diff --git "a/checkpoint-1000/trainer_state.json" "b/checkpoint-1000/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-1000/trainer_state.json"
@@ -0,0 +1,6099 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6925207756232687,
+  "eval_steps": 100,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.8181818181818174e-06,
+      "loss": 0.2935,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3636363636363635e-05,
+      "loss": 0.3591,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.045454545454545e-05,
+      "loss": 0.3512,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.727272727272727e-05,
+      "loss": 0.3309,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4090909090909085e-05,
+      "loss": 0.2684,
+      "step": 5
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.09090909090909e-05,
+      "loss": 0.2415,
+      "step": 6
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7727272727272724e-05,
+      "loss": 0.3725,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.454545454545454e-05,
+      "loss": 0.2985,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.136363636363636e-05,
+      "loss": 0.2536,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.818181818181817e-05,
+      "loss": 0.2281,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.5e-05,
+      "loss": 0.2784,
+      "step": 11
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.18181818181818e-05,
+      "loss": 0.3719,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.863636363636364e-05,
+      "loss": 0.2899,
+      "step": 13
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.545454545454545e-05,
+      "loss": 0.2462,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00010227272727272726,
+      "loss": 0.2554,
+      "step": 15
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00010909090909090908,
+      "loss": 0.332,
+      "step": 16
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.0001159090909090909,
+      "loss": 0.2812,
+      "step": 17
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00012272727272727272,
+      "loss": 0.2226,
+      "step": 18
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00012954545454545453,
+      "loss": 0.2564,
+      "step": 19
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00013636363636363634,
+      "loss": 0.3411,
+      "step": 20
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.00014318181818181818,
+      "loss": 0.3666,
+      "step": 21
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00015,
+      "loss": 0.243,
+      "step": 22
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001568181818181818,
+      "loss": 0.239,
+      "step": 23
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001636363636363636,
+      "loss": 0.2591,
+      "step": 24
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00017045454545454547,
+      "loss": 0.2167,
+      "step": 25
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00017727272727272728,
+      "loss": 0.3029,
+      "step": 26
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00018409090909090909,
+      "loss": 0.3964,
+      "step": 27
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001909090909090909,
+      "loss": 0.3917,
+      "step": 28
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001977272727272727,
+      "loss": 0.2781,
+      "step": 29
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0002045454545454545,
+      "loss": 0.247,
+      "step": 30
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00021136363636363635,
+      "loss": 0.3181,
+      "step": 31
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00021818181818181816,
+      "loss": 0.3593,
+      "step": 32
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.000225,
+      "loss": 0.3164,
+      "step": 33
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0002318181818181818,
+      "loss": 0.3297,
+      "step": 34
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0002386363636363636,
+      "loss": 0.3638,
+      "step": 35
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00024545454545454545,
+      "loss": 0.298,
+      "step": 36
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0002522727272727273,
+      "loss": 0.3642,
+      "step": 37
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00025909090909090907,
+      "loss": 0.2459,
+      "step": 38
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0002659090909090909,
+      "loss": 0.4347,
+      "step": 39
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0002727272727272727,
+      "loss": 0.232,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0002795454545454545,
+      "loss": 0.265,
+      "step": 41
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00028636363636363636,
+      "loss": 0.2941,
+      "step": 42
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00029318181818181814,
+      "loss": 0.3151,
+      "step": 43
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0003,
+      "loss": 0.3219,
+      "step": 44
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0002999996223367247,
+      "loss": 0.3614,
+      "step": 45
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0002999984893488008,
+      "loss": 0.3071,
+      "step": 46
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00029999660104193315,
+      "loss": 0.3223,
+      "step": 47
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00029999395742563066,
+      "loss": 0.3473,
+      "step": 48
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00029999055851320506,
+      "loss": 0.301,
+      "step": 49
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0002999864043217717,
+      "loss": 0.3275,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0002999814948722491,
+      "loss": 0.3894,
+      "step": 51
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0002999758301893587,
+      "loss": 0.3163,
+      "step": 52
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00029996941030162517,
+      "loss": 0.2869,
+      "step": 53
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.000299962235241376,
+      "loss": 0.3036,
+      "step": 54
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00029995430504474105,
+      "loss": 0.2893,
+      "step": 55
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00029994561975165313,
+      "loss": 0.3469,
+      "step": 56
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00029993617940584705,
+      "loss": 0.2517,
+      "step": 57
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0002999259840548597,
+      "loss": 0.2794,
+      "step": 58
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00029991503375003,
+      "loss": 0.3508,
+      "step": 59
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0002999033285464982,
+      "loss": 0.317,
+      "step": 60
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.000299890868503206,
+      "loss": 0.313,
+      "step": 61
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00029987765368289616,
+      "loss": 0.3155,
+      "step": 62
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00029986368415211196,
+      "loss": 0.3203,
+      "step": 63
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0002998489599811972,
+      "loss": 0.254,
+      "step": 64
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0002998334812442955,
+      "loss": 0.3749,
+      "step": 65
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029981724801935025,
+      "loss": 0.2615,
+      "step": 66
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029980026038810395,
+      "loss": 0.3369,
+      "step": 67
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029978251843609815,
+      "loss": 0.3073,
+      "step": 68
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029976402225267247,
+      "loss": 0.3998,
+      "step": 69
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0002997447719309647,
+      "loss": 0.3852,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0002997247675679101,
+      "loss": 0.3341,
+      "step": 71
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0002997040092642407,
+      "loss": 0.4505,
+      "step": 72
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029968249712448525,
+      "loss": 0.3565,
+      "step": 73
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029966023125696837,
+      "loss": 0.2858,
+      "step": 74
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029963721177381,
+      "loss": 0.2831,
+      "step": 75
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0002996134387909251,
+      "loss": 0.3214,
+      "step": 76
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0002995889124280226,
+      "loss": 0.3949,
+      "step": 77
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029956363280860535,
+      "loss": 0.2967,
+      "step": 78
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00029953760005996916,
+      "loss": 0.3067,
+      "step": 79
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029951081431320217,
+      "loss": 0.4727,
+      "step": 80
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029948327570318426,
+      "loss": 0.3245,
+      "step": 81
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0002994549843685864,
+      "loss": 0.2191,
+      "step": 82
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029942594045187,
+      "loss": 0.283,
+      "step": 83
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029939614409928584,
+      "loss": 0.3527,
+      "step": 84
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0002993655954608738,
+      "loss": 0.3712,
+      "step": 85
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.000299334294690462,
+      "loss": 0.3313,
+      "step": 86
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029930224194566564,
+      "loss": 0.2698,
+      "step": 87
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029926943738788666,
+      "loss": 0.2209,
+      "step": 88
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0002992358811823128,
+      "loss": 0.3313,
+      "step": 89
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0002992015734979167,
+      "loss": 0.3474,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029916651450745505,
+      "loss": 0.2964,
+      "step": 91
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029913070438746767,
+      "loss": 0.3489,
+      "step": 92
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029909414331827697,
+      "loss": 0.2803,
+      "step": 93
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002990568314839864,
+      "loss": 0.3481,
+      "step": 94
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029901876907248014,
+      "loss": 0.2366,
+      "step": 95
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002989799562754219,
+      "loss": 0.3413,
+      "step": 96
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029894039328825394,
+      "loss": 0.3224,
+      "step": 97
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.000298900080310196,
+      "loss": 0.3207,
+      "step": 98
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002988590175442446,
+      "loss": 0.3361,
+      "step": 99
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002988172051971717,
+      "loss": 0.3555,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "eval_loss": 0.3661580979824066,
+      "eval_runtime": 430.9338,
+      "eval_samples_per_second": 2.167,
+      "eval_steps_per_second": 0.272,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029877464347952365,
+      "loss": 0.2807,
+      "step": 101
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029873133260562067,
+      "loss": 0.3233,
+      "step": 102
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029868727279355495,
+      "loss": 0.3436,
+      "step": 103
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002986424642651902,
+      "loss": 0.3557,
+      "step": 104
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002985969072461602,
+      "loss": 0.3424,
+      "step": 105
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029855060196586774,
+      "loss": 0.3835,
+      "step": 106
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002985035486574836,
+      "loss": 0.2908,
+      "step": 107
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029845574755794517,
+      "loss": 0.3522,
+      "step": 108
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029840719890795547,
+      "loss": 0.2667,
+      "step": 109
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002983579029519816,
+      "loss": 0.3424,
+      "step": 110
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029830785993825395,
+      "loss": 0.28,
+      "step": 111
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002982570701187646,
+      "loss": 0.3091,
+      "step": 112
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002982055337492662,
+      "loss": 0.2888,
+      "step": 113
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029815325108927063,
+      "loss": 0.3948,
+      "step": 114
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002981002224020478,
+      "loss": 0.3093,
+      "step": 115
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029804644795462435,
+      "loss": 0.314,
+      "step": 116
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002979919280177818,
+      "loss": 0.3129,
+      "step": 117
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002979366628660561,
+      "loss": 0.3162,
+      "step": 118
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029788065277773534,
+      "loss": 0.3045,
+      "step": 119
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.000297823898034859,
+      "loss": 0.4251,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029776639892321606,
+      "loss": 0.382,
+      "step": 121
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029770815573234395,
+      "loss": 0.3747,
+      "step": 122
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002976491687555269,
+      "loss": 0.2517,
+      "step": 123
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029758943828979444,
+      "loss": 0.2771,
+      "step": 124
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029752896463591984,
+      "loss": 0.331,
+      "step": 125
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002974677480984189,
+      "loss": 0.3161,
+      "step": 126
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029740578898554816,
+      "loss": 0.3053,
+      "step": 127
