diff --git "a/checkpoint-1200/trainer_state.json" "b/checkpoint-1200/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-1200/trainer_state.json"
@@ -0,0 +1,8826 @@
+{
+  "best_metric": 0.0031958832405507565,
+  "best_model_checkpoint": "/home/paperspace/Data/models/akash_unifo_757/llm3br256/checkpoint-1200",
+  "epoch": 3.878787878787879,
+  "eval_steps": 25,
+  "global_step": 1200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0032323232323232323,
+      "grad_norm": 0.11503094434738159,
+      "learning_rate": 1.2936610608020701e-07,
+      "loss": 0.0131,
+      "step": 1
+    },
+    {
+      "epoch": 0.006464646464646465,
+      "grad_norm": 0.13206231594085693,
+      "learning_rate": 2.5873221216041403e-07,
+      "loss": 0.0155,
+      "step": 2
+    },
+    {
+      "epoch": 0.009696969696969697,
+      "grad_norm": 0.1161206066608429,
+      "learning_rate": 3.8809831824062096e-07,
+      "loss": 0.0245,
+      "step": 3
+    },
+    {
+      "epoch": 0.01292929292929293,
+      "grad_norm": 0.1274755299091339,
+      "learning_rate": 5.174644243208281e-07,
+      "loss": 0.0127,
+      "step": 4
+    },
+    {
+      "epoch": 0.01616161616161616,
+      "grad_norm": 0.12534309923648834,
+      "learning_rate": 6.468305304010349e-07,
+      "loss": 0.0137,
+      "step": 5
+    },
+    {
+      "epoch": 0.019393939393939394,
+      "grad_norm": 0.11610464006662369,
+      "learning_rate": 7.761966364812419e-07,
+      "loss": 0.0153,
+      "step": 6
+    },
+    {
+      "epoch": 0.022626262626262626,
+      "grad_norm": 0.10964969545602798,
+      "learning_rate": 9.055627425614489e-07,
+      "loss": 0.0205,
+      "step": 7
+    },
+    {
+      "epoch": 0.02585858585858586,
+      "grad_norm": 0.11367467045783997,
+      "learning_rate": 1.0349288486416561e-06,
+      "loss": 0.0167,
+      "step": 8
+    },
+    {
+      "epoch": 0.02909090909090909,
+      "grad_norm": 0.12127836048603058,
+      "learning_rate": 1.164294954721863e-06,
+      "loss": 0.0155,
+      "step": 9
+    },
+    {
+      "epoch": 0.03232323232323232,
+      "grad_norm": 0.12462533265352249,
+      "learning_rate": 1.2936610608020699e-06,
+      "loss": 0.0169,
+      "step": 10
+    },
+    {
+      "epoch": 0.035555555555555556,
+      "grad_norm": 0.13044221699237823,
+      "learning_rate": 1.423027166882277e-06,
+      "loss": 0.0104,
+      "step": 11
+    },
+    {
+      "epoch": 0.03878787878787879,
+      "grad_norm": 0.08799455314874649,
+      "learning_rate": 1.5523932729624839e-06,
+      "loss": 0.021,
+      "step": 12
+    },
+    {
+      "epoch": 0.04202020202020202,
+      "grad_norm": 0.09403660893440247,
+      "learning_rate": 1.6817593790426907e-06,
+      "loss": 0.0151,
+      "step": 13
+    },
+    {
+      "epoch": 0.04525252525252525,
+      "grad_norm": 0.07015115022659302,
+      "learning_rate": 1.8111254851228978e-06,
+      "loss": 0.0157,
+      "step": 14
+    },
+    {
+      "epoch": 0.048484848484848485,
+      "grad_norm": 0.07983457297086716,
+      "learning_rate": 1.9404915912031045e-06,
+      "loss": 0.0091,
+      "step": 15
+    },
+    {
+      "epoch": 0.05171717171717172,
+      "grad_norm": 0.06726181507110596,
+      "learning_rate": 2.0698576972833122e-06,
+      "loss": 0.0147,
+      "step": 16
+    },
+    {
+      "epoch": 0.05494949494949495,
+      "grad_norm": 0.06473492830991745,
+      "learning_rate": 2.199223803363519e-06,
+      "loss": 0.0091,
+      "step": 17
+    },
+    {
+      "epoch": 0.05818181818181818,
+      "grad_norm": 0.051432088017463684,
+      "learning_rate": 2.328589909443726e-06,
+      "loss": 0.0129,
+      "step": 18
+    },
+    {
+      "epoch": 0.061414141414141414,
+      "grad_norm": 0.048174694180488586,
+      "learning_rate": 2.457956015523933e-06,
+      "loss": 0.0119,
+      "step": 19
+    },
+    {
+      "epoch": 0.06464646464646465,
+      "grad_norm": 0.04263145476579666,
+      "learning_rate": 2.5873221216041398e-06,
+      "loss": 0.0112,
+      "step": 20
+    },
+    {
+      "epoch": 0.06787878787878789,
+      "grad_norm": 0.03820733353495598,
+      "learning_rate": 2.716688227684347e-06,
+      "loss": 0.0091,
+      "step": 21
+    },
+    {
+      "epoch": 0.07111111111111111,
+      "grad_norm": 0.03781511262059212,
+      "learning_rate": 2.846054333764554e-06,
+      "loss": 0.0103,
+      "step": 22
+    },
+    {
+      "epoch": 0.07434343434343435,
+      "grad_norm": 0.03252599388360977,
+      "learning_rate": 2.975420439844761e-06,
+      "loss": 0.01,
+      "step": 23
+    },
+    {
+      "epoch": 0.07757575757575758,
+      "grad_norm": 0.03773924708366394,
+      "learning_rate": 3.1047865459249677e-06,
+      "loss": 0.0115,
+      "step": 24
+    },
+    {
+      "epoch": 0.08080808080808081,
+      "grad_norm": 0.03268961235880852,
+      "learning_rate": 3.234152652005175e-06,
+      "loss": 0.0133,
+      "step": 25
+    },
+    {
+      "epoch": 0.08080808080808081,
+      "eval_loss": 0.01230227667838335,
+      "eval_runtime": 21.2392,
+      "eval_samples_per_second": 4.708,
+      "eval_steps_per_second": 1.177,
+      "step": 25
+    },
+    {
+      "epoch": 0.08404040404040404,
+      "grad_norm": 0.02927996963262558,
+      "learning_rate": 3.3635187580853815e-06,
+      "loss": 0.0092,
+      "step": 26
+    },
+    {
+      "epoch": 0.08727272727272728,
+      "grad_norm": 0.039579834789037704,
+      "learning_rate": 3.492884864165589e-06,
+      "loss": 0.0163,
+      "step": 27
+    },
+    {
+      "epoch": 0.0905050505050505,
+      "grad_norm": 0.024472083896398544,
+      "learning_rate": 3.6222509702457957e-06,
+      "loss": 0.0067,
+      "step": 28
+    },
+    {
+      "epoch": 0.09373737373737374,
+      "grad_norm": 0.03290821984410286,
+      "learning_rate": 3.751617076326003e-06,
+      "loss": 0.0127,
+      "step": 29
+    },
+    {
+      "epoch": 0.09696969696969697,
+      "grad_norm": 0.03189826384186745,
+      "learning_rate": 3.880983182406209e-06,
+      "loss": 0.0096,
+      "step": 30
+    },
+    {
+      "epoch": 0.10020202020202021,
+      "grad_norm": 0.030306048691272736,
+      "learning_rate": 4.010349288486417e-06,
+      "loss": 0.0082,
+      "step": 31
+    },
+    {
+      "epoch": 0.10343434343434343,
+      "grad_norm": 0.03980877622961998,
+      "learning_rate": 4.1397153945666245e-06,
+      "loss": 0.0133,
+      "step": 32
+    },
+    {
+      "epoch": 0.10666666666666667,
+      "grad_norm": 0.030887724831700325,
+      "learning_rate": 4.2690815006468305e-06,
+      "loss": 0.0074,
+      "step": 33
+    },
+    {
+      "epoch": 0.1098989898989899,
+      "grad_norm": 0.027738964185118675,
+      "learning_rate": 4.398447606727038e-06,
+      "loss": 0.0069,
+      "step": 34
+    },
+    {
+      "epoch": 0.11313131313131314,
+      "grad_norm": 0.03853800520300865,
+      "learning_rate": 4.527813712807244e-06,
+      "loss": 0.0128,
+      "step": 35
+    },
+    {
+      "epoch": 0.11636363636363636,
+      "grad_norm": 0.028064019978046417,
+      "learning_rate": 4.657179818887452e-06,
+      "loss": 0.0091,
+      "step": 36
+    },
+    {
+      "epoch": 0.1195959595959596,
+      "grad_norm": 0.026782216504216194,
+      "learning_rate": 4.786545924967659e-06,
+      "loss": 0.0071,
+      "step": 37
+    },
+    {
+      "epoch": 0.12282828282828283,
+      "grad_norm": 0.017248690128326416,
+      "learning_rate": 4.915912031047866e-06,
+      "loss": 0.0033,
+      "step": 38
+    },
+    {
+      "epoch": 0.12606060606060607,
+      "grad_norm": 0.0254344642162323,
+      "learning_rate": 5.045278137128073e-06,
+      "loss": 0.0076,
+      "step": 39
+    },
+    {
+      "epoch": 0.1292929292929293,
+      "grad_norm": 0.02050536312162876,
+      "learning_rate": 5.1746442432082795e-06,
+      "loss": 0.0072,
+      "step": 40
+    },
+    {
+      "epoch": 0.13252525252525252,
+      "grad_norm": 0.02892652526497841,
+      "learning_rate": 5.304010349288486e-06,
+      "loss": 0.0086,
+      "step": 41
+    },
+    {
+      "epoch": 0.13575757575757577,
+      "grad_norm": 0.025958560407161713,
+      "learning_rate": 5.433376455368694e-06,
+      "loss": 0.0092,
+      "step": 42
+    },
+    {
+      "epoch": 0.138989898989899,
+      "grad_norm": 0.026592133566737175,
+      "learning_rate": 5.5627425614489e-06,
+      "loss": 0.0077,
+      "step": 43
+    },
+    {
+      "epoch": 0.14222222222222222,
+      "grad_norm": 0.016826845705509186,
+      "learning_rate": 5.692108667529108e-06,
+      "loss": 0.0048,
+      "step": 44
+    },
+    {
+      "epoch": 0.14545454545454545,
+      "grad_norm": 0.02611837536096573,
+      "learning_rate": 5.821474773609315e-06,
+      "loss": 0.0103,
+      "step": 45
+    },
+    {
+      "epoch": 0.1486868686868687,
+      "grad_norm": 0.018496474251151085,
+      "learning_rate": 5.950840879689522e-06,
+      "loss": 0.0046,
+      "step": 46
+    },
+    {
+      "epoch": 0.15191919191919193,
+      "grad_norm": 0.016068018972873688,
+      "learning_rate": 6.0802069857697286e-06,
+      "loss": 0.0068,
+      "step": 47
+    },
+    {
+      "epoch": 0.15515151515151515,
+      "grad_norm": 0.023762917146086693,
+      "learning_rate": 6.2095730918499354e-06,
+      "loss": 0.0078,
+      "step": 48
+    },
+    {
+      "epoch": 0.15838383838383838,
+      "grad_norm": 0.022558651864528656,
+      "learning_rate": 6.338939197930142e-06,
+      "loss": 0.0079,
+      "step": 49
+    },
+    {
+      "epoch": 0.16161616161616163,
+      "grad_norm": 0.02249864488840103,
+      "learning_rate": 6.46830530401035e-06,
+      "loss": 0.0056,
+      "step": 50
+    },
+    {
+      "epoch": 0.16161616161616163,
+      "eval_loss": 0.009845715016126633,
+      "eval_runtime": 18.6136,
+      "eval_samples_per_second": 5.372,
+      "eval_steps_per_second": 1.343,
+      "step": 50
+    },
+    {
+      "epoch": 0.16484848484848486,
+      "grad_norm": 0.01815730892121792,
+      "learning_rate": 6.597671410090557e-06,
+      "loss": 0.0053,
+      "step": 51
+    },
+    {
+      "epoch": 0.16808080808080808,
+      "grad_norm": 0.023437149822711945,
+      "learning_rate": 6.727037516170763e-06,
+      "loss": 0.0089,
+      "step": 52
+    },
+    {
+      "epoch": 0.1713131313131313,
+      "grad_norm": 0.02191866748034954,
+      "learning_rate": 6.856403622250971e-06,
+      "loss": 0.0116,
+      "step": 53
+    },
+    {
+      "epoch": 0.17454545454545456,
+      "grad_norm": 0.018593238666653633,
+      "learning_rate": 6.985769728331178e-06,
+      "loss": 0.0067,
+      "step": 54
+    },
+    {
+      "epoch": 0.17777777777777778,
+      "grad_norm": 0.019710399210453033,
+      "learning_rate": 7.115135834411385e-06,
+      "loss": 0.0065,
+      "step": 55
+    },
+    {
+      "epoch": 0.181010101010101,
+      "grad_norm": 0.01702931523323059,
+      "learning_rate": 7.244501940491591e-06,
+      "loss": 0.0054,
+      "step": 56
+    },
+    {
+      "epoch": 0.18424242424242424,
+      "grad_norm": 0.020323824137449265,
+      "learning_rate": 7.373868046571798e-06,
+      "loss": 0.0072,
+      "step": 57
+    },
+    {
+      "epoch": 0.1874747474747475,
+      "grad_norm": 0.022320428863167763,
+      "learning_rate": 7.503234152652006e-06,
+      "loss": 0.0107,
+      "step": 58
+    },
+    {
+      "epoch": 0.1907070707070707,
+      "grad_norm": 0.01765402965247631,
+      "learning_rate": 7.632600258732213e-06,
+      "loss": 0.0055,
+      "step": 59
+    },
+    {
+      "epoch": 0.19393939393939394,
+      "grad_norm": 0.019662927836179733,
+      "learning_rate": 7.761966364812418e-06,
+      "loss": 0.0073,
+      "step": 60
+    },
+    {
+      "epoch": 0.19717171717171716,
+      "grad_norm": 0.016214415431022644,
+      "learning_rate": 7.891332470892627e-06,
+      "loss": 0.0083,
+      "step": 61
+    },
+    {
+      "epoch": 0.20040404040404042,
+      "grad_norm": 0.015475176274776459,
+      "learning_rate": 8.020698576972833e-06,
+      "loss": 0.0075,
+      "step": 62
+    },
+    {
+      "epoch": 0.20363636363636364,
+      "grad_norm": 0.016818247735500336,
+      "learning_rate": 8.15006468305304e-06,
+      "loss": 0.0081,
+      "step": 63
+    },
+    {
+      "epoch": 0.20686868686868687,
+      "grad_norm": 0.018314030021429062,
+      "learning_rate": 8.279430789133249e-06,
+      "loss": 0.0064,
+      "step": 64
+    },
+    {
+      "epoch": 0.2101010101010101,
+      "grad_norm": 0.021758757531642914,
+      "learning_rate": 8.408796895213454e-06,
+      "loss": 0.0105,
+      "step": 65
+    },
+    {
+      "epoch": 0.21333333333333335,
+      "grad_norm": 0.01629498228430748,
+      "learning_rate": 8.538163001293661e-06,
+      "loss": 0.0062,
+      "step": 66
+    },
+    {
+      "epoch": 0.21656565656565657,
+      "grad_norm": 0.017469746991991997,
+      "learning_rate": 8.66752910737387e-06,
+      "loss": 0.0058,
+      "step": 67
+    },
+    {
+      "epoch": 0.2197979797979798,
+      "grad_norm": 0.016880586743354797,
+      "learning_rate": 8.796895213454076e-06,
+      "loss": 0.0093,
+      "step": 68
+    },
+    {
+      "epoch": 0.22303030303030302,
+      "grad_norm": 0.020206857472658157,
+      "learning_rate": 8.926261319534282e-06,
+      "loss": 0.0068,
+      "step": 69
+    },
+    {
+      "epoch": 0.22626262626262628,
+      "grad_norm": 0.018911803141236305,
+      "learning_rate": 9.055627425614489e-06,
+      "loss": 0.008,
+      "step": 70
+    },
+    {
+      "epoch": 0.2294949494949495,
+      "grad_norm": 0.017788654193282127,
+      "learning_rate": 9.184993531694697e-06,
+      "loss": 0.0088,
+      "step": 71
+    },
+    {
+      "epoch": 0.23272727272727273,
+      "grad_norm": 0.013764932751655579,
+      "learning_rate": 9.314359637774904e-06,
+      "loss": 0.0053,
+      "step": 72
+    },
+    {
+      "epoch": 0.23595959595959595,
+      "grad_norm": 0.016450250521302223,
+      "learning_rate": 9.44372574385511e-06,
+      "loss": 0.0074,
+      "step": 73
+    },
+    {
+      "epoch": 0.2391919191919192,
+      "grad_norm": 0.02177259512245655,
+      "learning_rate": 9.573091849935318e-06,
+      "loss": 0.0119,
+      "step": 74
+    },
+    {
+      "epoch": 0.24242424242424243,
+      "grad_norm": 0.019282735884189606,
+      "learning_rate": 9.702457956015525e-06,
+      "loss": 0.0113,
+      "step": 75
+    },
+    {
+      "epoch": 0.24242424242424243,
+      "eval_loss": 0.00841269176453352,
+      "eval_runtime": 18.5455,
+      "eval_samples_per_second": 5.392,
+      "eval_steps_per_second": 1.348,
+      "step": 75
+    },
+    {
+      "epoch": 0.24565656565656566,
+      "grad_norm": 0.020014960318803787,
+      "learning_rate": 9.831824062095732e-06,
+      "loss": 0.0097,
+      "step": 76
+    },
+    {
+      "epoch": 0.24888888888888888,
+      "grad_norm": 0.021017547696828842,
+      "learning_rate": 9.961190168175938e-06,
+      "loss": 0.0085,
+      "step": 77
+    },
+    {
+      "epoch": 0.25212121212121213,
+      "grad_norm": 0.016342103481292725,
+      "learning_rate": 1.0090556274256145e-05,
+      "loss": 0.0065,
+      "step": 78
+    },
+    {
+      "epoch": 0.25535353535353533,
+      "grad_norm": 0.020554421469569206,
+      "learning_rate": 1.0219922380336352e-05,
+      "loss": 0.0092,
+      "step": 79
+    },
+    {
+      "epoch": 0.2585858585858586,
+      "grad_norm": 0.011228547431528568,
+      "learning_rate": 1.0349288486416559e-05,
+      "loss": 0.005,
+      "step": 80
+    },
+    {
+      "epoch": 0.26181818181818184,
+      "grad_norm": 0.01503695547580719,
+      "learning_rate": 1.0478654592496766e-05,
+      "loss": 0.0056,
+      "step": 81
+    },
+    {
+      "epoch": 0.26505050505050504,
+      "grad_norm": 0.016742516309022903,
+      "learning_rate": 1.0608020698576973e-05,
+      "loss": 0.0067,
+      "step": 82
+    },
+    {
+      "epoch": 0.2682828282828283,
+      "grad_norm": 0.015310220420360565,
+      "learning_rate": 1.073738680465718e-05,
+      "loss": 0.0084,
+      "step": 83
+    },
+    {
+      "epoch": 0.27151515151515154,
+      "grad_norm": 0.01845848560333252,
+      "learning_rate": 1.0866752910737388e-05,
+      "loss": 0.0069,
+      "step": 84
+    },
+    {
+      "epoch": 0.27474747474747474,
+      "grad_norm": 0.014536215923726559,
+      "learning_rate": 1.0996119016817593e-05,
+      "loss": 0.0057,
+      "step": 85
+    },
+    {
+      "epoch": 0.277979797979798,
+      "grad_norm": 0.016418730840086937,
+      "learning_rate": 1.11254851228978e-05,
+      "loss": 0.0064,
+      "step": 86
+    },
+    {
+      "epoch": 0.2812121212121212,
+      "grad_norm": 0.01798829808831215,
+      "learning_rate": 1.1254851228978009e-05,
+      "loss": 0.0048,
+      "step": 87
+    },
+    {
+      "epoch": 0.28444444444444444,
+      "grad_norm": 0.017619946971535683,
+      "learning_rate": 1.1384217335058216e-05,
+      "loss": 0.0112,
+      "step": 88
+    },
+    {
+      "epoch": 0.2876767676767677,
+      "grad_norm": 0.017026174813508987,
+      "learning_rate": 1.1513583441138421e-05,
+      "loss": 0.005,
+      "step": 89
+    },
+    {
+      "epoch": 0.2909090909090909,
+      "grad_norm": 0.019230445846915245,
+      "learning_rate": 1.164294954721863e-05,
+      "loss": 0.0108,
+      "step": 90
+    },
+    {
+      "epoch": 0.29414141414141415,
+      "grad_norm": 0.014893417246639729,
+      "learning_rate": 1.1772315653298836e-05,
+      "loss": 0.0067,
+      "step": 91
+    },
+    {
+      "epoch": 0.2973737373737374,
+      "grad_norm": 0.01982959918677807,
+      "learning_rate": 1.1901681759379043e-05,
+      "loss": 0.0092,
+      "step": 92
+    },
+    {
+      "epoch": 0.3006060606060606,
+      "grad_norm": 0.016334552317857742,
+      "learning_rate": 1.203104786545925e-05,
+      "loss": 0.0042,
+      "step": 93
+    },
+    {
+      "epoch": 0.30383838383838385,
+      "grad_norm": 0.012507308274507523,
+      "learning_rate": 1.2160413971539457e-05,
+      "loss": 0.0059,
+      "step": 94
+    },
+    {
+      "epoch": 0.30707070707070705,
+      "grad_norm": 0.014092699624598026,
+      "learning_rate": 1.2289780077619664e-05,
+      "loss": 0.0075,
+      "step": 95
+    },
+    {
+      "epoch": 0.3103030303030303,
+      "grad_norm": 0.01649198681116104,
+      "learning_rate": 1.2419146183699871e-05,
+      "loss": 0.0078,
+      "step": 96
+    },
+    {
+      "epoch": 0.31353535353535356,
+      "grad_norm": 0.017579909414052963,
+      "learning_rate": 1.254851228978008e-05,
+      "loss": 0.0067,
+      "step": 97
+    },
+    {
+      "epoch": 0.31676767676767675,
+      "grad_norm": 0.01603030413389206,
+      "learning_rate": 1.2677878395860285e-05,
+      "loss": 0.0056,
+      "step": 98
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.01683826372027397,
+      "learning_rate": 1.2807244501940493e-05,
+      "loss": 0.0087,
+      "step": 99
+    },
+    {
+      "epoch": 0.32323232323232326,
+      "grad_norm": 0.01677022874355316,
+      "learning_rate": 1.29366106080207e-05,
+      "loss": 0.0083,
+      "step": 100
+    },
+    {
+      "epoch": 0.32323232323232326,
+      "eval_loss": 0.00736634898930788,
+      "eval_runtime": 18.5672,
+      "eval_samples_per_second": 5.386,
+      "eval_steps_per_second": 1.346,
+      "step": 100
+    },
+    {
+      "epoch": 0.32646464646464646,
+      "grad_norm": 0.014008735306560993,
+      "learning_rate": 1.3065976714100905e-05,
+      "loss": 0.0043,
+      "step": 101
+    },
+    {
+      "epoch": 0.3296969696969697,
+      "grad_norm": 0.010679924860596657,
+      "learning_rate": 1.3195342820181114e-05,
+      "loss": 0.0034,
+      "step": 102
+    },
+    {
+      "epoch": 0.3329292929292929,
+      "grad_norm": 0.01666383258998394,
+      "learning_rate": 1.332470892626132e-05,
+      "loss": 0.0079,
+      "step": 103
+    },
+    {
+      "epoch": 0.33616161616161616,
+      "grad_norm": 0.016743313521146774,
+      "learning_rate": 1.3454075032341526e-05,
+      "loss": 0.005,
+      "step": 104
+    },
+    {
+      "epoch": 0.3393939393939394,
+      "grad_norm": 0.02069983258843422,
+      "learning_rate": 1.3583441138421735e-05,
+      "loss": 0.0103,
+      "step": 105
+    },
+    {
+      "epoch": 0.3426262626262626,
+      "grad_norm": 0.01835963875055313,
+      "learning_rate": 1.3712807244501941e-05,
+      "loss": 0.0048,
+      "step": 106
+    },
+    {
+      "epoch": 0.34585858585858587,
+      "grad_norm": 0.019786424934864044,
+      "learning_rate": 1.384217335058215e-05,
+      "loss": 0.0143,
+      "step": 107
+    },
+    {
+      "epoch": 0.3490909090909091,
+      "grad_norm": 0.019772058352828026,
+      "learning_rate": 1.3971539456662355e-05,
+      "loss": 0.009,
+      "step": 108
+    },
+    {
+      "epoch": 0.3523232323232323,
+      "grad_norm": 0.018820039927959442,
+      "learning_rate": 1.4100905562742562e-05,
+      "loss": 0.0103,
+      "step": 109
+    },
+    {
+      "epoch": 0.35555555555555557,
+      "grad_norm": 0.01463607419282198,
+      "learning_rate": 1.423027166882277e-05,
+      "loss": 0.0064,
+      "step": 110
+    },
+    {
+      "epoch": 0.35878787878787877,
+      "grad_norm": 0.015627332031726837,
+      "learning_rate": 1.4359637774902976e-05,
+      "loss": 0.0064,
+      "step": 111
+    },
+    {
+      "epoch": 0.362020202020202,
+      "grad_norm": 0.014721403829753399,
+      "learning_rate": 1.4489003880983183e-05,
+      "loss": 0.0043,
+      "step": 112
+    },
+    {
+      "epoch": 0.3652525252525253,
+      "grad_norm": 0.015796532854437828,
+      "learning_rate": 1.4618369987063391e-05,
+      "loss": 0.0072,
+      "step": 113
+    },
+    {
+      "epoch": 0.36848484848484847,
+      "grad_norm": 0.0158319603651762,
+      "learning_rate": 1.4747736093143596e-05,
+      "loss": 0.0059,
+      "step": 114
+    },
+    {
+      "epoch": 0.3717171717171717,
+      "grad_norm": 0.015959493815898895,
+      "learning_rate": 1.4877102199223805e-05,
+      "loss": 0.0051,
+      "step": 115
+    },
+    {
+      "epoch": 0.374949494949495,
+      "grad_norm": 0.016709132120013237,
+      "learning_rate": 1.5006468305304012e-05,
+      "loss": 0.0077,
+      "step": 116
+    },
+    {
+      "epoch": 0.3781818181818182,
+      "grad_norm": 0.02095826528966427,
+      "learning_rate": 1.5135834411384217e-05,
+      "loss": 0.012,
+      "step": 117
+    },
+    {
+      "epoch": 0.3814141414141414,
+      "grad_norm": 0.012002995237708092,
+      "learning_rate": 1.5265200517464426e-05,
+      "loss": 0.0048,
+      "step": 118
+    },
+    {
+      "epoch": 0.3846464646464646,
+      "grad_norm": 0.019866062328219414,
+      "learning_rate": 1.5394566623544633e-05,
+      "loss": 0.0137,
+      "step": 119
+    },
+    {
+      "epoch": 0.3878787878787879,
+      "grad_norm": 0.013328887522220612,
+      "learning_rate": 1.5523932729624836e-05,
+      "loss": 0.0042,
+      "step": 120
+    },
+    {
+      "epoch": 0.39111111111111113,
+      "grad_norm": 0.012495558708906174,
+      "learning_rate": 1.5653298835705046e-05,
+      "loss": 0.0029,
+      "step": 121
+    },
+    {
+      "epoch": 0.39434343434343433,
+      "grad_norm": 0.01434282772243023,
+      "learning_rate": 1.5782664941785253e-05,
+      "loss": 0.0043,
+      "step": 122
+    },
+    {
+      "epoch": 0.3975757575757576,
+      "grad_norm": 0.019523974508047104,
+      "learning_rate": 1.591203104786546e-05,
+      "loss": 0.0098,
+      "step": 123
+    },
+    {
+      "epoch": 0.40080808080808084,
+      "grad_norm": 0.013115576468408108,
+      "learning_rate": 1.6041397153945667e-05,
+      "loss": 0.0038,
+      "step": 124
+    },
+    {
+      "epoch": 0.40404040404040403,
+      "grad_norm": 0.016748910769820213,
+      "learning_rate": 1.6170763260025874e-05,
+      "loss": 0.0086,
+      "step": 125
+    },
+    {
+      "epoch": 0.40404040404040403,
+      "eval_loss": 0.006461107637733221,
+      "eval_runtime": 18.517,
+      "eval_samples_per_second": 5.4,
+      "eval_steps_per_second": 1.35,
+      "step": 125
+    },
+    {
+      "epoch": 0.4072727272727273,
+      "grad_norm": 0.013579982332885265,
+      "learning_rate": 1.630012936610608e-05,
+      "loss": 0.0054,
+      "step": 126
+    },
+    {
+      "epoch": 0.4105050505050505,
+      "grad_norm": 0.0175962932407856,
+      "learning_rate": 1.6429495472186288e-05,
+      "loss": 0.0058,
+      "step": 127
+    },
+    {
+      "epoch": 0.41373737373737374,
+      "grad_norm": 0.014661675319075584,
+      "learning_rate": 1.6558861578266498e-05,
+      "loss": 0.0067,
+      "step": 128
+    },
+    {
+      "epoch": 0.416969696969697,
+      "grad_norm": 0.016726892441511154,
+      "learning_rate": 1.66882276843467e-05,
+      "loss": 0.0079,
+      "step": 129
+    },
+    {
+      "epoch": 0.4202020202020202,
+      "grad_norm": 0.012960881926119328,
+      "learning_rate": 1.6817593790426908e-05,
+      "loss": 0.0053,
+      "step": 130
+    },
+    {
+      "epoch": 0.42343434343434344,
+      "grad_norm": 0.019023625180125237,
+      "learning_rate": 1.694695989650712e-05,
+      "loss": 0.0097,
+      "step": 131
+    },
+    {
+      "epoch": 0.4266666666666667,
+      "grad_norm": 0.018510041758418083,
+      "learning_rate": 1.7076326002587322e-05,
+      "loss": 0.0053,
+      "step": 132
+    },
+    {
+      "epoch": 0.4298989898989899,
+      "grad_norm": 0.016019567847251892,
+      "learning_rate": 1.720569210866753e-05,
+      "loss": 0.0063,
+      "step": 133
+    },
+    {
+      "epoch": 0.43313131313131314,
+      "grad_norm": 0.01722690835595131,
+      "learning_rate": 1.733505821474774e-05,
+      "loss": 0.0063,
+      "step": 134
+    },
+    {
+      "epoch": 0.43636363636363634,
+      "grad_norm": 0.015361025929450989,
+      "learning_rate": 1.7464424320827943e-05,
+      "loss": 0.0037,
+      "step": 135
+    },
+    {
+      "epoch": 0.4395959595959596,
+      "grad_norm": 0.014419769868254662,
+      "learning_rate": 1.7593790426908153e-05,
+      "loss": 0.0046,
+      "step": 136
+    },
+    {
+      "epoch": 0.44282828282828285,
+      "grad_norm": 0.01369231566786766,
+      "learning_rate": 1.7723156532988356e-05,
+      "loss": 0.0039,
+      "step": 137
+    },
+    {
+      "epoch": 0.44606060606060605,
+      "grad_norm": 0.01594926416873932,
+      "learning_rate": 1.7852522639068563e-05,
+      "loss": 0.0037,
+      "step": 138
+    },
+    {
+      "epoch": 0.4492929292929293,
+      "grad_norm": 0.01753450371325016,
+      "learning_rate": 1.7981888745148774e-05,
+      "loss": 0.0063,
+      "step": 139
+    },
+    {
+      "epoch": 0.45252525252525255,
+      "grad_norm": 0.012597791850566864,
+      "learning_rate": 1.8111254851228977e-05,
+      "loss": 0.0039,
+      "step": 140
+    },
+    {
+      "epoch": 0.45575757575757575,
+      "grad_norm": 0.015224005095660686,
+      "learning_rate": 1.8240620957309184e-05,
+      "loss": 0.008,
+      "step": 141
+    },
+    {
+      "epoch": 0.458989898989899,
+      "grad_norm": 0.01901715062558651,
+      "learning_rate": 1.8369987063389394e-05,
+      "loss": 0.0104,
+      "step": 142
+    },
+    {
+      "epoch": 0.4622222222222222,
+      "grad_norm": 0.014205937273800373,
+      "learning_rate": 1.8499353169469598e-05,
+      "loss": 0.006,
+      "step": 143
+    },
+    {
+      "epoch": 0.46545454545454545,
+      "grad_norm": 0.017999831587076187,
+      "learning_rate": 1.8628719275549808e-05,
+      "loss": 0.0067,
+      "step": 144
+    },
+    {
+      "epoch": 0.4686868686868687,
+      "grad_norm": 0.01627945713698864,
+      "learning_rate": 1.8758085381630015e-05,
+      "loss": 0.0066,
+      "step": 145
+    },
+    {
+      "epoch": 0.4719191919191919,
+      "grad_norm": 0.015483185648918152,
+      "learning_rate": 1.888745148771022e-05,
+      "loss": 0.0048,
+      "step": 146
+    },
+    {
+      "epoch": 0.47515151515151516,
+      "grad_norm": 0.01489502564072609,
+      "learning_rate": 1.901681759379043e-05,
+      "loss": 0.0042,
+      "step": 147
+    },
+    {
+      "epoch": 0.4783838383838384,
+      "grad_norm": 0.013736383058130741,
+      "learning_rate": 1.9146183699870636e-05,
+      "loss": 0.0054,
+      "step": 148
+    },
+    {
+      "epoch": 0.4816161616161616,
+      "grad_norm": 0.014781012199819088,
+      "learning_rate": 1.927554980595084e-05,
+      "loss": 0.0038,
+      "step": 149
+    },
+    {
