diff --git "a/checkpoint-750/trainer_state.json" "b/checkpoint-750/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-750/trainer_state.json"
@@ -0,0 +1,5284 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.24,
+  "eval_steps": 500,
+  "global_step": 750,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00032,
+      "grad_norm": 0.24713319540023804,
+      "learning_rate": 0.0,
+      "loss": 0.8884,
+      "step": 1
+    },
+    {
+      "epoch": 0.00064,
+      "grad_norm": 0.20603826642036438,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 0.8479,
+      "step": 2
+    },
+    {
+      "epoch": 0.00096,
+      "grad_norm": 0.24543610215187073,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.9853,
+      "step": 3
+    },
+    {
+      "epoch": 0.00128,
+      "grad_norm": 0.2051621973514557,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.8293,
+      "step": 4
+    },
+    {
+      "epoch": 0.0016,
+      "grad_norm": 0.23469021916389465,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 0.9811,
+      "step": 5
+    },
+    {
+      "epoch": 0.00192,
+      "grad_norm": 0.25180304050445557,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.039,
+      "step": 6
+    },
+    {
+      "epoch": 0.00224,
+      "grad_norm": 0.23135970532894135,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.8688,
+      "step": 7
+    },
+    {
+      "epoch": 0.00256,
+      "grad_norm": 0.22394850850105286,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.9214,
+      "step": 8
+    },
+    {
+      "epoch": 0.00288,
+      "grad_norm": 0.25566354393959045,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.8667,
+      "step": 9
+    },
+    {
+      "epoch": 0.0032,
+      "grad_norm": 0.2881408631801605,
+      "learning_rate": 1.2e-05,
+      "loss": 0.9544,
+      "step": 10
+    },
+    {
+      "epoch": 0.00352,
+      "grad_norm": 0.22555342316627502,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.9383,
+      "step": 11
+    },
+    {
+      "epoch": 0.00384,
+      "grad_norm": 0.23086762428283691,
+      "learning_rate": 1.4666666666666668e-05,
+      "loss": 0.8595,
+      "step": 12
+    },
+    {
+      "epoch": 0.00416,
+      "grad_norm": 0.2487766444683075,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.9469,
+      "step": 13
+    },
+    {
+      "epoch": 0.00448,
+      "grad_norm": 0.3970007598400116,
+      "learning_rate": 1.7333333333333336e-05,
+      "loss": 1.1923,
+      "step": 14
+    },
+    {
+      "epoch": 0.0048,
+      "grad_norm": 0.25204387307167053,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 0.8854,
+      "step": 15
+    },
+    {
+      "epoch": 0.00512,
+      "grad_norm": 0.23282122611999512,
+      "learning_rate": 2e-05,
+      "loss": 0.932,
+      "step": 16
+    },
+    {
+      "epoch": 0.00544,
+      "grad_norm": 0.2607545852661133,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 0.9094,
+      "step": 17
+    },
+    {
+      "epoch": 0.00576,
+      "grad_norm": 0.21402348577976227,
+      "learning_rate": 2.2666666666666668e-05,
+      "loss": 0.7879,
+      "step": 18
+    },
+    {
+      "epoch": 0.00608,
+      "grad_norm": 0.23922377824783325,
+      "learning_rate": 2.4e-05,
+      "loss": 0.9193,
+      "step": 19
+    },
+    {
+      "epoch": 0.0064,
+      "grad_norm": 0.18835392594337463,
+      "learning_rate": 2.5333333333333337e-05,
+      "loss": 0.9786,
+      "step": 20
+    },
+    {
+      "epoch": 0.00672,
+      "grad_norm": 0.1775645613670349,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.933,
+      "step": 21
+    },
+    {
+      "epoch": 0.00704,
+      "grad_norm": 0.20945614576339722,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 0.8171,
+      "step": 22
+    },
+    {
+      "epoch": 0.00736,
+      "grad_norm": 0.16134795546531677,
+      "learning_rate": 2.9333333333333336e-05,
+      "loss": 0.9665,
+      "step": 23
+    },
+    {
+      "epoch": 0.00768,
+      "grad_norm": 0.1454283446073532,
+      "learning_rate": 3.066666666666667e-05,
+      "loss": 0.7238,
+      "step": 24
+    },
+    {
+      "epoch": 0.008,
+      "grad_norm": 0.15920202434062958,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.7932,
+      "step": 25
+    },
+    {
+      "epoch": 0.00832,
+      "grad_norm": 0.16204868257045746,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.8175,
+      "step": 26
+    },
+    {
+      "epoch": 0.00864,
+      "grad_norm": 0.15179912745952606,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 0.9067,
+      "step": 27
+    },
+    {
+      "epoch": 0.00896,
+      "grad_norm": 0.1287967711687088,
+      "learning_rate": 3.6e-05,
+      "loss": 0.8039,
+      "step": 28
+    },
+    {
+      "epoch": 0.00928,
+      "grad_norm": 0.13528791069984436,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 0.8892,
+      "step": 29
+    },
+    {
+      "epoch": 0.0096,
+      "grad_norm": 0.13059762120246887,
+      "learning_rate": 3.866666666666667e-05,
+      "loss": 0.7023,
+      "step": 30
+    },
+    {
+      "epoch": 0.00992,
+      "grad_norm": 0.13119487464427948,
+      "learning_rate": 4e-05,
+      "loss": 0.7607,
+      "step": 31
+    },
+    {
+      "epoch": 0.01024,
+      "grad_norm": 0.12565594911575317,
+      "learning_rate": 4.133333333333333e-05,
+      "loss": 0.808,
+      "step": 32
+    },
+    {
+      "epoch": 0.01056,
+      "grad_norm": 0.1259136199951172,
+      "learning_rate": 4.266666666666667e-05,
+      "loss": 0.8704,
+      "step": 33
+    },
+    {
+      "epoch": 0.01088,
+      "grad_norm": 0.1319507360458374,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6421,
+      "step": 34
+    },
+    {
+      "epoch": 0.0112,
+      "grad_norm": 0.1317446082830429,
+      "learning_rate": 4.5333333333333335e-05,
+      "loss": 0.688,
+      "step": 35
+    },
+    {
+      "epoch": 0.01152,
+      "grad_norm": 0.12360246479511261,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.798,
+      "step": 36
+    },
+    {
+      "epoch": 0.01184,
+      "grad_norm": 0.11974922567605972,
+      "learning_rate": 4.8e-05,
+      "loss": 0.829,
+      "step": 37
+    },
+    {
+      "epoch": 0.01216,
+      "grad_norm": 0.13918066024780273,
+      "learning_rate": 4.933333333333334e-05,
+      "loss": 0.8626,
+      "step": 38
+    },
+    {
+      "epoch": 0.01248,
+      "grad_norm": 0.11004938930273056,
+      "learning_rate": 5.0666666666666674e-05,
+      "loss": 0.8056,
+      "step": 39
+    },
+    {
+      "epoch": 0.0128,
+      "grad_norm": 0.10854440927505493,
+      "learning_rate": 5.2000000000000004e-05,
+      "loss": 0.8773,
+      "step": 40
+    },
+    {
+      "epoch": 0.01312,
+      "grad_norm": 0.10877286642789841,
+      "learning_rate": 5.333333333333333e-05,
+      "loss": 0.8947,
+      "step": 41
+    },
+    {
+      "epoch": 0.01344,
+      "grad_norm": 0.12767033278942108,
+      "learning_rate": 5.466666666666666e-05,
+      "loss": 0.7892,
+      "step": 42
+    },
+    {
+      "epoch": 0.01376,
+      "grad_norm": 0.11700518429279327,
+      "learning_rate": 5.6000000000000006e-05,
+      "loss": 0.8598,
+      "step": 43
+    },
+    {
+      "epoch": 0.01408,
+      "grad_norm": 0.12549364566802979,
+      "learning_rate": 5.7333333333333336e-05,
+      "loss": 0.8498,
+      "step": 44
+    },
+    {
+      "epoch": 0.0144,
+      "grad_norm": 0.11921705305576324,
+      "learning_rate": 5.866666666666667e-05,
+      "loss": 0.7979,
+      "step": 45
+    },
+    {
+      "epoch": 0.01472,
+      "grad_norm": 0.11621945351362228,
+      "learning_rate": 6e-05,
+      "loss": 0.9252,
+      "step": 46
+    },
+    {
+      "epoch": 0.01504,
+      "grad_norm": 0.10912485420703888,
+      "learning_rate": 6.133333333333334e-05,
+      "loss": 0.7674,
+      "step": 47
+    },
+    {
+      "epoch": 0.01536,
+      "grad_norm": 0.11039536446332932,
+      "learning_rate": 6.266666666666667e-05,
+      "loss": 0.841,
+      "step": 48
+    },
+    {
+      "epoch": 0.01568,
+      "grad_norm": 0.10852135717868805,
+      "learning_rate": 6.400000000000001e-05,
+      "loss": 0.7991,
+      "step": 49
+    },
+    {
+      "epoch": 0.016,
+      "grad_norm": 0.09653560072183609,
+      "learning_rate": 6.533333333333334e-05,
+      "loss": 0.6292,
+      "step": 50
+    },
+    {
+      "epoch": 0.01632,
+      "grad_norm": 0.11932919919490814,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 0.7327,
+      "step": 51
+    },
+    {
+      "epoch": 0.01664,
+      "grad_norm": 0.10563201457262039,
+      "learning_rate": 6.800000000000001e-05,
+      "loss": 0.7041,
+      "step": 52
+    },
+    {
+      "epoch": 0.01696,
+      "grad_norm": 0.1063474714756012,
+      "learning_rate": 6.933333333333334e-05,
+      "loss": 0.7953,
+      "step": 53
+    },
+    {
+      "epoch": 0.01728,
+      "grad_norm": 0.11396024376153946,
+      "learning_rate": 7.066666666666667e-05,
+      "loss": 0.899,
+      "step": 54
+    },
+    {
+      "epoch": 0.0176,
+      "grad_norm": 0.10861669480800629,
+      "learning_rate": 7.2e-05,
+      "loss": 0.9343,
+      "step": 55
+    },
+    {
+      "epoch": 0.01792,
+      "grad_norm": 0.12476427853107452,
+      "learning_rate": 7.333333333333333e-05,
+      "loss": 0.8145,
+      "step": 56
+    },
+    {
+      "epoch": 0.01824,
+      "grad_norm": 0.1240817978978157,
+      "learning_rate": 7.466666666666667e-05,
+      "loss": 0.7133,
+      "step": 57
+    },
+    {
+      "epoch": 0.01856,
+      "grad_norm": 0.10977518558502197,
+      "learning_rate": 7.6e-05,
+      "loss": 0.7576,
+      "step": 58
+    },
+    {
+      "epoch": 0.01888,
+      "grad_norm": 0.11463093012571335,
+      "learning_rate": 7.733333333333333e-05,
+      "loss": 0.8552,
+      "step": 59
+    },
+    {
+      "epoch": 0.0192,
+      "grad_norm": 0.1143040657043457,
+      "learning_rate": 7.866666666666666e-05,
+      "loss": 0.9198,
+      "step": 60
+    },
+    {
+      "epoch": 0.01952,
+      "grad_norm": 0.12839296460151672,
+      "learning_rate": 8e-05,
+      "loss": 0.6644,
+      "step": 61
+    },
+    {
+      "epoch": 0.01984,
+      "grad_norm": 0.11903133243322372,
+      "learning_rate": 8.133333333333334e-05,
+      "loss": 0.9053,
+      "step": 62
+    },
+    {
+      "epoch": 0.02016,
+      "grad_norm": 0.12296755611896515,
+      "learning_rate": 8.266666666666667e-05,
+      "loss": 0.7695,
+      "step": 63
+    },
+    {
+      "epoch": 0.02048,
+      "grad_norm": 0.11100097000598907,
+      "learning_rate": 8.4e-05,
+      "loss": 0.8909,
+      "step": 64
+    },
+    {
+      "epoch": 0.0208,
+      "grad_norm": 0.11848391592502594,
+      "learning_rate": 8.533333333333334e-05,
+      "loss": 0.8204,
+      "step": 65
+    },
+    {
+      "epoch": 0.02112,
+      "grad_norm": 0.11903996020555496,
+      "learning_rate": 8.666666666666667e-05,
+      "loss": 0.7778,
+      "step": 66
+    },
+    {
+      "epoch": 0.02144,
+      "grad_norm": 0.10964814573526382,
+      "learning_rate": 8.800000000000001e-05,
+      "loss": 0.6879,
+      "step": 67
+    },
+    {
+      "epoch": 0.02176,
+      "grad_norm": 0.11036152392625809,
+      "learning_rate": 8.933333333333334e-05,
+      "loss": 0.8051,
+      "step": 68
+    },
+    {
+      "epoch": 0.02208,
+      "grad_norm": 0.10888101160526276,
+      "learning_rate": 9.066666666666667e-05,
+      "loss": 0.759,
+      "step": 69
+    },
+    {
+      "epoch": 0.0224,
+      "grad_norm": 0.11158827692270279,
+      "learning_rate": 9.200000000000001e-05,
+      "loss": 0.7412,
+      "step": 70
+    },
+    {
+      "epoch": 0.02272,
+      "grad_norm": 0.12085185945034027,
+      "learning_rate": 9.333333333333334e-05,
+      "loss": 0.7978,
+      "step": 71
+    },
+    {
+      "epoch": 0.02304,
+      "grad_norm": 0.12325625121593475,
+      "learning_rate": 9.466666666666667e-05,
+      "loss": 0.7175,
+      "step": 72
+    },
+    {
+      "epoch": 0.02336,
+      "grad_norm": 0.10695890337228775,
+      "learning_rate": 9.6e-05,
+      "loss": 0.868,
+      "step": 73
+    },
+    {
+      "epoch": 0.02368,
+      "grad_norm": 0.11983097344636917,
+      "learning_rate": 9.733333333333335e-05,
+      "loss": 0.7965,
+      "step": 74
+    },
+    {
+      "epoch": 0.024,
+      "grad_norm": 0.11521439999341965,
+      "learning_rate": 9.866666666666668e-05,
+      "loss": 0.9305,
+      "step": 75
+    },
+    {
+      "epoch": 0.02432,
+      "grad_norm": 0.1252303123474121,
+      "learning_rate": 0.0001,
+      "loss": 0.6807,
+      "step": 76
+    },
+    {
+      "epoch": 0.02464,
+      "grad_norm": 0.1265154629945755,
+      "learning_rate": 0.00010133333333333335,
+      "loss": 0.718,
+      "step": 77
+    },
+    {
+      "epoch": 0.02496,
+      "grad_norm": 0.1109502986073494,
+      "learning_rate": 0.00010266666666666666,
+      "loss": 0.769,
+      "step": 78
+    },
+    {
+      "epoch": 0.02528,
+      "grad_norm": 0.11158560961484909,
+      "learning_rate": 0.00010400000000000001,
+      "loss": 0.8579,
+      "step": 79
+    },
+    {
+      "epoch": 0.0256,
+      "grad_norm": 0.13526469469070435,
+      "learning_rate": 0.00010533333333333332,
+      "loss": 0.7704,
+      "step": 80
+    },
+    {
+      "epoch": 0.02592,
+      "grad_norm": 0.12420251965522766,
+      "learning_rate": 0.00010666666666666667,
+      "loss": 0.673,
+      "step": 81
+    },
+    {
+      "epoch": 0.02624,
+      "grad_norm": 0.11838296055793762,
+      "learning_rate": 0.00010800000000000001,
+      "loss": 0.8576,
+      "step": 82
+    },
+    {
+      "epoch": 0.02656,
+      "grad_norm": 0.11958430707454681,
+      "learning_rate": 0.00010933333333333333,
+      "loss": 0.8521,
+      "step": 83
+    },
+    {
+      "epoch": 0.02688,
+      "grad_norm": 0.12049078941345215,
+      "learning_rate": 0.00011066666666666667,
+      "loss": 0.6008,
+      "step": 84
+    },
+    {
+      "epoch": 0.0272,
+      "grad_norm": 0.10999640822410583,
+      "learning_rate": 0.00011200000000000001,
+      "loss": 0.7351,
+      "step": 85
+    },
+    {
+      "epoch": 0.02752,
+      "grad_norm": 0.11500285565853119,
+      "learning_rate": 0.00011333333333333334,
+      "loss": 0.7841,
+      "step": 86
+    },
+    {
+      "epoch": 0.02784,
+      "grad_norm": 0.11917781829833984,
+      "learning_rate": 0.00011466666666666667,
+      "loss": 0.8275,
+      "step": 87
+    },
+    {
+      "epoch": 0.02816,
+      "grad_norm": 0.11745665222406387,
+      "learning_rate": 0.000116,
+      "loss": 0.6096,
+      "step": 88
+    },
+    {
+      "epoch": 0.02848,
+      "grad_norm": 0.1208786815404892,
+      "learning_rate": 0.00011733333333333334,
+      "loss": 0.6842,
+      "step": 89
+    },
+    {
+      "epoch": 0.0288,
+      "grad_norm": 0.12839952111244202,
+      "learning_rate": 0.00011866666666666669,
+      "loss": 0.7467,
+      "step": 90
+    },
+    {
+      "epoch": 0.02912,
+      "grad_norm": 0.12688350677490234,
+      "learning_rate": 0.00012,
+      "loss": 0.6507,
+      "step": 91
+    },
+    {
+      "epoch": 0.02944,
+      "grad_norm": 0.11489767581224442,
+      "learning_rate": 0.00012133333333333335,
+      "loss": 0.8403,
+      "step": 92
+    },
+    {
+      "epoch": 0.02976,
+      "grad_norm": 0.1305433064699173,
+      "learning_rate": 0.00012266666666666668,
+      "loss": 0.748,
+      "step": 93
+    },
+    {
+      "epoch": 0.03008,
+      "grad_norm": 0.13992320001125336,
+      "learning_rate": 0.000124,
+      "loss": 0.796,
+      "step": 94
+    },
+    {
+      "epoch": 0.0304,
+      "grad_norm": 0.12120400369167328,
+      "learning_rate": 0.00012533333333333334,
+      "loss": 0.6445,
+      "step": 95
+    },
+    {
+      "epoch": 0.03072,
