diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,9597 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 1365,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002197802197802198,
+      "grad_norm": 0.07574141770601273,
+      "learning_rate": 2.18978102189781e-05,
+      "loss": 1.4133,
+      "step": 1
+    },
+    {
+      "epoch": 0.004395604395604396,
+      "grad_norm": 0.005070243496447802,
+      "learning_rate": 4.37956204379562e-05,
+      "loss": 1.2362,
+      "step": 2
+    },
+    {
+      "epoch": 0.006593406593406593,
+      "grad_norm": 0.007299270015209913,
+      "learning_rate": 6.569343065693432e-05,
+      "loss": 1.6253,
+      "step": 3
+    },
+    {
+      "epoch": 0.008791208791208791,
+      "grad_norm": 0.01405386533588171,
+      "learning_rate": 8.75912408759124e-05,
+      "loss": 1.4974,
+      "step": 4
+    },
+    {
+      "epoch": 0.01098901098901099,
+      "grad_norm": 0.0059348479844629765,
+      "learning_rate": 0.00010948905109489051,
+      "loss": 1.4851,
+      "step": 5
+    },
+    {
+      "epoch": 0.013186813186813187,
+      "grad_norm": 0.013611181639134884,
+      "learning_rate": 0.00013138686131386864,
+      "loss": 1.1586,
+      "step": 6
+    },
+    {
+      "epoch": 0.015384615384615385,
+      "grad_norm": 0.00719482870772481,
+      "learning_rate": 0.00015328467153284672,
+      "loss": 1.3258,
+      "step": 7
+    },
+    {
+      "epoch": 0.017582417582417582,
+      "grad_norm": 0.004739147145301104,
+      "learning_rate": 0.0001751824817518248,
+      "loss": 1.3883,
+      "step": 8
+    },
+    {
+      "epoch": 0.01978021978021978,
+      "grad_norm": 0.01157298032194376,
+      "learning_rate": 0.00019708029197080293,
+      "loss": 1.2708,
+      "step": 9
+    },
+    {
+      "epoch": 0.02197802197802198,
+      "grad_norm": 0.006943157874047756,
+      "learning_rate": 0.00021897810218978101,
+      "loss": 1.4275,
+      "step": 10
+    },
+    {
+      "epoch": 0.024175824175824177,
+      "grad_norm": 0.015182427130639553,
+      "learning_rate": 0.00024087591240875913,
+      "loss": 1.4422,
+      "step": 11
+    },
+    {
+      "epoch": 0.026373626373626374,
+      "grad_norm": 0.006962641607969999,
+      "learning_rate": 0.00026277372262773727,
+      "loss": 1.451,
+      "step": 12
+    },
+    {
+      "epoch": 0.02857142857142857,
+      "grad_norm": 0.005967486184090376,
+      "learning_rate": 0.00028467153284671533,
+      "loss": 1.3214,
+      "step": 13
+    },
+    {
+      "epoch": 0.03076923076923077,
+      "grad_norm": 0.009044559672474861,
+      "learning_rate": 0.00030656934306569345,
+      "loss": 1.4095,
+      "step": 14
+    },
+    {
+      "epoch": 0.03296703296703297,
+      "grad_norm": 0.006914634723216295,
+      "learning_rate": 0.00032846715328467156,
+      "loss": 1.6557,
+      "step": 15
+    },
+    {
+      "epoch": 0.035164835164835165,
+      "grad_norm": 0.009315505623817444,
+      "learning_rate": 0.0003503649635036496,
+      "loss": 1.2597,
+      "step": 16
+    },
+    {
+      "epoch": 0.03736263736263736,
+      "grad_norm": 0.008870800025761127,
+      "learning_rate": 0.00037226277372262774,
+      "loss": 1.4747,
+      "step": 17
+    },
+    {
+      "epoch": 0.03956043956043956,
+      "grad_norm": 0.015926798805594444,
+      "learning_rate": 0.00039416058394160585,
+      "loss": 1.4035,
+      "step": 18
+    },
+    {
+      "epoch": 0.041758241758241756,
+      "grad_norm": 0.14785222709178925,
+      "learning_rate": 0.00041605839416058397,
+      "loss": 1.547,
+      "step": 19
+    },
+    {
+      "epoch": 0.04395604395604396,
+      "grad_norm": 0.009723897092044353,
+      "learning_rate": 0.00043795620437956203,
+      "loss": 1.7155,
+      "step": 20
+    },
+    {
+      "epoch": 0.046153846153846156,
+      "grad_norm": 0.0053133368492126465,
+      "learning_rate": 0.0004598540145985402,
+      "loss": 1.2713,
+      "step": 21
+    },
+    {
+      "epoch": 0.04835164835164835,
+      "grad_norm": 0.010869111865758896,
+      "learning_rate": 0.00048175182481751826,
+      "loss": 1.3779,
+      "step": 22
+    },
+    {
+      "epoch": 0.05054945054945055,
+      "grad_norm": 0.008631070144474506,
+      "learning_rate": 0.0005036496350364964,
+      "loss": 1.4778,
+      "step": 23
+    },
+    {
+      "epoch": 0.05274725274725275,
+      "grad_norm": 1.0794625282287598,
+      "learning_rate": 0.0005255474452554745,
+      "loss": 1.354,
+      "step": 24
+    },
+    {
+      "epoch": 0.054945054945054944,
+      "grad_norm": 0.010598675347864628,
+      "learning_rate": 0.0005474452554744526,
+      "loss": 1.3122,
+      "step": 25
+    },
+    {
+      "epoch": 0.05714285714285714,
+      "grad_norm": 0.011011985130608082,
+      "learning_rate": 0.0005693430656934307,
+      "loss": 1.2156,
+      "step": 26
+    },
+    {
+      "epoch": 0.05934065934065934,
+      "grad_norm": 0.006993886549025774,
+      "learning_rate": 0.0005912408759124088,
+      "loss": 1.456,
+      "step": 27
+    },
+    {
+      "epoch": 0.06153846153846154,
+      "grad_norm": 0.25161153078079224,
+      "learning_rate": 0.0006131386861313869,
+      "loss": 1.4362,
+      "step": 28
+    },
+    {
+      "epoch": 0.06373626373626373,
+      "grad_norm": 0.006526893470436335,
+      "learning_rate": 0.000635036496350365,
+      "loss": 1.5982,
+      "step": 29
+    },
+    {
+      "epoch": 0.06593406593406594,
+      "grad_norm": 0.009625286795198917,
+      "learning_rate": 0.0006569343065693431,
+      "loss": 1.4703,
+      "step": 30
+    },
+    {
+      "epoch": 0.06813186813186813,
+      "grad_norm": 0.007690387777984142,
+      "learning_rate": 0.0006788321167883212,
+      "loss": 1.2864,
+      "step": 31
+    },
+    {
+      "epoch": 0.07032967032967033,
+      "grad_norm": 0.018430771306157112,
+      "learning_rate": 0.0007007299270072992,
+      "loss": 1.5381,
+      "step": 32
+    },
+    {
+      "epoch": 0.07252747252747253,
+      "grad_norm": 0.01161026768386364,
+      "learning_rate": 0.0007226277372262774,
+      "loss": 1.6495,
+      "step": 33
+    },
+    {
+      "epoch": 0.07472527472527472,
+      "grad_norm": 0.007151211611926556,
+      "learning_rate": 0.0007445255474452555,
+      "loss": 1.4165,
+      "step": 34
+    },
+    {
+      "epoch": 0.07692307692307693,
+      "grad_norm": 0.007144616916775703,
+      "learning_rate": 0.0007664233576642336,
+      "loss": 1.4281,
+      "step": 35
+    },
+    {
+      "epoch": 0.07912087912087912,
+      "grad_norm": 0.01014639250934124,
+      "learning_rate": 0.0007883211678832117,
+      "loss": 1.5729,
+      "step": 36
+    },
+    {
+      "epoch": 0.08131868131868132,
+      "grad_norm": 0.04061491787433624,
+      "learning_rate": 0.0008102189781021898,
+      "loss": 1.3339,
+      "step": 37
+    },
+    {
+      "epoch": 0.08351648351648351,
+      "grad_norm": 0.005060202442109585,
+      "learning_rate": 0.0008321167883211679,
+      "loss": 1.3054,
+      "step": 38
+    },
+    {
+      "epoch": 0.08571428571428572,
+      "grad_norm": 0.007262351457029581,
+      "learning_rate": 0.000854014598540146,
+      "loss": 1.3202,
+      "step": 39
+    },
+    {
+      "epoch": 0.08791208791208792,
+      "grad_norm": 0.00564504973590374,
+      "learning_rate": 0.0008759124087591241,
+      "loss": 1.5231,
+      "step": 40
+    },
+    {
+      "epoch": 0.09010989010989011,
+      "grad_norm": 0.005943926516920328,
+      "learning_rate": 0.0008978102189781023,
+      "loss": 1.2927,
+      "step": 41
+    },
+    {
+      "epoch": 0.09230769230769231,
+      "grad_norm": 0.010411452502012253,
+      "learning_rate": 0.0009197080291970804,
+      "loss": 1.4502,
+      "step": 42
+    },
+    {
+      "epoch": 0.0945054945054945,
+      "grad_norm": 0.03844933584332466,
+      "learning_rate": 0.0009416058394160585,
+      "loss": 1.6313,
+      "step": 43
+    },
+    {
+      "epoch": 0.0967032967032967,
+      "grad_norm": 0.0605161227285862,
+      "learning_rate": 0.0009635036496350365,
+      "loss": 1.394,
+      "step": 44
+    },
+    {
+      "epoch": 0.0989010989010989,
+      "grad_norm": 0.030179811641573906,
+      "learning_rate": 0.0009854014598540146,
+      "loss": 1.4689,
+      "step": 45
+    },
+    {
+      "epoch": 0.1010989010989011,
+      "grad_norm": 0.011401821859180927,
+      "learning_rate": 0.0010072992700729927,
+      "loss": 1.3952,
+      "step": 46
+    },
+    {
+      "epoch": 0.10329670329670329,
+      "grad_norm": 0.0072502028197050095,
+      "learning_rate": 0.001029197080291971,
+      "loss": 1.3456,
+      "step": 47
+    },
+    {
+      "epoch": 0.1054945054945055,
+      "grad_norm": 0.0062689585611224174,
+      "learning_rate": 0.001051094890510949,
+      "loss": 1.7539,
+      "step": 48
+    },
+    {
+      "epoch": 0.1076923076923077,
+      "grad_norm": 0.013934290036559105,
+      "learning_rate": 0.001072992700729927,
+      "loss": 1.4441,
+      "step": 49
+    },
+    {
+      "epoch": 0.10989010989010989,
+      "grad_norm": 0.04717353358864784,
+      "learning_rate": 0.0010948905109489052,
+      "loss": 1.2645,
+      "step": 50
+    },
+    {
+      "epoch": 0.11208791208791209,
+      "grad_norm": 0.00611092010512948,
+      "learning_rate": 0.0011167883211678832,
+      "loss": 1.4791,
+      "step": 51
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.014487843960523605,
+      "learning_rate": 0.0011386861313868613,
+      "loss": 1.7079,
+      "step": 52
+    },
+    {
+      "epoch": 0.11648351648351649,
+      "grad_norm": 0.0061956737190485,
+      "learning_rate": 0.0011605839416058395,
+      "loss": 1.7464,
+      "step": 53
+    },
+    {
+      "epoch": 0.11868131868131868,
+      "grad_norm": 0.025850724428892136,
+      "learning_rate": 0.0011824817518248177,
+      "loss": 1.5349,
+      "step": 54
+    },
+    {
+      "epoch": 0.12087912087912088,
+      "grad_norm": 0.007444498594850302,
+      "learning_rate": 0.0012043795620437956,
+      "loss": 1.4581,
+      "step": 55
+    },
+    {
+      "epoch": 0.12307692307692308,
+      "grad_norm": 0.007917330600321293,
+      "learning_rate": 0.0012262773722627738,
+      "loss": 1.1958,
+      "step": 56
+    },
+    {
+      "epoch": 0.12527472527472527,
+      "grad_norm": 0.0064592063426971436,
+      "learning_rate": 0.0012481751824817517,
+      "loss": 1.4113,
+      "step": 57
+    },
+    {
+      "epoch": 0.12747252747252746,
+      "grad_norm": 0.014220084063708782,
+      "learning_rate": 0.00127007299270073,
+      "loss": 1.4925,
+      "step": 58
+    },
+    {
+      "epoch": 0.12967032967032968,
+      "grad_norm": 0.028200611472129822,
+      "learning_rate": 0.001291970802919708,
+      "loss": 1.547,
+      "step": 59
+    },
+    {
+      "epoch": 0.13186813186813187,
+      "grad_norm": 0.010817811824381351,
+      "learning_rate": 0.0013138686131386862,
+      "loss": 1.536,
+      "step": 60
+    },
+    {
+      "epoch": 0.13406593406593406,
+      "grad_norm": 0.0063562411814928055,
+      "learning_rate": 0.0013357664233576642,
+      "loss": 1.4042,
+      "step": 61
+    },
+    {
+      "epoch": 0.13626373626373625,
+      "grad_norm": 0.03416217118501663,
+      "learning_rate": 0.0013576642335766424,
+      "loss": 1.4861,
+      "step": 62
+    },
+    {
+      "epoch": 0.13846153846153847,
+      "grad_norm": 0.010366651229560375,
+      "learning_rate": 0.0013795620437956205,
+      "loss": 1.2353,
+      "step": 63
+    },
+    {
+      "epoch": 0.14065934065934066,
+      "grad_norm": 0.010402251034975052,
+      "learning_rate": 0.0014014598540145985,
+      "loss": 1.469,
+      "step": 64
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 0.09900416433811188,
+      "learning_rate": 0.0014233576642335767,
+      "loss": 1.606,
+      "step": 65
+    },
+    {
+      "epoch": 0.14505494505494507,
+      "grad_norm": 0.02288840152323246,
+      "learning_rate": 0.0014452554744525548,
+      "loss": 1.4445,
+      "step": 66
+    },
+    {
+      "epoch": 0.14725274725274726,
+      "grad_norm": 0.010228401981294155,
+      "learning_rate": 0.001467153284671533,
+      "loss": 1.6161,
+      "step": 67
+    },
+    {
+      "epoch": 0.14945054945054945,
+      "grad_norm": 0.012993928045034409,
+      "learning_rate": 0.001489051094890511,
+      "loss": 1.4085,
+      "step": 68
+    },
+    {
+      "epoch": 0.15164835164835164,
+      "grad_norm": 0.016949519515037537,
+      "learning_rate": 0.0015109489051094893,
+      "loss": 1.3981,
+      "step": 69
+    },
+    {
+      "epoch": 0.15384615384615385,
+      "grad_norm": 0.02030923403799534,
+      "learning_rate": 0.0015328467153284673,
+      "loss": 1.5502,
+      "step": 70
+    },
+    {
+      "epoch": 0.15604395604395604,
+      "grad_norm": 0.009495298378169537,
+      "learning_rate": 0.0015547445255474455,
+      "loss": 1.5825,
+      "step": 71
+    },
+    {
+      "epoch": 0.15824175824175823,
+      "grad_norm": 0.020167484879493713,
+      "learning_rate": 0.0015766423357664234,
+      "loss": 1.4041,
+      "step": 72
+    },
+    {
+      "epoch": 0.16043956043956045,
+      "grad_norm": 0.013751978985965252,
+      "learning_rate": 0.0015985401459854016,
+      "loss": 1.7026,
+      "step": 73
+    },
+    {
+      "epoch": 0.16263736263736264,
+      "grad_norm": 0.00719607574865222,
+      "learning_rate": 0.0016204379562043795,
+      "loss": 1.4572,
+      "step": 74
+    },
+    {
+      "epoch": 0.16483516483516483,
+      "grad_norm": 0.134651780128479,
+      "learning_rate": 0.0016423357664233577,
+      "loss": 1.6772,
+      "step": 75
+    },
+    {
+      "epoch": 0.16703296703296702,
+      "grad_norm": 0.009734016843140125,
+      "learning_rate": 0.0016642335766423359,
+      "loss": 1.3254,
+      "step": 76
+    },
+    {
+      "epoch": 0.16923076923076924,
+      "grad_norm": 0.006087453570216894,
+      "learning_rate": 0.0016861313868613138,
+      "loss": 1.427,
+      "step": 77
+    },
+    {
+      "epoch": 0.17142857142857143,
+      "grad_norm": 0.013573980890214443,
+      "learning_rate": 0.001708029197080292,
+      "loss": 1.3918,
+      "step": 78
+    },
+    {
+      "epoch": 0.17362637362637362,
+      "grad_norm": 0.013283486478030682,
+      "learning_rate": 0.00172992700729927,
+      "loss": 1.322,
+      "step": 79
+    },
+    {
+      "epoch": 0.17582417582417584,
+      "grad_norm": 0.008982392027974129,
+      "learning_rate": 0.0017518248175182481,
+      "loss": 1.515,
+      "step": 80
+    },
+    {
+      "epoch": 0.17802197802197803,
+      "grad_norm": 0.013176189735531807,
+      "learning_rate": 0.0017737226277372265,
+      "loss": 1.5472,
+      "step": 81
+    },
+    {
+      "epoch": 0.18021978021978022,
+      "grad_norm": 0.004916726145893335,
+      "learning_rate": 0.0017956204379562047,
+      "loss": 1.3784,
+      "step": 82
+    },
+    {
+      "epoch": 0.1824175824175824,
+      "grad_norm": 0.012261349707841873,
+      "learning_rate": 0.0018175182481751826,
+      "loss": 1.5343,
+      "step": 83
+    },
+    {
+      "epoch": 0.18461538461538463,
+      "grad_norm": 0.022266652435064316,
+      "learning_rate": 0.0018394160583941608,
+      "loss": 1.4171,
+      "step": 84
+    },
+    {
+      "epoch": 0.18681318681318682,
+      "grad_norm": 0.005485480651259422,
+      "learning_rate": 0.0018613138686131387,
+      "loss": 1.4494,
+      "step": 85
+    },
+    {
+      "epoch": 0.189010989010989,
+      "grad_norm": 0.0106284124776721,
+      "learning_rate": 0.001883211678832117,
+      "loss": 1.7724,
+      "step": 86
+    },
+    {
+      "epoch": 0.1912087912087912,
+      "grad_norm": 0.008599367924034595,
+      "learning_rate": 0.0019051094890510949,
+      "loss": 1.1491,
+      "step": 87
+    },
+    {
+      "epoch": 0.1934065934065934,
+      "grad_norm": 0.008495704270899296,
+      "learning_rate": 0.001927007299270073,
+      "loss": 1.4948,
+      "step": 88
+    },
+    {
+      "epoch": 0.1956043956043956,
+      "grad_norm": 0.06533650308847427,
+      "learning_rate": 0.001948905109489051,
+      "loss": 1.4502,
+      "step": 89
+    },
+    {
+      "epoch": 0.1978021978021978,
+      "grad_norm": 0.00667093088850379,
+      "learning_rate": 0.001970802919708029,
+      "loss": 1.4318,
+      "step": 90
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.00814666785299778,
+      "learning_rate": 0.0019927007299270073,
+      "loss": 1.2298,
+      "step": 91
+    },
+    {
+      "epoch": 0.2021978021978022,
+      "grad_norm": 0.004924417473375797,
+      "learning_rate": 0.0020145985401459855,
+      "loss": 1.0947,
+      "step": 92
+    },
+    {
+      "epoch": 0.2043956043956044,
+      "grad_norm": 0.008014189079403877,
+      "learning_rate": 0.0020364963503649632,
+      "loss": 1.4832,
+      "step": 93
+    },
+    {
+      "epoch": 0.20659340659340658,
+      "grad_norm": 0.013895885087549686,
+      "learning_rate": 0.002058394160583942,
+      "loss": 1.6468,
+      "step": 94
+    },
+    {
+      "epoch": 0.2087912087912088,
+      "grad_norm": 0.008322709240019321,
+      "learning_rate": 0.00208029197080292,
+      "loss": 1.5204,
+      "step": 95
+    },
+    {
+      "epoch": 0.210989010989011,
+      "grad_norm": 0.010290144942700863,
+      "learning_rate": 0.002102189781021898,
+      "loss": 1.5306,
+      "step": 96
+    },
+    {
+      "epoch": 0.21318681318681318,
+      "grad_norm": 0.009071101434528828,
+      "learning_rate": 0.002124087591240876,
+      "loss": 1.4285,
+      "step": 97
+    },
+    {
+      "epoch": 0.2153846153846154,
+      "grad_norm": 0.003956215921789408,
+      "learning_rate": 0.002145985401459854,
+      "loss": 1.5215,
+      "step": 98
+    },
+    {
+      "epoch": 0.2175824175824176,
+      "grad_norm": 0.0071427118964493275,
+      "learning_rate": 0.0021678832116788322,
+      "loss": 1.7401,
+      "step": 99
+    },
+    {
+      "epoch": 0.21978021978021978,
+      "grad_norm": 0.009502488188445568,
+      "learning_rate": 0.0021897810218978104,
+      "loss": 1.5805,
+      "step": 100
+    },
+    {
+      "epoch": 0.22197802197802197,
+      "grad_norm": 0.009622444398701191,
+      "learning_rate": 0.002211678832116788,
+      "loss": 1.3939,
+      "step": 101
+    },
+    {
+      "epoch": 0.22417582417582418,
+      "grad_norm": 0.014163932763040066,
+      "learning_rate": 0.0022335766423357663,
+      "loss": 1.5483,
+      "step": 102
+    },
+    {
+      "epoch": 0.22637362637362637,
+      "grad_norm": 0.009929346852004528,
+      "learning_rate": 0.0022554744525547445,
+      "loss": 1.3761,
+      "step": 103
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.007974986918270588,
+      "learning_rate": 0.0022773722627737226,
+      "loss": 1.5046,
+      "step": 104
+    },
+    {
+      "epoch": 0.23076923076923078,
+      "grad_norm": 0.005453914403915405,
+      "learning_rate": 0.002299270072992701,
+      "loss": 1.527,
+      "step": 105
+    },
+    {
+      "epoch": 0.23296703296703297,
+      "grad_norm": 0.015407297760248184,
+      "learning_rate": 0.002321167883211679,
+      "loss": 1.3481,
+      "step": 106
+    },
+    {
+      "epoch": 0.23516483516483516,
+      "grad_norm": 0.02923862636089325,
+      "learning_rate": 0.002343065693430657,
+      "loss": 1.4483,
+      "step": 107
+    },
+    {
+      "epoch": 0.23736263736263735,
+      "grad_norm": 0.009559174999594688,
+      "learning_rate": 0.0023649635036496353,
+      "loss": 1.4847,
+      "step": 108
+    },
+    {
+      "epoch": 0.23956043956043957,
+      "grad_norm": 0.006819604896008968,
+      "learning_rate": 0.002386861313868613,
+      "loss": 1.3656,
+      "step": 109
+    },
+    {
+      "epoch": 0.24175824175824176,
+      "grad_norm": 0.02884874865412712,
+      "learning_rate": 0.0024087591240875912,
+      "loss": 1.3614,
+      "step": 110
+    },
+    {
+      "epoch": 0.24395604395604395,
+      "grad_norm": 0.006585326977074146,
+      "learning_rate": 0.0024306569343065694,
+      "loss": 1.4407,
+      "step": 111
+    },
+    {
+      "epoch": 0.24615384615384617,
+      "grad_norm": 0.006407666485756636,
+      "learning_rate": 0.0024525547445255476,
+      "loss": 1.6717,
+      "step": 112
+    },
+    {
+      "epoch": 0.24835164835164836,
+      "grad_norm": 0.008582557551562786,
+      "learning_rate": 0.0024744525547445257,
+      "loss": 1.2543,
+      "step": 113
+    },
+    {
+      "epoch": 0.25054945054945055,
+      "grad_norm": 0.007554011885076761,
+      "learning_rate": 0.0024963503649635035,
+      "loss": 1.6589,
+      "step": 114
+    },
+    {
+      "epoch": 0.25274725274725274,
+      "grad_norm": 0.006149493157863617,
+      "learning_rate": 0.0025182481751824816,
+      "loss": 1.4681,
+      "step": 115
+    },
+    {
+      "epoch": 0.2549450549450549,
+      "grad_norm": 0.015087589621543884,
+      "learning_rate": 0.00254014598540146,
+      "loss": 1.3143,
+      "step": 116
+    },
+    {
+      "epoch": 0.2571428571428571,
+      "grad_norm": 0.005657331086695194,
+      "learning_rate": 0.002562043795620438,
+      "loss": 1.3625,
+      "step": 117
+    },
+    {
+      "epoch": 0.25934065934065936,
+      "grad_norm": 0.042173877358436584,
+      "learning_rate": 0.002583941605839416,
+      "loss": 1.364,
+      "step": 118
+    },
+    {
+      "epoch": 0.26153846153846155,
+      "grad_norm": 0.011670319363474846,
+      "learning_rate": 0.0026058394160583943,
+      "loss": 1.2794,
+      "step": 119
+    },
+    {
+      "epoch": 0.26373626373626374,
+      "grad_norm": 0.007955322042107582,
+      "learning_rate": 0.0026277372262773725,
+      "loss": 1.3709,
+      "step": 120
+    },
+    {
+      "epoch": 0.26593406593406593,
+      "grad_norm": 0.00576738640666008,
+      "learning_rate": 0.0026496350364963507,
+      "loss": 1.3827,
+      "step": 121
+    },
+    {
+      "epoch": 0.2681318681318681,
+      "grad_norm": 0.023384226486086845,
+      "learning_rate": 0.0026715328467153284,
+      "loss": 1.4434,
+      "step": 122
+    },
+    {
+      "epoch": 0.2703296703296703,
+      "grad_norm": 0.010464617051184177,
+      "learning_rate": 0.0026934306569343066,
+      "loss": 1.2503,
+      "step": 123
+    },
+    {
+      "epoch": 0.2725274725274725,
+      "grad_norm": 0.007307030726224184,
+      "learning_rate": 0.0027153284671532847,
+      "loss": 1.3831,
+      "step": 124
+    },
+    {
+      "epoch": 0.27472527472527475,
+      "grad_norm": 0.021920938044786453,
+      "learning_rate": 0.002737226277372263,
+      "loss": 1.4207,
+      "step": 125
+    },
+    {
+      "epoch": 0.27692307692307694,
+      "grad_norm": 0.006052302662283182,
+      "learning_rate": 0.002759124087591241,
+      "loss": 1.4374,
+      "step": 126
+    },
+    {
+      "epoch": 0.27912087912087913,
+      "grad_norm": 0.008776125498116016,
+      "learning_rate": 0.002781021897810219,
+      "loss": 1.4179,
+      "step": 127
+    },
+    {
+      "epoch": 0.2813186813186813,
+      "grad_norm": 0.021074872463941574,
+      "learning_rate": 0.002802919708029197,
+      "loss": 1.5376,
+      "step": 128
+    },
+    {
+      "epoch": 0.2835164835164835,
+      "grad_norm": 0.005307208746671677,
+      "learning_rate": 0.002824817518248175,
+      "loss": 1.2318,
+      "step": 129
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 0.031820766627788544,
+      "learning_rate": 0.0028467153284671533,
+      "loss": 1.2594,
+      "step": 130
+    },
+    {
+      "epoch": 0.2879120879120879,
+      "grad_norm": 0.0091733792796731,
+      "learning_rate": 0.0028686131386861315,
+      "loss": 1.4086,
+      "step": 131
+    },
+    {
+      "epoch": 0.29010989010989013,
+      "grad_norm": 0.009230692870914936,
+      "learning_rate": 0.0028905109489051097,
+      "loss": 1.2843,
+      "step": 132
+    },
+    {
+      "epoch": 0.2923076923076923,
+      "grad_norm": 0.01663454808294773,
+      "learning_rate": 0.002912408759124088,
+      "loss": 1.3928,
+      "step": 133
+    },
+    {
+      "epoch": 0.2945054945054945,
+      "grad_norm": 0.00575832137838006,
+      "learning_rate": 0.002934306569343066,
+      "loss": 1.2317,
+      "step": 134
+    },
+    {
+      "epoch": 0.2967032967032967,
+      "grad_norm": 0.011704837903380394,
+      "learning_rate": 0.0029562043795620437,
+      "loss": 1.5327,
+      "step": 135
+    },
+    {
+      "epoch": 0.2989010989010989,
+      "grad_norm": 0.011859576217830181,
+      "learning_rate": 0.002978102189781022,
+      "loss": 1.613,
+      "step": 136
+    },
+    {
+      "epoch": 0.3010989010989011,
+      "grad_norm": 0.006614954210817814,
+      "learning_rate": 0.003,
+      "loss": 1.6398,
+      "step": 137
+    },
+    {
+      "epoch": 0.3032967032967033,
+      "grad_norm": 0.008094151504337788,
+      "learning_rate": 0.002997557003257329,
+      "loss": 1.3161,
+      "step": 138
+    },
+    {
+      "epoch": 0.3054945054945055,
+      "grad_norm": 0.005938228685408831,
+      "learning_rate": 0.002995114006514658,
+      "loss": 1.1945,
+      "step": 139
+    },
+    {
+      "epoch": 0.3076923076923077,
+      "grad_norm": 0.010452362708747387,
+      "learning_rate": 0.002992671009771987,
+      "loss": 1.5978,
+      "step": 140
+    },
+    {
+      "epoch": 0.3098901098901099,
+      "grad_norm": 0.004628652241080999,
+      "learning_rate": 0.002990228013029316,
+      "loss": 1.2548,
+      "step": 141
+    },
+    {
+      "epoch": 0.3120879120879121,
+      "grad_norm": 0.006251639220863581,
+      "learning_rate": 0.002987785016286645,
+      "loss": 1.4409,
+      "step": 142
+    },
+    {
+      "epoch": 0.3142857142857143,
+      "grad_norm": 0.008969109505414963,
+      "learning_rate": 0.002985342019543974,
+      "loss": 1.4967,
+      "step": 143
+    },
+    {
+      "epoch": 0.31648351648351647,
+      "grad_norm": 0.009970350190997124,
+      "learning_rate": 0.002982899022801303,
+      "loss": 1.173,
+      "step": 144
+    },
+    {
+      "epoch": 0.31868131868131866,
+      "grad_norm": 0.010100000537931919,
+      "learning_rate": 0.002980456026058632,
+      "loss": 1.3375,
+      "step": 145
+    },
+    {
+      "epoch": 0.3208791208791209,
+      "grad_norm": 0.005915519315749407,
+      "learning_rate": 0.002978013029315961,
+      "loss": 1.4281,
+      "step": 146
+    },
+    {
+      "epoch": 0.3230769230769231,
+      "grad_norm": 0.01975257322192192,
+      "learning_rate": 0.00297557003257329,
+      "loss": 1.6068,
+      "step": 147
+    },
+    {
+      "epoch": 0.3252747252747253,
+      "grad_norm": 0.008059101179242134,
+      "learning_rate": 0.0029731270358306187,
+      "loss": 1.6355,
+      "step": 148
+    },
+    {
+      "epoch": 0.3274725274725275,
+      "grad_norm": 0.060351669788360596,
+      "learning_rate": 0.002970684039087948,
+      "loss": 1.4176,
+      "step": 149
+    },
+    {
+      "epoch": 0.32967032967032966,
+      "grad_norm": 0.007178079336881638,
+      "learning_rate": 0.002968241042345277,
+      "loss": 1.4986,
+      "step": 150
+    },
+    {
+      "epoch": 0.33186813186813185,
+      "grad_norm": 0.0231226347386837,
+      "learning_rate": 0.002965798045602606,
+      "loss": 1.22,
+      "step": 151
+    },
+    {
+      "epoch": 0.33406593406593404,
+      "grad_norm": 0.006698589771986008,
+      "learning_rate": 0.0029633550488599347,
+      "loss": 1.4746,
+      "step": 152
+    },
+    {
+      "epoch": 0.3362637362637363,
+      "grad_norm": 0.03889273479580879,
+      "learning_rate": 0.002960912052117264,
+      "loss": 1.3884,
+      "step": 153
+    },
+    {
+      "epoch": 0.3384615384615385,
+      "grad_norm": 0.005003706552088261,
+      "learning_rate": 0.002958469055374593,
+      "loss": 1.3336,
+      "step": 154
+    },
+    {
+      "epoch": 0.34065934065934067,
+      "grad_norm": 0.008152004331350327,
+      "learning_rate": 0.002956026058631922,
+      "loss": 1.6016,
+      "step": 155
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.009225542657077312,
+      "learning_rate": 0.0029535830618892507,
+      "loss": 1.4132,
+      "step": 156
+    },
+    {
+      "epoch": 0.34505494505494505,
+      "grad_norm": 0.011879832483828068,
+      "learning_rate": 0.0029511400651465796,
+      "loss": 1.3636,
+      "step": 157
+    },
+    {
+      "epoch": 0.34725274725274724,
+      "grad_norm": 0.0076890247873961926,
+      "learning_rate": 0.002948697068403909,
+      "loss": 1.8009,
+      "step": 158
+    },
+    {
+      "epoch": 0.34945054945054943,
+      "grad_norm": 0.007640939671546221,
+      "learning_rate": 0.002946254071661238,
+      "loss": 1.2438,
+      "step": 159
+    },
+    {
+      "epoch": 0.3516483516483517,
+      "grad_norm": 0.012748723849654198,
+      "learning_rate": 0.0029438110749185667,
+      "loss": 1.6403,
+      "step": 160
+    },
+    {
+      "epoch": 0.35384615384615387,
+      "grad_norm": 0.041986215859651566,
+      "learning_rate": 0.0029413680781758956,
+      "loss": 1.4551,
+      "step": 161
+    },
+    {
+      "epoch": 0.35604395604395606,
+      "grad_norm": 0.008680409751832485,
+      "learning_rate": 0.002938925081433225,
+      "loss": 1.4634,
+      "step": 162
+    },
+    {
+      "epoch": 0.35824175824175825,
+      "grad_norm": 0.006498666945844889,
+      "learning_rate": 0.002936482084690554,
+      "loss": 1.3769,
+      "step": 163
+    },
+    {
+      "epoch": 0.36043956043956044,
+      "grad_norm": 0.008386346511542797,
+      "learning_rate": 0.0029340390879478827,
+      "loss": 1.2803,
+      "step": 164
+    },
+    {
+      "epoch": 0.3626373626373626,
+      "grad_norm": 0.137540802359581,
+      "learning_rate": 0.0029315960912052116,
+      "loss": 1.4055,
+      "step": 165
+    },
+    {
+      "epoch": 0.3648351648351648,
+      "grad_norm": 0.009009119123220444,
+      "learning_rate": 0.0029291530944625405,
+      "loss": 1.5068,
+      "step": 166
+    },
+    {
+      "epoch": 0.367032967032967,
+      "grad_norm": 0.014569335617125034,
+      "learning_rate": 0.00292671009771987,
+      "loss": 1.5839,
+      "step": 167
+    },
+    {
+      "epoch": 0.36923076923076925,
+      "grad_norm": 0.008042916655540466,
+      "learning_rate": 0.0029242671009771988,
+      "loss": 1.5584,
+      "step": 168
+    },
+    {
+      "epoch": 0.37142857142857144,
+      "grad_norm": 0.0038654771633446217,
+      "learning_rate": 0.0029218241042345276,
+      "loss": 1.2381,
+      "step": 169
+    },
+    {
+      "epoch": 0.37362637362637363,
+      "grad_norm": 0.055235568434000015,
+      "learning_rate": 0.0029193811074918565,
+      "loss": 1.4086,
+      "step": 170
+    },
+    {
+      "epoch": 0.3758241758241758,
