": 0.0774,
+      "step": 1610
+    },
+    {
+      "epoch": 0.69921875,
+      "grad_norm": 0.5597624778747559,
+      "learning_rate": 7.553582600833709e-06,
+      "loss": 0.0337,
+      "step": 1611
+    },
+    {
+      "epoch": 0.6996527777777778,
+      "grad_norm": 0.85475754737854,
+      "learning_rate": 7.550560061782986e-06,
+      "loss": 0.0528,
+      "step": 1612
+    },
+    {
+      "epoch": 0.7000868055555556,
+      "grad_norm": 0.5703578591346741,
+      "learning_rate": 7.5475362623143345e-06,
+      "loss": 0.027,
+      "step": 1613
+    },
+    {
+      "epoch": 0.7005208333333334,
+      "grad_norm": 0.6355430483818054,
+      "learning_rate": 7.544511203922028e-06,
+      "loss": 0.046,
+      "step": 1614
+    },
+    {
+      "epoch": 0.7009548611111112,
+      "grad_norm": 1.1655287742614746,
+      "learning_rate": 7.541484888100974e-06,
+      "loss": 0.0724,
+      "step": 1615
+    },
+    {
+      "epoch": 0.7013888888888888,
+      "grad_norm": 1.0697033405303955,
+      "learning_rate": 7.5384573163466945e-06,
+      "loss": 0.0626,
+      "step": 1616
+    },
+    {
+      "epoch": 0.7018229166666666,
+      "grad_norm": 0.6042693853378296,
+      "learning_rate": 7.535428490155332e-06,
+      "loss": 0.0394,
+      "step": 1617
+    },
+    {
+      "epoch": 0.7022569444444444,
+      "grad_norm": 1.1471283435821533,
+      "learning_rate": 7.532398411023654e-06,
+      "loss": 0.0676,
+      "step": 1618
+    },
+    {
+      "epoch": 0.7026909722222222,
+      "grad_norm": 0.6342034339904785,
+      "learning_rate": 7.5293670804490396e-06,
+      "loss": 0.0401,
+      "step": 1619
+    },
+    {
+      "epoch": 0.703125,
+      "grad_norm": 0.630740761756897,
+      "learning_rate": 7.526334499929495e-06,
+      "loss": 0.052,
+      "step": 1620
+    },
+    {
+      "epoch": 0.7035590277777778,
+      "grad_norm": 1.228358507156372,
+      "learning_rate": 7.523300670963636e-06,
+      "loss": 0.066,
+      "step": 1621
+    },
+    {
+      "epoch": 0.7039930555555556,
+      "grad_norm": 0.7143034934997559,
+      "learning_rate": 7.520265595050702e-06,
+      "loss": 0.0544,
+      "step": 1622
+    },
+    {
+      "epoch": 0.7044270833333334,
+      "grad_norm": 0.8387890458106995,
+      "learning_rate": 7.517229273690543e-06,
+      "loss": 0.0758,
+      "step": 1623
+    },
+    {
+      "epoch": 0.7048611111111112,
+      "grad_norm": 0.5054572820663452,
+      "learning_rate": 7.514191708383627e-06,
+      "loss": 0.0383,
+      "step": 1624
+    },
+    {
+      "epoch": 0.7052951388888888,
+      "grad_norm": 0.522240161895752,
+      "learning_rate": 7.511152900631038e-06,
+      "loss": 0.0473,
+      "step": 1625
+    },
+    {
+      "epoch": 0.7057291666666666,
+      "grad_norm": 0.5247807502746582,
+      "learning_rate": 7.508112851934475e-06,
+      "loss": 0.0567,
+      "step": 1626
+    },
+    {
+      "epoch": 0.7061631944444444,
+      "grad_norm": 0.579017698764801,
+      "learning_rate": 7.505071563796245e-06,
+      "loss": 0.0429,
+      "step": 1627
+    },
+    {
+      "epoch": 0.7065972222222222,
+      "grad_norm": 0.5940823554992676,
+      "learning_rate": 7.5020290377192696e-06,
+      "loss": 0.0493,
+      "step": 1628
+    },
+    {
+      "epoch": 0.70703125,
+      "grad_norm": 0.6563714742660522,
+      "learning_rate": 7.498985275207086e-06,
+      "loss": 0.0531,
+      "step": 1629
+    },
+    {
+      "epoch": 0.7074652777777778,
+      "grad_norm": 0.5494057536125183,
+      "learning_rate": 7.4959402777638365e-06,
+      "loss": 0.0298,
+      "step": 1630
+    },
+    {
+      "epoch": 0.7078993055555556,
+      "grad_norm": 0.4213770031929016,
+      "learning_rate": 7.492894046894279e-06,
+      "loss": 0.0349,
+      "step": 1631
+    },
+    {
+      "epoch": 0.7083333333333334,
+      "grad_norm": 0.6510403752326965,
+      "learning_rate": 7.489846584103778e-06,
+      "loss": 0.0414,
+      "step": 1632
+    },
+    {
+      "epoch": 0.7087673611111112,
+      "grad_norm": 0.7641744613647461,
+      "learning_rate": 7.486797890898307e-06,
+      "loss": 0.0448,
+      "step": 1633
+    },
+    {
+      "epoch": 0.7092013888888888,
+      "grad_norm": 0.8901313543319702,
+      "learning_rate": 7.483747968784449e-06,
+      "loss": 0.0624,
+      "step": 1634
+    },
+    {
+      "epoch": 0.7096354166666666,
+      "grad_norm": 0.6470595002174377,
+      "learning_rate": 7.480696819269391e-06,
+      "loss": 0.0396,
+      "step": 1635
+    },
+    {
+      "epoch": 0.7100694444444444,
+      "grad_norm": 1.072182059288025,
+      "learning_rate": 7.47764444386093e-06,
+      "loss": 0.0712,
+      "step": 1636
+    },
+    {
+      "epoch": 0.7105034722222222,
+      "grad_norm": 0.8364940881729126,
+      "learning_rate": 7.474590844067468e-06,
+      "loss": 0.0454,
+      "step": 1637
+    },
+    {
+      "epoch": 0.7109375,
+      "grad_norm": 0.7541149258613586,
+      "learning_rate": 7.471536021398009e-06,
+      "loss": 0.0518,
+      "step": 1638
+    },
+    {
+      "epoch": 0.7113715277777778,
+      "grad_norm": 0.7979973554611206,
+      "learning_rate": 7.468479977362169e-06,
+      "loss": 0.0439,
+      "step": 1639
+    },
+    {
+      "epoch": 0.7118055555555556,
+      "grad_norm": 0.6536715030670166,
+      "learning_rate": 7.465422713470156e-06,
+      "loss": 0.0482,
+      "step": 1640
+    },
+    {
+      "epoch": 0.7122395833333334,
+      "grad_norm": 1.1344897747039795,
+      "learning_rate": 7.462364231232794e-06,
+      "loss": 0.0427,
+      "step": 1641
+    },
+    {
+      "epoch": 0.7126736111111112,
+      "grad_norm": 0.6313517093658447,
+      "learning_rate": 7.459304532161498e-06,
+      "loss": 0.039,
+      "step": 1642
+    },
+    {
+      "epoch": 0.7131076388888888,
+      "grad_norm": 0.8508172631263733,
+      "learning_rate": 7.4562436177682885e-06,
+      "loss": 0.0594,
+      "step": 1643
+    },
+    {
+      "epoch": 0.7135416666666666,
+      "grad_norm": 0.7220401167869568,
+      "learning_rate": 7.4531814895657885e-06,
+      "loss": 0.0448,
+      "step": 1644
+    },
+    {
+      "epoch": 0.7139756944444444,
+      "grad_norm": 0.7740882635116577,
+      "learning_rate": 7.450118149067218e-06,
+      "loss": 0.0564,
+      "step": 1645
+    },
+    {
+      "epoch": 0.7144097222222222,
+      "grad_norm": 0.41988405585289,
+      "learning_rate": 7.447053597786398e-06,
+      "loss": 0.0383,
+      "step": 1646
+    },
+    {
+      "epoch": 0.71484375,
+      "grad_norm": 0.9427817463874817,
+      "learning_rate": 7.443987837237749e-06,
+      "loss": 0.0673,
+      "step": 1647
+    },
+    {
+      "epoch": 0.7152777777777778,
+      "grad_norm": 0.9876265525817871,
+      "learning_rate": 7.440920868936282e-06,
+      "loss": 0.0574,
+      "step": 1648
+    },
+    {
+      "epoch": 0.7157118055555556,
+      "grad_norm": 0.6729372143745422,
+      "learning_rate": 7.437852694397615e-06,
+      "loss": 0.0492,
+      "step": 1649
+    },
+    {
+      "epoch": 0.7161458333333334,
+      "grad_norm": 0.677540123462677,
+      "learning_rate": 7.434783315137953e-06,
+      "loss": 0.046,
+      "step": 1650
+    },
+    {
+      "epoch": 0.7165798611111112,
+      "grad_norm": 0.7205467820167542,
+      "learning_rate": 7.431712732674103e-06,
+      "loss": 0.0416,
+      "step": 1651
+    },
+    {
+      "epoch": 0.7170138888888888,
+      "grad_norm": 0.7260408401489258,
+      "learning_rate": 7.428640948523463e-06,
+      "loss": 0.0257,
+      "step": 1652
+    },
+    {
+      "epoch": 0.7174479166666666,
+      "grad_norm": 0.9158640503883362,
+      "learning_rate": 7.4255679642040265e-06,
+      "loss": 0.0706,
+      "step": 1653
+    },
+    {
+      "epoch": 0.7178819444444444,
+      "grad_norm": 1.2473180294036865,
+      "learning_rate": 7.422493781234379e-06,
+      "loss": 0.0516,
+      "step": 1654
+    },
+    {
+      "epoch": 0.7183159722222222,
+      "grad_norm": 0.8920367956161499,
+      "learning_rate": 7.419418401133698e-06,
+      "loss": 0.0554,
+      "step": 1655
+    },
+    {
+      "epoch": 0.71875,
+      "grad_norm": 0.6255639791488647,
+      "learning_rate": 7.416341825421755e-06,
+      "loss": 0.0463,
+      "step": 1656
+    },
+    {
+      "epoch": 0.7191840277777778,
+      "grad_norm": 0.8922591209411621,
+      "learning_rate": 7.413264055618909e-06,
+      "loss": 0.0701,
+      "step": 1657
+    },
+    {
+      "epoch": 0.7196180555555556,
+      "grad_norm": 0.8104597330093384,
+      "learning_rate": 7.4101850932461095e-06,
+      "loss": 0.0695,
+      "step": 1658
+    },
+    {
+      "epoch": 0.7200520833333334,
+      "grad_norm": 0.5503954887390137,
+      "learning_rate": 7.407104939824901e-06,
+      "loss": 0.0359,
+      "step": 1659
+    },
+    {
+      "epoch": 0.7204861111111112,
+      "grad_norm": 0.5614795684814453,
+      "learning_rate": 7.404023596877409e-06,
+      "loss": 0.0535,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7209201388888888,
+      "grad_norm": 1.2029560804367065,
+      "learning_rate": 7.40094106592635e-06,
+      "loss": 0.067,
+      "step": 1661
+    },
+    {
+      "epoch": 0.7213541666666666,
+      "grad_norm": 0.5232289433479309,
+      "learning_rate": 7.39785734849503e-06,
+      "loss": 0.0389,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7217881944444444,
+      "grad_norm": 0.6728552579879761,
+      "learning_rate": 7.394772446107335e-06,
+      "loss": 0.0629,
+      "step": 1663
+    },
+    {
+      "epoch": 0.7222222222222222,
+      "grad_norm": 0.64680415391922,
+      "learning_rate": 7.391686360287744e-06,
+      "loss": 0.044,
+      "step": 1664
+    },
+    {
+      "epoch": 0.72265625,
+      "grad_norm": 0.5645890831947327,
+      "learning_rate": 7.388599092561315e-06,
+      "loss": 0.0496,
+      "step": 1665
+    },
+    {
+      "epoch": 0.7230902777777778,
+      "grad_norm": 0.6096482872962952,
+      "learning_rate": 7.385510644453694e-06,
+      "loss": 0.0591,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7235243055555556,
+      "grad_norm": 0.7313439846038818,
+      "learning_rate": 7.382421017491108e-06,
+      "loss": 0.0697,
+      "step": 1667
+    },
+    {
+      "epoch": 0.7239583333333334,
+      "grad_norm": 0.6447915434837341,
+      "learning_rate": 7.3793302132003665e-06,
+      "loss": 0.0465,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7243923611111112,
+      "grad_norm": 0.6097691655158997,
+      "learning_rate": 7.376238233108863e-06,
+      "loss": 0.0352,
+      "step": 1669
+    },
+    {
+      "epoch": 0.7248263888888888,
+      "grad_norm": 0.7632279992103577,
+      "learning_rate": 7.37314507874457e-06,
+      "loss": 0.055,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7252604166666666,
+      "grad_norm": 0.7166815996170044,
+      "learning_rate": 7.370050751636041e-06,
+      "loss": 0.0426,
+      "step": 1671
+    },
+    {
+      "epoch": 0.7256944444444444,
+      "grad_norm": 0.9069613814353943,
+      "learning_rate": 7.366955253312407e-06,