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002973430876093033,
+      "loss": 0.2365,
+      "step": 128
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002972796442854177,
+      "loss": 0.2687,
+      "step": 129
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029721545933336104,
+      "loss": 0.2488,
+      "step": 130
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002971505330763372,
+      "loss": 0.3202,
+      "step": 131
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.000297084865841283,
+      "loss": 0.3289,
+      "step": 132
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029701845795886654,
+      "loss": 0.3061,
+      "step": 133
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002969513097634853,
+      "loss": 0.3474,
+      "step": 134
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002968834215932648,
+      "loss": 0.27,
+      "step": 135
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002968147937900567,
+      "loss": 0.2733,
+      "step": 136
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002967454266994368,
+      "loss": 0.3651,
+      "step": 137
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.000296675320670704,
+      "loss": 0.3448,
+      "step": 138
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00029660447605687785,
+      "loss": 0.2632,
+      "step": 139
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002965328932146971,
+      "loss": 0.3662,
+      "step": 140
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.000296460572504618,
+      "loss": 0.3963,
+      "step": 141
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002963875142908121,
+      "loss": 0.3186,
+      "step": 142
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002963137189411649,
+      "loss": 0.3109,
+      "step": 143
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002962391868272735,
+      "loss": 0.4337,
+      "step": 144
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002961639183244453,
+      "loss": 0.3443,
+      "step": 145
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002960879138116956,
+      "loss": 0.347,
+      "step": 146
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00029601117367174576,
+      "loss": 0.2632,
+      "step": 147
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002959336982910217,
+      "loss": 0.2897,
+      "step": 148
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002958554880596515,
+      "loss": 0.317,
+      "step": 149
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002957765433714634,
+      "loss": 0.3383,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002956968646239845,
+      "loss": 0.3299,
+      "step": 151
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002956164522184377,
+      "loss": 0.2869,
+      "step": 152
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029553530655974063,
+      "loss": 0.2665,
+      "step": 153
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.000295453428056503,
+      "loss": 0.3833,
+      "step": 154
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002953708171210249,
+      "loss": 0.2955,
+      "step": 155
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029528747416929463,
+      "loss": 0.307,
+      "step": 156
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002952033996209864,
+      "loss": 0.3072,
+      "step": 157
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029511859389945845,
+      "loss": 0.3135,
+      "step": 158
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002950330574317509,
+      "loss": 0.2607,
+      "step": 159
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002949467906485836,
+      "loss": 0.3032,
+      "step": 160
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029485979398435365,
+      "loss": 0.4594,
+      "step": 161
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002947720678771338,
+      "loss": 0.3875,
+      "step": 162
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002946836127686697,
+      "loss": 0.2992,
+      "step": 163
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029459442910437797,
+      "loss": 0.3454,
+      "step": 164
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002945045173333439,
+      "loss": 0.3506,
+      "step": 165
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002944138779083191,
+      "loss": 0.3296,
+      "step": 166
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002943225112857194,
+      "loss": 0.324,
+      "step": 167
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002942304179256223,
+      "loss": 0.3531,
+      "step": 168
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002941375982917649,
+      "loss": 0.2671,
+      "step": 169
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002940440528515414,
+      "loss": 0.3906,
+      "step": 170
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002939497820760009,
+      "loss": 0.3454,
+      "step": 171
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002938547864398448,
+      "loss": 0.3434,
+      "step": 172
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029375906642142467,
+      "loss": 0.2481,
+      "step": 173
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029366262250273955,
+      "loss": 0.335,
+      "step": 174
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029356545516943385,
+      "loss": 0.2842,
+      "step": 175
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029346756491079474,
+      "loss": 0.3256,
+      "step": 176
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029336895221974946,
+      "loss": 0.3419,
+      "step": 177
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029326961759286343,
+      "loss": 0.2448,
+      "step": 178
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029316956153033707,
+      "loss": 0.2276,
+      "step": 179
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029306878453600377,
+      "loss": 0.3307,
+      "step": 180
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029296728711732714,
+      "loss": 0.2659,
+      "step": 181
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002928650697853984,
+      "loss": 0.3261,
+      "step": 182
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.000292762133054934,
+      "loss": 0.3047,
+      "step": 183
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029265847744427303,
+      "loss": 0.337,
+      "step": 184
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002925541034753742,
+      "loss": 0.2641,
+      "step": 185
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002924490116738139,
+      "loss": 0.3339,
+      "step": 186
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029234320256878285,
+      "loss": 0.4126,
+      "step": 187
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002922366766930839,
+      "loss": 0.3095,
+      "step": 188
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002921294345831293,
+      "loss": 0.4463,
+      "step": 189
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002920214767789377,
+      "loss": 0.3103,
+      "step": 190
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002919128038241318,
+      "loss": 0.3747,
+      "step": 191
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002918034162659353,
+      "loss": 0.2582,
+      "step": 192
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002916933146551705,
+      "loss": 0.273,
+      "step": 193
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002915824995462551,
+      "loss": 0.3291,
+      "step": 194
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002914709714971998,
+      "loss": 0.2405,
+      "step": 195
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002913587310696052,
+      "loss": 0.2769,
+      "step": 196
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029124577882865915,
+      "loss": 0.2273,
+      "step": 197
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002911321153431338,
+      "loss": 0.4251,
+      "step": 198
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029101774118538284,
+      "loss": 0.323,
+      "step": 199
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029090265693133854,
+      "loss": 0.2576,
+      "step": 200
+    },
+    {
+      "epoch": 0.14,
+      "eval_loss": 0.3576963245868683,
+      "eval_runtime": 410.72,
+      "eval_samples_per_second": 2.274,
+      "eval_steps_per_second": 0.285,
+      "step": 200
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029078686316050874,
+      "loss": 0.3654,
+      "step": 201
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029067036045597435,
+      "loss": 0.342,
+      "step": 202
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029055314940438576,
+      "loss": 0.2253,
+      "step": 203
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029043523059596053,
+      "loss": 0.2887,
+      "step": 204
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002903166046244801,
+      "loss": 0.3511,
+      "step": 205
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002901972720872866,
+      "loss": 0.3639,
+      "step": 206
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002900772335852804,
+      "loss": 0.3686,
+      "step": 207
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002899564897229166,
+      "loss": 0.3076,
+      "step": 208
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002898350411082021,
+      "loss": 0.3183,
+      "step": 209
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00028971288835269275,
+      "loss": 0.2417,
+      "step": 210
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002895900320714899,
+      "loss": 0.3401,
+      "step": 211
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00028946647288323766,
+      "loss": 0.4682,
+      "step": 212
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002893422114101197,
+      "loss": 0.2697,
+      "step": 213
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002892172482778558,
+      "loss": 0.2754,
+      "step": 214
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002890915841156992,
+      "loss": 0.2631,
+      "step": 215
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002889652195564331,
+      "loss": 0.3227,
+      "step": 216
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00028883815523636746,
+      "loss": 0.3394,
+      "step": 217
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00028871039179533605,
+      "loss": 0.2743,
+      "step": 218
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.000288581929876693,
+      "loss": 0.2247,
+      "step": 219
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002884527701273096,
+      "loss": 0.3068,
+      "step": 220
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00028832291319757104,
+      "loss": 0.4094,
+      "step": 221
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00028819235974137335,
+      "loss": 0.3223,
+      "step": 222
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00028806111041611976,
+      "loss": 0.2296,
+      "step": 223
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0002879291658827176,
+      "loss": 0.2486,
+      "step": 224
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0002877965268055749,
+      "loss": 0.2759,
+      "step": 225
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028766319385259713,
+      "loss": 0.3165,
+      "step": 226
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0002875291676951838,
+      "loss": 0.3755,
+      "step": 227
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.000287394449008225,
+      "loss": 0.2859,
+      "step": 228
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0002872590384700979,
+      "loss": 0.3506,
+      "step": 229
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.000287122936762664,
+      "loss": 0.3166,
+      "step": 230
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028698614457126455,
+      "loss": 0.3258,
+      "step": 231
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028684866258471813,
+      "loss": 0.2901,
+      "step": 232
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028671049149531675,
+      "loss": 0.3573,
+      "step": 233
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028657163199882237,
+      "loss": 0.364,
+      "step": 234
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0002864320847944633,
+      "loss": 0.2158,
+      "step": 235
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028629185058493114,
+      "loss": 0.3299,
+      "step": 236
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028615093007637653,
+      "loss": 0.3535,
+      "step": 237
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00028600932397840616,
+      "loss": 0.2786,
+      "step": 238
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002858670330040791,
+      "loss": 0.3313,
+      "step": 239
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00028572405786990294,
+      "loss": 0.3865,
+      "step": 240
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002855803992958304,
+      "loss": 0.2902,
+      "step": 241
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00028543605800525584,
+      "loss": 0.2914,
+      "step": 242
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00028529103472501124,
+      "loss": 0.3619,
+      "step": 243
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002851453301853628,
+      "loss": 0.3738,
+      "step": 244
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002849989451200074,
+      "loss": 0.2853,
+      "step": 245
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00028485188026606834,
+      "loss": 0.2604,
+      "step": 246
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002847041363640923,
+      "loss": 0.2208,
+      "step": 247
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002845557141580453,
+      "loss": 0.2787,
+      "step": 248
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002844066143953088,
+      "loss": 0.2579,
+      "step": 249
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00028425683782667613,
+      "loss": 0.3502,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002841063852063489,
+      "loss": 0.2731,
+      "step": 251
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002839552572919328,
+      "loss": 0.3358,
+      "step": 252
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028380345484443393,
+      "loss": 0.2799,
+      "step": 253