+      "grad_norm": 0.057514674961566925,
+      "learning_rate": 0.002916938110749186,
+      "loss": 1.3622,
+      "step": 171
+    },
+    {
+      "epoch": 0.378021978021978,
+      "grad_norm": 0.009622988291084766,
+      "learning_rate": 0.0029144951140065148,
+      "loss": 1.3248,
+      "step": 172
+    },
+    {
+      "epoch": 0.3802197802197802,
+      "grad_norm": 0.006263973191380501,
+      "learning_rate": 0.0029120521172638436,
+      "loss": 1.2973,
+      "step": 173
+    },
+    {
+      "epoch": 0.3824175824175824,
+      "grad_norm": 0.005491574760526419,
+      "learning_rate": 0.0029096091205211725,
+      "loss": 1.4183,
+      "step": 174
+    },
+    {
+      "epoch": 0.38461538461538464,
+      "grad_norm": 0.005659735295921564,
+      "learning_rate": 0.002907166123778502,
+      "loss": 1.2996,
+      "step": 175
+    },
+    {
+      "epoch": 0.3868131868131868,
+      "grad_norm": 0.005487625487148762,
+      "learning_rate": 0.0029047231270358308,
+      "loss": 1.6294,
+      "step": 176
+    },
+    {
+      "epoch": 0.389010989010989,
+      "grad_norm": 0.0109627740457654,
+      "learning_rate": 0.0029022801302931596,
+      "loss": 1.4157,
+      "step": 177
+    },
+    {
+      "epoch": 0.3912087912087912,
+      "grad_norm": 0.005304614547640085,
+      "learning_rate": 0.0028998371335504885,
+      "loss": 1.3516,
+      "step": 178
+    },
+    {
+      "epoch": 0.3934065934065934,
+      "grad_norm": 0.11526429653167725,
+      "learning_rate": 0.0028973941368078174,
+      "loss": 1.3041,
+      "step": 179
+    },
+    {
+      "epoch": 0.3956043956043956,
+      "grad_norm": 0.031382568180561066,
+      "learning_rate": 0.0028949511400651468,
+      "loss": 1.2204,
+      "step": 180
+    },
+    {
+      "epoch": 0.3978021978021978,
+      "grad_norm": 0.011900527402758598,
+      "learning_rate": 0.0028925081433224756,
+      "loss": 1.4786,
+      "step": 181
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.014553237706422806,
+      "learning_rate": 0.0028900651465798045,
+      "loss": 1.5034,
+      "step": 182
+    },
+    {
+      "epoch": 0.4021978021978022,
+      "grad_norm": 0.011427970603108406,
+      "learning_rate": 0.0028876221498371334,
+      "loss": 1.3303,
+      "step": 183
+    },
+    {
+      "epoch": 0.4043956043956044,
+      "grad_norm": 0.005950626917183399,
+      "learning_rate": 0.0028851791530944628,
+      "loss": 1.5152,
+      "step": 184
+    },
+    {
+      "epoch": 0.4065934065934066,
+      "grad_norm": 0.012342427857220173,
+      "learning_rate": 0.0028827361563517916,
+      "loss": 1.407,
+      "step": 185
+    },
+    {
+      "epoch": 0.4087912087912088,
+      "grad_norm": 0.07044073939323425,
+      "learning_rate": 0.0028802931596091205,
+      "loss": 1.3078,
+      "step": 186
+    },
+    {
+      "epoch": 0.41098901098901097,
+      "grad_norm": 0.006731156725436449,
+      "learning_rate": 0.0028778501628664494,
+      "loss": 1.454,
+      "step": 187
+    },
+    {
+      "epoch": 0.41318681318681316,
+      "grad_norm": 0.007256140932440758,
+      "learning_rate": 0.0028754071661237783,
+      "loss": 1.4112,
+      "step": 188
+    },
+    {
+      "epoch": 0.4153846153846154,
+      "grad_norm": 0.0282784141600132,
+      "learning_rate": 0.0028729641693811077,
+      "loss": 1.4888,
+      "step": 189
+    },
+    {
+      "epoch": 0.4175824175824176,
+      "grad_norm": 0.011000944301486015,
+      "learning_rate": 0.0028705211726384365,
+      "loss": 1.5954,
+      "step": 190
+    },
+    {
+      "epoch": 0.4197802197802198,
+      "grad_norm": 0.011626143008470535,
+      "learning_rate": 0.0028680781758957654,
+      "loss": 1.5705,
+      "step": 191
+    },
+    {
+      "epoch": 0.421978021978022,
+      "grad_norm": 0.03467955440282822,
+      "learning_rate": 0.0028656351791530943,
+      "loss": 1.6255,
+      "step": 192
+    },
+    {
+      "epoch": 0.42417582417582417,
+      "grad_norm": 0.011410298757255077,
+      "learning_rate": 0.0028631921824104237,
+      "loss": 1.2728,
+      "step": 193
+    },
+    {
+      "epoch": 0.42637362637362636,
+      "grad_norm": 0.008856048807501793,
+      "learning_rate": 0.0028607491856677525,
+      "loss": 1.376,
+      "step": 194
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 0.014439216814935207,
+      "learning_rate": 0.0028583061889250814,
+      "loss": 1.4301,
+      "step": 195
+    },
+    {
+      "epoch": 0.4307692307692308,
+      "grad_norm": 0.018393777310848236,
+      "learning_rate": 0.0028558631921824103,
+      "loss": 1.389,
+      "step": 196
+    },
+    {
+      "epoch": 0.432967032967033,
+      "grad_norm": 0.015493788756430149,
+      "learning_rate": 0.0028534201954397392,
+      "loss": 1.3311,
+      "step": 197
+    },
+    {
+      "epoch": 0.4351648351648352,
+      "grad_norm": 0.14504477381706238,
+      "learning_rate": 0.0028509771986970685,
+      "loss": 1.243,
+      "step": 198
+    },
+    {
+      "epoch": 0.43736263736263736,
+      "grad_norm": 0.004661436192691326,
+      "learning_rate": 0.0028485342019543974,
+      "loss": 1.324,
+      "step": 199
+    },
+    {
+      "epoch": 0.43956043956043955,
+      "grad_norm": 0.011457816697657108,
+      "learning_rate": 0.0028460912052117263,
+      "loss": 1.337,
+      "step": 200
+    },
+    {
+      "epoch": 0.44175824175824174,
+      "grad_norm": 0.006620511878281832,
+      "learning_rate": 0.0028436482084690552,
+      "loss": 1.3078,
+      "step": 201
+    },
+    {
+      "epoch": 0.44395604395604393,
+      "grad_norm": 0.005717736668884754,
+      "learning_rate": 0.0028412052117263845,
+      "loss": 1.4814,
+      "step": 202
+    },
+    {
+      "epoch": 0.4461538461538462,
+      "grad_norm": 0.22640234231948853,
+      "learning_rate": 0.0028387622149837134,
+      "loss": 1.593,
+      "step": 203
+    },
+    {
+      "epoch": 0.44835164835164837,
+      "grad_norm": 0.006773973815143108,
+      "learning_rate": 0.0028363192182410423,
+      "loss": 1.2739,
+      "step": 204
+    },
+    {
+      "epoch": 0.45054945054945056,
+      "grad_norm": 0.02671419084072113,
+      "learning_rate": 0.0028338762214983712,
+      "loss": 1.5083,
+      "step": 205
+    },
+    {
+      "epoch": 0.45274725274725275,
+      "grad_norm": 0.005834756884723902,
+      "learning_rate": 0.0028314332247557,
+      "loss": 1.4515,
+      "step": 206
+    },
+    {
+      "epoch": 0.45494505494505494,
+      "grad_norm": 0.009998585097491741,
+      "learning_rate": 0.0028289902280130294,
+      "loss": 1.3591,
+      "step": 207
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.009909688495099545,
+      "learning_rate": 0.0028265472312703583,
+      "loss": 1.4143,
+      "step": 208
+    },
+    {
+      "epoch": 0.4593406593406593,
+      "grad_norm": 0.005463716108351946,
+      "learning_rate": 0.0028241042345276872,
+      "loss": 1.3672,
+      "step": 209
+    },
+    {
+      "epoch": 0.46153846153846156,
+      "grad_norm": 0.006725917104631662,
+      "learning_rate": 0.002821661237785016,
+      "loss": 1.6644,
+      "step": 210
+    },
+    {
+      "epoch": 0.46373626373626375,
+      "grad_norm": 0.0072638364508748055,
+      "learning_rate": 0.0028192182410423454,
+      "loss": 1.3402,
+      "step": 211
+    },
+    {
+      "epoch": 0.46593406593406594,
+      "grad_norm": 0.11351262778043747,
+      "learning_rate": 0.0028167752442996743,
+      "loss": 1.3117,
+      "step": 212
+    },
+    {
+      "epoch": 0.46813186813186813,
+      "grad_norm": 0.013904514722526073,
+      "learning_rate": 0.0028143322475570032,
+      "loss": 1.4583,
+      "step": 213
+    },
+    {
+      "epoch": 0.4703296703296703,
+      "grad_norm": 0.005298851057887077,
+      "learning_rate": 0.002811889250814332,
+      "loss": 1.3084,
+      "step": 214
+    },
+    {
+      "epoch": 0.4725274725274725,
+      "grad_norm": 0.01244547963142395,
+      "learning_rate": 0.002809446254071661,
+      "loss": 1.1843,
+      "step": 215
+    },
+    {
+      "epoch": 0.4747252747252747,
+      "grad_norm": 0.006712875794619322,
+      "learning_rate": 0.0028070032573289903,
+      "loss": 1.4614,
+      "step": 216
+    },
+    {
+      "epoch": 0.47692307692307695,
+      "grad_norm": 0.009297316893935204,
+      "learning_rate": 0.0028045602605863192,
+      "loss": 1.5011,
+      "step": 217
+    },
+    {
+      "epoch": 0.47912087912087914,
+      "grad_norm": 0.013585609383881092,
+      "learning_rate": 0.002802117263843648,
+      "loss": 1.3613,
+      "step": 218
+    },
+    {
+      "epoch": 0.48131868131868133,
+      "grad_norm": 0.011516177095472813,
+      "learning_rate": 0.002799674267100977,
+      "loss": 1.2949,
+      "step": 219
+    },
+    {
+      "epoch": 0.4835164835164835,
+      "grad_norm": 0.018927421420812607,
+      "learning_rate": 0.0027972312703583063,
+      "loss": 1.3893,
+      "step": 220
+    },
+    {
+      "epoch": 0.4857142857142857,
+      "grad_norm": 0.005685708485543728,
+      "learning_rate": 0.0027947882736156352,
+      "loss": 1.2811,
+      "step": 221
+    },
+    {
+      "epoch": 0.4879120879120879,
+      "grad_norm": 0.016761409118771553,
+      "learning_rate": 0.002792345276872964,
+      "loss": 1.3299,
+      "step": 222
+    },
+    {
+      "epoch": 0.4901098901098901,
+      "grad_norm": 0.0048350561410188675,
+      "learning_rate": 0.002789902280130293,
+      "loss": 1.1971,
+      "step": 223
+    },
+    {
+      "epoch": 0.49230769230769234,
+      "grad_norm": 0.006412938237190247,
+      "learning_rate": 0.002787459283387622,
+      "loss": 1.2833,
+      "step": 224
+    },
+    {
+      "epoch": 0.4945054945054945,
+      "grad_norm": 0.007817007601261139,
+      "learning_rate": 0.0027850162866449512,
+      "loss": 1.5996,
+      "step": 225
+    },
+    {
+      "epoch": 0.4967032967032967,
+      "grad_norm": 0.1362128108739853,
+      "learning_rate": 0.00278257328990228,
+      "loss": 1.214,
+      "step": 226
+    },
+    {
+      "epoch": 0.4989010989010989,
+      "grad_norm": 0.07182040065526962,
+      "learning_rate": 0.002780130293159609,
+      "loss": 1.2171,
+      "step": 227
+    },
+    {
+      "epoch": 0.5010989010989011,
+      "grad_norm": 0.01861986517906189,
+      "learning_rate": 0.002777687296416938,
+      "loss": 1.2822,
+      "step": 228
+    },
+    {
+      "epoch": 0.5032967032967033,
+      "grad_norm": 0.010009989142417908,
+      "learning_rate": 0.0027752442996742672,
+      "loss": 1.4208,
+      "step": 229
+    },
+    {
+      "epoch": 0.5054945054945055,
+      "grad_norm": 0.005975718144327402,
+      "learning_rate": 0.002772801302931596,
+      "loss": 1.3766,
+      "step": 230
+    },
+    {
+      "epoch": 0.5076923076923077,
+      "grad_norm": 0.014379371888935566,
+      "learning_rate": 0.002770358306188925,
+      "loss": 1.509,
+      "step": 231
+    },
+    {
+      "epoch": 0.5098901098901099,
+      "grad_norm": 0.0055440496653318405,
+      "learning_rate": 0.002767915309446254,
+      "loss": 1.372,
+      "step": 232
+    },
+    {
+      "epoch": 0.512087912087912,
+      "grad_norm": 0.011932543478906155,
+      "learning_rate": 0.002765472312703583,
+      "loss": 1.6592,
+      "step": 233
+    },
+    {
+      "epoch": 0.5142857142857142,
+      "grad_norm": 0.007669544313102961,
+      "learning_rate": 0.002763029315960912,
+      "loss": 1.3871,
+      "step": 234
+    },
+    {
+      "epoch": 0.5164835164835165,
+      "grad_norm": 0.010674835182726383,
+      "learning_rate": 0.002760586319218241,
+      "loss": 1.5464,
+      "step": 235
+    },
+    {
+      "epoch": 0.5186813186813187,
+      "grad_norm": 0.01164399366825819,
+      "learning_rate": 0.00275814332247557,
+      "loss": 1.4339,
+      "step": 236
+    },
+    {
+      "epoch": 0.5208791208791209,
+      "grad_norm": 0.007403552532196045,
+      "learning_rate": 0.002755700325732899,
+      "loss": 1.3027,
+      "step": 237
+    },
+    {
+      "epoch": 0.5230769230769231,
+      "grad_norm": 0.0042869155295193195,
+      "learning_rate": 0.002753257328990228,
+      "loss": 1.3542,
+      "step": 238
+    },
+    {
+      "epoch": 0.5252747252747253,
+      "grad_norm": 0.008637145161628723,
+      "learning_rate": 0.002750814332247557,
+      "loss": 1.2705,
+      "step": 239
+    },
+    {
+      "epoch": 0.5274725274725275,
+      "grad_norm": 0.00813508965075016,
+      "learning_rate": 0.002748371335504886,
+      "loss": 1.408,
+      "step": 240
+    },
+    {
+      "epoch": 0.5296703296703297,
+      "grad_norm": 0.006155966315418482,
+      "learning_rate": 0.002745928338762215,
+      "loss": 1.2112,
+      "step": 241
+    },
+    {
+      "epoch": 0.5318681318681319,
+      "grad_norm": 0.025280984118580818,
+      "learning_rate": 0.002743485342019544,
+      "loss": 1.4142,
+      "step": 242
+    },
+    {
+      "epoch": 0.5340659340659341,
+      "grad_norm": 0.061497267335653305,
+      "learning_rate": 0.002741042345276873,
+      "loss": 1.4127,
+      "step": 243
+    },
+    {
+      "epoch": 0.5362637362637362,
+      "grad_norm": 0.006200543604791164,
+      "learning_rate": 0.002738599348534202,
+      "loss": 1.3996,
+      "step": 244
+    },
+    {
+      "epoch": 0.5384615384615384,
+      "grad_norm": 0.006728803273290396,
+      "learning_rate": 0.002736156351791531,
+      "loss": 1.4453,
+      "step": 245
+    },
+    {
+      "epoch": 0.5406593406593406,
+      "grad_norm": 0.05864905193448067,
+      "learning_rate": 0.0027337133550488597,
+      "loss": 1.3365,
+      "step": 246
+    },
+    {
+      "epoch": 0.5428571428571428,
+      "grad_norm": 0.1165623739361763,
+      "learning_rate": 0.002731270358306189,
+      "loss": 1.7673,
+      "step": 247
+    },
+    {
+      "epoch": 0.545054945054945,
+      "grad_norm": 0.01865364983677864,
+      "learning_rate": 0.002728827361563518,
+      "loss": 1.464,
+      "step": 248
+    },
+    {
+      "epoch": 0.5472527472527473,
+      "grad_norm": 0.006465948652476072,
+      "learning_rate": 0.002726384364820847,
+      "loss": 1.165,
+      "step": 249
+    },
+    {
+      "epoch": 0.5494505494505495,
+      "grad_norm": 0.00611953204497695,
+      "learning_rate": 0.0027239413680781757,
+      "loss": 1.4508,
+      "step": 250
+    },
+    {
+      "epoch": 0.5516483516483517,
+      "grad_norm": 0.07096860557794571,
+      "learning_rate": 0.002721498371335505,
+      "loss": 1.3195,
+      "step": 251
+    },
+    {
+      "epoch": 0.5538461538461539,
+      "grad_norm": 0.013239827938377857,
+      "learning_rate": 0.002719055374592834,
+      "loss": 1.2598,
+      "step": 252
+    },
+    {
+      "epoch": 0.5560439560439561,
+      "grad_norm": 0.02189604938030243,
+      "learning_rate": 0.002716612377850163,
+      "loss": 1.394,
+      "step": 253
+    },
+    {
+      "epoch": 0.5582417582417583,
+      "grad_norm": 0.006946844048798084,
+      "learning_rate": 0.0027141693811074917,
+      "loss": 1.598,
+      "step": 254
+    },
+    {
+      "epoch": 0.5604395604395604,
+      "grad_norm": 0.010237875394523144,
+      "learning_rate": 0.0027117263843648206,
+      "loss": 1.3354,
+      "step": 255
+    },
+    {
+      "epoch": 0.5626373626373626,
+      "grad_norm": 0.007440627086907625,
+      "learning_rate": 0.00270928338762215,
+      "loss": 1.4352,
+      "step": 256
+    },
+    {
+      "epoch": 0.5648351648351648,
+      "grad_norm": 0.009911991655826569,
+      "learning_rate": 0.002706840390879479,
+      "loss": 1.3056,
+      "step": 257
+    },
+    {
+      "epoch": 0.567032967032967,
+      "grad_norm": 0.007230260409414768,
+      "learning_rate": 0.0027043973941368077,
+      "loss": 1.3746,
+      "step": 258
+    },
+    {
+      "epoch": 0.5692307692307692,
+      "grad_norm": 0.011950643733143806,
+      "learning_rate": 0.0027019543973941366,
+      "loss": 1.4084,
+      "step": 259
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.006813615094870329,
+      "learning_rate": 0.002699511400651466,
+      "loss": 1.3169,
+      "step": 260
+    },
+    {
+      "epoch": 0.5736263736263736,
+      "grad_norm": 0.012173987925052643,
+      "learning_rate": 0.002697068403908795,
+      "loss": 1.3836,
+      "step": 261
+    },
+    {
+      "epoch": 0.5758241758241758,
+      "grad_norm": 0.009055081754922867,
+      "learning_rate": 0.0026946254071661237,
+      "loss": 1.2159,
+      "step": 262
+    },
+    {
+      "epoch": 0.578021978021978,
+      "grad_norm": 0.004840587731450796,
+      "learning_rate": 0.0026921824104234526,
+      "loss": 1.2863,
+      "step": 263
+    },
+    {
+      "epoch": 0.5802197802197803,
+      "grad_norm": 0.02011577971279621,
+      "learning_rate": 0.0026897394136807815,
+      "loss": 1.345,
+      "step": 264
+    },
+    {
+      "epoch": 0.5824175824175825,
+      "grad_norm": 0.01080879382789135,
+      "learning_rate": 0.002687296416938111,
+      "loss": 1.429,
+      "step": 265
+    },
+    {
+      "epoch": 0.5846153846153846,
+      "grad_norm": 0.005254679825156927,
+      "learning_rate": 0.0026848534201954397,
+      "loss": 1.3179,
+      "step": 266
+    },
+    {
+      "epoch": 0.5868131868131868,
+      "grad_norm": 0.019347960129380226,
+      "learning_rate": 0.0026824104234527686,
+      "loss": 1.4902,
+      "step": 267
+    },
+    {
+      "epoch": 0.589010989010989,
+      "grad_norm": 0.00422753905877471,
+      "learning_rate": 0.0026799674267100975,
+      "loss": 1.394,
+      "step": 268
+    },
+    {
+      "epoch": 0.5912087912087912,
+      "grad_norm": 0.006875904742628336,
+      "learning_rate": 0.002677524429967427,
+      "loss": 1.2938,
+      "step": 269
+    },
+    {
+      "epoch": 0.5934065934065934,
+      "grad_norm": 0.007426791824400425,
+      "learning_rate": 0.0026750814332247557,
+      "loss": 1.2852,
+      "step": 270
+    },
+    {
+      "epoch": 0.5956043956043956,
+      "grad_norm": 0.06034421920776367,
+      "learning_rate": 0.0026726384364820846,
+      "loss": 1.3948,
+      "step": 271
+    },
+    {
+      "epoch": 0.5978021978021978,
+      "grad_norm": 0.05798085033893585,
+      "learning_rate": 0.0026701954397394135,
+      "loss": 1.382,
+      "step": 272
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.007921774871647358,
+      "learning_rate": 0.0026677524429967424,
+      "loss": 1.7273,
+      "step": 273
+    },
+    {
+      "epoch": 0.6021978021978022,
+      "grad_norm": 0.008947125636041164,
+      "learning_rate": 0.0026653094462540717,
+      "loss": 1.3259,
+      "step": 274
+    },
+    {
+      "epoch": 0.6043956043956044,
+      "grad_norm": 0.007731279823929071,
+      "learning_rate": 0.0026628664495114006,
+      "loss": 1.4904,
+      "step": 275
+    },
+    {
+      "epoch": 0.6065934065934065,
+      "grad_norm": 1.1332515478134155,
+      "learning_rate": 0.0026604234527687295,
+      "loss": 1.6831,
+      "step": 276
+    },
+    {
+      "epoch": 0.6087912087912087,
+      "grad_norm": 0.005166809074580669,
+      "learning_rate": 0.0026579804560260584,
+      "loss": 1.3199,
+      "step": 277
+    },
+    {
+      "epoch": 0.610989010989011,
+      "grad_norm": 0.00900485273450613,
+      "learning_rate": 0.0026555374592833877,
+      "loss": 1.1974,
+      "step": 278
+    },
+    {
+      "epoch": 0.6131868131868132,
+      "grad_norm": 0.01591731235384941,
+      "learning_rate": 0.0026530944625407166,
+      "loss": 1.3258,
+      "step": 279
+    },
+    {
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.007398122455924749,
+      "learning_rate": 0.0026506514657980455,
+      "loss": 1.3826,
+      "step": 280
+    },
+    {
+      "epoch": 0.6175824175824176,
+      "grad_norm": 0.00645383819937706,
+      "learning_rate": 0.0026482084690553744,
+      "loss": 1.2519,
+      "step": 281
+    },
+    {
+      "epoch": 0.6197802197802198,
+      "grad_norm": 0.009886123239994049,
+      "learning_rate": 0.0026457654723127033,
+      "loss": 1.1429,
+      "step": 282
+    },
+    {
+      "epoch": 0.621978021978022,
+      "grad_norm": 0.010514532215893269,
+      "learning_rate": 0.0026433224755700326,
+      "loss": 1.4621,
+      "step": 283
+    },
+    {
+      "epoch": 0.6241758241758242,
+      "grad_norm": 0.04716843366622925,
+      "learning_rate": 0.0026408794788273615,
+      "loss": 1.3415,
+      "step": 284
+    },
+    {
+      "epoch": 0.6263736263736264,
+      "grad_norm": 0.007175733335316181,
+      "learning_rate": 0.0026384364820846904,
+      "loss": 1.4425,
+      "step": 285
+    },
+    {
+      "epoch": 0.6285714285714286,
+      "grad_norm": 0.006638946942985058,
+      "learning_rate": 0.0026359934853420193,
+      "loss": 1.4221,
+      "step": 286
+    },
+    {
+      "epoch": 0.6307692307692307,
+      "grad_norm": 0.004308321978896856,
+      "learning_rate": 0.0026335504885993486,
+      "loss": 1.249,
+      "step": 287
+    },
+    {
+      "epoch": 0.6329670329670329,
+      "grad_norm": 0.008631543256342411,
+      "learning_rate": 0.0026311074918566775,
+      "loss": 1.5098,
+      "step": 288
+    },
+    {
+      "epoch": 0.6351648351648351,
+      "grad_norm": 0.00608354015275836,
+      "learning_rate": 0.0026286644951140064,
+      "loss": 1.3762,
+      "step": 289
+    },
+    {
+      "epoch": 0.6373626373626373,
+      "grad_norm": 0.02524668164551258,
+      "learning_rate": 0.0026262214983713353,
+      "loss": 1.3684,
+      "step": 290
+    },
+    {
+      "epoch": 0.6395604395604395,
+      "grad_norm": 0.005527480039745569,
+      "learning_rate": 0.0026237785016286646,
+      "loss": 1.2645,
+      "step": 291
+    },
+    {
+      "epoch": 0.6417582417582418,
+      "grad_norm": 0.008363848552107811,
+      "learning_rate": 0.0026213355048859935,
+      "loss": 1.3048,
+      "step": 292
+    },
+    {
+      "epoch": 0.643956043956044,
+      "grad_norm": 0.006506824400275946,
+      "learning_rate": 0.002618892508143323,
+      "loss": 1.2547,
+      "step": 293
+    },
+    {
+      "epoch": 0.6461538461538462,
+      "grad_norm": 0.009241343475878239,
+      "learning_rate": 0.0026164495114006517,
+      "loss": 1.2742,
+      "step": 294
+    },
+    {
+      "epoch": 0.6483516483516484,
+      "grad_norm": 0.00860565435141325,
+      "learning_rate": 0.0026140065146579806,
+      "loss": 1.2628,
+      "step": 295
+    },
+    {
+      "epoch": 0.6505494505494506,
+      "grad_norm": 0.008727004751563072,
+      "learning_rate": 0.0026115635179153095,
+      "loss": 1.3899,
+      "step": 296
+    },
+    {
+      "epoch": 0.6527472527472528,
+      "grad_norm": 0.004602030385285616,
+      "learning_rate": 0.002609120521172639,
+      "loss": 1.3138,
+      "step": 297
+    },
+    {
+      "epoch": 0.654945054945055,
+      "grad_norm": 0.0060708895325660706,
+      "learning_rate": 0.0026066775244299677,
+      "loss": 1.5001,
+      "step": 298
+    },
+    {
+      "epoch": 0.6571428571428571,
+      "grad_norm": 0.04194958508014679,
+      "learning_rate": 0.0026042345276872966,
+      "loss": 1.4954,
+      "step": 299
+    },
+    {
+      "epoch": 0.6593406593406593,
+      "grad_norm": 0.00692878570407629,
+      "learning_rate": 0.0026017915309446255,
+      "loss": 1.3652,
+      "step": 300
+    },
+    {
+      "epoch": 0.6615384615384615,
+      "grad_norm": 0.025298116728663445,
+      "learning_rate": 0.0025993485342019544,
+      "loss": 1.4154,
+      "step": 301
+    },
+    {
+      "epoch": 0.6637362637362637,
+      "grad_norm": 0.005875185132026672,
+      "learning_rate": 0.0025969055374592837,
+      "loss": 1.5054,
+      "step": 302
+    },
+    {
+      "epoch": 0.6659340659340659,
+      "grad_norm": 0.007307960651814938,
+      "learning_rate": 0.0025944625407166126,
+      "loss": 1.3102,
+      "step": 303
+    },
+    {
+      "epoch": 0.6681318681318681,
+      "grad_norm": 0.008598933927714825,
+      "learning_rate": 0.0025920195439739415,
+      "loss": 1.2383,
+      "step": 304
+    },
+    {
+      "epoch": 0.6703296703296703,
+      "grad_norm": 0.011500825174152851,
+      "learning_rate": 0.0025895765472312704,
+      "loss": 1.4187,
+      "step": 305
+    },
+    {
+      "epoch": 0.6725274725274726,
+      "grad_norm": 0.01613699458539486,
+      "learning_rate": 0.0025871335504885997,
+      "loss": 1.3789,
+      "step": 306
+    },
+    {
+      "epoch": 0.6747252747252748,
+      "grad_norm": 0.006763146258890629,
+      "learning_rate": 0.0025846905537459286,
+      "loss": 1.2362,
+      "step": 307
+    },
+    {
+      "epoch": 0.676923076923077,
+      "grad_norm": 0.14663328230381012,
+      "learning_rate": 0.0025822475570032575,
+      "loss": 1.5981,
+      "step": 308
+    },
+    {
+      "epoch": 0.6791208791208792,
+      "grad_norm": 0.009048673324286938,
+      "learning_rate": 0.0025798045602605864,
+      "loss": 1.2274,
+      "step": 309
+    },
+    {
+      "epoch": 0.6813186813186813,
+      "grad_norm": 0.005770419258624315,
+      "learning_rate": 0.0025773615635179153,
+      "loss": 1.338,
+      "step": 310
+    },
+    {
+      "epoch": 0.6835164835164835,
+      "grad_norm": 0.03282594680786133,
+      "learning_rate": 0.0025749185667752446,
+      "loss": 1.4783,
+      "step": 311
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.03209996223449707,
+      "learning_rate": 0.0025724755700325735,
+      "loss": 1.4697,
+      "step": 312
+    },
+    {
+      "epoch": 0.6879120879120879,
+      "grad_norm": 0.016078608110547066,
+      "learning_rate": 0.0025700325732899024,
+      "loss": 1.3134,
+      "step": 313
+    },
+    {
+      "epoch": 0.6901098901098901,
+      "grad_norm": 0.00448894826695323,
+      "learning_rate": 0.0025675895765472313,
+      "loss": 1.2063,
+      "step": 314
+    },
+    {
+      "epoch": 0.6923076923076923,
+      "grad_norm": 0.0286596417427063,
+      "learning_rate": 0.0025651465798045606,
+      "loss": 1.279,
+      "step": 315
+    },
+    {
+      "epoch": 0.6945054945054945,
+      "grad_norm": 0.007250271271914244,
+      "learning_rate": 0.0025627035830618895,
+      "loss": 1.2777,
+      "step": 316
+    },
+    {
+      "epoch": 0.6967032967032967,
+      "grad_norm": 0.008373573422431946,
+      "learning_rate": 0.0025602605863192184,
+      "loss": 1.5309,
+      "step": 317
+    },
+    {
+      "epoch": 0.6989010989010989,
+      "grad_norm": 0.005867587402462959,
+      "learning_rate": 0.0025578175895765473,
+      "loss": 1.3943,
+      "step": 318
+    },
+    {
+      "epoch": 0.701098901098901,
+      "grad_norm": 0.009243875741958618,
+      "learning_rate": 0.0025553745928338766,
+      "loss": 1.4646,
+      "step": 319
+    },
+    {
+      "epoch": 0.7032967032967034,
+      "grad_norm": 0.03471420705318451,
+      "learning_rate": 0.0025529315960912055,
+      "loss": 1.5,
+      "step": 320
+    },
+    {
+      "epoch": 0.7054945054945055,
+      "grad_norm": 0.009368221275508404,
+      "learning_rate": 0.0025504885993485344,
+      "loss": 1.4766,
+      "step": 321
+    },
+    {
+      "epoch": 0.7076923076923077,
+      "grad_norm": 0.007304329425096512,
+      "learning_rate": 0.0025480456026058633,
+      "loss": 1.3157,
+      "step": 322
+    },
+    {
+      "epoch": 0.7098901098901099,
+      "grad_norm": 0.01224322710186243,
+      "learning_rate": 0.002545602605863192,
+      "loss": 1.3634,
+      "step": 323
+    },
+    {
+      "epoch": 0.7120879120879121,
+      "grad_norm": 0.003647648496553302,
+      "learning_rate": 0.0025431596091205215,
+      "loss": 1.164,
+      "step": 324
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.00685991020873189,
+      "learning_rate": 0.0025407166123778504,
+      "loss": 1.4639,
+      "step": 325
+    },
+    {
+      "epoch": 0.7164835164835165,
+      "grad_norm": 0.00489164749160409,
+      "learning_rate": 0.0025382736156351793,
+      "loss": 1.316,
+      "step": 326
+    },
+    {
+      "epoch": 0.7186813186813187,
+      "grad_norm": 0.026210328564047813,
+      "learning_rate": 0.002535830618892508,
+      "loss": 1.5678,
+      "step": 327
+    },
+    {
+      "epoch": 0.7208791208791209,
+      "grad_norm": 0.005729404743760824,
+      "learning_rate": 0.0025333876221498375,
+      "loss": 1.3255,
+      "step": 328
+    },
+    {
+      "epoch": 0.7230769230769231,
+      "grad_norm": 0.007942290045320988,
+      "learning_rate": 0.0025309446254071664,
+      "loss": 1.1786,
+      "step": 329
+    },
+    {
+      "epoch": 0.7252747252747253,
+      "grad_norm": 0.010033193044364452,
+      "learning_rate": 0.0025285016286644953,
+      "loss": 1.4401,
+      "step": 330
+    },
+    {
+      "epoch": 0.7274725274725274,
+      "grad_norm": 0.010759457014501095,
+      "learning_rate": 0.002526058631921824,
+      "loss": 1.3325,
+      "step": 331
+    },
+    {
+      "epoch": 0.7296703296703296,
+      "grad_norm": 0.009265536442399025,
+      "learning_rate": 0.002523615635179153,
+      "loss": 1.1382,
+      "step": 332
+    },
+    {
+      "epoch": 0.7318681318681318,
+      "grad_norm": 0.005714711733162403,
+      "learning_rate": 0.0025211726384364824,
+      "loss": 1.3886,
+      "step": 333
+    },
+    {
+      "epoch": 0.734065934065934,
+      "grad_norm": 0.03787574544548988,
+      "learning_rate": 0.0025187296416938113,
+      "loss": 1.4105,
+      "step": 334
+    },
+    {
+      "epoch": 0.7362637362637363,
+      "grad_norm": 0.019143305718898773,
+      "learning_rate": 0.00251628664495114,
+      "loss": 1.4213,
+      "step": 335
+    },
+    {
+      "epoch": 0.7384615384615385,
+      "grad_norm": 0.014862719923257828,
+      "learning_rate": 0.002513843648208469,
+      "loss": 1.3195,
+      "step": 336
+    },
+    {
+      "epoch": 0.7406593406593407,
+      "grad_norm": 0.024634849280118942,
+      "learning_rate": 0.0025114006514657984,
+      "loss": 1.412,
+      "step": 337
+    },
+    {
+      "epoch": 0.7428571428571429,
+      "grad_norm": 0.012242087163031101,
+      "learning_rate": 0.0025089576547231273,
+      "loss": 1.5884,
+      "step": 338
+    },
+    {
+      "epoch": 0.7450549450549451,
+      "grad_norm": 0.008459597826004028,
+      "learning_rate": 0.002506514657980456,
+      "loss": 1.2119,
+      "step": 339
+    },
+    {
+      "epoch": 0.7472527472527473,
+      "grad_norm": 0.015496046282351017,
+      "learning_rate": 0.002504071661237785,
+      "loss": 1.3869,
+      "step": 340
+    },
+    {
+      "epoch": 0.7494505494505495,
+      "grad_norm": 0.03498433530330658,
+      "learning_rate": 0.002501628664495114,
+      "loss": 1.2898,
+      "step": 341
+    },
+    {
+      "epoch": 0.7516483516483516,
+      "grad_norm": 0.008740990422666073,
+      "learning_rate": 0.0024991856677524433,