+      "grad_norm": 0.13254240155220032,
+      "learning_rate": 0.00012666666666666666,
+      "loss": 0.7488,
+      "step": 96
+    },
+    {
+      "epoch": 0.03104,
+      "grad_norm": 0.13790103793144226,
+      "learning_rate": 0.00012800000000000002,
+      "loss": 0.8005,
+      "step": 97
+    },
+    {
+      "epoch": 0.03136,
+      "grad_norm": 0.12862953543663025,
+      "learning_rate": 0.00012933333333333332,
+      "loss": 0.7975,
+      "step": 98
+    },
+    {
+      "epoch": 0.03168,
+      "grad_norm": 0.14197330176830292,
+      "learning_rate": 0.00013066666666666668,
+      "loss": 0.9026,
+      "step": 99
+    },
+    {
+      "epoch": 0.032,
+      "grad_norm": 0.12494270503520966,
+      "learning_rate": 0.000132,
+      "loss": 0.7353,
+      "step": 100
+    },
+    {
+      "epoch": 0.03232,
+      "grad_norm": 0.12572987377643585,
+      "learning_rate": 0.00013333333333333334,
+      "loss": 0.9271,
+      "step": 101
+    },
+    {
+      "epoch": 0.03264,
+      "grad_norm": 0.1314753144979477,
+      "learning_rate": 0.00013466666666666667,
+      "loss": 0.7816,
+      "step": 102
+    },
+    {
+      "epoch": 0.03296,
+      "grad_norm": 0.12902331352233887,
+      "learning_rate": 0.00013600000000000003,
+      "loss": 0.7348,
+      "step": 103
+    },
+    {
+      "epoch": 0.03328,
+      "grad_norm": 0.14086449146270752,
+      "learning_rate": 0.00013733333333333333,
+      "loss": 0.831,
+      "step": 104
+    },
+    {
+      "epoch": 0.0336,
+      "grad_norm": 0.12464585155248642,
+      "learning_rate": 0.00013866666666666669,
+      "loss": 0.6945,
+      "step": 105
+    },
+    {
+      "epoch": 0.03392,
+      "grad_norm": 0.11436092108488083,
+      "learning_rate": 0.00014,
+      "loss": 0.8192,
+      "step": 106
+    },
+    {
+      "epoch": 0.03424,
+      "grad_norm": 0.11779410392045975,
+      "learning_rate": 0.00014133333333333334,
+      "loss": 0.7989,
+      "step": 107
+    },
+    {
+      "epoch": 0.03456,
+      "grad_norm": 0.14601927995681763,
+      "learning_rate": 0.00014266666666666667,
+      "loss": 0.6256,
+      "step": 108
+    },
+    {
+      "epoch": 0.03488,
+      "grad_norm": 0.13231013715267181,
+      "learning_rate": 0.000144,
+      "loss": 0.7445,
+      "step": 109
+    },
+    {
+      "epoch": 0.0352,
+      "grad_norm": 0.11249889433383942,
+      "learning_rate": 0.00014533333333333333,
+      "loss": 0.7338,
+      "step": 110
+    },
+    {
+      "epoch": 0.03552,
+      "grad_norm": 0.11275426298379898,
+      "learning_rate": 0.00014666666666666666,
+      "loss": 0.7993,
+      "step": 111
+    },
+    {
+      "epoch": 0.03584,
+      "grad_norm": 0.11678344756364822,
+      "learning_rate": 0.000148,
+      "loss": 0.9234,
+      "step": 112
+    },
+    {
+      "epoch": 0.03616,
+      "grad_norm": 0.10879474133253098,
+      "learning_rate": 0.00014933333333333335,
+      "loss": 0.8732,
+      "step": 113
+    },
+    {
+      "epoch": 0.03648,
+      "grad_norm": 0.13673344254493713,
+      "learning_rate": 0.00015066666666666668,
+      "loss": 0.8698,
+      "step": 114
+    },
+    {
+      "epoch": 0.0368,
+      "grad_norm": 0.12564410269260406,
+      "learning_rate": 0.000152,
+      "loss": 0.7913,
+      "step": 115
+    },
+    {
+      "epoch": 0.03712,
+      "grad_norm": 0.12272034585475922,
+      "learning_rate": 0.00015333333333333334,
+      "loss": 0.7244,
+      "step": 116
+    },
+    {
+      "epoch": 0.03744,
+      "grad_norm": 0.11841209977865219,
+      "learning_rate": 0.00015466666666666667,
+      "loss": 0.7648,
+      "step": 117
+    },
+    {
+      "epoch": 0.03776,
+      "grad_norm": 0.11938042938709259,
+      "learning_rate": 0.00015600000000000002,
+      "loss": 0.8381,
+      "step": 118
+    },
+    {
+      "epoch": 0.03808,
+      "grad_norm": 0.13476668298244476,
+      "learning_rate": 0.00015733333333333333,
+      "loss": 0.6931,
+      "step": 119
+    },
+    {
+      "epoch": 0.0384,
+      "grad_norm": 0.1274396926164627,
+      "learning_rate": 0.00015866666666666668,
+      "loss": 0.854,
+      "step": 120
+    },
+    {
+      "epoch": 0.03872,
+      "grad_norm": 0.11518494784832001,
+      "learning_rate": 0.00016,
+      "loss": 0.7591,
+      "step": 121
+    },
+    {
+      "epoch": 0.03904,
+      "grad_norm": 0.12739014625549316,
+      "learning_rate": 0.00016133333333333334,
+      "loss": 0.6956,
+      "step": 122
+    },
+    {
+      "epoch": 0.03936,
+      "grad_norm": 0.13384665548801422,
+      "learning_rate": 0.00016266666666666667,
+      "loss": 0.6487,
+      "step": 123
+    },
+    {
+      "epoch": 0.03968,
+      "grad_norm": 0.11407941579818726,
+      "learning_rate": 0.000164,
+      "loss": 0.6822,
+      "step": 124
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.12307750433683395,
+      "learning_rate": 0.00016533333333333333,
+      "loss": 0.7038,
+      "step": 125
+    },
+    {
+      "epoch": 0.04032,
+      "grad_norm": 0.12420140206813812,
+      "learning_rate": 0.0001666666666666667,
+      "loss": 0.6589,
+      "step": 126
+    },
+    {
+      "epoch": 0.04064,
+      "grad_norm": 0.1327451914548874,
+      "learning_rate": 0.000168,
+      "loss": 0.6002,
+      "step": 127
+    },
+    {
+      "epoch": 0.04096,
+      "grad_norm": 0.1248873695731163,
+      "learning_rate": 0.00016933333333333335,
+      "loss": 0.8812,
+      "step": 128
+    },
+    {
+      "epoch": 0.04128,
+      "grad_norm": 0.11638613790273666,
+      "learning_rate": 0.00017066666666666668,
+      "loss": 0.7455,
+      "step": 129
+    },
+    {
+      "epoch": 0.0416,
+      "grad_norm": 0.12473749369382858,
+      "learning_rate": 0.000172,
+      "loss": 0.6218,
+      "step": 130
+    },
+    {
+      "epoch": 0.04192,
+      "grad_norm": 0.1339467465877533,
+      "learning_rate": 0.00017333333333333334,
+      "loss": 0.729,
+      "step": 131
+    },
+    {
+      "epoch": 0.04224,
+      "grad_norm": 0.11078035831451416,
+      "learning_rate": 0.00017466666666666667,
+      "loss": 0.8507,
+      "step": 132
+    },
+    {
+      "epoch": 0.04256,
+      "grad_norm": 0.11885277926921844,
+      "learning_rate": 0.00017600000000000002,
+      "loss": 0.8817,
+      "step": 133
+    },
+    {
+      "epoch": 0.04288,
+      "grad_norm": 0.13378757238388062,
+      "learning_rate": 0.00017733333333333335,
+      "loss": 0.705,
+      "step": 134
+    },
+    {
+      "epoch": 0.0432,
+      "grad_norm": 0.11696331948041916,
+      "learning_rate": 0.00017866666666666668,
+      "loss": 0.7089,
+      "step": 135
+    },
+    {
+      "epoch": 0.04352,
+      "grad_norm": 0.12444671988487244,
+      "learning_rate": 0.00018,
+      "loss": 0.699,
+      "step": 136
+    },
+    {
+      "epoch": 0.04384,
+      "grad_norm": 0.11136961728334427,
+      "learning_rate": 0.00018133333333333334,
+      "loss": 0.7805,
+      "step": 137
+    },
+    {
+      "epoch": 0.04416,
+      "grad_norm": 0.13461412489414215,
+      "learning_rate": 0.00018266666666666667,
+      "loss": 0.7581,
+      "step": 138
+    },
+    {
+      "epoch": 0.04448,
+      "grad_norm": 0.11344683915376663,
+      "learning_rate": 0.00018400000000000003,
+      "loss": 0.7234,
+      "step": 139
+    },
+    {
+      "epoch": 0.0448,
+      "grad_norm": 0.12247714400291443,
+      "learning_rate": 0.00018533333333333333,
+      "loss": 0.9326,
+      "step": 140
+    },
+    {
+      "epoch": 0.04512,
+      "grad_norm": 0.12991179525852203,
+      "learning_rate": 0.0001866666666666667,
+      "loss": 0.7644,
+      "step": 141
+    },
+    {
+      "epoch": 0.04544,
+      "grad_norm": 0.1182120144367218,
+      "learning_rate": 0.000188,
+      "loss": 0.8437,
+      "step": 142
+    },
+    {
+      "epoch": 0.04576,
+      "grad_norm": 0.10737155377864838,
+      "learning_rate": 0.00018933333333333335,
+      "loss": 0.8047,
+      "step": 143
+    },
+    {
+      "epoch": 0.04608,
+      "grad_norm": 0.11993440985679626,
+      "learning_rate": 0.00019066666666666668,
+      "loss": 0.7247,
+      "step": 144
+    },
+    {
+      "epoch": 0.0464,
+      "grad_norm": 0.11705442517995834,
+      "learning_rate": 0.000192,
+      "loss": 0.7104,
+      "step": 145
+    },
+    {
+      "epoch": 0.04672,
+      "grad_norm": 0.12065937370061874,
+      "learning_rate": 0.00019333333333333333,
+      "loss": 0.7243,
+      "step": 146
+    },
+    {
+      "epoch": 0.04704,
+      "grad_norm": 0.1178629919886589,
+      "learning_rate": 0.0001946666666666667,
+      "loss": 0.6946,
+      "step": 147
+    },
+    {
+      "epoch": 0.04736,
+      "grad_norm": 0.12744340300559998,
+      "learning_rate": 0.000196,
+      "loss": 0.6946,
+      "step": 148
+    },
+    {
+      "epoch": 0.04768,
+      "grad_norm": 0.14068365097045898,
+      "learning_rate": 0.00019733333333333335,
+      "loss": 0.7905,
+      "step": 149
+    },
+    {
+      "epoch": 0.048,
+      "grad_norm": 0.11114205420017242,
+      "learning_rate": 0.00019866666666666668,
+      "loss": 0.8352,
+      "step": 150
+    },
+    {
+      "epoch": 0.04832,
+      "grad_norm": 0.10463167726993561,
+      "learning_rate": 0.0002,
+      "loss": 0.6664,
+      "step": 151
+    },
+    {
+      "epoch": 0.04864,
+      "grad_norm": 0.12237433344125748,
+      "learning_rate": 0.00019995876288659794,
+      "loss": 0.8114,
+      "step": 152
+    },
+    {
+      "epoch": 0.04896,
+      "grad_norm": 0.12069843709468842,
+      "learning_rate": 0.0001999175257731959,
+      "loss": 0.7901,
+      "step": 153
+    },
+    {
+      "epoch": 0.04928,
+      "grad_norm": 0.11398322135210037,
+      "learning_rate": 0.00019987628865979383,
+      "loss": 0.7131,
+      "step": 154
+    },
+    {
+      "epoch": 0.0496,
+      "grad_norm": 0.12183695286512375,
+      "learning_rate": 0.00019983505154639176,
+      "loss": 0.7609,
+      "step": 155
+    },
+    {
+      "epoch": 0.04992,
+      "grad_norm": 0.12869080901145935,
+      "learning_rate": 0.0001997938144329897,
+      "loss": 0.8103,
+      "step": 156
+    },
+    {
+      "epoch": 0.05024,
+      "grad_norm": 0.1119738221168518,
+      "learning_rate": 0.00019975257731958762,
+      "loss": 0.6731,
+      "step": 157
+    },
+    {
+      "epoch": 0.05056,
+      "grad_norm": 0.12156844139099121,
+      "learning_rate": 0.00019971134020618558,
+      "loss": 0.7535,
+      "step": 158
+    },
+    {
+      "epoch": 0.05088,
+      "grad_norm": 0.12900514900684357,
+      "learning_rate": 0.00019967010309278351,
+      "loss": 0.7378,
+      "step": 159
+    },
+    {
+      "epoch": 0.0512,
+      "grad_norm": 0.12743592262268066,
+      "learning_rate": 0.00019962886597938147,
+      "loss": 0.6197,
+      "step": 160
+    },
+    {
+      "epoch": 0.05152,
+      "grad_norm": 0.128151997923851,
+      "learning_rate": 0.0001995876288659794,
+      "loss": 0.7812,
+      "step": 161
+    },
+    {
+      "epoch": 0.05184,
+      "grad_norm": 0.10753507912158966,
+      "learning_rate": 0.00019954639175257733,
+      "loss": 0.5774,
+      "step": 162
+    },
+    {
+      "epoch": 0.05216,
+      "grad_norm": 0.11813949793577194,
+      "learning_rate": 0.00019950515463917527,
+      "loss": 0.6894,
+      "step": 163
+    },
+    {
+      "epoch": 0.05248,
+      "grad_norm": 0.1340881586074829,
+      "learning_rate": 0.0001994639175257732,
+      "loss": 0.7193,
+      "step": 164
+    },
+    {
+      "epoch": 0.0528,
+      "grad_norm": 0.10451763868331909,
+      "learning_rate": 0.00019942268041237116,
+      "loss": 0.572,
+      "step": 165
+    },
+    {
+      "epoch": 0.05312,
+      "grad_norm": 0.10894517600536346,
+      "learning_rate": 0.0001993814432989691,
+      "loss": 0.7693,
+      "step": 166
+    },
+    {
+      "epoch": 0.05344,
+      "grad_norm": 0.1243942379951477,
+      "learning_rate": 0.00019934020618556702,
+      "loss": 0.7805,
+      "step": 167
+    },
+    {
+      "epoch": 0.05376,
+      "grad_norm": 0.14128854870796204,
+      "learning_rate": 0.00019929896907216498,
+      "loss": 0.8595,
+      "step": 168
+    },
+    {
+      "epoch": 0.05408,
+      "grad_norm": 0.12148380279541016,
+      "learning_rate": 0.00019925773195876288,
+      "loss": 0.8954,
+      "step": 169
+    },
+    {
+      "epoch": 0.0544,
+      "grad_norm": 0.10860492289066315,
+      "learning_rate": 0.0001992164948453608,
+      "loss": 0.8101,
+      "step": 170
+    },
+    {
+      "epoch": 0.05472,
+      "grad_norm": 0.12292741239070892,
+      "learning_rate": 0.00019917525773195877,
+      "loss": 0.7965,
+      "step": 171
+    },
+    {
+      "epoch": 0.05504,
+      "grad_norm": 0.13840395212173462,
+      "learning_rate": 0.0001991340206185567,
+      "loss": 0.7231,
+      "step": 172
+    },
+    {
+      "epoch": 0.05536,
+      "grad_norm": 0.11291699856519699,
+      "learning_rate": 0.00019909278350515466,
+      "loss": 0.8247,
+      "step": 173
+    },
+    {
+      "epoch": 0.05568,
+      "grad_norm": 0.11811841279268265,
+      "learning_rate": 0.0001990515463917526,
+      "loss": 0.8504,
+      "step": 174
+    },
+    {
+      "epoch": 0.056,
+      "grad_norm": 0.11040613055229187,
+      "learning_rate": 0.00019901030927835052,
+      "loss": 0.6464,
+      "step": 175
+    },
+    {
+      "epoch": 0.05632,
+      "grad_norm": 0.10370033234357834,
+      "learning_rate": 0.00019896907216494845,
+      "loss": 0.8016,
+      "step": 176
+    },
+    {
+      "epoch": 0.05664,
+      "grad_norm": 0.11868111789226532,
+      "learning_rate": 0.00019892783505154639,
+      "loss": 0.7267,
+      "step": 177
+    },
+    {
+      "epoch": 0.05696,
+      "grad_norm": 0.11893659085035324,
+      "learning_rate": 0.00019888659793814434,
+      "loss": 0.814,
+      "step": 178
+    },
+    {
+      "epoch": 0.05728,
+      "grad_norm": 0.11829444020986557,
+      "learning_rate": 0.00019884536082474227,
+      "loss": 0.7852,
+      "step": 179
+    },
+    {
+      "epoch": 0.0576,
+      "grad_norm": 0.10493418574333191,
+      "learning_rate": 0.0001988041237113402,
+      "loss": 0.6154,
+      "step": 180
+    },
+    {
+      "epoch": 0.05792,
+      "grad_norm": 0.10895421355962753,
+      "learning_rate": 0.00019876288659793816,
+      "loss": 0.8663,
+      "step": 181
+    },
+    {
+      "epoch": 0.05824,
+      "grad_norm": 0.11782484501600266,
+      "learning_rate": 0.0001987216494845361,
+      "loss": 0.7902,
+      "step": 182
+    },
+    {
+      "epoch": 0.05856,
+      "grad_norm": 0.12639066576957703,
+      "learning_rate": 0.00019868041237113403,
+      "loss": 0.7381,
+      "step": 183
+    },
+    {
+      "epoch": 0.05888,
+      "grad_norm": 0.1321602165699005,
+      "learning_rate": 0.00019863917525773196,
+      "loss": 0.9283,
+      "step": 184
+    },
+    {
+      "epoch": 0.0592,
+      "grad_norm": 0.12468945235013962,
+      "learning_rate": 0.0001985979381443299,
+      "loss": 0.6641,
+      "step": 185
+    },
+    {
+      "epoch": 0.05952,
+      "grad_norm": 0.12850762903690338,
+      "learning_rate": 0.00019855670103092785,
+      "loss": 0.8694,
+      "step": 186
+    },
+    {
+      "epoch": 0.05984,
+      "grad_norm": 0.10586417466402054,
+      "learning_rate": 0.00019851546391752578,
+      "loss": 0.806,
+      "step": 187
+    },
+    {
+      "epoch": 0.06016,
+      "grad_norm": 0.10202761739492416,
+      "learning_rate": 0.00019847422680412374,
+      "loss": 0.684,
+      "step": 188
+    },
+    {
+      "epoch": 0.06048,
+      "grad_norm": 0.11566773056983948,
+      "learning_rate": 0.00019843298969072167,
+      "loss": 0.6737,
+      "step": 189
+    },
+    {
+      "epoch": 0.0608,
+      "grad_norm": 0.12295495718717575,
+      "learning_rate": 0.0001983917525773196,
+      "loss": 0.6539,
