diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,6042 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 106,
+  "global_step": 848,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0023584905660377358,
+      "grad_norm": 5.1080776865120034,
+      "learning_rate": 0.0,
+      "loss": 0.9592,
+      "step": 1
+    },
+    {
+      "epoch": 0.0023584905660377358,
+      "eval_loss": 0.9177566170692444,
+      "eval_runtime": 83.3901,
+      "eval_samples_per_second": 8.454,
+      "eval_steps_per_second": 0.54,
+      "step": 1
+    },
+    {
+      "epoch": 0.0047169811320754715,
+      "grad_norm": 5.216656877082355,
+      "learning_rate": 4e-08,
+      "loss": 0.9643,
+      "step": 2
+    },
+    {
+      "epoch": 0.007075471698113208,
+      "grad_norm": 5.86435572061478,
+      "learning_rate": 8e-08,
+      "loss": 0.9448,
+      "step": 3
+    },
+    {
+      "epoch": 0.009433962264150943,
+      "grad_norm": 4.885606218308886,
+      "learning_rate": 1.2e-07,
+      "loss": 0.9499,
+      "step": 4
+    },
+    {
+      "epoch": 0.01179245283018868,
+      "grad_norm": 4.863782725318177,
+      "learning_rate": 1.6e-07,
+      "loss": 0.9547,
+      "step": 5
+    },
+    {
+      "epoch": 0.014150943396226415,
+      "grad_norm": 4.84448661904324,
+      "learning_rate": 2e-07,
+      "loss": 0.9378,
+      "step": 6
+    },
+    {
+      "epoch": 0.01650943396226415,
+      "grad_norm": 4.594764533947918,
+      "learning_rate": 2.4e-07,
+      "loss": 0.967,
+      "step": 7
+    },
+    {
+      "epoch": 0.018867924528301886,
+      "grad_norm": 5.311507883731841,
+      "learning_rate": 2.8e-07,
+      "loss": 0.9534,
+      "step": 8
+    },
+    {
+      "epoch": 0.02122641509433962,
+      "grad_norm": 4.780331285112587,
+      "learning_rate": 3.2e-07,
+      "loss": 0.979,
+      "step": 9
+    },
+    {
+      "epoch": 0.02358490566037736,
+      "grad_norm": 5.172814213529754,
+      "learning_rate": 3.6e-07,
+      "loss": 0.9284,
+      "step": 10
+    },
+    {
+      "epoch": 0.025943396226415096,
+      "grad_norm": 4.693805464386623,
+      "learning_rate": 4e-07,
+      "loss": 0.9452,
+      "step": 11
+    },
+    {
+      "epoch": 0.02830188679245283,
+      "grad_norm": 5.378045080991313,
+      "learning_rate": 4.3999999999999997e-07,
+      "loss": 0.9575,
+      "step": 12
+    },
+    {
+      "epoch": 0.030660377358490566,
+      "grad_norm": 4.718804488820069,
+      "learning_rate": 4.8e-07,
+      "loss": 0.9601,
+      "step": 13
+    },
+    {
+      "epoch": 0.0330188679245283,
+      "grad_norm": 4.8688696728838785,
+      "learning_rate": 5.2e-07,
+      "loss": 0.9578,
+      "step": 14
+    },
+    {
+      "epoch": 0.03537735849056604,
+      "grad_norm": 4.41639309685216,
+      "learning_rate": 5.6e-07,
+      "loss": 0.9433,
+      "step": 15
+    },
+    {
+      "epoch": 0.03773584905660377,
+      "grad_norm": 4.330783791521376,
+      "learning_rate": 6e-07,
+      "loss": 0.9498,
+      "step": 16
+    },
+    {
+      "epoch": 0.04009433962264151,
+      "grad_norm": 4.580748421983631,
+      "learning_rate": 6.4e-07,
+      "loss": 0.9607,
+      "step": 17
+    },
+    {
+      "epoch": 0.04245283018867924,
+      "grad_norm": 3.895307917106911,
+      "learning_rate": 6.800000000000001e-07,
+      "loss": 0.9561,
+      "step": 18
+    },
+    {
+      "epoch": 0.04481132075471698,
+      "grad_norm": 3.9063376740559894,
+      "learning_rate": 7.2e-07,
+      "loss": 0.9423,
+      "step": 19
+    },
+    {
+      "epoch": 0.04716981132075472,
+      "grad_norm": 3.9785213934452335,
+      "learning_rate": 7.599999999999999e-07,
+      "loss": 0.9422,
+      "step": 20
+    },
+    {
+      "epoch": 0.049528301886792456,
+      "grad_norm": 3.5176355010251044,
+      "learning_rate": 8e-07,
+      "loss": 0.9289,
+      "step": 21
+    },
+    {
+      "epoch": 0.05188679245283019,
+      "grad_norm": 5.022294330418908,
+      "learning_rate": 8.399999999999999e-07,
+      "loss": 0.9446,
+      "step": 22
+    },
+    {
+      "epoch": 0.054245283018867926,
+      "grad_norm": 2.6661772621278637,
+      "learning_rate": 8.799999999999999e-07,
+      "loss": 0.9396,
+      "step": 23
+    },
+    {
+      "epoch": 0.05660377358490566,
+      "grad_norm": 2.8934776201699757,
+      "learning_rate": 9.2e-07,
+      "loss": 0.9424,
+      "step": 24
+    },
+    {
+      "epoch": 0.0589622641509434,
+      "grad_norm": 3.310951896187245,
+      "learning_rate": 9.6e-07,
+      "loss": 0.9317,
+      "step": 25
+    },
+    {
+      "epoch": 0.06132075471698113,
+      "grad_norm": 3.180088649692653,
+      "learning_rate": 1e-06,
+      "loss": 0.8733,
+      "step": 26
+    },
+    {
+      "epoch": 0.06367924528301887,
+      "grad_norm": 3.0819569242393032,
+      "learning_rate": 1.04e-06,
+      "loss": 0.9422,
+      "step": 27
+    },
+    {
+      "epoch": 0.0660377358490566,
+      "grad_norm": 4.5985187060206405,
+      "learning_rate": 1.08e-06,
+      "loss": 0.9522,
+      "step": 28
+    },
+    {
+      "epoch": 0.06839622641509434,
+      "grad_norm": 2.614037871400199,
+      "learning_rate": 1.12e-06,
+      "loss": 0.9168,
+      "step": 29
+    },
+    {
+      "epoch": 0.07075471698113207,
+      "grad_norm": 3.4637889884655007,
+      "learning_rate": 1.16e-06,
+      "loss": 0.8819,
+      "step": 30
+    },
+    {
+      "epoch": 0.07311320754716981,
+      "grad_norm": 2.7607684537358623,
+      "learning_rate": 1.2e-06,
+      "loss": 0.9443,
+      "step": 31
+    },
+    {
+      "epoch": 0.07547169811320754,
+      "grad_norm": 2.45835136955183,
+      "learning_rate": 1.24e-06,
+      "loss": 0.8971,
+      "step": 32
+    },
+    {
+      "epoch": 0.07783018867924528,
+      "grad_norm": 2.8319894880867724,
+      "learning_rate": 1.28e-06,
+      "loss": 0.9311,
+      "step": 33
+    },
+    {
+      "epoch": 0.08018867924528301,
+      "grad_norm": 1.880755586715356,
+      "learning_rate": 1.32e-06,
+      "loss": 0.895,
+      "step": 34
+    },
+    {
+      "epoch": 0.08254716981132075,
+      "grad_norm": 3.1062915440189713,
+      "learning_rate": 1.3600000000000001e-06,
+      "loss": 0.9074,
+      "step": 35
+    },
+    {
+      "epoch": 0.08490566037735849,
+      "grad_norm": 2.745579181683214,
+      "learning_rate": 1.4e-06,
+      "loss": 0.9066,
+      "step": 36
+    },
+    {
+      "epoch": 0.08726415094339622,
+      "grad_norm": 2.51965653055992,
+      "learning_rate": 1.44e-06,
+      "loss": 0.8737,
+      "step": 37
+    },
+    {
+      "epoch": 0.08962264150943396,
+      "grad_norm": 2.324394856248302,
+      "learning_rate": 1.48e-06,
+      "loss": 0.9239,
+      "step": 38
+    },
+    {
+      "epoch": 0.09198113207547169,
+      "grad_norm": 2.7275453116030834,
+      "learning_rate": 1.5199999999999998e-06,
+      "loss": 0.9029,
+      "step": 39
+    },
+    {
+      "epoch": 0.09433962264150944,
+      "grad_norm": 2.8394394816071498,
+      "learning_rate": 1.5599999999999999e-06,
+      "loss": 0.885,
+      "step": 40
+    },
+    {
+      "epoch": 0.09669811320754718,
+      "grad_norm": 2.3797648877001323,
+      "learning_rate": 1.6e-06,
+      "loss": 0.9084,
+      "step": 41
+    },
+    {
+      "epoch": 0.09905660377358491,
+      "grad_norm": 2.298049507841523,
+      "learning_rate": 1.6399999999999998e-06,
+      "loss": 0.8786,
+      "step": 42
+    },
+    {
+      "epoch": 0.10141509433962265,
+      "grad_norm": 2.0882957626327814,
+      "learning_rate": 1.6799999999999998e-06,
+      "loss": 0.885,
+      "step": 43
+    },
+    {
+      "epoch": 0.10377358490566038,
+      "grad_norm": 1.9805393143682433,
+      "learning_rate": 1.7199999999999998e-06,
+      "loss": 0.9202,
+      "step": 44
+    },
+    {
+      "epoch": 0.10613207547169812,
+      "grad_norm": 2.1136265646711605,
+      "learning_rate": 1.7599999999999999e-06,
+      "loss": 0.8515,
+      "step": 45
+    },
+    {
+      "epoch": 0.10849056603773585,
+      "grad_norm": 1.7161578870903316,
+      "learning_rate": 1.8e-06,
+      "loss": 0.9035,
+      "step": 46
+    },
+    {
+      "epoch": 0.11084905660377359,
+      "grad_norm": 2.2352887975077795,
+      "learning_rate": 1.84e-06,
+      "loss": 0.8615,
+      "step": 47
+    },
+    {
+      "epoch": 0.11320754716981132,
+      "grad_norm": 1.997343145683379,
+      "learning_rate": 1.8799999999999998e-06,
+      "loss": 0.8958,
+      "step": 48
+    },
+    {
+      "epoch": 0.11556603773584906,
+      "grad_norm": 2.1520613046888024,
+      "learning_rate": 1.92e-06,
+      "loss": 0.8706,
+      "step": 49
+    },
+    {
+      "epoch": 0.1179245283018868,
+      "grad_norm": 1.610078026570572,
+      "learning_rate": 1.96e-06,
+      "loss": 0.9127,
+      "step": 50
+    },
+    {
+      "epoch": 0.12028301886792453,
+      "grad_norm": 1.6267119325099246,
+      "learning_rate": 2e-06,
+      "loss": 0.8604,
+      "step": 51
+    },
+    {
+      "epoch": 0.12264150943396226,
+      "grad_norm": 1.7305415242899538,
+      "learning_rate": 1.99999225068337e-06,
+      "loss": 0.8704,
+      "step": 52
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 4.837245274184259,
+      "learning_rate": 1.9999690028535855e-06,
+      "loss": 0.9003,
+      "step": 53
+    },
+    {
+      "epoch": 0.12735849056603774,
+      "grad_norm": 2.006265922883068,
+      "learning_rate": 1.9999302568709546e-06,
+      "loss": 0.8759,
+      "step": 54
+    },
+    {
+      "epoch": 0.12971698113207547,
+      "grad_norm": 1.8749359565723283,
+      "learning_rate": 1.999876013335988e-06,
+      "loss": 0.8686,
+      "step": 55
+    },
+    {
+      "epoch": 0.1320754716981132,
+      "grad_norm": 1.5041642645696787,
+      "learning_rate": 1.9998062730893862e-06,
+      "loss": 0.8727,
+      "step": 56
+    },
+    {
+      "epoch": 0.13443396226415094,
+      "grad_norm": 1.4630749706626824,
+      "learning_rate": 1.9997210372120272e-06,
+      "loss": 0.8718,
+      "step": 57
+    },
+    {
+      "epoch": 0.13679245283018868,
+      "grad_norm": 1.4357346298179297,
+      "learning_rate": 1.9996203070249514e-06,
+      "loss": 0.9,
+      "step": 58
+    },
+    {
+      "epoch": 0.1391509433962264,
+      "grad_norm": 1.589349788137753,
+      "learning_rate": 1.9995040840893383e-06,
+      "loss": 0.8832,
+      "step": 59
+    },
+    {
+      "epoch": 0.14150943396226415,
+      "grad_norm": 1.3078655659788383,
+      "learning_rate": 1.9993723702064853e-06,
+      "loss": 0.8689,
+      "step": 60
+    },
+    {
+      "epoch": 0.14386792452830188,
+      "grad_norm": 1.2162726319357984,
+      "learning_rate": 1.9992251674177763e-06,
+      "loss": 0.8565,
+      "step": 61
+    },
+    {
+      "epoch": 0.14622641509433962,
+      "grad_norm": 1.4682400986060873,
+      "learning_rate": 1.999062478004655e-06,
+      "loss": 0.8768,
+      "step": 62
+    },
+    {
+      "epoch": 0.14858490566037735,
+      "grad_norm": 1.475835127346073,
+      "learning_rate": 1.9988843044885837e-06,
+      "loss": 0.8485,
+      "step": 63
+    },
+    {
+      "epoch": 0.1509433962264151,
+      "grad_norm": 1.313661603517591,
+      "learning_rate": 1.998690649631009e-06,
+      "loss": 0.8791,
+      "step": 64
+    },
+    {
+      "epoch": 0.15330188679245282,
+      "grad_norm": 1.2318794472546806,
+      "learning_rate": 1.998481516433316e-06,
+      "loss": 0.8418,
+      "step": 65
+    },
+    {
+      "epoch": 0.15566037735849056,
+      "grad_norm": 1.224129101101457,
+      "learning_rate": 1.9982569081367843e-06,
+      "loss": 0.8472,
+      "step": 66
+    },
+    {
+      "epoch": 0.1580188679245283,
+      "grad_norm": 1.4354409800877264,
+      "learning_rate": 1.9980168282225344e-06,
+      "loss": 0.8682,
+      "step": 67
+    },
+    {
+      "epoch": 0.16037735849056603,
+      "grad_norm": 1.3882277723705556,
+      "learning_rate": 1.9977612804114775e-06,
+      "loss": 0.8929,
+      "step": 68
+    },
+    {
+      "epoch": 0.16273584905660377,
+      "grad_norm": 1.1353588967534465,
+      "learning_rate": 1.9974902686642557e-06,
+      "loss": 0.8123,
+      "step": 69
+    },
+    {
+      "epoch": 0.1650943396226415,
+      "grad_norm": 1.1680380889234738,
+      "learning_rate": 1.9972037971811797e-06,
+      "loss": 0.8573,
+      "step": 70
+    },
+    {
+      "epoch": 0.16745283018867924,
+      "grad_norm": 1.1797181726680495,
+      "learning_rate": 1.9969018704021675e-06,
+      "loss": 0.8518,
+      "step": 71
+    },
+    {
+      "epoch": 0.16981132075471697,
+      "grad_norm": 1.3076313839386762,
+      "learning_rate": 1.9965844930066696e-06,
+      "loss": 0.8425,
+      "step": 72
+    },
+    {
+      "epoch": 0.1721698113207547,
+      "grad_norm": 1.2192501506435032,
+      "learning_rate": 1.9962516699136036e-06,
+      "loss": 0.8633,
+      "step": 73
+    },
+    {
+      "epoch": 0.17452830188679244,
+      "grad_norm": 1.1885172994924504,
+      "learning_rate": 1.9959034062812714e-06,
+      "loss": 0.8417,
+      "step": 74
+    },
+    {
+      "epoch": 0.17688679245283018,
+      "grad_norm": 1.0228274366877697,
+      "learning_rate": 1.9955397075072833e-06,
+      "loss": 0.8645,
+      "step": 75
+    },
+    {
+      "epoch": 0.1792452830188679,
+      "grad_norm": 1.0362433430443114,
+      "learning_rate": 1.9951605792284742e-06,
+      "loss": 0.8442,
+      "step": 76
+    },
+    {
+      "epoch": 0.18160377358490565,
+      "grad_norm": 1.3774246732359812,
+      "learning_rate": 1.9947660273208134e-06,
+      "loss": 0.8355,
+      "step": 77
+    },
+    {
+      "epoch": 0.18396226415094338,
+      "grad_norm": 1.3163064309206076,
+      "learning_rate": 1.9943560578993165e-06,
+      "loss": 0.8246,
+      "step": 78
+    },
+    {
+      "epoch": 0.18632075471698112,
+      "grad_norm": 1.0822503373501942,
+      "learning_rate": 1.9939306773179494e-06,
+      "loss": 0.8315,
+      "step": 79
+    },
+    {
+      "epoch": 0.18867924528301888,
+      "grad_norm": 1.0361871014520703,
+      "learning_rate": 1.9934898921695292e-06,
+      "loss": 0.8346,
+      "step": 80
+    },
+    {
+      "epoch": 0.19103773584905662,
+      "grad_norm": 1.0371978245512512,
+      "learning_rate": 1.993033709285624e-06,
+      "loss": 0.8706,
+      "step": 81
+    },
+    {
+      "epoch": 0.19339622641509435,
+      "grad_norm": 1.0574050613888015,
+      "learning_rate": 1.992562135736444e-06,
+      "loss": 0.8308,
+      "step": 82
+    },
+    {
+      "epoch": 0.1957547169811321,
+      "grad_norm": 1.241296877885062,
+      "learning_rate": 1.992075178830736e-06,
+      "loss": 0.872,
+      "step": 83
+    },
+    {
+      "epoch": 0.19811320754716982,
+      "grad_norm": 1.152024986745144,
+      "learning_rate": 1.9915728461156654e-06,
+      "loss": 0.8365,
+      "step": 84
+    },
+    {
+      "epoch": 0.20047169811320756,
+      "grad_norm": 1.0972286027771576,
+      "learning_rate": 1.991055145376703e-06,
+      "loss": 0.8511,
+      "step": 85
+    },
+    {
+      "epoch": 0.2028301886792453,
+      "grad_norm": 1.3363467615851807,
+      "learning_rate": 1.990522084637503e-06,
+      "loss": 0.8604,
+      "step": 86
+    },
+    {
+      "epoch": 0.20518867924528303,
+      "grad_norm": 0.9249543605316475,
+      "learning_rate": 1.9899736721597786e-06,
+      "loss": 0.8078,
+      "step": 87
+    },
+    {
+      "epoch": 0.20754716981132076,
+      "grad_norm": 1.028197337537026,
+      "learning_rate": 1.9894099164431722e-06,
+      "loss": 0.8572,
+      "step": 88
+    },
+    {
+      "epoch": 0.2099056603773585,
+      "grad_norm": 0.9345635713942412,
+      "learning_rate": 1.9888308262251284e-06,
+      "loss": 0.814,
+      "step": 89
+    },
+    {
+      "epoch": 0.21226415094339623,
+      "grad_norm": 1.0476770833253457,
+      "learning_rate": 1.9882364104807535e-06,
+      "loss": 0.8358,
+      "step": 90
+    },
+    {
+      "epoch": 0.21462264150943397,
+      "grad_norm": 1.0517946592201646,
+      "learning_rate": 1.9876266784226794e-06,
+      "loss": 0.8263,
+      "step": 91
+    },
+    {
+      "epoch": 0.2169811320754717,
+      "grad_norm": 1.1715532913432805,
+      "learning_rate": 1.987001639500919e-06,
+      "loss": 0.8268,
+      "step": 92
+    },
+    {
+      "epoch": 0.21933962264150944,
+      "grad_norm": 0.928474382958498,
+      "learning_rate": 1.9863613034027223e-06,
+      "loss": 0.8278,
+      "step": 93
+    },
+    {
+      "epoch": 0.22169811320754718,
+      "grad_norm": 0.9243352298229107,
+      "learning_rate": 1.985705680052423e-06,
+      "loss": 0.817,
+      "step": 94
+    },
+    {
+      "epoch": 0.2240566037735849,
+      "grad_norm": 0.9825897380291061,
+      "learning_rate": 1.985034779611287e-06,
+      "loss": 0.8185,
+      "step": 95
+    },
+    {
+      "epoch": 0.22641509433962265,
+      "grad_norm": 1.0129581531905947,
+      "learning_rate": 1.9843486124773543e-06,
+      "loss": 0.8261,
+      "step": 96
+    },
+    {
+      "epoch": 0.22877358490566038,
+      "grad_norm": 0.9634164693430555,
+      "learning_rate": 1.9836471892852777e-06,
+      "loss": 0.8448,
+      "step": 97
+    },
+    {
+      "epoch": 0.23113207547169812,
+      "grad_norm": 0.9208269074792377,
+      "learning_rate": 1.982930520906158e-06,
+      "loss": 0.8435,
+      "step": 98
+    },
+    {
+      "epoch": 0.23349056603773585,
+      "grad_norm": 1.2744902927155426,
+      "learning_rate": 1.9821986184473754e-06,
+      "loss": 0.811,
+      "step": 99
+    },
+    {
+      "epoch": 0.2358490566037736,
+      "grad_norm": 0.9132891996164993,
+      "learning_rate": 1.9814514932524176e-06,
+      "loss": 0.8629,
+      "step": 100
+    },
+    {
+      "epoch": 0.23820754716981132,
+      "grad_norm": 1.3924445022644105,
+      "learning_rate": 1.9806891569007048e-06,
+      "loss": 0.8157,
+      "step": 101
+    },
+    {
+      "epoch": 0.24056603773584906,
+      "grad_norm": 0.9910775720488755,
+      "learning_rate": 1.9799116212074075e-06,
+      "loss": 0.8133,
+      "step": 102
+    },
+    {
+      "epoch": 0.2429245283018868,
+      "grad_norm": 0.8521457201237292,
+      "learning_rate": 1.979118898223267e-06,
+      "loss": 0.818,
+      "step": 103
+    },
+    {
+      "epoch": 0.24528301886792453,
+      "grad_norm": 0.961507205526783,
+      "learning_rate": 1.978311000234406e-06,
+      "loss": 0.8312,
+      "step": 104
+    },
+    {
+      "epoch": 0.24764150943396226,
+      "grad_norm": 0.9234203353918131,
+      "learning_rate": 1.9774879397621383e-06,
+      "loss": 0.8307,
+      "step": 105
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.939970736839533,
+      "learning_rate": 1.9766497295627777e-06,
+      "loss": 0.8121,
+      "step": 106
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 0.7522591948509216,
+      "eval_runtime": 82.8066,
+      "eval_samples_per_second": 8.514,
+      "eval_steps_per_second": 0.543,
+      "step": 106
+    },
+    {
+      "epoch": 0.25235849056603776,
+      "grad_norm": 1.009636311779919,
+      "learning_rate": 1.9757963826274354e-06,
+      "loss": 0.8321,
+      "step": 107
+    },
+    {