+      "loss": 1.2335,
+      "step": 342
+    },
+    {
+      "epoch": 0.7538461538461538,
+      "grad_norm": 0.007045309524983168,
+      "learning_rate": 0.002496742671009772,
+      "loss": 1.2747,
+      "step": 343
+    },
+    {
+      "epoch": 0.756043956043956,
+      "grad_norm": 0.07084590941667557,
+      "learning_rate": 0.002494299674267101,
+      "loss": 1.3921,
+      "step": 344
+    },
+    {
+      "epoch": 0.7582417582417582,
+      "grad_norm": 0.06975960731506348,
+      "learning_rate": 0.00249185667752443,
+      "loss": 0.9892,
+      "step": 345
+    },
+    {
+      "epoch": 0.7604395604395604,
+      "grad_norm": 0.012194158509373665,
+      "learning_rate": 0.0024894136807817593,
+      "loss": 1.2227,
+      "step": 346
+    },
+    {
+      "epoch": 0.7626373626373626,
+      "grad_norm": 0.011062041856348515,
+      "learning_rate": 0.002486970684039088,
+      "loss": 1.4682,
+      "step": 347
+    },
+    {
+      "epoch": 0.7648351648351648,
+      "grad_norm": 0.013519876636564732,
+      "learning_rate": 0.002484527687296417,
+      "loss": 1.3041,
+      "step": 348
+    },
+    {
+      "epoch": 0.7670329670329671,
+      "grad_norm": 0.019174879416823387,
+      "learning_rate": 0.002482084690553746,
+      "loss": 1.2814,
+      "step": 349
+    },
+    {
+      "epoch": 0.7692307692307693,
+      "grad_norm": 0.005717654246836901,
+      "learning_rate": 0.002479641693811075,
+      "loss": 1.2942,
+      "step": 350
+    },
+    {
+      "epoch": 0.7714285714285715,
+      "grad_norm": 0.00967381987720728,
+      "learning_rate": 0.002477198697068404,
+      "loss": 1.4044,
+      "step": 351
+    },
+    {
+      "epoch": 0.7736263736263737,
+      "grad_norm": 0.005588676780462265,
+      "learning_rate": 0.002474755700325733,
+      "loss": 1.1513,
+      "step": 352
+    },
+    {
+      "epoch": 0.7758241758241758,
+      "grad_norm": 0.008063740096986294,
+      "learning_rate": 0.002472312703583062,
+      "loss": 1.4,
+      "step": 353
+    },
+    {
+      "epoch": 0.778021978021978,
+      "grad_norm": 0.1522638350725174,
+      "learning_rate": 0.002469869706840391,
+      "loss": 1.3363,
+      "step": 354
+    },
+    {
+      "epoch": 0.7802197802197802,
+      "grad_norm": 0.007826803252100945,
+      "learning_rate": 0.00246742671009772,
+      "loss": 1.4628,
+      "step": 355
+    },
+    {
+      "epoch": 0.7824175824175824,
+      "grad_norm": 0.007838739082217216,
+      "learning_rate": 0.002464983713355049,
+      "loss": 1.3455,
+      "step": 356
+    },
+    {
+      "epoch": 0.7846153846153846,
+      "grad_norm": 0.01171797700226307,
+      "learning_rate": 0.002462540716612378,
+      "loss": 1.2655,
+      "step": 357
+    },
+    {
+      "epoch": 0.7868131868131868,
+      "grad_norm": 0.009182740934193134,
+      "learning_rate": 0.002460097719869707,
+      "loss": 1.4408,
+      "step": 358
+    },
+    {
+      "epoch": 0.789010989010989,
+      "grad_norm": 0.009247886016964912,
+      "learning_rate": 0.0024576547231270358,
+      "loss": 1.3946,
+      "step": 359
+    },
+    {
+      "epoch": 0.7912087912087912,
+      "grad_norm": 0.009077033028006554,
+      "learning_rate": 0.002455211726384365,
+      "loss": 1.3463,
+      "step": 360
+    },
+    {
+      "epoch": 0.7934065934065934,
+      "grad_norm": 0.014478461816906929,
+      "learning_rate": 0.002452768729641694,
+      "loss": 1.3339,
+      "step": 361
+    },
+    {
+      "epoch": 0.7956043956043956,
+      "grad_norm": 0.01966753974556923,
+      "learning_rate": 0.002450325732899023,
+      "loss": 1.2821,
+      "step": 362
+    },
+    {
+      "epoch": 0.7978021978021979,
+      "grad_norm": 0.008981863968074322,
+      "learning_rate": 0.0024478827361563518,
+      "loss": 1.5259,
+      "step": 363
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.00753204058855772,
+      "learning_rate": 0.002445439739413681,
+      "loss": 1.2244,
+      "step": 364
+    },
+    {
+      "epoch": 0.8021978021978022,
+      "grad_norm": 0.015534610487520695,
+      "learning_rate": 0.00244299674267101,
+      "loss": 1.4722,
+      "step": 365
+    },
+    {
+      "epoch": 0.8043956043956044,
+      "grad_norm": 0.02946232073009014,
+      "learning_rate": 0.002440553745928339,
+      "loss": 1.4607,
+      "step": 366
+    },
+    {
+      "epoch": 0.8065934065934066,
+      "grad_norm": 0.005813276395201683,
+      "learning_rate": 0.0024381107491856678,
+      "loss": 1.2336,
+      "step": 367
+    },
+    {
+      "epoch": 0.8087912087912088,
+      "grad_norm": 0.0075688669458031654,
+      "learning_rate": 0.0024356677524429967,
+      "loss": 1.4462,
+      "step": 368
+    },
+    {
+      "epoch": 0.810989010989011,
+      "grad_norm": 0.007163495756685734,
+      "learning_rate": 0.002433224755700326,
+      "loss": 1.2909,
+      "step": 369
+    },
+    {
+      "epoch": 0.8131868131868132,
+      "grad_norm": 0.011231552809476852,
+      "learning_rate": 0.002430781758957655,
+      "loss": 1.342,
+      "step": 370
+    },
+    {
+      "epoch": 0.8153846153846154,
+      "grad_norm": 0.01787695288658142,
+      "learning_rate": 0.0024283387622149838,
+      "loss": 1.37,
+      "step": 371
+    },
+    {
+      "epoch": 0.8175824175824176,
+      "grad_norm": 0.03626333549618721,
+      "learning_rate": 0.0024258957654723127,
+      "loss": 1.3631,
+      "step": 372
+    },
+    {
+      "epoch": 0.8197802197802198,
+      "grad_norm": 0.0648449957370758,
+      "learning_rate": 0.002423452768729642,
+      "loss": 1.305,
+      "step": 373
+    },
+    {
+      "epoch": 0.8219780219780219,
+      "grad_norm": 0.009728919714689255,
+      "learning_rate": 0.002421009771986971,
+      "loss": 1.3603,
+      "step": 374
+    },
+    {
+      "epoch": 0.8241758241758241,
+      "grad_norm": 0.0502760075032711,
+      "learning_rate": 0.0024185667752442998,
+      "loss": 1.2815,
+      "step": 375
+    },
+    {
+      "epoch": 0.8263736263736263,
+      "grad_norm": 0.01963469572365284,
+      "learning_rate": 0.0024161237785016287,
+      "loss": 1.5425,
+      "step": 376
+    },
+    {
+      "epoch": 0.8285714285714286,
+      "grad_norm": 0.006639283616095781,
+      "learning_rate": 0.0024136807817589576,
+      "loss": 1.5299,
+      "step": 377
+    },
+    {
+      "epoch": 0.8307692307692308,
+      "grad_norm": 0.007913485169410706,
+      "learning_rate": 0.002411237785016287,
+      "loss": 1.3743,
+      "step": 378
+    },
+    {
+      "epoch": 0.832967032967033,
+      "grad_norm": 0.022798698395490646,
+      "learning_rate": 0.0024087947882736158,
+      "loss": 1.3395,
+      "step": 379
+    },
+    {
+      "epoch": 0.8351648351648352,
+      "grad_norm": 0.02612307108938694,
+      "learning_rate": 0.0024063517915309447,
+      "loss": 1.2235,
+      "step": 380
+    },
+    {
+      "epoch": 0.8373626373626374,
+      "grad_norm": 0.012525797821581364,
+      "learning_rate": 0.0024039087947882736,
+      "loss": 1.5225,
+      "step": 381
+    },
+    {
+      "epoch": 0.8395604395604396,
+      "grad_norm": 0.04603765904903412,
+      "learning_rate": 0.002401465798045603,
+      "loss": 1.2747,
+      "step": 382
+    },
+    {
+      "epoch": 0.8417582417582418,
+      "grad_norm": 0.017153330147266388,
+      "learning_rate": 0.0023990228013029318,
+      "loss": 1.358,
+      "step": 383
+    },
+    {
+      "epoch": 0.843956043956044,
+      "grad_norm": 0.021469995379447937,
+      "learning_rate": 0.0023965798045602607,
+      "loss": 1.3299,
+      "step": 384
+    },
+    {
+      "epoch": 0.8461538461538461,
+      "grad_norm": 0.008087101392447948,
+      "learning_rate": 0.0023941368078175896,
+      "loss": 1.3,
+      "step": 385
+    },
+    {
+      "epoch": 0.8483516483516483,
+      "grad_norm": 0.012908970937132835,
+      "learning_rate": 0.002391693811074919,
+      "loss": 1.3844,
+      "step": 386
+    },
+    {
+      "epoch": 0.8505494505494505,
+      "grad_norm": 0.09564747661352158,
+      "learning_rate": 0.0023892508143322478,
+      "loss": 1.603,
+      "step": 387
+    },
+    {
+      "epoch": 0.8527472527472527,
+      "grad_norm": 0.02637900598347187,
+      "learning_rate": 0.0023868078175895767,
+      "loss": 1.2457,
+      "step": 388
+    },
+    {
+      "epoch": 0.8549450549450549,
+      "grad_norm": 0.021444106474518776,
+      "learning_rate": 0.0023843648208469056,
+      "loss": 1.2989,
+      "step": 389
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 0.029872575774788857,
+      "learning_rate": 0.0023819218241042345,
+      "loss": 1.4851,
+      "step": 390
+    },
+    {
+      "epoch": 0.8593406593406593,
+      "grad_norm": 0.012310764752328396,
+      "learning_rate": 0.002379478827361564,
+      "loss": 1.3087,
+      "step": 391
+    },
+    {
+      "epoch": 0.8615384615384616,
+      "grad_norm": 0.020886827260255814,
+      "learning_rate": 0.0023770358306188927,
+      "loss": 1.3427,
+      "step": 392
+    },
+    {
+      "epoch": 0.8637362637362638,
+      "grad_norm": 0.01470887754112482,
+      "learning_rate": 0.0023745928338762216,
+      "loss": 1.2669,
+      "step": 393
+    },
+    {
+      "epoch": 0.865934065934066,
+      "grad_norm": 0.07828725129365921,
+      "learning_rate": 0.0023721498371335505,
+      "loss": 1.4278,
+      "step": 394
+    },
+    {
+      "epoch": 0.8681318681318682,
+      "grad_norm": 0.05607123672962189,
+      "learning_rate": 0.00236970684039088,
+      "loss": 1.2852,
+      "step": 395
+    },
+    {
+      "epoch": 0.8703296703296703,
+      "grad_norm": 0.018453998491168022,
+      "learning_rate": 0.0023672638436482087,
+      "loss": 1.412,
+      "step": 396
+    },
+    {
+      "epoch": 0.8725274725274725,
+      "grad_norm": 0.056366316974163055,
+      "learning_rate": 0.0023648208469055376,
+      "loss": 1.2388,
+      "step": 397
+    },
+    {
+      "epoch": 0.8747252747252747,
+      "grad_norm": 0.018255451694130898,
+      "learning_rate": 0.0023623778501628665,
+      "loss": 1.427,
+      "step": 398
+    },
+    {
+      "epoch": 0.8769230769230769,
+      "grad_norm": 0.01393771730363369,
+      "learning_rate": 0.0023599348534201954,
+      "loss": 1.2435,
+      "step": 399
+    },
+    {
+      "epoch": 0.8791208791208791,
+      "grad_norm": 0.027305787429213524,
+      "learning_rate": 0.0023574918566775247,
+      "loss": 1.291,
+      "step": 400
+    },
+    {
+      "epoch": 0.8813186813186813,
+      "grad_norm": 0.01781146600842476,
+      "learning_rate": 0.0023550488599348536,
+      "loss": 1.3849,
+      "step": 401
+    },
+    {
+      "epoch": 0.8835164835164835,
+      "grad_norm": 0.024978850036859512,
+      "learning_rate": 0.0023526058631921825,
+      "loss": 1.4405,
+      "step": 402
+    },
+    {
+      "epoch": 0.8857142857142857,
+      "grad_norm": 0.010903376154601574,
+      "learning_rate": 0.0023501628664495114,
+      "loss": 1.301,
+      "step": 403
+    },
+    {
+      "epoch": 0.8879120879120879,
+      "grad_norm": 0.01275777630507946,
+      "learning_rate": 0.0023477198697068407,
+      "loss": 1.2068,
+      "step": 404
+    },
+    {
+      "epoch": 0.8901098901098901,
+      "grad_norm": 0.015543562360107899,
+      "learning_rate": 0.0023452768729641696,
+      "loss": 1.2853,
+      "step": 405
+    },
+    {
+      "epoch": 0.8923076923076924,
+      "grad_norm": 0.011452585458755493,
+      "learning_rate": 0.0023428338762214985,
+      "loss": 1.276,
+      "step": 406
+    },
+    {
+      "epoch": 0.8945054945054945,
+      "grad_norm": 0.01002784725278616,
+      "learning_rate": 0.0023403908794788274,
+      "loss": 1.2248,
+      "step": 407
+    },
+    {
+      "epoch": 0.8967032967032967,
+      "grad_norm": 0.03246638923883438,
+      "learning_rate": 0.0023379478827361562,
+      "loss": 1.3455,
+      "step": 408
+    },
+    {
+      "epoch": 0.8989010989010989,
+      "grad_norm": 0.0223382581025362,
+      "learning_rate": 0.0023355048859934856,
+      "loss": 1.4396,
+      "step": 409
+    },
+    {
+      "epoch": 0.9010989010989011,
+      "grad_norm": 0.014086885377764702,
+      "learning_rate": 0.0023330618892508145,
+      "loss": 1.2226,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032967032967033,
+      "grad_norm": 0.012379745952785015,
+      "learning_rate": 0.0023306188925081434,
+      "loss": 1.3504,
+      "step": 411
+    },
+    {
+      "epoch": 0.9054945054945055,
+      "grad_norm": 0.023947101086378098,
+      "learning_rate": 0.0023281758957654723,
+      "loss": 1.4942,
+      "step": 412
+    },
+    {
+      "epoch": 0.9076923076923077,
+      "grad_norm": 0.017677150666713715,
+      "learning_rate": 0.0023257328990228016,
+      "loss": 1.3664,
+      "step": 413
+    },
+    {
+      "epoch": 0.9098901098901099,
+      "grad_norm": 0.01714276894927025,
+      "learning_rate": 0.0023232899022801305,
+      "loss": 1.3804,
+      "step": 414
+    },
+    {
+      "epoch": 0.9120879120879121,
+      "grad_norm": 0.011378195136785507,
+      "learning_rate": 0.0023208469055374594,
+      "loss": 1.3451,
+      "step": 415
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.022561779245734215,
+      "learning_rate": 0.0023184039087947883,
+      "loss": 1.4097,
+      "step": 416
+    },
+    {
+      "epoch": 0.9164835164835164,
+      "grad_norm": 0.012388916686177254,
+      "learning_rate": 0.002315960912052117,
+      "loss": 1.3248,
+      "step": 417
+    },
+    {
+      "epoch": 0.9186813186813186,
+      "grad_norm": 0.014059054665267467,
+      "learning_rate": 0.0023135179153094465,
+      "loss": 1.2953,
+      "step": 418
+    },
+    {
+      "epoch": 0.9208791208791208,
+      "grad_norm": 0.011394564993679523,
+      "learning_rate": 0.0023110749185667754,
+      "loss": 1.3313,
+      "step": 419
+    },
+    {
+      "epoch": 0.9230769230769231,
+      "grad_norm": 0.008186538703739643,
+      "learning_rate": 0.0023086319218241043,
+      "loss": 1.1804,
+      "step": 420
+    },
+    {
+      "epoch": 0.9252747252747253,
+      "grad_norm": 0.009855720214545727,
+      "learning_rate": 0.002306188925081433,
+      "loss": 1.1636,
+      "step": 421
+    },
+    {
+      "epoch": 0.9274725274725275,
+      "grad_norm": 0.013031002134084702,
+      "learning_rate": 0.0023037459283387625,
+      "loss": 1.2822,
+      "step": 422
+    },
+    {
+      "epoch": 0.9296703296703297,
+      "grad_norm": 0.013941977173089981,
+      "learning_rate": 0.0023013029315960914,
+      "loss": 1.2416,
+      "step": 423
+    },
+    {
+      "epoch": 0.9318681318681319,
+      "grad_norm": 0.01642168127000332,
+      "learning_rate": 0.0022988599348534203,
+      "loss": 1.3067,
+      "step": 424
+    },
+    {
+      "epoch": 0.9340659340659341,
+      "grad_norm": 0.014118148013949394,
+      "learning_rate": 0.002296416938110749,
+      "loss": 1.2555,
+      "step": 425
+    },
+    {
+      "epoch": 0.9362637362637363,
+      "grad_norm": 0.009554185904562473,
+      "learning_rate": 0.002293973941368078,
+      "loss": 1.2962,
+      "step": 426
+    },
+    {
+      "epoch": 0.9384615384615385,
+      "grad_norm": 0.014090403914451599,
+      "learning_rate": 0.0022915309446254074,
+      "loss": 1.4304,
+      "step": 427
+    },
+    {
+      "epoch": 0.9406593406593406,
+      "grad_norm": 0.011014874093234539,
+      "learning_rate": 0.0022890879478827363,
+      "loss": 1.2799,
+      "step": 428
+    },
+    {
+      "epoch": 0.9428571428571428,
+      "grad_norm": 0.015417378395795822,
+      "learning_rate": 0.002286644951140065,
+      "loss": 1.4022,
+      "step": 429
+    },
+    {
+      "epoch": 0.945054945054945,
+      "grad_norm": 0.00967811606824398,
+      "learning_rate": 0.002284201954397394,
+      "loss": 1.2082,
+      "step": 430
+    },
+    {
+      "epoch": 0.9472527472527472,
+      "grad_norm": 0.01139946561306715,
+      "learning_rate": 0.0022817589576547234,
+      "loss": 1.0332,
+      "step": 431
+    },
+    {
+      "epoch": 0.9494505494505494,
+      "grad_norm": 0.010407102294266224,
+      "learning_rate": 0.0022793159609120523,
+      "loss": 1.2744,
+      "step": 432
+    },
+    {
+      "epoch": 0.9516483516483516,
+      "grad_norm": 0.010214755311608315,
+      "learning_rate": 0.002276872964169381,
+      "loss": 1.0614,
+      "step": 433
+    },
+    {
+      "epoch": 0.9538461538461539,
+      "grad_norm": 0.01070279162377119,
+      "learning_rate": 0.00227442996742671,
+      "loss": 1.4112,
+      "step": 434
+    },
+    {
+      "epoch": 0.9560439560439561,
+      "grad_norm": 0.020081380382180214,
+      "learning_rate": 0.002271986970684039,
+      "loss": 1.1302,
+      "step": 435
+    },
+    {
+      "epoch": 0.9582417582417583,
+      "grad_norm": 0.010682860389351845,
+      "learning_rate": 0.0022695439739413683,
+      "loss": 1.3552,
+      "step": 436
+    },
+    {
+      "epoch": 0.9604395604395605,
+      "grad_norm": 0.013676096685230732,
+      "learning_rate": 0.002267100977198697,
+      "loss": 1.4859,
+      "step": 437
+    },
+    {
+      "epoch": 0.9626373626373627,
+      "grad_norm": 0.01134436670690775,
+      "learning_rate": 0.002264657980456026,
+      "loss": 1.2799,
+      "step": 438
+    },
+    {
+      "epoch": 0.9648351648351648,
+      "grad_norm": 0.01545412465929985,
+      "learning_rate": 0.002262214983713355,
+      "loss": 1.4259,
+      "step": 439
+    },
+    {
+      "epoch": 0.967032967032967,
+      "grad_norm": 0.06949850171804428,
+      "learning_rate": 0.0022597719869706843,
+      "loss": 1.3424,
+      "step": 440
+    },
+    {
+      "epoch": 0.9692307692307692,
+      "grad_norm": 0.012467940337955952,
+      "learning_rate": 0.002257328990228013,
+      "loss": 1.4349,
+      "step": 441
+    },
+    {
+      "epoch": 0.9714285714285714,
+      "grad_norm": 0.029872827231884003,
+      "learning_rate": 0.002254885993485342,
+      "loss": 1.2881,
+      "step": 442
+    },
+    {
+      "epoch": 0.9736263736263736,
+      "grad_norm": 0.008445368148386478,
+      "learning_rate": 0.002252442996742671,
+      "loss": 1.098,
+      "step": 443
+    },
+    {
+      "epoch": 0.9758241758241758,
+      "grad_norm": 0.013956134207546711,
+      "learning_rate": 0.0022500000000000003,
+      "loss": 1.3004,
+      "step": 444
+    },
+    {
+      "epoch": 0.978021978021978,
+      "grad_norm": 0.008711514994502068,
+      "learning_rate": 0.002247557003257329,
+      "loss": 1.2507,
+      "step": 445
+    },
+    {
+      "epoch": 0.9802197802197802,
+      "grad_norm": 0.01162966713309288,
+      "learning_rate": 0.002245114006514658,
+      "loss": 1.2834,
+      "step": 446
+    },
+    {
+      "epoch": 0.9824175824175824,
+      "grad_norm": 0.012029342353343964,
+      "learning_rate": 0.002242671009771987,
+      "loss": 1.216,
+      "step": 447
+    },
+    {
+      "epoch": 0.9846153846153847,
+      "grad_norm": 0.014662378467619419,
+      "learning_rate": 0.002240228013029316,
+      "loss": 1.4296,
+      "step": 448
+    },
+    {
+      "epoch": 0.9868131868131869,
+      "grad_norm": 0.01343568041920662,
+      "learning_rate": 0.002237785016286645,
+      "loss": 1.3317,
+      "step": 449
+    },
+    {
+      "epoch": 0.989010989010989,
+      "grad_norm": 0.01101224310696125,
+      "learning_rate": 0.002235342019543974,
+      "loss": 1.2614,
+      "step": 450
+    },
+    {
+      "epoch": 0.9912087912087912,
+      "grad_norm": 0.00980074517428875,
+      "learning_rate": 0.002232899022801303,
+      "loss": 1.3,
+      "step": 451
+    },
+    {
+      "epoch": 0.9934065934065934,
+      "grad_norm": 0.008838055655360222,
+      "learning_rate": 0.002230456026058632,
+      "loss": 1.4337,
+      "step": 452
+    },
+    {
+      "epoch": 0.9956043956043956,
+      "grad_norm": 0.01873493567109108,
+      "learning_rate": 0.002228013029315961,
+      "loss": 1.2978,
+      "step": 453
+    },
+    {
+      "epoch": 0.9978021978021978,
+      "grad_norm": 0.010824136435985565,
+      "learning_rate": 0.00222557003257329,
+      "loss": 1.3136,
+      "step": 454
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.014136227779090405,
+      "learning_rate": 0.002223127035830619,
+      "loss": 1.2696,
+      "step": 455
+    },
+    {
+      "epoch": 1.0021978021978022,
+      "grad_norm": 0.01353212259709835,
+      "learning_rate": 0.002220684039087948,
+      "loss": 1.1014,
+      "step": 456
+    },
+    {
+      "epoch": 1.0043956043956044,
+      "grad_norm": 0.01069607399404049,
+      "learning_rate": 0.0022182410423452767,
+      "loss": 1.2951,
+      "step": 457
+    },
+    {
+      "epoch": 1.0065934065934066,
+      "grad_norm": 0.015369105152785778,
+      "learning_rate": 0.002215798045602606,
+      "loss": 1.2353,
+      "step": 458
+    },
+    {
+      "epoch": 1.0087912087912088,
+      "grad_norm": 0.11915262043476105,
+      "learning_rate": 0.002213355048859935,
+      "loss": 1.3696,
+      "step": 459
+    },
+    {
+      "epoch": 1.010989010989011,
+      "grad_norm": 0.01160374190658331,
+      "learning_rate": 0.002210912052117264,
+      "loss": 1.392,
+      "step": 460
+    },
+    {
+      "epoch": 1.0131868131868131,
+      "grad_norm": 0.008986646309494972,
+      "learning_rate": 0.0022084690553745927,
+      "loss": 1.3128,
+      "step": 461
+    },
+    {
+      "epoch": 1.0153846153846153,
+      "grad_norm": 0.008956193923950195,
+      "learning_rate": 0.002206026058631922,
+      "loss": 1.3793,
+      "step": 462
+    },
+    {
+      "epoch": 1.0175824175824175,
+      "grad_norm": 0.014813029207289219,
+      "learning_rate": 0.002203583061889251,
+      "loss": 1.3163,
+      "step": 463
+    },
+    {
+      "epoch": 1.0197802197802197,
+      "grad_norm": 0.020007099956274033,
+      "learning_rate": 0.00220114006514658,
+      "loss": 1.3092,
+      "step": 464
+    },
+    {
+      "epoch": 1.021978021978022,
+      "grad_norm": 0.011827349662780762,
+      "learning_rate": 0.0021986970684039087,
+      "loss": 1.2834,
+      "step": 465
+    },
+    {
+      "epoch": 1.024175824175824,
+      "grad_norm": 0.009880859404802322,
+      "learning_rate": 0.0021962540716612376,
+      "loss": 1.2281,
+      "step": 466
+    },
+    {
+      "epoch": 1.0263736263736263,
+      "grad_norm": 0.010755722410976887,
+      "learning_rate": 0.002193811074918567,
+      "loss": 1.2618,
+      "step": 467
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.015370133332908154,
+      "learning_rate": 0.002191368078175896,
+      "loss": 1.2764,
+      "step": 468
+    },
+    {
+      "epoch": 1.0307692307692307,
+      "grad_norm": 0.011006797663867474,
+      "learning_rate": 0.0021889250814332247,
+      "loss": 1.167,
+      "step": 469
+    },
+    {
+      "epoch": 1.032967032967033,
+      "grad_norm": 0.008608494885265827,
+      "learning_rate": 0.0021864820846905536,
+      "loss": 1.2778,
+      "step": 470
+    },
+    {
+      "epoch": 1.0351648351648353,
+      "grad_norm": 0.009305492043495178,
+      "learning_rate": 0.002184039087947883,
+      "loss": 1.2399,
+      "step": 471
+    },
+    {
+      "epoch": 1.0373626373626375,
+      "grad_norm": 0.010387458838522434,
+      "learning_rate": 0.002181596091205212,
+      "loss": 1.5028,
+      "step": 472
+    },
+    {
+      "epoch": 1.0395604395604396,
+      "grad_norm": 0.02264990098774433,
+      "learning_rate": 0.0021791530944625407,
+      "loss": 1.1223,
+      "step": 473
+    },
+    {
+      "epoch": 1.0417582417582418,
+      "grad_norm": 0.013376729562878609,
+      "learning_rate": 0.0021767100977198696,
+      "loss": 1.2852,
+      "step": 474
+    },
+    {
+      "epoch": 1.043956043956044,
+      "grad_norm": 0.008080637082457542,
+      "learning_rate": 0.0021742671009771985,
+      "loss": 1.3472,
+      "step": 475
+    },
+    {
+      "epoch": 1.0461538461538462,
+      "grad_norm": 0.006970232352614403,
+      "learning_rate": 0.002171824104234528,
+      "loss": 1.3253,
+      "step": 476
+    },
+    {
+      "epoch": 1.0483516483516484,
+      "grad_norm": 0.010948806069791317,
+      "learning_rate": 0.0021693811074918567,
+      "loss": 1.393,
+      "step": 477
+    },
+    {
+      "epoch": 1.0505494505494506,
+      "grad_norm": 0.014537020586431026,
+      "learning_rate": 0.0021669381107491856,
+      "loss": 1.2401,
+      "step": 478
+    },
+    {
+      "epoch": 1.0527472527472528,
+      "grad_norm": 0.014503644779324532,
+      "learning_rate": 0.0021644951140065145,
+      "loss": 1.3407,
+      "step": 479
+    },
+    {
+      "epoch": 1.054945054945055,
+      "grad_norm": 0.01736668311059475,
+      "learning_rate": 0.002162052117263844,
+      "loss": 1.464,
+      "step": 480
+    },
+    {
+      "epoch": 1.0571428571428572,
+      "grad_norm": 0.00850729551166296,
+      "learning_rate": 0.0021596091205211727,
+      "loss": 1.1544,
+      "step": 481
+    },
+    {
+      "epoch": 1.0593406593406594,
+      "grad_norm": 0.011973443441092968,
+      "learning_rate": 0.0021571661237785016,
+      "loss": 1.1744,
+      "step": 482
+    },
+    {
+      "epoch": 1.0615384615384615,
+      "grad_norm": 0.007684681564569473,
+      "learning_rate": 0.0021547231270358305,
+      "loss": 1.1685,
+      "step": 483
+    },
+    {
+      "epoch": 1.0637362637362637,
+      "grad_norm": 0.02417232096195221,
+      "learning_rate": 0.0021522801302931594,
+      "loss": 1.2117,
+      "step": 484
+    },
+    {
+      "epoch": 1.065934065934066,
+      "grad_norm": 0.010968627408146858,
+      "learning_rate": 0.0021498371335504887,
+      "loss": 1.4078,
+      "step": 485
+    },
+    {
+      "epoch": 1.0681318681318681,
+      "grad_norm": 0.016391798853874207,
+      "learning_rate": 0.0021473941368078176,
+      "loss": 1.2833,
+      "step": 486
+    },
+    {
+      "epoch": 1.0703296703296703,
+      "grad_norm": 0.05708368495106697,
+      "learning_rate": 0.0021449511400651465,
+      "loss": 1.091,
+      "step": 487
+    },
+    {
+      "epoch": 1.0725274725274725,
+      "grad_norm": 0.01210484467446804,
+      "learning_rate": 0.0021425081433224754,
+      "loss": 1.12,
+      "step": 488
+    },
+    {
+      "epoch": 1.0747252747252747,
+      "grad_norm": 0.009857972152531147,
+      "learning_rate": 0.0021400651465798047,
+      "loss": 1.3086,
+      "step": 489
+    },
+    {
+      "epoch": 1.0769230769230769,
+      "grad_norm": 0.00905830878764391,
+      "learning_rate": 0.0021376221498371336,
+      "loss": 1.2364,
+      "step": 490
+    },
+    {
+      "epoch": 1.079120879120879,
+      "grad_norm": 0.02259766310453415,
+      "learning_rate": 0.0021351791530944625,
+      "loss": 1.3985,
+      "step": 491
+    },
+    {
+      "epoch": 1.0813186813186813,
+      "grad_norm": 0.015585063956677914,
+      "learning_rate": 0.0021327361563517914,
+      "loss": 1.4456,
+      "step": 492
+    },
+    {
+      "epoch": 1.0835164835164834,
+      "grad_norm": 0.01756209135055542,
+      "learning_rate": 0.0021302931596091203,
+      "loss": 1.4516,
+      "step": 493
+    },
+    {
+      "epoch": 1.0857142857142856,
+      "grad_norm": 0.01778987981379032,
+      "learning_rate": 0.0021278501628664496,
+      "loss": 1.1788,
+      "step": 494
+    },
+    {
+      "epoch": 1.0879120879120878,
+      "grad_norm": 0.017591632902622223,
+      "learning_rate": 0.0021254071661237785,
+      "loss": 1.2774,
+      "step": 495
+    },
+    {
+      "epoch": 1.09010989010989,
+      "grad_norm": 0.013083560392260551,
+      "learning_rate": 0.0021229641693811074,
+      "loss": 1.4037,
+      "step": 496
+    },
+    {
+      "epoch": 1.0923076923076924,
+      "grad_norm": 0.01581994816660881,
+      "learning_rate": 0.0021205211726384363,
+      "loss": 1.353,
+      "step": 497
+    },
+    {
+      "epoch": 1.0945054945054946,
+      "grad_norm": 0.022908881306648254,
+      "learning_rate": 0.0021180781758957656,
+      "loss": 1.2347,
+      "step": 498
+    },
+    {
+      "epoch": 1.0967032967032968,
+      "grad_norm": 0.015620234422385693,
+      "learning_rate": 0.0021156351791530945,
+      "loss": 1.2598,
+      "step": 499
+    },
+    {
+      "epoch": 1.098901098901099,
+      "grad_norm": 0.012264670804142952,
+      "learning_rate": 0.0021131921824104234,
+      "loss": 1.3748,
+      "step": 500
+    },
+    {
+      "epoch": 1.1010989010989012,
+      "grad_norm": 0.01436922512948513,
+      "learning_rate": 0.0021107491856677523,
+      "loss": 1.2022,
+      "step": 501
+    },
+    {
+      "epoch": 1.1032967032967034,
+      "grad_norm": 0.013385582715272903,
+      "learning_rate": 0.002108306188925081,
+      "loss": 1.3147,
+      "step": 502
+    },
+    {
+      "epoch": 1.1054945054945056,
+      "grad_norm": 0.008359020575881004,
+      "learning_rate": 0.0021058631921824105,
+      "loss": 1.3678,
+      "step": 503
+    },
+    {
+      "epoch": 1.1076923076923078,
+      "grad_norm": 0.0090394401922822,
+      "learning_rate": 0.0021034201954397394,
+      "loss": 1.2797,
+      "step": 504
+    },
+    {
+      "epoch": 1.10989010989011,
+      "grad_norm": 0.010517923161387444,
+      "learning_rate": 0.0021009771986970683,
+      "loss": 1.296,
+      "step": 505
+    },
+    {
+      "epoch": 1.1120879120879121,
+      "grad_norm": 0.01679222472012043,
+      "learning_rate": 0.002098534201954397,
+      "loss": 1.4157,
+      "step": 506
+    },
+    {
+      "epoch": 1.1142857142857143,
+      "grad_norm": 0.017458060756325722,
+      "learning_rate": 0.0020960912052117265,
+      "loss": 1.3804,
+      "step": 507
+    },
+    {
+      "epoch": 1.1164835164835165,
+      "grad_norm": 0.007606922648847103,
+      "learning_rate": 0.0020936482084690554,
+      "loss": 1.1815,
+      "step": 508
+    },
+    {
+      "epoch": 1.1186813186813187,
+      "grad_norm": 0.014373927377164364,
+      "learning_rate": 0.0020912052117263843,
+      "loss": 1.2441,
+      "step": 509
+    },
+    {
+      "epoch": 1.120879120879121,
+      "grad_norm": 0.011267592199146748,
+      "learning_rate": 0.002088762214983713,
+      "loss": 1.3663,
+      "step": 510
+    },
+    {
+      "epoch": 1.123076923076923,
+      "grad_norm": 0.01143614761531353,
+      "learning_rate": 0.0020863192182410425,
+      "loss": 1.587,
+      "step": 511
+    },
+    {
+      "epoch": 1.1252747252747253,
+      "grad_norm": 0.010744009166955948,
+      "learning_rate": 0.0020838762214983714,
+      "loss": 1.289,
+      "step": 512
+    },
+    {
+      "epoch": 1.1274725274725275,
+      "grad_norm": 0.00835670530796051,
+      "learning_rate": 0.0020814332247557003,
+      "loss": 1.3792,
+      "step": 513
+    },
+    {
+      "epoch": 1.1296703296703297,