+      "step": 190
+    },
+    {
+      "epoch": 0.06112,
+      "grad_norm": 0.14126336574554443,
+      "learning_rate": 0.00019835051546391753,
+      "loss": 0.8539,
+      "step": 191
+    },
+    {
+      "epoch": 0.06144,
+      "grad_norm": 0.1215134933590889,
+      "learning_rate": 0.00019830927835051546,
+      "loss": 0.7051,
+      "step": 192
+    },
+    {
+      "epoch": 0.06176,
+      "grad_norm": 0.10785145312547684,
+      "learning_rate": 0.00019826804123711342,
+      "loss": 0.7877,
+      "step": 193
+    },
+    {
+      "epoch": 0.06208,
+      "grad_norm": 0.1299588680267334,
+      "learning_rate": 0.00019822680412371135,
+      "loss": 0.7942,
+      "step": 194
+    },
+    {
+      "epoch": 0.0624,
+      "grad_norm": 0.1072564497590065,
+      "learning_rate": 0.00019818556701030928,
+      "loss": 0.7055,
+      "step": 195
+    },
+    {
+      "epoch": 0.06272,
+      "grad_norm": 0.12069325149059296,
+      "learning_rate": 0.00019814432989690724,
+      "loss": 0.9036,
+      "step": 196
+    },
+    {
+      "epoch": 0.06304,
+      "grad_norm": 0.14454863965511322,
+      "learning_rate": 0.00019810309278350517,
+      "loss": 0.7567,
+      "step": 197
+    },
+    {
+      "epoch": 0.06336,
+      "grad_norm": 0.11882466822862625,
+      "learning_rate": 0.0001980618556701031,
+      "loss": 0.5986,
+      "step": 198
+    },
+    {
+      "epoch": 0.06368,
+      "grad_norm": 0.12222633510828018,
+      "learning_rate": 0.00019802061855670104,
+      "loss": 0.8493,
+      "step": 199
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 0.10818106681108475,
+      "learning_rate": 0.00019797938144329897,
+      "loss": 0.7932,
+      "step": 200
+    },
+    {
+      "epoch": 0.06432,
+      "grad_norm": 0.1225774884223938,
+      "learning_rate": 0.00019793814432989693,
+      "loss": 0.7501,
+      "step": 201
+    },
+    {
+      "epoch": 0.06464,
+      "grad_norm": 0.13827428221702576,
+      "learning_rate": 0.00019789690721649486,
+      "loss": 0.818,
+      "step": 202
+    },
+    {
+      "epoch": 0.06496,
+      "grad_norm": 0.1145445853471756,
+      "learning_rate": 0.0001978556701030928,
+      "loss": 0.6433,
+      "step": 203
+    },
+    {
+      "epoch": 0.06528,
+      "grad_norm": 0.11928769201040268,
+      "learning_rate": 0.00019781443298969075,
+      "loss": 0.7352,
+      "step": 204
+    },
+    {
+      "epoch": 0.0656,
+      "grad_norm": 0.11696071177721024,
+      "learning_rate": 0.00019777319587628865,
+      "loss": 0.8499,
+      "step": 205
+    },
+    {
+      "epoch": 0.06592,
+      "grad_norm": 0.1281200349330902,
+      "learning_rate": 0.0001977319587628866,
+      "loss": 0.7526,
+      "step": 206
+    },
+    {
+      "epoch": 0.06624,
+      "grad_norm": 0.1270955204963684,
+      "learning_rate": 0.00019769072164948454,
+      "loss": 0.8456,
+      "step": 207
+    },
+    {
+      "epoch": 0.06656,
+      "grad_norm": 0.12278109043836594,
+      "learning_rate": 0.00019764948453608247,
+      "loss": 0.5972,
+      "step": 208
+    },
+    {
+      "epoch": 0.06688,
+      "grad_norm": 0.10781408101320267,
+      "learning_rate": 0.00019760824742268043,
+      "loss": 0.5967,
+      "step": 209
+    },
+    {
+      "epoch": 0.0672,
+      "grad_norm": 0.12237494438886642,
+      "learning_rate": 0.00019756701030927836,
+      "loss": 0.7306,
+      "step": 210
+    },
+    {
+      "epoch": 0.06752,
+      "grad_norm": 0.1100955531001091,
+      "learning_rate": 0.00019752577319587632,
+      "loss": 0.7579,
+      "step": 211
+    },
+    {
+      "epoch": 0.06784,
+      "grad_norm": 0.11036559194326401,
+      "learning_rate": 0.00019748453608247422,
+      "loss": 0.6668,
+      "step": 212
+    },
+    {
+      "epoch": 0.06816,
+      "grad_norm": 0.14552150666713715,
+      "learning_rate": 0.00019744329896907216,
+      "loss": 0.6668,
+      "step": 213
+    },
+    {
+      "epoch": 0.06848,
+      "grad_norm": 0.11125028133392334,
+      "learning_rate": 0.0001974020618556701,
+      "loss": 0.7342,
+      "step": 214
+    },
+    {
+      "epoch": 0.0688,
+      "grad_norm": 0.14329232275485992,
+      "learning_rate": 0.00019736082474226804,
+      "loss": 0.8708,
+      "step": 215
+    },
+    {
+      "epoch": 0.06912,
+      "grad_norm": 0.11911016702651978,
+      "learning_rate": 0.000197319587628866,
+      "loss": 0.7224,
+      "step": 216
+    },
+    {
+      "epoch": 0.06944,
+      "grad_norm": 0.12713277339935303,
+      "learning_rate": 0.00019727835051546393,
+      "loss": 0.7222,
+      "step": 217
+    },
+    {
+      "epoch": 0.06976,
+      "grad_norm": 0.1140274852514267,
+      "learning_rate": 0.00019723711340206187,
+      "loss": 0.627,
+      "step": 218
+    },
+    {
+      "epoch": 0.07008,
+      "grad_norm": 0.1070765033364296,
+      "learning_rate": 0.0001971958762886598,
+      "loss": 0.7514,
+      "step": 219
+    },
+    {
+      "epoch": 0.0704,
+      "grad_norm": 0.09937503188848495,
+      "learning_rate": 0.00019715463917525773,
+      "loss": 0.7539,
+      "step": 220
+    },
+    {
+      "epoch": 0.07072,
+      "grad_norm": 0.12137589603662491,
+      "learning_rate": 0.0001971134020618557,
+      "loss": 0.68,
+      "step": 221
+    },
+    {
+      "epoch": 0.07104,
+      "grad_norm": 0.11091487109661102,
+      "learning_rate": 0.00019707216494845362,
+      "loss": 0.6904,
+      "step": 222
+    },
+    {
+      "epoch": 0.07136,
+      "grad_norm": 0.10996535420417786,
+      "learning_rate": 0.00019703092783505155,
+      "loss": 0.6854,
+      "step": 223
+    },
+    {
+      "epoch": 0.07168,
+      "grad_norm": 0.129950150847435,
+      "learning_rate": 0.0001969896907216495,
+      "loss": 0.7023,
+      "step": 224
+    },
+    {
+      "epoch": 0.072,
+      "grad_norm": 0.13028115034103394,
+      "learning_rate": 0.00019694845360824744,
+      "loss": 0.6479,
+      "step": 225
+    },
+    {
+      "epoch": 0.07232,
+      "grad_norm": 0.1166912168264389,
+      "learning_rate": 0.00019690721649484537,
+      "loss": 0.6498,
+      "step": 226
+    },
+    {
+      "epoch": 0.07264,
+      "grad_norm": 0.1148250624537468,
+      "learning_rate": 0.0001968659793814433,
+      "loss": 0.7461,
+      "step": 227
+    },
+    {
+      "epoch": 0.07296,
+      "grad_norm": 0.12639757990837097,
+      "learning_rate": 0.00019682474226804123,
+      "loss": 0.712,
+      "step": 228
+    },
+    {
+      "epoch": 0.07328,
+      "grad_norm": 0.13229860365390778,
+      "learning_rate": 0.0001967835051546392,
+      "loss": 0.7447,
+      "step": 229
+    },
+    {
+      "epoch": 0.0736,
+      "grad_norm": 0.11198769509792328,
+      "learning_rate": 0.00019674226804123712,
+      "loss": 0.7329,
+      "step": 230
+    },
+    {
+      "epoch": 0.07392,
+      "grad_norm": 0.11425165086984634,
+      "learning_rate": 0.00019670103092783505,
+      "loss": 0.868,
+      "step": 231
+    },
+    {
+      "epoch": 0.07424,
+      "grad_norm": 0.12034732848405838,
+      "learning_rate": 0.000196659793814433,
+      "loss": 0.7276,
+      "step": 232
+    },
+    {
+      "epoch": 0.07456,
+      "grad_norm": 0.10866102576255798,
+      "learning_rate": 0.00019661855670103094,
+      "loss": 0.7338,
+      "step": 233
+    },
+    {
+      "epoch": 0.07488,
+      "grad_norm": 0.12432979047298431,
+      "learning_rate": 0.00019657731958762887,
+      "loss": 0.8257,
+      "step": 234
+    },
+    {
+      "epoch": 0.0752,
+      "grad_norm": 0.12356464564800262,
+      "learning_rate": 0.0001965360824742268,
+      "loss": 0.6813,
+      "step": 235
+    },
+    {
+      "epoch": 0.07552,
+      "grad_norm": 0.1002034917473793,
+      "learning_rate": 0.00019649484536082474,
+      "loss": 0.7406,
+      "step": 236
+    },
+    {
+      "epoch": 0.07584,
+      "grad_norm": 0.11528842151165009,
+      "learning_rate": 0.0001964536082474227,
+      "loss": 0.7597,
+      "step": 237
+    },
+    {
+      "epoch": 0.07616,
+      "grad_norm": 0.11711521446704865,
+      "learning_rate": 0.00019641237113402063,
+      "loss": 0.6925,
+      "step": 238
+    },
+    {
+      "epoch": 0.07648,
+      "grad_norm": 0.09948590397834778,
+      "learning_rate": 0.00019637113402061859,
+      "loss": 0.8017,
+      "step": 239
+    },
+    {
+      "epoch": 0.0768,
+      "grad_norm": 0.11264682561159134,
+      "learning_rate": 0.00019632989690721652,
+      "loss": 0.6871,
+      "step": 240
+    },
+    {
+      "epoch": 0.07712,
+      "grad_norm": 0.12238943576812744,
+      "learning_rate": 0.00019628865979381442,
+      "loss": 0.7113,
+      "step": 241
+    },
+    {
+      "epoch": 0.07744,
+      "grad_norm": 0.11449657380580902,
+      "learning_rate": 0.00019624742268041238,
+      "loss": 0.7144,
+      "step": 242
+    },
+    {
+      "epoch": 0.07776,
+      "grad_norm": 0.12212596088647842,
+      "learning_rate": 0.0001962061855670103,
+      "loss": 0.697,
+      "step": 243
+    },
+    {
+      "epoch": 0.07808,
+      "grad_norm": 0.12231374531984329,
+      "learning_rate": 0.00019616494845360827,
+      "loss": 0.6955,
+      "step": 244
+    },
+    {
+      "epoch": 0.0784,
+      "grad_norm": 0.11105109006166458,
+      "learning_rate": 0.0001961237113402062,
+      "loss": 0.8251,
+      "step": 245
+    },
+    {
+      "epoch": 0.07872,
+      "grad_norm": 0.12369677424430847,
+      "learning_rate": 0.00019608247422680413,
+      "loss": 0.8814,
+      "step": 246
+    },
+    {
+      "epoch": 0.07904,
+      "grad_norm": 0.12226764857769012,
+      "learning_rate": 0.0001960412371134021,
+      "loss": 0.7408,
+      "step": 247
+    },
+    {
+      "epoch": 0.07936,
+      "grad_norm": 0.10316195338964462,
+      "learning_rate": 0.000196,
+      "loss": 0.792,
+      "step": 248
+    },
+    {
+      "epoch": 0.07968,
+      "grad_norm": 0.12036872655153275,
+      "learning_rate": 0.00019595876288659795,
+      "loss": 0.6,
+      "step": 249
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.1140509843826294,
+      "learning_rate": 0.00019591752577319588,
+      "loss": 0.7599,
+      "step": 250
+    },
+    {
+      "epoch": 0.08032,
+      "grad_norm": 0.14200666546821594,
+      "learning_rate": 0.00019587628865979381,
+      "loss": 0.7723,
+      "step": 251
+    },
+    {
+      "epoch": 0.08064,
+      "grad_norm": 0.11529050767421722,
+      "learning_rate": 0.00019583505154639177,
+      "loss": 0.8305,
+      "step": 252
+    },
+    {
+      "epoch": 0.08096,
+      "grad_norm": 0.11324192583560944,
+      "learning_rate": 0.0001957938144329897,
+      "loss": 0.7913,
+      "step": 253
+    },
+    {
+      "epoch": 0.08128,
+      "grad_norm": 0.12280263006687164,
+      "learning_rate": 0.00019575257731958764,
+      "loss": 0.8388,
+      "step": 254
+    },
+    {
+      "epoch": 0.0816,
+      "grad_norm": 0.10810542851686478,
+      "learning_rate": 0.00019571134020618557,
+      "loss": 0.7025,
+      "step": 255
+    },
+    {
+      "epoch": 0.08192,
+      "grad_norm": 0.11314940452575684,
+      "learning_rate": 0.0001956701030927835,
+      "loss": 0.6696,
+      "step": 256
+    },
+    {
+      "epoch": 0.08224,
+      "grad_norm": 0.10368701070547104,
+      "learning_rate": 0.00019562886597938146,
+      "loss": 0.581,
+      "step": 257
+    },
+    {
+      "epoch": 0.08256,
+      "grad_norm": 0.11146122217178345,
+      "learning_rate": 0.0001955876288659794,
+      "loss": 0.8221,
+      "step": 258
+    },
+    {
+      "epoch": 0.08288,
+      "grad_norm": 0.11592487245798111,
+      "learning_rate": 0.00019554639175257732,
+      "loss": 0.8082,
+      "step": 259
+    },
+    {
+      "epoch": 0.0832,
+      "grad_norm": 0.11007897555828094,
+      "learning_rate": 0.00019550515463917528,
+      "loss": 0.8943,
+      "step": 260
+    },
+    {
+      "epoch": 0.08352,
+      "grad_norm": 0.11887793987989426,
+      "learning_rate": 0.0001954639175257732,
+      "loss": 0.7447,
+      "step": 261
+    },
+    {
+      "epoch": 0.08384,
+      "grad_norm": 0.10715821385383606,
+      "learning_rate": 0.00019542268041237114,
+      "loss": 0.6405,
+      "step": 262
+    },
+    {
+      "epoch": 0.08416,
+      "grad_norm": 0.11692032217979431,
+      "learning_rate": 0.00019538144329896907,
+      "loss": 0.863,
+      "step": 263
+    },
+    {
+      "epoch": 0.08448,
+      "grad_norm": 0.11917733401060104,
+      "learning_rate": 0.000195340206185567,
+      "loss": 0.6976,
+      "step": 264
+    },
+    {
+      "epoch": 0.0848,
+      "grad_norm": 0.1289709061384201,
+      "learning_rate": 0.00019529896907216496,
+      "loss": 0.9149,
+      "step": 265
+    },
+    {
+      "epoch": 0.08512,
+      "grad_norm": 0.10539647191762924,
+      "learning_rate": 0.0001952577319587629,
+      "loss": 0.8449,
+      "step": 266
+    },
+    {
+      "epoch": 0.08544,
+      "grad_norm": 0.10908807814121246,
+      "learning_rate": 0.00019521649484536085,
+      "loss": 0.7162,
+      "step": 267
+    },
+    {
+      "epoch": 0.08576,
+      "grad_norm": 0.10654058307409286,
+      "learning_rate": 0.00019517525773195878,
+      "loss": 0.6285,
+      "step": 268
+    },
+    {
+      "epoch": 0.08608,
+      "grad_norm": 0.111301951110363,
+      "learning_rate": 0.0001951340206185567,
+      "loss": 0.755,
+      "step": 269
+    },
+    {
+      "epoch": 0.0864,
+      "grad_norm": 0.10679474472999573,
+      "learning_rate": 0.00019509278350515464,
+      "loss": 0.7782,
+      "step": 270
+    },
+    {
+      "epoch": 0.08672,
+      "grad_norm": 0.10793033987283707,
+      "learning_rate": 0.00019505154639175258,
+      "loss": 0.7021,
+      "step": 271
+    },
+    {
+      "epoch": 0.08704,
+      "grad_norm": 0.13048402965068817,
+      "learning_rate": 0.00019501030927835053,
+      "loss": 0.7575,
+      "step": 272
+    },
+    {
+      "epoch": 0.08736,
+      "grad_norm": 0.11437942832708359,
+      "learning_rate": 0.00019496907216494847,
+      "loss": 0.7925,
+      "step": 273
+    },
+    {
+      "epoch": 0.08768,
+      "grad_norm": 0.1390543431043625,
+      "learning_rate": 0.0001949278350515464,
+      "loss": 0.8013,
+      "step": 274
+    },
+    {
+      "epoch": 0.088,
+      "grad_norm": 0.1289997398853302,
+      "learning_rate": 0.00019488659793814435,
+      "loss": 0.6053,
+      "step": 275
+    },
+    {
+      "epoch": 0.08832,
+      "grad_norm": 0.09682965278625488,
+      "learning_rate": 0.00019484536082474229,
+      "loss": 0.8756,
+      "step": 276
+    },
+    {
+      "epoch": 0.08864,
+      "grad_norm": 0.12752555310726166,
+      "learning_rate": 0.00019480412371134022,
+      "loss": 0.6314,
+      "step": 277
+    },
+    {
+      "epoch": 0.08896,
+      "grad_norm": 0.11673011630773544,
+      "learning_rate": 0.00019476288659793815,
+      "loss": 0.8842,
+      "step": 278
+    },
+    {
+      "epoch": 0.08928,
+      "grad_norm": 0.108588308095932,
+      "learning_rate": 0.00019472164948453608,
+      "loss": 0.6348,
+      "step": 279
+    },
+    {
+      "epoch": 0.0896,
+      "grad_norm": 0.1195591390132904,
+      "learning_rate": 0.00019468041237113404,
+      "loss": 0.8097,
+      "step": 280
+    },
+    {
+      "epoch": 0.08992,
+      "grad_norm": 0.11936333775520325,
+      "learning_rate": 0.00019463917525773197,
+      "loss": 0.7858,
+      "step": 281
+    },
+    {
+      "epoch": 0.09024,
+      "grad_norm": 0.11627551168203354,
+      "learning_rate": 0.00019459793814432993,
+      "loss": 0.7322,
+      "step": 282
+    },
+    {
+      "epoch": 0.09056,
+      "grad_norm": 0.10832615196704865,
+      "learning_rate": 0.00019455670103092786,
+      "loss": 0.6215,
+      "step": 283
+    },
+    {
+      "epoch": 0.09088,
+      "grad_norm": 0.11125192046165466,
