diff --git "a/checkpoint-1398/trainer_state.json" "b/checkpoint-1398/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-1398/trainer_state.json"
@@ -0,0 +1,8959 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 20,
+  "global_step": 1398,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 2e-05,
+      "loss": 3.8527,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4e-05,
+      "loss": 2.357,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6e-05,
+      "loss": 1.6448,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8e-05,
+      "loss": 3.7495,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.0001,
+      "loss": 3.6283,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00012,
+      "loss": 2.8617,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00014,
+      "loss": 2.1554,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00016,
+      "loss": 1.5715,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00018,
+      "loss": 1.162,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00019999988670134103,
+      "loss": 0.8023,
+      "step": 11
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00019999954680562074,
+      "loss": 0.9134,
+      "step": 12
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00019999898031360943,
+      "loss": 0.8533,
+      "step": 13
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00019999818722659068,
+      "loss": 0.792,
+      "step": 14
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00019999716754636165,
+      "loss": 0.9903,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00019999592127523287,
+      "loss": 0.7565,
+      "step": 16
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001999944484160284,
+      "loss": 0.8171,
+      "step": 17
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00019999274897208565,
+      "loss": 0.9344,
+      "step": 18
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00019999082294725555,
+      "loss": 0.8733,
+      "step": 19
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00019998867034590241,
+      "loss": 0.9758,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 0.6869800090789795,
+      "eval_runtime": 29.3692,
+      "eval_samples_per_second": 55.466,
+      "eval_steps_per_second": 27.75,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00019998629117290396,
+      "loss": 0.8391,
+      "step": 21
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0001999836854336514,
+      "loss": 1.032,
+      "step": 22
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00019998085313404916,
+      "loss": 0.9099,
+      "step": 23
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00019997779428051522,
+      "loss": 0.8684,
+      "step": 24
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019997450887998086,
+      "loss": 0.9439,
+      "step": 25
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0001999709969398907,
+      "loss": 1.0515,
+      "step": 26
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0001999672584682027,
+      "loss": 1.1631,
+      "step": 27
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019996329347338814,
+      "loss": 0.7389,
+      "step": 28
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019995910196443157,
+      "loss": 0.6839,
+      "step": 29
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019995468395083088,
+      "loss": 1.0289,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019995003944259715,
+      "loss": 0.9143,
+      "step": 31
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00019994516845025468,
+      "loss": 0.95,
+      "step": 32
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0001999400709848411,
+      "loss": 0.9214,
+      "step": 33
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.000199934747057907,
+      "loss": 0.9471,
+      "step": 34
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00019992919668151635,
+      "loss": 0.7858,
+      "step": 35
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00019992341986824612,
+      "loss": 0.6643,
+      "step": 36
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00019991741663118642,
+      "loss": 0.7344,
+      "step": 37
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00019991118698394042,
+      "loss": 0.7909,
+      "step": 38
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00019990473094062434,
+      "loss": 0.7732,
+      "step": 39
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00019989804851586743,
+      "loss": 0.7228,
+      "step": 40
+    },
+    {
+      "epoch": 0.06,
+      "eval_loss": 0.6790834069252014,
+      "eval_runtime": 29.3451,
+      "eval_samples_per_second": 55.512,
+      "eval_steps_per_second": 27.773,
+      "step": 40
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00019989113972481183,
+      "loss": 0.8713,
+      "step": 41
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0001998840045831127,
+      "loss": 0.6862,
+      "step": 42
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00019987664310693805,
+      "loss": 0.7524,
+      "step": 43
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00019986905531296884,
+      "loss": 0.7405,
+      "step": 44
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0001998612412183988,
+      "loss": 0.7113,
+      "step": 45
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00019985320084093443,
+      "loss": 0.7264,
+      "step": 46
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00019984493419879503,
+      "loss": 0.717,
+      "step": 47
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00019983644131071256,
+      "loss": 0.9136,
+      "step": 48
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00019982772219593172,
+      "loss": 0.662,
+      "step": 49
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00019981877687420975,
+      "loss": 0.6811,
+      "step": 50
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00019980960536581654,
+      "loss": 0.7089,
+      "step": 51
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0001998002076915345,
+      "loss": 0.774,
+      "step": 52
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019979058387265843,
+      "loss": 0.9975,
+      "step": 53
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0001997807339309957,
+      "loss": 0.8504,
+      "step": 54
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019977065788886602,
+      "loss": 0.7098,
+      "step": 55
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0001997603557691014,
+      "loss": 0.6549,
+      "step": 56
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019974982759504625,
+      "loss": 0.6731,
+      "step": 57
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.000199739073390557,
+      "loss": 0.8107,
+      "step": 58
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019972809318000246,
+      "loss": 0.7196,
+      "step": 59
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019971688698826353,
+      "loss": 0.6804,
+      "step": 60
+    },
+    {
+      "epoch": 0.09,
+      "eval_loss": 0.6613275408744812,
+      "eval_runtime": 29.3333,
+      "eval_samples_per_second": 55.534,
+      "eval_steps_per_second": 27.784,
+      "step": 60
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019970545484073306,
+      "loss": 0.7939,
+      "step": 61
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019969379676331602,
+      "loss": 0.7578,
+      "step": 62
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019968191278242934,
+      "loss": 0.792,
+      "step": 63
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019966980292500174,
+      "loss": 0.9361,
+      "step": 64
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019965746721847387,
+      "loss": 0.7841,
+      "step": 65
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00019964490569079812,
+      "loss": 0.7706,
+      "step": 66
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00019963211837043852,
+      "loss": 0.8221,
+      "step": 67
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00019961910528637088,
+      "loss": 0.6315,
+      "step": 68
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0001996058664680824,
+      "loss": 0.6358,
+      "step": 69
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00019959240194557197,
+      "loss": 0.7126,
+      "step": 70
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00019957871174934978,
+      "loss": 0.5788,
+      "step": 71
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00019956479591043752,
+      "loss": 0.7683,
+      "step": 72
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.000199550654460368,
+      "loss": 0.7092,
+      "step": 73
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00019953628743118546,
+      "loss": 0.7995,
+      "step": 74
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00019952169485544516,
+      "loss": 0.6204,
+      "step": 75
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00019950687676621352,
+      "loss": 0.6226,
+      "step": 76
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0001994918331970679,
+      "loss": 0.5133,
+      "step": 77
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0001994765641820966,
+      "loss": 0.9482,
+      "step": 78
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00019946106975589884,
+      "loss": 0.7825,
+      "step": 79
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00019944534995358458,
+      "loss": 0.8117,
+      "step": 80
+    },
+    {
+      "epoch": 0.11,
+      "eval_loss": 0.6359772682189941,
+      "eval_runtime": 29.2458,
+      "eval_samples_per_second": 55.7,
+      "eval_steps_per_second": 27.867,
+      "step": 80
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00019942940481077446,
+      "loss": 0.844,
+      "step": 81
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00019941323436359972,
+      "loss": 0.766,
+      "step": 82
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00019939683864870217,
+      "loss": 0.7071,
+      "step": 83
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001993802177032341,
+      "loss": 0.7645,
+      "step": 84
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001993633715648581,
+      "loss": 0.9787,
+      "step": 85
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00019934630027174707,
+      "loss": 0.6916,
+      "step": 86
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00019932900386258407,
+      "loss": 0.6979,
+      "step": 87
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0001993114823765623,
+      "loss": 0.688,
+      "step": 88
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00019929373585338508,
+      "loss": 0.8155,
+      "step": 89
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00019927576433326544,
+      "loss": 0.6204,
+      "step": 90
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0001992575678569264,
+      "loss": 0.8641,
+      "step": 91
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0001992391464656007,
+      "loss": 0.758,
+      "step": 92
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0001992205002010307,
+      "loss": 0.6035,
+      "step": 93
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00019920162910546833,
+      "loss": 0.857,
+      "step": 94
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00019918253322167502,
+      "loss": 0.6189,
+      "step": 95
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00019916321259292152,
+      "loss": 0.8429,
+      "step": 96
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00019914366726298782,
+      "loss": 0.9827,
+      "step": 97
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00019912389727616314,
+      "loss": 0.9278,
+      "step": 98
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00019910390267724576,
+      "loss": 0.6625,
+      "step": 99
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0001990836835115429,
+      "loss": 0.6458,
+      "step": 100
+    },
+    {
+      "epoch": 0.14,
+      "eval_loss": 0.6334865093231201,
+      "eval_runtime": 29.3599,
+      "eval_samples_per_second": 55.484,
+      "eval_steps_per_second": 27.759,
+      "step": 100
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0001990632398248706,
+      "loss": 0.6327,
+      "step": 101
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019904257166355376,
+      "loss": 0.719,
+      "step": 102
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019902167907442583,
+      "loss": 0.6946,
+      "step": 103
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019900056210482892,
+      "loss": 0.6667,
+      "step": 104
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019897922080261345,
+      "loss": 0.6793,
+      "step": 105
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0001989576552161383,
+      "loss": 0.7095,
+      "step": 106
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019893586539427048,
+      "loss": 0.6903,
+      "step": 107
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0001989138513863851,
+      "loss": 0.9088,
+      "step": 108
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0001988916132423654,
+      "loss": 0.7673,
+      "step": 109
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019886915101260234,
+      "loss": 0.671,
+      "step": 110
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019884646474799475,
+      "loss": 0.6958,
+      "step": 111
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019882355449994915,
+      "loss": 0.6109,
+      "step": 112
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019880042032037947,
+      "loss": 0.7404,
+      "step": 113
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0001987770622617072,
+      "loss": 0.7017,
+      "step": 114
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00019875348037686106,
+      "loss": 0.8272,
+      "step": 115
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019872967471927692,
+      "loss": 0.8391,
+      "step": 116
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019870564534289783,
+      "loss": 0.6695,
+      "step": 117
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0001986813923021737,
+      "loss": 0.7992,
+      "step": 118
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019865691565206122,
+      "loss": 0.7508,
+      "step": 119
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019863221544802386,
+      "loss": 0.7509,
+      "step": 120
+    },
+    {
+      "epoch": 0.17,
+      "eval_loss": 0.6245399713516235,
+      "eval_runtime": 29.3529,
+      "eval_samples_per_second": 55.497,
+      "eval_steps_per_second": 27.766,
+      "step": 120
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019860729174603163,
+      "loss": 0.6813,
+      "step": 121
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00019858214460256095,
+      "loss": 0.6432,
+      "step": 122
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019855677407459458,
+      "loss": 0.672,
+      "step": 123
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019853118021962148,
+      "loss": 0.7951,
+      "step": 124
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019850536309563656,
+      "loss": 0.7095,
+      "step": 125
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019847932276114083,
+      "loss": 0.6037,
+      "step": 126
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019845305927514094,
+      "loss": 0.5812,
+      "step": 127
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00019842657269714923,
+      "loss": 0.6124,
+      "step": 128
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0001983998630871836,
+      "loss": 0.6501,
+      "step": 129
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0001983729305057673,
+      "loss": 0.9065,
+      "step": 130
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00019834577501392885,
+      "loss": 0.8046,
+      "step": 131
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00019831839667320183,
+      "loss": 0.904,
+      "step": 132
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00019829079554562487,
+      "loss": 0.5364,
+      "step": 133
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0001982629716937414,
+      "loss": 0.4704,
+      "step": 134
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00019823492518059946,
+      "loss": 0.6457,
+      "step": 135
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00019820665606975175,
+      "loss": 0.752,
+      "step": 136
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019817816442525526,
+      "loss": 0.7523,
+      "step": 137
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019814945031167134,
+      "loss": 0.675,
+      "step": 138
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0001981205137940654,
+      "loss": 0.7782,
+      "step": 139
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019809135493800679,
+      "loss": 0.6174,
+      "step": 140
+    },
+    {
+      "epoch": 0.2,
+      "eval_loss": 0.6312826871871948,
+      "eval_runtime": 29.36,
+      "eval_samples_per_second": 55.484,
+      "eval_steps_per_second": 27.759,
+      "step": 140
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019806197380956874,
+      "loss": 0.5355,
+      "step": 141
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019803237047532802,
+      "loss": 0.5741,
+      "step": 142
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019800254500236503,
+      "loss": 0.7357,
+      "step": 143
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00019797249745826353,
+      "loss": 0.6641,
+      "step": 144
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0001979422279111104,
+      "loss": 0.5884,
+      "step": 145
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00019791173642949564,
+      "loss": 0.4902,
+      "step": 146
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00019788102308251212,
+      "loss": 0.6585,
+      "step": 147
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00019785008793975548,
+      "loss": 0.6445,
+      "step": 148
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00019781893107132393,
+      "loss": 0.8865,
+      "step": 149
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00019778755254781807,
+      "loss": 0.6996,
+      "step": 150
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00019775595244034077,
+      "loss": 0.5696,
+      "step": 151
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00019772413082049713,
+      "loss": 0.6353,
+      "step": 152
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00019769208776039397,
+      "loss": 0.6547,
+      "step": 153
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00019765982333264006,
+      "loss": 1.059,
+      "step": 154
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0001976273376103457,
+      "loss": 0.6234,
+      "step": 155
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0001975946306671227,
+      "loss": 0.5276,
+      "step": 156
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00019756170257708412,
+      "loss": 0.6942,
+      "step": 157
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0001975285534148441,
+      "loss": 0.7216,
+      "step": 158
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019749518325551778,
+      "loss": 0.6563,
+      "step": 159
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019746159217472097,
+      "loss": 0.7549,
+      "step": 160
+    },
+    {
+      "epoch": 0.23,
+      "eval_loss": 0.6180239319801331,
+      "eval_runtime": 29.3051,
+      "eval_samples_per_second": 55.588,
+      "eval_steps_per_second": 27.811,
+      "step": 160
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019742778024857028,
+      "loss": 0.6399,
+      "step": 161
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019739374755368253,
+      "loss": 0.6149,
+      "step": 162
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019735949416717493,
+      "loss": 0.5183,
+      "step": 163
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00019732502016666476,
+      "loss": 0.5566,
+      "step": 164
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00019729032563026914,
+      "loss": 0.7261,
+      "step": 165
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00019725541063660498,
+      "loss": 0.7432,
+      "step": 166
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001972202752647887,
+      "loss": 0.6235,
+      "step": 167
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00019718491959443615,
+      "loss": 0.8306,
+      "step": 168
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00019714934370566227,
+      "loss": 0.7739,
+      "step": 169
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00019711354767908113,
+      "loss": 0.5514,
+      "step": 170
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001970775315958055,
+      "loss": 0.6287,
+      "step": 171
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00019704129553744696,
+      "loss": 0.5172,
+      "step": 172
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00019700483958611534,
+      "loss": 0.5628,
+      "step": 173
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0001969681638244189,
+      "loss": 0.5371,
+      "step": 174
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00019693126833546392,
+      "loss": 0.5714,
+      "step": 175
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0001968941532028546,
+      "loss": 0.7978,
+      "step": 176
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00019685681851069285,
+      "loss": 0.7627,
+      "step": 177
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00019681926434357802,
+      "loss": 0.5672,
+      "step": 178
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00019678149078660692,
+      "loss": 0.639,
+      "step": 179