+      "grad_norm": 0.01791512221097946,
+      "learning_rate": 0.002078990228013029,
+      "loss": 1.178,
+      "step": 514
+    },
+    {
+      "epoch": 1.1318681318681318,
+      "grad_norm": 0.012324312701821327,
+      "learning_rate": 0.002076547231270358,
+      "loss": 1.3504,
+      "step": 515
+    },
+    {
+      "epoch": 1.134065934065934,
+      "grad_norm": 0.03248134255409241,
+      "learning_rate": 0.0020741042345276874,
+      "loss": 1.2569,
+      "step": 516
+    },
+    {
+      "epoch": 1.1362637362637362,
+      "grad_norm": 0.011701959185302258,
+      "learning_rate": 0.0020716612377850163,
+      "loss": 1.157,
+      "step": 517
+    },
+    {
+      "epoch": 1.1384615384615384,
+      "grad_norm": 0.012444762513041496,
+      "learning_rate": 0.002069218241042345,
+      "loss": 1.3604,
+      "step": 518
+    },
+    {
+      "epoch": 1.1406593406593406,
+      "grad_norm": 0.010032200254499912,
+      "learning_rate": 0.002066775244299674,
+      "loss": 1.2029,
+      "step": 519
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.01350658480077982,
+      "learning_rate": 0.0020643322475570034,
+      "loss": 1.2305,
+      "step": 520
+    },
+    {
+      "epoch": 1.145054945054945,
+      "grad_norm": 0.010027634911239147,
+      "learning_rate": 0.0020618892508143323,
+      "loss": 1.5027,
+      "step": 521
+    },
+    {
+      "epoch": 1.1472527472527472,
+      "grad_norm": 0.017331158742308617,
+      "learning_rate": 0.002059446254071661,
+      "loss": 1.4783,
+      "step": 522
+    },
+    {
+      "epoch": 1.1494505494505494,
+      "grad_norm": 0.015214362181723118,
+      "learning_rate": 0.00205700325732899,
+      "loss": 1.1963,
+      "step": 523
+    },
+    {
+      "epoch": 1.1516483516483516,
+      "grad_norm": 0.010057060047984123,
+      "learning_rate": 0.002054560260586319,
+      "loss": 1.3053,
+      "step": 524
+    },
+    {
+      "epoch": 1.1538461538461537,
+      "grad_norm": 0.012808633968234062,
+      "learning_rate": 0.0020521172638436483,
+      "loss": 1.1348,
+      "step": 525
+    },
+    {
+      "epoch": 1.156043956043956,
+      "grad_norm": 0.01069604605436325,
+      "learning_rate": 0.002049674267100977,
+      "loss": 1.2718,
+      "step": 526
+    },
+    {
+      "epoch": 1.1582417582417581,
+      "grad_norm": 0.009505991823971272,
+      "learning_rate": 0.002047231270358306,
+      "loss": 1.2312,
+      "step": 527
+    },
+    {
+      "epoch": 1.1604395604395605,
+      "grad_norm": 0.03593682125210762,
+      "learning_rate": 0.002044788273615635,
+      "loss": 1.3583,
+      "step": 528
+    },
+    {
+      "epoch": 1.1626373626373627,
+      "grad_norm": 0.012589525431394577,
+      "learning_rate": 0.0020423452768729643,
+      "loss": 1.3769,
+      "step": 529
+    },
+    {
+      "epoch": 1.164835164835165,
+      "grad_norm": 0.01110800076276064,
+      "learning_rate": 0.002039902280130293,
+      "loss": 1.2771,
+      "step": 530
+    },
+    {
+      "epoch": 1.167032967032967,
+      "grad_norm": 0.01574832759797573,
+      "learning_rate": 0.002037459283387622,
+      "loss": 1.28,
+      "step": 531
+    },
+    {
+      "epoch": 1.1692307692307693,
+      "grad_norm": 0.012825988233089447,
+      "learning_rate": 0.002035016286644951,
+      "loss": 1.5246,
+      "step": 532
+    },
+    {
+      "epoch": 1.1714285714285715,
+      "grad_norm": 0.012039689347147942,
+      "learning_rate": 0.00203257328990228,
+      "loss": 1.0989,
+      "step": 533
+    },
+    {
+      "epoch": 1.1736263736263737,
+      "grad_norm": 0.028587760403752327,
+      "learning_rate": 0.0020301302931596092,
+      "loss": 1.5238,
+      "step": 534
+    },
+    {
+      "epoch": 1.1758241758241759,
+      "grad_norm": 0.01644471473991871,
+      "learning_rate": 0.002027687296416938,
+      "loss": 1.3584,
+      "step": 535
+    },
+    {
+      "epoch": 1.178021978021978,
+      "grad_norm": 0.01879804953932762,
+      "learning_rate": 0.002025244299674267,
+      "loss": 1.2133,
+      "step": 536
+    },
+    {
+      "epoch": 1.1802197802197802,
+      "grad_norm": 0.008065084926784039,
+      "learning_rate": 0.002022801302931596,
+      "loss": 1.3137,
+      "step": 537
+    },
+    {
+      "epoch": 1.1824175824175824,
+      "grad_norm": 0.012157510034739971,
+      "learning_rate": 0.0020203583061889252,
+      "loss": 1.2536,
+      "step": 538
+    },
+    {
+      "epoch": 1.1846153846153846,
+      "grad_norm": 0.042824432253837585,
+      "learning_rate": 0.002017915309446254,
+      "loss": 1.5703,
+      "step": 539
+    },
+    {
+      "epoch": 1.1868131868131868,
+      "grad_norm": 0.016401996836066246,
+      "learning_rate": 0.002015472312703583,
+      "loss": 1.3031,
+      "step": 540
+    },
+    {
+      "epoch": 1.189010989010989,
+      "grad_norm": 0.012873569503426552,
+      "learning_rate": 0.002013029315960912,
+      "loss": 1.1816,
+      "step": 541
+    },
+    {
+      "epoch": 1.1912087912087912,
+      "grad_norm": 0.010174766182899475,
+      "learning_rate": 0.002010586319218241,
+      "loss": 1.2505,
+      "step": 542
+    },
+    {
+      "epoch": 1.1934065934065934,
+      "grad_norm": 0.009019955061376095,
+      "learning_rate": 0.00200814332247557,
+      "loss": 1.2286,
+      "step": 543
+    },
+    {
+      "epoch": 1.1956043956043956,
+      "grad_norm": 0.01039096713066101,
+      "learning_rate": 0.002005700325732899,
+      "loss": 1.3927,
+      "step": 544
+    },
+    {
+      "epoch": 1.1978021978021978,
+      "grad_norm": 0.02589944191277027,
+      "learning_rate": 0.002003257328990228,
+      "loss": 1.287,
+      "step": 545
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.013341677375137806,
+      "learning_rate": 0.002000814332247557,
+      "loss": 1.2736,
+      "step": 546
+    },
+    {
+      "epoch": 1.2021978021978021,
+      "grad_norm": 0.014910456724464893,
+      "learning_rate": 0.001998371335504886,
+      "loss": 1.3385,
+      "step": 547
+    },
+    {
+      "epoch": 1.2043956043956043,
+      "grad_norm": 0.012872068211436272,
+      "learning_rate": 0.001995928338762215,
+      "loss": 1.3101,
+      "step": 548
+    },
+    {
+      "epoch": 1.2065934065934065,
+      "grad_norm": 0.008229200728237629,
+      "learning_rate": 0.001993485342019544,
+      "loss": 1.255,
+      "step": 549
+    },
+    {
+      "epoch": 1.2087912087912087,
+      "grad_norm": 0.010236970148980618,
+      "learning_rate": 0.001991042345276873,
+      "loss": 1.0542,
+      "step": 550
+    },
+    {
+      "epoch": 1.210989010989011,
+      "grad_norm": 0.012508456595242023,
+      "learning_rate": 0.0019885993485342017,
+      "loss": 1.2771,
+      "step": 551
+    },
+    {
+      "epoch": 1.213186813186813,
+      "grad_norm": 0.00906662829220295,
+      "learning_rate": 0.001986156351791531,
+      "loss": 1.3153,
+      "step": 552
+    },
+    {
+      "epoch": 1.2153846153846155,
+      "grad_norm": 0.008852269500494003,
+      "learning_rate": 0.00198371335504886,
+      "loss": 1.1993,
+      "step": 553
+    },
+    {
+      "epoch": 1.2175824175824177,
+      "grad_norm": 0.008736698888242245,
+      "learning_rate": 0.001981270358306189,
+      "loss": 1.2479,
+      "step": 554
+    },
+    {
+      "epoch": 1.2197802197802199,
+      "grad_norm": 0.012984815053641796,
+      "learning_rate": 0.0019788273615635177,
+      "loss": 1.5013,
+      "step": 555
+    },
+    {
+      "epoch": 1.221978021978022,
+      "grad_norm": 0.009770010598003864,
+      "learning_rate": 0.001976384364820847,
+      "loss": 1.2782,
+      "step": 556
+    },
+    {
+      "epoch": 1.2241758241758243,
+      "grad_norm": 0.009296323172748089,
+      "learning_rate": 0.001973941368078176,
+      "loss": 1.2675,
+      "step": 557
+    },
+    {
+      "epoch": 1.2263736263736265,
+      "grad_norm": 0.00993712805211544,
+      "learning_rate": 0.001971498371335505,
+      "loss": 1.0881,
+      "step": 558
+    },
+    {
+      "epoch": 1.2285714285714286,
+      "grad_norm": 0.009717991575598717,
+      "learning_rate": 0.0019690553745928337,
+      "loss": 1.1752,
+      "step": 559
+    },
+    {
+      "epoch": 1.2307692307692308,
+      "grad_norm": 0.012592695653438568,
+      "learning_rate": 0.0019666123778501626,
+      "loss": 1.2719,
+      "step": 560
+    },
+    {
+      "epoch": 1.232967032967033,
+      "grad_norm": 0.011870156973600388,
+      "learning_rate": 0.001964169381107492,
+      "loss": 1.2585,
+      "step": 561
+    },
+    {
+      "epoch": 1.2351648351648352,
+      "grad_norm": 0.011108724400401115,
+      "learning_rate": 0.001961726384364821,
+      "loss": 1.3542,
+      "step": 562
+    },
+    {
+      "epoch": 1.2373626373626374,
+      "grad_norm": 0.009791416116058826,
+      "learning_rate": 0.0019592833876221497,
+      "loss": 1.2579,
+      "step": 563
+    },
+    {
+      "epoch": 1.2395604395604396,
+      "grad_norm": 0.015620584599673748,
+      "learning_rate": 0.0019568403908794786,
+      "loss": 1.5938,
+      "step": 564
+    },
+    {
+      "epoch": 1.2417582417582418,
+      "grad_norm": 0.011419607326388359,
+      "learning_rate": 0.001954397394136808,
+      "loss": 1.2828,
+      "step": 565
+    },
+    {
+      "epoch": 1.243956043956044,
+      "grad_norm": 0.027966516092419624,
+      "learning_rate": 0.0019519543973941368,
+      "loss": 1.2179,
+      "step": 566
+    },
+    {
+      "epoch": 1.2461538461538462,
+      "grad_norm": 0.011032767593860626,
+      "learning_rate": 0.0019495114006514657,
+      "loss": 1.3535,
+      "step": 567
+    },
+    {
+      "epoch": 1.2483516483516484,
+      "grad_norm": 0.009936105459928513,
+      "learning_rate": 0.0019470684039087948,
+      "loss": 1.1959,
+      "step": 568
+    },
+    {
+      "epoch": 1.2505494505494505,
+      "grad_norm": 0.011602498590946198,
+      "learning_rate": 0.0019446254071661237,
+      "loss": 1.2917,
+      "step": 569
+    },
+    {
+      "epoch": 1.2527472527472527,
+      "grad_norm": 0.00922809075564146,
+      "learning_rate": 0.0019421824104234526,
+      "loss": 1.3799,
+      "step": 570
+    },
+    {
+      "epoch": 1.254945054945055,
+      "grad_norm": 0.012401960790157318,
+      "learning_rate": 0.0019397394136807817,
+      "loss": 1.3032,
+      "step": 571
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.009585228748619556,
+      "learning_rate": 0.0019372964169381106,
+      "loss": 1.1589,
+      "step": 572
+    },
+    {
+      "epoch": 1.2593406593406593,
+      "grad_norm": 0.01102573610842228,
+      "learning_rate": 0.0019348534201954397,
+      "loss": 1.4175,
+      "step": 573
+    },
+    {
+      "epoch": 1.2615384615384615,
+      "grad_norm": 0.01276457216590643,
+      "learning_rate": 0.0019324104234527686,
+      "loss": 1.2553,
+      "step": 574
+    },
+    {
+      "epoch": 1.2637362637362637,
+      "grad_norm": 0.01171040814369917,
+      "learning_rate": 0.0019299674267100977,
+      "loss": 1.3639,
+      "step": 575
+    },
+    {
+      "epoch": 1.2659340659340659,
+      "grad_norm": 0.015174838714301586,
+      "learning_rate": 0.0019275244299674266,
+      "loss": 1.2978,
+      "step": 576
+    },
+    {
+      "epoch": 1.268131868131868,
+      "grad_norm": 0.007973062805831432,
+      "learning_rate": 0.0019250814332247557,
+      "loss": 1.2413,
+      "step": 577
+    },
+    {
+      "epoch": 1.2703296703296703,
+      "grad_norm": 0.013420450501143932,
+      "learning_rate": 0.0019226384364820846,
+      "loss": 1.3247,
+      "step": 578
+    },
+    {
+      "epoch": 1.2725274725274724,
+      "grad_norm": 0.00842191744595766,
+      "learning_rate": 0.0019201954397394135,
+      "loss": 1.0303,
+      "step": 579
+    },
+    {
+      "epoch": 1.2747252747252746,
+      "grad_norm": 0.010580204427242279,
+      "learning_rate": 0.0019177524429967426,
+      "loss": 1.2257,
+      "step": 580
+    },
+    {
+      "epoch": 1.2769230769230768,
+      "grad_norm": 0.010875397361814976,
+      "learning_rate": 0.0019153094462540715,
+      "loss": 1.0917,
+      "step": 581
+    },
+    {
+      "epoch": 1.279120879120879,
+      "grad_norm": 0.01389323454350233,
+      "learning_rate": 0.0019128664495114006,
+      "loss": 1.3981,
+      "step": 582
+    },
+    {
+      "epoch": 1.2813186813186812,
+      "grad_norm": 0.011721659451723099,
+      "learning_rate": 0.0019104234527687295,
+      "loss": 1.2346,
+      "step": 583
+    },
+    {
+      "epoch": 1.2835164835164834,
+      "grad_norm": 0.017351191490888596,
+      "learning_rate": 0.0019079804560260586,
+      "loss": 1.4502,
+      "step": 584
+    },
+    {
+      "epoch": 1.2857142857142856,
+      "grad_norm": 0.013639170676469803,
+      "learning_rate": 0.0019055374592833875,
+      "loss": 1.3452,
+      "step": 585
+    },
+    {
+      "epoch": 1.2879120879120878,
+      "grad_norm": 0.012638512998819351,
+      "learning_rate": 0.0019030944625407166,
+      "loss": 1.2404,
+      "step": 586
+    },
+    {
+      "epoch": 1.2901098901098902,
+      "grad_norm": 0.01941491849720478,
+      "learning_rate": 0.0019006514657980455,
+      "loss": 1.4858,
+      "step": 587
+    },
+    {
+      "epoch": 1.2923076923076924,
+      "grad_norm": 0.008536739274859428,
+      "learning_rate": 0.0018982084690553744,
+      "loss": 1.1853,
+      "step": 588
+    },
+    {
+      "epoch": 1.2945054945054946,
+      "grad_norm": 0.009980559349060059,
+      "learning_rate": 0.0018957654723127035,
+      "loss": 1.2789,
+      "step": 589
+    },
+    {
+      "epoch": 1.2967032967032968,
+      "grad_norm": 0.011733613908290863,
+      "learning_rate": 0.0018933224755700324,
+      "loss": 1.3,
+      "step": 590
+    },
+    {
+      "epoch": 1.298901098901099,
+      "grad_norm": 0.009111632592976093,
+      "learning_rate": 0.0018908794788273615,
+      "loss": 1.4136,
+      "step": 591
+    },
+    {
+      "epoch": 1.3010989010989011,
+      "grad_norm": 0.008808271028101444,
+      "learning_rate": 0.0018884364820846904,
+      "loss": 1.4198,
+      "step": 592
+    },
+    {
+      "epoch": 1.3032967032967033,
+      "grad_norm": 0.010935761965811253,
+      "learning_rate": 0.0018859934853420195,
+      "loss": 1.2683,
+      "step": 593
+    },
+    {
+      "epoch": 1.3054945054945055,
+      "grad_norm": 0.035847339779138565,
+      "learning_rate": 0.0018835504885993484,
+      "loss": 1.4785,
+      "step": 594
+    },
+    {
+      "epoch": 1.3076923076923077,
+      "grad_norm": 0.009762615896761417,
+      "learning_rate": 0.0018811074918566775,
+      "loss": 1.2748,
+      "step": 595
+    },
+    {
+      "epoch": 1.30989010989011,
+      "grad_norm": 0.012693176977336407,
+      "learning_rate": 0.0018786644951140064,
+      "loss": 1.0694,
+      "step": 596
+    },
+    {
+      "epoch": 1.312087912087912,
+      "grad_norm": 0.015002069994807243,
+      "learning_rate": 0.0018762214983713355,
+      "loss": 1.3531,
+      "step": 597
+    },
+    {
+      "epoch": 1.3142857142857143,
+      "grad_norm": 0.00767421955242753,
+      "learning_rate": 0.0018737785016286646,
+      "loss": 1.0878,
+      "step": 598
+    },
+    {
+      "epoch": 1.3164835164835165,
+      "grad_norm": 0.010209274478256702,
+      "learning_rate": 0.0018713355048859937,
+      "loss": 1.3334,
+      "step": 599
+    },
+    {
+      "epoch": 1.3186813186813187,
+      "grad_norm": 0.02255304530262947,
+      "learning_rate": 0.0018688925081433226,
+      "loss": 1.3849,
+      "step": 600
+    },
+    {
+      "epoch": 1.3208791208791208,
+      "grad_norm": 0.009391605854034424,
+      "learning_rate": 0.0018664495114006517,
+      "loss": 1.2814,
+      "step": 601
+    },
+    {
+      "epoch": 1.323076923076923,
+      "grad_norm": 0.010748128406703472,
+      "learning_rate": 0.0018640065146579806,
+      "loss": 1.169,
+      "step": 602
+    },
+    {
+      "epoch": 1.3252747252747252,
+      "grad_norm": 0.009680286049842834,
+      "learning_rate": 0.0018615635179153097,
+      "loss": 1.2152,
+      "step": 603
+    },
+    {
+      "epoch": 1.3274725274725274,
+      "grad_norm": 0.009965418837964535,
+      "learning_rate": 0.0018591205211726386,
+      "loss": 1.4544,
+      "step": 604
+    },
+    {
+      "epoch": 1.3296703296703296,
+      "grad_norm": 0.010849074460566044,
+      "learning_rate": 0.0018566775244299677,
+      "loss": 1.3177,
+      "step": 605
+    },
+    {
+      "epoch": 1.3318681318681318,
+      "grad_norm": 0.00995631329715252,
+      "learning_rate": 0.0018542345276872966,
+      "loss": 1.2398,
+      "step": 606
+    },
+    {
+      "epoch": 1.334065934065934,
+      "grad_norm": 0.009554999880492687,
+      "learning_rate": 0.0018517915309446257,
+      "loss": 1.334,
+      "step": 607
+    },
+    {
+      "epoch": 1.3362637362637364,
+      "grad_norm": 0.02414814382791519,
+      "learning_rate": 0.0018493485342019546,
+      "loss": 1.1903,
+      "step": 608
+    },
+    {
+      "epoch": 1.3384615384615386,
+      "grad_norm": 0.011140548624098301,
+      "learning_rate": 0.0018469055374592835,
+      "loss": 1.3571,
+      "step": 609
+    },
+    {
+      "epoch": 1.3406593406593408,
+      "grad_norm": 0.009847228415310383,
+      "learning_rate": 0.0018444625407166126,
+      "loss": 1.2523,
+      "step": 610
+    },
+    {
+      "epoch": 1.342857142857143,
+      "grad_norm": 0.010612218640744686,
+      "learning_rate": 0.0018420195439739415,
+      "loss": 1.2825,
+      "step": 611
+    },
+    {
+      "epoch": 1.3450549450549452,
+      "grad_norm": 0.01574314571917057,
+      "learning_rate": 0.0018395765472312706,
+      "loss": 1.1444,
+      "step": 612
+    },
+    {
+      "epoch": 1.3472527472527474,
+      "grad_norm": 0.013081908226013184,
+      "learning_rate": 0.0018371335504885995,
+      "loss": 1.1687,
+      "step": 613
+    },
+    {
+      "epoch": 1.3494505494505495,
+      "grad_norm": 0.013602767139673233,
+      "learning_rate": 0.0018346905537459286,
+      "loss": 1.3765,
+      "step": 614
+    },
+    {
+      "epoch": 1.3516483516483517,
+      "grad_norm": 0.012633330188691616,
+      "learning_rate": 0.0018322475570032575,
+      "loss": 1.0674,
+      "step": 615
+    },
+    {
+      "epoch": 1.353846153846154,
+      "grad_norm": 0.015483934432268143,
+      "learning_rate": 0.0018298045602605866,
+      "loss": 1.0577,
+      "step": 616
+    },
+    {
+      "epoch": 1.3560439560439561,
+      "grad_norm": 0.05298389494419098,
+      "learning_rate": 0.0018273615635179155,
+      "loss": 1.1539,
+      "step": 617
+    },
+    {
+      "epoch": 1.3582417582417583,
+      "grad_norm": 0.04067712277173996,
+      "learning_rate": 0.0018249185667752444,
+      "loss": 1.2335,
+      "step": 618
+    },
+    {
+      "epoch": 1.3604395604395605,
+      "grad_norm": 0.010437838733196259,
+      "learning_rate": 0.0018224755700325735,
+      "loss": 1.087,
+      "step": 619
+    },
+    {
+      "epoch": 1.3626373626373627,
+      "grad_norm": 0.011761399917304516,
+      "learning_rate": 0.0018200325732899024,
+      "loss": 1.2137,
+      "step": 620
+    },
+    {
+      "epoch": 1.3648351648351649,
+      "grad_norm": 0.011025581508874893,
+      "learning_rate": 0.0018175895765472315,
+      "loss": 1.3531,
+      "step": 621
+    },
+    {
+      "epoch": 1.367032967032967,
+      "grad_norm": 0.01060621626675129,
+      "learning_rate": 0.0018151465798045604,
+      "loss": 1.3065,
+      "step": 622
+    },
+    {
+      "epoch": 1.3692307692307693,
+      "grad_norm": 0.009669648483395576,
+      "learning_rate": 0.0018127035830618895,
+      "loss": 1.2851,
+      "step": 623
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.007756940554827452,
+      "learning_rate": 0.0018102605863192184,
+      "loss": 1.3061,
+      "step": 624
+    },
+    {
+      "epoch": 1.3736263736263736,
+      "grad_norm": 0.026689475402235985,
+      "learning_rate": 0.0018078175895765475,
+      "loss": 1.2658,
+      "step": 625
+    },
+    {
+      "epoch": 1.3758241758241758,
+      "grad_norm": 0.009520587511360645,
+      "learning_rate": 0.0018053745928338764,
+      "loss": 1.0932,
+      "step": 626
+    },
+    {
+      "epoch": 1.378021978021978,
+      "grad_norm": 0.014994572848081589,
+      "learning_rate": 0.0018029315960912053,
+      "loss": 1.3798,
+      "step": 627
+    },
+    {
+      "epoch": 1.3802197802197802,
+      "grad_norm": 0.010245769284665585,
+      "learning_rate": 0.0018004885993485344,
+      "loss": 1.2203,
+      "step": 628
+    },
+    {
+      "epoch": 1.3824175824175824,
+      "grad_norm": 0.010305079631507397,
+      "learning_rate": 0.0017980456026058633,
+      "loss": 1.2175,
+      "step": 629
+    },
+    {
+      "epoch": 1.3846153846153846,
+      "grad_norm": 0.014559204690158367,
+      "learning_rate": 0.0017956026058631924,
+      "loss": 1.2223,
+      "step": 630
+    },
+    {
+      "epoch": 1.3868131868131868,
+      "grad_norm": 0.0119517482817173,
+      "learning_rate": 0.0017931596091205213,
+      "loss": 1.3834,
+      "step": 631
+    },
+    {
+      "epoch": 1.389010989010989,
+      "grad_norm": 0.012630819343030453,
+      "learning_rate": 0.0017907166123778504,
+      "loss": 1.2999,
+      "step": 632
+    },
+    {
+      "epoch": 1.3912087912087912,
+      "grad_norm": 0.012436777353286743,
+      "learning_rate": 0.0017882736156351793,
+      "loss": 1.233,
+      "step": 633
+    },
+    {
+      "epoch": 1.3934065934065933,
+      "grad_norm": 0.009733055718243122,
+      "learning_rate": 0.0017858306188925084,
+      "loss": 1.2425,
+      "step": 634
+    },
+    {
+      "epoch": 1.3956043956043955,
+      "grad_norm": 0.014204381965100765,
+      "learning_rate": 0.0017833876221498373,
+      "loss": 1.2729,
+      "step": 635
+    },
+    {
+      "epoch": 1.3978021978021977,
+      "grad_norm": 0.00814998708665371,
+      "learning_rate": 0.0017809446254071664,
+      "loss": 1.1191,
+      "step": 636
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.015157910995185375,
+      "learning_rate": 0.0017785016286644953,
+      "loss": 1.2763,
+      "step": 637
+    },
+    {
+      "epoch": 1.402197802197802,
+      "grad_norm": 0.007780939806252718,
+      "learning_rate": 0.0017760586319218242,
+      "loss": 1.2336,
+      "step": 638
+    },
+    {
+      "epoch": 1.4043956043956043,
+      "grad_norm": 0.022356653586030006,
+      "learning_rate": 0.0017736156351791533,
+      "loss": 1.1384,
+      "step": 639
+    },
+    {
+      "epoch": 1.4065934065934065,
+      "grad_norm": 0.010843878611922264,
+      "learning_rate": 0.0017711726384364822,
+      "loss": 1.294,
+      "step": 640
+    },
+    {
+      "epoch": 1.4087912087912087,
+      "grad_norm": 0.01623634248971939,
+      "learning_rate": 0.0017687296416938113,
+      "loss": 1.3374,
+      "step": 641
+    },
+    {
+      "epoch": 1.4109890109890109,
+      "grad_norm": 0.011397762224078178,
+      "learning_rate": 0.0017662866449511402,
+      "loss": 1.2067,
+      "step": 642
+    },
+    {
+      "epoch": 1.413186813186813,
+      "grad_norm": 0.015096818096935749,
+      "learning_rate": 0.0017638436482084693,
+      "loss": 1.0455,
+      "step": 643
+    },
+    {
+      "epoch": 1.4153846153846155,
+      "grad_norm": 0.01465892605483532,
+      "learning_rate": 0.0017614006514657982,
+      "loss": 1.3199,
+      "step": 644
+    },
+    {
+      "epoch": 1.4175824175824177,
+      "grad_norm": 0.018846601247787476,
+      "learning_rate": 0.0017589576547231273,
+      "loss": 1.4854,
+      "step": 645
+    },
+    {
+      "epoch": 1.4197802197802198,
+      "grad_norm": 0.013873511925339699,
+      "learning_rate": 0.0017565146579804562,
+      "loss": 1.6688,
+      "step": 646
+    },
+    {
+      "epoch": 1.421978021978022,
+      "grad_norm": 0.011877002194523811,
+      "learning_rate": 0.001754071661237785,
+      "loss": 1.2959,
+      "step": 647
+    },
+    {
+      "epoch": 1.4241758241758242,
+      "grad_norm": 0.01166671421378851,
+      "learning_rate": 0.0017516286644951142,
+      "loss": 1.4232,
+      "step": 648
+    },
+    {
+      "epoch": 1.4263736263736264,
+      "grad_norm": 0.01266437117010355,
+      "learning_rate": 0.001749185667752443,
+      "loss": 1.216,
+      "step": 649
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.011775790713727474,
+      "learning_rate": 0.0017467426710097722,
+      "loss": 1.2749,
+      "step": 650
+    },
+    {
+      "epoch": 1.4307692307692308,
+      "grad_norm": 0.010579568333923817,
+      "learning_rate": 0.001744299674267101,
+      "loss": 1.1762,
+      "step": 651
+    },
+    {
+      "epoch": 1.432967032967033,
+      "grad_norm": 0.013902468606829643,
+      "learning_rate": 0.0017418566775244302,
+      "loss": 1.2661,
+      "step": 652
+    },
+    {
+      "epoch": 1.4351648351648352,
+      "grad_norm": 0.012006877921521664,
+      "learning_rate": 0.001739413680781759,
+      "loss": 1.3089,
+      "step": 653
+    },
+    {
+      "epoch": 1.4373626373626374,
+      "grad_norm": 0.011949055828154087,
+      "learning_rate": 0.0017369706840390882,
+      "loss": 1.2481,
+      "step": 654
+    },
+    {
+      "epoch": 1.4395604395604396,
+      "grad_norm": 0.019233338534832,
+      "learning_rate": 0.001734527687296417,
+      "loss": 1.0841,
+      "step": 655
+    },
+    {
+      "epoch": 1.4417582417582417,
+      "grad_norm": 0.01738724298775196,
+      "learning_rate": 0.001732084690553746,
+      "loss": 1.3804,
+      "step": 656
+    },
+    {
+      "epoch": 1.443956043956044,
+      "grad_norm": 0.01236459519714117,
+      "learning_rate": 0.001729641693811075,
+      "loss": 1.2248,
+      "step": 657
+    },
+    {
+      "epoch": 1.4461538461538461,
+      "grad_norm": 0.01071969885379076,
+      "learning_rate": 0.001727198697068404,
+      "loss": 1.3585,
+      "step": 658
+    },
+    {
+      "epoch": 1.4483516483516483,
+      "grad_norm": 0.015186529606580734,
+      "learning_rate": 0.001724755700325733,
+      "loss": 1.5402,
+      "step": 659
+    },
+    {
+      "epoch": 1.4505494505494505,
+      "grad_norm": 0.009707773104310036,
+      "learning_rate": 0.001722312703583062,
+      "loss": 1.382,
+      "step": 660
+    },
+    {
+      "epoch": 1.4527472527472527,
+      "grad_norm": 0.015624198131263256,
+      "learning_rate": 0.001719869706840391,
+      "loss": 1.2394,
+      "step": 661
+    },
+    {
+      "epoch": 1.4549450549450549,
+      "grad_norm": 0.012445087544620037,
+      "learning_rate": 0.00171742671009772,
+      "loss": 1.3037,
+      "step": 662
+    },
+    {
+      "epoch": 1.457142857142857,
+      "grad_norm": 0.010136253200471401,
+      "learning_rate": 0.001714983713355049,
+      "loss": 1.1378,
+      "step": 663
+    },
+    {
+      "epoch": 1.4593406593406593,
+      "grad_norm": 0.009542087092995644,
+      "learning_rate": 0.001712540716612378,
+      "loss": 1.2127,
+      "step": 664
+    },
+    {
+      "epoch": 1.4615384615384617,
+      "grad_norm": 0.015063334256410599,
+      "learning_rate": 0.0017100977198697069,
+      "loss": 1.3039,
+      "step": 665
+    },
+    {
+      "epoch": 1.4637362637362639,
+      "grad_norm": 0.010203385725617409,
+      "learning_rate": 0.001707654723127036,
+      "loss": 1.1287,
+      "step": 666
+    },
+    {
+      "epoch": 1.465934065934066,
+      "grad_norm": 0.015582417137920856,
+      "learning_rate": 0.0017052117263843649,
+      "loss": 1.3109,
+      "step": 667
+    },
+    {
+      "epoch": 1.4681318681318682,
+      "grad_norm": 0.015825500711798668,
+      "learning_rate": 0.001702768729641694,
+      "loss": 1.1692,
+      "step": 668
+    },
+    {
+      "epoch": 1.4703296703296704,
+      "grad_norm": 0.011124425567686558,
+      "learning_rate": 0.0017003257328990229,
+      "loss": 1.1786,
+      "step": 669
+    },
+    {
+      "epoch": 1.4725274725274726,
+      "grad_norm": 0.015450658276677132,
+      "learning_rate": 0.001697882736156352,
+      "loss": 1.3468,
+      "step": 670
+    },
+    {
+      "epoch": 1.4747252747252748,
+      "grad_norm": 0.016263052821159363,
+      "learning_rate": 0.0016954397394136809,
+      "loss": 1.3454,
+      "step": 671
+    },
+    {
+      "epoch": 1.476923076923077,
+      "grad_norm": 0.010463095270097256,
+      "learning_rate": 0.00169299674267101,
+      "loss": 1.1602,
+      "step": 672
+    },
+    {
+      "epoch": 1.4791208791208792,
+      "grad_norm": 0.014461630955338478,
+      "learning_rate": 0.0016905537459283389,
+      "loss": 1.3847,
+      "step": 673
+    },
+    {
+      "epoch": 1.4813186813186814,
+      "grad_norm": 0.009912253357470036,
+      "learning_rate": 0.001688110749185668,
+      "loss": 1.1022,
+      "step": 674
+    },
+    {
+      "epoch": 1.4835164835164836,
+      "grad_norm": 0.013759217225015163,
+      "learning_rate": 0.0016856677524429969,
+      "loss": 1.3166,
+      "step": 675
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.012520554475486279,
+      "learning_rate": 0.0016832247557003258,
+      "loss": 1.0888,
+      "step": 676
+    },
+    {
+      "epoch": 1.487912087912088,
+      "grad_norm": 0.014557498507201672,
+      "learning_rate": 0.0016807817589576549,
+      "loss": 1.318,
+      "step": 677
+    },
+    {
+      "epoch": 1.4901098901098901,
+      "grad_norm": 0.021435771137475967,
+      "learning_rate": 0.0016783387622149838,
+      "loss": 1.6007,
+      "step": 678
+    },
+    {
+      "epoch": 1.4923076923076923,
+      "grad_norm": 0.011264498345553875,
+      "learning_rate": 0.0016758957654723129,
+      "loss": 1.2962,
+      "step": 679
+    },
+    {
+      "epoch": 1.4945054945054945,
+      "grad_norm": 0.019099639728665352,
+      "learning_rate": 0.0016734527687296418,
+      "loss": 1.2819,
+      "step": 680
+    },
+    {
+      "epoch": 1.4967032967032967,
+      "grad_norm": 0.012928360141813755,
+      "learning_rate": 0.0016710097719869709,
+      "loss": 1.1882,
+      "step": 681
+    },
+    {
+      "epoch": 1.498901098901099,
+      "grad_norm": 0.012643392197787762,
+      "learning_rate": 0.0016685667752442998,
+      "loss": 1.515,
+      "step": 682
+    },
+    {
+      "epoch": 1.501098901098901,
+      "grad_norm": 0.01199061144143343,
+      "learning_rate": 0.0016661237785016289,
+      "loss": 1.2633,
+      "step": 683
+    },
+    {
+      "epoch": 1.5032967032967033,
+      "grad_norm": 0.017223527655005455,
+      "learning_rate": 0.0016636807817589578,
+      "loss": 1.3018,
+      "step": 684
+    },
+    {
+      "epoch": 1.5054945054945055,
+      "grad_norm": 0.039446037262678146,
+      "learning_rate": 0.0016612377850162867,