+      "learning_rate": 0.00019451546391752576,
+      "loss": 0.8058,
+      "step": 284
+    },
+    {
+      "epoch": 0.0912,
+      "grad_norm": 0.11165905743837357,
+      "learning_rate": 0.00019447422680412372,
+      "loss": 0.7104,
+      "step": 285
+    },
+    {
+      "epoch": 0.09152,
+      "grad_norm": 0.10910595208406448,
+      "learning_rate": 0.00019443298969072165,
+      "loss": 0.707,
+      "step": 286
+    },
+    {
+      "epoch": 0.09184,
+      "grad_norm": 0.11235277354717255,
+      "learning_rate": 0.0001943917525773196,
+      "loss": 0.8161,
+      "step": 287
+    },
+    {
+      "epoch": 0.09216,
+      "grad_norm": 0.11668789386749268,
+      "learning_rate": 0.00019435051546391754,
+      "loss": 0.6641,
+      "step": 288
+    },
+    {
+      "epoch": 0.09248,
+      "grad_norm": 0.11274772882461548,
+      "learning_rate": 0.00019430927835051547,
+      "loss": 0.6987,
+      "step": 289
+    },
+    {
+      "epoch": 0.0928,
+      "grad_norm": 0.12356577813625336,
+      "learning_rate": 0.0001942680412371134,
+      "loss": 0.7051,
+      "step": 290
+    },
+    {
+      "epoch": 0.09312,
+      "grad_norm": 0.0987682193517685,
+      "learning_rate": 0.00019422680412371134,
+      "loss": 0.7218,
+      "step": 291
+    },
+    {
+      "epoch": 0.09344,
+      "grad_norm": 0.12354995310306549,
+      "learning_rate": 0.00019418556701030927,
+      "loss": 0.7379,
+      "step": 292
+    },
+    {
+      "epoch": 0.09376,
+      "grad_norm": 0.1022811159491539,
+      "learning_rate": 0.00019414432989690723,
+      "loss": 0.65,
+      "step": 293
+    },
+    {
+      "epoch": 0.09408,
+      "grad_norm": 0.11507881432771683,
+      "learning_rate": 0.00019410309278350516,
+      "loss": 0.8681,
+      "step": 294
+    },
+    {
+      "epoch": 0.0944,
+      "grad_norm": 0.12279893457889557,
+      "learning_rate": 0.00019406185567010312,
+      "loss": 0.6837,
+      "step": 295
+    },
+    {
+      "epoch": 0.09472,
+      "grad_norm": 0.10548900812864304,
+      "learning_rate": 0.00019402061855670105,
+      "loss": 0.669,
+      "step": 296
+    },
+    {
+      "epoch": 0.09504,
+      "grad_norm": 0.1302396059036255,
+      "learning_rate": 0.00019397938144329898,
+      "loss": 0.7969,
+      "step": 297
+    },
+    {
+      "epoch": 0.09536,
+      "grad_norm": 0.11492603272199631,
+      "learning_rate": 0.0001939381443298969,
+      "loss": 1.0184,
+      "step": 298
+    },
+    {
+      "epoch": 0.09568,
+      "grad_norm": 0.131149023771286,
+      "learning_rate": 0.00019389690721649484,
+      "loss": 0.8521,
+      "step": 299
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 0.1252460777759552,
+      "learning_rate": 0.0001938556701030928,
+      "loss": 0.8791,
+      "step": 300
+    },
+    {
+      "epoch": 0.09632,
+      "grad_norm": 0.11322636157274246,
+      "learning_rate": 0.00019381443298969073,
+      "loss": 0.7305,
+      "step": 301
+    },
+    {
+      "epoch": 0.09664,
+      "grad_norm": 0.10373307764530182,
+      "learning_rate": 0.00019377319587628866,
+      "loss": 0.7844,
+      "step": 302
+    },
+    {
+      "epoch": 0.09696,
+      "grad_norm": 0.13182489573955536,
+      "learning_rate": 0.00019373195876288662,
+      "loss": 0.6581,
+      "step": 303
+    },
+    {
+      "epoch": 0.09728,
+      "grad_norm": 0.1276710480451584,
+      "learning_rate": 0.00019369072164948455,
+      "loss": 0.6664,
+      "step": 304
+    },
+    {
+      "epoch": 0.0976,
+      "grad_norm": 0.10770755261182785,
+      "learning_rate": 0.00019364948453608248,
+      "loss": 0.6315,
+      "step": 305
+    },
+    {
+      "epoch": 0.09792,
+      "grad_norm": 0.11755330115556717,
+      "learning_rate": 0.00019360824742268041,
+      "loss": 0.7562,
+      "step": 306
+    },
+    {
+      "epoch": 0.09824,
+      "grad_norm": 0.1252545267343521,
+      "learning_rate": 0.00019356701030927835,
+      "loss": 0.6606,
+      "step": 307
+    },
+    {
+      "epoch": 0.09856,
+      "grad_norm": 0.12284315377473831,
+      "learning_rate": 0.0001935257731958763,
+      "loss": 0.8037,
+      "step": 308
+    },
+    {
+      "epoch": 0.09888,
+      "grad_norm": 0.106113962829113,
+      "learning_rate": 0.00019348453608247424,
+      "loss": 0.6797,
+      "step": 309
+    },
+    {
+      "epoch": 0.0992,
+      "grad_norm": 0.09285733103752136,
+      "learning_rate": 0.0001934432989690722,
+      "loss": 0.6927,
+      "step": 310
+    },
+    {
+      "epoch": 0.09952,
+      "grad_norm": 0.10903103649616241,
+      "learning_rate": 0.00019340206185567012,
+      "loss": 0.7863,
+      "step": 311
+    },
+    {
+      "epoch": 0.09984,
+      "grad_norm": 0.12015190720558167,
+      "learning_rate": 0.00019336082474226806,
+      "loss": 0.6961,
+      "step": 312
+    },
+    {
+      "epoch": 0.10016,
+      "grad_norm": 0.12260061502456665,
+      "learning_rate": 0.000193319587628866,
+      "loss": 0.7965,
+      "step": 313
+    },
+    {
+      "epoch": 0.10048,
+      "grad_norm": 0.10297476500272751,
+      "learning_rate": 0.00019327835051546392,
+      "loss": 0.6935,
+      "step": 314
+    },
+    {
+      "epoch": 0.1008,
+      "grad_norm": 0.12614290416240692,
+      "learning_rate": 0.00019323711340206188,
+      "loss": 0.8751,
+      "step": 315
+    },
+    {
+      "epoch": 0.10112,
+      "grad_norm": 0.10241006314754486,
+      "learning_rate": 0.0001931958762886598,
+      "loss": 0.5642,
+      "step": 316
+    },
+    {
+      "epoch": 0.10144,
+      "grad_norm": 0.1053333505988121,
+      "learning_rate": 0.00019315463917525774,
+      "loss": 0.6425,
+      "step": 317
+    },
+    {
+      "epoch": 0.10176,
+      "grad_norm": 0.1024412289261818,
+      "learning_rate": 0.0001931134020618557,
+      "loss": 0.6414,
+      "step": 318
+    },
+    {
+      "epoch": 0.10208,
+      "grad_norm": 0.10628654062747955,
+      "learning_rate": 0.0001930721649484536,
+      "loss": 0.8259,
+      "step": 319
+    },
+    {
+      "epoch": 0.1024,
+      "grad_norm": 0.1291695237159729,
+      "learning_rate": 0.00019303092783505153,
+      "loss": 0.6887,
+      "step": 320
+    },
+    {
+      "epoch": 0.10272,
+      "grad_norm": 0.11315542459487915,
+      "learning_rate": 0.0001929896907216495,
+      "loss": 0.6316,
+      "step": 321
+    },
+    {
+      "epoch": 0.10304,
+      "grad_norm": 0.11541493982076645,
+      "learning_rate": 0.00019294845360824742,
+      "loss": 0.8023,
+      "step": 322
+    },
+    {
+      "epoch": 0.10336,
+      "grad_norm": 0.09769251197576523,
+      "learning_rate": 0.00019290721649484538,
+      "loss": 0.674,
+      "step": 323
+    },
+    {
+      "epoch": 0.10368,
+      "grad_norm": 0.10549357533454895,
+      "learning_rate": 0.0001928659793814433,
+      "loss": 0.7102,
+      "step": 324
+    },
+    {
+      "epoch": 0.104,
+      "grad_norm": 0.1344941407442093,
+      "learning_rate": 0.00019282474226804124,
+      "loss": 0.8861,
+      "step": 325
+    },
+    {
+      "epoch": 0.10432,
+      "grad_norm": 0.13625279068946838,
+      "learning_rate": 0.00019278350515463918,
+      "loss": 0.8333,
+      "step": 326
+    },
+    {
+      "epoch": 0.10464,
+      "grad_norm": 0.1136624813079834,
+      "learning_rate": 0.0001927422680412371,
+      "loss": 0.7187,
+      "step": 327
+    },
+    {
+      "epoch": 0.10496,
+      "grad_norm": 0.12351350486278534,
+      "learning_rate": 0.00019270103092783506,
+      "loss": 0.6361,
+      "step": 328
+    },
+    {
+      "epoch": 0.10528,
+      "grad_norm": 0.12927868962287903,
+      "learning_rate": 0.000192659793814433,
+      "loss": 0.6716,
+      "step": 329
+    },
+    {
+      "epoch": 0.1056,
+      "grad_norm": 0.11619272083044052,
+      "learning_rate": 0.00019261855670103093,
+      "loss": 0.8108,
+      "step": 330
+    },
+    {
+      "epoch": 0.10592,
+      "grad_norm": 0.12765595316886902,
+      "learning_rate": 0.00019257731958762889,
+      "loss": 0.77,
+      "step": 331
+    },
+    {
+      "epoch": 0.10624,
+      "grad_norm": 0.13606730103492737,
+      "learning_rate": 0.00019253608247422682,
+      "loss": 0.7572,
+      "step": 332
+    },
+    {
+      "epoch": 0.10656,
+      "grad_norm": 0.1074293777346611,
+      "learning_rate": 0.00019249484536082475,
+      "loss": 0.8137,
+      "step": 333
+    },
+    {
+      "epoch": 0.10688,
+      "grad_norm": 0.12371189147233963,
+      "learning_rate": 0.00019245360824742268,
+      "loss": 0.7414,
+      "step": 334
+    },
+    {
+      "epoch": 0.1072,
+      "grad_norm": 0.1253589391708374,
+      "learning_rate": 0.0001924123711340206,
+      "loss": 0.6796,
+      "step": 335
+    },
+    {
+      "epoch": 0.10752,
+      "grad_norm": 0.11865826696157455,
+      "learning_rate": 0.00019237113402061857,
+      "loss": 0.7051,
+      "step": 336
+    },
+    {
+      "epoch": 0.10784,
+      "grad_norm": 0.12180513888597488,
+      "learning_rate": 0.0001923298969072165,
+      "loss": 0.8126,
+      "step": 337
+    },
+    {
+      "epoch": 0.10816,
+      "grad_norm": 0.11744590848684311,
+      "learning_rate": 0.00019228865979381446,
+      "loss": 0.7553,
+      "step": 338
+    },
+    {
+      "epoch": 0.10848,
+      "grad_norm": 0.11796604096889496,
+      "learning_rate": 0.0001922474226804124,
+      "loss": 0.814,
+      "step": 339
+    },
+    {
+      "epoch": 0.1088,
+      "grad_norm": 0.125954270362854,
+      "learning_rate": 0.00019220618556701032,
+      "loss": 0.64,
+      "step": 340
+    },
+    {
+      "epoch": 0.10912,
+      "grad_norm": 0.12020507454872131,
+      "learning_rate": 0.00019216494845360825,
+      "loss": 0.6175,
+      "step": 341
+    },
+    {
+      "epoch": 0.10944,
+      "grad_norm": 0.10076280683279037,
+      "learning_rate": 0.00019212371134020618,
+      "loss": 0.6176,
+      "step": 342
+    },
+    {
+      "epoch": 0.10976,
+      "grad_norm": 0.11676807701587677,
+      "learning_rate": 0.00019208247422680414,
+      "loss": 0.677,
+      "step": 343
+    },
+    {
+      "epoch": 0.11008,
+      "grad_norm": 0.11364617198705673,
+      "learning_rate": 0.00019204123711340207,
+      "loss": 0.5619,
+      "step": 344
+    },
+    {
+      "epoch": 0.1104,
+      "grad_norm": 0.10261007398366928,
+      "learning_rate": 0.000192,
+      "loss": 0.614,
+      "step": 345
+    },
+    {
+      "epoch": 0.11072,
+      "grad_norm": 0.09645744413137436,
+      "learning_rate": 0.00019195876288659796,
+      "loss": 0.7274,
+      "step": 346
+    },
+    {
+      "epoch": 0.11104,
+      "grad_norm": 0.12451738864183426,
+      "learning_rate": 0.0001919175257731959,
+      "loss": 0.8735,
+      "step": 347
+    },
+    {
+      "epoch": 0.11136,
+      "grad_norm": 0.11062555015087128,
+      "learning_rate": 0.00019187628865979383,
+      "loss": 0.6268,
+      "step": 348
+    },
+    {
+      "epoch": 0.11168,
+      "grad_norm": 0.1201324462890625,
+      "learning_rate": 0.00019183505154639176,
+      "loss": 0.793,
+      "step": 349
+    },
+    {
+      "epoch": 0.112,
+      "grad_norm": 0.12797708809375763,
+      "learning_rate": 0.0001917938144329897,
+      "loss": 0.6075,
+      "step": 350
+    },
+    {
+      "epoch": 0.11232,
+      "grad_norm": 0.11388446390628815,
+      "learning_rate": 0.00019175257731958765,
+      "loss": 0.7343,
+      "step": 351
+    },
+    {
+      "epoch": 0.11264,
+      "grad_norm": 0.11379895359277725,
+      "learning_rate": 0.00019171134020618558,
+      "loss": 0.6767,
+      "step": 352
+    },
+    {
+      "epoch": 0.11296,
+      "grad_norm": 0.11296231299638748,
+      "learning_rate": 0.0001916701030927835,
+      "loss": 0.7473,
+      "step": 353
+    },
+    {
+      "epoch": 0.11328,
+      "grad_norm": 0.1344781517982483,
+      "learning_rate": 0.00019162886597938147,
+      "loss": 0.6533,
+      "step": 354
+    },
+    {
+      "epoch": 0.1136,
+      "grad_norm": 0.12438254803419113,
+      "learning_rate": 0.00019158762886597937,
+      "loss": 0.7511,
+      "step": 355
+    },
+    {
+      "epoch": 0.11392,
+      "grad_norm": 0.11343064159154892,
+      "learning_rate": 0.00019154639175257733,
+      "loss": 0.6876,
+      "step": 356
+    },
+    {
+      "epoch": 0.11424,
+      "grad_norm": 0.11751387268304825,
+      "learning_rate": 0.00019150515463917526,
+      "loss": 0.7198,
+      "step": 357
+    },
+    {
+      "epoch": 0.11456,
+      "grad_norm": 0.1194872185587883,
+      "learning_rate": 0.0001914639175257732,
+      "loss": 0.6149,
+      "step": 358
+    },
+    {
+      "epoch": 0.11488,
+      "grad_norm": 0.11200791597366333,
+      "learning_rate": 0.00019142268041237115,
+      "loss": 0.8605,
+      "step": 359
+    },
+    {
+      "epoch": 0.1152,
+      "grad_norm": 0.11708471924066544,
+      "learning_rate": 0.00019138144329896908,
+      "loss": 0.6603,
+      "step": 360
+    },
+    {
+      "epoch": 0.11552,
+      "grad_norm": 0.116248220205307,
+      "learning_rate": 0.00019134020618556704,
+      "loss": 0.7141,
+      "step": 361
+    },
+    {
+      "epoch": 0.11584,
+      "grad_norm": 0.11516683548688889,
+      "learning_rate": 0.00019129896907216494,
+      "loss": 0.7316,
+      "step": 362
+    },
+    {
+      "epoch": 0.11616,
+      "grad_norm": 0.10961434990167618,
+      "learning_rate": 0.00019125773195876288,
+      "loss": 0.6173,
+      "step": 363
+    },
+    {
+      "epoch": 0.11648,
+      "grad_norm": 0.12037073075771332,
+      "learning_rate": 0.00019121649484536083,
+      "loss": 0.6659,
+      "step": 364
+    },
+    {
+      "epoch": 0.1168,
+      "grad_norm": 0.12053418904542923,
+      "learning_rate": 0.00019117525773195877,
+      "loss": 0.6544,
+      "step": 365
+    },
+    {
+      "epoch": 0.11712,
+      "grad_norm": 0.12387484312057495,
+      "learning_rate": 0.00019113402061855672,
+      "loss": 0.6406,
+      "step": 366
+    },
+    {
+      "epoch": 0.11744,
+      "grad_norm": 0.12213553488254547,
+      "learning_rate": 0.00019109278350515466,
+      "loss": 0.6592,
+      "step": 367
+    },
+    {
+      "epoch": 0.11776,
+      "grad_norm": 0.1237708106637001,
+      "learning_rate": 0.0001910515463917526,
+      "loss": 0.779,
+      "step": 368
+    },
+    {
+      "epoch": 0.11808,
+      "grad_norm": 0.10254320502281189,
+      "learning_rate": 0.00019101030927835052,
+      "loss": 0.7242,
+      "step": 369
+    },
+    {
+      "epoch": 0.1184,
+      "grad_norm": 0.11179213225841522,
+      "learning_rate": 0.00019096907216494845,
+      "loss": 0.765,
+      "step": 370
+    },
+    {
+      "epoch": 0.11872,
+      "grad_norm": 0.11824121326208115,
+      "learning_rate": 0.0001909278350515464,
+      "loss": 0.6971,
+      "step": 371
+    },
+    {
+      "epoch": 0.11904,
+      "grad_norm": 0.13179510831832886,
+      "learning_rate": 0.00019088659793814434,
+      "loss": 0.7179,
+      "step": 372
+    },
+    {
+      "epoch": 0.11936,
+      "grad_norm": 0.1161201074719429,
+      "learning_rate": 0.00019084536082474227,
+      "loss": 0.6249,
+      "step": 373
+    },
+    {
+      "epoch": 0.11968,
+      "grad_norm": 0.12930230796337128,
+      "learning_rate": 0.00019080412371134023,
+      "loss": 0.6636,
+      "step": 374
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.12602809071540833,
+      "learning_rate": 0.00019076288659793816,
+      "loss": 0.7044,
+      "step": 375
+    },
+    {
+      "epoch": 0.12032,
+      "grad_norm": 0.1393459141254425,
+      "learning_rate": 0.0001907216494845361,
+      "loss": 0.6518,
+      "step": 376
+    },
+    {
+      "epoch": 0.12064,
+      "grad_norm": 0.11460117250680923,
+      "learning_rate": 0.00019068041237113402,
+      "loss": 0.7716,
+      "step": 377
+    },
+    {
