diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,9410 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9997177001976099,
+  "eval_steps": 133,
+  "global_step": 1328,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0007527994730403689,
+      "grad_norm": 36.36020673883064,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 2.0808,
+      "step": 1
+    },
+    {
+      "epoch": 0.0015055989460807378,
+      "grad_norm": 41.49766742673561,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 2.0787,
+      "step": 2
+    },
+    {
+      "epoch": 0.0022583984191211067,
+      "grad_norm": 29.396842470362714,
+      "learning_rate": 1.5e-06,
+      "loss": 1.9621,
+      "step": 3
+    },
+    {
+      "epoch": 0.0030111978921614755,
+      "grad_norm": 22.162520153974473,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.8945,
+      "step": 4
+    },
+    {
+      "epoch": 0.0037639973652018443,
+      "grad_norm": 29.318403975263223,
+      "learning_rate": 2.5e-06,
+      "loss": 2.1412,
+      "step": 5
+    },
+    {
+      "epoch": 0.0045167968382422135,
+      "grad_norm": 18.707385510288105,
+      "learning_rate": 3e-06,
+      "loss": 2.1158,
+      "step": 6
+    },
+    {
+      "epoch": 0.005269596311282582,
+      "grad_norm": 19.505666042431155,
+      "learning_rate": 3.5e-06,
+      "loss": 1.8249,
+      "step": 7
+    },
+    {
+      "epoch": 0.006022395784322951,
+      "grad_norm": 12.946325162286865,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.7453,
+      "step": 8
+    },
+    {
+      "epoch": 0.00677519525736332,
+      "grad_norm": 48.622192972274625,
+      "learning_rate": 4.5e-06,
+      "loss": 1.6962,
+      "step": 9
+    },
+    {
+      "epoch": 0.007527994730403689,
+      "grad_norm": 19.565469972794624,
+      "learning_rate": 5e-06,
+      "loss": 1.6878,
+      "step": 10
+    },
+    {
+      "epoch": 0.008280794203444058,
+      "grad_norm": 75.67524968927123,
+      "learning_rate": 5.500000000000001e-06,
+      "loss": 1.6624,
+      "step": 11
+    },
+    {
+      "epoch": 0.009033593676484427,
+      "grad_norm": 30.445369285190505,
+      "learning_rate": 6e-06,
+      "loss": 1.6926,
+      "step": 12
+    },
+    {
+      "epoch": 0.009786393149524796,
+      "grad_norm": 9.286678675822756,
+      "learning_rate": 6.5000000000000004e-06,
+      "loss": 1.493,
+      "step": 13
+    },
+    {
+      "epoch": 0.010539192622565164,
+      "grad_norm": 8.800345354001589,
+      "learning_rate": 7e-06,
+      "loss": 1.4314,
+      "step": 14
+    },
+    {
+      "epoch": 0.011291992095605533,
+      "grad_norm": 17.285960388196333,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 1.552,
+      "step": 15
+    },
+    {
+      "epoch": 0.012044791568645902,
+      "grad_norm": 9.20149362304735,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.4517,
+      "step": 16
+    },
+    {
+      "epoch": 0.012797591041686271,
+      "grad_norm": 12.357494553067204,
+      "learning_rate": 8.5e-06,
+      "loss": 1.5248,
+      "step": 17
+    },
+    {
+      "epoch": 0.01355039051472664,
+      "grad_norm": 5.930012552972285,
+      "learning_rate": 9e-06,
+      "loss": 1.3537,
+      "step": 18
+    },
+    {
+      "epoch": 0.014303189987767008,
+      "grad_norm": 7.406734075765084,
+      "learning_rate": 9.5e-06,
+      "loss": 1.437,
+      "step": 19
+    },
+    {
+      "epoch": 0.015055989460807377,
+      "grad_norm": 7.89125237928406,
+      "learning_rate": 1e-05,
+      "loss": 1.4079,
+      "step": 20
+    },
+    {
+      "epoch": 0.015808788933847746,
+      "grad_norm": 9.23489645591984,
+      "learning_rate": 1.0500000000000001e-05,
+      "loss": 1.4561,
+      "step": 21
+    },
+    {
+      "epoch": 0.016561588406888116,
+      "grad_norm": 6.965625824093193,
+      "learning_rate": 1.1000000000000001e-05,
+      "loss": 1.3439,
+      "step": 22
+    },
+    {
+      "epoch": 0.017314387879928485,
+      "grad_norm": 7.080158120837606,
+      "learning_rate": 1.15e-05,
+      "loss": 1.3862,
+      "step": 23
+    },
+    {
+      "epoch": 0.018067187352968854,
+      "grad_norm": 7.7009738518492235,
+      "learning_rate": 1.2e-05,
+      "loss": 1.3187,
+      "step": 24
+    },
+    {
+      "epoch": 0.018819986826009223,
+      "grad_norm": 6.953266502376254,
+      "learning_rate": 1.25e-05,
+      "loss": 1.4262,
+      "step": 25
+    },
+    {
+      "epoch": 0.019572786299049592,
+      "grad_norm": 5.092442953921055,
+      "learning_rate": 1.3000000000000001e-05,
+      "loss": 1.2701,
+      "step": 26
+    },
+    {
+      "epoch": 0.020325585772089958,
+      "grad_norm": 5.561143584438669,
+      "learning_rate": 1.3500000000000001e-05,
+      "loss": 1.4132,
+      "step": 27
+    },
+    {
+      "epoch": 0.021078385245130327,
+      "grad_norm": 9.291190855019563,
+      "learning_rate": 1.4e-05,
+      "loss": 1.3376,
+      "step": 28
+    },
+    {
+      "epoch": 0.021831184718170697,
+      "grad_norm": 6.741096041673811,
+      "learning_rate": 1.45e-05,
+      "loss": 1.3535,
+      "step": 29
+    },
+    {
+      "epoch": 0.022583984191211066,
+      "grad_norm": 6.345428967677617,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 1.281,
+      "step": 30
+    },
+    {
+      "epoch": 0.023336783664251435,
+      "grad_norm": 6.240759122218503,
+      "learning_rate": 1.55e-05,
+      "loss": 1.3036,
+      "step": 31
+    },
+    {
+      "epoch": 0.024089583137291804,
+      "grad_norm": 7.655198830811222,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 1.3841,
+      "step": 32
+    },
+    {
+      "epoch": 0.024842382610332173,
+      "grad_norm": 5.018774253201613,
+      "learning_rate": 1.65e-05,
+      "loss": 1.278,
+      "step": 33
+    },
+    {
+      "epoch": 0.025595182083372543,
+      "grad_norm": 5.460243707216422,
+      "learning_rate": 1.7e-05,
+      "loss": 1.3331,
+      "step": 34
+    },
+    {
+      "epoch": 0.026347981556412912,
+      "grad_norm": 4.624338654763614,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 1.2542,
+      "step": 35
+    },
+    {
+      "epoch": 0.02710078102945328,
+      "grad_norm": 7.701628120385831,
+      "learning_rate": 1.8e-05,
+      "loss": 1.298,
+      "step": 36
+    },
+    {
+      "epoch": 0.027853580502493647,
+      "grad_norm": 5.136489226357356,
+      "learning_rate": 1.8500000000000002e-05,
+      "loss": 1.1864,
+      "step": 37
+    },
+    {
+      "epoch": 0.028606379975534016,
+      "grad_norm": 6.874578456975717,
+      "learning_rate": 1.9e-05,
+      "loss": 1.2732,
+      "step": 38
+    },
+    {
+      "epoch": 0.029359179448574385,
+      "grad_norm": 5.811036013054745,
+      "learning_rate": 1.95e-05,
+      "loss": 1.2806,
+      "step": 39
+    },
+    {
+      "epoch": 0.030111978921614754,
+      "grad_norm": 5.51821322763505,
+      "learning_rate": 2e-05,
+      "loss": 1.3575,
+      "step": 40
+    },
+    {
+      "epoch": 0.030864778394655124,
+      "grad_norm": 6.6228648443339475,
+      "learning_rate": 1.999997025336748e-05,
+      "loss": 1.2004,
+      "step": 41
+    },
+    {
+      "epoch": 0.03161757786769549,
+      "grad_norm": 5.963103240408259,
+      "learning_rate": 1.9999881013646893e-05,
+      "loss": 1.2557,
+      "step": 42
+    },
+    {
+      "epoch": 0.03237037734073586,
+      "grad_norm": 6.519925159232462,
+      "learning_rate": 1.999973228136915e-05,
+      "loss": 1.2194,
+      "step": 43
+    },
+    {
+      "epoch": 0.03312317681377623,
+      "grad_norm": 5.898983728146134,
+      "learning_rate": 1.9999524057419116e-05,
+      "loss": 1.228,
+      "step": 44
+    },
+    {
+      "epoch": 0.0338759762868166,
+      "grad_norm": 7.997013591911161,
+      "learning_rate": 1.9999256343035577e-05,
+      "loss": 1.2152,
+      "step": 45
+    },
+    {
+      "epoch": 0.03462877575985697,
+      "grad_norm": 11.068558967470823,
+      "learning_rate": 1.9998929139811257e-05,
+      "loss": 1.2682,
+      "step": 46
+    },
+    {
+      "epoch": 0.03538157523289734,
+      "grad_norm": 6.025642126702577,
+      "learning_rate": 1.9998542449692794e-05,
+      "loss": 1.2186,
+      "step": 47
+    },
+    {
+      "epoch": 0.03613437470593771,
+      "grad_norm": 5.1234335564506015,
+      "learning_rate": 1.999809627498073e-05,
+      "loss": 1.0775,
+      "step": 48
+    },
+    {
+      "epoch": 0.03688717417897808,
+      "grad_norm": 6.9589319498297835,
+      "learning_rate": 1.9997590618329507e-05,
+      "loss": 1.1942,
+      "step": 49
+    },
+    {
+      "epoch": 0.037639973652018446,
+      "grad_norm": 4.899408391501564,
+      "learning_rate": 1.999702548274744e-05,
+      "loss": 1.1736,
+      "step": 50
+    },
+    {
+      "epoch": 0.038392773125058816,
+      "grad_norm": 6.523950562265359,
+      "learning_rate": 1.999640087159671e-05,
+      "loss": 1.1497,
+      "step": 51
+    },
+    {
+      "epoch": 0.039145572598099185,
+      "grad_norm": 6.080223023749499,
+      "learning_rate": 1.999571678859333e-05,
+      "loss": 1.1685,
+      "step": 52
+    },
+    {
+      "epoch": 0.03989837207113955,
+      "grad_norm": 4.810923884376871,
+      "learning_rate": 1.9994973237807133e-05,
+      "loss": 1.1577,
+      "step": 53
+    },
+    {
+      "epoch": 0.040651171544179916,
+      "grad_norm": 9.585902843233736,
+      "learning_rate": 1.999417022366174e-05,
+      "loss": 1.2533,
+      "step": 54
+    },
+    {
+      "epoch": 0.041403971017220285,
+      "grad_norm": 7.121805505321927,
+      "learning_rate": 1.9993307750934555e-05,
+      "loss": 1.2452,
+      "step": 55
+    },
+    {
+      "epoch": 0.042156770490260655,
+      "grad_norm": 5.119582554676219,
+      "learning_rate": 1.99923858247567e-05,
+      "loss": 1.1369,
+      "step": 56
+    },
+    {
+      "epoch": 0.042909569963301024,
+      "grad_norm": 6.992826625922773,
+      "learning_rate": 1.999140445061302e-05,
+      "loss": 1.0986,
+      "step": 57
+    },
+    {
+      "epoch": 0.04366236943634139,
+      "grad_norm": 4.249338096365068,
+      "learning_rate": 1.9990363634342032e-05,
+      "loss": 1.1033,
+      "step": 58
+    },
+    {
+      "epoch": 0.04441516890938176,
+      "grad_norm": 5.775070285941309,
+      "learning_rate": 1.9989263382135882e-05,
+      "loss": 1.1086,
+      "step": 59
+    },
+    {
+      "epoch": 0.04516796838242213,
+      "grad_norm": 6.663400587483444,
+      "learning_rate": 1.9988103700540345e-05,
+      "loss": 1.1803,
+      "step": 60
+    },
+    {
+      "epoch": 0.0459207678554625,
+      "grad_norm": 5.183870397182474,
+      "learning_rate": 1.998688459645473e-05,
+      "loss": 1.1531,
+      "step": 61
+    },
+    {
+      "epoch": 0.04667356732850287,
+      "grad_norm": 3.5395456165525427,
+      "learning_rate": 1.9985606077131895e-05,
+      "loss": 1.0054,
+      "step": 62
+    },
+    {
+      "epoch": 0.04742636680154324,
+      "grad_norm": 3.8150354882109503,
+      "learning_rate": 1.998426815017817e-05,
+      "loss": 1.1239,
+      "step": 63
+    },
+    {
+      "epoch": 0.04817916627458361,
+      "grad_norm": 5.03910810217068,
+      "learning_rate": 1.998287082355331e-05,
+      "loss": 1.1786,
+      "step": 64
+    },
+    {
+      "epoch": 0.04893196574762398,
+      "grad_norm": 5.5510267998080645,
+      "learning_rate": 1.9981414105570473e-05,
+      "loss": 1.1524,
+      "step": 65
+    },
+    {
+      "epoch": 0.04968476522066435,
+      "grad_norm": 4.565720771118459,
+      "learning_rate": 1.997989800489615e-05,
+      "loss": 1.154,
+      "step": 66
+    },
+    {
+      "epoch": 0.050437564693704716,
+      "grad_norm": 4.530477664369838,
+      "learning_rate": 1.997832253055012e-05,
+      "loss": 1.1039,
+      "step": 67
+    },
+    {
+      "epoch": 0.051190364166745085,
+      "grad_norm": 4.123668763760377,
+      "learning_rate": 1.9976687691905394e-05,
+      "loss": 1.15,
+      "step": 68
+    },
+    {
+      "epoch": 0.051943163639785454,
+      "grad_norm": 7.025385301726715,
+      "learning_rate": 1.997499349868816e-05,
+      "loss": 1.0612,
+      "step": 69
+    },
+    {
+      "epoch": 0.052695963112825824,
+      "grad_norm": 4.176310522867936,
+      "learning_rate": 1.997323996097772e-05,
+      "loss": 1.0894,
+      "step": 70
+    },
+    {
+      "epoch": 0.05344876258586619,
+      "grad_norm": 3.246977668511193,
+      "learning_rate": 1.9971427089206458e-05,
+      "loss": 1.1196,
+      "step": 71
+    },
+    {
+      "epoch": 0.05420156205890656,
+      "grad_norm": 4.176862011800632,
+      "learning_rate": 1.9969554894159723e-05,
+      "loss": 1.1921,
+      "step": 72
+    },
+    {
+      "epoch": 0.054954361531946924,
+      "grad_norm": 8.805463366473532,
+      "learning_rate": 1.9967623386975826e-05,
+      "loss": 1.0246,
+      "step": 73
+    },
+    {
+      "epoch": 0.05570716100498729,
+      "grad_norm": 3.64619595168792,
+      "learning_rate": 1.996563257914593e-05,
+      "loss": 1.0822,
+      "step": 74
+    },
+    {
+      "epoch": 0.05645996047802766,
+      "grad_norm": 4.217544603504828,
+      "learning_rate": 1.9963582482514003e-05,
+      "loss": 1.1187,
+      "step": 75
+    },
+    {
+      "epoch": 0.05721275995106803,
+      "grad_norm": 5.678985309075052,
+      "learning_rate": 1.9961473109276735e-05,
+      "loss": 1.2067,
+      "step": 76
+    },
+    {
+      "epoch": 0.0579655594241084,
+      "grad_norm": 5.001296565321351,
+      "learning_rate": 1.9959304471983477e-05,
+      "loss": 1.1527,
+      "step": 77
+    },
+    {
+      "epoch": 0.05871835889714877,
+      "grad_norm": 5.604515835465151,
+      "learning_rate": 1.9957076583536166e-05,
+      "loss": 1.1761,
+      "step": 78
+    },
+    {
+      "epoch": 0.05947115837018914,
+      "grad_norm": 5.095593228417783,
+      "learning_rate": 1.995478945718923e-05,
+      "loss": 1.1628,
+      "step": 79
+    },
+    {
+      "epoch": 0.06022395784322951,
+      "grad_norm": 6.597427071965454,
+      "learning_rate": 1.9952443106549535e-05,
+      "loss": 1.2414,
+      "step": 80
+    },
+    {
+      "epoch": 0.06097675731626988,
+      "grad_norm": 6.9875665495378225,
+      "learning_rate": 1.9950037545576288e-05,
+      "loss": 1.2225,
+      "step": 81
+    },
+    {
+      "epoch": 0.06172955678931025,
+      "grad_norm": 4.687344949995792,
+      "learning_rate": 1.994757278858095e-05,
+      "loss": 1.1402,
+      "step": 82
+    },
+    {
+      "epoch": 0.062482356262350616,
+      "grad_norm": 5.277541712626479,
+      "learning_rate": 1.994504885022717e-05,
+      "loss": 1.1022,
+      "step": 83
+    },
+    {
+      "epoch": 0.06323515573539099,
+      "grad_norm": 2.9552521360386907,
+      "learning_rate": 1.9942465745530687e-05,
+      "loss": 1.0843,
+      "step": 84
+    },
+    {
+      "epoch": 0.06398795520843135,
+      "grad_norm": 3.454331479730833,
+      "learning_rate": 1.9939823489859226e-05,
+      "loss": 1.0417,
+      "step": 85
+    },
+    {
+      "epoch": 0.06474075468147172,
+      "grad_norm": 6.494959979859043,
+      "learning_rate": 1.9937122098932428e-05,
+      "loss": 1.1774,
+      "step": 86
+    },
+    {
+      "epoch": 0.06549355415451209,
+      "grad_norm": 5.232583927463086,
+      "learning_rate": 1.9934361588821757e-05,
+      "loss": 1.1037,
+      "step": 87
+    },
+    {
+      "epoch": 0.06624635362755246,
+      "grad_norm": 4.2279381075948175,
+      "learning_rate": 1.993154197595038e-05,
+      "loss": 1.1474,
+      "step": 88
+    },
+    {
+      "epoch": 0.06699915310059283,
+      "grad_norm": 4.148924762168968,
+      "learning_rate": 1.99286632770931e-05,
+      "loss": 1.1898,
+      "step": 89
+    },
+    {
+      "epoch": 0.0677519525736332,
+      "grad_norm": 5.154636179011585,
+      "learning_rate": 1.9925725509376236e-05,
+      "loss": 1.1342,
+      "step": 90
+    },
+    {
+      "epoch": 0.06850475204667357,
+      "grad_norm": 5.217031244017699,
+      "learning_rate": 1.9922728690277528e-05,
+      "loss": 1.1692,
+      "step": 91
+    },
+    {
+      "epoch": 0.06925755151971394,
+      "grad_norm": 4.338496940394976,
+      "learning_rate": 1.991967283762603e-05,
+      "loss": 1.141,
+      "step": 92
+    },
+    {
+      "epoch": 0.07001035099275431,
+      "grad_norm": 4.197923670398518,
+      "learning_rate": 1.9916557969602007e-05,
+      "loss": 1.1289,
+      "step": 93
+    },
+    {
+      "epoch": 0.07076315046579468,
+      "grad_norm": 3.743275306079982,
+      "learning_rate": 1.991338410473683e-05,
+      "loss": 1.1073,
+      "step": 94
+    },
+    {
+      "epoch": 0.07151594993883505,
+      "grad_norm": 8.539568294151485,
+      "learning_rate": 1.991015126191285e-05,
+      "loss": 1.1257,
+      "step": 95
+    },
+    {
+      "epoch": 0.07226874941187542,
+      "grad_norm": 3.5627327495855856,
+      "learning_rate": 1.9906859460363307e-05,
+      "loss": 1.063,
+      "step": 96
+    },
+    {
+      "epoch": 0.07302154888491579,
+      "grad_norm": 3.7469430274257376,
+      "learning_rate": 1.9903508719672208e-05,
+      "loss": 1.041,
+      "step": 97
+    },
+    {
+      "epoch": 0.07377434835795615,
+      "grad_norm": 5.4214290416554025,
+      "learning_rate": 1.9900099059774197e-05,
+      "loss": 1.1798,
+      "step": 98
+    },
+    {
+      "epoch": 0.07452714783099652,
+      "grad_norm": 6.3484542327964615,
+      "learning_rate": 1.989663050095446e-05,
+      "loss": 1.0986,
+      "step": 99
+    },
+    {
+      "epoch": 0.07527994730403689,
+      "grad_norm": 3.8840684282304982,
+      "learning_rate": 1.989310306384858e-05,
+      "loss": 1.1509,
+      "step": 100
+    },
+    {
+      "epoch": 0.07603274677707726,
+      "grad_norm": 6.221356396939684,
+      "learning_rate": 1.9889516769442436e-05,
+      "loss": 1.0952,
+      "step": 101
+    },
+    {
+      "epoch": 0.07678554625011763,
+      "grad_norm": 4.917299687739317,
+      "learning_rate": 1.988587163907206e-05,
+      "loss": 1.1515,
+      "step": 102
+    },
+    {
+      "epoch": 0.077538345723158,
+      "grad_norm": 3.71254487396025,
+      "learning_rate": 1.988216769442353e-05,
+      "loss": 1.1358,
+      "step": 103
+    },
+    {
+      "epoch": 0.07829114519619837,
+      "grad_norm": 5.144719861197378,
+      "learning_rate": 1.9878404957532817e-05,
+      "loss": 1.1711,
+      "step": 104
+    },
+    {
+      "epoch": 0.07904394466923872,
+      "grad_norm": 3.9343091087557105,
+      "learning_rate": 1.987458345078567e-05,
+      "loss": 1.1338,
+      "step": 105
+    },
+    {
+      "epoch": 0.0797967441422791,
+      "grad_norm": 5.493391217443812,
+      "learning_rate": 1.9870703196917485e-05,
+      "loss": 1.1356,
+      "step": 106
+    },
+    {
+      "epoch": 0.08054954361531946,
+      "grad_norm": 4.196852899871084,
+      "learning_rate": 1.9866764219013154e-05,
+      "loss": 1.1412,
+      "step": 107
+    },
+    {
+      "epoch": 0.08130234308835983,
+      "grad_norm": 4.031891342725685,
+      "learning_rate": 1.986276654050695e-05,
+      "loss": 1.2285,
+      "step": 108
+    },
+    {
+      "epoch": 0.0820551425614002,
+      "grad_norm": 5.3508039600144865,
+      "learning_rate": 1.985871018518236e-05,
+      "loss": 1.2115,
+      "step": 109
+    },
+    {
+      "epoch": 0.08280794203444057,
+      "grad_norm": 4.108835495999991,
+      "learning_rate": 1.9854595177171968e-05,
+      "loss": 1.1451,
+      "step": 110
+    },
+    {
+      "epoch": 0.08356074150748094,
+      "grad_norm": 3.302036630754478,
+      "learning_rate": 1.9850421540957307e-05,
+      "loss": 1.0757,
+      "step": 111
+    },
+    {
+      "epoch": 0.08431354098052131,
+      "grad_norm": 4.414268844516725,
+      "learning_rate": 1.984618930136869e-05,
+      "loss": 1.0881,
+      "step": 112
+    },
+    {
+      "epoch": 0.08506634045356168,
+      "grad_norm": 3.700539962241975,
+      "learning_rate": 1.98418984835851e-05,
+      "loss": 1.144,
+      "step": 113
+    },
+    {
+      "epoch": 0.08581913992660205,
+      "grad_norm": 4.9358231906298204,
+      "learning_rate": 1.9837549113134015e-05,
+      "loss": 1.038,
+      "step": 114
+    },
+    {
+      "epoch": 0.08657193939964242,
+      "grad_norm": 5.645738001161152,
+      "learning_rate": 1.9833141215891253e-05,
+      "loss": 1.2081,
+      "step": 115
+    },
+    {
+      "epoch": 0.08732473887268279,
+      "grad_norm": 5.119377292254036,
+      "learning_rate": 1.9828674818080837e-05,
+      "loss": 1.1226,
+      "step": 116
+    },
+    {
+      "epoch": 0.08807753834572316,
+      "grad_norm": 4.1763020003723526,
+      "learning_rate": 1.9824149946274827e-05,
+      "loss": 1.1374,
+      "step": 117
+    },
+    {
+      "epoch": 0.08883033781876352,
+      "grad_norm": 5.503112263483898,
+      "learning_rate": 1.981956662739316e-05,
+      "loss": 1.1168,
+      "step": 118
+    },
+    {
+      "epoch": 0.0895831372918039,
+      "grad_norm": 3.9664193299971147,
+      "learning_rate": 1.98149248887035e-05,
+      "loss": 1.1612,
+      "step": 119
+    },
+    {
+      "epoch": 0.09033593676484426,
+      "grad_norm": 4.46872589770535,
+      "learning_rate": 1.9810224757821063e-05,
+      "loss": 1.0124,
+      "step": 120
+    },
+    {
+      "epoch": 0.09108873623788463,
+      "grad_norm": 3.7293231569422702,
+      "learning_rate": 1.9805466262708464e-05,
+      "loss": 1.1548,
+      "step": 121
+    },
+    {
+      "epoch": 0.091841535710925,
+      "grad_norm": 4.012485005147592,
+      "learning_rate": 1.9800649431675544e-05,
+      "loss": 1.1773,
+      "step": 122
+    },
+    {
+      "epoch": 0.09259433518396537,
+      "grad_norm": 3.5463279826981102,
+      "learning_rate": 1.9795774293379206e-05,
+      "loss": 1.11,
+      "step": 123
+    },
+    {
+      "epoch": 0.09334713465700574,
+      "grad_norm": 3.528951778129194,
+      "learning_rate": 1.979084087682323e-05,
+      "loss": 1.2064,
+      "step": 124
+    },
+    {
+      "epoch": 0.09409993413004611,
+      "grad_norm": 10.1058424194676,
+      "learning_rate": 1.9785849211358133e-05,
+      "loss": 1.1444,
+      "step": 125
+    },
+    {
+      "epoch": 0.09485273360308648,
+      "grad_norm": 6.315992672709579,
+      "learning_rate": 1.9780799326680956e-05,
+      "loss": 1.1438,
+      "step": 126
+    },
+    {
+      "epoch": 0.09560553307612685,
+      "grad_norm": 4.060283476245672,
+      "learning_rate": 1.9775691252835113e-05,
+      "loss": 1.0999,
+      "step": 127
+    },
+    {
+      "epoch": 0.09635833254916722,
+      "grad_norm": 3.8976057736287313,
+      "learning_rate": 1.9770525020210204e-05,
+      "loss": 1.1222,
+      "step": 128
+    },
+    {
+      "epoch": 0.09711113202220759,
+      "grad_norm": 3.9139807812132905,
+      "learning_rate": 1.9765300659541837e-05,
+      "loss": 1.1164,
+      "step": 129
+    },
+    {
+      "epoch": 0.09786393149524795,
+      "grad_norm": 4.995087522269767,
+      "learning_rate": 1.976001820191143e-05,
+      "loss": 1.085,
+      "step": 130
+    },
+    {
+      "epoch": 0.09861673096828832,
+      "grad_norm": 3.7833873291414823,
+      "learning_rate": 1.9754677678746064e-05,
+      "loss": 1.0634,
+      "step": 131
+    },
+    {
+      "epoch": 0.0993695304413287,
+      "grad_norm": 3.334670976301107,
+      "learning_rate": 1.9749279121818235e-05,
+      "loss": 1.1077,
+      "step": 132
+    },
+    {
+      "epoch": 0.10012232991436906,
+      "grad_norm": 7.0827146935611065,
+      "learning_rate": 1.9743822563245738e-05,
+      "loss": 1.1753,
+      "step": 133
+    },
+    {
+      "epoch": 0.10012232991436906,
+      "eval_loss": 1.0733484029769897,
+      "eval_runtime": 600.4408,
+      "eval_samples_per_second": 29.786,
+      "eval_steps_per_second": 0.466,
+      "step": 133
+    },
+    {
+      "epoch": 0.10087512938740943,
+      "grad_norm": 3.600867409556725,
+      "learning_rate": 1.973830803549141e-05,
+      "loss": 0.9933,
+      "step": 134
+    },
+    {
+      "epoch": 0.1016279288604498,
+      "grad_norm": 3.8689385067834148,
+      "learning_rate": 1.9732735571362985e-05,
+      "loss": 1.1034,
+      "step": 135
+    },
+    {
+      "epoch": 0.10238072833349017,
+      "grad_norm": 6.883779237321949,
+      "learning_rate": 1.972710520401287e-05,
+      "loss": 1.1141,
+      "step": 136
+    },
+    {
+      "epoch": 0.10313352780653054,
+      "grad_norm": 4.762968226804512,
+      "learning_rate": 1.972141696693795e-05,
+      "loss": 1.2243,
+      "step": 137
+    },
+    {
+      "epoch": 0.10388632727957091,
+      "grad_norm": 3.6511081137242902,
+      "learning_rate": 1.9715670893979416e-05,
+      "loss": 1.0188,
+      "step": 138
+    },
+    {
+      "epoch": 0.10463912675261128,
+      "grad_norm": 3.5691298823695914,
+      "learning_rate": 1.9709867019322528e-05,
+      "loss": 1.0551,
+      "step": 139
+    },
+    {
+      "epoch": 0.10539192622565165,
+      "grad_norm": 3.5475531225746444,
+      "learning_rate": 1.9704005377496428e-05,
+      "loss": 1.0743,
+      "step": 140
+    },
+    {
+      "epoch": 0.10614472569869202,
+      "grad_norm": 3.791897303162606,
+      "learning_rate": 1.969808600337394e-05,
+      "loss": 1.066,
+      "step": 141
+    },
+    {
+      "epoch": 0.10689752517173239,
+      "grad_norm": 4.693893833681282,
+      "learning_rate": 1.969210893217136e-05,
+      "loss": 1.1729,
+      "step": 142
+    },
+    {
+      "epoch": 0.10765032464477275,
+      "grad_norm": 3.2325270058743,
+      "learning_rate": 1.9686074199448222e-05,
+      "loss": 1.1003,
+      "step": 143
+    },
+    {
+      "epoch": 0.10840312411781312,
+      "grad_norm": 3.546773416480843,
+      "learning_rate": 1.967998184110713e-05,
+      "loss": 1.1224,
+      "step": 144
+    },
+    {
+      "epoch": 0.10915592359085349,
+      "grad_norm": 4.8504525234744245,
+      "learning_rate": 1.967383189339352e-05,
+      "loss": 1.2487,
+      "step": 145
+    },
+    {
+      "epoch": 0.10990872306389385,
+      "grad_norm": 3.224089352304579,
+      "learning_rate": 1.9667624392895423e-05,
+      "loss": 1.0354,
+      "step": 146
+    },
+    {
+      "epoch": 0.11066152253693422,
+      "grad_norm": 4.260944538348364,
+      "learning_rate": 1.9661359376543295e-05,
+      "loss": 1.1538,
+      "step": 147
+    },
+    {
+      "epoch": 0.11141432200997459,
+      "grad_norm": 3.7968699711076965,
+      "learning_rate": 1.9655036881609763e-05,
+      "loss": 1.1139,
+      "step": 148
+    },
+    {
+      "epoch": 0.11216712148301496,
+      "grad_norm": 6.106144924923444,
+      "learning_rate": 1.9648656945709413e-05,
+      "loss": 1.1584,
+      "step": 149
+    },
+    {
+      "epoch": 0.11291992095605533,
+      "grad_norm": 3.862077116537031,
+      "learning_rate": 1.9642219606798566e-05,
+      "loss": 1.0728,
+      "step": 150
+    },
+    {
+      "epoch": 0.1136727204290957,
+      "grad_norm": 3.984219630530602,
+      "learning_rate": 1.9635724903175055e-05,
+      "loss": 1.1206,
+      "step": 151
+    },
+    {
+      "epoch": 0.11442551990213606,
+      "grad_norm": 3.669494644891549,
+      "learning_rate": 1.9629172873477995e-05,
+      "loss": 1.083,
+      "step": 152
+    },
+    {
+      "epoch": 0.11517831937517643,
+      "grad_norm": 5.42829688475088,
+      "learning_rate": 1.9622563556687545e-05,
+      "loss": 1.1588,
+      "step": 153
+    },
+    {
+      "epoch": 0.1159311188482168,
+      "grad_norm": 4.79232246776243,
+      "learning_rate": 1.961589699212469e-05,
+      "loss": 1.2102,
+      "step": 154
+    },
+    {
+      "epoch": 0.11668391832125717,
+      "grad_norm": 11.405826556390412,
+      "learning_rate": 1.9609173219450998e-05,
+      "loss": 1.109,
+      "step": 155
+    },
+    {
+      "epoch": 0.11743671779429754,
+      "grad_norm": 5.458296218745196,
+      "learning_rate": 1.960239227866839e-05,
+      "loss": 1.0762,
+      "step": 156
+    },
+    {
+      "epoch": 0.11818951726733791,
+      "grad_norm": 3.7544350915582854,
+      "learning_rate": 1.9595554210118896e-05,
+      "loss": 1.1603,
+      "step": 157
+    },
+    {
+      "epoch": 0.11894231674037828,
+      "grad_norm": 22.120555226575323,
+      "learning_rate": 1.9588659054484417e-05,
+      "loss": 1.0958,
+      "step": 158
+    },
+    {
+      "epoch": 0.11969511621341865,
+      "grad_norm": 4.148885042117614,
+      "learning_rate": 1.9581706852786492e-05,
+      "loss": 1.1157,
+      "step": 159
+    },
+    {
+      "epoch": 0.12044791568645902,
+      "grad_norm": 3.9732316992633328,
+      "learning_rate": 1.9574697646386027e-05,
+      "loss": 1.1421,
+      "step": 160
+    },
+    {
+      "epoch": 0.12120071515949939,
+      "grad_norm": 3.4050988239013296,
+      "learning_rate": 1.9567631476983088e-05,
+      "loss": 1.1144,
+      "step": 161
+    },
+    {
+      "epoch": 0.12195351463253976,
+      "grad_norm": 3.6985764801817465,
+      "learning_rate": 1.9560508386616624e-05,
+      "loss": 1.0498,
+      "step": 162
+    },
+    {
+      "epoch": 0.12270631410558012,
+      "grad_norm": 3.8528386734401434,
+      "learning_rate": 1.9553328417664223e-05,
+      "loss": 1.0844,
+      "step": 163
+    },
+    {
+      "epoch": 0.1234591135786205,
+      "grad_norm": 3.2819992244778184,
+      "learning_rate": 1.954609161284186e-05,
+      "loss": 1.1187,
+      "step": 164
+    },
+    {
+      "epoch": 0.12421191305166086,
+      "grad_norm": 3.22953606650776,
+      "learning_rate": 1.953879801520366e-05,
+      "loss": 1.0609,
+      "step": 165
+    },
+    {
+      "epoch": 0.12496471252470123,
+      "grad_norm": 6.494288008441105,
+      "learning_rate": 1.953144766814161e-05,
+      "loss": 1.0842,
+      "step": 166
+    },
+    {
+      "epoch": 0.1257175119977416,
+      "grad_norm": 4.888963654063504,
+      "learning_rate": 1.9524040615385324e-05,
+      "loss": 1.1902,
+      "step": 167
+    },
+    {
+      "epoch": 0.12647031147078197,
+      "grad_norm": 3.6311602708602257,