+      "loss": 1.1636,
+      "step": 685
+    },
+    {
+      "epoch": 1.5076923076923077,
+      "grad_norm": 0.014351975172758102,
+      "learning_rate": 0.0016587947882736158,
+      "loss": 1.3693,
+      "step": 686
+    },
+    {
+      "epoch": 1.5098901098901099,
+      "grad_norm": 0.019068924710154533,
+      "learning_rate": 0.0016563517915309447,
+      "loss": 1.2441,
+      "step": 687
+    },
+    {
+      "epoch": 1.512087912087912,
+      "grad_norm": 0.016259146854281425,
+      "learning_rate": 0.0016539087947882738,
+      "loss": 1.367,
+      "step": 688
+    },
+    {
+      "epoch": 1.5142857142857142,
+      "grad_norm": 0.015941163524985313,
+      "learning_rate": 0.0016514657980456027,
+      "loss": 1.3291,
+      "step": 689
+    },
+    {
+      "epoch": 1.5164835164835164,
+      "grad_norm": 0.0135962488129735,
+      "learning_rate": 0.0016490228013029318,
+      "loss": 1.173,
+      "step": 690
+    },
+    {
+      "epoch": 1.5186813186813186,
+      "grad_norm": 0.02029673382639885,
+      "learning_rate": 0.0016465798045602607,
+      "loss": 1.0343,
+      "step": 691
+    },
+    {
+      "epoch": 1.5208791208791208,
+      "grad_norm": 0.016091369092464447,
+      "learning_rate": 0.0016441368078175898,
+      "loss": 1.2518,
+      "step": 692
+    },
+    {
+      "epoch": 1.523076923076923,
+      "grad_norm": 0.013290997594594955,
+      "learning_rate": 0.0016416938110749187,
+      "loss": 1.2229,
+      "step": 693
+    },
+    {
+      "epoch": 1.5252747252747252,
+      "grad_norm": 0.011686509475111961,
+      "learning_rate": 0.0016392508143322475,
+      "loss": 1.1743,
+      "step": 694
+    },
+    {
+      "epoch": 1.5274725274725274,
+      "grad_norm": 0.013623136095702648,
+      "learning_rate": 0.0016368078175895767,
+      "loss": 1.0528,
+      "step": 695
+    },
+    {
+      "epoch": 1.5296703296703296,
+      "grad_norm": 0.016562001779675484,
+      "learning_rate": 0.0016343648208469056,
+      "loss": 1.3477,
+      "step": 696
+    },
+    {
+      "epoch": 1.5318681318681318,
+      "grad_norm": 0.013674282468855381,
+      "learning_rate": 0.0016319218241042347,
+      "loss": 1.1116,
+      "step": 697
+    },
+    {
+      "epoch": 1.534065934065934,
+      "grad_norm": 0.015475422143936157,
+      "learning_rate": 0.0016294788273615636,
+      "loss": 1.2507,
+      "step": 698
+    },
+    {
+      "epoch": 1.5362637362637361,
+      "grad_norm": 0.014629727229475975,
+      "learning_rate": 0.0016270358306188927,
+      "loss": 1.1886,
+      "step": 699
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 0.013872419483959675,
+      "learning_rate": 0.0016245928338762216,
+      "loss": 1.1778,
+      "step": 700
+    },
+    {
+      "epoch": 1.5406593406593405,
+      "grad_norm": 0.017514893785119057,
+      "learning_rate": 0.0016221498371335507,
+      "loss": 1.5172,
+      "step": 701
+    },
+    {
+      "epoch": 1.5428571428571427,
+      "grad_norm": 0.013659069314599037,
+      "learning_rate": 0.0016197068403908796,
+      "loss": 1.2157,
+      "step": 702
+    },
+    {
+      "epoch": 1.545054945054945,
+      "grad_norm": 0.021774468943476677,
+      "learning_rate": 0.0016172638436482087,
+      "loss": 1.1799,
+      "step": 703
+    },
+    {
+      "epoch": 1.5472527472527473,
+      "grad_norm": 0.011277959682047367,
+      "learning_rate": 0.0016148208469055376,
+      "loss": 1.1414,
+      "step": 704
+    },
+    {
+      "epoch": 1.5494505494505495,
+      "grad_norm": 0.013084550388157368,
+      "learning_rate": 0.0016123778501628664,
+      "loss": 1.0694,
+      "step": 705
+    },
+    {
+      "epoch": 1.5516483516483517,
+      "grad_norm": 0.016685500741004944,
+      "learning_rate": 0.0016099348534201956,
+      "loss": 1.5639,
+      "step": 706
+    },
+    {
+      "epoch": 1.5538461538461539,
+      "grad_norm": 0.02061067521572113,
+      "learning_rate": 0.0016074918566775244,
+      "loss": 1.0824,
+      "step": 707
+    },
+    {
+      "epoch": 1.556043956043956,
+      "grad_norm": 0.015204605646431446,
+      "learning_rate": 0.0016050488599348536,
+      "loss": 1.2638,
+      "step": 708
+    },
+    {
+      "epoch": 1.5582417582417583,
+      "grad_norm": 0.012646040879189968,
+      "learning_rate": 0.0016026058631921824,
+      "loss": 1.1606,
+      "step": 709
+    },
+    {
+      "epoch": 1.5604395604395604,
+      "grad_norm": 0.02266811579465866,
+      "learning_rate": 0.0016001628664495116,
+      "loss": 1.369,
+      "step": 710
+    },
+    {
+      "epoch": 1.5626373626373626,
+      "grad_norm": 0.025244278833270073,
+      "learning_rate": 0.0015977198697068404,
+      "loss": 1.1696,
+      "step": 711
+    },
+    {
+      "epoch": 1.5648351648351648,
+      "grad_norm": 0.012283441610634327,
+      "learning_rate": 0.0015952768729641696,
+      "loss": 1.0804,
+      "step": 712
+    },
+    {
+      "epoch": 1.567032967032967,
+      "grad_norm": 0.028124062344431877,
+      "learning_rate": 0.0015928338762214984,
+      "loss": 1.3319,
+      "step": 713
+    },
+    {
+      "epoch": 1.5692307692307692,
+      "grad_norm": 0.015816954895853996,
+      "learning_rate": 0.0015903908794788273,
+      "loss": 1.2147,
+      "step": 714
+    },
+    {
+      "epoch": 1.5714285714285714,
+      "grad_norm": 0.016656775027513504,
+      "learning_rate": 0.0015879478827361564,
+      "loss": 1.3039,
+      "step": 715
+    },
+    {
+      "epoch": 1.5736263736263736,
+      "grad_norm": 0.015638362616300583,
+      "learning_rate": 0.0015855048859934853,
+      "loss": 1.3339,
+      "step": 716
+    },
+    {
+      "epoch": 1.5758241758241758,
+      "grad_norm": 0.014728598296642303,
+      "learning_rate": 0.0015830618892508145,
+      "loss": 1.311,
+      "step": 717
+    },
+    {
+      "epoch": 1.578021978021978,
+      "grad_norm": 0.017344551160931587,
+      "learning_rate": 0.0015806188925081433,
+      "loss": 1.2517,
+      "step": 718
+    },
+    {
+      "epoch": 1.5802197802197804,
+      "grad_norm": 0.014576535671949387,
+      "learning_rate": 0.0015781758957654725,
+      "loss": 1.3089,
+      "step": 719
+    },
+    {
+      "epoch": 1.5824175824175826,
+      "grad_norm": 0.02584868296980858,
+      "learning_rate": 0.0015757328990228013,
+      "loss": 1.2487,
+      "step": 720
+    },
+    {
+      "epoch": 1.5846153846153848,
+      "grad_norm": 0.02446175552904606,
+      "learning_rate": 0.0015732899022801305,
+      "loss": 1.3067,
+      "step": 721
+    },
+    {
+      "epoch": 1.586813186813187,
+      "grad_norm": 0.02153056673705578,
+      "learning_rate": 0.0015708469055374593,
+      "loss": 1.1984,
+      "step": 722
+    },
+    {
+      "epoch": 1.5890109890109891,
+      "grad_norm": 0.017252309247851372,
+      "learning_rate": 0.0015684039087947882,
+      "loss": 1.3886,
+      "step": 723
+    },
+    {
+      "epoch": 1.5912087912087913,
+      "grad_norm": 0.021748440340161324,
+      "learning_rate": 0.0015659609120521173,
+      "loss": 1.335,
+      "step": 724
+    },
+    {
+      "epoch": 1.5934065934065935,
+      "grad_norm": 0.01892728917300701,
+      "learning_rate": 0.0015635179153094462,
+      "loss": 1.4251,
+      "step": 725
+    },
+    {
+      "epoch": 1.5956043956043957,
+      "grad_norm": 0.015512732788920403,
+      "learning_rate": 0.0015610749185667753,
+      "loss": 1.3899,
+      "step": 726
+    },
+    {
+      "epoch": 1.597802197802198,
+      "grad_norm": 0.0180047657340765,
+      "learning_rate": 0.0015586319218241042,
+      "loss": 1.4812,
+      "step": 727
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.016533158719539642,
+      "learning_rate": 0.0015561889250814333,
+      "loss": 1.3076,
+      "step": 728
+    },
+    {
+      "epoch": 1.6021978021978023,
+      "grad_norm": 0.02126281149685383,
+      "learning_rate": 0.0015537459283387622,
+      "loss": 1.281,
+      "step": 729
+    },
+    {
+      "epoch": 1.6043956043956045,
+      "grad_norm": 0.017477162182331085,
+      "learning_rate": 0.0015513029315960913,
+      "loss": 1.2051,
+      "step": 730
+    },
+    {
+      "epoch": 1.6065934065934067,
+      "grad_norm": 0.015082748606801033,
+      "learning_rate": 0.0015488599348534202,
+      "loss": 1.0821,
+      "step": 731
+    },
+    {
+      "epoch": 1.6087912087912088,
+      "grad_norm": 0.012245159596204758,
+      "learning_rate": 0.0015464169381107491,
+      "loss": 1.4063,
+      "step": 732
+    },
+    {
+      "epoch": 1.610989010989011,
+      "grad_norm": 0.019378548488020897,
+      "learning_rate": 0.0015439739413680782,
+      "loss": 1.1071,
+      "step": 733
+    },
+    {
+      "epoch": 1.6131868131868132,
+      "grad_norm": 0.011383804492652416,
+      "learning_rate": 0.0015415309446254071,
+      "loss": 1.1662,
+      "step": 734
+    },
+    {
+      "epoch": 1.6153846153846154,
+      "grad_norm": 0.016531851142644882,
+      "learning_rate": 0.0015390879478827362,
+      "loss": 1.1172,
+      "step": 735
+    },
+    {
+      "epoch": 1.6175824175824176,
+      "grad_norm": 0.04639346897602081,
+      "learning_rate": 0.0015366449511400651,
+      "loss": 1.6241,
+      "step": 736
+    },
+    {
+      "epoch": 1.6197802197802198,
+      "grad_norm": 0.014337725937366486,
+      "learning_rate": 0.0015342019543973942,
+      "loss": 1.0386,
+      "step": 737
+    },
+    {
+      "epoch": 1.621978021978022,
+      "grad_norm": 0.015334147959947586,
+      "learning_rate": 0.0015317589576547231,
+      "loss": 1.243,
+      "step": 738
+    },
+    {
+      "epoch": 1.6241758241758242,
+      "grad_norm": 0.011797674000263214,
+      "learning_rate": 0.0015293159609120522,
+      "loss": 1.178,
+      "step": 739
+    },
+    {
+      "epoch": 1.6263736263736264,
+      "grad_norm": 0.01543725561350584,
+      "learning_rate": 0.0015268729641693811,
+      "loss": 1.1603,
+      "step": 740
+    },
+    {
+      "epoch": 1.6285714285714286,
+      "grad_norm": 0.018438948318362236,
+      "learning_rate": 0.0015244299674267102,
+      "loss": 1.1954,
+      "step": 741
+    },
+    {
+      "epoch": 1.6307692307692307,
+      "grad_norm": 0.016693588346242905,
+      "learning_rate": 0.0015219869706840391,
+      "loss": 1.0425,
+      "step": 742
+    },
+    {
+      "epoch": 1.632967032967033,
+      "grad_norm": 0.01162511482834816,
+      "learning_rate": 0.001519543973941368,
+      "loss": 1.1446,
+      "step": 743
+    },
+    {
+      "epoch": 1.6351648351648351,
+      "grad_norm": 0.014890515245497227,
+      "learning_rate": 0.0015171009771986971,
+      "loss": 0.9861,
+      "step": 744
+    },
+    {
+      "epoch": 1.6373626373626373,
+      "grad_norm": 0.01899428851902485,
+      "learning_rate": 0.001514657980456026,
+      "loss": 0.9586,
+      "step": 745
+    },
+    {
+      "epoch": 1.6395604395604395,
+      "grad_norm": 0.019463766366243362,
+      "learning_rate": 0.0015122149837133551,
+      "loss": 1.1511,
+      "step": 746
+    },
+    {
+      "epoch": 1.6417582417582417,
+      "grad_norm": 0.012637551873922348,
+      "learning_rate": 0.001509771986970684,
+      "loss": 1.0723,
+      "step": 747
+    },
+    {
+      "epoch": 1.6439560439560439,
+      "grad_norm": 0.02219577692449093,
+      "learning_rate": 0.0015073289902280131,
+      "loss": 1.1376,
+      "step": 748
+    },
+    {
+      "epoch": 1.646153846153846,
+      "grad_norm": 0.019987845793366432,
+      "learning_rate": 0.001504885993485342,
+      "loss": 0.8866,
+      "step": 749
+    },
+    {
+      "epoch": 1.6483516483516483,
+      "grad_norm": 0.01576593518257141,
+      "learning_rate": 0.0015024429967426711,
+      "loss": 1.0304,
+      "step": 750
+    },
+    {
+      "epoch": 1.6505494505494505,
+      "grad_norm": 0.01196097768843174,
+      "learning_rate": 0.0015,
+      "loss": 1.3228,
+      "step": 751
+    },
+    {
+      "epoch": 1.6527472527472526,
+      "grad_norm": 0.012350653298199177,
+      "learning_rate": 0.001497557003257329,
+      "loss": 1.1149,
+      "step": 752
+    },
+    {
+      "epoch": 1.6549450549450548,
+      "grad_norm": 0.013656220398843288,
+      "learning_rate": 0.001495114006514658,
+      "loss": 1.1428,
+      "step": 753
+    },
+    {
+      "epoch": 1.657142857142857,
+      "grad_norm": 0.023477014154195786,
+      "learning_rate": 0.001492671009771987,
+      "loss": 1.273,
+      "step": 754
+    },
+    {
+      "epoch": 1.6593406593406592,
+      "grad_norm": 0.023004047572612762,
+      "learning_rate": 0.001490228013029316,
+      "loss": 1.078,
+      "step": 755
+    },
+    {
+      "epoch": 1.6615384615384614,
+      "grad_norm": 0.01143412571400404,
+      "learning_rate": 0.001487785016286645,
+      "loss": 1.173,
+      "step": 756
+    },
+    {
+      "epoch": 1.6637362637362636,
+      "grad_norm": 0.02114829048514366,
+      "learning_rate": 0.001485342019543974,
+      "loss": 1.2063,
+      "step": 757
+    },
+    {
+      "epoch": 1.6659340659340658,
+      "grad_norm": 0.01743108034133911,
+      "learning_rate": 0.001482899022801303,
+      "loss": 1.4103,
+      "step": 758
+    },
+    {
+      "epoch": 1.668131868131868,
+      "grad_norm": 0.022327395156025887,
+      "learning_rate": 0.001480456026058632,
+      "loss": 1.258,
+      "step": 759
+    },
+    {
+      "epoch": 1.6703296703296702,
+      "grad_norm": 0.018606038764119148,
+      "learning_rate": 0.001478013029315961,
+      "loss": 1.2534,
+      "step": 760
+    },
+    {
+      "epoch": 1.6725274725274726,
+      "grad_norm": 0.021844081580638885,
+      "learning_rate": 0.0014755700325732898,
+      "loss": 0.9618,
+      "step": 761
+    },
+    {
+      "epoch": 1.6747252747252748,
+      "grad_norm": 0.013079552911221981,
+      "learning_rate": 0.001473127035830619,
+      "loss": 1.3177,
+      "step": 762
+    },
+    {
+      "epoch": 1.676923076923077,
+      "grad_norm": 0.013392949476838112,
+      "learning_rate": 0.0014706840390879478,
+      "loss": 1.2783,
+      "step": 763
+    },
+    {
+      "epoch": 1.6791208791208792,
+      "grad_norm": 0.014620695263147354,
+      "learning_rate": 0.001468241042345277,
+      "loss": 1.1731,
+      "step": 764
+    },
+    {
+      "epoch": 1.6813186813186813,
+      "grad_norm": 0.024165354669094086,
+      "learning_rate": 0.0014657980456026058,
+      "loss": 1.0531,
+      "step": 765
+    },
+    {
+      "epoch": 1.6835164835164835,
+      "grad_norm": 0.015794849023222923,
+      "learning_rate": 0.001463355048859935,
+      "loss": 1.0702,
+      "step": 766
+    },
+    {
+      "epoch": 1.6857142857142857,
+      "grad_norm": 0.013587509281933308,
+      "learning_rate": 0.0014609120521172638,
+      "loss": 1.0576,
+      "step": 767
+    },
+    {
+      "epoch": 1.687912087912088,
+      "grad_norm": 0.05401821434497833,
+      "learning_rate": 0.001458469055374593,
+      "loss": 1.1809,
+      "step": 768
+    },
+    {
+      "epoch": 1.69010989010989,
+      "grad_norm": 0.013319086283445358,
+      "learning_rate": 0.0014560260586319218,
+      "loss": 1.3439,
+      "step": 769
+    },
+    {
+      "epoch": 1.6923076923076923,
+      "grad_norm": 0.016835635527968407,
+      "learning_rate": 0.001453583061889251,
+      "loss": 1.2401,
+      "step": 770
+    },
+    {
+      "epoch": 1.6945054945054945,
+      "grad_norm": 0.016473611816763878,
+      "learning_rate": 0.0014511400651465798,
+      "loss": 0.9798,
+      "step": 771
+    },
+    {
+      "epoch": 1.6967032967032967,
+      "grad_norm": 0.02077752910554409,
+      "learning_rate": 0.0014486970684039087,
+      "loss": 1.3223,
+      "step": 772
+    },
+    {
+      "epoch": 1.6989010989010989,
+      "grad_norm": 0.013765125535428524,
+      "learning_rate": 0.0014462540716612378,
+      "loss": 1.1985,
+      "step": 773
+    },
+    {
+      "epoch": 1.701098901098901,
+      "grad_norm": 0.021739911288022995,
+      "learning_rate": 0.0014438110749185667,
+      "loss": 1.5051,
+      "step": 774
+    },
+    {
+      "epoch": 1.7032967032967035,
+      "grad_norm": 0.012776698917150497,
+      "learning_rate": 0.0014413680781758958,
+      "loss": 1.0245,
+      "step": 775
+    },
+    {
+      "epoch": 1.7054945054945057,
+      "grad_norm": 0.016143497079610825,
+      "learning_rate": 0.0014389250814332247,
+      "loss": 1.16,
+      "step": 776
+    },
+    {
+      "epoch": 1.7076923076923078,
+      "grad_norm": 0.03685399517416954,
+      "learning_rate": 0.0014364820846905538,
+      "loss": 1.2637,
+      "step": 777
+    },
+    {
+      "epoch": 1.70989010989011,
+      "grad_norm": 0.018249938264489174,
+      "learning_rate": 0.0014340390879478827,
+      "loss": 1.1584,
+      "step": 778
+    },
+    {
+      "epoch": 1.7120879120879122,
+      "grad_norm": 0.02126156911253929,
+      "learning_rate": 0.0014315960912052118,
+      "loss": 1.1492,
+      "step": 779
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.10813774168491364,
+      "learning_rate": 0.0014291530944625407,
+      "loss": 1.2917,
+      "step": 780
+    },
+    {
+      "epoch": 1.7164835164835166,
+      "grad_norm": 0.019032590091228485,
+      "learning_rate": 0.0014267100977198696,
+      "loss": 1.153,
+      "step": 781
+    },
+    {
+      "epoch": 1.7186813186813188,
+      "grad_norm": 0.01757064461708069,
+      "learning_rate": 0.0014242671009771987,
+      "loss": 1.2127,
+      "step": 782
+    },
+    {
+      "epoch": 1.720879120879121,
+      "grad_norm": 0.016893142834305763,
+      "learning_rate": 0.0014218241042345276,
+      "loss": 1.1105,
+      "step": 783
+    },
+    {
+      "epoch": 1.7230769230769232,
+      "grad_norm": 0.020693473517894745,
+      "learning_rate": 0.0014193811074918567,
+      "loss": 1.257,
+      "step": 784
+    },
+    {
+      "epoch": 1.7252747252747254,
+      "grad_norm": 0.016782060265541077,
+      "learning_rate": 0.0014169381107491856,
+      "loss": 1.194,
+      "step": 785
+    },
+    {
+      "epoch": 1.7274725274725276,
+      "grad_norm": 0.014398118481040001,
+      "learning_rate": 0.0014144951140065147,
+      "loss": 1.1436,
+      "step": 786
+    },
+    {
+      "epoch": 1.7296703296703297,
+      "grad_norm": 0.04177403450012207,
+      "learning_rate": 0.0014120521172638436,
+      "loss": 1.2818,
+      "step": 787
+    },
+    {
+      "epoch": 1.731868131868132,
+      "grad_norm": 0.01436767727136612,
+      "learning_rate": 0.0014096091205211727,
+      "loss": 1.1739,
+      "step": 788
+    },
+    {
+      "epoch": 1.7340659340659341,
+      "grad_norm": 0.015435214154422283,
+      "learning_rate": 0.0014071661237785016,
+      "loss": 1.2209,
+      "step": 789
+    },
+    {
+      "epoch": 1.7362637362637363,
+      "grad_norm": 0.019650137051939964,
+      "learning_rate": 0.0014047231270358305,
+      "loss": 1.2583,
+      "step": 790
+    },
+    {
+      "epoch": 1.7384615384615385,
+      "grad_norm": 0.02432181127369404,
+      "learning_rate": 0.0014022801302931596,
+      "loss": 1.1655,
+      "step": 791
+    },
+    {
+      "epoch": 1.7406593406593407,
+      "grad_norm": 0.08194996416568756,
+      "learning_rate": 0.0013998371335504885,
+      "loss": 1.286,
+      "step": 792
+    },
+    {
+      "epoch": 1.7428571428571429,
+      "grad_norm": 0.016385288909077644,
+      "learning_rate": 0.0013973941368078176,
+      "loss": 1.3949,
+      "step": 793
+    },
+    {
+      "epoch": 1.745054945054945,
+      "grad_norm": 0.018561335280537605,
+      "learning_rate": 0.0013949511400651465,
+      "loss": 1.1893,
+      "step": 794
+    },
+    {
+      "epoch": 1.7472527472527473,
+      "grad_norm": 0.015930531546473503,
+      "learning_rate": 0.0013925081433224756,
+      "loss": 1.3288,
+      "step": 795
+    },
+    {
+      "epoch": 1.7494505494505495,
+      "grad_norm": 0.011869377456605434,
+      "learning_rate": 0.0013900651465798045,
+      "loss": 1.1695,
+      "step": 796
+    },
+    {
+      "epoch": 1.7516483516483516,
+      "grad_norm": 0.019633185118436813,
+      "learning_rate": 0.0013876221498371336,
+      "loss": 1.1786,
+      "step": 797
+    },
+    {
+      "epoch": 1.7538461538461538,
+      "grad_norm": 0.017025187611579895,
+      "learning_rate": 0.0013851791530944625,
+      "loss": 1.0935,
+      "step": 798
+    },
+    {
+      "epoch": 1.756043956043956,
+      "grad_norm": 0.01677589677274227,
+      "learning_rate": 0.0013827361563517914,
+      "loss": 1.3436,
+      "step": 799
+    },
+    {
+      "epoch": 1.7582417582417582,
+      "grad_norm": 0.016136005520820618,
+      "learning_rate": 0.0013802931596091205,
+      "loss": 1.2418,
+      "step": 800
+    },
+    {
+      "epoch": 1.7604395604395604,
+      "grad_norm": 0.03322204574942589,
+      "learning_rate": 0.0013778501628664494,
+      "loss": 1.2403,
+      "step": 801
+    },
+    {
+      "epoch": 1.7626373626373626,
+      "grad_norm": 0.019656501710414886,
+      "learning_rate": 0.0013754071661237785,
+      "loss": 1.2068,
+      "step": 802
+    },
+    {
+      "epoch": 1.7648351648351648,
+      "grad_norm": 0.021833691745996475,
+      "learning_rate": 0.0013729641693811074,
+      "loss": 1.1749,
+      "step": 803
+    },
+    {
+      "epoch": 1.767032967032967,
+      "grad_norm": 0.01640917733311653,
+      "learning_rate": 0.0013705211726384365,
+      "loss": 1.1284,
+      "step": 804
+    },
+    {
+      "epoch": 1.7692307692307692,
+      "grad_norm": 0.017773520201444626,
+      "learning_rate": 0.0013680781758957654,
+      "loss": 1.1051,
+      "step": 805
+    },
+    {
+      "epoch": 1.7714285714285714,
+      "grad_norm": 0.013762084767222404,
+      "learning_rate": 0.0013656351791530945,
+      "loss": 1.1906,
+      "step": 806
+    },
+    {
+      "epoch": 1.7736263736263735,
+      "grad_norm": 0.017267320305109024,
+      "learning_rate": 0.0013631921824104234,
+      "loss": 1.3256,
+      "step": 807
+    },
+    {
+      "epoch": 1.7758241758241757,
+      "grad_norm": 0.01900363340973854,
+      "learning_rate": 0.0013607491856677525,
+      "loss": 1.2096,
+      "step": 808
+    },
+    {
+      "epoch": 1.778021978021978,
+      "grad_norm": 0.01512153074145317,
+      "learning_rate": 0.0013583061889250814,
+      "loss": 1.1332,
+      "step": 809
+    },
+    {
+      "epoch": 1.7802197802197801,
+      "grad_norm": 0.01562659628689289,
+      "learning_rate": 0.0013558631921824103,
+      "loss": 1.225,
+      "step": 810
+    },
+    {
+      "epoch": 1.7824175824175823,
+      "grad_norm": 0.027901241555809975,
+      "learning_rate": 0.0013534201954397394,
+      "loss": 1.2386,
+      "step": 811
+    },
+    {
+      "epoch": 1.7846153846153845,
+      "grad_norm": 0.015367643907666206,
+      "learning_rate": 0.0013509771986970683,
+      "loss": 1.0433,
+      "step": 812
+    },
+    {
+      "epoch": 1.7868131868131867,
+      "grad_norm": 0.016418293118476868,
+      "learning_rate": 0.0013485342019543974,
+      "loss": 0.9961,
+      "step": 813
+    },
+    {
+      "epoch": 1.7890109890109889,
+      "grad_norm": 0.016704102978110313,
+      "learning_rate": 0.0013460912052117263,
+      "loss": 1.2801,
+      "step": 814
+    },
+    {
+      "epoch": 1.791208791208791,
+      "grad_norm": 0.01672317087650299,
+      "learning_rate": 0.0013436482084690554,
+      "loss": 1.1826,
+      "step": 815
+    },
+    {
+      "epoch": 1.7934065934065933,
+      "grad_norm": 0.018016209825873375,
+      "learning_rate": 0.0013412052117263843,
+      "loss": 1.2681,
+      "step": 816
+    },
+    {
+      "epoch": 1.7956043956043954,
+      "grad_norm": 0.028898248448967934,
+      "learning_rate": 0.0013387622149837134,
+      "loss": 1.2122,
+      "step": 817
+    },
+    {
+      "epoch": 1.7978021978021979,
+      "grad_norm": 0.015268645249307156,
+      "learning_rate": 0.0013363192182410423,
+      "loss": 1.0178,
+      "step": 818
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 0.01428668387234211,
+      "learning_rate": 0.0013338762214983712,
+      "loss": 1.2147,
+      "step": 819
+    },
+    {
+      "epoch": 1.8021978021978022,
+      "grad_norm": 0.022538773715496063,
+      "learning_rate": 0.0013314332247557003,
+      "loss": 1.2517,
+      "step": 820
+    },
+    {
+      "epoch": 1.8043956043956044,
+      "grad_norm": 0.019880985841155052,
+      "learning_rate": 0.0013289902280130292,
+      "loss": 1.2598,
+      "step": 821
+    },
+    {
+      "epoch": 1.8065934065934066,
+      "grad_norm": 0.014048965647816658,
+      "learning_rate": 0.0013265472312703583,
+      "loss": 1.3735,
+      "step": 822
+    },
+    {
+      "epoch": 1.8087912087912088,
+      "grad_norm": 0.02725040353834629,
+      "learning_rate": 0.0013241042345276872,
+      "loss": 1.3355,
+      "step": 823
+    },
+    {
+      "epoch": 1.810989010989011,
+      "grad_norm": 0.013692625798285007,
+      "learning_rate": 0.0013216612377850163,
+      "loss": 1.1996,
+      "step": 824
+    },
+    {
+      "epoch": 1.8131868131868132,
+      "grad_norm": 0.01800549402832985,
+      "learning_rate": 0.0013192182410423452,
+      "loss": 1.2611,
+      "step": 825
+    },
+    {
+      "epoch": 1.8153846153846154,
+      "grad_norm": 0.034041475504636765,
+      "learning_rate": 0.0013167752442996743,
+      "loss": 1.1745,
+      "step": 826
+    },
+    {
+      "epoch": 1.8175824175824176,
+      "grad_norm": 0.019226646050810814,
+      "learning_rate": 0.0013143322475570032,
+      "loss": 1.141,
+      "step": 827
+    },
+    {
+      "epoch": 1.8197802197802198,
+      "grad_norm": 0.017994670197367668,
+      "learning_rate": 0.0013118892508143323,
+      "loss": 0.9783,
+      "step": 828
+    },
+    {
+      "epoch": 1.821978021978022,
+      "grad_norm": 0.22939659655094147,
+      "learning_rate": 0.0013094462540716614,
+      "loss": 1.2172,
+      "step": 829
+    },
+    {
+      "epoch": 1.8241758241758241,
+      "grad_norm": 0.02140214294195175,
+      "learning_rate": 0.0013070032573289903,
+      "loss": 1.1541,
+      "step": 830
+    },
+    {
+      "epoch": 1.8263736263736263,
+      "grad_norm": 0.01699877716600895,
+      "learning_rate": 0.0013045602605863194,
+      "loss": 1.0915,
+      "step": 831
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.01787002943456173,
+      "learning_rate": 0.0013021172638436483,
+      "loss": 1.251,
+      "step": 832
+    },
+    {
+      "epoch": 1.830769230769231,
+      "grad_norm": 0.02216591313481331,
+      "learning_rate": 0.0012996742671009772,
+      "loss": 1.0814,
+      "step": 833
+    },
+    {
+      "epoch": 1.8329670329670331,
+      "grad_norm": 0.02157304435968399,
+      "learning_rate": 0.0012972312703583063,
+      "loss": 1.2862,
+      "step": 834
+    },
+    {
+      "epoch": 1.8351648351648353,
+      "grad_norm": 0.020130475983023643,
+      "learning_rate": 0.0012947882736156352,
+      "loss": 1.046,
+      "step": 835
+    },
+    {
+      "epoch": 1.8373626373626375,
+      "grad_norm": 0.01731707900762558,
+      "learning_rate": 0.0012923452768729643,
+      "loss": 1.1206,
+      "step": 836
+    },
+    {
+      "epoch": 1.8395604395604397,
+      "grad_norm": 0.01451348140835762,
+      "learning_rate": 0.0012899022801302932,
+      "loss": 1.2654,
+      "step": 837
+    },
+    {
+      "epoch": 1.8417582417582419,
+      "grad_norm": 0.020306188613176346,
+      "learning_rate": 0.0012874592833876223,
+      "loss": 1.1652,
+      "step": 838
+    },
+    {
+      "epoch": 1.843956043956044,
+      "grad_norm": 0.013657951727509499,
+      "learning_rate": 0.0012850162866449512,
+      "loss": 1.1833,
+      "step": 839
+    },
+    {
+      "epoch": 1.8461538461538463,
+      "grad_norm": 0.02617635391652584,
+      "learning_rate": 0.0012825732899022803,
+      "loss": 1.442,
+      "step": 840
+    },
+    {
+      "epoch": 1.8483516483516484,
+      "grad_norm": 0.014833233319222927,
+      "learning_rate": 0.0012801302931596092,
+      "loss": 1.0684,
+      "step": 841
+    },
+    {
+      "epoch": 1.8505494505494506,
+      "grad_norm": 0.029542014002799988,
+      "learning_rate": 0.0012776872964169383,
+      "loss": 1.2331,
+      "step": 842
+    },
+    {
+      "epoch": 1.8527472527472528,
+      "grad_norm": 0.01990935206413269,
+      "learning_rate": 0.0012752442996742672,
+      "loss": 1.3486,
+      "step": 843
+    },
+    {
+      "epoch": 1.854945054945055,
+      "grad_norm": 0.017032798379659653,
+      "learning_rate": 0.001272801302931596,
+      "loss": 1.1677,
+      "step": 844
+    },
+    {
+      "epoch": 1.8571428571428572,
+      "grad_norm": 0.018776411190629005,
+      "learning_rate": 0.0012703583061889252,
+      "loss": 1.2736,
+      "step": 845
+    },
+    {
+      "epoch": 1.8593406593406594,
+      "grad_norm": 0.013690021820366383,
+      "learning_rate": 0.001267915309446254,
+      "loss": 1.084,
+      "step": 846
+    },
+    {
+      "epoch": 1.8615384615384616,
+      "grad_norm": 0.01719077304005623,
+      "learning_rate": 0.0012654723127035832,
+      "loss": 1.1187,
+      "step": 847
+    },
+    {
+      "epoch": 1.8637362637362638,
+      "grad_norm": 0.015042711980640888,
+      "learning_rate": 0.001263029315960912,
+      "loss": 1.2394,
+      "step": 848
+    },
+    {
+      "epoch": 1.865934065934066,
+      "grad_norm": 0.013787307776510715,
+      "learning_rate": 0.0012605863192182412,
+      "loss": 1.2836,
+      "step": 849
+    },
+    {
+      "epoch": 1.8681318681318682,
+      "grad_norm": 0.015249217860400677,
+      "learning_rate": 0.00125814332247557,
+      "loss": 1.2061,
+      "step": 850
+    },
+    {
+      "epoch": 1.8703296703296703,
+      "grad_norm": 0.016133125871419907,
+      "learning_rate": 0.0012557003257328992,
+      "loss": 1.1835,
+      "step": 851
+    },
+    {
+      "epoch": 1.8725274725274725,
+      "grad_norm": 0.015879381448030472,
+      "learning_rate": 0.001253257328990228,
+      "loss": 1.1391,
+      "step": 852
+    },
+    {
+      "epoch": 1.8747252747252747,
+      "grad_norm": 0.022001340985298157,
+      "learning_rate": 0.001250814332247557,
+      "loss": 1.2993,
+      "step": 853
+    },
+    {
+      "epoch": 1.876923076923077,
+      "grad_norm": 0.015539953485131264,
+      "learning_rate": 0.001248371335504886,
+      "loss": 1.2772,
+      "step": 854
+    },
+    {
+      "epoch": 1.879120879120879,
+      "grad_norm": 0.01636308990418911,
+      "learning_rate": 0.001245928338762215,
+      "loss": 1.2298,
+      "step": 855
+    },
+    {
+      "epoch": 1.8813186813186813,
+      "grad_norm": 0.01758064702153206,
+      "learning_rate": 0.001243485342019544,
+      "loss": 1.1145,
+      "step": 856
+    },