+      "epoch": 0.48484848484848486,
+      "grad_norm": 0.013498370535671711,
+      "learning_rate": 1.940491591203105e-05,
+      "loss": 0.0067,
+      "step": 150
+    },
+    {
+      "epoch": 0.48484848484848486,
+      "eval_loss": 0.006012982223182917,
+      "eval_runtime": 18.5388,
+      "eval_samples_per_second": 5.394,
+      "eval_steps_per_second": 1.349,
+      "step": 150
+    },
+    {
+      "epoch": 0.48808080808080806,
+      "grad_norm": 0.019391966983675957,
+      "learning_rate": 1.9534282018111256e-05,
+      "loss": 0.0062,
+      "step": 151
+    },
+    {
+      "epoch": 0.4913131313131313,
+      "grad_norm": 0.01793098822236061,
+      "learning_rate": 1.9663648124191463e-05,
+      "loss": 0.0048,
+      "step": 152
+    },
+    {
+      "epoch": 0.49454545454545457,
+      "grad_norm": 0.025553777813911438,
+      "learning_rate": 1.979301423027167e-05,
+      "loss": 0.0088,
+      "step": 153
+    },
+    {
+      "epoch": 0.49777777777777776,
+      "grad_norm": 0.015286175534129143,
+      "learning_rate": 1.9922380336351877e-05,
+      "loss": 0.0052,
+      "step": 154
+    },
+    {
+      "epoch": 0.501010101010101,
+      "grad_norm": 0.013409867882728577,
+      "learning_rate": 2.0051746442432084e-05,
+      "loss": 0.0029,
+      "step": 155
+    },
+    {
+      "epoch": 0.5042424242424243,
+      "grad_norm": 0.02216433919966221,
+      "learning_rate": 2.018111254851229e-05,
+      "loss": 0.0068,
+      "step": 156
+    },
+    {
+      "epoch": 0.5074747474747475,
+      "grad_norm": 0.01498087402433157,
+      "learning_rate": 2.0310478654592497e-05,
+      "loss": 0.0037,
+      "step": 157
+    },
+    {
+      "epoch": 0.5107070707070707,
+      "grad_norm": 0.06357154995203018,
+      "learning_rate": 2.0439844760672704e-05,
+      "loss": 0.0074,
+      "step": 158
+    },
+    {
+      "epoch": 0.5139393939393939,
+      "grad_norm": 0.014753567986190319,
+      "learning_rate": 2.056921086675291e-05,
+      "loss": 0.0042,
+      "step": 159
+    },
+    {
+      "epoch": 0.5171717171717172,
+      "grad_norm": 0.020152902230620384,
+      "learning_rate": 2.0698576972833118e-05,
+      "loss": 0.0072,
+      "step": 160
+    },
+    {
+      "epoch": 0.5204040404040404,
+      "grad_norm": 0.02215396985411644,
+      "learning_rate": 2.0827943078913325e-05,
+      "loss": 0.0063,
+      "step": 161
+    },
+    {
+      "epoch": 0.5236363636363637,
+      "grad_norm": 0.01737254299223423,
+      "learning_rate": 2.0957309184993532e-05,
+      "loss": 0.0045,
+      "step": 162
+    },
+    {
+      "epoch": 0.5268686868686868,
+      "grad_norm": 0.01855519600212574,
+      "learning_rate": 2.108667529107374e-05,
+      "loss": 0.0079,
+      "step": 163
+    },
+    {
+      "epoch": 0.5301010101010101,
+      "grad_norm": 0.0156853087246418,
+      "learning_rate": 2.1216041397153946e-05,
+      "loss": 0.0059,
+      "step": 164
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.015692561864852905,
+      "learning_rate": 2.1345407503234156e-05,
+      "loss": 0.0033,
+      "step": 165
+    },
+    {
+      "epoch": 0.5365656565656566,
+      "grad_norm": 0.01826147921383381,
+      "learning_rate": 2.147477360931436e-05,
+      "loss": 0.0082,
+      "step": 166
+    },
+    {
+      "epoch": 0.5397979797979798,
+      "grad_norm": 0.017402255907654762,
+      "learning_rate": 2.1604139715394566e-05,
+      "loss": 0.0085,
+      "step": 167
+    },
+    {
+      "epoch": 0.5430303030303031,
+      "grad_norm": 0.020618412643671036,
+      "learning_rate": 2.1733505821474777e-05,
+      "loss": 0.0057,
+      "step": 168
+    },
+    {
+      "epoch": 0.5462626262626262,
+      "grad_norm": 0.013431048952043056,
+      "learning_rate": 2.186287192755498e-05,
+      "loss": 0.0057,
+      "step": 169
+    },
+    {
+      "epoch": 0.5494949494949495,
+      "grad_norm": 0.020214086398482323,
+      "learning_rate": 2.1992238033635187e-05,
+      "loss": 0.0098,
+      "step": 170
+    },
+    {
+      "epoch": 0.5527272727272727,
+      "grad_norm": 0.018985526636242867,
+      "learning_rate": 2.2121604139715397e-05,
+      "loss": 0.0047,
+      "step": 171
+    },
+    {
+      "epoch": 0.555959595959596,
+      "grad_norm": 0.01286108884960413,
+      "learning_rate": 2.22509702457956e-05,
+      "loss": 0.0045,
+      "step": 172
+    },
+    {
+      "epoch": 0.5591919191919192,
+      "grad_norm": 0.017977848649024963,
+      "learning_rate": 2.238033635187581e-05,
+      "loss": 0.0075,
+      "step": 173
+    },
+    {
+      "epoch": 0.5624242424242424,
+      "grad_norm": 0.01748410426080227,
+      "learning_rate": 2.2509702457956018e-05,
+      "loss": 0.0054,
+      "step": 174
+    },
+    {
+      "epoch": 0.5656565656565656,
+      "grad_norm": 0.014239229261875153,
+      "learning_rate": 2.263906856403622e-05,
+      "loss": 0.0038,
+      "step": 175
+    },
+    {
+      "epoch": 0.5656565656565656,
+      "eval_loss": 0.00581002701073885,
+      "eval_runtime": 18.5437,
+      "eval_samples_per_second": 5.393,
+      "eval_steps_per_second": 1.348,
+      "step": 175
+    },
+    {
+      "epoch": 0.5688888888888889,
+      "grad_norm": 0.015481271781027317,
+      "learning_rate": 2.276843467011643e-05,
+      "loss": 0.0049,
+      "step": 176
+    },
+    {
+      "epoch": 0.5721212121212121,
+      "grad_norm": 0.01505062822252512,
+      "learning_rate": 2.289780077619664e-05,
+      "loss": 0.0042,
+      "step": 177
+    },
+    {
+      "epoch": 0.5753535353535354,
+      "grad_norm": 0.016980962827801704,
+      "learning_rate": 2.3027166882276842e-05,
+      "loss": 0.0068,
+      "step": 178
+    },
+    {
+      "epoch": 0.5785858585858585,
+      "grad_norm": 0.017680799588561058,
+      "learning_rate": 2.3156532988357052e-05,
+      "loss": 0.006,
+      "step": 179
+    },
+    {
+      "epoch": 0.5818181818181818,
+      "grad_norm": 0.017391802743077278,
+      "learning_rate": 2.328589909443726e-05,
+      "loss": 0.0088,
+      "step": 180
+    },
+    {
+      "epoch": 0.585050505050505,
+      "grad_norm": 0.015167100355029106,
+      "learning_rate": 2.3415265200517466e-05,
+      "loss": 0.0036,
+      "step": 181
+    },
+    {
+      "epoch": 0.5882828282828283,
+      "grad_norm": 0.0169716514647007,
+      "learning_rate": 2.3544631306597673e-05,
+      "loss": 0.0032,
+      "step": 182
+    },
+    {
+      "epoch": 0.5915151515151515,
+      "grad_norm": 0.013683251105248928,
+      "learning_rate": 2.367399741267788e-05,
+      "loss": 0.0026,
+      "step": 183
+    },
+    {
+      "epoch": 0.5947474747474748,
+      "grad_norm": 0.01659058965742588,
+      "learning_rate": 2.3803363518758087e-05,
+      "loss": 0.0058,
+      "step": 184
+    },
+    {
+      "epoch": 0.597979797979798,
+      "grad_norm": 0.019827809184789658,
+      "learning_rate": 2.3932729624838294e-05,
+      "loss": 0.007,
+      "step": 185
+    },
+    {
+      "epoch": 0.6012121212121212,
+      "grad_norm": 0.01573983021080494,
+      "learning_rate": 2.40620957309185e-05,
+      "loss": 0.0041,
+      "step": 186
+    },
+    {
+      "epoch": 0.6044444444444445,
+      "grad_norm": 0.016037194058299065,
+      "learning_rate": 2.4191461836998707e-05,
+      "loss": 0.0048,
+      "step": 187
+    },
+    {
+      "epoch": 0.6076767676767677,
+      "grad_norm": 0.017361117526888847,
+      "learning_rate": 2.4320827943078914e-05,
+      "loss": 0.0041,
+      "step": 188
+    },
+    {
+      "epoch": 0.610909090909091,
+      "grad_norm": 0.013419854454696178,
+      "learning_rate": 2.445019404915912e-05,
+      "loss": 0.0057,
+      "step": 189
+    },
+    {
+      "epoch": 0.6141414141414141,
+      "grad_norm": 0.014191025868058205,
+      "learning_rate": 2.4579560155239328e-05,
+      "loss": 0.0026,
+      "step": 190
+    },
+    {
+      "epoch": 0.6173737373737374,
+      "grad_norm": 0.017151927575469017,
+      "learning_rate": 2.4708926261319535e-05,
+      "loss": 0.0077,
+      "step": 191
+    },
+    {
+      "epoch": 0.6206060606060606,
+      "grad_norm": 0.01370987668633461,
+      "learning_rate": 2.4838292367399742e-05,
+      "loss": 0.0049,
+      "step": 192
+    },
+    {
+      "epoch": 0.6238383838383839,
+      "grad_norm": 0.015623660758137703,
+      "learning_rate": 2.496765847347995e-05,
+      "loss": 0.0038,
+      "step": 193
+    },
+    {
+      "epoch": 0.6270707070707071,
+      "grad_norm": 0.015143529511988163,
+      "learning_rate": 2.509702457956016e-05,
+      "loss": 0.0036,
+      "step": 194
+    },
+    {
+      "epoch": 0.6303030303030303,
+      "grad_norm": 0.025124063715338707,
+      "learning_rate": 2.5226390685640362e-05,
+      "loss": 0.0065,
+      "step": 195
+    },
+    {
+      "epoch": 0.6335353535353535,
+      "grad_norm": 0.016248000785708427,
+      "learning_rate": 2.535575679172057e-05,
+      "loss": 0.005,
+      "step": 196
+    },
+    {
+      "epoch": 0.6367676767676768,
+      "grad_norm": 0.018650708720088005,
+      "learning_rate": 2.548512289780078e-05,
+      "loss": 0.0038,
+      "step": 197
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.019356263801455498,
+      "learning_rate": 2.5614489003880986e-05,
+      "loss": 0.0034,
+      "step": 198
+    },
+    {
+      "epoch": 0.6432323232323233,
+      "grad_norm": 0.014984376728534698,
+      "learning_rate": 2.574385510996119e-05,
+      "loss": 0.0041,
+      "step": 199
+    },
+    {
+      "epoch": 0.6464646464646465,
+      "grad_norm": 0.014227643609046936,
+      "learning_rate": 2.58732212160414e-05,
+      "loss": 0.0043,
+      "step": 200
+    },
+    {
+      "epoch": 0.6464646464646465,
+      "eval_loss": 0.005420052912086248,
+      "eval_runtime": 18.5123,
+      "eval_samples_per_second": 5.402,
+      "eval_steps_per_second": 1.35,
+      "step": 200
+    },
+    {
+      "epoch": 0.6496969696969697,
+      "grad_norm": 0.01470093335956335,
+      "learning_rate": 2.6002587322121607e-05,
+      "loss": 0.0058,
+      "step": 201
+    },
+    {
+      "epoch": 0.6529292929292929,
+      "grad_norm": 0.01742161437869072,
+      "learning_rate": 2.613195342820181e-05,
+      "loss": 0.006,
+      "step": 202
+    },
+    {
+      "epoch": 0.6561616161616162,
+      "grad_norm": 0.01952281780540943,
+      "learning_rate": 2.626131953428202e-05,
+      "loss": 0.0035,
+      "step": 203
+    },
+    {
+      "epoch": 0.6593939393939394,
+      "grad_norm": 0.015216818079352379,
+      "learning_rate": 2.6390685640362228e-05,
+      "loss": 0.0052,
+      "step": 204
+    },
+    {
+      "epoch": 0.6626262626262627,
+      "grad_norm": 0.013172456994652748,
+      "learning_rate": 2.652005174644243e-05,
+      "loss": 0.0041,
+      "step": 205
+    },
+    {
+      "epoch": 0.6658585858585858,
+      "grad_norm": 0.01898268423974514,
+      "learning_rate": 2.664941785252264e-05,
+      "loss": 0.0068,
+      "step": 206
+    },
+    {
+      "epoch": 0.6690909090909091,
+      "grad_norm": 0.02104322984814644,
+      "learning_rate": 2.677878395860285e-05,
+      "loss": 0.0059,
+      "step": 207
+    },
+    {
+      "epoch": 0.6723232323232323,
+      "grad_norm": 0.014322024770081043,
+      "learning_rate": 2.6908150064683052e-05,
+      "loss": 0.0038,
+      "step": 208
+    },
+    {
+      "epoch": 0.6755555555555556,
+      "grad_norm": 0.011105583049356937,
+      "learning_rate": 2.7037516170763262e-05,
+      "loss": 0.0016,
+      "step": 209
+    },
+    {
+      "epoch": 0.6787878787878788,
+      "grad_norm": 0.012854715809226036,
+      "learning_rate": 2.716688227684347e-05,
+      "loss": 0.0036,
+      "step": 210
+    },
+    {
+      "epoch": 0.682020202020202,
+      "grad_norm": 0.018748968839645386,
+      "learning_rate": 2.7296248382923673e-05,
+      "loss": 0.0044,
+      "step": 211
+    },
+    {
+      "epoch": 0.6852525252525252,
+      "grad_norm": 0.013730528764426708,
+      "learning_rate": 2.7425614489003883e-05,
+      "loss": 0.0024,
+      "step": 212
+    },
+    {
+      "epoch": 0.6884848484848485,
+      "grad_norm": 0.01574782282114029,
+      "learning_rate": 2.755498059508409e-05,
+      "loss": 0.0049,
+      "step": 213
+    },
+    {
+      "epoch": 0.6917171717171717,
+      "grad_norm": 0.016125742346048355,
+      "learning_rate": 2.76843467011643e-05,
+      "loss": 0.0047,
+      "step": 214
+    },
+    {
+      "epoch": 0.694949494949495,
+      "grad_norm": 0.01681477390229702,
+      "learning_rate": 2.7813712807244503e-05,
+      "loss": 0.004,
+      "step": 215
+    },
+    {
+      "epoch": 0.6981818181818182,
+      "grad_norm": 0.013295124284923077,
+      "learning_rate": 2.794307891332471e-05,
+      "loss": 0.0027,
+      "step": 216
+    },
+    {
+      "epoch": 0.7014141414141414,
+      "grad_norm": 0.01294485479593277,
+      "learning_rate": 2.807244501940492e-05,
+      "loss": 0.0039,
+      "step": 217
+    },
+    {
+      "epoch": 0.7046464646464646,
+      "grad_norm": 0.011234375648200512,
+      "learning_rate": 2.8201811125485124e-05,
+      "loss": 0.0026,
+      "step": 218
+    },
+    {
+      "epoch": 0.7078787878787879,
+      "grad_norm": 0.01438144315034151,
+      "learning_rate": 2.833117723156533e-05,
+      "loss": 0.0052,
+      "step": 219
+    },
+    {
+      "epoch": 0.7111111111111111,
+      "grad_norm": 0.013788231648504734,
+      "learning_rate": 2.846054333764554e-05,
+      "loss": 0.0037,
+      "step": 220
+    },
+    {
+      "epoch": 0.7143434343434344,
+      "grad_norm": 0.010624570772051811,
+      "learning_rate": 2.8589909443725745e-05,
+      "loss": 0.0037,
+      "step": 221
+    },
+    {
+      "epoch": 0.7175757575757575,
+      "grad_norm": 0.01703670434653759,
+      "learning_rate": 2.871927554980595e-05,
+      "loss": 0.0035,
+      "step": 222
+    },
+    {
+      "epoch": 0.7208080808080808,
+      "grad_norm": 0.020146546885371208,
+      "learning_rate": 2.8848641655886162e-05,
+      "loss": 0.0057,
+      "step": 223
+    },
+    {
+      "epoch": 0.724040404040404,
+      "grad_norm": 0.014769370667636395,
+      "learning_rate": 2.8978007761966365e-05,
+      "loss": 0.007,
+      "step": 224
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "grad_norm": 0.017661113291978836,
+      "learning_rate": 2.9107373868046572e-05,
+      "loss": 0.004,
+      "step": 225
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "eval_loss": 0.0052576931193470955,
+      "eval_runtime": 18.5949,
+      "eval_samples_per_second": 5.378,
+      "eval_steps_per_second": 1.344,
+      "step": 225
+    },
+    {
+      "epoch": 0.7305050505050505,
+      "grad_norm": 0.01635902002453804,
+      "learning_rate": 2.9236739974126783e-05,
+      "loss": 0.0044,
+      "step": 226
+    },
+    {
+      "epoch": 0.7337373737373737,
+      "grad_norm": 0.013843000866472721,
+      "learning_rate": 2.936610608020699e-05,
+      "loss": 0.0051,
+      "step": 227
+    },
+    {
+      "epoch": 0.7369696969696969,
+      "grad_norm": 0.01753757894039154,
+      "learning_rate": 2.9495472186287193e-05,
+      "loss": 0.0056,
+      "step": 228
+    },
+    {
+      "epoch": 0.7402020202020202,
+      "grad_norm": 0.021675704047083855,
+      "learning_rate": 2.9624838292367403e-05,
+      "loss": 0.0084,
+      "step": 229
+    },
+    {
+      "epoch": 0.7434343434343434,
+      "grad_norm": 0.01598481833934784,
+      "learning_rate": 2.975420439844761e-05,
+      "loss": 0.006,
+      "step": 230
+    },
+    {
+      "epoch": 0.7466666666666667,
+      "grad_norm": 0.016229236498475075,
+      "learning_rate": 2.9883570504527814e-05,
+      "loss": 0.0081,
+      "step": 231
+    },
+    {
+      "epoch": 0.74989898989899,
+      "grad_norm": 0.017028771340847015,
+      "learning_rate": 3.0012936610608024e-05,
+      "loss": 0.0081,
+      "step": 232
+    },
+    {
+      "epoch": 0.7531313131313131,
+      "grad_norm": 0.009784836322069168,
+      "learning_rate": 3.014230271668823e-05,
+      "loss": 0.0022,
+      "step": 233
+    },
+    {
+      "epoch": 0.7563636363636363,
+      "grad_norm": 0.015256540849804878,
+      "learning_rate": 3.0271668822768434e-05,
+      "loss": 0.0091,
+      "step": 234
+    },
+    {
+      "epoch": 0.7595959595959596,
+      "grad_norm": 0.019257735460996628,
+      "learning_rate": 3.0401034928848644e-05,
+      "loss": 0.0096,
+      "step": 235
+    },
+    {
+      "epoch": 0.7628282828282829,
+      "grad_norm": 0.01391427218914032,
+      "learning_rate": 3.053040103492885e-05,
+      "loss": 0.0052,
+      "step": 236
+    },
+    {
+      "epoch": 0.7660606060606061,
+      "grad_norm": 0.012768765911459923,
+      "learning_rate": 3.0659767141009055e-05,
+      "loss": 0.0043,
+      "step": 237
+    },
+    {
+      "epoch": 0.7692929292929293,
+      "grad_norm": 0.015724869444966316,
+      "learning_rate": 3.0789133247089265e-05,
+      "loss": 0.0066,
+      "step": 238
+    },
+    {
+      "epoch": 0.7725252525252525,
+      "grad_norm": 0.016083568334579468,
+      "learning_rate": 3.0918499353169475e-05,
+      "loss": 0.0082,
+      "step": 239
+    },
+    {
+      "epoch": 0.7757575757575758,
+      "grad_norm": 0.01483457162976265,
+      "learning_rate": 3.104786545924967e-05,
+      "loss": 0.0046,
+      "step": 240
+    },
+    {
+      "epoch": 0.778989898989899,
+      "grad_norm": 0.015515637584030628,
+      "learning_rate": 3.117723156532988e-05,
+      "loss": 0.0048,
+      "step": 241
+    },
+    {
+      "epoch": 0.7822222222222223,
+      "grad_norm": 0.016584735363721848,
+      "learning_rate": 3.130659767141009e-05,
+      "loss": 0.0052,
+      "step": 242
+    },
+    {
+      "epoch": 0.7854545454545454,
+      "grad_norm": 0.014646550640463829,
+      "learning_rate": 3.14359637774903e-05,
+      "loss": 0.0022,
+      "step": 243
+    },
+    {
+      "epoch": 0.7886868686868687,
+      "grad_norm": 0.014621698297560215,
+      "learning_rate": 3.1565329883570506e-05,
+      "loss": 0.006,
+      "step": 244
+    },
+    {
+      "epoch": 0.7919191919191919,
+      "grad_norm": 0.015543024055659771,
+      "learning_rate": 3.169469598965072e-05,
+      "loss": 0.0051,
+      "step": 245
+    },
+    {
+      "epoch": 0.7951515151515152,
+      "grad_norm": 0.02005600556731224,
+      "learning_rate": 3.182406209573092e-05,
+      "loss": 0.0119,
+      "step": 246
+    },
+    {
+      "epoch": 0.7983838383838384,
+      "grad_norm": 0.01678292639553547,
+      "learning_rate": 3.1953428201811124e-05,
+      "loss": 0.0052,
+      "step": 247
+    },
+    {
+      "epoch": 0.8016161616161617,
+      "grad_norm": 0.013058966025710106,
+      "learning_rate": 3.2082794307891334e-05,
+      "loss": 0.0033,
+      "step": 248
+    },
+    {
+      "epoch": 0.8048484848484848,
+      "grad_norm": 0.019511310383677483,
+      "learning_rate": 3.2212160413971544e-05,
+      "loss": 0.0048,
+      "step": 249
+    },
+    {
+      "epoch": 0.8080808080808081,
+      "grad_norm": 0.0210106261074543,
+      "learning_rate": 3.234152652005175e-05,
+      "loss": 0.0095,
+      "step": 250
+    },
+    {
+      "epoch": 0.8080808080808081,
+      "eval_loss": 0.0050358218140900135,
+      "eval_runtime": 18.5332,
+      "eval_samples_per_second": 5.396,
+      "eval_steps_per_second": 1.349,
+      "step": 250
+    },
+    {
+      "epoch": 0.8113131313131313,
+      "grad_norm": 0.008065695874392986,
+      "learning_rate": 3.247089262613196e-05,
+      "loss": 0.0022,
+      "step": 251
+    },
+    {
+      "epoch": 0.8145454545454546,
+      "grad_norm": 0.013688714243471622,
+      "learning_rate": 3.260025873221216e-05,
+      "loss": 0.0042,
+      "step": 252
+    },
+    {
+      "epoch": 0.8177777777777778,
+      "grad_norm": 0.023383015766739845,
+      "learning_rate": 3.2729624838292365e-05,
+      "loss": 0.006,
+      "step": 253
+    },
+    {
+      "epoch": 0.821010101010101,
+      "grad_norm": 0.011116388253867626,
+      "learning_rate": 3.2858990944372575e-05,
+      "loss": 0.0029,
+      "step": 254
+    },
+    {
+      "epoch": 0.8242424242424242,
+      "grad_norm": 0.01456495188176632,
+      "learning_rate": 3.2988357050452786e-05,
+      "loss": 0.0038,
+      "step": 255
+    },
+    {
+      "epoch": 0.8274747474747475,
+      "grad_norm": 0.014656679704785347,
+      "learning_rate": 3.3117723156532996e-05,
+      "loss": 0.0045,
+      "step": 256
+    },
+    {
+      "epoch": 0.8307070707070707,
+      "grad_norm": 0.024107573553919792,
+      "learning_rate": 3.324708926261319e-05,
+      "loss": 0.0047,
+      "step": 257
+    },
+    {
+      "epoch": 0.833939393939394,
+      "grad_norm": 0.015572698786854744,
+      "learning_rate": 3.33764553686934e-05,
+      "loss": 0.0064,
+      "step": 258
+    },
+    {
+      "epoch": 0.8371717171717171,
+      "grad_norm": 0.01332643162459135,
+      "learning_rate": 3.350582147477361e-05,
+      "loss": 0.0048,
+      "step": 259
+    },
+    {
+      "epoch": 0.8404040404040404,
+      "grad_norm": 0.018154941499233246,
+      "learning_rate": 3.3635187580853817e-05,
+      "loss": 0.006,
+      "step": 260
+    },
+    {
+      "epoch": 0.8436363636363636,
+      "grad_norm": 0.015191690064966679,
+      "learning_rate": 3.376455368693403e-05,
+      "loss": 0.0039,
+      "step": 261
+    },
+    {
+      "epoch": 0.8468686868686869,
+      "grad_norm": 0.014015432447195053,
+      "learning_rate": 3.389391979301424e-05,
+      "loss": 0.0046,
+      "step": 262
+    },
+    {
+      "epoch": 0.8501010101010101,
+      "grad_norm": 0.022445395588874817,
+      "learning_rate": 3.4023285899094434e-05,
+      "loss": 0.0113,
+      "step": 263
+    },
+    {
+      "epoch": 0.8533333333333334,
+      "grad_norm": 0.015556816942989826,
+      "learning_rate": 3.4152652005174644e-05,
+      "loss": 0.0052,
+      "step": 264
+    },
+    {
+      "epoch": 0.8565656565656565,
+      "grad_norm": 0.01582123525440693,
+      "learning_rate": 3.4282018111254854e-05,
+      "loss": 0.0037,
+      "step": 265
+    },
+    {
+      "epoch": 0.8597979797979798,
+      "grad_norm": 0.014743310399353504,
+      "learning_rate": 3.441138421733506e-05,
+      "loss": 0.004,
+      "step": 266
+    },
+    {
+      "epoch": 0.863030303030303,
+      "grad_norm": 0.013264276087284088,
+      "learning_rate": 3.454075032341527e-05,
+      "loss": 0.0041,
+      "step": 267
+    },
+    {
+      "epoch": 0.8662626262626263,
+      "grad_norm": 0.01359622087329626,
+      "learning_rate": 3.467011642949548e-05,
+      "loss": 0.004,
+      "step": 268
+    },
+    {
+      "epoch": 0.8694949494949495,
+      "grad_norm": 0.01856253668665886,
+      "learning_rate": 3.4799482535575675e-05,
+      "loss": 0.0047,
+      "step": 269
+    },
+    {
+      "epoch": 0.8727272727272727,
+      "grad_norm": 0.01446174643933773,
+      "learning_rate": 3.4928848641655885e-05,
+      "loss": 0.0048,
+      "step": 270
+    },
+    {
+      "epoch": 0.8759595959595959,
+      "grad_norm": 0.01740093156695366,
+      "learning_rate": 3.5058214747736096e-05,
+      "loss": 0.0083,
+      "step": 271
+    },
+    {
+      "epoch": 0.8791919191919192,
+      "grad_norm": 0.021578310057520866,
+      "learning_rate": 3.5187580853816306e-05,
+      "loss": 0.0069,
+      "step": 272
+    },
+    {
+      "epoch": 0.8824242424242424,
+      "grad_norm": 0.013721848838031292,
+      "learning_rate": 3.531694695989651e-05,
+      "loss": 0.0047,
+      "step": 273
+    },
+    {
+      "epoch": 0.8856565656565657,
+      "grad_norm": 0.015622769482433796,
+      "learning_rate": 3.544631306597671e-05,
+      "loss": 0.0064,
+      "step": 274
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.01654888316988945,
+      "learning_rate": 3.557567917205692e-05,
+      "loss": 0.0071,
+      "step": 275
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "eval_loss": 0.004978961776942015,
+      "eval_runtime": 18.5395,
+      "eval_samples_per_second": 5.394,
+      "eval_steps_per_second": 1.348,
+      "step": 275
+    },
+    {
+      "epoch": 0.8921212121212121,
+      "grad_norm": 0.01643654704093933,
+      "learning_rate": 3.570504527813713e-05,
+      "loss": 0.0033,
+      "step": 276
+    },
+    {
+      "epoch": 0.8953535353535353,
+      "grad_norm": 0.015030475333333015,
+      "learning_rate": 3.583441138421734e-05,
+      "loss": 0.003,
+      "step": 277
+    },
+    {
+      "epoch": 0.8985858585858586,
+      "grad_norm": 0.010186896659433842,
+      "learning_rate": 3.596377749029755e-05,
+      "loss": 0.0035,
+      "step": 278
+    },
+    {
+      "epoch": 0.9018181818181819,
+      "grad_norm": 0.012999375350773335,
+      "learning_rate": 3.609314359637775e-05,
+      "loss": 0.0033,
+      "step": 279
+    },
+    {
+      "epoch": 0.9050505050505051,
+      "grad_norm": 0.015473639592528343,
+      "learning_rate": 3.6222509702457954e-05,
+      "loss": 0.0044,
+      "step": 280
+    },
+    {
+      "epoch": 0.9082828282828282,
+      "grad_norm": 0.011669456027448177,
+      "learning_rate": 3.6351875808538164e-05,
+      "loss": 0.0018,
+      "step": 281
+    },
+    {
+      "epoch": 0.9115151515151515,
+      "grad_norm": 0.011770956218242645,
+      "learning_rate": 3.648124191461837e-05,
+      "loss": 0.0036,
+      "step": 282
+    },
+    {
+      "epoch": 0.9147474747474748,
+      "grad_norm": 0.013007396832108498,
+      "learning_rate": 3.661060802069858e-05,
+      "loss": 0.005,
+      "step": 283
+    },
+    {
+      "epoch": 0.917979797979798,
+      "grad_norm": 0.012168454937636852,
+      "learning_rate": 3.673997412677879e-05,
+      "loss": 0.0043,
+      "step": 284
+    },
+    {
+      "epoch": 0.9212121212121213,
+      "grad_norm": 0.010970023460686207,
+      "learning_rate": 3.6869340232859e-05,
+      "loss": 0.0032,
+      "step": 285
+    },
+    {
+      "epoch": 0.9244444444444444,
+      "grad_norm": 0.012372443452477455,
+      "learning_rate": 3.6998706338939195e-05,
+      "loss": 0.0041,
+      "step": 286
+    },
+    {
+      "epoch": 0.9276767676767677,
+      "grad_norm": 0.013462092727422714,
+      "learning_rate": 3.7128072445019406e-05,
+      "loss": 0.0046,
+      "step": 287
+    },
+    {
+      "epoch": 0.9309090909090909,
+      "grad_norm": 0.01485317014157772,
+      "learning_rate": 3.7257438551099616e-05,
+      "loss": 0.0041,
+      "step": 288
+    },
+    {
+      "epoch": 0.9341414141414142,
+      "grad_norm": 0.014620418660342693,
+      "learning_rate": 3.738680465717982e-05,
+      "loss": 0.007,
+      "step": 289
+    },
+    {
+      "epoch": 0.9373737373737374,
+      "grad_norm": 0.008358921855688095,
+      "learning_rate": 3.751617076326003e-05,
+      "loss": 0.0018,
+      "step": 290
+    },
+    {
+      "epoch": 0.9406060606060606,
+      "grad_norm": 0.015606251545250416,
+      "learning_rate": 3.764553686934023e-05,
+      "loss": 0.0032,
+      "step": 291
+    },
+    {
+      "epoch": 0.9438383838383838,
+      "grad_norm": 0.017797425389289856,
+      "learning_rate": 3.777490297542044e-05,
+      "loss": 0.007,
+      "step": 292
+    },
+    {
+      "epoch": 0.9470707070707071,
+      "grad_norm": 0.011691603809595108,
+      "learning_rate": 3.790426908150065e-05,
+      "loss": 0.0042,
+      "step": 293
+    },
+    {
+      "epoch": 0.9503030303030303,
+      "grad_norm": 0.009671170264482498,
+      "learning_rate": 3.803363518758086e-05,
+      "loss": 0.0031,
+      "step": 294
+    },
+    {
+      "epoch": 0.9535353535353536,
+      "grad_norm": 0.010161872953176498,
+      "learning_rate": 3.816300129366106e-05,
+      "loss": 0.003,
+      "step": 295
+    },
+    {
+      "epoch": 0.9567676767676768,
+      "grad_norm": 0.014181291684508324,
+      "learning_rate": 3.829236739974127e-05,
+      "loss": 0.0039,
+      "step": 296
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.014730575494468212,
+      "learning_rate": 3.8421733505821475e-05,
+      "loss": 0.0048,
+      "step": 297
+    },
+    {
+      "epoch": 0.9632323232323232,
+      "grad_norm": 0.012808237224817276,
+      "learning_rate": 3.855109961190168e-05,
+      "loss": 0.005,
+      "step": 298
+    },
+    {
+      "epoch": 0.9664646464646465,
+      "grad_norm": 0.010238195769488811,
+      "learning_rate": 3.868046571798189e-05,
+      "loss": 0.0035,
+      "step": 299
+    },
+    {
+      "epoch": 0.9696969696969697,
+      "grad_norm": 0.01267747487872839,