+      "learning_rate": 1.951657690100178e-05,
+      "loss": 1.0839,
+      "step": 168
+    },
+    {
+      "epoch": 0.12722311094382233,
+      "grad_norm": 4.00682080860383,
+      "learning_rate": 1.950905656939505e-05,
+      "loss": 1.1657,
+      "step": 169
+    },
+    {
+      "epoch": 0.1279759104168627,
+      "grad_norm": 3.890665804371254,
+      "learning_rate": 1.9501479665306046e-05,
+      "loss": 1.052,
+      "step": 170
+    },
+    {
+      "epoch": 0.12872870988990306,
+      "grad_norm": 3.4307975760371585,
+      "learning_rate": 1.949384623381224e-05,
+      "loss": 1.0919,
+      "step": 171
+    },
+    {
+      "epoch": 0.12948150936294345,
+      "grad_norm": 4.0455339119695655,
+      "learning_rate": 1.9486156320327406e-05,
+      "loss": 1.1103,
+      "step": 172
+    },
+    {
+      "epoch": 0.1302343088359838,
+      "grad_norm": 3.4165115341264296,
+      "learning_rate": 1.947840997060136e-05,
+      "loss": 1.0946,
+      "step": 173
+    },
+    {
+      "epoch": 0.13098710830902419,
+      "grad_norm": 4.65681514038902,
+      "learning_rate": 1.9470607230719654e-05,
+      "loss": 1.1848,
+      "step": 174
+    },
+    {
+      "epoch": 0.13173990778206454,
+      "grad_norm": 3.9568518534057193,
+      "learning_rate": 1.9462748147103342e-05,
+      "loss": 1.1567,
+      "step": 175
+    },
+    {
+      "epoch": 0.13249270725510492,
+      "grad_norm": 4.955380650232782,
+      "learning_rate": 1.945483276650868e-05,
+      "loss": 1.2039,
+      "step": 176
+    },
+    {
+      "epoch": 0.13324550672814528,
+      "grad_norm": 3.9264372250335158,
+      "learning_rate": 1.9446861136026846e-05,
+      "loss": 1.04,
+      "step": 177
+    },
+    {
+      "epoch": 0.13399830620118566,
+      "grad_norm": 3.405903263698621,
+      "learning_rate": 1.9438833303083677e-05,
+      "loss": 1.058,
+      "step": 178
+    },
+    {
+      "epoch": 0.13475110567422602,
+      "grad_norm": 3.9674907430691544,
+      "learning_rate": 1.943074931543937e-05,
+      "loss": 1.1056,
+      "step": 179
+    },
+    {
+      "epoch": 0.1355039051472664,
+      "grad_norm": 3.5452086691685096,
+      "learning_rate": 1.9422609221188208e-05,
+      "loss": 1.0868,
+      "step": 180
+    },
+    {
+      "epoch": 0.13625670462030676,
+      "grad_norm": 7.650572420624207,
+      "learning_rate": 1.9414413068758266e-05,
+      "loss": 1.1428,
+      "step": 181
+    },
+    {
+      "epoch": 0.13700950409334714,
+      "grad_norm": 3.999012518334001,
+      "learning_rate": 1.9406160906911137e-05,
+      "loss": 1.1722,
+      "step": 182
+    },
+    {
+      "epoch": 0.1377623035663875,
+      "grad_norm": 3.038212939029898,
+      "learning_rate": 1.939785278474162e-05,
+      "loss": 1.1286,
+      "step": 183
+    },
+    {
+      "epoch": 0.13851510303942788,
+      "grad_norm": 3.906578885384695,
+      "learning_rate": 1.938948875167745e-05,
+      "loss": 1.1803,
+      "step": 184
+    },
+    {
+      "epoch": 0.13926790251246823,
+      "grad_norm": 3.3509854123706293,
+      "learning_rate": 1.9381068857478994e-05,
+      "loss": 1.0973,
+      "step": 185
+    },
+    {
+      "epoch": 0.14002070198550862,
+      "grad_norm": 4.003782189974324,
+      "learning_rate": 1.937259315223894e-05,
+      "loss": 1.0838,
+      "step": 186
+    },
+    {
+      "epoch": 0.14077350145854897,
+      "grad_norm": 3.8245109394744583,
+      "learning_rate": 1.9364061686382042e-05,
+      "loss": 1.1094,
+      "step": 187
+    },
+    {
+      "epoch": 0.14152630093158935,
+      "grad_norm": 3.013745925688022,
+      "learning_rate": 1.9355474510664763e-05,
+      "loss": 1.0676,
+      "step": 188
+    },
+    {
+      "epoch": 0.1422791004046297,
+      "grad_norm": 4.3642368726250425,
+      "learning_rate": 1.934683167617502e-05,
+      "loss": 1.1752,
+      "step": 189
+    },
+    {
+      "epoch": 0.1430318998776701,
+      "grad_norm": 4.122723903005718,
+      "learning_rate": 1.933813323433186e-05,
+      "loss": 1.1305,
+      "step": 190
+    },
+    {
+      "epoch": 0.14378469935071045,
+      "grad_norm": 4.253380654597697,
+      "learning_rate": 1.9329379236885145e-05,
+      "loss": 1.0763,
+      "step": 191
+    },
+    {
+      "epoch": 0.14453749882375083,
+      "grad_norm": 4.458868784252592,
+      "learning_rate": 1.9320569735915273e-05,
+      "loss": 1.0436,
+      "step": 192
+    },
+    {
+      "epoch": 0.1452902982967912,
+      "grad_norm": 3.901071568651419,
+      "learning_rate": 1.9311704783832835e-05,
+      "loss": 1.1949,
+      "step": 193
+    },
+    {
+      "epoch": 0.14604309776983157,
+      "grad_norm": 3.129172072359172,
+      "learning_rate": 1.9302784433378333e-05,
+      "loss": 1.1172,
+      "step": 194
+    },
+    {
+      "epoch": 0.14679589724287193,
+      "grad_norm": 3.8443786959818045,
+      "learning_rate": 1.9293808737621837e-05,
+      "loss": 1.2054,
+      "step": 195
+    },
+    {
+      "epoch": 0.1475486967159123,
+      "grad_norm": 6.00072077234027,
+      "learning_rate": 1.9284777749962696e-05,
+      "loss": 1.1121,
+      "step": 196
+    },
+    {
+      "epoch": 0.14830149618895266,
+      "grad_norm": 3.7432081921844773,
+      "learning_rate": 1.9275691524129203e-05,
+      "loss": 1.0584,
+      "step": 197
+    },
+    {
+      "epoch": 0.14905429566199305,
+      "grad_norm": 3.4135705273299934,
+      "learning_rate": 1.926655011417828e-05,
+      "loss": 1.0901,
+      "step": 198
+    },
+    {
+      "epoch": 0.1498070951350334,
+      "grad_norm": 3.4177123004615817,
+      "learning_rate": 1.9257353574495164e-05,
+      "loss": 1.1038,
+      "step": 199
+    },
+    {
+      "epoch": 0.15055989460807379,
+      "grad_norm": 3.0832234189207504,
+      "learning_rate": 1.9248101959793066e-05,
+      "loss": 1.0317,
+      "step": 200
+    },
+    {
+      "epoch": 0.15131269408111414,
+      "grad_norm": 3.291295738727208,
+      "learning_rate": 1.9238795325112867e-05,
+      "loss": 1.2094,
+      "step": 201
+    },
+    {
+      "epoch": 0.15206549355415452,
+      "grad_norm": 3.4703930043192,
+      "learning_rate": 1.9229433725822776e-05,
+      "loss": 1.1485,
+      "step": 202
+    },
+    {
+      "epoch": 0.15281829302719488,
+      "grad_norm": 3.927008914244205,
+      "learning_rate": 1.9220017217618006e-05,
+      "loss": 1.0951,
+      "step": 203
+    },
+    {
+      "epoch": 0.15357109250023526,
+      "grad_norm": 3.346783632612943,
+      "learning_rate": 1.921054585652043e-05,
+      "loss": 1.0405,
+      "step": 204
+    },
+    {
+      "epoch": 0.15432389197327562,
+      "grad_norm": 3.732718718454638,
+      "learning_rate": 1.9201019698878272e-05,
+      "loss": 1.072,
+      "step": 205
+    },
+    {
+      "epoch": 0.155076691446316,
+      "grad_norm": 2.8305022942366502,
+      "learning_rate": 1.9191438801365763e-05,
+      "loss": 1.047,
+      "step": 206
+    },
+    {
+      "epoch": 0.15582949091935636,
+      "grad_norm": 3.470154225042611,
+      "learning_rate": 1.9181803220982776e-05,
+      "loss": 1.1342,
+      "step": 207
+    },
+    {
+      "epoch": 0.15658229039239674,
+      "grad_norm": 3.2370363710969032,
+      "learning_rate": 1.917211301505453e-05,
+      "loss": 1.0424,
+      "step": 208
+    },
+    {
+      "epoch": 0.1573350898654371,
+      "grad_norm": 4.036275798333279,
+      "learning_rate": 1.916236824123123e-05,
+      "loss": 1.1305,
+      "step": 209
+    },
+    {
+      "epoch": 0.15808788933847745,
+      "grad_norm": 3.105762174117261,
+      "learning_rate": 1.915256895748771e-05,
+      "loss": 1.0866,
+      "step": 210
+    },
+    {
+      "epoch": 0.15884068881151783,
+      "grad_norm": 3.241485339921148,
+      "learning_rate": 1.914271522212311e-05,
+      "loss": 1.0438,
+      "step": 211
+    },
+    {
+      "epoch": 0.1595934882845582,
+      "grad_norm": 3.7672919645423084,
+      "learning_rate": 1.9132807093760523e-05,
+      "loss": 1.0428,
+      "step": 212
+    },
+    {
+      "epoch": 0.16034628775759857,
+      "grad_norm": 3.0743803531471614,
+      "learning_rate": 1.9122844631346632e-05,
+      "loss": 1.1082,
+      "step": 213
+    },
+    {
+      "epoch": 0.16109908723063893,
+      "grad_norm": 4.63250823250084,
+      "learning_rate": 1.9112827894151386e-05,
+      "loss": 1.0954,
+      "step": 214
+    },
+    {
+      "epoch": 0.1618518867036793,
+      "grad_norm": 3.122767531926535,
+      "learning_rate": 1.9102756941767625e-05,
+      "loss": 1.0046,
+      "step": 215
+    },
+    {
+      "epoch": 0.16260468617671966,
+      "grad_norm": 3.268308565573597,
+      "learning_rate": 1.9092631834110723e-05,
+      "loss": 1.0834,
+      "step": 216
+    },
+    {
+      "epoch": 0.16335748564976005,
+      "grad_norm": 3.106344816610958,
+      "learning_rate": 1.9082452631418265e-05,
+      "loss": 1.0597,
+      "step": 217
+    },
+    {
+      "epoch": 0.1641102851228004,
+      "grad_norm": 2.894735322482825,
+      "learning_rate": 1.9072219394249644e-05,
+      "loss": 1.0627,
+      "step": 218
+    },
+    {
+      "epoch": 0.1648630845958408,
+      "grad_norm": 3.1750259084537324,
+      "learning_rate": 1.9061932183485726e-05,
+      "loss": 0.9958,
+      "step": 219
+    },
+    {
+      "epoch": 0.16561588406888114,
+      "grad_norm": 3.3713532561993897,
+      "learning_rate": 1.9051591060328496e-05,
+      "loss": 1.1432,
+      "step": 220
+    },
+    {
+      "epoch": 0.16636868354192152,
+      "grad_norm": 3.1214606959314875,
+      "learning_rate": 1.9041196086300666e-05,
+      "loss": 1.1035,
+      "step": 221
+    },
+    {
+      "epoch": 0.16712148301496188,
+      "grad_norm": 4.111611609451444,
+      "learning_rate": 1.903074732324533e-05,
+      "loss": 1.1142,
+      "step": 222
+    },
+    {
+      "epoch": 0.16787428248800226,
+      "grad_norm": 6.34744510464487,
+      "learning_rate": 1.902024483332559e-05,
+      "loss": 1.1981,
+      "step": 223
+    },
+    {
+      "epoch": 0.16862708196104262,
+      "grad_norm": 3.853858890566725,
+      "learning_rate": 1.900968867902419e-05,
+      "loss": 1.1083,
+      "step": 224
+    },
+    {
+      "epoch": 0.169379881434083,
+      "grad_norm": 3.0972363845249316,
+      "learning_rate": 1.8999078923143142e-05,
+      "loss": 1.1272,
+      "step": 225
+    },
+    {
+      "epoch": 0.17013268090712336,
+      "grad_norm": 6.697298803805668,
+      "learning_rate": 1.8988415628803345e-05,
+      "loss": 1.078,
+      "step": 226
+    },
+    {
+      "epoch": 0.17088548038016374,
+      "grad_norm": 3.320222370823791,
+      "learning_rate": 1.8977698859444217e-05,
+      "loss": 1.0549,
+      "step": 227
+    },
+    {
+      "epoch": 0.1716382798532041,
+      "grad_norm": 3.7206790432455006,
+      "learning_rate": 1.8966928678823317e-05,
+      "loss": 1.1347,
+      "step": 228
+    },
+    {
+      "epoch": 0.17239107932624448,
+      "grad_norm": 3.290103204055493,
+      "learning_rate": 1.8956105151015966e-05,
+      "loss": 1.0179,
+      "step": 229
+    },
+    {
+      "epoch": 0.17314387879928483,
+      "grad_norm": 3.1386966405501564,
+      "learning_rate": 1.894522834041487e-05,
+      "loss": 1.1059,
+      "step": 230
+    },
+    {
+      "epoch": 0.17389667827232522,
+      "grad_norm": 3.2706699814852875,
+      "learning_rate": 1.8934298311729728e-05,
+      "loss": 1.0172,
+      "step": 231
+    },
+    {
+      "epoch": 0.17464947774536557,
+      "grad_norm": 3.8269561638109355,
+      "learning_rate": 1.8923315129986838e-05,
+      "loss": 1.1137,
+      "step": 232
+    },
+    {
+      "epoch": 0.17540227721840596,
+      "grad_norm": 3.772999914244335,
+      "learning_rate": 1.8912278860528742e-05,
+      "loss": 1.1287,
+      "step": 233
+    },
+    {
+      "epoch": 0.1761550766914463,
+      "grad_norm": 5.414372276284227,
+      "learning_rate": 1.890118956901381e-05,
+      "loss": 1.1443,
+      "step": 234
+    },
+    {
+      "epoch": 0.1769078761644867,
+      "grad_norm": 3.263109591052546,
+      "learning_rate": 1.8890047321415856e-05,
+      "loss": 1.0893,
+      "step": 235
+    },
+    {
+      "epoch": 0.17766067563752705,
+      "grad_norm": 3.3607503214338856,
+      "learning_rate": 1.8878852184023754e-05,
+      "loss": 1.0175,
+      "step": 236
+    },
+    {
+      "epoch": 0.17841347511056743,
+      "grad_norm": 4.631748841741912,
+      "learning_rate": 1.8867604223441027e-05,
+      "loss": 1.1528,
+      "step": 237
+    },
+    {
+      "epoch": 0.1791662745836078,
+      "grad_norm": 3.7210788472140837,
+      "learning_rate": 1.885630350658546e-05,
+      "loss": 1.1066,
+      "step": 238
+    },
+    {
+      "epoch": 0.17991907405664817,
+      "grad_norm": 3.1758770450597247,
+      "learning_rate": 1.884495010068872e-05,
+      "loss": 1.103,
+      "step": 239
+    },
+    {
+      "epoch": 0.18067187352968853,
+      "grad_norm": 3.4141865931283037,
+      "learning_rate": 1.8833544073295918e-05,
+      "loss": 1.091,
+      "step": 240
+    },
+    {
+      "epoch": 0.1814246730027289,
+      "grad_norm": 3.432096224130023,
+      "learning_rate": 1.8822085492265235e-05,
+      "loss": 1.1407,
+      "step": 241
+    },
+    {
+      "epoch": 0.18217747247576926,
+      "grad_norm": 3.3201010603309573,
+      "learning_rate": 1.8810574425767512e-05,
+      "loss": 1.1171,
+      "step": 242
+    },
+    {
+      "epoch": 0.18293027194880965,
+      "grad_norm": 3.142047941577179,
+      "learning_rate": 1.879901094228584e-05,
+      "loss": 1.0008,
+      "step": 243
+    },
+    {
+      "epoch": 0.18368307142185,
+      "grad_norm": 4.428870556760251,
+      "learning_rate": 1.8787395110615163e-05,
+      "loss": 1.1599,
+      "step": 244
+    },
+    {
+      "epoch": 0.18443587089489039,
+      "grad_norm": 3.42062935475352,
+      "learning_rate": 1.877572699986185e-05,
+      "loss": 1.0318,
+      "step": 245
+    },
+    {
+      "epoch": 0.18518867036793074,
+      "grad_norm": 2.532845389080576,
+      "learning_rate": 1.8764006679443306e-05,
+      "loss": 1.0513,
+      "step": 246
+    },
+    {
+      "epoch": 0.18594146984097112,
+      "grad_norm": 3.4902131357709663,
+      "learning_rate": 1.8752234219087538e-05,
+      "loss": 1.0851,
+      "step": 247
+    },
+    {
+      "epoch": 0.18669426931401148,
+      "grad_norm": 3.9584887752768396,
+      "learning_rate": 1.8740409688832762e-05,
+      "loss": 1.1399,
+      "step": 248
+    },
+    {
+      "epoch": 0.18744706878705186,
+      "grad_norm": 3.369982891463805,
+      "learning_rate": 1.8728533159026972e-05,
+      "loss": 1.182,
+      "step": 249
+    },
+    {
+      "epoch": 0.18819986826009222,
+      "grad_norm": 4.416503857133181,
+      "learning_rate": 1.8716604700327516e-05,
+      "loss": 1.0763,
+      "step": 250
+    },
+    {
+      "epoch": 0.18895266773313257,
+      "grad_norm": 2.7507954310960114,
+      "learning_rate": 1.8704624383700686e-05,
+      "loss": 1.1158,
+      "step": 251
+    },
+    {
+      "epoch": 0.18970546720617296,
+      "grad_norm": 3.1857061931174875,
+      "learning_rate": 1.8692592280421305e-05,
+      "loss": 1.1041,
+      "step": 252
+    },
+    {
+      "epoch": 0.1904582666792133,
+      "grad_norm": 4.04760329251056,
+      "learning_rate": 1.8680508462072282e-05,
+      "loss": 1.0831,
+      "step": 253
+    },
+    {
+      "epoch": 0.1912110661522537,
+      "grad_norm": 3.882827890480107,
+      "learning_rate": 1.8668373000544197e-05,
+      "loss": 1.1167,
+      "step": 254
+    },
+    {
+      "epoch": 0.19196386562529405,
+      "grad_norm": 3.287198676998183,
+      "learning_rate": 1.865618596803487e-05,
+      "loss": 1.0305,
+      "step": 255
+    },
+    {
+      "epoch": 0.19271666509833443,
+      "grad_norm": 3.604326797700127,
+      "learning_rate": 1.8643947437048944e-05,
+      "loss": 1.0032,
+      "step": 256
+    },
+    {
+      "epoch": 0.1934694645713748,
+      "grad_norm": 4.379872775610173,
+      "learning_rate": 1.863165748039743e-05,
+      "loss": 1.008,
+      "step": 257
+    },
+    {
+      "epoch": 0.19422226404441517,
+      "grad_norm": 3.369497382545173,
+      "learning_rate": 1.8619316171197292e-05,
+      "loss": 1.067,
+      "step": 258
+    },
+    {
+      "epoch": 0.19497506351745553,
+      "grad_norm": 4.120227014926847,
+      "learning_rate": 1.8606923582871007e-05,
+      "loss": 1.0974,
+      "step": 259
+    },
+    {
+      "epoch": 0.1957278629904959,
+      "grad_norm": 3.0494014526642608,
+      "learning_rate": 1.859447978914614e-05,
+      "loss": 1.0551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19648066246353627,
+      "grad_norm": 3.3910652042019724,
+      "learning_rate": 1.8581984864054866e-05,
+      "loss": 1.0932,
+      "step": 261
+    },
+    {
+      "epoch": 0.19723346193657665,
+      "grad_norm": 3.9039698571171138,
+      "learning_rate": 1.8569438881933587e-05,
+      "loss": 1.1064,
+      "step": 262
+    },
+    {
+      "epoch": 0.197986261409617,
+      "grad_norm": 3.268330588127502,
+      "learning_rate": 1.8556841917422443e-05,
+      "loss": 0.9784,
+      "step": 263
+    },
+    {
+      "epoch": 0.1987390608826574,
+      "grad_norm": 4.314068622976358,
+      "learning_rate": 1.8544194045464888e-05,
+      "loss": 1.0996,
+      "step": 264
+    },
+    {
+      "epoch": 0.19949186035569774,
+      "grad_norm": 3.1827098111819967,
+      "learning_rate": 1.853149534130724e-05,
+      "loss": 1.1221,
+      "step": 265
+    },
+    {
+      "epoch": 0.20024465982873813,
+      "grad_norm": 4.040702246051114,
+      "learning_rate": 1.8518745880498242e-05,
+      "loss": 1.0216,
+      "step": 266
+    },
+    {
+      "epoch": 0.20024465982873813,
+      "eval_loss": 1.0534603595733643,
+      "eval_runtime": 583.4736,
+      "eval_samples_per_second": 30.653,
+      "eval_steps_per_second": 0.48,
+      "step": 266
+    },
+    {
+      "epoch": 0.20099745930177848,
+      "grad_norm": 3.1584439216784266,
+      "learning_rate": 1.8505945738888593e-05,
+      "loss": 1.0341,
+      "step": 267
+    },
+    {
+      "epoch": 0.20175025877481886,
+      "grad_norm": 3.2691826063280294,
+      "learning_rate": 1.849309499263052e-05,
+      "loss": 1.14,
+      "step": 268
+    },
+    {
+      "epoch": 0.20250305824785922,
+      "grad_norm": 3.6898030714786794,
+      "learning_rate": 1.8480193718177305e-05,
+      "loss": 1.1653,
+      "step": 269
+    },
+    {
+      "epoch": 0.2032558577208996,
+      "grad_norm": 3.575806731417126,
+      "learning_rate": 1.8467241992282842e-05,
+      "loss": 1.1146,
+      "step": 270
+    },
+    {
+      "epoch": 0.20400865719393996,
+      "grad_norm": 3.1081441368569744,
+      "learning_rate": 1.845423989200118e-05,
+      "loss": 1.0278,
+      "step": 271
+    },
+    {
+      "epoch": 0.20476145666698034,
+      "grad_norm": 2.789678804862752,
+      "learning_rate": 1.8441187494686055e-05,
+      "loss": 1.0891,
+      "step": 272
+    },
+    {
+      "epoch": 0.2055142561400207,
+      "grad_norm": 3.985369579436919,
+      "learning_rate": 1.8428084877990443e-05,
+      "loss": 1.1268,
+      "step": 273
+    },
+    {
+      "epoch": 0.20626705561306108,
+      "grad_norm": 3.22661410744001,
+      "learning_rate": 1.841493211986609e-05,
+      "loss": 1.1072,
+      "step": 274
+    },
+    {
+      "epoch": 0.20701985508610143,
+      "grad_norm": 3.1835000593613123,
+      "learning_rate": 1.840172929856304e-05,
+      "loss": 1.0414,
+      "step": 275
+    },
+    {
+      "epoch": 0.20777265455914182,
+      "grad_norm": 3.305479541641861,
+      "learning_rate": 1.8388476492629198e-05,
+      "loss": 1.0443,
+      "step": 276
+    },
+    {
+      "epoch": 0.20852545403218217,
+      "grad_norm": 4.693809651108469,
+      "learning_rate": 1.837517378090983e-05,
+      "loss": 0.9937,
+      "step": 277
+    },
+    {
+      "epoch": 0.20927825350522256,
+      "grad_norm": 3.5123200132532566,
+      "learning_rate": 1.836182124254711e-05,
+      "loss": 1.0776,
+      "step": 278
+    },
+    {
+      "epoch": 0.2100310529782629,
+      "grad_norm": 3.3276868884741733,
+      "learning_rate": 1.834841895697965e-05,
+      "loss": 1.0845,
+      "step": 279
+    },
+    {
+      "epoch": 0.2107838524513033,
+      "grad_norm": 3.133629495172562,
+      "learning_rate": 1.833496700394202e-05,
+      "loss": 1.1334,
+      "step": 280
+    },
+    {
+      "epoch": 0.21153665192434365,
+      "grad_norm": 2.7777396926997957,
+      "learning_rate": 1.8321465463464287e-05,
+      "loss": 1.0504,
+      "step": 281
+    },
+    {
+      "epoch": 0.21228945139738403,
+      "grad_norm": 2.727044505380872,
+      "learning_rate": 1.8307914415871516e-05,
+      "loss": 0.9811,
+      "step": 282
+    },
+    {
+      "epoch": 0.2130422508704244,
+      "grad_norm": 3.9758603746678594,
+      "learning_rate": 1.829431394178332e-05,
+      "loss": 0.9935,
+      "step": 283
+    },
+    {
+      "epoch": 0.21379505034346477,
+      "grad_norm": 3.5295194163983625,
+      "learning_rate": 1.8280664122113356e-05,
+      "loss": 1.0375,
+      "step": 284
+    },
+    {
+      "epoch": 0.21454784981650513,
+      "grad_norm": 3.774125316395598,
+      "learning_rate": 1.8266965038068856e-05,
+      "loss": 1.1335,
+      "step": 285
+    },
+    {
+      "epoch": 0.2153006492895455,
+      "grad_norm": 4.651024677980998,
+      "learning_rate": 1.8253216771150153e-05,
+      "loss": 1.0207,
+      "step": 286
+    },
+    {
+      "epoch": 0.21605344876258586,
+      "grad_norm": 3.099286878703337,
+      "learning_rate": 1.823941940315017e-05,
+      "loss": 1.0212,
+      "step": 287
+    },
+    {
+      "epoch": 0.21680624823562625,
+      "grad_norm": 3.0097671512838096,
+      "learning_rate": 1.8225573016153945e-05,
+      "loss": 1.0461,
+      "step": 288
+    },
+    {
+      "epoch": 0.2175590477086666,
+      "grad_norm": 2.973750973417238,
+      "learning_rate": 1.821167769253817e-05,
+      "loss": 1.0388,
+      "step": 289
+    },
+    {
+      "epoch": 0.21831184718170699,
+      "grad_norm": 2.967274178762896,
+      "learning_rate": 1.8197733514970655e-05,
+      "loss": 1.0564,
+      "step": 290
+    },
+    {
+      "epoch": 0.21906464665474734,
+      "grad_norm": 2.5138035874146327,
+      "learning_rate": 1.8183740566409867e-05,
+      "loss": 1.073,
+      "step": 291
+    },
+    {
+      "epoch": 0.2198174461277877,
+      "grad_norm": 3.7001058284729433,
+      "learning_rate": 1.816969893010442e-05,
+      "loss": 1.2111,
+      "step": 292
+    },
+    {
+      "epoch": 0.22057024560082808,
+      "grad_norm": 3.5282758941825456,
+      "learning_rate": 1.8155608689592604e-05,
+      "loss": 1.183,
+      "step": 293
+    },
+    {
+      "epoch": 0.22132304507386844,
+      "grad_norm": 3.312941108339452,
+      "learning_rate": 1.8141469928701852e-05,
+      "loss": 1.0349,
+      "step": 294
+    },
+    {
+      "epoch": 0.22207584454690882,
+      "grad_norm": 2.9008531856700417,
+      "learning_rate": 1.812728273154827e-05,
+      "loss": 1.0123,
+      "step": 295
+    },
+    {
+      "epoch": 0.22282864401994917,
+      "grad_norm": 3.55932688032634,
+      "learning_rate": 1.8113047182536128e-05,
+      "loss": 1.1382,
+      "step": 296
+    },
+    {
+      "epoch": 0.22358144349298956,
+      "grad_norm": 4.056235257189877,
+      "learning_rate": 1.8098763366357354e-05,
+      "loss": 1.0822,
+      "step": 297
+    },
+    {
+      "epoch": 0.2243342429660299,
+      "grad_norm": 4.772955312310337,
+      "learning_rate": 1.8084431367991032e-05,
+      "loss": 1.1983,
+      "step": 298
+    },
+    {
+      "epoch": 0.2250870424390703,
+      "grad_norm": 4.2810326932753995,
+      "learning_rate": 1.8070051272702905e-05,
+      "loss": 1.1598,
+      "step": 299
+    },
+    {
+      "epoch": 0.22583984191211065,
+      "grad_norm": 3.7796974809301473,
+      "learning_rate": 1.8055623166044855e-05,
+      "loss": 1.1074,
+      "step": 300
+    },
+    {
+      "epoch": 0.22659264138515103,
+      "grad_norm": 3.6457585818347304,
+      "learning_rate": 1.804114713385439e-05,
+      "loss": 1.1222,
+      "step": 301
+    },
+    {
+      "epoch": 0.2273454408581914,
+      "grad_norm": 3.453904748558131,
+      "learning_rate": 1.8026623262254164e-05,
+      "loss": 1.1338,
+      "step": 302
+    },
+    {
+      "epoch": 0.22809824033123177,
+      "grad_norm": 3.532636263927634,
+      "learning_rate": 1.8012051637651423e-05,
+      "loss": 1.059,
+      "step": 303
+    },
+    {
+      "epoch": 0.22885103980427213,
+      "grad_norm": 3.9412069933016403,
+      "learning_rate": 1.7997432346737524e-05,
+      "loss": 1.0612,
+      "step": 304
+    },
+    {
+      "epoch": 0.2296038392773125,
+      "grad_norm": 3.3027808930961835,
+      "learning_rate": 1.7982765476487398e-05,
+      "loss": 1.1245,
+      "step": 305
+    },
+    {
+      "epoch": 0.23035663875035287,
+      "grad_norm": 3.36315843459813,
+      "learning_rate": 1.7968051114159046e-05,
+      "loss": 1.0142,
+      "step": 306
+    },
+    {
+      "epoch": 0.23110943822339325,
+      "grad_norm": 3.4294122694976426,
+      "learning_rate": 1.795328934729302e-05,
+      "loss": 1.0978,
+      "step": 307
+    },
+    {
+      "epoch": 0.2318622376964336,
+      "grad_norm": 3.09886747290162,
+      "learning_rate": 1.793848026371188e-05,
+      "loss": 1.0259,
+      "step": 308
+    },
+    {
+      "epoch": 0.232615037169474,
+      "grad_norm": 3.802369083047949,
+      "learning_rate": 1.7923623951519708e-05,
+      "loss": 1.0032,
+      "step": 309
+    },
+    {
+      "epoch": 0.23336783664251434,
+      "grad_norm": 3.459231482580547,
+      "learning_rate": 1.7908720499101552e-05,
+      "loss": 1.1666,
+      "step": 310
+    },
+    {
+      "epoch": 0.23412063611555473,
+      "grad_norm": 2.642651982530993,
+      "learning_rate": 1.7893769995122916e-05,
+      "loss": 1.084,
+      "step": 311
+    },
+    {
+      "epoch": 0.23487343558859508,
+      "grad_norm": 2.9773440456665208,
+      "learning_rate": 1.7878772528529232e-05,
+      "loss": 1.0423,
+      "step": 312
+    },
+    {
+      "epoch": 0.23562623506163546,
+      "grad_norm": 3.154468928265403,
+      "learning_rate": 1.7863728188545326e-05,
+      "loss": 1.1134,
+      "step": 313
+    },
+    {
+      "epoch": 0.23637903453467582,
+      "grad_norm": 3.7877452098752937,
+      "learning_rate": 1.7848637064674887e-05,
+      "loss": 1.1711,
+      "step": 314
+    },
+    {
+      "epoch": 0.2371318340077162,
+      "grad_norm": 3.378663421824846,
+      "learning_rate": 1.783349924669994e-05,
+      "loss": 1.0739,
+      "step": 315
+    },
+    {
+      "epoch": 0.23788463348075656,
+      "grad_norm": 4.410599739962371,
+      "learning_rate": 1.78183148246803e-05,
+      "loss": 1.0029,
+      "step": 316
+    },
+    {
+      "epoch": 0.23863743295379694,
+      "grad_norm": 3.6348238736424157,
+      "learning_rate": 1.7803083888953058e-05,
+      "loss": 1.0898,
+      "step": 317
+    },
+    {
+      "epoch": 0.2393902324268373,
+      "grad_norm": 2.957533946262562,
+      "learning_rate": 1.7787806530132022e-05,
+      "loss": 1.1463,
+      "step": 318
+    },
+    {
+      "epoch": 0.24014303189987768,
+      "grad_norm": 6.23103586941611,
+      "learning_rate": 1.777248283910719e-05,
+      "loss": 1.1414,
+      "step": 319
+    },
+    {
+      "epoch": 0.24089583137291803,
+      "grad_norm": 2.865456649973816,
+      "learning_rate": 1.77571129070442e-05,
+      "loss": 1.0555,
+      "step": 320
+    },
+    {
+      "epoch": 0.24164863084595842,
+      "grad_norm": 2.978113549024132,
+      "learning_rate": 1.7741696825383797e-05,
+      "loss": 1.1657,
+      "step": 321
+    },
+    {
+      "epoch": 0.24240143031899877,
+      "grad_norm": 3.6676086137258777,
+      "learning_rate": 1.7726234685841283e-05,
+      "loss": 1.1312,
+      "step": 322
+    },
+    {
+      "epoch": 0.24315422979203916,
+      "grad_norm": 3.8360009679535043,
+      "learning_rate": 1.7710726580405977e-05,
+      "loss": 1.1094,
+      "step": 323
+    },
+    {
+      "epoch": 0.2439070292650795,
+      "grad_norm": 4.893214293222856,
+      "learning_rate": 1.769517260134066e-05,
+      "loss": 1.1595,
+      "step": 324
+    },
+    {
+      "epoch": 0.2446598287381199,
+      "grad_norm": 2.9775759972530706,
+      "learning_rate": 1.7679572841181033e-05,
+      "loss": 1.0488,
+      "step": 325
+    },
+    {
+      "epoch": 0.24541262821116025,
+      "grad_norm": 3.2211167234681897,
+      "learning_rate": 1.766392739273516e-05,
+      "loss": 1.0147,
+      "step": 326
+    },
+    {
+      "epoch": 0.24616542768420063,
+      "grad_norm": 2.8555966297737934,
+      "learning_rate": 1.7648236349082928e-05,
+      "loss": 1.0554,
+      "step": 327
+    },
+    {
+      "epoch": 0.246918227157241,
+      "grad_norm": 3.001112904305115,
+      "learning_rate": 1.7632499803575473e-05,
+      "loss": 1.1109,
+      "step": 328
+    },
+    {
+      "epoch": 0.24767102663028137,
+      "grad_norm": 3.79725157083872,
+      "learning_rate": 1.7616717849834644e-05,
+      "loss": 1.0736,
+      "step": 329
+    },
+    {
+      "epoch": 0.24842382610332173,
+      "grad_norm": 3.1372105798656764,
+      "learning_rate": 1.7600890581752435e-05,
+      "loss": 1.08,
+      "step": 330
+    },
+    {
+      "epoch": 0.2491766255763621,
+      "grad_norm": 3.0523824969950866,
+      "learning_rate": 1.758501809349044e-05,
+      "loss": 1.0516,
+      "step": 331
+    },
+    {
+      "epoch": 0.24992942504940246,
+      "grad_norm": 3.090928065217594,
+      "learning_rate": 1.756910047947926e-05,
+      "loss": 0.967,
+      "step": 332
+    },
+    {
+      "epoch": 0.2506822245224428,
+      "grad_norm": 3.7066771841864985,
+      "learning_rate": 1.755313783441799e-05,
+      "loss": 1.0556,
+      "step": 333
+    },
+    {
+      "epoch": 0.2514350239954832,
+      "grad_norm": 3.0285861591388934,