+    {
+      "epoch": 1.8835164835164835,
+      "grad_norm": 0.022739041596651077,
+      "learning_rate": 0.001241042345276873,
+      "loss": 1.3818,
+      "step": 857
+    },
+    {
+      "epoch": 1.8857142857142857,
+      "grad_norm": 0.026938587427139282,
+      "learning_rate": 0.001238599348534202,
+      "loss": 1.0443,
+      "step": 858
+    },
+    {
+      "epoch": 1.8879120879120879,
+      "grad_norm": 0.02215585485100746,
+      "learning_rate": 0.001236156351791531,
+      "loss": 1.1341,
+      "step": 859
+    },
+    {
+      "epoch": 1.89010989010989,
+      "grad_norm": 0.022181663662195206,
+      "learning_rate": 0.00123371335504886,
+      "loss": 1.3252,
+      "step": 860
+    },
+    {
+      "epoch": 1.8923076923076922,
+      "grad_norm": 0.014212338253855705,
+      "learning_rate": 0.001231270358306189,
+      "loss": 1.2906,
+      "step": 861
+    },
+    {
+      "epoch": 1.8945054945054944,
+      "grad_norm": 0.01381717249751091,
+      "learning_rate": 0.0012288273615635179,
+      "loss": 1.107,
+      "step": 862
+    },
+    {
+      "epoch": 1.8967032967032966,
+      "grad_norm": 0.013119513168931007,
+      "learning_rate": 0.001226384364820847,
+      "loss": 1.006,
+      "step": 863
+    },
+    {
+      "epoch": 1.8989010989010988,
+      "grad_norm": 0.019712921231985092,
+      "learning_rate": 0.0012239413680781759,
+      "loss": 1.2199,
+      "step": 864
+    },
+    {
+      "epoch": 1.901098901098901,
+      "grad_norm": 0.0124557726085186,
+      "learning_rate": 0.001221498371335505,
+      "loss": 1.3084,
+      "step": 865
+    },
+    {
+      "epoch": 1.9032967032967032,
+      "grad_norm": 0.01526841800659895,
+      "learning_rate": 0.0012190553745928339,
+      "loss": 1.0554,
+      "step": 866
+    },
+    {
+      "epoch": 1.9054945054945054,
+      "grad_norm": 0.018715258687734604,
+      "learning_rate": 0.001216612377850163,
+      "loss": 1.2341,
+      "step": 867
+    },
+    {
+      "epoch": 1.9076923076923076,
+      "grad_norm": 0.02176375687122345,
+      "learning_rate": 0.0012141693811074919,
+      "loss": 1.2893,
+      "step": 868
+    },
+    {
+      "epoch": 1.9098901098901098,
+      "grad_norm": 0.017871743068099022,
+      "learning_rate": 0.001211726384364821,
+      "loss": 1.1321,
+      "step": 869
+    },
+    {
+      "epoch": 1.912087912087912,
+      "grad_norm": 0.019207222387194633,
+      "learning_rate": 0.0012092833876221499,
+      "loss": 1.2799,
+      "step": 870
+    },
+    {
+      "epoch": 1.9142857142857141,
+      "grad_norm": 0.015295089222490788,
+      "learning_rate": 0.0012068403908794788,
+      "loss": 1.0321,
+      "step": 871
+    },
+    {
+      "epoch": 1.9164835164835163,
+      "grad_norm": 0.01687150076031685,
+      "learning_rate": 0.0012043973941368079,
+      "loss": 1.2894,
+      "step": 872
+    },
+    {
+      "epoch": 1.9186813186813185,
+      "grad_norm": 0.01425098441541195,
+      "learning_rate": 0.0012019543973941368,
+      "loss": 1.4013,
+      "step": 873
+    },
+    {
+      "epoch": 1.9208791208791207,
+      "grad_norm": 0.01962149143218994,
+      "learning_rate": 0.0011995114006514659,
+      "loss": 1.1344,
+      "step": 874
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "grad_norm": 0.01427402999252081,
+      "learning_rate": 0.0011970684039087948,
+      "loss": 1.0857,
+      "step": 875
+    },
+    {
+      "epoch": 1.9252747252747253,
+      "grad_norm": 0.018701909109950066,
+      "learning_rate": 0.0011946254071661239,
+      "loss": 1.2489,
+      "step": 876
+    },
+    {
+      "epoch": 1.9274725274725275,
+      "grad_norm": 0.030532198026776314,
+      "learning_rate": 0.0011921824104234528,
+      "loss": 1.185,
+      "step": 877
+    },
+    {
+      "epoch": 1.9296703296703297,
+      "grad_norm": 0.01818496361374855,
+      "learning_rate": 0.001189739413680782,
+      "loss": 1.0666,
+      "step": 878
+    },
+    {
+      "epoch": 1.9318681318681319,
+      "grad_norm": 0.019826194271445274,
+      "learning_rate": 0.0011872964169381108,
+      "loss": 1.1458,
+      "step": 879
+    },
+    {
+      "epoch": 1.934065934065934,
+      "grad_norm": 0.019886046648025513,
+      "learning_rate": 0.00118485342019544,
+      "loss": 1.353,
+      "step": 880
+    },
+    {
+      "epoch": 1.9362637362637363,
+      "grad_norm": 0.01921088434755802,
+      "learning_rate": 0.0011824104234527688,
+      "loss": 1.124,
+      "step": 881
+    },
+    {
+      "epoch": 1.9384615384615385,
+      "grad_norm": 0.017668234184384346,
+      "learning_rate": 0.0011799674267100977,
+      "loss": 1.1711,
+      "step": 882
+    },
+    {
+      "epoch": 1.9406593406593406,
+      "grad_norm": 0.019154686480760574,
+      "learning_rate": 0.0011775244299674268,
+      "loss": 1.1891,
+      "step": 883
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.020730633288621902,
+      "learning_rate": 0.0011750814332247557,
+      "loss": 1.0312,
+      "step": 884
+    },
+    {
+      "epoch": 1.945054945054945,
+      "grad_norm": 0.02451159805059433,
+      "learning_rate": 0.0011726384364820848,
+      "loss": 1.2438,
+      "step": 885
+    },
+    {
+      "epoch": 1.9472527472527472,
+      "grad_norm": 0.01805088110268116,
+      "learning_rate": 0.0011701954397394137,
+      "loss": 1.1034,
+      "step": 886
+    },
+    {
+      "epoch": 1.9494505494505494,
+      "grad_norm": 0.01628929376602173,
+      "learning_rate": 0.0011677524429967428,
+      "loss": 1.1159,
+      "step": 887
+    },
+    {
+      "epoch": 1.9516483516483516,
+      "grad_norm": 0.018993090838193893,
+      "learning_rate": 0.0011653094462540717,
+      "loss": 1.1267,
+      "step": 888
+    },
+    {
+      "epoch": 1.953846153846154,
+      "grad_norm": 0.013723342679440975,
+      "learning_rate": 0.0011628664495114008,
+      "loss": 1.2315,
+      "step": 889
+    },
+    {
+      "epoch": 1.9560439560439562,
+      "grad_norm": 0.024077730253338814,
+      "learning_rate": 0.0011604234527687297,
+      "loss": 1.1441,
+      "step": 890
+    },
+    {
+      "epoch": 1.9582417582417584,
+      "grad_norm": 0.01715434156358242,
+      "learning_rate": 0.0011579804560260586,
+      "loss": 1.1853,
+      "step": 891
+    },
+    {
+      "epoch": 1.9604395604395606,
+      "grad_norm": 0.020522549748420715,
+      "learning_rate": 0.0011555374592833877,
+      "loss": 1.1933,
+      "step": 892
+    },
+    {
+      "epoch": 1.9626373626373628,
+      "grad_norm": 0.01317064929753542,
+      "learning_rate": 0.0011530944625407166,
+      "loss": 1.2317,
+      "step": 893
+    },
+    {
+      "epoch": 1.964835164835165,
+      "grad_norm": 0.01973244920372963,
+      "learning_rate": 0.0011506514657980457,
+      "loss": 1.0885,
+      "step": 894
+    },
+    {
+      "epoch": 1.9670329670329672,
+      "grad_norm": 0.016848191618919373,
+      "learning_rate": 0.0011482084690553746,
+      "loss": 1.0095,
+      "step": 895
+    },
+    {
+      "epoch": 1.9692307692307693,
+      "grad_norm": 0.021867195144295692,
+      "learning_rate": 0.0011457654723127037,
+      "loss": 1.2995,
+      "step": 896
+    },
+    {
+      "epoch": 1.9714285714285715,
+      "grad_norm": 0.019006947055459023,
+      "learning_rate": 0.0011433224755700326,
+      "loss": 1.2752,
+      "step": 897
+    },
+    {
+      "epoch": 1.9736263736263737,
+      "grad_norm": 0.02720695175230503,
+      "learning_rate": 0.0011408794788273617,
+      "loss": 1.4799,
+      "step": 898
+    },
+    {
+      "epoch": 1.975824175824176,
+      "grad_norm": 0.02766169235110283,
+      "learning_rate": 0.0011384364820846906,
+      "loss": 1.1028,
+      "step": 899
+    },
+    {
+      "epoch": 1.978021978021978,
+      "grad_norm": 0.023604700341820717,
+      "learning_rate": 0.0011359934853420195,
+      "loss": 1.4826,
+      "step": 900
+    },
+    {
+      "epoch": 1.9802197802197803,
+      "grad_norm": 0.017260747030377388,
+      "learning_rate": 0.0011335504885993486,
+      "loss": 1.1484,
+      "step": 901
+    },
+    {
+      "epoch": 1.9824175824175825,
+      "grad_norm": 0.020022496581077576,
+      "learning_rate": 0.0011311074918566775,
+      "loss": 1.1064,
+      "step": 902
+    },
+    {
+      "epoch": 1.9846153846153847,
+      "grad_norm": 0.017829999327659607,
+      "learning_rate": 0.0011286644951140066,
+      "loss": 1.1668,
+      "step": 903
+    },
+    {
+      "epoch": 1.9868131868131869,
+      "grad_norm": 0.018766185268759727,
+      "learning_rate": 0.0011262214983713355,
+      "loss": 1.2784,
+      "step": 904
+    },
+    {
+      "epoch": 1.989010989010989,
+      "grad_norm": 0.019597548991441727,
+      "learning_rate": 0.0011237785016286646,
+      "loss": 1.1796,
+      "step": 905
+    },
+    {
+      "epoch": 1.9912087912087912,
+      "grad_norm": 0.014065897092223167,
+      "learning_rate": 0.0011213355048859935,
+      "loss": 1.0838,
+      "step": 906
+    },
+    {
+      "epoch": 1.9934065934065934,
+      "grad_norm": 0.015758566558361053,
+      "learning_rate": 0.0011188925081433226,
+      "loss": 1.1525,
+      "step": 907
+    },
+    {
+      "epoch": 1.9956043956043956,
+      "grad_norm": 0.013296855613589287,
+      "learning_rate": 0.0011164495114006515,
+      "loss": 1.2282,
+      "step": 908
+    },
+    {
+      "epoch": 1.9978021978021978,
+      "grad_norm": 0.020694365724921227,
+      "learning_rate": 0.0011140065146579806,
+      "loss": 1.126,
+      "step": 909
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.021740898489952087,
+      "learning_rate": 0.0011115635179153095,
+      "loss": 1.1388,
+      "step": 910
+    },
+    {
+      "epoch": 2.002197802197802,
+      "grad_norm": 0.015985773876309395,
+      "learning_rate": 0.0011091205211726384,
+      "loss": 0.9864,
+      "step": 911
+    },
+    {
+      "epoch": 2.0043956043956044,
+      "grad_norm": 0.017542334273457527,
+      "learning_rate": 0.0011066775244299675,
+      "loss": 1.0761,
+      "step": 912
+    },
+    {
+      "epoch": 2.0065934065934066,
+      "grad_norm": 0.01890396513044834,
+      "learning_rate": 0.0011042345276872964,
+      "loss": 1.092,
+      "step": 913
+    },
+    {
+      "epoch": 2.0087912087912088,
+      "grad_norm": 0.015976088121533394,
+      "learning_rate": 0.0011017915309446255,
+      "loss": 1.0527,
+      "step": 914
+    },
+    {
+      "epoch": 2.010989010989011,
+      "grad_norm": 0.023014573380351067,
+      "learning_rate": 0.0010993485342019544,
+      "loss": 1.0045,
+      "step": 915
+    },
+    {
+      "epoch": 2.013186813186813,
+      "grad_norm": 0.014600986614823341,
+      "learning_rate": 0.0010969055374592835,
+      "loss": 1.08,
+      "step": 916
+    },
+    {
+      "epoch": 2.0153846153846153,
+      "grad_norm": 0.017879046499729156,
+      "learning_rate": 0.0010944625407166124,
+      "loss": 1.2091,
+      "step": 917
+    },
+    {
+      "epoch": 2.0175824175824175,
+      "grad_norm": 0.02753550559282303,
+      "learning_rate": 0.0010920195439739415,
+      "loss": 1.1158,
+      "step": 918
+    },
+    {
+      "epoch": 2.0197802197802197,
+      "grad_norm": 0.021567443385720253,
+      "learning_rate": 0.0010895765472312704,
+      "loss": 1.0905,
+      "step": 919
+    },
+    {
+      "epoch": 2.021978021978022,
+      "grad_norm": 0.041532788425683975,
+      "learning_rate": 0.0010871335504885993,
+      "loss": 1.2704,
+      "step": 920
+    },
+    {
+      "epoch": 2.024175824175824,
+      "grad_norm": 0.020661059767007828,
+      "learning_rate": 0.0010846905537459284,
+      "loss": 1.23,
+      "step": 921
+    },
+    {
+      "epoch": 2.0263736263736263,
+      "grad_norm": 0.018856240436434746,
+      "learning_rate": 0.0010822475570032573,
+      "loss": 1.0106,
+      "step": 922
+    },
+    {
+      "epoch": 2.0285714285714285,
+      "grad_norm": 0.021778056398034096,
+      "learning_rate": 0.0010798045602605864,
+      "loss": 1.2042,
+      "step": 923
+    },
+    {
+      "epoch": 2.0307692307692307,
+      "grad_norm": 0.017993081361055374,
+      "learning_rate": 0.0010773615635179153,
+      "loss": 1.0075,
+      "step": 924
+    },
+    {
+      "epoch": 2.032967032967033,
+      "grad_norm": 0.015567179769277573,
+      "learning_rate": 0.0010749185667752444,
+      "loss": 1.2155,
+      "step": 925
+    },
+    {
+      "epoch": 2.035164835164835,
+      "grad_norm": 0.021821482107043266,
+      "learning_rate": 0.0010724755700325733,
+      "loss": 1.2265,
+      "step": 926
+    },
+    {
+      "epoch": 2.0373626373626372,
+      "grad_norm": 0.019789891317486763,
+      "learning_rate": 0.0010700325732899024,
+      "loss": 1.1475,
+      "step": 927
+    },
+    {
+      "epoch": 2.0395604395604394,
+      "grad_norm": 0.0235284510999918,
+      "learning_rate": 0.0010675895765472313,
+      "loss": 1.2523,
+      "step": 928
+    },
+    {
+      "epoch": 2.0417582417582416,
+      "grad_norm": 0.016807036474347115,
+      "learning_rate": 0.0010651465798045602,
+      "loss": 0.9792,
+      "step": 929
+    },
+    {
+      "epoch": 2.043956043956044,
+      "grad_norm": 0.027279330417513847,
+      "learning_rate": 0.0010627035830618893,
+      "loss": 1.1641,
+      "step": 930
+    },
+    {
+      "epoch": 2.046153846153846,
+      "grad_norm": 0.01792175881564617,
+      "learning_rate": 0.0010602605863192182,
+      "loss": 1.1772,
+      "step": 931
+    },
+    {
+      "epoch": 2.048351648351648,
+      "grad_norm": 0.018325671553611755,
+      "learning_rate": 0.0010578175895765473,
+      "loss": 1.0031,
+      "step": 932
+    },
+    {
+      "epoch": 2.0505494505494504,
+      "grad_norm": 0.018724532797932625,
+      "learning_rate": 0.0010553745928338762,
+      "loss": 1.2929,
+      "step": 933
+    },
+    {
+      "epoch": 2.0527472527472526,
+      "grad_norm": 0.018978089094161987,
+      "learning_rate": 0.0010529315960912053,
+      "loss": 0.9517,
+      "step": 934
+    },
+    {
+      "epoch": 2.0549450549450547,
+      "grad_norm": 0.018275558948516846,
+      "learning_rate": 0.0010504885993485342,
+      "loss": 1.2158,
+      "step": 935
+    },
+    {
+      "epoch": 2.057142857142857,
+      "grad_norm": 0.016714967787265778,
+      "learning_rate": 0.0010480456026058633,
+      "loss": 1.1821,
+      "step": 936
+    },
+    {
+      "epoch": 2.059340659340659,
+      "grad_norm": 0.016149839386343956,
+      "learning_rate": 0.0010456026058631922,
+      "loss": 1.0688,
+      "step": 937
+    },
+    {
+      "epoch": 2.0615384615384613,
+      "grad_norm": 0.02975514717400074,
+      "learning_rate": 0.0010431596091205213,
+      "loss": 1.2164,
+      "step": 938
+    },
+    {
+      "epoch": 2.0637362637362635,
+      "grad_norm": 0.02220059372484684,
+      "learning_rate": 0.0010407166123778502,
+      "loss": 1.257,
+      "step": 939
+    },
+    {
+      "epoch": 2.065934065934066,
+      "grad_norm": 0.022131310775876045,
+      "learning_rate": 0.001038273615635179,
+      "loss": 1.042,
+      "step": 940
+    },
+    {
+      "epoch": 2.0681318681318683,
+      "grad_norm": 0.011823747307062149,
+      "learning_rate": 0.0010358306188925082,
+      "loss": 0.9709,
+      "step": 941
+    },
+    {
+      "epoch": 2.0703296703296705,
+      "grad_norm": 0.028340544551610947,
+      "learning_rate": 0.001033387622149837,
+      "loss": 1.2683,
+      "step": 942
+    },
+    {
+      "epoch": 2.0725274725274727,
+      "grad_norm": 0.02090267278254032,
+      "learning_rate": 0.0010309446254071662,
+      "loss": 1.3056,
+      "step": 943
+    },
+    {
+      "epoch": 2.074725274725275,
+      "grad_norm": 0.02798914723098278,
+      "learning_rate": 0.001028501628664495,
+      "loss": 1.1822,
+      "step": 944
+    },
+    {
+      "epoch": 2.076923076923077,
+      "grad_norm": 0.01783566363155842,
+      "learning_rate": 0.0010260586319218242,
+      "loss": 1.2385,
+      "step": 945
+    },
+    {
+      "epoch": 2.0791208791208793,
+      "grad_norm": 0.02041742019355297,
+      "learning_rate": 0.001023615635179153,
+      "loss": 1.2341,
+      "step": 946
+    },
+    {
+      "epoch": 2.0813186813186815,
+      "grad_norm": 0.015458942390978336,
+      "learning_rate": 0.0010211726384364822,
+      "loss": 1.1799,
+      "step": 947
+    },
+    {
+      "epoch": 2.0835164835164837,
+      "grad_norm": 0.02003684639930725,
+      "learning_rate": 0.001018729641693811,
+      "loss": 1.0475,
+      "step": 948
+    },
+    {
+      "epoch": 2.085714285714286,
+      "grad_norm": 0.02523563615977764,
+      "learning_rate": 0.00101628664495114,
+      "loss": 1.1299,
+      "step": 949
+    },
+    {
+      "epoch": 2.087912087912088,
+      "grad_norm": 0.02131020464003086,
+      "learning_rate": 0.001013843648208469,
+      "loss": 0.9519,
+      "step": 950
+    },
+    {
+      "epoch": 2.0901098901098902,
+      "grad_norm": 0.031957827508449554,
+      "learning_rate": 0.001011400651465798,
+      "loss": 1.1708,
+      "step": 951
+    },
+    {
+      "epoch": 2.0923076923076924,
+      "grad_norm": 0.015571853145956993,
+      "learning_rate": 0.001008957654723127,
+      "loss": 1.0664,
+      "step": 952
+    },
+    {
+      "epoch": 2.0945054945054946,
+      "grad_norm": 0.012164526619017124,
+      "learning_rate": 0.001006514657980456,
+      "loss": 1.094,
+      "step": 953
+    },
+    {
+      "epoch": 2.096703296703297,
+      "grad_norm": 0.01840994507074356,
+      "learning_rate": 0.001004071661237785,
+      "loss": 1.0969,
+      "step": 954
+    },
+    {
+      "epoch": 2.098901098901099,
+      "grad_norm": 0.019973980262875557,
+      "learning_rate": 0.001001628664495114,
+      "loss": 1.0811,
+      "step": 955
+    },
+    {
+      "epoch": 2.101098901098901,
+      "grad_norm": 0.019167479127645493,
+      "learning_rate": 0.000999185667752443,
+      "loss": 1.2261,
+      "step": 956
+    },
+    {
+      "epoch": 2.1032967032967034,
+      "grad_norm": 0.027639804407954216,
+      "learning_rate": 0.000996742671009772,
+      "loss": 1.2617,
+      "step": 957
+    },
+    {
+      "epoch": 2.1054945054945056,
+      "grad_norm": 0.016373364254832268,
+      "learning_rate": 0.0009942996742671008,
+      "loss": 1.0638,
+      "step": 958
+    },
+    {
+      "epoch": 2.1076923076923078,
+      "grad_norm": 0.018773552030324936,
+      "learning_rate": 0.00099185667752443,
+      "loss": 1.095,
+      "step": 959
+    },
+    {
+      "epoch": 2.10989010989011,
+      "grad_norm": 0.026365388184785843,
+      "learning_rate": 0.0009894136807817588,
+      "loss": 1.2694,
+      "step": 960
+    },
+    {
+      "epoch": 2.112087912087912,
+      "grad_norm": 0.020940013229846954,
+      "learning_rate": 0.000986970684039088,
+      "loss": 1.0801,
+      "step": 961
+    },
+    {
+      "epoch": 2.1142857142857143,
+      "grad_norm": 0.01942221261560917,
+      "learning_rate": 0.0009845276872964168,
+      "loss": 1.139,
+      "step": 962
+    },
+    {
+      "epoch": 2.1164835164835165,
+      "grad_norm": 0.015033043920993805,
+      "learning_rate": 0.000982084690553746,
+      "loss": 1.2243,
+      "step": 963
+    },
+    {
+      "epoch": 2.1186813186813187,
+      "grad_norm": 0.05060616880655289,
+      "learning_rate": 0.0009796416938110748,
+      "loss": 1.2195,
+      "step": 964
+    },
+    {
+      "epoch": 2.120879120879121,
+      "grad_norm": 0.01661854423582554,
+      "learning_rate": 0.000977198697068404,
+      "loss": 1.0418,
+      "step": 965
+    },
+    {
+      "epoch": 2.123076923076923,
+      "grad_norm": 0.017443183809518814,
+      "learning_rate": 0.0009747557003257328,
+      "loss": 1.0016,
+      "step": 966
+    },
+    {
+      "epoch": 2.1252747252747253,
+      "grad_norm": 0.018947936594486237,
+      "learning_rate": 0.0009723127035830618,
+      "loss": 1.3122,
+      "step": 967
+    },
+    {
+      "epoch": 2.1274725274725275,
+      "grad_norm": 0.013806606642901897,
+      "learning_rate": 0.0009698697068403908,
+      "loss": 0.9424,
+      "step": 968
+    },
+    {
+      "epoch": 2.1296703296703297,
+      "grad_norm": 0.0235359575599432,
+      "learning_rate": 0.0009674267100977198,
+      "loss": 1.1659,
+      "step": 969
+    },
+    {
+      "epoch": 2.131868131868132,
+      "grad_norm": 0.017178446054458618,
+      "learning_rate": 0.0009649837133550488,
+      "loss": 1.1362,
+      "step": 970
+    },
+    {
+      "epoch": 2.134065934065934,
+      "grad_norm": 0.020924653857946396,
+      "learning_rate": 0.0009625407166123778,
+      "loss": 1.0728,
+      "step": 971
+    },
+    {
+      "epoch": 2.1362637362637362,
+      "grad_norm": 0.01945706084370613,
+      "learning_rate": 0.0009600977198697067,
+      "loss": 1.2516,
+      "step": 972
+    },
+    {
+      "epoch": 2.1384615384615384,
+      "grad_norm": 0.027772869914770126,
+      "learning_rate": 0.0009576547231270357,
+      "loss": 1.0851,
+      "step": 973
+    },
+    {
+      "epoch": 2.1406593406593406,
+      "grad_norm": 0.01347445696592331,
+      "learning_rate": 0.0009552117263843647,
+      "loss": 0.9442,
+      "step": 974
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.02134643867611885,
+      "learning_rate": 0.0009527687296416937,
+      "loss": 1.3108,
+      "step": 975
+    },
+    {
+      "epoch": 2.145054945054945,
+      "grad_norm": 0.01867590844631195,
+      "learning_rate": 0.0009503257328990227,
+      "loss": 1.2942,
+      "step": 976
+    },
+    {
+      "epoch": 2.147252747252747,
+      "grad_norm": 0.01993655227124691,
+      "learning_rate": 0.0009478827361563517,
+      "loss": 1.202,
+      "step": 977
+    },
+    {
+      "epoch": 2.1494505494505494,
+      "grad_norm": 0.01862264983355999,
+      "learning_rate": 0.0009454397394136807,
+      "loss": 1.2061,
+      "step": 978
+    },
+    {
+      "epoch": 2.1516483516483516,
+      "grad_norm": 0.017521658912301064,
+      "learning_rate": 0.0009429967426710097,
+      "loss": 1.0542,
+      "step": 979
+    },
+    {
+      "epoch": 2.1538461538461537,
+      "grad_norm": 0.14997851848602295,
+      "learning_rate": 0.0009405537459283387,
+      "loss": 1.0695,
+      "step": 980
+    },
+    {
+      "epoch": 2.156043956043956,
+      "grad_norm": 0.02277497574687004,
+      "learning_rate": 0.0009381107491856677,
+      "loss": 1.1274,
+      "step": 981
+    },
+    {
+      "epoch": 2.158241758241758,
+      "grad_norm": 0.016072440892457962,
+      "learning_rate": 0.0009356677524429969,
+      "loss": 1.1174,
+      "step": 982
+    },
+    {
+      "epoch": 2.1604395604395603,
+      "grad_norm": 0.019632680341601372,
+      "learning_rate": 0.0009332247557003259,
+      "loss": 1.0602,
+      "step": 983
+    },
+    {
+      "epoch": 2.1626373626373625,
+      "grad_norm": 0.01908484473824501,
+      "learning_rate": 0.0009307817589576549,
+      "loss": 1.1257,
+      "step": 984
+    },
+    {
+      "epoch": 2.1648351648351647,
+      "grad_norm": 0.09126205742359161,
+      "learning_rate": 0.0009283387622149839,
+      "loss": 1.1271,
+      "step": 985
+    },
+    {
+      "epoch": 2.167032967032967,
+      "grad_norm": 0.01970268227159977,
+      "learning_rate": 0.0009258957654723129,
+      "loss": 0.9955,
+      "step": 986
+    },
+    {
+      "epoch": 2.169230769230769,
+      "grad_norm": 0.024221209809184074,
+      "learning_rate": 0.0009234527687296417,
+      "loss": 1.0863,
+      "step": 987
+    },
+    {
+      "epoch": 2.1714285714285713,
+      "grad_norm": 0.020160816609859467,
+      "learning_rate": 0.0009210097719869707,
+      "loss": 1.1382,
+      "step": 988
+    },
+    {
+      "epoch": 2.1736263736263735,
+      "grad_norm": 0.011731104925274849,
+      "learning_rate": 0.0009185667752442997,
+      "loss": 1.2569,
+      "step": 989
+    },
+    {
+      "epoch": 2.1758241758241756,
+      "grad_norm": 0.016863910481333733,
+      "learning_rate": 0.0009161237785016287,
+      "loss": 1.1988,
+      "step": 990
+    },
+    {
+      "epoch": 2.178021978021978,
+      "grad_norm": 0.025215644389390945,
+      "learning_rate": 0.0009136807817589577,
+      "loss": 1.1645,
+      "step": 991
+    },
+    {
+      "epoch": 2.18021978021978,
+      "grad_norm": 0.016928907483816147,
+      "learning_rate": 0.0009112377850162867,
+      "loss": 1.2163,
+      "step": 992
+    },
+    {
+      "epoch": 2.182417582417582,
+      "grad_norm": 0.011591528542339802,
+      "learning_rate": 0.0009087947882736157,
+      "loss": 1.2138,
+      "step": 993
+    },
+    {
+      "epoch": 2.184615384615385,
+      "grad_norm": 0.0187345240265131,
+      "learning_rate": 0.0009063517915309447,
+      "loss": 1.1811,
+      "step": 994
+    },
+    {
+      "epoch": 2.186813186813187,
+      "grad_norm": 0.025712179020047188,
+      "learning_rate": 0.0009039087947882737,
+      "loss": 0.9833,
+      "step": 995
+    },
+    {
+      "epoch": 2.1890109890109892,
+      "grad_norm": 0.017650732770562172,
+      "learning_rate": 0.0009014657980456026,
+      "loss": 1.1694,
+      "step": 996
+    },
+    {
+      "epoch": 2.1912087912087914,
+      "grad_norm": 0.01903453655540943,
+      "learning_rate": 0.0008990228013029316,
+      "loss": 1.3293,
+      "step": 997
+    },
+    {
+      "epoch": 2.1934065934065936,
+      "grad_norm": 0.024799736216664314,
+      "learning_rate": 0.0008965798045602606,
+      "loss": 1.1887,
+      "step": 998
+    },
+    {
+      "epoch": 2.195604395604396,
+      "grad_norm": 0.015706755220890045,
+      "learning_rate": 0.0008941368078175896,
+      "loss": 1.2347,
+      "step": 999
+    },
+    {
+      "epoch": 2.197802197802198,
+      "grad_norm": 0.019947845488786697,
+      "learning_rate": 0.0008916938110749186,
+      "loss": 1.3678,
+      "step": 1000
+    },
+    {
+      "epoch": 2.2,
+      "grad_norm": 0.01747577078640461,
+      "learning_rate": 0.0008892508143322476,
+      "loss": 1.1498,
+      "step": 1001
+    },
+    {
+      "epoch": 2.2021978021978024,
+      "grad_norm": 0.01697716861963272,
+      "learning_rate": 0.0008868078175895766,
+      "loss": 1.0266,
+      "step": 1002
+    },
+    {
+      "epoch": 2.2043956043956046,
+      "grad_norm": 0.012452851980924606,
+      "learning_rate": 0.0008843648208469056,
+      "loss": 1.1046,
+      "step": 1003
+    },
+    {
+      "epoch": 2.2065934065934067,
+      "grad_norm": 0.019205965101718903,
+      "learning_rate": 0.0008819218241042346,
+      "loss": 0.9209,
+      "step": 1004
+    },
+    {
+      "epoch": 2.208791208791209,
+      "grad_norm": 0.017458468675613403,
+      "learning_rate": 0.0008794788273615636,
+      "loss": 1.3788,
+      "step": 1005
+    },
+    {
+      "epoch": 2.210989010989011,
+      "grad_norm": 0.01975560374557972,
+      "learning_rate": 0.0008770358306188925,
+      "loss": 1.2747,
+      "step": 1006
+    },
+    {
+      "epoch": 2.2131868131868133,
+      "grad_norm": 0.018666088581085205,
+      "learning_rate": 0.0008745928338762215,
+      "loss": 1.1591,
+      "step": 1007
+    },
+    {
+      "epoch": 2.2153846153846155,
+      "grad_norm": 0.020304059609770775,
+      "learning_rate": 0.0008721498371335505,
+      "loss": 1.1949,
+      "step": 1008
+    },
+    {
+      "epoch": 2.2175824175824177,
+      "grad_norm": 0.019182439893484116,
+      "learning_rate": 0.0008697068403908795,
+      "loss": 1.2841,
+      "step": 1009
+    },
+    {
+      "epoch": 2.21978021978022,
+      "grad_norm": 0.022571206092834473,
+      "learning_rate": 0.0008672638436482085,
+      "loss": 1.0325,
+      "step": 1010
+    },
+    {
+      "epoch": 2.221978021978022,
+      "grad_norm": 0.012984970584511757,
+      "learning_rate": 0.0008648208469055375,
+      "loss": 1.025,
+      "step": 1011
+    },
+    {
+      "epoch": 2.2241758241758243,
+      "grad_norm": 0.021859358996152878,
+      "learning_rate": 0.0008623778501628665,
+      "loss": 1.1995,
+      "step": 1012
+    },
+    {
+      "epoch": 2.2263736263736265,
+      "grad_norm": 0.02292865701019764,
+      "learning_rate": 0.0008599348534201955,
+      "loss": 1.3006,
+      "step": 1013
+    },
+    {
+      "epoch": 2.2285714285714286,
+      "grad_norm": 0.020875519141554832,
+      "learning_rate": 0.0008574918566775245,
+      "loss": 1.4463,
+      "step": 1014
+    },
+    {
+      "epoch": 2.230769230769231,
+      "grad_norm": 0.030453646555542946,
+      "learning_rate": 0.0008550488599348534,
+      "loss": 1.3366,
+      "step": 1015
+    },
+    {
+      "epoch": 2.232967032967033,
+      "grad_norm": 0.013538055121898651,
+      "learning_rate": 0.0008526058631921824,
+      "loss": 1.1517,
+      "step": 1016
+    },
+    {
+      "epoch": 2.235164835164835,
+      "grad_norm": 0.03148024156689644,
+      "learning_rate": 0.0008501628664495114,
+      "loss": 0.9753,
+      "step": 1017
+    },
+    {
+      "epoch": 2.2373626373626374,
+      "grad_norm": 0.017705893144011497,
+      "learning_rate": 0.0008477198697068404,
+      "loss": 1.0008,
+      "step": 1018
+    },
+    {
+      "epoch": 2.2395604395604396,
+      "grad_norm": 0.03420194983482361,
+      "learning_rate": 0.0008452768729641694,
+      "loss": 1.079,
+      "step": 1019
+    },
+    {
+      "epoch": 2.241758241758242,
+      "grad_norm": 0.017806069925427437,
+      "learning_rate": 0.0008428338762214984,
+      "loss": 0.988,
+      "step": 1020
+    },
+    {
+      "epoch": 2.243956043956044,
+      "grad_norm": 0.019936196506023407,
+      "learning_rate": 0.0008403908794788274,
+      "loss": 1.2404,
+      "step": 1021
+    },
+    {
+      "epoch": 2.246153846153846,
+      "grad_norm": 0.02254304103553295,
+      "learning_rate": 0.0008379478827361564,
+      "loss": 1.1192,
+      "step": 1022
+    },
+    {
+      "epoch": 2.2483516483516484,
+      "grad_norm": 0.01460746768862009,
+      "learning_rate": 0.0008355048859934854,
+      "loss": 1.1115,
+      "step": 1023
+    },
+    {
+      "epoch": 2.2505494505494505,
+      "grad_norm": 0.017172737047076225,
+      "learning_rate": 0.0008330618892508144,
+      "loss": 1.1388,
+      "step": 1024
+    },
+    {
+      "epoch": 2.2527472527472527,
+      "grad_norm": 0.015675432980060577,
+      "learning_rate": 0.0008306188925081433,
+      "loss": 1.0202,
+      "step": 1025
+    },
+    {
+      "epoch": 2.254945054945055,
+      "grad_norm": 0.024180566892027855,
+      "learning_rate": 0.0008281758957654723,
+      "loss": 1.1281,
+      "step": 1026
+    },
+    {
+      "epoch": 2.257142857142857,
+      "grad_norm": 0.017173459753394127,
+      "learning_rate": 0.0008257328990228013,
+      "loss": 1.131,
+      "step": 1027
+    },
+    {
+      "epoch": 2.2593406593406593,