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028365097862825513,
+      "loss": 0.2887,
+      "step": 254
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028349782941119203,
+      "loss": 0.3689,
+      "step": 255
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028334400796442896,
+      "loss": 0.2289,
+      "step": 256
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0002831895150625355,
+      "loss": 0.3759,
+      "step": 257
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0002830343514834622,
+      "loss": 0.2855,
+      "step": 258
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028287851800853697,
+      "loss": 0.3864,
+      "step": 259
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028272201542246073,
+      "loss": 0.2584,
+      "step": 260
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028256484451330403,
+      "loss": 0.3531,
+      "step": 261
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028240700607250245,
+      "loss": 0.2778,
+      "step": 262
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0002822485008948532,
+      "loss": 0.3646,
+      "step": 263
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028208932977851067,
+      "loss": 0.3598,
+      "step": 264
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0002819294935249827,
+      "loss": 0.3691,
+      "step": 265
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00028176899293912643,
+      "loss": 0.2941,
+      "step": 266
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0002816078288291442,
+      "loss": 0.3349,
+      "step": 267
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002814460020065795,
+      "loss": 0.3111,
+      "step": 268
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00028128351328631304,
+      "loss": 0.3575,
+      "step": 269
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002811203634865584,
+      "loss": 0.2307,
+      "step": 270
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00028095655342885804,
+      "loss": 0.2248,
+      "step": 271
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00028079208393807934,
+      "loss": 0.301,
+      "step": 272
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00028062695584241,
+      "loss": 0.2811,
+      "step": 273
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002804611699733543,
+      "loss": 0.2851,
+      "step": 274
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00028029472716572867,
+      "loss": 0.3758,
+      "step": 275
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002801276282576576,
+      "loss": 0.2958,
+      "step": 276
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002799598740905694,
+      "loss": 0.285,
+      "step": 277
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002797914655091919,
+      "loss": 0.3346,
+      "step": 278
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002796224033615482,
+      "loss": 0.3573,
+      "step": 279
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002794526884989525,
+      "loss": 0.2312,
+      "step": 280
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00027928232177600585,
+      "loss": 0.3695,
+      "step": 281
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002791113040505915,
+      "loss": 0.3503,
+      "step": 282
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002789396361838711,
+      "loss": 0.359,
+      "step": 283
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002787673190402799,
+      "loss": 0.3637,
+      "step": 284
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00027859435348752263,
+      "loss": 0.3154,
+      "step": 285
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00027842074039656923,
+      "loss": 0.3313,
+      "step": 286
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002782464806416501,
+      "loss": 0.2544,
+      "step": 287
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00027807157510025205,
+      "loss": 0.3054,
+      "step": 288
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002778960246531138,
+      "loss": 0.423,
+      "step": 289
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00027771983018422145,
+      "loss": 0.2778,
+      "step": 290
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00027754299258080393,
+      "loss": 0.3546,
+      "step": 291
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.000277365512733329,
+      "loss": 0.3297,
+      "step": 292
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00027718739153549817,
+      "loss": 0.2973,
+      "step": 293
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002770086298842426,
+      "loss": 0.3164,
+      "step": 294
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002768292286797184,
+      "loss": 0.241,
+      "step": 295
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00027664918882530225,
+      "loss": 0.3432,
+      "step": 296
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002764685112275866,
+      "loss": 0.3224,
+      "step": 297
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002762871967963755,
+      "loss": 0.2248,
+      "step": 298
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002761052464446795,
+      "loss": 0.2908,
+      "step": 299
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00027592266108871156,
+      "loss": 0.3887,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "eval_loss": 0.3512147068977356,
+      "eval_runtime": 416.8184,
+      "eval_samples_per_second": 2.241,
+      "eval_steps_per_second": 0.281,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002757394416478821,
+      "loss": 0.2728,
+      "step": 301
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002755555890447945,
+      "loss": 0.3177,
+      "step": 302
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002753711042052405,
+      "loss": 0.3394,
+      "step": 303
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002751859880581954,
+      "loss": 0.2704,
+      "step": 304
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002750002415358134,
+      "loss": 0.2746,
+      "step": 305
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00027481386557342305,
+      "loss": 0.3615,
+      "step": 306
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002746268611095225,
+      "loss": 0.2759,
+      "step": 307
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002744392290857747,
+      "loss": 0.3264,
+      "step": 308
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00027425097044700246,
+      "loss": 0.2578,
+      "step": 309
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00027406208614118424,
+      "loss": 0.2866,
+      "step": 310
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002738725771194489,
+      "loss": 0.3237,
+      "step": 311
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002736824443360711,
+      "loss": 0.4085,
+      "step": 312
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00027349168874846636,
+      "loss": 0.2564,
+      "step": 313
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00027330031131718647,
+      "loss": 0.345,
+      "step": 314
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00027310831300591447,
+      "loss": 0.3996,
+      "step": 315
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002729156947814598,
+      "loss": 0.2875,
+      "step": 316
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002727224576137535,
+      "loss": 0.3282,
+      "step": 317
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00027252860247584343,
+      "loss": 0.3408,
+      "step": 318
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.000272334130343889,
+      "loss": 0.2905,
+      "step": 319
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00027213904219715676,
+      "loss": 0.3149,
+      "step": 320
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00027194333901801507,
+      "loss": 0.3763,
+      "step": 321
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002717470217919293,
+      "loss": 0.2918,
+      "step": 322
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00027155009150745674,
+      "loss": 0.364,
+      "step": 323
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002713525491562421,
+      "loss": 0.2722,
+      "step": 324
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002711543957330118,
+      "loss": 0.3424,
+      "step": 325
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002709556322355694,
+      "loss": 0.3069,
+      "step": 326
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00027075625966479074,
+      "loss": 0.2508,
+      "step": 327
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002705562790246183,
+      "loss": 0.2991,
+      "step": 328
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002703556913220567,
+      "loss": 0.4025,
+      "step": 329
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00027015449756716736,
+      "loss": 0.2895,
+      "step": 330
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00026995269877306356,
+      "loss": 0.3384,
+      "step": 331
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002697502959559052,
+      "loss": 0.3113,
+      "step": 332
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00026954729013489373,
+      "loss": 0.3183,
+      "step": 333
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00026934368233226714,
+      "loss": 0.3079,
+      "step": 334
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00026913947357329457,
+      "loss": 0.2294,
+      "step": 335
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002689346648862713,
+      "loss": 0.3304,
+      "step": 336
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002687292573025136,
+      "loss": 0.2668,
+      "step": 337
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00026852325185635354,
+      "loss": 0.367,
+      "step": 338
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00026831664958513367,
+      "loss": 0.2825,
+      "step": 339
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026810945152920177,
+      "loss": 0.2546,
+      "step": 340
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026790165873190585,
+      "loss": 0.3059,
+      "step": 341
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0002676932722395888,
+      "loss": 0.2582,
+      "step": 342
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026748429310158273,
+      "loss": 0.2572,
+      "step": 343
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026727472237020447,
+      "loss": 0.2955,
+      "step": 344
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026706456110074943,
+      "loss": 0.2978,
+      "step": 345
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.000266853810351487,
+      "loss": 0.3181,
+      "step": 346
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026664247118365463,
+      "loss": 0.2878,
+      "step": 347
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026643054466145294,
+      "loss": 0.3152,
+      "step": 348
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026621803185204013,
+      "loss": 0.3325,
+      "step": 349
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026600493382552664,
+      "loss": 0.2938,
+      "step": 350
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026579125165496984,
+      "loss": 0.2782,
+      "step": 351
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00026557698641636835,
+      "loss": 0.2884,
+      "step": 352
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0002653621391886571,
+      "loss": 0.3014,
+      "step": 353
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026514671105370166,
+      "loss": 0.3679,
+      "step": 354
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002649307030962924,
+      "loss": 0.2563,
+      "step": 355
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026471411640413973,
+      "loss": 0.3266,
+      "step": 356
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002644969520678682,
+      "loss": 0.293,
+      "step": 357
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026427921118101096,
+      "loss": 0.3993,
+      "step": 358
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026406089484000466,
+      "loss": 0.2686,
+      "step": 359
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002638420041441833,
+      "loss": 0.284,
+      "step": 360
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002636225401957734,
+      "loss": 0.256,
+      "step": 361
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002634025040998879,
+      "loss": 0.2845,
+      "step": 362
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026318189696452075,
+      "loss": 0.3791,
+      "step": 363
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026296071990054165,
+      "loss": 0.3166,
+      "step": 364
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002627389740216898,
+      "loss": 0.2813,
+      "step": 365
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002625166604445689,
+      "loss": 0.3218,
+      "step": 366
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026229378028864135,
+      "loss": 0.3286,
+      "step": 367
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00026207033467622243,
+      "loss": 0.2546,
+      "step": 368
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002618463247324748,
+      "loss": 0.252,
+      "step": 369
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00026162175158540293,
+      "loss": 0.3465,
+      "step": 370
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00026139661636584714,
+      "loss": 0.3192,
+      "step": 371
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002611709202074782,
+      "loss": 0.2571,
+      "step": 372
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00026094466424679146,
+      "loss": 0.3023,
+      "step": 373
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002607178496231011,
+      "loss": 0.2731,
+      "step": 374
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002604904774785345,
+      "loss": 0.2513,
+      "step": 375
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002602625489580265,
+      "loss": 0.3269,
+      "step": 376
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002600340652093135,
+      "loss": 0.2539,
+      "step": 377
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002598050273829278,
+      "loss": 0.3323,
+      "step": 378
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002595754366321916,
+      "loss": 0.2298,