+      "epoch": 0.25471698113207547,
+      "grad_norm": 0.8979616538385209,
+      "learning_rate": 1.9749279121818236e-06,
+      "loss": 0.8442,
+      "step": 108
+    },
+    {
+      "epoch": 0.25707547169811323,
+      "grad_norm": 0.953771354540708,
+      "learning_rate": 1.9740443316860463e-06,
+      "loss": 0.8484,
+      "step": 109
+    },
+    {
+      "epoch": 0.25943396226415094,
+      "grad_norm": 1.0957061062800166,
+      "learning_rate": 1.9731456548343944e-06,
+      "loss": 0.8204,
+      "step": 110
+    },
+    {
+      "epoch": 0.2617924528301887,
+      "grad_norm": 0.8834656211178301,
+      "learning_rate": 1.9722318955551303e-06,
+      "loss": 0.7817,
+      "step": 111
+    },
+    {
+      "epoch": 0.2641509433962264,
+      "grad_norm": 1.0070698705880765,
+      "learning_rate": 1.9713030680102743e-06,
+      "loss": 0.8309,
+      "step": 112
+    },
+    {
+      "epoch": 0.2665094339622642,
+      "grad_norm": 1.0103166521916906,
+      "learning_rate": 1.970359186595384e-06,
+      "loss": 0.8454,
+      "step": 113
+    },
+    {
+      "epoch": 0.2688679245283019,
+      "grad_norm": 0.7965830780073067,
+      "learning_rate": 1.9694002659393305e-06,
+      "loss": 0.7659,
+      "step": 114
+    },
+    {
+      "epoch": 0.27122641509433965,
+      "grad_norm": 0.9676729525657803,
+      "learning_rate": 1.968426320904074e-06,
+      "loss": 0.8076,
+      "step": 115
+    },
+    {
+      "epoch": 0.27358490566037735,
+      "grad_norm": 0.8547052227174852,
+      "learning_rate": 1.967437366584431e-06,
+      "loss": 0.8305,
+      "step": 116
+    },
+    {
+      "epoch": 0.2759433962264151,
+      "grad_norm": 0.9129048151834617,
+      "learning_rate": 1.9664334183078425e-06,
+      "loss": 0.8443,
+      "step": 117
+    },
+    {
+      "epoch": 0.2783018867924528,
+      "grad_norm": 0.8521299847115695,
+      "learning_rate": 1.965414491634134e-06,
+      "loss": 0.8244,
+      "step": 118
+    },
+    {
+      "epoch": 0.2806603773584906,
+      "grad_norm": 0.8875033746133125,
+      "learning_rate": 1.964380602355277e-06,
+      "loss": 0.7855,
+      "step": 119
+    },
+    {
+      "epoch": 0.2830188679245283,
+      "grad_norm": 0.8322961791869938,
+      "learning_rate": 1.9633317664951417e-06,
+      "loss": 0.8246,
+      "step": 120
+    },
+    {
+      "epoch": 0.28537735849056606,
+      "grad_norm": 0.8996547969201056,
+      "learning_rate": 1.9622680003092503e-06,
+      "loss": 0.7911,
+      "step": 121
+    },
+    {
+      "epoch": 0.28773584905660377,
+      "grad_norm": 0.8421734340822833,
+      "learning_rate": 1.9611893202845253e-06,
+      "loss": 0.8075,
+      "step": 122
+    },
+    {
+      "epoch": 0.29009433962264153,
+      "grad_norm": 0.9715000278619699,
+      "learning_rate": 1.9600957431390324e-06,
+      "loss": 0.8258,
+      "step": 123
+    },
+    {
+      "epoch": 0.29245283018867924,
+      "grad_norm": 0.9486047429972293,
+      "learning_rate": 1.9589872858217233e-06,
+      "loss": 0.8249,
+      "step": 124
+    },
+    {
+      "epoch": 0.294811320754717,
+      "grad_norm": 0.8708681162275084,
+      "learning_rate": 1.9578639655121707e-06,
+      "loss": 0.8061,
+      "step": 125
+    },
+    {
+      "epoch": 0.2971698113207547,
+      "grad_norm": 0.815600627546263,
+      "learning_rate": 1.9567257996203046e-06,
+      "loss": 0.8117,
+      "step": 126
+    },
+    {
+      "epoch": 0.29952830188679247,
+      "grad_norm": 0.899644949674199,
+      "learning_rate": 1.955572805786141e-06,
+      "loss": 0.8324,
+      "step": 127
+    },
+    {
+      "epoch": 0.3018867924528302,
+      "grad_norm": 0.7797228121513826,
+      "learning_rate": 1.9544050018795075e-06,
+      "loss": 0.8085,
+      "step": 128
+    },
+    {
+      "epoch": 0.30424528301886794,
+      "grad_norm": 1.4425571525030387,
+      "learning_rate": 1.953222405999769e-06,
+      "loss": 0.8043,
+      "step": 129
+    },
+    {
+      "epoch": 0.30660377358490565,
+      "grad_norm": 1.3224705998967385,
+      "learning_rate": 1.9520250364755458e-06,
+      "loss": 0.8104,
+      "step": 130
+    },
+    {
+      "epoch": 0.3089622641509434,
+      "grad_norm": 0.8148761244743041,
+      "learning_rate": 1.9508129118644293e-06,
+      "loss": 0.828,
+      "step": 131
+    },
+    {
+      "epoch": 0.3113207547169811,
+      "grad_norm": 1.0994942919129091,
+      "learning_rate": 1.949586050952693e-06,
+      "loss": 0.8007,
+      "step": 132
+    },
+    {
+      "epoch": 0.3136792452830189,
+      "grad_norm": 0.9838294161632027,
+      "learning_rate": 1.9483444727550054e-06,
+      "loss": 0.8304,
+      "step": 133
+    },
+    {
+      "epoch": 0.3160377358490566,
+      "grad_norm": 1.0206044534349468,
+      "learning_rate": 1.9470881965141307e-06,
+      "loss": 0.8044,
+      "step": 134
+    },
+    {
+      "epoch": 0.31839622641509435,
+      "grad_norm": 0.8186471659261382,
+      "learning_rate": 1.9458172417006346e-06,
+      "loss": 0.7921,
+      "step": 135
+    },
+    {
+      "epoch": 0.32075471698113206,
+      "grad_norm": 1.040687215576322,
+      "learning_rate": 1.944531628012579e-06,
+      "loss": 0.8309,
+      "step": 136
+    },
+    {
+      "epoch": 0.3231132075471698,
+      "grad_norm": 0.8067576806018053,
+      "learning_rate": 1.9432313753752194e-06,
+      "loss": 0.8006,
+      "step": 137
+    },
+    {
+      "epoch": 0.32547169811320753,
+      "grad_norm": 1.1895556658600497,
+      "learning_rate": 1.941916503940694e-06,
+      "loss": 0.8055,
+      "step": 138
+    },
+    {
+      "epoch": 0.3278301886792453,
+      "grad_norm": 0.877104803725933,
+      "learning_rate": 1.9405870340877135e-06,
+      "loss": 0.815,
+      "step": 139
+    },
+    {
+      "epoch": 0.330188679245283,
+      "grad_norm": 1.0755873765351904,
+      "learning_rate": 1.9392429864212433e-06,
+      "loss": 0.8357,
+      "step": 140
+    },
+    {
+      "epoch": 0.33254716981132076,
+      "grad_norm": 0.8389051921712227,
+      "learning_rate": 1.9378843817721854e-06,
+      "loss": 0.8054,
+      "step": 141
+    },
+    {
+      "epoch": 0.33490566037735847,
+      "grad_norm": 0.8492954330518299,
+      "learning_rate": 1.9365112411970546e-06,
+      "loss": 0.8125,
+      "step": 142
+    },
+    {
+      "epoch": 0.33726415094339623,
+      "grad_norm": 0.8117108742521268,
+      "learning_rate": 1.9351235859776537e-06,
+      "loss": 0.7996,
+      "step": 143
+    },
+    {
+      "epoch": 0.33962264150943394,
+      "grad_norm": 0.8734747925195555,
+      "learning_rate": 1.9337214376207417e-06,
+      "loss": 0.7987,
+      "step": 144
+    },
+    {
+      "epoch": 0.3419811320754717,
+      "grad_norm": 0.870843122026526,
+      "learning_rate": 1.932304817857702e-06,
+      "loss": 0.8071,
+      "step": 145
+    },
+    {
+      "epoch": 0.3443396226415094,
+      "grad_norm": 0.7823387766729867,
+      "learning_rate": 1.930873748644204e-06,
+      "loss": 0.781,
+      "step": 146
+    },
+    {
+      "epoch": 0.3466981132075472,
+      "grad_norm": 1.0441516146481582,
+      "learning_rate": 1.9294282521598657e-06,
+      "loss": 0.8211,
+      "step": 147
+    },
+    {
+      "epoch": 0.3490566037735849,
+      "grad_norm": 0.8689106447919821,
+      "learning_rate": 1.927968350807906e-06,
+      "loss": 0.7827,
+      "step": 148
+    },
+    {
+      "epoch": 0.35141509433962265,
+      "grad_norm": 0.8369054883974562,
+      "learning_rate": 1.9264940672148015e-06,
+      "loss": 0.7987,
+      "step": 149
+    },
+    {
+      "epoch": 0.35377358490566035,
+      "grad_norm": 1.628472390625535,
+      "learning_rate": 1.9250054242299326e-06,
+      "loss": 0.8245,
+      "step": 150
+    },
+    {
+      "epoch": 0.3561320754716981,
+      "grad_norm": 0.8210347316488275,
+      "learning_rate": 1.9235024449252305e-06,
+      "loss": 0.8026,
+      "step": 151
+    },
+    {
+      "epoch": 0.3584905660377358,
+      "grad_norm": 0.8976272281891767,
+      "learning_rate": 1.9219851525948203e-06,
+      "loss": 0.8343,
+      "step": 152
+    },
+    {
+      "epoch": 0.3608490566037736,
+      "grad_norm": 0.8688278315807118,
+      "learning_rate": 1.92045357075466e-06,
+      "loss": 0.8378,
+      "step": 153
+    },
+    {
+      "epoch": 0.3632075471698113,
+      "grad_norm": 0.8259924861279039,
+      "learning_rate": 1.9189077231421746e-06,
+      "loss": 0.771,
+      "step": 154
+    },
+    {
+      "epoch": 0.36556603773584906,
+      "grad_norm": 0.850617328037459,
+      "learning_rate": 1.917347633715889e-06,
+      "loss": 0.8116,
+      "step": 155
+    },
+    {
+      "epoch": 0.36792452830188677,
+      "grad_norm": 0.830159427983018,
+      "learning_rate": 1.915773326655057e-06,
+      "loss": 0.8142,
+      "step": 156
+    },
+    {
+      "epoch": 0.37028301886792453,
+      "grad_norm": 0.9047033822843099,
+      "learning_rate": 1.9141848263592873e-06,
+      "loss": 0.7867,
+      "step": 157
+    },
+    {
+      "epoch": 0.37264150943396224,
+      "grad_norm": 0.8043279871245154,
+      "learning_rate": 1.9125821574481623e-06,
+      "loss": 0.7821,
+      "step": 158
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 0.9855797219219302,
+      "learning_rate": 1.9109653447608605e-06,
+      "loss": 0.838,
+      "step": 159
+    },
+    {
+      "epoch": 0.37735849056603776,
+      "grad_norm": 0.8671907333946726,
+      "learning_rate": 1.909334413355768e-06,
+      "loss": 0.8199,
+      "step": 160
+    },
+    {
+      "epoch": 0.37971698113207547,
+      "grad_norm": 0.8038519157610383,
+      "learning_rate": 1.9076893885100934e-06,
+      "loss": 0.7914,
+      "step": 161
+    },
+    {
+      "epoch": 0.38207547169811323,
+      "grad_norm": 0.8534138715990678,
+      "learning_rate": 1.906030295719473e-06,
+      "loss": 0.803,
+      "step": 162
+    },
+    {
+      "epoch": 0.38443396226415094,
+      "grad_norm": 1.272101513607124,
+      "learning_rate": 1.9043571606975775e-06,
+      "loss": 0.8243,
+      "step": 163
+    },
+    {
+      "epoch": 0.3867924528301887,
+      "grad_norm": 0.8724688206578753,
+      "learning_rate": 1.9026700093757129e-06,
+      "loss": 0.8,
+      "step": 164
+    },
+    {
+      "epoch": 0.3891509433962264,
+      "grad_norm": 0.8272714642025372,
+      "learning_rate": 1.9009688679024189e-06,
+      "loss": 0.7941,
+      "step": 165
+    },
+    {
+      "epoch": 0.3915094339622642,
+      "grad_norm": 1.609360832798017,
+      "learning_rate": 1.8992537626430636e-06,
+      "loss": 0.8184,
+      "step": 166
+    },
+    {
+      "epoch": 0.3938679245283019,
+      "grad_norm": 1.068913128886596,
+      "learning_rate": 1.897524720179434e-06,
+      "loss": 0.7551,
+      "step": 167
+    },
+    {
+      "epoch": 0.39622641509433965,
+      "grad_norm": 0.876062488501413,
+      "learning_rate": 1.8957817673093256e-06,
+      "loss": 0.8202,
+      "step": 168
+    },
+    {
+      "epoch": 0.39858490566037735,
+      "grad_norm": 0.7804659272450435,
+      "learning_rate": 1.894024931046125e-06,
+      "loss": 0.8096,
+      "step": 169
+    },
+    {
+      "epoch": 0.4009433962264151,
+      "grad_norm": 1.1020958270083518,
+      "learning_rate": 1.8922542386183939e-06,
+      "loss": 0.8023,
+      "step": 170
+    },
+    {
+      "epoch": 0.4033018867924528,
+      "grad_norm": 0.8307091625451186,
+      "learning_rate": 1.8904697174694446e-06,
+      "loss": 0.797,
+      "step": 171
+    },
+    {
+      "epoch": 0.4056603773584906,
+      "grad_norm": 1.4660516537985,
+      "learning_rate": 1.8886713952569156e-06,
+      "loss": 0.8001,
+      "step": 172
+    },
+    {
+      "epoch": 0.4080188679245283,
+      "grad_norm": 0.9042037911584216,
+      "learning_rate": 1.8868592998523436e-06,
+      "loss": 0.7798,
+      "step": 173
+    },
+    {
+      "epoch": 0.41037735849056606,
+      "grad_norm": 0.8127436933722597,
+      "learning_rate": 1.885033459340731e-06,
+      "loss": 0.791,
+      "step": 174
+    },
+    {
+      "epoch": 0.41273584905660377,
+      "grad_norm": 0.8906980453056944,
+      "learning_rate": 1.8831939020201096e-06,
+      "loss": 0.8117,
+      "step": 175
+    },
+    {
+      "epoch": 0.41509433962264153,
+      "grad_norm": 0.7811131450862968,
+      "learning_rate": 1.8813406564011044e-06,
+      "loss": 0.742,
+      "step": 176
+    },
+    {
+      "epoch": 0.41745283018867924,
+      "grad_norm": 2.0285717851138974,
+      "learning_rate": 1.8794737512064888e-06,
+      "loss": 0.803,
+      "step": 177
+    },
+    {
+      "epoch": 0.419811320754717,
+      "grad_norm": 0.9681303046265977,
+      "learning_rate": 1.8775932153707426e-06,
+      "loss": 0.7857,
+      "step": 178
+    },
+    {
+      "epoch": 0.4221698113207547,
+      "grad_norm": 0.8312813398366425,
+      "learning_rate": 1.8756990780396006e-06,
+      "loss": 0.8091,
+      "step": 179
+    },
+    {
+      "epoch": 0.42452830188679247,
+      "grad_norm": 0.8325963532837448,
+      "learning_rate": 1.8737913685696027e-06,
+      "loss": 0.7936,
+      "step": 180
+    },
+    {
+      "epoch": 0.4268867924528302,
+      "grad_norm": 1.0242582555350113,
+      "learning_rate": 1.8718701165276383e-06,
+      "loss": 0.7679,
+      "step": 181
+    },
+    {
+      "epoch": 0.42924528301886794,
+      "grad_norm": 0.8568509432407337,
+      "learning_rate": 1.869935351690488e-06,
+      "loss": 0.7813,
+      "step": 182
+    },
+    {
+      "epoch": 0.43160377358490565,
+      "grad_norm": 1.1130973261677584,
+      "learning_rate": 1.867987104044363e-06,
+      "loss": 0.8173,
+      "step": 183
+    },
+    {
+      "epoch": 0.4339622641509434,
+      "grad_norm": 0.8076547906696275,
+      "learning_rate": 1.8660254037844386e-06,
+      "loss": 0.7823,
+      "step": 184
+    },
+    {
+      "epoch": 0.4363207547169811,
+      "grad_norm": 0.8462497637993398,
+      "learning_rate": 1.864050281314388e-06,
+      "loss": 0.7964,
+      "step": 185
+    },
+    {
+      "epoch": 0.4386792452830189,
+      "grad_norm": 0.8157104951412149,
+      "learning_rate": 1.8620617672459096e-06,
+      "loss": 0.7851,
+      "step": 186
+    },
+    {
+      "epoch": 0.4410377358490566,
+      "grad_norm": 0.8016398797843809,
+      "learning_rate": 1.8600598923982537e-06,
+      "loss": 0.7939,
+      "step": 187
+    },
+    {
+      "epoch": 0.44339622641509435,
+      "grad_norm": 0.9872324739847712,
+      "learning_rate": 1.858044687797745e-06,
+      "loss": 0.8105,
+      "step": 188
+    },
+    {
+      "epoch": 0.44575471698113206,
+      "grad_norm": 0.8070225439686687,
+      "learning_rate": 1.8560161846773e-06,
+      "loss": 0.7972,
+      "step": 189
+    },
+    {
+      "epoch": 0.4481132075471698,
+      "grad_norm": 0.8159111285487299,
+      "learning_rate": 1.8539744144759447e-06,
+      "loss": 0.7956,
+      "step": 190
+    },
+    {
+      "epoch": 0.45047169811320753,
+      "grad_norm": 0.8271007147512706,
+      "learning_rate": 1.851919408838327e-06,
+      "loss": 0.801,
+      "step": 191
+    },
+    {
+      "epoch": 0.4528301886792453,
+      "grad_norm": 0.8240075084440164,
+      "learning_rate": 1.8498511996142253e-06,
+      "loss": 0.8131,
+      "step": 192
+    },
+    {
+      "epoch": 0.455188679245283,
+      "grad_norm": 0.8100827961239685,
+      "learning_rate": 1.8477698188580557e-06,
+      "loss": 0.7953,
+      "step": 193
+    },
+    {
+      "epoch": 0.45754716981132076,
+      "grad_norm": 0.8090963621477647,
+      "learning_rate": 1.8456752988283757e-06,
+      "loss": 0.7713,
+      "step": 194
+    },
+    {
+      "epoch": 0.45990566037735847,
+      "grad_norm": 0.8646115465094427,
+      "learning_rate": 1.8435676719873827e-06,
+      "loss": 0.8218,
+      "step": 195
+    },
+    {
+      "epoch": 0.46226415094339623,
+      "grad_norm": 0.8283897647434588,
+      "learning_rate": 1.8414469710004124e-06,
+      "loss": 0.7937,
+      "step": 196
+    },
+    {
+      "epoch": 0.46462264150943394,
+      "grad_norm": 0.8130051258008277,
+      "learning_rate": 1.839313228735431e-06,
+      "loss": 0.7822,
+      "step": 197
+    },
+    {
+      "epoch": 0.4669811320754717,
+      "grad_norm": 0.8754673959382613,
+      "learning_rate": 1.8371664782625285e-06,
+      "loss": 0.7903,
+      "step": 198
+    },
+    {
+      "epoch": 0.4693396226415094,
+      "grad_norm": 1.287939397478769,
+      "learning_rate": 1.8350067528534024e-06,
+      "loss": 0.7718,
+      "step": 199
+    },
+    {
+      "epoch": 0.4716981132075472,
+      "grad_norm": 0.8236848990785427,
+      "learning_rate": 1.8328340859808446e-06,
+      "loss": 0.8012,
+      "step": 200
+    },
+    {
+      "epoch": 0.4740566037735849,
+      "grad_norm": 0.8938042537260998,
+      "learning_rate": 1.8306485113182229e-06,
+      "loss": 0.7721,
+      "step": 201
+    },
+    {
+      "epoch": 0.47641509433962265,
+      "grad_norm": 0.8496295751844541,
+      "learning_rate": 1.8284500627389567e-06,
+      "loss": 0.8043,
+      "step": 202
+    },
+    {
+      "epoch": 0.47877358490566035,
+      "grad_norm": 0.8325387527055562,
+      "learning_rate": 1.8262387743159948e-06,
+      "loss": 0.7936,
+      "step": 203
+    },
+    {
+      "epoch": 0.4811320754716981,
+      "grad_norm": 0.8207579274147835,
+      "learning_rate": 1.824014680321285e-06,
+      "loss": 0.8153,
+      "step": 204
+    },
+    {
+      "epoch": 0.4834905660377358,
+      "grad_norm": 0.9100175298073298,
+      "learning_rate": 1.821777815225245e-06,
+      "loss": 0.777,
+      "step": 205
+    },
+    {
+      "epoch": 0.4858490566037736,
+      "grad_norm": 0.9563101605978098,
+      "learning_rate": 1.8195282136962264e-06,
+      "loss": 0.7991,
+      "step": 206
+    },
+    {
+      "epoch": 0.4882075471698113,
+      "grad_norm": 0.8404681193002833,
+      "learning_rate": 1.817265910599978e-06,
+      "loss": 0.7968,
+      "step": 207
+    },
+    {
+      "epoch": 0.49056603773584906,
+      "grad_norm": 0.898412624742003,
+      "learning_rate": 1.814990940999106e-06,
+      "loss": 0.7585,
+      "step": 208
+    },
+    {
+      "epoch": 0.49292452830188677,
+      "grad_norm": 0.8495327539856999,
+      "learning_rate": 1.81270334015253e-06,
+      "loss": 0.8133,
+      "step": 209
+    },
+    {
+      "epoch": 0.49528301886792453,
+      "grad_norm": 1.1225882373174563,
+      "learning_rate": 1.8104031435149362e-06,
+      "loss": 0.8121,
+      "step": 210
+    },
+    {
+      "epoch": 0.49764150943396224,
+      "grad_norm": 0.9329252376646318,
+      "learning_rate": 1.8080903867362293e-06,
+      "loss": 0.815,
+      "step": 211
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 1.0591364512578068,
+      "learning_rate": 1.8057651056609782e-06,
+      "loss": 0.7649,
+      "step": 212
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 0.714023768901825,
+      "eval_runtime": 83.0267,
+      "eval_samples_per_second": 8.491,
+      "eval_steps_per_second": 0.542,
+      "step": 212
+    },
+    {
+      "epoch": 0.5023584905660378,
+      "grad_norm": 0.8614390398527076,
+      "learning_rate": 1.803427336327861e-06,