+      "epoch": 0.12096,
+      "grad_norm": 0.14296174049377441,
+      "learning_rate": 0.00019063917525773195,
+      "loss": 0.7899,
+      "step": 378
+    },
+    {
+      "epoch": 0.12128,
+      "grad_norm": 0.14445006847381592,
+      "learning_rate": 0.0001905979381443299,
+      "loss": 0.7617,
+      "step": 379
+    },
+    {
+      "epoch": 0.1216,
+      "grad_norm": 0.12754163146018982,
+      "learning_rate": 0.00019055670103092784,
+      "loss": 0.7888,
+      "step": 380
+    },
+    {
+      "epoch": 0.12192,
+      "grad_norm": 0.11828102916479111,
+      "learning_rate": 0.00019051546391752577,
+      "loss": 0.6641,
+      "step": 381
+    },
+    {
+      "epoch": 0.12224,
+      "grad_norm": 0.11433565616607666,
+      "learning_rate": 0.00019047422680412373,
+      "loss": 0.5984,
+      "step": 382
+    },
+    {
+      "epoch": 0.12256,
+      "grad_norm": 0.11538711935281754,
+      "learning_rate": 0.00019043298969072166,
+      "loss": 0.7464,
+      "step": 383
+    },
+    {
+      "epoch": 0.12288,
+      "grad_norm": 0.10493431240320206,
+      "learning_rate": 0.0001903917525773196,
+      "loss": 0.7104,
+      "step": 384
+    },
+    {
+      "epoch": 0.1232,
+      "grad_norm": 0.13114306330680847,
+      "learning_rate": 0.00019035051546391753,
+      "loss": 0.9104,
+      "step": 385
+    },
+    {
+      "epoch": 0.12352,
+      "grad_norm": 0.11584331095218658,
+      "learning_rate": 0.00019030927835051546,
+      "loss": 0.7205,
+      "step": 386
+    },
+    {
+      "epoch": 0.12384,
+      "grad_norm": 0.1123526394367218,
+      "learning_rate": 0.00019026804123711342,
+      "loss": 0.7061,
+      "step": 387
+    },
+    {
+      "epoch": 0.12416,
+      "grad_norm": 0.11471172422170639,
+      "learning_rate": 0.00019022680412371135,
+      "loss": 0.7491,
+      "step": 388
+    },
+    {
+      "epoch": 0.12448,
+      "grad_norm": 0.1361629068851471,
+      "learning_rate": 0.0001901855670103093,
+      "loss": 0.8168,
+      "step": 389
+    },
+    {
+      "epoch": 0.1248,
+      "grad_norm": 0.1272808462381363,
+      "learning_rate": 0.00019014432989690724,
+      "loss": 0.6907,
+      "step": 390
+    },
+    {
+      "epoch": 0.12512,
+      "grad_norm": 0.12672847509384155,
+      "learning_rate": 0.00019010309278350514,
+      "loss": 0.7511,
+      "step": 391
+    },
+    {
+      "epoch": 0.12544,
+      "grad_norm": 0.12205267697572708,
+      "learning_rate": 0.0001900618556701031,
+      "loss": 0.8164,
+      "step": 392
+    },
+    {
+      "epoch": 0.12576,
+      "grad_norm": 0.12806954979896545,
+      "learning_rate": 0.00019002061855670103,
+      "loss": 0.8247,
+      "step": 393
+    },
+    {
+      "epoch": 0.12608,
+      "grad_norm": 0.11856722086668015,
+      "learning_rate": 0.000189979381443299,
+      "loss": 0.7324,
+      "step": 394
+    },
+    {
+      "epoch": 0.1264,
+      "grad_norm": 0.11742661893367767,
+      "learning_rate": 0.00018993814432989692,
+      "loss": 0.8151,
+      "step": 395
+    },
+    {
+      "epoch": 0.12672,
+      "grad_norm": 0.11036640405654907,
+      "learning_rate": 0.00018989690721649485,
+      "loss": 0.5336,
+      "step": 396
+    },
+    {
+      "epoch": 0.12704,
+      "grad_norm": 0.11546391993761063,
+      "learning_rate": 0.0001898556701030928,
+      "loss": 0.8078,
+      "step": 397
+    },
+    {
+      "epoch": 0.12736,
+      "grad_norm": 0.11293994635343552,
+      "learning_rate": 0.00018981443298969071,
+      "loss": 0.7818,
+      "step": 398
+    },
+    {
+      "epoch": 0.12768,
+      "grad_norm": 0.11465566605329514,
+      "learning_rate": 0.00018977319587628867,
+      "loss": 0.8291,
+      "step": 399
+    },
+    {
+      "epoch": 0.128,
+      "grad_norm": 0.1251228302717209,
+      "learning_rate": 0.0001897319587628866,
+      "loss": 0.7199,
+      "step": 400
+    },
+    {
+      "epoch": 0.12832,
+      "grad_norm": 0.11204738169908524,
+      "learning_rate": 0.00018969072164948454,
+      "loss": 0.7193,
+      "step": 401
+    },
+    {
+      "epoch": 0.12864,
+      "grad_norm": 0.1072632372379303,
+      "learning_rate": 0.0001896494845360825,
+      "loss": 0.7763,
+      "step": 402
+    },
+    {
+      "epoch": 0.12896,
+      "grad_norm": 0.11516649276018143,
+      "learning_rate": 0.00018960824742268043,
+      "loss": 0.6853,
+      "step": 403
+    },
+    {
+      "epoch": 0.12928,
+      "grad_norm": 0.1176360473036766,
+      "learning_rate": 0.00018956701030927836,
+      "loss": 0.6184,
+      "step": 404
+    },
+    {
+      "epoch": 0.1296,
+      "grad_norm": 0.11547468602657318,
+      "learning_rate": 0.0001895257731958763,
+      "loss": 0.8048,
+      "step": 405
+    },
+    {
+      "epoch": 0.12992,
+      "grad_norm": 0.11308492720127106,
+      "learning_rate": 0.00018948453608247422,
+      "loss": 0.7213,
+      "step": 406
+    },
+    {
+      "epoch": 0.13024,
+      "grad_norm": 0.11973299831151962,
+      "learning_rate": 0.00018944329896907218,
+      "loss": 0.8809,
+      "step": 407
+    },
+    {
+      "epoch": 0.13056,
+      "grad_norm": 0.1213008239865303,
+      "learning_rate": 0.0001894020618556701,
+      "loss": 0.7794,
+      "step": 408
+    },
+    {
+      "epoch": 0.13088,
+      "grad_norm": 0.13353341817855835,
+      "learning_rate": 0.00018936082474226804,
+      "loss": 0.6674,
+      "step": 409
+    },
+    {
+      "epoch": 0.1312,
+      "grad_norm": 0.1155238002538681,
+      "learning_rate": 0.000189319587628866,
+      "loss": 0.7668,
+      "step": 410
+    },
+    {
+      "epoch": 0.13152,
+      "grad_norm": 0.13428455591201782,
+      "learning_rate": 0.00018927835051546393,
+      "loss": 0.6864,
+      "step": 411
+    },
+    {
+      "epoch": 0.13184,
+      "grad_norm": 0.13777495920658112,
+      "learning_rate": 0.00018923711340206186,
+      "loss": 0.8271,
+      "step": 412
+    },
+    {
+      "epoch": 0.13216,
+      "grad_norm": 0.12846726179122925,
+      "learning_rate": 0.0001891958762886598,
+      "loss": 0.9027,
+      "step": 413
+    },
+    {
+      "epoch": 0.13248,
+      "grad_norm": 0.11394570022821426,
+      "learning_rate": 0.00018915463917525772,
+      "loss": 0.7531,
+      "step": 414
+    },
+    {
+      "epoch": 0.1328,
+      "grad_norm": 0.11797945201396942,
+      "learning_rate": 0.00018911340206185568,
+      "loss": 0.7393,
+      "step": 415
+    },
+    {
+      "epoch": 0.13312,
+      "grad_norm": 0.11778296530246735,
+      "learning_rate": 0.0001890721649484536,
+      "loss": 0.8172,
+      "step": 416
+    },
+    {
+      "epoch": 0.13344,
+      "grad_norm": 0.12857432663440704,
+      "learning_rate": 0.00018903092783505157,
+      "loss": 0.5693,
+      "step": 417
+    },
+    {
+      "epoch": 0.13376,
+      "grad_norm": 0.1395915150642395,
+      "learning_rate": 0.0001889896907216495,
+      "loss": 0.7187,
+      "step": 418
+    },
+    {
+      "epoch": 0.13408,
+      "grad_norm": 0.12571577727794647,
+      "learning_rate": 0.00018894845360824743,
+      "loss": 0.7347,
+      "step": 419
+    },
+    {
+      "epoch": 0.1344,
+      "grad_norm": 0.11062180995941162,
+      "learning_rate": 0.00018890721649484537,
+      "loss": 0.8797,
+      "step": 420
+    },
+    {
+      "epoch": 0.13472,
+      "grad_norm": 0.12011471390724182,
+      "learning_rate": 0.0001888659793814433,
+      "loss": 0.7638,
+      "step": 421
+    },
+    {
+      "epoch": 0.13504,
+      "grad_norm": 0.11402884870767593,
+      "learning_rate": 0.00018882474226804126,
+      "loss": 0.6198,
+      "step": 422
+    },
+    {
+      "epoch": 0.13536,
+      "grad_norm": 0.11654789745807648,
+      "learning_rate": 0.00018878350515463919,
+      "loss": 0.8276,
+      "step": 423
+    },
+    {
+      "epoch": 0.13568,
+      "grad_norm": 0.110030896961689,
+      "learning_rate": 0.00018874226804123712,
+      "loss": 0.7828,
+      "step": 424
+    },
+    {
+      "epoch": 0.136,
+      "grad_norm": 0.1173168271780014,
+      "learning_rate": 0.00018870103092783508,
+      "loss": 0.6445,
+      "step": 425
+    },
+    {
+      "epoch": 0.13632,
+      "grad_norm": 0.11195468157529831,
+      "learning_rate": 0.000188659793814433,
+      "loss": 0.7336,
+      "step": 426
+    },
+    {
+      "epoch": 0.13664,
+      "grad_norm": 0.1138012707233429,
+      "learning_rate": 0.00018861855670103094,
+      "loss": 0.7085,
+      "step": 427
+    },
+    {
+      "epoch": 0.13696,
+      "grad_norm": 0.11757896840572357,
+      "learning_rate": 0.00018857731958762887,
+      "loss": 0.593,
+      "step": 428
+    },
+    {
+      "epoch": 0.13728,
+      "grad_norm": 0.13230633735656738,
+      "learning_rate": 0.0001885360824742268,
+      "loss": 0.705,
+      "step": 429
+    },
+    {
+      "epoch": 0.1376,
+      "grad_norm": 0.13503113389015198,
+      "learning_rate": 0.00018849484536082476,
+      "loss": 0.7134,
+      "step": 430
+    },
+    {
+      "epoch": 0.13792,
+      "grad_norm": 0.11426805704832077,
+      "learning_rate": 0.0001884536082474227,
+      "loss": 0.6781,
+      "step": 431
+    },
+    {
+      "epoch": 0.13824,
+      "grad_norm": 0.11934298276901245,
+      "learning_rate": 0.00018841237113402065,
+      "loss": 0.7727,
+      "step": 432
+    },
+    {
+      "epoch": 0.13856,
+      "grad_norm": 0.12700718641281128,
+      "learning_rate": 0.00018837113402061858,
+      "loss": 0.7153,
+      "step": 433
+    },
+    {
+      "epoch": 0.13888,
+      "grad_norm": 0.15089921653270721,
+      "learning_rate": 0.00018832989690721648,
+      "loss": 0.7326,
+      "step": 434
+    },
+    {
+      "epoch": 0.1392,
+      "grad_norm": 0.10799750685691833,
+      "learning_rate": 0.00018828865979381444,
+      "loss": 0.7738,
+      "step": 435
+    },
+    {
+      "epoch": 0.13952,
+      "grad_norm": 0.1076594665646553,
+      "learning_rate": 0.00018824742268041237,
+      "loss": 0.545,
+      "step": 436
+    },
+    {
+      "epoch": 0.13984,
+      "grad_norm": 0.1204753890633583,
+      "learning_rate": 0.00018820618556701033,
+      "loss": 0.73,
+      "step": 437
+    },
+    {
+      "epoch": 0.14016,
+      "grad_norm": 0.11598300188779831,
+      "learning_rate": 0.00018816494845360826,
+      "loss": 0.635,
+      "step": 438
+    },
+    {
+      "epoch": 0.14048,
+      "grad_norm": 0.11010400205850601,
+      "learning_rate": 0.0001881237113402062,
+      "loss": 0.7891,
+      "step": 439
+    },
+    {
+      "epoch": 0.1408,
+      "grad_norm": 0.12184228003025055,
+      "learning_rate": 0.00018808247422680413,
+      "loss": 0.7332,
+      "step": 440
+    },
+    {
+      "epoch": 0.14112,
+      "grad_norm": 0.11696261167526245,
+      "learning_rate": 0.00018804123711340206,
+      "loss": 0.8299,
+      "step": 441
+    },
+    {
+      "epoch": 0.14144,
+      "grad_norm": 0.11464802920818329,
+      "learning_rate": 0.000188,
+      "loss": 0.7107,
+      "step": 442
+    },
+    {
+      "epoch": 0.14176,
+      "grad_norm": 0.12120334804058075,
+      "learning_rate": 0.00018795876288659795,
+      "loss": 0.7729,
+      "step": 443
+    },
+    {
+      "epoch": 0.14208,
+      "grad_norm": 0.11571822315454483,
+      "learning_rate": 0.00018791752577319588,
+      "loss": 0.6551,
+      "step": 444
+    },
+    {
+      "epoch": 0.1424,
+      "grad_norm": 0.11872228980064392,
+      "learning_rate": 0.00018787628865979384,
+      "loss": 0.698,
+      "step": 445
+    },
+    {
+      "epoch": 0.14272,
+      "grad_norm": 0.12505149841308594,
+      "learning_rate": 0.00018783505154639177,
+      "loss": 0.751,
+      "step": 446
+    },
+    {
+      "epoch": 0.14304,
+      "grad_norm": 0.10618813335895538,
+      "learning_rate": 0.0001877938144329897,
+      "loss": 0.7299,
+      "step": 447
+    },
+    {
+      "epoch": 0.14336,
+      "grad_norm": 0.11824540048837662,
+      "learning_rate": 0.00018775257731958763,
+      "loss": 0.6796,
+      "step": 448
+    },
+    {
+      "epoch": 0.14368,
+      "grad_norm": 0.10695328563451767,
+      "learning_rate": 0.00018771134020618556,
+      "loss": 0.7424,
+      "step": 449
+    },
+    {
+      "epoch": 0.144,
+      "grad_norm": 0.12764234840869904,
+      "learning_rate": 0.00018767010309278352,
+      "loss": 0.7301,
+      "step": 450
+    },
+    {
+      "epoch": 0.14432,
+      "grad_norm": 0.12354765832424164,
+      "learning_rate": 0.00018762886597938145,
+      "loss": 0.7945,
+      "step": 451
+    },
+    {
+      "epoch": 0.14464,
+      "grad_norm": 0.12733551859855652,
+      "learning_rate": 0.00018758762886597938,
+      "loss": 0.7565,
+      "step": 452
+    },
+    {
+      "epoch": 0.14496,
+      "grad_norm": 0.1533960998058319,
+      "learning_rate": 0.00018754639175257734,
+      "loss": 0.7964,
+      "step": 453
+    },
+    {
+      "epoch": 0.14528,
+      "grad_norm": 0.11748246848583221,
+      "learning_rate": 0.00018750515463917527,
+      "loss": 0.7715,
+      "step": 454
+    },
+    {
+      "epoch": 0.1456,
+      "grad_norm": 0.12309489399194717,
+      "learning_rate": 0.0001874639175257732,
+      "loss": 0.7758,
+      "step": 455
+    },
+    {
+      "epoch": 0.14592,
+      "grad_norm": 0.10743677616119385,
+      "learning_rate": 0.00018742268041237114,
+      "loss": 0.8084,
+      "step": 456
+    },
+    {
+      "epoch": 0.14624,
+      "grad_norm": 0.15399467945098877,
+      "learning_rate": 0.00018738144329896907,
+      "loss": 0.7689,
+      "step": 457
+    },
+    {
+      "epoch": 0.14656,
+      "grad_norm": 0.1381152868270874,
+      "learning_rate": 0.00018734020618556702,
+      "loss": 0.7902,
+      "step": 458
+    },
+    {
+      "epoch": 0.14688,
+      "grad_norm": 0.12086453288793564,
+      "learning_rate": 0.00018729896907216496,
+      "loss": 0.739,
+      "step": 459
+    },
+    {
+      "epoch": 0.1472,
+      "grad_norm": 0.14193426072597504,
+      "learning_rate": 0.00018725773195876291,
+      "loss": 0.8245,
+      "step": 460
+    },
+    {
+      "epoch": 0.14752,
+      "grad_norm": 0.1121336966753006,
+      "learning_rate": 0.00018721649484536085,
+      "loss": 0.6877,
+      "step": 461
+    },
+    {
+      "epoch": 0.14784,
+      "grad_norm": 0.12167156487703323,
+      "learning_rate": 0.00018717525773195878,
+      "loss": 0.7017,
+      "step": 462
+    },
+    {
+      "epoch": 0.14816,
+      "grad_norm": 0.1287817507982254,
+      "learning_rate": 0.0001871340206185567,
+      "loss": 0.6229,
+      "step": 463
+    },
+    {
+      "epoch": 0.14848,
+      "grad_norm": 0.128828227519989,
+      "learning_rate": 0.00018709278350515464,
+      "loss": 0.7917,
+      "step": 464
+    },
+    {
+      "epoch": 0.1488,
+      "grad_norm": 0.13771773874759674,
+      "learning_rate": 0.0001870515463917526,
+      "loss": 0.6883,
+      "step": 465
+    },
+    {
+      "epoch": 0.14912,
+      "grad_norm": 0.10570216923952103,
+      "learning_rate": 0.00018701030927835053,
+      "loss": 0.7791,
+      "step": 466
+    },
+    {
+      "epoch": 0.14944,
+      "grad_norm": 0.1250445395708084,
+      "learning_rate": 0.00018696907216494846,
+      "loss": 0.7435,
+      "step": 467
+    },
+    {
+      "epoch": 0.14976,
+      "grad_norm": 0.1064952090382576,
+      "learning_rate": 0.00018692783505154642,
+      "loss": 0.7621,
+      "step": 468
+    },
+    {
+      "epoch": 0.15008,
+      "grad_norm": 0.10514791309833527,
+      "learning_rate": 0.00018688659793814432,
+      "loss": 0.6403,
+      "step": 469
+    },
+    {
+      "epoch": 0.1504,
+      "grad_norm": 0.10278081893920898,
+      "learning_rate": 0.00018684536082474225,
+      "loss": 0.8246,
+      "step": 470
+    },
+    {
+      "epoch": 0.15072,
+      "grad_norm": 0.11813073605298996,