+      "learning_rate": 1.7537130253273613e-05,
+      "loss": 1.1683,
+      "step": 334
+    },
+    {
+      "epoch": 0.2521878234685236,
+      "grad_norm": 2.80325314926076,
+      "learning_rate": 1.7521077831280453e-05,
+      "loss": 0.9991,
+      "step": 335
+    },
+    {
+      "epoch": 0.25294062294156394,
+      "grad_norm": 2.995638992681817,
+      "learning_rate": 1.7504980663939614e-05,
+      "loss": 1.0163,
+      "step": 336
+    },
+    {
+      "epoch": 0.2536934224146043,
+      "grad_norm": 3.139175462481918,
+      "learning_rate": 1.7488838847018397e-05,
+      "loss": 1.1129,
+      "step": 337
+    },
+    {
+      "epoch": 0.25444622188764465,
+      "grad_norm": 2.9364998673824485,
+      "learning_rate": 1.7472652476549747e-05,
+      "loss": 1.0008,
+      "step": 338
+    },
+    {
+      "epoch": 0.25519902136068506,
+      "grad_norm": 3.3899427096289974,
+      "learning_rate": 1.7456421648831658e-05,
+      "loss": 1.0741,
+      "step": 339
+    },
+    {
+      "epoch": 0.2559518208337254,
+      "grad_norm": 3.333680754858661,
+      "learning_rate": 1.744014646042663e-05,
+      "loss": 1.0719,
+      "step": 340
+    },
+    {
+      "epoch": 0.2567046203067658,
+      "grad_norm": 3.6173353051432273,
+      "learning_rate": 1.742382700816107e-05,
+      "loss": 1.0938,
+      "step": 341
+    },
+    {
+      "epoch": 0.25745741977980613,
+      "grad_norm": 3.43700516947627,
+      "learning_rate": 1.7407463389124728e-05,
+      "loss": 1.1011,
+      "step": 342
+    },
+    {
+      "epoch": 0.25821021925284654,
+      "grad_norm": 7.1179022156761365,
+      "learning_rate": 1.739105570067012e-05,
+      "loss": 1.1076,
+      "step": 343
+    },
+    {
+      "epoch": 0.2589630187258869,
+      "grad_norm": 3.6490493958884342,
+      "learning_rate": 1.7374604040411934e-05,
+      "loss": 0.9959,
+      "step": 344
+    },
+    {
+      "epoch": 0.25971581819892725,
+      "grad_norm": 3.990073166624143,
+      "learning_rate": 1.7358108506226477e-05,
+      "loss": 1.0756,
+      "step": 345
+    },
+    {
+      "epoch": 0.2604686176719676,
+      "grad_norm": 3.7673869409287044,
+      "learning_rate": 1.7341569196251065e-05,
+      "loss": 1.0857,
+      "step": 346
+    },
+    {
+      "epoch": 0.261221417145008,
+      "grad_norm": 3.830082557723633,
+      "learning_rate": 1.732498620888345e-05,
+      "loss": 1.0964,
+      "step": 347
+    },
+    {
+      "epoch": 0.26197421661804837,
+      "grad_norm": 7.568910658413351,
+      "learning_rate": 1.730835964278124e-05,
+      "loss": 1.0863,
+      "step": 348
+    },
+    {
+      "epoch": 0.2627270160910887,
+      "grad_norm": 5.225358882803191,
+      "learning_rate": 1.729168959686131e-05,
+      "loss": 1.1107,
+      "step": 349
+    },
+    {
+      "epoch": 0.2634798155641291,
+      "grad_norm": 3.3550233061746355,
+      "learning_rate": 1.7274976170299197e-05,
+      "loss": 1.1369,
+      "step": 350
+    },
+    {
+      "epoch": 0.2642326150371695,
+      "grad_norm": 3.5292404069420114,
+      "learning_rate": 1.7258219462528543e-05,
+      "loss": 1.0442,
+      "step": 351
+    },
+    {
+      "epoch": 0.26498541451020985,
+      "grad_norm": 4.527087799400158,
+      "learning_rate": 1.7241419573240463e-05,
+      "loss": 1.1139,
+      "step": 352
+    },
+    {
+      "epoch": 0.2657382139832502,
+      "grad_norm": 4.109247471170187,
+      "learning_rate": 1.7224576602382993e-05,
+      "loss": 1.1061,
+      "step": 353
+    },
+    {
+      "epoch": 0.26649101345629056,
+      "grad_norm": 3.2870155559806946,
+      "learning_rate": 1.720769065016046e-05,
+      "loss": 1.0581,
+      "step": 354
+    },
+    {
+      "epoch": 0.26724381292933097,
+      "grad_norm": 3.779535153125096,
+      "learning_rate": 1.719076181703291e-05,
+      "loss": 1.066,
+      "step": 355
+    },
+    {
+      "epoch": 0.2679966124023713,
+      "grad_norm": 5.005623870979084,
+      "learning_rate": 1.7173790203715494e-05,
+      "loss": 1.0451,
+      "step": 356
+    },
+    {
+      "epoch": 0.2687494118754117,
+      "grad_norm": 3.93851827850133,
+      "learning_rate": 1.7156775911177888e-05,
+      "loss": 1.0904,
+      "step": 357
+    },
+    {
+      "epoch": 0.26950221134845204,
+      "grad_norm": 4.547113741305462,
+      "learning_rate": 1.713971904064367e-05,
+      "loss": 1.113,
+      "step": 358
+    },
+    {
+      "epoch": 0.27025501082149245,
+      "grad_norm": 3.7171211560510686,
+      "learning_rate": 1.712261969358973e-05,
+      "loss": 1.0793,
+      "step": 359
+    },
+    {
+      "epoch": 0.2710078102945328,
+      "grad_norm": 3.5934038171786766,
+      "learning_rate": 1.7105477971745668e-05,
+      "loss": 1.004,
+      "step": 360
+    },
+    {
+      "epoch": 0.27176060976757316,
+      "grad_norm": 4.1748700140242985,
+      "learning_rate": 1.7088293977093187e-05,
+      "loss": 1.1683,
+      "step": 361
+    },
+    {
+      "epoch": 0.2725134092406135,
+      "grad_norm": 3.435875864417095,
+      "learning_rate": 1.7071067811865477e-05,
+      "loss": 1.0426,
+      "step": 362
+    },
+    {
+      "epoch": 0.2732662087136539,
+      "grad_norm": 3.4776749127597126,
+      "learning_rate": 1.7053799578546623e-05,
+      "loss": 1.0401,
+      "step": 363
+    },
+    {
+      "epoch": 0.2740190081866943,
+      "grad_norm": 3.2901799263911826,
+      "learning_rate": 1.7036489379870982e-05,
+      "loss": 1.115,
+      "step": 364
+    },
+    {
+      "epoch": 0.27477180765973463,
+      "grad_norm": 3.2846213361675645,
+      "learning_rate": 1.7019137318822577e-05,
+      "loss": 1.0098,
+      "step": 365
+    },
+    {
+      "epoch": 0.275524607132775,
+      "grad_norm": 3.5406251727492313,
+      "learning_rate": 1.7001743498634487e-05,
+      "loss": 1.0518,
+      "step": 366
+    },
+    {
+      "epoch": 0.2762774066058154,
+      "grad_norm": 4.997896297281951,
+      "learning_rate": 1.6984308022788227e-05,
+      "loss": 1.1053,
+      "step": 367
+    },
+    {
+      "epoch": 0.27703020607885576,
+      "grad_norm": 3.306159931490668,
+      "learning_rate": 1.6966830995013134e-05,
+      "loss": 1.0413,
+      "step": 368
+    },
+    {
+      "epoch": 0.2777830055518961,
+      "grad_norm": 3.5786298411885014,
+      "learning_rate": 1.694931251928575e-05,
+      "loss": 1.0284,
+      "step": 369
+    },
+    {
+      "epoch": 0.27853580502493647,
+      "grad_norm": 3.4314746136554755,
+      "learning_rate": 1.693175269982921e-05,
+      "loss": 1.0704,
+      "step": 370
+    },
+    {
+      "epoch": 0.2792886044979769,
+      "grad_norm": 3.802313145942647,
+      "learning_rate": 1.691415164111261e-05,
+      "loss": 1.0687,
+      "step": 371
+    },
+    {
+      "epoch": 0.28004140397101723,
+      "grad_norm": 3.8384682286598952,
+      "learning_rate": 1.689650944785041e-05,
+      "loss": 1.0777,
+      "step": 372
+    },
+    {
+      "epoch": 0.2807942034440576,
+      "grad_norm": 3.1221086188163487,
+      "learning_rate": 1.6878826225001756e-05,
+      "loss": 1.0765,
+      "step": 373
+    },
+    {
+      "epoch": 0.28154700291709794,
+      "grad_norm": 3.189592999644019,
+      "learning_rate": 1.686110207776993e-05,
+      "loss": 1.0747,
+      "step": 374
+    },
+    {
+      "epoch": 0.28229980239013835,
+      "grad_norm": 3.3796395188906714,
+      "learning_rate": 1.6843337111601663e-05,
+      "loss": 1.0966,
+      "step": 375
+    },
+    {
+      "epoch": 0.2830526018631787,
+      "grad_norm": 3.620502370043677,
+      "learning_rate": 1.6825531432186545e-05,
+      "loss": 1.0919,
+      "step": 376
+    },
+    {
+      "epoch": 0.28380540133621907,
+      "grad_norm": 3.342937529880184,
+      "learning_rate": 1.680768514545637e-05,
+      "loss": 1.0405,
+      "step": 377
+    },
+    {
+      "epoch": 0.2845582008092594,
+      "grad_norm": 3.44340611172516,
+      "learning_rate": 1.6789798357584524e-05,
+      "loss": 0.9996,
+      "step": 378
+    },
+    {
+      "epoch": 0.2853110002822998,
+      "grad_norm": 3.015178467631497,
+      "learning_rate": 1.677187117498536e-05,
+      "loss": 1.0534,
+      "step": 379
+    },
+    {
+      "epoch": 0.2860637997553402,
+      "grad_norm": 2.8540969925658084,
+      "learning_rate": 1.6753903704313527e-05,
+      "loss": 1.0674,
+      "step": 380
+    },
+    {
+      "epoch": 0.28681659922838054,
+      "grad_norm": 3.2861357016852284,
+      "learning_rate": 1.6735896052463384e-05,
+      "loss": 1.0382,
+      "step": 381
+    },
+    {
+      "epoch": 0.2875693987014209,
+      "grad_norm": 3.0596062908393047,
+      "learning_rate": 1.6717848326568327e-05,
+      "loss": 1.0231,
+      "step": 382
+    },
+    {
+      "epoch": 0.28832219817446125,
+      "grad_norm": 3.0906561170405005,
+      "learning_rate": 1.6699760634000166e-05,
+      "loss": 1.0676,
+      "step": 383
+    },
+    {
+      "epoch": 0.28907499764750166,
+      "grad_norm": 2.3789151459015434,
+      "learning_rate": 1.66816330823685e-05,
+      "loss": 1.0296,
+      "step": 384
+    },
+    {
+      "epoch": 0.289827797120542,
+      "grad_norm": 3.8593539388223843,
+      "learning_rate": 1.6663465779520042e-05,
+      "loss": 1.0445,
+      "step": 385
+    },
+    {
+      "epoch": 0.2905805965935824,
+      "grad_norm": 3.854702956207567,
+      "learning_rate": 1.6645258833538015e-05,
+      "loss": 1.1603,
+      "step": 386
+    },
+    {
+      "epoch": 0.29133339606662273,
+      "grad_norm": 3.383690393128223,
+      "learning_rate": 1.6627012352741482e-05,
+      "loss": 1.0015,
+      "step": 387
+    },
+    {
+      "epoch": 0.29208619553966314,
+      "grad_norm": 3.364330036008893,
+      "learning_rate": 1.6608726445684715e-05,
+      "loss": 1.0971,
+      "step": 388
+    },
+    {
+      "epoch": 0.2928389950127035,
+      "grad_norm": 3.114986206892385,
+      "learning_rate": 1.659040122115655e-05,
+      "loss": 0.9818,
+      "step": 389
+    },
+    {
+      "epoch": 0.29359179448574385,
+      "grad_norm": 3.0575994965283186,
+      "learning_rate": 1.6572036788179728e-05,
+      "loss": 1.0585,
+      "step": 390
+    },
+    {
+      "epoch": 0.2943445939587842,
+      "grad_norm": 3.0998367099456035,
+      "learning_rate": 1.6553633256010254e-05,
+      "loss": 0.9561,
+      "step": 391
+    },
+    {
+      "epoch": 0.2950973934318246,
+      "grad_norm": 3.3358383840570247,
+      "learning_rate": 1.653519073413675e-05,
+      "loss": 1.1339,
+      "step": 392
+    },
+    {
+      "epoch": 0.295850192904865,
+      "grad_norm": 3.6575458151303555,
+      "learning_rate": 1.6516709332279806e-05,
+      "loss": 1.1176,
+      "step": 393
+    },
+    {
+      "epoch": 0.29660299237790533,
+      "grad_norm": 3.223000019134338,
+      "learning_rate": 1.649818916039131e-05,
+      "loss": 0.985,
+      "step": 394
+    },
+    {
+      "epoch": 0.2973557918509457,
+      "grad_norm": 2.8833925903157156,
+      "learning_rate": 1.6479630328653814e-05,
+      "loss": 1.0436,
+      "step": 395
+    },
+    {
+      "epoch": 0.2981085913239861,
+      "grad_norm": 3.451911301191298,
+      "learning_rate": 1.646103294747987e-05,
+      "loss": 1.0887,
+      "step": 396
+    },
+    {
+      "epoch": 0.29886139079702645,
+      "grad_norm": 2.9713239489773073,
+      "learning_rate": 1.6442397127511366e-05,
+      "loss": 1.0673,
+      "step": 397
+    },
+    {
+      "epoch": 0.2996141902700668,
+      "grad_norm": 3.040445936124947,
+      "learning_rate": 1.6423722979618883e-05,
+      "loss": 1.0726,
+      "step": 398
+    },
+    {
+      "epoch": 0.30036698974310716,
+      "grad_norm": 3.468326113161527,
+      "learning_rate": 1.6405010614901017e-05,
+      "loss": 1.012,
+      "step": 399
+    },
+    {
+      "epoch": 0.30036698974310716,
+      "eval_loss": 1.0304744243621826,
+      "eval_runtime": 583.2125,
+      "eval_samples_per_second": 30.666,
+      "eval_steps_per_second": 0.48,
+      "step": 399
+    },
+    {
+      "epoch": 0.30111978921614757,
+      "grad_norm": 2.8552822081266305,
+      "learning_rate": 1.6386260144683744e-05,
+      "loss": 1.0157,
+      "step": 400
+    },
+    {
+      "epoch": 0.3018725886891879,
+      "grad_norm": 5.422850816552489,
+      "learning_rate": 1.6367471680519734e-05,
+      "loss": 1.0109,
+      "step": 401
+    },
+    {
+      "epoch": 0.3026253881622283,
+      "grad_norm": 3.6553640892361985,
+      "learning_rate": 1.6348645334187686e-05,
+      "loss": 1.1007,
+      "step": 402
+    },
+    {
+      "epoch": 0.30337818763526864,
+      "grad_norm": 3.9670270138761645,
+      "learning_rate": 1.632978121769169e-05,
+      "loss": 1.0345,
+      "step": 403
+    },
+    {
+      "epoch": 0.30413098710830905,
+      "grad_norm": 3.224743049795719,
+      "learning_rate": 1.631087944326053e-05,
+      "loss": 1.0532,
+      "step": 404
+    },
+    {
+      "epoch": 0.3048837865813494,
+      "grad_norm": 3.5913998466341286,
+      "learning_rate": 1.6291940123347033e-05,
+      "loss": 1.08,
+      "step": 405
+    },
+    {
+      "epoch": 0.30563658605438976,
+      "grad_norm": 2.906360104068391,
+      "learning_rate": 1.6272963370627398e-05,
+      "loss": 1.0642,
+      "step": 406
+    },
+    {
+      "epoch": 0.3063893855274301,
+      "grad_norm": 2.8309108092126616,
+      "learning_rate": 1.6253949298000527e-05,
+      "loss": 1.0076,
+      "step": 407
+    },
+    {
+      "epoch": 0.3071421850004705,
+      "grad_norm": 2.913978002284561,
+      "learning_rate": 1.6234898018587336e-05,
+      "loss": 1.0802,
+      "step": 408
+    },
+    {
+      "epoch": 0.3078949844735109,
+      "grad_norm": 3.2698030119279538,
+      "learning_rate": 1.6215809645730115e-05,
+      "loss": 1.1231,
+      "step": 409
+    },
+    {
+      "epoch": 0.30864778394655124,
+      "grad_norm": 3.1904018234043248,
+      "learning_rate": 1.6196684292991827e-05,
+      "loss": 1.0841,
+      "step": 410
+    },
+    {
+      "epoch": 0.3094005834195916,
+      "grad_norm": 3.0466834191213703,
+      "learning_rate": 1.6177522074155436e-05,
+      "loss": 1.0113,
+      "step": 411
+    },
+    {
+      "epoch": 0.310153382892632,
+      "grad_norm": 4.011803340356661,
+      "learning_rate": 1.615832310322324e-05,
+      "loss": 1.126,
+      "step": 412
+    },
+    {
+      "epoch": 0.31090618236567236,
+      "grad_norm": 2.761867409413426,
+      "learning_rate": 1.6139087494416184e-05,
+      "loss": 1.123,
+      "step": 413
+    },
+    {
+      "epoch": 0.3116589818387127,
+      "grad_norm": 2.8570455014305014,
+      "learning_rate": 1.6119815362173188e-05,
+      "loss": 1.0686,
+      "step": 414
+    },
+    {
+      "epoch": 0.31241178131175307,
+      "grad_norm": 3.1348742494088633,
+      "learning_rate": 1.6100506821150455e-05,
+      "loss": 1.0467,
+      "step": 415
+    },
+    {
+      "epoch": 0.3131645807847935,
+      "grad_norm": 3.072427636664424,
+      "learning_rate": 1.6081161986220807e-05,
+      "loss": 1.0526,
+      "step": 416
+    },
+    {
+      "epoch": 0.31391738025783383,
+      "grad_norm": 3.001432948323744,
+      "learning_rate": 1.6061780972472978e-05,
+      "loss": 1.0589,
+      "step": 417
+    },
+    {
+      "epoch": 0.3146701797308742,
+      "grad_norm": 2.876397896586471,
+      "learning_rate": 1.6042363895210948e-05,
+      "loss": 1.0391,
+      "step": 418
+    },
+    {
+      "epoch": 0.31542297920391454,
+      "grad_norm": 5.019687647291023,
+      "learning_rate": 1.6022910869953245e-05,
+      "loss": 1.0123,
+      "step": 419
+    },
+    {
+      "epoch": 0.3161757786769549,
+      "grad_norm": 2.758077610379088,
+      "learning_rate": 1.6003422012432275e-05,
+      "loss": 1.0637,
+      "step": 420
+    },
+    {
+      "epoch": 0.3169285781499953,
+      "grad_norm": 4.352286539206582,
+      "learning_rate": 1.5983897438593612e-05,
+      "loss": 1.0986,
+      "step": 421
+    },
+    {
+      "epoch": 0.31768137762303567,
+      "grad_norm": 5.425551091825354,
+      "learning_rate": 1.596433726459532e-05,
+      "loss": 1.0641,
+      "step": 422
+    },
+    {
+      "epoch": 0.318434177096076,
+      "grad_norm": 3.114326822208182,
+      "learning_rate": 1.5944741606807257e-05,
+      "loss": 1.0409,
+      "step": 423
+    },
+    {
+      "epoch": 0.3191869765691164,
+      "grad_norm": 3.4234417375142665,
+      "learning_rate": 1.5925110581810396e-05,
+      "loss": 1.1171,
+      "step": 424
+    },
+    {
+      "epoch": 0.3199397760421568,
+      "grad_norm": 2.5707654982863377,
+      "learning_rate": 1.590544430639611e-05,
+      "loss": 1.0763,
+      "step": 425
+    },
+    {
+      "epoch": 0.32069257551519714,
+      "grad_norm": 3.4413913127761586,
+      "learning_rate": 1.5885742897565494e-05,
+      "loss": 1.1053,
+      "step": 426
+    },
+    {
+      "epoch": 0.3214453749882375,
+      "grad_norm": 2.9975019877398124,
+      "learning_rate": 1.586600647252866e-05,
+      "loss": 0.9461,
+      "step": 427
+    },
+    {
+      "epoch": 0.32219817446127785,
+      "grad_norm": 3.190004846812769,
+      "learning_rate": 1.5846235148704047e-05,
+      "loss": 0.9774,
+      "step": 428
+    },
+    {
+      "epoch": 0.32295097393431826,
+      "grad_norm": 3.110338821378457,
+      "learning_rate": 1.5826429043717716e-05,
+      "loss": 1.1223,
+      "step": 429
+    },
+    {
+      "epoch": 0.3237037734073586,
+      "grad_norm": 2.8781286921631613,
+      "learning_rate": 1.580658827540265e-05,
+      "loss": 1.1501,
+      "step": 430
+    },
+    {
+      "epoch": 0.324456572880399,
+      "grad_norm": 3.3316597059677,
+      "learning_rate": 1.578671296179806e-05,
+      "loss": 1.0884,
+      "step": 431
+    },
+    {
+      "epoch": 0.32520937235343933,
+      "grad_norm": 3.3183585538326925,
+      "learning_rate": 1.5766803221148676e-05,
+      "loss": 1.1408,
+      "step": 432
+    },
+    {
+      "epoch": 0.32596217182647974,
+      "grad_norm": 3.2746550035987485,
+      "learning_rate": 1.574685917190404e-05,
+      "loss": 1.0604,
+      "step": 433
+    },
+    {
+      "epoch": 0.3267149712995201,
+      "grad_norm": 2.9205856949478544,
+      "learning_rate": 1.572688093271782e-05,
+      "loss": 1.0424,
+      "step": 434
+    },
+    {
+      "epoch": 0.32746777077256045,
+      "grad_norm": 2.8380294657039795,
+      "learning_rate": 1.5706868622447084e-05,
+      "loss": 1.0492,
+      "step": 435
+    },
+    {
+      "epoch": 0.3282205702456008,
+      "grad_norm": 3.753517198713471,
+      "learning_rate": 1.568682236015159e-05,
+      "loss": 1.0763,
+      "step": 436
+    },
+    {
+      "epoch": 0.3289733697186412,
+      "grad_norm": 3.450360877181349,
+      "learning_rate": 1.566674226509311e-05,
+      "loss": 1.058,
+      "step": 437
+    },
+    {
+      "epoch": 0.3297261691916816,
+      "grad_norm": 3.5110811328728104,
+      "learning_rate": 1.564662845673468e-05,
+      "loss": 1.0606,
+      "step": 438
+    },
+    {
+      "epoch": 0.33047896866472193,
+      "grad_norm": 3.927731470985521,
+      "learning_rate": 1.5626481054739916e-05,
+      "loss": 1.1387,
+      "step": 439
+    },
+    {
+      "epoch": 0.3312317681377623,
+      "grad_norm": 5.154292997922348,
+      "learning_rate": 1.560630017897229e-05,
+      "loss": 1.0563,
+      "step": 440
+    },
+    {
+      "epoch": 0.3319845676108027,
+      "grad_norm": 4.313155430470082,
+      "learning_rate": 1.558608594949441e-05,
+      "loss": 1.1021,
+      "step": 441
+    },
+    {
+      "epoch": 0.33273736708384305,
+      "grad_norm": 2.7409094235023224,
+      "learning_rate": 1.5565838486567343e-05,
+      "loss": 0.9847,
+      "step": 442
+    },
+    {
+      "epoch": 0.3334901665568834,
+      "grad_norm": 2.832698130182746,
+      "learning_rate": 1.554555791064985e-05,
+      "loss": 1.0113,
+      "step": 443
+    },
+    {
+      "epoch": 0.33424296602992376,
+      "grad_norm": 3.1151968557597627,
+      "learning_rate": 1.55252443423977e-05,
+      "loss": 1.0411,
+      "step": 444
+    },
+    {
+      "epoch": 0.33499576550296417,
+      "grad_norm": 2.9105383861881378,
+      "learning_rate": 1.550489790266294e-05,
+      "loss": 1.0841,
+      "step": 445
+    },
+    {
+      "epoch": 0.3357485649760045,
+      "grad_norm": 3.616458297606771,
+      "learning_rate": 1.5484518712493188e-05,
+      "loss": 1.0521,
+      "step": 446
+    },
+    {
+      "epoch": 0.3365013644490449,
+      "grad_norm": 3.5478639800192076,
+      "learning_rate": 1.5464106893130896e-05,
+      "loss": 1.124,
+      "step": 447
+    },
+    {
+      "epoch": 0.33725416392208524,
+      "grad_norm": 3.029607726981716,
+      "learning_rate": 1.5443662566012645e-05,
+      "loss": 1.0125,
+      "step": 448
+    },
+    {
+      "epoch": 0.33800696339512565,
+      "grad_norm": 2.72691607967082,
+      "learning_rate": 1.542318585276841e-05,
+      "loss": 1.055,
+      "step": 449
+    },
+    {
+      "epoch": 0.338759762868166,
+      "grad_norm": 4.1070664679867575,
+      "learning_rate": 1.5402676875220847e-05,
+      "loss": 1.1034,
+      "step": 450
+    },
+    {
+      "epoch": 0.33951256234120636,
+      "grad_norm": 2.8627575059209147,
+      "learning_rate": 1.5382135755384554e-05,
+      "loss": 1.0617,
+      "step": 451
+    },
+    {
+      "epoch": 0.3402653618142467,
+      "grad_norm": 2.846634773819146,
+      "learning_rate": 1.5361562615465366e-05,
+      "loss": 1.0299,
+      "step": 452
+    },
+    {
+      "epoch": 0.3410181612872871,
+      "grad_norm": 3.185640397378095,
+      "learning_rate": 1.5340957577859605e-05,
+      "loss": 1.0712,
+      "step": 453
+    },
+    {
+      "epoch": 0.3417709607603275,
+      "grad_norm": 2.9033149341984106,
+      "learning_rate": 1.5320320765153367e-05,
+      "loss": 1.0137,
+      "step": 454
+    },
+    {
+      "epoch": 0.34252376023336784,
+      "grad_norm": 2.6348325760822235,
+      "learning_rate": 1.5299652300121792e-05,
+      "loss": 1.0322,
+      "step": 455
+    },
+    {
+      "epoch": 0.3432765597064082,
+      "grad_norm": 3.141329980975792,
+      "learning_rate": 1.5278952305728325e-05,
+      "loss": 1.0285,
+      "step": 456
+    },
+    {
+      "epoch": 0.3440293591794486,
+      "grad_norm": 2.932905735802033,
+      "learning_rate": 1.5258220905123997e-05,
+      "loss": 0.9916,
+      "step": 457
+    },
+    {
+      "epoch": 0.34478215865248896,
+      "grad_norm": 3.91581873738193,
+      "learning_rate": 1.5237458221646668e-05,
+      "loss": 0.9755,
+      "step": 458
+    },
+    {
+      "epoch": 0.3455349581255293,
+      "grad_norm": 3.0788973407240112,
+      "learning_rate": 1.5216664378820327e-05,
+      "loss": 0.9842,
+      "step": 459
+    },
+    {
+      "epoch": 0.34628775759856967,
+      "grad_norm": 3.1508943897589785,
+      "learning_rate": 1.5195839500354337e-05,
+      "loss": 1.0143,
+      "step": 460
+    },
+    {
+      "epoch": 0.34704055707161,
+      "grad_norm": 2.755535031332719,
+      "learning_rate": 1.5174983710142694e-05,
+      "loss": 1.1033,
+      "step": 461
+    },
+    {
+      "epoch": 0.34779335654465043,
+      "grad_norm": 2.72908305859456,
+      "learning_rate": 1.515409713226331e-05,
+      "loss": 1.0267,
+      "step": 462
+    },
+    {
+      "epoch": 0.3485461560176908,
+      "grad_norm": 2.8179620108009664,
+      "learning_rate": 1.513317989097725e-05,
+      "loss": 1.0323,
+      "step": 463
+    },
+    {
+      "epoch": 0.34929895549073114,
+      "grad_norm": 3.414258478909775,
+      "learning_rate": 1.5112232110728016e-05,
+      "loss": 1.0828,
+      "step": 464
+    },
+    {
+      "epoch": 0.3500517549637715,
+      "grad_norm": 3.2784290028904954,
+      "learning_rate": 1.5091253916140789e-05,
+      "loss": 1.0864,
+      "step": 465
+    },
+    {
+      "epoch": 0.3508045544368119,
+      "grad_norm": 4.226014886336044,
+      "learning_rate": 1.5070245432021699e-05,
+      "loss": 1.0652,
+      "step": 466
+    },
+    {
+      "epoch": 0.35155735390985227,
+      "grad_norm": 2.4370397137721733,
+      "learning_rate": 1.5049206783357082e-05,
+      "loss": 0.9975,
+      "step": 467
+    },
+    {
+      "epoch": 0.3523101533828926,
+      "grad_norm": 3.5305399592217794,
+      "learning_rate": 1.502813809531272e-05,
+      "loss": 0.9722,
+      "step": 468
+    },
+    {
+      "epoch": 0.353062952855933,
+      "grad_norm": 3.6714227478521138,
+      "learning_rate": 1.5007039493233123e-05,
+      "loss": 1.0186,
+      "step": 469
+    },
+    {
+      "epoch": 0.3538157523289734,
+      "grad_norm": 3.2619665655127292,
+      "learning_rate": 1.4985911102640762e-05,
+      "loss": 1.0209,
+      "step": 470
+    },
+    {
+      "epoch": 0.35456855180201374,
+      "grad_norm": 2.8237326768852338,
+      "learning_rate": 1.4964753049235333e-05,
+      "loss": 1.0689,
+      "step": 471
+    },
+    {
+      "epoch": 0.3553213512750541,
+      "grad_norm": 3.0898996817280437,
+      "learning_rate": 1.4943565458892999e-05,
+      "loss": 1.0663,
+      "step": 472
+    },
+    {
+      "epoch": 0.35607415074809445,
+      "grad_norm": 3.0818327635483347,
+      "learning_rate": 1.4922348457665656e-05,
+      "loss": 1.1391,
+      "step": 473
+    },
+    {
+      "epoch": 0.35682695022113486,
+      "grad_norm": 2.6170688801613036,
+      "learning_rate": 1.4901102171780175e-05,
+      "loss": 1.0734,
+      "step": 474
+    },
+    {
+      "epoch": 0.3575797496941752,
+      "grad_norm": 2.6182095607098868,
+      "learning_rate": 1.487982672763764e-05,
+      "loss": 0.9822,
+      "step": 475
+    },
+    {
+      "epoch": 0.3583325491672156,
+      "grad_norm": 3.531969368207952,
+      "learning_rate": 1.4858522251812621e-05,
+      "loss": 1.1004,
+      "step": 476
+    },
+    {
+      "epoch": 0.35908534864025593,
+      "grad_norm": 3.3782017687196575,
+      "learning_rate": 1.4837188871052399e-05,
+      "loss": 1.0112,
+      "step": 477
+    },
+    {
+      "epoch": 0.35983814811329634,
+      "grad_norm": 3.0803648867618802,
+      "learning_rate": 1.481582671227622e-05,
+      "loss": 1.0364,
+      "step": 478
+    },
+    {
+      "epoch": 0.3605909475863367,
+      "grad_norm": 3.665787032584331,
+      "learning_rate": 1.4794435902574543e-05,
+      "loss": 1.085,
+      "step": 479
+    },
+    {
+      "epoch": 0.36134374705937705,
+      "grad_norm": 3.211770330714845,
+      "learning_rate": 1.4773016569208283e-05,
+      "loss": 1.1324,
+      "step": 480
+    },
+    {
+      "epoch": 0.3620965465324174,
+      "grad_norm": 3.4731502354467874,
+      "learning_rate": 1.4751568839608036e-05,
+      "loss": 1.0276,
+      "step": 481
+    },
+    {
+      "epoch": 0.3628493460054578,
+      "grad_norm": 3.5439955990261334,
+      "learning_rate": 1.4730092841373362e-05,
+      "loss": 1.0329,
+      "step": 482
+    },
+    {
+      "epoch": 0.3636021454784982,
+      "grad_norm": 2.778610271559237,
+      "learning_rate": 1.4708588702271978e-05,
+      "loss": 1.0138,
+      "step": 483
+    },
+    {
+      "epoch": 0.36435494495153853,
+      "grad_norm": 4.278298948928529,
+      "learning_rate": 1.468705655023903e-05,
+      "loss": 1.065,
+      "step": 484
+    },
+    {
+      "epoch": 0.3651077444245789,
+      "grad_norm": 3.5441846422629593,
+      "learning_rate": 1.466549651337632e-05,
+      "loss": 1.0446,
+      "step": 485
+    },
+    {
+      "epoch": 0.3658605438976193,
+      "grad_norm": 3.039844008569692,
+      "learning_rate": 1.4643908719951551e-05,
+      "loss": 1.085,
+      "step": 486
+    },
+    {
+      "epoch": 0.36661334337065965,
+      "grad_norm": 2.8411313966432052,
+      "learning_rate": 1.4622293298397554e-05,
+      "loss": 1.1288,
+      "step": 487
+    },
+    {
+      "epoch": 0.3673661428437,
+      "grad_norm": 3.747615895853388,
+      "learning_rate": 1.4600650377311523e-05,
+      "loss": 1.0776,
+      "step": 488
+    },
+    {
+      "epoch": 0.36811894231674036,
+      "grad_norm": 2.7460960649591524,
+      "learning_rate": 1.4578980085454268e-05,
+      "loss": 1.0613,
+      "step": 489
+    },
+    {
+      "epoch": 0.36887174178978077,
+      "grad_norm": 3.1896235531083286,
+      "learning_rate": 1.4557282551749428e-05,
+      "loss": 0.99,
+      "step": 490
+    },
+    {
+      "epoch": 0.3696245412628211,
+      "grad_norm": 3.159003490883495,
+      "learning_rate": 1.4535557905282716e-05,
+      "loss": 0.9487,
+      "step": 491
+    },
+    {
+      "epoch": 0.3703773407358615,
+      "grad_norm": 3.0652639342812833,
+      "learning_rate": 1.451380627530115e-05,
+      "loss": 0.9943,
+      "step": 492
+    },
+    {
+      "epoch": 0.37113014020890184,
+      "grad_norm": 3.550087346952904,
+      "learning_rate": 1.4492027791212275e-05,
+      "loss": 1.0086,
+      "step": 493
+    },
+    {
+      "epoch": 0.37188293968194225,
+      "grad_norm": 2.6049865869424704,
+      "learning_rate": 1.4470222582583404e-05,
+      "loss": 0.9236,
+      "step": 494
+    },
+    {
+      "epoch": 0.3726357391549826,
+      "grad_norm": 3.6347381036624995,
+      "learning_rate": 1.4448390779140844e-05,
+      "loss": 1.1121,
+      "step": 495
+    },
+    {
+      "epoch": 0.37338853862802296,
+      "grad_norm": 6.880850065504925,
+      "learning_rate": 1.442653251076912e-05,
+      "loss": 0.9888,
+      "step": 496
+    },
+    {
+      "epoch": 0.3741413381010633,
+      "grad_norm": 6.344279229708359,
+      "learning_rate": 1.4404647907510212e-05,
+      "loss": 1.0726,
+      "step": 497
+    },
+    {
+      "epoch": 0.3748941375741037,
+      "grad_norm": 2.855607928708251,
+      "learning_rate": 1.4382737099562765e-05,
+      "loss": 1.0838,
+      "step": 498
+    },
+    {
+      "epoch": 0.3756469370471441,
+      "grad_norm": 2.9832996495692075,
+      "learning_rate": 1.436080021728133e-05,
+      "loss": 0.9708,
+      "step": 499
+    },
+    {
+      "epoch": 0.37639973652018444,
+      "grad_norm": 2.751066005440165,
+      "learning_rate": 1.4338837391175582e-05,