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00019674349792537336,
+      "loss": 0.6015,
+      "step": 180
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.616681694984436,
+      "eval_runtime": 29.315,
+      "eval_samples_per_second": 55.569,
+      "eval_steps_per_second": 27.801,
+      "step": 180
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0001967052858459682,
+      "loss": 0.564,
+      "step": 181
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0001966668546349789,
+      "loss": 0.6838,
+      "step": 182
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00019662820437948967,
+      "loss": 0.6492,
+      "step": 183
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00019658933516708085,
+      "loss": 0.7017,
+      "step": 184
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00019655024708582904,
+      "loss": 0.7092,
+      "step": 185
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00019651094022430686,
+      "loss": 0.4604,
+      "step": 186
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0001964714146715826,
+      "loss": 0.6142,
+      "step": 187
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00019643167051722,
+      "loss": 0.6881,
+      "step": 188
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00019639170785127835,
+      "loss": 0.5784,
+      "step": 189
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0001963515267643119,
+      "loss": 0.8617,
+      "step": 190
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00019631112734737,
+      "loss": 0.7697,
+      "step": 191
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00019627050969199655,
+      "loss": 0.5587,
+      "step": 192
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019622967389023015,
+      "loss": 0.5915,
+      "step": 193
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0001961886200346036,
+      "loss": 0.9243,
+      "step": 194
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019614734821814383,
+      "loss": 0.7143,
+      "step": 195
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019610585853437165,
+      "loss": 0.9429,
+      "step": 196
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019606415107730165,
+      "loss": 0.5176,
+      "step": 197
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0001960222259414417,
+      "loss": 0.5437,
+      "step": 198
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019598008322179312,
+      "loss": 0.6687,
+      "step": 199
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00019593772301385016,
+      "loss": 0.716,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "eval_loss": 0.6165403127670288,
+      "eval_runtime": 29.3359,
+      "eval_samples_per_second": 55.529,
+      "eval_steps_per_second": 27.782,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0001958951454135999,
+      "loss": 0.6579,
+      "step": 201
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.000195852350517522,
+      "loss": 0.6,
+      "step": 202
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00019580933842258867,
+      "loss": 0.6689,
+      "step": 203
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00019576610922626402,
+      "loss": 0.679,
+      "step": 204
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00019572266302650434,
+      "loss": 0.8533,
+      "step": 205
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00019567899992175753,
+      "loss": 0.5989,
+      "step": 206
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.000195635120010963,
+      "loss": 0.5986,
+      "step": 207
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00019559102339355148,
+      "loss": 0.7082,
+      "step": 208
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00019554671016944474,
+      "loss": 0.6403,
+      "step": 209
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00019550218043905526,
+      "loss": 0.5886,
+      "step": 210
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00019545743430328632,
+      "loss": 0.5391,
+      "step": 211
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0001954124718635314,
+      "loss": 0.6303,
+      "step": 212
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0001953672932216742,
+      "loss": 0.7248,
+      "step": 213
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00019532189848008833,
+      "loss": 0.6166,
+      "step": 214
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00019527628774163705,
+      "loss": 0.5582,
+      "step": 215
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00019523046110967305,
+      "loss": 0.6682,
+      "step": 216
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00019518441868803828,
+      "loss": 0.6317,
+      "step": 217
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0001951381605810636,
+      "loss": 0.7702,
+      "step": 218
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00019509168689356866,
+      "loss": 0.584,
+      "step": 219
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0001950449977308616,
+      "loss": 0.6304,
+      "step": 220
+    },
+    {
+      "epoch": 0.31,
+      "eval_loss": 0.6014450192451477,
+      "eval_runtime": 29.301,
+      "eval_samples_per_second": 55.595,
+      "eval_steps_per_second": 27.815,
+      "step": 220
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019499809319873873,
+      "loss": 0.5839,
+      "step": 221
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019495097340348458,
+      "loss": 0.589,
+      "step": 222
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019490363845187125,
+      "loss": 0.8189,
+      "step": 223
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019485608845115854,
+      "loss": 0.5217,
+      "step": 224
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019480832350909344,
+      "loss": 0.6624,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019476034373391005,
+      "loss": 0.9488,
+      "step": 226
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0001947121492343292,
+      "loss": 0.7354,
+      "step": 227
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0001946637401195584,
+      "loss": 0.6767,
+      "step": 228
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00019461511649929137,
+      "loss": 0.6337,
+      "step": 229
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00019456627848370793,
+      "loss": 0.436,
+      "step": 230
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0001945172261834737,
+      "loss": 0.4473,
+      "step": 231
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00019446795970973993,
+      "loss": 0.5208,
+      "step": 232
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00019441847917414307,
+      "loss": 0.5907,
+      "step": 233
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0001943687846888047,
+      "loss": 0.6171,
+      "step": 234
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019431887636633125,
+      "loss": 0.7125,
+      "step": 235
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019426875431981355,
+      "loss": 0.46,
+      "step": 236
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019421841866282686,
+      "loss": 0.71,
+      "step": 237
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019416786950943044,
+      "loss": 0.6707,
+      "step": 238
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019411710697416726,
+      "loss": 0.4639,
+      "step": 239
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019406613117206397,
+      "loss": 0.5781,
+      "step": 240
+    },
+    {
+      "epoch": 0.34,
+      "eval_loss": 0.6106613874435425,
+      "eval_runtime": 29.3327,
+      "eval_samples_per_second": 55.535,
+      "eval_steps_per_second": 27.785,
+      "step": 240
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00019401494221863024,
+      "loss": 0.5778,
+      "step": 241
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00019396354022985896,
+      "loss": 0.6418,
+      "step": 242
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0001939119253222256,
+      "loss": 0.4679,
+      "step": 243
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00019386009761268821,
+      "loss": 0.6215,
+      "step": 244
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00019380805721868694,
+      "loss": 0.5621,
+      "step": 245
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00019375580425814396,
+      "loss": 0.5727,
+      "step": 246
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00019370333884946307,
+      "loss": 0.7134,
+      "step": 247
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0001936506611115295,
+      "loss": 0.6081,
+      "step": 248
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019359777116370955,
+      "loss": 0.9212,
+      "step": 249
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.0001935446691258504,
+      "loss": 0.5593,
+      "step": 250
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019349135511827995,
+      "loss": 0.6382,
+      "step": 251
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.0001934378292618062,
+      "loss": 0.6483,
+      "step": 252
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019338409167771734,
+      "loss": 0.5854,
+      "step": 253
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019333014248778133,
+      "loss": 0.6317,
+      "step": 254
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019327598181424557,
+      "loss": 0.556,
+      "step": 255
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0001932216097798367,
+      "loss": 0.7319,
+      "step": 256
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0001931670265077602,
+      "loss": 0.6195,
+      "step": 257
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00019311223212170045,
+      "loss": 0.5316,
+      "step": 258
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00019305722674581996,
+      "loss": 0.753,
+      "step": 259
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00019300201050475948,
+      "loss": 0.8,
+      "step": 260
+    },
+    {
+      "epoch": 0.37,
+      "eval_loss": 0.594850480556488,
+      "eval_runtime": 29.3304,
+      "eval_samples_per_second": 55.54,
+      "eval_steps_per_second": 27.787,
+      "step": 260
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0001929465835236375,
+      "loss": 0.5674,
+      "step": 261
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00019289094592805011,
+      "loss": 0.5302,
+      "step": 262
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019283509784407058,
+      "loss": 0.621,
+      "step": 263
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019277903939824914,
+      "loss": 0.6329,
+      "step": 264
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019272277071761282,
+      "loss": 0.5637,
+      "step": 265
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019266629192966485,
+      "loss": 0.5367,
+      "step": 266
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019260960316238467,
+      "loss": 0.6705,
+      "step": 267
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019255270454422756,
+      "loss": 0.7982,
+      "step": 268
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00019249559620412418,
+      "loss": 0.5728,
+      "step": 269
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019243827827148055,
+      "loss": 0.5841,
+      "step": 270
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019238075087617759,
+      "loss": 0.7056,
+      "step": 271
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019232301414857074,
+      "loss": 0.6554,
+      "step": 272
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019226506821948998,
+      "loss": 0.6277,
+      "step": 273
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019220691322023917,
+      "loss": 0.5815,
+      "step": 274
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019214854928259603,
+      "loss": 0.722,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00019208997653881164,
+      "loss": 0.6855,
+      "step": 276
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019203119512161023,
+      "loss": 0.532,
+      "step": 277
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019197220516418902,
+      "loss": 0.5008,
+      "step": 278
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019191300680021755,
+      "loss": 0.5297,
+      "step": 279
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0001918536001638378,
+      "loss": 0.6845,
+      "step": 280
+    },
+    {
+      "epoch": 0.4,
+      "eval_loss": 0.595288097858429,
+      "eval_runtime": 29.2376,
+      "eval_samples_per_second": 55.716,
+      "eval_steps_per_second": 27.875,
+      "step": 280
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019179398538966358,
+      "loss": 0.6998,
+      "step": 281
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00019173416261278044,
+      "loss": 0.8184,
+      "step": 282
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0001916741319687451,
+      "loss": 0.4835,
+      "step": 283
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0001916138935935854,
+      "loss": 0.7273,
+      "step": 284
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019155344762379994,
+      "loss": 0.854,
+      "step": 285
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0001914927941963576,
+      "loss": 0.6895,
+      "step": 286
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0001914319334486975,
+      "loss": 0.6412,
+      "step": 287
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0001913708655187284,
+      "loss": 0.5484,
+      "step": 288
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019130959054482858,
+      "loss": 0.5316,
+      "step": 289
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019124810866584554,
+      "loss": 0.5696,
+      "step": 290
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019118642002109552,
+      "loss": 0.6214,
+      "step": 291
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019112452475036337,
+      "loss": 0.5533,
+      "step": 292
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019106242299390212,
+      "loss": 0.4784,
+      "step": 293
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019100011489243263,
+      "loss": 0.856,
+      "step": 294
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019093760058714346,
+      "loss": 0.6587,
+      "step": 295
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019087488021969027,
+      "loss": 0.6849,
+      "step": 296
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0001908119539321958,
+      "loss": 0.5236,
+      "step": 297
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019074882186724928,
+      "loss": 0.584,
+      "step": 298
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0001906854841679063,
+      "loss": 0.5225,
+      "step": 299
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0001906219409776884,
+      "loss": 0.5857,
+      "step": 300
+    },
+    {
+      "epoch": 0.43,
+      "eval_loss": 0.5940439701080322,
+      "eval_runtime": 29.3378,
+      "eval_samples_per_second": 55.526,
+      "eval_steps_per_second": 27.78,
+      "step": 300
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019055819244058272,
+      "loss": 0.9588,
+      "step": 301
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019049423870104174,
+      "loss": 0.5345,
+      "step": 302
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019043007990398293,
+      "loss": 0.827,
+      "step": 303
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0001903657161947884,
+      "loss": 0.53,
+      "step": 304
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001903011477193046,
+      "loss": 0.469,
+      "step": 305
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00019023637462384194,
+      "loss": 0.6566,
+      "step": 306
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00019017139705517454,
+      "loss": 0.5136,
+      "step": 307
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00019010621516053977,
+      "loss": 0.6217,
+      "step": 308
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00019004082908763813,
+      "loss": 0.5953,
+      "step": 309
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018997523898463267,
+      "loss": 0.6747,
+      "step": 310
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018990944500014883,
+      "loss": 0.5683,
+      "step": 311
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018984344728327395,
+      "loss": 0.8048,
+      "step": 312
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018977724598355717,
+      "loss": 0.5998,
+      "step": 313
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018971084125100882,
+      "loss": 0.6533,
+      "step": 314
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018964423323610026,
+      "loss": 0.5986,
+      "step": 315
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018957742208976344,
+      "loss": 0.5504,
+      "step": 316
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0001895104079633906,
+      "loss": 0.6991,
+      "step": 317
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018944319100883404,
+      "loss": 0.566,
+      "step": 318
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001893757713784055,
+      "loss": 0.5838,
+      "step": 319
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00018930814922487607,
+      "loss": 0.6369,
+      "step": 320
+    },
+    {
+      "epoch": 0.46,
+      "eval_loss": 0.5889019966125488,
+      "eval_runtime": 29.353,
+      "eval_samples_per_second": 55.497,
+      "eval_steps_per_second": 27.765,
+      "step": 320
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00018924032470147575,
+      "loss": 0.6911,
+      "step": 321
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001891722979618931,
+      "loss": 0.839,
+      "step": 322
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001891040691602749,
+      "loss": 0.664,
+      "step": 323
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001890356384512257,
+      "loss": 0.7154,
+      "step": 324
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00018896700598980775,
+      "loss": 0.6249,
+      "step": 325
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001888981719315403,
+      "loss": 0.6035,
+      "step": 326
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00018882913643239953,
+      "loss": 0.542,
+      "step": 327
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00018875989964881797,
+      "loss": 0.7356,
+      "step": 328
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001886904617376844,
+      "loss": 0.5376,
+      "step": 329
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001886208228563432,
+      "loss": 0.5325,
+      "step": 330
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001885509831625942,
+      "loss": 0.4952,
+      "step": 331
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001884809428146923,
+      "loss": 0.5887,
+      "step": 332
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018841070197134706,
+      "loss": 0.5159,
+      "step": 333
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018834026079172237,
+      "loss": 0.6428,
+      "step": 334
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018826961943543594,
+      "loss": 0.624,
+      "step": 335
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018819877806255933,
+      "loss": 0.7188,
+      "step": 336
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018812773683361708,
+      "loss": 0.4654,
+      "step": 337
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018805649590958678,
+      "loss": 0.7584,
+      "step": 338
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018798505545189844,
+      "loss": 0.5568,
+      "step": 339
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00018791341562243418,
+      "loss": 0.4767,
+      "step": 340
+    },
+    {
+      "epoch": 0.49,
+      "eval_loss": 0.5945929288864136,
+      "eval_runtime": 29.3121,
+      "eval_samples_per_second": 55.574,
+      "eval_steps_per_second": 27.804,
+      "step": 340
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.000187841576583528,
+      "loss": 0.6751,
+      "step": 341
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00018776953849796514,
+      "loss": 0.4626,
+      "step": 342
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00018769730152898208,
+      "loss": 0.7323,
+      "step": 343
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00018762486584026578,
+      "loss": 0.4619,
+      "step": 344
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0001875522315959536,
+      "loss": 0.6054,
+      "step": 345
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00018747939896063276,
+      "loss": 0.6465,
+      "step": 346
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00018740636809934009,
+      "loss": 0.8281,
+      "step": 347
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00018733313917756154,
+      "loss": 0.6565,
+      "step": 348
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0001872597123612319,
+      "loss": 0.6819,
+      "step": 349
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0001871860878167344,
+      "loss": 0.7214,
+      "step": 350
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0001871122657109002,
+      "loss": 0.6414,
+      "step": 351
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00018703824621100825,
+      "loss": 0.3563,
+      "step": 352
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018696402948478475,
+      "loss": 0.6458,
+      "step": 353
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018688961570040283,
+      "loss": 0.4847,
+      "step": 354
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018681500502648214,
+      "loss": 0.6457,