+      "learning_rate": 3.88098318240621e-05,
+      "loss": 0.0056,
+      "step": 300
+    },
+    {
+      "epoch": 0.9696969696969697,
+      "eval_loss": 0.004724177997559309,
+      "eval_runtime": 18.5689,
+      "eval_samples_per_second": 5.385,
+      "eval_steps_per_second": 1.346,
+      "step": 300
+    },
+    {
+      "epoch": 0.972929292929293,
+      "grad_norm": 0.009855328127741814,
+      "learning_rate": 3.893919793014231e-05,
+      "loss": 0.0022,
+      "step": 301
+    },
+    {
+      "epoch": 0.9761616161616161,
+      "grad_norm": 0.011048916727304459,
+      "learning_rate": 3.906856403622251e-05,
+      "loss": 0.0035,
+      "step": 302
+    },
+    {
+      "epoch": 0.9793939393939394,
+      "grad_norm": 0.02079099602997303,
+      "learning_rate": 3.9197930142302716e-05,
+      "loss": 0.012,
+      "step": 303
+    },
+    {
+      "epoch": 0.9826262626262626,
+      "grad_norm": 0.01305049005895853,
+      "learning_rate": 3.9327296248382926e-05,
+      "loss": 0.0047,
+      "step": 304
+    },
+    {
+      "epoch": 0.9858585858585859,
+      "grad_norm": 0.010051904246211052,
+      "learning_rate": 3.945666235446313e-05,
+      "loss": 0.0025,
+      "step": 305
+    },
+    {
+      "epoch": 0.9890909090909091,
+      "grad_norm": 0.01657235063612461,
+      "learning_rate": 3.958602846054334e-05,
+      "loss": 0.008,
+      "step": 306
+    },
+    {
+      "epoch": 0.9923232323232323,
+      "grad_norm": 0.008980581536889076,
+      "learning_rate": 3.971539456662355e-05,
+      "loss": 0.0022,
+      "step": 307
+    },
+    {
+      "epoch": 0.9955555555555555,
+      "grad_norm": 0.01711704581975937,
+      "learning_rate": 3.9844760672703754e-05,
+      "loss": 0.0082,
+      "step": 308
+    },
+    {
+      "epoch": 0.9987878787878788,
+      "grad_norm": 0.01096993125975132,
+      "learning_rate": 3.997412677878396e-05,
+      "loss": 0.0042,
+      "step": 309
+    },
+    {
+      "epoch": 1.002020202020202,
+      "grad_norm": 0.023890549317002296,
+      "learning_rate": 4.010349288486417e-05,
+      "loss": 0.0076,
+      "step": 310
+    },
+    {
+      "epoch": 1.0052525252525253,
+      "grad_norm": 0.009565077722072601,
+      "learning_rate": 4.023285899094437e-05,
+      "loss": 0.003,
+      "step": 311
+    },
+    {
+      "epoch": 1.0084848484848485,
+      "grad_norm": 0.012166207656264305,
+      "learning_rate": 4.036222509702458e-05,
+      "loss": 0.0057,
+      "step": 312
+    },
+    {
+      "epoch": 1.0117171717171718,
+      "grad_norm": 0.007330482825636864,
+      "learning_rate": 4.049159120310479e-05,
+      "loss": 0.0021,
+      "step": 313
+    },
+    {
+      "epoch": 1.014949494949495,
+      "grad_norm": 0.010659433901309967,
+      "learning_rate": 4.0620957309184995e-05,
+      "loss": 0.0031,
+      "step": 314
+    },
+    {
+      "epoch": 1.018181818181818,
+      "grad_norm": 0.009105874225497246,
+      "learning_rate": 4.07503234152652e-05,
+      "loss": 0.0028,
+      "step": 315
+    },
+    {
+      "epoch": 1.0214141414141413,
+      "grad_norm": 0.011084525845944881,
+      "learning_rate": 4.087968952134541e-05,
+      "loss": 0.0029,
+      "step": 316
+    },
+    {
+      "epoch": 1.0246464646464646,
+      "grad_norm": 0.008576706051826477,
+      "learning_rate": 4.100905562742562e-05,
+      "loss": 0.0024,
+      "step": 317
+    },
+    {
+      "epoch": 1.0278787878787878,
+      "grad_norm": 0.01535963173955679,
+      "learning_rate": 4.113842173350582e-05,
+      "loss": 0.0049,
+      "step": 318
+    },
+    {
+      "epoch": 1.031111111111111,
+      "grad_norm": 0.013791275210678577,
+      "learning_rate": 4.126778783958603e-05,
+      "loss": 0.0028,
+      "step": 319
+    },
+    {
+      "epoch": 1.0343434343434343,
+      "grad_norm": 0.01699056848883629,
+      "learning_rate": 4.1397153945666236e-05,
+      "loss": 0.0075,
+      "step": 320
+    },
+    {
+      "epoch": 1.0375757575757576,
+      "grad_norm": 0.008655001409351826,
+      "learning_rate": 4.152652005174644e-05,
+      "loss": 0.0037,
+      "step": 321
+    },
+    {
+      "epoch": 1.0408080808080808,
+      "grad_norm": 0.009002985432744026,
+      "learning_rate": 4.165588615782665e-05,
+      "loss": 0.0031,
+      "step": 322
+    },
+    {
+      "epoch": 1.044040404040404,
+      "grad_norm": 0.008515028282999992,
+      "learning_rate": 4.178525226390686e-05,
+      "loss": 0.002,
+      "step": 323
+    },
+    {
+      "epoch": 1.0472727272727274,
+      "grad_norm": 0.011139890179038048,
+      "learning_rate": 4.1914618369987064e-05,
+      "loss": 0.0031,
+      "step": 324
+    },
+    {
+      "epoch": 1.0505050505050506,
+      "grad_norm": 0.008947103284299374,
+      "learning_rate": 4.2043984476067274e-05,
+      "loss": 0.0017,
+      "step": 325
+    },
+    {
+      "epoch": 1.0505050505050506,
+      "eval_loss": 0.004780053161084652,
+      "eval_runtime": 18.5718,
+      "eval_samples_per_second": 5.385,
+      "eval_steps_per_second": 1.346,
+      "step": 325
+    },
+    {
+      "epoch": 1.0537373737373736,
+      "grad_norm": 0.01565450057387352,
+      "learning_rate": 4.217335058214748e-05,
+      "loss": 0.0033,
+      "step": 326
+    },
+    {
+      "epoch": 1.056969696969697,
+      "grad_norm": 0.016629492864012718,
+      "learning_rate": 4.230271668822768e-05,
+      "loss": 0.0071,
+      "step": 327
+    },
+    {
+      "epoch": 1.0602020202020201,
+      "grad_norm": 0.012543603777885437,
+      "learning_rate": 4.243208279430789e-05,
+      "loss": 0.0042,
+      "step": 328
+    },
+    {
+      "epoch": 1.0634343434343434,
+      "grad_norm": 0.010482177138328552,
+      "learning_rate": 4.25614489003881e-05,
+      "loss": 0.0028,
+      "step": 329
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.010985253378748894,
+      "learning_rate": 4.269081500646831e-05,
+      "loss": 0.0033,
+      "step": 330
+    },
+    {
+      "epoch": 1.06989898989899,
+      "grad_norm": 0.010153640992939472,
+      "learning_rate": 4.2820181112548515e-05,
+      "loss": 0.0028,
+      "step": 331
+    },
+    {
+      "epoch": 1.0731313131313132,
+      "grad_norm": 0.0068151215091347694,
+      "learning_rate": 4.294954721862872e-05,
+      "loss": 0.0016,
+      "step": 332
+    },
+    {
+      "epoch": 1.0763636363636364,
+      "grad_norm": 0.01169463712722063,
+      "learning_rate": 4.307891332470893e-05,
+      "loss": 0.0029,
+      "step": 333
+    },
+    {
+      "epoch": 1.0795959595959597,
+      "grad_norm": 0.016815535724163055,
+      "learning_rate": 4.320827943078913e-05,
+      "loss": 0.0069,
+      "step": 334
+    },
+    {
+      "epoch": 1.082828282828283,
+      "grad_norm": 0.01823563128709793,
+      "learning_rate": 4.333764553686934e-05,
+      "loss": 0.0096,
+      "step": 335
+    },
+    {
+      "epoch": 1.086060606060606,
+      "grad_norm": 0.009779485873878002,
+      "learning_rate": 4.346701164294955e-05,
+      "loss": 0.0023,
+      "step": 336
+    },
+    {
+      "epoch": 1.0892929292929292,
+      "grad_norm": 0.014083399437367916,
+      "learning_rate": 4.359637774902976e-05,
+      "loss": 0.0032,
+      "step": 337
+    },
+    {
+      "epoch": 1.0925252525252525,
+      "grad_norm": 0.01473916880786419,
+      "learning_rate": 4.372574385510996e-05,
+      "loss": 0.0047,
+      "step": 338
+    },
+    {
+      "epoch": 1.0957575757575757,
+      "grad_norm": 0.010457643307745457,
+      "learning_rate": 4.385510996119017e-05,
+      "loss": 0.0031,
+      "step": 339
+    },
+    {
+      "epoch": 1.098989898989899,
+      "grad_norm": 0.010025047697126865,
+      "learning_rate": 4.3984476067270374e-05,
+      "loss": 0.0035,
+      "step": 340
+    },
+    {
+      "epoch": 1.1022222222222222,
+      "grad_norm": 0.013103731907904148,
+      "learning_rate": 4.4113842173350584e-05,
+      "loss": 0.0046,
+      "step": 341
+    },
+    {
+      "epoch": 1.1054545454545455,
+      "grad_norm": 0.011729112826287746,
+      "learning_rate": 4.4243208279430794e-05,
+      "loss": 0.0031,
+      "step": 342
+    },
+    {
+      "epoch": 1.1086868686868687,
+      "grad_norm": 0.010078197345137596,
+      "learning_rate": 4.4372574385511e-05,
+      "loss": 0.0037,
+      "step": 343
+    },
+    {
+      "epoch": 1.111919191919192,
+      "grad_norm": 0.01012883149087429,
+      "learning_rate": 4.45019404915912e-05,
+      "loss": 0.0027,
+      "step": 344
+    },
+    {
+      "epoch": 1.1151515151515152,
+      "grad_norm": 0.01668359898030758,
+      "learning_rate": 4.463130659767141e-05,
+      "loss": 0.0057,
+      "step": 345
+    },
+    {
+      "epoch": 1.1183838383838385,
+      "grad_norm": 0.011654009111225605,
+      "learning_rate": 4.476067270375162e-05,
+      "loss": 0.0029,
+      "step": 346
+    },
+    {
+      "epoch": 1.1216161616161617,
+      "grad_norm": 0.018832651898264885,
+      "learning_rate": 4.4890038809831825e-05,
+      "loss": 0.0056,
+      "step": 347
+    },
+    {
+      "epoch": 1.1248484848484848,
+      "grad_norm": 0.006298431195318699,
+      "learning_rate": 4.5019404915912036e-05,
+      "loss": 0.0012,
+      "step": 348
+    },
+    {
+      "epoch": 1.128080808080808,
+      "grad_norm": 0.018230654299259186,
+      "learning_rate": 4.514877102199224e-05,
+      "loss": 0.0065,
+      "step": 349
+    },
+    {
+      "epoch": 1.1313131313131313,
+      "grad_norm": 0.007927543483674526,
+      "learning_rate": 4.527813712807244e-05,
+      "loss": 0.0024,
+      "step": 350
+    },
+    {
+      "epoch": 1.1313131313131313,
+      "eval_loss": 0.004599680192768574,
+      "eval_runtime": 18.5134,
+      "eval_samples_per_second": 5.401,
+      "eval_steps_per_second": 1.35,
+      "step": 350
+    },
+    {
+      "epoch": 1.1345454545454545,
+      "grad_norm": 0.012782711535692215,
+      "learning_rate": 4.540750323415265e-05,
+      "loss": 0.0043,
+      "step": 351
+    },
+    {
+      "epoch": 1.1377777777777778,
+      "grad_norm": 0.01866583153605461,
+      "learning_rate": 4.553686934023286e-05,
+      "loss": 0.0083,
+      "step": 352
+    },
+    {
+      "epoch": 1.141010101010101,
+      "grad_norm": 0.010514589957892895,
+      "learning_rate": 4.566623544631307e-05,
+      "loss": 0.0033,
+      "step": 353
+    },
+    {
+      "epoch": 1.1442424242424243,
+      "grad_norm": 0.01175905391573906,
+      "learning_rate": 4.579560155239328e-05,
+      "loss": 0.0036,
+      "step": 354
+    },
+    {
+      "epoch": 1.1474747474747475,
+      "grad_norm": 0.01182876992970705,
+      "learning_rate": 4.592496765847348e-05,
+      "loss": 0.0051,
+      "step": 355
+    },
+    {
+      "epoch": 1.1507070707070708,
+      "grad_norm": 0.007950863800942898,
+      "learning_rate": 4.6054333764553684e-05,
+      "loss": 0.002,
+      "step": 356
+    },
+    {
+      "epoch": 1.1539393939393938,
+      "grad_norm": 0.009339825250208378,
+      "learning_rate": 4.6183699870633894e-05,
+      "loss": 0.0022,
+      "step": 357
+    },
+    {
+      "epoch": 1.157171717171717,
+      "grad_norm": 0.012703759595751762,
+      "learning_rate": 4.6313065976714105e-05,
+      "loss": 0.0039,
+      "step": 358
+    },
+    {
+      "epoch": 1.1604040404040403,
+      "grad_norm": 0.019440334290266037,
+      "learning_rate": 4.6442432082794315e-05,
+      "loss": 0.0086,
+      "step": 359
+    },
+    {
+      "epoch": 1.1636363636363636,
+      "grad_norm": 0.008004146628081799,
+      "learning_rate": 4.657179818887452e-05,
+      "loss": 0.0015,
+      "step": 360
+    },
+    {
+      "epoch": 1.1668686868686868,
+      "grad_norm": 0.017837535589933395,
+      "learning_rate": 4.670116429495472e-05,
+      "loss": 0.0055,
+      "step": 361
+    },
+    {
+      "epoch": 1.17010101010101,
+      "grad_norm": 0.011339561082422733,
+      "learning_rate": 4.683053040103493e-05,
+      "loss": 0.0035,
+      "step": 362
+    },
+    {
+      "epoch": 1.1733333333333333,
+      "grad_norm": 0.012849444523453712,
+      "learning_rate": 4.6959896507115136e-05,
+      "loss": 0.0046,
+      "step": 363
+    },
+    {
+      "epoch": 1.1765656565656566,
+      "grad_norm": 0.015334355644881725,
+      "learning_rate": 4.7089262613195346e-05,
+      "loss": 0.0061,
+      "step": 364
+    },
+    {
+      "epoch": 1.1797979797979798,
+      "grad_norm": 0.010206463746726513,
+      "learning_rate": 4.7218628719275556e-05,
+      "loss": 0.0031,
+      "step": 365
+    },
+    {
+      "epoch": 1.183030303030303,
+      "grad_norm": 0.004871445242315531,
+      "learning_rate": 4.734799482535576e-05,
+      "loss": 0.0016,
+      "step": 366
+    },
+    {
+      "epoch": 1.1862626262626264,
+      "grad_norm": 0.012882773764431477,
+      "learning_rate": 4.747736093143596e-05,
+      "loss": 0.003,
+      "step": 367
+    },
+    {
+      "epoch": 1.1894949494949496,
+      "grad_norm": 0.009307224303483963,
+      "learning_rate": 4.760672703751617e-05,
+      "loss": 0.0018,
+      "step": 368
+    },
+    {
+      "epoch": 1.1927272727272726,
+      "grad_norm": 0.011766292154788971,
+      "learning_rate": 4.773609314359638e-05,
+      "loss": 0.0049,
+      "step": 369
+    },
+    {
+      "epoch": 1.195959595959596,
+      "grad_norm": 0.006270453333854675,
+      "learning_rate": 4.786545924967659e-05,
+      "loss": 0.0016,
+      "step": 370
+    },
+    {
+      "epoch": 1.1991919191919191,
+      "grad_norm": 0.011882314458489418,
+      "learning_rate": 4.79948253557568e-05,
+      "loss": 0.0045,
+      "step": 371
+    },
+    {
+      "epoch": 1.2024242424242424,
+      "grad_norm": 0.012190425768494606,
+      "learning_rate": 4.8124191461837e-05,
+      "loss": 0.0029,
+      "step": 372
+    },
+    {
+      "epoch": 1.2056565656565656,
+      "grad_norm": 0.017565051093697548,
+      "learning_rate": 4.8253557567917204e-05,
+      "loss": 0.0066,
+      "step": 373
+    },
+    {
+      "epoch": 1.208888888888889,
+      "grad_norm": 0.01365733053535223,
+      "learning_rate": 4.8382923673997415e-05,
+      "loss": 0.003,
+      "step": 374
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "grad_norm": 0.00959896668791771,
+      "learning_rate": 4.8512289780077625e-05,
+      "loss": 0.0023,
+      "step": 375
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "eval_loss": 0.004647642374038696,
+      "eval_runtime": 18.5462,
+      "eval_samples_per_second": 5.392,
+      "eval_steps_per_second": 1.348,
+      "step": 375
+    },
+    {
+      "epoch": 1.2153535353535354,
+      "grad_norm": 0.008952205069363117,
+      "learning_rate": 4.864165588615783e-05,
+      "loss": 0.003,
+      "step": 376
+    },
+    {
+      "epoch": 1.2185858585858587,
+      "grad_norm": 0.01386015210300684,
+      "learning_rate": 4.877102199223804e-05,
+      "loss": 0.0046,
+      "step": 377
+    },
+    {
+      "epoch": 1.221818181818182,
+      "grad_norm": 0.01027520652860403,
+      "learning_rate": 4.890038809831824e-05,
+      "loss": 0.0031,
+      "step": 378
+    },
+    {
+      "epoch": 1.225050505050505,
+      "grad_norm": 0.014104802161455154,
+      "learning_rate": 4.9029754204398446e-05,
+      "loss": 0.0055,
+      "step": 379
+    },
+    {
+      "epoch": 1.2282828282828282,
+      "grad_norm": 0.012762988917529583,
+      "learning_rate": 4.9159120310478656e-05,
+      "loss": 0.0026,
+      "step": 380
+    },
+    {
+      "epoch": 1.2315151515151515,
+      "grad_norm": 0.010741367004811764,
+      "learning_rate": 4.9288486416558866e-05,
+      "loss": 0.0028,
+      "step": 381
+    },
+    {
+      "epoch": 1.2347474747474747,
+      "grad_norm": 0.008360295556485653,
+      "learning_rate": 4.941785252263907e-05,
+      "loss": 0.0022,
+      "step": 382
+    },
+    {
+      "epoch": 1.237979797979798,
+      "grad_norm": 0.013385162688791752,
+      "learning_rate": 4.954721862871928e-05,
+      "loss": 0.0032,
+      "step": 383
+    },
+    {
+      "epoch": 1.2412121212121212,
+      "grad_norm": 0.009576680138707161,
+      "learning_rate": 4.9676584734799483e-05,
+      "loss": 0.0024,
+      "step": 384
+    },
+    {
+      "epoch": 1.2444444444444445,
+      "grad_norm": 0.011253474280238152,
+      "learning_rate": 4.980595084087969e-05,
+      "loss": 0.0035,
+      "step": 385
+    },
+    {
+      "epoch": 1.2476767676767677,
+      "grad_norm": 0.013251024298369884,
+      "learning_rate": 4.99353169469599e-05,
+      "loss": 0.0037,
+      "step": 386
+    },
+    {
+      "epoch": 1.250909090909091,
+      "grad_norm": 0.0128615228459239,
+      "learning_rate": 5.006468305304011e-05,
+      "loss": 0.0042,
+      "step": 387
+    },
+    {
+      "epoch": 1.2541414141414142,
+      "grad_norm": 0.011205179616808891,
+      "learning_rate": 5.019404915912032e-05,
+      "loss": 0.0022,
+      "step": 388
+    },
+    {
+      "epoch": 1.2573737373737375,
+      "grad_norm": 0.009263547137379646,
+      "learning_rate": 5.032341526520052e-05,
+      "loss": 0.0033,
+      "step": 389
+    },
+    {
+      "epoch": 1.2606060606060607,
+      "grad_norm": 0.013434207998216152,
+      "learning_rate": 5.0452781371280725e-05,
+      "loss": 0.0041,
+      "step": 390
+    },
+    {
+      "epoch": 1.2638383838383838,
+      "grad_norm": 0.008763938210904598,
+      "learning_rate": 5.058214747736093e-05,
+      "loss": 0.0026,
+      "step": 391
+    },
+    {
+      "epoch": 1.267070707070707,
+      "grad_norm": 0.013437739573419094,
+      "learning_rate": 5.071151358344114e-05,
+      "loss": 0.003,
+      "step": 392
+    },
+    {
+      "epoch": 1.2703030303030303,
+      "grad_norm": 0.017890624701976776,
+      "learning_rate": 5.084087968952135e-05,
+      "loss": 0.0095,
+      "step": 393
+    },
+    {
+      "epoch": 1.2735353535353535,
+      "grad_norm": 0.008945245295763016,
+      "learning_rate": 5.097024579560156e-05,
+      "loss": 0.0022,
+      "step": 394
+    },
+    {
+      "epoch": 1.2767676767676768,
+      "grad_norm": 0.010177022777497768,
+      "learning_rate": 5.109961190168176e-05,
+      "loss": 0.0026,
+      "step": 395
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.010548125021159649,
+      "learning_rate": 5.122897800776197e-05,
+      "loss": 0.0022,
+      "step": 396
+    },
+    {
+      "epoch": 1.2832323232323233,
+      "grad_norm": 0.01482499111443758,
+      "learning_rate": 5.135834411384217e-05,
+      "loss": 0.0041,
+      "step": 397
+    },
+    {
+      "epoch": 1.2864646464646465,
+      "grad_norm": 0.009971357882022858,
+      "learning_rate": 5.148771021992238e-05,
+      "loss": 0.0032,
+      "step": 398
+    },
+    {
+      "epoch": 1.2896969696969696,
+      "grad_norm": 0.009518085978925228,
+      "learning_rate": 5.161707632600259e-05,
+      "loss": 0.0026,
+      "step": 399
+    },
+    {
+      "epoch": 1.2929292929292928,
+      "grad_norm": 0.010194380767643452,
+      "learning_rate": 5.17464424320828e-05,
+      "loss": 0.003,
+      "step": 400
+    },
+    {
+      "epoch": 1.2929292929292928,
+      "eval_loss": 0.004324791021645069,
+      "eval_runtime": 18.5425,
+      "eval_samples_per_second": 5.393,
+      "eval_steps_per_second": 1.348,
+      "step": 400
+    },
+    {
+      "epoch": 1.296161616161616,
+      "grad_norm": 0.013518190011382103,
+      "learning_rate": 5.1875808538163004e-05,
+      "loss": 0.0048,
+      "step": 401
+    },
+    {
+      "epoch": 1.2993939393939393,
+      "grad_norm": 0.009436222724616528,
+      "learning_rate": 5.2005174644243214e-05,
+      "loss": 0.0032,
+      "step": 402
+    },
+    {
+      "epoch": 1.3026262626262626,
+      "grad_norm": 0.014251810498535633,
+      "learning_rate": 5.213454075032341e-05,
+      "loss": 0.0038,
+      "step": 403
+    },
+    {
+      "epoch": 1.3058585858585858,
+      "grad_norm": 0.013866190798580647,
+      "learning_rate": 5.226390685640362e-05,
+      "loss": 0.0053,
+      "step": 404
+    },
+    {
+      "epoch": 1.309090909090909,
+      "grad_norm": 0.014752188697457314,
+      "learning_rate": 5.239327296248383e-05,
+      "loss": 0.0024,
+      "step": 405
+    },
+    {
+      "epoch": 1.3123232323232323,
+      "grad_norm": 0.010886342264711857,
+      "learning_rate": 5.252263906856404e-05,
+      "loss": 0.0041,
+      "step": 406
+    },
+    {
+      "epoch": 1.3155555555555556,
+      "grad_norm": 0.010374347679316998,
+      "learning_rate": 5.2652005174644245e-05,
+      "loss": 0.0037,
+      "step": 407
+    },
+    {
+      "epoch": 1.3187878787878788,
+      "grad_norm": 0.010170732624828815,
+      "learning_rate": 5.2781371280724455e-05,
+      "loss": 0.0051,
+      "step": 408
+    },
+    {
+      "epoch": 1.322020202020202,
+      "grad_norm": 0.010164180770516396,
+      "learning_rate": 5.2910737386804666e-05,
+      "loss": 0.004,
+      "step": 409
+    },
+    {
+      "epoch": 1.3252525252525253,
+      "grad_norm": 0.008564438670873642,
+      "learning_rate": 5.304010349288486e-05,
+      "loss": 0.0029,
+      "step": 410
+    },
+    {
+      "epoch": 1.3284848484848486,
+      "grad_norm": 0.008800626732409,
+      "learning_rate": 5.316946959896507e-05,
+      "loss": 0.003,
+      "step": 411
+    },
+    {
+      "epoch": 1.3317171717171719,
+      "grad_norm": 0.007803233340382576,
+      "learning_rate": 5.329883570504528e-05,
+      "loss": 0.0027,
+      "step": 412
+    },
+    {
+      "epoch": 1.3349494949494949,
+      "grad_norm": 0.00937716942280531,
+      "learning_rate": 5.3428201811125486e-05,
+      "loss": 0.0027,
+      "step": 413
+    },
+    {
+      "epoch": 1.3381818181818181,
+      "grad_norm": 0.008757539093494415,
+      "learning_rate": 5.35575679172057e-05,
+      "loss": 0.0031,
+      "step": 414
+    },
+    {
+      "epoch": 1.3414141414141414,
+      "grad_norm": 0.00794417317956686,
+      "learning_rate": 5.368693402328591e-05,
+      "loss": 0.0028,
+      "step": 415
+    },
+    {
+      "epoch": 1.3446464646464646,
+      "grad_norm": 0.011016054078936577,
+      "learning_rate": 5.3816300129366104e-05,
+      "loss": 0.0042,
+      "step": 416
+    },
+    {
+      "epoch": 1.347878787878788,
+      "grad_norm": 0.011397106572985649,
+      "learning_rate": 5.3945666235446314e-05,
+      "loss": 0.0043,
+      "step": 417
+    },
+    {
+      "epoch": 1.3511111111111112,
+      "grad_norm": 0.020235812291502953,
+      "learning_rate": 5.4075032341526524e-05,
+      "loss": 0.0082,
+      "step": 418
+    },
+    {
+      "epoch": 1.3543434343434344,
+      "grad_norm": 0.010009384714066982,
+      "learning_rate": 5.420439844760673e-05,
+      "loss": 0.0037,
+      "step": 419
+    },
+    {
+      "epoch": 1.3575757575757577,
+      "grad_norm": 0.010380405001342297,
+      "learning_rate": 5.433376455368694e-05,
+      "loss": 0.0032,
+      "step": 420
+    },
+    {
+      "epoch": 1.3608080808080807,
+      "grad_norm": 0.011236142367124557,
+      "learning_rate": 5.446313065976715e-05,
+      "loss": 0.0043,
+      "step": 421
+    },
+    {
+      "epoch": 1.364040404040404,
+      "grad_norm": 0.008906678296625614,
+      "learning_rate": 5.4592496765847345e-05,
+      "loss": 0.0025,
+      "step": 422
+    },
+    {
+      "epoch": 1.3672727272727272,
+      "grad_norm": 0.013497211039066315,
+      "learning_rate": 5.4721862871927555e-05,
+      "loss": 0.0039,
+      "step": 423
+    },
+    {
+      "epoch": 1.3705050505050504,
+      "grad_norm": 0.007696770131587982,
+      "learning_rate": 5.4851228978007766e-05,
+      "loss": 0.0028,
+      "step": 424
+    },
+    {
+      "epoch": 1.3737373737373737,
+      "grad_norm": 0.07293156534433365,
+      "learning_rate": 5.498059508408797e-05,
+      "loss": 0.005,
+      "step": 425
+    },
+    {
+      "epoch": 1.3737373737373737,
+      "eval_loss": 0.004379382357001305,
+      "eval_runtime": 18.5797,
+      "eval_samples_per_second": 5.382,
+      "eval_steps_per_second": 1.346,
+      "step": 425
+    },
+    {
+      "epoch": 1.376969696969697,
+      "grad_norm": 0.01241080742329359,
+      "learning_rate": 5.510996119016818e-05,
+      "loss": 0.0035,
+      "step": 426
+    },
+    {
+      "epoch": 1.3802020202020202,
+      "grad_norm": 0.010909633710980415,
+      "learning_rate": 5.523932729624839e-05,
+      "loss": 0.0036,
+      "step": 427
+    },
+    {
+      "epoch": 1.3834343434343435,
+      "grad_norm": 0.014454485848546028,
+      "learning_rate": 5.53686934023286e-05,
+      "loss": 0.0061,
+      "step": 428
+    },
+    {
+      "epoch": 1.3866666666666667,
+      "grad_norm": 0.006944994442164898,
+      "learning_rate": 5.5498059508408797e-05,
+      "loss": 0.0024,
+      "step": 429
+    },
+    {
+      "epoch": 1.38989898989899,
+      "grad_norm": 0.009738634340465069,
+      "learning_rate": 5.562742561448901e-05,
+      "loss": 0.0037,
+      "step": 430
+    },
+    {
+      "epoch": 1.3931313131313132,
+      "grad_norm": 0.01077584270387888,
+      "learning_rate": 5.575679172056921e-05,
+      "loss": 0.0041,
+      "step": 431
+    },
+    {
+      "epoch": 1.3963636363636365,
+      "grad_norm": 0.009695135056972504,
+      "learning_rate": 5.588615782664942e-05,
+      "loss": 0.0031,
+      "step": 432
+    },
+    {
+      "epoch": 1.3995959595959597,
+      "grad_norm": 0.014413881115615368,
+      "learning_rate": 5.601552393272963e-05,
+      "loss": 0.0037,
+      "step": 433
+    },
+    {
+      "epoch": 1.4028282828282828,
+      "grad_norm": 0.012725897133350372,
+      "learning_rate": 5.614489003880984e-05,
+      "loss": 0.0043,
+      "step": 434
+    },
+    {
+      "epoch": 1.406060606060606,
+      "grad_norm": 0.010477320291101933,
+      "learning_rate": 5.627425614489004e-05,
+      "loss": 0.003,
+      "step": 435
+    },
+    {
+      "epoch": 1.4092929292929293,
+      "grad_norm": 0.010203997604548931,
+      "learning_rate": 5.640362225097025e-05,
+      "loss": 0.0031,
+      "step": 436
+    },
+    {
+      "epoch": 1.4125252525252525,
+      "grad_norm": 0.011213167570531368,
+      "learning_rate": 5.653298835705045e-05,
+      "loss": 0.0045,
+      "step": 437
+    },
+    {
+      "epoch": 1.4157575757575758,
+      "grad_norm": 0.011191388592123985,
+      "learning_rate": 5.666235446313066e-05,
+      "loss": 0.0048,
+      "step": 438
+    },
+    {
+      "epoch": 1.418989898989899,
+      "grad_norm": 0.008038587868213654,
+      "learning_rate": 5.679172056921087e-05,
+      "loss": 0.0027,
+      "step": 439
+    },
+    {
+      "epoch": 1.4222222222222223,
+      "grad_norm": 0.010327471420168877,
+      "learning_rate": 5.692108667529108e-05,
+      "loss": 0.004,
+      "step": 440
+    },
+    {
+      "epoch": 1.4254545454545455,
+      "grad_norm": 0.01546016801148653,
+      "learning_rate": 5.7050452781371286e-05,
+      "loss": 0.0052,
+      "step": 441
+    },
+    {
+      "epoch": 1.4286868686868686,
+      "grad_norm": 0.00971939880400896,
+      "learning_rate": 5.717981888745149e-05,
+      "loss": 0.0029,
+      "step": 442
+    },
+    {
+      "epoch": 1.4319191919191918,
+      "grad_norm": 0.019665749743580818,
+      "learning_rate": 5.730918499353169e-05,
+      "loss": 0.01,
+      "step": 443
+    },
+    {
+      "epoch": 1.435151515151515,
+      "grad_norm": 0.00928762461990118,
+      "learning_rate": 5.74385510996119e-05,
+      "loss": 0.003,
+      "step": 444
+    },
+    {
+      "epoch": 1.4383838383838383,
+      "grad_norm": 0.009248577058315277,
+      "learning_rate": 5.7567917205692113e-05,
+      "loss": 0.0024,
+      "step": 445
+    },
+    {
+      "epoch": 1.4416161616161616,
+      "grad_norm": 0.006708938162773848,
+      "learning_rate": 5.7697283311772324e-05,
+      "loss": 0.0021,
+      "step": 446
+    },
+    {
+      "epoch": 1.4448484848484848,
+      "grad_norm": 0.014481152407824993,
+      "learning_rate": 5.782664941785253e-05,
+      "loss": 0.004,
+      "step": 447
+    },
+    {
+      "epoch": 1.448080808080808,
+      "grad_norm": 0.015891794115304947,
+      "learning_rate": 5.795601552393273e-05,
+      "loss": 0.0078,
+      "step": 448
+    },
+    {
+      "epoch": 1.4513131313131313,
+      "grad_norm": 0.01767084375023842,
+      "learning_rate": 5.8085381630012934e-05,
+      "loss": 0.0078,
+      "step": 449
+    },
+    {
+      "epoch": 1.4545454545454546,
+      "grad_norm": 0.009849028661847115,
+      "learning_rate": 5.8214747736093145e-05,
+      "loss": 0.0039,