+      "loss": 1.0608,
+      "step": 500
+    },
+    {
+      "epoch": 0.3771525359932248,
+      "grad_norm": 3.485859409386025,
+      "learning_rate": 1.431684875190955e-05,
+      "loss": 1.0913,
+      "step": 501
+    },
+    {
+      "epoch": 0.37790533546626515,
+      "grad_norm": 3.426005357359028,
+      "learning_rate": 1.4294834430300822e-05,
+      "loss": 1.0836,
+      "step": 502
+    },
+    {
+      "epoch": 0.37865813493930556,
+      "grad_norm": 3.9542615717855463,
+      "learning_rate": 1.4272794557319785e-05,
+      "loss": 1.1442,
+      "step": 503
+    },
+    {
+      "epoch": 0.3794109344123459,
+      "grad_norm": 2.622153474999588,
+      "learning_rate": 1.4250729264088845e-05,
+      "loss": 1.0587,
+      "step": 504
+    },
+    {
+      "epoch": 0.38016373388538627,
+      "grad_norm": 3.087065445678147,
+      "learning_rate": 1.4228638681881633e-05,
+      "loss": 1.0277,
+      "step": 505
+    },
+    {
+      "epoch": 0.3809165333584266,
+      "grad_norm": 3.1258129796224625,
+      "learning_rate": 1.4206522942122232e-05,
+      "loss": 1.0719,
+      "step": 506
+    },
+    {
+      "epoch": 0.38166933283146703,
+      "grad_norm": 3.5669441503070143,
+      "learning_rate": 1.4184382176384403e-05,
+      "loss": 1.1192,
+      "step": 507
+    },
+    {
+      "epoch": 0.3824221323045074,
+      "grad_norm": 2.631996666259307,
+      "learning_rate": 1.4162216516390787e-05,
+      "loss": 1.0321,
+      "step": 508
+    },
+    {
+      "epoch": 0.38317493177754774,
+      "grad_norm": 2.7083628383479414,
+      "learning_rate": 1.4140026094012136e-05,
+      "loss": 1.0325,
+      "step": 509
+    },
+    {
+      "epoch": 0.3839277312505881,
+      "grad_norm": 2.7679663041389624,
+      "learning_rate": 1.4117811041266518e-05,
+      "loss": 0.9971,
+      "step": 510
+    },
+    {
+      "epoch": 0.3846805307236285,
+      "grad_norm": 3.0521936330145834,
+      "learning_rate": 1.4095571490318532e-05,
+      "loss": 1.0261,
+      "step": 511
+    },
+    {
+      "epoch": 0.38543333019666887,
+      "grad_norm": 2.7908671493996233,
+      "learning_rate": 1.4073307573478528e-05,
+      "loss": 1.0099,
+      "step": 512
+    },
+    {
+      "epoch": 0.3861861296697092,
+      "grad_norm": 2.5789858700673647,
+      "learning_rate": 1.405101942320182e-05,
+      "loss": 1.0592,
+      "step": 513
+    },
+    {
+      "epoch": 0.3869389291427496,
+      "grad_norm": 4.887610637172058,
+      "learning_rate": 1.4028707172087885e-05,
+      "loss": 1.1322,
+      "step": 514
+    },
+    {
+      "epoch": 0.38769172861579,
+      "grad_norm": 3.542141348173503,
+      "learning_rate": 1.40063709528796e-05,
+      "loss": 1.0538,
+      "step": 515
+    },
+    {
+      "epoch": 0.38844452808883034,
+      "grad_norm": 3.290310094881258,
+      "learning_rate": 1.3984010898462417e-05,
+      "loss": 1.0527,
+      "step": 516
+    },
+    {
+      "epoch": 0.3891973275618707,
+      "grad_norm": 3.089847663466809,
+      "learning_rate": 1.3961627141863603e-05,
+      "loss": 1.0758,
+      "step": 517
+    },
+    {
+      "epoch": 0.38995012703491105,
+      "grad_norm": 3.046285439988099,
+      "learning_rate": 1.3939219816251435e-05,
+      "loss": 1.0543,
+      "step": 518
+    },
+    {
+      "epoch": 0.39070292650795146,
+      "grad_norm": 2.907476323401419,
+      "learning_rate": 1.3916789054934408e-05,
+      "loss": 1.0748,
+      "step": 519
+    },
+    {
+      "epoch": 0.3914557259809918,
+      "grad_norm": 2.9958486413284344,
+      "learning_rate": 1.3894334991360448e-05,
+      "loss": 1.1112,
+      "step": 520
+    },
+    {
+      "epoch": 0.3922085254540322,
+      "grad_norm": 2.66442614074179,
+      "learning_rate": 1.3871857759116104e-05,
+      "loss": 0.9338,
+      "step": 521
+    },
+    {
+      "epoch": 0.39296132492707253,
+      "grad_norm": 3.6446504858155215,
+      "learning_rate": 1.3849357491925779e-05,
+      "loss": 1.1278,
+      "step": 522
+    },
+    {
+      "epoch": 0.39371412440011294,
+      "grad_norm": 3.22553976414466,
+      "learning_rate": 1.3826834323650899e-05,
+      "loss": 1.0915,
+      "step": 523
+    },
+    {
+      "epoch": 0.3944669238731533,
+      "grad_norm": 3.3837299425627343,
+      "learning_rate": 1.3804288388289152e-05,
+      "loss": 0.9893,
+      "step": 524
+    },
+    {
+      "epoch": 0.39521972334619365,
+      "grad_norm": 4.383592843892888,
+      "learning_rate": 1.378171981997367e-05,
+      "loss": 1.1149,
+      "step": 525
+    },
+    {
+      "epoch": 0.395972522819234,
+      "grad_norm": 2.9287918069708967,
+      "learning_rate": 1.3759128752972229e-05,
+      "loss": 0.9805,
+      "step": 526
+    },
+    {
+      "epoch": 0.3967253222922744,
+      "grad_norm": 3.769542827664187,
+      "learning_rate": 1.3736515321686468e-05,
+      "loss": 0.999,
+      "step": 527
+    },
+    {
+      "epoch": 0.3974781217653148,
+      "grad_norm": 4.040693234235752,
+      "learning_rate": 1.3713879660651069e-05,
+      "loss": 1.0457,
+      "step": 528
+    },
+    {
+      "epoch": 0.39823092123835513,
+      "grad_norm": 3.0216912415405193,
+      "learning_rate": 1.3691221904532972e-05,
+      "loss": 1.0238,
+      "step": 529
+    },
+    {
+      "epoch": 0.3989837207113955,
+      "grad_norm": 3.2298304957599444,
+      "learning_rate": 1.3668542188130567e-05,
+      "loss": 1.1817,
+      "step": 530
+    },
+    {
+      "epoch": 0.3997365201844359,
+      "grad_norm": 3.205471685042934,
+      "learning_rate": 1.364584064637289e-05,
+      "loss": 1.052,
+      "step": 531
+    },
+    {
+      "epoch": 0.40048931965747625,
+      "grad_norm": 3.0582507683769897,
+      "learning_rate": 1.3623117414318827e-05,
+      "loss": 1.0515,
+      "step": 532
+    },
+    {
+      "epoch": 0.40048931965747625,
+      "eval_loss": 1.013999581336975,
+      "eval_runtime": 583.7469,
+      "eval_samples_per_second": 30.638,
+      "eval_steps_per_second": 0.48,
+      "step": 532
+    },
+    {
+      "epoch": 0.4012421191305166,
+      "grad_norm": 3.2249512274563905,
+      "learning_rate": 1.3600372627156304e-05,
+      "loss": 1.1174,
+      "step": 533
+    },
+    {
+      "epoch": 0.40199491860355696,
+      "grad_norm": 2.8980939124189193,
+      "learning_rate": 1.3577606420201483e-05,
+      "loss": 0.9254,
+      "step": 534
+    },
+    {
+      "epoch": 0.40274771807659737,
+      "grad_norm": 2.6685576816268,
+      "learning_rate": 1.3554818928897965e-05,
+      "loss": 1.0023,
+      "step": 535
+    },
+    {
+      "epoch": 0.4035005175496377,
+      "grad_norm": 3.2379228804073947,
+      "learning_rate": 1.353201028881598e-05,
+      "loss": 1.083,
+      "step": 536
+    },
+    {
+      "epoch": 0.4042533170226781,
+      "grad_norm": 3.0605669152172728,
+      "learning_rate": 1.350918063565157e-05,
+      "loss": 0.9634,
+      "step": 537
+    },
+    {
+      "epoch": 0.40500611649571844,
+      "grad_norm": 3.2347349427551078,
+      "learning_rate": 1.3486330105225797e-05,
+      "loss": 0.9626,
+      "step": 538
+    },
+    {
+      "epoch": 0.40575891596875885,
+      "grad_norm": 3.7810473941095264,
+      "learning_rate": 1.3463458833483923e-05,
+      "loss": 1.1268,
+      "step": 539
+    },
+    {
+      "epoch": 0.4065117154417992,
+      "grad_norm": 2.726200850867419,
+      "learning_rate": 1.344056695649462e-05,
+      "loss": 1.1052,
+      "step": 540
+    },
+    {
+      "epoch": 0.40726451491483956,
+      "grad_norm": 2.5107188818811923,
+      "learning_rate": 1.3417654610449131e-05,
+      "loss": 1.0178,
+      "step": 541
+    },
+    {
+      "epoch": 0.4080173143878799,
+      "grad_norm": 3.010490350395698,
+      "learning_rate": 1.3394721931660488e-05,
+      "loss": 0.9679,
+      "step": 542
+    },
+    {
+      "epoch": 0.40877011386092027,
+      "grad_norm": 2.465477343951642,
+      "learning_rate": 1.3371769056562683e-05,
+      "loss": 1.0452,
+      "step": 543
+    },
+    {
+      "epoch": 0.4095229133339607,
+      "grad_norm": 4.104294474246676,
+      "learning_rate": 1.3348796121709862e-05,
+      "loss": 1.0685,
+      "step": 544
+    },
+    {
+      "epoch": 0.41027571280700104,
+      "grad_norm": 2.918854588082266,
+      "learning_rate": 1.3325803263775521e-05,
+      "loss": 1.0147,
+      "step": 545
+    },
+    {
+      "epoch": 0.4110285122800414,
+      "grad_norm": 3.31220295919612,
+      "learning_rate": 1.3302790619551673e-05,
+      "loss": 1.0331,
+      "step": 546
+    },
+    {
+      "epoch": 0.41178131175308175,
+      "grad_norm": 4.183130206995228,
+      "learning_rate": 1.3279758325948054e-05,
+      "loss": 1.1075,
+      "step": 547
+    },
+    {
+      "epoch": 0.41253411122612216,
+      "grad_norm": 3.7338257554064422,
+      "learning_rate": 1.32567065199913e-05,
+      "loss": 1.0739,
+      "step": 548
+    },
+    {
+      "epoch": 0.4132869106991625,
+      "grad_norm": 3.1338772898350995,
+      "learning_rate": 1.3233635338824132e-05,
+      "loss": 1.0575,
+      "step": 549
+    },
+    {
+      "epoch": 0.41403971017220287,
+      "grad_norm": 2.923412077252181,
+      "learning_rate": 1.3210544919704539e-05,
+      "loss": 1.0108,
+      "step": 550
+    },
+    {
+      "epoch": 0.4147925096452432,
+      "grad_norm": 2.383268802641107,
+      "learning_rate": 1.318743540000496e-05,
+      "loss": 0.9357,
+      "step": 551
+    },
+    {
+      "epoch": 0.41554530911828363,
+      "grad_norm": 2.8657602020971678,
+      "learning_rate": 1.3164306917211475e-05,
+      "loss": 1.0201,
+      "step": 552
+    },
+    {
+      "epoch": 0.416298108591324,
+      "grad_norm": 6.229638816200601,
+      "learning_rate": 1.3141159608922984e-05,
+      "loss": 1.1212,
+      "step": 553
+    },
+    {
+      "epoch": 0.41705090806436435,
+      "grad_norm": 3.0334418266757806,
+      "learning_rate": 1.3117993612850377e-05,
+      "loss": 1.0709,
+      "step": 554
+    },
+    {
+      "epoch": 0.4178037075374047,
+      "grad_norm": 3.3777028783426464,
+      "learning_rate": 1.3094809066815731e-05,
+      "loss": 1.0202,
+      "step": 555
+    },
+    {
+      "epoch": 0.4185565070104451,
+      "grad_norm": 4.869952280964426,
+      "learning_rate": 1.3071606108751475e-05,
+      "loss": 1.066,
+      "step": 556
+    },
+    {
+      "epoch": 0.41930930648348547,
+      "grad_norm": 3.207610155370362,
+      "learning_rate": 1.3048384876699588e-05,
+      "loss": 1.0221,
+      "step": 557
+    },
+    {
+      "epoch": 0.4200621059565258,
+      "grad_norm": 4.465798720585036,
+      "learning_rate": 1.302514550881076e-05,
+      "loss": 0.9787,
+      "step": 558
+    },
+    {
+      "epoch": 0.4208149054295662,
+      "grad_norm": 3.803049002174094,
+      "learning_rate": 1.3001888143343578e-05,
+      "loss": 1.02,
+      "step": 559
+    },
+    {
+      "epoch": 0.4215677049026066,
+      "grad_norm": 3.441550189570075,
+      "learning_rate": 1.2978612918663702e-05,
+      "loss": 1.0819,
+      "step": 560
+    },
+    {
+      "epoch": 0.42232050437564694,
+      "grad_norm": 2.688153536798873,
+      "learning_rate": 1.2955319973243043e-05,
+      "loss": 1.0335,
+      "step": 561
+    },
+    {
+      "epoch": 0.4230733038486873,
+      "grad_norm": 3.4545648939837545,
+      "learning_rate": 1.293200944565894e-05,
+      "loss": 1.0494,
+      "step": 562
+    },
+    {
+      "epoch": 0.42382610332172765,
+      "grad_norm": 3.0549486245877584,
+      "learning_rate": 1.290868147459333e-05,
+      "loss": 1.0133,
+      "step": 563
+    },
+    {
+      "epoch": 0.42457890279476806,
+      "grad_norm": 2.6796979217078025,
+      "learning_rate": 1.2885336198831933e-05,
+      "loss": 1.0228,
+      "step": 564
+    },
+    {
+      "epoch": 0.4253317022678084,
+      "grad_norm": 2.564267991029905,
+      "learning_rate": 1.2861973757263416e-05,
+      "loss": 0.9532,
+      "step": 565
+    },
+    {
+      "epoch": 0.4260845017408488,
+      "grad_norm": 3.2351011961272897,
+      "learning_rate": 1.2838594288878567e-05,
+      "loss": 1.0146,
+      "step": 566
+    },
+    {
+      "epoch": 0.42683730121388913,
+      "grad_norm": 3.6439280454672107,
+      "learning_rate": 1.2815197932769486e-05,
+      "loss": 1.0859,
+      "step": 567
+    },
+    {
+      "epoch": 0.42759010068692954,
+      "grad_norm": 4.0189574487408395,
+      "learning_rate": 1.2791784828128727e-05,
+      "loss": 1.1235,
+      "step": 568
+    },
+    {
+      "epoch": 0.4283429001599699,
+      "grad_norm": 3.332167286806501,
+      "learning_rate": 1.2768355114248493e-05,
+      "loss": 1.008,
+      "step": 569
+    },
+    {
+      "epoch": 0.42909569963301025,
+      "grad_norm": 2.7378718594417157,
+      "learning_rate": 1.274490893051981e-05,
+      "loss": 0.9768,
+      "step": 570
+    },
+    {
+      "epoch": 0.4298484991060506,
+      "grad_norm": 3.5007868628431966,
+      "learning_rate": 1.2721446416431676e-05,
+      "loss": 1.0619,
+      "step": 571
+    },
+    {
+      "epoch": 0.430601298579091,
+      "grad_norm": 4.43305491920189,
+      "learning_rate": 1.2697967711570243e-05,
+      "loss": 1.1178,
+      "step": 572
+    },
+    {
+      "epoch": 0.4313540980521314,
+      "grad_norm": 3.963447493013407,
+      "learning_rate": 1.2674472955618001e-05,
+      "loss": 1.013,
+      "step": 573
+    },
+    {
+      "epoch": 0.43210689752517173,
+      "grad_norm": 2.773126731287594,
+      "learning_rate": 1.2650962288352916e-05,
+      "loss": 1.0851,
+      "step": 574
+    },
+    {
+      "epoch": 0.4328596969982121,
+      "grad_norm": 3.639961417112135,
+      "learning_rate": 1.2627435849647629e-05,
+      "loss": 1.0373,
+      "step": 575
+    },
+    {
+      "epoch": 0.4336124964712525,
+      "grad_norm": 3.0305761457545684,
+      "learning_rate": 1.2603893779468604e-05,
+      "loss": 1.0637,
+      "step": 576
+    },
+    {
+      "epoch": 0.43436529594429285,
+      "grad_norm": 2.4310238944919726,
+      "learning_rate": 1.2580336217875303e-05,
+      "loss": 1.0316,
+      "step": 577
+    },
+    {
+      "epoch": 0.4351180954173332,
+      "grad_norm": 3.3566129974599095,
+      "learning_rate": 1.2556763305019353e-05,
+      "loss": 1.002,
+      "step": 578
+    },
+    {
+      "epoch": 0.43587089489037356,
+      "grad_norm": 3.332157255311208,
+      "learning_rate": 1.2533175181143704e-05,
+      "loss": 1.0369,
+      "step": 579
+    },
+    {
+      "epoch": 0.43662369436341397,
+      "grad_norm": 3.296409695390347,
+      "learning_rate": 1.2509571986581814e-05,
+      "loss": 1.0249,
+      "step": 580
+    },
+    {
+      "epoch": 0.4373764938364543,
+      "grad_norm": 3.104522070073551,
+      "learning_rate": 1.248595386175679e-05,
+      "loss": 1.0295,
+      "step": 581
+    },
+    {
+      "epoch": 0.4381292933094947,
+      "grad_norm": 3.702230092409405,
+      "learning_rate": 1.2462320947180565e-05,
+      "loss": 1.0505,
+      "step": 582
+    },
+    {
+      "epoch": 0.43888209278253504,
+      "grad_norm": 2.9464845086562876,
+      "learning_rate": 1.2438673383453073e-05,
+      "loss": 1.0431,
+      "step": 583
+    },
+    {
+      "epoch": 0.4396348922555754,
+      "grad_norm": 3.245101732744253,
+      "learning_rate": 1.241501131126138e-05,
+      "loss": 1.0038,
+      "step": 584
+    },
+    {
+      "epoch": 0.4403876917286158,
+      "grad_norm": 3.1816104157227207,
+      "learning_rate": 1.239133487137889e-05,
+      "loss": 1.0419,
+      "step": 585
+    },
+    {
+      "epoch": 0.44114049120165616,
+      "grad_norm": 4.017669311902771,
+      "learning_rate": 1.2367644204664468e-05,
+      "loss": 1.0518,
+      "step": 586
+    },
+    {
+      "epoch": 0.4418932906746965,
+      "grad_norm": 3.2156530752080363,
+      "learning_rate": 1.2343939452061628e-05,
+      "loss": 1.0116,
+      "step": 587
+    },
+    {
+      "epoch": 0.44264609014773687,
+      "grad_norm": 2.7530854094979818,
+      "learning_rate": 1.232022075459768e-05,
+      "loss": 1.0057,
+      "step": 588
+    },
+    {
+      "epoch": 0.4433988896207773,
+      "grad_norm": 2.4845646778491157,
+      "learning_rate": 1.2296488253382902e-05,
+      "loss": 0.9812,
+      "step": 589
+    },
+    {
+      "epoch": 0.44415168909381764,
+      "grad_norm": 2.771993461799015,
+      "learning_rate": 1.2272742089609694e-05,
+      "loss": 1.0233,
+      "step": 590
+    },
+    {
+      "epoch": 0.444904488566858,
+      "grad_norm": 3.2638541440257174,
+      "learning_rate": 1.2248982404551733e-05,
+      "loss": 1.1117,
+      "step": 591
+    },
+    {
+      "epoch": 0.44565728803989835,
+      "grad_norm": 3.019401928494119,
+      "learning_rate": 1.2225209339563144e-05,
+      "loss": 1.0764,
+      "step": 592
+    },
+    {
+      "epoch": 0.44641008751293876,
+      "grad_norm": 2.5950884293437433,
+      "learning_rate": 1.2201423036077657e-05,
+      "loss": 0.9995,
+      "step": 593
+    },
+    {
+      "epoch": 0.4471628869859791,
+      "grad_norm": 3.1737029171437836,
+      "learning_rate": 1.2177623635607753e-05,
+      "loss": 1.0109,
+      "step": 594
+    },
+    {
+      "epoch": 0.44791568645901947,
+      "grad_norm": 2.73228114214597,
+      "learning_rate": 1.2153811279743841e-05,
+      "loss": 0.9672,
+      "step": 595
+    },
+    {
+      "epoch": 0.4486684859320598,
+      "grad_norm": 3.2642865460623693,
+      "learning_rate": 1.2129986110153395e-05,
+      "loss": 1.0576,
+      "step": 596
+    },
+    {
+      "epoch": 0.44942128540510023,
+      "grad_norm": 2.703014706115608,
+      "learning_rate": 1.210614826858013e-05,
+      "loss": 1.0316,
+      "step": 597
+    },
+    {
+      "epoch": 0.4501740848781406,
+      "grad_norm": 4.575951156022697,
+      "learning_rate": 1.208229789684315e-05,
+      "loss": 1.0492,
+      "step": 598
+    },
+    {
+      "epoch": 0.45092688435118095,
+      "grad_norm": 4.2735868522314036,
+      "learning_rate": 1.2058435136836101e-05,
+      "loss": 0.9975,
+      "step": 599
+    },
+    {
+      "epoch": 0.4516796838242213,
+      "grad_norm": 3.152733804810423,
+      "learning_rate": 1.2034560130526341e-05,
+      "loss": 1.0168,
+      "step": 600
+    },
+    {
+      "epoch": 0.4524324832972617,
+      "grad_norm": 4.0000821775666475,
+      "learning_rate": 1.201067301995407e-05,
+      "loss": 1.1154,
+      "step": 601
+    },
+    {
+      "epoch": 0.45318528277030207,
+      "grad_norm": 4.068503998540418,
+      "learning_rate": 1.1986773947231505e-05,
+      "loss": 1.0289,
+      "step": 602
+    },
+    {
+      "epoch": 0.4539380822433424,
+      "grad_norm": 2.6788756306717594,
+      "learning_rate": 1.1962863054542045e-05,
+      "loss": 0.9589,
+      "step": 603
+    },
+    {
+      "epoch": 0.4546908817163828,
+      "grad_norm": 5.012791771168218,
+      "learning_rate": 1.1938940484139387e-05,
+      "loss": 1.1051,
+      "step": 604
+    },
+    {
+      "epoch": 0.4554436811894232,
+      "grad_norm": 3.521967279714444,
+      "learning_rate": 1.1915006378346719e-05,
+      "loss": 1.0553,
+      "step": 605
+    },
+    {
+      "epoch": 0.45619648066246354,
+      "grad_norm": 3.616593755479914,
+      "learning_rate": 1.1891060879555847e-05,
+      "loss": 1.102,
+      "step": 606
+    },
+    {
+      "epoch": 0.4569492801355039,
+      "grad_norm": 2.9722037476669496,
+      "learning_rate": 1.1867104130226363e-05,
+      "loss": 1.0029,
+      "step": 607
+    },
+    {
+      "epoch": 0.45770207960854425,
+      "grad_norm": 3.4759851155141765,
+      "learning_rate": 1.1843136272884795e-05,
+      "loss": 1.0905,
+      "step": 608
+    },
+    {
+      "epoch": 0.45845487908158467,
+      "grad_norm": 2.7210451462852325,
+      "learning_rate": 1.1819157450123745e-05,
+      "loss": 1.0133,
+      "step": 609
+    },
+    {
+      "epoch": 0.459207678554625,
+      "grad_norm": 2.572254919121521,
+      "learning_rate": 1.1795167804601062e-05,
+      "loss": 0.9759,
+      "step": 610
+    },
+    {
+      "epoch": 0.4599604780276654,
+      "grad_norm": 2.724723345376094,
+      "learning_rate": 1.1771167479038978e-05,
+      "loss": 1.0532,
+      "step": 611
+    },
+    {
+      "epoch": 0.46071327750070573,
+      "grad_norm": 3.4328753233271163,
+      "learning_rate": 1.1747156616223272e-05,
+      "loss": 1.0223,
+      "step": 612
+    },
+    {
+      "epoch": 0.46146607697374614,
+      "grad_norm": 3.0252287414459125,
+      "learning_rate": 1.1723135359002403e-05,
+      "loss": 0.9849,
+      "step": 613
+    },
+    {
+      "epoch": 0.4622188764467865,
+      "grad_norm": 3.0032309037729648,
+      "learning_rate": 1.1699103850286668e-05,
+      "loss": 1.0028,
+      "step": 614
+    },
+    {
+      "epoch": 0.46297167591982685,
+      "grad_norm": 2.8243286910732905,
+      "learning_rate": 1.1675062233047365e-05,
+      "loss": 1.05,
+      "step": 615
+    },
+    {
+      "epoch": 0.4637244753928672,
+      "grad_norm": 3.5574655975394984,
+      "learning_rate": 1.1651010650315923e-05,
+      "loss": 1.0497,
+      "step": 616
+    },
+    {
+      "epoch": 0.4644772748659076,
+      "grad_norm": 3.022425375669109,
+      "learning_rate": 1.1626949245183061e-05,
+      "loss": 0.9807,
+      "step": 617
+    },
+    {
+      "epoch": 0.465230074338948,
+      "grad_norm": 2.771846295903394,
+      "learning_rate": 1.1602878160797936e-05,
+      "loss": 1.076,
+      "step": 618
+    },
+    {
+      "epoch": 0.46598287381198833,
+      "grad_norm": 2.505321802846061,
+      "learning_rate": 1.1578797540367284e-05,
+      "loss": 0.9784,
+      "step": 619
+    },
+    {
+      "epoch": 0.4667356732850287,
+      "grad_norm": 2.983970077564599,
+      "learning_rate": 1.155470752715458e-05,
+      "loss": 0.959,
+      "step": 620
+    },
+    {
+      "epoch": 0.4674884727580691,
+      "grad_norm": 4.383473009689115,
+      "learning_rate": 1.153060826447918e-05,
+      "loss": 0.986,
+      "step": 621
+    },
+    {
+      "epoch": 0.46824127223110945,
+      "grad_norm": 3.4488742706032367,
+      "learning_rate": 1.1506499895715462e-05,
+      "loss": 1.0913,
+      "step": 622
+    },
+    {
+      "epoch": 0.4689940717041498,
+      "grad_norm": 2.6140008129721517,
+      "learning_rate": 1.148238256429199e-05,
+      "loss": 0.9833,
+      "step": 623
+    },
+    {
+      "epoch": 0.46974687117719016,
+      "grad_norm": 2.92309907991184,
+      "learning_rate": 1.1458256413690634e-05,
+      "loss": 1.0083,
+      "step": 624
+    },
+    {
+      "epoch": 0.4704996706502306,
+      "grad_norm": 2.724708243419771,
+      "learning_rate": 1.1434121587445752e-05,
+      "loss": 1.0594,
+      "step": 625
+    },
+    {
+      "epoch": 0.47125247012327093,
+      "grad_norm": 3.0724226680282367,
+      "learning_rate": 1.1409978229143297e-05,
+      "loss": 1.0204,
+      "step": 626
+    },
+    {
+      "epoch": 0.4720052695963113,
+      "grad_norm": 3.0671644057682923,
+      "learning_rate": 1.1385826482419993e-05,
+      "loss": 1.0139,
+      "step": 627
+    },
+    {
+      "epoch": 0.47275806906935164,
+      "grad_norm": 3.1314240764279364,
+      "learning_rate": 1.1361666490962468e-05,
+      "loss": 1.0052,
+      "step": 628
+    },
+    {
+      "epoch": 0.473510868542392,
+      "grad_norm": 3.3855180309761237,
+      "learning_rate": 1.1337498398506397e-05,
+      "loss": 0.9997,
+      "step": 629
+    },
+    {
+      "epoch": 0.4742636680154324,
+      "grad_norm": 3.1909270618792958,
+      "learning_rate": 1.1313322348835658e-05,
+      "loss": 1.0612,
+      "step": 630
+    },
+    {
+      "epoch": 0.47501646748847276,
+      "grad_norm": 3.783620789630306,
+      "learning_rate": 1.1289138485781456e-05,
+      "loss": 1.0032,
+      "step": 631
+    },
+    {
+      "epoch": 0.4757692669615131,
+      "grad_norm": 4.229526141420337,
+      "learning_rate": 1.1264946953221496e-05,
+      "loss": 1.0331,
+      "step": 632
+    },
+    {
+      "epoch": 0.47652206643455347,
+      "grad_norm": 4.6378584634680715,
+      "learning_rate": 1.12407478950791e-05,
+      "loss": 1.0362,
+      "step": 633
+    },
+    {
+      "epoch": 0.4772748659075939,
+      "grad_norm": 2.6228761321729452,
+      "learning_rate": 1.1216541455322367e-05,
+      "loss": 1.0012,
+      "step": 634
+    },
+    {
+      "epoch": 0.47802766538063424,
+      "grad_norm": 2.9654511830251535,
+      "learning_rate": 1.1192327777963313e-05,
+      "loss": 1.0351,
+      "step": 635
+    },
+    {
+      "epoch": 0.4787804648536746,
+      "grad_norm": 2.5847394351824406,
+      "learning_rate": 1.1168107007057006e-05,
+      "loss": 0.9708,
+      "step": 636
+    },
+    {
+      "epoch": 0.47953326432671495,
+      "grad_norm": 3.838509319191613,
+      "learning_rate": 1.1143879286700723e-05,
+      "loss": 0.9432,
+      "step": 637
+    },
+    {
+      "epoch": 0.48028606379975536,
+      "grad_norm": 2.817996686530098,
+      "learning_rate": 1.1119644761033079e-05,
+      "loss": 0.9694,
+      "step": 638
+    },
+    {
+      "epoch": 0.4810388632727957,
+      "grad_norm": 3.531105963283091,
+      "learning_rate": 1.1095403574233185e-05,
+      "loss": 1.0909,
+      "step": 639
+    },
+    {
+      "epoch": 0.48179166274583607,
+      "grad_norm": 3.3417816286597213,
+      "learning_rate": 1.1071155870519777e-05,
+      "loss": 1.1137,
+      "step": 640
+    },
+    {
+      "epoch": 0.4825444622188764,
+      "grad_norm": 3.5562409238146317,
+      "learning_rate": 1.1046901794150358e-05,
+      "loss": 1.0303,
+      "step": 641
+    },
+    {
+      "epoch": 0.48329726169191684,
+      "grad_norm": 2.907782910105036,
+      "learning_rate": 1.1022641489420342e-05,
+      "loss": 1.0336,
+      "step": 642
+    },
+    {
+      "epoch": 0.4840500611649572,
+      "grad_norm": 2.970163177761591,
+      "learning_rate": 1.0998375100662215e-05,
+      "loss": 1.0605,
+      "step": 643
+    },
+    {
+      "epoch": 0.48480286063799755,
+      "grad_norm": 3.2772867501044582,
+      "learning_rate": 1.0974102772244638e-05,
+      "loss": 1.0235,
+      "step": 644
+    },
+    {
+      "epoch": 0.4855556601110379,
+      "grad_norm": 3.497468818702157,
+      "learning_rate": 1.094982464857162e-05,
+      "loss": 1.0848,
+      "step": 645
+    },
+    {
+      "epoch": 0.4863084595840783,
+      "grad_norm": 3.3225147360774403,
+      "learning_rate": 1.0925540874081649e-05,
+      "loss": 0.9684,
+      "step": 646
+    },
+    {
+      "epoch": 0.48706125905711867,
+      "grad_norm": 3.2301417809169575,
+      "learning_rate": 1.0901251593246822e-05,
+      "loss": 0.9827,
+      "step": 647
+    },
+    {
+      "epoch": 0.487814058530159,
+      "grad_norm": 3.636896054687057,
+      "learning_rate": 1.0876956950572006e-05,
+      "loss": 1.065,
+      "step": 648
+    },
+    {
+      "epoch": 0.4885668580031994,
+      "grad_norm": 2.932810236288748,
+      "learning_rate": 1.0852657090593961e-05,
+      "loss": 1.0698,
+      "step": 649
+    },
+    {
+      "epoch": 0.4893196574762398,
+      "grad_norm": 5.0895461182195385,
+      "learning_rate": 1.0828352157880489e-05,
+      "loss": 1.1485,
+      "step": 650
+    },
+    {
+      "epoch": 0.49007245694928014,
+      "grad_norm": 3.1415226619078616,
+      "learning_rate": 1.0804042297029567e-05,
+      "loss": 0.9901,
+      "step": 651
+    },
+    {
+      "epoch": 0.4908252564223205,
+      "grad_norm": 2.909255583040199,
+      "learning_rate": 1.0779727652668496e-05,
+      "loss": 1.0655,
+      "step": 652
+    },
+    {
+      "epoch": 0.49157805589536085,
+      "grad_norm": 3.1548858316494877,
+      "learning_rate": 1.075540836945304e-05,
+      "loss": 1.0486,
+      "step": 653
+    },
+    {
+      "epoch": 0.49233085536840127,
+      "grad_norm": 2.7012241871329614,
+      "learning_rate": 1.0731084592066548e-05,
+      "loss": 0.9917,
+      "step": 654
+    },
+    {
+      "epoch": 0.4930836548414416,
+      "grad_norm": 3.0157551767075796,
+      "learning_rate": 1.0706756465219114e-05,
+      "loss": 1.0596,
+      "step": 655
+    },
+    {
+      "epoch": 0.493836454314482,
+      "grad_norm": 2.8704467999311367,
+      "learning_rate": 1.0682424133646712e-05,
+      "loss": 1.0269,
+      "step": 656
+    },
+    {
+      "epoch": 0.49458925378752233,
+      "grad_norm": 3.0763751507732127,
+      "learning_rate": 1.0658087742110322e-05,
+      "loss": 1.0799,
+      "step": 657
+    },
+    {
+      "epoch": 0.49534205326056274,
+      "grad_norm": 3.9020236991450856,
+      "learning_rate": 1.063374743539509e-05,
+      "loss": 1.0025,
+      "step": 658
+    },
+    {
+      "epoch": 0.4960948527336031,
+      "grad_norm": 2.609346679016363,
+      "learning_rate": 1.060940335830944e-05,
+      "loss": 0.9835,
+      "step": 659
+    },
+    {
+      "epoch": 0.49684765220664345,
+      "grad_norm": 3.343369949839618,
+      "learning_rate": 1.058505565568424e-05,
+      "loss": 0.9486,
+      "step": 660
+    },
+    {
+      "epoch": 0.4976004516796838,
+      "grad_norm": 3.3313727642441795,
+      "learning_rate": 1.0560704472371919e-05,
+      "loss": 0.9556,
+      "step": 661
+    },
+    {
+      "epoch": 0.4983532511527242,
+      "grad_norm": 3.4155555321239355,
+      "learning_rate": 1.0536349953245622e-05,
+      "loss": 1.0095,
+      "step": 662
+    },
+    {
+      "epoch": 0.4991060506257646,
+      "grad_norm": 2.480958118144923,
+      "learning_rate": 1.0511992243198335e-05,
+      "loss": 0.9385,
+      "step": 663
+    },
+    {
+      "epoch": 0.49985885009880493,
+      "grad_norm": 3.5008683649636816,
+      "learning_rate": 1.0487631487142018e-05,
+      "loss": 1.0369,
+      "step": 664
+    },
+    {
+      "epoch": 0.5006116495718453,
+      "grad_norm": 2.663416310535519,
+      "learning_rate": 1.0463267830006779e-05,
+      "loss": 1.0242,
+      "step": 665
+    },
+    {
+      "epoch": 0.5006116495718453,
+      "eval_loss": 0.9954760074615479,
+      "eval_runtime": 584.4864,