+      "loss": 0.0528,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7261284722222222,
+      "grad_norm": 0.7017959356307983,
+      "learning_rate": 7.3638585853033826e-06,
+      "loss": 0.0519,
+      "step": 1673
+    },
+    {
+      "epoch": 0.7265625,
+      "grad_norm": 0.594621479511261,
+      "learning_rate": 7.360760749139257e-06,
+      "loss": 0.0357,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7269965277777778,
+      "grad_norm": 0.7259089350700378,
+      "learning_rate": 7.357661746350895e-06,
+      "loss": 0.0497,
+      "step": 1675
+    },
+    {
+      "epoch": 0.7274305555555556,
+      "grad_norm": 0.4349896311759949,
+      "learning_rate": 7.354561578469743e-06,
+      "loss": 0.0321,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7278645833333334,
+      "grad_norm": 1.0102602243423462,
+      "learning_rate": 7.351460247027818e-06,
+      "loss": 0.0527,
+      "step": 1677
+    },
+    {
+      "epoch": 0.7282986111111112,
+      "grad_norm": 0.8198878169059753,
+      "learning_rate": 7.348357753557716e-06,
+      "loss": 0.0447,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7287326388888888,
+      "grad_norm": 0.8680114150047302,
+      "learning_rate": 7.345254099592605e-06,
+      "loss": 0.0543,
+      "step": 1679
+    },
+    {
+      "epoch": 0.7291666666666666,
+      "grad_norm": 0.9011032581329346,
+      "learning_rate": 7.342149286666226e-06,
+      "loss": 0.0598,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7296006944444444,
+      "grad_norm": 0.6858422756195068,
+      "learning_rate": 7.339043316312895e-06,
+      "loss": 0.0359,
+      "step": 1681
+    },
+    {
+      "epoch": 0.7300347222222222,
+      "grad_norm": 0.875012993812561,
+      "learning_rate": 7.335936190067497e-06,
+      "loss": 0.0593,
+      "step": 1682
+    },
+    {
+      "epoch": 0.73046875,
+      "grad_norm": 0.6261489987373352,
+      "learning_rate": 7.332827909465493e-06,
+      "loss": 0.0505,
+      "step": 1683
+    },
+    {
+      "epoch": 0.7309027777777778,
+      "grad_norm": 0.7853638529777527,
+      "learning_rate": 7.329718476042908e-06,
+      "loss": 0.0413,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7313368055555556,
+      "grad_norm": 0.6131523251533508,
+      "learning_rate": 7.326607891336342e-06,
+      "loss": 0.0472,
+      "step": 1685
+    },
+    {
+      "epoch": 0.7317708333333334,
+      "grad_norm": 0.9483081102371216,
+      "learning_rate": 7.323496156882961e-06,
+      "loss": 0.0606,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7322048611111112,
+      "grad_norm": 0.7218665480613708,
+      "learning_rate": 7.320383274220502e-06,
+      "loss": 0.0458,
+      "step": 1687
+    },
+    {
+      "epoch": 0.7326388888888888,
+      "grad_norm": 0.6520138382911682,
+      "learning_rate": 7.317269244887265e-06,
+      "loss": 0.0495,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7330729166666666,
+      "grad_norm": 0.4401592016220093,
+      "learning_rate": 7.314154070422121e-06,
+      "loss": 0.031,
+      "step": 1689
+    },
+    {
+      "epoch": 0.7335069444444444,
+      "grad_norm": 0.7516753673553467,
+      "learning_rate": 7.311037752364505e-06,
+      "loss": 0.0407,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7339409722222222,
+      "grad_norm": 0.3969035744667053,
+      "learning_rate": 7.307920292254417e-06,
+      "loss": 0.0274,
+      "step": 1691
+    },
+    {
+      "epoch": 0.734375,
+      "grad_norm": 0.773316502571106,
+      "learning_rate": 7.304801691632421e-06,
+      "loss": 0.0572,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7348090277777778,
+      "grad_norm": 0.665576696395874,
+      "learning_rate": 7.301681952039645e-06,
+      "loss": 0.0467,
+      "step": 1693
+    },
+    {
+      "epoch": 0.7352430555555556,
+      "grad_norm": 0.601359486579895,
+      "learning_rate": 7.298561075017783e-06,
+      "loss": 0.0457,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7356770833333334,
+      "grad_norm": 0.6760073304176331,
+      "learning_rate": 7.295439062109085e-06,
+      "loss": 0.058,
+      "step": 1695
+    },
+    {
+      "epoch": 0.7361111111111112,
+      "grad_norm": 1.0305536985397339,
+      "learning_rate": 7.292315914856366e-06,
+      "loss": 0.0546,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7365451388888888,
+      "grad_norm": 0.6392028331756592,
+      "learning_rate": 7.289191634803002e-06,
+      "loss": 0.0425,
+      "step": 1697
+    },
+    {
+      "epoch": 0.7369791666666666,
+      "grad_norm": 0.985002875328064,
+      "learning_rate": 7.286066223492928e-06,
+      "loss": 0.0679,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7374131944444444,
+      "grad_norm": 0.6694068312644958,
+      "learning_rate": 7.282939682470637e-06,
+      "loss": 0.029,
+      "step": 1699
+    },
+    {
+      "epoch": 0.7378472222222222,
+      "grad_norm": 0.41032591462135315,
+      "learning_rate": 7.2798120132811815e-06,
+      "loss": 0.0295,
+      "step": 1700
+    },
+    {
+      "epoch": 0.73828125,
+      "grad_norm": 0.7770684957504272,
+      "learning_rate": 7.276683217470171e-06,
+      "loss": 0.064,
+      "step": 1701
+    },
+    {
+      "epoch": 0.7387152777777778,
+      "grad_norm": 0.552871584892273,
+      "learning_rate": 7.273553296583773e-06,
+      "loss": 0.0394,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7391493055555556,
+      "grad_norm": 1.2697641849517822,
+      "learning_rate": 7.270422252168708e-06,
+      "loss": 0.0613,
+      "step": 1703
+    },
+    {
+      "epoch": 0.7395833333333334,
+      "grad_norm": 0.560497522354126,
+      "learning_rate": 7.267290085772256e-06,
+      "loss": 0.0303,
+      "step": 1704
+    },
+    {
+      "epoch": 0.7400173611111112,
+      "grad_norm": 0.6473203301429749,
+      "learning_rate": 7.264156798942247e-06,
+      "loss": 0.0397,
+      "step": 1705
+    },
+    {
+      "epoch": 0.7404513888888888,
+      "grad_norm": 0.6359472870826721,
+      "learning_rate": 7.2610223932270665e-06,
+      "loss": 0.0434,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7408854166666666,
+      "grad_norm": 0.7363631725311279,
+      "learning_rate": 7.2578868701756564e-06,
+      "loss": 0.0445,
+      "step": 1707
+    },
+    {
+      "epoch": 0.7413194444444444,
+      "grad_norm": 0.9180881381034851,
+      "learning_rate": 7.254750231337503e-06,
+      "loss": 0.0514,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7417534722222222,
+      "grad_norm": 1.7444438934326172,
+      "learning_rate": 7.251612478262653e-06,
+      "loss": 0.0694,
+      "step": 1709
+    },
+    {
+      "epoch": 0.7421875,
+      "grad_norm": 0.7270860075950623,
+      "learning_rate": 7.2484736125016965e-06,
+      "loss": 0.0561,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7426215277777778,
+      "grad_norm": 0.9425753951072693,
+      "learning_rate": 7.245333635605778e-06,
+      "loss": 0.0519,
+      "step": 1711
+    },
+    {
+      "epoch": 0.7430555555555556,
+      "grad_norm": 0.7384446263313293,
+      "learning_rate": 7.242192549126586e-06,
+      "loss": 0.0374,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7434895833333334,
+      "grad_norm": 0.6277630925178528,
+      "learning_rate": 7.239050354616365e-06,
+      "loss": 0.0412,
+      "step": 1713
+    },
+    {
+      "epoch": 0.7439236111111112,
+      "grad_norm": 1.0086884498596191,
+      "learning_rate": 7.235907053627899e-06,
+      "loss": 0.065,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7443576388888888,
+      "grad_norm": 1.4292784929275513,
+      "learning_rate": 7.232762647714525e-06,
+      "loss": 0.092,
+      "step": 1715
+    },
+    {
+      "epoch": 0.7447916666666666,
+      "grad_norm": 0.9445415139198303,
+      "learning_rate": 7.229617138430123e-06,
+      "loss": 0.0444,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7452256944444444,
+      "grad_norm": 0.821932852268219,
+      "learning_rate": 7.226470527329117e-06,
+      "loss": 0.0449,
+      "step": 1717
+    },
+    {
+      "epoch": 0.7456597222222222,
+      "grad_norm": 0.6048581600189209,
+      "learning_rate": 7.22332281596648e-06,
+      "loss": 0.0496,
+      "step": 1718
+    },
+    {
+      "epoch": 0.74609375,
+      "grad_norm": 0.5893974900245667,
+      "learning_rate": 7.220174005897724e-06,
+      "loss": 0.0543,
+      "step": 1719
+    },
+    {
+      "epoch": 0.7465277777777778,
+      "grad_norm": 0.702113687992096,
+      "learning_rate": 7.217024098678908e-06,
+      "loss": 0.0479,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7469618055555556,
+      "grad_norm": 0.7570499777793884,
+      "learning_rate": 7.213873095866628e-06,
+      "loss": 0.0529,
+      "step": 1721
+    },
+    {
+      "epoch": 0.7473958333333334,
+      "grad_norm": 0.8346821069717407,
+      "learning_rate": 7.210720999018027e-06,
+      "loss": 0.0518,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7478298611111112,
+      "grad_norm": 0.5654697418212891,
+      "learning_rate": 7.207567809690786e-06,
+      "loss": 0.0538,
+      "step": 1723
+    },
+    {
+      "epoch": 0.7482638888888888,
+      "grad_norm": 1.3470139503479004,
+      "learning_rate": 7.204413529443125e-06,
+      "loss": 0.0699,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7486979166666666,
+      "grad_norm": 0.6355326771736145,
+      "learning_rate": 7.2012581598338035e-06,
+      "loss": 0.0601,
+      "step": 1725
+    },
+    {
+      "epoch": 0.7491319444444444,
+      "grad_norm": 0.6898193359375,
+      "learning_rate": 7.198101702422123e-06,
+      "loss": 0.0562,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7495659722222222,
+      "grad_norm": 0.7929469347000122,
+      "learning_rate": 7.194944158767916e-06,
+      "loss": 0.067,
+      "step": 1727
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.7281990051269531,
+      "learning_rate": 7.1917855304315575e-06,
+      "loss": 0.0668,
+      "step": 1728
+    },
+    {
+      "epoch": 0.7504340277777778,
+      "grad_norm": 0.45239919424057007,
+      "learning_rate": 7.188625818973955e-06,
+      "loss": 0.0265,
+      "step": 1729
+    },
+    {
+      "epoch": 0.7508680555555556,
+      "grad_norm": 0.8641472458839417,
+      "learning_rate": 7.185465025956555e-06,
+      "loss": 0.07,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7513020833333334,
+      "grad_norm": 0.6607235670089722,
+      "learning_rate": 7.182303152941334e-06,
+      "loss": 0.0557,
+      "step": 1731
+    },
+    {
+      "epoch": 0.7517361111111112,
+      "grad_norm": 0.5443063378334045,
+      "learning_rate": 7.179140201490805e-06,
+      "loss": 0.0493,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7521701388888888,
+      "grad_norm": 0.6223754286766052,
+      "learning_rate": 7.1759761731680135e-06,
+      "loss": 0.0585,
+      "step": 1733
+    },
+    {
+      "epoch": 0.7526041666666666,
+      "grad_norm": 0.5347004532814026,
+      "learning_rate": 7.172811069536536e-06,