+      "step": 355
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018674019763208842,
+      "loss": 0.7152,
+      "step": 356
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018666519368673324,
+      "loss": 0.8583,
+      "step": 357
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018658999336037356,
+      "loss": 0.7007,
+      "step": 358
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00018651459682341126,
+      "loss": 0.5519,
+      "step": 359
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018643900424669286,
+      "loss": 0.4848,
+      "step": 360
+    },
+    {
+      "epoch": 0.52,
+      "eval_loss": 0.5990561842918396,
+      "eval_runtime": 29.3397,
+      "eval_samples_per_second": 55.522,
+      "eval_steps_per_second": 27.778,
+      "step": 360
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018636321580150917,
+      "loss": 0.5851,
+      "step": 361
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0001862872316595947,
+      "loss": 0.5745,
+      "step": 362
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018621105199312753,
+      "loss": 0.5597,
+      "step": 363
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018613467697472876,
+      "loss": 0.6364,
+      "step": 364
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0001860581067774621,
+      "loss": 0.5668,
+      "step": 365
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018598134157483354,
+      "loss": 0.7182,
+      "step": 366
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.000185904381540791,
+      "loss": 0.5478,
+      "step": 367
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00018582722684972383,
+      "loss": 0.485,
+      "step": 368
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00018574987767646254,
+      "loss": 0.7902,
+      "step": 369
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00018567233419627824,
+      "loss": 0.6895,
+      "step": 370
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00018559459658488238,
+      "loss": 0.6304,
+      "step": 371
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00018551666501842636,
+      "loss": 0.4389,
+      "step": 372
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001854385396735009,
+      "loss": 0.7157,
+      "step": 373
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018536022072713606,
+      "loss": 0.552,
+      "step": 374
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018528170835680036,
+      "loss": 0.5701,
+      "step": 375
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018520300274040084,
+      "loss": 0.7629,
+      "step": 376
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018512410405628225,
+      "loss": 0.4954,
+      "step": 377
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018504501248322686,
+      "loss": 1.0569,
+      "step": 378
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018496572820045413,
+      "loss": 0.5473,
+      "step": 379
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00018488625138762008,
+      "loss": 0.9067,
+      "step": 380
+    },
+    {
+      "epoch": 0.54,
+      "eval_loss": 0.5943491458892822,
+      "eval_runtime": 29.3323,
+      "eval_samples_per_second": 55.536,
+      "eval_steps_per_second": 27.785,
+      "step": 380
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00018480658222481703,
+      "loss": 0.4678,
+      "step": 381
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.0001847267208925732,
+      "loss": 0.7051,
+      "step": 382
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00018464666757185216,
+      "loss": 0.3741,
+      "step": 383
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00018456642244405266,
+      "loss": 0.6583,
+      "step": 384
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.000184485985691008,
+      "loss": 0.4571,
+      "step": 385
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00018440535749498563,
+      "loss": 0.4962,
+      "step": 386
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00018432453803868696,
+      "loss": 0.6971,
+      "step": 387
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018424352750524668,
+      "loss": 0.7043,
+      "step": 388
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0001841623260782325,
+      "loss": 0.5788,
+      "step": 389
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018408093394164468,
+      "loss": 0.6969,
+      "step": 390
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018399935127991554,
+      "loss": 0.6253,
+      "step": 391
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018391757827790933,
+      "loss": 0.6763,
+      "step": 392
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018383561512092138,
+      "loss": 0.6583,
+      "step": 393
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018375346199467807,
+      "loss": 0.5904,
+      "step": 394
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00018367111908533615,
+      "loss": 0.8813,
+      "step": 395
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0001835885865794824,
+      "loss": 0.7437,
+      "step": 396
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00018350586466413336,
+      "loss": 0.6893,
+      "step": 397
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00018342295352673463,
+      "loss": 0.5746,
+      "step": 398
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00018333985335516057,
+      "loss": 0.7338,
+      "step": 399
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.000183256564337714,
+      "loss": 0.5943,
+      "step": 400
+    },
+    {
+      "epoch": 0.57,
+      "eval_loss": 0.5853814482688904,
+      "eval_runtime": 29.3481,
+      "eval_samples_per_second": 55.506,
+      "eval_steps_per_second": 27.77,
+      "step": 400
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0001831730866631256,
+      "loss": 0.631,
+      "step": 401
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00018308942052055354,
+      "loss": 0.5297,
+      "step": 402
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00018300556609958304,
+      "loss": 0.6393,
+      "step": 403
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00018292152359022595,
+      "loss": 0.4876,
+      "step": 404
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00018283729318292036,
+      "loss": 0.5634,
+      "step": 405
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00018275287506853017,
+      "loss": 0.4768,
+      "step": 406
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00018266826943834445,
+      "loss": 0.9219,
+      "step": 407
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0001825834764840774,
+      "loss": 0.5326,
+      "step": 408
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00018249849639786749,
+      "loss": 0.717,
+      "step": 409
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00018241332937227734,
+      "loss": 0.7521,
+      "step": 410
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0001823279756002932,
+      "loss": 0.8977,
+      "step": 411
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0001822424352753244,
+      "loss": 0.7813,
+      "step": 412
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.000182156708591203,
+      "loss": 0.8893,
+      "step": 413
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00018207079574218338,
+      "loss": 0.6064,
+      "step": 414
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00018198469692294174,
+      "loss": 0.8171,
+      "step": 415
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018189841232857571,
+      "loss": 0.6169,
+      "step": 416
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018181194215460388,
+      "loss": 0.5828,
+      "step": 417
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018172528659696533,
+      "loss": 0.5173,
+      "step": 418
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0001816384458520192,
+      "loss": 0.5217,
+      "step": 419
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018155142011654435,
+      "loss": 0.6999,
+      "step": 420
+    },
+    {
+      "epoch": 0.6,
+      "eval_loss": 0.5941495299339294,
+      "eval_runtime": 29.3194,
+      "eval_samples_per_second": 55.561,
+      "eval_steps_per_second": 27.797,
+      "step": 420
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0001814642095877387,
+      "loss": 0.5381,
+      "step": 421
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018137681446321903,
+      "loss": 0.7032,
+      "step": 422
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00018128923494102028,
+      "loss": 0.5898,
+      "step": 423
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00018120147121959536,
+      "loss": 0.4766,
+      "step": 424
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0001811135234978145,
+      "loss": 0.5916,
+      "step": 425
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00018102539197496482,
+      "loss": 0.5091,
+      "step": 426
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00018093707685075004,
+      "loss": 0.7035,
+      "step": 427
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00018084857832528987,
+      "loss": 0.7507,
+      "step": 428
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00018075989659911956,
+      "loss": 0.4905,
+      "step": 429
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00018067103187318956,
+      "loss": 0.5706,
+      "step": 430
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00018058198434886492,
+      "loss": 0.784,
+      "step": 431
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00018049275422792497,
+      "loss": 0.5877,
+      "step": 432
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00018040334171256277,
+      "loss": 0.6392,
+      "step": 433
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00018031374700538467,
+      "loss": 0.5623,
+      "step": 434
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00018022397030940986,
+      "loss": 0.6878,
+      "step": 435
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00018013401182806994,
+      "loss": 0.5297,
+      "step": 436
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00018004387176520843,
+      "loss": 0.4828,
+      "step": 437
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00017995355032508027,
+      "loss": 0.5836,
+      "step": 438
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00017986304771235143,
+      "loss": 0.5292,
+      "step": 439
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.0001797723641320984,
+      "loss": 0.5173,
+      "step": 440
+    },
+    {
+      "epoch": 0.63,
+      "eval_loss": 0.5887213349342346,
+      "eval_runtime": 29.3237,
+      "eval_samples_per_second": 55.552,
+      "eval_steps_per_second": 27.793,
+      "step": 440
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00017968149978980774,
+      "loss": 0.6018,
+      "step": 441
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00017959045489137566,
+      "loss": 0.5353,
+      "step": 442
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00017949922964310738,
+      "loss": 0.7697,
+      "step": 443
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00017940782425171693,
+      "loss": 0.4775,
+      "step": 444
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0001793162389243264,
+      "loss": 0.6834,
+      "step": 445
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00017922447386846578,
+      "loss": 0.8259,
+      "step": 446
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00017913252929207217,
+      "loss": 0.7376,
+      "step": 447
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0001790404054034895,
+      "loss": 0.7481,
+      "step": 448
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0001789481024114681,
+      "loss": 0.8861,
+      "step": 449
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00017885562052516398,
+      "loss": 0.4938,
+      "step": 450
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017876295995413867,
+      "loss": 0.5681,
+      "step": 451
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017867012090835854,
+      "loss": 0.4892,
+      "step": 452
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017857710359819436,
+      "loss": 0.6888,
+      "step": 453
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017848390823442086,
+      "loss": 0.5947,
+      "step": 454
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0001783905350282163,
+      "loss": 0.7089,
+      "step": 455
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017829698419116177,
+      "loss": 0.5473,
+      "step": 456
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00017820325593524098,
+      "loss": 0.6345,
+      "step": 457
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00017810935047283966,
+      "loss": 0.6343,
+      "step": 458
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00017801526801674506,
+      "loss": 0.5587,
+      "step": 459
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00017792100878014552,
+      "loss": 0.4201,
+      "step": 460
+    },
+    {
+      "epoch": 0.66,
+      "eval_loss": 0.5952056646347046,
+      "eval_runtime": 29.408,
+      "eval_samples_per_second": 55.393,
+      "eval_steps_per_second": 27.714,
+      "step": 460
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00017782657297662992,
+      "loss": 0.6726,
+      "step": 461
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00017773196082018728,
+      "loss": 0.5181,
+      "step": 462
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0001776371725252062,
+      "loss": 0.4734,
+      "step": 463
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0001775422083064744,
+      "loss": 0.5406,
+      "step": 464
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00017744706837917828,
+      "loss": 0.4941,
+      "step": 465
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00017735175295890233,
+      "loss": 0.5945,
+      "step": 466
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00017725626226162874,
+      "loss": 0.5451,
+      "step": 467
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0001771605965037369,
+      "loss": 0.6012,
+      "step": 468
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00017706475590200285,
+      "loss": 0.6645,
+      "step": 469
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0001769687406735988,
+      "loss": 0.8397,
+      "step": 470
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00017687255103609266,
+      "loss": 0.4745,
+      "step": 471
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017677618720744764,
+      "loss": 0.9309,
+      "step": 472
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0001766796494060215,
+      "loss": 0.5752,
+      "step": 473
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017658293785056638,
+      "loss": 0.846,
+      "step": 474
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017648605276022808,
+      "loss": 0.4726,
+      "step": 475
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017638899435454554,
+      "loss": 0.5713,
+      "step": 476
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0001762917628534506,
+      "loss": 0.5512,
+      "step": 477
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017619435847726712,
+      "loss": 0.538,
+      "step": 478
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017609678144671093,
+      "loss": 0.5254,
+      "step": 479
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017599903198288886,
+      "loss": 0.667,
+      "step": 480
+    },
+    {
+      "epoch": 0.69,
+      "eval_loss": 0.5801939964294434,
+      "eval_runtime": 29.3619,
+      "eval_samples_per_second": 55.48,
+      "eval_steps_per_second": 27.757,
+      "step": 480
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017590111030729862,
+      "loss": 0.8356,
+      "step": 481
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017580301664182812,
+      "loss": 0.6087,
+      "step": 482
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017570475120875495,
+      "loss": 0.5965,
+      "step": 483
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017560631423074591,
+      "loss": 0.5231,
+      "step": 484
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00017550770593085662,
+      "loss": 0.595,
+      "step": 485
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0001754089265325308,
+      "loss": 0.5015,
+      "step": 486
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00017530997625959992,
+      "loss": 0.4863,
+      "step": 487
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00017521085533628265,
+      "loss": 0.7477,
+      "step": 488
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00017511156398718439,
+      "loss": 0.5633,
+      "step": 489
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0001750121024372966,
+      "loss": 0.6316,
+      "step": 490
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0001749124709119965,
+      "loss": 0.4627,
+      "step": 491
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0001748126696370465,
+      "loss": 0.662,
+      "step": 492
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017471269883859356,
+      "loss": 0.6426,
+      "step": 493
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017461255874316885,
+      "loss": 0.6185,
+      "step": 494
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017451224957768717,
+      "loss": 0.6106,
+      "step": 495
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017441177156944635,
+      "loss": 0.7092,
+      "step": 496
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0001743111249461269,
+      "loss": 0.816,
+      "step": 497
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017421030993579133,
+      "loss": 0.8185,
+      "step": 498
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00017410932676688377,
+      "loss": 0.4779,
+      "step": 499
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017400817566822938,
+      "loss": 0.8568,
+      "step": 500
+    },
+    {
+      "epoch": 0.72,
+      "eval_loss": 0.5921781063079834,
+      "eval_runtime": 29.3482,
+      "eval_samples_per_second": 55.506,
+      "eval_steps_per_second": 27.77,
+      "step": 500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0001739068568690338,
+      "loss": 0.6461,
+      "step": 501
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017380537059888277,
+      "loss": 0.6208,
+      "step": 502
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017370371708774141,
+      "loss": 0.5665,
+      "step": 503
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0001736018965659539,
+      "loss": 0.5903,
+      "step": 504
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017349990926424273,
+      "loss": 0.5639,
+      "step": 505
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0001733977554137085,
+      "loss": 0.5138,
+      "step": 506
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00017329543524582898,
+      "loss": 0.5628,
+      "step": 507
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.000173192948992459,
+      "loss": 0.3835,
+      "step": 508
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00017309029688582963,
+      "loss": 0.6234,
+      "step": 509
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00017298747915854782,
+      "loss": 0.86,
+      "step": 510
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00017288449604359576,
+      "loss": 0.7151,
+      "step": 511
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00017278134777433044,
+      "loss": 0.554,
+      "step": 512
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00017267803458448307,
+      "loss": 0.583,
+      "step": 513
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017257455670815854,
+      "loss": 0.5642,
+      "step": 514
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017247091437983496,
+      "loss": 1.0091,
+      "step": 515
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017236710783436306,
+      "loss": 0.5834,
+      "step": 516
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017226313730696574,
+      "loss": 0.6862,
+      "step": 517
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017215900303323736,
+      "loss": 0.4787,
+      "step": 518
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017205470524914342,
+      "loss": 0.4951,
+      "step": 519
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00017195024419101987,
+      "loss": 0.515,
+      "step": 520
+    },
+    {
+      "epoch": 0.74,
+      "eval_loss": 0.5800089836120605,
+      "eval_runtime": 29.3282,
+      "eval_samples_per_second": 55.544,
+      "eval_steps_per_second": 27.789,
+      "step": 520
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00017184562009557271,
+      "loss": 0.6123,
+      "step": 521
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00017174083319987732,
+      "loss": 0.6898,
+      "step": 522
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.000171635883741378,
+      "loss": 0.487,
+      "step": 523
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0001715307719578874,
+      "loss": 0.6113,
+      "step": 524
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.000171425498087586,
+      "loss": 0.4276,
+      "step": 525
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00017132006236902155,
+      "loss": 0.5996,
+      "step": 526
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00017121446504110859,
+      "loss": 0.564,
+      "step": 527
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00017110870634312784,
+      "loss": 0.8961,
+      "step": 528
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00017100278651472562,
+      "loss": 0.4929,
+      "step": 529