+      "loss": 0.8039,
+      "step": 213
+    },
+    {
+      "epoch": 0.5047169811320755,
+      "grad_norm": 0.903956531352285,
+      "learning_rate": 1.8010771149691082e-06,
+      "loss": 0.784,
+      "step": 214
+    },
+    {
+      "epoch": 0.5070754716981132,
+      "grad_norm": 0.8574588324038701,
+      "learning_rate": 1.7987144780099376e-06,
+      "loss": 0.7956,
+      "step": 215
+    },
+    {
+      "epoch": 0.5094339622641509,
+      "grad_norm": 0.8976197887421247,
+      "learning_rate": 1.7963394620679942e-06,
+      "loss": 0.8261,
+      "step": 216
+    },
+    {
+      "epoch": 0.5117924528301887,
+      "grad_norm": 0.8410804453164143,
+      "learning_rate": 1.7939521039527781e-06,
+      "loss": 0.7816,
+      "step": 217
+    },
+    {
+      "epoch": 0.5141509433962265,
+      "grad_norm": 0.903488950014599,
+      "learning_rate": 1.7915524406650775e-06,
+      "loss": 0.7701,
+      "step": 218
+    },
+    {
+      "epoch": 0.5165094339622641,
+      "grad_norm": 0.8374711197328489,
+      "learning_rate": 1.7891405093963937e-06,
+      "loss": 0.7761,
+      "step": 219
+    },
+    {
+      "epoch": 0.5188679245283019,
+      "grad_norm": 0.938187408138231,
+      "learning_rate": 1.7867163475283646e-06,
+      "loss": 0.8247,
+      "step": 220
+    },
+    {
+      "epoch": 0.5212264150943396,
+      "grad_norm": 1.974177068847498,
+      "learning_rate": 1.7842799926321863e-06,
+      "loss": 0.7984,
+      "step": 221
+    },
+    {
+      "epoch": 0.5235849056603774,
+      "grad_norm": 0.8445314748349818,
+      "learning_rate": 1.7818314824680298e-06,
+      "loss": 0.7783,
+      "step": 222
+    },
+    {
+      "epoch": 0.5259433962264151,
+      "grad_norm": 0.8043844477824647,
+      "learning_rate": 1.779370854984456e-06,
+      "loss": 0.7748,
+      "step": 223
+    },
+    {
+      "epoch": 0.5283018867924528,
+      "grad_norm": 0.8199345888971309,
+      "learning_rate": 1.7768981483178279e-06,
+      "loss": 0.7961,
+      "step": 224
+    },
+    {
+      "epoch": 0.5306603773584906,
+      "grad_norm": 0.8198791332861125,
+      "learning_rate": 1.7744134007917194e-06,
+      "loss": 0.7809,
+      "step": 225
+    },
+    {
+      "epoch": 0.5330188679245284,
+      "grad_norm": 0.8038089067621491,
+      "learning_rate": 1.7719166509163208e-06,
+      "loss": 0.752,
+      "step": 226
+    },
+    {
+      "epoch": 0.535377358490566,
+      "grad_norm": 0.8115733374512414,
+      "learning_rate": 1.7694079373878433e-06,
+      "loss": 0.7977,
+      "step": 227
+    },
+    {
+      "epoch": 0.5377358490566038,
+      "grad_norm": 0.8325145007355951,
+      "learning_rate": 1.7668872990879173e-06,
+      "loss": 0.775,
+      "step": 228
+    },
+    {
+      "epoch": 0.5400943396226415,
+      "grad_norm": 0.8029142353641874,
+      "learning_rate": 1.7643547750829918e-06,
+      "loss": 0.7897,
+      "step": 229
+    },
+    {
+      "epoch": 0.5424528301886793,
+      "grad_norm": 0.8068710232788708,
+      "learning_rate": 1.7618104046237274e-06,
+      "loss": 0.7869,
+      "step": 230
+    },
+    {
+      "epoch": 0.5448113207547169,
+      "grad_norm": 0.8179053896349096,
+      "learning_rate": 1.7592542271443887e-06,
+      "loss": 0.7737,
+      "step": 231
+    },
+    {
+      "epoch": 0.5471698113207547,
+      "grad_norm": 0.9236859332969564,
+      "learning_rate": 1.7566862822622328e-06,
+      "loss": 0.7779,
+      "step": 232
+    },
+    {
+      "epoch": 0.5495283018867925,
+      "grad_norm": 0.8105651817226325,
+      "learning_rate": 1.754106609776896e-06,
+      "loss": 0.7841,
+      "step": 233
+    },
+    {
+      "epoch": 0.5518867924528302,
+      "grad_norm": 0.8955884533729873,
+      "learning_rate": 1.7515152496697763e-06,
+      "loss": 0.8077,
+      "step": 234
+    },
+    {
+      "epoch": 0.5542452830188679,
+      "grad_norm": 1.015222493683735,
+      "learning_rate": 1.748912242103413e-06,
+      "loss": 0.8013,
+      "step": 235
+    },
+    {
+      "epoch": 0.5566037735849056,
+      "grad_norm": 0.9309116956021466,
+      "learning_rate": 1.746297627420866e-06,
+      "loss": 0.7672,
+      "step": 236
+    },
+    {
+      "epoch": 0.5589622641509434,
+      "grad_norm": 0.9599117437990355,
+      "learning_rate": 1.7436714461450897e-06,
+      "loss": 0.7957,
+      "step": 237
+    },
+    {
+      "epoch": 0.5613207547169812,
+      "grad_norm": 0.8340801449927656,
+      "learning_rate": 1.7410337389783041e-06,
+      "loss": 0.7994,
+      "step": 238
+    },
+    {
+      "epoch": 0.5636792452830188,
+      "grad_norm": 0.788362523807564,
+      "learning_rate": 1.7383845468013654e-06,
+      "loss": 0.7707,
+      "step": 239
+    },
+    {
+      "epoch": 0.5660377358490566,
+      "grad_norm": 0.8572171206767668,
+      "learning_rate": 1.7357239106731317e-06,
+      "loss": 0.7931,
+      "step": 240
+    },
+    {
+      "epoch": 0.5683962264150944,
+      "grad_norm": 4.180247592376519,
+      "learning_rate": 1.733051871829826e-06,
+      "loss": 0.798,
+      "step": 241
+    },
+    {
+      "epoch": 0.5707547169811321,
+      "grad_norm": 0.8738642652249169,
+      "learning_rate": 1.7303684716843995e-06,
+      "loss": 0.7768,
+      "step": 242
+    },
+    {
+      "epoch": 0.5731132075471698,
+      "grad_norm": 0.8413513234438594,
+      "learning_rate": 1.727673751825886e-06,
+      "loss": 0.8139,
+      "step": 243
+    },
+    {
+      "epoch": 0.5754716981132075,
+      "grad_norm": 0.8032830413975406,
+      "learning_rate": 1.7249677540187609e-06,
+      "loss": 0.7877,
+      "step": 244
+    },
+    {
+      "epoch": 0.5778301886792453,
+      "grad_norm": 0.7992641395875911,
+      "learning_rate": 1.7222505202022913e-06,
+      "loss": 0.7712,
+      "step": 245
+    },
+    {
+      "epoch": 0.5801886792452831,
+      "grad_norm": 0.8865998909749021,
+      "learning_rate": 1.7195220924898882e-06,
+      "loss": 0.7584,
+      "step": 246
+    },
+    {
+      "epoch": 0.5825471698113207,
+      "grad_norm": 0.8256395052056026,
+      "learning_rate": 1.7167825131684511e-06,
+      "loss": 0.7905,
+      "step": 247
+    },
+    {
+      "epoch": 0.5849056603773585,
+      "grad_norm": 0.8134932247816357,
+      "learning_rate": 1.7140318246977163e-06,
+      "loss": 0.7932,
+      "step": 248
+    },
+    {
+      "epoch": 0.5872641509433962,
+      "grad_norm": 0.8289362753913454,
+      "learning_rate": 1.7112700697095953e-06,
+      "loss": 0.7727,
+      "step": 249
+    },
+    {
+      "epoch": 0.589622641509434,
+      "grad_norm": 0.7919440339387878,
+      "learning_rate": 1.7084972910075154e-06,
+      "loss": 0.819,
+      "step": 250
+    },
+    {
+      "epoch": 0.5919811320754716,
+      "grad_norm": 0.7892796730141778,
+      "learning_rate": 1.7057135315657567e-06,
+      "loss": 0.7994,
+      "step": 251
+    },
+    {
+      "epoch": 0.5943396226415094,
+      "grad_norm": 0.818593844058535,
+      "learning_rate": 1.7029188345287865e-06,
+      "loss": 0.7833,
+      "step": 252
+    },
+    {
+      "epoch": 0.5966981132075472,
+      "grad_norm": 0.8550235074282425,
+      "learning_rate": 1.7001132432105894e-06,
+      "loss": 0.7627,
+      "step": 253
+    },
+    {
+      "epoch": 0.5990566037735849,
+      "grad_norm": 0.7825131867584777,
+      "learning_rate": 1.6972968010939952e-06,
+      "loss": 0.7863,
+      "step": 254
+    },
+    {
+      "epoch": 0.6014150943396226,
+      "grad_norm": 0.7872637454110324,
+      "learning_rate": 1.6944695518300084e-06,
+      "loss": 0.8202,
+      "step": 255
+    },
+    {
+      "epoch": 0.6037735849056604,
+      "grad_norm": 0.8264865629926671,
+      "learning_rate": 1.6916315392371283e-06,
+      "loss": 0.7794,
+      "step": 256
+    },
+    {
+      "epoch": 0.6061320754716981,
+      "grad_norm": 0.8112171041686048,
+      "learning_rate": 1.688782807300671e-06,
+      "loss": 0.7975,
+      "step": 257
+    },
+    {
+      "epoch": 0.6084905660377359,
+      "grad_norm": 0.8017599214256371,
+      "learning_rate": 1.685923400172088e-06,
+      "loss": 0.7821,
+      "step": 258
+    },
+    {
+      "epoch": 0.6108490566037735,
+      "grad_norm": 0.7855915924335556,
+      "learning_rate": 1.683053362168282e-06,
+      "loss": 0.7834,
+      "step": 259
+    },
+    {
+      "epoch": 0.6132075471698113,
+      "grad_norm": 0.8227864852025443,
+      "learning_rate": 1.6801727377709191e-06,
+      "loss": 0.7715,
+      "step": 260
+    },
+    {
+      "epoch": 0.6155660377358491,
+      "grad_norm": 0.8347676603852344,
+      "learning_rate": 1.6772815716257411e-06,
+      "loss": 0.7783,
+      "step": 261
+    },
+    {
+      "epoch": 0.6179245283018868,
+      "grad_norm": 0.847840576553053,
+      "learning_rate": 1.6743799085418709e-06,
+      "loss": 0.7796,
+      "step": 262
+    },
+    {
+      "epoch": 0.6202830188679245,
+      "grad_norm": 0.8187554443504805,
+      "learning_rate": 1.671467793491121e-06,
+      "loss": 0.8223,
+      "step": 263
+    },
+    {
+      "epoch": 0.6226415094339622,
+      "grad_norm": 0.8560599665847224,
+      "learning_rate": 1.6685452716072942e-06,
+      "loss": 0.7769,
+      "step": 264
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 0.8756516850796994,
+      "learning_rate": 1.6656123881854858e-06,
+      "loss": 0.782,
+      "step": 265
+    },
+    {
+      "epoch": 0.6273584905660378,
+      "grad_norm": 0.861305723586375,
+      "learning_rate": 1.6626691886813802e-06,
+      "loss": 0.7907,
+      "step": 266
+    },
+    {
+      "epoch": 0.6297169811320755,
+      "grad_norm": 0.8565858633218812,
+      "learning_rate": 1.6597157187105474e-06,
+      "loss": 0.7586,
+      "step": 267
+    },
+    {
+      "epoch": 0.6320754716981132,
+      "grad_norm": 0.7796005038585514,
+      "learning_rate": 1.6567520240477343e-06,
+      "loss": 0.7996,
+      "step": 268
+    },
+    {
+      "epoch": 0.6344339622641509,
+      "grad_norm": 0.9853760454692906,
+      "learning_rate": 1.6537781506261586e-06,
+      "loss": 0.803,
+      "step": 269
+    },
+    {
+      "epoch": 0.6367924528301887,
+      "grad_norm": 0.8214336899114236,
+      "learning_rate": 1.6507941445367934e-06,
+      "loss": 0.7484,
+      "step": 270
+    },
+    {
+      "epoch": 0.6391509433962265,
+      "grad_norm": 0.8871384572382998,
+      "learning_rate": 1.647800052027655e-06,
+      "loss": 0.7876,
+      "step": 271
+    },
+    {
+      "epoch": 0.6415094339622641,
+      "grad_norm": 0.8564572450153076,
+      "learning_rate": 1.6447959195030849e-06,
+      "loss": 0.8054,
+      "step": 272
+    },
+    {
+      "epoch": 0.6438679245283019,
+      "grad_norm": 0.8408989291477506,
+      "learning_rate": 1.6417817935230316e-06,
+      "loss": 0.7668,
+      "step": 273
+    },
+    {
+      "epoch": 0.6462264150943396,
+      "grad_norm": 0.7945270947755096,
+      "learning_rate": 1.6387577208023279e-06,
+      "loss": 0.7893,
+      "step": 274
+    },
+    {
+      "epoch": 0.6485849056603774,
+      "grad_norm": 0.8844404937902336,
+      "learning_rate": 1.6357237482099683e-06,
+      "loss": 0.794,
+      "step": 275
+    },
+    {
+      "epoch": 0.6509433962264151,
+      "grad_norm": 0.8518070202862351,
+      "learning_rate": 1.6326799227683803e-06,
+      "loss": 0.7967,
+      "step": 276
+    },
+    {
+      "epoch": 0.6533018867924528,
+      "grad_norm": 0.819470621959587,
+      "learning_rate": 1.6296262916526995e-06,
+      "loss": 0.7633,
+      "step": 277
+    },
+    {
+      "epoch": 0.6556603773584906,
+      "grad_norm": 0.7989369195204531,
+      "learning_rate": 1.626562902190034e-06,
+      "loss": 0.807,
+      "step": 278
+    },
+    {
+      "epoch": 0.6580188679245284,
+      "grad_norm": 0.9412075176216419,
+      "learning_rate": 1.6234898018587336e-06,
+      "loss": 0.7752,
+      "step": 279
+    },
+    {
+      "epoch": 0.660377358490566,
+      "grad_norm": 0.7895838126966979,
+      "learning_rate": 1.6204070382876531e-06,
+      "loss": 0.7786,
+      "step": 280
+    },
+    {
+      "epoch": 0.6627358490566038,
+      "grad_norm": 0.7981016976096619,
+      "learning_rate": 1.6173146592554152e-06,
+      "loss": 0.7958,
+      "step": 281
+    },
+    {
+      "epoch": 0.6650943396226415,
+      "grad_norm": 0.8907323576889667,
+      "learning_rate": 1.6142127126896679e-06,
+      "loss": 0.807,
+      "step": 282
+    },
+    {
+      "epoch": 0.6674528301886793,
+      "grad_norm": 0.8053567442309949,
+      "learning_rate": 1.6111012466663434e-06,
+      "loss": 0.788,
+      "step": 283
+    },
+    {
+      "epoch": 0.6698113207547169,
+      "grad_norm": 0.8378739800286575,
+      "learning_rate": 1.6079803094089125e-06,
+      "loss": 0.736,
+      "step": 284
+    },
+    {
+      "epoch": 0.6721698113207547,
+      "grad_norm": 0.8858100985238717,
+      "learning_rate": 1.6048499492876375e-06,
+      "loss": 0.8106,
+      "step": 285
+    },
+    {
+      "epoch": 0.6745283018867925,
+      "grad_norm": 0.8276972316955101,
+      "learning_rate": 1.6017102148188215e-06,
+      "loss": 0.7835,
+      "step": 286
+    },
+    {
+      "epoch": 0.6768867924528302,
+      "grad_norm": 0.8853131510592382,
+      "learning_rate": 1.598561154664058e-06,
+      "loss": 0.7965,
+      "step": 287
+    },
+    {
+      "epoch": 0.6792452830188679,
+      "grad_norm": 0.8786321199638693,
+      "learning_rate": 1.5954028176294746e-06,
+      "loss": 0.8268,
+      "step": 288
+    },
+    {
+      "epoch": 0.6816037735849056,
+      "grad_norm": 0.8669272571254311,
+      "learning_rate": 1.5922352526649801e-06,
+      "loss": 0.7823,
+      "step": 289
+    },
+    {
+      "epoch": 0.6839622641509434,
+      "grad_norm": 0.9644127048111335,
+      "learning_rate": 1.589058508863501e-06,
+      "loss": 0.78,
+      "step": 290
+    },
+    {
+      "epoch": 0.6863207547169812,
+      "grad_norm": 0.8073310279373501,
+      "learning_rate": 1.5858726354602248e-06,
+      "loss": 0.7765,
+      "step": 291
+    },
+    {
+      "epoch": 0.6886792452830188,
+      "grad_norm": 0.8775721979922628,
+      "learning_rate": 1.5826776818318348e-06,
+      "loss": 0.7597,
+      "step": 292
+    },
+    {
+      "epoch": 0.6910377358490566,
+      "grad_norm": 1.015547726125361,
+      "learning_rate": 1.5794736974957464e-06,
+      "loss": 0.786,
+      "step": 293
+    },
+    {
+      "epoch": 0.6933962264150944,
+      "grad_norm": 0.7708601481727361,
+      "learning_rate": 1.5762607321093366e-06,
+      "loss": 0.7817,
+      "step": 294
+    },
+    {
+      "epoch": 0.6957547169811321,
+      "grad_norm": 0.8426421786890617,
+      "learning_rate": 1.5730388354691785e-06,
+      "loss": 0.7847,
+      "step": 295
+    },
+    {
+      "epoch": 0.6981132075471698,
+      "grad_norm": 0.7855014020345946,
+      "learning_rate": 1.569808057510266e-06,
+      "loss": 0.7748,
+      "step": 296
+    },
+    {
+      "epoch": 0.7004716981132075,
+      "grad_norm": 0.8389890263875862,
+      "learning_rate": 1.5665684483052424e-06,
+      "loss": 0.7856,
+      "step": 297
+    },
+    {
+      "epoch": 0.7028301886792453,
+      "grad_norm": 0.8482309028462507,
+      "learning_rate": 1.563320058063622e-06,
+      "loss": 0.7469,
+      "step": 298
+    },
+    {
+      "epoch": 0.7051886792452831,
+      "grad_norm": 1.0888639918368348,
+      "learning_rate": 1.5600629371310144e-06,
+      "loss": 0.7905,
+      "step": 299
+    },
+    {
+      "epoch": 0.7075471698113207,
+      "grad_norm": 0.8011947485403833,
+      "learning_rate": 1.556797135988342e-06,
+      "loss": 0.7902,
+      "step": 300
+    },
+    {
+      "epoch": 0.7099056603773585,
+      "grad_norm": 0.7584432625187374,
+      "learning_rate": 1.5535227052510588e-06,
+      "loss": 0.7935,
+      "step": 301
+    },
+    {
+      "epoch": 0.7122641509433962,
+      "grad_norm": 1.1353449141437797,
+      "learning_rate": 1.5502396956683667e-06,
+      "loss": 0.7992,
+      "step": 302
+    },
+    {
+      "epoch": 0.714622641509434,
+      "grad_norm": 0.8144382945981636,
+      "learning_rate": 1.5469481581224271e-06,
+      "loss": 0.761,
+      "step": 303
+    },
+    {
+      "epoch": 0.7169811320754716,
+      "grad_norm": 0.9448262210280223,
+      "learning_rate": 1.5436481436275724e-06,
+      "loss": 0.7839,
+      "step": 304
+    },
+    {
+      "epoch": 0.7193396226415094,
+      "grad_norm": 0.8145609006654517,
+      "learning_rate": 1.5403397033295179e-06,
+      "loss": 0.7975,
+      "step": 305
+    },
+    {
+      "epoch": 0.7216981132075472,
+      "grad_norm": 0.9825256520905712,
+      "learning_rate": 1.537022888504566e-06,
+      "loss": 0.7918,
+      "step": 306
+    },
+    {
+      "epoch": 0.7240566037735849,
+      "grad_norm": 0.9647183394075436,
+      "learning_rate": 1.5336977505588134e-06,
+      "loss": 0.7973,
+      "step": 307
+    },
+    {
+      "epoch": 0.7264150943396226,
+      "grad_norm": 0.802175509091533,
+      "learning_rate": 1.530364341027354e-06,
+      "loss": 0.7928,
+      "step": 308
+    },
+    {
+      "epoch": 0.7287735849056604,
+      "grad_norm": 0.8122620856594104,
+      "learning_rate": 1.5270227115734789e-06,
+      "loss": 0.7767,
+      "step": 309
+    },
+    {
+      "epoch": 0.7311320754716981,
+      "grad_norm": 0.8656607812881103,
+      "learning_rate": 1.5236729139878778e-06,
+      "loss": 0.7746,
+      "step": 310
+    },
+    {
+      "epoch": 0.7334905660377359,
+      "grad_norm": 0.8721224552709389,
+      "learning_rate": 1.5203150001878353e-06,
+      "loss": 0.7929,
+      "step": 311
+    },
+    {
+      "epoch": 0.7358490566037735,
+      "grad_norm": 0.7856050732773205,
+      "learning_rate": 1.5169490222164254e-06,
+      "loss": 0.7733,
+      "step": 312
+    },
+    {
+      "epoch": 0.7382075471698113,
+      "grad_norm": 0.8639502364522528,
+      "learning_rate": 1.5135750322417066e-06,
+      "loss": 0.7958,
+      "step": 313
+    },
+    {
+      "epoch": 0.7405660377358491,
+      "grad_norm": 0.9986722246195658,
+      "learning_rate": 1.5101930825559124e-06,
+      "loss": 0.7944,
+      "step": 314
+    },
+    {
+      "epoch": 0.7429245283018868,
+      "grad_norm": 0.8861549532629037,
+      "learning_rate": 1.50680322557464e-06,
+      "loss": 0.7887,
+      "step": 315
+    },
+    {
+      "epoch": 0.7452830188679245,
+      "grad_norm": 0.8526619540985027,
+      "learning_rate": 1.5034055138360398e-06,
+      "loss": 0.7388,
+      "step": 316
+    },
+    {
+      "epoch": 0.7476415094339622,
+      "grad_norm": 0.8760084110715984,
+      "learning_rate": 1.5e-06,
+      "loss": 0.7719,
+      "step": 317
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.8038573705056793,
+      "learning_rate": 1.4965867368473306e-06,
+      "loss": 0.7514,
+      "step": 318
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 0.6982797384262085,
+      "eval_runtime": 83.2645,
+      "eval_samples_per_second": 8.467,
+      "eval_steps_per_second": 0.54,
+      "step": 318
+    },
+    {
+      "epoch": 0.7523584905660378,
+      "grad_norm": 0.8193168846563599,