+      "step": 450
+    },
+    {
+      "epoch": 1.4545454545454546,
+      "eval_loss": 0.003997213672846556,
+      "eval_runtime": 18.599,
+      "eval_samples_per_second": 5.377,
+      "eval_steps_per_second": 1.344,
+      "step": 450
+    },
+    {
+      "epoch": 1.4577777777777778,
+      "grad_norm": 0.007672747131437063,
+      "learning_rate": 5.8344113842173355e-05,
+      "loss": 0.0024,
+      "step": 451
+    },
+    {
+      "epoch": 1.461010101010101,
+      "grad_norm": 0.0118905920535326,
+      "learning_rate": 5.8473479948253565e-05,
+      "loss": 0.0048,
+      "step": 452
+    },
+    {
+      "epoch": 1.4642424242424243,
+      "grad_norm": 0.01564439944922924,
+      "learning_rate": 5.860284605433377e-05,
+      "loss": 0.0084,
+      "step": 453
+    },
+    {
+      "epoch": 1.4674747474747476,
+      "grad_norm": 0.005980201996862888,
+      "learning_rate": 5.873221216041398e-05,
+      "loss": 0.0018,
+      "step": 454
+    },
+    {
+      "epoch": 1.4707070707070706,
+      "grad_norm": 0.007433287333697081,
+      "learning_rate": 5.8861578266494176e-05,
+      "loss": 0.0039,
+      "step": 455
+    },
+    {
+      "epoch": 1.4739393939393939,
+      "grad_norm": 0.010296525433659554,
+      "learning_rate": 5.8990944372574386e-05,
+      "loss": 0.0039,
+      "step": 456
+    },
+    {
+      "epoch": 1.4771717171717171,
+      "grad_norm": 0.009413332678377628,
+      "learning_rate": 5.9120310478654596e-05,
+      "loss": 0.0027,
+      "step": 457
+    },
+    {
+      "epoch": 1.4804040404040404,
+      "grad_norm": 0.012635589577257633,
+      "learning_rate": 5.9249676584734806e-05,
+      "loss": 0.0053,
+      "step": 458
+    },
+    {
+      "epoch": 1.4836363636363636,
+      "grad_norm": 0.0071015129797160625,
+      "learning_rate": 5.937904269081501e-05,
+      "loss": 0.0026,
+      "step": 459
+    },
+    {
+      "epoch": 1.486868686868687,
+      "grad_norm": 0.010815092362463474,
+      "learning_rate": 5.950840879689522e-05,
+      "loss": 0.004,
+      "step": 460
+    },
+    {
+      "epoch": 1.4901010101010101,
+      "grad_norm": 0.010755320079624653,
+      "learning_rate": 5.963777490297542e-05,
+      "loss": 0.0043,
+      "step": 461
+    },
+    {
+      "epoch": 1.4933333333333334,
+      "grad_norm": 0.0068894485011696815,
+      "learning_rate": 5.976714100905563e-05,
+      "loss": 0.0024,
+      "step": 462
+    },
+    {
+      "epoch": 1.4965656565656564,
+      "grad_norm": 0.010043000802397728,
+      "learning_rate": 5.989650711513584e-05,
+      "loss": 0.004,
+      "step": 463
+    },
+    {
+      "epoch": 1.4997979797979797,
+      "grad_norm": 0.013079461641609669,
+      "learning_rate": 6.002587322121605e-05,
+      "loss": 0.0061,
+      "step": 464
+    },
+    {
+      "epoch": 1.503030303030303,
+      "grad_norm": 0.012679477222263813,
+      "learning_rate": 6.015523932729625e-05,
+      "loss": 0.0036,
+      "step": 465
+    },
+    {
+      "epoch": 1.5062626262626262,
+      "grad_norm": 0.009493381716310978,
+      "learning_rate": 6.028460543337646e-05,
+      "loss": 0.0038,
+      "step": 466
+    },
+    {
+      "epoch": 1.5094949494949494,
+      "grad_norm": 0.016118116676807404,
+      "learning_rate": 6.041397153945667e-05,
+      "loss": 0.0058,
+      "step": 467
+    },
+    {
+      "epoch": 1.5127272727272727,
+      "grad_norm": 0.01638084277510643,
+      "learning_rate": 6.054333764553687e-05,
+      "loss": 0.0056,
+      "step": 468
+    },
+    {
+      "epoch": 1.515959595959596,
+      "grad_norm": 0.007027367129921913,
+      "learning_rate": 6.067270375161708e-05,
+      "loss": 0.002,
+      "step": 469
+    },
+    {
+      "epoch": 1.5191919191919192,
+      "grad_norm": 0.00825897790491581,
+      "learning_rate": 6.080206985769729e-05,
+      "loss": 0.0018,
+      "step": 470
+    },
+    {
+      "epoch": 1.5224242424242425,
+      "grad_norm": 0.00937443133443594,
+      "learning_rate": 6.093143596377749e-05,
+      "loss": 0.0025,
+      "step": 471
+    },
+    {
+      "epoch": 1.5256565656565657,
+      "grad_norm": 0.009296581149101257,
+      "learning_rate": 6.10608020698577e-05,
+      "loss": 0.0033,
+      "step": 472
+    },
+    {
+      "epoch": 1.528888888888889,
+      "grad_norm": 0.008974263444542885,
+      "learning_rate": 6.119016817593791e-05,
+      "loss": 0.0033,
+      "step": 473
+    },
+    {
+      "epoch": 1.5321212121212122,
+      "grad_norm": 0.012877869419753551,
+      "learning_rate": 6.131953428201811e-05,
+      "loss": 0.0049,
+      "step": 474
+    },
+    {
+      "epoch": 1.5353535353535355,
+      "grad_norm": 0.011777269653975964,
+      "learning_rate": 6.144890038809832e-05,
+      "loss": 0.0045,
+      "step": 475
+    },
+    {
+      "epoch": 1.5353535353535355,
+      "eval_loss": 0.004129638429731131,
+      "eval_runtime": 18.5241,
+      "eval_samples_per_second": 5.398,
+      "eval_steps_per_second": 1.35,
+      "step": 475
+    },
+    {
+      "epoch": 1.5385858585858587,
+      "grad_norm": 0.010810826905071735,
+      "learning_rate": 6.157826649417853e-05,
+      "loss": 0.0027,
+      "step": 476
+    },
+    {
+      "epoch": 1.541818181818182,
+      "grad_norm": 0.008260673843324184,
+      "learning_rate": 6.170763260025874e-05,
+      "loss": 0.0022,
+      "step": 477
+    },
+    {
+      "epoch": 1.545050505050505,
+      "grad_norm": 0.018339315429329872,
+      "learning_rate": 6.183699870633895e-05,
+      "loss": 0.0081,
+      "step": 478
+    },
+    {
+      "epoch": 1.5482828282828283,
+      "grad_norm": 0.010011550970375538,
+      "learning_rate": 6.196636481241915e-05,
+      "loss": 0.0053,
+      "step": 479
+    },
+    {
+      "epoch": 1.5515151515151515,
+      "grad_norm": 0.012307565659284592,
+      "learning_rate": 6.209573091849934e-05,
+      "loss": 0.0052,
+      "step": 480
+    },
+    {
+      "epoch": 1.5547474747474748,
+      "grad_norm": 0.0053881146013736725,
+      "learning_rate": 6.222509702457955e-05,
+      "loss": 0.002,
+      "step": 481
+    },
+    {
+      "epoch": 1.557979797979798,
+      "grad_norm": 0.00998155027627945,
+      "learning_rate": 6.235446313065976e-05,
+      "loss": 0.0034,
+      "step": 482
+    },
+    {
+      "epoch": 1.561212121212121,
+      "grad_norm": 0.010800400748848915,
+      "learning_rate": 6.248382923673998e-05,
+      "loss": 0.004,
+      "step": 483
+    },
+    {
+      "epoch": 1.5644444444444443,
+      "grad_norm": 0.008864471688866615,
+      "learning_rate": 6.261319534282019e-05,
+      "loss": 0.0037,
+      "step": 484
+    },
+    {
+      "epoch": 1.5676767676767676,
+      "grad_norm": 0.012302475981414318,
+      "learning_rate": 6.27425614489004e-05,
+      "loss": 0.0051,
+      "step": 485
+    },
+    {
+      "epoch": 1.5709090909090908,
+      "grad_norm": 0.007560106460005045,
+      "learning_rate": 6.28719275549806e-05,
+      "loss": 0.0022,
+      "step": 486
+    },
+    {
+      "epoch": 1.574141414141414,
+      "grad_norm": 0.018010618165135384,
+      "learning_rate": 6.30012936610608e-05,
+      "loss": 0.0067,
+      "step": 487
+    },
+    {
+      "epoch": 1.5773737373737373,
+      "grad_norm": 0.00866749044507742,
+      "learning_rate": 6.313065976714101e-05,
+      "loss": 0.0022,
+      "step": 488
+    },
+    {
+      "epoch": 1.5806060606060606,
+      "grad_norm": 0.01100226491689682,
+      "learning_rate": 6.326002587322122e-05,
+      "loss": 0.0043,
+      "step": 489
+    },
+    {
+      "epoch": 1.5838383838383838,
+      "grad_norm": 0.008842782117426395,
+      "learning_rate": 6.338939197930143e-05,
+      "loss": 0.0025,
+      "step": 490
+    },
+    {
+      "epoch": 1.587070707070707,
+      "grad_norm": 0.010475879535079002,
+      "learning_rate": 6.351875808538163e-05,
+      "loss": 0.003,
+      "step": 491
+    },
+    {
+      "epoch": 1.5903030303030303,
+      "grad_norm": 0.005673507694154978,
+      "learning_rate": 6.364812419146184e-05,
+      "loss": 0.0012,
+      "step": 492
+    },
+    {
+      "epoch": 1.5935353535353536,
+      "grad_norm": 0.009791476652026176,
+      "learning_rate": 6.377749029754204e-05,
+      "loss": 0.0019,
+      "step": 493
+    },
+    {
+      "epoch": 1.5967676767676768,
+      "grad_norm": 0.009763398207724094,
+      "learning_rate": 6.390685640362225e-05,
+      "loss": 0.0027,
+      "step": 494
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.009230737574398518,
+      "learning_rate": 6.403622250970246e-05,
+      "loss": 0.0023,
+      "step": 495
+    },
+    {
+      "epoch": 1.6032323232323233,
+      "grad_norm": 0.007090194150805473,
+      "learning_rate": 6.416558861578267e-05,
+      "loss": 0.002,
+      "step": 496
+    },
+    {
+      "epoch": 1.6064646464646466,
+      "grad_norm": 0.014568297192454338,
+      "learning_rate": 6.429495472186288e-05,
+      "loss": 0.0036,
+      "step": 497
+    },
+    {
+      "epoch": 1.6096969696969698,
+      "grad_norm": 0.011722132563591003,
+      "learning_rate": 6.442432082794309e-05,
+      "loss": 0.0035,
+      "step": 498
+    },
+    {
+      "epoch": 1.6129292929292929,
+      "grad_norm": 0.009481179527938366,
+      "learning_rate": 6.45536869340233e-05,
+      "loss": 0.0035,
+      "step": 499
+    },
+    {
+      "epoch": 1.6161616161616161,
+      "grad_norm": 0.011115781962871552,
+      "learning_rate": 6.46830530401035e-05,
+      "loss": 0.0032,
+      "step": 500
+    },
+    {
+      "epoch": 1.6161616161616161,
+      "eval_loss": 0.00416895467787981,
+      "eval_runtime": 18.5315,
+      "eval_samples_per_second": 5.396,
+      "eval_steps_per_second": 1.349,
+      "step": 500
+    },
+    {
+      "epoch": 1.6193939393939394,
+      "grad_norm": 0.010090421885251999,
+      "learning_rate": 6.48124191461837e-05,
+      "loss": 0.0036,
+      "step": 501
+    },
+    {
+      "epoch": 1.6226262626262626,
+      "grad_norm": 0.00535732414573431,
+      "learning_rate": 6.494178525226392e-05,
+      "loss": 0.0018,
+      "step": 502
+    },
+    {
+      "epoch": 1.625858585858586,
+      "grad_norm": 0.01589041016995907,
+      "learning_rate": 6.507115135834411e-05,
+      "loss": 0.0092,
+      "step": 503
+    },
+    {
+      "epoch": 1.6290909090909091,
+      "grad_norm": 0.011435211636126041,
+      "learning_rate": 6.520051746442432e-05,
+      "loss": 0.005,
+      "step": 504
+    },
+    {
+      "epoch": 1.6323232323232322,
+      "grad_norm": 0.008358100429177284,
+      "learning_rate": 6.532988357050453e-05,
+      "loss": 0.0032,
+      "step": 505
+    },
+    {
+      "epoch": 1.6355555555555554,
+      "grad_norm": 0.01174091175198555,
+      "learning_rate": 6.545924967658473e-05,
+      "loss": 0.0032,
+      "step": 506
+    },
+    {
+      "epoch": 1.6387878787878787,
+      "grad_norm": 0.011000048369169235,
+      "learning_rate": 6.558861578266494e-05,
+      "loss": 0.002,
+      "step": 507
+    },
+    {
+      "epoch": 1.642020202020202,
+      "grad_norm": 0.009679551236331463,
+      "learning_rate": 6.571798188874515e-05,
+      "loss": 0.0024,
+      "step": 508
+    },
+    {
+      "epoch": 1.6452525252525252,
+      "grad_norm": 0.01002445723861456,
+      "learning_rate": 6.584734799482536e-05,
+      "loss": 0.0019,
+      "step": 509
+    },
+    {
+      "epoch": 1.6484848484848484,
+      "grad_norm": 0.01417891588062048,
+      "learning_rate": 6.597671410090557e-05,
+      "loss": 0.0033,
+      "step": 510
+    },
+    {
+      "epoch": 1.6517171717171717,
+      "grad_norm": 0.013681537471711636,
+      "learning_rate": 6.610608020698578e-05,
+      "loss": 0.0036,
+      "step": 511
+    },
+    {
+      "epoch": 1.654949494949495,
+      "grad_norm": 0.009443574585020542,
+      "learning_rate": 6.623544631306599e-05,
+      "loss": 0.0023,
+      "step": 512
+    },
+    {
+      "epoch": 1.6581818181818182,
+      "grad_norm": 0.009754985570907593,
+      "learning_rate": 6.636481241914619e-05,
+      "loss": 0.0033,
+      "step": 513
+    },
+    {
+      "epoch": 1.6614141414141415,
+      "grad_norm": 0.010350651107728481,
+      "learning_rate": 6.649417852522638e-05,
+      "loss": 0.0024,
+      "step": 514
+    },
+    {
+      "epoch": 1.6646464646464647,
+      "grad_norm": 0.012001021765172482,
+      "learning_rate": 6.66235446313066e-05,
+      "loss": 0.0046,
+      "step": 515
+    },
+    {
+      "epoch": 1.667878787878788,
+      "grad_norm": 0.008749539963901043,
+      "learning_rate": 6.67529107373868e-05,
+      "loss": 0.0029,
+      "step": 516
+    },
+    {
+      "epoch": 1.6711111111111112,
+      "grad_norm": 0.012192701920866966,
+      "learning_rate": 6.688227684346702e-05,
+      "loss": 0.0057,
+      "step": 517
+    },
+    {
+      "epoch": 1.6743434343434345,
+      "grad_norm": 0.007326943334192038,
+      "learning_rate": 6.701164294954723e-05,
+      "loss": 0.0022,
+      "step": 518
+    },
+    {
+      "epoch": 1.6775757575757577,
+      "grad_norm": 0.01528987754136324,
+      "learning_rate": 6.714100905562742e-05,
+      "loss": 0.0081,
+      "step": 519
+    },
+    {
+      "epoch": 1.680808080808081,
+      "grad_norm": 0.011944162659347057,
+      "learning_rate": 6.727037516170763e-05,
+      "loss": 0.0041,
+      "step": 520
+    },
+    {
+      "epoch": 1.684040404040404,
+      "grad_norm": 0.007957994006574154,
+      "learning_rate": 6.739974126778784e-05,
+      "loss": 0.0029,
+      "step": 521
+    },
+    {
+      "epoch": 1.6872727272727273,
+      "grad_norm": 0.008053603582084179,
+      "learning_rate": 6.752910737386805e-05,
+      "loss": 0.0031,
+      "step": 522
+    },
+    {
+      "epoch": 1.6905050505050505,
+      "grad_norm": 0.008817918598651886,
+      "learning_rate": 6.765847347994826e-05,
+      "loss": 0.0045,
+      "step": 523
+    },
+    {
+      "epoch": 1.6937373737373738,
+      "grad_norm": 0.010544711723923683,
+      "learning_rate": 6.778783958602847e-05,
+      "loss": 0.0052,
+      "step": 524
+    },
+    {
+      "epoch": 1.696969696969697,
+      "grad_norm": 0.006746083032339811,
+      "learning_rate": 6.791720569210867e-05,
+      "loss": 0.0021,
+      "step": 525
+    },
+    {
+      "epoch": 1.696969696969697,
+      "eval_loss": 0.003925703000277281,
+      "eval_runtime": 18.5246,
+      "eval_samples_per_second": 5.398,
+      "eval_steps_per_second": 1.35,
+      "step": 525
+    },
+    {
+      "epoch": 1.70020202020202,
+      "grad_norm": 0.00789568293839693,
+      "learning_rate": 6.804657179818887e-05,
+      "loss": 0.0026,
+      "step": 526
+    },
+    {
+      "epoch": 1.7034343434343433,
+      "grad_norm": 0.004943890497088432,
+      "learning_rate": 6.817593790426908e-05,
+      "loss": 0.0016,
+      "step": 527
+    },
+    {
+      "epoch": 1.7066666666666666,
+      "grad_norm": 0.011704254895448685,
+      "learning_rate": 6.830530401034929e-05,
+      "loss": 0.0058,
+      "step": 528
+    },
+    {
+      "epoch": 1.7098989898989898,
+      "grad_norm": 0.010488653555512428,
+      "learning_rate": 6.84346701164295e-05,
+      "loss": 0.0038,
+      "step": 529
+    },
+    {
+      "epoch": 1.713131313131313,
+      "grad_norm": 0.014268319122493267,
+      "learning_rate": 6.856403622250971e-05,
+      "loss": 0.0035,
+      "step": 530
+    },
+    {
+      "epoch": 1.7163636363636363,
+      "grad_norm": 0.008911222219467163,
+      "learning_rate": 6.869340232858992e-05,
+      "loss": 0.0022,
+      "step": 531
+    },
+    {
+      "epoch": 1.7195959595959596,
+      "grad_norm": 0.007050775457173586,
+      "learning_rate": 6.882276843467012e-05,
+      "loss": 0.0024,
+      "step": 532
+    },
+    {
+      "epoch": 1.7228282828282828,
+      "grad_norm": 0.008809910155832767,
+      "learning_rate": 6.895213454075033e-05,
+      "loss": 0.0035,
+      "step": 533
+    },
+    {
+      "epoch": 1.726060606060606,
+      "grad_norm": 0.010169385932385921,
+      "learning_rate": 6.908150064683054e-05,
+      "loss": 0.0031,
+      "step": 534
+    },
+    {
+      "epoch": 1.7292929292929293,
+      "grad_norm": 0.00835514348000288,
+      "learning_rate": 6.921086675291075e-05,
+      "loss": 0.003,
+      "step": 535
+    },
+    {
+      "epoch": 1.7325252525252526,
+      "grad_norm": 0.011087946593761444,
+      "learning_rate": 6.934023285899096e-05,
+      "loss": 0.0047,
+      "step": 536
+    },
+    {
+      "epoch": 1.7357575757575758,
+      "grad_norm": 0.011055010370910168,
+      "learning_rate": 6.946959896507115e-05,
+      "loss": 0.003,
+      "step": 537
+    },
+    {
+      "epoch": 1.738989898989899,
+      "grad_norm": 0.011571435257792473,
+      "learning_rate": 6.959896507115135e-05,
+      "loss": 0.005,
+      "step": 538
+    },
+    {
+      "epoch": 1.7422222222222223,
+      "grad_norm": 0.010782378725707531,
+      "learning_rate": 6.972833117723156e-05,
+      "loss": 0.004,
+      "step": 539
+    },
+    {
+      "epoch": 1.7454545454545456,
+      "grad_norm": 0.013954983092844486,
+      "learning_rate": 6.985769728331177e-05,
+      "loss": 0.0061,
+      "step": 540
+    },
+    {
+      "epoch": 1.7486868686868688,
+      "grad_norm": 0.012455100193619728,
+      "learning_rate": 6.998706338939198e-05,
+      "loss": 0.0062,
+      "step": 541
+    },
+    {
+      "epoch": 1.7519191919191919,
+      "grad_norm": 0.012288283556699753,
+      "learning_rate": 7.011642949547219e-05,
+      "loss": 0.0035,
+      "step": 542
+    },
+    {
+      "epoch": 1.7551515151515151,
+      "grad_norm": 0.007455012295395136,
+      "learning_rate": 7.02457956015524e-05,
+      "loss": 0.0022,
+      "step": 543
+    },
+    {
+      "epoch": 1.7583838383838384,
+      "grad_norm": 0.00867515616118908,
+      "learning_rate": 7.037516170763261e-05,
+      "loss": 0.0035,
+      "step": 544
+    },
+    {
+      "epoch": 1.7616161616161616,
+      "grad_norm": 0.005005441606044769,
+      "learning_rate": 7.050452781371281e-05,
+      "loss": 0.002,
+      "step": 545
+    },
+    {
+      "epoch": 1.7648484848484849,
+      "grad_norm": 0.008555242791771889,
+      "learning_rate": 7.063389391979302e-05,
+      "loss": 0.0037,
+      "step": 546
+    },
+    {
+      "epoch": 1.768080808080808,
+      "grad_norm": 0.010418135672807693,
+      "learning_rate": 7.076326002587323e-05,
+      "loss": 0.0038,
+      "step": 547
+    },
+    {
+      "epoch": 1.7713131313131312,
+      "grad_norm": 0.006279077846556902,
+      "learning_rate": 7.089262613195343e-05,
+      "loss": 0.0017,
+      "step": 548
+    },
+    {
+      "epoch": 1.7745454545454544,
+      "grad_norm": 0.012131530791521072,
+      "learning_rate": 7.102199223803364e-05,
+      "loss": 0.0061,
+      "step": 549
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.009560024365782738,
+      "learning_rate": 7.115135834411385e-05,
+      "loss": 0.0028,
+      "step": 550
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "eval_loss": 0.0038468276616185904,
+      "eval_runtime": 18.5364,
+      "eval_samples_per_second": 5.395,
+      "eval_steps_per_second": 1.349,
+      "step": 550
+    },
+    {
+      "epoch": 1.781010101010101,
+      "grad_norm": 0.00908865500241518,
+      "learning_rate": 7.128072445019404e-05,
+      "loss": 0.0028,
+      "step": 551
+    },
+    {
+      "epoch": 1.7842424242424242,
+      "grad_norm": 0.010756799951195717,
+      "learning_rate": 7.141009055627425e-05,
+      "loss": 0.0044,
+      "step": 552
+    },
+    {
+      "epoch": 1.7874747474747474,
+      "grad_norm": 0.007486172951757908,
+      "learning_rate": 7.153945666235446e-05,
+      "loss": 0.0026,
+      "step": 553
+    },
+    {
+      "epoch": 1.7907070707070707,
+      "grad_norm": 0.011368786916136742,
+      "learning_rate": 7.166882276843467e-05,
+      "loss": 0.0029,
+      "step": 554
+    },
+    {
+      "epoch": 1.793939393939394,
+      "grad_norm": 0.009916623122990131,
+      "learning_rate": 7.179818887451488e-05,
+      "loss": 0.0041,
+      "step": 555
+    },
+    {
+      "epoch": 1.7971717171717172,
+      "grad_norm": 0.008696876466274261,
+      "learning_rate": 7.19275549805951e-05,
+      "loss": 0.0028,
+      "step": 556
+    },
+    {
+      "epoch": 1.8004040404040405,
+      "grad_norm": 0.013026717118918896,
+      "learning_rate": 7.20569210866753e-05,
+      "loss": 0.0081,
+      "step": 557
+    },
+    {
+      "epoch": 1.8036363636363637,
+      "grad_norm": 0.014385446906089783,
+      "learning_rate": 7.21862871927555e-05,
+      "loss": 0.0053,
+      "step": 558
+    },
+    {
+      "epoch": 1.806868686868687,
+      "grad_norm": 0.009371618740260601,
+      "learning_rate": 7.231565329883571e-05,
+      "loss": 0.0022,
+      "step": 559
+    },
+    {
+      "epoch": 1.8101010101010102,
+      "grad_norm": 0.00902082584798336,
+      "learning_rate": 7.244501940491591e-05,
+      "loss": 0.0043,
+      "step": 560
+    },
+    {
+      "epoch": 1.8133333333333335,
+      "grad_norm": 0.007532780524343252,
+      "learning_rate": 7.257438551099612e-05,
+      "loss": 0.0026,
+      "step": 561
+    },
+    {
+      "epoch": 1.8165656565656567,
+      "grad_norm": 0.011961591430008411,
+      "learning_rate": 7.270375161707633e-05,
+      "loss": 0.0048,
+      "step": 562
+    },
+    {
+      "epoch": 1.8197979797979797,
+      "grad_norm": 0.006793793756514788,
+      "learning_rate": 7.283311772315654e-05,
+      "loss": 0.0026,
+      "step": 563
+    },
+    {
+      "epoch": 1.823030303030303,
+      "grad_norm": 0.005821246188133955,
+      "learning_rate": 7.296248382923674e-05,
+      "loss": 0.0021,
+      "step": 564
+    },
+    {
+      "epoch": 1.8262626262626263,
+      "grad_norm": 0.010227235034108162,
+      "learning_rate": 7.309184993531695e-05,
+      "loss": 0.0036,
+      "step": 565
+    },
+    {
+      "epoch": 1.8294949494949495,
+      "grad_norm": 0.008356478065252304,
+      "learning_rate": 7.322121604139716e-05,
+      "loss": 0.0023,
+      "step": 566
+    },
+    {
+      "epoch": 1.8327272727272728,
+      "grad_norm": 0.009778296574950218,
+      "learning_rate": 7.335058214747737e-05,
+      "loss": 0.0037,
+      "step": 567
+    },
+    {
+      "epoch": 1.835959595959596,
+      "grad_norm": 0.005213058087974787,
+      "learning_rate": 7.347994825355758e-05,
+      "loss": 0.0015,
+      "step": 568
+    },
+    {
+      "epoch": 1.839191919191919,
+      "grad_norm": 0.008941391482949257,
+      "learning_rate": 7.360931435963779e-05,
+      "loss": 0.0042,
+      "step": 569
+    },
+    {
+      "epoch": 1.8424242424242423,
+      "grad_norm": 0.011023417115211487,
+      "learning_rate": 7.3738680465718e-05,
+      "loss": 0.0043,
+      "step": 570
+    },
+    {
+      "epoch": 1.8456565656565656,
+      "grad_norm": 0.007842346094548702,
+      "learning_rate": 7.38680465717982e-05,
+      "loss": 0.0022,
+      "step": 571
+    },
+    {
+      "epoch": 1.8488888888888888,
+      "grad_norm": 0.013547302223742008,
+      "learning_rate": 7.399741267787839e-05,
+      "loss": 0.0056,
+      "step": 572
+    },
+    {
+      "epoch": 1.852121212121212,
+      "grad_norm": 0.006040791980922222,
+      "learning_rate": 7.41267787839586e-05,
+      "loss": 0.0018,
+      "step": 573
+    },
+    {
+      "epoch": 1.8553535353535353,
+      "grad_norm": 0.014373473823070526,
+      "learning_rate": 7.425614489003881e-05,
+      "loss": 0.0065,
+      "step": 574
+    },
+    {
+      "epoch": 1.8585858585858586,
+      "grad_norm": 0.013441539369523525,
+      "learning_rate": 7.438551099611902e-05,
+      "loss": 0.0072,
+      "step": 575
+    },
+    {
+      "epoch": 1.8585858585858586,
+      "eval_loss": 0.0037343467120081186,
+      "eval_runtime": 18.5469,
+      "eval_samples_per_second": 5.392,
+      "eval_steps_per_second": 1.348,
+      "step": 575
+    },
+    {
+      "epoch": 1.8618181818181818,
+      "grad_norm": 0.010687317699193954,
+      "learning_rate": 7.451487710219923e-05,
+      "loss": 0.0046,
+      "step": 576
+    },
+    {
+      "epoch": 1.865050505050505,
+      "grad_norm": 0.007853105664253235,
+      "learning_rate": 7.464424320827943e-05,
+      "loss": 0.0034,
+      "step": 577
+    },
+    {
+      "epoch": 1.8682828282828283,
+      "grad_norm": 0.0073550548404455185,
+      "learning_rate": 7.477360931435964e-05,
+      "loss": 0.0023,
+      "step": 578
+    },
+    {
+      "epoch": 1.8715151515151516,
+      "grad_norm": 0.010772529058158398,
+      "learning_rate": 7.490297542043985e-05,
+      "loss": 0.0036,
+      "step": 579
+    },
+    {
+      "epoch": 1.8747474747474748,
+      "grad_norm": 0.008650882169604301,
+      "learning_rate": 7.503234152652006e-05,
+      "loss": 0.0028,
+      "step": 580
+    },
+    {
+      "epoch": 1.877979797979798,
+      "grad_norm": 0.0067758322693407536,
+      "learning_rate": 7.516170763260027e-05,
+      "loss": 0.0027,
+      "step": 581
+    },
+    {
+      "epoch": 1.8812121212121213,
+      "grad_norm": 0.006348081398755312,
+      "learning_rate": 7.529107373868047e-05,
+      "loss": 0.0022,
+      "step": 582
+    },
+    {
+      "epoch": 1.8844444444444446,
+      "grad_norm": 0.009605488739907742,
+      "learning_rate": 7.542043984476068e-05,
+      "loss": 0.0034,
+      "step": 583
+    },
+    {
+      "epoch": 1.8876767676767678,
+      "grad_norm": 0.007978033274412155,
+      "learning_rate": 7.554980595084087e-05,
+      "loss": 0.0026,
+      "step": 584
+    },
+    {
+      "epoch": 1.8909090909090909,
+      "grad_norm": 0.004183103330433369,
+      "learning_rate": 7.567917205692108e-05,
+      "loss": 0.0013,
+      "step": 585
+    },
+    {
+      "epoch": 1.8941414141414141,
+      "grad_norm": 0.009894490242004395,
+      "learning_rate": 7.58085381630013e-05,
+      "loss": 0.0036,
+      "step": 586
+    },
+    {
+      "epoch": 1.8973737373737374,
+      "grad_norm": 0.009437436237931252,
+      "learning_rate": 7.59379042690815e-05,
+      "loss": 0.0032,
+      "step": 587
+    },
+    {
+      "epoch": 1.9006060606060606,
+      "grad_norm": 0.007200904656201601,
+      "learning_rate": 7.606727037516171e-05,
+      "loss": 0.0026,
+      "step": 588
+    },
+    {
+      "epoch": 1.9038383838383839,
+      "grad_norm": 0.010573303326964378,
+      "learning_rate": 7.619663648124192e-05,
+      "loss": 0.003,
+      "step": 589
+    },
+    {
+      "epoch": 1.907070707070707,
+      "grad_norm": 0.011452763341367245,
+      "learning_rate": 7.632600258732212e-05,
+      "loss": 0.0066,
+      "step": 590
+    },
+    {
+      "epoch": 1.9103030303030302,
+      "grad_norm": 0.006278225686401129,
+      "learning_rate": 7.645536869340233e-05,
+      "loss": 0.0026,
+      "step": 591
+    },
+    {
+      "epoch": 1.9135353535353534,
+      "grad_norm": 0.009631357155740261,
+      "learning_rate": 7.658473479948254e-05,
+      "loss": 0.0028,
+      "step": 592
+    },
+    {
+      "epoch": 1.9167676767676767,
+      "grad_norm": 0.007342373952269554,
+      "learning_rate": 7.671410090556275e-05,
+      "loss": 0.003,
+      "step": 593
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.007070086430758238,
+      "learning_rate": 7.684346701164295e-05,
+      "loss": 0.0017,
+      "step": 594
+    },
+    {
+      "epoch": 1.9232323232323232,
+      "grad_norm": 0.00917192455381155,
+      "learning_rate": 7.697283311772316e-05,
+      "loss": 0.0017,
+      "step": 595
+    },
+    {
+      "epoch": 1.9264646464646464,
+      "grad_norm": 0.008652327582240105,
+      "learning_rate": 7.710219922380336e-05,
+      "loss": 0.0027,
+      "step": 596
+    },
+    {
+      "epoch": 1.9296969696969697,
+      "grad_norm": 0.012016909196972847,
+      "learning_rate": 7.723156532988357e-05,
+      "loss": 0.0036,
+      "step": 597
+    },
+    {
+      "epoch": 1.932929292929293,
+      "grad_norm": 0.006076544988900423,
+      "learning_rate": 7.736093143596378e-05,
+      "loss": 0.002,
+      "step": 598
+    },
+    {
+      "epoch": 1.9361616161616162,
+      "grad_norm": 0.004448462277650833,
+      "learning_rate": 7.749029754204399e-05,
+      "loss": 0.0015,
+      "step": 599
+    },
+    {
+      "epoch": 1.9393939393939394,
+      "grad_norm": 0.012907430529594421,
+      "learning_rate": 7.76196636481242e-05,
+      "loss": 0.0091,
+      "step": 600
+    },
+    {
+      "epoch": 1.9393939393939394,
+      "eval_loss": 0.004055679310113192,