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0001708967057959135,
+      "loss": 0.5149,
+      "step": 530
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00017079046442706748,
+      "loss": 0.4424,
+      "step": 531
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00017068406264892768,
+      "loss": 1.0212,
+      "step": 532
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00017057750070259765,
+      "loss": 0.4541,
+      "step": 533
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00017047077882954392,
+      "loss": 0.7563,
+      "step": 534
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0001703638972715954,
+      "loss": 0.644,
+      "step": 535
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00017025685627094283,
+      "loss": 0.4545,
+      "step": 536
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00017014965607013824,
+      "loss": 0.61,
+      "step": 537
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0001700422969120944,
+      "loss": 0.6925,
+      "step": 538
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00016993477904008432,
+      "loss": 0.4985,
+      "step": 539
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00016982710269774058,
+      "loss": 0.504,
+      "step": 540
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.5893893241882324,
+      "eval_runtime": 29.3252,
+      "eval_samples_per_second": 55.55,
+      "eval_steps_per_second": 27.792,
+      "step": 540
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00016971926812905488,
+      "loss": 0.3721,
+      "step": 541
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00016961127557837751,
+      "loss": 0.5396,
+      "step": 542
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00016950312529041663,
+      "loss": 0.4731,
+      "step": 543
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0001693948175102379,
+      "loss": 0.5411,
+      "step": 544
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0001692863524832639,
+      "loss": 0.6364,
+      "step": 545
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00016917773045527343,
+      "loss": 0.5857,
+      "step": 546
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00016906895167240112,
+      "loss": 0.7359,
+      "step": 547
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0001689600163811367,
+      "loss": 0.5994,
+      "step": 548
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0001688509248283247,
+      "loss": 0.5892,
+      "step": 549
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00016874167726116363,
+      "loss": 0.7209,
+      "step": 550
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00016863227392720554,
+      "loss": 0.5054,
+      "step": 551
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00016852271507435544,
+      "loss": 0.5956,
+      "step": 552
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00016841300095087077,
+      "loss": 0.5568,
+      "step": 553
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00016830313180536078,
+      "loss": 0.6406,
+      "step": 554
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00016819310788678603,
+      "loss": 0.6066,
+      "step": 555
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016808292944445774,
+      "loss": 0.4708,
+      "step": 556
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0001679725967280373,
+      "loss": 0.6026,
+      "step": 557
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016786210998753575,
+      "loss": 0.8752,
+      "step": 558
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016775146947331298,
+      "loss": 0.7405,
+      "step": 559
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016764067543607753,
+      "loss": 0.6361,
+      "step": 560
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 0.5982840657234192,
+      "eval_runtime": 29.3385,
+      "eval_samples_per_second": 55.524,
+      "eval_steps_per_second": 27.779,
+      "step": 560
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016752972812688564,
+      "loss": 0.6528,
+      "step": 561
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016741862779714098,
+      "loss": 0.5128,
+      "step": 562
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00016730737469859388,
+      "loss": 0.723,
+      "step": 563
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00016719596908334092,
+      "loss": 0.5223,
+      "step": 564
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.0001670844112038242,
+      "loss": 0.5353,
+      "step": 565
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.0001669727013128309,
+      "loss": 0.7645,
+      "step": 566
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00016686083966349266,
+      "loss": 0.6167,
+      "step": 567
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00016674882650928493,
+      "loss": 0.3836,
+      "step": 568
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00016663666210402656,
+      "loss": 0.6045,
+      "step": 569
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00016652434670187907,
+      "loss": 0.5973,
+      "step": 570
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0001664118805573461,
+      "loss": 0.5661,
+      "step": 571
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.000166299263925273,
+      "loss": 0.4528,
+      "step": 572
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00016618649706084596,
+      "loss": 0.6458,
+      "step": 573
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00016607358021959173,
+      "loss": 0.412,
+      "step": 574
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0001659605136573768,
+      "loss": 0.5031,
+      "step": 575
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00016584729763040697,
+      "loss": 0.4902,
+      "step": 576
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00016573393239522678,
+      "loss": 0.8064,
+      "step": 577
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00016562041820871874,
+      "loss": 0.5497,
+      "step": 578
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.000165506755328103,
+      "loss": 0.6602,
+      "step": 579
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00016539294401093658,
+      "loss": 0.4896,
+      "step": 580
+    },
+    {
+      "epoch": 0.83,
+      "eval_loss": 0.5770359039306641,
+      "eval_runtime": 29.3487,
+      "eval_samples_per_second": 55.505,
+      "eval_steps_per_second": 27.77,
+      "step": 580
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00016527898451511287,
+      "loss": 0.3522,
+      "step": 581
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00016516487709886105,
+      "loss": 0.6396,
+      "step": 582
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0001650506220207454,
+      "loss": 0.8431,
+      "step": 583
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016493621953966495,
+      "loss": 0.5146,
+      "step": 584
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016482166991485265,
+      "loss": 0.5748,
+      "step": 585
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016470697340587476,
+      "loss": 0.5961,
+      "step": 586
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016459213027263063,
+      "loss": 0.5165,
+      "step": 587
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016447714077535167,
+      "loss": 0.8974,
+      "step": 588
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.000164362005174601,
+      "loss": 0.839,
+      "step": 589
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00016424672373127277,
+      "loss": 0.5579,
+      "step": 590
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.00016413129670659167,
+      "loss": 0.832,
+      "step": 591
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.00016401572436211222,
+      "loss": 0.7778,
+      "step": 592
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0001639000069597183,
+      "loss": 0.4416,
+      "step": 593
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0001637841447616224,
+      "loss": 0.4954,
+      "step": 594
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0001636681380303652,
+      "loss": 0.487,
+      "step": 595
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.00016355198702881478,
+      "loss": 0.6773,
+      "step": 596
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0001634356920201662,
+      "loss": 0.5598,
+      "step": 597
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016331925326794087,
+      "loss": 0.4758,
+      "step": 598
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016320267103598585,
+      "loss": 0.5392,
+      "step": 599
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016308594558847337,
+      "loss": 0.6044,
+      "step": 600
+    },
+    {
+      "epoch": 0.86,
+      "eval_loss": 0.5716915726661682,
+      "eval_runtime": 29.3528,
+      "eval_samples_per_second": 55.497,
+      "eval_steps_per_second": 27.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016296907718990015,
+      "loss": 0.543,
+      "step": 601
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016285206610508685,
+      "loss": 0.677,
+      "step": 602
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00016273491259917745,
+      "loss": 0.8737,
+      "step": 603
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0001626176169376387,
+      "loss": 0.6001,
+      "step": 604
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0001625001793862593,
+      "loss": 0.6186,
+      "step": 605
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0001623826002111497,
+      "loss": 0.6821,
+      "step": 606
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00016226487967874116,
+      "loss": 0.4421,
+      "step": 607
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00016214701805578518,
+      "loss": 0.5087,
+      "step": 608
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0001620290156093531,
+      "loss": 0.465,
+      "step": 609
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00016191087260683523,
+      "loss": 0.5147,
+      "step": 610
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00016179258931594051,
+      "loss": 0.6099,
+      "step": 611
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0001616741660046957,
+      "loss": 0.6062,
+      "step": 612
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.00016155560294144479,
+      "loss": 0.5735,
+      "step": 613
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.00016143690039484857,
+      "loss": 0.6008,
+      "step": 614
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.00016131805863388378,
+      "loss": 0.6302,
+      "step": 615
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.00016119907792784267,
+      "loss": 0.5624,
+      "step": 616
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.00016107995854633235,
+      "loss": 0.5803,
+      "step": 617
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.00016096070075927415,
+      "loss": 0.5977,
+      "step": 618
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00016084130483690295,
+      "loss": 0.4958,
+      "step": 619
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0001607217710497668,
+      "loss": 0.4925,
+      "step": 620
+    },
+    {
+      "epoch": 0.89,
+      "eval_loss": 0.5715158581733704,
+      "eval_runtime": 29.3023,
+      "eval_samples_per_second": 55.593,
+      "eval_steps_per_second": 27.813,
+      "step": 620
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.000160602099668726,
+      "loss": 0.641,
+      "step": 621
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00016048229096495272,
+      "loss": 0.5944,
+      "step": 622
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00016036234520993024,
+      "loss": 0.655,
+      "step": 623
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0001602422626754524,
+      "loss": 0.3301,
+      "step": 624
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0001601220436336231,
+      "loss": 0.6563,
+      "step": 625
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00016000168835685535,
+      "loss": 0.6173,
+      "step": 626
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00015988119711787105,
+      "loss": 0.7029,
+      "step": 627
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.0001597605701897001,
+      "loss": 0.7588,
+      "step": 628
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00015963980784567986,
+      "loss": 0.527,
+      "step": 629
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00015951891035945464,
+      "loss": 0.6378,
+      "step": 630
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.0001593978780049748,
+      "loss": 0.5056,
+      "step": 631
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00015927671105649648,
+      "loss": 0.6517,
+      "step": 632
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015915540978858066,
+      "loss": 1.0716,
+      "step": 633
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015903397447609288,
+      "loss": 0.5948,
+      "step": 634
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0001589124053942022,
+      "loss": 0.4883,
+      "step": 635
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0001587907028183809,
+      "loss": 0.5172,
+      "step": 636
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015866886702440384,
+      "loss": 0.4618,
+      "step": 637
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015854689828834757,
+      "loss": 0.7591,
+      "step": 638
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015842479688659003,
+      "loss": 0.435,
+      "step": 639
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015830256309580968,
+      "loss": 0.4704,
+      "step": 640
+    },
+    {
+      "epoch": 0.92,
+      "eval_loss": 0.5707182288169861,
+      "eval_runtime": 29.3397,
+      "eval_samples_per_second": 55.522,
+      "eval_steps_per_second": 27.778,
+      "step": 640
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015818019719298504,
+      "loss": 0.6049,
+      "step": 641
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015805769945539394,
+      "loss": 0.5501,
+      "step": 642
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015793507016061305,
+      "loss": 0.7212,
+      "step": 643
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015781230958651694,
+      "loss": 0.6531,
+      "step": 644
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015768941801127783,
+      "loss": 0.9526,
+      "step": 645
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015756639571336476,
+      "loss": 0.5878,
+      "step": 646
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015744324297154293,
+      "loss": 0.46,
+      "step": 647
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015731996006487317,
+      "loss": 0.4893,
+      "step": 648
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015719654727271122,
+      "loss": 0.5149,
+      "step": 649
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015707300487470717,
+      "loss": 0.6435,
+      "step": 650
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015694933315080477,
+      "loss": 0.5271,
+      "step": 651
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015682553238124082,
+      "loss": 0.4587,
+      "step": 652
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00015670160284654458,
+      "loss": 0.4208,
+      "step": 653
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00015657754482753704,
+      "loss": 0.5887,
+      "step": 654
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0001564533586053303,
+      "loss": 0.6264,
+      "step": 655
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00015632904446132706,
+      "loss": 0.6527,
+      "step": 656
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00015620460267721983,
+      "loss": 0.6423,
+      "step": 657
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00015608003353499033,
+      "loss": 0.5313,
+      "step": 658
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0001559553373169089,
+      "loss": 0.58,
+      "step": 659
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00015583051430553385,
+      "loss": 0.5342,
+      "step": 660
+    },
+    {
+      "epoch": 0.94,
+      "eval_loss": 0.5747966766357422,
+      "eval_runtime": 29.3347,
+      "eval_samples_per_second": 55.531,
+      "eval_steps_per_second": 27.783,
+      "step": 660
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00015570556478371075,
+      "loss": 0.6231,
+      "step": 661
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0001555804890345719,
+      "loss": 0.3901,
+      "step": 662
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00015545528734153553,
+      "loss": 0.5303,
+      "step": 663
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0001553299599883054,
+      "loss": 0.5951,
+      "step": 664
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00015520450725886988,
+      "loss": 0.4837,
+      "step": 665
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00015507892943750147,
+      "loss": 0.6,
+      "step": 666
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0001549532268087562,
+      "loss": 0.5726,
+      "step": 667
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00015482739965747282,
+      "loss": 0.5652,
+      "step": 668
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0001547014482687723,
+      "loss": 0.8029,
+      "step": 669
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0001545753729280571,
+      "loss": 0.6563,
+      "step": 670
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00015444917392101054,
+      "loss": 0.5851,
+      "step": 671
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0001543228515335962,
+      "loss": 0.4526,
+      "step": 672
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00015419640605205727,
+      "loss": 0.6139,
+      "step": 673
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0001540698377629157,
+      "loss": 0.5471,
+      "step": 674
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.000153943146952972,
+      "loss": 0.5244,
+      "step": 675
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00015381633390930402,
+      "loss": 0.6256,
+      "step": 676
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.0001536893989192668,
+      "loss": 0.4401,
+      "step": 677
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00015356234227049154,
+      "loss": 0.5637,
+      "step": 678
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00015343516425088524,
+      "loss": 0.5212,
+      "step": 679
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.0001533078651486299,
+      "loss": 0.755,
+      "step": 680
+    },
+    {
+      "epoch": 0.97,
+      "eval_loss": 0.5672922730445862,
+      "eval_runtime": 29.3376,
+      "eval_samples_per_second": 55.526,
+      "eval_steps_per_second": 27.78,
+      "step": 680
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.0001531804452521818,
+      "loss": 0.6678,
+      "step": 681
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00015305290485027114,
+      "loss": 0.413,
+      "step": 682
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00015292524423190094,
+      "loss": 0.7693,
+      "step": 683
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00015279746368634673,
+      "loss": 0.7246,
+      "step": 684
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00015266956350315586,
+      "loss": 0.5489,
+      "step": 685
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.0001525415439721467,
+      "loss": 0.6732,
+      "step": 686
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00015241340538340808,
+      "loss": 0.4387,
+      "step": 687
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.0001522851480272986,
+      "loss": 0.7792,
+      "step": 688
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00015215677219444594,
+      "loss": 0.4398,
+      "step": 689
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.0001520282781757464,
+      "loss": 0.5612,
+      "step": 690
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00015189966626236385,
+      "loss": 0.7103,
+      "step": 691
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.0001517709367457295,
+      "loss": 0.6044,
+      "step": 692
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.000151642089917541,
+      "loss": 0.6622,
+      "step": 693
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.0001515131260697618,
+      "loss": 0.4819,
+      "step": 694
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00015138404549462053,
+      "loss": 0.5547,
+      "step": 695
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015125484848461026,
+      "loss": 0.6227,
+      "step": 696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.000151125535332488,
+      "loss": 0.4646,
+      "step": 697
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015099610633127387,
+      "loss": 0.5327,
+      "step": 698
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015086656177425048,
+      "loss": 0.5109,
+      "step": 699
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0001507369019549624,
+      "loss": 0.6547,
+      "step": 700
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.5720962285995483,
+      "eval_runtime": 29.362,
+      "eval_samples_per_second": 55.48,
+      "eval_steps_per_second": 27.757,
+      "step": 700
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015060712716721518,
+      "loss": 0.6987,
+      "step": 701
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015047723770507515,
+      "loss": 0.6133,
+      "step": 702
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00015034723386286828,
+      "loss": 0.5761,
+      "step": 703
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00015021711593517977,