+      "learning_rate": 1.4931657772789457e-06,
+      "loss": 0.7834,
+      "step": 319
+    },
+    {
+      "epoch": 0.7547169811320755,
+      "grad_norm": 0.9092282687501353,
+      "learning_rate": 1.4897371743150423e-06,
+      "loss": 0.753,
+      "step": 320
+    },
+    {
+      "epoch": 0.7570754716981132,
+      "grad_norm": 0.944599516853919,
+      "learning_rate": 1.4863009810942813e-06,
+      "loss": 0.7855,
+      "step": 321
+    },
+    {
+      "epoch": 0.7594339622641509,
+      "grad_norm": 0.9097780581691608,
+      "learning_rate": 1.4828572508729606e-06,
+      "loss": 0.7837,
+      "step": 322
+    },
+    {
+      "epoch": 0.7617924528301887,
+      "grad_norm": 0.8635246603088246,
+      "learning_rate": 1.479406037024192e-06,
+      "loss": 0.7862,
+      "step": 323
+    },
+    {
+      "epoch": 0.7641509433962265,
+      "grad_norm": 0.8041740773336183,
+      "learning_rate": 1.4759473930370736e-06,
+      "loss": 0.7705,
+      "step": 324
+    },
+    {
+      "epoch": 0.7665094339622641,
+      "grad_norm": 0.8563734416095982,
+      "learning_rate": 1.4724813725158596e-06,
+      "loss": 0.7944,
+      "step": 325
+    },
+    {
+      "epoch": 0.7688679245283019,
+      "grad_norm": 0.7885016820225573,
+      "learning_rate": 1.4690080291791312e-06,
+      "loss": 0.78,
+      "step": 326
+    },
+    {
+      "epoch": 0.7712264150943396,
+      "grad_norm": 0.9769999602785767,
+      "learning_rate": 1.4655274168589633e-06,
+      "loss": 0.7506,
+      "step": 327
+    },
+    {
+      "epoch": 0.7735849056603774,
+      "grad_norm": 0.8241115466345136,
+      "learning_rate": 1.462039589500089e-06,
+      "loss": 0.7826,
+      "step": 328
+    },
+    {
+      "epoch": 0.7759433962264151,
+      "grad_norm": 0.8261080001645498,
+      "learning_rate": 1.4585446011590658e-06,
+      "loss": 0.7592,
+      "step": 329
+    },
+    {
+      "epoch": 0.7783018867924528,
+      "grad_norm": 1.1493611817217861,
+      "learning_rate": 1.4550425060034365e-06,
+      "loss": 0.7674,
+      "step": 330
+    },
+    {
+      "epoch": 0.7806603773584906,
+      "grad_norm": 0.8343201494559767,
+      "learning_rate": 1.4515333583108893e-06,
+      "loss": 0.8072,
+      "step": 331
+    },
+    {
+      "epoch": 0.7830188679245284,
+      "grad_norm": 1.0664747829909942,
+      "learning_rate": 1.4480172124684172e-06,
+      "loss": 0.7802,
+      "step": 332
+    },
+    {
+      "epoch": 0.785377358490566,
+      "grad_norm": 0.7792110149695123,
+      "learning_rate": 1.4444941229714758e-06,
+      "loss": 0.7704,
+      "step": 333
+    },
+    {
+      "epoch": 0.7877358490566038,
+      "grad_norm": 0.8718279852088779,
+      "learning_rate": 1.440964144423136e-06,
+      "loss": 0.7963,
+      "step": 334
+    },
+    {
+      "epoch": 0.7900943396226415,
+      "grad_norm": 1.02769301319547,
+      "learning_rate": 1.4374273315332415e-06,
+      "loss": 0.7863,
+      "step": 335
+    },
+    {
+      "epoch": 0.7924528301886793,
+      "grad_norm": 0.7843647199877115,
+      "learning_rate": 1.433883739117558e-06,
+      "loss": 0.7705,
+      "step": 336
+    },
+    {
+      "epoch": 0.7948113207547169,
+      "grad_norm": 0.815818120801617,
+      "learning_rate": 1.430333422096925e-06,
+      "loss": 0.7651,
+      "step": 337
+    },
+    {
+      "epoch": 0.7971698113207547,
+      "grad_norm": 0.7835815188776669,
+      "learning_rate": 1.4267764354964037e-06,
+      "loss": 0.789,
+      "step": 338
+    },
+    {
+      "epoch": 0.7995283018867925,
+      "grad_norm": 0.8382956450308263,
+      "learning_rate": 1.423212834444425e-06,
+      "loss": 0.817,
+      "step": 339
+    },
+    {
+      "epoch": 0.8018867924528302,
+      "grad_norm": 0.9606287069411898,
+      "learning_rate": 1.4196426741719345e-06,
+      "loss": 0.7801,
+      "step": 340
+    },
+    {
+      "epoch": 0.8042452830188679,
+      "grad_norm": 0.7833416575809102,
+      "learning_rate": 1.4160660100115373e-06,
+      "loss": 0.8139,
+      "step": 341
+    },
+    {
+      "epoch": 0.8066037735849056,
+      "grad_norm": 0.8112490810199738,
+      "learning_rate": 1.4124828973966392e-06,
+      "loss": 0.7951,
+      "step": 342
+    },
+    {
+      "epoch": 0.8089622641509434,
+      "grad_norm": 0.8533925894384254,
+      "learning_rate": 1.4088933918605887e-06,
+      "loss": 0.7555,
+      "step": 343
+    },
+    {
+      "epoch": 0.8113207547169812,
+      "grad_norm": 0.8917123790523681,
+      "learning_rate": 1.405297549035816e-06,
+      "loss": 0.7941,
+      "step": 344
+    },
+    {
+      "epoch": 0.8136792452830188,
+      "grad_norm": 0.8690146736748873,
+      "learning_rate": 1.4016954246529694e-06,
+      "loss": 0.7966,
+      "step": 345
+    },
+    {
+      "epoch": 0.8160377358490566,
+      "grad_norm": 0.8372555162768984,
+      "learning_rate": 1.3980870745400544e-06,
+      "loss": 0.7972,
+      "step": 346
+    },
+    {
+      "epoch": 0.8183962264150944,
+      "grad_norm": 0.8356712411477044,
+      "learning_rate": 1.3944725546215662e-06,
+      "loss": 0.753,
+      "step": 347
+    },
+    {
+      "epoch": 0.8207547169811321,
+      "grad_norm": 0.8366058600346219,
+      "learning_rate": 1.3908519209176225e-06,
+      "loss": 0.7813,
+      "step": 348
+    },
+    {
+      "epoch": 0.8231132075471698,
+      "grad_norm": 0.8373524566281616,
+      "learning_rate": 1.3872252295430986e-06,
+      "loss": 0.7956,
+      "step": 349
+    },
+    {
+      "epoch": 0.8254716981132075,
+      "grad_norm": 0.7833296789989939,
+      "learning_rate": 1.3835925367067529e-06,
+      "loss": 0.7449,
+      "step": 350
+    },
+    {
+      "epoch": 0.8278301886792453,
+      "grad_norm": 0.7981265071211454,
+      "learning_rate": 1.37995389871036e-06,
+      "loss": 0.7756,
+      "step": 351
+    },
+    {
+      "epoch": 0.8301886792452831,
+      "grad_norm": 0.8840540972881347,
+      "learning_rate": 1.3763093719478357e-06,
+      "loss": 0.7984,
+      "step": 352
+    },
+    {
+      "epoch": 0.8325471698113207,
+      "grad_norm": 0.985313340190558,
+      "learning_rate": 1.372659012904363e-06,
+      "loss": 0.7821,
+      "step": 353
+    },
+    {
+      "epoch": 0.8349056603773585,
+      "grad_norm": 0.8138177618630202,
+      "learning_rate": 1.369002878155519e-06,
+      "loss": 0.7618,
+      "step": 354
+    },
+    {
+      "epoch": 0.8372641509433962,
+      "grad_norm": 0.7948448582342686,
+      "learning_rate": 1.3653410243663951e-06,
+      "loss": 0.7559,
+      "step": 355
+    },
+    {
+      "epoch": 0.839622641509434,
+      "grad_norm": 0.8151246949343306,
+      "learning_rate": 1.3616735082907196e-06,
+      "loss": 0.7722,
+      "step": 356
+    },
+    {
+      "epoch": 0.8419811320754716,
+      "grad_norm": 0.8162887157403834,
+      "learning_rate": 1.35800038676998e-06,
+      "loss": 0.7711,
+      "step": 357
+    },
+    {
+      "epoch": 0.8443396226415094,
+      "grad_norm": 1.1322690706169278,
+      "learning_rate": 1.3543217167325388e-06,
+      "loss": 0.762,
+      "step": 358
+    },
+    {
+      "epoch": 0.8466981132075472,
+      "grad_norm": 0.8102205098745584,
+      "learning_rate": 1.3506375551927544e-06,
+      "loss": 0.7532,
+      "step": 359
+    },
+    {
+      "epoch": 0.8490566037735849,
+      "grad_norm": 0.9369994865443904,
+      "learning_rate": 1.3469479592500951e-06,
+      "loss": 0.737,
+      "step": 360
+    },
+    {
+      "epoch": 0.8514150943396226,
+      "grad_norm": 0.9319609634032745,
+      "learning_rate": 1.3432529860882556e-06,
+      "loss": 0.8074,
+      "step": 361
+    },
+    {
+      "epoch": 0.8537735849056604,
+      "grad_norm": 0.844911038169109,
+      "learning_rate": 1.3395526929742691e-06,
+      "loss": 0.7576,
+      "step": 362
+    },
+    {
+      "epoch": 0.8561320754716981,
+      "grad_norm": 0.8213852435481517,
+      "learning_rate": 1.3358471372576227e-06,
+      "loss": 0.8091,
+      "step": 363
+    },
+    {
+      "epoch": 0.8584905660377359,
+      "grad_norm": 0.8679917359639049,
+      "learning_rate": 1.3321363763693643e-06,
+      "loss": 0.7649,
+      "step": 364
+    },
+    {
+      "epoch": 0.8608490566037735,
+      "grad_norm": 0.863728546344944,
+      "learning_rate": 1.3284204678212167e-06,
+      "loss": 0.7364,
+      "step": 365
+    },
+    {
+      "epoch": 0.8632075471698113,
+      "grad_norm": 0.9151925878031607,
+      "learning_rate": 1.3246994692046835e-06,
+      "loss": 0.7617,
+      "step": 366
+    },
+    {
+      "epoch": 0.8655660377358491,
+      "grad_norm": 0.7676651686086633,
+      "learning_rate": 1.3209734381901578e-06,
+      "loss": 0.7703,
+      "step": 367
+    },
+    {
+      "epoch": 0.8679245283018868,
+      "grad_norm": 0.79154626898034,
+      "learning_rate": 1.3172424325260272e-06,
+      "loss": 0.7816,
+      "step": 368
+    },
+    {
+      "epoch": 0.8702830188679245,
+      "grad_norm": 0.8831690649426513,
+      "learning_rate": 1.3135065100377814e-06,
+      "loss": 0.7646,
+      "step": 369
+    },
+    {
+      "epoch": 0.8726415094339622,
+      "grad_norm": 0.8098189405568664,
+      "learning_rate": 1.3097657286271116e-06,
+      "loss": 0.7673,
+      "step": 370
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 0.8071739591609438,
+      "learning_rate": 1.3060201462710176e-06,
+      "loss": 0.7438,
+      "step": 371
+    },
+    {
+      "epoch": 0.8773584905660378,
+      "grad_norm": 0.8864737282217798,
+      "learning_rate": 1.3022698210209066e-06,
+      "loss": 0.7393,
+      "step": 372
+    },
+    {
+      "epoch": 0.8797169811320755,
+      "grad_norm": 0.8193555244410853,
+      "learning_rate": 1.2985148110016947e-06,
+      "loss": 0.7658,
+      "step": 373
+    },
+    {
+      "epoch": 0.8820754716981132,
+      "grad_norm": 0.7819668232955883,
+      "learning_rate": 1.2947551744109043e-06,
+      "loss": 0.7876,
+      "step": 374
+    },
+    {
+      "epoch": 0.8844339622641509,
+      "grad_norm": 0.9661764152661253,
+      "learning_rate": 1.2909909695177645e-06,
+      "loss": 0.7556,
+      "step": 375
+    },
+    {
+      "epoch": 0.8867924528301887,
+      "grad_norm": 0.7916178501632968,
+      "learning_rate": 1.2872222546623065e-06,
+      "loss": 0.7776,
+      "step": 376
+    },
+    {
+      "epoch": 0.8891509433962265,
+      "grad_norm": 0.908391562439544,
+      "learning_rate": 1.2834490882544598e-06,
+      "loss": 0.7604,
+      "step": 377
+    },
+    {
+      "epoch": 0.8915094339622641,
+      "grad_norm": 0.7673617521747667,
+      "learning_rate": 1.2796715287731461e-06,
+      "loss": 0.7786,
+      "step": 378
+    },
+    {
+      "epoch": 0.8938679245283019,
+      "grad_norm": 1.1737766347467777,
+      "learning_rate": 1.2758896347653752e-06,
+      "loss": 0.7692,
+      "step": 379
+    },
+    {
+      "epoch": 0.8962264150943396,
+      "grad_norm": 0.7823714073715833,
+      "learning_rate": 1.272103464845335e-06,
+      "loss": 0.7389,
+      "step": 380
+    },
+    {
+      "epoch": 0.8985849056603774,
+      "grad_norm": 1.0030453246218272,
+      "learning_rate": 1.2683130776934848e-06,
+      "loss": 0.7556,
+      "step": 381
+    },
+    {
+      "epoch": 0.9009433962264151,
+      "grad_norm": 0.9688917314182516,
+      "learning_rate": 1.2645185320556444e-06,
+      "loss": 0.7415,
+      "step": 382
+    },
+    {
+      "epoch": 0.9033018867924528,
+      "grad_norm": 0.8539895113482738,
+      "learning_rate": 1.2607198867420858e-06,
+      "loss": 0.7772,
+      "step": 383
+    },
+    {
+      "epoch": 0.9056603773584906,
+      "grad_norm": 0.8011530846319137,
+      "learning_rate": 1.256917200626619e-06,
+      "loss": 0.7426,
+      "step": 384
+    },
+    {
+      "epoch": 0.9080188679245284,
+      "grad_norm": 0.9859270894449156,
+      "learning_rate": 1.253110532645682e-06,
+      "loss": 0.7942,
+      "step": 385
+    },
+    {
+      "epoch": 0.910377358490566,
+      "grad_norm": 0.7970067775774904,
+      "learning_rate": 1.2492999417974253e-06,
+      "loss": 0.7914,
+      "step": 386
+    },
+    {
+      "epoch": 0.9127358490566038,
+      "grad_norm": 0.9049836680757578,
+      "learning_rate": 1.245485487140799e-06,
+      "loss": 0.7787,
+      "step": 387
+    },
+    {
+      "epoch": 0.9150943396226415,
+      "grad_norm": 0.866322963165458,
+      "learning_rate": 1.2416672277946373e-06,
+      "loss": 0.781,
+      "step": 388
+    },
+    {
+      "epoch": 0.9174528301886793,
+      "grad_norm": 0.9269306886834423,
+      "learning_rate": 1.2378452229367413e-06,
+      "loss": 0.808,
+      "step": 389
+    },
+    {
+      "epoch": 0.9198113207547169,
+      "grad_norm": 0.8286820002413788,
+      "learning_rate": 1.2340195318029622e-06,
+      "loss": 0.7623,
+      "step": 390
+    },
+    {
+      "epoch": 0.9221698113207547,
+      "grad_norm": 0.8143258703482895,
+      "learning_rate": 1.2301902136862848e-06,
+      "loss": 0.7965,
+      "step": 391
+    },
+    {
+      "epoch": 0.9245283018867925,
+      "grad_norm": 0.8041175896213196,
+      "learning_rate": 1.2263573279359053e-06,
+      "loss": 0.7613,
+      "step": 392
+    },
+    {
+      "epoch": 0.9268867924528302,
+      "grad_norm": 0.9087369855701409,
+      "learning_rate": 1.2225209339563143e-06,
+      "loss": 0.7669,
+      "step": 393
+    },
+    {
+      "epoch": 0.9292452830188679,
+      "grad_norm": 0.8126237661274147,
+      "learning_rate": 1.2186810912063758e-06,
+      "loss": 0.7531,
+      "step": 394
+    },
+    {
+      "epoch": 0.9316037735849056,
+      "grad_norm": 0.8354891878817764,
+      "learning_rate": 1.214837859198404e-06,
+      "loss": 0.768,
+      "step": 395
+    },
+    {
+      "epoch": 0.9339622641509434,
+      "grad_norm": 0.7793219061787805,
+      "learning_rate": 1.2109912974972422e-06,
+      "loss": 0.7831,
+      "step": 396
+    },
+    {
+      "epoch": 0.9363207547169812,
+      "grad_norm": 0.7813175142444524,
+      "learning_rate": 1.2071414657193396e-06,
+      "loss": 0.7671,
+      "step": 397
+    },
+    {
+      "epoch": 0.9386792452830188,
+      "grad_norm": 0.8960835581240535,
+      "learning_rate": 1.2032884235318268e-06,
+      "loss": 0.775,
+      "step": 398
+    },
+    {
+      "epoch": 0.9410377358490566,
+      "grad_norm": 0.8773817072709057,
+      "learning_rate": 1.1994322306515925e-06,
+      "loss": 0.7915,
+      "step": 399
+    },
+    {
+      "epoch": 0.9433962264150944,
+      "grad_norm": 0.8191284697051531,
+      "learning_rate": 1.1955729468443546e-06,
+      "loss": 0.7548,
+      "step": 400
+    },
+    {
+      "epoch": 0.9457547169811321,
+      "grad_norm": 0.8704070228218153,
+      "learning_rate": 1.1917106319237384e-06,
+      "loss": 0.773,
+      "step": 401
+    },
+    {
+      "epoch": 0.9481132075471698,
+      "grad_norm": 0.8119790656590767,
+      "learning_rate": 1.1878453457503464e-06,
+      "loss": 0.7933,
+      "step": 402
+    },
+    {
+      "epoch": 0.9504716981132075,
+      "grad_norm": 0.8290570173312977,
+      "learning_rate": 1.1839771482308308e-06,
+      "loss": 0.7515,
+      "step": 403
+    },
+    {
+      "epoch": 0.9528301886792453,
+      "grad_norm": 0.7808059259378133,
+      "learning_rate": 1.1801060993169666e-06,
+      "loss": 0.771,
+      "step": 404
+    },
+    {
+      "epoch": 0.9551886792452831,
+      "grad_norm": 0.8554505156061943,
+      "learning_rate": 1.1762322590047219e-06,
+      "loss": 0.767,
+      "step": 405
+    },
+    {
+      "epoch": 0.9575471698113207,
+      "grad_norm": 0.8132652328205165,
+      "learning_rate": 1.172355687333326e-06,
+      "loss": 0.7467,
+      "step": 406
+    },
+    {
+      "epoch": 0.9599056603773585,
+      "grad_norm": 0.7767004491954446,
+      "learning_rate": 1.168476444384342e-06,
+      "loss": 0.7667,
+      "step": 407
+    },
+    {
+      "epoch": 0.9622641509433962,
+      "grad_norm": 0.8619052365176207,
+      "learning_rate": 1.164594590280734e-06,
+      "loss": 0.7993,
+      "step": 408
+    },
+    {
+      "epoch": 0.964622641509434,
+      "grad_norm": 0.8115144668833016,
+      "learning_rate": 1.1607101851859345e-06,
+      "loss": 0.7787,
+      "step": 409
+    },
+    {
+      "epoch": 0.9669811320754716,
+      "grad_norm": 0.8000429071733963,
+      "learning_rate": 1.156823289302914e-06,
+      "loss": 0.7767,
+      "step": 410
+    },
+    {
+      "epoch": 0.9693396226415094,
+      "grad_norm": 0.8228751842965996,
+      "learning_rate": 1.152933962873246e-06,
+      "loss": 0.7829,
+      "step": 411
+    },
+    {
+      "epoch": 0.9716981132075472,
+      "grad_norm": 0.8685312714228164,
+      "learning_rate": 1.1490422661761743e-06,
+      "loss": 0.7775,
+      "step": 412
+    },
+    {
+      "epoch": 0.9740566037735849,
+      "grad_norm": 0.9383805515060808,
+      "learning_rate": 1.1451482595276796e-06,
+      "loss": 0.7716,
+      "step": 413
+    },
+    {
+      "epoch": 0.9764150943396226,
+      "grad_norm": 0.978907741116596,
+      "learning_rate": 1.1412520032795419e-06,
+      "loss": 0.79,
+      "step": 414
+    },
+    {
+      "epoch": 0.9787735849056604,
+      "grad_norm": 0.936909408413621,
+      "learning_rate": 1.1373535578184082e-06,
+      "loss": 0.7463,
+      "step": 415
+    },
+    {
+      "epoch": 0.9811320754716981,
+      "grad_norm": 0.7779360289097986,
+      "learning_rate": 1.1334529835648552e-06,
+      "loss": 0.7544,
+      "step": 416
+    },
+    {
+      "epoch": 0.9834905660377359,
+      "grad_norm": 0.8442404415567627,
+      "learning_rate": 1.1295503409724525e-06,
+      "loss": 0.7968,
+      "step": 417
+    },
+    {
+      "epoch": 0.9858490566037735,
+      "grad_norm": 0.8856697438669298,
+      "learning_rate": 1.1256456905268263e-06,
+      "loss": 0.7622,
+      "step": 418
+    },
+    {
+      "epoch": 0.9882075471698113,
+      "grad_norm": 0.8362007813195715,
+      "learning_rate": 1.1217390927447225e-06,
+      "loss": 0.7475,
+      "step": 419
+    },
+    {
+      "epoch": 0.9905660377358491,
+      "grad_norm": 0.7824485352158383,
+      "learning_rate": 1.1178306081730664e-06,
+      "loss": 0.7705,
+      "step": 420
+    },
+    {
+      "epoch": 0.9929245283018868,
+      "grad_norm": 0.7992205810856007,
+      "learning_rate": 1.113920297388028e-06,
+      "loss": 0.7545,
+      "step": 421
+    },
+    {
+      "epoch": 0.9952830188679245,
+      "grad_norm": 0.785520038953663,
+      "learning_rate": 1.1100082209940793e-06,
+      "loss": 0.7647,
+      "step": 422
+    },
+    {
+      "epoch": 0.9976415094339622,
+      "grad_norm": 0.8327122327201606,
+      "learning_rate": 1.106094439623058e-06,
+      "loss": 0.7632,
+      "step": 423
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.7736149741657676,
+      "learning_rate": 1.1021790139332264e-06,
+      "loss": 0.7861,
+      "step": 424
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.6892534494400024,
+      "eval_runtime": 81.9429,
+      "eval_samples_per_second": 8.604,
+      "eval_steps_per_second": 0.549,
+      "step": 424
+    },
+    {
+      "epoch": 1.0023584905660377,
+      "grad_norm": 0.8035685998252425,