+      "loss": 0.0506,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7530381944444444,
+      "grad_norm": 0.509100079536438,
+      "learning_rate": 7.169644892160481e-06,
+      "loss": 0.0413,
+      "step": 1735
+    },
+    {
+      "epoch": 0.7534722222222222,
+      "grad_norm": 0.9070925712585449,
+      "learning_rate": 7.166477642604489e-06,
+      "loss": 0.0576,
+      "step": 1736
+    },
+    {
+      "epoch": 0.75390625,
+      "grad_norm": 0.46375447511672974,
+      "learning_rate": 7.163309322433728e-06,
+      "loss": 0.0442,
+      "step": 1737
+    },
+    {
+      "epoch": 0.7543402777777778,
+      "grad_norm": 1.7822531461715698,
+      "learning_rate": 7.160139933213899e-06,
+      "loss": 0.0693,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7547743055555556,
+      "grad_norm": 0.5103685855865479,
+      "learning_rate": 7.156969476511226e-06,
+      "loss": 0.0433,
+      "step": 1739
+    },
+    {
+      "epoch": 0.7552083333333334,
+      "grad_norm": 0.5952293276786804,
+      "learning_rate": 7.153797953892463e-06,
+      "loss": 0.0592,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7556423611111112,
+      "grad_norm": 0.8038705587387085,
+      "learning_rate": 7.150625366924893e-06,
+      "loss": 0.0575,
+      "step": 1741
+    },
+    {
+      "epoch": 0.7560763888888888,
+      "grad_norm": 1.7251653671264648,
+      "learning_rate": 7.14745171717632e-06,
+      "loss": 0.0634,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7565104166666666,
+      "grad_norm": 0.5570972561836243,
+      "learning_rate": 7.144277006215079e-06,
+      "loss": 0.048,
+      "step": 1743
+    },
+    {
+      "epoch": 0.7569444444444444,
+      "grad_norm": 0.6353147625923157,
+      "learning_rate": 7.141101235610025e-06,
+      "loss": 0.0506,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7573784722222222,
+      "grad_norm": 0.7536773085594177,
+      "learning_rate": 7.137924406930539e-06,
+      "loss": 0.0478,
+      "step": 1745
+    },
+    {
+      "epoch": 0.7578125,
+      "grad_norm": 0.6892389059066772,
+      "learning_rate": 7.134746521746521e-06,
+      "loss": 0.0473,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7582465277777778,
+      "grad_norm": 1.0137925148010254,
+      "learning_rate": 7.131567581628398e-06,
+      "loss": 0.0649,
+      "step": 1747
+    },
+    {
+      "epoch": 0.7586805555555556,
+      "grad_norm": 0.7530260682106018,
+      "learning_rate": 7.128387588147119e-06,
+      "loss": 0.0453,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7591145833333334,
+      "grad_norm": 0.8550921678543091,
+      "learning_rate": 7.125206542874146e-06,
+      "loss": 0.0603,
+      "step": 1749
+    },
+    {
+      "epoch": 0.7595486111111112,
+      "grad_norm": 0.42573830485343933,
+      "learning_rate": 7.122024447381469e-06,
+      "loss": 0.0342,
+      "step": 1750
+    },
+    {
+      "epoch": 0.7599826388888888,
+      "grad_norm": 0.6021143794059753,
+      "learning_rate": 7.118841303241594e-06,
+      "loss": 0.0416,
+      "step": 1751
+    },
+    {
+      "epoch": 0.7604166666666666,
+      "grad_norm": 0.6040723323822021,
+      "learning_rate": 7.115657112027543e-06,
+      "loss": 0.0495,
+      "step": 1752
+    },
+    {
+      "epoch": 0.7608506944444444,
+      "grad_norm": 0.5850236415863037,
+      "learning_rate": 7.112471875312858e-06,
+      "loss": 0.0542,
+      "step": 1753
+    },
+    {
+      "epoch": 0.7612847222222222,
+      "grad_norm": 0.5208216309547424,
+      "learning_rate": 7.109285594671599e-06,
+      "loss": 0.046,
+      "step": 1754
+    },
+    {
+      "epoch": 0.76171875,
+      "grad_norm": 0.6182832717895508,
+      "learning_rate": 7.106098271678337e-06,
+      "loss": 0.0562,
+      "step": 1755
+    },
+    {
+      "epoch": 0.7621527777777778,
+      "grad_norm": 0.6657155156135559,
+      "learning_rate": 7.102909907908161e-06,
+      "loss": 0.0701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7625868055555556,
+      "grad_norm": 0.62384033203125,
+      "learning_rate": 7.099720504936676e-06,
+      "loss": 0.06,
+      "step": 1757
+    },
+    {
+      "epoch": 0.7630208333333334,
+      "grad_norm": 0.8038895130157471,
+      "learning_rate": 7.096530064339997e-06,
+      "loss": 0.058,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7634548611111112,
+      "grad_norm": 0.999763548374176,
+      "learning_rate": 7.0933385876947566e-06,
+      "loss": 0.0506,
+      "step": 1759
+    },
+    {
+      "epoch": 0.7638888888888888,
+      "grad_norm": 0.6280719637870789,
+      "learning_rate": 7.090146076578095e-06,
+      "loss": 0.053,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7643229166666666,
+      "grad_norm": 0.7273646593093872,
+      "learning_rate": 7.086952532567664e-06,
+      "loss": 0.0512,
+      "step": 1761
+    },
+    {
+      "epoch": 0.7647569444444444,
+      "grad_norm": 0.822219967842102,
+      "learning_rate": 7.083757957241627e-06,
+      "loss": 0.0716,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7651909722222222,
+      "grad_norm": 1.076998233795166,
+      "learning_rate": 7.080562352178657e-06,
+      "loss": 0.0819,
+      "step": 1763
+    },
+    {
+      "epoch": 0.765625,
+      "grad_norm": 0.5817064642906189,
+      "learning_rate": 7.0773657189579355e-06,
+      "loss": 0.0407,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7660590277777778,
+      "grad_norm": 0.6132467985153198,
+      "learning_rate": 7.074168059159155e-06,
+      "loss": 0.0469,
+      "step": 1765
+    },
+    {
+      "epoch": 0.7664930555555556,
+      "grad_norm": 0.6296002864837646,
+      "learning_rate": 7.07096937436251e-06,
+      "loss": 0.045,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7669270833333334,
+      "grad_norm": 0.6748993992805481,
+      "learning_rate": 7.067769666148705e-06,
+      "loss": 0.0394,
+      "step": 1767
+    },
+    {
+      "epoch": 0.7673611111111112,
+      "grad_norm": 0.6105886697769165,
+      "learning_rate": 7.064568936098948e-06,
+      "loss": 0.0406,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7677951388888888,
+      "grad_norm": 0.8035129308700562,
+      "learning_rate": 7.061367185794956e-06,
+      "loss": 0.0612,
+      "step": 1769
+    },
+    {
+      "epoch": 0.7682291666666666,
+      "grad_norm": 1.034857153892517,
+      "learning_rate": 7.058164416818945e-06,
+      "loss": 0.0544,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7686631944444444,
+      "grad_norm": 0.629572331905365,
+      "learning_rate": 7.054960630753639e-06,
+      "loss": 0.0444,
+      "step": 1771
+    },
+    {
+      "epoch": 0.7690972222222222,
+      "grad_norm": 0.9120610356330872,
+      "learning_rate": 7.051755829182262e-06,
+      "loss": 0.0683,
+      "step": 1772
+    },
+    {
+      "epoch": 0.76953125,
+      "grad_norm": 0.6496632695198059,
+      "learning_rate": 7.04855001368854e-06,
+      "loss": 0.0565,
+      "step": 1773
+    },
+    {
+      "epoch": 0.7699652777777778,
+      "grad_norm": 0.5675666928291321,
+      "learning_rate": 7.045343185856701e-06,
+      "loss": 0.0415,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7703993055555556,
+      "grad_norm": 0.6537423133850098,
+      "learning_rate": 7.042135347271472e-06,
+      "loss": 0.0598,
+      "step": 1775
+    },
+    {
+      "epoch": 0.7708333333333334,
+      "grad_norm": 0.7533829808235168,
+      "learning_rate": 7.038926499518079e-06,
+      "loss": 0.0715,
+      "step": 1776
+    },
+    {
+      "epoch": 0.7712673611111112,
+      "grad_norm": 0.529955267906189,
+      "learning_rate": 7.035716644182251e-06,
+      "loss": 0.0459,
+      "step": 1777
+    },
+    {
+      "epoch": 0.7717013888888888,
+      "grad_norm": 0.5904744863510132,
+      "learning_rate": 7.032505782850209e-06,
+      "loss": 0.0398,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7721354166666666,
+      "grad_norm": 0.4565054178237915,
+      "learning_rate": 7.029293917108678e-06,
+      "loss": 0.0445,
+      "step": 1779
+    },
+    {
+      "epoch": 0.7725694444444444,
+      "grad_norm": 0.5179257392883301,
+      "learning_rate": 7.026081048544873e-06,
+      "loss": 0.0374,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7730034722222222,
+      "grad_norm": 0.4564122259616852,
+      "learning_rate": 7.022867178746506e-06,
+      "loss": 0.0397,
+      "step": 1781
+    },
+    {
+      "epoch": 0.7734375,
+      "grad_norm": 0.9230285286903381,
+      "learning_rate": 7.019652309301785e-06,
+      "loss": 0.0634,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7738715277777778,
+      "grad_norm": 0.6400481462478638,
+      "learning_rate": 7.016436441799412e-06,
+      "loss": 0.0603,
+      "step": 1783
+    },
+    {
+      "epoch": 0.7743055555555556,
+      "grad_norm": 0.7093133330345154,
+      "learning_rate": 7.0132195778285806e-06,
+      "loss": 0.0433,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7747395833333334,
+      "grad_norm": 0.5242016315460205,
+      "learning_rate": 7.010001718978981e-06,
+      "loss": 0.0424,
+      "step": 1785
+    },
+    {
+      "epoch": 0.7751736111111112,
+      "grad_norm": 0.5063726305961609,
+      "learning_rate": 7.006782866840791e-06,
+      "loss": 0.0332,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7756076388888888,
+      "grad_norm": 0.8927604556083679,
+      "learning_rate": 7.00356302300468e-06,
+      "loss": 0.0459,
+      "step": 1787
+    },
+    {
+      "epoch": 0.7760416666666666,
+      "grad_norm": 0.6253120303153992,
+      "learning_rate": 7.0003421890618065e-06,
+      "loss": 0.0442,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7764756944444444,
+      "grad_norm": 0.6660035848617554,
+      "learning_rate": 6.997120366603823e-06,
+      "loss": 0.0479,
+      "step": 1789
+    },
+    {
+      "epoch": 0.7769097222222222,
+      "grad_norm": 0.7846994996070862,
+      "learning_rate": 6.993897557222865e-06,
+      "loss": 0.0601,
+      "step": 1790
+    },
+    {
+      "epoch": 0.77734375,
+      "grad_norm": 0.4856984317302704,
+      "learning_rate": 6.990673762511557e-06,
+      "loss": 0.03,
+      "step": 1791
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.8482086062431335,
+      "learning_rate": 6.987448984063014e-06,
+      "loss": 0.0578,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7782118055555556,
+      "grad_norm": 1.2565405368804932,
+      "learning_rate": 6.984223223470834e-06,
+      "loss": 0.0745,
+      "step": 1793
+    },
+    {
+      "epoch": 0.7786458333333334,
+      "grad_norm": 1.0909104347229004,
+      "learning_rate": 6.9809964823291e-06,
+      "loss": 0.0482,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7790798611111112,
+      "grad_norm": 0.6057597398757935,
+      "learning_rate": 6.977768762232379e-06,
+      "loss": 0.0464,
+      "step": 1795
+    },
+    {
+      "epoch": 0.7795138888888888,
+      "grad_norm": 0.8874408602714539,
+      "learning_rate": 6.974540064775726e-06,
+      "loss": 0.0598,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7799479166666666,