+      "loss": 0.7156,
+      "step": 704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.0001500868842168534,
+      "loss": 0.5887,
+      "step": 705
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.0001499565390029907,
+      "loss": 0.3749,
+      "step": 706
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00014982608058895046,
+      "loss": 0.8819,
+      "step": 707
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00014969550927034796,
+      "loss": 0.4753,
+      "step": 708
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.0001495648253430543,
+      "loss": 0.7483,
+      "step": 709
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0001494340291031957,
+      "loss": 0.4904,
+      "step": 710
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00014930312084715302,
+      "loss": 0.5602,
+      "step": 711
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0001491721008715608,
+      "loss": 0.5305,
+      "step": 712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0001490409694733068,
+      "loss": 0.5839,
+      "step": 713
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00014890972694953124,
+      "loss": 0.5995,
+      "step": 714
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0001487783735976262,
+      "loss": 0.5653,
+      "step": 715
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0001486469097152348,
+      "loss": 0.7084,
+      "step": 716
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014851533560025074,
+      "loss": 0.4129,
+      "step": 717
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014838365155081737,
+      "loss": 0.6075,
+      "step": 718
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014825185786532724,
+      "loss": 0.5569,
+      "step": 719
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014811995484242127,
+      "loss": 0.6014,
+      "step": 720
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 0.5892384052276611,
+      "eval_runtime": 29.2514,
+      "eval_samples_per_second": 55.69,
+      "eval_steps_per_second": 27.862,
+      "step": 720
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014798794278098824,
+      "loss": 0.4306,
+      "step": 721
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014785582198016393,
+      "loss": 0.5909,
+      "step": 722
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00014772359273933046,
+      "loss": 0.5001,
+      "step": 723
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014759125535811581,
+      "loss": 0.4349,
+      "step": 724
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014745881013639294,
+      "loss": 0.5137,
+      "step": 725
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014732625737427912,
+      "loss": 0.5506,
+      "step": 726
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0001471935973721354,
+      "loss": 0.6122,
+      "step": 727
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014706083043056578,
+      "loss": 0.4386,
+      "step": 728
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014692795685041654,
+      "loss": 0.6179,
+      "step": 729
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0001467949769327757,
+      "loss": 0.502,
+      "step": 730
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00014666189097897218,
+      "loss": 0.5482,
+      "step": 731
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00014652869929057517,
+      "loss": 0.4443,
+      "step": 732
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0001463954021693935,
+      "loss": 0.5305,
+      "step": 733
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00014626199991747482,
+      "loss": 0.4034,
+      "step": 734
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0001461284928371051,
+      "loss": 0.7497,
+      "step": 735
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00014599488123080774,
+      "loss": 0.5743,
+      "step": 736
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0001458611654013431,
+      "loss": 0.6069,
+      "step": 737
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.0001457273456517077,
+      "loss": 0.4714,
+      "step": 738
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00014559342228513342,
+      "loss": 0.6079,
+      "step": 739
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.0001454593956050871,
+      "loss": 0.4692,
+      "step": 740
+    },
+    {
+      "epoch": 1.06,
+      "eval_loss": 0.5981099605560303,
+      "eval_runtime": 29.3369,
+      "eval_samples_per_second": 55.527,
+      "eval_steps_per_second": 27.781,
+      "step": 740
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.0001453252659152695,
+      "loss": 0.3255,
+      "step": 741
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.000145191033519615,
+      "loss": 0.3091,
+      "step": 742
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.0001450566987222906,
+      "loss": 0.4252,
+      "step": 743
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00014492226182769526,
+      "loss": 0.5188,
+      "step": 744
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00014478772314045947,
+      "loss": 0.5744,
+      "step": 745
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00014465308296544423,
+      "loss": 0.4112,
+      "step": 746
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00014451834160774066,
+      "loss": 0.4442,
+      "step": 747
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0001443834993726689,
+      "loss": 0.569,
+      "step": 748
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00014424855656577795,
+      "loss": 0.5295,
+      "step": 749
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00014411351349284462,
+      "loss": 0.5342,
+      "step": 750
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0001439783704598728,
+      "loss": 0.3478,
+      "step": 751
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014384312777309302,
+      "loss": 0.6724,
+      "step": 752
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.0001437077857389616,
+      "loss": 0.49,
+      "step": 753
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.0001435723446641599,
+      "loss": 0.5332,
+      "step": 754
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.0001434368048555939,
+      "loss": 0.4408,
+      "step": 755
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.000143301166620393,
+      "loss": 0.5218,
+      "step": 756
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014316543026590987,
+      "loss": 0.634,
+      "step": 757
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014302959609971947,
+      "loss": 0.5032,
+      "step": 758
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00014289366442961837,
+      "loss": 0.3714,
+      "step": 759
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00014275763556362406,
+      "loss": 0.407,
+      "step": 760
+    },
+    {
+      "epoch": 1.09,
+      "eval_loss": 0.599507749080658,
+      "eval_runtime": 29.31,
+      "eval_samples_per_second": 55.578,
+      "eval_steps_per_second": 27.806,
+      "step": 760
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00014262150980997438,
+      "loss": 0.3594,
+      "step": 761
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00014248528747712658,
+      "loss": 0.4647,
+      "step": 762
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00014234896887375674,
+      "loss": 0.5196,
+      "step": 763
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00014221255430875928,
+      "loss": 0.4629,
+      "step": 764
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00014207604409124589,
+      "loss": 0.3411,
+      "step": 765
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00014193943853054507,
+      "loss": 0.3671,
+      "step": 766
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.0001418027379362014,
+      "loss": 0.5559,
+      "step": 767
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.0001416659426179747,
+      "loss": 0.3496,
+      "step": 768
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00014152905288583947,
+      "loss": 0.6111,
+      "step": 769
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00014139206904998424,
+      "loss": 0.4747,
+      "step": 770
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00014125499142081065,
+      "loss": 0.5567,
+      "step": 771
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00014111782030893301,
+      "loss": 0.4069,
+      "step": 772
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.0001409805560251773,
+      "loss": 0.7049,
+      "step": 773
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00014084319888058077,
+      "loss": 0.4703,
+      "step": 774
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00014070574918639096,
+      "loss": 0.4031,
+      "step": 775
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00014056820725406524,
+      "loss": 0.3365,
+      "step": 776
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.0001404305733952699,
+      "loss": 0.3774,
+      "step": 777
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00014029284792187963,
+      "loss": 0.5119,
+      "step": 778
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.0001401550311459766,
+      "loss": 0.6868,
+      "step": 779
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00014001712337984997,
+      "loss": 0.5351,
+      "step": 780
+    },
+    {
+      "epoch": 1.12,
+      "eval_loss": 0.5948033928871155,
+      "eval_runtime": 29.1497,
+      "eval_samples_per_second": 55.884,
+      "eval_steps_per_second": 27.959,
+      "step": 780
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00013987912493599501,
+      "loss": 0.5557,
+      "step": 781
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00013974103612711253,
+      "loss": 0.57,
+      "step": 782
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00013960285726610802,
+      "loss": 0.4644,
+      "step": 783
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.0001394645886660911,
+      "loss": 0.6286,
+      "step": 784
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00013932623064037475,
+      "loss": 0.4457,
+      "step": 785
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00013918778350247441,
+      "loss": 0.4967,
+      "step": 786
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0001390492475661077,
+      "loss": 0.4602,
+      "step": 787
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00013891062314519336,
+      "loss": 0.5656,
+      "step": 788
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00013877191055385052,
+      "loss": 0.6678,
+      "step": 789
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00013863311010639824,
+      "loss": 0.6069,
+      "step": 790
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00013849422211735458,
+      "loss": 0.4265,
+      "step": 791
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.000138355246901436,
+      "loss": 0.5138,
+      "step": 792
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0001382161847735567,
+      "loss": 0.5329,
+      "step": 793
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00013807703604882762,
+      "loss": 0.501,
+      "step": 794
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.0001379378010425561,
+      "loss": 0.6914,
+      "step": 795
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00013779848007024493,
+      "loss": 0.4193,
+      "step": 796
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00013765907344759164,
+      "loss": 0.4554,
+      "step": 797
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00013751958149048793,
+      "loss": 0.6135,
+      "step": 798
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00013738000451501885,
+      "loss": 0.3686,
+      "step": 799
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.0001372403428374621,
+      "loss": 0.3004,
+      "step": 800
+    },
+    {
+      "epoch": 1.14,
+      "eval_loss": 0.5757906436920166,
+      "eval_runtime": 29.3301,
+      "eval_samples_per_second": 55.54,
+      "eval_steps_per_second": 27.787,
+      "step": 800
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00013710059677428727,
+      "loss": 0.5949,
+      "step": 801
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00013696076664215513,
+      "loss": 0.6565,
+      "step": 802
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0001368208527579171,
+      "loss": 0.585,
+      "step": 803
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00013668085543861429,
+      "loss": 0.7216,
+      "step": 804
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0001365407750014768,
+      "loss": 0.5711,
+      "step": 805
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0001364006117639232,
+      "loss": 0.4327,
+      "step": 806
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0001362603660435596,
+      "loss": 0.6132,
+      "step": 807
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00013612003815817905,
+      "loss": 0.3675,
+      "step": 808
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0001359796284257608,
+      "loss": 0.5741,
+      "step": 809
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00013583913716446946,
+      "loss": 0.3657,
+      "step": 810
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00013569856469265455,
+      "loss": 0.4624,
+      "step": 811
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0001355579113288495,
+      "loss": 0.4706,
+      "step": 812
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.000135417177391771,
+      "loss": 0.404,
+      "step": 813
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00013527636320031844,
+      "loss": 0.4095,
+      "step": 814
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00013513546907357302,
+      "loss": 0.4368,
+      "step": 815
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00013499449533079696,
+      "loss": 0.5346,
+      "step": 816
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00013485344229143307,
+      "loss": 0.3819,
+      "step": 817
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.0001347123102751037,
+      "loss": 0.6743,
+      "step": 818
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00013457109960161024,
+      "loss": 0.7041,
+      "step": 819
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00013442981059093228,
+      "loss": 0.554,
+      "step": 820
+    },
+    {
+      "epoch": 1.17,
+      "eval_loss": 0.5862205624580383,
+      "eval_runtime": 29.3331,
+      "eval_samples_per_second": 55.535,
+      "eval_steps_per_second": 27.784,
+      "step": 820
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.0001342884435632269,
+      "loss": 0.4807,
+      "step": 821
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00013414699883882805,
+      "loss": 0.4877,
+      "step": 822
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.0001340054767382457,
+      "loss": 0.5486,
+      "step": 823
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00013386387758216502,
+      "loss": 0.7256,
+      "step": 824
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.000133722201691446,
+      "loss": 0.562,
+      "step": 825
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00013358044938712235,
+      "loss": 0.41,
+      "step": 826
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00013343862099040107,
+      "loss": 0.4434,
+      "step": 827
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00013329671682266146,
+      "loss": 0.4361,
+      "step": 828
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013315473720545447,
+      "loss": 0.6426,
+      "step": 829
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013301268246050223,
+      "loss": 0.5431,
+      "step": 830
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013287055290969697,
+      "loss": 0.3048,
+      "step": 831
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013272834887510034,
+      "loss": 0.3661,
+      "step": 832
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013258607067894297,
+      "loss": 0.4582,
+      "step": 833
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013244371864362342,
+      "loss": 0.436,
+      "step": 834
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00013230129309170754,
+      "loss": 0.4206,
+      "step": 835
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013215879434592785,
+      "loss": 0.905,
+      "step": 836
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013201622272918264,
+      "loss": 0.5078,
+      "step": 837
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013187357856453544,
+      "loss": 0.4638,
+      "step": 838
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013173086217521408,
+      "loss": 0.4443,
+      "step": 839
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013158807388461,
+      "loss": 0.6394,
+      "step": 840
+    },
+    {
+      "epoch": 1.2,
+      "eval_loss": 0.585025429725647,
+      "eval_runtime": 29.3265,
+      "eval_samples_per_second": 55.547,
+      "eval_steps_per_second": 27.791,
+      "step": 840
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.0001314452140162777,
+      "loss": 0.4049,
+      "step": 841
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013130228289393383,
+      "loss": 0.4156,
+      "step": 842
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00013115928084145642,
+      "loss": 0.386,
+      "step": 843
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00013101620818288434,
+      "loss": 0.4081,
+      "step": 844
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0001308730652424164,
+      "loss": 0.5429,
+      "step": 845
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00013072985234441063,
+      "loss": 0.5267,
+      "step": 846
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00013058656981338357,
+      "loss": 0.4339,
+      "step": 847
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00013044321797400966,
+      "loss": 0.5478,
+      "step": 848
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00013029979715112035,
+      "loss": 0.3668,
+      "step": 849
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.0001301563076697033,
+      "loss": 0.4211,
+      "step": 850
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.0001300127498549019,
+      "loss": 0.3399,
+      "step": 851
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00012986912403201425,
+      "loss": 0.6452,
+      "step": 852
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00012972543052649265,
+      "loss": 0.5335,
+      "step": 853
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00012958166966394268,
+      "loss": 0.4113,
+      "step": 854
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00012943784177012267,
+      "loss": 0.4066,
+      "step": 855
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00012929394717094272,
+      "loss": 0.4953,
+      "step": 856
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012914998619246415,
+      "loss": 0.6675,
+      "step": 857
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012900595916089867,
+      "loss": 0.7448,
+      "step": 858
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012886186640260764,
+      "loss": 0.3679,
+      "step": 859
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012871770824410145,
+      "loss": 0.7135,
+      "step": 860
+    },
+    {
+      "epoch": 1.23,
+      "eval_loss": 0.5899772644042969,
+      "eval_runtime": 29.3159,
+      "eval_samples_per_second": 55.567,
+      "eval_steps_per_second": 27.801,
+      "step": 860
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012857348501203858,
+      "loss": 0.576,
+      "step": 861
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012842919703322505,
+      "loss": 0.4974,
+      "step": 862
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00012828484463461346,
+      "loss": 0.5556,
+      "step": 863
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012814042814330255,
+      "loss": 0.3399,
+      "step": 864
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012799594788653624,
+      "loss": 0.38,
+      "step": 865
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012785140419170282,
+      "loss": 0.5291,
+      "step": 866
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0001277067973863345,
+      "loss": 1.0151,
+      "step": 867
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012756212779810642,
+      "loss": 0.412,
+      "step": 868
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012741739575483592,
+      "loss": 0.4503,
+      "step": 869
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.000127272601584482,
+      "loss": 0.4727,
+      "step": 870
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00012712774561514432,
+      "loss": 0.3761,
+      "step": 871
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00012698282817506267,
+      "loss": 0.3192,
+      "step": 872
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00012683784959261607,
+      "loss": 0.6261,
+      "step": 873
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00012669281019632203,
+      "loss": 0.4313,
+      "step": 874
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00012654771031483602,
+      "loss": 0.764,
+      "step": 875
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00012640255027695046,
+      "loss": 0.2878,
+      "step": 876
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00012625733041159406,
+      "loss": 0.5484,
+      "step": 877
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00012611205104783122,
+      "loss": 0.4229,
+      "step": 878
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00012596671251486098,
+      "loss": 0.7992,
+      "step": 879
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00012582131514201665,