+      "learning_rate": 1.0982620046083305e-06,
+      "loss": 0.7506,
+      "step": 425
+    },
+    {
+      "epoch": 1.0047169811320755,
+      "grad_norm": 1.001273686570143,
+      "learning_rate": 1.0943434723566623e-06,
+      "loss": 0.783,
+      "step": 426
+    },
+    {
+      "epoch": 1.0070754716981132,
+      "grad_norm": 0.7620350953234064,
+      "learning_rate": 1.0904234779101158e-06,
+      "loss": 0.7436,
+      "step": 427
+    },
+    {
+      "epoch": 1.009433962264151,
+      "grad_norm": 0.8051824532909724,
+      "learning_rate": 1.0865020820232466e-06,
+      "loss": 0.7408,
+      "step": 428
+    },
+    {
+      "epoch": 1.0117924528301887,
+      "grad_norm": 1.1441709749307019,
+      "learning_rate": 1.0825793454723324e-06,
+      "loss": 0.7619,
+      "step": 429
+    },
+    {
+      "epoch": 1.0141509433962264,
+      "grad_norm": 0.934287576351282,
+      "learning_rate": 1.0786553290544278e-06,
+      "loss": 0.7291,
+      "step": 430
+    },
+    {
+      "epoch": 1.0165094339622642,
+      "grad_norm": 0.867146390883491,
+      "learning_rate": 1.0747300935864243e-06,
+      "loss": 0.7697,
+      "step": 431
+    },
+    {
+      "epoch": 1.0188679245283019,
+      "grad_norm": 0.7934743429540405,
+      "learning_rate": 1.070803699904107e-06,
+      "loss": 0.7774,
+      "step": 432
+    },
+    {
+      "epoch": 1.0212264150943395,
+      "grad_norm": 1.204890281675565,
+      "learning_rate": 1.0668762088612114e-06,
+      "loss": 0.7659,
+      "step": 433
+    },
+    {
+      "epoch": 1.0235849056603774,
+      "grad_norm": 0.7748660244184836,
+      "learning_rate": 1.0629476813284807e-06,
+      "loss": 0.7812,
+      "step": 434
+    },
+    {
+      "epoch": 1.025943396226415,
+      "grad_norm": 0.8559552505505363,
+      "learning_rate": 1.0590181781927227e-06,
+      "loss": 0.7641,
+      "step": 435
+    },
+    {
+      "epoch": 1.028301886792453,
+      "grad_norm": 0.8278708288686452,
+      "learning_rate": 1.0550877603558654e-06,
+      "loss": 0.7339,
+      "step": 436
+    },
+    {
+      "epoch": 1.0306603773584906,
+      "grad_norm": 0.8199114258570241,
+      "learning_rate": 1.0511564887340135e-06,
+      "loss": 0.7884,
+      "step": 437
+    },
+    {
+      "epoch": 1.0330188679245282,
+      "grad_norm": 0.8220578039981541,
+      "learning_rate": 1.0472244242565034e-06,
+      "loss": 0.7636,
+      "step": 438
+    },
+    {
+      "epoch": 1.0353773584905661,
+      "grad_norm": 0.9899659514811034,
+      "learning_rate": 1.043291627864961e-06,
+      "loss": 0.7507,
+      "step": 439
+    },
+    {
+      "epoch": 1.0377358490566038,
+      "grad_norm": 0.9164260231900845,
+      "learning_rate": 1.0393581605123552e-06,
+      "loss": 0.7604,
+      "step": 440
+    },
+    {
+      "epoch": 1.0400943396226414,
+      "grad_norm": 0.780514004370331,
+      "learning_rate": 1.035424083162054e-06,
+      "loss": 0.7661,
+      "step": 441
+    },
+    {
+      "epoch": 1.0424528301886793,
+      "grad_norm": 0.9654942156178127,
+      "learning_rate": 1.031489456786879e-06,
+      "loss": 0.732,
+      "step": 442
+    },
+    {
+      "epoch": 1.044811320754717,
+      "grad_norm": 0.9028629851443349,
+      "learning_rate": 1.0275543423681621e-06,
+      "loss": 0.7704,
+      "step": 443
+    },
+    {
+      "epoch": 1.0471698113207548,
+      "grad_norm": 0.8997990850767794,
+      "learning_rate": 1.0236188008947978e-06,
+      "loss": 0.7832,
+      "step": 444
+    },
+    {
+      "epoch": 1.0495283018867925,
+      "grad_norm": 0.8380399486601815,
+      "learning_rate": 1.0196828933623008e-06,
+      "loss": 0.7513,
+      "step": 445
+    },
+    {
+      "epoch": 1.0518867924528301,
+      "grad_norm": 0.818805158462816,
+      "learning_rate": 1.0157466807718577e-06,
+      "loss": 0.7595,
+      "step": 446
+    },
+    {
+      "epoch": 1.054245283018868,
+      "grad_norm": 0.8325448461017404,
+      "learning_rate": 1.0118102241293847e-06,
+      "loss": 0.7634,
+      "step": 447
+    },
+    {
+      "epoch": 1.0566037735849056,
+      "grad_norm": 0.7936938025099423,
+      "learning_rate": 1.0078735844445788e-06,
+      "loss": 0.768,
+      "step": 448
+    },
+    {
+      "epoch": 1.0589622641509433,
+      "grad_norm": 0.8378311364099181,
+      "learning_rate": 1.0039368227299753e-06,
+      "loss": 0.7599,
+      "step": 449
+    },
+    {
+      "epoch": 1.0613207547169812,
+      "grad_norm": 0.8299911067496856,
+      "learning_rate": 1e-06,
+      "loss": 0.7488,
+      "step": 450
+    },
+    {
+      "epoch": 1.0636792452830188,
+      "grad_norm": 0.8677844259185759,
+      "learning_rate": 9.960631772700248e-07,
+      "loss": 0.7698,
+      "step": 451
+    },
+    {
+      "epoch": 1.0660377358490567,
+      "grad_norm": 0.8463491375440975,
+      "learning_rate": 9.92126415555421e-07,
+      "loss": 0.7493,
+      "step": 452
+    },
+    {
+      "epoch": 1.0683962264150944,
+      "grad_norm": 0.8004441596711467,
+      "learning_rate": 9.881897758706154e-07,
+      "loss": 0.7366,
+      "step": 453
+    },
+    {
+      "epoch": 1.070754716981132,
+      "grad_norm": 1.2321632539336993,
+      "learning_rate": 9.842533192281422e-07,
+      "loss": 0.7481,
+      "step": 454
+    },
+    {
+      "epoch": 1.0731132075471699,
+      "grad_norm": 0.8194799035445371,
+      "learning_rate": 9.803171066376995e-07,
+      "loss": 0.7834,
+      "step": 455
+    },
+    {
+      "epoch": 1.0754716981132075,
+      "grad_norm": 0.7953471718099565,
+      "learning_rate": 9.763811991052019e-07,
+      "loss": 0.7508,
+      "step": 456
+    },
+    {
+      "epoch": 1.0778301886792452,
+      "grad_norm": 0.8500817934393053,
+      "learning_rate": 9.72445657631838e-07,
+      "loss": 0.7639,
+      "step": 457
+    },
+    {
+      "epoch": 1.080188679245283,
+      "grad_norm": 0.8053711636082244,
+      "learning_rate": 9.68510543213121e-07,
+      "loss": 0.7546,
+      "step": 458
+    },
+    {
+      "epoch": 1.0825471698113207,
+      "grad_norm": 1.1272698654628237,
+      "learning_rate": 9.645759168379461e-07,
+      "loss": 0.735,
+      "step": 459
+    },
+    {
+      "epoch": 1.0849056603773586,
+      "grad_norm": 0.7772007598787661,
+      "learning_rate": 9.606418394876447e-07,
+      "loss": 0.747,
+      "step": 460
+    },
+    {
+      "epoch": 1.0872641509433962,
+      "grad_norm": 0.8414360815423256,
+      "learning_rate": 9.567083721350388e-07,
+      "loss": 0.7411,
+      "step": 461
+    },
+    {
+      "epoch": 1.0896226415094339,
+      "grad_norm": 0.7791267607388218,
+      "learning_rate": 9.527755757434966e-07,
+      "loss": 0.7436,
+      "step": 462
+    },
+    {
+      "epoch": 1.0919811320754718,
+      "grad_norm": 0.80528466203603,
+      "learning_rate": 9.48843511265987e-07,
+      "loss": 0.7656,
+      "step": 463
+    },
+    {
+      "epoch": 1.0943396226415094,
+      "grad_norm": 0.7758351013188943,
+      "learning_rate": 9.449122396441343e-07,
+      "loss": 0.737,
+      "step": 464
+    },
+    {
+      "epoch": 1.0966981132075473,
+      "grad_norm": 1.2494162832647924,
+      "learning_rate": 9.409818218072772e-07,
+      "loss": 0.7613,
+      "step": 465
+    },
+    {
+      "epoch": 1.099056603773585,
+      "grad_norm": 0.8049303948390758,
+      "learning_rate": 9.370523186715194e-07,
+      "loss": 0.7628,
+      "step": 466
+    },
+    {
+      "epoch": 1.1014150943396226,
+      "grad_norm": 0.7896423211268341,
+      "learning_rate": 9.331237911387888e-07,
+      "loss": 0.7301,
+      "step": 467
+    },
+    {
+      "epoch": 1.1037735849056605,
+      "grad_norm": 0.7947232451183359,
+      "learning_rate": 9.291963000958931e-07,
+      "loss": 0.7803,
+      "step": 468
+    },
+    {
+      "epoch": 1.1061320754716981,
+      "grad_norm": 0.8128944094386984,
+      "learning_rate": 9.252699064135758e-07,
+      "loss": 0.7693,
+      "step": 469
+    },
+    {
+      "epoch": 1.1084905660377358,
+      "grad_norm": 0.9181658602259837,
+      "learning_rate": 9.213446709455722e-07,
+      "loss": 0.7794,
+      "step": 470
+    },
+    {
+      "epoch": 1.1108490566037736,
+      "grad_norm": 0.753282034988688,
+      "learning_rate": 9.174206545276677e-07,
+      "loss": 0.7383,
+      "step": 471
+    },
+    {
+      "epoch": 1.1132075471698113,
+      "grad_norm": 0.8890774956158611,
+      "learning_rate": 9.134979179767532e-07,
+      "loss": 0.7528,
+      "step": 472
+    },
+    {
+      "epoch": 1.115566037735849,
+      "grad_norm": 0.9503365963029163,
+      "learning_rate": 9.095765220898843e-07,
+      "loss": 0.7757,
+      "step": 473
+    },
+    {
+      "epoch": 1.1179245283018868,
+      "grad_norm": 0.9284151146362815,
+      "learning_rate": 9.056565276433377e-07,
+      "loss": 0.7442,
+      "step": 474
+    },
+    {
+      "epoch": 1.1202830188679245,
+      "grad_norm": 0.9704301526052668,
+      "learning_rate": 9.017379953916695e-07,
+      "loss": 0.7417,
+      "step": 475
+    },
+    {
+      "epoch": 1.1226415094339623,
+      "grad_norm": 0.8572079632472874,
+      "learning_rate": 8.978209860667738e-07,
+      "loss": 0.7724,
+      "step": 476
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 0.7912726011032718,
+      "learning_rate": 8.93905560376942e-07,
+      "loss": 0.7741,
+      "step": 477
+    },
+    {
+      "epoch": 1.1273584905660377,
+      "grad_norm": 0.7762605349979439,
+      "learning_rate": 8.899917790059207e-07,
+      "loss": 0.7211,
+      "step": 478
+    },
+    {
+      "epoch": 1.1297169811320755,
+      "grad_norm": 0.8625085844504403,
+      "learning_rate": 8.860797026119721e-07,
+      "loss": 0.7327,
+      "step": 479
+    },
+    {
+      "epoch": 1.1320754716981132,
+      "grad_norm": 0.8012017130865744,
+      "learning_rate": 8.821693918269333e-07,
+      "loss": 0.7293,
+      "step": 480
+    },
+    {
+      "epoch": 1.134433962264151,
+      "grad_norm": 0.7832218701492647,
+      "learning_rate": 8.782609072552776e-07,
+      "loss": 0.7584,
+      "step": 481
+    },
+    {
+      "epoch": 1.1367924528301887,
+      "grad_norm": 0.8223365605395953,
+      "learning_rate": 8.743543094731737e-07,
+      "loss": 0.757,
+      "step": 482
+    },
+    {
+      "epoch": 1.1391509433962264,
+      "grad_norm": 0.769269936037437,
+      "learning_rate": 8.704496590275477e-07,
+      "loss": 0.7265,
+      "step": 483
+    },
+    {
+      "epoch": 1.1415094339622642,
+      "grad_norm": 0.8470727703331676,
+      "learning_rate": 8.665470164351447e-07,
+      "loss": 0.7446,
+      "step": 484
+    },
+    {
+      "epoch": 1.1438679245283019,
+      "grad_norm": 0.8609116334367811,
+      "learning_rate": 8.626464421815918e-07,
+      "loss": 0.7357,
+      "step": 485
+    },
+    {
+      "epoch": 1.1462264150943395,
+      "grad_norm": 0.7644474626055862,
+      "learning_rate": 8.587479967204582e-07,
+      "loss": 0.7502,
+      "step": 486
+    },
+    {
+      "epoch": 1.1485849056603774,
+      "grad_norm": 0.7974217428241436,
+      "learning_rate": 8.548517404723206e-07,
+      "loss": 0.7255,
+      "step": 487
+    },
+    {
+      "epoch": 1.150943396226415,
+      "grad_norm": 0.8124702999828396,
+      "learning_rate": 8.509577338238254e-07,
+      "loss": 0.766,
+      "step": 488
+    },
+    {
+      "epoch": 1.1533018867924527,
+      "grad_norm": 0.8292245412487236,
+      "learning_rate": 8.47066037126754e-07,
+      "loss": 0.7302,
+      "step": 489
+    },
+    {
+      "epoch": 1.1556603773584906,
+      "grad_norm": 0.819810501024351,
+      "learning_rate": 8.43176710697086e-07,
+      "loss": 0.7163,
+      "step": 490
+    },
+    {
+      "epoch": 1.1580188679245282,
+      "grad_norm": 0.8385443524556274,
+      "learning_rate": 8.392898148140656e-07,
+      "loss": 0.7465,
+      "step": 491
+    },
+    {
+      "epoch": 1.1603773584905661,
+      "grad_norm": 0.778726751410171,
+      "learning_rate": 8.354054097192659e-07,
+      "loss": 0.7758,
+      "step": 492
+    },
+    {
+      "epoch": 1.1627358490566038,
+      "grad_norm": 0.788415584362694,
+      "learning_rate": 8.315235556156579e-07,
+      "loss": 0.7498,
+      "step": 493
+    },
+    {
+      "epoch": 1.1650943396226414,
+      "grad_norm": 0.7703907755533744,
+      "learning_rate": 8.276443126666742e-07,
+      "loss": 0.7276,
+      "step": 494
+    },
+    {
+      "epoch": 1.1674528301886793,
+      "grad_norm": 0.8704987882807484,
+      "learning_rate": 8.237677409952784e-07,
+      "loss": 0.766,
+      "step": 495
+    },
+    {
+      "epoch": 1.169811320754717,
+      "grad_norm": 0.7948667823022109,
+      "learning_rate": 8.198939006830333e-07,
+      "loss": 0.7656,
+      "step": 496
+    },
+    {
+      "epoch": 1.1721698113207548,
+      "grad_norm": 0.8090640349585008,
+      "learning_rate": 8.160228517691692e-07,
+      "loss": 0.7257,
+      "step": 497
+    },
+    {
+      "epoch": 1.1745283018867925,
+      "grad_norm": 1.159992161506865,
+      "learning_rate": 8.121546542496538e-07,
+      "loss": 0.7593,
+      "step": 498
+    },
+    {
+      "epoch": 1.1768867924528301,
+      "grad_norm": 0.8230774285047441,
+      "learning_rate": 8.082893680762618e-07,
+      "loss": 0.7511,
+      "step": 499
+    },
+    {
+      "epoch": 1.179245283018868,
+      "grad_norm": 0.8682718782355334,
+      "learning_rate": 8.044270531556452e-07,
+      "loss": 0.7647,
+      "step": 500
+    },
+    {
+      "epoch": 1.1816037735849056,
+      "grad_norm": 0.8274347964467509,
+      "learning_rate": 8.005677693484076e-07,
+      "loss": 0.7536,
+      "step": 501
+    },
+    {
+      "epoch": 1.1839622641509433,
+      "grad_norm": 0.8334334904558071,
+      "learning_rate": 7.967115764681731e-07,
+      "loss": 0.782,
+      "step": 502
+    },
+    {
+      "epoch": 1.1863207547169812,
+      "grad_norm": 0.8985376970213992,
+      "learning_rate": 7.928585342806607e-07,
+      "loss": 0.7461,
+      "step": 503
+    },
+    {
+      "epoch": 1.1886792452830188,
+      "grad_norm": 0.7846783242941104,
+      "learning_rate": 7.890087025027579e-07,
+      "loss": 0.7461,
+      "step": 504
+    },
+    {
+      "epoch": 1.1910377358490567,
+      "grad_norm": 0.800897718044774,
+      "learning_rate": 7.85162140801596e-07,
+      "loss": 0.7525,
+      "step": 505
+    },
+    {
+      "epoch": 1.1933962264150944,
+      "grad_norm": 0.7742939894445084,
+      "learning_rate": 7.813189087936242e-07,
+      "loss": 0.7082,
+      "step": 506
+    },
+    {
+      "epoch": 1.195754716981132,
+      "grad_norm": 0.8945053331857878,
+      "learning_rate": 7.774790660436857e-07,
+      "loss": 0.7328,
+      "step": 507
+    },
+    {
+      "epoch": 1.1981132075471699,
+      "grad_norm": 0.9098394537893619,
+      "learning_rate": 7.736426720640948e-07,
+      "loss": 0.7802,
+      "step": 508
+    },
+    {
+      "epoch": 1.2004716981132075,
+      "grad_norm": 0.7969225975663269,
+      "learning_rate": 7.698097863137152e-07,
+      "loss": 0.7423,
+      "step": 509
+    },
+    {
+      "epoch": 1.2028301886792452,
+      "grad_norm": 0.8867765640390939,
+      "learning_rate": 7.659804681970377e-07,
+      "loss": 0.7443,
+      "step": 510
+    },
+    {
+      "epoch": 1.205188679245283,
+      "grad_norm": 0.8936277250989298,
+      "learning_rate": 7.621547770632589e-07,
+      "loss": 0.7446,
+      "step": 511
+    },
+    {
+      "epoch": 1.2075471698113207,
+      "grad_norm": 0.7650448661091079,
+      "learning_rate": 7.583327722053626e-07,
+      "loss": 0.7437,
+      "step": 512
+    },
+    {
+      "epoch": 1.2099056603773586,
+      "grad_norm": 0.7718047147502091,
+      "learning_rate": 7.545145128592008e-07,
+      "loss": 0.7507,
+      "step": 513
+    },
+    {
+      "epoch": 1.2122641509433962,
+      "grad_norm": 0.9572262834139598,
+      "learning_rate": 7.507000582025748e-07,
+      "loss": 0.7329,
+      "step": 514
+    },
+    {
+      "epoch": 1.2146226415094339,
+      "grad_norm": 0.8375942206336273,
+      "learning_rate": 7.468894673543181e-07,
+      "loss": 0.7393,
+      "step": 515
+    },
+    {
+      "epoch": 1.2169811320754718,
+      "grad_norm": 0.8300278395620482,
+      "learning_rate": 7.430827993733808e-07,
+      "loss": 0.729,
+      "step": 516
+    },
+    {
+      "epoch": 1.2193396226415094,
+      "grad_norm": 0.924932117248118,
+      "learning_rate": 7.39280113257914e-07,
+      "loss": 0.7418,
+      "step": 517
+    },
+    {
+      "epoch": 1.2216981132075473,
+      "grad_norm": 0.8751602815649364,
+      "learning_rate": 7.354814679443556e-07,
+      "loss": 0.7695,
+      "step": 518
+    },
+    {
+      "epoch": 1.224056603773585,
+      "grad_norm": 0.7891993502461758,
+      "learning_rate": 7.316869223065155e-07,
+      "loss": 0.7755,
+      "step": 519
+    },
+    {
+      "epoch": 1.2264150943396226,
+      "grad_norm": 0.8088776711707174,
+      "learning_rate": 7.278965351546648e-07,
+      "loss": 0.7572,
+      "step": 520
+    },
+    {
+      "epoch": 1.2287735849056605,
+      "grad_norm": 0.7734655717390242,
+      "learning_rate": 7.241103652346248e-07,
+      "loss": 0.731,
+      "step": 521
+    },
+    {
+      "epoch": 1.2311320754716981,
+      "grad_norm": 0.8075042544518887,
+      "learning_rate": 7.20328471226854e-07,
+      "loss": 0.7681,
+      "step": 522
+    },
+    {
+      "epoch": 1.2334905660377358,
+      "grad_norm": 0.8682085195199817,
+      "learning_rate": 7.165509117455406e-07,
+      "loss": 0.7869,
+      "step": 523
+    },
+    {
+      "epoch": 1.2358490566037736,
+      "grad_norm": 0.7941356815577187,
+      "learning_rate": 7.127777453376935e-07,
+      "loss": 0.7839,
+      "step": 524
+    },
+    {
+      "epoch": 1.2382075471698113,
+      "grad_norm": 0.797011639565925,
+      "learning_rate": 7.090090304822355e-07,
+      "loss": 0.7704,
+      "step": 525
+    },
+    {
+      "epoch": 1.240566037735849,
+      "grad_norm": 0.778511764914176,
+      "learning_rate": 7.052448255890957e-07,
+      "loss": 0.7684,
+      "step": 526
+    },
+    {
+      "epoch": 1.2429245283018868,
+      "grad_norm": 0.9154469456152399,
+      "learning_rate": 7.014851889983057e-07,
+      "loss": 0.747,
+      "step": 527
+    },
+    {
+      "epoch": 1.2452830188679245,
+      "grad_norm": 0.7896606577750244,
+      "learning_rate": 6.97730178979093e-07,
+      "loss": 0.7243,
+      "step": 528
+    },
+    {
+      "epoch": 1.2476415094339623,
+      "grad_norm": 0.8076356582882437,
+      "learning_rate": 6.939798537289825e-07,
+      "loss": 0.7879,
+      "step": 529
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.8038075591560918,
+      "learning_rate": 6.902342713728886e-07,
+      "loss": 0.7304,
+      "step": 530
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 0.6839449405670166,
+      "eval_runtime": 82.7741,
+      "eval_samples_per_second": 8.517,
+      "eval_steps_per_second": 0.544,
+      "step": 530
+    },
+    {
+      "epoch": 1.2523584905660377,
+      "grad_norm": 0.7543374055980275,
+      "learning_rate": 6.864934899622191e-07,
+      "loss": 0.7671,