+      "eval_samples_per_second": 30.6,
+      "eval_steps_per_second": 0.479,
+      "step": 665
+    },
+    {
+      "epoch": 0.5013644490448856,
+      "grad_norm": 3.1777588089309208,
+      "learning_rate": 1.0438901416739955e-05,
+      "loss": 1.0186,
+      "step": 666
+    },
+    {
+      "epoch": 0.502117248517926,
+      "grad_norm": 2.7838827384508944,
+      "learning_rate": 1.0414532392305301e-05,
+      "loss": 0.9709,
+      "step": 667
+    },
+    {
+      "epoch": 0.5028700479909664,
+      "grad_norm": 2.921400217102283,
+      "learning_rate": 1.03901609016821e-05,
+      "loss": 1.0714,
+      "step": 668
+    },
+    {
+      "epoch": 0.5036228474640068,
+      "grad_norm": 2.7341839637863847,
+      "learning_rate": 1.0365787089864303e-05,
+      "loss": 0.9927,
+      "step": 669
+    },
+    {
+      "epoch": 0.5043756469370472,
+      "grad_norm": 4.914922149938216,
+      "learning_rate": 1.034141110185968e-05,
+      "loss": 1.1115,
+      "step": 670
+    },
+    {
+      "epoch": 0.5051284464100875,
+      "grad_norm": 2.989888013177945,
+      "learning_rate": 1.031703308268894e-05,
+      "loss": 0.9944,
+      "step": 671
+    },
+    {
+      "epoch": 0.5058812458831279,
+      "grad_norm": 2.579049180170171,
+      "learning_rate": 1.0292653177384878e-05,
+      "loss": 0.9751,
+      "step": 672
+    },
+    {
+      "epoch": 0.5066340453561683,
+      "grad_norm": 3.959139485335877,
+      "learning_rate": 1.0268271530991509e-05,
+      "loss": 1.0733,
+      "step": 673
+    },
+    {
+      "epoch": 0.5073868448292086,
+      "grad_norm": 2.600824437688288,
+      "learning_rate": 1.0243888288563213e-05,
+      "loss": 0.9694,
+      "step": 674
+    },
+    {
+      "epoch": 0.508139644302249,
+      "grad_norm": 2.5090320928335075,
+      "learning_rate": 1.0219503595163857e-05,
+      "loss": 1.0231,
+      "step": 675
+    },
+    {
+      "epoch": 0.5088924437752893,
+      "grad_norm": 2.6849288835400453,
+      "learning_rate": 1.0195117595865947e-05,
+      "loss": 1.0139,
+      "step": 676
+    },
+    {
+      "epoch": 0.5096452432483297,
+      "grad_norm": 3.682029203578967,
+      "learning_rate": 1.017073043574975e-05,
+      "loss": 0.9919,
+      "step": 677
+    },
+    {
+      "epoch": 0.5103980427213701,
+      "grad_norm": 3.275453418274055,
+      "learning_rate": 1.0146342259902446e-05,
+      "loss": 1.0953,
+      "step": 678
+    },
+    {
+      "epoch": 0.5111508421944104,
+      "grad_norm": 3.0421924924609907,
+      "learning_rate": 1.0121953213417256e-05,
+      "loss": 1.0479,
+      "step": 679
+    },
+    {
+      "epoch": 0.5119036416674508,
+      "grad_norm": 3.4230363471287273,
+      "learning_rate": 1.0097563441392582e-05,
+      "loss": 1.0352,
+      "step": 680
+    },
+    {
+      "epoch": 0.5126564411404912,
+      "grad_norm": 3.21566163229416,
+      "learning_rate": 1.0073173088931143e-05,
+      "loss": 0.9818,
+      "step": 681
+    },
+    {
+      "epoch": 0.5134092406135315,
+      "grad_norm": 2.729633411998688,
+      "learning_rate": 1.0048782301139102e-05,
+      "loss": 0.9991,
+      "step": 682
+    },
+    {
+      "epoch": 0.514162040086572,
+      "grad_norm": 2.9671555111696284,
+      "learning_rate": 1.0024391223125226e-05,
+      "loss": 1.0071,
+      "step": 683
+    },
+    {
+      "epoch": 0.5149148395596123,
+      "grad_norm": 3.8434580170431665,
+      "learning_rate": 1e-05,
+      "loss": 1.0889,
+      "step": 684
+    },
+    {
+      "epoch": 0.5156676390326527,
+      "grad_norm": 2.7977469550737184,
+      "learning_rate": 9.975608776874775e-06,
+      "loss": 1.029,
+      "step": 685
+    },
+    {
+      "epoch": 0.5164204385056931,
+      "grad_norm": 3.312999685049618,
+      "learning_rate": 9.951217698860902e-06,
+      "loss": 1.0791,
+      "step": 686
+    },
+    {
+      "epoch": 0.5171732379787334,
+      "grad_norm": 2.8531522744975706,
+      "learning_rate": 9.926826911068862e-06,
+      "loss": 0.9893,
+      "step": 687
+    },
+    {
+      "epoch": 0.5179260374517738,
+      "grad_norm": 3.2587348054443948,
+      "learning_rate": 9.90243655860742e-06,
+      "loss": 1.0307,
+      "step": 688
+    },
+    {
+      "epoch": 0.5186788369248142,
+      "grad_norm": 2.783794808143798,
+      "learning_rate": 9.878046786582745e-06,
+      "loss": 1.042,
+      "step": 689
+    },
+    {
+      "epoch": 0.5194316363978545,
+      "grad_norm": 2.6862927134575414,
+      "learning_rate": 9.853657740097558e-06,
+      "loss": 1.0261,
+      "step": 690
+    },
+    {
+      "epoch": 0.5201844358708949,
+      "grad_norm": 2.997479231338951,
+      "learning_rate": 9.829269564250254e-06,
+      "loss": 1.0427,
+      "step": 691
+    },
+    {
+      "epoch": 0.5209372353439352,
+      "grad_norm": 2.4242881837479464,
+      "learning_rate": 9.804882404134057e-06,
+      "loss": 0.928,
+      "step": 692
+    },
+    {
+      "epoch": 0.5216900348169756,
+      "grad_norm": 2.745092956089842,
+      "learning_rate": 9.780496404836146e-06,
+      "loss": 1.037,
+      "step": 693
+    },
+    {
+      "epoch": 0.522442834290016,
+      "grad_norm": 3.03917646598378,
+      "learning_rate": 9.75611171143679e-06,
+      "loss": 0.9335,
+      "step": 694
+    },
+    {
+      "epoch": 0.5231956337630563,
+      "grad_norm": 2.7717089194586713,
+      "learning_rate": 9.731728469008493e-06,
+      "loss": 1.0349,
+      "step": 695
+    },
+    {
+      "epoch": 0.5239484332360967,
+      "grad_norm": 4.154071155669152,
+      "learning_rate": 9.707346822615127e-06,
+      "loss": 1.0371,
+      "step": 696
+    },
+    {
+      "epoch": 0.5247012327091372,
+      "grad_norm": 2.888871286357065,
+      "learning_rate": 9.682966917311065e-06,
+      "loss": 0.9778,
+      "step": 697
+    },
+    {
+      "epoch": 0.5254540321821775,
+      "grad_norm": 2.598234744411881,
+      "learning_rate": 9.658588898140322e-06,
+      "loss": 0.9517,
+      "step": 698
+    },
+    {
+      "epoch": 0.5262068316552179,
+      "grad_norm": 2.87901406826843,
+      "learning_rate": 9.634212910135697e-06,
+      "loss": 1.0629,
+      "step": 699
+    },
+    {
+      "epoch": 0.5269596311282582,
+      "grad_norm": 2.3502091823270153,
+      "learning_rate": 9.609839098317902e-06,
+      "loss": 0.9472,
+      "step": 700
+    },
+    {
+      "epoch": 0.5277124306012986,
+      "grad_norm": 2.3727300076429043,
+      "learning_rate": 9.585467607694702e-06,
+      "loss": 0.9752,
+      "step": 701
+    },
+    {
+      "epoch": 0.528465230074339,
+      "grad_norm": 3.0992227105534766,
+      "learning_rate": 9.561098583260047e-06,
+      "loss": 1.0605,
+      "step": 702
+    },
+    {
+      "epoch": 0.5292180295473793,
+      "grad_norm": 2.998573259606581,
+      "learning_rate": 9.536732169993225e-06,
+      "loss": 1.0572,
+      "step": 703
+    },
+    {
+      "epoch": 0.5299708290204197,
+      "grad_norm": 3.0645514660070363,
+      "learning_rate": 9.512368512857983e-06,
+      "loss": 1.0209,
+      "step": 704
+    },
+    {
+      "epoch": 0.5307236284934601,
+      "grad_norm": 2.7686653578062446,
+      "learning_rate": 9.488007756801672e-06,
+      "loss": 0.9782,
+      "step": 705
+    },
+    {
+      "epoch": 0.5314764279665004,
+      "grad_norm": 2.419321874756083,
+      "learning_rate": 9.463650046754383e-06,
+      "loss": 0.9543,
+      "step": 706
+    },
+    {
+      "epoch": 0.5322292274395408,
+      "grad_norm": 3.2183835254514923,
+      "learning_rate": 9.439295527628083e-06,
+      "loss": 1.0494,
+      "step": 707
+    },
+    {
+      "epoch": 0.5329820269125811,
+      "grad_norm": 2.7493248848894773,
+      "learning_rate": 9.414944344315765e-06,
+      "loss": 1.0513,
+      "step": 708
+    },
+    {
+      "epoch": 0.5337348263856215,
+      "grad_norm": 2.5720306128631054,
+      "learning_rate": 9.390596641690563e-06,
+      "loss": 0.9475,
+      "step": 709
+    },
+    {
+      "epoch": 0.5344876258586619,
+      "grad_norm": 2.6083152242917804,
+      "learning_rate": 9.366252564604914e-06,
+      "loss": 0.9755,
+      "step": 710
+    },
+    {
+      "epoch": 0.5352404253317022,
+      "grad_norm": 2.49703677714488,
+      "learning_rate": 9.34191225788968e-06,
+      "loss": 0.9997,
+      "step": 711
+    },
+    {
+      "epoch": 0.5359932248047427,
+      "grad_norm": 3.556514244662279,
+      "learning_rate": 9.317575866353293e-06,
+      "loss": 1.0298,
+      "step": 712
+    },
+    {
+      "epoch": 0.536746024277783,
+      "grad_norm": 2.754748122306969,
+      "learning_rate": 9.293243534780887e-06,
+      "loss": 1.0231,
+      "step": 713
+    },
+    {
+      "epoch": 0.5374988237508234,
+      "grad_norm": 2.458781472253963,
+      "learning_rate": 9.268915407933457e-06,
+      "loss": 0.9993,
+      "step": 714
+    },
+    {
+      "epoch": 0.5382516232238638,
+      "grad_norm": 3.1095834663708013,
+      "learning_rate": 9.244591630546964e-06,
+      "loss": 1.0862,
+      "step": 715
+    },
+    {
+      "epoch": 0.5390044226969041,
+      "grad_norm": 3.35070855022117,
+      "learning_rate": 9.220272347331502e-06,
+      "loss": 0.9522,
+      "step": 716
+    },
+    {
+      "epoch": 0.5397572221699445,
+      "grad_norm": 4.631416487344075,
+      "learning_rate": 9.195957702970434e-06,
+      "loss": 1.0246,
+      "step": 717
+    },
+    {
+      "epoch": 0.5405100216429849,
+      "grad_norm": 2.7689299663326112,
+      "learning_rate": 9.171647842119515e-06,
+      "loss": 1.0256,
+      "step": 718
+    },
+    {
+      "epoch": 0.5412628211160252,
+      "grad_norm": 3.049609437321965,
+      "learning_rate": 9.14734290940604e-06,
+      "loss": 1.0002,
+      "step": 719
+    },
+    {
+      "epoch": 0.5420156205890656,
+      "grad_norm": 2.795960963250181,
+      "learning_rate": 9.123043049427996e-06,
+      "loss": 1.0337,
+      "step": 720
+    },
+    {
+      "epoch": 0.5427684200621059,
+      "grad_norm": 2.8255346166814914,
+      "learning_rate": 9.098748406753181e-06,
+      "loss": 0.9985,
+      "step": 721
+    },
+    {
+      "epoch": 0.5435212195351463,
+      "grad_norm": 2.7816767930156137,
+      "learning_rate": 9.074459125918356e-06,
+      "loss": 1.0088,
+      "step": 722
+    },
+    {
+      "epoch": 0.5442740190081867,
+      "grad_norm": 2.9163237872445054,
+      "learning_rate": 9.050175351428381e-06,
+      "loss": 0.9896,
+      "step": 723
+    },
+    {
+      "epoch": 0.545026818481227,
+      "grad_norm": 2.7100301901021266,
+      "learning_rate": 9.025897227755367e-06,
+      "loss": 1.0298,
+      "step": 724
+    },
+    {
+      "epoch": 0.5457796179542674,
+      "grad_norm": 3.6617428934101586,
+      "learning_rate": 9.001624899337785e-06,
+      "loss": 0.9469,
+      "step": 725
+    },
+    {
+      "epoch": 0.5465324174273078,
+      "grad_norm": 3.2124650336329474,
+      "learning_rate": 8.977358510579658e-06,
+      "loss": 1.0152,
+      "step": 726
+    },
+    {
+      "epoch": 0.5472852169003481,
+      "grad_norm": 2.7297134092753077,
+      "learning_rate": 8.953098205849647e-06,
+      "loss": 1.0301,
+      "step": 727
+    },
+    {
+      "epoch": 0.5480380163733886,
+      "grad_norm": 3.555015522849158,
+      "learning_rate": 8.928844129480228e-06,
+      "loss": 1.0335,
+      "step": 728
+    },
+    {
+      "epoch": 0.5487908158464289,
+      "grad_norm": 2.9032937240322525,
+      "learning_rate": 8.904596425766817e-06,
+      "loss": 1.067,
+      "step": 729
+    },
+    {
+      "epoch": 0.5495436153194693,
+      "grad_norm": 2.8183123864129644,
+      "learning_rate": 8.880355238966923e-06,
+      "loss": 0.9862,
+      "step": 730
+    },
+    {
+      "epoch": 0.5502964147925097,
+      "grad_norm": 3.380735555760305,
+      "learning_rate": 8.856120713299284e-06,
+      "loss": 1.1157,
+      "step": 731
+    },
+    {
+      "epoch": 0.55104921426555,
+      "grad_norm": 3.4648885692214804,
+      "learning_rate": 8.831892992943e-06,
+      "loss": 1.0331,
+      "step": 732
+    },
+    {
+      "epoch": 0.5518020137385904,
+      "grad_norm": 3.0630177296042516,
+      "learning_rate": 8.807672222036692e-06,
+      "loss": 1.0538,
+      "step": 733
+    },
+    {
+      "epoch": 0.5525548132116308,
+      "grad_norm": 2.549871690933666,
+      "learning_rate": 8.783458544677633e-06,
+      "loss": 1.0284,
+      "step": 734
+    },
+    {
+      "epoch": 0.5533076126846711,
+      "grad_norm": 3.211702233203899,
+      "learning_rate": 8.7592521049209e-06,
+      "loss": 1.0029,
+      "step": 735
+    },
+    {
+      "epoch": 0.5540604121577115,
+      "grad_norm": 3.274729428072235,
+      "learning_rate": 8.735053046778506e-06,
+      "loss": 0.9601,
+      "step": 736
+    },
+    {
+      "epoch": 0.5548132116307518,
+      "grad_norm": 3.115868299982563,
+      "learning_rate": 8.710861514218545e-06,
+      "loss": 0.9472,
+      "step": 737
+    },
+    {
+      "epoch": 0.5555660111037922,
+      "grad_norm": 2.5731947785457154,
+      "learning_rate": 8.686677651164345e-06,
+      "loss": 0.9086,
+      "step": 738
+    },
+    {
+      "epoch": 0.5563188105768326,
+      "grad_norm": 2.5740093256448833,
+      "learning_rate": 8.662501601493607e-06,
+      "loss": 1.0467,
+      "step": 739
+    },
+    {
+      "epoch": 0.5570716100498729,
+      "grad_norm": 4.663822637060085,
+      "learning_rate": 8.638333509037537e-06,
+      "loss": 1.0923,
+      "step": 740
+    },
+    {
+      "epoch": 0.5578244095229133,
+      "grad_norm": 2.5224339575220482,
+      "learning_rate": 8.61417351758001e-06,
+      "loss": 1.0176,
+      "step": 741
+    },
+    {
+      "epoch": 0.5585772089959538,
+      "grad_norm": 2.6320547385384403,
+      "learning_rate": 8.590021770856708e-06,
+      "loss": 1.0732,
+      "step": 742
+    },
+    {
+      "epoch": 0.559330008468994,
+      "grad_norm": 3.0609203768889,
+      "learning_rate": 8.565878412554251e-06,
+      "loss": 1.0025,
+      "step": 743
+    },
+    {
+      "epoch": 0.5600828079420345,
+      "grad_norm": 3.0482117277870797,
+      "learning_rate": 8.541743586309366e-06,
+      "loss": 1.0,
+      "step": 744
+    },
+    {
+      "epoch": 0.5608356074150748,
+      "grad_norm": 2.9710905213118557,
+      "learning_rate": 8.517617435708011e-06,
+      "loss": 1.0132,
+      "step": 745
+    },
+    {
+      "epoch": 0.5615884068881152,
+      "grad_norm": 3.3175141891426647,
+      "learning_rate": 8.493500104284539e-06,
+      "loss": 0.9257,
+      "step": 746
+    },
+    {
+      "epoch": 0.5623412063611556,
+      "grad_norm": 2.508169904735289,
+      "learning_rate": 8.469391735520824e-06,
+      "loss": 0.938,
+      "step": 747
+    },
+    {
+      "epoch": 0.5630940058341959,
+      "grad_norm": 2.707186967130032,
+      "learning_rate": 8.445292472845423e-06,
+      "loss": 1.0384,
+      "step": 748
+    },
+    {
+      "epoch": 0.5638468053072363,
+      "grad_norm": 2.477550044042414,
+      "learning_rate": 8.42120245963272e-06,
+      "loss": 0.9575,
+      "step": 749
+    },
+    {
+      "epoch": 0.5645996047802767,
+      "grad_norm": 2.968333454620037,
+      "learning_rate": 8.397121839202069e-06,
+      "loss": 0.9827,
+      "step": 750
+    },
+    {
+      "epoch": 0.565352404253317,
+      "grad_norm": 6.9956574767627435,
+      "learning_rate": 8.373050754816942e-06,
+      "loss": 1.0719,
+      "step": 751
+    },
+    {
+      "epoch": 0.5661052037263574,
+      "grad_norm": 2.9911439956329686,
+      "learning_rate": 8.348989349684077e-06,
+      "loss": 0.9937,
+      "step": 752
+    },
+    {
+      "epoch": 0.5668580031993977,
+      "grad_norm": 2.6387837783947052,
+      "learning_rate": 8.324937766952638e-06,
+      "loss": 0.9247,
+      "step": 753
+    },
+    {
+      "epoch": 0.5676108026724381,
+      "grad_norm": 4.394239465779068,
+      "learning_rate": 8.300896149713334e-06,
+      "loss": 1.0567,
+      "step": 754
+    },
+    {
+      "epoch": 0.5683636021454785,
+      "grad_norm": 3.0051427224594613,
+      "learning_rate": 8.276864640997602e-06,
+      "loss": 1.0641,
+      "step": 755
+    },
+    {
+      "epoch": 0.5691164016185188,
+      "grad_norm": 3.028211050401354,
+      "learning_rate": 8.252843383776731e-06,
+      "loss": 0.9739,
+      "step": 756
+    },
+    {
+      "epoch": 0.5698692010915593,
+      "grad_norm": 2.526597328033783,
+      "learning_rate": 8.228832520961023e-06,
+      "loss": 0.9153,
+      "step": 757
+    },
+    {
+      "epoch": 0.5706220005645996,
+      "grad_norm": 2.701644840333796,
+      "learning_rate": 8.204832195398941e-06,
+      "loss": 1.0112,
+      "step": 758
+    },
+    {
+      "epoch": 0.57137480003764,
+      "grad_norm": 3.7978352979490464,
+      "learning_rate": 8.18084254987626e-06,
+      "loss": 0.9748,
+      "step": 759
+    },
+    {
+      "epoch": 0.5721275995106804,
+      "grad_norm": 3.758272789518827,
+      "learning_rate": 8.15686372711521e-06,
+      "loss": 1.0222,
+      "step": 760
+    },
+    {
+      "epoch": 0.5728803989837207,
+      "grad_norm": 2.5711172001094353,
+      "learning_rate": 8.132895869773638e-06,
+      "loss": 1.0119,
+      "step": 761
+    },
+    {
+      "epoch": 0.5736331984567611,
+      "grad_norm": 2.9242712243012012,
+      "learning_rate": 8.108939120444154e-06,
+      "loss": 0.9453,
+      "step": 762
+    },
+    {
+      "epoch": 0.5743859979298015,
+      "grad_norm": 3.010900564143939,
+      "learning_rate": 8.084993621653283e-06,
+      "loss": 0.9995,
+      "step": 763
+    },
+    {
+      "epoch": 0.5751387974028418,
+      "grad_norm": 3.013143165423945,
+      "learning_rate": 8.061059515860616e-06,
+      "loss": 0.9736,
+      "step": 764
+    },
+    {
+      "epoch": 0.5758915968758822,
+      "grad_norm": 3.600869577617628,
+      "learning_rate": 8.037136945457959e-06,
+      "loss": 1.0206,
+      "step": 765
+    },
+    {
+      "epoch": 0.5766443963489225,
+      "grad_norm": 2.8526015661570656,
+      "learning_rate": 8.013226052768498e-06,
+      "loss": 0.9567,
+      "step": 766
+    },
+    {
+      "epoch": 0.5773971958219629,
+      "grad_norm": 2.9267086899123322,
+      "learning_rate": 7.989326980045937e-06,
+      "loss": 1.021,
+      "step": 767
+    },
+    {
+      "epoch": 0.5781499952950033,
+      "grad_norm": 2.489653573594032,
+      "learning_rate": 7.965439869473664e-06,
+      "loss": 1.0382,
+      "step": 768
+    },
+    {
+      "epoch": 0.5789027947680436,
+      "grad_norm": 2.7327345835934804,
+      "learning_rate": 7.941564863163899e-06,
+      "loss": 1.058,
+      "step": 769
+    },
+    {
+      "epoch": 0.579655594241084,
+      "grad_norm": 3.209412795643892,
+      "learning_rate": 7.91770210315685e-06,
+      "loss": 1.0648,
+      "step": 770
+    },
+    {
+      "epoch": 0.5804083937141244,
+      "grad_norm": 2.9843151463835755,
+      "learning_rate": 7.893851731419872e-06,
+      "loss": 1.0647,
+      "step": 771
+    },
+    {
+      "epoch": 0.5811611931871647,
+      "grad_norm": 3.647581304182318,
+      "learning_rate": 7.870013889846608e-06,
+      "loss": 1.0173,
+      "step": 772
+    },
+    {
+      "epoch": 0.5819139926602052,
+      "grad_norm": 2.367016172308078,
+      "learning_rate": 7.846188720256162e-06,
+      "loss": 0.9627,
+      "step": 773
+    },
+    {
+      "epoch": 0.5826667921332455,
+      "grad_norm": 2.5064853047353646,
+      "learning_rate": 7.822376364392248e-06,
+      "loss": 0.9736,
+      "step": 774
+    },
+    {
+      "epoch": 0.5834195916062859,
+      "grad_norm": 3.3689106910977555,
+      "learning_rate": 7.798576963922347e-06,
+      "loss": 1.033,
+      "step": 775
+    },
+    {
+      "epoch": 0.5841723910793263,
+      "grad_norm": 2.545780240126541,
+      "learning_rate": 7.774790660436857e-06,
+      "loss": 1.0125,
+      "step": 776
+    },
+    {
+      "epoch": 0.5849251905523666,
+      "grad_norm": 2.827075772715696,
+      "learning_rate": 7.75101759544827e-06,
+      "loss": 1.1027,
+      "step": 777
+    },
+    {
+      "epoch": 0.585677990025407,
+      "grad_norm": 2.70846832173852,
+      "learning_rate": 7.727257910390309e-06,
+      "loss": 0.9252,
+      "step": 778
+    },
+    {
+      "epoch": 0.5864307894984474,
+      "grad_norm": 2.8457104629747123,
+      "learning_rate": 7.703511746617098e-06,
+      "loss": 0.8725,
+      "step": 779
+    },
+    {
+      "epoch": 0.5871835889714877,
+      "grad_norm": 3.248698548728837,
+      "learning_rate": 7.679779245402321e-06,
+      "loss": 0.9749,
+      "step": 780
+    },
+    {
+      "epoch": 0.5879363884445281,
+      "grad_norm": 3.1331071482094903,
+      "learning_rate": 7.656060547938375e-06,
+      "loss": 1.0044,
+      "step": 781
+    },
+    {
+      "epoch": 0.5886891879175684,
+      "grad_norm": 3.230980661848815,
+      "learning_rate": 7.632355795335533e-06,
+      "loss": 1.0545,
+      "step": 782
+    },
+    {
+      "epoch": 0.5894419873906088,
+      "grad_norm": 2.8480133523980578,
+      "learning_rate": 7.608665128621111e-06,
+      "loss": 0.9804,
+      "step": 783
+    },
+    {
+      "epoch": 0.5901947868636492,
+      "grad_norm": 3.6918447213822363,
+      "learning_rate": 7.584988688738622e-06,
+      "loss": 0.9664,
+      "step": 784
+    },
+    {
+      "epoch": 0.5909475863366895,
+      "grad_norm": 3.673578400225882,
+      "learning_rate": 7.561326616546932e-06,
+      "loss": 1.0105,
+      "step": 785
+    },
+    {
+      "epoch": 0.59170038580973,
+      "grad_norm": 3.0681486083575766,
+      "learning_rate": 7.5376790528194354e-06,
+      "loss": 0.9658,
+      "step": 786
+    },
+    {
+      "epoch": 0.5924531852827704,
+      "grad_norm": 2.969194265691157,
+      "learning_rate": 7.514046138243211e-06,
+      "loss": 1.037,
+      "step": 787
+    },
+    {
+      "epoch": 0.5932059847558107,
+      "grad_norm": 3.041477785697227,
+      "learning_rate": 7.490428013418187e-06,
+      "loss": 1.033,
+      "step": 788
+    },
+    {
+      "epoch": 0.5939587842288511,
+      "grad_norm": 2.8679574117693085,
+      "learning_rate": 7.466824818856296e-06,
+      "loss": 1.004,
+      "step": 789
+    },
+    {
+      "epoch": 0.5947115837018914,
+      "grad_norm": 2.2901391036080345,
+      "learning_rate": 7.443236694980649e-06,
+      "loss": 0.8783,
+      "step": 790
+    },
+    {
+      "epoch": 0.5954643831749318,
+      "grad_norm": 3.3270607439893025,
+      "learning_rate": 7.4196637821247e-06,
+      "loss": 1.0179,
+      "step": 791
+    },
+    {
+      "epoch": 0.5962171826479722,
+      "grad_norm": 2.358053180305501,
+      "learning_rate": 7.396106220531398e-06,
+      "loss": 0.9435,
+      "step": 792
+    },
+    {
+      "epoch": 0.5969699821210125,
+      "grad_norm": 3.076010451529784,
+      "learning_rate": 7.372564150352373e-06,
+      "loss": 0.9858,
+      "step": 793
+    },
+    {
+      "epoch": 0.5977227815940529,
+      "grad_norm": 3.20081193030361,
+      "learning_rate": 7.349037711647089e-06,
+      "loss": 1.007,
+      "step": 794
+    },
+    {
+      "epoch": 0.5984755810670932,
+      "grad_norm": 2.6824310401334444,
+      "learning_rate": 7.325527044382004e-06,
+      "loss": 0.9502,
+      "step": 795
+    },
+    {
+      "epoch": 0.5992283805401336,
+      "grad_norm": 2.967747752671785,
+      "learning_rate": 7.3020322884297565e-06,
+      "loss": 0.9533,
+      "step": 796
+    },
+    {
+      "epoch": 0.599981180013174,
+      "grad_norm": 2.3419025363845094,
+      "learning_rate": 7.278553583568326e-06,
+      "loss": 1.0294,
+      "step": 797
+    },
+    {
+      "epoch": 0.6007339794862143,
+      "grad_norm": 2.8919180147573744,
+      "learning_rate": 7.2550910694801905e-06,
+      "loss": 1.0084,
+      "step": 798
+    },
+    {
+      "epoch": 0.6007339794862143,
+      "eval_loss": 0.9782312512397766,
+      "eval_runtime": 585.6058,
+      "eval_samples_per_second": 30.541,
+      "eval_steps_per_second": 0.478,
+      "step": 798
+    },
+    {
+      "epoch": 0.6014867789592547,
+      "grad_norm": 3.0960911393551,
+      "learning_rate": 7.2316448857515076e-06,
+      "loss": 0.9845,
+      "step": 799
+    },
+    {
+      "epoch": 0.6022395784322951,
+      "grad_norm": 2.5561402924270347,
+      "learning_rate": 7.208215171871277e-06,
+      "loss": 0.9745,
+      "step": 800
+    },
+    {
+      "epoch": 0.6029923779053354,
+      "grad_norm": 2.5104780203751322,
+      "learning_rate": 7.184802067230518e-06,
+      "loss": 0.9585,
+      "step": 801
+    },
+    {
+      "epoch": 0.6037451773783759,
+      "grad_norm": 2.412017836012546,
+      "learning_rate": 7.161405711121436e-06,
+      "loss": 0.874,
+      "step": 802
+    },
+    {
+      "epoch": 0.6044979768514162,
+      "grad_norm": 2.5834562133722523,
+      "learning_rate": 7.1380262427365885e-06,
+      "loss": 0.8987,
+      "step": 803
+    },
+    {
+      "epoch": 0.6052507763244566,
+      "grad_norm": 4.380554220513787,
+      "learning_rate": 7.114663801168073e-06,
+      "loss": 1.0493,
+      "step": 804
+    },
+    {
+      "epoch": 0.606003575797497,
+      "grad_norm": 2.7164301877546166,
+      "learning_rate": 7.091318525406671e-06,
+      "loss": 1.0521,
+      "step": 805
+    },
+    {
+      "epoch": 0.6067563752705373,
+      "grad_norm": 2.940119831487498,
+      "learning_rate": 7.067990554341064e-06,
+      "loss": 0.9458,
+      "step": 806
+    },
+    {
+      "epoch": 0.6075091747435777,
+      "grad_norm": 2.7406400760630176,
+      "learning_rate": 7.04468002675696e-06,
+      "loss": 1.0437,
+      "step": 807
+    },
+    {
+      "epoch": 0.6082619742166181,
+      "grad_norm": 2.8146647858748937,
+      "learning_rate": 7.021387081336302e-06,
+      "loss": 0.9932,
+      "step": 808
+    },
+    {
+      "epoch": 0.6090147736896584,
+      "grad_norm": 3.1861410008902857,
+      "learning_rate": 6.998111856656427e-06,
+      "loss": 1.0304,
+      "step": 809
+    },
+    {
+      "epoch": 0.6097675731626988,
+      "grad_norm": 3.3122100772434804,
+      "learning_rate": 6.974854491189243e-06,
+      "loss": 0.9891,
+      "step": 810
+    },
+    {
+      "epoch": 0.6105203726357391,
+      "grad_norm": 2.892082290412832,
+      "learning_rate": 6.951615123300415e-06,
+      "loss": 0.9414,
+      "step": 811
+    },
+    {
+      "epoch": 0.6112731721087795,
+      "grad_norm": 2.4416813602015925,
+      "learning_rate": 6.928393891248529e-06,
+      "loss": 1.004,
+      "step": 812
+    },
+    {
+      "epoch": 0.6120259715818199,
+      "grad_norm": 3.8141614574322893,
+      "learning_rate": 6.905190933184275e-06,
+      "loss": 0.965,
+      "step": 813
+    },
+    {
+      "epoch": 0.6127787710548602,
+      "grad_norm": 2.7579831000313115,
+      "learning_rate": 6.882006387149625e-06,
+      "loss": 1.0356,
+      "step": 814
+    },
+    {
+      "epoch": 0.6135315705279006,
+      "grad_norm": 3.092633092511484,
+      "learning_rate": 6.858840391077017e-06,
+      "loss": 0.9903,
+      "step": 815
+    },
+    {
+      "epoch": 0.614284370000941,
+      "grad_norm": 2.9786953925685755,
+      "learning_rate": 6.8356930827885256e-06,
+      "loss": 1.0204,
+      "step": 816
+    },
+    {
+      "epoch": 0.6150371694739813,
+      "grad_norm": 2.495282333549744,
+      "learning_rate": 6.812564599995042e-06,
+      "loss": 0.9635,
+      "step": 817
+    },
+    {
+      "epoch": 0.6157899689470218,
+      "grad_norm": 2.663824091107566,
+      "learning_rate": 6.789455080295464e-06,
+      "loss": 1.0795,
+      "step": 818
+    },
+    {
+      "epoch": 0.6165427684200621,
+      "grad_norm": 2.8628446717330647,
+      "learning_rate": 6.766364661175872e-06,
+      "loss": 0.9932,
+      "step": 819
+    },
+    {
+      "epoch": 0.6172955678931025,
+      "grad_norm": 2.7547900025423324,
+      "learning_rate": 6.743293480008703e-06,
+      "loss": 0.9901,
+      "step": 820
+    },
+    {
+      "epoch": 0.6180483673661429,
+      "grad_norm": 2.9160986440400074,
+      "learning_rate": 6.720241674051948e-06,
+      "loss": 1.0101,
+      "step": 821
+    },
+    {
+      "epoch": 0.6188011668391832,
+      "grad_norm": 2.7815364890305236,
+      "learning_rate": 6.697209380448333e-06,
+      "loss": 1.0153,
+      "step": 822
+    },
+    {
+      "epoch": 0.6195539663122236,
+      "grad_norm": 2.814498812667104,
+      "learning_rate": 6.674196736224481e-06,
+      "loss": 1.0124,
+      "step": 823
+    },
+    {
+      "epoch": 0.620306765785264,
+      "grad_norm": 3.284851022195675,
+      "learning_rate": 6.651203878290139e-06,
+      "loss": 1.0318,
+      "step": 824
+    },
+    {
+      "epoch": 0.6210595652583043,
+      "grad_norm": 4.043376728595679,
+      "learning_rate": 6.628230943437319e-06,
+      "loss": 1.0221,
+      "step": 825
+    },
+    {
+      "epoch": 0.6218123647313447,
+      "grad_norm": 2.724946690541462,
+      "learning_rate": 6.605278068339516e-06,
+      "loss": 0.9727,
+      "step": 826
+    },
+    {
+      "epoch": 0.622565164204385,
+      "grad_norm": 3.175680771426865,
+      "learning_rate": 6.58234538955087e-06,
+      "loss": 0.9562,
+      "step": 827
+    },
+    {
+      "epoch": 0.6233179636774254,
+      "grad_norm": 3.337757521832563,
+      "learning_rate": 6.559433043505383e-06,
+      "loss": 1.0422,
+      "step": 828
+    },
+    {
+      "epoch": 0.6240707631504658,
+      "grad_norm": 3.0583079706035456,
+      "learning_rate": 6.536541166516079e-06,
+      "loss": 0.9973,
+      "step": 829
+    },
+    {
+      "epoch": 0.6248235626235061,
+      "grad_norm": 3.4140875903459422,
+      "learning_rate": 6.513669894774209e-06,
+      "loss": 1.0313,
+      "step": 830
+    },
+    {
+      "epoch": 0.6255763620965465,
+      "grad_norm": 2.963263472232558,
+      "learning_rate": 6.490819364348434e-06,
+      "loss": 0.9431,
+      "step": 831
+    },
+    {
+      "epoch": 0.626329161569587,
+      "grad_norm": 3.885267790943743,
+      "learning_rate": 6.467989711184021e-06,
+      "loss": 0.9676,