+      "loss": 0.6323,
+      "step": 880
+    },
+    {
+      "epoch": 1.26,
+      "eval_loss": 0.5930511355400085,
+      "eval_runtime": 29.3079,
+      "eval_samples_per_second": 55.582,
+      "eval_steps_per_second": 27.808,
+      "step": 880
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.0001256758592587647,
+      "loss": 0.5132,
+      "step": 881
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00012553034519470434,
+      "loss": 0.6001,
+      "step": 882
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00012538477327956647,
+      "loss": 0.5153,
+      "step": 883
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.0001252391438432132,
+      "loss": 0.5863,
+      "step": 884
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00012509345721563683,
+      "loss": 0.5854,
+      "step": 885
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00012494771372695946,
+      "loss": 0.5447,
+      "step": 886
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00012480191370743189,
+      "loss": 0.5849,
+      "step": 887
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.000124656057487433,
+      "loss": 0.4259,
+      "step": 888
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0001245101453974692,
+      "loss": 0.3614,
+      "step": 889
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00012436417776817322,
+      "loss": 0.515,
+      "step": 890
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0001242181549303039,
+      "loss": 0.662,
+      "step": 891
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012407207721474504,
+      "loss": 0.5037,
+      "step": 892
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.0001239259449525048,
+      "loss": 0.5135,
+      "step": 893
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012377975847471502,
+      "loss": 0.4479,
+      "step": 894
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012363351811263033,
+      "loss": 0.3363,
+      "step": 895
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012348722419762743,
+      "loss": 0.8034,
+      "step": 896
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.0001233408770612044,
+      "loss": 0.8467,
+      "step": 897
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012319447703497996,
+      "loss": 0.5276,
+      "step": 898
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00012304802445069267,
+      "loss": 0.4586,
+      "step": 899
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00012290151964020006,
+      "loss": 0.3257,
+      "step": 900
+    },
+    {
+      "epoch": 1.29,
+      "eval_loss": 0.590179979801178,
+      "eval_runtime": 29.354,
+      "eval_samples_per_second": 55.495,
+      "eval_steps_per_second": 27.765,
+      "step": 900
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00012275496293547825,
+      "loss": 0.4204,
+      "step": 901
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.0001226083546686207,
+      "loss": 0.7001,
+      "step": 902
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00012246169517183777,
+      "loss": 0.5542,
+      "step": 903
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00012231498477745603,
+      "loss": 0.5463,
+      "step": 904
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00012216822381791727,
+      "loss": 0.4087,
+      "step": 905
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012202141262577791,
+      "loss": 0.5635,
+      "step": 906
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012187455153370817,
+      "loss": 0.5886,
+      "step": 907
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012172764087449129,
+      "loss": 0.4857,
+      "step": 908
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012158068098102293,
+      "loss": 0.4152,
+      "step": 909
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.0001214336721863103,
+      "loss": 0.6229,
+      "step": 910
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012128661482347133,
+      "loss": 0.7343,
+      "step": 911
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00012113950922573407,
+      "loss": 0.3661,
+      "step": 912
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012099235572643585,
+      "loss": 0.4915,
+      "step": 913
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012084515465902262,
+      "loss": 0.3756,
+      "step": 914
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012069790635704797,
+      "loss": 0.5162,
+      "step": 915
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012055061115417261,
+      "loss": 0.5953,
+      "step": 916
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.0001204032693841636,
+      "loss": 0.5942,
+      "step": 917
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012025588138089334,
+      "loss": 0.4016,
+      "step": 918
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00012010844747833914,
+      "loss": 0.4787,
+      "step": 919
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011996096801058225,
+      "loss": 0.5183,
+      "step": 920
+    },
+    {
+      "epoch": 1.32,
+      "eval_loss": 0.5762912631034851,
+      "eval_runtime": 29.3663,
+      "eval_samples_per_second": 55.472,
+      "eval_steps_per_second": 27.753,
+      "step": 920
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011981344331180724,
+      "loss": 0.2845,
+      "step": 921
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011966587371630106,
+      "loss": 0.9301,
+      "step": 922
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.0001195182595584525,
+      "loss": 0.6466,
+      "step": 923
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011937060117275123,
+      "loss": 0.4812,
+      "step": 924
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011922289889378726,
+      "loss": 0.5352,
+      "step": 925
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011907515305624992,
+      "loss": 0.5356,
+      "step": 926
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011892736399492735,
+      "loss": 0.4959,
+      "step": 927
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011877953204470562,
+      "loss": 0.3936,
+      "step": 928
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011863165754056796,
+      "loss": 0.4165,
+      "step": 929
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.000118483740817594,
+      "loss": 0.4413,
+      "step": 930
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011833578221095905,
+      "loss": 0.3603,
+      "step": 931
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.0001181877820559334,
+      "loss": 0.5058,
+      "step": 932
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011803974068788142,
+      "loss": 0.5413,
+      "step": 933
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011789165844226087,
+      "loss": 0.6285,
+      "step": 934
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011774353565462213,
+      "loss": 0.5791,
+      "step": 935
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011759537266060747,
+      "loss": 0.5654,
+      "step": 936
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011744716979595032,
+      "loss": 0.7473,
+      "step": 937
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011729892739647427,
+      "loss": 0.477,
+      "step": 938
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011715064579809277,
+      "loss": 0.5739,
+      "step": 939
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011700232533680786,
+      "loss": 0.5383,
+      "step": 940
+    },
+    {
+      "epoch": 1.34,
+      "eval_loss": 0.5842157602310181,
+      "eval_runtime": 29.3073,
+      "eval_samples_per_second": 55.583,
+      "eval_steps_per_second": 27.809,
+      "step": 940
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011685396634870971,
+      "loss": 0.4622,
+      "step": 941
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011670556916997584,
+      "loss": 0.4992,
+      "step": 942
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011655713413687026,
+      "loss": 1.0426,
+      "step": 943
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011640866158574284,
+      "loss": 0.3919,
+      "step": 944
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011626015185302832,
+      "loss": 0.7385,
+      "step": 945
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.0001161116052752458,
+      "loss": 0.3194,
+      "step": 946
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011596302218899783,
+      "loss": 0.4549,
+      "step": 947
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011581440293096974,
+      "loss": 0.4434,
+      "step": 948
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011566574783792872,
+      "loss": 0.5585,
+      "step": 949
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011551705724672326,
+      "loss": 0.3396,
+      "step": 950
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011536833149428226,
+      "loss": 0.4078,
+      "step": 951
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011521957091761429,
+      "loss": 0.4217,
+      "step": 952
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011507077585380677,
+      "loss": 0.4353,
+      "step": 953
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011492194664002538,
+      "loss": 0.4128,
+      "step": 954
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.0001147730836135131,
+      "loss": 0.6397,
+      "step": 955
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.0001146241871115896,
+      "loss": 0.5282,
+      "step": 956
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011447525747165033,
+      "loss": 0.7034,
+      "step": 957
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011432629503116581,
+      "loss": 0.8259,
+      "step": 958
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011417730012768102,
+      "loss": 0.4297,
+      "step": 959
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011402827309881434,
+      "loss": 0.453,
+      "step": 960
+    },
+    {
+      "epoch": 1.37,
+      "eval_loss": 0.5878216624259949,
+      "eval_runtime": 29.3409,
+      "eval_samples_per_second": 55.52,
+      "eval_steps_per_second": 27.777,
+      "step": 960
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011387921428225707,
+      "loss": 0.4299,
+      "step": 961
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011373012401577252,
+      "loss": 0.8583,
+      "step": 962
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011358100263719516,
+      "loss": 0.5711,
+      "step": 963
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011343185048443005,
+      "loss": 0.3149,
+      "step": 964
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011328266789545202,
+      "loss": 0.4907,
+      "step": 965
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011313345520830474,
+      "loss": 0.5732,
+      "step": 966
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011298421276110023,
+      "loss": 0.7914,
+      "step": 967
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011283494089201785,
+      "loss": 0.4798,
+      "step": 968
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011268563993930361,
+      "loss": 0.4937,
+      "step": 969
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011253631024126949,
+      "loss": 0.6512,
+      "step": 970
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011238695213629265,
+      "loss": 0.5409,
+      "step": 971
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011223756596281443,
+      "loss": 0.4722,
+      "step": 972
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011208815205933997,
+      "loss": 0.673,
+      "step": 973
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011193871076443716,
+      "loss": 0.596,
+      "step": 974
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011178924241673595,
+      "loss": 0.5037,
+      "step": 975
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011163974735492757,
+      "loss": 0.4995,
+      "step": 976
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011149022591776388,
+      "loss": 0.5033,
+      "step": 977
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011134067844405647,
+      "loss": 0.4973,
+      "step": 978
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011119110527267584,
+      "loss": 0.4618,
+      "step": 979
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.0001110415067425508,
+      "loss": 0.5305,
+      "step": 980
+    },
+    {
+      "epoch": 1.4,
+      "eval_loss": 0.5975159406661987,
+      "eval_runtime": 29.322,
+      "eval_samples_per_second": 55.555,
+      "eval_steps_per_second": 27.795,
+      "step": 980
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.0001108918831926676,
+      "loss": 0.4924,
+      "step": 981
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011074223496206926,
+      "loss": 0.5541,
+      "step": 982
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00011059256238985456,
+      "loss": 0.3688,
+      "step": 983
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.0001104428658151776,
+      "loss": 0.5225,
+      "step": 984
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00011029314557724675,
+      "loss": 0.6157,
+      "step": 985
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00011014340201532411,
+      "loss": 0.3331,
+      "step": 986
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00010999363546872455,
+      "loss": 0.3965,
+      "step": 987
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00010984384627681504,
+      "loss": 0.3695,
+      "step": 988
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00010969403477901388,
+      "loss": 0.3555,
+      "step": 989
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.0001095442013147899,
+      "loss": 0.3879,
+      "step": 990
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010939434622366174,
+      "loss": 0.4132,
+      "step": 991
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010924446984519696,
+      "loss": 0.4322,
+      "step": 992
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010909457251901147,
+      "loss": 0.5689,
+      "step": 993
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010894465458476854,
+      "loss": 0.5678,
+      "step": 994
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010879471638217826,
+      "loss": 0.7025,
+      "step": 995
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010864475825099651,
+      "loss": 0.2963,
+      "step": 996
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010849478053102444,
+      "loss": 0.5467,
+      "step": 997
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010834478356210748,
+      "loss": 0.4298,
+      "step": 998
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.0001081947676841348,
+      "loss": 0.6676,
+      "step": 999
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.0001080447332370383,
+      "loss": 0.4316,
+      "step": 1000
+    },
+    {
+      "epoch": 1.43,
+      "eval_loss": 0.5829254984855652,
+      "eval_runtime": 29.332,
+      "eval_samples_per_second": 55.537,
+      "eval_steps_per_second": 27.785,
+      "step": 1000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010789468056079211,
+      "loss": 0.7808,
+      "step": 1001
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010774460999541145,
+      "loss": 0.45,
+      "step": 1002
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010759452188095224,
+      "loss": 0.2693,
+      "step": 1003
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010744441655751016,
+      "loss": 0.5441,
+      "step": 1004
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010729429436521983,
+      "loss": 0.3714,
+      "step": 1005
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010714415564425408,
+      "loss": 0.6015,
+      "step": 1006
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010699400073482327,
+      "loss": 0.309,
+      "step": 1007
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010684382997717434,
+      "loss": 0.429,
+      "step": 1008
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010669364371159027,
+      "loss": 0.7312,
+      "step": 1009
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010654344227838906,
+      "loss": 0.6466,
+      "step": 1010
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010639322601792314,
+      "loss": 0.343,
+      "step": 1011
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010624299527057852,
+      "loss": 0.4511,
+      "step": 1012
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010609275037677409,
+      "loss": 0.5177,
+      "step": 1013
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010594249167696068,
+      "loss": 0.4192,
+      "step": 1014
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.0001057922195116205,
+      "loss": 0.4013,
+      "step": 1015
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010564193422126629,
+      "loss": 0.5443,
+      "step": 1016
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010549163614644044,
+      "loss": 0.4555,
+      "step": 1017
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010534132562771437,
+      "loss": 0.3793,
+      "step": 1018
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010519100300568767,
+      "loss": 0.5876,
+      "step": 1019
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010504066862098738,
+      "loss": 0.5992,
+      "step": 1020
+    },
+    {
+      "epoch": 1.46,
+      "eval_loss": 0.5800718069076538,
+      "eval_runtime": 29.3972,
+      "eval_samples_per_second": 55.413,
+      "eval_steps_per_second": 27.724,
+      "step": 1020
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010489032281426719,
+      "loss": 0.2647,
+      "step": 1021
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010473996592620666,
+      "loss": 0.4988,
+      "step": 1022
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010458959829751048,
+      "loss": 0.5841,
+      "step": 1023
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010443922026890764,
+      "loss": 0.4789,
+      "step": 1024
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010428883218115073,
+      "loss": 0.4238,
+      "step": 1025
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010413843437501514,
+      "loss": 0.3631,
+      "step": 1026
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010398802719129823,
+      "loss": 0.5576,
+      "step": 1027
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010383761097081867,
+      "loss": 0.4903,
+      "step": 1028
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010368718605441559,
+      "loss": 0.4636,
+      "step": 1029
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010353675278294776,
+      "loss": 0.5042,
+      "step": 1030
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010338631149729304,
+      "loss": 0.6723,
+      "step": 1031
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010323586253834723,
+      "loss": 0.5232,
+      "step": 1032
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010308540624702373,
+      "loss": 0.452,
+      "step": 1033
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010293494296425242,
+      "loss": 0.6632,
+      "step": 1034
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010278447303097906,
+      "loss": 0.4307,
+      "step": 1035
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.0001026339967881645,
+      "loss": 0.6222,
+      "step": 1036
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010248351457678383,
+      "loss": 0.5757,
+      "step": 1037
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010233302673782576,
+      "loss": 0.4118,
+      "step": 1038
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010218253361229169,
+      "loss": 0.5929,
+      "step": 1039
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010203203554119498,
+      "loss": 0.5043,
+      "step": 1040
+    },
+    {
+      "epoch": 1.49,
+      "eval_loss": 0.573050320148468,
+      "eval_runtime": 29.4258,
+      "eval_samples_per_second": 55.36,
+      "eval_steps_per_second": 27.697,
+      "step": 1040
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010188153286556021,
+      "loss": 0.4027,
+      "step": 1041
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010173102592642245,
+      "loss": 0.5068,
+      "step": 1042
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010158051506482637,
+      "loss": 0.8552,
+      "step": 1043
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010143000062182554,
+      "loss": 0.5423,
+      "step": 1044
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010127948293848166,
+      "loss": 0.5602,
+      "step": 1045
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00010112896235586378,
+      "loss": 0.4201,
+      "step": 1046
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00010097843921504744,
+      "loss": 0.2884,
+      "step": 1047
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00010082791385711408,
+      "loss": 0.753,
+      "step": 1048
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00010067738662315012,
+      "loss": 0.6389,
+      "step": 1049
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00010052685785424628,
+      "loss": 0.4706,
+      "step": 1050
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00010037632789149664,
+      "loss": 0.3332,
+      "step": 1051
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00010022579707599808,
+      "loss": 0.4749,
+      "step": 1052
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00010007526574884942,
+      "loss": 0.6428,
+      "step": 1053
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.99247342511506e-05,
+      "loss": 0.4818,
+      "step": 1054