+      "eval_runtime": 18.6094,
+      "eval_samples_per_second": 5.374,
+      "eval_steps_per_second": 1.343,
+      "step": 600
+    },
+    {
+      "epoch": 1.9426262626262627,
+      "grad_norm": 0.011052297428250313,
+      "learning_rate": 7.774902975420441e-05,
+      "loss": 0.0041,
+      "step": 601
+    },
+    {
+      "epoch": 1.945858585858586,
+      "grad_norm": 0.008796362206339836,
+      "learning_rate": 7.787839586028462e-05,
+      "loss": 0.0036,
+      "step": 602
+    },
+    {
+      "epoch": 1.9490909090909092,
+      "grad_norm": 0.014682842418551445,
+      "learning_rate": 7.800776196636481e-05,
+      "loss": 0.0048,
+      "step": 603
+    },
+    {
+      "epoch": 1.9523232323232325,
+      "grad_norm": 0.008703047409653664,
+      "learning_rate": 7.813712807244502e-05,
+      "loss": 0.0034,
+      "step": 604
+    },
+    {
+      "epoch": 1.9555555555555557,
+      "grad_norm": 0.006767407990992069,
+      "learning_rate": 7.826649417852523e-05,
+      "loss": 0.0024,
+      "step": 605
+    },
+    {
+      "epoch": 1.9587878787878787,
+      "grad_norm": 0.008383658714592457,
+      "learning_rate": 7.839586028460543e-05,
+      "loss": 0.0032,
+      "step": 606
+    },
+    {
+      "epoch": 1.962020202020202,
+      "grad_norm": 0.005494570825248957,
+      "learning_rate": 7.852522639068564e-05,
+      "loss": 0.0019,
+      "step": 607
+    },
+    {
+      "epoch": 1.9652525252525253,
+      "grad_norm": 0.009956276044249535,
+      "learning_rate": 7.865459249676585e-05,
+      "loss": 0.0032,
+      "step": 608
+    },
+    {
+      "epoch": 1.9684848484848485,
+      "grad_norm": 0.007628526072949171,
+      "learning_rate": 7.878395860284605e-05,
+      "loss": 0.0028,
+      "step": 609
+    },
+    {
+      "epoch": 1.9717171717171718,
+      "grad_norm": 0.007804389111697674,
+      "learning_rate": 7.891332470892626e-05,
+      "loss": 0.0029,
+      "step": 610
+    },
+    {
+      "epoch": 1.9749494949494948,
+      "grad_norm": 0.007861903868615627,
+      "learning_rate": 7.904269081500647e-05,
+      "loss": 0.0032,
+      "step": 611
+    },
+    {
+      "epoch": 1.978181818181818,
+      "grad_norm": 0.004911143332719803,
+      "learning_rate": 7.917205692108668e-05,
+      "loss": 0.0015,
+      "step": 612
+    },
+    {
+      "epoch": 1.9814141414141413,
+      "grad_norm": 0.009105902165174484,
+      "learning_rate": 7.930142302716689e-05,
+      "loss": 0.0052,
+      "step": 613
+    },
+    {
+      "epoch": 1.9846464646464645,
+      "grad_norm": 0.006514673586934805,
+      "learning_rate": 7.94307891332471e-05,
+      "loss": 0.0022,
+      "step": 614
+    },
+    {
+      "epoch": 1.9878787878787878,
+      "grad_norm": 0.009500369429588318,
+      "learning_rate": 7.956015523932731e-05,
+      "loss": 0.0035,
+      "step": 615
+    },
+    {
+      "epoch": 1.991111111111111,
+      "grad_norm": 0.008826248347759247,
+      "learning_rate": 7.968952134540751e-05,
+      "loss": 0.0035,
+      "step": 616
+    },
+    {
+      "epoch": 1.9943434343434343,
+      "grad_norm": 0.018535856157541275,
+      "learning_rate": 7.981888745148772e-05,
+      "loss": 0.011,
+      "step": 617
+    },
+    {
+      "epoch": 1.9975757575757576,
+      "grad_norm": 0.00722027663141489,
+      "learning_rate": 7.994825355756791e-05,
+      "loss": 0.0024,
+      "step": 618
+    },
+    {
+      "epoch": 2.000808080808081,
+      "grad_norm": 0.015815898776054382,
+      "learning_rate": 8.007761966364812e-05,
+      "loss": 0.0062,
+      "step": 619
+    },
+    {
+      "epoch": 2.004040404040404,
+      "grad_norm": 0.01083781011402607,
+      "learning_rate": 8.020698576972833e-05,
+      "loss": 0.004,
+      "step": 620
+    },
+    {
+      "epoch": 2.0072727272727273,
+      "grad_norm": 0.005312093999236822,
+      "learning_rate": 8.033635187580855e-05,
+      "loss": 0.0018,
+      "step": 621
+    },
+    {
+      "epoch": 2.0105050505050506,
+      "grad_norm": 0.006571098230779171,
+      "learning_rate": 8.046571798188874e-05,
+      "loss": 0.0022,
+      "step": 622
+    },
+    {
+      "epoch": 2.013737373737374,
+      "grad_norm": 0.005624288227409124,
+      "learning_rate": 8.059508408796895e-05,
+      "loss": 0.0017,
+      "step": 623
+    },
+    {
+      "epoch": 2.016969696969697,
+      "grad_norm": 0.007645309902727604,
+      "learning_rate": 8.072445019404916e-05,
+      "loss": 0.0025,
+      "step": 624
+    },
+    {
+      "epoch": 2.0202020202020203,
+      "grad_norm": 0.007025252562016249,
+      "learning_rate": 8.085381630012937e-05,
+      "loss": 0.0021,
+      "step": 625
+    },
+    {
+      "epoch": 2.0202020202020203,
+      "eval_loss": 0.003941759467124939,
+      "eval_runtime": 18.549,
+      "eval_samples_per_second": 5.391,
+      "eval_steps_per_second": 1.348,
+      "step": 625
+    },
+    {
+      "epoch": 2.0234343434343436,
+      "grad_norm": 0.006916250567883253,
+      "learning_rate": 8.098318240620958e-05,
+      "loss": 0.0023,
+      "step": 626
+    },
+    {
+      "epoch": 2.026666666666667,
+      "grad_norm": 0.007594529539346695,
+      "learning_rate": 8.111254851228979e-05,
+      "loss": 0.0028,
+      "step": 627
+    },
+    {
+      "epoch": 2.02989898989899,
+      "grad_norm": 0.010098805651068687,
+      "learning_rate": 8.124191461836999e-05,
+      "loss": 0.0042,
+      "step": 628
+    },
+    {
+      "epoch": 2.0331313131313133,
+      "grad_norm": 0.006069192662835121,
+      "learning_rate": 8.13712807244502e-05,
+      "loss": 0.0019,
+      "step": 629
+    },
+    {
+      "epoch": 2.036363636363636,
+      "grad_norm": 0.008519073016941547,
+      "learning_rate": 8.15006468305304e-05,
+      "loss": 0.0025,
+      "step": 630
+    },
+    {
+      "epoch": 2.0395959595959594,
+      "grad_norm": 0.007799296174198389,
+      "learning_rate": 8.163001293661061e-05,
+      "loss": 0.0028,
+      "step": 631
+    },
+    {
+      "epoch": 2.0428282828282827,
+      "grad_norm": 0.007101105526089668,
+      "learning_rate": 8.175937904269082e-05,
+      "loss": 0.0023,
+      "step": 632
+    },
+    {
+      "epoch": 2.046060606060606,
+      "grad_norm": 0.008387326262891293,
+      "learning_rate": 8.188874514877103e-05,
+      "loss": 0.0023,
+      "step": 633
+    },
+    {
+      "epoch": 2.049292929292929,
+      "grad_norm": 0.00731281191110611,
+      "learning_rate": 8.201811125485124e-05,
+      "loss": 0.002,
+      "step": 634
+    },
+    {
+      "epoch": 2.0525252525252524,
+      "grad_norm": 0.008349014446139336,
+      "learning_rate": 8.214747736093143e-05,
+      "loss": 0.0028,
+      "step": 635
+    },
+    {
+      "epoch": 2.0557575757575757,
+      "grad_norm": 0.006019700318574905,
+      "learning_rate": 8.227684346701164e-05,
+      "loss": 0.0014,
+      "step": 636
+    },
+    {
+      "epoch": 2.058989898989899,
+      "grad_norm": 0.010838819667696953,
+      "learning_rate": 8.240620957309186e-05,
+      "loss": 0.0038,
+      "step": 637
+    },
+    {
+      "epoch": 2.062222222222222,
+      "grad_norm": 0.007151686120778322,
+      "learning_rate": 8.253557567917207e-05,
+      "loss": 0.0027,
+      "step": 638
+    },
+    {
+      "epoch": 2.0654545454545454,
+      "grad_norm": 0.006484883837401867,
+      "learning_rate": 8.266494178525228e-05,
+      "loss": 0.0018,
+      "step": 639
+    },
+    {
+      "epoch": 2.0686868686868687,
+      "grad_norm": 0.008207079023122787,
+      "learning_rate": 8.279430789133247e-05,
+      "loss": 0.0028,
+      "step": 640
+    },
+    {
+      "epoch": 2.071919191919192,
+      "grad_norm": 0.005906389560550451,
+      "learning_rate": 8.292367399741268e-05,
+      "loss": 0.0018,
+      "step": 641
+    },
+    {
+      "epoch": 2.075151515151515,
+      "grad_norm": 0.010167015716433525,
+      "learning_rate": 8.305304010349288e-05,
+      "loss": 0.0033,
+      "step": 642
+    },
+    {
+      "epoch": 2.0783838383838384,
+      "grad_norm": 0.01006457768380642,
+      "learning_rate": 8.318240620957309e-05,
+      "loss": 0.0034,
+      "step": 643
+    },
+    {
+      "epoch": 2.0816161616161617,
+      "grad_norm": 0.012126877903938293,
+      "learning_rate": 8.33117723156533e-05,
+      "loss": 0.0032,
+      "step": 644
+    },
+    {
+      "epoch": 2.084848484848485,
+      "grad_norm": 0.0077040912583470345,
+      "learning_rate": 8.344113842173351e-05,
+      "loss": 0.003,
+      "step": 645
+    },
+    {
+      "epoch": 2.088080808080808,
+      "grad_norm": 0.006703955586999655,
+      "learning_rate": 8.357050452781372e-05,
+      "loss": 0.0023,
+      "step": 646
+    },
+    {
+      "epoch": 2.0913131313131315,
+      "grad_norm": 0.009514384903013706,
+      "learning_rate": 8.369987063389393e-05,
+      "loss": 0.0028,
+      "step": 647
+    },
+    {
+      "epoch": 2.0945454545454547,
+      "grad_norm": 0.008247634395956993,
+      "learning_rate": 8.382923673997413e-05,
+      "loss": 0.0015,
+      "step": 648
+    },
+    {
+      "epoch": 2.097777777777778,
+      "grad_norm": 0.006758189760148525,
+      "learning_rate": 8.395860284605434e-05,
+      "loss": 0.0017,
+      "step": 649
+    },
+    {
+      "epoch": 2.101010101010101,
+      "grad_norm": 0.00795058161020279,
+      "learning_rate": 8.408796895213455e-05,
+      "loss": 0.0026,
+      "step": 650
+    },
+    {
+      "epoch": 2.101010101010101,
+      "eval_loss": 0.004011084325611591,
+      "eval_runtime": 18.5451,
+      "eval_samples_per_second": 5.392,
+      "eval_steps_per_second": 1.348,
+      "step": 650
+    },
+    {
+      "epoch": 2.1042424242424245,
+      "grad_norm": 0.006842897739261389,
+      "learning_rate": 8.421733505821476e-05,
+      "loss": 0.0019,
+      "step": 651
+    },
+    {
+      "epoch": 2.1074747474747473,
+      "grad_norm": 0.008785001002252102,
+      "learning_rate": 8.434670116429496e-05,
+      "loss": 0.0018,
+      "step": 652
+    },
+    {
+      "epoch": 2.1107070707070705,
+      "grad_norm": 0.012285329401493073,
+      "learning_rate": 8.447606727037517e-05,
+      "loss": 0.0041,
+      "step": 653
+    },
+    {
+      "epoch": 2.113939393939394,
+      "grad_norm": 0.0077829184010624886,
+      "learning_rate": 8.460543337645536e-05,
+      "loss": 0.0025,
+      "step": 654
+    },
+    {
+      "epoch": 2.117171717171717,
+      "grad_norm": 0.008003472350537777,
+      "learning_rate": 8.473479948253557e-05,
+      "loss": 0.0032,
+      "step": 655
+    },
+    {
+      "epoch": 2.1204040404040403,
+      "grad_norm": 0.004586879629641771,
+      "learning_rate": 8.486416558861578e-05,
+      "loss": 0.0012,
+      "step": 656
+    },
+    {
+      "epoch": 2.1236363636363635,
+      "grad_norm": 0.005405848380178213,
+      "learning_rate": 8.499353169469599e-05,
+      "loss": 0.0018,
+      "step": 657
+    },
+    {
+      "epoch": 2.126868686868687,
+      "grad_norm": 0.004980962257832289,
+      "learning_rate": 8.51228978007762e-05,
+      "loss": 0.0021,
+      "step": 658
+    },
+    {
+      "epoch": 2.13010101010101,
+      "grad_norm": 0.007102985866367817,
+      "learning_rate": 8.525226390685641e-05,
+      "loss": 0.0027,
+      "step": 659
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 0.0070281801745295525,
+      "learning_rate": 8.538163001293662e-05,
+      "loss": 0.0024,
+      "step": 660
+    },
+    {
+      "epoch": 2.1365656565656566,
+      "grad_norm": 0.01416573766618967,
+      "learning_rate": 8.551099611901682e-05,
+      "loss": 0.0028,
+      "step": 661
+    },
+    {
+      "epoch": 2.13979797979798,
+      "grad_norm": 0.00932216364890337,
+      "learning_rate": 8.564036222509703e-05,
+      "loss": 0.0048,
+      "step": 662
+    },
+    {
+      "epoch": 2.143030303030303,
+      "grad_norm": 0.008925393223762512,
+      "learning_rate": 8.576972833117724e-05,
+      "loss": 0.0023,
+      "step": 663
+    },
+    {
+      "epoch": 2.1462626262626263,
+      "grad_norm": 0.006588651333004236,
+      "learning_rate": 8.589909443725744e-05,
+      "loss": 0.0028,
+      "step": 664
+    },
+    {
+      "epoch": 2.1494949494949496,
+      "grad_norm": 0.004025767557322979,
+      "learning_rate": 8.602846054333765e-05,
+      "loss": 0.0013,
+      "step": 665
+    },
+    {
+      "epoch": 2.152727272727273,
+      "grad_norm": 0.005784463603049517,
+      "learning_rate": 8.615782664941786e-05,
+      "loss": 0.0022,
+      "step": 666
+    },
+    {
+      "epoch": 2.155959595959596,
+      "grad_norm": 0.008596932515501976,
+      "learning_rate": 8.628719275549805e-05,
+      "loss": 0.0037,
+      "step": 667
+    },
+    {
+      "epoch": 2.1591919191919193,
+      "grad_norm": 0.006371675059199333,
+      "learning_rate": 8.641655886157827e-05,
+      "loss": 0.0016,
+      "step": 668
+    },
+    {
+      "epoch": 2.1624242424242426,
+      "grad_norm": 0.00666641304269433,
+      "learning_rate": 8.654592496765848e-05,
+      "loss": 0.0028,
+      "step": 669
+    },
+    {
+      "epoch": 2.165656565656566,
+      "grad_norm": 0.006822609808295965,
+      "learning_rate": 8.667529107373869e-05,
+      "loss": 0.0029,
+      "step": 670
+    },
+    {
+      "epoch": 2.168888888888889,
+      "grad_norm": 0.012946972623467445,
+      "learning_rate": 8.68046571798189e-05,
+      "loss": 0.0053,
+      "step": 671
+    },
+    {
+      "epoch": 2.172121212121212,
+      "grad_norm": 0.005195168312638998,
+      "learning_rate": 8.69340232858991e-05,
+      "loss": 0.0013,
+      "step": 672
+    },
+    {
+      "epoch": 2.175353535353535,
+      "grad_norm": 0.004933821968734264,
+      "learning_rate": 8.706338939197932e-05,
+      "loss": 0.0014,
+      "step": 673
+    },
+    {
+      "epoch": 2.1785858585858584,
+      "grad_norm": 0.00683283805847168,
+      "learning_rate": 8.719275549805951e-05,
+      "loss": 0.0027,
+      "step": 674
+    },
+    {
+      "epoch": 2.1818181818181817,
+      "grad_norm": 0.008817669935524464,
+      "learning_rate": 8.732212160413972e-05,
+      "loss": 0.0028,
+      "step": 675
+    },
+    {
+      "epoch": 2.1818181818181817,
+      "eval_loss": 0.00383288087323308,
+      "eval_runtime": 18.5346,
+      "eval_samples_per_second": 5.395,
+      "eval_steps_per_second": 1.349,
+      "step": 675
+    },
+    {
+      "epoch": 2.185050505050505,
+      "grad_norm": 0.004791957791894674,
+      "learning_rate": 8.745148771021992e-05,
+      "loss": 0.0014,
+      "step": 676
+    },
+    {
+      "epoch": 2.188282828282828,
+      "grad_norm": 0.0061204745434224606,
+      "learning_rate": 8.758085381630013e-05,
+      "loss": 0.0018,
+      "step": 677
+    },
+    {
+      "epoch": 2.1915151515151514,
+      "grad_norm": 0.011844735592603683,
+      "learning_rate": 8.771021992238034e-05,
+      "loss": 0.0048,
+      "step": 678
+    },
+    {
+      "epoch": 2.1947474747474747,
+      "grad_norm": 0.006783795543015003,
+      "learning_rate": 8.783958602846055e-05,
+      "loss": 0.0019,
+      "step": 679
+    },
+    {
+      "epoch": 2.197979797979798,
+      "grad_norm": 0.011243957094848156,
+      "learning_rate": 8.796895213454075e-05,
+      "loss": 0.0036,
+      "step": 680
+    },
+    {
+      "epoch": 2.201212121212121,
+      "grad_norm": 0.007772068493068218,
+      "learning_rate": 8.809831824062096e-05,
+      "loss": 0.0022,
+      "step": 681
+    },
+    {
+      "epoch": 2.2044444444444444,
+      "grad_norm": 0.0051286122761666775,
+      "learning_rate": 8.822768434670117e-05,
+      "loss": 0.0013,
+      "step": 682
+    },
+    {
+      "epoch": 2.2076767676767677,
+      "grad_norm": 0.01016032975167036,
+      "learning_rate": 8.835705045278138e-05,
+      "loss": 0.0039,
+      "step": 683
+    },
+    {
+      "epoch": 2.210909090909091,
+      "grad_norm": 0.006838168017566204,
+      "learning_rate": 8.848641655886159e-05,
+      "loss": 0.002,
+      "step": 684
+    },
+    {
+      "epoch": 2.214141414141414,
+      "grad_norm": 0.008824942633509636,
+      "learning_rate": 8.86157826649418e-05,
+      "loss": 0.0024,
+      "step": 685
+    },
+    {
+      "epoch": 2.2173737373737374,
+      "grad_norm": 0.00688344007357955,
+      "learning_rate": 8.8745148771022e-05,
+      "loss": 0.002,
+      "step": 686
+    },
+    {
+      "epoch": 2.2206060606060607,
+      "grad_norm": 0.0066194599494338036,
+      "learning_rate": 8.88745148771022e-05,
+      "loss": 0.0019,
+      "step": 687
+    },
+    {
+      "epoch": 2.223838383838384,
+      "grad_norm": 0.007829937152564526,
+      "learning_rate": 8.90038809831824e-05,
+      "loss": 0.0022,
+      "step": 688
+    },
+    {
+      "epoch": 2.227070707070707,
+      "grad_norm": 0.009328142739832401,
+      "learning_rate": 8.913324708926261e-05,
+      "loss": 0.002,
+      "step": 689
+    },
+    {
+      "epoch": 2.2303030303030305,
+      "grad_norm": 0.009020456112921238,
+      "learning_rate": 8.926261319534282e-05,
+      "loss": 0.0031,
+      "step": 690
+    },
+    {
+      "epoch": 2.2335353535353537,
+      "grad_norm": 0.00872059352695942,
+      "learning_rate": 8.939197930142303e-05,
+      "loss": 0.0026,
+      "step": 691
+    },
+    {
+      "epoch": 2.236767676767677,
+      "grad_norm": 0.009462382644414902,
+      "learning_rate": 8.952134540750324e-05,
+      "loss": 0.0034,
+      "step": 692
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.0071613360196352005,
+      "learning_rate": 8.965071151358344e-05,
+      "loss": 0.002,
+      "step": 693
+    },
+    {
+      "epoch": 2.2432323232323235,
+      "grad_norm": 0.00632578507065773,
+      "learning_rate": 8.978007761966365e-05,
+      "loss": 0.0019,
+      "step": 694
+    },
+    {
+      "epoch": 2.2464646464646463,
+      "grad_norm": 0.007571411784738302,
+      "learning_rate": 8.990944372574386e-05,
+      "loss": 0.0023,
+      "step": 695
+    },
+    {
+      "epoch": 2.2496969696969695,
+      "grad_norm": 0.009794604033231735,
+      "learning_rate": 9.003880983182407e-05,
+      "loss": 0.0032,
+      "step": 696
+    },
+    {
+      "epoch": 2.252929292929293,
+      "grad_norm": 0.006740176118910313,
+      "learning_rate": 9.016817593790428e-05,
+      "loss": 0.002,
+      "step": 697
+    },
+    {
+      "epoch": 2.256161616161616,
+      "grad_norm": 0.008458360098302364,
+      "learning_rate": 9.029754204398448e-05,
+      "loss": 0.0022,
+      "step": 698
+    },
+    {
+      "epoch": 2.2593939393939393,
+      "grad_norm": 0.010913528501987457,
+      "learning_rate": 9.042690815006469e-05,
+      "loss": 0.0027,
+      "step": 699
+    },
+    {
+      "epoch": 2.2626262626262625,
+      "grad_norm": 0.00851623248308897,
+      "learning_rate": 9.055627425614489e-05,
+      "loss": 0.0027,
+      "step": 700
+    },
+    {
+      "epoch": 2.2626262626262625,
+      "eval_loss": 0.0037586898542940617,
+      "eval_runtime": 18.6064,
+      "eval_samples_per_second": 5.375,
+      "eval_steps_per_second": 1.344,
+      "step": 700
+    },
+    {
+      "epoch": 2.265858585858586,
+      "grad_norm": 0.008736968040466309,
+      "learning_rate": 9.06856403622251e-05,
+      "loss": 0.0026,
+      "step": 701
+    },
+    {
+      "epoch": 2.269090909090909,
+      "grad_norm": 0.007413066923618317,
+      "learning_rate": 9.08150064683053e-05,
+      "loss": 0.002,
+      "step": 702
+    },
+    {
+      "epoch": 2.2723232323232323,
+      "grad_norm": 0.009513185359537601,
+      "learning_rate": 9.094437257438552e-05,
+      "loss": 0.0029,
+      "step": 703
+    },
+    {
+      "epoch": 2.2755555555555556,
+      "grad_norm": 0.006239297799766064,
+      "learning_rate": 9.107373868046573e-05,
+      "loss": 0.0024,
+      "step": 704
+    },
+    {
+      "epoch": 2.278787878787879,
+      "grad_norm": 0.009737716056406498,
+      "learning_rate": 9.120310478654594e-05,
+      "loss": 0.0032,
+      "step": 705
+    },
+    {
+      "epoch": 2.282020202020202,
+      "grad_norm": 0.005764431785792112,
+      "learning_rate": 9.133247089262613e-05,
+      "loss": 0.0021,
+      "step": 706
+    },
+    {
+      "epoch": 2.2852525252525253,
+      "grad_norm": 0.01445918157696724,
+      "learning_rate": 9.146183699870634e-05,
+      "loss": 0.0041,
+      "step": 707
+    },
+    {
+      "epoch": 2.2884848484848486,
+      "grad_norm": 0.007700520567595959,
+      "learning_rate": 9.159120310478655e-05,
+      "loss": 0.0024,
+      "step": 708
+    },
+    {
+      "epoch": 2.291717171717172,
+      "grad_norm": 0.007032734341919422,
+      "learning_rate": 9.172056921086676e-05,
+      "loss": 0.002,
+      "step": 709
+    },
+    {
+      "epoch": 2.294949494949495,
+      "grad_norm": 0.012682800181210041,
+      "learning_rate": 9.184993531694696e-05,
+      "loss": 0.0023,
+      "step": 710
+    },
+    {
+      "epoch": 2.2981818181818183,
+      "grad_norm": 0.01120555866509676,
+      "learning_rate": 9.197930142302717e-05,
+      "loss": 0.0065,
+      "step": 711
+    },
+    {
+      "epoch": 2.3014141414141416,
+      "grad_norm": 0.009766733273863792,
+      "learning_rate": 9.210866752910737e-05,
+      "loss": 0.0034,
+      "step": 712
+    },
+    {
+      "epoch": 2.304646464646465,
+      "grad_norm": 0.009006375446915627,
+      "learning_rate": 9.223803363518758e-05,
+      "loss": 0.0028,
+      "step": 713
+    },
+    {
+      "epoch": 2.3078787878787876,
+      "grad_norm": 0.0110854497179389,
+      "learning_rate": 9.236739974126779e-05,
+      "loss": 0.0044,
+      "step": 714
+    },
+    {
+      "epoch": 2.311111111111111,
+      "grad_norm": 0.00787447765469551,
+      "learning_rate": 9.2496765847348e-05,
+      "loss": 0.0028,
+      "step": 715
+    },
+    {
+      "epoch": 2.314343434343434,
+      "grad_norm": 0.007719083689153194,
+      "learning_rate": 9.262613195342821e-05,
+      "loss": 0.0022,
+      "step": 716
+    },
+    {
+      "epoch": 2.3175757575757574,
+      "grad_norm": 0.004776225425302982,
+      "learning_rate": 9.275549805950842e-05,
+      "loss": 0.0014,
+      "step": 717
+    },
+    {
+      "epoch": 2.3208080808080807,
+      "grad_norm": 0.009256124496459961,
+      "learning_rate": 9.288486416558863e-05,
+      "loss": 0.0034,
+      "step": 718
+    },
+    {
+      "epoch": 2.324040404040404,
+      "grad_norm": 0.00416655233129859,
+      "learning_rate": 9.301423027166883e-05,
+      "loss": 0.0014,
+      "step": 719
+    },
+    {
+      "epoch": 2.327272727272727,
+      "grad_norm": 0.010706130415201187,
+      "learning_rate": 9.314359637774904e-05,
+      "loss": 0.0064,
+      "step": 720
+    },
+    {
+      "epoch": 2.3305050505050504,
+      "grad_norm": 0.010033998638391495,
+      "learning_rate": 9.327296248382925e-05,
+      "loss": 0.0032,
+      "step": 721
+    },
+    {
+      "epoch": 2.3337373737373737,
+      "grad_norm": 0.010281972587108612,
+      "learning_rate": 9.340232858990944e-05,
+      "loss": 0.0046,
+      "step": 722
+    },
+    {
+      "epoch": 2.336969696969697,
+      "grad_norm": 0.007038958836346865,
+      "learning_rate": 9.353169469598965e-05,
+      "loss": 0.0025,
+      "step": 723
+    },
+    {
+      "epoch": 2.34020202020202,
+      "grad_norm": 0.006714510731399059,
+      "learning_rate": 9.366106080206986e-05,
+      "loss": 0.0023,
+      "step": 724
+    },
+    {
+      "epoch": 2.3434343434343434,
+      "grad_norm": 0.008812773041427135,
+      "learning_rate": 9.379042690815006e-05,
+      "loss": 0.0032,
+      "step": 725
+    },
+    {
+      "epoch": 2.3434343434343434,
+      "eval_loss": 0.0038471692241728306,
+      "eval_runtime": 18.6172,
+      "eval_samples_per_second": 5.371,
+      "eval_steps_per_second": 1.343,
+      "step": 725
+    },
+    {
+      "epoch": 2.3466666666666667,
+      "grad_norm": 0.00827405508607626,
+      "learning_rate": 9.391979301423027e-05,
+      "loss": 0.0024,
+      "step": 726
+    },
+    {
+      "epoch": 2.34989898989899,
+      "grad_norm": 0.009304560720920563,
+      "learning_rate": 9.404915912031048e-05,
+      "loss": 0.0032,
+      "step": 727
+    },
+    {
+      "epoch": 2.353131313131313,
+      "grad_norm": 0.004189759958535433,
+      "learning_rate": 9.417852522639069e-05,
+      "loss": 0.001,
+      "step": 728
+    },
+    {
+      "epoch": 2.3563636363636364,
+      "grad_norm": 0.010822804644703865,
+      "learning_rate": 9.43078913324709e-05,
+      "loss": 0.0037,
+      "step": 729
+    },
+    {
+      "epoch": 2.3595959595959597,
+      "grad_norm": 0.010418107733130455,
+      "learning_rate": 9.443725743855111e-05,
+      "loss": 0.0029,
+      "step": 730
+    },
+    {
+      "epoch": 2.362828282828283,
+      "grad_norm": 0.013821220025420189,
+      "learning_rate": 9.456662354463132e-05,
+      "loss": 0.0037,
+      "step": 731
+    },
+    {
+      "epoch": 2.366060606060606,
+      "grad_norm": 0.006959732621908188,
+      "learning_rate": 9.469598965071152e-05,
+      "loss": 0.0021,
+      "step": 732
+    },
+    {
+      "epoch": 2.3692929292929295,
+      "grad_norm": 0.010071862488985062,
+      "learning_rate": 9.482535575679173e-05,
+      "loss": 0.0019,
+      "step": 733
+    },
+    {
+      "epoch": 2.3725252525252527,
+      "grad_norm": 0.010580536909401417,
+      "learning_rate": 9.495472186287193e-05,
+      "loss": 0.0041,
+      "step": 734
+    },
+    {
+      "epoch": 2.375757575757576,
+      "grad_norm": 0.009700599126517773,
+      "learning_rate": 9.508408796895214e-05,
+      "loss": 0.0036,
+      "step": 735
+    },
+    {
+      "epoch": 2.378989898989899,
+      "grad_norm": 0.008695641532540321,
+      "learning_rate": 9.521345407503235e-05,
+      "loss": 0.0034,
+      "step": 736
+    },
+    {
+      "epoch": 2.3822222222222225,
+      "grad_norm": 0.007013495545834303,
+      "learning_rate": 9.534282018111256e-05,
+      "loss": 0.0024,
+      "step": 737
+    },
+    {
+      "epoch": 2.3854545454545453,
+      "grad_norm": 0.007236323785036802,
+      "learning_rate": 9.547218628719275e-05,
+      "loss": 0.002,
+      "step": 738
+    },
+    {
+      "epoch": 2.3886868686868685,
+      "grad_norm": 0.005592238157987595,
+      "learning_rate": 9.560155239327296e-05,
+      "loss": 0.0016,
+      "step": 739
+    },
+    {
+      "epoch": 2.391919191919192,
+      "grad_norm": 0.00680518476292491,
+      "learning_rate": 9.573091849935317e-05,
+      "loss": 0.002,
+      "step": 740
+    },
+    {
+      "epoch": 2.395151515151515,
+      "grad_norm": 0.008009014651179314,
+      "learning_rate": 9.586028460543338e-05,
+      "loss": 0.0018,
+      "step": 741
+    },
+    {
+      "epoch": 2.3983838383838383,
+      "grad_norm": 0.006378913763910532,
+      "learning_rate": 9.59896507115136e-05,
+      "loss": 0.0019,
+      "step": 742
+    },
+    {
+      "epoch": 2.4016161616161615,
+      "grad_norm": 0.006395281758159399,
+      "learning_rate": 9.61190168175938e-05,
+      "loss": 0.002,
+      "step": 743
+    },
+    {
+      "epoch": 2.404848484848485,
+      "grad_norm": 0.008712049573659897,
+      "learning_rate": 9.6248382923674e-05,
+      "loss": 0.0026,
+      "step": 744
+    },
+    {
+      "epoch": 2.408080808080808,
+      "grad_norm": 0.009031414985656738,
+      "learning_rate": 9.63777490297542e-05,
+      "loss": 0.0029,
+      "step": 745
+    },
+    {
+      "epoch": 2.4113131313131313,
+      "grad_norm": 0.010614071041345596,
+      "learning_rate": 9.650711513583441e-05,
+      "loss": 0.0039,
+      "step": 746
+    },
+    {
+      "epoch": 2.4145454545454546,
+      "grad_norm": 0.006484184879809618,
+      "learning_rate": 9.663648124191462e-05,
+      "loss": 0.0015,
+      "step": 747
+    },
+    {
+      "epoch": 2.417777777777778,
+      "grad_norm": 0.005564129911363125,
+      "learning_rate": 9.676584734799483e-05,
+      "loss": 0.0018,
+      "step": 748
+    },
+    {
+      "epoch": 2.421010101010101,
+      "grad_norm": 0.00510470662266016,
+      "learning_rate": 9.689521345407504e-05,
+      "loss": 0.0013,
+      "step": 749
+    },
+    {
+      "epoch": 2.4242424242424243,
+      "grad_norm": 0.007081560790538788,
+      "learning_rate": 9.702457956015525e-05,
+      "loss": 0.0022,
+      "step": 750
+    },
+    {
+      "epoch": 2.4242424242424243,
+      "eval_loss": 0.0037562157958745956,
+      "eval_runtime": 18.5443,
+      "eval_samples_per_second": 5.393,
+      "eval_steps_per_second": 1.348,
+      "step": 750
+    },