+      "step": 379
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002593452941132117,
+      "loss": 0.2867,
+      "step": 380
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00025911460098487305,
+      "loss": 0.3437,
+      "step": 381
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002588833584088333,
+      "loss": 0.339,
+      "step": 382
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002586515675495167,
+      "loss": 0.3775,
+      "step": 383
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002584192295741087,
+      "loss": 0.3152,
+      "step": 384
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00025818634565254954,
+      "loss": 0.271,
+      "step": 385
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00025795291695752856,
+      "loss": 0.2948,
+      "step": 386
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002577189446644783,
+      "loss": 0.2434,
+      "step": 387
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002574844299515688,
+      "loss": 0.3804,
+      "step": 388
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002572493739997013,
+      "loss": 0.3665,
+      "step": 389
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002570137779925024,
+      "loss": 0.2785,
+      "step": 390
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00025677764311631825,
+      "loss": 0.2946,
+      "step": 391
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002565409705602085,
+      "loss": 0.3341,
+      "step": 392
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002563037615159403,
+      "loss": 0.219,
+      "step": 393
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00025606601717798207,
+      "loss": 0.3707,
+      "step": 394
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00025582773874349804,
+      "loss": 0.2568,
+      "step": 395
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002555889274123417,
+      "loss": 0.2648,
+      "step": 396
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00025534958438704997,
+      "loss": 0.3418,
+      "step": 397
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00025510971087283705,
+      "loss": 0.1983,
+      "step": 398
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002548693080775886,
+      "loss": 0.3219,
+      "step": 399
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00025462837721185535,
+      "loss": 0.3084,
+      "step": 400
+    },
+    {
+      "epoch": 0.28,
+      "eval_loss": 0.33674970269203186,
+      "eval_runtime": 416.1091,
+      "eval_samples_per_second": 2.245,
+      "eval_steps_per_second": 0.281,
+      "step": 400
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002543869194888471,
+      "loss": 0.3648,
+      "step": 401
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002541449361244268,
+      "loss": 0.3832,
+      "step": 402
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002539024283371042,
+      "loss": 0.3058,
+      "step": 403
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002536593973480297,
+      "loss": 0.2936,
+      "step": 404
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00025341584438098836,
+      "loss": 0.3428,
+      "step": 405
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00025317177066239366,
+      "loss": 0.253,
+      "step": 406
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002529271774212814,
+      "loss": 0.2826,
+      "step": 407
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002526820658893033,
+      "loss": 0.2316,
+      "step": 408
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.000252436437300721,
+      "loss": 0.2631,
+      "step": 409
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002521902928923999,
+      "loss": 0.2844,
+      "step": 410
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00025194363390380263,
+      "loss": 0.2828,
+      "step": 411
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002516964615769831,
+      "loss": 0.2879,
+      "step": 412
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002514487771565801,
+      "loss": 0.2843,
+      "step": 413
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00025120058188981116,
+      "loss": 0.2321,
+      "step": 414
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00025095187702646595,
+      "loss": 0.3335,
+      "step": 415
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00025070266381890056,
+      "loss": 0.2888,
+      "step": 416
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002504529435220306,
+      "loss": 0.3596,
+      "step": 417
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002502027173933252,
+      "loss": 0.2776,
+      "step": 418
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002499519866928006,
+      "loss": 0.2877,
+      "step": 419
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00024970075268301384,
+      "loss": 0.3577,
+      "step": 420
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002494490166290564,
+      "loss": 0.2311,
+      "step": 421
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00024919677979854776,
+      "loss": 0.2853,
+      "step": 422
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002489440434616291,
+      "loss": 0.3409,
+      "step": 423
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002486908088909569,
+      "loss": 0.334,
+      "step": 424
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002484370773616964,
+      "loss": 0.2979,
+      "step": 425
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00024818285015151527,
+      "loss": 0.2281,
+      "step": 426
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002479281285405774,
+      "loss": 0.3019,
+      "step": 427
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.000247672913811536,
+      "loss": 0.2755,
+      "step": 428
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002474172072495275,
+      "loss": 0.2195,
+      "step": 429
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00024716101014216497,
+      "loss": 0.308,
+      "step": 430
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00024690432377953155,
+      "loss": 0.3277,
+      "step": 431
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00024664714945417415,
+      "loss": 0.2556,
+      "step": 432
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00024638948846109656,
+      "loss": 0.295,
+      "step": 433
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002461313420977536,
+      "loss": 0.3255,
+      "step": 434
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002458727116640438,
+      "loss": 0.3477,
+      "step": 435
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002456135984623034,
+      "loss": 0.3645,
+      "step": 436
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002453540037972997,
+      "loss": 0.267,
+      "step": 437
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002450939289762242,
+      "loss": 0.2784,
+      "step": 438
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00024483337530868644,
+      "loss": 0.2993,
+      "step": 439
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.000244572344106707,
+      "loss": 0.2219,
+      "step": 440
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024431083668471134,
+      "loss": 0.3632,
+      "step": 441
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024404885435952265,
+      "loss": 0.3313,
+      "step": 442
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024378639845035579,
+      "loss": 0.3702,
+      "step": 443
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024352347027881003,
+      "loss": 0.2512,
+      "step": 444
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024326007116886296,
+      "loss": 0.3034,
+      "step": 445
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024299620244686357,
+      "loss": 0.2334,
+      "step": 446
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0002427318654415255,
+      "loss": 0.3378,
+      "step": 447
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024246706148392045,
+      "loss": 0.325,
+      "step": 448
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024220179190747156,
+      "loss": 0.3234,
+      "step": 449
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024193605804794646,
+      "loss": 0.3642,
+      "step": 450
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024166986124345076,
+      "loss": 0.3163,
+      "step": 451
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024140320283442122,
+      "loss": 0.3386,
+      "step": 452
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024113608416361898,
+      "loss": 0.2377,
+      "step": 453
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00024086850657612297,
+      "loss": 0.2411,
+      "step": 454
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00024060047141932276,
+      "loss": 0.2512,
+      "step": 455
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00024033198004291232,
+      "loss": 0.2785,
+      "step": 456
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002400630337988826,
+      "loss": 0.3279,
+      "step": 457
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023979363404151525,
+      "loss": 0.2682,
+      "step": 458
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023952378212737554,
+      "loss": 0.2706,
+      "step": 459
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023925347941530556,
+      "loss": 0.1878,
+      "step": 460
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023898272726641741,
+      "loss": 0.3695,
+      "step": 461
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002387115270440863,
+      "loss": 0.3103,
+      "step": 462
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002384398801139439,
+      "loss": 0.3707,
+      "step": 463
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023816778784387094,
+      "loss": 0.2876,
+      "step": 464
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023789525160399096,
+      "loss": 0.2444,
+      "step": 465
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.000237622272766663,
+      "loss": 0.2575,
+      "step": 466
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023734885270647484,
+      "loss": 0.2286,
+      "step": 467
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.000237074992800236,
+      "loss": 0.2691,
+      "step": 468
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00023680069442697088,
+      "loss": 0.2277,
+      "step": 469
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002365259589679117,
+      "loss": 0.3302,
+      "step": 470
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023625078780649178,
+      "loss": 0.2591,
+      "step": 471
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023597518232833833,
+      "loss": 0.2451,
+      "step": 472
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023569914392126558,
+      "loss": 0.3542,
+      "step": 473
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023542267397526776,
+      "loss": 0.2733,
+      "step": 474
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023514577388251224,
+      "loss": 0.3362,
+      "step": 475
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023486844503733216,
+      "loss": 0.356,
+      "step": 476
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023459068883621985,
+      "loss": 0.2722,
+      "step": 477
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023431250667781958,
+      "loss": 0.303,
+      "step": 478
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002340338999629204,
+      "loss": 0.2613,
+      "step": 479
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023375487009444934,
+      "loss": 0.3344,
+      "step": 480
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023347541847746424,
+      "loss": 0.2513,
+      "step": 481
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002331955465191465,
+      "loss": 0.3072,
+      "step": 482
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00023291525562879435,
+      "loss": 0.3652,
+      "step": 483
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023263454721781537,
+      "loss": 0.3671,
+      "step": 484
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023235342269971974,
+      "loss": 0.2753,
+      "step": 485
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023207188349011286,
+      "loss": 0.2894,
+      "step": 486
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023178993100668837,
+      "loss": 0.2629,
+      "step": 487
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023150756666922087,
+      "loss": 0.283,
+      "step": 488
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023122479189955885,
+      "loss": 0.1945,
+      "step": 489
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023094160812161765,
+      "loss": 0.2772,
+      "step": 490
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023065801676137217,
+      "loss": 0.2786,
+      "step": 491
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00023037401924684946,
+      "loss": 0.3114,
+      "step": 492
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.0002300896170081221,
+      "loss": 0.2696,
+      "step": 493
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00022980481147730046,
+      "loss": 0.2742,
+      "step": 494
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00022951960408852575,
+      "loss": 0.3294,
+      "step": 495
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00022923399627796265,
+      "loss": 0.2375,
+      "step": 496
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00022894798948379228,
+      "loss": 0.2628,
+      "step": 497
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00022866158514620483,
+      "loss": 0.3652,
+      "step": 498
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002283747847073923,
+      "loss": 0.3074,
+      "step": 499
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022808758961154129,
+      "loss": 0.2767,
+      "step": 500
+    },
+    {
+      "epoch": 0.35,
+      "eval_loss": 0.32663506269454956,
+      "eval_runtime": 430.4483,
+      "eval_samples_per_second": 2.17,
+      "eval_steps_per_second": 0.272,
+      "step": 500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022780000130482559,
+      "loss": 0.3318,
+      "step": 501
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022751202123539915,
+      "loss": 0.245,
+      "step": 502
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022722365085338858,