+      "step": 832
+    },
+    {
+      "epoch": 0.6270819610426273,
+      "grad_norm": 2.986705312141074,
+      "learning_rate": 6.445181071102034e-06,
+      "loss": 0.9332,
+      "step": 833
+    },
+    {
+      "epoch": 0.6278347605156677,
+      "grad_norm": 2.4434737482142848,
+      "learning_rate": 6.422393579798519e-06,
+      "loss": 0.9472,
+      "step": 834
+    },
+    {
+      "epoch": 0.628587559988708,
+      "grad_norm": 2.9933676178531203,
+      "learning_rate": 6.399627372843699e-06,
+      "loss": 1.0365,
+      "step": 835
+    },
+    {
+      "epoch": 0.6293403594617484,
+      "grad_norm": 2.423565098566723,
+      "learning_rate": 6.376882585681174e-06,
+      "loss": 0.9378,
+      "step": 836
+    },
+    {
+      "epoch": 0.6300931589347888,
+      "grad_norm": 3.930550350027819,
+      "learning_rate": 6.354159353627114e-06,
+      "loss": 1.0059,
+      "step": 837
+    },
+    {
+      "epoch": 0.6308459584078291,
+      "grad_norm": 3.4001506334649667,
+      "learning_rate": 6.331457811869437e-06,
+      "loss": 0.9484,
+      "step": 838
+    },
+    {
+      "epoch": 0.6315987578808695,
+      "grad_norm": 2.9835541064513698,
+      "learning_rate": 6.3087780954670306e-06,
+      "loss": 0.9999,
+      "step": 839
+    },
+    {
+      "epoch": 0.6323515573539098,
+      "grad_norm": 2.9869824794113113,
+      "learning_rate": 6.286120339348935e-06,
+      "loss": 1.0178,
+      "step": 840
+    },
+    {
+      "epoch": 0.6331043568269502,
+      "grad_norm": 3.336008765122204,
+      "learning_rate": 6.263484678313536e-06,
+      "loss": 1.0082,
+      "step": 841
+    },
+    {
+      "epoch": 0.6338571562999906,
+      "grad_norm": 2.6343658023196146,
+      "learning_rate": 6.240871247027774e-06,
+      "loss": 0.9284,
+      "step": 842
+    },
+    {
+      "epoch": 0.6346099557730309,
+      "grad_norm": 2.8587661519934744,
+      "learning_rate": 6.2182801800263325e-06,
+      "loss": 0.9708,
+      "step": 843
+    },
+    {
+      "epoch": 0.6353627552460713,
+      "grad_norm": 2.68509448532749,
+      "learning_rate": 6.195711611710851e-06,
+      "loss": 0.973,
+      "step": 844
+    },
+    {
+      "epoch": 0.6361155547191117,
+      "grad_norm": 2.436855010449512,
+      "learning_rate": 6.173165676349103e-06,
+      "loss": 0.9015,
+      "step": 845
+    },
+    {
+      "epoch": 0.636868354192152,
+      "grad_norm": 3.2521498003954026,
+      "learning_rate": 6.150642508074225e-06,
+      "loss": 1.0108,
+      "step": 846
+    },
+    {
+      "epoch": 0.6376211536651925,
+      "grad_norm": 3.1006751282783402,
+      "learning_rate": 6.128142240883899e-06,
+      "loss": 0.9579,
+      "step": 847
+    },
+    {
+      "epoch": 0.6383739531382328,
+      "grad_norm": 2.8978862891079054,
+      "learning_rate": 6.105665008639557e-06,
+      "loss": 1.015,
+      "step": 848
+    },
+    {
+      "epoch": 0.6391267526112732,
+      "grad_norm": 2.279326073157125,
+      "learning_rate": 6.083210945065595e-06,
+      "loss": 0.9875,
+      "step": 849
+    },
+    {
+      "epoch": 0.6398795520843136,
+      "grad_norm": 3.0675504012666654,
+      "learning_rate": 6.0607801837485665e-06,
+      "loss": 0.9362,
+      "step": 850
+    },
+    {
+      "epoch": 0.6406323515573539,
+      "grad_norm": 2.4679963317993,
+      "learning_rate": 6.038372858136401e-06,
+      "loss": 1.0029,
+      "step": 851
+    },
+    {
+      "epoch": 0.6413851510303943,
+      "grad_norm": 2.318094472802182,
+      "learning_rate": 6.015989101537586e-06,
+      "loss": 0.9346,
+      "step": 852
+    },
+    {
+      "epoch": 0.6421379505034347,
+      "grad_norm": 3.5138374389839737,
+      "learning_rate": 5.993629047120403e-06,
+      "loss": 1.0261,
+      "step": 853
+    },
+    {
+      "epoch": 0.642890749976475,
+      "grad_norm": 2.854935679385697,
+      "learning_rate": 5.971292827912117e-06,
+      "loss": 0.9759,
+      "step": 854
+    },
+    {
+      "epoch": 0.6436435494495154,
+      "grad_norm": 2.4090683744161834,
+      "learning_rate": 5.9489805767981845e-06,
+      "loss": 0.9546,
+      "step": 855
+    },
+    {
+      "epoch": 0.6443963489225557,
+      "grad_norm": 2.7843495775159233,
+      "learning_rate": 5.926692426521474e-06,
+      "loss": 0.8722,
+      "step": 856
+    },
+    {
+      "epoch": 0.6451491483955961,
+      "grad_norm": 2.740863877578159,
+      "learning_rate": 5.904428509681473e-06,
+      "loss": 0.9348,
+      "step": 857
+    },
+    {
+      "epoch": 0.6459019478686365,
+      "grad_norm": 3.0461938473498833,
+      "learning_rate": 5.882188958733488e-06,
+      "loss": 0.9922,
+      "step": 858
+    },
+    {
+      "epoch": 0.6466547473416768,
+      "grad_norm": 3.0394454770711294,
+      "learning_rate": 5.859973905987866e-06,
+      "loss": 0.9624,
+      "step": 859
+    },
+    {
+      "epoch": 0.6474075468147172,
+      "grad_norm": 2.723188343497309,
+      "learning_rate": 5.837783483609214e-06,
+      "loss": 1.0125,
+      "step": 860
+    },
+    {
+      "epoch": 0.6481603462877576,
+      "grad_norm": 2.3691610151759432,
+      "learning_rate": 5.815617823615599e-06,
+      "loss": 0.9457,
+      "step": 861
+    },
+    {
+      "epoch": 0.648913145760798,
+      "grad_norm": 2.667516135651494,
+      "learning_rate": 5.79347705787777e-06,
+      "loss": 1.0045,
+      "step": 862
+    },
+    {
+      "epoch": 0.6496659452338384,
+      "grad_norm": 3.301425067030213,
+      "learning_rate": 5.771361318118374e-06,
+      "loss": 1.0249,
+      "step": 863
+    },
+    {
+      "epoch": 0.6504187447068787,
+      "grad_norm": 3.075114609965823,
+      "learning_rate": 5.749270735911159e-06,
+      "loss": 1.0088,
+      "step": 864
+    },
+    {
+      "epoch": 0.6511715441799191,
+      "grad_norm": 2.349924278634797,
+      "learning_rate": 5.727205442680218e-06,
+      "loss": 0.9746,
+      "step": 865
+    },
+    {
+      "epoch": 0.6519243436529595,
+      "grad_norm": 2.9453643370218003,
+      "learning_rate": 5.7051655696991825e-06,
+      "loss": 0.9492,
+      "step": 866
+    },
+    {
+      "epoch": 0.6526771431259998,
+      "grad_norm": 3.003085836693138,
+      "learning_rate": 5.683151248090455e-06,
+      "loss": 1.0076,
+      "step": 867
+    },
+    {
+      "epoch": 0.6534299425990402,
+      "grad_norm": 4.541919041618707,
+      "learning_rate": 5.66116260882442e-06,
+      "loss": 0.9974,
+      "step": 868
+    },
+    {
+      "epoch": 0.6541827420720806,
+      "grad_norm": 3.3380757605339038,
+      "learning_rate": 5.639199782718674e-06,
+      "loss": 1.0176,
+      "step": 869
+    },
+    {
+      "epoch": 0.6549355415451209,
+      "grad_norm": 7.9191564125396345,
+      "learning_rate": 5.617262900437239e-06,
+      "loss": 1.0649,
+      "step": 870
+    },
+    {
+      "epoch": 0.6556883410181613,
+      "grad_norm": 3.1847304138480963,
+      "learning_rate": 5.595352092489791e-06,
+      "loss": 1.0244,
+      "step": 871
+    },
+    {
+      "epoch": 0.6564411404912016,
+      "grad_norm": 2.8005119319971072,
+      "learning_rate": 5.573467489230879e-06,
+      "loss": 0.9347,
+      "step": 872
+    },
+    {
+      "epoch": 0.657193939964242,
+      "grad_norm": 3.7323965387138887,
+      "learning_rate": 5.55160922085916e-06,
+      "loss": 1.0172,
+      "step": 873
+    },
+    {
+      "epoch": 0.6579467394372824,
+      "grad_norm": 3.100690883893798,
+      "learning_rate": 5.529777417416599e-06,
+      "loss": 1.0025,
+      "step": 874
+    },
+    {
+      "epoch": 0.6586995389103227,
+      "grad_norm": 3.061825155807451,
+      "learning_rate": 5.507972208787728e-06,
+      "loss": 0.9475,
+      "step": 875
+    },
+    {
+      "epoch": 0.6594523383833631,
+      "grad_norm": 3.0772915102988456,
+      "learning_rate": 5.486193724698854e-06,
+      "loss": 0.899,
+      "step": 876
+    },
+    {
+      "epoch": 0.6602051378564034,
+      "grad_norm": 2.4561283709078596,
+      "learning_rate": 5.464442094717281e-06,
+      "loss": 0.9297,
+      "step": 877
+    },
+    {
+      "epoch": 0.6609579373294439,
+      "grad_norm": 3.2450122492959106,
+      "learning_rate": 5.442717448250574e-06,
+      "loss": 1.0168,
+      "step": 878
+    },
+    {
+      "epoch": 0.6617107368024843,
+      "grad_norm": 2.6488414083706213,
+      "learning_rate": 5.421019914545735e-06,
+      "loss": 0.9117,
+      "step": 879
+    },
+    {
+      "epoch": 0.6624635362755246,
+      "grad_norm": 2.3123514940320677,
+      "learning_rate": 5.399349622688479e-06,
+      "loss": 0.985,
+      "step": 880
+    },
+    {
+      "epoch": 0.663216335748565,
+      "grad_norm": 3.4193564013787157,
+      "learning_rate": 5.3777067016024495e-06,
+      "loss": 1.0232,
+      "step": 881
+    },
+    {
+      "epoch": 0.6639691352216054,
+      "grad_norm": 2.6262517727005013,
+      "learning_rate": 5.35609128004845e-06,
+      "loss": 0.9809,
+      "step": 882
+    },
+    {
+      "epoch": 0.6647219346946457,
+      "grad_norm": 3.544497471768593,
+      "learning_rate": 5.33450348662368e-06,
+      "loss": 1.0073,
+      "step": 883
+    },
+    {
+      "epoch": 0.6654747341676861,
+      "grad_norm": 2.208983215094343,
+      "learning_rate": 5.312943449760975e-06,
+      "loss": 0.9091,
+      "step": 884
+    },
+    {
+      "epoch": 0.6662275336407264,
+      "grad_norm": 2.655612493732194,
+      "learning_rate": 5.291411297728027e-06,
+      "loss": 0.9758,
+      "step": 885
+    },
+    {
+      "epoch": 0.6669803331137668,
+      "grad_norm": 3.0194835999472405,
+      "learning_rate": 5.269907158626639e-06,
+      "loss": 1.0229,
+      "step": 886
+    },
+    {
+      "epoch": 0.6677331325868072,
+      "grad_norm": 3.113229923263648,
+      "learning_rate": 5.248431160391963e-06,
+      "loss": 0.9857,
+      "step": 887
+    },
+    {
+      "epoch": 0.6684859320598475,
+      "grad_norm": 2.870839406559322,
+      "learning_rate": 5.226983430791722e-06,
+      "loss": 1.027,
+      "step": 888
+    },
+    {
+      "epoch": 0.6692387315328879,
+      "grad_norm": 2.652991785215087,
+      "learning_rate": 5.205564097425458e-06,
+      "loss": 1.0588,
+      "step": 889
+    },
+    {
+      "epoch": 0.6699915310059283,
+      "grad_norm": 2.895387116149638,
+      "learning_rate": 5.184173287723782e-06,
+      "loss": 0.9812,
+      "step": 890
+    },
+    {
+      "epoch": 0.6707443304789686,
+      "grad_norm": 2.764343371936672,
+      "learning_rate": 5.1628111289476025e-06,
+      "loss": 1.0278,
+      "step": 891
+    },
+    {
+      "epoch": 0.671497129952009,
+      "grad_norm": 3.649390125582586,
+      "learning_rate": 5.14147774818738e-06,
+      "loss": 1.0463,
+      "step": 892
+    },
+    {
+      "epoch": 0.6722499294250494,
+      "grad_norm": 2.6370061008680907,
+      "learning_rate": 5.120173272362361e-06,
+      "loss": 1.0512,
+      "step": 893
+    },
+    {
+      "epoch": 0.6730027288980898,
+      "grad_norm": 3.2489475334028652,
+      "learning_rate": 5.098897828219831e-06,
+      "loss": 0.9963,
+      "step": 894
+    },
+    {
+      "epoch": 0.6737555283711302,
+      "grad_norm": 2.86598289535442,
+      "learning_rate": 5.0776515423343445e-06,
+      "loss": 0.9304,
+      "step": 895
+    },
+    {
+      "epoch": 0.6745083278441705,
+      "grad_norm": 2.963772691083135,
+      "learning_rate": 5.0564345411070025e-06,
+      "loss": 1.0188,
+      "step": 896
+    },
+    {
+      "epoch": 0.6752611273172109,
+      "grad_norm": 2.5796584317987916,
+      "learning_rate": 5.03524695076467e-06,
+      "loss": 0.9741,
+      "step": 897
+    },
+    {
+      "epoch": 0.6760139267902513,
+      "grad_norm": 2.51416923212948,
+      "learning_rate": 5.014088897359242e-06,
+      "loss": 1.0096,
+      "step": 898
+    },
+    {
+      "epoch": 0.6767667262632916,
+      "grad_norm": 2.6665940615567187,
+      "learning_rate": 4.992960506766881e-06,
+      "loss": 0.9346,
+      "step": 899
+    },
+    {
+      "epoch": 0.677519525736332,
+      "grad_norm": 3.065335474625922,
+      "learning_rate": 4.971861904687283e-06,
+      "loss": 1.0338,
+      "step": 900
+    },
+    {
+      "epoch": 0.6782723252093723,
+      "grad_norm": 2.9639586464498913,
+      "learning_rate": 4.950793216642923e-06,
+      "loss": 0.9781,
+      "step": 901
+    },
+    {
+      "epoch": 0.6790251246824127,
+      "grad_norm": 3.040680837970922,
+      "learning_rate": 4.929754567978303e-06,
+      "loss": 1.0059,
+      "step": 902
+    },
+    {
+      "epoch": 0.6797779241554531,
+      "grad_norm": 2.6333146143761716,
+      "learning_rate": 4.908746083859214e-06,
+      "loss": 0.985,
+      "step": 903
+    },
+    {
+      "epoch": 0.6805307236284934,
+      "grad_norm": 3.292933520792098,
+      "learning_rate": 4.887767889271987e-06,
+      "loss": 1.0661,
+      "step": 904
+    },
+    {
+      "epoch": 0.6812835231015338,
+      "grad_norm": 2.652137597319615,
+      "learning_rate": 4.866820109022752e-06,
+      "loss": 1.0072,
+      "step": 905
+    },
+    {
+      "epoch": 0.6820363225745742,
+      "grad_norm": 3.1886078375655136,
+      "learning_rate": 4.845902867736692e-06,
+      "loss": 1.0616,
+      "step": 906
+    },
+    {
+      "epoch": 0.6827891220476145,
+      "grad_norm": 2.6580714465346036,
+      "learning_rate": 4.8250162898573046e-06,
+      "loss": 0.9878,
+      "step": 907
+    },
+    {
+      "epoch": 0.683541921520655,
+      "grad_norm": 3.5025560301930834,
+      "learning_rate": 4.804160499645667e-06,
+      "loss": 1.0037,
+      "step": 908
+    },
+    {
+      "epoch": 0.6842947209936953,
+      "grad_norm": 3.0446356739651215,
+      "learning_rate": 4.783335621179675e-06,
+      "loss": 0.9986,
+      "step": 909
+    },
+    {
+      "epoch": 0.6850475204667357,
+      "grad_norm": 2.5280116263957524,
+      "learning_rate": 4.762541778353337e-06,
+      "loss": 0.9486,
+      "step": 910
+    },
+    {
+      "epoch": 0.6858003199397761,
+      "grad_norm": 3.185618921006941,
+      "learning_rate": 4.741779094876009e-06,
+      "loss": 1.0347,
+      "step": 911
+    },
+    {
+      "epoch": 0.6865531194128164,
+      "grad_norm": 3.2600894992341525,
+      "learning_rate": 4.721047694271676e-06,
+      "loss": 1.1032,
+      "step": 912
+    },
+    {
+      "epoch": 0.6873059188858568,
+      "grad_norm": 2.661744895585514,
+      "learning_rate": 4.700347699878211e-06,
+      "loss": 0.9778,
+      "step": 913
+    },
+    {
+      "epoch": 0.6880587183588972,
+      "grad_norm": 2.806546351632779,
+      "learning_rate": 4.679679234846636e-06,
+      "loss": 0.9729,
+      "step": 914
+    },
+    {
+      "epoch": 0.6888115178319375,
+      "grad_norm": 3.2198890803288136,
+      "learning_rate": 4.659042422140399e-06,
+      "loss": 1.0176,
+      "step": 915
+    },
+    {
+      "epoch": 0.6895643173049779,
+      "grad_norm": 3.1995454384698245,
+      "learning_rate": 4.6384373845346375e-06,
+      "loss": 0.9504,
+      "step": 916
+    },
+    {
+      "epoch": 0.6903171167780182,
+      "grad_norm": 2.206607456533358,
+      "learning_rate": 4.617864244615448e-06,
+      "loss": 0.9667,
+      "step": 917
+    },
+    {
+      "epoch": 0.6910699162510586,
+      "grad_norm": 2.7105101530672986,
+      "learning_rate": 4.597323124779155e-06,
+      "loss": 0.9501,
+      "step": 918
+    },
+    {
+      "epoch": 0.691822715724099,
+      "grad_norm": 2.486726858098457,
+      "learning_rate": 4.576814147231594e-06,
+      "loss": 0.9677,
+      "step": 919
+    },
+    {
+      "epoch": 0.6925755151971393,
+      "grad_norm": 3.5397984883778983,
+      "learning_rate": 4.556337433987359e-06,
+      "loss": 1.0204,
+      "step": 920
+    },
+    {
+      "epoch": 0.6933283146701797,
+      "grad_norm": 3.0384236542438336,
+      "learning_rate": 4.535893106869107e-06,
+      "loss": 1.1224,
+      "step": 921
+    },
+    {
+      "epoch": 0.69408111414322,
+      "grad_norm": 2.8509939761812233,
+      "learning_rate": 4.515481287506811e-06,
+      "loss": 0.9704,
+      "step": 922
+    },
+    {
+      "epoch": 0.6948339136162605,
+      "grad_norm": 2.612764095608806,
+      "learning_rate": 4.495102097337062e-06,
+      "loss": 0.9182,
+      "step": 923
+    },
+    {
+      "epoch": 0.6955867130893009,
+      "grad_norm": 2.991760219708861,
+      "learning_rate": 4.474755657602303e-06,
+      "loss": 0.9583,
+      "step": 924
+    },
+    {
+      "epoch": 0.6963395125623412,
+      "grad_norm": 2.680080595323768,
+      "learning_rate": 4.454442089350151e-06,
+      "loss": 1.0236,
+      "step": 925
+    },
+    {
+      "epoch": 0.6970923120353816,
+      "grad_norm": 2.4653672558221067,
+      "learning_rate": 4.434161513432659e-06,
+      "loss": 0.8888,
+      "step": 926
+    },
+    {
+      "epoch": 0.697845111508422,
+      "grad_norm": 3.4498774018544447,
+      "learning_rate": 4.413914050505591e-06,
+      "loss": 0.9749,
+      "step": 927
+    },
+    {
+      "epoch": 0.6985979109814623,
+      "grad_norm": 2.9269586402584724,
+      "learning_rate": 4.393699821027716e-06,
+      "loss": 0.9258,
+      "step": 928
+    },
+    {
+      "epoch": 0.6993507104545027,
+      "grad_norm": 3.2070405609265764,
+      "learning_rate": 4.37351894526009e-06,
+      "loss": 1.0338,
+      "step": 929
+    },
+    {
+      "epoch": 0.700103509927543,
+      "grad_norm": 2.8651840370763333,
+      "learning_rate": 4.35337154326532e-06,
+      "loss": 0.8729,
+      "step": 930
+    },
+    {
+      "epoch": 0.7008563094005834,
+      "grad_norm": 2.9774465542399424,
+      "learning_rate": 4.333257734906889e-06,
+      "loss": 0.9809,
+      "step": 931
+    },
+    {
+      "epoch": 0.7008563094005834,
+      "eval_loss": 0.9652193188667297,
+      "eval_runtime": 585.081,
+      "eval_samples_per_second": 30.568,
+      "eval_steps_per_second": 0.479,
+      "step": 931
+    },
+    {
+      "epoch": 0.7016091088736238,
+      "grad_norm": 3.7083258629749887,
+      "learning_rate": 4.313177639848408e-06,
+      "loss": 1.044,
+      "step": 932
+    },
+    {
+      "epoch": 0.7023619083466641,
+      "grad_norm": 3.682326422466505,
+      "learning_rate": 4.293131377552923e-06,
+      "loss": 0.9859,
+      "step": 933
+    },
+    {
+      "epoch": 0.7031147078197045,
+      "grad_norm": 2.6777610705109494,
+      "learning_rate": 4.273119067282184e-06,
+      "loss": 0.9582,
+      "step": 934
+    },
+    {
+      "epoch": 0.7038675072927449,
+      "grad_norm": 2.473756981986894,
+      "learning_rate": 4.253140828095964e-06,
+      "loss": 0.9545,
+      "step": 935
+    },
+    {
+      "epoch": 0.7046203067657852,
+      "grad_norm": 2.9701523969106978,
+      "learning_rate": 4.2331967788513295e-06,
+      "loss": 0.9585,
+      "step": 936
+    },
+    {
+      "epoch": 0.7053731062388257,
+      "grad_norm": 2.7968419484911715,
+      "learning_rate": 4.213287038201943e-06,
+      "loss": 0.9302,
+      "step": 937
+    },
+    {
+      "epoch": 0.706125905711866,
+      "grad_norm": 3.068880113580643,
+      "learning_rate": 4.193411724597352e-06,
+      "loss": 1.0344,
+      "step": 938
+    },
+    {
+      "epoch": 0.7068787051849064,
+      "grad_norm": 2.6284431085135656,
+      "learning_rate": 4.173570956282286e-06,
+      "loss": 0.9452,
+      "step": 939
+    },
+    {
+      "epoch": 0.7076315046579468,
+      "grad_norm": 4.4448629730253915,
+      "learning_rate": 4.153764851295954e-06,
+      "loss": 0.9522,
+      "step": 940
+    },
+    {
+      "epoch": 0.7083843041309871,
+      "grad_norm": 3.262892448671552,
+      "learning_rate": 4.1339935274713404e-06,
+      "loss": 0.9437,
+      "step": 941
+    },
+    {
+      "epoch": 0.7091371036040275,
+      "grad_norm": 2.523698427815705,
+      "learning_rate": 4.114257102434508e-06,
+      "loss": 0.9616,
+      "step": 942
+    },
+    {
+      "epoch": 0.7098899030770679,
+      "grad_norm": 3.44474459550001,
+      "learning_rate": 4.094555693603891e-06,
+      "loss": 1.0122,
+      "step": 943
+    },
+    {
+      "epoch": 0.7106427025501082,
+      "grad_norm": 2.950294841102305,
+      "learning_rate": 4.074889418189608e-06,
+      "loss": 0.9568,
+      "step": 944
+    },
+    {
+      "epoch": 0.7113955020231486,
+      "grad_norm": 4.539750688224925,
+      "learning_rate": 4.055258393192746e-06,
+      "loss": 0.9975,
+      "step": 945
+    },
+    {
+      "epoch": 0.7121483014961889,
+      "grad_norm": 3.2693568806029996,
+      "learning_rate": 4.0356627354046854e-06,
+      "loss": 1.0896,
+      "step": 946
+    },
+    {
+      "epoch": 0.7129011009692293,
+      "grad_norm": 2.8301942529564275,
+      "learning_rate": 4.016102561406392e-06,
+      "loss": 0.9751,
+      "step": 947
+    },
+    {
+      "epoch": 0.7136539004422697,
+      "grad_norm": 3.279627709424991,
+      "learning_rate": 3.996577987567727e-06,
+      "loss": 0.914,
+      "step": 948
+    },
+    {
+      "epoch": 0.71440669991531,
+      "grad_norm": 2.832349441872171,
+      "learning_rate": 3.977089130046756e-06,
+      "loss": 0.9785,
+      "step": 949
+    },
+    {
+      "epoch": 0.7151594993883504,
+      "grad_norm": 3.1766100419605885,
+      "learning_rate": 3.957636104789056e-06,
+      "loss": 1.0199,
+      "step": 950
+    },
+    {
+      "epoch": 0.7159122988613909,
+      "grad_norm": 3.1393936203952255,
+      "learning_rate": 3.938219027527023e-06,
+      "loss": 1.0026,
+      "step": 951
+    },
+    {
+      "epoch": 0.7166650983344311,
+      "grad_norm": 2.4038458329860415,
+      "learning_rate": 3.9188380137791934e-06,
+      "loss": 0.9613,
+      "step": 952
+    },
+    {
+      "epoch": 0.7174178978074716,
+      "grad_norm": 2.7702463519768736,
+      "learning_rate": 3.899493178849544e-06,
+      "loss": 0.9076,
+      "step": 953
+    },
+    {
+      "epoch": 0.7181706972805119,
+      "grad_norm": 2.831283963811367,
+      "learning_rate": 3.880184637826816e-06,
+      "loss": 0.9818,
+      "step": 954
+    },
+    {
+      "epoch": 0.7189234967535523,
+      "grad_norm": 2.483000422386397,
+      "learning_rate": 3.860912505583819e-06,
+      "loss": 0.8982,
+      "step": 955
+    },
+    {
+      "epoch": 0.7196762962265927,
+      "grad_norm": 2.699549276031205,
+      "learning_rate": 3.841676896776764e-06,
+      "loss": 1.0386,
+      "step": 956
+    },
+    {
+      "epoch": 0.720429095699633,
+      "grad_norm": 3.118505420829208,
+      "learning_rate": 3.822477925844564e-06,
+      "loss": 0.956,
+      "step": 957
+    },
+    {
+      "epoch": 0.7211818951726734,
+      "grad_norm": 3.4585379790093245,
+      "learning_rate": 3.803315707008176e-06,
+      "loss": 0.9622,
+      "step": 958
+    },
+    {
+      "epoch": 0.7219346946457137,
+      "grad_norm": 2.8026653218513333,
+      "learning_rate": 3.7841903542698855e-06,
+      "loss": 1.0212,
+      "step": 959
+    },
+    {
+      "epoch": 0.7226874941187541,
+      "grad_norm": 3.049945822314557,
+      "learning_rate": 3.7651019814126656e-06,
+      "loss": 0.9308,
+      "step": 960
+    },
+    {
+      "epoch": 0.7234402935917945,
+      "grad_norm": 3.878397734938858,
+      "learning_rate": 3.7460507019994775e-06,
+      "loss": 1.0046,
+      "step": 961
+    },
+    {
+      "epoch": 0.7241930930648348,
+      "grad_norm": 2.9601126857355506,
+      "learning_rate": 3.7270366293726033e-06,
+      "loss": 0.955,
+      "step": 962
+    },
+    {
+      "epoch": 0.7249458925378752,
+      "grad_norm": 2.844017367671424,
+      "learning_rate": 3.7080598766529686e-06,
+      "loss": 0.9088,
+      "step": 963
+    },
+    {
+      "epoch": 0.7256986920109156,
+      "grad_norm": 4.1384896382910625,
+      "learning_rate": 3.689120556739475e-06,
+      "loss": 1.0619,
+      "step": 964
+    },
+    {
+      "epoch": 0.7264514914839559,
+      "grad_norm": 3.8764277193054784,
+      "learning_rate": 3.6702187823083147e-06,
+      "loss": 0.9416,
+      "step": 965
+    },
+    {
+      "epoch": 0.7272042909569963,
+      "grad_norm": 2.9803869936914675,
+      "learning_rate": 3.651354665812313e-06,
+      "loss": 1.0522,
+      "step": 966
+    },
+    {
+      "epoch": 0.7279570904300366,
+      "grad_norm": 3.3176825382488926,
+      "learning_rate": 3.6325283194802675e-06,
+      "loss": 0.9702,
+      "step": 967
+    },
+    {
+      "epoch": 0.7287098899030771,
+      "grad_norm": 2.6696911093544173,
+      "learning_rate": 3.613739855316257e-06,
+      "loss": 0.8698,
+      "step": 968
+    },
+    {
+      "epoch": 0.7294626893761175,
+      "grad_norm": 3.1150460867558003,
+      "learning_rate": 3.594989385098985e-06,
+      "loss": 0.9889,
+      "step": 969
+    },
+    {
+      "epoch": 0.7302154888491578,
+      "grad_norm": 2.935099853350304,
+      "learning_rate": 3.5762770203811225e-06,
+      "loss": 0.9321,
+      "step": 970
+    },
+    {
+      "epoch": 0.7309682883221982,
+      "grad_norm": 2.44072267113799,
+      "learning_rate": 3.557602872488638e-06,
+      "loss": 0.9667,
+      "step": 971
+    },
+    {
+      "epoch": 0.7317210877952386,
+      "grad_norm": 3.0638151270341902,
+      "learning_rate": 3.5389670525201335e-06,
+      "loss": 0.9692,
+      "step": 972
+    },
+    {
+      "epoch": 0.7324738872682789,
+      "grad_norm": 3.5159680803026143,
+      "learning_rate": 3.5203696713461866e-06,
+      "loss": 1.0826,
+      "step": 973
+    },
+    {
+      "epoch": 0.7332266867413193,
+      "grad_norm": 2.476026867492786,
+      "learning_rate": 3.5018108396086945e-06,
+      "loss": 1.0051,
+      "step": 974
+    },
+    {
+      "epoch": 0.7339794862143596,
+      "grad_norm": 2.376830294565953,
+      "learning_rate": 3.483290667720196e-06,
+      "loss": 1.0273,
+      "step": 975
+    },
+    {
+      "epoch": 0.7347322856874,
+      "grad_norm": 2.2455010627110514,
+      "learning_rate": 3.4648092658632506e-06,
+      "loss": 0.9793,
+      "step": 976
+    },
+    {
+      "epoch": 0.7354850851604404,
+      "grad_norm": 2.6970914455002535,
+      "learning_rate": 3.4463667439897486e-06,
+      "loss": 0.9663,
+      "step": 977
+    },
+    {
+      "epoch": 0.7362378846334807,
+      "grad_norm": 3.3981774487654017,
+      "learning_rate": 3.4279632118202744e-06,
+      "loss": 0.9573,
+      "step": 978
+    },
+    {
+      "epoch": 0.7369906841065211,
+      "grad_norm": 2.5187307744033776,
+      "learning_rate": 3.4095987788434538e-06,
+      "loss": 0.9323,
+      "step": 979
+    },
+    {
+      "epoch": 0.7377434835795615,
+      "grad_norm": 2.6118138428418853,
+      "learning_rate": 3.3912735543152864e-06,
+      "loss": 0.9362,
+      "step": 980
+    },
+    {
+      "epoch": 0.7384962830526018,
+      "grad_norm": 2.751134189690385,
+      "learning_rate": 3.372987647258521e-06,
+      "loss": 0.9826,
+      "step": 981
+    },
+    {
+      "epoch": 0.7392490825256423,
+      "grad_norm": 2.5860913540515322,
+      "learning_rate": 3.354741166461989e-06,
+      "loss": 0.9628,
+      "step": 982
+    },
+    {
+      "epoch": 0.7400018819986826,
+      "grad_norm": 2.9459531689153193,
+      "learning_rate": 3.3365342204799613e-06,
+      "loss": 0.9987,
+      "step": 983
+    },
+    {
+      "epoch": 0.740754681471723,
+      "grad_norm": 3.1706073449088676,
+      "learning_rate": 3.3183669176315046e-06,
+      "loss": 0.9705,
+      "step": 984
+    },
+    {
+      "epoch": 0.7415074809447634,
+      "grad_norm": 3.400712366138258,
+      "learning_rate": 3.3002393659998357e-06,
+      "loss": 1.0182,
+      "step": 985
+    },
+    {
+      "epoch": 0.7422602804178037,
+      "grad_norm": 2.162383584339622,
+      "learning_rate": 3.2821516734316772e-06,
+      "loss": 0.9337,
+      "step": 986
+    },
+    {
+      "epoch": 0.7430130798908441,
+      "grad_norm": 2.252372753973197,
+      "learning_rate": 3.264103947536619e-06,
+      "loss": 0.9497,
+      "step": 987
+    },
+    {
+      "epoch": 0.7437658793638845,
+      "grad_norm": 3.400700187647833,
+      "learning_rate": 3.2460962956864727e-06,
+      "loss": 0.9949,
+      "step": 988
+    },
+    {
+      "epoch": 0.7445186788369248,
+      "grad_norm": 3.1714265140894504,
+      "learning_rate": 3.2281288250146447e-06,
+      "loss": 1.0138,
+      "step": 989
+    },
+    {
+      "epoch": 0.7452714783099652,
+      "grad_norm": 2.4482936996207814,
+      "learning_rate": 3.210201642415477e-06,
+      "loss": 0.9914,
+      "step": 990
+    },
+    {
+      "epoch": 0.7460242777830055,
+      "grad_norm": 2.524241576825987,
+      "learning_rate": 3.1923148545436357e-06,
+      "loss": 0.8905,
+      "step": 991
+    },
+    {
+      "epoch": 0.7467770772560459,
+      "grad_norm": 3.2756497769927715,
+      "learning_rate": 3.174468567813461e-06,
+      "loss": 0.9682,
+      "step": 992
+    },
+    {
+      "epoch": 0.7475298767290863,
+      "grad_norm": 2.991414090084408,
+      "learning_rate": 3.1566628883983395e-06,
+      "loss": 0.983,
+      "step": 993
+    },
+    {
+      "epoch": 0.7482826762021266,
+      "grad_norm": 2.8418499968354376,
+      "learning_rate": 3.138897922230074e-06,
+      "loss": 0.9612,
+      "step": 994
+    },
+    {
+      "epoch": 0.749035475675167,
+      "grad_norm": 3.5608060404370234,
+      "learning_rate": 3.121173774998245e-06,
+      "loss": 1.0149,
+      "step": 995
+    },
+    {
+      "epoch": 0.7497882751482075,
+      "grad_norm": 3.1105127346719152,
+      "learning_rate": 3.103490552149595e-06,
+      "loss": 0.9507,
+      "step": 996
+    },
+    {
+      "epoch": 0.7505410746212478,
+      "grad_norm": 2.8041971684545537,
+      "learning_rate": 3.0858483588873878e-06,
+      "loss": 0.9625,
+      "step": 997
+    },
+    {
+      "epoch": 0.7512938740942882,
+      "grad_norm": 2.5647827030571397,
+      "learning_rate": 3.0682473001707925e-06,
+      "loss": 0.8918,
+      "step": 998
+    },
+    {
+      "epoch": 0.7520466735673285,
+      "grad_norm": 3.6591031416487105,
+      "learning_rate": 3.050687480714256e-06,