+    {
+      "epoch": 2.4274747474747476,
+      "grad_norm": 0.009013883769512177,
+      "learning_rate": 9.715394566623545e-05,
+      "loss": 0.0031,
+      "step": 751
+    },
+    {
+      "epoch": 2.430707070707071,
+      "grad_norm": 0.009731669910252094,
+      "learning_rate": 9.728331177231566e-05,
+      "loss": 0.0041,
+      "step": 752
+    },
+    {
+      "epoch": 2.433939393939394,
+      "grad_norm": 0.010123556479811668,
+      "learning_rate": 9.741267787839587e-05,
+      "loss": 0.0042,
+      "step": 753
+    },
+    {
+      "epoch": 2.4371717171717173,
+      "grad_norm": 0.005915546324104071,
+      "learning_rate": 9.754204398447608e-05,
+      "loss": 0.0024,
+      "step": 754
+    },
+    {
+      "epoch": 2.4404040404040406,
+      "grad_norm": 0.008191730827093124,
+      "learning_rate": 9.767141009055629e-05,
+      "loss": 0.0032,
+      "step": 755
+    },
+    {
+      "epoch": 2.443636363636364,
+      "grad_norm": 0.005690063815563917,
+      "learning_rate": 9.780077619663648e-05,
+      "loss": 0.0021,
+      "step": 756
+    },
+    {
+      "epoch": 2.4468686868686866,
+      "grad_norm": 0.008176354691386223,
+      "learning_rate": 9.793014230271668e-05,
+      "loss": 0.0027,
+      "step": 757
+    },
+    {
+      "epoch": 2.45010101010101,
+      "grad_norm": 0.007075440138578415,
+      "learning_rate": 9.805950840879689e-05,
+      "loss": 0.0023,
+      "step": 758
+    },
+    {
+      "epoch": 2.453333333333333,
+      "grad_norm": 0.006414720788598061,
+      "learning_rate": 9.81888745148771e-05,
+      "loss": 0.0018,
+      "step": 759
+    },
+    {
+      "epoch": 2.4565656565656564,
+      "grad_norm": 0.011210867203772068,
+      "learning_rate": 9.831824062095731e-05,
+      "loss": 0.0039,
+      "step": 760
+    },
+    {
+      "epoch": 2.4597979797979797,
+      "grad_norm": 0.006865155417472124,
+      "learning_rate": 9.844760672703752e-05,
+      "loss": 0.002,
+      "step": 761
+    },
+    {
+      "epoch": 2.463030303030303,
+      "grad_norm": 0.010836309753358364,
+      "learning_rate": 9.857697283311773e-05,
+      "loss": 0.0036,
+      "step": 762
+    },
+    {
+      "epoch": 2.466262626262626,
+      "grad_norm": 0.007952204905450344,
+      "learning_rate": 9.870633893919794e-05,
+      "loss": 0.003,
+      "step": 763
+    },
+    {
+      "epoch": 2.4694949494949494,
+      "grad_norm": 0.007584880106151104,
+      "learning_rate": 9.883570504527814e-05,
+      "loss": 0.0023,
+      "step": 764
+    },
+    {
+      "epoch": 2.4727272727272727,
+      "grad_norm": 0.00835363008081913,
+      "learning_rate": 9.896507115135835e-05,
+      "loss": 0.0016,
+      "step": 765
+    },
+    {
+      "epoch": 2.475959595959596,
+      "grad_norm": 0.007696302607655525,
+      "learning_rate": 9.909443725743856e-05,
+      "loss": 0.0023,
+      "step": 766
+    },
+    {
+      "epoch": 2.479191919191919,
+      "grad_norm": 0.008753744885325432,
+      "learning_rate": 9.922380336351877e-05,
+      "loss": 0.0027,
+      "step": 767
+    },
+    {
+      "epoch": 2.4824242424242424,
+      "grad_norm": 0.007579338271170855,
+      "learning_rate": 9.935316946959897e-05,
+      "loss": 0.0028,
+      "step": 768
+    },
+    {
+      "epoch": 2.4856565656565657,
+      "grad_norm": 0.011279474943876266,
+      "learning_rate": 9.948253557567918e-05,
+      "loss": 0.0042,
+      "step": 769
+    },
+    {
+      "epoch": 2.488888888888889,
+      "grad_norm": 0.008233487606048584,
+      "learning_rate": 9.961190168175937e-05,
+      "loss": 0.0027,
+      "step": 770
+    },
+    {
+      "epoch": 2.492121212121212,
+      "grad_norm": 0.006320327054709196,
+      "learning_rate": 9.974126778783958e-05,
+      "loss": 0.0014,
+      "step": 771
+    },
+    {
+      "epoch": 2.4953535353535354,
+      "grad_norm": 0.007094175089150667,
+      "learning_rate": 9.98706338939198e-05,
+      "loss": 0.0019,
+      "step": 772
+    },
+    {
+      "epoch": 2.4985858585858587,
+      "grad_norm": 0.007033037953078747,
+      "learning_rate": 0.0001,
+      "loss": 0.0021,
+      "step": 773
+    },
+    {
+      "epoch": 2.501818181818182,
+      "grad_norm": 0.008634313941001892,
+      "learning_rate": 9.999999489471233e-05,
+      "loss": 0.0027,
+      "step": 774
+    },
+    {
+      "epoch": 2.505050505050505,
+      "grad_norm": 0.00898764282464981,
+      "learning_rate": 9.99999795788503e-05,
+      "loss": 0.0024,
+      "step": 775
+    },
+    {
+      "epoch": 2.505050505050505,
+      "eval_loss": 0.0035804486833512783,
+      "eval_runtime": 18.5794,
+      "eval_samples_per_second": 5.382,
+      "eval_steps_per_second": 1.346,
+      "step": 775
+    },
+    {
+      "epoch": 2.5082828282828284,
+      "grad_norm": 0.007207226939499378,
+      "learning_rate": 9.99999540524171e-05,
+      "loss": 0.0022,
+      "step": 776
+    },
+    {
+      "epoch": 2.5115151515151517,
+      "grad_norm": 0.012771569192409515,
+      "learning_rate": 9.999991831541789e-05,
+      "loss": 0.0045,
+      "step": 777
+    },
+    {
+      "epoch": 2.514747474747475,
+      "grad_norm": 0.0067327446304261684,
+      "learning_rate": 9.999987236786e-05,
+      "loss": 0.002,
+      "step": 778
+    },
+    {
+      "epoch": 2.517979797979798,
+      "grad_norm": 0.009945794008672237,
+      "learning_rate": 9.999981620975281e-05,
+      "loss": 0.0034,
+      "step": 779
+    },
+    {
+      "epoch": 2.5212121212121215,
+      "grad_norm": 0.006773203145712614,
+      "learning_rate": 9.999974984110779e-05,
+      "loss": 0.0023,
+      "step": 780
+    },
+    {
+      "epoch": 2.5244444444444447,
+      "grad_norm": 0.00785363744944334,
+      "learning_rate": 9.999967326193847e-05,
+      "loss": 0.0019,
+      "step": 781
+    },
+    {
+      "epoch": 2.5276767676767675,
+      "grad_norm": 0.010476037859916687,
+      "learning_rate": 9.999958647226049e-05,
+      "loss": 0.0033,
+      "step": 782
+    },
+    {
+      "epoch": 2.5309090909090908,
+      "grad_norm": 0.007686255965381861,
+      "learning_rate": 9.999948947209162e-05,
+      "loss": 0.0026,
+      "step": 783
+    },
+    {
+      "epoch": 2.534141414141414,
+      "grad_norm": 0.007367600221186876,
+      "learning_rate": 9.999938226145161e-05,
+      "loss": 0.0023,
+      "step": 784
+    },
+    {
+      "epoch": 2.5373737373737373,
+      "grad_norm": 0.00861043855547905,
+      "learning_rate": 9.999926484036237e-05,
+      "loss": 0.0026,
+      "step": 785
+    },
+    {
+      "epoch": 2.5406060606060605,
+      "grad_norm": 0.009773798286914825,
+      "learning_rate": 9.999913720884791e-05,
+      "loss": 0.0025,
+      "step": 786
+    },
+    {
+      "epoch": 2.543838383838384,
+      "grad_norm": 0.006338477600365877,
+      "learning_rate": 9.999899936693426e-05,
+      "loss": 0.0018,
+      "step": 787
+    },
+    {
+      "epoch": 2.547070707070707,
+      "grad_norm": 0.012471425347030163,
+      "learning_rate": 9.99988513146496e-05,
+      "loss": 0.0053,
+      "step": 788
+    },
+    {
+      "epoch": 2.5503030303030303,
+      "grad_norm": 0.009621040895581245,
+      "learning_rate": 9.999869305202412e-05,
+      "loss": 0.0031,
+      "step": 789
+    },
+    {
+      "epoch": 2.5535353535353535,
+      "grad_norm": 0.006214372348040342,
+      "learning_rate": 9.999852457909018e-05,
+      "loss": 0.0016,
+      "step": 790
+    },
+    {
+      "epoch": 2.556767676767677,
+      "grad_norm": 0.011059516109526157,
+      "learning_rate": 9.999834589588217e-05,
+      "loss": 0.0024,
+      "step": 791
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.012596976943314075,
+      "learning_rate": 9.999815700243656e-05,
+      "loss": 0.0061,
+      "step": 792
+    },
+    {
+      "epoch": 2.5632323232323233,
+      "grad_norm": 0.012062342837452888,
+      "learning_rate": 9.999795789879196e-05,
+      "loss": 0.0031,
+      "step": 793
+    },
+    {
+      "epoch": 2.5664646464646466,
+      "grad_norm": 0.012593230232596397,
+      "learning_rate": 9.9997748584989e-05,
+      "loss": 0.0016,
+      "step": 794
+    },
+    {
+      "epoch": 2.56969696969697,
+      "grad_norm": 0.0052862451411783695,
+      "learning_rate": 9.999752906107042e-05,
+      "loss": 0.0017,
+      "step": 795
+    },
+    {
+      "epoch": 2.572929292929293,
+      "grad_norm": 0.009504672139883041,
+      "learning_rate": 9.999729932708109e-05,
+      "loss": 0.0031,
+      "step": 796
+    },
+    {
+      "epoch": 2.5761616161616163,
+      "grad_norm": 0.006016214843839407,
+      "learning_rate": 9.999705938306789e-05,
+      "loss": 0.0022,
+      "step": 797
+    },
+    {
+      "epoch": 2.579393939393939,
+      "grad_norm": 0.009827526286244392,
+      "learning_rate": 9.999680922907982e-05,
+      "loss": 0.0028,
+      "step": 798
+    },
+    {
+      "epoch": 2.5826262626262624,
+      "grad_norm": 0.00939889531582594,
+      "learning_rate": 9.999654886516798e-05,
+      "loss": 0.0025,
+      "step": 799
+    },
+    {
+      "epoch": 2.5858585858585856,
+      "grad_norm": 0.004873231053352356,
+      "learning_rate": 9.999627829138554e-05,
+      "loss": 0.0015,
+      "step": 800
+    },
+    {
+      "epoch": 2.5858585858585856,
+      "eval_loss": 0.003381554502993822,
+      "eval_runtime": 18.5553,
+      "eval_samples_per_second": 5.389,
+      "eval_steps_per_second": 1.347,
+      "step": 800
+    },
+    {
+      "epoch": 2.589090909090909,
+      "grad_norm": 0.011554384604096413,
+      "learning_rate": 9.999599750778772e-05,
+      "loss": 0.0024,
+      "step": 801
+    },
+    {
+      "epoch": 2.592323232323232,
+      "grad_norm": 0.00788146536797285,
+      "learning_rate": 9.999570651443191e-05,
+      "loss": 0.0028,
+      "step": 802
+    },
+    {
+      "epoch": 2.5955555555555554,
+      "grad_norm": 0.007881752215325832,
+      "learning_rate": 9.99954053113775e-05,
+      "loss": 0.003,
+      "step": 803
+    },
+    {
+      "epoch": 2.5987878787878786,
+      "grad_norm": 0.00717515591531992,
+      "learning_rate": 9.9995093898686e-05,
+      "loss": 0.0025,
+      "step": 804
+    },
+    {
+      "epoch": 2.602020202020202,
+      "grad_norm": 0.008232329972088337,
+      "learning_rate": 9.999477227642103e-05,
+      "loss": 0.0029,
+      "step": 805
+    },
+    {
+      "epoch": 2.605252525252525,
+      "grad_norm": 0.017594264820218086,
+      "learning_rate": 9.999444044464823e-05,
+      "loss": 0.0038,
+      "step": 806
+    },
+    {
+      "epoch": 2.6084848484848484,
+      "grad_norm": 0.006009047385305166,
+      "learning_rate": 9.999409840343539e-05,
+      "loss": 0.0023,
+      "step": 807
+    },
+    {
+      "epoch": 2.6117171717171717,
+      "grad_norm": 0.005063756834715605,
+      "learning_rate": 9.999374615285236e-05,
+      "loss": 0.0018,
+      "step": 808
+    },
+    {
+      "epoch": 2.614949494949495,
+      "grad_norm": 0.007926912046968937,
+      "learning_rate": 9.999338369297106e-05,
+      "loss": 0.003,
+      "step": 809
+    },
+    {
+      "epoch": 2.618181818181818,
+      "grad_norm": 0.00994369387626648,
+      "learning_rate": 9.999301102386553e-05,
+      "loss": 0.0031,
+      "step": 810
+    },
+    {
+      "epoch": 2.6214141414141414,
+      "grad_norm": 0.004684642422944307,
+      "learning_rate": 9.999262814561185e-05,
+      "loss": 0.0017,
+      "step": 811
+    },
+    {
+      "epoch": 2.6246464646464647,
+      "grad_norm": 0.01245308667421341,
+      "learning_rate": 9.999223505828821e-05,
+      "loss": 0.0037,
+      "step": 812
+    },
+    {
+      "epoch": 2.627878787878788,
+      "grad_norm": 0.00802707765251398,
+      "learning_rate": 9.999183176197491e-05,
+      "loss": 0.0027,
+      "step": 813
+    },
+    {
+      "epoch": 2.631111111111111,
+      "grad_norm": 0.008055022917687893,
+      "learning_rate": 9.999141825675426e-05,
+      "loss": 0.0036,
+      "step": 814
+    },
+    {
+      "epoch": 2.6343434343434344,
+      "grad_norm": 0.005763540044426918,
+      "learning_rate": 9.999099454271074e-05,
+      "loss": 0.0016,
+      "step": 815
+    },
+    {
+      "epoch": 2.6375757575757577,
+      "grad_norm": 0.008480647578835487,
+      "learning_rate": 9.999056061993089e-05,
+      "loss": 0.0027,
+      "step": 816
+    },
+    {
+      "epoch": 2.640808080808081,
+      "grad_norm": 0.010444406419992447,
+      "learning_rate": 9.999011648850329e-05,
+      "loss": 0.0025,
+      "step": 817
+    },
+    {
+      "epoch": 2.644040404040404,
+      "grad_norm": 0.010159408673644066,
+      "learning_rate": 9.998966214851864e-05,
+      "loss": 0.0042,
+      "step": 818
+    },
+    {
+      "epoch": 2.6472727272727274,
+      "grad_norm": 0.009814529679715633,
+      "learning_rate": 9.998919760006972e-05,
+      "loss": 0.0036,
+      "step": 819
+    },
+    {
+      "epoch": 2.6505050505050507,
+      "grad_norm": 0.009227470494806767,
+      "learning_rate": 9.998872284325142e-05,
+      "loss": 0.0027,
+      "step": 820
+    },
+    {
+      "epoch": 2.653737373737374,
+      "grad_norm": 0.008259695954620838,
+      "learning_rate": 9.998823787816066e-05,
+      "loss": 0.003,
+      "step": 821
+    },
+    {
+      "epoch": 2.656969696969697,
+      "grad_norm": 0.010044812224805355,
+      "learning_rate": 9.99877427048965e-05,
+      "loss": 0.0023,
+      "step": 822
+    },
+    {
+      "epoch": 2.6602020202020205,
+      "grad_norm": 0.007520666811615229,
+      "learning_rate": 9.998723732356006e-05,
+      "loss": 0.0026,
+      "step": 823
+    },
+    {
+      "epoch": 2.6634343434343437,
+      "grad_norm": 0.012261404655873775,
+      "learning_rate": 9.998672173425452e-05,
+      "loss": 0.0053,
+      "step": 824
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.006036568898707628,
+      "learning_rate": 9.998619593708518e-05,
+      "loss": 0.0022,
+      "step": 825
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "eval_loss": 0.0035520135425031185,
+      "eval_runtime": 18.6008,
+      "eval_samples_per_second": 5.376,
+      "eval_steps_per_second": 1.344,
+      "step": 825
+    },
+    {
+      "epoch": 2.6698989898989898,
+      "grad_norm": 0.006456975359469652,
+      "learning_rate": 9.998565993215943e-05,
+      "loss": 0.0024,
+      "step": 826
+    },
+    {
+      "epoch": 2.673131313131313,
+      "grad_norm": 0.00519346771761775,
+      "learning_rate": 9.998511371958672e-05,
+      "loss": 0.0016,
+      "step": 827
+    },
+    {
+      "epoch": 2.6763636363636363,
+      "grad_norm": 0.007795286830514669,
+      "learning_rate": 9.998455729947858e-05,
+      "loss": 0.0036,
+      "step": 828
+    },
+    {
+      "epoch": 2.6795959595959595,
+      "grad_norm": 0.009498896077275276,
+      "learning_rate": 9.998399067194864e-05,
+      "loss": 0.0035,
+      "step": 829
+    },
+    {
+      "epoch": 2.682828282828283,
+      "grad_norm": 0.005344277247786522,
+      "learning_rate": 9.998341383711263e-05,
+      "loss": 0.0019,
+      "step": 830
+    },
+    {
+      "epoch": 2.686060606060606,
+      "grad_norm": 0.006172938738018274,
+      "learning_rate": 9.998282679508835e-05,
+      "loss": 0.0021,
+      "step": 831
+    },
+    {
+      "epoch": 2.6892929292929293,
+      "grad_norm": 0.006847708486020565,
+      "learning_rate": 9.998222954599563e-05,
+      "loss": 0.0018,
+      "step": 832
+    },
+    {
+      "epoch": 2.6925252525252525,
+      "grad_norm": 0.007936537265777588,
+      "learning_rate": 9.99816220899565e-05,
+      "loss": 0.0025,
+      "step": 833
+    },
+    {
+      "epoch": 2.695757575757576,
+      "grad_norm": 0.0077203791588544846,
+      "learning_rate": 9.998100442709497e-05,
+      "loss": 0.003,
+      "step": 834
+    },
+    {
+      "epoch": 2.698989898989899,
+      "grad_norm": 0.007416009437292814,
+      "learning_rate": 9.998037655753717e-05,
+      "loss": 0.0027,
+      "step": 835
+    },
+    {
+      "epoch": 2.7022222222222223,
+      "grad_norm": 0.006032218690961599,
+      "learning_rate": 9.997973848141137e-05,
+      "loss": 0.0019,
+      "step": 836
+    },
+    {
+      "epoch": 2.7054545454545456,
+      "grad_norm": 0.007939686998724937,
+      "learning_rate": 9.997909019884781e-05,
+      "loss": 0.0022,
+      "step": 837
+    },
+    {
+      "epoch": 2.708686868686869,
+      "grad_norm": 0.007059860974550247,
+      "learning_rate": 9.99784317099789e-05,
+      "loss": 0.003,
+      "step": 838
+    },
+    {
+      "epoch": 2.711919191919192,
+      "grad_norm": 0.01049652136862278,
+      "learning_rate": 9.997776301493914e-05,
+      "loss": 0.0034,
+      "step": 839
+    },
+    {
+      "epoch": 2.7151515151515153,
+      "grad_norm": 0.010508295148611069,
+      "learning_rate": 9.997708411386501e-05,
+      "loss": 0.0031,
+      "step": 840
+    },
+    {
+      "epoch": 2.718383838383838,
+      "grad_norm": 0.010989880189299583,
+      "learning_rate": 9.997639500689523e-05,
+      "loss": 0.004,
+      "step": 841
+    },
+    {
+      "epoch": 2.7216161616161614,
+      "grad_norm": 0.0070852129720151424,
+      "learning_rate": 9.997569569417049e-05,
+      "loss": 0.0022,
+      "step": 842
+    },
+    {
+      "epoch": 2.7248484848484846,
+      "grad_norm": 0.007061573676764965,
+      "learning_rate": 9.997498617583358e-05,
+      "loss": 0.0018,
+      "step": 843
+    },
+    {
+      "epoch": 2.728080808080808,
+      "grad_norm": 0.004920569248497486,
+      "learning_rate": 9.997426645202943e-05,
+      "loss": 0.0015,
+      "step": 844
+    },
+    {
+      "epoch": 2.731313131313131,
+      "grad_norm": 0.009148934856057167,
+      "learning_rate": 9.9973536522905e-05,
+      "loss": 0.0028,
+      "step": 845
+    },
+    {
+      "epoch": 2.7345454545454544,
+      "grad_norm": 0.006100498139858246,
+      "learning_rate": 9.997279638860933e-05,
+      "loss": 0.002,
+      "step": 846
+    },
+    {
+      "epoch": 2.7377777777777776,
+      "grad_norm": 0.010787500068545341,
+      "learning_rate": 9.99720460492936e-05,
+      "loss": 0.0022,
+      "step": 847
+    },
+    {
+      "epoch": 2.741010101010101,
+      "grad_norm": 0.007626068778336048,
+      "learning_rate": 9.997128550511099e-05,
+      "loss": 0.0019,
+      "step": 848
+    },
+    {
+      "epoch": 2.744242424242424,
+      "grad_norm": 0.007720778696238995,
+      "learning_rate": 9.997051475621687e-05,
+      "loss": 0.002,
+      "step": 849
+    },
+    {
+      "epoch": 2.7474747474747474,
+      "grad_norm": 0.010044184513390064,
+      "learning_rate": 9.996973380276857e-05,
+      "loss": 0.0045,
+      "step": 850
+    },
+    {
+      "epoch": 2.7474747474747474,
+      "eval_loss": 0.0034316859673708677,
+      "eval_runtime": 18.5217,
+      "eval_samples_per_second": 5.399,
+      "eval_steps_per_second": 1.35,
+      "step": 850
+    },
+    {
+      "epoch": 2.7507070707070707,
+      "grad_norm": 0.008457653224468231,
+      "learning_rate": 9.996894264492563e-05,
+      "loss": 0.0027,
+      "step": 851
+    },
+    {
+      "epoch": 2.753939393939394,
+      "grad_norm": 0.0105837257578969,
+      "learning_rate": 9.99681412828496e-05,
+      "loss": 0.0038,
+      "step": 852
+    },
+    {
+      "epoch": 2.757171717171717,
+      "grad_norm": 0.012494977563619614,
+      "learning_rate": 9.996732971670408e-05,
+      "loss": 0.0053,
+      "step": 853
+    },
+    {
+      "epoch": 2.7604040404040404,
+      "grad_norm": 0.006976769305765629,
+      "learning_rate": 9.996650794665487e-05,
+      "loss": 0.0018,
+      "step": 854
+    },
+    {
+      "epoch": 2.7636363636363637,
+      "grad_norm": 0.00825168564915657,
+      "learning_rate": 9.996567597286974e-05,
+      "loss": 0.0026,
+      "step": 855
+    },
+    {
+      "epoch": 2.766868686868687,
+      "grad_norm": 0.007835673168301582,
+      "learning_rate": 9.996483379551861e-05,
+      "loss": 0.0022,
+      "step": 856
+    },
+    {
+      "epoch": 2.77010101010101,
+      "grad_norm": 0.007407719735056162,
+      "learning_rate": 9.996398141477344e-05,
+      "loss": 0.0023,
+      "step": 857
+    },
+    {
+      "epoch": 2.7733333333333334,
+      "grad_norm": 0.008899042382836342,
+      "learning_rate": 9.996311883080832e-05,
+      "loss": 0.0032,
+      "step": 858
+    },
+    {
+      "epoch": 2.7765656565656567,
+      "grad_norm": 0.008769405074417591,
+      "learning_rate": 9.996224604379938e-05,
+      "loss": 0.0042,
+      "step": 859
+    },
+    {
+      "epoch": 2.77979797979798,
+      "grad_norm": 0.007881446741521358,
+      "learning_rate": 9.996136305392487e-05,
+      "loss": 0.0027,
+      "step": 860
+    },
+    {
+      "epoch": 2.783030303030303,
+      "grad_norm": 0.005930191837251186,
+      "learning_rate": 9.996046986136509e-05,
+      "loss": 0.0023,
+      "step": 861
+    },
+    {
+      "epoch": 2.7862626262626264,
+      "grad_norm": 0.007934244349598885,
+      "learning_rate": 9.995956646630246e-05,
+      "loss": 0.0027,
+      "step": 862
+    },
+    {
+      "epoch": 2.7894949494949497,
+      "grad_norm": 0.0088659031316638,
+      "learning_rate": 9.995865286892145e-05,
+      "loss": 0.0035,
+      "step": 863
+    },
+    {
+      "epoch": 2.792727272727273,
+      "grad_norm": 0.01819589175283909,
+      "learning_rate": 9.995772906940864e-05,
+      "loss": 0.0076,
+      "step": 864
+    },
+    {
+      "epoch": 2.795959595959596,
+      "grad_norm": 0.005735949147492647,
+      "learning_rate": 9.995679506795264e-05,
+      "loss": 0.0021,
+      "step": 865
+    },
+    {
+      "epoch": 2.7991919191919195,
+      "grad_norm": 0.008254305459558964,
+      "learning_rate": 9.995585086474424e-05,
+      "loss": 0.0029,
+      "step": 866
+    },
+    {
+      "epoch": 2.8024242424242423,
+      "grad_norm": 0.007419264875352383,
+      "learning_rate": 9.995489645997622e-05,
+      "loss": 0.0016,
+      "step": 867
+    },
+    {
+      "epoch": 2.8056565656565655,
+      "grad_norm": 0.013274376280605793,
+      "learning_rate": 9.99539318538435e-05,
+      "loss": 0.0042,
+      "step": 868
+    },
+    {
+      "epoch": 2.8088888888888888,
+      "grad_norm": 0.008460251614451408,
+      "learning_rate": 9.995295704654304e-05,
+      "loss": 0.0028,
+      "step": 869
+    },
+    {
+      "epoch": 2.812121212121212,
+      "grad_norm": 0.007807243149727583,
+      "learning_rate": 9.995197203827393e-05,
+      "loss": 0.0032,
+      "step": 870
+    },
+    {
+      "epoch": 2.8153535353535353,
+      "grad_norm": 0.008562683127820492,
+      "learning_rate": 9.995097682923733e-05,
+      "loss": 0.0019,
+      "step": 871
+    },
+    {
+      "epoch": 2.8185858585858585,
+      "grad_norm": 0.0074541945941746235,
+      "learning_rate": 9.994997141963644e-05,
+      "loss": 0.002,
+      "step": 872
+    },
+    {
+      "epoch": 2.821818181818182,
+      "grad_norm": 0.006698752287775278,
+      "learning_rate": 9.994895580967658e-05,
+      "loss": 0.0025,
+      "step": 873
+    },
+    {
+      "epoch": 2.825050505050505,
+      "grad_norm": 0.015189897269010544,
+      "learning_rate": 9.994792999956518e-05,
+      "loss": 0.0063,
+      "step": 874
+    },
+    {
+      "epoch": 2.8282828282828283,
+      "grad_norm": 0.009802022948861122,
+      "learning_rate": 9.994689398951169e-05,
+      "loss": 0.004,
+      "step": 875
+    },
+    {
+      "epoch": 2.8282828282828283,
+      "eval_loss": 0.003472257638350129,
+      "eval_runtime": 18.5556,
+      "eval_samples_per_second": 5.389,
+      "eval_steps_per_second": 1.347,
+      "step": 875
+    },
+    {
+      "epoch": 2.8315151515151515,
+      "grad_norm": 0.008691331371665001,
+      "learning_rate": 9.994584777972769e-05,
+      "loss": 0.0038,
+      "step": 876
+    },
+    {
+      "epoch": 2.834747474747475,
+      "grad_norm": 0.004085318651050329,
+      "learning_rate": 9.994479137042683e-05,
+      "loss": 0.0017,
+      "step": 877
+    },
+    {
+      "epoch": 2.837979797979798,
+      "grad_norm": 0.01081239152699709,
+      "learning_rate": 9.994372476182484e-05,
+      "loss": 0.0051,
+      "step": 878
+    },
+    {
+      "epoch": 2.8412121212121213,
+      "grad_norm": 0.00917616207152605,
+      "learning_rate": 9.994264795413953e-05,
+      "loss": 0.0027,
+      "step": 879
+    },
+    {
+      "epoch": 2.8444444444444446,
+      "grad_norm": 0.0047592418268322945,
+      "learning_rate": 9.99415609475908e-05,
+      "loss": 0.0014,
+      "step": 880
+    },
+    {
+      "epoch": 2.847676767676768,
+      "grad_norm": 0.009527614340186119,
+      "learning_rate": 9.994046374240062e-05,
+      "loss": 0.0032,
+      "step": 881
+    },
+    {
+      "epoch": 2.850909090909091,
+      "grad_norm": 0.007115239277482033,
+      "learning_rate": 9.993935633879306e-05,
+      "loss": 0.0027,
+      "step": 882
+    },
+    {
+      "epoch": 2.854141414141414,
+      "grad_norm": 0.008802005089819431,
+      "learning_rate": 9.993823873699426e-05,
+      "loss": 0.0026,
+      "step": 883
+    },
+    {
+      "epoch": 2.857373737373737,
+      "grad_norm": 0.008757531642913818,
+      "learning_rate": 9.993711093723245e-05,
+      "loss": 0.0036,
+      "step": 884
+    },
+    {
+      "epoch": 2.8606060606060604,
+      "grad_norm": 0.0062814452685415745,
+      "learning_rate": 9.993597293973796e-05,
+      "loss": 0.0026,
+      "step": 885
+    },
+    {
+      "epoch": 2.8638383838383836,
+      "grad_norm": 0.007916122674942017,
+      "learning_rate": 9.993482474474314e-05,
+      "loss": 0.0026,
+      "step": 886
+    },
+    {
+      "epoch": 2.867070707070707,
+      "grad_norm": 0.012285426259040833,
+      "learning_rate": 9.99336663524825e-05,
+      "loss": 0.0048,
+      "step": 887
+    },
+    {
+      "epoch": 2.87030303030303,
+      "grad_norm": 0.007117332424968481,
+      "learning_rate": 9.993249776319258e-05,
+      "loss": 0.0014,
+      "step": 888
+    },
+    {
+      "epoch": 2.8735353535353534,
+      "grad_norm": 0.007871873676776886,
+      "learning_rate": 9.993131897711202e-05,
+      "loss": 0.0029,
+      "step": 889
+    },
+    {
+      "epoch": 2.8767676767676766,
+      "grad_norm": 0.009144186042249203,
+      "learning_rate": 9.993012999448154e-05,
+      "loss": 0.003,
+      "step": 890
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.006271008867770433,
+      "learning_rate": 9.992893081554397e-05,
+      "loss": 0.0026,
+      "step": 891
+    },
+    {
+      "epoch": 2.883232323232323,
+      "grad_norm": 0.004877634346485138,
+      "learning_rate": 9.992772144054415e-05,
+      "loss": 0.0017,
+      "step": 892
+    },
+    {
+      "epoch": 2.8864646464646464,
+      "grad_norm": 0.01159873977303505,
+      "learning_rate": 9.992650186972909e-05,
+      "loss": 0.0025,
+      "step": 893
+    },
+    {
+      "epoch": 2.8896969696969697,
+      "grad_norm": 0.006861403118818998,
+      "learning_rate": 9.99252721033478e-05,
+      "loss": 0.0022,
+      "step": 894
+    },
+    {
+      "epoch": 2.892929292929293,
+      "grad_norm": 0.009415537118911743,
+      "learning_rate": 9.992403214165147e-05,
+      "loss": 0.0036,
+      "step": 895
+    },
+    {
+      "epoch": 2.896161616161616,
+      "grad_norm": 0.007355588488280773,
+      "learning_rate": 9.992278198489327e-05,
+      "loss": 0.0035,
+      "step": 896
+    },
+    {
+      "epoch": 2.8993939393939394,
+      "grad_norm": 0.003762519918382168,
+      "learning_rate": 9.99215216333285e-05,
+      "loss": 0.0013,
+      "step": 897
+    },
+    {
+      "epoch": 2.9026262626262627,
+      "grad_norm": 0.007687287870794535,
+      "learning_rate": 9.992025108721454e-05,
+      "loss": 0.0023,
+      "step": 898
+    },
+    {
+      "epoch": 2.905858585858586,
+      "grad_norm": 0.006643611006438732,
+      "learning_rate": 9.991897034681087e-05,
+      "loss": 0.0024,
+      "step": 899
+    },
+    {
+      "epoch": 2.909090909090909,
+      "grad_norm": 0.006356996949762106,
+      "learning_rate": 9.9917679412379e-05,
+      "loss": 0.0026,
+      "step": 900
+    },
+    {
+      "epoch": 2.909090909090909,
+      "eval_loss": 0.0033850145991891623,
+      "eval_runtime": 18.5793,
+      "eval_samples_per_second": 5.382,
+      "eval_steps_per_second": 1.346,
+      "step": 900
+    },
+    {
+      "epoch": 2.9123232323232324,
+      "grad_norm": 0.004237829241901636,
+      "learning_rate": 9.99163782841826e-05,
+      "loss": 0.0018,
+      "step": 901