+      "loss": 0.288,
+      "step": 503
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002269348916108859,
+      "loss": 0.2872,
+      "step": 504
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002266457449619413,
+      "loss": 0.2428,
+      "step": 505
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022635621236255567,
+      "loss": 0.2677,
+      "step": 506
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022606629527067344,
+      "loss": 0.318,
+      "step": 507
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022577599514617515,
+      "loss": 0.3151,
+      "step": 508
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022548531345087,
+      "loss": 0.2295,
+      "step": 509
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022519425164848874,
+      "loss": 0.3515,
+      "step": 510
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022490281120467605,
+      "loss": 0.3882,
+      "step": 511
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00022461099358698328,
+      "loss": 0.2507,
+      "step": 512
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022431880026486112,
+      "loss": 0.3211,
+      "step": 513
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.000224026232709652,
+      "loss": 0.3136,
+      "step": 514
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022373329239458296,
+      "loss": 0.2906,
+      "step": 515
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022343998079475786,
+      "loss": 0.3354,
+      "step": 516
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022314629938715037,
+      "loss": 0.2776,
+      "step": 517
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022285224965059623,
+      "loss": 0.3228,
+      "step": 518
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022255783306578594,
+      "loss": 0.2954,
+      "step": 519
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022226305111525726,
+      "loss": 0.3041,
+      "step": 520
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022196790528338777,
+      "loss": 0.2367,
+      "step": 521
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022167239705638732,
+      "loss": 0.3199,
+      "step": 522
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022137652792229062,
+      "loss": 0.2585,
+      "step": 523
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.0002210802993709498,
+      "loss": 0.2257,
+      "step": 524
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022078371289402677,
+      "loss": 0.2149,
+      "step": 525
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00022048676998498578,
+      "loss": 0.282,
+      "step": 526
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.000220189472139086,
+      "loss": 0.2475,
+      "step": 527
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002198918208533737,
+      "loss": 0.2864,
+      "step": 528
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002195938176266751,
+      "loss": 0.2675,
+      "step": 529
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021929546395958846,
+      "loss": 0.2711,
+      "step": 530
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021899676135447675,
+      "loss": 0.296,
+      "step": 531
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021869771131546013,
+      "loss": 0.3615,
+      "step": 532
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021839831534840802,
+      "loss": 0.2738,
+      "step": 533
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021809857496093199,
+      "loss": 0.3909,
+      "step": 534
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021779849166237786,
+      "loss": 0.2874,
+      "step": 535
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021749806696381814,
+      "loss": 0.201,
+      "step": 536
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021719730237804456,
+      "loss": 0.2585,
+      "step": 537
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021689619941956024,
+      "loss": 0.3246,
+      "step": 538
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002165947596045723,
+      "loss": 0.2626,
+      "step": 539
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.000216292984450984,
+      "loss": 0.308,
+      "step": 540
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00021599087547838727,
+      "loss": 0.2015,
+      "step": 541
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021568843420805494,
+      "loss": 0.3287,
+      "step": 542
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021538566216293317,
+      "loss": 0.3935,
+      "step": 543
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021508256086763368,
+      "loss": 0.2501,
+      "step": 544
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021477913184842629,
+      "loss": 0.2809,
+      "step": 545
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.0002144753766332308,
+      "loss": 0.3167,
+      "step": 546
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021417129675160995,
+      "loss": 0.247,
+      "step": 547
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021386689373476087,
+      "loss": 0.2614,
+      "step": 548
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021356216911550829,
+      "loss": 0.2917,
+      "step": 549
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021325712442829613,
+      "loss": 0.2902,
+      "step": 550
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021295176120918001,
+      "loss": 0.3168,
+      "step": 551
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021264608099581963,
+      "loss": 0.2715,
+      "step": 552
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021234008532747094,
+      "loss": 0.2379,
+      "step": 553
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021203377574497808,
+      "loss": 0.2701,
+      "step": 554
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00021172715379076632,
+      "loss": 0.315,
+      "step": 555
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00021142022100883358,
+      "loss": 0.2802,
+      "step": 556
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00021111297894474307,
+      "loss": 0.3304,
+      "step": 557
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0002108054291456154,
+      "loss": 0.2461,
+      "step": 558
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00021049757316012077,
+      "loss": 0.2741,
+      "step": 559
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00021018941253847113,
+      "loss": 0.3782,
+      "step": 560
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00020988094883241253,
+      "loss": 0.2922,
+      "step": 561
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00020957218359521706,
+      "loss": 0.2909,
+      "step": 562
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0002092631183816754,
+      "loss": 0.2427,
+      "step": 563
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00020895375474808852,
+      "loss": 0.3164,
+      "step": 564
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00020864409425226023,
+      "loss": 0.2946,
+      "step": 565
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00020833413845348922,
+      "loss": 0.2347,
+      "step": 566
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.000208023888912561,
+      "loss": 0.2967,
+      "step": 567
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00020771334719174039,
+      "loss": 0.3025,
+      "step": 568
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00020740251485476345,
+      "loss": 0.2864,
+      "step": 569
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0002070913934668296,
+      "loss": 0.2973,
+      "step": 570
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020677998459459385,
+      "loss": 0.3408,
+      "step": 571
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020646828980615877,
+      "loss": 0.3269,
+      "step": 572
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002061563106710667,
+      "loss": 0.2433,
+      "step": 573
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020584404876029176,
+      "loss": 0.3203,
+      "step": 574
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002055315056462321,
+      "loss": 0.3005,
+      "step": 575
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002052186829027017,
+      "loss": 0.2235,
+      "step": 576
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020490558210492277,
+      "loss": 0.322,
+      "step": 577
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020459220482951765,
+      "loss": 0.3228,
+      "step": 578
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020427855265450081,
+      "loss": 0.2743,
+      "step": 579
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020396462715927105,
+      "loss": 0.2584,
+      "step": 580
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002036504299246034,
+      "loss": 0.3438,
+      "step": 581
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020333596253264125,
+      "loss": 0.2986,
+      "step": 582
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00020302122656688858,
+      "loss": 0.2532,
+      "step": 583
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002027062236122014,
+      "loss": 0.3165,
+      "step": 584
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0002023909552547804,
+      "loss": 0.3445,
+      "step": 585
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00020207542308216277,
+      "loss": 0.223,
+      "step": 586
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0002017596286832139,
+      "loss": 0.3348,
+      "step": 587
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00020144357364811972,
+      "loss": 0.2205,
+      "step": 588
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00020112725956837873,
+      "loss": 0.3823,
+      "step": 589
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00020081068803679371,
+      "loss": 0.3206,
+      "step": 590
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00020049386064746383,
+      "loss": 0.3496,
+      "step": 591
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0002001767789957768,
+      "loss": 0.3115,
+      "step": 592
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019985944467840045,
+      "loss": 0.3348,
+      "step": 593
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019954185929327506,
+      "loss": 0.2746,
+      "step": 594
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019922402443960509,
+      "loss": 0.3618,
+      "step": 595
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019890594171785125,
+      "loss": 0.2744,
+      "step": 596
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0001985876127297224,
+      "loss": 0.2409,
+      "step": 597
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019826903907816743,
+      "loss": 0.2236,
+      "step": 598
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0001979502223673672,
+      "loss": 0.1992,
+      "step": 599
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019763116420272664,
+      "loss": 0.2262,
+      "step": 600
+    },
+    {
+      "epoch": 0.42,
+      "eval_loss": 0.3170485198497772,
+      "eval_runtime": 435.7355,
+      "eval_samples_per_second": 2.144,
+      "eval_steps_per_second": 0.269,
+      "step": 600
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0001973118661908664,
+      "loss": 0.2505,
+      "step": 601
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019699232993961488,
+      "loss": 0.2638,
+      "step": 602
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019667255705800024,
+      "loss": 0.3205,
+      "step": 603
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0001963525491562421,
+      "loss": 0.2979,
+      "step": 604
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019603230784574356,
+      "loss": 0.2554,
+      "step": 605
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019571183473908308,
+      "loss": 0.2829,
+      "step": 606
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019539113145000623,
+      "loss": 0.2975,
+      "step": 607
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019507019959341776,
+      "loss": 0.3208,
+      "step": 608
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0001947490407853734,
+      "loss": 0.2956,
+      "step": 609
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019442765664307165,
+      "loss": 0.2717,
+      "step": 610
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019410604878484556,
+      "loss": 0.2846,
+      "step": 611
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019378421883015502,
+      "loss": 0.2934,
+      "step": 612
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00019346216839957788,
+      "loss": 0.3237,
+      "step": 613
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019313989911480257,
+      "loss": 0.3288,
+      "step": 614
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019281741259861938,
+      "loss": 0.3011,
+      "step": 615
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019249471047491239,
+      "loss": 0.3063,
+      "step": 616
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0001921717943686516,
+      "loss": 0.2326,
+      "step": 617
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019184866590588439,
+      "loss": 0.3847,
+      "step": 618
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019152532671372737,
+      "loss": 0.2826,
+      "step": 619
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019120177842035853,
+      "loss": 0.3045,
+      "step": 620
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019087802265500852,
+      "loss": 0.2749,
+      "step": 621
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0001905540610479529,
+      "loss": 0.2744,
+      "step": 622
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00019022989523050373,
+      "loss": 0.1984,
+      "step": 623
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00018990552683500125,
+      "loss": 0.2372,
+      "step": 624
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00018958095749480592,
+      "loss": 0.2468,
+      "step": 625
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00018925618884429,
+      "loss": 0.2496,
+      "step": 626
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00018893122251882936,
+      "loss": 0.2924,
+      "step": 627
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00018860606015479536,
+      "loss": 0.2795,