+      "step": 531
+    },
+    {
+      "epoch": 1.2547169811320755,
+      "grad_norm": 0.7940675058566147,
+      "learning_rate": 6.827575674739725e-07,
+      "loss": 0.7509,
+      "step": 532
+    },
+    {
+      "epoch": 1.2570754716981132,
+      "grad_norm": 0.7898168085125467,
+      "learning_rate": 6.790265618098423e-07,
+      "loss": 0.7819,
+      "step": 533
+    },
+    {
+      "epoch": 1.259433962264151,
+      "grad_norm": 0.8290907307209445,
+      "learning_rate": 6.753005307953165e-07,
+      "loss": 0.7433,
+      "step": 534
+    },
+    {
+      "epoch": 1.2617924528301887,
+      "grad_norm": 0.8051377335366591,
+      "learning_rate": 6.715795321787836e-07,
+      "loss": 0.7583,
+      "step": 535
+    },
+    {
+      "epoch": 1.2641509433962264,
+      "grad_norm": 0.7964463188550859,
+      "learning_rate": 6.678636236306357e-07,
+      "loss": 0.7733,
+      "step": 536
+    },
+    {
+      "epoch": 1.2665094339622642,
+      "grad_norm": 1.0183118136986125,
+      "learning_rate": 6.641528627423774e-07,
+      "loss": 0.7648,
+      "step": 537
+    },
+    {
+      "epoch": 1.2688679245283019,
+      "grad_norm": 0.7893378901295127,
+      "learning_rate": 6.604473070257308e-07,
+      "loss": 0.7543,
+      "step": 538
+    },
+    {
+      "epoch": 1.2712264150943398,
+      "grad_norm": 0.962839563526039,
+      "learning_rate": 6.567470139117447e-07,
+      "loss": 0.7815,
+      "step": 539
+    },
+    {
+      "epoch": 1.2735849056603774,
+      "grad_norm": 0.779909096149425,
+      "learning_rate": 6.530520407499049e-07,
+      "loss": 0.7321,
+      "step": 540
+    },
+    {
+      "epoch": 1.275943396226415,
+      "grad_norm": 1.0157939390615558,
+      "learning_rate": 6.493624448072457e-07,
+      "loss": 0.7622,
+      "step": 541
+    },
+    {
+      "epoch": 1.2783018867924527,
+      "grad_norm": 0.9839155144434717,
+      "learning_rate": 6.456782832674613e-07,
+      "loss": 0.7498,
+      "step": 542
+    },
+    {
+      "epoch": 1.2806603773584906,
+      "grad_norm": 0.8287088174891809,
+      "learning_rate": 6.419996132300203e-07,
+      "loss": 0.7479,
+      "step": 543
+    },
+    {
+      "epoch": 1.2830188679245282,
+      "grad_norm": 0.8401170082742024,
+      "learning_rate": 6.383264917092801e-07,
+      "loss": 0.7398,
+      "step": 544
+    },
+    {
+      "epoch": 1.2853773584905661,
+      "grad_norm": 0.8649864952648277,
+      "learning_rate": 6.34658975633605e-07,
+      "loss": 0.7629,
+      "step": 545
+    },
+    {
+      "epoch": 1.2877358490566038,
+      "grad_norm": 0.7978032312823685,
+      "learning_rate": 6.30997121844481e-07,
+      "loss": 0.7455,
+      "step": 546
+    },
+    {
+      "epoch": 1.2900943396226414,
+      "grad_norm": 0.78443421199377,
+      "learning_rate": 6.273409870956369e-07,
+      "loss": 0.7512,
+      "step": 547
+    },
+    {
+      "epoch": 1.2924528301886793,
+      "grad_norm": 0.9742229809355082,
+      "learning_rate": 6.236906280521646e-07,
+      "loss": 0.7629,
+      "step": 548
+    },
+    {
+      "epoch": 1.294811320754717,
+      "grad_norm": 0.8605023916238441,
+      "learning_rate": 6.200461012896401e-07,
+      "loss": 0.7389,
+      "step": 549
+    },
+    {
+      "epoch": 1.2971698113207548,
+      "grad_norm": 0.7976345004555616,
+      "learning_rate": 6.164074632932472e-07,
+      "loss": 0.7675,
+      "step": 550
+    },
+    {
+      "epoch": 1.2995283018867925,
+      "grad_norm": 0.8766199870243025,
+      "learning_rate": 6.127747704569015e-07,
+      "loss": 0.7392,
+      "step": 551
+    },
+    {
+      "epoch": 1.3018867924528301,
+      "grad_norm": 0.8388074881151663,
+      "learning_rate": 6.091480790823771e-07,
+      "loss": 0.7594,
+      "step": 552
+    },
+    {
+      "epoch": 1.304245283018868,
+      "grad_norm": 0.7816816297148448,
+      "learning_rate": 6.055274453784338e-07,
+      "loss": 0.7754,
+      "step": 553
+    },
+    {
+      "epoch": 1.3066037735849056,
+      "grad_norm": 0.8051939468008078,
+      "learning_rate": 6.019129254599456e-07,
+      "loss": 0.7456,
+      "step": 554
+    },
+    {
+      "epoch": 1.3089622641509435,
+      "grad_norm": 0.7954437915413475,
+      "learning_rate": 5.983045753470307e-07,
+      "loss": 0.7804,
+      "step": 555
+    },
+    {
+      "epoch": 1.3113207547169812,
+      "grad_norm": 0.7995393304495568,
+      "learning_rate": 5.947024509641841e-07,
+      "loss": 0.7346,
+      "step": 556
+    },
+    {
+      "epoch": 1.3136792452830188,
+      "grad_norm": 0.7666328856774342,
+      "learning_rate": 5.911066081394112e-07,
+      "loss": 0.7641,
+      "step": 557
+    },
+    {
+      "epoch": 1.3160377358490565,
+      "grad_norm": 0.9135245763682434,
+      "learning_rate": 5.875171026033608e-07,
+      "loss": 0.7493,
+      "step": 558
+    },
+    {
+      "epoch": 1.3183962264150944,
+      "grad_norm": 0.7843209916172604,
+      "learning_rate": 5.839339899884628e-07,
+      "loss": 0.754,
+      "step": 559
+    },
+    {
+      "epoch": 1.320754716981132,
+      "grad_norm": 0.7798421629061796,
+      "learning_rate": 5.803573258280654e-07,
+      "loss": 0.7453,
+      "step": 560
+    },
+    {
+      "epoch": 1.3231132075471699,
+      "grad_norm": 0.8026516589712169,
+      "learning_rate": 5.76787165555575e-07,
+      "loss": 0.7791,
+      "step": 561
+    },
+    {
+      "epoch": 1.3254716981132075,
+      "grad_norm": 0.8161986502002226,
+      "learning_rate": 5.732235645035963e-07,
+      "loss": 0.7435,
+      "step": 562
+    },
+    {
+      "epoch": 1.3278301886792452,
+      "grad_norm": 0.7982394811556922,
+      "learning_rate": 5.696665779030749e-07,
+      "loss": 0.7312,
+      "step": 563
+    },
+    {
+      "epoch": 1.330188679245283,
+      "grad_norm": 0.8424060129449961,
+      "learning_rate": 5.661162608824419e-07,
+      "loss": 0.7209,
+      "step": 564
+    },
+    {
+      "epoch": 1.3325471698113207,
+      "grad_norm": 0.8393522976074216,
+      "learning_rate": 5.625726684667585e-07,
+      "loss": 0.7497,
+      "step": 565
+    },
+    {
+      "epoch": 1.3349056603773586,
+      "grad_norm": 0.8458006058371306,
+      "learning_rate": 5.590358555768642e-07,
+      "loss": 0.7415,
+      "step": 566
+    },
+    {
+      "epoch": 1.3372641509433962,
+      "grad_norm": 0.7731963506173513,
+      "learning_rate": 5.555058770285246e-07,
+      "loss": 0.7289,
+      "step": 567
+    },
+    {
+      "epoch": 1.3396226415094339,
+      "grad_norm": 0.919340196876583,
+      "learning_rate": 5.519827875315823e-07,
+      "loss": 0.7372,
+      "step": 568
+    },
+    {
+      "epoch": 1.3419811320754718,
+      "grad_norm": 0.8047418954501285,
+      "learning_rate": 5.484666416891108e-07,
+      "loss": 0.7854,
+      "step": 569
+    },
+    {
+      "epoch": 1.3443396226415094,
+      "grad_norm": 0.8420862799710107,
+      "learning_rate": 5.449574939965636e-07,
+      "loss": 0.7487,
+      "step": 570
+    },
+    {
+      "epoch": 1.3466981132075473,
+      "grad_norm": 0.862354237686273,
+      "learning_rate": 5.414553988409343e-07,
+      "loss": 0.7615,
+      "step": 571
+    },
+    {
+      "epoch": 1.349056603773585,
+      "grad_norm": 0.764673226355618,
+      "learning_rate": 5.379604104999109e-07,
+      "loss": 0.7402,
+      "step": 572
+    },
+    {
+      "epoch": 1.3514150943396226,
+      "grad_norm": 0.9229852280599696,
+      "learning_rate": 5.344725831410368e-07,
+      "loss": 0.746,
+      "step": 573
+    },
+    {
+      "epoch": 1.3537735849056602,
+      "grad_norm": 0.7749939321603743,
+      "learning_rate": 5.309919708208686e-07,
+      "loss": 0.7393,
+      "step": 574
+    },
+    {
+      "epoch": 1.3561320754716981,
+      "grad_norm": 1.0855469699332105,
+      "learning_rate": 5.275186274841404e-07,
+      "loss": 0.7292,
+      "step": 575
+    },
+    {
+      "epoch": 1.3584905660377358,
+      "grad_norm": 1.3079826302270743,
+      "learning_rate": 5.240526069629264e-07,
+      "loss": 0.7551,
+      "step": 576
+    },
+    {
+      "epoch": 1.3608490566037736,
+      "grad_norm": 0.8097304218867772,
+      "learning_rate": 5.205939629758078e-07,
+      "loss": 0.7554,
+      "step": 577
+    },
+    {
+      "epoch": 1.3632075471698113,
+      "grad_norm": 0.7629186235386943,
+      "learning_rate": 5.171427491270393e-07,
+      "loss": 0.7448,
+      "step": 578
+    },
+    {
+      "epoch": 1.365566037735849,
+      "grad_norm": 0.8231834244100196,
+      "learning_rate": 5.136990189057187e-07,
+      "loss": 0.7675,
+      "step": 579
+    },
+    {
+      "epoch": 1.3679245283018868,
+      "grad_norm": 0.7758543745731015,
+      "learning_rate": 5.102628256849575e-07,
+      "loss": 0.7643,
+      "step": 580
+    },
+    {
+      "epoch": 1.3702830188679245,
+      "grad_norm": 0.7915005392947032,
+      "learning_rate": 5.068342227210545e-07,
+      "loss": 0.732,
+      "step": 581
+    },
+    {
+      "epoch": 1.3726415094339623,
+      "grad_norm": 0.821233361833331,
+      "learning_rate": 5.034132631526695e-07,
+      "loss": 0.7145,
+      "step": 582
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 0.8288831867947671,
+      "learning_rate": 5.000000000000002e-07,
+      "loss": 0.7717,
+      "step": 583
+    },
+    {
+      "epoch": 1.3773584905660377,
+      "grad_norm": 0.9380885957065556,
+      "learning_rate": 4.9659448616396e-07,
+      "loss": 0.7655,
+      "step": 584
+    },
+    {
+      "epoch": 1.3797169811320755,
+      "grad_norm": 0.769393956851887,
+      "learning_rate": 4.9319677442536e-07,
+      "loss": 0.7481,
+      "step": 585
+    },
+    {
+      "epoch": 1.3820754716981132,
+      "grad_norm": 0.7624074760874653,
+      "learning_rate": 4.89806917444088e-07,
+      "loss": 0.7408,
+      "step": 586
+    },
+    {
+      "epoch": 1.384433962264151,
+      "grad_norm": 0.8836661905432738,
+      "learning_rate": 4.864249677582935e-07,
+      "loss": 0.7423,
+      "step": 587
+    },
+    {
+      "epoch": 1.3867924528301887,
+      "grad_norm": 1.7675171950723623,
+      "learning_rate": 4.830509777835744e-07,
+      "loss": 0.7414,
+      "step": 588
+    },
+    {
+      "epoch": 1.3891509433962264,
+      "grad_norm": 0.8297208161937744,
+      "learning_rate": 4.796849998121647e-07,
+      "loss": 0.7571,
+      "step": 589
+    },
+    {
+      "epoch": 1.3915094339622642,
+      "grad_norm": 0.8105952844684988,
+      "learning_rate": 4.7632708601212215e-07,
+      "loss": 0.7473,
+      "step": 590
+    },
+    {
+      "epoch": 1.3938679245283019,
+      "grad_norm": 0.943401465741068,
+      "learning_rate": 4.7297728842652116e-07,
+      "loss": 0.7885,
+      "step": 591
+    },
+    {
+      "epoch": 1.3962264150943398,
+      "grad_norm": 0.8407405967900344,
+      "learning_rate": 4.6963565897264623e-07,
+      "loss": 0.7768,
+      "step": 592
+    },
+    {
+      "epoch": 1.3985849056603774,
+      "grad_norm": 0.7742620223687944,
+      "learning_rate": 4.663022494411866e-07,
+      "loss": 0.7679,
+      "step": 593
+    },
+    {
+      "epoch": 1.400943396226415,
+      "grad_norm": 0.7984390061772534,
+      "learning_rate": 4.6297711149543405e-07,
+      "loss": 0.7503,
+      "step": 594
+    },
+    {
+      "epoch": 1.4033018867924527,
+      "grad_norm": 0.7687242314529935,
+      "learning_rate": 4.596602966704822e-07,
+      "loss": 0.7593,
+      "step": 595
+    },
+    {
+      "epoch": 1.4056603773584906,
+      "grad_norm": 0.7752552260078952,
+      "learning_rate": 4.5635185637242735e-07,
+      "loss": 0.7236,
+      "step": 596
+    },
+    {
+      "epoch": 1.4080188679245282,
+      "grad_norm": 0.7909608031012165,
+      "learning_rate": 4.530518418775733e-07,
+      "loss": 0.7487,
+      "step": 597
+    },
+    {
+      "epoch": 1.4103773584905661,
+      "grad_norm": 1.4293658648035819,
+      "learning_rate": 4.4976030433163337e-07,
+      "loss": 0.7619,
+      "step": 598
+    },
+    {
+      "epoch": 1.4127358490566038,
+      "grad_norm": 0.9601629598020603,
+      "learning_rate": 4.4647729474894123e-07,
+      "loss": 0.7301,
+      "step": 599
+    },
+    {
+      "epoch": 1.4150943396226414,
+      "grad_norm": 0.8524971604621474,
+      "learning_rate": 4.432028640116581e-07,
+      "loss": 0.7743,
+      "step": 600
+    },
+    {
+      "epoch": 1.4174528301886793,
+      "grad_norm": 0.7955465630229851,
+      "learning_rate": 4.399370628689857e-07,
+      "loss": 0.7549,
+      "step": 601
+    },
+    {
+      "epoch": 1.419811320754717,
+      "grad_norm": 0.8461044127574078,
+      "learning_rate": 4.366799419363779e-07,
+      "loss": 0.7923,
+      "step": 602
+    },
+    {
+      "epoch": 1.4221698113207548,
+      "grad_norm": 0.7959852549975004,
+      "learning_rate": 4.3343155169475797e-07,
+      "loss": 0.7313,
+      "step": 603
+    },
+    {
+      "epoch": 1.4245283018867925,
+      "grad_norm": 0.8115147827054038,
+      "learning_rate": 4.3019194248973377e-07,
+      "loss": 0.7353,
+      "step": 604
+    },
+    {
+      "epoch": 1.4268867924528301,
+      "grad_norm": 0.7884615566733161,
+      "learning_rate": 4.269611645308214e-07,
+      "loss": 0.7396,
+      "step": 605
+    },
+    {
+      "epoch": 1.429245283018868,
+      "grad_norm": 0.8709874327087314,
+      "learning_rate": 4.237392678906633e-07,
+      "loss": 0.7557,
+      "step": 606
+    },
+    {
+      "epoch": 1.4316037735849056,
+      "grad_norm": 0.9486373793118293,
+      "learning_rate": 4.205263025042538e-07,
+      "loss": 0.7397,
+      "step": 607
+    },
+    {
+      "epoch": 1.4339622641509435,
+      "grad_norm": 0.8162741407892988,
+      "learning_rate": 4.173223181681651e-07,
+      "loss": 0.7327,
+      "step": 608
+    },
+    {
+      "epoch": 1.4363207547169812,
+      "grad_norm": 0.8564525506081095,
+      "learning_rate": 4.141273645397754e-07,
+      "loss": 0.7745,
+      "step": 609
+    },
+    {
+      "epoch": 1.4386792452830188,
+      "grad_norm": 0.8291423725092533,
+      "learning_rate": 4.109414911364992e-07,
+      "loss": 0.761,
+      "step": 610
+    },
+    {
+      "epoch": 1.4410377358490565,
+      "grad_norm": 0.9110035252316134,
+      "learning_rate": 4.0776474733502007e-07,
+      "loss": 0.7227,
+      "step": 611
+    },
+    {
+      "epoch": 1.4433962264150944,
+      "grad_norm": 7.248992388941307,
+      "learning_rate": 4.045971823705249e-07,
+      "loss": 0.7515,
+      "step": 612
+    },
+    {
+      "epoch": 1.445754716981132,
+      "grad_norm": 0.737128592265643,
+      "learning_rate": 4.0143884533594197e-07,
+      "loss": 0.7308,
+      "step": 613
+    },
+    {
+      "epoch": 1.4481132075471699,
+      "grad_norm": 0.7878766896489899,
+      "learning_rate": 3.982897851811786e-07,
+      "loss": 0.7323,
+      "step": 614
+    },
+    {
+      "epoch": 1.4504716981132075,
+      "grad_norm": 0.7859241609615767,
+      "learning_rate": 3.951500507123627e-07,
+      "loss": 0.7814,
+      "step": 615
+    },
+    {
+      "epoch": 1.4528301886792452,
+      "grad_norm": 0.8312739174249127,
+      "learning_rate": 3.920196905910873e-07,
+      "loss": 0.7446,
+      "step": 616
+    },
+    {
+      "epoch": 1.455188679245283,
+      "grad_norm": 0.8160254075021761,
+      "learning_rate": 3.888987533336566e-07,
+      "loss": 0.745,
+      "step": 617
+    },
+    {
+      "epoch": 1.4575471698113207,
+      "grad_norm": 0.9271092228733877,
+      "learning_rate": 3.8578728731033214e-07,
+      "loss": 0.7875,
+      "step": 618
+    },
+    {
+      "epoch": 1.4599056603773586,
+      "grad_norm": 1.3059448042170938,
+      "learning_rate": 3.826853407445848e-07,
+      "loss": 0.7758,
+      "step": 619
+    },
+    {
+      "epoch": 1.4622641509433962,
+      "grad_norm": 0.7940556961073607,
+      "learning_rate": 3.795929617123468e-07,
+      "loss": 0.7289,
+      "step": 620
+    },
+    {
+      "epoch": 1.4646226415094339,
+      "grad_norm": 0.7908440739021586,
+      "learning_rate": 3.765101981412665e-07,
+      "loss": 0.7247,
+      "step": 621
+    },
+    {
+      "epoch": 1.4669811320754718,
+      "grad_norm": 0.807553140785391,
+      "learning_rate": 3.7343709780996614e-07,
+      "loss": 0.7365,
+      "step": 622
+    },
+    {
+      "epoch": 1.4693396226415094,
+      "grad_norm": 2.5050420828496787,
+      "learning_rate": 3.703737083473005e-07,
+      "loss": 0.7729,
+      "step": 623
+    },
+    {
+      "epoch": 1.4716981132075473,
+      "grad_norm": 0.8163112108481482,
+      "learning_rate": 3.673200772316193e-07,
+      "loss": 0.7269,
+      "step": 624
+    },
+    {
+      "epoch": 1.474056603773585,
+      "grad_norm": 0.8451238760005333,
+      "learning_rate": 3.6427625179003217e-07,
+      "loss": 0.7435,
+      "step": 625
+    },
+    {
+      "epoch": 1.4764150943396226,
+      "grad_norm": 1.0290141575096419,
+      "learning_rate": 3.6124227919767227e-07,
+      "loss": 0.7488,
+      "step": 626
+    },
+    {
+      "epoch": 1.4787735849056602,
+      "grad_norm": 0.7955100632691958,
+      "learning_rate": 3.5821820647696864e-07,
+      "loss": 0.7556,
+      "step": 627
+    },
+    {
+      "epoch": 1.4811320754716981,
+      "grad_norm": 0.8238206028160229,
+      "learning_rate": 3.552040804969149e-07,
+      "loss": 0.7541,
+      "step": 628
+    },
+    {
+      "epoch": 1.4834905660377358,
+      "grad_norm": 0.8519020092733177,
+      "learning_rate": 3.5219994797234476e-07,
+      "loss": 0.7783,
+      "step": 629
+    },
+    {
+      "epoch": 1.4858490566037736,
+      "grad_norm": 0.7845182991866497,
+      "learning_rate": 3.4920585546320625e-07,
+      "loss": 0.7457,
+      "step": 630
+    },
+    {
+      "epoch": 1.4882075471698113,
+      "grad_norm": 0.7643446889279868,
+      "learning_rate": 3.4622184937384156e-07,
+      "loss": 0.7637,
+      "step": 631
+    },
+    {
+      "epoch": 1.490566037735849,
+      "grad_norm": 0.8986522134788724,
+      "learning_rate": 3.4324797595226564e-07,
+      "loss": 0.7431,
+      "step": 632
+    },
+    {
+      "epoch": 1.4929245283018868,
+      "grad_norm": 0.788015002889114,
+      "learning_rate": 3.4028428128945286e-07,
+      "loss": 0.746,
+      "step": 633
+    },
+    {
+      "epoch": 1.4952830188679245,
+      "grad_norm": 0.7754842750109397,
+      "learning_rate": 3.3733081131861975e-07,
+      "loss": 0.7688,
+      "step": 634
+    },
+    {
+      "epoch": 1.4976415094339623,
+      "grad_norm": 0.8009965193531273,
+      "learning_rate": 3.343876118145141e-07,
+      "loss": 0.7501,
+      "step": 635
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 1.006640510210976,
+      "learning_rate": 3.314547283927057e-07,
+      "loss": 0.7239,
+      "step": 636
+    },
+    {
+      "epoch": 1.5,
+      "eval_loss": 0.6808757185935974,
+      "eval_runtime": 83.2321,
+      "eval_samples_per_second": 8.47,
+      "eval_steps_per_second": 0.541,
+      "step": 636
+    },
+    {
+      "epoch": 1.5023584905660377,
+      "grad_norm": 0.8598327085719165,
+      "learning_rate": 3.2853220650887913e-07,
+      "loss": 0.7149,
+      "step": 637
+    },
+    {