+      "grad_norm": 0.01447053998708725,
+      "learning_rate": 0.0008232899022801303,
+      "loss": 1.1188,
+      "step": 1028
+    },
+    {
+      "epoch": 2.2615384615384615,
+      "grad_norm": 0.015481141395866871,
+      "learning_rate": 0.0008208469055374593,
+      "loss": 0.9977,
+      "step": 1029
+    },
+    {
+      "epoch": 2.2637362637362637,
+      "grad_norm": 0.028231805190443993,
+      "learning_rate": 0.0008184039087947883,
+      "loss": 1.4557,
+      "step": 1030
+    },
+    {
+      "epoch": 2.265934065934066,
+      "grad_norm": 0.020987428724765778,
+      "learning_rate": 0.0008159609120521173,
+      "loss": 1.0213,
+      "step": 1031
+    },
+    {
+      "epoch": 2.268131868131868,
+      "grad_norm": 0.013762211427092552,
+      "learning_rate": 0.0008135179153094463,
+      "loss": 1.1412,
+      "step": 1032
+    },
+    {
+      "epoch": 2.2703296703296703,
+      "grad_norm": 0.01937873475253582,
+      "learning_rate": 0.0008110749185667753,
+      "loss": 1.2743,
+      "step": 1033
+    },
+    {
+      "epoch": 2.2725274725274724,
+      "grad_norm": 0.01655028760433197,
+      "learning_rate": 0.0008086319218241043,
+      "loss": 1.1157,
+      "step": 1034
+    },
+    {
+      "epoch": 2.2747252747252746,
+      "grad_norm": 0.019710851833224297,
+      "learning_rate": 0.0008061889250814332,
+      "loss": 1.3361,
+      "step": 1035
+    },
+    {
+      "epoch": 2.276923076923077,
+      "grad_norm": 0.02316323108971119,
+      "learning_rate": 0.0008037459283387622,
+      "loss": 1.2362,
+      "step": 1036
+    },
+    {
+      "epoch": 2.279120879120879,
+      "grad_norm": 0.016647597774863243,
+      "learning_rate": 0.0008013029315960912,
+      "loss": 1.0208,
+      "step": 1037
+    },
+    {
+      "epoch": 2.281318681318681,
+      "grad_norm": 0.022860154509544373,
+      "learning_rate": 0.0007988599348534202,
+      "loss": 1.1466,
+      "step": 1038
+    },
+    {
+      "epoch": 2.2835164835164834,
+      "grad_norm": 0.018142597749829292,
+      "learning_rate": 0.0007964169381107492,
+      "loss": 0.9089,
+      "step": 1039
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.01826503500342369,
+      "learning_rate": 0.0007939739413680782,
+      "loss": 0.9947,
+      "step": 1040
+    },
+    {
+      "epoch": 2.2879120879120878,
+      "grad_norm": 0.018456997349858284,
+      "learning_rate": 0.0007915309446254072,
+      "loss": 1.3844,
+      "step": 1041
+    },
+    {
+      "epoch": 2.29010989010989,
+      "grad_norm": 0.025633038952946663,
+      "learning_rate": 0.0007890879478827362,
+      "loss": 1.2288,
+      "step": 1042
+    },
+    {
+      "epoch": 2.292307692307692,
+      "grad_norm": 0.02109592966735363,
+      "learning_rate": 0.0007866449511400652,
+      "loss": 1.2134,
+      "step": 1043
+    },
+    {
+      "epoch": 2.2945054945054943,
+      "grad_norm": 0.01989317126572132,
+      "learning_rate": 0.0007842019543973941,
+      "loss": 1.2604,
+      "step": 1044
+    },
+    {
+      "epoch": 2.2967032967032965,
+      "grad_norm": 0.01797320879995823,
+      "learning_rate": 0.0007817589576547231,
+      "loss": 1.0806,
+      "step": 1045
+    },
+    {
+      "epoch": 2.2989010989010987,
+      "grad_norm": 0.038766615092754364,
+      "learning_rate": 0.0007793159609120521,
+      "loss": 1.5164,
+      "step": 1046
+    },
+    {
+      "epoch": 2.301098901098901,
+      "grad_norm": 0.018925610929727554,
+      "learning_rate": 0.0007768729641693811,
+      "loss": 1.2116,
+      "step": 1047
+    },
+    {
+      "epoch": 2.303296703296703,
+      "grad_norm": 0.013326489366590977,
+      "learning_rate": 0.0007744299674267101,
+      "loss": 1.1329,
+      "step": 1048
+    },
+    {
+      "epoch": 2.3054945054945053,
+      "grad_norm": 0.015729276463389397,
+      "learning_rate": 0.0007719869706840391,
+      "loss": 1.1074,
+      "step": 1049
+    },
+    {
+      "epoch": 2.3076923076923075,
+      "grad_norm": 0.01916268840432167,
+      "learning_rate": 0.0007695439739413681,
+      "loss": 1.039,
+      "step": 1050
+    },
+    {
+      "epoch": 2.3098901098901097,
+      "grad_norm": 0.0200406014919281,
+      "learning_rate": 0.0007671009771986971,
+      "loss": 1.1827,
+      "step": 1051
+    },
+    {
+      "epoch": 2.312087912087912,
+      "grad_norm": 0.017016170546412468,
+      "learning_rate": 0.0007646579804560261,
+      "loss": 1.0092,
+      "step": 1052
+    },
+    {
+      "epoch": 2.314285714285714,
+      "grad_norm": 0.02630026452243328,
+      "learning_rate": 0.0007622149837133551,
+      "loss": 1.0426,
+      "step": 1053
+    },
+    {
+      "epoch": 2.3164835164835162,
+      "grad_norm": 0.022458499297499657,
+      "learning_rate": 0.000759771986970684,
+      "loss": 1.2238,
+      "step": 1054
+    },
+    {
+      "epoch": 2.3186813186813184,
+      "grad_norm": 0.02120788022875786,
+      "learning_rate": 0.000757328990228013,
+      "loss": 1.1652,
+      "step": 1055
+    },
+    {
+      "epoch": 2.320879120879121,
+      "grad_norm": 0.018224399536848068,
+      "learning_rate": 0.000754885993485342,
+      "loss": 1.0352,
+      "step": 1056
+    },
+    {
+      "epoch": 2.3230769230769233,
+      "grad_norm": 0.01742849312722683,
+      "learning_rate": 0.000752442996742671,
+      "loss": 1.156,
+      "step": 1057
+    },
+    {
+      "epoch": 2.3252747252747255,
+      "grad_norm": 0.04613817110657692,
+      "learning_rate": 0.00075,
+      "loss": 1.1251,
+      "step": 1058
+    },
+    {
+      "epoch": 2.3274725274725276,
+      "grad_norm": 0.03896471858024597,
+      "learning_rate": 0.000747557003257329,
+      "loss": 1.0487,
+      "step": 1059
+    },
+    {
+      "epoch": 2.32967032967033,
+      "grad_norm": 0.020400717854499817,
+      "learning_rate": 0.000745114006514658,
+      "loss": 1.3357,
+      "step": 1060
+    },
+    {
+      "epoch": 2.331868131868132,
+      "grad_norm": 0.027905935421586037,
+      "learning_rate": 0.000742671009771987,
+      "loss": 1.1314,
+      "step": 1061
+    },
+    {
+      "epoch": 2.334065934065934,
+      "grad_norm": 0.01940222643315792,
+      "learning_rate": 0.000740228013029316,
+      "loss": 1.0656,
+      "step": 1062
+    },
+    {
+      "epoch": 2.3362637362637364,
+      "grad_norm": 0.015771683305501938,
+      "learning_rate": 0.0007377850162866449,
+      "loss": 1.3808,
+      "step": 1063
+    },
+    {
+      "epoch": 2.3384615384615386,
+      "grad_norm": 0.012769527733325958,
+      "learning_rate": 0.0007353420195439739,
+      "loss": 1.2289,
+      "step": 1064
+    },
+    {
+      "epoch": 2.340659340659341,
+      "grad_norm": 0.02021734043955803,
+      "learning_rate": 0.0007328990228013029,
+      "loss": 1.0901,
+      "step": 1065
+    },
+    {
+      "epoch": 2.342857142857143,
+      "grad_norm": 0.021800443530082703,
+      "learning_rate": 0.0007304560260586319,
+      "loss": 1.2761,
+      "step": 1066
+    },
+    {
+      "epoch": 2.345054945054945,
+      "grad_norm": 0.023716017603874207,
+      "learning_rate": 0.0007280130293159609,
+      "loss": 1.053,
+      "step": 1067
+    },
+    {
+      "epoch": 2.3472527472527474,
+      "grad_norm": 0.017524084076285362,
+      "learning_rate": 0.0007255700325732899,
+      "loss": 1.044,
+      "step": 1068
+    },
+    {
+      "epoch": 2.3494505494505495,
+      "grad_norm": 0.021461596712470055,
+      "learning_rate": 0.0007231270358306189,
+      "loss": 1.2012,
+      "step": 1069
+    },
+    {
+      "epoch": 2.3516483516483517,
+      "grad_norm": 0.019119570031762123,
+      "learning_rate": 0.0007206840390879479,
+      "loss": 1.1331,
+      "step": 1070
+    },
+    {
+      "epoch": 2.353846153846154,
+      "grad_norm": 0.014990235678851604,
+      "learning_rate": 0.0007182410423452769,
+      "loss": 1.1254,
+      "step": 1071
+    },
+    {
+      "epoch": 2.356043956043956,
+      "grad_norm": 0.02588587813079357,
+      "learning_rate": 0.0007157980456026059,
+      "loss": 1.135,
+      "step": 1072
+    },
+    {
+      "epoch": 2.3582417582417583,
+      "grad_norm": 0.01744583435356617,
+      "learning_rate": 0.0007133550488599348,
+      "loss": 1.2629,
+      "step": 1073
+    },
+    {
+      "epoch": 2.3604395604395605,
+      "grad_norm": 0.0167690422385931,
+      "learning_rate": 0.0007109120521172638,
+      "loss": 1.1254,
+      "step": 1074
+    },
+    {
+      "epoch": 2.3626373626373627,
+      "grad_norm": 0.01779271475970745,
+      "learning_rate": 0.0007084690553745928,
+      "loss": 0.8616,
+      "step": 1075
+    },
+    {
+      "epoch": 2.364835164835165,
+      "grad_norm": 0.017713597044348717,
+      "learning_rate": 0.0007060260586319218,
+      "loss": 1.1907,
+      "step": 1076
+    },
+    {
+      "epoch": 2.367032967032967,
+      "grad_norm": 0.022040650248527527,
+      "learning_rate": 0.0007035830618892508,
+      "loss": 1.0079,
+      "step": 1077
+    },
+    {
+      "epoch": 2.3692307692307693,
+      "grad_norm": 0.014824488200247288,
+      "learning_rate": 0.0007011400651465798,
+      "loss": 1.0933,
+      "step": 1078
+    },
+    {
+      "epoch": 2.3714285714285714,
+      "grad_norm": 0.018874352797865868,
+      "learning_rate": 0.0006986970684039088,
+      "loss": 1.184,
+      "step": 1079
+    },
+    {
+      "epoch": 2.3736263736263736,
+      "grad_norm": 0.023572541773319244,
+      "learning_rate": 0.0006962540716612378,
+      "loss": 1.0725,
+      "step": 1080
+    },
+    {
+      "epoch": 2.375824175824176,
+      "grad_norm": 0.017929786816239357,
+      "learning_rate": 0.0006938110749185668,
+      "loss": 1.0833,
+      "step": 1081
+    },
+    {
+      "epoch": 2.378021978021978,
+      "grad_norm": 0.028589854016900063,
+      "learning_rate": 0.0006913680781758957,
+      "loss": 1.2893,
+      "step": 1082
+    },
+    {
+      "epoch": 2.38021978021978,
+      "grad_norm": 0.020735429599881172,
+      "learning_rate": 0.0006889250814332247,
+      "loss": 1.1547,
+      "step": 1083
+    },
+    {
+      "epoch": 2.3824175824175824,
+      "grad_norm": 0.02688845433294773,
+      "learning_rate": 0.0006864820846905537,
+      "loss": 1.2127,
+      "step": 1084
+    },
+    {
+      "epoch": 2.3846153846153846,
+      "grad_norm": 0.016168169677257538,
+      "learning_rate": 0.0006840390879478827,
+      "loss": 0.992,
+      "step": 1085
+    },
+    {
+      "epoch": 2.3868131868131868,
+      "grad_norm": 0.024221330881118774,
+      "learning_rate": 0.0006815960912052117,
+      "loss": 1.2556,
+      "step": 1086
+    },
+    {
+      "epoch": 2.389010989010989,
+      "grad_norm": 0.028019916266202927,
+      "learning_rate": 0.0006791530944625407,
+      "loss": 1.2694,
+      "step": 1087
+    },
+    {
+      "epoch": 2.391208791208791,
+      "grad_norm": 0.019569087773561478,
+      "learning_rate": 0.0006767100977198697,
+      "loss": 1.4699,
+      "step": 1088
+    },
+    {
+      "epoch": 2.3934065934065933,
+      "grad_norm": 0.015849817544221878,
+      "learning_rate": 0.0006742671009771987,
+      "loss": 1.0758,
+      "step": 1089
+    },
+    {
+      "epoch": 2.3956043956043955,
+      "grad_norm": 0.02135786972939968,
+      "learning_rate": 0.0006718241042345277,
+      "loss": 1.0792,
+      "step": 1090
+    },
+    {
+      "epoch": 2.3978021978021977,
+      "grad_norm": 0.036413662135601044,
+      "learning_rate": 0.0006693811074918567,
+      "loss": 1.3153,
+      "step": 1091
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.02495344914495945,
+      "learning_rate": 0.0006669381107491856,
+      "loss": 1.11,
+      "step": 1092
+    },
+    {
+      "epoch": 2.402197802197802,
+      "grad_norm": 0.02176428772509098,
+      "learning_rate": 0.0006644951140065146,
+      "loss": 1.3306,
+      "step": 1093
+    },
+    {
+      "epoch": 2.4043956043956043,
+      "grad_norm": 0.02090856432914734,
+      "learning_rate": 0.0006620521172638436,
+      "loss": 1.4705,
+      "step": 1094
+    },
+    {
+      "epoch": 2.4065934065934065,
+      "grad_norm": 0.030693447217345238,
+      "learning_rate": 0.0006596091205211726,
+      "loss": 1.0606,
+      "step": 1095
+    },
+    {
+      "epoch": 2.4087912087912087,
+      "grad_norm": 0.03181895986199379,
+      "learning_rate": 0.0006571661237785016,
+      "loss": 1.302,
+      "step": 1096
+    },
+    {
+      "epoch": 2.410989010989011,
+      "grad_norm": 0.017742475494742393,
+      "learning_rate": 0.0006547231270358307,
+      "loss": 1.2135,
+      "step": 1097
+    },
+    {
+      "epoch": 2.413186813186813,
+      "grad_norm": 0.014439833350479603,
+      "learning_rate": 0.0006522801302931597,
+      "loss": 0.9945,
+      "step": 1098
+    },
+    {
+      "epoch": 2.4153846153846152,
+      "grad_norm": 0.016706665977835655,
+      "learning_rate": 0.0006498371335504886,
+      "loss": 1.0518,
+      "step": 1099
+    },
+    {
+      "epoch": 2.4175824175824174,
+      "grad_norm": 0.01884816586971283,
+      "learning_rate": 0.0006473941368078176,
+      "loss": 1.0885,
+      "step": 1100
+    },
+    {
+      "epoch": 2.4197802197802196,
+      "grad_norm": 0.01415912713855505,
+      "learning_rate": 0.0006449511400651466,
+      "loss": 1.196,
+      "step": 1101
+    },
+    {
+      "epoch": 2.421978021978022,
+      "grad_norm": 0.0182810015976429,
+      "learning_rate": 0.0006425081433224756,
+      "loss": 1.2036,
+      "step": 1102
+    },
+    {
+      "epoch": 2.424175824175824,
+      "grad_norm": 0.025705775246024132,
+      "learning_rate": 0.0006400651465798046,
+      "loss": 1.156,
+      "step": 1103
+    },
+    {
+      "epoch": 2.426373626373626,
+      "grad_norm": 0.01893295906484127,
+      "learning_rate": 0.0006376221498371336,
+      "loss": 1.0455,
+      "step": 1104
+    },
+    {
+      "epoch": 2.4285714285714284,
+      "grad_norm": 0.030085502192378044,
+      "learning_rate": 0.0006351791530944626,
+      "loss": 1.3925,
+      "step": 1105
+    },
+    {
+      "epoch": 2.430769230769231,
+      "grad_norm": 0.019910903647542,
+      "learning_rate": 0.0006327361563517916,
+      "loss": 0.9559,
+      "step": 1106
+    },
+    {
+      "epoch": 2.432967032967033,
+      "grad_norm": 0.015571760013699532,
+      "learning_rate": 0.0006302931596091206,
+      "loss": 1.2722,
+      "step": 1107
+    },
+    {
+      "epoch": 2.4351648351648354,
+      "grad_norm": 0.017694668844342232,
+      "learning_rate": 0.0006278501628664496,
+      "loss": 1.0783,
+      "step": 1108
+    },
+    {
+      "epoch": 2.4373626373626376,
+      "grad_norm": 0.025843292474746704,
+      "learning_rate": 0.0006254071661237785,
+      "loss": 1.104,
+      "step": 1109
+    },
+    {
+      "epoch": 2.4395604395604398,
+      "grad_norm": 0.038322899490594864,
+      "learning_rate": 0.0006229641693811075,
+      "loss": 1.0829,
+      "step": 1110
+    },
+    {
+      "epoch": 2.441758241758242,
+      "grad_norm": 0.020965848118066788,
+      "learning_rate": 0.0006205211726384365,
+      "loss": 0.973,
+      "step": 1111
+    },
+    {
+      "epoch": 2.443956043956044,
+      "grad_norm": 0.02236080914735794,
+      "learning_rate": 0.0006180781758957655,
+      "loss": 1.2991,
+      "step": 1112
+    },
+    {
+      "epoch": 2.4461538461538463,
+      "grad_norm": 0.017847837880253792,
+      "learning_rate": 0.0006156351791530945,
+      "loss": 1.0402,
+      "step": 1113
+    },
+    {
+      "epoch": 2.4483516483516485,
+      "grad_norm": 0.021520404145121574,
+      "learning_rate": 0.0006131921824104235,
+      "loss": 1.1893,
+      "step": 1114
+    },
+    {
+      "epoch": 2.4505494505494507,
+      "grad_norm": 0.017562136054039,
+      "learning_rate": 0.0006107491856677525,
+      "loss": 1.0834,
+      "step": 1115
+    },
+    {
+      "epoch": 2.452747252747253,
+      "grad_norm": 0.020595835521817207,
+      "learning_rate": 0.0006083061889250815,
+      "loss": 1.1056,
+      "step": 1116
+    },
+    {
+      "epoch": 2.454945054945055,
+      "grad_norm": 0.018546195700764656,
+      "learning_rate": 0.0006058631921824105,
+      "loss": 1.1208,
+      "step": 1117
+    },
+    {
+      "epoch": 2.4571428571428573,
+      "grad_norm": 0.022387731820344925,
+      "learning_rate": 0.0006034201954397394,
+      "loss": 1.1354,
+      "step": 1118
+    },
+    {
+      "epoch": 2.4593406593406595,
+      "grad_norm": 0.0266229547560215,
+      "learning_rate": 0.0006009771986970684,
+      "loss": 0.9855,
+      "step": 1119
+    },
+    {
+      "epoch": 2.4615384615384617,
+      "grad_norm": 0.01836404763162136,
+      "learning_rate": 0.0005985342019543974,
+      "loss": 1.1539,
+      "step": 1120
+    },
+    {
+      "epoch": 2.463736263736264,
+      "grad_norm": 0.023547504097223282,
+      "learning_rate": 0.0005960912052117264,
+      "loss": 0.9446,
+      "step": 1121
+    },
+    {
+      "epoch": 2.465934065934066,
+      "grad_norm": 0.02130022831261158,
+      "learning_rate": 0.0005936482084690554,
+      "loss": 1.173,
+      "step": 1122
+    },
+    {
+      "epoch": 2.4681318681318682,
+      "grad_norm": 0.020415248349308968,
+      "learning_rate": 0.0005912052117263844,
+      "loss": 1.1616,
+      "step": 1123
+    },
+    {
+      "epoch": 2.4703296703296704,
+      "grad_norm": 0.02320365607738495,
+      "learning_rate": 0.0005887622149837134,
+      "loss": 1.3361,
+      "step": 1124
+    },
+    {
+      "epoch": 2.4725274725274726,
+      "grad_norm": 0.05857139080762863,
+      "learning_rate": 0.0005863192182410424,
+      "loss": 1.0714,
+      "step": 1125
+    },
+    {
+      "epoch": 2.474725274725275,
+      "grad_norm": 0.020381705835461617,
+      "learning_rate": 0.0005838762214983714,
+      "loss": 1.1217,
+      "step": 1126
+    },
+    {
+      "epoch": 2.476923076923077,
+      "grad_norm": 0.02237487956881523,
+      "learning_rate": 0.0005814332247557004,
+      "loss": 1.1702,
+      "step": 1127
+    },
+    {
+      "epoch": 2.479120879120879,
+      "grad_norm": 0.025580093264579773,
+      "learning_rate": 0.0005789902280130293,
+      "loss": 1.2631,
+      "step": 1128
+    },
+    {
+      "epoch": 2.4813186813186814,
+      "grad_norm": 0.019088264554739,
+      "learning_rate": 0.0005765472312703583,
+      "loss": 1.0985,
+      "step": 1129
+    },
+    {
+      "epoch": 2.4835164835164836,
+      "grad_norm": 0.02450934611260891,
+      "learning_rate": 0.0005741042345276873,
+      "loss": 1.2496,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4857142857142858,
+      "grad_norm": 0.024226488545536995,
+      "learning_rate": 0.0005716612377850163,
+      "loss": 1.1853,
+      "step": 1131
+    },
+    {
+      "epoch": 2.487912087912088,
+      "grad_norm": 0.023824702948331833,
+      "learning_rate": 0.0005692182410423453,
+      "loss": 0.9933,
+      "step": 1132
+    },
+    {
+      "epoch": 2.49010989010989,
+      "grad_norm": 0.02103614993393421,
+      "learning_rate": 0.0005667752442996743,
+      "loss": 1.0172,
+      "step": 1133
+    },
+    {
+      "epoch": 2.4923076923076923,
+      "grad_norm": 0.01793731562793255,
+      "learning_rate": 0.0005643322475570033,
+      "loss": 1.2776,
+      "step": 1134
+    },
+    {
+      "epoch": 2.4945054945054945,
+      "grad_norm": 0.02265056036412716,
+      "learning_rate": 0.0005618892508143323,
+      "loss": 0.8895,
+      "step": 1135
+    },
+    {
+      "epoch": 2.4967032967032967,
+      "grad_norm": 0.025548463687300682,
+      "learning_rate": 0.0005594462540716613,
+      "loss": 1.0964,
+      "step": 1136
+    },
+    {
+      "epoch": 2.498901098901099,
+      "grad_norm": 0.017929360270500183,
+      "learning_rate": 0.0005570032573289903,
+      "loss": 0.9181,
+      "step": 1137
+    },
+    {
+      "epoch": 2.501098901098901,
+      "grad_norm": 0.01569174788892269,
+      "learning_rate": 0.0005545602605863192,
+      "loss": 1.004,
+      "step": 1138
+    },
+    {
+      "epoch": 2.5032967032967033,
+      "grad_norm": 0.021732257679104805,
+      "learning_rate": 0.0005521172638436482,
+      "loss": 1.0817,
+      "step": 1139
+    },
+    {
+      "epoch": 2.5054945054945055,
+      "grad_norm": 0.02816474623978138,
+      "learning_rate": 0.0005496742671009772,
+      "loss": 1.2192,
+      "step": 1140
+    },
+    {
+      "epoch": 2.5076923076923077,
+      "grad_norm": 0.02665177546441555,
+      "learning_rate": 0.0005472312703583062,
+      "loss": 1.1058,
+      "step": 1141
+    },
+    {
+      "epoch": 2.50989010989011,
+      "grad_norm": 0.029243988916277885,
+      "learning_rate": 0.0005447882736156352,
+      "loss": 1.3018,
+      "step": 1142
+    },
+    {
+      "epoch": 2.512087912087912,
+      "grad_norm": 0.01623457483947277,
+      "learning_rate": 0.0005423452768729642,
+      "loss": 1.0747,
+      "step": 1143
+    },
+    {
+      "epoch": 2.5142857142857142,
+      "grad_norm": 0.01885387860238552,
+      "learning_rate": 0.0005399022801302932,
+      "loss": 1.1252,
+      "step": 1144
+    },
+    {
+      "epoch": 2.5164835164835164,
+      "grad_norm": 0.016283603385090828,
+      "learning_rate": 0.0005374592833876222,
+      "loss": 1.2849,
+      "step": 1145
+    },
+    {
+      "epoch": 2.5186813186813186,
+      "grad_norm": 0.020428769290447235,
+      "learning_rate": 0.0005350162866449512,
+      "loss": 0.9323,
+      "step": 1146
+    },
+    {
+      "epoch": 2.520879120879121,
+      "grad_norm": 0.018850408494472504,
+      "learning_rate": 0.0005325732899022801,
+      "loss": 0.9047,
+      "step": 1147
+    },
+    {
+      "epoch": 2.523076923076923,
+      "grad_norm": 0.0196260716766119,
+      "learning_rate": 0.0005301302931596091,
+      "loss": 1.0813,
+      "step": 1148
+    },
+    {
+      "epoch": 2.525274725274725,
+      "grad_norm": 0.016515124589204788,
+      "learning_rate": 0.0005276872964169381,
+      "loss": 0.9664,
+      "step": 1149
+    },
+    {
+      "epoch": 2.5274725274725274,
+      "grad_norm": 0.03192732855677605,
+      "learning_rate": 0.0005252442996742671,
+      "loss": 1.3195,
+      "step": 1150
+    },
+    {
+      "epoch": 2.5296703296703296,
+      "grad_norm": 0.017747916281223297,
+      "learning_rate": 0.0005228013029315961,
+      "loss": 0.9984,
+      "step": 1151
+    },
+    {
+      "epoch": 2.5318681318681318,
+      "grad_norm": 0.025288138538599014,
+      "learning_rate": 0.0005203583061889251,
+      "loss": 1.0928,
+      "step": 1152
+    },
+    {
+      "epoch": 2.534065934065934,
+      "grad_norm": 0.042984798550605774,
+      "learning_rate": 0.0005179153094462541,
+      "loss": 1.1261,
+      "step": 1153
+    },
+    {
+      "epoch": 2.536263736263736,
+      "grad_norm": 0.016622595489025116,
+      "learning_rate": 0.0005154723127035831,
+      "loss": 1.1741,
+      "step": 1154
+    },
+    {
+      "epoch": 2.5384615384615383,
+      "grad_norm": 0.02410488948225975,
+      "learning_rate": 0.0005130293159609121,
+      "loss": 1.274,
+      "step": 1155
+    },
+    {
+      "epoch": 2.5406593406593405,
+      "grad_norm": 0.023818153887987137,
+      "learning_rate": 0.0005105863192182411,
+      "loss": 1.1372,
+      "step": 1156
+    },
+    {
+      "epoch": 2.5428571428571427,
+      "grad_norm": 0.016220534220337868,
+      "learning_rate": 0.00050814332247557,
+      "loss": 1.1193,
+      "step": 1157
+    },
+    {
+      "epoch": 2.545054945054945,
+      "grad_norm": 0.03744957596063614,
+      "learning_rate": 0.000505700325732899,
+      "loss": 0.9971,
+      "step": 1158
+    },
+    {
+      "epoch": 2.547252747252747,
+      "grad_norm": 0.02002684399485588,
+      "learning_rate": 0.000503257328990228,
+      "loss": 1.1277,
+      "step": 1159
+    },
+    {
+      "epoch": 2.5494505494505493,
+      "grad_norm": 0.01999819092452526,
+      "learning_rate": 0.000500814332247557,
+      "loss": 0.9847,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5516483516483515,
+      "grad_norm": 0.03147386014461517,
+      "learning_rate": 0.000498371335504886,
+      "loss": 0.989,
+      "step": 1161
+    },
+    {
+      "epoch": 2.5538461538461537,
+      "grad_norm": 0.021901557222008705,
+      "learning_rate": 0.000495928338762215,
+      "loss": 1.0147,
+      "step": 1162
+    },
+    {
+      "epoch": 2.556043956043956,
+      "grad_norm": 0.023956071585416794,
+      "learning_rate": 0.000493485342019544,
+      "loss": 1.1717,
+      "step": 1163
+    },
+    {
+      "epoch": 2.558241758241758,
+      "grad_norm": 0.02171126939356327,
+      "learning_rate": 0.000491042345276873,
+      "loss": 1.1482,
+      "step": 1164
+    },
+    {
+      "epoch": 2.5604395604395602,
+      "grad_norm": 0.01914544776082039,
+      "learning_rate": 0.000488599348534202,
+      "loss": 1.1629,
+      "step": 1165
+    },
+    {
+      "epoch": 2.5626373626373624,
+      "grad_norm": 0.031010208651423454,
+      "learning_rate": 0.0004861563517915309,
+      "loss": 1.2863,
+      "step": 1166
+    },
+    {
+      "epoch": 2.5648351648351646,
+      "grad_norm": 0.024695992469787598,
+      "learning_rate": 0.0004837133550488599,
+      "loss": 1.074,
+      "step": 1167
+    },
+    {
+      "epoch": 2.567032967032967,
+      "grad_norm": 0.01731940545141697,
+      "learning_rate": 0.0004812703583061889,
+      "loss": 1.0579,
+      "step": 1168
+    },
+    {
+      "epoch": 2.569230769230769,
+      "grad_norm": 0.019721299409866333,
+      "learning_rate": 0.00047882736156351787,
+      "loss": 1.2086,
+      "step": 1169
+    },
+    {
+      "epoch": 2.571428571428571,
+      "grad_norm": 0.018155694007873535,
+      "learning_rate": 0.00047638436482084687,
+      "loss": 1.0503,
+      "step": 1170
+    },
+    {
+      "epoch": 2.5736263736263734,
+      "grad_norm": 0.028678542003035545,
+      "learning_rate": 0.00047394136807817587,
+      "loss": 1.1104,
+      "step": 1171
+    },
+    {
+      "epoch": 2.5758241758241756,
+      "grad_norm": 0.022171618416905403,
+      "learning_rate": 0.00047149837133550487,
+      "loss": 1.0336,
+      "step": 1172
+    },
+    {
+      "epoch": 2.5780219780219777,
+      "grad_norm": 0.01968390867114067,
+      "learning_rate": 0.00046905537459283387,
+      "loss": 1.3027,
+      "step": 1173
+    },
+    {
+      "epoch": 2.5802197802197804,
+      "grad_norm": 0.03449544310569763,
+      "learning_rate": 0.0004666123778501629,
+      "loss": 1.1038,
+      "step": 1174
+    },
+    {
+      "epoch": 2.5824175824175826,
+      "grad_norm": 0.020968087017536163,
+      "learning_rate": 0.0004641693811074919,
+      "loss": 1.4044,
+      "step": 1175
+    },
+    {
+      "epoch": 2.5846153846153848,
+      "grad_norm": 0.018007470294833183,
+      "learning_rate": 0.00046172638436482087,
+      "loss": 1.0727,
+      "step": 1176
+    },
+    {
+      "epoch": 2.586813186813187,
+      "grad_norm": 0.020947923883795738,
+      "learning_rate": 0.00045928338762214987,
+      "loss": 1.2497,
+      "step": 1177
+    },
+    {
+      "epoch": 2.589010989010989,
+      "grad_norm": 0.026972908526659012,
+      "learning_rate": 0.0004568403908794789,
+      "loss": 1.0305,
+      "step": 1178
+    },
+    {
+      "epoch": 2.5912087912087913,
+      "grad_norm": 0.01737581379711628,
+      "learning_rate": 0.0004543973941368079,
+      "loss": 1.0768,
+      "step": 1179
+    },
+    {
+      "epoch": 2.5934065934065935,
+      "grad_norm": 0.01553143747150898,
+      "learning_rate": 0.0004519543973941369,
+      "loss": 1.1006,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5956043956043957,
+      "grad_norm": 0.02098107896745205,
+      "learning_rate": 0.0004495114006514658,
+      "loss": 1.2585,
+      "step": 1181
+    },
+    {
+      "epoch": 2.597802197802198,
+      "grad_norm": 0.013780457898974419,
+      "learning_rate": 0.0004470684039087948,
+      "loss": 1.0853,
+      "step": 1182
+    },
+    {
+      "epoch": 2.6,
+      "grad_norm": 0.021120892837643623,
+      "learning_rate": 0.0004446254071661238,
+      "loss": 1.0937,
+      "step": 1183
+    },
+    {
+      "epoch": 2.6021978021978023,
+      "grad_norm": 0.016106463968753815,
+      "learning_rate": 0.0004421824104234528,
+      "loss": 0.9388,
+      "step": 1184
+    },
+    {
+      "epoch": 2.6043956043956045,
+      "grad_norm": 0.025462951511144638,
+      "learning_rate": 0.0004397394136807818,
+      "loss": 1.042,
+      "step": 1185
+    },
+    {
+      "epoch": 2.6065934065934067,
+      "grad_norm": 0.02200544998049736,
+      "learning_rate": 0.00043729641693811077,
+      "loss": 1.15,
+      "step": 1186
+    },
+    {
+      "epoch": 2.608791208791209,
+      "grad_norm": 0.013440352864563465,
+      "learning_rate": 0.00043485342019543977,
+      "loss": 1.1463,
+      "step": 1187
+    },
+    {
+      "epoch": 2.610989010989011,
+      "grad_norm": 0.014734337106347084,
+      "learning_rate": 0.00043241042345276877,
+      "loss": 0.9765,
+      "step": 1188
+    },
+    {
+      "epoch": 2.6131868131868132,
+      "grad_norm": 0.016969390213489532,
+      "learning_rate": 0.00042996742671009777,
+      "loss": 1.2223,
+      "step": 1189
+    },
+    {
+      "epoch": 2.6153846153846154,
+      "grad_norm": 0.021582217887043953,
+      "learning_rate": 0.0004275244299674267,
+      "loss": 1.068,
+      "step": 1190
+    },
+    {
+      "epoch": 2.6175824175824176,
+      "grad_norm": 0.014704504981637001,
+      "learning_rate": 0.0004250814332247557,
+      "loss": 1.0251,
+      "step": 1191
+    },
+    {
+      "epoch": 2.61978021978022,
+      "grad_norm": 0.03158725053071976,
+      "learning_rate": 0.0004226384364820847,
+      "loss": 1.0486,
+      "step": 1192
+    },
+    {
+      "epoch": 2.621978021978022,
+      "grad_norm": 0.015141256153583527,
+      "learning_rate": 0.0004201954397394137,
+      "loss": 1.0953,
+      "step": 1193
+    },
+    {
+      "epoch": 2.624175824175824,
+      "grad_norm": 0.025808772072196007,
+      "learning_rate": 0.0004177524429967427,
+      "loss": 1.0225,
+      "step": 1194
+    },
+    {
+      "epoch": 2.6263736263736264,
+      "grad_norm": 0.016076676547527313,
+      "learning_rate": 0.00041530944625407166,
+      "loss": 1.194,
+      "step": 1195
+    },
+    {
+      "epoch": 2.6285714285714286,
+      "grad_norm": 0.02463696338236332,
+      "learning_rate": 0.00041286644951140066,
+      "loss": 0.9915,
+      "step": 1196
+    },
+    {
+      "epoch": 2.6307692307692307,
+      "grad_norm": 0.020694071426987648,
+      "learning_rate": 0.00041042345276872966,
+      "loss": 0.987,
+      "step": 1197
+    },
+    {
+      "epoch": 2.632967032967033,
+      "grad_norm": 0.017329517751932144,