+      "loss": 0.998,
+      "step": 999
+    },
+    {
+      "epoch": 0.7527994730403689,
+      "grad_norm": 2.9910839287028064,
+      "learning_rate": 3.0331690049868733e-06,
+      "loss": 0.9591,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7535522725134093,
+      "grad_norm": 2.5843459545257903,
+      "learning_rate": 3.0156919772117788e-06,
+      "loss": 0.97,
+      "step": 1001
+    },
+    {
+      "epoch": 0.7543050719864496,
+      "grad_norm": 2.625453019920782,
+      "learning_rate": 2.998256501365514e-06,
+      "loss": 0.9809,
+      "step": 1002
+    },
+    {
+      "epoch": 0.75505787145949,
+      "grad_norm": 2.710505597656893,
+      "learning_rate": 2.9808626811774222e-06,
+      "loss": 0.9726,
+      "step": 1003
+    },
+    {
+      "epoch": 0.7558106709325303,
+      "grad_norm": 2.611841535968574,
+      "learning_rate": 2.963510620129021e-06,
+      "loss": 0.9266,
+      "step": 1004
+    },
+    {
+      "epoch": 0.7565634704055707,
+      "grad_norm": 3.2583815916558203,
+      "learning_rate": 2.9462004214533803e-06,
+      "loss": 0.9847,
+      "step": 1005
+    },
+    {
+      "epoch": 0.7573162698786111,
+      "grad_norm": 2.532735200287804,
+      "learning_rate": 2.9289321881345257e-06,
+      "loss": 1.0084,
+      "step": 1006
+    },
+    {
+      "epoch": 0.7580690693516514,
+      "grad_norm": 2.857674586577938,
+      "learning_rate": 2.911706022906816e-06,
+      "loss": 1.0016,
+      "step": 1007
+    },
+    {
+      "epoch": 0.7588218688246918,
+      "grad_norm": 2.533187369034248,
+      "learning_rate": 2.894522028254334e-06,
+      "loss": 0.94,
+      "step": 1008
+    },
+    {
+      "epoch": 0.7595746682977322,
+      "grad_norm": 2.594116896625209,
+      "learning_rate": 2.8773803064102758e-06,
+      "loss": 0.8854,
+      "step": 1009
+    },
+    {
+      "epoch": 0.7603274677707725,
+      "grad_norm": 3.120940328980785,
+      "learning_rate": 2.860280959356336e-06,
+      "loss": 0.9751,
+      "step": 1010
+    },
+    {
+      "epoch": 0.761080267243813,
+      "grad_norm": 2.4297166404754544,
+      "learning_rate": 2.843224088822113e-06,
+      "loss": 0.9551,
+      "step": 1011
+    },
+    {
+      "epoch": 0.7618330667168532,
+      "grad_norm": 2.7126291505696583,
+      "learning_rate": 2.8262097962845058e-06,
+      "loss": 1.009,
+      "step": 1012
+    },
+    {
+      "epoch": 0.7625858661898937,
+      "grad_norm": 3.1508191398932888,
+      "learning_rate": 2.809238182967092e-06,
+      "loss": 0.9555,
+      "step": 1013
+    },
+    {
+      "epoch": 0.7633386656629341,
+      "grad_norm": 2.540908640400342,
+      "learning_rate": 2.7923093498395438e-06,
+      "loss": 0.934,
+      "step": 1014
+    },
+    {
+      "epoch": 0.7640914651359744,
+      "grad_norm": 2.524639624628765,
+      "learning_rate": 2.77542339761701e-06,
+      "loss": 1.022,
+      "step": 1015
+    },
+    {
+      "epoch": 0.7648442646090148,
+      "grad_norm": 2.6011323382599048,
+      "learning_rate": 2.7585804267595383e-06,
+      "loss": 0.9177,
+      "step": 1016
+    },
+    {
+      "epoch": 0.7655970640820552,
+      "grad_norm": 2.4744466687181235,
+      "learning_rate": 2.74178053747146e-06,
+      "loss": 0.9263,
+      "step": 1017
+    },
+    {
+      "epoch": 0.7663498635550955,
+      "grad_norm": 3.1963960205551545,
+      "learning_rate": 2.7250238297008026e-06,
+      "loss": 1.0093,
+      "step": 1018
+    },
+    {
+      "epoch": 0.7671026630281359,
+      "grad_norm": 3.240522495596184,
+      "learning_rate": 2.708310403138692e-06,
+      "loss": 1.0007,
+      "step": 1019
+    },
+    {
+      "epoch": 0.7678554625011762,
+      "grad_norm": 3.056025997152738,
+      "learning_rate": 2.691640357218759e-06,
+      "loss": 0.9285,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7686082619742166,
+      "grad_norm": 3.010243843142797,
+      "learning_rate": 2.675013791116551e-06,
+      "loss": 0.9169,
+      "step": 1021
+    },
+    {
+      "epoch": 0.769361061447257,
+      "grad_norm": 2.9212925265086818,
+      "learning_rate": 2.658430803748936e-06,
+      "loss": 0.9661,
+      "step": 1022
+    },
+    {
+      "epoch": 0.7701138609202973,
+      "grad_norm": 2.6746112820249532,
+      "learning_rate": 2.6418914937735228e-06,
+      "loss": 0.9974,
+      "step": 1023
+    },
+    {
+      "epoch": 0.7708666603933377,
+      "grad_norm": 2.8886622989206763,
+      "learning_rate": 2.625395959588067e-06,
+      "loss": 1.0527,
+      "step": 1024
+    },
+    {
+      "epoch": 0.7716194598663781,
+      "grad_norm": 3.198567662487964,
+      "learning_rate": 2.6089442993298854e-06,
+      "loss": 1.0188,
+      "step": 1025
+    },
+    {
+      "epoch": 0.7723722593394184,
+      "grad_norm": 3.01161211798399,
+      "learning_rate": 2.592536610875275e-06,
+      "loss": 0.9717,
+      "step": 1026
+    },
+    {
+      "epoch": 0.7731250588124589,
+      "grad_norm": 3.2539981490647474,
+      "learning_rate": 2.576172991838933e-06,
+      "loss": 0.9739,
+      "step": 1027
+    },
+    {
+      "epoch": 0.7738778582854992,
+      "grad_norm": 2.1667397313948373,
+      "learning_rate": 2.5598535395733735e-06,
+      "loss": 0.9793,
+      "step": 1028
+    },
+    {
+      "epoch": 0.7746306577585396,
+      "grad_norm": 2.5896902701922144,
+      "learning_rate": 2.5435783511683444e-06,
+      "loss": 0.9285,
+      "step": 1029
+    },
+    {
+      "epoch": 0.77538345723158,
+      "grad_norm": 3.199288571835217,
+      "learning_rate": 2.5273475234502565e-06,
+      "loss": 1.0095,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7761362567046203,
+      "grad_norm": 2.7782162251256883,
+      "learning_rate": 2.511161152981604e-06,
+      "loss": 0.9286,
+      "step": 1031
+    },
+    {
+      "epoch": 0.7768890561776607,
+      "grad_norm": 2.3999364661213196,
+      "learning_rate": 2.4950193360603868e-06,
+      "loss": 0.991,
+      "step": 1032
+    },
+    {
+      "epoch": 0.7776418556507011,
+      "grad_norm": 3.4885352475618334,
+      "learning_rate": 2.4789221687195473e-06,
+      "loss": 0.9268,
+      "step": 1033
+    },
+    {
+      "epoch": 0.7783946551237414,
+      "grad_norm": 2.702835220531536,
+      "learning_rate": 2.4628697467263916e-06,
+      "loss": 0.918,
+      "step": 1034
+    },
+    {
+      "epoch": 0.7791474545967818,
+      "grad_norm": 2.3672163089187066,
+      "learning_rate": 2.4468621655820125e-06,
+      "loss": 1.0267,
+      "step": 1035
+    },
+    {
+      "epoch": 0.7799002540698221,
+      "grad_norm": 3.3617116952028856,
+      "learning_rate": 2.430899520520741e-06,
+      "loss": 1.1033,
+      "step": 1036
+    },
+    {
+      "epoch": 0.7806530535428625,
+      "grad_norm": 2.605061703191369,
+      "learning_rate": 2.414981906509565e-06,
+      "loss": 0.9952,
+      "step": 1037
+    },
+    {
+      "epoch": 0.7814058530159029,
+      "grad_norm": 2.7895797271741127,
+      "learning_rate": 2.399109418247563e-06,
+      "loss": 0.95,
+      "step": 1038
+    },
+    {
+      "epoch": 0.7821586524889432,
+      "grad_norm": 2.9660703069176355,
+      "learning_rate": 2.383282150165358e-06,
+      "loss": 0.9452,
+      "step": 1039
+    },
+    {
+      "epoch": 0.7829114519619836,
+      "grad_norm": 2.5443722224387058,
+      "learning_rate": 2.367500196424529e-06,
+      "loss": 0.9476,
+      "step": 1040
+    },
+    {
+      "epoch": 0.783664251435024,
+      "grad_norm": 3.0632555247984987,
+      "learning_rate": 2.351763650917074e-06,
+      "loss": 0.9636,
+      "step": 1041
+    },
+    {
+      "epoch": 0.7844170509080644,
+      "grad_norm": 2.574855917383529,
+      "learning_rate": 2.33607260726484e-06,
+      "loss": 0.9424,
+      "step": 1042
+    },
+    {
+      "epoch": 0.7851698503811048,
+      "grad_norm": 2.6815683468237927,
+      "learning_rate": 2.3204271588189685e-06,
+      "loss": 0.9793,
+      "step": 1043
+    },
+    {
+      "epoch": 0.7859226498541451,
+      "grad_norm": 2.4551608958468827,
+      "learning_rate": 2.304827398659342e-06,
+      "loss": 0.8961,
+      "step": 1044
+    },
+    {
+      "epoch": 0.7866754493271855,
+      "grad_norm": 3.1263512494254813,
+      "learning_rate": 2.289273419594027e-06,
+      "loss": 0.9426,
+      "step": 1045
+    },
+    {
+      "epoch": 0.7874282488002259,
+      "grad_norm": 3.586836506696256,
+      "learning_rate": 2.2737653141587203e-06,
+      "loss": 1.021,
+      "step": 1046
+    },
+    {
+      "epoch": 0.7881810482732662,
+      "grad_norm": 2.6987281971760178,
+      "learning_rate": 2.258303174616204e-06,
+      "loss": 0.999,
+      "step": 1047
+    },
+    {
+      "epoch": 0.7889338477463066,
+      "grad_norm": 2.8798605799038755,
+      "learning_rate": 2.2428870929558012e-06,
+      "loss": 0.9594,
+      "step": 1048
+    },
+    {
+      "epoch": 0.7896866472193469,
+      "grad_norm": 3.113718806953596,
+      "learning_rate": 2.2275171608928124e-06,
+      "loss": 0.9488,
+      "step": 1049
+    },
+    {
+      "epoch": 0.7904394466923873,
+      "grad_norm": 4.380703171601769,
+      "learning_rate": 2.2121934698679793e-06,
+      "loss": 0.963,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7911922461654277,
+      "grad_norm": 2.9209377610680325,
+      "learning_rate": 2.196916111046944e-06,
+      "loss": 0.9774,
+      "step": 1051
+    },
+    {
+      "epoch": 0.791945045638468,
+      "grad_norm": 2.7181240542221903,
+      "learning_rate": 2.1816851753197023e-06,
+      "loss": 0.8549,
+      "step": 1052
+    },
+    {
+      "epoch": 0.7926978451115084,
+      "grad_norm": 2.4755549085650697,
+      "learning_rate": 2.166500753300065e-06,
+      "loss": 0.93,
+      "step": 1053
+    },
+    {
+      "epoch": 0.7934506445845488,
+      "grad_norm": 2.757372825800587,
+      "learning_rate": 2.151362935325115e-06,
+      "loss": 1.0114,
+      "step": 1054
+    },
+    {
+      "epoch": 0.7942034440575891,
+      "grad_norm": 2.915932886320286,
+      "learning_rate": 2.1362718114546777e-06,
+      "loss": 1.0286,
+      "step": 1055
+    },
+    {
+      "epoch": 0.7949562435306295,
+      "grad_norm": 2.1476012635980957,
+      "learning_rate": 2.121227471470768e-06,
+      "loss": 0.9105,
+      "step": 1056
+    },
+    {
+      "epoch": 0.7957090430036698,
+      "grad_norm": 3.423807719412449,
+      "learning_rate": 2.1062300048770847e-06,
+      "loss": 0.9917,
+      "step": 1057
+    },
+    {
+      "epoch": 0.7964618424767103,
+      "grad_norm": 2.640799998383572,
+      "learning_rate": 2.09127950089845e-06,
+      "loss": 1.076,
+      "step": 1058
+    },
+    {
+      "epoch": 0.7972146419497507,
+      "grad_norm": 2.5105814202222456,
+      "learning_rate": 2.0763760484802966e-06,
+      "loss": 0.9166,
+      "step": 1059
+    },
+    {
+      "epoch": 0.797967441422791,
+      "grad_norm": 2.4939451327072546,
+      "learning_rate": 2.0615197362881234e-06,
+      "loss": 0.9807,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7987202408958314,
+      "grad_norm": 2.618028999667555,
+      "learning_rate": 2.046710652706985e-06,
+      "loss": 0.9361,
+      "step": 1061
+    },
+    {
+      "epoch": 0.7994730403688718,
+      "grad_norm": 2.7939434505437695,
+      "learning_rate": 2.0319488858409552e-06,
+      "loss": 0.9447,
+      "step": 1062
+    },
+    {
+      "epoch": 0.8002258398419121,
+      "grad_norm": 4.901271577317021,
+      "learning_rate": 2.0172345235126043e-06,
+      "loss": 1.0124,
+      "step": 1063
+    },
+    {
+      "epoch": 0.8009786393149525,
+      "grad_norm": 3.2082041335316602,
+      "learning_rate": 2.0025676532624794e-06,
+      "loss": 0.9451,
+      "step": 1064
+    },
+    {
+      "epoch": 0.8009786393149525,
+      "eval_loss": 0.955357015132904,
+      "eval_runtime": 584.7085,
+      "eval_samples_per_second": 30.588,
+      "eval_steps_per_second": 0.479,
+      "step": 1064
+    },
+    {
+      "epoch": 0.8017314387879928,
+      "grad_norm": 2.386836563345872,
+      "learning_rate": 1.9879483623485786e-06,
+      "loss": 0.9693,
+      "step": 1065
+    },
+    {
+      "epoch": 0.8024842382610332,
+      "grad_norm": 2.7219810252216106,
+      "learning_rate": 1.9733767377458377e-06,
+      "loss": 0.9822,
+      "step": 1066
+    },
+    {
+      "epoch": 0.8032370377340736,
+      "grad_norm": 2.718832993746013,
+      "learning_rate": 1.9588528661456087e-06,
+      "loss": 0.9315,
+      "step": 1067
+    },
+    {
+      "epoch": 0.8039898372071139,
+      "grad_norm": 2.6930921754180974,
+      "learning_rate": 1.944376833955147e-06,
+      "loss": 1.0321,
+      "step": 1068
+    },
+    {
+      "epoch": 0.8047426366801543,
+      "grad_norm": 2.6093370687591784,
+      "learning_rate": 1.929948727297096e-06,
+      "loss": 0.9255,
+      "step": 1069
+    },
+    {
+      "epoch": 0.8054954361531947,
+      "grad_norm": 2.638629387462051,
+      "learning_rate": 1.9155686320089684e-06,
+      "loss": 0.937,
+      "step": 1070
+    },
+    {
+      "epoch": 0.806248235626235,
+      "grad_norm": 2.7812671767987975,
+      "learning_rate": 1.901236633642649e-06,
+      "loss": 0.9696,
+      "step": 1071
+    },
+    {
+      "epoch": 0.8070010350992755,
+      "grad_norm": 2.5334036447302144,
+      "learning_rate": 1.8869528174638752e-06,
+      "loss": 0.9251,
+      "step": 1072
+    },
+    {
+      "epoch": 0.8077538345723158,
+      "grad_norm": 5.233440136651286,
+      "learning_rate": 1.8727172684517325e-06,
+      "loss": 1.0172,
+      "step": 1073
+    },
+    {
+      "epoch": 0.8085066340453562,
+      "grad_norm": 3.2625149189009814,
+      "learning_rate": 1.8585300712981514e-06,
+      "loss": 0.9732,
+      "step": 1074
+    },
+    {
+      "epoch": 0.8092594335183966,
+      "grad_norm": 2.59047640894747,
+      "learning_rate": 1.8443913104073984e-06,
+      "loss": 1.0292,
+      "step": 1075
+    },
+    {
+      "epoch": 0.8100122329914369,
+      "grad_norm": 3.406989263786069,
+      "learning_rate": 1.8303010698955803e-06,
+      "loss": 0.9894,
+      "step": 1076
+    },
+    {
+      "epoch": 0.8107650324644773,
+      "grad_norm": 2.901678111605894,
+      "learning_rate": 1.8162594335901363e-06,
+      "loss": 1.0178,
+      "step": 1077
+    },
+    {
+      "epoch": 0.8115178319375177,
+      "grad_norm": 3.799031519554636,
+      "learning_rate": 1.802266485029347e-06,
+      "loss": 1.0145,
+      "step": 1078
+    },
+    {
+      "epoch": 0.812270631410558,
+      "grad_norm": 2.4781380441389986,
+      "learning_rate": 1.7883223074618316e-06,
+      "loss": 0.9426,
+      "step": 1079
+    },
+    {
+      "epoch": 0.8130234308835984,
+      "grad_norm": 2.887790196949554,
+      "learning_rate": 1.774426983846058e-06,
+      "loss": 0.9036,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8137762303566387,
+      "grad_norm": 3.0308047796368385,
+      "learning_rate": 1.760580596849838e-06,
+      "loss": 0.9406,
+      "step": 1081
+    },
+    {
+      "epoch": 0.8145290298296791,
+      "grad_norm": 2.888803583084765,
+      "learning_rate": 1.746783228849851e-06,
+      "loss": 0.9907,
+      "step": 1082
+    },
+    {
+      "epoch": 0.8152818293027195,
+      "grad_norm": 3.0358192779029864,
+      "learning_rate": 1.7330349619311415e-06,
+      "loss": 0.9547,
+      "step": 1083
+    },
+    {
+      "epoch": 0.8160346287757598,
+      "grad_norm": 2.5236660699183138,
+      "learning_rate": 1.7193358778866464e-06,
+      "loss": 0.968,
+      "step": 1084
+    },
+    {
+      "epoch": 0.8167874282488002,
+      "grad_norm": 4.185827467185533,
+      "learning_rate": 1.7056860582166823e-06,
+      "loss": 1.0663,
+      "step": 1085
+    },
+    {
+      "epoch": 0.8175402277218405,
+      "grad_norm": 2.6918458931338716,
+      "learning_rate": 1.6920855841284844e-06,
+      "loss": 0.9741,
+      "step": 1086
+    },
+    {
+      "epoch": 0.818293027194881,
+      "grad_norm": 3.3467900392243717,
+      "learning_rate": 1.6785345365357153e-06,
+      "loss": 0.9688,
+      "step": 1087
+    },
+    {
+      "epoch": 0.8190458266679214,
+      "grad_norm": 2.574503648578952,
+      "learning_rate": 1.6650329960579792e-06,
+      "loss": 0.8948,
+      "step": 1088
+    },
+    {
+      "epoch": 0.8197986261409617,
+      "grad_norm": 2.906207704777791,
+      "learning_rate": 1.6515810430203516e-06,
+      "loss": 0.98,
+      "step": 1089
+    },
+    {
+      "epoch": 0.8205514256140021,
+      "grad_norm": 2.3838491968983204,
+      "learning_rate": 1.638178757452894e-06,
+      "loss": 0.908,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8213042250870425,
+      "grad_norm": 2.6732775130808295,
+      "learning_rate": 1.624826219090172e-06,
+      "loss": 0.9864,
+      "step": 1091
+    },
+    {
+      "epoch": 0.8220570245600828,
+      "grad_norm": 2.6917421373345367,
+      "learning_rate": 1.6115235073708024e-06,
+      "loss": 0.9564,
+      "step": 1092
+    },
+    {
+      "epoch": 0.8228098240331232,
+      "grad_norm": 2.8465083001648837,
+      "learning_rate": 1.5982707014369603e-06,
+      "loss": 0.9958,
+      "step": 1093
+    },
+    {
+      "epoch": 0.8235626235061635,
+      "grad_norm": 2.859612454624599,
+      "learning_rate": 1.585067880133916e-06,
+      "loss": 0.996,
+      "step": 1094
+    },
+    {
+      "epoch": 0.8243154229792039,
+      "grad_norm": 2.8110918785044547,
+      "learning_rate": 1.5719151220095596e-06,
+      "loss": 0.9513,
+      "step": 1095
+    },
+    {
+      "epoch": 0.8250682224522443,
+      "grad_norm": 3.041627129196771,
+      "learning_rate": 1.558812505313947e-06,
+      "loss": 0.9961,
+      "step": 1096
+    },
+    {
+      "epoch": 0.8258210219252846,
+      "grad_norm": 3.120842891505949,
+      "learning_rate": 1.5457601079988226e-06,
+      "loss": 0.9401,
+      "step": 1097
+    },
+    {
+      "epoch": 0.826573821398325,
+      "grad_norm": 2.753472364352601,
+      "learning_rate": 1.5327580077171589e-06,
+      "loss": 0.9639,
+      "step": 1098
+    },
+    {
+      "epoch": 0.8273266208713654,
+      "grad_norm": 3.3039694472199703,
+      "learning_rate": 1.5198062818226967e-06,
+      "loss": 0.9655,
+      "step": 1099
+    },
+    {
+      "epoch": 0.8280794203444057,
+      "grad_norm": 2.566689484677852,
+      "learning_rate": 1.5069050073694813e-06,
+      "loss": 0.8969,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8288322198174461,
+      "grad_norm": 2.783130384357118,
+      "learning_rate": 1.4940542611114073e-06,
+      "loss": 0.9337,
+      "step": 1101
+    },
+    {
+      "epoch": 0.8295850192904864,
+      "grad_norm": 2.6527066025848516,
+      "learning_rate": 1.4812541195017593e-06,
+      "loss": 1.0317,
+      "step": 1102
+    },
+    {
+      "epoch": 0.8303378187635269,
+      "grad_norm": 3.0204607327918818,
+      "learning_rate": 1.4685046586927598e-06,
+      "loss": 0.9498,
+      "step": 1103
+    },
+    {
+      "epoch": 0.8310906182365673,
+      "grad_norm": 3.8488215000090986,
+      "learning_rate": 1.4558059545351144e-06,
+      "loss": 0.928,
+      "step": 1104
+    },
+    {
+      "epoch": 0.8318434177096076,
+      "grad_norm": 2.6329112944573865,
+      "learning_rate": 1.4431580825775604e-06,
+      "loss": 0.9369,
+      "step": 1105
+    },
+    {
+      "epoch": 0.832596217182648,
+      "grad_norm": 3.142878133761454,
+      "learning_rate": 1.4305611180664157e-06,
+      "loss": 1.0072,
+      "step": 1106
+    },
+    {
+      "epoch": 0.8333490166556884,
+      "grad_norm": 4.196882021549821,
+      "learning_rate": 1.4180151359451367e-06,
+      "loss": 0.9551,
+      "step": 1107
+    },
+    {
+      "epoch": 0.8341018161287287,
+      "grad_norm": 2.6280348669160283,
+      "learning_rate": 1.4055202108538657e-06,
+      "loss": 0.9148,
+      "step": 1108
+    },
+    {
+      "epoch": 0.8348546156017691,
+      "grad_norm": 2.51099856856607,
+      "learning_rate": 1.3930764171289935e-06,
+      "loss": 0.9765,
+      "step": 1109
+    },
+    {
+      "epoch": 0.8356074150748094,
+      "grad_norm": 2.6019099844031737,
+      "learning_rate": 1.3806838288027113e-06,
+      "loss": 1.0087,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8363602145478498,
+      "grad_norm": 3.4496452020581043,
+      "learning_rate": 1.3683425196025734e-06,
+      "loss": 0.8536,
+      "step": 1111
+    },
+    {
+      "epoch": 0.8371130140208902,
+      "grad_norm": 2.8099366285977863,
+      "learning_rate": 1.3560525629510567e-06,
+      "loss": 0.9274,
+      "step": 1112
+    },
+    {
+      "epoch": 0.8378658134939305,
+      "grad_norm": 2.8308845223177586,
+      "learning_rate": 1.3438140319651283e-06,
+      "loss": 0.9104,
+      "step": 1113
+    },
+    {
+      "epoch": 0.8386186129669709,
+      "grad_norm": 2.3951322032617295,
+      "learning_rate": 1.331626999455804e-06,
+      "loss": 0.9188,
+      "step": 1114
+    },
+    {
+      "epoch": 0.8393714124400113,
+      "grad_norm": 3.35662443745714,
+      "learning_rate": 1.3194915379277195e-06,
+      "loss": 0.9918,
+      "step": 1115
+    },
+    {
+      "epoch": 0.8401242119130516,
+      "grad_norm": 3.588345323887508,
+      "learning_rate": 1.307407719578696e-06,
+      "loss": 0.9311,
+      "step": 1116
+    },
+    {
+      "epoch": 0.840877011386092,
+      "grad_norm": 2.7850061242387656,
+      "learning_rate": 1.2953756162993158e-06,
+      "loss": 0.9796,
+      "step": 1117
+    },
+    {
+      "epoch": 0.8416298108591324,
+      "grad_norm": 2.571025177959364,
+      "learning_rate": 1.2833952996724864e-06,
+      "loss": 0.9475,
+      "step": 1118
+    },
+    {
+      "epoch": 0.8423826103321728,
+      "grad_norm": 3.3900850223254237,
+      "learning_rate": 1.2714668409730312e-06,
+      "loss": 0.9553,
+      "step": 1119
+    },
+    {
+      "epoch": 0.8431354098052132,
+      "grad_norm": 3.1347132178889106,
+      "learning_rate": 1.259590311167238e-06,
+      "loss": 1.0453,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8438882092782535,
+      "grad_norm": 2.876010919809789,
+      "learning_rate": 1.2477657809124632e-06,
+      "loss": 0.8553,
+      "step": 1121
+    },
+    {
+      "epoch": 0.8446410087512939,
+      "grad_norm": 2.887022175621213,
+      "learning_rate": 1.2359933205566987e-06,
+      "loss": 0.9759,
+      "step": 1122
+    },
+    {
+      "epoch": 0.8453938082243343,
+      "grad_norm": 2.434068422278726,
+      "learning_rate": 1.2242730001381532e-06,
+      "loss": 0.9128,
+      "step": 1123
+    },
+    {
+      "epoch": 0.8461466076973746,
+      "grad_norm": 2.736589284541001,
+      "learning_rate": 1.2126048893848396e-06,
+      "loss": 0.9591,
+      "step": 1124
+    },
+    {
+      "epoch": 0.846899407170415,
+      "grad_norm": 2.4174009434081496,
+      "learning_rate": 1.2009890577141625e-06,
+      "loss": 1.0533,
+      "step": 1125
+    },
+    {
+      "epoch": 0.8476522066434553,
+      "grad_norm": 2.854192293105523,
+      "learning_rate": 1.189425574232491e-06,
+      "loss": 0.9179,
+      "step": 1126
+    },
+    {
+      "epoch": 0.8484050061164957,
+      "grad_norm": 2.7270460687471627,
+      "learning_rate": 1.1779145077347653e-06,
+      "loss": 0.92,
+      "step": 1127
+    },
+    {
+      "epoch": 0.8491578055895361,
+      "grad_norm": 3.292903949444069,
+      "learning_rate": 1.1664559267040821e-06,
+      "loss": 1.0056,
+      "step": 1128
+    },
+    {
+      "epoch": 0.8499106050625764,
+      "grad_norm": 3.1573702260917984,
+      "learning_rate": 1.1550498993112812e-06,
+      "loss": 0.9981,
+      "step": 1129
+    },
+    {
+      "epoch": 0.8506634045356168,
+      "grad_norm": 2.566537819994616,
+      "learning_rate": 1.143696493414539e-06,
+      "loss": 1.0057,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8514162040086571,
+      "grad_norm": 3.1010812728125057,
+      "learning_rate": 1.1323957765589766e-06,
+      "loss": 0.9657,
+      "step": 1131
+    },
+    {
+      "epoch": 0.8521690034816976,
+      "grad_norm": 3.7067774109664624,
+      "learning_rate": 1.121147815976248e-06,
+      "loss": 1.0551,
+      "step": 1132
+    },
+    {
+      "epoch": 0.852921802954738,
+      "grad_norm": 2.7515838856907906,
+      "learning_rate": 1.109952678584144e-06,
+      "loss": 0.9818,
+      "step": 1133
+    },
+    {
+      "epoch": 0.8536746024277783,
+      "grad_norm": 3.4640444745995205,
+      "learning_rate": 1.0988104309861913e-06,
+      "loss": 0.8982,
+      "step": 1134
+    },
+    {
+      "epoch": 0.8544274019008187,
+      "grad_norm": 3.7011907670434656,
+      "learning_rate": 1.0877211394712617e-06,
+      "loss": 1.0956,
+      "step": 1135
+    },
+    {
+      "epoch": 0.8551802013738591,
+      "grad_norm": 3.187205634209682,
+      "learning_rate": 1.076684870013165e-06,
+      "loss": 0.9818,
+      "step": 1136
+    },
+    {
+      "epoch": 0.8559330008468994,
+      "grad_norm": 3.0989118973167518,
+      "learning_rate": 1.0657016882702764e-06,
+      "loss": 0.9672,
+      "step": 1137
+    },
+    {
+      "epoch": 0.8566858003199398,
+      "grad_norm": 2.7075586038664605,
+      "learning_rate": 1.0547716595851298e-06,
+      "loss": 0.9506,
+      "step": 1138
+    },
+    {
+      "epoch": 0.8574385997929801,
+      "grad_norm": 3.1360256431016587,
+      "learning_rate": 1.0438948489840327e-06,
+      "loss": 0.9602,
+      "step": 1139
+    },
+    {
+      "epoch": 0.8581913992660205,
+      "grad_norm": 3.260413556431712,
+      "learning_rate": 1.0330713211766864e-06,
+      "loss": 0.936,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8589441987390609,
+      "grad_norm": 2.762546618163618,
+      "learning_rate": 1.022301140555787e-06,
+      "loss": 0.9092,
+      "step": 1141
+    },
+    {
+      "epoch": 0.8596969982121012,
+      "grad_norm": 3.027720830975655,
+      "learning_rate": 1.0115843711966577e-06,
+      "loss": 0.9336,
+      "step": 1142
+    },
+    {
+      "epoch": 0.8604497976851416,
+      "grad_norm": 2.4551963085300543,
+      "learning_rate": 1.000921076856859e-06,
+      "loss": 0.9253,
+      "step": 1143
+    },
+    {
+      "epoch": 0.861202597158182,
+      "grad_norm": 2.5990984905624566,
+      "learning_rate": 9.903113209758098e-07,
+      "loss": 0.9247,
+      "step": 1144
+    },
+    {
+      "epoch": 0.8619553966312223,
+      "grad_norm": 2.520291676745646,
+      "learning_rate": 9.79755166674411e-07,
+      "loss": 0.9327,
+      "step": 1145
+    },
+    {
+      "epoch": 0.8627081961042627,
+      "grad_norm": 2.9740303315319645,
+      "learning_rate": 9.692526767546727e-07,
+      "loss": 0.936,
+      "step": 1146
+    },
+    {
+      "epoch": 0.863460995577303,
+      "grad_norm": 2.8172674563090876,
+      "learning_rate": 9.588039136993366e-07,
+      "loss": 0.9249,
+      "step": 1147
+    },
+    {
+      "epoch": 0.8642137950503435,
+      "grad_norm": 2.6866056844439314,
+      "learning_rate": 9.484089396715057e-07,
+      "loss": 0.9751,
+      "step": 1148
+    },
+    {
+      "epoch": 0.8649665945233839,
+      "grad_norm": 2.4806642645953416,
+      "learning_rate": 9.380678165142732e-07,
+      "loss": 0.9758,
+      "step": 1149
+    },
+    {
+      "epoch": 0.8657193939964242,
+      "grad_norm": 2.673649491913748,
+      "learning_rate": 9.277806057503592e-07,
+      "loss": 0.8811,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8664721934694646,
+      "grad_norm": 4.406342851150164,
+      "learning_rate": 9.175473685817371e-07,
+      "loss": 0.8679,
+      "step": 1151
+    },
+    {
+      "epoch": 0.867224992942505,
+      "grad_norm": 3.7985140736623815,
+      "learning_rate": 9.073681658892775e-07,
+      "loss": 1.0464,
+      "step": 1152
+    },
+    {
+      "epoch": 0.8679777924155453,
+      "grad_norm": 2.5754256214115125,
+      "learning_rate": 8.972430582323788e-07,
+      "loss": 0.978,
+      "step": 1153
+    },
+    {
+      "epoch": 0.8687305918885857,
+      "grad_norm": 2.6776748283021594,
+      "learning_rate": 8.871721058486149e-07,
+      "loss": 0.9401,
+      "step": 1154
+    },
+    {
+      "epoch": 0.869483391361626,
+      "grad_norm": 2.5262861152205613,
+      "learning_rate": 8.771553686533684e-07,
+      "loss": 0.9202,
+      "step": 1155
+    },
+    {
+      "epoch": 0.8702361908346664,
+      "grad_norm": 2.5512587089365097,
+      "learning_rate": 8.671929062394802e-07,
+      "loss": 0.9661,
+      "step": 1156
+    },
+    {
+      "epoch": 0.8709889903077068,
+      "grad_norm": 3.0309053039828875,
+      "learning_rate": 8.572847778768912e-07,
+      "loss": 1.0411,
+      "step": 1157
+    },
+    {
+      "epoch": 0.8717417897807471,
+      "grad_norm": 2.870089467163631,
+      "learning_rate": 8.474310425122923e-07,
+      "loss": 0.9452,
+      "step": 1158
+    },
+    {
+      "epoch": 0.8724945892537875,
+      "grad_norm": 2.7251037878757014,
+      "learning_rate": 8.376317587687721e-07,
+      "loss": 0.9643,
+      "step": 1159
+    },
+    {
+      "epoch": 0.8732473887268279,
+      "grad_norm": 3.0515241331021206,
+      "learning_rate": 8.278869849454718e-07,
+      "loss": 0.9511,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8740001881998682,
+      "grad_norm": 2.7559306109031327,
+      "learning_rate": 8.181967790172274e-07,
+      "loss": 0.9655,
+      "step": 1161
+    },
+    {
+      "epoch": 0.8747529876729087,
+      "grad_norm": 3.005111485431956,
+      "learning_rate": 8.085611986342423e-07,
+      "loss": 0.9264,
+      "step": 1162
+    },
+    {
+      "epoch": 0.875505787145949,
+      "grad_norm": 3.183291159476253,
+      "learning_rate": 7.989803011217256e-07,
+      "loss": 0.9755,
+      "step": 1163
+    },
+    {
+      "epoch": 0.8762585866189894,
+      "grad_norm": 3.6302734872126976,
+      "learning_rate": 7.8945414347957e-07,
+      "loss": 1.0719,
+      "step": 1164
+    },
+    {
+      "epoch": 0.8770113860920298,
+      "grad_norm": 2.200336710545509,
+      "learning_rate": 7.799827823819972e-07,