+      "epoch": 1.5047169811320755,
+      "grad_norm": 0.8482092774985639,
+      "learning_rate": 3.256200914581292e-07,
+      "loss": 0.7556,
+      "step": 638
+    },
+    {
+      "epoch": 1.5070754716981132,
+      "grad_norm": 0.7837630460537407,
+      "learning_rate": 3.227184283742591e-07,
+      "loss": 0.7432,
+      "step": 639
+    },
+    {
+      "epoch": 1.509433962264151,
+      "grad_norm": 0.8392002898875265,
+      "learning_rate": 3.198272622290804e-07,
+      "loss": 0.7395,
+      "step": 640
+    },
+    {
+      "epoch": 1.5117924528301887,
+      "grad_norm": 0.826051018451035,
+      "learning_rate": 3.169466378317177e-07,
+      "loss": 0.7849,
+      "step": 641
+    },
+    {
+      "epoch": 1.5141509433962264,
+      "grad_norm": 1.0059361369860251,
+      "learning_rate": 3.1407659982791204e-07,
+      "loss": 0.7541,
+      "step": 642
+    },
+    {
+      "epoch": 1.516509433962264,
+      "grad_norm": 0.8092490329577119,
+      "learning_rate": 3.112171926993291e-07,
+      "loss": 0.7348,
+      "step": 643
+    },
+    {
+      "epoch": 1.5188679245283019,
+      "grad_norm": 0.7280559308125196,
+      "learning_rate": 3.0836846076287146e-07,
+      "loss": 0.6873,
+      "step": 644
+    },
+    {
+      "epoch": 1.5212264150943398,
+      "grad_norm": 0.8517206567349668,
+      "learning_rate": 3.055304481699913e-07,
+      "loss": 0.7839,
+      "step": 645
+    },
+    {
+      "epoch": 1.5235849056603774,
+      "grad_norm": 0.8340097914857253,
+      "learning_rate": 3.027031989060046e-07,
+      "loss": 0.7561,
+      "step": 646
+    },
+    {
+      "epoch": 1.525943396226415,
+      "grad_norm": 0.8002849424485019,
+      "learning_rate": 2.998867567894108e-07,
+      "loss": 0.7747,
+      "step": 647
+    },
+    {
+      "epoch": 1.5283018867924527,
+      "grad_norm": 0.8623306309909063,
+      "learning_rate": 2.970811654712133e-07,
+      "loss": 0.7337,
+      "step": 648
+    },
+    {
+      "epoch": 1.5306603773584906,
+      "grad_norm": 0.8335419959901207,
+      "learning_rate": 2.942864684342432e-07,
+      "loss": 0.741,
+      "step": 649
+    },
+    {
+      "epoch": 1.5330188679245285,
+      "grad_norm": 0.8731062418731657,
+      "learning_rate": 2.91502708992485e-07,
+      "loss": 0.7424,
+      "step": 650
+    },
+    {
+      "epoch": 1.5353773584905661,
+      "grad_norm": 0.819811746630834,
+      "learning_rate": 2.8872993029040506e-07,
+      "loss": 0.7355,
+      "step": 651
+    },
+    {
+      "epoch": 1.5377358490566038,
+      "grad_norm": 0.8074813557073929,
+      "learning_rate": 2.859681753022838e-07,
+      "loss": 0.7594,
+      "step": 652
+    },
+    {
+      "epoch": 1.5400943396226414,
+      "grad_norm": 0.820619090005774,
+      "learning_rate": 2.8321748683154887e-07,
+      "loss": 0.7443,
+      "step": 653
+    },
+    {
+      "epoch": 1.5424528301886793,
+      "grad_norm": 0.8554565499942881,
+      "learning_rate": 2.8047790751011216e-07,
+      "loss": 0.6991,
+      "step": 654
+    },
+    {
+      "epoch": 1.544811320754717,
+      "grad_norm": 0.7784690870870672,
+      "learning_rate": 2.777494797977088e-07,
+      "loss": 0.7328,
+      "step": 655
+    },
+    {
+      "epoch": 1.5471698113207548,
+      "grad_norm": 1.7923607795927756,
+      "learning_rate": 2.7503224598123895e-07,
+      "loss": 0.7266,
+      "step": 656
+    },
+    {
+      "epoch": 1.5495283018867925,
+      "grad_norm": 0.7677443512564135,
+      "learning_rate": 2.7232624817411376e-07,
+      "loss": 0.737,
+      "step": 657
+    },
+    {
+      "epoch": 1.5518867924528301,
+      "grad_norm": 0.8055758349794541,
+      "learning_rate": 2.6963152831560066e-07,
+      "loss": 0.7444,
+      "step": 658
+    },
+    {
+      "epoch": 1.5542452830188678,
+      "grad_norm": 0.9008016637756693,
+      "learning_rate": 2.6694812817017387e-07,
+      "loss": 0.7128,
+      "step": 659
+    },
+    {
+      "epoch": 1.5566037735849056,
+      "grad_norm": 0.8106690468627914,
+      "learning_rate": 2.642760893268684e-07,
+      "loss": 0.7457,
+      "step": 660
+    },
+    {
+      "epoch": 1.5589622641509435,
+      "grad_norm": 1.027228042593183,
+      "learning_rate": 2.616154531986345e-07,
+      "loss": 0.7251,
+      "step": 661
+    },
+    {
+      "epoch": 1.5613207547169812,
+      "grad_norm": 0.7957920778904417,
+      "learning_rate": 2.5896626102169594e-07,
+      "loss": 0.7481,
+      "step": 662
+    },
+    {
+      "epoch": 1.5636792452830188,
+      "grad_norm": 0.7514802697133819,
+      "learning_rate": 2.5632855385491037e-07,
+      "loss": 0.757,
+      "step": 663
+    },
+    {
+      "epoch": 1.5660377358490565,
+      "grad_norm": 0.8814866050056973,
+      "learning_rate": 2.53702372579134e-07,
+      "loss": 0.7361,
+      "step": 664
+    },
+    {
+      "epoch": 1.5683962264150944,
+      "grad_norm": 0.8346312832230348,
+      "learning_rate": 2.51087757896587e-07,
+      "loss": 0.7378,
+      "step": 665
+    },
+    {
+      "epoch": 1.5707547169811322,
+      "grad_norm": 0.8164772219257521,
+      "learning_rate": 2.4848475033022377e-07,
+      "loss": 0.7631,
+      "step": 666
+    },
+    {
+      "epoch": 1.5731132075471699,
+      "grad_norm": 0.7933267558316177,
+      "learning_rate": 2.458933902231038e-07,
+      "loss": 0.7288,
+      "step": 667
+    },
+    {
+      "epoch": 1.5754716981132075,
+      "grad_norm": 0.8420080383383726,
+      "learning_rate": 2.4331371773776687e-07,
+      "loss": 0.77,
+      "step": 668
+    },
+    {
+      "epoch": 1.5778301886792452,
+      "grad_norm": 5.289720197423799,
+      "learning_rate": 2.407457728556115e-07,
+      "loss": 0.7184,
+      "step": 669
+    },
+    {
+      "epoch": 1.580188679245283,
+      "grad_norm": 1.0842649090634284,
+      "learning_rate": 2.3818959537627282e-07,
+      "loss": 0.7383,
+      "step": 670
+    },
+    {
+      "epoch": 1.5825471698113207,
+      "grad_norm": 0.8519744691939684,
+      "learning_rate": 2.3564522491700833e-07,
+      "loss": 0.769,
+      "step": 671
+    },
+    {
+      "epoch": 1.5849056603773586,
+      "grad_norm": 0.8168668326966967,
+      "learning_rate": 2.3311270091208256e-07,
+      "loss": 0.7444,
+      "step": 672
+    },
+    {
+      "epoch": 1.5872641509433962,
+      "grad_norm": 0.7514913975722562,
+      "learning_rate": 2.3059206261215668e-07,
+      "loss": 0.691,
+      "step": 673
+    },
+    {
+      "epoch": 1.5896226415094339,
+      "grad_norm": 0.7519874181403705,
+      "learning_rate": 2.2808334908367909e-07,
+      "loss": 0.7425,
+      "step": 674
+    },
+    {
+      "epoch": 1.5919811320754715,
+      "grad_norm": 0.8353526410754343,
+      "learning_rate": 2.2558659920828095e-07,
+      "loss": 0.7616,
+      "step": 675
+    },
+    {
+      "epoch": 1.5943396226415094,
+      "grad_norm": 0.7738579301027763,
+      "learning_rate": 2.2310185168217212e-07,
+      "loss": 0.7756,
+      "step": 676
+    },
+    {
+      "epoch": 1.5966981132075473,
+      "grad_norm": 0.813046680487735,
+      "learning_rate": 2.206291450155441e-07,
+      "loss": 0.7675,
+      "step": 677
+    },
+    {
+      "epoch": 1.599056603773585,
+      "grad_norm": 0.7815916852088219,
+      "learning_rate": 2.181685175319702e-07,
+      "loss": 0.7459,
+      "step": 678
+    },
+    {
+      "epoch": 1.6014150943396226,
+      "grad_norm": 0.7482293961455181,
+      "learning_rate": 2.157200073678137e-07,
+      "loss": 0.7237,
+      "step": 679
+    },
+    {
+      "epoch": 1.6037735849056602,
+      "grad_norm": 0.9475759536537903,
+      "learning_rate": 2.132836524716355e-07,
+      "loss": 0.718,
+      "step": 680
+    },
+    {
+      "epoch": 1.6061320754716981,
+      "grad_norm": 0.8731724054191227,
+      "learning_rate": 2.1085949060360653e-07,
+      "loss": 0.7344,
+      "step": 681
+    },
+    {
+      "epoch": 1.608490566037736,
+      "grad_norm": 0.8124879077769539,
+      "learning_rate": 2.0844755933492263e-07,
+      "loss": 0.7515,
+      "step": 682
+    },
+    {
+      "epoch": 1.6108490566037736,
+      "grad_norm": 0.7772794085957034,
+      "learning_rate": 2.0604789604722205e-07,
+      "loss": 0.7471,
+      "step": 683
+    },
+    {
+      "epoch": 1.6132075471698113,
+      "grad_norm": 0.8144653370194169,
+      "learning_rate": 2.0366053793200565e-07,
+      "loss": 0.7724,
+      "step": 684
+    },
+    {
+      "epoch": 1.615566037735849,
+      "grad_norm": 1.1024190577460233,
+      "learning_rate": 2.0128552199006198e-07,
+      "loss": 0.7389,
+      "step": 685
+    },
+    {
+      "epoch": 1.6179245283018868,
+      "grad_norm": 0.7813510804481604,
+      "learning_rate": 1.9892288503089205e-07,
+      "loss": 0.7688,
+      "step": 686
+    },
+    {
+      "epoch": 1.6202830188679245,
+      "grad_norm": 0.7649385235808642,
+      "learning_rate": 1.9657266367213898e-07,
+      "loss": 0.7279,
+      "step": 687
+    },
+    {
+      "epoch": 1.6226415094339623,
+      "grad_norm": 0.8664030315613933,
+      "learning_rate": 1.9423489433902184e-07,
+      "loss": 0.7604,
+      "step": 688
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 0.8134285929715391,
+      "learning_rate": 1.9190961326377053e-07,
+      "loss": 0.7628,
+      "step": 689
+    },
+    {
+      "epoch": 1.6273584905660377,
+      "grad_norm": 0.8126410627826297,
+      "learning_rate": 1.8959685648506362e-07,
+      "loss": 0.7479,
+      "step": 690
+    },
+    {
+      "epoch": 1.6297169811320755,
+      "grad_norm": 0.8243887159653667,
+      "learning_rate": 1.8729665984747e-07,
+      "loss": 0.743,
+      "step": 691
+    },
+    {
+      "epoch": 1.6320754716981132,
+      "grad_norm": 0.8107223249523531,
+      "learning_rate": 1.8500905900089403e-07,
+      "loss": 0.7601,
+      "step": 692
+    },
+    {
+      "epoch": 1.634433962264151,
+      "grad_norm": 0.8038068007057135,
+      "learning_rate": 1.82734089400022e-07,
+      "loss": 0.7643,
+      "step": 693
+    },
+    {
+      "epoch": 1.6367924528301887,
+      "grad_norm": 0.9712605139438796,
+      "learning_rate": 1.804717863037737e-07,
+      "loss": 0.7445,
+      "step": 694
+    },
+    {
+      "epoch": 1.6391509433962264,
+      "grad_norm": 0.7879085057169681,
+      "learning_rate": 1.7822218477475494e-07,
+      "loss": 0.7485,
+      "step": 695
+    },
+    {
+      "epoch": 1.641509433962264,
+      "grad_norm": 0.8550428267297568,
+      "learning_rate": 1.7598531967871465e-07,
+      "loss": 0.7783,
+      "step": 696
+    },
+    {
+      "epoch": 1.6438679245283019,
+      "grad_norm": 0.7795732830836477,
+      "learning_rate": 1.737612256840053e-07,
+      "loss": 0.7484,
+      "step": 697
+    },
+    {
+      "epoch": 1.6462264150943398,
+      "grad_norm": 0.8133414594705825,
+      "learning_rate": 1.7154993726104328e-07,
+      "loss": 0.7751,
+      "step": 698
+    },
+    {
+      "epoch": 1.6485849056603774,
+      "grad_norm": 0.7994545997274072,
+      "learning_rate": 1.6935148868177718e-07,
+      "loss": 0.7501,
+      "step": 699
+    },
+    {
+      "epoch": 1.650943396226415,
+      "grad_norm": 0.7496147135106944,
+      "learning_rate": 1.6716591401915502e-07,
+      "loss": 0.759,
+      "step": 700
+    },
+    {
+      "epoch": 1.6533018867924527,
+      "grad_norm": 0.7274661448787512,
+      "learning_rate": 1.6499324714659758e-07,
+      "loss": 0.735,
+      "step": 701
+    },
+    {
+      "epoch": 1.6556603773584906,
+      "grad_norm": 0.8892069878489233,
+      "learning_rate": 1.6283352173747146e-07,
+      "loss": 0.7694,
+      "step": 702
+    },
+    {
+      "epoch": 1.6580188679245285,
+      "grad_norm": 0.7851702666302485,
+      "learning_rate": 1.6068677126456897e-07,
+      "loss": 0.7373,
+      "step": 703
+    },
+    {
+      "epoch": 1.6603773584905661,
+      "grad_norm": 0.7771975549062173,
+      "learning_rate": 1.585530289995878e-07,
+      "loss": 0.7101,
+      "step": 704
+    },
+    {
+      "epoch": 1.6627358490566038,
+      "grad_norm": 0.8852742002247123,
+      "learning_rate": 1.564323280126173e-07,
+      "loss": 0.7525,
+      "step": 705
+    },
+    {
+      "epoch": 1.6650943396226414,
+      "grad_norm": 1.0210362257937011,
+      "learning_rate": 1.5432470117162433e-07,
+      "loss": 0.7752,
+      "step": 706
+    },
+    {
+      "epoch": 1.6674528301886793,
+      "grad_norm": 0.8077571397037019,
+      "learning_rate": 1.522301811419442e-07,
+      "loss": 0.7395,
+      "step": 707
+    },
+    {
+      "epoch": 1.669811320754717,
+      "grad_norm": 0.8331764729151993,
+      "learning_rate": 1.5014880038577482e-07,
+      "loss": 0.7416,
+      "step": 708
+    },
+    {
+      "epoch": 1.6721698113207548,
+      "grad_norm": 2.1774846943729935,
+      "learning_rate": 1.4808059116167303e-07,
+      "loss": 0.7492,
+      "step": 709
+    },
+    {
+      "epoch": 1.6745283018867925,
+      "grad_norm": 0.8939523687224223,
+      "learning_rate": 1.460255855240552e-07,
+      "loss": 0.7358,
+      "step": 710
+    },
+    {
+      "epoch": 1.6768867924528301,
+      "grad_norm": 0.7563678829167328,
+      "learning_rate": 1.4398381532269998e-07,
+      "loss": 0.7593,
+      "step": 711
+    },
+    {
+      "epoch": 1.6792452830188678,
+      "grad_norm": 0.7668313174457733,
+      "learning_rate": 1.4195531220225487e-07,
+      "loss": 0.7275,
+      "step": 712
+    },
+    {
+      "epoch": 1.6816037735849056,
+      "grad_norm": 0.7791231762469476,
+      "learning_rate": 1.39940107601746e-07,
+      "loss": 0.7692,
+      "step": 713
+    },
+    {
+      "epoch": 1.6839622641509435,
+      "grad_norm": 1.014530101254706,
+      "learning_rate": 1.3793823275409066e-07,
+      "loss": 0.7475,
+      "step": 714
+    },
+    {
+      "epoch": 1.6863207547169812,
+      "grad_norm": 0.754925190142359,
+      "learning_rate": 1.3594971868561232e-07,
+      "loss": 0.7248,
+      "step": 715
+    },
+    {
+      "epoch": 1.6886792452830188,
+      "grad_norm": 0.7608021380299098,
+      "learning_rate": 1.3397459621556128e-07,
+      "loss": 0.7558,
+      "step": 716
+    },
+    {
+      "epoch": 1.6910377358490565,
+      "grad_norm": 1.320307190976547,
+      "learning_rate": 1.320128959556369e-07,
+      "loss": 0.7393,
+      "step": 717
+    },
+    {
+      "epoch": 1.6933962264150944,
+      "grad_norm": 0.8339214047495644,
+      "learning_rate": 1.300646483095118e-07,
+      "loss": 0.7597,
+      "step": 718
+    },
+    {
+      "epoch": 1.6957547169811322,
+      "grad_norm": 0.8826119280765871,
+      "learning_rate": 1.2812988347236166e-07,
+      "loss": 0.7416,
+      "step": 719
+    },
+    {
+      "epoch": 1.6981132075471699,
+      "grad_norm": 0.8928251184591149,
+      "learning_rate": 1.262086314303973e-07,
+      "loss": 0.7318,
+      "step": 720
+    },
+    {
+      "epoch": 1.7004716981132075,
+      "grad_norm": 0.7951715947182522,
+      "learning_rate": 1.243009219603993e-07,
+      "loss": 0.7482,
+      "step": 721
+    },
+    {
+      "epoch": 1.7028301886792452,
+      "grad_norm": 0.8741704408992066,
+      "learning_rate": 1.2240678462925723e-07,
+      "loss": 0.7287,
+      "step": 722
+    },
+    {
+      "epoch": 1.705188679245283,
+      "grad_norm": 0.7786335970762307,
+      "learning_rate": 1.2052624879351103e-07,
+      "loss": 0.719,
+      "step": 723
+    },
+    {
+      "epoch": 1.7075471698113207,
+      "grad_norm": 0.9985291709831805,
+      "learning_rate": 1.1865934359889573e-07,
+      "loss": 0.7453,
+      "step": 724
+    },
+    {
+      "epoch": 1.7099056603773586,
+      "grad_norm": 0.7859405679363147,
+      "learning_rate": 1.1680609797989038e-07,
+      "loss": 0.7452,
+      "step": 725
+    },
+    {
+      "epoch": 1.7122641509433962,
+      "grad_norm": 0.8028725796240722,
+      "learning_rate": 1.1496654065926925e-07,
+      "loss": 0.7173,
+      "step": 726
+    },
+    {
+      "epoch": 1.7146226415094339,
+      "grad_norm": 0.8165982522602734,
+      "learning_rate": 1.1314070014765642e-07,
+      "loss": 0.7218,
+      "step": 727
+    },
+    {
+      "epoch": 1.7169811320754715,
+      "grad_norm": 0.7860295163288017,
+      "learning_rate": 1.1132860474308436e-07,
+      "loss": 0.7614,
+      "step": 728
+    },
+    {
+      "epoch": 1.7193396226415094,
+      "grad_norm": 0.7889724037710595,
+      "learning_rate": 1.0953028253055541e-07,
+      "loss": 0.7195,
+      "step": 729
+    },
+    {
+      "epoch": 1.7216981132075473,
+      "grad_norm": 0.8067588097960763,
+      "learning_rate": 1.0774576138160596e-07,
+      "loss": 0.7482,
+      "step": 730
+    },
+    {
+      "epoch": 1.724056603773585,
+      "grad_norm": 0.786061208841395,
+      "learning_rate": 1.0597506895387499e-07,
+      "loss": 0.7472,
+      "step": 731
+    },
+    {
+      "epoch": 1.7264150943396226,
+      "grad_norm": 0.7917438727246212,
+      "learning_rate": 1.0421823269067442e-07,
+      "loss": 0.7751,
+      "step": 732
+    },
+    {
+      "epoch": 1.7287735849056602,
+      "grad_norm": 0.7907162871021037,
+      "learning_rate": 1.024752798205658e-07,
+      "loss": 0.7202,
+      "step": 733
+    },
+    {
+      "epoch": 1.7311320754716981,
+      "grad_norm": 0.9995843557820588,
+      "learning_rate": 1.0074623735693633e-07,
+      "loss": 0.7471,
+      "step": 734
+    },
+    {
+      "epoch": 1.733490566037736,
+      "grad_norm": 1.3558698569505305,
+      "learning_rate": 9.903113209758096e-08,
+      "loss": 0.7564,
+      "step": 735
+    },
+    {
+      "epoch": 1.7358490566037736,
+      "grad_norm": 0.7870396429020904,
+      "learning_rate": 9.732999062428704e-08,
+      "loss": 0.7641,
+      "step": 736
+    },
+    {
+      "epoch": 1.7382075471698113,
+      "grad_norm": 0.8525068172515468,
+      "learning_rate": 9.564283930242257e-08,
+      "loss": 0.7404,
+      "step": 737
+    },
+    {
+      "epoch": 1.740566037735849,
+      "grad_norm": 0.8014286873652903,
+      "learning_rate": 9.396970428052697e-08,
+      "loss": 0.7194,
+      "step": 738
+    },
+    {
+      "epoch": 1.7429245283018868,
+      "grad_norm": 0.7676759180901973,
+      "learning_rate": 9.231061148990648e-08,
+      "loss": 0.7221,
+      "step": 739
+    },
+    {
+      "epoch": 1.7452830188679245,
+      "grad_norm": 0.9131948150344564,
+      "learning_rate": 9.066558664423163e-08,
+      "loss": 0.7572,
+      "step": 740
+    },
+    {
+      "epoch": 1.7476415094339623,
+      "grad_norm": 0.7734044180311592,
+      "learning_rate": 8.903465523913955e-08,
+      "loss": 0.7757,
+      "step": 741
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 0.755029746442279,
+      "learning_rate": 8.741784255183759e-08,
+      "loss": 0.7411,
+      "step": 742
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 0.6796497702598572,
+      "eval_runtime": 82.4746,
+      "eval_samples_per_second": 8.548,
+      "eval_steps_per_second": 0.546,
+      "step": 742
+    },
+    {
+      "epoch": 1.7523584905660377,
+      "grad_norm": 0.8497097319379918,
+      "learning_rate": 8.581517364071267e-08,
+      "loss": 0.7203,
+      "step": 743
+    },
+    {