+      "step": 628
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018828070338954638,
+      "loss": 0.2819,
+      "step": 629
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018795515386141974,
+      "loss": 0.2744,
+      "step": 630
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001876294132097236,
+      "loss": 0.2761,
+      "step": 631
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018730348307472824,
+      "loss": 0.3291,
+      "step": 632
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018697736509765822,
+      "loss": 0.2946,
+      "step": 633
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018665106092068408,
+      "loss": 0.2336,
+      "step": 634
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018632457218691384,
+      "loss": 0.2842,
+      "step": 635
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018599790054038486,
+      "loss": 0.3071,
+      "step": 636
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018567104762605562,
+      "loss": 0.3817,
+      "step": 637
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001853440150897974,
+      "loss": 0.4117,
+      "step": 638
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001850168045783858,
+      "loss": 0.2398,
+      "step": 639
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018468941773949282,
+      "loss": 0.2952,
+      "step": 640
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018436185622167824,
+      "loss": 0.3224,
+      "step": 641
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00018403412167438147,
+      "loss": 0.2629,
+      "step": 642
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018370621574791322,
+      "loss": 0.3025,
+      "step": 643
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018337814009344714,
+      "loss": 0.3066,
+      "step": 644
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018304989636301166,
+      "loss": 0.1789,
+      "step": 645
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0001827214862094814,
+      "loss": 0.3176,
+      "step": 646
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018239291128656907,
+      "loss": 0.2573,
+      "step": 647
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018206417324881716,
+      "loss": 0.3049,
+      "step": 648
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0001817352737515894,
+      "loss": 0.3125,
+      "step": 649
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018140621445106263,
+      "loss": 0.1662,
+      "step": 650
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018107699700421845,
+      "loss": 0.2741,
+      "step": 651
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018074762306883457,
+      "loss": 0.2982,
+      "step": 652
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018041809430347684,
+      "loss": 0.3202,
+      "step": 653
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018008841236749092,
+      "loss": 0.2496,
+      "step": 654
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00017975857892099347,
+      "loss": 0.2719,
+      "step": 655
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00017942859562486427,
+      "loss": 0.307,
+      "step": 656
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00017909846414073763,
+      "loss": 0.3083,
+      "step": 657
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.000178768186130994,
+      "loss": 0.3173,
+      "step": 658
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017843776325875176,
+      "loss": 0.2997,
+      "step": 659
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001781071971878587,
+      "loss": 0.2361,
+      "step": 660
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017777648958288363,
+      "loss": 0.3227,
+      "step": 661
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001774456421091082,
+      "loss": 0.2766,
+      "step": 662
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017711465643251823,
+      "loss": 0.2536,
+      "step": 663
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017678353421979548,
+      "loss": 0.2971,
+      "step": 664
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017645227713830936,
+      "loss": 0.2555,
+      "step": 665
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001761208868561083,
+      "loss": 0.3386,
+      "step": 666
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0001757893650419114,
+      "loss": 0.2256,
+      "step": 667
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017545771336510032,
+      "loss": 0.2979,
+      "step": 668
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017512593349571043,
+      "loss": 0.2183,
+      "step": 669
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017479402710442274,
+      "loss": 0.2082,
+      "step": 670
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017446199586255528,
+      "loss": 0.2664,
+      "step": 671
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017412984144205476,
+      "loss": 0.2694,
+      "step": 672
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017379756551548825,
+      "loss": 0.2347,
+      "step": 673
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017346516975603462,
+      "loss": 0.3269,
+      "step": 674
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017313265583747615,
+      "loss": 0.2471,
+      "step": 675
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001728000254341901,
+      "loss": 0.3183,
+      "step": 676
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017246728022114025,
+      "loss": 0.2714,
+      "step": 677
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017213442187386863,
+      "loss": 0.2985,
+      "step": 678
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001718014520684869,
+      "loss": 0.2923,
+      "step": 679
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001714683724816679,
+      "loss": 0.206,
+      "step": 680
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017113518479063738,
+      "loss": 0.2784,
+      "step": 681
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001708018906731654,
+      "loss": 0.3425,
+      "step": 682
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017046849180755793,
+      "loss": 0.229,
+      "step": 683
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00017013498987264832,
+      "loss": 0.2058,
+      "step": 684
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001698013865477891,
+      "loss": 0.3237,
+      "step": 685
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016946768351284327,
+      "loss": 0.258,
+      "step": 686
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016913388244817586,
+      "loss": 0.2244,
+      "step": 687
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016879998503464561,
+      "loss": 0.2383,
+      "step": 688
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016846599295359643,
+      "loss": 0.2196,
+      "step": 689
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016813190788684884,
+      "loss": 0.3826,
+      "step": 690
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016779773151669161,
+      "loss": 0.3966,
+      "step": 691
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0001674634655258734,
+      "loss": 0.2805,
+      "step": 692
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016712911159759406,
+      "loss": 0.2495,
+      "step": 693
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016679467141549617,
+      "loss": 0.2076,
+      "step": 694
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016646014666365675,
+      "loss": 0.3564,
+      "step": 695
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016612553902657876,
+      "loss": 0.2004,
+      "step": 696
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016579085018918224,
+      "loss": 0.2654,
+      "step": 697
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016545608183679637,
+      "loss": 0.2892,
+      "step": 698
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016512123565515065,
+      "loss": 0.3076,
+      "step": 699
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00016478631333036653,
+      "loss": 0.2703,
+      "step": 700
+    },
+    {
+      "epoch": 0.48,
+      "eval_loss": 0.3060886263847351,
+      "eval_runtime": 432.8445,
+      "eval_samples_per_second": 2.158,
+      "eval_steps_per_second": 0.27,
+      "step": 700
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016445131654894876,
+      "loss": 0.3737,
+      "step": 701
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016411624699777717,
+      "loss": 0.196,
+      "step": 702
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016378110636409784,
+      "loss": 0.2705,
+      "step": 703
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016344589633551502,
+      "loss": 0.28,
+      "step": 704
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016311061859998222,
+      "loss": 0.3013,
+      "step": 705
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0001627752748457939,
+      "loss": 0.3199,
+      "step": 706
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016243986676157704,
+      "loss": 0.2383,
+      "step": 707
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0001621043960362826,
+      "loss": 0.2129,
+      "step": 708
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016176886435917675,
+      "loss": 0.208,
+      "step": 709
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016143327341983275,
+      "loss": 0.2469,
+      "step": 710
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016109762490812228,
+      "loss": 0.2663,
+      "step": 711
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016076192051420687,
+      "loss": 0.3564,
+      "step": 712
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0001604261619285295,
+      "loss": 0.2445,
+      "step": 713
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00016009035084180596,
+      "loss": 0.2528,
+      "step": 714
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015975448894501643,
+      "loss": 0.2414,
+      "step": 715
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.000159418577929397,
+      "loss": 0.382,
+      "step": 716
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015908261948643107,
+      "loss": 0.2513,
+      "step": 717
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0001587466153078408,
+      "loss": 0.2635,
+      "step": 718
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015841056708557874,
+      "loss": 0.2655,
+      "step": 719
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015807447651181922,
+      "loss": 0.274,
+      "step": 720
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0001577383452789497,
+      "loss": 0.22,
+      "step": 721
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015740217507956267,
+      "loss": 0.309,
+      "step": 722
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015706596760644637,
+      "loss": 0.3358,
+      "step": 723
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015672972455257723,
+      "loss": 0.2,
+      "step": 724
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015639344761111054,
+      "loss": 0.2409,
+      "step": 725
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015605713847537226,
+      "loss": 0.239,
+      "step": 726
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015572079883885063,
+      "loss": 0.3201,
+      "step": 727
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00015538443039518735,
+      "loss": 0.3946,
+      "step": 728
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0001550480348381691,
+      "loss": 0.3094,
+      "step": 729
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015471161386171922,
+      "loss": 0.2782,
+      "step": 730
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.0001543751691598891,
+      "loss": 0.282,
+      "step": 731
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.0001540387024268494,
+      "loss": 0.2884,
+      "step": 732
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015370221535688193,
+      "loss": 0.3097,
+      "step": 733
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015336570964437075,
+      "loss": 0.2159,
+      "step": 734
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015302918698379387,
+      "loss": 0.2569,
+      "step": 735
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015269264906971457,
+      "loss": 0.2641,
+      "step": 736
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.0001523560975967731,
+      "loss": 0.3836,
+      "step": 737
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.0001520195342596778,
+      "loss": 0.2721,
+      "step": 738
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015168296075319688,
+      "loss": 0.302,
+      "step": 739
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015134637877214967,
+      "loss": 0.3011,
+      "step": 740
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015100979001139828,
+      "loss": 0.2276,
+      "step": 741
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00015067319616583885,
+      "loss": 0.3098,
+      "step": 742
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.0001503365989303931,
+      "loss": 0.1844,
+      "step": 743
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00015,
+      "loss": 0.1585,
+      "step": 744
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0001496634010696069,
+      "loss": 0.2755,
+      "step": 745
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014932680383416118,
+      "loss": 0.2363,
+      "step": 746
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014899020998860175,
+      "loss": 0.2918,
+      "step": 747
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0001486536212278503,
+      "loss": 0.2651,
+      "step": 748
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0001483170392468031,
+      "loss": 0.2688,
+      "step": 749
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014798046574032215,
+      "loss": 0.2422,
+      "step": 750
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014764390240322691,
+      "loss": 0.2634,
+      "step": 751
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014730735093028543,
+      "loss": 0.2553,
+      "step": 752
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014697081301620615,
+      "loss": 0.2172,
+      "step": 753
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014663429035562925,