+      "grad_norm": 0.5312131643295288,
+      "learning_rate": 6.971310391554671e-06,
+      "loss": 0.0401,
+      "step": 1797
+    },
+    {
+      "epoch": 0.7803819444444444,
+      "grad_norm": 0.46000969409942627,
+      "learning_rate": 6.968079744165238e-06,
+      "loss": 0.0337,
+      "step": 1798
+    },
+    {
+      "epoch": 0.7808159722222222,
+      "grad_norm": 0.6547852158546448,
+      "learning_rate": 6.96484812420392e-06,
+      "loss": 0.0497,
+      "step": 1799
+    },
+    {
+      "epoch": 0.78125,
+      "grad_norm": 0.7223719358444214,
+      "learning_rate": 6.961615533267701e-06,
+      "loss": 0.0327,
+      "step": 1800
+    },
+    {
+      "epoch": 0.78125,
+      "eval_layout_rm_rank_test_accuracy": 0.9675206041167634,
+      "eval_layout_rm_rank_test_chosen_scores": 0.6561079025268555,
+      "eval_layout_rm_rank_test_loss": 0.04797226935625076,
+      "eval_layout_rm_rank_test_rejected_scores": 0.10819125175476074,
+      "eval_layout_rm_rank_test_runtime": 1182.7169,
+      "eval_layout_rm_rank_test_samples_per_second": 20.826,
+      "eval_layout_rm_rank_test_steps_per_second": 0.651,
+      "step": 1800
+    },
+    {
+      "epoch": 0.7816840277777778,
+      "grad_norm": 0.7490624785423279,
+      "learning_rate": 6.958381972954038e-06,
+      "loss": 0.0452,
+      "step": 1801
+    },
+    {
+      "epoch": 0.7821180555555556,
+      "grad_norm": 0.7055108547210693,
+      "learning_rate": 6.95514744486087e-06,
+      "loss": 0.0483,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7825520833333334,
+      "grad_norm": 0.7814860939979553,
+      "learning_rate": 6.9519119505866125e-06,
+      "loss": 0.0674,
+      "step": 1803
+    },
+    {
+      "epoch": 0.7829861111111112,
+      "grad_norm": 0.649394690990448,
+      "learning_rate": 6.948675491730159e-06,
+      "loss": 0.0329,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7834201388888888,
+      "grad_norm": 0.5071682929992676,
+      "learning_rate": 6.945438069890881e-06,
+      "loss": 0.0267,
+      "step": 1805
+    },
+    {
+      "epoch": 0.7838541666666666,
+      "grad_norm": 0.6328599452972412,
+      "learning_rate": 6.942199686668625e-06,
+      "loss": 0.0353,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7842881944444444,
+      "grad_norm": 0.7604709267616272,
+      "learning_rate": 6.938960343663712e-06,
+      "loss": 0.0465,
+      "step": 1807
+    },
+    {
+      "epoch": 0.7847222222222222,
+      "grad_norm": 0.693663477897644,
+      "learning_rate": 6.935720042476939e-06,
+      "loss": 0.0404,
+      "step": 1808
+    },
+    {
+      "epoch": 0.78515625,
+      "grad_norm": 0.9226562976837158,
+      "learning_rate": 6.932478784709574e-06,
+      "loss": 0.034,
+      "step": 1809
+    },
+    {
+      "epoch": 0.7855902777777778,
+      "grad_norm": 1.2657535076141357,
+      "learning_rate": 6.9292365719633605e-06,
+      "loss": 0.0932,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7860243055555556,
+      "grad_norm": 0.7438696026802063,
+      "learning_rate": 6.9259934058405124e-06,
+      "loss": 0.0393,
+      "step": 1811
+    },
+    {
+      "epoch": 0.7864583333333334,
+      "grad_norm": 0.6592585444450378,
+      "learning_rate": 6.922749287943711e-06,
+      "loss": 0.0406,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7868923611111112,
+      "grad_norm": 0.8676554560661316,
+      "learning_rate": 6.919504219876117e-06,
+      "loss": 0.0396,
+      "step": 1813
+    },
+    {
+      "epoch": 0.7873263888888888,
+      "grad_norm": 0.7734280228614807,
+      "learning_rate": 6.916258203241355e-06,
+      "loss": 0.0504,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7877604166666666,
+      "grad_norm": 0.8438120484352112,
+      "learning_rate": 6.913011239643516e-06,
+      "loss": 0.0368,
+      "step": 1815
+    },
+    {
+      "epoch": 0.7881944444444444,
+      "grad_norm": 0.975790798664093,
+      "learning_rate": 6.909763330687162e-06,
+      "loss": 0.0591,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7886284722222222,
+      "grad_norm": 0.7018232345581055,
+      "learning_rate": 6.906514477977324e-06,
+      "loss": 0.0397,
+      "step": 1817
+    },
+    {
+      "epoch": 0.7890625,
+      "grad_norm": 0.914828360080719,
+      "learning_rate": 6.903264683119496e-06,
+      "loss": 0.0456,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7894965277777778,
+      "grad_norm": 0.9112297296524048,
+      "learning_rate": 6.900013947719639e-06,
+      "loss": 0.0513,
+      "step": 1819
+    },
+    {
+      "epoch": 0.7899305555555556,
+      "grad_norm": 0.5047491788864136,
+      "learning_rate": 6.896762273384179e-06,
+      "loss": 0.0494,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7903645833333334,
+      "grad_norm": 0.7444157004356384,
+      "learning_rate": 6.893509661720005e-06,
+      "loss": 0.0515,
+      "step": 1821
+    },
+    {
+      "epoch": 0.7907986111111112,
+      "grad_norm": 0.61995929479599,
+      "learning_rate": 6.890256114334472e-06,
+      "loss": 0.0615,
+      "step": 1822
+    },
+    {
+      "epoch": 0.7912326388888888,
+      "grad_norm": 1.0646355152130127,
+      "learning_rate": 6.887001632835392e-06,
+      "loss": 0.0932,
+      "step": 1823
+    },
+    {
+      "epoch": 0.7916666666666666,
+      "grad_norm": 1.0587164163589478,
+      "learning_rate": 6.883746218831045e-06,
+      "loss": 0.0614,
+      "step": 1824
+    },
+    {
+      "epoch": 0.7921006944444444,
+      "grad_norm": 0.7038317322731018,
+      "learning_rate": 6.8804898739301654e-06,
+      "loss": 0.0428,
+      "step": 1825
+    },
+    {
+      "epoch": 0.7925347222222222,
+      "grad_norm": 0.5331775546073914,
+      "learning_rate": 6.877232599741953e-06,
+      "loss": 0.0479,
+      "step": 1826
+    },
+    {
+      "epoch": 0.79296875,
+      "grad_norm": 0.43616053462028503,
+      "learning_rate": 6.873974397876067e-06,
+      "loss": 0.0347,
+      "step": 1827
+    },
+    {
+      "epoch": 0.7934027777777778,
+      "grad_norm": 0.5002326369285583,
+      "learning_rate": 6.8707152699426184e-06,
+      "loss": 0.0447,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7938368055555556,
+      "grad_norm": 0.6660779714584351,
+      "learning_rate": 6.867455217552183e-06,
+      "loss": 0.0476,
+      "step": 1829
+    },
+    {
+      "epoch": 0.7942708333333334,
+      "grad_norm": 0.5385332703590393,
+      "learning_rate": 6.864194242315788e-06,
+      "loss": 0.0496,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7947048611111112,
+      "grad_norm": 0.5744720697402954,
+      "learning_rate": 6.8609323458449236e-06,
+      "loss": 0.047,
+      "step": 1831
+    },
+    {
+      "epoch": 0.7951388888888888,
+      "grad_norm": 0.7200570106506348,
+      "learning_rate": 6.857669529751529e-06,
+      "loss": 0.0519,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7955729166666666,
+      "grad_norm": 0.5372190475463867,
+      "learning_rate": 6.854405795647997e-06,
+      "loss": 0.0387,
+      "step": 1833
+    },
+    {
+      "epoch": 0.7960069444444444,
+      "grad_norm": 0.5767813324928284,
+      "learning_rate": 6.851141145147181e-06,
+      "loss": 0.0334,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7964409722222222,
+      "grad_norm": 1.3028239011764526,
+      "learning_rate": 6.847875579862382e-06,
+      "loss": 0.0646,
+      "step": 1835
+    },
+    {
+      "epoch": 0.796875,
+      "grad_norm": 0.8541083931922913,
+      "learning_rate": 6.844609101407354e-06,
+      "loss": 0.0516,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7973090277777778,
+      "grad_norm": 0.6067532300949097,
+      "learning_rate": 6.841341711396301e-06,
+      "loss": 0.0381,
+      "step": 1837
+    },
+    {
+      "epoch": 0.7977430555555556,
+      "grad_norm": 0.9461897015571594,
+      "learning_rate": 6.838073411443882e-06,
+      "loss": 0.0478,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7981770833333334,
+      "grad_norm": 1.1896733045578003,
+      "learning_rate": 6.834804203165199e-06,
+      "loss": 0.0594,
+      "step": 1839
+    },
+    {
+      "epoch": 0.7986111111111112,
+      "grad_norm": 0.8283300995826721,
+      "learning_rate": 6.831534088175807e-06,
+      "loss": 0.0567,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7990451388888888,
+      "grad_norm": 0.8821940422058105,
+      "learning_rate": 6.828263068091711e-06,
+      "loss": 0.0367,
+      "step": 1841
+    },
+    {
+      "epoch": 0.7994791666666666,
+      "grad_norm": 0.7667244672775269,
+      "learning_rate": 6.824991144529362e-06,
+      "loss": 0.054,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7999131944444444,
+      "grad_norm": 0.5846320986747742,
+      "learning_rate": 6.8217183191056524e-06,
+      "loss": 0.0365,
+      "step": 1843
+    },
+    {
+      "epoch": 0.8003472222222222,
+      "grad_norm": 0.6740960478782654,
+      "learning_rate": 6.8184445934379255e-06,
+      "loss": 0.0477,
+      "step": 1844
+    },
+    {
+      "epoch": 0.80078125,
+      "grad_norm": 0.7076245546340942,
+      "learning_rate": 6.815169969143969e-06,
+      "loss": 0.0455,
+      "step": 1845
+    },
+    {
+      "epoch": 0.8012152777777778,
+      "grad_norm": 0.6552544236183167,
+      "learning_rate": 6.811894447842012e-06,
+      "loss": 0.0389,
+      "step": 1846
+    },
+    {
+      "epoch": 0.8016493055555556,
+      "grad_norm": 0.912534773349762,
+      "learning_rate": 6.80861803115073e-06,
+      "loss": 0.094,
+      "step": 1847
+    },
+    {
+      "epoch": 0.8020833333333334,
+      "grad_norm": 0.5448792576789856,
+      "learning_rate": 6.805340720689241e-06,
+      "loss": 0.0419,
+      "step": 1848
+    },
+    {
+      "epoch": 0.8025173611111112,
+      "grad_norm": 0.6828315258026123,
+      "learning_rate": 6.802062518077101e-06,
+      "loss": 0.0493,
+      "step": 1849
+    },
+    {
+      "epoch": 0.8029513888888888,
+      "grad_norm": 0.5973538756370544,
+      "learning_rate": 6.7987834249343096e-06,
+      "loss": 0.0523,
+      "step": 1850
+    },
+    {
+      "epoch": 0.8033854166666666,
+      "grad_norm": 0.7848672866821289,
+      "learning_rate": 6.795503442881307e-06,
+      "loss": 0.0468,
+      "step": 1851
+    },
+    {
+      "epoch": 0.8038194444444444,
+      "grad_norm": 0.4816887676715851,
+      "learning_rate": 6.7922225735389726e-06,
+      "loss": 0.0414,
+      "step": 1852
+    },
+    {
+      "epoch": 0.8042534722222222,
+      "grad_norm": 1.0304285287857056,
+      "learning_rate": 6.788940818528622e-06,
+      "loss": 0.0619,
+      "step": 1853
+    },
+    {
+      "epoch": 0.8046875,
+      "grad_norm": 0.5743433833122253,
+      "learning_rate": 6.78565817947201e-06,
+      "loss": 0.0493,
+      "step": 1854
+    },
+    {
+      "epoch": 0.8051215277777778,
+      "grad_norm": 0.6866853833198547,
+      "learning_rate": 6.782374657991329e-06,
+      "loss": 0.0394,
+      "step": 1855
+    },
+    {
+      "epoch": 0.8055555555555556,
+      "grad_norm": 0.8010555505752563,