+      "learning_rate": 0.00040798045602605866,
+      "loss": 1.1712,
+      "step": 1198
+    },
+    {
+      "epoch": 2.635164835164835,
+      "grad_norm": 0.026437675580382347,
+      "learning_rate": 0.00040553745928338767,
+      "loss": 1.2967,
+      "step": 1199
+    },
+    {
+      "epoch": 2.6373626373626373,
+      "grad_norm": 0.03410179167985916,
+      "learning_rate": 0.0004030944625407166,
+      "loss": 1.1723,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6395604395604395,
+      "grad_norm": 0.02042745240032673,
+      "learning_rate": 0.0004006514657980456,
+      "loss": 1.0671,
+      "step": 1201
+    },
+    {
+      "epoch": 2.6417582417582417,
+      "grad_norm": 0.023282315582036972,
+      "learning_rate": 0.0003982084690553746,
+      "loss": 1.1967,
+      "step": 1202
+    },
+    {
+      "epoch": 2.643956043956044,
+      "grad_norm": 0.01871698722243309,
+      "learning_rate": 0.0003957654723127036,
+      "loss": 1.1376,
+      "step": 1203
+    },
+    {
+      "epoch": 2.646153846153846,
+      "grad_norm": 0.024699589237570763,
+      "learning_rate": 0.0003933224755700326,
+      "loss": 0.9756,
+      "step": 1204
+    },
+    {
+      "epoch": 2.6483516483516483,
+      "grad_norm": 0.023694438859820366,
+      "learning_rate": 0.00039087947882736156,
+      "loss": 1.1792,
+      "step": 1205
+    },
+    {
+      "epoch": 2.6505494505494505,
+      "grad_norm": 0.025342661887407303,
+      "learning_rate": 0.00038843648208469056,
+      "loss": 1.2202,
+      "step": 1206
+    },
+    {
+      "epoch": 2.6527472527472526,
+      "grad_norm": 0.021758759394288063,
+      "learning_rate": 0.00038599348534201956,
+      "loss": 0.9959,
+      "step": 1207
+    },
+    {
+      "epoch": 2.654945054945055,
+      "grad_norm": 0.023840008303523064,
+      "learning_rate": 0.00038355048859934856,
+      "loss": 1.102,
+      "step": 1208
+    },
+    {
+      "epoch": 2.657142857142857,
+      "grad_norm": 0.0237457025796175,
+      "learning_rate": 0.00038110749185667756,
+      "loss": 1.358,
+      "step": 1209
+    },
+    {
+      "epoch": 2.659340659340659,
+      "grad_norm": 0.022617178037762642,
+      "learning_rate": 0.0003786644951140065,
+      "loss": 1.1728,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6615384615384614,
+      "grad_norm": 0.020305756479501724,
+      "learning_rate": 0.0003762214983713355,
+      "loss": 1.1682,
+      "step": 1211
+    },
+    {
+      "epoch": 2.6637362637362636,
+      "grad_norm": 0.017555296421051025,
+      "learning_rate": 0.0003737785016286645,
+      "loss": 0.9821,
+      "step": 1212
+    },
+    {
+      "epoch": 2.665934065934066,
+      "grad_norm": 0.0160040445625782,
+      "learning_rate": 0.0003713355048859935,
+      "loss": 1.0197,
+      "step": 1213
+    },
+    {
+      "epoch": 2.668131868131868,
+      "grad_norm": 0.022772135213017464,
+      "learning_rate": 0.00036889250814332245,
+      "loss": 0.9273,
+      "step": 1214
+    },
+    {
+      "epoch": 2.67032967032967,
+      "grad_norm": 0.025266308337450027,
+      "learning_rate": 0.00036644951140065146,
+      "loss": 1.0534,
+      "step": 1215
+    },
+    {
+      "epoch": 2.672527472527473,
+      "grad_norm": 0.020648887380957603,
+      "learning_rate": 0.00036400651465798046,
+      "loss": 0.9759,
+      "step": 1216
+    },
+    {
+      "epoch": 2.674725274725275,
+      "grad_norm": 0.019274771213531494,
+      "learning_rate": 0.00036156351791530946,
+      "loss": 1.0128,
+      "step": 1217
+    },
+    {
+      "epoch": 2.676923076923077,
+      "grad_norm": 0.019681645557284355,
+      "learning_rate": 0.00035912052117263846,
+      "loss": 1.2158,
+      "step": 1218
+    },
+    {
+      "epoch": 2.6791208791208794,
+      "grad_norm": 0.021705498918890953,
+      "learning_rate": 0.0003566775244299674,
+      "loss": 0.9156,
+      "step": 1219
+    },
+    {
+      "epoch": 2.6813186813186816,
+      "grad_norm": 0.025703178718686104,
+      "learning_rate": 0.0003542345276872964,
+      "loss": 1.0891,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6835164835164838,
+      "grad_norm": 0.023441698402166367,
+      "learning_rate": 0.0003517915309446254,
+      "loss": 1.0403,
+      "step": 1221
+    },
+    {
+      "epoch": 2.685714285714286,
+      "grad_norm": 0.018052132800221443,
+      "learning_rate": 0.0003493485342019544,
+      "loss": 1.2011,
+      "step": 1222
+    },
+    {
+      "epoch": 2.687912087912088,
+      "grad_norm": 0.01904676854610443,
+      "learning_rate": 0.0003469055374592834,
+      "loss": 1.2523,
+      "step": 1223
+    },
+    {
+      "epoch": 2.6901098901098903,
+      "grad_norm": 0.03153200447559357,
+      "learning_rate": 0.00034446254071661235,
+      "loss": 1.0952,
+      "step": 1224
+    },
+    {
+      "epoch": 2.6923076923076925,
+      "grad_norm": 0.017920130863785744,
+      "learning_rate": 0.00034201954397394135,
+      "loss": 1.184,
+      "step": 1225
+    },
+    {
+      "epoch": 2.6945054945054947,
+      "grad_norm": 0.014197424054145813,
+      "learning_rate": 0.00033957654723127035,
+      "loss": 0.9072,
+      "step": 1226
+    },
+    {
+      "epoch": 2.696703296703297,
+      "grad_norm": 0.019953910261392593,
+      "learning_rate": 0.00033713355048859935,
+      "loss": 0.8608,
+      "step": 1227
+    },
+    {
+      "epoch": 2.698901098901099,
+      "grad_norm": 0.02067406475543976,
+      "learning_rate": 0.00033469055374592835,
+      "loss": 0.961,
+      "step": 1228
+    },
+    {
+      "epoch": 2.7010989010989013,
+      "grad_norm": 0.022154590114951134,
+      "learning_rate": 0.0003322475570032573,
+      "loss": 1.3714,
+      "step": 1229
+    },
+    {
+      "epoch": 2.7032967032967035,
+      "grad_norm": 0.01952856406569481,
+      "learning_rate": 0.0003298045602605863,
+      "loss": 1.321,
+      "step": 1230
+    },
+    {
+      "epoch": 2.7054945054945057,
+      "grad_norm": 0.014930257573723793,
+      "learning_rate": 0.00032736156351791535,
+      "loss": 1.0971,
+      "step": 1231
+    },
+    {
+      "epoch": 2.707692307692308,
+      "grad_norm": 0.02209659107029438,
+      "learning_rate": 0.0003249185667752443,
+      "loss": 1.0883,
+      "step": 1232
+    },
+    {
+      "epoch": 2.70989010989011,
+      "grad_norm": 0.02391032502055168,
+      "learning_rate": 0.0003224755700325733,
+      "loss": 1.291,
+      "step": 1233
+    },
+    {
+      "epoch": 2.7120879120879122,
+      "grad_norm": 0.023111382499337196,
+      "learning_rate": 0.0003200325732899023,
+      "loss": 1.2173,
+      "step": 1234
+    },
+    {
+      "epoch": 2.7142857142857144,
+      "grad_norm": 0.024029018357396126,
+      "learning_rate": 0.0003175895765472313,
+      "loss": 1.0457,
+      "step": 1235
+    },
+    {
+      "epoch": 2.7164835164835166,
+      "grad_norm": 0.01494946051388979,
+      "learning_rate": 0.0003151465798045603,
+      "loss": 0.8903,
+      "step": 1236
+    },
+    {
+      "epoch": 2.718681318681319,
+      "grad_norm": 0.028328321874141693,
+      "learning_rate": 0.00031270358306188925,
+      "loss": 0.9648,
+      "step": 1237
+    },
+    {
+      "epoch": 2.720879120879121,
+      "grad_norm": 0.01757773384451866,
+      "learning_rate": 0.00031026058631921825,
+      "loss": 1.2363,
+      "step": 1238
+    },
+    {
+      "epoch": 2.723076923076923,
+      "grad_norm": 0.02536613680422306,
+      "learning_rate": 0.00030781758957654725,
+      "loss": 1.1908,
+      "step": 1239
+    },
+    {
+      "epoch": 2.7252747252747254,
+      "grad_norm": 0.025294583290815353,
+      "learning_rate": 0.00030537459283387625,
+      "loss": 1.1992,
+      "step": 1240
+    },
+    {
+      "epoch": 2.7274725274725276,
+      "grad_norm": 0.02701612561941147,
+      "learning_rate": 0.00030293159609120525,
+      "loss": 0.9271,
+      "step": 1241
+    },
+    {
+      "epoch": 2.7296703296703297,
+      "grad_norm": 0.015726754441857338,
+      "learning_rate": 0.0003004885993485342,
+      "loss": 0.9276,
+      "step": 1242
+    },
+    {
+      "epoch": 2.731868131868132,
+      "grad_norm": 0.014120040461421013,
+      "learning_rate": 0.0002980456026058632,
+      "loss": 1.0405,
+      "step": 1243
+    },
+    {
+      "epoch": 2.734065934065934,
+      "grad_norm": 0.03073623962700367,
+      "learning_rate": 0.0002956026058631922,
+      "loss": 1.1676,
+      "step": 1244
+    },
+    {
+      "epoch": 2.7362637362637363,
+      "grad_norm": 0.02590056136250496,
+      "learning_rate": 0.0002931596091205212,
+      "loss": 1.0897,
+      "step": 1245
+    },
+    {
+      "epoch": 2.7384615384615385,
+      "grad_norm": 0.022374441847205162,
+      "learning_rate": 0.0002907166123778502,
+      "loss": 1.0671,
+      "step": 1246
+    },
+    {
+      "epoch": 2.7406593406593407,
+      "grad_norm": 0.02750556170940399,
+      "learning_rate": 0.00028827361563517914,
+      "loss": 1.0858,
+      "step": 1247
+    },
+    {
+      "epoch": 2.742857142857143,
+      "grad_norm": 0.017792342230677605,
+      "learning_rate": 0.00028583061889250814,
+      "loss": 1.0654,
+      "step": 1248
+    },
+    {
+      "epoch": 2.745054945054945,
+      "grad_norm": 0.029480930417776108,
+      "learning_rate": 0.00028338762214983714,
+      "loss": 1.192,
+      "step": 1249
+    },
+    {
+      "epoch": 2.7472527472527473,
+      "grad_norm": 0.01793689839541912,
+      "learning_rate": 0.00028094462540716614,
+      "loss": 0.9745,
+      "step": 1250
+    },
+    {
+      "epoch": 2.7494505494505495,
+      "grad_norm": 0.06284184008836746,
+      "learning_rate": 0.00027850162866449514,
+      "loss": 1.0377,
+      "step": 1251
+    },
+    {
+      "epoch": 2.7516483516483516,
+      "grad_norm": 0.02327103540301323,
+      "learning_rate": 0.0002760586319218241,
+      "loss": 1.2121,
+      "step": 1252
+    },
+    {
+      "epoch": 2.753846153846154,
+      "grad_norm": 0.020397521555423737,
+      "learning_rate": 0.0002736156351791531,
+      "loss": 1.1769,
+      "step": 1253
+    },
+    {
+      "epoch": 2.756043956043956,
+      "grad_norm": 0.02766312286257744,
+      "learning_rate": 0.0002711726384364821,
+      "loss": 1.0883,
+      "step": 1254
+    },
+    {
+      "epoch": 2.758241758241758,
+      "grad_norm": 0.02165943756699562,
+      "learning_rate": 0.0002687296416938111,
+      "loss": 1.2648,
+      "step": 1255
+    },
+    {
+      "epoch": 2.7604395604395604,
+      "grad_norm": 0.019348718225955963,
+      "learning_rate": 0.00026628664495114004,
+      "loss": 1.0231,
+      "step": 1256
+    },
+    {
+      "epoch": 2.7626373626373626,
+      "grad_norm": 0.01911984384059906,
+      "learning_rate": 0.00026384364820846904,
+      "loss": 1.12,
+      "step": 1257
+    },
+    {
+      "epoch": 2.764835164835165,
+      "grad_norm": 0.015475914813578129,
+      "learning_rate": 0.00026140065146579804,
+      "loss": 1.0043,
+      "step": 1258
+    },
+    {
+      "epoch": 2.767032967032967,
+      "grad_norm": 0.019358325749635696,
+      "learning_rate": 0.00025895765472312704,
+      "loss": 0.9168,
+      "step": 1259
+    },
+    {
+      "epoch": 2.769230769230769,
+      "grad_norm": 0.01625392585992813,
+      "learning_rate": 0.00025651465798045604,
+      "loss": 1.1224,
+      "step": 1260
+    },
+    {
+      "epoch": 2.7714285714285714,
+      "grad_norm": 0.019652750343084335,
+      "learning_rate": 0.000254071661237785,
+      "loss": 1.3172,
+      "step": 1261
+    },
+    {
+      "epoch": 2.7736263736263735,
+      "grad_norm": 0.01908297650516033,
+      "learning_rate": 0.000251628664495114,
+      "loss": 1.0036,
+      "step": 1262
+    },
+    {
+      "epoch": 2.7758241758241757,
+      "grad_norm": 0.017116932198405266,
+      "learning_rate": 0.000249185667752443,
+      "loss": 1.1737,
+      "step": 1263
+    },
+    {
+      "epoch": 2.778021978021978,
+      "grad_norm": 0.019146107137203217,
+      "learning_rate": 0.000246742671009772,
+      "loss": 1.0328,
+      "step": 1264
+    },
+    {
+      "epoch": 2.78021978021978,
+      "grad_norm": 0.018499931320548058,
+      "learning_rate": 0.000244299674267101,
+      "loss": 1.0762,
+      "step": 1265
+    },
+    {
+      "epoch": 2.7824175824175823,
+      "grad_norm": 0.018627038225531578,
+      "learning_rate": 0.00024185667752442996,
+      "loss": 1.0807,
+      "step": 1266
+    },
+    {
+      "epoch": 2.7846153846153845,
+      "grad_norm": 0.01747470535337925,
+      "learning_rate": 0.00023941368078175893,
+      "loss": 1.0272,
+      "step": 1267
+    },
+    {
+      "epoch": 2.7868131868131867,
+      "grad_norm": 0.017504315823316574,
+      "learning_rate": 0.00023697068403908794,
+      "loss": 1.1309,
+      "step": 1268
+    },
+    {
+      "epoch": 2.789010989010989,
+      "grad_norm": 0.03496870398521423,
+      "learning_rate": 0.00023452768729641694,
+      "loss": 1.0693,
+      "step": 1269
+    },
+    {
+      "epoch": 2.791208791208791,
+      "grad_norm": 0.013090156018733978,
+      "learning_rate": 0.00023208469055374596,
+      "loss": 1.098,
+      "step": 1270
+    },
+    {
+      "epoch": 2.7934065934065933,
+      "grad_norm": 0.025052325800061226,
+      "learning_rate": 0.00022964169381107494,
+      "loss": 1.2679,
+      "step": 1271
+    },
+    {
+      "epoch": 2.7956043956043954,
+      "grad_norm": 0.019536377862095833,
+      "learning_rate": 0.00022719869706840394,
+      "loss": 1.145,
+      "step": 1272
+    },
+    {
+      "epoch": 2.7978021978021976,
+      "grad_norm": 0.018812142312526703,
+      "learning_rate": 0.0002247557003257329,
+      "loss": 1.1093,
+      "step": 1273
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 0.023722905665636063,
+      "learning_rate": 0.0002223127035830619,
+      "loss": 1.0906,
+      "step": 1274
+    },
+    {
+      "epoch": 2.802197802197802,
+      "grad_norm": 0.015723684802651405,
+      "learning_rate": 0.0002198697068403909,
+      "loss": 1.1906,
+      "step": 1275
+    },
+    {
+      "epoch": 2.804395604395604,
+      "grad_norm": 0.021724017336964607,
+      "learning_rate": 0.00021742671009771988,
+      "loss": 0.9663,
+      "step": 1276
+    },
+    {
+      "epoch": 2.8065934065934064,
+      "grad_norm": 0.23949511349201202,
+      "learning_rate": 0.00021498371335504888,
+      "loss": 1.2406,
+      "step": 1277
+    },
+    {
+      "epoch": 2.8087912087912086,
+      "grad_norm": 0.015856703743338585,
+      "learning_rate": 0.00021254071661237786,
+      "loss": 1.0885,
+      "step": 1278
+    },
+    {
+      "epoch": 2.8109890109890108,
+      "grad_norm": 0.013400726020336151,
+      "learning_rate": 0.00021009771986970686,
+      "loss": 1.0476,
+      "step": 1279
+    },
+    {
+      "epoch": 2.813186813186813,
+      "grad_norm": 0.1020999327301979,
+      "learning_rate": 0.00020765472312703583,
+      "loss": 1.1456,
+      "step": 1280
+    },
+    {
+      "epoch": 2.815384615384615,
+      "grad_norm": 0.01777460426092148,
+      "learning_rate": 0.00020521172638436483,
+      "loss": 0.9238,
+      "step": 1281
+    },
+    {
+      "epoch": 2.8175824175824173,
+      "grad_norm": 0.01749647967517376,
+      "learning_rate": 0.00020276872964169383,
+      "loss": 1.3069,
+      "step": 1282
+    },
+    {
+      "epoch": 2.8197802197802195,
+      "grad_norm": 0.020597370341420174,
+      "learning_rate": 0.0002003257328990228,
+      "loss": 1.0467,
+      "step": 1283
+    },
+    {
+      "epoch": 2.8219780219780217,
+      "grad_norm": 0.025279099121689796,
+      "learning_rate": 0.0001978827361563518,
+      "loss": 1.166,
+      "step": 1284
+    },
+    {
+      "epoch": 2.824175824175824,
+      "grad_norm": 0.031879622489213943,
+      "learning_rate": 0.00019543973941368078,
+      "loss": 1.0254,
+      "step": 1285
+    },
+    {
+      "epoch": 2.826373626373626,
+      "grad_norm": 0.017444491386413574,
+      "learning_rate": 0.00019299674267100978,
+      "loss": 1.1358,
+      "step": 1286
+    },
+    {
+      "epoch": 2.8285714285714287,
+      "grad_norm": 0.020547732710838318,
+      "learning_rate": 0.00019055374592833878,
+      "loss": 0.8457,
+      "step": 1287
+    },
+    {
+      "epoch": 2.830769230769231,
+      "grad_norm": 0.015599466860294342,
+      "learning_rate": 0.00018811074918566775,
+      "loss": 1.0596,
+      "step": 1288
+    },
+    {
+      "epoch": 2.832967032967033,
+      "grad_norm": 0.01673927903175354,
+      "learning_rate": 0.00018566775244299675,
+      "loss": 1.1575,
+      "step": 1289
+    },
+    {
+      "epoch": 2.8351648351648353,
+      "grad_norm": 0.01843508705496788,
+      "learning_rate": 0.00018322475570032573,
+      "loss": 1.1421,
+      "step": 1290
+    },
+    {
+      "epoch": 2.8373626373626375,
+      "grad_norm": 0.018654989078640938,
+      "learning_rate": 0.00018078175895765473,
+      "loss": 1.1222,
+      "step": 1291
+    },
+    {
+      "epoch": 2.8395604395604397,
+      "grad_norm": 0.014196369796991348,
+      "learning_rate": 0.0001783387622149837,
+      "loss": 0.9923,
+      "step": 1292
+    },
+    {
+      "epoch": 2.841758241758242,
+      "grad_norm": 0.06032684072852135,
+      "learning_rate": 0.0001758957654723127,
+      "loss": 1.084,
+      "step": 1293
+    },
+    {
+      "epoch": 2.843956043956044,
+      "grad_norm": 0.01826374977827072,
+      "learning_rate": 0.0001734527687296417,
+      "loss": 1.1926,
+      "step": 1294
+    },
+    {
+      "epoch": 2.8461538461538463,
+      "grad_norm": 0.019515972584486008,
+      "learning_rate": 0.00017100977198697068,
+      "loss": 1.0409,
+      "step": 1295
+    },
+    {
+      "epoch": 2.8483516483516484,
+      "grad_norm": 0.055204443633556366,
+      "learning_rate": 0.00016856677524429968,
+      "loss": 0.9682,
+      "step": 1296
+    },
+    {
+      "epoch": 2.8505494505494506,
+      "grad_norm": 0.019304707646369934,
+      "learning_rate": 0.00016612377850162865,
+      "loss": 0.9495,
+      "step": 1297
+    },
+    {
+      "epoch": 2.852747252747253,
+      "grad_norm": 0.01615406945347786,
+      "learning_rate": 0.00016368078175895768,
+      "loss": 1.211,
+      "step": 1298
+    },
+    {
+      "epoch": 2.854945054945055,
+      "grad_norm": 0.01531881745904684,
+      "learning_rate": 0.00016123778501628665,
+      "loss": 1.3195,
+      "step": 1299
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.01931050606071949,
+      "learning_rate": 0.00015879478827361565,
+      "loss": 1.004,
+      "step": 1300
+    },
+    {
+      "epoch": 2.8593406593406594,
+      "grad_norm": 0.02026810124516487,
+      "learning_rate": 0.00015635179153094462,
+      "loss": 1.2654,
+      "step": 1301
+    },
+    {
+      "epoch": 2.8615384615384616,
+      "grad_norm": 0.027087172493338585,
+      "learning_rate": 0.00015390879478827362,
+      "loss": 1.2733,
+      "step": 1302
+    },
+    {
+      "epoch": 2.8637362637362638,
+      "grad_norm": 0.018786976113915443,
+      "learning_rate": 0.00015146579804560262,
+      "loss": 0.9927,
+      "step": 1303
+    },
+    {
+      "epoch": 2.865934065934066,
+      "grad_norm": 0.017147710546851158,
+      "learning_rate": 0.0001490228013029316,
+      "loss": 0.9431,
+      "step": 1304
+    },
+    {
+      "epoch": 2.868131868131868,
+      "grad_norm": 0.019477317109704018,
+      "learning_rate": 0.0001465798045602606,
+      "loss": 1.0948,
+      "step": 1305
+    },
+    {
+      "epoch": 2.8703296703296703,
+      "grad_norm": 0.018405893817543983,
+      "learning_rate": 0.00014413680781758957,
+      "loss": 1.2517,
+      "step": 1306
+    },
+    {
+      "epoch": 2.8725274725274725,
+      "grad_norm": 0.039821021258831024,
+      "learning_rate": 0.00014169381107491857,
+      "loss": 1.1753,
+      "step": 1307
+    },
+    {
+      "epoch": 2.8747252747252747,
+      "grad_norm": 0.01612751930952072,
+      "learning_rate": 0.00013925081433224757,
+      "loss": 1.3547,
+      "step": 1308
+    },
+    {
+      "epoch": 2.876923076923077,
+      "grad_norm": 0.02193351648747921,
+      "learning_rate": 0.00013680781758957655,
+      "loss": 1.3638,
+      "step": 1309
+    },
+    {
+      "epoch": 2.879120879120879,
+      "grad_norm": 0.02079048380255699,
+      "learning_rate": 0.00013436482084690555,
+      "loss": 0.8659,
+      "step": 1310
+    },
+    {
+      "epoch": 2.8813186813186813,
+      "grad_norm": 0.018918242305517197,
+      "learning_rate": 0.00013192182410423452,
+      "loss": 1.1849,
+      "step": 1311
+    },
+    {
+      "epoch": 2.8835164835164835,
+      "grad_norm": 0.02591584622859955,
+      "learning_rate": 0.00012947882736156352,
+      "loss": 1.236,
+      "step": 1312
+    },
+    {
+      "epoch": 2.8857142857142857,
+      "grad_norm": 0.01779877208173275,
+      "learning_rate": 0.0001270358306188925,
+      "loss": 0.9413,
+      "step": 1313
+    },
+    {
+      "epoch": 2.887912087912088,
+      "grad_norm": 0.024049490690231323,
+      "learning_rate": 0.0001245928338762215,
+      "loss": 1.0344,
+      "step": 1314
+    },
+    {
+      "epoch": 2.89010989010989,
+      "grad_norm": 0.02717575989663601,
+      "learning_rate": 0.0001221498371335505,
+      "loss": 1.1877,
+      "step": 1315
+    },
+    {
+      "epoch": 2.8923076923076922,
+      "grad_norm": 0.0154302092269063,
+      "learning_rate": 0.00011970684039087947,
+      "loss": 1.0586,
+      "step": 1316
+    },
+    {
+      "epoch": 2.8945054945054944,
+      "grad_norm": 0.020416175946593285,
+      "learning_rate": 0.00011726384364820847,
+      "loss": 0.9731,
+      "step": 1317
+    },
+    {
+      "epoch": 2.8967032967032966,
+      "grad_norm": 0.01942111738026142,
+      "learning_rate": 0.00011482084690553747,
+      "loss": 1.1593,
+      "step": 1318
+    },
+    {
+      "epoch": 2.898901098901099,
+      "grad_norm": 0.017026664689183235,
+      "learning_rate": 0.00011237785016286646,
+      "loss": 0.9078,
+      "step": 1319
+    },
+    {
+      "epoch": 2.901098901098901,
+      "grad_norm": 0.021696975454688072,
+      "learning_rate": 0.00010993485342019546,
+      "loss": 1.0012,
+      "step": 1320
+    },
+    {
+      "epoch": 2.903296703296703,
+      "grad_norm": 0.01822924241423607,
+      "learning_rate": 0.00010749185667752444,
+      "loss": 1.0037,
+      "step": 1321
+    },
+    {
+      "epoch": 2.9054945054945054,
+      "grad_norm": 0.024573974311351776,
+      "learning_rate": 0.00010504885993485343,
+      "loss": 1.1209,
+      "step": 1322
+    },
+    {
+      "epoch": 2.9076923076923076,
+      "grad_norm": 0.028301198035478592,
+      "learning_rate": 0.00010260586319218242,
+      "loss": 1.0656,
+      "step": 1323
+    },
+    {
+      "epoch": 2.9098901098901098,
+      "grad_norm": 0.020846789702773094,
+      "learning_rate": 0.0001001628664495114,
+      "loss": 1.1696,
+      "step": 1324
+    },
+    {
+      "epoch": 2.912087912087912,
+      "grad_norm": 0.016432736068964005,
+      "learning_rate": 9.771986970684039e-05,
+      "loss": 1.0706,
+      "step": 1325
+    },
+    {
+      "epoch": 2.914285714285714,
+      "grad_norm": 0.02147461473941803,
+      "learning_rate": 9.527687296416939e-05,
+      "loss": 1.0445,
+      "step": 1326
+    },
+    {
+      "epoch": 2.9164835164835163,
+      "grad_norm": 0.01872866041958332,
+      "learning_rate": 9.283387622149838e-05,
+      "loss": 0.9919,
+      "step": 1327
+    },
+    {
+      "epoch": 2.9186813186813185,
+      "grad_norm": 0.017467539757490158,
+      "learning_rate": 9.039087947882736e-05,
+      "loss": 0.9842,
+      "step": 1328
+    },
+    {
+      "epoch": 2.9208791208791207,
+      "grad_norm": 0.02007579803466797,
+      "learning_rate": 8.794788273615635e-05,
+      "loss": 1.1538,
+      "step": 1329
+    },
+    {
+      "epoch": 2.9230769230769234,
+      "grad_norm": 0.019204875454306602,
+      "learning_rate": 8.550488599348534e-05,
+      "loss": 1.3953,
+      "step": 1330
+    },
+    {
+      "epoch": 2.9252747252747255,
+      "grad_norm": 0.013325495645403862,
+      "learning_rate": 8.306188925081432e-05,
+      "loss": 1.0486,
+      "step": 1331
+    },
+    {
+      "epoch": 2.9274725274725277,
+      "grad_norm": 0.017742281779646873,
+      "learning_rate": 8.061889250814332e-05,
+      "loss": 1.2016,
+      "step": 1332
+    },
+    {
+      "epoch": 2.92967032967033,
+      "grad_norm": 0.01945466734468937,
+      "learning_rate": 7.817589576547231e-05,
+      "loss": 1.0292,
+      "step": 1333
+    },
+    {
+      "epoch": 2.931868131868132,
+      "grad_norm": 0.018728459253907204,
+      "learning_rate": 7.573289902280131e-05,
+      "loss": 1.0562,
+      "step": 1334
+    },
+    {
+      "epoch": 2.9340659340659343,
+      "grad_norm": 0.013764962553977966,
+      "learning_rate": 7.32899022801303e-05,
+      "loss": 1.0709,
+      "step": 1335
+    },
+    {
+      "epoch": 2.9362637362637365,
+      "grad_norm": 0.013915336690843105,
+      "learning_rate": 7.084690553745929e-05,
+      "loss": 0.86,
+      "step": 1336
+    },
+    {
+      "epoch": 2.9384615384615387,
+      "grad_norm": 0.017405368387699127,
+      "learning_rate": 6.840390879478827e-05,
+      "loss": 0.99,
+      "step": 1337
+    },
+    {
+      "epoch": 2.940659340659341,
+      "grad_norm": 0.029991112649440765,
+      "learning_rate": 6.596091205211726e-05,
+      "loss": 1.2901,
+      "step": 1338
+    },
+    {
+      "epoch": 2.942857142857143,
+      "grad_norm": 0.01561530027538538,
+      "learning_rate": 6.351791530944625e-05,
+      "loss": 1.1527,
+      "step": 1339
+    },
+    {
+      "epoch": 2.9450549450549453,
+      "grad_norm": 0.016143502667546272,
+      "learning_rate": 6.107491856677525e-05,
+      "loss": 1.1665,
+      "step": 1340
+    },
+    {
+      "epoch": 2.9472527472527474,
+      "grad_norm": 0.02236691489815712,
+      "learning_rate": 5.8631921824104234e-05,
+      "loss": 1.1047,
+      "step": 1341
+    },
+    {
+      "epoch": 2.9494505494505496,
+      "grad_norm": 0.02110566943883896,
+      "learning_rate": 5.618892508143323e-05,
+      "loss": 1.2325,
+      "step": 1342
+    },
+    {
+      "epoch": 2.951648351648352,
+      "grad_norm": 0.015632957220077515,
+      "learning_rate": 5.374592833876222e-05,
+      "loss": 1.1655,
+      "step": 1343
+    },
+    {
+      "epoch": 2.953846153846154,
+      "grad_norm": 0.030875615775585175,
+      "learning_rate": 5.130293159609121e-05,
+      "loss": 1.1668,
+      "step": 1344
+    },
+    {
+      "epoch": 2.956043956043956,
+      "grad_norm": 0.017714468762278557,
+      "learning_rate": 4.8859934853420195e-05,
+      "loss": 0.9718,
+      "step": 1345
+    },
+    {
+      "epoch": 2.9582417582417584,
+      "grad_norm": 0.025245748460292816,
+      "learning_rate": 4.641693811074919e-05,
+      "loss": 1.2785,
+      "step": 1346
+    },
+    {
+      "epoch": 2.9604395604395606,
+      "grad_norm": 0.026781179010868073,
+      "learning_rate": 4.3973941368078175e-05,
+      "loss": 1.1881,
+      "step": 1347
+    },
+    {
+      "epoch": 2.9626373626373628,
+      "grad_norm": 0.02544347569346428,
+      "learning_rate": 4.153094462540716e-05,
+      "loss": 1.2019,
+      "step": 1348
+    },
+    {
+      "epoch": 2.964835164835165,
+      "grad_norm": 0.021691257134079933,
+      "learning_rate": 3.9087947882736156e-05,
+      "loss": 0.9757,
+      "step": 1349
+    },
+    {
+      "epoch": 2.967032967032967,
+      "grad_norm": 0.015637267380952835,
+      "learning_rate": 3.664495114006515e-05,
+      "loss": 1.0991,
+      "step": 1350
+    },
+    {
+      "epoch": 2.9692307692307693,
+      "grad_norm": 0.03807251900434494,
+      "learning_rate": 3.4201954397394136e-05,
+      "loss": 1.1907,
+      "step": 1351
+    },
+    {
+      "epoch": 2.9714285714285715,
+      "grad_norm": 0.035527005791664124,
+      "learning_rate": 3.175895765472312e-05,
+      "loss": 0.9645,
+      "step": 1352
+    },
+    {
+      "epoch": 2.9736263736263737,
+      "grad_norm": 0.015676917508244514,
+      "learning_rate": 2.9315960912052117e-05,
+      "loss": 1.2146,
+      "step": 1353
+    },
+    {
+      "epoch": 2.975824175824176,
+      "grad_norm": 0.016830801963806152,
+      "learning_rate": 2.687296416938111e-05,
+      "loss": 1.0477,
+      "step": 1354
+    },
+    {
+      "epoch": 2.978021978021978,
+      "grad_norm": 0.016681453213095665,
+      "learning_rate": 2.4429967426710097e-05,
+      "loss": 1.0765,
+      "step": 1355
+    },
+    {
+      "epoch": 2.9802197802197803,
+      "grad_norm": 0.016072623431682587,
+      "learning_rate": 2.1986970684039088e-05,
+      "loss": 1.083,
+      "step": 1356
+    },
+    {
+      "epoch": 2.9824175824175825,
+      "grad_norm": 0.025591861456632614,
+      "learning_rate": 1.9543973941368078e-05,
+      "loss": 1.2528,
+      "step": 1357
+    },
+    {
+      "epoch": 2.9846153846153847,
+      "grad_norm": 0.02123359404504299,
+      "learning_rate": 1.7100977198697068e-05,
+      "loss": 1.0207,
+      "step": 1358
+    },
+    {
+      "epoch": 2.986813186813187,
+      "grad_norm": 0.017488786950707436,
+      "learning_rate": 1.4657980456026058e-05,
+      "loss": 0.902,
+      "step": 1359
+    },
+    {
+      "epoch": 2.989010989010989,
+      "grad_norm": 0.011685109697282314,
+      "learning_rate": 1.2214983713355049e-05,
+      "loss": 1.0178,
+      "step": 1360
+    },
+    {
+      "epoch": 2.9912087912087912,
+      "grad_norm": 0.014467261731624603,
+      "learning_rate": 9.771986970684039e-06,
+      "loss": 0.9425,
+      "step": 1361
+    },
+    {
+      "epoch": 2.9934065934065934,
+      "grad_norm": 0.016747958958148956,
+      "learning_rate": 7.328990228013029e-06,
+      "loss": 1.1684,
+      "step": 1362
+    },
+    {
+      "epoch": 2.9956043956043956,
+      "grad_norm": 0.016586847603321075,
+      "learning_rate": 4.8859934853420195e-06,
+      "loss": 1.0848,
+      "step": 1363
+    },
+    {
+      "epoch": 2.997802197802198,
+      "grad_norm": 0.017365796491503716,
+      "learning_rate": 2.4429967426710097e-06,
+      "loss": 1.078,
+      "step": 1364
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.03259021043777466,
+      "learning_rate": 0.0,
+      "loss": 1.1344,
+      "step": 1365
+    },
+    {
+      "epoch": 3.0,
+      "step": 1365,
+      "total_flos": 4.592889669447844e+17,
+      "train_loss": 1.2525357378708137,
+      "train_runtime": 3291.1671,
+      "train_samples_per_second": 6.636,
+      "train_steps_per_second": 0.415
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 1365,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.592889669447844e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}