+    },
+    {
+      "epoch": 2.9155555555555557,
+      "grad_norm": 0.006980978883802891,
+      "learning_rate": 9.991506696248731e-05,
+      "loss": 0.0022,
+      "step": 902
+    },
+    {
+      "epoch": 2.918787878787879,
+      "grad_norm": 0.007011194713413715,
+      "learning_rate": 9.991374544756098e-05,
+      "loss": 0.0034,
+      "step": 903
+    },
+    {
+      "epoch": 2.922020202020202,
+      "grad_norm": 0.006265897769480944,
+      "learning_rate": 9.991241373967344e-05,
+      "loss": 0.0023,
+      "step": 904
+    },
+    {
+      "epoch": 2.9252525252525254,
+      "grad_norm": 0.006889102049171925,
+      "learning_rate": 9.991107183909664e-05,
+      "loss": 0.0021,
+      "step": 905
+    },
+    {
+      "epoch": 2.9284848484848487,
+      "grad_norm": 0.00801191944628954,
+      "learning_rate": 9.990971974610466e-05,
+      "loss": 0.0028,
+      "step": 906
+    },
+    {
+      "epoch": 2.931717171717172,
+      "grad_norm": 0.006524492986500263,
+      "learning_rate": 9.990835746097356e-05,
+      "loss": 0.0018,
+      "step": 907
+    },
+    {
+      "epoch": 2.934949494949495,
+      "grad_norm": 0.007707702461630106,
+      "learning_rate": 9.990698498398155e-05,
+      "loss": 0.003,
+      "step": 908
+    },
+    {
+      "epoch": 2.9381818181818184,
+      "grad_norm": 0.006807916797697544,
+      "learning_rate": 9.990560231540889e-05,
+      "loss": 0.0019,
+      "step": 909
+    },
+    {
+      "epoch": 2.9414141414141413,
+      "grad_norm": 0.01063486561179161,
+      "learning_rate": 9.990420945553797e-05,
+      "loss": 0.0041,
+      "step": 910
+    },
+    {
+      "epoch": 2.9446464646464645,
+      "grad_norm": 0.007649898063391447,
+      "learning_rate": 9.990280640465321e-05,
+      "loss": 0.0021,
+      "step": 911
+    },
+    {
+      "epoch": 2.9478787878787878,
+      "grad_norm": 0.0075135622173547745,
+      "learning_rate": 9.990139316304112e-05,
+      "loss": 0.0018,
+      "step": 912
+    },
+    {
+      "epoch": 2.951111111111111,
+      "grad_norm": 0.008089892566204071,
+      "learning_rate": 9.989996973099032e-05,
+      "loss": 0.0018,
+      "step": 913
+    },
+    {
+      "epoch": 2.9543434343434343,
+      "grad_norm": 0.006691002286970615,
+      "learning_rate": 9.989853610879147e-05,
+      "loss": 0.0018,
+      "step": 914
+    },
+    {
+      "epoch": 2.9575757575757575,
+      "grad_norm": 0.008318419568240643,
+      "learning_rate": 9.989709229673736e-05,
+      "loss": 0.0012,
+      "step": 915
+    },
+    {
+      "epoch": 2.9608080808080808,
+      "grad_norm": 0.009121793322265148,
+      "learning_rate": 9.98956382951228e-05,
+      "loss": 0.003,
+      "step": 916
+    },
+    {
+      "epoch": 2.964040404040404,
+      "grad_norm": 0.008950279094278812,
+      "learning_rate": 9.989417410424475e-05,
+      "loss": 0.003,
+      "step": 917
+    },
+    {
+      "epoch": 2.9672727272727273,
+      "grad_norm": 0.014811480417847633,
+      "learning_rate": 9.98926997244022e-05,
+      "loss": 0.0041,
+      "step": 918
+    },
+    {
+      "epoch": 2.9705050505050505,
+      "grad_norm": 0.007069775369018316,
+      "learning_rate": 9.989121515589622e-05,
+      "loss": 0.003,
+      "step": 919
+    },
+    {
+      "epoch": 2.973737373737374,
+      "grad_norm": 0.0060707684606313705,
+      "learning_rate": 9.988972039902997e-05,
+      "loss": 0.0019,
+      "step": 920
+    },
+    {
+      "epoch": 2.976969696969697,
+      "grad_norm": 0.005915610119700432,
+      "learning_rate": 9.988821545410874e-05,
+      "loss": 0.002,
+      "step": 921
+    },
+    {
+      "epoch": 2.9802020202020203,
+      "grad_norm": 0.008888405747711658,
+      "learning_rate": 9.988670032143981e-05,
+      "loss": 0.0033,
+      "step": 922
+    },
+    {
+      "epoch": 2.9834343434343435,
+      "grad_norm": 0.0058620586059987545,
+      "learning_rate": 9.988517500133262e-05,
+      "loss": 0.0019,
+      "step": 923
+    },
+    {
+      "epoch": 2.986666666666667,
+      "grad_norm": 0.005796400364488363,
+      "learning_rate": 9.988363949409865e-05,
+      "loss": 0.0021,
+      "step": 924
+    },
+    {
+      "epoch": 2.98989898989899,
+      "grad_norm": 0.0063505168072879314,
+      "learning_rate": 9.988209380005144e-05,
+      "loss": 0.0019,
+      "step": 925
+    },
+    {
+      "epoch": 2.98989898989899,
+      "eval_loss": 0.003320756834000349,
+      "eval_runtime": 18.5511,
+      "eval_samples_per_second": 5.391,
+      "eval_steps_per_second": 1.348,
+      "step": 925
+    },
+    {
+      "epoch": 2.993131313131313,
+      "grad_norm": 0.008493457920849323,
+      "learning_rate": 9.98805379195067e-05,
+      "loss": 0.0033,
+      "step": 926
+    },
+    {
+      "epoch": 2.996363636363636,
+      "grad_norm": 0.009364234283566475,
+      "learning_rate": 9.987897185278208e-05,
+      "loss": 0.0048,
+      "step": 927
+    },
+    {
+      "epoch": 2.9995959595959594,
+      "grad_norm": 0.009802762418985367,
+      "learning_rate": 9.987739560019746e-05,
+      "loss": 0.0028,
+      "step": 928
+    },
+    {
+      "epoch": 3.0028282828282826,
+      "grad_norm": 0.018954146653413773,
+      "learning_rate": 9.987580916207468e-05,
+      "loss": 0.0086,
+      "step": 929
+    },
+    {
+      "epoch": 3.006060606060606,
+      "grad_norm": 0.006144994404166937,
+      "learning_rate": 9.987421253873775e-05,
+      "loss": 0.0018,
+      "step": 930
+    },
+    {
+      "epoch": 3.009292929292929,
+      "grad_norm": 0.004907108843326569,
+      "learning_rate": 9.987260573051269e-05,
+      "loss": 0.0015,
+      "step": 931
+    },
+    {
+      "epoch": 3.0125252525252524,
+      "grad_norm": 0.006426096893846989,
+      "learning_rate": 9.987098873772763e-05,
+      "loss": 0.0023,
+      "step": 932
+    },
+    {
+      "epoch": 3.0157575757575756,
+      "grad_norm": 0.006343123037368059,
+      "learning_rate": 9.986936156071278e-05,
+      "loss": 0.0018,
+      "step": 933
+    },
+    {
+      "epoch": 3.018989898989899,
+      "grad_norm": 0.010373006574809551,
+      "learning_rate": 9.986772419980044e-05,
+      "loss": 0.0034,
+      "step": 934
+    },
+    {
+      "epoch": 3.022222222222222,
+      "grad_norm": 0.004722005221992731,
+      "learning_rate": 9.986607665532497e-05,
+      "loss": 0.0012,
+      "step": 935
+    },
+    {
+      "epoch": 3.0254545454545454,
+      "grad_norm": 0.007018386386334896,
+      "learning_rate": 9.986441892762281e-05,
+      "loss": 0.0019,
+      "step": 936
+    },
+    {
+      "epoch": 3.0286868686868686,
+      "grad_norm": 0.006261017639189959,
+      "learning_rate": 9.98627510170325e-05,
+      "loss": 0.0014,
+      "step": 937
+    },
+    {
+      "epoch": 3.031919191919192,
+      "grad_norm": 0.008190566673874855,
+      "learning_rate": 9.986107292389464e-05,
+      "loss": 0.0017,
+      "step": 938
+    },
+    {
+      "epoch": 3.035151515151515,
+      "grad_norm": 0.006314157508313656,
+      "learning_rate": 9.985938464855191e-05,
+      "loss": 0.0015,
+      "step": 939
+    },
+    {
+      "epoch": 3.0383838383838384,
+      "grad_norm": 0.008036741055548191,
+      "learning_rate": 9.985768619134909e-05,
+      "loss": 0.0021,
+      "step": 940
+    },
+    {
+      "epoch": 3.0416161616161617,
+      "grad_norm": 0.005572419613599777,
+      "learning_rate": 9.985597755263302e-05,
+      "loss": 0.0019,
+      "step": 941
+    },
+    {
+      "epoch": 3.044848484848485,
+      "grad_norm": 0.007106456905603409,
+      "learning_rate": 9.985425873275263e-05,
+      "loss": 0.0017,
+      "step": 942
+    },
+    {
+      "epoch": 3.048080808080808,
+      "grad_norm": 0.006766545120626688,
+      "learning_rate": 9.98525297320589e-05,
+      "loss": 0.0016,
+      "step": 943
+    },
+    {
+      "epoch": 3.0513131313131314,
+      "grad_norm": 0.00921205710619688,
+      "learning_rate": 9.985079055090493e-05,
+      "loss": 0.0012,
+      "step": 944
+    },
+    {
+      "epoch": 3.0545454545454547,
+      "grad_norm": 0.00547370221465826,
+      "learning_rate": 9.984904118964588e-05,
+      "loss": 0.0017,
+      "step": 945
+    },
+    {
+      "epoch": 3.057777777777778,
+      "grad_norm": 0.007727344986051321,
+      "learning_rate": 9.984728164863898e-05,
+      "loss": 0.0016,
+      "step": 946
+    },
+    {
+      "epoch": 3.061010101010101,
+      "grad_norm": 0.006054291035979986,
+      "learning_rate": 9.984551192824355e-05,
+      "loss": 0.0017,
+      "step": 947
+    },
+    {
+      "epoch": 3.0642424242424244,
+      "grad_norm": 0.00898781418800354,
+      "learning_rate": 9.9843732028821e-05,
+      "loss": 0.0022,
+      "step": 948
+    },
+    {
+      "epoch": 3.0674747474747477,
+      "grad_norm": 0.0060675074346363544,
+      "learning_rate": 9.98419419507348e-05,
+      "loss": 0.0013,
+      "step": 949
+    },
+    {
+      "epoch": 3.0707070707070705,
+      "grad_norm": 0.007023282814770937,
+      "learning_rate": 9.98401416943505e-05,
+      "loss": 0.0015,
+      "step": 950
+    },
+    {
+      "epoch": 3.0707070707070705,
+      "eval_loss": 0.003585340455174446,
+      "eval_runtime": 18.5936,
+      "eval_samples_per_second": 5.378,
+      "eval_steps_per_second": 1.345,
+      "step": 950
+    },
+    {
+      "epoch": 3.0739393939393937,
+      "grad_norm": 0.007739969063550234,
+      "learning_rate": 9.983833126003572e-05,
+      "loss": 0.0022,
+      "step": 951
+    },
+    {
+      "epoch": 3.077171717171717,
+      "grad_norm": 0.008393503725528717,
+      "learning_rate": 9.98365106481602e-05,
+      "loss": 0.0019,
+      "step": 952
+    },
+    {
+      "epoch": 3.0804040404040403,
+      "grad_norm": 0.006369095295667648,
+      "learning_rate": 9.983467985909573e-05,
+      "loss": 0.0011,
+      "step": 953
+    },
+    {
+      "epoch": 3.0836363636363635,
+      "grad_norm": 0.008163919672369957,
+      "learning_rate": 9.983283889321615e-05,
+      "loss": 0.0018,
+      "step": 954
+    },
+    {
+      "epoch": 3.0868686868686868,
+      "grad_norm": 0.018781553953886032,
+      "learning_rate": 9.983098775089742e-05,
+      "loss": 0.0038,
+      "step": 955
+    },
+    {
+      "epoch": 3.09010101010101,
+      "grad_norm": 0.012334289960563183,
+      "learning_rate": 9.982912643251757e-05,
+      "loss": 0.004,
+      "step": 956
+    },
+    {
+      "epoch": 3.0933333333333333,
+      "grad_norm": 0.005684965290129185,
+      "learning_rate": 9.98272549384567e-05,
+      "loss": 0.0016,
+      "step": 957
+    },
+    {
+      "epoch": 3.0965656565656565,
+      "grad_norm": 0.00594430323690176,
+      "learning_rate": 9.982537326909697e-05,
+      "loss": 0.0015,
+      "step": 958
+    },
+    {
+      "epoch": 3.0997979797979798,
+      "grad_norm": 0.006609255447983742,
+      "learning_rate": 9.982348142482269e-05,
+      "loss": 0.0016,
+      "step": 959
+    },
+    {
+      "epoch": 3.103030303030303,
+      "grad_norm": 0.010460656136274338,
+      "learning_rate": 9.982157940602014e-05,
+      "loss": 0.0039,
+      "step": 960
+    },
+    {
+      "epoch": 3.1062626262626263,
+      "grad_norm": 0.006908283568918705,
+      "learning_rate": 9.981966721307778e-05,
+      "loss": 0.0025,
+      "step": 961
+    },
+    {
+      "epoch": 3.1094949494949495,
+      "grad_norm": 0.006505319382995367,
+      "learning_rate": 9.981774484638606e-05,
+      "loss": 0.0018,
+      "step": 962
+    },
+    {
+      "epoch": 3.112727272727273,
+      "grad_norm": 0.005758680868893862,
+      "learning_rate": 9.981581230633758e-05,
+      "loss": 0.0014,
+      "step": 963
+    },
+    {
+      "epoch": 3.115959595959596,
+      "grad_norm": 0.004635549616068602,
+      "learning_rate": 9.981386959332697e-05,
+      "loss": 0.0013,
+      "step": 964
+    },
+    {
+      "epoch": 3.1191919191919193,
+      "grad_norm": 0.010312603786587715,
+      "learning_rate": 9.981191670775097e-05,
+      "loss": 0.0029,
+      "step": 965
+    },
+    {
+      "epoch": 3.1224242424242425,
+      "grad_norm": 0.006679642479866743,
+      "learning_rate": 9.980995365000836e-05,
+      "loss": 0.0022,
+      "step": 966
+    },
+    {
+      "epoch": 3.125656565656566,
+      "grad_norm": 0.009541639126837254,
+      "learning_rate": 9.980798042050004e-05,
+      "loss": 0.0024,
+      "step": 967
+    },
+    {
+      "epoch": 3.128888888888889,
+      "grad_norm": 0.006234496366232634,
+      "learning_rate": 9.980599701962896e-05,
+      "loss": 0.0018,
+      "step": 968
+    },
+    {
+      "epoch": 3.1321212121212123,
+      "grad_norm": 0.007063133176416159,
+      "learning_rate": 9.980400344780015e-05,
+      "loss": 0.0023,
+      "step": 969
+    },
+    {
+      "epoch": 3.1353535353535356,
+      "grad_norm": 0.009879599325358868,
+      "learning_rate": 9.98019997054207e-05,
+      "loss": 0.0037,
+      "step": 970
+    },
+    {
+      "epoch": 3.1385858585858584,
+      "grad_norm": 0.006942450068891048,
+      "learning_rate": 9.979998579289984e-05,
+      "loss": 0.0018,
+      "step": 971
+    },
+    {
+      "epoch": 3.1418181818181816,
+      "grad_norm": 0.007005956023931503,
+      "learning_rate": 9.979796171064881e-05,
+      "loss": 0.0015,
+      "step": 972
+    },
+    {
+      "epoch": 3.145050505050505,
+      "grad_norm": 0.009488895535469055,
+      "learning_rate": 9.979592745908095e-05,
+      "loss": 0.0022,
+      "step": 973
+    },
+    {
+      "epoch": 3.148282828282828,
+      "grad_norm": 0.00824445765465498,
+      "learning_rate": 9.979388303861169e-05,
+      "loss": 0.0016,
+      "step": 974
+    },
+    {
+      "epoch": 3.1515151515151514,
+      "grad_norm": 0.006121410988271236,
+      "learning_rate": 9.97918284496585e-05,
+      "loss": 0.0018,
+      "step": 975
+    },
+    {
+      "epoch": 3.1515151515151514,
+      "eval_loss": 0.0034363586455583572,
+      "eval_runtime": 18.5459,
+      "eval_samples_per_second": 5.392,
+      "eval_steps_per_second": 1.348,
+      "step": 975
+    },
+    {
+      "epoch": 3.1547474747474746,
+      "grad_norm": 0.003954809159040451,
+      "learning_rate": 9.978976369264098e-05,
+      "loss": 0.0008,
+      "step": 976
+    },
+    {
+      "epoch": 3.157979797979798,
+      "grad_norm": 0.0071476600132882595,
+      "learning_rate": 9.978768876798075e-05,
+      "loss": 0.0019,
+      "step": 977
+    },
+    {
+      "epoch": 3.161212121212121,
+      "grad_norm": 0.006608667783439159,
+      "learning_rate": 9.978560367610156e-05,
+      "loss": 0.0015,
+      "step": 978
+    },
+    {
+      "epoch": 3.1644444444444444,
+      "grad_norm": 0.010076618753373623,
+      "learning_rate": 9.978350841742919e-05,
+      "loss": 0.0031,
+      "step": 979
+    },
+    {
+      "epoch": 3.1676767676767676,
+      "grad_norm": 0.008147617802023888,
+      "learning_rate": 9.978140299239152e-05,
+      "loss": 0.0023,
+      "step": 980
+    },
+    {
+      "epoch": 3.170909090909091,
+      "grad_norm": 0.006681856233626604,
+      "learning_rate": 9.977928740141851e-05,
+      "loss": 0.0021,
+      "step": 981
+    },
+    {
+      "epoch": 3.174141414141414,
+      "grad_norm": 0.007071407046169043,
+      "learning_rate": 9.977716164494217e-05,
+      "loss": 0.0018,
+      "step": 982
+    },
+    {
+      "epoch": 3.1773737373737374,
+      "grad_norm": 0.009117268025875092,
+      "learning_rate": 9.977502572339664e-05,
+      "loss": 0.0023,
+      "step": 983
+    },
+    {
+      "epoch": 3.1806060606060607,
+      "grad_norm": 0.007354568224400282,
+      "learning_rate": 9.977287963721804e-05,
+      "loss": 0.002,
+      "step": 984
+    },
+    {
+      "epoch": 3.183838383838384,
+      "grad_norm": 0.003036913461983204,
+      "learning_rate": 9.977072338684469e-05,
+      "loss": 0.0007,
+      "step": 985
+    },
+    {
+      "epoch": 3.187070707070707,
+      "grad_norm": 0.011151008307933807,
+      "learning_rate": 9.976855697271689e-05,
+      "loss": 0.0026,
+      "step": 986
+    },
+    {
+      "epoch": 3.1903030303030304,
+      "grad_norm": 0.008813554421067238,
+      "learning_rate": 9.976638039527704e-05,
+      "loss": 0.0023,
+      "step": 987
+    },
+    {
+      "epoch": 3.1935353535353537,
+      "grad_norm": 0.005291069392114878,
+      "learning_rate": 9.976419365496963e-05,
+      "loss": 0.0013,
+      "step": 988
+    },
+    {
+      "epoch": 3.196767676767677,
+      "grad_norm": 0.007516716606914997,
+      "learning_rate": 9.976199675224123e-05,
+      "loss": 0.0021,
+      "step": 989
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.0058667948469519615,
+      "learning_rate": 9.975978968754045e-05,
+      "loss": 0.0014,
+      "step": 990
+    },
+    {
+      "epoch": 3.2032323232323234,
+      "grad_norm": 0.011559627018868923,
+      "learning_rate": 9.975757246131803e-05,
+      "loss": 0.0027,
+      "step": 991
+    },
+    {
+      "epoch": 3.2064646464646467,
+      "grad_norm": 0.008682847954332829,
+      "learning_rate": 9.975534507402671e-05,
+      "loss": 0.0025,
+      "step": 992
+    },
+    {
+      "epoch": 3.2096969696969695,
+      "grad_norm": 0.009858380071818829,
+      "learning_rate": 9.975310752612137e-05,
+      "loss": 0.0035,
+      "step": 993
+    },
+    {
+      "epoch": 3.2129292929292927,
+      "grad_norm": 0.00856295507401228,
+      "learning_rate": 9.975085981805897e-05,
+      "loss": 0.0021,
+      "step": 994
+    },
+    {
+      "epoch": 3.216161616161616,
+      "grad_norm": 0.0047248350456357,
+      "learning_rate": 9.974860195029847e-05,
+      "loss": 0.0013,
+      "step": 995
+    },
+    {
+      "epoch": 3.2193939393939393,
+      "grad_norm": 0.008117076009511948,
+      "learning_rate": 9.974633392330097e-05,
+      "loss": 0.0019,
+      "step": 996
+    },
+    {
+      "epoch": 3.2226262626262625,
+      "grad_norm": 0.008703903295099735,
+      "learning_rate": 9.974405573752965e-05,
+      "loss": 0.0027,
+      "step": 997
+    },
+    {
+      "epoch": 3.2258585858585858,
+      "grad_norm": 0.011345437727868557,
+      "learning_rate": 9.974176739344971e-05,
+      "loss": 0.0033,
+      "step": 998
+    },
+    {
+      "epoch": 3.229090909090909,
+      "grad_norm": 0.006451399996876717,
+      "learning_rate": 9.973946889152847e-05,
+      "loss": 0.0015,
+      "step": 999
+    },
+    {
+      "epoch": 3.2323232323232323,
+      "grad_norm": 0.004938072524964809,
+      "learning_rate": 9.973716023223531e-05,
+      "loss": 0.0013,
+      "step": 1000
+    },
+    {
+      "epoch": 3.2323232323232323,
+      "eval_loss": 0.003583682468160987,
+      "eval_runtime": 18.5886,
+      "eval_samples_per_second": 5.38,
+      "eval_steps_per_second": 1.345,
+      "step": 1000
+    },
+    {
+      "epoch": 3.2355555555555555,
+      "grad_norm": 0.008080260828137398,
+      "learning_rate": 9.97348414160417e-05,
+      "loss": 0.0021,
+      "step": 1001
+    },
+    {
+      "epoch": 3.2387878787878788,
+      "grad_norm": 0.00879309605807066,
+      "learning_rate": 9.973251244342114e-05,
+      "loss": 0.0022,
+      "step": 1002
+    },
+    {
+      "epoch": 3.242020202020202,
+      "grad_norm": 0.008178070187568665,
+      "learning_rate": 9.973017331484926e-05,
+      "loss": 0.0021,
+      "step": 1003
+    },
+    {
+      "epoch": 3.2452525252525253,
+      "grad_norm": 0.004717908333986998,
+      "learning_rate": 9.972782403080372e-05,
+      "loss": 0.0014,
+      "step": 1004
+    },
+    {
+      "epoch": 3.2484848484848485,
+      "grad_norm": 0.006190045271068811,
+      "learning_rate": 9.972546459176425e-05,
+      "loss": 0.0017,
+      "step": 1005
+    },
+    {
+      "epoch": 3.251717171717172,
+      "grad_norm": 0.010530155152082443,
+      "learning_rate": 9.972309499821273e-05,
+      "loss": 0.0034,
+      "step": 1006
+    },
+    {
+      "epoch": 3.254949494949495,
+      "grad_norm": 0.00893255416303873,
+      "learning_rate": 9.972071525063303e-05,
+      "loss": 0.0025,
+      "step": 1007
+    },
+    {
+      "epoch": 3.2581818181818183,
+      "grad_norm": 0.007108526304364204,
+      "learning_rate": 9.971832534951108e-05,
+      "loss": 0.0014,
+      "step": 1008
+    },
+    {
+      "epoch": 3.2614141414141415,
+      "grad_norm": 0.00837770476937294,
+      "learning_rate": 9.9715925295335e-05,
+      "loss": 0.0025,
+      "step": 1009
+    },
+    {
+      "epoch": 3.264646464646465,
+      "grad_norm": 0.0038234281819313765,
+      "learning_rate": 9.971351508859488e-05,
+      "loss": 0.0012,
+      "step": 1010
+    },
+    {
+      "epoch": 3.267878787878788,
+      "grad_norm": 0.00693462323397398,
+      "learning_rate": 9.971109472978288e-05,
+      "loss": 0.0022,
+      "step": 1011
+    },
+    {
+      "epoch": 3.2711111111111113,
+      "grad_norm": 0.006061007268726826,
+      "learning_rate": 9.97086642193933e-05,
+      "loss": 0.0017,
+      "step": 1012
+    },
+    {
+      "epoch": 3.274343434343434,
+      "grad_norm": 0.0076158661395311356,
+      "learning_rate": 9.970622355792247e-05,
+      "loss": 0.0019,
+      "step": 1013
+    },
+    {
+      "epoch": 3.2775757575757574,
+      "grad_norm": 0.003440310014411807,
+      "learning_rate": 9.970377274586879e-05,
+      "loss": 0.0008,
+      "step": 1014
+    },
+    {
+      "epoch": 3.2808080808080806,
+      "grad_norm": 0.006996039301156998,
+      "learning_rate": 9.970131178373277e-05,
+      "loss": 0.0015,
+      "step": 1015
+    },
+    {
+      "epoch": 3.284040404040404,
+      "grad_norm": 0.006500132381916046,
+      "learning_rate": 9.969884067201695e-05,
+      "loss": 0.002,
+      "step": 1016
+    },
+    {
+      "epoch": 3.287272727272727,
+      "grad_norm": 0.009818831458687782,
+      "learning_rate": 9.969635941122595e-05,
+      "loss": 0.0021,
+      "step": 1017
+    },
+    {
+      "epoch": 3.2905050505050504,
+      "grad_norm": 0.005738329142332077,
+      "learning_rate": 9.969386800186649e-05,
+      "loss": 0.0015,
+      "step": 1018
+    },
+    {
+      "epoch": 3.2937373737373736,
+      "grad_norm": 0.005722105037420988,
+      "learning_rate": 9.969136644444731e-05,
+      "loss": 0.0014,
+      "step": 1019
+    },
+    {
+      "epoch": 3.296969696969697,
+      "grad_norm": 0.005793727934360504,
+      "learning_rate": 9.968885473947932e-05,
+      "loss": 0.0016,
+      "step": 1020
+    },
+    {
+      "epoch": 3.30020202020202,
+      "grad_norm": 0.004094158299267292,
+      "learning_rate": 9.968633288747539e-05,
+      "loss": 0.001,
+      "step": 1021
+    },
+    {
+      "epoch": 3.3034343434343434,
+      "grad_norm": 0.011295527219772339,
+      "learning_rate": 9.968380088895052e-05,
+      "loss": 0.0015,
+      "step": 1022
+    },
+    {
+      "epoch": 3.3066666666666666,
+      "grad_norm": 0.006014000624418259,
+      "learning_rate": 9.968125874442179e-05,
+      "loss": 0.002,
+      "step": 1023
+    },
+    {
+      "epoch": 3.30989898989899,
+      "grad_norm": 0.010188030079007149,
+      "learning_rate": 9.96787064544083e-05,
+      "loss": 0.0022,
+      "step": 1024
+    },
+    {
+      "epoch": 3.313131313131313,
+      "grad_norm": 0.00868601817637682,
+      "learning_rate": 9.96761440194313e-05,
+      "loss": 0.0019,
+      "step": 1025
+    },
+    {
+      "epoch": 3.313131313131313,
+      "eval_loss": 0.0033993881661444902,
+      "eval_runtime": 18.5483,
+      "eval_samples_per_second": 5.391,
+      "eval_steps_per_second": 1.348,
+      "step": 1025
+    },
+    {
+      "epoch": 3.3163636363636364,
+      "grad_norm": 0.007088824175298214,
+      "learning_rate": 9.967357144001403e-05,
+      "loss": 0.002,
+      "step": 1026
+    },
+    {
+      "epoch": 3.3195959595959597,
+      "grad_norm": 0.007010690867900848,
+      "learning_rate": 9.967098871668186e-05,
+      "loss": 0.0017,
+      "step": 1027
+    },
+    {
+      "epoch": 3.322828282828283,
+      "grad_norm": 0.005562958773225546,
+      "learning_rate": 9.966839584996222e-05,
+      "loss": 0.0016,
+      "step": 1028
+    },
+    {
+      "epoch": 3.326060606060606,
+      "grad_norm": 0.0047673918306827545,
+      "learning_rate": 9.96657928403846e-05,
+      "loss": 0.0013,
+      "step": 1029
+    },
+    {
+      "epoch": 3.3292929292929294,
+      "grad_norm": 0.007019753102213144,
+      "learning_rate": 9.966317968848054e-05,
+      "loss": 0.0019,
+      "step": 1030
+    },
+    {
+      "epoch": 3.3325252525252527,
+      "grad_norm": 0.006734140682965517,
+      "learning_rate": 9.966055639478369e-05,
+      "loss": 0.0021,
+      "step": 1031
+    },
+    {
+      "epoch": 3.335757575757576,
+      "grad_norm": 0.008648472838103771,
+      "learning_rate": 9.965792295982978e-05,
+      "loss": 0.0033,
+      "step": 1032
+    },
+    {
+      "epoch": 3.338989898989899,
+      "grad_norm": 0.004571566358208656,
+      "learning_rate": 9.965527938415655e-05,
+      "loss": 0.0014,
+      "step": 1033
+    },
+    {
+      "epoch": 3.3422222222222224,
+      "grad_norm": 0.007610399276018143,
+      "learning_rate": 9.965262566830388e-05,
+      "loss": 0.0026,
+      "step": 1034
+    },
+    {
+      "epoch": 3.3454545454545457,
+      "grad_norm": 0.006182728800922632,
+      "learning_rate": 9.964996181281367e-05,
+      "loss": 0.0022,
+      "step": 1035
+    },
+    {
+      "epoch": 3.348686868686869,
+      "grad_norm": 0.006889553740620613,
+      "learning_rate": 9.964728781822992e-05,
+      "loss": 0.0018,
+      "step": 1036
+    },
+    {
+      "epoch": 3.3519191919191917,
+      "grad_norm": 0.005316969007253647,
+      "learning_rate": 9.964460368509867e-05,
+      "loss": 0.0013,
+      "step": 1037
+    },
+    {
+      "epoch": 3.355151515151515,
+      "grad_norm": 0.014395351521670818,
+      "learning_rate": 9.964190941396808e-05,
+      "loss": 0.0057,
+      "step": 1038
+    },
+    {
+      "epoch": 3.3583838383838382,
+      "grad_norm": 0.007097650784999132,
+      "learning_rate": 9.963920500538834e-05,
+      "loss": 0.0015,
+      "step": 1039
+    },
+    {
+      "epoch": 3.3616161616161615,
+      "grad_norm": 0.009330140426754951,
+      "learning_rate": 9.963649045991173e-05,
+      "loss": 0.0029,
+      "step": 1040
+    },
+    {
+      "epoch": 3.3648484848484848,
+      "grad_norm": 0.009492140263319016,
+      "learning_rate": 9.963376577809256e-05,
+      "loss": 0.0016,
+      "step": 1041
+    },
+    {
+      "epoch": 3.368080808080808,
+      "grad_norm": 0.00905213225632906,
+      "learning_rate": 9.963103096048728e-05,
+      "loss": 0.0027,
+      "step": 1042
+    },
+    {
+      "epoch": 3.3713131313131313,
+      "grad_norm": 0.010693477466702461,
+      "learning_rate": 9.962828600765433e-05,
+      "loss": 0.003,
+      "step": 1043
+    },
+    {
+      "epoch": 3.3745454545454545,
+      "grad_norm": 0.006187601946294308,
+      "learning_rate": 9.96255309201543e-05,
+      "loss": 0.0014,
+      "step": 1044
+    },
+    {
+      "epoch": 3.3777777777777778,
+      "grad_norm": 0.00784679502248764,
+      "learning_rate": 9.962276569854977e-05,
+      "loss": 0.0019,
+      "step": 1045
+    },
+    {
+      "epoch": 3.381010101010101,
+      "grad_norm": 0.022418344393372536,
+      "learning_rate": 9.96199903434055e-05,
+      "loss": 0.0017,
+      "step": 1046
+    },
+    {
+      "epoch": 3.3842424242424243,
+      "grad_norm": 0.005133874248713255,
+      "learning_rate": 9.961720485528819e-05,
+      "loss": 0.0013,
+      "step": 1047
+    },
+    {
+      "epoch": 3.3874747474747475,
+      "grad_norm": 0.007564987521618605,
+      "learning_rate": 9.961440923476666e-05,
+      "loss": 0.0019,
+      "step": 1048
+    },
+    {
+      "epoch": 3.390707070707071,
+      "grad_norm": 0.008043873123824596,
+      "learning_rate": 9.961160348241185e-05,
+      "loss": 0.0016,
+      "step": 1049
+    },
+    {
+      "epoch": 3.393939393939394,
+      "grad_norm": 0.0051131471991539,
+      "learning_rate": 9.96087875987967e-05,
+      "loss": 0.0012,
+      "step": 1050
+    },
+    {
+      "epoch": 3.393939393939394,
+      "eval_loss": 0.003347641322761774,
+      "eval_runtime": 18.537,
+      "eval_samples_per_second": 5.395,
+      "eval_steps_per_second": 1.349,
+      "step": 1050
+    },
+    {
+      "epoch": 3.3971717171717173,
+      "grad_norm": 0.008095295168459415,
+      "learning_rate": 9.960596158449627e-05,
+      "loss": 0.002,
+      "step": 1051
+    },
+    {
+      "epoch": 3.4004040404040405,