+      "learning_rate": 6.779090255709205e-06,
+      "loss": 0.0506,
+      "step": 1856
+    },
+    {
+      "epoch": 0.8059895833333334,
+      "grad_norm": 0.6958445906639099,
+      "learning_rate": 6.775804974248703e-06,
+      "loss": 0.0535,
+      "step": 1857
+    },
+    {
+      "epoch": 0.8064236111111112,
+      "grad_norm": 0.6616341471672058,
+      "learning_rate": 6.77251881523332e-06,
+      "loss": 0.0438,
+      "step": 1858
+    },
+    {
+      "epoch": 0.8068576388888888,
+      "grad_norm": 0.6229743957519531,
+      "learning_rate": 6.769231780286987e-06,
+      "loss": 0.0401,
+      "step": 1859
+    },
+    {
+      "epoch": 0.8072916666666666,
+      "grad_norm": 0.7217589020729065,
+      "learning_rate": 6.7659438710340654e-06,
+      "loss": 0.054,
+      "step": 1860
+    },
+    {
+      "epoch": 0.8077256944444444,
+      "grad_norm": 0.5400440096855164,
+      "learning_rate": 6.762655089099353e-06,
+      "loss": 0.0401,
+      "step": 1861
+    },
+    {
+      "epoch": 0.8081597222222222,
+      "grad_norm": 1.5119469165802002,
+      "learning_rate": 6.759365436108079e-06,
+      "loss": 0.0661,
+      "step": 1862
+    },
+    {
+      "epoch": 0.80859375,
+      "grad_norm": 0.8494582772254944,
+      "learning_rate": 6.7560749136859005e-06,
+      "loss": 0.0482,
+      "step": 1863
+    },
+    {
+      "epoch": 0.8090277777777778,
+      "grad_norm": 2.100743055343628,
+      "learning_rate": 6.752783523458903e-06,
+      "loss": 0.0657,
+      "step": 1864
+    },
+    {
+      "epoch": 0.8094618055555556,
+      "grad_norm": 0.5947306156158447,
+      "learning_rate": 6.749491267053604e-06,
+      "loss": 0.0436,
+      "step": 1865
+    },
+    {
+      "epoch": 0.8098958333333334,
+      "grad_norm": 0.47346755862236023,
+      "learning_rate": 6.746198146096947e-06,
+      "loss": 0.0392,
+      "step": 1866
+    },
+    {
+      "epoch": 0.8103298611111112,
+      "grad_norm": 0.7026275992393494,
+      "learning_rate": 6.742904162216304e-06,
+      "loss": 0.0376,
+      "step": 1867
+    },
+    {
+      "epoch": 0.8107638888888888,
+      "grad_norm": 0.6437444686889648,
+      "learning_rate": 6.739609317039471e-06,
+      "loss": 0.0507,
+      "step": 1868
+    },
+    {
+      "epoch": 0.8111979166666666,
+      "grad_norm": 0.5360928177833557,
+      "learning_rate": 6.736313612194676e-06,
+      "loss": 0.0499,
+      "step": 1869
+    },
+    {
+      "epoch": 0.8116319444444444,
+      "grad_norm": 0.6127115488052368,
+      "learning_rate": 6.733017049310563e-06,
+      "loss": 0.0353,
+      "step": 1870
+    },
+    {
+      "epoch": 0.8120659722222222,
+      "grad_norm": 0.7498536109924316,
+      "learning_rate": 6.729719630016206e-06,
+      "loss": 0.0312,
+      "step": 1871
+    },
+    {
+      "epoch": 0.8125,
+      "grad_norm": 0.8961223363876343,
+      "learning_rate": 6.7264213559411e-06,
+      "loss": 0.0715,
+      "step": 1872
+    },
+    {
+      "epoch": 0.8129340277777778,
+      "grad_norm": 0.6380618214607239,
+      "learning_rate": 6.723122228715167e-06,
+      "loss": 0.0335,
+      "step": 1873
+    },
+    {
+      "epoch": 0.8133680555555556,
+      "grad_norm": 0.7774998545646667,
+      "learning_rate": 6.719822249968739e-06,
+      "loss": 0.0327,
+      "step": 1874
+    },
+    {
+      "epoch": 0.8138020833333334,
+      "grad_norm": 0.8536900281906128,
+      "learning_rate": 6.716521421332582e-06,
+      "loss": 0.0489,
+      "step": 1875
+    },
+    {
+      "epoch": 0.8142361111111112,
+      "grad_norm": 1.3191429376602173,
+      "learning_rate": 6.713219744437876e-06,
+      "loss": 0.1005,
+      "step": 1876
+    },
+    {
+      "epoch": 0.8146701388888888,
+      "grad_norm": 0.8960707783699036,
+      "learning_rate": 6.70991722091622e-06,
+      "loss": 0.0474,
+      "step": 1877
+    },
+    {
+      "epoch": 0.8151041666666666,
+      "grad_norm": 0.7469573020935059,
+      "learning_rate": 6.706613852399632e-06,
+      "loss": 0.0513,
+      "step": 1878
+    },
+    {
+      "epoch": 0.8155381944444444,
+      "grad_norm": 0.6806800365447998,
+      "learning_rate": 6.703309640520546e-06,
+      "loss": 0.0354,
+      "step": 1879
+    },
+    {
+      "epoch": 0.8159722222222222,
+      "grad_norm": 0.880378246307373,
+      "learning_rate": 6.7000045869118154e-06,
+      "loss": 0.0615,
+      "step": 1880
+    },
+    {
+      "epoch": 0.81640625,
+      "grad_norm": 0.8611704707145691,
+      "learning_rate": 6.696698693206708e-06,
+      "loss": 0.0499,
+      "step": 1881
+    },
+    {
+      "epoch": 0.8168402777777778,
+      "grad_norm": 0.6044244170188904,
+      "learning_rate": 6.6933919610389086e-06,
+      "loss": 0.0374,
+      "step": 1882
+    },
+    {
+      "epoch": 0.8172743055555556,
+      "grad_norm": 0.675598680973053,
+      "learning_rate": 6.690084392042514e-06,
+      "loss": 0.0549,
+      "step": 1883
+    },
+    {
+      "epoch": 0.8177083333333334,
+      "grad_norm": 0.8580988645553589,
+      "learning_rate": 6.686775987852037e-06,
+      "loss": 0.0593,
+      "step": 1884
+    },
+    {
+      "epoch": 0.8181423611111112,
+      "grad_norm": 0.4779009521007538,
+      "learning_rate": 6.683466750102399e-06,
+      "loss": 0.039,
+      "step": 1885
+    },
+    {
+      "epoch": 0.8185763888888888,
+      "grad_norm": 0.6775459051132202,
+      "learning_rate": 6.680156680428938e-06,
+      "loss": 0.0457,
+      "step": 1886
+    },
+    {
+      "epoch": 0.8190104166666666,
+      "grad_norm": 0.649315357208252,
+      "learning_rate": 6.6768457804674e-06,
+      "loss": 0.05,
+      "step": 1887
+    },
+    {
+      "epoch": 0.8194444444444444,
+      "grad_norm": 0.7105356454849243,
+      "learning_rate": 6.673534051853942e-06,
+      "loss": 0.0519,
+      "step": 1888
+    },
+    {
+      "epoch": 0.8198784722222222,
+      "grad_norm": 0.5776419043540955,
+      "learning_rate": 6.670221496225133e-06,
+      "loss": 0.0439,
+      "step": 1889
+    },
+    {
+      "epoch": 0.8203125,
+      "grad_norm": 0.6759132146835327,
+      "learning_rate": 6.666908115217947e-06,
+      "loss": 0.0544,
+      "step": 1890
+    },
+    {
+      "epoch": 0.8207465277777778,
+      "grad_norm": 0.6925734281539917,
+      "learning_rate": 6.6635939104697655e-06,
+      "loss": 0.039,
+      "step": 1891
+    },
+    {
+      "epoch": 0.8211805555555556,
+      "grad_norm": 0.3945949673652649,
+      "learning_rate": 6.660278883618384e-06,
+      "loss": 0.0246,
+      "step": 1892
+    },
+    {
+      "epoch": 0.8216145833333334,
+      "grad_norm": 1.0184353590011597,
+      "learning_rate": 6.656963036301995e-06,
+      "loss": 0.0539,
+      "step": 1893
+    },
+    {
+      "epoch": 0.8220486111111112,
+      "grad_norm": 0.7225958108901978,
+      "learning_rate": 6.653646370159204e-06,
+      "loss": 0.0419,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8224826388888888,
+      "grad_norm": 0.8667299151420593,
+      "learning_rate": 6.650328886829015e-06,
+      "loss": 0.0489,
+      "step": 1895
+    },
+    {
+      "epoch": 0.8229166666666666,
+      "grad_norm": 0.597955584526062,
+      "learning_rate": 6.647010587950838e-06,
+      "loss": 0.0338,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8233506944444444,
+      "grad_norm": 0.8427351117134094,
+      "learning_rate": 6.643691475164491e-06,
+      "loss": 0.0369,
+      "step": 1897
+    },
+    {
+      "epoch": 0.8237847222222222,
+      "grad_norm": 0.7786532640457153,
+      "learning_rate": 6.6403715501101875e-06,
+      "loss": 0.0411,
+      "step": 1898
+    },
+    {
+      "epoch": 0.82421875,
+      "grad_norm": 0.8530535101890564,
+      "learning_rate": 6.637050814428544e-06,
+      "loss": 0.0494,
+      "step": 1899
+    },
+    {
+      "epoch": 0.8246527777777778,
+      "grad_norm": 1.1018110513687134,
+      "learning_rate": 6.6337292697605796e-06,
+      "loss": 0.0582,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8250868055555556,
+      "grad_norm": 0.9957329630851746,
+      "learning_rate": 6.63040691774771e-06,
+      "loss": 0.0593,
+      "step": 1901
+    },
+    {
+      "epoch": 0.8255208333333334,
+      "grad_norm": 1.0684702396392822,
+      "learning_rate": 6.627083760031755e-06,
+      "loss": 0.0597,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8259548611111112,
+      "grad_norm": 1.0910362005233765,
+      "learning_rate": 6.623759798254927e-06,
+      "loss": 0.0489,
+      "step": 1903
+    },
+    {
+      "epoch": 0.8263888888888888,
+      "grad_norm": 0.6104505062103271,
+      "learning_rate": 6.62043503405984e-06,
+      "loss": 0.0385,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8268229166666666,
+      "grad_norm": 1.3332817554473877,
+      "learning_rate": 6.617109469089502e-06,
+      "loss": 0.0629,
+      "step": 1905
+    },
+    {
+      "epoch": 0.8272569444444444,
+      "grad_norm": 0.4675104022026062,
+      "learning_rate": 6.613783104987318e-06,
+      "loss": 0.0468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8276909722222222,
+      "grad_norm": 0.4829596281051636,
+      "learning_rate": 6.610455943397086e-06,
+      "loss": 0.036,
+      "step": 1907
+    },
+    {
+      "epoch": 0.828125,
+      "grad_norm": 0.6301221251487732,
+      "learning_rate": 6.607127985963e-06,
+      "loss": 0.0441,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8285590277777778,
+      "grad_norm": 0.8131287097930908,
+      "learning_rate": 6.603799234329646e-06,
+      "loss": 0.0526,
+      "step": 1909
+    },
+    {
+      "epoch": 0.8289930555555556,
+      "grad_norm": 0.43327629566192627,
+      "learning_rate": 6.600469690142005e-06,
+      "loss": 0.0357,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8294270833333334,
+      "grad_norm": 0.3900514841079712,
+      "learning_rate": 6.597139355045448e-06,
+      "loss": 0.0245,
+      "step": 1911
+    },
+    {
+      "epoch": 0.8298611111111112,
+      "grad_norm": 0.6519450545310974,
+      "learning_rate": 6.593808230685733e-06,
+      "loss": 0.0433,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8302951388888888,
+      "grad_norm": 0.6983678340911865,
+      "learning_rate": 6.590476318709016e-06,
+      "loss": 0.0445,
+      "step": 1913
+    },
+    {
+      "epoch": 0.8307291666666666,
+      "grad_norm": 1.0150282382965088,
+      "learning_rate": 6.587143620761835e-06,
+      "loss": 0.0482,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8311631944444444,
+      "grad_norm": 0.6016936898231506,
+      "learning_rate": 6.583810138491122e-06,
+      "loss": 0.0355,
+      "step": 1915
+    },
+    {
+      "epoch": 0.8315972222222222,
+      "grad_norm": 0.873637318611145,
+      "learning_rate": 6.580475873544193e-06,
+      "loss": 0.0539,
+      "step": 1916
+    },
+    {
+      "epoch": 0.83203125,
+      "grad_norm": 0.7255651354789734,
+      "learning_rate": 6.57714082756875e-06,
+      "loss": 0.0502,
+      "step": 1917
+    },
+    {
+      "epoch": 0.8324652777777778,
+      "grad_norm": 0.9832391738891602,