+      "loss": 0.3092,
+      "step": 754
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014629778464311804,
+      "loss": 0.2259,
+      "step": 755
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0001459612975731506,
+      "loss": 0.331,
+      "step": 756
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014562483084011092,
+      "loss": 0.2816,
+      "step": 757
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00014528838613828076,
+      "loss": 0.2568,
+      "step": 758
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014495196516183095,
+      "loss": 0.2572,
+      "step": 759
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014461556960481265,
+      "loss": 0.3297,
+      "step": 760
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014427920116114935,
+      "loss": 0.3141,
+      "step": 761
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001439428615246277,
+      "loss": 0.3239,
+      "step": 762
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014360655238888947,
+      "loss": 0.3344,
+      "step": 763
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001432702754474228,
+      "loss": 0.2538,
+      "step": 764
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001429340323935536,
+      "loss": 0.2107,
+      "step": 765
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014259782492043738,
+      "loss": 0.3295,
+      "step": 766
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014226165472105021,
+      "loss": 0.2634,
+      "step": 767
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014192552348818078,
+      "loss": 0.2797,
+      "step": 768
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001415894329144212,
+      "loss": 0.149,
+      "step": 769
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0001412533846921592,
+      "loss": 0.2558,
+      "step": 770
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00014091738051356896,
+      "loss": 0.3599,
+      "step": 771
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.000140581422070603,
+      "loss": 0.2474,
+      "step": 772
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0001402455110549836,
+      "loss": 0.2546,
+      "step": 773
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013990964915819404,
+      "loss": 0.238,
+      "step": 774
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013957383807147053,
+      "loss": 0.3249,
+      "step": 775
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013923807948579308,
+      "loss": 0.2205,
+      "step": 776
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0001389023750918777,
+      "loss": 0.2654,
+      "step": 777
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013856672658016728,
+      "loss": 0.2748,
+      "step": 778
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013823113564082325,
+      "loss": 0.2586,
+      "step": 779
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0001378956039637174,
+      "loss": 0.2715,
+      "step": 780
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013756013323842288,
+      "loss": 0.1985,
+      "step": 781
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0001372247251542061,
+      "loss": 0.2243,
+      "step": 782
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0001368893814000178,
+      "loss": 0.2082,
+      "step": 783
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013655410366448498,
+      "loss": 0.3412,
+      "step": 784
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013621889363590214,
+      "loss": 0.2684,
+      "step": 785
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.00013588375300222283,
+      "loss": 0.3061,
+      "step": 786
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013554868345105124,
+      "loss": 0.2027,
+      "step": 787
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013521368666963345,
+      "loss": 0.2806,
+      "step": 788
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013487876434484933,
+      "loss": 0.2321,
+      "step": 789
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013454391816320366,
+      "loss": 0.2878,
+      "step": 790
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013420914981081774,
+      "loss": 0.221,
+      "step": 791
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013387446097342127,
+      "loss": 0.2298,
+      "step": 792
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.0001335398533363432,
+      "loss": 0.2317,
+      "step": 793
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.0001332053285845038,
+      "loss": 0.3981,
+      "step": 794
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013287088840240591,
+      "loss": 0.3128,
+      "step": 795
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013253653447412657,
+      "loss": 0.3695,
+      "step": 796
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013220226848330839,
+      "loss": 0.188,
+      "step": 797
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013186809211315116,
+      "loss": 0.3359,
+      "step": 798
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013153400704640357,
+      "loss": 0.2317,
+      "step": 799
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013120001496535433,
+      "loss": 0.2897,
+      "step": 800
+    },
+    {
+      "epoch": 0.55,
+      "eval_loss": 0.2967790365219116,
+      "eval_runtime": 413.6599,
+      "eval_samples_per_second": 2.258,
+      "eval_steps_per_second": 0.283,
+      "step": 800
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00013086611755182414,
+      "loss": 0.2827,
+      "step": 801
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0001305323164871567,
+      "loss": 0.271,
+      "step": 802
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00013019861345221087,
+      "loss": 0.2886,
+      "step": 803
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0001298650101273517,
+      "loss": 0.2194,
+      "step": 804
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012953150819244207,
+      "loss": 0.2513,
+      "step": 805
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012919810932683463,
+      "loss": 0.2318,
+      "step": 806
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0001288648152093626,
+      "loss": 0.2832,
+      "step": 807
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012853162751833206,
+      "loss": 0.2341,
+      "step": 808
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012819854793151312,
+      "loss": 0.256,
+      "step": 809
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012786557812613137,
+      "loss": 0.3107,
+      "step": 810
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012753271977885975,
+      "loss": 0.1994,
+      "step": 811
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0001271999745658099,
+      "loss": 0.3033,
+      "step": 812
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012686734416252385,
+      "loss": 0.2753,
+      "step": 813
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012653483024396533,
+      "loss": 0.285,
+      "step": 814
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00012620243448451175,
+      "loss": 0.2635,
+      "step": 815
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012587015855794524,
+      "loss": 0.29,
+      "step": 816
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012553800413744472,
+      "loss": 0.2186,
+      "step": 817
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012520597289557728,
+      "loss": 0.2064,
+      "step": 818
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012487406650428954,
+      "loss": 0.2735,
+      "step": 819
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012454228663489965,
+      "loss": 0.2375,
+      "step": 820
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012421063495808853,
+      "loss": 0.2282,
+      "step": 821
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0001238791131438917,
+      "loss": 0.3249,
+      "step": 822
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012354772286169065,
+      "loss": 0.2508,
+      "step": 823
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0001232164657802045,
+      "loss": 0.2783,
+      "step": 824
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0001228853435674818,
+      "loss": 0.2948,
+      "step": 825
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0001225543578908918,
+      "loss": 0.2618,
+      "step": 826
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012222351041711637,
+      "loss": 0.215,
+      "step": 827
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012189280281214126,
+      "loss": 0.2082,
+      "step": 828
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00012156223674124823,
+      "loss": 0.2574,
+      "step": 829
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.000121231813869006,
+      "loss": 0.2574,
+      "step": 830
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00012090153585926237,
+      "loss": 0.2667,
+      "step": 831
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00012057140437513575,
+      "loss": 0.2521,
+      "step": 832
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00012024142107900652,
+      "loss": 0.241,
+      "step": 833
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011991158763250908,
+      "loss": 0.2791,
+      "step": 834
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011958190569652316,
+      "loss": 0.243,
+      "step": 835
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011925237693116544,
+      "loss": 0.2643,
+      "step": 836
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011892300299578159,
+      "loss": 0.2315,
+      "step": 837
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011859378554893731,
+      "loss": 0.2066,
+      "step": 838
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011826472624841058,
+      "loss": 0.1797,
+      "step": 839
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0001179358267511828,
+      "loss": 0.281,
+      "step": 840
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011760708871343093,
+      "loss": 0.2792,
+      "step": 841
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011727851379051865,
+      "loss": 0.2355,
+      "step": 842
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011695010363698834,
+      "loss": 0.2957,
+      "step": 843
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00011662185990655284,
+      "loss": 0.4294,
+      "step": 844
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011629378425208677,
+      "loss": 0.3326,
+      "step": 845
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011596587832561854,
+      "loss": 0.2172,
+      "step": 846
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011563814377832175,
+      "loss": 0.3538,
+      "step": 847
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011531058226050718,
+      "loss": 0.2216,
+      "step": 848
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011498319542161423,
+      "loss": 0.2975,
+      "step": 849
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011465598491020261,
+      "loss": 0.2169,
+      "step": 850
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011432895237394435,
+      "loss": 0.3459,
+      "step": 851
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0001140020994596151,
+      "loss": 0.2275,
+      "step": 852
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011367542781308613,
+      "loss": 0.278,
+      "step": 853
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011334893907931586,
+      "loss": 0.2121,
+      "step": 854
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011302263490234176,
+      "loss": 0.2383,
+      "step": 855
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0001126965169252718,
+      "loss": 0.2471,
+      "step": 856
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011237058679027641,
+      "loss": 0.2681,
+      "step": 857
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00011204484613858025,
+      "loss": 0.2053,
+      "step": 858
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0001117192966104536,
+      "loss": 0.2396,
+      "step": 859
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00011139393984520465,
+      "loss": 0.2365,
+      "step": 860
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00011106877748117063,
+      "loss": 0.1745,
+      "step": 861
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00011074381115570998,
+      "loss": 0.253,
+      "step": 862
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0001104190425051941,
+      "loss": 0.2984,
+      "step": 863
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00011009447316499873,
+      "loss": 0.2862,
+      "step": 864
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010977010476949628,
+      "loss": 0.2951,
+      "step": 865
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010944593895204705,
+      "loss": 0.2645,
+      "step": 866
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010912197734499144,
+      "loss": 0.285,
+      "step": 867
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010879822157964149,
+      "loss": 0.2188,
+      "step": 868
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010847467328627259,
+      "loss": 0.2755,
+      "step": 869
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010815133409411562,
+      "loss": 0.2713,
+      "step": 870
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010782820563134835,
+      "loss": 0.2568,
+      "step": 871
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010750528952508762,
+      "loss": 0.2412,
+      "step": 872
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00010718258740138059,
+      "loss": 0.2863,
+      "step": 873
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0001068601008851974,
+      "loss": 0.2702,
+      "step": 874
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0001065378316004221,
+      "loss": 0.1762,
+      "step": 875
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.000106215781169845,
+      "loss": 0.3286,
+      "step": 876
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010589395121515441,
+      "loss": 0.1931,
+      "step": 877
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010557234335692835,
+      "loss": 0.16,
+      "step": 878
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010525095921462659,
+      "loss": 0.2984,
+      "step": 879