+      "epoch": 1.7547169811320755,
+      "grad_norm": 0.8020779183255858,
+      "learning_rate": 8.422667334494249e-08,
+      "loss": 0.7558,
+      "step": 744
+    },
+    {
+      "epoch": 1.7570754716981132,
+      "grad_norm": 0.7616565389684601,
+      "learning_rate": 8.265236628411087e-08,
+      "loss": 0.7422,
+      "step": 745
+    },
+    {
+      "epoch": 1.759433962264151,
+      "grad_norm": 0.9323853208771872,
+      "learning_rate": 8.109227685782538e-08,
+      "loss": 0.7819,
+      "step": 746
+    },
+    {
+      "epoch": 1.7617924528301887,
+      "grad_norm": 1.350644299549973,
+      "learning_rate": 7.954642924533994e-08,
+      "loss": 0.7492,
+      "step": 747
+    },
+    {
+      "epoch": 1.7641509433962264,
+      "grad_norm": 0.8092647636710026,
+      "learning_rate": 7.801484740517939e-08,
+      "loss": 0.7638,
+      "step": 748
+    },
+    {
+      "epoch": 1.766509433962264,
+      "grad_norm": 0.8308445659034376,
+      "learning_rate": 7.649755507476952e-08,
+      "loss": 0.7444,
+      "step": 749
+    },
+    {
+      "epoch": 1.7688679245283019,
+      "grad_norm": 2.5649621576701587,
+      "learning_rate": 7.499457577006751e-08,
+      "loss": 0.7713,
+      "step": 750
+    },
+    {
+      "epoch": 1.7712264150943398,
+      "grad_norm": 0.786323334872225,
+      "learning_rate": 7.350593278519823e-08,
+      "loss": 0.7426,
+      "step": 751
+    },
+    {
+      "epoch": 1.7735849056603774,
+      "grad_norm": 0.7795649745069329,
+      "learning_rate": 7.203164919209359e-08,
+      "loss": 0.7674,
+      "step": 752
+    },
+    {
+      "epoch": 1.775943396226415,
+      "grad_norm": 0.8003158325581134,
+      "learning_rate": 7.057174784013431e-08,
+      "loss": 0.7531,
+      "step": 753
+    },
+    {
+      "epoch": 1.7783018867924527,
+      "grad_norm": 0.7834816769548146,
+      "learning_rate": 6.912625135579586e-08,
+      "loss": 0.7212,
+      "step": 754
+    },
+    {
+      "epoch": 1.7806603773584906,
+      "grad_norm": 0.8289452420732738,
+      "learning_rate": 6.76951821422982e-08,
+      "loss": 0.7715,
+      "step": 755
+    },
+    {
+      "epoch": 1.7830188679245285,
+      "grad_norm": 0.8410465172590653,
+      "learning_rate": 6.627856237925811e-08,
+      "loss": 0.7276,
+      "step": 756
+    },
+    {
+      "epoch": 1.7853773584905661,
+      "grad_norm": 0.8394202851068477,
+      "learning_rate": 6.487641402234612e-08,
+      "loss": 0.744,
+      "step": 757
+    },
+    {
+      "epoch": 1.7877358490566038,
+      "grad_norm": 0.8383760649776106,
+      "learning_rate": 6.348875880294535e-08,
+      "loss": 0.7384,
+      "step": 758
+    },
+    {
+      "epoch": 1.7900943396226414,
+      "grad_norm": 1.453114388460075,
+      "learning_rate": 6.211561822781474e-08,
+      "loss": 0.7765,
+      "step": 759
+    },
+    {
+      "epoch": 1.7924528301886793,
+      "grad_norm": 0.8499896987869552,
+      "learning_rate": 6.075701357875662e-08,
+      "loss": 0.7284,
+      "step": 760
+    },
+    {
+      "epoch": 1.794811320754717,
+      "grad_norm": 0.8652146496280481,
+      "learning_rate": 5.9412965912286396e-08,
+      "loss": 0.7561,
+      "step": 761
+    },
+    {
+      "epoch": 1.7971698113207548,
+      "grad_norm": 0.8299032649447099,
+      "learning_rate": 5.808349605930585e-08,
+      "loss": 0.7565,
+      "step": 762
+    },
+    {
+      "epoch": 1.7995283018867925,
+      "grad_norm": 0.7812451643671983,
+      "learning_rate": 5.6768624624780604e-08,
+      "loss": 0.725,
+      "step": 763
+    },
+    {
+      "epoch": 1.8018867924528301,
+      "grad_norm": 0.8477713992666978,
+      "learning_rate": 5.5468371987420936e-08,
+      "loss": 0.7466,
+      "step": 764
+    },
+    {
+      "epoch": 1.8042452830188678,
+      "grad_norm": 0.8337239533995923,
+      "learning_rate": 5.4182758299365364e-08,
+      "loss": 0.735,
+      "step": 765
+    },
+    {
+      "epoch": 1.8066037735849056,
+      "grad_norm": 0.7691176408664865,
+      "learning_rate": 5.29118034858691e-08,
+      "loss": 0.772,
+      "step": 766
+    },
+    {
+      "epoch": 1.8089622641509435,
+      "grad_norm": 0.8049526588971233,
+      "learning_rate": 5.165552724499478e-08,
+      "loss": 0.7315,
+      "step": 767
+    },
+    {
+      "epoch": 1.8113207547169812,
+      "grad_norm": 0.8319464046786564,
+      "learning_rate": 5.0413949047306894e-08,
+      "loss": 0.7726,
+      "step": 768
+    },
+    {
+      "epoch": 1.8136792452830188,
+      "grad_norm": 0.8474092435937203,
+      "learning_rate": 4.918708813557093e-08,
+      "loss": 0.7487,
+      "step": 769
+    },
+    {
+      "epoch": 1.8160377358490565,
+      "grad_norm": 0.7943675010786753,
+      "learning_rate": 4.797496352445396e-08,
+      "loss": 0.7691,
+      "step": 770
+    },
+    {
+      "epoch": 1.8183962264150944,
+      "grad_norm": 0.8217628363291084,
+      "learning_rate": 4.677759400023085e-08,
+      "loss": 0.7437,
+      "step": 771
+    },
+    {
+      "epoch": 1.8207547169811322,
+      "grad_norm": 0.832358842867652,
+      "learning_rate": 4.55949981204925e-08,
+      "loss": 0.7568,
+      "step": 772
+    },
+    {
+      "epoch": 1.8231132075471699,
+      "grad_norm": 0.8613869764376921,
+      "learning_rate": 4.442719421385921e-08,
+      "loss": 0.7274,
+      "step": 773
+    },
+    {
+      "epoch": 1.8254716981132075,
+      "grad_norm": 0.7931363978514411,
+      "learning_rate": 4.3274200379695315e-08,
+      "loss": 0.7209,
+      "step": 774
+    },
+    {
+      "epoch": 1.8278301886792452,
+      "grad_norm": 0.7551383801460104,
+      "learning_rate": 4.213603448782932e-08,
+      "loss": 0.7533,
+      "step": 775
+    },
+    {
+      "epoch": 1.830188679245283,
+      "grad_norm": 0.8476707134264969,
+      "learning_rate": 4.101271417827668e-08,
+      "loss": 0.7585,
+      "step": 776
+    },
+    {
+      "epoch": 1.8325471698113207,
+      "grad_norm": 0.8877158557106878,
+      "learning_rate": 3.9904256860967433e-08,
+      "loss": 0.7222,
+      "step": 777
+    },
+    {
+      "epoch": 1.8349056603773586,
+      "grad_norm": 0.8633047498966683,
+      "learning_rate": 3.881067971547469e-08,
+      "loss": 0.7502,
+      "step": 778
+    },
+    {
+      "epoch": 1.8372641509433962,
+      "grad_norm": 1.0520376373477054,
+      "learning_rate": 3.7731999690749585e-08,
+      "loss": 0.7443,
+      "step": 779
+    },
+    {
+      "epoch": 1.8396226415094339,
+      "grad_norm": 0.7492875546723151,
+      "learning_rate": 3.666823350485848e-08,
+      "loss": 0.7781,
+      "step": 780
+    },
+    {
+      "epoch": 1.8419811320754715,
+      "grad_norm": 0.7847944050831147,
+      "learning_rate": 3.561939764472299e-08,
+      "loss": 0.737,
+      "step": 781
+    },
+    {
+      "epoch": 1.8443396226415094,
+      "grad_norm": 1.368419762468144,
+      "learning_rate": 3.458550836586582e-08,
+      "loss": 0.7523,
+      "step": 782
+    },
+    {
+      "epoch": 1.8466981132075473,
+      "grad_norm": 0.8704033001785062,
+      "learning_rate": 3.356658169215743e-08,
+      "loss": 0.742,
+      "step": 783
+    },
+    {
+      "epoch": 1.849056603773585,
+      "grad_norm": 0.764412444543419,
+      "learning_rate": 3.2562633415568754e-08,
+      "loss": 0.7327,
+      "step": 784
+    },
+    {
+      "epoch": 1.8514150943396226,
+      "grad_norm": 0.8695554372256408,
+      "learning_rate": 3.157367909592601e-08,
+      "loss": 0.7338,
+      "step": 785
+    },
+    {
+      "epoch": 1.8537735849056602,
+      "grad_norm": 0.7794614378925147,
+      "learning_rate": 3.0599734060669626e-08,
+      "loss": 0.7446,
+      "step": 786
+    },
+    {
+      "epoch": 1.8561320754716981,
+      "grad_norm": 0.7712181969029962,
+      "learning_rate": 2.9640813404616327e-08,
+      "loss": 0.7377,
+      "step": 787
+    },
+    {
+      "epoch": 1.858490566037736,
+      "grad_norm": 0.8826462854163208,
+      "learning_rate": 2.869693198972556e-08,
+      "loss": 0.7555,
+      "step": 788
+    },
+    {
+      "epoch": 1.8608490566037736,
+      "grad_norm": 1.029113085742105,
+      "learning_rate": 2.7768104444869434e-08,
+      "loss": 0.7795,
+      "step": 789
+    },
+    {
+      "epoch": 1.8632075471698113,
+      "grad_norm": 0.8318363964351398,
+      "learning_rate": 2.6854345165605474e-08,
+      "loss": 0.7351,
+      "step": 790
+    },
+    {
+      "epoch": 1.865566037735849,
+      "grad_norm": 0.7865924156566725,
+      "learning_rate": 2.595566831395346e-08,
+      "loss": 0.7222,
+      "step": 791
+    },
+    {
+      "epoch": 1.8679245283018868,
+      "grad_norm": 0.7955098420784998,
+      "learning_rate": 2.507208781817638e-08,
+      "loss": 0.7515,
+      "step": 792
+    },
+    {
+      "epoch": 1.8702830188679245,
+      "grad_norm": 0.9146875621357804,
+      "learning_rate": 2.4203617372564378e-08,
+      "loss": 0.7173,
+      "step": 793
+    },
+    {
+      "epoch": 1.8726415094339623,
+      "grad_norm": 0.7544835268896075,
+      "learning_rate": 2.3350270437222374e-08,
+      "loss": 0.7307,
+      "step": 794
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 1.806767663849715,
+      "learning_rate": 2.2512060237861452e-08,
+      "loss": 0.7301,
+      "step": 795
+    },
+    {
+      "epoch": 1.8773584905660377,
+      "grad_norm": 0.804925906379938,
+      "learning_rate": 2.1688999765594018e-08,
+      "loss": 0.7552,
+      "step": 796
+    },
+    {
+      "epoch": 1.8797169811320755,
+      "grad_norm": 0.7773879448134918,
+      "learning_rate": 2.0881101776732967e-08,
+      "loss": 0.7619,
+      "step": 797
+    },
+    {
+      "epoch": 1.8820754716981132,
+      "grad_norm": 0.871709115043609,
+      "learning_rate": 2.0088378792592286e-08,
+      "loss": 0.7686,
+      "step": 798
+    },
+    {
+      "epoch": 1.884433962264151,
+      "grad_norm": 0.7446217889352331,
+      "learning_rate": 1.9310843099295204e-08,
+      "loss": 0.7128,
+      "step": 799
+    },
+    {
+      "epoch": 1.8867924528301887,
+      "grad_norm": 0.7661065857274754,
+      "learning_rate": 1.8548506747582128e-08,
+      "loss": 0.7284,
+      "step": 800
+    },
+    {
+      "epoch": 1.8891509433962264,
+      "grad_norm": 0.7916406522727306,
+      "learning_rate": 1.780138155262456e-08,
+      "loss": 0.7321,
+      "step": 801
+    },
+    {
+      "epoch": 1.891509433962264,
+      "grad_norm": 0.9641121942586747,
+      "learning_rate": 1.7069479093842042e-08,
+      "loss": 0.7411,
+      "step": 802
+    },
+    {
+      "epoch": 1.8938679245283019,
+      "grad_norm": 0.7930873801533905,
+      "learning_rate": 1.6352810714722387e-08,
+      "loss": 0.7345,
+      "step": 803
+    },
+    {
+      "epoch": 1.8962264150943398,
+      "grad_norm": 0.9298677041430008,
+      "learning_rate": 1.565138752264572e-08,
+      "loss": 0.7616,
+      "step": 804
+    },
+    {
+      "epoch": 1.8985849056603774,
+      "grad_norm": 0.885574237101463,
+      "learning_rate": 1.496522038871295e-08,
+      "loss": 0.7572,
+      "step": 805
+    },
+    {
+      "epoch": 1.900943396226415,
+      "grad_norm": 0.8153461825360933,
+      "learning_rate": 1.4294319947577017e-08,
+      "loss": 0.75,
+      "step": 806
+    },
+    {
+      "epoch": 1.9033018867924527,
+      "grad_norm": 0.7660700579693493,
+      "learning_rate": 1.3638696597277677e-08,
+      "loss": 0.7421,
+      "step": 807
+    },
+    {
+      "epoch": 1.9056603773584906,
+      "grad_norm": 0.805964323807633,
+      "learning_rate": 1.2998360499080763e-08,
+      "loss": 0.7958,
+      "step": 808
+    },
+    {
+      "epoch": 1.9080188679245285,
+      "grad_norm": 0.9429626916004815,
+      "learning_rate": 1.2373321577320628e-08,
+      "loss": 0.734,
+      "step": 809
+    },
+    {
+      "epoch": 1.9103773584905661,
+      "grad_norm": 0.7904102497155711,
+      "learning_rate": 1.1763589519246387e-08,
+      "loss": 0.7478,
+      "step": 810
+    },
+    {
+      "epoch": 1.9127358490566038,
+      "grad_norm": 0.7928898152839706,
+      "learning_rate": 1.1169173774871477e-08,
+      "loss": 0.7373,
+      "step": 811
+    },
+    {
+      "epoch": 1.9150943396226414,
+      "grad_norm": 0.7814723884700895,
+      "learning_rate": 1.0590083556827556e-08,
+      "loss": 0.7615,
+      "step": 812
+    },
+    {
+      "epoch": 1.9174528301886793,
+      "grad_norm": 0.7802344182322299,
+      "learning_rate": 1.0026327840221727e-08,
+      "loss": 0.7503,
+      "step": 813
+    },
+    {
+      "epoch": 1.919811320754717,
+      "grad_norm": 1.0812085791485797,
+      "learning_rate": 9.477915362496758e-09,
+      "loss": 0.7614,
+      "step": 814
+    },
+    {
+      "epoch": 1.9221698113207548,
+      "grad_norm": 0.820203697559199,
+      "learning_rate": 8.94485462329675e-09,
+      "loss": 0.7374,
+      "step": 815
+    },
+    {
+      "epoch": 1.9245283018867925,
+      "grad_norm": 0.8722587060636673,
+      "learning_rate": 8.42715388433446e-09,
+      "loss": 0.7328,
+      "step": 816
+    },
+    {
+      "epoch": 1.9268867924528301,
+      "grad_norm": 0.7840984315269908,
+      "learning_rate": 7.924821169263963e-09,
+      "loss": 0.7542,
+      "step": 817
+    },
+    {
+      "epoch": 1.9292452830188678,
+      "grad_norm": 0.8440330231818731,
+      "learning_rate": 7.437864263555638e-09,
+      "loss": 0.7317,
+      "step": 818
+    },
+    {
+      "epoch": 1.9316037735849056,
+      "grad_norm": 0.895812932209063,
+      "learning_rate": 6.966290714375933e-09,
+      "loss": 0.7409,
+      "step": 819
+    },
+    {
+      "epoch": 1.9339622641509435,
+      "grad_norm": 0.899638041623908,
+      "learning_rate": 6.510107830470568e-09,
+      "loss": 0.7446,
+      "step": 820
+    },
+    {
+      "epoch": 1.9363207547169812,
+      "grad_norm": 0.8152793437016046,
+      "learning_rate": 6.069322682050515e-09,
+      "loss": 0.7634,
+      "step": 821
+    },
+    {
+      "epoch": 1.9386792452830188,
+      "grad_norm": 0.7735814965641481,
+      "learning_rate": 5.643942100683308e-09,
+      "loss": 0.7809,
+      "step": 822
+    },
+    {
+      "epoch": 1.9410377358490565,
+      "grad_norm": 0.8351447564551893,
+      "learning_rate": 5.23397267918646e-09,
+      "loss": 0.7588,
+      "step": 823
+    },
+    {
+      "epoch": 1.9433962264150944,
+      "grad_norm": 0.7614597204440534,
+      "learning_rate": 4.83942077152577e-09,
+      "loss": 0.7231,
+      "step": 824
+    },
+    {
+      "epoch": 1.9457547169811322,
+      "grad_norm": 0.8548976791386363,
+      "learning_rate": 4.460292492716511e-09,
+      "loss": 0.754,
+      "step": 825
+    },
+    {
+      "epoch": 1.9481132075471699,
+      "grad_norm": 0.8381107415157564,
+      "learning_rate": 4.0965937187287246e-09,
+      "loss": 0.7351,
+      "step": 826
+    },
+    {
+      "epoch": 1.9504716981132075,
+      "grad_norm": 5.772455437843555,
+      "learning_rate": 3.748330086396523e-09,
+      "loss": 0.7657,
+      "step": 827
+    },
+    {
+      "epoch": 1.9528301886792452,
+      "grad_norm": 0.797010233582818,
+      "learning_rate": 3.415506993330153e-09,
+      "loss": 0.7248,
+      "step": 828
+    },
+    {
+      "epoch": 1.955188679245283,
+      "grad_norm": 0.777781094479457,
+      "learning_rate": 3.0981295978326216e-09,
+      "loss": 0.7604,
+      "step": 829
+    },
+    {
+      "epoch": 1.9575471698113207,
+      "grad_norm": 0.8449725908943951,
+      "learning_rate": 2.7962028188198706e-09,
+      "loss": 0.7736,
+      "step": 830
+    },
+    {
+      "epoch": 1.9599056603773586,
+      "grad_norm": 0.7801120467589358,
+      "learning_rate": 2.5097313357442806e-09,
+      "loss": 0.7178,
+      "step": 831
+    },
+    {
+      "epoch": 1.9622641509433962,
+      "grad_norm": 0.7762312098767981,
+      "learning_rate": 2.2387195885221756e-09,
+      "loss": 0.7575,
+      "step": 832
+    },
+    {
+      "epoch": 1.9646226415094339,
+      "grad_norm": 0.824362916729911,
+      "learning_rate": 1.983171777465431e-09,
+      "loss": 0.76,
+      "step": 833
+    },
+    {
+      "epoch": 1.9669811320754715,
+      "grad_norm": 0.9204175438402101,
+      "learning_rate": 1.743091863215751e-09,
+      "loss": 0.7312,
+      "step": 834
+    },
+    {
+      "epoch": 1.9693396226415094,
+      "grad_norm": 0.8426151332993567,
+      "learning_rate": 1.5184835666838258e-09,
+      "loss": 0.7489,
+      "step": 835
+    },
+    {
+      "epoch": 1.9716981132075473,
+      "grad_norm": 0.7997774511020088,
+      "learning_rate": 1.3093503689910467e-09,
+      "loss": 0.7243,
+      "step": 836
+    },
+    {
+      "epoch": 1.974056603773585,
+      "grad_norm": 1.0915014977180026,
+      "learning_rate": 1.1156955114162147e-09,
+      "loss": 0.7411,
+      "step": 837
+    },
+    {
+      "epoch": 1.9764150943396226,
+      "grad_norm": 0.7394147838620203,
+      "learning_rate": 9.375219953450253e-10,
+      "loss": 0.746,
+      "step": 838
+    },
+    {
+      "epoch": 1.9787735849056602,
+      "grad_norm": 0.7690476142622722,
+      "learning_rate": 7.748325822234392e-10,
+      "loss": 0.729,
+      "step": 839
+    },
+    {
+      "epoch": 1.9811320754716981,
+      "grad_norm": 0.9039884352435014,
+      "learning_rate": 6.276297935149388e-10,
+      "loss": 0.7569,
+      "step": 840
+    },
+    {
+      "epoch": 1.983490566037736,
+      "grad_norm": 1.2814279255523484,
+      "learning_rate": 4.959159106615596e-10,
+      "loss": 0.7351,
+      "step": 841
+    },
+    {
+      "epoch": 1.9858490566037736,
+      "grad_norm": 0.8004748840113347,
+      "learning_rate": 3.7969297504858443e-10,
+      "loss": 0.724,
+      "step": 842
+    },
+    {
+      "epoch": 1.9882075471698113,
+      "grad_norm": 0.9234344448728469,
+      "learning_rate": 2.789627879725698e-10,
+      "loss": 0.749,
+      "step": 843
+    },
+    {
+      "epoch": 1.990566037735849,
+      "grad_norm": 1.0793509856974828,
+      "learning_rate": 1.9372691061381175e-10,
+      "loss": 0.7296,
+      "step": 844
+    },
+    {
+      "epoch": 1.9929245283018868,
+      "grad_norm": 0.8234018950018985,
+      "learning_rate": 1.2398666401181035e-10,
+      "loss": 0.738,
+      "step": 845
+    },
+    {
+      "epoch": 1.9952830188679245,
+      "grad_norm": 0.7591071692242605,
+      "learning_rate": 6.974312904517443e-11,
+      "loss": 0.7713,
+      "step": 846
+    },
+    {
+      "epoch": 1.9976415094339623,
+      "grad_norm": 0.8367565544655438,
+      "learning_rate": 3.099714641452422e-11,
+      "loss": 0.7812,
+      "step": 847
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8006380089754,
+      "learning_rate": 7.749316629612756e-12,
+      "loss": 0.769,
+      "step": 848
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6794618964195251,
+      "eval_runtime": 82.3418,
+      "eval_samples_per_second": 8.562,
+      "eval_steps_per_second": 0.547,
+      "step": 848
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 848,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 212,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.964690701765837e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}