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.977420292400191e-05,
+      "loss": 0.5046,
+      "step": 1055
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.962367210850338e-05,
+      "loss": 0.6038,
+      "step": 1056
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.947314214575379e-05,
+      "loss": 0.5406,
+      "step": 1057
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.932261337684988e-05,
+      "loss": 0.4457,
+      "step": 1058
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.917208614288596e-05,
+      "loss": 0.6406,
+      "step": 1059
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.902156078495261e-05,
+      "loss": 0.4566,
+      "step": 1060
+    },
+    {
+      "epoch": 1.52,
+      "eval_loss": 0.5777260065078735,
+      "eval_runtime": 29.3855,
+      "eval_samples_per_second": 55.436,
+      "eval_steps_per_second": 27.735,
+      "step": 1060
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.887103764413626e-05,
+      "loss": 0.5897,
+      "step": 1061
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.872051706151836e-05,
+      "loss": 0.4523,
+      "step": 1062
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.856999937817448e-05,
+      "loss": 0.57,
+      "step": 1063
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.841948493517364e-05,
+      "loss": 0.4778,
+      "step": 1064
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.826897407357756e-05,
+      "loss": 0.5249,
+      "step": 1065
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.811846713443977e-05,
+      "loss": 0.3702,
+      "step": 1066
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.796796445880503e-05,
+      "loss": 0.8632,
+      "step": 1067
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.781746638770834e-05,
+      "loss": 0.447,
+      "step": 1068
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.766697326217425e-05,
+      "loss": 0.6032,
+      "step": 1069
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.751648542321618e-05,
+      "loss": 0.3924,
+      "step": 1070
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.736600321183556e-05,
+      "loss": 0.3984,
+      "step": 1071
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.721552696902095e-05,
+      "loss": 0.3234,
+      "step": 1072
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.70650570357476e-05,
+      "loss": 0.4285,
+      "step": 1073
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.691459375297628e-05,
+      "loss": 0.7067,
+      "step": 1074
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.676413746165278e-05,
+      "loss": 0.446,
+      "step": 1075
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.661368850270701e-05,
+      "loss": 0.5859,
+      "step": 1076
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.646324721705224e-05,
+      "loss": 0.3135,
+      "step": 1077
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.631281394558443e-05,
+      "loss": 0.8627,
+      "step": 1078
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.616238902918135e-05,
+      "loss": 0.6596,
+      "step": 1079
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.601197280870178e-05,
+      "loss": 0.4879,
+      "step": 1080
+    },
+    {
+      "epoch": 1.55,
+      "eval_loss": 0.5784857869148254,
+      "eval_runtime": 29.3461,
+      "eval_samples_per_second": 55.51,
+      "eval_steps_per_second": 27.772,
+      "step": 1080
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.58615656249849e-05,
+      "loss": 0.3706,
+      "step": 1081
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.57111678188493e-05,
+      "loss": 0.6009,
+      "step": 1082
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.556077973109237e-05,
+      "loss": 0.5445,
+      "step": 1083
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.541040170248953e-05,
+      "loss": 0.3572,
+      "step": 1084
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.526003407379335e-05,
+      "loss": 0.8192,
+      "step": 1085
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.510967718573282e-05,
+      "loss": 0.2684,
+      "step": 1086
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.495933137901265e-05,
+      "loss": 0.3552,
+      "step": 1087
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.480899699431233e-05,
+      "loss": 0.4733,
+      "step": 1088
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.465867437228566e-05,
+      "loss": 0.459,
+      "step": 1089
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.450836385355958e-05,
+      "loss": 0.4819,
+      "step": 1090
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.435806577873373e-05,
+      "loss": 0.4337,
+      "step": 1091
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.420778048837951e-05,
+      "loss": 0.3997,
+      "step": 1092
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.405750832303931e-05,
+      "loss": 0.5851,
+      "step": 1093
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.390724962322592e-05,
+      "loss": 0.4391,
+      "step": 1094
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.375700472942149e-05,
+      "loss": 0.4987,
+      "step": 1095
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.360677398207687e-05,
+      "loss": 0.5934,
+      "step": 1096
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.345655772161095e-05,
+      "loss": 0.5977,
+      "step": 1097
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.330635628840976e-05,
+      "loss": 0.5727,
+      "step": 1098
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.315617002282565e-05,
+      "loss": 0.3471,
+      "step": 1099
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.300599926517674e-05,
+      "loss": 0.7149,
+      "step": 1100
+    },
+    {
+      "epoch": 1.57,
+      "eval_loss": 0.572716474533081,
+      "eval_runtime": 29.3506,
+      "eval_samples_per_second": 55.501,
+      "eval_steps_per_second": 27.768,
+      "step": 1100
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.285584435574594e-05,
+      "loss": 0.406,
+      "step": 1101
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.27057056347802e-05,
+      "loss": 0.5829,
+      "step": 1102
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.255558344248987e-05,
+      "loss": 0.5076,
+      "step": 1103
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.240547811904775e-05,
+      "loss": 0.7703,
+      "step": 1104
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.225539000458858e-05,
+      "loss": 0.49,
+      "step": 1105
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.210531943920795e-05,
+      "loss": 0.5655,
+      "step": 1106
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.19552667629617e-05,
+      "loss": 0.5238,
+      "step": 1107
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.180523231586524e-05,
+      "loss": 0.3763,
+      "step": 1108
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.165521643789257e-05,
+      "loss": 0.6826,
+      "step": 1109
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.150521946897558e-05,
+      "loss": 0.3059,
+      "step": 1110
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.135524174900351e-05,
+      "loss": 0.52,
+      "step": 1111
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.120528361782175e-05,
+      "loss": 0.4649,
+      "step": 1112
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.105534541523147e-05,
+      "loss": 0.4375,
+      "step": 1113
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.090542748098857e-05,
+      "loss": 0.415,
+      "step": 1114
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.075553015480305e-05,
+      "loss": 0.4202,
+      "step": 1115
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.06056537763383e-05,
+      "loss": 0.4125,
+      "step": 1116
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.045579868521012e-05,
+      "loss": 0.3053,
+      "step": 1117
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.030596522098614e-05,
+      "loss": 0.4462,
+      "step": 1118
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.0156153723185e-05,
+      "loss": 0.4434,
+      "step": 1119
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.00063645312755e-05,
+      "loss": 0.4555,
+      "step": 1120
+    },
+    {
+      "epoch": 1.6,
+      "eval_loss": 0.5823623538017273,
+      "eval_runtime": 29.3348,
+      "eval_samples_per_second": 55.531,
+      "eval_steps_per_second": 27.783,
+      "step": 1120
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.985659798467591e-05,
+      "loss": 0.3941,
+      "step": 1121
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.970685442275328e-05,
+      "loss": 0.505,
+      "step": 1122
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.955713418482243e-05,
+      "loss": 0.5765,
+      "step": 1123
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.940743761014547e-05,
+      "loss": 0.6552,
+      "step": 1124
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.92577650379308e-05,
+      "loss": 0.464,
+      "step": 1125
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.910811680733239e-05,
+      "loss": 0.5661,
+      "step": 1126
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.895849325744922e-05,
+      "loss": 0.6224,
+      "step": 1127
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 8.880889472732419e-05,
+      "loss": 0.5228,
+      "step": 1128
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.865932155594355e-05,
+      "loss": 0.4026,
+      "step": 1129
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.850977408223612e-05,
+      "loss": 0.4856,
+      "step": 1130
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.836025264507243e-05,
+      "loss": 0.6195,
+      "step": 1131
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.821075758326409e-05,
+      "loss": 0.6123,
+      "step": 1132
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.806128923556288e-05,
+      "loss": 0.2697,
+      "step": 1133
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.791184794066005e-05,
+      "loss": 0.465,
+      "step": 1134
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 8.77624340371856e-05,
+      "loss": 0.7926,
+      "step": 1135
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.761304786370741e-05,
+      "loss": 0.3856,
+      "step": 1136
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.746368975873049e-05,
+      "loss": 0.5178,
+      "step": 1137
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.731436006069641e-05,
+      "loss": 0.3967,
+      "step": 1138
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.716505910798219e-05,
+      "loss": 0.418,
+      "step": 1139
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.701578723889978e-05,
+      "loss": 0.5248,
+      "step": 1140
+    },
+    {
+      "epoch": 1.63,
+      "eval_loss": 0.582135796546936,
+      "eval_runtime": 29.3891,
+      "eval_samples_per_second": 55.429,
+      "eval_steps_per_second": 27.731,
+      "step": 1140
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.686654479169528e-05,
+      "loss": 0.3618,
+      "step": 1141
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 8.6717332104548e-05,
+      "loss": 0.681,
+      "step": 1142
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.656814951556996e-05,
+      "loss": 0.6408,
+      "step": 1143
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.641899736280488e-05,
+      "loss": 0.4655,
+      "step": 1144
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.626987598422751e-05,
+      "loss": 0.4905,
+      "step": 1145
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.612078571774294e-05,
+      "loss": 0.3638,
+      "step": 1146
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.59717269011857e-05,
+      "loss": 0.4311,
+      "step": 1147
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.5822699872319e-05,
+      "loss": 0.409,
+      "step": 1148
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.56737049688342e-05,
+      "loss": 0.2719,
+      "step": 1149
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.55247425283497e-05,
+      "loss": 0.5915,
+      "step": 1150
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.537581288841042e-05,
+      "loss": 0.3778,
+      "step": 1151
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.522691638648692e-05,
+      "loss": 0.4242,
+      "step": 1152
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.507805335997462e-05,
+      "loss": 0.5354,
+      "step": 1153
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.492922414619324e-05,
+      "loss": 0.4193,
+      "step": 1154
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.478042908238575e-05,
+      "loss": 0.62,
+      "step": 1155
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.463166850571773e-05,
+      "loss": 0.7998,
+      "step": 1156
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.448294275327677e-05,
+      "loss": 0.6347,
+      "step": 1157
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.433425216207132e-05,
+      "loss": 0.4113,
+      "step": 1158
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.418559706903029e-05,
+      "loss": 0.4017,
+      "step": 1159
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.403697781100218e-05,
+      "loss": 0.4981,
+      "step": 1160
+    },
+    {
+      "epoch": 1.66,
+      "eval_loss": 0.5711487531661987,
+      "eval_runtime": 29.3529,
+      "eval_samples_per_second": 55.497,
+      "eval_steps_per_second": 27.766,
+      "step": 1160
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.388839472475419e-05,
+      "loss": 0.3636,
+      "step": 1161
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.373984814697169e-05,
+      "loss": 0.6068,
+      "step": 1162
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.359133841425718e-05,
+      "loss": 0.8062,
+      "step": 1163
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.344286586312972e-05,
+      "loss": 0.3631,
+      "step": 1164
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.329443083002418e-05,
+      "loss": 0.3347,
+      "step": 1165
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.314603365129034e-05,
+      "loss": 0.5381,
+      "step": 1166
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.299767466319217e-05,
+      "loss": 0.5155,
+      "step": 1167
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.284935420190724e-05,
+      "loss": 0.4295,
+      "step": 1168
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.270107260352571e-05,
+      "loss": 0.4726,
+      "step": 1169
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.255283020404972e-05,
+      "loss": 0.5529,
+      "step": 1170
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.240462733939254e-05,
+      "loss": 0.3748,
+      "step": 1171
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.225646434537787e-05,
+      "loss": 0.4301,
+      "step": 1172
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.210834155773914e-05,
+      "loss": 0.3765,
+      "step": 1173
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.196025931211859e-05,
+      "loss": 0.4272,
+      "step": 1174
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.181221794406661e-05,
+      "loss": 0.4169,
+      "step": 1175
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.166421778904096e-05,
+      "loss": 0.5331,
+      "step": 1176
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.151625918240606e-05,
+      "loss": 0.4877,
+      "step": 1177
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.136834245943205e-05,
+      "loss": 0.6338,
+      "step": 1178
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.122046795529439e-05,
+      "loss": 0.6108,
+      "step": 1179
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.107263600507265e-05,
+      "loss": 0.5595,
+      "step": 1180
+    },
+    {
+      "epoch": 1.69,
+      "eval_loss": 0.5931345224380493,
+      "eval_runtime": 29.1774,
+      "eval_samples_per_second": 55.831,
+      "eval_steps_per_second": 27.933,
+      "step": 1180
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.09248469437501e-05,
+      "loss": 0.3121,
+      "step": 1181
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.077710110621279e-05,
+      "loss": 0.5753,
+      "step": 1182
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.062939882724876e-05,
+      "loss": 0.5813,
+      "step": 1183
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.048174044154753e-05,
+      "loss": 0.4526,
+      "step": 1184
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.033412628369896e-05,
+      "loss": 0.505,
+      "step": 1185
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.018655668819277e-05,
+      "loss": 0.5573,
+      "step": 1186
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.003903198941776e-05,
+      "loss": 0.5836,
+      "step": 1187
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.98915525216609e-05,
+      "loss": 0.454,
+      "step": 1188
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.974411861910667e-05,
+      "loss": 0.6324,
+      "step": 1189
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.959673061583644e-05,
+      "loss": 0.3298,
+      "step": 1190
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.94493888458274e-05,
+      "loss": 0.2667,
+      "step": 1191
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.930209364295207e-05,
+      "loss": 0.5494,
+      "step": 1192
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.915484534097743e-05,
+      "loss": 0.3648,
+      "step": 1193
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.900764427356413e-05,
+      "loss": 0.3719,
+      "step": 1194
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.886049077426595e-05,
+      "loss": 0.5492,
+      "step": 1195
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.871338517652871e-05,
+      "loss": 0.5015,
+      "step": 1196
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.856632781368972e-05,
+      "loss": 0.6613,
+      "step": 1197
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.841931901897708e-05,
+      "loss": 0.4753,
+      "step": 1198
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.82723591255087e-05,
+      "loss": 0.4504,
+      "step": 1199
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.812544846629185e-05,
+      "loss": 0.577,
+      "step": 1200
+    },
+    {
+      "epoch": 1.72,
+      "eval_loss": 0.589836061000824,
+      "eval_runtime": 29.6055,
+      "eval_samples_per_second": 55.024,
+      "eval_steps_per_second": 27.529,
+      "step": 1200
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.79785873742221e-05,
+      "loss": 0.4482,
+      "step": 1201
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.783177618208273e-05,
+      "loss": 0.4069,
+      "step": 1202
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.7685015222544e-05,
+      "loss": 0.3776,
+      "step": 1203
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.753830482816227e-05,
+      "loss": 0.5216,
+      "step": 1204
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.739164533137935e-05,
+      "loss": 0.609,
+      "step": 1205
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.724503706452179e-05,
+      "loss": 0.4599,
+      "step": 1206
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.709848035979994e-05,
+      "loss": 0.4509,
+      "step": 1207
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.695197554930737e-05,
+      "loss": 0.3666,
+      "step": 1208
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.680552296502006e-05,
+      "loss": 0.4005,
+      "step": 1209
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.66591229387956e-05,
+      "loss": 0.5558,
+      "step": 1210
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.651277580237262e-05,
+      "loss": 0.3661,
+      "step": 1211
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.636648188736971e-05,
+      "loss": 0.4188,
+      "step": 1212
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.6220241525285e-05,
+      "loss": 0.4188,
+      "step": 1213
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.607405504749521e-05,
+      "loss": 0.6393,
+      "step": 1214
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.592792278525501e-05,
+      "loss": 0.3854,
+      "step": 1215
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.578184506969612e-05,
+      "loss": 0.3448,
+      "step": 1216
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.563582223182679e-05,
+      "loss": 0.4549,
+      "step": 1217
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.548985460253084e-05,
+      "loss": 0.4321,
+      "step": 1218
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.5343942512567e-05,
+      "loss": 0.6051,
+      "step": 1219
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.519808629256816e-05,
+      "loss": 0.3202,
+      "step": 1220
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 0.5774584412574768,
+      "eval_runtime": 29.3458,
+      "eval_samples_per_second": 55.511,
+      "eval_steps_per_second": 27.772,
+      "step": 1220
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.505228627304055e-05,
+      "loss": 0.4445,
+      "step": 1221
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.490654278436318e-05,
+      "loss": 0.6725,
+      "step": 1222
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.476085615678686e-05,
+      "loss": 0.4999,
+      "step": 1223
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.461522672043355e-05,
+      "loss": 0.4731,
+      "step": 1224
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.44696548052957e-05,
+      "loss": 0.52,
+      "step": 1225
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.432414074123533e-05,
+      "loss": 0.4356,
+      "step": 1226