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0001049298004065822,
+      "loss": 0.2614,
+      "step": 880
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010460886854999375,
+      "loss": 0.3418,
+      "step": 881
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010428816526091697,
+      "loss": 0.2817,
+      "step": 882
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010396769215425642,
+      "loss": 0.3755,
+      "step": 883
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0001036474508437579,
+      "loss": 0.254,
+      "step": 884
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010332744294199972,
+      "loss": 0.3858,
+      "step": 885
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010300767006038513,
+      "loss": 0.2092,
+      "step": 886
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010268813380913363,
+      "loss": 0.2513,
+      "step": 887
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00010236883579727336,
+      "loss": 0.2815,
+      "step": 888
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00010204977763263279,
+      "loss": 0.2847,
+      "step": 889
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00010173096092183255,
+      "loss": 0.286,
+      "step": 890
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00010141238727027761,
+      "loss": 0.2514,
+      "step": 891
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00010109405828214871,
+      "loss": 0.32,
+      "step": 892
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00010077597556039489,
+      "loss": 0.3244,
+      "step": 893
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00010045814070672498,
+      "loss": 0.2519,
+      "step": 894
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00010014055532159956,
+      "loss": 0.1773,
+      "step": 895
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.982322100422319e-05,
+      "loss": 0.2859,
+      "step": 896
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.950613935253612e-05,
+      "loss": 0.1867,
+      "step": 897
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.918931196320629e-05,
+      "loss": 0.2626,
+      "step": 898
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.887274043162123e-05,
+      "loss": 0.3173,
+      "step": 899
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.855642635188028e-05,
+      "loss": 0.3112,
+      "step": 900
+    },
+    {
+      "epoch": 0.62,
+      "eval_loss": 0.2862403690814972,
+      "eval_runtime": 414.2075,
+      "eval_samples_per_second": 2.255,
+      "eval_steps_per_second": 0.282,
+      "step": 900
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.824037131678615e-05,
+      "loss": 0.3049,
+      "step": 901
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.792457691783723e-05,
+      "loss": 0.256,
+      "step": 902
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.76090447452196e-05,
+      "loss": 0.2294,
+      "step": 903
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.729377638779857e-05,
+      "loss": 0.286,
+      "step": 904
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.697877343311144e-05,
+      "loss": 0.281,
+      "step": 905
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.666403746735868e-05,
+      "loss": 0.2442,
+      "step": 906
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.634957007539659e-05,
+      "loss": 0.2457,
+      "step": 907
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.603537284072899e-05,
+      "loss": 0.2488,
+      "step": 908
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.572144734549919e-05,
+      "loss": 0.2111,
+      "step": 909
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.540779517048235e-05,
+      "loss": 0.2211,
+      "step": 910
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.50944178950772e-05,
+      "loss": 0.3047,
+      "step": 911
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.47813170972983e-05,
+      "loss": 0.2119,
+      "step": 912
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.446849435376795e-05,
+      "loss": 0.3416,
+      "step": 913
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.415595123970821e-05,
+      "loss": 0.2621,
+      "step": 914
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.38436893289333e-05,
+      "loss": 0.3434,
+      "step": 915
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.353171019384119e-05,
+      "loss": 0.2789,
+      "step": 916
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.322001540540614e-05,
+      "loss": 0.2642,
+      "step": 917
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.290860653317035e-05,
+      "loss": 0.2013,
+      "step": 918
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.259748514523653e-05,
+      "loss": 0.2641,
+      "step": 919
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.228665280825961e-05,
+      "loss": 0.3053,
+      "step": 920
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.197611108743899e-05,
+      "loss": 0.2645,
+      "step": 921
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.166586154651082e-05,
+      "loss": 0.2962,
+      "step": 922
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.135590574773973e-05,
+      "loss": 0.2297,
+      "step": 923
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.104624525191145e-05,
+      "loss": 0.2757,
+      "step": 924
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.073688161832457e-05,
+      "loss": 0.2522,
+      "step": 925
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.042781640478291e-05,
+      "loss": 0.1648,
+      "step": 926
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.011905116758752e-05,
+      "loss": 0.2364,
+      "step": 927
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 8.981058746152887e-05,
+      "loss": 0.3065,
+      "step": 928
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 8.950242683987925e-05,
+      "loss": 0.2621,
+      "step": 929
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 8.919457085438458e-05,
+      "loss": 0.2498,
+      "step": 930
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 8.888702105525695e-05,
+      "loss": 0.2306,
+      "step": 931
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.857977899116645e-05,
+      "loss": 0.2246,
+      "step": 932
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.827284620923368e-05,
+      "loss": 0.228,
+      "step": 933
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.796622425502192e-05,
+      "loss": 0.1776,
+      "step": 934
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.765991467252908e-05,
+      "loss": 0.2613,
+      "step": 935
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.735391900418033e-05,
+      "loss": 0.2944,
+      "step": 936
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.704823879081997e-05,
+      "loss": 0.2655,
+      "step": 937
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.67428755717039e-05,
+      "loss": 0.3042,
+      "step": 938
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.643783088449169e-05,
+      "loss": 0.262,
+      "step": 939
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.613310626523909e-05,
+      "loss": 0.2258,
+      "step": 940
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.58287032483901e-05,
+      "loss": 0.2142,
+      "step": 941
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.552462336676914e-05,
+      "loss": 0.2212,
+      "step": 942
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.522086815157373e-05,
+      "loss": 0.2267,
+      "step": 943
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.491743913236628e-05,
+      "loss": 0.2237,
+      "step": 944
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 8.461433783706682e-05,
+      "loss": 0.2641,
+      "step": 945
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.431156579194507e-05,
+      "loss": 0.3007,
+      "step": 946
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.400912452161271e-05,
+      "loss": 0.2414,
+      "step": 947
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.370701554901602e-05,
+      "loss": 0.2293,
+      "step": 948
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.340524039542769e-05,
+      "loss": 0.2484,
+      "step": 949
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.310380058043972e-05,
+      "loss": 0.331,
+      "step": 950
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.280269762195539e-05,
+      "loss": 0.2931,
+      "step": 951
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.250193303618182e-05,
+      "loss": 0.293,
+      "step": 952
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.220150833762215e-05,
+      "loss": 0.1975,
+      "step": 953
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.190142503906798e-05,
+      "loss": 0.2335,
+      "step": 954
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.160168465159199e-05,
+      "loss": 0.208,
+      "step": 955
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.130228868453988e-05,
+      "loss": 0.3428,
+      "step": 956
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.10032386455232e-05,
+      "loss": 0.2472,
+      "step": 957
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.070453604041155e-05,
+      "loss": 0.3039,
+      "step": 958
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.04061823733249e-05,
+      "loss": 0.2255,
+      "step": 959
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.010817914662631e-05,
+      "loss": 0.2257,
+      "step": 960
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.9810527860914e-05,
+      "loss": 0.3138,
+      "step": 961
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.951323001501418e-05,
+      "loss": 0.1823,
+      "step": 962
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.92162871059732e-05,
+      "loss": 0.2379,
+      "step": 963
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.89197006290502e-05,
+      "loss": 0.2631,
+      "step": 964
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.86234720777094e-05,
+      "loss": 0.206,
+      "step": 965
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.83276029436127e-05,
+      "loss": 0.2091,
+      "step": 966
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.803209471661223e-05,
+      "loss": 0.2355,
+      "step": 967
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.773694888474267e-05,
+      "loss": 0.1977,
+      "step": 968
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.744216693421402e-05,
+      "loss": 0.2937,
+      "step": 969
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.714775034940373e-05,
+      "loss": 0.263,
+      "step": 970
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.685370061284964e-05,
+      "loss": 0.2564,
+      "step": 971
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.656001920524219e-05,
+      "loss": 0.2825,
+      "step": 972
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.626670760541701e-05,
+      "loss": 0.2408,
+      "step": 973
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 7.597376729034797e-05,
+      "loss": 0.3018,
+      "step": 974
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.568119973513885e-05,
+      "loss": 0.2988,
+      "step": 975
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.538900641301672e-05,
+      "loss": 0.1808,
+      "step": 976
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.509718879532394e-05,
+      "loss": 0.2324,
+      "step": 977
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.480574835151128e-05,
+      "loss": 0.1946,
+      "step": 978
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.451468654912999e-05,
+      "loss": 0.2703,
+      "step": 979
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.422400485382483e-05,
+      "loss": 0.2354,
+      "step": 980
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.393370472932654e-05,
+      "loss": 0.2754,
+      "step": 981
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.364378763744429e-05,
+      "loss": 0.3297,
+      "step": 982
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.335425503805872e-05,
+      "loss": 0.2702,
+      "step": 983
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.306510838911407e-05,
+      "loss": 0.2039,
+      "step": 984
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.27763491466114e-05,
+      "loss": 0.2706,
+      "step": 985
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.248797876460085e-05,
+      "loss": 0.2029,
+      "step": 986
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.219999869517439e-05,
+      "loss": 0.3196,
+      "step": 987
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.191241038845874e-05,
+      "loss": 0.1675,
+      "step": 988
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.162521529260767e-05,
+      "loss": 0.1959,
+      "step": 989
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 7.133841485379512e-05,
+      "loss": 0.2223,
+      "step": 990
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 7.105201051620772e-05,
+      "loss": 0.2514,
+      "step": 991
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 7.076600372203735e-05,
+      "loss": 0.2726,
+      "step": 992
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 7.04803959114743e-05,
+      "loss": 0.2536,
+      "step": 993
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 7.019518852269953e-05,
+      "loss": 0.2934,
+      "step": 994
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 6.99103829918779e-05,
+      "loss": 0.2486,
+      "step": 995
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 6.962598075315046e-05,
+      "loss": 0.2011,
+      "step": 996
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 6.934198323862784e-05,
+      "loss": 0.2214,
+      "step": 997
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 6.905839187838235e-05,
+      "loss": 0.2271,
+      "step": 998
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 6.877520810044114e-05,
+      "loss": 0.293,
+      "step": 999
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 6.849243333077919e-05,
+      "loss": 0.2479,
+      "step": 1000
+    },
+    {
+      "epoch": 0.69,
+      "eval_loss": 0.27964773774147034,
+      "eval_runtime": 411.8246,
+      "eval_samples_per_second": 2.268,
+      "eval_steps_per_second": 0.284,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1444,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 1.1151496043737252e+18,
+  "trial_name": null,
+  "trial_params": null
+}