+      "learning_rate": 6.573805002212884e-06,
+      "loss": 0.0773,
+      "step": 1918
+    },
+    {
+      "epoch": 0.8328993055555556,
+      "grad_norm": 0.8161395192146301,
+      "learning_rate": 6.570468399125071e-06,
+      "loss": 0.0571,
+      "step": 1919
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 0.6000357270240784,
+      "learning_rate": 6.567131019954168e-06,
+      "loss": 0.0406,
+      "step": 1920
+    },
+    {
+      "epoch": 0.8337673611111112,
+      "grad_norm": 0.5935305953025818,
+      "learning_rate": 6.56379286634942e-06,
+      "loss": 0.0428,
+      "step": 1921
+    },
+    {
+      "epoch": 0.8342013888888888,
+      "grad_norm": 0.9758431315422058,
+      "learning_rate": 6.56045393996045e-06,
+      "loss": 0.0543,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8346354166666666,
+      "grad_norm": 0.7438514232635498,
+      "learning_rate": 6.557114242437266e-06,
+      "loss": 0.0494,
+      "step": 1923
+    },
+    {
+      "epoch": 0.8350694444444444,
+      "grad_norm": 0.5088344216346741,
+      "learning_rate": 6.553773775430256e-06,
+      "loss": 0.0467,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8355034722222222,
+      "grad_norm": 0.7656986713409424,
+      "learning_rate": 6.550432540590192e-06,
+      "loss": 0.0609,
+      "step": 1925
+    },
+    {
+      "epoch": 0.8359375,
+      "grad_norm": 0.5904448628425598,
+      "learning_rate": 6.5470905395682186e-06,
+      "loss": 0.0364,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8363715277777778,
+      "grad_norm": 0.7087653875350952,
+      "learning_rate": 6.543747774015863e-06,
+      "loss": 0.0426,
+      "step": 1927
+    },
+    {
+      "epoch": 0.8368055555555556,
+      "grad_norm": 0.838172435760498,
+      "learning_rate": 6.5404042455850306e-06,
+      "loss": 0.0612,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8372395833333334,
+      "grad_norm": 0.5433200597763062,
+      "learning_rate": 6.537059955928003e-06,
+      "loss": 0.0479,
+      "step": 1929
+    },
+    {
+      "epoch": 0.8376736111111112,
+      "grad_norm": 0.5060282349586487,
+      "learning_rate": 6.5337149066974345e-06,
+      "loss": 0.0388,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8381076388888888,
+      "grad_norm": 0.6122724413871765,
+      "learning_rate": 6.530369099546363e-06,
+      "loss": 0.042,
+      "step": 1931
+    },
+    {
+      "epoch": 0.8385416666666666,
+      "grad_norm": 0.9471703767776489,
+      "learning_rate": 6.5270225361281925e-06,
+      "loss": 0.0543,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8389756944444444,
+      "grad_norm": 0.8253403306007385,
+      "learning_rate": 6.523675218096707e-06,
+      "loss": 0.0689,
+      "step": 1933
+    },
+    {
+      "epoch": 0.8394097222222222,
+      "grad_norm": 1.0257596969604492,
+      "learning_rate": 6.520327147106059e-06,
+      "loss": 0.062,
+      "step": 1934
+    },
+    {
+      "epoch": 0.83984375,
+      "grad_norm": 0.5009595155715942,
+      "learning_rate": 6.516978324810777e-06,
+      "loss": 0.0353,
+      "step": 1935
+    },
+    {
+      "epoch": 0.8402777777777778,
+      "grad_norm": 0.7189043760299683,
+      "learning_rate": 6.513628752865755e-06,
+      "loss": 0.0679,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8407118055555556,
+      "grad_norm": 0.9218528270721436,
+      "learning_rate": 6.5102784329262626e-06,
+      "loss": 0.051,
+      "step": 1937
+    },
+    {
+      "epoch": 0.8411458333333334,
+      "grad_norm": 0.6909580230712891,
+      "learning_rate": 6.506927366647937e-06,
+      "loss": 0.0682,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8415798611111112,
+      "grad_norm": 0.5428255796432495,
+      "learning_rate": 6.503575555686787e-06,
+      "loss": 0.042,
+      "step": 1939
+    },
+    {
+      "epoch": 0.8420138888888888,
+      "grad_norm": 0.5663094520568848,
+      "learning_rate": 6.500223001699187e-06,
+      "loss": 0.0321,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8424479166666666,
+      "grad_norm": 0.5367885231971741,
+      "learning_rate": 6.496869706341875e-06,
+      "loss": 0.0563,
+      "step": 1941
+    },
+    {
+      "epoch": 0.8428819444444444,
+      "grad_norm": 0.682534396648407,
+      "learning_rate": 6.4935156712719615e-06,
+      "loss": 0.0369,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8433159722222222,
+      "grad_norm": 0.9464619159698486,
+      "learning_rate": 6.490160898146919e-06,
+      "loss": 0.0597,
+      "step": 1943
+    },
+    {
+      "epoch": 0.84375,
+      "grad_norm": 0.6500828862190247,
+      "learning_rate": 6.4868053886245885e-06,
+      "loss": 0.0506,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8441840277777778,
+      "grad_norm": 0.5619100332260132,
+      "learning_rate": 6.483449144363169e-06,
+      "loss": 0.0408,
+      "step": 1945
+    },
+    {
+      "epoch": 0.8446180555555556,
+      "grad_norm": 0.41894733905792236,
+      "learning_rate": 6.480092167021228e-06,
+      "loss": 0.0241,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8450520833333334,
+      "grad_norm": 0.7337756752967834,
+      "learning_rate": 6.476734458257692e-06,
+      "loss": 0.0448,
+      "step": 1947
+    },
+    {
+      "epoch": 0.8454861111111112,
+      "grad_norm": 0.7293131947517395,
+      "learning_rate": 6.473376019731851e-06,
+      "loss": 0.0435,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8459201388888888,
+      "grad_norm": 0.6051151752471924,
+      "learning_rate": 6.470016853103353e-06,
+      "loss": 0.0513,
+      "step": 1949
+    },
+    {
+      "epoch": 0.8463541666666666,
+      "grad_norm": 0.7551562190055847,
+      "learning_rate": 6.466656960032209e-06,
+      "loss": 0.0318,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8467881944444444,
+      "grad_norm": 0.818610668182373,
+      "learning_rate": 6.463296342178787e-06,
+      "loss": 0.0586,
+      "step": 1951
+    },
+    {
+      "epoch": 0.8472222222222222,
+      "grad_norm": 0.9881995320320129,
+      "learning_rate": 6.459935001203812e-06,
+      "loss": 0.0387,
+      "step": 1952
+    },
+    {
+      "epoch": 0.84765625,
+      "grad_norm": 0.7172938585281372,
+      "learning_rate": 6.456572938768371e-06,
+      "loss": 0.0432,
+      "step": 1953
+    },
+    {
+      "epoch": 0.8480902777777778,
+      "grad_norm": 0.7845926880836487,
+      "learning_rate": 6.453210156533902e-06,
+      "loss": 0.0489,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8485243055555556,
+      "grad_norm": 2.5467822551727295,
+      "learning_rate": 6.449846656162205e-06,
+      "loss": 0.0298,
+      "step": 1955
+    },
+    {
+      "epoch": 0.8489583333333334,
+      "grad_norm": 1.1665350198745728,
+      "learning_rate": 6.446482439315426e-06,
+      "loss": 0.0506,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8493923611111112,
+      "grad_norm": 1.032641053199768,
+      "learning_rate": 6.443117507656072e-06,
+      "loss": 0.044,
+      "step": 1957
+    },
+    {
+      "epoch": 0.8498263888888888,
+      "grad_norm": 0.5748589038848877,
+      "learning_rate": 6.439751862847002e-06,
+      "loss": 0.0415,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8502604166666666,
+      "grad_norm": 0.5190004706382751,
+      "learning_rate": 6.436385506551428e-06,
+      "loss": 0.0369,
+      "step": 1959
+    },
+    {
+      "epoch": 0.8506944444444444,
+      "grad_norm": 0.5892989039421082,
+      "learning_rate": 6.43301844043291e-06,
+      "loss": 0.0316,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8511284722222222,
+      "grad_norm": 0.8164686560630798,
+      "learning_rate": 6.429650666155363e-06,
+      "loss": 0.0389,
+      "step": 1961
+    },
+    {
+      "epoch": 0.8515625,
+      "grad_norm": 0.7017279267311096,
+      "learning_rate": 6.426282185383049e-06,
+      "loss": 0.0376,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8519965277777778,
+      "grad_norm": 0.5410183668136597,
+      "learning_rate": 6.422912999780579e-06,
+      "loss": 0.0429,
+      "step": 1963
+    },
+    {
+      "epoch": 0.8524305555555556,
+      "grad_norm": 1.0295459032058716,
+      "learning_rate": 6.4195431110129184e-06,
+      "loss": 0.0523,
+      "step": 1964
+    },
+    {
+      "epoch": 0.8528645833333334,
+      "grad_norm": 0.6248032450675964,
+      "learning_rate": 6.4161725207453694e-06,
+      "loss": 0.027,
+      "step": 1965
+    },
+    {
+      "epoch": 0.8532986111111112,
+      "grad_norm": 0.671219527721405,
+      "learning_rate": 6.41280123064359e-06,
+      "loss": 0.0543,
+      "step": 1966
+    },
+    {
+      "epoch": 0.8537326388888888,
+      "grad_norm": 1.6697989702224731,
+      "learning_rate": 6.409429242373581e-06,
+      "loss": 0.0504,
+      "step": 1967
+    },
+    {
+      "epoch": 0.8541666666666666,
+      "grad_norm": 1.516735315322876,
+      "learning_rate": 6.4060565576016875e-06,
+      "loss": 0.0404,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8546006944444444,
+      "grad_norm": 3.2779290676116943,
+      "learning_rate": 6.402683177994599e-06,
+      "loss": 0.0635,
+      "step": 1969
+    },
+    {
+      "epoch": 0.8550347222222222,
+      "grad_norm": 1.1508862972259521,
+      "learning_rate": 6.399309105219347e-06,
+      "loss": 0.0595,
+      "step": 1970
+    },
+    {
+      "epoch": 0.85546875,
+      "grad_norm": 0.636896014213562,
+      "learning_rate": 6.395934340943311e-06,
+      "loss": 0.0267,
+      "step": 1971
+    },
+    {
+      "epoch": 0.8559027777777778,
+      "grad_norm": 0.9543007016181946,
+      "learning_rate": 6.3925588868342035e-06,
+      "loss": 0.0464,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8563368055555556,
+      "grad_norm": 2.7885780334472656,
+      "learning_rate": 6.389182744560085e-06,
+      "loss": 0.0358,
+      "step": 1973
+    },
+    {
+      "epoch": 0.8567708333333334,
+      "grad_norm": 1.1232353448867798,
+      "learning_rate": 6.385805915789356e-06,
+      "loss": 0.0319,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8572048611111112,
+      "grad_norm": 17.793001174926758,
+      "learning_rate": 6.382428402190753e-06,
+      "loss": 0.0765,
+      "step": 1975
+    },
+    {
+      "epoch": 0.8576388888888888,
+      "grad_norm": 5.3929443359375,
+      "learning_rate": 6.3790502054333515e-06,
+      "loss": 0.0385,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8580729166666666,
+      "grad_norm": 2.506985902786255,
+      "learning_rate": 6.375671327186565e-06,
+      "loss": 0.0561,
+      "step": 1977
+    },
+    {
+      "epoch": 0.8585069444444444,
+      "grad_norm": 1.9946200847625732,
+      "learning_rate": 6.372291769120143e-06,
+      "loss": 0.0296,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8589409722222222,
+      "grad_norm": 3.731701374053955,
+      "learning_rate": 6.368911532904173e-06,
+      "loss": 0.0384,
+      "step": 1979
+    },
+    {
+      "epoch": 0.859375,
+      "grad_norm": 1.0285123586654663,
+      "learning_rate": 6.365530620209077e-06,