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.417868485798336e-05,
+      "loss": 0.5952,
+      "step": 1227
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.403328748513902e-05,
+      "loss": 0.3216,
+      "step": 1228
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.388794895216879e-05,
+      "loss": 0.401,
+      "step": 1229
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.374266958840595e-05,
+      "loss": 0.5764,
+      "step": 1230
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.359744972304958e-05,
+      "loss": 0.3612,
+      "step": 1231
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.345228968516397e-05,
+      "loss": 0.4412,
+      "step": 1232
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.330718980367798e-05,
+      "loss": 0.5644,
+      "step": 1233
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.316215040738399e-05,
+      "loss": 0.3483,
+      "step": 1234
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.301717182493734e-05,
+      "loss": 0.5694,
+      "step": 1235
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.28722543848557e-05,
+      "loss": 0.3305,
+      "step": 1236
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.272739841551801e-05,
+      "loss": 0.3934,
+      "step": 1237
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.25826042451641e-05,
+      "loss": 0.4048,
+      "step": 1238
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.243787220189363e-05,
+      "loss": 0.4196,
+      "step": 1239
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.229320261366552e-05,
+      "loss": 0.7182,
+      "step": 1240
+    },
+    {
+      "epoch": 1.77,
+      "eval_loss": 0.5799919366836548,
+      "eval_runtime": 29.3314,
+      "eval_samples_per_second": 55.538,
+      "eval_steps_per_second": 27.786,
+      "step": 1240
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.21485958082972e-05,
+      "loss": 0.5384,
+      "step": 1241
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.200405211346381e-05,
+      "loss": 0.6711,
+      "step": 1242
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.185957185669743e-05,
+      "loss": 0.3968,
+      "step": 1243
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.171515536538653e-05,
+      "loss": 0.4168,
+      "step": 1244
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.157080296677499e-05,
+      "loss": 0.4861,
+      "step": 1245
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.142651498796142e-05,
+      "loss": 0.5064,
+      "step": 1246
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.128229175589857e-05,
+      "loss": 0.4599,
+      "step": 1247
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.113813359739235e-05,
+      "loss": 0.4744,
+      "step": 1248
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.099404083910137e-05,
+      "loss": 0.3751,
+      "step": 1249
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.085001380753589e-05,
+      "loss": 0.6402,
+      "step": 1250
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.070605282905731e-05,
+      "loss": 0.3915,
+      "step": 1251
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.056215822987737e-05,
+      "loss": 0.578,
+      "step": 1252
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.041833033605735e-05,
+      "loss": 0.5009,
+      "step": 1253
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.027456947350738e-05,
+      "loss": 0.478,
+      "step": 1254
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.013087596798578e-05,
+      "loss": 0.4641,
+      "step": 1255
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.998725014509811e-05,
+      "loss": 0.4216,
+      "step": 1256
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.984369233029672e-05,
+      "loss": 0.4131,
+      "step": 1257
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.970020284887969e-05,
+      "loss": 0.511,
+      "step": 1258
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.955678202599033e-05,
+      "loss": 0.716,
+      "step": 1259
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.941343018661645e-05,
+      "loss": 0.5608,
+      "step": 1260
+    },
+    {
+      "epoch": 1.8,
+      "eval_loss": 0.5667746067047119,
+      "eval_runtime": 29.369,
+      "eval_samples_per_second": 55.467,
+      "eval_steps_per_second": 27.75,
+      "step": 1260
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.927014765558943e-05,
+      "loss": 0.6544,
+      "step": 1261
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.912693475758361e-05,
+      "loss": 0.5252,
+      "step": 1262
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.898379181711567e-05,
+      "loss": 0.3907,
+      "step": 1263
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.88407191585436e-05,
+      "loss": 0.4148,
+      "step": 1264
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.869771710606619e-05,
+      "loss": 0.7751,
+      "step": 1265
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.855478598372232e-05,
+      "loss": 0.48,
+      "step": 1266
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.841192611539e-05,
+      "loss": 0.3844,
+      "step": 1267
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.826913782478597e-05,
+      "loss": 0.3519,
+      "step": 1268
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.812642143546459e-05,
+      "loss": 0.7342,
+      "step": 1269
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.798377727081736e-05,
+      "loss": 0.4145,
+      "step": 1270
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.78412056540722e-05,
+      "loss": 0.4057,
+      "step": 1271
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.76987069082925e-05,
+      "loss": 0.4354,
+      "step": 1272
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.75562813563766e-05,
+      "loss": 0.352,
+      "step": 1273
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.741392932105704e-05,
+      "loss": 0.3982,
+      "step": 1274
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 6.727165112489966e-05,
+      "loss": 0.3127,
+      "step": 1275
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.712944709030307e-05,
+      "loss": 0.4363,
+      "step": 1276
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.698731753949779e-05,
+      "loss": 0.4633,
+      "step": 1277
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.684526279454551e-05,
+      "loss": 0.5377,
+      "step": 1278
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.670328317733858e-05,
+      "loss": 0.4467,
+      "step": 1279
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.656137900959896e-05,
+      "loss": 0.5677,
+      "step": 1280
+    },
+    {
+      "epoch": 1.83,
+      "eval_loss": 0.5796756148338318,
+      "eval_runtime": 29.0247,
+      "eval_samples_per_second": 56.125,
+      "eval_steps_per_second": 28.08,
+      "step": 1280
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.641955061287763e-05,
+      "loss": 0.5424,
+      "step": 1281
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.627779830855403e-05,
+      "loss": 0.4185,
+      "step": 1282
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.613612241783502e-05,
+      "loss": 0.5242,
+      "step": 1283
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.599452326175433e-05,
+      "loss": 0.4354,
+      "step": 1284
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.585300116117194e-05,
+      "loss": 0.4879,
+      "step": 1285
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.571155643677309e-05,
+      "loss": 0.3639,
+      "step": 1286
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.557018940906775e-05,
+      "loss": 0.4756,
+      "step": 1287
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.54289003983898e-05,
+      "loss": 0.4869,
+      "step": 1288
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.528768972489631e-05,
+      "loss": 0.2529,
+      "step": 1289
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.514655770856695e-05,
+      "loss": 0.5598,
+      "step": 1290
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.500550466920305e-05,
+      "loss": 0.5447,
+      "step": 1291
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.486453092642702e-05,
+      "loss": 0.32,
+      "step": 1292
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.472363679968156e-05,
+      "loss": 0.5504,
+      "step": 1293
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.458282260822898e-05,
+      "loss": 0.5915,
+      "step": 1294
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.444208867115052e-05,
+      "loss": 0.4419,
+      "step": 1295
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.430143530734546e-05,
+      "loss": 0.506,
+      "step": 1296
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.416086283553053e-05,
+      "loss": 0.4827,
+      "step": 1297
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.402037157423923e-05,
+      "loss": 0.4564,
+      "step": 1298
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.387996184182098e-05,
+      "loss": 0.3509,
+      "step": 1299
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.37396339564404e-05,
+      "loss": 0.5046,
+      "step": 1300
+    },
+    {
+      "epoch": 1.86,
+      "eval_loss": 0.5725200772285461,
+      "eval_runtime": 29.4189,
+      "eval_samples_per_second": 55.373,
+      "eval_steps_per_second": 27.703,
+      "step": 1300
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.359938823607682e-05,
+      "loss": 0.61,
+      "step": 1301
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.345922499852322e-05,
+      "loss": 0.6042,
+      "step": 1302
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.331914456138574e-05,
+      "loss": 0.382,
+      "step": 1303
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.317914724208292e-05,
+      "loss": 0.3951,
+      "step": 1304
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.303923335784485e-05,
+      "loss": 0.4017,
+      "step": 1305
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.289940322571277e-05,
+      "loss": 0.3473,
+      "step": 1306
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.275965716253793e-05,
+      "loss": 0.4138,
+      "step": 1307
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.261999548498114e-05,
+      "loss": 0.5744,
+      "step": 1308
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.248041850951208e-05,
+      "loss": 0.3429,
+      "step": 1309
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.23409265524084e-05,
+      "loss": 0.43,
+      "step": 1310
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.220151992975509e-05,
+      "loss": 0.4251,
+      "step": 1311
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.20621989574439e-05,
+      "loss": 0.5128,
+      "step": 1312
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.192296395117239e-05,
+      "loss": 0.4152,
+      "step": 1313
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.17838152264433e-05,
+      "loss": 0.5217,
+      "step": 1314
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.164475309856399e-05,
+      "loss": 0.7992,
+      "step": 1315
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.150577788264542e-05,
+      "loss": 0.3642,
+      "step": 1316
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.136688989360178e-05,
+      "loss": 0.4655,
+      "step": 1317
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.12280894461495e-05,
+      "loss": 0.4123,
+      "step": 1318
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.108937685480666e-05,
+      "loss": 0.6143,
+      "step": 1319
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.0950752433892297e-05,
+      "loss": 0.5165,
+      "step": 1320
+    },
+    {
+      "epoch": 1.89,
+      "eval_loss": 0.5709309577941895,
+      "eval_runtime": 29.3531,
+      "eval_samples_per_second": 55.497,
+      "eval_steps_per_second": 27.765,
+      "step": 1320
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.081221649752562e-05,
+      "loss": 0.3735,
+      "step": 1321
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.06737693596253e-05,
+      "loss": 0.3759,
+      "step": 1322
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.053541133390892e-05,
+      "loss": 0.6508,
+      "step": 1323
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.0397142733891985e-05,
+      "loss": 0.3797,
+      "step": 1324
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.0258963872887495e-05,
+      "loss": 0.5026,
+      "step": 1325
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.012087506400501e-05,
+      "loss": 0.3529,
+      "step": 1326
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.998287662015003e-05,
+      "loss": 0.359,
+      "step": 1327
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.984496885402341e-05,
+      "loss": 0.451,
+      "step": 1328
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.97071520781204e-05,
+      "loss": 0.482,
+      "step": 1329
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.95694266047301e-05,
+      "loss": 0.4186,
+      "step": 1330
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.94317927459348e-05,
+      "loss": 0.431,
+      "step": 1331
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.9294250813609084e-05,
+      "loss": 0.4495,
+      "step": 1332
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.915680111941926e-05,
+      "loss": 0.4742,
+      "step": 1333
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.901944397482272e-05,
+      "loss": 0.4781,
+      "step": 1334
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.888217969106701e-05,
+      "loss": 0.4772,
+      "step": 1335
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.874500857918936e-05,
+      "loss": 0.7348,
+      "step": 1336
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.8607930950015796e-05,
+      "loss": 0.4883,
+      "step": 1337
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 5.8470947114160544e-05,
+      "loss": 0.6178,
+      "step": 1338
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.8334057382025354e-05,
+      "loss": 0.2811,
+      "step": 1339
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.819726206379864e-05,
+      "loss": 0.6432,
+      "step": 1340
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 0.5817170143127441,
+      "eval_runtime": 29.3316,
+      "eval_samples_per_second": 55.537,
+      "eval_steps_per_second": 27.786,
+      "step": 1340
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.806056146945491e-05,
+      "loss": 0.3731,
+      "step": 1341
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.7923955908754104e-05,
+      "loss": 0.4952,
+      "step": 1342
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.778744569124074e-05,
+      "loss": 0.4035,
+      "step": 1343
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.765103112624328e-05,
+      "loss": 0.4831,
+      "step": 1344
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.7514712522873506e-05,
+      "loss": 0.4987,
+      "step": 1345
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.7378490190025634e-05,
+      "loss": 0.8284,
+      "step": 1346
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.724236443637594e-05,
+      "loss": 0.6285,
+      "step": 1347
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.710633557038166e-05,
+      "loss": 0.5372,
+      "step": 1348
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.6970403900280536e-05,
+      "loss": 0.5845,
+      "step": 1349
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.683456973409015e-05,
+      "loss": 0.437,
+      "step": 1350
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.669883337960704e-05,
+      "loss": 0.732,
+      "step": 1351
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.6563195144406135e-05,
+      "loss": 0.4329,
+      "step": 1352
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.642765533584008e-05,
+      "loss": 0.6346,
+      "step": 1353
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.6292214261038425e-05,
+      "loss": 0.4879,
+      "step": 1354
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.6156872226906995e-05,
+      "loss": 0.2465,
+      "step": 1355
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.602162954012724e-05,
+      "loss": 0.617,
+      "step": 1356
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.58864865071554e-05,
+      "loss": 0.3693,
+      "step": 1357
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.575144343422205e-05,
+      "loss": 0.4881,
+      "step": 1358
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.561650062733114e-05,
+      "loss": 0.6001,
+      "step": 1359
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.5481658392259385e-05,
+      "loss": 0.4973,
+      "step": 1360
+    },
+    {
+      "epoch": 1.95,
+      "eval_loss": 0.5695145130157471,
+      "eval_runtime": 29.331,
+      "eval_samples_per_second": 55.539,
+      "eval_steps_per_second": 27.786,
+      "step": 1360
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.534691703455577e-05,
+      "loss": 0.3581,
+      "step": 1361
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.5212276859540515e-05,
+      "loss": 0.5293,
+      "step": 1362
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.5077738172304737e-05,
+      "loss": 0.3298,
+      "step": 1363
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.494330127770944e-05,
+      "loss": 0.2673,
+      "step": 1364
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.480896648038498e-05,
+      "loss": 0.5709,
+      "step": 1365
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.467473408473053e-05,
+      "loss": 0.398,
+      "step": 1366
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.454060439491297e-05,
+      "loss": 0.5055,
+      "step": 1367
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.4406577714866614e-05,
+      "loss": 0.637,
+      "step": 1368
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.427265434829236e-05,
+      "loss": 0.5337,
+      "step": 1369
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.413883459865694e-05,
+      "loss": 0.4031,
+      "step": 1370
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.400511876919228e-05,
+      "loss": 0.5186,
+      "step": 1371
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.387150716289495e-05,
+      "loss": 0.4191,
+      "step": 1372
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.3738000082525186e-05,
+      "loss": 0.4557,
+      "step": 1373
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.3604597830606515e-05,
+      "loss": 0.5968,
+      "step": 1374
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.347130070942483e-05,
+      "loss": 0.4447,
+      "step": 1375
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.3338109021027804e-05,
+      "loss": 0.3534,
+      "step": 1376
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.32050230672243e-05,
+      "loss": 0.4481,
+      "step": 1377
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.307204314958347e-05,
+      "loss": 0.5887,
+      "step": 1378
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.2939169569434275e-05,
+      "loss": 0.4474,
+      "step": 1379
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.280640262786465e-05,
+      "loss": 0.2903,
+      "step": 1380
+    },
+    {
+      "epoch": 1.97,
+      "eval_loss": 0.5762348771095276,
+      "eval_runtime": 29.5913,
+      "eval_samples_per_second": 55.05,
+      "eval_steps_per_second": 27.542,
+      "step": 1380
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.267374262572089e-05,
+      "loss": 0.8783,
+      "step": 1381
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.25411898636071e-05,
+      "loss": 0.8051,
+      "step": 1382
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.240874464188421e-05,
+      "loss": 0.4557,
+      "step": 1383
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.227640726066954e-05,
+      "loss": 0.689,
+      "step": 1384
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.214417801983611e-05,
+      "loss": 0.4886,
+      "step": 1385
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.201205721901178e-05,
+      "loss": 0.4497,
+      "step": 1386
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.188004515757872e-05,
+      "loss": 0.3783,
+      "step": 1387
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.1748142134672785e-05,
+      "loss": 0.4163,
+      "step": 1388
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.161634844918266e-05,
+      "loss": 0.4249,
+      "step": 1389
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.1484664399749305e-05,
+      "loss": 0.6207,
+      "step": 1390
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.135309028476523e-05,
+      "loss": 0.3058,
+      "step": 1391
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.12216264023738e-05,
+      "loss": 0.7745,
+      "step": 1392
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.1090273050468765e-05,
+      "loss": 0.4179,
+      "step": 1393
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.095903052669323e-05,
+      "loss": 0.4612,
+      "step": 1394
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.08278991284392e-05,
+      "loss": 0.3968,
+      "step": 1395
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.069687915284699e-05,
+      "loss": 0.35,
+      "step": 1396
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.0565970896804306e-05,
+      "loss": 0.4985,
+      "step": 1397
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.043517465694569e-05,
+      "loss": 0.4872,
+      "step": 1398
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2097,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 1.8380571191531274e+18,
+  "trial_name": null,
+  "trial_params": null
+}