+      "learning_rate": 0.0001868041237113402,
+      "loss": 0.8403,
+      "step": 471
+    },
+    {
+      "epoch": 0.15104,
+      "grad_norm": 0.10654576867818832,
+      "learning_rate": 0.00018676288659793814,
+      "loss": 0.7483,
+      "step": 472
+    },
+    {
+      "epoch": 0.15136,
+      "grad_norm": 0.11642228811979294,
+      "learning_rate": 0.0001867216494845361,
+      "loss": 0.6114,
+      "step": 473
+    },
+    {
+      "epoch": 0.15168,
+      "grad_norm": 0.12933100759983063,
+      "learning_rate": 0.00018668041237113403,
+      "loss": 0.7902,
+      "step": 474
+    },
+    {
+      "epoch": 0.152,
+      "grad_norm": 0.1200195848941803,
+      "learning_rate": 0.00018663917525773196,
+      "loss": 0.7298,
+      "step": 475
+    },
+    {
+      "epoch": 0.15232,
+      "grad_norm": 0.10747142881155014,
+      "learning_rate": 0.0001865979381443299,
+      "loss": 0.7376,
+      "step": 476
+    },
+    {
+      "epoch": 0.15264,
+      "grad_norm": 0.10868332535028458,
+      "learning_rate": 0.00018655670103092783,
+      "loss": 0.7013,
+      "step": 477
+    },
+    {
+      "epoch": 0.15296,
+      "grad_norm": 0.12750379741191864,
+      "learning_rate": 0.00018651546391752579,
+      "loss": 0.7718,
+      "step": 478
+    },
+    {
+      "epoch": 0.15328,
+      "grad_norm": 0.10860800743103027,
+      "learning_rate": 0.00018647422680412372,
+      "loss": 0.5887,
+      "step": 479
+    },
+    {
+      "epoch": 0.1536,
+      "grad_norm": 0.11445030570030212,
+      "learning_rate": 0.00018643298969072165,
+      "loss": 0.7955,
+      "step": 480
+    },
+    {
+      "epoch": 0.15392,
+      "grad_norm": 0.11774427443742752,
+      "learning_rate": 0.0001863917525773196,
+      "loss": 0.7095,
+      "step": 481
+    },
+    {
+      "epoch": 0.15424,
+      "grad_norm": 0.13853546977043152,
+      "learning_rate": 0.00018635051546391754,
+      "loss": 0.9395,
+      "step": 482
+    },
+    {
+      "epoch": 0.15456,
+      "grad_norm": 0.120772585272789,
+      "learning_rate": 0.00018630927835051547,
+      "loss": 0.8,
+      "step": 483
+    },
+    {
+      "epoch": 0.15488,
+      "grad_norm": 0.13263949751853943,
+      "learning_rate": 0.0001862680412371134,
+      "loss": 0.9118,
+      "step": 484
+    },
+    {
+      "epoch": 0.1552,
+      "grad_norm": 0.12179119139909744,
+      "learning_rate": 0.00018622680412371133,
+      "loss": 0.6567,
+      "step": 485
+    },
+    {
+      "epoch": 0.15552,
+      "grad_norm": 0.1244128867983818,
+      "learning_rate": 0.0001861855670103093,
+      "loss": 0.7702,
+      "step": 486
+    },
+    {
+      "epoch": 0.15584,
+      "grad_norm": 0.12456309050321579,
+      "learning_rate": 0.00018614432989690722,
+      "loss": 0.7687,
+      "step": 487
+    },
+    {
+      "epoch": 0.15616,
+      "grad_norm": 0.12738095223903656,
+      "learning_rate": 0.00018610309278350518,
+      "loss": 0.7489,
+      "step": 488
+    },
+    {
+      "epoch": 0.15648,
+      "grad_norm": 0.11148428916931152,
+      "learning_rate": 0.0001860618556701031,
+      "loss": 0.7362,
+      "step": 489
+    },
+    {
+      "epoch": 0.1568,
+      "grad_norm": 0.1029975637793541,
+      "learning_rate": 0.00018602061855670104,
+      "loss": 0.5808,
+      "step": 490
+    },
+    {
+      "epoch": 0.15712,
+      "grad_norm": 0.12028798460960388,
+      "learning_rate": 0.00018597938144329897,
+      "loss": 0.8541,
+      "step": 491
+    },
+    {
+      "epoch": 0.15744,
+      "grad_norm": 0.12013960629701614,
+      "learning_rate": 0.0001859381443298969,
+      "loss": 0.6814,
+      "step": 492
+    },
+    {
+      "epoch": 0.15776,
+      "grad_norm": 0.12026184052228928,
+      "learning_rate": 0.00018589690721649486,
+      "loss": 0.8444,
+      "step": 493
+    },
+    {
+      "epoch": 0.15808,
+      "grad_norm": 0.11788841336965561,
+      "learning_rate": 0.0001858556701030928,
+      "loss": 0.5718,
+      "step": 494
+    },
+    {
+      "epoch": 0.1584,
+      "grad_norm": 0.11122492700815201,
+      "learning_rate": 0.00018581443298969073,
+      "loss": 0.5172,
+      "step": 495
+    },
+    {
+      "epoch": 0.15872,
+      "grad_norm": 0.12228988111019135,
+      "learning_rate": 0.00018577319587628868,
+      "loss": 0.724,
+      "step": 496
+    },
+    {
+      "epoch": 0.15904,
+      "grad_norm": 0.12170293927192688,
+      "learning_rate": 0.00018573195876288662,
+      "loss": 0.6863,
+      "step": 497
+    },
+    {
+      "epoch": 0.15936,
+      "grad_norm": 0.12947559356689453,
+      "learning_rate": 0.00018569072164948455,
+      "loss": 0.897,
+      "step": 498
+    },
+    {
+      "epoch": 0.15968,
+      "grad_norm": 0.13217338919639587,
+      "learning_rate": 0.00018564948453608248,
+      "loss": 0.7,
+      "step": 499
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.12307247519493103,
+      "learning_rate": 0.0001856082474226804,
+      "loss": 0.6686,
+      "step": 500
+    },
+    {
+      "epoch": 0.16032,
+      "grad_norm": 0.12643639743328094,
+      "learning_rate": 0.00018556701030927837,
+      "loss": 0.6166,
+      "step": 501
+    },
+    {
+      "epoch": 0.16064,
+      "grad_norm": 0.10645804554224014,
+      "learning_rate": 0.0001855257731958763,
+      "loss": 0.785,
+      "step": 502
+    },
+    {
+      "epoch": 0.16096,
+      "grad_norm": 0.12195251882076263,
+      "learning_rate": 0.00018548453608247423,
+      "loss": 0.6931,
+      "step": 503
+    },
+    {
+      "epoch": 0.16128,
+      "grad_norm": 0.10586483776569366,
+      "learning_rate": 0.0001854432989690722,
+      "loss": 0.6172,
+      "step": 504
+    },
+    {
+      "epoch": 0.1616,
+      "grad_norm": 0.12337654829025269,
+      "learning_rate": 0.0001854020618556701,
+      "loss": 0.7811,
+      "step": 505
+    },
+    {
+      "epoch": 0.16192,
+      "grad_norm": 0.11356624215841293,
+      "learning_rate": 0.00018536082474226805,
+      "loss": 0.7117,
+      "step": 506
+    },
+    {
+      "epoch": 0.16224,
+      "grad_norm": 0.11793960630893707,
+      "learning_rate": 0.00018531958762886598,
+      "loss": 0.7778,
+      "step": 507
+    },
+    {
+      "epoch": 0.16256,
+      "grad_norm": 0.12677061557769775,
+      "learning_rate": 0.00018527835051546391,
+      "loss": 0.6465,
+      "step": 508
+    },
+    {
+      "epoch": 0.16288,
+      "grad_norm": 0.12064032256603241,
+      "learning_rate": 0.00018523711340206187,
+      "loss": 0.7658,
+      "step": 509
+    },
+    {
+      "epoch": 0.1632,
+      "grad_norm": 0.12942875921726227,
+      "learning_rate": 0.0001851958762886598,
+      "loss": 0.7735,
+      "step": 510
+    },
+    {
+      "epoch": 0.16352,
+      "grad_norm": 0.1350182294845581,
+      "learning_rate": 0.00018515463917525776,
+      "loss": 0.7512,
+      "step": 511
+    },
+    {
+      "epoch": 0.16384,
+      "grad_norm": 0.12585826218128204,
+      "learning_rate": 0.00018511340206185567,
+      "loss": 0.5842,
+      "step": 512
+    },
+    {
+      "epoch": 0.16416,
+      "grad_norm": 0.1233866736292839,
+      "learning_rate": 0.0001850721649484536,
+      "loss": 0.6707,
+      "step": 513
+    },
+    {
+      "epoch": 0.16448,
+      "grad_norm": 0.13590243458747864,
+      "learning_rate": 0.00018503092783505156,
+      "loss": 0.645,
+      "step": 514
+    },
+    {
+      "epoch": 0.1648,
+      "grad_norm": 0.11508861184120178,
+      "learning_rate": 0.0001849896907216495,
+      "loss": 0.698,
+      "step": 515
+    },
+    {
+      "epoch": 0.16512,
+      "grad_norm": 0.11506541818380356,
+      "learning_rate": 0.00018494845360824745,
+      "loss": 0.6834,
+      "step": 516
+    },
+    {
+      "epoch": 0.16544,
+      "grad_norm": 0.1177835762500763,
+      "learning_rate": 0.00018490721649484538,
+      "loss": 0.8327,
+      "step": 517
+    },
+    {
+      "epoch": 0.16576,
+      "grad_norm": 0.12268577516078949,
+      "learning_rate": 0.0001848659793814433,
+      "loss": 0.7925,
+      "step": 518
+    },
+    {
+      "epoch": 0.16608,
+      "grad_norm": 0.11970438808202744,
+      "learning_rate": 0.00018482474226804124,
+      "loss": 0.9855,
+      "step": 519
+    },
+    {
+      "epoch": 0.1664,
+      "grad_norm": 0.11444945633411407,
+      "learning_rate": 0.00018478350515463917,
+      "loss": 0.6759,
+      "step": 520
+    },
+    {
+      "epoch": 0.16672,
+      "grad_norm": 0.1250583827495575,
+      "learning_rate": 0.00018474226804123713,
+      "loss": 0.7223,
+      "step": 521
+    },
+    {
+      "epoch": 0.16704,
+      "grad_norm": 0.128667950630188,
+      "learning_rate": 0.00018470103092783506,
+      "loss": 0.7038,
+      "step": 522
+    },
+    {
+      "epoch": 0.16736,
+      "grad_norm": 0.1111631840467453,
+      "learning_rate": 0.000184659793814433,
+      "loss": 0.8037,
+      "step": 523
+    },
+    {
+      "epoch": 0.16768,
+      "grad_norm": 0.10735559463500977,
+      "learning_rate": 0.00018461855670103095,
+      "loss": 0.7687,
+      "step": 524
+    },
+    {
+      "epoch": 0.168,
+      "grad_norm": 0.12646177411079407,
+      "learning_rate": 0.00018457731958762888,
+      "loss": 0.7467,
+      "step": 525
+    },
+    {
+      "epoch": 0.16832,
+      "grad_norm": 0.11589492112398148,
+      "learning_rate": 0.0001845360824742268,
+      "loss": 0.5764,
+      "step": 526
+    },
+    {
+      "epoch": 0.16864,
+      "grad_norm": 0.11384910345077515,
+      "learning_rate": 0.00018449484536082474,
+      "loss": 0.7205,
+      "step": 527
+    },
+    {
+      "epoch": 0.16896,
+      "grad_norm": 0.11103667318820953,
+      "learning_rate": 0.00018445360824742267,
+      "loss": 0.7596,
+      "step": 528
+    },
+    {
+      "epoch": 0.16928,
+      "grad_norm": 0.11777044087648392,
+      "learning_rate": 0.00018441237113402063,
+      "loss": 0.8103,
+      "step": 529
+    },
+    {
+      "epoch": 0.1696,
+      "grad_norm": 0.13625359535217285,
+      "learning_rate": 0.00018437113402061856,
+      "loss": 0.8057,
+      "step": 530
+    },
+    {
+      "epoch": 0.16992,
+      "grad_norm": 0.12384586781263351,
+      "learning_rate": 0.0001843298969072165,
+      "loss": 0.6888,
+      "step": 531
+    },
+    {
+      "epoch": 0.17024,
+      "grad_norm": 0.12003818899393082,
+      "learning_rate": 0.00018428865979381445,
+      "loss": 0.6938,
+      "step": 532
+    },
+    {
+      "epoch": 0.17056,
+      "grad_norm": 0.11006835103034973,
+      "learning_rate": 0.00018424742268041239,
+      "loss": 0.8389,
+      "step": 533
+    },
+    {
+      "epoch": 0.17088,
+      "grad_norm": 0.1308056265115738,
+      "learning_rate": 0.00018420618556701032,
+      "loss": 0.6838,
+      "step": 534
+    },
+    {
+      "epoch": 0.1712,
+      "grad_norm": 0.13694193959236145,
+      "learning_rate": 0.00018416494845360825,
+      "loss": 0.7372,
+      "step": 535
+    },
+    {
+      "epoch": 0.17152,
+      "grad_norm": 0.12474847584962845,
+      "learning_rate": 0.00018412371134020618,
+      "loss": 0.7316,
+      "step": 536
+    },
+    {
+      "epoch": 0.17184,
+      "grad_norm": 0.12851357460021973,
+      "learning_rate": 0.00018408247422680414,
+      "loss": 0.8742,
+      "step": 537
+    },
+    {
+      "epoch": 0.17216,
+      "grad_norm": 0.12235444784164429,
+      "learning_rate": 0.00018404123711340207,
+      "loss": 0.798,
+      "step": 538
+    },
+    {
+      "epoch": 0.17248,
+      "grad_norm": 0.13520556688308716,
+      "learning_rate": 0.00018400000000000003,
+      "loss": 0.8435,
+      "step": 539
+    },
+    {
+      "epoch": 0.1728,
+      "grad_norm": 0.10702179372310638,
+      "learning_rate": 0.00018395876288659796,
+      "loss": 0.5605,
+      "step": 540
+    },
+    {
+      "epoch": 0.17312,
+      "grad_norm": 0.13347023725509644,
+      "learning_rate": 0.00018391752577319586,
+      "loss": 0.7539,
+      "step": 541
+    },
+    {
+      "epoch": 0.17344,
+      "grad_norm": 0.12964722514152527,
+      "learning_rate": 0.00018387628865979382,
+      "loss": 0.7302,
+      "step": 542
+    },
+    {
+      "epoch": 0.17376,
+      "grad_norm": 0.1171535775065422,
+      "learning_rate": 0.00018383505154639175,
+      "loss": 0.6735,
+      "step": 543
+    },
+    {
+      "epoch": 0.17408,
+      "grad_norm": 0.11912510544061661,
+      "learning_rate": 0.0001837938144329897,
+      "loss": 0.7924,
+      "step": 544
+    },
+    {
+      "epoch": 0.1744,
+      "grad_norm": 0.13499023020267487,
+      "learning_rate": 0.00018375257731958764,
+      "loss": 0.6884,
+      "step": 545
+    },
+    {
+      "epoch": 0.17472,
+      "grad_norm": 0.11606519669294357,
+      "learning_rate": 0.00018371134020618557,
+      "loss": 0.6533,
+      "step": 546
+    },
+    {
+      "epoch": 0.17504,
+      "grad_norm": 0.11444912850856781,
+      "learning_rate": 0.00018367010309278353,
+      "loss": 0.6173,
+      "step": 547
+    },
+    {
+      "epoch": 0.17536,
+      "grad_norm": 0.12102066725492477,
+      "learning_rate": 0.00018362886597938144,
+      "loss": 0.7732,
+      "step": 548
+    },
+    {
+      "epoch": 0.17568,
+      "grad_norm": 0.12623833119869232,
+      "learning_rate": 0.0001835876288659794,
+      "loss": 0.8299,
+      "step": 549
+    },
+    {
+      "epoch": 0.176,
+      "grad_norm": 0.13104446232318878,
+      "learning_rate": 0.00018354639175257733,
+      "loss": 0.6679,
+      "step": 550
+    },
+    {
+      "epoch": 0.17632,
+      "grad_norm": 0.12092466652393341,
+      "learning_rate": 0.00018350515463917526,
+      "loss": 0.7737,
+      "step": 551
+    },
+    {
+      "epoch": 0.17664,
+      "grad_norm": 0.11604610830545425,
+      "learning_rate": 0.00018346391752577322,
+      "loss": 0.8411,
+      "step": 552
+    },
+    {
+      "epoch": 0.17696,
+      "grad_norm": 0.11258059740066528,
+      "learning_rate": 0.00018342268041237115,
+      "loss": 0.7016,
+      "step": 553
+    },
+    {
+      "epoch": 0.17728,
+      "grad_norm": 0.11493982374668121,
+      "learning_rate": 0.00018338144329896908,
+      "loss": 0.7129,
+      "step": 554
+    },
+    {
+      "epoch": 0.1776,
+      "grad_norm": 0.13074208796024323,
+      "learning_rate": 0.000183340206185567,
+      "loss": 0.7522,
+      "step": 555
+    },
+    {
+      "epoch": 0.17792,
+      "grad_norm": 0.1120564267039299,
+      "learning_rate": 0.00018329896907216494,
+      "loss": 0.6846,
+      "step": 556
+    },
+    {
+      "epoch": 0.17824,
+      "grad_norm": 0.1323213428258896,
+      "learning_rate": 0.0001832577319587629,
+      "loss": 0.8367,
+      "step": 557
+    },
+    {
+      "epoch": 0.17856,
+      "grad_norm": 0.14067520201206207,
+      "learning_rate": 0.00018321649484536083,
+      "loss": 0.9111,
+      "step": 558
+    },
+    {
+      "epoch": 0.17888,
+      "grad_norm": 0.12386526167392731,
+      "learning_rate": 0.0001831752577319588,
+      "loss": 0.7513,
+      "step": 559
+    },
+    {
+      "epoch": 0.1792,
+      "grad_norm": 0.11267178505659103,
+      "learning_rate": 0.00018313402061855672,
+      "loss": 0.7561,
+      "step": 560
+    },
+    {
+      "epoch": 0.17952,
+      "grad_norm": 0.12407040596008301,
+      "learning_rate": 0.00018309278350515465,
+      "loss": 0.9464,
+      "step": 561
+    },
+    {
+      "epoch": 0.17984,
+      "grad_norm": 0.12141172587871552,
+      "learning_rate": 0.00018305154639175258,
+      "loss": 0.8564,
+      "step": 562
+    },
+    {
+      "epoch": 0.18016,
+      "grad_norm": 0.11525171250104904,
+      "learning_rate": 0.0001830103092783505,
+      "loss": 0.7085,
+      "step": 563
+    },
+    {
+      "epoch": 0.18048,
+      "grad_norm": 0.1141604334115982,
+      "learning_rate": 0.00018296907216494844,
+      "loss": 0.89,
+      "step": 564
+    },