+      "loss": 0.047,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8598090277777778,
+      "grad_norm": 14.954253196716309,
+      "learning_rate": 6.36214903270561e-06,
+      "loss": 0.0549,
+      "step": 1981
+    },
+    {
+      "epoch": 0.8602430555555556,
+      "grad_norm": 4.2624125480651855,
+      "learning_rate": 6.3587667720648626e-06,
+      "loss": 0.0373,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8606770833333334,
+      "grad_norm": 5.046442031860352,
+      "learning_rate": 6.355383839958255e-06,
+      "loss": 0.0531,
+      "step": 1983
+    },
+    {
+      "epoch": 0.8611111111111112,
+      "grad_norm": 3.881274938583374,
+      "learning_rate": 6.3520002380575395e-06,
+      "loss": 0.0815,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8615451388888888,
+      "grad_norm": 5.472319602966309,
+      "learning_rate": 6.3486159680348046e-06,
+      "loss": 0.0767,
+      "step": 1985
+    },
+    {
+      "epoch": 0.8619791666666666,
+      "grad_norm": 3.93212890625,
+      "learning_rate": 6.345231031562464e-06,
+      "loss": 0.0518,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8624131944444444,
+      "grad_norm": 5.016967296600342,
+      "learning_rate": 6.341845430313259e-06,
+      "loss": 0.0534,
+      "step": 1987
+    },
+    {
+      "epoch": 0.8628472222222222,
+      "grad_norm": 2.7708795070648193,
+      "learning_rate": 6.338459165960266e-06,
+      "loss": 0.0419,
+      "step": 1988
+    },
+    {
+      "epoch": 0.86328125,
+      "grad_norm": 1.2714332342147827,
+      "learning_rate": 6.335072240176885e-06,
+      "loss": 0.0386,
+      "step": 1989
+    },
+    {
+      "epoch": 0.8637152777777778,
+      "grad_norm": 3.638699769973755,
+      "learning_rate": 6.331684654636843e-06,
+      "loss": 0.0483,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8641493055555556,
+      "grad_norm": 2.74619197845459,
+      "learning_rate": 6.328296411014191e-06,
+      "loss": 0.0311,
+      "step": 1991
+    },
+    {
+      "epoch": 0.8645833333333334,
+      "grad_norm": 2.6659770011901855,
+      "learning_rate": 6.32490751098331e-06,
+      "loss": 0.0388,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8650173611111112,
+      "grad_norm": 15.077685356140137,
+      "learning_rate": 6.321517956218901e-06,
+      "loss": 0.0531,
+      "step": 1993
+    },
+    {
+      "epoch": 0.8654513888888888,
+      "grad_norm": 3.0722298622131348,
+      "learning_rate": 6.318127748395992e-06,
+      "loss": 0.05,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8658854166666666,
+      "grad_norm": 3.180548667907715,
+      "learning_rate": 6.314736889189932e-06,
+      "loss": 0.0521,
+      "step": 1995
+    },
+    {
+      "epoch": 0.8663194444444444,
+      "grad_norm": 2.55326771736145,
+      "learning_rate": 6.311345380276393e-06,
+      "loss": 0.0493,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8667534722222222,
+      "grad_norm": 3.489673137664795,
+      "learning_rate": 6.307953223331366e-06,
+      "loss": 0.0536,
+      "step": 1997
+    },
+    {
+      "epoch": 0.8671875,
+      "grad_norm": 1.6632792949676514,
+      "learning_rate": 6.304560420031163e-06,
+      "loss": 0.0673,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8676215277777778,
+      "grad_norm": 3.4524927139282227,
+      "learning_rate": 6.301166972052416e-06,
+      "loss": 0.0699,
+      "step": 1999
+    },
+    {
+      "epoch": 0.8680555555555556,
+      "grad_norm": 2.278571367263794,
+      "learning_rate": 6.297772881072075e-06,
+      "loss": 0.0658,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8684895833333334,
+      "grad_norm": 2.340599298477173,
+      "learning_rate": 6.29437814876741e-06,
+      "loss": 0.0627,
+      "step": 2001
+    },
+    {
+      "epoch": 0.8689236111111112,
+      "grad_norm": 1.5224546194076538,
+      "learning_rate": 6.290982776816005e-06,
+      "loss": 0.0373,
+      "step": 2002
+    },
+    {
+      "epoch": 0.8693576388888888,
+      "grad_norm": 6.141655445098877,
+      "learning_rate": 6.2875867668957604e-06,
+      "loss": 0.0631,
+      "step": 2003
+    },
+    {
+      "epoch": 0.8697916666666666,
+      "grad_norm": 1.6914006471633911,
+      "learning_rate": 6.284190120684892e-06,
+      "loss": 0.0556,
+      "step": 2004
+    },
+    {
+      "epoch": 0.8702256944444444,
+      "grad_norm": 2.2304089069366455,
+      "learning_rate": 6.280792839861935e-06,
+      "loss": 0.0217,
+      "step": 2005
+    },
+    {
+      "epoch": 0.8706597222222222,
+      "grad_norm": 1.3991873264312744,
+      "learning_rate": 6.277394926105729e-06,
+      "loss": 0.046,
+      "step": 2006
+    },
+    {
+      "epoch": 0.87109375,
+      "grad_norm": 1.3106783628463745,
+      "learning_rate": 6.2739963810954325e-06,
+      "loss": 0.0483,
+      "step": 2007
+    },
+    {
+      "epoch": 0.8715277777777778,
+      "grad_norm": 0.8954133987426758,
+      "learning_rate": 6.270597206510516e-06,
+      "loss": 0.0584,
+      "step": 2008
+    },
+    {
+      "epoch": 0.8719618055555556,
+      "grad_norm": 0.5888851881027222,
+      "learning_rate": 6.2671974040307585e-06,
+      "loss": 0.0375,
+      "step": 2009
+    },
+    {
+      "epoch": 0.8723958333333334,
+      "grad_norm": 9.936369895935059,
+      "learning_rate": 6.2637969753362515e-06,
+      "loss": 0.0443,
+      "step": 2010
+    },
+    {
+      "epoch": 0.8728298611111112,
+      "grad_norm": 1.5567840337753296,
+      "learning_rate": 6.260395922107397e-06,
+      "loss": 0.0389,
+      "step": 2011
+    },
+    {
+      "epoch": 0.8732638888888888,
+      "grad_norm": 0.9838268756866455,
+      "learning_rate": 6.256994246024899e-06,
+      "loss": 0.0556,
+      "step": 2012
+    },
+    {
+      "epoch": 0.8736979166666666,
+      "grad_norm": 5.1909708976745605,
+      "learning_rate": 6.253591948769777e-06,
+      "loss": 0.0542,
+      "step": 2013
+    },
+    {
+      "epoch": 0.8741319444444444,
+      "grad_norm": 0.4206882119178772,
+      "learning_rate": 6.250189032023353e-06,
+      "loss": 0.0216,
+      "step": 2014
+    },
+    {
+      "epoch": 0.8745659722222222,
+      "grad_norm": 1.2391408681869507,
+      "learning_rate": 6.246785497467257e-06,
+      "loss": 0.0398,
+      "step": 2015
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 0.8819404244422913,
+      "learning_rate": 6.243381346783426e-06,
+      "loss": 0.0544,
+      "step": 2016
+    },
+    {
+      "epoch": 0.8754340277777778,
+      "grad_norm": 0.6158362627029419,
+      "learning_rate": 6.239976581654098e-06,
+      "loss": 0.0337,
+      "step": 2017
+    },
+    {
+      "epoch": 0.8758680555555556,
+      "grad_norm": 1.3593721389770508,
+      "learning_rate": 6.236571203761815e-06,
+      "loss": 0.0291,
+      "step": 2018
+    },
+    {
+      "epoch": 0.8763020833333334,
+      "grad_norm": 1.1812952756881714,
+      "learning_rate": 6.233165214789424e-06,
+      "loss": 0.0333,
+      "step": 2019
+    },
+    {
+      "epoch": 0.8767361111111112,
+      "grad_norm": 0.7267976403236389,
+      "learning_rate": 6.229758616420073e-06,
+      "loss": 0.0264,
+      "step": 2020
+    },
+    {
+      "epoch": 0.8771701388888888,
+      "grad_norm": 1.3565468788146973,
+      "learning_rate": 6.226351410337209e-06,
+      "loss": 0.0292,
+      "step": 2021
+    },
+    {
+      "epoch": 0.8776041666666666,
+      "grad_norm": 0.8340696096420288,
+      "learning_rate": 6.222943598224583e-06,
+      "loss": 0.0287,
+      "step": 2022
+    },
+    {
+      "epoch": 0.8780381944444444,
+      "grad_norm": 1.3752747774124146,
+      "learning_rate": 6.219535181766244e-06,
+      "loss": 0.0588,
+      "step": 2023
+    },
+    {
+      "epoch": 0.8784722222222222,
+      "grad_norm": 0.6889845132827759,
+      "learning_rate": 6.216126162646539e-06,
+      "loss": 0.0421,
+      "step": 2024
+    },
+    {
+      "epoch": 0.87890625,
+      "grad_norm": 1.2558727264404297,
+      "learning_rate": 6.212716542550112e-06,
+      "loss": 0.0365,
+      "step": 2025
+    },
+    {
+      "epoch": 0.8793402777777778,
+      "grad_norm": 1.1363980770111084,
+      "learning_rate": 6.209306323161908e-06,
+      "loss": 0.0447,
+      "step": 2026
+    },
+    {
+      "epoch": 0.8797743055555556,
+      "grad_norm": 0.8089574575424194,
+      "learning_rate": 6.2058955061671625e-06,
+      "loss": 0.0524,
+      "step": 2027
+    },
+    {
+      "epoch": 0.8802083333333334,
+      "grad_norm": 0.8970324397087097,
+      "learning_rate": 6.20248409325141e-06,
+      "loss": 0.0503,
+      "step": 2028
+    },
+    {
+      "epoch": 0.8806423611111112,
+      "grad_norm": 4.890934944152832,
+      "learning_rate": 6.199072086100479e-06,
+      "loss": 0.0463,
+      "step": 2029
+    },
+    {
+      "epoch": 0.8810763888888888,
+      "grad_norm": 1.5201784372329712,
+      "learning_rate": 6.195659486400491e-06,
+      "loss": 0.0384,
+      "step": 2030
+    },
+    {
+      "epoch": 0.8815104166666666,
+      "grad_norm": 0.8894219398498535,
+      "learning_rate": 6.192246295837863e-06,
+      "loss": 0.0539,
+      "step": 2031
+    },
+    {
+      "epoch": 0.8819444444444444,
+      "grad_norm": 1.5229333639144897,
+      "learning_rate": 6.188832516099298e-06,
+      "loss": 0.0447,
+      "step": 2032
+    },
+    {
+      "epoch": 0.8823784722222222,
+      "grad_norm": 0.9418681263923645,
+      "learning_rate": 6.185418148871795e-06,
+      "loss": 0.0322,
+      "step": 2033
+    },
+    {
+      "epoch": 0.8828125,
+      "grad_norm": 1.105890154838562,
+      "learning_rate": 6.182003195842643e-06,
+      "loss": 0.0517,
+      "step": 2034
+    },
+    {
+      "epoch": 0.8832465277777778,
+      "grad_norm": 0.7091906666755676,
+      "learning_rate": 6.178587658699418e-06,
+      "loss": 0.0399,
+      "step": 2035
+    },
+    {
+      "epoch": 0.8836805555555556,
+      "grad_norm": 1.3659592866897583,
+      "learning_rate": 6.175171539129988e-06,
+      "loss": 0.0459,
+      "step": 2036
+    },
+    {
+      "epoch": 0.8841145833333334,
+      "grad_norm": 0.693271815776825,
+      "learning_rate": 6.171754838822508e-06,
+      "loss": 0.0476,
+      "step": 2037
+    },
+    {
+      "epoch": 0.8845486111111112,
+      "grad_norm": 0.6045472025871277,
+      "learning_rate": 6.168337559465416e-06,
+      "loss": 0.0399,
+      "step": 2038
+    },
+    {
+      "epoch": 0.8849826388888888,
+      "grad_norm": 0.8418381810188293,
+      "learning_rate": 6.164919702747442e-06,
+      "loss": 0.049,
+      "step": 2039
+    },
+    {
+      "epoch": 0.8854166666666666,
+      "grad_norm": 0.8624420166015625,
+      "learning_rate": 6.161501270357597e-06,
+      "loss": 0.0531,
+      "step": 2040
+    },
+    {
+      "epoch": 0.8858506944444444,
+      "grad_norm": 0.5789508819580078,
+      "learning_rate": 6.158082263985178e-06,
+      "loss": 0.0421,
+      "step": 2041
+    },
+    {
+      "epoch": 0.8862847222222222,
+      "grad_norm": 0.6958301067352295,
+      "learning_rate": 6.154662685319764e-06,