+      "loss": 0.8843,
+      "step": 1165
+    },
+    {
+      "epoch": 0.8777641855650701,
+      "grad_norm": 2.7624309633223345,
+      "learning_rate": 7.705662741772235e-07,
+      "loss": 1.0182,
+      "step": 1166
+    },
+    {
+      "epoch": 0.8785169850381105,
+      "grad_norm": 3.1072712239860523,
+      "learning_rate": 7.612046748871327e-07,
+      "loss": 1.0004,
+      "step": 1167
+    },
+    {
+      "epoch": 0.8792697845111508,
+      "grad_norm": 2.65712921508899,
+      "learning_rate": 7.518980402069354e-07,
+      "loss": 0.9294,
+      "step": 1168
+    },
+    {
+      "epoch": 0.8800225839841912,
+      "grad_norm": 2.7368564706264813,
+      "learning_rate": 7.426464255048393e-07,
+      "loss": 0.9607,
+      "step": 1169
+    },
+    {
+      "epoch": 0.8807753834572316,
+      "grad_norm": 2.441360830618783,
+      "learning_rate": 7.334498858217231e-07,
+      "loss": 0.9321,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815281829302719,
+      "grad_norm": 2.788678307102662,
+      "learning_rate": 7.243084758708007e-07,
+      "loss": 0.989,
+      "step": 1171
+    },
+    {
+      "epoch": 0.8822809824033123,
+      "grad_norm": 2.5226447327049355,
+      "learning_rate": 7.152222500373052e-07,
+      "loss": 0.9305,
+      "step": 1172
+    },
+    {
+      "epoch": 0.8830337818763527,
+      "grad_norm": 2.7207043507041107,
+      "learning_rate": 7.06191262378163e-07,
+      "loss": 0.9421,
+      "step": 1173
+    },
+    {
+      "epoch": 0.883786581349393,
+      "grad_norm": 2.623692509214404,
+      "learning_rate": 6.972155666216684e-07,
+      "loss": 1.031,
+      "step": 1174
+    },
+    {
+      "epoch": 0.8845393808224334,
+      "grad_norm": 2.599723287251219,
+      "learning_rate": 6.882952161671652e-07,
+      "loss": 0.9094,
+      "step": 1175
+    },
+    {
+      "epoch": 0.8852921802954737,
+      "grad_norm": 2.4580215454865586,
+      "learning_rate": 6.794302640847294e-07,
+      "loss": 0.9175,
+      "step": 1176
+    },
+    {
+      "epoch": 0.8860449797685142,
+      "grad_norm": 2.4645666417160097,
+      "learning_rate": 6.706207631148564e-07,
+      "loss": 0.8983,
+      "step": 1177
+    },
+    {
+      "epoch": 0.8867977792415546,
+      "grad_norm": 3.0348308308305962,
+      "learning_rate": 6.618667656681444e-07,
+      "loss": 0.9817,
+      "step": 1178
+    },
+    {
+      "epoch": 0.8875505787145949,
+      "grad_norm": 2.4790311726644663,
+      "learning_rate": 6.531683238249809e-07,
+      "loss": 0.9532,
+      "step": 1179
+    },
+    {
+      "epoch": 0.8883033781876353,
+      "grad_norm": 2.6924152713824956,
+      "learning_rate": 6.445254893352381e-07,
+      "loss": 1.0677,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890561776606757,
+      "grad_norm": 2.6608700924933135,
+      "learning_rate": 6.359383136179598e-07,
+      "loss": 0.8722,
+      "step": 1181
+    },
+    {
+      "epoch": 0.889808977133716,
+      "grad_norm": 2.6182698749250695,
+      "learning_rate": 6.274068477610584e-07,
+      "loss": 0.9527,
+      "step": 1182
+    },
+    {
+      "epoch": 0.8905617766067564,
+      "grad_norm": 6.3305830668644845,
+      "learning_rate": 6.189311425210087e-07,
+      "loss": 0.9959,
+      "step": 1183
+    },
+    {
+      "epoch": 0.8913145760797967,
+      "grad_norm": 3.0187373017741392,
+      "learning_rate": 6.105112483225495e-07,
+      "loss": 1.0358,
+      "step": 1184
+    },
+    {
+      "epoch": 0.8920673755528371,
+      "grad_norm": 2.7315069757579913,
+      "learning_rate": 6.021472152583818e-07,
+      "loss": 0.9743,
+      "step": 1185
+    },
+    {
+      "epoch": 0.8928201750258775,
+      "grad_norm": 2.827226835891074,
+      "learning_rate": 5.938390930888671e-07,
+      "loss": 0.928,
+      "step": 1186
+    },
+    {
+      "epoch": 0.8935729744989178,
+      "grad_norm": 2.7665857256735125,
+      "learning_rate": 5.855869312417362e-07,
+      "loss": 0.8687,
+      "step": 1187
+    },
+    {
+      "epoch": 0.8943257739719582,
+      "grad_norm": 2.5980919886852676,
+      "learning_rate": 5.77390778811796e-07,
+      "loss": 0.9564,
+      "step": 1188
+    },
+    {
+      "epoch": 0.8950785734449986,
+      "grad_norm": 2.6210409082659245,
+      "learning_rate": 5.692506845606327e-07,
+      "loss": 0.9545,
+      "step": 1189
+    },
+    {
+      "epoch": 0.8958313729180389,
+      "grad_norm": 3.662360614683105,
+      "learning_rate": 5.611666969163243e-07,
+      "loss": 0.9616,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965841723910793,
+      "grad_norm": 3.147025077582227,
+      "learning_rate": 5.53138863973155e-07,
+      "loss": 1.0046,
+      "step": 1191
+    },
+    {
+      "epoch": 0.8973369718641196,
+      "grad_norm": 3.1345035339900438,
+      "learning_rate": 5.451672334913216e-07,
+      "loss": 0.9031,
+      "step": 1192
+    },
+    {
+      "epoch": 0.8980897713371601,
+      "grad_norm": 2.8518474768796778,
+      "learning_rate": 5.372518528966575e-07,
+      "loss": 0.9732,
+      "step": 1193
+    },
+    {
+      "epoch": 0.8988425708102005,
+      "grad_norm": 3.062727839155735,
+      "learning_rate": 5.293927692803458e-07,
+      "loss": 1.0282,
+      "step": 1194
+    },
+    {
+      "epoch": 0.8995953702832408,
+      "grad_norm": 2.9353100401019323,
+      "learning_rate": 5.215900293986431e-07,
+      "loss": 1.0028,
+      "step": 1195
+    },
+    {
+      "epoch": 0.9003481697562812,
+      "grad_norm": 2.6913340260060594,
+      "learning_rate": 5.138436796725942e-07,
+      "loss": 0.9665,
+      "step": 1196
+    },
+    {
+      "epoch": 0.9011009692293216,
+      "grad_norm": 2.8934435608082842,
+      "learning_rate": 5.061537661877636e-07,
+      "loss": 1.0446,
+      "step": 1197
+    },
+    {
+      "epoch": 0.9011009692293216,
+      "eval_loss": 0.9507451057434082,
+      "eval_runtime": 583.6153,
+      "eval_samples_per_second": 30.645,
+      "eval_steps_per_second": 0.48,
+      "step": 1197
+    },
+    {
+      "epoch": 0.9018537687023619,
+      "grad_norm": 2.553104219592107,
+      "learning_rate": 4.98520334693956e-07,
+      "loss": 0.9547,
+      "step": 1198
+    },
+    {
+      "epoch": 0.9026065681754023,
+      "grad_norm": 2.7253032781503657,
+      "learning_rate": 4.909434306049487e-07,
+      "loss": 0.9249,
+      "step": 1199
+    },
+    {
+      "epoch": 0.9033593676484426,
+      "grad_norm": 3.2709141351515285,
+      "learning_rate": 4.834230989982214e-07,
+      "loss": 0.9728,
+      "step": 1200
+    },
+    {
+      "epoch": 0.904112167121483,
+      "grad_norm": 4.401571356782588,
+      "learning_rate": 4.7595938461467706e-07,
+      "loss": 0.9535,
+      "step": 1201
+    },
+    {
+      "epoch": 0.9048649665945234,
+      "grad_norm": 2.5275547478412093,
+      "learning_rate": 4.6855233185839175e-07,
+      "loss": 0.9745,
+      "step": 1202
+    },
+    {
+      "epoch": 0.9056177660675637,
+      "grad_norm": 2.5001698261652954,
+      "learning_rate": 4.6120198479634117e-07,
+      "loss": 0.9821,
+      "step": 1203
+    },
+    {
+      "epoch": 0.9063705655406041,
+      "grad_norm": 2.6567511968251454,
+      "learning_rate": 4.5390838715813956e-07,
+      "loss": 0.9181,
+      "step": 1204
+    },
+    {
+      "epoch": 0.9071233650136445,
+      "grad_norm": 2.462427713605323,
+      "learning_rate": 4.4667158233577925e-07,
+      "loss": 0.9857,
+      "step": 1205
+    },
+    {
+      "epoch": 0.9078761644866848,
+      "grad_norm": 2.5236048684689476,
+      "learning_rate": 4.394916133833782e-07,
+      "loss": 0.9388,
+      "step": 1206
+    },
+    {
+      "epoch": 0.9086289639597253,
+      "grad_norm": 2.5204539009176443,
+      "learning_rate": 4.323685230169128e-07,
+      "loss": 0.9548,
+      "step": 1207
+    },
+    {
+      "epoch": 0.9093817634327656,
+      "grad_norm": 2.7491122618434414,
+      "learning_rate": 4.253023536139733e-07,
+      "loss": 0.937,
+      "step": 1208
+    },
+    {
+      "epoch": 0.910134562905806,
+      "grad_norm": 2.5076418947884376,
+      "learning_rate": 4.1829314721351213e-07,
+      "loss": 0.9159,
+      "step": 1209
+    },
+    {
+      "epoch": 0.9108873623788464,
+      "grad_norm": 3.1696179972875766,
+      "learning_rate": 4.113409455155837e-07,
+      "loss": 0.877,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9116401618518867,
+      "grad_norm": 2.8529238109300548,
+      "learning_rate": 4.0444578988110715e-07,
+      "loss": 0.9462,
+      "step": 1211
+    },
+    {
+      "epoch": 0.9123929613249271,
+      "grad_norm": 2.90898356943298,
+      "learning_rate": 3.976077213316132e-07,
+      "loss": 0.9328,
+      "step": 1212
+    },
+    {
+      "epoch": 0.9131457607979674,
+      "grad_norm": 2.6862075191363117,
+      "learning_rate": 3.908267805490051e-07,
+      "loss": 1.0102,
+      "step": 1213
+    },
+    {
+      "epoch": 0.9138985602710078,
+      "grad_norm": 2.8255615762352377,
+      "learning_rate": 3.8410300787531385e-07,
+      "loss": 0.9589,
+      "step": 1214
+    },
+    {
+      "epoch": 0.9146513597440482,
+      "grad_norm": 2.862843093830216,
+      "learning_rate": 3.774364433124578e-07,
+      "loss": 0.9288,
+      "step": 1215
+    },
+    {
+      "epoch": 0.9154041592170885,
+      "grad_norm": 2.6717135457398036,
+      "learning_rate": 3.708271265220087e-07,
+      "loss": 0.9827,
+      "step": 1216
+    },
+    {
+      "epoch": 0.9161569586901289,
+      "grad_norm": 2.98813894982846,
+      "learning_rate": 3.642750968249442e-07,
+      "loss": 1.0676,
+      "step": 1217
+    },
+    {
+      "epoch": 0.9169097581631693,
+      "grad_norm": 3.367070226516106,
+      "learning_rate": 3.5778039320143456e-07,
+      "loss": 0.9641,
+      "step": 1218
+    },
+    {
+      "epoch": 0.9176625576362096,
+      "grad_norm": 3.312081763624184,
+      "learning_rate": 3.5134305429058935e-07,
+      "loss": 0.955,
+      "step": 1219
+    },
+    {
+      "epoch": 0.91841535710925,
+      "grad_norm": 3.0015816399724873,
+      "learning_rate": 3.4496311839024133e-07,
+      "loss": 0.9972,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9191681565822903,
+      "grad_norm": 2.851669734888555,
+      "learning_rate": 3.386406234567086e-07,
+      "loss": 0.8808,
+      "step": 1221
+    },
+    {
+      "epoch": 0.9199209560553308,
+      "grad_norm": 3.579793360196461,
+      "learning_rate": 3.3237560710458137e-07,
+      "loss": 0.9801,
+      "step": 1222
+    },
+    {
+      "epoch": 0.9206737555283712,
+      "grad_norm": 2.569101602670157,
+      "learning_rate": 3.261681066064859e-07,
+      "loss": 0.9929,
+      "step": 1223
+    },
+    {
+      "epoch": 0.9214265550014115,
+      "grad_norm": 4.085329324706736,
+      "learning_rate": 3.2001815889286856e-07,
+      "loss": 0.91,
+      "step": 1224
+    },
+    {
+      "epoch": 0.9221793544744519,
+      "grad_norm": 2.713777124027884,
+      "learning_rate": 3.1392580055177867e-07,
+      "loss": 1.0687,
+      "step": 1225
+    },
+    {
+      "epoch": 0.9229321539474923,
+      "grad_norm": 2.680582171824577,
+      "learning_rate": 3.0789106782864285e-07,
+      "loss": 0.9431,
+      "step": 1226
+    },
+    {
+      "epoch": 0.9236849534205326,
+      "grad_norm": 2.7796394938594244,
+      "learning_rate": 3.019139966260587e-07,
+      "loss": 0.9721,
+      "step": 1227
+    },
+    {
+      "epoch": 0.924437752893573,
+      "grad_norm": 3.0975647783515483,
+      "learning_rate": 2.959946225035726e-07,
+      "loss": 0.9531,
+      "step": 1228
+    },
+    {
+      "epoch": 0.9251905523666133,
+      "grad_norm": 2.7977963864759587,
+      "learning_rate": 2.901329806774744e-07,
+      "loss": 0.9394,
+      "step": 1229
+    },
+    {
+      "epoch": 0.9259433518396537,
+      "grad_norm": 2.5453356161001284,
+      "learning_rate": 2.843291060205855e-07,
+      "loss": 0.904,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9266961513126941,
+      "grad_norm": 2.7541961024247144,
+      "learning_rate": 2.785830330620509e-07,
+      "loss": 0.9774,
+      "step": 1231
+    },
+    {
+      "epoch": 0.9274489507857344,
+      "grad_norm": 3.084595078404382,
+      "learning_rate": 2.728947959871353e-07,
+      "loss": 0.8789,
+      "step": 1232
+    },
+    {
+      "epoch": 0.9282017502587748,
+      "grad_norm": 2.838093099116681,
+      "learning_rate": 2.672644286370163e-07,
+      "loss": 0.9272,
+      "step": 1233
+    },
+    {
+      "epoch": 0.9289545497318152,
+      "grad_norm": 3.2906398456754453,
+      "learning_rate": 2.616919645085902e-07,
+      "loss": 0.9266,
+      "step": 1234
+    },
+    {
+      "epoch": 0.9297073492048555,
+      "grad_norm": 2.5775786586459555,
+      "learning_rate": 2.5617743675426354e-07,
+      "loss": 0.8926,
+      "step": 1235
+    },
+    {
+      "epoch": 0.930460148677896,
+      "grad_norm": 2.6722830170773344,
+      "learning_rate": 2.507208781817638e-07,
+      "loss": 0.9202,
+      "step": 1236
+    },
+    {
+      "epoch": 0.9312129481509362,
+      "grad_norm": 2.4434315642032716,
+      "learning_rate": 2.453223212539391e-07,
+      "loss": 0.9665,
+      "step": 1237
+    },
+    {
+      "epoch": 0.9319657476239767,
+      "grad_norm": 3.06220669668657,
+      "learning_rate": 2.399817980885677e-07,
+      "loss": 0.9852,
+      "step": 1238
+    },
+    {
+      "epoch": 0.9327185470970171,
+      "grad_norm": 2.980887110954531,
+      "learning_rate": 2.3469934045816435e-07,
+      "loss": 0.9432,
+      "step": 1239
+    },
+    {
+      "epoch": 0.9334713465700574,
+      "grad_norm": 2.856068250126512,
+      "learning_rate": 2.294749797897955e-07,
+      "loss": 1.0187,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9342241460430978,
+      "grad_norm": 3.5481294720376697,
+      "learning_rate": 2.243087471648886e-07,
+      "loss": 0.9678,
+      "step": 1241
+    },
+    {
+      "epoch": 0.9349769455161382,
+      "grad_norm": 2.990365041826747,
+      "learning_rate": 2.192006733190466e-07,
+      "loss": 0.9186,
+      "step": 1242
+    },
+    {
+      "epoch": 0.9357297449891785,
+      "grad_norm": 2.7209981946105963,
+      "learning_rate": 2.1415078864187034e-07,
+      "loss": 0.8997,
+      "step": 1243
+    },
+    {
+      "epoch": 0.9364825444622189,
+      "grad_norm": 2.633970676325884,
+      "learning_rate": 2.091591231767709e-07,
+      "loss": 0.9298,
+      "step": 1244
+    },
+    {
+      "epoch": 0.9372353439352592,
+      "grad_norm": 2.5883350194236416,
+      "learning_rate": 2.0422570662079866e-07,
+      "loss": 0.9546,
+      "step": 1245
+    },
+    {
+      "epoch": 0.9379881434082996,
+      "grad_norm": 3.2796875101881517,
+      "learning_rate": 1.9935056832445676e-07,
+      "loss": 0.9693,
+      "step": 1246
+    },
+    {
+      "epoch": 0.93874094288134,
+      "grad_norm": 2.5962393649368614,
+      "learning_rate": 1.945337372915368e-07,
+      "loss": 1.0175,
+      "step": 1247
+    },
+    {
+      "epoch": 0.9394937423543803,
+      "grad_norm": 2.7526882993306154,
+      "learning_rate": 1.8977524217893782e-07,
+      "loss": 0.9715,
+      "step": 1248
+    },
+    {
+      "epoch": 0.9402465418274207,
+      "grad_norm": 2.8840274447370993,
+      "learning_rate": 1.85075111296501e-07,
+      "loss": 0.8795,
+      "step": 1249
+    },
+    {
+      "epoch": 0.9409993413004611,
+      "grad_norm": 2.9301887630100554,
+      "learning_rate": 1.804333726068408e-07,
+      "loss": 0.9652,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9417521407735014,
+      "grad_norm": 2.6456393593543144,
+      "learning_rate": 1.7585005372517504e-07,
+      "loss": 0.9918,
+      "step": 1251
+    },
+    {
+      "epoch": 0.9425049402465419,
+      "grad_norm": 2.8411309787841863,
+      "learning_rate": 1.7132518191916413e-07,
+      "loss": 0.9837,
+      "step": 1252
+    },
+    {
+      "epoch": 0.9432577397195822,
+      "grad_norm": 3.2850349406325594,
+      "learning_rate": 1.6685878410874768e-07,
+      "loss": 0.9916,
+      "step": 1253
+    },
+    {
+      "epoch": 0.9440105391926226,
+      "grad_norm": 2.964090383970478,
+      "learning_rate": 1.6245088686598686e-07,
+      "loss": 0.9974,
+      "step": 1254
+    },
+    {
+      "epoch": 0.944763338665663,
+      "grad_norm": 2.6196165472195543,
+      "learning_rate": 1.5810151641489912e-07,
+      "loss": 0.9014,
+      "step": 1255
+    },
+    {
+      "epoch": 0.9455161381387033,
+      "grad_norm": 2.7772230721575615,
+      "learning_rate": 1.5381069863131037e-07,
+      "loss": 0.9592,
+      "step": 1256
+    },
+    {
+      "epoch": 0.9462689376117437,
+      "grad_norm": 2.5463035833372043,
+      "learning_rate": 1.495784590426963e-07,
+      "loss": 0.9416,
+      "step": 1257
+    },
+    {
+      "epoch": 0.947021737084784,
+      "grad_norm": 3.1995799318479943,
+      "learning_rate": 1.4540482282803136e-07,
+      "loss": 1.0492,
+      "step": 1258
+    },
+    {
+      "epoch": 0.9477745365578244,
+      "grad_norm": 3.5523139139820374,
+      "learning_rate": 1.4128981481764115e-07,
+      "loss": 1.016,
+      "step": 1259
+    },
+    {
+      "epoch": 0.9485273360308648,
+      "grad_norm": 3.262601645233588,
+      "learning_rate": 1.3723345949305245e-07,
+      "loss": 0.9621,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9492801355039051,
+      "grad_norm": 2.456450672228116,
+      "learning_rate": 1.3323578098684565e-07,
+      "loss": 0.8501,
+      "step": 1261
+    },
+    {
+      "epoch": 0.9500329349769455,
+      "grad_norm": 3.7186153382386595,
+      "learning_rate": 1.292968030825159e-07,
+      "loss": 0.9515,
+      "step": 1262
+    },
+    {
+      "epoch": 0.9507857344499859,
+      "grad_norm": 3.049161674324443,
+      "learning_rate": 1.2541654921432998e-07,
+      "loss": 1.0811,
+      "step": 1263
+    },
+    {
+      "epoch": 0.9515385339230262,
+      "grad_norm": 3.291544360767655,
+      "learning_rate": 1.2159504246718522e-07,
+      "loss": 0.9972,
+      "step": 1264
+    },
+    {
+      "epoch": 0.9522913333960666,
+      "grad_norm": 3.0656467336632085,
+      "learning_rate": 1.1783230557647075e-07,
+      "loss": 0.9184,
+      "step": 1265
+    },
+    {
+      "epoch": 0.9530441328691069,
+      "grad_norm": 2.628146994913141,
+      "learning_rate": 1.1412836092793977e-07,
+      "loss": 0.8738,
+      "step": 1266
+    },
+    {
+      "epoch": 0.9537969323421474,
+      "grad_norm": 2.8202267098554836,
+      "learning_rate": 1.1048323055756649e-07,
+      "loss": 0.9663,
+      "step": 1267
+    },
+    {
+      "epoch": 0.9545497318151878,
+      "grad_norm": 2.430990179547932,
+      "learning_rate": 1.068969361514216e-07,
+      "loss": 0.9611,
+      "step": 1268
+    },
+    {
+      "epoch": 0.9553025312882281,
+      "grad_norm": 3.248130764229673,
+      "learning_rate": 1.033694990455425e-07,
+      "loss": 0.9268,
+      "step": 1269
+    },
+    {
+      "epoch": 0.9560553307612685,
+      "grad_norm": 3.7522100183518363,
+      "learning_rate": 9.990094022580332e-08,
+      "loss": 0.9114,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9568081302343089,
+      "grad_norm": 3.053183197199831,
+      "learning_rate": 9.649128032779287e-08,
+      "loss": 0.9182,
+      "step": 1271
+    },
+    {
+      "epoch": 0.9575609297073492,
+      "grad_norm": 2.7055521182554965,
+      "learning_rate": 9.314053963669245e-08,
+      "loss": 0.9798,
+      "step": 1272
+    },
+    {
+      "epoch": 0.9583137291803896,
+      "grad_norm": 4.422847847788482,
+      "learning_rate": 8.984873808715155e-08,
+      "loss": 0.9754,
+      "step": 1273
+    },
+    {
+      "epoch": 0.9590665286534299,
+      "grad_norm": 2.4097712053201854,
+      "learning_rate": 8.661589526317238e-08,
+      "loss": 1.0184,
+      "step": 1274
+    },
+    {
+      "epoch": 0.9598193281264703,
+      "grad_norm": 3.3193315006269564,
+      "learning_rate": 8.344203039799214e-08,
+      "loss": 0.9705,
+      "step": 1275
+    },
+    {
+      "epoch": 0.9605721275995107,
+      "grad_norm": 3.230643135053207,
+      "learning_rate": 8.032716237396987e-08,
+      "loss": 1.0181,
+      "step": 1276
+    },
+    {
+      "epoch": 0.961324927072551,
+      "grad_norm": 2.748071452140581,
+      "learning_rate": 7.727130972247199e-08,
+      "loss": 0.9488,
+      "step": 1277
+    },
+    {
+      "epoch": 0.9620777265455914,
+      "grad_norm": 4.801952883628051,
+      "learning_rate": 7.427449062376468e-08,
+      "loss": 0.9536,
+      "step": 1278
+    },
+    {
+      "epoch": 0.9628305260186318,
+      "grad_norm": 2.2950527457208594,
+      "learning_rate": 7.133672290690064e-08,
+      "loss": 1.0118,
+      "step": 1279
+    },
+    {
+      "epoch": 0.9635833254916721,
+      "grad_norm": 2.5220242298362603,
+      "learning_rate": 6.845802404962243e-08,
+      "loss": 0.8761,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9643361249647125,
+      "grad_norm": 3.113693252903444,
+      "learning_rate": 6.56384111782482e-08,
+      "loss": 0.9248,
+      "step": 1281
+    },
+    {
+      "epoch": 0.9650889244377528,
+      "grad_norm": 3.275264466855922,
+      "learning_rate": 6.287790106757396e-08,
+      "loss": 0.9332,
+      "step": 1282
+    },
+    {
+      "epoch": 0.9658417239107933,
+      "grad_norm": 3.1198895526228316,
+      "learning_rate": 6.017651014077807e-08,
+      "loss": 1.0051,
+      "step": 1283
+    },
+    {
+      "epoch": 0.9665945233838337,
+      "grad_norm": 3.2659243106544054,
+      "learning_rate": 5.753425446931582e-08,
+      "loss": 0.9792,
+      "step": 1284
+    },
+    {
+      "epoch": 0.967347322856874,
+      "grad_norm": 2.7005554860553174,
+      "learning_rate": 5.495114977282945e-08,
+      "loss": 1.0284,
+      "step": 1285
+    },
+    {
+      "epoch": 0.9681001223299144,
+      "grad_norm": 2.231612566361707,
+      "learning_rate": 5.2427211419051605e-08,
+      "loss": 0.8459,
+      "step": 1286
+    },
+    {
+      "epoch": 0.9688529218029548,
+      "grad_norm": 2.470625102204559,
+      "learning_rate": 4.99624544237165e-08,
+      "loss": 0.9918,
+      "step": 1287
+    },
+    {
+      "epoch": 0.9696057212759951,
+      "grad_norm": 2.7251044172160417,
+      "learning_rate": 4.7556893450466656e-08,
+      "loss": 0.9955,
+      "step": 1288
+    },
+    {
+      "epoch": 0.9703585207490355,
+      "grad_norm": 2.662846919070589,
+      "learning_rate": 4.5210542810771864e-08,
+      "loss": 0.9875,
+      "step": 1289
+    },
+    {
+      "epoch": 0.9711113202220758,
+      "grad_norm": 3.1981248551397847,
+      "learning_rate": 4.292341646383813e-08,
+      "loss": 1.0478,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9718641196951162,
+      "grad_norm": 2.738474070013934,
+      "learning_rate": 4.069552801652443e-08,
+      "loss": 0.9877,
+      "step": 1291
+    },
+    {
+      "epoch": 0.9726169191681566,
+      "grad_norm": 2.7898654824264213,
+      "learning_rate": 3.852689072326832e-08,
+      "loss": 0.9641,
+      "step": 1292
+    },
+    {
+      "epoch": 0.9733697186411969,
+      "grad_norm": 2.942924173502617,
+      "learning_rate": 3.641751748600042e-08,
+      "loss": 0.9441,
+      "step": 1293
+    },
+    {
+      "epoch": 0.9741225181142373,
+      "grad_norm": 2.767562137073869,
+      "learning_rate": 3.436742085407119e-08,
+      "loss": 0.9486,
+      "step": 1294
+    },
+    {
+      "epoch": 0.9748753175872776,
+      "grad_norm": 3.186982969064621,
+      "learning_rate": 3.2376613024175384e-08,
+      "loss": 0.9654,
+      "step": 1295
+    },
+    {
+      "epoch": 0.975628117060318,
+      "grad_norm": 2.7143614500651503,
+      "learning_rate": 3.044510584027771e-08,
+      "loss": 0.9494,
+      "step": 1296
+    },
+    {
+      "epoch": 0.9763809165333585,
+      "grad_norm": 4.735419194732088,
+      "learning_rate": 2.8572910793546183e-08,
+      "loss": 0.9178,
+      "step": 1297
+    },
+    {
+      "epoch": 0.9771337160063988,
+      "grad_norm": 2.683092581543925,
+      "learning_rate": 2.676003902227886e-08,
+      "loss": 1.047,
+      "step": 1298
+    },
+    {
+      "epoch": 0.9778865154794392,
+      "grad_norm": 3.1217343085874654,
+      "learning_rate": 2.50065013118439e-08,
+      "loss": 0.8718,
+      "step": 1299
+    },
+    {
+      "epoch": 0.9786393149524796,
+      "grad_norm": 2.3338305591023634,
+      "learning_rate": 2.3312308094607382e-08,
+      "loss": 0.9517,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9793921144255199,
+      "grad_norm": 3.498090322618458,
+      "learning_rate": 2.167746944988114e-08,
+      "loss": 0.9942,
+      "step": 1301
+    },
+    {
+      "epoch": 0.9801449138985603,
+      "grad_norm": 3.1250744843303195,
+      "learning_rate": 2.010199510385058e-08,
+      "loss": 0.9257,
+      "step": 1302
+    },
+    {
+      "epoch": 0.9808977133716006,
+      "grad_norm": 3.0363120214015034,
+      "learning_rate": 1.8585894429528073e-08,
+      "loss": 0.975,
+      "step": 1303
+    },
+    {
+      "epoch": 0.981650512844641,
+      "grad_norm": 2.3790341152687255,
+      "learning_rate": 1.7129176446692986e-08,
+      "loss": 0.9287,
+      "step": 1304
+    },
+    {
+      "epoch": 0.9824033123176814,
+      "grad_norm": 2.7257619356627076,
+      "learning_rate": 1.5731849821833955e-08,
+      "loss": 0.892,
+      "step": 1305
+    },
+    {
+      "epoch": 0.9831561117907217,
+      "grad_norm": 2.36317553749754,
+      "learning_rate": 1.4393922868105591e-08,
+      "loss": 0.9281,
+      "step": 1306
+    },
+    {
+      "epoch": 0.9839089112637621,
+      "grad_norm": 2.3089545627122505,
+      "learning_rate": 1.3115403545270744e-08,
+      "loss": 0.8987,
+      "step": 1307
+    },
+    {
+      "epoch": 0.9846617107368025,
+      "grad_norm": 3.1710248965049357,
+      "learning_rate": 1.1896299459658311e-08,
+      "loss": 0.9173,
+      "step": 1308
+    },
+    {
+      "epoch": 0.9854145102098428,
+      "grad_norm": 2.597055385822197,
+      "learning_rate": 1.0736617864117727e-08,
+      "loss": 0.9617,
+      "step": 1309
+    },
+    {
+      "epoch": 0.9861673096828832,
+      "grad_norm": 3.916236816005277,
+      "learning_rate": 9.636365657971215e-09,
+      "loss": 1.023,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9869201091559235,
+      "grad_norm": 2.688668108409379,
+      "learning_rate": 8.595549386981595e-09,
+      "loss": 0.9199,
+      "step": 1311
+    },
+    {
+      "epoch": 0.987672908628964,
+      "grad_norm": 2.4623215137317254,
+      "learning_rate": 7.614175243301213e-09,
+      "loss": 0.9765,
+      "step": 1312
+    },
+    {
+      "epoch": 0.9884257081020044,
+      "grad_norm": 3.3372772901322794,
+      "learning_rate": 6.692249065447521e-09,
+      "loss": 0.8604,
+      "step": 1313
+    },
+    {
+      "epoch": 0.9891785075750447,
+      "grad_norm": 3.3515115025764595,
+      "learning_rate": 5.8297763382597625e-09,
+      "loss": 1.0037,
+      "step": 1314
+    },
+    {
+      "epoch": 0.9899313070480851,
+      "grad_norm": 2.604424720606584,
+      "learning_rate": 5.026762192870127e-09,
+      "loss": 0.9879,
+      "step": 1315
+    },
+    {
+      "epoch": 0.9906841065211255,
+      "grad_norm": 3.544062299090655,
+      "learning_rate": 4.283211406670429e-09,
+      "loss": 1.0411,
+      "step": 1316
+    },
+    {
+      "epoch": 0.9914369059941658,
+      "grad_norm": 3.267680350517861,
+      "learning_rate": 3.5991284032899087e-09,
+      "loss": 0.976,
+      "step": 1317
+    },
+    {
+      "epoch": 0.9921897054672062,
+      "grad_norm": 3.0886627412078256,
+      "learning_rate": 2.974517252558595e-09,
+      "loss": 0.8727,
+      "step": 1318
+    },
+    {
+      "epoch": 0.9929425049402465,
+      "grad_norm": 2.4720359733436275,
+      "learning_rate": 2.4093816704950924e-09,
+      "loss": 0.9837,
+      "step": 1319
+    },
+    {
+      "epoch": 0.9936953044132869,
+      "grad_norm": 2.5467003523165035,
+      "learning_rate": 1.9037250192732728e-09,
+      "loss": 0.9094,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9944481038863273,
+      "grad_norm": 2.76588303003768,
+      "learning_rate": 1.4575503072100649e-09,
+      "loss": 1.019,
+      "step": 1321
+    },
+    {
+      "epoch": 0.9952009033593676,
+      "grad_norm": 2.894873158632333,
+      "learning_rate": 1.0708601887454706e-09,
+      "loss": 0.9081,
+      "step": 1322
+    },
+    {
+      "epoch": 0.995953702832408,
+      "grad_norm": 2.6572995358290408,
+      "learning_rate": 7.43656964423689e-10,
+      "loss": 0.9697,
+      "step": 1323
+    },
+    {
+      "epoch": 0.9967065023054484,
+      "grad_norm": 3.158217365660322,
+      "learning_rate": 4.759425808853468e-10,
+      "loss": 0.9996,
+      "step": 1324
+    },
+    {
+      "epoch": 0.9974593017784887,
+      "grad_norm": 4.329882531340572,
+      "learning_rate": 2.677186308497337e-10,
+      "loss": 0.9162,
+      "step": 1325
+    },
+    {
+      "epoch": 0.9982121012515291,
+      "grad_norm": 3.0070450446433963,
+      "learning_rate": 1.1898635310925167e-10,
+      "loss": 0.9543,
+      "step": 1326
+    },
+    {
+      "epoch": 0.9989649007245694,
+      "grad_norm": 2.7526840917387676,
+      "learning_rate": 2.9746632520533116e-11,
+      "loss": 0.9748,
+      "step": 1327
+    },
+    {
+      "epoch": 0.9997177001976099,
+      "grad_norm": 2.4986435103048477,
+      "learning_rate": 0.0,
+      "loss": 0.9028,
+      "step": 1328
+    },
+    {
+      "epoch": 0.9997177001976099,
+      "step": 1328,
+      "total_flos": 1676734831263744.0,
+      "train_loss": 1.0399419161300343,
+      "train_runtime": 48886.61,
+      "train_samples_per_second": 6.956,
+      "train_steps_per_second": 0.027
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 1328,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 133,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1676734831263744.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}