+    {
+      "epoch": 0.1808,
+      "grad_norm": 0.117414191365242,
+      "learning_rate": 0.0001829278350515464,
+      "loss": 0.6754,
+      "step": 565
+    },
+    {
+      "epoch": 0.18112,
+      "grad_norm": 0.13193415105342865,
+      "learning_rate": 0.00018288659793814433,
+      "loss": 0.8465,
+      "step": 566
+    },
+    {
+      "epoch": 0.18144,
+      "grad_norm": 0.12107145041227341,
+      "learning_rate": 0.0001828453608247423,
+      "loss": 0.7464,
+      "step": 567
+    },
+    {
+      "epoch": 0.18176,
+      "grad_norm": 0.11851692944765091,
+      "learning_rate": 0.00018280412371134022,
+      "loss": 0.6608,
+      "step": 568
+    },
+    {
+      "epoch": 0.18208,
+      "grad_norm": 0.11890149861574173,
+      "learning_rate": 0.00018276288659793816,
+      "loss": 0.9179,
+      "step": 569
+    },
+    {
+      "epoch": 0.1824,
+      "grad_norm": 0.11981701850891113,
+      "learning_rate": 0.0001827216494845361,
+      "loss": 0.7348,
+      "step": 570
+    },
+    {
+      "epoch": 0.18272,
+      "grad_norm": 0.13188837468624115,
+      "learning_rate": 0.00018268041237113402,
+      "loss": 0.8143,
+      "step": 571
+    },
+    {
+      "epoch": 0.18304,
+      "grad_norm": 0.12455637753009796,
+      "learning_rate": 0.00018263917525773198,
+      "loss": 0.5797,
+      "step": 572
+    },
+    {
+      "epoch": 0.18336,
+      "grad_norm": 0.12442684173583984,
+      "learning_rate": 0.0001825979381443299,
+      "loss": 0.8271,
+      "step": 573
+    },
+    {
+      "epoch": 0.18368,
+      "grad_norm": 0.12401217222213745,
+      "learning_rate": 0.00018255670103092784,
+      "loss": 0.6963,
+      "step": 574
+    },
+    {
+      "epoch": 0.184,
+      "grad_norm": 0.12561100721359253,
+      "learning_rate": 0.0001825154639175258,
+      "loss": 0.7436,
+      "step": 575
+    },
+    {
+      "epoch": 0.18432,
+      "grad_norm": 0.1348034292459488,
+      "learning_rate": 0.00018247422680412373,
+      "loss": 0.677,
+      "step": 576
+    },
+    {
+      "epoch": 0.18464,
+      "grad_norm": 0.12489503622055054,
+      "learning_rate": 0.00018243298969072166,
+      "loss": 0.7475,
+      "step": 577
+    },
+    {
+      "epoch": 0.18496,
+      "grad_norm": 0.11799902468919754,
+      "learning_rate": 0.0001823917525773196,
+      "loss": 0.6301,
+      "step": 578
+    },
+    {
+      "epoch": 0.18528,
+      "grad_norm": 0.12443645298480988,
+      "learning_rate": 0.00018235051546391752,
+      "loss": 0.8342,
+      "step": 579
+    },
+    {
+      "epoch": 0.1856,
+      "grad_norm": 0.1418801099061966,
+      "learning_rate": 0.00018230927835051548,
+      "loss": 0.7459,
+      "step": 580
+    },
+    {
+      "epoch": 0.18592,
+      "grad_norm": 0.13329613208770752,
+      "learning_rate": 0.0001822680412371134,
+      "loss": 0.8183,
+      "step": 581
+    },
+    {
+      "epoch": 0.18624,
+      "grad_norm": 0.11498089879751205,
+      "learning_rate": 0.00018222680412371137,
+      "loss": 0.586,
+      "step": 582
+    },
+    {
+      "epoch": 0.18656,
+      "grad_norm": 0.11994834244251251,
+      "learning_rate": 0.0001821855670103093,
+      "loss": 0.6292,
+      "step": 583
+    },
+    {
+      "epoch": 0.18688,
+      "grad_norm": 0.11894108355045319,
+      "learning_rate": 0.0001821443298969072,
+      "loss": 0.79,
+      "step": 584
+    },
+    {
+      "epoch": 0.1872,
+      "grad_norm": 0.13187873363494873,
+      "learning_rate": 0.00018210309278350516,
+      "loss": 0.6799,
+      "step": 585
+    },
+    {
+      "epoch": 0.18752,
+      "grad_norm": 0.12553033232688904,
+      "learning_rate": 0.0001820618556701031,
+      "loss": 0.7286,
+      "step": 586
+    },
+    {
+      "epoch": 0.18784,
+      "grad_norm": 0.1164742261171341,
+      "learning_rate": 0.00018202061855670105,
+      "loss": 0.6721,
+      "step": 587
+    },
+    {
+      "epoch": 0.18816,
+      "grad_norm": 0.10284867882728577,
+      "learning_rate": 0.00018197938144329898,
+      "loss": 0.5752,
+      "step": 588
+    },
+    {
+      "epoch": 0.18848,
+      "grad_norm": 0.14761145412921906,
+      "learning_rate": 0.00018193814432989692,
+      "loss": 0.8899,
+      "step": 589
+    },
+    {
+      "epoch": 0.1888,
+      "grad_norm": 0.11473073065280914,
+      "learning_rate": 0.00018189690721649485,
+      "loss": 0.856,
+      "step": 590
+    },
+    {
+      "epoch": 0.18912,
+      "grad_norm": 0.12335352599620819,
+      "learning_rate": 0.00018185567010309278,
+      "loss": 0.7842,
+      "step": 591
+    },
+    {
+      "epoch": 0.18944,
+      "grad_norm": 0.10649198293685913,
+      "learning_rate": 0.0001818144329896907,
+      "loss": 0.7395,
+      "step": 592
+    },
+    {
+      "epoch": 0.18976,
+      "grad_norm": 0.11945925652980804,
+      "learning_rate": 0.00018177319587628867,
+      "loss": 0.6553,
+      "step": 593
+    },
+    {
+      "epoch": 0.19008,
+      "grad_norm": 0.1164558157324791,
+      "learning_rate": 0.0001817319587628866,
+      "loss": 0.6454,
+      "step": 594
+    },
+    {
+      "epoch": 0.1904,
+      "grad_norm": 0.12806792557239532,
+      "learning_rate": 0.00018169072164948456,
+      "loss": 0.7041,
+      "step": 595
+    },
+    {
+      "epoch": 0.19072,
+      "grad_norm": 0.12284845858812332,
+      "learning_rate": 0.0001816494845360825,
+      "loss": 0.8651,
+      "step": 596
+    },
+    {
+      "epoch": 0.19104,
+      "grad_norm": 0.12949353456497192,
+      "learning_rate": 0.00018160824742268042,
+      "loss": 0.761,
+      "step": 597
+    },
+    {
+      "epoch": 0.19136,
+      "grad_norm": 0.11860032379627228,
+      "learning_rate": 0.00018156701030927835,
+      "loss": 0.7484,
+      "step": 598
+    },
+    {
+      "epoch": 0.19168,
+      "grad_norm": 0.12866483628749847,
+      "learning_rate": 0.00018152577319587628,
+      "loss": 0.7962,
+      "step": 599
+    },
+    {
+      "epoch": 0.192,
+      "grad_norm": 0.1290092021226883,
+      "learning_rate": 0.00018148453608247424,
+      "loss": 0.6357,
+      "step": 600
+    },
+    {
+      "epoch": 0.19232,
+      "grad_norm": 0.12456108629703522,
+      "learning_rate": 0.00018144329896907217,
+      "loss": 0.8116,
+      "step": 601
+    },
+    {
+      "epoch": 0.19264,
+      "grad_norm": 0.11560390144586563,
+      "learning_rate": 0.0001814020618556701,
+      "loss": 0.6152,
+      "step": 602
+    },
+    {
+      "epoch": 0.19296,
+      "grad_norm": 0.1171826645731926,
+      "learning_rate": 0.00018136082474226806,
+      "loss": 0.7314,
+      "step": 603
+    },
+    {
+      "epoch": 0.19328,
+      "grad_norm": 0.11141692847013474,
+      "learning_rate": 0.000181319587628866,
+      "loss": 0.831,
+      "step": 604
+    },
+    {
+      "epoch": 0.1936,
+      "grad_norm": 0.12226671725511551,
+      "learning_rate": 0.00018127835051546393,
+      "loss": 0.6946,
+      "step": 605
+    },
+    {
+      "epoch": 0.19392,
+      "grad_norm": 0.12684857845306396,
+      "learning_rate": 0.00018123711340206186,
+      "loss": 0.7299,
+      "step": 606
+    },
+    {
+      "epoch": 0.19424,
+      "grad_norm": 0.12957026064395905,
+      "learning_rate": 0.0001811958762886598,
+      "loss": 0.7208,
+      "step": 607
+    },
+    {
+      "epoch": 0.19456,
+      "grad_norm": 0.11711089313030243,
+      "learning_rate": 0.00018115463917525775,
+      "loss": 0.6967,
+      "step": 608
+    },
+    {
+      "epoch": 0.19488,
+      "grad_norm": 0.12728916108608246,
+      "learning_rate": 0.00018111340206185568,
+      "loss": 0.6866,
+      "step": 609
+    },
+    {
+      "epoch": 0.1952,
+      "grad_norm": 0.11292556673288345,
+      "learning_rate": 0.00018107216494845364,
+      "loss": 0.6493,
+      "step": 610
+    },
+    {
+      "epoch": 0.19552,
+      "grad_norm": 0.11602400243282318,
+      "learning_rate": 0.00018103092783505157,
+      "loss": 0.7802,
+      "step": 611
+    },
+    {
+      "epoch": 0.19584,
+      "grad_norm": 0.12061775475740433,
+      "learning_rate": 0.0001809896907216495,
+      "loss": 0.7336,
+      "step": 612
+    },
+    {
+      "epoch": 0.19616,
+      "grad_norm": 0.13388070464134216,
+      "learning_rate": 0.00018094845360824743,
+      "loss": 0.6978,
+      "step": 613
+    },
+    {
+      "epoch": 0.19648,
+      "grad_norm": 0.11619807779788971,
+      "learning_rate": 0.00018090721649484536,
+      "loss": 0.7327,
+      "step": 614
+    },
+    {
+      "epoch": 0.1968,
+      "grad_norm": 0.12670999765396118,
+      "learning_rate": 0.00018086597938144332,
+      "loss": 0.7561,
+      "step": 615
+    },
+    {
+      "epoch": 0.19712,
+      "grad_norm": 0.11523107439279556,
+      "learning_rate": 0.00018082474226804125,
+      "loss": 0.7343,
+      "step": 616
+    },
+    {
+      "epoch": 0.19744,
+      "grad_norm": 0.13320648670196533,
+      "learning_rate": 0.00018078350515463918,
+      "loss": 0.8011,
+      "step": 617
+    },
+    {
+      "epoch": 0.19776,
+      "grad_norm": 0.1562347114086151,
+      "learning_rate": 0.00018074226804123714,
+      "loss": 0.8,
+      "step": 618
+    },
+    {
+      "epoch": 0.19808,
+      "grad_norm": 0.13111111521720886,
+      "learning_rate": 0.00018070103092783507,
+      "loss": 0.7242,
+      "step": 619
+    },
+    {
+      "epoch": 0.1984,
+      "grad_norm": 0.13651488721370697,
+      "learning_rate": 0.000180659793814433,
+      "loss": 0.7653,
+      "step": 620
+    },
+    {
+      "epoch": 0.19872,
+      "grad_norm": 0.12414918839931488,
+      "learning_rate": 0.00018061855670103093,
+      "loss": 0.803,
+      "step": 621
+    },
+    {
+      "epoch": 0.19904,
+      "grad_norm": 0.11437831819057465,
+      "learning_rate": 0.00018057731958762887,
+      "loss": 0.6214,
+      "step": 622
+    },
+    {
+      "epoch": 0.19936,
+      "grad_norm": 0.10947408527135849,
+      "learning_rate": 0.00018053608247422682,
+      "loss": 0.6069,
+      "step": 623
+    },
+    {
+      "epoch": 0.19968,
+      "grad_norm": 0.11896886676549911,
+      "learning_rate": 0.00018049484536082475,
+      "loss": 0.8071,
+      "step": 624
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.10557591170072556,
+      "learning_rate": 0.00018045360824742269,
+      "loss": 0.6428,
+      "step": 625
+    },
+    {
+      "epoch": 0.20032,
+      "grad_norm": 0.11610836535692215,
+      "learning_rate": 0.00018041237113402062,
+      "loss": 0.6034,
+      "step": 626
+    },
+    {
+      "epoch": 0.20064,
+      "grad_norm": 0.11742173135280609,
+      "learning_rate": 0.00018037113402061855,
+      "loss": 0.7368,
+      "step": 627
+    },
+    {
+      "epoch": 0.20096,
+      "grad_norm": 0.11513591557741165,
+      "learning_rate": 0.0001803298969072165,
+      "loss": 0.5753,
+      "step": 628
+    },
+    {
+      "epoch": 0.20128,
+      "grad_norm": 0.12790197134017944,
+      "learning_rate": 0.00018028865979381444,
+      "loss": 0.6109,
+      "step": 629
+    },
+    {
+      "epoch": 0.2016,
+      "grad_norm": 0.12305860221385956,
+      "learning_rate": 0.00018024742268041237,
+      "loss": 0.6894,
+      "step": 630
+    },
+    {
+      "epoch": 0.20192,
+      "grad_norm": 0.1050051599740982,
+      "learning_rate": 0.00018020618556701033,
+      "loss": 0.7507,
+      "step": 631
+    },
+    {
+      "epoch": 0.20224,
+      "grad_norm": 0.12011159211397171,
+      "learning_rate": 0.00018016494845360826,
+      "loss": 0.6234,
+      "step": 632
+    },
+    {
+      "epoch": 0.20256,
+      "grad_norm": 0.13481462001800537,
+      "learning_rate": 0.0001801237113402062,
+      "loss": 0.7408,
+      "step": 633
+    },
+    {
+      "epoch": 0.20288,
+      "grad_norm": 0.11771773546934128,
+      "learning_rate": 0.00018008247422680412,
+      "loss": 0.7283,
+      "step": 634
+    },
+    {
+      "epoch": 0.2032,
+      "grad_norm": 0.12805534899234772,
+      "learning_rate": 0.00018004123711340205,
+      "loss": 0.6377,
+      "step": 635
+    },
+    {
+      "epoch": 0.20352,
+      "grad_norm": 0.1451328545808792,
+      "learning_rate": 0.00018,
+      "loss": 0.6929,
+      "step": 636
+    },
+    {
+      "epoch": 0.20384,
+      "grad_norm": 0.116309255361557,
+      "learning_rate": 0.00017995876288659794,
+      "loss": 0.8133,
+      "step": 637
+    },
+    {
+      "epoch": 0.20416,
+      "grad_norm": 0.11883087456226349,
+      "learning_rate": 0.0001799175257731959,
+      "loss": 0.7036,
+      "step": 638
+    },
+    {
+      "epoch": 0.20448,
+      "grad_norm": 0.11787105351686478,
+      "learning_rate": 0.00017987628865979383,
+      "loss": 0.7313,
+      "step": 639
+    },
+    {
+      "epoch": 0.2048,
+      "grad_norm": 0.10839900374412537,
+      "learning_rate": 0.00017983505154639176,
+      "loss": 0.5725,
+      "step": 640
+    },
+    {
+      "epoch": 0.20512,
+      "grad_norm": 0.12540479004383087,
+      "learning_rate": 0.0001797938144329897,
+      "loss": 0.6349,
+      "step": 641
+    },
+    {
+      "epoch": 0.20544,
+      "grad_norm": 0.1231970563530922,
+      "learning_rate": 0.00017975257731958763,
+      "loss": 0.7068,
+      "step": 642
+    },
+    {
+      "epoch": 0.20576,
+      "grad_norm": 0.1263824850320816,
+      "learning_rate": 0.00017971134020618558,
+      "loss": 0.861,
+      "step": 643
+    },
+    {
+      "epoch": 0.20608,
+      "grad_norm": 0.134733647108078,
+      "learning_rate": 0.00017967010309278352,
+      "loss": 0.67,
+      "step": 644
+    },
+    {
+      "epoch": 0.2064,
+      "grad_norm": 0.14626337587833405,
+      "learning_rate": 0.00017962886597938145,
+      "loss": 0.6905,
+      "step": 645
+    },
+    {
+      "epoch": 0.20672,
+      "grad_norm": 0.11077899485826492,
+      "learning_rate": 0.0001795876288659794,
+      "loss": 0.6866,
+      "step": 646
+    },
+    {
+      "epoch": 0.20704,
+      "grad_norm": 0.12598447501659393,
+      "learning_rate": 0.00017954639175257734,
+      "loss": 0.7583,
+      "step": 647
+    },
+    {
+      "epoch": 0.20736,
+      "grad_norm": 0.12656626105308533,
+      "learning_rate": 0.00017950515463917527,
+      "loss": 0.679,
+      "step": 648
+    },
+    {
+      "epoch": 0.20768,
+      "grad_norm": 0.1403527706861496,
+      "learning_rate": 0.0001794639175257732,
+      "loss": 0.8081,
+      "step": 649
+    },
+    {
+      "epoch": 0.208,
+      "grad_norm": 0.11713801324367523,
+      "learning_rate": 0.00017942268041237113,
+      "loss": 0.7322,
+      "step": 650
+    },
+    {
+      "epoch": 0.20832,
+      "grad_norm": 0.1281721144914627,
+      "learning_rate": 0.0001793814432989691,
+      "loss": 0.8457,
+      "step": 651
+    },
+    {
+      "epoch": 0.20864,
+      "grad_norm": 0.11331729590892792,
+      "learning_rate": 0.00017934020618556702,
+      "loss": 0.6479,
+      "step": 652
+    },
+    {
+      "epoch": 0.20896,
+      "grad_norm": 0.11600172519683838,
+      "learning_rate": 0.00017929896907216495,
+      "loss": 0.7144,
+      "step": 653
+    },
+    {
+      "epoch": 0.20928,
+      "grad_norm": 0.13482129573822021,
+      "learning_rate": 0.0001792577319587629,
+      "loss": 0.7639,
+      "step": 654
+    },
+    {
+      "epoch": 0.2096,
+      "grad_norm": 0.12466888129711151,
+      "learning_rate": 0.00017921649484536081,
+      "loss": 0.6413,
+      "step": 655
+    },
+    {
+      "epoch": 0.20992,
+      "grad_norm": 0.11935215443372726,
+      "learning_rate": 0.00017917525773195877,
+      "loss": 0.7551,
+      "step": 656
+    },
+    {
+      "epoch": 0.21024,
+      "grad_norm": 0.12447444349527359,
+      "learning_rate": 0.0001791340206185567,
+      "loss": 0.7171,
+      "step": 657
+    },
+    {
+      "epoch": 0.21056,
+      "grad_norm": 0.1334541141986847,