+      "loss": 0.0552,
+      "step": 2042
+    },
+    {
+      "epoch": 0.88671875,
+      "grad_norm": 1.0128082036972046,
+      "learning_rate": 6.151242536051223e-06,
+      "loss": 0.0613,
+      "step": 2043
+    },
+    {
+      "epoch": 0.8871527777777778,
+      "grad_norm": 1.0239746570587158,
+      "learning_rate": 6.147821817869697e-06,
+      "loss": 0.0594,
+      "step": 2044
+    },
+    {
+      "epoch": 0.8875868055555556,
+      "grad_norm": 0.4922561049461365,
+      "learning_rate": 6.144400532465612e-06,
+      "loss": 0.0313,
+      "step": 2045
+    },
+    {
+      "epoch": 0.8880208333333334,
+      "grad_norm": 0.44895923137664795,
+      "learning_rate": 6.140978681529677e-06,
+      "loss": 0.0386,
+      "step": 2046
+    },
+    {
+      "epoch": 0.8884548611111112,
+      "grad_norm": 0.47623518109321594,
+      "learning_rate": 6.137556266752878e-06,
+      "loss": 0.0354,
+      "step": 2047
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7948447465896606,
+      "learning_rate": 6.134133289826481e-06,
+      "loss": 0.0587,
+      "step": 2048
+    },
+    {
+      "epoch": 0.8893229166666666,
+      "grad_norm": 0.6529760956764221,
+      "learning_rate": 6.130709752442024e-06,
+      "loss": 0.045,
+      "step": 2049
+    },
+    {
+      "epoch": 0.8897569444444444,
+      "grad_norm": 1.3702458143234253,
+      "learning_rate": 6.127285656291333e-06,
+      "loss": 0.0638,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8901909722222222,
+      "grad_norm": 0.9024243354797363,
+      "learning_rate": 6.1238610030665014e-06,
+      "loss": 0.0619,
+      "step": 2051
+    },
+    {
+      "epoch": 0.890625,
+      "grad_norm": 1.0158090591430664,
+      "learning_rate": 6.120435794459899e-06,
+      "loss": 0.0539,
+      "step": 2052
+    },
+    {
+      "epoch": 0.8910590277777778,
+      "grad_norm": 0.49644342064857483,
+      "learning_rate": 6.117010032164174e-06,
+      "loss": 0.0376,
+      "step": 2053
+    },
+    {
+      "epoch": 0.8914930555555556,
+      "grad_norm": 0.5801069736480713,
+      "learning_rate": 6.113583717872243e-06,
+      "loss": 0.0466,
+      "step": 2054
+    },
+    {
+      "epoch": 0.8919270833333334,
+      "grad_norm": 0.5090112686157227,
+      "learning_rate": 6.110156853277298e-06,
+      "loss": 0.0435,
+      "step": 2055
+    },
+    {
+      "epoch": 0.8923611111111112,
+      "grad_norm": 0.7452555298805237,
+      "learning_rate": 6.106729440072804e-06,
+      "loss": 0.0458,
+      "step": 2056
+    },
+    {
+      "epoch": 0.8927951388888888,
+      "grad_norm": 0.6473926305770874,
+      "learning_rate": 6.103301479952497e-06,
+      "loss": 0.0483,
+      "step": 2057
+    },
+    {
+      "epoch": 0.8932291666666666,
+      "grad_norm": 0.6398634314537048,
+      "learning_rate": 6.099872974610382e-06,
+      "loss": 0.0356,
+      "step": 2058
+    },
+    {
+      "epoch": 0.8936631944444444,
+      "grad_norm": 1.5123109817504883,
+      "learning_rate": 6.0964439257407335e-06,
+      "loss": 0.0769,
+      "step": 2059
+    },
+    {
+      "epoch": 0.8940972222222222,
+      "grad_norm": 1.6154077053070068,
+      "learning_rate": 6.093014335038094e-06,
+      "loss": 0.0598,
+      "step": 2060
+    },
+    {
+      "epoch": 0.89453125,
+      "grad_norm": 0.6833464503288269,
+      "learning_rate": 6.089584204197277e-06,
+      "loss": 0.0298,
+      "step": 2061
+    },
+    {
+      "epoch": 0.8949652777777778,
+      "grad_norm": 1.5443462133407593,
+      "learning_rate": 6.086153534913359e-06,
+      "loss": 0.0563,
+      "step": 2062
+    },
+    {
+      "epoch": 0.8953993055555556,
+      "grad_norm": 0.786438524723053,
+      "learning_rate": 6.082722328881682e-06,
+      "loss": 0.0319,
+      "step": 2063
+    },
+    {
+      "epoch": 0.8958333333333334,
+      "grad_norm": 0.7036506533622742,
+      "learning_rate": 6.079290587797858e-06,
+      "loss": 0.0528,
+      "step": 2064
+    },
+    {
+      "epoch": 0.8962673611111112,
+      "grad_norm": 0.4531269669532776,
+      "learning_rate": 6.075858313357762e-06,
+      "loss": 0.0254,
+      "step": 2065
+    },
+    {
+      "epoch": 0.8967013888888888,
+      "grad_norm": 0.7851917743682861,
+      "learning_rate": 6.072425507257528e-06,
+      "loss": 0.0423,
+      "step": 2066
+    },
+    {
+      "epoch": 0.8971354166666666,
+      "grad_norm": 0.6986308097839355,
+      "learning_rate": 6.068992171193558e-06,
+      "loss": 0.0198,
+      "step": 2067
+    },
+    {
+      "epoch": 0.8975694444444444,
+      "grad_norm": 3.0546317100524902,
+      "learning_rate": 6.065558306862514e-06,
+      "loss": 0.0408,
+      "step": 2068
+    },
+    {
+      "epoch": 0.8980034722222222,
+      "grad_norm": 0.8991721272468567,
+      "learning_rate": 6.06212391596132e-06,
+      "loss": 0.0408,
+      "step": 2069
+    },
+    {
+      "epoch": 0.8984375,
+      "grad_norm": 1.241905689239502,
+      "learning_rate": 6.058689000187157e-06,
+      "loss": 0.0447,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8988715277777778,
+      "grad_norm": 1.7747160196304321,
+      "learning_rate": 6.055253561237469e-06,
+      "loss": 0.0491,
+      "step": 2071
+    },
+    {
+      "epoch": 0.8993055555555556,
+      "grad_norm": 0.8431347608566284,
+      "learning_rate": 6.051817600809957e-06,
+      "loss": 0.0391,
+      "step": 2072
+    },
+    {
+      "epoch": 0.8997395833333334,
+      "grad_norm": 1.0173333883285522,
+      "learning_rate": 6.0483811206025815e-06,
+      "loss": 0.0517,
+      "step": 2073
+    },
+    {
+      "epoch": 0.9001736111111112,
+      "grad_norm": 1.2025859355926514,
+      "learning_rate": 6.044944122313554e-06,
+      "loss": 0.0228,
+      "step": 2074
+    },
+    {
+      "epoch": 0.9006076388888888,
+      "grad_norm": 0.8313187956809998,
+      "learning_rate": 6.041506607641351e-06,
+      "loss": 0.0427,
+      "step": 2075
+    },
+    {
+      "epoch": 0.9010416666666666,
+      "grad_norm": 0.8221765160560608,
+      "learning_rate": 6.038068578284696e-06,
+      "loss": 0.0401,
+      "step": 2076
+    },
+    {
+      "epoch": 0.9014756944444444,
+      "grad_norm": 1.3112930059432983,
+      "learning_rate": 6.034630035942573e-06,
+      "loss": 0.0426,
+      "step": 2077
+    },
+    {
+      "epoch": 0.9019097222222222,
+      "grad_norm": 1.079392671585083,
+      "learning_rate": 6.0311909823142124e-06,
+      "loss": 0.0376,
+      "step": 2078
+    },
+    {
+      "epoch": 0.90234375,
+      "grad_norm": 1.1555027961730957,
+      "learning_rate": 6.027751419099107e-06,
+      "loss": 0.0456,
+      "step": 2079
+    },
+    {
+      "epoch": 0.9027777777777778,
+      "grad_norm": 4.339806079864502,
+      "learning_rate": 6.024311347996994e-06,
+      "loss": 0.0472,
+      "step": 2080
+    },
+    {
+      "epoch": 0.9032118055555556,
+      "grad_norm": 1.6129090785980225,
+      "learning_rate": 6.020870770707864e-06,
+      "loss": 0.0475,
+      "step": 2081
+    },
+    {
+      "epoch": 0.9036458333333334,
+      "grad_norm": 1.724150538444519,
+      "learning_rate": 6.017429688931956e-06,
+      "loss": 0.0534,
+      "step": 2082
+    },
+    {
+      "epoch": 0.9040798611111112,
+      "grad_norm": 0.8326488733291626,
+      "learning_rate": 6.013988104369762e-06,
+      "loss": 0.0406,
+      "step": 2083
+    },
+    {
+      "epoch": 0.9045138888888888,
+      "grad_norm": 0.8307406306266785,
+      "learning_rate": 6.010546018722017e-06,
+      "loss": 0.0391,
+      "step": 2084
+    },
+    {
+      "epoch": 0.9049479166666666,
+      "grad_norm": 0.7936608195304871,
+      "learning_rate": 6.007103433689709e-06,
+      "loss": 0.0573,
+      "step": 2085
+    },
+    {
+      "epoch": 0.9053819444444444,
+      "grad_norm": 0.5702849626541138,
+      "learning_rate": 6.00366035097407e-06,
+      "loss": 0.0295,
+      "step": 2086
+    },
+    {
+      "epoch": 0.9058159722222222,
+      "grad_norm": 1.1666098833084106,
+      "learning_rate": 6.000216772276579e-06,
+      "loss": 0.0506,
+      "step": 2087
+    },
+    {
+      "epoch": 0.90625,
+      "grad_norm": 0.600796639919281,
+      "learning_rate": 5.99677269929896e-06,
+      "loss": 0.0366,
+      "step": 2088
+    },
+    {
+      "epoch": 0.9066840277777778,
+      "grad_norm": 0.7910858988761902,
+      "learning_rate": 5.993328133743182e-06,
+      "loss": 0.0378,
+      "step": 2089
+    },
+    {
+      "epoch": 0.9071180555555556,
+      "grad_norm": 1.3797601461410522,
+      "learning_rate": 5.989883077311453e-06,
+      "loss": 0.0715,
+      "step": 2090
+    },
+    {
+      "epoch": 0.9075520833333334,
+      "grad_norm": 1.3978376388549805,
+      "learning_rate": 5.986437531706226e-06,
+      "loss": 0.075,
+      "step": 2091
+    },
+    {
+      "epoch": 0.9079861111111112,
+      "grad_norm": 0.7575281262397766,
+      "learning_rate": 5.982991498630201e-06,
+      "loss": 0.0324,
+      "step": 2092
+    },
+    {
+      "epoch": 0.9084201388888888,
+      "grad_norm": 1.9478482007980347,
+      "learning_rate": 5.979544979786315e-06,
+      "loss": 0.0549,
+      "step": 2093
+    },
+    {
+      "epoch": 0.9088541666666666,
+      "grad_norm": 1.209311604499817,
+      "learning_rate": 5.97609797687774e-06,
+      "loss": 0.0308,
+      "step": 2094
+    },
+    {
+      "epoch": 0.9092881944444444,
+      "grad_norm": 1.0717582702636719,
+      "learning_rate": 5.972650491607896e-06,
+      "loss": 0.0683,
+      "step": 2095
+    },
+    {
+      "epoch": 0.9097222222222222,
+      "grad_norm": 0.743563711643219,
+      "learning_rate": 5.9692025256804335e-06,
+      "loss": 0.0484,
+      "step": 2096
+    },
+    {
+      "epoch": 0.91015625,
+      "grad_norm": 0.5490623712539673,
+      "learning_rate": 5.965754080799244e-06,
+      "loss": 0.0312,
+      "step": 2097
+    },
+    {
+      "epoch": 0.9105902777777778,
+      "grad_norm": 0.6579229235649109,
+      "learning_rate": 5.96230515866846e-06,
+      "loss": 0.0568,
+      "step": 2098
+    },
+    {
+      "epoch": 0.9110243055555556,
+      "grad_norm": 0.5728437900543213,
+      "learning_rate": 5.9588557609924425e-06,
+      "loss": 0.0421,
+      "step": 2099
+    },
+    {
+      "epoch": 0.9114583333333334,
+      "grad_norm": 1.7242991924285889,
+      "learning_rate": 5.955405889475793e-06,
+      "loss": 0.0537,
+      "step": 2100
+    },
+    {
+      "epoch": 0.9114583333333334,
+      "eval_layout_rm_rank_test_accuracy": 0.9735698916000163,
+      "eval_layout_rm_rank_test_chosen_scores": 0.5979847311973572,
+      "eval_layout_rm_rank_test_loss": 0.04800070822238922,
+      "eval_layout_rm_rank_test_rejected_scores": 0.08641856908798218,
+      "eval_layout_rm_rank_test_runtime": 1197.1207,
+      "eval_layout_rm_rank_test_samples_per_second": 20.575,
+      "eval_layout_rm_rank_test_steps_per_second": 0.643,
+      "step": 2100
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 4608,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 300,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}