+      "grad_norm": 0.006727342493832111,
+      "learning_rate": 9.960312544008763e-05,
+      "loss": 0.0018,
+      "step": 1052
+    },
+    {
+      "epoch": 3.403636363636364,
+      "grad_norm": 0.006898078136146069,
+      "learning_rate": 9.960027916614998e-05,
+      "loss": 0.0012,
+      "step": 1053
+    },
+    {
+      "epoch": 3.406868686868687,
+      "grad_norm": 0.007435730658471584,
+      "learning_rate": 9.959742276326456e-05,
+      "loss": 0.0017,
+      "step": 1054
+    },
+    {
+      "epoch": 3.41010101010101,
+      "grad_norm": 0.008618218824267387,
+      "learning_rate": 9.959455623201465e-05,
+      "loss": 0.0021,
+      "step": 1055
+    },
+    {
+      "epoch": 3.413333333333333,
+      "grad_norm": 0.005248959641903639,
+      "learning_rate": 9.959167957298568e-05,
+      "loss": 0.0015,
+      "step": 1056
+    },
+    {
+      "epoch": 3.4165656565656564,
+      "grad_norm": 0.006079430691897869,
+      "learning_rate": 9.958879278676506e-05,
+      "loss": 0.0015,
+      "step": 1057
+    },
+    {
+      "epoch": 3.4197979797979796,
+      "grad_norm": 0.007255829870700836,
+      "learning_rate": 9.958589587394231e-05,
+      "loss": 0.002,
+      "step": 1058
+    },
+    {
+      "epoch": 3.423030303030303,
+      "grad_norm": 0.00781642273068428,
+      "learning_rate": 9.958298883510903e-05,
+      "loss": 0.0021,
+      "step": 1059
+    },
+    {
+      "epoch": 3.426262626262626,
+      "grad_norm": 0.008929956704378128,
+      "learning_rate": 9.958007167085886e-05,
+      "loss": 0.0018,
+      "step": 1060
+    },
+    {
+      "epoch": 3.4294949494949494,
+      "grad_norm": 0.008798833936452866,
+      "learning_rate": 9.95771443817875e-05,
+      "loss": 0.0025,
+      "step": 1061
+    },
+    {
+      "epoch": 3.4327272727272726,
+      "grad_norm": 0.008338157087564468,
+      "learning_rate": 9.957420696849275e-05,
+      "loss": 0.0021,
+      "step": 1062
+    },
+    {
+      "epoch": 3.435959595959596,
+      "grad_norm": 0.011783271096646786,
+      "learning_rate": 9.957125943157448e-05,
+      "loss": 0.0024,
+      "step": 1063
+    },
+    {
+      "epoch": 3.439191919191919,
+      "grad_norm": 0.005160772707313299,
+      "learning_rate": 9.956830177163461e-05,
+      "loss": 0.0014,
+      "step": 1064
+    },
+    {
+      "epoch": 3.4424242424242424,
+      "grad_norm": 0.007435562089085579,
+      "learning_rate": 9.95653339892771e-05,
+      "loss": 0.0025,
+      "step": 1065
+    },
+    {
+      "epoch": 3.4456565656565656,
+      "grad_norm": 0.009998551569879055,
+      "learning_rate": 9.956235608510802e-05,
+      "loss": 0.0026,
+      "step": 1066
+    },
+    {
+      "epoch": 3.448888888888889,
+      "grad_norm": 0.008799971081316471,
+      "learning_rate": 9.95593680597355e-05,
+      "loss": 0.0028,
+      "step": 1067
+    },
+    {
+      "epoch": 3.452121212121212,
+      "grad_norm": 0.013296165503561497,
+      "learning_rate": 9.955636991376971e-05,
+      "loss": 0.0044,
+      "step": 1068
+    },
+    {
+      "epoch": 3.4553535353535354,
+      "grad_norm": 0.008160554803907871,
+      "learning_rate": 9.955336164782292e-05,
+      "loss": 0.002,
+      "step": 1069
+    },
+    {
+      "epoch": 3.4585858585858587,
+      "grad_norm": 0.010213302448391914,
+      "learning_rate": 9.955034326250946e-05,
+      "loss": 0.0027,
+      "step": 1070
+    },
+    {
+      "epoch": 3.461818181818182,
+      "grad_norm": 0.009601346217095852,
+      "learning_rate": 9.954731475844571e-05,
+      "loss": 0.0036,
+      "step": 1071
+    },
+    {
+      "epoch": 3.465050505050505,
+      "grad_norm": 0.009020958095788956,
+      "learning_rate": 9.954427613625013e-05,
+      "loss": 0.0027,
+      "step": 1072
+    },
+    {
+      "epoch": 3.4682828282828284,
+      "grad_norm": 0.005136935506016016,
+      "learning_rate": 9.95412273965432e-05,
+      "loss": 0.0016,
+      "step": 1073
+    },
+    {
+      "epoch": 3.4715151515151517,
+      "grad_norm": 0.004412552807480097,
+      "learning_rate": 9.953816853994759e-05,
+      "loss": 0.0012,
+      "step": 1074
+    },
+    {
+      "epoch": 3.474747474747475,
+      "grad_norm": 0.00772328395396471,
+      "learning_rate": 9.953509956708789e-05,
+      "loss": 0.0018,
+      "step": 1075
+    },
+    {
+      "epoch": 3.474747474747475,
+      "eval_loss": 0.0034066771622747183,
+      "eval_runtime": 18.5582,
+      "eval_samples_per_second": 5.388,
+      "eval_steps_per_second": 1.347,
+      "step": 1075
+    },
+    {
+      "epoch": 3.477979797979798,
+      "grad_norm": 0.010046589188277721,
+      "learning_rate": 9.953202047859085e-05,
+      "loss": 0.0035,
+      "step": 1076
+    },
+    {
+      "epoch": 3.4812121212121214,
+      "grad_norm": 0.008048112504184246,
+      "learning_rate": 9.952893127508522e-05,
+      "loss": 0.0021,
+      "step": 1077
+    },
+    {
+      "epoch": 3.4844444444444447,
+      "grad_norm": 0.0077101015485823154,
+      "learning_rate": 9.95258319572019e-05,
+      "loss": 0.002,
+      "step": 1078
+    },
+    {
+      "epoch": 3.4876767676767675,
+      "grad_norm": 0.00719053577631712,
+      "learning_rate": 9.952272252557378e-05,
+      "loss": 0.002,
+      "step": 1079
+    },
+    {
+      "epoch": 3.4909090909090907,
+      "grad_norm": 0.006908205803483725,
+      "learning_rate": 9.951960298083583e-05,
+      "loss": 0.0017,
+      "step": 1080
+    },
+    {
+      "epoch": 3.494141414141414,
+      "grad_norm": 0.006123875733464956,
+      "learning_rate": 9.95164733236251e-05,
+      "loss": 0.0019,
+      "step": 1081
+    },
+    {
+      "epoch": 3.4973737373737372,
+      "grad_norm": 0.00980414543300867,
+      "learning_rate": 9.951333355458072e-05,
+      "loss": 0.0022,
+      "step": 1082
+    },
+    {
+      "epoch": 3.5006060606060605,
+      "grad_norm": 0.004448724910616875,
+      "learning_rate": 9.951018367434386e-05,
+      "loss": 0.0012,
+      "step": 1083
+    },
+    {
+      "epoch": 3.5038383838383838,
+      "grad_norm": 0.0064595406875014305,
+      "learning_rate": 9.950702368355775e-05,
+      "loss": 0.0012,
+      "step": 1084
+    },
+    {
+      "epoch": 3.507070707070707,
+      "grad_norm": 0.005067575257271528,
+      "learning_rate": 9.950385358286772e-05,
+      "loss": 0.0014,
+      "step": 1085
+    },
+    {
+      "epoch": 3.5103030303030303,
+      "grad_norm": 0.006261507514864206,
+      "learning_rate": 9.950067337292112e-05,
+      "loss": 0.0018,
+      "step": 1086
+    },
+    {
+      "epoch": 3.5135353535353535,
+      "grad_norm": 0.008754471316933632,
+      "learning_rate": 9.949748305436741e-05,
+      "loss": 0.0027,
+      "step": 1087
+    },
+    {
+      "epoch": 3.5167676767676768,
+      "grad_norm": 0.011027954518795013,
+      "learning_rate": 9.949428262785805e-05,
+      "loss": 0.0028,
+      "step": 1088
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 0.012316351756453514,
+      "learning_rate": 9.949107209404665e-05,
+      "loss": 0.0025,
+      "step": 1089
+    },
+    {
+      "epoch": 3.5232323232323233,
+      "grad_norm": 0.005655081011354923,
+      "learning_rate": 9.948785145358879e-05,
+      "loss": 0.0014,
+      "step": 1090
+    },
+    {
+      "epoch": 3.5264646464646465,
+      "grad_norm": 0.010539629496634007,
+      "learning_rate": 9.948462070714219e-05,
+      "loss": 0.002,
+      "step": 1091
+    },
+    {
+      "epoch": 3.5296969696969698,
+      "grad_norm": 0.006462043151259422,
+      "learning_rate": 9.948137985536662e-05,
+      "loss": 0.0022,
+      "step": 1092
+    },
+    {
+      "epoch": 3.532929292929293,
+      "grad_norm": 0.007820974104106426,
+      "learning_rate": 9.947812889892387e-05,
+      "loss": 0.0016,
+      "step": 1093
+    },
+    {
+      "epoch": 3.5361616161616163,
+      "grad_norm": 0.006513669621199369,
+      "learning_rate": 9.947486783847784e-05,
+      "loss": 0.0019,
+      "step": 1094
+    },
+    {
+      "epoch": 3.5393939393939395,
+      "grad_norm": 0.007371378131210804,
+      "learning_rate": 9.947159667469446e-05,
+      "loss": 0.0014,
+      "step": 1095
+    },
+    {
+      "epoch": 3.542626262626263,
+      "grad_norm": 0.005782526917755604,
+      "learning_rate": 9.946831540824175e-05,
+      "loss": 0.0018,
+      "step": 1096
+    },
+    {
+      "epoch": 3.5458585858585856,
+      "grad_norm": 0.00670381635427475,
+      "learning_rate": 9.94650240397898e-05,
+      "loss": 0.0023,
+      "step": 1097
+    },
+    {
+      "epoch": 3.549090909090909,
+      "grad_norm": 0.010472368448972702,
+      "learning_rate": 9.946172257001069e-05,
+      "loss": 0.0031,
+      "step": 1098
+    },
+    {
+      "epoch": 3.552323232323232,
+      "grad_norm": 0.01276701781898737,
+      "learning_rate": 9.945841099957869e-05,
+      "loss": 0.0037,
+      "step": 1099
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 0.00712077459320426,
+      "learning_rate": 9.945508932917001e-05,
+      "loss": 0.0015,
+      "step": 1100
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "eval_loss": 0.003400927409529686,
+      "eval_runtime": 18.5727,
+      "eval_samples_per_second": 5.384,
+      "eval_steps_per_second": 1.346,
+      "step": 1100
+    },
+    {
+      "epoch": 3.5587878787878786,
+      "grad_norm": 0.00810240674763918,
+      "learning_rate": 9.9451757559463e-05,
+      "loss": 0.0023,
+      "step": 1101
+    },
+    {
+      "epoch": 3.562020202020202,
+      "grad_norm": 0.011555838398635387,
+      "learning_rate": 9.944841569113803e-05,
+      "loss": 0.0029,
+      "step": 1102
+    },
+    {
+      "epoch": 3.565252525252525,
+      "grad_norm": 0.006031879689544439,
+      "learning_rate": 9.944506372487754e-05,
+      "loss": 0.0015,
+      "step": 1103
+    },
+    {
+      "epoch": 3.5684848484848484,
+      "grad_norm": 0.007457388564944267,
+      "learning_rate": 9.944170166136607e-05,
+      "loss": 0.0016,
+      "step": 1104
+    },
+    {
+      "epoch": 3.5717171717171716,
+      "grad_norm": 0.00727875716984272,
+      "learning_rate": 9.943832950129018e-05,
+      "loss": 0.0017,
+      "step": 1105
+    },
+    {
+      "epoch": 3.574949494949495,
+      "grad_norm": 0.014444559812545776,
+      "learning_rate": 9.943494724533848e-05,
+      "loss": 0.0041,
+      "step": 1106
+    },
+    {
+      "epoch": 3.578181818181818,
+      "grad_norm": 0.01081145741045475,
+      "learning_rate": 9.943155489420169e-05,
+      "loss": 0.0018,
+      "step": 1107
+    },
+    {
+      "epoch": 3.5814141414141414,
+      "grad_norm": 0.01200890727341175,
+      "learning_rate": 9.942815244857256e-05,
+      "loss": 0.0026,
+      "step": 1108
+    },
+    {
+      "epoch": 3.5846464646464646,
+      "grad_norm": 0.008189111016690731,
+      "learning_rate": 9.942473990914593e-05,
+      "loss": 0.003,
+      "step": 1109
+    },
+    {
+      "epoch": 3.587878787878788,
+      "grad_norm": 0.0054262722842395306,
+      "learning_rate": 9.942131727661863e-05,
+      "loss": 0.0014,
+      "step": 1110
+    },
+    {
+      "epoch": 3.591111111111111,
+      "grad_norm": 0.009128454141318798,
+      "learning_rate": 9.941788455168965e-05,
+      "loss": 0.0023,
+      "step": 1111
+    },
+    {
+      "epoch": 3.5943434343434344,
+      "grad_norm": 0.007988560944795609,
+      "learning_rate": 9.941444173505997e-05,
+      "loss": 0.003,
+      "step": 1112
+    },
+    {
+      "epoch": 3.5975757575757576,
+      "grad_norm": 0.0045745945535600185,
+      "learning_rate": 9.941098882743267e-05,
+      "loss": 0.0015,
+      "step": 1113
+    },
+    {
+      "epoch": 3.600808080808081,
+      "grad_norm": 0.009445126168429852,
+      "learning_rate": 9.940752582951283e-05,
+      "loss": 0.0025,
+      "step": 1114
+    },
+    {
+      "epoch": 3.604040404040404,
+      "grad_norm": 0.007550603710114956,
+      "learning_rate": 9.940405274200769e-05,
+      "loss": 0.0021,
+      "step": 1115
+    },
+    {
+      "epoch": 3.6072727272727274,
+      "grad_norm": 0.006470720749348402,
+      "learning_rate": 9.940056956562645e-05,
+      "loss": 0.002,
+      "step": 1116
+    },
+    {
+      "epoch": 3.6105050505050507,
+      "grad_norm": 0.006469183601438999,
+      "learning_rate": 9.939707630108044e-05,
+      "loss": 0.0014,
+      "step": 1117
+    },
+    {
+      "epoch": 3.613737373737374,
+      "grad_norm": 0.008634147234261036,
+      "learning_rate": 9.939357294908301e-05,
+      "loss": 0.0021,
+      "step": 1118
+    },
+    {
+      "epoch": 3.616969696969697,
+      "grad_norm": 0.0053526172414422035,
+      "learning_rate": 9.939005951034959e-05,
+      "loss": 0.0013,
+      "step": 1119
+    },
+    {
+      "epoch": 3.6202020202020204,
+      "grad_norm": 0.007559166289865971,
+      "learning_rate": 9.938653598559769e-05,
+      "loss": 0.0013,
+      "step": 1120
+    },
+    {
+      "epoch": 3.6234343434343437,
+      "grad_norm": 0.005990424659103155,
+      "learning_rate": 9.93830023755468e-05,
+      "loss": 0.0014,
+      "step": 1121
+    },
+    {
+      "epoch": 3.626666666666667,
+      "grad_norm": 0.009027826599776745,
+      "learning_rate": 9.937945868091856e-05,
+      "loss": 0.002,
+      "step": 1122
+    },
+    {
+      "epoch": 3.6298989898989897,
+      "grad_norm": 0.006613972131162882,
+      "learning_rate": 9.937590490243665e-05,
+      "loss": 0.0019,
+      "step": 1123
+    },
+    {
+      "epoch": 3.633131313131313,
+      "grad_norm": 0.009335087612271309,
+      "learning_rate": 9.937234104082676e-05,
+      "loss": 0.0025,
+      "step": 1124
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 0.005268748849630356,
+      "learning_rate": 9.936876709681668e-05,
+      "loss": 0.0012,
+      "step": 1125
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "eval_loss": 0.0034194523468613625,
+      "eval_runtime": 18.5538,
+      "eval_samples_per_second": 5.39,
+      "eval_steps_per_second": 1.347,
+      "step": 1125
+    },
+    {
+      "epoch": 3.6395959595959595,
+      "grad_norm": 0.008097901940345764,
+      "learning_rate": 9.936518307113625e-05,
+      "loss": 0.0019,
+      "step": 1126
+    },
+    {
+      "epoch": 3.6428282828282827,
+      "grad_norm": 0.007424204144626856,
+      "learning_rate": 9.936158896451737e-05,
+      "loss": 0.0015,
+      "step": 1127
+    },
+    {
+      "epoch": 3.646060606060606,
+      "grad_norm": 0.00813052337616682,
+      "learning_rate": 9.9357984777694e-05,
+      "loss": 0.0026,
+      "step": 1128
+    },
+    {
+      "epoch": 3.6492929292929293,
+      "grad_norm": 0.010218325071036816,
+      "learning_rate": 9.935437051140216e-05,
+      "loss": 0.0021,
+      "step": 1129
+    },
+    {
+      "epoch": 3.6525252525252525,
+      "grad_norm": 0.006201726384460926,
+      "learning_rate": 9.935074616637992e-05,
+      "loss": 0.0015,
+      "step": 1130
+    },
+    {
+      "epoch": 3.6557575757575758,
+      "grad_norm": 0.010994481854140759,
+      "learning_rate": 9.934711174336742e-05,
+      "loss": 0.0045,
+      "step": 1131
+    },
+    {
+      "epoch": 3.658989898989899,
+      "grad_norm": 0.005597412586212158,
+      "learning_rate": 9.934346724310684e-05,
+      "loss": 0.0015,
+      "step": 1132
+    },
+    {
+      "epoch": 3.6622222222222223,
+      "grad_norm": 0.008483976125717163,
+      "learning_rate": 9.933981266634243e-05,
+      "loss": 0.0016,
+      "step": 1133
+    },
+    {
+      "epoch": 3.6654545454545455,
+      "grad_norm": 0.008453707210719585,
+      "learning_rate": 9.93361480138205e-05,
+      "loss": 0.0026,
+      "step": 1134
+    },
+    {
+      "epoch": 3.6686868686868688,
+      "grad_norm": 0.007491501048207283,
+      "learning_rate": 9.933247328628944e-05,
+      "loss": 0.0016,
+      "step": 1135
+    },
+    {
+      "epoch": 3.671919191919192,
+      "grad_norm": 0.00986399594694376,
+      "learning_rate": 9.93287884844996e-05,
+      "loss": 0.0028,
+      "step": 1136
+    },
+    {
+      "epoch": 3.6751515151515153,
+      "grad_norm": 0.0068638077937066555,
+      "learning_rate": 9.932509360920353e-05,
+      "loss": 0.0015,
+      "step": 1137
+    },
+    {
+      "epoch": 3.6783838383838385,
+      "grad_norm": 0.005731069948524237,
+      "learning_rate": 9.932138866115574e-05,
+      "loss": 0.0021,
+      "step": 1138
+    },
+    {
+      "epoch": 3.6816161616161613,
+      "grad_norm": 0.006409571971744299,
+      "learning_rate": 9.931767364111283e-05,
+      "loss": 0.0018,
+      "step": 1139
+    },
+    {
+      "epoch": 3.6848484848484846,
+      "grad_norm": 0.010908740572631359,
+      "learning_rate": 9.931394854983345e-05,
+      "loss": 0.002,
+      "step": 1140
+    },
+    {
+      "epoch": 3.688080808080808,
+      "grad_norm": 0.011212184093892574,
+      "learning_rate": 9.931021338807828e-05,
+      "loss": 0.004,
+      "step": 1141
+    },
+    {
+      "epoch": 3.691313131313131,
+      "grad_norm": 0.011185558512806892,
+      "learning_rate": 9.93064681566101e-05,
+      "loss": 0.0032,
+      "step": 1142
+    },
+    {
+      "epoch": 3.6945454545454544,
+      "grad_norm": 0.00570765370503068,
+      "learning_rate": 9.930271285619376e-05,
+      "loss": 0.0015,
+      "step": 1143
+    },
+    {
+      "epoch": 3.6977777777777776,
+      "grad_norm": 0.008618151769042015,
+      "learning_rate": 9.92989474875961e-05,
+      "loss": 0.0027,
+      "step": 1144
+    },
+    {
+      "epoch": 3.701010101010101,
+      "grad_norm": 0.009267282672226429,
+      "learning_rate": 9.929517205158605e-05,
+      "loss": 0.0032,
+      "step": 1145
+    },
+    {
+      "epoch": 3.704242424242424,
+      "grad_norm": 0.010019971989095211,
+      "learning_rate": 9.929138654893462e-05,
+      "loss": 0.0027,
+      "step": 1146
+    },
+    {
+      "epoch": 3.7074747474747474,
+      "grad_norm": 0.0062004695646464825,
+      "learning_rate": 9.928759098041482e-05,
+      "loss": 0.0018,
+      "step": 1147
+    },
+    {
+      "epoch": 3.7107070707070706,
+      "grad_norm": 0.005688016768544912,
+      "learning_rate": 9.928378534680178e-05,
+      "loss": 0.0012,
+      "step": 1148
+    },
+    {
+      "epoch": 3.713939393939394,
+      "grad_norm": 0.008614295162260532,
+      "learning_rate": 9.927996964887265e-05,
+      "loss": 0.0023,
+      "step": 1149
+    },
+    {
+      "epoch": 3.717171717171717,
+      "grad_norm": 0.004473026841878891,
+      "learning_rate": 9.927614388740663e-05,
+      "loss": 0.0018,
+      "step": 1150
+    },
+    {
+      "epoch": 3.717171717171717,
+      "eval_loss": 0.0032353622373193502,
+      "eval_runtime": 18.57,
+      "eval_samples_per_second": 5.385,
+      "eval_steps_per_second": 1.346,
+      "step": 1150
+    },
+    {
+      "epoch": 3.7204040404040404,
+      "grad_norm": 0.0077863396145403385,
+      "learning_rate": 9.9272308063185e-05,
+      "loss": 0.0016,
+      "step": 1151
+    },
+    {
+      "epoch": 3.7236363636363636,
+      "grad_norm": 0.010232211090624332,
+      "learning_rate": 9.926846217699104e-05,
+      "loss": 0.0028,
+      "step": 1152
+    },
+    {
+      "epoch": 3.726868686868687,
+      "grad_norm": 0.006621774286031723,
+      "learning_rate": 9.926460622961016e-05,
+      "loss": 0.002,
+      "step": 1153
+    },
+    {
+      "epoch": 3.73010101010101,
+      "grad_norm": 0.004830366000533104,
+      "learning_rate": 9.926074022182979e-05,
+      "loss": 0.0012,
+      "step": 1154
+    },
+    {
+      "epoch": 3.7333333333333334,
+      "grad_norm": 0.006570787634700537,
+      "learning_rate": 9.92568641544394e-05,
+      "loss": 0.0019,
+      "step": 1155
+    },
+    {
+      "epoch": 3.7365656565656566,
+      "grad_norm": 0.005988228600472212,
+      "learning_rate": 9.925297802823054e-05,
+      "loss": 0.0015,
+      "step": 1156
+    },
+    {
+      "epoch": 3.73979797979798,
+      "grad_norm": 0.007999586872756481,
+      "learning_rate": 9.924908184399677e-05,
+      "loss": 0.0024,
+      "step": 1157
+    },
+    {
+      "epoch": 3.743030303030303,
+      "grad_norm": 0.010493471287190914,
+      "learning_rate": 9.924517560253378e-05,
+      "loss": 0.002,
+      "step": 1158
+    },
+    {
+      "epoch": 3.7462626262626264,
+      "grad_norm": 0.006136655807495117,
+      "learning_rate": 9.924125930463924e-05,
+      "loss": 0.0013,
+      "step": 1159
+    },
+    {
+      "epoch": 3.7494949494949497,
+      "grad_norm": 0.0069641610607504845,
+      "learning_rate": 9.92373329511129e-05,
+      "loss": 0.0018,
+      "step": 1160
+    },
+    {
+      "epoch": 3.752727272727273,
+      "grad_norm": 0.0039301421493291855,
+      "learning_rate": 9.92333965427566e-05,
+      "loss": 0.0014,
+      "step": 1161
+    },
+    {
+      "epoch": 3.755959595959596,
+      "grad_norm": 0.006330995354801416,
+      "learning_rate": 9.922945008037417e-05,
+      "loss": 0.0018,
+      "step": 1162
+    },
+    {
+      "epoch": 3.7591919191919194,
+      "grad_norm": 0.010062163695693016,
+      "learning_rate": 9.922549356477152e-05,
+      "loss": 0.0026,
+      "step": 1163
+    },
+    {
+      "epoch": 3.7624242424242427,
+      "grad_norm": 0.00648066122084856,
+      "learning_rate": 9.922152699675664e-05,
+      "loss": 0.0015,
+      "step": 1164
+    },
+    {
+      "epoch": 3.765656565656566,
+      "grad_norm": 0.005979966837912798,
+      "learning_rate": 9.921755037713952e-05,
+      "loss": 0.0017,
+      "step": 1165
+    },
+    {
+      "epoch": 3.7688888888888887,
+      "grad_norm": 0.005628496408462524,
+      "learning_rate": 9.921356370673225e-05,
+      "loss": 0.0017,
+      "step": 1166
+    },
+    {
+      "epoch": 3.772121212121212,
+      "grad_norm": 0.008315288461744785,
+      "learning_rate": 9.920956698634896e-05,
+      "loss": 0.0019,
+      "step": 1167
+    },
+    {
+      "epoch": 3.7753535353535352,
+      "grad_norm": 0.005807152483612299,
+      "learning_rate": 9.92055602168058e-05,
+      "loss": 0.0014,
+      "step": 1168
+    },
+    {
+      "epoch": 3.7785858585858585,
+      "grad_norm": 0.006952639203518629,
+      "learning_rate": 9.920154339892104e-05,
+      "loss": 0.0016,
+      "step": 1169
+    },
+    {
+      "epoch": 3.7818181818181817,
+      "grad_norm": 0.0038634459488093853,
+      "learning_rate": 9.919751653351493e-05,
+      "loss": 0.0011,
+      "step": 1170
+    },
+    {
+      "epoch": 3.785050505050505,
+      "grad_norm": 0.007332623470574617,
+      "learning_rate": 9.919347962140979e-05,
+      "loss": 0.0023,
+      "step": 1171
+    },
+    {
+      "epoch": 3.7882828282828283,
+      "grad_norm": 0.008147547952830791,
+      "learning_rate": 9.918943266343004e-05,
+      "loss": 0.0023,
+      "step": 1172
+    },
+    {
+      "epoch": 3.7915151515151515,
+      "grad_norm": 0.010149548761546612,
+      "learning_rate": 9.91853756604021e-05,
+      "loss": 0.0022,
+      "step": 1173
+    },
+    {
+      "epoch": 3.7947474747474748,
+      "grad_norm": 0.009069012477993965,
+      "learning_rate": 9.918130861315444e-05,
+      "loss": 0.0026,
+      "step": 1174
+    },
+    {
+      "epoch": 3.797979797979798,
+      "grad_norm": 0.0058683850802481174,
+      "learning_rate": 9.91772315225176e-05,
+      "loss": 0.002,
+      "step": 1175
+    },
+    {
+      "epoch": 3.797979797979798,
+      "eval_loss": 0.003167589195072651,
+      "eval_runtime": 18.5504,
+      "eval_samples_per_second": 5.391,
+      "eval_steps_per_second": 1.348,
+      "step": 1175
+    },
+    {
+      "epoch": 3.8012121212121213,
+      "grad_norm": 0.007021645549684763,
+      "learning_rate": 9.917314438932421e-05,
+      "loss": 0.0021,
+      "step": 1176
+    },
+    {
+      "epoch": 3.8044444444444445,
+      "grad_norm": 0.006806994788348675,
+      "learning_rate": 9.916904721440887e-05,
+      "loss": 0.0032,
+      "step": 1177
+    },
+    {
+      "epoch": 3.8076767676767678,
+      "grad_norm": 0.004510834813117981,
+      "learning_rate": 9.916493999860828e-05,
+      "loss": 0.0018,
+      "step": 1178
+    },
+    {
+      "epoch": 3.810909090909091,
+      "grad_norm": 0.005397962871938944,
+      "learning_rate": 9.916082274276117e-05,
+      "loss": 0.0022,
+      "step": 1179
+    },
+    {
+      "epoch": 3.8141414141414143,
+      "grad_norm": 0.0048407320864498615,
+      "learning_rate": 9.915669544770836e-05,
+      "loss": 0.0012,
+      "step": 1180
+    },
+    {
+      "epoch": 3.8173737373737375,
+      "grad_norm": 0.008071119897067547,
+      "learning_rate": 9.915255811429267e-05,
+      "loss": 0.0027,
+      "step": 1181
+    },
+    {
+      "epoch": 3.8206060606060603,
+      "grad_norm": 0.005428985692560673,
+      "learning_rate": 9.914841074335898e-05,
+      "loss": 0.002,
+      "step": 1182
+    },
+    {
+      "epoch": 3.8238383838383836,
+      "grad_norm": 0.00699599226936698,
+      "learning_rate": 9.914425333575426e-05,
+      "loss": 0.0019,
+      "step": 1183
+    },
+    {
+      "epoch": 3.827070707070707,
+      "grad_norm": 0.007088671904057264,
+      "learning_rate": 9.914008589232749e-05,
+      "loss": 0.0024,
+      "step": 1184
+    },
+    {
+      "epoch": 3.83030303030303,
+      "grad_norm": 0.008141590282320976,
+      "learning_rate": 9.91359084139297e-05,
+      "loss": 0.0028,
+      "step": 1185
+    },
+    {
+      "epoch": 3.8335353535353534,
+      "grad_norm": 0.006589834112673998,
+      "learning_rate": 9.913172090141399e-05,
+      "loss": 0.002,
+      "step": 1186
+    },
+    {
+      "epoch": 3.8367676767676766,
+      "grad_norm": 0.008189876563847065,
+      "learning_rate": 9.912752335563548e-05,
+      "loss": 0.0019,
+      "step": 1187
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 0.005289820954203606,
+      "learning_rate": 9.912331577745138e-05,
+      "loss": 0.0017,
+      "step": 1188
+    },
+    {
+      "epoch": 3.843232323232323,
+      "grad_norm": 0.0077620758675038815,
+      "learning_rate": 9.911909816772091e-05,
+      "loss": 0.0024,
+      "step": 1189
+    },
+    {
+      "epoch": 3.8464646464646464,
+      "grad_norm": 0.006231298670172691,
+      "learning_rate": 9.911487052730537e-05,
+      "loss": 0.0016,
+      "step": 1190
+    },
+    {
+      "epoch": 3.8496969696969696,
+      "grad_norm": 0.007889009080827236,
+      "learning_rate": 9.911063285706808e-05,
+      "loss": 0.002,
+      "step": 1191
+    },
+    {
+      "epoch": 3.852929292929293,
+      "grad_norm": 0.01032167673110962,
+      "learning_rate": 9.910638515787442e-05,
+      "loss": 0.0014,
+      "step": 1192
+    },
+    {
+      "epoch": 3.856161616161616,
+      "grad_norm": 0.007993797771632671,
+      "learning_rate": 9.910212743059182e-05,
+      "loss": 0.0022,
+      "step": 1193
+    },
+    {
+      "epoch": 3.8593939393939394,
+      "grad_norm": 0.005151242949068546,
+      "learning_rate": 9.909785967608977e-05,
+      "loss": 0.0015,
+      "step": 1194
+    },
+    {
+      "epoch": 3.8626262626262626,
+      "grad_norm": 0.007885465398430824,
+      "learning_rate": 9.909358189523978e-05,
+      "loss": 0.0024,
+      "step": 1195
+    },
+    {
+      "epoch": 3.865858585858586,
+      "grad_norm": 0.009114793501794338,
+      "learning_rate": 9.908929408891542e-05,
+      "loss": 0.0015,
+      "step": 1196
+    },
+    {
+      "epoch": 3.869090909090909,
+      "grad_norm": 0.00829376745969057,
+      "learning_rate": 9.908499625799235e-05,
+      "loss": 0.002,
+      "step": 1197
+    },
+    {
+      "epoch": 3.8723232323232324,
+      "grad_norm": 0.007342406082898378,
+      "learning_rate": 9.908068840334818e-05,
+      "loss": 0.0021,
+      "step": 1198
+    },
+    {
+      "epoch": 3.8755555555555556,
+      "grad_norm": 0.0075923847034573555,
+      "learning_rate": 9.907637052586265e-05,
+      "loss": 0.0019,
+      "step": 1199
+    },
+    {
+      "epoch": 3.878787878787879,
+      "grad_norm": 0.008298894390463829,
+      "learning_rate": 9.907204262641751e-05,
+      "loss": 0.002,
+      "step": 1200
+    },
+    {
+      "epoch": 3.878787878787879,
+      "eval_loss": 0.0031958832405507565,
+      "eval_runtime": 18.5457,
+      "eval_samples_per_second": 5.392,
+      "eval_steps_per_second": 1.348,
+      "step": 1200
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 7725,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 25,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.684921203405947e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}