+      "learning_rate": 0.00017909278350515463,
+      "loss": 0.9159,
+      "step": 658
+    },
+    {
+      "epoch": 0.21088,
+      "grad_norm": 0.111027792096138,
+      "learning_rate": 0.0001790515463917526,
+      "loss": 0.6894,
+      "step": 659
+    },
+    {
+      "epoch": 0.2112,
+      "grad_norm": 0.12192652374505997,
+      "learning_rate": 0.00017901030927835052,
+      "loss": 0.7354,
+      "step": 660
+    },
+    {
+      "epoch": 0.21152,
+      "grad_norm": 0.12837322056293488,
+      "learning_rate": 0.00017896907216494848,
+      "loss": 0.7535,
+      "step": 661
+    },
+    {
+      "epoch": 0.21184,
+      "grad_norm": 0.1143856793642044,
+      "learning_rate": 0.0001789278350515464,
+      "loss": 0.6189,
+      "step": 662
+    },
+    {
+      "epoch": 0.21216,
+      "grad_norm": 0.12996436655521393,
+      "learning_rate": 0.00017888659793814432,
+      "loss": 0.7194,
+      "step": 663
+    },
+    {
+      "epoch": 0.21248,
+      "grad_norm": 0.13088072836399078,
+      "learning_rate": 0.00017884536082474228,
+      "loss": 0.6356,
+      "step": 664
+    },
+    {
+      "epoch": 0.2128,
+      "grad_norm": 0.11904153227806091,
+      "learning_rate": 0.0001788041237113402,
+      "loss": 0.5998,
+      "step": 665
+    },
+    {
+      "epoch": 0.21312,
+      "grad_norm": 0.11976154148578644,
+      "learning_rate": 0.00017876288659793817,
+      "loss": 0.7417,
+      "step": 666
+    },
+    {
+      "epoch": 0.21344,
+      "grad_norm": 0.1175193041563034,
+      "learning_rate": 0.0001787216494845361,
+      "loss": 0.741,
+      "step": 667
+    },
+    {
+      "epoch": 0.21376,
+      "grad_norm": 0.1151885837316513,
+      "learning_rate": 0.00017868041237113403,
+      "loss": 0.7779,
+      "step": 668
+    },
+    {
+      "epoch": 0.21408,
+      "grad_norm": 0.1385083645582199,
+      "learning_rate": 0.00017863917525773196,
+      "loss": 0.7296,
+      "step": 669
+    },
+    {
+      "epoch": 0.2144,
+      "grad_norm": 0.1304243803024292,
+      "learning_rate": 0.0001785979381443299,
+      "loss": 0.7085,
+      "step": 670
+    },
+    {
+      "epoch": 0.21472,
+      "grad_norm": 0.11901744455099106,
+      "learning_rate": 0.00017855670103092785,
+      "loss": 0.7005,
+      "step": 671
+    },
+    {
+      "epoch": 0.21504,
+      "grad_norm": 0.11635670065879822,
+      "learning_rate": 0.00017851546391752578,
+      "loss": 0.7824,
+      "step": 672
+    },
+    {
+      "epoch": 0.21536,
+      "grad_norm": 0.14305877685546875,
+      "learning_rate": 0.0001784742268041237,
+      "loss": 0.6263,
+      "step": 673
+    },
+    {
+      "epoch": 0.21568,
+      "grad_norm": 0.11984144151210785,
+      "learning_rate": 0.00017843298969072167,
+      "loss": 0.6518,
+      "step": 674
+    },
+    {
+      "epoch": 0.216,
+      "grad_norm": 0.10570491850376129,
+      "learning_rate": 0.0001783917525773196,
+      "loss": 0.6311,
+      "step": 675
+    },
+    {
+      "epoch": 0.21632,
+      "grad_norm": 0.13266734778881073,
+      "learning_rate": 0.00017835051546391753,
+      "loss": 0.7975,
+      "step": 676
+    },
+    {
+      "epoch": 0.21664,
+      "grad_norm": 0.12022270262241364,
+      "learning_rate": 0.00017830927835051546,
+      "loss": 0.6691,
+      "step": 677
+    },
+    {
+      "epoch": 0.21696,
+      "grad_norm": 0.1266927272081375,
+      "learning_rate": 0.0001782680412371134,
+      "loss": 0.7211,
+      "step": 678
+    },
+    {
+      "epoch": 0.21728,
+      "grad_norm": 0.13459637761116028,
+      "learning_rate": 0.00017822680412371135,
+      "loss": 0.7911,
+      "step": 679
+    },
+    {
+      "epoch": 0.2176,
+      "grad_norm": 0.14075347781181335,
+      "learning_rate": 0.00017818556701030929,
+      "loss": 0.8012,
+      "step": 680
+    },
+    {
+      "epoch": 0.21792,
+      "grad_norm": 0.12798281013965607,
+      "learning_rate": 0.00017814432989690724,
+      "loss": 0.6135,
+      "step": 681
+    },
+    {
+      "epoch": 0.21824,
+      "grad_norm": 0.14813600480556488,
+      "learning_rate": 0.00017810309278350518,
+      "loss": 0.7988,
+      "step": 682
+    },
+    {
+      "epoch": 0.21856,
+      "grad_norm": 0.14158020913600922,
+      "learning_rate": 0.0001780618556701031,
+      "loss": 0.7725,
+      "step": 683
+    },
+    {
+      "epoch": 0.21888,
+      "grad_norm": 0.10573446750640869,
+      "learning_rate": 0.00017802061855670104,
+      "loss": 0.7956,
+      "step": 684
+    },
+    {
+      "epoch": 0.2192,
+      "grad_norm": 0.14443108439445496,
+      "learning_rate": 0.00017797938144329897,
+      "loss": 0.8615,
+      "step": 685
+    },
+    {
+      "epoch": 0.21952,
+      "grad_norm": 0.1228407770395279,
+      "learning_rate": 0.0001779381443298969,
+      "loss": 0.706,
+      "step": 686
+    },
+    {
+      "epoch": 0.21984,
+      "grad_norm": 0.1218361034989357,
+      "learning_rate": 0.00017789690721649486,
+      "loss": 0.6525,
+      "step": 687
+    },
+    {
+      "epoch": 0.22016,
+      "grad_norm": 0.12609677016735077,
+      "learning_rate": 0.0001778556701030928,
+      "loss": 0.5638,
+      "step": 688
+    },
+    {
+      "epoch": 0.22048,
+      "grad_norm": 0.1489102840423584,
+      "learning_rate": 0.00017781443298969075,
+      "loss": 0.641,
+      "step": 689
+    },
+    {
+      "epoch": 0.2208,
+      "grad_norm": 0.11327382177114487,
+      "learning_rate": 0.00017777319587628868,
+      "loss": 0.7372,
+      "step": 690
+    },
+    {
+      "epoch": 0.22112,
+      "grad_norm": 0.12614741921424866,
+      "learning_rate": 0.00017773195876288658,
+      "loss": 0.7819,
+      "step": 691
+    },
+    {
+      "epoch": 0.22144,
+      "grad_norm": 0.11078184098005295,
+      "learning_rate": 0.00017769072164948454,
+      "loss": 0.8376,
+      "step": 692
+    },
+    {
+      "epoch": 0.22176,
+      "grad_norm": 0.13330969214439392,
+      "learning_rate": 0.00017764948453608247,
+      "loss": 0.771,
+      "step": 693
+    },
+    {
+      "epoch": 0.22208,
+      "grad_norm": 0.14074283838272095,
+      "learning_rate": 0.00017760824742268043,
+      "loss": 0.7675,
+      "step": 694
+    },
+    {
+      "epoch": 0.2224,
+      "grad_norm": 0.12653911113739014,
+      "learning_rate": 0.00017756701030927836,
+      "loss": 0.7578,
+      "step": 695
+    },
+    {
+      "epoch": 0.22272,
+      "grad_norm": 0.13229238986968994,
+      "learning_rate": 0.0001775257731958763,
+      "loss": 0.5843,
+      "step": 696
+    },
+    {
+      "epoch": 0.22304,
+      "grad_norm": 0.14447958767414093,
+      "learning_rate": 0.00017748453608247425,
+      "loss": 0.8251,
+      "step": 697
+    },
+    {
+      "epoch": 0.22336,
+      "grad_norm": 0.12437491863965988,
+      "learning_rate": 0.00017744329896907216,
+      "loss": 0.7624,
+      "step": 698
+    },
+    {
+      "epoch": 0.22368,
+      "grad_norm": 0.13665013015270233,
+      "learning_rate": 0.00017740206185567012,
+      "loss": 0.6546,
+      "step": 699
+    },
+    {
+      "epoch": 0.224,
+      "grad_norm": 0.12415925413370132,
+      "learning_rate": 0.00017736082474226805,
+      "loss": 0.5989,
+      "step": 700
+    },
+    {
+      "epoch": 0.22432,
+      "grad_norm": 0.13429298996925354,
+      "learning_rate": 0.00017731958762886598,
+      "loss": 0.7283,
+      "step": 701
+    },
+    {
+      "epoch": 0.22464,
+      "grad_norm": 0.11220762878656387,
+      "learning_rate": 0.00017727835051546394,
+      "loss": 0.844,
+      "step": 702
+    },
+    {
+      "epoch": 0.22496,
+      "grad_norm": 0.15322859585285187,
+      "learning_rate": 0.00017723711340206187,
+      "loss": 0.744,
+      "step": 703
+    },
+    {
+      "epoch": 0.22528,
+      "grad_norm": 0.10617350786924362,
+      "learning_rate": 0.0001771958762886598,
+      "loss": 0.7462,
+      "step": 704
+    },
+    {
+      "epoch": 0.2256,
+      "grad_norm": 0.12569446861743927,
+      "learning_rate": 0.00017715463917525773,
+      "loss": 0.7465,
+      "step": 705
+    },
+    {
+      "epoch": 0.22592,
+      "grad_norm": 0.13956156373023987,
+      "learning_rate": 0.00017711340206185566,
+      "loss": 0.7465,
+      "step": 706
+    },
+    {
+      "epoch": 0.22624,
+      "grad_norm": 0.13465113937854767,
+      "learning_rate": 0.00017707216494845362,
+      "loss": 0.8502,
+      "step": 707
+    },
+    {
+      "epoch": 0.22656,
+      "grad_norm": 0.11806920915842056,
+      "learning_rate": 0.00017703092783505155,
+      "loss": 0.6246,
+      "step": 708
+    },
+    {
+      "epoch": 0.22688,
+      "grad_norm": 0.1319315880537033,
+      "learning_rate": 0.0001769896907216495,
+      "loss": 0.6639,
+      "step": 709
+    },
+    {
+      "epoch": 0.2272,
+      "grad_norm": 0.14387674629688263,
+      "learning_rate": 0.00017694845360824744,
+      "loss": 0.8698,
+      "step": 710
+    },
+    {
+      "epoch": 0.22752,
+      "grad_norm": 0.12522293627262115,
+      "learning_rate": 0.00017690721649484537,
+      "loss": 0.7884,
+      "step": 711
+    },
+    {
+      "epoch": 0.22784,
+      "grad_norm": 0.11482946574687958,
+      "learning_rate": 0.0001768659793814433,
+      "loss": 0.7562,
+      "step": 712
+    },
+    {
+      "epoch": 0.22816,
+      "grad_norm": 0.11533328890800476,
+      "learning_rate": 0.00017682474226804123,
+      "loss": 0.6228,
+      "step": 713
+    },
+    {
+      "epoch": 0.22848,
+      "grad_norm": 0.12137199193239212,
+      "learning_rate": 0.00017678350515463917,
+      "loss": 0.8835,
+      "step": 714
+    },
+    {
+      "epoch": 0.2288,
+      "grad_norm": 0.11885856837034225,
+      "learning_rate": 0.00017674226804123712,
+      "loss": 0.6241,
+      "step": 715
+    },
+    {
+      "epoch": 0.22912,
+      "grad_norm": 0.1262994110584259,
+      "learning_rate": 0.00017670103092783506,
+      "loss": 0.715,
+      "step": 716
+    },
+    {
+      "epoch": 0.22944,
+      "grad_norm": 0.1313248574733734,
+      "learning_rate": 0.00017665979381443301,
+      "loss": 0.8368,
+      "step": 717
+    },
+    {
+      "epoch": 0.22976,
+      "grad_norm": 0.15882809460163116,
+      "learning_rate": 0.00017661855670103095,
+      "loss": 0.8688,
+      "step": 718
+    },
+    {
+      "epoch": 0.23008,
+      "grad_norm": 0.13610002398490906,
+      "learning_rate": 0.00017657731958762888,
+      "loss": 0.7367,
+      "step": 719
+    },
+    {
+      "epoch": 0.2304,
+      "grad_norm": 0.11154987663030624,
+      "learning_rate": 0.0001765360824742268,
+      "loss": 0.8026,
+      "step": 720
+    },
+    {
+      "epoch": 0.23072,
+      "grad_norm": 0.14375466108322144,
+      "learning_rate": 0.00017649484536082474,
+      "loss": 0.8038,
+      "step": 721
+    },
+    {
+      "epoch": 0.23104,
+      "grad_norm": 0.13212132453918457,
+      "learning_rate": 0.0001764536082474227,
+      "loss": 0.9008,
+      "step": 722
+    },
+    {
+      "epoch": 0.23136,
+      "grad_norm": 0.1395123153924942,
+      "learning_rate": 0.00017641237113402063,
+      "loss": 0.7491,
+      "step": 723
+    },
+    {
+      "epoch": 0.23168,
+      "grad_norm": 0.11533952504396439,
+      "learning_rate": 0.00017637113402061856,
+      "loss": 0.6714,
+      "step": 724
+    },
+    {
+      "epoch": 0.232,
+      "grad_norm": 0.10333327949047089,
+      "learning_rate": 0.00017632989690721652,
+      "loss": 0.6643,
+      "step": 725
+    },
+    {
+      "epoch": 0.23232,
+      "grad_norm": 0.12025081366300583,
+      "learning_rate": 0.00017628865979381445,
+      "loss": 0.6639,
+      "step": 726
+    },
+    {
+      "epoch": 0.23264,
+      "grad_norm": 0.13901305198669434,
+      "learning_rate": 0.00017624742268041238,
+      "loss": 0.7138,
+      "step": 727
+    },
+    {
+      "epoch": 0.23296,
+      "grad_norm": 0.14389030635356903,
+      "learning_rate": 0.0001762061855670103,
+      "loss": 0.7785,
+      "step": 728
+    },
+    {
+      "epoch": 0.23328,
+      "grad_norm": 0.11535743623971939,
+      "learning_rate": 0.00017616494845360824,
+      "loss": 0.715,
+      "step": 729
+    },
+    {
+      "epoch": 0.2336,
+      "grad_norm": 0.11448106914758682,
+      "learning_rate": 0.0001761237113402062,
+      "loss": 0.7739,
+      "step": 730
+    },
+    {
+      "epoch": 0.23392,
+      "grad_norm": 0.1345527321100235,
+      "learning_rate": 0.00017608247422680413,
+      "loss": 0.6493,
+      "step": 731
+    },
+    {
+      "epoch": 0.23424,
+      "grad_norm": 0.12693467736244202,
+      "learning_rate": 0.0001760412371134021,
+      "loss": 0.7164,
+      "step": 732
+    },
+    {
+      "epoch": 0.23456,
+      "grad_norm": 0.12556055188179016,
+      "learning_rate": 0.00017600000000000002,
+      "loss": 0.7162,
+      "step": 733
+    },
+    {
+      "epoch": 0.23488,
+      "grad_norm": 0.13427414000034332,
+      "learning_rate": 0.00017595876288659793,
+      "loss": 0.7169,
+      "step": 734
+    },
+    {
+      "epoch": 0.2352,
+      "grad_norm": 0.1384212076663971,
+      "learning_rate": 0.00017591752577319589,
+      "loss": 0.8806,
+      "step": 735
+    },
+    {
+      "epoch": 0.23552,
+      "grad_norm": 0.1685527116060257,
+      "learning_rate": 0.00017587628865979382,
+      "loss": 0.8755,
+      "step": 736
+    },
+    {
+      "epoch": 0.23584,
+      "grad_norm": 0.12363595515489578,
+      "learning_rate": 0.00017583505154639177,
+      "loss": 0.7556,
+      "step": 737
+    },
+    {
+      "epoch": 0.23616,
+      "grad_norm": 0.121871218085289,
+      "learning_rate": 0.0001757938144329897,
+      "loss": 0.7021,
+      "step": 738
+    },
+    {
+      "epoch": 0.23648,
+      "grad_norm": 0.133682981133461,
+      "learning_rate": 0.00017575257731958764,
+      "loss": 0.8063,
+      "step": 739
+    },
+    {
+      "epoch": 0.2368,
+      "grad_norm": 0.10897425562143326,
+      "learning_rate": 0.00017571134020618557,
+      "loss": 0.7153,
+      "step": 740
+    },
+    {
+      "epoch": 0.23712,
+      "grad_norm": 0.1330791562795639,
+      "learning_rate": 0.0001756701030927835,
+      "loss": 0.7005,
+      "step": 741
+    },
+    {
+      "epoch": 0.23744,
+      "grad_norm": 0.13002124428749084,
+      "learning_rate": 0.00017562886597938146,
+      "loss": 0.7188,
+      "step": 742
+    },
+    {
+      "epoch": 0.23776,
+      "grad_norm": 0.11166704446077347,
+      "learning_rate": 0.0001755876288659794,
+      "loss": 0.6922,
+      "step": 743
+    },
+    {
+      "epoch": 0.23808,
+      "grad_norm": 0.1195744052529335,
+      "learning_rate": 0.00017554639175257732,
+      "loss": 0.6846,
+      "step": 744
+    },
+    {
+      "epoch": 0.2384,
+      "grad_norm": 0.12304572016000748,
+      "learning_rate": 0.00017550515463917528,
+      "loss": 0.7575,
+      "step": 745
+    },
+    {
+      "epoch": 0.23872,
+      "grad_norm": 0.15100136399269104,
+      "learning_rate": 0.0001754639175257732,
+      "loss": 0.9129,
+      "step": 746
+    },
+    {
+      "epoch": 0.23904,
+      "grad_norm": 0.11369301378726959,
+      "learning_rate": 0.00017542268041237114,
+      "loss": 0.8557,
+      "step": 747
+    },
+    {
+      "epoch": 0.23936,
+      "grad_norm": 0.14192970097064972,
+      "learning_rate": 0.00017538144329896907,
+      "loss": 0.7254,
+      "step": 748
+    },
+    {
+      "epoch": 0.23968,
+      "grad_norm": 0.11873924732208252,
+      "learning_rate": 0.000175340206185567,
+      "loss": 0.7028,
+      "step": 749
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.1388227790594101,
+      "learning_rate": 0.00017529896907216496,
+      "loss": 0.8156,
+      "step": 750
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 5000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.228197673132073e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}